0 files changed, 0 insertions, 0 deletions
diff --git a/drivers/net/ethernet/3com/3c509.c b/drivers/net/ethernet/3com/3c509.c
index db8592d412ab..fb68339e1511 100644
--- a/drivers/net/ethernet/3com/3c509.c
+++ b/drivers/net/ethernet/3com/3c509.c
@@ -60,8 +60,6 @@
 */
 
 #define DRV_NAME	"3c509"
-#define DRV_VERSION	"1.20"
-#define DRV_RELDATE	"04Feb2008"
 
 /* A few values that may be tweaked. */
 
@@ -92,8 +90,6 @@
 #include <asm/io.h>
 #include <asm/irq.h>
 
-static char version[] = DRV_NAME ".c:" DRV_VERSION " " DRV_RELDATE " becker@scyld.com\n";
-
 #ifdef EL3_DEBUG
 static int el3_debug = EL3_DEBUG;
 #else
@@ -196,7 +192,7 @@ static struct net_device_stats *el3_get_stats(struct net_device *dev);
 static int el3_rx(struct net_device *dev);
 static int el3_close(struct net_device *dev);
 static void set_multicast_list(struct net_device *dev);
-static void el3_tx_timeout (struct net_device *dev);
+static void el3_tx_timeout (struct net_device *dev, unsigned int txqueue);
 static void el3_down(struct net_device *dev);
 static void el3_up(struct net_device *dev);
 static const struct ethtool_ops ethtool_ops;
@@ -274,7 +270,7 @@ static void el3_dev_fill(struct net_device *dev, __be16 *phys_addr, int ioaddr,
 {
 	struct el3_private *lp = netdev_priv(dev);
 
-	memcpy(dev->dev_addr, phys_addr, ETH_ALEN);
+	eth_hw_addr_set(dev, (u8 *)phys_addr);
 	dev->base_addr = ioaddr;
 	dev->irq = irq;
 	dev->if_port = if_port;
@@ -306,7 +302,6 @@ static int el3_isa_match(struct device *pdev, unsigned int ndev)
 		return -ENOMEM;
 
 	SET_NETDEV_DEV(dev, pdev);
-	netdev_boot_setup_check(dev);
 
 	if (!request_region(ioaddr, EL3_IO_EXTENT, "3c509-isa")) {
 		free_netdev(dev);
@@ -339,12 +334,11 @@ static int el3_isa_match(struct device *pdev, unsigned int ndev)
 	return 1;
 }
 
-static int el3_isa_remove(struct device *pdev,
+static void el3_isa_remove(struct device *pdev,
 				    unsigned int ndev)
 {
 	el3_device_remove(pdev);
 	dev_set_drvdata(pdev, NULL);
-	return 0;
 }
 
 #ifdef CONFIG_PM
@@ -392,7 +386,7 @@ static struct isa_driver el3_isa_driver = {
 static int isa_registered;
 
 #ifdef CONFIG_PNP
-static struct pnp_device_id el3_pnp_ids[] = {
+static const struct pnp_device_id el3_pnp_ids[] = {
 	{ .id = "TCM5090" }, /* 3Com Etherlink III (TP) */
 	{ .id = "TCM5091" }, /* 3Com Etherlink III */
 	{ .id = "TCM5094" }, /* 3Com Etherlink III (combo) */
@@ -426,7 +420,6 @@ static int el3_pnp_probe(struct pnp_dev *pdev, const struct pnp_device_id *id)
 		return -ENOMEM;
 	}
 	SET_NETDEV_DEV(dev, &pdev->dev);
-	netdev_boot_setup_check(dev);
 
 	el3_dev_fill(dev, phys_addr, ioaddr, irq, if_port, EL3_PNP);
 	pnp_set_drvdata(pdev, dev);
@@ -474,7 +467,7 @@ static int pnp_registered;
 #endif /* CONFIG_PNP */
 
 #ifdef CONFIG_EISA
-static struct eisa_device_id el3_eisa_ids[] = {
+static const struct eisa_device_id el3_eisa_ids[] = {
 		{ "TCM5090" },
 		{ "TCM5091" },
 		{ "TCM5092" },
@@ -519,7 +512,9 @@ static int el3_common_init(struct net_device *dev)
 {
 	struct el3_private *lp = netdev_priv(dev);
 	int err;
-	const char *if_names[] = {"10baseT", "AUI", "undefined", "BNC"};
+	static const char * const if_names[] = {
+		"10baseT", "AUI", "undefined", "BNC"
+	};
 
 	spin_lock_init(&lp->lock);
 
@@ -547,8 +542,6 @@ static int el3_common_init(struct net_device *dev)
 	       dev->name, dev->base_addr, if_names[(dev->if_port & 0x03)],
 	       dev->dev_addr, dev->irq);
 
-	if (el3_debug > 0)
-		pr_info("%s", version);
 	return 0;
 
 }
@@ -595,7 +588,6 @@ static int el3_eisa_probe(struct device *device)
 	}
 
 	SET_NETDEV_DEV(dev, device);
-	netdev_boot_setup_check(dev);
 
 	el3_dev_fill(dev, phys_addr, ioaddr, irq, if_port, EL3_EISA);
 	eisa_set_drvdata (edev, dev);
@@ -689,7 +681,7 @@ el3_open(struct net_device *dev)
 }
 
 static void
-el3_tx_timeout (struct net_device *dev)
+el3_tx_timeout (struct net_device *dev, unsigned int txqueue)
 {
 	int ioaddr = dev->base_addr;
 
@@ -1039,7 +1031,7 @@ el3_link_ok(struct net_device *dev)
 	return tmp & (1<<11);
 }
 
-static int
+static void
 el3_netdev_get_ecmd(struct net_device *dev, struct ethtool_link_ksettings *cmd)
 {
 	u16 tmp;
@@ -1058,6 +1050,7 @@ el3_netdev_get_ecmd(struct net_device *dev, struct ethtool_link_ksettings *cmd)
 		break;
 	case 3:
 		cmd->base.port = PORT_BNC;
+		break;
 	default:
 		break;
 	}
@@ -1082,7 +1075,6 @@ el3_netdev_get_ecmd(struct net_device *dev, struct ethtool_link_ksettings *cmd)
 						supported);
 	cmd->base.speed = SPEED_10;
 	EL3WINDOW(1);
-	return 0;
 }
 
 static int
@@ -1143,20 +1135,18 @@ el3_netdev_set_ecmd(struct net_device *dev,
 
 static void el3_get_drvinfo(struct net_device *dev, struct ethtool_drvinfo *info)
 {
-	strlcpy(info->driver, DRV_NAME, sizeof(info->driver));
-	strlcpy(info->version, DRV_VERSION, sizeof(info->version));
+	strscpy(info->driver, DRV_NAME, sizeof(info->driver));
 }
 
 static int el3_get_link_ksettings(struct net_device *dev,
 				  struct ethtool_link_ksettings *cmd)
 {
 	struct el3_private *lp = netdev_priv(dev);
-	int ret;
 
 	spin_lock_irq(&lp->lock);
-	ret = el3_netdev_get_ecmd(dev, cmd);
+	el3_netdev_get_ecmd(dev, cmd);
 	spin_unlock_irq(&lp->lock);
-	return ret;
+	return 0;
 }
 
 static int el3_set_link_ksettings(struct net_device *dev,
@@ -1268,12 +1258,14 @@ el3_up(struct net_device *dev)
 					pr_cont("Forcing 3c5x9b full-duplex mode");
 					break;
 				}
+				fallthrough;
 			case 8:
 				/* set full-duplex mode based on eeprom config setting */
 				if ((sw_info & 0x000f) && (sw_info & 0x8000)) {
 					pr_cont("Setting 3c5x9b full-duplex mode (from EEPROM configuration bit)");
 					break;
 				}
+				fallthrough;
 			default:
 				/* xcvr=(0 || 4) OR user has an old 3c5x9 non "B" model */
 				pr_cont("Setting 3c5x9/3c5x9B half-duplex mode");
diff --git a/drivers/net/ethernet/3com/3c515.c b/drivers/net/ethernet/3com/3c515.c
index e7b1fa56b290..ecdea58e6a21 100644
--- a/drivers/net/ethernet/3com/3c515.c
+++ b/drivers/net/ethernet/3com/3c515.c
@@ -23,11 +23,6 @@
 */
 
 #define DRV_NAME		"3c515"
-#define DRV_VERSION		"0.99t-ac"
-#define DRV_RELDATE		"28-Oct-2002"
-
-static char *version =
-DRV_NAME ".c:v" DRV_VERSION " " DRV_RELDATE " becker@scyld.com and others\n";
 
 #define CORKSCREW 1
 
@@ -36,9 +31,6 @@ DRV_NAME ".c:v" DRV_VERSION " " DRV_RELDATE " becker@scyld.com and others\n";
    Setting to > 1512 effectively disables this feature. */
 static int rx_copybreak = 200;
 
-/* Allow setting MTU to a larger size, bypassing the normal ethernet setup. */
-static const int mtu = 1500;
-
 /* Maximum events (Rx packets, etc.) to handle at each interrupt. */
 static int max_interrupt_work = 20;
 
@@ -71,8 +63,10 @@ static int max_interrupt_work = 20;
 #include <linux/timer.h>
 #include <linux/ethtool.h>
 #include <linux/bitops.h>
-
 #include <linux/uaccess.h>
+
+#include <net/Space.h>
+
 #include <asm/io.h>
 #include <asm/dma.h>
 
@@ -84,7 +78,6 @@ static int max_interrupt_work = 20;
 MODULE_AUTHOR("Donald Becker <becker@scyld.com>");
 MODULE_DESCRIPTION("3Com 3c515 Corkscrew driver");
 MODULE_LICENSE("GPL");
-MODULE_VERSION(DRV_VERSION);
 
 /* "Knobs" for adjusting internal parameters. */
 /* Put out somewhat more debugging messages. (0 - no msg, 1 minimal msgs). */
@@ -367,11 +360,11 @@ static struct net_device *corkscrew_scan(int unit);
 static int corkscrew_setup(struct net_device *dev, int ioaddr,
 			    struct pnp_dev *idev, int card_number);
 static int corkscrew_open(struct net_device *dev);
-static void corkscrew_timer(unsigned long arg);
+static void corkscrew_timer(struct timer_list *t);
 static netdev_tx_t corkscrew_start_xmit(struct sk_buff *skb,
 					struct net_device *dev);
 static int corkscrew_rx(struct net_device *dev);
-static void corkscrew_timeout(struct net_device *dev);
+static void corkscrew_timeout(struct net_device *dev, unsigned int txqueue);
 static int boomerang_rx(struct net_device *dev);
 static irqreturn_t corkscrew_interrupt(int irq, void *dev_id);
 static int corkscrew_close(struct net_device *dev);
@@ -413,32 +406,25 @@ MODULE_PARM_DESC(max_interrupt_work, "3c515 maximum events handled per interrupt
 /* we will need locking (and refcounting) if we ever use it for more */
 static LIST_HEAD(root_corkscrew_dev);
 
-int init_module(void)
+static int corkscrew_init_module(void)
 {
 	int found = 0;
 	if (debug >= 0)
 		corkscrew_debug = debug;
-	if (corkscrew_debug)
-		pr_debug("%s", version);
 	while (corkscrew_scan(-1))
 		found++;
 	return found ? 0 : -ENODEV;
 }
+module_init(corkscrew_init_module);
 
 #else
 struct net_device *tc515_probe(int unit)
 {
 	struct net_device *dev = corkscrew_scan(unit);
-	static int printed;
 
 	if (!dev)
 		return ERR_PTR(-ENODEV);
 
-	if (corkscrew_debug > 0 && !printed) {
-		printed = 1;
-		pr_debug("%s", version);
-	}
-
 	return dev;
 }
 #endif				/* not MODULE */
@@ -580,6 +566,7 @@ static int corkscrew_setup(struct net_device *dev, int ioaddr,
 {
 	struct corkscrew_private *vp = netdev_priv(dev);
 	unsigned int eeprom[0x40], checksum = 0;	/* EEPROM contents */
+	__be16 addr[ETH_ALEN / 2];
 	int i;
 	int irq;
 
@@ -627,12 +614,11 @@ static int corkscrew_setup(struct net_device *dev, int ioaddr,
 
 	spin_lock_init(&vp->lock);
 
-	setup_timer(&vp->timer, corkscrew_timer, (unsigned long) dev);
+	timer_setup(&vp->timer, corkscrew_timer, 0);
 
 	/* Read the station address from the EEPROM. */
 	EL3WINDOW(0);
 	for (i = 0; i < 0x18; i++) {
-		__be16 *phys_addr = (__be16 *) dev->dev_addr;
 		int timer;
 		outw(EEPROM_Read + i, ioaddr + Wn0EepromCmd);
 		/* Pause for at least 162 us. for the read to take place. */
@@ -644,8 +630,9 @@ static int corkscrew_setup(struct net_device *dev, int ioaddr,
 		eeprom[i] = inw(ioaddr + Wn0EepromData);
 		checksum ^= eeprom[i];
 		if (i < 3)
-			phys_addr[i] = htons(eeprom[i]);
+			addr[i] = htons(eeprom[i]);
 	}
+	eth_hw_addr_set(dev, (u8 *)addr);
 	checksum = (checksum ^ (checksum >> 8)) & 0xff;
 	if (checksum != 0x00)
 		pr_cont(" ***INVALID CHECKSUM %4.4x*** ", checksum);
@@ -869,11 +856,11 @@ static int corkscrew_open(struct net_device *dev)
 	return 0;
 }
 
-static void corkscrew_timer(unsigned long data)
+static void corkscrew_timer(struct timer_list *t)
 {
 #ifdef AUTOMEDIA
-	struct net_device *dev = (struct net_device *) data;
-	struct corkscrew_private *vp = netdev_priv(dev);
+	struct corkscrew_private *vp = timer_container_of(vp, t, timer);
+	struct net_device *dev = vp->our_dev;
 	int ioaddr = dev->base_addr;
 	unsigned long flags;
 	int ok = 0;
@@ -961,7 +948,7 @@ static void corkscrew_timer(unsigned long data)
 #endif				/* AUTOMEDIA */
 }
 
-static void corkscrew_timeout(struct net_device *dev)
+static void corkscrew_timeout(struct net_device *dev, unsigned int txqueue)
 {
 	int i;
 	struct corkscrew_private *vp = netdev_priv(dev);
@@ -1063,7 +1050,7 @@ static netdev_tx_t corkscrew_start_xmit(struct sk_buff *skb,
 #ifdef VORTEX_BUS_MASTER
 	if (vp->bus_master) {
 		/* Set the bus-master controller to transfer the packet. */
-		outl((int) (skb->data), ioaddr + Wn7_MasterAddr);
+		outl(isa_virt_to_bus(skb->data), ioaddr + Wn7_MasterAddr);
 		outw((skb->len + 3) & ~3, ioaddr + Wn7_MasterLen);
 		vp->tx_skb = skb;
 		outw(StartDMADown, ioaddr + EL3_CMD);
@@ -1177,7 +1164,7 @@ static irqreturn_t corkscrew_interrupt(int irq, void *dev_id)
 				if (inl(ioaddr + DownListPtr) == isa_virt_to_bus(&lp->tx_ring[entry]))
 					break;	/* It still hasn't been processed. */
 				if (lp->tx_skbuff[entry]) {
-					dev_kfree_skb_irq(lp->tx_skbuff[entry]);
+					dev_consume_skb_irq(lp->tx_skbuff[entry]);
 					lp->tx_skbuff[entry] = NULL;
 				}
 				dirty_tx++;
@@ -1192,7 +1179,7 @@ static irqreturn_t corkscrew_interrupt(int irq, void *dev_id)
 #ifdef VORTEX_BUS_MASTER
 		if (status & DMADone) {
 			outw(0x1000, ioaddr + Wn7_MasterStatus);	/* Ack the event. */
-			dev_kfree_skb_irq(lp->tx_skb);	/* Release the transferred buffer */
+			dev_consume_skb_irq(lp->tx_skb);	/* Release the transferred buffer */
 			netif_wake_queue(dev);
 		}
 #endif
@@ -1370,9 +1357,9 @@ static int boomerang_rx(struct net_device *dev)
 			    (skb = netdev_alloc_skb(dev, pkt_len + 4)) != NULL) {
 				skb_reserve(skb, 2);	/* Align IP on 16 byte boundaries */
 				/* 'skb_put()' points to the start of sk_buff data area. */
-				memcpy(skb_put(skb, pkt_len),
-				       isa_bus_to_virt(vp->rx_ring[entry].
-						   addr), pkt_len);
+				skb_put_data(skb,
+					     isa_bus_to_virt(vp->rx_ring[entry].addr),
+					     pkt_len);
 				rx_copy++;
 			} else {
 				void *temp;
@@ -1427,7 +1414,7 @@ static int corkscrew_close(struct net_device *dev)
 			dev->name, rx_nocopy, rx_copy, queued_packet);
 	}
 
-	del_timer_sync(&vp->timer);
+	timer_delete_sync(&vp->timer);
 
 	/* Turn off statistics ASAP.  We update lp->stats below. */
 	outw(StatsDisable, ioaddr + EL3_CMD);
@@ -1521,7 +1508,7 @@ static void update_stats(int ioaddr, struct net_device *dev)
 static void set_rx_mode(struct net_device *dev)
 {
 	int ioaddr = dev->base_addr;
-	short new_mode;
+	unsigned short new_mode;
 
 	if (dev->flags & IFF_PROMISC) {
 		if (corkscrew_debug > 3)
@@ -1539,8 +1526,7 @@ static void set_rx_mode(struct net_device *dev)
 static void netdev_get_drvinfo(struct net_device *dev,
 			       struct ethtool_drvinfo *info)
 {
-	strlcpy(info->driver, DRV_NAME, sizeof(info->driver));
-	strlcpy(info->version, DRV_VERSION, sizeof(info->version));
+	strscpy(info->driver, DRV_NAME, sizeof(info->driver));
 	snprintf(info->bus_info, sizeof(info->bus_info), "ISA 0x%lx",
 		 dev->base_addr);
 }
diff --git a/drivers/net/ethernet/3com/3c574_cs.c b/drivers/net/ethernet/3com/3c574_cs.c
index 47c844cc9d27..1f2070497a75 100644
--- a/drivers/net/ethernet/3com/3c574_cs.c
+++ b/drivers/net/ethernet/3com/3c574_cs.c
@@ -225,7 +225,7 @@ static unsigned short read_eeprom(unsigned int ioaddr, int index);
 static void tc574_wait_for_completion(struct net_device *dev, int cmd);
 
 static void tc574_reset(struct net_device *dev);
-static void media_check(unsigned long arg);
+static void media_check(struct timer_list *t);
 static int el3_open(struct net_device *dev);
 static netdev_tx_t el3_start_xmit(struct sk_buff *skb,
 					struct net_device *dev);
@@ -234,7 +234,7 @@ static void update_stats(struct net_device *dev);
 static struct net_device_stats *el3_get_stats(struct net_device *dev);
 static int el3_rx(struct net_device *dev, int worklimit);
 static int el3_close(struct net_device *dev);
-static void el3_tx_timeout(struct net_device *dev);
+static void el3_tx_timeout(struct net_device *dev, unsigned int txqueue);
 static int el3_ioctl(struct net_device *dev, struct ifreq *rq, int cmd);
 static void set_rx_mode(struct net_device *dev);
 static void set_multicast_list(struct net_device *dev);
@@ -252,7 +252,7 @@ static const struct net_device_ops el3_netdev_ops = {
 	.ndo_start_xmit		= el3_start_xmit,
 	.ndo_tx_timeout 	= el3_tx_timeout,
 	.ndo_get_stats		= el3_get_stats,
-	.ndo_do_ioctl		= el3_ioctl,
+	.ndo_eth_ioctl		= el3_ioctl,
 	.ndo_set_rx_mode	= set_multicast_list,
 	.ndo_set_mac_address 	= eth_mac_addr,
 	.ndo_validate_addr	= eth_validate_addr,
@@ -305,15 +305,13 @@ static int tc574_config(struct pcmcia_device *link)
 	struct net_device *dev = link->priv;
 	struct el3_private *lp = netdev_priv(dev);
 	int ret, i, j;
+	__be16 addr[ETH_ALEN / 2];
 	unsigned int ioaddr;
-	__be16 *phys_addr;
 	char *cardname;
 	__u32 config;
 	u8 *buf;
 	size_t len;
 
-	phys_addr = (__be16 *)dev->dev_addr;
-
 	dev_dbg(&link->dev, "3c574_config()\n");
 
 	link->io_lines = 16;
@@ -347,19 +345,20 @@ static int tc574_config(struct pcmcia_device *link)
 	len = pcmcia_get_tuple(link, 0x88, &buf);
 	if (buf && len >= 6) {
 		for (i = 0; i < 3; i++)
-			phys_addr[i] = htons(le16_to_cpu(buf[i * 2]));
+			addr[i] = htons(le16_to_cpu(buf[i * 2]));
 		kfree(buf);
 	} else {
 		kfree(buf); /* 0 < len < 6 */
 		EL3WINDOW(0);
 		for (i = 0; i < 3; i++)
-			phys_addr[i] = htons(read_eeprom(ioaddr, i + 10));
-		if (phys_addr[0] == htons(0x6060)) {
+			addr[i] = htons(read_eeprom(ioaddr, i + 10));
+		if (addr[0] == htons(0x6060)) {
 			pr_notice("IO port conflict at 0x%03lx-0x%03lx\n",
 				  dev->base_addr, dev->base_addr+15);
 			goto failed;
 		}
 	}
+	eth_hw_addr_set(dev, (u8 *)addr);
 	if (link->prod_id[1])
 		cardname = link->prod_id[1];
 	else
@@ -377,7 +376,7 @@ static int tc574_config(struct pcmcia_device *link)
 		lp->autoselect = config & Autoselect ? 1 : 0;
 	}
 
-	init_timer(&lp->media);
+	timer_setup(&lp->media, media_check, 0);
 
 	{
 		int phy;
@@ -681,8 +680,6 @@ static int el3_open(struct net_device *dev)
 	netif_start_queue(dev);
 	
 	tc574_reset(dev);
-	lp->media.function = media_check;
-	lp->media.data = (unsigned long) dev;
 	lp->media.expires = jiffies + HZ;
 	add_timer(&lp->media);
 	
@@ -692,7 +689,7 @@ static int el3_open(struct net_device *dev)
 	return 0;
 }
 
-static void el3_tx_timeout(struct net_device *dev)
+static void el3_tx_timeout(struct net_device *dev, unsigned int txqueue)
 {
 	unsigned int ioaddr = dev->base_addr;
 	
@@ -859,10 +856,10 @@ static irqreturn_t el3_interrupt(int irq, void *dev_id)
 	(and as a last resort, poll the NIC for events), and to monitor
 	the MII, reporting changes in cable status.
 */
-static void media_check(unsigned long arg)
+static void media_check(struct timer_list *t)
 {
-	struct net_device *dev = (struct net_device *) arg;
-	struct el3_private *lp = netdev_priv(dev);
+	struct el3_private *lp = timer_container_of(lp, t, media);
+	struct net_device *dev = lp->p_dev->priv;
 	unsigned int ioaddr = dev->base_addr;
 	unsigned long flags;
 	unsigned short /* cable, */ media, partner;
@@ -953,7 +950,7 @@ static struct net_device_stats *el3_get_stats(struct net_device *dev)
 static void update_stats(struct net_device *dev)
 {
 	unsigned int ioaddr = dev->base_addr;
-	u8 rx, tx, up;
+	u8 up;
 
 	pr_debug("%s: updating the statistics.\n", dev->name);
 
@@ -974,8 +971,8 @@ static void update_stats(struct net_device *dev)
 	dev->stats.tx_packets			+= (up&0x30) << 4;
 	/* Rx packets   */			   inb(ioaddr + 7);
 	/* Tx deferrals */			   inb(ioaddr + 8);
-	rx		 			 = inw(ioaddr + 10);
-	tx					 = inw(ioaddr + 12);
+	/* rx */				   inw(ioaddr + 10);
+	/* tx */				   inw(ioaddr + 12);
 
 	EL3WINDOW(4);
 	/* BadSSD */				   inb(ioaddr + 12);
@@ -1048,6 +1045,7 @@ static int el3_ioctl(struct net_device *dev, struct ifreq *rq, int cmd)
 	switch(cmd) {
 	case SIOCGMIIPHY:		/* Get the address of the PHY in use. */
 		data->phy_id = phy;
+		fallthrough;
 	case SIOCGMIIREG:		/* Read the specified MII register. */
 		{
 			int saved_window;
@@ -1142,7 +1140,7 @@ static int el3_close(struct net_device *dev)
 
 	link->open--;
 	netif_stop_queue(dev);
-	del_timer_sync(&lp->media);
+	timer_delete_sync(&lp->media);
 
 	return 0;
 }
diff --git a/drivers/net/ethernet/3com/3c589_cs.c b/drivers/net/ethernet/3com/3c589_cs.c
index e28254a00599..ea49be43b8c3 100644
--- a/drivers/net/ethernet/3com/3c589_cs.c
+++ b/drivers/net/ethernet/3com/3c589_cs.c
@@ -23,7 +23,6 @@
 #define pr_fmt(fmt) KBUILD_MODNAME ": " fmt
 
 #define DRV_NAME	"3c589_cs"
-#define DRV_VERSION	"1.162-ac"
 
 #include <linux/module.h>
 #include <linux/kernel.h>
@@ -163,7 +162,7 @@ static void tc589_release(struct pcmcia_device *link);
 
 static u16 read_eeprom(unsigned int ioaddr, int index);
 static void tc589_reset(struct net_device *dev);
-static void media_check(unsigned long arg);
+static void media_check(struct timer_list *t);
 static int el3_config(struct net_device *dev, struct ifmap *map);
 static int el3_open(struct net_device *dev);
 static netdev_tx_t el3_start_xmit(struct sk_buff *skb,
@@ -173,7 +172,7 @@ static void update_stats(struct net_device *dev);
 static struct net_device_stats *el3_get_stats(struct net_device *dev);
 static int el3_rx(struct net_device *dev);
 static int el3_close(struct net_device *dev);
-static void el3_tx_timeout(struct net_device *dev);
+static void el3_tx_timeout(struct net_device *dev, unsigned int txqueue);
 static void set_rx_mode(struct net_device *dev);
 static void set_multicast_list(struct net_device *dev);
 static const struct ethtool_ops netdev_ethtool_ops;
@@ -196,6 +195,7 @@ static int tc589_probe(struct pcmcia_device *link)
 {
 	struct el3_private *lp;
 	struct net_device *dev;
+	int ret;
 
 	dev_dbg(&link->dev, "3c589_attach()\n");
 
@@ -219,7 +219,15 @@ static int tc589_probe(struct pcmcia_device *link)
 
 	dev->ethtool_ops = &netdev_ethtool_ops;
 
-	return tc589_config(link);
+	ret = tc589_config(link);
+	if (ret)
+		goto err_free_netdev;
+
+	return 0;
+
+err_free_netdev:
+	free_netdev(dev);
+	return ret;
 }
 
 static void tc589_detach(struct pcmcia_device *link)
@@ -238,8 +246,8 @@ static void tc589_detach(struct pcmcia_device *link)
 static int tc589_config(struct pcmcia_device *link)
 {
 	struct net_device *dev = link->priv;
-	__be16 *phys_addr;
 	int ret, i, j, multi = 0, fifo;
+	__be16 addr[ETH_ALEN / 2];
 	unsigned int ioaddr;
 	static const char * const ram_split[] = {"5:3", "3:1", "1:1", "3:5"};
 	u8 *buf;
@@ -247,7 +255,6 @@ static int tc589_config(struct pcmcia_device *link)
 
 	dev_dbg(&link->dev, "3c589_config\n");
 
-	phys_addr = (__be16 *)dev->dev_addr;
 	/* Is this a 3c562? */
 	if (link->manf_id != MANFID_3COM)
 		dev_info(&link->dev, "hmmm, is this really a 3Com card??\n");
@@ -286,18 +293,19 @@ static int tc589_config(struct pcmcia_device *link)
 	len = pcmcia_get_tuple(link, 0x88, &buf);
 	if (buf && len >= 6) {
 		for (i = 0; i < 3; i++)
-			phys_addr[i] = htons(le16_to_cpu(buf[i*2]));
+			addr[i] = htons(le16_to_cpu(buf[i*2]));
 		kfree(buf);
 	} else {
 		kfree(buf); /* 0 < len < 6 */
 		for (i = 0; i < 3; i++)
-			phys_addr[i] = htons(read_eeprom(ioaddr, i));
-		if (phys_addr[0] == htons(0x6060)) {
+			addr[i] = htons(read_eeprom(ioaddr, i));
+		if (addr[0] == htons(0x6060)) {
 			dev_err(&link->dev, "IO port conflict at 0x%03lx-0x%03lx\n",
 					dev->base_addr, dev->base_addr+15);
 			goto failed;
 		}
 	}
+	eth_hw_addr_set(dev, (u8 *)addr);
 
 	/* The address and resource configuration register aren't loaded from
 	 * the EEPROM and *must* be set to 0 and IRQ3 for the PCMCIA version.
@@ -481,8 +489,7 @@ static void tc589_reset(struct net_device *dev)
 static void netdev_get_drvinfo(struct net_device *dev,
 			       struct ethtool_drvinfo *info)
 {
-	strlcpy(info->driver, DRV_NAME, sizeof(info->driver));
-	strlcpy(info->version, DRV_VERSION, sizeof(info->version));
+	strscpy(info->driver, DRV_NAME, sizeof(info->driver));
 	snprintf(info->bus_info, sizeof(info->bus_info),
 		"PCMCIA 0x%lx", dev->base_addr);
 }
@@ -495,7 +502,7 @@ static int el3_config(struct net_device *dev, struct ifmap *map)
 {
 	if ((map->port != (u_char)(-1)) && (map->port != dev->if_port)) {
 		if (map->port <= 3) {
-			dev->if_port = map->port;
+			WRITE_ONCE(dev->if_port, map->port);
 			netdev_info(dev, "switched to %s port\n", if_names[dev->if_port]);
 			tc589_set_xcvr(dev, dev->if_port);
 		} else {
@@ -517,7 +524,7 @@ static int el3_open(struct net_device *dev)
 	netif_start_queue(dev);
 
 	tc589_reset(dev);
-	setup_timer(&lp->media, media_check, (unsigned long)dev);
+	timer_setup(&lp->media, media_check, 0);
 	mod_timer(&lp->media, jiffies + HZ);
 
 	dev_dbg(&link->dev, "%s: opened, status %4.4x.\n",
@@ -526,7 +533,7 @@ static int el3_open(struct net_device *dev)
 	return 0;
 }
 
-static void el3_tx_timeout(struct net_device *dev)
+static void el3_tx_timeout(struct net_device *dev, unsigned int txqueue)
 {
 	unsigned int ioaddr = dev->base_addr;
 
@@ -676,10 +683,10 @@ static irqreturn_t el3_interrupt(int irq, void *dev_id)
 	return IRQ_RETVAL(handled);
 }
 
-static void media_check(unsigned long arg)
+static void media_check(struct timer_list *t)
 {
-	struct net_device *dev = (struct net_device *)(arg);
-	struct el3_private *lp = netdev_priv(dev);
+	struct el3_private *lp = timer_container_of(lp, t, media);
+	struct net_device *dev = lp->p_dev->priv;
 	unsigned int ioaddr = dev->base_addr;
 	u16 media, errs;
 	unsigned long flags;
@@ -939,7 +946,7 @@ static int el3_close(struct net_device *dev)
 
 	link->open--;
 	netif_stop_queue(dev);
-	del_timer_sync(&lp->media);
+	timer_delete_sync(&lp->media);
 
 	return 0;
 }
diff --git a/drivers/net/ethernet/3com/3c59x.c b/drivers/net/ethernet/3com/3c59x.c
index e41245a54f8b..8c9cc97efd4e 100644
--- a/drivers/net/ethernet/3com/3c59x.c
+++ b/drivers/net/ethernet/3com/3c59x.c
@@ -602,7 +602,7 @@ struct vortex_private {
 	struct sk_buff* rx_skbuff[RX_RING_SIZE];
 	struct sk_buff* tx_skbuff[TX_RING_SIZE];
 	unsigned int cur_rx, cur_tx;		/* The next free ring entry */
-	unsigned int dirty_rx, dirty_tx;	/* The ring entries to be free()ed. */
+	unsigned int dirty_tx;	/* The ring entries to be free()ed. */
 	struct vortex_extra_stats xstats;	/* NIC-specific extra stats */
 	struct sk_buff *tx_skb;				/* Packet being eaten by bus master ctrl.  */
 	dma_addr_t tx_skb_dma;				/* Allocated DMA address for bus master ctrl DMA.   */
@@ -618,7 +618,6 @@ struct vortex_private {
 
 	/* The remainder are related to chip state, mostly media selection. */
 	struct timer_list timer;			/* Media selection timer. */
-	struct timer_list rx_oom_timer;		/* Rx skb allocation retry timer */
 	int options;						/* User-settable misc. driver options. */
 	unsigned int media_override:4, 		/* Passed-in media type. */
 		default_media:4,				/* Read from the EEPROM/Wn3_Config. */
@@ -759,16 +758,16 @@ static int vortex_open(struct net_device *dev);
 static void mdio_sync(struct vortex_private *vp, int bits);
 static int mdio_read(struct net_device *dev, int phy_id, int location);
 static void mdio_write(struct net_device *vp, int phy_id, int location, int value);
-static void vortex_timer(unsigned long arg);
-static void rx_oom_timer(unsigned long arg);
+static void vortex_timer(struct timer_list *t);
 static netdev_tx_t vortex_start_xmit(struct sk_buff *skb,
 				     struct net_device *dev);
 static netdev_tx_t boomerang_start_xmit(struct sk_buff *skb,
 					struct net_device *dev);
 static int vortex_rx(struct net_device *dev);
 static int boomerang_rx(struct net_device *dev);
-static irqreturn_t vortex_interrupt(int irq, void *dev_id);
-static irqreturn_t boomerang_interrupt(int irq, void *dev_id);
+static irqreturn_t vortex_boomerang_interrupt(int irq, void *dev_id);
+static irqreturn_t _vortex_interrupt(int irq, struct net_device *dev);
+static irqreturn_t _boomerang_interrupt(int irq, struct net_device *dev);
 static int vortex_close(struct net_device *dev);
 static void dump_tx_ring(struct net_device *dev);
 static void update_stats(void __iomem *ioaddr, struct net_device *dev);
@@ -777,7 +776,7 @@ static void set_rx_mode(struct net_device *dev);
 #ifdef CONFIG_PCI
 static int vortex_ioctl(struct net_device *dev, struct ifreq *rq, int cmd);
 #endif
-static void vortex_tx_timeout(struct net_device *dev);
+static void vortex_tx_timeout(struct net_device *dev, unsigned int txqueue);
 static void acpi_set_WOL(struct net_device *dev);
 static const struct ethtool_ops vortex_ethtool_ops;
 static void set_8021q_mode(struct net_device *dev, int enable);
@@ -840,11 +839,7 @@ MODULE_PARM_DESC(use_mmio, "3c59x: use memory-mapped PCI I/O resource (0-1)");
 #ifdef CONFIG_NET_POLL_CONTROLLER
 static void poll_vortex(struct net_device *dev)
 {
-	struct vortex_private *vp = netdev_priv(dev);
-	unsigned long flags;
-	local_irq_save(flags);
-	(vp->full_bus_master_rx ? boomerang_interrupt:vortex_interrupt)(dev->irq,dev);
-	local_irq_restore(flags);
+	vortex_boomerang_interrupt(dev->irq, dev);
 }
 #endif
 
@@ -852,8 +847,7 @@ static void poll_vortex(struct net_device *dev)
 
 static int vortex_suspend(struct device *dev)
 {
-	struct pci_dev *pdev = to_pci_dev(dev);
-	struct net_device *ndev = pci_get_drvdata(pdev);
+	struct net_device *ndev = dev_get_drvdata(dev);
 
 	if (!ndev || !netif_running(ndev))
 		return 0;
@@ -866,8 +860,7 @@ static int vortex_suspend(struct device *dev)
 
 static int vortex_resume(struct device *dev)
 {
-	struct pci_dev *pdev = to_pci_dev(dev);
-	struct net_device *ndev = pci_get_drvdata(pdev);
+	struct net_device *ndev = dev_get_drvdata(dev);
 	int err;
 
 	if (!ndev || !netif_running(ndev))
@@ -900,7 +893,7 @@ static const struct dev_pm_ops vortex_pm_ops = {
 #endif /* !CONFIG_PM */
 
 #ifdef CONFIG_EISA
-static struct eisa_device_id vortex_eisa_ids[] = {
+static const struct eisa_device_id vortex_eisa_ids[] = {
 	{ "TCM5920", CH_3C592 },
 	{ "TCM5970", CH_3C597 },
 	{ "" }
@@ -1059,7 +1052,7 @@ static const struct net_device_ops boomrang_netdev_ops = {
 	.ndo_tx_timeout		= vortex_tx_timeout,
 	.ndo_get_stats		= vortex_get_stats,
 #ifdef CONFIG_PCI
-	.ndo_do_ioctl 		= vortex_ioctl,
+	.ndo_eth_ioctl		= vortex_ioctl,
 #endif
 	.ndo_set_rx_mode	= set_rx_mode,
 	.ndo_set_mac_address 	= eth_mac_addr,
@@ -1076,7 +1069,7 @@ static const struct net_device_ops vortex_netdev_ops = {
 	.ndo_tx_timeout		= vortex_tx_timeout,
 	.ndo_get_stats		= vortex_get_stats,
 #ifdef CONFIG_PCI
-	.ndo_do_ioctl 		= vortex_ioctl,
+	.ndo_eth_ioctl		= vortex_ioctl,
 #endif
 	.ndo_set_rx_mode	= set_rx_mode,
 	.ndo_set_mac_address 	= eth_mac_addr,
@@ -1098,6 +1091,7 @@ static int vortex_probe1(struct device *gendev, void __iomem *ioaddr, int irq,
 	struct vortex_private *vp;
 	int option;
 	unsigned int eeprom[0x40], checksum = 0;		/* EEPROM contents */
+	__be16 addr[ETH_ALEN / 2];
 	int i, step;
 	struct net_device *dev;
 	static int printed_version;
@@ -1156,7 +1150,7 @@ static int vortex_probe1(struct device *gendev, void __iomem *ioaddr, int irq,
 
 	print_info = (vortex_debug > 1);
 	if (print_info)
-		pr_info("See Documentation/networking/vortex.txt\n");
+		pr_info("See Documentation/networking/device_drivers/ethernet/3com/vortex.rst\n");
 
 	pr_info("%s: 3Com %s %s at %p.\n",
 	       print_name,
@@ -1214,9 +1208,9 @@ static int vortex_probe1(struct device *gendev, void __iomem *ioaddr, int irq,
 	vp->mii.reg_num_mask = 0x1f;
 
 	/* Makes sure rings are at least 16 byte aligned. */
-	vp->rx_ring = pci_alloc_consistent(pdev, sizeof(struct boom_rx_desc) * RX_RING_SIZE
+	vp->rx_ring = dma_alloc_coherent(gendev, sizeof(struct boom_rx_desc) * RX_RING_SIZE
 					   + sizeof(struct boom_tx_desc) * TX_RING_SIZE,
-					   &vp->rx_ring_dma);
+					   &vp->rx_ring_dma, GFP_KERNEL);
 	retval = -ENOMEM;
 	if (!vp->rx_ring)
 		goto free_device;
@@ -1291,7 +1285,8 @@ static int vortex_probe1(struct device *gendev, void __iomem *ioaddr, int irq,
 	if ((checksum != 0x00) && !(vci->drv_flags & IS_TORNADO))
 		pr_cont(" ***INVALID CHECKSUM %4.4x*** ", checksum);
 	for (i = 0; i < 3; i++)
-		((__be16 *)dev->dev_addr)[i] = htons(eeprom[i + 10]);
+		addr[i] = htons(eeprom[i + 10]);
+	eth_hw_addr_set(dev, (u8 *)addr);
 	if (print_info)
 		pr_cont(" %pM", dev->dev_addr);
 	/* Unfortunately an all zero eeprom passes the checksum and this
@@ -1307,7 +1302,7 @@ static int vortex_probe1(struct device *gendev, void __iomem *ioaddr, int irq,
 	if (print_info)
 		pr_cont(", IRQ %d\n", dev->irq);
 	/* Tell them about an invalid IRQ. */
-	if (dev->irq <= 0 || dev->irq >= nr_irqs)
+	if (dev->irq <= 0 || dev->irq >= irq_get_nr_irqs())
 		pr_warn(" *** Warning: IRQ %d is unlikely to work! ***\n",
 			dev->irq);
 
@@ -1471,18 +1466,17 @@ static int vortex_probe1(struct device *gendev, void __iomem *ioaddr, int irq,
 	if (pdev) {
 		vp->pm_state_valid = 1;
 		pci_save_state(pdev);
- 		acpi_set_WOL(dev);
+		acpi_set_WOL(dev);
 	}
 	retval = register_netdev(dev);
 	if (retval == 0)
 		return 0;
 
 free_ring:
-	pci_free_consistent(pdev,
-						sizeof(struct boom_rx_desc) * RX_RING_SIZE
-							+ sizeof(struct boom_tx_desc) * TX_RING_SIZE,
-						vp->rx_ring,
-						vp->rx_ring_dma);
+	dma_free_coherent(&pdev->dev,
+		sizeof(struct boom_rx_desc) * RX_RING_SIZE +
+		sizeof(struct boom_tx_desc) * TX_RING_SIZE,
+		vp->rx_ring, vp->rx_ring_dma);
 free_device:
 	free_netdev(dev);
 	pr_err(PFX "vortex_probe1 fails.  Returns %d\n", retval);
@@ -1556,7 +1550,7 @@ vortex_up(struct net_device *dev)
 	struct vortex_private *vp = netdev_priv(dev);
 	void __iomem *ioaddr = vp->ioaddr;
 	unsigned int config;
-	int i, mii_reg1, mii_reg5, err = 0;
+	int i, mii_reg5, err = 0;
 
 	if (VORTEX_PCI(vp)) {
 		pci_set_power_state(VORTEX_PCI(vp), PCI_D0);	/* Go active */
@@ -1599,9 +1593,8 @@ vortex_up(struct net_device *dev)
 				dev->name, media_tbl[dev->if_port].name);
 	}
 
-	setup_timer(&vp->timer, vortex_timer, (unsigned long)dev);
+	timer_setup(&vp->timer, vortex_timer, 0);
 	mod_timer(&vp->timer, RUN_AT(media_tbl[dev->if_port].wait));
-	setup_timer(&vp->rx_oom_timer, rx_oom_timer, (unsigned long)dev);
 
 	if (vortex_debug > 1)
 		pr_debug("%s: Initial media type %s.\n",
@@ -1614,7 +1607,7 @@ vortex_up(struct net_device *dev)
 	window_write32(vp, config, 3, Wn3_Config);
 
 	if (dev->if_port == XCVR_MII || dev->if_port == XCVR_NWAY) {
-		mii_reg1 = mdio_read(dev, vp->phys[0], MII_BMSR);
+		mdio_read(dev, vp->phys[0], MII_BMSR);
 		mii_reg5 = mdio_read(dev, vp->phys[0], MII_LPA);
 		vp->partner_flow_ctrl = ((mii_reg5 & 0x0400) != 0);
 		vp->mii.full_duplex = vp->full_duplex;
@@ -1676,7 +1669,7 @@ vortex_up(struct net_device *dev)
 	window_write16(vp, 0x0040, 4, Wn4_NetDiag);
 
 	if (vp->full_bus_master_rx) { /* Boomerang bus master. */
-		vp->cur_rx = vp->dirty_rx = 0;
+		vp->cur_rx = 0;
 		/* Initialize the RxEarly register as recommended. */
 		iowrite16(SetRxThreshold + (1536>>2), ioaddr + EL3_CMD);
 		iowrite32(0x0020, ioaddr + PktStatus);
@@ -1729,10 +1722,10 @@ vortex_open(struct net_device *dev)
 	struct vortex_private *vp = netdev_priv(dev);
 	int i;
 	int retval;
+	dma_addr_t dma;
 
 	/* Use the now-standard shared IRQ implementation. */
-	if ((retval = request_irq(dev->irq, vp->full_bus_master_rx ?
-				boomerang_interrupt : vortex_interrupt, IRQF_SHARED, dev->name, dev))) {
+	if ((retval = request_irq(dev->irq, vortex_boomerang_interrupt, IRQF_SHARED, dev->name, dev))) {
 		pr_err("%s: Could not reserve IRQ %d\n", dev->name, dev->irq);
 		goto err;
 	}
@@ -1753,7 +1746,11 @@ vortex_open(struct net_device *dev)
 				break;			/* Bad news!  */
 
 			skb_reserve(skb, NET_IP_ALIGN);	/* Align IP on 16 byte boundaries */
-			vp->rx_ring[i].addr = cpu_to_le32(pci_map_single(VORTEX_PCI(vp), skb->data, PKT_BUF_SZ, PCI_DMA_FROMDEVICE));
+			dma = dma_map_single(vp->gendev, skb->data,
+					     PKT_BUF_SZ, DMA_FROM_DEVICE);
+			if (dma_mapping_error(vp->gendev, dma))
+				break;
+			vp->rx_ring[i].addr = cpu_to_le32(dma);
 		}
 		if (i != RX_RING_SIZE) {
 			pr_emerg("%s: no memory for rx ring\n", dev->name);
@@ -1784,10 +1781,10 @@ out:
 }
 
 static void
-vortex_timer(unsigned long data)
+vortex_timer(struct timer_list *t)
 {
-	struct net_device *dev = (struct net_device *)data;
-	struct vortex_private *vp = netdev_priv(dev);
+	struct vortex_private *vp = timer_container_of(vp, t, timer);
+	struct net_device *dev = vp->mii.dev;
 	void __iomem *ioaddr = vp->ioaddr;
 	int next_tick = 60*HZ;
 	int ok = 0;
@@ -1882,7 +1879,7 @@ leave_media_alone:
 		iowrite16(FakeIntr, ioaddr + EL3_CMD);
 }
 
-static void vortex_tx_timeout(struct net_device *dev)
+static void vortex_tx_timeout(struct net_device *dev, unsigned int txqueue)
 {
 	struct vortex_private *vp = netdev_priv(dev);
 	void __iomem *ioaddr = vp->ioaddr;
@@ -1903,18 +1900,7 @@ static void vortex_tx_timeout(struct net_device *dev)
 		pr_err("%s: Interrupt posted but not delivered --"
 			   " IRQ blocked by another device?\n", dev->name);
 		/* Bad idea here.. but we might as well handle a few events. */
-		{
-			/*
-			 * Block interrupts because vortex_interrupt does a bare spin_lock()
-			 */
-			unsigned long flags;
-			local_irq_save(flags);
-			if (vp->full_bus_master_tx)
-				boomerang_interrupt(dev->irq, dev);
-			else
-				vortex_interrupt(dev->irq, dev);
-			local_irq_restore(flags);
-		}
+		vortex_boomerang_interrupt(dev->irq, dev);
 	}
 
 	if (vortex_debug > 0)
@@ -1970,7 +1956,7 @@ vortex_error(struct net_device *dev, int status)
 				   dev->name, tx_status);
 			if (tx_status == 0x82) {
 				pr_err("Probably a duplex mismatch.  See "
-						"Documentation/networking/vortex.txt\n");
+						"Documentation/networking/device_drivers/ethernet/3com/vortex.rst\n");
 			}
 			dump_tx_ring(dev);
 		}
@@ -2065,8 +2051,14 @@ vortex_start_xmit(struct sk_buff *skb, struct net_device *dev)
 	if (vp->bus_master) {
 		/* Set the bus-master controller to transfer the packet. */
 		int len = (skb->len + 3) & ~3;
-		vp->tx_skb_dma = pci_map_single(VORTEX_PCI(vp), skb->data, len,
-						PCI_DMA_TODEVICE);
+		vp->tx_skb_dma = dma_map_single(vp->gendev, skb->data, len,
+						DMA_TO_DEVICE);
+		if (dma_mapping_error(vp->gendev, vp->tx_skb_dma)) {
+			dev_kfree_skb_any(skb);
+			dev->stats.tx_dropped++;
+			return NETDEV_TX_OK;
+		}
+
 		spin_lock_irq(&vp->window_lock);
 		window_set(vp, 7);
 		iowrite32(vp->tx_skb_dma, ioaddr + Wn7_MasterAddr);
@@ -2160,9 +2152,9 @@ boomerang_start_xmit(struct sk_buff *skb, struct net_device *dev)
 			vp->tx_ring[entry].status = cpu_to_le32(skb->len | TxIntrUploaded | AddTCPChksum | AddUDPChksum);
 
 	if (!skb_shinfo(skb)->nr_frags) {
-		dma_addr = pci_map_single(VORTEX_PCI(vp), skb->data, skb->len,
-					  PCI_DMA_TODEVICE);
-		if (dma_mapping_error(&VORTEX_PCI(vp)->dev, dma_addr))
+		dma_addr = dma_map_single(vp->gendev, skb->data, skb->len,
+					  DMA_TO_DEVICE);
+		if (dma_mapping_error(vp->gendev, dma_addr))
 			goto out_dma_err;
 
 		vp->tx_ring[entry].frag[0].addr = cpu_to_le32(dma_addr);
@@ -2170,9 +2162,9 @@ boomerang_start_xmit(struct sk_buff *skb, struct net_device *dev)
 	} else {
 		int i;
 
-		dma_addr = pci_map_single(VORTEX_PCI(vp), skb->data,
-					  skb_headlen(skb), PCI_DMA_TODEVICE);
-		if (dma_mapping_error(&VORTEX_PCI(vp)->dev, dma_addr))
+		dma_addr = dma_map_single(vp->gendev, skb->data,
+					  skb_headlen(skb), DMA_TO_DEVICE);
+		if (dma_mapping_error(vp->gendev, dma_addr))
 			goto out_dma_err;
 
 		vp->tx_ring[entry].frag[0].addr = cpu_to_le32(dma_addr);
@@ -2181,21 +2173,21 @@ boomerang_start_xmit(struct sk_buff *skb, struct net_device *dev)
 		for (i = 0; i < skb_shinfo(skb)->nr_frags; i++) {
 			skb_frag_t *frag = &skb_shinfo(skb)->frags[i];
 
-			dma_addr = skb_frag_dma_map(&VORTEX_PCI(vp)->dev, frag,
+			dma_addr = skb_frag_dma_map(vp->gendev, frag,
 						    0,
-						    frag->size,
+						    skb_frag_size(frag),
 						    DMA_TO_DEVICE);
-			if (dma_mapping_error(&VORTEX_PCI(vp)->dev, dma_addr)) {
+			if (dma_mapping_error(vp->gendev, dma_addr)) {
 				for(i = i-1; i >= 0; i--)
-					dma_unmap_page(&VORTEX_PCI(vp)->dev,
+					dma_unmap_page(vp->gendev,
 						       le32_to_cpu(vp->tx_ring[entry].frag[i+1].addr),
 						       le32_to_cpu(vp->tx_ring[entry].frag[i+1].length),
 						       DMA_TO_DEVICE);
 
-				pci_unmap_single(VORTEX_PCI(vp),
+				dma_unmap_single(vp->gendev,
 						 le32_to_cpu(vp->tx_ring[entry].frag[0].addr),
 						 le32_to_cpu(vp->tx_ring[entry].frag[0].length),
-						 PCI_DMA_TODEVICE);
+						 DMA_TO_DEVICE);
 
 				goto out_dma_err;
 			}
@@ -2210,8 +2202,8 @@ boomerang_start_xmit(struct sk_buff *skb, struct net_device *dev)
 		}
 	}
 #else
-	dma_addr = pci_map_single(VORTEX_PCI(vp), skb->data, skb->len, PCI_DMA_TODEVICE);
-	if (dma_mapping_error(&VORTEX_PCI(vp)->dev, dma_addr))
+	dma_addr = dma_map_single(vp->gendev, skb->data, skb->len, DMA_TO_DEVICE);
+	if (dma_mapping_error(vp->gendev, dma_addr))
 		goto out_dma_err;
 	vp->tx_ring[entry].addr = cpu_to_le32(dma_addr);
 	vp->tx_ring[entry].length = cpu_to_le32(skb->len | LAST_FRAG);
@@ -2246,7 +2238,7 @@ boomerang_start_xmit(struct sk_buff *skb, struct net_device *dev)
 out:
 	return NETDEV_TX_OK;
 out_dma_err:
-	dev_err(&VORTEX_PCI(vp)->dev, "Error mapping dma buffer\n");
+	dev_err(vp->gendev, "Error mapping dma buffer\n");
 	goto out;
 }
 
@@ -2259,9 +2251,8 @@ out_dma_err:
  */
 
 static irqreturn_t
-vortex_interrupt(int irq, void *dev_id)
+_vortex_interrupt(int irq, struct net_device *dev)
 {
-	struct net_device *dev = dev_id;
 	struct vortex_private *vp = netdev_priv(dev);
 	void __iomem *ioaddr;
 	int status;
@@ -2270,7 +2261,6 @@ vortex_interrupt(int irq, void *dev_id)
 	unsigned int bytes_compl = 0, pkts_compl = 0;
 
 	ioaddr = vp->ioaddr;
-	spin_lock(&vp->lock);
 
 	status = ioread16(ioaddr + EL3_STATUS);
 
@@ -2314,10 +2304,10 @@ vortex_interrupt(int irq, void *dev_id)
 		if (status & DMADone) {
 			if (ioread16(ioaddr + Wn7_MasterStatus) & 0x1000) {
 				iowrite16(0x1000, ioaddr + Wn7_MasterStatus); /* Ack the event. */
-				pci_unmap_single(VORTEX_PCI(vp), vp->tx_skb_dma, (vp->tx_skb->len + 3) & ~3, PCI_DMA_TODEVICE);
+				dma_unmap_single(vp->gendev, vp->tx_skb_dma, (vp->tx_skb->len + 3) & ~3, DMA_TO_DEVICE);
 				pkts_compl++;
 				bytes_compl += vp->tx_skb->len;
-				dev_kfree_skb_irq(vp->tx_skb); /* Release the transferred buffer */
+				dev_consume_skb_irq(vp->tx_skb); /* Release the transferred buffer */
 				if (ioread16(ioaddr + TxFree) > 1536) {
 					/*
 					 * AKPM: FIXME: I don't think we need this.  If the queue was stopped due to
@@ -2368,7 +2358,6 @@ vortex_interrupt(int irq, void *dev_id)
 		pr_debug("%s: exiting interrupt, status %4.4x.\n",
 			   dev->name, status);
 handler_exit:
-	spin_unlock(&vp->lock);
 	return IRQ_RETVAL(handled);
 }
 
@@ -2378,9 +2367,8 @@ handler_exit:
  */
 
 static irqreturn_t
-boomerang_interrupt(int irq, void *dev_id)
+_boomerang_interrupt(int irq, struct net_device *dev)
 {
-	struct net_device *dev = dev_id;
 	struct vortex_private *vp = netdev_priv(dev);
 	void __iomem *ioaddr;
 	int status;
@@ -2390,12 +2378,6 @@ boomerang_interrupt(int irq, void *dev_id)
 
 	ioaddr = vp->ioaddr;
 
-
-	/*
-	 * It seems dopey to put the spinlock this early, but we could race against vortex_tx_timeout
-	 * and boomerang_start_xmit
-	 */
-	spin_lock(&vp->lock);
 	vp->handling_irq = 1;
 
 	status = ioread16(ioaddr + EL3_STATUS);
@@ -2451,23 +2433,23 @@ boomerang_interrupt(int irq, void *dev_id)
 					struct sk_buff *skb = vp->tx_skbuff[entry];
 #if DO_ZEROCOPY
 					int i;
-					pci_unmap_single(VORTEX_PCI(vp),
+					dma_unmap_single(vp->gendev,
 							le32_to_cpu(vp->tx_ring[entry].frag[0].addr),
 							le32_to_cpu(vp->tx_ring[entry].frag[0].length)&0xFFF,
-							PCI_DMA_TODEVICE);
+							DMA_TO_DEVICE);
 
 					for (i=1; i<=skb_shinfo(skb)->nr_frags; i++)
-							pci_unmap_page(VORTEX_PCI(vp),
+							dma_unmap_page(vp->gendev,
 											 le32_to_cpu(vp->tx_ring[entry].frag[i].addr),
 											 le32_to_cpu(vp->tx_ring[entry].frag[i].length)&0xFFF,
-											 PCI_DMA_TODEVICE);
+											 DMA_TO_DEVICE);
 #else
-					pci_unmap_single(VORTEX_PCI(vp),
-						le32_to_cpu(vp->tx_ring[entry].addr), skb->len, PCI_DMA_TODEVICE);
+					dma_unmap_single(vp->gendev,
+						le32_to_cpu(vp->tx_ring[entry].addr), skb->len, DMA_TO_DEVICE);
 #endif
 					pkts_compl++;
 					bytes_compl += skb->len;
-					dev_kfree_skb_irq(skb);
+					dev_consume_skb_irq(skb);
 					vp->tx_skbuff[entry] = NULL;
 				} else {
 					pr_debug("boomerang_interrupt: no skb!\n");
@@ -2514,10 +2496,29 @@ boomerang_interrupt(int irq, void *dev_id)
 			   dev->name, status);
 handler_exit:
 	vp->handling_irq = 0;
-	spin_unlock(&vp->lock);
 	return IRQ_RETVAL(handled);
 }
 
+static irqreturn_t
+vortex_boomerang_interrupt(int irq, void *dev_id)
+{
+	struct net_device *dev = dev_id;
+	struct vortex_private *vp = netdev_priv(dev);
+	unsigned long flags;
+	irqreturn_t ret;
+
+	spin_lock_irqsave(&vp->lock, flags);
+
+	if (vp->full_bus_master_rx)
+		ret = _boomerang_interrupt(dev->irq, dev);
+	else
+		ret = _vortex_interrupt(dev->irq, dev);
+
+	spin_unlock_irqrestore(&vp->lock, flags);
+
+	return ret;
+}
+
 static int vortex_rx(struct net_device *dev)
 {
 	struct vortex_private *vp = netdev_priv(dev);
@@ -2553,14 +2554,14 @@ static int vortex_rx(struct net_device *dev)
 				/* 'skb_put()' points to the start of sk_buff data area. */
 				if (vp->bus_master &&
 					! (ioread16(ioaddr + Wn7_MasterStatus) & 0x8000)) {
-					dma_addr_t dma = pci_map_single(VORTEX_PCI(vp), skb_put(skb, pkt_len),
-									   pkt_len, PCI_DMA_FROMDEVICE);
+					dma_addr_t dma = dma_map_single(vp->gendev, skb_put(skb, pkt_len),
+									   pkt_len, DMA_FROM_DEVICE);
 					iowrite32(dma, ioaddr + Wn7_MasterAddr);
 					iowrite16((skb->len + 3) & ~3, ioaddr + Wn7_MasterLen);
 					iowrite16(StartDMAUp, ioaddr + EL3_CMD);
 					while (ioread16(ioaddr + Wn7_MasterStatus) & 0x8000)
 						;
-					pci_unmap_single(VORTEX_PCI(vp), dma, pkt_len, PCI_DMA_FROMDEVICE);
+					dma_unmap_single(vp->gendev, dma, pkt_len, DMA_FROM_DEVICE);
 				} else {
 					ioread32_rep(ioaddr + RX_FIFO,
 					             skb_put(skb, pkt_len),
@@ -2593,7 +2594,7 @@ boomerang_rx(struct net_device *dev)
 	int entry = vp->cur_rx % RX_RING_SIZE;
 	void __iomem *ioaddr = vp->ioaddr;
 	int rx_status;
-	int rx_work_limit = vp->dirty_rx + RX_RING_SIZE - vp->cur_rx;
+	int rx_work_limit = RX_RING_SIZE;
 
 	if (vortex_debug > 5)
 		pr_debug("boomerang_rx(): status %4.4x\n", ioread16(ioaddr+EL3_STATUS));
@@ -2614,7 +2615,8 @@ boomerang_rx(struct net_device *dev)
 		} else {
 			/* The packet length: up to 4.5K!. */
 			int pkt_len = rx_status & 0x1fff;
-			struct sk_buff *skb;
+			struct sk_buff *skb, *newskb;
+			dma_addr_t newdma;
 			dma_addr_t dma = le32_to_cpu(vp->rx_ring[entry].addr);
 
 			if (vortex_debug > 4)
@@ -2626,19 +2628,36 @@ boomerang_rx(struct net_device *dev)
 			if (pkt_len < rx_copybreak &&
 			    (skb = netdev_alloc_skb(dev, pkt_len + 2)) != NULL) {
 				skb_reserve(skb, 2);	/* Align IP on 16 byte boundaries */
-				pci_dma_sync_single_for_cpu(VORTEX_PCI(vp), dma, PKT_BUF_SZ, PCI_DMA_FROMDEVICE);
+				dma_sync_single_for_cpu(vp->gendev, dma, PKT_BUF_SZ, DMA_FROM_DEVICE);
 				/* 'skb_put()' points to the start of sk_buff data area. */
-				memcpy(skb_put(skb, pkt_len),
-					   vp->rx_skbuff[entry]->data,
-					   pkt_len);
-				pci_dma_sync_single_for_device(VORTEX_PCI(vp), dma, PKT_BUF_SZ, PCI_DMA_FROMDEVICE);
+				skb_put_data(skb, vp->rx_skbuff[entry]->data,
+					     pkt_len);
+				dma_sync_single_for_device(vp->gendev, dma, PKT_BUF_SZ, DMA_FROM_DEVICE);
 				vp->rx_copy++;
 			} else {
+				/* Pre-allocate the replacement skb.  If it or its
+				 * mapping fails then recycle the buffer thats already
+				 * in place
+				 */
+				newskb = netdev_alloc_skb_ip_align(dev, PKT_BUF_SZ);
+				if (!newskb) {
+					dev->stats.rx_dropped++;
+					goto clear_complete;
+				}
+				newdma = dma_map_single(vp->gendev, newskb->data,
+							PKT_BUF_SZ, DMA_FROM_DEVICE);
+				if (dma_mapping_error(vp->gendev, newdma)) {
+					dev->stats.rx_dropped++;
+					consume_skb(newskb);
+					goto clear_complete;
+				}
+
 				/* Pass up the skbuff already on the Rx ring. */
 				skb = vp->rx_skbuff[entry];
-				vp->rx_skbuff[entry] = NULL;
+				vp->rx_skbuff[entry] = newskb;
+				vp->rx_ring[entry].addr = cpu_to_le32(newdma);
 				skb_put(skb, pkt_len);
-				pci_unmap_single(VORTEX_PCI(vp), dma, PKT_BUF_SZ, PCI_DMA_FROMDEVICE);
+				dma_unmap_single(vp->gendev, dma, PKT_BUF_SZ, DMA_FROM_DEVICE);
 				vp->rx_nocopy++;
 			}
 			skb->protocol = eth_type_trans(skb, dev);
@@ -2654,55 +2673,15 @@ boomerang_rx(struct net_device *dev)
 			netif_rx(skb);
 			dev->stats.rx_packets++;
 		}
-		entry = (++vp->cur_rx) % RX_RING_SIZE;
-	}
-	/* Refill the Rx ring buffers. */
-	for (; vp->cur_rx - vp->dirty_rx > 0; vp->dirty_rx++) {
-		struct sk_buff *skb;
-		entry = vp->dirty_rx % RX_RING_SIZE;
-		if (vp->rx_skbuff[entry] == NULL) {
-			skb = netdev_alloc_skb_ip_align(dev, PKT_BUF_SZ);
-			if (skb == NULL) {
-				static unsigned long last_jif;
-				if (time_after(jiffies, last_jif + 10 * HZ)) {
-					pr_warn("%s: memory shortage\n",
-						dev->name);
-					last_jif = jiffies;
-				}
-				if ((vp->cur_rx - vp->dirty_rx) == RX_RING_SIZE)
-					mod_timer(&vp->rx_oom_timer, RUN_AT(HZ * 1));
-				break;			/* Bad news!  */
-			}
 
-			vp->rx_ring[entry].addr = cpu_to_le32(pci_map_single(VORTEX_PCI(vp), skb->data, PKT_BUF_SZ, PCI_DMA_FROMDEVICE));
-			vp->rx_skbuff[entry] = skb;
-		}
+clear_complete:
 		vp->rx_ring[entry].status = 0;	/* Clear complete bit. */
 		iowrite16(UpUnstall, ioaddr + EL3_CMD);
+		entry = (++vp->cur_rx) % RX_RING_SIZE;
 	}
 	return 0;
 }
 
-/*
- * If we've hit a total OOM refilling the Rx ring we poll once a second
- * for some memory.  Otherwise there is no way to restart the rx process.
- */
-static void
-rx_oom_timer(unsigned long arg)
-{
-	struct net_device *dev = (struct net_device *)arg;
-	struct vortex_private *vp = netdev_priv(dev);
-
-	spin_lock_irq(&vp->lock);
-	if ((vp->cur_rx - vp->dirty_rx) == RX_RING_SIZE)	/* This test is redundant, but makes me feel good */
-		boomerang_rx(dev);
-	if (vortex_debug > 1) {
-		pr_debug("%s: rx_oom_timer %s\n", dev->name,
-			((vp->cur_rx - vp->dirty_rx) != RX_RING_SIZE) ? "succeeded" : "retrying");
-	}
-	spin_unlock_irq(&vp->lock);
-}
-
 static void
 vortex_down(struct net_device *dev, int final_down)
 {
@@ -2712,8 +2691,7 @@ vortex_down(struct net_device *dev, int final_down)
 	netdev_reset_queue(dev);
 	netif_stop_queue(dev);
 
-	del_timer_sync(&vp->rx_oom_timer);
-	del_timer_sync(&vp->timer);
+	timer_delete_sync(&vp->timer);
 
 	/* Turn off statistics ASAP.  We update dev->stats below. */
 	iowrite16(StatsDisable, ioaddr + EL3_CMD);
@@ -2776,8 +2754,8 @@ vortex_close(struct net_device *dev)
 	if (vp->full_bus_master_rx) { /* Free Boomerang bus master Rx buffers. */
 		for (i = 0; i < RX_RING_SIZE; i++)
 			if (vp->rx_skbuff[i]) {
-				pci_unmap_single(	VORTEX_PCI(vp), le32_to_cpu(vp->rx_ring[i].addr),
-									PKT_BUF_SZ, PCI_DMA_FROMDEVICE);
+				dma_unmap_single(vp->gendev, le32_to_cpu(vp->rx_ring[i].addr),
+									PKT_BUF_SZ, DMA_FROM_DEVICE);
 				dev_kfree_skb(vp->rx_skbuff[i]);
 				vp->rx_skbuff[i] = NULL;
 			}
@@ -2790,12 +2768,12 @@ vortex_close(struct net_device *dev)
 				int k;
 
 				for (k=0; k<=skb_shinfo(skb)->nr_frags; k++)
-						pci_unmap_single(VORTEX_PCI(vp),
+						dma_unmap_single(vp->gendev,
 										 le32_to_cpu(vp->tx_ring[i].frag[k].addr),
 										 le32_to_cpu(vp->tx_ring[i].frag[k].length)&0xFFF,
-										 PCI_DMA_TODEVICE);
+										 DMA_TO_DEVICE);
 #else
-				pci_unmap_single(VORTEX_PCI(vp), le32_to_cpu(vp->tx_ring[i].addr), skb->len, PCI_DMA_TODEVICE);
+				dma_unmap_single(vp->gendev, le32_to_cpu(vp->tx_ring[i].addr), skb->len, DMA_TO_DEVICE);
 #endif
 				dev_kfree_skb(skb);
 				vp->tx_skbuff[i] = NULL;
@@ -2810,7 +2788,7 @@ static void
 dump_tx_ring(struct net_device *dev)
 {
 	if (vortex_debug > 0) {
-	struct vortex_private *vp = netdev_priv(dev);
+		struct vortex_private *vp = netdev_priv(dev);
 		void __iomem *ioaddr = vp->ioaddr;
 
 		if (vp->full_bus_master_tx) {
@@ -2912,7 +2890,9 @@ static int vortex_get_link_ksettings(struct net_device *dev,
 {
 	struct vortex_private *vp = netdev_priv(dev);
 
-	return mii_ethtool_get_link_ksettings(&vp->mii, cmd);
+	mii_ethtool_get_link_ksettings(&vp->mii, cmd);
+
+	return 0;
 }
 
 static int vortex_set_link_ksettings(struct net_device *dev,
@@ -2979,13 +2959,13 @@ static void vortex_get_drvinfo(struct net_device *dev,
 {
 	struct vortex_private *vp = netdev_priv(dev);
 
-	strlcpy(info->driver, DRV_NAME, sizeof(info->driver));
+	strscpy(info->driver, DRV_NAME, sizeof(info->driver));
 	if (VORTEX_PCI(vp)) {
-		strlcpy(info->bus_info, pci_name(VORTEX_PCI(vp)),
+		strscpy(info->bus_info, pci_name(VORTEX_PCI(vp)),
 			sizeof(info->bus_info));
 	} else {
 		if (VORTEX_EISA(vp))
-			strlcpy(info->bus_info, dev_name(vp->gendev),
+			strscpy(info->bus_info, dev_name(vp->gendev),
 				sizeof(info->bus_info));
 		else
 			snprintf(info->bus_info, sizeof(info->bus_info),
@@ -3301,11 +3281,10 @@ static void vortex_remove_one(struct pci_dev *pdev)
 
 	pci_iounmap(pdev, vp->ioaddr);
 
-	pci_free_consistent(pdev,
-						sizeof(struct boom_rx_desc) * RX_RING_SIZE
-							+ sizeof(struct boom_tx_desc) * TX_RING_SIZE,
-						vp->rx_ring,
-						vp->rx_ring_dma);
+	dma_free_coherent(&pdev->dev,
+			sizeof(struct boom_rx_desc) * RX_RING_SIZE +
+			sizeof(struct boom_tx_desc) * TX_RING_SIZE,
+			vp->rx_ring, vp->rx_ring_dma);
 
 	pci_release_regions(pdev);
 
diff --git a/drivers/net/ethernet/3com/Kconfig b/drivers/net/ethernet/3com/Kconfig
index 5b7658bcf020..1fbab79e2be4 100644
--- a/drivers/net/ethernet/3com/Kconfig
+++ b/drivers/net/ethernet/3com/Kconfig
@@ -1,3 +1,4 @@
+# SPDX-License-Identifier: GPL-2.0-only
 #
 # 3Com Ethernet device configuration
 #
@@ -6,7 +7,7 @@ config NET_VENDOR_3COM
 	bool "3Com devices"
 	default y
 	depends on ISA || EISA || PCI || PCMCIA
-	---help---
+	help
 	  If you have a network (Ethernet) card belonging to this class, say Y.
 
 	  Note that the answer to this question doesn't directly affect the
@@ -19,7 +20,7 @@ if NET_VENDOR_3COM
 config EL3
 	tristate "3c509/3c579 \"EtherLink III\" support"
 	depends on (ISA || EISA)
-	---help---
+	help
 	  If you have a network (Ethernet) card belonging to the 3Com
 	  EtherLinkIII series, say Y here.
 
@@ -32,8 +33,9 @@ config EL3
 
 config 3C515
 	tristate "3c515 ISA \"Fast EtherLink\""
-	depends on ISA && ISA_DMA_API
-	---help---
+	depends on ISA && ISA_DMA_API && !PPC32
+	select NETDEV_LEGACY_INIT
+	help
 	  If you have a 3Com ISA EtherLink XL "Corkscrew" 3c515 Fast Ethernet
 	  network card, say Y here.
 
@@ -42,8 +44,8 @@ config 3C515
 
 config PCMCIA_3C574
 	tristate "3Com 3c574 PCMCIA support"
-	depends on PCMCIA
-	---help---
+	depends on PCMCIA && HAS_IOPORT
+	help
 	  Say Y here if you intend to attach a 3Com 3c574 or compatible PCMCIA
 	  (PC-card) Fast Ethernet card to your computer.
 
@@ -52,8 +54,8 @@ config PCMCIA_3C574
 
 config PCMCIA_3C589
 	tristate "3Com 3c589 PCMCIA support"
-	depends on PCMCIA
-	---help---
+	depends on PCMCIA && HAS_IOPORT
+	help
 	  Say Y here if you intend to attach a 3Com 3c589 or compatible PCMCIA
 	  (PC-card) Ethernet card to your computer.
 
@@ -64,7 +66,7 @@ config VORTEX
 	tristate "3c590/3c900 series (592/595/597) \"Vortex/Boomerang\" support"
 	depends on (PCI || EISA) && HAS_IOPORT_MAP
 	select MII
-	---help---
+	help
 	  This option enables driver support for a large number of 10Mbps and
 	  10/100Mbps EISA, PCI and Cardbus 3Com network cards:
 
@@ -75,8 +77,9 @@ config VORTEX
 	  "Hurricane" (3c555/3cSOHO)                           PCI
 
 	  If you have such a card, say Y here.  More specific information is in
-	  <file:Documentation/networking/vortex.txt> and in the comments at
-	  the beginning of <file:drivers/net/ethernet/3com/3c59x.c>.
+	  <file:Documentation/networking/device_drivers/ethernet/3com/vortex.rst>
+	  and in the comments at the beginning of
+	  <file:drivers/net/ethernet/3com/3c59x.c>.
 
 	  To compile this support as a module, choose M here.
 
@@ -84,7 +87,7 @@ config TYPHOON
 	tristate "3cr990 series \"Typhoon\" support"
 	depends on PCI
 	select CRC32
-	---help---
+	help
 	  This option enables driver support for the 3cr990 series of cards:
 
 	  3C990-TX, 3CR990-TX-95, 3CR990-TX-97, 3CR990-FX-95, 3CR990-FX-97,
diff --git a/drivers/net/ethernet/3com/Makefile b/drivers/net/ethernet/3com/Makefile
index 74046afab993..f8b73babc510 100644
--- a/drivers/net/ethernet/3com/Makefile
+++ b/drivers/net/ethernet/3com/Makefile
@@ -1,3 +1,4 @@
+# SPDX-License-Identifier: GPL-2.0
 #
 # Makefile for the 3Com Ethernet device drivers
 #
diff --git a/drivers/net/ethernet/3com/typhoon.c b/drivers/net/ethernet/3com/typhoon.c
index be823c186517..aaaff3ba43ef 100644
--- a/drivers/net/ethernet/3com/typhoon.c
+++ b/drivers/net/ethernet/3com/typhoon.c
@@ -127,7 +127,6 @@ static const int multicast_filter_limit = 32;
 #include "typhoon.h"
 
 MODULE_AUTHOR("David Dillow <dave@thedillows.org>");
-MODULE_VERSION("1.0");
 MODULE_LICENSE("GPL");
 MODULE_FIRMWARE(FIRMWARE_NAME);
 MODULE_DESCRIPTION("3Com Typhoon Family (3C990, 3CR990, and variants)");
@@ -139,11 +138,6 @@ MODULE_PARM_DESC(use_mmio, "Use MMIO (1) or PIO(0) to access the NIC. "
 module_param(rx_copybreak, int, 0);
 module_param(use_mmio, int, 0);
 
-#if defined(NETIF_F_TSO) && MAX_SKB_FRAGS > 32
-#warning Typhoon only supports 32 entries in its SG list for TSO, disabling TSO
-#undef NETIF_F_TSO
-#endif
-
 #if TXLO_ENTRIES <= (2 * MAX_SKB_FRAGS)
 #error TX ring too small!
 #endif
@@ -311,7 +305,7 @@ enum state_values {
  * cannot pass a read, so this forces current writes to post.
  */
 #define typhoon_post_pci_writes(x) \
-	do { if(likely(use_mmio)) ioread32(x+TYPHOON_REG_HEARTBEAT); } while(0)
+	do { if (likely(use_mmio)) ioread32(x+TYPHOON_REG_HEARTBEAT); } while (0)
 
 /* We'll wait up to six seconds for a reset, and half a second normally.
  */
@@ -381,7 +375,7 @@ typhoon_reset(void __iomem *ioaddr, int wait_type)
 	int i, err = 0;
 	int timeout;
 
-	if(wait_type == WaitNoSleep)
+	if (wait_type == WaitNoSleep)
 		timeout = TYPHOON_RESET_TIMEOUT_NOSLEEP;
 	else
 		timeout = TYPHOON_RESET_TIMEOUT_SLEEP;
@@ -394,13 +388,13 @@ typhoon_reset(void __iomem *ioaddr, int wait_type)
 	udelay(1);
 	iowrite32(TYPHOON_RESET_NONE, ioaddr + TYPHOON_REG_SOFT_RESET);
 
-	if(wait_type != NoWait) {
-		for(i = 0; i < timeout; i++) {
-			if(ioread32(ioaddr + TYPHOON_REG_STATUS) ==
+	if (wait_type != NoWait) {
+		for (i = 0; i < timeout; i++) {
+			if (ioread32(ioaddr + TYPHOON_REG_STATUS) ==
 			   TYPHOON_STATUS_WAITING_FOR_HOST)
 				goto out;
 
-			if(wait_type == WaitSleep)
+			if (wait_type == WaitSleep)
 				schedule_timeout_uninterruptible(1);
 			else
 				udelay(TYPHOON_UDELAY);
@@ -423,7 +417,7 @@ out:
 	 * which should be enough (I've see it work well at 100us, but still
 	 * saw occasional problems.)
 	 */
-	if(wait_type == WaitSleep)
+	if (wait_type == WaitSleep)
 		msleep(5);
 	else
 		udelay(500);
@@ -435,8 +429,8 @@ typhoon_wait_status(void __iomem *ioaddr, u32 wait_value)
 {
 	int i, err = 0;
 
-	for(i = 0; i < TYPHOON_WAIT_TIMEOUT; i++) {
-		if(ioread32(ioaddr + TYPHOON_REG_STATUS) == wait_value)
+	for (i = 0; i < TYPHOON_WAIT_TIMEOUT; i++) {
+		if (ioread32(ioaddr + TYPHOON_REG_STATUS) == wait_value)
 			goto out;
 		udelay(TYPHOON_UDELAY);
 	}
@@ -450,7 +444,7 @@ out:
 static inline void
 typhoon_media_status(struct net_device *dev, struct resp_desc *resp)
 {
-	if(resp->parm1 & TYPHOON_MEDIA_STAT_NO_LINK)
+	if (resp->parm1 & TYPHOON_MEDIA_STAT_NO_LINK)
 		netif_carrier_off(dev);
 	else
 		netif_carrier_on(dev);
@@ -466,7 +460,7 @@ typhoon_hello(struct typhoon *tp)
 	 * card in a long while. If the lock is held, then we're in the
 	 * process of issuing a command, so we don't need to respond.
 	 */
-	if(spin_trylock(&tp->command_lock)) {
+	if (spin_trylock(&tp->command_lock)) {
 		cmd = (struct cmd_desc *)(ring->ringBase + ring->lastWrite);
 		typhoon_inc_cmd_index(&ring->lastWrite, 1);
 
@@ -490,32 +484,32 @@ typhoon_process_response(struct typhoon *tp, int resp_size,
 
 	cleared = le32_to_cpu(indexes->respCleared);
 	ready = le32_to_cpu(indexes->respReady);
-	while(cleared != ready) {
+	while (cleared != ready) {
 		resp = (struct resp_desc *)(base + cleared);
 		count = resp->numDesc + 1;
-		if(resp_save && resp->seqNo) {
-			if(count > resp_size) {
+		if (resp_save && resp->seqNo) {
+			if (count > resp_size) {
 				resp_save->flags = TYPHOON_RESP_ERROR;
 				goto cleanup;
 			}
 
 			wrap_len = 0;
 			len = count * sizeof(*resp);
-			if(unlikely(cleared + len > RESPONSE_RING_SIZE)) {
+			if (unlikely(cleared + len > RESPONSE_RING_SIZE)) {
 				wrap_len = cleared + len - RESPONSE_RING_SIZE;
 				len = RESPONSE_RING_SIZE - cleared;
 			}
 
 			memcpy(resp_save, resp, len);
-			if(unlikely(wrap_len)) {
+			if (unlikely(wrap_len)) {
 				resp_save += len / sizeof(*resp);
 				memcpy(resp_save, base, wrap_len);
 			}
 
 			resp_save = NULL;
-		} else if(resp->cmd == TYPHOON_CMD_READ_MEDIA_STATUS) {
+		} else if (resp->cmd == TYPHOON_CMD_READ_MEDIA_STATUS) {
 			typhoon_media_status(tp->dev, resp);
-		} else if(resp->cmd == TYPHOON_CMD_HELLO_RESP) {
+		} else if (resp->cmd == TYPHOON_CMD_HELLO_RESP) {
 			typhoon_hello(tp);
 		} else {
 			netdev_err(tp->dev,
@@ -589,19 +583,19 @@ typhoon_issue_command(struct typhoon *tp, int num_cmd, struct cmd_desc *cmd,
 	freeCmd = typhoon_num_free_cmd(tp);
 	freeResp = typhoon_num_free_resp(tp);
 
-	if(freeCmd < num_cmd || freeResp < num_resp) {
+	if (freeCmd < num_cmd || freeResp < num_resp) {
 		netdev_err(tp->dev, "no descs for cmd, had (needed) %d (%d) cmd, %d (%d) resp\n",
 			   freeCmd, num_cmd, freeResp, num_resp);
 		err = -ENOMEM;
 		goto out;
 	}
 
-	if(cmd->flags & TYPHOON_CMD_RESPOND) {
+	if (cmd->flags & TYPHOON_CMD_RESPOND) {
 		/* If we're expecting a response, but the caller hasn't given
 		 * us a place to put it, we'll provide one.
 		 */
 		tp->awaiting_resp = 1;
-		if(resp == NULL) {
+		if (resp == NULL) {
 			resp = &local_resp;
 			num_resp = 1;
 		}
@@ -609,13 +603,13 @@ typhoon_issue_command(struct typhoon *tp, int num_cmd, struct cmd_desc *cmd,
 
 	wrap_len = 0;
 	len = num_cmd * sizeof(*cmd);
-	if(unlikely(ring->lastWrite + len > COMMAND_RING_SIZE)) {
+	if (unlikely(ring->lastWrite + len > COMMAND_RING_SIZE)) {
 		wrap_len = ring->lastWrite + len - COMMAND_RING_SIZE;
 		len = COMMAND_RING_SIZE - ring->lastWrite;
 	}
 
 	memcpy(ring->ringBase + ring->lastWrite, cmd, len);
-	if(unlikely(wrap_len)) {
+	if (unlikely(wrap_len)) {
 		struct cmd_desc *wrap_ptr = cmd;
 		wrap_ptr += len / sizeof(*cmd);
 		memcpy(ring->ringBase, wrap_ptr, wrap_len);
@@ -629,7 +623,7 @@ typhoon_issue_command(struct typhoon *tp, int num_cmd, struct cmd_desc *cmd,
 	iowrite32(ring->lastWrite, tp->ioaddr + TYPHOON_REG_CMD_READY);
 	typhoon_post_pci_writes(tp->ioaddr);
 
-	if((cmd->flags & TYPHOON_CMD_RESPOND) == 0)
+	if ((cmd->flags & TYPHOON_CMD_RESPOND) == 0)
 		goto out;
 
 	/* Ugh. We'll be here about 8ms, spinning our thumbs, unable to
@@ -649,14 +643,14 @@ typhoon_issue_command(struct typhoon *tp, int num_cmd, struct cmd_desc *cmd,
 	 * wait here.
 	 */
 	got_resp = 0;
-	for(i = 0; i < TYPHOON_WAIT_TIMEOUT && !got_resp; i++) {
-		if(indexes->respCleared != indexes->respReady)
+	for (i = 0; i < TYPHOON_WAIT_TIMEOUT && !got_resp; i++) {
+		if (indexes->respCleared != indexes->respReady)
 			got_resp = typhoon_process_response(tp, num_resp,
 								resp);
 		udelay(TYPHOON_UDELAY);
 	}
 
-	if(!got_resp) {
+	if (!got_resp) {
 		err = -ETIMEDOUT;
 		goto out;
 	}
@@ -664,11 +658,11 @@ typhoon_issue_command(struct typhoon *tp, int num_cmd, struct cmd_desc *cmd,
 	/* Collect the error response even if we don't care about the
 	 * rest of the response
 	 */
-	if(resp->flags & TYPHOON_RESP_ERROR)
+	if (resp->flags & TYPHOON_RESP_ERROR)
 		err = -EIO;
 
 out:
-	if(tp->awaiting_resp) {
+	if (tp->awaiting_resp) {
 		tp->awaiting_resp = 0;
 		smp_wmb();
 
@@ -679,7 +673,7 @@ out:
 		 * time. So, check for it, and interrupt ourselves if this
 		 * is the case.
 		 */
-		if(indexes->respCleared != indexes->respReady)
+		if (indexes->respCleared != indexes->respReady)
 			iowrite32(1, tp->ioaddr + TYPHOON_REG_SELF_INTERRUPT);
 	}
 
@@ -749,7 +743,7 @@ typhoon_start_tx(struct sk_buff *skb, struct net_device *dev)
 	 * between marking the queue awake and updating the cleared index.
 	 * Just loop and it will appear. This comes from the acenic driver.
 	 */
-	while(unlikely(typhoon_num_free_tx(txRing) < (numDesc + 2)))
+	while (unlikely(typhoon_num_free_tx(txRing) < (numDesc + 2)))
 		smp_rmb();
 
 	first_txd = (struct tx_desc *) (txRing->ringBase + txRing->lastWrite);
@@ -761,7 +755,7 @@ typhoon_start_tx(struct sk_buff *skb, struct net_device *dev)
 	first_txd->tx_addr = (u64)((unsigned long) skb);
 	first_txd->processFlags = 0;
 
-	if(skb->ip_summed == CHECKSUM_PARTIAL) {
+	if (skb->ip_summed == CHECKSUM_PARTIAL) {
 		/* The 3XP will figure out if this is UDP/TCP */
 		first_txd->processFlags |= TYPHOON_TX_PF_TCP_CHKSUM;
 		first_txd->processFlags |= TYPHOON_TX_PF_UDP_CHKSUM;
@@ -789,9 +783,9 @@ typhoon_start_tx(struct sk_buff *skb, struct net_device *dev)
 	/* No need to worry about padding packet -- the firmware pads
 	 * it with zeros to ETH_ZLEN for us.
 	 */
-	if(skb_shinfo(skb)->nr_frags == 0) {
-		skb_dma = pci_map_single(tp->tx_pdev, skb->data, skb->len,
-				       PCI_DMA_TODEVICE);
+	if (skb_shinfo(skb)->nr_frags == 0) {
+		skb_dma = dma_map_single(&tp->tx_pdev->dev, skb->data,
+					 skb->len, DMA_TO_DEVICE);
 		txd->flags = TYPHOON_FRAG_DESC | TYPHOON_DESC_VALID;
 		txd->len = cpu_to_le16(skb->len);
 		txd->frag.addr = cpu_to_le32(skb_dma);
@@ -801,8 +795,8 @@ typhoon_start_tx(struct sk_buff *skb, struct net_device *dev)
 		int i, len;
 
 		len = skb_headlen(skb);
-		skb_dma = pci_map_single(tp->tx_pdev, skb->data, len,
-				         PCI_DMA_TODEVICE);
+		skb_dma = dma_map_single(&tp->tx_pdev->dev, skb->data, len,
+					 DMA_TO_DEVICE);
 		txd->flags = TYPHOON_FRAG_DESC | TYPHOON_DESC_VALID;
 		txd->len = cpu_to_le16(len);
 		txd->frag.addr = cpu_to_le32(skb_dma);
@@ -819,8 +813,8 @@ typhoon_start_tx(struct sk_buff *skb, struct net_device *dev)
 
 			len = skb_frag_size(frag);
 			frag_addr = skb_frag_address(frag);
-			skb_dma = pci_map_single(tp->tx_pdev, frag_addr, len,
-					 PCI_DMA_TODEVICE);
+			skb_dma = dma_map_single(&tp->tx_pdev->dev, frag_addr,
+						 len, DMA_TO_DEVICE);
 			txd->flags = TYPHOON_FRAG_DESC | TYPHOON_DESC_VALID;
 			txd->len = cpu_to_le16(len);
 			txd->frag.addr = cpu_to_le32(skb_dma);
@@ -841,14 +835,14 @@ typhoon_start_tx(struct sk_buff *skb, struct net_device *dev)
 	 */
 	numDesc = MAX_SKB_FRAGS + TSO_NUM_DESCRIPTORS + 1;
 
-	if(typhoon_num_free_tx(txRing) < (numDesc + 2)) {
+	if (typhoon_num_free_tx(txRing) < (numDesc + 2)) {
 		netif_stop_queue(dev);
 
 		/* A Tx complete IRQ could have gotten between, making
 		 * the ring free again. Only need to recheck here, since
 		 * Tx is serialized.
 		 */
-		if(typhoon_num_free_tx(txRing) >= (numDesc + 2))
+		if (typhoon_num_free_tx(txRing) >= (numDesc + 2))
 			netif_wake_queue(dev);
 	}
 
@@ -864,7 +858,7 @@ typhoon_set_rx_mode(struct net_device *dev)
 	__le16 filter;
 
 	filter = TYPHOON_RX_FILTER_DIRECTED | TYPHOON_RX_FILTER_BROADCAST;
-	if(dev->flags & IFF_PROMISC) {
+	if (dev->flags & IFF_PROMISC) {
 		filter |= TYPHOON_RX_FILTER_PROMISCOUS;
 	} else if ((netdev_mc_count(dev) > multicast_filter_limit) ||
 		  (dev->flags & IFF_ALLMULTI)) {
@@ -906,7 +900,7 @@ typhoon_do_get_stats(struct typhoon *tp)
 
 	INIT_COMMAND_WITH_RESPONSE(&xp_cmd, TYPHOON_CMD_READ_STATS);
 	err = typhoon_issue_command(tp, 1, &xp_cmd, 7, xp_resp);
-	if(err < 0)
+	if (err < 0)
 		return err;
 
 	/* 3Com's Linux driver uses txMultipleCollisions as it's
@@ -954,10 +948,10 @@ typhoon_get_stats(struct net_device *dev)
 	struct net_device_stats *saved = &tp->stats_saved;
 
 	smp_rmb();
-	if(tp->card_state == Sleeping)
+	if (tp->card_state == Sleeping)
 		return saved;
 
-	if(typhoon_do_get_stats(tp) < 0) {
+	if (typhoon_do_get_stats(tp) < 0) {
 		netdev_err(dev, "error getting stats\n");
 		return saved;
 	}
@@ -974,13 +968,13 @@ typhoon_get_drvinfo(struct net_device *dev, struct ethtool_drvinfo *info)
 	struct resp_desc xp_resp[3];
 
 	smp_rmb();
-	if(tp->card_state == Sleeping) {
-		strlcpy(info->fw_version, "Sleep image",
+	if (tp->card_state == Sleeping) {
+		strscpy(info->fw_version, "Sleep image",
 			sizeof(info->fw_version));
 	} else {
 		INIT_COMMAND_WITH_RESPONSE(&xp_cmd, TYPHOON_CMD_READ_VERSIONS);
-		if(typhoon_issue_command(tp, 1, &xp_cmd, 3, xp_resp) < 0) {
-			strlcpy(info->fw_version, "Unknown runtime",
+		if (typhoon_issue_command(tp, 1, &xp_cmd, 3, xp_resp) < 0) {
+			strscpy(info->fw_version, "Unknown runtime",
 				sizeof(info->fw_version));
 		} else {
 			u32 sleep_ver = le32_to_cpu(xp_resp[0].parm2);
@@ -990,8 +984,8 @@ typhoon_get_drvinfo(struct net_device *dev, struct ethtool_drvinfo *info)
 		}
 	}
 
-	strlcpy(info->driver, KBUILD_MODNAME, sizeof(info->driver));
-	strlcpy(info->bus_info, pci_name(pci_dev), sizeof(info->bus_info));
+	strscpy(info->driver, KBUILD_MODNAME, sizeof(info->driver));
+	strscpy(info->bus_info, pci_name(pci_dev), sizeof(info->bus_info));
 }
 
 static int
@@ -1026,7 +1020,7 @@ typhoon_get_link_ksettings(struct net_device *dev,
 		break;
 	}
 
-	if(tp->capabilities & TYPHOON_FIBER) {
+	if (tp->capabilities & TYPHOON_FIBER) {
 		supported |= SUPPORTED_FIBRE;
 		advertising |= ADVERTISED_FIBRE;
 		cmd->base.port = PORT_FIBRE;
@@ -1043,7 +1037,7 @@ typhoon_get_link_ksettings(struct net_device *dev,
 	cmd->base.speed = tp->speed;
 	cmd->base.duplex = tp->duplex;
 	cmd->base.phy_address = 0;
-	if(tp->xcvr_select == TYPHOON_XCVR_AUTONEG)
+	if (tp->xcvr_select == TYPHOON_XCVR_AUTONEG)
 		cmd->base.autoneg = AUTONEG_ENABLE;
 	else
 		cmd->base.autoneg = AUTONEG_DISABLE;
@@ -1091,7 +1085,7 @@ typhoon_set_link_ksettings(struct net_device *dev,
 	INIT_COMMAND_NO_RESPONSE(&xp_cmd, TYPHOON_CMD_XCVR_SELECT);
 	xp_cmd.parm1 = xcvr;
 	err = typhoon_issue_command(tp, 1, &xp_cmd, 0, NULL);
-	if(err < 0)
+	if (err < 0)
 		goto out;
 
 	tp->xcvr_select = xcvr;
@@ -1114,9 +1108,9 @@ typhoon_get_wol(struct net_device *dev, struct ethtool_wolinfo *wol)
 
 	wol->supported = WAKE_PHY | WAKE_MAGIC;
 	wol->wolopts = 0;
-	if(tp->wol_events & TYPHOON_WAKE_LINK_EVENT)
+	if (tp->wol_events & TYPHOON_WAKE_LINK_EVENT)
 		wol->wolopts |= WAKE_PHY;
-	if(tp->wol_events & TYPHOON_WAKE_MAGIC_PKT)
+	if (tp->wol_events & TYPHOON_WAKE_MAGIC_PKT)
 		wol->wolopts |= WAKE_MAGIC;
 	memset(&wol->sopass, 0, sizeof(wol->sopass));
 }
@@ -1126,20 +1120,22 @@ typhoon_set_wol(struct net_device *dev, struct ethtool_wolinfo *wol)
 {
 	struct typhoon *tp = netdev_priv(dev);
 
-	if(wol->wolopts & ~(WAKE_PHY | WAKE_MAGIC))
+	if (wol->wolopts & ~(WAKE_PHY | WAKE_MAGIC))
 		return -EINVAL;
 
 	tp->wol_events = 0;
-	if(wol->wolopts & WAKE_PHY)
+	if (wol->wolopts & WAKE_PHY)
 		tp->wol_events |= TYPHOON_WAKE_LINK_EVENT;
-	if(wol->wolopts & WAKE_MAGIC)
+	if (wol->wolopts & WAKE_MAGIC)
 		tp->wol_events |= TYPHOON_WAKE_MAGIC_PKT;
 
 	return 0;
 }
 
 static void
-typhoon_get_ringparam(struct net_device *dev, struct ethtool_ringparam *ering)
+typhoon_get_ringparam(struct net_device *dev, struct ethtool_ringparam *ering,
+		      struct kernel_ethtool_ringparam *kernel_ering,
+		      struct netlink_ext_ack *extack)
 {
 	ering->rx_max_pending = RXENT_ENTRIES;
 	ering->tx_max_pending = TXLO_ENTRIES - 1;
@@ -1163,8 +1159,8 @@ typhoon_wait_interrupt(void __iomem *ioaddr)
 {
 	int i, err = 0;
 
-	for(i = 0; i < TYPHOON_WAIT_TIMEOUT; i++) {
-		if(ioread32(ioaddr + TYPHOON_REG_INTR_STATUS) &
+	for (i = 0; i < TYPHOON_WAIT_TIMEOUT; i++) {
+		if (ioread32(ioaddr + TYPHOON_REG_INTR_STATUS) &
 		   TYPHOON_INTR_BOOTCMD)
 			goto out;
 		udelay(TYPHOON_UDELAY);
@@ -1350,13 +1346,13 @@ typhoon_download_firmware(struct typhoon *tp)
 	image_data = typhoon_fw->data;
 	fHdr = (struct typhoon_file_header *) image_data;
 
-	/* Cannot just map the firmware image using pci_map_single() as
+	/* Cannot just map the firmware image using dma_map_single() as
 	 * the firmware is vmalloc()'d and may not be physically contiguous,
-	 * so we allocate some consistent memory to copy the sections into.
+	 * so we allocate some coherent memory to copy the sections into.
 	 */
 	err = -ENOMEM;
-	dpage = pci_alloc_consistent(pdev, PAGE_SIZE, &dpage_dma);
-	if(!dpage) {
+	dpage = dma_alloc_coherent(&pdev->dev, PAGE_SIZE, &dpage_dma, GFP_ATOMIC);
+	if (!dpage) {
 		netdev_err(tp->dev, "no DMA mem for firmware\n");
 		goto err_out;
 	}
@@ -1369,7 +1365,7 @@ typhoon_download_firmware(struct typhoon *tp)
 	       ioaddr + TYPHOON_REG_INTR_MASK);
 
 	err = -ETIMEDOUT;
-	if(typhoon_wait_status(ioaddr, TYPHOON_STATUS_WAITING_FOR_HOST) < 0) {
+	if (typhoon_wait_status(ioaddr, TYPHOON_STATUS_WAITING_FOR_HOST) < 0) {
 		netdev_err(tp->dev, "card ready timeout\n");
 		goto err_out_irq;
 	}
@@ -1398,16 +1394,16 @@ typhoon_download_firmware(struct typhoon *tp)
 	 * last write to the command register to post, so
 	 * we don't need a typhoon_post_pci_writes() after it.
 	 */
-	for(i = 0; i < numSections; i++) {
+	for (i = 0; i < numSections; i++) {
 		sHdr = (struct typhoon_section_header *) image_data;
 		image_data += sizeof(struct typhoon_section_header);
 		load_addr = le32_to_cpu(sHdr->startAddr);
 		section_len = le32_to_cpu(sHdr->len);
 
-		while(section_len) {
+		while (section_len) {
 			len = min_t(u32, section_len, PAGE_SIZE);
 
-			if(typhoon_wait_interrupt(ioaddr) < 0 ||
+			if (typhoon_wait_interrupt(ioaddr) < 0 ||
 			   ioread32(ioaddr + TYPHOON_REG_STATUS) !=
 			   TYPHOON_STATUS_WAITING_FOR_SEGMENT) {
 				netdev_err(tp->dev, "segment ready timeout\n");
@@ -1420,8 +1416,7 @@ typhoon_download_firmware(struct typhoon *tp)
 			 * the checksum, we can do this once, at the end.
 			 */
 			csum = csum_fold(csum_partial_copy_nocheck(image_data,
-								   dpage, len,
-								   0));
+								   dpage, len));
 
 			iowrite32(len, ioaddr + TYPHOON_REG_BOOT_LENGTH);
 			iowrite32(le16_to_cpu((__force __le16)csum),
@@ -1440,7 +1435,7 @@ typhoon_download_firmware(struct typhoon *tp)
 		}
 	}
 
-	if(typhoon_wait_interrupt(ioaddr) < 0 ||
+	if (typhoon_wait_interrupt(ioaddr) < 0 ||
 	   ioread32(ioaddr + TYPHOON_REG_STATUS) !=
 	   TYPHOON_STATUS_WAITING_FOR_SEGMENT) {
 		netdev_err(tp->dev, "final segment ready timeout\n");
@@ -1449,7 +1444,7 @@ typhoon_download_firmware(struct typhoon *tp)
 
 	iowrite32(TYPHOON_BOOTCMD_DNLD_COMPLETE, ioaddr + TYPHOON_REG_COMMAND);
 
-	if(typhoon_wait_status(ioaddr, TYPHOON_STATUS_WAITING_FOR_BOOT) < 0) {
+	if (typhoon_wait_status(ioaddr, TYPHOON_STATUS_WAITING_FOR_BOOT) < 0) {
 		netdev_err(tp->dev, "boot ready timeout, status 0x%0x\n",
 			   ioread32(ioaddr + TYPHOON_REG_STATUS));
 		goto err_out_irq;
@@ -1461,7 +1456,7 @@ err_out_irq:
 	iowrite32(irqMasked, ioaddr + TYPHOON_REG_INTR_MASK);
 	iowrite32(irqEnabled, ioaddr + TYPHOON_REG_INTR_ENABLE);
 
-	pci_free_consistent(pdev, PAGE_SIZE, dpage, dpage_dma);
+	dma_free_coherent(&pdev->dev, PAGE_SIZE, dpage, dpage_dma);
 
 err_out:
 	return err;
@@ -1472,7 +1467,7 @@ typhoon_boot_3XP(struct typhoon *tp, u32 initial_status)
 {
 	void __iomem *ioaddr = tp->ioaddr;
 
-	if(typhoon_wait_status(ioaddr, initial_status) < 0) {
+	if (typhoon_wait_status(ioaddr, initial_status) < 0) {
 		netdev_err(tp->dev, "boot ready timeout\n");
 		goto out_timeout;
 	}
@@ -1483,7 +1478,7 @@ typhoon_boot_3XP(struct typhoon *tp, u32 initial_status)
 	iowrite32(TYPHOON_BOOTCMD_REG_BOOT_RECORD,
 				ioaddr + TYPHOON_REG_COMMAND);
 
-	if(typhoon_wait_status(ioaddr, TYPHOON_STATUS_RUNNING) < 0) {
+	if (typhoon_wait_status(ioaddr, TYPHOON_STATUS_RUNNING) < 0) {
 		netdev_err(tp->dev, "boot finish timeout (status 0x%x)\n",
 			   ioread32(ioaddr + TYPHOON_REG_STATUS));
 		goto out_timeout;
@@ -1513,23 +1508,23 @@ typhoon_clean_tx(struct typhoon *tp, struct transmit_ring *txRing,
 	int dma_len;
 	int type;
 
-	while(lastRead != le32_to_cpu(*index)) {
+	while (lastRead != le32_to_cpu(*index)) {
 		tx = (struct tx_desc *) (txRing->ringBase + lastRead);
 		type = tx->flags & TYPHOON_TYPE_MASK;
 
-		if(type == TYPHOON_TX_DESC) {
+		if (type == TYPHOON_TX_DESC) {
 			/* This tx_desc describes a packet.
 			 */
 			unsigned long ptr = tx->tx_addr;
 			struct sk_buff *skb = (struct sk_buff *) ptr;
 			dev_kfree_skb_irq(skb);
-		} else if(type == TYPHOON_FRAG_DESC) {
+		} else if (type == TYPHOON_FRAG_DESC) {
 			/* This tx_desc describes a memory mapping. Free it.
 			 */
 			skb_dma = (dma_addr_t) le32_to_cpu(tx->frag.addr);
 			dma_len = le16_to_cpu(tx->len);
-			pci_unmap_single(tp->pdev, skb_dma, dma_len,
-				       PCI_DMA_TODEVICE);
+			dma_unmap_single(&tp->pdev->dev, skb_dma, dma_len,
+					 DMA_TO_DEVICE);
 		}
 
 		tx->flags = 0;
@@ -1548,7 +1543,7 @@ typhoon_tx_complete(struct typhoon *tp, struct transmit_ring *txRing,
 
 	/* This will need changing if we start to use the Hi Tx ring. */
 	lastRead = typhoon_clean_tx(tp, txRing, index);
-	if(netif_queue_stopped(tp->dev) && typhoon_num_free(txRing->lastWrite,
+	if (netif_queue_stopped(tp->dev) && typhoon_num_free(txRing->lastWrite,
 				lastRead, TXLO_ENTRIES) > (numDesc + 2))
 		netif_wake_queue(tp->dev);
 
@@ -1564,7 +1559,7 @@ typhoon_recycle_rx_skb(struct typhoon *tp, u32 idx)
 	struct basic_ring *ring = &tp->rxBuffRing;
 	struct rx_free *r;
 
-	if((ring->lastWrite + sizeof(*r)) % (RXFREE_ENTRIES * sizeof(*r)) ==
+	if ((ring->lastWrite + sizeof(*r)) % (RXFREE_ENTRIES * sizeof(*r)) ==
 				le32_to_cpu(indexes->rxBuffCleared)) {
 		/* no room in ring, just drop the skb
 		 */
@@ -1595,12 +1590,12 @@ typhoon_alloc_rx_skb(struct typhoon *tp, u32 idx)
 
 	rxb->skb = NULL;
 
-	if((ring->lastWrite + sizeof(*r)) % (RXFREE_ENTRIES * sizeof(*r)) ==
+	if ((ring->lastWrite + sizeof(*r)) % (RXFREE_ENTRIES * sizeof(*r)) ==
 				le32_to_cpu(indexes->rxBuffCleared))
 		return -ENOMEM;
 
 	skb = netdev_alloc_skb(tp->dev, PKT_BUF_SZ);
-	if(!skb)
+	if (!skb)
 		return -ENOMEM;
 
 #if 0
@@ -1610,8 +1605,8 @@ typhoon_alloc_rx_skb(struct typhoon *tp, u32 idx)
 	skb_reserve(skb, 2);
 #endif
 
-	dma_addr = pci_map_single(tp->pdev, skb->data,
-				  PKT_BUF_SZ, PCI_DMA_FROMDEVICE);
+	dma_addr = dma_map_single(&tp->pdev->dev, skb->data, PKT_BUF_SZ,
+				  DMA_FROM_DEVICE);
 
 	/* Since no card does 64 bit DAC, the high bits will never
 	 * change from zero.
@@ -1647,7 +1642,7 @@ typhoon_rx(struct typhoon *tp, struct basic_ring *rxRing, volatile __le32 * read
 	received = 0;
 	local_ready = le32_to_cpu(*ready);
 	rxaddr = le32_to_cpu(*cleared);
-	while(rxaddr != local_ready && budget > 0) {
+	while (rxaddr != local_ready && budget > 0) {
 		rx = (struct rx_desc *) (rxRing->ringBase + rxaddr);
 		idx = rx->addr;
 		rxb = &tp->rxbuffers[idx];
@@ -1656,36 +1651,35 @@ typhoon_rx(struct typhoon *tp, struct basic_ring *rxRing, volatile __le32 * read
 
 		typhoon_inc_rx_index(&rxaddr, 1);
 
-		if(rx->flags & TYPHOON_RX_ERROR) {
+		if (rx->flags & TYPHOON_RX_ERROR) {
 			typhoon_recycle_rx_skb(tp, idx);
 			continue;
 		}
 
 		pkt_len = le16_to_cpu(rx->frameLen);
 
-		if(pkt_len < rx_copybreak &&
+		if (pkt_len < rx_copybreak &&
 		   (new_skb = netdev_alloc_skb(tp->dev, pkt_len + 2)) != NULL) {
 			skb_reserve(new_skb, 2);
-			pci_dma_sync_single_for_cpu(tp->pdev, dma_addr,
-						    PKT_BUF_SZ,
-						    PCI_DMA_FROMDEVICE);
+			dma_sync_single_for_cpu(&tp->pdev->dev, dma_addr,
+						PKT_BUF_SZ, DMA_FROM_DEVICE);
 			skb_copy_to_linear_data(new_skb, skb->data, pkt_len);
-			pci_dma_sync_single_for_device(tp->pdev, dma_addr,
-						       PKT_BUF_SZ,
-						       PCI_DMA_FROMDEVICE);
+			dma_sync_single_for_device(&tp->pdev->dev, dma_addr,
+						   PKT_BUF_SZ,
+						   DMA_FROM_DEVICE);
 			skb_put(new_skb, pkt_len);
 			typhoon_recycle_rx_skb(tp, idx);
 		} else {
 			new_skb = skb;
 			skb_put(new_skb, pkt_len);
-			pci_unmap_single(tp->pdev, dma_addr, PKT_BUF_SZ,
-				       PCI_DMA_FROMDEVICE);
+			dma_unmap_single(&tp->pdev->dev, dma_addr, PKT_BUF_SZ,
+					 DMA_FROM_DEVICE);
 			typhoon_alloc_rx_skb(tp, idx);
 		}
 		new_skb->protocol = eth_type_trans(new_skb, tp->dev);
 		csum_bits = rx->rxStatus & (TYPHOON_RX_IP_CHK_GOOD |
 			TYPHOON_RX_UDP_CHK_GOOD | TYPHOON_RX_TCP_CHK_GOOD);
-		if(csum_bits ==
+		if (csum_bits ==
 		   (TYPHOON_RX_IP_CHK_GOOD | TYPHOON_RX_TCP_CHK_GOOD) ||
 		   csum_bits ==
 		   (TYPHOON_RX_IP_CHK_GOOD | TYPHOON_RX_UDP_CHK_GOOD)) {
@@ -1711,11 +1705,11 @@ typhoon_fill_free_ring(struct typhoon *tp)
 {
 	u32 i;
 
-	for(i = 0; i < RXENT_ENTRIES; i++) {
+	for (i = 0; i < RXENT_ENTRIES; i++) {
 		struct rxbuff_ent *rxb = &tp->rxbuffers[i];
-		if(rxb->skb)
+		if (rxb->skb)
 			continue;
-		if(typhoon_alloc_rx_skb(tp, i) < 0)
+		if (typhoon_alloc_rx_skb(tp, i) < 0)
 			break;
 	}
 }
@@ -1728,25 +1722,25 @@ typhoon_poll(struct napi_struct *napi, int budget)
 	int work_done;
 
 	rmb();
-	if(!tp->awaiting_resp && indexes->respReady != indexes->respCleared)
+	if (!tp->awaiting_resp && indexes->respReady != indexes->respCleared)
 			typhoon_process_response(tp, 0, NULL);
 
-	if(le32_to_cpu(indexes->txLoCleared) != tp->txLoRing.lastRead)
+	if (le32_to_cpu(indexes->txLoCleared) != tp->txLoRing.lastRead)
 		typhoon_tx_complete(tp, &tp->txLoRing, &indexes->txLoCleared);
 
 	work_done = 0;
 
-	if(indexes->rxHiCleared != indexes->rxHiReady) {
+	if (indexes->rxHiCleared != indexes->rxHiReady) {
 		work_done += typhoon_rx(tp, &tp->rxHiRing, &indexes->rxHiReady,
 			   		&indexes->rxHiCleared, budget);
 	}
 
-	if(indexes->rxLoCleared != indexes->rxLoReady) {
+	if (indexes->rxLoCleared != indexes->rxLoReady) {
 		work_done += typhoon_rx(tp, &tp->rxLoRing, &indexes->rxLoReady,
 					&indexes->rxLoCleared, budget - work_done);
 	}
 
-	if(le32_to_cpu(indexes->rxBuffCleared) == tp->rxBuffRing.lastWrite) {
+	if (le32_to_cpu(indexes->rxBuffCleared) == tp->rxBuffRing.lastWrite) {
 		/* rxBuff ring is empty, try to fill it. */
 		typhoon_fill_free_ring(tp);
 	}
@@ -1770,7 +1764,7 @@ typhoon_interrupt(int irq, void *dev_instance)
 	u32 intr_status;
 
 	intr_status = ioread32(ioaddr + TYPHOON_REG_INTR_STATUS);
-	if(!(intr_status & TYPHOON_INTR_HOST_INT))
+	if (!(intr_status & TYPHOON_INTR_HOST_INT))
 		return IRQ_NONE;
 
 	iowrite32(intr_status, ioaddr + TYPHOON_REG_INTR_STATUS);
@@ -1790,11 +1784,11 @@ typhoon_free_rx_rings(struct typhoon *tp)
 {
 	u32 i;
 
-	for(i = 0; i < RXENT_ENTRIES; i++) {
+	for (i = 0; i < RXENT_ENTRIES; i++) {
 		struct rxbuff_ent *rxb = &tp->rxbuffers[i];
-		if(rxb->skb) {
-			pci_unmap_single(tp->pdev, rxb->dma_addr, PKT_BUF_SZ,
-				       PCI_DMA_FROMDEVICE);
+		if (rxb->skb) {
+			dma_unmap_single(&tp->pdev->dev, rxb->dma_addr,
+					 PKT_BUF_SZ, DMA_FROM_DEVICE);
 			dev_kfree_skb(rxb->skb);
 			rxb->skb = NULL;
 		}
@@ -1802,9 +1796,8 @@ typhoon_free_rx_rings(struct typhoon *tp)
 }
 
 static int
-typhoon_sleep(struct typhoon *tp, pci_power_t state, __le16 events)
+typhoon_sleep_early(struct typhoon *tp, __le16 events)
 {
-	struct pci_dev *pdev = tp->pdev;
 	void __iomem *ioaddr = tp->ioaddr;
 	struct cmd_desc xp_cmd;
 	int err;
@@ -1812,7 +1805,7 @@ typhoon_sleep(struct typhoon *tp, pci_power_t state, __le16 events)
 	INIT_COMMAND_WITH_RESPONSE(&xp_cmd, TYPHOON_CMD_ENABLE_WAKE_EVENTS);
 	xp_cmd.parm1 = events;
 	err = typhoon_issue_command(tp, 1, &xp_cmd, 0, NULL);
-	if(err < 0) {
+	if (err < 0) {
 		netdev_err(tp->dev, "typhoon_sleep(): wake events cmd err %d\n",
 			   err);
 		return err;
@@ -1820,12 +1813,12 @@ typhoon_sleep(struct typhoon *tp, pci_power_t state, __le16 events)
 
 	INIT_COMMAND_NO_RESPONSE(&xp_cmd, TYPHOON_CMD_GOTO_SLEEP);
 	err = typhoon_issue_command(tp, 1, &xp_cmd, 0, NULL);
-	if(err < 0) {
+	if (err < 0) {
 		netdev_err(tp->dev, "typhoon_sleep(): sleep cmd err %d\n", err);
 		return err;
 	}
 
-	if(typhoon_wait_status(ioaddr, TYPHOON_STATUS_SLEEPING) < 0)
+	if (typhoon_wait_status(ioaddr, TYPHOON_STATUS_SLEEPING) < 0)
 		return -ETIMEDOUT;
 
 	/* Since we cannot monitor the status of the link while sleeping,
@@ -1833,26 +1826,35 @@ typhoon_sleep(struct typhoon *tp, pci_power_t state, __le16 events)
 	 */
 	netif_carrier_off(tp->dev);
 
+	return 0;
+}
+
+static int
+typhoon_sleep(struct typhoon *tp, pci_power_t state, __le16 events)
+{
+	int err;
+
+	err = typhoon_sleep_early(tp, events);
+
+	if (err)
+		return err;
+
 	pci_enable_wake(tp->pdev, state, 1);
-	pci_disable_device(pdev);
-	return pci_set_power_state(pdev, state);
+	pci_disable_device(tp->pdev);
+	return pci_set_power_state(tp->pdev, state);
 }
 
 static int
 typhoon_wakeup(struct typhoon *tp, int wait_type)
 {
-	struct pci_dev *pdev = tp->pdev;
 	void __iomem *ioaddr = tp->ioaddr;
 
-	pci_set_power_state(pdev, PCI_D0);
-	pci_restore_state(pdev);
-
 	/* Post 2.x.x versions of the Sleep Image require a reset before
 	 * we can download the Runtime Image. But let's not make users of
 	 * the old firmware pay for the reset.
 	 */
 	iowrite32(TYPHOON_BOOTCMD_WAKEUP, ioaddr + TYPHOON_REG_COMMAND);
-	if(typhoon_wait_status(ioaddr, TYPHOON_STATUS_WAITING_FOR_HOST) < 0 ||
+	if (typhoon_wait_status(ioaddr, TYPHOON_STATUS_WAITING_FOR_HOST) < 0 ||
 			(tp->capabilities & TYPHOON_WAKEUP_NEEDS_RESET))
 		return typhoon_reset(ioaddr, wait_type);
 
@@ -1871,12 +1873,12 @@ typhoon_start_runtime(struct typhoon *tp)
 	typhoon_fill_free_ring(tp);
 
 	err = typhoon_download_firmware(tp);
-	if(err < 0) {
+	if (err < 0) {
 		netdev_err(tp->dev, "cannot load runtime on 3XP\n");
 		goto error_out;
 	}
 
-	if(typhoon_boot_3XP(tp, TYPHOON_STATUS_WAITING_FOR_BOOT) < 0) {
+	if (typhoon_boot_3XP(tp, TYPHOON_STATUS_WAITING_FOR_BOOT) < 0) {
 		netdev_err(tp->dev, "cannot boot 3XP\n");
 		err = -EIO;
 		goto error_out;
@@ -1885,14 +1887,14 @@ typhoon_start_runtime(struct typhoon *tp)
 	INIT_COMMAND_NO_RESPONSE(&xp_cmd, TYPHOON_CMD_SET_MAX_PKT_SIZE);
 	xp_cmd.parm1 = cpu_to_le16(PKT_BUF_SZ);
 	err = typhoon_issue_command(tp, 1, &xp_cmd, 0, NULL);
-	if(err < 0)
+	if (err < 0)
 		goto error_out;
 
 	INIT_COMMAND_NO_RESPONSE(&xp_cmd, TYPHOON_CMD_SET_MAC_ADDRESS);
 	xp_cmd.parm1 = cpu_to_le16(ntohs(*(__be16 *)&dev->dev_addr[0]));
 	xp_cmd.parm2 = cpu_to_le32(ntohl(*(__be32 *)&dev->dev_addr[2]));
 	err = typhoon_issue_command(tp, 1, &xp_cmd, 0, NULL);
-	if(err < 0)
+	if (err < 0)
 		goto error_out;
 
 	/* Disable IRQ coalescing -- we can reenable it when 3Com gives
@@ -1901,38 +1903,38 @@ typhoon_start_runtime(struct typhoon *tp)
 	INIT_COMMAND_WITH_RESPONSE(&xp_cmd, TYPHOON_CMD_IRQ_COALESCE_CTRL);
 	xp_cmd.parm1 = 0;
 	err = typhoon_issue_command(tp, 1, &xp_cmd, 0, NULL);
-	if(err < 0)
+	if (err < 0)
 		goto error_out;
 
 	INIT_COMMAND_NO_RESPONSE(&xp_cmd, TYPHOON_CMD_XCVR_SELECT);
 	xp_cmd.parm1 = tp->xcvr_select;
 	err = typhoon_issue_command(tp, 1, &xp_cmd, 0, NULL);
-	if(err < 0)
+	if (err < 0)
 		goto error_out;
 
 	INIT_COMMAND_NO_RESPONSE(&xp_cmd, TYPHOON_CMD_VLAN_TYPE_WRITE);
 	xp_cmd.parm1 = cpu_to_le16(ETH_P_8021Q);
 	err = typhoon_issue_command(tp, 1, &xp_cmd, 0, NULL);
-	if(err < 0)
+	if (err < 0)
 		goto error_out;
 
 	INIT_COMMAND_NO_RESPONSE(&xp_cmd, TYPHOON_CMD_SET_OFFLOAD_TASKS);
 	xp_cmd.parm2 = tp->offload;
 	xp_cmd.parm3 = tp->offload;
 	err = typhoon_issue_command(tp, 1, &xp_cmd, 0, NULL);
-	if(err < 0)
+	if (err < 0)
 		goto error_out;
 
 	typhoon_set_rx_mode(dev);
 
 	INIT_COMMAND_NO_RESPONSE(&xp_cmd, TYPHOON_CMD_TX_ENABLE);
 	err = typhoon_issue_command(tp, 1, &xp_cmd, 0, NULL);
-	if(err < 0)
+	if (err < 0)
 		goto error_out;
 
 	INIT_COMMAND_WITH_RESPONSE(&xp_cmd, TYPHOON_CMD_RX_ENABLE);
 	err = typhoon_issue_command(tp, 1, &xp_cmd, 0, NULL);
-	if(err < 0)
+	if (err < 0)
 		goto error_out;
 
 	tp->card_state = Running;
@@ -1972,13 +1974,13 @@ typhoon_stop_runtime(struct typhoon *tp, int wait_type)
 	/* Wait 1/2 sec for any outstanding transmits to occur
 	 * We'll cleanup after the reset if this times out.
 	 */
-	for(i = 0; i < TYPHOON_WAIT_TIMEOUT; i++) {
-		if(indexes->txLoCleared == cpu_to_le32(txLo->lastWrite))
+	for (i = 0; i < TYPHOON_WAIT_TIMEOUT; i++) {
+		if (indexes->txLoCleared == cpu_to_le32(txLo->lastWrite))
 			break;
 		udelay(TYPHOON_UDELAY);
 	}
 
-	if(i == TYPHOON_WAIT_TIMEOUT)
+	if (i == TYPHOON_WAIT_TIMEOUT)
 		netdev_err(tp->dev, "halt timed out waiting for Tx to complete\n");
 
 	INIT_COMMAND_NO_RESPONSE(&xp_cmd, TYPHOON_CMD_TX_DISABLE);
@@ -1995,16 +1997,16 @@ typhoon_stop_runtime(struct typhoon *tp, int wait_type)
 	INIT_COMMAND_NO_RESPONSE(&xp_cmd, TYPHOON_CMD_HALT);
 	typhoon_issue_command(tp, 1, &xp_cmd, 0, NULL);
 
-	if(typhoon_wait_status(ioaddr, TYPHOON_STATUS_HALTED) < 0)
+	if (typhoon_wait_status(ioaddr, TYPHOON_STATUS_HALTED) < 0)
 		netdev_err(tp->dev, "timed out waiting for 3XP to halt\n");
 
-	if(typhoon_reset(ioaddr, wait_type) < 0) {
+	if (typhoon_reset(ioaddr, wait_type) < 0) {
 		netdev_err(tp->dev, "unable to reset 3XP\n");
 		return -ETIMEDOUT;
 	}
 
 	/* cleanup any outstanding Tx packets */
-	if(indexes->txLoCleared != cpu_to_le32(txLo->lastWrite)) {
+	if (indexes->txLoCleared != cpu_to_le32(txLo->lastWrite)) {
 		indexes->txLoCleared = cpu_to_le32(txLo->lastWrite);
 		typhoon_clean_tx(tp, &tp->txLoRing, &indexes->txLoCleared);
 	}
@@ -2013,11 +2015,11 @@ typhoon_stop_runtime(struct typhoon *tp, int wait_type)
 }
 
 static void
-typhoon_tx_timeout(struct net_device *dev)
+typhoon_tx_timeout(struct net_device *dev, unsigned int txqueue)
 {
 	struct typhoon *tp = netdev_priv(dev);
 
-	if(typhoon_reset(tp->ioaddr, WaitNoSleep) < 0) {
+	if (typhoon_reset(tp->ioaddr, WaitNoSleep) < 0) {
 		netdev_warn(dev, "could not reset in tx timeout\n");
 		goto truly_dead;
 	}
@@ -2026,7 +2028,7 @@ typhoon_tx_timeout(struct net_device *dev)
 	typhoon_clean_tx(tp, &tp->txLoRing, &tp->indexes->txLoCleared);
 	typhoon_free_rx_rings(tp);
 
-	if(typhoon_start_runtime(tp) < 0) {
+	if (typhoon_start_runtime(tp) < 0) {
 		netdev_err(dev, "could not start runtime in tx timeout\n");
 		goto truly_dead;
         }
@@ -2050,21 +2052,24 @@ typhoon_open(struct net_device *dev)
 	if (err)
 		goto out;
 
+	pci_set_power_state(tp->pdev, PCI_D0);
+	pci_restore_state(tp->pdev);
+
 	err = typhoon_wakeup(tp, WaitSleep);
-	if(err < 0) {
+	if (err < 0) {
 		netdev_err(dev, "unable to wakeup device\n");
 		goto out_sleep;
 	}
 
 	err = request_irq(dev->irq, typhoon_interrupt, IRQF_SHARED,
 				dev->name, dev);
-	if(err < 0)
+	if (err < 0)
 		goto out_sleep;
 
 	napi_enable(&tp->napi);
 
 	err = typhoon_start_runtime(tp);
-	if(err < 0) {
+	if (err < 0) {
 		napi_disable(&tp->napi);
 		goto out_irq;
 	}
@@ -2076,13 +2081,13 @@ out_irq:
 	free_irq(dev->irq, dev);
 
 out_sleep:
-	if(typhoon_boot_3XP(tp, TYPHOON_STATUS_WAITING_FOR_HOST) < 0) {
+	if (typhoon_boot_3XP(tp, TYPHOON_STATUS_WAITING_FOR_HOST) < 0) {
 		netdev_err(dev, "unable to reboot into sleep img\n");
 		typhoon_reset(tp->ioaddr, NoWait);
 		goto out;
 	}
 
-	if(typhoon_sleep(tp, PCI_D3hot, 0) < 0)
+	if (typhoon_sleep(tp, PCI_D3hot, 0) < 0)
 		netdev_err(dev, "unable to go back to sleep\n");
 
 out:
@@ -2097,7 +2102,7 @@ typhoon_close(struct net_device *dev)
 	netif_stop_queue(dev);
 	napi_disable(&tp->napi);
 
-	if(typhoon_stop_runtime(tp, WaitSleep) < 0)
+	if (typhoon_stop_runtime(tp, WaitSleep) < 0)
 		netdev_err(dev, "unable to stop runtime\n");
 
 	/* Make sure there is no irq handler running on a different CPU. */
@@ -2106,33 +2111,32 @@ typhoon_close(struct net_device *dev)
 	typhoon_free_rx_rings(tp);
 	typhoon_init_rings(tp);
 
-	if(typhoon_boot_3XP(tp, TYPHOON_STATUS_WAITING_FOR_HOST) < 0)
+	if (typhoon_boot_3XP(tp, TYPHOON_STATUS_WAITING_FOR_HOST) < 0)
 		netdev_err(dev, "unable to boot sleep image\n");
 
-	if(typhoon_sleep(tp, PCI_D3hot, 0) < 0)
+	if (typhoon_sleep(tp, PCI_D3hot, 0) < 0)
 		netdev_err(dev, "unable to put card to sleep\n");
 
 	return 0;
 }
 
-#ifdef CONFIG_PM
-static int
-typhoon_resume(struct pci_dev *pdev)
+static int __maybe_unused
+typhoon_resume(struct device *dev_d)
 {
-	struct net_device *dev = pci_get_drvdata(pdev);
+	struct net_device *dev = dev_get_drvdata(dev_d);
 	struct typhoon *tp = netdev_priv(dev);
 
 	/* If we're down, resume when we are upped.
 	 */
-	if(!netif_running(dev))
+	if (!netif_running(dev))
 		return 0;
 
-	if(typhoon_wakeup(tp, WaitNoSleep) < 0) {
+	if (typhoon_wakeup(tp, WaitNoSleep) < 0) {
 		netdev_err(dev, "critical: could not wake up in resume\n");
 		goto reset;
 	}
 
-	if(typhoon_start_runtime(tp) < 0) {
+	if (typhoon_start_runtime(tp) < 0) {
 		netdev_err(dev, "critical: could not start runtime in resume\n");
 		goto reset;
 	}
@@ -2145,25 +2149,26 @@ reset:
 	return -EBUSY;
 }
 
-static int
-typhoon_suspend(struct pci_dev *pdev, pm_message_t state)
+static int __maybe_unused
+typhoon_suspend(struct device *dev_d)
 {
+	struct pci_dev *pdev = to_pci_dev(dev_d);
 	struct net_device *dev = pci_get_drvdata(pdev);
 	struct typhoon *tp = netdev_priv(dev);
 	struct cmd_desc xp_cmd;
 
 	/* If we're down, we're already suspended.
 	 */
-	if(!netif_running(dev))
+	if (!netif_running(dev))
 		return 0;
 
 	/* TYPHOON_OFFLOAD_VLAN is always on now, so this doesn't work */
-	if(tp->wol_events & TYPHOON_WAKE_MAGIC_PKT)
+	if (tp->wol_events & TYPHOON_WAKE_MAGIC_PKT)
 		netdev_warn(dev, "cannot do WAKE_MAGIC with VLAN offloading\n");
 
 	netif_device_detach(dev);
 
-	if(typhoon_stop_runtime(tp, WaitNoSleep) < 0) {
+	if (typhoon_stop_runtime(tp, WaitNoSleep) < 0) {
 		netdev_err(dev, "unable to stop runtime\n");
 		goto need_resume;
 	}
@@ -2171,7 +2176,7 @@ typhoon_suspend(struct pci_dev *pdev, pm_message_t state)
 	typhoon_free_rx_rings(tp);
 	typhoon_init_rings(tp);
 
-	if(typhoon_boot_3XP(tp, TYPHOON_STATUS_WAITING_FOR_HOST) < 0) {
+	if (typhoon_boot_3XP(tp, TYPHOON_STATUS_WAITING_FOR_HOST) < 0) {
 		netdev_err(dev, "unable to boot sleep image\n");
 		goto need_resume;
 	}
@@ -2179,30 +2184,31 @@ typhoon_suspend(struct pci_dev *pdev, pm_message_t state)
 	INIT_COMMAND_NO_RESPONSE(&xp_cmd, TYPHOON_CMD_SET_MAC_ADDRESS);
 	xp_cmd.parm1 = cpu_to_le16(ntohs(*(__be16 *)&dev->dev_addr[0]));
 	xp_cmd.parm2 = cpu_to_le32(ntohl(*(__be32 *)&dev->dev_addr[2]));
-	if(typhoon_issue_command(tp, 1, &xp_cmd, 0, NULL) < 0) {
+	if (typhoon_issue_command(tp, 1, &xp_cmd, 0, NULL) < 0) {
 		netdev_err(dev, "unable to set mac address in suspend\n");
 		goto need_resume;
 	}
 
 	INIT_COMMAND_NO_RESPONSE(&xp_cmd, TYPHOON_CMD_SET_RX_FILTER);
 	xp_cmd.parm1 = TYPHOON_RX_FILTER_DIRECTED | TYPHOON_RX_FILTER_BROADCAST;
-	if(typhoon_issue_command(tp, 1, &xp_cmd, 0, NULL) < 0) {
+	if (typhoon_issue_command(tp, 1, &xp_cmd, 0, NULL) < 0) {
 		netdev_err(dev, "unable to set rx filter in suspend\n");
 		goto need_resume;
 	}
 
-	if(typhoon_sleep(tp, pci_choose_state(pdev, state), tp->wol_events) < 0) {
+	if (typhoon_sleep_early(tp, tp->wol_events) < 0) {
 		netdev_err(dev, "unable to put card to sleep\n");
 		goto need_resume;
 	}
 
+	device_wakeup_enable(dev_d);
+
 	return 0;
 
 need_resume:
-	typhoon_resume(pdev);
+	typhoon_resume(dev_d);
 	return -EBUSY;
 }
-#endif
 
 static int
 typhoon_test_mmio(struct pci_dev *pdev)
@@ -2211,10 +2217,10 @@ typhoon_test_mmio(struct pci_dev *pdev)
 	int mode = 0;
 	u32 val;
 
-	if(!ioaddr)
+	if (!ioaddr)
 		goto out;
 
-	if(ioread32(ioaddr + TYPHOON_REG_STATUS) !=
+	if (ioread32(ioaddr + TYPHOON_REG_STATUS) !=
 				TYPHOON_STATUS_WAITING_FOR_HOST)
 		goto out_unmap;
 
@@ -2227,12 +2233,12 @@ typhoon_test_mmio(struct pci_dev *pdev)
 	 * The 50usec delay is arbitrary -- it could probably be smaller.
 	 */
 	val = ioread32(ioaddr + TYPHOON_REG_INTR_STATUS);
-	if((val & TYPHOON_INTR_SELF) == 0) {
+	if ((val & TYPHOON_INTR_SELF) == 0) {
 		iowrite32(1, ioaddr + TYPHOON_REG_SELF_INTERRUPT);
 		ioread32(ioaddr + TYPHOON_REG_INTR_STATUS);
 		udelay(50);
 		val = ioread32(ioaddr + TYPHOON_REG_INTR_STATUS);
-		if(val & TYPHOON_INTR_SELF)
+		if (val & TYPHOON_INTR_SELF)
 			mode = 1;
 	}
 
@@ -2245,14 +2251,33 @@ out_unmap:
 	pci_iounmap(pdev, ioaddr);
 
 out:
-	if(!mode)
+	if (!mode)
 		pr_info("%s: falling back to port IO\n", pci_name(pdev));
 	return mode;
 }
 
+#if MAX_SKB_FRAGS > 32
+
+#include <net/vxlan.h>
+
+static netdev_features_t typhoon_features_check(struct sk_buff *skb,
+						struct net_device *dev,
+						netdev_features_t features)
+{
+	if (skb_shinfo(skb)->nr_frags > 32 && skb_is_gso(skb))
+		features &= ~NETIF_F_GSO_MASK;
+
+	features = vlan_features_check(skb, features);
+	return vxlan_features_check(skb, features);
+}
+#endif
+
 static const struct net_device_ops typhoon_netdev_ops = {
 	.ndo_open		= typhoon_open,
 	.ndo_stop		= typhoon_close,
+#if MAX_SKB_FRAGS > 32
+	.ndo_features_check	= typhoon_features_check,
+#endif
 	.ndo_start_xmit		= typhoon_start_tx,
 	.ndo_set_rx_mode	= typhoon_set_rx_mode,
 	.ndo_tx_timeout		= typhoon_tx_timeout,
@@ -2267,6 +2292,7 @@ typhoon_init_one(struct pci_dev *pdev, const struct pci_device_id *ent)
 	struct net_device *dev;
 	struct typhoon *tp;
 	int card_id = (int) ent->driver_data;
+	u8 addr[ETH_ALEN] __aligned(4);
 	void __iomem *ioaddr;
 	void *shared;
 	dma_addr_t shared_dma;
@@ -2276,7 +2302,7 @@ typhoon_init_one(struct pci_dev *pdev, const struct pci_device_id *ent)
 	const char *err_msg;
 
 	dev = alloc_etherdev(sizeof(*tp));
-	if(dev == NULL) {
+	if (dev == NULL) {
 		err_msg = "unable to alloc new net device";
 		err = -ENOMEM;
 		goto error_out;
@@ -2284,55 +2310,55 @@ typhoon_init_one(struct pci_dev *pdev, const struct pci_device_id *ent)
 	SET_NETDEV_DEV(dev, &pdev->dev);
 
 	err = pci_enable_device(pdev);
-	if(err < 0) {
+	if (err < 0) {
 		err_msg = "unable to enable device";
 		goto error_out_dev;
 	}
 
 	err = pci_set_mwi(pdev);
-	if(err < 0) {
+	if (err < 0) {
 		err_msg = "unable to set MWI";
 		goto error_out_disable;
 	}
 
-	err = pci_set_dma_mask(pdev, DMA_BIT_MASK(32));
-	if(err < 0) {
+	err = dma_set_mask(&pdev->dev, DMA_BIT_MASK(32));
+	if (err < 0) {
 		err_msg = "No usable DMA configuration";
 		goto error_out_mwi;
 	}
 
 	/* sanity checks on IO and MMIO BARs
 	 */
-	if(!(pci_resource_flags(pdev, 0) & IORESOURCE_IO)) {
+	if (!(pci_resource_flags(pdev, 0) & IORESOURCE_IO)) {
 		err_msg = "region #1 not a PCI IO resource, aborting";
 		err = -ENODEV;
 		goto error_out_mwi;
 	}
-	if(pci_resource_len(pdev, 0) < 128) {
+	if (pci_resource_len(pdev, 0) < 128) {
 		err_msg = "Invalid PCI IO region size, aborting";
 		err = -ENODEV;
 		goto error_out_mwi;
 	}
-	if(!(pci_resource_flags(pdev, 1) & IORESOURCE_MEM)) {
+	if (!(pci_resource_flags(pdev, 1) & IORESOURCE_MEM)) {
 		err_msg = "region #1 not a PCI MMIO resource, aborting";
 		err = -ENODEV;
 		goto error_out_mwi;
 	}
-	if(pci_resource_len(pdev, 1) < 128) {
+	if (pci_resource_len(pdev, 1) < 128) {
 		err_msg = "Invalid PCI MMIO region size, aborting";
 		err = -ENODEV;
 		goto error_out_mwi;
 	}
 
 	err = pci_request_regions(pdev, KBUILD_MODNAME);
-	if(err < 0) {
+	if (err < 0) {
 		err_msg = "could not request regions";
 		goto error_out_mwi;
 	}
 
 	/* map our registers
 	 */
-	if(use_mmio != 0 && use_mmio != 1)
+	if (use_mmio != 0 && use_mmio != 1)
 		use_mmio = typhoon_test_mmio(pdev);
 
 	ioaddr = pci_iomap(pdev, use_mmio, 128);
@@ -2344,9 +2370,9 @@ typhoon_init_one(struct pci_dev *pdev, const struct pci_device_id *ent)
 
 	/* allocate pci dma space for rx and tx descriptor rings
 	 */
-	shared = pci_alloc_consistent(pdev, sizeof(struct typhoon_shared),
-				      &shared_dma);
-	if(!shared) {
+	shared = dma_alloc_coherent(&pdev->dev, sizeof(struct typhoon_shared),
+				    &shared_dma, GFP_KERNEL);
+	if (!shared) {
 		err_msg = "could not allocate DMA memory";
 		err = -ENOMEM;
 		goto error_out_remap;
@@ -2398,8 +2424,9 @@ typhoon_init_one(struct pci_dev *pdev, const struct pci_device_id *ent)
 		goto error_out_reset;
 	}
 
-	*(__be16 *)&dev->dev_addr[0] = htons(le16_to_cpu(xp_resp[0].parm1));
-	*(__be32 *)&dev->dev_addr[2] = htonl(le32_to_cpu(xp_resp[0].parm2));
+	*(__be16 *)&addr[0] = htons(le16_to_cpu(xp_resp[0].parm1));
+	*(__be32 *)&addr[2] = htonl(le32_to_cpu(xp_resp[0].parm2));
+	eth_hw_addr_set(dev, addr);
 
 	if (!is_valid_ether_addr(dev->dev_addr)) {
 		err_msg = "Could not obtain valid ethernet address, aborting";
@@ -2426,7 +2453,7 @@ typhoon_init_one(struct pci_dev *pdev, const struct pci_device_id *ent)
 	 * seem to need a little extra help to get started. Since we don't
 	 * know how to nudge it along, just kick it.
 	 */
-	if(xp_resp[0].numDesc != 0)
+	if (xp_resp[0].numDesc != 0)
 		tp->capabilities |= TYPHOON_WAKEUP_NEEDS_RESET;
 
 	err = typhoon_sleep(tp, PCI_D3hot, 0);
@@ -2437,7 +2464,7 @@ typhoon_init_one(struct pci_dev *pdev, const struct pci_device_id *ent)
 
 	/* The chip-specific entries in the device structure. */
 	dev->netdev_ops		= &typhoon_netdev_ops;
-	netif_napi_add(dev, &tp->napi, typhoon_poll, 16);
+	netif_napi_add_weight(dev, &tp->napi, typhoon_poll, 16);
 	dev->watchdog_timeo	= TX_TIMEOUT;
 
 	dev->ethtool_ops = &typhoon_ethtool_ops;
@@ -2471,14 +2498,14 @@ typhoon_init_one(struct pci_dev *pdev, const struct pci_device_id *ent)
 	/* xp_resp still contains the response to the READ_VERSIONS command.
 	 * For debugging, let the user know what version he has.
 	 */
-	if(xp_resp[0].numDesc == 0) {
+	if (xp_resp[0].numDesc == 0) {
 		/* This is the Typhoon 1.0 type Sleep Image, last 16 bits
 		 * of version is Month/Day of build.
 		 */
 		u16 monthday = le32_to_cpu(xp_resp[0].parm2) & 0xffff;
 		netdev_info(dev, "Typhoon 1.0 Sleep Image built %02u/%02u/2000\n",
 			    monthday >> 8, monthday & 0xff);
-	} else if(xp_resp[0].numDesc == 2) {
+	} else if (xp_resp[0].numDesc == 2) {
 		/* This is the Typhoon 1.1+ type Sleep Image
 		 */
 		u32 sleep_ver = le32_to_cpu(xp_resp[0].parm2);
@@ -2498,8 +2525,8 @@ error_out_reset:
 	typhoon_reset(ioaddr, NoWait);
 
 error_out_dma:
-	pci_free_consistent(pdev, sizeof(struct typhoon_shared),
-			    shared, shared_dma);
+	dma_free_coherent(&pdev->dev, sizeof(struct typhoon_shared), shared,
+			  shared_dma);
 error_out_remap:
 	pci_iounmap(pdev, ioaddr);
 error_out_regions:
@@ -2526,23 +2553,22 @@ typhoon_remove_one(struct pci_dev *pdev)
 	pci_restore_state(pdev);
 	typhoon_reset(tp->ioaddr, NoWait);
 	pci_iounmap(pdev, tp->ioaddr);
-	pci_free_consistent(pdev, sizeof(struct typhoon_shared),
-			    tp->shared, tp->shared_dma);
+	dma_free_coherent(&pdev->dev, sizeof(struct typhoon_shared),
+			  tp->shared, tp->shared_dma);
 	pci_release_regions(pdev);
 	pci_clear_mwi(pdev);
 	pci_disable_device(pdev);
 	free_netdev(dev);
 }
 
+static SIMPLE_DEV_PM_OPS(typhoon_pm_ops, typhoon_suspend, typhoon_resume);
+
 static struct pci_driver typhoon_driver = {
 	.name		= KBUILD_MODNAME,
 	.id_table	= typhoon_pci_tbl,
 	.probe		= typhoon_init_one,
 	.remove		= typhoon_remove_one,
-#ifdef CONFIG_PM
-	.suspend	= typhoon_suspend,
-	.resume		= typhoon_resume,
-#endif
+	.driver.pm	= &typhoon_pm_ops,
 };
 
 static int __init
diff --git a/drivers/net/ethernet/3com/typhoon.h b/drivers/net/ethernet/3com/typhoon.h
index 88187fc84aa3..38e6dcab4e94 100644
--- a/drivers/net/ethernet/3com/typhoon.h
+++ b/drivers/net/ethernet/3com/typhoon.h
@@ -33,7 +33,7 @@ struct basic_ring {
 	u32 lastWrite;
 };
 
-/* The Typoon transmit ring -- same as a basic ring, plus:
+/* The Typhoon transmit ring -- same as a basic ring, plus:
  * lastRead:      where we're at in regard to cleaning up the ring
  * writeRegister: register to use for writing (different for Hi & Lo rings)
  */
@@ -366,7 +366,7 @@ struct resp_desc {
 		memset(_ptr, 0, sizeof(struct cmd_desc));		\
 		_ptr->flags = TYPHOON_CMD_DESC | TYPHOON_DESC_VALID;	\
 		_ptr->cmd = command;					\
-	} while(0)
+	} while (0)
 
 /* We set seqNo to 1 if we're expecting a response from this command */
 #define INIT_COMMAND_WITH_RESPONSE(x, command)				\
@@ -376,7 +376,7 @@ struct resp_desc {
 		_ptr->flags |= TYPHOON_DESC_VALID; 			\
 		_ptr->cmd = command;					\
 		_ptr->seqNo = 1;					\
-	} while(0)
+	} while (0)
 
 /* TYPHOON_CMD_SET_RX_FILTER filter bits (cmd.parm1)
  */
diff --git a/drivers/net/ethernet/8390/8390.c b/drivers/net/ethernet/8390/8390.c
index a43544af257b..c5636245f1ca 100644
--- a/drivers/net/ethernet/8390/8390.c
+++ b/drivers/net/ethernet/8390/8390.c
@@ -1,3 +1,4 @@
+// SPDX-License-Identifier: GPL-2.0-only
 /* 8390 core for usual drivers */
 
 static const char version[] =
@@ -35,9 +36,9 @@ void ei_set_multicast_list(struct net_device *dev)
 }
 EXPORT_SYMBOL(ei_set_multicast_list);
 
-void ei_tx_timeout(struct net_device *dev)
+void ei_tx_timeout(struct net_device *dev, unsigned int txqueue)
 {
-	__ei_tx_timeout(dev);
+	__ei_tx_timeout(dev, txqueue);
 }
 EXPORT_SYMBOL(ei_tx_timeout);
 
@@ -99,4 +100,5 @@ static void __exit ns8390_module_exit(void)
 module_init(ns8390_module_init);
 module_exit(ns8390_module_exit);
 #endif /* MODULE */
+MODULE_DESCRIPTION("National Semiconductor 8390 core driver");
 MODULE_LICENSE("GPL");
diff --git a/drivers/net/ethernet/8390/8390.h b/drivers/net/ethernet/8390/8390.h
index 3e2f2c2e7b58..f784a6e2ab0e 100644
--- a/drivers/net/ethernet/8390/8390.h
+++ b/drivers/net/ethernet/8390/8390.h
@@ -1,8 +1,12 @@
+/* SPDX-License-Identifier: GPL-1.0+ */
+
 /* Generic NS8390 register definitions. */
+
 /* This file is part of Donald Becker's 8390 drivers, and is distributed
-   under the same license. Auto-loading of 8390.o only in v2.2 - Paul G.
-   Some of these names and comments originated from the Crynwr
-   packet drivers, which are distributed under the GPL. */
+ * under the same license. Auto-loading of 8390.o only in v2.2 - Paul G.
+ * Some of these names and comments originated from the Crynwr
+ * packet drivers, which are distributed under the GPL.
+ */
 
 #ifndef _8390_h
 #define _8390_h
@@ -16,9 +20,9 @@
 
 /* The 8390 specific per-packet-header format. */
 struct e8390_pkt_hdr {
-  unsigned char status; /* status */
-  unsigned char next;   /* pointer to next packet. */
-  unsigned short count; /* header + packet length in bytes */
+	unsigned char status; /* status */
+	unsigned char next;   /* pointer to next packet. */
+	unsigned short count; /* header + packet length in bytes */
 };
 
 #ifdef CONFIG_NET_POLL_CONTROLLER
@@ -32,7 +36,7 @@ void NS8390_init(struct net_device *dev, int startp);
 int ei_open(struct net_device *dev);
 int ei_close(struct net_device *dev);
 irqreturn_t ei_interrupt(int irq, void *dev_id);
-void ei_tx_timeout(struct net_device *dev);
+void ei_tx_timeout(struct net_device *dev, unsigned int txqueue);
 netdev_tx_t ei_start_xmit(struct sk_buff *skb, struct net_device *dev);
 void ei_set_multicast_list(struct net_device *dev);
 struct net_device_stats *ei_get_stats(struct net_device *dev);
@@ -50,7 +54,7 @@ void NS8390p_init(struct net_device *dev, int startp);
 int eip_open(struct net_device *dev);
 int eip_close(struct net_device *dev);
 irqreturn_t eip_interrupt(int irq, void *dev_id);
-void eip_tx_timeout(struct net_device *dev);
+void eip_tx_timeout(struct net_device *dev, unsigned int txqueue);
 netdev_tx_t eip_start_xmit(struct sk_buff *skb, struct net_device *dev);
 void eip_set_multicast_list(struct net_device *dev);
 struct net_device_stats *eip_get_stats(struct net_device *dev);
@@ -66,18 +70,24 @@ static inline struct net_device *alloc_eip_netdev(void)
 /* You have one of these per-board */
 struct ei_device {
 	const char *name;
-	void (*reset_8390)(struct net_device *);
-	void (*get_8390_hdr)(struct net_device *, struct e8390_pkt_hdr *, int);
-	void (*block_output)(struct net_device *, int, const unsigned char *, int);
-	void (*block_input)(struct net_device *, int, struct sk_buff *, int);
+	void (*reset_8390)(struct net_device *dev);
+	void (*get_8390_hdr)(struct net_device *dev,
+			     struct e8390_pkt_hdr *hdr, int ring_page);
+	void (*block_output)(struct net_device *dev, int count,
+			     const unsigned char *buf, int start_page);
+	void (*block_input)(struct net_device *dev, int count,
+			    struct sk_buff *skb, int ring_offset);
 	unsigned long rmem_start;
 	unsigned long rmem_end;
 	void __iomem *mem;
 	unsigned char mcfilter[8];
 	unsigned open:1;
-	unsigned word16:1;  		/* We have the 16-bit (vs 8-bit) version of the card. */
-	unsigned bigendian:1;		/* 16-bit big endian mode. Do NOT */
-					/* set this on random 8390 clones! */
+	unsigned word16:1;		/* We have the 16-bit (vs 8-bit)
+					 * version of the card.
+					 */
+	unsigned bigendian:1;		/* 16-bit big endian mode. Do NOT
+					 * set this on random 8390 clones!
+					 */
 	unsigned txing:1;		/* Transmit Active */
 	unsigned irqlock:1;		/* 8390's intrs disabled when '1'. */
 	unsigned dmaing:1;		/* Remote DMA Active */
@@ -115,12 +125,16 @@ struct ei_device {
 #define E8390_RXCONFIG		(ei_status.rxcr_base | 0x04)
 #define E8390_RXOFF		(ei_status.rxcr_base | 0x20)
 #else
-#define E8390_RXCONFIG		0x4	/* EN0_RXCR: broadcasts, no multicast,errors */
-#define E8390_RXOFF		0x20	/* EN0_RXCR: Accept no packets */
+/* EN0_RXCR: broadcasts, no multicast,errors */
+#define E8390_RXCONFIG		0x4
+/* EN0_RXCR: Accept no packets */
+#define E8390_RXOFF		0x20
 #endif
 
-#define E8390_TXCONFIG		0x00	/* EN0_TXCR: Normal transmit mode */
-#define E8390_TXOFF		0x02	/* EN0_TXCR: Transmitter off */
+/* EN0_TXCR: Normal transmit mode */
+#define E8390_TXCONFIG		0x00
+/* EN0_TXCR: Transmitter off */
+#define E8390_TXOFF		0x02
 
 
 /*  Register accessed at EN_CMD, the 8390 base addr.  */
@@ -134,17 +148,16 @@ struct ei_device {
 #define E8390_PAGE1	0x40	/* using the two high-order bits */
 #define E8390_PAGE2	0x80	/* Page 3 is invalid. */
 
-/*
- *	Only generate indirect loads given a machine that needs them.
- *      - removed AMIGA_PCMCIA from this list, handled as ISA io now
- *	- the _p for generates no delay by default 8390p.c overrides this.
+/* Only generate indirect loads given a machine that needs them.
+ * - removed AMIGA_PCMCIA from this list, handled as ISA io now
+ * - the _p for generates no delay by default 8390p.c overrides this.
  */
 
 #ifndef ei_inb
 #define ei_inb(_p)	inb(_p)
-#define ei_outb(_v,_p)	outb(_v,_p)
+#define ei_outb(_v, _p)	outb(_v, _p)
 #define ei_inb_p(_p)	inb(_p)
-#define ei_outb_p(_v,_p) outb(_v,_p)
+#define ei_outb_p(_v, _p) outb(_v, _p)
 #endif
 
 #ifndef EI_SHIFT
@@ -153,9 +166,9 @@ struct ei_device {
 
 #define E8390_CMD	EI_SHIFT(0x00)  /* The command register (for all pages) */
 /* Page 0 register offsets. */
-#define EN0_CLDALO	EI_SHIFT(0x01)	/* Low byte of current local dma addr  RD */
+#define EN0_CLDALO	EI_SHIFT(0x01)	/* Low byte of current local dma addr RD */
 #define EN0_STARTPG	EI_SHIFT(0x01)	/* Starting page of ring bfr WR */
-#define EN0_CLDAHI	EI_SHIFT(0x02)	/* High byte of current local dma addr  RD */
+#define EN0_CLDAHI	EI_SHIFT(0x02)	/* High byte of current local dma addr RD */
 #define EN0_STOPPG	EI_SHIFT(0x02)	/* Ending page +1 of ring bfr WR */
 #define EN0_BOUNDARY	EI_SHIFT(0x03)	/* Boundary page of ring bfr RD WR */
 #define EN0_TSR		EI_SHIFT(0x04)	/* Transmit status reg RD */
diff --git a/drivers/net/ethernet/8390/8390p.c b/drivers/net/ethernet/8390/8390p.c
index 46d2257c4430..6d429b11e9c6 100644
--- a/drivers/net/ethernet/8390/8390p.c
+++ b/drivers/net/ethernet/8390/8390p.c
@@ -1,3 +1,4 @@
+// SPDX-License-Identifier: GPL-2.0-only
 /* 8390 core for ISA devices needing bus delays */
 
 static const char version[] =
@@ -40,9 +41,9 @@ void eip_set_multicast_list(struct net_device *dev)
 }
 EXPORT_SYMBOL(eip_set_multicast_list);
 
-void eip_tx_timeout(struct net_device *dev)
+void eip_tx_timeout(struct net_device *dev, unsigned int txqueue)
 {
-	__ei_tx_timeout(dev);
+	__ei_tx_timeout(dev, txqueue);
 }
 EXPORT_SYMBOL(eip_tx_timeout);
 
@@ -101,4 +102,5 @@ static void __exit NS8390p_cleanup_module(void)
 
 module_init(NS8390p_init_module);
 module_exit(NS8390p_cleanup_module);
+MODULE_DESCRIPTION("National Semiconductor 8390 core for ISA driver");
 MODULE_LICENSE("GPL");
diff --git a/drivers/net/ethernet/8390/Kconfig b/drivers/net/ethernet/8390/Kconfig
index 29c3075bfb05..345f250781c6 100644
--- a/drivers/net/ethernet/8390/Kconfig
+++ b/drivers/net/ethernet/8390/Kconfig
@@ -1,25 +1,26 @@
+# SPDX-License-Identifier: GPL-2.0-only
 #
 # 8390 device configuration
 #
 
 config NET_VENDOR_8390
-	bool "National Semi-conductor 8390 devices"
+	bool "National Semiconductor 8390 devices"
 	default y
 	depends on NET_VENDOR_NATSEMI
-	---help---
+	help
 	  If you have a network (Ethernet) card belonging to this class, say Y.
 
 	  Note that the answer to this question doesn't directly affect the
 	  kernel: saying N will just cause the configurator to skip all
-	  the questions about Western Digital cards. If you say Y, you will be
-	  asked for your specific card in the following questions.
+	  the questions about National Semiconductor 8390 cards. If you say Y,
+	  you will be asked for your specific card in the following questions.
 
 if NET_VENDOR_8390
 
 config PCMCIA_AXNET
 	tristate "Asix AX88190 PCMCIA support"
-	depends on PCMCIA
-	---help---
+	depends on PCMCIA && HAS_IOPORT
+	help
 	  Say Y here if you intend to attach an Asix AX88190-based PCMCIA
 	  (PC-card) Fast Ethernet card to your computer.  These cards are
 	  nearly NE2000 compatible but need a separate driver due to a few
@@ -29,12 +30,12 @@ config PCMCIA_AXNET
 	  called axnet_cs.  If unsure, say N.
 
 config AX88796
-	tristate "ASIX AX88796 NE2000 clone support"
-	depends on (ARM || MIPS || SUPERH)
+	tristate "ASIX AX88796 NE2000 clone support" if !ZORRO
+	depends on (ARM || MIPS || SUPERH || ZORRO || COMPILE_TEST)
 	select CRC32
 	select PHYLIB
 	select MDIO_BITBANG
-	---help---
+	help
 	  AX88796 driver, using platform bus to provide
 	  chip detection and resources
 
@@ -42,14 +43,27 @@ config AX88796_93CX6
 	bool "ASIX AX88796 external 93CX6 eeprom support"
 	depends on AX88796
 	select EEPROM_93CX6
-	---help---
+	help
 	  Select this if your platform comes with an external 93CX6 eeprom.
 
+config XSURF100
+	tristate "Amiga XSurf 100 AX88796/NE2000 clone support"
+	depends on ZORRO
+	select AX88796
+	select AX88796B_PHY
+	help
+	  This driver is for the Individual Computers X-Surf 100 Ethernet
+	  card (based on the Asix AX88796 chip). If you have such a card,
+	  say Y. Otherwise, say N.
+
+	  To compile this driver as a module, choose M here: the module
+	  will be called xsurf100.
+
 config HYDRA
 	tristate "Hydra support"
 	depends on ZORRO
 	select CRC32
-	---help---
+	help
 	  If you have a Hydra Ethernet adapter, say Y. Otherwise, say N.
 
 	  To compile this driver as a module, choose M here: the module
@@ -59,7 +73,7 @@ config ARM_ETHERH
 	tristate "I-cubed EtherH/ANT EtherM support"
 	depends on ARM && ARCH_ACORN
 	select CRC32
-	---help---
+	help
 	  If you have an Acorn system with one of these network cards, you
 	  should say Y to this option if you wish to use it with Linux.
 
@@ -67,7 +81,7 @@ config MAC8390
 	tristate "Macintosh NS 8390 based ethernet cards"
 	depends on MAC
 	select CRC32
-	---help---
+	help
 	  If you want to include a driver to support Nubus or LC-PDS
 	  Ethernet cards using an NS8390 chipset or its equivalent, say Y.
 
@@ -75,7 +89,7 @@ config MCF8390
 	tristate "ColdFire NS8390 based Ethernet support"
 	depends on COLDFIRE
 	select CRC32
-	---help---
+	help
 	  This driver is for Ethernet devices using an NS8390-compatible
 	  chipset on many common ColdFire CPU based boards. Many of the older
 	  Freescale dev boards use this, and some other common boards like
@@ -87,10 +101,10 @@ config MCF8390
 
 config NE2000
 	tristate "NE2000/NE1000 support"
-	depends on (ISA || (Q40 && m) || M32R || MACH_TX49XX || \
-		    ATARI_ETHERNEC)
+	depends on (ISA || (Q40 && m) || MACH_TX49XX || ATARI_ETHERNEC)
+	select NETDEV_LEGACY_INIT if ISA
 	select CRC32
-	---help---
+	help
 	  If you have a network (Ethernet) card of this type, say Y here.
 	  Many Ethernet cards without a specific driver are compatible with
 	  the NE2000.
@@ -103,9 +117,9 @@ config NE2000
 
 config NE2K_PCI
 	tristate "PCI NE2000 and clones support (see help)"
-	depends on PCI
+	depends on PCI && HAS_IOPORT
 	select CRC32
-	---help---
+	help
 	  This driver is for NE2000 compatible PCI cards. It will not work
 	  with ISA NE2000 cards (they have their own driver, "NE2000/NE1000
 	  support" below). If you have a PCI NE2000 network (Ethernet) card,
@@ -123,7 +137,7 @@ config APNE
 	tristate "PCMCIA NE2000 support"
 	depends on AMIGA_PCMCIA
 	select CRC32
-	---help---
+	help
 	  If you have a PCMCIA NE2000 compatible adapter, say Y.  Otherwise,
 	  say N.
 
@@ -132,9 +146,9 @@ config APNE
 
 config PCMCIA_PCNET
 	tristate "NE2000 compatible PCMCIA support"
-	depends on PCMCIA
+	depends on PCMCIA && HAS_IOPORT
 	select CRC32
-	---help---
+	help
 	  Say Y here if you intend to attach an NE2000 compatible PCMCIA
 	  (PC-card) Ethernet or Fast Ethernet card to your computer.
 
@@ -145,7 +159,7 @@ config STNIC
 	tristate "National DP83902AV  support"
 	depends on SUPERH
 	select CRC32
-	---help---
+	help
 	  Support for cards based on the National Semiconductor DP83902AV
 	  ST-NIC Serial Network Interface Controller for Twisted Pair.  This
 	  is a 10Mbit/sec Ethernet controller.  Product overview and specs at
@@ -156,8 +170,9 @@ config STNIC
 config ULTRA
 	tristate "SMC Ultra support"
 	depends on ISA
+	select NETDEV_LEGACY_INIT
 	select CRC32
-	---help---
+	help
 	  If you have a network (Ethernet) card of this type, say Y here.
 
 	  Important: There have been many reports that, with some motherboards
@@ -173,8 +188,9 @@ config ULTRA
 config WD80x3
 	tristate "WD80*3 support"
 	depends on ISA
+	select NETDEV_LEGACY_INIT
 	select CRC32
-	---help---
+	help
 	  If you have a network (Ethernet) card of this type, say Y here.
 
 	  To compile this driver as a module, choose M here. The module
@@ -184,7 +200,7 @@ config ZORRO8390
 	tristate "Zorro NS8390-based Ethernet support"
 	depends on ZORRO
 	select CRC32
-	---help---
+	help
 	  This driver is for Zorro Ethernet cards using an NS8390-compatible
 	  chipset, like the Village Tronic Ariadne II and the Individual
 	  Computers X-Surf Ethernet cards. If you have such a card, say Y.
diff --git a/drivers/net/ethernet/8390/Makefile b/drivers/net/ethernet/8390/Makefile
index ff3b31894188..85c83c566ec6 100644
--- a/drivers/net/ethernet/8390/Makefile
+++ b/drivers/net/ethernet/8390/Makefile
@@ -1,3 +1,4 @@
+# SPDX-License-Identifier: GPL-2.0
 #
 # Makefile for the 8390 network device drivers.
 #
@@ -6,8 +7,8 @@ obj-$(CONFIG_MAC8390) += mac8390.o
 obj-$(CONFIG_APNE) += apne.o 8390.o
 obj-$(CONFIG_ARM_ETHERH) += etherh.o
 obj-$(CONFIG_AX88796) += ax88796.o
-obj-$(CONFIG_HYDRA) += hydra.o 8390.o
-obj-$(CONFIG_MCF8390) += mcf8390.o 8390.o
+obj-$(CONFIG_HYDRA) += hydra.o
+obj-$(CONFIG_MCF8390) += mcf8390.o
 obj-$(CONFIG_NE2000) += ne.o 8390p.o
 obj-$(CONFIG_NE2K_PCI) += ne2k-pci.o 8390.o
 obj-$(CONFIG_PCMCIA_AXNET) += axnet_cs.o 8390.o
@@ -15,4 +16,5 @@ obj-$(CONFIG_PCMCIA_PCNET) += pcnet_cs.o 8390.o
 obj-$(CONFIG_STNIC) += stnic.o 8390.o
 obj-$(CONFIG_ULTRA) += smc-ultra.o 8390.o
 obj-$(CONFIG_WD80x3) += wd.o 8390.o
-obj-$(CONFIG_ZORRO8390) += zorro8390.o 8390.o
+obj-$(CONFIG_XSURF100) += xsurf100.o
+obj-$(CONFIG_ZORRO8390) += zorro8390.o
diff --git a/drivers/net/ethernet/8390/apne.c b/drivers/net/ethernet/8390/apne.c
index c56ac9ebc08f..828edca8d30c 100644
--- a/drivers/net/ethernet/8390/apne.c
+++ b/drivers/net/ethernet/8390/apne.c
@@ -1,3 +1,4 @@
+// SPDX-License-Identifier: GPL-2.0-only
 /*
  * Amiga Linux/68k 8390 based PCMCIA Ethernet Driver for the Amiga 1200
  *
@@ -19,12 +20,6 @@
  *
  * ----------------------------------------------------------------------------
  *
- * This file is subject to the terms and conditions of the GNU General Public
- * License.  See the file COPYING in the main directory of the Linux
- * distribution for more details.
- *
- * ----------------------------------------------------------------------------
- *
  */
 
 
@@ -75,7 +70,6 @@
 #define NESM_STOP_PG	0x80	/* Last page +1 of RX ring */
 
 
-struct net_device * __init apne_probe(int unit);
 static int apne_probe1(struct net_device *dev, int ioaddr);
 
 static void apne_reset_8390(struct net_device *dev);
@@ -117,10 +111,10 @@ static const char version[] =
 static int apne_owned;	/* signal if card already owned */
 
 static u32 apne_msg_enable;
-module_param_named(msg_enable, apne_msg_enable, uint, (S_IRUSR|S_IRGRP|S_IROTH));
+module_param_named(msg_enable, apne_msg_enable, uint, 0444);
 MODULE_PARM_DESC(msg_enable, "Debug message level (see linux/netdevice.h for bitmap)");
 
-struct net_device * __init apne_probe(int unit)
+static struct net_device * __init apne_probe(void)
 {
 	struct net_device *dev;
 	struct ei_device *ei_local;
@@ -150,10 +144,6 @@ struct net_device * __init apne_probe(int unit)
 	dev = alloc_ei_netdev();
 	if (!dev)
 		return ERR_PTR(-ENOMEM);
-	if (unit >= 0) {
-		sprintf(dev->name, "eth%d", unit);
-		netdev_boot_setup_check(dev);
-	}
 	ei_local = netdev_priv(dev);
 	ei_local->msg_enable = apne_msg_enable;
 
@@ -325,8 +315,7 @@ static int __init apne_probe1(struct net_device *dev, int ioaddr)
     i = request_irq(dev->irq, apne_interrupt, IRQF_SHARED, DRV_NAME, dev);
     if (i) return i;
 
-    for (i = 0; i < ETH_ALEN; i++)
-	dev->dev_addr[i] = SA_prom[i];
+    eth_hw_addr_set(dev, SA_prom);
 
     pr_cont(" %pM\n", dev->dev_addr);
 
@@ -554,12 +543,11 @@ static irqreturn_t apne_interrupt(int irq, void *dev_id)
     return IRQ_HANDLED;
 }
 
-#ifdef MODULE
 static struct net_device *apne_dev;
 
 static int __init apne_module_init(void)
 {
-	apne_dev = apne_probe(-1);
+	apne_dev = apne_probe();
 	return PTR_ERR_OR_ZERO(apne_dev);
 }
 
@@ -579,7 +567,6 @@ static void __exit apne_module_exit(void)
 }
 module_init(apne_module_init);
 module_exit(apne_module_exit);
-#endif
 
 static int init_pcmcia(void)
 {
@@ -623,4 +610,5 @@ static int init_pcmcia(void)
 	return 1;
 }
 
+MODULE_DESCRIPTION("National Semiconductor 8390 Amiga PCMCIA ethernet driver");
 MODULE_LICENSE("GPL");
diff --git a/drivers/net/ethernet/8390/ax88796.c b/drivers/net/ethernet/8390/ax88796.c
index 05d9d3e2e92e..e1695d0fbd8b 100644
--- a/drivers/net/ethernet/8390/ax88796.c
+++ b/drivers/net/ethernet/8390/ax88796.c
@@ -1,3 +1,4 @@
+// SPDX-License-Identifier: GPL-2.0-only
 /* drivers/net/ethernet/8390/ax88796.c
  *
  * Copyright 2005,2007 Simtec Electronics
@@ -5,10 +6,6 @@
  *
  * Asix AX88796 10/100 Ethernet controller support
  *	Based on ne.c, by Donald Becker, et-al.
- *
- * This program is free software; you can redistribute it and/or modify
- * it under the terms of the GNU General Public License version 2 as
- * published by the Free Software Foundation.
  */
 
 #include <linux/module.h>
@@ -77,8 +74,6 @@ static unsigned char version[] = "ax88796.c: Copyright 2005,2007 Simtec Electron
 
 #define AX_GPOC_PPDSET	BIT(6)
 
-static u32 ax_msg_enable;
-
 /* device private data */
 
 struct ax_device {
@@ -106,6 +101,13 @@ static inline struct ax_device *to_ax_dev(struct net_device *dev)
 	return (struct ax_device *)(ei_local + 1);
 }
 
+void ax_NS8390_reinit(struct net_device *dev)
+{
+	ax_NS8390_init(dev, 1);
+}
+
+EXPORT_SYMBOL_GPL(ax_NS8390_reinit);
+
 /*
  * ax_initial_check
  *
@@ -165,6 +167,21 @@ static void ax_reset_8390(struct net_device *dev)
 	ei_outb(ENISR_RESET, addr + EN0_ISR);	/* Ack intr. */
 }
 
+/* Wrapper for __ei_interrupt for platforms that have a platform-specific
+ * way to find out whether the interrupt request might be caused by
+ * the ax88796 chip.
+ */
+static irqreturn_t ax_ei_interrupt_filtered(int irq, void *dev_id)
+{
+	struct net_device *dev = dev_id;
+	struct ax_device *ax = to_ax_dev(dev);
+	struct platform_device *pdev = to_platform_device(dev->dev.parent);
+
+	if (!ax->plat->check_irq(pdev))
+		return IRQ_NONE;
+
+	return ax_ei_interrupt(irq, dev_id);
+}
 
 static void ax_get_8390_hdr(struct net_device *dev, struct e8390_pkt_hdr *hdr,
 			    int ring_page)
@@ -364,9 +381,7 @@ static int ax_mii_probe(struct net_device *dev)
 		return ret;
 	}
 
-	/* mask with MAC supported features */
-	phy_dev->supported &= PHY_BASIC_FEATURES;
-	phy_dev->advertising = phy_dev->supported;
+	phy_set_max_speed(phy_dev, SPEED_100);
 
 	netdev_info(dev, "PHY driver [%s] (mii_bus:phy_addr=%s, irq=%d)\n",
 		    phy_dev->drv->name, phydev_name(phy_dev), phy_dev->irq);
@@ -389,6 +404,90 @@ static void ax_phy_switch(struct net_device *dev, int on)
 	ei_outb(reg_gpoc, ei_local->mem + EI_SHIFT(0x17));
 }
 
+static void ax_bb_mdc(struct mdiobb_ctrl *ctrl, int level)
+{
+	struct ax_device *ax = container_of(ctrl, struct ax_device, bb_ctrl);
+
+	if (level)
+		ax->reg_memr |= AX_MEMR_MDC;
+	else
+		ax->reg_memr &= ~AX_MEMR_MDC;
+
+	ei_outb(ax->reg_memr, ax->addr_memr);
+}
+
+static void ax_bb_dir(struct mdiobb_ctrl *ctrl, int output)
+{
+	struct ax_device *ax = container_of(ctrl, struct ax_device, bb_ctrl);
+
+	if (output)
+		ax->reg_memr &= ~AX_MEMR_MDIR;
+	else
+		ax->reg_memr |= AX_MEMR_MDIR;
+
+	ei_outb(ax->reg_memr, ax->addr_memr);
+}
+
+static void ax_bb_set_data(struct mdiobb_ctrl *ctrl, int value)
+{
+	struct ax_device *ax = container_of(ctrl, struct ax_device, bb_ctrl);
+
+	if (value)
+		ax->reg_memr |= AX_MEMR_MDO;
+	else
+		ax->reg_memr &= ~AX_MEMR_MDO;
+
+	ei_outb(ax->reg_memr, ax->addr_memr);
+}
+
+static int ax_bb_get_data(struct mdiobb_ctrl *ctrl)
+{
+	struct ax_device *ax = container_of(ctrl, struct ax_device, bb_ctrl);
+	int reg_memr = ei_inb(ax->addr_memr);
+
+	return reg_memr & AX_MEMR_MDI ? 1 : 0;
+}
+
+static const struct mdiobb_ops bb_ops = {
+	.owner = THIS_MODULE,
+	.set_mdc = ax_bb_mdc,
+	.set_mdio_dir = ax_bb_dir,
+	.set_mdio_data = ax_bb_set_data,
+	.get_mdio_data = ax_bb_get_data,
+};
+
+static int ax_mii_init(struct net_device *dev)
+{
+	struct platform_device *pdev = to_platform_device(dev->dev.parent);
+	struct ei_device *ei_local = netdev_priv(dev);
+	struct ax_device *ax = to_ax_dev(dev);
+	int err;
+
+	ax->bb_ctrl.ops = &bb_ops;
+	ax->addr_memr = ei_local->mem + AX_MEMR;
+	ax->mii_bus = alloc_mdio_bitbang(&ax->bb_ctrl);
+	if (!ax->mii_bus) {
+		err = -ENOMEM;
+		goto out;
+	}
+
+	ax->mii_bus->name = "ax88796_mii_bus";
+	ax->mii_bus->parent = dev->dev.parent;
+	snprintf(ax->mii_bus->id, MII_BUS_ID_SIZE, "%s-%x",
+		 pdev->name, pdev->id);
+
+	err = mdiobus_register(ax->mii_bus);
+	if (err)
+		goto out_free_mdio_bitbang;
+
+	return 0;
+
+ out_free_mdio_bitbang:
+	free_mdio_bitbang(ax->mii_bus);
+ out:
+	return err;
+}
+
 static int ax_open(struct net_device *dev)
 {
 	struct ax_device *ax = to_ax_dev(dev);
@@ -396,8 +495,16 @@ static int ax_open(struct net_device *dev)
 
 	netdev_dbg(dev, "open\n");
 
-	ret = request_irq(dev->irq, ax_ei_interrupt, ax->irqflags,
-			  dev->name, dev);
+	ret = ax_mii_init(dev);
+	if (ret)
+		goto failed_mii;
+
+	if (ax->plat->check_irq)
+		ret = request_irq(dev->irq, ax_ei_interrupt_filtered,
+				  ax->irqflags, dev->name, dev);
+	else
+		ret = request_irq(dev->irq, ax_ei_interrupt, ax->irqflags,
+				  dev->name, dev);
 	if (ret)
 		goto failed_request_irq;
 
@@ -423,6 +530,10 @@ static int ax_open(struct net_device *dev)
 	ax_phy_switch(dev, 0);
 	free_irq(dev->irq, dev);
  failed_request_irq:
+	/* unregister mdiobus */
+	mdiobus_unregister(ax->mii_bus);
+	free_mdio_bitbang(ax->mii_bus);
+ failed_mii:
 	return ret;
 }
 
@@ -442,6 +553,9 @@ static int ax_close(struct net_device *dev)
 	phy_disconnect(dev->phydev);
 
 	free_irq(dev->irq, dev);
+
+	mdiobus_unregister(ax->mii_bus);
+	free_mdio_bitbang(ax->mii_bus);
 	return 0;
 }
 
@@ -465,9 +579,9 @@ static void ax_get_drvinfo(struct net_device *dev,
 {
 	struct platform_device *pdev = to_platform_device(dev->dev.parent);
 
-	strlcpy(info->driver, DRV_NAME, sizeof(info->driver));
-	strlcpy(info->version, DRV_VERSION, sizeof(info->version));
-	strlcpy(info->bus_info, pdev->name, sizeof(info->bus_info));
+	strscpy(info->driver, DRV_NAME, sizeof(info->driver));
+	strscpy(info->version, DRV_VERSION, sizeof(info->version));
+	strscpy(info->bus_info, pdev->name, sizeof(info->bus_info));
 }
 
 static u32 ax_get_msglevel(struct net_device *dev)
@@ -528,7 +642,7 @@ static void ax_eeprom_register_write(struct eeprom_93cx6 *eeprom)
 static const struct net_device_ops ax_netdev_ops = {
 	.ndo_open		= ax_open,
 	.ndo_stop		= ax_close,
-	.ndo_do_ioctl		= ax_ioctl,
+	.ndo_eth_ioctl		= ax_ioctl,
 
 	.ndo_start_xmit		= ax_ei_start_xmit,
 	.ndo_tx_timeout		= ax_ei_tx_timeout,
@@ -541,92 +655,8 @@ static const struct net_device_ops ax_netdev_ops = {
 #endif
 };
 
-static void ax_bb_mdc(struct mdiobb_ctrl *ctrl, int level)
-{
-	struct ax_device *ax = container_of(ctrl, struct ax_device, bb_ctrl);
-
-	if (level)
-		ax->reg_memr |= AX_MEMR_MDC;
-	else
-		ax->reg_memr &= ~AX_MEMR_MDC;
-
-	ei_outb(ax->reg_memr, ax->addr_memr);
-}
-
-static void ax_bb_dir(struct mdiobb_ctrl *ctrl, int output)
-{
-	struct ax_device *ax = container_of(ctrl, struct ax_device, bb_ctrl);
-
-	if (output)
-		ax->reg_memr &= ~AX_MEMR_MDIR;
-	else
-		ax->reg_memr |= AX_MEMR_MDIR;
-
-	ei_outb(ax->reg_memr, ax->addr_memr);
-}
-
-static void ax_bb_set_data(struct mdiobb_ctrl *ctrl, int value)
-{
-	struct ax_device *ax = container_of(ctrl, struct ax_device, bb_ctrl);
-
-	if (value)
-		ax->reg_memr |= AX_MEMR_MDO;
-	else
-		ax->reg_memr &= ~AX_MEMR_MDO;
-
-	ei_outb(ax->reg_memr, ax->addr_memr);
-}
-
-static int ax_bb_get_data(struct mdiobb_ctrl *ctrl)
-{
-	struct ax_device *ax = container_of(ctrl, struct ax_device, bb_ctrl);
-	int reg_memr = ei_inb(ax->addr_memr);
-
-	return reg_memr & AX_MEMR_MDI ? 1 : 0;
-}
-
-static struct mdiobb_ops bb_ops = {
-	.owner = THIS_MODULE,
-	.set_mdc = ax_bb_mdc,
-	.set_mdio_dir = ax_bb_dir,
-	.set_mdio_data = ax_bb_set_data,
-	.get_mdio_data = ax_bb_get_data,
-};
-
 /* setup code */
 
-static int ax_mii_init(struct net_device *dev)
-{
-	struct platform_device *pdev = to_platform_device(dev->dev.parent);
-	struct ei_device *ei_local = netdev_priv(dev);
-	struct ax_device *ax = to_ax_dev(dev);
-	int err;
-
-	ax->bb_ctrl.ops = &bb_ops;
-	ax->addr_memr = ei_local->mem + AX_MEMR;
-	ax->mii_bus = alloc_mdio_bitbang(&ax->bb_ctrl);
-	if (!ax->mii_bus) {
-		err = -ENOMEM;
-		goto out;
-	}
-
-	ax->mii_bus->name = "ax88796_mii_bus";
-	ax->mii_bus->parent = dev->dev.parent;
-	snprintf(ax->mii_bus->id, MII_BUS_ID_SIZE, "%s-%x",
-		pdev->name, pdev->id);
-
-	err = mdiobus_register(ax->mii_bus);
-	if (err)
-		goto out_free_mdio_bitbang;
-
-	return 0;
-
- out_free_mdio_bitbang:
-	free_mdio_bitbang(ax->mii_bus);
- out:
-	return err;
-}
-
 static void ax_initial_setup(struct net_device *dev, struct ei_device *ei_local)
 {
 	void __iomem *ioaddr = ei_local->mem;
@@ -671,16 +701,22 @@ static int ax_init_dev(struct net_device *dev)
 	if (ax->plat->flags & AXFLG_HAS_EEPROM) {
 		unsigned char SA_prom[32];
 
+		ei_outb(6, ioaddr + EN0_RCNTLO);
+		ei_outb(0, ioaddr + EN0_RCNTHI);
+		ei_outb(0, ioaddr + EN0_RSARLO);
+		ei_outb(0, ioaddr + EN0_RSARHI);
+		ei_outb(E8390_RREAD + E8390_START, ioaddr + NE_CMD);
 		for (i = 0; i < sizeof(SA_prom); i += 2) {
 			SA_prom[i] = ei_inb(ioaddr + NE_DATAPORT);
 			SA_prom[i + 1] = ei_inb(ioaddr + NE_DATAPORT);
 		}
+		ei_outb(ENISR_RDC, ioaddr + EN0_ISR);	/* Ack intr. */
 
 		if (ax->plat->wordlength == 2)
 			for (i = 0; i < 16; i++)
 				SA_prom[i] = SA_prom[i+i];
 
-		memcpy(dev->dev_addr, SA_prom, ETH_ALEN);
+		eth_hw_addr_set(dev, SA_prom);
 	}
 
 #ifdef CONFIG_AX88796_93CX6
@@ -697,7 +733,7 @@ static int ax_init_dev(struct net_device *dev)
 				       (__le16 __force *)mac_addr,
 				       sizeof(mac_addr) >> 1);
 
-		memcpy(dev->dev_addr, mac_addr, ETH_ALEN);
+		eth_hw_addr_set(dev, mac_addr);
 	}
 #endif
 	if (ax->plat->wordlength == 2) {
@@ -712,16 +748,18 @@ static int ax_init_dev(struct net_device *dev)
 
 	/* load the mac-address from the device */
 	if (ax->plat->flags & AXFLG_MAC_FROMDEV) {
+		u8 addr[ETH_ALEN];
+
 		ei_outb(E8390_NODMA + E8390_PAGE1 + E8390_STOP,
 			ei_local->mem + E8390_CMD); /* 0x61 */
 		for (i = 0; i < ETH_ALEN; i++)
-			dev->dev_addr[i] =
-				ei_inb(ioaddr + EN1_PHYS_SHIFT(i));
+			addr[i] = ei_inb(ioaddr + EN1_PHYS_SHIFT(i));
+		eth_hw_addr_set(dev, addr);
 	}
 
 	if ((ax->plat->flags & AXFLG_MAC_FROMPLATFORM) &&
 	    ax->plat->mac_addr)
-		memcpy(dev->dev_addr, ax->plat->mac_addr, ETH_ALEN);
+		eth_hw_addr_set(dev, ax->plat->mac_addr);
 
 	if (!is_valid_ether_addr(dev->dev_addr)) {
 		eth_hw_addr_random(dev);
@@ -743,19 +781,20 @@ static int ax_init_dev(struct net_device *dev)
 #endif
 
 	ei_local->reset_8390 = &ax_reset_8390;
-	ei_local->block_input = &ax_block_input;
-	ei_local->block_output = &ax_block_output;
+	if (ax->plat->block_input)
+		ei_local->block_input = ax->plat->block_input;
+	else
+		ei_local->block_input = &ax_block_input;
+	if (ax->plat->block_output)
+		ei_local->block_output = ax->plat->block_output;
+	else
+		ei_local->block_output = &ax_block_output;
 	ei_local->get_8390_hdr = &ax_get_8390_hdr;
 	ei_local->priv = 0;
-	ei_local->msg_enable = ax_msg_enable;
 
 	dev->netdev_ops = &ax_netdev_ops;
 	dev->ethtool_ops = &ax_ethtool_ops;
 
-	ret = ax_mii_init(dev);
-	if (ret)
-		goto err_out;
-
 	ax_NS8390_init(dev, 0);
 
 	ret = register_netdev(dev);
@@ -772,7 +811,7 @@ static int ax_init_dev(struct net_device *dev)
 	return ret;
 }
 
-static int ax_remove(struct platform_device *pdev)
+static void ax_remove(struct platform_device *pdev)
 {
 	struct net_device *dev = platform_get_drvdata(pdev);
 	struct ei_device *ei_local = netdev_priv(dev);
@@ -780,7 +819,6 @@ static int ax_remove(struct platform_device *pdev)
 	struct resource *mem;
 
 	unregister_netdev(dev);
-	free_irq(dev->irq, dev);
 
 	iounmap(ei_local->mem);
 	mem = platform_get_resource(pdev, IORESOURCE_MEM, 0);
@@ -792,9 +830,8 @@ static int ax_remove(struct platform_device *pdev)
 		release_mem_region(mem->start, resource_size(mem));
 	}
 
+	platform_set_drvdata(pdev, NULL);
 	free_netdev(dev);
-
-	return 0;
 }
 
 /*
@@ -838,6 +875,9 @@ static int ax_probe(struct platform_device *pdev)
 	dev->irq = irq->start;
 	ax->irqflags = irq->flags & IRQF_TRIGGER_MASK;
 
+	if (irq->flags &  IORESOURCE_IRQ_SHAREABLE)
+		ax->irqflags |= IRQF_SHARED;
+
 	mem = platform_get_resource(pdev, IORESOURCE_MEM, 0);
 	if (!mem) {
 		dev_err(&pdev->dev, "no MEM specified\n");
@@ -922,6 +962,7 @@ static int ax_probe(struct platform_device *pdev)
 	release_mem_region(mem->start, mem_size);
 
  exit_mem:
+	platform_set_drvdata(pdev, NULL);
 	free_netdev(dev);
 
 	return ret;
diff --git a/drivers/net/ethernet/8390/axnet_cs.c b/drivers/net/ethernet/8390/axnet_cs.c
index 3da1fc539ef9..7c8213011b5c 100644
--- a/drivers/net/ethernet/8390/axnet_cs.c
+++ b/drivers/net/ethernet/8390/axnet_cs.c
@@ -1,3 +1,5 @@
+// SPDX-License-Identifier: GPL-1.0+
+
 /*======================================================================
 
     A PCMCIA ethernet driver for Asix AX88190-based cards
@@ -17,9 +19,7 @@
 
     Written 1992,1993 by Donald Becker.
     Copyright 1993 United States Government as represented by the
-    Director, National Security Agency.  This software may be used and
-    distributed according to the terms of the GNU General Public License,
-    incorporated herein by reference.
+    Director, National Security Agency.
     Donald Becker may be reached at becker@scyld.com
 
 ======================================================================*/
@@ -83,9 +83,9 @@ static netdev_tx_t axnet_start_xmit(struct sk_buff *skb,
 					  struct net_device *dev);
 static struct net_device_stats *get_stats(struct net_device *dev);
 static void set_multicast_list(struct net_device *dev);
-static void axnet_tx_timeout(struct net_device *dev);
+static void axnet_tx_timeout(struct net_device *dev, unsigned int txqueue);
 static irqreturn_t ei_irq_wrapper(int irq, void *dev_id);
-static void ei_watchdog(u_long arg);
+static void ei_watchdog(struct timer_list *t);
 static void axnet_reset_8390(struct net_device *dev);
 
 static int mdio_read(unsigned int addr, int phy_id, int loc);
@@ -104,7 +104,6 @@ static void AX88190_init(struct net_device *dev, int startp);
 static int ax_open(struct net_device *dev);
 static int ax_close(struct net_device *dev);
 static irqreturn_t ax_interrupt(int irq, void *dev_id);
-static u32 axnet_msg_enable;
 
 /*====================================================================*/
 
@@ -129,7 +128,7 @@ static inline struct axnet_dev *PRIV(struct net_device *dev)
 static const struct net_device_ops axnet_netdev_ops = {
 	.ndo_open 		= axnet_open,
 	.ndo_stop		= axnet_close,
-	.ndo_do_ioctl		= axnet_ioctl,
+	.ndo_eth_ioctl		= axnet_ioctl,
 	.ndo_start_xmit		= axnet_start_xmit,
 	.ndo_tx_timeout		= axnet_tx_timeout,
 	.ndo_get_stats		= get_stats,
@@ -151,7 +150,6 @@ static int axnet_probe(struct pcmcia_device *link)
 	return -ENOMEM;
 
     ei_local = netdev_priv(dev);
-    ei_local->msg_enable = axnet_msg_enable;
     spin_lock_init(&ei_local->page_lock);
 
     info = PRIV(dev);
@@ -189,6 +187,7 @@ static int get_prom(struct pcmcia_device *link)
 {
     struct net_device *dev = link->priv;
     unsigned int ioaddr = dev->base_addr;
+    u8 addr[ETH_ALEN];
     int i, j;
 
     /* This is based on drivers/net/ethernet/8390/ne.c */
@@ -222,9 +221,11 @@ static int get_prom(struct pcmcia_device *link)
 
     for (i = 0; i < 6; i += 2) {
 	j = inw(ioaddr + AXNET_DATAPORT);
-	dev->dev_addr[i] = j & 0xff;
-	dev->dev_addr[i+1] = j >> 8;
+	addr[i] = j & 0xff;
+	addr[i+1] = j >> 8;
     }
+    eth_hw_addr_set(dev, addr);
+
     return 1;
 } /* get_prom */
 
@@ -483,7 +484,7 @@ static int axnet_open(struct net_device *dev)
     link->open++;
 
     info->link_status = 0x00;
-    setup_timer(&info->watchdog, ei_watchdog, (u_long)dev);
+    timer_setup(&info->watchdog, ei_watchdog, 0);
     mod_timer(&info->watchdog, jiffies + HZ);
 
     return ax_open(dev);
@@ -503,7 +504,7 @@ static int axnet_close(struct net_device *dev)
     
     link->open--;
     netif_stop_queue(dev);
-    del_timer_sync(&info->watchdog);
+    timer_delete_sync(&info->watchdog);
 
     return 0;
 } /* axnet_close */
@@ -547,10 +548,10 @@ static irqreturn_t ei_irq_wrapper(int irq, void *dev_id)
     return ax_interrupt(irq, dev_id);
 }
 
-static void ei_watchdog(u_long arg)
+static void ei_watchdog(struct timer_list *t)
 {
-    struct net_device *dev = (struct net_device *)(arg);
-    struct axnet_dev *info = PRIV(dev);
+    struct axnet_dev *info = timer_container_of(info, t, watchdog);
+    struct net_device *dev = info->p_dev->priv;
     unsigned int nic_base = dev->base_addr;
     unsigned int mii_addr = nic_base + AXNET_MII_EEP;
     u_short link;
@@ -612,6 +613,7 @@ static int axnet_ioctl(struct net_device *dev, struct ifreq *rq, int cmd)
     switch (cmd) {
     case SIOCGMIIPHY:
 	data->phy_id = info->phy_id;
+	fallthrough;
     case SIOCGMIIREG:		/* Read MII PHY register. */
 	data->val_out = mdio_read(mii_addr, data->phy_id, data->reg_num & 0x1f);
 	return 0;
@@ -648,7 +650,6 @@ static void block_input(struct net_device *dev, int count,
 {
     unsigned int nic_base = dev->base_addr;
     struct ei_device *ei_local = netdev_priv(dev);
-    int xfer_count = count;
     char *buf = skb->data;
 
     if ((netif_msg_rx_status(ei_local)) && (count != 4))
@@ -658,9 +659,9 @@ static void block_input(struct net_device *dev, int count,
     outb_p(E8390_RREAD+E8390_START, nic_base + AXNET_CMD);
 
     insw(nic_base + AXNET_DATAPORT,buf,count>>1);
-    if (count & 0x01)
-	buf[count-1] = inb(nic_base + AXNET_DATAPORT), xfer_count++;
-
+    if (count & 0x01) {
+	buf[count-1] = inb(nic_base + AXNET_DATAPORT);
+    }
 }
 
 /*====================================================================*/
@@ -766,7 +767,7 @@ module_pcmcia_driver(axnet_cs_driver);
   Paul Gortmaker	: tweak ANK's above multicast changes a bit.
   Paul Gortmaker	: update packet statistics for v2.1.x
   Alan Cox		: support arbitrary stupid port mappings on the
-  			  68K Macintosh. Support >16bit I/O spaces
+			  68K Macintosh. Support >16bit I/O spaces
   Paul Gortmaker	: add kmod support for auto-loading of the 8390
 			  module by all drivers that require it.
   Alan Cox		: Spinlocking work, added 'BUG_83C690'
@@ -899,12 +900,13 @@ static int ax_close(struct net_device *dev)
 /**
  * axnet_tx_timeout - handle transmit time out condition
  * @dev: network device which has apparently fallen asleep
+ * @txqueue: unused
  *
  * Called by kernel when device never acknowledges a transmit has
  * completed (or failed) - i.e. never posted a Tx related interrupt.
  */
 
-static void axnet_tx_timeout(struct net_device *dev)
+static void axnet_tx_timeout(struct net_device *dev, unsigned int txqueue)
 {
 	long e8390_base = dev->base_addr;
 	struct ei_device *ei_local = netdev_priv(dev);
@@ -1089,7 +1091,7 @@ static irqreturn_t ax_interrupt(int irq, void *dev_id)
 	long e8390_base;
 	int interrupts, nr_serviced = 0, i;
 	struct ei_device *ei_local;
-    	int handled = 0;
+	int handled = 0;
 	unsigned long flags;
 
 	e8390_base = dev->base_addr;
@@ -1270,10 +1272,12 @@ static void ei_tx_intr(struct net_device *dev)
 			ei_local->txing = 1;
 			NS8390_trigger_send(dev, ei_local->tx2, ei_local->tx_start_page + 6);
 			netif_trans_update(dev);
-			ei_local->tx2 = -1,
+			ei_local->tx2 = -1;
 			ei_local->lasttx = 2;
+		} else {
+			ei_local->lasttx = 20;
+			ei_local->txing = 0;
 		}
-		else ei_local->lasttx = 20, ei_local->txing = 0;	
 	}
 	else if (ei_local->tx2 < 0) 
 	{
@@ -1289,9 +1293,10 @@ static void ei_tx_intr(struct net_device *dev)
 			netif_trans_update(dev);
 			ei_local->tx1 = -1;
 			ei_local->lasttx = 1;
+		} else {
+			ei_local->lasttx = 10;
+			ei_local->txing = 0;
 		}
-		else
-			ei_local->lasttx = 10, ei_local->txing = 0;
 	}
 //	else
 //		netdev_warn(dev, "unexpected TX-done interrupt, lasttx=%d\n",
@@ -1582,12 +1587,12 @@ static void do_set_multicast_list(struct net_device *dev)
 	}
 	outb_p(E8390_NODMA + E8390_PAGE0, e8390_base + E8390_CMD);
 
-  	if(dev->flags&IFF_PROMISC)
-  		outb_p(E8390_RXCONFIG | 0x58, e8390_base + EN0_RXCR);
+	if(dev->flags&IFF_PROMISC)
+		outb_p(E8390_RXCONFIG | 0x58, e8390_base + EN0_RXCR);
 	else if (dev->flags & IFF_ALLMULTI || !netdev_mc_empty(dev))
-  		outb_p(E8390_RXCONFIG | 0x48, e8390_base + EN0_RXCR);
-  	else
-  		outb_p(E8390_RXCONFIG | 0x40, e8390_base + EN0_RXCR);
+		outb_p(E8390_RXCONFIG | 0x48, e8390_base + EN0_RXCR);
+	else
+		outb_p(E8390_RXCONFIG | 0x40, e8390_base + EN0_RXCR);
 
 	outb_p(E8390_NODMA+E8390_PAGE0+E8390_START, e8390_base+E8390_CMD);
 }
diff --git a/drivers/net/ethernet/8390/etherh.c b/drivers/net/ethernet/8390/etherh.c
index 11cbf22ad201..e876fe52399a 100644
--- a/drivers/net/ethernet/8390/etherh.c
+++ b/drivers/net/ethernet/8390/etherh.c
@@ -1,12 +1,9 @@
+// SPDX-License-Identifier: GPL-2.0-only
 /*
  *  linux/drivers/acorn/net/etherh.c
  *
  *  Copyright (C) 2000-2002 Russell King
  *
- * This program is free software; you can redistribute it and/or modify
- * it under the terms of the GNU General Public License version 2 as
- * published by the Free Software Foundation.
- *
  * NS8390 I-cubed EtherH and ANT EtherM specific driver
  * Thanks to I-Cubed for information on their cards.
  * EtherM conversion (C) 1999 Chris Kemp and Tim Watterton
@@ -64,8 +61,6 @@ static char version[] =
 
 #include "lib8390.c"
 
-static u32 etherh_msg_enable;
-
 struct etherh_priv {
 	void __iomem	*ioc_fast;
 	void __iomem	*memc;
@@ -263,7 +258,7 @@ static int etherh_set_config(struct net_device *dev, struct ifmap *map)
 		 * media type, turn off automedia detection.
 		 */
 		dev->flags &= ~IFF_AUTOMEDIA;
-		dev->if_port = map->port;
+		WRITE_ONCE(dev->if_port, map->port);
 		break;
 
 	default:
@@ -502,18 +497,6 @@ etherh_close(struct net_device *dev)
 }
 
 /*
- * Initialisation
- */
-
-static void __init etherh_banner(void)
-{
-	static int version_printed;
-
-	if ((etherh_msg_enable & NETIF_MSG_DRV) && (version_printed++ == 0))
-		pr_info("%s", version);
-}
-
-/*
  * Read the ethernet address string from the on board rom.
  * This is an ascii string...
  */
@@ -572,32 +555,35 @@ static int __init etherm_addr(char *addr)
 
 static void etherh_get_drvinfo(struct net_device *dev, struct ethtool_drvinfo *info)
 {
-	strlcpy(info->driver, DRV_NAME, sizeof(info->driver));
-	strlcpy(info->version, DRV_VERSION, sizeof(info->version));
-	strlcpy(info->bus_info, dev_name(dev->dev.parent),
+	strscpy(info->driver, DRV_NAME, sizeof(info->driver));
+	strscpy(info->version, DRV_VERSION, sizeof(info->version));
+	strscpy(info->bus_info, dev_name(dev->dev.parent),
 		sizeof(info->bus_info));
 }
 
-static int etherh_get_settings(struct net_device *dev, struct ethtool_cmd *cmd)
+static int etherh_get_link_ksettings(struct net_device *dev,
+				     struct ethtool_link_ksettings *cmd)
 {
-	cmd->supported	= etherh_priv(dev)->supported;
-	ethtool_cmd_speed_set(cmd, SPEED_10);
-	cmd->duplex	= DUPLEX_HALF;
-	cmd->port	= dev->if_port == IF_PORT_10BASET ? PORT_TP : PORT_BNC;
-	cmd->autoneg	= (dev->flags & IFF_AUTOMEDIA ?
-			   AUTONEG_ENABLE : AUTONEG_DISABLE);
+	ethtool_convert_legacy_u32_to_link_mode(cmd->link_modes.supported,
+						etherh_priv(dev)->supported);
+	cmd->base.speed = SPEED_10;
+	cmd->base.duplex = DUPLEX_HALF;
+	cmd->base.port = dev->if_port == IF_PORT_10BASET ? PORT_TP : PORT_BNC;
+	cmd->base.autoneg = (dev->flags & IFF_AUTOMEDIA ? AUTONEG_ENABLE :
+							  AUTONEG_DISABLE);
 	return 0;
 }
 
-static int etherh_set_settings(struct net_device *dev, struct ethtool_cmd *cmd)
+static int etherh_set_link_ksettings(struct net_device *dev,
+				     const struct ethtool_link_ksettings *cmd)
 {
-	switch (cmd->autoneg) {
+	switch (cmd->base.autoneg) {
 	case AUTONEG_ENABLE:
 		dev->flags |= IFF_AUTOMEDIA;
 		break;
 
 	case AUTONEG_DISABLE:
-		switch (cmd->port) {
+		switch (cmd->base.port) {
 		case PORT_TP:
 			dev->if_port = IF_PORT_10BASET;
 			break;
@@ -636,12 +622,12 @@ static void etherh_set_msglevel(struct net_device *dev, u32 v)
 }
 
 static const struct ethtool_ops etherh_ethtool_ops = {
-	.get_settings	= etherh_get_settings,
-	.set_settings	= etherh_set_settings,
-	.get_drvinfo	= etherh_get_drvinfo,
-	.get_ts_info	= ethtool_op_get_ts_info,
-	.get_msglevel	= etherh_get_msglevel,
-	.set_msglevel	= etherh_set_msglevel,
+	.get_drvinfo		= etherh_get_drvinfo,
+	.get_ts_info		= ethtool_op_get_ts_info,
+	.get_msglevel		= etherh_get_msglevel,
+	.set_msglevel		= etherh_set_msglevel,
+	.get_link_ksettings	= etherh_get_link_ksettings,
+	.set_link_ksettings	= etherh_set_link_ksettings,
 };
 
 static const struct net_device_ops etherh_netdev_ops = {
@@ -669,10 +655,9 @@ etherh_probe(struct expansion_card *ec, const struct ecard_id *id)
 	struct ei_device *ei_local;
 	struct net_device *dev;
 	struct etherh_priv *eh;
+	u8 addr[ETH_ALEN];
 	int ret;
 
-	etherh_banner();
-
 	ret = ecard_request_resources(ec);
 	if (ret)
 		goto out;
@@ -740,12 +725,13 @@ etherh_probe(struct expansion_card *ec, const struct ecard_id *id)
 	spin_lock_init(&ei_local->page_lock);
 
 	if (ec->cid.product == PROD_ANT_ETHERM) {
-		etherm_addr(dev->dev_addr);
+		etherm_addr(addr);
 		ei_local->reg_offset = etherm_regoffsets;
 	} else {
-		etherh_addr(dev->dev_addr, ec);
+		etherh_addr(addr, ec);
 		ei_local->reg_offset = etherh_regoffsets;
 	}
+	eth_hw_addr_set(dev, addr);
 
 	ei_local->name          = dev->name;
 	ei_local->word16        = 1;
@@ -757,7 +743,6 @@ etherh_probe(struct expansion_card *ec, const struct ecard_id *id)
 	ei_local->block_output  = etherh_block_output;
 	ei_local->get_8390_hdr  = etherh_get_header;
 	ei_local->interface_num = 0;
-	ei_local->msg_enable = etherh_msg_enable;
 
 	etherh_reset(dev);
 	__NS8390_init(dev, 0);
diff --git a/drivers/net/ethernet/8390/hydra.c b/drivers/net/ethernet/8390/hydra.c
index 8ae249195301..fd9dcdc356e6 100644
--- a/drivers/net/ethernet/8390/hydra.c
+++ b/drivers/net/ethernet/8390/hydra.c
@@ -1,10 +1,8 @@
+// SPDX-License-Identifier: GPL-2.0-only
+
 /* New Hydra driver using generic 8390 core */
 /* Based on old hydra driver by Topi Kanerva (topi@susanna.oulu.fi) */
 
-/* This file is subject to the terms and conditions of the GNU General      */
-/* Public License.  See the file COPYING in the main directory of the       */
-/* Linux distribution for more details.                                     */
-
 /* Peter De Schrijver (p2@mind.be) */
 /* Oldenburg 2000 */
 
@@ -66,7 +64,6 @@ static void hydra_block_input(struct net_device *dev, int count,
 static void hydra_block_output(struct net_device *dev, int count,
 			       const unsigned char *buf, int start_page);
 static void hydra_remove_one(struct zorro_dev *z);
-static u32 hydra_msg_enable;
 
 static struct zorro_device_id hydra_zorro_tbl[] = {
     { ZORRO_PROD_HYDRA_SYSTEMS_AMIGANET },
@@ -117,9 +114,9 @@ static int hydra_init(struct zorro_dev *z)
     unsigned long ioaddr = board+HYDRA_NIC_BASE;
     const char name[] = "NE2000";
     int start_page, stop_page;
+    u8 macaddr[ETH_ALEN];
     int j;
     int err;
-    struct ei_device *ei_local;
 
     static u32 hydra_offsets[16] = {
 	0x00, 0x02, 0x04, 0x06, 0x08, 0x0a, 0x0c, 0x0e,
@@ -131,15 +128,14 @@ static int hydra_init(struct zorro_dev *z)
 	return -ENOMEM;
 
     for (j = 0; j < ETH_ALEN; j++)
-	dev->dev_addr[j] = *((u8 *)(board + HYDRA_ADDRPROM + 2*j));
+	macaddr[j] = *((u8 *)(board + HYDRA_ADDRPROM + 2*j));
+    eth_hw_addr_set(dev, macaddr);
 
     /* We must set the 8390 for word mode. */
     z_writeb(0x4b, ioaddr + NE_EN0_DCFG);
     start_page = NESM_START_PG;
     stop_page = NESM_STOP_PG;
 
-    ei_local = netdev_priv(dev);
-    ei_local->msg_enable = hydra_msg_enable;
     dev->base_addr = ioaddr;
     dev->irq = IRQ_AMIGA_PORTS;
 
@@ -274,4 +270,5 @@ static void __exit hydra_cleanup_module(void)
 module_init(hydra_init_module);
 module_exit(hydra_cleanup_module);
 
+MODULE_DESCRIPTION("Zorro-II Hydra 8390 ethernet driver");
 MODULE_LICENSE("GPL");
diff --git a/drivers/net/ethernet/8390/lib8390.c b/drivers/net/ethernet/8390/lib8390.c
index 60f8e2c8e726..84aeb8054304 100644
--- a/drivers/net/ethernet/8390/lib8390.c
+++ b/drivers/net/ethernet/8390/lib8390.c
@@ -1,3 +1,5 @@
+// SPDX-License-Identifier: GPL-1.0+
+
 /* 8390.c: A general NS8390 ethernet driver core for linux. */
 /*
 	Written 1992-94 by Donald Becker.
@@ -5,9 +7,6 @@
 	Copyright 1993 United States Government as represented by the
 	Director, National Security Agency.
 
-	This software may be used and distributed according to the terms
-	of the GNU General Public License, incorporated herein by reference.
-
 	The author may be reached as becker@scyld.com, or C/O
 	Scyld Computing Corporation
 	410 Severn Ave., Suite 210
@@ -50,6 +49,7 @@
 
   */
 
+#include <linux/build_bug.h>
 #include <linux/module.h>
 #include <linux/kernel.h>
 #include <linux/jiffies.h>
@@ -112,8 +112,10 @@ static void do_set_multicast_list(struct net_device *dev);
 static void __NS8390_init(struct net_device *dev, int startp);
 
 static unsigned version_printed;
-static u32 msg_enable;
-module_param(msg_enable, uint, (S_IRUSR|S_IRGRP|S_IROTH));
+static int msg_enable;
+static const int default_msg_level = (NETIF_MSG_DRV | NETIF_MSG_PROBE | NETIF_MSG_RX_ERR |
+				     NETIF_MSG_TX_ERR);
+module_param(msg_enable, int, 0444);
 MODULE_PARM_DESC(msg_enable, "Debug message level (see linux/netdevice.h for bitmap)");
 
 /*
@@ -251,7 +253,7 @@ static int __ei_close(struct net_device *dev)
  * completed (or failed) - i.e. never posted a Tx related interrupt.
  */
 
-static void __ei_tx_timeout(struct net_device *dev)
+static void __ei_tx_timeout(struct net_device *dev, unsigned int txqueue)
 {
 	unsigned long e8390_base = dev->base_addr;
 	struct ei_device *ei_local = netdev_priv(dev);
@@ -597,10 +599,12 @@ static void ei_tx_intr(struct net_device *dev)
 			ei_local->txing = 1;
 			NS8390_trigger_send(dev, ei_local->tx2, ei_local->tx_start_page + 6);
 			netif_trans_update(dev);
-			ei_local->tx2 = -1,
+			ei_local->tx2 = -1;
 			ei_local->lasttx = 2;
-		} else
-			ei_local->lasttx = 20, ei_local->txing = 0;
+		} else {
+			ei_local->lasttx = 20;
+			ei_local->txing = 0;
+		}
 	} else if (ei_local->tx2 < 0) {
 		if (ei_local->lasttx != 2  &&  ei_local->lasttx != -2)
 			pr_err("%s: bogus last_tx_buffer %d, tx2=%d\n",
@@ -612,8 +616,10 @@ static void ei_tx_intr(struct net_device *dev)
 			netif_trans_update(dev);
 			ei_local->tx1 = -1;
 			ei_local->lasttx = 1;
-		} else
-			ei_local->lasttx = 10, ei_local->txing = 0;
+		} else {
+			ei_local->lasttx = 10;
+			ei_local->txing = 0;
+		}
 	} /* else
 		netdev_warn(dev, "unexpected TX-done interrupt, lasttx=%d\n",
 			    ei_local->lasttx);
@@ -969,12 +975,14 @@ static void ethdev_setup(struct net_device *dev)
 {
 	struct ei_device *ei_local = netdev_priv(dev);
 
-	if ((msg_enable & NETIF_MSG_DRV) && (version_printed++ == 0))
-		pr_info("%s", version);
-
 	ether_setup(dev);
 
 	spin_lock_init(&ei_local->page_lock);
+
+	ei_local->msg_enable = netif_msg_init(msg_enable, default_msg_level);
+
+	if (netif_msg_drv(ei_local) && (version_printed++ == 0))
+		pr_info("%s", version);
 }
 
 /**
@@ -1012,8 +1020,7 @@ static void __NS8390_init(struct net_device *dev, int startp)
 	    ? (0x48 | ENDCFG_WTS | (ei_local->bigendian ? ENDCFG_BOS : 0))
 	    : 0x48;
 
-	if (sizeof(struct e8390_pkt_hdr) != 4)
-		panic("8390.c: header struct mispacked\n");
+	BUILD_BUG_ON(sizeof(struct e8390_pkt_hdr) != 4);
 	/* Follow National Semi's recommendations for initing the DP83902. */
 	ei_outb_p(E8390_NODMA+E8390_PAGE0+E8390_STOP, e8390_base+E8390_CMD); /* 0x21 */
 	ei_outb_p(endcfg, e8390_base + EN0_DCFG);	/* 0x48 or 0x49 */
diff --git a/drivers/net/ethernet/8390/mac8390.c b/drivers/net/ethernet/8390/mac8390.c
index 9497f18eaba0..4a0a095a1a8a 100644
--- a/drivers/net/ethernet/8390/mac8390.c
+++ b/drivers/net/ethernet/8390/mac8390.c
@@ -1,11 +1,9 @@
+// SPDX-License-Identifier: GPL-1.0+
 /* mac8390.c: New driver for 8390-based Nubus (or Nubus-alike)
    Ethernet cards on Linux */
 /* Based on the former daynaport.c driver, by Alan Cox.  Some code
    taken from or inspired by skeleton.c by Donald Becker, acenic.c by
-   Jes Sorensen, and ne2k-pci.c by Donald Becker and Paul Gortmaker.
-
-   This software may be used and distributed according to the terms of
-   the GNU Public License, incorporated herein by reference.  */
+   Jes Sorensen, and ne2k-pci.c by Donald Becker and Paul Gortmaker. */
 
 /* 2000-02-28: support added for Dayna and Kinetics cards by
    A.G.deWijn@phys.uu.nl */
@@ -123,7 +121,7 @@ enum mac8390_access {
 };
 
 extern int mac8390_memtest(struct net_device *dev);
-static int mac8390_initdev(struct net_device *dev, struct nubus_dev *ndev,
+static int mac8390_initdev(struct net_device *dev, struct nubus_board *board,
 			   enum mac8390_type type);
 
 static int mac8390_open(struct net_device *dev);
@@ -153,11 +151,6 @@ static void dayna_block_input(struct net_device *dev, int count,
 static void dayna_block_output(struct net_device *dev, int count,
 			       const unsigned char *buf, int start_page);
 
-#define memcpy_fromio(a, b, c)	memcpy((a), (void *)(b), (c))
-#define memcpy_toio(a, b, c)	memcpy((void *)(a), (b), (c))
-
-#define memcmp_withio(a, b, c)	memcmp((a), (void *)(b), (c))
-
 /* Slow Sane (16-bit chunk memory read/write) Cabletron uses this */
 static void slow_sane_get_8390_hdr(struct net_device *dev,
 				   struct e8390_pkt_hdr *hdr, int ring_page);
@@ -167,13 +160,12 @@ static void slow_sane_block_output(struct net_device *dev, int count,
 				   const unsigned char *buf, int start_page);
 static void word_memcpy_tocard(unsigned long tp, const void *fp, int count);
 static void word_memcpy_fromcard(void *tp, unsigned long fp, int count);
-static u32 mac8390_msg_enable;
 
-static enum mac8390_type __init mac8390_ident(struct nubus_dev *dev)
+static enum mac8390_type mac8390_ident(struct nubus_rsrc *fres)
 {
-	switch (dev->dr_sw) {
+	switch (fres->dr_sw) {
 	case NUBUS_DRSW_3COM:
-		switch (dev->dr_hw) {
+		switch (fres->dr_hw) {
 		case NUBUS_DRHW_APPLE_SONIC_NB:
 		case NUBUS_DRHW_APPLE_SONIC_LC:
 		case NUBUS_DRHW_SONNET:
@@ -181,10 +173,9 @@ static enum mac8390_type __init mac8390_ident(struct nubus_dev *dev)
 		default:
 			return MAC8390_APPLE;
 		}
-		break;
 
 	case NUBUS_DRSW_APPLE:
-		switch (dev->dr_hw) {
+		switch (fres->dr_hw) {
 		case NUBUS_DRHW_ASANTE_LC:
 			return MAC8390_NONE;
 		case NUBUS_DRHW_CABLETRON:
@@ -192,68 +183,69 @@ static enum mac8390_type __init mac8390_ident(struct nubus_dev *dev)
 		default:
 			return MAC8390_APPLE;
 		}
-		break;
 
 	case NUBUS_DRSW_ASANTE:
 		return MAC8390_ASANTE;
-		break;
 
 	case NUBUS_DRSW_TECHWORKS:
 	case NUBUS_DRSW_DAYNA2:
 	case NUBUS_DRSW_DAYNA_LC:
-		if (dev->dr_hw == NUBUS_DRHW_CABLETRON)
+		if (fres->dr_hw == NUBUS_DRHW_CABLETRON)
 			return MAC8390_CABLETRON;
 		else
 			return MAC8390_APPLE;
-		break;
 
 	case NUBUS_DRSW_FARALLON:
 		return MAC8390_FARALLON;
-		break;
 
 	case NUBUS_DRSW_KINETICS:
-		switch (dev->dr_hw) {
+		switch (fres->dr_hw) {
 		case NUBUS_DRHW_INTERLAN:
 			return MAC8390_INTERLAN;
 		default:
 			return MAC8390_KINETICS;
 		}
-		break;
 
 	case NUBUS_DRSW_DAYNA:
 		/*
 		 * These correspond to Dayna Sonic cards
 		 * which use the macsonic driver
 		 */
-		if (dev->dr_hw == NUBUS_DRHW_SMC9194 ||
-		    dev->dr_hw == NUBUS_DRHW_INTERLAN)
+		if (fres->dr_hw == NUBUS_DRHW_SMC9194 ||
+		    fres->dr_hw == NUBUS_DRHW_INTERLAN)
 			return MAC8390_NONE;
 		else
 			return MAC8390_DAYNA;
-		break;
 	}
 	return MAC8390_NONE;
 }
 
-static enum mac8390_access __init mac8390_testio(volatile unsigned long membase)
+static enum mac8390_access mac8390_testio(unsigned long membase)
 {
-	unsigned long outdata = 0xA5A0B5B0;
-	unsigned long indata =  0x00000000;
+	u32 outdata = 0xA5A0B5B0;
+	u32 indata = 0;
+
 	/* Try writing 32 bits */
-	memcpy_toio(membase, &outdata, 4);
-	/* Now compare them */
-	if (memcmp_withio(&outdata, membase, 4) == 0)
+	nubus_writel(outdata, membase);
+	/* Now read it back */
+	indata = nubus_readl(membase);
+	if (outdata == indata)
 		return ACCESS_32;
+
+	outdata = 0xC5C0D5D0;
+	indata = 0;
+
 	/* Write 16 bit output */
 	word_memcpy_tocard(membase, &outdata, 4);
 	/* Now read it back */
 	word_memcpy_fromcard(&indata, membase, 4);
 	if (outdata == indata)
 		return ACCESS_16;
+
 	return ACCESS_UNKNOWN;
 }
 
-static int __init mac8390_memsize(unsigned long membase)
+static int mac8390_memsize(unsigned long membase)
 {
 	unsigned long flags;
 	int i, j;
@@ -289,46 +281,47 @@ static int __init mac8390_memsize(unsigned long membase)
 	return i * 0x1000;
 }
 
-static bool __init mac8390_init(struct net_device *dev, struct nubus_dev *ndev,
-				enum mac8390_type cardtype)
+static bool mac8390_rsrc_init(struct net_device *dev,
+			      struct nubus_rsrc *fres,
+			      enum mac8390_type cardtype)
 {
+	struct nubus_board *board = fres->board;
 	struct nubus_dir dir;
 	struct nubus_dirent ent;
 	int offset;
 	volatile unsigned short *i;
+	u8 addr[ETH_ALEN];
 
-	printk_once(KERN_INFO pr_fmt("%s"), version);
-
-	dev->irq = SLOT2IRQ(ndev->board->slot);
+	dev->irq = SLOT2IRQ(board->slot);
 	/* This is getting to be a habit */
-	dev->base_addr = (ndev->board->slot_addr |
-			  ((ndev->board->slot & 0xf) << 20));
+	dev->base_addr = board->slot_addr | ((board->slot & 0xf) << 20);
 
 	/*
 	 * Get some Nubus info - we will trust the card's idea
 	 * of where its memory and registers are.
 	 */
 
-	if (nubus_get_func_dir(ndev, &dir) == -1) {
-		pr_err("%s: Unable to get Nubus functional directory for slot %X!\n",
-		       dev->name, ndev->board->slot);
+	if (nubus_get_func_dir(fres, &dir) == -1) {
+		dev_err(&board->dev,
+			"Unable to get Nubus functional directory\n");
 		return false;
 	}
 
 	/* Get the MAC address */
 	if (nubus_find_rsrc(&dir, NUBUS_RESID_MAC_ADDRESS, &ent) == -1) {
-		pr_info("%s: Couldn't get MAC address!\n", dev->name);
+		dev_info(&board->dev, "MAC address resource not found\n");
 		return false;
 	}
 
-	nubus_get_rsrc_mem(dev->dev_addr, &ent, 6);
+	nubus_get_rsrc_mem(addr, &ent, 6);
+	eth_hw_addr_set(dev, addr);
 
 	if (useresources[cardtype] == 1) {
 		nubus_rewinddir(&dir);
 		if (nubus_find_rsrc(&dir, NUBUS_RESID_MINOR_BASEOS,
 				    &ent) == -1) {
-			pr_err("%s: Memory offset resource for slot %X not found!\n",
-			       dev->name, ndev->board->slot);
+			dev_err(&board->dev,
+				"Memory offset resource not found\n");
 			return false;
 		}
 		nubus_get_rsrc_mem(&offset, &ent, 4);
@@ -338,8 +331,8 @@ static bool __init mac8390_init(struct net_device *dev, struct nubus_dev *ndev,
 		nubus_rewinddir(&dir);
 		if (nubus_find_rsrc(&dir, NUBUS_RESID_MINOR_LENGTH,
 				    &ent) == -1) {
-			pr_info("%s: Memory length resource for slot %X not found, probing\n",
-				dev->name, ndev->board->slot);
+			dev_info(&board->dev,
+				 "Memory length resource not found, probing\n");
 			offset = mac8390_memsize(dev->mem_start);
 		} else {
 			nubus_get_rsrc_mem(&offset, &ent, 4);
@@ -349,25 +342,25 @@ static bool __init mac8390_init(struct net_device *dev, struct nubus_dev *ndev,
 		switch (cardtype) {
 		case MAC8390_KINETICS:
 		case MAC8390_DAYNA: /* it's the same */
-			dev->base_addr = (int)(ndev->board->slot_addr +
+			dev->base_addr = (int)(board->slot_addr +
 					       DAYNA_8390_BASE);
-			dev->mem_start = (int)(ndev->board->slot_addr +
+			dev->mem_start = (int)(board->slot_addr +
 					       DAYNA_8390_MEM);
 			dev->mem_end = dev->mem_start +
 				       mac8390_memsize(dev->mem_start);
 			break;
 		case MAC8390_INTERLAN:
-			dev->base_addr = (int)(ndev->board->slot_addr +
+			dev->base_addr = (int)(board->slot_addr +
 					       INTERLAN_8390_BASE);
-			dev->mem_start = (int)(ndev->board->slot_addr +
+			dev->mem_start = (int)(board->slot_addr +
 					       INTERLAN_8390_MEM);
 			dev->mem_end = dev->mem_start +
 				       mac8390_memsize(dev->mem_start);
 			break;
 		case MAC8390_CABLETRON:
-			dev->base_addr = (int)(ndev->board->slot_addr +
+			dev->base_addr = (int)(board->slot_addr +
 					       CABLETRON_8390_BASE);
-			dev->mem_start = (int)(ndev->board->slot_addr +
+			dev->mem_start = (int)(board->slot_addr +
 					       CABLETRON_8390_MEM);
 			/* The base address is unreadable if 0x00
 			 * has been written to the command register
@@ -382,8 +375,8 @@ static bool __init mac8390_init(struct net_device *dev, struct nubus_dev *ndev,
 			break;
 
 		default:
-			pr_err("Card type %s is unsupported, sorry\n",
-			       ndev->board->name);
+			dev_err(&board->dev,
+				"No known base address for card type\n");
 			return false;
 		}
 	}
@@ -391,88 +384,82 @@ static bool __init mac8390_init(struct net_device *dev, struct nubus_dev *ndev,
 	return true;
 }
 
-struct net_device * __init mac8390_probe(int unit)
+static int mac8390_device_probe(struct nubus_board *board)
 {
 	struct net_device *dev;
-	struct nubus_dev *ndev = NULL;
 	int err = -ENODEV;
-	struct ei_device *ei_local;
-
-	static unsigned int slots;
-
-	enum mac8390_type cardtype;
-
-	/* probably should check for Nubus instead */
-
-	if (!MACH_IS_MAC)
-		return ERR_PTR(-ENODEV);
+	struct nubus_rsrc *fres;
+	enum mac8390_type cardtype = MAC8390_NONE;
 
 	dev = ____alloc_ei_netdev(0);
 	if (!dev)
-		return ERR_PTR(-ENOMEM);
+		return -ENOMEM;
 
-	if (unit >= 0)
-		sprintf(dev->name, "eth%d", unit);
+	SET_NETDEV_DEV(dev, &board->dev);
 
-	while ((ndev = nubus_find_type(NUBUS_CAT_NETWORK, NUBUS_TYPE_ETHERNET,
-				       ndev))) {
-		/* Have we seen it already? */
-		if (slots & (1 << ndev->board->slot))
+	for_each_board_func_rsrc(board, fres) {
+		if (fres->category != NUBUS_CAT_NETWORK ||
+		    fres->type != NUBUS_TYPE_ETHERNET)
 			continue;
-		slots |= 1 << ndev->board->slot;
 
-		cardtype = mac8390_ident(ndev);
+		cardtype = mac8390_ident(fres);
 		if (cardtype == MAC8390_NONE)
 			continue;
 
-		if (!mac8390_init(dev, ndev, cardtype))
-			continue;
-
-		/* Do the nasty 8390 stuff */
-		if (!mac8390_initdev(dev, ndev, cardtype))
+		if (mac8390_rsrc_init(dev, fres, cardtype))
 			break;
 	}
-
-	if (!ndev)
+	if (!fres)
 		goto out;
 
-	 ei_local = netdev_priv(dev);
-	 ei_local->msg_enable = mac8390_msg_enable;
+	err = mac8390_initdev(dev, board, cardtype);
+	if (err)
+		goto out;
 
 	err = register_netdev(dev);
 	if (err)
 		goto out;
-	return dev;
+
+	nubus_set_drvdata(board, dev);
+	return 0;
 
 out:
 	free_netdev(dev);
-	return ERR_PTR(err);
+	return err;
 }
 
-#ifdef MODULE
+static void mac8390_device_remove(struct nubus_board *board)
+{
+	struct net_device *dev = nubus_get_drvdata(board);
+
+	unregister_netdev(dev);
+	free_netdev(dev);
+}
+
+static struct nubus_driver mac8390_driver = {
+	.probe = mac8390_device_probe,
+	.remove = mac8390_device_remove,
+	.driver = {
+		.name = KBUILD_MODNAME,
+		.owner = THIS_MODULE,
+	}
+};
+
 MODULE_AUTHOR("David Huggins-Daines <dhd@debian.org> and others");
 MODULE_DESCRIPTION("Macintosh NS8390-based Nubus Ethernet driver");
 MODULE_LICENSE("GPL");
 
-static struct net_device *dev_mac8390;
-
-int __init init_module(void)
+static int __init mac8390_init(void)
 {
-	dev_mac8390 = mac8390_probe(-1);
-	if (IS_ERR(dev_mac8390)) {
-		pr_warn("mac8390: No card found\n");
-		return PTR_ERR(dev_mac8390);
-	}
-	return 0;
+	return nubus_driver_register(&mac8390_driver);
 }
+module_init(mac8390_init);
 
-void __exit cleanup_module(void)
+static void __exit mac8390_exit(void)
 {
-	unregister_netdev(dev_mac8390);
-	free_netdev(dev_mac8390);
+	nubus_driver_unregister(&mac8390_driver);
 }
-
-#endif /* MODULE */
+module_exit(mac8390_exit);
 
 static const struct net_device_ops mac8390_netdev_ops = {
 	.ndo_open 		= mac8390_open,
@@ -488,9 +475,8 @@ static const struct net_device_ops mac8390_netdev_ops = {
 #endif
 };
 
-static int __init mac8390_initdev(struct net_device *dev,
-				  struct nubus_dev *ndev,
-				  enum mac8390_type type)
+static int mac8390_initdev(struct net_device *dev, struct nubus_board *board,
+			   enum mac8390_type type)
 {
 	static u32 fwrd4_offsets[16] = {
 		0,      4,      8,      12,
@@ -541,7 +527,8 @@ static int __init mac8390_initdev(struct net_device *dev,
 	case MAC8390_APPLE:
 		switch (mac8390_testio(dev->mem_start)) {
 		case ACCESS_UNKNOWN:
-			pr_err("Don't know how to access card memory!\n");
+			dev_err(&board->dev,
+				"Don't know how to access card memory\n");
 			return -ENODEV;
 
 		case ACCESS_16:
@@ -607,21 +594,18 @@ static int __init mac8390_initdev(struct net_device *dev,
 		break;
 
 	default:
-		pr_err("Card type %s is unsupported, sorry\n",
-		       ndev->board->name);
+		dev_err(&board->dev, "Unsupported card type\n");
 		return -ENODEV;
 	}
 
 	__NS8390_init(dev, 0);
 
 	/* Good, done, now spit out some messages */
-	pr_info("%s: %s in slot %X (type %s)\n",
-		dev->name, ndev->board->name, ndev->board->slot,
-		cardname[type]);
-	pr_info("MAC %pM IRQ %d, %d KB shared memory at %#lx, %d-bit access.\n",
-		dev->dev_addr, dev->irq,
-		(unsigned int)(dev->mem_end - dev->mem_start) >> 10,
-		dev->mem_start, access_bitmode ? 32 : 16);
+	dev_info(&board->dev, "%s (type %s)\n", board->name, cardname[type]);
+	dev_info(&board->dev, "MAC %pM, IRQ %d, %d KB shared memory at %#lx, %d-bit access.\n",
+		 dev->dev_addr, dev->irq,
+		 (unsigned int)(dev->mem_end - dev->mem_start) >> 10,
+		 dev->mem_start, access_bitmode ? 32 : 16);
 	return 0;
 }
 
@@ -721,7 +705,7 @@ static void sane_get_8390_hdr(struct net_device *dev,
 			      struct e8390_pkt_hdr *hdr, int ring_page)
 {
 	unsigned long hdr_start = (ring_page - WD_START_PG)<<8;
-	memcpy_fromio(hdr, dev->mem_start + hdr_start, 4);
+	memcpy_fromio(hdr, (void __iomem *)dev->mem_start + hdr_start, 4);
 	/* Fix endianness */
 	hdr->count = swab16(hdr->count);
 }
@@ -735,13 +719,16 @@ static void sane_block_input(struct net_device *dev, int count,
 	if (xfer_start + count > ei_status.rmem_end) {
 		/* We must wrap the input move. */
 		int semi_count = ei_status.rmem_end - xfer_start;
-		memcpy_fromio(skb->data, dev->mem_start + xfer_base,
+		memcpy_fromio(skb->data,
+			      (void __iomem *)dev->mem_start + xfer_base,
 			      semi_count);
 		count -= semi_count;
-		memcpy_fromio(skb->data + semi_count, ei_status.rmem_start,
-			      count);
+		memcpy_fromio(skb->data + semi_count,
+			      (void __iomem *)ei_status.rmem_start, count);
 	} else {
-		memcpy_fromio(skb->data, dev->mem_start + xfer_base, count);
+		memcpy_fromio(skb->data,
+			      (void __iomem *)dev->mem_start + xfer_base,
+			      count);
 	}
 }
 
@@ -750,7 +737,7 @@ static void sane_block_output(struct net_device *dev, int count,
 {
 	long shmem = (start_page - WD_START_PG)<<8;
 
-	memcpy_toio(dev->mem_start + shmem, buf, count);
+	memcpy_toio((void __iomem *)dev->mem_start + shmem, buf, count);
 }
 
 /* dayna block input/output */
diff --git a/drivers/net/ethernet/8390/mcf8390.c b/drivers/net/ethernet/8390/mcf8390.c
index 4bb967bc879e..94ff8364cdf0 100644
--- a/drivers/net/ethernet/8390/mcf8390.c
+++ b/drivers/net/ethernet/8390/mcf8390.c
@@ -1,3 +1,4 @@
+// SPDX-License-Identifier: GPL-2.0-only
 /*
  *  Support for ColdFire CPU based boards using a NS8390 Ethernet device.
  *
@@ -5,9 +6,6 @@
  *
  *  (C) Copyright 2012,  Greg Ungerer <gerg@uclinux.org>
  *
- *  This file is subject to the terms and conditions of the GNU General Public
- *  License.  See the file COPYING in the main directory of the Linux
- *  distribution for more details.
  */
 
 #include <linux/module.h>
@@ -38,7 +36,6 @@ static const char version[] =
 
 #define NESM_START_PG	0x40	/* First page of TX buffer */
 #define NESM_STOP_PG	0x80	/* Last page +1 of RX ring */
-static u32 mcf8390_msg_enable;
 
 #ifdef NE2000_ODDOFFSET
 /*
@@ -375,8 +372,7 @@ static int mcf8390_init(struct net_device *dev)
 	if (ret)
 		return ret;
 
-	for (i = 0; i < ETH_ALEN; i++)
-		dev->dev_addr[i] = SA_prom[i];
+	eth_hw_addr_set(dev, SA_prom);
 
 	netdev_dbg(dev, "Found ethernet address: %pM\n", dev->dev_addr);
 
@@ -407,16 +403,13 @@ static int mcf8390_init(struct net_device *dev)
 static int mcf8390_probe(struct platform_device *pdev)
 {
 	struct net_device *dev;
-	struct ei_device *ei_local;
-	struct resource *mem, *irq;
+	struct resource *mem;
 	resource_size_t msize;
-	int ret;
+	int ret, irq;
 
-	irq = platform_get_resource(pdev, IORESOURCE_IRQ, 0);
-	if (irq == NULL) {
-		dev_err(&pdev->dev, "no IRQ specified?\n");
+	irq = platform_get_irq(pdev, 0);
+	if (irq < 0)
 		return -ENXIO;
-	}
 
 	mem = platform_get_resource(pdev, IORESOURCE_MEM, 0);
 	if (mem == NULL) {
@@ -435,10 +428,8 @@ static int mcf8390_probe(struct platform_device *pdev)
 
 	SET_NETDEV_DEV(dev, &pdev->dev);
 	platform_set_drvdata(pdev, dev);
-	ei_local = netdev_priv(dev);
-	ei_local->msg_enable = mcf8390_msg_enable;
 
-	dev->irq = irq->start;
+	dev->irq = irq;
 	dev->base_addr = mem->start;
 
 	ret = mcf8390_init(dev);
@@ -450,17 +441,15 @@ static int mcf8390_probe(struct platform_device *pdev)
 	return 0;
 }
 
-static int mcf8390_remove(struct platform_device *pdev)
+static void mcf8390_remove(struct platform_device *pdev)
 {
 	struct net_device *dev = platform_get_drvdata(pdev);
 	struct resource *mem;
 
 	unregister_netdev(dev);
 	mem = platform_get_resource(pdev, IORESOURCE_MEM, 0);
-	if (mem)
-		release_mem_region(mem->start, resource_size(mem));
+	release_mem_region(mem->start, resource_size(mem));
 	free_netdev(dev);
-	return 0;
 }
 
 static struct platform_driver mcf8390_drv = {
diff --git a/drivers/net/ethernet/8390/ne.c b/drivers/net/ethernet/8390/ne.c
index 66f47987e2a2..961019c32842 100644
--- a/drivers/net/ethernet/8390/ne.c
+++ b/drivers/net/ethernet/8390/ne.c
@@ -1,3 +1,4 @@
+// SPDX-License-Identifier: GPL-1.0+
 /* ne.c: A general non-shared-memory NS8390 ethernet driver for linux. */
 /*
     Written 1992-94 by Donald Becker.
@@ -5,9 +6,6 @@
     Copyright 1993 United States Government as represented by the
     Director, National Security Agency.
 
-    This software may be used and distributed according to the terms
-    of the GNU General Public License, incorporated herein by reference.
-
     The author may be reached as becker@scyld.com, or C/O
     Scyld Computing Corporation, 410 Severn Ave., Suite 210, Annapolis MD 21403
 
@@ -52,6 +50,7 @@ static const char version2[] =
 #include <linux/etherdevice.h>
 #include <linux/jiffies.h>
 #include <linux/platform_device.h>
+#include <net/Space.h>
 
 #include <asm/io.h>
 
@@ -77,7 +76,7 @@ static u32 ne_msg_enable;
 module_param_hw_array(io, int, ioport, NULL, 0);
 module_param_hw_array(irq, int, irq, NULL, 0);
 module_param_array(bad, int, NULL, 0);
-module_param_named(msg_enable, ne_msg_enable, uint, (S_IRUSR|S_IRGRP|S_IROTH));
+module_param_named(msg_enable, ne_msg_enable, uint, 0444);
 MODULE_PARM_DESC(io, "I/O base address(es),required");
 MODULE_PARM_DESC(irq, "IRQ number(s)");
 MODULE_PARM_DESC(bad, "Accept card(s) with bad signatures");
@@ -99,7 +98,7 @@ MODULE_LICENSE("GPL");
 that the ne2k probe is the last 8390 based probe to take place (as it
 is at boot) and so the probe will get confused by any other 8390 cards.
 ISA device autoprobes on a running machine are not recommended anyway. */
-#if !defined(MODULE) && (defined(CONFIG_ISA) || defined(CONFIG_M32R))
+#if !defined(MODULE) && defined(CONFIG_ISA)
 /* Do we need a portlist for the ISA auto-probe ? */
 #define NEEDS_PORTLIST
 #endif
@@ -164,10 +163,7 @@ bad_clone_list[] __initdata = {
 #define NESM_START_PG	0x40	/* First page of TX buffer */
 #define NESM_STOP_PG	0x80	/* Last page +1 of RX ring */
 
-#if defined(CONFIG_PLAT_MAPPI)
-#  define DCR_VAL 0x4b
-#elif defined(CONFIG_PLAT_OAKS32R)  || \
-   defined(CONFIG_MACH_TX49XX)
+#if defined(CONFIG_MACH_TX49XX)
 #  define DCR_VAL 0x48		/* 8-bit mode */
 #elif defined(CONFIG_ATARI)	/* 8-bit mode on Atari, normal on Q40 */
 #  define DCR_VAL (MACH_IS_ATARI ? 0x48 : 0x49)
@@ -422,12 +418,7 @@ static int __init ne_probe1(struct net_device *dev, unsigned long ioaddr)
 		stop_page  = NE1SM_STOP_PG;
 	}
 
-#if  defined(CONFIG_PLAT_MAPPI) || defined(CONFIG_PLAT_OAKS32R)
-	neX000 = ((SA_prom[14] == 0x57  &&  SA_prom[15] == 0x57)
-		|| (SA_prom[14] == 0x42 && SA_prom[15] == 0x42));
-#else
 	neX000 = (SA_prom[14] == 0x57  &&  SA_prom[15] == 0x57);
-#endif
 	ctron =  (SA_prom[0] == 0x00 && SA_prom[1] == 0x00 && SA_prom[2] == 0x1d);
 	copam =  (SA_prom[14] == 0x49 && SA_prom[15] == 0x00);
 
@@ -485,7 +476,7 @@ static int __init ne_probe1(struct net_device *dev, unsigned long ioaddr)
 		mdelay(10);		/* wait 10ms for interrupt to propagate */
 		outb_p(0x00, ioaddr + EN0_IMR); 		/* Mask it again. */
 		dev->irq = probe_irq_off(cookie);
-		if (netif_msg_probe(ei_local))
+		if (ne_msg_enable & NETIF_MSG_PROBE)
 			pr_cont(" autoirq is %d", dev->irq);
 	} else if (dev->irq == 2)
 		/* Fixup for users that don't know that IRQ 2 is really IRQ 9,
@@ -508,18 +499,7 @@ static int __init ne_probe1(struct net_device *dev, unsigned long ioaddr)
 
 	dev->base_addr = ioaddr;
 
-#ifdef CONFIG_PLAT_MAPPI
-	outb_p(E8390_NODMA + E8390_PAGE1 + E8390_STOP,
-		ioaddr + E8390_CMD); /* 0x61 */
-	for (i = 0; i < ETH_ALEN; i++) {
-		dev->dev_addr[i] = SA_prom[i]
-			= inb_p(ioaddr + EN1_PHYS_SHIFT(i));
-	}
-#else
-	for (i = 0; i < ETH_ALEN; i++) {
-		dev->dev_addr[i] = SA_prom[i];
-	}
-#endif
+	eth_hw_addr_set(dev, SA_prom);
 
 	pr_cont("%pM\n", dev->dev_addr);
 
@@ -727,7 +707,7 @@ static void ne_block_output(struct net_device *dev, int count,
 retry:
 #endif
 
-#ifdef NE8390_RW_BUGFIX
+#ifdef NE_RW_BUGFIX
 	/* Handle the read-before-write bug the same way as the
 	   Crynwr packet driver -- the NatSemi method doesn't work.
 	   Actually this doesn't always work either, but if you have
@@ -843,7 +823,7 @@ static int __init ne_drv_probe(struct platform_device *pdev)
 	return 0;
 }
 
-static int ne_drv_remove(struct platform_device *pdev)
+static void ne_drv_remove(struct platform_device *pdev)
 {
 	struct net_device *dev = platform_get_drvdata(pdev);
 
@@ -862,7 +842,6 @@ static int ne_drv_remove(struct platform_device *pdev)
 		release_region(dev->base_addr, NE_IO_EXTENT);
 		free_netdev(dev);
 	}
-	return 0;
 }
 
 /* Remove unused devices or all if true. */
@@ -939,13 +918,16 @@ static void __init ne_add_devices(void)
 	}
 }
 
-#ifdef MODULE
-int __init init_module(void)
+static int __init ne_init(void)
 {
 	int retval;
-	ne_add_devices();
+
+	if (IS_MODULE(CONFIG_NE2000))
+		ne_add_devices();
+
 	retval = platform_driver_probe(&ne_driver, ne_drv_probe);
-	if (retval) {
+
+	if (IS_MODULE(CONFIG_NE2000) && retval) {
 		if (io[0] == 0)
 			pr_notice("ne.c: You must supply \"io=0xNNN\""
 			       " value(s) for ISA cards.\n");
@@ -957,17 +939,9 @@ int __init init_module(void)
 	ne_loop_rm_unreg(0);
 	return retval;
 }
-#else /* MODULE */
-static int __init ne_init(void)
-{
-	int retval = platform_driver_probe(&ne_driver, ne_drv_probe);
-
-	/* Unregister unused platform_devices. */
-	ne_loop_rm_unreg(0);
-	return retval;
-}
 module_init(ne_init);
 
+#if !defined(MODULE) && defined(CONFIG_NETDEV_LEGACY_INIT)
 struct net_device * __init ne_probe(int unit)
 {
 	int this_dev;
@@ -1008,7 +982,7 @@ struct net_device * __init ne_probe(int unit)
 
 	return ERR_PTR(-ENODEV);
 }
-#endif /* MODULE */
+#endif
 
 static void __exit ne_exit(void)
 {
diff --git a/drivers/net/ethernet/8390/ne2k-pci.c b/drivers/net/ethernet/8390/ne2k-pci.c
index 1bdea746926c..1a34da07c0db 100644
--- a/drivers/net/ethernet/8390/ne2k-pci.c
+++ b/drivers/net/ethernet/8390/ne2k-pci.c
@@ -1,39 +1,44 @@
-/* ne2k-pci.c: A NE2000 clone on PCI bus driver for Linux. */
-/*
-	A Linux device driver for PCI NE2000 clones.
-
-	Authors and other copyright holders:
-	1992-2000 by Donald Becker, NE2000 core and various modifications.
-	1995-1998 by Paul Gortmaker, core modifications and PCI support.
-	Copyright 1993 assigned to the United States Government as represented
-	by the Director, National Security Agency.
-
-	This software may be used and distributed according to the terms of
-	the GNU General Public License (GPL), incorporated herein by reference.
-	Drivers based on or derived from this code fall under the GPL and must
-	retain the authorship, copyright and license notice.  This file is not
-	a complete program and may only be used when the entire operating
-	system is licensed under the GPL.
-
-	The author may be reached as becker@scyld.com, or C/O
-	Scyld Computing Corporation
-	410 Severn Ave., Suite 210
-	Annapolis MD 21403
-
-	Issues remaining:
-	People are making PCI ne2000 clones! Oh the horror, the horror...
-	Limited full-duplex support.
-*/
+// SPDX-License-Identifier: GPL-1.0+
+/* A Linux device driver for PCI NE2000 clones.
+ *
+ * Authors and other copyright holders:
+ * 1992-2000 by Donald Becker, NE2000 core and various modifications.
+ * 1995-1998 by Paul Gortmaker, core modifications and PCI support.
+ * Copyright 1993 assigned to the United States Government as represented
+ * by the Director, National Security Agency.
+ *
+ * This software may be used and distributed according to the terms of
+ * the GNU General Public License (GPL), incorporated herein by reference.
+ * Drivers based on or derived from this code fall under the GPL and must
+ * retain the authorship, copyright and license notice.  This file is not
+ * a complete program and may only be used when the entire operating
+ * system is licensed under the GPL.
+ *
+ * The author may be reached as becker@scyld.com, or C/O
+ * Scyld Computing Corporation
+ * 410 Severn Ave., Suite 210
+ * Annapolis MD 21403
+ *
+ * Issues remaining:
+ * People are making PCI NE2000 clones! Oh the horror, the horror...
+ * Limited full-duplex support.
+ */
 
 #define DRV_NAME	"ne2k-pci"
+#define DRV_DESCRIPTION	"PCI NE2000 clone driver"
+#define DRV_AUTHOR	"Donald Becker / Paul Gortmaker"
 #define DRV_VERSION	"1.03"
 #define DRV_RELDATE	"9/22/2003"
 
+#define pr_fmt(fmt) KBUILD_MODNAME ": " fmt
 
 /* The user-configurable values.
-   These may be modified when a driver module is loaded.*/
+ * These may be modified when a driver module is loaded.
+ */
+
+/* More are supported, limit only on options */
+#define MAX_UNITS 8
 
-#define MAX_UNITS 8				/* More are supported, limit only on options */
 /* Used to pass the full-duplex flag, etc. */
 static int full_duplex[MAX_UNITS];
 static int options[MAX_UNITS];
@@ -52,38 +57,36 @@ static int options[MAX_UNITS];
 #include <linux/netdevice.h>
 #include <linux/etherdevice.h>
 
-#include <asm/io.h>
+#include <linux/io.h>
 #include <asm/irq.h>
 #include <linux/uaccess.h>
 
 #include "8390.h"
 
-static u32 ne2k_msg_enable;
+static int ne2k_msg_enable;
 
-/* These identify the driver base version and may not be removed. */
-static const char version[] =
-	KERN_INFO DRV_NAME ".c:v" DRV_VERSION " " DRV_RELDATE
-	" D. Becker/P. Gortmaker\n";
+static const int default_msg_level = (NETIF_MSG_DRV | NETIF_MSG_PROBE |
+				      NETIF_MSG_RX_ERR | NETIF_MSG_TX_ERR);
 
 #if defined(__powerpc__)
 #define inl_le(addr)  le32_to_cpu(inl(addr))
 #define inw_le(addr)  le16_to_cpu(inw(addr))
 #endif
 
-#define PFX DRV_NAME ": "
-
-MODULE_AUTHOR("Donald Becker / Paul Gortmaker");
-MODULE_DESCRIPTION("PCI NE2000 clone driver");
+MODULE_AUTHOR(DRV_AUTHOR);
+MODULE_DESCRIPTION(DRV_DESCRIPTION);
+MODULE_VERSION(DRV_VERSION);
 MODULE_LICENSE("GPL");
 
-module_param_named(msg_enable, ne2k_msg_enable, uint, (S_IRUSR|S_IRGRP|S_IROTH));
+module_param_named(msg_enable, ne2k_msg_enable, int, 0444);
 module_param_array(options, int, NULL, 0);
 module_param_array(full_duplex, int, NULL, 0);
 MODULE_PARM_DESC(msg_enable, "Debug message level (see linux/netdevice.h for bitmap)");
 MODULE_PARM_DESC(options, "Bit 5: full duplex");
 MODULE_PARM_DESC(full_duplex, "full duplex setting(s) (1)");
 
-/* Some defines that people can play with if so inclined. */
+/* Some defines that people can play with if so inclined.
+ */
 
 /* Use 32 bit data-movement operations instead of 16 bit. */
 #define USE_LONGIO
@@ -91,14 +94,18 @@ MODULE_PARM_DESC(full_duplex, "full duplex setting(s) (1)");
 /* Do we implement the read before write bugfix ? */
 /* #define NE_RW_BUGFIX */
 
-/* Flags.  We rename an existing ei_status field to store flags! */
-/* Thus only the low 8 bits are usable for non-init-time flags. */
+/* Flags.  We rename an existing ei_status field to store flags!
+ * Thus only the low 8 bits are usable for non-init-time flags.
+ */
 #define ne2k_flags reg0
+
 enum {
-	ONLY_16BIT_IO=8, ONLY_32BIT_IO=4,	/* Chip can do only 16/32-bit xfers. */
-	FORCE_FDX=0x20,						/* User override. */
-	REALTEK_FDX=0x40, HOLTEK_FDX=0x80,
-	STOP_PG_0x60=0x100,
+	/* Chip can do only 16/32-bit xfers. */
+	ONLY_16BIT_IO = 8, ONLY_32BIT_IO = 4,
+	/* User override. */
+	FORCE_FDX = 0x20,
+	REALTEK_FDX = 0x40, HOLTEK_FDX = 0x80,
+	STOP_PG_0x60 = 0x100,
 };
 
 enum ne2k_pci_chipsets {
@@ -120,7 +127,7 @@ static struct {
 	char *name;
 	int flags;
 } pci_clone_list[] = {
-	{"RealTek RTL-8029", REALTEK_FDX},
+	{"RealTek RTL-8029(AS)", REALTEK_FDX},
 	{"Winbond 89C940", 0},
 	{"Compex RL2000", 0},
 	{"KTI ET32P2", 0},
@@ -149,13 +156,14 @@ static const struct pci_device_id ne2k_pci_tbl[] = {
 	{ 0x8c4a, 0x1980, PCI_ANY_ID, PCI_ANY_ID, 0, 0, CH_Winbond_89C940_8c4a },
 	{ 0, }
 };
+
 MODULE_DEVICE_TABLE(pci, ne2k_pci_tbl);
 
 
 /* ---- No user-serviceable parts below ---- */
 
 #define NE_BASE	 (dev->base_addr)
-#define NE_CMD	 	0x00
+#define NE_CMD		0x00
 #define NE_DATAPORT	0x10	/* NatSemi-defined port window offset. */
 #define NE_RESET	0x1f	/* Issue a read to reset, a write to clear. */
 #define NE_IO_EXTENT	0x20
@@ -168,37 +176,27 @@ static int ne2k_pci_open(struct net_device *dev);
 static int ne2k_pci_close(struct net_device *dev);
 
 static void ne2k_pci_reset_8390(struct net_device *dev);
-static void ne2k_pci_get_8390_hdr(struct net_device *dev, struct e8390_pkt_hdr *hdr,
-			  int ring_page);
+static void ne2k_pci_get_8390_hdr(struct net_device *dev,
+				  struct e8390_pkt_hdr *hdr, int ring_page);
 static void ne2k_pci_block_input(struct net_device *dev, int count,
-			  struct sk_buff *skb, int ring_offset);
+				 struct sk_buff *skb, int ring_offset);
 static void ne2k_pci_block_output(struct net_device *dev, const int count,
-		const unsigned char *buf, const int start_page);
+				  const unsigned char *buf,
+				  const int start_page);
 static const struct ethtool_ops ne2k_pci_ethtool_ops;
 
 
-
-/* There is no room in the standard 8390 structure for extra info we need,
-   so we build a meta/outer-wrapper structure.. */
-struct ne2k_pci_card {
-	struct net_device *dev;
-	struct pci_dev *pci_dev;
-};
-
-
-
-/*
-  NEx000-clone boards have a Station Address (SA) PROM (SAPROM) in the packet
-  buffer memory space.  By-the-spec NE2000 clones have 0x57,0x57 in bytes
-  0x0e,0x0f of the SAPROM, while other supposed NE2000 clones must be
-  detected by their SA prefix.
-
-  Reading the SAPROM from a word-wide card with the 8390 set in byte-wide
-  mode results in doubled values, which can be detected and compensated for.
-
-  The probe is also responsible for initializing the card and filling
-  in the 'dev' and 'ei_status' structures.
-*/
+/* NEx000-clone boards have a Station Address (SA) PROM (SAPROM) in the packet
+ * buffer memory space.  By-the-spec NE2000 clones have 0x57,0x57 in bytes
+ * 0x0e,0x0f of the SAPROM, while other supposed NE2000 clones must be
+ * detected by their SA prefix.
+ *
+ * Reading the SAPROM from a word-wide card with the 8390 set in byte-wide
+ * mode results in doubled values, which can be detected and compensated for.
+ *
+ * The probe is also responsible for initializing the card and filling
+ * in the 'dev' and 'ei_status' structures.
+ */
 
 static const struct net_device_ops ne2k_netdev_ops = {
 	.ndo_open		= ne2k_pci_open,
@@ -208,7 +206,7 @@ static const struct net_device_ops ne2k_netdev_ops = {
 	.ndo_get_stats		= ei_get_stats,
 	.ndo_set_rx_mode	= ei_set_multicast_list,
 	.ndo_validate_addr	= eth_validate_addr,
-	.ndo_set_mac_address 	= eth_mac_addr,
+	.ndo_set_mac_address	= eth_mac_addr,
 #ifdef CONFIG_NET_POLL_CONTROLLER
 	.ndo_poll_controller = ei_poll,
 #endif
@@ -227,28 +225,21 @@ static int ne2k_pci_init_one(struct pci_dev *pdev,
 	int flags = pci_clone_list[chip_idx].flags;
 	struct ei_device *ei_local;
 
-/* when built into the kernel, we only print version if device is found */
-#ifndef MODULE
-	static int printed_version;
-	if (!printed_version++)
-		printk(version);
-#endif
-
 	fnd_cnt++;
 
-	i = pci_enable_device (pdev);
+	i = pci_enable_device(pdev);
 	if (i)
 		return i;
 
-	ioaddr = pci_resource_start (pdev, 0);
+	ioaddr = pci_resource_start(pdev, 0);
 	irq = pdev->irq;
 
-	if (!ioaddr || ((pci_resource_flags (pdev, 0) & IORESOURCE_IO) == 0)) {
+	if (!ioaddr || ((pci_resource_flags(pdev, 0) & IORESOURCE_IO) == 0)) {
 		dev_err(&pdev->dev, "no I/O resource at PCI BAR #0\n");
 		goto err_out;
 	}
 
-	if (request_region (ioaddr, NE_IO_EXTENT, DRV_NAME) == NULL) {
+	if (!request_region(ioaddr, NE_IO_EXTENT, DRV_NAME)) {
 		dev_err(&pdev->dev, "I/O resource 0x%x @ 0x%lx busy\n",
 			NE_IO_EXTENT, ioaddr);
 		goto err_out;
@@ -261,14 +252,17 @@ static int ne2k_pci_init_one(struct pci_dev *pdev,
 	/* Do a preliminary verification that we have a 8390. */
 	{
 		int regd;
-		outb(E8390_NODMA+E8390_PAGE1+E8390_STOP, ioaddr + E8390_CMD);
+
+		outb(E8390_NODMA + E8390_PAGE1 + E8390_STOP, ioaddr + E8390_CMD);
 		regd = inb(ioaddr + 0x0d);
 		outb(0xff, ioaddr + 0x0d);
-		outb(E8390_NODMA+E8390_PAGE0, ioaddr + E8390_CMD);
-		inb(ioaddr + EN0_COUNTER0); /* Clear the counter by reading. */
+		outb(E8390_NODMA + E8390_PAGE0, ioaddr + E8390_CMD);
+		/* Clear the counter by reading. */
+		inb(ioaddr + EN0_COUNTER0);
 		if (inb(ioaddr + EN0_COUNTER0) != 0) {
 			outb(reg0, ioaddr);
-			outb(regd, ioaddr + 0x0d);	/* Restore the old values. */
+			/*  Restore the old values. */
+			outb(regd, ioaddr + 0x0d);
 			goto err_out_free_res;
 		}
 	}
@@ -281,7 +275,7 @@ static int ne2k_pci_init_one(struct pci_dev *pdev,
 	}
 	dev->netdev_ops = &ne2k_netdev_ops;
 	ei_local = netdev_priv(dev);
-	ei_local->msg_enable = ne2k_msg_enable;
+	ei_local->msg_enable = netif_msg_init(ne2k_msg_enable, default_msg_level);
 
 	SET_NETDEV_DEV(dev, &pdev->dev);
 
@@ -291,9 +285,9 @@ static int ne2k_pci_init_one(struct pci_dev *pdev,
 
 		outb(inb(ioaddr + NE_RESET), ioaddr + NE_RESET);
 
-		/* This looks like a horrible timing loop, but it should never take
-		   more than a few cycles.
-		*/
+		/* This looks like a horrible timing loop, but it should never
+		 * take more than a few cycles.
+		 */
 		while ((inb(ioaddr + EN0_ISR) & ENISR_RESET) == 0)
 			/* Limit wait: '2' avoids jiffy roll-over. */
 			if (jiffies - reset_start_time > 2) {
@@ -301,42 +295,53 @@ static int ne2k_pci_init_one(struct pci_dev *pdev,
 					"Card failure (no reset ack).\n");
 				goto err_out_free_netdev;
 			}
-
-		outb(0xff, ioaddr + EN0_ISR);		/* Ack all intr. */
+		/* Ack all intr. */
+		outb(0xff, ioaddr + EN0_ISR);
 	}
 
 	/* Read the 16 bytes of station address PROM.
-	   We must first initialize registers, similar to NS8390_init(eifdev, 0).
-	   We can't reliably read the SAPROM address without this.
-	   (I learned the hard way!). */
+	 * We must first initialize registers, similar
+	 * to NS8390_init(eifdev, 0).
+	 * We can't reliably read the SAPROM address without this.
+	 * (I learned the hard way!).
+	 */
 	{
 		struct {unsigned char value, offset; } program_seq[] = {
-			{E8390_NODMA+E8390_PAGE0+E8390_STOP, E8390_CMD}, /* Select page 0*/
-			{0x49,	EN0_DCFG},	/* Set word-wide access. */
-			{0x00,	EN0_RCNTLO},	/* Clear the count regs. */
+			/* Select page 0 */
+			{E8390_NODMA + E8390_PAGE0 + E8390_STOP, E8390_CMD},
+			/* Set word-wide access */
+			{0x49,	EN0_DCFG},
+			/* Clear the count regs. */
+			{0x00,	EN0_RCNTLO},
+			/* Mask completion IRQ */
 			{0x00,	EN0_RCNTHI},
-			{0x00,	EN0_IMR},	/* Mask completion irq. */
+			{0x00,	EN0_IMR},
 			{0xFF,	EN0_ISR},
-			{E8390_RXOFF, EN0_RXCR},	/* 0x20  Set to monitor */
-			{E8390_TXOFF, EN0_TXCR},	/* 0x02  and loopback mode. */
+			/* 0x20 Set to monitor */
+			{E8390_RXOFF, EN0_RXCR},
+			/* 0x02 and loopback mode */
+			{E8390_TXOFF, EN0_TXCR},
 			{32,	EN0_RCNTLO},
 			{0x00,	EN0_RCNTHI},
-			{0x00,	EN0_RSARLO},	/* DMA starting at 0x0000. */
+			/* DMA starting at 0x0000 */
+			{0x00,	EN0_RSARLO},
 			{0x00,	EN0_RSARHI},
 			{E8390_RREAD+E8390_START, E8390_CMD},
 		};
 		for (i = 0; i < ARRAY_SIZE(program_seq); i++)
-			outb(program_seq[i].value, ioaddr + program_seq[i].offset);
+			outb(program_seq[i].value,
+			     ioaddr + program_seq[i].offset);
 
 	}
 
 	/* Note: all PCI cards have at least 16 bit access, so we don't have
-	   to check for 8 bit cards.  Most cards permit 32 bit access. */
+	 * to check for 8 bit cards.  Most cards permit 32 bit access.
+	 */
 	if (flags & ONLY_32BIT_IO) {
 		for (i = 0; i < 4 ; i++)
 			((u32 *)SA_prom)[i] = le32_to_cpu(inl(ioaddr + NE_DATAPORT));
 	} else
-		for(i = 0; i < 32 /*sizeof(SA_prom)*/; i++)
+		for (i = 0; i < 32 /* sizeof(SA_prom )*/; i++)
 			SA_prom[i] = inb(ioaddr + NE_DATAPORT);
 
 	/* We always set the 8390 registers for word mode. */
@@ -356,7 +361,7 @@ static int ne2k_pci_init_one(struct pci_dev *pdev,
 	ei_status.word16 = 1;
 	ei_status.ne2k_flags = flags;
 	if (fnd_cnt < MAX_UNITS) {
-		if (full_duplex[fnd_cnt] > 0  ||  (options[fnd_cnt] & FORCE_FDX))
+		if (full_duplex[fnd_cnt] > 0 || (options[fnd_cnt] & FORCE_FDX))
 			ei_status.ne2k_flags |= FORCE_FDX;
 	}
 
@@ -375,7 +380,7 @@ static int ne2k_pci_init_one(struct pci_dev *pdev,
 	dev->ethtool_ops = &ne2k_pci_ethtool_ops;
 	NS8390_init(dev, 0);
 
-	memcpy(dev->dev_addr, SA_prom, dev->addr_len);
+	eth_hw_addr_set(dev, SA_prom);
 
 	i = register_netdev(dev);
 	if (i)
@@ -388,16 +393,15 @@ static int ne2k_pci_init_one(struct pci_dev *pdev,
 	return 0;
 
 err_out_free_netdev:
-	free_netdev (dev);
+	free_netdev(dev);
 err_out_free_res:
-	release_region (ioaddr, NE_IO_EXTENT);
+	release_region(ioaddr, NE_IO_EXTENT);
 err_out:
 	pci_disable_device(pdev);
 	return -ENODEV;
 }
 
-/*
- * Magic incantation sequence for full duplex on the supported cards.
+/* Magic incantation sequence for full duplex on the supported cards.
  */
 static inline int set_realtek_fdx(struct net_device *dev)
 {
@@ -431,7 +435,9 @@ static int ne2k_pci_set_fdx(struct net_device *dev)
 
 static int ne2k_pci_open(struct net_device *dev)
 {
-	int ret = request_irq(dev->irq, ei_interrupt, IRQF_SHARED, dev->name, dev);
+	int ret = request_irq(dev->irq, ei_interrupt, IRQF_SHARED,
+			      dev->name, dev);
+
 	if (ret)
 		return ret;
 
@@ -450,7 +456,8 @@ static int ne2k_pci_close(struct net_device *dev)
 }
 
 /* Hard reset the card.  This used to pause for the same period that a
-   8390 reset command required, but that shouldn't be necessary. */
+ * 8390 reset command required, but that shouldn't be necessary.
+ */
 static void ne2k_pci_reset_8390(struct net_device *dev)
 {
 	unsigned long reset_start_time = jiffies;
@@ -467,31 +474,34 @@ static void ne2k_pci_reset_8390(struct net_device *dev)
 	/* This check _should_not_ be necessary, omit eventually. */
 	while ((inb(NE_BASE+EN0_ISR) & ENISR_RESET) == 0)
 		if (jiffies - reset_start_time > 2) {
-			netdev_err(dev, "ne2k_pci_reset_8390() did not complete.\n");
+			netdev_err(dev, "%s did not complete.\n", __func__);
 			break;
 		}
-	outb(ENISR_RESET, NE_BASE + EN0_ISR);	/* Ack intr. */
+	/* Ack intr. */
+	outb(ENISR_RESET, NE_BASE + EN0_ISR);
 }
 
 /* Grab the 8390 specific header. Similar to the block_input routine, but
-   we don't need to be concerned with ring wrap as the header will be at
-   the start of a page, so we optimize accordingly. */
+ * we don't need to be concerned with ring wrap as the header will be at
+ * the start of a page, so we optimize accordingly.
+ */
 
-static void ne2k_pci_get_8390_hdr(struct net_device *dev, struct e8390_pkt_hdr *hdr, int ring_page)
+static void ne2k_pci_get_8390_hdr(struct net_device *dev,
+				  struct e8390_pkt_hdr *hdr, int ring_page)
 {
 
 	long nic_base = dev->base_addr;
 
-	/* This *shouldn't* happen. If it does, it's the last thing you'll see */
+	/* This *shouldn't* happen. If it does, it's the last thing you'll see
+	 */
 	if (ei_status.dmaing) {
-		netdev_err(dev, "DMAing conflict in ne2k_pci_get_8390_hdr "
-			   "[DMAstat:%d][irqlock:%d].\n",
-			   ei_status.dmaing, ei_status.irqlock);
+		netdev_err(dev, "DMAing conflict in %s [DMAstat:%d][irqlock:%d].\n",
+			   __func__, ei_status.dmaing, ei_status.irqlock);
 		return;
 	}
 
 	ei_status.dmaing |= 0x01;
-	outb(E8390_NODMA+E8390_PAGE0+E8390_START, nic_base+ NE_CMD);
+	outb(E8390_NODMA + E8390_PAGE0 + E8390_START, nic_base + NE_CMD);
 	outb(sizeof(struct e8390_pkt_hdr), nic_base + EN0_RCNTLO);
 	outb(0, nic_base + EN0_RCNTHI);
 	outb(0, nic_base + EN0_RSARLO);		/* On page boundary */
@@ -499,20 +509,22 @@ static void ne2k_pci_get_8390_hdr(struct net_device *dev, struct e8390_pkt_hdr *
 	outb(E8390_RREAD+E8390_START, nic_base + NE_CMD);
 
 	if (ei_status.ne2k_flags & ONLY_16BIT_IO) {
-		insw(NE_BASE + NE_DATAPORT, hdr, sizeof(struct e8390_pkt_hdr)>>1);
+		insw(NE_BASE + NE_DATAPORT, hdr,
+		     sizeof(struct e8390_pkt_hdr) >> 1);
 	} else {
-		*(u32*)hdr = le32_to_cpu(inl(NE_BASE + NE_DATAPORT));
+		*(u32 *)hdr = le32_to_cpu(inl(NE_BASE + NE_DATAPORT));
 		le16_to_cpus(&hdr->count);
 	}
-
-	outb(ENISR_RDC, nic_base + EN0_ISR);	/* Ack intr. */
+	/* Ack intr. */
+	outb(ENISR_RDC, nic_base + EN0_ISR);
 	ei_status.dmaing &= ~0x01;
 }
 
 /* Block input and output, similar to the Crynwr packet driver.  If you
-   are porting to a new ethercard, look at the packet driver source for hints.
-   The NEx000 doesn't share the on-board packet memory -- you have to put
-   the packet out through the "remote DMA" dataport using outb. */
+ *are porting to a new ethercard, look at the packet driver source for hints.
+ *The NEx000 doesn't share the on-board packet memory -- you have to put
+ *the packet out through the "remote DMA" dataport using outb.
+ */
 
 static void ne2k_pci_block_input(struct net_device *dev, int count,
 				 struct sk_buff *skb, int ring_offset)
@@ -520,30 +532,30 @@ static void ne2k_pci_block_input(struct net_device *dev, int count,
 	long nic_base = dev->base_addr;
 	char *buf = skb->data;
 
-	/* This *shouldn't* happen. If it does, it's the last thing you'll see */
+	/* This *shouldn't* happen.
+	 * If it does, it's the last thing you'll see.
+	 */
 	if (ei_status.dmaing) {
-		netdev_err(dev, "DMAing conflict in ne2k_pci_block_input "
-			   "[DMAstat:%d][irqlock:%d].\n",
-			   ei_status.dmaing, ei_status.irqlock);
+		netdev_err(dev, "DMAing conflict in %s [DMAstat:%d][irqlock:%d]\n",
+			   __func__, ei_status.dmaing, ei_status.irqlock);
 		return;
 	}
 	ei_status.dmaing |= 0x01;
 	if (ei_status.ne2k_flags & ONLY_32BIT_IO)
 		count = (count + 3) & 0xFFFC;
-	outb(E8390_NODMA+E8390_PAGE0+E8390_START, nic_base+ NE_CMD);
+	outb(E8390_NODMA + E8390_PAGE0 + E8390_START, nic_base + NE_CMD);
 	outb(count & 0xff, nic_base + EN0_RCNTLO);
 	outb(count >> 8, nic_base + EN0_RCNTHI);
 	outb(ring_offset & 0xff, nic_base + EN0_RSARLO);
 	outb(ring_offset >> 8, nic_base + EN0_RSARHI);
-	outb(E8390_RREAD+E8390_START, nic_base + NE_CMD);
+	outb(E8390_RREAD + E8390_START, nic_base + NE_CMD);
 
 	if (ei_status.ne2k_flags & ONLY_16BIT_IO) {
-		insw(NE_BASE + NE_DATAPORT,buf,count>>1);
-		if (count & 0x01) {
+		insw(NE_BASE + NE_DATAPORT, buf, count >> 1);
+		if (count & 0x01)
 			buf[count-1] = inb(NE_BASE + NE_DATAPORT);
-		}
 	} else {
-		insl(NE_BASE + NE_DATAPORT, buf, count>>2);
+		insl(NE_BASE + NE_DATAPORT, buf, count >> 2);
 		if (count & 3) {
 			buf += count & ~3;
 			if (count & 2) {
@@ -556,41 +568,44 @@ static void ne2k_pci_block_input(struct net_device *dev, int count,
 				*buf = inb(NE_BASE + NE_DATAPORT);
 		}
 	}
-
-	outb(ENISR_RDC, nic_base + EN0_ISR);	/* Ack intr. */
+	/* Ack intr. */
+	outb(ENISR_RDC, nic_base + EN0_ISR);
 	ei_status.dmaing &= ~0x01;
 }
 
 static void ne2k_pci_block_output(struct net_device *dev, int count,
-				  const unsigned char *buf, const int start_page)
+		const unsigned char *buf, const int start_page)
 {
 	long nic_base = NE_BASE;
 	unsigned long dma_start;
 
 	/* On little-endian it's always safe to round the count up for
-	   word writes. */
+	 * word writes.
+	 */
 	if (ei_status.ne2k_flags & ONLY_32BIT_IO)
 		count = (count + 3) & 0xFFFC;
 	else
 		if (count & 0x01)
 			count++;
 
-	/* This *shouldn't* happen. If it does, it's the last thing you'll see */
+	/* This *shouldn't* happen.
+	 * If it does, it's the last thing you'll see.
+	 */
 	if (ei_status.dmaing) {
-		netdev_err(dev, "DMAing conflict in ne2k_pci_block_output."
-			   "[DMAstat:%d][irqlock:%d]\n",
-			   ei_status.dmaing, ei_status.irqlock);
+		netdev_err(dev, "DMAing conflict in %s [DMAstat:%d][irqlock:%d]\n",
+			   __func__, ei_status.dmaing, ei_status.irqlock);
 		return;
 	}
 	ei_status.dmaing |= 0x01;
 	/* We should already be in page 0, but to be safe... */
 	outb(E8390_PAGE0+E8390_START+E8390_NODMA, nic_base + NE_CMD);
 
-#ifdef NE8390_RW_BUGFIX
+#ifdef NE_RW_BUGFIX
 	/* Handle the read-before-write bug the same way as the
-	   Crynwr packet driver -- the NatSemi method doesn't work.
-	   Actually this doesn't always work either, but if you have
-	   problems with your NEx000 this is better than nothing! */
+	 * Crynwr packet driver -- the NatSemi method doesn't work.
+	 * Actually this doesn't always work either, but if you have
+	 * problems with your NEx000 this is better than nothing!
+	 */
 	outb(0x42, nic_base + EN0_RCNTLO);
 	outb(0x00, nic_base + EN0_RCNTHI);
 	outb(0x42, nic_base + EN0_RSARLO);
@@ -599,16 +614,16 @@ static void ne2k_pci_block_output(struct net_device *dev, int count,
 #endif
 	outb(ENISR_RDC, nic_base + EN0_ISR);
 
-   /* Now the normal output. */
+	/* Now the normal output. */
 	outb(count & 0xff, nic_base + EN0_RCNTLO);
 	outb(count >> 8,   nic_base + EN0_RCNTHI);
 	outb(0x00, nic_base + EN0_RSARLO);
 	outb(start_page, nic_base + EN0_RSARHI);
 	outb(E8390_RWRITE+E8390_START, nic_base + NE_CMD);
 	if (ei_status.ne2k_flags & ONLY_16BIT_IO) {
-		outsw(NE_BASE + NE_DATAPORT, buf, count>>1);
+		outsw(NE_BASE + NE_DATAPORT, buf, count >> 1);
 	} else {
-		outsl(NE_BASE + NE_DATAPORT, buf, count>>2);
+		outsl(NE_BASE + NE_DATAPORT, buf, count >> 2);
 		if (count & 3) {
 			buf += count & ~3;
 			if (count & 2) {
@@ -623,14 +638,15 @@ static void ne2k_pci_block_output(struct net_device *dev, int count,
 	dma_start = jiffies;
 
 	while ((inb(nic_base + EN0_ISR) & ENISR_RDC) == 0)
-		if (jiffies - dma_start > 2) {			/* Avoid clock roll-over. */
+		/* Avoid clock roll-over. */
+		if (jiffies - dma_start > 2) {
 			netdev_warn(dev, "timeout waiting for Tx RDC.\n");
 			ne2k_pci_reset_8390(dev);
-			NS8390_init(dev,1);
+			NS8390_init(dev, 1);
 			break;
 		}
-
-	outb(ENISR_RDC, nic_base + EN0_ISR);	/* Ack intr. */
+	/* Ack intr. */
+	outb(ENISR_RDC, nic_base + EN0_ISR);
 	ei_status.dmaing &= ~0x01;
 }
 
@@ -640,9 +656,9 @@ static void ne2k_pci_get_drvinfo(struct net_device *dev,
 	struct ei_device *ei = netdev_priv(dev);
 	struct pci_dev *pci_dev = (struct pci_dev *) ei->priv;
 
-	strlcpy(info->driver, DRV_NAME, sizeof(info->driver));
-	strlcpy(info->version, DRV_VERSION, sizeof(info->version));
-	strlcpy(info->bus_info, pci_name(pci_dev), sizeof(info->bus_info));
+	strscpy(info->driver, DRV_NAME, sizeof(info->driver));
+	strscpy(info->version, DRV_VERSION, sizeof(info->version));
+	strscpy(info->bus_info, pci_name(pci_dev), sizeof(info->bus_info));
 }
 
 static u32 ne2k_pci_get_msglevel(struct net_device *dev)
@@ -676,30 +692,18 @@ static void ne2k_pci_remove_one(struct pci_dev *pdev)
 	pci_disable_device(pdev);
 }
 
-#ifdef CONFIG_PM
-static int ne2k_pci_suspend (struct pci_dev *pdev, pm_message_t state)
+static int __maybe_unused ne2k_pci_suspend(struct device *dev_d)
 {
-	struct net_device *dev = pci_get_drvdata (pdev);
+	struct net_device *dev = dev_get_drvdata(dev_d);
 
 	netif_device_detach(dev);
-	pci_save_state(pdev);
-	pci_disable_device(pdev);
-	pci_set_power_state(pdev, pci_choose_state(pdev, state));
 
 	return 0;
 }
 
-static int ne2k_pci_resume (struct pci_dev *pdev)
+static int __maybe_unused ne2k_pci_resume(struct device *dev_d)
 {
-	struct net_device *dev = pci_get_drvdata (pdev);
-	int rc;
-
-	pci_set_power_state(pdev, PCI_D0);
-	pci_restore_state(pdev);
-
-	rc = pci_enable_device(pdev);
-	if (rc)
-		return rc;
+	struct net_device *dev = dev_get_drvdata(dev_d);
 
 	NS8390_init(dev, 1);
 	netif_device_attach(dev);
@@ -707,36 +711,13 @@ static int ne2k_pci_resume (struct pci_dev *pdev)
 	return 0;
 }
 
-#endif /* CONFIG_PM */
-
+static SIMPLE_DEV_PM_OPS(ne2k_pci_pm_ops, ne2k_pci_suspend, ne2k_pci_resume);
 
 static struct pci_driver ne2k_driver = {
 	.name		= DRV_NAME,
 	.probe		= ne2k_pci_init_one,
 	.remove		= ne2k_pci_remove_one,
 	.id_table	= ne2k_pci_tbl,
-#ifdef CONFIG_PM
-	.suspend	= ne2k_pci_suspend,
-	.resume		= ne2k_pci_resume,
-#endif /* CONFIG_PM */
-
+	.driver.pm	= &ne2k_pci_pm_ops,
 };
-
-
-static int __init ne2k_pci_init(void)
-{
-/* when a module, this is printed whether or not devices are found in probe */
-#ifdef MODULE
-	printk(version);
-#endif
-	return pci_register_driver(&ne2k_driver);
-}
-
-
-static void __exit ne2k_pci_cleanup(void)
-{
-	pci_unregister_driver (&ne2k_driver);
-}
-
-module_init(ne2k_pci_init);
-module_exit(ne2k_pci_cleanup);
+module_pci_driver(ne2k_driver);
diff --git a/drivers/net/ethernet/8390/pcnet_cs.c b/drivers/net/ethernet/8390/pcnet_cs.c
index bd0a2a14b649..19f9c5db3f3b 100644
--- a/drivers/net/ethernet/8390/pcnet_cs.c
+++ b/drivers/net/ethernet/8390/pcnet_cs.c
@@ -1,3 +1,4 @@
+// SPDX-License-Identifier: GPL-1.0+
 /*======================================================================
 
     A PCMCIA ethernet driver for NS8390-based cards
@@ -17,9 +18,7 @@
 
     Written 1992,1993 by Donald Becker.
     Copyright 1993 United States Government as represented by the
-    Director, National Security Agency.  This software may be used and
-    distributed according to the terms of the GNU General Public License,
-    incorporated herein by reference.
+    Director, National Security Agency.
     Donald Becker may be reached at becker@scyld.com
 
     Based also on Keith Moore's changes to Don Becker's code, for IBM
@@ -66,7 +65,6 @@
 #define PCNET_RDC_TIMEOUT (2*HZ/100)	/* Max wait in jiffies for Tx RDC */
 
 static const char *if_names[] = { "auto", "10baseT", "10base2"};
-static u32 pcnet_msg_enable;
 
 /*====================================================================*/
 
@@ -99,7 +97,7 @@ static int pcnet_open(struct net_device *dev);
 static int pcnet_close(struct net_device *dev);
 static int ei_ioctl(struct net_device *dev, struct ifreq *rq, int cmd);
 static irqreturn_t ei_irq_wrapper(int irq, void *dev_id);
-static void ei_watchdog(u_long arg);
+static void ei_watchdog(struct timer_list *t);
 static void pcnet_reset_8390(struct net_device *dev);
 static int set_config(struct net_device *dev, struct ifmap *map);
 static int setup_shmem_window(struct pcmcia_device *link, int start_pg,
@@ -224,7 +222,7 @@ static const struct net_device_ops pcnet_netdev_ops = {
 	.ndo_set_config		= set_config,
 	.ndo_start_xmit 	= ei_start_xmit,
 	.ndo_get_stats		= ei_get_stats,
-	.ndo_do_ioctl 		= ei_ioctl,
+	.ndo_eth_ioctl		= ei_ioctl,
 	.ndo_set_rx_mode	= ei_set_multicast_list,
 	.ndo_tx_timeout 	= ei_tx_timeout,
 	.ndo_set_mac_address 	= eth_mac_addr,
@@ -279,6 +277,7 @@ static struct hw_info *get_hwinfo(struct pcmcia_device *link)
 {
     struct net_device *dev = link->priv;
     u_char __iomem *base, *virt;
+    u8 addr[ETH_ALEN];
     int i, j;
 
     /* Allocate a small memory window */
@@ -290,6 +289,11 @@ static struct hw_info *get_hwinfo(struct pcmcia_device *link)
 
     virt = ioremap(link->resource[2]->start,
 	    resource_size(link->resource[2]));
+    if (unlikely(!virt)) {
+	    pcmcia_release_window(link, link->resource[2]);
+	    return NULL;
+    }
+
     for (i = 0; i < NR_INFO; i++) {
 	pcmcia_map_mem_page(link, link->resource[2],
 		hw_info[i].offset & ~(resource_size(link->resource[2])-1));
@@ -298,7 +302,8 @@ static struct hw_info *get_hwinfo(struct pcmcia_device *link)
 	    (readb(base+2) == hw_info[i].a1) &&
 	    (readb(base+4) == hw_info[i].a2)) {
 		for (j = 0; j < 6; j++)
-		    dev->dev_addr[j] = readb(base + (j<<1));
+			addr[j] = readb(base + (j<<1));
+		eth_hw_addr_set(dev, addr);
 		break;
 	}
     }
@@ -320,6 +325,7 @@ static struct hw_info *get_prom(struct pcmcia_device *link)
 {
     struct net_device *dev = link->priv;
     unsigned int ioaddr = dev->base_addr;
+    u8 addr[ETH_ALEN];
     u_char prom[32];
     int i, j;
 
@@ -358,7 +364,8 @@ static struct hw_info *get_prom(struct pcmcia_device *link)
     }
     if ((i < NR_INFO) || ((prom[28] == 0x57) && (prom[30] == 0x57))) {
 	for (j = 0; j < 6; j++)
-	    dev->dev_addr[j] = prom[j<<1];
+	    addr[j] = prom[j<<1];
+	eth_hw_addr_set(dev, addr);
 	return (i < NR_INFO) ? hw_info+i : &default_info;
     }
     return NULL;
@@ -373,6 +380,7 @@ static struct hw_info *get_prom(struct pcmcia_device *link)
 static struct hw_info *get_dl10019(struct pcmcia_device *link)
 {
     struct net_device *dev = link->priv;
+    u8 addr[ETH_ALEN];
     int i;
     u_char sum;
 
@@ -381,7 +389,8 @@ static struct hw_info *get_dl10019(struct pcmcia_device *link)
     if (sum != 0xff)
 	return NULL;
     for (i = 0; i < 6; i++)
-	dev->dev_addr[i] = inb_p(dev->base_addr + 0x14 + i);
+	addr[i] = inb_p(dev->base_addr + 0x14 + i);
+    eth_hw_addr_set(dev, addr);
     i = inb(dev->base_addr + 0x1f);
     return ((i == 0x91)||(i == 0x99)) ? &dl10022_info : &dl10019_info;
 }
@@ -396,6 +405,7 @@ static struct hw_info *get_ax88190(struct pcmcia_device *link)
 {
     struct net_device *dev = link->priv;
     unsigned int ioaddr = dev->base_addr;
+    u8 addr[ETH_ALEN];
     int i, j;
 
     /* Not much of a test, but the alternatives are messy */
@@ -409,9 +419,10 @@ static struct hw_info *get_ax88190(struct pcmcia_device *link)
 
     for (i = 0; i < 6; i += 2) {
 	j = inw(ioaddr + PCNET_DATAPORT);
-	dev->dev_addr[i] = j & 0xff;
-	dev->dev_addr[i+1] = j >> 8;
+	addr[i] = j & 0xff;
+	addr[i+1] = j >> 8;
     }
+    eth_hw_addr_set(dev, addr);
     return NULL;
 }
 
@@ -426,6 +437,7 @@ static struct hw_info *get_ax88190(struct pcmcia_device *link)
 static struct hw_info *get_hwired(struct pcmcia_device *link)
 {
     struct net_device *dev = link->priv;
+    u8 addr[ETH_ALEN];
     int i;
 
     for (i = 0; i < 6; i++)
@@ -434,7 +446,8 @@ static struct hw_info *get_hwired(struct pcmcia_device *link)
 	return NULL;
 
     for (i = 0; i < 6; i++)
-	dev->dev_addr[i] = hw_addr[i];
+	addr[i] = hw_addr[i];
+    eth_hw_addr_set(dev, addr);
 
     return &default_info;
 } /* get_hwired */
@@ -556,7 +569,6 @@ static int pcnet_config(struct pcmcia_device *link)
     int start_pg, stop_pg, cm_offset;
     int has_shmem = 0;
     struct hw_info *local_hw_info;
-    struct ei_device *ei_local;
 
     dev_dbg(&link->dev, "pcnet_config\n");
 
@@ -606,8 +618,6 @@ static int pcnet_config(struct pcmcia_device *link)
 	mii_phy_probe(dev);
 
     SET_NETDEV_DEV(dev, &link->dev);
-    ei_local = netdev_priv(dev);
-    ei_local->msg_enable = pcnet_msg_enable;
 
     if (register_netdev(dev) != 0) {
 	pr_notice("register_netdev() failed\n");
@@ -917,7 +927,7 @@ static int pcnet_open(struct net_device *dev)
 
     info->phy_id = info->eth_phy;
     info->link_status = 0x00;
-    setup_timer(&info->watchdog, ei_watchdog, (u_long)dev);
+    timer_setup(&info->watchdog, ei_watchdog, 0);
     mod_timer(&info->watchdog, jiffies + HZ);
 
     return ei_open(dev);
@@ -937,7 +947,7 @@ static int pcnet_close(struct net_device *dev)
 
     link->open--;
     netif_stop_queue(dev);
-    del_timer_sync(&info->watchdog);
+    timer_delete_sync(&info->watchdog);
 
     return 0;
 } /* pcnet_close */
@@ -984,7 +994,7 @@ static int set_config(struct net_device *dev, struct ifmap *map)
 	    return -EOPNOTSUPP;
 	else if ((map->port < 1) || (map->port > 2))
 	    return -EINVAL;
-	dev->if_port = map->port;
+	WRITE_ONCE(dev->if_port, map->port);
 	netdev_info(dev, "switched to %s port\n", if_names[dev->if_port]);
 	NS8390_init(dev, 1);
     }
@@ -1006,10 +1016,10 @@ static irqreturn_t ei_irq_wrapper(int irq, void *dev_id)
     return ret;
 }
 
-static void ei_watchdog(u_long arg)
+static void ei_watchdog(struct timer_list *t)
 {
-    struct net_device *dev = (struct net_device *)arg;
-    struct pcnet_dev *info = PRIV(dev);
+    struct pcnet_dev *info = timer_container_of(info, t, watchdog);
+    struct net_device *dev = info->p_dev->priv;
     unsigned int nic_base = dev->base_addr;
     unsigned int mii_addr = nic_base + DLINK_GPIO;
     u_short link;
@@ -1107,6 +1117,7 @@ static int ei_ioctl(struct net_device *dev, struct ifreq *rq, int cmd)
     switch (cmd) {
     case SIOCGMIIPHY:
 	data->phy_id = info->phy_id;
+	fallthrough;
     case SIOCGMIIREG:		/* Read MII PHY register. */
 	data->val_out = mdio_read(mii_addr, data->phy_id, data->reg_num & 0x1f);
 	return 0;
@@ -1176,8 +1187,10 @@ static void dma_block_input(struct net_device *dev, int count,
     outb_p(E8390_RREAD+E8390_START, nic_base + PCNET_CMD);
 
     insw(nic_base + PCNET_DATAPORT,buf,count>>1);
-    if (count & 0x01)
-	buf[count-1] = inb(nic_base + PCNET_DATAPORT), xfer_count++;
+    if (count & 0x01) {
+	buf[count-1] = inb(nic_base + PCNET_DATAPORT);
+	xfer_count++;
+    }
 
     /* This was for the ALPHA version only, but enough people have been
        encountering problems that it is still here. */
@@ -1426,6 +1439,11 @@ static int setup_shmem_window(struct pcmcia_device *link, int start_pg,
     /* Try scribbling on the buffer */
     info->base = ioremap(link->resource[3]->start,
 			resource_size(link->resource[3]));
+    if (unlikely(!info->base)) {
+	    ret = -ENOMEM;
+	    goto failed;
+    }
+
     for (i = 0; i < (TX_PAGES<<8); i += 2)
 	__raw_writew((i>>1), info->base+offset+i);
     udelay(100);
@@ -1518,7 +1536,7 @@ static const struct pcmcia_device_id pcnet_ids[] = {
 	PCMCIA_DEVICE_PROD_ID12("ACCTON", "EN2216-PCMCIA-ETHERNET", 0xdfc6b5b2, 0x5542bfff),
 	PCMCIA_DEVICE_PROD_ID12("Allied Telesis, K.K.", "CentreCOM LA100-PCM-T V2 100/10M LAN PC Card", 0xbb7fbdd7, 0xcd91cc68),
 	PCMCIA_DEVICE_PROD_ID12("Allied Telesis K.K.", "LA100-PCM V2", 0x36634a66, 0xc6d05997),
-  	PCMCIA_DEVICE_PROD_ID12("Allied Telesis, K.K.", "CentreCOM LA-PCM_V2", 0xbb7fBdd7, 0x28e299f8),
+	PCMCIA_DEVICE_PROD_ID12("Allied Telesis, K.K.", "CentreCOM LA-PCM_V2", 0xbb7fBdd7, 0x28e299f8),
 	PCMCIA_DEVICE_PROD_ID12("Allied Telesis K.K.", "LA-PCM V3", 0x36634a66, 0x62241d96),
 	PCMCIA_DEVICE_PROD_ID12("AmbiCom", "AMB8010", 0x5070a7f9, 0x82f96e96),
 	PCMCIA_DEVICE_PROD_ID12("AmbiCom", "AMB8610", 0x5070a7f9, 0x86741224),
diff --git a/drivers/net/ethernet/8390/smc-ultra.c b/drivers/net/ethernet/8390/smc-ultra.c
index 4e02f6a23575..22ca804b2e95 100644
--- a/drivers/net/ethernet/8390/smc-ultra.c
+++ b/drivers/net/ethernet/8390/smc-ultra.c
@@ -1,3 +1,4 @@
+// SPDX-License-Identifier: GPL-1.0+
 /* smc-ultra.c: A SMC Ultra ethernet driver for linux. */
 /*
 	This is a driver for the SMC Ultra and SMC EtherEZ ISA ethercards.
@@ -7,9 +8,6 @@
 	Copyright 1993 United States Government as represented by the
 	Director, National Security Agency.
 
-	This software may be used and distributed according to the terms
-	of the GNU General Public License, incorporated herein by reference.
-
 	The author may be reached as becker@scyld.com, or C/O
 	Scyld Computing Corporation
 	410 Severn Ave., Suite 210
@@ -66,6 +64,7 @@ static const char version[] =
 #include <linux/isapnp.h>
 #include <linux/netdevice.h>
 #include <linux/etherdevice.h>
+#include <net/Space.h>
 
 #include <asm/io.h>
 #include <asm/irq.h>
@@ -204,6 +203,7 @@ static int __init ultra_probe1(struct net_device *dev, int ioaddr)
 {
 	int i, retval;
 	int checksum = 0;
+	u8 macaddr[ETH_ALEN];
 	const char *model_name;
 	unsigned char eeprom_irq = 0;
 	static unsigned version_printed;
@@ -239,7 +239,8 @@ static int __init ultra_probe1(struct net_device *dev, int ioaddr)
 	model_name = (idreg & 0xF0) == 0x20 ? "SMC Ultra" : "SMC EtherEZ";
 
 	for (i = 0; i < 6; i++)
-		dev->dev_addr[i] = inb(ioaddr + 8 + i);
+		macaddr[i] = inb(ioaddr + 8 + i);
+	eth_hw_addr_set(dev, macaddr);
 
 	netdev_info(dev, "%s at %#3x, %pM", model_name,
 		    ioaddr, dev->dev_addr);
@@ -347,11 +348,11 @@ static int __init ultra_probe_isapnp(struct net_device *dev)
                                             idev))) {
                         /* Avoid already found cards from previous calls */
                         if (pnp_device_attach(idev) < 0)
-                        	continue;
+				continue;
                         if (pnp_activate_dev(idev) < 0) {
                               __again:
-                        	pnp_device_detach(idev);
-                        	continue;
+				pnp_device_detach(idev);
+				continue;
                         }
 			/* if no io and irq, search for next */
 			if (!pnp_port_valid(idev, 0) || !pnp_irq_valid(idev, 0))
@@ -522,7 +523,6 @@ static void ultra_pio_input(struct net_device *dev, int count,
 	/* We know skbuffs are padded to at least word alignment. */
 	insw(ioaddr + IOPD, buf, (count+1)>>1);
 }
-
 static void ultra_pio_output(struct net_device *dev, int count,
 							const unsigned char *buf, const int start_page)
 {
@@ -563,7 +563,7 @@ static int irq[MAX_ULTRA_CARDS];
 
 module_param_hw_array(io, int, ioport, NULL, 0);
 module_param_hw_array(irq, int, irq, NULL, 0);
-module_param_named(msg_enable, ultra_msg_enable, uint, (S_IRUSR|S_IRGRP|S_IROTH));
+module_param_named(msg_enable, ultra_msg_enable, uint, 0444);
 MODULE_PARM_DESC(io, "I/O base address(es)");
 MODULE_PARM_DESC(irq, "IRQ number(s) (assigned)");
 MODULE_PARM_DESC(msg_enable, "Debug message level (see linux/netdevice.h for bitmap)");
@@ -572,8 +572,7 @@ MODULE_LICENSE("GPL");
 
 /* This is set up so that only a single autoprobe takes place per call.
 ISA device autoprobes on a running machine are not recommended. */
-int __init
-init_module(void)
+static int __init ultra_init_module(void)
 {
 	struct net_device *dev;
 	int this_dev, found = 0;
@@ -600,6 +599,7 @@ init_module(void)
 		return 0;
 	return -ENXIO;
 }
+module_init(ultra_init_module);
 
 static void cleanup_card(struct net_device *dev)
 {
@@ -613,8 +613,7 @@ static void cleanup_card(struct net_device *dev)
 	iounmap(ei_status.mem);
 }
 
-void __exit
-cleanup_module(void)
+static void __exit ultra_cleanup_module(void)
 {
 	int this_dev;
 
@@ -627,4 +626,5 @@ cleanup_module(void)
 		}
 	}
 }
+module_exit(ultra_cleanup_module);
 #endif /* MODULE */
diff --git a/drivers/net/ethernet/8390/stnic.c b/drivers/net/ethernet/8390/stnic.c
index aca957d4e121..6cc0e190aa79 100644
--- a/drivers/net/ethernet/8390/stnic.c
+++ b/drivers/net/ethernet/8390/stnic.c
@@ -1,9 +1,6 @@
+// SPDX-License-Identifier: GPL-2.0-only
 /* stnic.c : A SH7750 specific part of driver for NS DP83902A ST-NIC.
  *
- * This file is subject to the terms and conditions of the GNU General Public
- * License.  See the file "COPYING" in the main directory of this archive
- * for more details.
- *
  * Copyright (C) 1999 kaz Kojima
  */
 
@@ -71,7 +68,7 @@ static void stnic_init (struct net_device *dev);
 
 static u32 stnic_msg_enable;
 
-module_param_named(msg_enable, stnic_msg_enable, uint, (S_IRUSR|S_IRGRP|S_IROTH));
+module_param_named(msg_enable, stnic_msg_enable, uint, 0444);
 MODULE_PARM_DESC(msg_enable, "Debug message level (see linux/netdevice.h for bitmap)");
 
 /* SH7750 specific read/write io. */
@@ -104,8 +101,8 @@ STNIC_WRITE (int reg, byte val)
 static int __init stnic_probe(void)
 {
   struct net_device *dev;
-  int i, err;
   struct ei_device *ei_local;
+  int err;
 
   /* If we are not running on a SolutionEngine, give up now */
   if (! MACH_SE)
@@ -114,13 +111,12 @@ static int __init stnic_probe(void)
   /* New style probing API */
   dev = alloc_ei_netdev();
   if (!dev)
-  	return -ENOMEM;
+	return -ENOMEM;
 
 #ifdef CONFIG_SH_STANDARD_BIOS
   sh_bios_get_node_addr (stnic_eadr);
 #endif
-  for (i = 0; i < ETH_ALEN; i++)
-    dev->dev_addr[i] = stnic_eadr[i];
+  eth_hw_addr_set(dev, stnic_eadr);
 
   /* Set the base address to point to the NIC, not the "real" base! */
   dev->base_addr = 0x1000;
@@ -300,4 +296,5 @@ static void __exit stnic_cleanup(void)
 
 module_init(stnic_probe);
 module_exit(stnic_cleanup);
+MODULE_DESCRIPTION("National Semiconductor DP83902AV ethernet driver");
 MODULE_LICENSE("GPL");
diff --git a/drivers/net/ethernet/8390/wd.c b/drivers/net/ethernet/8390/wd.c
index 6efa2722f850..ffd639477dfc 100644
--- a/drivers/net/ethernet/8390/wd.c
+++ b/drivers/net/ethernet/8390/wd.c
@@ -1,3 +1,4 @@
+// SPDX-License-Identifier: GPL-1.0+
 /* wd.c: A WD80x3 ethernet driver for linux. */
 /*
 	Written 1993-94 by Donald Becker.
@@ -5,9 +6,6 @@
 	Copyright 1993 United States Government as represented by the
 	Director, National Security Agency.
 
-	This software may be used and distributed according to the terms
-	of the GNU General Public License, incorporated herein by reference.
-
 	The author may be reached as becker@scyld.com, or C/O
 	Scyld Computing Corporation
 	410 Severn Ave., Suite 210
@@ -37,6 +35,7 @@ static const char version[] =
 #include <linux/delay.h>
 #include <linux/netdevice.h>
 #include <linux/etherdevice.h>
+#include <net/Space.h>
 
 #include <asm/io.h>
 
@@ -168,6 +167,7 @@ static int __init wd_probe1(struct net_device *dev, int ioaddr)
 	int checksum = 0;
 	int ancient = 0;			/* An old card without config registers. */
 	int word16 = 0;				/* 0 = 8 bit, 1 = 16 bit */
+	u8 addr[ETH_ALEN];
 	const char *model_name;
 	static unsigned version_printed;
 	struct ei_device *ei_local = netdev_priv(dev);
@@ -191,7 +191,8 @@ static int __init wd_probe1(struct net_device *dev, int ioaddr)
 		netdev_info(dev, version);
 
 	for (i = 0; i < 6; i++)
-		dev->dev_addr[i] = inb(ioaddr + 8 + i);
+		addr[i] = inb(ioaddr + 8 + i);
+	eth_hw_addr_set(dev, addr);
 
 	netdev_info(dev, "WD80x3 at %#3x, %pM", ioaddr, dev->dev_addr);
 
@@ -299,7 +300,7 @@ static int __init wd_probe1(struct net_device *dev, int ioaddr)
 
 			outb_p(0x00, nic_addr+EN0_IMR);	/* Mask all intrs. again. */
 
-			if (netif_msg_drv(ei_local))
+			if (wd_msg_enable & NETIF_MSG_PROBE)
 				pr_cont(" autoirq is %d", dev->irq);
 			if (dev->irq < 2)
 				dev->irq = word16 ? 10 : 5;
@@ -507,7 +508,7 @@ module_param_hw_array(io, int, ioport, NULL, 0);
 module_param_hw_array(irq, int, irq, NULL, 0);
 module_param_hw_array(mem, int, iomem, NULL, 0);
 module_param_hw_array(mem_end, int, iomem, NULL, 0);
-module_param_named(msg_enable, wd_msg_enable, uint, (S_IRUSR|S_IRGRP|S_IROTH));
+module_param_named(msg_enable, wd_msg_enable, uint, 0444);
 MODULE_PARM_DESC(io, "I/O base address(es)");
 MODULE_PARM_DESC(irq, "IRQ number(s) (ignored for PureData boards)");
 MODULE_PARM_DESC(mem, "memory base address(es)(ignored for PureData boards)");
@@ -519,7 +520,7 @@ MODULE_LICENSE("GPL");
 /* This is set up so that only a single autoprobe takes place per call.
 ISA device autoprobes on a running machine are not recommended. */
 
-int __init init_module(void)
+static int __init wd_init_module(void)
 {
 	struct net_device *dev;
 	int this_dev, found = 0;
@@ -548,6 +549,7 @@ int __init init_module(void)
 		return 0;
 	return -ENXIO;
 }
+module_init(wd_init_module);
 
 static void cleanup_card(struct net_device *dev)
 {
@@ -556,8 +558,7 @@ static void cleanup_card(struct net_device *dev)
 	iounmap(ei_status.mem);
 }
 
-void __exit
-cleanup_module(void)
+static void __exit wd_cleanup_module(void)
 {
 	int this_dev;
 
@@ -570,4 +571,5 @@ cleanup_module(void)
 		}
 	}
 }
+module_exit(wd_cleanup_module);
 #endif /* MODULE */
diff --git a/drivers/net/ethernet/8390/xsurf100.c b/drivers/net/ethernet/8390/xsurf100.c
new file mode 100644
index 000000000000..fe7a74707aa4
--- /dev/null
+++ b/drivers/net/ethernet/8390/xsurf100.c
@@ -0,0 +1,377 @@
+// SPDX-License-Identifier: GPL-2.0
+#include <linux/module.h>
+#include <linux/netdevice.h>
+#include <linux/platform_device.h>
+#include <linux/zorro.h>
+#include <net/ax88796.h>
+#include <asm/amigaints.h>
+
+#define ZORRO_PROD_INDIVIDUAL_COMPUTERS_X_SURF100 \
+		ZORRO_ID(INDIVIDUAL_COMPUTERS, 0x64, 0)
+
+#define XS100_IRQSTATUS_BASE 0x40
+#define XS100_8390_BASE 0x800
+
+/* Longword-access area. Translated to 2 16-bit access cycles by the
+ * X-Surf 100 FPGA
+ */
+#define XS100_8390_DATA32_BASE 0x8000
+#define XS100_8390_DATA32_SIZE 0x2000
+/* Sub-Areas for fast data register access; addresses relative to area begin */
+#define XS100_8390_DATA_READ32_BASE 0x0880
+#define XS100_8390_DATA_WRITE32_BASE 0x0C80
+#define XS100_8390_DATA_AREA_SIZE 0x80
+
+/* force unsigned long back to 'void __iomem *' */
+#define ax_convert_addr(_a) ((void __force __iomem *)(_a))
+
+#define ei_inb(_a) z_readb(ax_convert_addr(_a))
+#define ei_outb(_v, _a) z_writeb(_v, ax_convert_addr(_a))
+
+#define ei_inw(_a) z_readw(ax_convert_addr(_a))
+#define ei_outw(_v, _a) z_writew(_v, ax_convert_addr(_a))
+
+#define ei_inb_p(_a) ei_inb(_a)
+#define ei_outb_p(_v, _a) ei_outb(_v, _a)
+
+/* define EI_SHIFT() to take into account our register offsets */
+#define EI_SHIFT(x) (ei_local->reg_offset[(x)])
+
+/* Ensure we have our RCR base value */
+#define AX88796_PLATFORM
+
+#include "8390.h"
+
+/* from ne.c */
+#define NE_CMD		EI_SHIFT(0x00)
+#define NE_RESET	EI_SHIFT(0x1f)
+#define NE_DATAPORT	EI_SHIFT(0x10)
+
+struct xsurf100_ax_plat_data {
+	struct ax_plat_data ax;
+	void __iomem *base_regs;
+	void __iomem *data_area;
+};
+
+static int is_xsurf100_network_irq(struct platform_device *pdev)
+{
+	struct xsurf100_ax_plat_data *xs100 = dev_get_platdata(&pdev->dev);
+
+	return (readw(xs100->base_regs + XS100_IRQSTATUS_BASE) & 0xaaaa) != 0;
+}
+
+/* These functions guarantee that the iomem is accessed with 32 bit
+ * cycles only. z_memcpy_fromio / z_memcpy_toio don't
+ */
+static void z_memcpy_fromio32(void *dst, const void __iomem *src, size_t bytes)
+{
+	while (bytes > 32) {
+		asm __volatile__
+		   ("movem.l (%0)+,%%d0-%%d7\n"
+		    "movem.l %%d0-%%d7,(%1)\n"
+		    "adda.l #32,%1" : "=a"(src), "=a"(dst)
+		    : "0"(src), "1"(dst) : "d0", "d1", "d2", "d3", "d4",
+					   "d5", "d6", "d7", "memory");
+		bytes -= 32;
+	}
+	while (bytes) {
+		*(uint32_t *)dst = z_readl(src);
+		src += 4;
+		dst += 4;
+		bytes -= 4;
+	}
+}
+
+static void z_memcpy_toio32(void __iomem *dst, const void *src, size_t bytes)
+{
+	while (bytes) {
+		z_writel(*(const uint32_t *)src, dst);
+		src += 4;
+		dst += 4;
+		bytes -= 4;
+	}
+}
+
+static void xs100_write(struct net_device *dev, const void *src,
+			unsigned int count)
+{
+	struct ei_device *ei_local = netdev_priv(dev);
+	struct platform_device *pdev = to_platform_device(dev->dev.parent);
+	struct xsurf100_ax_plat_data *xs100 = dev_get_platdata(&pdev->dev);
+
+	/* copy whole blocks */
+	while (count > XS100_8390_DATA_AREA_SIZE) {
+		z_memcpy_toio32(xs100->data_area +
+				XS100_8390_DATA_WRITE32_BASE, src,
+				XS100_8390_DATA_AREA_SIZE);
+		src += XS100_8390_DATA_AREA_SIZE;
+		count -= XS100_8390_DATA_AREA_SIZE;
+	}
+	/* copy whole dwords */
+	z_memcpy_toio32(xs100->data_area + XS100_8390_DATA_WRITE32_BASE,
+			src, count & ~3);
+	src += count & ~3;
+	if (count & 2) {
+		ei_outw(*(uint16_t *)src, ei_local->mem + NE_DATAPORT);
+		src += 2;
+	}
+	if (count & 1)
+		ei_outb(*(uint8_t *)src, ei_local->mem + NE_DATAPORT);
+}
+
+static void xs100_read(struct net_device *dev, void *dst, unsigned int count)
+{
+	struct ei_device *ei_local = netdev_priv(dev);
+	struct platform_device *pdev = to_platform_device(dev->dev.parent);
+	struct xsurf100_ax_plat_data *xs100 = dev_get_platdata(&pdev->dev);
+
+	/* copy whole blocks */
+	while (count > XS100_8390_DATA_AREA_SIZE) {
+		z_memcpy_fromio32(dst, xs100->data_area +
+				  XS100_8390_DATA_READ32_BASE,
+				  XS100_8390_DATA_AREA_SIZE);
+		dst += XS100_8390_DATA_AREA_SIZE;
+		count -= XS100_8390_DATA_AREA_SIZE;
+	}
+	/* copy whole dwords */
+	z_memcpy_fromio32(dst, xs100->data_area + XS100_8390_DATA_READ32_BASE,
+			  count & ~3);
+	dst += count & ~3;
+	if (count & 2) {
+		*(uint16_t *)dst = ei_inw(ei_local->mem + NE_DATAPORT);
+		dst += 2;
+	}
+	if (count & 1)
+		*(uint8_t *)dst = ei_inb(ei_local->mem + NE_DATAPORT);
+}
+
+/* Block input and output, similar to the Crynwr packet driver. If
+ * you are porting to a new ethercard, look at the packet driver
+ * source for hints. The NEx000 doesn't share the on-board packet
+ * memory -- you have to put the packet out through the "remote DMA"
+ * dataport using ei_outb.
+ */
+static void xs100_block_input(struct net_device *dev, int count,
+			      struct sk_buff *skb, int ring_offset)
+{
+	struct ei_device *ei_local = netdev_priv(dev);
+	void __iomem *nic_base = ei_local->mem;
+	char *buf = skb->data;
+
+	if (ei_local->dmaing) {
+		netdev_err(dev,
+			   "DMAing conflict in %s [DMAstat:%d][irqlock:%d]\n",
+			   __func__,
+			   ei_local->dmaing, ei_local->irqlock);
+		return;
+	}
+
+	ei_local->dmaing |= 0x01;
+
+	ei_outb(E8390_NODMA + E8390_PAGE0 + E8390_START, nic_base + NE_CMD);
+	ei_outb(count & 0xff, nic_base + EN0_RCNTLO);
+	ei_outb(count >> 8, nic_base + EN0_RCNTHI);
+	ei_outb(ring_offset & 0xff, nic_base + EN0_RSARLO);
+	ei_outb(ring_offset >> 8, nic_base + EN0_RSARHI);
+	ei_outb(E8390_RREAD + E8390_START, nic_base + NE_CMD);
+
+	xs100_read(dev, buf, count);
+
+	ei_local->dmaing &= ~1;
+}
+
+static void xs100_block_output(struct net_device *dev, int count,
+			       const unsigned char *buf, const int start_page)
+{
+	struct ei_device *ei_local = netdev_priv(dev);
+	void __iomem *nic_base = ei_local->mem;
+	unsigned long dma_start;
+
+	/* Round the count up for word writes. Do we need to do this?
+	 * What effect will an odd byte count have on the 8390?  I
+	 * should check someday.
+	 */
+	if (ei_local->word16 && (count & 0x01))
+		count++;
+
+	/* This *shouldn't* happen. If it does, it's the last thing
+	 * you'll see
+	 */
+	if (ei_local->dmaing) {
+		netdev_err(dev,
+			   "DMAing conflict in %s [DMAstat:%d][irqlock:%d]\n",
+			   __func__,
+			   ei_local->dmaing, ei_local->irqlock);
+		return;
+	}
+
+	ei_local->dmaing |= 0x01;
+	/* We should already be in page 0, but to be safe... */
+	ei_outb(E8390_PAGE0 + E8390_START + E8390_NODMA, nic_base + NE_CMD);
+
+	ei_outb(ENISR_RDC, nic_base + EN0_ISR);
+
+	/* Now the normal output. */
+	ei_outb(count & 0xff, nic_base + EN0_RCNTLO);
+	ei_outb(count >> 8, nic_base + EN0_RCNTHI);
+	ei_outb(0x00, nic_base + EN0_RSARLO);
+	ei_outb(start_page, nic_base + EN0_RSARHI);
+
+	ei_outb(E8390_RWRITE + E8390_START, nic_base + NE_CMD);
+
+	xs100_write(dev, buf, count);
+
+	dma_start = jiffies;
+
+	while ((ei_inb(nic_base + EN0_ISR) & ENISR_RDC) == 0) {
+		if (jiffies - dma_start > 2 * HZ / 100) {	/* 20ms */
+			netdev_warn(dev, "timeout waiting for Tx RDC.\n");
+			ei_local->reset_8390(dev);
+			ax_NS8390_reinit(dev);
+			break;
+		}
+	}
+
+	ei_outb(ENISR_RDC, nic_base + EN0_ISR);	/* Ack intr. */
+	ei_local->dmaing &= ~0x01;
+}
+
+static int xsurf100_probe(struct zorro_dev *zdev,
+			  const struct zorro_device_id *ent)
+{
+	struct platform_device *pdev;
+	struct xsurf100_ax_plat_data ax88796_data;
+	struct resource res[2] = {
+		DEFINE_RES_NAMED(IRQ_AMIGA_PORTS, 1, NULL,
+				 IORESOURCE_IRQ | IORESOURCE_IRQ_SHAREABLE),
+		DEFINE_RES_MEM(zdev->resource.start + XS100_8390_BASE,
+			       4 * 0x20)
+	};
+	int reg;
+	/* This table is referenced in the device structure, so it must
+	 * outlive the scope of xsurf100_probe.
+	 */
+	static u32 reg_offsets[32];
+	int ret = 0;
+
+	/* X-Surf 100 control and 32 bit ring buffer data access areas.
+	 * These resources are not used by the ax88796 driver, so must
+	 * be requested here and passed via platform data.
+	 */
+
+	if (!request_mem_region(zdev->resource.start, 0x100, zdev->name)) {
+		dev_err(&zdev->dev, "cannot reserve X-Surf 100 control registers\n");
+		return -ENXIO;
+	}
+
+	if (!request_mem_region(zdev->resource.start +
+				XS100_8390_DATA32_BASE,
+				XS100_8390_DATA32_SIZE,
+				"X-Surf 100 32-bit data access")) {
+		dev_err(&zdev->dev, "cannot reserve 32-bit area\n");
+		ret = -ENXIO;
+		goto exit_req;
+	}
+
+	for (reg = 0; reg < 0x20; reg++)
+		reg_offsets[reg] = 4 * reg;
+
+	memset(&ax88796_data, 0, sizeof(ax88796_data));
+	ax88796_data.ax.flags = AXFLG_HAS_EEPROM;
+	ax88796_data.ax.wordlength = 2;
+	ax88796_data.ax.dcr_val = 0x48;
+	ax88796_data.ax.rcr_val = 0x40;
+	ax88796_data.ax.reg_offsets = reg_offsets;
+	ax88796_data.ax.check_irq = is_xsurf100_network_irq;
+	ax88796_data.base_regs = ioremap(zdev->resource.start, 0x100);
+
+	/* error handling for ioremap regs */
+	if (!ax88796_data.base_regs) {
+		dev_err(&zdev->dev, "Cannot ioremap area %pR (registers)\n",
+			&zdev->resource);
+
+		ret = -ENXIO;
+		goto exit_req2;
+	}
+
+	ax88796_data.data_area = ioremap(zdev->resource.start +
+			XS100_8390_DATA32_BASE, XS100_8390_DATA32_SIZE);
+
+	/* error handling for ioremap data */
+	if (!ax88796_data.data_area) {
+		dev_err(&zdev->dev,
+			"Cannot ioremap area %pR offset %x (32-bit access)\n",
+			&zdev->resource,  XS100_8390_DATA32_BASE);
+
+		ret = -ENXIO;
+		goto exit_mem;
+	}
+
+	ax88796_data.ax.block_output = xs100_block_output;
+	ax88796_data.ax.block_input = xs100_block_input;
+
+	pdev = platform_device_register_resndata(&zdev->dev, "ax88796",
+						 zdev->slotaddr, res, 2,
+						 &ax88796_data,
+						 sizeof(ax88796_data));
+
+	if (IS_ERR(pdev)) {
+		dev_err(&zdev->dev, "cannot register platform device\n");
+		ret = -ENXIO;
+		goto exit_mem2;
+	}
+
+	zorro_set_drvdata(zdev, pdev);
+
+	if (!ret)
+		return 0;
+
+ exit_mem2:
+	iounmap(ax88796_data.data_area);
+
+ exit_mem:
+	iounmap(ax88796_data.base_regs);
+
+ exit_req2:
+	release_mem_region(zdev->resource.start + XS100_8390_DATA32_BASE,
+			   XS100_8390_DATA32_SIZE);
+
+ exit_req:
+	release_mem_region(zdev->resource.start, 0x100);
+
+	return ret;
+}
+
+static void xsurf100_remove(struct zorro_dev *zdev)
+{
+	struct platform_device *pdev = zorro_get_drvdata(zdev);
+	struct xsurf100_ax_plat_data *xs100 = dev_get_platdata(&pdev->dev);
+
+	platform_device_unregister(pdev);
+
+	iounmap(xs100->base_regs);
+	release_mem_region(zdev->resource.start, 0x100);
+	iounmap(xs100->data_area);
+	release_mem_region(zdev->resource.start + XS100_8390_DATA32_BASE,
+			   XS100_8390_DATA32_SIZE);
+}
+
+static const struct zorro_device_id xsurf100_zorro_tbl[] = {
+	{ ZORRO_PROD_INDIVIDUAL_COMPUTERS_X_SURF100, },
+	{ 0 }
+};
+
+MODULE_DEVICE_TABLE(zorro, xsurf100_zorro_tbl);
+
+static struct zorro_driver xsurf100_driver = {
+	.name           = "xsurf100",
+	.id_table       = xsurf100_zorro_tbl,
+	.probe          = xsurf100_probe,
+	.remove         = xsurf100_remove,
+};
+
+module_driver(xsurf100_driver, zorro_register_driver, zorro_unregister_driver);
+
+MODULE_DESCRIPTION("X-Surf 100 driver");
+MODULE_AUTHOR("Michael Karcher <kernel@mkarcher.dialup.fu-berlin.de>");
+MODULE_LICENSE("GPL v2");
diff --git a/drivers/net/ethernet/8390/zorro8390.c b/drivers/net/ethernet/8390/zorro8390.c
index 6d93956b293b..c24dd4fe7a10 100644
--- a/drivers/net/ethernet/8390/zorro8390.c
+++ b/drivers/net/ethernet/8390/zorro8390.c
@@ -1,3 +1,4 @@
+// SPDX-License-Identifier: GPL-2.0-only
 /*
  *  Amiga Linux/m68k and Linux/PPC Zorro NS8390 Ethernet Driver
  *
@@ -9,12 +10,6 @@
  *
  *  ---------------------------------------------------------------------------
  *
- *  This file is subject to the terms and conditions of the GNU General Public
- *  License.  See the file COPYING in the main directory of the Linux
- *  distribution for more details.
- *
- *  ---------------------------------------------------------------------------
- *
  *  The Ariadne II and X-Surf are Zorro-II boards containing Realtek RTL8019AS
  *  Ethernet Controllers.
  */
@@ -44,8 +39,6 @@
 static const char version[] =
 	"8390.c:v1.10cvs 9/23/94 Donald Becker (becker@cesdis.gsfc.nasa.gov)\n";
 
-static u32 zorro8390_msg_enable;
-
 #include "lib8390.c"
 
 #define DRV_NAME	"zorro8390"
@@ -296,7 +289,6 @@ static int zorro8390_init(struct net_device *dev, unsigned long board,
 	int err;
 	unsigned char SA_prom[32];
 	int start_page, stop_page;
-	struct ei_device *ei_local = netdev_priv(dev);
 	static u32 zorro8390_offsets[16] = {
 		0x00, 0x02, 0x04, 0x06, 0x08, 0x0a, 0x0c, 0x0e,
 		0x10, 0x12, 0x14, 0x16, 0x18, 0x1a, 0x1c, 0x1e,
@@ -367,8 +359,7 @@ static int zorro8390_init(struct net_device *dev, unsigned long board,
 	if (i)
 		return i;
 
-	for (i = 0; i < ETH_ALEN; i++)
-		dev->dev_addr[i] = SA_prom[i];
+	eth_hw_addr_set(dev, SA_prom);
 
 	pr_debug("Found ethernet address: %pM\n", dev->dev_addr);
 
@@ -388,8 +379,6 @@ static int zorro8390_init(struct net_device *dev, unsigned long board,
 	dev->netdev_ops = &zorro8390_netdev_ops;
 	__NS8390_init(dev, 0);
 
-	ei_local->msg_enable = zorro8390_msg_enable;
-
 	err = register_netdev(dev);
 	if (err) {
 		free_irq(IRQ_AMIGA_PORTS, dev);
@@ -454,4 +443,5 @@ static void __exit zorro8390_cleanup_module(void)
 module_init(zorro8390_init_module);
 module_exit(zorro8390_cleanup_module);
 
+MODULE_DESCRIPTION("Zorro NS8390-based ethernet driver");
 MODULE_LICENSE("GPL");
diff --git a/drivers/net/ethernet/Kconfig b/drivers/net/ethernet/Kconfig
index edae15ac0e98..aead145dd91d 100644
--- a/drivers/net/ethernet/Kconfig
+++ b/drivers/net/ethernet/Kconfig
@@ -1,3 +1,4 @@
+# SPDX-License-Identifier: GPL-2.0-only
 #
 # Ethernet LAN device configuration
 #
@@ -6,7 +7,7 @@ menuconfig ETHERNET
 	bool "Ethernet driver support"
 	depends on NET
 	default y
-	---help---
+	help
 	  This section contains all the Ethernet device drivers.
 
 if ETHERNET
@@ -14,13 +15,12 @@ if ETHERNET
 config MDIO
 	tristate
 
-config SUNGEM_PHY
-	tristate
-
 source "drivers/net/ethernet/3com/Kconfig"
+source "drivers/net/ethernet/actions/Kconfig"
 source "drivers/net/ethernet/adaptec/Kconfig"
 source "drivers/net/ethernet/aeroflex/Kconfig"
 source "drivers/net/ethernet/agere/Kconfig"
+source "drivers/net/ethernet/airoha/Kconfig"
 source "drivers/net/ethernet/alacritech/Kconfig"
 source "drivers/net/ethernet/allwinner/Kconfig"
 source "drivers/net/ethernet/alteon/Kconfig"
@@ -31,23 +31,14 @@ source "drivers/net/ethernet/apm/Kconfig"
 source "drivers/net/ethernet/apple/Kconfig"
 source "drivers/net/ethernet/aquantia/Kconfig"
 source "drivers/net/ethernet/arc/Kconfig"
+source "drivers/net/ethernet/asix/Kconfig"
 source "drivers/net/ethernet/atheros/Kconfig"
-source "drivers/net/ethernet/aurora/Kconfig"
-source "drivers/net/ethernet/cadence/Kconfig"
-source "drivers/net/ethernet/adi/Kconfig"
-source "drivers/net/ethernet/broadcom/Kconfig"
-source "drivers/net/ethernet/brocade/Kconfig"
-source "drivers/net/ethernet/calxeda/Kconfig"
-source "drivers/net/ethernet/cavium/Kconfig"
-source "drivers/net/ethernet/chelsio/Kconfig"
-source "drivers/net/ethernet/cirrus/Kconfig"
-source "drivers/net/ethernet/cisco/Kconfig"
 
 config CX_ECAT
 	tristate "Beckhoff CX5020 EtherCAT master support"
 	depends on PCI
 	depends on X86 || COMPILE_TEST
-	---help---
+	help
 	  Driver for EtherCAT master module located on CCAT FPGA
 	  that can be found on Beckhoff CX5020, and possibly other of CX
 	  Beckhoff CX series industrial PCs.
@@ -55,13 +46,21 @@ config CX_ECAT
 	  To compile this driver as a module, choose M here. The module
 	  will be called ec_bhf.
 
+source "drivers/net/ethernet/broadcom/Kconfig"
+source "drivers/net/ethernet/cadence/Kconfig"
+source "drivers/net/ethernet/calxeda/Kconfig"
+source "drivers/net/ethernet/cavium/Kconfig"
+source "drivers/net/ethernet/chelsio/Kconfig"
+source "drivers/net/ethernet/cirrus/Kconfig"
+source "drivers/net/ethernet/cisco/Kconfig"
+source "drivers/net/ethernet/cortina/Kconfig"
 source "drivers/net/ethernet/davicom/Kconfig"
 
 config DNET
 	tristate "Dave ethernet support (DNET)"
 	depends on HAS_IOMEM
 	select PHYLIB
-	---help---
+	help
 	  The Dave ethernet interface (DNET) is found on Qong Board FPGA.
 	  Say Y to include support for the DNET chip.
 
@@ -71,16 +70,18 @@ config DNET
 source "drivers/net/ethernet/dec/Kconfig"
 source "drivers/net/ethernet/dlink/Kconfig"
 source "drivers/net/ethernet/emulex/Kconfig"
+source "drivers/net/ethernet/engleder/Kconfig"
 source "drivers/net/ethernet/ezchip/Kconfig"
-source "drivers/net/ethernet/neterion/Kconfig"
 source "drivers/net/ethernet/faraday/Kconfig"
 source "drivers/net/ethernet/freescale/Kconfig"
 source "drivers/net/ethernet/fujitsu/Kconfig"
+source "drivers/net/ethernet/fungible/Kconfig"
+source "drivers/net/ethernet/google/Kconfig"
 source "drivers/net/ethernet/hisilicon/Kconfig"
-source "drivers/net/ethernet/hp/Kconfig"
+source "drivers/net/ethernet/huawei/Kconfig"
+source "drivers/net/ethernet/i825xx/Kconfig"
 source "drivers/net/ethernet/ibm/Kconfig"
 source "drivers/net/ethernet/intel/Kconfig"
-source "drivers/net/ethernet/i825xx/Kconfig"
 source "drivers/net/ethernet/xscale/Kconfig"
 
 config JME
@@ -88,7 +89,7 @@ config JME
 	depends on PCI
 	select CRC32
 	select MII
-	---help---
+	help
 	  This driver supports the PCI-Express gigabit ethernet adapters
 	  based on JMicron JMC250 chipset.
 
@@ -97,22 +98,36 @@ config JME
 
 config KORINA
 	tristate "Korina (IDT RC32434) Ethernet support"
-	depends on MIKROTIK_RB532
-	---help---
+	depends on MIKROTIK_RB532 || COMPILE_TEST
+	select CRC32
+	select MII
+	help
 	  If you have a Mikrotik RouterBoard 500 or IDT RC32434
 	  based system say Y. Otherwise say N.
 
 config LANTIQ_ETOP
 	tristate "Lantiq SoC ETOP driver"
 	depends on SOC_TYPE_XWAY
-	---help---
+	help
 	  Support for the MII0 inside the Lantiq SoC
 
+config LANTIQ_XRX200
+	tristate "Lantiq / Intel xRX200 PMAC network driver"
+	depends on SOC_TYPE_XWAY
+	help
+	  Support for the PMAC of the Gigabit switch (GSWIP) inside the
+	  Lantiq / Intel VRX200 VDSL SoC
+
+source "drivers/net/ethernet/adi/Kconfig"
+source "drivers/net/ethernet/litex/Kconfig"
 source "drivers/net/ethernet/marvell/Kconfig"
 source "drivers/net/ethernet/mediatek/Kconfig"
 source "drivers/net/ethernet/mellanox/Kconfig"
+source "drivers/net/ethernet/meta/Kconfig"
 source "drivers/net/ethernet/micrel/Kconfig"
 source "drivers/net/ethernet/microchip/Kconfig"
+source "drivers/net/ethernet/mscc/Kconfig"
+source "drivers/net/ethernet/microsoft/Kconfig"
 source "drivers/net/ethernet/moxa/Kconfig"
 source "drivers/net/ethernet/myricom/Kconfig"
 
@@ -121,25 +136,15 @@ config FEALNX
 	depends on PCI
 	select CRC32
 	select MII
-	---help---
+	help
 	  Say Y here to support the Myson MTD-800 family of PCI-based Ethernet
 	  cards. <http://www.myson.com.tw/>
 
+source "drivers/net/ethernet/ni/Kconfig"
 source "drivers/net/ethernet/natsemi/Kconfig"
+source "drivers/net/ethernet/neterion/Kconfig"
 source "drivers/net/ethernet/netronome/Kconfig"
 source "drivers/net/ethernet/8390/Kconfig"
-
-config NET_NETX
-	tristate "NetX Ethernet support"
-	select MII
-	depends on ARCH_NETX
-	---help---
-	  This is support for the Hilscher netX builtin Ethernet ports
-
-	  To compile this driver as a module, choose M here. The module
-	  will be called netx-eth.
-
-source "drivers/net/ethernet/nuvoton/Kconfig"
 source "drivers/net/ethernet/nvidia/Kconfig"
 source "drivers/net/ethernet/nxp/Kconfig"
 source "drivers/net/ethernet/oki-semi/Kconfig"
@@ -151,35 +156,52 @@ config ETHOC
 	select PHYLIB
 	select CRC32
 	select BITREVERSE
-	---help---
+	help
 	  Say Y here if you want to use the OpenCores 10/100 Mbps Ethernet MAC.
 
+config OA_TC6
+	tristate "OPEN Alliance TC6 10BASE-T1x MAC-PHY support" if COMPILE_TEST
+	depends on SPI
+	select PHYLIB
+	help
+	  This library implements OPEN Alliance TC6 10BASE-T1x MAC-PHY
+	  Serial Interface protocol for supporting 10BASE-T1x MAC-PHYs.
+
+	  To know the implementation details, refer documentation in
+	  <file:Documentation/networking/oa-tc6-framework.rst>.
+
 source "drivers/net/ethernet/packetengines/Kconfig"
 source "drivers/net/ethernet/pasemi/Kconfig"
+source "drivers/net/ethernet/pensando/Kconfig"
 source "drivers/net/ethernet/qlogic/Kconfig"
+source "drivers/net/ethernet/brocade/Kconfig"
 source "drivers/net/ethernet/qualcomm/Kconfig"
+source "drivers/net/ethernet/rdc/Kconfig"
 source "drivers/net/ethernet/realtek/Kconfig"
 source "drivers/net/ethernet/renesas/Kconfig"
-source "drivers/net/ethernet/rdc/Kconfig"
 source "drivers/net/ethernet/rocker/Kconfig"
 source "drivers/net/ethernet/samsung/Kconfig"
 source "drivers/net/ethernet/seeq/Kconfig"
+source "drivers/net/ethernet/sgi/Kconfig"
 source "drivers/net/ethernet/silan/Kconfig"
 source "drivers/net/ethernet/sis/Kconfig"
 source "drivers/net/ethernet/sfc/Kconfig"
-source "drivers/net/ethernet/sgi/Kconfig"
 source "drivers/net/ethernet/smsc/Kconfig"
+source "drivers/net/ethernet/socionext/Kconfig"
+source "drivers/net/ethernet/spacemit/Kconfig"
 source "drivers/net/ethernet/stmicro/Kconfig"
 source "drivers/net/ethernet/sun/Kconfig"
+source "drivers/net/ethernet/sunplus/Kconfig"
+source "drivers/net/ethernet/synopsys/Kconfig"
 source "drivers/net/ethernet/tehuti/Kconfig"
 source "drivers/net/ethernet/ti/Kconfig"
-source "drivers/net/ethernet/tile/Kconfig"
 source "drivers/net/ethernet/toshiba/Kconfig"
 source "drivers/net/ethernet/tundra/Kconfig"
+source "drivers/net/ethernet/vertexcom/Kconfig"
 source "drivers/net/ethernet/via/Kconfig"
+source "drivers/net/ethernet/wangxun/Kconfig"
 source "drivers/net/ethernet/wiznet/Kconfig"
 source "drivers/net/ethernet/xilinx/Kconfig"
 source "drivers/net/ethernet/xircom/Kconfig"
-source "drivers/net/ethernet/synopsys/Kconfig"
 
 endif # ETHERNET
diff --git a/drivers/net/ethernet/Makefile b/drivers/net/ethernet/Makefile
index bf7f4502cabc..998dd628b202 100644
--- a/drivers/net/ethernet/Makefile
+++ b/drivers/net/ethernet/Makefile
@@ -1,12 +1,16 @@
+# SPDX-License-Identifier: GPL-2.0
 #
 # Makefile for the Linux network Ethernet device drivers.
 #
 
 obj-$(CONFIG_NET_VENDOR_3COM) += 3com/
 obj-$(CONFIG_NET_VENDOR_8390) += 8390/
+obj-$(CONFIG_NET_VENDOR_ACTIONS) += actions/
 obj-$(CONFIG_NET_VENDOR_ADAPTEC) += adaptec/
 obj-$(CONFIG_GRETH) += aeroflex/
+obj-$(CONFIG_NET_VENDOR_ADI) += adi/
 obj-$(CONFIG_NET_VENDOR_AGERE) += agere/
+obj-$(CONFIG_NET_VENDOR_AIROHA) += airoha/
 obj-$(CONFIG_NET_VENDOR_ALACRITECH) += alacritech/
 obj-$(CONFIG_NET_VENDOR_ALLWINNER) += allwinner/
 obj-$(CONFIG_NET_VENDOR_ALTEON) += alteon/
@@ -17,10 +21,9 @@ obj-$(CONFIG_NET_XGENE) += apm/
 obj-$(CONFIG_NET_VENDOR_APPLE) += apple/
 obj-$(CONFIG_NET_VENDOR_AQUANTIA) += aquantia/
 obj-$(CONFIG_NET_VENDOR_ARC) += arc/
+obj-$(CONFIG_NET_VENDOR_ASIX) += asix/
 obj-$(CONFIG_NET_VENDOR_ATHEROS) += atheros/
-obj-$(CONFIG_NET_VENDOR_AURORA) += aurora/
-obj-$(CONFIG_NET_CADENCE) += cadence/
-obj-$(CONFIG_NET_BFIN) += adi/
+obj-$(CONFIG_NET_VENDOR_CADENCE) += cadence/
 obj-$(CONFIG_NET_VENDOR_BROADCOM) += broadcom/
 obj-$(CONFIG_NET_VENDOR_BROCADE) += brocade/
 obj-$(CONFIG_NET_CALXEDA_XGMAC) += calxeda/
@@ -28,43 +31,51 @@ obj-$(CONFIG_NET_VENDOR_CAVIUM) += cavium/
 obj-$(CONFIG_NET_VENDOR_CHELSIO) += chelsio/
 obj-$(CONFIG_NET_VENDOR_CIRRUS) += cirrus/
 obj-$(CONFIG_NET_VENDOR_CISCO) += cisco/
+obj-$(CONFIG_NET_VENDOR_CORTINA) += cortina/
 obj-$(CONFIG_CX_ECAT) += ec_bhf.o
 obj-$(CONFIG_DM9000) += davicom/
 obj-$(CONFIG_DNET) += dnet.o
 obj-$(CONFIG_NET_VENDOR_DEC) += dec/
 obj-$(CONFIG_NET_VENDOR_DLINK) += dlink/
 obj-$(CONFIG_NET_VENDOR_EMULEX) += emulex/
+obj-$(CONFIG_NET_VENDOR_ENGLEDER) += engleder/
 obj-$(CONFIG_NET_VENDOR_EZCHIP) += ezchip/
-obj-$(CONFIG_NET_VENDOR_EXAR) += neterion/
 obj-$(CONFIG_NET_VENDOR_FARADAY) += faraday/
 obj-$(CONFIG_NET_VENDOR_FREESCALE) += freescale/
 obj-$(CONFIG_NET_VENDOR_FUJITSU) += fujitsu/
+obj-$(CONFIG_NET_VENDOR_FUNGIBLE) += fungible/
+obj-$(CONFIG_NET_VENDOR_GOOGLE) += google/
 obj-$(CONFIG_NET_VENDOR_HISILICON) += hisilicon/
-obj-$(CONFIG_NET_VENDOR_HP) += hp/
+obj-$(CONFIG_NET_VENDOR_HUAWEI) += huawei/
 obj-$(CONFIG_NET_VENDOR_IBM) += ibm/
 obj-$(CONFIG_NET_VENDOR_INTEL) += intel/
 obj-$(CONFIG_NET_VENDOR_I825XX) += i825xx/
+obj-$(CONFIG_NET_VENDOR_MICROSOFT) += microsoft/
 obj-$(CONFIG_NET_VENDOR_XSCALE) += xscale/
 obj-$(CONFIG_JME) += jme.o
 obj-$(CONFIG_KORINA) += korina.o
 obj-$(CONFIG_LANTIQ_ETOP) += lantiq_etop.o
+obj-$(CONFIG_LANTIQ_XRX200) += lantiq_xrx200.o
+obj-$(CONFIG_NET_VENDOR_LITEX) += litex/
 obj-$(CONFIG_NET_VENDOR_MARVELL) += marvell/
 obj-$(CONFIG_NET_VENDOR_MEDIATEK) += mediatek/
 obj-$(CONFIG_NET_VENDOR_MELLANOX) += mellanox/
+obj-$(CONFIG_NET_VENDOR_META) += meta/
 obj-$(CONFIG_NET_VENDOR_MICREL) += micrel/
 obj-$(CONFIG_NET_VENDOR_MICROCHIP) += microchip/
+obj-$(CONFIG_NET_VENDOR_MICROSEMI) += mscc/
 obj-$(CONFIG_NET_VENDOR_MOXART) += moxa/
 obj-$(CONFIG_NET_VENDOR_MYRI) += myricom/
 obj-$(CONFIG_FEALNX) += fealnx.o
 obj-$(CONFIG_NET_VENDOR_NATSEMI) += natsemi/
+obj-$(CONFIG_NET_VENDOR_NETERION) += neterion/
 obj-$(CONFIG_NET_VENDOR_NETRONOME) += netronome/
-obj-$(CONFIG_NET_NETX) += netx-eth.o
-obj-$(CONFIG_NET_VENDOR_NUVOTON) += nuvoton/
+obj-$(CONFIG_NET_VENDOR_NI) += ni/
 obj-$(CONFIG_NET_VENDOR_NVIDIA) += nvidia/
 obj-$(CONFIG_LPC_ENET) += nxp/
 obj-$(CONFIG_NET_VENDOR_OKI) += oki-semi/
 obj-$(CONFIG_ETHOC) += ethoc.o
-obj-$(CONFIG_NET_PACKET_ENGINE) += packetengines/
+obj-$(CONFIG_NET_VENDOR_PACKET_ENGINES) += packetengines/
 obj-$(CONFIG_NET_VENDOR_PASEMI) += pasemi/
 obj-$(CONFIG_NET_VENDOR_QLOGIC) += qlogic/
 obj-$(CONFIG_NET_VENDOR_QUALCOMM) += qualcomm/
@@ -76,19 +87,24 @@ obj-$(CONFIG_NET_VENDOR_SAMSUNG) += samsung/
 obj-$(CONFIG_NET_VENDOR_SEEQ) += seeq/
 obj-$(CONFIG_NET_VENDOR_SILAN) += silan/
 obj-$(CONFIG_NET_VENDOR_SIS) += sis/
-obj-$(CONFIG_SFC) += sfc/
-obj-$(CONFIG_SFC_FALCON) += sfc/falcon/
+obj-$(CONFIG_NET_VENDOR_SOLARFLARE) += sfc/
 obj-$(CONFIG_NET_VENDOR_SGI) += sgi/
 obj-$(CONFIG_NET_VENDOR_SMSC) += smsc/
+obj-$(CONFIG_NET_VENDOR_SOCIONEXT) += socionext/
+obj-$(CONFIG_NET_VENDOR_SPACEMIT) += spacemit/
 obj-$(CONFIG_NET_VENDOR_STMICRO) += stmicro/
 obj-$(CONFIG_NET_VENDOR_SUN) += sun/
+obj-$(CONFIG_NET_VENDOR_SUNPLUS) += sunplus/
 obj-$(CONFIG_NET_VENDOR_TEHUTI) += tehuti/
 obj-$(CONFIG_NET_VENDOR_TI) += ti/
-obj-$(CONFIG_TILE_NET) += tile/
 obj-$(CONFIG_NET_VENDOR_TOSHIBA) += toshiba/
 obj-$(CONFIG_NET_VENDOR_TUNDRA) += tundra/
+obj-$(CONFIG_NET_VENDOR_VERTEXCOM) += vertexcom/
 obj-$(CONFIG_NET_VENDOR_VIA) += via/
+obj-$(CONFIG_NET_VENDOR_WANGXUN) += wangxun/
 obj-$(CONFIG_NET_VENDOR_WIZNET) += wiznet/
 obj-$(CONFIG_NET_VENDOR_XILINX) += xilinx/
 obj-$(CONFIG_NET_VENDOR_XIRCOM) += xircom/
 obj-$(CONFIG_NET_VENDOR_SYNOPSYS) += synopsys/
+obj-$(CONFIG_NET_VENDOR_PENSANDO) += pensando/
+obj-$(CONFIG_OA_TC6) += oa_tc6.o
diff --git a/drivers/net/ethernet/actions/Kconfig b/drivers/net/ethernet/actions/Kconfig
new file mode 100644
index 000000000000..f630cac2ab6c
--- /dev/null
+++ b/drivers/net/ethernet/actions/Kconfig
@@ -0,0 +1,26 @@
+# SPDX-License-Identifier: GPL-2.0-only
+
+config NET_VENDOR_ACTIONS
+	bool "Actions Semi devices"
+	depends on ARCH_ACTIONS || COMPILE_TEST
+	default ARCH_ACTIONS
+	help
+	  If you have a network (Ethernet) card belonging to this class, say Y.
+
+	  Note that the answer to this question doesn't directly affect the
+	  kernel: saying N will just cause the configurator to skip all the
+	  questions about Actions Semi devices.  If you say Y, you will be
+	  asked for your specific card in the following questions.
+
+if NET_VENDOR_ACTIONS
+
+config OWL_EMAC
+	tristate "Actions Semi Owl Ethernet MAC support"
+	select PHYLIB
+	help
+	  This driver supports the Actions Semi Ethernet Media Access
+	  Controller (EMAC) found on the S500 and S900 SoCs.  The controller
+	  is compliant with the IEEE 802.3 CSMA/CD standard and supports
+	  both half-duplex and full-duplex operation modes at 10/100 Mb/s.
+
+endif # NET_VENDOR_ACTIONS
diff --git a/drivers/net/ethernet/actions/Makefile b/drivers/net/ethernet/actions/Makefile
new file mode 100644
index 000000000000..fde8001d538a
--- /dev/null
+++ b/drivers/net/ethernet/actions/Makefile
@@ -0,0 +1,6 @@
+# SPDX-License-Identifier: GPL-2.0-only
+#
+# Makefile for the Actions Semi Owl SoCs built-in ethernet macs
+#
+
+obj-$(CONFIG_OWL_EMAC) += owl-emac.o
diff --git a/drivers/net/ethernet/actions/owl-emac.c b/drivers/net/ethernet/actions/owl-emac.c
new file mode 100644
index 000000000000..0a08da799255
--- /dev/null
+++ b/drivers/net/ethernet/actions/owl-emac.c
@@ -0,0 +1,1612 @@
+// SPDX-License-Identifier: GPL-2.0-or-later
+/*
+ * Actions Semi Owl SoCs Ethernet MAC driver
+ *
+ * Copyright (c) 2012 Actions Semi Inc.
+ * Copyright (c) 2021 Cristian Ciocaltea <cristian.ciocaltea@gmail.com>
+ */
+
+#include <linux/circ_buf.h>
+#include <linux/clk.h>
+#include <linux/dma-mapping.h>
+#include <linux/etherdevice.h>
+#include <linux/of_mdio.h>
+#include <linux/of_net.h>
+#include <linux/platform_device.h>
+#include <linux/pm.h>
+#include <linux/reset.h>
+
+#include "owl-emac.h"
+
+#define OWL_EMAC_DEFAULT_MSG_ENABLE	(NETIF_MSG_DRV | \
+					 NETIF_MSG_PROBE | \
+					 NETIF_MSG_LINK)
+
+static u32 owl_emac_reg_read(struct owl_emac_priv *priv, u32 reg)
+{
+	return readl(priv->base + reg);
+}
+
+static void owl_emac_reg_write(struct owl_emac_priv *priv, u32 reg, u32 data)
+{
+	writel(data, priv->base + reg);
+}
+
+static u32 owl_emac_reg_update(struct owl_emac_priv *priv,
+			       u32 reg, u32 mask, u32 val)
+{
+	u32 data, old_val;
+
+	data = owl_emac_reg_read(priv, reg);
+	old_val = data & mask;
+
+	data &= ~mask;
+	data |= val & mask;
+
+	owl_emac_reg_write(priv, reg, data);
+
+	return old_val;
+}
+
+static void owl_emac_reg_set(struct owl_emac_priv *priv, u32 reg, u32 bits)
+{
+	owl_emac_reg_update(priv, reg, bits, bits);
+}
+
+static void owl_emac_reg_clear(struct owl_emac_priv *priv, u32 reg, u32 bits)
+{
+	owl_emac_reg_update(priv, reg, bits, 0);
+}
+
+static struct device *owl_emac_get_dev(struct owl_emac_priv *priv)
+{
+	return priv->netdev->dev.parent;
+}
+
+static void owl_emac_irq_enable(struct owl_emac_priv *priv)
+{
+	/* Enable all interrupts except TU.
+	 *
+	 * Note the NIE and AIE bits shall also be set in order to actually
+	 * enable the selected interrupts.
+	 */
+	owl_emac_reg_write(priv, OWL_EMAC_REG_MAC_CSR7,
+			   OWL_EMAC_BIT_MAC_CSR7_NIE |
+			   OWL_EMAC_BIT_MAC_CSR7_AIE |
+			   OWL_EMAC_BIT_MAC_CSR7_ALL_NOT_TUE);
+}
+
+static void owl_emac_irq_disable(struct owl_emac_priv *priv)
+{
+	/* Disable all interrupts.
+	 *
+	 * WARNING: Unset only the NIE and AIE bits in CSR7 to workaround an
+	 * unexpected side effect (MAC hardware bug?!) where some bits in the
+	 * status register (CSR5) are cleared automatically before being able
+	 * to read them via owl_emac_irq_clear().
+	 */
+	owl_emac_reg_write(priv, OWL_EMAC_REG_MAC_CSR7,
+			   OWL_EMAC_BIT_MAC_CSR7_ALL_NOT_TUE);
+}
+
+static u32 owl_emac_irq_status(struct owl_emac_priv *priv)
+{
+	return owl_emac_reg_read(priv, OWL_EMAC_REG_MAC_CSR5);
+}
+
+static u32 owl_emac_irq_clear(struct owl_emac_priv *priv)
+{
+	u32 val = owl_emac_irq_status(priv);
+
+	owl_emac_reg_write(priv, OWL_EMAC_REG_MAC_CSR5, val);
+
+	return val;
+}
+
+static dma_addr_t owl_emac_dma_map_rx(struct owl_emac_priv *priv,
+				      struct sk_buff *skb)
+{
+	struct device *dev = owl_emac_get_dev(priv);
+
+	/* Buffer pointer for the RX DMA descriptor must be word aligned. */
+	return dma_map_single(dev, skb_tail_pointer(skb),
+			      skb_tailroom(skb), DMA_FROM_DEVICE);
+}
+
+static void owl_emac_dma_unmap_rx(struct owl_emac_priv *priv,
+				  struct sk_buff *skb, dma_addr_t dma_addr)
+{
+	struct device *dev = owl_emac_get_dev(priv);
+
+	dma_unmap_single(dev, dma_addr, skb_tailroom(skb), DMA_FROM_DEVICE);
+}
+
+static dma_addr_t owl_emac_dma_map_tx(struct owl_emac_priv *priv,
+				      struct sk_buff *skb)
+{
+	struct device *dev = owl_emac_get_dev(priv);
+
+	return dma_map_single(dev, skb->data, skb_headlen(skb), DMA_TO_DEVICE);
+}
+
+static void owl_emac_dma_unmap_tx(struct owl_emac_priv *priv,
+				  struct sk_buff *skb, dma_addr_t dma_addr)
+{
+	struct device *dev = owl_emac_get_dev(priv);
+
+	dma_unmap_single(dev, dma_addr, skb_headlen(skb), DMA_TO_DEVICE);
+}
+
+static unsigned int owl_emac_ring_num_unused(struct owl_emac_ring *ring)
+{
+	return CIRC_SPACE(ring->head, ring->tail, ring->size);
+}
+
+static unsigned int owl_emac_ring_get_next(struct owl_emac_ring *ring,
+					   unsigned int cur)
+{
+	return (cur + 1) & (ring->size - 1);
+}
+
+static void owl_emac_ring_push_head(struct owl_emac_ring *ring)
+{
+	ring->head = owl_emac_ring_get_next(ring, ring->head);
+}
+
+static void owl_emac_ring_pop_tail(struct owl_emac_ring *ring)
+{
+	ring->tail = owl_emac_ring_get_next(ring, ring->tail);
+}
+
+static struct sk_buff *owl_emac_alloc_skb(struct net_device *netdev)
+{
+	struct sk_buff *skb;
+	int offset;
+
+	skb = netdev_alloc_skb(netdev, OWL_EMAC_RX_FRAME_MAX_LEN +
+			       OWL_EMAC_SKB_RESERVE);
+	if (unlikely(!skb))
+		return NULL;
+
+	/* Ensure 4 bytes DMA alignment. */
+	offset = ((uintptr_t)skb->data) & (OWL_EMAC_SKB_ALIGN - 1);
+	if (unlikely(offset))
+		skb_reserve(skb, OWL_EMAC_SKB_ALIGN - offset);
+
+	return skb;
+}
+
+static int owl_emac_ring_prepare_rx(struct owl_emac_priv *priv)
+{
+	struct owl_emac_ring *ring = &priv->rx_ring;
+	struct device *dev = owl_emac_get_dev(priv);
+	struct net_device *netdev = priv->netdev;
+	struct owl_emac_ring_desc *desc;
+	struct sk_buff *skb;
+	dma_addr_t dma_addr;
+	int i;
+
+	for (i = 0; i < ring->size; i++) {
+		skb = owl_emac_alloc_skb(netdev);
+		if (!skb)
+			return -ENOMEM;
+
+		dma_addr = owl_emac_dma_map_rx(priv, skb);
+		if (dma_mapping_error(dev, dma_addr)) {
+			dev_kfree_skb(skb);
+			return -ENOMEM;
+		}
+
+		desc = &ring->descs[i];
+		desc->status = OWL_EMAC_BIT_RDES0_OWN;
+		desc->control = skb_tailroom(skb) & OWL_EMAC_MSK_RDES1_RBS1;
+		desc->buf_addr = dma_addr;
+		desc->reserved = 0;
+
+		ring->skbs[i] = skb;
+		ring->skbs_dma[i] = dma_addr;
+	}
+
+	desc->control |= OWL_EMAC_BIT_RDES1_RER;
+
+	ring->head = 0;
+	ring->tail = 0;
+
+	return 0;
+}
+
+static void owl_emac_ring_prepare_tx(struct owl_emac_priv *priv)
+{
+	struct owl_emac_ring *ring = &priv->tx_ring;
+	struct owl_emac_ring_desc *desc;
+	int i;
+
+	for (i = 0; i < ring->size; i++) {
+		desc = &ring->descs[i];
+
+		desc->status = 0;
+		desc->control = OWL_EMAC_BIT_TDES1_IC;
+		desc->buf_addr = 0;
+		desc->reserved = 0;
+	}
+
+	desc->control |= OWL_EMAC_BIT_TDES1_TER;
+
+	memset(ring->skbs_dma, 0, sizeof(dma_addr_t) * ring->size);
+
+	ring->head = 0;
+	ring->tail = 0;
+}
+
+static void owl_emac_ring_unprepare_rx(struct owl_emac_priv *priv)
+{
+	struct owl_emac_ring *ring = &priv->rx_ring;
+	int i;
+
+	for (i = 0; i < ring->size; i++) {
+		ring->descs[i].status = 0;
+
+		if (!ring->skbs_dma[i])
+			continue;
+
+		owl_emac_dma_unmap_rx(priv, ring->skbs[i], ring->skbs_dma[i]);
+		ring->skbs_dma[i] = 0;
+
+		dev_kfree_skb(ring->skbs[i]);
+		ring->skbs[i] = NULL;
+	}
+}
+
+static void owl_emac_ring_unprepare_tx(struct owl_emac_priv *priv)
+{
+	struct owl_emac_ring *ring = &priv->tx_ring;
+	int i;
+
+	for (i = 0; i < ring->size; i++) {
+		ring->descs[i].status = 0;
+
+		if (!ring->skbs_dma[i])
+			continue;
+
+		owl_emac_dma_unmap_tx(priv, ring->skbs[i], ring->skbs_dma[i]);
+		ring->skbs_dma[i] = 0;
+
+		dev_kfree_skb(ring->skbs[i]);
+		ring->skbs[i] = NULL;
+	}
+}
+
+static int owl_emac_ring_alloc(struct device *dev, struct owl_emac_ring *ring,
+			       unsigned int size)
+{
+	ring->descs = dmam_alloc_coherent(dev,
+					  sizeof(struct owl_emac_ring_desc) * size,
+					  &ring->descs_dma, GFP_KERNEL);
+	if (!ring->descs)
+		return -ENOMEM;
+
+	ring->skbs = devm_kcalloc(dev, size, sizeof(struct sk_buff *),
+				  GFP_KERNEL);
+	if (!ring->skbs)
+		return -ENOMEM;
+
+	ring->skbs_dma = devm_kcalloc(dev, size, sizeof(dma_addr_t),
+				      GFP_KERNEL);
+	if (!ring->skbs_dma)
+		return -ENOMEM;
+
+	ring->size = size;
+
+	return 0;
+}
+
+static void owl_emac_dma_cmd_resume_rx(struct owl_emac_priv *priv)
+{
+	owl_emac_reg_write(priv, OWL_EMAC_REG_MAC_CSR2,
+			   OWL_EMAC_VAL_MAC_CSR2_RPD);
+}
+
+static void owl_emac_dma_cmd_resume_tx(struct owl_emac_priv *priv)
+{
+	owl_emac_reg_write(priv, OWL_EMAC_REG_MAC_CSR1,
+			   OWL_EMAC_VAL_MAC_CSR1_TPD);
+}
+
+static u32 owl_emac_dma_cmd_set_tx(struct owl_emac_priv *priv, u32 status)
+{
+	return owl_emac_reg_update(priv, OWL_EMAC_REG_MAC_CSR6,
+				   OWL_EMAC_BIT_MAC_CSR6_ST, status);
+}
+
+static u32 owl_emac_dma_cmd_start_tx(struct owl_emac_priv *priv)
+{
+	return owl_emac_dma_cmd_set_tx(priv, ~0);
+}
+
+static u32 owl_emac_dma_cmd_set(struct owl_emac_priv *priv, u32 status)
+{
+	return owl_emac_reg_update(priv, OWL_EMAC_REG_MAC_CSR6,
+				   OWL_EMAC_MSK_MAC_CSR6_STSR, status);
+}
+
+static u32 owl_emac_dma_cmd_start(struct owl_emac_priv *priv)
+{
+	return owl_emac_dma_cmd_set(priv, ~0);
+}
+
+static u32 owl_emac_dma_cmd_stop(struct owl_emac_priv *priv)
+{
+	return owl_emac_dma_cmd_set(priv, 0);
+}
+
+static void owl_emac_set_hw_mac_addr(struct net_device *netdev)
+{
+	struct owl_emac_priv *priv = netdev_priv(netdev);
+	const u8 *mac_addr = netdev->dev_addr;
+	u32 addr_high, addr_low;
+
+	addr_high = mac_addr[0] << 8 | mac_addr[1];
+	addr_low = mac_addr[2] << 24 | mac_addr[3] << 16 |
+		   mac_addr[4] << 8 | mac_addr[5];
+
+	owl_emac_reg_write(priv, OWL_EMAC_REG_MAC_CSR17, addr_high);
+	owl_emac_reg_write(priv, OWL_EMAC_REG_MAC_CSR16, addr_low);
+}
+
+static void owl_emac_update_link_state(struct owl_emac_priv *priv)
+{
+	u32 val, status;
+
+	if (priv->pause) {
+		val = OWL_EMAC_BIT_MAC_CSR20_FCE | OWL_EMAC_BIT_MAC_CSR20_TUE;
+		val |= OWL_EMAC_BIT_MAC_CSR20_TPE | OWL_EMAC_BIT_MAC_CSR20_RPE;
+		val |= OWL_EMAC_BIT_MAC_CSR20_BPE;
+	} else {
+		val = 0;
+	}
+
+	/* Update flow control. */
+	owl_emac_reg_write(priv, OWL_EMAC_REG_MAC_CSR20, val);
+
+	val = (priv->speed == SPEED_100) ? OWL_EMAC_VAL_MAC_CSR6_SPEED_100M :
+					   OWL_EMAC_VAL_MAC_CSR6_SPEED_10M;
+	val <<= OWL_EMAC_OFF_MAC_CSR6_SPEED;
+
+	if (priv->duplex == DUPLEX_FULL)
+		val |= OWL_EMAC_BIT_MAC_CSR6_FD;
+
+	spin_lock_bh(&priv->lock);
+
+	/* Temporarily stop DMA TX & RX. */
+	status = owl_emac_dma_cmd_stop(priv);
+
+	/* Update operation modes. */
+	owl_emac_reg_update(priv, OWL_EMAC_REG_MAC_CSR6,
+			    OWL_EMAC_MSK_MAC_CSR6_SPEED |
+			    OWL_EMAC_BIT_MAC_CSR6_FD, val);
+
+	/* Restore DMA TX & RX status. */
+	owl_emac_dma_cmd_set(priv, status);
+
+	spin_unlock_bh(&priv->lock);
+}
+
+static void owl_emac_adjust_link(struct net_device *netdev)
+{
+	struct owl_emac_priv *priv = netdev_priv(netdev);
+	struct phy_device *phydev = netdev->phydev;
+	bool state_changed = false;
+
+	if (phydev->link) {
+		if (!priv->link) {
+			priv->link = phydev->link;
+			state_changed = true;
+		}
+
+		if (priv->speed != phydev->speed) {
+			priv->speed = phydev->speed;
+			state_changed = true;
+		}
+
+		if (priv->duplex != phydev->duplex) {
+			priv->duplex = phydev->duplex;
+			state_changed = true;
+		}
+
+		if (priv->pause != phydev->pause) {
+			priv->pause = phydev->pause;
+			state_changed = true;
+		}
+	} else {
+		if (priv->link) {
+			priv->link = phydev->link;
+			state_changed = true;
+		}
+	}
+
+	if (state_changed) {
+		if (phydev->link)
+			owl_emac_update_link_state(priv);
+
+		if (netif_msg_link(priv))
+			phy_print_status(phydev);
+	}
+}
+
+static irqreturn_t owl_emac_handle_irq(int irq, void *data)
+{
+	struct net_device *netdev = data;
+	struct owl_emac_priv *priv = netdev_priv(netdev);
+
+	if (netif_running(netdev)) {
+		owl_emac_irq_disable(priv);
+		napi_schedule(&priv->napi);
+	}
+
+	return IRQ_HANDLED;
+}
+
+static void owl_emac_ether_addr_push(u8 **dst, const u8 *src)
+{
+	u32 *a = (u32 *)(*dst);
+	const u16 *b = (const u16 *)src;
+
+	a[0] = b[0];
+	a[1] = b[1];
+	a[2] = b[2];
+
+	*dst += 12;
+}
+
+static void
+owl_emac_setup_frame_prepare(struct owl_emac_priv *priv, struct sk_buff *skb)
+{
+	const u8 bcast_addr[] = { 0xFF, 0xFF, 0xFF, 0xFF, 0xFF, 0xFF };
+	const u8 *mac_addr = priv->netdev->dev_addr;
+	u8 *frame;
+	int i;
+
+	skb_put(skb, OWL_EMAC_SETUP_FRAME_LEN);
+
+	frame = skb->data;
+	memset(frame, 0, skb->len);
+
+	owl_emac_ether_addr_push(&frame, mac_addr);
+	owl_emac_ether_addr_push(&frame, bcast_addr);
+
+	/* Fill multicast addresses. */
+	WARN_ON(priv->mcaddr_list.count >= OWL_EMAC_MAX_MULTICAST_ADDRS);
+	for (i = 0; i < priv->mcaddr_list.count; i++) {
+		mac_addr = priv->mcaddr_list.addrs[i];
+		owl_emac_ether_addr_push(&frame, mac_addr);
+	}
+}
+
+/* The setup frame is a special descriptor which is used to provide physical
+ * addresses (i.e. mac, broadcast and multicast) to the MAC hardware for
+ * filtering purposes. To be recognized as a setup frame, the TDES1_SET bit
+ * must be set in the TX descriptor control field.
+ */
+static int owl_emac_setup_frame_xmit(struct owl_emac_priv *priv)
+{
+	struct owl_emac_ring *ring = &priv->tx_ring;
+	struct net_device *netdev = priv->netdev;
+	struct owl_emac_ring_desc *desc;
+	struct sk_buff *skb;
+	unsigned int tx_head;
+	u32 status, control;
+	dma_addr_t dma_addr;
+	int ret;
+
+	skb = owl_emac_alloc_skb(netdev);
+	if (!skb)
+		return -ENOMEM;
+
+	owl_emac_setup_frame_prepare(priv, skb);
+
+	dma_addr = owl_emac_dma_map_tx(priv, skb);
+	if (dma_mapping_error(owl_emac_get_dev(priv), dma_addr)) {
+		ret = -ENOMEM;
+		goto err_free_skb;
+	}
+
+	spin_lock_bh(&priv->lock);
+
+	tx_head = ring->head;
+	desc = &ring->descs[tx_head];
+
+	status = READ_ONCE(desc->status);
+	control = READ_ONCE(desc->control);
+	dma_rmb(); /* Ensure data has been read before used. */
+
+	if (unlikely(status & OWL_EMAC_BIT_TDES0_OWN) ||
+	    !owl_emac_ring_num_unused(ring)) {
+		spin_unlock_bh(&priv->lock);
+		owl_emac_dma_unmap_tx(priv, skb, dma_addr);
+		ret = -EBUSY;
+		goto err_free_skb;
+	}
+
+	ring->skbs[tx_head] = skb;
+	ring->skbs_dma[tx_head] = dma_addr;
+
+	control &= OWL_EMAC_BIT_TDES1_IC | OWL_EMAC_BIT_TDES1_TER; /* Maintain bits */
+	control |= OWL_EMAC_BIT_TDES1_SET;
+	control |= OWL_EMAC_MSK_TDES1_TBS1 & skb->len;
+
+	WRITE_ONCE(desc->control, control);
+	WRITE_ONCE(desc->buf_addr, dma_addr);
+	dma_wmb(); /* Flush descriptor before changing ownership. */
+	WRITE_ONCE(desc->status, OWL_EMAC_BIT_TDES0_OWN);
+
+	owl_emac_ring_push_head(ring);
+
+	/* Temporarily enable DMA TX. */
+	status = owl_emac_dma_cmd_start_tx(priv);
+
+	/* Trigger setup frame processing. */
+	owl_emac_dma_cmd_resume_tx(priv);
+
+	/* Restore DMA TX status. */
+	owl_emac_dma_cmd_set_tx(priv, status);
+
+	/* Stop regular TX until setup frame is processed. */
+	netif_stop_queue(netdev);
+
+	spin_unlock_bh(&priv->lock);
+
+	return 0;
+
+err_free_skb:
+	dev_kfree_skb(skb);
+	return ret;
+}
+
+static netdev_tx_t owl_emac_ndo_start_xmit(struct sk_buff *skb,
+					   struct net_device *netdev)
+{
+	struct owl_emac_priv *priv = netdev_priv(netdev);
+	struct device *dev = owl_emac_get_dev(priv);
+	struct owl_emac_ring *ring = &priv->tx_ring;
+	struct owl_emac_ring_desc *desc;
+	unsigned int tx_head;
+	u32 status, control;
+	dma_addr_t dma_addr;
+
+	dma_addr = owl_emac_dma_map_tx(priv, skb);
+	if (dma_mapping_error(dev, dma_addr)) {
+		dev_err_ratelimited(&netdev->dev, "TX DMA mapping failed\n");
+		dev_kfree_skb(skb);
+		netdev->stats.tx_dropped++;
+		return NETDEV_TX_OK;
+	}
+
+	spin_lock_bh(&priv->lock);
+
+	tx_head = ring->head;
+	desc = &ring->descs[tx_head];
+
+	status = READ_ONCE(desc->status);
+	control = READ_ONCE(desc->control);
+	dma_rmb(); /* Ensure data has been read before used. */
+
+	if (!owl_emac_ring_num_unused(ring) ||
+	    unlikely(status & OWL_EMAC_BIT_TDES0_OWN)) {
+		netif_stop_queue(netdev);
+		spin_unlock_bh(&priv->lock);
+
+		dev_dbg_ratelimited(&netdev->dev, "TX buffer full, status=0x%08x\n",
+				    owl_emac_irq_status(priv));
+		owl_emac_dma_unmap_tx(priv, skb, dma_addr);
+		netdev->stats.tx_dropped++;
+		return NETDEV_TX_BUSY;
+	}
+
+	ring->skbs[tx_head] = skb;
+	ring->skbs_dma[tx_head] = dma_addr;
+
+	control &= OWL_EMAC_BIT_TDES1_IC | OWL_EMAC_BIT_TDES1_TER; /* Maintain bits */
+	control |= OWL_EMAC_BIT_TDES1_FS | OWL_EMAC_BIT_TDES1_LS;
+	control |= OWL_EMAC_MSK_TDES1_TBS1 & skb->len;
+
+	WRITE_ONCE(desc->control, control);
+	WRITE_ONCE(desc->buf_addr, dma_addr);
+	dma_wmb(); /* Flush descriptor before changing ownership. */
+	WRITE_ONCE(desc->status, OWL_EMAC_BIT_TDES0_OWN);
+
+	owl_emac_dma_cmd_resume_tx(priv);
+	owl_emac_ring_push_head(ring);
+
+	/* FIXME: The transmission is currently restricted to a single frame
+	 * at a time as a workaround for a MAC hardware bug that causes random
+	 * freeze of the TX queue processor.
+	 */
+	netif_stop_queue(netdev);
+
+	spin_unlock_bh(&priv->lock);
+
+	return NETDEV_TX_OK;
+}
+
+static bool owl_emac_tx_complete_tail(struct owl_emac_priv *priv)
+{
+	struct owl_emac_ring *ring = &priv->tx_ring;
+	struct net_device *netdev = priv->netdev;
+	struct owl_emac_ring_desc *desc;
+	struct sk_buff *skb;
+	unsigned int tx_tail;
+	u32 status;
+
+	tx_tail = ring->tail;
+	desc = &ring->descs[tx_tail];
+
+	status = READ_ONCE(desc->status);
+	dma_rmb(); /* Ensure data has been read before used. */
+
+	if (status & OWL_EMAC_BIT_TDES0_OWN)
+		return false;
+
+	/* Check for errors. */
+	if (status & OWL_EMAC_BIT_TDES0_ES) {
+		dev_dbg_ratelimited(&netdev->dev,
+				    "TX complete error status: 0x%08x\n",
+				    status);
+
+		netdev->stats.tx_errors++;
+
+		if (status & OWL_EMAC_BIT_TDES0_UF)
+			netdev->stats.tx_fifo_errors++;
+
+		if (status & OWL_EMAC_BIT_TDES0_EC)
+			netdev->stats.tx_aborted_errors++;
+
+		if (status & OWL_EMAC_BIT_TDES0_LC)
+			netdev->stats.tx_window_errors++;
+
+		if (status & OWL_EMAC_BIT_TDES0_NC)
+			netdev->stats.tx_heartbeat_errors++;
+
+		if (status & OWL_EMAC_BIT_TDES0_LO)
+			netdev->stats.tx_carrier_errors++;
+	} else {
+		netdev->stats.tx_packets++;
+		netdev->stats.tx_bytes += ring->skbs[tx_tail]->len;
+	}
+
+	/* Some collisions occurred, but pkt has been transmitted. */
+	if (status & OWL_EMAC_BIT_TDES0_DE)
+		netdev->stats.collisions++;
+
+	skb = ring->skbs[tx_tail];
+	owl_emac_dma_unmap_tx(priv, skb, ring->skbs_dma[tx_tail]);
+	dev_kfree_skb(skb);
+
+	ring->skbs[tx_tail] = NULL;
+	ring->skbs_dma[tx_tail] = 0;
+
+	owl_emac_ring_pop_tail(ring);
+
+	if (unlikely(netif_queue_stopped(netdev)))
+		netif_wake_queue(netdev);
+
+	return true;
+}
+
+static void owl_emac_tx_complete(struct owl_emac_priv *priv)
+{
+	struct owl_emac_ring *ring = &priv->tx_ring;
+	struct net_device *netdev = priv->netdev;
+	unsigned int tx_next;
+	u32 status;
+
+	spin_lock(&priv->lock);
+
+	while (ring->tail != ring->head) {
+		if (!owl_emac_tx_complete_tail(priv))
+			break;
+	}
+
+	/* FIXME: This is a workaround for a MAC hardware bug not clearing
+	 * (sometimes) the OWN bit for a transmitted frame descriptor.
+	 *
+	 * At this point, when TX queue is full, the tail descriptor has the
+	 * OWN bit set, which normally means the frame has not been processed
+	 * or transmitted yet. But if there is at least one descriptor in the
+	 * queue having the OWN bit cleared, we can safely assume the tail
+	 * frame has been also processed by the MAC hardware.
+	 *
+	 * If that's the case, let's force the frame completion by manually
+	 * clearing the OWN bit.
+	 */
+	if (unlikely(!owl_emac_ring_num_unused(ring))) {
+		tx_next = ring->tail;
+
+		while ((tx_next = owl_emac_ring_get_next(ring, tx_next)) != ring->head) {
+			status = READ_ONCE(ring->descs[tx_next].status);
+			dma_rmb(); /* Ensure data has been read before used. */
+
+			if (status & OWL_EMAC_BIT_TDES0_OWN)
+				continue;
+
+			netdev_dbg(netdev, "Found uncleared TX desc OWN bit\n");
+
+			status = READ_ONCE(ring->descs[ring->tail].status);
+			dma_rmb(); /* Ensure data has been read before used. */
+			status &= ~OWL_EMAC_BIT_TDES0_OWN;
+			WRITE_ONCE(ring->descs[ring->tail].status, status);
+
+			owl_emac_tx_complete_tail(priv);
+			break;
+		}
+	}
+
+	spin_unlock(&priv->lock);
+}
+
+static int owl_emac_rx_process(struct owl_emac_priv *priv, int budget)
+{
+	struct owl_emac_ring *ring = &priv->rx_ring;
+	struct device *dev = owl_emac_get_dev(priv);
+	struct net_device *netdev = priv->netdev;
+	struct owl_emac_ring_desc *desc;
+	struct sk_buff *curr_skb, *new_skb;
+	dma_addr_t curr_dma, new_dma;
+	unsigned int rx_tail, len;
+	u32 status;
+	int recv = 0;
+
+	while (recv < budget) {
+		spin_lock(&priv->lock);
+
+		rx_tail = ring->tail;
+		desc = &ring->descs[rx_tail];
+
+		status = READ_ONCE(desc->status);
+		dma_rmb(); /* Ensure data has been read before used. */
+
+		if (status & OWL_EMAC_BIT_RDES0_OWN) {
+			spin_unlock(&priv->lock);
+			break;
+		}
+
+		curr_skb = ring->skbs[rx_tail];
+		curr_dma = ring->skbs_dma[rx_tail];
+		owl_emac_ring_pop_tail(ring);
+
+		spin_unlock(&priv->lock);
+
+		if (status & (OWL_EMAC_BIT_RDES0_DE | OWL_EMAC_BIT_RDES0_RF |
+		    OWL_EMAC_BIT_RDES0_TL | OWL_EMAC_BIT_RDES0_CS |
+		    OWL_EMAC_BIT_RDES0_DB | OWL_EMAC_BIT_RDES0_CE |
+		    OWL_EMAC_BIT_RDES0_ZERO)) {
+			dev_dbg_ratelimited(&netdev->dev,
+					    "RX desc error status: 0x%08x\n",
+					    status);
+
+			if (status & OWL_EMAC_BIT_RDES0_DE)
+				netdev->stats.rx_over_errors++;
+
+			if (status & (OWL_EMAC_BIT_RDES0_RF | OWL_EMAC_BIT_RDES0_DB))
+				netdev->stats.rx_frame_errors++;
+
+			if (status & OWL_EMAC_BIT_RDES0_TL)
+				netdev->stats.rx_length_errors++;
+
+			if (status & OWL_EMAC_BIT_RDES0_CS)
+				netdev->stats.collisions++;
+
+			if (status & OWL_EMAC_BIT_RDES0_CE)
+				netdev->stats.rx_crc_errors++;
+
+			if (status & OWL_EMAC_BIT_RDES0_ZERO)
+				netdev->stats.rx_fifo_errors++;
+
+			goto drop_skb;
+		}
+
+		len = (status & OWL_EMAC_MSK_RDES0_FL) >> OWL_EMAC_OFF_RDES0_FL;
+		if (unlikely(len > OWL_EMAC_RX_FRAME_MAX_LEN)) {
+			netdev->stats.rx_length_errors++;
+			netdev_err(netdev, "invalid RX frame len: %u\n", len);
+			goto drop_skb;
+		}
+
+		/* Prepare new skb before receiving the current one. */
+		new_skb = owl_emac_alloc_skb(netdev);
+		if (unlikely(!new_skb))
+			goto drop_skb;
+
+		new_dma = owl_emac_dma_map_rx(priv, new_skb);
+		if (dma_mapping_error(dev, new_dma)) {
+			dev_kfree_skb(new_skb);
+			netdev_err(netdev, "RX DMA mapping failed\n");
+			goto drop_skb;
+		}
+
+		owl_emac_dma_unmap_rx(priv, curr_skb, curr_dma);
+
+		skb_put(curr_skb, len - ETH_FCS_LEN);
+		curr_skb->ip_summed = CHECKSUM_NONE;
+		curr_skb->protocol = eth_type_trans(curr_skb, netdev);
+		curr_skb->dev = netdev;
+
+		netif_receive_skb(curr_skb);
+
+		netdev->stats.rx_packets++;
+		netdev->stats.rx_bytes += len;
+		recv++;
+		goto push_skb;
+
+drop_skb:
+		netdev->stats.rx_dropped++;
+		netdev->stats.rx_errors++;
+		/* Reuse the current skb. */
+		new_skb = curr_skb;
+		new_dma = curr_dma;
+
+push_skb:
+		spin_lock(&priv->lock);
+
+		ring->skbs[ring->head] = new_skb;
+		ring->skbs_dma[ring->head] = new_dma;
+
+		WRITE_ONCE(desc->buf_addr, new_dma);
+		dma_wmb(); /* Flush descriptor before changing ownership. */
+		WRITE_ONCE(desc->status, OWL_EMAC_BIT_RDES0_OWN);
+
+		owl_emac_ring_push_head(ring);
+
+		spin_unlock(&priv->lock);
+	}
+
+	return recv;
+}
+
+static int owl_emac_poll(struct napi_struct *napi, int budget)
+{
+	int work_done = 0, ru_cnt = 0, recv;
+	static int tx_err_cnt, rx_err_cnt;
+	struct owl_emac_priv *priv;
+	u32 status, proc_status;
+
+	priv = container_of(napi, struct owl_emac_priv, napi);
+
+	while ((status = owl_emac_irq_clear(priv)) &
+	       (OWL_EMAC_BIT_MAC_CSR5_NIS | OWL_EMAC_BIT_MAC_CSR5_AIS)) {
+		recv = 0;
+
+		/* TX setup frame raises ETI instead of TI. */
+		if (status & (OWL_EMAC_BIT_MAC_CSR5_TI | OWL_EMAC_BIT_MAC_CSR5_ETI)) {
+			owl_emac_tx_complete(priv);
+			tx_err_cnt = 0;
+
+			/* Count MAC internal RX errors. */
+			proc_status = status & OWL_EMAC_MSK_MAC_CSR5_RS;
+			proc_status >>= OWL_EMAC_OFF_MAC_CSR5_RS;
+			if (proc_status == OWL_EMAC_VAL_MAC_CSR5_RS_DATA ||
+			    proc_status == OWL_EMAC_VAL_MAC_CSR5_RS_CDES ||
+			    proc_status == OWL_EMAC_VAL_MAC_CSR5_RS_FDES)
+				rx_err_cnt++;
+		}
+
+		if (status & OWL_EMAC_BIT_MAC_CSR5_RI) {
+			recv = owl_emac_rx_process(priv, budget - work_done);
+			rx_err_cnt = 0;
+
+			/* Count MAC internal TX errors. */
+			proc_status = status & OWL_EMAC_MSK_MAC_CSR5_TS;
+			proc_status >>= OWL_EMAC_OFF_MAC_CSR5_TS;
+			if (proc_status == OWL_EMAC_VAL_MAC_CSR5_TS_DATA ||
+			    proc_status == OWL_EMAC_VAL_MAC_CSR5_TS_CDES)
+				tx_err_cnt++;
+		} else if (status & OWL_EMAC_BIT_MAC_CSR5_RU) {
+			/* MAC AHB is in suspended state, will return to RX
+			 * descriptor processing when the host changes ownership
+			 * of the descriptor and either an RX poll demand CMD is
+			 * issued or a new frame is recognized by the MAC AHB.
+			 */
+			if (++ru_cnt == 2)
+				owl_emac_dma_cmd_resume_rx(priv);
+
+			recv = owl_emac_rx_process(priv, budget - work_done);
+
+			/* Guard against too many RU interrupts. */
+			if (ru_cnt > 3)
+				break;
+		}
+
+		work_done += recv;
+		if (work_done >= budget)
+			break;
+	}
+
+	if (work_done < budget) {
+		napi_complete_done(napi, work_done);
+		owl_emac_irq_enable(priv);
+	}
+
+	/* Reset MAC when getting too many internal TX or RX errors. */
+	if (tx_err_cnt > 10 || rx_err_cnt > 10) {
+		netdev_dbg(priv->netdev, "%s error status: 0x%08x\n",
+			   tx_err_cnt > 10 ? "TX" : "RX", status);
+		rx_err_cnt = 0;
+		tx_err_cnt = 0;
+		schedule_work(&priv->mac_reset_task);
+	}
+
+	return work_done;
+}
+
+static void owl_emac_mdio_clock_enable(struct owl_emac_priv *priv)
+{
+	u32 val;
+
+	/* Enable MDC clock generation by adjusting CLKDIV according to
+	 * the vendor implementation of the original driver.
+	 */
+	val = owl_emac_reg_read(priv, OWL_EMAC_REG_MAC_CSR10);
+	val &= OWL_EMAC_MSK_MAC_CSR10_CLKDIV;
+	val |= OWL_EMAC_VAL_MAC_CSR10_CLKDIV_128 << OWL_EMAC_OFF_MAC_CSR10_CLKDIV;
+
+	val |= OWL_EMAC_BIT_MAC_CSR10_SB;
+	val |= OWL_EMAC_VAL_MAC_CSR10_OPCODE_CDS << OWL_EMAC_OFF_MAC_CSR10_OPCODE;
+	owl_emac_reg_write(priv, OWL_EMAC_REG_MAC_CSR10, val);
+}
+
+static void owl_emac_core_hw_reset(struct owl_emac_priv *priv)
+{
+	/* Trigger hardware reset. */
+	reset_control_assert(priv->reset);
+	usleep_range(10, 20);
+	reset_control_deassert(priv->reset);
+	usleep_range(100, 200);
+}
+
+static int owl_emac_core_sw_reset(struct owl_emac_priv *priv)
+{
+	u32 val;
+	int ret;
+
+	/* Trigger software reset. */
+	owl_emac_reg_set(priv, OWL_EMAC_REG_MAC_CSR0, OWL_EMAC_BIT_MAC_CSR0_SWR);
+	ret = readl_poll_timeout(priv->base + OWL_EMAC_REG_MAC_CSR0,
+				 val, !(val & OWL_EMAC_BIT_MAC_CSR0_SWR),
+				 OWL_EMAC_POLL_DELAY_USEC,
+				 OWL_EMAC_RESET_POLL_TIMEOUT_USEC);
+	if (ret)
+		return ret;
+
+	if (priv->phy_mode == PHY_INTERFACE_MODE_RMII) {
+		/* Enable RMII and use the 50MHz rmii clk as output to PHY. */
+		val = 0;
+	} else {
+		/* Enable SMII and use the 125MHz rmii clk as output to PHY.
+		 * Additionally set SMII SYNC delay to 4 half cycle.
+		 */
+		val = 0x04 << OWL_EMAC_OFF_MAC_CTRL_SSDC;
+		val |= OWL_EMAC_BIT_MAC_CTRL_RSIS;
+	}
+	owl_emac_reg_write(priv, OWL_EMAC_REG_MAC_CTRL, val);
+
+	/* MDC is disabled after reset. */
+	owl_emac_mdio_clock_enable(priv);
+
+	/* Set FIFO pause & restart threshold levels. */
+	val = 0x40 << OWL_EMAC_OFF_MAC_CSR19_FPTL;
+	val |= 0x10 << OWL_EMAC_OFF_MAC_CSR19_FRTL;
+	owl_emac_reg_write(priv, OWL_EMAC_REG_MAC_CSR19, val);
+
+	/* Set flow control pause quanta time to ~100 ms. */
+	val = 0x4FFF << OWL_EMAC_OFF_MAC_CSR18_PQT;
+	owl_emac_reg_write(priv, OWL_EMAC_REG_MAC_CSR18, val);
+
+	/* Setup interrupt mitigation. */
+	val = 7 << OWL_EMAC_OFF_MAC_CSR11_NRP;
+	val |= 4 << OWL_EMAC_OFF_MAC_CSR11_RT;
+	owl_emac_reg_write(priv, OWL_EMAC_REG_MAC_CSR11, val);
+
+	/* Set RX/TX rings base addresses. */
+	owl_emac_reg_write(priv, OWL_EMAC_REG_MAC_CSR3,
+			   (u32)(priv->rx_ring.descs_dma));
+	owl_emac_reg_write(priv, OWL_EMAC_REG_MAC_CSR4,
+			   (u32)(priv->tx_ring.descs_dma));
+
+	/* Setup initial operation mode. */
+	val = OWL_EMAC_VAL_MAC_CSR6_SPEED_100M << OWL_EMAC_OFF_MAC_CSR6_SPEED;
+	val |= OWL_EMAC_BIT_MAC_CSR6_FD;
+	owl_emac_reg_update(priv, OWL_EMAC_REG_MAC_CSR6,
+			    OWL_EMAC_MSK_MAC_CSR6_SPEED |
+			    OWL_EMAC_BIT_MAC_CSR6_FD, val);
+	owl_emac_reg_clear(priv, OWL_EMAC_REG_MAC_CSR6,
+			   OWL_EMAC_BIT_MAC_CSR6_PR | OWL_EMAC_BIT_MAC_CSR6_PM);
+
+	priv->link = 0;
+	priv->speed = SPEED_UNKNOWN;
+	priv->duplex = DUPLEX_UNKNOWN;
+	priv->pause = 0;
+	priv->mcaddr_list.count = 0;
+
+	return 0;
+}
+
+static int owl_emac_enable(struct net_device *netdev, bool start_phy)
+{
+	struct owl_emac_priv *priv = netdev_priv(netdev);
+	int ret;
+
+	owl_emac_dma_cmd_stop(priv);
+	owl_emac_irq_disable(priv);
+	owl_emac_irq_clear(priv);
+
+	owl_emac_ring_prepare_tx(priv);
+	ret = owl_emac_ring_prepare_rx(priv);
+	if (ret)
+		goto err_unprep;
+
+	ret = owl_emac_core_sw_reset(priv);
+	if (ret) {
+		netdev_err(netdev, "failed to soft reset MAC core: %d\n", ret);
+		goto err_unprep;
+	}
+
+	owl_emac_set_hw_mac_addr(netdev);
+	owl_emac_setup_frame_xmit(priv);
+
+	netdev_reset_queue(netdev);
+	napi_enable(&priv->napi);
+
+	owl_emac_irq_enable(priv);
+	owl_emac_dma_cmd_start(priv);
+
+	if (start_phy)
+		phy_start(netdev->phydev);
+
+	netif_start_queue(netdev);
+
+	return 0;
+
+err_unprep:
+	owl_emac_ring_unprepare_rx(priv);
+	owl_emac_ring_unprepare_tx(priv);
+
+	return ret;
+}
+
+static void owl_emac_disable(struct net_device *netdev, bool stop_phy)
+{
+	struct owl_emac_priv *priv = netdev_priv(netdev);
+
+	owl_emac_dma_cmd_stop(priv);
+	owl_emac_irq_disable(priv);
+
+	netif_stop_queue(netdev);
+	napi_disable(&priv->napi);
+
+	if (stop_phy)
+		phy_stop(netdev->phydev);
+
+	owl_emac_ring_unprepare_rx(priv);
+	owl_emac_ring_unprepare_tx(priv);
+}
+
+static int owl_emac_ndo_open(struct net_device *netdev)
+{
+	return owl_emac_enable(netdev, true);
+}
+
+static int owl_emac_ndo_stop(struct net_device *netdev)
+{
+	owl_emac_disable(netdev, true);
+
+	return 0;
+}
+
+static void owl_emac_set_multicast(struct net_device *netdev, int count)
+{
+	struct owl_emac_priv *priv = netdev_priv(netdev);
+	struct netdev_hw_addr *ha;
+	int index = 0;
+
+	if (count <= 0) {
+		priv->mcaddr_list.count = 0;
+		return;
+	}
+
+	netdev_for_each_mc_addr(ha, netdev) {
+		if (!is_multicast_ether_addr(ha->addr))
+			continue;
+
+		WARN_ON(index >= OWL_EMAC_MAX_MULTICAST_ADDRS);
+		ether_addr_copy(priv->mcaddr_list.addrs[index++], ha->addr);
+	}
+
+	priv->mcaddr_list.count = index;
+
+	owl_emac_setup_frame_xmit(priv);
+}
+
+static void owl_emac_ndo_set_rx_mode(struct net_device *netdev)
+{
+	struct owl_emac_priv *priv = netdev_priv(netdev);
+	u32 status, val = 0;
+	int mcast_count = 0;
+
+	if (netdev->flags & IFF_PROMISC) {
+		val = OWL_EMAC_BIT_MAC_CSR6_PR;
+	} else if (netdev->flags & IFF_ALLMULTI) {
+		val = OWL_EMAC_BIT_MAC_CSR6_PM;
+	} else if (netdev->flags & IFF_MULTICAST) {
+		mcast_count = netdev_mc_count(netdev);
+
+		if (mcast_count > OWL_EMAC_MAX_MULTICAST_ADDRS) {
+			val = OWL_EMAC_BIT_MAC_CSR6_PM;
+			mcast_count = 0;
+		}
+	}
+
+	spin_lock_bh(&priv->lock);
+
+	/* Temporarily stop DMA TX & RX. */
+	status = owl_emac_dma_cmd_stop(priv);
+
+	/* Update operation modes. */
+	owl_emac_reg_update(priv, OWL_EMAC_REG_MAC_CSR6,
+			    OWL_EMAC_BIT_MAC_CSR6_PR | OWL_EMAC_BIT_MAC_CSR6_PM,
+			    val);
+
+	/* Restore DMA TX & RX status. */
+	owl_emac_dma_cmd_set(priv, status);
+
+	spin_unlock_bh(&priv->lock);
+
+	/* Set/reset multicast addr list. */
+	owl_emac_set_multicast(netdev, mcast_count);
+}
+
+static int owl_emac_ndo_set_mac_addr(struct net_device *netdev, void *addr)
+{
+	struct sockaddr *skaddr = addr;
+
+	if (!is_valid_ether_addr(skaddr->sa_data))
+		return -EADDRNOTAVAIL;
+
+	if (netif_running(netdev))
+		return -EBUSY;
+
+	eth_hw_addr_set(netdev, skaddr->sa_data);
+	owl_emac_set_hw_mac_addr(netdev);
+
+	return owl_emac_setup_frame_xmit(netdev_priv(netdev));
+}
+
+static int owl_emac_ndo_eth_ioctl(struct net_device *netdev,
+				  struct ifreq *req, int cmd)
+{
+	if (!netif_running(netdev))
+		return -EINVAL;
+
+	return phy_mii_ioctl(netdev->phydev, req, cmd);
+}
+
+static void owl_emac_ndo_tx_timeout(struct net_device *netdev,
+				    unsigned int txqueue)
+{
+	struct owl_emac_priv *priv = netdev_priv(netdev);
+
+	schedule_work(&priv->mac_reset_task);
+}
+
+static void owl_emac_reset_task(struct work_struct *work)
+{
+	struct owl_emac_priv *priv;
+
+	priv = container_of(work, struct owl_emac_priv, mac_reset_task);
+
+	netdev_dbg(priv->netdev, "resetting MAC\n");
+	owl_emac_disable(priv->netdev, false);
+	owl_emac_enable(priv->netdev, false);
+}
+
+static struct net_device_stats *
+owl_emac_ndo_get_stats(struct net_device *netdev)
+{
+	/* FIXME: If possible, try to get stats from MAC hardware registers
+	 * instead of tracking them manually in the driver.
+	 */
+
+	return &netdev->stats;
+}
+
+static const struct net_device_ops owl_emac_netdev_ops = {
+	.ndo_open		= owl_emac_ndo_open,
+	.ndo_stop		= owl_emac_ndo_stop,
+	.ndo_start_xmit		= owl_emac_ndo_start_xmit,
+	.ndo_set_rx_mode	= owl_emac_ndo_set_rx_mode,
+	.ndo_set_mac_address	= owl_emac_ndo_set_mac_addr,
+	.ndo_validate_addr	= eth_validate_addr,
+	.ndo_eth_ioctl		= owl_emac_ndo_eth_ioctl,
+	.ndo_tx_timeout         = owl_emac_ndo_tx_timeout,
+	.ndo_get_stats		= owl_emac_ndo_get_stats,
+};
+
+static void owl_emac_ethtool_get_drvinfo(struct net_device *dev,
+					 struct ethtool_drvinfo *info)
+{
+	strscpy(info->driver, OWL_EMAC_DRVNAME, sizeof(info->driver));
+}
+
+static u32 owl_emac_ethtool_get_msglevel(struct net_device *netdev)
+{
+	struct owl_emac_priv *priv = netdev_priv(netdev);
+
+	return priv->msg_enable;
+}
+
+static void owl_emac_ethtool_set_msglevel(struct net_device *ndev, u32 val)
+{
+	struct owl_emac_priv *priv = netdev_priv(ndev);
+
+	priv->msg_enable = val;
+}
+
+static const struct ethtool_ops owl_emac_ethtool_ops = {
+	.get_drvinfo		= owl_emac_ethtool_get_drvinfo,
+	.get_link		= ethtool_op_get_link,
+	.get_link_ksettings	= phy_ethtool_get_link_ksettings,
+	.set_link_ksettings	= phy_ethtool_set_link_ksettings,
+	.get_msglevel		= owl_emac_ethtool_get_msglevel,
+	.set_msglevel		= owl_emac_ethtool_set_msglevel,
+};
+
+static int owl_emac_mdio_wait(struct owl_emac_priv *priv)
+{
+	u32 val;
+
+	/* Wait while data transfer is in progress. */
+	return readl_poll_timeout(priv->base + OWL_EMAC_REG_MAC_CSR10,
+				  val, !(val & OWL_EMAC_BIT_MAC_CSR10_SB),
+				  OWL_EMAC_POLL_DELAY_USEC,
+				  OWL_EMAC_MDIO_POLL_TIMEOUT_USEC);
+}
+
+static int owl_emac_mdio_read(struct mii_bus *bus, int addr, int regnum)
+{
+	struct owl_emac_priv *priv = bus->priv;
+	u32 data, tmp;
+	int ret;
+
+	data = OWL_EMAC_BIT_MAC_CSR10_SB;
+	data |= OWL_EMAC_VAL_MAC_CSR10_OPCODE_RD << OWL_EMAC_OFF_MAC_CSR10_OPCODE;
+
+	tmp = addr << OWL_EMAC_OFF_MAC_CSR10_PHYADD;
+	data |= tmp & OWL_EMAC_MSK_MAC_CSR10_PHYADD;
+
+	tmp = regnum << OWL_EMAC_OFF_MAC_CSR10_REGADD;
+	data |= tmp & OWL_EMAC_MSK_MAC_CSR10_REGADD;
+
+	owl_emac_reg_write(priv, OWL_EMAC_REG_MAC_CSR10, data);
+
+	ret = owl_emac_mdio_wait(priv);
+	if (ret)
+		return ret;
+
+	data = owl_emac_reg_read(priv, OWL_EMAC_REG_MAC_CSR10);
+	data &= OWL_EMAC_MSK_MAC_CSR10_DATA;
+
+	return data;
+}
+
+static int
+owl_emac_mdio_write(struct mii_bus *bus, int addr, int regnum, u16 val)
+{
+	struct owl_emac_priv *priv = bus->priv;
+	u32 data, tmp;
+
+	data = OWL_EMAC_BIT_MAC_CSR10_SB;
+	data |= OWL_EMAC_VAL_MAC_CSR10_OPCODE_WR << OWL_EMAC_OFF_MAC_CSR10_OPCODE;
+
+	tmp = addr << OWL_EMAC_OFF_MAC_CSR10_PHYADD;
+	data |= tmp & OWL_EMAC_MSK_MAC_CSR10_PHYADD;
+
+	tmp = regnum << OWL_EMAC_OFF_MAC_CSR10_REGADD;
+	data |= tmp & OWL_EMAC_MSK_MAC_CSR10_REGADD;
+
+	data |= val & OWL_EMAC_MSK_MAC_CSR10_DATA;
+
+	owl_emac_reg_write(priv, OWL_EMAC_REG_MAC_CSR10, data);
+
+	return owl_emac_mdio_wait(priv);
+}
+
+static int owl_emac_mdio_init(struct net_device *netdev)
+{
+	struct owl_emac_priv *priv = netdev_priv(netdev);
+	struct device *dev = owl_emac_get_dev(priv);
+	struct device_node *mdio_node;
+	int ret;
+
+	mdio_node = of_get_available_child_by_name(dev->of_node, "mdio");
+	if (!mdio_node)
+		return -ENODEV;
+
+	priv->mii = devm_mdiobus_alloc(dev);
+	if (!priv->mii) {
+		ret = -ENOMEM;
+		goto err_put_node;
+	}
+
+	snprintf(priv->mii->id, MII_BUS_ID_SIZE, "%s", dev_name(dev));
+	priv->mii->name = "owl-emac-mdio";
+	priv->mii->parent = dev;
+	priv->mii->read = owl_emac_mdio_read;
+	priv->mii->write = owl_emac_mdio_write;
+	priv->mii->phy_mask = ~0; /* Mask out all PHYs from auto probing. */
+	priv->mii->priv = priv;
+
+	ret = devm_of_mdiobus_register(dev, priv->mii, mdio_node);
+
+err_put_node:
+	of_node_put(mdio_node);
+	return ret;
+}
+
+static int owl_emac_phy_init(struct net_device *netdev)
+{
+	struct owl_emac_priv *priv = netdev_priv(netdev);
+	struct device *dev = owl_emac_get_dev(priv);
+	struct phy_device *phy;
+
+	phy = of_phy_get_and_connect(netdev, dev->of_node,
+				     owl_emac_adjust_link);
+	if (!phy)
+		return -ENODEV;
+
+	phy_set_sym_pause(phy, true, true, true);
+
+	if (netif_msg_link(priv))
+		phy_attached_info(phy);
+
+	return 0;
+}
+
+static void owl_emac_get_mac_addr(struct net_device *netdev)
+{
+	struct device *dev = netdev->dev.parent;
+	int ret;
+
+	ret = platform_get_ethdev_address(dev, netdev);
+	if (!ret && is_valid_ether_addr(netdev->dev_addr))
+		return;
+
+	eth_hw_addr_random(netdev);
+	dev_warn(dev, "using random MAC address %pM\n", netdev->dev_addr);
+}
+
+static __maybe_unused int owl_emac_suspend(struct device *dev)
+{
+	struct net_device *netdev = dev_get_drvdata(dev);
+	struct owl_emac_priv *priv = netdev_priv(netdev);
+
+	disable_irq(netdev->irq);
+
+	if (netif_running(netdev)) {
+		owl_emac_disable(netdev, true);
+		netif_device_detach(netdev);
+	}
+
+	clk_bulk_disable_unprepare(OWL_EMAC_NCLKS, priv->clks);
+
+	return 0;
+}
+
+static __maybe_unused int owl_emac_resume(struct device *dev)
+{
+	struct net_device *netdev = dev_get_drvdata(dev);
+	struct owl_emac_priv *priv = netdev_priv(netdev);
+	int ret;
+
+	ret = clk_bulk_prepare_enable(OWL_EMAC_NCLKS, priv->clks);
+	if (ret)
+		return ret;
+
+	if (netif_running(netdev)) {
+		owl_emac_core_hw_reset(priv);
+		owl_emac_core_sw_reset(priv);
+
+		ret = owl_emac_enable(netdev, true);
+		if (ret) {
+			clk_bulk_disable_unprepare(OWL_EMAC_NCLKS, priv->clks);
+			return ret;
+		}
+
+		netif_device_attach(netdev);
+	}
+
+	enable_irq(netdev->irq);
+
+	return 0;
+}
+
+static void owl_emac_clk_disable_unprepare(void *data)
+{
+	struct owl_emac_priv *priv = data;
+
+	clk_bulk_disable_unprepare(OWL_EMAC_NCLKS, priv->clks);
+}
+
+static int owl_emac_clk_set_rate(struct owl_emac_priv *priv)
+{
+	struct device *dev = owl_emac_get_dev(priv);
+	unsigned long rate;
+	int ret;
+
+	switch (priv->phy_mode) {
+	case PHY_INTERFACE_MODE_RMII:
+		rate = 50000000;
+		break;
+
+	case PHY_INTERFACE_MODE_SMII:
+		rate = 125000000;
+		break;
+
+	default:
+		dev_err(dev, "unsupported phy interface mode %d\n",
+			priv->phy_mode);
+		return -EOPNOTSUPP;
+	}
+
+	ret = clk_set_rate(priv->clks[OWL_EMAC_CLK_RMII].clk, rate);
+	if (ret)
+		dev_err(dev, "failed to set RMII clock rate: %d\n", ret);
+
+	return ret;
+}
+
+static int owl_emac_probe(struct platform_device *pdev)
+{
+	struct device *dev = &pdev->dev;
+	struct net_device *netdev;
+	struct owl_emac_priv *priv;
+	int ret, i;
+
+	netdev = devm_alloc_etherdev(dev, sizeof(*priv));
+	if (!netdev)
+		return -ENOMEM;
+
+	platform_set_drvdata(pdev, netdev);
+	SET_NETDEV_DEV(netdev, dev);
+
+	priv = netdev_priv(netdev);
+	priv->netdev = netdev;
+	priv->msg_enable = netif_msg_init(-1, OWL_EMAC_DEFAULT_MSG_ENABLE);
+
+	ret = of_get_phy_mode(dev->of_node, &priv->phy_mode);
+	if (ret) {
+		dev_err(dev, "failed to get phy mode: %d\n", ret);
+		return ret;
+	}
+
+	spin_lock_init(&priv->lock);
+
+	ret = dma_set_mask_and_coherent(dev, DMA_BIT_MASK(32));
+	if (ret) {
+		dev_err(dev, "unsupported DMA mask\n");
+		return ret;
+	}
+
+	ret = owl_emac_ring_alloc(dev, &priv->rx_ring, OWL_EMAC_RX_RING_SIZE);
+	if (ret)
+		return ret;
+
+	ret = owl_emac_ring_alloc(dev, &priv->tx_ring, OWL_EMAC_TX_RING_SIZE);
+	if (ret)
+		return ret;
+
+	priv->base = devm_platform_ioremap_resource(pdev, 0);
+	if (IS_ERR(priv->base))
+		return PTR_ERR(priv->base);
+
+	netdev->irq = platform_get_irq(pdev, 0);
+	if (netdev->irq < 0)
+		return netdev->irq;
+
+	ret = devm_request_irq(dev, netdev->irq, owl_emac_handle_irq,
+			       IRQF_SHARED, netdev->name, netdev);
+	if (ret) {
+		dev_err(dev, "failed to request irq: %d\n", netdev->irq);
+		return ret;
+	}
+
+	for (i = 0; i < OWL_EMAC_NCLKS; i++)
+		priv->clks[i].id = owl_emac_clk_names[i];
+
+	ret = devm_clk_bulk_get(dev, OWL_EMAC_NCLKS, priv->clks);
+	if (ret)
+		return ret;
+
+	ret = clk_bulk_prepare_enable(OWL_EMAC_NCLKS, priv->clks);
+	if (ret)
+		return ret;
+
+	ret = devm_add_action_or_reset(dev, owl_emac_clk_disable_unprepare, priv);
+	if (ret)
+		return ret;
+
+	ret = owl_emac_clk_set_rate(priv);
+	if (ret)
+		return ret;
+
+	priv->reset = devm_reset_control_get_exclusive(dev, NULL);
+	if (IS_ERR(priv->reset))
+		return dev_err_probe(dev, PTR_ERR(priv->reset),
+				     "failed to get reset control");
+
+	owl_emac_get_mac_addr(netdev);
+
+	owl_emac_core_hw_reset(priv);
+	owl_emac_mdio_clock_enable(priv);
+
+	ret = owl_emac_mdio_init(netdev);
+	if (ret) {
+		dev_err(dev, "failed to initialize MDIO bus\n");
+		return ret;
+	}
+
+	ret = owl_emac_phy_init(netdev);
+	if (ret) {
+		dev_err(dev, "failed to initialize PHY\n");
+		return ret;
+	}
+
+	INIT_WORK(&priv->mac_reset_task, owl_emac_reset_task);
+
+	netdev->min_mtu = OWL_EMAC_MTU_MIN;
+	netdev->max_mtu = OWL_EMAC_MTU_MAX;
+	netdev->watchdog_timeo = OWL_EMAC_TX_TIMEOUT;
+	netdev->netdev_ops = &owl_emac_netdev_ops;
+	netdev->ethtool_ops = &owl_emac_ethtool_ops;
+	netif_napi_add(netdev, &priv->napi, owl_emac_poll);
+
+	ret = devm_register_netdev(dev, netdev);
+	if (ret) {
+		netif_napi_del(&priv->napi);
+		phy_disconnect(netdev->phydev);
+		return ret;
+	}
+
+	return 0;
+}
+
+static void owl_emac_remove(struct platform_device *pdev)
+{
+	struct owl_emac_priv *priv = platform_get_drvdata(pdev);
+
+	netif_napi_del(&priv->napi);
+	phy_disconnect(priv->netdev->phydev);
+	cancel_work_sync(&priv->mac_reset_task);
+}
+
+static const struct of_device_id owl_emac_of_match[] = {
+	{ .compatible = "actions,owl-emac", },
+	{ }
+};
+MODULE_DEVICE_TABLE(of, owl_emac_of_match);
+
+static SIMPLE_DEV_PM_OPS(owl_emac_pm_ops,
+			 owl_emac_suspend, owl_emac_resume);
+
+static struct platform_driver owl_emac_driver = {
+	.driver = {
+		.name = OWL_EMAC_DRVNAME,
+		.of_match_table = owl_emac_of_match,
+		.pm = &owl_emac_pm_ops,
+	},
+	.probe = owl_emac_probe,
+	.remove = owl_emac_remove,
+};
+module_platform_driver(owl_emac_driver);
+
+MODULE_DESCRIPTION("Actions Semi Owl SoCs Ethernet MAC Driver");
+MODULE_AUTHOR("Actions Semi Inc.");
+MODULE_AUTHOR("Cristian Ciocaltea <cristian.ciocaltea@gmail.com>");
+MODULE_LICENSE("GPL");
diff --git a/drivers/net/ethernet/actions/owl-emac.h b/drivers/net/ethernet/actions/owl-emac.h
new file mode 100644
index 000000000000..9eb0d1a30242
--- /dev/null
+++ b/drivers/net/ethernet/actions/owl-emac.h
@@ -0,0 +1,280 @@
+/* SPDX-License-Identifier: GPL-2.0-or-later */
+/*
+ * Actions Semi Owl SoCs Ethernet MAC driver
+ *
+ * Copyright (c) 2012 Actions Semi Inc.
+ * Copyright (c) 2021 Cristian Ciocaltea <cristian.ciocaltea@gmail.com>
+ */
+
+#ifndef __OWL_EMAC_H__
+#define __OWL_EMAC_H__
+
+#define OWL_EMAC_DRVNAME			"owl-emac"
+
+#define OWL_EMAC_POLL_DELAY_USEC		5
+#define OWL_EMAC_MDIO_POLL_TIMEOUT_USEC		1000
+#define OWL_EMAC_RESET_POLL_TIMEOUT_USEC	2000
+#define OWL_EMAC_TX_TIMEOUT			(2 * HZ)
+
+#define OWL_EMAC_MTU_MIN			ETH_MIN_MTU
+#define OWL_EMAC_MTU_MAX			ETH_DATA_LEN
+#define OWL_EMAC_RX_FRAME_MAX_LEN		(ETH_FRAME_LEN + ETH_FCS_LEN)
+#define OWL_EMAC_SKB_ALIGN			4
+#define OWL_EMAC_SKB_RESERVE			18
+
+#define OWL_EMAC_MAX_MULTICAST_ADDRS		14
+#define OWL_EMAC_SETUP_FRAME_LEN		192
+
+#define OWL_EMAC_RX_RING_SIZE			64
+#define OWL_EMAC_TX_RING_SIZE			32
+
+/* Bus mode register */
+#define OWL_EMAC_REG_MAC_CSR0			0x0000
+#define OWL_EMAC_BIT_MAC_CSR0_SWR		BIT(0)	/* Software reset */
+
+/* Transmit/receive poll demand registers */
+#define OWL_EMAC_REG_MAC_CSR1			0x0008
+#define OWL_EMAC_VAL_MAC_CSR1_TPD		0x01
+#define OWL_EMAC_REG_MAC_CSR2			0x0010
+#define OWL_EMAC_VAL_MAC_CSR2_RPD		0x01
+
+/* Receive/transmit descriptor list base address registers */
+#define OWL_EMAC_REG_MAC_CSR3			0x0018
+#define OWL_EMAC_REG_MAC_CSR4			0x0020
+
+/* Status register */
+#define OWL_EMAC_REG_MAC_CSR5			0x0028
+#define OWL_EMAC_MSK_MAC_CSR5_TS		GENMASK(22, 20)	/* Transmit process state */
+#define OWL_EMAC_OFF_MAC_CSR5_TS		20
+#define OWL_EMAC_VAL_MAC_CSR5_TS_DATA		0x03	/* Transferring data HOST -> FIFO */
+#define OWL_EMAC_VAL_MAC_CSR5_TS_CDES		0x07	/* Closing transmit descriptor */
+#define OWL_EMAC_MSK_MAC_CSR5_RS		GENMASK(19, 17)	/* Receive process state */
+#define OWL_EMAC_OFF_MAC_CSR5_RS		17
+#define OWL_EMAC_VAL_MAC_CSR5_RS_FDES		0x01	/* Fetching receive descriptor */
+#define OWL_EMAC_VAL_MAC_CSR5_RS_CDES		0x05	/* Closing receive descriptor */
+#define OWL_EMAC_VAL_MAC_CSR5_RS_DATA		0x07	/* Transferring data FIFO -> HOST */
+#define OWL_EMAC_BIT_MAC_CSR5_NIS		BIT(16)	/* Normal interrupt summary */
+#define OWL_EMAC_BIT_MAC_CSR5_AIS		BIT(15)	/* Abnormal interrupt summary */
+#define OWL_EMAC_BIT_MAC_CSR5_ERI		BIT(14)	/* Early receive interrupt */
+#define OWL_EMAC_BIT_MAC_CSR5_GTE		BIT(11)	/* General-purpose timer expiration */
+#define OWL_EMAC_BIT_MAC_CSR5_ETI		BIT(10)	/* Early transmit interrupt */
+#define OWL_EMAC_BIT_MAC_CSR5_RPS		BIT(8)	/* Receive process stopped */
+#define OWL_EMAC_BIT_MAC_CSR5_RU		BIT(7)	/* Receive buffer unavailable */
+#define OWL_EMAC_BIT_MAC_CSR5_RI		BIT(6)	/* Receive interrupt */
+#define OWL_EMAC_BIT_MAC_CSR5_UNF		BIT(5)	/* Transmit underflow */
+#define OWL_EMAC_BIT_MAC_CSR5_LCIS		BIT(4)	/* Link change status */
+#define OWL_EMAC_BIT_MAC_CSR5_LCIQ		BIT(3)	/* Link change interrupt */
+#define OWL_EMAC_BIT_MAC_CSR5_TU		BIT(2)	/* Transmit buffer unavailable */
+#define OWL_EMAC_BIT_MAC_CSR5_TPS		BIT(1)	/* Transmit process stopped */
+#define OWL_EMAC_BIT_MAC_CSR5_TI		BIT(0)	/* Transmit interrupt */
+
+/* Operation mode register */
+#define OWL_EMAC_REG_MAC_CSR6			0x0030
+#define OWL_EMAC_BIT_MAC_CSR6_RA		BIT(30)	/* Receive all */
+#define OWL_EMAC_BIT_MAC_CSR6_TTM		BIT(22)	/* Transmit threshold mode */
+#define OWL_EMAC_BIT_MAC_CSR6_SF		BIT(21)	/* Store and forward */
+#define OWL_EMAC_MSK_MAC_CSR6_SPEED		GENMASK(17, 16)	/* Eth speed selection */
+#define OWL_EMAC_OFF_MAC_CSR6_SPEED		16
+#define OWL_EMAC_VAL_MAC_CSR6_SPEED_100M	0x00
+#define OWL_EMAC_VAL_MAC_CSR6_SPEED_10M		0x02
+#define OWL_EMAC_BIT_MAC_CSR6_ST		BIT(13)	/* Start/stop transmit command */
+#define OWL_EMAC_BIT_MAC_CSR6_LP		BIT(10)	/* Loopback mode */
+#define OWL_EMAC_BIT_MAC_CSR6_FD		BIT(9)	/* Full duplex mode */
+#define OWL_EMAC_BIT_MAC_CSR6_PM		BIT(7)	/* Pass all multicast */
+#define OWL_EMAC_BIT_MAC_CSR6_PR		BIT(6)	/* Promiscuous mode */
+#define OWL_EMAC_BIT_MAC_CSR6_IF		BIT(4)	/* Inverse filtering */
+#define OWL_EMAC_BIT_MAC_CSR6_PB		BIT(3)	/* Pass bad frames */
+#define OWL_EMAC_BIT_MAC_CSR6_HO		BIT(2)	/* Hash only filtering mode */
+#define OWL_EMAC_BIT_MAC_CSR6_SR		BIT(1)	/* Start/stop receive command */
+#define OWL_EMAC_BIT_MAC_CSR6_HP		BIT(0)	/* Hash/perfect receive filtering mode */
+#define OWL_EMAC_MSK_MAC_CSR6_STSR	       (OWL_EMAC_BIT_MAC_CSR6_ST | \
+						OWL_EMAC_BIT_MAC_CSR6_SR)
+
+/* Interrupt enable register */
+#define OWL_EMAC_REG_MAC_CSR7			0x0038
+#define OWL_EMAC_BIT_MAC_CSR7_NIE		BIT(16)	/* Normal interrupt summary enable */
+#define OWL_EMAC_BIT_MAC_CSR7_AIE		BIT(15)	/* Abnormal interrupt summary enable */
+#define OWL_EMAC_BIT_MAC_CSR7_ERE		BIT(14)	/* Early receive interrupt enable */
+#define OWL_EMAC_BIT_MAC_CSR7_GTE		BIT(11)	/* General-purpose timer overflow */
+#define OWL_EMAC_BIT_MAC_CSR7_ETE		BIT(10)	/* Early transmit interrupt enable */
+#define OWL_EMAC_BIT_MAC_CSR7_RSE		BIT(8)	/* Receive stopped enable */
+#define OWL_EMAC_BIT_MAC_CSR7_RUE		BIT(7)	/* Receive buffer unavailable enable */
+#define OWL_EMAC_BIT_MAC_CSR7_RIE		BIT(6)	/* Receive interrupt enable */
+#define OWL_EMAC_BIT_MAC_CSR7_UNE		BIT(5)	/* Underflow interrupt enable */
+#define OWL_EMAC_BIT_MAC_CSR7_TUE		BIT(2)	/* Transmit buffer unavailable enable */
+#define OWL_EMAC_BIT_MAC_CSR7_TSE		BIT(1)	/* Transmit stopped enable */
+#define OWL_EMAC_BIT_MAC_CSR7_TIE		BIT(0)	/* Transmit interrupt enable */
+#define OWL_EMAC_BIT_MAC_CSR7_ALL_NOT_TUE      (OWL_EMAC_BIT_MAC_CSR7_ERE | \
+						OWL_EMAC_BIT_MAC_CSR7_GTE | \
+						OWL_EMAC_BIT_MAC_CSR7_ETE | \
+						OWL_EMAC_BIT_MAC_CSR7_RSE | \
+						OWL_EMAC_BIT_MAC_CSR7_RUE | \
+						OWL_EMAC_BIT_MAC_CSR7_RIE | \
+						OWL_EMAC_BIT_MAC_CSR7_UNE | \
+						OWL_EMAC_BIT_MAC_CSR7_TSE | \
+						OWL_EMAC_BIT_MAC_CSR7_TIE)
+
+/* Missed frames and overflow counter register */
+#define OWL_EMAC_REG_MAC_CSR8			0x0040
+/* MII management and serial ROM register */
+#define OWL_EMAC_REG_MAC_CSR9			0x0048
+
+/* MII serial management register */
+#define OWL_EMAC_REG_MAC_CSR10			0x0050
+#define OWL_EMAC_BIT_MAC_CSR10_SB		BIT(31)	/* Start transfer or busy */
+#define OWL_EMAC_MSK_MAC_CSR10_CLKDIV		GENMASK(30, 28)	/* Clock divider */
+#define OWL_EMAC_OFF_MAC_CSR10_CLKDIV		28
+#define OWL_EMAC_VAL_MAC_CSR10_CLKDIV_128	0x04
+#define OWL_EMAC_VAL_MAC_CSR10_OPCODE_WR	0x01	/* Register write command */
+#define OWL_EMAC_OFF_MAC_CSR10_OPCODE		26	/* Operation mode */
+#define OWL_EMAC_VAL_MAC_CSR10_OPCODE_DCG	0x00	/* Disable clock generation */
+#define OWL_EMAC_VAL_MAC_CSR10_OPCODE_WR	0x01	/* Register write command */
+#define OWL_EMAC_VAL_MAC_CSR10_OPCODE_RD	0x02	/* Register read command */
+#define OWL_EMAC_VAL_MAC_CSR10_OPCODE_CDS	0x03	/* Clock divider set */
+#define OWL_EMAC_MSK_MAC_CSR10_PHYADD		GENMASK(25, 21)	/* Physical layer address */
+#define OWL_EMAC_OFF_MAC_CSR10_PHYADD		21
+#define OWL_EMAC_MSK_MAC_CSR10_REGADD		GENMASK(20, 16)	/* Register address */
+#define OWL_EMAC_OFF_MAC_CSR10_REGADD		16
+#define OWL_EMAC_MSK_MAC_CSR10_DATA		GENMASK(15, 0)	/* Register data */
+
+/* General-purpose timer and interrupt mitigation control register */
+#define OWL_EMAC_REG_MAC_CSR11			0x0058
+#define OWL_EMAC_OFF_MAC_CSR11_TT		27	/* Transmit timer */
+#define OWL_EMAC_OFF_MAC_CSR11_NTP		24	/* No. of transmit packets */
+#define OWL_EMAC_OFF_MAC_CSR11_RT		20	/* Receive timer */
+#define OWL_EMAC_OFF_MAC_CSR11_NRP		17	/* No. of receive packets */
+
+/* MAC address low/high registers */
+#define OWL_EMAC_REG_MAC_CSR16			0x0080
+#define OWL_EMAC_REG_MAC_CSR17			0x0088
+
+/* Pause time & cache thresholds register */
+#define OWL_EMAC_REG_MAC_CSR18			0x0090
+#define OWL_EMAC_OFF_MAC_CSR18_CPTL		24	/* Cache pause threshold level */
+#define OWL_EMAC_OFF_MAC_CSR18_CRTL		16	/* Cache restart threshold level */
+#define OWL_EMAC_OFF_MAC_CSR18_PQT		0	/* Flow control pause quanta time */
+
+/* FIFO pause & restart threshold register */
+#define OWL_EMAC_REG_MAC_CSR19			0x0098
+#define OWL_EMAC_OFF_MAC_CSR19_FPTL		16	/* FIFO pause threshold level */
+#define OWL_EMAC_OFF_MAC_CSR19_FRTL		0	/* FIFO restart threshold level */
+
+/* Flow control setup & status register */
+#define OWL_EMAC_REG_MAC_CSR20			0x00A0
+#define OWL_EMAC_BIT_MAC_CSR20_FCE		BIT(31)	/* Flow Control Enable */
+#define OWL_EMAC_BIT_MAC_CSR20_TUE		BIT(30)	/* Transmit Un-pause frames Enable */
+#define OWL_EMAC_BIT_MAC_CSR20_TPE		BIT(29)	/* Transmit Pause frames Enable */
+#define OWL_EMAC_BIT_MAC_CSR20_RPE		BIT(28)	/* Receive Pause frames Enable */
+#define OWL_EMAC_BIT_MAC_CSR20_BPE		BIT(27)	/* Back pressure (half-duplex) Enable */
+
+/* MII control register */
+#define OWL_EMAC_REG_MAC_CTRL			0x00B0
+#define OWL_EMAC_BIT_MAC_CTRL_RRSB		BIT(8)	/* RMII_REFCLK select bit */
+#define OWL_EMAC_OFF_MAC_CTRL_SSDC		4	/* SMII SYNC delay cycle */
+#define OWL_EMAC_BIT_MAC_CTRL_RCPS		BIT(1)	/* REF_CLK phase select */
+#define OWL_EMAC_BIT_MAC_CTRL_RSIS		BIT(0)	/* RMII/SMII interface select */
+
+/* Receive descriptor status field */
+#define OWL_EMAC_BIT_RDES0_OWN			BIT(31)	/* Ownership bit */
+#define OWL_EMAC_BIT_RDES0_FF			BIT(30)	/* Filtering fail */
+#define OWL_EMAC_MSK_RDES0_FL			GENMASK(29, 16)	/* Frame length */
+#define OWL_EMAC_OFF_RDES0_FL			16
+#define OWL_EMAC_BIT_RDES0_ES			BIT(15)	/* Error summary */
+#define OWL_EMAC_BIT_RDES0_DE			BIT(14)	/* Descriptor error */
+#define OWL_EMAC_BIT_RDES0_RF			BIT(11)	/* Runt frame */
+#define OWL_EMAC_BIT_RDES0_MF			BIT(10)	/* Multicast frame */
+#define OWL_EMAC_BIT_RDES0_FS			BIT(9)	/* First descriptor */
+#define OWL_EMAC_BIT_RDES0_LS			BIT(8)	/* Last descriptor */
+#define OWL_EMAC_BIT_RDES0_TL			BIT(7)	/* Frame too long */
+#define OWL_EMAC_BIT_RDES0_CS			BIT(6)	/* Collision seen */
+#define OWL_EMAC_BIT_RDES0_FT			BIT(5)	/* Frame type */
+#define OWL_EMAC_BIT_RDES0_RE			BIT(3)	/* Report on MII error */
+#define OWL_EMAC_BIT_RDES0_DB			BIT(2)	/* Dribbling bit */
+#define OWL_EMAC_BIT_RDES0_CE			BIT(1)	/* CRC error */
+#define OWL_EMAC_BIT_RDES0_ZERO			BIT(0)	/* Legal frame length indicator */
+
+/* Receive descriptor control and count field */
+#define OWL_EMAC_BIT_RDES1_RER			BIT(25)	/* Receive end of ring */
+#define OWL_EMAC_MSK_RDES1_RBS1			GENMASK(10, 0) /* Buffer 1 size */
+
+/* Transmit descriptor status field */
+#define OWL_EMAC_BIT_TDES0_OWN			BIT(31)	/* Ownership bit */
+#define OWL_EMAC_BIT_TDES0_ES			BIT(15)	/* Error summary */
+#define OWL_EMAC_BIT_TDES0_LO			BIT(11)	/* Loss of carrier */
+#define OWL_EMAC_BIT_TDES0_NC			BIT(10)	/* No carrier */
+#define OWL_EMAC_BIT_TDES0_LC			BIT(9)	/* Late collision */
+#define OWL_EMAC_BIT_TDES0_EC			BIT(8)	/* Excessive collisions */
+#define OWL_EMAC_MSK_TDES0_CC			GENMASK(6, 3) /* Collision count */
+#define OWL_EMAC_BIT_TDES0_UF			BIT(1)	/* Underflow error */
+#define OWL_EMAC_BIT_TDES0_DE			BIT(0)	/* Deferred */
+
+/* Transmit descriptor control and count field */
+#define OWL_EMAC_BIT_TDES1_IC			BIT(31)	/* Interrupt on completion */
+#define OWL_EMAC_BIT_TDES1_LS			BIT(30)	/* Last descriptor */
+#define OWL_EMAC_BIT_TDES1_FS			BIT(29)	/* First descriptor */
+#define OWL_EMAC_BIT_TDES1_FT1			BIT(28)	/* Filtering type */
+#define OWL_EMAC_BIT_TDES1_SET			BIT(27)	/* Setup packet */
+#define OWL_EMAC_BIT_TDES1_AC			BIT(26)	/* Add CRC disable */
+#define OWL_EMAC_BIT_TDES1_TER			BIT(25)	/* Transmit end of ring */
+#define OWL_EMAC_BIT_TDES1_DPD			BIT(23)	/* Disabled padding */
+#define OWL_EMAC_BIT_TDES1_FT0			BIT(22)	/* Filtering type */
+#define OWL_EMAC_MSK_TDES1_TBS1			GENMASK(10, 0) /* Buffer 1 size */
+
+static const char *const owl_emac_clk_names[] = { "eth", "rmii" };
+#define OWL_EMAC_NCLKS ARRAY_SIZE(owl_emac_clk_names)
+
+enum owl_emac_clk_map {
+	OWL_EMAC_CLK_ETH = 0,
+	OWL_EMAC_CLK_RMII
+};
+
+struct owl_emac_addr_list {
+	u8 addrs[OWL_EMAC_MAX_MULTICAST_ADDRS][ETH_ALEN];
+	int count;
+};
+
+/* TX/RX descriptors */
+struct owl_emac_ring_desc {
+	u32 status;
+	u32 control;
+	u32 buf_addr;
+	u32 reserved;		/* 2nd buffer address is not used */
+};
+
+struct owl_emac_ring {
+	struct owl_emac_ring_desc *descs;
+	dma_addr_t descs_dma;
+	struct sk_buff **skbs;
+	dma_addr_t *skbs_dma;
+	unsigned int size;
+	unsigned int head;
+	unsigned int tail;
+};
+
+struct owl_emac_priv {
+	struct net_device *netdev;
+	void __iomem *base;
+
+	struct clk_bulk_data clks[OWL_EMAC_NCLKS];
+	struct reset_control *reset;
+
+	struct owl_emac_ring rx_ring;
+	struct owl_emac_ring tx_ring;
+
+	struct mii_bus *mii;
+	struct napi_struct napi;
+
+	phy_interface_t phy_mode;
+	unsigned int link;
+	int speed;
+	int duplex;
+	int pause;
+	struct owl_emac_addr_list mcaddr_list;
+
+	struct work_struct mac_reset_task;
+
+	u32 msg_enable;		/* Debug message level */
+	spinlock_t lock;	/* Sync concurrent ring access */
+};
+
+#endif /* __OWL_EMAC_H__ */
diff --git a/drivers/net/ethernet/adaptec/Kconfig b/drivers/net/ethernet/adaptec/Kconfig
index 822cffb4174c..c96edc2e582f 100644
--- a/drivers/net/ethernet/adaptec/Kconfig
+++ b/drivers/net/ethernet/adaptec/Kconfig
@@ -1,3 +1,4 @@
+# SPDX-License-Identifier: GPL-2.0-only
 #
 # Adaptec network device configuration
 #
@@ -6,7 +7,7 @@ config NET_VENDOR_ADAPTEC
 	bool "Adaptec devices"
 	default y
 	depends on PCI
-	---help---
+	help
 	  If you have a network (Ethernet) card belonging to this class, say Y.
 
 	  Note that the answer to this question doesn't directly affect the
@@ -21,7 +22,7 @@ config ADAPTEC_STARFIRE
 	depends on PCI
 	select CRC32
 	select MII
-	---help---
+	help
 	  Say Y here if you have an Adaptec Starfire (or DuraLAN) PCI network
 	  adapter. The DuraLAN chip is used on the 64 bit PCI boards from
 	  Adaptec e.g. the ANA-6922A. The older 32 bit boards use the tulip
diff --git a/drivers/net/ethernet/adaptec/Makefile b/drivers/net/ethernet/adaptec/Makefile
index 6c07b758ac0a..d84138c8a9ea 100644
--- a/drivers/net/ethernet/adaptec/Makefile
+++ b/drivers/net/ethernet/adaptec/Makefile
@@ -1,3 +1,4 @@
+# SPDX-License-Identifier: GPL-2.0-only
 #
 # Makefile for the Adaptec network device drivers.
 #
diff --git a/drivers/net/ethernet/adaptec/starfire.c b/drivers/net/ethernet/adaptec/starfire.c
index 3872ab96b80a..e1b8794b14c9 100644
--- a/drivers/net/ethernet/adaptec/starfire.c
+++ b/drivers/net/ethernet/adaptec/starfire.c
@@ -27,8 +27,6 @@
 */
 
 #define DRV_NAME	"starfire"
-#define DRV_VERSION	"2.1"
-#define DRV_RELDATE	"July  6, 2008"
 
 #include <linux/interrupt.h>
 #include <linux/module.h>
@@ -165,15 +163,9 @@ static int rx_copybreak /* = 0 */;
 #define FIRMWARE_RX	"adaptec/starfire_rx.bin"
 #define FIRMWARE_TX	"adaptec/starfire_tx.bin"
 
-/* These identify the driver base version and may not be removed. */
-static const char version[] =
-KERN_INFO "starfire.c:v1.03 7/26/2000  Written by Donald Becker <becker@scyld.com>\n"
-" (unofficial 2.2/2.4 kernel port, version " DRV_VERSION ", " DRV_RELDATE ")\n";
-
 MODULE_AUTHOR("Donald Becker <becker@scyld.com>");
 MODULE_DESCRIPTION("Adaptec Starfire Ethernet driver");
 MODULE_LICENSE("GPL");
-MODULE_VERSION(DRV_VERSION);
 MODULE_FIRMWARE(FIRMWARE_RX);
 MODULE_FIRMWARE(FIRMWARE_TX);
 
@@ -449,14 +441,6 @@ enum rx_desc_bits {
 };
 
 /* Completion queue entry. */
-struct short_rx_done_desc {
-	__le32 status;			/* Low 16 bits is length. */
-};
-struct basic_rx_done_desc {
-	__le32 status;			/* Low 16 bits is length. */
-	__le16 vlanid;
-	__le16 status2;
-};
 struct csum_rx_done_desc {
 	__le32 status;			/* Low 16 bits is length. */
 	__le16 csum;			/* Partial checksum */
@@ -576,7 +560,7 @@ static int	mdio_read(struct net_device *dev, int phy_id, int location);
 static void	mdio_write(struct net_device *dev, int phy_id, int location, int value);
 static int	netdev_open(struct net_device *dev);
 static void	check_duplex(struct net_device *dev);
-static void	tx_timeout(struct net_device *dev);
+static void	tx_timeout(struct net_device *dev, unsigned int txqueue);
 static void	init_ring(struct net_device *dev);
 static netdev_tx_t start_tx(struct sk_buff *skb, struct net_device *dev);
 static irqreturn_t intr_handler(int irq, void *dev_instance);
@@ -633,7 +617,7 @@ static const struct net_device_ops netdev_ops = {
 	.ndo_tx_timeout		= tx_timeout,
 	.ndo_get_stats		= get_stats,
 	.ndo_set_rx_mode	= set_rx_mode,
-	.ndo_do_ioctl		= netdev_ioctl,
+	.ndo_eth_ioctl		= netdev_ioctl,
 	.ndo_set_mac_address	= eth_mac_addr,
 	.ndo_validate_addr	= eth_validate_addr,
 #ifdef VLAN_SUPPORT
@@ -649,18 +633,12 @@ static int starfire_init_one(struct pci_dev *pdev,
 	struct netdev_private *np;
 	int i, irq, chip_idx = ent->driver_data;
 	struct net_device *dev;
+	u8 addr[ETH_ALEN];
 	long ioaddr;
 	void __iomem *base;
 	int drv_flags, io_size;
 	int boguscnt;
 
-/* when built into the kernel, we only print version if device is found */
-#ifndef MODULE
-	static int printed_version;
-	if (!printed_version++)
-		printk(version);
-#endif
-
 	if (pci_enable_device (pdev))
 		return -EIO;
 
@@ -711,7 +689,8 @@ static int starfire_init_one(struct pci_dev *pdev,
 
 	/* Serial EEPROM reads are hidden by the hardware. */
 	for (i = 0; i < 6; i++)
-		dev->dev_addr[i] = readb(base + EEPROMCtrl + 20 - i);
+		addr[i] = readb(base + EEPROMCtrl + 20 - i);
+	eth_hw_addr_set(dev, addr);
 
 #if ! defined(final_version) /* Dump the EEPROM contents during development. */
 	if (debug > 4)
@@ -785,7 +764,7 @@ static int starfire_init_one(struct pci_dev *pdev,
 	dev->watchdog_timeo = TX_TIMEOUT;
 	dev->ethtool_ops = &ethtool_ops;
 
-	netif_napi_add(dev, &np->napi, netdev_poll, max_interrupt_work);
+	netif_napi_add_weight(dev, &np->napi, netdev_poll, max_interrupt_work);
 
 	if (mtu)
 		dev->mtu = mtu;
@@ -802,7 +781,7 @@ static int starfire_init_one(struct pci_dev *pdev,
 		int mii_status;
 		for (phy = 0; phy < 32 && phy_idx < PHY_CNT; phy++) {
 			mdio_write(dev, phy, MII_BMCR, BMCR_RESET);
-			mdelay(100);
+			msleep(100);
 			boguscnt = 1000;
 			while (--boguscnt > 0)
 				if ((mdio_read(dev, phy, MII_BMCR) & BMCR_RESET) == 0)
@@ -901,7 +880,9 @@ static int netdev_open(struct net_device *dev)
 		tx_ring_size = ((sizeof(starfire_tx_desc) * TX_RING_SIZE + QUEUE_ALIGN - 1) / QUEUE_ALIGN) * QUEUE_ALIGN;
 		rx_ring_size = sizeof(struct starfire_rx_desc) * RX_RING_SIZE;
 		np->queue_mem_size = tx_done_q_size + rx_done_q_size + tx_ring_size + rx_ring_size;
-		np->queue_mem = pci_alloc_consistent(np->pci_dev, np->queue_mem_size, &np->queue_mem_dma);
+		np->queue_mem = dma_alloc_coherent(&np->pci_dev->dev,
+						   np->queue_mem_size,
+						   &np->queue_mem_dma, GFP_ATOMIC);
 		if (np->queue_mem == NULL) {
 			free_irq(irq, dev);
 			return -ENOMEM;
@@ -968,7 +949,7 @@ static int netdev_open(struct net_device *dev)
 	writew(0, ioaddr + PerfFilterTable + 4);
 	writew(0, ioaddr + PerfFilterTable + 8);
 	for (i = 1; i < 16; i++) {
-		__be16 *eaddrs = (__be16 *)dev->dev_addr;
+		const __be16 *eaddrs = (const __be16 *)dev->dev_addr;
 		void __iomem *setup_frm = ioaddr + PerfFilterTable + i * 16;
 		writew(be16_to_cpu(eaddrs[2]), setup_frm); setup_frm += 4;
 		writew(be16_to_cpu(eaddrs[1]), setup_frm); setup_frm += 4;
@@ -1105,7 +1086,7 @@ static void check_duplex(struct net_device *dev)
 }
 
 
-static void tx_timeout(struct net_device *dev)
+static void tx_timeout(struct net_device *dev, unsigned int txqueue)
 {
 	struct netdev_private *np = netdev_priv(dev);
 	void __iomem *ioaddr = np->base;
@@ -1151,9 +1132,11 @@ static void init_ring(struct net_device *dev)
 		np->rx_info[i].skb = skb;
 		if (skb == NULL)
 			break;
-		np->rx_info[i].mapping = pci_map_single(np->pci_dev, skb->data, np->rx_buf_sz, PCI_DMA_FROMDEVICE);
-		if (pci_dma_mapping_error(np->pci_dev,
-					  np->rx_info[i].mapping)) {
+		np->rx_info[i].mapping = dma_map_single(&np->pci_dev->dev,
+							skb->data,
+							np->rx_buf_sz,
+							DMA_FROM_DEVICE);
+		if (dma_mapping_error(&np->pci_dev->dev, np->rx_info[i].mapping)) {
 			dev_kfree_skb(skb);
 			np->rx_info[i].skb = NULL;
 			break;
@@ -1232,18 +1215,19 @@ static netdev_tx_t start_tx(struct sk_buff *skb, struct net_device *dev)
 			status |= skb_first_frag_len(skb) | (skb_num_frags(skb) << 16);
 
 			np->tx_info[entry].mapping =
-				pci_map_single(np->pci_dev, skb->data, skb_first_frag_len(skb), PCI_DMA_TODEVICE);
+				dma_map_single(&np->pci_dev->dev, skb->data,
+					       skb_first_frag_len(skb),
+					       DMA_TO_DEVICE);
 		} else {
 			const skb_frag_t *this_frag = &skb_shinfo(skb)->frags[i - 1];
 			status |= skb_frag_size(this_frag);
 			np->tx_info[entry].mapping =
-				pci_map_single(np->pci_dev,
+				dma_map_single(&np->pci_dev->dev,
 					       skb_frag_address(this_frag),
 					       skb_frag_size(this_frag),
-					       PCI_DMA_TODEVICE);
+					       DMA_TO_DEVICE);
 		}
-		if (pci_dma_mapping_error(np->pci_dev,
-					  np->tx_info[entry].mapping)) {
+		if (dma_mapping_error(&np->pci_dev->dev, np->tx_info[entry].mapping)) {
 			dev->stats.tx_dropped++;
 			goto err_out;
 		}
@@ -1286,18 +1270,16 @@ err_out:
 	entry = prev_tx % TX_RING_SIZE;
 	np->tx_info[entry].skb = NULL;
 	if (i > 0) {
-		pci_unmap_single(np->pci_dev,
+		dma_unmap_single(&np->pci_dev->dev,
 				 np->tx_info[entry].mapping,
-				 skb_first_frag_len(skb),
-				 PCI_DMA_TODEVICE);
+				 skb_first_frag_len(skb), DMA_TO_DEVICE);
 		np->tx_info[entry].mapping = 0;
 		entry = (entry + np->tx_info[entry].used_slots) % TX_RING_SIZE;
 		for (j = 1; j < i; j++) {
-			pci_unmap_single(np->pci_dev,
+			dma_unmap_single(&np->pci_dev->dev,
 					 np->tx_info[entry].mapping,
-					 skb_frag_size(
-						&skb_shinfo(skb)->frags[j-1]),
-					 PCI_DMA_TODEVICE);
+					 skb_frag_size(&skb_shinfo(skb)->frags[j - 1]),
+					 DMA_TO_DEVICE);
 			entry++;
 		}
 	}
@@ -1371,26 +1353,26 @@ static irqreturn_t intr_handler(int irq, void *dev_instance)
 				u16 entry = (tx_status & 0x7fff) / sizeof(starfire_tx_desc);
 				struct sk_buff *skb = np->tx_info[entry].skb;
 				np->tx_info[entry].skb = NULL;
-				pci_unmap_single(np->pci_dev,
+				dma_unmap_single(&np->pci_dev->dev,
 						 np->tx_info[entry].mapping,
 						 skb_first_frag_len(skb),
-						 PCI_DMA_TODEVICE);
+						 DMA_TO_DEVICE);
 				np->tx_info[entry].mapping = 0;
 				np->dirty_tx += np->tx_info[entry].used_slots;
 				entry = (entry + np->tx_info[entry].used_slots) % TX_RING_SIZE;
 				{
 					int i;
 					for (i = 0; i < skb_shinfo(skb)->nr_frags; i++) {
-						pci_unmap_single(np->pci_dev,
+						dma_unmap_single(&np->pci_dev->dev,
 								 np->tx_info[entry].mapping,
 								 skb_frag_size(&skb_shinfo(skb)->frags[i]),
-								 PCI_DMA_TODEVICE);
+								 DMA_TO_DEVICE);
 						np->dirty_tx++;
 						entry++;
 					}
 				}
 
-				dev_kfree_skb_irq(skb);
+				dev_consume_skb_irq(skb);
 			}
 			np->tx_done_q[np->tx_done].status = 0;
 			np->tx_done = (np->tx_done + 1) % DONE_Q_SIZE;
@@ -1476,16 +1458,18 @@ static int __netdev_rx(struct net_device *dev, int *quota)
 		if (pkt_len < rx_copybreak &&
 		    (skb = netdev_alloc_skb(dev, pkt_len + 2)) != NULL) {
 			skb_reserve(skb, 2);	/* 16 byte align the IP header */
-			pci_dma_sync_single_for_cpu(np->pci_dev,
-						    np->rx_info[entry].mapping,
-						    pkt_len, PCI_DMA_FROMDEVICE);
+			dma_sync_single_for_cpu(&np->pci_dev->dev,
+						np->rx_info[entry].mapping,
+						pkt_len, DMA_FROM_DEVICE);
 			skb_copy_to_linear_data(skb, np->rx_info[entry].skb->data, pkt_len);
-			pci_dma_sync_single_for_device(np->pci_dev,
-						       np->rx_info[entry].mapping,
-						       pkt_len, PCI_DMA_FROMDEVICE);
+			dma_sync_single_for_device(&np->pci_dev->dev,
+						   np->rx_info[entry].mapping,
+						   pkt_len, DMA_FROM_DEVICE);
 			skb_put(skb, pkt_len);
 		} else {
-			pci_unmap_single(np->pci_dev, np->rx_info[entry].mapping, np->rx_buf_sz, PCI_DMA_FROMDEVICE);
+			dma_unmap_single(&np->pci_dev->dev,
+					 np->rx_info[entry].mapping,
+					 np->rx_buf_sz, DMA_FROM_DEVICE);
 			skb = np->rx_info[entry].skb;
 			skb_put(skb, pkt_len);
 			np->rx_info[entry].skb = NULL;
@@ -1603,9 +1587,9 @@ static void refill_rx_ring(struct net_device *dev)
 			if (skb == NULL)
 				break;	/* Better luck next round. */
 			np->rx_info[entry].mapping =
-				pci_map_single(np->pci_dev, skb->data, np->rx_buf_sz, PCI_DMA_FROMDEVICE);
-			if (pci_dma_mapping_error(np->pci_dev,
-						np->rx_info[entry].mapping)) {
+				dma_map_single(&np->pci_dev->dev, skb->data,
+					       np->rx_buf_sz, DMA_FROM_DEVICE);
+			if (dma_mapping_error(&np->pci_dev->dev, np->rx_info[entry].mapping)) {
 				dev_kfree_skb(skb);
 				np->rx_info[entry].skb = NULL;
 				break;
@@ -1797,14 +1781,14 @@ static void set_rx_mode(struct net_device *dev)
 	} else if (netdev_mc_count(dev) <= 14) {
 		/* Use the 16 element perfect filter, skip first two entries. */
 		void __iomem *filter_addr = ioaddr + PerfFilterTable + 2 * 16;
-		__be16 *eaddrs;
+		const __be16 *eaddrs;
 		netdev_for_each_mc_addr(ha, dev) {
 			eaddrs = (__be16 *) ha->addr;
 			writew(be16_to_cpu(eaddrs[2]), filter_addr); filter_addr += 4;
 			writew(be16_to_cpu(eaddrs[1]), filter_addr); filter_addr += 4;
 			writew(be16_to_cpu(eaddrs[0]), filter_addr); filter_addr += 8;
 		}
-		eaddrs = (__be16 *)dev->dev_addr;
+		eaddrs = (const __be16 *)dev->dev_addr;
 		i = netdev_mc_count(dev) + 2;
 		while (i++ < 16) {
 			writew(be16_to_cpu(eaddrs[0]), filter_addr); filter_addr += 4;
@@ -1815,7 +1799,7 @@ static void set_rx_mode(struct net_device *dev)
 	} else {
 		/* Must use a multicast hash table. */
 		void __iomem *filter_addr;
-		__be16 *eaddrs;
+		const __be16 *eaddrs;
 		__le16 mc_filter[32] __attribute__ ((aligned(sizeof(long))));	/* Multicast hash filter */
 
 		memset(mc_filter, 0, sizeof(mc_filter));
@@ -1829,7 +1813,7 @@ static void set_rx_mode(struct net_device *dev)
 		}
 		/* Clear the perfect filter list, skip first two entries. */
 		filter_addr = ioaddr + PerfFilterTable + 2 * 16;
-		eaddrs = (__be16 *)dev->dev_addr;
+		eaddrs = (const __be16 *)dev->dev_addr;
 		for (i = 2; i < 16; i++) {
 			writew(be16_to_cpu(eaddrs[0]), filter_addr); filter_addr += 4;
 			writew(be16_to_cpu(eaddrs[1]), filter_addr); filter_addr += 4;
@@ -1852,9 +1836,8 @@ static int check_if_running(struct net_device *dev)
 static void get_drvinfo(struct net_device *dev, struct ethtool_drvinfo *info)
 {
 	struct netdev_private *np = netdev_priv(dev);
-	strlcpy(info->driver, DRV_NAME, sizeof(info->driver));
-	strlcpy(info->version, DRV_VERSION, sizeof(info->version));
-	strlcpy(info->bus_info, pci_name(np->pci_dev), sizeof(info->bus_info));
+	strscpy(info->driver, DRV_NAME, sizeof(info->driver));
+	strscpy(info->bus_info, pci_name(np->pci_dev), sizeof(info->bus_info));
 }
 
 static int get_link_ksettings(struct net_device *dev,
@@ -1979,7 +1962,9 @@ static int netdev_close(struct net_device *dev)
 	for (i = 0; i < RX_RING_SIZE; i++) {
 		np->rx_ring[i].rxaddr = cpu_to_dma(0xBADF00D0); /* An invalid address. */
 		if (np->rx_info[i].skb != NULL) {
-			pci_unmap_single(np->pci_dev, np->rx_info[i].mapping, np->rx_buf_sz, PCI_DMA_FROMDEVICE);
+			dma_unmap_single(&np->pci_dev->dev,
+					 np->rx_info[i].mapping,
+					 np->rx_buf_sz, DMA_FROM_DEVICE);
 			dev_kfree_skb(np->rx_info[i].skb);
 		}
 		np->rx_info[i].skb = NULL;
@@ -1989,9 +1974,8 @@ static int netdev_close(struct net_device *dev)
 		struct sk_buff *skb = np->tx_info[i].skb;
 		if (skb == NULL)
 			continue;
-		pci_unmap_single(np->pci_dev,
-				 np->tx_info[i].mapping,
-				 skb_first_frag_len(skb), PCI_DMA_TODEVICE);
+		dma_unmap_single(&np->pci_dev->dev, np->tx_info[i].mapping,
+				 skb_first_frag_len(skb), DMA_TO_DEVICE);
 		np->tx_info[i].mapping = 0;
 		dev_kfree_skb(skb);
 		np->tx_info[i].skb = NULL;
@@ -2000,28 +1984,21 @@ static int netdev_close(struct net_device *dev)
 	return 0;
 }
 
-#ifdef CONFIG_PM
-static int starfire_suspend(struct pci_dev *pdev, pm_message_t state)
+static int __maybe_unused starfire_suspend(struct device *dev_d)
 {
-	struct net_device *dev = pci_get_drvdata(pdev);
+	struct net_device *dev = dev_get_drvdata(dev_d);
 
 	if (netif_running(dev)) {
 		netif_device_detach(dev);
 		netdev_close(dev);
 	}
 
-	pci_save_state(pdev);
-	pci_set_power_state(pdev, pci_choose_state(pdev,state));
-
 	return 0;
 }
 
-static int starfire_resume(struct pci_dev *pdev)
+static int __maybe_unused starfire_resume(struct device *dev_d)
 {
-	struct net_device *dev = pci_get_drvdata(pdev);
-
-	pci_set_power_state(pdev, PCI_D0);
-	pci_restore_state(pdev);
+	struct net_device *dev = dev_get_drvdata(dev_d);
 
 	if (netif_running(dev)) {
 		netdev_open(dev);
@@ -2030,8 +2007,6 @@ static int starfire_resume(struct pci_dev *pdev)
 
 	return 0;
 }
-#endif /* CONFIG_PM */
-
 
 static void starfire_remove_one(struct pci_dev *pdev)
 {
@@ -2043,7 +2018,8 @@ static void starfire_remove_one(struct pci_dev *pdev)
 	unregister_netdev(dev);
 
 	if (np->queue_mem)
-		pci_free_consistent(pdev, np->queue_mem_size, np->queue_mem, np->queue_mem_dma);
+		dma_free_coherent(&pdev->dev, np->queue_mem_size,
+				  np->queue_mem, np->queue_mem_dma);
 
 
 	/* XXX: add wakeup code -- requires firmware for MagicPacket */
@@ -2056,15 +2032,13 @@ static void starfire_remove_one(struct pci_dev *pdev)
 	free_netdev(dev);			/* Will also free np!! */
 }
 
+static SIMPLE_DEV_PM_OPS(starfire_pm_ops, starfire_suspend, starfire_resume);
 
 static struct pci_driver starfire_driver = {
 	.name		= DRV_NAME,
 	.probe		= starfire_init_one,
 	.remove		= starfire_remove_one,
-#ifdef CONFIG_PM
-	.suspend	= starfire_suspend,
-	.resume		= starfire_resume,
-#endif /* CONFIG_PM */
+	.driver.pm	= &starfire_pm_ops,
 	.id_table	= starfire_pci_tbl,
 };
 
@@ -2073,8 +2047,6 @@ static int __init starfire_init (void)
 {
 /* when a module, this is printed whether or not devices are found in probe */
 #ifdef MODULE
-	printk(version);
-
 	printk(KERN_INFO DRV_NAME ": polling (NAPI) enabled\n");
 #endif
 
@@ -2092,11 +2064,3 @@ static void __exit starfire_cleanup (void)
 
 module_init(starfire_init);
 module_exit(starfire_cleanup);
-
-
-/*
- * Local variables:
- *  c-basic-offset: 8
- *  tab-width: 8
- * End:
- */
diff --git a/drivers/net/ethernet/adi/Kconfig b/drivers/net/ethernet/adi/Kconfig
index 98cc8f535021..760a9a60bc15 100644
--- a/drivers/net/ethernet/adi/Kconfig
+++ b/drivers/net/ethernet/adi/Kconfig
@@ -1,66 +1,29 @@
+# SPDX-License-Identifier: GPL-2.0 OR BSD-2-Clause
 #
-# Blackfin device configuration
+# Analog Devices device configuration
 #
 
-config NET_BFIN
-	bool "Blackfin devices"
-	depends on BF516 || BF518 || BF526 || BF527 || BF536 || BF537
-	---help---
+config NET_VENDOR_ADI
+	bool "Analog Devices devices"
+	default y
+	depends on SPI
+	help
 	  If you have a network (Ethernet) card belonging to this class, say Y.
 
-	  If unsure, say Y.
-
 	  Note that the answer to this question doesn't directly affect the
 	  kernel: saying N will just cause the configurator to skip all
-	  the remaining Blackfin card questions. If you say Y, you will be
-	  asked for your specific card in the following questions.
+	  the questions about ADI devices. If you say Y, you will be asked
+	  for your specific card in the following questions.
 
-if NET_BFIN
+if NET_VENDOR_ADI
 
-config BFIN_MAC
-	tristate "Blackfin on-chip MAC support"
-	depends on (BF516 || BF518 || BF526 || BF527 || BF536 || BF537)
-	select CRC32
-	select MII
+config ADIN1110
+	tristate "Analog Devices ADIN1110 MAC-PHY"
+	depends on SPI && NET_SWITCHDEV
+	select CRC8
 	select PHYLIB
-	select BFIN_MAC_USE_L1 if DMA_UNCACHED_NONE
-	---help---
-	  This is the driver for Blackfin on-chip mac device. Say Y if you want
-	  it compiled into the kernel. This driver is also available as a
-	  module ( = code which can be inserted in and removed from the running
-	  kernel whenever you want). The module will be called bfin_mac.
-
-config BFIN_MAC_USE_L1
-	bool "Use L1 memory for rx/tx packets"
-	depends on BFIN_MAC && (BF527 || BF537)
-	default y
-	---help---
-	  To get maximum network performance, you should use L1 memory as rx/tx
-	  buffers. Say N here if you want to reserve L1 memory for other uses.
-
-config BFIN_TX_DESC_NUM
-	int "Number of transmit buffer packets"
-	depends on BFIN_MAC
-	range 6 10 if BFIN_MAC_USE_L1
-	range 10 100
-	default "10"
-	---help---
-	  Set the number of buffer packets used in driver.
-
-config BFIN_RX_DESC_NUM
-	int "Number of receive buffer packets"
-	depends on BFIN_MAC
-	range 20 64
-	default "20"
-	---help---
-	  Set the number of buffer packets used in driver.
-
-config BFIN_MAC_USE_HWSTAMP
-	bool "Use IEEE 1588 hwstamp"
-	depends on BFIN_MAC && BF518
-	imply PTP_1588_CLOCK
-	default y
-	---help---
-	  To support the IEEE 1588 Precision Time Protocol (PTP), select y here
+	help
+	  Say yes here to build support for Analog Devices ADIN1110
+	  Low Power 10BASE-T1L Ethernet MAC-PHY.
 
-endif # NET_BFIN
+endif # NET_VENDOR_ADI
diff --git a/drivers/net/ethernet/adi/Makefile b/drivers/net/ethernet/adi/Makefile
index b1fbe195d0e8..d0383d94303c 100644
--- a/drivers/net/ethernet/adi/Makefile
+++ b/drivers/net/ethernet/adi/Makefile
@@ -1,5 +1,6 @@
+# SPDX-License-Identifier: GPL-2.0 OR BSD-2-Clause
 #
-# Makefile for the Blackfin device drivers.
+# Makefile for the Analog Devices network device drivers.
 #
 
-obj-$(CONFIG_BFIN_MAC) += bfin_mac.o
+obj-$(CONFIG_ADIN1110) += adin1110.o
diff --git a/drivers/net/ethernet/adi/adin1110.c b/drivers/net/ethernet/adi/adin1110.c
new file mode 100644
index 000000000000..30f9d271e595
--- /dev/null
+++ b/drivers/net/ethernet/adi/adin1110.c
@@ -0,0 +1,1737 @@
+// SPDX-License-Identifier: GPL-2.0 OR BSD-2-Clause
+/* ADIN1110 Low Power 10BASE-T1L Ethernet MAC-PHY
+ * ADIN2111 2-Port Ethernet Switch with Integrated 10BASE-T1L PHY
+ *
+ * Copyright 2021 Analog Devices Inc.
+ */
+
+#include <linux/bitfield.h>
+#include <linux/bits.h>
+#include <linux/cache.h>
+#include <linux/crc8.h>
+#include <linux/etherdevice.h>
+#include <linux/ethtool.h>
+#include <linux/gpio/consumer.h>
+#include <linux/if_bridge.h>
+#include <linux/interrupt.h>
+#include <linux/iopoll.h>
+#include <linux/kernel.h>
+#include <linux/mii.h>
+#include <linux/module.h>
+#include <linux/netdevice.h>
+#include <linux/regulator/consumer.h>
+#include <linux/phy.h>
+#include <linux/property.h>
+#include <linux/spi/spi.h>
+
+#include <net/switchdev.h>
+
+#include <linux/unaligned.h>
+
+#define ADIN1110_PHY_ID				0x1
+
+#define ADIN1110_RESET				0x03
+#define   ADIN1110_SWRESET			BIT(0)
+
+#define ADIN1110_CONFIG1			0x04
+#define   ADIN1110_CONFIG1_SYNC			BIT(15)
+
+#define ADIN1110_CONFIG2			0x06
+#define   ADIN2111_P2_FWD_UNK2HOST		BIT(12)
+#define   ADIN2111_PORT_CUT_THRU_EN		BIT(11)
+#define   ADIN1110_CRC_APPEND			BIT(5)
+#define   ADIN1110_FWD_UNK2HOST			BIT(2)
+
+#define ADIN1110_STATUS0			0x08
+
+#define ADIN1110_STATUS1			0x09
+#define   ADIN2111_P2_RX_RDY			BIT(17)
+#define   ADIN1110_SPI_ERR			BIT(10)
+#define   ADIN1110_RX_RDY			BIT(4)
+
+#define ADIN1110_IMASK1				0x0D
+#define   ADIN2111_RX_RDY_IRQ			BIT(17)
+#define   ADIN1110_SPI_ERR_IRQ			BIT(10)
+#define   ADIN1110_RX_RDY_IRQ			BIT(4)
+#define   ADIN1110_TX_RDY_IRQ			BIT(3)
+
+#define ADIN1110_MDIOACC			0x20
+#define   ADIN1110_MDIO_TRDONE			BIT(31)
+#define   ADIN1110_MDIO_ST			GENMASK(29, 28)
+#define   ADIN1110_MDIO_OP			GENMASK(27, 26)
+#define   ADIN1110_MDIO_PRTAD			GENMASK(25, 21)
+#define   ADIN1110_MDIO_DEVAD			GENMASK(20, 16)
+#define   ADIN1110_MDIO_DATA			GENMASK(15, 0)
+
+#define ADIN1110_TX_FSIZE			0x30
+#define ADIN1110_TX				0x31
+#define ADIN1110_TX_SPACE			0x32
+
+#define ADIN1110_MAC_ADDR_FILTER_UPR		0x50
+#define   ADIN2111_MAC_ADDR_APPLY2PORT2		BIT(31)
+#define   ADIN1110_MAC_ADDR_APPLY2PORT		BIT(30)
+#define   ADIN2111_MAC_ADDR_TO_OTHER_PORT	BIT(17)
+#define   ADIN1110_MAC_ADDR_TO_HOST		BIT(16)
+
+#define ADIN1110_MAC_ADDR_FILTER_LWR		0x51
+
+#define ADIN1110_MAC_ADDR_MASK_UPR		0x70
+#define ADIN1110_MAC_ADDR_MASK_LWR		0x71
+
+#define ADIN1110_RX_FSIZE			0x90
+#define ADIN1110_RX				0x91
+
+#define ADIN2111_RX_P2_FSIZE			0xC0
+#define ADIN2111_RX_P2				0xC1
+
+#define ADIN1110_CLEAR_STATUS0			0xFFF
+
+/* MDIO_OP codes */
+#define ADIN1110_MDIO_OP_WR			0x1
+#define ADIN1110_MDIO_OP_RD			0x3
+
+#define ADIN1110_CD				BIT(7)
+#define ADIN1110_WRITE				BIT(5)
+
+#define ADIN1110_MAX_BUFF			2048
+#define ADIN1110_MAX_FRAMES_READ		64
+#define ADIN1110_WR_HEADER_LEN			2
+#define ADIN1110_FRAME_HEADER_LEN		2
+#define ADIN1110_INTERNAL_SIZE_HEADER_LEN	2
+#define ADIN1110_RD_HEADER_LEN			3
+#define ADIN1110_REG_LEN			4
+#define ADIN1110_FEC_LEN			4
+
+#define ADIN1110_PHY_ID_VAL			0x0283BC91
+#define ADIN2111_PHY_ID_VAL			0x0283BCA1
+
+#define ADIN_MAC_MAX_PORTS			2
+#define ADIN_MAC_MAX_ADDR_SLOTS			16
+
+#define ADIN_MAC_MULTICAST_ADDR_SLOT		0
+#define ADIN_MAC_BROADCAST_ADDR_SLOT		1
+#define ADIN_MAC_P1_ADDR_SLOT			2
+#define ADIN_MAC_P2_ADDR_SLOT			3
+#define ADIN_MAC_FDB_ADDR_SLOT			4
+
+DECLARE_CRC8_TABLE(adin1110_crc_table);
+
+enum adin1110_chips_id {
+	ADIN1110_MAC = 0,
+	ADIN2111_MAC,
+};
+
+struct adin1110_cfg {
+	enum adin1110_chips_id	id;
+	char			name[MDIO_NAME_SIZE];
+	u32			phy_ids[PHY_MAX_ADDR];
+	u32			ports_nr;
+	u32			phy_id_val;
+};
+
+struct adin1110_port_priv {
+	struct adin1110_priv		*priv;
+	struct net_device		*netdev;
+	struct net_device		*bridge;
+	struct phy_device		*phydev;
+	struct work_struct		tx_work;
+	u64				rx_packets;
+	u64				tx_packets;
+	u64				rx_bytes;
+	u64				tx_bytes;
+	struct work_struct		rx_mode_work;
+	u32				flags;
+	struct sk_buff_head		txq;
+	u32				nr;
+	u32				state;
+	struct adin1110_cfg		*cfg;
+};
+
+struct adin1110_priv {
+	struct mutex			lock; /* protect spi */
+	spinlock_t			state_lock; /* protect RX mode */
+	struct mii_bus			*mii_bus;
+	struct spi_device		*spidev;
+	bool				append_crc;
+	struct adin1110_cfg		*cfg;
+	u32				tx_space;
+	u32				irq_mask;
+	bool				forwarding;
+	int				irq;
+	struct adin1110_port_priv	*ports[ADIN_MAC_MAX_PORTS];
+	char				mii_bus_name[MII_BUS_ID_SIZE];
+	u8				data[ADIN1110_MAX_BUFF] ____cacheline_aligned;
+};
+
+struct adin1110_switchdev_event_work {
+	struct work_struct work;
+	struct switchdev_notifier_fdb_info fdb_info;
+	struct adin1110_port_priv *port_priv;
+	unsigned long event;
+};
+
+static struct adin1110_cfg adin1110_cfgs[] = {
+	{
+		.id = ADIN1110_MAC,
+		.name = "adin1110",
+		.phy_ids = {1},
+		.ports_nr = 1,
+		.phy_id_val = ADIN1110_PHY_ID_VAL,
+	},
+	{
+		.id = ADIN2111_MAC,
+		.name = "adin2111",
+		.phy_ids = {1, 2},
+		.ports_nr = 2,
+		.phy_id_val = ADIN2111_PHY_ID_VAL,
+	},
+};
+
+static u8 adin1110_crc_data(u8 *data, u32 len)
+{
+	return crc8(adin1110_crc_table, data, len, 0);
+}
+
+static int adin1110_read_reg(struct adin1110_priv *priv, u16 reg, u32 *val)
+{
+	u32 header_len = ADIN1110_RD_HEADER_LEN;
+	u32 read_len = ADIN1110_REG_LEN;
+	struct spi_transfer t = {0};
+	int ret;
+
+	priv->data[0] = ADIN1110_CD | FIELD_GET(GENMASK(12, 8), reg);
+	priv->data[1] = FIELD_GET(GENMASK(7, 0), reg);
+	priv->data[2] = 0x00;
+
+	if (priv->append_crc) {
+		priv->data[2] = adin1110_crc_data(&priv->data[0], 2);
+		priv->data[3] = 0x00;
+		header_len++;
+	}
+
+	if (priv->append_crc)
+		read_len++;
+
+	memset(&priv->data[header_len], 0, read_len);
+	t.tx_buf = &priv->data[0];
+	t.rx_buf = &priv->data[0];
+	t.len = read_len + header_len;
+
+	ret = spi_sync_transfer(priv->spidev, &t, 1);
+	if (ret)
+		return ret;
+
+	if (priv->append_crc) {
+		u8 recv_crc;
+		u8 crc;
+
+		crc = adin1110_crc_data(&priv->data[header_len],
+					ADIN1110_REG_LEN);
+		recv_crc = priv->data[header_len + ADIN1110_REG_LEN];
+
+		if (crc != recv_crc) {
+			dev_err_ratelimited(&priv->spidev->dev, "CRC error.");
+			return -EBADMSG;
+		}
+	}
+
+	*val = get_unaligned_be32(&priv->data[header_len]);
+
+	return ret;
+}
+
+static int adin1110_write_reg(struct adin1110_priv *priv, u16 reg, u32 val)
+{
+	u32 header_len = ADIN1110_WR_HEADER_LEN;
+	u32 write_len = ADIN1110_REG_LEN;
+
+	priv->data[0] = ADIN1110_CD | ADIN1110_WRITE | FIELD_GET(GENMASK(12, 8), reg);
+	priv->data[1] = FIELD_GET(GENMASK(7, 0), reg);
+
+	if (priv->append_crc) {
+		priv->data[2] = adin1110_crc_data(&priv->data[0], header_len);
+		header_len++;
+	}
+
+	put_unaligned_be32(val, &priv->data[header_len]);
+	if (priv->append_crc) {
+		priv->data[header_len + write_len] = adin1110_crc_data(&priv->data[header_len],
+								       write_len);
+		write_len++;
+	}
+
+	return spi_write(priv->spidev, &priv->data[0], header_len + write_len);
+}
+
+static int adin1110_set_bits(struct adin1110_priv *priv, u16 reg,
+			     unsigned long mask, unsigned long val)
+{
+	u32 write_val;
+	int ret;
+
+	ret = adin1110_read_reg(priv, reg, &write_val);
+	if (ret < 0)
+		return ret;
+
+	set_mask_bits(&write_val, mask, val);
+
+	return adin1110_write_reg(priv, reg, write_val);
+}
+
+static int adin1110_round_len(int len)
+{
+	/* can read/write only mutiples of 4 bytes of payload */
+	len = ALIGN(len, 4);
+
+	/* NOTE: ADIN1110_WR_HEADER_LEN should be used for write ops. */
+	if (len + ADIN1110_RD_HEADER_LEN > ADIN1110_MAX_BUFF)
+		return -EINVAL;
+
+	return len;
+}
+
+static int adin1110_read_fifo(struct adin1110_port_priv *port_priv)
+{
+	struct adin1110_priv *priv = port_priv->priv;
+	u32 header_len = ADIN1110_RD_HEADER_LEN;
+	struct spi_transfer t = {0};
+	u32 frame_size_no_fcs;
+	struct sk_buff *rxb;
+	u32 frame_size;
+	int round_len;
+	u16 reg;
+	int ret;
+
+	if (!port_priv->nr) {
+		reg = ADIN1110_RX;
+		ret = adin1110_read_reg(priv, ADIN1110_RX_FSIZE, &frame_size);
+	} else {
+		reg = ADIN2111_RX_P2;
+		ret = adin1110_read_reg(priv, ADIN2111_RX_P2_FSIZE,
+					&frame_size);
+	}
+
+	if (ret < 0)
+		return ret;
+
+	/* The read frame size includes the extra 2 bytes
+	 * from the  ADIN1110 frame header.
+	 */
+	if (frame_size < ADIN1110_FRAME_HEADER_LEN + ADIN1110_FEC_LEN)
+		return -EINVAL;
+
+	round_len = adin1110_round_len(frame_size);
+	if (round_len < 0)
+		return -EINVAL;
+
+	frame_size_no_fcs = frame_size - ADIN1110_FRAME_HEADER_LEN - ADIN1110_FEC_LEN;
+	memset(priv->data, 0, ADIN1110_RD_HEADER_LEN);
+
+	priv->data[0] = ADIN1110_CD | FIELD_GET(GENMASK(12, 8), reg);
+	priv->data[1] = FIELD_GET(GENMASK(7, 0), reg);
+
+	if (priv->append_crc) {
+		priv->data[2] = adin1110_crc_data(&priv->data[0], 2);
+		header_len++;
+	}
+
+	rxb = netdev_alloc_skb(port_priv->netdev, round_len + header_len);
+	if (!rxb)
+		return -ENOMEM;
+
+	skb_put(rxb, frame_size_no_fcs + header_len + ADIN1110_FRAME_HEADER_LEN);
+
+	t.tx_buf = &priv->data[0];
+	t.rx_buf = &rxb->data[0];
+	t.len = header_len + round_len;
+
+	ret = spi_sync_transfer(priv->spidev, &t, 1);
+	if (ret) {
+		kfree_skb(rxb);
+		return ret;
+	}
+
+	skb_pull(rxb, header_len + ADIN1110_FRAME_HEADER_LEN);
+	rxb->protocol = eth_type_trans(rxb, port_priv->netdev);
+
+	if ((port_priv->flags & IFF_ALLMULTI && rxb->pkt_type == PACKET_MULTICAST) ||
+	    (port_priv->flags & IFF_BROADCAST && rxb->pkt_type == PACKET_BROADCAST))
+		rxb->offload_fwd_mark = port_priv->priv->forwarding;
+
+	netif_rx(rxb);
+
+	port_priv->rx_bytes += frame_size - ADIN1110_FRAME_HEADER_LEN;
+	port_priv->rx_packets++;
+
+	return 0;
+}
+
+static int adin1110_write_fifo(struct adin1110_port_priv *port_priv,
+			       struct sk_buff *txb)
+{
+	struct adin1110_priv *priv = port_priv->priv;
+	u32 header_len = ADIN1110_WR_HEADER_LEN;
+	__be16 frame_header;
+	int padding = 0;
+	int padded_len;
+	int round_len;
+	int ret;
+
+	/* Pad frame to 64 byte length,
+	 * MAC nor PHY will otherwise add the
+	 * required padding.
+	 * The FEC will be added by the MAC internally.
+	 */
+	if (txb->len + ADIN1110_FEC_LEN < 64)
+		padding = 64 - (txb->len + ADIN1110_FEC_LEN);
+
+	padded_len = txb->len + padding + ADIN1110_FRAME_HEADER_LEN;
+
+	round_len = adin1110_round_len(padded_len);
+	if (round_len < 0)
+		return round_len;
+
+	ret = adin1110_write_reg(priv, ADIN1110_TX_FSIZE, padded_len);
+	if (ret < 0)
+		return ret;
+
+	memset(priv->data, 0, round_len + ADIN1110_WR_HEADER_LEN);
+
+	priv->data[0] = ADIN1110_CD | ADIN1110_WRITE;
+	priv->data[0] |= FIELD_GET(GENMASK(12, 8), ADIN1110_TX);
+	priv->data[1] = FIELD_GET(GENMASK(7, 0), ADIN1110_TX);
+	if (priv->append_crc) {
+		priv->data[2] = adin1110_crc_data(&priv->data[0], 2);
+		header_len++;
+	}
+
+	/* mention the port on which to send the frame in the frame header */
+	frame_header = cpu_to_be16(port_priv->nr);
+	memcpy(&priv->data[header_len], &frame_header,
+	       ADIN1110_FRAME_HEADER_LEN);
+
+	memcpy(&priv->data[header_len + ADIN1110_FRAME_HEADER_LEN],
+	       txb->data, txb->len);
+
+	ret = spi_write(priv->spidev, &priv->data[0], round_len + header_len);
+	if (ret < 0)
+		return ret;
+
+	port_priv->tx_bytes += txb->len;
+	port_priv->tx_packets++;
+
+	return 0;
+}
+
+static int adin1110_read_mdio_acc(struct adin1110_priv *priv)
+{
+	u32 val;
+	int ret;
+
+	mutex_lock(&priv->lock);
+	ret = adin1110_read_reg(priv, ADIN1110_MDIOACC, &val);
+	mutex_unlock(&priv->lock);
+	if (ret < 0)
+		return 0;
+
+	return val;
+}
+
+static int adin1110_mdio_read(struct mii_bus *bus, int phy_id, int reg)
+{
+	struct adin1110_priv *priv = bus->priv;
+	u32 val = 0;
+	int ret;
+
+	if (mdio_phy_id_is_c45(phy_id))
+		return -EOPNOTSUPP;
+
+	val |= FIELD_PREP(ADIN1110_MDIO_OP, ADIN1110_MDIO_OP_RD);
+	val |= FIELD_PREP(ADIN1110_MDIO_ST, 0x1);
+	val |= FIELD_PREP(ADIN1110_MDIO_PRTAD, phy_id);
+	val |= FIELD_PREP(ADIN1110_MDIO_DEVAD, reg);
+
+	/* write the clause 22 read command to the chip */
+	mutex_lock(&priv->lock);
+	ret = adin1110_write_reg(priv, ADIN1110_MDIOACC, val);
+	mutex_unlock(&priv->lock);
+	if (ret < 0)
+		return ret;
+
+	/* ADIN1110_MDIO_TRDONE BIT of the ADIN1110_MDIOACC
+	 * register is set when the read is done.
+	 * After the transaction is done, ADIN1110_MDIO_DATA
+	 * bitfield of ADIN1110_MDIOACC register will contain
+	 * the requested register value.
+	 */
+	ret = readx_poll_timeout_atomic(adin1110_read_mdio_acc, priv, val,
+					(val & ADIN1110_MDIO_TRDONE),
+					100, 30000);
+	if (ret < 0)
+		return ret;
+
+	return (val & ADIN1110_MDIO_DATA);
+}
+
+static int adin1110_mdio_write(struct mii_bus *bus, int phy_id,
+			       int reg, u16 reg_val)
+{
+	struct adin1110_priv *priv = bus->priv;
+	u32 val = 0;
+	int ret;
+
+	if (mdio_phy_id_is_c45(phy_id))
+		return -EOPNOTSUPP;
+
+	val |= FIELD_PREP(ADIN1110_MDIO_OP, ADIN1110_MDIO_OP_WR);
+	val |= FIELD_PREP(ADIN1110_MDIO_ST, 0x1);
+	val |= FIELD_PREP(ADIN1110_MDIO_PRTAD, phy_id);
+	val |= FIELD_PREP(ADIN1110_MDIO_DEVAD, reg);
+	val |= FIELD_PREP(ADIN1110_MDIO_DATA, reg_val);
+
+	/* write the clause 22 write command to the chip */
+	mutex_lock(&priv->lock);
+	ret = adin1110_write_reg(priv, ADIN1110_MDIOACC, val);
+	mutex_unlock(&priv->lock);
+	if (ret < 0)
+		return ret;
+
+	return readx_poll_timeout_atomic(adin1110_read_mdio_acc, priv, val,
+					 (val & ADIN1110_MDIO_TRDONE),
+					 100, 30000);
+}
+
+/* ADIN1110 MAC-PHY contains an ADIN1100 PHY.
+ * ADIN2111 MAC-PHY contains two ADIN1100 PHYs.
+ * By registering a new MDIO bus we allow the PAL to discover
+ * the encapsulated PHY and probe the ADIN1100 driver.
+ */
+static int adin1110_register_mdiobus(struct adin1110_priv *priv,
+				     struct device *dev)
+{
+	struct mii_bus *mii_bus;
+	int ret;
+
+	mii_bus = devm_mdiobus_alloc(dev);
+	if (!mii_bus)
+		return -ENOMEM;
+
+	snprintf(priv->mii_bus_name, MII_BUS_ID_SIZE, "%s-%u",
+		 priv->cfg->name, spi_get_chipselect(priv->spidev, 0));
+
+	mii_bus->name = priv->mii_bus_name;
+	mii_bus->read = adin1110_mdio_read;
+	mii_bus->write = adin1110_mdio_write;
+	mii_bus->priv = priv;
+	mii_bus->parent = dev;
+	mii_bus->phy_mask = ~((u32)GENMASK(2, 0));
+	snprintf(mii_bus->id, MII_BUS_ID_SIZE, "%s", dev_name(dev));
+
+	ret = devm_mdiobus_register(dev, mii_bus);
+	if (ret)
+		return ret;
+
+	priv->mii_bus = mii_bus;
+
+	return 0;
+}
+
+static bool adin1110_port_rx_ready(struct adin1110_port_priv *port_priv,
+				   u32 status)
+{
+	if (!netif_oper_up(port_priv->netdev))
+		return false;
+
+	if (!port_priv->nr)
+		return !!(status & ADIN1110_RX_RDY);
+	else
+		return !!(status & ADIN2111_P2_RX_RDY);
+}
+
+static void adin1110_read_frames(struct adin1110_port_priv *port_priv,
+				 unsigned int budget)
+{
+	struct adin1110_priv *priv = port_priv->priv;
+	u32 status1;
+	int ret;
+
+	while (budget) {
+		ret = adin1110_read_reg(priv, ADIN1110_STATUS1, &status1);
+		if (ret < 0)
+			return;
+
+		if (!adin1110_port_rx_ready(port_priv, status1))
+			break;
+
+		ret = adin1110_read_fifo(port_priv);
+		if (ret < 0)
+			return;
+
+		budget--;
+	}
+}
+
+static void adin1110_wake_queues(struct adin1110_priv *priv)
+{
+	int i;
+
+	for (i = 0; i < priv->cfg->ports_nr; i++)
+		netif_wake_queue(priv->ports[i]->netdev);
+}
+
+static irqreturn_t adin1110_irq(int irq, void *p)
+{
+	struct adin1110_priv *priv = p;
+	u32 status1;
+	u32 val;
+	int ret;
+	int i;
+
+	mutex_lock(&priv->lock);
+
+	ret = adin1110_read_reg(priv, ADIN1110_STATUS1, &status1);
+	if (ret < 0)
+		goto out;
+
+	if (priv->append_crc && (status1 & ADIN1110_SPI_ERR))
+		dev_warn_ratelimited(&priv->spidev->dev,
+				     "SPI CRC error on write.\n");
+
+	ret = adin1110_read_reg(priv, ADIN1110_TX_SPACE, &val);
+	if (ret < 0)
+		goto out;
+
+	/* TX FIFO space is expressed in half-words */
+	priv->tx_space = 2 * val;
+
+	for (i = 0; i < priv->cfg->ports_nr; i++) {
+		if (adin1110_port_rx_ready(priv->ports[i], status1))
+			adin1110_read_frames(priv->ports[i],
+					     ADIN1110_MAX_FRAMES_READ);
+	}
+
+	/* clear IRQ sources */
+	adin1110_write_reg(priv, ADIN1110_STATUS0, ADIN1110_CLEAR_STATUS0);
+	adin1110_write_reg(priv, ADIN1110_STATUS1, priv->irq_mask);
+
+out:
+	mutex_unlock(&priv->lock);
+
+	if (priv->tx_space > 0 && ret >= 0)
+		adin1110_wake_queues(priv);
+
+	return IRQ_HANDLED;
+}
+
+/* ADIN1110 can filter up to 16 MAC addresses, mac_nr here is the slot used */
+static int adin1110_write_mac_address(struct adin1110_port_priv *port_priv,
+				      int mac_nr, const u8 *addr,
+				      u8 *mask, u32 port_rules)
+{
+	struct adin1110_priv *priv = port_priv->priv;
+	u32 offset = mac_nr * 2;
+	u32 port_rules_mask;
+	int ret;
+	u32 val;
+
+	if (!port_priv->nr)
+		port_rules_mask = ADIN1110_MAC_ADDR_APPLY2PORT;
+	else
+		port_rules_mask = ADIN2111_MAC_ADDR_APPLY2PORT2;
+
+	if (port_rules & port_rules_mask)
+		port_rules_mask |= ADIN1110_MAC_ADDR_TO_HOST | ADIN2111_MAC_ADDR_TO_OTHER_PORT;
+
+	port_rules_mask |= GENMASK(15, 0);
+	val = port_rules | get_unaligned_be16(&addr[0]);
+	ret = adin1110_set_bits(priv, ADIN1110_MAC_ADDR_FILTER_UPR + offset,
+				port_rules_mask, val);
+	if (ret < 0)
+		return ret;
+
+	val = get_unaligned_be32(&addr[2]);
+	ret =  adin1110_write_reg(priv,
+				  ADIN1110_MAC_ADDR_FILTER_LWR + offset, val);
+	if (ret < 0)
+		return ret;
+
+	/* Only the first two MAC address slots support masking. */
+	if (mac_nr < ADIN_MAC_P1_ADDR_SLOT) {
+		val = get_unaligned_be16(&mask[0]);
+		ret = adin1110_write_reg(priv,
+					 ADIN1110_MAC_ADDR_MASK_UPR + offset,
+					 val);
+		if (ret < 0)
+			return ret;
+
+		val = get_unaligned_be32(&mask[2]);
+		return adin1110_write_reg(priv,
+					  ADIN1110_MAC_ADDR_MASK_LWR + offset,
+					  val);
+	}
+
+	return 0;
+}
+
+static int adin1110_clear_mac_address(struct adin1110_priv *priv, int mac_nr)
+{
+	u32 offset = mac_nr * 2;
+	int ret;
+
+	ret = adin1110_write_reg(priv, ADIN1110_MAC_ADDR_FILTER_UPR + offset, 0);
+	if (ret < 0)
+		return ret;
+
+	ret =  adin1110_write_reg(priv, ADIN1110_MAC_ADDR_FILTER_LWR + offset, 0);
+	if (ret < 0)
+		return ret;
+
+	/* only the first two MAC address slots are maskable */
+	if (mac_nr <= 1) {
+		ret = adin1110_write_reg(priv, ADIN1110_MAC_ADDR_MASK_UPR + offset, 0);
+		if (ret < 0)
+			return ret;
+
+		ret = adin1110_write_reg(priv, ADIN1110_MAC_ADDR_MASK_LWR + offset, 0);
+	}
+
+	return ret;
+}
+
+static u32 adin1110_port_rules(struct adin1110_port_priv *port_priv,
+			       bool fw_to_host,
+			       bool fw_to_other_port)
+{
+	u32 port_rules = 0;
+
+	if (!port_priv->nr)
+		port_rules |= ADIN1110_MAC_ADDR_APPLY2PORT;
+	else
+		port_rules |= ADIN2111_MAC_ADDR_APPLY2PORT2;
+
+	if (fw_to_host)
+		port_rules |= ADIN1110_MAC_ADDR_TO_HOST;
+
+	if (fw_to_other_port && port_priv->priv->forwarding)
+		port_rules |= ADIN2111_MAC_ADDR_TO_OTHER_PORT;
+
+	return port_rules;
+}
+
+static int adin1110_multicast_filter(struct adin1110_port_priv *port_priv,
+				     int mac_nr, bool accept_multicast)
+{
+	u8 mask[ETH_ALEN] = {0};
+	u8 mac[ETH_ALEN] = {0};
+	u32 port_rules = 0;
+
+	mask[0] = BIT(0);
+	mac[0] = BIT(0);
+
+	if (accept_multicast && port_priv->state == BR_STATE_FORWARDING)
+		port_rules = adin1110_port_rules(port_priv, true, true);
+
+	return adin1110_write_mac_address(port_priv, mac_nr, mac,
+					  mask, port_rules);
+}
+
+static int adin1110_broadcasts_filter(struct adin1110_port_priv *port_priv,
+				      int mac_nr, bool accept_broadcast)
+{
+	u32 port_rules = 0;
+	u8 mask[ETH_ALEN];
+
+	eth_broadcast_addr(mask);
+
+	if (accept_broadcast && port_priv->state == BR_STATE_FORWARDING)
+		port_rules = adin1110_port_rules(port_priv, true, true);
+
+	return adin1110_write_mac_address(port_priv, mac_nr, mask,
+					  mask, port_rules);
+}
+
+static int adin1110_set_mac_address(struct net_device *netdev,
+				    const unsigned char *dev_addr)
+{
+	struct adin1110_port_priv *port_priv = netdev_priv(netdev);
+	u8 mask[ETH_ALEN];
+	u32 port_rules;
+	u32 mac_slot;
+
+	if (!is_valid_ether_addr(dev_addr))
+		return -EADDRNOTAVAIL;
+
+	eth_hw_addr_set(netdev, dev_addr);
+	eth_broadcast_addr(mask);
+
+	mac_slot = (!port_priv->nr) ?  ADIN_MAC_P1_ADDR_SLOT : ADIN_MAC_P2_ADDR_SLOT;
+	port_rules = adin1110_port_rules(port_priv, true, false);
+
+	return adin1110_write_mac_address(port_priv, mac_slot, netdev->dev_addr,
+					  mask, port_rules);
+}
+
+static int adin1110_ndo_set_mac_address(struct net_device *netdev, void *addr)
+{
+	struct sockaddr *sa = addr;
+	int ret;
+
+	ret = eth_prepare_mac_addr_change(netdev, addr);
+	if (ret < 0)
+		return ret;
+
+	return adin1110_set_mac_address(netdev, sa->sa_data);
+}
+
+static int adin1110_ioctl(struct net_device *netdev, struct ifreq *rq, int cmd)
+{
+	if (!netif_running(netdev))
+		return -EINVAL;
+
+	return phy_do_ioctl(netdev, rq, cmd);
+}
+
+static int adin1110_set_promisc_mode(struct adin1110_port_priv *port_priv,
+				     bool promisc)
+{
+	struct adin1110_priv *priv = port_priv->priv;
+	u32 mask;
+
+	if (port_priv->state != BR_STATE_FORWARDING)
+		promisc = false;
+
+	if (!port_priv->nr)
+		mask = ADIN1110_FWD_UNK2HOST;
+	else
+		mask = ADIN2111_P2_FWD_UNK2HOST;
+
+	return adin1110_set_bits(priv, ADIN1110_CONFIG2,
+				 mask, promisc ? mask : 0);
+}
+
+static int adin1110_setup_rx_mode(struct adin1110_port_priv *port_priv)
+{
+	int ret;
+
+	ret = adin1110_set_promisc_mode(port_priv,
+					!!(port_priv->flags & IFF_PROMISC));
+	if (ret < 0)
+		return ret;
+
+	ret = adin1110_multicast_filter(port_priv, ADIN_MAC_MULTICAST_ADDR_SLOT,
+					!!(port_priv->flags & IFF_ALLMULTI));
+	if (ret < 0)
+		return ret;
+
+	ret = adin1110_broadcasts_filter(port_priv,
+					 ADIN_MAC_BROADCAST_ADDR_SLOT,
+					 !!(port_priv->flags & IFF_BROADCAST));
+	if (ret < 0)
+		return ret;
+
+	return adin1110_set_bits(port_priv->priv, ADIN1110_CONFIG1,
+				 ADIN1110_CONFIG1_SYNC, ADIN1110_CONFIG1_SYNC);
+}
+
+static bool adin1110_can_offload_forwarding(struct adin1110_priv *priv)
+{
+	int i;
+
+	if (priv->cfg->id != ADIN2111_MAC)
+		return false;
+
+	/* Can't enable forwarding if ports do not belong to the same bridge */
+	if (priv->ports[0]->bridge != priv->ports[1]->bridge || !priv->ports[0]->bridge)
+		return false;
+
+	/* Can't enable forwarding if there is a port
+	 * that has been blocked by STP.
+	 */
+	for (i = 0; i < priv->cfg->ports_nr; i++) {
+		if (priv->ports[i]->state != BR_STATE_FORWARDING)
+			return false;
+	}
+
+	return true;
+}
+
+static void adin1110_rx_mode_work(struct work_struct *work)
+{
+	struct adin1110_port_priv *port_priv;
+	struct adin1110_priv *priv;
+
+	port_priv = container_of(work, struct adin1110_port_priv, rx_mode_work);
+	priv = port_priv->priv;
+
+	mutex_lock(&priv->lock);
+	adin1110_setup_rx_mode(port_priv);
+	mutex_unlock(&priv->lock);
+}
+
+static void adin1110_set_rx_mode(struct net_device *dev)
+{
+	struct adin1110_port_priv *port_priv = netdev_priv(dev);
+	struct adin1110_priv *priv = port_priv->priv;
+
+	spin_lock(&priv->state_lock);
+
+	port_priv->flags = dev->flags;
+	schedule_work(&port_priv->rx_mode_work);
+
+	spin_unlock(&priv->state_lock);
+}
+
+static int adin1110_net_open(struct net_device *net_dev)
+{
+	struct adin1110_port_priv *port_priv = netdev_priv(net_dev);
+	struct adin1110_priv *priv = port_priv->priv;
+	u32 val;
+	int ret;
+
+	mutex_lock(&priv->lock);
+
+	/* Configure MAC to compute and append the FCS itself. */
+	ret = adin1110_write_reg(priv, ADIN1110_CONFIG2, ADIN1110_CRC_APPEND);
+	if (ret < 0)
+		goto out;
+
+	val = ADIN1110_TX_RDY_IRQ | ADIN1110_RX_RDY_IRQ | ADIN1110_SPI_ERR_IRQ;
+	if (priv->cfg->id == ADIN2111_MAC)
+		val |= ADIN2111_RX_RDY_IRQ;
+
+	priv->irq_mask = val;
+	ret = adin1110_write_reg(priv, ADIN1110_IMASK1, ~val);
+	if (ret < 0) {
+		netdev_err(net_dev, "Failed to enable chip IRQs: %d\n", ret);
+		goto out;
+	}
+
+	ret = adin1110_read_reg(priv, ADIN1110_TX_SPACE, &val);
+	if (ret < 0) {
+		netdev_err(net_dev, "Failed to read TX FIFO space: %d\n", ret);
+		goto out;
+	}
+
+	priv->tx_space = 2 * val;
+
+	port_priv->state = BR_STATE_FORWARDING;
+	ret = adin1110_set_mac_address(net_dev, net_dev->dev_addr);
+	if (ret < 0) {
+		netdev_err(net_dev, "Could not set MAC address: %pM, %d\n",
+			   net_dev->dev_addr, ret);
+		goto out;
+	}
+
+	ret = adin1110_set_bits(priv, ADIN1110_CONFIG1, ADIN1110_CONFIG1_SYNC,
+				ADIN1110_CONFIG1_SYNC);
+
+out:
+	mutex_unlock(&priv->lock);
+
+	if (ret < 0)
+		return ret;
+
+	phy_start(port_priv->phydev);
+
+	netif_start_queue(net_dev);
+
+	return 0;
+}
+
+static int adin1110_net_stop(struct net_device *net_dev)
+{
+	struct adin1110_port_priv *port_priv = netdev_priv(net_dev);
+	struct adin1110_priv *priv = port_priv->priv;
+	u32 mask;
+	int ret;
+
+	mask = !port_priv->nr ? ADIN2111_RX_RDY_IRQ : ADIN1110_RX_RDY_IRQ;
+
+	/* Disable RX RDY IRQs */
+	mutex_lock(&priv->lock);
+	ret = adin1110_set_bits(priv, ADIN1110_IMASK1, mask, mask);
+	mutex_unlock(&priv->lock);
+	if (ret < 0)
+		return ret;
+
+	netif_stop_queue(port_priv->netdev);
+	flush_work(&port_priv->tx_work);
+	phy_stop(port_priv->phydev);
+
+	return 0;
+}
+
+static void adin1110_tx_work(struct work_struct *work)
+{
+	struct adin1110_port_priv *port_priv;
+	struct adin1110_priv *priv;
+	struct sk_buff *txb;
+	int ret;
+
+	port_priv = container_of(work, struct adin1110_port_priv, tx_work);
+	priv = port_priv->priv;
+
+	mutex_lock(&priv->lock);
+
+	while ((txb = skb_dequeue(&port_priv->txq))) {
+		ret = adin1110_write_fifo(port_priv, txb);
+		if (ret < 0)
+			dev_err_ratelimited(&priv->spidev->dev,
+					    "Frame write error: %d\n", ret);
+
+		dev_kfree_skb(txb);
+	}
+
+	mutex_unlock(&priv->lock);
+}
+
+static netdev_tx_t adin1110_start_xmit(struct sk_buff *skb, struct net_device *dev)
+{
+	struct adin1110_port_priv *port_priv = netdev_priv(dev);
+	struct adin1110_priv *priv = port_priv->priv;
+	netdev_tx_t netdev_ret = NETDEV_TX_OK;
+	u32 tx_space_needed;
+
+	tx_space_needed = skb->len + ADIN1110_FRAME_HEADER_LEN + ADIN1110_INTERNAL_SIZE_HEADER_LEN;
+	if (tx_space_needed > priv->tx_space) {
+		netif_stop_queue(dev);
+		netdev_ret = NETDEV_TX_BUSY;
+	} else {
+		priv->tx_space -= tx_space_needed;
+		skb_queue_tail(&port_priv->txq, skb);
+	}
+
+	schedule_work(&port_priv->tx_work);
+
+	return netdev_ret;
+}
+
+static void adin1110_ndo_get_stats64(struct net_device *dev,
+				     struct rtnl_link_stats64 *storage)
+{
+	struct adin1110_port_priv *port_priv = netdev_priv(dev);
+
+	storage->rx_packets = port_priv->rx_packets;
+	storage->tx_packets = port_priv->tx_packets;
+
+	storage->rx_bytes = port_priv->rx_bytes;
+	storage->tx_bytes = port_priv->tx_bytes;
+}
+
+static int adin1110_port_get_port_parent_id(struct net_device *dev,
+					    struct netdev_phys_item_id *ppid)
+{
+	struct adin1110_port_priv *port_priv = netdev_priv(dev);
+	struct adin1110_priv *priv = port_priv->priv;
+
+	ppid->id_len = strnlen(priv->mii_bus_name, MAX_PHYS_ITEM_ID_LEN);
+	memcpy(ppid->id, priv->mii_bus_name, ppid->id_len);
+
+	return 0;
+}
+
+static int adin1110_ndo_get_phys_port_name(struct net_device *dev,
+					   char *name, size_t len)
+{
+	struct adin1110_port_priv *port_priv = netdev_priv(dev);
+	int err;
+
+	err = snprintf(name, len, "p%d", port_priv->nr);
+	if (err >= len)
+		return -EINVAL;
+
+	return 0;
+}
+
+static const struct net_device_ops adin1110_netdev_ops = {
+	.ndo_open		= adin1110_net_open,
+	.ndo_stop		= adin1110_net_stop,
+	.ndo_eth_ioctl		= adin1110_ioctl,
+	.ndo_start_xmit		= adin1110_start_xmit,
+	.ndo_set_mac_address	= adin1110_ndo_set_mac_address,
+	.ndo_set_rx_mode	= adin1110_set_rx_mode,
+	.ndo_validate_addr	= eth_validate_addr,
+	.ndo_get_stats64	= adin1110_ndo_get_stats64,
+	.ndo_get_port_parent_id	= adin1110_port_get_port_parent_id,
+	.ndo_get_phys_port_name	= adin1110_ndo_get_phys_port_name,
+};
+
+static void adin1110_get_drvinfo(struct net_device *dev,
+				 struct ethtool_drvinfo *di)
+{
+	strscpy(di->driver, "ADIN1110", sizeof(di->driver));
+	strscpy(di->bus_info, dev_name(dev->dev.parent), sizeof(di->bus_info));
+}
+
+static const struct ethtool_ops adin1110_ethtool_ops = {
+	.get_drvinfo		= adin1110_get_drvinfo,
+	.get_link		= ethtool_op_get_link,
+	.get_link_ksettings	= phy_ethtool_get_link_ksettings,
+	.set_link_ksettings	= phy_ethtool_set_link_ksettings,
+};
+
+static void adin1110_adjust_link(struct net_device *dev)
+{
+	struct phy_device *phydev = dev->phydev;
+
+	if (!phydev->link)
+		phy_print_status(phydev);
+}
+
+/* PHY ID is stored in the MAC registers too,
+ * check spi connection by reading it.
+ */
+static int adin1110_check_spi(struct adin1110_priv *priv)
+{
+	struct gpio_desc *reset_gpio;
+	int ret;
+	u32 val;
+
+	reset_gpio = devm_gpiod_get_optional(&priv->spidev->dev, "reset",
+					     GPIOD_OUT_LOW);
+	if (reset_gpio) {
+		/* MISO pin is used for internal configuration, can't have
+		 * anyone else disturbing the SDO line.
+		 */
+		spi_bus_lock(priv->spidev->controller);
+
+		gpiod_set_value(reset_gpio, 1);
+		fsleep(10000);
+		gpiod_set_value(reset_gpio, 0);
+
+		/* Need to wait 90 ms before interacting with
+		 * the MAC after a HW reset.
+		 */
+		fsleep(90000);
+
+		spi_bus_unlock(priv->spidev->controller);
+	}
+
+	ret = adin1110_read_reg(priv, ADIN1110_PHY_ID, &val);
+	if (ret < 0)
+		return ret;
+
+	if (val != priv->cfg->phy_id_val) {
+		dev_err(&priv->spidev->dev, "PHY ID expected: %x, read: %x\n",
+			priv->cfg->phy_id_val, val);
+		return -EIO;
+	}
+
+	return 0;
+}
+
+static int adin1110_hw_forwarding(struct adin1110_priv *priv, bool enable)
+{
+	int ret;
+	int i;
+
+	priv->forwarding = enable;
+
+	if (!priv->forwarding) {
+		for (i = ADIN_MAC_FDB_ADDR_SLOT; i < ADIN_MAC_MAX_ADDR_SLOTS; i++) {
+			ret = adin1110_clear_mac_address(priv, i);
+			if (ret < 0)
+				return ret;
+		}
+	}
+
+	/* Forwarding is optimised when MAC runs in Cut Through mode. */
+	ret = adin1110_set_bits(priv, ADIN1110_CONFIG2,
+				ADIN2111_PORT_CUT_THRU_EN,
+				priv->forwarding ? ADIN2111_PORT_CUT_THRU_EN : 0);
+	if (ret < 0)
+		return ret;
+
+	for (i = 0; i < priv->cfg->ports_nr; i++) {
+		ret = adin1110_setup_rx_mode(priv->ports[i]);
+		if (ret < 0)
+			return ret;
+	}
+
+	return ret;
+}
+
+static int adin1110_port_bridge_join(struct adin1110_port_priv *port_priv,
+				     struct net_device *bridge)
+{
+	struct adin1110_priv *priv = port_priv->priv;
+	int ret;
+
+	port_priv->bridge = bridge;
+
+	if (adin1110_can_offload_forwarding(priv)) {
+		mutex_lock(&priv->lock);
+		ret = adin1110_hw_forwarding(priv, true);
+		mutex_unlock(&priv->lock);
+
+		if (ret < 0)
+			return ret;
+	}
+
+	return adin1110_set_mac_address(port_priv->netdev, bridge->dev_addr);
+}
+
+static int adin1110_port_bridge_leave(struct adin1110_port_priv *port_priv,
+				      struct net_device *bridge)
+{
+	struct adin1110_priv *priv = port_priv->priv;
+	int ret;
+
+	port_priv->bridge = NULL;
+
+	mutex_lock(&priv->lock);
+	ret = adin1110_hw_forwarding(priv, false);
+	mutex_unlock(&priv->lock);
+
+	return ret;
+}
+
+static bool adin1110_port_dev_check(const struct net_device *dev)
+{
+	return dev->netdev_ops == &adin1110_netdev_ops;
+}
+
+static int adin1110_netdevice_event(struct notifier_block *unused,
+				    unsigned long event, void *ptr)
+{
+	struct net_device *dev = netdev_notifier_info_to_dev(ptr);
+	struct adin1110_port_priv *port_priv = netdev_priv(dev);
+	struct netdev_notifier_changeupper_info *info = ptr;
+	int ret = 0;
+
+	if (!adin1110_port_dev_check(dev))
+		return NOTIFY_DONE;
+
+	switch (event) {
+	case NETDEV_CHANGEUPPER:
+		if (netif_is_bridge_master(info->upper_dev)) {
+			if (info->linking)
+				ret = adin1110_port_bridge_join(port_priv, info->upper_dev);
+			else
+				ret = adin1110_port_bridge_leave(port_priv, info->upper_dev);
+		}
+		break;
+	default:
+		break;
+	}
+
+	return notifier_from_errno(ret);
+}
+
+static struct notifier_block adin1110_netdevice_nb = {
+	.notifier_call = adin1110_netdevice_event,
+};
+
+static void adin1110_disconnect_phy(void *data)
+{
+	phy_disconnect(data);
+}
+
+static int adin1110_port_set_forwarding_state(struct adin1110_port_priv *port_priv)
+{
+	struct adin1110_priv *priv = port_priv->priv;
+	int ret;
+
+	port_priv->state = BR_STATE_FORWARDING;
+
+	mutex_lock(&priv->lock);
+	ret = adin1110_set_mac_address(port_priv->netdev,
+				       port_priv->netdev->dev_addr);
+	if (ret < 0)
+		goto out;
+
+	if (adin1110_can_offload_forwarding(priv))
+		ret = adin1110_hw_forwarding(priv, true);
+	else
+		ret = adin1110_setup_rx_mode(port_priv);
+out:
+	mutex_unlock(&priv->lock);
+
+	return ret;
+}
+
+static int adin1110_port_set_blocking_state(struct adin1110_port_priv *port_priv)
+{
+	u8 mac[ETH_ALEN] = {0x01, 0x80, 0xC2, 0x00, 0x00, 0x00};
+	struct adin1110_priv *priv = port_priv->priv;
+	u8 mask[ETH_ALEN];
+	u32 port_rules;
+	int mac_slot;
+	int ret;
+
+	port_priv->state = BR_STATE_BLOCKING;
+
+	mutex_lock(&priv->lock);
+
+	mac_slot = (!port_priv->nr) ?  ADIN_MAC_P1_ADDR_SLOT : ADIN_MAC_P2_ADDR_SLOT;
+	ret = adin1110_clear_mac_address(priv, mac_slot);
+	if (ret < 0)
+		goto out;
+
+	ret = adin1110_hw_forwarding(priv, false);
+	if (ret < 0)
+		goto out;
+
+	/* Allow only BPDUs to be passed to the CPU */
+	eth_broadcast_addr(mask);
+	port_rules = adin1110_port_rules(port_priv, true, false);
+	ret = adin1110_write_mac_address(port_priv, mac_slot, mac,
+					 mask, port_rules);
+out:
+	mutex_unlock(&priv->lock);
+
+	return ret;
+}
+
+/* ADIN1110/2111 does not have any native STP support.
+ * Listen for bridge core state changes and
+ * allow all frames to pass or only the BPDUs.
+ */
+static int adin1110_port_attr_stp_state_set(struct adin1110_port_priv *port_priv,
+					    u8 state)
+{
+	switch (state) {
+	case BR_STATE_FORWARDING:
+		return adin1110_port_set_forwarding_state(port_priv);
+	case BR_STATE_LEARNING:
+	case BR_STATE_LISTENING:
+	case BR_STATE_DISABLED:
+	case BR_STATE_BLOCKING:
+		return adin1110_port_set_blocking_state(port_priv);
+	default:
+		return -EINVAL;
+	}
+}
+
+static int adin1110_port_attr_set(struct net_device *dev, const void *ctx,
+				  const struct switchdev_attr *attr,
+				  struct netlink_ext_ack *extack)
+{
+	struct adin1110_port_priv *port_priv = netdev_priv(dev);
+
+	switch (attr->id) {
+	case SWITCHDEV_ATTR_ID_PORT_STP_STATE:
+		return adin1110_port_attr_stp_state_set(port_priv,
+							attr->u.stp_state);
+	default:
+		return -EOPNOTSUPP;
+	}
+}
+
+static int adin1110_switchdev_blocking_event(struct notifier_block *unused,
+					     unsigned long event,
+					     void *ptr)
+{
+	struct net_device *netdev = switchdev_notifier_info_to_dev(ptr);
+	int ret;
+
+	if (event == SWITCHDEV_PORT_ATTR_SET) {
+		ret = switchdev_handle_port_attr_set(netdev, ptr,
+						     adin1110_port_dev_check,
+						     adin1110_port_attr_set);
+
+		return notifier_from_errno(ret);
+	}
+
+	return NOTIFY_DONE;
+}
+
+static struct notifier_block adin1110_switchdev_blocking_notifier = {
+	.notifier_call = adin1110_switchdev_blocking_event,
+};
+
+static void adin1110_fdb_offload_notify(struct net_device *netdev,
+					struct switchdev_notifier_fdb_info *rcv)
+{
+	struct switchdev_notifier_fdb_info info = {};
+
+	info.addr = rcv->addr;
+	info.vid = rcv->vid;
+	info.offloaded = true;
+	call_switchdev_notifiers(SWITCHDEV_FDB_OFFLOADED,
+				 netdev, &info.info, NULL);
+}
+
+static int adin1110_fdb_add(struct adin1110_port_priv *port_priv,
+			    struct switchdev_notifier_fdb_info *fdb)
+{
+	struct adin1110_priv *priv = port_priv->priv;
+	struct adin1110_port_priv *other_port;
+	u8 mask[ETH_ALEN];
+	u32 port_rules;
+	int mac_nr;
+	u32 val;
+	int ret;
+
+	netdev_dbg(port_priv->netdev,
+		   "DEBUG: %s: MACID = %pM vid = %u flags = %u %u -- port %d\n",
+		    __func__, fdb->addr, fdb->vid, fdb->added_by_user,
+		    fdb->offloaded, port_priv->nr);
+
+	if (!priv->forwarding)
+		return 0;
+
+	if (fdb->is_local)
+		return -EINVAL;
+
+	/* Find free FDB slot on device. */
+	for (mac_nr = ADIN_MAC_FDB_ADDR_SLOT; mac_nr < ADIN_MAC_MAX_ADDR_SLOTS; mac_nr++) {
+		ret = adin1110_read_reg(priv, ADIN1110_MAC_ADDR_FILTER_UPR + (mac_nr * 2), &val);
+		if (ret < 0)
+			return ret;
+		if (!val)
+			break;
+	}
+
+	if (mac_nr == ADIN_MAC_MAX_ADDR_SLOTS)
+		return -ENOMEM;
+
+	other_port = priv->ports[!port_priv->nr];
+	port_rules = adin1110_port_rules(other_port, false, true);
+	eth_broadcast_addr(mask);
+
+	return adin1110_write_mac_address(other_port, mac_nr, (u8 *)fdb->addr,
+					  mask, port_rules);
+}
+
+static int adin1110_read_mac(struct adin1110_priv *priv, int mac_nr, u8 *addr)
+{
+	u32 val;
+	int ret;
+
+	ret = adin1110_read_reg(priv, ADIN1110_MAC_ADDR_FILTER_UPR + (mac_nr * 2), &val);
+	if (ret < 0)
+		return ret;
+
+	put_unaligned_be16(val, addr);
+
+	ret = adin1110_read_reg(priv, ADIN1110_MAC_ADDR_FILTER_LWR + (mac_nr * 2), &val);
+	if (ret < 0)
+		return ret;
+
+	put_unaligned_be32(val, addr + 2);
+
+	return 0;
+}
+
+static int adin1110_fdb_del(struct adin1110_port_priv *port_priv,
+			    struct switchdev_notifier_fdb_info *fdb)
+{
+	struct adin1110_priv *priv = port_priv->priv;
+	u8 addr[ETH_ALEN];
+	int mac_nr;
+	int ret;
+
+	netdev_dbg(port_priv->netdev,
+		   "DEBUG: %s: MACID = %pM vid = %u flags = %u %u -- port %d\n",
+		   __func__, fdb->addr, fdb->vid, fdb->added_by_user,
+		   fdb->offloaded, port_priv->nr);
+
+	if (fdb->is_local)
+		return -EINVAL;
+
+	for (mac_nr = ADIN_MAC_FDB_ADDR_SLOT; mac_nr < ADIN_MAC_MAX_ADDR_SLOTS; mac_nr++) {
+		ret = adin1110_read_mac(priv, mac_nr, addr);
+		if (ret < 0)
+			return ret;
+
+		if (ether_addr_equal(addr, fdb->addr)) {
+			ret = adin1110_clear_mac_address(priv, mac_nr);
+			if (ret < 0)
+				return ret;
+		}
+	}
+
+	return 0;
+}
+
+static void adin1110_switchdev_event_work(struct work_struct *work)
+{
+	struct adin1110_switchdev_event_work *switchdev_work;
+	struct adin1110_port_priv *port_priv;
+	int ret;
+
+	switchdev_work = container_of(work, struct adin1110_switchdev_event_work, work);
+	port_priv = switchdev_work->port_priv;
+
+	mutex_lock(&port_priv->priv->lock);
+
+	switch (switchdev_work->event) {
+	case SWITCHDEV_FDB_ADD_TO_DEVICE:
+		ret = adin1110_fdb_add(port_priv, &switchdev_work->fdb_info);
+		if (!ret)
+			adin1110_fdb_offload_notify(port_priv->netdev,
+						    &switchdev_work->fdb_info);
+		break;
+	case SWITCHDEV_FDB_DEL_TO_DEVICE:
+		adin1110_fdb_del(port_priv, &switchdev_work->fdb_info);
+		break;
+	default:
+		break;
+	}
+
+	mutex_unlock(&port_priv->priv->lock);
+
+	kfree(switchdev_work->fdb_info.addr);
+	kfree(switchdev_work);
+	dev_put(port_priv->netdev);
+}
+
+/* called under rcu_read_lock() */
+static int adin1110_switchdev_event(struct notifier_block *unused,
+				    unsigned long event, void *ptr)
+{
+	struct net_device *netdev = switchdev_notifier_info_to_dev(ptr);
+	struct adin1110_port_priv *port_priv = netdev_priv(netdev);
+	struct adin1110_switchdev_event_work *switchdev_work;
+	struct switchdev_notifier_fdb_info *fdb_info = ptr;
+
+	if (!adin1110_port_dev_check(netdev))
+		return NOTIFY_DONE;
+
+	switchdev_work = kzalloc(sizeof(*switchdev_work), GFP_ATOMIC);
+	if (WARN_ON(!switchdev_work))
+		return NOTIFY_BAD;
+
+	INIT_WORK(&switchdev_work->work, adin1110_switchdev_event_work);
+	switchdev_work->port_priv = port_priv;
+	switchdev_work->event = event;
+
+	switch (event) {
+	case SWITCHDEV_FDB_ADD_TO_DEVICE:
+	case SWITCHDEV_FDB_DEL_TO_DEVICE:
+		memcpy(&switchdev_work->fdb_info, ptr,
+		       sizeof(switchdev_work->fdb_info));
+		switchdev_work->fdb_info.addr = kzalloc(ETH_ALEN, GFP_ATOMIC);
+
+		if (!switchdev_work->fdb_info.addr)
+			goto err_addr_alloc;
+
+		ether_addr_copy((u8 *)switchdev_work->fdb_info.addr,
+				fdb_info->addr);
+		dev_hold(netdev);
+		break;
+	default:
+		kfree(switchdev_work);
+		return NOTIFY_DONE;
+	}
+
+	queue_work(system_long_wq, &switchdev_work->work);
+
+	return NOTIFY_DONE;
+
+err_addr_alloc:
+	kfree(switchdev_work);
+	return NOTIFY_BAD;
+}
+
+static struct notifier_block adin1110_switchdev_notifier = {
+	.notifier_call = adin1110_switchdev_event,
+};
+
+static void adin1110_unregister_notifiers(void)
+{
+	unregister_switchdev_blocking_notifier(&adin1110_switchdev_blocking_notifier);
+	unregister_switchdev_notifier(&adin1110_switchdev_notifier);
+	unregister_netdevice_notifier(&adin1110_netdevice_nb);
+}
+
+static int adin1110_setup_notifiers(void)
+{
+	int ret;
+
+	ret = register_netdevice_notifier(&adin1110_netdevice_nb);
+	if (ret < 0)
+		return ret;
+
+	ret = register_switchdev_notifier(&adin1110_switchdev_notifier);
+	if (ret < 0)
+		goto err_netdev;
+
+	ret = register_switchdev_blocking_notifier(&adin1110_switchdev_blocking_notifier);
+	if (ret < 0)
+		goto err_sdev;
+
+	return 0;
+
+err_sdev:
+	unregister_switchdev_notifier(&adin1110_switchdev_notifier);
+
+err_netdev:
+	unregister_netdevice_notifier(&adin1110_netdevice_nb);
+
+	return ret;
+}
+
+static int adin1110_probe_netdevs(struct adin1110_priv *priv)
+{
+	struct device *dev = &priv->spidev->dev;
+	struct adin1110_port_priv *port_priv;
+	struct net_device *netdev;
+	int ret;
+	int i;
+
+	for (i = 0; i < priv->cfg->ports_nr; i++) {
+		netdev = devm_alloc_etherdev(dev, sizeof(*port_priv));
+		if (!netdev)
+			return -ENOMEM;
+
+		port_priv = netdev_priv(netdev);
+		port_priv->netdev = netdev;
+		port_priv->priv = priv;
+		port_priv->cfg = priv->cfg;
+		port_priv->nr = i;
+		priv->ports[i] = port_priv;
+		SET_NETDEV_DEV(netdev, dev);
+
+		ret = device_get_ethdev_address(dev, netdev);
+		if (ret < 0)
+			return ret;
+
+		netdev->irq = priv->spidev->irq;
+		INIT_WORK(&port_priv->tx_work, adin1110_tx_work);
+		INIT_WORK(&port_priv->rx_mode_work, adin1110_rx_mode_work);
+		skb_queue_head_init(&port_priv->txq);
+
+		netif_carrier_off(netdev);
+
+		netdev->if_port = IF_PORT_10BASET;
+		netdev->netdev_ops = &adin1110_netdev_ops;
+		netdev->ethtool_ops = &adin1110_ethtool_ops;
+		netdev->priv_flags |= IFF_UNICAST_FLT;
+		netdev->netns_immutable = true;
+
+		port_priv->phydev = get_phy_device(priv->mii_bus, i + 1, false);
+		if (IS_ERR(port_priv->phydev)) {
+			netdev_err(netdev, "Could not find PHY with device address: %d.\n", i);
+			return PTR_ERR(port_priv->phydev);
+		}
+
+		port_priv->phydev = phy_connect(netdev,
+						phydev_name(port_priv->phydev),
+						adin1110_adjust_link,
+						PHY_INTERFACE_MODE_INTERNAL);
+		if (IS_ERR(port_priv->phydev)) {
+			netdev_err(netdev, "Could not connect PHY with device address: %d.\n", i);
+			return PTR_ERR(port_priv->phydev);
+		}
+
+		ret = devm_add_action_or_reset(dev, adin1110_disconnect_phy,
+					       port_priv->phydev);
+		if (ret < 0)
+			return ret;
+	}
+
+	/* ADIN1110 INT_N pin will be used to signal the host */
+	ret = devm_request_threaded_irq(dev, priv->spidev->irq, NULL,
+					adin1110_irq,
+					IRQF_TRIGGER_LOW | IRQF_ONESHOT,
+					dev_name(dev), priv);
+	if (ret < 0)
+		return ret;
+
+	for (i = 0; i < priv->cfg->ports_nr; i++) {
+		ret = devm_register_netdev(dev, priv->ports[i]->netdev);
+		if (ret < 0) {
+			dev_err(dev, "Failed to register network device.\n");
+			return ret;
+		}
+	}
+
+	return 0;
+}
+
+static int adin1110_probe(struct spi_device *spi)
+{
+	const struct spi_device_id *dev_id = spi_get_device_id(spi);
+	struct device *dev = &spi->dev;
+	struct adin1110_priv *priv;
+	int ret;
+
+	priv = devm_kzalloc(dev, sizeof(struct adin1110_priv), GFP_KERNEL);
+	if (!priv)
+		return -ENOMEM;
+
+	priv->spidev = spi;
+	priv->cfg = &adin1110_cfgs[dev_id->driver_data];
+	spi->bits_per_word = 8;
+	spi->mode = SPI_MODE_0;
+
+	mutex_init(&priv->lock);
+	spin_lock_init(&priv->state_lock);
+
+	/* use of CRC on control and data transactions is pin dependent */
+	priv->append_crc = device_property_read_bool(dev, "adi,spi-crc");
+	if (priv->append_crc)
+		crc8_populate_msb(adin1110_crc_table, 0x7);
+
+	ret = adin1110_check_spi(priv);
+	if (ret < 0) {
+		dev_err(dev, "Probe SPI Read check failed: %d\n", ret);
+		return ret;
+	}
+
+	ret = adin1110_write_reg(priv, ADIN1110_RESET, ADIN1110_SWRESET);
+	if (ret < 0)
+		return ret;
+
+	ret = adin1110_register_mdiobus(priv, dev);
+	if (ret < 0) {
+		dev_err(dev, "Could not register MDIO bus %d\n", ret);
+		return ret;
+	}
+
+	return adin1110_probe_netdevs(priv);
+}
+
+static const struct of_device_id adin1110_match_table[] = {
+	{ .compatible = "adi,adin1110" },
+	{ .compatible = "adi,adin2111" },
+	{ }
+};
+MODULE_DEVICE_TABLE(of, adin1110_match_table);
+
+static const struct spi_device_id adin1110_spi_id[] = {
+	{ .name = "adin1110", .driver_data = ADIN1110_MAC },
+	{ .name = "adin2111", .driver_data = ADIN2111_MAC },
+	{ }
+};
+MODULE_DEVICE_TABLE(spi, adin1110_spi_id);
+
+static struct spi_driver adin1110_driver = {
+	.driver = {
+		.name = "adin1110",
+		.of_match_table = adin1110_match_table,
+	},
+	.probe = adin1110_probe,
+	.id_table = adin1110_spi_id,
+};
+
+static int __init adin1110_driver_init(void)
+{
+	int ret;
+
+	ret = adin1110_setup_notifiers();
+	if (ret < 0)
+		return ret;
+
+	ret = spi_register_driver(&adin1110_driver);
+	if (ret < 0) {
+		adin1110_unregister_notifiers();
+		return ret;
+	}
+
+	return 0;
+}
+
+static void __exit adin1110_exit(void)
+{
+	adin1110_unregister_notifiers();
+	spi_unregister_driver(&adin1110_driver);
+}
+module_init(adin1110_driver_init);
+module_exit(adin1110_exit);
+
+MODULE_DESCRIPTION("ADIN1110 Network driver");
+MODULE_AUTHOR("Alexandru Tachici <alexandru.tachici@analog.com>");
+MODULE_LICENSE("Dual BSD/GPL");
diff --git a/drivers/net/ethernet/adi/bfin_mac.c b/drivers/net/ethernet/adi/bfin_mac.c
deleted file mode 100644
index a9ac58c351a0..000000000000
--- a/drivers/net/ethernet/adi/bfin_mac.c
+++ /dev/null
@@ -1,1881 +0,0 @@
-/*
- * Blackfin On-Chip MAC Driver
- *
- * Copyright 2004-2010 Analog Devices Inc.
- *
- * Enter bugs at http://blackfin.uclinux.org/
- *
- * Licensed under the GPL-2 or later.
- */
-
-#define DRV_VERSION	"1.1"
-#define DRV_DESC	"Blackfin on-chip Ethernet MAC driver"
-
-#define pr_fmt(fmt) KBUILD_MODNAME ": " fmt
-
-#include <linux/init.h>
-#include <linux/module.h>
-#include <linux/kernel.h>
-#include <linux/sched.h>
-#include <linux/slab.h>
-#include <linux/delay.h>
-#include <linux/timer.h>
-#include <linux/errno.h>
-#include <linux/irq.h>
-#include <linux/io.h>
-#include <linux/ioport.h>
-#include <linux/crc32.h>
-#include <linux/device.h>
-#include <linux/spinlock.h>
-#include <linux/mii.h>
-#include <linux/netdevice.h>
-#include <linux/etherdevice.h>
-#include <linux/ethtool.h>
-#include <linux/skbuff.h>
-#include <linux/platform_device.h>
-
-#include <asm/dma.h>
-#include <linux/dma-mapping.h>
-
-#include <asm/div64.h>
-#include <asm/dpmc.h>
-#include <asm/blackfin.h>
-#include <asm/cacheflush.h>
-#include <asm/portmux.h>
-#include <mach/pll.h>
-
-#include "bfin_mac.h"
-
-MODULE_AUTHOR("Bryan Wu, Luke Yang");
-MODULE_LICENSE("GPL");
-MODULE_DESCRIPTION(DRV_DESC);
-MODULE_ALIAS("platform:bfin_mac");
-
-#if defined(CONFIG_BFIN_MAC_USE_L1)
-# define bfin_mac_alloc(dma_handle, size, num)  l1_data_sram_zalloc(size*num)
-# define bfin_mac_free(dma_handle, ptr, num)    l1_data_sram_free(ptr)
-#else
-# define bfin_mac_alloc(dma_handle, size, num) \
-	dma_alloc_coherent(NULL, size*num, dma_handle, GFP_KERNEL)
-# define bfin_mac_free(dma_handle, ptr, num) \
-	dma_free_coherent(NULL, sizeof(*ptr)*num, ptr, dma_handle)
-#endif
-
-#define PKT_BUF_SZ 1580
-
-#define MAX_TIMEOUT_CNT	500
-
-/* pointers to maintain transmit list */
-static struct net_dma_desc_tx *tx_list_head;
-static struct net_dma_desc_tx *tx_list_tail;
-static struct net_dma_desc_rx *rx_list_head;
-static struct net_dma_desc_rx *rx_list_tail;
-static struct net_dma_desc_rx *current_rx_ptr;
-static struct net_dma_desc_tx *current_tx_ptr;
-static struct net_dma_desc_tx *tx_desc;
-static struct net_dma_desc_rx *rx_desc;
-
-static void desc_list_free(void)
-{
-	struct net_dma_desc_rx *r;
-	struct net_dma_desc_tx *t;
-	int i;
-#if !defined(CONFIG_BFIN_MAC_USE_L1)
-	dma_addr_t dma_handle = 0;
-#endif
-
-	if (tx_desc) {
-		t = tx_list_head;
-		for (i = 0; i < CONFIG_BFIN_TX_DESC_NUM; i++) {
-			if (t) {
-				if (t->skb) {
-					dev_kfree_skb(t->skb);
-					t->skb = NULL;
-				}
-				t = t->next;
-			}
-		}
-		bfin_mac_free(dma_handle, tx_desc, CONFIG_BFIN_TX_DESC_NUM);
-	}
-
-	if (rx_desc) {
-		r = rx_list_head;
-		for (i = 0; i < CONFIG_BFIN_RX_DESC_NUM; i++) {
-			if (r) {
-				if (r->skb) {
-					dev_kfree_skb(r->skb);
-					r->skb = NULL;
-				}
-				r = r->next;
-			}
-		}
-		bfin_mac_free(dma_handle, rx_desc, CONFIG_BFIN_RX_DESC_NUM);
-	}
-}
-
-static int desc_list_init(struct net_device *dev)
-{
-	int i;
-	struct sk_buff *new_skb;
-#if !defined(CONFIG_BFIN_MAC_USE_L1)
-	/*
-	 * This dma_handle is useless in Blackfin dma_alloc_coherent().
-	 * The real dma handler is the return value of dma_alloc_coherent().
-	 */
-	dma_addr_t dma_handle;
-#endif
-
-	tx_desc = bfin_mac_alloc(&dma_handle,
-				sizeof(struct net_dma_desc_tx),
-				CONFIG_BFIN_TX_DESC_NUM);
-	if (tx_desc == NULL)
-		goto init_error;
-
-	rx_desc = bfin_mac_alloc(&dma_handle,
-				sizeof(struct net_dma_desc_rx),
-				CONFIG_BFIN_RX_DESC_NUM);
-	if (rx_desc == NULL)
-		goto init_error;
-
-	/* init tx_list */
-	tx_list_head = tx_list_tail = tx_desc;
-
-	for (i = 0; i < CONFIG_BFIN_TX_DESC_NUM; i++) {
-		struct net_dma_desc_tx *t = tx_desc + i;
-		struct dma_descriptor *a = &(t->desc_a);
-		struct dma_descriptor *b = &(t->desc_b);
-
-		/*
-		 * disable DMA
-		 * read from memory WNR = 0
-		 * wordsize is 32 bits
-		 * 6 half words is desc size
-		 * large desc flow
-		 */
-		a->config = WDSIZE_32 | NDSIZE_6 | DMAFLOW_LARGE;
-		a->start_addr = (unsigned long)t->packet;
-		a->x_count = 0;
-		a->next_dma_desc = b;
-
-		/*
-		 * enabled DMA
-		 * write to memory WNR = 1
-		 * wordsize is 32 bits
-		 * disable interrupt
-		 * 6 half words is desc size
-		 * large desc flow
-		 */
-		b->config = DMAEN | WNR | WDSIZE_32 | NDSIZE_6 | DMAFLOW_LARGE;
-		b->start_addr = (unsigned long)(&(t->status));
-		b->x_count = 0;
-
-		t->skb = NULL;
-		tx_list_tail->desc_b.next_dma_desc = a;
-		tx_list_tail->next = t;
-		tx_list_tail = t;
-	}
-	tx_list_tail->next = tx_list_head;	/* tx_list is a circle */
-	tx_list_tail->desc_b.next_dma_desc = &(tx_list_head->desc_a);
-	current_tx_ptr = tx_list_head;
-
-	/* init rx_list */
-	rx_list_head = rx_list_tail = rx_desc;
-
-	for (i = 0; i < CONFIG_BFIN_RX_DESC_NUM; i++) {
-		struct net_dma_desc_rx *r = rx_desc + i;
-		struct dma_descriptor *a = &(r->desc_a);
-		struct dma_descriptor *b = &(r->desc_b);
-
-		/* allocate a new skb for next time receive */
-		new_skb = netdev_alloc_skb(dev, PKT_BUF_SZ + NET_IP_ALIGN);
-		if (!new_skb)
-			goto init_error;
-
-		skb_reserve(new_skb, NET_IP_ALIGN);
-		/* Invalidate the data cache of skb->data range when it is write back
-		 * cache. It will prevent overwriting the new data from DMA
-		 */
-		blackfin_dcache_invalidate_range((unsigned long)new_skb->head,
-					 (unsigned long)new_skb->end);
-		r->skb = new_skb;
-
-		/*
-		 * enabled DMA
-		 * write to memory WNR = 1
-		 * wordsize is 32 bits
-		 * disable interrupt
-		 * 6 half words is desc size
-		 * large desc flow
-		 */
-		a->config = DMAEN | WNR | WDSIZE_32 | NDSIZE_6 | DMAFLOW_LARGE;
-		/* since RXDWA is enabled */
-		a->start_addr = (unsigned long)new_skb->data - 2;
-		a->x_count = 0;
-		a->next_dma_desc = b;
-
-		/*
-		 * enabled DMA
-		 * write to memory WNR = 1
-		 * wordsize is 32 bits
-		 * enable interrupt
-		 * 6 half words is desc size
-		 * large desc flow
-		 */
-		b->config = DMAEN | WNR | WDSIZE_32 | DI_EN |
-				NDSIZE_6 | DMAFLOW_LARGE;
-		b->start_addr = (unsigned long)(&(r->status));
-		b->x_count = 0;
-
-		rx_list_tail->desc_b.next_dma_desc = a;
-		rx_list_tail->next = r;
-		rx_list_tail = r;
-	}
-	rx_list_tail->next = rx_list_head;	/* rx_list is a circle */
-	rx_list_tail->desc_b.next_dma_desc = &(rx_list_head->desc_a);
-	current_rx_ptr = rx_list_head;
-
-	return 0;
-
-init_error:
-	desc_list_free();
-	pr_err("kmalloc failed\n");
-	return -ENOMEM;
-}
-
-
-/*---PHY CONTROL AND CONFIGURATION-----------------------------------------*/
-
-/*
- * MII operations
- */
-/* Wait until the previous MDC/MDIO transaction has completed */
-static int bfin_mdio_poll(void)
-{
-	int timeout_cnt = MAX_TIMEOUT_CNT;
-
-	/* poll the STABUSY bit */
-	while ((bfin_read_EMAC_STAADD()) & STABUSY) {
-		udelay(1);
-		if (timeout_cnt-- < 0) {
-			pr_err("wait MDC/MDIO transaction to complete timeout\n");
-			return -ETIMEDOUT;
-		}
-	}
-
-	return 0;
-}
-
-/* Read an off-chip register in a PHY through the MDC/MDIO port */
-static int bfin_mdiobus_read(struct mii_bus *bus, int phy_addr, int regnum)
-{
-	int ret;
-
-	ret = bfin_mdio_poll();
-	if (ret)
-		return ret;
-
-	/* read mode */
-	bfin_write_EMAC_STAADD(SET_PHYAD((u16) phy_addr) |
-				SET_REGAD((u16) regnum) |
-				STABUSY);
-
-	ret = bfin_mdio_poll();
-	if (ret)
-		return ret;
-
-	return (int) bfin_read_EMAC_STADAT();
-}
-
-/* Write an off-chip register in a PHY through the MDC/MDIO port */
-static int bfin_mdiobus_write(struct mii_bus *bus, int phy_addr, int regnum,
-			      u16 value)
-{
-	int ret;
-
-	ret = bfin_mdio_poll();
-	if (ret)
-		return ret;
-
-	bfin_write_EMAC_STADAT((u32) value);
-
-	/* write mode */
-	bfin_write_EMAC_STAADD(SET_PHYAD((u16) phy_addr) |
-				SET_REGAD((u16) regnum) |
-				STAOP |
-				STABUSY);
-
-	return bfin_mdio_poll();
-}
-
-static void bfin_mac_adjust_link(struct net_device *dev)
-{
-	struct bfin_mac_local *lp = netdev_priv(dev);
-	struct phy_device *phydev = dev->phydev;
-	unsigned long flags;
-	int new_state = 0;
-
-	spin_lock_irqsave(&lp->lock, flags);
-	if (phydev->link) {
-		/* Now we make sure that we can be in full duplex mode.
-		 * If not, we operate in half-duplex mode. */
-		if (phydev->duplex != lp->old_duplex) {
-			u32 opmode = bfin_read_EMAC_OPMODE();
-			new_state = 1;
-
-			if (phydev->duplex)
-				opmode |= FDMODE;
-			else
-				opmode &= ~(FDMODE);
-
-			bfin_write_EMAC_OPMODE(opmode);
-			lp->old_duplex = phydev->duplex;
-		}
-
-		if (phydev->speed != lp->old_speed) {
-			if (phydev->interface == PHY_INTERFACE_MODE_RMII) {
-				u32 opmode = bfin_read_EMAC_OPMODE();
-				switch (phydev->speed) {
-				case 10:
-					opmode |= RMII_10;
-					break;
-				case 100:
-					opmode &= ~RMII_10;
-					break;
-				default:
-					netdev_warn(dev,
-						"Ack! Speed (%d) is not 10/100!\n",
-						phydev->speed);
-					break;
-				}
-				bfin_write_EMAC_OPMODE(opmode);
-			}
-
-			new_state = 1;
-			lp->old_speed = phydev->speed;
-		}
-
-		if (!lp->old_link) {
-			new_state = 1;
-			lp->old_link = 1;
-		}
-	} else if (lp->old_link) {
-		new_state = 1;
-		lp->old_link = 0;
-		lp->old_speed = 0;
-		lp->old_duplex = -1;
-	}
-
-	if (new_state) {
-		u32 opmode = bfin_read_EMAC_OPMODE();
-		phy_print_status(phydev);
-		pr_debug("EMAC_OPMODE = 0x%08x\n", opmode);
-	}
-
-	spin_unlock_irqrestore(&lp->lock, flags);
-}
-
-/* MDC  = 2.5 MHz */
-#define MDC_CLK 2500000
-
-static int mii_probe(struct net_device *dev, int phy_mode)
-{
-	struct bfin_mac_local *lp = netdev_priv(dev);
-	struct phy_device *phydev;
-	unsigned short sysctl;
-	u32 sclk, mdc_div;
-
-	/* Enable PHY output early */
-	if (!(bfin_read_VR_CTL() & CLKBUFOE))
-		bfin_write_VR_CTL(bfin_read_VR_CTL() | CLKBUFOE);
-
-	sclk = get_sclk();
-	mdc_div = ((sclk / MDC_CLK) / 2) - 1;
-
-	sysctl = bfin_read_EMAC_SYSCTL();
-	sysctl = (sysctl & ~MDCDIV) | SET_MDCDIV(mdc_div);
-	bfin_write_EMAC_SYSCTL(sysctl);
-
-	phydev = phy_find_first(lp->mii_bus);
-	if (!phydev) {
-		netdev_err(dev, "no phy device found\n");
-		return -ENODEV;
-	}
-
-	if (phy_mode != PHY_INTERFACE_MODE_RMII &&
-		phy_mode != PHY_INTERFACE_MODE_MII) {
-		netdev_err(dev, "invalid phy interface mode\n");
-		return -EINVAL;
-	}
-
-	phydev = phy_connect(dev, phydev_name(phydev),
-			     &bfin_mac_adjust_link, phy_mode);
-
-	if (IS_ERR(phydev)) {
-		netdev_err(dev, "could not attach PHY\n");
-		return PTR_ERR(phydev);
-	}
-
-	/* mask with MAC supported features */
-	phydev->supported &= (SUPPORTED_10baseT_Half
-			      | SUPPORTED_10baseT_Full
-			      | SUPPORTED_100baseT_Half
-			      | SUPPORTED_100baseT_Full
-			      | SUPPORTED_Autoneg
-			      | SUPPORTED_Pause | SUPPORTED_Asym_Pause
-			      | SUPPORTED_MII
-			      | SUPPORTED_TP);
-
-	phydev->advertising = phydev->supported;
-
-	lp->old_link = 0;
-	lp->old_speed = 0;
-	lp->old_duplex = -1;
-
-	phy_attached_print(phydev, "mdc_clk=%dHz(mdc_div=%d)@sclk=%dMHz)\n",
-			   MDC_CLK, mdc_div, sclk / 1000000);
-
-	return 0;
-}
-
-/*
- * Ethtool support
- */
-
-/*
- * interrupt routine for magic packet wakeup
- */
-static irqreturn_t bfin_mac_wake_interrupt(int irq, void *dev_id)
-{
-	return IRQ_HANDLED;
-}
-
-static void bfin_mac_ethtool_getdrvinfo(struct net_device *dev,
-					struct ethtool_drvinfo *info)
-{
-	strlcpy(info->driver, KBUILD_MODNAME, sizeof(info->driver));
-	strlcpy(info->version, DRV_VERSION, sizeof(info->version));
-	strlcpy(info->fw_version, "N/A", sizeof(info->fw_version));
-	strlcpy(info->bus_info, dev_name(&dev->dev), sizeof(info->bus_info));
-}
-
-static void bfin_mac_ethtool_getwol(struct net_device *dev,
-	struct ethtool_wolinfo *wolinfo)
-{
-	struct bfin_mac_local *lp = netdev_priv(dev);
-
-	wolinfo->supported = WAKE_MAGIC;
-	wolinfo->wolopts = lp->wol;
-}
-
-static int bfin_mac_ethtool_setwol(struct net_device *dev,
-	struct ethtool_wolinfo *wolinfo)
-{
-	struct bfin_mac_local *lp = netdev_priv(dev);
-	int rc;
-
-	if (wolinfo->wolopts & (WAKE_MAGICSECURE |
-				WAKE_UCAST |
-				WAKE_MCAST |
-				WAKE_BCAST |
-				WAKE_ARP))
-		return -EOPNOTSUPP;
-
-	lp->wol = wolinfo->wolopts;
-
-	if (lp->wol && !lp->irq_wake_requested) {
-		/* register wake irq handler */
-		rc = request_irq(IRQ_MAC_WAKEDET, bfin_mac_wake_interrupt,
-				 0, "EMAC_WAKE", dev);
-		if (rc)
-			return rc;
-		lp->irq_wake_requested = true;
-	}
-
-	if (!lp->wol && lp->irq_wake_requested) {
-		free_irq(IRQ_MAC_WAKEDET, dev);
-		lp->irq_wake_requested = false;
-	}
-
-	/* Make sure the PHY driver doesn't suspend */
-	device_init_wakeup(&dev->dev, lp->wol);
-
-	return 0;
-}
-
-#ifdef CONFIG_BFIN_MAC_USE_HWSTAMP
-static int bfin_mac_ethtool_get_ts_info(struct net_device *dev,
-	struct ethtool_ts_info *info)
-{
-	struct bfin_mac_local *lp = netdev_priv(dev);
-
-	info->so_timestamping =
-		SOF_TIMESTAMPING_TX_HARDWARE |
-		SOF_TIMESTAMPING_RX_HARDWARE |
-		SOF_TIMESTAMPING_RAW_HARDWARE;
-	info->phc_index = lp->phc_index;
-	info->tx_types =
-		(1 << HWTSTAMP_TX_OFF) |
-		(1 << HWTSTAMP_TX_ON);
-	info->rx_filters =
-		(1 << HWTSTAMP_FILTER_NONE) |
-		(1 << HWTSTAMP_FILTER_PTP_V1_L4_EVENT) |
-		(1 << HWTSTAMP_FILTER_PTP_V2_L2_EVENT) |
-		(1 << HWTSTAMP_FILTER_PTP_V2_L4_EVENT);
-	return 0;
-}
-#endif
-
-static const struct ethtool_ops bfin_mac_ethtool_ops = {
-	.get_link = ethtool_op_get_link,
-	.get_drvinfo = bfin_mac_ethtool_getdrvinfo,
-	.get_wol = bfin_mac_ethtool_getwol,
-	.set_wol = bfin_mac_ethtool_setwol,
-#ifdef CONFIG_BFIN_MAC_USE_HWSTAMP
-	.get_ts_info = bfin_mac_ethtool_get_ts_info,
-#endif
-	.get_link_ksettings = phy_ethtool_get_link_ksettings,
-	.set_link_ksettings = phy_ethtool_set_link_ksettings,
-};
-
-/**************************************************************************/
-static void setup_system_regs(struct net_device *dev)
-{
-	struct bfin_mac_local *lp = netdev_priv(dev);
-	int i;
-	unsigned short sysctl;
-
-	/*
-	 * Odd word alignment for Receive Frame DMA word
-	 * Configure checksum support and rcve frame word alignment
-	 */
-	sysctl = bfin_read_EMAC_SYSCTL();
-	/*
-	 * check if interrupt is requested for any PHY,
-	 * enable PHY interrupt only if needed
-	 */
-	for (i = 0; i < PHY_MAX_ADDR; ++i)
-		if (lp->mii_bus->irq[i] != PHY_POLL)
-			break;
-	if (i < PHY_MAX_ADDR)
-		sysctl |= PHYIE;
-	sysctl |= RXDWA;
-#if defined(BFIN_MAC_CSUM_OFFLOAD)
-	sysctl |= RXCKS;
-#else
-	sysctl &= ~RXCKS;
-#endif
-	bfin_write_EMAC_SYSCTL(sysctl);
-
-	bfin_write_EMAC_MMC_CTL(RSTC | CROLL);
-
-	/* Set vlan regs to let 1522 bytes long packets pass through */
-	bfin_write_EMAC_VLAN1(lp->vlan1_mask);
-	bfin_write_EMAC_VLAN2(lp->vlan2_mask);
-
-	/* Initialize the TX DMA channel registers */
-	bfin_write_DMA2_X_COUNT(0);
-	bfin_write_DMA2_X_MODIFY(4);
-	bfin_write_DMA2_Y_COUNT(0);
-	bfin_write_DMA2_Y_MODIFY(0);
-
-	/* Initialize the RX DMA channel registers */
-	bfin_write_DMA1_X_COUNT(0);
-	bfin_write_DMA1_X_MODIFY(4);
-	bfin_write_DMA1_Y_COUNT(0);
-	bfin_write_DMA1_Y_MODIFY(0);
-}
-
-static void setup_mac_addr(u8 *mac_addr)
-{
-	u32 addr_low = le32_to_cpu(*(__le32 *) & mac_addr[0]);
-	u16 addr_hi = le16_to_cpu(*(__le16 *) & mac_addr[4]);
-
-	/* this depends on a little-endian machine */
-	bfin_write_EMAC_ADDRLO(addr_low);
-	bfin_write_EMAC_ADDRHI(addr_hi);
-}
-
-static int bfin_mac_set_mac_address(struct net_device *dev, void *p)
-{
-	struct sockaddr *addr = p;
-	if (netif_running(dev))
-		return -EBUSY;
-	memcpy(dev->dev_addr, addr->sa_data, dev->addr_len);
-	setup_mac_addr(dev->dev_addr);
-	return 0;
-}
-
-#ifdef CONFIG_BFIN_MAC_USE_HWSTAMP
-#define bfin_mac_hwtstamp_is_none(cfg) ((cfg) == HWTSTAMP_FILTER_NONE)
-
-static u32 bfin_select_phc_clock(u32 input_clk, unsigned int *shift_result)
-{
-	u32 ipn = 1000000000UL / input_clk;
-	u32 ppn = 1;
-	unsigned int shift = 0;
-
-	while (ppn <= ipn) {
-		ppn <<= 1;
-		shift++;
-	}
-	*shift_result = shift;
-	return 1000000000UL / ppn;
-}
-
-static int bfin_mac_hwtstamp_set(struct net_device *netdev,
-				 struct ifreq *ifr)
-{
-	struct hwtstamp_config config;
-	struct bfin_mac_local *lp = netdev_priv(netdev);
-	u16 ptpctl;
-	u32 ptpfv1, ptpfv2, ptpfv3, ptpfoff;
-
-	if (copy_from_user(&config, ifr->ifr_data, sizeof(config)))
-		return -EFAULT;
-
-	pr_debug("%s config flag:0x%x, tx_type:0x%x, rx_filter:0x%x\n",
-			__func__, config.flags, config.tx_type, config.rx_filter);
-
-	/* reserved for future extensions */
-	if (config.flags)
-		return -EINVAL;
-
-	if ((config.tx_type != HWTSTAMP_TX_OFF) &&
-			(config.tx_type != HWTSTAMP_TX_ON))
-		return -ERANGE;
-
-	ptpctl = bfin_read_EMAC_PTP_CTL();
-
-	switch (config.rx_filter) {
-	case HWTSTAMP_FILTER_NONE:
-		/*
-		 * Dont allow any timestamping
-		 */
-		ptpfv3 = 0xFFFFFFFF;
-		bfin_write_EMAC_PTP_FV3(ptpfv3);
-		break;
-	case HWTSTAMP_FILTER_PTP_V1_L4_EVENT:
-	case HWTSTAMP_FILTER_PTP_V1_L4_SYNC:
-	case HWTSTAMP_FILTER_PTP_V1_L4_DELAY_REQ:
-		/*
-		 * Clear the five comparison mask bits (bits[12:8]) in EMAC_PTP_CTL)
-		 * to enable all the field matches.
-		 */
-		ptpctl &= ~0x1F00;
-		bfin_write_EMAC_PTP_CTL(ptpctl);
-		/*
-		 * Keep the default values of the EMAC_PTP_FOFF register.
-		 */
-		ptpfoff = 0x4A24170C;
-		bfin_write_EMAC_PTP_FOFF(ptpfoff);
-		/*
-		 * Keep the default values of the EMAC_PTP_FV1 and EMAC_PTP_FV2
-		 * registers.
-		 */
-		ptpfv1 = 0x11040800;
-		bfin_write_EMAC_PTP_FV1(ptpfv1);
-		ptpfv2 = 0x0140013F;
-		bfin_write_EMAC_PTP_FV2(ptpfv2);
-		/*
-		 * The default value (0xFFFC) allows the timestamping of both
-		 * received Sync messages and Delay_Req messages.
-		 */
-		ptpfv3 = 0xFFFFFFFC;
-		bfin_write_EMAC_PTP_FV3(ptpfv3);
-
-		config.rx_filter = HWTSTAMP_FILTER_PTP_V1_L4_EVENT;
-		break;
-	case HWTSTAMP_FILTER_PTP_V2_L4_EVENT:
-	case HWTSTAMP_FILTER_PTP_V2_L4_SYNC:
-	case HWTSTAMP_FILTER_PTP_V2_DELAY_REQ:
-		/* Clear all five comparison mask bits (bits[12:8]) in the
-		 * EMAC_PTP_CTL register to enable all the field matches.
-		 */
-		ptpctl &= ~0x1F00;
-		bfin_write_EMAC_PTP_CTL(ptpctl);
-		/*
-		 * Keep the default values of the EMAC_PTP_FOFF register, except set
-		 * the PTPCOF field to 0x2A.
-		 */
-		ptpfoff = 0x2A24170C;
-		bfin_write_EMAC_PTP_FOFF(ptpfoff);
-		/*
-		 * Keep the default values of the EMAC_PTP_FV1 and EMAC_PTP_FV2
-		 * registers.
-		 */
-		ptpfv1 = 0x11040800;
-		bfin_write_EMAC_PTP_FV1(ptpfv1);
-		ptpfv2 = 0x0140013F;
-		bfin_write_EMAC_PTP_FV2(ptpfv2);
-		/*
-		 * To allow the timestamping of Pdelay_Req and Pdelay_Resp, set
-		 * the value to 0xFFF0.
-		 */
-		ptpfv3 = 0xFFFFFFF0;
-		bfin_write_EMAC_PTP_FV3(ptpfv3);
-
-		config.rx_filter = HWTSTAMP_FILTER_PTP_V2_L4_EVENT;
-		break;
-	case HWTSTAMP_FILTER_PTP_V2_L2_EVENT:
-	case HWTSTAMP_FILTER_PTP_V2_L2_SYNC:
-	case HWTSTAMP_FILTER_PTP_V2_L2_DELAY_REQ:
-		/*
-		 * Clear bits 8 and 12 of the EMAC_PTP_CTL register to enable only the
-		 * EFTM and PTPCM field comparison.
-		 */
-		ptpctl &= ~0x1100;
-		bfin_write_EMAC_PTP_CTL(ptpctl);
-		/*
-		 * Keep the default values of all the fields of the EMAC_PTP_FOFF
-		 * register, except set the PTPCOF field to 0x0E.
-		 */
-		ptpfoff = 0x0E24170C;
-		bfin_write_EMAC_PTP_FOFF(ptpfoff);
-		/*
-		 * Program bits [15:0] of the EMAC_PTP_FV1 register to 0x88F7, which
-		 * corresponds to PTP messages on the MAC layer.
-		 */
-		ptpfv1 = 0x110488F7;
-		bfin_write_EMAC_PTP_FV1(ptpfv1);
-		ptpfv2 = 0x0140013F;
-		bfin_write_EMAC_PTP_FV2(ptpfv2);
-		/*
-		 * To allow the timestamping of Pdelay_Req and Pdelay_Resp
-		 * messages, set the value to 0xFFF0.
-		 */
-		ptpfv3 = 0xFFFFFFF0;
-		bfin_write_EMAC_PTP_FV3(ptpfv3);
-
-		config.rx_filter = HWTSTAMP_FILTER_PTP_V2_L2_EVENT;
-		break;
-	default:
-		return -ERANGE;
-	}
-
-	if (config.tx_type == HWTSTAMP_TX_OFF &&
-	    bfin_mac_hwtstamp_is_none(config.rx_filter)) {
-		ptpctl &= ~PTP_EN;
-		bfin_write_EMAC_PTP_CTL(ptpctl);
-
-		SSYNC();
-	} else {
-		ptpctl |= PTP_EN;
-		bfin_write_EMAC_PTP_CTL(ptpctl);
-
-		/*
-		 * clear any existing timestamp
-		 */
-		bfin_read_EMAC_PTP_RXSNAPLO();
-		bfin_read_EMAC_PTP_RXSNAPHI();
-
-		bfin_read_EMAC_PTP_TXSNAPLO();
-		bfin_read_EMAC_PTP_TXSNAPHI();
-
-		SSYNC();
-	}
-
-	lp->stamp_cfg = config;
-	return copy_to_user(ifr->ifr_data, &config, sizeof(config)) ?
-		-EFAULT : 0;
-}
-
-static int bfin_mac_hwtstamp_get(struct net_device *netdev,
-				 struct ifreq *ifr)
-{
-	struct bfin_mac_local *lp = netdev_priv(netdev);
-
-	return copy_to_user(ifr->ifr_data, &lp->stamp_cfg,
-			    sizeof(lp->stamp_cfg)) ?
-		-EFAULT : 0;
-}
-
-static void bfin_tx_hwtstamp(struct net_device *netdev, struct sk_buff *skb)
-{
-	struct bfin_mac_local *lp = netdev_priv(netdev);
-
-	if (skb_shinfo(skb)->tx_flags & SKBTX_HW_TSTAMP) {
-		int timeout_cnt = MAX_TIMEOUT_CNT;
-
-		/* When doing time stamping, keep the connection to the socket
-		 * a while longer
-		 */
-		skb_shinfo(skb)->tx_flags |= SKBTX_IN_PROGRESS;
-
-		/*
-		 * The timestamping is done at the EMAC module's MII/RMII interface
-		 * when the module sees the Start of Frame of an event message packet. This
-		 * interface is the closest possible place to the physical Ethernet transmission
-		 * medium, providing the best timing accuracy.
-		 */
-		while ((!(bfin_read_EMAC_PTP_ISTAT() & TXTL)) && (--timeout_cnt))
-			udelay(1);
-		if (timeout_cnt == 0)
-			netdev_err(netdev, "timestamp the TX packet failed\n");
-		else {
-			struct skb_shared_hwtstamps shhwtstamps;
-			u64 ns;
-			u64 regval;
-
-			regval = bfin_read_EMAC_PTP_TXSNAPLO();
-			regval |= (u64)bfin_read_EMAC_PTP_TXSNAPHI() << 32;
-			memset(&shhwtstamps, 0, sizeof(shhwtstamps));
-			ns = regval << lp->shift;
-			shhwtstamps.hwtstamp = ns_to_ktime(ns);
-			skb_tstamp_tx(skb, &shhwtstamps);
-		}
-	}
-}
-
-static void bfin_rx_hwtstamp(struct net_device *netdev, struct sk_buff *skb)
-{
-	struct bfin_mac_local *lp = netdev_priv(netdev);
-	u32 valid;
-	u64 regval, ns;
-	struct skb_shared_hwtstamps *shhwtstamps;
-
-	if (bfin_mac_hwtstamp_is_none(lp->stamp_cfg.rx_filter))
-		return;
-
-	valid = bfin_read_EMAC_PTP_ISTAT() & RXEL;
-	if (!valid)
-		return;
-
-	shhwtstamps = skb_hwtstamps(skb);
-
-	regval = bfin_read_EMAC_PTP_RXSNAPLO();
-	regval |= (u64)bfin_read_EMAC_PTP_RXSNAPHI() << 32;
-	ns = regval << lp->shift;
-	memset(shhwtstamps, 0, sizeof(*shhwtstamps));
-	shhwtstamps->hwtstamp = ns_to_ktime(ns);
-}
-
-static void bfin_mac_hwtstamp_init(struct net_device *netdev)
-{
-	struct bfin_mac_local *lp = netdev_priv(netdev);
-	u64 addend, ppb;
-	u32 input_clk, phc_clk;
-
-	/* Initialize hardware timer */
-	input_clk = get_sclk();
-	phc_clk = bfin_select_phc_clock(input_clk, &lp->shift);
-	addend = phc_clk * (1ULL << 32);
-	do_div(addend, input_clk);
-	bfin_write_EMAC_PTP_ADDEND((u32)addend);
-
-	lp->addend = addend;
-	ppb = 1000000000ULL * input_clk;
-	do_div(ppb, phc_clk);
-	lp->max_ppb = ppb - 1000000000ULL - 1ULL;
-
-	/* Initialize hwstamp config */
-	lp->stamp_cfg.rx_filter = HWTSTAMP_FILTER_NONE;
-	lp->stamp_cfg.tx_type = HWTSTAMP_TX_OFF;
-}
-
-static u64 bfin_ptp_time_read(struct bfin_mac_local *lp)
-{
-	u64 ns;
-	u32 lo, hi;
-
-	lo = bfin_read_EMAC_PTP_TIMELO();
-	hi = bfin_read_EMAC_PTP_TIMEHI();
-
-	ns = ((u64) hi) << 32;
-	ns |= lo;
-	ns <<= lp->shift;
-
-	return ns;
-}
-
-static void bfin_ptp_time_write(struct bfin_mac_local *lp, u64 ns)
-{
-	u32 hi, lo;
-
-	ns >>= lp->shift;
-	hi = ns >> 32;
-	lo = ns & 0xffffffff;
-
-	bfin_write_EMAC_PTP_TIMELO(lo);
-	bfin_write_EMAC_PTP_TIMEHI(hi);
-}
-
-/* PTP Hardware Clock operations */
-
-static int bfin_ptp_adjfreq(struct ptp_clock_info *ptp, s32 ppb)
-{
-	u64 adj;
-	u32 diff, addend;
-	int neg_adj = 0;
-	struct bfin_mac_local *lp =
-		container_of(ptp, struct bfin_mac_local, caps);
-
-	if (ppb < 0) {
-		neg_adj = 1;
-		ppb = -ppb;
-	}
-	addend = lp->addend;
-	adj = addend;
-	adj *= ppb;
-	diff = div_u64(adj, 1000000000ULL);
-
-	addend = neg_adj ? addend - diff : addend + diff;
-
-	bfin_write_EMAC_PTP_ADDEND(addend);
-
-	return 0;
-}
-
-static int bfin_ptp_adjtime(struct ptp_clock_info *ptp, s64 delta)
-{
-	s64 now;
-	unsigned long flags;
-	struct bfin_mac_local *lp =
-		container_of(ptp, struct bfin_mac_local, caps);
-
-	spin_lock_irqsave(&lp->phc_lock, flags);
-
-	now = bfin_ptp_time_read(lp);
-	now += delta;
-	bfin_ptp_time_write(lp, now);
-
-	spin_unlock_irqrestore(&lp->phc_lock, flags);
-
-	return 0;
-}
-
-static int bfin_ptp_gettime(struct ptp_clock_info *ptp, struct timespec64 *ts)
-{
-	u64 ns;
-	unsigned long flags;
-	struct bfin_mac_local *lp =
-		container_of(ptp, struct bfin_mac_local, caps);
-
-	spin_lock_irqsave(&lp->phc_lock, flags);
-
-	ns = bfin_ptp_time_read(lp);
-
-	spin_unlock_irqrestore(&lp->phc_lock, flags);
-
-	*ts = ns_to_timespec64(ns);
-
-	return 0;
-}
-
-static int bfin_ptp_settime(struct ptp_clock_info *ptp,
-			   const struct timespec64 *ts)
-{
-	u64 ns;
-	unsigned long flags;
-	struct bfin_mac_local *lp =
-		container_of(ptp, struct bfin_mac_local, caps);
-
-	ns = timespec64_to_ns(ts);
-
-	spin_lock_irqsave(&lp->phc_lock, flags);
-
-	bfin_ptp_time_write(lp, ns);
-
-	spin_unlock_irqrestore(&lp->phc_lock, flags);
-
-	return 0;
-}
-
-static int bfin_ptp_enable(struct ptp_clock_info *ptp,
-			  struct ptp_clock_request *rq, int on)
-{
-	return -EOPNOTSUPP;
-}
-
-static struct ptp_clock_info bfin_ptp_caps = {
-	.owner		= THIS_MODULE,
-	.name		= "BF518 clock",
-	.max_adj	= 0,
-	.n_alarm	= 0,
-	.n_ext_ts	= 0,
-	.n_per_out	= 0,
-	.n_pins		= 0,
-	.pps		= 0,
-	.adjfreq	= bfin_ptp_adjfreq,
-	.adjtime	= bfin_ptp_adjtime,
-	.gettime64	= bfin_ptp_gettime,
-	.settime64	= bfin_ptp_settime,
-	.enable		= bfin_ptp_enable,
-};
-
-static int bfin_phc_init(struct net_device *netdev, struct device *dev)
-{
-	struct bfin_mac_local *lp = netdev_priv(netdev);
-
-	lp->caps = bfin_ptp_caps;
-	lp->caps.max_adj = lp->max_ppb;
-	lp->clock = ptp_clock_register(&lp->caps, dev);
-	if (IS_ERR(lp->clock))
-		return PTR_ERR(lp->clock);
-
-	lp->phc_index = ptp_clock_index(lp->clock);
-	spin_lock_init(&lp->phc_lock);
-
-	return 0;
-}
-
-static void bfin_phc_release(struct bfin_mac_local *lp)
-{
-	ptp_clock_unregister(lp->clock);
-}
-
-#else
-# define bfin_mac_hwtstamp_is_none(cfg) 0
-# define bfin_mac_hwtstamp_init(dev)
-# define bfin_mac_hwtstamp_set(dev, ifr) (-EOPNOTSUPP)
-# define bfin_mac_hwtstamp_get(dev, ifr) (-EOPNOTSUPP)
-# define bfin_rx_hwtstamp(dev, skb)
-# define bfin_tx_hwtstamp(dev, skb)
-# define bfin_phc_init(netdev, dev) 0
-# define bfin_phc_release(lp)
-#endif
-
-static inline void _tx_reclaim_skb(void)
-{
-	do {
-		tx_list_head->desc_a.config &= ~DMAEN;
-		tx_list_head->status.status_word = 0;
-		if (tx_list_head->skb) {
-			dev_consume_skb_any(tx_list_head->skb);
-			tx_list_head->skb = NULL;
-		}
-		tx_list_head = tx_list_head->next;
-
-	} while (tx_list_head->status.status_word != 0);
-}
-
-static void tx_reclaim_skb(struct bfin_mac_local *lp)
-{
-	int timeout_cnt = MAX_TIMEOUT_CNT;
-
-	if (tx_list_head->status.status_word != 0)
-		_tx_reclaim_skb();
-
-	if (current_tx_ptr->next == tx_list_head) {
-		while (tx_list_head->status.status_word == 0) {
-			/* slow down polling to avoid too many queue stop. */
-			udelay(10);
-			/* reclaim skb if DMA is not running. */
-			if (!(bfin_read_DMA2_IRQ_STATUS() & DMA_RUN))
-				break;
-			if (timeout_cnt-- < 0)
-				break;
-		}
-
-		if (timeout_cnt >= 0)
-			_tx_reclaim_skb();
-		else
-			netif_stop_queue(lp->ndev);
-	}
-
-	if (current_tx_ptr->next != tx_list_head &&
-		netif_queue_stopped(lp->ndev))
-		netif_wake_queue(lp->ndev);
-
-	if (tx_list_head != current_tx_ptr) {
-		/* shorten the timer interval if tx queue is stopped */
-		if (netif_queue_stopped(lp->ndev))
-			lp->tx_reclaim_timer.expires =
-				jiffies + (TX_RECLAIM_JIFFIES >> 4);
-		else
-			lp->tx_reclaim_timer.expires =
-				jiffies + TX_RECLAIM_JIFFIES;
-
-		mod_timer(&lp->tx_reclaim_timer,
-			lp->tx_reclaim_timer.expires);
-	}
-
-	return;
-}
-
-static void tx_reclaim_skb_timeout(unsigned long lp)
-{
-	tx_reclaim_skb((struct bfin_mac_local *)lp);
-}
-
-static int bfin_mac_hard_start_xmit(struct sk_buff *skb,
-				struct net_device *dev)
-{
-	struct bfin_mac_local *lp = netdev_priv(dev);
-	u16 *data;
-	u32 data_align = (unsigned long)(skb->data) & 0x3;
-
-	current_tx_ptr->skb = skb;
-
-	if (data_align == 0x2) {
-		/* move skb->data to current_tx_ptr payload */
-		data = (u16 *)(skb->data) - 1;
-		*data = (u16)(skb->len);
-		/*
-		 * When transmitting an Ethernet packet, the PTP_TSYNC module requires
-		 * a DMA_Length_Word field associated with the packet. The lower 12 bits
-		 * of this field are the length of the packet payload in bytes and the higher
-		 * 4 bits are the timestamping enable field.
-		 */
-		if (skb_shinfo(skb)->tx_flags & SKBTX_HW_TSTAMP)
-			*data |= 0x1000;
-
-		current_tx_ptr->desc_a.start_addr = (u32)data;
-		/* this is important! */
-		blackfin_dcache_flush_range((u32)data,
-				(u32)((u8 *)data + skb->len + 4));
-	} else {
-		*((u16 *)(current_tx_ptr->packet)) = (u16)(skb->len);
-		/* enable timestamping for the sent packet */
-		if (skb_shinfo(skb)->tx_flags & SKBTX_HW_TSTAMP)
-			*((u16 *)(current_tx_ptr->packet)) |= 0x1000;
-		memcpy((u8 *)(current_tx_ptr->packet + 2), skb->data,
-			skb->len);
-		current_tx_ptr->desc_a.start_addr =
-			(u32)current_tx_ptr->packet;
-		blackfin_dcache_flush_range(
-			(u32)current_tx_ptr->packet,
-			(u32)(current_tx_ptr->packet + skb->len + 2));
-	}
-
-	/* make sure the internal data buffers in the core are drained
-	 * so that the DMA descriptors are completely written when the
-	 * DMA engine goes to fetch them below
-	 */
-	SSYNC();
-
-	/* always clear status buffer before start tx dma */
-	current_tx_ptr->status.status_word = 0;
-
-	/* enable this packet's dma */
-	current_tx_ptr->desc_a.config |= DMAEN;
-
-	/* tx dma is running, just return */
-	if (bfin_read_DMA2_IRQ_STATUS() & DMA_RUN)
-		goto out;
-
-	/* tx dma is not running */
-	bfin_write_DMA2_NEXT_DESC_PTR(&(current_tx_ptr->desc_a));
-	/* dma enabled, read from memory, size is 6 */
-	bfin_write_DMA2_CONFIG(current_tx_ptr->desc_a.config);
-	/* Turn on the EMAC tx */
-	bfin_write_EMAC_OPMODE(bfin_read_EMAC_OPMODE() | TE);
-
-out:
-	bfin_tx_hwtstamp(dev, skb);
-
-	current_tx_ptr = current_tx_ptr->next;
-	dev->stats.tx_packets++;
-	dev->stats.tx_bytes += (skb->len);
-
-	tx_reclaim_skb(lp);
-
-	return NETDEV_TX_OK;
-}
-
-#define IP_HEADER_OFF  0
-#define RX_ERROR_MASK (RX_LONG | RX_ALIGN | RX_CRC | RX_LEN | \
-	RX_FRAG | RX_ADDR | RX_DMAO | RX_PHY | RX_LATE | RX_RANGE)
-
-static void bfin_mac_rx(struct bfin_mac_local *lp)
-{
-	struct net_device *dev = lp->ndev;
-	struct sk_buff *skb, *new_skb;
-	unsigned short len;
-#if defined(BFIN_MAC_CSUM_OFFLOAD)
-	unsigned int i;
-	unsigned char fcs[ETH_FCS_LEN + 1];
-#endif
-
-	/* check if frame status word reports an error condition
-	 * we which case we simply drop the packet
-	 */
-	if (current_rx_ptr->status.status_word & RX_ERROR_MASK) {
-		netdev_notice(dev, "rx: receive error - packet dropped\n");
-		dev->stats.rx_dropped++;
-		goto out;
-	}
-
-	/* allocate a new skb for next time receive */
-	skb = current_rx_ptr->skb;
-
-	new_skb = netdev_alloc_skb(dev, PKT_BUF_SZ + NET_IP_ALIGN);
-	if (!new_skb) {
-		dev->stats.rx_dropped++;
-		goto out;
-	}
-	/* reserve 2 bytes for RXDWA padding */
-	skb_reserve(new_skb, NET_IP_ALIGN);
-	/* Invalidate the data cache of skb->data range when it is write back
-	 * cache. It will prevent overwriting the new data from DMA
-	 */
-	blackfin_dcache_invalidate_range((unsigned long)new_skb->head,
-					 (unsigned long)new_skb->end);
-
-	current_rx_ptr->skb = new_skb;
-	current_rx_ptr->desc_a.start_addr = (unsigned long)new_skb->data - 2;
-
-	len = (unsigned short)(current_rx_ptr->status.status_word & RX_FRLEN);
-	/* Deduce Ethernet FCS length from Ethernet payload length */
-	len -= ETH_FCS_LEN;
-	skb_put(skb, len);
-
-	skb->protocol = eth_type_trans(skb, dev);
-
-	bfin_rx_hwtstamp(dev, skb);
-
-#if defined(BFIN_MAC_CSUM_OFFLOAD)
-	/* Checksum offloading only works for IPv4 packets with the standard IP header
-	 * length of 20 bytes, because the blackfin MAC checksum calculation is
-	 * based on that assumption. We must NOT use the calculated checksum if our
-	 * IP version or header break that assumption.
-	 */
-	if (skb->data[IP_HEADER_OFF] == 0x45) {
-		skb->csum = current_rx_ptr->status.ip_payload_csum;
-		/*
-		 * Deduce Ethernet FCS from hardware generated IP payload checksum.
-		 * IP checksum is based on 16-bit one's complement algorithm.
-		 * To deduce a value from checksum is equal to add its inversion.
-		 * If the IP payload len is odd, the inversed FCS should also
-		 * begin from odd address and leave first byte zero.
-		 */
-		if (skb->len % 2) {
-			fcs[0] = 0;
-			for (i = 0; i < ETH_FCS_LEN; i++)
-				fcs[i + 1] = ~skb->data[skb->len + i];
-			skb->csum = csum_partial(fcs, ETH_FCS_LEN + 1, skb->csum);
-		} else {
-			for (i = 0; i < ETH_FCS_LEN; i++)
-				fcs[i] = ~skb->data[skb->len + i];
-			skb->csum = csum_partial(fcs, ETH_FCS_LEN, skb->csum);
-		}
-		skb->ip_summed = CHECKSUM_COMPLETE;
-	}
-#endif
-
-	napi_gro_receive(&lp->napi, skb);
-
-	dev->stats.rx_packets++;
-	dev->stats.rx_bytes += len;
-out:
-	current_rx_ptr->status.status_word = 0x00000000;
-	current_rx_ptr = current_rx_ptr->next;
-}
-
-static int bfin_mac_poll(struct napi_struct *napi, int budget)
-{
-	int i = 0;
-	struct bfin_mac_local *lp = container_of(napi,
-						 struct bfin_mac_local,
-						 napi);
-
-	while (current_rx_ptr->status.status_word != 0 && i < budget) {
-		bfin_mac_rx(lp);
-		i++;
-	}
-
-	if (i < budget) {
-		napi_complete_done(napi, i);
-		if (test_and_clear_bit(BFIN_MAC_RX_IRQ_DISABLED, &lp->flags))
-			enable_irq(IRQ_MAC_RX);
-	}
-
-	return i;
-}
-
-/* interrupt routine to handle rx and error signal */
-static irqreturn_t bfin_mac_interrupt(int irq, void *dev_id)
-{
-	struct bfin_mac_local *lp = netdev_priv(dev_id);
-	u32 status;
-
-	status = bfin_read_DMA1_IRQ_STATUS();
-
-	bfin_write_DMA1_IRQ_STATUS(status | DMA_DONE | DMA_ERR);
-	if (status & DMA_DONE) {
-		disable_irq_nosync(IRQ_MAC_RX);
-		set_bit(BFIN_MAC_RX_IRQ_DISABLED, &lp->flags);
-		napi_schedule(&lp->napi);
-	}
-
-	return IRQ_HANDLED;
-}
-
-#ifdef CONFIG_NET_POLL_CONTROLLER
-static void bfin_mac_poll_controller(struct net_device *dev)
-{
-	struct bfin_mac_local *lp = netdev_priv(dev);
-
-	bfin_mac_interrupt(IRQ_MAC_RX, dev);
-	tx_reclaim_skb(lp);
-}
-#endif				/* CONFIG_NET_POLL_CONTROLLER */
-
-static void bfin_mac_disable(void)
-{
-	unsigned int opmode;
-
-	opmode = bfin_read_EMAC_OPMODE();
-	opmode &= (~RE);
-	opmode &= (~TE);
-	/* Turn off the EMAC */
-	bfin_write_EMAC_OPMODE(opmode);
-}
-
-/*
- * Enable Interrupts, Receive, and Transmit
- */
-static int bfin_mac_enable(struct phy_device *phydev)
-{
-	int ret;
-	u32 opmode;
-
-	pr_debug("%s\n", __func__);
-
-	/* Set RX DMA */
-	bfin_write_DMA1_NEXT_DESC_PTR(&(rx_list_head->desc_a));
-	bfin_write_DMA1_CONFIG(rx_list_head->desc_a.config);
-
-	/* Wait MII done */
-	ret = bfin_mdio_poll();
-	if (ret)
-		return ret;
-
-	/* We enable only RX here */
-	/* ASTP   : Enable Automatic Pad Stripping
-	   PR     : Promiscuous Mode for test
-	   PSF    : Receive frames with total length less than 64 bytes.
-	   FDMODE : Full Duplex Mode
-	   LB     : Internal Loopback for test
-	   RE     : Receiver Enable */
-	opmode = bfin_read_EMAC_OPMODE();
-	if (opmode & FDMODE)
-		opmode |= PSF;
-	else
-		opmode |= DRO | DC | PSF;
-	opmode |= RE;
-
-	if (phydev->interface == PHY_INTERFACE_MODE_RMII) {
-		opmode |= RMII; /* For Now only 100MBit are supported */
-#if defined(CONFIG_BF537) || defined(CONFIG_BF536)
-		if (__SILICON_REVISION__ < 3) {
-			/*
-			 * This isn't publicly documented (fun times!), but in
-			 * silicon <=0.2, the RX and TX pins are clocked together.
-			 * So in order to recv, we must enable the transmit side
-			 * as well.  This will cause a spurious TX interrupt too,
-			 * but we can easily consume that.
-			 */
-			opmode |= TE;
-		}
-#endif
-	}
-
-	/* Turn on the EMAC rx */
-	bfin_write_EMAC_OPMODE(opmode);
-
-	return 0;
-}
-
-/* Our watchdog timed out. Called by the networking layer */
-static void bfin_mac_timeout(struct net_device *dev)
-{
-	struct bfin_mac_local *lp = netdev_priv(dev);
-
-	pr_debug("%s: %s\n", dev->name, __func__);
-
-	bfin_mac_disable();
-
-	del_timer(&lp->tx_reclaim_timer);
-
-	/* reset tx queue and free skb */
-	while (tx_list_head != current_tx_ptr) {
-		tx_list_head->desc_a.config &= ~DMAEN;
-		tx_list_head->status.status_word = 0;
-		if (tx_list_head->skb) {
-			dev_kfree_skb(tx_list_head->skb);
-			tx_list_head->skb = NULL;
-		}
-		tx_list_head = tx_list_head->next;
-	}
-
-	if (netif_queue_stopped(dev))
-		netif_wake_queue(dev);
-
-	bfin_mac_enable(dev->phydev);
-
-	/* We can accept TX packets again */
-	netif_trans_update(dev); /* prevent tx timeout */
-}
-
-static void bfin_mac_multicast_hash(struct net_device *dev)
-{
-	u32 emac_hashhi, emac_hashlo;
-	struct netdev_hw_addr *ha;
-	u32 crc;
-
-	emac_hashhi = emac_hashlo = 0;
-
-	netdev_for_each_mc_addr(ha, dev) {
-		crc = ether_crc(ETH_ALEN, ha->addr);
-		crc >>= 26;
-
-		if (crc & 0x20)
-			emac_hashhi |= 1 << (crc & 0x1f);
-		else
-			emac_hashlo |= 1 << (crc & 0x1f);
-	}
-
-	bfin_write_EMAC_HASHHI(emac_hashhi);
-	bfin_write_EMAC_HASHLO(emac_hashlo);
-}
-
-/*
- * This routine will, depending on the values passed to it,
- * either make it accept multicast packets, go into
- * promiscuous mode (for TCPDUMP and cousins) or accept
- * a select set of multicast packets
- */
-static void bfin_mac_set_multicast_list(struct net_device *dev)
-{
-	u32 sysctl;
-
-	if (dev->flags & IFF_PROMISC) {
-		netdev_info(dev, "set promisc mode\n");
-		sysctl = bfin_read_EMAC_OPMODE();
-		sysctl |= PR;
-		bfin_write_EMAC_OPMODE(sysctl);
-	} else if (dev->flags & IFF_ALLMULTI) {
-		/* accept all multicast */
-		sysctl = bfin_read_EMAC_OPMODE();
-		sysctl |= PAM;
-		bfin_write_EMAC_OPMODE(sysctl);
-	} else if (!netdev_mc_empty(dev)) {
-		/* set up multicast hash table */
-		sysctl = bfin_read_EMAC_OPMODE();
-		sysctl |= HM;
-		bfin_write_EMAC_OPMODE(sysctl);
-		bfin_mac_multicast_hash(dev);
-	} else {
-		/* clear promisc or multicast mode */
-		sysctl = bfin_read_EMAC_OPMODE();
-		sysctl &= ~(RAF | PAM);
-		bfin_write_EMAC_OPMODE(sysctl);
-	}
-}
-
-static int bfin_mac_ioctl(struct net_device *netdev, struct ifreq *ifr, int cmd)
-{
-	if (!netif_running(netdev))
-		return -EINVAL;
-
-	switch (cmd) {
-	case SIOCSHWTSTAMP:
-		return bfin_mac_hwtstamp_set(netdev, ifr);
-	case SIOCGHWTSTAMP:
-		return bfin_mac_hwtstamp_get(netdev, ifr);
-	default:
-		if (netdev->phydev)
-			return phy_mii_ioctl(netdev->phydev, ifr, cmd);
-		else
-			return -EOPNOTSUPP;
-	}
-}
-
-/*
- * this puts the device in an inactive state
- */
-static void bfin_mac_shutdown(struct net_device *dev)
-{
-	/* Turn off the EMAC */
-	bfin_write_EMAC_OPMODE(0x00000000);
-	/* Turn off the EMAC RX DMA */
-	bfin_write_DMA1_CONFIG(0x0000);
-	bfin_write_DMA2_CONFIG(0x0000);
-}
-
-/*
- * Open and Initialize the interface
- *
- * Set up everything, reset the card, etc..
- */
-static int bfin_mac_open(struct net_device *dev)
-{
-	struct bfin_mac_local *lp = netdev_priv(dev);
-	int ret;
-	pr_debug("%s: %s\n", dev->name, __func__);
-
-	/*
-	 * Check that the address is valid.  If its not, refuse
-	 * to bring the device up.  The user must specify an
-	 * address using ifconfig eth0 hw ether xx:xx:xx:xx:xx:xx
-	 */
-	if (!is_valid_ether_addr(dev->dev_addr)) {
-		netdev_warn(dev, "no valid ethernet hw addr\n");
-		return -EINVAL;
-	}
-
-	/* initial rx and tx list */
-	ret = desc_list_init(dev);
-	if (ret)
-		return ret;
-
-	phy_start(dev->phydev);
-	setup_system_regs(dev);
-	setup_mac_addr(dev->dev_addr);
-
-	bfin_mac_disable();
-	ret = bfin_mac_enable(dev->phydev);
-	if (ret)
-		return ret;
-	pr_debug("hardware init finished\n");
-
-	napi_enable(&lp->napi);
-	netif_start_queue(dev);
-	netif_carrier_on(dev);
-
-	return 0;
-}
-
-/*
- * this makes the board clean up everything that it can
- * and not talk to the outside world.   Caused by
- * an 'ifconfig ethX down'
- */
-static int bfin_mac_close(struct net_device *dev)
-{
-	struct bfin_mac_local *lp = netdev_priv(dev);
-	pr_debug("%s: %s\n", dev->name, __func__);
-
-	netif_stop_queue(dev);
-	napi_disable(&lp->napi);
-	netif_carrier_off(dev);
-
-	phy_stop(dev->phydev);
-	phy_write(dev->phydev, MII_BMCR, BMCR_PDOWN);
-
-	/* clear everything */
-	bfin_mac_shutdown(dev);
-
-	/* free the rx/tx buffers */
-	desc_list_free();
-
-	return 0;
-}
-
-static const struct net_device_ops bfin_mac_netdev_ops = {
-	.ndo_open		= bfin_mac_open,
-	.ndo_stop		= bfin_mac_close,
-	.ndo_start_xmit		= bfin_mac_hard_start_xmit,
-	.ndo_set_mac_address	= bfin_mac_set_mac_address,
-	.ndo_tx_timeout		= bfin_mac_timeout,
-	.ndo_set_rx_mode	= bfin_mac_set_multicast_list,
-	.ndo_do_ioctl           = bfin_mac_ioctl,
-	.ndo_validate_addr	= eth_validate_addr,
-#ifdef CONFIG_NET_POLL_CONTROLLER
-	.ndo_poll_controller	= bfin_mac_poll_controller,
-#endif
-};
-
-static int bfin_mac_probe(struct platform_device *pdev)
-{
-	struct net_device *ndev;
-	struct bfin_mac_local *lp;
-	struct platform_device *pd;
-	struct bfin_mii_bus_platform_data *mii_bus_data;
-	int rc;
-
-	ndev = alloc_etherdev(sizeof(struct bfin_mac_local));
-	if (!ndev)
-		return -ENOMEM;
-
-	SET_NETDEV_DEV(ndev, &pdev->dev);
-	platform_set_drvdata(pdev, ndev);
-	lp = netdev_priv(ndev);
-	lp->ndev = ndev;
-
-	/* Grab the MAC address in the MAC */
-	*(__le32 *) (&(ndev->dev_addr[0])) = cpu_to_le32(bfin_read_EMAC_ADDRLO());
-	*(__le16 *) (&(ndev->dev_addr[4])) = cpu_to_le16((u16) bfin_read_EMAC_ADDRHI());
-
-	/* probe mac */
-	/*todo: how to probe? which is revision_register */
-	bfin_write_EMAC_ADDRLO(0x12345678);
-	if (bfin_read_EMAC_ADDRLO() != 0x12345678) {
-		dev_err(&pdev->dev, "Cannot detect Blackfin on-chip ethernet MAC controller!\n");
-		rc = -ENODEV;
-		goto out_err_probe_mac;
-	}
-
-
-	/*
-	 * Is it valid? (Did bootloader initialize it?)
-	 * Grab the MAC from the board somehow
-	 * this is done in the arch/blackfin/mach-bfxxx/boards/eth_mac.c
-	 */
-	if (!is_valid_ether_addr(ndev->dev_addr)) {
-		if (bfin_get_ether_addr(ndev->dev_addr) ||
-		     !is_valid_ether_addr(ndev->dev_addr)) {
-			/* Still not valid, get a random one */
-			netdev_warn(ndev, "Setting Ethernet MAC to a random one\n");
-			eth_hw_addr_random(ndev);
-		}
-	}
-
-	setup_mac_addr(ndev->dev_addr);
-
-	if (!dev_get_platdata(&pdev->dev)) {
-		dev_err(&pdev->dev, "Cannot get platform device bfin_mii_bus!\n");
-		rc = -ENODEV;
-		goto out_err_probe_mac;
-	}
-	pd = dev_get_platdata(&pdev->dev);
-	lp->mii_bus = platform_get_drvdata(pd);
-	if (!lp->mii_bus) {
-		dev_err(&pdev->dev, "Cannot get mii_bus!\n");
-		rc = -ENODEV;
-		goto out_err_probe_mac;
-	}
-	lp->mii_bus->priv = ndev;
-	mii_bus_data = dev_get_platdata(&pd->dev);
-
-	rc = mii_probe(ndev, mii_bus_data->phy_mode);
-	if (rc) {
-		dev_err(&pdev->dev, "MII Probe failed!\n");
-		goto out_err_mii_probe;
-	}
-
-	lp->vlan1_mask = ETH_P_8021Q | mii_bus_data->vlan1_mask;
-	lp->vlan2_mask = ETH_P_8021Q | mii_bus_data->vlan2_mask;
-
-	ndev->netdev_ops = &bfin_mac_netdev_ops;
-	ndev->ethtool_ops = &bfin_mac_ethtool_ops;
-
-	init_timer(&lp->tx_reclaim_timer);
-	lp->tx_reclaim_timer.data = (unsigned long)lp;
-	lp->tx_reclaim_timer.function = tx_reclaim_skb_timeout;
-
-	lp->flags = 0;
-	netif_napi_add(ndev, &lp->napi, bfin_mac_poll, CONFIG_BFIN_RX_DESC_NUM);
-
-	spin_lock_init(&lp->lock);
-
-	/* now, enable interrupts */
-	/* register irq handler */
-	rc = request_irq(IRQ_MAC_RX, bfin_mac_interrupt,
-			0, "EMAC_RX", ndev);
-	if (rc) {
-		dev_err(&pdev->dev, "Cannot request Blackfin MAC RX IRQ!\n");
-		rc = -EBUSY;
-		goto out_err_request_irq;
-	}
-
-	rc = register_netdev(ndev);
-	if (rc) {
-		dev_err(&pdev->dev, "Cannot register net device!\n");
-		goto out_err_reg_ndev;
-	}
-
-	bfin_mac_hwtstamp_init(ndev);
-	rc = bfin_phc_init(ndev, &pdev->dev);
-	if (rc) {
-		dev_err(&pdev->dev, "Cannot register PHC device!\n");
-		goto out_err_phc;
-	}
-
-	/* now, print out the card info, in a short format.. */
-	netdev_info(ndev, "%s, Version %s\n", DRV_DESC, DRV_VERSION);
-
-	return 0;
-
-out_err_phc:
-out_err_reg_ndev:
-	free_irq(IRQ_MAC_RX, ndev);
-out_err_request_irq:
-	netif_napi_del(&lp->napi);
-out_err_mii_probe:
-	mdiobus_unregister(lp->mii_bus);
-	mdiobus_free(lp->mii_bus);
-out_err_probe_mac:
-	free_netdev(ndev);
-
-	return rc;
-}
-
-static int bfin_mac_remove(struct platform_device *pdev)
-{
-	struct net_device *ndev = platform_get_drvdata(pdev);
-	struct bfin_mac_local *lp = netdev_priv(ndev);
-
-	bfin_phc_release(lp);
-
-	lp->mii_bus->priv = NULL;
-
-	unregister_netdev(ndev);
-
-	netif_napi_del(&lp->napi);
-
-	free_irq(IRQ_MAC_RX, ndev);
-
-	free_netdev(ndev);
-
-	return 0;
-}
-
-#ifdef CONFIG_PM
-static int bfin_mac_suspend(struct platform_device *pdev, pm_message_t mesg)
-{
-	struct net_device *net_dev = platform_get_drvdata(pdev);
-	struct bfin_mac_local *lp = netdev_priv(net_dev);
-
-	if (lp->wol) {
-		bfin_write_EMAC_OPMODE((bfin_read_EMAC_OPMODE() & ~TE) | RE);
-		bfin_write_EMAC_WKUP_CTL(MPKE);
-		enable_irq_wake(IRQ_MAC_WAKEDET);
-	} else {
-		if (netif_running(net_dev))
-			bfin_mac_close(net_dev);
-	}
-
-	return 0;
-}
-
-static int bfin_mac_resume(struct platform_device *pdev)
-{
-	struct net_device *net_dev = platform_get_drvdata(pdev);
-	struct bfin_mac_local *lp = netdev_priv(net_dev);
-
-	if (lp->wol) {
-		bfin_write_EMAC_OPMODE(bfin_read_EMAC_OPMODE() | TE);
-		bfin_write_EMAC_WKUP_CTL(0);
-		disable_irq_wake(IRQ_MAC_WAKEDET);
-	} else {
-		if (netif_running(net_dev))
-			bfin_mac_open(net_dev);
-	}
-
-	return 0;
-}
-#else
-#define bfin_mac_suspend NULL
-#define bfin_mac_resume NULL
-#endif	/* CONFIG_PM */
-
-static int bfin_mii_bus_probe(struct platform_device *pdev)
-{
-	struct mii_bus *miibus;
-	struct bfin_mii_bus_platform_data *mii_bus_pd;
-	const unsigned short *pin_req;
-	int rc, i;
-
-	mii_bus_pd = dev_get_platdata(&pdev->dev);
-	if (!mii_bus_pd) {
-		dev_err(&pdev->dev, "No peripherals in platform data!\n");
-		return -EINVAL;
-	}
-
-	/*
-	 * We are setting up a network card,
-	 * so set the GPIO pins to Ethernet mode
-	 */
-	pin_req = mii_bus_pd->mac_peripherals;
-	rc = peripheral_request_list(pin_req, KBUILD_MODNAME);
-	if (rc) {
-		dev_err(&pdev->dev, "Requesting peripherals failed!\n");
-		return rc;
-	}
-
-	rc = -ENOMEM;
-	miibus = mdiobus_alloc();
-	if (miibus == NULL)
-		goto out_err_alloc;
-	miibus->read = bfin_mdiobus_read;
-	miibus->write = bfin_mdiobus_write;
-
-	miibus->parent = &pdev->dev;
-	miibus->name = "bfin_mii_bus";
-	miibus->phy_mask = mii_bus_pd->phy_mask;
-
-	snprintf(miibus->id, MII_BUS_ID_SIZE, "%s-%x",
-		pdev->name, pdev->id);
-
-	rc = clamp(mii_bus_pd->phydev_number, 0, PHY_MAX_ADDR);
-	if (rc != mii_bus_pd->phydev_number)
-		dev_err(&pdev->dev, "Invalid number (%i) of phydevs\n",
-			mii_bus_pd->phydev_number);
-	for (i = 0; i < rc; ++i) {
-		unsigned short phyaddr = mii_bus_pd->phydev_data[i].addr;
-		if (phyaddr < PHY_MAX_ADDR)
-			miibus->irq[phyaddr] = mii_bus_pd->phydev_data[i].irq;
-		else
-			dev_err(&pdev->dev,
-				"Invalid PHY address %i for phydev %i\n",
-				phyaddr, i);
-	}
-
-	rc = mdiobus_register(miibus);
-	if (rc) {
-		dev_err(&pdev->dev, "Cannot register MDIO bus!\n");
-		goto out_err_irq_alloc;
-	}
-
-	platform_set_drvdata(pdev, miibus);
-	return 0;
-
-out_err_irq_alloc:
-	mdiobus_free(miibus);
-out_err_alloc:
-	peripheral_free_list(pin_req);
-
-	return rc;
-}
-
-static int bfin_mii_bus_remove(struct platform_device *pdev)
-{
-	struct mii_bus *miibus = platform_get_drvdata(pdev);
-	struct bfin_mii_bus_platform_data *mii_bus_pd =
-		dev_get_platdata(&pdev->dev);
-
-	mdiobus_unregister(miibus);
-	mdiobus_free(miibus);
-	peripheral_free_list(mii_bus_pd->mac_peripherals);
-
-	return 0;
-}
-
-static struct platform_driver bfin_mii_bus_driver = {
-	.probe = bfin_mii_bus_probe,
-	.remove = bfin_mii_bus_remove,
-	.driver = {
-		.name = "bfin_mii_bus",
-	},
-};
-
-static struct platform_driver bfin_mac_driver = {
-	.probe = bfin_mac_probe,
-	.remove = bfin_mac_remove,
-	.resume = bfin_mac_resume,
-	.suspend = bfin_mac_suspend,
-	.driver = {
-		.name = KBUILD_MODNAME,
-	},
-};
-
-static struct platform_driver * const drivers[] = {
-	&bfin_mii_bus_driver,
-	&bfin_mac_driver,
-};
-
-static int __init bfin_mac_init(void)
-{
-	return platform_register_drivers(drivers, ARRAY_SIZE(drivers));
-}
-
-module_init(bfin_mac_init);
-
-static void __exit bfin_mac_cleanup(void)
-{
-	platform_unregister_drivers(drivers, ARRAY_SIZE(drivers));
-}
-
-module_exit(bfin_mac_cleanup);
-
diff --git a/drivers/net/ethernet/adi/bfin_mac.h b/drivers/net/ethernet/adi/bfin_mac.h
deleted file mode 100644
index 4ad5b9be3f84..000000000000
--- a/drivers/net/ethernet/adi/bfin_mac.h
+++ /dev/null
@@ -1,104 +0,0 @@
-/*
- * Blackfin On-Chip MAC Driver
- *
- * Copyright 2004-2007 Analog Devices Inc.
- *
- * Enter bugs at http://blackfin.uclinux.org/
- *
- * Licensed under the GPL-2 or later.
- */
-#ifndef _BFIN_MAC_H_
-#define _BFIN_MAC_H_
-
-#include <linux/net_tstamp.h>
-#include <linux/ptp_clock_kernel.h>
-#include <linux/timer.h>
-#include <linux/etherdevice.h>
-#include <linux/bfin_mac.h>
-
-/*
- * Disable hardware checksum for bug #5600 if writeback cache is
- * enabled. Otherwize, corrupted RX packet will be sent up stack
- * without error mark.
- */
-#ifndef CONFIG_BFIN_EXTMEM_WRITEBACK
-#define BFIN_MAC_CSUM_OFFLOAD
-#endif
-
-#define TX_RECLAIM_JIFFIES (HZ / 5)
-#define BFIN_MAC_RX_IRQ_DISABLED	1
-
-struct dma_descriptor {
-	struct dma_descriptor *next_dma_desc;
-	unsigned long start_addr;
-	unsigned short config;
-	unsigned short x_count;
-};
-
-struct status_area_rx {
-#if defined(BFIN_MAC_CSUM_OFFLOAD)
-	unsigned short ip_hdr_csum;	/* ip header checksum */
-	/* ip payload(udp or tcp or others) checksum */
-	unsigned short ip_payload_csum;
-#endif
-	unsigned long status_word;	/* the frame status word */
-};
-
-struct status_area_tx {
-	unsigned long status_word;	/* the frame status word */
-};
-
-/* use two descriptors for a packet */
-struct net_dma_desc_rx {
-	struct net_dma_desc_rx *next;
-	struct sk_buff *skb;
-	struct dma_descriptor desc_a;
-	struct dma_descriptor desc_b;
-	struct status_area_rx status;
-};
-
-/* use two descriptors for a packet */
-struct net_dma_desc_tx {
-	struct net_dma_desc_tx *next;
-	struct sk_buff *skb;
-	struct dma_descriptor desc_a;
-	struct dma_descriptor desc_b;
-	unsigned char packet[1560];
-	struct status_area_tx status;
-};
-
-struct bfin_mac_local {
-	spinlock_t lock;
-
-	int wol;		/* Wake On Lan */
-	int irq_wake_requested;
-	struct timer_list tx_reclaim_timer;
-	struct net_device *ndev;
-	struct napi_struct napi;
-	unsigned long flags;
-
-	/* Data for EMAC_VLAN1 regs */
-	u16 vlan1_mask, vlan2_mask;
-
-	/* MII and PHY stuffs */
-	int old_link;          /* used by bf537_adjust_link */
-	int old_speed;
-	int old_duplex;
-
-	struct mii_bus *mii_bus;
-
-#if defined(CONFIG_BFIN_MAC_USE_HWSTAMP)
-	u32 addend;
-	unsigned int shift;
-	s32 max_ppb;
-	struct hwtstamp_config stamp_cfg;
-	struct ptp_clock_info caps;
-	struct ptp_clock *clock;
-	int phc_index;
-	spinlock_t phc_lock; /* protects time lo/hi registers */
-#endif
-};
-
-int bfin_get_ether_addr(char *addr);
-
-#endif
diff --git a/drivers/net/ethernet/aeroflex/Kconfig b/drivers/net/ethernet/aeroflex/Kconfig
index 4f4a8d78fd54..ea4be67f787b 100644
--- a/drivers/net/ethernet/aeroflex/Kconfig
+++ b/drivers/net/ethernet/aeroflex/Kconfig
@@ -1,3 +1,4 @@
+# SPDX-License-Identifier: GPL-2.0-only
 #
 # Aeroflex Gaisler network device configuration
 #
@@ -7,5 +8,5 @@ config GRETH
 	depends on SPARC
 	select PHYLIB
 	select CRC32
-	---help---
+	help
 	  Say Y here if you want to use the Aeroflex Gaisler GRETH Ethernet MAC.
diff --git a/drivers/net/ethernet/aeroflex/Makefile b/drivers/net/ethernet/aeroflex/Makefile
index 6e62a679282f..1b18ef0a5389 100644
--- a/drivers/net/ethernet/aeroflex/Makefile
+++ b/drivers/net/ethernet/aeroflex/Makefile
@@ -1,3 +1,4 @@
+# SPDX-License-Identifier: GPL-2.0-only
 #
 # Makefile for the Aeroflex Gaisler network device drivers.
 #
diff --git a/drivers/net/ethernet/aeroflex/greth.c b/drivers/net/ethernet/aeroflex/greth.c
index d8e133ced7b8..a593adc16c78 100644
--- a/drivers/net/ethernet/aeroflex/greth.c
+++ b/drivers/net/ethernet/aeroflex/greth.c
@@ -1,3 +1,4 @@
+// SPDX-License-Identifier: GPL-2.0-or-later
 /*
  * Aeroflex Gaisler GRETH 10/100/1G Ethernet MAC.
  *
@@ -7,16 +8,11 @@
  * available in the GRLIB VHDL IP core library.
  *
  * Full documentation of both cores can be found here:
- * http://www.gaisler.com/products/grlib/grip.pdf
+ * https://www.gaisler.com/products/grlib/grip.pdf
  *
  * The Gigabit version supports scatter/gather DMA, any alignment of
  * buffers and checksum offloading.
  *
- * This program is free software; you can redistribute it and/or modify it
- * under the terms of the GNU General Public License as published by the
- * Free Software Foundation; either version 2 of the License, or (at your
- * option) any later version.
- *
  * Contributors: Kristoffer Glembo
  *               Daniel Hellstrom
  *               Marko Isomaki
@@ -33,9 +29,9 @@
 #include <linux/io.h>
 #include <linux/crc32.h>
 #include <linux/mii.h>
-#include <linux/of_device.h>
+#include <linux/of.h>
 #include <linux/of_net.h>
-#include <linux/of_platform.h>
+#include <linux/platform_device.h>
 #include <linux/slab.h>
 #include <asm/cacheflush.h>
 #include <asm/byteorder.h>
@@ -114,7 +110,7 @@ static void greth_print_tx_packet(struct sk_buff *skb)
 
 		print_hex_dump(KERN_DEBUG, "TX: ", DUMP_PREFIX_OFFSET, 16, 1,
 			       skb_frag_address(&skb_shinfo(skb)->frags[i]),
-			       skb_shinfo(skb)->frags[i].size, true);
+			       skb_frag_size(&skb_shinfo(skb)->frags[i]), true);
 	}
 }
 
@@ -262,6 +258,7 @@ static int greth_init_rings(struct greth_private *greth)
 			if (dma_mapping_error(greth->dev, dma_addr)) {
 				if (netif_msg_ifup(greth))
 					dev_err(greth->dev, "Could not create initial DMA mapping\n");
+				dev_kfree_skb(skb);
 				goto cleanup;
 			}
 			greth->rx_skbuff[i] = skb;
@@ -487,7 +484,7 @@ greth_start_xmit_gbit(struct sk_buff *skb, struct net_device *dev)
 
 	if (unlikely(skb->len > MAX_FRAME_SIZE)) {
 		dev->stats.tx_errors++;
-		goto out;
+		goto len_error;
 	}
 
 	/* Save skb pointer. */
@@ -578,6 +575,7 @@ frag_map_error:
 map_error:
 	if (net_ratelimit())
 		dev_warn(greth->dev, "Could not create TX DMA mapping\n");
+len_error:
 	dev_kfree_skb(skb);
 out:
 	return err;
@@ -613,7 +611,6 @@ static irqreturn_t greth_interrupt(int irq, void *dev_id)
 		napi_schedule(&greth->napi);
 	}
 
-	mmiowb();
 	spin_unlock(&greth->devlock);
 
 	return retval;
@@ -807,7 +804,8 @@ static int greth_rx(struct net_device *dev, int limit)
 				if (netif_msg_pktdata(greth))
 					greth_print_rx_packet(phys_to_virt(dma_addr), pkt_len);
 
-				memcpy(skb_put(skb, pkt_len), phys_to_virt(dma_addr), pkt_len);
+				skb_put_data(skb, phys_to_virt(dma_addr),
+					     pkt_len);
 
 				skb->protocol = eth_type_trans(skb, dev);
 				dev->stats.rx_bytes += pkt_len;
@@ -1029,7 +1027,7 @@ static int greth_set_mac_add(struct net_device *dev, void *p)
 	if (!is_valid_ether_addr(addr->sa_data))
 		return -EADDRNOTAVAIL;
 
-	memcpy(dev->dev_addr, addr->sa_data, dev->addr_len);
+	eth_hw_addr_set(dev, addr->sa_data);
 	GRETH_REGSAVE(regs->esa_msb, dev->dev_addr[0] << 8 | dev->dev_addr[1]);
 	GRETH_REGSAVE(regs->esa_lsb, dev->dev_addr[2] << 24 | dev->dev_addr[3] << 16 |
 		      dev->dev_addr[4] << 8 | dev->dev_addr[5]);
@@ -1116,11 +1114,9 @@ static void greth_get_drvinfo(struct net_device *dev, struct ethtool_drvinfo *in
 {
 	struct greth_private *greth = netdev_priv(dev);
 
-	strlcpy(info->driver, dev_driver_string(greth->dev),
+	strscpy(info->driver, dev_driver_string(greth->dev),
 		sizeof(info->driver));
-	strlcpy(info->version, "revision: 1.0", sizeof(info->version));
-	strlcpy(info->bus_info, greth->dev->bus->name, sizeof(info->bus_info));
-	strlcpy(info->fw_version, "N/A", sizeof(info->fw_version));
+	strscpy(info->bus_info, greth->dev->bus->name, sizeof(info->bus_info));
 }
 
 static void greth_get_regs(struct net_device *dev, struct ethtool_regs *regs, void *p)
@@ -1278,11 +1274,11 @@ static int greth_mdio_probe(struct net_device *dev)
 	}
 
 	if (greth->gbit_mac)
-		phy->supported &= PHY_GBIT_FEATURES;
+		phy_set_max_speed(phy, SPEED_1000);
 	else
-		phy->supported &= PHY_BASIC_FEATURES;
+		phy_set_max_speed(phy, SPEED_100);
 
-	phy->advertising = phy->supported;
+	linkmode_copy(phy->advertising, phy->supported);
 
 	greth->link = 0;
 	greth->speed = 0;
@@ -1352,6 +1348,7 @@ static int greth_of_probe(struct platform_device *ofdev)
 	int i;
 	int err;
 	int tmp;
+	u8 addr[ETH_ALEN];
 	unsigned long timeout;
 
 	dev = alloc_etherdev(sizeof(struct greth_private));
@@ -1432,18 +1429,18 @@ static int greth_of_probe(struct platform_device *ofdev)
 	}
 
 	/* Allocate TX descriptor ring in coherent memory */
-	greth->tx_bd_base = dma_zalloc_coherent(greth->dev, 1024,
-						&greth->tx_bd_base_phys,
-						GFP_KERNEL);
+	greth->tx_bd_base = dma_alloc_coherent(greth->dev, 1024,
+					       &greth->tx_bd_base_phys,
+					       GFP_KERNEL);
 	if (!greth->tx_bd_base) {
 		err = -ENOMEM;
 		goto error3;
 	}
 
 	/* Allocate RX descriptor ring in coherent memory */
-	greth->rx_bd_base = dma_zalloc_coherent(greth->dev, 1024,
-						&greth->rx_bd_base_phys,
-						GFP_KERNEL);
+	greth->rx_bd_base = dma_alloc_coherent(greth->dev, 1024,
+					       &greth->rx_bd_base_phys,
+					       GFP_KERNEL);
 	if (!greth->rx_bd_base) {
 		err = -ENOMEM;
 		goto error4;
@@ -1455,10 +1452,8 @@ static int greth_of_probe(struct platform_device *ofdev)
 			break;
 	}
 	if (i == 6) {
-		const u8 *addr;
-
-		addr = of_get_mac_address(ofdev->dev.of_node);
-		if (addr) {
+		err = of_get_mac_address(ofdev->dev.of_node, addr);
+		if (!err) {
 			for (i = 0; i < 6; i++)
 				macaddr[i] = (unsigned int) addr[i];
 		} else {
@@ -1470,7 +1465,8 @@ static int greth_of_probe(struct platform_device *ofdev)
 	}
 
 	for (i = 0; i < 6; i++)
-		dev->dev_addr[i] = macaddr[i];
+		addr[i] = macaddr[i];
+	eth_hw_addr_set(dev, addr);
 
 	macaddr[5]++;
 
@@ -1513,7 +1509,7 @@ static int greth_of_probe(struct platform_device *ofdev)
 	}
 
 	/* setup NAPI */
-	netif_napi_add(dev, &greth->napi, greth_poll, 64);
+	netif_napi_add(dev, &greth->napi, greth_poll);
 
 	return 0;
 
@@ -1530,7 +1526,7 @@ error1:
 	return err;
 }
 
-static int greth_of_remove(struct platform_device *of_dev)
+static void greth_of_remove(struct platform_device *of_dev)
 {
 	struct net_device *ndev = platform_get_drvdata(of_dev);
 	struct greth_private *greth = netdev_priv(ndev);
@@ -1545,11 +1541,10 @@ static int greth_of_remove(struct platform_device *of_dev)
 	mdiobus_unregister(greth->mdio);
 
 	unregister_netdev(ndev);
-	free_netdev(ndev);
 
 	of_iounmap(&of_dev->resource[0], greth->regs, resource_size(&of_dev->resource[0]));
 
-	return 0;
+	free_netdev(ndev);
 }
 
 static const struct of_device_id greth_of_match[] = {
diff --git a/drivers/net/ethernet/aeroflex/greth.h b/drivers/net/ethernet/aeroflex/greth.h
index 9c07140a5d8d..a1e04c9e932e 100644
--- a/drivers/net/ethernet/aeroflex/greth.h
+++ b/drivers/net/ethernet/aeroflex/greth.h
@@ -1,3 +1,4 @@
+/* SPDX-License-Identifier: GPL-2.0 */
 #ifndef GRETH_H
 #define GRETH_H
 
diff --git a/drivers/net/ethernet/agere/Kconfig b/drivers/net/ethernet/agere/Kconfig
index b6fe9200355a..9cd750184947 100644
--- a/drivers/net/ethernet/agere/Kconfig
+++ b/drivers/net/ethernet/agere/Kconfig
@@ -1,3 +1,4 @@
+# SPDX-License-Identifier: GPL-2.0-only
 #
 # Agere device configuration
 #
@@ -6,7 +7,7 @@ config NET_VENDOR_AGERE
 	bool "Agere devices"
 	default y
 	depends on PCI
-	---help---
+	help
 	  If you have a network (Ethernet) card belonging to this class, say Y.
 
 	  Note that the answer to this question doesn't directly affect the
@@ -20,7 +21,8 @@ config ET131X
 	tristate "Agere ET-1310 Gigabit Ethernet support"
 	depends on PCI
 	select PHYLIB
-	---help---
+	select CRC32
+	help
 	  This driver supports Agere ET-1310 ethernet adapters.
 
 	  To compile this driver as a module, choose M here. The module
diff --git a/drivers/net/ethernet/agere/Makefile b/drivers/net/ethernet/agere/Makefile
index 027ff9453fe1..8dbdf666b994 100644
--- a/drivers/net/ethernet/agere/Makefile
+++ b/drivers/net/ethernet/agere/Makefile
@@ -1,3 +1,4 @@
+# SPDX-License-Identifier: GPL-2.0-only
 #
 # Makefile for the Agere ET-131x ethernet driver
 #
diff --git a/drivers/net/ethernet/agere/et131x.c b/drivers/net/ethernet/agere/et131x.c
index 87a11b9f0ea5..5c8217638dda 100644
--- a/drivers/net/ethernet/agere/et131x.c
+++ b/drivers/net/ethernet/agere/et131x.c
@@ -714,11 +714,11 @@ static int et131x_init_eeprom(struct et131x_adapter *adapter)
 			 * gather additional information that normally would
 			 * come from the eeprom, like MAC Address
 			 */
-			adapter->has_eeprom = 0;
+			adapter->has_eeprom = false;
 			return -EIO;
 		}
 	}
-	adapter->has_eeprom = 1;
+	adapter->has_eeprom = true;
 
 	/* Read the EEPROM for information regarding LED behavior. Refer to
 	 * et131x_xcvr_init() for its use.
@@ -950,7 +950,6 @@ static void et1310_setup_device_for_multicast(struct et131x_adapter *adapter)
 	u32 hash2 = 0;
 	u32 hash3 = 0;
 	u32 hash4 = 0;
-	u32 pm_csr;
 
 	/* If ET131X_PACKET_TYPE_MULTICAST is specified, then we provision
 	 * the multi-cast LIST.  If it is NOT specified, (and "ALL" is not
@@ -984,7 +983,6 @@ static void et1310_setup_device_for_multicast(struct et131x_adapter *adapter)
 	}
 
 	/* Write out the new hash to the device */
-	pm_csr = readl(&adapter->regs->global.pm_csr);
 	if (!et1310_in_phy_coma(adapter)) {
 		writel(hash1, &rxmac->multi_hash1);
 		writel(hash2, &rxmac->multi_hash2);
@@ -999,7 +997,6 @@ static void et1310_setup_device_for_unicast(struct et131x_adapter *adapter)
 	u32 uni_pf1;
 	u32 uni_pf2;
 	u32 uni_pf3;
-	u32 pm_csr;
 
 	/* Set up unicast packet filter reg 3 to be the first two octets of
 	 * the MAC address for both address
@@ -1025,7 +1022,6 @@ static void et1310_setup_device_for_unicast(struct et131x_adapter *adapter)
 		  (adapter->addr[4] << ET_RX_UNI_PF_ADDR1_5_SHIFT) |
 		   adapter->addr[5];
 
-	pm_csr = readl(&adapter->regs->global.pm_csr);
 	if (!et1310_in_phy_coma(adapter)) {
 		writel(uni_pf1, &rxmac->uni_pf_addr1);
 		writel(uni_pf2, &rxmac->uni_pf_addr2);
@@ -1110,7 +1106,7 @@ static void et1310_config_rxmac_regs(struct et131x_adapter *adapter)
 	writel(0, &rxmac->mif_ctrl);
 	writel(0, &rxmac->space_avail);
 
-	/* Initialize the the mif_ctrl register
+	/* Initialize the mif_ctrl register
 	 * bit 3:  Receive code error. One or more nibbles were signaled as
 	 *	   errors  during the reception of the packet.  Clear this
 	 *	   bit in Gigabit, set it in 100Mbit.  This was derived
@@ -2282,7 +2278,7 @@ static struct rfd *nic_rx_pkts(struct et131x_adapter *adapter)
 
 	adapter->netdev->stats.rx_bytes += rfd->len;
 
-	memcpy(skb_put(skb, rfd->len), fbr->virt[buff_index], rfd->len);
+	skb_put_data(skb, fbr->virt[buff_index], rfd->len);
 
 	skb->protocol = eth_type_trans(skb, adapter->netdev);
 	skb->ip_summed = CHECKSUM_NONE;
@@ -2362,7 +2358,7 @@ static int et131x_tx_dma_memory_alloc(struct et131x_adapter *adapter)
 
 	/* Allocate memory for the TCB's (Transmit Control Block) */
 	tx_ring->tcb_ring = kcalloc(NUM_TCB, sizeof(struct tcb),
-				    GFP_ATOMIC | GFP_DMA);
+				    GFP_KERNEL | GFP_DMA);
 	if (!tx_ring->tcb_ring)
 		return -ENOMEM;
 
@@ -2417,16 +2413,18 @@ static void et131x_tx_dma_memory_free(struct et131x_adapter *adapter)
 	kfree(tx_ring->tcb_ring);
 }
 
+#define MAX_TX_DESC_PER_PKT 24
+
 /* nic_send_packet - NIC specific send handler for version B silicon. */
 static int nic_send_packet(struct et131x_adapter *adapter, struct tcb *tcb)
 {
 	u32 i;
-	struct tx_desc desc[24];
+	struct tx_desc desc[MAX_TX_DESC_PER_PKT];
 	u32 frag = 0;
 	u32 thiscopy, remainder;
 	struct sk_buff *skb = tcb->skb;
 	u32 nr_frags = skb_shinfo(skb)->nr_frags + 1;
-	struct skb_frag_struct *frags = &skb_shinfo(skb)->frags[0];
+	skb_frag_t *frags = &skb_shinfo(skb)->frags[0];
 	struct phy_device *phydev = adapter->netdev->phydev;
 	dma_addr_t dma_addr;
 	struct tx_ring *tx_ring = &adapter->tx_ring;
@@ -2436,9 +2434,6 @@ static int nic_send_packet(struct et131x_adapter *adapter, struct tcb *tcb)
 	 * more than 5 fragments.
 	 */
 
-	/* nr_frags should be no more than 18. */
-	BUILD_BUG_ON(MAX_SKB_FRAGS + 1 > 23);
-
 	memset(desc, 0, sizeof(struct tx_desc) * (nr_frags + 1));
 
 	for (i = 0; i < nr_frags; i++) {
@@ -2464,6 +2459,10 @@ static int nic_send_packet(struct et131x_adapter *adapter, struct tcb *tcb)
 							  skb->data,
 							  skb_headlen(skb),
 							  DMA_TO_DEVICE);
+				if (dma_mapping_error(&adapter->pdev->dev,
+						      dma_addr))
+					return -ENOMEM;
+
 				desc[frag].addr_lo = lower_32_bits(dma_addr);
 				desc[frag].addr_hi = upper_32_bits(dma_addr);
 				frag++;
@@ -2473,6 +2472,10 @@ static int nic_send_packet(struct et131x_adapter *adapter, struct tcb *tcb)
 							  skb->data,
 							  skb_headlen(skb) / 2,
 							  DMA_TO_DEVICE);
+				if (dma_mapping_error(&adapter->pdev->dev,
+						      dma_addr))
+					return -ENOMEM;
+
 				desc[frag].addr_lo = lower_32_bits(dma_addr);
 				desc[frag].addr_hi = upper_32_bits(dma_addr);
 				frag++;
@@ -2483,17 +2486,24 @@ static int nic_send_packet(struct et131x_adapter *adapter, struct tcb *tcb)
 							  skb_headlen(skb) / 2,
 							  skb_headlen(skb) / 2,
 							  DMA_TO_DEVICE);
+				if (dma_mapping_error(&adapter->pdev->dev,
+						      dma_addr))
+					goto unmap_first_out;
+
 				desc[frag].addr_lo = lower_32_bits(dma_addr);
 				desc[frag].addr_hi = upper_32_bits(dma_addr);
 				frag++;
 			}
 		} else {
-			desc[frag].len_vlan = frags[i - 1].size;
+			desc[frag].len_vlan = skb_frag_size(&frags[i - 1]);
 			dma_addr = skb_frag_dma_map(&adapter->pdev->dev,
 						    &frags[i - 1],
 						    0,
-						    frags[i - 1].size,
+						    desc[frag].len_vlan,
 						    DMA_TO_DEVICE);
+			if (dma_mapping_error(&adapter->pdev->dev, dma_addr))
+				goto unmap_out;
+
 			desc[frag].addr_lo = lower_32_bits(dma_addr);
 			desc[frag].addr_hi = upper_32_bits(dma_addr);
 			frag++;
@@ -2583,6 +2593,27 @@ static int nic_send_packet(struct et131x_adapter *adapter, struct tcb *tcb)
 		       &adapter->regs->global.watchdog_timer);
 	}
 	return 0;
+
+unmap_out:
+	// Unmap the body of the packet with map_page
+	while (--i) {
+		frag--;
+		dma_addr = desc[frag].addr_lo;
+		dma_addr |= (u64)desc[frag].addr_hi << 32;
+		dma_unmap_page(&adapter->pdev->dev, dma_addr,
+			       desc[frag].len_vlan, DMA_TO_DEVICE);
+	}
+
+unmap_first_out:
+	// Unmap the header with map_single
+	while (frag--) {
+		dma_addr = desc[frag].addr_lo;
+		dma_addr |= (u64)desc[frag].addr_hi << 32;
+		dma_unmap_single(&adapter->pdev->dev, dma_addr,
+				 desc[frag].len_vlan, DMA_TO_DEVICE);
+	}
+
+	return -ENOMEM;
 }
 
 static int send_packet(struct sk_buff *skb, struct et131x_adapter *adapter)
@@ -2957,9 +2988,8 @@ static void et131x_get_drvinfo(struct net_device *netdev,
 {
 	struct et131x_adapter *adapter = netdev_priv(netdev);
 
-	strlcpy(info->driver, DRIVER_NAME, sizeof(info->driver));
-	strlcpy(info->version, DRIVER_VERSION, sizeof(info->version));
-	strlcpy(info->bus_info, pci_name(adapter->pdev),
+	strscpy(info->driver, DRIVER_NAME, sizeof(info->driver));
+	strscpy(info->bus_info, pci_name(adapter->pdev),
 		sizeof(info->bus_info));
 }
 
@@ -3080,9 +3110,10 @@ err_out:
  * The routine called when the error timer expires, to track the number of
  * recurring errors.
  */
-static void et131x_error_timer_handler(unsigned long data)
+static void et131x_error_timer_handler(struct timer_list *t)
 {
-	struct et131x_adapter *adapter = (struct et131x_adapter *)data;
+	struct et131x_adapter *adapter = timer_container_of(adapter, t,
+							    error_timer);
 	struct phy_device *phydev = adapter->netdev->phydev;
 
 	if (et1310_in_phy_coma(adapter)) {
@@ -3258,19 +3289,11 @@ static int et131x_mii_probe(struct net_device *netdev)
 		return PTR_ERR(phydev);
 	}
 
-	phydev->supported &= (SUPPORTED_10baseT_Half |
-			      SUPPORTED_10baseT_Full |
-			      SUPPORTED_100baseT_Half |
-			      SUPPORTED_100baseT_Full |
-			      SUPPORTED_Autoneg |
-			      SUPPORTED_MII |
-			      SUPPORTED_TP);
+	phy_set_max_speed(phydev, SPEED_100);
 
 	if (adapter->pdev->device != ET131X_PCI_DEVICE_ID_FAST)
-		phydev->supported |= SUPPORTED_1000baseT_Half |
-				     SUPPORTED_1000baseT_Full;
+		phy_set_max_speed(phydev, SPEED_1000);
 
-	phydev->advertising = phydev->supported;
 	phydev->autoneg = AUTONEG_ENABLE;
 
 	phy_attached_info(phydev);
@@ -3452,12 +3475,9 @@ static irqreturn_t et131x_isr(int irq, void *dev_id)
 		 * send a pause packet, otherwise just exit
 		 */
 		if (adapter->flow == FLOW_TXONLY || adapter->flow == FLOW_BOTH) {
-			u32 pm_csr;
-
 			/* Tell the device to send a pause packet via the back
 			 * pressure register (bp req and bp xon/xoff)
 			 */
-			pm_csr = readl(&iomem->global.pm_csr);
 			if (!et1310_in_phy_coma(adapter))
 				writel(3, &iomem->txmac.bp_ctrl);
 		}
@@ -3624,11 +3644,9 @@ static int et131x_open(struct net_device *netdev)
 	int result;
 
 	/* Start the timer to track NIC errors */
-	init_timer(&adapter->error_timer);
+	timer_setup(&adapter->error_timer, et131x_error_timer_handler, 0);
 	adapter->error_timer.expires = jiffies +
 		msecs_to_jiffies(TX_ERROR_PERIOD);
-	adapter->error_timer.function = et131x_error_timer_handler;
-	adapter->error_timer.data = (unsigned long)adapter;
 	add_timer(&adapter->error_timer);
 
 	result = request_irq(irq, et131x_isr,
@@ -3658,16 +3676,7 @@ static int et131x_close(struct net_device *netdev)
 	free_irq(adapter->pdev->irq, netdev);
 
 	/* Stop the error timer */
-	return del_timer_sync(&adapter->error_timer);
-}
-
-static int et131x_ioctl(struct net_device *netdev, struct ifreq *reqbuf,
-			int cmd)
-{
-	if (!netdev->phydev)
-		return -EINVAL;
-
-	return phy_mii_ioctl(netdev->phydev, reqbuf, cmd);
+	return timer_delete_sync(&adapter->error_timer);
 }
 
 /* et131x_set_packet_filter - Configures the Rx Packet filtering */
@@ -3789,6 +3798,13 @@ static netdev_tx_t et131x_tx(struct sk_buff *skb, struct net_device *netdev)
 	struct et131x_adapter *adapter = netdev_priv(netdev);
 	struct tx_ring *tx_ring = &adapter->tx_ring;
 
+	/* This driver does not support TSO, it is very unlikely
+	 * this condition is true.
+	 */
+	if (unlikely(skb_shinfo(skb)->nr_frags > MAX_TX_DESC_PER_PKT - 2)) {
+		if (skb_linearize(skb))
+			goto drop_err;
+	}
 	/* stop the queue if it's getting full */
 	if (tx_ring->used >= NUM_TCB - 1 && !netif_queue_stopped(netdev))
 		netif_stop_queue(netdev);
@@ -3821,7 +3837,7 @@ drop_err:
  * specified by the 'tx_timeo" element in the net_device structure (see
  * et131x_alloc_device() to see how this value is set).
  */
-static void et131x_tx_timeout(struct net_device *netdev)
+static void et131x_tx_timeout(struct net_device *netdev, unsigned int txqueue)
 {
 	struct et131x_adapter *adapter = netdev_priv(netdev);
 	struct tx_ring *tx_ring = &adapter->tx_ring;
@@ -3873,7 +3889,7 @@ static int et131x_change_mtu(struct net_device *netdev, int new_mtu)
 
 	et131x_disable_txrx(netdev);
 
-	netdev->mtu = new_mtu;
+	WRITE_ONCE(netdev->mtu, new_mtu);
 
 	et131x_adapter_memory_free(adapter);
 
@@ -3890,7 +3906,7 @@ static int et131x_change_mtu(struct net_device *netdev, int new_mtu)
 
 	et131x_init_send(adapter);
 	et131x_hwaddr_init(adapter);
-	ether_addr_copy(netdev->dev_addr, adapter->addr);
+	eth_hw_addr_set(netdev, adapter->addr);
 
 	/* Init the device with the new settings */
 	et131x_adapter_setup(adapter);
@@ -3909,7 +3925,7 @@ static const struct net_device_ops et131x_netdev_ops = {
 	.ndo_set_mac_address	= eth_mac_addr,
 	.ndo_validate_addr	= eth_validate_addr,
 	.ndo_get_stats		= et131x_stats,
-	.ndo_do_ioctl		= et131x_ioctl,
+	.ndo_eth_ioctl		= phy_do_ioctl,
 };
 
 static int et131x_pci_setup(struct pci_dev *pdev,
@@ -3941,10 +3957,9 @@ static int et131x_pci_setup(struct pci_dev *pdev,
 	pci_set_master(pdev);
 
 	/* Check the DMA addressing support of this device */
-	if (dma_set_mask_and_coherent(&pdev->dev, DMA_BIT_MASK(64)) &&
-	    dma_set_mask_and_coherent(&pdev->dev, DMA_BIT_MASK(32))) {
+	rc = dma_set_mask_and_coherent(&pdev->dev, DMA_BIT_MASK(64));
+	if (rc) {
 		dev_err(&pdev->dev, "No usable DMA addressing method\n");
-		rc = -EIO;
 		goto err_release_res;
 	}
 
@@ -3991,9 +4006,9 @@ static int et131x_pci_setup(struct pci_dev *pdev,
 
 	et131x_init_send(adapter);
 
-	netif_napi_add(netdev, &adapter->napi, et131x_poll, 64);
+	netif_napi_add(netdev, &adapter->napi, et131x_poll);
 
-	ether_addr_copy(netdev->dev_addr, adapter->addr);
+	eth_hw_addr_set(netdev, adapter->addr);
 
 	rc = -ENOMEM;
 
@@ -4004,8 +4019,7 @@ static int et131x_pci_setup(struct pci_dev *pdev,
 	}
 
 	adapter->mii_bus->name = "et131x_eth_mii";
-	snprintf(adapter->mii_bus->id, MII_BUS_ID_SIZE, "%x",
-		 (adapter->pdev->bus->number << 8) | adapter->pdev->devfn);
+	snprintf(adapter->mii_bus->id, MII_BUS_ID_SIZE, "%x", pci_dev_id(adapter->pdev));
 	adapter->mii_bus->priv = netdev;
 	adapter->mii_bus->read = et131x_mdio_read;
 	adapter->mii_bus->write = et131x_mdio_write;
diff --git a/drivers/net/ethernet/agere/et131x.h b/drivers/net/ethernet/agere/et131x.h
index be9a11c02526..d0e922584d8a 100644
--- a/drivers/net/ethernet/agere/et131x.h
+++ b/drivers/net/ethernet/agere/et131x.h
@@ -46,7 +46,6 @@
  */
 
 #define DRIVER_NAME "et131x"
-#define DRIVER_VERSION "v2.0"
 
 /* EEPROM registers */
 
diff --git a/drivers/net/ethernet/airoha/Kconfig b/drivers/net/ethernet/airoha/Kconfig
new file mode 100644
index 000000000000..ad3ce501e7a5
--- /dev/null
+++ b/drivers/net/ethernet/airoha/Kconfig
@@ -0,0 +1,34 @@
+# SPDX-License-Identifier: GPL-2.0-only
+config NET_VENDOR_AIROHA
+	bool "Airoha devices"
+	depends on ARCH_AIROHA || COMPILE_TEST
+	help
+	  If you have a Airoha SoC with ethernet, say Y.
+
+if NET_VENDOR_AIROHA
+
+config NET_AIROHA_NPU
+	tristate "Airoha NPU support"
+	select WANT_DEV_COREDUMP
+	select REGMAP_MMIO
+	help
+	  This driver supports Airoha Network Processor (NPU) available
+	  on the Airoha Soc family.
+
+config NET_AIROHA
+	tristate "Airoha SoC Gigabit Ethernet support"
+	depends on NET_DSA || !NET_DSA
+	select NET_AIROHA_NPU
+	select PAGE_POOL
+	help
+	  This driver supports the gigabit ethernet MACs in the
+	  Airoha SoC family.
+
+config NET_AIROHA_FLOW_STATS
+	default y
+	bool "Airoha flow stats"
+	depends on NET_AIROHA && NET_AIROHA_NPU
+	help
+	  Enable Aiorha flowtable statistic counters.
+
+endif #NET_VENDOR_AIROHA
diff --git a/drivers/net/ethernet/airoha/Makefile b/drivers/net/ethernet/airoha/Makefile
new file mode 100644
index 000000000000..94468053e34b
--- /dev/null
+++ b/drivers/net/ethernet/airoha/Makefile
@@ -0,0 +1,9 @@
+# SPDX-License-Identifier: GPL-2.0-only
+#
+# Airoha for the Mediatek SoCs built-in ethernet macs
+#
+
+obj-$(CONFIG_NET_AIROHA) += airoha-eth.o
+airoha-eth-y := airoha_eth.o airoha_ppe.o
+airoha-eth-$(CONFIG_DEBUG_FS) += airoha_ppe_debugfs.o
+obj-$(CONFIG_NET_AIROHA_NPU) += airoha_npu.o
diff --git a/drivers/net/ethernet/airoha/airoha_eth.c b/drivers/net/ethernet/airoha/airoha_eth.c
new file mode 100644
index 000000000000..833dd911980b
--- /dev/null
+++ b/drivers/net/ethernet/airoha/airoha_eth.c
@@ -0,0 +1,3050 @@
+// SPDX-License-Identifier: GPL-2.0-only
+/*
+ * Copyright (c) 2024 AIROHA Inc
+ * Author: Lorenzo Bianconi <lorenzo@kernel.org>
+ */
+#include <linux/of.h>
+#include <linux/of_net.h>
+#include <linux/of_reserved_mem.h>
+#include <linux/platform_device.h>
+#include <linux/tcp.h>
+#include <linux/u64_stats_sync.h>
+#include <net/dst_metadata.h>
+#include <net/page_pool/helpers.h>
+#include <net/pkt_cls.h>
+#include <uapi/linux/ppp_defs.h>
+
+#include "airoha_regs.h"
+#include "airoha_eth.h"
+
+u32 airoha_rr(void __iomem *base, u32 offset)
+{
+	return readl(base + offset);
+}
+
+void airoha_wr(void __iomem *base, u32 offset, u32 val)
+{
+	writel(val, base + offset);
+}
+
+u32 airoha_rmw(void __iomem *base, u32 offset, u32 mask, u32 val)
+{
+	val |= (airoha_rr(base, offset) & ~mask);
+	airoha_wr(base, offset, val);
+
+	return val;
+}
+
+static void airoha_qdma_set_irqmask(struct airoha_irq_bank *irq_bank,
+				    int index, u32 clear, u32 set)
+{
+	struct airoha_qdma *qdma = irq_bank->qdma;
+	int bank = irq_bank - &qdma->irq_banks[0];
+	unsigned long flags;
+
+	if (WARN_ON_ONCE(index >= ARRAY_SIZE(irq_bank->irqmask)))
+		return;
+
+	spin_lock_irqsave(&irq_bank->irq_lock, flags);
+
+	irq_bank->irqmask[index] &= ~clear;
+	irq_bank->irqmask[index] |= set;
+	airoha_qdma_wr(qdma, REG_INT_ENABLE(bank, index),
+		       irq_bank->irqmask[index]);
+	/* Read irq_enable register in order to guarantee the update above
+	 * completes in the spinlock critical section.
+	 */
+	airoha_qdma_rr(qdma, REG_INT_ENABLE(bank, index));
+
+	spin_unlock_irqrestore(&irq_bank->irq_lock, flags);
+}
+
+static void airoha_qdma_irq_enable(struct airoha_irq_bank *irq_bank,
+				   int index, u32 mask)
+{
+	airoha_qdma_set_irqmask(irq_bank, index, 0, mask);
+}
+
+static void airoha_qdma_irq_disable(struct airoha_irq_bank *irq_bank,
+				    int index, u32 mask)
+{
+	airoha_qdma_set_irqmask(irq_bank, index, mask, 0);
+}
+
+static void airoha_set_macaddr(struct airoha_gdm_port *port, const u8 *addr)
+{
+	struct airoha_eth *eth = port->qdma->eth;
+	u32 val, reg;
+
+	reg = airhoa_is_lan_gdm_port(port) ? REG_FE_LAN_MAC_H
+					   : REG_FE_WAN_MAC_H;
+	val = (addr[0] << 16) | (addr[1] << 8) | addr[2];
+	airoha_fe_wr(eth, reg, val);
+
+	val = (addr[3] << 16) | (addr[4] << 8) | addr[5];
+	airoha_fe_wr(eth, REG_FE_MAC_LMIN(reg), val);
+	airoha_fe_wr(eth, REG_FE_MAC_LMAX(reg), val);
+
+	airoha_ppe_init_upd_mem(port);
+}
+
+static void airoha_set_gdm_port_fwd_cfg(struct airoha_eth *eth, u32 addr,
+					u32 val)
+{
+	airoha_fe_rmw(eth, addr, GDM_OCFQ_MASK,
+		      FIELD_PREP(GDM_OCFQ_MASK, val));
+	airoha_fe_rmw(eth, addr, GDM_MCFQ_MASK,
+		      FIELD_PREP(GDM_MCFQ_MASK, val));
+	airoha_fe_rmw(eth, addr, GDM_BCFQ_MASK,
+		      FIELD_PREP(GDM_BCFQ_MASK, val));
+	airoha_fe_rmw(eth, addr, GDM_UCFQ_MASK,
+		      FIELD_PREP(GDM_UCFQ_MASK, val));
+}
+
+static int airoha_set_vip_for_gdm_port(struct airoha_gdm_port *port,
+				       bool enable)
+{
+	struct airoha_eth *eth = port->qdma->eth;
+	u32 vip_port;
+
+	switch (port->id) {
+	case 3:
+		/* FIXME: handle XSI_PCIE1_PORT */
+		vip_port = XSI_PCIE0_VIP_PORT_MASK;
+		break;
+	case 4:
+		/* FIXME: handle XSI_USB_PORT */
+		vip_port = XSI_ETH_VIP_PORT_MASK;
+		break;
+	default:
+		return 0;
+	}
+
+	if (enable) {
+		airoha_fe_set(eth, REG_FE_VIP_PORT_EN, vip_port);
+		airoha_fe_set(eth, REG_FE_IFC_PORT_EN, vip_port);
+	} else {
+		airoha_fe_clear(eth, REG_FE_VIP_PORT_EN, vip_port);
+		airoha_fe_clear(eth, REG_FE_IFC_PORT_EN, vip_port);
+	}
+
+	return 0;
+}
+
+static void airoha_fe_maccr_init(struct airoha_eth *eth)
+{
+	int p;
+
+	for (p = 1; p <= ARRAY_SIZE(eth->ports); p++)
+		airoha_fe_set(eth, REG_GDM_FWD_CFG(p),
+			      GDM_TCP_CKSUM | GDM_UDP_CKSUM | GDM_IP4_CKSUM |
+			      GDM_DROP_CRC_ERR);
+
+	airoha_fe_rmw(eth, REG_CDM1_VLAN_CTRL, CDM1_VLAN_MASK,
+		      FIELD_PREP(CDM1_VLAN_MASK, 0x8100));
+
+	airoha_fe_set(eth, REG_FE_CPORT_CFG, FE_CPORT_PAD);
+}
+
+static void airoha_fe_vip_setup(struct airoha_eth *eth)
+{
+	airoha_fe_wr(eth, REG_FE_VIP_PATN(3), ETH_P_PPP_DISC);
+	airoha_fe_wr(eth, REG_FE_VIP_EN(3), PATN_FCPU_EN_MASK | PATN_EN_MASK);
+
+	airoha_fe_wr(eth, REG_FE_VIP_PATN(4), PPP_LCP);
+	airoha_fe_wr(eth, REG_FE_VIP_EN(4),
+		     PATN_FCPU_EN_MASK | FIELD_PREP(PATN_TYPE_MASK, 1) |
+		     PATN_EN_MASK);
+
+	airoha_fe_wr(eth, REG_FE_VIP_PATN(6), PPP_IPCP);
+	airoha_fe_wr(eth, REG_FE_VIP_EN(6),
+		     PATN_FCPU_EN_MASK | FIELD_PREP(PATN_TYPE_MASK, 1) |
+		     PATN_EN_MASK);
+
+	airoha_fe_wr(eth, REG_FE_VIP_PATN(7), PPP_CHAP);
+	airoha_fe_wr(eth, REG_FE_VIP_EN(7),
+		     PATN_FCPU_EN_MASK | FIELD_PREP(PATN_TYPE_MASK, 1) |
+		     PATN_EN_MASK);
+
+	/* BOOTP (0x43) */
+	airoha_fe_wr(eth, REG_FE_VIP_PATN(8), 0x43);
+	airoha_fe_wr(eth, REG_FE_VIP_EN(8),
+		     PATN_FCPU_EN_MASK | PATN_SP_EN_MASK |
+		     FIELD_PREP(PATN_TYPE_MASK, 4) | PATN_EN_MASK);
+
+	/* BOOTP (0x44) */
+	airoha_fe_wr(eth, REG_FE_VIP_PATN(9), 0x44);
+	airoha_fe_wr(eth, REG_FE_VIP_EN(9),
+		     PATN_FCPU_EN_MASK | PATN_SP_EN_MASK |
+		     FIELD_PREP(PATN_TYPE_MASK, 4) | PATN_EN_MASK);
+
+	/* ISAKMP */
+	airoha_fe_wr(eth, REG_FE_VIP_PATN(10), 0x1f401f4);
+	airoha_fe_wr(eth, REG_FE_VIP_EN(10),
+		     PATN_FCPU_EN_MASK | PATN_DP_EN_MASK | PATN_SP_EN_MASK |
+		     FIELD_PREP(PATN_TYPE_MASK, 4) | PATN_EN_MASK);
+
+	airoha_fe_wr(eth, REG_FE_VIP_PATN(11), PPP_IPV6CP);
+	airoha_fe_wr(eth, REG_FE_VIP_EN(11),
+		     PATN_FCPU_EN_MASK | FIELD_PREP(PATN_TYPE_MASK, 1) |
+		     PATN_EN_MASK);
+
+	/* DHCPv6 */
+	airoha_fe_wr(eth, REG_FE_VIP_PATN(12), 0x2220223);
+	airoha_fe_wr(eth, REG_FE_VIP_EN(12),
+		     PATN_FCPU_EN_MASK | PATN_DP_EN_MASK | PATN_SP_EN_MASK |
+		     FIELD_PREP(PATN_TYPE_MASK, 4) | PATN_EN_MASK);
+
+	airoha_fe_wr(eth, REG_FE_VIP_PATN(19), PPP_PAP);
+	airoha_fe_wr(eth, REG_FE_VIP_EN(19),
+		     PATN_FCPU_EN_MASK | FIELD_PREP(PATN_TYPE_MASK, 1) |
+		     PATN_EN_MASK);
+
+	/* ETH->ETH_P_1905 (0x893a) */
+	airoha_fe_wr(eth, REG_FE_VIP_PATN(20), 0x893a);
+	airoha_fe_wr(eth, REG_FE_VIP_EN(20),
+		     PATN_FCPU_EN_MASK | PATN_EN_MASK);
+
+	airoha_fe_wr(eth, REG_FE_VIP_PATN(21), ETH_P_LLDP);
+	airoha_fe_wr(eth, REG_FE_VIP_EN(21),
+		     PATN_FCPU_EN_MASK | PATN_EN_MASK);
+}
+
+static u32 airoha_fe_get_pse_queue_rsv_pages(struct airoha_eth *eth,
+					     u32 port, u32 queue)
+{
+	u32 val;
+
+	airoha_fe_rmw(eth, REG_FE_PSE_QUEUE_CFG_WR,
+		      PSE_CFG_PORT_ID_MASK | PSE_CFG_QUEUE_ID_MASK,
+		      FIELD_PREP(PSE_CFG_PORT_ID_MASK, port) |
+		      FIELD_PREP(PSE_CFG_QUEUE_ID_MASK, queue));
+	val = airoha_fe_rr(eth, REG_FE_PSE_QUEUE_CFG_VAL);
+
+	return FIELD_GET(PSE_CFG_OQ_RSV_MASK, val);
+}
+
+static void airoha_fe_set_pse_queue_rsv_pages(struct airoha_eth *eth,
+					      u32 port, u32 queue, u32 val)
+{
+	airoha_fe_rmw(eth, REG_FE_PSE_QUEUE_CFG_VAL, PSE_CFG_OQ_RSV_MASK,
+		      FIELD_PREP(PSE_CFG_OQ_RSV_MASK, val));
+	airoha_fe_rmw(eth, REG_FE_PSE_QUEUE_CFG_WR,
+		      PSE_CFG_PORT_ID_MASK | PSE_CFG_QUEUE_ID_MASK |
+		      PSE_CFG_WR_EN_MASK | PSE_CFG_OQRSV_SEL_MASK,
+		      FIELD_PREP(PSE_CFG_PORT_ID_MASK, port) |
+		      FIELD_PREP(PSE_CFG_QUEUE_ID_MASK, queue) |
+		      PSE_CFG_WR_EN_MASK | PSE_CFG_OQRSV_SEL_MASK);
+}
+
+static u32 airoha_fe_get_pse_all_rsv(struct airoha_eth *eth)
+{
+	u32 val = airoha_fe_rr(eth, REG_FE_PSE_BUF_SET);
+
+	return FIELD_GET(PSE_ALLRSV_MASK, val);
+}
+
+static int airoha_fe_set_pse_oq_rsv(struct airoha_eth *eth,
+				    u32 port, u32 queue, u32 val)
+{
+	u32 orig_val = airoha_fe_get_pse_queue_rsv_pages(eth, port, queue);
+	u32 tmp, all_rsv, fq_limit;
+
+	airoha_fe_set_pse_queue_rsv_pages(eth, port, queue, val);
+
+	/* modify all rsv */
+	all_rsv = airoha_fe_get_pse_all_rsv(eth);
+	all_rsv += (val - orig_val);
+	airoha_fe_rmw(eth, REG_FE_PSE_BUF_SET, PSE_ALLRSV_MASK,
+		      FIELD_PREP(PSE_ALLRSV_MASK, all_rsv));
+
+	/* modify hthd */
+	tmp = airoha_fe_rr(eth, PSE_FQ_CFG);
+	fq_limit = FIELD_GET(PSE_FQ_LIMIT_MASK, tmp);
+	tmp = fq_limit - all_rsv - 0x20;
+	airoha_fe_rmw(eth, REG_PSE_SHARE_USED_THD,
+		      PSE_SHARE_USED_HTHD_MASK,
+		      FIELD_PREP(PSE_SHARE_USED_HTHD_MASK, tmp));
+
+	tmp = fq_limit - all_rsv - 0x100;
+	airoha_fe_rmw(eth, REG_PSE_SHARE_USED_THD,
+		      PSE_SHARE_USED_MTHD_MASK,
+		      FIELD_PREP(PSE_SHARE_USED_MTHD_MASK, tmp));
+	tmp = (3 * tmp) >> 2;
+	airoha_fe_rmw(eth, REG_FE_PSE_BUF_SET,
+		      PSE_SHARE_USED_LTHD_MASK,
+		      FIELD_PREP(PSE_SHARE_USED_LTHD_MASK, tmp));
+
+	return 0;
+}
+
+static void airoha_fe_pse_ports_init(struct airoha_eth *eth)
+{
+	const u32 pse_port_num_queues[] = {
+		[FE_PSE_PORT_CDM1] = 6,
+		[FE_PSE_PORT_GDM1] = 6,
+		[FE_PSE_PORT_GDM2] = 32,
+		[FE_PSE_PORT_GDM3] = 6,
+		[FE_PSE_PORT_PPE1] = 4,
+		[FE_PSE_PORT_CDM2] = 6,
+		[FE_PSE_PORT_CDM3] = 8,
+		[FE_PSE_PORT_CDM4] = 10,
+		[FE_PSE_PORT_PPE2] = 4,
+		[FE_PSE_PORT_GDM4] = 2,
+		[FE_PSE_PORT_CDM5] = 2,
+	};
+	u32 all_rsv;
+	int q;
+
+	all_rsv = airoha_fe_get_pse_all_rsv(eth);
+	/* hw misses PPE2 oq rsv */
+	all_rsv += PSE_RSV_PAGES * pse_port_num_queues[FE_PSE_PORT_PPE2];
+	airoha_fe_set(eth, REG_FE_PSE_BUF_SET, all_rsv);
+
+	/* CMD1 */
+	for (q = 0; q < pse_port_num_queues[FE_PSE_PORT_CDM1]; q++)
+		airoha_fe_set_pse_oq_rsv(eth, FE_PSE_PORT_CDM1, q,
+					 PSE_QUEUE_RSV_PAGES);
+	/* GMD1 */
+	for (q = 0; q < pse_port_num_queues[FE_PSE_PORT_GDM1]; q++)
+		airoha_fe_set_pse_oq_rsv(eth, FE_PSE_PORT_GDM1, q,
+					 PSE_QUEUE_RSV_PAGES);
+	/* GMD2 */
+	for (q = 6; q < pse_port_num_queues[FE_PSE_PORT_GDM2]; q++)
+		airoha_fe_set_pse_oq_rsv(eth, FE_PSE_PORT_GDM2, q, 0);
+	/* GMD3 */
+	for (q = 0; q < pse_port_num_queues[FE_PSE_PORT_GDM3]; q++)
+		airoha_fe_set_pse_oq_rsv(eth, FE_PSE_PORT_GDM3, q,
+					 PSE_QUEUE_RSV_PAGES);
+	/* PPE1 */
+	for (q = 0; q < pse_port_num_queues[FE_PSE_PORT_PPE1]; q++) {
+		if (q < pse_port_num_queues[FE_PSE_PORT_PPE1])
+			airoha_fe_set_pse_oq_rsv(eth, FE_PSE_PORT_PPE1, q,
+						 PSE_QUEUE_RSV_PAGES);
+		else
+			airoha_fe_set_pse_oq_rsv(eth, FE_PSE_PORT_PPE1, q, 0);
+	}
+	/* CDM2 */
+	for (q = 0; q < pse_port_num_queues[FE_PSE_PORT_CDM2]; q++)
+		airoha_fe_set_pse_oq_rsv(eth, FE_PSE_PORT_CDM2, q,
+					 PSE_QUEUE_RSV_PAGES);
+	/* CDM3 */
+	for (q = 0; q < pse_port_num_queues[FE_PSE_PORT_CDM3] - 1; q++)
+		airoha_fe_set_pse_oq_rsv(eth, FE_PSE_PORT_CDM3, q, 0);
+	/* CDM4 */
+	for (q = 4; q < pse_port_num_queues[FE_PSE_PORT_CDM4]; q++)
+		airoha_fe_set_pse_oq_rsv(eth, FE_PSE_PORT_CDM4, q,
+					 PSE_QUEUE_RSV_PAGES);
+	/* PPE2 */
+	for (q = 0; q < pse_port_num_queues[FE_PSE_PORT_PPE2]; q++) {
+		if (q < pse_port_num_queues[FE_PSE_PORT_PPE2] / 2)
+			airoha_fe_set_pse_oq_rsv(eth, FE_PSE_PORT_PPE2, q,
+						 PSE_QUEUE_RSV_PAGES);
+		else
+			airoha_fe_set_pse_oq_rsv(eth, FE_PSE_PORT_PPE2, q, 0);
+	}
+	/* GMD4 */
+	for (q = 0; q < pse_port_num_queues[FE_PSE_PORT_GDM4]; q++)
+		airoha_fe_set_pse_oq_rsv(eth, FE_PSE_PORT_GDM4, q,
+					 PSE_QUEUE_RSV_PAGES);
+	/* CDM5 */
+	for (q = 0; q < pse_port_num_queues[FE_PSE_PORT_CDM5]; q++)
+		airoha_fe_set_pse_oq_rsv(eth, FE_PSE_PORT_CDM5, q,
+					 PSE_QUEUE_RSV_PAGES);
+}
+
+static int airoha_fe_mc_vlan_clear(struct airoha_eth *eth)
+{
+	int i;
+
+	for (i = 0; i < AIROHA_FE_MC_MAX_VLAN_TABLE; i++) {
+		int err, j;
+		u32 val;
+
+		airoha_fe_wr(eth, REG_MC_VLAN_DATA, 0x0);
+
+		val = FIELD_PREP(MC_VLAN_CFG_TABLE_ID_MASK, i) |
+		      MC_VLAN_CFG_TABLE_SEL_MASK | MC_VLAN_CFG_RW_MASK;
+		airoha_fe_wr(eth, REG_MC_VLAN_CFG, val);
+		err = read_poll_timeout(airoha_fe_rr, val,
+					val & MC_VLAN_CFG_CMD_DONE_MASK,
+					USEC_PER_MSEC, 5 * USEC_PER_MSEC,
+					false, eth, REG_MC_VLAN_CFG);
+		if (err)
+			return err;
+
+		for (j = 0; j < AIROHA_FE_MC_MAX_VLAN_PORT; j++) {
+			airoha_fe_wr(eth, REG_MC_VLAN_DATA, 0x0);
+
+			val = FIELD_PREP(MC_VLAN_CFG_TABLE_ID_MASK, i) |
+			      FIELD_PREP(MC_VLAN_CFG_PORT_ID_MASK, j) |
+			      MC_VLAN_CFG_RW_MASK;
+			airoha_fe_wr(eth, REG_MC_VLAN_CFG, val);
+			err = read_poll_timeout(airoha_fe_rr, val,
+						val & MC_VLAN_CFG_CMD_DONE_MASK,
+						USEC_PER_MSEC,
+						5 * USEC_PER_MSEC, false, eth,
+						REG_MC_VLAN_CFG);
+			if (err)
+				return err;
+		}
+	}
+
+	return 0;
+}
+
+static void airoha_fe_crsn_qsel_init(struct airoha_eth *eth)
+{
+	/* CDM1_CRSN_QSEL */
+	airoha_fe_rmw(eth, REG_CDM1_CRSN_QSEL(CRSN_22 >> 2),
+		      CDM1_CRSN_QSEL_REASON_MASK(CRSN_22),
+		      FIELD_PREP(CDM1_CRSN_QSEL_REASON_MASK(CRSN_22),
+				 CDM_CRSN_QSEL_Q1));
+	airoha_fe_rmw(eth, REG_CDM1_CRSN_QSEL(CRSN_08 >> 2),
+		      CDM1_CRSN_QSEL_REASON_MASK(CRSN_08),
+		      FIELD_PREP(CDM1_CRSN_QSEL_REASON_MASK(CRSN_08),
+				 CDM_CRSN_QSEL_Q1));
+	airoha_fe_rmw(eth, REG_CDM1_CRSN_QSEL(CRSN_21 >> 2),
+		      CDM1_CRSN_QSEL_REASON_MASK(CRSN_21),
+		      FIELD_PREP(CDM1_CRSN_QSEL_REASON_MASK(CRSN_21),
+				 CDM_CRSN_QSEL_Q1));
+	airoha_fe_rmw(eth, REG_CDM1_CRSN_QSEL(CRSN_24 >> 2),
+		      CDM1_CRSN_QSEL_REASON_MASK(CRSN_24),
+		      FIELD_PREP(CDM1_CRSN_QSEL_REASON_MASK(CRSN_24),
+				 CDM_CRSN_QSEL_Q6));
+	airoha_fe_rmw(eth, REG_CDM1_CRSN_QSEL(CRSN_25 >> 2),
+		      CDM1_CRSN_QSEL_REASON_MASK(CRSN_25),
+		      FIELD_PREP(CDM1_CRSN_QSEL_REASON_MASK(CRSN_25),
+				 CDM_CRSN_QSEL_Q1));
+	/* CDM2_CRSN_QSEL */
+	airoha_fe_rmw(eth, REG_CDM2_CRSN_QSEL(CRSN_08 >> 2),
+		      CDM2_CRSN_QSEL_REASON_MASK(CRSN_08),
+		      FIELD_PREP(CDM2_CRSN_QSEL_REASON_MASK(CRSN_08),
+				 CDM_CRSN_QSEL_Q1));
+	airoha_fe_rmw(eth, REG_CDM2_CRSN_QSEL(CRSN_21 >> 2),
+		      CDM2_CRSN_QSEL_REASON_MASK(CRSN_21),
+		      FIELD_PREP(CDM2_CRSN_QSEL_REASON_MASK(CRSN_21),
+				 CDM_CRSN_QSEL_Q1));
+	airoha_fe_rmw(eth, REG_CDM2_CRSN_QSEL(CRSN_22 >> 2),
+		      CDM2_CRSN_QSEL_REASON_MASK(CRSN_22),
+		      FIELD_PREP(CDM2_CRSN_QSEL_REASON_MASK(CRSN_22),
+				 CDM_CRSN_QSEL_Q1));
+	airoha_fe_rmw(eth, REG_CDM2_CRSN_QSEL(CRSN_24 >> 2),
+		      CDM2_CRSN_QSEL_REASON_MASK(CRSN_24),
+		      FIELD_PREP(CDM2_CRSN_QSEL_REASON_MASK(CRSN_24),
+				 CDM_CRSN_QSEL_Q6));
+	airoha_fe_rmw(eth, REG_CDM2_CRSN_QSEL(CRSN_25 >> 2),
+		      CDM2_CRSN_QSEL_REASON_MASK(CRSN_25),
+		      FIELD_PREP(CDM2_CRSN_QSEL_REASON_MASK(CRSN_25),
+				 CDM_CRSN_QSEL_Q1));
+}
+
+static int airoha_fe_init(struct airoha_eth *eth)
+{
+	airoha_fe_maccr_init(eth);
+
+	/* PSE IQ reserve */
+	airoha_fe_rmw(eth, REG_PSE_IQ_REV1, PSE_IQ_RES1_P2_MASK,
+		      FIELD_PREP(PSE_IQ_RES1_P2_MASK, 0x10));
+	airoha_fe_rmw(eth, REG_PSE_IQ_REV2,
+		      PSE_IQ_RES2_P5_MASK | PSE_IQ_RES2_P4_MASK,
+		      FIELD_PREP(PSE_IQ_RES2_P5_MASK, 0x40) |
+		      FIELD_PREP(PSE_IQ_RES2_P4_MASK, 0x34));
+
+	/* enable FE copy engine for MC/KA/DPI */
+	airoha_fe_wr(eth, REG_FE_PCE_CFG,
+		     PCE_DPI_EN_MASK | PCE_KA_EN_MASK | PCE_MC_EN_MASK);
+	/* set vip queue selection to ring 1 */
+	airoha_fe_rmw(eth, REG_CDM1_FWD_CFG, CDM1_VIP_QSEL_MASK,
+		      FIELD_PREP(CDM1_VIP_QSEL_MASK, 0x4));
+	airoha_fe_rmw(eth, REG_CDM2_FWD_CFG, CDM2_VIP_QSEL_MASK,
+		      FIELD_PREP(CDM2_VIP_QSEL_MASK, 0x4));
+	/* set GDM4 source interface offset to 8 */
+	airoha_fe_rmw(eth, REG_GDM4_SRC_PORT_SET,
+		      GDM4_SPORT_OFF2_MASK |
+		      GDM4_SPORT_OFF1_MASK |
+		      GDM4_SPORT_OFF0_MASK,
+		      FIELD_PREP(GDM4_SPORT_OFF2_MASK, 8) |
+		      FIELD_PREP(GDM4_SPORT_OFF1_MASK, 8) |
+		      FIELD_PREP(GDM4_SPORT_OFF0_MASK, 8));
+
+	/* set PSE Page as 128B */
+	airoha_fe_rmw(eth, REG_FE_DMA_GLO_CFG,
+		      FE_DMA_GLO_L2_SPACE_MASK | FE_DMA_GLO_PG_SZ_MASK,
+		      FIELD_PREP(FE_DMA_GLO_L2_SPACE_MASK, 2) |
+		      FE_DMA_GLO_PG_SZ_MASK);
+	airoha_fe_wr(eth, REG_FE_RST_GLO_CFG,
+		     FE_RST_CORE_MASK | FE_RST_GDM3_MBI_ARB_MASK |
+		     FE_RST_GDM4_MBI_ARB_MASK);
+	usleep_range(1000, 2000);
+
+	/* connect RxRing1 and RxRing15 to PSE Port0 OQ-1
+	 * connect other rings to PSE Port0 OQ-0
+	 */
+	airoha_fe_wr(eth, REG_FE_CDM1_OQ_MAP0, BIT(4));
+	airoha_fe_wr(eth, REG_FE_CDM1_OQ_MAP1, BIT(28));
+	airoha_fe_wr(eth, REG_FE_CDM1_OQ_MAP2, BIT(4));
+	airoha_fe_wr(eth, REG_FE_CDM1_OQ_MAP3, BIT(28));
+
+	airoha_fe_vip_setup(eth);
+	airoha_fe_pse_ports_init(eth);
+
+	airoha_fe_set(eth, REG_GDM_MISC_CFG,
+		      GDM2_RDM_ACK_WAIT_PREF_MASK |
+		      GDM2_CHN_VLD_MODE_MASK);
+	airoha_fe_rmw(eth, REG_CDM2_FWD_CFG, CDM2_OAM_QSEL_MASK,
+		      FIELD_PREP(CDM2_OAM_QSEL_MASK, 15));
+
+	/* init fragment and assemble Force Port */
+	/* NPU Core-3, NPU Bridge Channel-3 */
+	airoha_fe_rmw(eth, REG_IP_FRAG_FP,
+		      IP_FRAGMENT_PORT_MASK | IP_FRAGMENT_NBQ_MASK,
+		      FIELD_PREP(IP_FRAGMENT_PORT_MASK, 6) |
+		      FIELD_PREP(IP_FRAGMENT_NBQ_MASK, 3));
+	/* QDMA LAN, RX Ring-22 */
+	airoha_fe_rmw(eth, REG_IP_FRAG_FP,
+		      IP_ASSEMBLE_PORT_MASK | IP_ASSEMBLE_NBQ_MASK,
+		      FIELD_PREP(IP_ASSEMBLE_PORT_MASK, 0) |
+		      FIELD_PREP(IP_ASSEMBLE_NBQ_MASK, 22));
+
+	airoha_fe_set(eth, REG_GDM3_FWD_CFG, GDM3_PAD_EN_MASK);
+	airoha_fe_set(eth, REG_GDM4_FWD_CFG, GDM4_PAD_EN_MASK);
+
+	airoha_fe_crsn_qsel_init(eth);
+
+	airoha_fe_clear(eth, REG_FE_CPORT_CFG, FE_CPORT_QUEUE_XFC_MASK);
+	airoha_fe_set(eth, REG_FE_CPORT_CFG, FE_CPORT_PORT_XFC_MASK);
+
+	/* default aging mode for mbi unlock issue */
+	airoha_fe_rmw(eth, REG_GDM2_CHN_RLS,
+		      MBI_RX_AGE_SEL_MASK | MBI_TX_AGE_SEL_MASK,
+		      FIELD_PREP(MBI_RX_AGE_SEL_MASK, 3) |
+		      FIELD_PREP(MBI_TX_AGE_SEL_MASK, 3));
+
+	/* disable IFC by default */
+	airoha_fe_clear(eth, REG_FE_CSR_IFC_CFG, FE_IFC_EN_MASK);
+
+	airoha_fe_wr(eth, REG_PPE_DFT_CPORT0(0),
+		     FIELD_PREP(DFT_CPORT_MASK(7), FE_PSE_PORT_CDM1) |
+		     FIELD_PREP(DFT_CPORT_MASK(6), FE_PSE_PORT_CDM1) |
+		     FIELD_PREP(DFT_CPORT_MASK(5), FE_PSE_PORT_CDM1) |
+		     FIELD_PREP(DFT_CPORT_MASK(4), FE_PSE_PORT_CDM1) |
+		     FIELD_PREP(DFT_CPORT_MASK(3), FE_PSE_PORT_CDM1) |
+		     FIELD_PREP(DFT_CPORT_MASK(2), FE_PSE_PORT_CDM1) |
+		     FIELD_PREP(DFT_CPORT_MASK(1), FE_PSE_PORT_CDM1) |
+		     FIELD_PREP(DFT_CPORT_MASK(0), FE_PSE_PORT_CDM1));
+	airoha_fe_wr(eth, REG_PPE_DFT_CPORT0(1),
+		     FIELD_PREP(DFT_CPORT_MASK(7), FE_PSE_PORT_CDM2) |
+		     FIELD_PREP(DFT_CPORT_MASK(6), FE_PSE_PORT_CDM2) |
+		     FIELD_PREP(DFT_CPORT_MASK(5), FE_PSE_PORT_CDM2) |
+		     FIELD_PREP(DFT_CPORT_MASK(4), FE_PSE_PORT_CDM2) |
+		     FIELD_PREP(DFT_CPORT_MASK(3), FE_PSE_PORT_CDM2) |
+		     FIELD_PREP(DFT_CPORT_MASK(2), FE_PSE_PORT_CDM2) |
+		     FIELD_PREP(DFT_CPORT_MASK(1), FE_PSE_PORT_CDM2) |
+		     FIELD_PREP(DFT_CPORT_MASK(0), FE_PSE_PORT_CDM2));
+
+	/* enable 1:N vlan action, init vlan table */
+	airoha_fe_set(eth, REG_MC_VLAN_EN, MC_VLAN_EN_MASK);
+
+	return airoha_fe_mc_vlan_clear(eth);
+}
+
+static int airoha_qdma_fill_rx_queue(struct airoha_queue *q)
+{
+	struct airoha_qdma *qdma = q->qdma;
+	int qid = q - &qdma->q_rx[0];
+	int nframes = 0;
+
+	while (q->queued < q->ndesc - 1) {
+		struct airoha_queue_entry *e = &q->entry[q->head];
+		struct airoha_qdma_desc *desc = &q->desc[q->head];
+		struct page *page;
+		int offset;
+		u32 val;
+
+		page = page_pool_dev_alloc_frag(q->page_pool, &offset,
+						q->buf_size);
+		if (!page)
+			break;
+
+		q->head = (q->head + 1) % q->ndesc;
+		q->queued++;
+		nframes++;
+
+		e->buf = page_address(page) + offset;
+		e->dma_addr = page_pool_get_dma_addr(page) + offset;
+		e->dma_len = SKB_WITH_OVERHEAD(q->buf_size);
+
+		val = FIELD_PREP(QDMA_DESC_LEN_MASK, e->dma_len);
+		WRITE_ONCE(desc->ctrl, cpu_to_le32(val));
+		WRITE_ONCE(desc->addr, cpu_to_le32(e->dma_addr));
+		val = FIELD_PREP(QDMA_DESC_NEXT_ID_MASK, q->head);
+		WRITE_ONCE(desc->data, cpu_to_le32(val));
+		WRITE_ONCE(desc->msg0, 0);
+		WRITE_ONCE(desc->msg1, 0);
+		WRITE_ONCE(desc->msg2, 0);
+		WRITE_ONCE(desc->msg3, 0);
+
+		airoha_qdma_rmw(qdma, REG_RX_CPU_IDX(qid),
+				RX_RING_CPU_IDX_MASK,
+				FIELD_PREP(RX_RING_CPU_IDX_MASK, q->head));
+	}
+
+	return nframes;
+}
+
+static int airoha_qdma_get_gdm_port(struct airoha_eth *eth,
+				    struct airoha_qdma_desc *desc)
+{
+	u32 port, sport, msg1 = le32_to_cpu(desc->msg1);
+
+	sport = FIELD_GET(QDMA_ETH_RXMSG_SPORT_MASK, msg1);
+	switch (sport) {
+	case 0x10 ... 0x14:
+		port = 0;
+		break;
+	case 0x2 ... 0x4:
+		port = sport - 1;
+		break;
+	default:
+		return -EINVAL;
+	}
+
+	return port >= ARRAY_SIZE(eth->ports) ? -EINVAL : port;
+}
+
+static int airoha_qdma_rx_process(struct airoha_queue *q, int budget)
+{
+	enum dma_data_direction dir = page_pool_get_dma_dir(q->page_pool);
+	struct airoha_qdma *qdma = q->qdma;
+	struct airoha_eth *eth = qdma->eth;
+	int qid = q - &qdma->q_rx[0];
+	int done = 0;
+
+	while (done < budget) {
+		struct airoha_queue_entry *e = &q->entry[q->tail];
+		struct airoha_qdma_desc *desc = &q->desc[q->tail];
+		u32 hash, reason, msg1 = le32_to_cpu(desc->msg1);
+		struct page *page = virt_to_head_page(e->buf);
+		u32 desc_ctrl = le32_to_cpu(desc->ctrl);
+		struct airoha_gdm_port *port;
+		int data_len, len, p;
+
+		if (!(desc_ctrl & QDMA_DESC_DONE_MASK))
+			break;
+
+		q->tail = (q->tail + 1) % q->ndesc;
+		q->queued--;
+
+		dma_sync_single_for_cpu(eth->dev, e->dma_addr,
+					SKB_WITH_OVERHEAD(q->buf_size), dir);
+
+		len = FIELD_GET(QDMA_DESC_LEN_MASK, desc_ctrl);
+		data_len = q->skb ? q->buf_size
+				  : SKB_WITH_OVERHEAD(q->buf_size);
+		if (!len || data_len < len)
+			goto free_frag;
+
+		p = airoha_qdma_get_gdm_port(eth, desc);
+		if (p < 0 || !eth->ports[p])
+			goto free_frag;
+
+		port = eth->ports[p];
+		if (!q->skb) { /* first buffer */
+			q->skb = napi_build_skb(e->buf, q->buf_size);
+			if (!q->skb)
+				goto free_frag;
+
+			__skb_put(q->skb, len);
+			skb_mark_for_recycle(q->skb);
+			q->skb->dev = port->dev;
+			q->skb->protocol = eth_type_trans(q->skb, port->dev);
+			q->skb->ip_summed = CHECKSUM_UNNECESSARY;
+			skb_record_rx_queue(q->skb, qid);
+		} else { /* scattered frame */
+			struct skb_shared_info *shinfo = skb_shinfo(q->skb);
+			int nr_frags = shinfo->nr_frags;
+
+			if (nr_frags >= ARRAY_SIZE(shinfo->frags))
+				goto free_frag;
+
+			skb_add_rx_frag(q->skb, nr_frags, page,
+					e->buf - page_address(page), len,
+					q->buf_size);
+		}
+
+		if (FIELD_GET(QDMA_DESC_MORE_MASK, desc_ctrl))
+			continue;
+
+		if (netdev_uses_dsa(port->dev)) {
+			/* PPE module requires untagged packets to work
+			 * properly and it provides DSA port index via the
+			 * DMA descriptor. Report DSA tag to the DSA stack
+			 * via skb dst info.
+			 */
+			u32 sptag = FIELD_GET(QDMA_ETH_RXMSG_SPTAG,
+					      le32_to_cpu(desc->msg0));
+
+			if (sptag < ARRAY_SIZE(port->dsa_meta) &&
+			    port->dsa_meta[sptag])
+				skb_dst_set_noref(q->skb,
+						  &port->dsa_meta[sptag]->dst);
+		}
+
+		hash = FIELD_GET(AIROHA_RXD4_FOE_ENTRY, msg1);
+		if (hash != AIROHA_RXD4_FOE_ENTRY)
+			skb_set_hash(q->skb, jhash_1word(hash, 0),
+				     PKT_HASH_TYPE_L4);
+
+		reason = FIELD_GET(AIROHA_RXD4_PPE_CPU_REASON, msg1);
+		if (reason == PPE_CPU_REASON_HIT_UNBIND_RATE_REACHED)
+			airoha_ppe_check_skb(&eth->ppe->dev, q->skb, hash,
+					     false);
+
+		done++;
+		napi_gro_receive(&q->napi, q->skb);
+		q->skb = NULL;
+		continue;
+free_frag:
+		if (q->skb) {
+			dev_kfree_skb(q->skb);
+			q->skb = NULL;
+		} else {
+			page_pool_put_full_page(q->page_pool, page, true);
+		}
+	}
+	airoha_qdma_fill_rx_queue(q);
+
+	return done;
+}
+
+static int airoha_qdma_rx_napi_poll(struct napi_struct *napi, int budget)
+{
+	struct airoha_queue *q = container_of(napi, struct airoha_queue, napi);
+	int cur, done = 0;
+
+	do {
+		cur = airoha_qdma_rx_process(q, budget - done);
+		done += cur;
+	} while (cur && done < budget);
+
+	if (done < budget && napi_complete(napi)) {
+		struct airoha_qdma *qdma = q->qdma;
+		int i, qid = q - &qdma->q_rx[0];
+		int intr_reg = qid < RX_DONE_HIGH_OFFSET ? QDMA_INT_REG_IDX1
+							 : QDMA_INT_REG_IDX2;
+
+		for (i = 0; i < ARRAY_SIZE(qdma->irq_banks); i++) {
+			if (!(BIT(qid) & RX_IRQ_BANK_PIN_MASK(i)))
+				continue;
+
+			airoha_qdma_irq_enable(&qdma->irq_banks[i], intr_reg,
+					       BIT(qid % RX_DONE_HIGH_OFFSET));
+		}
+	}
+
+	return done;
+}
+
+static int airoha_qdma_init_rx_queue(struct airoha_queue *q,
+				     struct airoha_qdma *qdma, int ndesc)
+{
+	const struct page_pool_params pp_params = {
+		.order = 0,
+		.pool_size = 256,
+		.flags = PP_FLAG_DMA_MAP | PP_FLAG_DMA_SYNC_DEV,
+		.dma_dir = DMA_FROM_DEVICE,
+		.max_len = PAGE_SIZE,
+		.nid = NUMA_NO_NODE,
+		.dev = qdma->eth->dev,
+		.napi = &q->napi,
+	};
+	struct airoha_eth *eth = qdma->eth;
+	int qid = q - &qdma->q_rx[0], thr;
+	dma_addr_t dma_addr;
+
+	q->buf_size = PAGE_SIZE / 2;
+	q->ndesc = ndesc;
+	q->qdma = qdma;
+
+	q->entry = devm_kzalloc(eth->dev, q->ndesc * sizeof(*q->entry),
+				GFP_KERNEL);
+	if (!q->entry)
+		return -ENOMEM;
+
+	q->page_pool = page_pool_create(&pp_params);
+	if (IS_ERR(q->page_pool)) {
+		int err = PTR_ERR(q->page_pool);
+
+		q->page_pool = NULL;
+		return err;
+	}
+
+	q->desc = dmam_alloc_coherent(eth->dev, q->ndesc * sizeof(*q->desc),
+				      &dma_addr, GFP_KERNEL);
+	if (!q->desc)
+		return -ENOMEM;
+
+	netif_napi_add(eth->napi_dev, &q->napi, airoha_qdma_rx_napi_poll);
+
+	airoha_qdma_wr(qdma, REG_RX_RING_BASE(qid), dma_addr);
+	airoha_qdma_rmw(qdma, REG_RX_RING_SIZE(qid),
+			RX_RING_SIZE_MASK,
+			FIELD_PREP(RX_RING_SIZE_MASK, ndesc));
+
+	thr = clamp(ndesc >> 3, 1, 32);
+	airoha_qdma_rmw(qdma, REG_RX_RING_SIZE(qid), RX_RING_THR_MASK,
+			FIELD_PREP(RX_RING_THR_MASK, thr));
+	airoha_qdma_rmw(qdma, REG_RX_DMA_IDX(qid), RX_RING_DMA_IDX_MASK,
+			FIELD_PREP(RX_RING_DMA_IDX_MASK, q->head));
+	airoha_qdma_set(qdma, REG_RX_SCATTER_CFG(qid), RX_RING_SG_EN_MASK);
+
+	airoha_qdma_fill_rx_queue(q);
+
+	return 0;
+}
+
+static void airoha_qdma_cleanup_rx_queue(struct airoha_queue *q)
+{
+	struct airoha_eth *eth = q->qdma->eth;
+
+	while (q->queued) {
+		struct airoha_queue_entry *e = &q->entry[q->tail];
+		struct page *page = virt_to_head_page(e->buf);
+
+		dma_sync_single_for_cpu(eth->dev, e->dma_addr, e->dma_len,
+					page_pool_get_dma_dir(q->page_pool));
+		page_pool_put_full_page(q->page_pool, page, false);
+		q->tail = (q->tail + 1) % q->ndesc;
+		q->queued--;
+	}
+}
+
+static int airoha_qdma_init_rx(struct airoha_qdma *qdma)
+{
+	int i;
+
+	for (i = 0; i < ARRAY_SIZE(qdma->q_rx); i++) {
+		int err;
+
+		if (!(RX_DONE_INT_MASK & BIT(i))) {
+			/* rx-queue not binded to irq */
+			continue;
+		}
+
+		err = airoha_qdma_init_rx_queue(&qdma->q_rx[i], qdma,
+						RX_DSCP_NUM(i));
+		if (err)
+			return err;
+	}
+
+	return 0;
+}
+
+static int airoha_qdma_tx_napi_poll(struct napi_struct *napi, int budget)
+{
+	struct airoha_tx_irq_queue *irq_q;
+	int id, done = 0, irq_queued;
+	struct airoha_qdma *qdma;
+	struct airoha_eth *eth;
+	u32 status, head;
+
+	irq_q = container_of(napi, struct airoha_tx_irq_queue, napi);
+	qdma = irq_q->qdma;
+	id = irq_q - &qdma->q_tx_irq[0];
+	eth = qdma->eth;
+
+	status = airoha_qdma_rr(qdma, REG_IRQ_STATUS(id));
+	head = FIELD_GET(IRQ_HEAD_IDX_MASK, status);
+	head = head % irq_q->size;
+	irq_queued = FIELD_GET(IRQ_ENTRY_LEN_MASK, status);
+
+	while (irq_queued > 0 && done < budget) {
+		u32 qid, val = irq_q->q[head];
+		struct airoha_qdma_desc *desc;
+		struct airoha_queue_entry *e;
+		struct airoha_queue *q;
+		u32 index, desc_ctrl;
+		struct sk_buff *skb;
+
+		if (val == 0xff)
+			break;
+
+		irq_q->q[head] = 0xff; /* mark as done */
+		head = (head + 1) % irq_q->size;
+		irq_queued--;
+		done++;
+
+		qid = FIELD_GET(IRQ_RING_IDX_MASK, val);
+		if (qid >= ARRAY_SIZE(qdma->q_tx))
+			continue;
+
+		q = &qdma->q_tx[qid];
+		if (!q->ndesc)
+			continue;
+
+		index = FIELD_GET(IRQ_DESC_IDX_MASK, val);
+		if (index >= q->ndesc)
+			continue;
+
+		spin_lock_bh(&q->lock);
+
+		if (!q->queued)
+			goto unlock;
+
+		desc = &q->desc[index];
+		desc_ctrl = le32_to_cpu(desc->ctrl);
+
+		if (!(desc_ctrl & QDMA_DESC_DONE_MASK) &&
+		    !(desc_ctrl & QDMA_DESC_DROP_MASK))
+			goto unlock;
+
+		e = &q->entry[index];
+		skb = e->skb;
+
+		dma_unmap_single(eth->dev, e->dma_addr, e->dma_len,
+				 DMA_TO_DEVICE);
+		memset(e, 0, sizeof(*e));
+		WRITE_ONCE(desc->msg0, 0);
+		WRITE_ONCE(desc->msg1, 0);
+		q->queued--;
+
+		/* completion ring can report out-of-order indexes if hw QoS
+		 * is enabled and packets with different priority are queued
+		 * to same DMA ring. Take into account possible out-of-order
+		 * reports incrementing DMA ring tail pointer
+		 */
+		while (q->tail != q->head && !q->entry[q->tail].dma_addr)
+			q->tail = (q->tail + 1) % q->ndesc;
+
+		if (skb) {
+			u16 queue = skb_get_queue_mapping(skb);
+			struct netdev_queue *txq;
+
+			txq = netdev_get_tx_queue(skb->dev, queue);
+			netdev_tx_completed_queue(txq, 1, skb->len);
+			if (netif_tx_queue_stopped(txq) &&
+			    q->ndesc - q->queued >= q->free_thr)
+				netif_tx_wake_queue(txq);
+
+			dev_kfree_skb_any(skb);
+		}
+unlock:
+		spin_unlock_bh(&q->lock);
+	}
+
+	if (done) {
+		int i, len = done >> 7;
+
+		for (i = 0; i < len; i++)
+			airoha_qdma_rmw(qdma, REG_IRQ_CLEAR_LEN(id),
+					IRQ_CLEAR_LEN_MASK, 0x80);
+		airoha_qdma_rmw(qdma, REG_IRQ_CLEAR_LEN(id),
+				IRQ_CLEAR_LEN_MASK, (done & 0x7f));
+	}
+
+	if (done < budget && napi_complete(napi))
+		airoha_qdma_irq_enable(&qdma->irq_banks[0], QDMA_INT_REG_IDX0,
+				       TX_DONE_INT_MASK(id));
+
+	return done;
+}
+
+static int airoha_qdma_init_tx_queue(struct airoha_queue *q,
+				     struct airoha_qdma *qdma, int size)
+{
+	struct airoha_eth *eth = qdma->eth;
+	int i, qid = q - &qdma->q_tx[0];
+	dma_addr_t dma_addr;
+
+	spin_lock_init(&q->lock);
+	q->ndesc = size;
+	q->qdma = qdma;
+	q->free_thr = 1 + MAX_SKB_FRAGS;
+
+	q->entry = devm_kzalloc(eth->dev, q->ndesc * sizeof(*q->entry),
+				GFP_KERNEL);
+	if (!q->entry)
+		return -ENOMEM;
+
+	q->desc = dmam_alloc_coherent(eth->dev, q->ndesc * sizeof(*q->desc),
+				      &dma_addr, GFP_KERNEL);
+	if (!q->desc)
+		return -ENOMEM;
+
+	for (i = 0; i < q->ndesc; i++) {
+		u32 val;
+
+		val = FIELD_PREP(QDMA_DESC_DONE_MASK, 1);
+		WRITE_ONCE(q->desc[i].ctrl, cpu_to_le32(val));
+	}
+
+	/* xmit ring drop default setting */
+	airoha_qdma_set(qdma, REG_TX_RING_BLOCKING(qid),
+			TX_RING_IRQ_BLOCKING_TX_DROP_EN_MASK);
+
+	airoha_qdma_wr(qdma, REG_TX_RING_BASE(qid), dma_addr);
+	airoha_qdma_rmw(qdma, REG_TX_CPU_IDX(qid), TX_RING_CPU_IDX_MASK,
+			FIELD_PREP(TX_RING_CPU_IDX_MASK, q->head));
+	airoha_qdma_rmw(qdma, REG_TX_DMA_IDX(qid), TX_RING_DMA_IDX_MASK,
+			FIELD_PREP(TX_RING_DMA_IDX_MASK, q->head));
+
+	return 0;
+}
+
+static int airoha_qdma_tx_irq_init(struct airoha_tx_irq_queue *irq_q,
+				   struct airoha_qdma *qdma, int size)
+{
+	int id = irq_q - &qdma->q_tx_irq[0];
+	struct airoha_eth *eth = qdma->eth;
+	dma_addr_t dma_addr;
+
+	netif_napi_add_tx(eth->napi_dev, &irq_q->napi,
+			  airoha_qdma_tx_napi_poll);
+	irq_q->q = dmam_alloc_coherent(eth->dev, size * sizeof(u32),
+				       &dma_addr, GFP_KERNEL);
+	if (!irq_q->q)
+		return -ENOMEM;
+
+	memset(irq_q->q, 0xff, size * sizeof(u32));
+	irq_q->size = size;
+	irq_q->qdma = qdma;
+
+	airoha_qdma_wr(qdma, REG_TX_IRQ_BASE(id), dma_addr);
+	airoha_qdma_rmw(qdma, REG_TX_IRQ_CFG(id), TX_IRQ_DEPTH_MASK,
+			FIELD_PREP(TX_IRQ_DEPTH_MASK, size));
+	airoha_qdma_rmw(qdma, REG_TX_IRQ_CFG(id), TX_IRQ_THR_MASK,
+			FIELD_PREP(TX_IRQ_THR_MASK, 1));
+
+	return 0;
+}
+
+static int airoha_qdma_init_tx(struct airoha_qdma *qdma)
+{
+	int i, err;
+
+	for (i = 0; i < ARRAY_SIZE(qdma->q_tx_irq); i++) {
+		err = airoha_qdma_tx_irq_init(&qdma->q_tx_irq[i], qdma,
+					      IRQ_QUEUE_LEN(i));
+		if (err)
+			return err;
+	}
+
+	for (i = 0; i < ARRAY_SIZE(qdma->q_tx); i++) {
+		err = airoha_qdma_init_tx_queue(&qdma->q_tx[i], qdma,
+						TX_DSCP_NUM);
+		if (err)
+			return err;
+	}
+
+	return 0;
+}
+
+static void airoha_qdma_cleanup_tx_queue(struct airoha_queue *q)
+{
+	struct airoha_eth *eth = q->qdma->eth;
+
+	spin_lock_bh(&q->lock);
+	while (q->queued) {
+		struct airoha_queue_entry *e = &q->entry[q->tail];
+
+		dma_unmap_single(eth->dev, e->dma_addr, e->dma_len,
+				 DMA_TO_DEVICE);
+		dev_kfree_skb_any(e->skb);
+		e->skb = NULL;
+
+		q->tail = (q->tail + 1) % q->ndesc;
+		q->queued--;
+	}
+	spin_unlock_bh(&q->lock);
+}
+
+static int airoha_qdma_init_hfwd_queues(struct airoha_qdma *qdma)
+{
+	int size, index, num_desc = HW_DSCP_NUM;
+	struct airoha_eth *eth = qdma->eth;
+	int id = qdma - &eth->qdma[0];
+	u32 status, buf_size;
+	dma_addr_t dma_addr;
+	const char *name;
+
+	name = devm_kasprintf(eth->dev, GFP_KERNEL, "qdma%d-buf", id);
+	if (!name)
+		return -ENOMEM;
+
+	buf_size = id ? AIROHA_MAX_PACKET_SIZE / 2 : AIROHA_MAX_PACKET_SIZE;
+	index = of_property_match_string(eth->dev->of_node,
+					 "memory-region-names", name);
+	if (index >= 0) {
+		struct reserved_mem *rmem;
+		struct device_node *np;
+
+		/* Consume reserved memory for hw forwarding buffers queue if
+		 * available in the DTS
+		 */
+		np = of_parse_phandle(eth->dev->of_node, "memory-region",
+				      index);
+		if (!np)
+			return -ENODEV;
+
+		rmem = of_reserved_mem_lookup(np);
+		of_node_put(np);
+		dma_addr = rmem->base;
+		/* Compute the number of hw descriptors according to the
+		 * reserved memory size and the payload buffer size
+		 */
+		num_desc = div_u64(rmem->size, buf_size);
+	} else {
+		size = buf_size * num_desc;
+		if (!dmam_alloc_coherent(eth->dev, size, &dma_addr,
+					 GFP_KERNEL))
+			return -ENOMEM;
+	}
+
+	airoha_qdma_wr(qdma, REG_FWD_BUF_BASE, dma_addr);
+
+	size = num_desc * sizeof(struct airoha_qdma_fwd_desc);
+	if (!dmam_alloc_coherent(eth->dev, size, &dma_addr, GFP_KERNEL))
+		return -ENOMEM;
+
+	airoha_qdma_wr(qdma, REG_FWD_DSCP_BASE, dma_addr);
+	/* QDMA0: 2KB. QDMA1: 1KB */
+	airoha_qdma_rmw(qdma, REG_HW_FWD_DSCP_CFG,
+			HW_FWD_DSCP_PAYLOAD_SIZE_MASK,
+			FIELD_PREP(HW_FWD_DSCP_PAYLOAD_SIZE_MASK, !!id));
+	airoha_qdma_rmw(qdma, REG_FWD_DSCP_LOW_THR, FWD_DSCP_LOW_THR_MASK,
+			FIELD_PREP(FWD_DSCP_LOW_THR_MASK, 128));
+	airoha_qdma_rmw(qdma, REG_LMGR_INIT_CFG,
+			LMGR_INIT_START | LMGR_SRAM_MODE_MASK |
+			HW_FWD_DESC_NUM_MASK,
+			FIELD_PREP(HW_FWD_DESC_NUM_MASK, num_desc) |
+			LMGR_INIT_START | LMGR_SRAM_MODE_MASK);
+
+	return read_poll_timeout(airoha_qdma_rr, status,
+				 !(status & LMGR_INIT_START), USEC_PER_MSEC,
+				 30 * USEC_PER_MSEC, true, qdma,
+				 REG_LMGR_INIT_CFG);
+}
+
+static void airoha_qdma_init_qos(struct airoha_qdma *qdma)
+{
+	airoha_qdma_clear(qdma, REG_TXWRR_MODE_CFG, TWRR_WEIGHT_SCALE_MASK);
+	airoha_qdma_set(qdma, REG_TXWRR_MODE_CFG, TWRR_WEIGHT_BASE_MASK);
+
+	airoha_qdma_clear(qdma, REG_PSE_BUF_USAGE_CFG,
+			  PSE_BUF_ESTIMATE_EN_MASK);
+
+	airoha_qdma_set(qdma, REG_EGRESS_RATE_METER_CFG,
+			EGRESS_RATE_METER_EN_MASK |
+			EGRESS_RATE_METER_EQ_RATE_EN_MASK);
+	/* 2047us x 31 = 63.457ms */
+	airoha_qdma_rmw(qdma, REG_EGRESS_RATE_METER_CFG,
+			EGRESS_RATE_METER_WINDOW_SZ_MASK,
+			FIELD_PREP(EGRESS_RATE_METER_WINDOW_SZ_MASK, 0x1f));
+	airoha_qdma_rmw(qdma, REG_EGRESS_RATE_METER_CFG,
+			EGRESS_RATE_METER_TIMESLICE_MASK,
+			FIELD_PREP(EGRESS_RATE_METER_TIMESLICE_MASK, 0x7ff));
+
+	/* ratelimit init */
+	airoha_qdma_set(qdma, REG_GLB_TRTCM_CFG, GLB_TRTCM_EN_MASK);
+	/* fast-tick 25us */
+	airoha_qdma_rmw(qdma, REG_GLB_TRTCM_CFG, GLB_FAST_TICK_MASK,
+			FIELD_PREP(GLB_FAST_TICK_MASK, 25));
+	airoha_qdma_rmw(qdma, REG_GLB_TRTCM_CFG, GLB_SLOW_TICK_RATIO_MASK,
+			FIELD_PREP(GLB_SLOW_TICK_RATIO_MASK, 40));
+
+	airoha_qdma_set(qdma, REG_EGRESS_TRTCM_CFG, EGRESS_TRTCM_EN_MASK);
+	airoha_qdma_rmw(qdma, REG_EGRESS_TRTCM_CFG, EGRESS_FAST_TICK_MASK,
+			FIELD_PREP(EGRESS_FAST_TICK_MASK, 25));
+	airoha_qdma_rmw(qdma, REG_EGRESS_TRTCM_CFG,
+			EGRESS_SLOW_TICK_RATIO_MASK,
+			FIELD_PREP(EGRESS_SLOW_TICK_RATIO_MASK, 40));
+
+	airoha_qdma_set(qdma, REG_INGRESS_TRTCM_CFG, INGRESS_TRTCM_EN_MASK);
+	airoha_qdma_clear(qdma, REG_INGRESS_TRTCM_CFG,
+			  INGRESS_TRTCM_MODE_MASK);
+	airoha_qdma_rmw(qdma, REG_INGRESS_TRTCM_CFG, INGRESS_FAST_TICK_MASK,
+			FIELD_PREP(INGRESS_FAST_TICK_MASK, 125));
+	airoha_qdma_rmw(qdma, REG_INGRESS_TRTCM_CFG,
+			INGRESS_SLOW_TICK_RATIO_MASK,
+			FIELD_PREP(INGRESS_SLOW_TICK_RATIO_MASK, 8));
+
+	airoha_qdma_set(qdma, REG_SLA_TRTCM_CFG, SLA_TRTCM_EN_MASK);
+	airoha_qdma_rmw(qdma, REG_SLA_TRTCM_CFG, SLA_FAST_TICK_MASK,
+			FIELD_PREP(SLA_FAST_TICK_MASK, 25));
+	airoha_qdma_rmw(qdma, REG_SLA_TRTCM_CFG, SLA_SLOW_TICK_RATIO_MASK,
+			FIELD_PREP(SLA_SLOW_TICK_RATIO_MASK, 40));
+}
+
+static void airoha_qdma_init_qos_stats(struct airoha_qdma *qdma)
+{
+	int i;
+
+	for (i = 0; i < AIROHA_NUM_QOS_CHANNELS; i++) {
+		/* Tx-cpu transferred count */
+		airoha_qdma_wr(qdma, REG_CNTR_VAL(i << 1), 0);
+		airoha_qdma_wr(qdma, REG_CNTR_CFG(i << 1),
+			       CNTR_EN_MASK | CNTR_ALL_QUEUE_EN_MASK |
+			       CNTR_ALL_DSCP_RING_EN_MASK |
+			       FIELD_PREP(CNTR_CHAN_MASK, i));
+		/* Tx-fwd transferred count */
+		airoha_qdma_wr(qdma, REG_CNTR_VAL((i << 1) + 1), 0);
+		airoha_qdma_wr(qdma, REG_CNTR_CFG(i << 1),
+			       CNTR_EN_MASK | CNTR_ALL_QUEUE_EN_MASK |
+			       CNTR_ALL_DSCP_RING_EN_MASK |
+			       FIELD_PREP(CNTR_SRC_MASK, 1) |
+			       FIELD_PREP(CNTR_CHAN_MASK, i));
+	}
+}
+
+static int airoha_qdma_hw_init(struct airoha_qdma *qdma)
+{
+	int i;
+
+	for (i = 0; i < ARRAY_SIZE(qdma->irq_banks); i++) {
+		/* clear pending irqs */
+		airoha_qdma_wr(qdma, REG_INT_STATUS(i), 0xffffffff);
+		/* setup rx irqs */
+		airoha_qdma_irq_enable(&qdma->irq_banks[i], QDMA_INT_REG_IDX0,
+				       INT_RX0_MASK(RX_IRQ_BANK_PIN_MASK(i)));
+		airoha_qdma_irq_enable(&qdma->irq_banks[i], QDMA_INT_REG_IDX1,
+				       INT_RX1_MASK(RX_IRQ_BANK_PIN_MASK(i)));
+		airoha_qdma_irq_enable(&qdma->irq_banks[i], QDMA_INT_REG_IDX2,
+				       INT_RX2_MASK(RX_IRQ_BANK_PIN_MASK(i)));
+		airoha_qdma_irq_enable(&qdma->irq_banks[i], QDMA_INT_REG_IDX3,
+				       INT_RX3_MASK(RX_IRQ_BANK_PIN_MASK(i)));
+	}
+	/* setup tx irqs */
+	airoha_qdma_irq_enable(&qdma->irq_banks[0], QDMA_INT_REG_IDX0,
+			       TX_COHERENT_LOW_INT_MASK | INT_TX_MASK);
+	airoha_qdma_irq_enable(&qdma->irq_banks[0], QDMA_INT_REG_IDX4,
+			       TX_COHERENT_HIGH_INT_MASK);
+
+	/* setup irq binding */
+	for (i = 0; i < ARRAY_SIZE(qdma->q_tx); i++) {
+		if (!qdma->q_tx[i].ndesc)
+			continue;
+
+		if (TX_RING_IRQ_BLOCKING_MAP_MASK & BIT(i))
+			airoha_qdma_set(qdma, REG_TX_RING_BLOCKING(i),
+					TX_RING_IRQ_BLOCKING_CFG_MASK);
+		else
+			airoha_qdma_clear(qdma, REG_TX_RING_BLOCKING(i),
+					  TX_RING_IRQ_BLOCKING_CFG_MASK);
+	}
+
+	airoha_qdma_wr(qdma, REG_QDMA_GLOBAL_CFG,
+		       FIELD_PREP(GLOBAL_CFG_DMA_PREFERENCE_MASK, 3) |
+		       GLOBAL_CFG_CPU_TXR_RR_MASK |
+		       GLOBAL_CFG_PAYLOAD_BYTE_SWAP_MASK |
+		       GLOBAL_CFG_MULTICAST_MODIFY_FP_MASK |
+		       GLOBAL_CFG_MULTICAST_EN_MASK |
+		       GLOBAL_CFG_IRQ0_EN_MASK | GLOBAL_CFG_IRQ1_EN_MASK |
+		       GLOBAL_CFG_TX_WB_DONE_MASK |
+		       FIELD_PREP(GLOBAL_CFG_MAX_ISSUE_NUM_MASK, 2));
+
+	airoha_qdma_init_qos(qdma);
+
+	/* disable qdma rx delay interrupt */
+	for (i = 0; i < ARRAY_SIZE(qdma->q_rx); i++) {
+		if (!qdma->q_rx[i].ndesc)
+			continue;
+
+		airoha_qdma_clear(qdma, REG_RX_DELAY_INT_IDX(i),
+				  RX_DELAY_INT_MASK);
+	}
+
+	airoha_qdma_set(qdma, REG_TXQ_CNGST_CFG,
+			TXQ_CNGST_DROP_EN | TXQ_CNGST_DEI_DROP_EN);
+	airoha_qdma_init_qos_stats(qdma);
+
+	return 0;
+}
+
+static irqreturn_t airoha_irq_handler(int irq, void *dev_instance)
+{
+	struct airoha_irq_bank *irq_bank = dev_instance;
+	struct airoha_qdma *qdma = irq_bank->qdma;
+	u32 rx_intr_mask = 0, rx_intr1, rx_intr2;
+	u32 intr[ARRAY_SIZE(irq_bank->irqmask)];
+	int i;
+
+	for (i = 0; i < ARRAY_SIZE(intr); i++) {
+		intr[i] = airoha_qdma_rr(qdma, REG_INT_STATUS(i));
+		intr[i] &= irq_bank->irqmask[i];
+		airoha_qdma_wr(qdma, REG_INT_STATUS(i), intr[i]);
+	}
+
+	if (!test_bit(DEV_STATE_INITIALIZED, &qdma->eth->state))
+		return IRQ_NONE;
+
+	rx_intr1 = intr[1] & RX_DONE_LOW_INT_MASK;
+	if (rx_intr1) {
+		airoha_qdma_irq_disable(irq_bank, QDMA_INT_REG_IDX1, rx_intr1);
+		rx_intr_mask |= rx_intr1;
+	}
+
+	rx_intr2 = intr[2] & RX_DONE_HIGH_INT_MASK;
+	if (rx_intr2) {
+		airoha_qdma_irq_disable(irq_bank, QDMA_INT_REG_IDX2, rx_intr2);
+		rx_intr_mask |= (rx_intr2 << 16);
+	}
+
+	for (i = 0; rx_intr_mask && i < ARRAY_SIZE(qdma->q_rx); i++) {
+		if (!qdma->q_rx[i].ndesc)
+			continue;
+
+		if (rx_intr_mask & BIT(i))
+			napi_schedule(&qdma->q_rx[i].napi);
+	}
+
+	if (intr[0] & INT_TX_MASK) {
+		for (i = 0; i < ARRAY_SIZE(qdma->q_tx_irq); i++) {
+			if (!(intr[0] & TX_DONE_INT_MASK(i)))
+				continue;
+
+			airoha_qdma_irq_disable(irq_bank, QDMA_INT_REG_IDX0,
+						TX_DONE_INT_MASK(i));
+			napi_schedule(&qdma->q_tx_irq[i].napi);
+		}
+	}
+
+	return IRQ_HANDLED;
+}
+
+static int airoha_qdma_init_irq_banks(struct platform_device *pdev,
+				      struct airoha_qdma *qdma)
+{
+	struct airoha_eth *eth = qdma->eth;
+	int i, id = qdma - &eth->qdma[0];
+
+	for (i = 0; i < ARRAY_SIZE(qdma->irq_banks); i++) {
+		struct airoha_irq_bank *irq_bank = &qdma->irq_banks[i];
+		int err, irq_index = 4 * id + i;
+		const char *name;
+
+		spin_lock_init(&irq_bank->irq_lock);
+		irq_bank->qdma = qdma;
+
+		irq_bank->irq = platform_get_irq(pdev, irq_index);
+		if (irq_bank->irq < 0)
+			return irq_bank->irq;
+
+		name = devm_kasprintf(eth->dev, GFP_KERNEL,
+				      KBUILD_MODNAME ".%d", irq_index);
+		if (!name)
+			return -ENOMEM;
+
+		err = devm_request_irq(eth->dev, irq_bank->irq,
+				       airoha_irq_handler, IRQF_SHARED, name,
+				       irq_bank);
+		if (err)
+			return err;
+	}
+
+	return 0;
+}
+
+static int airoha_qdma_init(struct platform_device *pdev,
+			    struct airoha_eth *eth,
+			    struct airoha_qdma *qdma)
+{
+	int err, id = qdma - &eth->qdma[0];
+	const char *res;
+
+	qdma->eth = eth;
+	res = devm_kasprintf(eth->dev, GFP_KERNEL, "qdma%d", id);
+	if (!res)
+		return -ENOMEM;
+
+	qdma->regs = devm_platform_ioremap_resource_byname(pdev, res);
+	if (IS_ERR(qdma->regs))
+		return dev_err_probe(eth->dev, PTR_ERR(qdma->regs),
+				     "failed to iomap qdma%d regs\n", id);
+
+	err = airoha_qdma_init_irq_banks(pdev, qdma);
+	if (err)
+		return err;
+
+	err = airoha_qdma_init_rx(qdma);
+	if (err)
+		return err;
+
+	err = airoha_qdma_init_tx(qdma);
+	if (err)
+		return err;
+
+	err = airoha_qdma_init_hfwd_queues(qdma);
+	if (err)
+		return err;
+
+	return airoha_qdma_hw_init(qdma);
+}
+
+static int airoha_hw_init(struct platform_device *pdev,
+			  struct airoha_eth *eth)
+{
+	int err, i;
+
+	/* disable xsi */
+	err = reset_control_bulk_assert(ARRAY_SIZE(eth->xsi_rsts),
+					eth->xsi_rsts);
+	if (err)
+		return err;
+
+	err = reset_control_bulk_assert(ARRAY_SIZE(eth->rsts), eth->rsts);
+	if (err)
+		return err;
+
+	msleep(20);
+	err = reset_control_bulk_deassert(ARRAY_SIZE(eth->rsts), eth->rsts);
+	if (err)
+		return err;
+
+	msleep(20);
+	err = airoha_fe_init(eth);
+	if (err)
+		return err;
+
+	for (i = 0; i < ARRAY_SIZE(eth->qdma); i++) {
+		err = airoha_qdma_init(pdev, eth, &eth->qdma[i]);
+		if (err)
+			return err;
+	}
+
+	err = airoha_ppe_init(eth);
+	if (err)
+		return err;
+
+	set_bit(DEV_STATE_INITIALIZED, &eth->state);
+
+	return 0;
+}
+
+static void airoha_hw_cleanup(struct airoha_qdma *qdma)
+{
+	int i;
+
+	for (i = 0; i < ARRAY_SIZE(qdma->q_rx); i++) {
+		if (!qdma->q_rx[i].ndesc)
+			continue;
+
+		netif_napi_del(&qdma->q_rx[i].napi);
+		airoha_qdma_cleanup_rx_queue(&qdma->q_rx[i]);
+		if (qdma->q_rx[i].page_pool)
+			page_pool_destroy(qdma->q_rx[i].page_pool);
+	}
+
+	for (i = 0; i < ARRAY_SIZE(qdma->q_tx_irq); i++)
+		netif_napi_del(&qdma->q_tx_irq[i].napi);
+
+	for (i = 0; i < ARRAY_SIZE(qdma->q_tx); i++) {
+		if (!qdma->q_tx[i].ndesc)
+			continue;
+
+		airoha_qdma_cleanup_tx_queue(&qdma->q_tx[i]);
+	}
+}
+
+static void airoha_qdma_start_napi(struct airoha_qdma *qdma)
+{
+	int i;
+
+	for (i = 0; i < ARRAY_SIZE(qdma->q_tx_irq); i++)
+		napi_enable(&qdma->q_tx_irq[i].napi);
+
+	for (i = 0; i < ARRAY_SIZE(qdma->q_rx); i++) {
+		if (!qdma->q_rx[i].ndesc)
+			continue;
+
+		napi_enable(&qdma->q_rx[i].napi);
+	}
+}
+
+static void airoha_qdma_stop_napi(struct airoha_qdma *qdma)
+{
+	int i;
+
+	for (i = 0; i < ARRAY_SIZE(qdma->q_tx_irq); i++)
+		napi_disable(&qdma->q_tx_irq[i].napi);
+
+	for (i = 0; i < ARRAY_SIZE(qdma->q_rx); i++) {
+		if (!qdma->q_rx[i].ndesc)
+			continue;
+
+		napi_disable(&qdma->q_rx[i].napi);
+	}
+}
+
+static void airoha_update_hw_stats(struct airoha_gdm_port *port)
+{
+	struct airoha_eth *eth = port->qdma->eth;
+	u32 val, i = 0;
+
+	spin_lock(&port->stats.lock);
+	u64_stats_update_begin(&port->stats.syncp);
+
+	/* TX */
+	val = airoha_fe_rr(eth, REG_FE_GDM_TX_OK_PKT_CNT_H(port->id));
+	port->stats.tx_ok_pkts += ((u64)val << 32);
+	val = airoha_fe_rr(eth, REG_FE_GDM_TX_OK_PKT_CNT_L(port->id));
+	port->stats.tx_ok_pkts += val;
+
+	val = airoha_fe_rr(eth, REG_FE_GDM_TX_OK_BYTE_CNT_H(port->id));
+	port->stats.tx_ok_bytes += ((u64)val << 32);
+	val = airoha_fe_rr(eth, REG_FE_GDM_TX_OK_BYTE_CNT_L(port->id));
+	port->stats.tx_ok_bytes += val;
+
+	val = airoha_fe_rr(eth, REG_FE_GDM_TX_ETH_DROP_CNT(port->id));
+	port->stats.tx_drops += val;
+
+	val = airoha_fe_rr(eth, REG_FE_GDM_TX_ETH_BC_CNT(port->id));
+	port->stats.tx_broadcast += val;
+
+	val = airoha_fe_rr(eth, REG_FE_GDM_TX_ETH_MC_CNT(port->id));
+	port->stats.tx_multicast += val;
+
+	val = airoha_fe_rr(eth, REG_FE_GDM_TX_ETH_RUNT_CNT(port->id));
+	port->stats.tx_len[i] += val;
+
+	val = airoha_fe_rr(eth, REG_FE_GDM_TX_ETH_E64_CNT_H(port->id));
+	port->stats.tx_len[i] += ((u64)val << 32);
+	val = airoha_fe_rr(eth, REG_FE_GDM_TX_ETH_E64_CNT_L(port->id));
+	port->stats.tx_len[i++] += val;
+
+	val = airoha_fe_rr(eth, REG_FE_GDM_TX_ETH_L64_CNT_H(port->id));
+	port->stats.tx_len[i] += ((u64)val << 32);
+	val = airoha_fe_rr(eth, REG_FE_GDM_TX_ETH_L64_CNT_L(port->id));
+	port->stats.tx_len[i++] += val;
+
+	val = airoha_fe_rr(eth, REG_FE_GDM_TX_ETH_L127_CNT_H(port->id));
+	port->stats.tx_len[i] += ((u64)val << 32);
+	val = airoha_fe_rr(eth, REG_FE_GDM_TX_ETH_L127_CNT_L(port->id));
+	port->stats.tx_len[i++] += val;
+
+	val = airoha_fe_rr(eth, REG_FE_GDM_TX_ETH_L255_CNT_H(port->id));
+	port->stats.tx_len[i] += ((u64)val << 32);
+	val = airoha_fe_rr(eth, REG_FE_GDM_TX_ETH_L255_CNT_L(port->id));
+	port->stats.tx_len[i++] += val;
+
+	val = airoha_fe_rr(eth, REG_FE_GDM_TX_ETH_L511_CNT_H(port->id));
+	port->stats.tx_len[i] += ((u64)val << 32);
+	val = airoha_fe_rr(eth, REG_FE_GDM_TX_ETH_L511_CNT_L(port->id));
+	port->stats.tx_len[i++] += val;
+
+	val = airoha_fe_rr(eth, REG_FE_GDM_TX_ETH_L1023_CNT_H(port->id));
+	port->stats.tx_len[i] += ((u64)val << 32);
+	val = airoha_fe_rr(eth, REG_FE_GDM_TX_ETH_L1023_CNT_L(port->id));
+	port->stats.tx_len[i++] += val;
+
+	val = airoha_fe_rr(eth, REG_FE_GDM_TX_ETH_LONG_CNT(port->id));
+	port->stats.tx_len[i++] += val;
+
+	/* RX */
+	val = airoha_fe_rr(eth, REG_FE_GDM_RX_OK_PKT_CNT_H(port->id));
+	port->stats.rx_ok_pkts += ((u64)val << 32);
+	val = airoha_fe_rr(eth, REG_FE_GDM_RX_OK_PKT_CNT_L(port->id));
+	port->stats.rx_ok_pkts += val;
+
+	val = airoha_fe_rr(eth, REG_FE_GDM_RX_OK_BYTE_CNT_H(port->id));
+	port->stats.rx_ok_bytes += ((u64)val << 32);
+	val = airoha_fe_rr(eth, REG_FE_GDM_RX_OK_BYTE_CNT_L(port->id));
+	port->stats.rx_ok_bytes += val;
+
+	val = airoha_fe_rr(eth, REG_FE_GDM_RX_ETH_DROP_CNT(port->id));
+	port->stats.rx_drops += val;
+
+	val = airoha_fe_rr(eth, REG_FE_GDM_RX_ETH_BC_CNT(port->id));
+	port->stats.rx_broadcast += val;
+
+	val = airoha_fe_rr(eth, REG_FE_GDM_RX_ETH_MC_CNT(port->id));
+	port->stats.rx_multicast += val;
+
+	val = airoha_fe_rr(eth, REG_FE_GDM_RX_ERROR_DROP_CNT(port->id));
+	port->stats.rx_errors += val;
+
+	val = airoha_fe_rr(eth, REG_FE_GDM_RX_ETH_CRC_ERR_CNT(port->id));
+	port->stats.rx_crc_error += val;
+
+	val = airoha_fe_rr(eth, REG_FE_GDM_RX_OVERFLOW_DROP_CNT(port->id));
+	port->stats.rx_over_errors += val;
+
+	val = airoha_fe_rr(eth, REG_FE_GDM_RX_ETH_FRAG_CNT(port->id));
+	port->stats.rx_fragment += val;
+
+	val = airoha_fe_rr(eth, REG_FE_GDM_RX_ETH_JABBER_CNT(port->id));
+	port->stats.rx_jabber += val;
+
+	i = 0;
+	val = airoha_fe_rr(eth, REG_FE_GDM_RX_ETH_RUNT_CNT(port->id));
+	port->stats.rx_len[i] += val;
+
+	val = airoha_fe_rr(eth, REG_FE_GDM_RX_ETH_E64_CNT_H(port->id));
+	port->stats.rx_len[i] += ((u64)val << 32);
+	val = airoha_fe_rr(eth, REG_FE_GDM_RX_ETH_E64_CNT_L(port->id));
+	port->stats.rx_len[i++] += val;
+
+	val = airoha_fe_rr(eth, REG_FE_GDM_RX_ETH_L64_CNT_H(port->id));
+	port->stats.rx_len[i] += ((u64)val << 32);
+	val = airoha_fe_rr(eth, REG_FE_GDM_RX_ETH_L64_CNT_L(port->id));
+	port->stats.rx_len[i++] += val;
+
+	val = airoha_fe_rr(eth, REG_FE_GDM_RX_ETH_L127_CNT_H(port->id));
+	port->stats.rx_len[i] += ((u64)val << 32);
+	val = airoha_fe_rr(eth, REG_FE_GDM_RX_ETH_L127_CNT_L(port->id));
+	port->stats.rx_len[i++] += val;
+
+	val = airoha_fe_rr(eth, REG_FE_GDM_RX_ETH_L255_CNT_H(port->id));
+	port->stats.rx_len[i] += ((u64)val << 32);
+	val = airoha_fe_rr(eth, REG_FE_GDM_RX_ETH_L255_CNT_L(port->id));
+	port->stats.rx_len[i++] += val;
+
+	val = airoha_fe_rr(eth, REG_FE_GDM_RX_ETH_L511_CNT_H(port->id));
+	port->stats.rx_len[i] += ((u64)val << 32);
+	val = airoha_fe_rr(eth, REG_FE_GDM_RX_ETH_L511_CNT_L(port->id));
+	port->stats.rx_len[i++] += val;
+
+	val = airoha_fe_rr(eth, REG_FE_GDM_RX_ETH_L1023_CNT_H(port->id));
+	port->stats.rx_len[i] += ((u64)val << 32);
+	val = airoha_fe_rr(eth, REG_FE_GDM_RX_ETH_L1023_CNT_L(port->id));
+	port->stats.rx_len[i++] += val;
+
+	val = airoha_fe_rr(eth, REG_FE_GDM_RX_ETH_LONG_CNT(port->id));
+	port->stats.rx_len[i++] += val;
+
+	/* reset mib counters */
+	airoha_fe_set(eth, REG_FE_GDM_MIB_CLEAR(port->id),
+		      FE_GDM_MIB_RX_CLEAR_MASK | FE_GDM_MIB_TX_CLEAR_MASK);
+
+	u64_stats_update_end(&port->stats.syncp);
+	spin_unlock(&port->stats.lock);
+}
+
+static int airoha_dev_open(struct net_device *dev)
+{
+	int err, len = ETH_HLEN + dev->mtu + ETH_FCS_LEN;
+	struct airoha_gdm_port *port = netdev_priv(dev);
+	struct airoha_qdma *qdma = port->qdma;
+
+	netif_tx_start_all_queues(dev);
+	err = airoha_set_vip_for_gdm_port(port, true);
+	if (err)
+		return err;
+
+	if (netdev_uses_dsa(dev))
+		airoha_fe_set(qdma->eth, REG_GDM_INGRESS_CFG(port->id),
+			      GDM_STAG_EN_MASK);
+	else
+		airoha_fe_clear(qdma->eth, REG_GDM_INGRESS_CFG(port->id),
+				GDM_STAG_EN_MASK);
+
+	airoha_fe_rmw(qdma->eth, REG_GDM_LEN_CFG(port->id),
+		      GDM_SHORT_LEN_MASK | GDM_LONG_LEN_MASK,
+		      FIELD_PREP(GDM_SHORT_LEN_MASK, 60) |
+		      FIELD_PREP(GDM_LONG_LEN_MASK, len));
+
+	airoha_qdma_set(qdma, REG_QDMA_GLOBAL_CFG,
+			GLOBAL_CFG_TX_DMA_EN_MASK |
+			GLOBAL_CFG_RX_DMA_EN_MASK);
+	atomic_inc(&qdma->users);
+
+	return 0;
+}
+
+static int airoha_dev_stop(struct net_device *dev)
+{
+	struct airoha_gdm_port *port = netdev_priv(dev);
+	struct airoha_qdma *qdma = port->qdma;
+	int i, err;
+
+	netif_tx_disable(dev);
+	err = airoha_set_vip_for_gdm_port(port, false);
+	if (err)
+		return err;
+
+	for (i = 0; i < ARRAY_SIZE(qdma->q_tx); i++)
+		netdev_tx_reset_subqueue(dev, i);
+
+	if (atomic_dec_and_test(&qdma->users)) {
+		airoha_qdma_clear(qdma, REG_QDMA_GLOBAL_CFG,
+				  GLOBAL_CFG_TX_DMA_EN_MASK |
+				  GLOBAL_CFG_RX_DMA_EN_MASK);
+
+		for (i = 0; i < ARRAY_SIZE(qdma->q_tx); i++) {
+			if (!qdma->q_tx[i].ndesc)
+				continue;
+
+			airoha_qdma_cleanup_tx_queue(&qdma->q_tx[i]);
+		}
+	}
+
+	return 0;
+}
+
+static int airoha_dev_set_macaddr(struct net_device *dev, void *p)
+{
+	struct airoha_gdm_port *port = netdev_priv(dev);
+	int err;
+
+	err = eth_mac_addr(dev, p);
+	if (err)
+		return err;
+
+	airoha_set_macaddr(port, dev->dev_addr);
+
+	return 0;
+}
+
+static void airhoha_set_gdm2_loopback(struct airoha_gdm_port *port)
+{
+	u32 pse_port = port->id == 3 ? FE_PSE_PORT_GDM3 : FE_PSE_PORT_GDM4;
+	struct airoha_eth *eth = port->qdma->eth;
+	u32 chan = port->id == 3 ? 4 : 0;
+
+	/* Forward the traffic to the proper GDM port */
+	airoha_set_gdm_port_fwd_cfg(eth, REG_GDM_FWD_CFG(2), pse_port);
+	airoha_fe_clear(eth, REG_GDM_FWD_CFG(2), GDM_STRIP_CRC);
+
+	/* Enable GDM2 loopback */
+	airoha_fe_wr(eth, REG_GDM_TXCHN_EN(2), 0xffffffff);
+	airoha_fe_wr(eth, REG_GDM_RXCHN_EN(2), 0xffff);
+	airoha_fe_rmw(eth, REG_GDM_LPBK_CFG(2),
+		      LPBK_CHAN_MASK | LPBK_MODE_MASK | LPBK_EN_MASK,
+		      FIELD_PREP(LPBK_CHAN_MASK, chan) |
+		      LBK_GAP_MODE_MASK | LBK_LEN_MODE_MASK |
+		      LBK_CHAN_MODE_MASK | LPBK_EN_MASK);
+	airoha_fe_rmw(eth, REG_GDM_LEN_CFG(2),
+		      GDM_SHORT_LEN_MASK | GDM_LONG_LEN_MASK,
+		      FIELD_PREP(GDM_SHORT_LEN_MASK, 60) |
+		      FIELD_PREP(GDM_LONG_LEN_MASK, AIROHA_MAX_MTU));
+
+	/* Disable VIP and IFC for GDM2 */
+	airoha_fe_clear(eth, REG_FE_VIP_PORT_EN, BIT(2));
+	airoha_fe_clear(eth, REG_FE_IFC_PORT_EN, BIT(2));
+
+	if (port->id == 3) {
+		/* FIXME: handle XSI_PCE1_PORT */
+		airoha_fe_rmw(eth, REG_FE_WAN_PORT,
+			      WAN1_EN_MASK | WAN1_MASK | WAN0_MASK,
+			      FIELD_PREP(WAN0_MASK, HSGMII_LAN_PCIE0_SRCPORT));
+		airoha_fe_rmw(eth,
+			      REG_SP_DFT_CPORT(HSGMII_LAN_PCIE0_SRCPORT >> 3),
+			      SP_CPORT_PCIE0_MASK,
+			      FIELD_PREP(SP_CPORT_PCIE0_MASK,
+					 FE_PSE_PORT_CDM2));
+	} else {
+		/* FIXME: handle XSI_USB_PORT */
+		airoha_fe_rmw(eth, REG_SRC_PORT_FC_MAP6,
+			      FC_ID_OF_SRC_PORT24_MASK,
+			      FIELD_PREP(FC_ID_OF_SRC_PORT24_MASK, 2));
+		airoha_fe_rmw(eth, REG_FE_WAN_PORT,
+			      WAN1_EN_MASK | WAN1_MASK | WAN0_MASK,
+			      FIELD_PREP(WAN0_MASK, HSGMII_LAN_ETH_SRCPORT));
+		airoha_fe_rmw(eth,
+			      REG_SP_DFT_CPORT(HSGMII_LAN_ETH_SRCPORT >> 3),
+			      SP_CPORT_ETH_MASK,
+			      FIELD_PREP(SP_CPORT_ETH_MASK, FE_PSE_PORT_CDM2));
+	}
+}
+
+static int airoha_dev_init(struct net_device *dev)
+{
+	struct airoha_gdm_port *port = netdev_priv(dev);
+	struct airoha_eth *eth = port->qdma->eth;
+	u32 pse_port;
+
+	airoha_set_macaddr(port, dev->dev_addr);
+
+	switch (port->id) {
+	case 3:
+	case 4:
+		/* If GDM2 is active we can't enable loopback */
+		if (!eth->ports[1])
+			airhoha_set_gdm2_loopback(port);
+		fallthrough;
+	case 2:
+		pse_port = FE_PSE_PORT_PPE2;
+		break;
+	default:
+		pse_port = FE_PSE_PORT_PPE1;
+		break;
+	}
+
+	airoha_set_gdm_port_fwd_cfg(eth, REG_GDM_FWD_CFG(port->id), pse_port);
+
+	return 0;
+}
+
+static void airoha_dev_get_stats64(struct net_device *dev,
+				   struct rtnl_link_stats64 *storage)
+{
+	struct airoha_gdm_port *port = netdev_priv(dev);
+	unsigned int start;
+
+	airoha_update_hw_stats(port);
+	do {
+		start = u64_stats_fetch_begin(&port->stats.syncp);
+		storage->rx_packets = port->stats.rx_ok_pkts;
+		storage->tx_packets = port->stats.tx_ok_pkts;
+		storage->rx_bytes = port->stats.rx_ok_bytes;
+		storage->tx_bytes = port->stats.tx_ok_bytes;
+		storage->multicast = port->stats.rx_multicast;
+		storage->rx_errors = port->stats.rx_errors;
+		storage->rx_dropped = port->stats.rx_drops;
+		storage->tx_dropped = port->stats.tx_drops;
+		storage->rx_crc_errors = port->stats.rx_crc_error;
+		storage->rx_over_errors = port->stats.rx_over_errors;
+	} while (u64_stats_fetch_retry(&port->stats.syncp, start));
+}
+
+static int airoha_dev_change_mtu(struct net_device *dev, int mtu)
+{
+	struct airoha_gdm_port *port = netdev_priv(dev);
+	struct airoha_eth *eth = port->qdma->eth;
+	u32 len = ETH_HLEN + mtu + ETH_FCS_LEN;
+
+	airoha_fe_rmw(eth, REG_GDM_LEN_CFG(port->id),
+		      GDM_LONG_LEN_MASK,
+		      FIELD_PREP(GDM_LONG_LEN_MASK, len));
+	WRITE_ONCE(dev->mtu, mtu);
+
+	return 0;
+}
+
+static u16 airoha_dev_select_queue(struct net_device *dev, struct sk_buff *skb,
+				   struct net_device *sb_dev)
+{
+	struct airoha_gdm_port *port = netdev_priv(dev);
+	int queue, channel;
+
+	/* For dsa device select QoS channel according to the dsa user port
+	 * index, rely on port id otherwise. Select QoS queue based on the
+	 * skb priority.
+	 */
+	channel = netdev_uses_dsa(dev) ? skb_get_queue_mapping(skb) : port->id;
+	channel = channel % AIROHA_NUM_QOS_CHANNELS;
+	queue = (skb->priority - 1) % AIROHA_NUM_QOS_QUEUES; /* QoS queue */
+	queue = channel * AIROHA_NUM_QOS_QUEUES + queue;
+
+	return queue < dev->num_tx_queues ? queue : 0;
+}
+
+static u32 airoha_get_dsa_tag(struct sk_buff *skb, struct net_device *dev)
+{
+#if IS_ENABLED(CONFIG_NET_DSA)
+	struct ethhdr *ehdr;
+	u8 xmit_tpid;
+	u16 tag;
+
+	if (!netdev_uses_dsa(dev))
+		return 0;
+
+	if (dev->dsa_ptr->tag_ops->proto != DSA_TAG_PROTO_MTK)
+		return 0;
+
+	if (skb_cow_head(skb, 0))
+		return 0;
+
+	ehdr = (struct ethhdr *)skb->data;
+	tag = be16_to_cpu(ehdr->h_proto);
+	xmit_tpid = tag >> 8;
+
+	switch (xmit_tpid) {
+	case MTK_HDR_XMIT_TAGGED_TPID_8100:
+		ehdr->h_proto = cpu_to_be16(ETH_P_8021Q);
+		tag &= ~(MTK_HDR_XMIT_TAGGED_TPID_8100 << 8);
+		break;
+	case MTK_HDR_XMIT_TAGGED_TPID_88A8:
+		ehdr->h_proto = cpu_to_be16(ETH_P_8021AD);
+		tag &= ~(MTK_HDR_XMIT_TAGGED_TPID_88A8 << 8);
+		break;
+	default:
+		/* PPE module requires untagged DSA packets to work properly,
+		 * so move DSA tag to DMA descriptor.
+		 */
+		memmove(skb->data + MTK_HDR_LEN, skb->data, 2 * ETH_ALEN);
+		__skb_pull(skb, MTK_HDR_LEN);
+		break;
+	}
+
+	return tag;
+#else
+	return 0;
+#endif
+}
+
+static netdev_tx_t airoha_dev_xmit(struct sk_buff *skb,
+				   struct net_device *dev)
+{
+	struct airoha_gdm_port *port = netdev_priv(dev);
+	struct airoha_qdma *qdma = port->qdma;
+	u32 nr_frags, tag, msg0, msg1, len;
+	struct netdev_queue *txq;
+	struct airoha_queue *q;
+	void *data;
+	int i, qid;
+	u16 index;
+	u8 fport;
+
+	qid = skb_get_queue_mapping(skb) % ARRAY_SIZE(qdma->q_tx);
+	tag = airoha_get_dsa_tag(skb, dev);
+
+	msg0 = FIELD_PREP(QDMA_ETH_TXMSG_CHAN_MASK,
+			  qid / AIROHA_NUM_QOS_QUEUES) |
+	       FIELD_PREP(QDMA_ETH_TXMSG_QUEUE_MASK,
+			  qid % AIROHA_NUM_QOS_QUEUES) |
+	       FIELD_PREP(QDMA_ETH_TXMSG_SP_TAG_MASK, tag);
+	if (skb->ip_summed == CHECKSUM_PARTIAL)
+		msg0 |= FIELD_PREP(QDMA_ETH_TXMSG_TCO_MASK, 1) |
+			FIELD_PREP(QDMA_ETH_TXMSG_UCO_MASK, 1) |
+			FIELD_PREP(QDMA_ETH_TXMSG_ICO_MASK, 1);
+
+	/* TSO: fill MSS info in tcp checksum field */
+	if (skb_is_gso(skb)) {
+		if (skb_cow_head(skb, 0))
+			goto error;
+
+		if (skb_shinfo(skb)->gso_type & (SKB_GSO_TCPV4 |
+						 SKB_GSO_TCPV6)) {
+			__be16 csum = cpu_to_be16(skb_shinfo(skb)->gso_size);
+
+			tcp_hdr(skb)->check = (__force __sum16)csum;
+			msg0 |= FIELD_PREP(QDMA_ETH_TXMSG_TSO_MASK, 1);
+		}
+	}
+
+	fport = port->id == 4 ? FE_PSE_PORT_GDM4 : port->id;
+	msg1 = FIELD_PREP(QDMA_ETH_TXMSG_FPORT_MASK, fport) |
+	       FIELD_PREP(QDMA_ETH_TXMSG_METER_MASK, 0x7f);
+
+	q = &qdma->q_tx[qid];
+	if (WARN_ON_ONCE(!q->ndesc))
+		goto error;
+
+	spin_lock_bh(&q->lock);
+
+	txq = netdev_get_tx_queue(dev, qid);
+	nr_frags = 1 + skb_shinfo(skb)->nr_frags;
+
+	if (q->queued + nr_frags > q->ndesc) {
+		/* not enough space in the queue */
+		netif_tx_stop_queue(txq);
+		spin_unlock_bh(&q->lock);
+		return NETDEV_TX_BUSY;
+	}
+
+	len = skb_headlen(skb);
+	data = skb->data;
+	index = q->head;
+
+	for (i = 0; i < nr_frags; i++) {
+		struct airoha_qdma_desc *desc = &q->desc[index];
+		struct airoha_queue_entry *e = &q->entry[index];
+		skb_frag_t *frag = &skb_shinfo(skb)->frags[i];
+		dma_addr_t addr;
+		u32 val;
+
+		addr = dma_map_single(dev->dev.parent, data, len,
+				      DMA_TO_DEVICE);
+		if (unlikely(dma_mapping_error(dev->dev.parent, addr)))
+			goto error_unmap;
+
+		index = (index + 1) % q->ndesc;
+
+		val = FIELD_PREP(QDMA_DESC_LEN_MASK, len);
+		if (i < nr_frags - 1)
+			val |= FIELD_PREP(QDMA_DESC_MORE_MASK, 1);
+		WRITE_ONCE(desc->ctrl, cpu_to_le32(val));
+		WRITE_ONCE(desc->addr, cpu_to_le32(addr));
+		val = FIELD_PREP(QDMA_DESC_NEXT_ID_MASK, index);
+		WRITE_ONCE(desc->data, cpu_to_le32(val));
+		WRITE_ONCE(desc->msg0, cpu_to_le32(msg0));
+		WRITE_ONCE(desc->msg1, cpu_to_le32(msg1));
+		WRITE_ONCE(desc->msg2, cpu_to_le32(0xffff));
+
+		e->skb = i ? NULL : skb;
+		e->dma_addr = addr;
+		e->dma_len = len;
+
+		data = skb_frag_address(frag);
+		len = skb_frag_size(frag);
+	}
+
+	q->head = index;
+	q->queued += i;
+
+	skb_tx_timestamp(skb);
+	netdev_tx_sent_queue(txq, skb->len);
+
+	if (netif_xmit_stopped(txq) || !netdev_xmit_more())
+		airoha_qdma_rmw(qdma, REG_TX_CPU_IDX(qid),
+				TX_RING_CPU_IDX_MASK,
+				FIELD_PREP(TX_RING_CPU_IDX_MASK, q->head));
+
+	if (q->ndesc - q->queued < q->free_thr)
+		netif_tx_stop_queue(txq);
+
+	spin_unlock_bh(&q->lock);
+
+	return NETDEV_TX_OK;
+
+error_unmap:
+	for (i--; i >= 0; i--) {
+		index = (q->head + i) % q->ndesc;
+		dma_unmap_single(dev->dev.parent, q->entry[index].dma_addr,
+				 q->entry[index].dma_len, DMA_TO_DEVICE);
+	}
+
+	spin_unlock_bh(&q->lock);
+error:
+	dev_kfree_skb_any(skb);
+	dev->stats.tx_dropped++;
+
+	return NETDEV_TX_OK;
+}
+
+static void airoha_ethtool_get_drvinfo(struct net_device *dev,
+				       struct ethtool_drvinfo *info)
+{
+	struct airoha_gdm_port *port = netdev_priv(dev);
+	struct airoha_eth *eth = port->qdma->eth;
+
+	strscpy(info->driver, eth->dev->driver->name, sizeof(info->driver));
+	strscpy(info->bus_info, dev_name(eth->dev), sizeof(info->bus_info));
+}
+
+static void airoha_ethtool_get_mac_stats(struct net_device *dev,
+					 struct ethtool_eth_mac_stats *stats)
+{
+	struct airoha_gdm_port *port = netdev_priv(dev);
+	unsigned int start;
+
+	airoha_update_hw_stats(port);
+	do {
+		start = u64_stats_fetch_begin(&port->stats.syncp);
+		stats->MulticastFramesXmittedOK = port->stats.tx_multicast;
+		stats->BroadcastFramesXmittedOK = port->stats.tx_broadcast;
+		stats->BroadcastFramesReceivedOK = port->stats.rx_broadcast;
+	} while (u64_stats_fetch_retry(&port->stats.syncp, start));
+}
+
+static const struct ethtool_rmon_hist_range airoha_ethtool_rmon_ranges[] = {
+	{    0,    64 },
+	{   65,   127 },
+	{  128,   255 },
+	{  256,   511 },
+	{  512,  1023 },
+	{ 1024,  1518 },
+	{ 1519, 10239 },
+	{},
+};
+
+static void
+airoha_ethtool_get_rmon_stats(struct net_device *dev,
+			      struct ethtool_rmon_stats *stats,
+			      const struct ethtool_rmon_hist_range **ranges)
+{
+	struct airoha_gdm_port *port = netdev_priv(dev);
+	struct airoha_hw_stats *hw_stats = &port->stats;
+	unsigned int start;
+
+	BUILD_BUG_ON(ARRAY_SIZE(airoha_ethtool_rmon_ranges) !=
+		     ARRAY_SIZE(hw_stats->tx_len) + 1);
+	BUILD_BUG_ON(ARRAY_SIZE(airoha_ethtool_rmon_ranges) !=
+		     ARRAY_SIZE(hw_stats->rx_len) + 1);
+
+	*ranges = airoha_ethtool_rmon_ranges;
+	airoha_update_hw_stats(port);
+	do {
+		int i;
+
+		start = u64_stats_fetch_begin(&port->stats.syncp);
+		stats->fragments = hw_stats->rx_fragment;
+		stats->jabbers = hw_stats->rx_jabber;
+		for (i = 0; i < ARRAY_SIZE(airoha_ethtool_rmon_ranges) - 1;
+		     i++) {
+			stats->hist[i] = hw_stats->rx_len[i];
+			stats->hist_tx[i] = hw_stats->tx_len[i];
+		}
+	} while (u64_stats_fetch_retry(&port->stats.syncp, start));
+}
+
+static int airoha_qdma_set_chan_tx_sched(struct airoha_gdm_port *port,
+					 int channel, enum tx_sched_mode mode,
+					 const u16 *weights, u8 n_weights)
+{
+	int i;
+
+	for (i = 0; i < AIROHA_NUM_TX_RING; i++)
+		airoha_qdma_clear(port->qdma, REG_QUEUE_CLOSE_CFG(channel),
+				  TXQ_DISABLE_CHAN_QUEUE_MASK(channel, i));
+
+	for (i = 0; i < n_weights; i++) {
+		u32 status;
+		int err;
+
+		airoha_qdma_wr(port->qdma, REG_TXWRR_WEIGHT_CFG,
+			       TWRR_RW_CMD_MASK |
+			       FIELD_PREP(TWRR_CHAN_IDX_MASK, channel) |
+			       FIELD_PREP(TWRR_QUEUE_IDX_MASK, i) |
+			       FIELD_PREP(TWRR_VALUE_MASK, weights[i]));
+		err = read_poll_timeout(airoha_qdma_rr, status,
+					status & TWRR_RW_CMD_DONE,
+					USEC_PER_MSEC, 10 * USEC_PER_MSEC,
+					true, port->qdma,
+					REG_TXWRR_WEIGHT_CFG);
+		if (err)
+			return err;
+	}
+
+	airoha_qdma_rmw(port->qdma, REG_CHAN_QOS_MODE(channel >> 3),
+			CHAN_QOS_MODE_MASK(channel),
+			mode << __ffs(CHAN_QOS_MODE_MASK(channel)));
+
+	return 0;
+}
+
+static int airoha_qdma_set_tx_prio_sched(struct airoha_gdm_port *port,
+					 int channel)
+{
+	static const u16 w[AIROHA_NUM_QOS_QUEUES] = {};
+
+	return airoha_qdma_set_chan_tx_sched(port, channel, TC_SCH_SP, w,
+					     ARRAY_SIZE(w));
+}
+
+static int airoha_qdma_set_tx_ets_sched(struct airoha_gdm_port *port,
+					int channel,
+					struct tc_ets_qopt_offload *opt)
+{
+	struct tc_ets_qopt_offload_replace_params *p = &opt->replace_params;
+	enum tx_sched_mode mode = TC_SCH_SP;
+	u16 w[AIROHA_NUM_QOS_QUEUES] = {};
+	int i, nstrict = 0;
+
+	if (p->bands > AIROHA_NUM_QOS_QUEUES)
+		return -EINVAL;
+
+	for (i = 0; i < p->bands; i++) {
+		if (!p->quanta[i])
+			nstrict++;
+	}
+
+	/* this configuration is not supported by the hw */
+	if (nstrict == AIROHA_NUM_QOS_QUEUES - 1)
+		return -EINVAL;
+
+	/* EN7581 SoC supports fixed QoS band priority where WRR queues have
+	 * lowest priorities with respect to SP ones.
+	 * e.g: WRR0, WRR1, .., WRRm, SP0, SP1, .., SPn
+	 */
+	for (i = 0; i < nstrict; i++) {
+		if (p->priomap[p->bands - i - 1] != i)
+			return -EINVAL;
+	}
+
+	for (i = 0; i < p->bands - nstrict; i++) {
+		if (p->priomap[i] != nstrict + i)
+			return -EINVAL;
+
+		w[i] = p->weights[nstrict + i];
+	}
+
+	if (!nstrict)
+		mode = TC_SCH_WRR8;
+	else if (nstrict < AIROHA_NUM_QOS_QUEUES - 1)
+		mode = nstrict + 1;
+
+	return airoha_qdma_set_chan_tx_sched(port, channel, mode, w,
+					     ARRAY_SIZE(w));
+}
+
+static int airoha_qdma_get_tx_ets_stats(struct airoha_gdm_port *port,
+					int channel,
+					struct tc_ets_qopt_offload *opt)
+{
+	u64 cpu_tx_packets = airoha_qdma_rr(port->qdma,
+					    REG_CNTR_VAL(channel << 1));
+	u64 fwd_tx_packets = airoha_qdma_rr(port->qdma,
+					    REG_CNTR_VAL((channel << 1) + 1));
+	u64 tx_packets = (cpu_tx_packets - port->cpu_tx_packets) +
+			 (fwd_tx_packets - port->fwd_tx_packets);
+	_bstats_update(opt->stats.bstats, 0, tx_packets);
+
+	port->cpu_tx_packets = cpu_tx_packets;
+	port->fwd_tx_packets = fwd_tx_packets;
+
+	return 0;
+}
+
+static int airoha_tc_setup_qdisc_ets(struct airoha_gdm_port *port,
+				     struct tc_ets_qopt_offload *opt)
+{
+	int channel;
+
+	if (opt->parent == TC_H_ROOT)
+		return -EINVAL;
+
+	channel = TC_H_MAJ(opt->handle) >> 16;
+	channel = channel % AIROHA_NUM_QOS_CHANNELS;
+
+	switch (opt->command) {
+	case TC_ETS_REPLACE:
+		return airoha_qdma_set_tx_ets_sched(port, channel, opt);
+	case TC_ETS_DESTROY:
+		/* PRIO is default qdisc scheduler */
+		return airoha_qdma_set_tx_prio_sched(port, channel);
+	case TC_ETS_STATS:
+		return airoha_qdma_get_tx_ets_stats(port, channel, opt);
+	default:
+		return -EOPNOTSUPP;
+	}
+}
+
+static int airoha_qdma_get_rl_param(struct airoha_qdma *qdma, int queue_id,
+				    u32 addr, enum trtcm_param_type param,
+				    u32 *val_low, u32 *val_high)
+{
+	u32 idx = QDMA_METER_IDX(queue_id), group = QDMA_METER_GROUP(queue_id);
+	u32 val, config = FIELD_PREP(RATE_LIMIT_PARAM_TYPE_MASK, param) |
+			  FIELD_PREP(RATE_LIMIT_METER_GROUP_MASK, group) |
+			  FIELD_PREP(RATE_LIMIT_PARAM_INDEX_MASK, idx);
+
+	airoha_qdma_wr(qdma, REG_TRTCM_CFG_PARAM(addr), config);
+	if (read_poll_timeout(airoha_qdma_rr, val,
+			      val & RATE_LIMIT_PARAM_RW_DONE_MASK,
+			      USEC_PER_MSEC, 10 * USEC_PER_MSEC, true, qdma,
+			      REG_TRTCM_CFG_PARAM(addr)))
+		return -ETIMEDOUT;
+
+	*val_low = airoha_qdma_rr(qdma, REG_TRTCM_DATA_LOW(addr));
+	if (val_high)
+		*val_high = airoha_qdma_rr(qdma, REG_TRTCM_DATA_HIGH(addr));
+
+	return 0;
+}
+
+static int airoha_qdma_set_rl_param(struct airoha_qdma *qdma, int queue_id,
+				    u32 addr, enum trtcm_param_type param,
+				    u32 val)
+{
+	u32 idx = QDMA_METER_IDX(queue_id), group = QDMA_METER_GROUP(queue_id);
+	u32 config = RATE_LIMIT_PARAM_RW_MASK |
+		     FIELD_PREP(RATE_LIMIT_PARAM_TYPE_MASK, param) |
+		     FIELD_PREP(RATE_LIMIT_METER_GROUP_MASK, group) |
+		     FIELD_PREP(RATE_LIMIT_PARAM_INDEX_MASK, idx);
+
+	airoha_qdma_wr(qdma, REG_TRTCM_DATA_LOW(addr), val);
+	airoha_qdma_wr(qdma, REG_TRTCM_CFG_PARAM(addr), config);
+
+	return read_poll_timeout(airoha_qdma_rr, val,
+				 val & RATE_LIMIT_PARAM_RW_DONE_MASK,
+				 USEC_PER_MSEC, 10 * USEC_PER_MSEC, true,
+				 qdma, REG_TRTCM_CFG_PARAM(addr));
+}
+
+static int airoha_qdma_set_rl_config(struct airoha_qdma *qdma, int queue_id,
+				     u32 addr, bool enable, u32 enable_mask)
+{
+	u32 val;
+	int err;
+
+	err = airoha_qdma_get_rl_param(qdma, queue_id, addr, TRTCM_MISC_MODE,
+				       &val, NULL);
+	if (err)
+		return err;
+
+	val = enable ? val | enable_mask : val & ~enable_mask;
+
+	return airoha_qdma_set_rl_param(qdma, queue_id, addr, TRTCM_MISC_MODE,
+					val);
+}
+
+static int airoha_qdma_set_rl_token_bucket(struct airoha_qdma *qdma,
+					   int queue_id, u32 rate_val,
+					   u32 bucket_size)
+{
+	u32 val, config, tick, unit, rate, rate_frac;
+	int err;
+
+	err = airoha_qdma_get_rl_param(qdma, queue_id, REG_INGRESS_TRTCM_CFG,
+				       TRTCM_MISC_MODE, &config, NULL);
+	if (err)
+		return err;
+
+	val = airoha_qdma_rr(qdma, REG_INGRESS_TRTCM_CFG);
+	tick = FIELD_GET(INGRESS_FAST_TICK_MASK, val);
+	if (config & TRTCM_TICK_SEL)
+		tick *= FIELD_GET(INGRESS_SLOW_TICK_RATIO_MASK, val);
+	if (!tick)
+		return -EINVAL;
+
+	unit = (config & TRTCM_PKT_MODE) ? 1000000 / tick : 8000 / tick;
+	if (!unit)
+		return -EINVAL;
+
+	rate = rate_val / unit;
+	rate_frac = rate_val % unit;
+	rate_frac = FIELD_PREP(TRTCM_TOKEN_RATE_MASK, rate_frac) / unit;
+	rate = FIELD_PREP(TRTCM_TOKEN_RATE_MASK, rate) |
+	       FIELD_PREP(TRTCM_TOKEN_RATE_FRACTION_MASK, rate_frac);
+
+	err = airoha_qdma_set_rl_param(qdma, queue_id, REG_INGRESS_TRTCM_CFG,
+				       TRTCM_TOKEN_RATE_MODE, rate);
+	if (err)
+		return err;
+
+	val = bucket_size;
+	if (!(config & TRTCM_PKT_MODE))
+		val = max_t(u32, val, MIN_TOKEN_SIZE);
+	val = min_t(u32, __fls(val), MAX_TOKEN_SIZE_OFFSET);
+
+	return airoha_qdma_set_rl_param(qdma, queue_id, REG_INGRESS_TRTCM_CFG,
+					TRTCM_BUCKETSIZE_SHIFT_MODE, val);
+}
+
+static int airoha_qdma_init_rl_config(struct airoha_qdma *qdma, int queue_id,
+				      bool enable, enum trtcm_unit_type unit)
+{
+	bool tick_sel = queue_id == 0 || queue_id == 2 || queue_id == 8;
+	enum trtcm_param mode = TRTCM_METER_MODE;
+	int err;
+
+	mode |= unit == TRTCM_PACKET_UNIT ? TRTCM_PKT_MODE : 0;
+	err = airoha_qdma_set_rl_config(qdma, queue_id, REG_INGRESS_TRTCM_CFG,
+					enable, mode);
+	if (err)
+		return err;
+
+	return airoha_qdma_set_rl_config(qdma, queue_id, REG_INGRESS_TRTCM_CFG,
+					 tick_sel, TRTCM_TICK_SEL);
+}
+
+static int airoha_qdma_get_trtcm_param(struct airoha_qdma *qdma, int channel,
+				       u32 addr, enum trtcm_param_type param,
+				       enum trtcm_mode_type mode,
+				       u32 *val_low, u32 *val_high)
+{
+	u32 idx = QDMA_METER_IDX(channel), group = QDMA_METER_GROUP(channel);
+	u32 val, config = FIELD_PREP(TRTCM_PARAM_TYPE_MASK, param) |
+			  FIELD_PREP(TRTCM_METER_GROUP_MASK, group) |
+			  FIELD_PREP(TRTCM_PARAM_INDEX_MASK, idx) |
+			  FIELD_PREP(TRTCM_PARAM_RATE_TYPE_MASK, mode);
+
+	airoha_qdma_wr(qdma, REG_TRTCM_CFG_PARAM(addr), config);
+	if (read_poll_timeout(airoha_qdma_rr, val,
+			      val & TRTCM_PARAM_RW_DONE_MASK,
+			      USEC_PER_MSEC, 10 * USEC_PER_MSEC, true,
+			      qdma, REG_TRTCM_CFG_PARAM(addr)))
+		return -ETIMEDOUT;
+
+	*val_low = airoha_qdma_rr(qdma, REG_TRTCM_DATA_LOW(addr));
+	if (val_high)
+		*val_high = airoha_qdma_rr(qdma, REG_TRTCM_DATA_HIGH(addr));
+
+	return 0;
+}
+
+static int airoha_qdma_set_trtcm_param(struct airoha_qdma *qdma, int channel,
+				       u32 addr, enum trtcm_param_type param,
+				       enum trtcm_mode_type mode, u32 val)
+{
+	u32 idx = QDMA_METER_IDX(channel), group = QDMA_METER_GROUP(channel);
+	u32 config = TRTCM_PARAM_RW_MASK |
+		     FIELD_PREP(TRTCM_PARAM_TYPE_MASK, param) |
+		     FIELD_PREP(TRTCM_METER_GROUP_MASK, group) |
+		     FIELD_PREP(TRTCM_PARAM_INDEX_MASK, idx) |
+		     FIELD_PREP(TRTCM_PARAM_RATE_TYPE_MASK, mode);
+
+	airoha_qdma_wr(qdma, REG_TRTCM_DATA_LOW(addr), val);
+	airoha_qdma_wr(qdma, REG_TRTCM_CFG_PARAM(addr), config);
+
+	return read_poll_timeout(airoha_qdma_rr, val,
+				 val & TRTCM_PARAM_RW_DONE_MASK,
+				 USEC_PER_MSEC, 10 * USEC_PER_MSEC, true,
+				 qdma, REG_TRTCM_CFG_PARAM(addr));
+}
+
+static int airoha_qdma_set_trtcm_config(struct airoha_qdma *qdma, int channel,
+					u32 addr, enum trtcm_mode_type mode,
+					bool enable, u32 enable_mask)
+{
+	u32 val;
+
+	if (airoha_qdma_get_trtcm_param(qdma, channel, addr, TRTCM_MISC_MODE,
+					mode, &val, NULL))
+		return -EINVAL;
+
+	val = enable ? val | enable_mask : val & ~enable_mask;
+
+	return airoha_qdma_set_trtcm_param(qdma, channel, addr, TRTCM_MISC_MODE,
+					   mode, val);
+}
+
+static int airoha_qdma_set_trtcm_token_bucket(struct airoha_qdma *qdma,
+					      int channel, u32 addr,
+					      enum trtcm_mode_type mode,
+					      u32 rate_val, u32 bucket_size)
+{
+	u32 val, config, tick, unit, rate, rate_frac;
+	int err;
+
+	if (airoha_qdma_get_trtcm_param(qdma, channel, addr, TRTCM_MISC_MODE,
+					mode, &config, NULL))
+		return -EINVAL;
+
+	val = airoha_qdma_rr(qdma, addr);
+	tick = FIELD_GET(INGRESS_FAST_TICK_MASK, val);
+	if (config & TRTCM_TICK_SEL)
+		tick *= FIELD_GET(INGRESS_SLOW_TICK_RATIO_MASK, val);
+	if (!tick)
+		return -EINVAL;
+
+	unit = (config & TRTCM_PKT_MODE) ? 1000000 / tick : 8000 / tick;
+	if (!unit)
+		return -EINVAL;
+
+	rate = rate_val / unit;
+	rate_frac = rate_val % unit;
+	rate_frac = FIELD_PREP(TRTCM_TOKEN_RATE_MASK, rate_frac) / unit;
+	rate = FIELD_PREP(TRTCM_TOKEN_RATE_MASK, rate) |
+	       FIELD_PREP(TRTCM_TOKEN_RATE_FRACTION_MASK, rate_frac);
+
+	err = airoha_qdma_set_trtcm_param(qdma, channel, addr,
+					  TRTCM_TOKEN_RATE_MODE, mode, rate);
+	if (err)
+		return err;
+
+	val = max_t(u32, bucket_size, MIN_TOKEN_SIZE);
+	val = min_t(u32, __fls(val), MAX_TOKEN_SIZE_OFFSET);
+
+	return airoha_qdma_set_trtcm_param(qdma, channel, addr,
+					   TRTCM_BUCKETSIZE_SHIFT_MODE,
+					   mode, val);
+}
+
+static int airoha_qdma_set_tx_rate_limit(struct airoha_gdm_port *port,
+					 int channel, u32 rate,
+					 u32 bucket_size)
+{
+	int i, err;
+
+	for (i = 0; i <= TRTCM_PEAK_MODE; i++) {
+		err = airoha_qdma_set_trtcm_config(port->qdma, channel,
+						   REG_EGRESS_TRTCM_CFG, i,
+						   !!rate, TRTCM_METER_MODE);
+		if (err)
+			return err;
+
+		err = airoha_qdma_set_trtcm_token_bucket(port->qdma, channel,
+							 REG_EGRESS_TRTCM_CFG,
+							 i, rate, bucket_size);
+		if (err)
+			return err;
+	}
+
+	return 0;
+}
+
+static int airoha_tc_htb_alloc_leaf_queue(struct airoha_gdm_port *port,
+					  struct tc_htb_qopt_offload *opt)
+{
+	u32 channel = TC_H_MIN(opt->classid) % AIROHA_NUM_QOS_CHANNELS;
+	u32 rate = div_u64(opt->rate, 1000) << 3; /* kbps */
+	struct net_device *dev = port->dev;
+	int num_tx_queues = dev->real_num_tx_queues;
+	int err;
+
+	if (opt->parent_classid != TC_HTB_CLASSID_ROOT) {
+		NL_SET_ERR_MSG_MOD(opt->extack, "invalid parent classid");
+		return -EINVAL;
+	}
+
+	err = airoha_qdma_set_tx_rate_limit(port, channel, rate, opt->quantum);
+	if (err) {
+		NL_SET_ERR_MSG_MOD(opt->extack,
+				   "failed configuring htb offload");
+		return err;
+	}
+
+	if (opt->command == TC_HTB_NODE_MODIFY)
+		return 0;
+
+	err = netif_set_real_num_tx_queues(dev, num_tx_queues + 1);
+	if (err) {
+		airoha_qdma_set_tx_rate_limit(port, channel, 0, opt->quantum);
+		NL_SET_ERR_MSG_MOD(opt->extack,
+				   "failed setting real_num_tx_queues");
+		return err;
+	}
+
+	set_bit(channel, port->qos_sq_bmap);
+	opt->qid = AIROHA_NUM_TX_RING + channel;
+
+	return 0;
+}
+
+static int airoha_qdma_set_rx_meter(struct airoha_gdm_port *port,
+				    u32 rate, u32 bucket_size,
+				    enum trtcm_unit_type unit_type)
+{
+	struct airoha_qdma *qdma = port->qdma;
+	int i;
+
+	for (i = 0; i < ARRAY_SIZE(qdma->q_rx); i++) {
+		int err;
+
+		if (!qdma->q_rx[i].ndesc)
+			continue;
+
+		err = airoha_qdma_init_rl_config(qdma, i, !!rate, unit_type);
+		if (err)
+			return err;
+
+		err = airoha_qdma_set_rl_token_bucket(qdma, i, rate,
+						      bucket_size);
+		if (err)
+			return err;
+	}
+
+	return 0;
+}
+
+static int airoha_tc_matchall_act_validate(struct tc_cls_matchall_offload *f)
+{
+	const struct flow_action *actions = &f->rule->action;
+	const struct flow_action_entry *act;
+
+	if (!flow_action_has_entries(actions)) {
+		NL_SET_ERR_MSG_MOD(f->common.extack,
+				   "filter run with no actions");
+		return -EINVAL;
+	}
+
+	if (!flow_offload_has_one_action(actions)) {
+		NL_SET_ERR_MSG_MOD(f->common.extack,
+				   "only once action per filter is supported");
+		return -EOPNOTSUPP;
+	}
+
+	act = &actions->entries[0];
+	if (act->id != FLOW_ACTION_POLICE) {
+		NL_SET_ERR_MSG_MOD(f->common.extack, "unsupported action");
+		return -EOPNOTSUPP;
+	}
+
+	if (act->police.exceed.act_id != FLOW_ACTION_DROP) {
+		NL_SET_ERR_MSG_MOD(f->common.extack,
+				   "invalid exceed action id");
+		return -EOPNOTSUPP;
+	}
+
+	if (act->police.notexceed.act_id != FLOW_ACTION_ACCEPT) {
+		NL_SET_ERR_MSG_MOD(f->common.extack,
+				   "invalid notexceed action id");
+		return -EOPNOTSUPP;
+	}
+
+	if (act->police.notexceed.act_id == FLOW_ACTION_ACCEPT &&
+	    !flow_action_is_last_entry(actions, act)) {
+		NL_SET_ERR_MSG_MOD(f->common.extack,
+				   "action accept must be last");
+		return -EOPNOTSUPP;
+	}
+
+	if (act->police.peakrate_bytes_ps || act->police.avrate ||
+	    act->police.overhead || act->police.mtu) {
+		NL_SET_ERR_MSG_MOD(f->common.extack,
+				   "peakrate/avrate/overhead/mtu unsupported");
+		return -EOPNOTSUPP;
+	}
+
+	return 0;
+}
+
+static int airoha_dev_tc_matchall(struct net_device *dev,
+				  struct tc_cls_matchall_offload *f)
+{
+	enum trtcm_unit_type unit_type = TRTCM_BYTE_UNIT;
+	struct airoha_gdm_port *port = netdev_priv(dev);
+	u32 rate = 0, bucket_size = 0;
+
+	switch (f->command) {
+	case TC_CLSMATCHALL_REPLACE: {
+		const struct flow_action_entry *act;
+		int err;
+
+		err = airoha_tc_matchall_act_validate(f);
+		if (err)
+			return err;
+
+		act = &f->rule->action.entries[0];
+		if (act->police.rate_pkt_ps) {
+			rate = act->police.rate_pkt_ps;
+			bucket_size = act->police.burst_pkt;
+			unit_type = TRTCM_PACKET_UNIT;
+		} else {
+			rate = div_u64(act->police.rate_bytes_ps, 1000);
+			rate = rate << 3; /* Kbps */
+			bucket_size = act->police.burst;
+		}
+		fallthrough;
+	}
+	case TC_CLSMATCHALL_DESTROY:
+		return airoha_qdma_set_rx_meter(port, rate, bucket_size,
+						unit_type);
+	default:
+		return -EOPNOTSUPP;
+	}
+}
+
+static int airoha_dev_setup_tc_block_cb(enum tc_setup_type type,
+					void *type_data, void *cb_priv)
+{
+	struct net_device *dev = cb_priv;
+	struct airoha_gdm_port *port = netdev_priv(dev);
+	struct airoha_eth *eth = port->qdma->eth;
+
+	if (!tc_can_offload(dev))
+		return -EOPNOTSUPP;
+
+	switch (type) {
+	case TC_SETUP_CLSFLOWER:
+		return airoha_ppe_setup_tc_block_cb(&eth->ppe->dev, type_data);
+	case TC_SETUP_CLSMATCHALL:
+		return airoha_dev_tc_matchall(dev, type_data);
+	default:
+		return -EOPNOTSUPP;
+	}
+}
+
+static int airoha_dev_setup_tc_block(struct airoha_gdm_port *port,
+				     struct flow_block_offload *f)
+{
+	flow_setup_cb_t *cb = airoha_dev_setup_tc_block_cb;
+	static LIST_HEAD(block_cb_list);
+	struct flow_block_cb *block_cb;
+
+	if (f->binder_type != FLOW_BLOCK_BINDER_TYPE_CLSACT_INGRESS)
+		return -EOPNOTSUPP;
+
+	f->driver_block_list = &block_cb_list;
+	switch (f->command) {
+	case FLOW_BLOCK_BIND:
+		block_cb = flow_block_cb_lookup(f->block, cb, port->dev);
+		if (block_cb) {
+			flow_block_cb_incref(block_cb);
+			return 0;
+		}
+		block_cb = flow_block_cb_alloc(cb, port->dev, port->dev, NULL);
+		if (IS_ERR(block_cb))
+			return PTR_ERR(block_cb);
+
+		flow_block_cb_incref(block_cb);
+		flow_block_cb_add(block_cb, f);
+		list_add_tail(&block_cb->driver_list, &block_cb_list);
+		return 0;
+	case FLOW_BLOCK_UNBIND:
+		block_cb = flow_block_cb_lookup(f->block, cb, port->dev);
+		if (!block_cb)
+			return -ENOENT;
+
+		if (!flow_block_cb_decref(block_cb)) {
+			flow_block_cb_remove(block_cb, f);
+			list_del(&block_cb->driver_list);
+		}
+		return 0;
+	default:
+		return -EOPNOTSUPP;
+	}
+}
+
+static void airoha_tc_remove_htb_queue(struct airoha_gdm_port *port, int queue)
+{
+	struct net_device *dev = port->dev;
+
+	netif_set_real_num_tx_queues(dev, dev->real_num_tx_queues - 1);
+	airoha_qdma_set_tx_rate_limit(port, queue + 1, 0, 0);
+	clear_bit(queue, port->qos_sq_bmap);
+}
+
+static int airoha_tc_htb_delete_leaf_queue(struct airoha_gdm_port *port,
+					   struct tc_htb_qopt_offload *opt)
+{
+	u32 channel = TC_H_MIN(opt->classid) % AIROHA_NUM_QOS_CHANNELS;
+
+	if (!test_bit(channel, port->qos_sq_bmap)) {
+		NL_SET_ERR_MSG_MOD(opt->extack, "invalid queue id");
+		return -EINVAL;
+	}
+
+	airoha_tc_remove_htb_queue(port, channel);
+
+	return 0;
+}
+
+static int airoha_tc_htb_destroy(struct airoha_gdm_port *port)
+{
+	int q;
+
+	for_each_set_bit(q, port->qos_sq_bmap, AIROHA_NUM_QOS_CHANNELS)
+		airoha_tc_remove_htb_queue(port, q);
+
+	return 0;
+}
+
+static int airoha_tc_get_htb_get_leaf_queue(struct airoha_gdm_port *port,
+					    struct tc_htb_qopt_offload *opt)
+{
+	u32 channel = TC_H_MIN(opt->classid) % AIROHA_NUM_QOS_CHANNELS;
+
+	if (!test_bit(channel, port->qos_sq_bmap)) {
+		NL_SET_ERR_MSG_MOD(opt->extack, "invalid queue id");
+		return -EINVAL;
+	}
+
+	opt->qid = AIROHA_NUM_TX_RING + channel;
+
+	return 0;
+}
+
+static int airoha_tc_setup_qdisc_htb(struct airoha_gdm_port *port,
+				     struct tc_htb_qopt_offload *opt)
+{
+	switch (opt->command) {
+	case TC_HTB_CREATE:
+		break;
+	case TC_HTB_DESTROY:
+		return airoha_tc_htb_destroy(port);
+	case TC_HTB_NODE_MODIFY:
+	case TC_HTB_LEAF_ALLOC_QUEUE:
+		return airoha_tc_htb_alloc_leaf_queue(port, opt);
+	case TC_HTB_LEAF_DEL:
+	case TC_HTB_LEAF_DEL_LAST:
+	case TC_HTB_LEAF_DEL_LAST_FORCE:
+		return airoha_tc_htb_delete_leaf_queue(port, opt);
+	case TC_HTB_LEAF_QUERY_QUEUE:
+		return airoha_tc_get_htb_get_leaf_queue(port, opt);
+	default:
+		return -EOPNOTSUPP;
+	}
+
+	return 0;
+}
+
+static int airoha_dev_tc_setup(struct net_device *dev, enum tc_setup_type type,
+			       void *type_data)
+{
+	struct airoha_gdm_port *port = netdev_priv(dev);
+
+	switch (type) {
+	case TC_SETUP_QDISC_ETS:
+		return airoha_tc_setup_qdisc_ets(port, type_data);
+	case TC_SETUP_QDISC_HTB:
+		return airoha_tc_setup_qdisc_htb(port, type_data);
+	case TC_SETUP_BLOCK:
+	case TC_SETUP_FT:
+		return airoha_dev_setup_tc_block(port, type_data);
+	default:
+		return -EOPNOTSUPP;
+	}
+}
+
+static const struct net_device_ops airoha_netdev_ops = {
+	.ndo_init		= airoha_dev_init,
+	.ndo_open		= airoha_dev_open,
+	.ndo_stop		= airoha_dev_stop,
+	.ndo_change_mtu		= airoha_dev_change_mtu,
+	.ndo_select_queue	= airoha_dev_select_queue,
+	.ndo_start_xmit		= airoha_dev_xmit,
+	.ndo_get_stats64        = airoha_dev_get_stats64,
+	.ndo_set_mac_address	= airoha_dev_set_macaddr,
+	.ndo_setup_tc		= airoha_dev_tc_setup,
+};
+
+static const struct ethtool_ops airoha_ethtool_ops = {
+	.get_drvinfo		= airoha_ethtool_get_drvinfo,
+	.get_eth_mac_stats      = airoha_ethtool_get_mac_stats,
+	.get_rmon_stats		= airoha_ethtool_get_rmon_stats,
+};
+
+static int airoha_metadata_dst_alloc(struct airoha_gdm_port *port)
+{
+	int i;
+
+	for (i = 0; i < ARRAY_SIZE(port->dsa_meta); i++) {
+		struct metadata_dst *md_dst;
+
+		md_dst = metadata_dst_alloc(0, METADATA_HW_PORT_MUX,
+					    GFP_KERNEL);
+		if (!md_dst)
+			return -ENOMEM;
+
+		md_dst->u.port_info.port_id = i;
+		port->dsa_meta[i] = md_dst;
+	}
+
+	return 0;
+}
+
+static void airoha_metadata_dst_free(struct airoha_gdm_port *port)
+{
+	int i;
+
+	for (i = 0; i < ARRAY_SIZE(port->dsa_meta); i++) {
+		if (!port->dsa_meta[i])
+			continue;
+
+		metadata_dst_free(port->dsa_meta[i]);
+	}
+}
+
+bool airoha_is_valid_gdm_port(struct airoha_eth *eth,
+			      struct airoha_gdm_port *port)
+{
+	int i;
+
+	for (i = 0; i < ARRAY_SIZE(eth->ports); i++) {
+		if (eth->ports[i] == port)
+			return true;
+	}
+
+	return false;
+}
+
+static int airoha_alloc_gdm_port(struct airoha_eth *eth,
+				 struct device_node *np, int index)
+{
+	const __be32 *id_ptr = of_get_property(np, "reg", NULL);
+	struct airoha_gdm_port *port;
+	struct airoha_qdma *qdma;
+	struct net_device *dev;
+	int err, p;
+	u32 id;
+
+	if (!id_ptr) {
+		dev_err(eth->dev, "missing gdm port id\n");
+		return -EINVAL;
+	}
+
+	id = be32_to_cpup(id_ptr);
+	p = id - 1;
+
+	if (!id || id > ARRAY_SIZE(eth->ports)) {
+		dev_err(eth->dev, "invalid gdm port id: %d\n", id);
+		return -EINVAL;
+	}
+
+	if (eth->ports[p]) {
+		dev_err(eth->dev, "duplicate gdm port id: %d\n", id);
+		return -EINVAL;
+	}
+
+	dev = devm_alloc_etherdev_mqs(eth->dev, sizeof(*port),
+				      AIROHA_NUM_NETDEV_TX_RINGS,
+				      AIROHA_NUM_RX_RING);
+	if (!dev) {
+		dev_err(eth->dev, "alloc_etherdev failed\n");
+		return -ENOMEM;
+	}
+
+	qdma = &eth->qdma[index % AIROHA_MAX_NUM_QDMA];
+	dev->netdev_ops = &airoha_netdev_ops;
+	dev->ethtool_ops = &airoha_ethtool_ops;
+	dev->max_mtu = AIROHA_MAX_MTU;
+	dev->watchdog_timeo = 5 * HZ;
+	dev->hw_features = NETIF_F_IP_CSUM | NETIF_F_RXCSUM |
+			   NETIF_F_TSO6 | NETIF_F_IPV6_CSUM |
+			   NETIF_F_SG | NETIF_F_TSO |
+			   NETIF_F_HW_TC;
+	dev->features |= dev->hw_features;
+	dev->vlan_features = dev->hw_features;
+	dev->dev.of_node = np;
+	dev->irq = qdma->irq_banks[0].irq;
+	SET_NETDEV_DEV(dev, eth->dev);
+
+	/* reserve hw queues for HTB offloading */
+	err = netif_set_real_num_tx_queues(dev, AIROHA_NUM_TX_RING);
+	if (err)
+		return err;
+
+	err = of_get_ethdev_address(np, dev);
+	if (err) {
+		if (err == -EPROBE_DEFER)
+			return err;
+
+		eth_hw_addr_random(dev);
+		dev_info(eth->dev, "generated random MAC address %pM\n",
+			 dev->dev_addr);
+	}
+
+	port = netdev_priv(dev);
+	u64_stats_init(&port->stats.syncp);
+	spin_lock_init(&port->stats.lock);
+	port->qdma = qdma;
+	port->dev = dev;
+	port->id = id;
+	eth->ports[p] = port;
+
+	err = airoha_metadata_dst_alloc(port);
+	if (err)
+		return err;
+
+	err = register_netdev(dev);
+	if (err)
+		goto free_metadata_dst;
+
+	return 0;
+
+free_metadata_dst:
+	airoha_metadata_dst_free(port);
+	return err;
+}
+
+static int airoha_probe(struct platform_device *pdev)
+{
+	struct device_node *np;
+	struct airoha_eth *eth;
+	int i, err;
+
+	eth = devm_kzalloc(&pdev->dev, sizeof(*eth), GFP_KERNEL);
+	if (!eth)
+		return -ENOMEM;
+
+	eth->dev = &pdev->dev;
+
+	err = dma_set_mask_and_coherent(eth->dev, DMA_BIT_MASK(32));
+	if (err) {
+		dev_err(eth->dev, "failed configuring DMA mask\n");
+		return err;
+	}
+
+	eth->fe_regs = devm_platform_ioremap_resource_byname(pdev, "fe");
+	if (IS_ERR(eth->fe_regs))
+		return dev_err_probe(eth->dev, PTR_ERR(eth->fe_regs),
+				     "failed to iomap fe regs\n");
+
+	eth->rsts[0].id = "fe";
+	eth->rsts[1].id = "pdma";
+	eth->rsts[2].id = "qdma";
+	err = devm_reset_control_bulk_get_exclusive(eth->dev,
+						    ARRAY_SIZE(eth->rsts),
+						    eth->rsts);
+	if (err) {
+		dev_err(eth->dev, "failed to get bulk reset lines\n");
+		return err;
+	}
+
+	eth->xsi_rsts[0].id = "xsi-mac";
+	eth->xsi_rsts[1].id = "hsi0-mac";
+	eth->xsi_rsts[2].id = "hsi1-mac";
+	eth->xsi_rsts[3].id = "hsi-mac";
+	eth->xsi_rsts[4].id = "xfp-mac";
+	err = devm_reset_control_bulk_get_exclusive(eth->dev,
+						    ARRAY_SIZE(eth->xsi_rsts),
+						    eth->xsi_rsts);
+	if (err) {
+		dev_err(eth->dev, "failed to get bulk xsi reset lines\n");
+		return err;
+	}
+
+	eth->napi_dev = alloc_netdev_dummy(0);
+	if (!eth->napi_dev)
+		return -ENOMEM;
+
+	/* Enable threaded NAPI by default */
+	eth->napi_dev->threaded = true;
+	strscpy(eth->napi_dev->name, "qdma_eth", sizeof(eth->napi_dev->name));
+	platform_set_drvdata(pdev, eth);
+
+	err = airoha_hw_init(pdev, eth);
+	if (err)
+		goto error_hw_cleanup;
+
+	for (i = 0; i < ARRAY_SIZE(eth->qdma); i++)
+		airoha_qdma_start_napi(&eth->qdma[i]);
+
+	i = 0;
+	for_each_child_of_node(pdev->dev.of_node, np) {
+		if (!of_device_is_compatible(np, "airoha,eth-mac"))
+			continue;
+
+		if (!of_device_is_available(np))
+			continue;
+
+		err = airoha_alloc_gdm_port(eth, np, i++);
+		if (err) {
+			of_node_put(np);
+			goto error_napi_stop;
+		}
+	}
+
+	return 0;
+
+error_napi_stop:
+	for (i = 0; i < ARRAY_SIZE(eth->qdma); i++)
+		airoha_qdma_stop_napi(&eth->qdma[i]);
+	airoha_ppe_deinit(eth);
+error_hw_cleanup:
+	for (i = 0; i < ARRAY_SIZE(eth->qdma); i++)
+		airoha_hw_cleanup(&eth->qdma[i]);
+
+	for (i = 0; i < ARRAY_SIZE(eth->ports); i++) {
+		struct airoha_gdm_port *port = eth->ports[i];
+
+		if (port && port->dev->reg_state == NETREG_REGISTERED) {
+			unregister_netdev(port->dev);
+			airoha_metadata_dst_free(port);
+		}
+	}
+	free_netdev(eth->napi_dev);
+	platform_set_drvdata(pdev, NULL);
+
+	return err;
+}
+
+static void airoha_remove(struct platform_device *pdev)
+{
+	struct airoha_eth *eth = platform_get_drvdata(pdev);
+	int i;
+
+	for (i = 0; i < ARRAY_SIZE(eth->qdma); i++) {
+		airoha_qdma_stop_napi(&eth->qdma[i]);
+		airoha_hw_cleanup(&eth->qdma[i]);
+	}
+
+	for (i = 0; i < ARRAY_SIZE(eth->ports); i++) {
+		struct airoha_gdm_port *port = eth->ports[i];
+
+		if (!port)
+			continue;
+
+		airoha_dev_stop(port->dev);
+		unregister_netdev(port->dev);
+		airoha_metadata_dst_free(port);
+	}
+	free_netdev(eth->napi_dev);
+
+	airoha_ppe_deinit(eth);
+	platform_set_drvdata(pdev, NULL);
+}
+
+static const struct of_device_id of_airoha_match[] = {
+	{ .compatible = "airoha,en7581-eth" },
+	{ /* sentinel */ }
+};
+MODULE_DEVICE_TABLE(of, of_airoha_match);
+
+static struct platform_driver airoha_driver = {
+	.probe = airoha_probe,
+	.remove = airoha_remove,
+	.driver = {
+		.name = KBUILD_MODNAME,
+		.of_match_table = of_airoha_match,
+	},
+};
+module_platform_driver(airoha_driver);
+
+MODULE_LICENSE("GPL");
+MODULE_AUTHOR("Lorenzo Bianconi <lorenzo@kernel.org>");
+MODULE_DESCRIPTION("Ethernet driver for Airoha SoC");
diff --git a/drivers/net/ethernet/airoha/airoha_eth.h b/drivers/net/ethernet/airoha/airoha_eth.h
new file mode 100644
index 000000000000..cd13c1c1224f
--- /dev/null
+++ b/drivers/net/ethernet/airoha/airoha_eth.h
@@ -0,0 +1,643 @@
+/* SPDX-License-Identifier: GPL-2.0-only */
+/*
+ * Copyright (c) 2024 AIROHA Inc
+ * Author: Lorenzo Bianconi <lorenzo@kernel.org>
+ */
+
+#ifndef AIROHA_ETH_H
+#define AIROHA_ETH_H
+
+#include <linux/debugfs.h>
+#include <linux/etherdevice.h>
+#include <linux/iopoll.h>
+#include <linux/kernel.h>
+#include <linux/netdevice.h>
+#include <linux/reset.h>
+#include <linux/soc/airoha/airoha_offload.h>
+#include <net/dsa.h>
+
+#define AIROHA_MAX_NUM_GDM_PORTS	4
+#define AIROHA_MAX_NUM_QDMA		2
+#define AIROHA_MAX_NUM_IRQ_BANKS	4
+#define AIROHA_MAX_DSA_PORTS		7
+#define AIROHA_MAX_NUM_RSTS		3
+#define AIROHA_MAX_NUM_XSI_RSTS		5
+#define AIROHA_MAX_MTU			9216
+#define AIROHA_MAX_PACKET_SIZE		2048
+#define AIROHA_NUM_QOS_CHANNELS		4
+#define AIROHA_NUM_QOS_QUEUES		8
+#define AIROHA_NUM_TX_RING		32
+#define AIROHA_NUM_RX_RING		32
+#define AIROHA_NUM_NETDEV_TX_RINGS	(AIROHA_NUM_TX_RING + \
+					 AIROHA_NUM_QOS_CHANNELS)
+#define AIROHA_FE_MC_MAX_VLAN_TABLE	64
+#define AIROHA_FE_MC_MAX_VLAN_PORT	16
+#define AIROHA_NUM_TX_IRQ		2
+#define HW_DSCP_NUM			2048
+#define IRQ_QUEUE_LEN(_n)		((_n) ? 1024 : 2048)
+#define TX_DSCP_NUM			1024
+#define RX_DSCP_NUM(_n)			\
+	((_n) ==  2 ? 128 :		\
+	 (_n) == 11 ? 128 :		\
+	 (_n) == 15 ? 128 :		\
+	 (_n) ==  0 ? 1024 : 16)
+
+#define PSE_RSV_PAGES			128
+#define PSE_QUEUE_RSV_PAGES		64
+
+#define QDMA_METER_IDX(_n)		((_n) & 0xff)
+#define QDMA_METER_GROUP(_n)		(((_n) >> 8) & 0x3)
+
+#define PPE_NUM				2
+#define PPE1_SRAM_NUM_ENTRIES		(8 * 1024)
+#define PPE_SRAM_NUM_ENTRIES		(2 * PPE1_SRAM_NUM_ENTRIES)
+#ifdef CONFIG_NET_AIROHA_FLOW_STATS
+#define PPE1_STATS_NUM_ENTRIES		(4 * 1024)
+#else
+#define PPE1_STATS_NUM_ENTRIES		0
+#endif /* CONFIG_NET_AIROHA_FLOW_STATS */
+#define PPE_STATS_NUM_ENTRIES		(2 * PPE1_STATS_NUM_ENTRIES)
+#define PPE1_SRAM_NUM_DATA_ENTRIES	(PPE1_SRAM_NUM_ENTRIES - PPE1_STATS_NUM_ENTRIES)
+#define PPE_SRAM_NUM_DATA_ENTRIES	(2 * PPE1_SRAM_NUM_DATA_ENTRIES)
+#define PPE_DRAM_NUM_ENTRIES		(16 * 1024)
+#define PPE_NUM_ENTRIES			(PPE_SRAM_NUM_ENTRIES + PPE_DRAM_NUM_ENTRIES)
+#define PPE_HASH_MASK			(PPE_NUM_ENTRIES - 1)
+#define PPE_ENTRY_SIZE			80
+#define PPE_RAM_NUM_ENTRIES_SHIFT(_n)	(__ffs((_n) >> 10))
+
+#define MTK_HDR_LEN			4
+#define MTK_HDR_XMIT_TAGGED_TPID_8100	1
+#define MTK_HDR_XMIT_TAGGED_TPID_88A8	2
+
+enum {
+	QDMA_INT_REG_IDX0,
+	QDMA_INT_REG_IDX1,
+	QDMA_INT_REG_IDX2,
+	QDMA_INT_REG_IDX3,
+	QDMA_INT_REG_IDX4,
+	QDMA_INT_REG_MAX
+};
+
+enum {
+	HSGMII_LAN_PCIE0_SRCPORT = 0x16,
+	HSGMII_LAN_PCIE1_SRCPORT,
+	HSGMII_LAN_ETH_SRCPORT,
+	HSGMII_LAN_USB_SRCPORT,
+};
+
+enum {
+	XSI_PCIE0_VIP_PORT_MASK	= BIT(22),
+	XSI_PCIE1_VIP_PORT_MASK	= BIT(23),
+	XSI_USB_VIP_PORT_MASK	= BIT(25),
+	XSI_ETH_VIP_PORT_MASK	= BIT(24),
+};
+
+enum {
+	DEV_STATE_INITIALIZED,
+};
+
+enum {
+	CDM_CRSN_QSEL_Q1 = 1,
+	CDM_CRSN_QSEL_Q5 = 5,
+	CDM_CRSN_QSEL_Q6 = 6,
+	CDM_CRSN_QSEL_Q15 = 15,
+};
+
+enum {
+	CRSN_08 = 0x8,
+	CRSN_21 = 0x15, /* KA */
+	CRSN_22 = 0x16, /* hit bind and force route to CPU */
+	CRSN_24 = 0x18,
+	CRSN_25 = 0x19,
+};
+
+enum {
+	FE_PSE_PORT_CDM1,
+	FE_PSE_PORT_GDM1,
+	FE_PSE_PORT_GDM2,
+	FE_PSE_PORT_GDM3,
+	FE_PSE_PORT_PPE1,
+	FE_PSE_PORT_CDM2,
+	FE_PSE_PORT_CDM3,
+	FE_PSE_PORT_CDM4,
+	FE_PSE_PORT_PPE2,
+	FE_PSE_PORT_GDM4,
+	FE_PSE_PORT_CDM5,
+	FE_PSE_PORT_DROP = 0xf,
+};
+
+enum tx_sched_mode {
+	TC_SCH_WRR8,
+	TC_SCH_SP,
+	TC_SCH_WRR7,
+	TC_SCH_WRR6,
+	TC_SCH_WRR5,
+	TC_SCH_WRR4,
+	TC_SCH_WRR3,
+	TC_SCH_WRR2,
+};
+
+enum trtcm_unit_type {
+	TRTCM_BYTE_UNIT,
+	TRTCM_PACKET_UNIT,
+};
+
+enum trtcm_param_type {
+	TRTCM_MISC_MODE, /* meter_en, pps_mode, tick_sel */
+	TRTCM_TOKEN_RATE_MODE,
+	TRTCM_BUCKETSIZE_SHIFT_MODE,
+	TRTCM_BUCKET_COUNTER_MODE,
+};
+
+enum trtcm_mode_type {
+	TRTCM_COMMIT_MODE,
+	TRTCM_PEAK_MODE,
+};
+
+enum trtcm_param {
+	TRTCM_TICK_SEL = BIT(0),
+	TRTCM_PKT_MODE = BIT(1),
+	TRTCM_METER_MODE = BIT(2),
+};
+
+#define MIN_TOKEN_SIZE				4096
+#define MAX_TOKEN_SIZE_OFFSET			17
+#define TRTCM_TOKEN_RATE_MASK			GENMASK(23, 6)
+#define TRTCM_TOKEN_RATE_FRACTION_MASK		GENMASK(5, 0)
+
+struct airoha_queue_entry {
+	union {
+		void *buf;
+		struct sk_buff *skb;
+	};
+	dma_addr_t dma_addr;
+	u16 dma_len;
+};
+
+struct airoha_queue {
+	struct airoha_qdma *qdma;
+
+	/* protect concurrent queue accesses */
+	spinlock_t lock;
+	struct airoha_queue_entry *entry;
+	struct airoha_qdma_desc *desc;
+	u16 head;
+	u16 tail;
+
+	int queued;
+	int ndesc;
+	int free_thr;
+	int buf_size;
+
+	struct napi_struct napi;
+	struct page_pool *page_pool;
+	struct sk_buff *skb;
+};
+
+struct airoha_tx_irq_queue {
+	struct airoha_qdma *qdma;
+
+	struct napi_struct napi;
+
+	int size;
+	u32 *q;
+};
+
+struct airoha_hw_stats {
+	/* protect concurrent hw_stats accesses */
+	spinlock_t lock;
+	struct u64_stats_sync syncp;
+
+	/* get_stats64 */
+	u64 rx_ok_pkts;
+	u64 tx_ok_pkts;
+	u64 rx_ok_bytes;
+	u64 tx_ok_bytes;
+	u64 rx_multicast;
+	u64 rx_errors;
+	u64 rx_drops;
+	u64 tx_drops;
+	u64 rx_crc_error;
+	u64 rx_over_errors;
+	/* ethtool stats */
+	u64 tx_broadcast;
+	u64 tx_multicast;
+	u64 tx_len[7];
+	u64 rx_broadcast;
+	u64 rx_fragment;
+	u64 rx_jabber;
+	u64 rx_len[7];
+};
+
+enum {
+	AIROHA_FOE_STATE_INVALID,
+	AIROHA_FOE_STATE_UNBIND,
+	AIROHA_FOE_STATE_BIND,
+	AIROHA_FOE_STATE_FIN
+};
+
+enum {
+	PPE_PKT_TYPE_IPV4_HNAPT = 0,
+	PPE_PKT_TYPE_IPV4_ROUTE = 1,
+	PPE_PKT_TYPE_BRIDGE = 2,
+	PPE_PKT_TYPE_IPV4_DSLITE = 3,
+	PPE_PKT_TYPE_IPV6_ROUTE_3T = 4,
+	PPE_PKT_TYPE_IPV6_ROUTE_5T = 5,
+	PPE_PKT_TYPE_IPV6_6RD = 7,
+};
+
+#define AIROHA_FOE_MAC_SMAC_ID		GENMASK(20, 16)
+#define AIROHA_FOE_MAC_PPPOE_ID		GENMASK(15, 0)
+
+#define AIROHA_FOE_MAC_WDMA_QOS		GENMASK(15, 12)
+#define AIROHA_FOE_MAC_WDMA_BAND	BIT(11)
+#define AIROHA_FOE_MAC_WDMA_WCID	GENMASK(10, 0)
+
+struct airoha_foe_mac_info_common {
+	u16 vlan1;
+	u16 etype;
+
+	u32 dest_mac_hi;
+
+	u16 vlan2;
+	u16 dest_mac_lo;
+
+	u32 src_mac_hi;
+};
+
+struct airoha_foe_mac_info {
+	struct airoha_foe_mac_info_common common;
+
+	u16 pppoe_id;
+	u16 src_mac_lo;
+
+	u32 meter;
+};
+
+#define AIROHA_FOE_IB1_UNBIND_PREBIND		BIT(24)
+#define AIROHA_FOE_IB1_UNBIND_PACKETS		GENMASK(23, 8)
+#define AIROHA_FOE_IB1_UNBIND_TIMESTAMP		GENMASK(7, 0)
+
+#define AIROHA_FOE_IB1_BIND_STATIC		BIT(31)
+#define AIROHA_FOE_IB1_BIND_UDP			BIT(30)
+#define AIROHA_FOE_IB1_BIND_STATE		GENMASK(29, 28)
+#define AIROHA_FOE_IB1_BIND_PACKET_TYPE		GENMASK(27, 25)
+#define AIROHA_FOE_IB1_BIND_TTL			BIT(24)
+#define AIROHA_FOE_IB1_BIND_TUNNEL_DECAP	BIT(23)
+#define AIROHA_FOE_IB1_BIND_PPPOE		BIT(22)
+#define AIROHA_FOE_IB1_BIND_VPM			GENMASK(21, 20)
+#define AIROHA_FOE_IB1_BIND_VLAN_LAYER		GENMASK(19, 16)
+#define AIROHA_FOE_IB1_BIND_KEEPALIVE		BIT(15)
+#define AIROHA_FOE_IB1_BIND_TIMESTAMP		GENMASK(14, 0)
+
+#define AIROHA_FOE_IB2_DSCP			GENMASK(31, 24)
+#define AIROHA_FOE_IB2_PORT_AG			GENMASK(23, 13)
+#define AIROHA_FOE_IB2_PCP			BIT(12)
+#define AIROHA_FOE_IB2_MULTICAST		BIT(11)
+#define AIROHA_FOE_IB2_FAST_PATH		BIT(10)
+#define AIROHA_FOE_IB2_PSE_QOS			BIT(9)
+#define AIROHA_FOE_IB2_PSE_PORT			GENMASK(8, 5)
+#define AIROHA_FOE_IB2_NBQ			GENMASK(4, 0)
+
+#define AIROHA_FOE_ACTDP			GENMASK(31, 24)
+#define AIROHA_FOE_SHAPER_ID			GENMASK(23, 16)
+#define AIROHA_FOE_CHANNEL			GENMASK(15, 11)
+#define AIROHA_FOE_QID				GENMASK(10, 8)
+#define AIROHA_FOE_DPI				BIT(7)
+#define AIROHA_FOE_TUNNEL			BIT(6)
+#define AIROHA_FOE_TUNNEL_ID			GENMASK(5, 0)
+
+#define AIROHA_FOE_TUNNEL_MTU			GENMASK(31, 16)
+#define AIROHA_FOE_ACNT_GRP3			GENMASK(15, 9)
+#define AIROHA_FOE_METER_GRP3			GENMASK(8, 5)
+#define AIROHA_FOE_METER_GRP2			GENMASK(4, 0)
+
+struct airoha_foe_bridge {
+	u32 dest_mac_hi;
+
+	u16 src_mac_hi;
+	u16 dest_mac_lo;
+
+	u32 src_mac_lo;
+
+	u32 ib2;
+
+	u32 rsv[5];
+
+	u32 data;
+
+	struct airoha_foe_mac_info l2;
+};
+
+struct airoha_foe_ipv4_tuple {
+	u32 src_ip;
+	u32 dest_ip;
+	union {
+		struct {
+			u16 dest_port;
+			u16 src_port;
+		};
+		struct {
+			u8 protocol;
+			u8 _pad[3]; /* fill with 0xa5a5a5 */
+		};
+		u32 ports;
+	};
+};
+
+struct airoha_foe_ipv4 {
+	struct airoha_foe_ipv4_tuple orig_tuple;
+
+	u32 ib2;
+
+	struct airoha_foe_ipv4_tuple new_tuple;
+
+	u32 rsv[2];
+
+	u32 data;
+
+	struct airoha_foe_mac_info l2;
+};
+
+struct airoha_foe_ipv4_dslite {
+	struct airoha_foe_ipv4_tuple ip4;
+
+	u32 ib2;
+
+	u8 flow_label[3];
+	u8 priority;
+
+	u32 rsv[4];
+
+	u32 data;
+
+	struct airoha_foe_mac_info l2;
+};
+
+struct airoha_foe_ipv6 {
+	u32 src_ip[4];
+	u32 dest_ip[4];
+
+	union {
+		struct {
+			u16 dest_port;
+			u16 src_port;
+		};
+		struct {
+			u8 protocol;
+			u8 pad[3];
+		};
+		u32 ports;
+	};
+
+	u32 data;
+
+	u32 ib2;
+
+	struct airoha_foe_mac_info_common l2;
+
+	u32 meter;
+};
+
+struct airoha_foe_entry {
+	union {
+		struct {
+			u32 ib1;
+			union {
+				struct airoha_foe_bridge bridge;
+				struct airoha_foe_ipv4 ipv4;
+				struct airoha_foe_ipv4_dslite dslite;
+				struct airoha_foe_ipv6 ipv6;
+				DECLARE_FLEX_ARRAY(u32, d);
+			};
+		};
+		u8 data[PPE_ENTRY_SIZE];
+	};
+};
+
+struct airoha_foe_stats {
+	u32 bytes;
+	u32 packets;
+};
+
+struct airoha_foe_stats64 {
+	u64 bytes;
+	u64 packets;
+};
+
+struct airoha_flow_data {
+	struct ethhdr eth;
+
+	union {
+		struct {
+			__be32 src_addr;
+			__be32 dst_addr;
+		} v4;
+
+		struct {
+			struct in6_addr src_addr;
+			struct in6_addr dst_addr;
+		} v6;
+	};
+
+	__be16 src_port;
+	__be16 dst_port;
+
+	struct {
+		struct {
+			u16 id;
+			__be16 proto;
+		} hdr[2];
+		u8 num;
+	} vlan;
+	struct {
+		u16 sid;
+		u8 num;
+	} pppoe;
+};
+
+enum airoha_flow_entry_type {
+	FLOW_TYPE_L4,
+	FLOW_TYPE_L2,
+	FLOW_TYPE_L2_SUBFLOW,
+};
+
+struct airoha_flow_table_entry {
+	union {
+		struct hlist_node list; /* PPE L3 flow entry */
+		struct {
+			struct rhash_head l2_node;  /* L2 flow entry */
+			struct hlist_head l2_flows; /* PPE L2 subflows list */
+		};
+	};
+
+	struct hlist_node l2_subflow_node; /* PPE L2 subflow entry */
+	u32 hash;
+
+	struct airoha_foe_stats64 stats;
+	enum airoha_flow_entry_type type;
+
+	struct rhash_head node;
+	unsigned long cookie;
+
+	/* Must be last --ends in a flexible-array member. */
+	struct airoha_foe_entry data;
+};
+
+struct airoha_wdma_info {
+	u8 idx;
+	u8 queue;
+	u16 wcid;
+	u8 bss;
+};
+
+/* RX queue to IRQ mapping: BIT(q) in IRQ(n) */
+#define RX_IRQ0_BANK_PIN_MASK			0x839f
+#define RX_IRQ1_BANK_PIN_MASK			0x7fe00000
+#define RX_IRQ2_BANK_PIN_MASK			0x20
+#define RX_IRQ3_BANK_PIN_MASK			0x40
+#define RX_IRQ_BANK_PIN_MASK(_n)		\
+	(((_n) == 3) ? RX_IRQ3_BANK_PIN_MASK :	\
+	 ((_n) == 2) ? RX_IRQ2_BANK_PIN_MASK :	\
+	 ((_n) == 1) ? RX_IRQ1_BANK_PIN_MASK :	\
+	 RX_IRQ0_BANK_PIN_MASK)
+
+struct airoha_irq_bank {
+	struct airoha_qdma *qdma;
+
+	/* protect concurrent irqmask accesses */
+	spinlock_t irq_lock;
+	u32 irqmask[QDMA_INT_REG_MAX];
+	int irq;
+};
+
+struct airoha_qdma {
+	struct airoha_eth *eth;
+	void __iomem *regs;
+
+	atomic_t users;
+
+	struct airoha_irq_bank irq_banks[AIROHA_MAX_NUM_IRQ_BANKS];
+
+	struct airoha_tx_irq_queue q_tx_irq[AIROHA_NUM_TX_IRQ];
+
+	struct airoha_queue q_tx[AIROHA_NUM_TX_RING];
+	struct airoha_queue q_rx[AIROHA_NUM_RX_RING];
+};
+
+struct airoha_gdm_port {
+	struct airoha_qdma *qdma;
+	struct net_device *dev;
+	int id;
+
+	struct airoha_hw_stats stats;
+
+	DECLARE_BITMAP(qos_sq_bmap, AIROHA_NUM_QOS_CHANNELS);
+
+	/* qos stats counters */
+	u64 cpu_tx_packets;
+	u64 fwd_tx_packets;
+
+	struct metadata_dst *dsa_meta[AIROHA_MAX_DSA_PORTS];
+};
+
+#define AIROHA_RXD4_PPE_CPU_REASON	GENMASK(20, 16)
+#define AIROHA_RXD4_FOE_ENTRY		GENMASK(15, 0)
+
+struct airoha_ppe {
+	struct airoha_ppe_dev dev;
+	struct airoha_eth *eth;
+
+	void *foe;
+	dma_addr_t foe_dma;
+
+	struct rhashtable l2_flows;
+
+	struct hlist_head *foe_flow;
+	u16 foe_check_time[PPE_NUM_ENTRIES];
+
+	struct airoha_foe_stats *foe_stats;
+	dma_addr_t foe_stats_dma;
+
+	struct dentry *debugfs_dir;
+};
+
+struct airoha_eth {
+	struct device *dev;
+
+	unsigned long state;
+	void __iomem *fe_regs;
+
+	struct airoha_npu __rcu *npu;
+
+	struct airoha_ppe *ppe;
+	struct rhashtable flow_table;
+
+	struct reset_control_bulk_data rsts[AIROHA_MAX_NUM_RSTS];
+	struct reset_control_bulk_data xsi_rsts[AIROHA_MAX_NUM_XSI_RSTS];
+
+	struct net_device *napi_dev;
+
+	struct airoha_qdma qdma[AIROHA_MAX_NUM_QDMA];
+	struct airoha_gdm_port *ports[AIROHA_MAX_NUM_GDM_PORTS];
+};
+
+u32 airoha_rr(void __iomem *base, u32 offset);
+void airoha_wr(void __iomem *base, u32 offset, u32 val);
+u32 airoha_rmw(void __iomem *base, u32 offset, u32 mask, u32 val);
+
+#define airoha_fe_rr(eth, offset)				\
+	airoha_rr((eth)->fe_regs, (offset))
+#define airoha_fe_wr(eth, offset, val)				\
+	airoha_wr((eth)->fe_regs, (offset), (val))
+#define airoha_fe_rmw(eth, offset, mask, val)			\
+	airoha_rmw((eth)->fe_regs, (offset), (mask), (val))
+#define airoha_fe_set(eth, offset, val)				\
+	airoha_rmw((eth)->fe_regs, (offset), 0, (val))
+#define airoha_fe_clear(eth, offset, val)			\
+	airoha_rmw((eth)->fe_regs, (offset), (val), 0)
+
+#define airoha_qdma_rr(qdma, offset)				\
+	airoha_rr((qdma)->regs, (offset))
+#define airoha_qdma_wr(qdma, offset, val)			\
+	airoha_wr((qdma)->regs, (offset), (val))
+#define airoha_qdma_rmw(qdma, offset, mask, val)		\
+	airoha_rmw((qdma)->regs, (offset), (mask), (val))
+#define airoha_qdma_set(qdma, offset, val)			\
+	airoha_rmw((qdma)->regs, (offset), 0, (val))
+#define airoha_qdma_clear(qdma, offset, val)			\
+	airoha_rmw((qdma)->regs, (offset), (val), 0)
+
+static inline bool airhoa_is_lan_gdm_port(struct airoha_gdm_port *port)
+{
+	/* GDM1 port on EN7581 SoC is connected to the lan dsa switch.
+	 * GDM{2,3,4} can be used as wan port connected to an external
+	 * phy module.
+	 */
+	return port->id == 1;
+}
+
+bool airoha_is_valid_gdm_port(struct airoha_eth *eth,
+			      struct airoha_gdm_port *port);
+
+void airoha_ppe_check_skb(struct airoha_ppe_dev *dev, struct sk_buff *skb,
+			  u16 hash, bool rx_wlan);
+int airoha_ppe_setup_tc_block_cb(struct airoha_ppe_dev *dev, void *type_data);
+int airoha_ppe_init(struct airoha_eth *eth);
+void airoha_ppe_deinit(struct airoha_eth *eth);
+void airoha_ppe_init_upd_mem(struct airoha_gdm_port *port);
+struct airoha_foe_entry *airoha_ppe_foe_get_entry(struct airoha_ppe *ppe,
+						  u32 hash);
+void airoha_ppe_foe_entry_get_stats(struct airoha_ppe *ppe, u32 hash,
+				    struct airoha_foe_stats64 *stats);
+
+#ifdef CONFIG_DEBUG_FS
+int airoha_ppe_debugfs_init(struct airoha_ppe *ppe);
+#else
+static inline int airoha_ppe_debugfs_init(struct airoha_ppe *ppe)
+{
+	return 0;
+}
+#endif
+
+#endif /* AIROHA_ETH_H */
diff --git a/drivers/net/ethernet/airoha/airoha_npu.c b/drivers/net/ethernet/airoha/airoha_npu.c
new file mode 100644
index 000000000000..8c883f2b2d36
--- /dev/null
+++ b/drivers/net/ethernet/airoha/airoha_npu.c
@@ -0,0 +1,742 @@
+// SPDX-License-Identifier: GPL-2.0-only
+/*
+ * Copyright (c) 2025 AIROHA Inc
+ * Author: Lorenzo Bianconi <lorenzo@kernel.org>
+ */
+
+#include <linux/devcoredump.h>
+#include <linux/firmware.h>
+#include <linux/platform_device.h>
+#include <linux/of_net.h>
+#include <linux/of_platform.h>
+#include <linux/of_reserved_mem.h>
+#include <linux/regmap.h>
+
+#include "airoha_eth.h"
+
+#define NPU_EN7581_FIRMWARE_DATA		"airoha/en7581_npu_data.bin"
+#define NPU_EN7581_FIRMWARE_RV32		"airoha/en7581_npu_rv32.bin"
+#define NPU_EN7581_FIRMWARE_RV32_MAX_SIZE	0x200000
+#define NPU_EN7581_FIRMWARE_DATA_MAX_SIZE	0x10000
+#define NPU_DUMP_SIZE				512
+
+#define REG_NPU_LOCAL_SRAM		0x0
+
+#define NPU_PC_BASE_ADDR		0x305000
+#define REG_PC_DBG(_n)			(0x305000 + ((_n) * 0x100))
+
+#define NPU_CLUSTER_BASE_ADDR		0x306000
+
+#define REG_CR_BOOT_TRIGGER		(NPU_CLUSTER_BASE_ADDR + 0x000)
+#define REG_CR_BOOT_CONFIG		(NPU_CLUSTER_BASE_ADDR + 0x004)
+#define REG_CR_BOOT_BASE(_n)		(NPU_CLUSTER_BASE_ADDR + 0x020 + ((_n) << 2))
+
+#define NPU_MBOX_BASE_ADDR		0x30c000
+
+#define REG_CR_MBOX_INT_STATUS		(NPU_MBOX_BASE_ADDR + 0x000)
+#define MBOX_INT_STATUS_MASK		BIT(8)
+
+#define REG_CR_MBOX_INT_MASK(_n)	(NPU_MBOX_BASE_ADDR + 0x004 + ((_n) << 2))
+#define REG_CR_MBQ0_CTRL(_n)		(NPU_MBOX_BASE_ADDR + 0x030 + ((_n) << 2))
+#define REG_CR_MBQ8_CTRL(_n)		(NPU_MBOX_BASE_ADDR + 0x0b0 + ((_n) << 2))
+#define REG_CR_NPU_MIB(_n)		(NPU_MBOX_BASE_ADDR + 0x140 + ((_n) << 2))
+
+#define NPU_WLAN_BASE_ADDR		0x30d000
+
+#define REG_IRQ_STATUS			(NPU_WLAN_BASE_ADDR + 0x030)
+#define REG_IRQ_RXDONE(_n)		(NPU_WLAN_BASE_ADDR + ((_n) << 2) + 0x034)
+#define NPU_IRQ_RX_MASK(_n)		((_n) == 1 ? BIT(17) : BIT(16))
+
+#define REG_TX_BASE(_n)			(NPU_WLAN_BASE_ADDR + ((_n) << 4) + 0x080)
+#define REG_TX_DSCP_NUM(_n)		(NPU_WLAN_BASE_ADDR + ((_n) << 4) + 0x084)
+#define REG_TX_CPU_IDX(_n)		(NPU_WLAN_BASE_ADDR + ((_n) << 4) + 0x088)
+#define REG_TX_DMA_IDX(_n)		(NPU_WLAN_BASE_ADDR + ((_n) << 4) + 0x08c)
+
+#define REG_RX_BASE(_n)			(NPU_WLAN_BASE_ADDR + ((_n) << 4) + 0x180)
+#define REG_RX_DSCP_NUM(_n)		(NPU_WLAN_BASE_ADDR + ((_n) << 4) + 0x184)
+#define REG_RX_CPU_IDX(_n)		(NPU_WLAN_BASE_ADDR + ((_n) << 4) + 0x188)
+#define REG_RX_DMA_IDX(_n)		(NPU_WLAN_BASE_ADDR + ((_n) << 4) + 0x18c)
+
+#define NPU_TIMER_BASE_ADDR		0x310100
+#define REG_WDT_TIMER_CTRL(_n)		(NPU_TIMER_BASE_ADDR + ((_n) * 0x100))
+#define WDT_EN_MASK			BIT(25)
+#define WDT_INTR_MASK			BIT(21)
+
+enum {
+	NPU_OP_SET = 1,
+	NPU_OP_SET_NO_WAIT,
+	NPU_OP_GET,
+	NPU_OP_GET_NO_WAIT,
+};
+
+enum {
+	NPU_FUNC_WIFI,
+	NPU_FUNC_TUNNEL,
+	NPU_FUNC_NOTIFY,
+	NPU_FUNC_DBA,
+	NPU_FUNC_TR471,
+	NPU_FUNC_PPE,
+};
+
+enum {
+	NPU_MBOX_ERROR,
+	NPU_MBOX_SUCCESS,
+};
+
+enum {
+	PPE_FUNC_SET_WAIT,
+	PPE_FUNC_SET_WAIT_HWNAT_INIT,
+	PPE_FUNC_SET_WAIT_HWNAT_DEINIT,
+	PPE_FUNC_SET_WAIT_API,
+	PPE_FUNC_SET_WAIT_FLOW_STATS_SETUP,
+};
+
+enum {
+	PPE2_SRAM_SET_ENTRY,
+	PPE_SRAM_SET_ENTRY,
+	PPE_SRAM_SET_VAL,
+	PPE_SRAM_RESET_VAL,
+};
+
+enum {
+	QDMA_WAN_ETHER = 1,
+	QDMA_WAN_PON_XDSL,
+};
+
+#define MBOX_MSG_FUNC_ID	GENMASK(14, 11)
+#define MBOX_MSG_STATIC_BUF	BIT(5)
+#define MBOX_MSG_STATUS		GENMASK(4, 2)
+#define MBOX_MSG_DONE		BIT(1)
+#define MBOX_MSG_WAIT_RSP	BIT(0)
+
+#define PPE_TYPE_L2B_IPV4	2
+#define PPE_TYPE_L2B_IPV4_IPV6	3
+
+struct ppe_mbox_data {
+	u32 func_type;
+	u32 func_id;
+	union {
+		struct {
+			u8 cds;
+			u8 xpon_hal_api;
+			u8 wan_xsi;
+			u8 ct_joyme4;
+			u8 max_packet;
+			u8 rsv[3];
+			u32 ppe_type;
+			u32 wan_mode;
+			u32 wan_sel;
+		} init_info;
+		struct {
+			u32 func_id;
+			u32 size;
+			u32 data;
+		} set_info;
+		struct {
+			u32 npu_stats_addr;
+			u32 foe_stats_addr;
+		} stats_info;
+	};
+};
+
+struct wlan_mbox_data {
+	u32 ifindex:4;
+	u32 func_type:4;
+	u32 func_id;
+	DECLARE_FLEX_ARRAY(u8, d);
+};
+
+static int airoha_npu_send_msg(struct airoha_npu *npu, int func_id,
+			       void *p, int size)
+{
+	u16 core = 0; /* FIXME */
+	u32 val, offset = core << 4;
+	dma_addr_t dma_addr;
+	int ret;
+
+	dma_addr = dma_map_single(npu->dev, p, size, DMA_TO_DEVICE);
+	ret = dma_mapping_error(npu->dev, dma_addr);
+	if (ret)
+		return ret;
+
+	spin_lock_bh(&npu->cores[core].lock);
+
+	regmap_write(npu->regmap, REG_CR_MBQ0_CTRL(0) + offset, dma_addr);
+	regmap_write(npu->regmap, REG_CR_MBQ0_CTRL(1) + offset, size);
+	regmap_read(npu->regmap, REG_CR_MBQ0_CTRL(2) + offset, &val);
+	regmap_write(npu->regmap, REG_CR_MBQ0_CTRL(2) + offset, val + 1);
+	val = FIELD_PREP(MBOX_MSG_FUNC_ID, func_id) | MBOX_MSG_WAIT_RSP;
+	regmap_write(npu->regmap, REG_CR_MBQ0_CTRL(3) + offset, val);
+
+	ret = regmap_read_poll_timeout_atomic(npu->regmap,
+					      REG_CR_MBQ0_CTRL(3) + offset,
+					      val, (val & MBOX_MSG_DONE),
+					      100, 100 * MSEC_PER_SEC);
+	if (!ret && FIELD_GET(MBOX_MSG_STATUS, val) != NPU_MBOX_SUCCESS)
+		ret = -EINVAL;
+
+	spin_unlock_bh(&npu->cores[core].lock);
+
+	dma_unmap_single(npu->dev, dma_addr, size, DMA_TO_DEVICE);
+
+	return ret;
+}
+
+static int airoha_npu_run_firmware(struct device *dev, void __iomem *base,
+				   struct resource *res)
+{
+	const struct firmware *fw;
+	void __iomem *addr;
+	int ret;
+
+	ret = request_firmware(&fw, NPU_EN7581_FIRMWARE_RV32, dev);
+	if (ret)
+		return ret == -ENOENT ? -EPROBE_DEFER : ret;
+
+	if (fw->size > NPU_EN7581_FIRMWARE_RV32_MAX_SIZE) {
+		dev_err(dev, "%s: fw size too overlimit (%zu)\n",
+			NPU_EN7581_FIRMWARE_RV32, fw->size);
+		ret = -E2BIG;
+		goto out;
+	}
+
+	addr = devm_ioremap_resource(dev, res);
+	if (IS_ERR(addr)) {
+		ret = PTR_ERR(addr);
+		goto out;
+	}
+
+	memcpy_toio(addr, fw->data, fw->size);
+	release_firmware(fw);
+
+	ret = request_firmware(&fw, NPU_EN7581_FIRMWARE_DATA, dev);
+	if (ret)
+		return ret == -ENOENT ? -EPROBE_DEFER : ret;
+
+	if (fw->size > NPU_EN7581_FIRMWARE_DATA_MAX_SIZE) {
+		dev_err(dev, "%s: fw size too overlimit (%zu)\n",
+			NPU_EN7581_FIRMWARE_DATA, fw->size);
+		ret = -E2BIG;
+		goto out;
+	}
+
+	memcpy_toio(base + REG_NPU_LOCAL_SRAM, fw->data, fw->size);
+out:
+	release_firmware(fw);
+
+	return ret;
+}
+
+static irqreturn_t airoha_npu_mbox_handler(int irq, void *npu_instance)
+{
+	struct airoha_npu *npu = npu_instance;
+
+	/* clear mbox interrupt status */
+	regmap_write(npu->regmap, REG_CR_MBOX_INT_STATUS,
+		     MBOX_INT_STATUS_MASK);
+
+	/* acknowledge npu */
+	regmap_update_bits(npu->regmap, REG_CR_MBQ8_CTRL(3),
+			   MBOX_MSG_STATUS | MBOX_MSG_DONE, MBOX_MSG_DONE);
+
+	return IRQ_HANDLED;
+}
+
+static void airoha_npu_wdt_work(struct work_struct *work)
+{
+	struct airoha_npu_core *core;
+	struct airoha_npu *npu;
+	void *dump;
+	u32 val[3];
+	int c;
+
+	core = container_of(work, struct airoha_npu_core, wdt_work);
+	npu = core->npu;
+
+	dump = vzalloc(NPU_DUMP_SIZE);
+	if (!dump)
+		return;
+
+	c = core - &npu->cores[0];
+	regmap_bulk_read(npu->regmap, REG_PC_DBG(c), val, ARRAY_SIZE(val));
+	snprintf(dump, NPU_DUMP_SIZE, "PC: %08x SP: %08x LR: %08x\n",
+		 val[0], val[1], val[2]);
+
+	dev_coredumpv(npu->dev, dump, NPU_DUMP_SIZE, GFP_KERNEL);
+}
+
+static irqreturn_t airoha_npu_wdt_handler(int irq, void *core_instance)
+{
+	struct airoha_npu_core *core = core_instance;
+	struct airoha_npu *npu = core->npu;
+	int c = core - &npu->cores[0];
+	u32 val;
+
+	regmap_set_bits(npu->regmap, REG_WDT_TIMER_CTRL(c), WDT_INTR_MASK);
+	if (!regmap_read(npu->regmap, REG_WDT_TIMER_CTRL(c), &val) &&
+	    FIELD_GET(WDT_EN_MASK, val))
+		schedule_work(&core->wdt_work);
+
+	return IRQ_HANDLED;
+}
+
+static int airoha_npu_ppe_init(struct airoha_npu *npu)
+{
+	struct ppe_mbox_data *ppe_data;
+	int err;
+
+	ppe_data = kzalloc(sizeof(*ppe_data), GFP_KERNEL);
+	if (!ppe_data)
+		return -ENOMEM;
+
+	ppe_data->func_type = NPU_OP_SET;
+	ppe_data->func_id = PPE_FUNC_SET_WAIT_HWNAT_INIT;
+	ppe_data->init_info.ppe_type = PPE_TYPE_L2B_IPV4_IPV6;
+	ppe_data->init_info.wan_mode = QDMA_WAN_ETHER;
+
+	err = airoha_npu_send_msg(npu, NPU_FUNC_PPE, ppe_data,
+				  sizeof(*ppe_data));
+	kfree(ppe_data);
+
+	return err;
+}
+
+static int airoha_npu_ppe_deinit(struct airoha_npu *npu)
+{
+	struct ppe_mbox_data *ppe_data;
+	int err;
+
+	ppe_data = kzalloc(sizeof(*ppe_data), GFP_KERNEL);
+	if (!ppe_data)
+		return -ENOMEM;
+
+	ppe_data->func_type = NPU_OP_SET;
+	ppe_data->func_id = PPE_FUNC_SET_WAIT_HWNAT_DEINIT;
+
+	err = airoha_npu_send_msg(npu, NPU_FUNC_PPE, ppe_data,
+				  sizeof(*ppe_data));
+	kfree(ppe_data);
+
+	return err;
+}
+
+static int airoha_npu_ppe_flush_sram_entries(struct airoha_npu *npu,
+					     dma_addr_t foe_addr,
+					     int sram_num_entries)
+{
+	struct ppe_mbox_data *ppe_data;
+	int err;
+
+	ppe_data = kzalloc(sizeof(*ppe_data), GFP_KERNEL);
+	if (!ppe_data)
+		return -ENOMEM;
+
+	ppe_data->func_type = NPU_OP_SET;
+	ppe_data->func_id = PPE_FUNC_SET_WAIT_API;
+	ppe_data->set_info.func_id = PPE_SRAM_RESET_VAL;
+	ppe_data->set_info.data = foe_addr;
+	ppe_data->set_info.size = sram_num_entries;
+
+	err = airoha_npu_send_msg(npu, NPU_FUNC_PPE, ppe_data,
+				  sizeof(*ppe_data));
+	kfree(ppe_data);
+
+	return err;
+}
+
+static int airoha_npu_foe_commit_entry(struct airoha_npu *npu,
+				       dma_addr_t foe_addr,
+				       u32 entry_size, u32 hash, bool ppe2)
+{
+	struct ppe_mbox_data *ppe_data;
+	int err;
+
+	ppe_data = kzalloc(sizeof(*ppe_data), GFP_ATOMIC);
+	if (!ppe_data)
+		return -ENOMEM;
+
+	ppe_data->func_type = NPU_OP_SET;
+	ppe_data->func_id = PPE_FUNC_SET_WAIT_API;
+	ppe_data->set_info.data = foe_addr;
+	ppe_data->set_info.size = entry_size;
+	ppe_data->set_info.func_id = ppe2 ? PPE2_SRAM_SET_ENTRY
+					  : PPE_SRAM_SET_ENTRY;
+
+	err = airoha_npu_send_msg(npu, NPU_FUNC_PPE, ppe_data,
+				  sizeof(*ppe_data));
+	if (err)
+		goto out;
+
+	ppe_data->set_info.func_id = PPE_SRAM_SET_VAL;
+	ppe_data->set_info.data = hash;
+	ppe_data->set_info.size = sizeof(u32);
+
+	err = airoha_npu_send_msg(npu, NPU_FUNC_PPE, ppe_data,
+				  sizeof(*ppe_data));
+out:
+	kfree(ppe_data);
+
+	return err;
+}
+
+static int airoha_npu_ppe_stats_setup(struct airoha_npu *npu,
+				      dma_addr_t foe_stats_addr,
+				      u32 num_stats_entries)
+{
+	int err, size = num_stats_entries * sizeof(*npu->stats);
+	struct ppe_mbox_data *ppe_data;
+
+	ppe_data = kzalloc(sizeof(*ppe_data), GFP_ATOMIC);
+	if (!ppe_data)
+		return -ENOMEM;
+
+	ppe_data->func_type = NPU_OP_SET;
+	ppe_data->func_id = PPE_FUNC_SET_WAIT_FLOW_STATS_SETUP;
+	ppe_data->stats_info.foe_stats_addr = foe_stats_addr;
+
+	err = airoha_npu_send_msg(npu, NPU_FUNC_PPE, ppe_data,
+				  sizeof(*ppe_data));
+	if (err)
+		goto out;
+
+	npu->stats = devm_ioremap(npu->dev,
+				  ppe_data->stats_info.npu_stats_addr,
+				  size);
+	if (!npu->stats)
+		err = -ENOMEM;
+out:
+	kfree(ppe_data);
+
+	return err;
+}
+
+static int airoha_npu_wlan_msg_send(struct airoha_npu *npu, int ifindex,
+				    enum airoha_npu_wlan_set_cmd func_id,
+				    void *data, int data_len, gfp_t gfp)
+{
+	struct wlan_mbox_data *wlan_data;
+	int err, len;
+
+	len = sizeof(*wlan_data) + data_len;
+	wlan_data = kzalloc(len, gfp);
+	if (!wlan_data)
+		return -ENOMEM;
+
+	wlan_data->ifindex = ifindex;
+	wlan_data->func_type = NPU_OP_SET;
+	wlan_data->func_id = func_id;
+	memcpy(wlan_data->d, data, data_len);
+
+	err = airoha_npu_send_msg(npu, NPU_FUNC_WIFI, wlan_data, len);
+	kfree(wlan_data);
+
+	return err;
+}
+
+static int airoha_npu_wlan_msg_get(struct airoha_npu *npu, int ifindex,
+				   enum airoha_npu_wlan_get_cmd func_id,
+				   void *data, int data_len, gfp_t gfp)
+{
+	struct wlan_mbox_data *wlan_data;
+	int err, len;
+
+	len = sizeof(*wlan_data) + data_len;
+	wlan_data = kzalloc(len, gfp);
+	if (!wlan_data)
+		return -ENOMEM;
+
+	wlan_data->ifindex = ifindex;
+	wlan_data->func_type = NPU_OP_GET;
+	wlan_data->func_id = func_id;
+
+	err = airoha_npu_send_msg(npu, NPU_FUNC_WIFI, wlan_data, len);
+	if (!err)
+		memcpy(data, wlan_data->d, data_len);
+	kfree(wlan_data);
+
+	return err;
+}
+
+static int
+airoha_npu_wlan_set_reserved_memory(struct airoha_npu *npu,
+				    int ifindex, const char *name,
+				    enum airoha_npu_wlan_set_cmd func_id)
+{
+	struct device *dev = npu->dev;
+	struct resource res;
+	int err;
+	u32 val;
+
+	err = of_reserved_mem_region_to_resource_byname(dev->of_node, name,
+							&res);
+	if (err)
+		return err;
+
+	val = res.start;
+	return airoha_npu_wlan_msg_send(npu, ifindex, func_id, &val,
+					sizeof(val), GFP_KERNEL);
+}
+
+static int airoha_npu_wlan_init_memory(struct airoha_npu *npu)
+{
+	enum airoha_npu_wlan_set_cmd cmd = WLAN_FUNC_SET_WAIT_NPU_BAND0_ONCPU;
+	u32 val = 0;
+	int err;
+
+	err = airoha_npu_wlan_msg_send(npu, 1, cmd, &val, sizeof(val),
+				       GFP_KERNEL);
+	if (err)
+		return err;
+
+	cmd = WLAN_FUNC_SET_WAIT_TX_BUF_CHECK_ADDR;
+	err = airoha_npu_wlan_set_reserved_memory(npu, 0, "tx-bufid", cmd);
+	if (err)
+		return err;
+
+	cmd = WLAN_FUNC_SET_WAIT_PKT_BUF_ADDR;
+	err = airoha_npu_wlan_set_reserved_memory(npu, 0, "pkt", cmd);
+	if (err)
+		return err;
+
+	cmd = WLAN_FUNC_SET_WAIT_TX_PKT_BUF_ADDR;
+	err = airoha_npu_wlan_set_reserved_memory(npu, 0, "tx-pkt", cmd);
+	if (err)
+		return err;
+
+	cmd = WLAN_FUNC_SET_WAIT_IS_FORCE_TO_CPU;
+	return airoha_npu_wlan_msg_send(npu, 0, cmd, &val, sizeof(val),
+					GFP_KERNEL);
+}
+
+static u32 airoha_npu_wlan_queue_addr_get(struct airoha_npu *npu, int qid,
+					  bool xmit)
+{
+	if (xmit)
+		return REG_TX_BASE(qid + 2);
+
+	return REG_RX_BASE(qid);
+}
+
+static void airoha_npu_wlan_irq_status_set(struct airoha_npu *npu, u32 val)
+{
+	regmap_write(npu->regmap, REG_IRQ_STATUS, val);
+}
+
+static u32 airoha_npu_wlan_irq_status_get(struct airoha_npu *npu, int q)
+{
+	u32 val;
+
+	regmap_read(npu->regmap, REG_IRQ_STATUS, &val);
+	return val;
+}
+
+static void airoha_npu_wlan_irq_enable(struct airoha_npu *npu, int q)
+{
+	regmap_set_bits(npu->regmap, REG_IRQ_RXDONE(q), NPU_IRQ_RX_MASK(q));
+}
+
+static void airoha_npu_wlan_irq_disable(struct airoha_npu *npu, int q)
+{
+	regmap_clear_bits(npu->regmap, REG_IRQ_RXDONE(q), NPU_IRQ_RX_MASK(q));
+}
+
+struct airoha_npu *airoha_npu_get(struct device *dev)
+{
+	struct platform_device *pdev;
+	struct device_node *np;
+	struct airoha_npu *npu;
+
+	np = of_parse_phandle(dev->of_node, "airoha,npu", 0);
+	if (!np)
+		return ERR_PTR(-ENODEV);
+
+	pdev = of_find_device_by_node(np);
+
+	if (!pdev) {
+		dev_err(dev, "cannot find device node %s\n", np->name);
+		of_node_put(np);
+		return ERR_PTR(-ENODEV);
+	}
+	of_node_put(np);
+
+	if (!try_module_get(THIS_MODULE)) {
+		dev_err(dev, "failed to get the device driver module\n");
+		npu = ERR_PTR(-ENODEV);
+		goto error_pdev_put;
+	}
+
+	npu = platform_get_drvdata(pdev);
+	if (!npu) {
+		npu = ERR_PTR(-ENODEV);
+		goto error_module_put;
+	}
+
+	if (!device_link_add(dev, &pdev->dev, DL_FLAG_AUTOREMOVE_SUPPLIER)) {
+		dev_err(&pdev->dev,
+			"failed to create device link to consumer %s\n",
+			dev_name(dev));
+		npu = ERR_PTR(-EINVAL);
+		goto error_module_put;
+	}
+
+	return npu;
+
+error_module_put:
+	module_put(THIS_MODULE);
+error_pdev_put:
+	platform_device_put(pdev);
+
+	return npu;
+}
+EXPORT_SYMBOL_GPL(airoha_npu_get);
+
+void airoha_npu_put(struct airoha_npu *npu)
+{
+	module_put(THIS_MODULE);
+	put_device(npu->dev);
+}
+EXPORT_SYMBOL_GPL(airoha_npu_put);
+
+static const struct of_device_id of_airoha_npu_match[] = {
+	{ .compatible = "airoha,en7581-npu" },
+	{ /* sentinel */ }
+};
+MODULE_DEVICE_TABLE(of, of_airoha_npu_match);
+
+static const struct regmap_config regmap_config = {
+	.name			= "npu",
+	.reg_bits		= 32,
+	.val_bits		= 32,
+	.reg_stride		= 4,
+	.disable_locking	= true,
+};
+
+static int airoha_npu_probe(struct platform_device *pdev)
+{
+	struct device *dev = &pdev->dev;
+	struct airoha_npu *npu;
+	struct resource res;
+	void __iomem *base;
+	int i, irq, err;
+
+	base = devm_platform_ioremap_resource(pdev, 0);
+	if (IS_ERR(base))
+		return PTR_ERR(base);
+
+	npu = devm_kzalloc(dev, sizeof(*npu), GFP_KERNEL);
+	if (!npu)
+		return -ENOMEM;
+
+	npu->dev = dev;
+	npu->ops.ppe_init = airoha_npu_ppe_init;
+	npu->ops.ppe_deinit = airoha_npu_ppe_deinit;
+	npu->ops.ppe_init_stats = airoha_npu_ppe_stats_setup;
+	npu->ops.ppe_flush_sram_entries = airoha_npu_ppe_flush_sram_entries;
+	npu->ops.ppe_foe_commit_entry = airoha_npu_foe_commit_entry;
+	npu->ops.wlan_init_reserved_memory = airoha_npu_wlan_init_memory;
+	npu->ops.wlan_send_msg = airoha_npu_wlan_msg_send;
+	npu->ops.wlan_get_msg = airoha_npu_wlan_msg_get;
+	npu->ops.wlan_get_queue_addr = airoha_npu_wlan_queue_addr_get;
+	npu->ops.wlan_set_irq_status = airoha_npu_wlan_irq_status_set;
+	npu->ops.wlan_get_irq_status = airoha_npu_wlan_irq_status_get;
+	npu->ops.wlan_enable_irq = airoha_npu_wlan_irq_enable;
+	npu->ops.wlan_disable_irq = airoha_npu_wlan_irq_disable;
+
+	npu->regmap = devm_regmap_init_mmio(dev, base, &regmap_config);
+	if (IS_ERR(npu->regmap))
+		return PTR_ERR(npu->regmap);
+
+	err = of_reserved_mem_region_to_resource(dev->of_node, 0, &res);
+	if (err)
+		return err;
+
+	irq = platform_get_irq(pdev, 0);
+	if (irq < 0)
+		return irq;
+
+	err = devm_request_irq(dev, irq, airoha_npu_mbox_handler,
+			       IRQF_SHARED, "airoha-npu-mbox", npu);
+	if (err)
+		return err;
+
+	for (i = 0; i < ARRAY_SIZE(npu->cores); i++) {
+		struct airoha_npu_core *core = &npu->cores[i];
+
+		spin_lock_init(&core->lock);
+		core->npu = npu;
+
+		irq = platform_get_irq(pdev, i + 1);
+		if (irq < 0)
+			return irq;
+
+		err = devm_request_irq(dev, irq, airoha_npu_wdt_handler,
+				       IRQF_SHARED, "airoha-npu-wdt", core);
+		if (err)
+			return err;
+
+		INIT_WORK(&core->wdt_work, airoha_npu_wdt_work);
+	}
+
+	/* wlan IRQ lines */
+	for (i = 0; i < ARRAY_SIZE(npu->irqs); i++) {
+		irq = platform_get_irq(pdev, i + ARRAY_SIZE(npu->cores) + 1);
+		if (irq < 0)
+			return irq;
+
+		npu->irqs[i] = irq;
+	}
+
+	err = dma_set_coherent_mask(dev, DMA_BIT_MASK(32));
+	if (err)
+		return err;
+
+	err = airoha_npu_run_firmware(dev, base, &res);
+	if (err)
+		return dev_err_probe(dev, err, "failed to run npu firmware\n");
+
+	regmap_write(npu->regmap, REG_CR_NPU_MIB(10),
+		     res.start + NPU_EN7581_FIRMWARE_RV32_MAX_SIZE);
+	regmap_write(npu->regmap, REG_CR_NPU_MIB(11), 0x40000); /* SRAM 256K */
+	regmap_write(npu->regmap, REG_CR_NPU_MIB(12), 0);
+	regmap_write(npu->regmap, REG_CR_NPU_MIB(21), 1);
+	msleep(100);
+
+	/* setting booting address */
+	for (i = 0; i < NPU_NUM_CORES; i++)
+		regmap_write(npu->regmap, REG_CR_BOOT_BASE(i), res.start);
+	usleep_range(1000, 2000);
+
+	/* enable NPU cores */
+	regmap_write(npu->regmap, REG_CR_BOOT_CONFIG, 0xff);
+	regmap_write(npu->regmap, REG_CR_BOOT_TRIGGER, 0x1);
+	msleep(100);
+
+	platform_set_drvdata(pdev, npu);
+
+	return 0;
+}
+
+static void airoha_npu_remove(struct platform_device *pdev)
+{
+	struct airoha_npu *npu = platform_get_drvdata(pdev);
+	int i;
+
+	for (i = 0; i < ARRAY_SIZE(npu->cores); i++)
+		cancel_work_sync(&npu->cores[i].wdt_work);
+}
+
+static struct platform_driver airoha_npu_driver = {
+	.probe = airoha_npu_probe,
+	.remove = airoha_npu_remove,
+	.driver = {
+		.name = "airoha-npu",
+		.of_match_table = of_airoha_npu_match,
+	},
+};
+module_platform_driver(airoha_npu_driver);
+
+MODULE_FIRMWARE(NPU_EN7581_FIRMWARE_DATA);
+MODULE_FIRMWARE(NPU_EN7581_FIRMWARE_RV32);
+MODULE_LICENSE("GPL");
+MODULE_AUTHOR("Lorenzo Bianconi <lorenzo@kernel.org>");
+MODULE_DESCRIPTION("Airoha Network Processor Unit driver");
diff --git a/drivers/net/ethernet/airoha/airoha_ppe.c b/drivers/net/ethernet/airoha/airoha_ppe.c
new file mode 100644
index 000000000000..691361b25407
--- /dev/null
+++ b/drivers/net/ethernet/airoha/airoha_ppe.c
@@ -0,0 +1,1480 @@
+// SPDX-License-Identifier: GPL-2.0-only
+/*
+ * Copyright (c) 2025 AIROHA Inc
+ * Author: Lorenzo Bianconi <lorenzo@kernel.org>
+ */
+
+#include <linux/ip.h>
+#include <linux/ipv6.h>
+#include <linux/of_platform.h>
+#include <linux/platform_device.h>
+#include <linux/rhashtable.h>
+#include <net/ipv6.h>
+#include <net/pkt_cls.h>
+
+#include "airoha_regs.h"
+#include "airoha_eth.h"
+
+static DEFINE_MUTEX(flow_offload_mutex);
+static DEFINE_SPINLOCK(ppe_lock);
+
+static const struct rhashtable_params airoha_flow_table_params = {
+	.head_offset = offsetof(struct airoha_flow_table_entry, node),
+	.key_offset = offsetof(struct airoha_flow_table_entry, cookie),
+	.key_len = sizeof(unsigned long),
+	.automatic_shrinking = true,
+};
+
+static const struct rhashtable_params airoha_l2_flow_table_params = {
+	.head_offset = offsetof(struct airoha_flow_table_entry, l2_node),
+	.key_offset = offsetof(struct airoha_flow_table_entry, data.bridge),
+	.key_len = 2 * ETH_ALEN,
+	.automatic_shrinking = true,
+};
+
+static bool airoha_ppe2_is_enabled(struct airoha_eth *eth)
+{
+	return airoha_fe_rr(eth, REG_PPE_GLO_CFG(1)) & PPE_GLO_CFG_EN_MASK;
+}
+
+static u32 airoha_ppe_get_timestamp(struct airoha_ppe *ppe)
+{
+	u16 timestamp = airoha_fe_rr(ppe->eth, REG_FE_FOE_TS);
+
+	return FIELD_GET(AIROHA_FOE_IB1_BIND_TIMESTAMP, timestamp);
+}
+
+static void airoha_ppe_hw_init(struct airoha_ppe *ppe)
+{
+	u32 sram_tb_size, sram_num_entries, dram_num_entries;
+	struct airoha_eth *eth = ppe->eth;
+	int i;
+
+	sram_tb_size = PPE_SRAM_NUM_ENTRIES * sizeof(struct airoha_foe_entry);
+	dram_num_entries = PPE_RAM_NUM_ENTRIES_SHIFT(PPE_DRAM_NUM_ENTRIES);
+
+	for (i = 0; i < PPE_NUM; i++) {
+		int p;
+
+		airoha_fe_wr(eth, REG_PPE_TB_BASE(i),
+			     ppe->foe_dma + sram_tb_size);
+
+		airoha_fe_rmw(eth, REG_PPE_BND_AGE0(i),
+			      PPE_BIND_AGE0_DELTA_NON_L4 |
+			      PPE_BIND_AGE0_DELTA_UDP,
+			      FIELD_PREP(PPE_BIND_AGE0_DELTA_NON_L4, 1) |
+			      FIELD_PREP(PPE_BIND_AGE0_DELTA_UDP, 12));
+		airoha_fe_rmw(eth, REG_PPE_BND_AGE1(i),
+			      PPE_BIND_AGE1_DELTA_TCP_FIN |
+			      PPE_BIND_AGE1_DELTA_TCP,
+			      FIELD_PREP(PPE_BIND_AGE1_DELTA_TCP_FIN, 1) |
+			      FIELD_PREP(PPE_BIND_AGE1_DELTA_TCP, 7));
+
+		airoha_fe_rmw(eth, REG_PPE_TB_HASH_CFG(i),
+			      PPE_SRAM_TABLE_EN_MASK |
+			      PPE_SRAM_HASH1_EN_MASK |
+			      PPE_DRAM_TABLE_EN_MASK |
+			      PPE_SRAM_HASH0_MODE_MASK |
+			      PPE_SRAM_HASH1_MODE_MASK |
+			      PPE_DRAM_HASH0_MODE_MASK |
+			      PPE_DRAM_HASH1_MODE_MASK,
+			      FIELD_PREP(PPE_SRAM_TABLE_EN_MASK, 1) |
+			      FIELD_PREP(PPE_SRAM_HASH1_EN_MASK, 1) |
+			      FIELD_PREP(PPE_SRAM_HASH1_MODE_MASK, 1) |
+			      FIELD_PREP(PPE_DRAM_HASH1_MODE_MASK, 3));
+
+		airoha_fe_rmw(eth, REG_PPE_TB_CFG(i),
+			      PPE_TB_CFG_SEARCH_MISS_MASK |
+			      PPE_TB_CFG_KEEPALIVE_MASK |
+			      PPE_TB_ENTRY_SIZE_MASK,
+			      FIELD_PREP(PPE_TB_CFG_SEARCH_MISS_MASK, 3) |
+			      FIELD_PREP(PPE_TB_ENTRY_SIZE_MASK, 0));
+
+		airoha_fe_wr(eth, REG_PPE_HASH_SEED(i), PPE_HASH_SEED);
+
+		for (p = 0; p < ARRAY_SIZE(eth->ports); p++)
+			airoha_fe_rmw(eth, REG_PPE_MTU(i, p),
+				      FP0_EGRESS_MTU_MASK |
+				      FP1_EGRESS_MTU_MASK,
+				      FIELD_PREP(FP0_EGRESS_MTU_MASK,
+						 AIROHA_MAX_MTU) |
+				      FIELD_PREP(FP1_EGRESS_MTU_MASK,
+						 AIROHA_MAX_MTU));
+	}
+
+	if (airoha_ppe2_is_enabled(eth)) {
+		sram_num_entries =
+			PPE_RAM_NUM_ENTRIES_SHIFT(PPE1_SRAM_NUM_DATA_ENTRIES);
+		airoha_fe_rmw(eth, REG_PPE_TB_CFG(0),
+			      PPE_SRAM_TB_NUM_ENTRY_MASK |
+			      PPE_DRAM_TB_NUM_ENTRY_MASK,
+			      FIELD_PREP(PPE_SRAM_TB_NUM_ENTRY_MASK,
+					 sram_num_entries) |
+			      FIELD_PREP(PPE_DRAM_TB_NUM_ENTRY_MASK,
+					 dram_num_entries));
+		airoha_fe_rmw(eth, REG_PPE_TB_CFG(1),
+			      PPE_SRAM_TB_NUM_ENTRY_MASK |
+			      PPE_DRAM_TB_NUM_ENTRY_MASK,
+			      FIELD_PREP(PPE_SRAM_TB_NUM_ENTRY_MASK,
+					 sram_num_entries) |
+			      FIELD_PREP(PPE_DRAM_TB_NUM_ENTRY_MASK,
+					 dram_num_entries));
+	} else {
+		sram_num_entries =
+			PPE_RAM_NUM_ENTRIES_SHIFT(PPE_SRAM_NUM_DATA_ENTRIES);
+		airoha_fe_rmw(eth, REG_PPE_TB_CFG(0),
+			      PPE_SRAM_TB_NUM_ENTRY_MASK |
+			      PPE_DRAM_TB_NUM_ENTRY_MASK,
+			      FIELD_PREP(PPE_SRAM_TB_NUM_ENTRY_MASK,
+					 sram_num_entries) |
+			      FIELD_PREP(PPE_DRAM_TB_NUM_ENTRY_MASK,
+					 dram_num_entries));
+	}
+}
+
+static void airoha_ppe_flow_mangle_eth(const struct flow_action_entry *act, void *eth)
+{
+	void *dest = eth + act->mangle.offset;
+	const void *src = &act->mangle.val;
+
+	if (act->mangle.offset > 8)
+		return;
+
+	if (act->mangle.mask == 0xffff) {
+		src += 2;
+		dest += 2;
+	}
+
+	memcpy(dest, src, act->mangle.mask ? 2 : 4);
+}
+
+static int airoha_ppe_flow_mangle_ports(const struct flow_action_entry *act,
+					struct airoha_flow_data *data)
+{
+	u32 val = be32_to_cpu((__force __be32)act->mangle.val);
+
+	switch (act->mangle.offset) {
+	case 0:
+		if ((__force __be32)act->mangle.mask == ~cpu_to_be32(0xffff))
+			data->dst_port = cpu_to_be16(val);
+		else
+			data->src_port = cpu_to_be16(val >> 16);
+		break;
+	case 2:
+		data->dst_port = cpu_to_be16(val);
+		break;
+	default:
+		return -EINVAL;
+	}
+
+	return 0;
+}
+
+static int airoha_ppe_flow_mangle_ipv4(const struct flow_action_entry *act,
+				       struct airoha_flow_data *data)
+{
+	__be32 *dest;
+
+	switch (act->mangle.offset) {
+	case offsetof(struct iphdr, saddr):
+		dest = &data->v4.src_addr;
+		break;
+	case offsetof(struct iphdr, daddr):
+		dest = &data->v4.dst_addr;
+		break;
+	default:
+		return -EINVAL;
+	}
+
+	memcpy(dest, &act->mangle.val, sizeof(u32));
+
+	return 0;
+}
+
+static int airoha_ppe_get_wdma_info(struct net_device *dev, const u8 *addr,
+				    struct airoha_wdma_info *info)
+{
+	struct net_device_path_stack stack;
+	struct net_device_path *path;
+	int err;
+
+	if (!dev)
+		return -ENODEV;
+
+	err = dev_fill_forward_path(dev, addr, &stack);
+	if (err)
+		return err;
+
+	path = &stack.path[stack.num_paths - 1];
+	if (path->type != DEV_PATH_MTK_WDMA)
+		return -1;
+
+	info->idx = path->mtk_wdma.wdma_idx;
+	info->bss = path->mtk_wdma.bss;
+	info->wcid = path->mtk_wdma.wcid;
+
+	return 0;
+}
+
+static int airoha_get_dsa_port(struct net_device **dev)
+{
+#if IS_ENABLED(CONFIG_NET_DSA)
+	struct dsa_port *dp = dsa_port_from_netdev(*dev);
+
+	if (IS_ERR(dp))
+		return -ENODEV;
+
+	*dev = dsa_port_to_conduit(dp);
+	return dp->index;
+#else
+	return -ENODEV;
+#endif
+}
+
+static void airoha_ppe_foe_set_bridge_addrs(struct airoha_foe_bridge *br,
+					    struct ethhdr *eh)
+{
+	br->dest_mac_hi = get_unaligned_be32(eh->h_dest);
+	br->dest_mac_lo = get_unaligned_be16(eh->h_dest + 4);
+	br->src_mac_hi = get_unaligned_be16(eh->h_source);
+	br->src_mac_lo = get_unaligned_be32(eh->h_source + 2);
+}
+
+static int airoha_ppe_foe_entry_prepare(struct airoha_eth *eth,
+					struct airoha_foe_entry *hwe,
+					struct net_device *dev, int type,
+					struct airoha_flow_data *data,
+					int l4proto)
+{
+	u32 qdata = FIELD_PREP(AIROHA_FOE_SHAPER_ID, 0x7f), ports_pad, val;
+	int wlan_etype = -EINVAL, dsa_port = airoha_get_dsa_port(&dev);
+	struct airoha_foe_mac_info_common *l2;
+	u8 smac_id = 0xf;
+
+	memset(hwe, 0, sizeof(*hwe));
+
+	val = FIELD_PREP(AIROHA_FOE_IB1_BIND_STATE, AIROHA_FOE_STATE_BIND) |
+	      FIELD_PREP(AIROHA_FOE_IB1_BIND_PACKET_TYPE, type) |
+	      FIELD_PREP(AIROHA_FOE_IB1_BIND_UDP, l4proto == IPPROTO_UDP) |
+	      FIELD_PREP(AIROHA_FOE_IB1_BIND_VLAN_LAYER, data->vlan.num) |
+	      FIELD_PREP(AIROHA_FOE_IB1_BIND_VPM, data->vlan.num) |
+	      FIELD_PREP(AIROHA_FOE_IB1_BIND_PPPOE, data->pppoe.num) |
+	      AIROHA_FOE_IB1_BIND_TTL;
+	hwe->ib1 = val;
+
+	val = FIELD_PREP(AIROHA_FOE_IB2_PORT_AG, 0x1f);
+	if (dev) {
+		struct airoha_wdma_info info = {};
+
+		if (!airoha_ppe_get_wdma_info(dev, data->eth.h_dest, &info)) {
+			val |= FIELD_PREP(AIROHA_FOE_IB2_NBQ, info.idx) |
+			       FIELD_PREP(AIROHA_FOE_IB2_PSE_PORT,
+					  FE_PSE_PORT_CDM4);
+			qdata |= FIELD_PREP(AIROHA_FOE_ACTDP, info.bss);
+			wlan_etype = FIELD_PREP(AIROHA_FOE_MAC_WDMA_BAND,
+						info.idx) |
+				     FIELD_PREP(AIROHA_FOE_MAC_WDMA_WCID,
+						info.wcid);
+		} else {
+			struct airoha_gdm_port *port = netdev_priv(dev);
+			u8 pse_port;
+
+			if (!airoha_is_valid_gdm_port(eth, port))
+				return -EINVAL;
+
+			if (dsa_port >= 0)
+				pse_port = port->id == 4 ? FE_PSE_PORT_GDM4
+							 : port->id;
+			else
+				pse_port = 2; /* uplink relies on GDM2
+					       * loopback
+					       */
+
+			val |= FIELD_PREP(AIROHA_FOE_IB2_PSE_PORT, pse_port) |
+			       AIROHA_FOE_IB2_PSE_QOS;
+			/* For downlink traffic consume SRAM memory for hw
+			 * forwarding descriptors queue.
+			 */
+			if (airhoa_is_lan_gdm_port(port))
+				val |= AIROHA_FOE_IB2_FAST_PATH;
+			if (dsa_port >= 0)
+				val |= FIELD_PREP(AIROHA_FOE_IB2_NBQ,
+						  dsa_port);
+
+			smac_id = port->id;
+		}
+	}
+
+	if (is_multicast_ether_addr(data->eth.h_dest))
+		val |= AIROHA_FOE_IB2_MULTICAST;
+
+	ports_pad = 0xa5a5a500 | (l4proto & 0xff);
+	if (type == PPE_PKT_TYPE_IPV4_ROUTE)
+		hwe->ipv4.orig_tuple.ports = ports_pad;
+	if (type == PPE_PKT_TYPE_IPV6_ROUTE_3T)
+		hwe->ipv6.ports = ports_pad;
+
+	if (type == PPE_PKT_TYPE_BRIDGE) {
+		airoha_ppe_foe_set_bridge_addrs(&hwe->bridge, &data->eth);
+		hwe->bridge.data = qdata;
+		hwe->bridge.ib2 = val;
+		l2 = &hwe->bridge.l2.common;
+	} else if (type >= PPE_PKT_TYPE_IPV6_ROUTE_3T) {
+		hwe->ipv6.data = qdata;
+		hwe->ipv6.ib2 = val;
+		l2 = &hwe->ipv6.l2;
+		l2->etype = ETH_P_IPV6;
+	} else {
+		hwe->ipv4.data = qdata;
+		hwe->ipv4.ib2 = val;
+		l2 = &hwe->ipv4.l2.common;
+		l2->etype = ETH_P_IP;
+	}
+
+	l2->dest_mac_hi = get_unaligned_be32(data->eth.h_dest);
+	l2->dest_mac_lo = get_unaligned_be16(data->eth.h_dest + 4);
+	if (type <= PPE_PKT_TYPE_IPV4_DSLITE) {
+		struct airoha_foe_mac_info *mac_info;
+
+		l2->src_mac_hi = get_unaligned_be32(data->eth.h_source);
+		hwe->ipv4.l2.src_mac_lo =
+			get_unaligned_be16(data->eth.h_source + 4);
+
+		mac_info = (struct airoha_foe_mac_info *)l2;
+		mac_info->pppoe_id = data->pppoe.sid;
+	} else {
+		l2->src_mac_hi = FIELD_PREP(AIROHA_FOE_MAC_SMAC_ID, smac_id) |
+				 FIELD_PREP(AIROHA_FOE_MAC_PPPOE_ID,
+					    data->pppoe.sid);
+	}
+
+	if (data->vlan.num) {
+		l2->vlan1 = data->vlan.hdr[0].id;
+		if (data->vlan.num == 2)
+			l2->vlan2 = data->vlan.hdr[1].id;
+	}
+
+	if (wlan_etype >= 0) {
+		l2->etype = wlan_etype;
+	} else if (dsa_port >= 0) {
+		l2->etype = BIT(dsa_port);
+		l2->etype |= !data->vlan.num ? BIT(15) : 0;
+	} else if (data->pppoe.num) {
+		l2->etype = ETH_P_PPP_SES;
+	}
+
+	return 0;
+}
+
+static int airoha_ppe_foe_entry_set_ipv4_tuple(struct airoha_foe_entry *hwe,
+					       struct airoha_flow_data *data,
+					       bool egress)
+{
+	int type = FIELD_GET(AIROHA_FOE_IB1_BIND_PACKET_TYPE, hwe->ib1);
+	struct airoha_foe_ipv4_tuple *t;
+
+	switch (type) {
+	case PPE_PKT_TYPE_IPV4_HNAPT:
+		if (egress) {
+			t = &hwe->ipv4.new_tuple;
+			break;
+		}
+		fallthrough;
+	case PPE_PKT_TYPE_IPV4_DSLITE:
+	case PPE_PKT_TYPE_IPV4_ROUTE:
+		t = &hwe->ipv4.orig_tuple;
+		break;
+	default:
+		WARN_ON_ONCE(1);
+		return -EINVAL;
+	}
+
+	t->src_ip = be32_to_cpu(data->v4.src_addr);
+	t->dest_ip = be32_to_cpu(data->v4.dst_addr);
+
+	if (type != PPE_PKT_TYPE_IPV4_ROUTE) {
+		t->src_port = be16_to_cpu(data->src_port);
+		t->dest_port = be16_to_cpu(data->dst_port);
+	}
+
+	return 0;
+}
+
+static int airoha_ppe_foe_entry_set_ipv6_tuple(struct airoha_foe_entry *hwe,
+					       struct airoha_flow_data *data)
+
+{
+	int type = FIELD_GET(AIROHA_FOE_IB1_BIND_PACKET_TYPE, hwe->ib1);
+	u32 *src, *dest;
+
+	switch (type) {
+	case PPE_PKT_TYPE_IPV6_ROUTE_5T:
+	case PPE_PKT_TYPE_IPV6_6RD:
+		hwe->ipv6.src_port = be16_to_cpu(data->src_port);
+		hwe->ipv6.dest_port = be16_to_cpu(data->dst_port);
+		fallthrough;
+	case PPE_PKT_TYPE_IPV6_ROUTE_3T:
+		src = hwe->ipv6.src_ip;
+		dest = hwe->ipv6.dest_ip;
+		break;
+	default:
+		WARN_ON_ONCE(1);
+		return -EINVAL;
+	}
+
+	ipv6_addr_be32_to_cpu(src, data->v6.src_addr.s6_addr32);
+	ipv6_addr_be32_to_cpu(dest, data->v6.dst_addr.s6_addr32);
+
+	return 0;
+}
+
+static u32 airoha_ppe_foe_get_entry_hash(struct airoha_foe_entry *hwe)
+{
+	int type = FIELD_GET(AIROHA_FOE_IB1_BIND_PACKET_TYPE, hwe->ib1);
+	u32 hash, hv1, hv2, hv3;
+
+	switch (type) {
+	case PPE_PKT_TYPE_IPV4_ROUTE:
+	case PPE_PKT_TYPE_IPV4_HNAPT:
+		hv1 = hwe->ipv4.orig_tuple.ports;
+		hv2 = hwe->ipv4.orig_tuple.dest_ip;
+		hv3 = hwe->ipv4.orig_tuple.src_ip;
+		break;
+	case PPE_PKT_TYPE_IPV6_ROUTE_3T:
+	case PPE_PKT_TYPE_IPV6_ROUTE_5T:
+		hv1 = hwe->ipv6.src_ip[3] ^ hwe->ipv6.dest_ip[3];
+		hv1 ^= hwe->ipv6.ports;
+
+		hv2 = hwe->ipv6.src_ip[2] ^ hwe->ipv6.dest_ip[2];
+		hv2 ^= hwe->ipv6.dest_ip[0];
+
+		hv3 = hwe->ipv6.src_ip[1] ^ hwe->ipv6.dest_ip[1];
+		hv3 ^= hwe->ipv6.src_ip[0];
+		break;
+	case PPE_PKT_TYPE_BRIDGE: {
+		struct airoha_foe_mac_info *l2 = &hwe->bridge.l2;
+
+		hv1 = l2->common.src_mac_hi & 0xffff;
+		hv1 = hv1 << 16 | l2->src_mac_lo;
+
+		hv2 = l2->common.dest_mac_lo;
+		hv2 = hv2 << 16;
+		hv2 = hv2 | ((l2->common.src_mac_hi & 0xffff0000) >> 16);
+
+		hv3 = l2->common.dest_mac_hi;
+		break;
+	}
+	case PPE_PKT_TYPE_IPV4_DSLITE:
+	case PPE_PKT_TYPE_IPV6_6RD:
+	default:
+		WARN_ON_ONCE(1);
+		return PPE_HASH_MASK;
+	}
+
+	hash = (hv1 & hv2) | ((~hv1) & hv3);
+	hash = (hash >> 24) | ((hash & 0xffffff) << 8);
+	hash ^= hv1 ^ hv2 ^ hv3;
+	hash ^= hash >> 16;
+	hash &= PPE_NUM_ENTRIES - 1;
+
+	return hash;
+}
+
+static u32 airoha_ppe_foe_get_flow_stats_index(struct airoha_ppe *ppe, u32 hash)
+{
+	if (!airoha_ppe2_is_enabled(ppe->eth))
+		return hash;
+
+	return hash >= PPE_STATS_NUM_ENTRIES ? hash - PPE1_STATS_NUM_ENTRIES
+					     : hash;
+}
+
+static void airoha_ppe_foe_flow_stat_entry_reset(struct airoha_ppe *ppe,
+						 struct airoha_npu *npu,
+						 int index)
+{
+	memset_io(&npu->stats[index], 0, sizeof(*npu->stats));
+	memset(&ppe->foe_stats[index], 0, sizeof(*ppe->foe_stats));
+}
+
+static void airoha_ppe_foe_flow_stats_reset(struct airoha_ppe *ppe,
+					    struct airoha_npu *npu)
+{
+	int i;
+
+	for (i = 0; i < PPE_STATS_NUM_ENTRIES; i++)
+		airoha_ppe_foe_flow_stat_entry_reset(ppe, npu, i);
+}
+
+static void airoha_ppe_foe_flow_stats_update(struct airoha_ppe *ppe,
+					     struct airoha_npu *npu,
+					     struct airoha_foe_entry *hwe,
+					     u32 hash)
+{
+	int type = FIELD_GET(AIROHA_FOE_IB1_BIND_PACKET_TYPE, hwe->ib1);
+	u32 index, pse_port, val, *data, *ib2, *meter;
+	u8 nbq;
+
+	index = airoha_ppe_foe_get_flow_stats_index(ppe, hash);
+	if (index >= PPE_STATS_NUM_ENTRIES)
+		return;
+
+	if (type == PPE_PKT_TYPE_BRIDGE) {
+		data = &hwe->bridge.data;
+		ib2 = &hwe->bridge.ib2;
+		meter = &hwe->bridge.l2.meter;
+	} else if (type >= PPE_PKT_TYPE_IPV6_ROUTE_3T) {
+		data = &hwe->ipv6.data;
+		ib2 = &hwe->ipv6.ib2;
+		meter = &hwe->ipv6.meter;
+	} else {
+		data = &hwe->ipv4.data;
+		ib2 = &hwe->ipv4.ib2;
+		meter = &hwe->ipv4.l2.meter;
+	}
+
+	pse_port = FIELD_GET(AIROHA_FOE_IB2_PSE_PORT, *ib2);
+	if (pse_port == FE_PSE_PORT_CDM4)
+		return;
+
+	airoha_ppe_foe_flow_stat_entry_reset(ppe, npu, index);
+
+	val = FIELD_GET(AIROHA_FOE_CHANNEL | AIROHA_FOE_QID, *data);
+	*data = (*data & ~AIROHA_FOE_ACTDP) |
+		FIELD_PREP(AIROHA_FOE_ACTDP, val);
+
+	val = *ib2 & (AIROHA_FOE_IB2_NBQ | AIROHA_FOE_IB2_PSE_PORT |
+		      AIROHA_FOE_IB2_PSE_QOS | AIROHA_FOE_IB2_FAST_PATH);
+	*meter |= FIELD_PREP(AIROHA_FOE_TUNNEL_MTU, val);
+
+	nbq = pse_port == 1 ? 6 : 5;
+	*ib2 &= ~(AIROHA_FOE_IB2_NBQ | AIROHA_FOE_IB2_PSE_PORT |
+		  AIROHA_FOE_IB2_PSE_QOS);
+	*ib2 |= FIELD_PREP(AIROHA_FOE_IB2_PSE_PORT, 6) |
+		FIELD_PREP(AIROHA_FOE_IB2_NBQ, nbq);
+}
+
+static struct airoha_foe_entry *
+airoha_ppe_foe_get_entry_locked(struct airoha_ppe *ppe, u32 hash)
+{
+	lockdep_assert_held(&ppe_lock);
+
+	if (hash < PPE_SRAM_NUM_ENTRIES) {
+		u32 *hwe = ppe->foe + hash * sizeof(struct airoha_foe_entry);
+		struct airoha_eth *eth = ppe->eth;
+		bool ppe2;
+		u32 val;
+		int i;
+
+		ppe2 = airoha_ppe2_is_enabled(ppe->eth) &&
+		       hash >= PPE1_SRAM_NUM_ENTRIES;
+		airoha_fe_wr(ppe->eth, REG_PPE_RAM_CTRL(ppe2),
+			     FIELD_PREP(PPE_SRAM_CTRL_ENTRY_MASK, hash) |
+			     PPE_SRAM_CTRL_REQ_MASK);
+		if (read_poll_timeout_atomic(airoha_fe_rr, val,
+					     val & PPE_SRAM_CTRL_ACK_MASK,
+					     10, 100, false, eth,
+					     REG_PPE_RAM_CTRL(ppe2)))
+			return NULL;
+
+		for (i = 0; i < sizeof(struct airoha_foe_entry) / 4; i++)
+			hwe[i] = airoha_fe_rr(eth,
+					      REG_PPE_RAM_ENTRY(ppe2, i));
+	}
+
+	return ppe->foe + hash * sizeof(struct airoha_foe_entry);
+}
+
+struct airoha_foe_entry *airoha_ppe_foe_get_entry(struct airoha_ppe *ppe,
+						  u32 hash)
+{
+	struct airoha_foe_entry *hwe;
+
+	spin_lock_bh(&ppe_lock);
+	hwe = airoha_ppe_foe_get_entry_locked(ppe, hash);
+	spin_unlock_bh(&ppe_lock);
+
+	return hwe;
+}
+
+static bool airoha_ppe_foe_compare_entry(struct airoha_flow_table_entry *e,
+					 struct airoha_foe_entry *hwe)
+{
+	int type = FIELD_GET(AIROHA_FOE_IB1_BIND_PACKET_TYPE, e->data.ib1);
+	int len;
+
+	if ((hwe->ib1 ^ e->data.ib1) & AIROHA_FOE_IB1_BIND_UDP)
+		return false;
+
+	if (type > PPE_PKT_TYPE_IPV4_DSLITE)
+		len = offsetof(struct airoha_foe_entry, ipv6.data);
+	else
+		len = offsetof(struct airoha_foe_entry, ipv4.ib2);
+
+	return !memcmp(&e->data.d, &hwe->d, len - sizeof(hwe->ib1));
+}
+
+static int airoha_ppe_foe_commit_entry(struct airoha_ppe *ppe,
+				       struct airoha_foe_entry *e,
+				       u32 hash, bool rx_wlan)
+{
+	struct airoha_foe_entry *hwe = ppe->foe + hash * sizeof(*hwe);
+	u32 ts = airoha_ppe_get_timestamp(ppe);
+	struct airoha_eth *eth = ppe->eth;
+	struct airoha_npu *npu;
+	int err = 0;
+
+	memcpy(&hwe->d, &e->d, sizeof(*hwe) - sizeof(hwe->ib1));
+	wmb();
+
+	e->ib1 &= ~AIROHA_FOE_IB1_BIND_TIMESTAMP;
+	e->ib1 |= FIELD_PREP(AIROHA_FOE_IB1_BIND_TIMESTAMP, ts);
+	hwe->ib1 = e->ib1;
+
+	rcu_read_lock();
+
+	npu = rcu_dereference(eth->npu);
+	if (!npu) {
+		err = -ENODEV;
+		goto unlock;
+	}
+
+	if (!rx_wlan)
+		airoha_ppe_foe_flow_stats_update(ppe, npu, hwe, hash);
+
+	if (hash < PPE_SRAM_NUM_ENTRIES) {
+		dma_addr_t addr = ppe->foe_dma + hash * sizeof(*hwe);
+		bool ppe2 = airoha_ppe2_is_enabled(eth) &&
+			    hash >= PPE1_SRAM_NUM_ENTRIES;
+
+		err = npu->ops.ppe_foe_commit_entry(npu, addr, sizeof(*hwe),
+						    hash, ppe2);
+	}
+unlock:
+	rcu_read_unlock();
+
+	return err;
+}
+
+static void airoha_ppe_foe_remove_flow(struct airoha_ppe *ppe,
+				       struct airoha_flow_table_entry *e)
+{
+	lockdep_assert_held(&ppe_lock);
+
+	hlist_del_init(&e->list);
+	if (e->hash != 0xffff) {
+		e->data.ib1 &= ~AIROHA_FOE_IB1_BIND_STATE;
+		e->data.ib1 |= FIELD_PREP(AIROHA_FOE_IB1_BIND_STATE,
+					  AIROHA_FOE_STATE_INVALID);
+		airoha_ppe_foe_commit_entry(ppe, &e->data, e->hash, false);
+		e->hash = 0xffff;
+	}
+	if (e->type == FLOW_TYPE_L2_SUBFLOW) {
+		hlist_del_init(&e->l2_subflow_node);
+		kfree(e);
+	}
+}
+
+static void airoha_ppe_foe_remove_l2_flow(struct airoha_ppe *ppe,
+					  struct airoha_flow_table_entry *e)
+{
+	struct hlist_head *head = &e->l2_flows;
+	struct hlist_node *n;
+
+	lockdep_assert_held(&ppe_lock);
+
+	rhashtable_remove_fast(&ppe->l2_flows, &e->l2_node,
+			       airoha_l2_flow_table_params);
+	hlist_for_each_entry_safe(e, n, head, l2_subflow_node)
+		airoha_ppe_foe_remove_flow(ppe, e);
+}
+
+static void airoha_ppe_foe_flow_remove_entry(struct airoha_ppe *ppe,
+					     struct airoha_flow_table_entry *e)
+{
+	spin_lock_bh(&ppe_lock);
+
+	if (e->type == FLOW_TYPE_L2)
+		airoha_ppe_foe_remove_l2_flow(ppe, e);
+	else
+		airoha_ppe_foe_remove_flow(ppe, e);
+
+	spin_unlock_bh(&ppe_lock);
+}
+
+static int
+airoha_ppe_foe_commit_subflow_entry(struct airoha_ppe *ppe,
+				    struct airoha_flow_table_entry *e,
+				    u32 hash, bool rx_wlan)
+{
+	u32 mask = AIROHA_FOE_IB1_BIND_PACKET_TYPE | AIROHA_FOE_IB1_BIND_UDP;
+	struct airoha_foe_entry *hwe_p, hwe;
+	struct airoha_flow_table_entry *f;
+	int type;
+
+	hwe_p = airoha_ppe_foe_get_entry_locked(ppe, hash);
+	if (!hwe_p)
+		return -EINVAL;
+
+	f = kzalloc(sizeof(*f), GFP_ATOMIC);
+	if (!f)
+		return -ENOMEM;
+
+	hlist_add_head(&f->l2_subflow_node, &e->l2_flows);
+	f->type = FLOW_TYPE_L2_SUBFLOW;
+	f->hash = hash;
+
+	memcpy(&hwe, hwe_p, sizeof(*hwe_p));
+	hwe.ib1 = (hwe.ib1 & mask) | (e->data.ib1 & ~mask);
+
+	type = FIELD_GET(AIROHA_FOE_IB1_BIND_PACKET_TYPE, hwe.ib1);
+	if (type >= PPE_PKT_TYPE_IPV6_ROUTE_3T) {
+		memcpy(&hwe.ipv6.l2, &e->data.bridge.l2, sizeof(hwe.ipv6.l2));
+		hwe.ipv6.ib2 = e->data.bridge.ib2;
+		/* setting smac_id to 0xf instruct the hw to keep original
+		 * source mac address
+		 */
+		hwe.ipv6.l2.src_mac_hi = FIELD_PREP(AIROHA_FOE_MAC_SMAC_ID,
+						    0xf);
+	} else {
+		memcpy(&hwe.bridge.l2, &e->data.bridge.l2,
+		       sizeof(hwe.bridge.l2));
+		hwe.bridge.ib2 = e->data.bridge.ib2;
+		if (type == PPE_PKT_TYPE_IPV4_HNAPT)
+			memcpy(&hwe.ipv4.new_tuple, &hwe.ipv4.orig_tuple,
+			       sizeof(hwe.ipv4.new_tuple));
+	}
+
+	hwe.bridge.data = e->data.bridge.data;
+	airoha_ppe_foe_commit_entry(ppe, &hwe, hash, rx_wlan);
+
+	return 0;
+}
+
+static void airoha_ppe_foe_insert_entry(struct airoha_ppe *ppe,
+					struct sk_buff *skb,
+					u32 hash, bool rx_wlan)
+{
+	struct airoha_flow_table_entry *e;
+	struct airoha_foe_bridge br = {};
+	struct airoha_foe_entry *hwe;
+	bool commit_done = false;
+	struct hlist_node *n;
+	u32 index, state;
+
+	spin_lock_bh(&ppe_lock);
+
+	hwe = airoha_ppe_foe_get_entry_locked(ppe, hash);
+	if (!hwe)
+		goto unlock;
+
+	state = FIELD_GET(AIROHA_FOE_IB1_BIND_STATE, hwe->ib1);
+	if (state == AIROHA_FOE_STATE_BIND)
+		goto unlock;
+
+	index = airoha_ppe_foe_get_entry_hash(hwe);
+	hlist_for_each_entry_safe(e, n, &ppe->foe_flow[index], list) {
+		if (e->type == FLOW_TYPE_L2_SUBFLOW) {
+			state = FIELD_GET(AIROHA_FOE_IB1_BIND_STATE, hwe->ib1);
+			if (state != AIROHA_FOE_STATE_BIND) {
+				e->hash = 0xffff;
+				airoha_ppe_foe_remove_flow(ppe, e);
+			}
+			continue;
+		}
+
+		if (!airoha_ppe_foe_compare_entry(e, hwe))
+			continue;
+
+		airoha_ppe_foe_commit_entry(ppe, &e->data, hash, rx_wlan);
+		commit_done = true;
+		e->hash = hash;
+	}
+
+	if (commit_done)
+		goto unlock;
+
+	airoha_ppe_foe_set_bridge_addrs(&br, eth_hdr(skb));
+	e = rhashtable_lookup_fast(&ppe->l2_flows, &br,
+				   airoha_l2_flow_table_params);
+	if (e)
+		airoha_ppe_foe_commit_subflow_entry(ppe, e, hash, rx_wlan);
+unlock:
+	spin_unlock_bh(&ppe_lock);
+}
+
+static int
+airoha_ppe_foe_l2_flow_commit_entry(struct airoha_ppe *ppe,
+				    struct airoha_flow_table_entry *e)
+{
+	struct airoha_flow_table_entry *prev;
+
+	e->type = FLOW_TYPE_L2;
+	prev = rhashtable_lookup_get_insert_fast(&ppe->l2_flows, &e->l2_node,
+						 airoha_l2_flow_table_params);
+	if (!prev)
+		return 0;
+
+	if (IS_ERR(prev))
+		return PTR_ERR(prev);
+
+	return rhashtable_replace_fast(&ppe->l2_flows, &prev->l2_node,
+				       &e->l2_node,
+				       airoha_l2_flow_table_params);
+}
+
+static int airoha_ppe_foe_flow_commit_entry(struct airoha_ppe *ppe,
+					    struct airoha_flow_table_entry *e)
+{
+	int type = FIELD_GET(AIROHA_FOE_IB1_BIND_PACKET_TYPE, e->data.ib1);
+	u32 hash;
+
+	if (type == PPE_PKT_TYPE_BRIDGE)
+		return airoha_ppe_foe_l2_flow_commit_entry(ppe, e);
+
+	hash = airoha_ppe_foe_get_entry_hash(&e->data);
+	e->type = FLOW_TYPE_L4;
+	e->hash = 0xffff;
+
+	spin_lock_bh(&ppe_lock);
+	hlist_add_head(&e->list, &ppe->foe_flow[hash]);
+	spin_unlock_bh(&ppe_lock);
+
+	return 0;
+}
+
+static int airoha_ppe_get_entry_idle_time(struct airoha_ppe *ppe, u32 ib1)
+{
+	u32 state = FIELD_GET(AIROHA_FOE_IB1_BIND_STATE, ib1);
+	u32 ts, ts_mask, now = airoha_ppe_get_timestamp(ppe);
+	int idle;
+
+	if (state == AIROHA_FOE_STATE_BIND) {
+		ts = FIELD_GET(AIROHA_FOE_IB1_BIND_TIMESTAMP, ib1);
+		ts_mask = AIROHA_FOE_IB1_BIND_TIMESTAMP;
+	} else {
+		ts = FIELD_GET(AIROHA_FOE_IB1_UNBIND_TIMESTAMP, ib1);
+		now = FIELD_GET(AIROHA_FOE_IB1_UNBIND_TIMESTAMP, now);
+		ts_mask = AIROHA_FOE_IB1_UNBIND_TIMESTAMP;
+	}
+	idle = now - ts;
+
+	return idle < 0 ? idle + ts_mask + 1 : idle;
+}
+
+static void
+airoha_ppe_foe_flow_l2_entry_update(struct airoha_ppe *ppe,
+				    struct airoha_flow_table_entry *e)
+{
+	int min_idle = airoha_ppe_get_entry_idle_time(ppe, e->data.ib1);
+	struct airoha_flow_table_entry *iter;
+	struct hlist_node *n;
+
+	lockdep_assert_held(&ppe_lock);
+
+	hlist_for_each_entry_safe(iter, n, &e->l2_flows, l2_subflow_node) {
+		struct airoha_foe_entry *hwe;
+		u32 ib1, state;
+		int idle;
+
+		hwe = airoha_ppe_foe_get_entry_locked(ppe, iter->hash);
+		if (!hwe)
+			continue;
+
+		ib1 = READ_ONCE(hwe->ib1);
+		state = FIELD_GET(AIROHA_FOE_IB1_BIND_STATE, ib1);
+		if (state != AIROHA_FOE_STATE_BIND) {
+			iter->hash = 0xffff;
+			airoha_ppe_foe_remove_flow(ppe, iter);
+			continue;
+		}
+
+		idle = airoha_ppe_get_entry_idle_time(ppe, ib1);
+		if (idle >= min_idle)
+			continue;
+
+		min_idle = idle;
+		e->data.ib1 &= ~AIROHA_FOE_IB1_BIND_TIMESTAMP;
+		e->data.ib1 |= ib1 & AIROHA_FOE_IB1_BIND_TIMESTAMP;
+	}
+}
+
+static void airoha_ppe_foe_flow_entry_update(struct airoha_ppe *ppe,
+					     struct airoha_flow_table_entry *e)
+{
+	struct airoha_foe_entry *hwe_p, hwe = {};
+
+	spin_lock_bh(&ppe_lock);
+
+	if (e->type == FLOW_TYPE_L2) {
+		airoha_ppe_foe_flow_l2_entry_update(ppe, e);
+		goto unlock;
+	}
+
+	if (e->hash == 0xffff)
+		goto unlock;
+
+	hwe_p = airoha_ppe_foe_get_entry_locked(ppe, e->hash);
+	if (!hwe_p)
+		goto unlock;
+
+	memcpy(&hwe, hwe_p, sizeof(*hwe_p));
+	if (!airoha_ppe_foe_compare_entry(e, &hwe)) {
+		e->hash = 0xffff;
+		goto unlock;
+	}
+
+	e->data.ib1 = hwe.ib1;
+unlock:
+	spin_unlock_bh(&ppe_lock);
+}
+
+static int airoha_ppe_entry_idle_time(struct airoha_ppe *ppe,
+				      struct airoha_flow_table_entry *e)
+{
+	airoha_ppe_foe_flow_entry_update(ppe, e);
+
+	return airoha_ppe_get_entry_idle_time(ppe, e->data.ib1);
+}
+
+static int airoha_ppe_flow_offload_replace(struct airoha_eth *eth,
+					   struct flow_cls_offload *f)
+{
+	struct flow_rule *rule = flow_cls_offload_flow_rule(f);
+	struct airoha_flow_table_entry *e;
+	struct airoha_flow_data data = {};
+	struct net_device *odev = NULL;
+	struct flow_action_entry *act;
+	struct airoha_foe_entry hwe;
+	int err, i, offload_type;
+	u16 addr_type = 0;
+	u8 l4proto = 0;
+
+	if (rhashtable_lookup(&eth->flow_table, &f->cookie,
+			      airoha_flow_table_params))
+		return -EEXIST;
+
+	if (!flow_rule_match_key(rule, FLOW_DISSECTOR_KEY_META))
+		return -EOPNOTSUPP;
+
+	if (flow_rule_match_key(rule, FLOW_DISSECTOR_KEY_CONTROL)) {
+		struct flow_match_control match;
+
+		flow_rule_match_control(rule, &match);
+		addr_type = match.key->addr_type;
+		if (flow_rule_has_control_flags(match.mask->flags,
+						f->common.extack))
+			return -EOPNOTSUPP;
+	} else {
+		return -EOPNOTSUPP;
+	}
+
+	if (flow_rule_match_key(rule, FLOW_DISSECTOR_KEY_BASIC)) {
+		struct flow_match_basic match;
+
+		flow_rule_match_basic(rule, &match);
+		l4proto = match.key->ip_proto;
+	} else {
+		return -EOPNOTSUPP;
+	}
+
+	switch (addr_type) {
+	case 0:
+		offload_type = PPE_PKT_TYPE_BRIDGE;
+		if (flow_rule_match_key(rule, FLOW_DISSECTOR_KEY_ETH_ADDRS)) {
+			struct flow_match_eth_addrs match;
+
+			flow_rule_match_eth_addrs(rule, &match);
+			memcpy(data.eth.h_dest, match.key->dst, ETH_ALEN);
+			memcpy(data.eth.h_source, match.key->src, ETH_ALEN);
+		} else {
+			return -EOPNOTSUPP;
+		}
+		break;
+	case FLOW_DISSECTOR_KEY_IPV4_ADDRS:
+		offload_type = PPE_PKT_TYPE_IPV4_HNAPT;
+		break;
+	case FLOW_DISSECTOR_KEY_IPV6_ADDRS:
+		offload_type = PPE_PKT_TYPE_IPV6_ROUTE_5T;
+		break;
+	default:
+		return -EOPNOTSUPP;
+	}
+
+	flow_action_for_each(i, act, &rule->action) {
+		switch (act->id) {
+		case FLOW_ACTION_MANGLE:
+			if (offload_type == PPE_PKT_TYPE_BRIDGE)
+				return -EOPNOTSUPP;
+
+			if (act->mangle.htype == FLOW_ACT_MANGLE_HDR_TYPE_ETH)
+				airoha_ppe_flow_mangle_eth(act, &data.eth);
+			break;
+		case FLOW_ACTION_REDIRECT:
+			odev = act->dev;
+			break;
+		case FLOW_ACTION_CSUM:
+			break;
+		case FLOW_ACTION_VLAN_PUSH:
+			if (data.vlan.num == 2 ||
+			    act->vlan.proto != htons(ETH_P_8021Q))
+				return -EOPNOTSUPP;
+
+			data.vlan.hdr[data.vlan.num].id = act->vlan.vid;
+			data.vlan.hdr[data.vlan.num].proto = act->vlan.proto;
+			data.vlan.num++;
+			break;
+		case FLOW_ACTION_VLAN_POP:
+			break;
+		case FLOW_ACTION_PPPOE_PUSH:
+			if (data.pppoe.num == 1 || data.vlan.num == 2)
+				return -EOPNOTSUPP;
+
+			data.pppoe.sid = act->pppoe.sid;
+			data.pppoe.num++;
+			break;
+		default:
+			return -EOPNOTSUPP;
+		}
+	}
+
+	if (!is_valid_ether_addr(data.eth.h_source) ||
+	    !is_valid_ether_addr(data.eth.h_dest))
+		return -EINVAL;
+
+	err = airoha_ppe_foe_entry_prepare(eth, &hwe, odev, offload_type,
+					   &data, l4proto);
+	if (err)
+		return err;
+
+	if (flow_rule_match_key(rule, FLOW_DISSECTOR_KEY_PORTS)) {
+		struct flow_match_ports ports;
+
+		if (offload_type == PPE_PKT_TYPE_BRIDGE)
+			return -EOPNOTSUPP;
+
+		flow_rule_match_ports(rule, &ports);
+		data.src_port = ports.key->src;
+		data.dst_port = ports.key->dst;
+	} else if (offload_type != PPE_PKT_TYPE_BRIDGE) {
+		return -EOPNOTSUPP;
+	}
+
+	if (addr_type == FLOW_DISSECTOR_KEY_IPV4_ADDRS) {
+		struct flow_match_ipv4_addrs addrs;
+
+		flow_rule_match_ipv4_addrs(rule, &addrs);
+		data.v4.src_addr = addrs.key->src;
+		data.v4.dst_addr = addrs.key->dst;
+		airoha_ppe_foe_entry_set_ipv4_tuple(&hwe, &data, false);
+	}
+
+	if (addr_type == FLOW_DISSECTOR_KEY_IPV6_ADDRS) {
+		struct flow_match_ipv6_addrs addrs;
+
+		flow_rule_match_ipv6_addrs(rule, &addrs);
+
+		data.v6.src_addr = addrs.key->src;
+		data.v6.dst_addr = addrs.key->dst;
+		airoha_ppe_foe_entry_set_ipv6_tuple(&hwe, &data);
+	}
+
+	flow_action_for_each(i, act, &rule->action) {
+		if (act->id != FLOW_ACTION_MANGLE)
+			continue;
+
+		if (offload_type == PPE_PKT_TYPE_BRIDGE)
+			return -EOPNOTSUPP;
+
+		switch (act->mangle.htype) {
+		case FLOW_ACT_MANGLE_HDR_TYPE_TCP:
+		case FLOW_ACT_MANGLE_HDR_TYPE_UDP:
+			err = airoha_ppe_flow_mangle_ports(act, &data);
+			break;
+		case FLOW_ACT_MANGLE_HDR_TYPE_IP4:
+			err = airoha_ppe_flow_mangle_ipv4(act, &data);
+			break;
+		case FLOW_ACT_MANGLE_HDR_TYPE_ETH:
+			/* handled earlier */
+			break;
+		default:
+			return -EOPNOTSUPP;
+		}
+
+		if (err)
+			return err;
+	}
+
+	if (addr_type == FLOW_DISSECTOR_KEY_IPV4_ADDRS) {
+		err = airoha_ppe_foe_entry_set_ipv4_tuple(&hwe, &data, true);
+		if (err)
+			return err;
+	}
+
+	e = kzalloc(sizeof(*e), GFP_KERNEL);
+	if (!e)
+		return -ENOMEM;
+
+	e->cookie = f->cookie;
+	memcpy(&e->data, &hwe, sizeof(e->data));
+
+	err = airoha_ppe_foe_flow_commit_entry(eth->ppe, e);
+	if (err)
+		goto free_entry;
+
+	err = rhashtable_insert_fast(&eth->flow_table, &e->node,
+				     airoha_flow_table_params);
+	if (err < 0)
+		goto remove_foe_entry;
+
+	return 0;
+
+remove_foe_entry:
+	airoha_ppe_foe_flow_remove_entry(eth->ppe, e);
+free_entry:
+	kfree(e);
+
+	return err;
+}
+
+static int airoha_ppe_flow_offload_destroy(struct airoha_eth *eth,
+					   struct flow_cls_offload *f)
+{
+	struct airoha_flow_table_entry *e;
+
+	e = rhashtable_lookup(&eth->flow_table, &f->cookie,
+			      airoha_flow_table_params);
+	if (!e)
+		return -ENOENT;
+
+	airoha_ppe_foe_flow_remove_entry(eth->ppe, e);
+	rhashtable_remove_fast(&eth->flow_table, &e->node,
+			       airoha_flow_table_params);
+	kfree(e);
+
+	return 0;
+}
+
+void airoha_ppe_foe_entry_get_stats(struct airoha_ppe *ppe, u32 hash,
+				    struct airoha_foe_stats64 *stats)
+{
+	u32 index = airoha_ppe_foe_get_flow_stats_index(ppe, hash);
+	struct airoha_eth *eth = ppe->eth;
+	struct airoha_npu *npu;
+
+	if (index >= PPE_STATS_NUM_ENTRIES)
+		return;
+
+	rcu_read_lock();
+
+	npu = rcu_dereference(eth->npu);
+	if (npu) {
+		u64 packets = ppe->foe_stats[index].packets;
+		u64 bytes = ppe->foe_stats[index].bytes;
+		struct airoha_foe_stats npu_stats;
+
+		memcpy_fromio(&npu_stats, &npu->stats[index],
+			      sizeof(*npu->stats));
+		stats->packets = packets << 32 | npu_stats.packets;
+		stats->bytes = bytes << 32 | npu_stats.bytes;
+	}
+
+	rcu_read_unlock();
+}
+
+static int airoha_ppe_flow_offload_stats(struct airoha_eth *eth,
+					 struct flow_cls_offload *f)
+{
+	struct airoha_flow_table_entry *e;
+	u32 idle;
+
+	e = rhashtable_lookup(&eth->flow_table, &f->cookie,
+			      airoha_flow_table_params);
+	if (!e)
+		return -ENOENT;
+
+	idle = airoha_ppe_entry_idle_time(eth->ppe, e);
+	f->stats.lastused = jiffies - idle * HZ;
+
+	if (e->hash != 0xffff) {
+		struct airoha_foe_stats64 stats = {};
+
+		airoha_ppe_foe_entry_get_stats(eth->ppe, e->hash, &stats);
+		f->stats.pkts += (stats.packets - e->stats.packets);
+		f->stats.bytes += (stats.bytes - e->stats.bytes);
+		e->stats = stats;
+	}
+
+	return 0;
+}
+
+static int airoha_ppe_flow_offload_cmd(struct airoha_eth *eth,
+				       struct flow_cls_offload *f)
+{
+	switch (f->command) {
+	case FLOW_CLS_REPLACE:
+		return airoha_ppe_flow_offload_replace(eth, f);
+	case FLOW_CLS_DESTROY:
+		return airoha_ppe_flow_offload_destroy(eth, f);
+	case FLOW_CLS_STATS:
+		return airoha_ppe_flow_offload_stats(eth, f);
+	default:
+		break;
+	}
+
+	return -EOPNOTSUPP;
+}
+
+static int airoha_ppe_flush_sram_entries(struct airoha_ppe *ppe,
+					 struct airoha_npu *npu)
+{
+	int i, sram_num_entries = PPE_SRAM_NUM_ENTRIES;
+	struct airoha_foe_entry *hwe = ppe->foe;
+
+	if (airoha_ppe2_is_enabled(ppe->eth))
+		sram_num_entries = sram_num_entries / 2;
+
+	for (i = 0; i < sram_num_entries; i++)
+		memset(&hwe[i], 0, sizeof(*hwe));
+
+	return npu->ops.ppe_flush_sram_entries(npu, ppe->foe_dma,
+					       PPE_SRAM_NUM_ENTRIES);
+}
+
+static struct airoha_npu *airoha_ppe_npu_get(struct airoha_eth *eth)
+{
+	struct airoha_npu *npu = airoha_npu_get(eth->dev);
+
+	if (IS_ERR(npu)) {
+		request_module("airoha-npu");
+		npu = airoha_npu_get(eth->dev);
+	}
+
+	return npu;
+}
+
+static int airoha_ppe_offload_setup(struct airoha_eth *eth)
+{
+	struct airoha_npu *npu = airoha_ppe_npu_get(eth);
+	struct airoha_ppe *ppe = eth->ppe;
+	int err;
+
+	if (IS_ERR(npu))
+		return PTR_ERR(npu);
+
+	err = npu->ops.ppe_init(npu);
+	if (err)
+		goto error_npu_put;
+
+	if (PPE_STATS_NUM_ENTRIES) {
+		err = npu->ops.ppe_init_stats(npu, ppe->foe_stats_dma,
+					      PPE_STATS_NUM_ENTRIES);
+		if (err)
+			goto error_npu_put;
+	}
+
+	airoha_ppe_hw_init(ppe);
+	err = airoha_ppe_flush_sram_entries(ppe, npu);
+	if (err)
+		goto error_npu_put;
+
+	airoha_ppe_foe_flow_stats_reset(ppe, npu);
+
+	rcu_assign_pointer(eth->npu, npu);
+	synchronize_rcu();
+
+	return 0;
+
+error_npu_put:
+	airoha_npu_put(npu);
+
+	return err;
+}
+
+int airoha_ppe_setup_tc_block_cb(struct airoha_ppe_dev *dev, void *type_data)
+{
+	struct airoha_ppe *ppe = dev->priv;
+	struct airoha_eth *eth = ppe->eth;
+	int err = 0;
+
+	mutex_lock(&flow_offload_mutex);
+
+	if (!eth->npu)
+		err = airoha_ppe_offload_setup(eth);
+	if (!err)
+		err = airoha_ppe_flow_offload_cmd(eth, type_data);
+
+	mutex_unlock(&flow_offload_mutex);
+
+	return err;
+}
+
+void airoha_ppe_check_skb(struct airoha_ppe_dev *dev, struct sk_buff *skb,
+			  u16 hash, bool rx_wlan)
+{
+	struct airoha_ppe *ppe = dev->priv;
+	u16 now, diff;
+
+	if (hash > PPE_HASH_MASK)
+		return;
+
+	now = (u16)jiffies;
+	diff = now - ppe->foe_check_time[hash];
+	if (diff < HZ / 10)
+		return;
+
+	ppe->foe_check_time[hash] = now;
+	airoha_ppe_foe_insert_entry(ppe, skb, hash, rx_wlan);
+}
+
+void airoha_ppe_init_upd_mem(struct airoha_gdm_port *port)
+{
+	struct airoha_eth *eth = port->qdma->eth;
+	struct net_device *dev = port->dev;
+	const u8 *addr = dev->dev_addr;
+	u32 val;
+
+	val = (addr[2] << 24) | (addr[3] << 16) | (addr[4] << 8) | addr[5];
+	airoha_fe_wr(eth, REG_UPDMEM_DATA(0), val);
+	airoha_fe_wr(eth, REG_UPDMEM_CTRL(0),
+		     FIELD_PREP(PPE_UPDMEM_ADDR_MASK, port->id) |
+		     PPE_UPDMEM_WR_MASK | PPE_UPDMEM_REQ_MASK);
+
+	val = (addr[0] << 8) | addr[1];
+	airoha_fe_wr(eth, REG_UPDMEM_DATA(0), val);
+	airoha_fe_wr(eth, REG_UPDMEM_CTRL(0),
+		     FIELD_PREP(PPE_UPDMEM_ADDR_MASK, port->id) |
+		     FIELD_PREP(PPE_UPDMEM_OFFSET_MASK, 1) |
+		     PPE_UPDMEM_WR_MASK | PPE_UPDMEM_REQ_MASK);
+}
+
+struct airoha_ppe_dev *airoha_ppe_get_dev(struct device *dev)
+{
+	struct platform_device *pdev;
+	struct device_node *np;
+	struct airoha_eth *eth;
+
+	np = of_parse_phandle(dev->of_node, "airoha,eth", 0);
+	if (!np)
+		return ERR_PTR(-ENODEV);
+
+	pdev = of_find_device_by_node(np);
+	if (!pdev) {
+		dev_err(dev, "cannot find device node %s\n", np->name);
+		of_node_put(np);
+		return ERR_PTR(-ENODEV);
+	}
+	of_node_put(np);
+
+	if (!try_module_get(THIS_MODULE)) {
+		dev_err(dev, "failed to get the device driver module\n");
+		goto error_pdev_put;
+	}
+
+	eth = platform_get_drvdata(pdev);
+	if (!eth)
+		goto error_module_put;
+
+	if (!device_link_add(dev, &pdev->dev, DL_FLAG_AUTOREMOVE_SUPPLIER)) {
+		dev_err(&pdev->dev,
+			"failed to create device link to consumer %s\n",
+			dev_name(dev));
+		goto error_module_put;
+	}
+
+	return &eth->ppe->dev;
+
+error_module_put:
+	module_put(THIS_MODULE);
+error_pdev_put:
+	platform_device_put(pdev);
+
+	return ERR_PTR(-ENODEV);
+}
+EXPORT_SYMBOL_GPL(airoha_ppe_get_dev);
+
+void airoha_ppe_put_dev(struct airoha_ppe_dev *dev)
+{
+	struct airoha_ppe *ppe = dev->priv;
+	struct airoha_eth *eth = ppe->eth;
+
+	module_put(THIS_MODULE);
+	put_device(eth->dev);
+}
+EXPORT_SYMBOL_GPL(airoha_ppe_put_dev);
+
+int airoha_ppe_init(struct airoha_eth *eth)
+{
+	struct airoha_ppe *ppe;
+	int foe_size, err;
+
+	ppe = devm_kzalloc(eth->dev, sizeof(*ppe), GFP_KERNEL);
+	if (!ppe)
+		return -ENOMEM;
+
+	ppe->dev.ops.setup_tc_block_cb = airoha_ppe_setup_tc_block_cb;
+	ppe->dev.ops.check_skb = airoha_ppe_check_skb;
+	ppe->dev.priv = ppe;
+
+	foe_size = PPE_NUM_ENTRIES * sizeof(struct airoha_foe_entry);
+	ppe->foe = dmam_alloc_coherent(eth->dev, foe_size, &ppe->foe_dma,
+				       GFP_KERNEL);
+	if (!ppe->foe)
+		return -ENOMEM;
+
+	ppe->eth = eth;
+	eth->ppe = ppe;
+
+	ppe->foe_flow = devm_kzalloc(eth->dev,
+				     PPE_NUM_ENTRIES * sizeof(*ppe->foe_flow),
+				     GFP_KERNEL);
+	if (!ppe->foe_flow)
+		return -ENOMEM;
+
+	foe_size = PPE_STATS_NUM_ENTRIES * sizeof(*ppe->foe_stats);
+	if (foe_size) {
+		ppe->foe_stats = dmam_alloc_coherent(eth->dev, foe_size,
+						     &ppe->foe_stats_dma,
+						     GFP_KERNEL);
+		if (!ppe->foe_stats)
+			return -ENOMEM;
+	}
+
+	err = rhashtable_init(&eth->flow_table, &airoha_flow_table_params);
+	if (err)
+		return err;
+
+	err = rhashtable_init(&ppe->l2_flows, &airoha_l2_flow_table_params);
+	if (err)
+		goto error_flow_table_destroy;
+
+	err = airoha_ppe_debugfs_init(ppe);
+	if (err)
+		goto error_l2_flow_table_destroy;
+
+	return 0;
+
+error_l2_flow_table_destroy:
+	rhashtable_destroy(&ppe->l2_flows);
+error_flow_table_destroy:
+	rhashtable_destroy(&eth->flow_table);
+
+	return err;
+}
+
+void airoha_ppe_deinit(struct airoha_eth *eth)
+{
+	struct airoha_npu *npu;
+
+	rcu_read_lock();
+	npu = rcu_dereference(eth->npu);
+	if (npu) {
+		npu->ops.ppe_deinit(npu);
+		airoha_npu_put(npu);
+	}
+	rcu_read_unlock();
+
+	rhashtable_destroy(&eth->ppe->l2_flows);
+	rhashtable_destroy(&eth->flow_table);
+	debugfs_remove(eth->ppe->debugfs_dir);
+}
diff --git a/drivers/net/ethernet/airoha/airoha_ppe_debugfs.c b/drivers/net/ethernet/airoha/airoha_ppe_debugfs.c
new file mode 100644
index 000000000000..05a756233f6a
--- /dev/null
+++ b/drivers/net/ethernet/airoha/airoha_ppe_debugfs.c
@@ -0,0 +1,186 @@
+// SPDX-License-Identifier: GPL-2.0-only
+/*
+ * Copyright (c) 2025 AIROHA Inc
+ * Author: Lorenzo Bianconi <lorenzo@kernel.org>
+ */
+
+#include "airoha_eth.h"
+
+static void airoha_debugfs_ppe_print_tuple(struct seq_file *m,
+					   void *src_addr, void *dest_addr,
+					   u16 *src_port, u16 *dest_port,
+					   bool ipv6)
+{
+	__be32 n_addr[IPV6_ADDR_WORDS];
+
+	if (ipv6) {
+		ipv6_addr_cpu_to_be32(n_addr, src_addr);
+		seq_printf(m, "%pI6", n_addr);
+	} else {
+		seq_printf(m, "%pI4h", src_addr);
+	}
+	if (src_port)
+		seq_printf(m, ":%d", *src_port);
+
+	seq_puts(m, "->");
+
+	if (ipv6) {
+		ipv6_addr_cpu_to_be32(n_addr, dest_addr);
+		seq_printf(m, "%pI6", n_addr);
+	} else {
+		seq_printf(m, "%pI4h", dest_addr);
+	}
+	if (dest_port)
+		seq_printf(m, ":%d", *dest_port);
+}
+
+static int airoha_ppe_debugfs_foe_show(struct seq_file *m, void *private,
+				       bool bind)
+{
+	static const char *const ppe_type_str[] = {
+		[PPE_PKT_TYPE_IPV4_HNAPT] = "IPv4 5T",
+		[PPE_PKT_TYPE_IPV4_ROUTE] = "IPv4 3T",
+		[PPE_PKT_TYPE_BRIDGE] = "L2B",
+		[PPE_PKT_TYPE_IPV4_DSLITE] = "DS-LITE",
+		[PPE_PKT_TYPE_IPV6_ROUTE_3T] = "IPv6 3T",
+		[PPE_PKT_TYPE_IPV6_ROUTE_5T] = "IPv6 5T",
+		[PPE_PKT_TYPE_IPV6_6RD] = "6RD",
+	};
+	static const char *const ppe_state_str[] = {
+		[AIROHA_FOE_STATE_INVALID] = "INV",
+		[AIROHA_FOE_STATE_UNBIND] = "UNB",
+		[AIROHA_FOE_STATE_BIND] = "BND",
+		[AIROHA_FOE_STATE_FIN] = "FIN",
+	};
+	struct airoha_ppe *ppe = m->private;
+	int i;
+
+	for (i = 0; i < PPE_NUM_ENTRIES; i++) {
+		const char *state_str, *type_str = "UNKNOWN";
+		void *src_addr = NULL, *dest_addr = NULL;
+		u16 *src_port = NULL, *dest_port = NULL;
+		struct airoha_foe_mac_info_common *l2;
+		unsigned char h_source[ETH_ALEN] = {};
+		struct airoha_foe_stats64 stats = {};
+		unsigned char h_dest[ETH_ALEN];
+		struct airoha_foe_entry *hwe;
+		u32 type, state, ib2, data;
+		bool ipv6 = false;
+
+		hwe = airoha_ppe_foe_get_entry(ppe, i);
+		if (!hwe)
+			continue;
+
+		state = FIELD_GET(AIROHA_FOE_IB1_BIND_STATE, hwe->ib1);
+		if (!state)
+			continue;
+
+		if (bind && state != AIROHA_FOE_STATE_BIND)
+			continue;
+
+		state_str = ppe_state_str[state % ARRAY_SIZE(ppe_state_str)];
+		type = FIELD_GET(AIROHA_FOE_IB1_BIND_PACKET_TYPE, hwe->ib1);
+		if (type < ARRAY_SIZE(ppe_type_str) && ppe_type_str[type])
+			type_str = ppe_type_str[type];
+
+		seq_printf(m, "%05x %s %7s", i, state_str, type_str);
+
+		switch (type) {
+		case PPE_PKT_TYPE_IPV4_HNAPT:
+		case PPE_PKT_TYPE_IPV4_DSLITE:
+			src_port = &hwe->ipv4.orig_tuple.src_port;
+			dest_port = &hwe->ipv4.orig_tuple.dest_port;
+			fallthrough;
+		case PPE_PKT_TYPE_IPV4_ROUTE:
+			src_addr = &hwe->ipv4.orig_tuple.src_ip;
+			dest_addr = &hwe->ipv4.orig_tuple.dest_ip;
+			break;
+		case PPE_PKT_TYPE_IPV6_ROUTE_5T:
+			src_port = &hwe->ipv6.src_port;
+			dest_port = &hwe->ipv6.dest_port;
+			fallthrough;
+		case PPE_PKT_TYPE_IPV6_ROUTE_3T:
+		case PPE_PKT_TYPE_IPV6_6RD:
+			src_addr = &hwe->ipv6.src_ip;
+			dest_addr = &hwe->ipv6.dest_ip;
+			ipv6 = true;
+			break;
+		default:
+			break;
+		}
+
+		if (src_addr && dest_addr) {
+			seq_puts(m, " orig=");
+			airoha_debugfs_ppe_print_tuple(m, src_addr, dest_addr,
+						       src_port, dest_port, ipv6);
+		}
+
+		switch (type) {
+		case PPE_PKT_TYPE_IPV4_HNAPT:
+		case PPE_PKT_TYPE_IPV4_DSLITE:
+			src_port = &hwe->ipv4.new_tuple.src_port;
+			dest_port = &hwe->ipv4.new_tuple.dest_port;
+			fallthrough;
+		case PPE_PKT_TYPE_IPV4_ROUTE:
+			src_addr = &hwe->ipv4.new_tuple.src_ip;
+			dest_addr = &hwe->ipv4.new_tuple.dest_ip;
+			seq_puts(m, " new=");
+			airoha_debugfs_ppe_print_tuple(m, src_addr, dest_addr,
+						       src_port, dest_port,
+						       ipv6);
+			break;
+		default:
+			break;
+		}
+
+		if (type >= PPE_PKT_TYPE_IPV6_ROUTE_3T) {
+			data = hwe->ipv6.data;
+			ib2 = hwe->ipv6.ib2;
+			l2 = &hwe->ipv6.l2;
+		} else {
+			data = hwe->ipv4.data;
+			ib2 = hwe->ipv4.ib2;
+			l2 = &hwe->ipv4.l2.common;
+			*((__be16 *)&h_source[4]) =
+				cpu_to_be16(hwe->ipv4.l2.src_mac_lo);
+		}
+
+		airoha_ppe_foe_entry_get_stats(ppe, i, &stats);
+
+		*((__be32 *)h_dest) = cpu_to_be32(l2->dest_mac_hi);
+		*((__be16 *)&h_dest[4]) = cpu_to_be16(l2->dest_mac_lo);
+		*((__be32 *)h_source) = cpu_to_be32(l2->src_mac_hi);
+
+		seq_printf(m, " eth=%pM->%pM etype=%04x data=%08x"
+			      " vlan=%d,%d ib1=%08x ib2=%08x"
+			      " packets=%llu bytes=%llu\n",
+			   h_source, h_dest, l2->etype, data,
+			   l2->vlan1, l2->vlan2, hwe->ib1, ib2,
+			   stats.packets, stats.bytes);
+	}
+
+	return 0;
+}
+
+static int airoha_ppe_debugfs_foe_all_show(struct seq_file *m, void *private)
+{
+	return airoha_ppe_debugfs_foe_show(m, private, false);
+}
+DEFINE_SHOW_ATTRIBUTE(airoha_ppe_debugfs_foe_all);
+
+static int airoha_ppe_debugfs_foe_bind_show(struct seq_file *m, void *private)
+{
+	return airoha_ppe_debugfs_foe_show(m, private, true);
+}
+DEFINE_SHOW_ATTRIBUTE(airoha_ppe_debugfs_foe_bind);
+
+int airoha_ppe_debugfs_init(struct airoha_ppe *ppe)
+{
+	ppe->debugfs_dir = debugfs_create_dir("ppe", NULL);
+	debugfs_create_file("entries", 0444, ppe->debugfs_dir, ppe,
+			    &airoha_ppe_debugfs_foe_all_fops);
+	debugfs_create_file("bind", 0444, ppe->debugfs_dir, ppe,
+			    &airoha_ppe_debugfs_foe_bind_fops);
+
+	return 0;
+}
diff --git a/drivers/net/ethernet/airoha/airoha_regs.h b/drivers/net/ethernet/airoha/airoha_regs.h
new file mode 100644
index 000000000000..69c5a143db8c
--- /dev/null
+++ b/drivers/net/ethernet/airoha/airoha_regs.h
@@ -0,0 +1,936 @@
+/* SPDX-License-Identifier: GPL-2.0-only */
+/*
+ * Copyright (c) 2024 AIROHA Inc
+ * Author: Lorenzo Bianconi <lorenzo@kernel.org>
+ */
+
+#ifndef AIROHA_REGS_H
+#define AIROHA_REGS_H
+
+#include <linux/types.h>
+
+/* FE */
+#define PSE_BASE			0x0100
+#define CSR_IFC_BASE			0x0200
+#define CDM1_BASE			0x0400
+#define GDM1_BASE			0x0500
+#define PPE1_BASE			0x0c00
+#define PPE2_BASE			0x1c00
+
+#define CDM2_BASE			0x1400
+#define GDM2_BASE			0x1500
+
+#define GDM3_BASE			0x1100
+#define GDM4_BASE			0x2500
+
+#define GDM_BASE(_n)			\
+	((_n) == 4 ? GDM4_BASE :	\
+	 (_n) == 3 ? GDM3_BASE :	\
+	 (_n) == 2 ? GDM2_BASE : GDM1_BASE)
+
+#define REG_FE_DMA_GLO_CFG		0x0000
+#define FE_DMA_GLO_L2_SPACE_MASK	GENMASK(7, 4)
+#define FE_DMA_GLO_PG_SZ_MASK		BIT(3)
+
+#define REG_FE_RST_GLO_CFG		0x0004
+#define FE_RST_GDM4_MBI_ARB_MASK	BIT(3)
+#define FE_RST_GDM3_MBI_ARB_MASK	BIT(2)
+#define FE_RST_CORE_MASK		BIT(0)
+
+#define REG_FE_FOE_TS			0x0010
+
+#define REG_FE_WAN_PORT			0x0024
+#define WAN1_EN_MASK			BIT(16)
+#define WAN1_MASK			GENMASK(12, 8)
+#define WAN0_MASK			GENMASK(4, 0)
+
+#define REG_FE_WAN_MAC_H		0x0030
+#define REG_FE_LAN_MAC_H		0x0040
+
+#define REG_FE_MAC_LMIN(_n)		((_n) + 0x04)
+#define REG_FE_MAC_LMAX(_n)		((_n) + 0x08)
+
+#define REG_FE_CDM1_OQ_MAP0		0x0050
+#define REG_FE_CDM1_OQ_MAP1		0x0054
+#define REG_FE_CDM1_OQ_MAP2		0x0058
+#define REG_FE_CDM1_OQ_MAP3		0x005c
+
+#define REG_FE_PCE_CFG			0x0070
+#define PCE_DPI_EN_MASK			BIT(2)
+#define PCE_KA_EN_MASK			BIT(1)
+#define PCE_MC_EN_MASK			BIT(0)
+
+#define REG_FE_PSE_QUEUE_CFG_WR		0x0080
+#define PSE_CFG_PORT_ID_MASK		GENMASK(27, 24)
+#define PSE_CFG_QUEUE_ID_MASK		GENMASK(20, 16)
+#define PSE_CFG_WR_EN_MASK		BIT(8)
+#define PSE_CFG_OQRSV_SEL_MASK		BIT(0)
+
+#define REG_FE_PSE_QUEUE_CFG_VAL	0x0084
+#define PSE_CFG_OQ_RSV_MASK		GENMASK(13, 0)
+
+#define PSE_FQ_CFG			0x008c
+#define PSE_FQ_LIMIT_MASK		GENMASK(14, 0)
+
+#define REG_FE_PSE_BUF_SET		0x0090
+#define PSE_SHARE_USED_LTHD_MASK	GENMASK(31, 16)
+#define PSE_ALLRSV_MASK			GENMASK(14, 0)
+
+#define REG_PSE_SHARE_USED_THD		0x0094
+#define PSE_SHARE_USED_MTHD_MASK	GENMASK(31, 16)
+#define PSE_SHARE_USED_HTHD_MASK	GENMASK(15, 0)
+
+#define REG_GDM_MISC_CFG		0x0148
+#define GDM2_RDM_ACK_WAIT_PREF_MASK	BIT(9)
+#define GDM2_CHN_VLD_MODE_MASK		BIT(5)
+
+#define REG_FE_CSR_IFC_CFG		CSR_IFC_BASE
+#define FE_IFC_EN_MASK			BIT(0)
+
+#define REG_FE_VIP_PORT_EN		0x01f0
+#define REG_FE_IFC_PORT_EN		0x01f4
+
+#define REG_PSE_IQ_REV1			(PSE_BASE + 0x08)
+#define PSE_IQ_RES1_P2_MASK		GENMASK(23, 16)
+
+#define REG_PSE_IQ_REV2			(PSE_BASE + 0x0c)
+#define PSE_IQ_RES2_P5_MASK		GENMASK(15, 8)
+#define PSE_IQ_RES2_P4_MASK		GENMASK(7, 0)
+
+#define REG_FE_VIP_EN(_n)		(0x0300 + ((_n) << 3))
+#define PATN_FCPU_EN_MASK		BIT(7)
+#define PATN_SWP_EN_MASK		BIT(6)
+#define PATN_DP_EN_MASK			BIT(5)
+#define PATN_SP_EN_MASK			BIT(4)
+#define PATN_TYPE_MASK			GENMASK(3, 1)
+#define PATN_EN_MASK			BIT(0)
+
+#define REG_FE_VIP_PATN(_n)		(0x0304 + ((_n) << 3))
+#define PATN_DP_MASK			GENMASK(31, 16)
+#define PATN_SP_MASK			GENMASK(15, 0)
+
+#define REG_CDM1_VLAN_CTRL		CDM1_BASE
+#define CDM1_VLAN_MASK			GENMASK(31, 16)
+
+#define REG_CDM1_FWD_CFG		(CDM1_BASE + 0x08)
+#define CDM1_VIP_QSEL_MASK		GENMASK(24, 20)
+
+#define REG_CDM1_CRSN_QSEL(_n)		(CDM1_BASE + 0x10 + ((_n) << 2))
+#define CDM1_CRSN_QSEL_REASON_MASK(_n)	\
+	GENMASK(4 + (((_n) % 4) << 3),	(((_n) % 4) << 3))
+
+#define REG_CDM2_FWD_CFG		(CDM2_BASE + 0x08)
+#define CDM2_OAM_QSEL_MASK		GENMASK(31, 27)
+#define CDM2_VIP_QSEL_MASK		GENMASK(24, 20)
+
+#define REG_CDM2_CRSN_QSEL(_n)		(CDM2_BASE + 0x10 + ((_n) << 2))
+#define CDM2_CRSN_QSEL_REASON_MASK(_n)	\
+	GENMASK(4 + (((_n) % 4) << 3),	(((_n) % 4) << 3))
+
+#define REG_GDM_FWD_CFG(_n)		GDM_BASE(_n)
+#define GDM_DROP_CRC_ERR		BIT(23)
+#define GDM_IP4_CKSUM			BIT(22)
+#define GDM_TCP_CKSUM			BIT(21)
+#define GDM_UDP_CKSUM			BIT(20)
+#define GDM_STRIP_CRC			BIT(16)
+#define GDM_UCFQ_MASK			GENMASK(15, 12)
+#define GDM_BCFQ_MASK			GENMASK(11, 8)
+#define GDM_MCFQ_MASK			GENMASK(7, 4)
+#define GDM_OCFQ_MASK			GENMASK(3, 0)
+
+#define REG_GDM_INGRESS_CFG(_n)		(GDM_BASE(_n) + 0x10)
+#define GDM_INGRESS_FC_EN_MASK		BIT(1)
+#define GDM_STAG_EN_MASK		BIT(0)
+
+#define REG_GDM_LEN_CFG(_n)		(GDM_BASE(_n) + 0x14)
+#define GDM_SHORT_LEN_MASK		GENMASK(13, 0)
+#define GDM_LONG_LEN_MASK		GENMASK(29, 16)
+
+#define REG_GDM_LPBK_CFG(_n)		(GDM_BASE(_n) + 0x1c)
+#define LPBK_GAP_MASK			GENMASK(31, 24)
+#define LPBK_LEN_MASK			GENMASK(23, 10)
+#define LPBK_CHAN_MASK			GENMASK(8, 4)
+#define LPBK_MODE_MASK			GENMASK(3, 1)
+#define LBK_GAP_MODE_MASK		BIT(3)
+#define LBK_LEN_MODE_MASK		BIT(2)
+#define LBK_CHAN_MODE_MASK		BIT(1)
+#define LPBK_EN_MASK			BIT(0)
+
+#define REG_GDM_TXCHN_EN(_n)		(GDM_BASE(_n) + 0x24)
+#define REG_GDM_RXCHN_EN(_n)		(GDM_BASE(_n) + 0x28)
+
+#define REG_FE_CPORT_CFG		(GDM1_BASE + 0x40)
+#define FE_CPORT_PAD			BIT(26)
+#define FE_CPORT_PORT_XFC_MASK		BIT(25)
+#define FE_CPORT_QUEUE_XFC_MASK		BIT(24)
+
+#define REG_FE_GDM_MIB_CLEAR(_n)	(GDM_BASE(_n) + 0xf0)
+#define FE_GDM_MIB_RX_CLEAR_MASK	BIT(1)
+#define FE_GDM_MIB_TX_CLEAR_MASK	BIT(0)
+
+#define REG_FE_GDM1_MIB_CFG		(GDM1_BASE + 0xf4)
+#define FE_STRICT_RFC2819_MODE_MASK	BIT(31)
+#define FE_GDM1_TX_MIB_SPLIT_EN_MASK	BIT(17)
+#define FE_GDM1_RX_MIB_SPLIT_EN_MASK	BIT(16)
+#define FE_TX_MIB_ID_MASK		GENMASK(15, 8)
+#define FE_RX_MIB_ID_MASK		GENMASK(7, 0)
+
+#define REG_FE_GDM_TX_OK_PKT_CNT_L(_n)		(GDM_BASE(_n) + 0x104)
+#define REG_FE_GDM_TX_OK_BYTE_CNT_L(_n)		(GDM_BASE(_n) + 0x10c)
+#define REG_FE_GDM_TX_ETH_PKT_CNT_L(_n)		(GDM_BASE(_n) + 0x110)
+#define REG_FE_GDM_TX_ETH_BYTE_CNT_L(_n)	(GDM_BASE(_n) + 0x114)
+#define REG_FE_GDM_TX_ETH_DROP_CNT(_n)		(GDM_BASE(_n) + 0x118)
+#define REG_FE_GDM_TX_ETH_BC_CNT(_n)		(GDM_BASE(_n) + 0x11c)
+#define REG_FE_GDM_TX_ETH_MC_CNT(_n)		(GDM_BASE(_n) + 0x120)
+#define REG_FE_GDM_TX_ETH_RUNT_CNT(_n)		(GDM_BASE(_n) + 0x124)
+#define REG_FE_GDM_TX_ETH_LONG_CNT(_n)		(GDM_BASE(_n) + 0x128)
+#define REG_FE_GDM_TX_ETH_E64_CNT_L(_n)		(GDM_BASE(_n) + 0x12c)
+#define REG_FE_GDM_TX_ETH_L64_CNT_L(_n)		(GDM_BASE(_n) + 0x130)
+#define REG_FE_GDM_TX_ETH_L127_CNT_L(_n)	(GDM_BASE(_n) + 0x134)
+#define REG_FE_GDM_TX_ETH_L255_CNT_L(_n)	(GDM_BASE(_n) + 0x138)
+#define REG_FE_GDM_TX_ETH_L511_CNT_L(_n)	(GDM_BASE(_n) + 0x13c)
+#define REG_FE_GDM_TX_ETH_L1023_CNT_L(_n)	(GDM_BASE(_n) + 0x140)
+
+#define REG_FE_GDM_RX_OK_PKT_CNT_L(_n)		(GDM_BASE(_n) + 0x148)
+#define REG_FE_GDM_RX_FC_DROP_CNT(_n)		(GDM_BASE(_n) + 0x14c)
+#define REG_FE_GDM_RX_RC_DROP_CNT(_n)		(GDM_BASE(_n) + 0x150)
+#define REG_FE_GDM_RX_OVERFLOW_DROP_CNT(_n)	(GDM_BASE(_n) + 0x154)
+#define REG_FE_GDM_RX_ERROR_DROP_CNT(_n)	(GDM_BASE(_n) + 0x158)
+#define REG_FE_GDM_RX_OK_BYTE_CNT_L(_n)		(GDM_BASE(_n) + 0x15c)
+#define REG_FE_GDM_RX_ETH_PKT_CNT_L(_n)		(GDM_BASE(_n) + 0x160)
+#define REG_FE_GDM_RX_ETH_BYTE_CNT_L(_n)	(GDM_BASE(_n) + 0x164)
+#define REG_FE_GDM_RX_ETH_DROP_CNT(_n)		(GDM_BASE(_n) + 0x168)
+#define REG_FE_GDM_RX_ETH_BC_CNT(_n)		(GDM_BASE(_n) + 0x16c)
+#define REG_FE_GDM_RX_ETH_MC_CNT(_n)		(GDM_BASE(_n) + 0x170)
+#define REG_FE_GDM_RX_ETH_CRC_ERR_CNT(_n)	(GDM_BASE(_n) + 0x174)
+#define REG_FE_GDM_RX_ETH_FRAG_CNT(_n)		(GDM_BASE(_n) + 0x178)
+#define REG_FE_GDM_RX_ETH_JABBER_CNT(_n)	(GDM_BASE(_n) + 0x17c)
+#define REG_FE_GDM_RX_ETH_RUNT_CNT(_n)		(GDM_BASE(_n) + 0x180)
+#define REG_FE_GDM_RX_ETH_LONG_CNT(_n)		(GDM_BASE(_n) + 0x184)
+#define REG_FE_GDM_RX_ETH_E64_CNT_L(_n)		(GDM_BASE(_n) + 0x188)
+#define REG_FE_GDM_RX_ETH_L64_CNT_L(_n)		(GDM_BASE(_n) + 0x18c)
+#define REG_FE_GDM_RX_ETH_L127_CNT_L(_n)	(GDM_BASE(_n) + 0x190)
+#define REG_FE_GDM_RX_ETH_L255_CNT_L(_n)	(GDM_BASE(_n) + 0x194)
+#define REG_FE_GDM_RX_ETH_L511_CNT_L(_n)	(GDM_BASE(_n) + 0x198)
+#define REG_FE_GDM_RX_ETH_L1023_CNT_L(_n)	(GDM_BASE(_n) + 0x19c)
+
+#define REG_PPE_GLO_CFG(_n)			(((_n) ? PPE2_BASE : PPE1_BASE) + 0x200)
+#define PPE_GLO_CFG_BUSY_MASK			BIT(31)
+#define PPE_GLO_CFG_FLOW_DROP_UPDATE_MASK	BIT(9)
+#define PPE_GLO_CFG_PSE_HASH_OFS_MASK		BIT(6)
+#define PPE_GLO_CFG_PPE_BSWAP_MASK		BIT(5)
+#define PPE_GLO_CFG_TTL_DROP_MASK		BIT(4)
+#define PPE_GLO_CFG_IP4_CS_DROP_MASK		BIT(3)
+#define PPE_GLO_CFG_IP4_L4_CS_DROP_MASK		BIT(2)
+#define PPE_GLO_CFG_EN_MASK			BIT(0)
+
+#define REG_PPE_PPE_FLOW_CFG(_n)		(((_n) ? PPE2_BASE : PPE1_BASE) + 0x204)
+#define PPE_FLOW_CFG_IP6_HASH_GRE_KEY_MASK	BIT(20)
+#define PPE_FLOW_CFG_IP4_HASH_GRE_KEY_MASK	BIT(19)
+#define PPE_FLOW_CFG_IP4_HASH_FLOW_LABEL_MASK	BIT(18)
+#define PPE_FLOW_CFG_IP4_NAT_FRAG_MASK		BIT(17)
+#define PPE_FLOW_CFG_IP_PROTO_BLACKLIST_MASK	BIT(16)
+#define PPE_FLOW_CFG_IP4_DSLITE_MASK		BIT(14)
+#define PPE_FLOW_CFG_IP4_NAPT_MASK		BIT(13)
+#define PPE_FLOW_CFG_IP4_NAT_MASK		BIT(12)
+#define PPE_FLOW_CFG_IP6_6RD_MASK		BIT(10)
+#define PPE_FLOW_CFG_IP6_5T_ROUTE_MASK		BIT(9)
+#define PPE_FLOW_CFG_IP6_3T_ROUTE_MASK		BIT(8)
+#define PPE_FLOW_CFG_IP4_UDP_FRAG_MASK		BIT(7)
+#define PPE_FLOW_CFG_IP4_TCP_FRAG_MASK		BIT(6)
+
+#define REG_PPE_IP_PROTO_CHK(_n)		(((_n) ? PPE2_BASE : PPE1_BASE) + 0x208)
+#define PPE_IP_PROTO_CHK_IPV4_MASK		GENMASK(31, 16)
+#define PPE_IP_PROTO_CHK_IPV6_MASK		GENMASK(15, 0)
+
+#define REG_PPE_TB_CFG(_n)			(((_n) ? PPE2_BASE : PPE1_BASE) + 0x21c)
+#define PPE_SRAM_TB_NUM_ENTRY_MASK		GENMASK(26, 24)
+#define PPE_TB_CFG_KEEPALIVE_MASK		GENMASK(13, 12)
+#define PPE_TB_CFG_AGE_TCP_FIN_MASK		BIT(11)
+#define PPE_TB_CFG_AGE_UDP_MASK			BIT(10)
+#define PPE_TB_CFG_AGE_TCP_MASK			BIT(9)
+#define PPE_TB_CFG_AGE_UNBIND_MASK		BIT(8)
+#define PPE_TB_CFG_AGE_NON_L4_MASK		BIT(7)
+#define PPE_TB_CFG_AGE_PREBIND_MASK		BIT(6)
+#define PPE_TB_CFG_SEARCH_MISS_MASK		GENMASK(5, 4)
+#define PPE_TB_ENTRY_SIZE_MASK			BIT(3)
+#define PPE_DRAM_TB_NUM_ENTRY_MASK		GENMASK(2, 0)
+
+#define REG_PPE_TB_BASE(_n)			(((_n) ? PPE2_BASE : PPE1_BASE) + 0x220)
+
+#define REG_PPE_BIND_RATE(_n)			(((_n) ? PPE2_BASE : PPE1_BASE) + 0x228)
+#define PPE_BIND_RATE_L2B_BIND_MASK		GENMASK(31, 16)
+#define PPE_BIND_RATE_BIND_MASK			GENMASK(15, 0)
+
+#define REG_PPE_BIND_LIMIT0(_n)			(((_n) ? PPE2_BASE : PPE1_BASE) + 0x22c)
+#define PPE_BIND_LIMIT0_HALF_MASK		GENMASK(29, 16)
+#define PPE_BIND_LIMIT0_QUARTER_MASK		GENMASK(13, 0)
+
+#define REG_PPE_BIND_LIMIT1(_n)			(((_n) ? PPE2_BASE : PPE1_BASE) + 0x230)
+#define PPE_BIND_LIMIT1_NON_L4_MASK		GENMASK(23, 16)
+#define PPE_BIND_LIMIT1_FULL_MASK		GENMASK(13, 0)
+
+#define REG_PPE_BND_AGE0(_n)			(((_n) ? PPE2_BASE : PPE1_BASE) + 0x23c)
+#define PPE_BIND_AGE0_DELTA_NON_L4		GENMASK(30, 16)
+#define PPE_BIND_AGE0_DELTA_UDP			GENMASK(14, 0)
+
+#define REG_PPE_UNBIND_AGE(_n)			(((_n) ? PPE2_BASE : PPE1_BASE) + 0x238)
+#define PPE_UNBIND_AGE_MIN_PACKETS_MASK		GENMASK(31, 16)
+#define PPE_UNBIND_AGE_DELTA_MASK		GENMASK(7, 0)
+
+#define REG_PPE_BND_AGE1(_n)			(((_n) ? PPE2_BASE : PPE1_BASE) + 0x240)
+#define PPE_BIND_AGE1_DELTA_TCP_FIN		GENMASK(30, 16)
+#define PPE_BIND_AGE1_DELTA_TCP			GENMASK(14, 0)
+
+#define REG_PPE_HASH_SEED(_n)			(((_n) ? PPE2_BASE : PPE1_BASE) + 0x244)
+#define PPE_HASH_SEED				0x12345678
+
+#define REG_PPE_DFT_CPORT0(_n)			(((_n) ? PPE2_BASE : PPE1_BASE) + 0x248)
+#define DFT_CPORT_MASK(_n)			GENMASK(3 + ((_n) << 2), ((_n) << 2))
+
+#define REG_PPE_DFT_CPORT1(_n)			(((_n) ? PPE2_BASE : PPE1_BASE) + 0x24c)
+
+#define REG_PPE_TB_HASH_CFG(_n)			(((_n) ? PPE2_BASE : PPE1_BASE) + 0x250)
+#define PPE_DRAM_HASH1_MODE_MASK		GENMASK(31, 28)
+#define PPE_DRAM_HASH1_EN_MASK			BIT(24)
+#define PPE_DRAM_HASH0_MODE_MASK		GENMASK(23, 20)
+#define PPE_DRAM_TABLE_EN_MASK			BIT(16)
+#define PPE_SRAM_HASH1_MODE_MASK		GENMASK(15, 12)
+#define PPE_SRAM_HASH1_EN_MASK			BIT(8)
+#define PPE_SRAM_HASH0_MODE_MASK		GENMASK(7, 4)
+#define PPE_SRAM_TABLE_EN_MASK			BIT(0)
+
+#define REG_PPE_MTU_BASE(_n)			(((_n) ? PPE2_BASE : PPE1_BASE) + 0x304)
+#define REG_PPE_MTU(_m, _n)			(REG_PPE_MTU_BASE(_m) + ((_n) << 2))
+#define FP1_EGRESS_MTU_MASK			GENMASK(29, 16)
+#define FP0_EGRESS_MTU_MASK			GENMASK(13, 0)
+
+#define REG_PPE_RAM_CTRL(_n)			(((_n) ? PPE2_BASE : PPE1_BASE) + 0x31c)
+#define PPE_SRAM_CTRL_ACK_MASK			BIT(31)
+#define PPE_SRAM_CTRL_DUAL_SUCESS_MASK		BIT(30)
+#define PPE_SRAM_CTRL_ENTRY_MASK		GENMASK(23, 8)
+#define PPE_SRAM_WR_DUAL_DIRECTION_MASK		BIT(2)
+#define PPE_SRAM_CTRL_WR_MASK			BIT(1)
+#define PPE_SRAM_CTRL_REQ_MASK			BIT(0)
+
+#define REG_PPE_RAM_BASE(_n)			(((_n) ? PPE2_BASE : PPE1_BASE) + 0x320)
+#define REG_PPE_RAM_ENTRY(_m, _n)		(REG_PPE_RAM_BASE(_m) + ((_n) << 2))
+
+#define REG_UPDMEM_CTRL(_n)			(((_n) ? PPE2_BASE : PPE1_BASE) + 0x370)
+#define PPE_UPDMEM_ACK_MASK			BIT(31)
+#define PPE_UPDMEM_ADDR_MASK			GENMASK(11, 8)
+#define PPE_UPDMEM_OFFSET_MASK			GENMASK(7, 4)
+#define PPE_UPDMEM_SEL_MASK			GENMASK(3, 2)
+#define PPE_UPDMEM_WR_MASK			BIT(1)
+#define PPE_UPDMEM_REQ_MASK			BIT(0)
+
+#define REG_UPDMEM_DATA(_n)			(((_n) ? PPE2_BASE : PPE1_BASE) + 0x374)
+
+#define REG_FE_GDM_TX_OK_PKT_CNT_H(_n)		(GDM_BASE(_n) + 0x280)
+#define REG_FE_GDM_TX_OK_BYTE_CNT_H(_n)		(GDM_BASE(_n) + 0x284)
+#define REG_FE_GDM_TX_ETH_PKT_CNT_H(_n)		(GDM_BASE(_n) + 0x288)
+#define REG_FE_GDM_TX_ETH_BYTE_CNT_H(_n)	(GDM_BASE(_n) + 0x28c)
+
+#define REG_FE_GDM_RX_OK_PKT_CNT_H(_n)		(GDM_BASE(_n) + 0x290)
+#define REG_FE_GDM_RX_OK_BYTE_CNT_H(_n)		(GDM_BASE(_n) + 0x294)
+#define REG_FE_GDM_RX_ETH_PKT_CNT_H(_n)		(GDM_BASE(_n) + 0x298)
+#define REG_FE_GDM_RX_ETH_BYTE_CNT_H(_n)	(GDM_BASE(_n) + 0x29c)
+#define REG_FE_GDM_TX_ETH_E64_CNT_H(_n)		(GDM_BASE(_n) + 0x2b8)
+#define REG_FE_GDM_TX_ETH_L64_CNT_H(_n)		(GDM_BASE(_n) + 0x2bc)
+#define REG_FE_GDM_TX_ETH_L127_CNT_H(_n)	(GDM_BASE(_n) + 0x2c0)
+#define REG_FE_GDM_TX_ETH_L255_CNT_H(_n)	(GDM_BASE(_n) + 0x2c4)
+#define REG_FE_GDM_TX_ETH_L511_CNT_H(_n)	(GDM_BASE(_n) + 0x2c8)
+#define REG_FE_GDM_TX_ETH_L1023_CNT_H(_n)	(GDM_BASE(_n) + 0x2cc)
+#define REG_FE_GDM_RX_ETH_E64_CNT_H(_n)		(GDM_BASE(_n) + 0x2e8)
+#define REG_FE_GDM_RX_ETH_L64_CNT_H(_n)		(GDM_BASE(_n) + 0x2ec)
+#define REG_FE_GDM_RX_ETH_L127_CNT_H(_n)	(GDM_BASE(_n) + 0x2f0)
+#define REG_FE_GDM_RX_ETH_L255_CNT_H(_n)	(GDM_BASE(_n) + 0x2f4)
+#define REG_FE_GDM_RX_ETH_L511_CNT_H(_n)	(GDM_BASE(_n) + 0x2f8)
+#define REG_FE_GDM_RX_ETH_L1023_CNT_H(_n)	(GDM_BASE(_n) + 0x2fc)
+
+#define REG_GDM2_CHN_RLS		(GDM2_BASE + 0x20)
+#define MBI_RX_AGE_SEL_MASK		GENMASK(26, 25)
+#define MBI_TX_AGE_SEL_MASK		GENMASK(18, 17)
+
+#define REG_GDM3_FWD_CFG		GDM3_BASE
+#define GDM3_PAD_EN_MASK		BIT(28)
+
+#define REG_GDM4_FWD_CFG		GDM4_BASE
+#define GDM4_PAD_EN_MASK		BIT(28)
+#define GDM4_SPORT_OFFSET0_MASK		GENMASK(11, 8)
+
+#define REG_GDM4_SRC_PORT_SET		(GDM4_BASE + 0x23c)
+#define GDM4_SPORT_OFF2_MASK		GENMASK(19, 16)
+#define GDM4_SPORT_OFF1_MASK		GENMASK(15, 12)
+#define GDM4_SPORT_OFF0_MASK		GENMASK(11, 8)
+
+#define REG_IP_FRAG_FP			0x2010
+#define IP_ASSEMBLE_PORT_MASK		GENMASK(24, 21)
+#define IP_ASSEMBLE_NBQ_MASK		GENMASK(20, 16)
+#define IP_FRAGMENT_PORT_MASK		GENMASK(8, 5)
+#define IP_FRAGMENT_NBQ_MASK		GENMASK(4, 0)
+
+#define REG_MC_VLAN_EN			0x2100
+#define MC_VLAN_EN_MASK			BIT(0)
+
+#define REG_MC_VLAN_CFG			0x2104
+#define MC_VLAN_CFG_CMD_DONE_MASK	BIT(31)
+#define MC_VLAN_CFG_TABLE_ID_MASK	GENMASK(21, 16)
+#define MC_VLAN_CFG_PORT_ID_MASK	GENMASK(11, 8)
+#define MC_VLAN_CFG_TABLE_SEL_MASK	BIT(4)
+#define MC_VLAN_CFG_RW_MASK		BIT(0)
+
+#define REG_MC_VLAN_DATA		0x2108
+
+#define REG_SP_DFT_CPORT(_n)		(0x20e0 + ((_n) << 2))
+#define SP_CPORT_PCIE1_MASK		GENMASK(31, 28)
+#define SP_CPORT_PCIE0_MASK		GENMASK(27, 24)
+#define SP_CPORT_USB_MASK		GENMASK(7, 4)
+#define SP_CPORT_ETH_MASK		GENMASK(7, 4)
+
+#define REG_SRC_PORT_FC_MAP6		0x2298
+#define FC_ID_OF_SRC_PORT27_MASK	GENMASK(28, 24)
+#define FC_ID_OF_SRC_PORT26_MASK	GENMASK(20, 16)
+#define FC_ID_OF_SRC_PORT25_MASK	GENMASK(12, 8)
+#define FC_ID_OF_SRC_PORT24_MASK	GENMASK(4, 0)
+
+#define REG_CDM5_RX_OQ1_DROP_CNT	0x29d4
+
+/* QDMA */
+#define REG_QDMA_GLOBAL_CFG			0x0004
+#define GLOBAL_CFG_RX_2B_OFFSET_MASK		BIT(31)
+#define GLOBAL_CFG_DMA_PREFERENCE_MASK		GENMASK(30, 29)
+#define GLOBAL_CFG_CPU_TXR_RR_MASK		BIT(28)
+#define GLOBAL_CFG_DSCP_BYTE_SWAP_MASK		BIT(27)
+#define GLOBAL_CFG_PAYLOAD_BYTE_SWAP_MASK	BIT(26)
+#define GLOBAL_CFG_MULTICAST_MODIFY_FP_MASK	BIT(25)
+#define GLOBAL_CFG_OAM_MODIFY_MASK		BIT(24)
+#define GLOBAL_CFG_RESET_MASK			BIT(23)
+#define GLOBAL_CFG_RESET_DONE_MASK		BIT(22)
+#define GLOBAL_CFG_MULTICAST_EN_MASK		BIT(21)
+#define GLOBAL_CFG_IRQ1_EN_MASK			BIT(20)
+#define GLOBAL_CFG_IRQ0_EN_MASK			BIT(19)
+#define GLOBAL_CFG_LOOPCNT_EN_MASK		BIT(18)
+#define GLOBAL_CFG_RD_BYPASS_WR_MASK		BIT(17)
+#define GLOBAL_CFG_QDMA_LOOPBACK_MASK		BIT(16)
+#define GLOBAL_CFG_LPBK_RXQ_SEL_MASK		GENMASK(13, 8)
+#define GLOBAL_CFG_CHECK_DONE_MASK		BIT(7)
+#define GLOBAL_CFG_TX_WB_DONE_MASK		BIT(6)
+#define GLOBAL_CFG_MAX_ISSUE_NUM_MASK		GENMASK(5, 4)
+#define GLOBAL_CFG_RX_DMA_BUSY_MASK		BIT(3)
+#define GLOBAL_CFG_RX_DMA_EN_MASK		BIT(2)
+#define GLOBAL_CFG_TX_DMA_BUSY_MASK		BIT(1)
+#define GLOBAL_CFG_TX_DMA_EN_MASK		BIT(0)
+
+#define REG_FWD_DSCP_BASE			0x0010
+#define REG_FWD_BUF_BASE			0x0014
+
+#define REG_HW_FWD_DSCP_CFG			0x0018
+#define HW_FWD_DSCP_PAYLOAD_SIZE_MASK		GENMASK(29, 28)
+#define HW_FWD_DSCP_SCATTER_LEN_MASK		GENMASK(17, 16)
+#define HW_FWD_DSCP_MIN_SCATTER_LEN_MASK	GENMASK(15, 0)
+
+#define REG_INT_STATUS(_n)		\
+	(((_n) == 4) ? 0x0730 :		\
+	 ((_n) == 3) ? 0x0724 :		\
+	 ((_n) == 2) ? 0x0720 :		\
+	 ((_n) == 1) ? 0x0024 : 0x0020)
+
+#define REG_INT_ENABLE(_b, _n)		\
+	(((_n) == 4) ? 0x0750 + ((_b) << 5) :	\
+	 ((_n) == 3) ? 0x0744 + ((_b) << 5) :	\
+	 ((_n) == 2) ? 0x0740 + ((_b) << 5) :	\
+	 ((_n) == 1) ? 0x002c + ((_b) << 3) :	\
+		       0x0028 + ((_b) << 3))
+
+/* QDMA_CSR_INT_ENABLE1 */
+#define RX15_COHERENT_INT_MASK		BIT(31)
+#define RX14_COHERENT_INT_MASK		BIT(30)
+#define RX13_COHERENT_INT_MASK		BIT(29)
+#define RX12_COHERENT_INT_MASK		BIT(28)
+#define RX11_COHERENT_INT_MASK		BIT(27)
+#define RX10_COHERENT_INT_MASK		BIT(26)
+#define RX9_COHERENT_INT_MASK		BIT(25)
+#define RX8_COHERENT_INT_MASK		BIT(24)
+#define RX7_COHERENT_INT_MASK		BIT(23)
+#define RX6_COHERENT_INT_MASK		BIT(22)
+#define RX5_COHERENT_INT_MASK		BIT(21)
+#define RX4_COHERENT_INT_MASK		BIT(20)
+#define RX3_COHERENT_INT_MASK		BIT(19)
+#define RX2_COHERENT_INT_MASK		BIT(18)
+#define RX1_COHERENT_INT_MASK		BIT(17)
+#define RX0_COHERENT_INT_MASK		BIT(16)
+#define TX7_COHERENT_INT_MASK		BIT(15)
+#define TX6_COHERENT_INT_MASK		BIT(14)
+#define TX5_COHERENT_INT_MASK		BIT(13)
+#define TX4_COHERENT_INT_MASK		BIT(12)
+#define TX3_COHERENT_INT_MASK		BIT(11)
+#define TX2_COHERENT_INT_MASK		BIT(10)
+#define TX1_COHERENT_INT_MASK		BIT(9)
+#define TX0_COHERENT_INT_MASK		BIT(8)
+#define CNT_OVER_FLOW_INT_MASK		BIT(7)
+#define IRQ1_FULL_INT_MASK		BIT(5)
+#define IRQ1_INT_MASK			BIT(4)
+#define HWFWD_DSCP_LOW_INT_MASK		BIT(3)
+#define HWFWD_DSCP_EMPTY_INT_MASK	BIT(2)
+#define IRQ0_FULL_INT_MASK		BIT(1)
+#define IRQ0_INT_MASK			BIT(0)
+
+#define RX_COHERENT_LOW_INT_MASK				\
+	(RX15_COHERENT_INT_MASK | RX14_COHERENT_INT_MASK |	\
+	 RX13_COHERENT_INT_MASK | RX12_COHERENT_INT_MASK |	\
+	 RX11_COHERENT_INT_MASK | RX10_COHERENT_INT_MASK |	\
+	 RX9_COHERENT_INT_MASK | RX8_COHERENT_INT_MASK |	\
+	 RX7_COHERENT_INT_MASK | RX6_COHERENT_INT_MASK |	\
+	 RX5_COHERENT_INT_MASK | RX4_COHERENT_INT_MASK |	\
+	 RX3_COHERENT_INT_MASK | RX2_COHERENT_INT_MASK |	\
+	 RX1_COHERENT_INT_MASK | RX0_COHERENT_INT_MASK)
+
+#define RX_COHERENT_LOW_OFFSET	__ffs(RX_COHERENT_LOW_INT_MASK)
+#define INT_RX0_MASK(_n)					\
+	(((_n) << RX_COHERENT_LOW_OFFSET) & RX_COHERENT_LOW_INT_MASK)
+
+#define TX_COHERENT_LOW_INT_MASK				\
+	(TX7_COHERENT_INT_MASK | TX6_COHERENT_INT_MASK |	\
+	 TX5_COHERENT_INT_MASK | TX4_COHERENT_INT_MASK |	\
+	 TX3_COHERENT_INT_MASK | TX2_COHERENT_INT_MASK |	\
+	 TX1_COHERENT_INT_MASK | TX0_COHERENT_INT_MASK)
+
+#define TX_DONE_INT_MASK(_n)					\
+	((_n) ? IRQ1_INT_MASK | IRQ1_FULL_INT_MASK		\
+	      : IRQ0_INT_MASK | IRQ0_FULL_INT_MASK)
+
+#define INT_TX_MASK						\
+	(IRQ1_INT_MASK | IRQ1_FULL_INT_MASK |			\
+	 IRQ0_INT_MASK | IRQ0_FULL_INT_MASK)
+
+/* QDMA_CSR_INT_ENABLE2 */
+#define RX15_NO_CPU_DSCP_INT_MASK	BIT(31)
+#define RX14_NO_CPU_DSCP_INT_MASK	BIT(30)
+#define RX13_NO_CPU_DSCP_INT_MASK	BIT(29)
+#define RX12_NO_CPU_DSCP_INT_MASK	BIT(28)
+#define RX11_NO_CPU_DSCP_INT_MASK	BIT(27)
+#define RX10_NO_CPU_DSCP_INT_MASK	BIT(26)
+#define RX9_NO_CPU_DSCP_INT_MASK	BIT(25)
+#define RX8_NO_CPU_DSCP_INT_MASK	BIT(24)
+#define RX7_NO_CPU_DSCP_INT_MASK	BIT(23)
+#define RX6_NO_CPU_DSCP_INT_MASK	BIT(22)
+#define RX5_NO_CPU_DSCP_INT_MASK	BIT(21)
+#define RX4_NO_CPU_DSCP_INT_MASK	BIT(20)
+#define RX3_NO_CPU_DSCP_INT_MASK	BIT(19)
+#define RX2_NO_CPU_DSCP_INT_MASK	BIT(18)
+#define RX1_NO_CPU_DSCP_INT_MASK	BIT(17)
+#define RX0_NO_CPU_DSCP_INT_MASK	BIT(16)
+#define RX15_DONE_INT_MASK		BIT(15)
+#define RX14_DONE_INT_MASK		BIT(14)
+#define RX13_DONE_INT_MASK		BIT(13)
+#define RX12_DONE_INT_MASK		BIT(12)
+#define RX11_DONE_INT_MASK		BIT(11)
+#define RX10_DONE_INT_MASK		BIT(10)
+#define RX9_DONE_INT_MASK		BIT(9)
+#define RX8_DONE_INT_MASK		BIT(8)
+#define RX7_DONE_INT_MASK		BIT(7)
+#define RX6_DONE_INT_MASK		BIT(6)
+#define RX5_DONE_INT_MASK		BIT(5)
+#define RX4_DONE_INT_MASK		BIT(4)
+#define RX3_DONE_INT_MASK		BIT(3)
+#define RX2_DONE_INT_MASK		BIT(2)
+#define RX1_DONE_INT_MASK		BIT(1)
+#define RX0_DONE_INT_MASK		BIT(0)
+
+#define RX_NO_CPU_DSCP_LOW_INT_MASK					\
+	(RX15_NO_CPU_DSCP_INT_MASK | RX14_NO_CPU_DSCP_INT_MASK |	\
+	 RX13_NO_CPU_DSCP_INT_MASK | RX12_NO_CPU_DSCP_INT_MASK |	\
+	 RX11_NO_CPU_DSCP_INT_MASK | RX10_NO_CPU_DSCP_INT_MASK |	\
+	 RX9_NO_CPU_DSCP_INT_MASK | RX8_NO_CPU_DSCP_INT_MASK |		\
+	 RX7_NO_CPU_DSCP_INT_MASK | RX6_NO_CPU_DSCP_INT_MASK |		\
+	 RX5_NO_CPU_DSCP_INT_MASK | RX4_NO_CPU_DSCP_INT_MASK |		\
+	 RX3_NO_CPU_DSCP_INT_MASK | RX2_NO_CPU_DSCP_INT_MASK |		\
+	 RX1_NO_CPU_DSCP_INT_MASK | RX0_NO_CPU_DSCP_INT_MASK)
+
+#define RX_DONE_LOW_INT_MASK				\
+	(RX15_DONE_INT_MASK | RX14_DONE_INT_MASK |	\
+	 RX13_DONE_INT_MASK | RX12_DONE_INT_MASK |	\
+	 RX11_DONE_INT_MASK | RX10_DONE_INT_MASK |	\
+	 RX9_DONE_INT_MASK | RX8_DONE_INT_MASK |	\
+	 RX7_DONE_INT_MASK | RX6_DONE_INT_MASK |	\
+	 RX5_DONE_INT_MASK | RX4_DONE_INT_MASK |	\
+	 RX3_DONE_INT_MASK | RX2_DONE_INT_MASK |	\
+	 RX1_DONE_INT_MASK | RX0_DONE_INT_MASK)
+
+#define RX_NO_CPU_DSCP_LOW_OFFSET	__ffs(RX_NO_CPU_DSCP_LOW_INT_MASK)
+#define INT_RX1_MASK(_n)							\
+	((((_n) << RX_NO_CPU_DSCP_LOW_OFFSET) & RX_NO_CPU_DSCP_LOW_INT_MASK) |	\
+	 (RX_DONE_LOW_INT_MASK & (_n)))
+
+/* QDMA_CSR_INT_ENABLE3 */
+#define RX31_NO_CPU_DSCP_INT_MASK	BIT(31)
+#define RX30_NO_CPU_DSCP_INT_MASK	BIT(30)
+#define RX29_NO_CPU_DSCP_INT_MASK	BIT(29)
+#define RX28_NO_CPU_DSCP_INT_MASK	BIT(28)
+#define RX27_NO_CPU_DSCP_INT_MASK	BIT(27)
+#define RX26_NO_CPU_DSCP_INT_MASK	BIT(26)
+#define RX25_NO_CPU_DSCP_INT_MASK	BIT(25)
+#define RX24_NO_CPU_DSCP_INT_MASK	BIT(24)
+#define RX23_NO_CPU_DSCP_INT_MASK	BIT(23)
+#define RX22_NO_CPU_DSCP_INT_MASK	BIT(22)
+#define RX21_NO_CPU_DSCP_INT_MASK	BIT(21)
+#define RX20_NO_CPU_DSCP_INT_MASK	BIT(20)
+#define RX19_NO_CPU_DSCP_INT_MASK	BIT(19)
+#define RX18_NO_CPU_DSCP_INT_MASK	BIT(18)
+#define RX17_NO_CPU_DSCP_INT_MASK	BIT(17)
+#define RX16_NO_CPU_DSCP_INT_MASK	BIT(16)
+#define RX31_DONE_INT_MASK		BIT(15)
+#define RX30_DONE_INT_MASK		BIT(14)
+#define RX29_DONE_INT_MASK		BIT(13)
+#define RX28_DONE_INT_MASK		BIT(12)
+#define RX27_DONE_INT_MASK		BIT(11)
+#define RX26_DONE_INT_MASK		BIT(10)
+#define RX25_DONE_INT_MASK		BIT(9)
+#define RX24_DONE_INT_MASK		BIT(8)
+#define RX23_DONE_INT_MASK		BIT(7)
+#define RX22_DONE_INT_MASK		BIT(6)
+#define RX21_DONE_INT_MASK		BIT(5)
+#define RX20_DONE_INT_MASK		BIT(4)
+#define RX19_DONE_INT_MASK		BIT(3)
+#define RX18_DONE_INT_MASK		BIT(2)
+#define RX17_DONE_INT_MASK		BIT(1)
+#define RX16_DONE_INT_MASK		BIT(0)
+
+#define RX_NO_CPU_DSCP_HIGH_INT_MASK					\
+	(RX31_NO_CPU_DSCP_INT_MASK | RX30_NO_CPU_DSCP_INT_MASK |	\
+	 RX29_NO_CPU_DSCP_INT_MASK | RX28_NO_CPU_DSCP_INT_MASK |	\
+	 RX27_NO_CPU_DSCP_INT_MASK | RX26_NO_CPU_DSCP_INT_MASK |	\
+	 RX25_NO_CPU_DSCP_INT_MASK | RX24_NO_CPU_DSCP_INT_MASK |	\
+	 RX23_NO_CPU_DSCP_INT_MASK | RX22_NO_CPU_DSCP_INT_MASK |	\
+	 RX21_NO_CPU_DSCP_INT_MASK | RX20_NO_CPU_DSCP_INT_MASK |	\
+	 RX19_NO_CPU_DSCP_INT_MASK | RX18_NO_CPU_DSCP_INT_MASK |	\
+	 RX17_NO_CPU_DSCP_INT_MASK | RX16_NO_CPU_DSCP_INT_MASK)
+
+#define RX_DONE_HIGH_INT_MASK				\
+	(RX31_DONE_INT_MASK | RX30_DONE_INT_MASK |	\
+	 RX29_DONE_INT_MASK | RX28_DONE_INT_MASK |	\
+	 RX27_DONE_INT_MASK | RX26_DONE_INT_MASK |	\
+	 RX25_DONE_INT_MASK | RX24_DONE_INT_MASK |	\
+	 RX23_DONE_INT_MASK | RX22_DONE_INT_MASK |	\
+	 RX21_DONE_INT_MASK | RX20_DONE_INT_MASK |	\
+	 RX19_DONE_INT_MASK | RX18_DONE_INT_MASK |	\
+	 RX17_DONE_INT_MASK | RX16_DONE_INT_MASK)
+
+#define RX_DONE_HIGH_OFFSET	fls(RX_DONE_HIGH_INT_MASK)
+#define RX_DONE_INT_MASK	\
+	((RX_DONE_HIGH_INT_MASK << RX_DONE_HIGH_OFFSET) | RX_DONE_LOW_INT_MASK)
+
+#define INT_RX2_MASK(_n)				\
+	((RX_NO_CPU_DSCP_HIGH_INT_MASK & (_n)) |	\
+	 (((_n) >> RX_DONE_HIGH_OFFSET) & RX_DONE_HIGH_INT_MASK))
+
+/* QDMA_CSR_INT_ENABLE4 */
+#define RX31_COHERENT_INT_MASK		BIT(31)
+#define RX30_COHERENT_INT_MASK		BIT(30)
+#define RX29_COHERENT_INT_MASK		BIT(29)
+#define RX28_COHERENT_INT_MASK		BIT(28)
+#define RX27_COHERENT_INT_MASK		BIT(27)
+#define RX26_COHERENT_INT_MASK		BIT(26)
+#define RX25_COHERENT_INT_MASK		BIT(25)
+#define RX24_COHERENT_INT_MASK		BIT(24)
+#define RX23_COHERENT_INT_MASK		BIT(23)
+#define RX22_COHERENT_INT_MASK		BIT(22)
+#define RX21_COHERENT_INT_MASK		BIT(21)
+#define RX20_COHERENT_INT_MASK		BIT(20)
+#define RX19_COHERENT_INT_MASK		BIT(19)
+#define RX18_COHERENT_INT_MASK		BIT(18)
+#define RX17_COHERENT_INT_MASK		BIT(17)
+#define RX16_COHERENT_INT_MASK		BIT(16)
+
+#define RX_COHERENT_HIGH_INT_MASK				\
+	(RX31_COHERENT_INT_MASK | RX30_COHERENT_INT_MASK |	\
+	 RX29_COHERENT_INT_MASK | RX28_COHERENT_INT_MASK |	\
+	 RX27_COHERENT_INT_MASK | RX26_COHERENT_INT_MASK |	\
+	 RX25_COHERENT_INT_MASK | RX24_COHERENT_INT_MASK |	\
+	 RX23_COHERENT_INT_MASK | RX22_COHERENT_INT_MASK |	\
+	 RX21_COHERENT_INT_MASK | RX20_COHERENT_INT_MASK |	\
+	 RX19_COHERENT_INT_MASK | RX18_COHERENT_INT_MASK |	\
+	 RX17_COHERENT_INT_MASK | RX16_COHERENT_INT_MASK)
+
+#define INT_RX3_MASK(_n)	(RX_COHERENT_HIGH_INT_MASK & (_n))
+
+/* QDMA_CSR_INT_ENABLE5 */
+#define TX31_COHERENT_INT_MASK		BIT(31)
+#define TX30_COHERENT_INT_MASK		BIT(30)
+#define TX29_COHERENT_INT_MASK		BIT(29)
+#define TX28_COHERENT_INT_MASK		BIT(28)
+#define TX27_COHERENT_INT_MASK		BIT(27)
+#define TX26_COHERENT_INT_MASK		BIT(26)
+#define TX25_COHERENT_INT_MASK		BIT(25)
+#define TX24_COHERENT_INT_MASK		BIT(24)
+#define TX23_COHERENT_INT_MASK		BIT(23)
+#define TX22_COHERENT_INT_MASK		BIT(22)
+#define TX21_COHERENT_INT_MASK		BIT(21)
+#define TX20_COHERENT_INT_MASK		BIT(20)
+#define TX19_COHERENT_INT_MASK		BIT(19)
+#define TX18_COHERENT_INT_MASK		BIT(18)
+#define TX17_COHERENT_INT_MASK		BIT(17)
+#define TX16_COHERENT_INT_MASK		BIT(16)
+#define TX15_COHERENT_INT_MASK		BIT(15)
+#define TX14_COHERENT_INT_MASK		BIT(14)
+#define TX13_COHERENT_INT_MASK		BIT(13)
+#define TX12_COHERENT_INT_MASK		BIT(12)
+#define TX11_COHERENT_INT_MASK		BIT(11)
+#define TX10_COHERENT_INT_MASK		BIT(10)
+#define TX9_COHERENT_INT_MASK		BIT(9)
+#define TX8_COHERENT_INT_MASK		BIT(8)
+
+#define TX_COHERENT_HIGH_INT_MASK				\
+	(TX31_COHERENT_INT_MASK | TX30_COHERENT_INT_MASK |	\
+	 TX29_COHERENT_INT_MASK | TX28_COHERENT_INT_MASK |	\
+	 TX27_COHERENT_INT_MASK | TX26_COHERENT_INT_MASK |	\
+	 TX25_COHERENT_INT_MASK | TX24_COHERENT_INT_MASK |	\
+	 TX23_COHERENT_INT_MASK | TX22_COHERENT_INT_MASK |	\
+	 TX21_COHERENT_INT_MASK | TX20_COHERENT_INT_MASK |	\
+	 TX19_COHERENT_INT_MASK | TX18_COHERENT_INT_MASK |	\
+	 TX17_COHERENT_INT_MASK | TX16_COHERENT_INT_MASK |	\
+	 TX15_COHERENT_INT_MASK | TX14_COHERENT_INT_MASK |	\
+	 TX13_COHERENT_INT_MASK | TX12_COHERENT_INT_MASK |	\
+	 TX11_COHERENT_INT_MASK | TX10_COHERENT_INT_MASK |	\
+	 TX9_COHERENT_INT_MASK | TX8_COHERENT_INT_MASK)
+
+#define REG_TX_IRQ_BASE(_n)		((_n) ? 0x0048 : 0x0050)
+
+#define REG_TX_IRQ_CFG(_n)		((_n) ? 0x004c : 0x0054)
+#define TX_IRQ_THR_MASK			GENMASK(27, 16)
+#define TX_IRQ_DEPTH_MASK		GENMASK(11, 0)
+
+#define REG_IRQ_CLEAR_LEN(_n)		((_n) ? 0x0064 : 0x0058)
+#define IRQ_CLEAR_LEN_MASK		GENMASK(7, 0)
+
+#define REG_IRQ_STATUS(_n)		((_n) ? 0x0068 : 0x005c)
+#define IRQ_ENTRY_LEN_MASK		GENMASK(27, 16)
+#define IRQ_HEAD_IDX_MASK		GENMASK(11, 0)
+
+#define REG_TX_RING_BASE(_n)	\
+	(((_n) < 8) ? 0x0100 + ((_n) << 5) : 0x0b00 + (((_n) - 8) << 5))
+
+#define REG_TX_RING_BLOCKING(_n)	\
+	(((_n) < 8) ? 0x0104 + ((_n) << 5) : 0x0b04 + (((_n) - 8) << 5))
+
+#define TX_RING_IRQ_BLOCKING_MAP_MASK			BIT(6)
+#define TX_RING_IRQ_BLOCKING_CFG_MASK			BIT(4)
+#define TX_RING_IRQ_BLOCKING_TX_DROP_EN_MASK		BIT(2)
+#define TX_RING_IRQ_BLOCKING_MAX_TH_TXRING_EN_MASK	BIT(1)
+#define TX_RING_IRQ_BLOCKING_MIN_TH_TXRING_EN_MASK	BIT(0)
+
+#define REG_TX_CPU_IDX(_n)	\
+	(((_n) < 8) ? 0x0108 + ((_n) << 5) : 0x0b08 + (((_n) - 8) << 5))
+
+#define TX_RING_CPU_IDX_MASK		GENMASK(15, 0)
+
+#define REG_TX_DMA_IDX(_n)	\
+	(((_n) < 8) ? 0x010c + ((_n) << 5) : 0x0b0c + (((_n) - 8) << 5))
+
+#define TX_RING_DMA_IDX_MASK		GENMASK(15, 0)
+
+#define IRQ_RING_IDX_MASK		GENMASK(20, 16)
+#define IRQ_DESC_IDX_MASK		GENMASK(15, 0)
+
+#define REG_RX_RING_BASE(_n)	\
+	(((_n) < 16) ? 0x0200 + ((_n) << 5) : 0x0e00 + (((_n) - 16) << 5))
+
+#define REG_RX_RING_SIZE(_n)	\
+	(((_n) < 16) ? 0x0204 + ((_n) << 5) : 0x0e04 + (((_n) - 16) << 5))
+
+#define RX_RING_THR_MASK		GENMASK(31, 16)
+#define RX_RING_SIZE_MASK		GENMASK(15, 0)
+
+#define REG_RX_CPU_IDX(_n)	\
+	(((_n) < 16) ? 0x0208 + ((_n) << 5) : 0x0e08 + (((_n) - 16) << 5))
+
+#define RX_RING_CPU_IDX_MASK		GENMASK(15, 0)
+
+#define REG_RX_DMA_IDX(_n)	\
+	(((_n) < 16) ? 0x020c + ((_n) << 5) : 0x0e0c + (((_n) - 16) << 5))
+
+#define REG_RX_DELAY_INT_IDX(_n)	\
+	(((_n) < 16) ? 0x0210 + ((_n) << 5) : 0x0e10 + (((_n) - 16) << 5))
+
+#define REG_RX_SCATTER_CFG(_n)	\
+	(((_n) < 16) ? 0x0214 + ((_n) << 5) : 0x0e14 + (((_n) - 16) << 5))
+
+#define RX_DELAY_INT_MASK		GENMASK(15, 0)
+
+#define RX_RING_DMA_IDX_MASK		GENMASK(15, 0)
+
+#define RX_RING_SG_EN_MASK		BIT(0)
+
+#define REG_INGRESS_TRTCM_CFG		0x0070
+#define INGRESS_TRTCM_EN_MASK		BIT(31)
+#define INGRESS_TRTCM_MODE_MASK		BIT(30)
+#define INGRESS_SLOW_TICK_RATIO_MASK	GENMASK(29, 16)
+#define INGRESS_FAST_TICK_MASK		GENMASK(15, 0)
+
+#define REG_QUEUE_CLOSE_CFG(_n)		(0x00a0 + ((_n) & 0xfc))
+#define TXQ_DISABLE_CHAN_QUEUE_MASK(_n, _m)	BIT((_m) + (((_n) & 0x3) << 3))
+
+#define REG_TXQ_DIS_CFG_BASE(_n)	((_n) ? 0x20a0 : 0x00a0)
+#define REG_TXQ_DIS_CFG(_n, _m)		(REG_TXQ_DIS_CFG_BASE((_n)) + (_m) << 2)
+
+#define REG_CNTR_CFG(_n)		(0x0400 + ((_n) << 3))
+#define CNTR_EN_MASK			BIT(31)
+#define CNTR_ALL_CHAN_EN_MASK		BIT(30)
+#define CNTR_ALL_QUEUE_EN_MASK		BIT(29)
+#define CNTR_ALL_DSCP_RING_EN_MASK	BIT(28)
+#define CNTR_SRC_MASK			GENMASK(27, 24)
+#define CNTR_DSCP_RING_MASK		GENMASK(20, 16)
+#define CNTR_CHAN_MASK			GENMASK(7, 3)
+#define CNTR_QUEUE_MASK			GENMASK(2, 0)
+
+#define REG_CNTR_VAL(_n)		(0x0404 + ((_n) << 3))
+
+#define REG_LMGR_INIT_CFG		0x1000
+#define LMGR_INIT_START			BIT(31)
+#define LMGR_SRAM_MODE_MASK		BIT(30)
+#define HW_FWD_PKTSIZE_OVERHEAD_MASK	GENMASK(27, 20)
+#define HW_FWD_DESC_NUM_MASK		GENMASK(16, 0)
+
+#define REG_FWD_DSCP_LOW_THR		0x1004
+#define FWD_DSCP_LOW_THR_MASK		GENMASK(17, 0)
+
+#define REG_EGRESS_RATE_METER_CFG		0x100c
+#define EGRESS_RATE_METER_EN_MASK		BIT(31)
+#define EGRESS_RATE_METER_EQ_RATE_EN_MASK	BIT(17)
+#define EGRESS_RATE_METER_WINDOW_SZ_MASK	GENMASK(16, 12)
+#define EGRESS_RATE_METER_TIMESLICE_MASK	GENMASK(10, 0)
+
+#define REG_EGRESS_TRTCM_CFG		0x1010
+#define EGRESS_TRTCM_EN_MASK		BIT(31)
+#define EGRESS_TRTCM_MODE_MASK		BIT(30)
+#define EGRESS_SLOW_TICK_RATIO_MASK	GENMASK(29, 16)
+#define EGRESS_FAST_TICK_MASK		GENMASK(15, 0)
+
+#define TRTCM_PARAM_RW_MASK		BIT(31)
+#define TRTCM_PARAM_RW_DONE_MASK	BIT(30)
+#define TRTCM_PARAM_TYPE_MASK		GENMASK(29, 28)
+#define TRTCM_METER_GROUP_MASK		GENMASK(27, 26)
+#define TRTCM_PARAM_INDEX_MASK		GENMASK(23, 17)
+#define TRTCM_PARAM_RATE_TYPE_MASK	BIT(16)
+
+#define REG_TRTCM_CFG_PARAM(_n)		((_n) + 0x4)
+#define REG_TRTCM_DATA_LOW(_n)		((_n) + 0x8)
+#define REG_TRTCM_DATA_HIGH(_n)		((_n) + 0xc)
+
+#define RATE_LIMIT_PARAM_RW_MASK	BIT(31)
+#define RATE_LIMIT_PARAM_RW_DONE_MASK	BIT(30)
+#define RATE_LIMIT_PARAM_TYPE_MASK	GENMASK(29, 28)
+#define RATE_LIMIT_METER_GROUP_MASK	GENMASK(27, 26)
+#define RATE_LIMIT_PARAM_INDEX_MASK	GENMASK(23, 16)
+
+#define REG_TXWRR_MODE_CFG		0x1020
+#define TWRR_WEIGHT_SCALE_MASK		BIT(31)
+#define TWRR_WEIGHT_BASE_MASK		BIT(3)
+
+#define REG_TXWRR_WEIGHT_CFG		0x1024
+#define TWRR_RW_CMD_MASK		BIT(31)
+#define TWRR_RW_CMD_DONE		BIT(30)
+#define TWRR_CHAN_IDX_MASK		GENMASK(23, 19)
+#define TWRR_QUEUE_IDX_MASK		GENMASK(18, 16)
+#define TWRR_VALUE_MASK			GENMASK(15, 0)
+
+#define REG_PSE_BUF_USAGE_CFG		0x1028
+#define PSE_BUF_ESTIMATE_EN_MASK	BIT(29)
+
+#define REG_CHAN_QOS_MODE(_n)		(0x1040 + ((_n) << 2))
+#define CHAN_QOS_MODE_MASK(_n)		GENMASK(2 + ((_n) << 2), (_n) << 2)
+
+#define REG_GLB_TRTCM_CFG		0x1080
+#define GLB_TRTCM_EN_MASK		BIT(31)
+#define GLB_TRTCM_MODE_MASK		BIT(30)
+#define GLB_SLOW_TICK_RATIO_MASK	GENMASK(29, 16)
+#define GLB_FAST_TICK_MASK		GENMASK(15, 0)
+
+#define REG_TXQ_CNGST_CFG		0x10a0
+#define TXQ_CNGST_DROP_EN		BIT(31)
+#define TXQ_CNGST_DEI_DROP_EN		BIT(30)
+
+#define REG_SLA_TRTCM_CFG		0x1150
+#define SLA_TRTCM_EN_MASK		BIT(31)
+#define SLA_TRTCM_MODE_MASK		BIT(30)
+#define SLA_SLOW_TICK_RATIO_MASK	GENMASK(29, 16)
+#define SLA_FAST_TICK_MASK		GENMASK(15, 0)
+
+/* CTRL */
+#define QDMA_DESC_DONE_MASK		BIT(31)
+#define QDMA_DESC_DROP_MASK		BIT(30) /* tx: drop - rx: overflow */
+#define QDMA_DESC_MORE_MASK		BIT(29) /* more SG elements */
+#define QDMA_DESC_DEI_MASK		BIT(25)
+#define QDMA_DESC_NO_DROP_MASK		BIT(24)
+#define QDMA_DESC_LEN_MASK		GENMASK(15, 0)
+/* DATA */
+#define QDMA_DESC_NEXT_ID_MASK		GENMASK(15, 0)
+/* TX MSG0 */
+#define QDMA_ETH_TXMSG_MIC_IDX_MASK	BIT(30)
+#define QDMA_ETH_TXMSG_SP_TAG_MASK	GENMASK(29, 14)
+#define QDMA_ETH_TXMSG_ICO_MASK		BIT(13)
+#define QDMA_ETH_TXMSG_UCO_MASK		BIT(12)
+#define QDMA_ETH_TXMSG_TCO_MASK		BIT(11)
+#define QDMA_ETH_TXMSG_TSO_MASK		BIT(10)
+#define QDMA_ETH_TXMSG_FAST_MASK	BIT(9)
+#define QDMA_ETH_TXMSG_OAM_MASK		BIT(8)
+#define QDMA_ETH_TXMSG_CHAN_MASK	GENMASK(7, 3)
+#define QDMA_ETH_TXMSG_QUEUE_MASK	GENMASK(2, 0)
+/* TX MSG1 */
+#define QDMA_ETH_TXMSG_NO_DROP		BIT(31)
+#define QDMA_ETH_TXMSG_METER_MASK	GENMASK(30, 24)	/* 0x7f no meters */
+#define QDMA_ETH_TXMSG_FPORT_MASK	GENMASK(23, 20)
+#define QDMA_ETH_TXMSG_NBOQ_MASK	GENMASK(19, 15)
+#define QDMA_ETH_TXMSG_HWF_MASK		BIT(14)
+#define QDMA_ETH_TXMSG_HOP_MASK		BIT(13)
+#define QDMA_ETH_TXMSG_PTP_MASK		BIT(12)
+#define QDMA_ETH_TXMSG_ACNT_G1_MASK	GENMASK(10, 6)	/* 0x1f do not count */
+#define QDMA_ETH_TXMSG_ACNT_G0_MASK	GENMASK(5, 0)	/* 0x3f do not count */
+
+/* RX MSG0 */
+#define QDMA_ETH_RXMSG_SPTAG		GENMASK(21, 14)
+/* RX MSG1 */
+#define QDMA_ETH_RXMSG_DEI_MASK		BIT(31)
+#define QDMA_ETH_RXMSG_IP6_MASK		BIT(30)
+#define QDMA_ETH_RXMSG_IP4_MASK		BIT(29)
+#define QDMA_ETH_RXMSG_IP4F_MASK	BIT(28)
+#define QDMA_ETH_RXMSG_L4_VALID_MASK	BIT(27)
+#define QDMA_ETH_RXMSG_L4F_MASK		BIT(26)
+#define QDMA_ETH_RXMSG_SPORT_MASK	GENMASK(25, 21)
+#define QDMA_ETH_RXMSG_CRSN_MASK	GENMASK(20, 16)
+#define QDMA_ETH_RXMSG_PPE_ENTRY_MASK	GENMASK(15, 0)
+
+struct airoha_qdma_desc {
+	__le32 rsv;
+	__le32 ctrl;
+	__le32 addr;
+	__le32 data;
+	__le32 msg0;
+	__le32 msg1;
+	__le32 msg2;
+	__le32 msg3;
+};
+
+/* CTRL0 */
+#define QDMA_FWD_DESC_CTX_MASK		BIT(31)
+#define QDMA_FWD_DESC_RING_MASK		GENMASK(30, 28)
+#define QDMA_FWD_DESC_IDX_MASK		GENMASK(27, 16)
+#define QDMA_FWD_DESC_LEN_MASK		GENMASK(15, 0)
+/* CTRL1 */
+#define QDMA_FWD_DESC_FIRST_IDX_MASK	GENMASK(15, 0)
+/* CTRL2 */
+#define QDMA_FWD_DESC_MORE_PKT_NUM_MASK	GENMASK(2, 0)
+
+struct airoha_qdma_fwd_desc {
+	__le32 addr;
+	__le32 ctrl0;
+	__le32 ctrl1;
+	__le32 ctrl2;
+	__le32 msg0;
+	__le32 msg1;
+	__le32 rsv0;
+	__le32 rsv1;
+};
+
+#endif /* AIROHA_REGS_H */
diff --git a/drivers/net/ethernet/alacritech/Kconfig b/drivers/net/ethernet/alacritech/Kconfig
index 09496e18cdc5..5f285e18faf7 100644
--- a/drivers/net/ethernet/alacritech/Kconfig
+++ b/drivers/net/ethernet/alacritech/Kconfig
@@ -1,7 +1,8 @@
+# SPDX-License-Identifier: GPL-2.0-only
 config NET_VENDOR_ALACRITECH
 	bool "Alacritech devices"
 	default y
-	---help---
+	help
 	  If you have a network (Ethernet) card belonging to this class, say Y.
 
 	  Note that the answer to this question doesn't directly affect the
@@ -15,7 +16,7 @@ config SLICOSS
 	tristate "Alacritech Slicoss support"
 	depends on PCI
 	select CRC32
-	---help---
+	help
 	  This driver supports Gigabit Ethernet adapters based on the
 	  Session Layer Interface (SLIC) technology by Alacritech.
 
diff --git a/drivers/net/ethernet/alacritech/Makefile b/drivers/net/ethernet/alacritech/Makefile
index 8790e9ed8496..4378aadf895b 100644
--- a/drivers/net/ethernet/alacritech/Makefile
+++ b/drivers/net/ethernet/alacritech/Makefile
@@ -1,3 +1,4 @@
+# SPDX-License-Identifier: GPL-2.0-only
 #
 # Makefile for the Alacritech Slicoss driver
 #
diff --git a/drivers/net/ethernet/alacritech/slic.h b/drivers/net/ethernet/alacritech/slic.h
index 08931b4afc96..82071d0e5f7f 100644
--- a/drivers/net/ethernet/alacritech/slic.h
+++ b/drivers/net/ethernet/alacritech/slic.h
@@ -1,3 +1,4 @@
+/* SPDX-License-Identifier: GPL-2.0 */
 
 #ifndef _SLIC_H
 #define _SLIC_H
@@ -7,7 +8,6 @@
 #include <linux/spinlock_types.h>
 #include <linux/dma-mapping.h>
 #include <linux/pci.h>
-#include <linux/netdevice.h>
 #include <linux/list.h>
 #include <linux/u64_stats_sync.h>
 
@@ -265,8 +265,6 @@
 #define SLIC_NUM_STAT_DESC_ARRAYS	4
 #define SLIC_INVALID_STAT_DESC_IDX	0xffffffff
 
-#define SLIC_NAPI_WEIGHT		64
-
 #define SLIC_UPR_LSTAT			0
 #define SLIC_UPR_CONFIG			1
 
@@ -290,13 +288,13 @@ do {						\
 	u64_stats_update_end(&(st)->syncp);	\
 } while (0)
 
-#define SLIC_GET_STATS_COUNTER(newst, st, counter)			\
-{									\
-	unsigned int start;						\
+#define SLIC_GET_STATS_COUNTER(newst, st, counter)		\
+{								\
+	unsigned int start;					\
 	do {							\
-		start = u64_stats_fetch_begin_irq(&(st)->syncp);	\
-		newst = (st)->counter;					\
-	} while (u64_stats_fetch_retry_irq(&(st)->syncp, start));	\
+		start = u64_stats_fetch_begin(&(st)->syncp);	\
+		newst = (st)->counter;				\
+	} while (u64_stats_fetch_retry(&(st)->syncp, start));	\
 }
 
 struct slic_upr {
diff --git a/drivers/net/ethernet/alacritech/slicoss.c b/drivers/net/ethernet/alacritech/slicoss.c
index 15a8096c60df..f62851708d4f 100644
--- a/drivers/net/ethernet/alacritech/slicoss.c
+++ b/drivers/net/ethernet/alacritech/slicoss.c
@@ -1,19 +1,10 @@
+// SPDX-License-Identifier: GPL-2.0-only
 /*
  * Driver for Gigabit Ethernet adapters based on the Session Layer
  * Interface (SLIC) technology by Alacritech. The driver does not
  * support the hardware acceleration features provided by these cards.
  *
  * Copyright (C) 2016 Lino Sanfilippo <LinoSanfilippo@gmx.de>
- *
- * This program is free software; you can redistribute it and/or modify
- * it under the terms of the GNU General Public License as published by
- * the Free Software Foundation; either version 2 of the License.
- *
- * This program is distributed in the hope that it will be useful,
- * but WITHOUT ANY WARRANTY; without even the implied warranty of
- * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
- * GNU General Public License for more details.
- *
  */
 
 #include <linux/kernel.h>
@@ -35,7 +26,6 @@
 #include "slic.h"
 
 #define DRV_NAME			"slicoss"
-#define DRV_VERSION			"1.0"
 
 static const struct pci_device_id slic_id_tbl[] = {
 	{ PCI_DEVICE(PCI_VENDOR_ID_ALACRITECH,
@@ -345,8 +335,6 @@ static void slic_set_rx_mode(struct net_device *dev)
 	if (sdev->promisc != set_promisc) {
 		sdev->promisc = set_promisc;
 		slic_configure_rcv(sdev);
-		/* make sure writes to receiver cant leak out of the lock */
-		mmiowb();
 	}
 	spin_unlock_bh(&sdev->link_lock);
 }
@@ -355,10 +343,10 @@ static void slic_xmit_complete(struct slic_device *sdev)
 {
 	struct slic_tx_queue *txq = &sdev->txq;
 	struct net_device *dev = sdev->netdev;
-	unsigned int idx = txq->done_idx;
 	struct slic_tx_buffer *buff;
 	unsigned int frames = 0;
 	unsigned int bytes = 0;
+	unsigned int idx;
 
 	/* Limit processing to SLIC_MAX_TX_COMPLETIONS frames to avoid that new
 	 * completions during processing keeps the loop running endlessly.
@@ -795,8 +783,8 @@ static int slic_init_stat_queue(struct slic_device *sdev)
 	size = stq->len * sizeof(*descs) + DESC_ALIGN_MASK;
 
 	for (i = 0; i < SLIC_NUM_STAT_DESC_ARRAYS; i++) {
-		descs = dma_zalloc_coherent(&sdev->pdev->dev, size, &paddr,
-					    GFP_KERNEL);
+		descs = dma_alloc_coherent(&sdev->pdev->dev, size, &paddr,
+					   GFP_KERNEL);
 		if (!descs) {
 			netdev_err(sdev->netdev,
 				   "failed to allocate status descriptors\n");
@@ -1020,7 +1008,7 @@ static void slic_set_link_autoneg(struct slic_device *sdev)
 
 static void slic_set_mac_address(struct slic_device *sdev)
 {
-	u8 *addr = sdev->netdev->dev_addr;
+	const u8 *addr = sdev->netdev->dev_addr;
 	u32 val;
 
 	val = addr[5] | addr[4] << 8 | addr[3] << 16 | addr[2] << 24;
@@ -1240,8 +1228,8 @@ static int slic_init_shmem(struct slic_device *sdev)
 	struct slic_shmem_data *sm_data;
 	dma_addr_t paddr;
 
-	sm_data = dma_zalloc_coherent(&sdev->pdev->dev, sizeof(*sm_data),
-				      &paddr, GFP_KERNEL);
+	sm_data = dma_alloc_coherent(&sdev->pdev->dev, sizeof(*sm_data),
+				     &paddr, GFP_KERNEL);
 	if (!sm_data) {
 		dev_err(&sdev->pdev->dev, "failed to allocate shared memory\n");
 		return -ENOMEM;
@@ -1461,8 +1449,6 @@ static netdev_tx_t slic_xmit(struct sk_buff *skb, struct net_device *dev)
 
 	if (slic_get_free_tx_descs(txq) < SLIC_MAX_REQ_TX_DESCS)
 		netif_stop_queue(dev);
-	/* make sure writes to io-memory cant leak out of tx queue lock */
-	mmiowb();
 
 	return NETDEV_TX_OK;
 drop_skb:
@@ -1534,10 +1520,8 @@ static void slic_get_ethtool_stats(struct net_device *dev,
 
 static void slic_get_strings(struct net_device *dev, u32 stringset, u8 *data)
 {
-	if (stringset == ETH_SS_STATS) {
+	if (stringset == ETH_SS_STATS)
 		memcpy(data, slic_stats_strings, sizeof(slic_stats_strings));
-		data += sizeof(slic_stats_strings);
-	}
 }
 
 static void slic_get_drvinfo(struct net_device *dev,
@@ -1545,9 +1529,8 @@ static void slic_get_drvinfo(struct net_device *dev,
 {
 	struct slic_device *sdev = netdev_priv(dev);
 
-	strlcpy(info->driver, DRV_NAME, sizeof(info->driver));
-	strlcpy(info->version, DRV_VERSION, sizeof(info->version));
-	strlcpy(info->bus_info, pci_name(sdev->pdev), sizeof(info->bus_info));
+	strscpy(info->driver, DRV_NAME, sizeof(info->driver));
+	strscpy(info->bus_info, pci_name(sdev->pdev), sizeof(info->bus_info));
 }
 
 static const struct ethtool_ops slic_ethtool_ops = {
@@ -1621,8 +1604,8 @@ static int slic_read_eeprom(struct slic_device *sdev)
 	int err = 0;
 	u8 *mac[2];
 
-	eeprom = dma_zalloc_coherent(&sdev->pdev->dev, SLIC_EEPROM_SIZE,
-				     &paddr, GFP_KERNEL);
+	eeprom = dma_alloc_coherent(&sdev->pdev->dev, SLIC_EEPROM_SIZE,
+				    &paddr, GFP_KERNEL);
 	if (!eeprom)
 		return -ENOMEM;
 
@@ -1675,7 +1658,7 @@ static int slic_read_eeprom(struct slic_device *sdev)
 		goto free_eeprom;
 	}
 	/* set mac address */
-	ether_addr_copy(sdev->netdev->dev_addr, mac[devfn]);
+	eth_hw_addr_set(sdev->netdev, mac[devfn]);
 free_eeprom:
 	dma_free_coherent(&sdev->pdev->dev, SLIC_EEPROM_SIZE, eeprom, paddr);
 
@@ -1695,17 +1678,15 @@ static int slic_init(struct slic_device *sdev)
 	slic_card_reset(sdev);
 
 	err = slic_load_firmware(sdev);
-	if (err) {
-		dev_err(&sdev->pdev->dev, "failed to load firmware\n");
-		return err;
-	}
+	if (err)
+		return dev_err_probe(&sdev->pdev->dev, err,
+			"failed to load firmware\n");
 
 	/* we need the shared memory to read EEPROM so set it up temporarily */
 	err = slic_init_shmem(sdev);
-	if (err) {
-		dev_err(&sdev->pdev->dev, "failed to init shared memory\n");
-		return err;
-	}
+	if (err)
+		return dev_err_probe(&sdev->pdev->dev, err,
+			"failed to init shared memory\n");
 
 	err = slic_read_eeprom(sdev);
 	if (err) {
@@ -1727,13 +1708,13 @@ static bool slic_is_fiber(unsigned short subdev)
 {
 	switch (subdev) {
 	/* Mojave */
-	case PCI_SUBDEVICE_ID_ALACRITECH_1000X1F: /* fallthrough */
-	case PCI_SUBDEVICE_ID_ALACRITECH_SES1001F: /* fallthrough */
+	case PCI_SUBDEVICE_ID_ALACRITECH_1000X1F:
+	case PCI_SUBDEVICE_ID_ALACRITECH_SES1001F: fallthrough;
 	/* Oasis */
-	case PCI_SUBDEVICE_ID_ALACRITECH_SEN2002XF: /* fallthrough */
-	case PCI_SUBDEVICE_ID_ALACRITECH_SEN2001XF: /* fallthrough */
-	case PCI_SUBDEVICE_ID_ALACRITECH_SEN2104EF: /* fallthrough */
-	case PCI_SUBDEVICE_ID_ALACRITECH_SEN2102EF: /* fallthrough */
+	case PCI_SUBDEVICE_ID_ALACRITECH_SEN2002XF:
+	case PCI_SUBDEVICE_ID_ALACRITECH_SEN2001XF:
+	case PCI_SUBDEVICE_ID_ALACRITECH_SEN2104EF:
+	case PCI_SUBDEVICE_ID_ALACRITECH_SEN2102EF:
 		return true;
 	}
 	return false;
@@ -1758,10 +1739,9 @@ static int slic_probe(struct pci_dev *pdev, const struct pci_device_id *ent)
 	int err;
 
 	err = pci_enable_device(pdev);
-	if (err) {
-		dev_err(&pdev->dev, "failed to enable PCI device\n");
-		return err;
-	}
+	if (err)
+		return dev_err_probe(&pdev->dev, err,
+			"failed to enable PCI device\n");
 
 	pci_set_master(pdev);
 	pci_try_set_mwi(pdev);
@@ -1804,7 +1784,7 @@ static int slic_probe(struct pci_dev *pdev, const struct pci_device_id *ent)
 	sdev->is_fiber = slic_is_fiber(pdev->subsystem_device);
 	sdev->pdev = pdev;
 	sdev->netdev = dev;
-	sdev->regs = ioremap_nocache(pci_resource_start(pdev, 0),
+	sdev->regs = ioremap(pci_resource_start(pdev, 0),
 				     pci_resource_len(pdev, 0));
 	if (!sdev->regs) {
 		dev_err(&pdev->dev, "failed to map registers\n");
@@ -1818,7 +1798,7 @@ static int slic_probe(struct pci_dev *pdev, const struct pci_device_id *ent)
 		goto unmap;
 	}
 
-	netif_napi_add(dev, &sdev->napi, slic_poll, SLIC_NAPI_WEIGHT);
+	netif_napi_add(dev, &sdev->napi, slic_poll);
 	netif_carrier_off(dev);
 
 	err = register_netdev(dev);
@@ -1865,4 +1845,3 @@ module_pci_driver(slic_driver);
 MODULE_DESCRIPTION("Alacritech non-accelerated SLIC driver");
 MODULE_AUTHOR("Lino Sanfilippo <LinoSanfilippo@gmx.de>");
 MODULE_LICENSE("GPL");
-MODULE_VERSION(DRV_VERSION);
diff --git a/drivers/net/ethernet/allwinner/Kconfig b/drivers/net/ethernet/allwinner/Kconfig
index 47da7e7a5b6a..3e81059f8693 100644
--- a/drivers/net/ethernet/allwinner/Kconfig
+++ b/drivers/net/ethernet/allwinner/Kconfig
@@ -1,3 +1,4 @@
+# SPDX-License-Identifier: GPL-2.0-only
 #
 # Allwinner device configuration
 #
@@ -7,7 +8,7 @@ config NET_VENDOR_ALLWINNER
 	default y
 
 	depends on ARCH_SUNXI
-	---help---
+	help
 	  If you have a network (Ethernet) card belonging to this
 	  class, say Y here.
 
@@ -20,17 +21,17 @@ config NET_VENDOR_ALLWINNER
 if NET_VENDOR_ALLWINNER
 
 config SUN4I_EMAC
-        tristate "Allwinner A10 EMAC support"
+	tristate "Allwinner A10 EMAC support"
 	depends on ARCH_SUNXI
 	depends on OF
 	select CRC32
 	select MII
 	select PHYLIB
 	select MDIO_SUN4I
-        ---help---
-          Support for Allwinner A10 EMAC ethernet driver.
+	help
+	  Support for Allwinner A10 EMAC ethernet driver.
 
-          To compile this driver as a module, choose M here.  The module
-          will be called sun4i-emac.
+	  To compile this driver as a module, choose M here.  The module
+	  will be called sun4i-emac.
 
 endif # NET_VENDOR_ALLWINNER
diff --git a/drivers/net/ethernet/allwinner/Makefile b/drivers/net/ethernet/allwinner/Makefile
index 03129f796514..ddd5a5079e8a 100644
--- a/drivers/net/ethernet/allwinner/Makefile
+++ b/drivers/net/ethernet/allwinner/Makefile
@@ -1,3 +1,4 @@
+# SPDX-License-Identifier: GPL-2.0-only
 #
 # Makefile for the Allwinner device drivers.
 #
diff --git a/drivers/net/ethernet/allwinner/sun4i-emac.c b/drivers/net/ethernet/allwinner/sun4i-emac.c
index c8f4d26fc9d4..2f516b950f4e 100644
--- a/drivers/net/ethernet/allwinner/sun4i-emac.c
+++ b/drivers/net/ethernet/allwinner/sun4i-emac.c
@@ -29,11 +29,11 @@
 #include <linux/platform_device.h>
 #include <linux/phy.h>
 #include <linux/soc/sunxi/sunxi_sram.h>
+#include <linux/dmaengine.h>
 
 #include "sun4i-emac.h"
 
 #define DRV_NAME		"sun4i-emac"
-#define DRV_VERSION		"1.02"
 
 #define EMAC_MAX_FRAME_LEN	0x0600
 
@@ -77,7 +77,6 @@ struct emac_board_info {
 	void __iomem		*membase;
 	u32			msg_enable;
 	struct net_device	*ndev;
-	struct sk_buff		*skb_last;
 	u16			tx_fifo_stat;
 
 	int			emacrx_completed_flag;
@@ -88,6 +87,16 @@ struct emac_board_info {
 	unsigned int		duplex;
 
 	phy_interface_t		phy_interface;
+	struct dma_chan	*rx_chan;
+	phys_addr_t emac_rx_fifo;
+};
+
+struct emac_dma_req {
+	struct emac_board_info *db;
+	struct dma_async_tx_descriptor *desc;
+	struct sk_buff *skb;
+	dma_addr_t rxbuf;
+	int count;
 };
 
 static void emac_update_speed(struct net_device *dev)
@@ -97,9 +106,9 @@ static void emac_update_speed(struct net_device *dev)
 
 	/* set EMAC SPEED, depend on PHY  */
 	reg_val = readl(db->membase + EMAC_MAC_SUPP_REG);
-	reg_val &= ~(0x1 << 8);
+	reg_val &= ~EMAC_MAC_SUPP_100M;
 	if (db->speed == SPEED_100)
-		reg_val |= 1 << 8;
+		reg_val |= EMAC_MAC_SUPP_100M;
 	writel(reg_val, db->membase + EMAC_MAC_SUPP_REG);
 }
 
@@ -172,8 +181,7 @@ static int emac_mdio_probe(struct net_device *dev)
 	}
 
 	/* mask with MAC supported features */
-	phydev->supported &= PHY_BASIC_FEATURES;
-	phydev->advertising = phydev->supported;
+	phy_set_max_speed(phydev, SPEED_100);
 
 	db->link = 0;
 	db->speed = 0;
@@ -208,26 +216,123 @@ static void emac_inblk_32bit(void __iomem *reg, void *data, int count)
 	readsl(reg, data, round_up(count, 4) / 4);
 }
 
-static int emac_ioctl(struct net_device *dev, struct ifreq *rq, int cmd)
+static struct emac_dma_req *
+emac_alloc_dma_req(struct emac_board_info *db,
+		   struct dma_async_tx_descriptor *desc, struct sk_buff *skb,
+		   dma_addr_t rxbuf, int count)
 {
-	struct phy_device *phydev = dev->phydev;
+	struct emac_dma_req *req;
+
+	req = kzalloc(sizeof(struct emac_dma_req), GFP_ATOMIC);
+	if (!req)
+		return NULL;
+
+	req->db = db;
+	req->desc = desc;
+	req->skb = skb;
+	req->rxbuf = rxbuf;
+	req->count = count;
+	return req;
+}
 
-	if (!netif_running(dev))
-		return -EINVAL;
+static void emac_free_dma_req(struct emac_dma_req *req)
+{
+	kfree(req);
+}
 
-	if (!phydev)
-		return -ENODEV;
+static void emac_dma_done_callback(void *arg)
+{
+	struct emac_dma_req *req = arg;
+	struct emac_board_info *db = req->db;
+	struct sk_buff *skb = req->skb;
+	struct net_device *dev = db->ndev;
+	int rxlen = req->count;
+	u32 reg_val;
+
+	dma_unmap_single(db->dev, req->rxbuf, rxlen, DMA_FROM_DEVICE);
+
+	skb->protocol = eth_type_trans(skb, dev);
+	netif_rx(skb);
+	dev->stats.rx_bytes += rxlen;
+	/* Pass to upper layer */
+	dev->stats.rx_packets++;
+
+	/* re enable cpu receive */
+	reg_val = readl(db->membase + EMAC_RX_CTL_REG);
+	reg_val &= ~EMAC_RX_CTL_DMA_EN;
+	writel(reg_val, db->membase + EMAC_RX_CTL_REG);
+
+	/* re enable interrupt */
+	reg_val = readl(db->membase + EMAC_INT_CTL_REG);
+	reg_val |= EMAC_INT_CTL_RX_EN;
+	writel(reg_val, db->membase + EMAC_INT_CTL_REG);
+
+	db->emacrx_completed_flag = 1;
+	emac_free_dma_req(req);
+}
+
+static int emac_dma_inblk_32bit(struct emac_board_info *db,
+		struct sk_buff *skb, void *rdptr, int count)
+{
+	struct dma_async_tx_descriptor *desc;
+	dma_cookie_t cookie;
+	dma_addr_t rxbuf;
+	struct emac_dma_req *req;
+	int ret = 0;
+
+	rxbuf = dma_map_single(db->dev, rdptr, count, DMA_FROM_DEVICE);
+	ret = dma_mapping_error(db->dev, rxbuf);
+	if (ret) {
+		dev_err(db->dev, "dma mapping error.\n");
+		return ret;
+	}
+
+	desc = dmaengine_prep_slave_single(db->rx_chan, rxbuf, count,
+					   DMA_DEV_TO_MEM,
+					   DMA_PREP_INTERRUPT | DMA_CTRL_ACK);
+	if (!desc) {
+		dev_err(db->dev, "prepare slave single failed\n");
+		ret = -ENOMEM;
+		goto prepare_err;
+	}
+
+	req = emac_alloc_dma_req(db, desc, skb, rxbuf, count);
+	if (!req) {
+		dev_err(db->dev, "alloc emac dma req error.\n");
+		ret = -ENOMEM;
+		goto alloc_req_err;
+	}
+
+	desc->callback_param = req;
+	desc->callback = emac_dma_done_callback;
+
+	cookie = dmaengine_submit(desc);
+	ret = dma_submit_error(cookie);
+	if (ret) {
+		dev_err(db->dev, "dma submit error.\n");
+		goto submit_err;
+	}
+
+	dma_async_issue_pending(db->rx_chan);
+	return ret;
+
+submit_err:
+	emac_free_dma_req(req);
+
+alloc_req_err:
+	dmaengine_desc_free(desc);
 
-	return phy_mii_ioctl(phydev, rq, cmd);
+prepare_err:
+	dma_unmap_single(db->dev, rxbuf, count, DMA_FROM_DEVICE);
+	return ret;
 }
 
 /* ethtool ops */
 static void emac_get_drvinfo(struct net_device *dev,
 			      struct ethtool_drvinfo *info)
 {
-	strlcpy(info->driver, DRV_NAME, sizeof(DRV_NAME));
-	strlcpy(info->version, DRV_VERSION, sizeof(DRV_VERSION));
-	strlcpy(info->bus_info, dev_name(&dev->dev), sizeof(info->bus_info));
+	strscpy(info->driver, DRV_NAME, sizeof(info->driver));
+	strscpy(info->bus_info, dev_name(&dev->dev), sizeof(info->bus_info));
 }
 
 static u32 emac_get_msglevel(struct net_device *dev)
@@ -324,7 +429,7 @@ static unsigned int emac_powerup(struct net_device *ndev)
 	/* initial EMAC */
 	/* flush RX FIFO */
 	reg_val = readl(db->membase + EMAC_RX_CTL_REG);
-	reg_val |= 0x8;
+	reg_val |= EMAC_RX_CTL_FLUSH_FIFO;
 	writel(reg_val, db->membase + EMAC_RX_CTL_REG);
 	udelay(1);
 
@@ -336,8 +441,8 @@ static unsigned int emac_powerup(struct net_device *ndev)
 
 	/* set MII clock */
 	reg_val = readl(db->membase + EMAC_MAC_MCFG_REG);
-	reg_val &= (~(0xf << 2));
-	reg_val |= (0xD << 2);
+	reg_val &= ~EMAC_MAC_MCFG_MII_CLKD_MASK;
+	reg_val |= EMAC_MAC_MCFG_MII_CLKD_72;
 	writel(reg_val, db->membase + EMAC_MAC_MCFG_REG);
 
 	/* clear RX counter */
@@ -372,7 +477,7 @@ static int emac_set_mac_address(struct net_device *dev, void *p)
 	if (netif_running(dev))
 		return -EBUSY;
 
-	memcpy(dev->dev_addr, addr->sa_data, ETH_ALEN);
+	eth_hw_addr_set(dev, addr->sa_data);
 
 	writel(dev->dev_addr[0] << 16 | dev->dev_addr[1] << 8 | dev->
 	       dev_addr[2], db->membase + EMAC_MAC_A1_REG);
@@ -401,14 +506,14 @@ static void emac_init_device(struct net_device *dev)
 
 	/* enable RX/TX0/RX Hlevel interrup */
 	reg_val = readl(db->membase + EMAC_INT_CTL_REG);
-	reg_val |= (0xf << 0) | (0x01 << 8);
+	reg_val |= (EMAC_INT_CTL_TX_EN | EMAC_INT_CTL_TX_ABRT_EN | EMAC_INT_CTL_RX_EN);
 	writel(reg_val, db->membase + EMAC_INT_CTL_REG);
 
 	spin_unlock_irqrestore(&db->lock, flags);
 }
 
 /* Our watchdog timed out. Called by the networking layer */
-static void emac_timeout(struct net_device *dev)
+static void emac_timeout(struct net_device *dev, unsigned int txqueue)
 {
 	struct emac_board_info *db = netdev_priv(dev);
 	unsigned long flags;
@@ -433,7 +538,7 @@ static void emac_timeout(struct net_device *dev)
 /* Hardware start transmission.
  * Send a packet to media from the upper layer.
  */
-static int emac_start_xmit(struct sk_buff *skb, struct net_device *dev)
+static netdev_tx_t emac_start_xmit(struct sk_buff *skb, struct net_device *dev)
 {
 	struct emac_board_info *db = netdev_priv(dev);
 	unsigned long channel;
@@ -441,7 +546,7 @@ static int emac_start_xmit(struct sk_buff *skb, struct net_device *dev)
 
 	channel = db->tx_fifo_stat & 3;
 	if (channel == 3)
-		return 1;
+		return NETDEV_TX_BUSY;
 
 	channel = (channel == 1 ? 1 : 0);
 
@@ -515,7 +620,6 @@ static void emac_rx(struct net_device *dev)
 	struct sk_buff *skb;
 	u8 *rdptr;
 	bool good_packet;
-	static int rxlen_last;
 	unsigned int reg_val;
 	u32 rxhdr, rxstatus, rxcount, rxlen;
 
@@ -530,26 +634,12 @@ static void emac_rx(struct net_device *dev)
 		if (netif_msg_rx_status(db))
 			dev_dbg(db->dev, "RXCount: %x\n", rxcount);
 
-		if ((db->skb_last != NULL) && (rxlen_last > 0)) {
-			dev->stats.rx_bytes += rxlen_last;
-
-			/* Pass to upper layer */
-			db->skb_last->protocol = eth_type_trans(db->skb_last,
-								dev);
-			netif_rx(db->skb_last);
-			dev->stats.rx_packets++;
-			db->skb_last = NULL;
-			rxlen_last = 0;
-
-			reg_val = readl(db->membase + EMAC_RX_CTL_REG);
-			reg_val &= ~EMAC_RX_CTL_DMA_EN;
-			writel(reg_val, db->membase + EMAC_RX_CTL_REG);
-		}
-
 		if (!rxcount) {
 			db->emacrx_completed_flag = 1;
 			reg_val = readl(db->membase + EMAC_INT_CTL_REG);
-			reg_val |= (0xf << 0) | (0x01 << 8);
+			reg_val |= (EMAC_INT_CTL_TX_EN |
+					EMAC_INT_CTL_TX_ABRT_EN |
+					EMAC_INT_CTL_RX_EN);
 			writel(reg_val, db->membase + EMAC_INT_CTL_REG);
 
 			/* had one stuck? */
@@ -581,7 +671,9 @@ static void emac_rx(struct net_device *dev)
 			writel(reg_val | EMAC_CTL_RX_EN,
 			       db->membase + EMAC_CTL_REG);
 			reg_val = readl(db->membase + EMAC_INT_CTL_REG);
-			reg_val |= (0xf << 0) | (0x01 << 8);
+			reg_val |= (EMAC_INT_CTL_TX_EN |
+					EMAC_INT_CTL_TX_ABRT_EN |
+					EMAC_INT_CTL_RX_EN);
 			writel(reg_val, db->membase + EMAC_INT_CTL_REG);
 
 			db->emacrx_completed_flag = 1;
@@ -633,12 +725,25 @@ static void emac_rx(struct net_device *dev)
 			if (!skb)
 				continue;
 			skb_reserve(skb, 2);
-			rdptr = (u8 *) skb_put(skb, rxlen - 4);
+			rdptr = skb_put(skb, rxlen - 4);
 
 			/* Read received packet from RX SRAM */
 			if (netif_msg_rx_status(db))
 				dev_dbg(db->dev, "RxLen %x\n", rxlen);
 
+			if (rxlen >= dev->mtu && db->rx_chan) {
+				reg_val = readl(db->membase + EMAC_RX_CTL_REG);
+				reg_val |= EMAC_RX_CTL_DMA_EN;
+				writel(reg_val, db->membase + EMAC_RX_CTL_REG);
+				if (!emac_dma_inblk_32bit(db, skb, rdptr, rxlen))
+					break;
+
+				/* re enable cpu receive. then try to receive by emac_inblk_32bit */
+				reg_val = readl(db->membase + EMAC_RX_CTL_REG);
+				reg_val &= ~EMAC_RX_CTL_DMA_EN;
+				writel(reg_val, db->membase + EMAC_RX_CTL_REG);
+			}
+
 			emac_inblk_32bit(db->membase + EMAC_RX_IO_DATA_REG,
 					rdptr, rxlen);
 			dev->stats.rx_bytes += rxlen;
@@ -656,13 +761,11 @@ static irqreturn_t emac_interrupt(int irq, void *dev_id)
 	struct net_device *dev = dev_id;
 	struct emac_board_info *db = netdev_priv(dev);
 	int int_status;
-	unsigned long flags;
 	unsigned int reg_val;
 
 	/* A real interrupt coming */
 
-	/* holders of db->lock must always block IRQs */
-	spin_lock_irqsave(&db->lock, flags);
+	spin_lock(&db->lock);
 
 	/* Disable all interrupts */
 	writel(0, db->membase + EMAC_INT_CTL_REG);
@@ -684,19 +787,24 @@ static irqreturn_t emac_interrupt(int irq, void *dev_id)
 	}
 
 	/* Transmit Interrupt check */
-	if (int_status & (0x01 | 0x02))
+	if (int_status & EMAC_INT_STA_TX_COMPLETE)
 		emac_tx_done(dev, db, int_status);
 
-	if (int_status & (0x04 | 0x08))
+	if (int_status & EMAC_INT_STA_TX_ABRT)
 		netdev_info(dev, " ab : %x\n", int_status);
 
 	/* Re-enable interrupt mask */
 	if (db->emacrx_completed_flag == 1) {
 		reg_val = readl(db->membase + EMAC_INT_CTL_REG);
-		reg_val |= (0xf << 0) | (0x01 << 8);
+		reg_val |= (EMAC_INT_CTL_TX_EN | EMAC_INT_CTL_TX_ABRT_EN | EMAC_INT_CTL_RX_EN);
+		writel(reg_val, db->membase + EMAC_INT_CTL_REG);
+	} else {
+		reg_val = readl(db->membase + EMAC_INT_CTL_REG);
+		reg_val |= (EMAC_INT_CTL_TX_EN | EMAC_INT_CTL_TX_ABRT_EN);
 		writel(reg_val, db->membase + EMAC_INT_CTL_REG);
 	}
-	spin_unlock_irqrestore(&db->lock, flags);
+
+	spin_unlock(&db->lock);
 
 	return IRQ_HANDLED;
 }
@@ -792,7 +900,7 @@ static const struct net_device_ops emac_netdev_ops = {
 	.ndo_start_xmit		= emac_start_xmit,
 	.ndo_tx_timeout		= emac_timeout,
 	.ndo_set_rx_mode	= emac_set_rx_mode,
-	.ndo_do_ioctl		= emac_ioctl,
+	.ndo_eth_ioctl		= phy_do_ioctl_running,
 	.ndo_validate_addr	= eth_validate_addr,
 	.ndo_set_mac_address	= emac_set_mac_address,
 #ifdef CONFIG_NET_POLL_CONTROLLER
@@ -800,6 +908,58 @@ static const struct net_device_ops emac_netdev_ops = {
 #endif
 };
 
+static int emac_configure_dma(struct emac_board_info *db)
+{
+	struct platform_device *pdev = db->pdev;
+	struct net_device *ndev = db->ndev;
+	struct dma_slave_config conf = {};
+	struct resource *regs;
+	int err = 0;
+
+	regs = platform_get_resource(pdev, IORESOURCE_MEM, 0);
+	if (!regs) {
+		netdev_err(ndev, "get io resource from device failed.\n");
+		err = -ENOMEM;
+		goto out_clear_chan;
+	}
+
+	netdev_info(ndev, "get io resource from device: %pa, size = %u\n",
+		    &regs->start, (unsigned int)resource_size(regs));
+	db->emac_rx_fifo = regs->start + EMAC_RX_IO_DATA_REG;
+
+	db->rx_chan = dma_request_chan(&pdev->dev, "rx");
+	if (IS_ERR(db->rx_chan)) {
+		netdev_err(ndev,
+			   "failed to request dma channel. dma is disabled\n");
+		err = PTR_ERR(db->rx_chan);
+		goto out_clear_chan;
+	}
+
+	conf.direction = DMA_DEV_TO_MEM;
+	conf.dst_addr_width = DMA_SLAVE_BUSWIDTH_4_BYTES;
+	conf.src_addr_width = DMA_SLAVE_BUSWIDTH_4_BYTES;
+	conf.src_addr = db->emac_rx_fifo;
+	conf.dst_maxburst = 4;
+	conf.src_maxburst = 4;
+	conf.device_fc = false;
+
+	err = dmaengine_slave_config(db->rx_chan, &conf);
+	if (err) {
+		netdev_err(ndev, "config dma slave failed\n");
+		err = -EINVAL;
+		goto out_slave_configure_err;
+	}
+
+	return err;
+
+out_slave_configure_err:
+	dma_release_channel(db->rx_chan);
+
+out_clear_chan:
+	db->rx_chan = NULL;
+	return err;
+}
+
 /* Search EMAC board, allocate space and register it
  */
 static int emac_probe(struct platform_device *pdev)
@@ -808,7 +968,6 @@ static int emac_probe(struct platform_device *pdev)
 	struct emac_board_info *db;
 	struct net_device *ndev;
 	int ret = 0;
-	const char *mac_addr;
 
 	ndev = alloc_etherdev(sizeof(struct emac_board_info));
 	if (!ndev) {
@@ -819,7 +978,6 @@ static int emac_probe(struct platform_device *pdev)
 	SET_NETDEV_DEV(ndev, &pdev->dev);
 
 	db = netdev_priv(ndev);
-	memset(db, 0, sizeof(*db));
 
 	db->dev = &pdev->dev;
 	db->ndev = ndev;
@@ -844,16 +1002,19 @@ static int emac_probe(struct platform_device *pdev)
 		goto out_iounmap;
 	}
 
+	if (emac_configure_dma(db))
+		netdev_info(ndev, "configure dma failed. disable dma.\n");
+
 	db->clk = devm_clk_get(&pdev->dev, NULL);
 	if (IS_ERR(db->clk)) {
 		ret = PTR_ERR(db->clk);
-		goto out_iounmap;
+		goto out_dispose_mapping;
 	}
 
 	ret = clk_prepare_enable(db->clk);
 	if (ret) {
 		dev_err(&pdev->dev, "Error couldn't enable clock (%d)\n", ret);
-		goto out_iounmap;
+		goto out_dispose_mapping;
 	}
 
 	ret = sunxi_sram_claim(&pdev->dev);
@@ -862,7 +1023,9 @@ static int emac_probe(struct platform_device *pdev)
 		goto out_clk_disable_unprepare;
 	}
 
-	db->phy_node = of_parse_phandle(np, "phy", 0);
+	db->phy_node = of_parse_phandle(np, "phy-handle", 0);
+	if (!db->phy_node)
+		db->phy_node = of_parse_phandle(np, "phy", 0);
 	if (!db->phy_node) {
 		dev_err(&pdev->dev, "no associated PHY\n");
 		ret = -ENODEV;
@@ -870,12 +1033,9 @@ static int emac_probe(struct platform_device *pdev)
 	}
 
 	/* Read MAC-address from DT */
-	mac_addr = of_get_mac_address(np);
-	if (mac_addr)
-		memcpy(ndev->dev_addr, mac_addr, ETH_ALEN);
-
-	/* Check if the MAC address is valid, if not get a random one */
-	if (!is_valid_ether_addr(ndev->dev_addr)) {
+	ret = of_get_ethdev_address(np, ndev);
+	if (ret) {
+		/* if the MAC address is invalid get a random one */
 		eth_hw_addr_random(ndev);
 		dev_warn(&pdev->dev, "using random MAC address %pM\n",
 			 ndev->dev_addr);
@@ -910,6 +1070,9 @@ out_release_sram:
 	sunxi_sram_release(&pdev->dev);
 out_clk_disable_unprepare:
 	clk_disable_unprepare(db->clk);
+out_dispose_mapping:
+	irq_dispose_mapping(ndev->irq);
+	dma_release_channel(db->rx_chan);
 out_iounmap:
 	iounmap(db->membase);
 out:
@@ -920,19 +1083,24 @@ out:
 	return ret;
 }
 
-static int emac_remove(struct platform_device *pdev)
+static void emac_remove(struct platform_device *pdev)
 {
 	struct net_device *ndev = platform_get_drvdata(pdev);
 	struct emac_board_info *db = netdev_priv(ndev);
 
+	if (db->rx_chan) {
+		dmaengine_terminate_all(db->rx_chan);
+		dma_release_channel(db->rx_chan);
+	}
+
 	unregister_netdev(ndev);
 	sunxi_sram_release(&pdev->dev);
 	clk_disable_unprepare(db->clk);
+	irq_dispose_mapping(ndev->irq);
 	iounmap(db->membase);
 	free_netdev(ndev);
 
 	dev_dbg(&pdev->dev, "released and freed device\n");
-	return 0;
 }
 
 static int emac_suspend(struct platform_device *dev, pm_message_t state)
diff --git a/drivers/net/ethernet/allwinner/sun4i-emac.h b/drivers/net/ethernet/allwinner/sun4i-emac.h
index 38c72d9ec600..90bd9ad77607 100644
--- a/drivers/net/ethernet/allwinner/sun4i-emac.h
+++ b/drivers/net/ethernet/allwinner/sun4i-emac.h
@@ -38,6 +38,7 @@
 #define EMAC_RX_CTL_REG		(0x3c)
 #define EMAC_RX_CTL_AUTO_DRQ_EN		(1 << 1)
 #define EMAC_RX_CTL_DMA_EN		(1 << 2)
+#define EMAC_RX_CTL_FLUSH_FIFO		(1 << 3)
 #define EMAC_RX_CTL_PASS_ALL_EN		(1 << 4)
 #define EMAC_RX_CTL_PASS_CTL_EN		(1 << 5)
 #define EMAC_RX_CTL_PASS_CRC_ERR_EN	(1 << 6)
@@ -61,7 +62,21 @@
 #define EMAC_RX_IO_DATA_STATUS_OK	(1 << 7)
 #define EMAC_RX_FBC_REG		(0x50)
 #define EMAC_INT_CTL_REG	(0x54)
+#define EMAC_INT_CTL_RX_EN	(1 << 8)
+#define EMAC_INT_CTL_TX0_EN	(1)
+#define EMAC_INT_CTL_TX1_EN	(1 << 1)
+#define EMAC_INT_CTL_TX_EN	(EMAC_INT_CTL_TX0_EN | EMAC_INT_CTL_TX1_EN)
+#define EMAC_INT_CTL_TX0_ABRT_EN	(0x1 << 2)
+#define EMAC_INT_CTL_TX1_ABRT_EN	(0x1 << 3)
+#define EMAC_INT_CTL_TX_ABRT_EN	(EMAC_INT_CTL_TX0_ABRT_EN | EMAC_INT_CTL_TX1_ABRT_EN)
 #define EMAC_INT_STA_REG	(0x58)
+#define EMAC_INT_STA_TX0_COMPLETE	(0x1)
+#define EMAC_INT_STA_TX1_COMPLETE	(0x1 << 1)
+#define EMAC_INT_STA_TX_COMPLETE	(EMAC_INT_STA_TX0_COMPLETE | EMAC_INT_STA_TX1_COMPLETE)
+#define EMAC_INT_STA_TX0_ABRT	(0x1 << 2)
+#define EMAC_INT_STA_TX1_ABRT	(0x1 << 3)
+#define EMAC_INT_STA_TX_ABRT	(EMAC_INT_STA_TX0_ABRT | EMAC_INT_STA_TX1_ABRT)
+#define EMAC_INT_STA_RX_COMPLETE	(0x1 << 8)
 #define EMAC_MAC_CTL0_REG	(0x5c)
 #define EMAC_MAC_CTL0_RX_FLOW_CTL_EN	(1 << 2)
 #define EMAC_MAC_CTL0_TX_FLOW_CTL_EN	(1 << 3)
@@ -87,8 +102,11 @@
 #define EMAC_MAC_CLRT_RM		(0x0f)
 #define EMAC_MAC_MAXF_REG	(0x70)
 #define EMAC_MAC_SUPP_REG	(0x74)
+#define EMAC_MAC_SUPP_100M	(0x1 << 8)
 #define EMAC_MAC_TEST_REG	(0x78)
 #define EMAC_MAC_MCFG_REG	(0x7c)
+#define EMAC_MAC_MCFG_MII_CLKD_MASK	(0xff << 2)
+#define EMAC_MAC_MCFG_MII_CLKD_72	(0x0d << 2)
 #define EMAC_MAC_A0_REG		(0x98)
 #define EMAC_MAC_A1_REG		(0x9c)
 #define EMAC_MAC_A2_REG		(0xa0)
diff --git a/drivers/net/ethernet/alteon/Kconfig b/drivers/net/ethernet/alteon/Kconfig
index e06ccab354b5..cfe1f3159d61 100644
--- a/drivers/net/ethernet/alteon/Kconfig
+++ b/drivers/net/ethernet/alteon/Kconfig
@@ -1,3 +1,4 @@
+# SPDX-License-Identifier: GPL-2.0-only
 #
 # Alteon network device configuration
 #
@@ -6,7 +7,7 @@ config NET_VENDOR_ALTEON
 	bool "Alteon devices"
 	default y
 	depends on PCI
-	---help---
+	help
 	  If you have a network (Ethernet) card belonging to this class, say Y.
 
 	  Note that the answer to this question doesn't directly affect the
@@ -19,7 +20,7 @@ if NET_VENDOR_ALTEON
 config ACENIC
 	tristate "Alteon AceNIC/3Com 3C985/NetGear GA620 Gigabit support"
 	depends on PCI
-	---help---
+	help
 	  Say Y here if you have an Alteon AceNIC, 3Com 3C985(B), NetGear
 	  GA620, SGI Gigabit or Farallon PN9000-SX PCI Gigabit Ethernet
 	  adapter. The driver allows for using the Jumbo Frame option (9000
@@ -33,7 +34,7 @@ config ACENIC
 config ACENIC_OMIT_TIGON_I
 	bool "Omit support for old Tigon I based AceNICs"
 	depends on ACENIC
-	---help---
+	help
 	  Say Y here if you only have Tigon II based AceNICs and want to leave
 	  out support for the older Tigon I based cards which are no longer
 	  being sold (ie. the original Alteon AceNIC and 3Com 3C985 (non B
diff --git a/drivers/net/ethernet/alteon/Makefile b/drivers/net/ethernet/alteon/Makefile
index a2ca173f2a50..be5225559b6d 100644
--- a/drivers/net/ethernet/alteon/Makefile
+++ b/drivers/net/ethernet/alteon/Makefile
@@ -1,3 +1,4 @@
+# SPDX-License-Identifier: GPL-2.0-only
 #
 # Makefile for the Alteon network device drivers.
 #
diff --git a/drivers/net/ethernet/alteon/acenic.c b/drivers/net/ethernet/alteon/acenic.c
index a1a52eb53b14..9e6f91df2ba0 100644
--- a/drivers/net/ethernet/alteon/acenic.c
+++ b/drivers/net/ethernet/alteon/acenic.c
@@ -1,3 +1,4 @@
+// SPDX-License-Identifier: GPL-2.0-or-later
 /*
  * acenic.c: Linux driver for the Alteon AceNIC Gigabit Ethernet card
  *           and other Tigon based cards.
@@ -12,11 +13,6 @@
  * about the driver. Send mail to linux-acenic-help@sunsite.auc.dk to
  * see how to subscribe.
  *
- * This program is free software; you can redistribute it and/or modify
- * it under the terms of the GNU General Public License as published by
- * the Free Software Foundation; either version 2 of the License, or
- * (at your option) any later version.
- *
  * Additional credits:
  *   Pete Wyckoff <wyckoff@ca.sandia.gov>: Initial Linux/Alpha and trace
  *       dump support. The trace dump support has not been
@@ -441,7 +437,7 @@ static const struct ethtool_ops ace_ethtool_ops = {
 	.set_link_ksettings = ace_set_link_ksettings,
 };
 
-static void ace_watchdog(struct net_device *dev);
+static void ace_watchdog(struct net_device *dev, unsigned int txqueue);
 
 static const struct net_device_ops ace_netdev_ops = {
 	.ndo_open		= ace_open,
@@ -469,6 +465,7 @@ static int acenic_probe_one(struct pci_dev *pdev,
 	SET_NETDEV_DEV(dev, &pdev->dev);
 
 	ap = netdev_priv(dev);
+	ap->ndev = dev;
 	ap->pdev = pdev;
 	ap->name = pci_name(pdev);
 
@@ -551,6 +548,7 @@ static int acenic_probe_one(struct pci_dev *pdev,
 			       ap->name);
 			break;
 		}
+		fallthrough;
 	case PCI_VENDOR_ID_SGI:
 		printk(KERN_INFO "%s: SGI AceNIC ", ap->name);
 		break;
@@ -591,8 +589,7 @@ static int acenic_probe_one(struct pci_dev *pdev,
 	}
 	ap->name = dev->name;
 
-	if (ap->pci_using_dac)
-		dev->features |= NETIF_F_HIGHDMA;
+	dev->features |= NETIF_F_HIGHDMA;
 
 	pci_set_drvdata(pdev, dev);
 
@@ -645,9 +642,8 @@ static void acenic_remove_one(struct pci_dev *pdev)
 
 			ringp = &ap->skb->rx_std_skbuff[i];
 			mapping = dma_unmap_addr(ringp, mapping);
-			pci_unmap_page(ap->pdev, mapping,
-				       ACE_STD_BUFSIZE,
-				       PCI_DMA_FROMDEVICE);
+			dma_unmap_page(&ap->pdev->dev, mapping,
+				       ACE_STD_BUFSIZE, DMA_FROM_DEVICE);
 
 			ap->rx_std_ring[i].size = 0;
 			ap->skb->rx_std_skbuff[i].skb = NULL;
@@ -665,9 +661,9 @@ static void acenic_remove_one(struct pci_dev *pdev)
 
 				ringp = &ap->skb->rx_mini_skbuff[i];
 				mapping = dma_unmap_addr(ringp,mapping);
-				pci_unmap_page(ap->pdev, mapping,
+				dma_unmap_page(&ap->pdev->dev, mapping,
 					       ACE_MINI_BUFSIZE,
-					       PCI_DMA_FROMDEVICE);
+					       DMA_FROM_DEVICE);
 
 				ap->rx_mini_ring[i].size = 0;
 				ap->skb->rx_mini_skbuff[i].skb = NULL;
@@ -684,9 +680,8 @@ static void acenic_remove_one(struct pci_dev *pdev)
 
 			ringp = &ap->skb->rx_jumbo_skbuff[i];
 			mapping = dma_unmap_addr(ringp, mapping);
-			pci_unmap_page(ap->pdev, mapping,
-				       ACE_JUMBO_BUFSIZE,
-				       PCI_DMA_FROMDEVICE);
+			dma_unmap_page(&ap->pdev->dev, mapping,
+				       ACE_JUMBO_BUFSIZE, DMA_FROM_DEVICE);
 
 			ap->rx_jumbo_ring[i].size = 0;
 			ap->skb->rx_jumbo_skbuff[i].skb = NULL;
@@ -716,8 +711,8 @@ static void ace_free_descriptors(struct net_device *dev)
 			 RX_JUMBO_RING_ENTRIES +
 			 RX_MINI_RING_ENTRIES +
 			 RX_RETURN_RING_ENTRIES));
-		pci_free_consistent(ap->pdev, size, ap->rx_std_ring,
-				    ap->rx_ring_base_dma);
+		dma_free_coherent(&ap->pdev->dev, size, ap->rx_std_ring,
+				  ap->rx_ring_base_dma);
 		ap->rx_std_ring = NULL;
 		ap->rx_jumbo_ring = NULL;
 		ap->rx_mini_ring = NULL;
@@ -725,31 +720,30 @@ static void ace_free_descriptors(struct net_device *dev)
 	}
 	if (ap->evt_ring != NULL) {
 		size = (sizeof(struct event) * EVT_RING_ENTRIES);
-		pci_free_consistent(ap->pdev, size, ap->evt_ring,
-				    ap->evt_ring_dma);
+		dma_free_coherent(&ap->pdev->dev, size, ap->evt_ring,
+				  ap->evt_ring_dma);
 		ap->evt_ring = NULL;
 	}
 	if (ap->tx_ring != NULL && !ACE_IS_TIGON_I(ap)) {
 		size = (sizeof(struct tx_desc) * MAX_TX_RING_ENTRIES);
-		pci_free_consistent(ap->pdev, size, ap->tx_ring,
-				    ap->tx_ring_dma);
+		dma_free_coherent(&ap->pdev->dev, size, ap->tx_ring,
+				  ap->tx_ring_dma);
 	}
 	ap->tx_ring = NULL;
 
 	if (ap->evt_prd != NULL) {
-		pci_free_consistent(ap->pdev, sizeof(u32),
-				    (void *)ap->evt_prd, ap->evt_prd_dma);
+		dma_free_coherent(&ap->pdev->dev, sizeof(u32),
+				  (void *)ap->evt_prd, ap->evt_prd_dma);
 		ap->evt_prd = NULL;
 	}
 	if (ap->rx_ret_prd != NULL) {
-		pci_free_consistent(ap->pdev, sizeof(u32),
-				    (void *)ap->rx_ret_prd,
-				    ap->rx_ret_prd_dma);
+		dma_free_coherent(&ap->pdev->dev, sizeof(u32),
+				  (void *)ap->rx_ret_prd, ap->rx_ret_prd_dma);
 		ap->rx_ret_prd = NULL;
 	}
 	if (ap->tx_csm != NULL) {
-		pci_free_consistent(ap->pdev, sizeof(u32),
-				    (void *)ap->tx_csm, ap->tx_csm_dma);
+		dma_free_coherent(&ap->pdev->dev, sizeof(u32),
+				  (void *)ap->tx_csm, ap->tx_csm_dma);
 		ap->tx_csm = NULL;
 	}
 }
@@ -766,8 +760,8 @@ static int ace_allocate_descriptors(struct net_device *dev)
 		 RX_MINI_RING_ENTRIES +
 		 RX_RETURN_RING_ENTRIES));
 
-	ap->rx_std_ring = pci_alloc_consistent(ap->pdev, size,
-					       &ap->rx_ring_base_dma);
+	ap->rx_std_ring = dma_alloc_coherent(&ap->pdev->dev, size,
+					     &ap->rx_ring_base_dma, GFP_KERNEL);
 	if (ap->rx_std_ring == NULL)
 		goto fail;
 
@@ -777,7 +771,8 @@ static int ace_allocate_descriptors(struct net_device *dev)
 
 	size = (sizeof(struct event) * EVT_RING_ENTRIES);
 
-	ap->evt_ring = pci_alloc_consistent(ap->pdev, size, &ap->evt_ring_dma);
+	ap->evt_ring = dma_alloc_coherent(&ap->pdev->dev, size,
+					  &ap->evt_ring_dma, GFP_KERNEL);
 
 	if (ap->evt_ring == NULL)
 		goto fail;
@@ -789,25 +784,25 @@ static int ace_allocate_descriptors(struct net_device *dev)
 	if (!ACE_IS_TIGON_I(ap)) {
 		size = (sizeof(struct tx_desc) * MAX_TX_RING_ENTRIES);
 
-		ap->tx_ring = pci_alloc_consistent(ap->pdev, size,
-						   &ap->tx_ring_dma);
+		ap->tx_ring = dma_alloc_coherent(&ap->pdev->dev, size,
+						 &ap->tx_ring_dma, GFP_KERNEL);
 
 		if (ap->tx_ring == NULL)
 			goto fail;
 	}
 
-	ap->evt_prd = pci_alloc_consistent(ap->pdev, sizeof(u32),
-					   &ap->evt_prd_dma);
+	ap->evt_prd = dma_alloc_coherent(&ap->pdev->dev, sizeof(u32),
+					 &ap->evt_prd_dma, GFP_KERNEL);
 	if (ap->evt_prd == NULL)
 		goto fail;
 
-	ap->rx_ret_prd = pci_alloc_consistent(ap->pdev, sizeof(u32),
-					      &ap->rx_ret_prd_dma);
+	ap->rx_ret_prd = dma_alloc_coherent(&ap->pdev->dev, sizeof(u32),
+					    &ap->rx_ret_prd_dma, GFP_KERNEL);
 	if (ap->rx_ret_prd == NULL)
 		goto fail;
 
-	ap->tx_csm = pci_alloc_consistent(ap->pdev, sizeof(u32),
-					  &ap->tx_csm_dma);
+	ap->tx_csm = dma_alloc_coherent(&ap->pdev->dev, sizeof(u32),
+					&ap->tx_csm_dma, GFP_KERNEL);
 	if (ap->tx_csm == NULL)
 		goto fail;
 
@@ -833,8 +828,8 @@ static void ace_init_cleanup(struct net_device *dev)
 	ace_free_descriptors(dev);
 
 	if (ap->info)
-		pci_free_consistent(ap->pdev, sizeof(struct ace_info),
-				    ap->info, ap->info_dma);
+		dma_free_coherent(&ap->pdev->dev, sizeof(struct ace_info),
+				  ap->info, ap->info_dma);
 	kfree(ap->skb);
 	kfree(ap->trace_buf);
 
@@ -873,6 +868,7 @@ static int ace_init(struct net_device *dev)
 	int board_idx, ecode = 0;
 	short i;
 	unsigned char cache_size;
+	u8 addr[ETH_ALEN];
 
 	ap = netdev_priv(dev);
 	regs = ap->regs;
@@ -992,12 +988,13 @@ static int ace_init(struct net_device *dev)
 	writel(mac1, &regs->MacAddrHi);
 	writel(mac2, &regs->MacAddrLo);
 
-	dev->dev_addr[0] = (mac1 >> 8) & 0xff;
-	dev->dev_addr[1] = mac1 & 0xff;
-	dev->dev_addr[2] = (mac2 >> 24) & 0xff;
-	dev->dev_addr[3] = (mac2 >> 16) & 0xff;
-	dev->dev_addr[4] = (mac2 >> 8) & 0xff;
-	dev->dev_addr[5] = mac2 & 0xff;
+	addr[0] = (mac1 >> 8) & 0xff;
+	addr[1] = mac1 & 0xff;
+	addr[2] = (mac2 >> 24) & 0xff;
+	addr[3] = (mac2 >> 16) & 0xff;
+	addr[4] = (mac2 >> 8) & 0xff;
+	addr[5] = mac2 & 0xff;
+	eth_hw_addr_set(dev, addr);
 
 	printk("MAC: %pM\n", dev->dev_addr);
 
@@ -1132,11 +1129,7 @@ static int ace_init(struct net_device *dev)
 	/*
 	 * Configure DMA attributes.
 	 */
-	if (!pci_set_dma_mask(pdev, DMA_BIT_MASK(64))) {
-		ap->pci_using_dac = 1;
-	} else if (!pci_set_dma_mask(pdev, DMA_BIT_MASK(32))) {
-		ap->pci_using_dac = 0;
-	} else {
+	if (dma_set_mask(&pdev->dev, DMA_BIT_MASK(64))) {
 		ecode = -ENODEV;
 		goto init_error;
 	}
@@ -1146,8 +1139,8 @@ static int ace_init(struct net_device *dev)
 	 * and the control blocks for the transmit and receive rings
 	 * as they need to be setup once and for all.
 	 */
-	if (!(info = pci_alloc_consistent(ap->pdev, sizeof(struct ace_info),
-					  &ap->info_dma))) {
+	if (!(info = dma_alloc_coherent(&ap->pdev->dev, sizeof(struct ace_info),
+					&ap->info_dma, GFP_KERNEL))) {
 		ecode = -EAGAIN;
 		goto init_error;
 	}
@@ -1156,7 +1149,7 @@ static int ace_init(struct net_device *dev)
 	/*
 	 * Get the memory for the skb rings.
 	 */
-	if (!(ap->skb = kmalloc(sizeof(struct ace_skb), GFP_KERNEL))) {
+	if (!(ap->skb = kzalloc(sizeof(struct ace_skb), GFP_KERNEL))) {
 		ecode = -EAGAIN;
 		goto init_error;
 	}
@@ -1177,9 +1170,6 @@ static int ace_init(struct net_device *dev)
 	ap->last_mini_rx = 0;
 #endif
 
-	memset(ap->info, 0, sizeof(struct ace_info));
-	memset(ap->skb, 0, sizeof(struct ace_skb));
-
 	ecode = ace_load_firmware(dev);
 	if (ecode)
 		goto init_error;
@@ -1436,13 +1426,13 @@ static int ace_init(struct net_device *dev)
 	ace_set_txprd(regs, ap, 0);
 	writel(0, &regs->RxRetCsm);
 
-       /*
-	* Enable DMA engine now.
-	* If we do this sooner, Mckinley box pukes.
-	* I assume it's because Tigon II DMA engine wants to check
-	* *something* even before the CPU is started.
-	*/
-       writel(1, &regs->AssistState);  /* enable DMA */
+	/*
+	 * Enable DMA engine now.
+	 * If we do this sooner, Mckinley box pukes.
+	 * I assume it's because Tigon II DMA engine wants to check
+	 * *something* even before the CPU is started.
+	 */
+	writel(1, &regs->AssistState);  /* enable DMA */
 
 	/*
 	 * Start the NIC CPU
@@ -1545,7 +1535,7 @@ static void ace_set_rxtx_parms(struct net_device *dev, int jumbo)
 }
 
 
-static void ace_watchdog(struct net_device *data)
+static void ace_watchdog(struct net_device *data, unsigned int txqueue)
 {
 	struct net_device *dev = data;
 	struct ace_private *ap = netdev_priv(dev);
@@ -1570,10 +1560,10 @@ static void ace_watchdog(struct net_device *data)
 }
 
 
-static void ace_tasklet(unsigned long arg)
+static void ace_bh_work(struct work_struct *work)
 {
-	struct net_device *dev = (struct net_device *) arg;
-	struct ace_private *ap = netdev_priv(dev);
+	struct ace_private *ap = from_work(ap, work, ace_bh_work);
+	struct net_device *dev = ap->ndev;
 	int cur_size;
 
 	cur_size = atomic_read(&ap->cur_rx_bufs);
@@ -1605,7 +1595,7 @@ static void ace_tasklet(unsigned long arg)
 #endif
 		ace_load_jumbo_rx_ring(dev, RX_JUMBO_SIZE - cur_size);
 	}
-	ap->tasklet_pending = 0;
+	ap->bh_work_pending = 0;
 }
 
 
@@ -1627,7 +1617,7 @@ static void ace_dump_trace(struct ace_private *ap)
  *
  * Loading rings is safe without holding the spin lock since this is
  * done only before the device is enabled, thus no interrupts are
- * generated and by the interrupt handler/tasklet handler.
+ * generated and by the interrupt handler/bh handler.
  */
 static void ace_load_std_rx_ring(struct net_device *dev, int nr_bufs)
 {
@@ -1649,10 +1639,10 @@ static void ace_load_std_rx_ring(struct net_device *dev, int nr_bufs)
 		if (!skb)
 			break;
 
-		mapping = pci_map_page(ap->pdev, virt_to_page(skb->data),
+		mapping = dma_map_page(&ap->pdev->dev,
+				       virt_to_page(skb->data),
 				       offset_in_page(skb->data),
-				       ACE_STD_BUFSIZE,
-				       PCI_DMA_FROMDEVICE);
+				       ACE_STD_BUFSIZE, DMA_FROM_DEVICE);
 		ap->skb->rx_std_skbuff[idx].skb = skb;
 		dma_unmap_addr_set(&ap->skb->rx_std_skbuff[idx],
 				   mapping, mapping);
@@ -1710,10 +1700,10 @@ static void ace_load_mini_rx_ring(struct net_device *dev, int nr_bufs)
 		if (!skb)
 			break;
 
-		mapping = pci_map_page(ap->pdev, virt_to_page(skb->data),
+		mapping = dma_map_page(&ap->pdev->dev,
+				       virt_to_page(skb->data),
 				       offset_in_page(skb->data),
-				       ACE_MINI_BUFSIZE,
-				       PCI_DMA_FROMDEVICE);
+				       ACE_MINI_BUFSIZE, DMA_FROM_DEVICE);
 		ap->skb->rx_mini_skbuff[idx].skb = skb;
 		dma_unmap_addr_set(&ap->skb->rx_mini_skbuff[idx],
 				   mapping, mapping);
@@ -1766,10 +1756,10 @@ static void ace_load_jumbo_rx_ring(struct net_device *dev, int nr_bufs)
 		if (!skb)
 			break;
 
-		mapping = pci_map_page(ap->pdev, virt_to_page(skb->data),
+		mapping = dma_map_page(&ap->pdev->dev,
+				       virt_to_page(skb->data),
 				       offset_in_page(skb->data),
-				       ACE_JUMBO_BUFSIZE,
-				       PCI_DMA_FROMDEVICE);
+				       ACE_JUMBO_BUFSIZE, DMA_FROM_DEVICE);
 		ap->skb->rx_jumbo_skbuff[idx].skb = skb;
 		dma_unmap_addr_set(&ap->skb->rx_jumbo_skbuff[idx],
 				   mapping, mapping);
@@ -1890,16 +1880,16 @@ static u32 ace_handle_event(struct net_device *dev, u32 evtcsm, u32 evtprd)
 				}
 			}
 
- 			if (ACE_IS_TIGON_I(ap)) {
- 				struct cmd cmd;
- 				cmd.evt = C_SET_RX_JUMBO_PRD_IDX;
- 				cmd.code = 0;
- 				cmd.idx = 0;
- 				ace_issue_cmd(ap->regs, &cmd);
- 			} else {
- 				writel(0, &((ap->regs)->RxJumboPrd));
- 				wmb();
- 			}
+			if (ACE_IS_TIGON_I(ap)) {
+				struct cmd cmd;
+				cmd.evt = C_SET_RX_JUMBO_PRD_IDX;
+				cmd.code = 0;
+				cmd.idx = 0;
+				ace_issue_cmd(ap->regs, &cmd);
+			} else {
+				writel(0, &((ap->regs)->RxJumboPrd));
+				wmb();
+			}
 
 			ap->jumbo = 0;
 			ap->rx_jumbo_skbprd = 0;
@@ -1933,7 +1923,7 @@ static void ace_rx_int(struct net_device *dev, u32 rxretprd, u32 rxretcsm)
 	while (idx != rxretprd) {
 		struct ring_info *rip;
 		struct sk_buff *skb;
-		struct rx_desc *rxdesc, *retdesc;
+		struct rx_desc *retdesc;
 		u32 skbidx;
 		int bd_flags, desc_type, mapsize;
 		u16 csum;
@@ -1959,19 +1949,16 @@ static void ace_rx_int(struct net_device *dev, u32 rxretprd, u32 rxretcsm)
 		case 0:
 			rip = &ap->skb->rx_std_skbuff[skbidx];
 			mapsize = ACE_STD_BUFSIZE;
-			rxdesc = &ap->rx_std_ring[skbidx];
 			std_count++;
 			break;
 		case BD_FLG_JUMBO:
 			rip = &ap->skb->rx_jumbo_skbuff[skbidx];
 			mapsize = ACE_JUMBO_BUFSIZE;
-			rxdesc = &ap->rx_jumbo_ring[skbidx];
 			atomic_dec(&ap->cur_jumbo_bufs);
 			break;
 		case BD_FLG_MINI:
 			rip = &ap->skb->rx_mini_skbuff[skbidx];
 			mapsize = ACE_MINI_BUFSIZE;
-			rxdesc = &ap->rx_mini_ring[skbidx];
 			mini_count++;
 			break;
 		default:
@@ -1983,10 +1970,8 @@ static void ace_rx_int(struct net_device *dev, u32 rxretprd, u32 rxretcsm)
 
 		skb = rip->skb;
 		rip->skb = NULL;
-		pci_unmap_page(ap->pdev,
-			       dma_unmap_addr(rip, mapping),
-			       mapsize,
-			       PCI_DMA_FROMDEVICE);
+		dma_unmap_page(&ap->pdev->dev, dma_unmap_addr(rip, mapping),
+			       mapsize, DMA_FROM_DEVICE);
 		skb_put(skb, retdesc->size);
 
 		/*
@@ -2052,16 +2037,17 @@ static inline void ace_tx_int(struct net_device *dev,
 		skb = info->skb;
 
 		if (dma_unmap_len(info, maplen)) {
-			pci_unmap_page(ap->pdev, dma_unmap_addr(info, mapping),
+			dma_unmap_page(&ap->pdev->dev,
+				       dma_unmap_addr(info, mapping),
 				       dma_unmap_len(info, maplen),
-				       PCI_DMA_TODEVICE);
+				       DMA_TO_DEVICE);
 			dma_unmap_len_set(info, maplen, 0);
 		}
 
 		if (skb) {
 			dev->stats.tx_packets++;
 			dev->stats.tx_bytes += skb->len;
-			dev_kfree_skb_irq(skb);
+			dev_consume_skb_irq(skb);
 			info->skb = NULL;
 		}
 
@@ -2174,7 +2160,7 @@ static irqreturn_t ace_interrupt(int irq, void *dev_id)
 	 */
 	if (netif_running(dev)) {
 		int cur_size;
-		int run_tasklet = 0;
+		int run_bh_work = 0;
 
 		cur_size = atomic_read(&ap->cur_rx_bufs);
 		if (cur_size < RX_LOW_STD_THRES) {
@@ -2186,7 +2172,7 @@ static irqreturn_t ace_interrupt(int irq, void *dev_id)
 				ace_load_std_rx_ring(dev,
 						     RX_RING_SIZE - cur_size);
 			} else
-				run_tasklet = 1;
+				run_bh_work = 1;
 		}
 
 		if (!ACE_IS_TIGON_I(ap)) {
@@ -2202,7 +2188,7 @@ static irqreturn_t ace_interrupt(int irq, void *dev_id)
 					ace_load_mini_rx_ring(dev,
 							      RX_MINI_SIZE - cur_size);
 				} else
-					run_tasklet = 1;
+					run_bh_work = 1;
 			}
 		}
 
@@ -2219,12 +2205,12 @@ static irqreturn_t ace_interrupt(int irq, void *dev_id)
 					ace_load_jumbo_rx_ring(dev,
 							       RX_JUMBO_SIZE - cur_size);
 				} else
-					run_tasklet = 1;
+					run_bh_work = 1;
 			}
 		}
-		if (run_tasklet && !ap->tasklet_pending) {
-			ap->tasklet_pending = 1;
-			tasklet_schedule(&ap->ace_tasklet);
+		if (run_bh_work && !ap->bh_work_pending) {
+			ap->bh_work_pending = 1;
+			queue_work(system_bh_wq, &ap->ace_bh_work);
 		}
 	}
 
@@ -2281,7 +2267,7 @@ static int ace_open(struct net_device *dev)
 	/*
 	 * Setup the bottom half rx ring refill handler
 	 */
-	tasklet_init(&ap->ace_tasklet, ace_tasklet, (unsigned long)dev);
+	INIT_WORK(&ap->ace_bh_work, ace_bh_work);
 	return 0;
 }
 
@@ -2315,7 +2301,7 @@ static int ace_close(struct net_device *dev)
 	cmd.idx = 0;
 	ace_issue_cmd(regs, &cmd);
 
-	tasklet_kill(&ap->ace_tasklet);
+	cancel_work_sync(&ap->ace_bh_work);
 
 	/*
 	 * Make sure one CPU is not processing packets while
@@ -2343,9 +2329,10 @@ static int ace_close(struct net_device *dev)
 			} else
 				memset(ap->tx_ring + i, 0,
 				       sizeof(struct tx_desc));
-			pci_unmap_page(ap->pdev, dma_unmap_addr(info, mapping),
+			dma_unmap_page(&ap->pdev->dev,
+				       dma_unmap_addr(info, mapping),
 				       dma_unmap_len(info, maplen),
-				       PCI_DMA_TODEVICE);
+				       DMA_TO_DEVICE);
 			dma_unmap_len_set(info, maplen, 0);
 		}
 		if (skb) {
@@ -2375,9 +2362,9 @@ ace_map_tx_skb(struct ace_private *ap, struct sk_buff *skb,
 	dma_addr_t mapping;
 	struct tx_ring_info *info;
 
-	mapping = pci_map_page(ap->pdev, virt_to_page(skb->data),
-			       offset_in_page(skb->data),
-			       skb->len, PCI_DMA_TODEVICE);
+	mapping = dma_map_page(&ap->pdev->dev, virt_to_page(skb->data),
+			       offset_in_page(skb->data), skb->len,
+			       DMA_TO_DEVICE);
 
 	info = ap->skb->tx_skbuff + idx;
 	info->skb = tail;
@@ -2448,7 +2435,7 @@ restart:
 	} else {
 		dma_addr_t mapping;
 		u32 vlan_tag = 0;
-		int i, len = 0;
+		int i;
 
 		mapping = ace_map_tx_skb(ap, skb, NULL, idx);
 		flagsize = (skb_headlen(skb) << 16);
@@ -2467,7 +2454,6 @@ restart:
 			const skb_frag_t *frag = &skb_shinfo(skb)->frags[i];
 			struct tx_ring_info *info;
 
-			len += skb_frag_size(frag);
 			info = ap->skb->tx_skbuff + idx;
 			desc = ap->tx_ring + idx;
 
@@ -2499,9 +2485,9 @@ restart:
 		}
 	}
 
- 	wmb();
- 	ap->tx_prd = idx;
- 	ace_set_txprd(regs, ap, idx);
+	wmb();
+	ap->tx_prd = idx;
+	ace_set_txprd(regs, ap, idx);
 
 	if (flagsize & BD_FLG_COAL_NOW) {
 		netif_stop_queue(dev);
@@ -2553,7 +2539,7 @@ static int ace_change_mtu(struct net_device *dev, int new_mtu)
 	struct ace_regs __iomem *regs = ap->regs;
 
 	writel(new_mtu + ETH_HLEN + 4, &regs->IfMtu);
-	dev->mtu = new_mtu;
+	WRITE_ONCE(dev->mtu, new_mtu);
 
 	if (new_mtu > ACE_STD_MTU) {
 		if (!(ap->jumbo)) {
@@ -2704,13 +2690,12 @@ static void ace_get_drvinfo(struct net_device *dev,
 {
 	struct ace_private *ap = netdev_priv(dev);
 
-	strlcpy(info->driver, "acenic", sizeof(info->driver));
-	snprintf(info->version, sizeof(info->version), "%i.%i.%i",
-		 ap->firmware_major, ap->firmware_minor,
-		 ap->firmware_fix);
+	strscpy(info->driver, "acenic", sizeof(info->driver));
+	snprintf(info->fw_version, sizeof(info->version), "%i.%i.%i",
+		 ap->firmware_major, ap->firmware_minor, ap->firmware_fix);
 
 	if (ap->pdev)
-		strlcpy(info->bus_info, pci_name(ap->pdev),
+		strscpy(info->bus_info, pci_name(ap->pdev),
 			sizeof(info->bus_info));
 
 }
@@ -2723,15 +2708,15 @@ static int ace_set_mac_addr(struct net_device *dev, void *p)
 	struct ace_private *ap = netdev_priv(dev);
 	struct ace_regs __iomem *regs = ap->regs;
 	struct sockaddr *addr=p;
-	u8 *da;
+	const u8 *da;
 	struct cmd cmd;
 
 	if(netif_running(dev))
 		return -EBUSY;
 
-	memcpy(dev->dev_addr, addr->sa_data,dev->addr_len);
+	eth_hw_addr_set(dev, addr->sa_data);
 
-	da = (u8 *)dev->dev_addr;
+	da = (const u8 *)dev->dev_addr;
 
 	writel(da[0] << 8 | da[1], &regs->MacAddrHi);
 	writel((da[2] << 24) | (da[3] << 16) | (da[4] << 8) | da[5],
diff --git a/drivers/net/ethernet/alteon/acenic.h b/drivers/net/ethernet/alteon/acenic.h
index 51c486cfbb8c..0e45a97b9c9b 100644
--- a/drivers/net/ethernet/alteon/acenic.h
+++ b/drivers/net/ethernet/alteon/acenic.h
@@ -1,7 +1,8 @@
+/* SPDX-License-Identifier: GPL-2.0 */
 #ifndef _ACENIC_H_
 #define _ACENIC_H_
 #include <linux/interrupt.h>
-
+#include <linux/workqueue.h>
 
 /*
  * Generate TX index update each time, when TX ring is closed.
@@ -632,6 +633,7 @@ struct ace_skb
  */
 struct ace_private
 {
+	struct net_device	*ndev;		/* backpointer */
 	struct ace_info		*info;
 	struct ace_regs	__iomem	*regs;		/* register base */
 	struct ace_skb		*skb;
@@ -665,8 +667,8 @@ struct ace_private
 	struct rx_desc		*rx_mini_ring;
 	struct rx_desc		*rx_return_ring;
 
-	int			tasklet_pending, jumbo;
-	struct tasklet_struct	ace_tasklet;
+	int			bh_work_pending, jumbo;
+	struct work_struct	ace_bh_work;
 
 	struct event		*evt_ring;
 
@@ -690,7 +692,6 @@ struct ace_private
 				__attribute__ ((aligned (SMP_CACHE_BYTES)));
 	u32			last_tx, last_std_rx, last_mini_rx;
 #endif
-	int			pci_using_dac;
 	u8			firmware_major;
 	u8			firmware_minor;
 	u8			firmware_fix;
@@ -775,7 +776,7 @@ static int ace_open(struct net_device *dev);
 static netdev_tx_t ace_start_xmit(struct sk_buff *skb,
 				  struct net_device *dev);
 static int ace_close(struct net_device *dev);
-static void ace_tasklet(unsigned long dev);
+static void ace_bh_work(struct work_struct *work);
 static void ace_dump_trace(struct ace_private *ap);
 static void ace_set_multicast_list(struct net_device *dev);
 static int ace_change_mtu(struct net_device *dev, int new_mtu);
diff --git a/drivers/net/ethernet/altera/Kconfig b/drivers/net/ethernet/altera/Kconfig
index fdddba51473e..4ef819a9a1ad 100644
--- a/drivers/net/ethernet/altera/Kconfig
+++ b/drivers/net/ethernet/altera/Kconfig
@@ -1,8 +1,14 @@
+# SPDX-License-Identifier: GPL-2.0-only
 config ALTERA_TSE
 	tristate "Altera Triple-Speed Ethernet MAC support"
 	depends on HAS_DMA
+	depends on HAS_IOMEM
 	select PHYLIB
-	---help---
+	select PHYLINK
+	select PCS_LYNX
+	select MDIO_REGMAP
+	select REGMAP_MMIO
+	help
 	  This driver supports the Altera Triple-Speed (TSE) Ethernet MAC.
 
 	  To compile this driver as a module, choose M here. The module
diff --git a/drivers/net/ethernet/altera/Makefile b/drivers/net/ethernet/altera/Makefile
index d4a187e45369..a52db80aee9f 100644
--- a/drivers/net/ethernet/altera/Makefile
+++ b/drivers/net/ethernet/altera/Makefile
@@ -1,3 +1,4 @@
+# SPDX-License-Identifier: GPL-2.0-only
 #
 # Makefile for the Altera device drivers.
 #
diff --git a/drivers/net/ethernet/altera/altera_msgdma.c b/drivers/net/ethernet/altera/altera_msgdma.c
index 0fb986ba3290..ac1efd08267a 100644
--- a/drivers/net/ethernet/altera/altera_msgdma.c
+++ b/drivers/net/ethernet/altera/altera_msgdma.c
@@ -1,17 +1,6 @@
+// SPDX-License-Identifier: GPL-2.0-only
 /* Altera TSE SGDMA and MSGDMA Linux driver
  * Copyright (C) 2014 Altera Corporation. All rights reserved
- *
- * This program is free software; you can redistribute it and/or modify it
- * under the terms and conditions of the GNU General Public License,
- * version 2, as published by the Free Software Foundation.
- *
- * This program is distributed in the hope it will be useful, but WITHOUT
- * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
- * FITNESS FOR A PARTICULAR PURPOSE.  See the GNU General Public License for
- * more details.
- *
- * You should have received a copy of the GNU General Public License along with
- * this program.  If not, see <http://www.gnu.org/licenses/>.
  */
 
 #include <linux/netdevice.h>
@@ -145,7 +134,8 @@ u32 msgdma_tx_completions(struct altera_tse_private *priv)
 			& 0xffff;
 
 	if (inuse) { /* Tx FIFO is not empty */
-		ready = priv->tx_prod - priv->tx_cons - inuse - 1;
+		ready = max_t(int,
+			      priv->tx_prod - priv->tx_cons - inuse - 1, 0);
 	} else {
 		/* Check for buffered last packet */
 		status = csrrd32(priv->tx_dma_csr, msgdma_csroffs(status));
diff --git a/drivers/net/ethernet/altera/altera_msgdma.h b/drivers/net/ethernet/altera/altera_msgdma.h
index 42cf61c81057..9813fbfff4d3 100644
--- a/drivers/net/ethernet/altera/altera_msgdma.h
+++ b/drivers/net/ethernet/altera/altera_msgdma.h
@@ -1,17 +1,6 @@
+/* SPDX-License-Identifier: GPL-2.0-only */
 /* Altera TSE SGDMA and MSGDMA Linux driver
  * Copyright (C) 2014 Altera Corporation. All rights reserved
- *
- * This program is free software; you can redistribute it and/or modify it
- * under the terms and conditions of the GNU General Public License,
- * version 2, as published by the Free Software Foundation.
- *
- * This program is distributed in the hope it will be useful, but WITHOUT
- * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
- * FITNESS FOR A PARTICULAR PURPOSE.  See the GNU General Public License for
- * more details.
- *
- * You should have received a copy of the GNU General Public License along with
- * this program.  If not, see <http://www.gnu.org/licenses/>.
  */
 
 #ifndef __ALTERA_MSGDMA_H__
diff --git a/drivers/net/ethernet/altera/altera_msgdmahw.h b/drivers/net/ethernet/altera/altera_msgdmahw.h
index 89cd11d86642..019f5a12630e 100644
--- a/drivers/net/ethernet/altera/altera_msgdmahw.h
+++ b/drivers/net/ethernet/altera/altera_msgdmahw.h
@@ -1,17 +1,6 @@
+/* SPDX-License-Identifier: GPL-2.0-only */
 /* Altera TSE SGDMA and MSGDMA Linux driver
  * Copyright (C) 2014 Altera Corporation. All rights reserved
- *
- * This program is free software; you can redistribute it and/or modify it
- * under the terms and conditions of the GNU General Public License,
- * version 2, as published by the Free Software Foundation.
- *
- * This program is distributed in the hope it will be useful, but WITHOUT
- * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
- * FITNESS FOR A PARTICULAR PURPOSE.  See the GNU General Public License for
- * more details.
- *
- * You should have received a copy of the GNU General Public License along with
- * this program.  If not, see <http://www.gnu.org/licenses/>.
  */
 
 #ifndef __ALTERA_MSGDMAHW_H__
diff --git a/drivers/net/ethernet/altera/altera_sgdma.c b/drivers/net/ethernet/altera/altera_sgdma.c
index 88ef67a998b4..7f247ccbe6ba 100644
--- a/drivers/net/ethernet/altera/altera_sgdma.c
+++ b/drivers/net/ethernet/altera/altera_sgdma.c
@@ -1,17 +1,6 @@
+// SPDX-License-Identifier: GPL-2.0-only
 /* Altera TSE SGDMA and MSGDMA Linux driver
  * Copyright (C) 2014 Altera Corporation. All rights reserved
- *
- * This program is free software; you can redistribute it and/or modify it
- * under the terms and conditions of the GNU General Public License,
- * version 2, as published by the Free Software Foundation.
- *
- * This program is distributed in the hope it will be useful, but WITHOUT
- * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
- * FITNESS FOR A PARTICULAR PURPOSE.  See the GNU General Public License for
- * more details.
- *
- * You should have received a copy of the GNU General Public License along with
- * this program.  If not, see <http://www.gnu.org/licenses/>.
  */
 
 #include <linux/list.h>
@@ -524,7 +513,7 @@ static int sgdma_txbusy(struct altera_tse_private *priv)
 {
 	int delay = 0;
 
-	/* if DMA is busy, wait for current transactino to finish */
+	/* if DMA is busy, wait for current transaction to finish */
 	while ((csrrd32(priv->tx_dma_csr, sgdma_csroffs(status))
 		& SGDMA_STSREG_BUSY) && (delay++ < 100))
 		udelay(1);
diff --git a/drivers/net/ethernet/altera/altera_sgdma.h b/drivers/net/ethernet/altera/altera_sgdma.h
index 584977e29ef9..08afe1c9994f 100644
--- a/drivers/net/ethernet/altera/altera_sgdma.h
+++ b/drivers/net/ethernet/altera/altera_sgdma.h
@@ -1,17 +1,6 @@
+/* SPDX-License-Identifier: GPL-2.0-only */
 /* Altera TSE SGDMA and MSGDMA Linux driver
  * Copyright (C) 2014 Altera Corporation. All rights reserved
- *
- * This program is free software; you can redistribute it and/or modify it
- * under the terms and conditions of the GNU General Public License,
- * version 2, as published by the Free Software Foundation.
- *
- * This program is distributed in the hope it will be useful, but WITHOUT
- * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
- * FITNESS FOR A PARTICULAR PURPOSE.  See the GNU General Public License for
- * more details.
- *
- * You should have received a copy of the GNU General Public License along with
- * this program.  If not, see <http://www.gnu.org/licenses/>.
  */
 
 #ifndef __ALTERA_SGDMA_H__
diff --git a/drivers/net/ethernet/altera/altera_sgdmahw.h b/drivers/net/ethernet/altera/altera_sgdmahw.h
index bbd52f02330b..3304518e2cbf 100644
--- a/drivers/net/ethernet/altera/altera_sgdmahw.h
+++ b/drivers/net/ethernet/altera/altera_sgdmahw.h
@@ -1,17 +1,6 @@
+/* SPDX-License-Identifier: GPL-2.0-only */
 /* Altera TSE SGDMA and MSGDMA Linux driver
  * Copyright (C) 2014 Altera Corporation. All rights reserved
- *
- * This program is free software; you can redistribute it and/or modify it
- * under the terms and conditions of the GNU General Public License,
- * version 2, as published by the Free Software Foundation.
- *
- * This program is distributed in the hope it will be useful, but WITHOUT
- * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
- * FITNESS FOR A PARTICULAR PURPOSE.  See the GNU General Public License for
- * more details.
- *
- * You should have received a copy of the GNU General Public License along with
- * this program.  If not, see <http://www.gnu.org/licenses/>.
  */
 
 #ifndef __ALTERA_SGDMAHW_H__
diff --git a/drivers/net/ethernet/altera/altera_tse.h b/drivers/net/ethernet/altera/altera_tse.h
index e2feee87180a..82f2363a45cd 100644
--- a/drivers/net/ethernet/altera/altera_tse.h
+++ b/drivers/net/ethernet/altera/altera_tse.h
@@ -1,3 +1,4 @@
+/* SPDX-License-Identifier: GPL-2.0-only */
 /* Altera Triple-Speed Ethernet MAC driver
  * Copyright (C) 2008-2014 Altera Corporation. All rights reserved
  *
@@ -14,18 +15,6 @@
  *
  * Original driver contributed by SLS.
  * Major updates contributed by GlobalLogic
- *
- * This program is free software; you can redistribute it and/or modify it
- * under the terms and conditions of the GNU General Public License,
- * version 2, as published by the Free Software Foundation.
- *
- * This program is distributed in the hope it will be useful, but WITHOUT
- * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
- * FITNESS FOR A PARTICULAR PURPOSE.  See the GNU General Public License for
- * more details.
- *
- * You should have received a copy of the GNU General Public License along with
- * this program.  If not, see <http://www.gnu.org/licenses/>.
  */
 
 #ifndef __ALTERA_TSE_H__
@@ -38,6 +27,7 @@
 #include <linux/list.h>
 #include <linux/netdevice.h>
 #include <linux/phy.h>
+#include <linux/phylink.h>
 
 #define ALTERA_TSE_SW_RESET_WATCHDOG_CNTR	10000
 #define ALTERA_TSE_MAC_FIFO_WIDTH		4	/* TX/RX FIFO width in
@@ -120,17 +110,6 @@
 #define MAC_CMDCFG_DISABLE_READ_TIMEOUT_GET(v)	GET_BIT_VALUE(v, 27)
 #define MAC_CMDCFG_CNT_RESET_GET(v)		GET_BIT_VALUE(v, 31)
 
-/* SGMII PCS register addresses
- */
-#define SGMII_PCS_SCRATCH	0x10
-#define SGMII_PCS_REV		0x11
-#define SGMII_PCS_LINK_TIMER_0	0x12
-#define SGMII_PCS_LINK_TIMER_1	0x13
-#define SGMII_PCS_IF_MODE	0x14
-#define SGMII_PCS_DIS_READ_TO	0x15
-#define SGMII_PCS_READ_TO	0x16
-#define SGMII_PCS_SW_RESET_TIMEOUT 100 /* usecs */
-
 /* MDIO registers within MAC register Space
  */
 struct altera_tse_mdio {
@@ -434,6 +413,9 @@ struct altera_tse_private {
 	void __iomem *tx_dma_csr;
 	void __iomem *tx_dma_desc;
 
+	/* SGMII PCS address space */
+	void __iomem *pcs_base;
+
 	/* Rx buffers queue */
 	struct tse_buffer *rx_ring;
 	u32 rx_cons;
@@ -490,7 +472,11 @@ struct altera_tse_private {
 	/* ethtool msglvl option */
 	u32 msg_enable;
 
-	struct altera_dmaops *dmaops;
+	const struct altera_dmaops *dmaops;
+
+	struct phylink *phylink;
+	struct phylink_config phylink_config;
+	struct phylink_pcs *pcs;
 };
 
 /* Function prototypes
diff --git a/drivers/net/ethernet/altera/altera_tse_ethtool.c b/drivers/net/ethernet/altera/altera_tse_ethtool.c
index 7c367713c3e6..81313c85833e 100644
--- a/drivers/net/ethernet/altera/altera_tse_ethtool.c
+++ b/drivers/net/ethernet/altera/altera_tse_ethtool.c
@@ -1,3 +1,4 @@
+// SPDX-License-Identifier: GPL-2.0-only
 /* Ethtool support for Altera Triple-Speed Ethernet MAC driver
  * Copyright (C) 2008-2014 Altera Corporation. All rights reserved
  *
@@ -13,18 +14,6 @@
  *
  * Original driver contributed by SLS.
  * Major updates contributed by GlobalLogic
- *
- * This program is free software; you can redistribute it and/or modify it
- * under the terms and conditions of the GNU General Public License,
- * version 2, as published by the Free Software Foundation.
- *
- * This program is distributed in the hope it will be useful, but WITHOUT
- * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
- * FITNESS FOR A PARTICULAR PURPOSE.  See the GNU General Public License for
- * more details.
- *
- * You should have received a copy of the GNU General Public License along with
- * this program.  If not, see <http://www.gnu.org/licenses/>.
  */
 
 #include <linux/ethtool.h>
@@ -78,7 +67,6 @@ static void tse_get_drvinfo(struct net_device *dev,
 	u32 rev = ioread32(&priv->mac_dev->megacore_revision);
 
 	strcpy(info->driver, "altera_tse");
-	strcpy(info->version, "v8.0");
 	snprintf(info->fw_version, ETHTOOL_FWVERS_LEN, "v%d.%d",
 		 rev & 0xFFFF, (rev & 0xFFFF0000) >> 16);
 	sprintf(info->bus_info, "platform");
@@ -211,9 +199,9 @@ static int tse_reglen(struct net_device *dev)
 static void tse_get_regs(struct net_device *dev, struct ethtool_regs *regs,
 			 void *regbuf)
 {
-	int i;
 	struct altera_tse_private *priv = netdev_priv(dev);
 	u32 *buf = regbuf;
+	int i;
 
 	/* Set version to a known value, so ethtool knows
 	 * how to do any special formatting of this data.
@@ -233,6 +221,22 @@ static void tse_get_regs(struct net_device *dev, struct ethtool_regs *regs,
 		buf[i] = csrrd32(priv->mac_dev, i * 4);
 }
 
+static int tse_ethtool_set_link_ksettings(struct net_device *dev,
+					  const struct ethtool_link_ksettings *cmd)
+{
+	struct altera_tse_private *priv = netdev_priv(dev);
+
+	return phylink_ethtool_ksettings_set(priv->phylink, cmd);
+}
+
+static int tse_ethtool_get_link_ksettings(struct net_device *dev,
+					  struct ethtool_link_ksettings *cmd)
+{
+	struct altera_tse_private *priv = netdev_priv(dev);
+
+	return phylink_ethtool_ksettings_get(priv->phylink, cmd);
+}
+
 static const struct ethtool_ops tse_ethtool_ops = {
 	.get_drvinfo = tse_get_drvinfo,
 	.get_regs_len = tse_reglen,
@@ -243,8 +247,9 @@ static const struct ethtool_ops tse_ethtool_ops = {
 	.get_ethtool_stats = tse_fill_stats,
 	.get_msglevel = tse_get_msglevel,
 	.set_msglevel = tse_set_msglevel,
-	.get_link_ksettings = phy_ethtool_get_link_ksettings,
-	.set_link_ksettings = phy_ethtool_set_link_ksettings,
+	.get_link_ksettings = tse_ethtool_get_link_ksettings,
+	.set_link_ksettings = tse_ethtool_set_link_ksettings,
+	.get_ts_info = ethtool_op_get_ts_info,
 };
 
 void altera_tse_set_ethtool_ops(struct net_device *netdev)
diff --git a/drivers/net/ethernet/altera/altera_tse_main.c b/drivers/net/ethernet/altera/altera_tse_main.c
index 527908c7e384..3f6204de9e6b 100644
--- a/drivers/net/ethernet/altera/altera_tse_main.c
+++ b/drivers/net/ethernet/altera/altera_tse_main.c
@@ -1,3 +1,4 @@
+// SPDX-License-Identifier: GPL-2.0-only
 /* Altera Triple-Speed Ethernet MAC driver
  * Copyright (C) 2008-2014 Altera Corporation. All rights reserved
  *
@@ -14,18 +15,6 @@
  *
  * Original driver contributed by SLS.
  * Major updates contributed by GlobalLogic
- *
- * This program is free software; you can redistribute it and/or modify it
- * under the terms and conditions of the GNU General Public License,
- * version 2, as published by the Free Software Foundation.
- *
- * This program is distributed in the hope it will be useful, but WITHOUT
- * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
- * FITNESS FOR A PARTICULAR PURPOSE.  See the GNU General Public License for
- * more details.
- *
- * You should have received a copy of the GNU General Public License along with
- * this program.  If not, see <http://www.gnu.org/licenses/>.
  */
 
 #include <linux/atomic.h>
@@ -38,13 +27,16 @@
 #include <linux/kernel.h>
 #include <linux/module.h>
 #include <linux/mii.h>
+#include <linux/mdio/mdio-regmap.h>
 #include <linux/netdevice.h>
-#include <linux/of_device.h>
+#include <linux/of.h>
 #include <linux/of_mdio.h>
 #include <linux/of_net.h>
-#include <linux/of_platform.h>
+#include <linux/pcs-lynx.h>
 #include <linux/phy.h>
 #include <linux/platform_device.h>
+#include <linux/property.h>
+#include <linux/regmap.h>
 #include <linux/skbuff.h>
 #include <asm/cacheflush.h>
 
@@ -56,7 +48,7 @@
 static atomic_t instance_count = ATOMIC_INIT(~0);
 /* Module parameters */
 static int debug = -1;
-module_param(debug, int, S_IRUGO | S_IWUSR);
+module_param(debug, int, 0644);
 MODULE_PARM_DESC(debug, "Message Level (-1: default, 0: no output, 16: all)");
 
 static const u32 default_msg_level = (NETIF_MSG_DRV | NETIF_MSG_PROBE |
@@ -65,12 +57,12 @@ static const u32 default_msg_level = (NETIF_MSG_DRV | NETIF_MSG_PROBE |
 
 #define RX_DESCRIPTORS 64
 static int dma_rx_num = RX_DESCRIPTORS;
-module_param(dma_rx_num, int, S_IRUGO | S_IWUSR);
+module_param(dma_rx_num, int, 0644);
 MODULE_PARM_DESC(dma_rx_num, "Number of descriptors in the RX list");
 
 #define TX_DESCRIPTORS 64
 static int dma_tx_num = TX_DESCRIPTORS;
-module_param(dma_tx_num, int, S_IRUGO | S_IWUSR);
+module_param(dma_tx_num, int, 0644);
 MODULE_PARM_DESC(dma_tx_num, "Number of descriptors in the TX list");
 
 
@@ -83,41 +75,18 @@ MODULE_PARM_DESC(dma_tx_num, "Number of descriptors in the TX list");
  */
 #define ALTERA_RXDMABUFFER_SIZE	2048
 
-/* Allow network stack to resume queueing packets after we've
+/* Allow network stack to resume queuing packets after we've
  * finished transmitting at least 1/4 of the packets in the queue.
  */
 #define TSE_TX_THRESH(x)	(x->tx_ring_size / 4)
 
 #define TXQUEUESTOP_THRESHHOLD	2
 
-static const struct of_device_id altera_tse_ids[];
-
 static inline u32 tse_tx_avail(struct altera_tse_private *priv)
 {
 	return priv->tx_cons + priv->tx_ring_size - priv->tx_prod - 1;
 }
 
-/* PCS Register read/write functions
- */
-static u16 sgmii_pcs_read(struct altera_tse_private *priv, int regnum)
-{
-	return csrrd32(priv->mac_dev,
-		       tse_csroffs(mdio_phy0) + regnum * 4) & 0xffff;
-}
-
-static void sgmii_pcs_write(struct altera_tse_private *priv, int regnum,
-				u16 value)
-{
-	csrwr32(value, priv->mac_dev, tse_csroffs(mdio_phy0) + regnum * 4);
-}
-
-/* Check PCS scratch memory */
-static int sgmii_pcs_scratch_test(struct altera_tse_private *priv, u16 value)
-{
-	sgmii_pcs_write(priv, SGMII_PCS_SCRATCH, value);
-	return (sgmii_pcs_read(priv, SGMII_PCS_SCRATCH) == value);
-}
-
 /* MDIO specific functions
  */
 static int altera_tse_mdio_read(struct mii_bus *bus, int mii_id, int regnum)
@@ -152,10 +121,10 @@ static int altera_tse_mdio_write(struct mii_bus *bus, int mii_id, int regnum,
 static int altera_tse_mdio_create(struct net_device *dev, unsigned int id)
 {
 	struct altera_tse_private *priv = netdev_priv(dev);
-	int ret;
 	struct device_node *mdio_node = NULL;
-	struct mii_bus *mdio = NULL;
 	struct device_node *child_node = NULL;
+	struct mii_bus *mdio = NULL;
+	int ret;
 
 	for_each_child_of_node(priv->device->of_node, child_node) {
 		if (of_device_is_compatible(child_node, "altr,tse-mdio")) {
@@ -174,7 +143,8 @@ static int altera_tse_mdio_create(struct net_device *dev, unsigned int id)
 	mdio = mdiobus_alloc();
 	if (mdio == NULL) {
 		netdev_err(dev, "Error allocating MDIO bus\n");
-		return -ENOMEM;
+		ret = -ENOMEM;
+		goto put_node;
 	}
 
 	mdio->name = ALTERA_TSE_RESOURCE_NAME;
@@ -191,6 +161,7 @@ static int altera_tse_mdio_create(struct net_device *dev, unsigned int id)
 			   mdio->id);
 		goto out_free_mdio;
 	}
+	of_node_put(mdio_node);
 
 	if (netif_msg_drv(priv))
 		netdev_info(dev, "MDIO bus %s: created\n", mdio->id);
@@ -200,6 +171,8 @@ static int altera_tse_mdio_create(struct net_device *dev, unsigned int id)
 out_free_mdio:
 	mdiobus_free(mdio);
 	mdio = NULL;
+put_node:
+	of_node_put(mdio_node);
 	return ret;
 }
 
@@ -243,8 +216,8 @@ static int tse_init_rx_buffer(struct altera_tse_private *priv,
 static void tse_free_rx_buffer(struct altera_tse_private *priv,
 			       struct tse_buffer *rxbuffer)
 {
-	struct sk_buff *skb = rxbuffer->skb;
 	dma_addr_t dma_addr = rxbuffer->dma_addr;
+	struct sk_buff *skb = rxbuffer->skb;
 
 	if (skb != NULL) {
 		if (dma_addr)
@@ -365,6 +338,7 @@ static inline void tse_rx_vlan(struct net_device *dev, struct sk_buff *skb)
 {
 	struct ethhdr *eth_hdr;
 	u16 vid;
+
 	if ((dev->features & NETIF_F_HW_VLAN_CTAG_RX) &&
 	    !__vlan_get_tag(skb, &vid)) {
 		eth_hdr = (struct ethhdr *)skb->data;
@@ -378,10 +352,10 @@ static inline void tse_rx_vlan(struct net_device *dev, struct sk_buff *skb)
  */
 static int tse_rx(struct altera_tse_private *priv, int limit)
 {
-	unsigned int count = 0;
+	unsigned int entry = priv->rx_cons % priv->rx_ring_size;
 	unsigned int next_entry;
+	unsigned int count = 0;
 	struct sk_buff *skb;
-	unsigned int entry = priv->rx_cons % priv->rx_ring_size;
 	u32 rxstatus;
 	u16 pktlength;
 	u16 pktstatus;
@@ -401,7 +375,7 @@ static int tse_rx(struct altera_tse_private *priv, int limit)
 				   "RCV pktstatus %08X pktlength %08X\n",
 				   pktstatus, pktlength);
 
-		/* DMA trasfer from TSE starts with 2 aditional bytes for
+		/* DMA transfer from TSE starts with 2 additional bytes for
 		 * IP payload alignment. Status returned by get_rx_status()
 		 * contains DMA transfer length. Packet is 2 bytes shorter.
 		 */
@@ -455,10 +429,10 @@ static int tse_rx(struct altera_tse_private *priv, int limit)
 static int tse_tx_complete(struct altera_tse_private *priv)
 {
 	unsigned int txsize = priv->tx_ring_size;
-	u32 ready;
-	unsigned int entry;
 	struct tse_buffer *tx_buff;
+	unsigned int entry;
 	int txcomplete = 0;
+	u32 ready;
 
 	spin_lock(&priv->tx_lock);
 
@@ -504,8 +478,8 @@ static int tse_poll(struct napi_struct *napi, int budget)
 {
 	struct altera_tse_private *priv =
 			container_of(napi, struct altera_tse_private, napi);
-	int rxcomplete = 0;
 	unsigned long int flags;
+	int rxcomplete = 0;
 
 	tse_tx_complete(priv);
 
@@ -565,16 +539,16 @@ static irqreturn_t altera_isr(int irq, void *dev_id)
  * physically contiguous fragment starting at
  * skb->data, for length of skb_headlen(skb).
  */
-static int tse_start_xmit(struct sk_buff *skb, struct net_device *dev)
+static netdev_tx_t tse_start_xmit(struct sk_buff *skb, struct net_device *dev)
 {
 	struct altera_tse_private *priv = netdev_priv(dev);
+	unsigned int nopaged_len = skb_headlen(skb);
 	unsigned int txsize = priv->tx_ring_size;
-	unsigned int entry;
-	struct tse_buffer *buffer = NULL;
 	int nfrags = skb_shinfo(skb)->nr_frags;
-	unsigned int nopaged_len = skb_headlen(skb);
-	enum netdev_tx ret = NETDEV_TX_OK;
+	struct tse_buffer *buffer = NULL;
+	netdev_tx_t ret = NETDEV_TX_OK;
 	dma_addr_t dma_addr;
+	unsigned int entry;
 
 	spin_lock_bh(&priv->tx_lock);
 
@@ -626,125 +600,16 @@ out:
 	return ret;
 }
 
-/* Called every time the controller might need to be made
- * aware of new link state.  The PHY code conveys this
- * information through variables in the phydev structure, and this
- * function converts those variables into the appropriate
- * register values, and can bring down the device if needed.
- */
-static void altera_tse_adjust_link(struct net_device *dev)
-{
-	struct altera_tse_private *priv = netdev_priv(dev);
-	struct phy_device *phydev = dev->phydev;
-	int new_state = 0;
-
-	/* only change config if there is a link */
-	spin_lock(&priv->mac_cfg_lock);
-	if (phydev->link) {
-		/* Read old config */
-		u32 cfg_reg = ioread32(&priv->mac_dev->command_config);
-
-		/* Check duplex */
-		if (phydev->duplex != priv->oldduplex) {
-			new_state = 1;
-			if (!(phydev->duplex))
-				cfg_reg |= MAC_CMDCFG_HD_ENA;
-			else
-				cfg_reg &= ~MAC_CMDCFG_HD_ENA;
-
-			netdev_dbg(priv->dev, "%s: Link duplex = 0x%x\n",
-				   dev->name, phydev->duplex);
-
-			priv->oldduplex = phydev->duplex;
-		}
-
-		/* Check speed */
-		if (phydev->speed != priv->oldspeed) {
-			new_state = 1;
-			switch (phydev->speed) {
-			case 1000:
-				cfg_reg |= MAC_CMDCFG_ETH_SPEED;
-				cfg_reg &= ~MAC_CMDCFG_ENA_10;
-				break;
-			case 100:
-				cfg_reg &= ~MAC_CMDCFG_ETH_SPEED;
-				cfg_reg &= ~MAC_CMDCFG_ENA_10;
-				break;
-			case 10:
-				cfg_reg &= ~MAC_CMDCFG_ETH_SPEED;
-				cfg_reg |= MAC_CMDCFG_ENA_10;
-				break;
-			default:
-				if (netif_msg_link(priv))
-					netdev_warn(dev, "Speed (%d) is not 10/100/1000!\n",
-						    phydev->speed);
-				break;
-			}
-			priv->oldspeed = phydev->speed;
-		}
-		iowrite32(cfg_reg, &priv->mac_dev->command_config);
-
-		if (!priv->oldlink) {
-			new_state = 1;
-			priv->oldlink = 1;
-		}
-	} else if (priv->oldlink) {
-		new_state = 1;
-		priv->oldlink = 0;
-		priv->oldspeed = 0;
-		priv->oldduplex = -1;
-	}
-
-	if (new_state && netif_msg_link(priv))
-		phy_print_status(phydev);
-
-	spin_unlock(&priv->mac_cfg_lock);
-}
-static struct phy_device *connect_local_phy(struct net_device *dev)
-{
-	struct altera_tse_private *priv = netdev_priv(dev);
-	struct phy_device *phydev = NULL;
-	char phy_id_fmt[MII_BUS_ID_SIZE + 3];
-
-	if (priv->phy_addr != POLL_PHY) {
-		snprintf(phy_id_fmt, MII_BUS_ID_SIZE + 3, PHY_ID_FMT,
-			 priv->mdio->id, priv->phy_addr);
-
-		netdev_dbg(dev, "trying to attach to %s\n", phy_id_fmt);
-
-		phydev = phy_connect(dev, phy_id_fmt, &altera_tse_adjust_link,
-				     priv->phy_iface);
-		if (IS_ERR(phydev))
-			netdev_err(dev, "Could not attach to PHY\n");
-
-	} else {
-		int ret;
-		phydev = phy_find_first(priv->mdio);
-		if (phydev == NULL) {
-			netdev_err(dev, "No PHY found\n");
-			return phydev;
-		}
-
-		ret = phy_connect_direct(dev, phydev, &altera_tse_adjust_link,
-				priv->phy_iface);
-		if (ret != 0) {
-			netdev_err(dev, "Could not attach to PHY\n");
-			phydev = NULL;
-		}
-	}
-	return phydev;
-}
-
 static int altera_tse_phy_get_addr_mdio_create(struct net_device *dev)
 {
 	struct altera_tse_private *priv = netdev_priv(dev);
 	struct device_node *np = priv->device->of_node;
-	int ret = 0;
+	int ret;
 
-	priv->phy_iface = of_get_phy_mode(np);
+	ret = of_get_phy_mode(np, &priv->phy_iface);
 
 	/* Avoid get phy addr and create mdio if no phy is present */
-	if (!priv->phy_iface)
+	if (ret)
 		return 0;
 
 	/* try to get PHY address from device tree, use PHY autodetection if
@@ -773,95 +638,7 @@ static int altera_tse_phy_get_addr_mdio_create(struct net_device *dev)
 	return 0;
 }
 
-/* Initialize driver's PHY state, and attach to the PHY
- */
-static int init_phy(struct net_device *dev)
-{
-	struct altera_tse_private *priv = netdev_priv(dev);
-	struct phy_device *phydev;
-	struct device_node *phynode;
-	bool fixed_link = false;
-	int rc = 0;
-
-	/* Avoid init phy in case of no phy present */
-	if (!priv->phy_iface)
-		return 0;
-
-	priv->oldlink = 0;
-	priv->oldspeed = 0;
-	priv->oldduplex = -1;
-
-	phynode = of_parse_phandle(priv->device->of_node, "phy-handle", 0);
-
-	if (!phynode) {
-		/* check if a fixed-link is defined in device-tree */
-		if (of_phy_is_fixed_link(priv->device->of_node)) {
-			rc = of_phy_register_fixed_link(priv->device->of_node);
-			if (rc < 0) {
-				netdev_err(dev, "cannot register fixed PHY\n");
-				return rc;
-			}
-
-			/* In the case of a fixed PHY, the DT node associated
-			 * to the PHY is the Ethernet MAC DT node.
-			 */
-			phynode = of_node_get(priv->device->of_node);
-			fixed_link = true;
-
-			netdev_dbg(dev, "fixed-link detected\n");
-			phydev = of_phy_connect(dev, phynode,
-						&altera_tse_adjust_link,
-						0, priv->phy_iface);
-		} else {
-			netdev_dbg(dev, "no phy-handle found\n");
-			if (!priv->mdio) {
-				netdev_err(dev, "No phy-handle nor local mdio specified\n");
-				return -ENODEV;
-			}
-			phydev = connect_local_phy(dev);
-		}
-	} else {
-		netdev_dbg(dev, "phy-handle found\n");
-		phydev = of_phy_connect(dev, phynode,
-			&altera_tse_adjust_link, 0, priv->phy_iface);
-	}
-	of_node_put(phynode);
-
-	if (!phydev) {
-		netdev_err(dev, "Could not find the PHY\n");
-		if (fixed_link)
-			of_phy_deregister_fixed_link(priv->device->of_node);
-		return -ENODEV;
-	}
-
-	/* Stop Advertising 1000BASE Capability if interface is not GMII
-	 * Note: Checkpatch throws CHECKs for the camel case defines below,
-	 * it's ok to ignore.
-	 */
-	if ((priv->phy_iface == PHY_INTERFACE_MODE_MII) ||
-	    (priv->phy_iface == PHY_INTERFACE_MODE_RMII))
-		phydev->advertising &= ~(SUPPORTED_1000baseT_Half |
-					 SUPPORTED_1000baseT_Full);
-
-	/* Broken HW is sometimes missing the pull-up resistor on the
-	 * MDIO line, which results in reads to non-existent devices returning
-	 * 0 rather than 0xffff. Catch this here and treat 0 as a non-existent
-	 * device as well. If a fixed-link is used the phy_id is always 0.
-	 * Note: phydev->phy_id is the result of reading the UID PHY registers.
-	 */
-	if ((phydev->phy_id == 0) && !fixed_link) {
-		netdev_err(dev, "Bad PHY UID 0x%08x\n", phydev->phy_id);
-		phy_disconnect(phydev);
-		return -ENODEV;
-	}
-
-	netdev_dbg(dev, "attached to PHY %d UID 0x%08x Link = %d\n",
-		   phydev->mdio.addr, phydev->phy_id, phydev->link);
-
-	return 0;
-}
-
-static void tse_update_mac_addr(struct altera_tse_private *priv, u8 *addr)
+static void tse_update_mac_addr(struct altera_tse_private *priv, const u8 *addr)
 {
 	u32 msb;
 	u32 lsb;
@@ -1011,7 +788,7 @@ static int tse_change_mtu(struct net_device *dev, int new_mtu)
 		return -EBUSY;
 	}
 
-	dev->mtu = new_mtu;
+	WRITE_ONCE(dev->mtu, new_mtu);
 	netdev_update_features(dev);
 
 	return 0;
@@ -1020,8 +797,8 @@ static int tse_change_mtu(struct net_device *dev, int new_mtu)
 static void altera_tse_set_mcfilter(struct net_device *dev)
 {
 	struct altera_tse_private *priv = netdev_priv(dev);
-	int i;
 	struct netdev_hw_addr *ha;
+	int i;
 
 	/* clear the hash filter */
 	for (i = 0; i < 64; i++)
@@ -1056,7 +833,7 @@ static void altera_tse_set_mcfilterall(struct net_device *dev)
 		csrwr32(1, priv->mac_dev, tse_csroffs(hash_table) + i * 4);
 }
 
-/* Set or clear the multicast filter for this adaptor
+/* Set or clear the multicast filter for this adapter
  */
 static void tse_set_rx_mode_hashfilter(struct net_device *dev)
 {
@@ -1076,7 +853,7 @@ static void tse_set_rx_mode_hashfilter(struct net_device *dev)
 	spin_unlock(&priv->mac_cfg_lock);
 }
 
-/* Set or clear the multicast filter for this adaptor
+/* Set or clear the multicast filter for this adapter
  */
 static void tse_set_rx_mode(struct net_device *dev)
 {
@@ -1095,74 +872,14 @@ static void tse_set_rx_mode(struct net_device *dev)
 	spin_unlock(&priv->mac_cfg_lock);
 }
 
-/* Initialise (if necessary) the SGMII PCS component
- */
-static int init_sgmii_pcs(struct net_device *dev)
-{
-	struct altera_tse_private *priv = netdev_priv(dev);
-	int n;
-	unsigned int tmp_reg = 0;
-
-	if (priv->phy_iface != PHY_INTERFACE_MODE_SGMII)
-		return 0; /* Nothing to do, not in SGMII mode */
-
-	/* The TSE SGMII PCS block looks a little like a PHY, it is
-	 * mapped into the zeroth MDIO space of the MAC and it has
-	 * ID registers like a PHY would.  Sadly this is often
-	 * configured to zeroes, so don't be surprised if it does
-	 * show 0x00000000.
-	 */
-
-	if (sgmii_pcs_scratch_test(priv, 0x0000) &&
-		sgmii_pcs_scratch_test(priv, 0xffff) &&
-		sgmii_pcs_scratch_test(priv, 0xa5a5) &&
-		sgmii_pcs_scratch_test(priv, 0x5a5a)) {
-		netdev_info(dev, "PCS PHY ID: 0x%04x%04x\n",
-				sgmii_pcs_read(priv, MII_PHYSID1),
-				sgmii_pcs_read(priv, MII_PHYSID2));
-	} else {
-		netdev_err(dev, "SGMII PCS Scratch memory test failed.\n");
-		return -ENOMEM;
-	}
-
-	/* Starting on page 5-29 of the MegaCore Function User Guide
-	 * Set SGMII Link timer to 1.6ms
-	 */
-	sgmii_pcs_write(priv, SGMII_PCS_LINK_TIMER_0, 0x0D40);
-	sgmii_pcs_write(priv, SGMII_PCS_LINK_TIMER_1, 0x03);
-
-	/* Enable SGMII Interface and Enable SGMII Auto Negotiation */
-	sgmii_pcs_write(priv, SGMII_PCS_IF_MODE, 0x3);
-
-	/* Enable Autonegotiation */
-	tmp_reg = sgmii_pcs_read(priv, MII_BMCR);
-	tmp_reg |= (BMCR_SPEED1000 | BMCR_FULLDPLX | BMCR_ANENABLE);
-	sgmii_pcs_write(priv, MII_BMCR, tmp_reg);
-
-	/* Reset PCS block */
-	tmp_reg |= BMCR_RESET;
-	sgmii_pcs_write(priv, MII_BMCR, tmp_reg);
-	for (n = 0; n < SGMII_PCS_SW_RESET_TIMEOUT; n++) {
-		if (!(sgmii_pcs_read(priv, MII_BMCR) & BMCR_RESET)) {
-			netdev_info(dev, "SGMII PCS block initialised OK\n");
-			return 0;
-		}
-		udelay(1);
-	}
-
-	/* We failed to reset the block, return a timeout */
-	netdev_err(dev, "SGMII PCS block reset failed.\n");
-	return -ETIMEDOUT;
-}
-
 /* Open and initialize the interface
  */
 static int tse_open(struct net_device *dev)
 {
 	struct altera_tse_private *priv = netdev_priv(dev);
+	unsigned long flags;
 	int ret = 0;
 	int i;
-	unsigned long int flags;
 
 	/* Reset and configure TSE MAC and probe associated PHY */
 	ret = priv->dmaops->init_dma(priv);
@@ -1179,14 +896,6 @@ static int tse_open(struct net_device *dev)
 		netdev_warn(dev, "TSE revision %x\n", priv->revision);
 
 	spin_lock(&priv->mac_cfg_lock);
-	/* no-op if MAC not operating in SGMII mode*/
-	ret = init_sgmii_pcs(dev);
-	if (ret) {
-		netdev_err(dev,
-			   "Cannot init the SGMII PCS (error: %d)\n", ret);
-		spin_unlock(&priv->mac_cfg_lock);
-		goto phy_error;
-	}
 
 	ret = reset_mac(priv);
 	/* Note that reset_mac will fail if the clocks are gated by the PHY
@@ -1244,8 +953,12 @@ static int tse_open(struct net_device *dev)
 
 	spin_unlock_irqrestore(&priv->rxdma_irq_lock, flags);
 
-	if (dev->phydev)
-		phy_start(dev->phydev);
+	ret = phylink_of_phy_connect(priv->phylink, priv->device->of_node, 0);
+	if (ret) {
+		netdev_err(dev, "could not connect phylink (%d)\n", ret);
+		goto tx_request_irq_error;
+	}
+	phylink_start(priv->phylink);
 
 	napi_enable(&priv->napi);
 	netif_start_queue(dev);
@@ -1273,13 +986,11 @@ phy_error:
 static int tse_shutdown(struct net_device *dev)
 {
 	struct altera_tse_private *priv = netdev_priv(dev);
-	int ret;
 	unsigned long int flags;
+	int ret;
 
-	/* Stop the PHY */
-	if (dev->phydev)
-		phy_stop(dev->phydev);
-
+	phylink_stop(priv->phylink);
+	phylink_disconnect_phy(priv->phylink);
 	netif_stop_queue(dev);
 	napi_disable(&priv->napi);
 
@@ -1325,11 +1036,73 @@ static struct net_device_ops altera_tse_netdev_ops = {
 	.ndo_validate_addr	= eth_validate_addr,
 };
 
+static void alt_tse_mac_config(struct phylink_config *config, unsigned int mode,
+			       const struct phylink_link_state *state)
+{
+	struct net_device *ndev = to_net_dev(config->dev);
+	struct altera_tse_private *priv = netdev_priv(ndev);
+
+	spin_lock(&priv->mac_cfg_lock);
+	reset_mac(priv);
+	tse_set_mac(priv, true);
+	spin_unlock(&priv->mac_cfg_lock);
+}
+
+static void alt_tse_mac_link_down(struct phylink_config *config,
+				  unsigned int mode, phy_interface_t interface)
+{
+}
+
+static void alt_tse_mac_link_up(struct phylink_config *config,
+				struct phy_device *phy, unsigned int mode,
+				phy_interface_t interface, int speed,
+				int duplex, bool tx_pause, bool rx_pause)
+{
+	struct net_device *ndev = to_net_dev(config->dev);
+	struct altera_tse_private *priv = netdev_priv(ndev);
+	u32 ctrl;
+
+	ctrl = csrrd32(priv->mac_dev, tse_csroffs(command_config));
+	ctrl &= ~(MAC_CMDCFG_ENA_10 | MAC_CMDCFG_ETH_SPEED | MAC_CMDCFG_HD_ENA);
+
+	if (duplex == DUPLEX_HALF)
+		ctrl |= MAC_CMDCFG_HD_ENA;
+
+	if (speed == SPEED_1000)
+		ctrl |= MAC_CMDCFG_ETH_SPEED;
+	else if (speed == SPEED_10)
+		ctrl |= MAC_CMDCFG_ENA_10;
+
+	spin_lock(&priv->mac_cfg_lock);
+	csrwr32(ctrl, priv->mac_dev, tse_csroffs(command_config));
+	spin_unlock(&priv->mac_cfg_lock);
+}
+
+static struct phylink_pcs *alt_tse_select_pcs(struct phylink_config *config,
+					      phy_interface_t interface)
+{
+	struct net_device *ndev = to_net_dev(config->dev);
+	struct altera_tse_private *priv = netdev_priv(ndev);
+
+	if (interface == PHY_INTERFACE_MODE_SGMII ||
+	    interface == PHY_INTERFACE_MODE_1000BASEX)
+		return priv->pcs;
+	else
+		return NULL;
+}
+
+static const struct phylink_mac_ops alt_tse_phylink_ops = {
+	.mac_config = alt_tse_mac_config,
+	.mac_link_down = alt_tse_mac_link_down,
+	.mac_link_up = alt_tse_mac_link_up,
+	.mac_select_pcs = alt_tse_select_pcs,
+};
+
 static int request_and_map(struct platform_device *pdev, const char *name,
 			   struct resource **res, void __iomem **ptr)
 {
-	struct resource *region;
 	struct device *device = &pdev->dev;
+	struct resource *region;
 
 	*res = platform_get_resource_byname(pdev, IORESOURCE_MEM, name);
 	if (*res == NULL) {
@@ -1344,10 +1117,10 @@ static int request_and_map(struct platform_device *pdev, const char *name,
 		return -EBUSY;
 	}
 
-	*ptr = devm_ioremap_nocache(device, region->start,
+	*ptr = devm_ioremap(device, region->start,
 				    resource_size(region));
 	if (*ptr == NULL) {
-		dev_err(device, "ioremap_nocache of %s failed!", name);
+		dev_err(device, "ioremap of %s failed!", name);
 		return -ENOMEM;
 	}
 
@@ -1358,14 +1131,17 @@ static int request_and_map(struct platform_device *pdev, const char *name,
  */
 static int altera_tse_probe(struct platform_device *pdev)
 {
-	struct net_device *ndev;
-	int ret = -ENODEV;
+	struct regmap_config pcs_regmap_cfg;
+	struct altera_tse_private *priv;
+	struct mdio_regmap_config mrc;
 	struct resource *control_port;
+	struct regmap *pcs_regmap;
 	struct resource *dma_res;
-	struct altera_tse_private *priv;
-	const unsigned char *macaddr;
+	struct resource *pcs_res;
+	struct mii_bus *pcs_bus;
+	struct net_device *ndev;
 	void __iomem *descmap;
-	const struct of_device_id *of_id = NULL;
+	int ret = -ENODEV;
 
 	ndev = alloc_etherdev(sizeof(struct altera_tse_private));
 	if (!ndev) {
@@ -1380,11 +1156,7 @@ static int altera_tse_probe(struct platform_device *pdev)
 	priv->dev = ndev;
 	priv->msg_enable = netif_msg_init(debug, default_msg_level);
 
-	of_id = of_match_device(altera_tse_ids, &pdev->dev);
-
-	if (of_id)
-		priv->dmaops = (struct altera_dmaops *)of_id->data;
-
+	priv->dmaops = device_get_match_data(&pdev->dev);
 
 	if (priv->dmaops &&
 	    priv->dmaops->altera_dtype == ALTERA_DTYPE_SGDMA) {
@@ -1443,16 +1215,19 @@ static int altera_tse_probe(struct platform_device *pdev)
 		priv->rxdescmem_busaddr = dma_res->start;
 
 	} else {
+		ret = -ENODEV;
 		goto err_free_netdev;
 	}
 
-	if (!dma_set_mask(priv->device, DMA_BIT_MASK(priv->dmaops->dmamask)))
+	if (!dma_set_mask(priv->device, DMA_BIT_MASK(priv->dmaops->dmamask))) {
 		dma_set_coherent_mask(priv->device,
 				      DMA_BIT_MASK(priv->dmaops->dmamask));
-	else if (!dma_set_mask(priv->device, DMA_BIT_MASK(32)))
+	} else if (!dma_set_mask(priv->device, DMA_BIT_MASK(32))) {
 		dma_set_coherent_mask(priv->device, DMA_BIT_MASK(32));
-	else
+	} else {
+		ret = -EIO;
 		goto err_free_netdev;
+	}
 
 	/* MAC address space */
 	ret = request_and_map(pdev, "control_port", &control_port,
@@ -1473,6 +1248,40 @@ static int altera_tse_probe(struct platform_device *pdev)
 	if (ret)
 		goto err_free_netdev;
 
+	memset(&pcs_regmap_cfg, 0, sizeof(pcs_regmap_cfg));
+	memset(&mrc, 0, sizeof(mrc));
+	/* SGMII PCS address space. The location can vary depending on how the
+	 * IP is integrated. We can have a resource dedicated to it at a specific
+	 * address space, but if it's not the case, we fallback to the mdiophy0
+	 * from the MAC's address space
+	 */
+	ret = request_and_map(pdev, "pcs", &pcs_res, &priv->pcs_base);
+	if (ret) {
+		/* If we can't find a dedicated resource for the PCS, fallback
+		 * to the internal PCS, that has a different address stride
+		 */
+		priv->pcs_base = priv->mac_dev + tse_csroffs(mdio_phy0);
+		pcs_regmap_cfg.reg_bits = 32;
+		/* Values are MDIO-like values, on 16 bits */
+		pcs_regmap_cfg.val_bits = 16;
+		pcs_regmap_cfg.reg_shift = REGMAP_UPSHIFT(2);
+	} else {
+		pcs_regmap_cfg.reg_bits = 16;
+		pcs_regmap_cfg.val_bits = 16;
+		pcs_regmap_cfg.reg_shift = REGMAP_UPSHIFT(1);
+	}
+
+	/* Create a regmap for the PCS so that it can be used by the PCS driver */
+	pcs_regmap = devm_regmap_init_mmio(&pdev->dev, priv->pcs_base,
+					   &pcs_regmap_cfg);
+	if (IS_ERR(pcs_regmap)) {
+		ret = PTR_ERR(pcs_regmap);
+		goto err_free_netdev;
+	}
+	mrc.regmap = pcs_regmap;
+	mrc.parent = &pdev->dev;
+	mrc.valid_addr = 0x0;
+	mrc.autoscan = false;
 
 	/* Rx IRQ */
 	priv->rx_irq = platform_get_irq_byname(pdev, "rx_irq");
@@ -1537,10 +1346,8 @@ static int altera_tse_probe(struct platform_device *pdev)
 	priv->rx_dma_buf_sz = ALTERA_RXDMABUFFER_SIZE;
 
 	/* get default MAC address from device tree */
-	macaddr = of_get_mac_address(pdev->dev.of_node);
-	if (macaddr)
-		ether_addr_copy(ndev->dev_addr, macaddr);
-	else
+	ret = of_get_ethdev_address(pdev->dev.of_node, ndev);
+	if (ret)
 		eth_hw_addr_random(ndev);
 
 	/* get phy addr and create mdio */
@@ -1574,7 +1381,7 @@ static int altera_tse_probe(struct platform_device *pdev)
 	ndev->features |= NETIF_F_HW_VLAN_CTAG_RX;
 
 	/* setup NAPI interface */
-	netif_napi_add(ndev, &priv->napi, tse_poll, NAPI_POLL_WEIGHT);
+	netif_napi_add(ndev, &priv->napi, tse_poll);
 
 	spin_lock_init(&priv->mac_cfg_lock);
 	spin_lock_init(&priv->tx_lock);
@@ -1598,14 +1405,47 @@ static int altera_tse_probe(struct platform_device *pdev)
 			 (unsigned long) control_port->start, priv->rx_irq,
 			 priv->tx_irq);
 
-	ret = init_phy(ndev);
-	if (ret != 0) {
-		netdev_err(ndev, "Cannot attach to PHY (error: %d)\n", ret);
-		goto err_init_phy;
+	snprintf(mrc.name, MII_BUS_ID_SIZE, "%s-pcs-mii", ndev->name);
+	pcs_bus = devm_mdio_regmap_register(&pdev->dev, &mrc);
+	if (IS_ERR(pcs_bus)) {
+		ret = PTR_ERR(pcs_bus);
+		goto err_init_pcs;
 	}
-	return 0;
 
-err_init_phy:
+	priv->pcs = lynx_pcs_create_mdiodev(pcs_bus, 0);
+	if (IS_ERR(priv->pcs)) {
+		ret = PTR_ERR(priv->pcs);
+		goto err_init_pcs;
+	}
+
+	priv->phylink_config.dev = &ndev->dev;
+	priv->phylink_config.type = PHYLINK_NETDEV;
+	priv->phylink_config.mac_capabilities = MAC_SYM_PAUSE | MAC_10 |
+						MAC_100 | MAC_1000FD;
+
+	phy_interface_set_rgmii(priv->phylink_config.supported_interfaces);
+	__set_bit(PHY_INTERFACE_MODE_MII,
+		  priv->phylink_config.supported_interfaces);
+	__set_bit(PHY_INTERFACE_MODE_GMII,
+		  priv->phylink_config.supported_interfaces);
+	__set_bit(PHY_INTERFACE_MODE_SGMII,
+		  priv->phylink_config.supported_interfaces);
+	__set_bit(PHY_INTERFACE_MODE_1000BASEX,
+		  priv->phylink_config.supported_interfaces);
+
+	priv->phylink = phylink_create(&priv->phylink_config,
+				       of_fwnode_handle(priv->device->of_node),
+				       priv->phy_iface, &alt_tse_phylink_ops);
+	if (IS_ERR(priv->phylink)) {
+		dev_err(&pdev->dev, "failed to create phylink\n");
+		ret = PTR_ERR(priv->phylink);
+		goto err_init_phylink;
+	}
+
+	return 0;
+err_init_phylink:
+	lynx_pcs_destroy(priv->pcs);
+err_init_pcs:
 	unregister_netdev(ndev);
 err_register_netdev:
 	netif_napi_del(&priv->napi);
@@ -1617,24 +1457,18 @@ err_free_netdev:
 
 /* Remove Altera TSE MAC device
  */
-static int altera_tse_remove(struct platform_device *pdev)
+static void altera_tse_remove(struct platform_device *pdev)
 {
 	struct net_device *ndev = platform_get_drvdata(pdev);
 	struct altera_tse_private *priv = netdev_priv(ndev);
 
-	if (ndev->phydev) {
-		phy_disconnect(ndev->phydev);
-
-		if (of_phy_is_fixed_link(priv->device->of_node))
-			of_phy_deregister_fixed_link(priv->device->of_node);
-	}
-
 	platform_set_drvdata(pdev, NULL);
 	altera_tse_mdio_destroy(ndev);
 	unregister_netdev(ndev);
-	free_netdev(ndev);
+	phylink_destroy(priv->phylink);
+	lynx_pcs_destroy(priv->pcs);
 
-	return 0;
+	free_netdev(ndev);
 }
 
 static const struct altera_dmaops altera_dtype_sgdma = {
diff --git a/drivers/net/ethernet/altera/altera_utils.c b/drivers/net/ethernet/altera/altera_utils.c
index d7eeb1713ad2..e6a7fc9d8fb1 100644
--- a/drivers/net/ethernet/altera/altera_utils.c
+++ b/drivers/net/ethernet/altera/altera_utils.c
@@ -1,17 +1,6 @@
+// SPDX-License-Identifier: GPL-2.0-only
 /* Altera TSE SGDMA and MSGDMA Linux driver
  * Copyright (C) 2014 Altera Corporation. All rights reserved
- *
- * This program is free software; you can redistribute it and/or modify it
- * under the terms and conditions of the GNU General Public License,
- * version 2, as published by the Free Software Foundation.
- *
- * This program is distributed in the hope it will be useful, but WITHOUT
- * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
- * FITNESS FOR A PARTICULAR PURPOSE.  See the GNU General Public License for
- * more details.
- *
- * You should have received a copy of the GNU General Public License along with
- * this program.  If not, see <http://www.gnu.org/licenses/>.
  */
 
 #include "altera_tse.h"
diff --git a/drivers/net/ethernet/altera/altera_utils.h b/drivers/net/ethernet/altera/altera_utils.h
index baf100ccf587..3c2e32fb7389 100644
--- a/drivers/net/ethernet/altera/altera_utils.h
+++ b/drivers/net/ethernet/altera/altera_utils.h
@@ -1,24 +1,14 @@
+/* SPDX-License-Identifier: GPL-2.0-only */
 /* Altera TSE SGDMA and MSGDMA Linux driver
  * Copyright (C) 2014 Altera Corporation. All rights reserved
- *
- * This program is free software; you can redistribute it and/or modify it
- * under the terms and conditions of the GNU General Public License,
- * version 2, as published by the Free Software Foundation.
- *
- * This program is distributed in the hope it will be useful, but WITHOUT
- * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
- * FITNESS FOR A PARTICULAR PURPOSE.  See the GNU General Public License for
- * more details.
- *
- * You should have received a copy of the GNU General Public License along with
- * this program.  If not, see <http://www.gnu.org/licenses/>.
  */
 
-#include <linux/kernel.h>
-
 #ifndef __ALTERA_UTILS_H__
 #define __ALTERA_UTILS_H__
 
+#include <linux/compiler.h>
+#include <linux/types.h>
+
 void tse_set_bit(void __iomem *ioaddr, size_t offs, u32 bit_mask);
 void tse_clear_bit(void __iomem *ioaddr, size_t offs, u32 bit_mask);
 int tse_bit_is_set(void __iomem *ioaddr, size_t offs, u32 bit_mask);
diff --git a/drivers/net/ethernet/amazon/Kconfig b/drivers/net/ethernet/amazon/Kconfig
index 99b30353541a..95dcc3969f0c 100644
--- a/drivers/net/ethernet/amazon/Kconfig
+++ b/drivers/net/ethernet/amazon/Kconfig
@@ -1,3 +1,4 @@
+# SPDX-License-Identifier: GPL-2.0-only
 #
 # Amazon network device configuration
 #
@@ -5,7 +6,7 @@
 config NET_VENDOR_AMAZON
 	bool "Amazon Devices"
 	default y
-	---help---
+	help
 	  If you have a network (Ethernet) device belonging to this class, say Y.
 
 	  Note that the answer to this question doesn't directly affect the
@@ -17,8 +18,11 @@ if NET_VENDOR_AMAZON
 
 config ENA_ETHERNET
 	tristate "Elastic Network Adapter (ENA) support"
-	depends on (PCI_MSI && X86)
-	---help---
+	depends on PCI_MSI && !CPU_BIG_ENDIAN
+	depends on PTP_1588_CLOCK_OPTIONAL
+	select DIMLIB
+	select NET_DEVLINK
+	help
 	  This driver supports Elastic Network Adapter (ENA)"
 
 	  To compile this driver as a module, choose M here.
diff --git a/drivers/net/ethernet/amazon/Makefile b/drivers/net/ethernet/amazon/Makefile
index 8e0b73f60d51..f614f23ec549 100644
--- a/drivers/net/ethernet/amazon/Makefile
+++ b/drivers/net/ethernet/amazon/Makefile
@@ -1,3 +1,4 @@
+# SPDX-License-Identifier: GPL-2.0-only
 #
 # Makefile for the Amazon network device drivers.
 #
diff --git a/drivers/net/ethernet/amazon/ena/Makefile b/drivers/net/ethernet/amazon/ena/Makefile
index eaeeae06c5d9..6d8036bc1823 100644
--- a/drivers/net/ethernet/amazon/ena/Makefile
+++ b/drivers/net/ethernet/amazon/ena/Makefile
@@ -1,7 +1,8 @@
+# SPDX-License-Identifier: GPL-2.0-only
 #
 # Makefile for the Elastic Network Adapter (ENA) device drivers.
 #
 
 obj-$(CONFIG_ENA_ETHERNET) += ena.o
 
-ena-y := ena_netdev.o ena_com.o ena_eth_com.o ena_ethtool.o
+ena-y := ena_netdev.o ena_com.o ena_eth_com.o ena_ethtool.o ena_xdp.o ena_phc.o ena_devlink.o ena_debugfs.o
diff --git a/drivers/net/ethernet/amazon/ena/ena_admin_defs.h b/drivers/net/ethernet/amazon/ena/ena_admin_defs.h
index 5b6509d59716..898ecd96b96a 100644
--- a/drivers/net/ethernet/amazon/ena/ena_admin_defs.h
+++ b/drivers/net/ethernet/amazon/ena/ena_admin_defs.h
@@ -1,144 +1,112 @@
+/* SPDX-License-Identifier: GPL-2.0 OR Linux-OpenIB */
 /*
- * Copyright 2015 - 2016 Amazon.com, Inc. or its affiliates.
- *
- * This software is available to you under a choice of one of two
- * licenses.  You may choose to be licensed under the terms of the GNU
- * General Public License (GPL) Version 2, available from the file
- * COPYING in the main directory of this source tree, or the
- * BSD license below:
- *
- *     Redistribution and use in source and binary forms, with or
- *     without modification, are permitted provided that the following
- *     conditions are met:
- *
- *      - Redistributions of source code must retain the above
- *        copyright notice, this list of conditions and the following
- *        disclaimer.
- *
- *      - Redistributions in binary form must reproduce the above
- *        copyright notice, this list of conditions and the following
- *        disclaimer in the documentation and/or other materials
- *        provided with the distribution.
- *
- * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND,
- * EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF
- * MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND
- * NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS
- * BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN
- * ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN
- * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
- * SOFTWARE.
+ * Copyright 2015-2020 Amazon.com, Inc. or its affiliates. All rights reserved.
  */
 #ifndef _ENA_ADMIN_H_
 #define _ENA_ADMIN_H_
 
-enum ena_admin_aq_opcode {
-	ENA_ADMIN_CREATE_SQ	= 1,
-
-	ENA_ADMIN_DESTROY_SQ	= 2,
-
-	ENA_ADMIN_CREATE_CQ	= 3,
-
-	ENA_ADMIN_DESTROY_CQ	= 4,
-
-	ENA_ADMIN_GET_FEATURE	= 8,
-
-	ENA_ADMIN_SET_FEATURE	= 9,
+#define ENA_ADMIN_RSS_KEY_PARTS              10
+
+#define ENA_ADMIN_CUSTOMER_METRICS_SUPPORT_MASK 0x3F
+#define ENA_ADMIN_CUSTOMER_METRICS_MIN_SUPPORT_MASK 0x1F
+
+ /* customer metrics - in correlation with
+  * ENA_ADMIN_CUSTOMER_METRICS_SUPPORT_MASK
+  */
+enum ena_admin_customer_metrics_id {
+	ENA_ADMIN_BW_IN_ALLOWANCE_EXCEEDED         = 0,
+	ENA_ADMIN_BW_OUT_ALLOWANCE_EXCEEDED        = 1,
+	ENA_ADMIN_PPS_ALLOWANCE_EXCEEDED           = 2,
+	ENA_ADMIN_CONNTRACK_ALLOWANCE_EXCEEDED     = 3,
+	ENA_ADMIN_LINKLOCAL_ALLOWANCE_EXCEEDED     = 4,
+	ENA_ADMIN_CONNTRACK_ALLOWANCE_AVAILABLE    = 5,
+};
 
-	ENA_ADMIN_GET_STATS	= 11,
+enum ena_admin_aq_opcode {
+	ENA_ADMIN_CREATE_SQ                         = 1,
+	ENA_ADMIN_DESTROY_SQ                        = 2,
+	ENA_ADMIN_CREATE_CQ                         = 3,
+	ENA_ADMIN_DESTROY_CQ                        = 4,
+	ENA_ADMIN_GET_FEATURE                       = 8,
+	ENA_ADMIN_SET_FEATURE                       = 9,
+	ENA_ADMIN_GET_STATS                         = 11,
 };
 
 enum ena_admin_aq_completion_status {
-	ENA_ADMIN_SUCCESS			= 0,
-
-	ENA_ADMIN_RESOURCE_ALLOCATION_FAILURE	= 1,
-
-	ENA_ADMIN_BAD_OPCODE			= 2,
-
-	ENA_ADMIN_UNSUPPORTED_OPCODE		= 3,
-
-	ENA_ADMIN_MALFORMED_REQUEST		= 4,
-
+	ENA_ADMIN_SUCCESS                           = 0,
+	ENA_ADMIN_RESOURCE_ALLOCATION_FAILURE       = 1,
+	ENA_ADMIN_BAD_OPCODE                        = 2,
+	ENA_ADMIN_UNSUPPORTED_OPCODE                = 3,
+	ENA_ADMIN_MALFORMED_REQUEST                 = 4,
 	/* Additional status is provided in ACQ entry extended_status */
-	ENA_ADMIN_ILLEGAL_PARAMETER		= 5,
-
-	ENA_ADMIN_UNKNOWN_ERROR			= 6,
+	ENA_ADMIN_ILLEGAL_PARAMETER                 = 5,
+	ENA_ADMIN_UNKNOWN_ERROR                     = 6,
+	ENA_ADMIN_RESOURCE_BUSY                     = 7,
 };
 
+/* subcommands for the set/get feature admin commands */
 enum ena_admin_aq_feature_id {
-	ENA_ADMIN_DEVICE_ATTRIBUTES		= 1,
-
-	ENA_ADMIN_MAX_QUEUES_NUM		= 2,
-
-	ENA_ADMIN_RSS_HASH_FUNCTION		= 10,
-
-	ENA_ADMIN_STATELESS_OFFLOAD_CONFIG	= 11,
-
-	ENA_ADMIN_RSS_REDIRECTION_TABLE_CONFIG	= 12,
-
-	ENA_ADMIN_MTU				= 14,
-
-	ENA_ADMIN_RSS_HASH_INPUT		= 18,
-
-	ENA_ADMIN_INTERRUPT_MODERATION		= 20,
-
-	ENA_ADMIN_AENQ_CONFIG			= 26,
-
-	ENA_ADMIN_LINK_CONFIG			= 27,
-
-	ENA_ADMIN_HOST_ATTR_CONFIG		= 28,
+	ENA_ADMIN_DEVICE_ATTRIBUTES                 = 1,
+	ENA_ADMIN_MAX_QUEUES_NUM                    = 2,
+	ENA_ADMIN_HW_HINTS                          = 3,
+	ENA_ADMIN_LLQ                               = 4,
+	ENA_ADMIN_MAX_QUEUES_EXT                    = 7,
+	ENA_ADMIN_RSS_HASH_FUNCTION                 = 10,
+	ENA_ADMIN_STATELESS_OFFLOAD_CONFIG          = 11,
+	ENA_ADMIN_RSS_INDIRECTION_TABLE_CONFIG      = 12,
+	ENA_ADMIN_MTU                               = 14,
+	ENA_ADMIN_RSS_HASH_INPUT                    = 18,
+	ENA_ADMIN_INTERRUPT_MODERATION              = 20,
+	ENA_ADMIN_AENQ_CONFIG                       = 26,
+	ENA_ADMIN_LINK_CONFIG                       = 27,
+	ENA_ADMIN_HOST_ATTR_CONFIG                  = 28,
+	ENA_ADMIN_PHC_CONFIG                        = 29,
+	ENA_ADMIN_FEATURES_OPCODE_NUM               = 32,
+};
 
-	ENA_ADMIN_FEATURES_OPCODE_NUM		= 32,
+/* device capabilities */
+enum ena_admin_aq_caps_id {
+	ENA_ADMIN_ENI_STATS                         = 0,
+	/* ENA SRD customer metrics */
+	ENA_ADMIN_ENA_SRD_INFO                      = 1,
+	ENA_ADMIN_CUSTOMER_METRICS                  = 2,
 };
 
 enum ena_admin_placement_policy_type {
 	/* descriptors and headers are in host memory */
-	ENA_ADMIN_PLACEMENT_POLICY_HOST	= 1,
-
+	ENA_ADMIN_PLACEMENT_POLICY_HOST             = 1,
 	/* descriptors and headers are in device memory (a.k.a Low Latency
 	 * Queue)
 	 */
-	ENA_ADMIN_PLACEMENT_POLICY_DEV	= 3,
+	ENA_ADMIN_PLACEMENT_POLICY_DEV              = 3,
 };
 
 enum ena_admin_link_types {
-	ENA_ADMIN_LINK_SPEED_1G		= 0x1,
-
-	ENA_ADMIN_LINK_SPEED_2_HALF_G	= 0x2,
-
-	ENA_ADMIN_LINK_SPEED_5G		= 0x4,
-
-	ENA_ADMIN_LINK_SPEED_10G	= 0x8,
-
-	ENA_ADMIN_LINK_SPEED_25G	= 0x10,
-
-	ENA_ADMIN_LINK_SPEED_40G	= 0x20,
-
-	ENA_ADMIN_LINK_SPEED_50G	= 0x40,
-
-	ENA_ADMIN_LINK_SPEED_100G	= 0x80,
-
-	ENA_ADMIN_LINK_SPEED_200G	= 0x100,
-
-	ENA_ADMIN_LINK_SPEED_400G	= 0x200,
+	ENA_ADMIN_LINK_SPEED_1G                     = 0x1,
+	ENA_ADMIN_LINK_SPEED_2_HALF_G               = 0x2,
+	ENA_ADMIN_LINK_SPEED_5G                     = 0x4,
+	ENA_ADMIN_LINK_SPEED_10G                    = 0x8,
+	ENA_ADMIN_LINK_SPEED_25G                    = 0x10,
+	ENA_ADMIN_LINK_SPEED_40G                    = 0x20,
+	ENA_ADMIN_LINK_SPEED_50G                    = 0x40,
+	ENA_ADMIN_LINK_SPEED_100G                   = 0x80,
+	ENA_ADMIN_LINK_SPEED_200G                   = 0x100,
+	ENA_ADMIN_LINK_SPEED_400G                   = 0x200,
 };
 
 enum ena_admin_completion_policy_type {
 	/* completion queue entry for each sq descriptor */
-	ENA_ADMIN_COMPLETION_POLICY_DESC		= 0,
-
+	ENA_ADMIN_COMPLETION_POLICY_DESC            = 0,
 	/* completion queue entry upon request in sq descriptor */
-	ENA_ADMIN_COMPLETION_POLICY_DESC_ON_DEMAND	= 1,
-
+	ENA_ADMIN_COMPLETION_POLICY_DESC_ON_DEMAND  = 1,
 	/* current queue head pointer is updated in OS memory upon sq
 	 * descriptor request
 	 */
-	ENA_ADMIN_COMPLETION_POLICY_HEAD_ON_DEMAND	= 2,
-
+	ENA_ADMIN_COMPLETION_POLICY_HEAD_ON_DEMAND  = 2,
 	/* current queue head pointer is updated in OS memory for each sq
 	 * descriptor
 	 */
-	ENA_ADMIN_COMPLETION_POLICY_HEAD		= 3,
+	ENA_ADMIN_COMPLETION_POLICY_HEAD            = 3,
 };
 
 /* basic stats return ena_admin_basic_stats while extanded stats return a
@@ -146,15 +114,36 @@ enum ena_admin_completion_policy_type {
  * device id
  */
 enum ena_admin_get_stats_type {
-	ENA_ADMIN_GET_STATS_TYPE_BASIC		= 0,
-
-	ENA_ADMIN_GET_STATS_TYPE_EXTENDED	= 1,
+	ENA_ADMIN_GET_STATS_TYPE_BASIC              = 0,
+	ENA_ADMIN_GET_STATS_TYPE_EXTENDED           = 1,
+	/* extra HW stats for specific network interface */
+	ENA_ADMIN_GET_STATS_TYPE_ENI                = 2,
+	/* extra HW stats for ENA SRD */
+	ENA_ADMIN_GET_STATS_TYPE_ENA_SRD            = 3,
+	ENA_ADMIN_GET_STATS_TYPE_CUSTOMER_METRICS   = 4,
 };
 
 enum ena_admin_get_stats_scope {
-	ENA_ADMIN_SPECIFIC_QUEUE	= 0,
+	ENA_ADMIN_SPECIFIC_QUEUE                    = 0,
+	ENA_ADMIN_ETH_TRAFFIC                       = 1,
+};
+
+enum ena_admin_phc_type {
+	ENA_ADMIN_PHC_TYPE_READLESS                 = 0,
+};
+
+enum ena_admin_phc_error_flags {
+	ENA_ADMIN_PHC_ERROR_FLAG_TIMESTAMP   = BIT(0),
+};
 
-	ENA_ADMIN_ETH_TRAFFIC		= 1,
+/* ENA SRD configuration for ENI */
+enum ena_admin_ena_srd_flags {
+	/* Feature enabled */
+	ENA_ADMIN_ENA_SRD_ENABLED                   = BIT(0),
+	/* UDP support enabled */
+	ENA_ADMIN_ENA_SRD_UDP_ENABLED               = BIT(1),
+	/* Bypass Rx UDP ordering */
+	ENA_ADMIN_ENA_SRD_UDP_ORDERING_BYPASS_ENABLED = BIT(2),
 };
 
 struct ena_admin_aq_common_desc {
@@ -225,7 +214,9 @@ struct ena_admin_acq_common_desc {
 
 	u16 extended_status;
 
-	/* serves as a hint what AQ entries can be revoked */
+	/* indicates to the driver which AQ entry has been consumed by the
+	 * device and could be reused
+	 */
 	u16 sq_head_indx;
 };
 
@@ -269,8 +260,8 @@ struct ena_admin_aq_create_sq_cmd {
 	 */
 	u8 sq_caps_3;
 
-	/* associated completion queue id. This CQ must be created prior to
-	 *    SQ creation
+	/* associated completion queue id. This CQ must be created prior to SQ
+	 * creation
 	 */
 	u16 cq_idx;
 
@@ -294,9 +285,8 @@ struct ena_admin_aq_create_sq_cmd {
 };
 
 enum ena_admin_sq_direction {
-	ENA_ADMIN_SQ_DIRECTION_TX	= 1,
-
-	ENA_ADMIN_SQ_DIRECTION_RX	= 2,
+	ENA_ADMIN_SQ_DIRECTION_TX                   = 1,
+	ENA_ADMIN_SQ_DIRECTION_RX                   = 2,
 };
 
 struct ena_admin_acq_create_sq_resp_desc {
@@ -410,9 +400,12 @@ struct ena_admin_aq_get_stats_cmd {
 	u16 queue_idx;
 
 	/* device id, value 0xFFFF means mine. only privileged device can get
-	 *    stats of other device
+	 * stats of other device
 	 */
 	u16 device_id;
+
+	/* a bitmap representing the requested metric values */
+	u64 requested_metrics;
 };
 
 /* Basic Statistics Command. */
@@ -436,12 +429,87 @@ struct ena_admin_basic_stats {
 	u32 rx_drops_low;
 
 	u32 rx_drops_high;
+
+	u32 tx_drops_low;
+
+	u32 tx_drops_high;
+};
+
+/* ENI Statistics Command. */
+struct ena_admin_eni_stats {
+	/* The number of packets shaped due to inbound aggregate BW
+	 * allowance being exceeded
+	 */
+	u64 bw_in_allowance_exceeded;
+
+	/* The number of packets shaped due to outbound aggregate BW
+	 * allowance being exceeded
+	 */
+	u64 bw_out_allowance_exceeded;
+
+	/* The number of packets shaped due to PPS allowance being exceeded */
+	u64 pps_allowance_exceeded;
+
+	/* The number of packets shaped due to connection tracking
+	 * allowance being exceeded and leading to failure in establishment
+	 * of new connections
+	 */
+	u64 conntrack_allowance_exceeded;
+
+	/* The number of packets shaped due to linklocal packet rate
+	 * allowance being exceeded
+	 */
+	u64 linklocal_allowance_exceeded;
+};
+
+struct ena_admin_ena_srd_stats {
+	/* Number of packets transmitted over ENA SRD */
+	u64 ena_srd_tx_pkts;
+
+	/* Number of packets transmitted or could have been
+	 * transmitted over ENA SRD
+	 */
+	u64 ena_srd_eligible_tx_pkts;
+
+	/* Number of packets received over ENA SRD */
+	u64 ena_srd_rx_pkts;
+
+	/* Percentage of the ENA SRD resources that is in use */
+	u64 ena_srd_resource_utilization;
+};
+
+/* ENA SRD Statistics Command */
+struct ena_admin_ena_srd_info {
+	/* ENA SRD configuration bitmap. See ena_admin_ena_srd_flags for
+	 * details
+	 */
+	u64 flags;
+
+	struct ena_admin_ena_srd_stats ena_srd_stats;
+};
+
+/* Customer Metrics Command. */
+struct ena_admin_customer_metrics {
+	/* A bitmap representing the reported customer metrics according to
+	 * the order they are reported
+	 */
+	u64 reported_metrics;
 };
 
 struct ena_admin_acq_get_stats_resp {
 	struct ena_admin_acq_common_desc acq_common_desc;
 
-	struct ena_admin_basic_stats basic_stats;
+	union {
+		u64 raw[7];
+
+		struct ena_admin_basic_stats basic_stats;
+
+		struct ena_admin_eni_stats eni_stats;
+
+		struct ena_admin_ena_srd_info ena_srd_info;
+
+		struct ena_admin_customer_metrics customer_metrics;
+	} u;
 };
 
 struct ena_admin_get_set_feature_common_desc {
@@ -454,7 +522,13 @@ struct ena_admin_get_set_feature_common_desc {
 	/* as appears in ena_admin_aq_feature_id */
 	u8 feature_id;
 
-	u16 reserved16;
+	/* The driver specifies the max feature version it supports and the
+	 * device responds with the currently supported feature version. The
+	 * field is zero based
+	 */
+	u8 feature_version;
+
+	u8 reserved8;
 };
 
 struct ena_admin_device_attr_feature_desc {
@@ -462,10 +536,15 @@ struct ena_admin_device_attr_feature_desc {
 
 	u32 device_version;
 
-	/* bitmap of ena_admin_aq_feature_id */
+	/* bitmap of ena_admin_aq_feature_id, which represents supported
+	 * subcommands for the set/get feature admin commands.
+	 */
 	u32 supported_features;
 
-	u32 reserved3;
+	/* bitmap of ena_admin_aq_caps_id, which represents device
+	 * capabilities.
+	 */
+	u32 capabilities;
 
 	/* Indicates how many bits are used physical address access. */
 	u32 phys_addr_width;
@@ -481,8 +560,149 @@ struct ena_admin_device_attr_feature_desc {
 	u32 max_mtu;
 };
 
+enum ena_admin_llq_header_location {
+	/* header is in descriptor list */
+	ENA_ADMIN_INLINE_HEADER                     = 1,
+	/* header in a separate ring, implies 16B descriptor list entry */
+	ENA_ADMIN_HEADER_RING                       = 2,
+};
+
+enum ena_admin_llq_ring_entry_size {
+	ENA_ADMIN_LIST_ENTRY_SIZE_128B              = 1,
+	ENA_ADMIN_LIST_ENTRY_SIZE_192B              = 2,
+	ENA_ADMIN_LIST_ENTRY_SIZE_256B              = 4,
+};
+
+enum ena_admin_llq_num_descs_before_header {
+	ENA_ADMIN_LLQ_NUM_DESCS_BEFORE_HEADER_0     = 0,
+	ENA_ADMIN_LLQ_NUM_DESCS_BEFORE_HEADER_1     = 1,
+	ENA_ADMIN_LLQ_NUM_DESCS_BEFORE_HEADER_2     = 2,
+	ENA_ADMIN_LLQ_NUM_DESCS_BEFORE_HEADER_4     = 4,
+	ENA_ADMIN_LLQ_NUM_DESCS_BEFORE_HEADER_8     = 8,
+};
+
+/* packet descriptor list entry always starts with one or more descriptors,
+ * followed by a header. The rest of the descriptors are located in the
+ * beginning of the subsequent entry. Stride refers to how the rest of the
+ * descriptors are placed. This field is relevant only for inline header
+ * mode
+ */
+enum ena_admin_llq_stride_ctrl {
+	ENA_ADMIN_SINGLE_DESC_PER_ENTRY             = 1,
+	ENA_ADMIN_MULTIPLE_DESCS_PER_ENTRY          = 2,
+};
+
+enum ena_admin_accel_mode_feat {
+	ENA_ADMIN_DISABLE_META_CACHING              = 0,
+	ENA_ADMIN_LIMIT_TX_BURST                    = 1,
+};
+
+struct ena_admin_accel_mode_get {
+	/* bit field of enum ena_admin_accel_mode_feat */
+	u16 supported_flags;
+
+	/* maximum burst size between two doorbells. The size is in bytes */
+	u16 max_tx_burst_size;
+};
+
+struct ena_admin_accel_mode_set {
+	/* bit field of enum ena_admin_accel_mode_feat */
+	u16 enabled_flags;
+
+	u16 reserved;
+};
+
+struct ena_admin_accel_mode_req {
+	union {
+		u32 raw[2];
+
+		struct ena_admin_accel_mode_get get;
+
+		struct ena_admin_accel_mode_set set;
+	} u;
+};
+
+struct ena_admin_feature_llq_desc {
+	u32 max_llq_num;
+
+	u32 max_llq_depth;
+
+	/* specify the header locations the device supports. bitfield of enum
+	 * ena_admin_llq_header_location.
+	 */
+	u16 header_location_ctrl_supported;
+
+	/* the header location the driver selected to use. */
+	u16 header_location_ctrl_enabled;
+
+	/* if inline header is specified - this is the size of descriptor list
+	 * entry. If header in a separate ring is specified - this is the size
+	 * of header ring entry. bitfield of enum ena_admin_llq_ring_entry_size.
+	 * specify the entry sizes the device supports
+	 */
+	u16 entry_size_ctrl_supported;
+
+	/* the entry size the driver selected to use. */
+	u16 entry_size_ctrl_enabled;
+
+	/* valid only if inline header is specified. First entry associated with
+	 * the packet includes descriptors and header. Rest of the entries
+	 * occupied by descriptors. This parameter defines the max number of
+	 * descriptors precedding the header in the first entry. The field is
+	 * bitfield of enum ena_admin_llq_num_descs_before_header and specify
+	 * the values the device supports
+	 */
+	u16 desc_num_before_header_supported;
+
+	/* the desire field the driver selected to use */
+	u16 desc_num_before_header_enabled;
+
+	/* valid only if inline was chosen. bitfield of enum
+	 * ena_admin_llq_stride_ctrl
+	 */
+	u16 descriptors_stride_ctrl_supported;
+
+	/* the stride control the driver selected to use */
+	u16 descriptors_stride_ctrl_enabled;
+
+	/* reserved */
+	u32 reserved1;
+
+	/* accelerated low latency queues requirement. driver needs to
+	 * support those requirements in order to use accelerated llq
+	 */
+	struct ena_admin_accel_mode_req accel_mode;
+};
+
+struct ena_admin_queue_ext_feature_fields {
+	u32 max_tx_sq_num;
+
+	u32 max_tx_cq_num;
+
+	u32 max_rx_sq_num;
+
+	u32 max_rx_cq_num;
+
+	u32 max_tx_sq_depth;
+
+	u32 max_tx_cq_depth;
+
+	u32 max_rx_sq_depth;
+
+	u32 max_rx_cq_depth;
+
+	u32 max_tx_header_size;
+
+	/* Maximum Descriptors number, including meta descriptor, allowed for a
+	 * single Tx packet
+	 */
+	u16 max_per_packet_tx_descs;
+
+	/* Maximum Descriptors number allowed for a single Rx packet */
+	u16 max_per_packet_rx_descs;
+};
+
 struct ena_admin_queue_feature_desc {
-	/* including LLQs */
 	u32 max_sq_num;
 
 	u32 max_sq_depth;
@@ -491,14 +711,14 @@ struct ena_admin_queue_feature_desc {
 
 	u32 max_cq_depth;
 
-	u32 max_llq_num;
+	u32 max_legacy_llq_num;
 
-	u32 max_llq_depth;
+	u32 max_legacy_llq_depth;
 
 	u32 max_header_size;
 
-	/* Maximum Descriptors number, including meta descriptor, allowed for
-	 *    a single Tx packet
+	/* Maximum Descriptors number, including meta descriptor, allowed for a
+	 * single Tx packet
 	 */
 	u16 max_packet_tx_descs;
 
@@ -581,17 +801,16 @@ struct ena_admin_feature_offload_desc {
 };
 
 enum ena_admin_hash_functions {
-	ENA_ADMIN_TOEPLITZ	= 1,
-
-	ENA_ADMIN_CRC32		= 2,
+	ENA_ADMIN_TOEPLITZ                          = 1,
+	ENA_ADMIN_CRC32                             = 2,
 };
 
 struct ena_admin_feature_rss_flow_hash_control {
-	u32 keys_num;
+	u32 key_parts;
 
 	u32 reserved;
 
-	u32 key[10];
+	u32 key[ENA_ADMIN_RSS_KEY_PARTS];
 };
 
 struct ena_admin_feature_rss_flow_hash_function {
@@ -609,44 +828,35 @@ struct ena_admin_feature_rss_flow_hash_function {
 
 /* RSS flow hash protocols */
 enum ena_admin_flow_hash_proto {
-	ENA_ADMIN_RSS_TCP4	= 0,
-
-	ENA_ADMIN_RSS_UDP4	= 1,
-
-	ENA_ADMIN_RSS_TCP6	= 2,
-
-	ENA_ADMIN_RSS_UDP6	= 3,
-
-	ENA_ADMIN_RSS_IP4	= 4,
-
-	ENA_ADMIN_RSS_IP6	= 5,
-
-	ENA_ADMIN_RSS_IP4_FRAG	= 6,
-
-	ENA_ADMIN_RSS_NOT_IP	= 7,
-
-	ENA_ADMIN_RSS_PROTO_NUM	= 16,
+	ENA_ADMIN_RSS_TCP4                          = 0,
+	ENA_ADMIN_RSS_UDP4                          = 1,
+	ENA_ADMIN_RSS_TCP6                          = 2,
+	ENA_ADMIN_RSS_UDP6                          = 3,
+	ENA_ADMIN_RSS_IP4                           = 4,
+	ENA_ADMIN_RSS_IP6                           = 5,
+	ENA_ADMIN_RSS_IP4_FRAG                      = 6,
+	ENA_ADMIN_RSS_NOT_IP                        = 7,
+	/* TCPv6 with extension header */
+	ENA_ADMIN_RSS_TCP6_EX                       = 8,
+	/* IPv6 with extension header */
+	ENA_ADMIN_RSS_IP6_EX                        = 9,
+	ENA_ADMIN_RSS_PROTO_NUM                     = 16,
 };
 
 /* RSS flow hash fields */
 enum ena_admin_flow_hash_fields {
 	/* Ethernet Dest Addr */
-	ENA_ADMIN_RSS_L2_DA	= BIT(0),
-
+	ENA_ADMIN_RSS_L2_DA                         = BIT(0),
 	/* Ethernet Src Addr */
-	ENA_ADMIN_RSS_L2_SA	= BIT(1),
-
+	ENA_ADMIN_RSS_L2_SA                         = BIT(1),
 	/* ipv4/6 Dest Addr */
-	ENA_ADMIN_RSS_L3_DA	= BIT(2),
-
+	ENA_ADMIN_RSS_L3_DA                         = BIT(2),
 	/* ipv4/6 Src Addr */
-	ENA_ADMIN_RSS_L3_SA	= BIT(3),
-
+	ENA_ADMIN_RSS_L3_SA                         = BIT(3),
 	/* tcp/udp Dest Port */
-	ENA_ADMIN_RSS_L4_DP	= BIT(4),
-
+	ENA_ADMIN_RSS_L4_DP                         = BIT(4),
 	/* tcp/udp Src Port */
-	ENA_ADMIN_RSS_L4_SP	= BIT(5),
+	ENA_ADMIN_RSS_L4_SP                         = BIT(5),
 };
 
 struct ena_admin_proto_input {
@@ -685,15 +895,13 @@ struct ena_admin_feature_rss_flow_hash_input {
 };
 
 enum ena_admin_os_type {
-	ENA_ADMIN_OS_LINUX	= 1,
-
-	ENA_ADMIN_OS_WIN	= 2,
-
-	ENA_ADMIN_OS_DPDK	= 3,
-
-	ENA_ADMIN_OS_FREEBSD	= 4,
-
-	ENA_ADMIN_OS_IPXE	= 5,
+	ENA_ADMIN_OS_LINUX                          = 1,
+	ENA_ADMIN_OS_WIN                            = 2,
+	ENA_ADMIN_OS_DPDK                           = 3,
+	ENA_ADMIN_OS_FREEBSD                        = 4,
+	ENA_ADMIN_OS_IPXE                           = 5,
+	ENA_ADMIN_OS_ESXI                           = 6,
+	ENA_ADMIN_OS_GROUPS_NUM                     = 6,
 };
 
 struct ena_admin_host_info {
@@ -715,11 +923,40 @@ struct ena_admin_host_info {
 	/* 7:0 : major
 	 * 15:8 : minor
 	 * 23:16 : sub_minor
+	 * 31:24 : module_type
 	 */
 	u32 driver_version;
 
 	/* features bitmap */
-	u32 supported_network_features[4];
+	u32 supported_network_features[2];
+
+	/* ENA spec version of driver */
+	u16 ena_spec_version;
+
+	/* ENA device's Bus, Device and Function
+	 * 2:0 : function
+	 * 7:3 : device
+	 * 15:8 : bus
+	 */
+	u16 bdf;
+
+	/* Number of CPUs */
+	u16 num_cpus;
+
+	u16 reserved;
+
+	/* 0 : reserved
+	 * 1 : rx_offset
+	 * 2 : interrupt_moderation
+	 * 3 : rx_buf_mirroring
+	 * 4 : rss_configurable_function_key
+	 * 5 : reserved
+	 * 6 : rx_page_reuse
+	 * 7 : reserved
+	 * 8 : phc
+	 * 31:9 : reserved
+	 */
+	u32 driver_supported_features;
 };
 
 struct ena_admin_rss_ind_table_entry {
@@ -749,6 +986,31 @@ struct ena_admin_feature_rss_ind_table {
 	struct ena_admin_rss_ind_table_entry inline_entry;
 };
 
+/* When hint value is 0, driver should use its own predefined value */
+struct ena_admin_ena_hw_hints {
+	/* value in ms */
+	u16 mmio_read_timeout;
+
+	/* value in ms */
+	u16 driver_watchdog_timeout;
+
+	/* Per packet tx completion timeout. value in ms */
+	u16 missing_tx_completion_timeout;
+
+	u16 missed_tx_completion_count_threshold_to_reset;
+
+	/* value in ms */
+	u16 admin_completion_tx_timeout;
+
+	u16 netdev_wd_timeout;
+
+	u16 max_tx_sgl_size;
+
+	u16 max_rx_sgl_size;
+
+	u16 reserved[8];
+};
+
 struct ena_admin_get_feat_cmd {
 	struct ena_admin_aq_common_desc aq_common_descriptor;
 
@@ -759,6 +1021,56 @@ struct ena_admin_get_feat_cmd {
 	u32 raw[11];
 };
 
+struct ena_admin_queue_ext_feature_desc {
+	/* version */
+	u8 version;
+
+	u8 reserved1[3];
+
+	union {
+		struct ena_admin_queue_ext_feature_fields max_queue_ext;
+
+		u32 raw[10];
+	};
+};
+
+struct ena_admin_feature_phc_desc {
+	/* PHC type as defined in enum ena_admin_get_phc_type,
+	 * used only for GET command.
+	 */
+	u8 type;
+
+	/* Reserved - MBZ */
+	u8 reserved1[3];
+
+	/* PHC doorbell address as an offset to PCIe MMIO REG BAR,
+	 * used only for GET command.
+	 */
+	u32 doorbell_offset;
+
+	/* Max time for valid PHC retrieval, passing this threshold will
+	 * fail the get-time request and block PHC requests for
+	 * block_timeout_usec, used only for GET command.
+	 */
+	u32 expire_timeout_usec;
+
+	/* PHC requests block period, blocking starts if PHC request expired
+	 * in order to prevent floods on busy device,
+	 * used only for GET command.
+	 */
+	u32 block_timeout_usec;
+
+	/* Shared PHC physical address (ena_admin_phc_resp),
+	 * used only for SET command.
+	 */
+	struct ena_common_mem_addr output_address;
+
+	/* Shared PHC Size (ena_admin_phc_resp),
+	 * used only for SET command.
+	 */
+	u32 output_length;
+};
+
 struct ena_admin_get_feat_resp {
 	struct ena_admin_acq_common_desc acq_common_desc;
 
@@ -767,8 +1079,12 @@ struct ena_admin_get_feat_resp {
 
 		struct ena_admin_device_attr_feature_desc dev_attr;
 
+		struct ena_admin_feature_llq_desc llq;
+
 		struct ena_admin_queue_feature_desc max_queue;
 
+		struct ena_admin_queue_ext_feature_desc max_queue_ext;
+
 		struct ena_admin_feature_aenq_desc aenq;
 
 		struct ena_admin_get_feature_link_desc link;
@@ -782,6 +1098,10 @@ struct ena_admin_get_feat_resp {
 		struct ena_admin_feature_rss_ind_table ind_table;
 
 		struct ena_admin_feature_intr_moder_desc intr_moderation;
+
+		struct ena_admin_ena_hw_hints hw_hints;
+
+		struct ena_admin_feature_phc_desc phc;
 	} u;
 };
 
@@ -812,6 +1132,12 @@ struct ena_admin_set_feat_cmd {
 
 		/* rss indirection table */
 		struct ena_admin_feature_rss_ind_table ind_table;
+
+		/* LLQ configuration */
+		struct ena_admin_feature_llq_desc llq;
+
+		/* PHC configuration */
+		struct ena_admin_feature_phc_desc phc;
 	} u;
 };
 
@@ -826,9 +1152,11 @@ struct ena_admin_set_feat_resp {
 struct ena_admin_aenq_common_desc {
 	u16 group;
 
-	u16 syndrom;
+	u16 syndrome;
 
-	/* 0 : phase */
+	/* 0 : phase
+	 * 7:1 : reserved - MBZ
+	 */
 	u8 flags;
 
 	u8 reserved1[3];
@@ -840,23 +1168,16 @@ struct ena_admin_aenq_common_desc {
 
 /* asynchronous event notification groups */
 enum ena_admin_aenq_group {
-	ENA_ADMIN_LINK_CHANGE		= 0,
-
-	ENA_ADMIN_FATAL_ERROR		= 1,
-
-	ENA_ADMIN_WARNING		= 2,
-
-	ENA_ADMIN_NOTIFICATION		= 3,
-
-	ENA_ADMIN_KEEP_ALIVE		= 4,
-
-	ENA_ADMIN_AENQ_GROUPS_NUM	= 5,
+	ENA_ADMIN_LINK_CHANGE                       = 0,
+	ENA_ADMIN_FATAL_ERROR                       = 1,
+	ENA_ADMIN_WARNING                           = 2,
+	ENA_ADMIN_NOTIFICATION                      = 3,
+	ENA_ADMIN_KEEP_ALIVE                        = 4,
+	ENA_ADMIN_AENQ_GROUPS_NUM                   = 5,
 };
 
-enum ena_admin_aenq_notification_syndrom {
-	ENA_ADMIN_SUSPEND	= 0,
-
-	ENA_ADMIN_RESUME	= 1,
+enum ena_admin_aenq_notification_syndrome {
+	ENA_ADMIN_UPDATE_HINTS                      = 2,
 };
 
 struct ena_admin_aenq_entry {
@@ -879,6 +1200,10 @@ struct ena_admin_aenq_keep_alive_desc {
 	u32 rx_drops_low;
 
 	u32 rx_drops_high;
+
+	u32 tx_drops_low;
+
+	u32 tx_drops_high;
 };
 
 struct ena_admin_ena_mmio_req_read_less_resp {
@@ -890,28 +1215,45 @@ struct ena_admin_ena_mmio_req_read_less_resp {
 	u32 reg_val;
 };
 
+struct ena_admin_phc_resp {
+	/* Request Id, received from DB register */
+	u16 req_id;
+
+	u8 reserved1[6];
+
+	/* PHC timestamp (nsec) */
+	u64 timestamp;
+
+	u8 reserved2[12];
+
+	/* Bit field of enum ena_admin_phc_error_flags */
+	u32 error_flags;
+
+	u8 reserved3[32];
+};
+
 /* aq_common_desc */
-#define ENA_ADMIN_AQ_COMMON_DESC_COMMAND_ID_MASK GENMASK(11, 0)
-#define ENA_ADMIN_AQ_COMMON_DESC_PHASE_MASK BIT(0)
-#define ENA_ADMIN_AQ_COMMON_DESC_CTRL_DATA_SHIFT 1
-#define ENA_ADMIN_AQ_COMMON_DESC_CTRL_DATA_MASK BIT(1)
-#define ENA_ADMIN_AQ_COMMON_DESC_CTRL_DATA_INDIRECT_SHIFT 2
-#define ENA_ADMIN_AQ_COMMON_DESC_CTRL_DATA_INDIRECT_MASK BIT(2)
+#define ENA_ADMIN_AQ_COMMON_DESC_COMMAND_ID_MASK            GENMASK(11, 0)
+#define ENA_ADMIN_AQ_COMMON_DESC_PHASE_MASK                 BIT(0)
+#define ENA_ADMIN_AQ_COMMON_DESC_CTRL_DATA_SHIFT            1
+#define ENA_ADMIN_AQ_COMMON_DESC_CTRL_DATA_MASK             BIT(1)
+#define ENA_ADMIN_AQ_COMMON_DESC_CTRL_DATA_INDIRECT_SHIFT   2
+#define ENA_ADMIN_AQ_COMMON_DESC_CTRL_DATA_INDIRECT_MASK    BIT(2)
 
 /* sq */
-#define ENA_ADMIN_SQ_SQ_DIRECTION_SHIFT 5
-#define ENA_ADMIN_SQ_SQ_DIRECTION_MASK GENMASK(7, 5)
+#define ENA_ADMIN_SQ_SQ_DIRECTION_SHIFT                     5
+#define ENA_ADMIN_SQ_SQ_DIRECTION_MASK                      GENMASK(7, 5)
 
 /* acq_common_desc */
-#define ENA_ADMIN_ACQ_COMMON_DESC_COMMAND_ID_MASK GENMASK(11, 0)
-#define ENA_ADMIN_ACQ_COMMON_DESC_PHASE_MASK BIT(0)
+#define ENA_ADMIN_ACQ_COMMON_DESC_COMMAND_ID_MASK           GENMASK(11, 0)
+#define ENA_ADMIN_ACQ_COMMON_DESC_PHASE_MASK                BIT(0)
 
 /* aq_create_sq_cmd */
-#define ENA_ADMIN_AQ_CREATE_SQ_CMD_SQ_DIRECTION_SHIFT 5
-#define ENA_ADMIN_AQ_CREATE_SQ_CMD_SQ_DIRECTION_MASK GENMASK(7, 5)
-#define ENA_ADMIN_AQ_CREATE_SQ_CMD_PLACEMENT_POLICY_MASK GENMASK(3, 0)
-#define ENA_ADMIN_AQ_CREATE_SQ_CMD_COMPLETION_POLICY_SHIFT 4
-#define ENA_ADMIN_AQ_CREATE_SQ_CMD_COMPLETION_POLICY_MASK GENMASK(6, 4)
+#define ENA_ADMIN_AQ_CREATE_SQ_CMD_SQ_DIRECTION_SHIFT       5
+#define ENA_ADMIN_AQ_CREATE_SQ_CMD_SQ_DIRECTION_MASK        GENMASK(7, 5)
+#define ENA_ADMIN_AQ_CREATE_SQ_CMD_PLACEMENT_POLICY_MASK    GENMASK(3, 0)
+#define ENA_ADMIN_AQ_CREATE_SQ_CMD_COMPLETION_POLICY_SHIFT  4
+#define ENA_ADMIN_AQ_CREATE_SQ_CMD_COMPLETION_POLICY_MASK   GENMASK(6, 4)
 #define ENA_ADMIN_AQ_CREATE_SQ_CMD_IS_PHYSICALLY_CONTIGUOUS_MASK BIT(0)
 
 /* aq_create_cq_cmd */
@@ -920,12 +1262,12 @@ struct ena_admin_ena_mmio_req_read_less_resp {
 #define ENA_ADMIN_AQ_CREATE_CQ_CMD_CQ_ENTRY_SIZE_WORDS_MASK GENMASK(4, 0)
 
 /* get_set_feature_common_desc */
-#define ENA_ADMIN_GET_SET_FEATURE_COMMON_DESC_SELECT_MASK GENMASK(1, 0)
+#define ENA_ADMIN_GET_SET_FEATURE_COMMON_DESC_SELECT_MASK   GENMASK(1, 0)
 
 /* get_feature_link_desc */
-#define ENA_ADMIN_GET_FEATURE_LINK_DESC_AUTONEG_MASK BIT(0)
-#define ENA_ADMIN_GET_FEATURE_LINK_DESC_DUPLEX_SHIFT 1
-#define ENA_ADMIN_GET_FEATURE_LINK_DESC_DUPLEX_MASK BIT(1)
+#define ENA_ADMIN_GET_FEATURE_LINK_DESC_AUTONEG_MASK        BIT(0)
+#define ENA_ADMIN_GET_FEATURE_LINK_DESC_DUPLEX_SHIFT        1
+#define ENA_ADMIN_GET_FEATURE_LINK_DESC_DUPLEX_MASK         BIT(1)
 
 /* feature_offload_desc */
 #define ENA_ADMIN_FEATURE_OFFLOAD_DESC_TX_L3_CSUM_IPV4_MASK BIT(0)
@@ -937,19 +1279,19 @@ struct ena_admin_ena_mmio_req_read_less_resp {
 #define ENA_ADMIN_FEATURE_OFFLOAD_DESC_TX_L4_IPV6_CSUM_PART_MASK BIT(3)
 #define ENA_ADMIN_FEATURE_OFFLOAD_DESC_TX_L4_IPV6_CSUM_FULL_SHIFT 4
 #define ENA_ADMIN_FEATURE_OFFLOAD_DESC_TX_L4_IPV6_CSUM_FULL_MASK BIT(4)
-#define ENA_ADMIN_FEATURE_OFFLOAD_DESC_TSO_IPV4_SHIFT 5
-#define ENA_ADMIN_FEATURE_OFFLOAD_DESC_TSO_IPV4_MASK BIT(5)
-#define ENA_ADMIN_FEATURE_OFFLOAD_DESC_TSO_IPV6_SHIFT 6
-#define ENA_ADMIN_FEATURE_OFFLOAD_DESC_TSO_IPV6_MASK BIT(6)
-#define ENA_ADMIN_FEATURE_OFFLOAD_DESC_TSO_ECN_SHIFT 7
-#define ENA_ADMIN_FEATURE_OFFLOAD_DESC_TSO_ECN_MASK BIT(7)
+#define ENA_ADMIN_FEATURE_OFFLOAD_DESC_TSO_IPV4_SHIFT       5
+#define ENA_ADMIN_FEATURE_OFFLOAD_DESC_TSO_IPV4_MASK        BIT(5)
+#define ENA_ADMIN_FEATURE_OFFLOAD_DESC_TSO_IPV6_SHIFT       6
+#define ENA_ADMIN_FEATURE_OFFLOAD_DESC_TSO_IPV6_MASK        BIT(6)
+#define ENA_ADMIN_FEATURE_OFFLOAD_DESC_TSO_ECN_SHIFT        7
+#define ENA_ADMIN_FEATURE_OFFLOAD_DESC_TSO_ECN_MASK         BIT(7)
 #define ENA_ADMIN_FEATURE_OFFLOAD_DESC_RX_L3_CSUM_IPV4_MASK BIT(0)
 #define ENA_ADMIN_FEATURE_OFFLOAD_DESC_RX_L4_IPV4_CSUM_SHIFT 1
 #define ENA_ADMIN_FEATURE_OFFLOAD_DESC_RX_L4_IPV4_CSUM_MASK BIT(1)
 #define ENA_ADMIN_FEATURE_OFFLOAD_DESC_RX_L4_IPV6_CSUM_SHIFT 2
 #define ENA_ADMIN_FEATURE_OFFLOAD_DESC_RX_L4_IPV6_CSUM_MASK BIT(2)
-#define ENA_ADMIN_FEATURE_OFFLOAD_DESC_RX_HASH_SHIFT 3
-#define ENA_ADMIN_FEATURE_OFFLOAD_DESC_RX_HASH_MASK BIT(3)
+#define ENA_ADMIN_FEATURE_OFFLOAD_DESC_RX_HASH_SHIFT        3
+#define ENA_ADMIN_FEATURE_OFFLOAD_DESC_RX_HASH_MASK         BIT(3)
 
 /* feature_rss_flow_hash_function */
 #define ENA_ADMIN_FEATURE_RSS_FLOW_HASH_FUNCTION_FUNCS_MASK GENMASK(7, 0)
@@ -957,25 +1299,44 @@ struct ena_admin_ena_mmio_req_read_less_resp {
 
 /* feature_rss_flow_hash_input */
 #define ENA_ADMIN_FEATURE_RSS_FLOW_HASH_INPUT_L3_SORT_SHIFT 1
-#define ENA_ADMIN_FEATURE_RSS_FLOW_HASH_INPUT_L3_SORT_MASK BIT(1)
+#define ENA_ADMIN_FEATURE_RSS_FLOW_HASH_INPUT_L3_SORT_MASK  BIT(1)
 #define ENA_ADMIN_FEATURE_RSS_FLOW_HASH_INPUT_L4_SORT_SHIFT 2
-#define ENA_ADMIN_FEATURE_RSS_FLOW_HASH_INPUT_L4_SORT_MASK BIT(2)
+#define ENA_ADMIN_FEATURE_RSS_FLOW_HASH_INPUT_L4_SORT_MASK  BIT(2)
 #define ENA_ADMIN_FEATURE_RSS_FLOW_HASH_INPUT_ENABLE_L3_SORT_SHIFT 1
 #define ENA_ADMIN_FEATURE_RSS_FLOW_HASH_INPUT_ENABLE_L3_SORT_MASK BIT(1)
 #define ENA_ADMIN_FEATURE_RSS_FLOW_HASH_INPUT_ENABLE_L4_SORT_SHIFT 2
 #define ENA_ADMIN_FEATURE_RSS_FLOW_HASH_INPUT_ENABLE_L4_SORT_MASK BIT(2)
 
 /* host_info */
-#define ENA_ADMIN_HOST_INFO_MAJOR_MASK GENMASK(7, 0)
-#define ENA_ADMIN_HOST_INFO_MINOR_SHIFT 8
-#define ENA_ADMIN_HOST_INFO_MINOR_MASK GENMASK(15, 8)
-#define ENA_ADMIN_HOST_INFO_SUB_MINOR_SHIFT 16
-#define ENA_ADMIN_HOST_INFO_SUB_MINOR_MASK GENMASK(23, 16)
+#define ENA_ADMIN_HOST_INFO_MAJOR_MASK                      GENMASK(7, 0)
+#define ENA_ADMIN_HOST_INFO_MINOR_SHIFT                     8
+#define ENA_ADMIN_HOST_INFO_MINOR_MASK                      GENMASK(15, 8)
+#define ENA_ADMIN_HOST_INFO_SUB_MINOR_SHIFT                 16
+#define ENA_ADMIN_HOST_INFO_SUB_MINOR_MASK                  GENMASK(23, 16)
+#define ENA_ADMIN_HOST_INFO_MODULE_TYPE_SHIFT               24
+#define ENA_ADMIN_HOST_INFO_MODULE_TYPE_MASK                GENMASK(31, 24)
+#define ENA_ADMIN_HOST_INFO_FUNCTION_MASK                   GENMASK(2, 0)
+#define ENA_ADMIN_HOST_INFO_DEVICE_SHIFT                    3
+#define ENA_ADMIN_HOST_INFO_DEVICE_MASK                     GENMASK(7, 3)
+#define ENA_ADMIN_HOST_INFO_BUS_SHIFT                       8
+#define ENA_ADMIN_HOST_INFO_BUS_MASK                        GENMASK(15, 8)
+#define ENA_ADMIN_HOST_INFO_RX_OFFSET_SHIFT                 1
+#define ENA_ADMIN_HOST_INFO_RX_OFFSET_MASK                  BIT(1)
+#define ENA_ADMIN_HOST_INFO_INTERRUPT_MODERATION_SHIFT      2
+#define ENA_ADMIN_HOST_INFO_INTERRUPT_MODERATION_MASK       BIT(2)
+#define ENA_ADMIN_HOST_INFO_RX_BUF_MIRRORING_SHIFT          3
+#define ENA_ADMIN_HOST_INFO_RX_BUF_MIRRORING_MASK           BIT(3)
+#define ENA_ADMIN_HOST_INFO_RSS_CONFIGURABLE_FUNCTION_KEY_SHIFT 4
+#define ENA_ADMIN_HOST_INFO_RSS_CONFIGURABLE_FUNCTION_KEY_MASK BIT(4)
+#define ENA_ADMIN_HOST_INFO_RX_PAGE_REUSE_SHIFT             6
+#define ENA_ADMIN_HOST_INFO_RX_PAGE_REUSE_MASK              BIT(6)
+#define ENA_ADMIN_HOST_INFO_PHC_SHIFT                       8
+#define ENA_ADMIN_HOST_INFO_PHC_MASK                        BIT(8)
 
 /* aenq_common_desc */
-#define ENA_ADMIN_AENQ_COMMON_DESC_PHASE_MASK BIT(0)
+#define ENA_ADMIN_AENQ_COMMON_DESC_PHASE_MASK               BIT(0)
 
 /* aenq_link_change_desc */
-#define ENA_ADMIN_AENQ_LINK_CHANGE_DESC_LINK_STATUS_MASK BIT(0)
+#define ENA_ADMIN_AENQ_LINK_CHANGE_DESC_LINK_STATUS_MASK    BIT(0)
 
-#endif /*_ENA_ADMIN_H_ */
+#endif /* _ENA_ADMIN_H_ */
diff --git a/drivers/net/ethernet/amazon/ena/ena_com.c b/drivers/net/ethernet/amazon/ena/ena_com.c
index 08d11cede9c9..e67b592e5697 100644
--- a/drivers/net/ethernet/amazon/ena/ena_com.c
+++ b/drivers/net/ethernet/amazon/ena/ena_com.c
@@ -1,33 +1,6 @@
+// SPDX-License-Identifier: GPL-2.0 OR Linux-OpenIB
 /*
- * Copyright 2015 Amazon.com, Inc. or its affiliates.
- *
- * This software is available to you under a choice of one of two
- * licenses.  You may choose to be licensed under the terms of the GNU
- * General Public License (GPL) Version 2, available from the file
- * COPYING in the main directory of this source tree, or the
- * BSD license below:
- *
- *     Redistribution and use in source and binary forms, with or
- *     without modification, are permitted provided that the following
- *     conditions are met:
- *
- *      - Redistributions of source code must retain the above
- *        copyright notice, this list of conditions and the following
- *        disclaimer.
- *
- *      - Redistributions in binary form must reproduce the above
- *        copyright notice, this list of conditions and the following
- *        disclaimer in the documentation and/or other materials
- *        provided with the distribution.
- *
- * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND,
- * EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF
- * MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND
- * NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS
- * BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN
- * ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN
- * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
- * SOFTWARE.
+ * Copyright 2015-2020 Amazon.com, Inc. or its affiliates. All rights reserved.
  */
 
 #include "ena_com.h"
@@ -41,9 +14,6 @@
 #define ENA_ASYNC_QUEUE_DEPTH 16
 #define ENA_ADMIN_QUEUE_DEPTH 32
 
-#define MIN_ENA_VER (((ENA_COMMON_SPEC_VERSION_MAJOR) << \
-		ENA_REGS_VERSION_MAJOR_VERSION_SHIFT) \
-		| (ENA_COMMON_SPEC_VERSION_MINOR))
 
 #define ENA_CTRL_MAJOR		0
 #define ENA_CTRL_MINOR		0
@@ -61,6 +31,22 @@
 
 #define ENA_MMIO_READ_TIMEOUT 0xFFFFFFFF
 
+#define ENA_COM_BOUNCE_BUFFER_CNTRL_CNT	4
+
+#define ENA_REGS_ADMIN_INTR_MASK 1
+
+#define ENA_MAX_BACKOFF_DELAY_EXP 16U
+
+#define ENA_MIN_ADMIN_POLL_US 100
+
+#define ENA_MAX_ADMIN_POLL_US 5000
+
+/* PHC definitions */
+#define ENA_PHC_DEFAULT_EXPIRE_TIMEOUT_USEC 10
+#define ENA_PHC_DEFAULT_BLOCK_TIMEOUT_USEC 1000
+#define ENA_PHC_REQ_ID_OFFSET 0xDEAD
+#define ENA_PHC_ERROR_FLAGS (ENA_ADMIN_PHC_ERROR_FLAG_TIMESTAMP)
+
 /*****************************************************************************/
 /*****************************************************************************/
 /*****************************************************************************/
@@ -88,31 +74,32 @@ struct ena_com_stats_ctx {
 	struct ena_admin_acq_get_stats_resp get_resp;
 };
 
-static inline int ena_com_mem_addr_set(struct ena_com_dev *ena_dev,
+static int ena_com_mem_addr_set(struct ena_com_dev *ena_dev,
 				       struct ena_common_mem_addr *ena_addr,
 				       dma_addr_t addr)
 {
 	if ((addr & GENMASK_ULL(ena_dev->dma_addr_bits - 1, 0)) != addr) {
-		pr_err("dma address has more bits that the device supports\n");
+		netdev_err(ena_dev->net_device,
+			   "DMA address has more bits that the device supports\n");
 		return -EINVAL;
 	}
 
-	ena_addr->mem_addr_low = (u32)addr;
-	ena_addr->mem_addr_high = (u64)addr >> 32;
+	ena_addr->mem_addr_low = lower_32_bits(addr);
+	ena_addr->mem_addr_high = (u16)upper_32_bits(addr);
 
 	return 0;
 }
 
-static int ena_com_admin_init_sq(struct ena_com_admin_queue *queue)
+static int ena_com_admin_init_sq(struct ena_com_admin_queue *admin_queue)
 {
-	struct ena_com_admin_sq *sq = &queue->sq;
-	u16 size = ADMIN_SQ_SIZE(queue->q_depth);
+	struct ena_com_dev *ena_dev = admin_queue->ena_dev;
+	struct ena_com_admin_sq *sq = &admin_queue->sq;
+	u16 size = ADMIN_SQ_SIZE(admin_queue->q_depth);
 
-	sq->entries = dma_zalloc_coherent(queue->q_dmadev, size, &sq->dma_addr,
-					  GFP_KERNEL);
+	sq->entries = dma_alloc_coherent(admin_queue->q_dmadev, size, &sq->dma_addr, GFP_KERNEL);
 
 	if (!sq->entries) {
-		pr_err("memory allocation failed");
+		netdev_err(ena_dev->net_device, "Memory allocation failed\n");
 		return -ENOMEM;
 	}
 
@@ -125,16 +112,16 @@ static int ena_com_admin_init_sq(struct ena_com_admin_queue *queue)
 	return 0;
 }
 
-static int ena_com_admin_init_cq(struct ena_com_admin_queue *queue)
+static int ena_com_admin_init_cq(struct ena_com_admin_queue *admin_queue)
 {
-	struct ena_com_admin_cq *cq = &queue->cq;
-	u16 size = ADMIN_CQ_SIZE(queue->q_depth);
+	struct ena_com_dev *ena_dev = admin_queue->ena_dev;
+	struct ena_com_admin_cq *cq = &admin_queue->cq;
+	u16 size = ADMIN_CQ_SIZE(admin_queue->q_depth);
 
-	cq->entries = dma_zalloc_coherent(queue->q_dmadev, size, &cq->dma_addr,
-					  GFP_KERNEL);
+	cq->entries = dma_alloc_coherent(admin_queue->q_dmadev, size, &cq->dma_addr, GFP_KERNEL);
 
 	if (!cq->entries) {
-		pr_err("memory allocation failed");
+		netdev_err(ena_dev->net_device, "Memory allocation failed\n");
 		return -ENOMEM;
 	}
 
@@ -144,20 +131,19 @@ static int ena_com_admin_init_cq(struct ena_com_admin_queue *queue)
 	return 0;
 }
 
-static int ena_com_admin_init_aenq(struct ena_com_dev *dev,
+static int ena_com_admin_init_aenq(struct ena_com_dev *ena_dev,
 				   struct ena_aenq_handlers *aenq_handlers)
 {
-	struct ena_com_aenq *aenq = &dev->aenq;
+	struct ena_com_aenq *aenq = &ena_dev->aenq;
 	u32 addr_low, addr_high, aenq_caps;
 	u16 size;
 
-	dev->aenq.q_depth = ENA_ASYNC_QUEUE_DEPTH;
+	ena_dev->aenq.q_depth = ENA_ASYNC_QUEUE_DEPTH;
 	size = ADMIN_AENQ_SIZE(ENA_ASYNC_QUEUE_DEPTH);
-	aenq->entries = dma_zalloc_coherent(dev->dmadev, size, &aenq->dma_addr,
-					    GFP_KERNEL);
+	aenq->entries = dma_alloc_coherent(ena_dev->dmadev, size, &aenq->dma_addr, GFP_KERNEL);
 
 	if (!aenq->entries) {
-		pr_err("memory allocation failed");
+		netdev_err(ena_dev->net_device, "Memory allocation failed\n");
 		return -ENOMEM;
 	}
 
@@ -167,18 +153,18 @@ static int ena_com_admin_init_aenq(struct ena_com_dev *dev,
 	addr_low = ENA_DMA_ADDR_TO_UINT32_LOW(aenq->dma_addr);
 	addr_high = ENA_DMA_ADDR_TO_UINT32_HIGH(aenq->dma_addr);
 
-	writel(addr_low, dev->reg_bar + ENA_REGS_AENQ_BASE_LO_OFF);
-	writel(addr_high, dev->reg_bar + ENA_REGS_AENQ_BASE_HI_OFF);
+	writel(addr_low, ena_dev->reg_bar + ENA_REGS_AENQ_BASE_LO_OFF);
+	writel(addr_high, ena_dev->reg_bar + ENA_REGS_AENQ_BASE_HI_OFF);
 
 	aenq_caps = 0;
-	aenq_caps |= dev->aenq.q_depth & ENA_REGS_AENQ_CAPS_AENQ_DEPTH_MASK;
-	aenq_caps |= (sizeof(struct ena_admin_aenq_entry)
-		      << ENA_REGS_AENQ_CAPS_AENQ_ENTRY_SIZE_SHIFT) &
-		     ENA_REGS_AENQ_CAPS_AENQ_ENTRY_SIZE_MASK;
-	writel(aenq_caps, dev->reg_bar + ENA_REGS_AENQ_CAPS_OFF);
+	aenq_caps |= ena_dev->aenq.q_depth & ENA_REGS_AENQ_CAPS_AENQ_DEPTH_MASK;
+	aenq_caps |=
+		(sizeof(struct ena_admin_aenq_entry) << ENA_REGS_AENQ_CAPS_AENQ_ENTRY_SIZE_SHIFT) &
+		ENA_REGS_AENQ_CAPS_AENQ_ENTRY_SIZE_MASK;
+	writel(aenq_caps, ena_dev->reg_bar + ENA_REGS_AENQ_CAPS_OFF);
 
 	if (unlikely(!aenq_handlers)) {
-		pr_err("aenq handlers pointer is NULL\n");
+		netdev_err(ena_dev->net_device, "AENQ handlers pointer is NULL\n");
 		return -EINVAL;
 	}
 
@@ -187,33 +173,39 @@ static int ena_com_admin_init_aenq(struct ena_com_dev *dev,
 	return 0;
 }
 
-static inline void comp_ctxt_release(struct ena_com_admin_queue *queue,
+static void comp_ctxt_release(struct ena_com_admin_queue *queue,
 				     struct ena_comp_ctx *comp_ctx)
 {
 	comp_ctx->occupied = false;
 	atomic_dec(&queue->outstanding_cmds);
 }
 
-static struct ena_comp_ctx *get_comp_ctxt(struct ena_com_admin_queue *queue,
+static struct ena_comp_ctx *get_comp_ctxt(struct ena_com_admin_queue *admin_queue,
 					  u16 command_id, bool capture)
 {
-	if (unlikely(command_id >= queue->q_depth)) {
-		pr_err("command id is larger than the queue size. cmd_id: %u queue size %d\n",
-		       command_id, queue->q_depth);
+	if (unlikely(command_id >= admin_queue->q_depth)) {
+		netdev_err(admin_queue->ena_dev->net_device,
+			   "Command id is larger than the queue size. cmd_id: %u queue size %d\n",
+			   command_id, admin_queue->q_depth);
 		return NULL;
 	}
 
-	if (unlikely(queue->comp_ctx[command_id].occupied && capture)) {
-		pr_err("Completion context is occupied\n");
+	if (unlikely(!admin_queue->comp_ctx)) {
+		netdev_err(admin_queue->ena_dev->net_device, "Completion context is NULL\n");
+		return NULL;
+	}
+
+	if (unlikely(admin_queue->comp_ctx[command_id].occupied && capture)) {
+		netdev_err(admin_queue->ena_dev->net_device, "Completion context is occupied\n");
 		return NULL;
 	}
 
 	if (capture) {
-		atomic_inc(&queue->outstanding_cmds);
-		queue->comp_ctx[command_id].occupied = true;
+		atomic_inc(&admin_queue->outstanding_cmds);
+		admin_queue->comp_ctx[command_id].occupied = true;
 	}
 
-	return &queue->comp_ctx[command_id];
+	return &admin_queue->comp_ctx[command_id];
 }
 
 static struct ena_comp_ctx *__ena_com_submit_admin_cmd(struct ena_com_admin_queue *admin_queue,
@@ -232,11 +224,9 @@ static struct ena_comp_ctx *__ena_com_submit_admin_cmd(struct ena_com_admin_queu
 	tail_masked = admin_queue->sq.tail & queue_size_mask;
 
 	/* In case of queue FULL */
-	cnt = admin_queue->sq.tail - admin_queue->sq.head;
+	cnt = (u16)atomic_read(&admin_queue->outstanding_cmds);
 	if (cnt >= admin_queue->q_depth) {
-		pr_debug("admin queue is FULL (tail %d head %d depth: %d)\n",
-			 admin_queue->sq.tail, admin_queue->sq.head,
-			 admin_queue->q_depth);
+		netdev_dbg(admin_queue->ena_dev->net_device, "Admin queue is full.\n");
 		admin_queue->stats.out_of_space++;
 		return ERR_PTR(-ENOSPC);
 	}
@@ -276,20 +266,21 @@ static struct ena_comp_ctx *__ena_com_submit_admin_cmd(struct ena_com_admin_queu
 	return comp_ctx;
 }
 
-static inline int ena_com_init_comp_ctxt(struct ena_com_admin_queue *queue)
+static int ena_com_init_comp_ctxt(struct ena_com_admin_queue *admin_queue)
 {
-	size_t size = queue->q_depth * sizeof(struct ena_comp_ctx);
+	struct ena_com_dev *ena_dev = admin_queue->ena_dev;
+	size_t size = admin_queue->q_depth * sizeof(struct ena_comp_ctx);
 	struct ena_comp_ctx *comp_ctx;
 	u16 i;
 
-	queue->comp_ctx = devm_kzalloc(queue->q_dmadev, size, GFP_KERNEL);
-	if (unlikely(!queue->comp_ctx)) {
-		pr_err("memory allocation failed");
+	admin_queue->comp_ctx = devm_kzalloc(admin_queue->q_dmadev, size, GFP_KERNEL);
+	if (unlikely(!admin_queue->comp_ctx)) {
+		netdev_err(ena_dev->net_device, "Memory allocation failed\n");
 		return -ENOMEM;
 	}
 
-	for (i = 0; i < queue->q_depth; i++) {
-		comp_ctx = get_comp_ctxt(queue, i, false);
+	for (i = 0; i < admin_queue->q_depth; i++) {
+		comp_ctx = get_comp_ctxt(admin_queue, i, false);
 		if (comp_ctx)
 			init_completion(&comp_ctx->wait_event);
 	}
@@ -303,7 +294,7 @@ static struct ena_comp_ctx *ena_com_submit_admin_cmd(struct ena_com_admin_queue
 						     struct ena_admin_acq_entry *comp,
 						     size_t comp_size_in_bytes)
 {
-	unsigned long flags;
+	unsigned long flags = 0;
 	struct ena_comp_ctx *comp_ctx;
 
 	spin_lock_irqsave(&admin_queue->q_lock, flags);
@@ -315,7 +306,7 @@ static struct ena_comp_ctx *ena_com_submit_admin_cmd(struct ena_com_admin_queue
 					      cmd_size_in_bytes,
 					      comp,
 					      comp_size_in_bytes);
-	if (unlikely(IS_ERR(comp_ctx)))
+	if (IS_ERR(comp_ctx))
 		admin_queue->running_state = false;
 	spin_unlock_irqrestore(&admin_queue->q_lock, flags);
 
@@ -327,10 +318,10 @@ static int ena_com_init_io_sq(struct ena_com_dev *ena_dev,
 			      struct ena_com_io_sq *io_sq)
 {
 	size_t size;
-	int dev_node = 0;
 
-	memset(&io_sq->desc_addr, 0x0, sizeof(struct ena_com_io_desc_addr));
+	memset(&io_sq->desc_addr, 0x0, sizeof(io_sq->desc_addr));
 
+	io_sq->dma_addr_bits = (u8)ena_dev->dma_addr_bits;
 	io_sq->desc_entry_size =
 		(io_sq->direction == ENA_COM_IO_QUEUE_DIRECTION_TX) ?
 		sizeof(struct ena_eth_io_tx_desc) :
@@ -339,34 +330,58 @@ static int ena_com_init_io_sq(struct ena_com_dev *ena_dev,
 	size = io_sq->desc_entry_size * io_sq->q_depth;
 
 	if (io_sq->mem_queue_type == ENA_ADMIN_PLACEMENT_POLICY_HOST) {
-		dev_node = dev_to_node(ena_dev->dmadev);
-		set_dev_node(ena_dev->dmadev, ctx->numa_node);
 		io_sq->desc_addr.virt_addr =
-			dma_zalloc_coherent(ena_dev->dmadev, size,
-					    &io_sq->desc_addr.phys_addr,
-					    GFP_KERNEL);
-		set_dev_node(ena_dev->dmadev, dev_node);
+			dma_alloc_coherent(ena_dev->dmadev, size, &io_sq->desc_addr.phys_addr,
+					   GFP_KERNEL);
 		if (!io_sq->desc_addr.virt_addr) {
 			io_sq->desc_addr.virt_addr =
-				dma_zalloc_coherent(ena_dev->dmadev, size,
-						    &io_sq->desc_addr.phys_addr,
-						    GFP_KERNEL);
+				dma_alloc_coherent(ena_dev->dmadev, size,
+						   &io_sq->desc_addr.phys_addr, GFP_KERNEL);
 		}
-	} else {
-		dev_node = dev_to_node(ena_dev->dmadev);
-		set_dev_node(ena_dev->dmadev, ctx->numa_node);
-		io_sq->desc_addr.virt_addr =
-			devm_kzalloc(ena_dev->dmadev, size, GFP_KERNEL);
-		set_dev_node(ena_dev->dmadev, dev_node);
+
 		if (!io_sq->desc_addr.virt_addr) {
-			io_sq->desc_addr.virt_addr =
-				devm_kzalloc(ena_dev->dmadev, size, GFP_KERNEL);
+			netdev_err(ena_dev->net_device, "Memory allocation failed\n");
+			return -ENOMEM;
 		}
 	}
 
-	if (!io_sq->desc_addr.virt_addr) {
-		pr_err("memory allocation failed");
-		return -ENOMEM;
+	if (io_sq->mem_queue_type == ENA_ADMIN_PLACEMENT_POLICY_DEV) {
+		/* Allocate bounce buffers */
+		io_sq->bounce_buf_ctrl.buffer_size =
+			ena_dev->llq_info.desc_list_entry_size;
+		io_sq->bounce_buf_ctrl.buffers_num =
+			ENA_COM_BOUNCE_BUFFER_CNTRL_CNT;
+		io_sq->bounce_buf_ctrl.next_to_use = 0;
+
+		size = (size_t)io_sq->bounce_buf_ctrl.buffer_size *
+			io_sq->bounce_buf_ctrl.buffers_num;
+
+		io_sq->bounce_buf_ctrl.base_buffer = devm_kzalloc(ena_dev->dmadev, size, GFP_KERNEL);
+		if (!io_sq->bounce_buf_ctrl.base_buffer)
+			io_sq->bounce_buf_ctrl.base_buffer =
+				devm_kzalloc(ena_dev->dmadev, size, GFP_KERNEL);
+
+		if (!io_sq->bounce_buf_ctrl.base_buffer) {
+			netdev_err(ena_dev->net_device, "Bounce buffer memory allocation failed\n");
+			return -ENOMEM;
+		}
+
+		memcpy(&io_sq->llq_info, &ena_dev->llq_info,
+		       sizeof(io_sq->llq_info));
+
+		/* Initiate the first bounce buffer */
+		io_sq->llq_buf_ctrl.curr_bounce_buf =
+			ena_com_get_next_bounce_buffer(&io_sq->bounce_buf_ctrl);
+		memset(io_sq->llq_buf_ctrl.curr_bounce_buf,
+		       0x0, io_sq->llq_info.desc_list_entry_size);
+		io_sq->llq_buf_ctrl.descs_left_in_line =
+			io_sq->llq_info.descs_num_before_header;
+		io_sq->disable_meta_caching =
+			io_sq->llq_info.disable_meta_caching;
+
+		if (io_sq->llq_info.max_entries_in_tx_burst > 0)
+			io_sq->entries_in_tx_burst_left =
+				io_sq->llq_info.max_entries_in_tx_burst;
 	}
 
 	io_sq->tail = 0;
@@ -381,9 +396,8 @@ static int ena_com_init_io_cq(struct ena_com_dev *ena_dev,
 			      struct ena_com_io_cq *io_cq)
 {
 	size_t size;
-	int prev_node = 0;
 
-	memset(&io_cq->cdesc_addr, 0x0, sizeof(struct ena_com_io_desc_addr));
+	memset(&io_cq->cdesc_addr, 0x0, sizeof(io_cq->cdesc_addr));
 
 	/* Use the basic completion descriptor for Rx */
 	io_cq->cdesc_entry_size_in_bytes =
@@ -393,21 +407,16 @@ static int ena_com_init_io_cq(struct ena_com_dev *ena_dev,
 
 	size = io_cq->cdesc_entry_size_in_bytes * io_cq->q_depth;
 
-	prev_node = dev_to_node(ena_dev->dmadev);
-	set_dev_node(ena_dev->dmadev, ctx->numa_node);
 	io_cq->cdesc_addr.virt_addr =
-		dma_zalloc_coherent(ena_dev->dmadev, size,
-				    &io_cq->cdesc_addr.phys_addr, GFP_KERNEL);
-	set_dev_node(ena_dev->dmadev, prev_node);
+		dma_alloc_coherent(ena_dev->dmadev, size, &io_cq->cdesc_addr.phys_addr, GFP_KERNEL);
 	if (!io_cq->cdesc_addr.virt_addr) {
 		io_cq->cdesc_addr.virt_addr =
-			dma_zalloc_coherent(ena_dev->dmadev, size,
-					    &io_cq->cdesc_addr.phys_addr,
-					    GFP_KERNEL);
+			dma_alloc_coherent(ena_dev->dmadev, size, &io_cq->cdesc_addr.phys_addr,
+					   GFP_KERNEL);
 	}
 
 	if (!io_cq->cdesc_addr.virt_addr) {
-		pr_err("memory allocation failed");
+		netdev_err(ena_dev->net_device, "Memory allocation failed\n");
 		return -ENOMEM;
 	}
 
@@ -428,7 +437,8 @@ static void ena_com_handle_single_admin_completion(struct ena_com_admin_queue *a
 
 	comp_ctx = get_comp_ctxt(admin_queue, cmd_id, false);
 	if (unlikely(!comp_ctx)) {
-		pr_err("comp_ctx is NULL. Changing the admin queue running state\n");
+		netdev_err(admin_queue->ena_dev->net_device,
+			   "comp_ctx is NULL. Changing the admin queue running state\n");
 		admin_queue->running_state = false;
 		return;
 	}
@@ -456,12 +466,12 @@ static void ena_com_handle_admin_completion(struct ena_com_admin_queue *admin_qu
 	cqe = &admin_queue->cq.entries[head_masked];
 
 	/* Go over all the completions */
-	while ((cqe->acq_common_descriptor.flags &
-			ENA_ADMIN_ACQ_COMMON_DESC_PHASE_MASK) == phase) {
+	while ((READ_ONCE(cqe->acq_common_descriptor.flags) &
+		ENA_ADMIN_ACQ_COMMON_DESC_PHASE_MASK) == phase) {
 		/* Do not read the rest of the completion entry before the
 		 * phase bit was validated
 		 */
-		rmb();
+		dma_rmb();
 		ena_com_handle_single_admin_completion(admin_queue, cqe);
 
 		head_masked++;
@@ -480,13 +490,12 @@ static void ena_com_handle_admin_completion(struct ena_com_admin_queue *admin_qu
 	admin_queue->stats.completed_cmd += comp_num;
 }
 
-static int ena_com_comp_status_to_errno(u8 comp_status)
+static int ena_com_comp_status_to_errno(struct ena_com_admin_queue *admin_queue,
+					u8 comp_status)
 {
 	if (unlikely(comp_status != 0))
-		pr_err("admin command failed[%u]\n", comp_status);
-
-	if (unlikely(comp_status > ENA_ADMIN_UNKNOWN_ERROR))
-		return -EINVAL;
+		netdev_err(admin_queue->ena_dev->net_device, "Admin command failed[%u]\n",
+			   comp_status);
 
 	switch (comp_status) {
 	case ENA_ADMIN_SUCCESS:
@@ -494,30 +503,48 @@ static int ena_com_comp_status_to_errno(u8 comp_status)
 	case ENA_ADMIN_RESOURCE_ALLOCATION_FAILURE:
 		return -ENOMEM;
 	case ENA_ADMIN_UNSUPPORTED_OPCODE:
-		return -EPERM;
+		return -EOPNOTSUPP;
 	case ENA_ADMIN_BAD_OPCODE:
 	case ENA_ADMIN_MALFORMED_REQUEST:
 	case ENA_ADMIN_ILLEGAL_PARAMETER:
 	case ENA_ADMIN_UNKNOWN_ERROR:
 		return -EINVAL;
+	case ENA_ADMIN_RESOURCE_BUSY:
+		return -EAGAIN;
 	}
 
-	return 0;
+	return -EINVAL;
+}
+
+static void ena_delay_exponential_backoff_us(u32 exp, u32 delay_us)
+{
+	exp = min_t(u32, exp, ENA_MAX_BACKOFF_DELAY_EXP);
+	delay_us = max_t(u32, ENA_MIN_ADMIN_POLL_US, delay_us);
+	delay_us = min_t(u32, delay_us * (1U << exp), ENA_MAX_ADMIN_POLL_US);
+	usleep_range(delay_us, 2 * delay_us);
 }
 
 static int ena_com_wait_and_process_admin_cq_polling(struct ena_comp_ctx *comp_ctx,
 						     struct ena_com_admin_queue *admin_queue)
 {
-	unsigned long flags;
-	u32 start_time;
+	unsigned long flags = 0;
+	unsigned long timeout;
 	int ret;
+	u32 exp = 0;
+
+	timeout = jiffies + usecs_to_jiffies(admin_queue->completion_timeout);
 
-	start_time = ((u32)jiffies_to_usecs(jiffies));
+	while (1) {
+		spin_lock_irqsave(&admin_queue->q_lock, flags);
+		ena_com_handle_admin_completion(admin_queue);
+		spin_unlock_irqrestore(&admin_queue->q_lock, flags);
 
-	while (comp_ctx->status == ENA_CMD_SUBMITTED) {
-		if ((((u32)jiffies_to_usecs(jiffies)) - start_time) >
-		    ADMIN_CMD_TIMEOUT_US) {
-			pr_err("Wait for completion (polling) timeout\n");
+		if (comp_ctx->status != ENA_CMD_SUBMITTED)
+			break;
+
+		if (time_is_before_jiffies(timeout)) {
+			netdev_err(admin_queue->ena_dev->net_device,
+				   "Wait for completion (polling) timeout\n");
 			/* ENA didn't have any completion */
 			spin_lock_irqsave(&admin_queue->q_lock, flags);
 			admin_queue->stats.no_completion++;
@@ -528,15 +555,12 @@ static int ena_com_wait_and_process_admin_cq_polling(struct ena_comp_ctx *comp_c
 			goto err;
 		}
 
-		spin_lock_irqsave(&admin_queue->q_lock, flags);
-		ena_com_handle_admin_completion(admin_queue);
-		spin_unlock_irqrestore(&admin_queue->q_lock, flags);
-
-		msleep(100);
+		ena_delay_exponential_backoff_us(exp++,
+						 admin_queue->ena_dev->ena_min_poll_delay_us);
 	}
 
 	if (unlikely(comp_ctx->status == ENA_CMD_ABORTED)) {
-		pr_err("Command was aborted\n");
+		netdev_err(admin_queue->ena_dev->net_device, "Command was aborted\n");
 		spin_lock_irqsave(&admin_queue->q_lock, flags);
 		admin_queue->stats.aborted_cmd++;
 		spin_unlock_irqrestore(&admin_queue->q_lock, flags);
@@ -544,23 +568,193 @@ static int ena_com_wait_and_process_admin_cq_polling(struct ena_comp_ctx *comp_c
 		goto err;
 	}
 
-	WARN(comp_ctx->status != ENA_CMD_COMPLETED, "Invalid comp status %d\n",
-	     comp_ctx->status);
+	WARN(comp_ctx->status != ENA_CMD_COMPLETED, "Invalid comp status %d\n", comp_ctx->status);
 
-	ret = ena_com_comp_status_to_errno(comp_ctx->comp_status);
+	ret = ena_com_comp_status_to_errno(admin_queue, comp_ctx->comp_status);
 err:
 	comp_ctxt_release(admin_queue, comp_ctx);
 	return ret;
 }
 
+/*
+ * Set the LLQ configurations of the firmware
+ *
+ * The driver provides only the enabled feature values to the device,
+ * which in turn, checks if they are supported.
+ */
+static int ena_com_set_llq(struct ena_com_dev *ena_dev)
+{
+	struct ena_com_admin_queue *admin_queue;
+	struct ena_admin_set_feat_cmd cmd;
+	struct ena_admin_set_feat_resp resp;
+	struct ena_com_llq_info *llq_info = &ena_dev->llq_info;
+	int ret;
+
+	memset(&cmd, 0x0, sizeof(cmd));
+	admin_queue = &ena_dev->admin_queue;
+
+	cmd.aq_common_descriptor.opcode = ENA_ADMIN_SET_FEATURE;
+	cmd.feat_common.feature_id = ENA_ADMIN_LLQ;
+
+	cmd.u.llq.header_location_ctrl_enabled = llq_info->header_location_ctrl;
+	cmd.u.llq.entry_size_ctrl_enabled = llq_info->desc_list_entry_size_ctrl;
+	cmd.u.llq.desc_num_before_header_enabled = llq_info->descs_num_before_header;
+	cmd.u.llq.descriptors_stride_ctrl_enabled = llq_info->desc_stride_ctrl;
+
+	cmd.u.llq.accel_mode.u.set.enabled_flags =
+		BIT(ENA_ADMIN_DISABLE_META_CACHING) |
+		BIT(ENA_ADMIN_LIMIT_TX_BURST);
+
+	ret = ena_com_execute_admin_command(admin_queue,
+					    (struct ena_admin_aq_entry *)&cmd,
+					    sizeof(cmd),
+					    (struct ena_admin_acq_entry *)&resp,
+					    sizeof(resp));
+
+	if (unlikely(ret))
+		netdev_err(ena_dev->net_device, "Failed to set LLQ configurations: %d\n", ret);
+
+	return ret;
+}
+
+static int ena_com_config_llq_info(struct ena_com_dev *ena_dev,
+				   struct ena_admin_feature_llq_desc *llq_features,
+				   struct ena_llq_configurations *llq_default_cfg)
+{
+	struct ena_com_llq_info *llq_info = &ena_dev->llq_info;
+	struct ena_admin_accel_mode_get llq_accel_mode_get;
+	u16 supported_feat;
+	int rc;
+
+	memset(llq_info, 0, sizeof(*llq_info));
+
+	supported_feat = llq_features->header_location_ctrl_supported;
+
+	if (likely(supported_feat & llq_default_cfg->llq_header_location)) {
+		llq_info->header_location_ctrl =
+			llq_default_cfg->llq_header_location;
+	} else {
+		netdev_err(ena_dev->net_device,
+			   "Invalid header location control, supported: 0x%x\n", supported_feat);
+		return -EINVAL;
+	}
+
+	if (likely(llq_info->header_location_ctrl == ENA_ADMIN_INLINE_HEADER)) {
+		supported_feat = llq_features->descriptors_stride_ctrl_supported;
+		if (likely(supported_feat & llq_default_cfg->llq_stride_ctrl)) {
+			llq_info->desc_stride_ctrl = llq_default_cfg->llq_stride_ctrl;
+		} else	{
+			if (supported_feat & ENA_ADMIN_MULTIPLE_DESCS_PER_ENTRY) {
+				llq_info->desc_stride_ctrl = ENA_ADMIN_MULTIPLE_DESCS_PER_ENTRY;
+			} else if (supported_feat & ENA_ADMIN_SINGLE_DESC_PER_ENTRY) {
+				llq_info->desc_stride_ctrl = ENA_ADMIN_SINGLE_DESC_PER_ENTRY;
+			} else {
+				netdev_err(ena_dev->net_device,
+					   "Invalid desc_stride_ctrl, supported: 0x%x\n",
+					   supported_feat);
+				return -EINVAL;
+			}
+
+			netdev_err(ena_dev->net_device,
+				   "Default llq stride ctrl is not supported, performing fallback, default: 0x%x, supported: 0x%x, used: 0x%x\n",
+				   llq_default_cfg->llq_stride_ctrl, supported_feat,
+				   llq_info->desc_stride_ctrl);
+		}
+	} else {
+		llq_info->desc_stride_ctrl = 0;
+	}
+
+	supported_feat = llq_features->entry_size_ctrl_supported;
+	if (likely(supported_feat & llq_default_cfg->llq_ring_entry_size)) {
+		llq_info->desc_list_entry_size_ctrl = llq_default_cfg->llq_ring_entry_size;
+		llq_info->desc_list_entry_size = llq_default_cfg->llq_ring_entry_size_value;
+	} else {
+		if (supported_feat & ENA_ADMIN_LIST_ENTRY_SIZE_128B) {
+			llq_info->desc_list_entry_size_ctrl = ENA_ADMIN_LIST_ENTRY_SIZE_128B;
+			llq_info->desc_list_entry_size = 128;
+		} else if (supported_feat & ENA_ADMIN_LIST_ENTRY_SIZE_192B) {
+			llq_info->desc_list_entry_size_ctrl = ENA_ADMIN_LIST_ENTRY_SIZE_192B;
+			llq_info->desc_list_entry_size = 192;
+		} else if (supported_feat & ENA_ADMIN_LIST_ENTRY_SIZE_256B) {
+			llq_info->desc_list_entry_size_ctrl = ENA_ADMIN_LIST_ENTRY_SIZE_256B;
+			llq_info->desc_list_entry_size = 256;
+		} else {
+			netdev_err(ena_dev->net_device,
+				   "Invalid entry_size_ctrl, supported: 0x%x\n", supported_feat);
+			return -EINVAL;
+		}
+
+		netdev_err(ena_dev->net_device,
+			   "Default llq ring entry size is not supported, performing fallback, default: 0x%x, supported: 0x%x, used: 0x%x\n",
+			   llq_default_cfg->llq_ring_entry_size, supported_feat,
+			   llq_info->desc_list_entry_size);
+	}
+	if (unlikely(llq_info->desc_list_entry_size & 0x7)) {
+		/* The desc list entry size should be whole multiply of 8
+		 * This requirement comes from __iowrite64_copy()
+		 */
+		netdev_err(ena_dev->net_device, "Illegal entry size %d\n",
+			   llq_info->desc_list_entry_size);
+		return -EINVAL;
+	}
+
+	if (llq_info->desc_stride_ctrl == ENA_ADMIN_MULTIPLE_DESCS_PER_ENTRY)
+		llq_info->descs_per_entry = llq_info->desc_list_entry_size /
+			sizeof(struct ena_eth_io_tx_desc);
+	else
+		llq_info->descs_per_entry = 1;
+
+	supported_feat = llq_features->desc_num_before_header_supported;
+	if (likely(supported_feat & llq_default_cfg->llq_num_decs_before_header)) {
+		llq_info->descs_num_before_header = llq_default_cfg->llq_num_decs_before_header;
+	} else {
+		if (supported_feat & ENA_ADMIN_LLQ_NUM_DESCS_BEFORE_HEADER_2) {
+			llq_info->descs_num_before_header = ENA_ADMIN_LLQ_NUM_DESCS_BEFORE_HEADER_2;
+		} else if (supported_feat & ENA_ADMIN_LLQ_NUM_DESCS_BEFORE_HEADER_1) {
+			llq_info->descs_num_before_header = ENA_ADMIN_LLQ_NUM_DESCS_BEFORE_HEADER_1;
+		} else if (supported_feat & ENA_ADMIN_LLQ_NUM_DESCS_BEFORE_HEADER_4) {
+			llq_info->descs_num_before_header = ENA_ADMIN_LLQ_NUM_DESCS_BEFORE_HEADER_4;
+		} else if (supported_feat & ENA_ADMIN_LLQ_NUM_DESCS_BEFORE_HEADER_8) {
+			llq_info->descs_num_before_header = ENA_ADMIN_LLQ_NUM_DESCS_BEFORE_HEADER_8;
+		} else {
+			netdev_err(ena_dev->net_device,
+				   "Invalid descs_num_before_header, supported: 0x%x\n",
+				   supported_feat);
+			return -EINVAL;
+		}
+
+		netdev_err(ena_dev->net_device,
+			   "Default llq num descs before header is not supported, performing fallback, default: 0x%x, supported: 0x%x, used: 0x%x\n",
+			   llq_default_cfg->llq_num_decs_before_header, supported_feat,
+			   llq_info->descs_num_before_header);
+	}
+	/* Check for accelerated queue supported */
+	llq_accel_mode_get = llq_features->accel_mode.u.get;
+
+	llq_info->disable_meta_caching =
+		!!(llq_accel_mode_get.supported_flags &
+		   BIT(ENA_ADMIN_DISABLE_META_CACHING));
+
+	if (llq_accel_mode_get.supported_flags & BIT(ENA_ADMIN_LIMIT_TX_BURST))
+		llq_info->max_entries_in_tx_burst =
+			llq_accel_mode_get.max_tx_burst_size /
+			llq_default_cfg->llq_ring_entry_size_value;
+
+	rc = ena_com_set_llq(ena_dev);
+	if (rc)
+		netdev_err(ena_dev->net_device, "Cannot set LLQ configuration: %d\n", rc);
+
+	return rc;
+}
+
 static int ena_com_wait_and_process_admin_cq_interrupts(struct ena_comp_ctx *comp_ctx,
 							struct ena_com_admin_queue *admin_queue)
 {
-	unsigned long flags;
+	unsigned long flags = 0;
 	int ret;
 
 	wait_for_completion_timeout(&comp_ctx->wait_event,
-				    usecs_to_jiffies(ADMIN_CMD_TIMEOUT_US));
+				    usecs_to_jiffies(admin_queue->completion_timeout));
 
 	/* In case the command wasn't completed find out the root cause.
 	 * There might be 2 kinds of errors
@@ -573,19 +767,21 @@ static int ena_com_wait_and_process_admin_cq_interrupts(struct ena_comp_ctx *com
 		admin_queue->stats.no_completion++;
 		spin_unlock_irqrestore(&admin_queue->q_lock, flags);
 
-		if (comp_ctx->status == ENA_CMD_COMPLETED)
-			pr_err("The ena device have completion but the driver didn't receive any MSI-X interrupt (cmd %d)\n",
-			       comp_ctx->cmd_opcode);
-		else
-			pr_err("The ena device doesn't send any completion for the admin cmd %d status %d\n",
-			       comp_ctx->cmd_opcode, comp_ctx->status);
-
+		if (comp_ctx->status == ENA_CMD_COMPLETED) {
+			netdev_err(admin_queue->ena_dev->net_device,
+				   "The ena device sent a completion but the driver didn't receive a MSI-X interrupt (cmd %d)\n",
+				   comp_ctx->cmd_opcode);
+		} else {
+			netdev_err(admin_queue->ena_dev->net_device,
+				   "The ena device didn't send a completion for the admin cmd %d status %d\n",
+				   comp_ctx->cmd_opcode, comp_ctx->status);
+		}
 		admin_queue->running_state = false;
 		ret = -ETIME;
 		goto err;
 	}
 
-	ret = ena_com_comp_status_to_errno(comp_ctx->comp_status);
+	ret = ena_com_comp_status_to_errno(admin_queue, comp_ctx->comp_status);
 err:
 	comp_ctxt_release(admin_queue, comp_ctx);
 	return ret;
@@ -600,12 +796,15 @@ static u32 ena_com_reg_bar_read32(struct ena_com_dev *ena_dev, u16 offset)
 	struct ena_com_mmio_read *mmio_read = &ena_dev->mmio_read;
 	volatile struct ena_admin_ena_mmio_req_read_less_resp *read_resp =
 		mmio_read->read_resp;
-	u32 mmio_read_reg, ret;
-	unsigned long flags;
-	int i;
+	u32 mmio_read_reg, ret, i;
+	unsigned long flags = 0;
+	u32 timeout = mmio_read->reg_read_to;
 
 	might_sleep();
 
+	if (timeout == 0)
+		timeout = ENA_REG_READ_TIMEOUT;
+
 	/* If readless is disabled, perform regular read */
 	if (!mmio_read->readless_supported)
 		return readl(ena_dev->reg_bar + offset);
@@ -619,30 +818,25 @@ static u32 ena_com_reg_bar_read32(struct ena_com_dev *ena_dev, u16 offset)
 	mmio_read_reg |= mmio_read->seq_num &
 			ENA_REGS_MMIO_REG_READ_REQ_ID_MASK;
 
-	/* make sure read_resp->req_id get updated before the hw can write
-	 * there
-	 */
-	wmb();
-
 	writel(mmio_read_reg, ena_dev->reg_bar + ENA_REGS_MMIO_REG_READ_OFF);
 
-	for (i = 0; i < ENA_REG_READ_TIMEOUT; i++) {
-		if (read_resp->req_id == mmio_read->seq_num)
+	for (i = 0; i < timeout; i++) {
+		if (READ_ONCE(read_resp->req_id) == mmio_read->seq_num)
 			break;
 
 		udelay(1);
 	}
 
-	if (unlikely(i == ENA_REG_READ_TIMEOUT)) {
-		pr_err("reading reg failed for timeout. expected: req id[%hu] offset[%hu] actual: req id[%hu] offset[%hu]\n",
-		       mmio_read->seq_num, offset, read_resp->req_id,
-		       read_resp->reg_off);
+	if (unlikely(i == timeout)) {
+		netdev_err(ena_dev->net_device,
+			   "Reading reg failed for timeout. expected: req id[%u] offset[%u] actual: req id[%u] offset[%u]\n",
+			   mmio_read->seq_num, offset, read_resp->req_id, read_resp->reg_off);
 		ret = ENA_MMIO_READ_TIMEOUT;
 		goto err;
 	}
 
 	if (read_resp->reg_off != offset) {
-		pr_err("Read failure: wrong offset provided");
+		netdev_err(ena_dev->net_device, "Read failure: wrong offset provided\n");
 		ret = ENA_MMIO_READ_TIMEOUT;
 	} else {
 		ret = read_resp->reg_val;
@@ -680,7 +874,7 @@ static int ena_com_destroy_io_sq(struct ena_com_dev *ena_dev,
 	u8 direction;
 	int ret;
 
-	memset(&destroy_cmd, 0x0, sizeof(struct ena_admin_aq_destroy_sq_cmd));
+	memset(&destroy_cmd, 0x0, sizeof(destroy_cmd));
 
 	if (io_sq->direction == ENA_COM_IO_QUEUE_DIRECTION_TX)
 		direction = ENA_ADMIN_SQ_DIRECTION_TX;
@@ -701,7 +895,7 @@ static int ena_com_destroy_io_sq(struct ena_com_dev *ena_dev,
 					    sizeof(destroy_resp));
 
 	if (unlikely(ret && (ret != -ENODEV)))
-		pr_err("failed to destroy io sq error: %d\n", ret);
+		netdev_err(ena_dev->net_device, "Failed to destroy io sq error: %d\n", ret);
 
 	return ret;
 }
@@ -715,8 +909,7 @@ static void ena_com_io_queue_free(struct ena_com_dev *ena_dev,
 	if (io_cq->cdesc_addr.virt_addr) {
 		size = io_cq->cdesc_entry_size_in_bytes * io_cq->q_depth;
 
-		dma_free_coherent(ena_dev->dmadev, size,
-				  io_cq->cdesc_addr.virt_addr,
+		dma_free_coherent(ena_dev->dmadev, size, io_cq->cdesc_addr.virt_addr,
 				  io_cq->cdesc_addr.phys_addr);
 
 		io_cq->cdesc_addr.virt_addr = NULL;
@@ -725,27 +918,32 @@ static void ena_com_io_queue_free(struct ena_com_dev *ena_dev,
 	if (io_sq->desc_addr.virt_addr) {
 		size = io_sq->desc_entry_size * io_sq->q_depth;
 
-		if (io_sq->mem_queue_type == ENA_ADMIN_PLACEMENT_POLICY_HOST)
-			dma_free_coherent(ena_dev->dmadev, size,
-					  io_sq->desc_addr.virt_addr,
-					  io_sq->desc_addr.phys_addr);
-		else
-			devm_kfree(ena_dev->dmadev, io_sq->desc_addr.virt_addr);
+		dma_free_coherent(ena_dev->dmadev, size, io_sq->desc_addr.virt_addr,
+				  io_sq->desc_addr.phys_addr);
 
 		io_sq->desc_addr.virt_addr = NULL;
 	}
+
+	if (io_sq->bounce_buf_ctrl.base_buffer) {
+		devm_kfree(ena_dev->dmadev, io_sq->bounce_buf_ctrl.base_buffer);
+		io_sq->bounce_buf_ctrl.base_buffer = NULL;
+	}
 }
 
 static int wait_for_reset_state(struct ena_com_dev *ena_dev, u32 timeout,
 				u16 exp_state)
 {
-	u32 val, i;
+	u32 val, exp = 0;
+	unsigned long timeout_stamp;
 
-	for (i = 0; i < timeout; i++) {
+	/* Convert timeout from resolution of 100ms to us resolution. */
+	timeout_stamp = jiffies + usecs_to_jiffies(100 * 1000 * timeout);
+
+	while (1) {
 		val = ena_com_reg_bar_read32(ena_dev, ENA_REGS_DEV_STS_OFF);
 
 		if (unlikely(val == ENA_MMIO_READ_TIMEOUT)) {
-			pr_err("Reg read timeout occurred\n");
+			netdev_err(ena_dev->net_device, "Reg read timeout occurred\n");
 			return -ETIME;
 		}
 
@@ -753,11 +951,11 @@ static int wait_for_reset_state(struct ena_com_dev *ena_dev, u32 timeout,
 			exp_state)
 			return 0;
 
-		/* The resolution of the timeout is 100ms */
-		msleep(100);
-	}
+		if (time_is_before_jiffies(timeout_stamp))
+			return -ETIME;
 
-	return -ETIME;
+		ena_delay_exponential_backoff_us(exp++, ena_dev->ena_min_poll_delay_us);
+	}
 }
 
 static bool ena_com_check_supported_feature_id(struct ena_com_dev *ena_dev,
@@ -777,15 +975,16 @@ static int ena_com_get_feature_ex(struct ena_com_dev *ena_dev,
 				  struct ena_admin_get_feat_resp *get_resp,
 				  enum ena_admin_aq_feature_id feature_id,
 				  dma_addr_t control_buf_dma_addr,
-				  u32 control_buff_size)
+				  u32 control_buff_size,
+				  u8 feature_ver)
 {
 	struct ena_com_admin_queue *admin_queue;
 	struct ena_admin_get_feat_cmd get_cmd;
 	int ret;
 
 	if (!ena_com_check_supported_feature_id(ena_dev, feature_id)) {
-		pr_debug("Feature %d isn't supported\n", feature_id);
-		return -EPERM;
+		netdev_dbg(ena_dev->net_device, "Feature %d isn't supported\n", feature_id);
+		return -EOPNOTSUPP;
 	}
 
 	memset(&get_cmd, 0x0, sizeof(get_cmd));
@@ -803,12 +1002,12 @@ static int ena_com_get_feature_ex(struct ena_com_dev *ena_dev,
 				   &get_cmd.control_buffer.address,
 				   control_buf_dma_addr);
 	if (unlikely(ret)) {
-		pr_err("memory address set failed\n");
+		netdev_err(ena_dev->net_device, "Memory address set failed\n");
 		return ret;
 	}
 
 	get_cmd.control_buffer.length = control_buff_size;
-
+	get_cmd.feat_common.feature_version = feature_ver;
 	get_cmd.feat_common.feature_id = feature_id;
 
 	ret = ena_com_execute_admin_command(admin_queue,
@@ -820,30 +1019,51 @@ static int ena_com_get_feature_ex(struct ena_com_dev *ena_dev,
 					    sizeof(*get_resp));
 
 	if (unlikely(ret))
-		pr_err("Failed to submit get_feature command %d error: %d\n",
-		       feature_id, ret);
+		netdev_err(ena_dev->net_device,
+			   "Failed to submit get_feature command %d error: %d\n", feature_id, ret);
 
 	return ret;
 }
 
 static int ena_com_get_feature(struct ena_com_dev *ena_dev,
 			       struct ena_admin_get_feat_resp *get_resp,
-			       enum ena_admin_aq_feature_id feature_id)
+			       enum ena_admin_aq_feature_id feature_id,
+			       u8 feature_ver)
 {
 	return ena_com_get_feature_ex(ena_dev,
 				      get_resp,
 				      feature_id,
 				      0,
-				      0);
+				      0,
+				      feature_ver);
+}
+
+int ena_com_get_current_hash_function(struct ena_com_dev *ena_dev)
+{
+	return ena_dev->rss.hash_func;
+}
+
+static void ena_com_hash_key_fill_default_key(struct ena_com_dev *ena_dev)
+{
+	struct ena_admin_feature_rss_flow_hash_control *hash_key =
+		(ena_dev->rss).hash_key;
+
+	netdev_rss_key_fill(&hash_key->key, sizeof(hash_key->key));
+	/* The key buffer is stored in the device in an array of
+	 * uint32 elements.
+	 */
+	hash_key->key_parts = ENA_ADMIN_RSS_KEY_PARTS;
 }
 
 static int ena_com_hash_key_allocate(struct ena_com_dev *ena_dev)
 {
 	struct ena_rss *rss = &ena_dev->rss;
 
-	rss->hash_key =
-		dma_zalloc_coherent(ena_dev->dmadev, sizeof(*rss->hash_key),
-				    &rss->hash_key_dma_addr, GFP_KERNEL);
+	if (!ena_com_check_supported_feature_id(ena_dev, ENA_ADMIN_RSS_HASH_FUNCTION))
+		return -EOPNOTSUPP;
+
+	rss->hash_key = dma_alloc_coherent(ena_dev->dmadev, sizeof(*rss->hash_key),
+					   &rss->hash_key_dma_addr, GFP_KERNEL);
 
 	if (unlikely(!rss->hash_key))
 		return -ENOMEM;
@@ -856,8 +1076,8 @@ static void ena_com_hash_key_destroy(struct ena_com_dev *ena_dev)
 	struct ena_rss *rss = &ena_dev->rss;
 
 	if (rss->hash_key)
-		dma_free_coherent(ena_dev->dmadev, sizeof(*rss->hash_key),
-				  rss->hash_key, rss->hash_key_dma_addr);
+		dma_free_coherent(ena_dev->dmadev, sizeof(*rss->hash_key), rss->hash_key,
+				  rss->hash_key_dma_addr);
 	rss->hash_key = NULL;
 }
 
@@ -865,9 +1085,8 @@ static int ena_com_hash_ctrl_init(struct ena_com_dev *ena_dev)
 {
 	struct ena_rss *rss = &ena_dev->rss;
 
-	rss->hash_ctrl =
-		dma_zalloc_coherent(ena_dev->dmadev, sizeof(*rss->hash_ctrl),
-				    &rss->hash_ctrl_dma_addr, GFP_KERNEL);
+	rss->hash_ctrl = dma_alloc_coherent(ena_dev->dmadev, sizeof(*rss->hash_ctrl),
+					    &rss->hash_ctrl_dma_addr, GFP_KERNEL);
 
 	if (unlikely(!rss->hash_ctrl))
 		return -ENOMEM;
@@ -880,8 +1099,8 @@ static void ena_com_hash_ctrl_destroy(struct ena_com_dev *ena_dev)
 	struct ena_rss *rss = &ena_dev->rss;
 
 	if (rss->hash_ctrl)
-		dma_free_coherent(ena_dev->dmadev, sizeof(*rss->hash_ctrl),
-				  rss->hash_ctrl, rss->hash_ctrl_dma_addr);
+		dma_free_coherent(ena_dev->dmadev, sizeof(*rss->hash_ctrl), rss->hash_ctrl,
+				  rss->hash_ctrl_dma_addr);
 	rss->hash_ctrl = NULL;
 }
 
@@ -894,30 +1113,29 @@ static int ena_com_indirect_table_allocate(struct ena_com_dev *ena_dev,
 	int ret;
 
 	ret = ena_com_get_feature(ena_dev, &get_resp,
-				  ENA_ADMIN_RSS_REDIRECTION_TABLE_CONFIG);
+				  ENA_ADMIN_RSS_INDIRECTION_TABLE_CONFIG, 0);
 	if (unlikely(ret))
 		return ret;
 
 	if ((get_resp.u.ind_table.min_size > log_size) ||
 	    (get_resp.u.ind_table.max_size < log_size)) {
-		pr_err("indirect table size doesn't fit. requested size: %d while min is:%d and max %d\n",
-		       1 << log_size, 1 << get_resp.u.ind_table.min_size,
-		       1 << get_resp.u.ind_table.max_size);
+		netdev_err(ena_dev->net_device,
+			   "Indirect table size doesn't fit. requested size: %d while min is:%d and max %d\n",
+			   1 << log_size, 1 << get_resp.u.ind_table.min_size,
+			   1 << get_resp.u.ind_table.max_size);
 		return -EINVAL;
 	}
 
 	tbl_size = (1ULL << log_size) *
 		sizeof(struct ena_admin_rss_ind_table_entry);
 
-	rss->rss_ind_tbl =
-		dma_zalloc_coherent(ena_dev->dmadev, tbl_size,
-				    &rss->rss_ind_tbl_dma_addr, GFP_KERNEL);
+	rss->rss_ind_tbl = dma_alloc_coherent(ena_dev->dmadev, tbl_size, &rss->rss_ind_tbl_dma_addr,
+					      GFP_KERNEL);
 	if (unlikely(!rss->rss_ind_tbl))
 		goto mem_err1;
 
 	tbl_size = (1ULL << log_size) * sizeof(u16);
-	rss->host_rss_ind_tbl =
-		devm_kzalloc(ena_dev->dmadev, tbl_size, GFP_KERNEL);
+	rss->host_rss_ind_tbl = devm_kzalloc(ena_dev->dmadev, tbl_size, GFP_KERNEL);
 	if (unlikely(!rss->host_rss_ind_tbl))
 		goto mem_err2;
 
@@ -929,8 +1147,7 @@ mem_err2:
 	tbl_size = (1ULL << log_size) *
 		sizeof(struct ena_admin_rss_ind_table_entry);
 
-	dma_free_coherent(ena_dev->dmadev, tbl_size, rss->rss_ind_tbl,
-			  rss->rss_ind_tbl_dma_addr);
+	dma_free_coherent(ena_dev->dmadev, tbl_size, rss->rss_ind_tbl, rss->rss_ind_tbl_dma_addr);
 	rss->rss_ind_tbl = NULL;
 mem_err1:
 	rss->tbl_log_size = 0;
@@ -962,7 +1179,7 @@ static int ena_com_create_io_sq(struct ena_com_dev *ena_dev,
 	u8 direction;
 	int ret;
 
-	memset(&create_cmd, 0x0, sizeof(struct ena_admin_aq_create_sq_cmd));
+	memset(&create_cmd, 0x0, sizeof(create_cmd));
 
 	create_cmd.aq_common_descriptor.opcode = ENA_ADMIN_CREATE_SQ;
 
@@ -993,7 +1210,7 @@ static int ena_com_create_io_sq(struct ena_com_dev *ena_dev,
 					   &create_cmd.sq_ba,
 					   io_sq->desc_addr.phys_addr);
 		if (unlikely(ret)) {
-			pr_err("memory address set failed\n");
+			netdev_err(ena_dev->net_device, "Memory address set failed\n");
 			return ret;
 		}
 	}
@@ -1004,7 +1221,7 @@ static int ena_com_create_io_sq(struct ena_com_dev *ena_dev,
 					    (struct ena_admin_acq_entry *)&cmd_completion,
 					    sizeof(cmd_completion));
 	if (unlikely(ret)) {
-		pr_err("Failed to create IO SQ. error: %d\n", ret);
+		netdev_err(ena_dev->net_device, "Failed to create IO SQ. error: %d\n", ret);
 		return ret;
 	}
 
@@ -1014,15 +1231,12 @@ static int ena_com_create_io_sq(struct ena_com_dev *ena_dev,
 		(uintptr_t)cmd_completion.sq_doorbell_offset);
 
 	if (io_sq->mem_queue_type == ENA_ADMIN_PLACEMENT_POLICY_DEV) {
-		io_sq->header_addr = (u8 __iomem *)((uintptr_t)ena_dev->mem_bar
-				+ cmd_completion.llq_headers_offset);
-
 		io_sq->desc_addr.pbuf_dev_addr =
 			(u8 __iomem *)((uintptr_t)ena_dev->mem_bar +
 			cmd_completion.llq_descriptors_offset);
 	}
 
-	pr_debug("created sq[%u], depth[%u]\n", io_sq->idx, io_sq->q_depth);
+	netdev_dbg(ena_dev->net_device, "Created sq[%u], depth[%u]\n", io_sq->idx, io_sq->q_depth);
 
 	return ret;
 }
@@ -1050,64 +1264,30 @@ static int ena_com_ind_tbl_convert_to_device(struct ena_com_dev *ena_dev)
 	return 0;
 }
 
-static int ena_com_ind_tbl_convert_from_device(struct ena_com_dev *ena_dev)
-{
-	u16 dev_idx_to_host_tbl[ENA_TOTAL_NUM_QUEUES] = { (u16)-1 };
-	struct ena_rss *rss = &ena_dev->rss;
-	u8 idx;
-	u16 i;
-
-	for (i = 0; i < ENA_TOTAL_NUM_QUEUES; i++)
-		dev_idx_to_host_tbl[ena_dev->io_sq_queues[i].idx] = i;
-
-	for (i = 0; i < 1 << rss->tbl_log_size; i++) {
-		if (rss->rss_ind_tbl[i].cq_idx > ENA_TOTAL_NUM_QUEUES)
-			return -EINVAL;
-		idx = (u8)rss->rss_ind_tbl[i].cq_idx;
-
-		if (dev_idx_to_host_tbl[idx] > ENA_TOTAL_NUM_QUEUES)
-			return -EINVAL;
-
-		rss->host_rss_ind_tbl[i] = dev_idx_to_host_tbl[idx];
-	}
-
-	return 0;
-}
-
-static int ena_com_init_interrupt_moderation_table(struct ena_com_dev *ena_dev)
-{
-	size_t size;
-
-	size = sizeof(struct ena_intr_moder_entry) * ENA_INTR_MAX_NUM_OF_LEVELS;
-
-	ena_dev->intr_moder_tbl =
-		devm_kzalloc(ena_dev->dmadev, size, GFP_KERNEL);
-	if (!ena_dev->intr_moder_tbl)
-		return -ENOMEM;
-
-	ena_com_config_default_interrupt_moderation_table(ena_dev);
-
-	return 0;
-}
-
 static void ena_com_update_intr_delay_resolution(struct ena_com_dev *ena_dev,
 						 u16 intr_delay_resolution)
 {
-	struct ena_intr_moder_entry *intr_moder_tbl = ena_dev->intr_moder_tbl;
-	unsigned int i;
+	u16 prev_intr_delay_resolution = ena_dev->intr_delay_resolution;
 
-	if (!intr_delay_resolution) {
-		pr_err("Illegal intr_delay_resolution provided. Going to use default 1 usec resolution\n");
-		intr_delay_resolution = 1;
+	if (unlikely(!intr_delay_resolution)) {
+		netdev_err(ena_dev->net_device,
+			   "Illegal intr_delay_resolution provided. Going to use default 1 usec resolution\n");
+		intr_delay_resolution = ENA_DEFAULT_INTR_DELAY_RESOLUTION;
 	}
-	ena_dev->intr_delay_resolution = intr_delay_resolution;
 
 	/* update Rx */
-	for (i = 0; i < ENA_INTR_MAX_NUM_OF_LEVELS; i++)
-		intr_moder_tbl[i].intr_moder_interval /= intr_delay_resolution;
+	ena_dev->intr_moder_rx_interval =
+		ena_dev->intr_moder_rx_interval *
+		prev_intr_delay_resolution /
+		intr_delay_resolution;
 
 	/* update Tx */
-	ena_dev->intr_moder_tx_interval /= intr_delay_resolution;
+	ena_dev->intr_moder_tx_interval =
+		ena_dev->intr_moder_tx_interval *
+		prev_intr_delay_resolution /
+		intr_delay_resolution;
+
+	ena_dev->intr_delay_resolution = intr_delay_resolution;
 }
 
 /*****************************************************************************/
@@ -1125,23 +1305,26 @@ int ena_com_execute_admin_command(struct ena_com_admin_queue *admin_queue,
 
 	comp_ctx = ena_com_submit_admin_cmd(admin_queue, cmd, cmd_size,
 					    comp, comp_size);
-	if (unlikely(IS_ERR(comp_ctx))) {
-		if (comp_ctx == ERR_PTR(-ENODEV))
-			pr_debug("Failed to submit command [%ld]\n",
-				 PTR_ERR(comp_ctx));
+	if (IS_ERR(comp_ctx)) {
+		ret = PTR_ERR(comp_ctx);
+		if (ret == -ENODEV)
+			netdev_dbg(admin_queue->ena_dev->net_device,
+				   "Failed to submit command [%d]\n", ret);
 		else
-			pr_err("Failed to submit command [%ld]\n",
-			       PTR_ERR(comp_ctx));
+			netdev_err(admin_queue->ena_dev->net_device,
+				   "Failed to submit command [%d]\n", ret);
 
-		return PTR_ERR(comp_ctx);
+		return ret;
 	}
 
 	ret = ena_com_wait_and_process_admin_cq(comp_ctx, admin_queue);
 	if (unlikely(ret)) {
 		if (admin_queue->running_state)
-			pr_err("Failed to process command. ret = %d\n", ret);
+			netdev_err(admin_queue->ena_dev->net_device,
+				   "Failed to process command. ret = %d\n", ret);
 		else
-			pr_debug("Failed to process command. ret = %d\n", ret);
+			netdev_dbg(admin_queue->ena_dev->net_device,
+				   "Failed to process command. ret = %d\n", ret);
 	}
 	return ret;
 }
@@ -1154,7 +1337,7 @@ int ena_com_create_io_cq(struct ena_com_dev *ena_dev,
 	struct ena_admin_acq_create_cq_resp_desc cmd_completion;
 	int ret;
 
-	memset(&create_cmd, 0x0, sizeof(struct ena_admin_aq_create_cq_cmd));
+	memset(&create_cmd, 0x0, sizeof(create_cmd));
 
 	create_cmd.aq_common_descriptor.opcode = ENA_ADMIN_CREATE_CQ;
 
@@ -1170,7 +1353,7 @@ int ena_com_create_io_cq(struct ena_com_dev *ena_dev,
 				   &create_cmd.cq_ba,
 				   io_cq->cdesc_addr.phys_addr);
 	if (unlikely(ret)) {
-		pr_err("memory address set failed\n");
+		netdev_err(ena_dev->net_device, "Memory address set failed\n");
 		return ret;
 	}
 
@@ -1180,7 +1363,7 @@ int ena_com_create_io_cq(struct ena_com_dev *ena_dev,
 					    (struct ena_admin_acq_entry *)&cmd_completion,
 					    sizeof(cmd_completion));
 	if (unlikely(ret)) {
-		pr_err("Failed to create IO CQ. error: %d\n", ret);
+		netdev_err(ena_dev->net_device, "Failed to create IO CQ. error: %d\n", ret);
 		return ret;
 	}
 
@@ -1189,17 +1372,12 @@ int ena_com_create_io_cq(struct ena_com_dev *ena_dev,
 	io_cq->unmask_reg = (u32 __iomem *)((uintptr_t)ena_dev->reg_bar +
 		cmd_completion.cq_interrupt_unmask_register_offset);
 
-	if (cmd_completion.cq_head_db_register_offset)
-		io_cq->cq_head_db_reg =
-			(u32 __iomem *)((uintptr_t)ena_dev->reg_bar +
-			cmd_completion.cq_head_db_register_offset);
-
 	if (cmd_completion.numa_node_register_offset)
 		io_cq->numa_node_cfg_reg =
 			(u32 __iomem *)((uintptr_t)ena_dev->reg_bar +
 			cmd_completion.numa_node_register_offset);
 
-	pr_debug("created cq[%u], depth[%u]\n", io_cq->idx, io_cq->q_depth);
+	netdev_dbg(ena_dev->net_device, "Created cq[%u], depth[%u]\n", io_cq->idx, io_cq->q_depth);
 
 	return ret;
 }
@@ -1209,8 +1387,8 @@ int ena_com_get_io_handlers(struct ena_com_dev *ena_dev, u16 qid,
 			    struct ena_com_io_cq **io_cq)
 {
 	if (qid >= ENA_TOTAL_NUM_QUEUES) {
-		pr_err("Invalid queue number %d but the max is %d\n", qid,
-		       ENA_TOTAL_NUM_QUEUES);
+		netdev_err(ena_dev->net_device, "Invalid queue number %d but the max is %d\n", qid,
+			   ENA_TOTAL_NUM_QUEUES);
 		return -EINVAL;
 	}
 
@@ -1243,12 +1421,13 @@ void ena_com_abort_admin_commands(struct ena_com_dev *ena_dev)
 void ena_com_wait_for_abort_completion(struct ena_com_dev *ena_dev)
 {
 	struct ena_com_admin_queue *admin_queue = &ena_dev->admin_queue;
-	unsigned long flags;
+	unsigned long flags = 0;
+	u32 exp = 0;
 
 	spin_lock_irqsave(&admin_queue->q_lock, flags);
 	while (atomic_read(&admin_queue->outstanding_cmds) != 0) {
 		spin_unlock_irqrestore(&admin_queue->q_lock, flags);
-		msleep(20);
+		ena_delay_exponential_backoff_us(exp++, ena_dev->ena_min_poll_delay_us);
 		spin_lock_irqsave(&admin_queue->q_lock, flags);
 	}
 	spin_unlock_irqrestore(&admin_queue->q_lock, flags);
@@ -1262,7 +1441,7 @@ int ena_com_destroy_io_cq(struct ena_com_dev *ena_dev,
 	struct ena_admin_acq_destroy_cq_resp_desc destroy_resp;
 	int ret;
 
-	memset(&destroy_cmd, 0x0, sizeof(struct ena_admin_aq_destroy_sq_cmd));
+	memset(&destroy_cmd, 0x0, sizeof(destroy_cmd));
 
 	destroy_cmd.cq_idx = io_cq->idx;
 	destroy_cmd.aq_common_descriptor.opcode = ENA_ADMIN_DESTROY_CQ;
@@ -1274,7 +1453,7 @@ int ena_com_destroy_io_cq(struct ena_com_dev *ena_dev,
 					    sizeof(destroy_resp));
 
 	if (unlikely(ret && (ret != -ENODEV)))
-		pr_err("Failed to destroy IO CQ. error: %d\n", ret);
+		netdev_err(ena_dev->net_device, "Failed to destroy IO CQ. error: %d\n", ret);
 
 	return ret;
 }
@@ -1287,7 +1466,7 @@ bool ena_com_get_admin_running_state(struct ena_com_dev *ena_dev)
 void ena_com_set_admin_running_state(struct ena_com_dev *ena_dev, bool state)
 {
 	struct ena_com_admin_queue *admin_queue = &ena_dev->admin_queue;
-	unsigned long flags;
+	unsigned long flags = 0;
 
 	spin_lock_irqsave(&admin_queue->q_lock, flags);
 	ena_dev->admin_queue.running_state = state;
@@ -1314,16 +1493,17 @@ int ena_com_set_aenq_config(struct ena_com_dev *ena_dev, u32 groups_flag)
 	struct ena_admin_get_feat_resp get_resp;
 	int ret;
 
-	ret = ena_com_get_feature(ena_dev, &get_resp, ENA_ADMIN_AENQ_CONFIG);
+	ret = ena_com_get_feature(ena_dev, &get_resp, ENA_ADMIN_AENQ_CONFIG, 0);
 	if (ret) {
-		pr_info("Can't get aenq configuration\n");
+		dev_info(ena_dev->dmadev, "Can't get aenq configuration\n");
 		return ret;
 	}
 
 	if ((get_resp.u.aenq.supported_groups & groups_flag) != groups_flag) {
-		pr_warn("Trying to set unsupported aenq events. supported flag: %x asked flag: %x\n",
-			get_resp.u.aenq.supported_groups, groups_flag);
-		return -EPERM;
+		netdev_warn(ena_dev->net_device,
+			    "Trying to set unsupported aenq events. supported flag: 0x%x asked flag: 0x%x\n",
+			    get_resp.u.aenq.supported_groups, groups_flag);
+		return -EOPNOTSUPP;
 	}
 
 	memset(&cmd, 0x0, sizeof(cmd));
@@ -1341,7 +1521,7 @@ int ena_com_set_aenq_config(struct ena_com_dev *ena_dev, u32 groups_flag)
 					    sizeof(resp));
 
 	if (unlikely(ret))
-		pr_err("Failed to config AENQ ret: %d\n", ret);
+		netdev_err(ena_dev->net_device, "Failed to config AENQ ret: %d\n", ret);
 
 	return ret;
 }
@@ -1349,20 +1529,20 @@ int ena_com_set_aenq_config(struct ena_com_dev *ena_dev, u32 groups_flag)
 int ena_com_get_dma_width(struct ena_com_dev *ena_dev)
 {
 	u32 caps = ena_com_reg_bar_read32(ena_dev, ENA_REGS_CAPS_OFF);
-	int width;
+	u32 width;
 
 	if (unlikely(caps == ENA_MMIO_READ_TIMEOUT)) {
-		pr_err("Reg read timeout occurred\n");
+		netdev_err(ena_dev->net_device, "Reg read timeout occurred\n");
 		return -ETIME;
 	}
 
 	width = (caps & ENA_REGS_CAPS_DMA_ADDR_WIDTH_MASK) >>
 		ENA_REGS_CAPS_DMA_ADDR_WIDTH_SHIFT;
 
-	pr_debug("ENA dma width: %d\n", width);
+	netdev_dbg(ena_dev->net_device, "ENA dma width: %d\n", width);
 
 	if ((width < 32) || width > ENA_MAX_PHYS_ADDR_SIZE_BITS) {
-		pr_err("DMA width illegal value: %d\n", width);
+		netdev_err(ena_dev->net_device, "DMA width illegal value: %d\n", width);
 		return -EINVAL;
 	}
 
@@ -1384,30 +1564,23 @@ int ena_com_validate_version(struct ena_com_dev *ena_dev)
 	ctrl_ver = ena_com_reg_bar_read32(ena_dev,
 					  ENA_REGS_CONTROLLER_VERSION_OFF);
 
-	if (unlikely((ver == ENA_MMIO_READ_TIMEOUT) ||
-		     (ctrl_ver == ENA_MMIO_READ_TIMEOUT))) {
-		pr_err("Reg read timeout occurred\n");
+	if (unlikely((ver == ENA_MMIO_READ_TIMEOUT) || (ctrl_ver == ENA_MMIO_READ_TIMEOUT))) {
+		netdev_err(ena_dev->net_device, "Reg read timeout occurred\n");
 		return -ETIME;
 	}
 
-	pr_info("ena device version: %d.%d\n",
-		(ver & ENA_REGS_VERSION_MAJOR_VERSION_MASK) >>
-			ENA_REGS_VERSION_MAJOR_VERSION_SHIFT,
-		ver & ENA_REGS_VERSION_MINOR_VERSION_MASK);
-
-	if (ver < MIN_ENA_VER) {
-		pr_err("ENA version is lower than the minimal version the driver supports\n");
-		return -1;
-	}
+	dev_info(ena_dev->dmadev, "ENA device version: %d.%d\n",
+		 (ver & ENA_REGS_VERSION_MAJOR_VERSION_MASK) >> ENA_REGS_VERSION_MAJOR_VERSION_SHIFT,
+		 ver & ENA_REGS_VERSION_MINOR_VERSION_MASK);
 
-	pr_info("ena controller version: %d.%d.%d implementation version %d\n",
-		(ctrl_ver & ENA_REGS_CONTROLLER_VERSION_MAJOR_VERSION_MASK) >>
-			ENA_REGS_CONTROLLER_VERSION_MAJOR_VERSION_SHIFT,
-		(ctrl_ver & ENA_REGS_CONTROLLER_VERSION_MINOR_VERSION_MASK) >>
-			ENA_REGS_CONTROLLER_VERSION_MINOR_VERSION_SHIFT,
-		(ctrl_ver & ENA_REGS_CONTROLLER_VERSION_SUBMINOR_VERSION_MASK),
-		(ctrl_ver & ENA_REGS_CONTROLLER_VERSION_IMPL_ID_MASK) >>
-			ENA_REGS_CONTROLLER_VERSION_IMPL_ID_SHIFT);
+	dev_info(ena_dev->dmadev, "ENA controller version: %d.%d.%d implementation version %d\n",
+		 (ctrl_ver & ENA_REGS_CONTROLLER_VERSION_MAJOR_VERSION_MASK) >>
+			 ENA_REGS_CONTROLLER_VERSION_MAJOR_VERSION_SHIFT,
+		 (ctrl_ver & ENA_REGS_CONTROLLER_VERSION_MINOR_VERSION_MASK) >>
+			 ENA_REGS_CONTROLLER_VERSION_MINOR_VERSION_SHIFT,
+		 (ctrl_ver & ENA_REGS_CONTROLLER_VERSION_SUBMINOR_VERSION_MASK),
+		 (ctrl_ver & ENA_REGS_CONTROLLER_VERSION_IMPL_ID_MASK) >>
+			 ENA_REGS_CONTROLLER_VERSION_IMPL_ID_SHIFT);
 
 	ctrl_ver_masked =
 		(ctrl_ver & ENA_REGS_CONTROLLER_VERSION_MAJOR_VERSION_MASK) |
@@ -1416,13 +1589,27 @@ int ena_com_validate_version(struct ena_com_dev *ena_dev)
 
 	/* Validate the ctrl version without the implementation ID */
 	if (ctrl_ver_masked < MIN_ENA_CTRL_VER) {
-		pr_err("ENA ctrl version is lower than the minimal ctrl version the driver supports\n");
+		netdev_err(ena_dev->net_device,
+			   "ENA ctrl version is lower than the minimal ctrl version the driver supports\n");
 		return -1;
 	}
 
 	return 0;
 }
 
+static void
+ena_com_free_ena_admin_queue_comp_ctx(struct ena_com_dev *ena_dev,
+				      struct ena_com_admin_queue *admin_queue)
+
+{
+	if (!admin_queue->comp_ctx)
+		return;
+
+	devm_kfree(ena_dev->dmadev, admin_queue->comp_ctx);
+
+	admin_queue->comp_ctx = NULL;
+}
+
 void ena_com_admin_destroy(struct ena_com_dev *ena_dev)
 {
 	struct ena_com_admin_queue *admin_queue = &ena_dev->admin_queue;
@@ -1431,44 +1618,305 @@ void ena_com_admin_destroy(struct ena_com_dev *ena_dev)
 	struct ena_com_aenq *aenq = &ena_dev->aenq;
 	u16 size;
 
-	if (admin_queue->comp_ctx)
-		devm_kfree(ena_dev->dmadev, admin_queue->comp_ctx);
-	admin_queue->comp_ctx = NULL;
+	ena_com_free_ena_admin_queue_comp_ctx(ena_dev, admin_queue);
+
 	size = ADMIN_SQ_SIZE(admin_queue->q_depth);
 	if (sq->entries)
-		dma_free_coherent(ena_dev->dmadev, size, sq->entries,
-				  sq->dma_addr);
+		dma_free_coherent(ena_dev->dmadev, size, sq->entries, sq->dma_addr);
 	sq->entries = NULL;
 
 	size = ADMIN_CQ_SIZE(admin_queue->q_depth);
 	if (cq->entries)
-		dma_free_coherent(ena_dev->dmadev, size, cq->entries,
-				  cq->dma_addr);
+		dma_free_coherent(ena_dev->dmadev, size, cq->entries, cq->dma_addr);
 	cq->entries = NULL;
 
 	size = ADMIN_AENQ_SIZE(aenq->q_depth);
 	if (ena_dev->aenq.entries)
-		dma_free_coherent(ena_dev->dmadev, size, aenq->entries,
-				  aenq->dma_addr);
+		dma_free_coherent(ena_dev->dmadev, size, aenq->entries, aenq->dma_addr);
 	aenq->entries = NULL;
 }
 
 void ena_com_set_admin_polling_mode(struct ena_com_dev *ena_dev, bool polling)
 {
+	u32 mask_value = 0;
+
+	if (polling)
+		mask_value = ENA_REGS_ADMIN_INTR_MASK;
+
+	writel(mask_value, ena_dev->reg_bar + ENA_REGS_INTR_MASK_OFF);
 	ena_dev->admin_queue.polling = polling;
 }
 
+bool ena_com_phc_supported(struct ena_com_dev *ena_dev)
+{
+	return ena_com_check_supported_feature_id(ena_dev, ENA_ADMIN_PHC_CONFIG);
+}
+
+int ena_com_phc_init(struct ena_com_dev *ena_dev)
+{
+	struct ena_com_phc_info *phc = &ena_dev->phc;
+
+	memset(phc, 0x0, sizeof(*phc));
+
+	/* Allocate shared mem used PHC timestamp retrieved from device */
+	phc->virt_addr = dma_alloc_coherent(ena_dev->dmadev,
+					    sizeof(*phc->virt_addr),
+					    &phc->phys_addr,
+					    GFP_KERNEL);
+	if (unlikely(!phc->virt_addr))
+		return -ENOMEM;
+
+	spin_lock_init(&phc->lock);
+
+	phc->virt_addr->req_id = 0;
+	phc->virt_addr->timestamp = 0;
+
+	return 0;
+}
+
+int ena_com_phc_config(struct ena_com_dev *ena_dev)
+{
+	struct ena_com_phc_info *phc = &ena_dev->phc;
+	struct ena_admin_get_feat_resp get_feat_resp;
+	struct ena_admin_set_feat_resp set_feat_resp;
+	struct ena_admin_set_feat_cmd set_feat_cmd;
+	int ret = 0;
+
+	/* Get device PHC default configuration */
+	ret = ena_com_get_feature(ena_dev,
+				  &get_feat_resp,
+				  ENA_ADMIN_PHC_CONFIG,
+				  0);
+	if (unlikely(ret)) {
+		netdev_err(ena_dev->net_device,
+			   "Failed to get PHC feature configuration, error: %d\n",
+			   ret);
+		return ret;
+	}
+
+	/* Supporting only readless PHC retrieval */
+	if (get_feat_resp.u.phc.type != ENA_ADMIN_PHC_TYPE_READLESS) {
+		netdev_err(ena_dev->net_device,
+			   "Unsupported PHC type, error: %d\n",
+			   -EOPNOTSUPP);
+		return -EOPNOTSUPP;
+	}
+
+	/* Update PHC doorbell offset according to device value,
+	 * used to write req_id to PHC bar
+	 */
+	phc->doorbell_offset = get_feat_resp.u.phc.doorbell_offset;
+
+	/* Update PHC expire timeout according to device
+	 * or default driver value
+	 */
+	phc->expire_timeout_usec = (get_feat_resp.u.phc.expire_timeout_usec) ?
+				    get_feat_resp.u.phc.expire_timeout_usec :
+				    ENA_PHC_DEFAULT_EXPIRE_TIMEOUT_USEC;
+
+	/* Update PHC block timeout according to device
+	 * or default driver value
+	 */
+	phc->block_timeout_usec = (get_feat_resp.u.phc.block_timeout_usec) ?
+				   get_feat_resp.u.phc.block_timeout_usec :
+				   ENA_PHC_DEFAULT_BLOCK_TIMEOUT_USEC;
+
+	/* Sanity check - expire timeout must not exceed block timeout */
+	if (phc->expire_timeout_usec > phc->block_timeout_usec)
+		phc->expire_timeout_usec = phc->block_timeout_usec;
+
+	/* Prepare PHC feature command */
+	memset(&set_feat_cmd, 0x0, sizeof(set_feat_cmd));
+	set_feat_cmd.aq_common_descriptor.opcode = ENA_ADMIN_SET_FEATURE;
+	set_feat_cmd.feat_common.feature_id = ENA_ADMIN_PHC_CONFIG;
+	set_feat_cmd.u.phc.output_length = sizeof(*phc->virt_addr);
+	ret = ena_com_mem_addr_set(ena_dev,
+				   &set_feat_cmd.u.phc.output_address,
+				   phc->phys_addr);
+	if (unlikely(ret)) {
+		netdev_err(ena_dev->net_device,
+			   "Failed setting PHC output address, error: %d\n",
+			   ret);
+		return ret;
+	}
+
+	/* Send PHC feature command to the device */
+	ret = ena_com_execute_admin_command(&ena_dev->admin_queue,
+					    (struct ena_admin_aq_entry *)&set_feat_cmd,
+					    sizeof(set_feat_cmd),
+					    (struct ena_admin_acq_entry *)&set_feat_resp,
+					    sizeof(set_feat_resp));
+
+	if (unlikely(ret)) {
+		netdev_err(ena_dev->net_device,
+			   "Failed to enable PHC, error: %d\n",
+			   ret);
+		return ret;
+	}
+
+	phc->active = true;
+	netdev_dbg(ena_dev->net_device, "PHC is active in the device\n");
+
+	return ret;
+}
+
+void ena_com_phc_destroy(struct ena_com_dev *ena_dev)
+{
+	struct ena_com_phc_info *phc = &ena_dev->phc;
+	unsigned long flags = 0;
+
+	/* In case PHC is not supported by the device, silently exiting */
+	if (!phc->virt_addr)
+		return;
+
+	spin_lock_irqsave(&phc->lock, flags);
+	phc->active = false;
+	spin_unlock_irqrestore(&phc->lock, flags);
+
+	dma_free_coherent(ena_dev->dmadev,
+			  sizeof(*phc->virt_addr),
+			  phc->virt_addr,
+			  phc->phys_addr);
+	phc->virt_addr = NULL;
+}
+
+int ena_com_phc_get_timestamp(struct ena_com_dev *ena_dev, u64 *timestamp)
+{
+	volatile struct ena_admin_phc_resp *resp = ena_dev->phc.virt_addr;
+	const ktime_t zero_system_time = ktime_set(0, 0);
+	struct ena_com_phc_info *phc = &ena_dev->phc;
+	ktime_t expire_time;
+	ktime_t block_time;
+	unsigned long flags = 0;
+	int ret = 0;
+
+	if (!phc->active) {
+		netdev_err(ena_dev->net_device, "PHC feature is not active in the device\n");
+		return -EOPNOTSUPP;
+	}
+
+	spin_lock_irqsave(&phc->lock, flags);
+
+	/* Check if PHC is in blocked state */
+	if (unlikely(ktime_compare(phc->system_time, zero_system_time))) {
+		/* Check if blocking time expired */
+		block_time = ktime_add_us(phc->system_time, phc->block_timeout_usec);
+		if (!ktime_after(ktime_get(), block_time)) {
+			/* PHC is still in blocked state, skip PHC request */
+			phc->stats.phc_skp++;
+			ret = -EBUSY;
+			goto skip;
+		}
+
+		/* PHC is in active state, update statistics according
+		 * to req_id and error_flags
+		 */
+		if (READ_ONCE(resp->req_id) != phc->req_id) {
+			/* Device didn't update req_id during blocking time,
+			 * this indicates on a device error
+			 */
+			netdev_err(ena_dev->net_device,
+				   "PHC get time request 0x%x failed (device error)\n",
+				   phc->req_id);
+			phc->stats.phc_err_dv++;
+		} else if (resp->error_flags & ENA_PHC_ERROR_FLAGS) {
+			/* Device updated req_id during blocking time but got
+			 * a PHC error, this occurs if device:
+			 * - exceeded the get time request limit
+			 * - received an invalid timestamp
+			 */
+			netdev_err(ena_dev->net_device,
+				   "PHC get time request 0x%x failed (error 0x%x)\n",
+				   phc->req_id,
+				   resp->error_flags);
+			phc->stats.phc_err_ts += !!(resp->error_flags &
+				ENA_ADMIN_PHC_ERROR_FLAG_TIMESTAMP);
+		} else {
+			/* Device updated req_id during blocking time
+			 * with valid timestamp
+			 */
+			phc->stats.phc_exp++;
+		}
+	}
+
+	/* Setting relative timeouts */
+	phc->system_time = ktime_get();
+	block_time = ktime_add_us(phc->system_time, phc->block_timeout_usec);
+	expire_time = ktime_add_us(phc->system_time, phc->expire_timeout_usec);
+
+	/* We expect the device to return this req_id once
+	 * the new PHC timestamp is updated
+	 */
+	phc->req_id++;
+
+	/* Initialize PHC shared memory with different req_id value
+	 * to be able to identify once the device changes it to req_id
+	 */
+	resp->req_id = phc->req_id + ENA_PHC_REQ_ID_OFFSET;
+
+	/* Writing req_id to PHC bar */
+	writel(phc->req_id, ena_dev->reg_bar + phc->doorbell_offset);
+
+	/* Stalling until the device updates req_id */
+	while (1) {
+		if (unlikely(ktime_after(ktime_get(), expire_time))) {
+			/* Gave up waiting for updated req_id, PHC enters into
+			 * blocked state until passing blocking time,
+			 * during this time any get PHC timestamp will fail with
+			 * device busy error
+			 */
+			ret = -EBUSY;
+			break;
+		}
+
+		/* Check if req_id was updated by the device */
+		if (READ_ONCE(resp->req_id) != phc->req_id) {
+			/* req_id was not updated by the device yet,
+			 * check again on next loop
+			 */
+			continue;
+		}
+
+		/* req_id was updated by the device which indicates that
+		 * PHC timestamp and error_flags are updated too,
+		 * checking errors before retrieving timestamp
+		 */
+		if (unlikely(resp->error_flags & ENA_PHC_ERROR_FLAGS)) {
+			/* Retrieved invalid PHC timestamp, PHC enters into
+			 * blocked state until passing blocking time,
+			 * during this time any get PHC timestamp requests
+			 * will fail with device busy error
+			 */
+			ret = -EBUSY;
+			break;
+		}
+
+		/* PHC timestamp value is returned to the caller */
+		*timestamp = resp->timestamp;
+
+		/* Update statistic on valid PHC timestamp retrieval */
+		phc->stats.phc_cnt++;
+
+		/* This indicates PHC state is active */
+		phc->system_time = zero_system_time;
+		break;
+	}
+
+skip:
+	spin_unlock_irqrestore(&phc->lock, flags);
+
+	return ret;
+}
+
 int ena_com_mmio_reg_read_request_init(struct ena_com_dev *ena_dev)
 {
 	struct ena_com_mmio_read *mmio_read = &ena_dev->mmio_read;
 
 	spin_lock_init(&mmio_read->lock);
-	mmio_read->read_resp =
-		dma_zalloc_coherent(ena_dev->dmadev,
-				    sizeof(*mmio_read->read_resp),
-				    &mmio_read->read_resp_dma_addr, GFP_KERNEL);
+	mmio_read->read_resp = dma_alloc_coherent(ena_dev->dmadev, sizeof(*mmio_read->read_resp),
+						  &mmio_read->read_resp_dma_addr, GFP_KERNEL);
 	if (unlikely(!mmio_read->read_resp))
-		return -ENOMEM;
+		goto err;
 
 	ena_com_mmio_reg_read_request_write_dev_addr(ena_dev);
 
@@ -1477,6 +1925,10 @@ int ena_com_mmio_reg_read_request_init(struct ena_com_dev *ena_dev)
 	mmio_read->readless_supported = true;
 
 	return 0;
+
+err:
+
+	return -ENOMEM;
 }
 
 void ena_com_set_mmio_read_mode(struct ena_com_dev *ena_dev, bool readless_supported)
@@ -1493,8 +1945,8 @@ void ena_com_mmio_reg_read_request_destroy(struct ena_com_dev *ena_dev)
 	writel(0x0, ena_dev->reg_bar + ENA_REGS_MMIO_RESP_LO_OFF);
 	writel(0x0, ena_dev->reg_bar + ENA_REGS_MMIO_RESP_HI_OFF);
 
-	dma_free_coherent(ena_dev->dmadev, sizeof(*mmio_read->read_resp),
-			  mmio_read->read_resp, mmio_read->read_resp_dma_addr);
+	dma_free_coherent(ena_dev->dmadev, sizeof(*mmio_read->read_resp), mmio_read->read_resp,
+			  mmio_read->read_resp_dma_addr);
 
 	mmio_read->read_resp = NULL;
 }
@@ -1512,8 +1964,7 @@ void ena_com_mmio_reg_read_request_write_dev_addr(struct ena_com_dev *ena_dev)
 }
 
 int ena_com_admin_init(struct ena_com_dev *ena_dev,
-		       struct ena_aenq_handlers *aenq_handlers,
-		       bool init_spinlock)
+		       struct ena_aenq_handlers *aenq_handlers)
 {
 	struct ena_com_admin_queue *admin_queue = &ena_dev->admin_queue;
 	u32 aq_caps, acq_caps, dev_sts, addr_low, addr_high;
@@ -1522,12 +1973,12 @@ int ena_com_admin_init(struct ena_com_dev *ena_dev,
 	dev_sts = ena_com_reg_bar_read32(ena_dev, ENA_REGS_DEV_STS_OFF);
 
 	if (unlikely(dev_sts == ENA_MMIO_READ_TIMEOUT)) {
-		pr_err("Reg read timeout occurred\n");
+		netdev_err(ena_dev->net_device, "Reg read timeout occurred\n");
 		return -ETIME;
 	}
 
 	if (!(dev_sts & ENA_REGS_DEV_STS_READY_MASK)) {
-		pr_err("Device isn't ready, abort com init\n");
+		netdev_err(ena_dev->net_device, "Device isn't ready, abort com init\n");
 		return -ENODEV;
 	}
 
@@ -1539,8 +1990,7 @@ int ena_com_admin_init(struct ena_com_dev *ena_dev,
 
 	atomic_set(&admin_queue->outstanding_cmds, 0);
 
-	if (init_spinlock)
-		spin_lock_init(&admin_queue->q_lock);
+	spin_lock_init(&admin_queue->q_lock);
 
 	ret = ena_com_init_comp_ctxt(admin_queue);
 	if (ret)
@@ -1587,6 +2037,7 @@ int ena_com_admin_init(struct ena_com_dev *ena_dev,
 	if (ret)
 		goto error;
 
+	admin_queue->ena_dev = ena_dev;
 	admin_queue->running_state = true;
 
 	return 0;
@@ -1604,16 +2055,16 @@ int ena_com_create_io_queue(struct ena_com_dev *ena_dev,
 	int ret;
 
 	if (ctx->qid >= ENA_TOTAL_NUM_QUEUES) {
-		pr_err("Qid (%d) is bigger than max num of queues (%d)\n",
-		       ctx->qid, ENA_TOTAL_NUM_QUEUES);
+		netdev_err(ena_dev->net_device, "Qid (%d) is bigger than max num of queues (%d)\n",
+			   ctx->qid, ENA_TOTAL_NUM_QUEUES);
 		return -EINVAL;
 	}
 
 	io_sq = &ena_dev->io_sq_queues[ctx->qid];
 	io_cq = &ena_dev->io_cq_queues[ctx->qid];
 
-	memset(io_sq, 0x0, sizeof(struct ena_com_io_sq));
-	memset(io_cq, 0x0, sizeof(struct ena_com_io_cq));
+	memset(io_sq, 0x0, sizeof(*io_sq));
+	memset(io_cq, 0x0, sizeof(*io_cq));
 
 	/* Init CQ */
 	io_cq->q_depth = ctx->queue_size;
@@ -1630,8 +2081,7 @@ int ena_com_create_io_queue(struct ena_com_dev *ena_dev,
 
 	if (ctx->direction == ENA_COM_IO_QUEUE_DIRECTION_TX)
 		/* header length is limited to 8 bits */
-		io_sq->tx_max_header_size =
-			min_t(u32, ena_dev->tx_max_header_size, SZ_256);
+		io_sq->tx_max_header_size = min_t(u32, ena_dev->tx_max_header_size, SZ_256);
 
 	ret = ena_com_init_io_sq(ena_dev, ctx, io_sq);
 	if (ret)
@@ -1663,8 +2113,8 @@ void ena_com_destroy_io_queue(struct ena_com_dev *ena_dev, u16 qid)
 	struct ena_com_io_cq *io_cq;
 
 	if (qid >= ENA_TOTAL_NUM_QUEUES) {
-		pr_err("Qid (%d) is bigger than max num of queues (%d)\n", qid,
-		       ENA_TOTAL_NUM_QUEUES);
+		netdev_err(ena_dev->net_device, "Qid (%d) is bigger than max num of queues (%d)\n",
+			   qid, ENA_TOTAL_NUM_QUEUES);
 		return;
 	}
 
@@ -1680,7 +2130,57 @@ void ena_com_destroy_io_queue(struct ena_com_dev *ena_dev, u16 qid)
 int ena_com_get_link_params(struct ena_com_dev *ena_dev,
 			    struct ena_admin_get_feat_resp *resp)
 {
-	return ena_com_get_feature(ena_dev, resp, ENA_ADMIN_LINK_CONFIG);
+	return ena_com_get_feature(ena_dev, resp, ENA_ADMIN_LINK_CONFIG, 0);
+}
+
+static int ena_get_dev_stats(struct ena_com_dev *ena_dev,
+			     struct ena_com_stats_ctx *ctx,
+			     enum ena_admin_get_stats_type type)
+{
+	struct ena_admin_acq_get_stats_resp *get_resp = &ctx->get_resp;
+	struct ena_admin_aq_get_stats_cmd *get_cmd = &ctx->get_cmd;
+	struct ena_com_admin_queue *admin_queue;
+	int ret;
+
+	admin_queue = &ena_dev->admin_queue;
+
+	get_cmd->aq_common_descriptor.opcode = ENA_ADMIN_GET_STATS;
+	get_cmd->aq_common_descriptor.flags = 0;
+	get_cmd->type = type;
+
+	ret = ena_com_execute_admin_command(admin_queue,
+					    (struct ena_admin_aq_entry *)get_cmd,
+					    sizeof(*get_cmd),
+					    (struct ena_admin_acq_entry *)get_resp,
+					    sizeof(*get_resp));
+
+	if (unlikely(ret))
+		netdev_err(ena_dev->net_device, "Failed to get stats. error: %d\n", ret);
+
+	return ret;
+}
+
+static void ena_com_set_supported_customer_metrics(struct ena_com_dev *ena_dev)
+{
+	struct ena_customer_metrics *customer_metrics;
+	struct ena_com_stats_ctx ctx;
+	int ret;
+
+	customer_metrics = &ena_dev->customer_metrics;
+	if (!ena_com_get_cap(ena_dev, ENA_ADMIN_CUSTOMER_METRICS)) {
+		customer_metrics->supported_metrics = ENA_ADMIN_CUSTOMER_METRICS_MIN_SUPPORT_MASK;
+		return;
+	}
+
+	memset(&ctx, 0x0, sizeof(ctx));
+	ctx.get_cmd.requested_metrics = ENA_ADMIN_CUSTOMER_METRICS_SUPPORT_MASK;
+	ret = ena_get_dev_stats(ena_dev, &ctx, ENA_ADMIN_GET_STATS_TYPE_CUSTOMER_METRICS);
+	if (likely(ret == 0))
+		customer_metrics->supported_metrics =
+			ctx.get_resp.u.customer_metrics.reported_metrics;
+	else
+		netdev_err(ena_dev->net_device,
+			   "Failed to query customer metrics support. error: %d\n", ret);
 }
 
 int ena_com_get_dev_attr_feat(struct ena_com_dev *ena_dev,
@@ -1690,25 +2190,44 @@ int ena_com_get_dev_attr_feat(struct ena_com_dev *ena_dev,
 	int rc;
 
 	rc = ena_com_get_feature(ena_dev, &get_resp,
-				 ENA_ADMIN_DEVICE_ATTRIBUTES);
+				 ENA_ADMIN_DEVICE_ATTRIBUTES, 0);
 	if (rc)
 		return rc;
 
 	memcpy(&get_feat_ctx->dev_attr, &get_resp.u.dev_attr,
 	       sizeof(get_resp.u.dev_attr));
+
 	ena_dev->supported_features = get_resp.u.dev_attr.supported_features;
+	ena_dev->capabilities = get_resp.u.dev_attr.capabilities;
 
-	rc = ena_com_get_feature(ena_dev, &get_resp,
-				 ENA_ADMIN_MAX_QUEUES_NUM);
-	if (rc)
-		return rc;
+	if (ena_dev->supported_features & BIT(ENA_ADMIN_MAX_QUEUES_EXT)) {
+		rc = ena_com_get_feature(ena_dev, &get_resp,
+					 ENA_ADMIN_MAX_QUEUES_EXT,
+					 ENA_FEATURE_MAX_QUEUE_EXT_VER);
+		if (rc)
+			return rc;
 
-	memcpy(&get_feat_ctx->max_queues, &get_resp.u.max_queue,
-	       sizeof(get_resp.u.max_queue));
-	ena_dev->tx_max_header_size = get_resp.u.max_queue.max_header_size;
+		if (get_resp.u.max_queue_ext.version != ENA_FEATURE_MAX_QUEUE_EXT_VER)
+			return -EINVAL;
+
+		memcpy(&get_feat_ctx->max_queue_ext, &get_resp.u.max_queue_ext,
+		       sizeof(get_resp.u.max_queue_ext));
+		ena_dev->tx_max_header_size =
+			get_resp.u.max_queue_ext.max_queue_ext.max_tx_header_size;
+	} else {
+		rc = ena_com_get_feature(ena_dev, &get_resp,
+					 ENA_ADMIN_MAX_QUEUES_NUM, 0);
+		memcpy(&get_feat_ctx->max_queues, &get_resp.u.max_queue,
+		       sizeof(get_resp.u.max_queue));
+		ena_dev->tx_max_header_size =
+			get_resp.u.max_queue.max_header_size;
+
+		if (rc)
+			return rc;
+	}
 
 	rc = ena_com_get_feature(ena_dev, &get_resp,
-				 ENA_ADMIN_AENQ_CONFIG);
+				 ENA_ADMIN_AENQ_CONFIG, 0);
 	if (rc)
 		return rc;
 
@@ -1716,13 +2235,35 @@ int ena_com_get_dev_attr_feat(struct ena_com_dev *ena_dev,
 	       sizeof(get_resp.u.aenq));
 
 	rc = ena_com_get_feature(ena_dev, &get_resp,
-				 ENA_ADMIN_STATELESS_OFFLOAD_CONFIG);
+				 ENA_ADMIN_STATELESS_OFFLOAD_CONFIG, 0);
 	if (rc)
 		return rc;
 
 	memcpy(&get_feat_ctx->offload, &get_resp.u.offload,
 	       sizeof(get_resp.u.offload));
 
+	/* Driver hints isn't mandatory admin command. So in case the
+	 * command isn't supported set driver hints to 0
+	 */
+	rc = ena_com_get_feature(ena_dev, &get_resp, ENA_ADMIN_HW_HINTS, 0);
+
+	if (!rc)
+		memcpy(&get_feat_ctx->hw_hints, &get_resp.u.hw_hints, sizeof(get_resp.u.hw_hints));
+	else if (rc == -EOPNOTSUPP)
+		memset(&get_feat_ctx->hw_hints, 0x0, sizeof(get_feat_ctx->hw_hints));
+	else
+		return rc;
+
+	rc = ena_com_get_feature(ena_dev, &get_resp, ENA_ADMIN_LLQ, 0);
+	if (!rc)
+		memcpy(&get_feat_ctx->llq, &get_resp.u.llq, sizeof(get_resp.u.llq));
+	else if (rc == -EOPNOTSUPP)
+		memset(&get_feat_ctx->llq, 0x0, sizeof(get_feat_ctx->llq));
+	else
+		return rc;
+
+	ena_com_set_supported_customer_metrics(ena_dev);
+
 	return 0;
 }
 
@@ -1734,10 +2275,10 @@ void ena_com_admin_q_comp_intr_handler(struct ena_com_dev *ena_dev)
 /* ena_handle_specific_aenq_event:
  * return the handler that is relevant to the specific event group
  */
-static ena_aenq_handler ena_com_get_specific_aenq_cb(struct ena_com_dev *dev,
+static ena_aenq_handler ena_com_get_specific_aenq_cb(struct ena_com_dev *ena_dev,
 						     u16 group)
 {
-	struct ena_aenq_handlers *aenq_handlers = dev->aenq.aenq_handlers;
+	struct ena_aenq_handlers *aenq_handlers = ena_dev->aenq.aenq_handlers;
 
 	if ((group < ENA_MAX_HANDLERS) && aenq_handlers->handlers[group])
 		return aenq_handlers->handlers[group];
@@ -1749,11 +2290,12 @@ static ena_aenq_handler ena_com_get_specific_aenq_cb(struct ena_com_dev *dev,
  * handles the aenq incoming events.
  * pop events from the queue and apply the specific handler
  */
-void ena_com_aenq_intr_handler(struct ena_com_dev *dev, void *data)
+void ena_com_aenq_intr_handler(struct ena_com_dev *ena_dev, void *data)
 {
 	struct ena_admin_aenq_entry *aenq_e;
 	struct ena_admin_aenq_common_desc *aenq_common;
-	struct ena_com_aenq *aenq  = &dev->aenq;
+	struct ena_com_aenq *aenq  = &ena_dev->aenq;
+	u64 timestamp;
 	ena_aenq_handler handler_cb;
 	u16 masked_head, processed = 0;
 	u8 phase;
@@ -1764,15 +2306,20 @@ void ena_com_aenq_intr_handler(struct ena_com_dev *dev, void *data)
 	aenq_common = &aenq_e->aenq_common_desc;
 
 	/* Go over all the events */
-	while ((aenq_common->flags & ENA_ADMIN_AENQ_COMMON_DESC_PHASE_MASK) ==
-	       phase) {
-		pr_debug("AENQ! Group[%x] Syndrom[%x] timestamp: [%llus]\n",
-			 aenq_common->group, aenq_common->syndrom,
-			 (u64)aenq_common->timestamp_low +
-				 ((u64)aenq_common->timestamp_high << 32));
+	while ((READ_ONCE(aenq_common->flags) & ENA_ADMIN_AENQ_COMMON_DESC_PHASE_MASK) == phase) {
+		/* Make sure the phase bit (ownership) is as expected before
+		 * reading the rest of the descriptor.
+		 */
+		dma_rmb();
+
+		timestamp = (u64)aenq_common->timestamp_low |
+			((u64)aenq_common->timestamp_high << 32);
+
+		netdev_dbg(ena_dev->net_device, "AENQ! Group[%x] Syndrome[%x] timestamp: [%llus]\n",
+			   aenq_common->group, aenq_common->syndrome, timestamp);
 
 		/* Handle specific event*/
-		handler_cb = ena_com_get_specific_aenq_cb(dev,
+		handler_cb = ena_com_get_specific_aenq_cb(ena_dev,
 							  aenq_common->group);
 		handler_cb(data, aenq_e); /* call the actual event handler*/
 
@@ -1797,10 +2344,11 @@ void ena_com_aenq_intr_handler(struct ena_com_dev *dev, void *data)
 
 	/* write the aenq doorbell after all AENQ descriptors were read */
 	mb();
-	writel((u32)aenq->head, dev->reg_bar + ENA_REGS_AENQ_HEAD_DB_OFF);
+	writel_relaxed((u32)aenq->head, ena_dev->reg_bar + ENA_REGS_AENQ_HEAD_DB_OFF);
 }
 
-int ena_com_dev_reset(struct ena_com_dev *ena_dev)
+int ena_com_dev_reset(struct ena_com_dev *ena_dev,
+		      enum ena_regs_reset_reason_types reset_reason)
 {
 	u32 stat, timeout, cap, reset_val;
 	int rc;
@@ -1808,26 +2356,27 @@ int ena_com_dev_reset(struct ena_com_dev *ena_dev)
 	stat = ena_com_reg_bar_read32(ena_dev, ENA_REGS_DEV_STS_OFF);
 	cap = ena_com_reg_bar_read32(ena_dev, ENA_REGS_CAPS_OFF);
 
-	if (unlikely((stat == ENA_MMIO_READ_TIMEOUT) ||
-		     (cap == ENA_MMIO_READ_TIMEOUT))) {
-		pr_err("Reg read32 timeout occurred\n");
+	if (unlikely((stat == ENA_MMIO_READ_TIMEOUT) || (cap == ENA_MMIO_READ_TIMEOUT))) {
+		netdev_err(ena_dev->net_device, "Reg read32 timeout occurred\n");
 		return -ETIME;
 	}
 
 	if ((stat & ENA_REGS_DEV_STS_READY_MASK) == 0) {
-		pr_err("Device isn't ready, can't reset device\n");
+		netdev_err(ena_dev->net_device, "Device isn't ready, can't reset device\n");
 		return -EINVAL;
 	}
 
 	timeout = (cap & ENA_REGS_CAPS_RESET_TIMEOUT_MASK) >>
 			ENA_REGS_CAPS_RESET_TIMEOUT_SHIFT;
 	if (timeout == 0) {
-		pr_err("Invalid timeout value\n");
+		netdev_err(ena_dev->net_device, "Invalid timeout value\n");
 		return -EINVAL;
 	}
 
 	/* start reset */
 	reset_val = ENA_REGS_DEV_CTL_DEV_RESET_MASK;
+	reset_val |= (reset_reason << ENA_REGS_DEV_CTL_RESET_REASON_SHIFT) &
+		     ENA_REGS_DEV_CTL_RESET_REASON_MASK;
 	writel(reset_val, ena_dev->reg_bar + ENA_REGS_DEV_CTL_OFF);
 
 	/* Write again the MMIO read request address */
@@ -1836,7 +2385,7 @@ int ena_com_dev_reset(struct ena_com_dev *ena_dev)
 	rc = wait_for_reset_state(ena_dev, timeout,
 				  ENA_REGS_DEV_STS_RESET_IN_PROGRESS_MASK);
 	if (rc != 0) {
-		pr_err("Reset indication didn't turn on\n");
+		netdev_err(ena_dev->net_device, "Reset indication didn't turn on\n");
 		return rc;
 	}
 
@@ -1844,56 +2393,108 @@ int ena_com_dev_reset(struct ena_com_dev *ena_dev)
 	writel(0, ena_dev->reg_bar + ENA_REGS_DEV_CTL_OFF);
 	rc = wait_for_reset_state(ena_dev, timeout, 0);
 	if (rc != 0) {
-		pr_err("Reset indication didn't turn off\n");
+		netdev_err(ena_dev->net_device, "Reset indication didn't turn off\n");
 		return rc;
 	}
 
+	timeout = (cap & ENA_REGS_CAPS_ADMIN_CMD_TO_MASK) >>
+		ENA_REGS_CAPS_ADMIN_CMD_TO_SHIFT;
+	if (timeout)
+		/* the resolution of timeout reg is 100ms */
+		ena_dev->admin_queue.completion_timeout = timeout * 100000;
+	else
+		ena_dev->admin_queue.completion_timeout = ADMIN_CMD_TIMEOUT_US;
+
 	return 0;
 }
 
-static int ena_get_dev_stats(struct ena_com_dev *ena_dev,
-			     struct ena_com_stats_ctx *ctx,
-			     enum ena_admin_get_stats_type type)
+int ena_com_get_eni_stats(struct ena_com_dev *ena_dev,
+			  struct ena_admin_eni_stats *stats)
 {
-	struct ena_admin_aq_get_stats_cmd *get_cmd = &ctx->get_cmd;
-	struct ena_admin_acq_get_stats_resp *get_resp = &ctx->get_resp;
-	struct ena_com_admin_queue *admin_queue;
+	struct ena_com_stats_ctx ctx;
 	int ret;
 
-	admin_queue = &ena_dev->admin_queue;
+	if (!ena_com_get_cap(ena_dev, ENA_ADMIN_ENI_STATS)) {
+		netdev_err(ena_dev->net_device, "Capability %d isn't supported\n",
+			   ENA_ADMIN_ENI_STATS);
+		return -EOPNOTSUPP;
+	}
 
-	get_cmd->aq_common_descriptor.opcode = ENA_ADMIN_GET_STATS;
-	get_cmd->aq_common_descriptor.flags = 0;
-	get_cmd->type = type;
+	memset(&ctx, 0x0, sizeof(ctx));
+	ret = ena_get_dev_stats(ena_dev, &ctx, ENA_ADMIN_GET_STATS_TYPE_ENI);
+	if (likely(ret == 0))
+		memcpy(stats, &ctx.get_resp.u.eni_stats,
+		       sizeof(ctx.get_resp.u.eni_stats));
 
-	ret =  ena_com_execute_admin_command(admin_queue,
-					     (struct ena_admin_aq_entry *)get_cmd,
-					     sizeof(*get_cmd),
-					     (struct ena_admin_acq_entry *)get_resp,
-					     sizeof(*get_resp));
+	return ret;
+}
 
-	if (unlikely(ret))
-		pr_err("Failed to get stats. error: %d\n", ret);
+int ena_com_get_ena_srd_info(struct ena_com_dev *ena_dev,
+			     struct ena_admin_ena_srd_info *info)
+{
+	struct ena_com_stats_ctx ctx;
+	int ret;
+
+	if (!ena_com_get_cap(ena_dev, ENA_ADMIN_ENA_SRD_INFO)) {
+		netdev_err(ena_dev->net_device, "Capability %d isn't supported\n",
+			   ENA_ADMIN_ENA_SRD_INFO);
+		return -EOPNOTSUPP;
+	}
+
+	memset(&ctx, 0x0, sizeof(ctx));
+	ret = ena_get_dev_stats(ena_dev, &ctx, ENA_ADMIN_GET_STATS_TYPE_ENA_SRD);
+	if (likely(ret == 0))
+		memcpy(info, &ctx.get_resp.u.ena_srd_info,
+		       sizeof(ctx.get_resp.u.ena_srd_info));
 
 	return ret;
 }
 
-int ena_com_get_dev_basic_stats(struct ena_com_dev *ena_dev,
-				struct ena_admin_basic_stats *stats)
+int ena_com_get_customer_metrics(struct ena_com_dev *ena_dev, char *buffer, u32 len)
 {
+	struct ena_admin_aq_get_stats_cmd *get_cmd;
 	struct ena_com_stats_ctx ctx;
 	int ret;
 
+	if (unlikely(len > ena_dev->customer_metrics.buffer_len)) {
+		netdev_err(ena_dev->net_device,
+			   "Invalid buffer size %u. The given buffer is too big.\n", len);
+		return -EINVAL;
+	}
+
+	if (!ena_com_get_cap(ena_dev, ENA_ADMIN_CUSTOMER_METRICS)) {
+		netdev_err(ena_dev->net_device, "Capability %d not supported.\n",
+			   ENA_ADMIN_CUSTOMER_METRICS);
+		return -EOPNOTSUPP;
+	}
+
+	if (!ena_dev->customer_metrics.supported_metrics) {
+		netdev_err(ena_dev->net_device, "No supported customer metrics.\n");
+		return -EOPNOTSUPP;
+	}
+
+	get_cmd = &ctx.get_cmd;
 	memset(&ctx, 0x0, sizeof(ctx));
-	ret = ena_get_dev_stats(ena_dev, &ctx, ENA_ADMIN_GET_STATS_TYPE_BASIC);
+	ret = ena_com_mem_addr_set(ena_dev,
+				   &get_cmd->u.control_buffer.address,
+				   ena_dev->customer_metrics.buffer_dma_addr);
+	if (unlikely(ret)) {
+		netdev_err(ena_dev->net_device, "Memory address set failed.\n");
+		return ret;
+	}
+
+	get_cmd->u.control_buffer.length = ena_dev->customer_metrics.buffer_len;
+	get_cmd->requested_metrics = ena_dev->customer_metrics.supported_metrics;
+	ret = ena_get_dev_stats(ena_dev, &ctx, ENA_ADMIN_GET_STATS_TYPE_CUSTOMER_METRICS);
 	if (likely(ret == 0))
-		memcpy(stats, &ctx.get_resp.basic_stats,
-		       sizeof(ctx.get_resp.basic_stats));
+		memcpy(buffer, ena_dev->customer_metrics.buffer_virt_addr, len);
+	else
+		netdev_err(ena_dev->net_device, "Failed to get customer metrics. error: %d\n", ret);
 
 	return ret;
 }
 
-int ena_com_set_dev_mtu(struct ena_com_dev *ena_dev, int mtu)
+int ena_com_set_dev_mtu(struct ena_com_dev *ena_dev, u32 mtu)
 {
 	struct ena_com_admin_queue *admin_queue;
 	struct ena_admin_set_feat_cmd cmd;
@@ -1901,8 +2502,8 @@ int ena_com_set_dev_mtu(struct ena_com_dev *ena_dev, int mtu)
 	int ret;
 
 	if (!ena_com_check_supported_feature_id(ena_dev, ENA_ADMIN_MTU)) {
-		pr_debug("Feature %d isn't supported\n", ENA_ADMIN_MTU);
-		return -EPERM;
+		netdev_dbg(ena_dev->net_device, "Feature %d isn't supported\n", ENA_ADMIN_MTU);
+		return -EOPNOTSUPP;
 	}
 
 	memset(&cmd, 0x0, sizeof(cmd));
@@ -1920,29 +2521,11 @@ int ena_com_set_dev_mtu(struct ena_com_dev *ena_dev, int mtu)
 					    sizeof(resp));
 
 	if (unlikely(ret))
-		pr_err("Failed to set mtu %d. error: %d\n", mtu, ret);
+		netdev_err(ena_dev->net_device, "Failed to set mtu %d. error: %d\n", mtu, ret);
 
 	return ret;
 }
 
-int ena_com_get_offload_settings(struct ena_com_dev *ena_dev,
-				 struct ena_admin_feature_offload_desc *offload)
-{
-	int ret;
-	struct ena_admin_get_feat_resp resp;
-
-	ret = ena_com_get_feature(ena_dev, &resp,
-				  ENA_ADMIN_STATELESS_OFFLOAD_CONFIG);
-	if (unlikely(ret)) {
-		pr_err("Failed to get offload capabilities %d\n", ret);
-		return ret;
-	}
-
-	memcpy(offload, &resp.u.offload, sizeof(resp.u.offload));
-
-	return 0;
-}
-
 int ena_com_set_hash_function(struct ena_com_dev *ena_dev)
 {
 	struct ena_com_admin_queue *admin_queue = &ena_dev->admin_queue;
@@ -1952,23 +2535,22 @@ int ena_com_set_hash_function(struct ena_com_dev *ena_dev)
 	struct ena_admin_get_feat_resp get_resp;
 	int ret;
 
-	if (!ena_com_check_supported_feature_id(ena_dev,
-						ENA_ADMIN_RSS_HASH_FUNCTION)) {
-		pr_debug("Feature %d isn't supported\n",
-			 ENA_ADMIN_RSS_HASH_FUNCTION);
-		return -EPERM;
+	if (!ena_com_check_supported_feature_id(ena_dev, ENA_ADMIN_RSS_HASH_FUNCTION)) {
+		netdev_dbg(ena_dev->net_device, "Feature %d isn't supported\n",
+			   ENA_ADMIN_RSS_HASH_FUNCTION);
+		return -EOPNOTSUPP;
 	}
 
 	/* Validate hash function is supported */
 	ret = ena_com_get_feature(ena_dev, &get_resp,
-				  ENA_ADMIN_RSS_HASH_FUNCTION);
+				  ENA_ADMIN_RSS_HASH_FUNCTION, 0);
 	if (unlikely(ret))
 		return ret;
 
-	if (get_resp.u.flow_hash_func.supported_func & (1 << rss->hash_func)) {
-		pr_err("Func hash %d isn't supported by device, abort\n",
-		       rss->hash_func);
-		return -EPERM;
+	if (!(get_resp.u.flow_hash_func.supported_func & BIT(rss->hash_func))) {
+		netdev_err(ena_dev->net_device, "Func hash %d isn't supported by device, abort\n",
+			   rss->hash_func);
+		return -EOPNOTSUPP;
 	}
 
 	memset(&cmd, 0x0, sizeof(cmd));
@@ -1984,7 +2566,7 @@ int ena_com_set_hash_function(struct ena_com_dev *ena_dev)
 				   &cmd.control_buffer.address,
 				   rss->hash_key_dma_addr);
 	if (unlikely(ret)) {
-		pr_err("memory address set failed\n");
+		netdev_err(ena_dev->net_device, "Memory address set failed\n");
 		return ret;
 	}
 
@@ -1996,8 +2578,8 @@ int ena_com_set_hash_function(struct ena_com_dev *ena_dev)
 					    (struct ena_admin_acq_entry *)&resp,
 					    sizeof(resp));
 	if (unlikely(ret)) {
-		pr_err("Failed to set hash function %d. error: %d\n",
-		       rss->hash_func, ret);
+		netdev_err(ena_dev->net_device, "Failed to set hash function %d. error: %d\n",
+			   rss->hash_func, ret);
 		return -EINVAL;
 	}
 
@@ -2008,12 +2590,14 @@ int ena_com_fill_hash_function(struct ena_com_dev *ena_dev,
 			       enum ena_admin_hash_functions func,
 			       const u8 *key, u16 key_len, u32 init_val)
 {
-	struct ena_rss *rss = &ena_dev->rss;
+	struct ena_admin_feature_rss_flow_hash_control *hash_key;
 	struct ena_admin_get_feat_resp get_resp;
-	struct ena_admin_feature_rss_flow_hash_control *hash_key =
-		rss->hash_key;
+	enum ena_admin_hash_functions old_func;
+	struct ena_rss *rss = &ena_dev->rss;
 	int rc;
 
+	hash_key = rss->hash_key;
+
 	/* Make sure size is a mult of DWs */
 	if (unlikely(key_len & 0x3))
 		return -EINVAL;
@@ -2021,67 +2605,73 @@ int ena_com_fill_hash_function(struct ena_com_dev *ena_dev,
 	rc = ena_com_get_feature_ex(ena_dev, &get_resp,
 				    ENA_ADMIN_RSS_HASH_FUNCTION,
 				    rss->hash_key_dma_addr,
-				    sizeof(*rss->hash_key));
+				    sizeof(*rss->hash_key), 0);
 	if (unlikely(rc))
 		return rc;
 
-	if (!((1 << func) & get_resp.u.flow_hash_func.supported_func)) {
-		pr_err("Flow hash function %d isn't supported\n", func);
-		return -EPERM;
+	if (!(BIT(func) & get_resp.u.flow_hash_func.supported_func)) {
+		netdev_err(ena_dev->net_device, "Flow hash function %d isn't supported\n", func);
+		return -EOPNOTSUPP;
 	}
 
-	switch (func) {
-	case ENA_ADMIN_TOEPLITZ:
-		if (key_len > sizeof(hash_key->key)) {
-			pr_err("key len (%hu) is bigger than the max supported (%zu)\n",
-			       key_len, sizeof(hash_key->key));
+	if ((func == ENA_ADMIN_TOEPLITZ) && key) {
+		if (key_len != sizeof(hash_key->key)) {
+			netdev_err(ena_dev->net_device,
+				   "key len (%u) doesn't equal the supported size (%zu)\n", key_len,
+				   sizeof(hash_key->key));
 			return -EINVAL;
 		}
-
 		memcpy(hash_key->key, key, key_len);
-		rss->hash_init_val = init_val;
-		hash_key->keys_num = key_len >> 2;
-		break;
-	case ENA_ADMIN_CRC32:
-		rss->hash_init_val = init_val;
-		break;
-	default:
-		pr_err("Invalid hash function (%d)\n", func);
-		return -EINVAL;
+		hash_key->key_parts = key_len / sizeof(hash_key->key[0]);
 	}
 
+	rss->hash_init_val = init_val;
+	old_func = rss->hash_func;
+	rss->hash_func = func;
 	rc = ena_com_set_hash_function(ena_dev);
 
 	/* Restore the old function */
 	if (unlikely(rc))
-		ena_com_get_hash_function(ena_dev, NULL, NULL);
+		rss->hash_func = old_func;
 
 	return rc;
 }
 
 int ena_com_get_hash_function(struct ena_com_dev *ena_dev,
-			      enum ena_admin_hash_functions *func,
-			      u8 *key)
+			      enum ena_admin_hash_functions *func)
 {
 	struct ena_rss *rss = &ena_dev->rss;
 	struct ena_admin_get_feat_resp get_resp;
-	struct ena_admin_feature_rss_flow_hash_control *hash_key =
-		rss->hash_key;
 	int rc;
 
+	if (unlikely(!func))
+		return -EINVAL;
+
 	rc = ena_com_get_feature_ex(ena_dev, &get_resp,
 				    ENA_ADMIN_RSS_HASH_FUNCTION,
 				    rss->hash_key_dma_addr,
-				    sizeof(*rss->hash_key));
+				    sizeof(*rss->hash_key), 0);
 	if (unlikely(rc))
 		return rc;
 
-	rss->hash_func = get_resp.u.flow_hash_func.selected_func;
-	if (func)
-		*func = rss->hash_func;
+	/* ffs() returns 1 in case the lsb is set */
+	rss->hash_func = ffs(get_resp.u.flow_hash_func.selected_func);
+	if (rss->hash_func)
+		rss->hash_func--;
+
+	*func = rss->hash_func;
+
+	return 0;
+}
+
+int ena_com_get_hash_key(struct ena_com_dev *ena_dev, u8 *key)
+{
+	struct ena_admin_feature_rss_flow_hash_control *hash_key =
+		ena_dev->rss.hash_key;
 
 	if (key)
-		memcpy(key, hash_key->key, (size_t)(hash_key->keys_num) << 2);
+		memcpy(key, hash_key->key,
+		       (size_t)(hash_key->key_parts) * sizeof(hash_key->key[0]));
 
 	return 0;
 }
@@ -2097,7 +2687,7 @@ int ena_com_get_hash_ctrl(struct ena_com_dev *ena_dev,
 	rc = ena_com_get_feature_ex(ena_dev, &get_resp,
 				    ENA_ADMIN_RSS_HASH_INPUT,
 				    rss->hash_ctrl_dma_addr,
-				    sizeof(*rss->hash_ctrl));
+				    sizeof(*rss->hash_ctrl), 0);
 	if (unlikely(rc))
 		return rc;
 
@@ -2116,11 +2706,10 @@ int ena_com_set_hash_ctrl(struct ena_com_dev *ena_dev)
 	struct ena_admin_set_feat_resp resp;
 	int ret;
 
-	if (!ena_com_check_supported_feature_id(ena_dev,
-						ENA_ADMIN_RSS_HASH_INPUT)) {
-		pr_debug("Feature %d isn't supported\n",
-			 ENA_ADMIN_RSS_HASH_INPUT);
-		return -EPERM;
+	if (!ena_com_check_supported_feature_id(ena_dev, ENA_ADMIN_RSS_HASH_INPUT)) {
+		netdev_dbg(ena_dev->net_device, "Feature %d isn't supported\n",
+			   ENA_ADMIN_RSS_HASH_INPUT);
+		return -EOPNOTSUPP;
 	}
 
 	memset(&cmd, 0x0, sizeof(cmd));
@@ -2137,7 +2726,7 @@ int ena_com_set_hash_ctrl(struct ena_com_dev *ena_dev)
 				   &cmd.control_buffer.address,
 				   rss->hash_ctrl_dma_addr);
 	if (unlikely(ret)) {
-		pr_err("memory address set failed\n");
+		netdev_err(ena_dev->net_device, "Memory address set failed\n");
 		return ret;
 	}
 	cmd.control_buffer.length = sizeof(*hash_ctrl);
@@ -2148,7 +2737,7 @@ int ena_com_set_hash_ctrl(struct ena_com_dev *ena_dev)
 					    (struct ena_admin_acq_entry *)&resp,
 					    sizeof(resp));
 	if (unlikely(ret))
-		pr_err("Failed to set hash input. error: %d\n", ret);
+		netdev_err(ena_dev->net_device, "Failed to set hash input. error: %d\n", ret);
 
 	return ret;
 }
@@ -2198,10 +2787,11 @@ int ena_com_set_default_hash_ctrl(struct ena_com_dev *ena_dev)
 		available_fields = hash_ctrl->selected_fields[i].fields &
 				hash_ctrl->supported_fields[i].fields;
 		if (available_fields != hash_ctrl->selected_fields[i].fields) {
-			pr_err("hash control doesn't support all the desire configuration. proto %x supported %x selected %x\n",
-			       i, hash_ctrl->supported_fields[i].fields,
-			       hash_ctrl->selected_fields[i].fields);
-			return -EPERM;
+			netdev_err(ena_dev->net_device,
+				   "Hash control doesn't support all the desire configuration. proto %x supported %x selected %x\n",
+				   i, hash_ctrl->supported_fields[i].fields,
+				   hash_ctrl->selected_fields[i].fields);
+			return -EOPNOTSUPP;
 		}
 	}
 
@@ -2224,7 +2814,7 @@ int ena_com_fill_hash_ctrl(struct ena_com_dev *ena_dev,
 	int rc;
 
 	if (proto >= ENA_ADMIN_RSS_PROTO_NUM) {
-		pr_err("Invalid proto num (%u)\n", proto);
+		netdev_err(ena_dev->net_device, "Invalid proto num (%u)\n", proto);
 		return -EINVAL;
 	}
 
@@ -2236,8 +2826,9 @@ int ena_com_fill_hash_ctrl(struct ena_com_dev *ena_dev,
 	/* Make sure all the fields are supported */
 	supported_fields = hash_ctrl->supported_fields[proto].fields;
 	if ((hash_fields & supported_fields) != hash_fields) {
-		pr_err("proto %d doesn't support the required fields %x. supports only: %x\n",
-		       proto, hash_fields, supported_fields);
+		netdev_err(ena_dev->net_device,
+			   "Proto %d doesn't support the required fields %x. supports only: %x\n",
+			   proto, hash_fields, supported_fields);
 	}
 
 	hash_ctrl->selected_fields[proto].fields = hash_fields;
@@ -2275,16 +2866,16 @@ int ena_com_indirect_table_set(struct ena_com_dev *ena_dev)
 	struct ena_admin_set_feat_resp resp;
 	int ret;
 
-	if (!ena_com_check_supported_feature_id(
-		    ena_dev, ENA_ADMIN_RSS_REDIRECTION_TABLE_CONFIG)) {
-		pr_debug("Feature %d isn't supported\n",
-			 ENA_ADMIN_RSS_REDIRECTION_TABLE_CONFIG);
-		return -EPERM;
+	if (!ena_com_check_supported_feature_id(ena_dev, ENA_ADMIN_RSS_INDIRECTION_TABLE_CONFIG)) {
+		netdev_dbg(ena_dev->net_device, "Feature %d isn't supported\n",
+			   ENA_ADMIN_RSS_INDIRECTION_TABLE_CONFIG);
+		return -EOPNOTSUPP;
 	}
 
 	ret = ena_com_ind_tbl_convert_to_device(ena_dev);
 	if (ret) {
-		pr_err("Failed to convert host indirection table to device table\n");
+		netdev_err(ena_dev->net_device,
+			   "Failed to convert host indirection table to device table\n");
 		return ret;
 	}
 
@@ -2293,7 +2884,7 @@ int ena_com_indirect_table_set(struct ena_com_dev *ena_dev)
 	cmd.aq_common_descriptor.opcode = ENA_ADMIN_SET_FEATURE;
 	cmd.aq_common_descriptor.flags =
 		ENA_ADMIN_AQ_COMMON_DESC_CTRL_DATA_INDIRECT_MASK;
-	cmd.feat_common.feature_id = ENA_ADMIN_RSS_REDIRECTION_TABLE_CONFIG;
+	cmd.feat_common.feature_id = ENA_ADMIN_RSS_INDIRECTION_TABLE_CONFIG;
 	cmd.u.ind_table.size = rss->tbl_log_size;
 	cmd.u.ind_table.inline_index = 0xFFFFFFFF;
 
@@ -2301,7 +2892,7 @@ int ena_com_indirect_table_set(struct ena_com_dev *ena_dev)
 				   &cmd.control_buffer.address,
 				   rss->rss_ind_tbl_dma_addr);
 	if (unlikely(ret)) {
-		pr_err("memory address set failed\n");
+		netdev_err(ena_dev->net_device, "Memory address set failed\n");
 		return ret;
 	}
 
@@ -2315,7 +2906,7 @@ int ena_com_indirect_table_set(struct ena_com_dev *ena_dev)
 					    sizeof(resp));
 
 	if (unlikely(ret))
-		pr_err("Failed to set indirect table. error: %d\n", ret);
+		netdev_err(ena_dev->net_device, "Failed to set indirect table. error: %d\n", ret);
 
 	return ret;
 }
@@ -2331,19 +2922,15 @@ int ena_com_indirect_table_get(struct ena_com_dev *ena_dev, u32 *ind_tbl)
 		sizeof(struct ena_admin_rss_ind_table_entry);
 
 	rc = ena_com_get_feature_ex(ena_dev, &get_resp,
-				    ENA_ADMIN_RSS_REDIRECTION_TABLE_CONFIG,
+				    ENA_ADMIN_RSS_INDIRECTION_TABLE_CONFIG,
 				    rss->rss_ind_tbl_dma_addr,
-				    tbl_size);
+				    tbl_size, 0);
 	if (unlikely(rc))
 		return rc;
 
 	if (!ind_tbl)
 		return 0;
 
-	rc = ena_com_ind_tbl_convert_from_device(ena_dev);
-	if (unlikely(rc))
-		return rc;
-
 	for (i = 0; i < (1 << rss->tbl_log_size); i++)
 		ind_tbl[i] = rss->host_rss_ind_tbl[i];
 
@@ -2360,8 +2947,14 @@ int ena_com_rss_init(struct ena_com_dev *ena_dev, u16 indr_tbl_log_size)
 	if (unlikely(rc))
 		goto err_indr_tbl;
 
+	/* The following function might return unsupported in case the
+	 * device doesn't support setting the key / hash function. We can safely
+	 * ignore this error and have indirection table support only.
+	 */
 	rc = ena_com_hash_key_allocate(ena_dev);
-	if (unlikely(rc))
+	if (likely(!rc))
+		ena_com_hash_key_fill_default_key(ena_dev);
+	else if (rc != -EOPNOTSUPP)
 		goto err_hash_key;
 
 	rc = ena_com_hash_ctrl_init(ena_dev);
@@ -2392,12 +2985,15 @@ int ena_com_allocate_host_info(struct ena_com_dev *ena_dev)
 {
 	struct ena_host_attribute *host_attr = &ena_dev->host_attr;
 
-	host_attr->host_info =
-		dma_zalloc_coherent(ena_dev->dmadev, SZ_4K,
-				    &host_attr->host_info_dma_addr, GFP_KERNEL);
+	host_attr->host_info = dma_alloc_coherent(ena_dev->dmadev, SZ_4K,
+						  &host_attr->host_info_dma_addr, GFP_KERNEL);
 	if (unlikely(!host_attr->host_info))
 		return -ENOMEM;
 
+	host_attr->host_info->ena_spec_version = ((ENA_COMMON_SPEC_VERSION_MAJOR <<
+		ENA_REGS_VERSION_MAJOR_VERSION_SHIFT) |
+		(ENA_COMMON_SPEC_VERSION_MINOR));
+
 	return 0;
 }
 
@@ -2407,8 +3003,8 @@ int ena_com_allocate_debug_area(struct ena_com_dev *ena_dev,
 	struct ena_host_attribute *host_attr = &ena_dev->host_attr;
 
 	host_attr->debug_area_virt_addr =
-		dma_zalloc_coherent(ena_dev->dmadev, debug_area_size,
-				    &host_attr->debug_area_dma_addr, GFP_KERNEL);
+		dma_alloc_coherent(ena_dev->dmadev, debug_area_size,
+				   &host_attr->debug_area_dma_addr, GFP_KERNEL);
 	if (unlikely(!host_attr->debug_area_virt_addr)) {
 		host_attr->debug_area_size = 0;
 		return -ENOMEM;
@@ -2419,6 +3015,24 @@ int ena_com_allocate_debug_area(struct ena_com_dev *ena_dev,
 	return 0;
 }
 
+int ena_com_allocate_customer_metrics_buffer(struct ena_com_dev *ena_dev)
+{
+	struct ena_customer_metrics *customer_metrics = &ena_dev->customer_metrics;
+
+	customer_metrics->buffer_len = ENA_CUSTOMER_METRICS_BUFFER_SIZE;
+	customer_metrics->buffer_virt_addr = NULL;
+
+	customer_metrics->buffer_virt_addr =
+		dma_alloc_coherent(ena_dev->dmadev, customer_metrics->buffer_len,
+				   &customer_metrics->buffer_dma_addr, GFP_KERNEL);
+	if (!customer_metrics->buffer_virt_addr) {
+		customer_metrics->buffer_len = 0;
+		return -ENOMEM;
+	}
+
+	return 0;
+}
+
 void ena_com_delete_host_info(struct ena_com_dev *ena_dev)
 {
 	struct ena_host_attribute *host_attr = &ena_dev->host_attr;
@@ -2436,12 +3050,24 @@ void ena_com_delete_debug_area(struct ena_com_dev *ena_dev)
 
 	if (host_attr->debug_area_virt_addr) {
 		dma_free_coherent(ena_dev->dmadev, host_attr->debug_area_size,
-				  host_attr->debug_area_virt_addr,
-				  host_attr->debug_area_dma_addr);
+				  host_attr->debug_area_virt_addr, host_attr->debug_area_dma_addr);
 		host_attr->debug_area_virt_addr = NULL;
 	}
 }
 
+void ena_com_delete_customer_metrics_buffer(struct ena_com_dev *ena_dev)
+{
+	struct ena_customer_metrics *customer_metrics = &ena_dev->customer_metrics;
+
+	if (customer_metrics->buffer_virt_addr) {
+		dma_free_coherent(ena_dev->dmadev, customer_metrics->buffer_len,
+				  customer_metrics->buffer_virt_addr,
+				  customer_metrics->buffer_dma_addr);
+		customer_metrics->buffer_virt_addr = NULL;
+		customer_metrics->buffer_len = 0;
+	}
+}
+
 int ena_com_set_host_attributes(struct ena_com_dev *ena_dev)
 {
 	struct ena_host_attribute *host_attr = &ena_dev->host_attr;
@@ -2465,7 +3091,7 @@ int ena_com_set_host_attributes(struct ena_com_dev *ena_dev)
 				   &cmd.u.host_attr.debug_ba,
 				   host_attr->debug_area_dma_addr);
 	if (unlikely(ret)) {
-		pr_err("memory address set failed\n");
+		netdev_err(ena_dev->net_device, "Memory address set failed\n");
 		return ret;
 	}
 
@@ -2473,7 +3099,7 @@ int ena_com_set_host_attributes(struct ena_com_dev *ena_dev)
 				   &cmd.u.host_attr.os_info_ba,
 				   host_attr->host_info_dma_addr);
 	if (unlikely(ret)) {
-		pr_err("memory address set failed\n");
+		netdev_err(ena_dev->net_device, "Memory address set failed\n");
 		return ret;
 	}
 
@@ -2486,7 +3112,7 @@ int ena_com_set_host_attributes(struct ena_com_dev *ena_dev)
 					    sizeof(resp));
 
 	if (unlikely(ret))
-		pr_err("Failed to set host attributes: %d\n", ret);
+		netdev_err(ena_dev->net_device, "Failed to set host attributes: %d\n", ret);
 
 	return ret;
 }
@@ -2498,42 +3124,37 @@ bool ena_com_interrupt_moderation_supported(struct ena_com_dev *ena_dev)
 						  ENA_ADMIN_INTERRUPT_MODERATION);
 }
 
-int ena_com_update_nonadaptive_moderation_interval_tx(struct ena_com_dev *ena_dev,
-						      u32 tx_coalesce_usecs)
+static int ena_com_update_nonadaptive_moderation_interval(struct ena_com_dev *ena_dev,
+							  u32 coalesce_usecs,
+							  u32 intr_delay_resolution,
+							  u32 *intr_moder_interval)
 {
-	if (!ena_dev->intr_delay_resolution) {
-		pr_err("Illegal interrupt delay granularity value\n");
+	if (!intr_delay_resolution) {
+		netdev_err(ena_dev->net_device, "Illegal interrupt delay granularity value\n");
 		return -EFAULT;
 	}
 
-	ena_dev->intr_moder_tx_interval = tx_coalesce_usecs /
-		ena_dev->intr_delay_resolution;
+	*intr_moder_interval = coalesce_usecs / intr_delay_resolution;
 
 	return 0;
 }
 
-int ena_com_update_nonadaptive_moderation_interval_rx(struct ena_com_dev *ena_dev,
-						      u32 rx_coalesce_usecs)
+int ena_com_update_nonadaptive_moderation_interval_tx(struct ena_com_dev *ena_dev,
+						      u32 tx_coalesce_usecs)
 {
-	if (!ena_dev->intr_delay_resolution) {
-		pr_err("Illegal interrupt delay granularity value\n");
-		return -EFAULT;
-	}
-
-	/* We use LOWEST entry of moderation table for storing
-	 * nonadaptive interrupt coalescing values
-	 */
-	ena_dev->intr_moder_tbl[ENA_INTR_MODER_LOWEST].intr_moder_interval =
-		rx_coalesce_usecs / ena_dev->intr_delay_resolution;
-
-	return 0;
+	return ena_com_update_nonadaptive_moderation_interval(ena_dev,
+							      tx_coalesce_usecs,
+							      ena_dev->intr_delay_resolution,
+							      &ena_dev->intr_moder_tx_interval);
 }
 
-void ena_com_destroy_interrupt_moderation(struct ena_com_dev *ena_dev)
+int ena_com_update_nonadaptive_moderation_interval_rx(struct ena_com_dev *ena_dev,
+						      u32 rx_coalesce_usecs)
 {
-	if (ena_dev->intr_moder_tbl)
-		devm_kfree(ena_dev->dmadev, ena_dev->intr_moder_tbl);
-	ena_dev->intr_moder_tbl = NULL;
+	return ena_com_update_nonadaptive_moderation_interval(ena_dev,
+							      rx_coalesce_usecs,
+							      ena_dev->intr_delay_resolution,
+							      &ena_dev->intr_moder_rx_interval);
 }
 
 int ena_com_init_interrupt_moderation(struct ena_com_dev *ena_dev)
@@ -2543,16 +3164,16 @@ int ena_com_init_interrupt_moderation(struct ena_com_dev *ena_dev)
 	int rc;
 
 	rc = ena_com_get_feature(ena_dev, &get_resp,
-				 ENA_ADMIN_INTERRUPT_MODERATION);
+				 ENA_ADMIN_INTERRUPT_MODERATION, 0);
 
 	if (rc) {
-		if (rc == -EPERM) {
-			pr_debug("Feature %d isn't supported\n",
-				 ENA_ADMIN_INTERRUPT_MODERATION);
+		if (rc == -EOPNOTSUPP) {
+			netdev_dbg(ena_dev->net_device, "Feature %d isn't supported\n",
+				   ENA_ADMIN_INTERRUPT_MODERATION);
 			rc = 0;
 		} else {
-			pr_err("Failed to get interrupt moderation admin cmd. rc: %d\n",
-			       rc);
+			netdev_err(ena_dev->net_device,
+				   "Failed to get interrupt moderation admin cmd. rc: %d\n", rc);
 		}
 
 		/* no moderation supported, disable adaptive support */
@@ -2560,62 +3181,14 @@ int ena_com_init_interrupt_moderation(struct ena_com_dev *ena_dev)
 		return rc;
 	}
 
-	rc = ena_com_init_interrupt_moderation_table(ena_dev);
-	if (rc)
-		goto err;
-
 	/* if moderation is supported by device we set adaptive moderation */
 	delay_resolution = get_resp.u.intr_moderation.intr_delay_resolution;
 	ena_com_update_intr_delay_resolution(ena_dev, delay_resolution);
-	ena_com_enable_adaptive_moderation(ena_dev);
 
-	return 0;
-err:
-	ena_com_destroy_interrupt_moderation(ena_dev);
-	return rc;
-}
-
-void ena_com_config_default_interrupt_moderation_table(struct ena_com_dev *ena_dev)
-{
-	struct ena_intr_moder_entry *intr_moder_tbl = ena_dev->intr_moder_tbl;
+	/* Disable adaptive moderation by default - can be enabled later */
+	ena_com_disable_adaptive_moderation(ena_dev);
 
-	if (!intr_moder_tbl)
-		return;
-
-	intr_moder_tbl[ENA_INTR_MODER_LOWEST].intr_moder_interval =
-		ENA_INTR_LOWEST_USECS;
-	intr_moder_tbl[ENA_INTR_MODER_LOWEST].pkts_per_interval =
-		ENA_INTR_LOWEST_PKTS;
-	intr_moder_tbl[ENA_INTR_MODER_LOWEST].bytes_per_interval =
-		ENA_INTR_LOWEST_BYTES;
-
-	intr_moder_tbl[ENA_INTR_MODER_LOW].intr_moder_interval =
-		ENA_INTR_LOW_USECS;
-	intr_moder_tbl[ENA_INTR_MODER_LOW].pkts_per_interval =
-		ENA_INTR_LOW_PKTS;
-	intr_moder_tbl[ENA_INTR_MODER_LOW].bytes_per_interval =
-		ENA_INTR_LOW_BYTES;
-
-	intr_moder_tbl[ENA_INTR_MODER_MID].intr_moder_interval =
-		ENA_INTR_MID_USECS;
-	intr_moder_tbl[ENA_INTR_MODER_MID].pkts_per_interval =
-		ENA_INTR_MID_PKTS;
-	intr_moder_tbl[ENA_INTR_MODER_MID].bytes_per_interval =
-		ENA_INTR_MID_BYTES;
-
-	intr_moder_tbl[ENA_INTR_MODER_HIGH].intr_moder_interval =
-		ENA_INTR_HIGH_USECS;
-	intr_moder_tbl[ENA_INTR_MODER_HIGH].pkts_per_interval =
-		ENA_INTR_HIGH_PKTS;
-	intr_moder_tbl[ENA_INTR_MODER_HIGH].bytes_per_interval =
-		ENA_INTR_HIGH_BYTES;
-
-	intr_moder_tbl[ENA_INTR_MODER_HIGHEST].intr_moder_interval =
-		ENA_INTR_HIGHEST_USECS;
-	intr_moder_tbl[ENA_INTR_MODER_HIGHEST].pkts_per_interval =
-		ENA_INTR_HIGHEST_PKTS;
-	intr_moder_tbl[ENA_INTR_MODER_HIGHEST].bytes_per_interval =
-		ENA_INTR_HIGHEST_BYTES;
+	return 0;
 }
 
 unsigned int ena_com_get_nonadaptive_moderation_interval_tx(struct ena_com_dev *ena_dev)
@@ -2625,47 +3198,34 @@ unsigned int ena_com_get_nonadaptive_moderation_interval_tx(struct ena_com_dev *
 
 unsigned int ena_com_get_nonadaptive_moderation_interval_rx(struct ena_com_dev *ena_dev)
 {
-	struct ena_intr_moder_entry *intr_moder_tbl = ena_dev->intr_moder_tbl;
-
-	if (intr_moder_tbl)
-		return intr_moder_tbl[ENA_INTR_MODER_LOWEST].intr_moder_interval;
-
-	return 0;
+	return ena_dev->intr_moder_rx_interval;
 }
 
-void ena_com_init_intr_moderation_entry(struct ena_com_dev *ena_dev,
-					enum ena_intr_moder_level level,
-					struct ena_intr_moder_entry *entry)
+int ena_com_config_dev_mode(struct ena_com_dev *ena_dev,
+			    struct ena_admin_feature_llq_desc *llq_features,
+			    struct ena_llq_configurations *llq_default_cfg)
 {
-	struct ena_intr_moder_entry *intr_moder_tbl = ena_dev->intr_moder_tbl;
+	struct ena_com_llq_info *llq_info = &ena_dev->llq_info;
+	int rc;
 
-	if (level >= ENA_INTR_MAX_NUM_OF_LEVELS)
-		return;
+	if (!llq_features->max_llq_num) {
+		ena_dev->tx_mem_queue_type = ENA_ADMIN_PLACEMENT_POLICY_HOST;
+		return 0;
+	}
 
-	intr_moder_tbl[level].intr_moder_interval = entry->intr_moder_interval;
-	if (ena_dev->intr_delay_resolution)
-		intr_moder_tbl[level].intr_moder_interval /=
-			ena_dev->intr_delay_resolution;
-	intr_moder_tbl[level].pkts_per_interval = entry->pkts_per_interval;
+	rc = ena_com_config_llq_info(ena_dev, llq_features, llq_default_cfg);
+	if (rc)
+		return rc;
 
-	/* use hardcoded value until ethtool supports bytecount parameter */
-	if (entry->bytes_per_interval != ENA_INTR_BYTE_COUNT_NOT_SUPPORTED)
-		intr_moder_tbl[level].bytes_per_interval = entry->bytes_per_interval;
-}
+	ena_dev->tx_max_header_size = llq_info->desc_list_entry_size -
+		(llq_info->descs_num_before_header * sizeof(struct ena_eth_io_tx_desc));
 
-void ena_com_get_intr_moderation_entry(struct ena_com_dev *ena_dev,
-				       enum ena_intr_moder_level level,
-				       struct ena_intr_moder_entry *entry)
-{
-	struct ena_intr_moder_entry *intr_moder_tbl = ena_dev->intr_moder_tbl;
+	if (unlikely(ena_dev->tx_max_header_size == 0)) {
+		netdev_err(ena_dev->net_device, "The size of the LLQ entry is smaller than needed\n");
+		return -EINVAL;
+	}
 
-	if (level >= ENA_INTR_MAX_NUM_OF_LEVELS)
-		return;
+	ena_dev->tx_mem_queue_type = ENA_ADMIN_PLACEMENT_POLICY_DEV;
 
-	entry->intr_moder_interval = intr_moder_tbl[level].intr_moder_interval;
-	if (ena_dev->intr_delay_resolution)
-		entry->intr_moder_interval *= ena_dev->intr_delay_resolution;
-	entry->pkts_per_interval =
-	intr_moder_tbl[level].pkts_per_interval;
-	entry->bytes_per_interval = intr_moder_tbl[level].bytes_per_interval;
+	return 0;
 }
diff --git a/drivers/net/ethernet/amazon/ena/ena_com.h b/drivers/net/ethernet/amazon/ena/ena_com.h
index c9b33ee5f258..64df2c48c9a6 100644
--- a/drivers/net/ethernet/amazon/ena/ena_com.h
+++ b/drivers/net/ethernet/amazon/ena/ena_com.h
@@ -1,33 +1,6 @@
+/* SPDX-License-Identifier: GPL-2.0 OR Linux-OpenIB */
 /*
- * Copyright 2015 Amazon.com, Inc. or its affiliates.
- *
- * This software is available to you under a choice of one of two
- * licenses.  You may choose to be licensed under the terms of the GNU
- * General Public License (GPL) Version 2, available from the file
- * COPYING in the main directory of this source tree, or the
- * BSD license below:
- *
- *     Redistribution and use in source and binary forms, with or
- *     without modification, are permitted provided that the following
- *     conditions are met:
- *
- *      - Redistributions of source code must retain the above
- *        copyright notice, this list of conditions and the following
- *        disclaimer.
- *
- *      - Redistributions in binary form must reproduce the above
- *        copyright notice, this list of conditions and the following
- *        disclaimer in the documentation and/or other materials
- *        provided with the distribution.
- *
- * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND,
- * EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF
- * MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND
- * NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS
- * BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN
- * ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN
- * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
- * SOFTWARE.
+ * Copyright 2015-2020 Amazon.com, Inc. or its affiliates. All rights reserved.
  */
 
 #ifndef ENA_COM
@@ -37,11 +10,14 @@
 #include <linux/delay.h>
 #include <linux/dma-mapping.h>
 #include <linux/gfp.h>
+#include <linux/io.h>
+#include <linux/prefetch.h>
 #include <linux/sched.h>
 #include <linux/sizes.h>
 #include <linux/spinlock.h>
 #include <linux/types.h>
 #include <linux/wait.h>
+#include <linux/netdevice.h>
 
 #include "ena_common_defs.h"
 #include "ena_admin_defs.h"
@@ -51,9 +27,9 @@
 #undef pr_fmt
 #define pr_fmt(fmt) KBUILD_MODNAME ": " fmt
 
-#define ENA_MAX_NUM_IO_QUEUES		128U
+#define ENA_MAX_NUM_IO_QUEUES 128U
 /* We need to queues for each IO (on for Tx and one for Rx) */
-#define ENA_TOTAL_NUM_QUEUES		(2 * (ENA_MAX_NUM_IO_QUEUES))
+#define ENA_TOTAL_NUM_QUEUES (2 * (ENA_MAX_NUM_IO_QUEUES))
 
 #define ENA_MAX_HANDLERS 256
 
@@ -66,50 +42,28 @@
 #define ADMIN_CQ_SIZE(depth)	((depth) * sizeof(struct ena_admin_acq_entry))
 #define ADMIN_AENQ_SIZE(depth)	((depth) * sizeof(struct ena_admin_aenq_entry))
 
+#define ENA_CUSTOMER_METRICS_BUFFER_SIZE 512
+
 /*****************************************************************************/
 /*****************************************************************************/
 /* ENA adaptive interrupt moderation settings */
 
-#define ENA_INTR_LOWEST_USECS           (0)
-#define ENA_INTR_LOWEST_PKTS            (3)
-#define ENA_INTR_LOWEST_BYTES           (2 * 1524)
-
-#define ENA_INTR_LOW_USECS              (32)
-#define ENA_INTR_LOW_PKTS               (12)
-#define ENA_INTR_LOW_BYTES              (16 * 1024)
-
-#define ENA_INTR_MID_USECS              (80)
-#define ENA_INTR_MID_PKTS               (48)
-#define ENA_INTR_MID_BYTES              (64 * 1024)
-
-#define ENA_INTR_HIGH_USECS             (128)
-#define ENA_INTR_HIGH_PKTS              (96)
-#define ENA_INTR_HIGH_BYTES             (128 * 1024)
-
-#define ENA_INTR_HIGHEST_USECS          (192)
-#define ENA_INTR_HIGHEST_PKTS           (128)
-#define ENA_INTR_HIGHEST_BYTES          (192 * 1024)
-
-#define ENA_INTR_INITIAL_TX_INTERVAL_USECS		196
-#define ENA_INTR_INITIAL_RX_INTERVAL_USECS		4
-#define ENA_INTR_DELAY_OLD_VALUE_WEIGHT			6
-#define ENA_INTR_DELAY_NEW_VALUE_WEIGHT			4
-#define ENA_INTR_MODER_LEVEL_STRIDE			2
-#define ENA_INTR_BYTE_COUNT_NOT_SUPPORTED		0xFFFFFF
-
-enum ena_intr_moder_level {
-	ENA_INTR_MODER_LOWEST = 0,
-	ENA_INTR_MODER_LOW,
-	ENA_INTR_MODER_MID,
-	ENA_INTR_MODER_HIGH,
-	ENA_INTR_MODER_HIGHEST,
-	ENA_INTR_MAX_NUM_OF_LEVELS,
-};
+#define ENA_INTR_INITIAL_TX_INTERVAL_USECS 64
+#define ENA_INTR_INITIAL_RX_INTERVAL_USECS 20
+#define ENA_DEFAULT_INTR_DELAY_RESOLUTION 1
+
+#define ENA_HASH_KEY_SIZE 40
+
+#define ENA_HW_HINTS_NO_TIMEOUT	0xFFFF
 
-struct ena_intr_moder_entry {
-	unsigned int intr_moder_interval;
-	unsigned int pkts_per_interval;
-	unsigned int bytes_per_interval;
+#define ENA_FEATURE_MAX_QUEUE_EXT_VER 1
+
+struct ena_llq_configurations {
+	enum ena_admin_llq_header_location llq_header_location;
+	enum ena_admin_llq_ring_entry_size llq_ring_entry_size;
+	enum ena_admin_llq_stride_ctrl  llq_stride_ctrl;
+	enum ena_admin_llq_num_descs_before_header llq_num_decs_before_header;
+	u16 llq_ring_entry_size_value;
 };
 
 enum queue_direction {
@@ -140,28 +94,36 @@ struct ena_com_tx_meta {
 	u16 l4_hdr_len; /* In words */
 };
 
+struct ena_com_llq_info {
+	u16 header_location_ctrl;
+	u16 desc_stride_ctrl;
+	u16 desc_list_entry_size_ctrl;
+	u16 desc_list_entry_size;
+	u16 descs_num_before_header;
+	u16 descs_per_entry;
+	u16 max_entries_in_tx_burst;
+	bool disable_meta_caching;
+};
+
 struct ena_com_io_cq {
 	struct ena_com_io_desc_addr cdesc_addr;
 
 	/* Interrupt unmask register */
 	u32 __iomem *unmask_reg;
 
-	/* The completion queue head doorbell register */
-	u32 __iomem *cq_head_db_reg;
-
 	/* numa configuration register (for TPH) */
 	u32 __iomem *numa_node_cfg_reg;
 
 	/* The value to write to the above register to unmask
 	 * the interrupt of this queue
 	 */
-	u32 msix_vector;
+	u32 msix_vector ____cacheline_aligned;
 
 	enum queue_direction direction;
 
 	/* holds the number of cdesc of the current packet */
 	u16 cur_rx_pkt_cdesc_count;
-	/* save the firt cdesc idx of the current packet */
+	/* save the first cdesc idx of the current packet */
 	u16 cur_rx_pkt_cdesc_start_idx;
 
 	u16 q_depth;
@@ -171,23 +133,40 @@ struct ena_com_io_cq {
 	/* Device queue index */
 	u16 idx;
 	u16 head;
-	u16 last_head_update;
 	u8 phase;
 	u8 cdesc_entry_size_in_bytes;
 
 } ____cacheline_aligned;
 
+struct ena_com_io_bounce_buffer_control {
+	u8 *base_buffer;
+	u16 next_to_use;
+	u16 buffer_size;
+	u16 buffers_num;  /* Must be a power of 2 */
+};
+
+/* This struct is to keep tracking the current location of the next llq entry */
+struct ena_com_llq_pkt_ctrl {
+	u8 *curr_bounce_buf;
+	u16 idx;
+	u16 descs_left_in_line;
+};
+
 struct ena_com_io_sq {
 	struct ena_com_io_desc_addr desc_addr;
 
 	u32 __iomem *db_addr;
-	u8 __iomem *header_addr;
 
 	enum queue_direction direction;
 	enum ena_admin_placement_policy_type mem_queue_type;
 
+	bool disable_meta_caching;
+
 	u32 msix_vector;
 	struct ena_com_tx_meta cached_tx_meta;
+	struct ena_com_llq_info llq_info;
+	struct ena_com_llq_pkt_ctrl llq_buf_ctrl;
+	struct ena_com_io_bounce_buffer_control bounce_buf_ctrl;
 
 	u16 q_depth;
 	u16 qid;
@@ -195,10 +174,12 @@ struct ena_com_io_sq {
 	u16 idx;
 	u16 tail;
 	u16 next_to_comp;
+	u16 llq_last_copy_tail;
 	u32 tx_max_header_size;
 	u8 phase;
 	u8 desc_entry_size;
 	u8 dma_addr_bits;
+	u16 entries_in_tx_burst_left;
 } ____cacheline_aligned;
 
 struct ena_com_admin_cq {
@@ -222,17 +203,28 @@ struct ena_com_admin_sq {
 };
 
 struct ena_com_stats_admin {
-	u32 aborted_cmd;
-	u32 submitted_cmd;
-	u32 completed_cmd;
-	u32 out_of_space;
-	u32 no_completion;
+	u64 aborted_cmd;
+	u64 submitted_cmd;
+	u64 completed_cmd;
+	u64 out_of_space;
+	u64 no_completion;
+};
+
+struct ena_com_stats_phc {
+	u64 phc_cnt;
+	u64 phc_exp;
+	u64 phc_skp;
+	u64 phc_err_dv;
+	u64 phc_err_ts;
 };
 
 struct ena_com_admin_queue {
 	void *q_dmadev;
+	struct ena_com_dev *ena_dev;
 	spinlock_t q_lock; /* spinlock for the admin queue */
+
 	struct ena_comp_ctx *comp_ctx;
+	u32 completion_timeout;
 	u16 q_depth;
 	struct ena_com_admin_cq cq;
 	struct ena_com_admin_sq sq;
@@ -267,12 +259,54 @@ struct ena_com_aenq {
 struct ena_com_mmio_read {
 	struct ena_admin_ena_mmio_req_read_less_resp *read_resp;
 	dma_addr_t read_resp_dma_addr;
+	u32 reg_read_to; /* in us */
 	u16 seq_num;
 	bool readless_supported;
 	/* spin lock to ensure a single outstanding read */
 	spinlock_t lock;
 };
 
+/* PTP hardware clock (PHC) MMIO read data info */
+struct ena_com_phc_info {
+	/* Internal PHC statistics */
+	struct ena_com_stats_phc stats;
+
+	/* PHC shared memory - virtual address */
+	struct ena_admin_phc_resp *virt_addr;
+
+	/* System time of last PHC request */
+	ktime_t system_time;
+
+	/* Spin lock to ensure a single outstanding PHC read */
+	spinlock_t lock;
+
+	/* PHC doorbell address as an offset to PCIe MMIO REG BAR */
+	u32 doorbell_offset;
+
+	/* Shared memory read expire timeout (usec)
+	 * Max time for valid PHC retrieval, passing this threshold will fail
+	 * the get time request and block new PHC requests for block_timeout_usec
+	 * in order to prevent floods on busy device
+	 */
+	u32 expire_timeout_usec;
+
+	/* Shared memory read abort timeout (usec)
+	 * PHC requests block period, blocking starts once PHC request expired
+	 * in order to prevent floods on busy device,
+	 * any PHC requests during block period will be skipped
+	 */
+	u32 block_timeout_usec;
+
+	/* PHC shared memory - physical address */
+	dma_addr_t phys_addr;
+
+	/* Request id sent to the device */
+	u16 req_id;
+
+	/* True if PHC is active in the device */
+	bool active;
+};
+
 struct ena_rss {
 	/* Indirect table */
 	u16 *host_rss_ind_tbl;
@@ -292,6 +326,16 @@ struct ena_rss {
 
 };
 
+struct ena_customer_metrics {
+	/* in correlation with ENA_ADMIN_CUSTOMER_METRICS_SUPPORT_MASK
+	 * and ena_admin_customer_metrics_id
+	 */
+	u64 supported_metrics;
+	dma_addr_t buffer_dma_addr;
+	void *buffer_virt_addr;
+	u32 buffer_len;
+};
+
 struct ena_host_attribute {
 	/* Debug area */
 	u8 *debug_area_virt_addr;
@@ -312,30 +356,48 @@ struct ena_com_dev {
 	u8 __iomem *reg_bar;
 	void __iomem *mem_bar;
 	void *dmadev;
+	struct net_device *net_device;
 
 	enum ena_admin_placement_policy_type tx_mem_queue_type;
 	u32 tx_max_header_size;
 	u16 stats_func; /* Selected function for extended statistic dump */
 	u16 stats_queue; /* Selected queue for extended statistic dump */
 
+	u32 ena_min_poll_delay_us;
+
 	struct ena_com_mmio_read mmio_read;
+	struct ena_com_phc_info phc;
 
 	struct ena_rss rss;
 	u32 supported_features;
+	u32 capabilities;
 	u32 dma_addr_bits;
 
 	struct ena_host_attribute host_attr;
 	bool adaptive_coalescing;
 	u16 intr_delay_resolution;
+
+	/* interrupt moderation intervals are in usec divided by
+	 * intr_delay_resolution, which is supplied by the device.
+	 */
 	u32 intr_moder_tx_interval;
+	u32 intr_moder_rx_interval;
+
 	struct ena_intr_moder_entry *intr_moder_tbl;
+
+	struct ena_com_llq_info llq_info;
+
+	struct ena_customer_metrics customer_metrics;
 };
 
 struct ena_com_dev_get_features_ctx {
 	struct ena_admin_queue_feature_desc max_queues;
+	struct ena_admin_queue_ext_feature_desc max_queue_ext;
 	struct ena_admin_device_attr_feature_desc dev_attr;
 	struct ena_admin_feature_aenq_desc aenq;
 	struct ena_admin_feature_offload_desc offload;
+	struct ena_admin_ena_hw_hints hw_hints;
+	struct ena_admin_feature_llq_desc llq;
 };
 
 struct ena_com_create_io_ctx {
@@ -370,7 +432,41 @@ struct ena_aenq_handlers {
  */
 int ena_com_mmio_reg_read_request_init(struct ena_com_dev *ena_dev);
 
-/* ena_com_set_mmio_read_mode - Enable/disable the mmio reg read mechanism
+/* ena_com_phc_init - Allocate and initialize PHC feature
+ * @ena_dev: ENA communication layer struct
+ * @note: This method assumes PHC is supported by the device
+ * @return - 0 on success, negative value on failure
+ */
+int ena_com_phc_init(struct ena_com_dev *ena_dev);
+
+/* ena_com_phc_supported - Return if PHC feature is supported by the device
+ * @ena_dev: ENA communication layer struct
+ * @note: This method must be called after getting supported features
+ * @return - supported or not
+ */
+bool ena_com_phc_supported(struct ena_com_dev *ena_dev);
+
+/* ena_com_phc_config - Configure PHC feature
+ * @ena_dev: ENA communication layer struct
+ * Configure PHC feature in driver and device
+ * @note: This method assumes PHC is supported by the device
+ * @return - 0 on success, negative value on failure
+ */
+int ena_com_phc_config(struct ena_com_dev *ena_dev);
+
+/* ena_com_phc_destroy - Destroy PHC feature
+ * @ena_dev: ENA communication layer struct
+ */
+void ena_com_phc_destroy(struct ena_com_dev *ena_dev);
+
+/* ena_com_phc_get_timestamp - Retrieve PHC timestamp
+ * @ena_dev: ENA communication layer struct
+ * @timestamp: Retrieved PHC timestamp
+ * @return - 0 on success, negative value on failure
+ */
+int ena_com_phc_get_timestamp(struct ena_com_dev *ena_dev, u64 *timestamp);
+
+/* ena_com_set_mmio_read_mode - Enable/disable the indirect mmio reg read mechanism
  * @ena_dev: ENA communication layer struct
  * @readless_supported: readless mode (enable/disable)
  */
@@ -391,8 +487,6 @@ void ena_com_mmio_reg_read_request_destroy(struct ena_com_dev *ena_dev);
 /* ena_com_admin_init - Init the admin and the async queues
  * @ena_dev: ENA communication layer struct
  * @aenq_handlers: Those handlers to be called upon event.
- * @init_spinlock: Indicate if this method should init the admin spinlock or
- * the spinlock was init before (for example, in a case of FLR).
  *
  * Initialize the admin submission and completion queues.
  * Initialize the asynchronous events notification queues.
@@ -400,8 +494,7 @@ void ena_com_mmio_reg_read_request_destroy(struct ena_com_dev *ena_dev);
  * @return - 0 on success, negative value on failure.
  */
 int ena_com_admin_init(struct ena_com_dev *ena_dev,
-		       struct ena_aenq_handlers *aenq_handlers,
-		       bool init_spinlock);
+		       struct ena_aenq_handlers *aenq_handlers);
 
 /* ena_com_admin_destroy - Destroy the admin and the async events queues.
  * @ena_dev: ENA communication layer struct
@@ -414,10 +507,12 @@ void ena_com_admin_destroy(struct ena_com_dev *ena_dev);
 
 /* ena_com_dev_reset - Perform device FLR to the device.
  * @ena_dev: ENA communication layer struct
+ * @reset_reason: Specify what is the trigger for the reset in case of an error.
  *
  * @return - 0 on success, negative value on failure.
  */
-int ena_com_dev_reset(struct ena_com_dev *ena_dev);
+int ena_com_dev_reset(struct ena_com_dev *ena_dev,
+		      enum ena_regs_reset_reason_types reset_reason);
 
 /* ena_com_create_io_queue - Create io queue.
  * @ena_dev: ENA communication layer struct
@@ -479,22 +574,10 @@ bool ena_com_get_admin_running_state(struct ena_com_dev *ena_dev);
  */
 void ena_com_set_admin_polling_mode(struct ena_com_dev *ena_dev, bool polling);
 
-/* ena_com_set_admin_polling_mode - Get the admin completion queue polling mode
- * @ena_dev: ENA communication layer struct
- *
- * Get the admin completion mode.
- * If polling mode is on, ena_com_execute_admin_command will perform a
- * polling on the admin completion queue for the commands completion,
- * otherwise it will wait on wait event.
- *
- * @return state
- */
-bool ena_com_get_ena_admin_polling_mode(struct ena_com_dev *ena_dev);
-
 /* ena_com_admin_q_comp_intr_handler - admin queue interrupt handler
  * @ena_dev: ENA communication layer struct
  *
- * This method go over the admin completion queue and wake up all the pending
+ * This method goes over the admin completion queue and wakes up all the pending
  * threads that wait on the commands wait event.
  *
  * @note: Should be called after MSI-X interrupt.
@@ -504,10 +587,10 @@ void ena_com_admin_q_comp_intr_handler(struct ena_com_dev *ena_dev);
 /* ena_com_aenq_intr_handler - AENQ interrupt handler
  * @ena_dev: ENA communication layer struct
  *
- * This method go over the async event notification queue and call the proper
+ * This method goes over the async event notification queue and calls the proper
  * aenq handler.
  */
-void ena_com_aenq_intr_handler(struct ena_com_dev *dev, void *data);
+void ena_com_aenq_intr_handler(struct ena_com_dev *ena_dev, void *data);
 
 /* ena_com_abort_admin_commands - Abort all the outstanding admin commands.
  * @ena_dev: ENA communication layer struct
@@ -521,14 +604,14 @@ void ena_com_abort_admin_commands(struct ena_com_dev *ena_dev);
 /* ena_com_wait_for_abort_completion - Wait for admin commands abort.
  * @ena_dev: ENA communication layer struct
  *
- * This method wait until all the outstanding admin commands will be completed.
+ * This method waits until all the outstanding admin commands are completed.
  */
 void ena_com_wait_for_abort_completion(struct ena_com_dev *ena_dev);
 
 /* ena_com_validate_version - Validate the device parameters
  * @ena_dev: ENA communication layer struct
  *
- * This method validate the device parameters are the same as the saved
+ * This method verifies the device parameters are the same as the saved
  * parameters in ena_dev.
  * This method is useful after device reset, to validate the device mac address
  * and the device offloads are the same as before the reset.
@@ -578,31 +661,40 @@ int ena_com_set_aenq_config(struct ena_com_dev *ena_dev, u32 groups_flag);
 int ena_com_get_dev_attr_feat(struct ena_com_dev *ena_dev,
 			      struct ena_com_dev_get_features_ctx *get_feat_ctx);
 
-/* ena_com_get_dev_basic_stats - Get device basic statistics
+/* ena_com_get_eni_stats - Get extended network interface statistics
  * @ena_dev: ENA communication layer struct
  * @stats: stats return value
  *
  * @return: 0 on Success and negative value otherwise.
  */
-int ena_com_get_dev_basic_stats(struct ena_com_dev *ena_dev,
-				struct ena_admin_basic_stats *stats);
+int ena_com_get_eni_stats(struct ena_com_dev *ena_dev,
+			  struct ena_admin_eni_stats *stats);
 
-/* ena_com_set_dev_mtu - Configure the device mtu.
+/* ena_com_get_ena_srd_info - Get ENA SRD network interface statistics
  * @ena_dev: ENA communication layer struct
- * @mtu: mtu value
+ * @info: ena srd stats and flags
  *
  * @return: 0 on Success and negative value otherwise.
  */
-int ena_com_set_dev_mtu(struct ena_com_dev *ena_dev, int mtu);
+int ena_com_get_ena_srd_info(struct ena_com_dev *ena_dev,
+			     struct ena_admin_ena_srd_info *info);
 
-/* ena_com_get_offload_settings - Retrieve the device offloads capabilities
+/* ena_com_get_customer_metrics - Get customer metrics for network interface
  * @ena_dev: ENA communication layer struct
- * @offlad: offload return value
+ * @buffer: buffer for returned customer metrics
+ * @len: size of the buffer
  *
  * @return: 0 on Success and negative value otherwise.
  */
-int ena_com_get_offload_settings(struct ena_com_dev *ena_dev,
-				 struct ena_admin_feature_offload_desc *offload);
+int ena_com_get_customer_metrics(struct ena_com_dev *ena_dev, char *buffer, u32 len);
+
+/* ena_com_set_dev_mtu - Configure the device mtu.
+ * @ena_dev: ENA communication layer struct
+ * @mtu: mtu value
+ *
+ * @return: 0 on Success and negative value otherwise.
+ */
+int ena_com_set_dev_mtu(struct ena_com_dev *ena_dev, u32 mtu);
 
 /* ena_com_rss_init - Init RSS
  * @ena_dev: ENA communication layer struct
@@ -623,6 +715,14 @@ int ena_com_rss_init(struct ena_com_dev *ena_dev, u16 log_size);
  */
 void ena_com_rss_destroy(struct ena_com_dev *ena_dev);
 
+/* ena_com_get_current_hash_function - Get RSS hash function
+ * @ena_dev: ENA communication layer struct
+ *
+ * Return the current hash function.
+ * @return: 0 or one of the ena_admin_hash_functions values.
+ */
+int ena_com_get_current_hash_function(struct ena_com_dev *ena_dev);
+
 /* ena_com_fill_hash_function - Fill RSS hash function
  * @ena_dev: ENA communication layer struct
  * @func: The hash function (Toeplitz or crc)
@@ -654,23 +754,32 @@ int ena_com_fill_hash_function(struct ena_com_dev *ena_dev,
  */
 int ena_com_set_hash_function(struct ena_com_dev *ena_dev);
 
-/* ena_com_get_hash_function - Retrieve the hash function and the hash key
- * from the device.
+/* ena_com_get_hash_function - Retrieve the hash function from the device.
  * @ena_dev: ENA communication layer struct
  * @func: hash function
- * @key: hash key
  *
- * Retrieve the hash function and the hash key from the device.
+ * Retrieve the hash function from the device.
  *
- * @note: If the caller called ena_com_fill_hash_function but didn't flash
+ * @note: If the caller called ena_com_fill_hash_function but didn't flush
  * it to the device, the new configuration will be lost.
  *
  * @return: 0 on Success and negative value otherwise.
  */
 int ena_com_get_hash_function(struct ena_com_dev *ena_dev,
-			      enum ena_admin_hash_functions *func,
-			      u8 *key);
+			      enum ena_admin_hash_functions *func);
 
+/* ena_com_get_hash_key - Retrieve the hash key
+ * @ena_dev: ENA communication layer struct
+ * @key: hash key
+ *
+ * Retrieve the hash key.
+ *
+ * @note: If the caller called ena_com_fill_hash_key but didn't flush
+ * it to the device, the new configuration will be lost.
+ *
+ * @return: 0 on Success and negative value otherwise.
+ */
+int ena_com_get_hash_key(struct ena_com_dev *ena_dev, u8 *key);
 /* ena_com_fill_hash_ctrl - Fill RSS hash control
  * @ena_dev: ENA communication layer struct.
  * @proto: The protocol to configure.
@@ -705,7 +814,7 @@ int ena_com_set_hash_ctrl(struct ena_com_dev *ena_dev);
  *
  * Retrieve the hash control from the device.
  *
- * @note, If the caller called ena_com_fill_hash_ctrl but didn't flash
+ * @note: If the caller called ena_com_fill_hash_ctrl but didn't flush
  * it to the device, the new configuration will be lost.
  *
  * @return: 0 on Success and negative value otherwise.
@@ -757,7 +866,7 @@ int ena_com_indirect_table_set(struct ena_com_dev *ena_dev);
  *
  * Retrieve the RSS indirection table from the device.
  *
- * @note: If the caller called ena_com_indirect_table_fill_entry but didn't flash
+ * @note: If the caller called ena_com_indirect_table_fill_entry but didn't flush
  * it to the device, the new configuration will be lost.
  *
  * @return: 0 on Success and negative value otherwise.
@@ -780,20 +889,34 @@ int ena_com_allocate_host_info(struct ena_com_dev *ena_dev);
 int ena_com_allocate_debug_area(struct ena_com_dev *ena_dev,
 				u32 debug_area_size);
 
+/* ena_com_allocate_customer_metrics_buffer - Allocate customer metrics resources.
+ * @ena_dev: ENA communication layer struct
+ *
+ * @return: 0 on Success and negative value otherwise.
+ */
+int ena_com_allocate_customer_metrics_buffer(struct ena_com_dev *ena_dev);
+
 /* ena_com_delete_debug_area - Free the debug area resources.
  * @ena_dev: ENA communication layer struct
  *
- * Free the allocate debug area.
+ * Free the allocated debug area.
  */
 void ena_com_delete_debug_area(struct ena_com_dev *ena_dev);
 
 /* ena_com_delete_host_info - Free the host info resources.
  * @ena_dev: ENA communication layer struct
  *
- * Free the allocate host info.
+ * Free the allocated host info.
  */
 void ena_com_delete_host_info(struct ena_com_dev *ena_dev);
 
+/* ena_com_delete_customer_metrics_buffer - Free the customer metrics resources.
+ * @ena_dev: ENA communication layer struct
+ *
+ * Free the allocated customer metrics area.
+ */
+void ena_com_delete_customer_metrics_buffer(struct ena_com_dev *ena_dev);
+
 /* ena_com_set_host_attributes - Update the device with the host
  * attributes (debug area and host info) base address.
  * @ena_dev: ENA communication layer struct
@@ -831,9 +954,9 @@ int ena_com_destroy_io_cq(struct ena_com_dev *ena_dev,
  * @cmd_completion: command completion return value.
  * @cmd_comp_size: command completion size.
 
- * Submit an admin command and then wait until the device will return a
+ * Submit an admin command and then wait until the device returns a
  * completion.
- * The completion will be copyed into cmd_comp.
+ * The completion will be copied into cmd_comp.
  *
  * @return - 0 on success, negative value on failure.
  */
@@ -850,11 +973,6 @@ int ena_com_execute_admin_command(struct ena_com_admin_queue *admin_queue,
  */
 int ena_com_init_interrupt_moderation(struct ena_com_dev *ena_dev);
 
-/* ena_com_destroy_interrupt_moderation - Destroy interrupt moderation resources
- * @ena_dev: ENA communication layer struct
- */
-void ena_com_destroy_interrupt_moderation(struct ena_com_dev *ena_dev);
-
 /* ena_com_interrupt_moderation_supported - Return if interrupt moderation
  * capability is supported by the device.
  *
@@ -862,12 +980,6 @@ void ena_com_destroy_interrupt_moderation(struct ena_com_dev *ena_dev);
  */
 bool ena_com_interrupt_moderation_supported(struct ena_com_dev *ena_dev);
 
-/* ena_com_config_default_interrupt_moderation_table - Restore the interrupt
- * moderation table back to the default parameters.
- * @ena_dev: ENA communication layer struct
- */
-void ena_com_config_default_interrupt_moderation_table(struct ena_com_dev *ena_dev);
-
 /* ena_com_update_nonadaptive_moderation_interval_tx - Update the
  * non-adaptive interval in Tx direction.
  * @ena_dev: ENA communication layer struct
@@ -904,28 +1016,35 @@ unsigned int ena_com_get_nonadaptive_moderation_interval_tx(struct ena_com_dev *
  */
 unsigned int ena_com_get_nonadaptive_moderation_interval_rx(struct ena_com_dev *ena_dev);
 
-/* ena_com_init_intr_moderation_entry - Update a single entry in the interrupt
- * moderation table.
+/* ena_com_config_dev_mode - Configure the placement policy of the device.
  * @ena_dev: ENA communication layer struct
- * @level: Interrupt moderation table level
- * @entry: Entry value
+ * @llq_features: LLQ feature descriptor, retrieve via
+ *		   ena_com_get_dev_attr_feat.
+ * @ena_llq_config: The default driver LLQ parameters configurations
+ */
+int ena_com_config_dev_mode(struct ena_com_dev *ena_dev,
+			    struct ena_admin_feature_llq_desc *llq_features,
+			    struct ena_llq_configurations *llq_default_config);
+
+/* ena_com_io_sq_to_ena_dev - Extract ena_com_dev using contained field io_sq.
+ * @io_sq: IO submit queue struct
  *
- * Update a single entry in the interrupt moderation table.
+ * @return - ena_com_dev struct extracted from io_sq
  */
-void ena_com_init_intr_moderation_entry(struct ena_com_dev *ena_dev,
-					enum ena_intr_moder_level level,
-					struct ena_intr_moder_entry *entry);
+static inline struct ena_com_dev *ena_com_io_sq_to_ena_dev(struct ena_com_io_sq *io_sq)
+{
+	return container_of(io_sq, struct ena_com_dev, io_sq_queues[io_sq->qid]);
+}
 
-/* ena_com_get_intr_moderation_entry - Init ena_intr_moder_entry.
- * @ena_dev: ENA communication layer struct
- * @level: Interrupt moderation table level
- * @entry: Entry to fill.
+/* ena_com_io_cq_to_ena_dev - Extract ena_com_dev using contained field io_cq.
+ * @io_sq: IO submit queue struct
  *
- * Initialize the entry according to the adaptive interrupt moderation table.
+ * @return - ena_com_dev struct extracted from io_sq
  */
-void ena_com_get_intr_moderation_entry(struct ena_com_dev *ena_dev,
-				       enum ena_intr_moder_level level,
-				       struct ena_intr_moder_entry *entry);
+static inline struct ena_com_dev *ena_com_io_cq_to_ena_dev(struct ena_com_io_cq *io_cq)
+{
+	return container_of(io_cq, struct ena_com_dev, io_cq_queues[io_cq->qid]);
+}
 
 static inline bool ena_com_get_adaptive_moderation_enabled(struct ena_com_dev *ena_dev)
 {
@@ -942,80 +1061,45 @@ static inline void ena_com_disable_adaptive_moderation(struct ena_com_dev *ena_d
 	ena_dev->adaptive_coalescing = false;
 }
 
-/* ena_com_calculate_interrupt_delay - Calculate new interrupt delay
+/* ena_com_get_cap - query whether device supports a capability.
  * @ena_dev: ENA communication layer struct
- * @pkts: Number of packets since the last update
- * @bytes: Number of bytes received since the last update.
- * @smoothed_interval: Returned interval
- * @moder_tbl_idx: Current table level as input update new level as return
- * value.
- */
-static inline void ena_com_calculate_interrupt_delay(struct ena_com_dev *ena_dev,
-						     unsigned int pkts,
-						     unsigned int bytes,
-						     unsigned int *smoothed_interval,
-						     unsigned int *moder_tbl_idx)
+ * @cap_id: enum value representing the capability
+ *
+ * @return - true if capability is supported or false otherwise
+ */
+static inline bool ena_com_get_cap(struct ena_com_dev *ena_dev,
+				   enum ena_admin_aq_caps_id cap_id)
 {
-	enum ena_intr_moder_level curr_moder_idx, new_moder_idx;
-	struct ena_intr_moder_entry *curr_moder_entry;
-	struct ena_intr_moder_entry *pred_moder_entry;
-	struct ena_intr_moder_entry *new_moder_entry;
-	struct ena_intr_moder_entry *intr_moder_tbl = ena_dev->intr_moder_tbl;
-	unsigned int interval;
-
-	/* We apply adaptive moderation on Rx path only.
-	 * Tx uses static interrupt moderation.
-	 */
-	if (!pkts || !bytes)
-		/* Tx interrupt, or spurious interrupt,
-		 * in both cases we just use same delay values
-		 */
-		return;
-
-	curr_moder_idx = (enum ena_intr_moder_level)(*moder_tbl_idx);
-	if (unlikely(curr_moder_idx >= ENA_INTR_MAX_NUM_OF_LEVELS)) {
-		pr_err("Wrong moderation index %u\n", curr_moder_idx);
-		return;
-	}
-
-	curr_moder_entry = &intr_moder_tbl[curr_moder_idx];
-	new_moder_idx = curr_moder_idx;
-
-	if (curr_moder_idx == ENA_INTR_MODER_LOWEST) {
-		if ((pkts > curr_moder_entry->pkts_per_interval) ||
-		    (bytes > curr_moder_entry->bytes_per_interval))
-			new_moder_idx =
-				(enum ena_intr_moder_level)(curr_moder_idx + ENA_INTR_MODER_LEVEL_STRIDE);
-	} else {
-		pred_moder_entry = &intr_moder_tbl[curr_moder_idx - ENA_INTR_MODER_LEVEL_STRIDE];
-
-		if ((pkts <= pred_moder_entry->pkts_per_interval) ||
-		    (bytes <= pred_moder_entry->bytes_per_interval))
-			new_moder_idx =
-				(enum ena_intr_moder_level)(curr_moder_idx - ENA_INTR_MODER_LEVEL_STRIDE);
-		else if ((pkts > curr_moder_entry->pkts_per_interval) ||
-			 (bytes > curr_moder_entry->bytes_per_interval)) {
-			if (curr_moder_idx != ENA_INTR_MODER_HIGHEST)
-				new_moder_idx =
-					(enum ena_intr_moder_level)(curr_moder_idx + ENA_INTR_MODER_LEVEL_STRIDE);
-		}
-	}
-	new_moder_entry = &intr_moder_tbl[new_moder_idx];
-
-	interval = new_moder_entry->intr_moder_interval;
-	*smoothed_interval = (
-		(interval * ENA_INTR_DELAY_NEW_VALUE_WEIGHT +
-		ENA_INTR_DELAY_OLD_VALUE_WEIGHT * (*smoothed_interval)) + 5) /
-		10;
-
-	*moder_tbl_idx = new_moder_idx;
+	return !!(ena_dev->capabilities & BIT(cap_id));
+}
+
+/* ena_com_get_customer_metric_support - query whether device supports a given customer metric.
+ * @ena_dev: ENA communication layer struct
+ * @metric_id: enum value representing the customer metric
+ *
+ * @return - true if customer metric is supported or false otherwise
+ */
+static inline bool ena_com_get_customer_metric_support(struct ena_com_dev *ena_dev,
+						       enum ena_admin_customer_metrics_id metric_id)
+{
+	return !!(ena_dev->customer_metrics.supported_metrics & BIT(metric_id));
+}
+
+/* ena_com_get_customer_metric_count - return the number of supported customer metrics.
+ * @ena_dev: ENA communication layer struct
+ *
+ * @return - the number of supported customer metrics
+ */
+static inline int ena_com_get_customer_metric_count(struct ena_com_dev *ena_dev)
+{
+	return hweight64(ena_dev->customer_metrics.supported_metrics);
 }
 
 /* ena_com_update_intr_reg - Prepare interrupt register
  * @intr_reg: interrupt register to update.
  * @rx_delay_interval: Rx interval in usecs
  * @tx_delay_interval: Tx interval in usecs
- * @unmask: unask enable/disable
+ * @unmask: unmask enable/disable
  *
  * Prepare interrupt update register with the supplied parameters.
  */
@@ -1036,4 +1120,21 @@ static inline void ena_com_update_intr_reg(struct ena_eth_io_intr_reg *intr_reg,
 		intr_reg->intr_control |= ENA_ETH_IO_INTR_REG_INTR_UNMASK_MASK;
 }
 
+static inline u8 *ena_com_get_next_bounce_buffer(struct ena_com_io_bounce_buffer_control *bounce_buf_ctrl)
+{
+	u16 size, buffers_num;
+	u8 *buf;
+
+	size = bounce_buf_ctrl->buffer_size;
+	buffers_num = bounce_buf_ctrl->buffers_num;
+
+	buf = bounce_buf_ctrl->base_buffer +
+		(bounce_buf_ctrl->next_to_use++ & (buffers_num - 1)) * size;
+
+	prefetchw(bounce_buf_ctrl->base_buffer +
+		(bounce_buf_ctrl->next_to_use & (buffers_num - 1)) * size);
+
+	return buf;
+}
+
 #endif /* !(ENA_COM) */
diff --git a/drivers/net/ethernet/amazon/ena/ena_common_defs.h b/drivers/net/ethernet/amazon/ena/ena_common_defs.h
index bb8d73676eab..e210c8a81fc0 100644
--- a/drivers/net/ethernet/amazon/ena/ena_common_defs.h
+++ b/drivers/net/ethernet/amazon/ena/ena_common_defs.h
@@ -1,39 +1,12 @@
+/* SPDX-License-Identifier: GPL-2.0 OR Linux-OpenIB */
 /*
- * Copyright 2015 - 2016 Amazon.com, Inc. or its affiliates.
- *
- * This software is available to you under a choice of one of two
- * licenses.  You may choose to be licensed under the terms of the GNU
- * General Public License (GPL) Version 2, available from the file
- * COPYING in the main directory of this source tree, or the
- * BSD license below:
- *
- *     Redistribution and use in source and binary forms, with or
- *     without modification, are permitted provided that the following
- *     conditions are met:
- *
- *      - Redistributions of source code must retain the above
- *        copyright notice, this list of conditions and the following
- *        disclaimer.
- *
- *      - Redistributions in binary form must reproduce the above
- *        copyright notice, this list of conditions and the following
- *        disclaimer in the documentation and/or other materials
- *        provided with the distribution.
- *
- * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND,
- * EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF
- * MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND
- * NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS
- * BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN
- * ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN
- * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
- * SOFTWARE.
+ * Copyright 2015-2020 Amazon.com, Inc. or its affiliates. All rights reserved.
  */
 #ifndef _ENA_COMMON_H_
 #define _ENA_COMMON_H_
 
-#define ENA_COMMON_SPEC_VERSION_MAJOR	0 /*  */
-#define ENA_COMMON_SPEC_VERSION_MINOR	10 /*  */
+#define ENA_COMMON_SPEC_VERSION_MAJOR        2
+#define ENA_COMMON_SPEC_VERSION_MINOR        0
 
 /* ENA operates with 48-bit memory addresses. ena_mem_addr_t */
 struct ena_common_mem_addr {
@@ -45,4 +18,4 @@ struct ena_common_mem_addr {
 	u16 reserved16;
 };
 
-#endif /*_ENA_COMMON_H_ */
+#endif /* _ENA_COMMON_H_ */
diff --git a/drivers/net/ethernet/amazon/ena/ena_debugfs.c b/drivers/net/ethernet/amazon/ena/ena_debugfs.c
new file mode 100644
index 000000000000..46ed80986724
--- /dev/null
+++ b/drivers/net/ethernet/amazon/ena/ena_debugfs.c
@@ -0,0 +1,62 @@
+// SPDX-License-Identifier: GPL-2.0 OR Linux-OpenIB
+/* Copyright (c) Amazon.com, Inc. or its affiliates.
+ * All rights reserved.
+ */
+
+#ifdef CONFIG_DEBUG_FS
+
+#include <linux/seq_file.h>
+#include <linux/pci.h>
+#include "ena_debugfs.h"
+#include "ena_phc.h"
+
+static int phc_stats_show(struct seq_file *file, void *priv)
+{
+	struct ena_adapter *adapter = file->private;
+
+	if (!ena_phc_is_active(adapter))
+		return 0;
+
+	seq_printf(file,
+		   "phc_cnt: %llu\n",
+		   adapter->ena_dev->phc.stats.phc_cnt);
+	seq_printf(file,
+		   "phc_exp: %llu\n",
+		   adapter->ena_dev->phc.stats.phc_exp);
+	seq_printf(file,
+		   "phc_skp: %llu\n",
+		   adapter->ena_dev->phc.stats.phc_skp);
+	seq_printf(file,
+		   "phc_err_dv: %llu\n",
+		   adapter->ena_dev->phc.stats.phc_err_dv);
+	seq_printf(file,
+		   "phc_err_ts: %llu\n",
+		   adapter->ena_dev->phc.stats.phc_err_ts);
+
+	return 0;
+}
+
+DEFINE_SHOW_ATTRIBUTE(phc_stats);
+
+void ena_debugfs_init(struct net_device *dev)
+{
+	struct ena_adapter *adapter = netdev_priv(dev);
+
+	adapter->debugfs_base =
+		debugfs_create_dir(dev_name(&adapter->pdev->dev), NULL);
+
+	debugfs_create_file("phc_stats",
+			    0400,
+			    adapter->debugfs_base,
+			    adapter,
+			    &phc_stats_fops);
+}
+
+void ena_debugfs_terminate(struct net_device *dev)
+{
+	struct ena_adapter *adapter = netdev_priv(dev);
+
+	debugfs_remove_recursive(adapter->debugfs_base);
+}
+
+#endif /* CONFIG_DEBUG_FS */
diff --git a/drivers/net/ethernet/amazon/ena/ena_debugfs.h b/drivers/net/ethernet/amazon/ena/ena_debugfs.h
new file mode 100644
index 000000000000..dc61dd998867
--- /dev/null
+++ b/drivers/net/ethernet/amazon/ena/ena_debugfs.h
@@ -0,0 +1,27 @@
+/* SPDX-License-Identifier: GPL-2.0 OR Linux-OpenIB */
+/* Copyright (c) Amazon.com, Inc. or its affiliates.
+ * All rights reserved.
+ */
+
+#ifndef __ENA_DEBUGFS_H__
+#define __ENA_DEBUGFS_H__
+
+#include <linux/debugfs.h>
+#include <linux/netdevice.h>
+#include "ena_netdev.h"
+
+#ifdef CONFIG_DEBUG_FS
+
+void ena_debugfs_init(struct net_device *dev);
+
+void ena_debugfs_terminate(struct net_device *dev);
+
+#else /* CONFIG_DEBUG_FS */
+
+static inline void ena_debugfs_init(struct net_device *dev) {}
+
+static inline void ena_debugfs_terminate(struct net_device *dev) {}
+
+#endif /* CONFIG_DEBUG_FS */
+
+#endif /* __ENA_DEBUGFS_H__ */
diff --git a/drivers/net/ethernet/amazon/ena/ena_devlink.c b/drivers/net/ethernet/amazon/ena/ena_devlink.c
new file mode 100644
index 000000000000..ac81c24016dd
--- /dev/null
+++ b/drivers/net/ethernet/amazon/ena/ena_devlink.c
@@ -0,0 +1,210 @@
+// SPDX-License-Identifier: GPL-2.0 OR Linux-OpenIB
+/* Copyright (c) Amazon.com, Inc. or its affiliates.
+ * All rights reserved.
+ */
+
+#include "linux/pci.h"
+#include "ena_devlink.h"
+#include "ena_phc.h"
+
+static int ena_devlink_enable_phc_validate(struct devlink *devlink, u32 id,
+					   union devlink_param_value val,
+					   struct netlink_ext_ack *extack)
+{
+	struct ena_adapter *adapter = ENA_DEVLINK_PRIV(devlink);
+
+	if (!val.vbool)
+		return 0;
+
+	if (!ena_com_phc_supported(adapter->ena_dev)) {
+		NL_SET_ERR_MSG_MOD(extack, "Device doesn't support PHC");
+		return -EOPNOTSUPP;
+	}
+
+	return 0;
+}
+
+static const struct devlink_param ena_devlink_params[] = {
+	DEVLINK_PARAM_GENERIC(ENABLE_PHC,
+			      BIT(DEVLINK_PARAM_CMODE_DRIVERINIT),
+			      NULL,
+			      NULL,
+			      ena_devlink_enable_phc_validate),
+};
+
+void ena_devlink_params_get(struct devlink *devlink)
+{
+	struct ena_adapter *adapter = ENA_DEVLINK_PRIV(devlink);
+	union devlink_param_value val;
+	int err;
+
+	err = devl_param_driverinit_value_get(devlink,
+					      DEVLINK_PARAM_GENERIC_ID_ENABLE_PHC,
+					      &val);
+	if (err) {
+		netdev_err(adapter->netdev, "Failed to query PHC param\n");
+		return;
+	}
+
+	ena_phc_enable(adapter, val.vbool);
+}
+
+void ena_devlink_disable_phc_param(struct devlink *devlink)
+{
+	union devlink_param_value value;
+
+	value.vbool = false;
+	devl_param_driverinit_value_set(devlink,
+					DEVLINK_PARAM_GENERIC_ID_ENABLE_PHC,
+					value);
+}
+
+static void ena_devlink_port_register(struct devlink *devlink)
+{
+	struct ena_adapter *adapter = ENA_DEVLINK_PRIV(devlink);
+	struct devlink_port_attrs attrs = {};
+
+	attrs.flavour = DEVLINK_PORT_FLAVOUR_PHYSICAL;
+	devlink_port_attrs_set(&adapter->devlink_port, &attrs);
+	devl_port_register(devlink, &adapter->devlink_port, 0);
+}
+
+static void ena_devlink_port_unregister(struct devlink *devlink)
+{
+	struct ena_adapter *adapter = ENA_DEVLINK_PRIV(devlink);
+
+	devl_port_unregister(&adapter->devlink_port);
+}
+
+static int ena_devlink_reload_down(struct devlink *devlink,
+				   bool netns_change,
+				   enum devlink_reload_action action,
+				   enum devlink_reload_limit limit,
+				   struct netlink_ext_ack *extack)
+{
+	struct ena_adapter *adapter = ENA_DEVLINK_PRIV(devlink);
+
+	if (netns_change) {
+		NL_SET_ERR_MSG_MOD(extack,
+				   "Namespace change is not supported");
+		return -EOPNOTSUPP;
+	}
+
+	ena_devlink_port_unregister(devlink);
+
+	rtnl_lock();
+	ena_destroy_device(adapter, false);
+	rtnl_unlock();
+
+	return 0;
+}
+
+static int ena_devlink_reload_up(struct devlink *devlink,
+				 enum devlink_reload_action action,
+				 enum devlink_reload_limit limit,
+				 u32 *actions_performed,
+				 struct netlink_ext_ack *extack)
+{
+	struct ena_adapter *adapter = ENA_DEVLINK_PRIV(devlink);
+	int err = 0;
+
+	rtnl_lock();
+	/* Check that no other routine initialized the device (e.g.
+	 * ena_fw_reset_device()). Also we're under devlink_mutex here,
+	 * so devlink isn't freed under our feet.
+	 */
+	if (!test_bit(ENA_FLAG_DEVICE_RUNNING, &adapter->flags))
+		err = ena_restore_device(adapter);
+
+	rtnl_unlock();
+
+	ena_devlink_port_register(devlink);
+
+	if (!err)
+		*actions_performed = BIT(DEVLINK_RELOAD_ACTION_DRIVER_REINIT);
+
+	return err;
+}
+
+static const struct devlink_ops ena_devlink_ops = {
+	.reload_actions = BIT(DEVLINK_RELOAD_ACTION_DRIVER_REINIT),
+	.reload_down	= ena_devlink_reload_down,
+	.reload_up	= ena_devlink_reload_up,
+};
+
+static int ena_devlink_configure_params(struct devlink *devlink)
+{
+	struct ena_adapter *adapter = ENA_DEVLINK_PRIV(devlink);
+	union devlink_param_value value;
+	int rc;
+
+	rc = devlink_params_register(devlink, ena_devlink_params,
+				     ARRAY_SIZE(ena_devlink_params));
+	if (rc) {
+		netdev_err(adapter->netdev, "Failed to register devlink params\n");
+		return rc;
+	}
+
+	value.vbool = ena_phc_is_enabled(adapter);
+	devl_param_driverinit_value_set(devlink,
+					DEVLINK_PARAM_GENERIC_ID_ENABLE_PHC,
+					value);
+
+	return 0;
+}
+
+struct devlink *ena_devlink_alloc(struct ena_adapter *adapter)
+{
+	struct device *dev = &adapter->pdev->dev;
+	struct devlink *devlink;
+
+	devlink = devlink_alloc(&ena_devlink_ops,
+				sizeof(struct ena_adapter *),
+				dev);
+	if (!devlink) {
+		netdev_err(adapter->netdev,
+			   "Failed to allocate devlink struct\n");
+		return NULL;
+	}
+
+	ENA_DEVLINK_PRIV(devlink) = adapter;
+	adapter->devlink = devlink;
+
+	if (ena_devlink_configure_params(devlink))
+		goto free_devlink;
+
+	return devlink;
+
+free_devlink:
+	devlink_free(devlink);
+	return NULL;
+}
+
+static void ena_devlink_configure_params_clean(struct devlink *devlink)
+{
+	devlink_params_unregister(devlink, ena_devlink_params,
+				  ARRAY_SIZE(ena_devlink_params));
+}
+
+void ena_devlink_free(struct devlink *devlink)
+{
+	ena_devlink_configure_params_clean(devlink);
+
+	devlink_free(devlink);
+}
+
+void ena_devlink_register(struct devlink *devlink, struct device *dev)
+{
+	devl_lock(devlink);
+	ena_devlink_port_register(devlink);
+	devl_register(devlink);
+	devl_unlock(devlink);
+}
+
+void ena_devlink_unregister(struct devlink *devlink)
+{
+	devl_lock(devlink);
+	ena_devlink_port_unregister(devlink);
+	devl_unregister(devlink);
+	devl_unlock(devlink);
+}
diff --git a/drivers/net/ethernet/amazon/ena/ena_devlink.h b/drivers/net/ethernet/amazon/ena/ena_devlink.h
new file mode 100644
index 000000000000..7a19ce4830d9
--- /dev/null
+++ b/drivers/net/ethernet/amazon/ena/ena_devlink.h
@@ -0,0 +1,21 @@
+/* SPDX-License-Identifier: GPL-2.0 OR Linux-OpenIB */
+/* Copyright (c) Amazon.com, Inc. or its affiliates.
+ * All rights reserved.
+ */
+#ifndef DEVLINK_H
+#define DEVLINK_H
+
+#include "ena_netdev.h"
+#include <net/devlink.h>
+
+#define ENA_DEVLINK_PRIV(devlink) \
+	(*(struct ena_adapter **)devlink_priv(devlink))
+
+struct devlink *ena_devlink_alloc(struct ena_adapter *adapter);
+void ena_devlink_free(struct devlink *devlink);
+void ena_devlink_register(struct devlink *devlink, struct device *dev);
+void ena_devlink_unregister(struct devlink *devlink);
+void ena_devlink_params_get(struct devlink *devlink);
+void ena_devlink_disable_phc_param(struct devlink *devlink);
+
+#endif /* DEVLINK_H */
diff --git a/drivers/net/ethernet/amazon/ena/ena_eth_com.c b/drivers/net/ethernet/amazon/ena/ena_eth_com.c
index f999305e1363..4c6e07aa4bbb 100644
--- a/drivers/net/ethernet/amazon/ena/ena_eth_com.c
+++ b/drivers/net/ethernet/amazon/ena/ena_eth_com.c
@@ -1,38 +1,11 @@
+// SPDX-License-Identifier: GPL-2.0 OR Linux-OpenIB
 /*
- * Copyright 2015 Amazon.com, Inc. or its affiliates.
- *
- * This software is available to you under a choice of one of two
- * licenses.  You may choose to be licensed under the terms of the GNU
- * General Public License (GPL) Version 2, available from the file
- * COPYING in the main directory of this source tree, or the
- * BSD license below:
- *
- *     Redistribution and use in source and binary forms, with or
- *     without modification, are permitted provided that the following
- *     conditions are met:
- *
- *      - Redistributions of source code must retain the above
- *        copyright notice, this list of conditions and the following
- *        disclaimer.
- *
- *      - Redistributions in binary form must reproduce the above
- *        copyright notice, this list of conditions and the following
- *        disclaimer in the documentation and/or other materials
- *        provided with the distribution.
- *
- * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND,
- * EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF
- * MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND
- * NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS
- * BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN
- * ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN
- * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
- * SOFTWARE.
+ * Copyright 2015-2020 Amazon.com, Inc. or its affiliates. All rights reserved.
  */
 
 #include "ena_eth_com.h"
 
-static inline struct ena_eth_io_rx_cdesc_base *ena_com_get_next_rx_cdesc(
+static struct ena_eth_io_rx_cdesc_base *ena_com_get_next_rx_cdesc(
 	struct ena_com_io_cq *io_cq)
 {
 	struct ena_eth_io_rx_cdesc_base *cdesc;
@@ -46,24 +19,20 @@ static inline struct ena_eth_io_rx_cdesc_base *ena_com_get_next_rx_cdesc(
 			+ (head_masked * io_cq->cdesc_entry_size_in_bytes));
 
 	desc_phase = (READ_ONCE(cdesc->status) & ENA_ETH_IO_RX_CDESC_BASE_PHASE_MASK) >>
-			ENA_ETH_IO_RX_CDESC_BASE_PHASE_SHIFT;
+		     ENA_ETH_IO_RX_CDESC_BASE_PHASE_SHIFT;
 
 	if (desc_phase != expected_phase)
 		return NULL;
 
-	return cdesc;
-}
-
-static inline void ena_com_cq_inc_head(struct ena_com_io_cq *io_cq)
-{
-	io_cq->head++;
+	/* Make sure we read the rest of the descriptor after the phase bit
+	 * has been read
+	 */
+	dma_rmb();
 
-	/* Switch phase bit in case of wrap around */
-	if (unlikely((io_cq->head & (io_cq->q_depth - 1)) == 0))
-		io_cq->phase ^= 1;
+	return cdesc;
 }
 
-static inline void *get_sq_desc(struct ena_com_io_sq *io_sq)
+static void *get_sq_desc_regular_queue(struct ena_com_io_sq *io_sq)
 {
 	u16 tail_masked;
 	u32 offset;
@@ -75,50 +44,183 @@ static inline void *get_sq_desc(struct ena_com_io_sq *io_sq)
 	return (void *)((uintptr_t)io_sq->desc_addr.virt_addr + offset);
 }
 
-static inline void ena_com_copy_curr_sq_desc_to_dev(struct ena_com_io_sq *io_sq)
+static int ena_com_write_bounce_buffer_to_dev(struct ena_com_io_sq *io_sq,
+						     u8 *bounce_buffer)
 {
-	u16 tail_masked = io_sq->tail & (io_sq->q_depth - 1);
-	u32 offset = tail_masked * io_sq->desc_entry_size;
+	struct ena_com_llq_info *llq_info = &io_sq->llq_info;
 
-	/* In case this queue isn't a LLQ */
-	if (io_sq->mem_queue_type == ENA_ADMIN_PLACEMENT_POLICY_HOST)
-		return;
+	u16 dst_tail_mask;
+	u32 dst_offset;
 
-	memcpy_toio(io_sq->desc_addr.pbuf_dev_addr + offset,
-		    io_sq->desc_addr.virt_addr + offset,
-		    io_sq->desc_entry_size);
-}
+	dst_tail_mask = io_sq->tail & (io_sq->q_depth - 1);
+	dst_offset = dst_tail_mask * llq_info->desc_list_entry_size;
+
+	if (is_llq_max_tx_burst_exists(io_sq)) {
+		if (unlikely(!io_sq->entries_in_tx_burst_left)) {
+			netdev_err(ena_com_io_sq_to_ena_dev(io_sq)->net_device,
+				   "Error: trying to send more packets than tx burst allows\n");
+			return -ENOSPC;
+		}
+
+		io_sq->entries_in_tx_burst_left--;
+		netdev_dbg(ena_com_io_sq_to_ena_dev(io_sq)->net_device,
+			   "Decreasing entries_in_tx_burst_left of queue %d to %d\n", io_sq->qid,
+			   io_sq->entries_in_tx_burst_left);
+	}
+
+	/* Make sure everything was written into the bounce buffer before
+	 * writing the bounce buffer to the device
+	 */
+	wmb();
+
+	/* The line is completed. Copy it to dev */
+	__iowrite64_copy(io_sq->desc_addr.pbuf_dev_addr + dst_offset, bounce_buffer,
+			 (llq_info->desc_list_entry_size) / 8);
 
-static inline void ena_com_sq_update_tail(struct ena_com_io_sq *io_sq)
-{
 	io_sq->tail++;
 
 	/* Switch phase bit in case of wrap around */
 	if (unlikely((io_sq->tail & (io_sq->q_depth - 1)) == 0))
 		io_sq->phase ^= 1;
+
+	return 0;
 }
 
-static inline int ena_com_write_header(struct ena_com_io_sq *io_sq,
-				       u8 *head_src, u16 header_len)
+static int ena_com_write_header_to_bounce(struct ena_com_io_sq *io_sq,
+						 u8 *header_src,
+						 u16 header_len)
 {
-	u16 tail_masked = io_sq->tail & (io_sq->q_depth - 1);
-	u8 __iomem *dev_head_addr =
-		io_sq->header_addr + (tail_masked * io_sq->tx_max_header_size);
+	struct ena_com_llq_pkt_ctrl *pkt_ctrl = &io_sq->llq_buf_ctrl;
+	struct ena_com_llq_info *llq_info = &io_sq->llq_info;
+	u8 *bounce_buffer = pkt_ctrl->curr_bounce_buf;
+	u16 header_offset;
 
-	if (io_sq->mem_queue_type == ENA_ADMIN_PLACEMENT_POLICY_HOST)
+	if (unlikely(io_sq->mem_queue_type == ENA_ADMIN_PLACEMENT_POLICY_HOST))
 		return 0;
 
-	if (unlikely(!io_sq->header_addr)) {
-		pr_err("Push buffer header ptr is NULL\n");
-		return -EINVAL;
+	header_offset =
+		llq_info->descs_num_before_header * io_sq->desc_entry_size;
+
+	if (unlikely((header_offset + header_len) > llq_info->desc_list_entry_size)) {
+		netdev_err(ena_com_io_sq_to_ena_dev(io_sq)->net_device,
+			   "Trying to write header larger than llq entry can accommodate\n");
+		return -EFAULT;
+	}
+
+	if (unlikely(!bounce_buffer)) {
+		netdev_err(ena_com_io_sq_to_ena_dev(io_sq)->net_device, "Bounce buffer is NULL\n");
+		return -EFAULT;
+	}
+
+	memcpy(bounce_buffer + header_offset, header_src, header_len);
+
+	return 0;
+}
+
+static void *get_sq_desc_llq(struct ena_com_io_sq *io_sq)
+{
+	struct ena_com_llq_pkt_ctrl *pkt_ctrl = &io_sq->llq_buf_ctrl;
+	u8 *bounce_buffer;
+	void *sq_desc;
+
+	bounce_buffer = pkt_ctrl->curr_bounce_buf;
+
+	if (unlikely(!bounce_buffer)) {
+		netdev_err(ena_com_io_sq_to_ena_dev(io_sq)->net_device, "Bounce buffer is NULL\n");
+		return NULL;
 	}
 
-	memcpy_toio(dev_head_addr, head_src, header_len);
+	sq_desc = bounce_buffer + pkt_ctrl->idx * io_sq->desc_entry_size;
+	pkt_ctrl->idx++;
+	pkt_ctrl->descs_left_in_line--;
+
+	return sq_desc;
+}
+
+static int ena_com_close_bounce_buffer(struct ena_com_io_sq *io_sq)
+{
+	struct ena_com_llq_pkt_ctrl *pkt_ctrl = &io_sq->llq_buf_ctrl;
+	struct ena_com_llq_info *llq_info = &io_sq->llq_info;
+	int rc;
+
+	if (unlikely(io_sq->mem_queue_type == ENA_ADMIN_PLACEMENT_POLICY_HOST))
+		return 0;
+
+	/* bounce buffer was used, so write it and get a new one */
+	if (likely(pkt_ctrl->idx)) {
+		rc = ena_com_write_bounce_buffer_to_dev(io_sq,
+							pkt_ctrl->curr_bounce_buf);
+		if (unlikely(rc)) {
+			netdev_err(ena_com_io_sq_to_ena_dev(io_sq)->net_device,
+				   "Failed to write bounce buffer to device\n");
+			return rc;
+		}
+
+		pkt_ctrl->curr_bounce_buf =
+			ena_com_get_next_bounce_buffer(&io_sq->bounce_buf_ctrl);
+		memset(io_sq->llq_buf_ctrl.curr_bounce_buf,
+		       0x0, llq_info->desc_list_entry_size);
+	}
+
+	pkt_ctrl->idx = 0;
+	pkt_ctrl->descs_left_in_line = llq_info->descs_num_before_header;
+	return 0;
+}
+
+static void *get_sq_desc(struct ena_com_io_sq *io_sq)
+{
+	if (io_sq->mem_queue_type == ENA_ADMIN_PLACEMENT_POLICY_DEV)
+		return get_sq_desc_llq(io_sq);
+
+	return get_sq_desc_regular_queue(io_sq);
+}
+
+static int ena_com_sq_update_llq_tail(struct ena_com_io_sq *io_sq)
+{
+	struct ena_com_llq_pkt_ctrl *pkt_ctrl = &io_sq->llq_buf_ctrl;
+	struct ena_com_llq_info *llq_info = &io_sq->llq_info;
+	int rc;
+
+	if (!pkt_ctrl->descs_left_in_line) {
+		rc = ena_com_write_bounce_buffer_to_dev(io_sq,
+							pkt_ctrl->curr_bounce_buf);
+		if (unlikely(rc)) {
+			netdev_err(ena_com_io_sq_to_ena_dev(io_sq)->net_device,
+				   "Failed to write bounce buffer to device\n");
+			return rc;
+		}
+
+		pkt_ctrl->curr_bounce_buf =
+			ena_com_get_next_bounce_buffer(&io_sq->bounce_buf_ctrl);
+		memset(io_sq->llq_buf_ctrl.curr_bounce_buf,
+		       0x0, llq_info->desc_list_entry_size);
+
+		pkt_ctrl->idx = 0;
+		if (unlikely(llq_info->desc_stride_ctrl == ENA_ADMIN_SINGLE_DESC_PER_ENTRY))
+			pkt_ctrl->descs_left_in_line = 1;
+		else
+			pkt_ctrl->descs_left_in_line =
+			llq_info->desc_list_entry_size / io_sq->desc_entry_size;
+	}
+
+	return 0;
+}
+
+static int ena_com_sq_update_tail(struct ena_com_io_sq *io_sq)
+{
+	if (io_sq->mem_queue_type == ENA_ADMIN_PLACEMENT_POLICY_DEV)
+		return ena_com_sq_update_llq_tail(io_sq);
+
+	io_sq->tail++;
+
+	/* Switch phase bit in case of wrap around */
+	if (unlikely((io_sq->tail & (io_sq->q_depth - 1)) == 0))
+		io_sq->phase ^= 1;
 
 	return 0;
 }
 
-static inline struct ena_eth_io_rx_cdesc_base *
+static struct ena_eth_io_rx_cdesc_base *
 	ena_com_rx_cdesc_idx_to_ptr(struct ena_com_io_cq *io_cq, u16 idx)
 {
 	idx &= (io_cq->q_depth - 1);
@@ -127,67 +229,66 @@ static inline struct ena_eth_io_rx_cdesc_base *
 		idx * io_cq->cdesc_entry_size_in_bytes);
 }
 
-static inline u16 ena_com_cdesc_rx_pkt_get(struct ena_com_io_cq *io_cq,
-					   u16 *first_cdesc_idx)
+static int ena_com_cdesc_rx_pkt_get(struct ena_com_io_cq *io_cq,
+				    u16 *first_cdesc_idx,
+				    u16 *num_descs)
 {
+	u16 count = io_cq->cur_rx_pkt_cdesc_count, head_masked;
 	struct ena_eth_io_rx_cdesc_base *cdesc;
-	u16 count = 0, head_masked;
 	u32 last = 0;
 
 	do {
+		u32 status;
+
 		cdesc = ena_com_get_next_rx_cdesc(io_cq);
 		if (!cdesc)
 			break;
+		status = READ_ONCE(cdesc->status);
 
 		ena_com_cq_inc_head(io_cq);
+		if (unlikely((status & ENA_ETH_IO_RX_CDESC_BASE_FIRST_MASK) >>
+		    ENA_ETH_IO_RX_CDESC_BASE_FIRST_SHIFT && count != 0)) {
+			struct ena_com_dev *dev = ena_com_io_cq_to_ena_dev(io_cq);
+
+			netdev_err(dev->net_device,
+				   "First bit is on in descriptor #%d on q_id: %d, req_id: %u\n",
+				   count, io_cq->qid, cdesc->req_id);
+			return -EFAULT;
+		}
 		count++;
-		last = (READ_ONCE(cdesc->status) & ENA_ETH_IO_RX_CDESC_BASE_LAST_MASK) >>
+		last = (status & ENA_ETH_IO_RX_CDESC_BASE_LAST_MASK) >>
 			ENA_ETH_IO_RX_CDESC_BASE_LAST_SHIFT;
 	} while (!last);
 
 	if (last) {
 		*first_cdesc_idx = io_cq->cur_rx_pkt_cdesc_start_idx;
-		count += io_cq->cur_rx_pkt_cdesc_count;
 
 		head_masked = io_cq->head & (io_cq->q_depth - 1);
 
+		*num_descs = count;
 		io_cq->cur_rx_pkt_cdesc_count = 0;
 		io_cq->cur_rx_pkt_cdesc_start_idx = head_masked;
 
-		pr_debug("ena q_id: %d packets were completed. first desc idx %u descs# %d\n",
-			 io_cq->qid, *first_cdesc_idx, count);
+		netdev_dbg(ena_com_io_cq_to_ena_dev(io_cq)->net_device,
+			   "ENA q_id: %d packets were completed. first desc idx %u descs# %d\n",
+			   io_cq->qid, *first_cdesc_idx, count);
 	} else {
-		io_cq->cur_rx_pkt_cdesc_count += count;
-		count = 0;
+		io_cq->cur_rx_pkt_cdesc_count = count;
+		*num_descs = 0;
 	}
 
-	return count;
-}
-
-static inline bool ena_com_meta_desc_changed(struct ena_com_io_sq *io_sq,
-					     struct ena_com_tx_ctx *ena_tx_ctx)
-{
-	int rc;
-
-	if (ena_tx_ctx->meta_valid) {
-		rc = memcmp(&io_sq->cached_tx_meta,
-			    &ena_tx_ctx->ena_meta,
-			    sizeof(struct ena_com_tx_meta));
-
-		if (unlikely(rc != 0))
-			return true;
-	}
-
-	return false;
+	return 0;
 }
 
-static inline void ena_com_create_and_store_tx_meta_desc(struct ena_com_io_sq *io_sq,
-							 struct ena_com_tx_ctx *ena_tx_ctx)
+static int ena_com_create_meta(struct ena_com_io_sq *io_sq,
+			       struct ena_com_tx_meta *ena_meta)
 {
 	struct ena_eth_io_tx_meta_desc *meta_desc = NULL;
-	struct ena_com_tx_meta *ena_meta = &ena_tx_ctx->ena_meta;
 
 	meta_desc = get_sq_desc(io_sq);
+	if (unlikely(!meta_desc))
+		return -EFAULT;
+
 	memset(meta_desc, 0x0, sizeof(struct ena_eth_io_tx_meta_desc));
 
 	meta_desc->len_ctrl |= ENA_ETH_IO_TX_META_DESC_META_DESC_MASK;
@@ -195,7 +296,7 @@ static inline void ena_com_create_and_store_tx_meta_desc(struct ena_com_io_sq *i
 	meta_desc->len_ctrl |= ENA_ETH_IO_TX_META_DESC_EXT_VALID_MASK;
 
 	/* bits 0-9 of the mss */
-	meta_desc->word2 |= (ena_meta->mss <<
+	meta_desc->word2 |= ((u32)ena_meta->mss <<
 		ENA_ETH_IO_TX_META_DESC_MSS_LO_SHIFT) &
 		ENA_ETH_IO_TX_META_DESC_MSS_LO_MASK;
 	/* bits 10-13 of the mss */
@@ -205,34 +306,55 @@ static inline void ena_com_create_and_store_tx_meta_desc(struct ena_com_io_sq *i
 
 	/* Extended meta desc */
 	meta_desc->len_ctrl |= ENA_ETH_IO_TX_META_DESC_ETH_META_TYPE_MASK;
-	meta_desc->len_ctrl |= ENA_ETH_IO_TX_META_DESC_META_STORE_MASK;
-	meta_desc->len_ctrl |= (io_sq->phase <<
+	meta_desc->len_ctrl |= ((u32)io_sq->phase <<
 		ENA_ETH_IO_TX_META_DESC_PHASE_SHIFT) &
 		ENA_ETH_IO_TX_META_DESC_PHASE_MASK;
 
 	meta_desc->len_ctrl |= ENA_ETH_IO_TX_META_DESC_FIRST_MASK;
+	meta_desc->len_ctrl |= ENA_ETH_IO_TX_META_DESC_META_STORE_MASK;
+
 	meta_desc->word2 |= ena_meta->l3_hdr_len &
 		ENA_ETH_IO_TX_META_DESC_L3_HDR_LEN_MASK;
 	meta_desc->word2 |= (ena_meta->l3_hdr_offset <<
 		ENA_ETH_IO_TX_META_DESC_L3_HDR_OFF_SHIFT) &
 		ENA_ETH_IO_TX_META_DESC_L3_HDR_OFF_MASK;
 
-	meta_desc->word2 |= (ena_meta->l4_hdr_len <<
+	meta_desc->word2 |= ((u32)ena_meta->l4_hdr_len <<
 		ENA_ETH_IO_TX_META_DESC_L4_HDR_LEN_IN_WORDS_SHIFT) &
 		ENA_ETH_IO_TX_META_DESC_L4_HDR_LEN_IN_WORDS_MASK;
 
-	meta_desc->len_ctrl |= ENA_ETH_IO_TX_META_DESC_META_STORE_MASK;
+	return ena_com_sq_update_tail(io_sq);
+}
+
+static int ena_com_create_and_store_tx_meta_desc(struct ena_com_io_sq *io_sq,
+						 struct ena_com_tx_ctx *ena_tx_ctx,
+						 bool *have_meta)
+{
+	struct ena_com_tx_meta *ena_meta = &ena_tx_ctx->ena_meta;
 
-	/* Cached the meta desc */
-	memcpy(&io_sq->cached_tx_meta, ena_meta,
-	       sizeof(struct ena_com_tx_meta));
+	/* When disable meta caching is set, don't bother to save the meta and
+	 * compare it to the stored version, just create the meta
+	 */
+	if (io_sq->disable_meta_caching) {
+		*have_meta = true;
+		return ena_com_create_meta(io_sq, ena_meta);
+	}
+
+	if (ena_com_meta_desc_changed(io_sq, ena_tx_ctx)) {
+		*have_meta = true;
+		/* Cache the meta desc */
+		memcpy(&io_sq->cached_tx_meta, ena_meta,
+		       sizeof(struct ena_com_tx_meta));
+		return ena_com_create_meta(io_sq, ena_meta);
+	}
 
-	ena_com_copy_curr_sq_desc_to_dev(io_sq);
-	ena_com_sq_update_tail(io_sq);
+	*have_meta = false;
+	return 0;
 }
 
-static inline void ena_com_rx_set_flags(struct ena_com_rx_ctx *ena_rx_ctx,
-					struct ena_eth_io_rx_cdesc_base *cdesc)
+static void ena_com_rx_set_flags(struct ena_com_io_cq *io_cq,
+				 struct ena_com_rx_ctx *ena_rx_ctx,
+				 struct ena_eth_io_rx_cdesc_base *cdesc)
 {
 	ena_rx_ctx->l3_proto = cdesc->status &
 		ENA_ETH_IO_RX_CDESC_BASE_L3_PROTO_IDX_MASK;
@@ -240,20 +362,23 @@ static inline void ena_com_rx_set_flags(struct ena_com_rx_ctx *ena_rx_ctx,
 		(cdesc->status & ENA_ETH_IO_RX_CDESC_BASE_L4_PROTO_IDX_MASK) >>
 		ENA_ETH_IO_RX_CDESC_BASE_L4_PROTO_IDX_SHIFT;
 	ena_rx_ctx->l3_csum_err =
-		(cdesc->status & ENA_ETH_IO_RX_CDESC_BASE_L3_CSUM_ERR_MASK) >>
-		ENA_ETH_IO_RX_CDESC_BASE_L3_CSUM_ERR_SHIFT;
+		!!((cdesc->status & ENA_ETH_IO_RX_CDESC_BASE_L3_CSUM_ERR_MASK) >>
+		ENA_ETH_IO_RX_CDESC_BASE_L3_CSUM_ERR_SHIFT);
 	ena_rx_ctx->l4_csum_err =
-		(cdesc->status & ENA_ETH_IO_RX_CDESC_BASE_L4_CSUM_ERR_MASK) >>
-		ENA_ETH_IO_RX_CDESC_BASE_L4_CSUM_ERR_SHIFT;
+		!!((cdesc->status & ENA_ETH_IO_RX_CDESC_BASE_L4_CSUM_ERR_MASK) >>
+		ENA_ETH_IO_RX_CDESC_BASE_L4_CSUM_ERR_SHIFT);
+	ena_rx_ctx->l4_csum_checked =
+		!!((cdesc->status & ENA_ETH_IO_RX_CDESC_BASE_L4_CSUM_CHECKED_MASK) >>
+		ENA_ETH_IO_RX_CDESC_BASE_L4_CSUM_CHECKED_SHIFT);
 	ena_rx_ctx->hash = cdesc->hash;
 	ena_rx_ctx->frag =
 		(cdesc->status & ENA_ETH_IO_RX_CDESC_BASE_IPV4_FRAG_MASK) >>
 		ENA_ETH_IO_RX_CDESC_BASE_IPV4_FRAG_SHIFT;
 
-	pr_debug("ena_rx_ctx->l3_proto %d ena_rx_ctx->l4_proto %d\nena_rx_ctx->l3_csum_err %d ena_rx_ctx->l4_csum_err %d\nhash frag %d frag: %d cdesc_status: %x\n",
-		 ena_rx_ctx->l3_proto, ena_rx_ctx->l4_proto,
-		 ena_rx_ctx->l3_csum_err, ena_rx_ctx->l4_csum_err,
-		 ena_rx_ctx->hash, ena_rx_ctx->frag, cdesc->status);
+	netdev_dbg(ena_com_io_cq_to_ena_dev(io_cq)->net_device,
+		   "l3_proto %d l4_proto %d l3_csum_err %d l4_csum_err %d hash %d frag %d cdesc_status %x\n",
+		   ena_rx_ctx->l3_proto, ena_rx_ctx->l4_proto, ena_rx_ctx->l3_csum_err,
+		   ena_rx_ctx->l4_csum_err, ena_rx_ctx->hash, ena_rx_ctx->frag, cdesc->status);
 }
 
 /*****************************************************************************/
@@ -266,60 +391,76 @@ int ena_com_prepare_tx(struct ena_com_io_sq *io_sq,
 {
 	struct ena_eth_io_tx_desc *desc = NULL;
 	struct ena_com_buf *ena_bufs = ena_tx_ctx->ena_bufs;
-	void *push_header = ena_tx_ctx->push_header;
+	void *buffer_to_push = ena_tx_ctx->push_header;
 	u16 header_len = ena_tx_ctx->header_len;
 	u16 num_bufs = ena_tx_ctx->num_bufs;
-	int total_desc, i, rc;
+	u16 start_tail = io_sq->tail;
+	int i, rc;
 	bool have_meta;
 	u64 addr_hi;
 
 	WARN(io_sq->direction != ENA_COM_IO_QUEUE_DIRECTION_TX, "wrong Q type");
 
 	/* num_bufs +1 for potential meta desc */
-	if (ena_com_sq_empty_space(io_sq) < (num_bufs + 1)) {
-		pr_err("Not enough space in the tx queue\n");
+	if (unlikely(!ena_com_sq_have_enough_space(io_sq, num_bufs + 1))) {
+		netdev_dbg(ena_com_io_sq_to_ena_dev(io_sq)->net_device,
+			   "Not enough space in the tx queue\n");
 		return -ENOMEM;
 	}
 
 	if (unlikely(header_len > io_sq->tx_max_header_size)) {
-		pr_err("header size is too large %d max header: %d\n",
-		       header_len, io_sq->tx_max_header_size);
+		netdev_err(ena_com_io_sq_to_ena_dev(io_sq)->net_device,
+			   "Header size is too large %d max header: %d\n", header_len,
+			   io_sq->tx_max_header_size);
+		return -EINVAL;
+	}
+
+	if (unlikely(io_sq->mem_queue_type == ENA_ADMIN_PLACEMENT_POLICY_DEV && !buffer_to_push)) {
+		netdev_err(ena_com_io_sq_to_ena_dev(io_sq)->net_device,
+			   "Push header wasn't provided in LLQ mode\n");
 		return -EINVAL;
 	}
 
-	/* start with pushing the header (if needed) */
-	rc = ena_com_write_header(io_sq, push_header, header_len);
+	rc = ena_com_write_header_to_bounce(io_sq, buffer_to_push, header_len);
 	if (unlikely(rc))
 		return rc;
 
-	have_meta = ena_tx_ctx->meta_valid && ena_com_meta_desc_changed(io_sq,
-			ena_tx_ctx);
-	if (have_meta)
-		ena_com_create_and_store_tx_meta_desc(io_sq, ena_tx_ctx);
+	rc = ena_com_create_and_store_tx_meta_desc(io_sq, ena_tx_ctx, &have_meta);
+	if (unlikely(rc)) {
+		netdev_err(ena_com_io_sq_to_ena_dev(io_sq)->net_device,
+			   "Failed to create and store tx meta desc\n");
+		return rc;
+	}
 
-	/* If the caller doesn't want send packets */
+	/* If the caller doesn't want to send packets */
 	if (unlikely(!num_bufs && !header_len)) {
-		*nb_hw_desc = have_meta ? 0 : 1;
-		return 0;
+		rc = ena_com_close_bounce_buffer(io_sq);
+		if (rc)
+			netdev_err(ena_com_io_sq_to_ena_dev(io_sq)->net_device,
+				   "Failed to write buffers to LLQ\n");
+		*nb_hw_desc = io_sq->tail - start_tail;
+		return rc;
 	}
 
 	desc = get_sq_desc(io_sq);
+	if (unlikely(!desc))
+		return -EFAULT;
 	memset(desc, 0x0, sizeof(struct ena_eth_io_tx_desc));
 
 	/* Set first desc when we don't have meta descriptor */
 	if (!have_meta)
 		desc->len_ctrl |= ENA_ETH_IO_TX_DESC_FIRST_MASK;
 
-	desc->buff_addr_hi_hdr_sz |= (header_len <<
+	desc->buff_addr_hi_hdr_sz |= ((u32)header_len <<
 		ENA_ETH_IO_TX_DESC_HEADER_LENGTH_SHIFT) &
 		ENA_ETH_IO_TX_DESC_HEADER_LENGTH_MASK;
-	desc->len_ctrl |= (io_sq->phase << ENA_ETH_IO_TX_DESC_PHASE_SHIFT) &
+	desc->len_ctrl |= ((u32)io_sq->phase << ENA_ETH_IO_TX_DESC_PHASE_SHIFT) &
 		ENA_ETH_IO_TX_DESC_PHASE_MASK;
 
 	desc->len_ctrl |= ENA_ETH_IO_TX_DESC_COMP_REQ_MASK;
 
 	/* Bits 0-9 */
-	desc->meta_ctrl |= (ena_tx_ctx->req_id <<
+	desc->meta_ctrl |= ((u32)ena_tx_ctx->req_id <<
 		ENA_ETH_IO_TX_DESC_REQ_ID_LO_SHIFT) &
 		ENA_ETH_IO_TX_DESC_REQ_ID_LO_MASK;
 
@@ -355,13 +496,20 @@ int ena_com_prepare_tx(struct ena_com_io_sq *io_sq,
 	for (i = 0; i < num_bufs; i++) {
 		/* The first desc share the same desc as the header */
 		if (likely(i != 0)) {
-			ena_com_copy_curr_sq_desc_to_dev(io_sq);
-			ena_com_sq_update_tail(io_sq);
+			rc = ena_com_sq_update_tail(io_sq);
+			if (unlikely(rc)) {
+				netdev_err(ena_com_io_sq_to_ena_dev(io_sq)->net_device,
+					   "Failed to update sq tail\n");
+				return rc;
+			}
 
 			desc = get_sq_desc(io_sq);
+			if (unlikely(!desc))
+				return -EFAULT;
+
 			memset(desc, 0x0, sizeof(struct ena_eth_io_tx_desc));
 
-			desc->len_ctrl |= (io_sq->phase <<
+			desc->len_ctrl |= ((u32)io_sq->phase <<
 				ENA_ETH_IO_TX_DESC_PHASE_SHIFT) &
 				ENA_ETH_IO_TX_DESC_PHASE_MASK;
 		}
@@ -381,15 +529,17 @@ int ena_com_prepare_tx(struct ena_com_io_sq *io_sq,
 	/* set the last desc indicator */
 	desc->len_ctrl |= ENA_ETH_IO_TX_DESC_LAST_MASK;
 
-	ena_com_copy_curr_sq_desc_to_dev(io_sq);
-
-	ena_com_sq_update_tail(io_sq);
+	rc = ena_com_sq_update_tail(io_sq);
+	if (unlikely(rc)) {
+		netdev_err(ena_com_io_sq_to_ena_dev(io_sq)->net_device,
+			   "Failed to update sq tail of the last descriptor\n");
+		return rc;
+	}
 
-	total_desc = max_t(u16, num_bufs, 1);
-	total_desc += have_meta ? 1 : 0;
+	rc = ena_com_close_bounce_buffer(io_sq);
 
-	*nb_hw_desc = total_desc;
-	return 0;
+	*nb_hw_desc = io_sq->tail - start_tail;
+	return rc;
 }
 
 int ena_com_rx_pkt(struct ena_com_io_cq *io_cq,
@@ -398,45 +548,60 @@ int ena_com_rx_pkt(struct ena_com_io_cq *io_cq,
 {
 	struct ena_com_rx_buf_info *ena_buf = &ena_rx_ctx->ena_bufs[0];
 	struct ena_eth_io_rx_cdesc_base *cdesc = NULL;
+	u16 q_depth = io_cq->q_depth;
 	u16 cdesc_idx = 0;
 	u16 nb_hw_desc;
-	u16 i;
+	u16 i = 0;
+	int rc;
 
 	WARN(io_cq->direction != ENA_COM_IO_QUEUE_DIRECTION_RX, "wrong Q type");
 
-	nb_hw_desc = ena_com_cdesc_rx_pkt_get(io_cq, &cdesc_idx);
+	rc = ena_com_cdesc_rx_pkt_get(io_cq, &cdesc_idx, &nb_hw_desc);
+	if (unlikely(rc != 0))
+		return -EFAULT;
+
 	if (nb_hw_desc == 0) {
 		ena_rx_ctx->descs = nb_hw_desc;
 		return 0;
 	}
 
-	pr_debug("fetch rx packet: queue %d completed desc: %d\n", io_cq->qid,
-		 nb_hw_desc);
+	netdev_dbg(ena_com_io_cq_to_ena_dev(io_cq)->net_device,
+		   "Fetch rx packet: queue %d completed desc: %d\n", io_cq->qid, nb_hw_desc);
 
 	if (unlikely(nb_hw_desc > ena_rx_ctx->max_bufs)) {
-		pr_err("Too many RX cdescs (%d) > MAX(%d)\n", nb_hw_desc,
-		       ena_rx_ctx->max_bufs);
+		netdev_err(ena_com_io_cq_to_ena_dev(io_cq)->net_device,
+			   "Too many RX cdescs (%d) > MAX(%d)\n", nb_hw_desc, ena_rx_ctx->max_bufs);
 		return -ENOSPC;
 	}
 
-	for (i = 0; i < nb_hw_desc; i++) {
+	cdesc = ena_com_rx_cdesc_idx_to_ptr(io_cq, cdesc_idx);
+	ena_rx_ctx->pkt_offset = cdesc->offset;
+
+	do {
+		ena_buf[i].len = cdesc->length;
+		ena_buf[i].req_id = cdesc->req_id;
+		if (unlikely(ena_buf[i].req_id >= q_depth))
+			return -EIO;
+
+		if (++i >= nb_hw_desc)
+			break;
+
 		cdesc = ena_com_rx_cdesc_idx_to_ptr(io_cq, cdesc_idx + i);
 
-		ena_buf->len = cdesc->length;
-		ena_buf->req_id = cdesc->req_id;
-		ena_buf++;
-	}
+	} while (1);
 
 	/* Update SQ head ptr */
 	io_sq->next_to_comp += nb_hw_desc;
 
-	pr_debug("[%s][QID#%d] Updating SQ head to: %d\n", __func__, io_sq->qid,
-		 io_sq->next_to_comp);
+	netdev_dbg(ena_com_io_cq_to_ena_dev(io_cq)->net_device,
+		   "[%s][QID#%d] Updating SQ head to: %d\n", __func__, io_sq->qid,
+		   io_sq->next_to_comp);
 
 	/* Get rx flags from the last pkt */
-	ena_com_rx_set_flags(ena_rx_ctx, cdesc);
+	ena_com_rx_set_flags(io_cq, ena_rx_ctx, cdesc);
 
 	ena_rx_ctx->descs = nb_hw_desc;
+
 	return 0;
 }
 
@@ -448,54 +613,42 @@ int ena_com_add_single_rx_desc(struct ena_com_io_sq *io_sq,
 
 	WARN(io_sq->direction != ENA_COM_IO_QUEUE_DIRECTION_RX, "wrong Q type");
 
-	if (unlikely(ena_com_sq_empty_space(io_sq) == 0))
+	if (unlikely(!ena_com_sq_have_enough_space(io_sq, 1)))
 		return -ENOSPC;
 
 	desc = get_sq_desc(io_sq);
+	if (unlikely(!desc))
+		return -EFAULT;
+
 	memset(desc, 0x0, sizeof(struct ena_eth_io_rx_desc));
 
 	desc->length = ena_buf->len;
 
-	desc->ctrl |= ENA_ETH_IO_RX_DESC_FIRST_MASK;
-	desc->ctrl |= ENA_ETH_IO_RX_DESC_LAST_MASK;
-	desc->ctrl |= io_sq->phase & ENA_ETH_IO_RX_DESC_PHASE_MASK;
-	desc->ctrl |= ENA_ETH_IO_RX_DESC_COMP_REQ_MASK;
+	desc->ctrl = ENA_ETH_IO_RX_DESC_FIRST_MASK |
+		     ENA_ETH_IO_RX_DESC_LAST_MASK |
+		     ENA_ETH_IO_RX_DESC_COMP_REQ_MASK |
+		     (io_sq->phase & ENA_ETH_IO_RX_DESC_PHASE_MASK);
 
 	desc->req_id = req_id;
 
+	netdev_dbg(ena_com_io_sq_to_ena_dev(io_sq)->net_device,
+		   "[%s] Adding single RX desc, Queue: %u, req_id: %u\n", __func__, io_sq->qid,
+		   req_id);
+
 	desc->buff_addr_lo = (u32)ena_buf->paddr;
 	desc->buff_addr_hi =
 		((ena_buf->paddr & GENMASK_ULL(io_sq->dma_addr_bits - 1, 32)) >> 32);
 
-	ena_com_sq_update_tail(io_sq);
-
-	return 0;
+	return ena_com_sq_update_tail(io_sq);
 }
 
-int ena_com_tx_comp_req_id_get(struct ena_com_io_cq *io_cq, u16 *req_id)
+bool ena_com_cq_empty(struct ena_com_io_cq *io_cq)
 {
-	u8 expected_phase, cdesc_phase;
-	struct ena_eth_io_tx_cdesc *cdesc;
-	u16 masked_head;
-
-	masked_head = io_cq->head & (io_cq->q_depth - 1);
-	expected_phase = io_cq->phase;
-
-	cdesc = (struct ena_eth_io_tx_cdesc *)
-		((uintptr_t)io_cq->cdesc_addr.virt_addr +
-		(masked_head * io_cq->cdesc_entry_size_in_bytes));
-
-	/* When the current completion descriptor phase isn't the same as the
-	 * expected, it mean that the device still didn't update
-	 * this completion.
-	 */
-	cdesc_phase = READ_ONCE(cdesc->flags) & ENA_ETH_IO_TX_CDESC_PHASE_MASK;
-	if (cdesc_phase != expected_phase)
-		return -EAGAIN;
-
-	ena_com_cq_inc_head(io_cq);
-
-	*req_id = READ_ONCE(cdesc->req_id);
+	struct ena_eth_io_rx_cdesc_base *cdesc;
 
-	return 0;
+	cdesc = ena_com_get_next_rx_cdesc(io_cq);
+	if (cdesc)
+		return false;
+	else
+		return true;
 }
diff --git a/drivers/net/ethernet/amazon/ena/ena_eth_com.h b/drivers/net/ethernet/amazon/ena/ena_eth_com.h
index bb53c3a4f8e9..449bc4960ccc 100644
--- a/drivers/net/ethernet/amazon/ena/ena_eth_com.h
+++ b/drivers/net/ethernet/amazon/ena/ena_eth_com.h
@@ -1,33 +1,6 @@
+/* SPDX-License-Identifier: GPL-2.0 OR Linux-OpenIB */
 /*
- * Copyright 2015 Amazon.com, Inc. or its affiliates.
- *
- * This software is available to you under a choice of one of two
- * licenses.  You may choose to be licensed under the terms of the GNU
- * General Public License (GPL) Version 2, available from the file
- * COPYING in the main directory of this source tree, or the
- * BSD license below:
- *
- *     Redistribution and use in source and binary forms, with or
- *     without modification, are permitted provided that the following
- *     conditions are met:
- *
- *      - Redistributions of source code must retain the above
- *        copyright notice, this list of conditions and the following
- *        disclaimer.
- *
- *      - Redistributions in binary form must reproduce the above
- *        copyright notice, this list of conditions and the following
- *        disclaimer in the documentation and/or other materials
- *        provided with the distribution.
- *
- * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND,
- * EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF
- * MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND
- * NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS
- * BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN
- * ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN
- * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
- * SOFTWARE.
+ * Copyright 2015-2020 Amazon.com, Inc. or its affiliates. All rights reserved.
  */
 
 #ifndef ENA_ETH_COM_H_
@@ -35,8 +8,10 @@
 
 #include "ena_com.h"
 
-/* head update threshold in units of (queue size / ENA_COMP_HEAD_THRESH) */
-#define ENA_COMP_HEAD_THRESH 4
+/* we allow 2 DMA descriptors per LLQ entry */
+#define ENA_LLQ_ENTRY_DESC_CHUNK_SIZE	(2 * sizeof(struct ena_eth_io_tx_desc))
+#define ENA_LLQ_HEADER		(128UL - ENA_LLQ_ENTRY_DESC_CHUNK_SIZE)
+#define ENA_LLQ_LARGE_HEADER	(256UL - ENA_LLQ_ENTRY_DESC_CHUNK_SIZE)
 
 struct ena_com_tx_ctx {
 	struct ena_com_tx_meta ena_meta;
@@ -67,11 +42,13 @@ struct ena_com_rx_ctx {
 	enum ena_eth_io_l4_proto_index l4_proto;
 	bool l3_csum_err;
 	bool l4_csum_err;
+	u8 l4_csum_checked;
 	/* fragmented packet */
 	bool frag;
 	u32 hash;
 	u16 descs;
-	int max_bufs;
+	u16 max_bufs;
+	u8 pkt_offset;
 };
 
 int ena_com_prepare_tx(struct ena_com_io_sq *io_sq,
@@ -86,7 +63,7 @@ int ena_com_add_single_rx_desc(struct ena_com_io_sq *io_sq,
 			       struct ena_com_buf *ena_buf,
 			       u16 req_id);
 
-int ena_com_tx_comp_req_id_get(struct ena_com_io_cq *io_cq, u16 *req_id);
+bool ena_com_cq_empty(struct ena_com_io_cq *io_cq);
 
 static inline void ena_com_unmask_intr(struct ena_com_io_cq *io_cq,
 				       struct ena_eth_io_intr_reg *intr_reg)
@@ -94,7 +71,7 @@ static inline void ena_com_unmask_intr(struct ena_com_io_cq *io_cq,
 	writel(intr_reg->intr_control, io_cq->unmask_reg);
 }
 
-static inline int ena_com_sq_empty_space(struct ena_com_io_sq *io_sq)
+static inline int ena_com_free_q_entries(struct ena_com_io_sq *io_sq)
 {
 	u16 tail, next_to_comp, cnt;
 
@@ -105,34 +82,88 @@ static inline int ena_com_sq_empty_space(struct ena_com_io_sq *io_sq)
 	return io_sq->q_depth - 1 - cnt;
 }
 
-static inline int ena_com_write_sq_doorbell(struct ena_com_io_sq *io_sq)
+/* Check if the submission queue has enough space to hold required_buffers */
+static inline bool ena_com_sq_have_enough_space(struct ena_com_io_sq *io_sq,
+						u16 required_buffers)
 {
-	u16 tail;
+	int temp;
 
-	tail = io_sq->tail;
+	if (io_sq->mem_queue_type == ENA_ADMIN_PLACEMENT_POLICY_HOST)
+		return ena_com_free_q_entries(io_sq) >= required_buffers;
 
-	pr_debug("write submission queue doorbell for queue: %d tail: %d\n",
-		 io_sq->qid, tail);
+	/* This calculation doesn't need to be 100% accurate. So to reduce
+	 * the calculation overhead just Subtract 2 lines from the free descs
+	 * (one for the header line and one to compensate the devision
+	 * down calculation.
+	 */
+	temp = required_buffers / io_sq->llq_info.descs_per_entry + 2;
 
-	writel(tail, io_sq->db_addr);
+	return ena_com_free_q_entries(io_sq) > temp;
+}
 
-	return 0;
+static inline bool ena_com_meta_desc_changed(struct ena_com_io_sq *io_sq,
+					     struct ena_com_tx_ctx *ena_tx_ctx)
+{
+	if (!ena_tx_ctx->meta_valid)
+		return false;
+
+	return !!memcmp(&io_sq->cached_tx_meta,
+			&ena_tx_ctx->ena_meta,
+			sizeof(struct ena_com_tx_meta));
+}
+
+static inline bool is_llq_max_tx_burst_exists(struct ena_com_io_sq *io_sq)
+{
+	return (io_sq->mem_queue_type == ENA_ADMIN_PLACEMENT_POLICY_DEV) &&
+	       io_sq->llq_info.max_entries_in_tx_burst > 0;
 }
 
-static inline int ena_com_update_dev_comp_head(struct ena_com_io_cq *io_cq)
+static inline bool ena_com_is_doorbell_needed(struct ena_com_io_sq *io_sq,
+					      struct ena_com_tx_ctx *ena_tx_ctx)
 {
-	u16 unreported_comp, head;
-	bool need_update;
-
-	head = io_cq->head;
-	unreported_comp = head - io_cq->last_head_update;
-	need_update = unreported_comp > (io_cq->q_depth / ENA_COMP_HEAD_THRESH);
-
-	if (io_cq->cq_head_db_reg && need_update) {
-		pr_debug("Write completion queue doorbell for queue %d: head: %d\n",
-			 io_cq->qid, head);
-		writel(head, io_cq->cq_head_db_reg);
-		io_cq->last_head_update = head;
+	struct ena_com_llq_info *llq_info;
+	int descs_after_first_entry;
+	int num_entries_needed = 1;
+	u16 num_descs;
+
+	if (!is_llq_max_tx_burst_exists(io_sq))
+		return false;
+
+	llq_info = &io_sq->llq_info;
+	num_descs = ena_tx_ctx->num_bufs;
+
+	if (llq_info->disable_meta_caching ||
+	    unlikely(ena_com_meta_desc_changed(io_sq, ena_tx_ctx)))
+		++num_descs;
+
+	if (num_descs > llq_info->descs_num_before_header) {
+		descs_after_first_entry = num_descs - llq_info->descs_num_before_header;
+		num_entries_needed += DIV_ROUND_UP(descs_after_first_entry,
+						   llq_info->descs_per_entry);
+	}
+
+	netdev_dbg(ena_com_io_sq_to_ena_dev(io_sq)->net_device,
+		   "Queue: %d num_descs: %d num_entries_needed: %d\n", io_sq->qid, num_descs,
+		   num_entries_needed);
+
+	return num_entries_needed > io_sq->entries_in_tx_burst_left;
+}
+
+static inline int ena_com_write_sq_doorbell(struct ena_com_io_sq *io_sq)
+{
+	u16 max_entries_in_tx_burst = io_sq->llq_info.max_entries_in_tx_burst;
+	u16 tail = io_sq->tail;
+
+	netdev_dbg(ena_com_io_sq_to_ena_dev(io_sq)->net_device,
+		   "Write submission queue doorbell for queue: %d tail: %d\n", io_sq->qid, tail);
+
+	writel(tail, io_sq->db_addr);
+
+	if (is_llq_max_tx_burst_exists(io_sq)) {
+		netdev_dbg(ena_com_io_sq_to_ena_dev(io_sq)->net_device,
+			   "Reset available entries in tx burst for queue %d to %d\n", io_sq->qid,
+			   max_entries_in_tx_burst);
+		io_sq->entries_in_tx_burst_left = max_entries_in_tx_burst;
 	}
 
 	return 0;
@@ -157,4 +188,49 @@ static inline void ena_com_comp_ack(struct ena_com_io_sq *io_sq, u16 elem)
 	io_sq->next_to_comp += elem;
 }
 
+static inline void ena_com_cq_inc_head(struct ena_com_io_cq *io_cq)
+{
+	io_cq->head++;
+
+	/* Switch phase bit in case of wrap around */
+	if (unlikely((io_cq->head & (io_cq->q_depth - 1)) == 0))
+		io_cq->phase ^= 1;
+}
+
+static inline int ena_com_tx_comp_req_id_get(struct ena_com_io_cq *io_cq,
+					     u16 *req_id)
+{
+	u8 expected_phase, cdesc_phase;
+	struct ena_eth_io_tx_cdesc *cdesc;
+	u16 masked_head;
+
+	masked_head = io_cq->head & (io_cq->q_depth - 1);
+	expected_phase = io_cq->phase;
+
+	cdesc = (struct ena_eth_io_tx_cdesc *)
+		((uintptr_t)io_cq->cdesc_addr.virt_addr +
+		(masked_head * io_cq->cdesc_entry_size_in_bytes));
+
+	/* When the current completion descriptor phase isn't the same as the
+	 * expected, it mean that the device still didn't update
+	 * this completion.
+	 */
+	cdesc_phase = READ_ONCE(cdesc->flags) & ENA_ETH_IO_TX_CDESC_PHASE_MASK;
+	if (cdesc_phase != expected_phase)
+		return -EAGAIN;
+
+	dma_rmb();
+
+	*req_id = READ_ONCE(cdesc->req_id);
+	if (unlikely(*req_id >= io_cq->q_depth)) {
+		netdev_err(ena_com_io_cq_to_ena_dev(io_cq)->net_device, "Invalid req id %d\n",
+			   cdesc->req_id);
+		return -EINVAL;
+	}
+
+	ena_com_cq_inc_head(io_cq);
+
+	return 0;
+}
+
 #endif /* ENA_ETH_COM_H_ */
diff --git a/drivers/net/ethernet/amazon/ena/ena_eth_io_defs.h b/drivers/net/ethernet/amazon/ena/ena_eth_io_defs.h
index f320c58793a5..332ac0d28ac7 100644
--- a/drivers/net/ethernet/amazon/ena/ena_eth_io_defs.h
+++ b/drivers/net/ethernet/amazon/ena/ena_eth_io_defs.h
@@ -1,57 +1,23 @@
+/* SPDX-License-Identifier: GPL-2.0 OR Linux-OpenIB */
 /*
- * Copyright 2015 - 2016 Amazon.com, Inc. or its affiliates.
- *
- * This software is available to you under a choice of one of two
- * licenses.  You may choose to be licensed under the terms of the GNU
- * General Public License (GPL) Version 2, available from the file
- * COPYING in the main directory of this source tree, or the
- * BSD license below:
- *
- *     Redistribution and use in source and binary forms, with or
- *     without modification, are permitted provided that the following
- *     conditions are met:
- *
- *      - Redistributions of source code must retain the above
- *        copyright notice, this list of conditions and the following
- *        disclaimer.
- *
- *      - Redistributions in binary form must reproduce the above
- *        copyright notice, this list of conditions and the following
- *        disclaimer in the documentation and/or other materials
- *        provided with the distribution.
- *
- * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND,
- * EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF
- * MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND
- * NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS
- * BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN
- * ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN
- * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
- * SOFTWARE.
+ * Copyright 2015-2020 Amazon.com, Inc. or its affiliates. All rights reserved.
  */
 #ifndef _ENA_ETH_IO_H_
 #define _ENA_ETH_IO_H_
 
 enum ena_eth_io_l3_proto_index {
-	ENA_ETH_IO_L3_PROTO_UNKNOWN	= 0,
-
-	ENA_ETH_IO_L3_PROTO_IPV4	= 8,
-
-	ENA_ETH_IO_L3_PROTO_IPV6	= 11,
-
-	ENA_ETH_IO_L3_PROTO_FCOE	= 21,
-
-	ENA_ETH_IO_L3_PROTO_ROCE	= 22,
+	ENA_ETH_IO_L3_PROTO_UNKNOWN                 = 0,
+	ENA_ETH_IO_L3_PROTO_IPV4                    = 8,
+	ENA_ETH_IO_L3_PROTO_IPV6                    = 11,
+	ENA_ETH_IO_L3_PROTO_FCOE                    = 21,
+	ENA_ETH_IO_L3_PROTO_ROCE                    = 22,
 };
 
 enum ena_eth_io_l4_proto_index {
-	ENA_ETH_IO_L4_PROTO_UNKNOWN		= 0,
-
-	ENA_ETH_IO_L4_PROTO_TCP			= 12,
-
-	ENA_ETH_IO_L4_PROTO_UDP			= 13,
-
-	ENA_ETH_IO_L4_PROTO_ROUTEABLE_ROCE	= 23,
+	ENA_ETH_IO_L4_PROTO_UNKNOWN                 = 0,
+	ENA_ETH_IO_L4_PROTO_TCP                     = 12,
+	ENA_ETH_IO_L4_PROTO_UDP                     = 13,
+	ENA_ETH_IO_L4_PROTO_ROUTEABLE_ROCE          = 23,
 };
 
 struct ena_eth_io_tx_desc {
@@ -242,9 +208,13 @@ struct ena_eth_io_rx_cdesc_base {
 	 *    checksum error detected, or, the controller didn't
 	 *    validate the checksum. This bit is valid only when
 	 *    l4_proto_idx indicates TCP/UDP packet, and,
-	 *    ipv4_frag is not set
+	 *    ipv4_frag is not set. This bit is valid only when
+	 *    l4_csum_checked below is set.
 	 * 15 : ipv4_frag - Indicates IPv4 fragmented packet
-	 * 23:16 : reserved16
+	 * 16 : l4_csum_checked - L4 checksum was verified
+	 *    (could be OK or error), when cleared the status of
+	 *    checksum is unknown
+	 * 23:17 : reserved17 - MBZ
 	 * 24 : phase
 	 * 25 : l3_csum2 - second checksum engine result
 	 * 26 : first - Indicates first descriptor in
@@ -267,7 +237,9 @@ struct ena_eth_io_rx_cdesc_base {
 
 	u16 sub_qid;
 
-	u16 reserved;
+	u8 offset;
+
+	u8 reserved;
 };
 
 /* 8-word format */
@@ -303,114 +275,116 @@ struct ena_eth_io_numa_node_cfg_reg {
 };
 
 /* tx_desc */
-#define ENA_ETH_IO_TX_DESC_LENGTH_MASK GENMASK(15, 0)
-#define ENA_ETH_IO_TX_DESC_REQ_ID_HI_SHIFT 16
-#define ENA_ETH_IO_TX_DESC_REQ_ID_HI_MASK GENMASK(21, 16)
-#define ENA_ETH_IO_TX_DESC_META_DESC_SHIFT 23
-#define ENA_ETH_IO_TX_DESC_META_DESC_MASK BIT(23)
-#define ENA_ETH_IO_TX_DESC_PHASE_SHIFT 24
-#define ENA_ETH_IO_TX_DESC_PHASE_MASK BIT(24)
-#define ENA_ETH_IO_TX_DESC_FIRST_SHIFT 26
-#define ENA_ETH_IO_TX_DESC_FIRST_MASK BIT(26)
-#define ENA_ETH_IO_TX_DESC_LAST_SHIFT 27
-#define ENA_ETH_IO_TX_DESC_LAST_MASK BIT(27)
-#define ENA_ETH_IO_TX_DESC_COMP_REQ_SHIFT 28
-#define ENA_ETH_IO_TX_DESC_COMP_REQ_MASK BIT(28)
-#define ENA_ETH_IO_TX_DESC_L3_PROTO_IDX_MASK GENMASK(3, 0)
-#define ENA_ETH_IO_TX_DESC_DF_SHIFT 4
-#define ENA_ETH_IO_TX_DESC_DF_MASK BIT(4)
-#define ENA_ETH_IO_TX_DESC_TSO_EN_SHIFT 7
-#define ENA_ETH_IO_TX_DESC_TSO_EN_MASK BIT(7)
-#define ENA_ETH_IO_TX_DESC_L4_PROTO_IDX_SHIFT 8
-#define ENA_ETH_IO_TX_DESC_L4_PROTO_IDX_MASK GENMASK(12, 8)
-#define ENA_ETH_IO_TX_DESC_L3_CSUM_EN_SHIFT 13
-#define ENA_ETH_IO_TX_DESC_L3_CSUM_EN_MASK BIT(13)
-#define ENA_ETH_IO_TX_DESC_L4_CSUM_EN_SHIFT 14
-#define ENA_ETH_IO_TX_DESC_L4_CSUM_EN_MASK BIT(14)
-#define ENA_ETH_IO_TX_DESC_ETHERNET_FCS_DIS_SHIFT 15
-#define ENA_ETH_IO_TX_DESC_ETHERNET_FCS_DIS_MASK BIT(15)
-#define ENA_ETH_IO_TX_DESC_L4_CSUM_PARTIAL_SHIFT 17
-#define ENA_ETH_IO_TX_DESC_L4_CSUM_PARTIAL_MASK BIT(17)
-#define ENA_ETH_IO_TX_DESC_REQ_ID_LO_SHIFT 22
-#define ENA_ETH_IO_TX_DESC_REQ_ID_LO_MASK GENMASK(31, 22)
-#define ENA_ETH_IO_TX_DESC_ADDR_HI_MASK GENMASK(15, 0)
-#define ENA_ETH_IO_TX_DESC_HEADER_LENGTH_SHIFT 24
-#define ENA_ETH_IO_TX_DESC_HEADER_LENGTH_MASK GENMASK(31, 24)
+#define ENA_ETH_IO_TX_DESC_LENGTH_MASK                      GENMASK(15, 0)
+#define ENA_ETH_IO_TX_DESC_REQ_ID_HI_SHIFT                  16
+#define ENA_ETH_IO_TX_DESC_REQ_ID_HI_MASK                   GENMASK(21, 16)
+#define ENA_ETH_IO_TX_DESC_META_DESC_SHIFT                  23
+#define ENA_ETH_IO_TX_DESC_META_DESC_MASK                   BIT(23)
+#define ENA_ETH_IO_TX_DESC_PHASE_SHIFT                      24
+#define ENA_ETH_IO_TX_DESC_PHASE_MASK                       BIT(24)
+#define ENA_ETH_IO_TX_DESC_FIRST_SHIFT                      26
+#define ENA_ETH_IO_TX_DESC_FIRST_MASK                       BIT(26)
+#define ENA_ETH_IO_TX_DESC_LAST_SHIFT                       27
+#define ENA_ETH_IO_TX_DESC_LAST_MASK                        BIT(27)
+#define ENA_ETH_IO_TX_DESC_COMP_REQ_SHIFT                   28
+#define ENA_ETH_IO_TX_DESC_COMP_REQ_MASK                    BIT(28)
+#define ENA_ETH_IO_TX_DESC_L3_PROTO_IDX_MASK                GENMASK(3, 0)
+#define ENA_ETH_IO_TX_DESC_DF_SHIFT                         4
+#define ENA_ETH_IO_TX_DESC_DF_MASK                          BIT(4)
+#define ENA_ETH_IO_TX_DESC_TSO_EN_SHIFT                     7
+#define ENA_ETH_IO_TX_DESC_TSO_EN_MASK                      BIT(7)
+#define ENA_ETH_IO_TX_DESC_L4_PROTO_IDX_SHIFT               8
+#define ENA_ETH_IO_TX_DESC_L4_PROTO_IDX_MASK                GENMASK(12, 8)
+#define ENA_ETH_IO_TX_DESC_L3_CSUM_EN_SHIFT                 13
+#define ENA_ETH_IO_TX_DESC_L3_CSUM_EN_MASK                  BIT(13)
+#define ENA_ETH_IO_TX_DESC_L4_CSUM_EN_SHIFT                 14
+#define ENA_ETH_IO_TX_DESC_L4_CSUM_EN_MASK                  BIT(14)
+#define ENA_ETH_IO_TX_DESC_ETHERNET_FCS_DIS_SHIFT           15
+#define ENA_ETH_IO_TX_DESC_ETHERNET_FCS_DIS_MASK            BIT(15)
+#define ENA_ETH_IO_TX_DESC_L4_CSUM_PARTIAL_SHIFT            17
+#define ENA_ETH_IO_TX_DESC_L4_CSUM_PARTIAL_MASK             BIT(17)
+#define ENA_ETH_IO_TX_DESC_REQ_ID_LO_SHIFT                  22
+#define ENA_ETH_IO_TX_DESC_REQ_ID_LO_MASK                   GENMASK(31, 22)
+#define ENA_ETH_IO_TX_DESC_ADDR_HI_MASK                     GENMASK(15, 0)
+#define ENA_ETH_IO_TX_DESC_HEADER_LENGTH_SHIFT              24
+#define ENA_ETH_IO_TX_DESC_HEADER_LENGTH_MASK               GENMASK(31, 24)
 
 /* tx_meta_desc */
-#define ENA_ETH_IO_TX_META_DESC_REQ_ID_LO_MASK GENMASK(9, 0)
-#define ENA_ETH_IO_TX_META_DESC_EXT_VALID_SHIFT 14
-#define ENA_ETH_IO_TX_META_DESC_EXT_VALID_MASK BIT(14)
-#define ENA_ETH_IO_TX_META_DESC_MSS_HI_SHIFT 16
-#define ENA_ETH_IO_TX_META_DESC_MSS_HI_MASK GENMASK(19, 16)
-#define ENA_ETH_IO_TX_META_DESC_ETH_META_TYPE_SHIFT 20
-#define ENA_ETH_IO_TX_META_DESC_ETH_META_TYPE_MASK BIT(20)
-#define ENA_ETH_IO_TX_META_DESC_META_STORE_SHIFT 21
-#define ENA_ETH_IO_TX_META_DESC_META_STORE_MASK BIT(21)
-#define ENA_ETH_IO_TX_META_DESC_META_DESC_SHIFT 23
-#define ENA_ETH_IO_TX_META_DESC_META_DESC_MASK BIT(23)
-#define ENA_ETH_IO_TX_META_DESC_PHASE_SHIFT 24
-#define ENA_ETH_IO_TX_META_DESC_PHASE_MASK BIT(24)
-#define ENA_ETH_IO_TX_META_DESC_FIRST_SHIFT 26
-#define ENA_ETH_IO_TX_META_DESC_FIRST_MASK BIT(26)
-#define ENA_ETH_IO_TX_META_DESC_LAST_SHIFT 27
-#define ENA_ETH_IO_TX_META_DESC_LAST_MASK BIT(27)
-#define ENA_ETH_IO_TX_META_DESC_COMP_REQ_SHIFT 28
-#define ENA_ETH_IO_TX_META_DESC_COMP_REQ_MASK BIT(28)
-#define ENA_ETH_IO_TX_META_DESC_REQ_ID_HI_MASK GENMASK(5, 0)
-#define ENA_ETH_IO_TX_META_DESC_L3_HDR_LEN_MASK GENMASK(7, 0)
-#define ENA_ETH_IO_TX_META_DESC_L3_HDR_OFF_SHIFT 8
-#define ENA_ETH_IO_TX_META_DESC_L3_HDR_OFF_MASK GENMASK(15, 8)
-#define ENA_ETH_IO_TX_META_DESC_L4_HDR_LEN_IN_WORDS_SHIFT 16
-#define ENA_ETH_IO_TX_META_DESC_L4_HDR_LEN_IN_WORDS_MASK GENMASK(21, 16)
-#define ENA_ETH_IO_TX_META_DESC_MSS_LO_SHIFT 22
-#define ENA_ETH_IO_TX_META_DESC_MSS_LO_MASK GENMASK(31, 22)
+#define ENA_ETH_IO_TX_META_DESC_REQ_ID_LO_MASK              GENMASK(9, 0)
+#define ENA_ETH_IO_TX_META_DESC_EXT_VALID_SHIFT             14
+#define ENA_ETH_IO_TX_META_DESC_EXT_VALID_MASK              BIT(14)
+#define ENA_ETH_IO_TX_META_DESC_MSS_HI_SHIFT                16
+#define ENA_ETH_IO_TX_META_DESC_MSS_HI_MASK                 GENMASK(19, 16)
+#define ENA_ETH_IO_TX_META_DESC_ETH_META_TYPE_SHIFT         20
+#define ENA_ETH_IO_TX_META_DESC_ETH_META_TYPE_MASK          BIT(20)
+#define ENA_ETH_IO_TX_META_DESC_META_STORE_SHIFT            21
+#define ENA_ETH_IO_TX_META_DESC_META_STORE_MASK             BIT(21)
+#define ENA_ETH_IO_TX_META_DESC_META_DESC_SHIFT             23
+#define ENA_ETH_IO_TX_META_DESC_META_DESC_MASK              BIT(23)
+#define ENA_ETH_IO_TX_META_DESC_PHASE_SHIFT                 24
+#define ENA_ETH_IO_TX_META_DESC_PHASE_MASK                  BIT(24)
+#define ENA_ETH_IO_TX_META_DESC_FIRST_SHIFT                 26
+#define ENA_ETH_IO_TX_META_DESC_FIRST_MASK                  BIT(26)
+#define ENA_ETH_IO_TX_META_DESC_LAST_SHIFT                  27
+#define ENA_ETH_IO_TX_META_DESC_LAST_MASK                   BIT(27)
+#define ENA_ETH_IO_TX_META_DESC_COMP_REQ_SHIFT              28
+#define ENA_ETH_IO_TX_META_DESC_COMP_REQ_MASK               BIT(28)
+#define ENA_ETH_IO_TX_META_DESC_REQ_ID_HI_MASK              GENMASK(5, 0)
+#define ENA_ETH_IO_TX_META_DESC_L3_HDR_LEN_MASK             GENMASK(7, 0)
+#define ENA_ETH_IO_TX_META_DESC_L3_HDR_OFF_SHIFT            8
+#define ENA_ETH_IO_TX_META_DESC_L3_HDR_OFF_MASK             GENMASK(15, 8)
+#define ENA_ETH_IO_TX_META_DESC_L4_HDR_LEN_IN_WORDS_SHIFT   16
+#define ENA_ETH_IO_TX_META_DESC_L4_HDR_LEN_IN_WORDS_MASK    GENMASK(21, 16)
+#define ENA_ETH_IO_TX_META_DESC_MSS_LO_SHIFT                22
+#define ENA_ETH_IO_TX_META_DESC_MSS_LO_MASK                 GENMASK(31, 22)
 
 /* tx_cdesc */
-#define ENA_ETH_IO_TX_CDESC_PHASE_MASK BIT(0)
+#define ENA_ETH_IO_TX_CDESC_PHASE_MASK                      BIT(0)
 
 /* rx_desc */
-#define ENA_ETH_IO_RX_DESC_PHASE_MASK BIT(0)
-#define ENA_ETH_IO_RX_DESC_FIRST_SHIFT 2
-#define ENA_ETH_IO_RX_DESC_FIRST_MASK BIT(2)
-#define ENA_ETH_IO_RX_DESC_LAST_SHIFT 3
-#define ENA_ETH_IO_RX_DESC_LAST_MASK BIT(3)
-#define ENA_ETH_IO_RX_DESC_COMP_REQ_SHIFT 4
-#define ENA_ETH_IO_RX_DESC_COMP_REQ_MASK BIT(4)
+#define ENA_ETH_IO_RX_DESC_PHASE_MASK                       BIT(0)
+#define ENA_ETH_IO_RX_DESC_FIRST_SHIFT                      2
+#define ENA_ETH_IO_RX_DESC_FIRST_MASK                       BIT(2)
+#define ENA_ETH_IO_RX_DESC_LAST_SHIFT                       3
+#define ENA_ETH_IO_RX_DESC_LAST_MASK                        BIT(3)
+#define ENA_ETH_IO_RX_DESC_COMP_REQ_SHIFT                   4
+#define ENA_ETH_IO_RX_DESC_COMP_REQ_MASK                    BIT(4)
 
 /* rx_cdesc_base */
-#define ENA_ETH_IO_RX_CDESC_BASE_L3_PROTO_IDX_MASK GENMASK(4, 0)
-#define ENA_ETH_IO_RX_CDESC_BASE_SRC_VLAN_CNT_SHIFT 5
-#define ENA_ETH_IO_RX_CDESC_BASE_SRC_VLAN_CNT_MASK GENMASK(6, 5)
-#define ENA_ETH_IO_RX_CDESC_BASE_L4_PROTO_IDX_SHIFT 8
-#define ENA_ETH_IO_RX_CDESC_BASE_L4_PROTO_IDX_MASK GENMASK(12, 8)
-#define ENA_ETH_IO_RX_CDESC_BASE_L3_CSUM_ERR_SHIFT 13
-#define ENA_ETH_IO_RX_CDESC_BASE_L3_CSUM_ERR_MASK BIT(13)
-#define ENA_ETH_IO_RX_CDESC_BASE_L4_CSUM_ERR_SHIFT 14
-#define ENA_ETH_IO_RX_CDESC_BASE_L4_CSUM_ERR_MASK BIT(14)
-#define ENA_ETH_IO_RX_CDESC_BASE_IPV4_FRAG_SHIFT 15
-#define ENA_ETH_IO_RX_CDESC_BASE_IPV4_FRAG_MASK BIT(15)
-#define ENA_ETH_IO_RX_CDESC_BASE_PHASE_SHIFT 24
-#define ENA_ETH_IO_RX_CDESC_BASE_PHASE_MASK BIT(24)
-#define ENA_ETH_IO_RX_CDESC_BASE_L3_CSUM2_SHIFT 25
-#define ENA_ETH_IO_RX_CDESC_BASE_L3_CSUM2_MASK BIT(25)
-#define ENA_ETH_IO_RX_CDESC_BASE_FIRST_SHIFT 26
-#define ENA_ETH_IO_RX_CDESC_BASE_FIRST_MASK BIT(26)
-#define ENA_ETH_IO_RX_CDESC_BASE_LAST_SHIFT 27
-#define ENA_ETH_IO_RX_CDESC_BASE_LAST_MASK BIT(27)
-#define ENA_ETH_IO_RX_CDESC_BASE_BUFFER_SHIFT 30
-#define ENA_ETH_IO_RX_CDESC_BASE_BUFFER_MASK BIT(30)
+#define ENA_ETH_IO_RX_CDESC_BASE_L3_PROTO_IDX_MASK          GENMASK(4, 0)
+#define ENA_ETH_IO_RX_CDESC_BASE_SRC_VLAN_CNT_SHIFT         5
+#define ENA_ETH_IO_RX_CDESC_BASE_SRC_VLAN_CNT_MASK          GENMASK(6, 5)
+#define ENA_ETH_IO_RX_CDESC_BASE_L4_PROTO_IDX_SHIFT         8
+#define ENA_ETH_IO_RX_CDESC_BASE_L4_PROTO_IDX_MASK          GENMASK(12, 8)
+#define ENA_ETH_IO_RX_CDESC_BASE_L3_CSUM_ERR_SHIFT          13
+#define ENA_ETH_IO_RX_CDESC_BASE_L3_CSUM_ERR_MASK           BIT(13)
+#define ENA_ETH_IO_RX_CDESC_BASE_L4_CSUM_ERR_SHIFT          14
+#define ENA_ETH_IO_RX_CDESC_BASE_L4_CSUM_ERR_MASK           BIT(14)
+#define ENA_ETH_IO_RX_CDESC_BASE_IPV4_FRAG_SHIFT            15
+#define ENA_ETH_IO_RX_CDESC_BASE_IPV4_FRAG_MASK             BIT(15)
+#define ENA_ETH_IO_RX_CDESC_BASE_L4_CSUM_CHECKED_SHIFT      16
+#define ENA_ETH_IO_RX_CDESC_BASE_L4_CSUM_CHECKED_MASK       BIT(16)
+#define ENA_ETH_IO_RX_CDESC_BASE_PHASE_SHIFT                24
+#define ENA_ETH_IO_RX_CDESC_BASE_PHASE_MASK                 BIT(24)
+#define ENA_ETH_IO_RX_CDESC_BASE_L3_CSUM2_SHIFT             25
+#define ENA_ETH_IO_RX_CDESC_BASE_L3_CSUM2_MASK              BIT(25)
+#define ENA_ETH_IO_RX_CDESC_BASE_FIRST_SHIFT                26
+#define ENA_ETH_IO_RX_CDESC_BASE_FIRST_MASK                 BIT(26)
+#define ENA_ETH_IO_RX_CDESC_BASE_LAST_SHIFT                 27
+#define ENA_ETH_IO_RX_CDESC_BASE_LAST_MASK                  BIT(27)
+#define ENA_ETH_IO_RX_CDESC_BASE_BUFFER_SHIFT               30
+#define ENA_ETH_IO_RX_CDESC_BASE_BUFFER_MASK                BIT(30)
 
 /* intr_reg */
-#define ENA_ETH_IO_INTR_REG_RX_INTR_DELAY_MASK GENMASK(14, 0)
-#define ENA_ETH_IO_INTR_REG_TX_INTR_DELAY_SHIFT 15
-#define ENA_ETH_IO_INTR_REG_TX_INTR_DELAY_MASK GENMASK(29, 15)
-#define ENA_ETH_IO_INTR_REG_INTR_UNMASK_SHIFT 30
-#define ENA_ETH_IO_INTR_REG_INTR_UNMASK_MASK BIT(30)
+#define ENA_ETH_IO_INTR_REG_RX_INTR_DELAY_MASK              GENMASK(14, 0)
+#define ENA_ETH_IO_INTR_REG_TX_INTR_DELAY_SHIFT             15
+#define ENA_ETH_IO_INTR_REG_TX_INTR_DELAY_MASK              GENMASK(29, 15)
+#define ENA_ETH_IO_INTR_REG_INTR_UNMASK_SHIFT               30
+#define ENA_ETH_IO_INTR_REG_INTR_UNMASK_MASK                BIT(30)
 
 /* numa_node_cfg_reg */
-#define ENA_ETH_IO_NUMA_NODE_CFG_REG_NUMA_MASK GENMASK(7, 0)
-#define ENA_ETH_IO_NUMA_NODE_CFG_REG_ENABLED_SHIFT 31
-#define ENA_ETH_IO_NUMA_NODE_CFG_REG_ENABLED_MASK BIT(31)
+#define ENA_ETH_IO_NUMA_NODE_CFG_REG_NUMA_MASK              GENMASK(7, 0)
+#define ENA_ETH_IO_NUMA_NODE_CFG_REG_ENABLED_SHIFT          31
+#define ENA_ETH_IO_NUMA_NODE_CFG_REG_ENABLED_MASK           BIT(31)
 
-#endif /*_ENA_ETH_IO_H_ */
+#endif /* _ENA_ETH_IO_H_ */
diff --git a/drivers/net/ethernet/amazon/ena/ena_ethtool.c b/drivers/net/ethernet/amazon/ena/ena_ethtool.c
index 67b2338f8fb3..fe3479b84a1f 100644
--- a/drivers/net/ethernet/amazon/ena/ena_ethtool.c
+++ b/drivers/net/ethernet/amazon/ena/ena_ethtool.c
@@ -1,52 +1,38 @@
+// SPDX-License-Identifier: GPL-2.0 OR Linux-OpenIB
 /*
- * Copyright 2015 Amazon.com, Inc. or its affiliates.
- *
- * This software is available to you under a choice of one of two
- * licenses.  You may choose to be licensed under the terms of the GNU
- * General Public License (GPL) Version 2, available from the file
- * COPYING in the main directory of this source tree, or the
- * BSD license below:
- *
- *     Redistribution and use in source and binary forms, with or
- *     without modification, are permitted provided that the following
- *     conditions are met:
- *
- *      - Redistributions of source code must retain the above
- *        copyright notice, this list of conditions and the following
- *        disclaimer.
- *
- *      - Redistributions in binary form must reproduce the above
- *        copyright notice, this list of conditions and the following
- *        disclaimer in the documentation and/or other materials
- *        provided with the distribution.
- *
- * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND,
- * EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF
- * MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND
- * NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS
- * BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN
- * ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN
- * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
- * SOFTWARE.
+ * Copyright 2015-2020 Amazon.com, Inc. or its affiliates. All rights reserved.
  */
 
+#include <linux/ethtool.h>
 #include <linux/pci.h>
+#include <linux/net_tstamp.h>
 
 #include "ena_netdev.h"
+#include "ena_xdp.h"
+#include "ena_phc.h"
 
 struct ena_stats {
 	char name[ETH_GSTRING_LEN];
 	int stat_offset;
 };
 
+struct ena_hw_metrics {
+	char name[ETH_GSTRING_LEN];
+};
+
 #define ENA_STAT_ENA_COM_ENTRY(stat) { \
 	.name = #stat, \
-	.stat_offset = offsetof(struct ena_com_stats_admin, stat) \
+	.stat_offset = offsetof(struct ena_com_stats_admin, stat) / sizeof(u64) \
 }
 
 #define ENA_STAT_ENTRY(stat, stat_type) { \
 	.name = #stat, \
-	.stat_offset = offsetof(struct ena_stats_##stat_type, stat) \
+	.stat_offset = offsetof(struct ena_stats_##stat_type, stat) / sizeof(u64) \
+}
+
+#define ENA_STAT_HW_ENTRY(stat, stat_type) { \
+	.name = #stat, \
+	.stat_offset = offsetof(struct ena_admin_##stat_type, stat) / sizeof(u64) \
 }
 
 #define ENA_STAT_RX_ENTRY(stat) \
@@ -58,14 +44,58 @@ struct ena_stats {
 #define ENA_STAT_GLOBAL_ENTRY(stat) \
 	ENA_STAT_ENTRY(stat, dev)
 
+#define ENA_STAT_ENI_ENTRY(stat) \
+	ENA_STAT_HW_ENTRY(stat, eni_stats)
+
+#define ENA_STAT_ENA_SRD_ENTRY(stat) \
+	ENA_STAT_HW_ENTRY(stat, ena_srd_stats)
+
+#define ENA_STAT_ENA_SRD_MODE_ENTRY(stat) { \
+	.name = #stat, \
+	.stat_offset = offsetof(struct ena_admin_ena_srd_info, flags) / sizeof(u64) \
+}
+
+#define ENA_METRIC_ENI_ENTRY(stat) { \
+	.name = #stat \
+}
+
 static const struct ena_stats ena_stats_global_strings[] = {
 	ENA_STAT_GLOBAL_ENTRY(tx_timeout),
-	ENA_STAT_GLOBAL_ENTRY(io_suspend),
-	ENA_STAT_GLOBAL_ENTRY(io_resume),
+	ENA_STAT_GLOBAL_ENTRY(suspend),
+	ENA_STAT_GLOBAL_ENTRY(resume),
 	ENA_STAT_GLOBAL_ENTRY(wd_expired),
 	ENA_STAT_GLOBAL_ENTRY(interface_up),
 	ENA_STAT_GLOBAL_ENTRY(interface_down),
 	ENA_STAT_GLOBAL_ENTRY(admin_q_pause),
+	ENA_STAT_GLOBAL_ENTRY(reset_fail),
+};
+
+/* A partial list of hw stats. Used when admin command
+ * with type ENA_ADMIN_GET_STATS_TYPE_CUSTOMER_METRICS is not supported
+ */
+static const struct ena_stats ena_stats_eni_strings[] = {
+	ENA_STAT_ENI_ENTRY(bw_in_allowance_exceeded),
+	ENA_STAT_ENI_ENTRY(bw_out_allowance_exceeded),
+	ENA_STAT_ENI_ENTRY(pps_allowance_exceeded),
+	ENA_STAT_ENI_ENTRY(conntrack_allowance_exceeded),
+	ENA_STAT_ENI_ENTRY(linklocal_allowance_exceeded),
+};
+
+static const struct ena_hw_metrics ena_hw_stats_strings[] = {
+	ENA_METRIC_ENI_ENTRY(bw_in_allowance_exceeded),
+	ENA_METRIC_ENI_ENTRY(bw_out_allowance_exceeded),
+	ENA_METRIC_ENI_ENTRY(pps_allowance_exceeded),
+	ENA_METRIC_ENI_ENTRY(conntrack_allowance_exceeded),
+	ENA_METRIC_ENI_ENTRY(linklocal_allowance_exceeded),
+	ENA_METRIC_ENI_ENTRY(conntrack_allowance_available),
+};
+
+static const struct ena_stats ena_srd_info_strings[] = {
+	ENA_STAT_ENA_SRD_MODE_ENTRY(ena_srd_mode),
+	ENA_STAT_ENA_SRD_ENTRY(ena_srd_tx_pkts),
+	ENA_STAT_ENA_SRD_ENTRY(ena_srd_eligible_tx_pkts),
+	ENA_STAT_ENA_SRD_ENTRY(ena_srd_rx_pkts),
+	ENA_STAT_ENA_SRD_ENTRY(ena_srd_resource_utilization)
 };
 
 static const struct ena_stats ena_stats_tx_strings[] = {
@@ -80,20 +110,32 @@ static const struct ena_stats ena_stats_tx_strings[] = {
 	ENA_STAT_TX_ENTRY(tx_poll),
 	ENA_STAT_TX_ENTRY(doorbells),
 	ENA_STAT_TX_ENTRY(prepare_ctx_err),
-	ENA_STAT_TX_ENTRY(missing_tx_comp),
 	ENA_STAT_TX_ENTRY(bad_req_id),
+	ENA_STAT_TX_ENTRY(llq_buffer_copy),
+	ENA_STAT_TX_ENTRY(missed_tx),
+	ENA_STAT_TX_ENTRY(unmask_interrupt),
 };
 
 static const struct ena_stats ena_stats_rx_strings[] = {
 	ENA_STAT_RX_ENTRY(cnt),
 	ENA_STAT_RX_ENTRY(bytes),
+	ENA_STAT_RX_ENTRY(rx_copybreak_pkt),
+	ENA_STAT_RX_ENTRY(csum_good),
 	ENA_STAT_RX_ENTRY(refil_partial),
-	ENA_STAT_RX_ENTRY(bad_csum),
+	ENA_STAT_RX_ENTRY(csum_bad),
 	ENA_STAT_RX_ENTRY(page_alloc_fail),
 	ENA_STAT_RX_ENTRY(skb_alloc_fail),
 	ENA_STAT_RX_ENTRY(dma_mapping_err),
 	ENA_STAT_RX_ENTRY(bad_desc_num),
-	ENA_STAT_RX_ENTRY(rx_copybreak_pkt),
+	ENA_STAT_RX_ENTRY(bad_req_id),
+	ENA_STAT_RX_ENTRY(empty_rx_ring),
+	ENA_STAT_RX_ENTRY(csum_unchecked),
+	ENA_STAT_RX_ENTRY(xdp_aborted),
+	ENA_STAT_RX_ENTRY(xdp_drop),
+	ENA_STAT_RX_ENTRY(xdp_pass),
+	ENA_STAT_RX_ENTRY(xdp_tx),
+	ENA_STAT_RX_ENTRY(xdp_invalid),
+	ENA_STAT_RX_ENTRY(xdp_redirect),
 };
 
 static const struct ena_stats ena_stats_ena_com_strings[] = {
@@ -104,10 +146,13 @@ static const struct ena_stats ena_stats_ena_com_strings[] = {
 	ENA_STAT_ENA_COM_ENTRY(no_completion),
 };
 
-#define ENA_STATS_ARRAY_GLOBAL	ARRAY_SIZE(ena_stats_global_strings)
-#define ENA_STATS_ARRAY_TX	ARRAY_SIZE(ena_stats_tx_strings)
-#define ENA_STATS_ARRAY_RX	ARRAY_SIZE(ena_stats_rx_strings)
-#define ENA_STATS_ARRAY_ENA_COM	ARRAY_SIZE(ena_stats_ena_com_strings)
+#define ENA_STATS_ARRAY_GLOBAL		ARRAY_SIZE(ena_stats_global_strings)
+#define ENA_STATS_ARRAY_TX		ARRAY_SIZE(ena_stats_tx_strings)
+#define ENA_STATS_ARRAY_RX		ARRAY_SIZE(ena_stats_rx_strings)
+#define ENA_STATS_ARRAY_ENA_COM		ARRAY_SIZE(ena_stats_ena_com_strings)
+#define ENA_STATS_ARRAY_ENI		ARRAY_SIZE(ena_stats_eni_strings)
+#define ENA_STATS_ARRAY_ENA_SRD		ARRAY_SIZE(ena_srd_info_strings)
+#define ENA_METRICS_ARRAY_ENI		ARRAY_SIZE(ena_hw_stats_strings)
 
 static void ena_safe_update_stat(u64 *src, u64 *dst,
 				 struct u64_stats_sync *syncp)
@@ -115,9 +160,60 @@ static void ena_safe_update_stat(u64 *src, u64 *dst,
 	unsigned int start;
 
 	do {
-		start = u64_stats_fetch_begin_irq(syncp);
+		start = u64_stats_fetch_begin(syncp);
 		*(dst) = *src;
-	} while (u64_stats_fetch_retry_irq(syncp, start));
+	} while (u64_stats_fetch_retry(syncp, start));
+}
+
+static void ena_metrics_stats(struct ena_adapter *adapter, u64 **data)
+{
+	struct ena_com_dev *dev = adapter->ena_dev;
+	const struct ena_stats *ena_stats;
+	u64 *ptr;
+	int i;
+
+	if (ena_com_get_cap(dev, ENA_ADMIN_CUSTOMER_METRICS)) {
+		u32 supported_metrics_count;
+		int len;
+
+		supported_metrics_count = ena_com_get_customer_metric_count(dev);
+		len = supported_metrics_count * sizeof(u64);
+
+		/* Fill the data buffer, and advance its pointer */
+		ena_com_get_customer_metrics(dev, (char *)(*data), len);
+		(*data) += supported_metrics_count;
+
+	} else if (ena_com_get_cap(dev, ENA_ADMIN_ENI_STATS)) {
+		ena_com_get_eni_stats(dev, &adapter->eni_stats);
+		/* Updating regardless of rc - once we told ethtool how many stats we have
+		 * it will print that much stats. We can't leave holes in the stats
+		 */
+		for (i = 0; i < ENA_STATS_ARRAY_ENI; i++) {
+			ena_stats = &ena_stats_eni_strings[i];
+
+			ptr = (u64 *)&adapter->eni_stats +
+				ena_stats->stat_offset;
+
+			ena_safe_update_stat(ptr, (*data)++, &adapter->syncp);
+		}
+	}
+
+	if (ena_com_get_cap(dev, ENA_ADMIN_ENA_SRD_INFO)) {
+		ena_com_get_ena_srd_info(dev, &adapter->ena_srd_info);
+		/* Get ENA SRD mode */
+		ptr = (u64 *)&adapter->ena_srd_info;
+		ena_safe_update_stat(ptr, (*data)++, &adapter->syncp);
+		for (i = 1; i < ENA_STATS_ARRAY_ENA_SRD; i++) {
+			ena_stats = &ena_srd_info_strings[i];
+			/* Wrapped within an outer struct - need to accommodate an
+			 * additional offset of the ENA SRD mode that was already processed
+			 */
+			ptr = (u64 *)&adapter->ena_srd_info +
+				ena_stats->stat_offset + 1;
+
+			ena_safe_update_stat(ptr, (*data)++, &adapter->syncp);
+		}
+	}
 }
 
 static void ena_queue_stats(struct ena_adapter *adapter, u64 **data)
@@ -128,29 +224,30 @@ static void ena_queue_stats(struct ena_adapter *adapter, u64 **data)
 	u64 *ptr;
 	int i, j;
 
-	for (i = 0; i < adapter->num_queues; i++) {
+	for (i = 0; i < adapter->num_io_queues + adapter->xdp_num_queues; i++) {
 		/* Tx stats */
 		ring = &adapter->tx_ring[i];
 
 		for (j = 0; j < ENA_STATS_ARRAY_TX; j++) {
 			ena_stats = &ena_stats_tx_strings[j];
 
-			ptr = (u64 *)((uintptr_t)&ring->tx_stats +
-				(uintptr_t)ena_stats->stat_offset);
+			ptr = (u64 *)&ring->tx_stats + ena_stats->stat_offset;
 
 			ena_safe_update_stat(ptr, (*data)++, &ring->syncp);
 		}
+		/* XDP TX queues don't have a RX queue counterpart */
+		if (!ENA_IS_XDP_INDEX(adapter, i)) {
+			/* Rx stats */
+			ring = &adapter->rx_ring[i];
 
-		/* Rx stats */
-		ring = &adapter->rx_ring[i];
+			for (j = 0; j < ENA_STATS_ARRAY_RX; j++) {
+				ena_stats = &ena_stats_rx_strings[j];
 
-		for (j = 0; j < ENA_STATS_ARRAY_RX; j++) {
-			ena_stats = &ena_stats_rx_strings[j];
+				ptr = (u64 *)&ring->rx_stats +
+					ena_stats->stat_offset;
 
-			ptr = (u64 *)((uintptr_t)&ring->rx_stats +
-				(uintptr_t)ena_stats->stat_offset);
-
-			ena_safe_update_stat(ptr, (*data)++, &ring->syncp);
+				ena_safe_update_stat(ptr, (*data)++, &ring->syncp);
+			}
 		}
 	}
 }
@@ -158,24 +255,23 @@ static void ena_queue_stats(struct ena_adapter *adapter, u64 **data)
 static void ena_dev_admin_queue_stats(struct ena_adapter *adapter, u64 **data)
 {
 	const struct ena_stats *ena_stats;
-	u32 *ptr;
+	u64 *ptr;
 	int i;
 
 	for (i = 0; i < ENA_STATS_ARRAY_ENA_COM; i++) {
 		ena_stats = &ena_stats_ena_com_strings[i];
 
-		ptr = (u32 *)((uintptr_t)&adapter->ena_dev->admin_queue.stats +
-			(uintptr_t)ena_stats->stat_offset);
+		ptr = (u64 *)&adapter->ena_dev->admin_queue.stats +
+			ena_stats->stat_offset;
 
 		*(*data)++ = *ptr;
 	}
 }
 
-static void ena_get_ethtool_stats(struct net_device *netdev,
-				  struct ethtool_stats *stats,
-				  u64 *data)
+static void ena_get_stats(struct ena_adapter *adapter,
+			  u64 *data,
+			  bool hw_stats_needed)
 {
-	struct ena_adapter *adapter = netdev_priv(netdev);
 	const struct ena_stats *ena_stats;
 	u64 *ptr;
 	int i;
@@ -183,48 +279,129 @@ static void ena_get_ethtool_stats(struct net_device *netdev,
 	for (i = 0; i < ENA_STATS_ARRAY_GLOBAL; i++) {
 		ena_stats = &ena_stats_global_strings[i];
 
-		ptr = (u64 *)((uintptr_t)&adapter->dev_stats +
-			(uintptr_t)ena_stats->stat_offset);
+		ptr = (u64 *)&adapter->dev_stats + ena_stats->stat_offset;
 
 		ena_safe_update_stat(ptr, data++, &adapter->syncp);
 	}
 
+	if (hw_stats_needed)
+		ena_metrics_stats(adapter, &data);
+
 	ena_queue_stats(adapter, &data);
 	ena_dev_admin_queue_stats(adapter, &data);
 }
 
-int ena_get_sset_count(struct net_device *netdev, int sset)
+static void ena_get_ethtool_stats(struct net_device *netdev,
+				  struct ethtool_stats *stats,
+				  u64 *data)
 {
 	struct ena_adapter *adapter = netdev_priv(netdev);
 
-	if (sset != ETH_SS_STATS)
-		return -EOPNOTSUPP;
+	ena_get_stats(adapter, data, true);
+}
 
-	return  adapter->num_queues * (ENA_STATS_ARRAY_TX + ENA_STATS_ARRAY_RX)
+static int ena_get_ts_info(struct net_device *netdev,
+			   struct kernel_ethtool_ts_info *info)
+{
+	struct ena_adapter *adapter = netdev_priv(netdev);
+
+	info->so_timestamping = SOF_TIMESTAMPING_TX_SOFTWARE;
+
+	info->phc_index = ena_phc_get_index(adapter);
+
+	return 0;
+}
+
+static int ena_get_sw_stats_count(struct ena_adapter *adapter)
+{
+	return adapter->num_io_queues * (ENA_STATS_ARRAY_TX + ENA_STATS_ARRAY_RX)
+		+ adapter->xdp_num_queues * ENA_STATS_ARRAY_TX
 		+ ENA_STATS_ARRAY_GLOBAL + ENA_STATS_ARRAY_ENA_COM;
 }
 
+static int ena_get_hw_stats_count(struct ena_adapter *adapter)
+{
+	struct ena_com_dev *dev = adapter->ena_dev;
+	int count;
+
+	count = ENA_STATS_ARRAY_ENA_SRD * ena_com_get_cap(dev, ENA_ADMIN_ENA_SRD_INFO);
+
+	if (ena_com_get_cap(dev, ENA_ADMIN_CUSTOMER_METRICS))
+		count += ena_com_get_customer_metric_count(dev);
+	else if (ena_com_get_cap(dev, ENA_ADMIN_ENI_STATS))
+		count += ENA_STATS_ARRAY_ENI;
+
+	return count;
+}
+
+int ena_get_sset_count(struct net_device *netdev, int sset)
+{
+	struct ena_adapter *adapter = netdev_priv(netdev);
+
+	switch (sset) {
+	case ETH_SS_STATS:
+		return ena_get_sw_stats_count(adapter) +
+		       ena_get_hw_stats_count(adapter);
+	}
+
+	return -EOPNOTSUPP;
+}
+
+static void ena_metrics_stats_strings(struct ena_adapter *adapter, u8 **data)
+{
+	struct ena_com_dev *dev = adapter->ena_dev;
+	const struct ena_hw_metrics *ena_metrics;
+	const struct ena_stats *ena_stats;
+	int i;
+
+	if (ena_com_get_cap(dev, ENA_ADMIN_CUSTOMER_METRICS)) {
+		for (i = 0; i < ENA_METRICS_ARRAY_ENI; i++) {
+			if (ena_com_get_customer_metric_support(dev, i)) {
+				ena_metrics = &ena_hw_stats_strings[i];
+				ethtool_puts(data, ena_metrics->name);
+			}
+		}
+	} else if (ena_com_get_cap(dev, ENA_ADMIN_ENI_STATS)) {
+		for (i = 0; i < ENA_STATS_ARRAY_ENI; i++) {
+			ena_stats = &ena_stats_eni_strings[i];
+			ethtool_puts(data, ena_stats->name);
+		}
+	}
+
+	if (ena_com_get_cap(dev, ENA_ADMIN_ENA_SRD_INFO)) {
+		for (i = 0; i < ENA_STATS_ARRAY_ENA_SRD; i++) {
+			ena_stats = &ena_srd_info_strings[i];
+			ethtool_puts(data, ena_stats->name);
+		}
+	}
+}
+
 static void ena_queue_strings(struct ena_adapter *adapter, u8 **data)
 {
 	const struct ena_stats *ena_stats;
+	bool is_xdp;
 	int i, j;
 
-	for (i = 0; i < adapter->num_queues; i++) {
+	for (i = 0; i < adapter->num_io_queues + adapter->xdp_num_queues; i++) {
+		is_xdp = ENA_IS_XDP_INDEX(adapter, i);
 		/* Tx stats */
 		for (j = 0; j < ENA_STATS_ARRAY_TX; j++) {
 			ena_stats = &ena_stats_tx_strings[j];
 
-			snprintf(*data, ETH_GSTRING_LEN,
-				 "queue_%u_tx_%s", i, ena_stats->name);
-			 (*data) += ETH_GSTRING_LEN;
+			ethtool_sprintf(data,
+					"queue_%u_%s_%s", i,
+					is_xdp ? "xdp_tx" : "tx",
+					ena_stats->name);
 		}
-		/* Rx stats */
+
+		/* In XDP there isn't an RX queue counterpart */
+		if (is_xdp)
+			continue;
+
 		for (j = 0; j < ENA_STATS_ARRAY_RX; j++) {
 			ena_stats = &ena_stats_rx_strings[j];
 
-			snprintf(*data, ETH_GSTRING_LEN,
-				 "queue_%u_rx_%s", i, ena_stats->name);
-			(*data) += ETH_GSTRING_LEN;
+			ethtool_sprintf(data, "queue_%u_rx_%s", i, ena_stats->name);
 		}
 	}
 }
@@ -237,32 +414,43 @@ static void ena_com_dev_strings(u8 **data)
 	for (i = 0; i < ENA_STATS_ARRAY_ENA_COM; i++) {
 		ena_stats = &ena_stats_ena_com_strings[i];
 
-		snprintf(*data, ETH_GSTRING_LEN,
-			 "ena_admin_q_%s", ena_stats->name);
-		(*data) += ETH_GSTRING_LEN;
+		ethtool_sprintf(data,
+				"ena_admin_q_%s", ena_stats->name);
 	}
 }
 
-static void ena_get_strings(struct net_device *netdev, u32 sset, u8 *data)
+static void ena_get_strings(struct ena_adapter *adapter,
+			    u8 *data,
+			    bool hw_stats_needed)
 {
-	struct ena_adapter *adapter = netdev_priv(netdev);
 	const struct ena_stats *ena_stats;
 	int i;
 
-	if (sset != ETH_SS_STATS)
-		return;
-
 	for (i = 0; i < ENA_STATS_ARRAY_GLOBAL; i++) {
 		ena_stats = &ena_stats_global_strings[i];
-
-		memcpy(data, ena_stats->name, ETH_GSTRING_LEN);
-		data += ETH_GSTRING_LEN;
+		ethtool_puts(&data, ena_stats->name);
 	}
 
+	if (hw_stats_needed)
+		ena_metrics_stats_strings(adapter, &data);
+
 	ena_queue_strings(adapter, &data);
 	ena_com_dev_strings(&data);
 }
 
+static void ena_get_ethtool_strings(struct net_device *netdev,
+				    u32 sset,
+				    u8 *data)
+{
+	struct ena_adapter *adapter = netdev_priv(netdev);
+
+	switch (sset) {
+	case ETH_SS_STATS:
+		ena_get_strings(adapter, data, true);
+		break;
+	}
+}
+
 static int ena_get_link_ksettings(struct net_device *netdev,
 				  struct ethtool_link_ksettings *link_ksettings)
 {
@@ -296,120 +484,85 @@ static int ena_get_link_ksettings(struct net_device *netdev,
 }
 
 static int ena_get_coalesce(struct net_device *net_dev,
-			    struct ethtool_coalesce *coalesce)
+			    struct ethtool_coalesce *coalesce,
+			    struct kernel_ethtool_coalesce *kernel_coal,
+			    struct netlink_ext_ack *extack)
 {
 	struct ena_adapter *adapter = netdev_priv(net_dev);
 	struct ena_com_dev *ena_dev = adapter->ena_dev;
-	struct ena_intr_moder_entry intr_moder_entry;
 
-	if (!ena_com_interrupt_moderation_supported(ena_dev)) {
-		/* the devie doesn't support interrupt moderation */
+	if (!ena_com_interrupt_moderation_supported(ena_dev))
 		return -EOPNOTSUPP;
-	}
+
 	coalesce->tx_coalesce_usecs =
-		ena_com_get_nonadaptive_moderation_interval_tx(ena_dev) /
+		ena_com_get_nonadaptive_moderation_interval_tx(ena_dev) *
 			ena_dev->intr_delay_resolution;
-	if (!ena_com_get_adaptive_moderation_enabled(ena_dev)) {
-		coalesce->rx_coalesce_usecs =
-			ena_com_get_nonadaptive_moderation_interval_rx(ena_dev)
-			/ ena_dev->intr_delay_resolution;
-	} else {
-		ena_com_get_intr_moderation_entry(adapter->ena_dev, ENA_INTR_MODER_LOWEST, &intr_moder_entry);
-		coalesce->rx_coalesce_usecs_low = intr_moder_entry.intr_moder_interval;
-		coalesce->rx_max_coalesced_frames_low = intr_moder_entry.pkts_per_interval;
 
-		ena_com_get_intr_moderation_entry(adapter->ena_dev, ENA_INTR_MODER_MID, &intr_moder_entry);
-		coalesce->rx_coalesce_usecs = intr_moder_entry.intr_moder_interval;
-		coalesce->rx_max_coalesced_frames = intr_moder_entry.pkts_per_interval;
+	coalesce->rx_coalesce_usecs =
+		ena_com_get_nonadaptive_moderation_interval_rx(ena_dev)
+		* ena_dev->intr_delay_resolution;
 
-		ena_com_get_intr_moderation_entry(adapter->ena_dev, ENA_INTR_MODER_HIGHEST, &intr_moder_entry);
-		coalesce->rx_coalesce_usecs_high = intr_moder_entry.intr_moder_interval;
-		coalesce->rx_max_coalesced_frames_high = intr_moder_entry.pkts_per_interval;
-	}
 	coalesce->use_adaptive_rx_coalesce =
 		ena_com_get_adaptive_moderation_enabled(ena_dev);
 
 	return 0;
 }
 
-static void ena_update_tx_rings_intr_moderation(struct ena_adapter *adapter)
+static void ena_update_tx_rings_nonadaptive_intr_moderation(struct ena_adapter *adapter)
 {
 	unsigned int val;
 	int i;
 
 	val = ena_com_get_nonadaptive_moderation_interval_tx(adapter->ena_dev);
 
-	for (i = 0; i < adapter->num_queues; i++)
+	for (i = 0; i < adapter->num_io_queues; i++)
 		adapter->tx_ring[i].smoothed_interval = val;
 }
 
+static void ena_update_rx_rings_nonadaptive_intr_moderation(struct ena_adapter *adapter)
+{
+	unsigned int val;
+	int i;
+
+	val = ena_com_get_nonadaptive_moderation_interval_rx(adapter->ena_dev);
+
+	for (i = 0; i < adapter->num_io_queues; i++)
+		adapter->rx_ring[i].smoothed_interval = val;
+}
+
 static int ena_set_coalesce(struct net_device *net_dev,
-			    struct ethtool_coalesce *coalesce)
+			    struct ethtool_coalesce *coalesce,
+			    struct kernel_ethtool_coalesce *kernel_coal,
+			    struct netlink_ext_ack *extack)
 {
 	struct ena_adapter *adapter = netdev_priv(net_dev);
 	struct ena_com_dev *ena_dev = adapter->ena_dev;
-	struct ena_intr_moder_entry intr_moder_entry;
 	int rc;
 
-	if (!ena_com_interrupt_moderation_supported(ena_dev)) {
-		/* the devie doesn't support interrupt moderation */
+	if (!ena_com_interrupt_moderation_supported(ena_dev))
 		return -EOPNOTSUPP;
-	}
-
-	if (coalesce->rx_coalesce_usecs_irq ||
-	    coalesce->rx_max_coalesced_frames_irq ||
-	    coalesce->tx_coalesce_usecs_irq ||
-	    coalesce->tx_max_coalesced_frames ||
-	    coalesce->tx_max_coalesced_frames_irq ||
-	    coalesce->stats_block_coalesce_usecs ||
-	    coalesce->use_adaptive_tx_coalesce ||
-	    coalesce->pkt_rate_low ||
-	    coalesce->tx_coalesce_usecs_low ||
-	    coalesce->tx_max_coalesced_frames_low ||
-	    coalesce->pkt_rate_high ||
-	    coalesce->tx_coalesce_usecs_high ||
-	    coalesce->tx_max_coalesced_frames_high ||
-	    coalesce->rate_sample_interval)
-		return -EINVAL;
 
 	rc = ena_com_update_nonadaptive_moderation_interval_tx(ena_dev,
 							       coalesce->tx_coalesce_usecs);
 	if (rc)
 		return rc;
 
-	ena_update_tx_rings_intr_moderation(adapter);
+	ena_update_tx_rings_nonadaptive_intr_moderation(adapter);
 
-	if (ena_com_get_adaptive_moderation_enabled(ena_dev)) {
-		if (!coalesce->use_adaptive_rx_coalesce) {
-			ena_com_disable_adaptive_moderation(ena_dev);
-			rc = ena_com_update_nonadaptive_moderation_interval_rx(ena_dev,
-									       coalesce->rx_coalesce_usecs);
-			return rc;
-		}
-	} else { /* was in non-adaptive mode */
-		if (coalesce->use_adaptive_rx_coalesce) {
-			ena_com_enable_adaptive_moderation(ena_dev);
-		} else {
-			rc = ena_com_update_nonadaptive_moderation_interval_rx(ena_dev,
-									       coalesce->rx_coalesce_usecs);
-			return rc;
-		}
-	}
+	rc = ena_com_update_nonadaptive_moderation_interval_rx(ena_dev,
+							       coalesce->rx_coalesce_usecs);
+	if (rc)
+		return rc;
 
-	intr_moder_entry.intr_moder_interval = coalesce->rx_coalesce_usecs_low;
-	intr_moder_entry.pkts_per_interval = coalesce->rx_max_coalesced_frames_low;
-	intr_moder_entry.bytes_per_interval = ENA_INTR_BYTE_COUNT_NOT_SUPPORTED;
-	ena_com_init_intr_moderation_entry(adapter->ena_dev, ENA_INTR_MODER_LOWEST, &intr_moder_entry);
+	ena_update_rx_rings_nonadaptive_intr_moderation(adapter);
 
-	intr_moder_entry.intr_moder_interval = coalesce->rx_coalesce_usecs;
-	intr_moder_entry.pkts_per_interval = coalesce->rx_max_coalesced_frames;
-	intr_moder_entry.bytes_per_interval = ENA_INTR_BYTE_COUNT_NOT_SUPPORTED;
-	ena_com_init_intr_moderation_entry(adapter->ena_dev, ENA_INTR_MODER_MID, &intr_moder_entry);
+	if (coalesce->use_adaptive_rx_coalesce &&
+	    !ena_com_get_adaptive_moderation_enabled(ena_dev))
+		ena_com_enable_adaptive_moderation(ena_dev);
 
-	intr_moder_entry.intr_moder_interval = coalesce->rx_coalesce_usecs_high;
-	intr_moder_entry.pkts_per_interval = coalesce->rx_max_coalesced_frames_high;
-	intr_moder_entry.bytes_per_interval = ENA_INTR_BYTE_COUNT_NOT_SUPPORTED;
-	ena_com_init_intr_moderation_entry(adapter->ena_dev, ENA_INTR_MODER_HIGHEST, &intr_moder_entry);
+	if (!coalesce->use_adaptive_rx_coalesce &&
+	    ena_com_get_adaptive_moderation_enabled(ena_dev))
+		ena_com_disable_adaptive_moderation(ena_dev);
 
 	return 0;
 }
@@ -432,24 +585,110 @@ static void ena_get_drvinfo(struct net_device *dev,
 			    struct ethtool_drvinfo *info)
 {
 	struct ena_adapter *adapter = netdev_priv(dev);
-
-	strlcpy(info->driver, DRV_MODULE_NAME, sizeof(info->driver));
-	strlcpy(info->version, DRV_MODULE_VERSION, sizeof(info->version));
-	strlcpy(info->bus_info, pci_name(adapter->pdev),
-		sizeof(info->bus_info));
+	ssize_t ret = 0;
+
+	ret = strscpy(info->driver, DRV_MODULE_NAME, sizeof(info->driver));
+	if (ret < 0)
+		netif_dbg(adapter, drv, dev,
+			  "module name will be truncated, status = %zd\n", ret);
+
+	ret = strscpy(info->bus_info, pci_name(adapter->pdev),
+		      sizeof(info->bus_info));
+	if (ret < 0)
+		netif_dbg(adapter, drv, dev,
+			  "bus info will be truncated, status = %zd\n", ret);
 }
 
 static void ena_get_ringparam(struct net_device *netdev,
-			      struct ethtool_ringparam *ring)
+			      struct ethtool_ringparam *ring,
+			      struct kernel_ethtool_ringparam *kernel_ring,
+			      struct netlink_ext_ack *extack)
 {
 	struct ena_adapter *adapter = netdev_priv(netdev);
-	struct ena_ring *tx_ring = &adapter->tx_ring[0];
-	struct ena_ring *rx_ring = &adapter->rx_ring[0];
 
-	ring->rx_max_pending = rx_ring->ring_size;
-	ring->tx_max_pending = tx_ring->ring_size;
-	ring->rx_pending = rx_ring->ring_size;
-	ring->tx_pending = tx_ring->ring_size;
+	ring->tx_max_pending = adapter->max_tx_ring_size;
+	ring->rx_max_pending = adapter->max_rx_ring_size;
+	if (adapter->ena_dev->tx_mem_queue_type == ENA_ADMIN_PLACEMENT_POLICY_DEV) {
+		bool large_llq_supported = adapter->large_llq_header_supported;
+
+		kernel_ring->tx_push = true;
+		kernel_ring->tx_push_buf_len = adapter->ena_dev->tx_max_header_size;
+		if (large_llq_supported)
+			kernel_ring->tx_push_buf_max_len = ENA_LLQ_LARGE_HEADER;
+		else
+			kernel_ring->tx_push_buf_max_len = ENA_LLQ_HEADER;
+	} else {
+		kernel_ring->tx_push = false;
+		kernel_ring->tx_push_buf_max_len = 0;
+		kernel_ring->tx_push_buf_len = 0;
+	}
+
+	ring->tx_pending = adapter->tx_ring[0].ring_size;
+	ring->rx_pending = adapter->rx_ring[0].ring_size;
+}
+
+static int ena_set_ringparam(struct net_device *netdev,
+			     struct ethtool_ringparam *ring,
+			     struct kernel_ethtool_ringparam *kernel_ring,
+			     struct netlink_ext_ack *extack)
+{
+	struct ena_adapter *adapter = netdev_priv(netdev);
+	u32 new_tx_size, new_rx_size, new_tx_push_buf_len;
+	bool changed = false;
+
+	new_tx_size = ring->tx_pending < ENA_MIN_RING_SIZE ?
+			ENA_MIN_RING_SIZE : ring->tx_pending;
+	new_tx_size = rounddown_pow_of_two(new_tx_size);
+
+	new_rx_size = ring->rx_pending < ENA_MIN_RING_SIZE ?
+			ENA_MIN_RING_SIZE : ring->rx_pending;
+	new_rx_size = rounddown_pow_of_two(new_rx_size);
+
+	changed |= new_tx_size != adapter->requested_tx_ring_size ||
+		   new_rx_size != adapter->requested_rx_ring_size;
+
+	/* This value is ignored if LLQ is not supported */
+	new_tx_push_buf_len = adapter->ena_dev->tx_max_header_size;
+
+	if ((adapter->ena_dev->tx_mem_queue_type == ENA_ADMIN_PLACEMENT_POLICY_DEV) !=
+	    kernel_ring->tx_push) {
+		NL_SET_ERR_MSG_MOD(extack, "Push mode state cannot be modified");
+		return -EINVAL;
+	}
+
+	/* Validate that the push buffer is supported on the underlying device */
+	if (kernel_ring->tx_push_buf_len) {
+		enum ena_admin_placement_policy_type placement;
+
+		new_tx_push_buf_len = kernel_ring->tx_push_buf_len;
+
+		placement = adapter->ena_dev->tx_mem_queue_type;
+		if (placement == ENA_ADMIN_PLACEMENT_POLICY_HOST)
+			return -EOPNOTSUPP;
+
+		if (new_tx_push_buf_len != ENA_LLQ_HEADER &&
+		    new_tx_push_buf_len != ENA_LLQ_LARGE_HEADER) {
+			bool large_llq_sup = adapter->large_llq_header_supported;
+			char large_llq_size_str[40];
+
+			snprintf(large_llq_size_str, 40, ", %lu", ENA_LLQ_LARGE_HEADER);
+
+			NL_SET_ERR_MSG_FMT_MOD(extack,
+					       "Supported tx push buff values: [%lu%s]",
+					       ENA_LLQ_HEADER,
+					       large_llq_sup ? large_llq_size_str : "");
+
+			return -EINVAL;
+		}
+
+		changed |= new_tx_push_buf_len != adapter->ena_dev->tx_max_header_size;
+	}
+
+	if (!changed)
+		return 0;
+
+	return ena_update_queue_params(adapter, new_tx_size, new_rx_size,
+				       new_tx_push_buf_len);
 }
 
 static u32 ena_flow_hash_to_flow_type(u16 hash_fields)
@@ -496,9 +735,11 @@ static u16 ena_flow_data_to_flow_hash(u32 hash_fields)
 	return data;
 }
 
-static int ena_get_rss_hash(struct ena_com_dev *ena_dev,
-			    struct ethtool_rxnfc *cmd)
+static int ena_get_rxfh_fields(struct net_device *netdev,
+			       struct ethtool_rxfh_fields *cmd)
 {
+	struct ena_adapter *adapter = netdev_priv(netdev);
+	struct ena_com_dev *ena_dev = adapter->ena_dev;
 	enum ena_admin_flow_hash_proto proto;
 	u16 hash_fields;
 	int rc;
@@ -539,21 +780,20 @@ static int ena_get_rss_hash(struct ena_com_dev *ena_dev,
 	}
 
 	rc = ena_com_get_hash_ctrl(ena_dev, proto, &hash_fields);
-	if (rc) {
-		/* If device don't have permission, return unsupported */
-		if (rc == -EPERM)
-			rc = -EOPNOTSUPP;
+	if (rc)
 		return rc;
-	}
 
 	cmd->data = ena_flow_hash_to_flow_type(hash_fields);
 
 	return 0;
 }
 
-static int ena_set_rss_hash(struct ena_com_dev *ena_dev,
-			    struct ethtool_rxnfc *cmd)
+static int ena_set_rxfh_fields(struct net_device *netdev,
+			       const struct ethtool_rxfh_fields *cmd,
+			       struct netlink_ext_ack *extack)
 {
+	struct ena_adapter *adapter = netdev_priv(netdev);
+	struct ena_com_dev *ena_dev = adapter->ena_dev;
 	enum ena_admin_flow_hash_proto proto;
 	u16 hash_fields;
 
@@ -595,26 +835,6 @@ static int ena_set_rss_hash(struct ena_com_dev *ena_dev,
 	return ena_com_fill_hash_ctrl(ena_dev, proto, hash_fields);
 }
 
-static int ena_set_rxnfc(struct net_device *netdev, struct ethtool_rxnfc *info)
-{
-	struct ena_adapter *adapter = netdev_priv(netdev);
-	int rc = 0;
-
-	switch (info->cmd) {
-	case ETHTOOL_SRXFH:
-		rc = ena_set_rss_hash(adapter->ena_dev, info);
-		break;
-	case ETHTOOL_SRXCLSRLDEL:
-	case ETHTOOL_SRXCLSRLINS:
-	default:
-		netif_err(adapter, drv, netdev,
-			  "Command parameter %d is not supported\n", info->cmd);
-		rc = -EOPNOTSUPP;
-	}
-
-	return (rc == -EPERM) ? -EOPNOTSUPP : rc;
-}
-
 static int ena_get_rxnfc(struct net_device *netdev, struct ethtool_rxnfc *info,
 			 u32 *rules)
 {
@@ -623,12 +843,9 @@ static int ena_get_rxnfc(struct net_device *netdev, struct ethtool_rxnfc *info,
 
 	switch (info->cmd) {
 	case ETHTOOL_GRXRINGS:
-		info->data = adapter->num_queues;
+		info->data = adapter->num_io_queues;
 		rc = 0;
 		break;
-	case ETHTOOL_GRXFH:
-		rc = ena_get_rss_hash(adapter->ena_dev, info);
-		break;
 	case ETHTOOL_GRXCLSRLCNT:
 	case ETHTOOL_GRXCLSRULE:
 	case ETHTOOL_GRXCLSRLALL:
@@ -638,7 +855,7 @@ static int ena_get_rxnfc(struct net_device *netdev, struct ethtool_rxnfc *info,
 		rc = -EOPNOTSUPP;
 	}
 
-	return (rc == -EPERM) ? -EOPNOTSUPP : rc;
+	return rc;
 }
 
 static u32 ena_get_rxfh_indir_size(struct net_device *netdev)
@@ -648,22 +865,84 @@ static u32 ena_get_rxfh_indir_size(struct net_device *netdev)
 
 static u32 ena_get_rxfh_key_size(struct net_device *netdev)
 {
-	return ENA_HASH_KEY_SIZE;
+	struct ena_adapter *adapter = netdev_priv(netdev);
+	struct ena_rss *rss = &adapter->ena_dev->rss;
+
+	return rss->hash_key ? ENA_HASH_KEY_SIZE : 0;
 }
 
-static int ena_get_rxfh(struct net_device *netdev, u32 *indir, u8 *key,
-			u8 *hfunc)
+static int ena_indirection_table_set(struct ena_adapter *adapter,
+				     const u32 *indir)
+{
+	struct ena_com_dev *ena_dev = adapter->ena_dev;
+	int i, rc;
+
+	for (i = 0; i < ENA_RX_RSS_TABLE_SIZE; i++) {
+		rc = ena_com_indirect_table_fill_entry(ena_dev,
+						       i,
+						       ENA_IO_RXQ_IDX(indir[i]));
+		if (unlikely(rc)) {
+			netif_err(adapter, drv, adapter->netdev,
+				  "Cannot fill indirect table (index is too large)\n");
+			return rc;
+		}
+	}
+
+	rc = ena_com_indirect_table_set(ena_dev);
+	if (rc) {
+		netif_err(adapter, drv, adapter->netdev,
+			  "Cannot set indirect table\n");
+		return rc == -EPERM ? -EOPNOTSUPP : rc;
+	}
+	return rc;
+}
+
+static int ena_indirection_table_get(struct ena_adapter *adapter, u32 *indir)
+{
+	struct ena_com_dev *ena_dev = adapter->ena_dev;
+	int i, rc;
+
+	if (!indir)
+		return 0;
+
+	rc = ena_com_indirect_table_get(ena_dev, indir);
+	if (rc)
+		return rc;
+
+	/* Our internal representation of the indices is: even indices
+	 * for Tx and uneven indices for Rx. We need to convert the Rx
+	 * indices to be consecutive
+	 */
+	for (i = 0; i < ENA_RX_RSS_TABLE_SIZE; i++)
+		indir[i] = ENA_IO_RXQ_IDX_TO_COMBINED_IDX(indir[i]);
+
+	return rc;
+}
+
+static int ena_get_rxfh(struct net_device *netdev,
+			struct ethtool_rxfh_param *rxfh)
 {
 	struct ena_adapter *adapter = netdev_priv(netdev);
 	enum ena_admin_hash_functions ena_func;
 	u8 func;
 	int rc;
 
-	rc = ena_com_indirect_table_get(adapter->ena_dev, indir);
+	rc = ena_indirection_table_get(adapter, rxfh->indir);
 	if (rc)
 		return rc;
 
-	rc = ena_com_get_hash_function(adapter->ena_dev, &ena_func, key);
+	/* We call this function in order to check if the device
+	 * supports getting/setting the hash function.
+	 */
+	rc = ena_com_get_hash_function(adapter->ena_dev, &ena_func);
+	if (rc) {
+		if (rc == -EOPNOTSUPP)
+			rc = 0;
+
+		return rc;
+	}
+
+	rc = ena_com_get_hash_key(adapter->ena_dev, rxfh->key);
 	if (rc)
 		return rc;
 
@@ -672,7 +951,7 @@ static int ena_get_rxfh(struct net_device *netdev, u32 *indir, u8 *key,
 		func = ETH_RSS_HASH_TOP;
 		break;
 	case ENA_ADMIN_CRC32:
-		func = ETH_RSS_HASH_XOR;
+		func = ETH_RSS_HASH_CRC32;
 		break;
 	default:
 		netif_err(adapter, drv, netdev,
@@ -680,55 +959,44 @@ static int ena_get_rxfh(struct net_device *netdev, u32 *indir, u8 *key,
 		return -EOPNOTSUPP;
 	}
 
-	if (hfunc)
-		*hfunc = func;
+	rxfh->hfunc = func;
 
-	return rc;
+	return 0;
 }
 
-static int ena_set_rxfh(struct net_device *netdev, const u32 *indir,
-			const u8 *key, const u8 hfunc)
+static int ena_set_rxfh(struct net_device *netdev,
+			struct ethtool_rxfh_param *rxfh,
+			struct netlink_ext_ack *extack)
 {
 	struct ena_adapter *adapter = netdev_priv(netdev);
 	struct ena_com_dev *ena_dev = adapter->ena_dev;
-	enum ena_admin_hash_functions func;
-	int rc, i;
-
-	if (indir) {
-		for (i = 0; i < ENA_RX_RSS_TABLE_SIZE; i++) {
-			rc = ena_com_indirect_table_fill_entry(ena_dev,
-							       ENA_IO_RXQ_IDX(indir[i]),
-							       i);
-			if (unlikely(rc)) {
-				netif_err(adapter, drv, netdev,
-					  "Cannot fill indirect table (index is too large)\n");
-				return rc;
-			}
-		}
+	enum ena_admin_hash_functions func = 0;
+	int rc;
 
-		rc = ena_com_indirect_table_set(ena_dev);
-		if (rc) {
-			netif_err(adapter, drv, netdev,
-				  "Cannot set indirect table\n");
-			return rc == -EPERM ? -EOPNOTSUPP : rc;
-		}
+	if (rxfh->indir) {
+		rc = ena_indirection_table_set(adapter, rxfh->indir);
+		if (rc)
+			return rc;
 	}
 
-	switch (hfunc) {
+	switch (rxfh->hfunc) {
+	case ETH_RSS_HASH_NO_CHANGE:
+		func = ena_com_get_current_hash_function(ena_dev);
+		break;
 	case ETH_RSS_HASH_TOP:
 		func = ENA_ADMIN_TOEPLITZ;
 		break;
-	case ETH_RSS_HASH_XOR:
+	case ETH_RSS_HASH_CRC32:
 		func = ENA_ADMIN_CRC32;
 		break;
 	default:
 		netif_err(adapter, drv, netdev, "Unsupported hfunc %d\n",
-			  hfunc);
+			  rxfh->hfunc);
 		return -EOPNOTSUPP;
 	}
 
-	if (key) {
-		rc = ena_com_fill_hash_function(ena_dev, func, key,
+	if (rxfh->key || func) {
+		rc = ena_com_fill_hash_function(ena_dev, func, rxfh->key,
 						ENA_HASH_KEY_SIZE,
 						0xFFFFFFFF);
 		if (unlikely(rc)) {
@@ -745,14 +1013,31 @@ static void ena_get_channels(struct net_device *netdev,
 {
 	struct ena_adapter *adapter = netdev_priv(netdev);
 
-	channels->max_rx = ENA_MAX_NUM_IO_QUEUES;
-	channels->max_tx = ENA_MAX_NUM_IO_QUEUES;
-	channels->max_other = 0;
-	channels->max_combined = 0;
-	channels->rx_count = adapter->num_queues;
-	channels->tx_count = adapter->num_queues;
-	channels->other_count = 0;
-	channels->combined_count = 0;
+	channels->max_combined = adapter->max_num_io_queues;
+	channels->combined_count = adapter->num_io_queues;
+}
+
+static int ena_set_channels(struct net_device *netdev,
+			    struct ethtool_channels *channels)
+{
+	struct ena_adapter *adapter = netdev_priv(netdev);
+	u32 count = channels->combined_count;
+	/* The check for max value is already done in ethtool */
+	if (count < ENA_MIN_NUM_IO_QUEUES)
+		return -EINVAL;
+
+	if (!ena_xdp_legal_queue_count(adapter, count)) {
+		if (ena_xdp_present(adapter))
+			return -EINVAL;
+
+		xdp_clear_features_flag(netdev);
+	} else {
+		xdp_set_features_flag(netdev,
+				      NETDEV_XDP_ACT_BASIC |
+				      NETDEV_XDP_ACT_REDIRECT);
+	}
+
+	return ena_update_queue_count(adapter, count);
 }
 
 static int ena_get_tunable(struct net_device *netdev,
@@ -784,11 +1069,7 @@ static int ena_set_tunable(struct net_device *netdev,
 	switch (tuna->id) {
 	case ETHTOOL_RX_COPYBREAK:
 		len = *(u32 *)data;
-		if (len > adapter->netdev->mtu) {
-			ret = -EINVAL;
-			break;
-		}
-		adapter->rx_copybreak = len;
+		ret = ena_set_rx_copybreak(adapter, len);
 		break;
 	default:
 		ret = -EINVAL;
@@ -799,6 +1080,10 @@ static int ena_set_tunable(struct net_device *netdev,
 }
 
 static const struct ethtool_ops ena_ethtool_ops = {
+	.supported_coalesce_params = ETHTOOL_COALESCE_USECS |
+				     ETHTOOL_COALESCE_USE_ADAPTIVE_RX,
+	.supported_ring_params	= ETHTOOL_RING_USE_TX_PUSH_BUF_LEN |
+				  ETHTOOL_RING_USE_TX_PUSH,
 	.get_link_ksettings	= ena_get_link_ksettings,
 	.get_drvinfo		= ena_get_drvinfo,
 	.get_msglevel		= ena_get_msglevel,
@@ -807,18 +1092,22 @@ static const struct ethtool_ops ena_ethtool_ops = {
 	.get_coalesce		= ena_get_coalesce,
 	.set_coalesce		= ena_set_coalesce,
 	.get_ringparam		= ena_get_ringparam,
+	.set_ringparam		= ena_set_ringparam,
 	.get_sset_count         = ena_get_sset_count,
-	.get_strings		= ena_get_strings,
+	.get_strings		= ena_get_ethtool_strings,
 	.get_ethtool_stats      = ena_get_ethtool_stats,
 	.get_rxnfc		= ena_get_rxnfc,
-	.set_rxnfc		= ena_set_rxnfc,
 	.get_rxfh_indir_size    = ena_get_rxfh_indir_size,
 	.get_rxfh_key_size	= ena_get_rxfh_key_size,
 	.get_rxfh		= ena_get_rxfh,
 	.set_rxfh		= ena_set_rxfh,
+	.get_rxfh_fields	= ena_get_rxfh_fields,
+	.set_rxfh_fields	= ena_set_rxfh_fields,
 	.get_channels		= ena_get_channels,
+	.set_channels		= ena_set_channels,
 	.get_tunable		= ena_get_tunable,
 	.set_tunable		= ena_set_tunable,
+	.get_ts_info		= ena_get_ts_info,
 };
 
 void ena_set_ethtool_ops(struct net_device *netdev)
@@ -834,33 +1123,29 @@ static void ena_dump_stats_ex(struct ena_adapter *adapter, u8 *buf)
 	int strings_num;
 	int i, rc;
 
-	strings_num = ena_get_sset_count(netdev, ETH_SS_STATS);
+	strings_num = ena_get_sw_stats_count(adapter);
 	if (strings_num <= 0) {
 		netif_err(adapter, drv, netdev, "Can't get stats num\n");
 		return;
 	}
 
-	strings_buf = devm_kzalloc(&adapter->pdev->dev,
-				   strings_num * ETH_GSTRING_LEN,
-				   GFP_ATOMIC);
+	strings_buf = kcalloc(strings_num, ETH_GSTRING_LEN, GFP_ATOMIC);
 	if (!strings_buf) {
 		netif_err(adapter, drv, netdev,
-			  "failed to alloc strings_buf\n");
+			  "Failed to allocate strings_buf\n");
 		return;
 	}
 
-	data_buf = devm_kzalloc(&adapter->pdev->dev,
-				strings_num * sizeof(u64),
-				GFP_ATOMIC);
+	data_buf = kcalloc(strings_num, sizeof(u64), GFP_ATOMIC);
 	if (!data_buf) {
 		netif_err(adapter, drv, netdev,
-			  "failed to allocate data buf\n");
-		devm_kfree(&adapter->pdev->dev, strings_buf);
+			  "Failed to allocate data buf\n");
+		kfree(strings_buf);
 		return;
 	}
 
-	ena_get_strings(netdev, ETH_SS_STATS, strings_buf);
-	ena_get_ethtool_stats(netdev, NULL, data_buf);
+	ena_get_strings(adapter, strings_buf, false);
+	ena_get_stats(adapter, data_buf, false);
 
 	/* If there is a buffer, dump stats, otherwise print them to dmesg */
 	if (buf)
@@ -877,8 +1162,8 @@ static void ena_dump_stats_ex(struct ena_adapter *adapter, u8 *buf)
 				  strings_buf + i * ETH_GSTRING_LEN,
 				  data_buf[i]);
 
-	devm_kfree(&adapter->pdev->dev, strings_buf);
-	devm_kfree(&adapter->pdev->dev, data_buf);
+	kfree(strings_buf);
+	kfree(data_buf);
 }
 
 void ena_dump_stats_to_buf(struct ena_adapter *adapter, u8 *buf)
diff --git a/drivers/net/ethernet/amazon/ena/ena_netdev.c b/drivers/net/ethernet/amazon/ena/ena_netdev.c
index 7c1214d78855..92d149d4f091 100644
--- a/drivers/net/ethernet/amazon/ena/ena_netdev.c
+++ b/drivers/net/ethernet/amazon/ena/ena_netdev.c
@@ -1,45 +1,13 @@
+// SPDX-License-Identifier: GPL-2.0 OR Linux-OpenIB
 /*
- * Copyright 2015 Amazon.com, Inc. or its affiliates.
- *
- * This software is available to you under a choice of one of two
- * licenses.  You may choose to be licensed under the terms of the GNU
- * General Public License (GPL) Version 2, available from the file
- * COPYING in the main directory of this source tree, or the
- * BSD license below:
- *
- *     Redistribution and use in source and binary forms, with or
- *     without modification, are permitted provided that the following
- *     conditions are met:
- *
- *      - Redistributions of source code must retain the above
- *        copyright notice, this list of conditions and the following
- *        disclaimer.
- *
- *      - Redistributions in binary form must reproduce the above
- *        copyright notice, this list of conditions and the following
- *        disclaimer in the documentation and/or other materials
- *        provided with the distribution.
- *
- * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND,
- * EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF
- * MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND
- * NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS
- * BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN
- * ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN
- * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
- * SOFTWARE.
+ * Copyright 2015-2020 Amazon.com, Inc. or its affiliates. All rights reserved.
  */
 
 #define pr_fmt(fmt) KBUILD_MODNAME ": " fmt
 
-#ifdef CONFIG_RFS_ACCEL
-#include <linux/cpu_rmap.h>
-#endif /* CONFIG_RFS_ACCEL */
 #include <linux/ethtool.h>
-#include <linux/if_vlan.h>
 #include <linux/kernel.h>
 #include <linux/module.h>
-#include <linux/moduleparam.h>
 #include <linux/numa.h>
 #include <linux/pci.h>
 #include <linux/utsname.h>
@@ -49,24 +17,25 @@
 
 #include "ena_netdev.h"
 #include "ena_pci_id_tbl.h"
+#include "ena_xdp.h"
 
-static char version[] = DEVICE_NAME " v" DRV_MODULE_VERSION "\n";
+#include "ena_phc.h"
+
+#include "ena_devlink.h"
+
+#include "ena_debugfs.h"
 
 MODULE_AUTHOR("Amazon.com, Inc. or its affiliates");
 MODULE_DESCRIPTION(DEVICE_NAME);
 MODULE_LICENSE("GPL");
-MODULE_VERSION(DRV_MODULE_VERSION);
 
 /* Time in jiffies before concluding the transmitter is hung. */
 #define TX_TIMEOUT  (5 * HZ)
 
-#define ENA_NAPI_BUDGET 64
+#define ENA_MAX_RINGS min_t(unsigned int, ENA_MAX_NUM_IO_QUEUES, num_possible_cpus())
 
 #define DEFAULT_MSG_ENABLE (NETIF_MSG_DRV | NETIF_MSG_PROBE | NETIF_MSG_IFUP | \
-		NETIF_MSG_TX_DONE | NETIF_MSG_TX_ERR | NETIF_MSG_RX_ERR)
-static int debug = -1;
-module_param(debug, int, 0);
-MODULE_PARM_DESC(debug, "Debug level (0=none,...,16=all)");
+		NETIF_MSG_IFDOWN | NETIF_MSG_TX_ERR | NETIF_MSG_RX_ERR)
 
 static struct ena_aenq_handlers aenq_handlers;
 
@@ -75,30 +44,55 @@ static struct workqueue_struct *ena_wq;
 MODULE_DEVICE_TABLE(pci, ena_pci_tbl);
 
 static int ena_rss_init_default(struct ena_adapter *adapter);
+static void check_for_admin_com_state(struct ena_adapter *adapter);
 
-static void ena_tx_timeout(struct net_device *dev)
+static void ena_tx_timeout(struct net_device *dev, unsigned int txqueue)
 {
+	enum ena_regs_reset_reason_types reset_reason = ENA_REGS_RESET_OS_NETDEV_WD;
 	struct ena_adapter *adapter = netdev_priv(dev);
+	unsigned int time_since_last_napi, threshold;
+	struct ena_ring *tx_ring;
+	int napi_scheduled;
 
+	if (txqueue >= adapter->num_io_queues) {
+		netdev_err(dev, "TX timeout on invalid queue %u\n", txqueue);
+		goto schedule_reset;
+	}
+
+	threshold = jiffies_to_usecs(dev->watchdog_timeo);
+	tx_ring = &adapter->tx_ring[txqueue];
+
+	time_since_last_napi = jiffies_to_usecs(jiffies - tx_ring->tx_stats.last_napi_jiffies);
+	napi_scheduled = !!(tx_ring->napi->state & NAPIF_STATE_SCHED);
+
+	netdev_err(dev,
+		   "TX q %d is paused for too long (threshold %u). Time since last napi %u usec. napi scheduled: %d\n",
+		   txqueue,
+		   threshold,
+		   time_since_last_napi,
+		   napi_scheduled);
+
+	if (threshold < time_since_last_napi && napi_scheduled) {
+		netdev_err(dev,
+			   "napi handler hasn't been called for a long time but is scheduled\n");
+		reset_reason = ENA_REGS_RESET_SUSPECTED_POLL_STARVATION;
+	}
+schedule_reset:
 	/* Change the state of the device to trigger reset
 	 * Check that we are not in the middle or a trigger already
 	 */
-
 	if (test_and_set_bit(ENA_FLAG_TRIGGER_RESET, &adapter->flags))
 		return;
 
-	u64_stats_update_begin(&adapter->syncp);
-	adapter->dev_stats.tx_timeout++;
-	u64_stats_update_end(&adapter->syncp);
-
-	netif_err(adapter, tx_err, dev, "Transmit time out\n");
+	ena_reset_device(adapter, reset_reason);
+	ena_increase_stat(&adapter->dev_stats.tx_timeout, 1, &adapter->syncp);
 }
 
 static void update_rx_ring_mtu(struct ena_adapter *adapter, int mtu)
 {
 	int i;
 
-	for (i = 0; i < adapter->num_queues; i++)
+	for (i = 0; i < adapter->num_io_queues; i++)
 		adapter->rx_ring[i].mtu = mtu;
 }
 
@@ -109,9 +103,9 @@ static int ena_change_mtu(struct net_device *dev, int new_mtu)
 
 	ret = ena_com_set_dev_mtu(adapter->ena_dev, new_mtu);
 	if (!ret) {
-		netif_dbg(adapter, drv, dev, "set MTU to %d\n", new_mtu);
+		netif_dbg(adapter, drv, dev, "Set MTU to %d\n", new_mtu);
 		update_rx_ring_mtu(adapter, new_mtu);
-		dev->mtu = new_mtu;
+		WRITE_ONCE(dev->mtu, new_mtu);
 	} else {
 		netif_err(adapter, drv, dev, "Failed to set MTU to %d\n",
 			  new_mtu);
@@ -120,27 +114,52 @@ static int ena_change_mtu(struct net_device *dev, int new_mtu)
 	return ret;
 }
 
-static int ena_init_rx_cpu_rmap(struct ena_adapter *adapter)
+int ena_xmit_common(struct ena_adapter *adapter,
+		    struct ena_ring *ring,
+		    struct ena_tx_buffer *tx_info,
+		    struct ena_com_tx_ctx *ena_tx_ctx,
+		    u16 next_to_use,
+		    u32 bytes)
 {
-#ifdef CONFIG_RFS_ACCEL
-	u32 i;
-	int rc;
+	int rc, nb_hw_desc;
+
+	if (unlikely(ena_com_is_doorbell_needed(ring->ena_com_io_sq,
+						ena_tx_ctx))) {
+		netif_dbg(adapter, tx_queued, adapter->netdev,
+			  "llq tx max burst size of queue %d achieved, writing doorbell to send burst\n",
+			  ring->qid);
+		ena_ring_tx_doorbell(ring);
+	}
 
-	adapter->netdev->rx_cpu_rmap = alloc_irq_cpu_rmap(adapter->num_queues);
-	if (!adapter->netdev->rx_cpu_rmap)
-		return -ENOMEM;
-	for (i = 0; i < adapter->num_queues; i++) {
-		int irq_idx = ENA_IO_IRQ_IDX(i);
+	/* prepare the packet's descriptors to dma engine */
+	rc = ena_com_prepare_tx(ring->ena_com_io_sq, ena_tx_ctx,
+				&nb_hw_desc);
 
-		rc = irq_cpu_rmap_add(adapter->netdev->rx_cpu_rmap,
-				      pci_irq_vector(adapter->pdev, irq_idx));
-		if (rc) {
-			free_irq_cpu_rmap(adapter->netdev->rx_cpu_rmap);
-			adapter->netdev->rx_cpu_rmap = NULL;
-			return rc;
-		}
+	/* In case there isn't enough space in the queue for the packet,
+	 * we simply drop it. All other failure reasons of
+	 * ena_com_prepare_tx() are fatal and therefore require a device reset.
+	 */
+	if (unlikely(rc)) {
+		netif_err(adapter, tx_queued, adapter->netdev,
+			  "Failed to prepare tx bufs\n");
+		ena_increase_stat(&ring->tx_stats.prepare_ctx_err, 1, &ring->syncp);
+		if (rc != -ENOMEM)
+			ena_reset_device(adapter, ENA_REGS_RESET_DRIVER_INVALID_STATE);
+		return rc;
 	}
-#endif /* CONFIG_RFS_ACCEL */
+
+	u64_stats_update_begin(&ring->syncp);
+	ring->tx_stats.cnt++;
+	ring->tx_stats.bytes += bytes;
+	u64_stats_update_end(&ring->syncp);
+
+	tx_info->tx_descs = nb_hw_desc;
+	tx_info->total_tx_size = bytes;
+	tx_info->last_jiffies = jiffies;
+	tx_info->print_once = 0;
+
+	ring->next_to_use = ENA_TX_RING_IDX_NEXT(next_to_use,
+						 ring->ring_size);
 	return 0;
 }
 
@@ -155,12 +174,14 @@ static void ena_init_io_rings_common(struct ena_adapter *adapter,
 	ring->adapter = adapter;
 	ring->ena_dev = adapter->ena_dev;
 	ring->per_napi_packets = 0;
-	ring->per_napi_bytes = 0;
 	ring->cpu = 0;
+	ring->numa_node = 0;
+	ring->no_interrupt_event_cnt = 0;
 	u64_stats_init(&ring->syncp);
 }
 
-static void ena_init_io_rings(struct ena_adapter *adapter)
+void ena_init_io_rings(struct ena_adapter *adapter,
+		       int first_index, int count)
 {
 	struct ena_com_dev *ena_dev;
 	struct ena_ring *txr, *rxr;
@@ -168,28 +189,39 @@ static void ena_init_io_rings(struct ena_adapter *adapter)
 
 	ena_dev = adapter->ena_dev;
 
-	for (i = 0; i < adapter->num_queues; i++) {
+	for (i = first_index; i < first_index + count; i++) {
 		txr = &adapter->tx_ring[i];
 		rxr = &adapter->rx_ring[i];
 
-		/* TX/RX common ring state */
+		/* TX common ring state */
 		ena_init_io_rings_common(adapter, txr, i);
-		ena_init_io_rings_common(adapter, rxr, i);
 
 		/* TX specific ring state */
-		txr->ring_size = adapter->tx_ring_size;
+		txr->ring_size = adapter->requested_tx_ring_size;
 		txr->tx_max_header_size = ena_dev->tx_max_header_size;
 		txr->tx_mem_queue_type = ena_dev->tx_mem_queue_type;
 		txr->sgl_size = adapter->max_tx_sgl_size;
 		txr->smoothed_interval =
 			ena_com_get_nonadaptive_moderation_interval_tx(ena_dev);
-
-		/* RX specific ring state */
-		rxr->ring_size = adapter->rx_ring_size;
-		rxr->rx_copybreak = adapter->rx_copybreak;
-		rxr->sgl_size = adapter->max_rx_sgl_size;
-		rxr->smoothed_interval =
-			ena_com_get_nonadaptive_moderation_interval_rx(ena_dev);
+		txr->disable_meta_caching = adapter->disable_meta_caching;
+		spin_lock_init(&txr->xdp_tx_lock);
+
+		/* Don't init RX queues for xdp queues */
+		if (!ENA_IS_XDP_INDEX(adapter, i)) {
+			/* RX common ring state */
+			ena_init_io_rings_common(adapter, rxr, i);
+
+			/* RX specific ring state */
+			rxr->ring_size = adapter->requested_rx_ring_size;
+			rxr->rx_copybreak = adapter->rx_copybreak;
+			rxr->sgl_size = adapter->max_rx_sgl_size;
+			rxr->smoothed_interval =
+				ena_com_get_nonadaptive_moderation_interval_rx(ena_dev);
+			rxr->empty_rx_queue = 0;
+			rxr->rx_headroom = NET_SKB_PAD;
+			adapter->ena_napi[i].dim.mode = DIM_CQ_PERIOD_MODE_START_FROM_EQE;
+			rxr->xdp_ring = &adapter->tx_ring[i + adapter->num_io_queues];
+		}
 	}
 }
 
@@ -218,22 +250,28 @@ static int ena_setup_tx_resources(struct ena_adapter *adapter, int qid)
 	if (!tx_ring->tx_buffer_info) {
 		tx_ring->tx_buffer_info = vzalloc(size);
 		if (!tx_ring->tx_buffer_info)
-			return -ENOMEM;
+			goto err_tx_buffer_info;
 	}
 
 	size = sizeof(u16) * tx_ring->ring_size;
-	tx_ring->free_tx_ids = vzalloc_node(size, node);
-	if (!tx_ring->free_tx_ids) {
-		tx_ring->free_tx_ids = vzalloc(size);
-		if (!tx_ring->free_tx_ids) {
-			vfree(tx_ring->tx_buffer_info);
-			return -ENOMEM;
-		}
+	tx_ring->free_ids = vzalloc_node(size, node);
+	if (!tx_ring->free_ids) {
+		tx_ring->free_ids = vzalloc(size);
+		if (!tx_ring->free_ids)
+			goto err_tx_free_ids;
+	}
+
+	size = tx_ring->tx_max_header_size;
+	tx_ring->push_buf_intermediate_buf = vzalloc_node(size, node);
+	if (!tx_ring->push_buf_intermediate_buf) {
+		tx_ring->push_buf_intermediate_buf = vzalloc(size);
+		if (!tx_ring->push_buf_intermediate_buf)
+			goto err_push_buf_intermediate_buf;
 	}
 
 	/* Req id ring for TX out of order completions */
 	for (i = 0; i < tx_ring->ring_size; i++)
-		tx_ring->free_tx_ids[i] = i;
+		tx_ring->free_ids[i] = i;
 
 	/* Reset tx statistics */
 	memset(&tx_ring->tx_stats, 0x0, sizeof(tx_ring->tx_stats));
@@ -241,7 +279,17 @@ static int ena_setup_tx_resources(struct ena_adapter *adapter, int qid)
 	tx_ring->next_to_use = 0;
 	tx_ring->next_to_clean = 0;
 	tx_ring->cpu = ena_irq->cpu;
+	tx_ring->numa_node = node;
 	return 0;
+
+err_push_buf_intermediate_buf:
+	vfree(tx_ring->free_ids);
+	tx_ring->free_ids = NULL;
+err_tx_free_ids:
+	vfree(tx_ring->tx_buffer_info);
+	tx_ring->tx_buffer_info = NULL;
+err_tx_buffer_info:
+	return -ENOMEM;
 }
 
 /* ena_free_tx_resources - Free I/O Tx Resources per Queue
@@ -257,20 +305,19 @@ static void ena_free_tx_resources(struct ena_adapter *adapter, int qid)
 	vfree(tx_ring->tx_buffer_info);
 	tx_ring->tx_buffer_info = NULL;
 
-	vfree(tx_ring->free_tx_ids);
-	tx_ring->free_tx_ids = NULL;
+	vfree(tx_ring->free_ids);
+	tx_ring->free_ids = NULL;
+
+	vfree(tx_ring->push_buf_intermediate_buf);
+	tx_ring->push_buf_intermediate_buf = NULL;
 }
 
-/* ena_setup_all_tx_resources - allocate I/O Tx queues resources for All queues
- * @adapter: private structure
- *
- * Return 0 on success, negative on failure
- */
-static int ena_setup_all_tx_resources(struct ena_adapter *adapter)
+int ena_setup_tx_resources_in_range(struct ena_adapter *adapter,
+				    int first_index, int count)
 {
 	int i, rc = 0;
 
-	for (i = 0; i < adapter->num_queues; i++) {
+	for (i = first_index; i < first_index + count; i++) {
 		rc = ena_setup_tx_resources(adapter, i);
 		if (rc)
 			goto err_setup_tx;
@@ -284,22 +331,31 @@ err_setup_tx:
 		  "Tx queue %d: allocation failed\n", i);
 
 	/* rewind the index freeing the rings as we go */
-	while (i--)
+	while (first_index < i--)
 		ena_free_tx_resources(adapter, i);
 	return rc;
 }
 
+void ena_free_all_io_tx_resources_in_range(struct ena_adapter *adapter,
+					   int first_index, int count)
+{
+	int i;
+
+	for (i = first_index; i < first_index + count; i++)
+		ena_free_tx_resources(adapter, i);
+}
+
 /* ena_free_all_io_tx_resources - Free I/O Tx Resources for All Queues
  * @adapter: board private structure
  *
  * Free all transmit software resources
  */
-static void ena_free_all_io_tx_resources(struct ena_adapter *adapter)
+void ena_free_all_io_tx_resources(struct ena_adapter *adapter)
 {
-	int i;
-
-	for (i = 0; i < adapter->num_queues; i++)
-		ena_free_tx_resources(adapter, i);
+	ena_free_all_io_tx_resources_in_range(adapter,
+					      0,
+					      adapter->xdp_num_queues +
+					      adapter->num_io_queues);
 }
 
 /* ena_setup_rx_resources - allocate I/O Rx resources (Descriptors)
@@ -313,7 +369,7 @@ static int ena_setup_rx_resources(struct ena_adapter *adapter,
 {
 	struct ena_ring *rx_ring = &adapter->rx_ring[qid];
 	struct ena_irq *ena_irq = &adapter->irq_tbl[ENA_IO_IRQ_IDX(qid)];
-	int size, node;
+	int size, node, i;
 
 	if (rx_ring->rx_buffer_info) {
 		netif_err(adapter, ifup, adapter->netdev,
@@ -334,12 +390,28 @@ static int ena_setup_rx_resources(struct ena_adapter *adapter,
 			return -ENOMEM;
 	}
 
+	size = sizeof(u16) * rx_ring->ring_size;
+	rx_ring->free_ids = vzalloc_node(size, node);
+	if (!rx_ring->free_ids) {
+		rx_ring->free_ids = vzalloc(size);
+		if (!rx_ring->free_ids) {
+			vfree(rx_ring->rx_buffer_info);
+			rx_ring->rx_buffer_info = NULL;
+			return -ENOMEM;
+		}
+	}
+
+	/* Req id ring for receiving RX pkts out of order */
+	for (i = 0; i < rx_ring->ring_size; i++)
+		rx_ring->free_ids[i] = i;
+
 	/* Reset rx statistics */
 	memset(&rx_ring->rx_stats, 0x0, sizeof(rx_ring->rx_stats));
 
 	rx_ring->next_to_clean = 0;
 	rx_ring->next_to_use = 0;
 	rx_ring->cpu = ena_irq->cpu;
+	rx_ring->numa_node = node;
 
 	return 0;
 }
@@ -357,6 +429,9 @@ static void ena_free_rx_resources(struct ena_adapter *adapter,
 
 	vfree(rx_ring->rx_buffer_info);
 	rx_ring->rx_buffer_info = NULL;
+
+	vfree(rx_ring->free_ids);
+	rx_ring->free_ids = NULL;
 }
 
 /* ena_setup_all_rx_resources - allocate I/O Rx queues resources for all queues
@@ -368,7 +443,7 @@ static int ena_setup_all_rx_resources(struct ena_adapter *adapter)
 {
 	int i, rc = 0;
 
-	for (i = 0; i < adapter->num_queues; i++) {
+	for (i = 0; i < adapter->num_io_queues; i++) {
 		rc = ena_setup_rx_resources(adapter, i);
 		if (rc)
 			goto err_setup_rx;
@@ -396,56 +471,87 @@ static void ena_free_all_io_rx_resources(struct ena_adapter *adapter)
 {
 	int i;
 
-	for (i = 0; i < adapter->num_queues; i++)
+	for (i = 0; i < adapter->num_io_queues; i++)
 		ena_free_rx_resources(adapter, i);
 }
 
-static inline int ena_alloc_rx_page(struct ena_ring *rx_ring,
-				    struct ena_rx_buffer *rx_info, gfp_t gfp)
+static struct page *ena_alloc_map_page(struct ena_ring *rx_ring,
+				       dma_addr_t *dma)
 {
+	struct page *page;
+
+	/* This would allocate the page on the same NUMA node the executing code
+	 * is running on.
+	 */
+	page = dev_alloc_page();
+	if (!page) {
+		ena_increase_stat(&rx_ring->rx_stats.page_alloc_fail, 1, &rx_ring->syncp);
+		return ERR_PTR(-ENOSPC);
+	}
+
+	/* To enable NIC-side port-mirroring, AKA SPAN port,
+	 * we make the buffer readable from the nic as well
+	 */
+	*dma = dma_map_page(rx_ring->dev, page, 0, ENA_PAGE_SIZE,
+			    DMA_BIDIRECTIONAL);
+	if (unlikely(dma_mapping_error(rx_ring->dev, *dma))) {
+		ena_increase_stat(&rx_ring->rx_stats.dma_mapping_err, 1,
+				  &rx_ring->syncp);
+		__free_page(page);
+		return ERR_PTR(-EIO);
+	}
+
+	return page;
+}
+
+static int ena_alloc_rx_buffer(struct ena_ring *rx_ring,
+			       struct ena_rx_buffer *rx_info)
+{
+	int headroom = rx_ring->rx_headroom;
 	struct ena_com_buf *ena_buf;
 	struct page *page;
 	dma_addr_t dma;
+	int tailroom;
+
+	/* restore page offset value in case it has been changed by device */
+	rx_info->buf_offset = headroom;
 
 	/* if previous allocated page is not used */
 	if (unlikely(rx_info->page))
 		return 0;
 
-	page = alloc_page(gfp);
-	if (unlikely(!page)) {
-		u64_stats_update_begin(&rx_ring->syncp);
-		rx_ring->rx_stats.page_alloc_fail++;
-		u64_stats_update_end(&rx_ring->syncp);
-		return -ENOMEM;
-	}
-
-	dma = dma_map_page(rx_ring->dev, page, 0, PAGE_SIZE,
-			   DMA_FROM_DEVICE);
-	if (unlikely(dma_mapping_error(rx_ring->dev, dma))) {
-		u64_stats_update_begin(&rx_ring->syncp);
-		rx_ring->rx_stats.dma_mapping_err++;
-		u64_stats_update_end(&rx_ring->syncp);
+	/* We handle DMA here */
+	page = ena_alloc_map_page(rx_ring, &dma);
+	if (IS_ERR(page))
+		return PTR_ERR(page);
 
-		__free_page(page);
-		return -EIO;
-	}
 	netif_dbg(rx_ring->adapter, rx_status, rx_ring->netdev,
-		  "alloc page %p, rx_info %p\n", page, rx_info);
+		  "Allocate page %p, rx_info %p\n", page, rx_info);
+
+	tailroom = SKB_DATA_ALIGN(sizeof(struct skb_shared_info));
 
 	rx_info->page = page;
+	rx_info->dma_addr = dma;
 	rx_info->page_offset = 0;
 	ena_buf = &rx_info->ena_buf;
-	ena_buf->paddr = dma;
-	ena_buf->len = PAGE_SIZE;
+	ena_buf->paddr = dma + headroom;
+	ena_buf->len = ENA_PAGE_SIZE - headroom - tailroom;
 
 	return 0;
 }
 
+static void ena_unmap_rx_buff_attrs(struct ena_ring *rx_ring,
+				    struct ena_rx_buffer *rx_info,
+				    unsigned long attrs)
+{
+	dma_unmap_page_attrs(rx_ring->dev, rx_info->dma_addr, ENA_PAGE_SIZE, DMA_BIDIRECTIONAL,
+			     attrs);
+}
+
 static void ena_free_rx_page(struct ena_ring *rx_ring,
 			     struct ena_rx_buffer *rx_info)
 {
 	struct page *page = rx_info->page;
-	struct ena_com_buf *ena_buf = &rx_info->ena_buf;
 
 	if (unlikely(!page)) {
 		netif_warn(rx_ring->adapter, rx_err, rx_ring->netdev,
@@ -453,8 +559,7 @@ static void ena_free_rx_page(struct ena_ring *rx_ring,
 		return;
 	}
 
-	dma_unmap_page(rx_ring->dev, ena_buf->paddr, PAGE_SIZE,
-		       DMA_FROM_DEVICE);
+	ena_unmap_rx_buff_attrs(rx_ring, rx_info, 0);
 
 	__free_page(page);
 	rx_info->page = NULL;
@@ -462,30 +567,32 @@ static void ena_free_rx_page(struct ena_ring *rx_ring,
 
 static int ena_refill_rx_bufs(struct ena_ring *rx_ring, u32 num)
 {
-	u16 next_to_use;
+	u16 next_to_use, req_id;
 	u32 i;
 	int rc;
 
 	next_to_use = rx_ring->next_to_use;
 
 	for (i = 0; i < num; i++) {
-		struct ena_rx_buffer *rx_info =
-			&rx_ring->rx_buffer_info[next_to_use];
+		struct ena_rx_buffer *rx_info;
+
+		req_id = rx_ring->free_ids[next_to_use];
 
-		rc = ena_alloc_rx_page(rx_ring, rx_info,
-				       __GFP_COLD | GFP_ATOMIC | __GFP_COMP);
+		rx_info = &rx_ring->rx_buffer_info[req_id];
+
+		rc = ena_alloc_rx_buffer(rx_ring, rx_info);
 		if (unlikely(rc < 0)) {
 			netif_warn(rx_ring->adapter, rx_err, rx_ring->netdev,
-				   "failed to alloc buffer for rx queue %d\n",
+				   "Failed to allocate buffer for rx queue %d\n",
 				   rx_ring->qid);
 			break;
 		}
 		rc = ena_com_add_single_rx_desc(rx_ring->ena_com_io_sq,
 						&rx_info->ena_buf,
-						next_to_use);
+						req_id);
 		if (unlikely(rc)) {
 			netif_warn(rx_ring->adapter, rx_status, rx_ring->netdev,
-				   "failed to add buffer for rx queue %d\n",
+				   "Failed to add buffer for rx queue %d\n",
 				   rx_ring->qid);
 			break;
 		}
@@ -494,21 +601,16 @@ static int ena_refill_rx_bufs(struct ena_ring *rx_ring, u32 num)
 	}
 
 	if (unlikely(i < num)) {
-		u64_stats_update_begin(&rx_ring->syncp);
-		rx_ring->rx_stats.refil_partial++;
-		u64_stats_update_end(&rx_ring->syncp);
-		netdev_warn(rx_ring->netdev,
-			    "refilled rx qid %d with only %d buffers (from %d)\n",
-			    rx_ring->qid, i, num);
+		ena_increase_stat(&rx_ring->rx_stats.refil_partial, 1,
+				  &rx_ring->syncp);
+		netif_warn(rx_ring->adapter, rx_err, rx_ring->netdev,
+			   "Refilled rx qid %d with only %d buffers (from %d)\n",
+			   rx_ring->qid, i, num);
 	}
 
-	if (likely(i)) {
-		/* Add memory barrier to make sure the desc were written before
-		 * issue a doorbell
-		 */
-		wmb();
+	/* ena_com_write_sq_doorbell issues a wmb() */
+	if (likely(i))
 		ena_com_write_sq_doorbell(rx_ring->ena_com_io_sq);
-	}
 
 	rx_ring->next_to_use = next_to_use;
 
@@ -531,21 +633,20 @@ static void ena_free_rx_bufs(struct ena_adapter *adapter,
 
 /* ena_refill_all_rx_bufs - allocate all queues Rx buffers
  * @adapter: board private structure
- *
  */
 static void ena_refill_all_rx_bufs(struct ena_adapter *adapter)
 {
 	struct ena_ring *rx_ring;
 	int i, rc, bufs_num;
 
-	for (i = 0; i < adapter->num_queues; i++) {
+	for (i = 0; i < adapter->num_io_queues; i++) {
 		rx_ring = &adapter->rx_ring[i];
 		bufs_num = rx_ring->ring_size - 1;
 		rc = ena_refill_rx_bufs(rx_ring, bufs_num);
 
 		if (unlikely(rc != bufs_num))
 			netif_warn(rx_ring->adapter, rx_status, rx_ring->netdev,
-				   "refilling Queue %d failed. allocated %d buffers from: %d\n",
+				   "Refilling Queue %d failed. allocated %d buffers from: %d\n",
 				   i, rc, bufs_num);
 	}
 }
@@ -554,58 +655,79 @@ static void ena_free_all_rx_bufs(struct ena_adapter *adapter)
 {
 	int i;
 
-	for (i = 0; i < adapter->num_queues; i++)
+	for (i = 0; i < adapter->num_io_queues; i++)
 		ena_free_rx_bufs(adapter, i);
 }
 
+void ena_unmap_tx_buff(struct ena_ring *tx_ring,
+		       struct ena_tx_buffer *tx_info)
+{
+	struct ena_com_buf *ena_buf;
+	u32 cnt;
+	int i;
+
+	ena_buf = tx_info->bufs;
+	cnt = tx_info->num_of_bufs;
+
+	if (unlikely(!cnt))
+		return;
+
+	if (tx_info->map_linear_data) {
+		dma_unmap_single(tx_ring->dev,
+				 dma_unmap_addr(ena_buf, paddr),
+				 dma_unmap_len(ena_buf, len),
+				 DMA_TO_DEVICE);
+		ena_buf++;
+		cnt--;
+	}
+
+	/* unmap remaining mapped pages */
+	for (i = 0; i < cnt; i++) {
+		dma_unmap_page(tx_ring->dev, dma_unmap_addr(ena_buf, paddr),
+			       dma_unmap_len(ena_buf, len), DMA_TO_DEVICE);
+		ena_buf++;
+	}
+}
+
 /* ena_free_tx_bufs - Free Tx Buffers per Queue
  * @tx_ring: TX ring for which buffers be freed
  */
 static void ena_free_tx_bufs(struct ena_ring *tx_ring)
 {
 	bool print_once = true;
+	bool is_xdp_ring;
 	u32 i;
 
+	is_xdp_ring = ENA_IS_XDP_INDEX(tx_ring->adapter, tx_ring->qid);
+
 	for (i = 0; i < tx_ring->ring_size; i++) {
 		struct ena_tx_buffer *tx_info = &tx_ring->tx_buffer_info[i];
-		struct ena_com_buf *ena_buf;
-		int nr_frags;
-		int j;
 
 		if (!tx_info->skb)
 			continue;
 
 		if (print_once) {
-			netdev_notice(tx_ring->netdev,
-				      "free uncompleted tx skb qid %d idx 0x%x\n",
-				      tx_ring->qid, i);
+			netif_notice(tx_ring->adapter, ifdown, tx_ring->netdev,
+				     "Free uncompleted tx skb qid %d idx 0x%x\n",
+				     tx_ring->qid, i);
 			print_once = false;
 		} else {
-			netdev_dbg(tx_ring->netdev,
-				   "free uncompleted tx skb qid %d idx 0x%x\n",
-				   tx_ring->qid, i);
+			netif_dbg(tx_ring->adapter, ifdown, tx_ring->netdev,
+				  "Free uncompleted tx skb qid %d idx 0x%x\n",
+				  tx_ring->qid, i);
 		}
 
-		ena_buf = tx_info->bufs;
-		dma_unmap_single(tx_ring->dev,
-				 ena_buf->paddr,
-				 ena_buf->len,
-				 DMA_TO_DEVICE);
-
-		/* unmap remaining mapped pages */
-		nr_frags = tx_info->num_of_bufs - 1;
-		for (j = 0; j < nr_frags; j++) {
-			ena_buf++;
-			dma_unmap_page(tx_ring->dev,
-				       ena_buf->paddr,
-				       ena_buf->len,
-				       DMA_TO_DEVICE);
-		}
+		ena_unmap_tx_buff(tx_ring, tx_info);
 
-		dev_kfree_skb_any(tx_info->skb);
+		if (is_xdp_ring)
+			xdp_return_frame(tx_info->xdpf);
+		else
+			dev_kfree_skb_any(tx_info->skb);
 	}
-	netdev_tx_reset_queue(netdev_get_tx_queue(tx_ring->netdev,
-						  tx_ring->qid));
+
+	if (!is_xdp_ring)
+		netdev_tx_reset_queue(netdev_get_tx_queue(tx_ring->netdev,
+							  tx_ring->qid));
 }
 
 static void ena_free_all_tx_bufs(struct ena_adapter *adapter)
@@ -613,7 +735,7 @@ static void ena_free_all_tx_bufs(struct ena_adapter *adapter)
 	struct ena_ring *tx_ring;
 	int i;
 
-	for (i = 0; i < adapter->num_queues; i++) {
+	for (i = 0; i < adapter->num_io_queues + adapter->xdp_num_queues; i++) {
 		tx_ring = &adapter->tx_ring[i];
 		ena_free_tx_bufs(tx_ring);
 	}
@@ -624,7 +746,7 @@ static void ena_destroy_all_tx_queues(struct ena_adapter *adapter)
 	u16 ena_qid;
 	int i;
 
-	for (i = 0; i < adapter->num_queues; i++) {
+	for (i = 0; i < adapter->num_io_queues + adapter->xdp_num_queues; i++) {
 		ena_qid = ENA_IO_TXQ_IDX(i);
 		ena_com_destroy_io_queue(adapter->ena_dev, ena_qid);
 	}
@@ -635,8 +757,10 @@ static void ena_destroy_all_rx_queues(struct ena_adapter *adapter)
 	u16 ena_qid;
 	int i;
 
-	for (i = 0; i < adapter->num_queues; i++) {
+	for (i = 0; i < adapter->num_io_queues; i++) {
 		ena_qid = ENA_IO_RXQ_IDX(i);
+		cancel_work_sync(&adapter->ena_napi[i].dim.work);
+		ena_xdp_unregister_rxq_info(&adapter->rx_ring[i]);
 		ena_com_destroy_io_queue(adapter->ena_dev, ena_qid);
 	}
 }
@@ -647,32 +771,39 @@ static void ena_destroy_all_io_queues(struct ena_adapter *adapter)
 	ena_destroy_all_rx_queues(adapter);
 }
 
-static int validate_tx_req_id(struct ena_ring *tx_ring, u16 req_id)
+int handle_invalid_req_id(struct ena_ring *ring, u16 req_id,
+			  struct ena_tx_buffer *tx_info, bool is_xdp)
 {
-	struct ena_tx_buffer *tx_info = NULL;
-
-	if (likely(req_id < tx_ring->ring_size)) {
-		tx_info = &tx_ring->tx_buffer_info[req_id];
-		if (likely(tx_info->skb))
-			return 0;
-	}
-
 	if (tx_info)
-		netif_err(tx_ring->adapter, tx_done, tx_ring->netdev,
-			  "tx_info doesn't have valid skb\n");
+		netif_err(ring->adapter,
+			  tx_done,
+			  ring->netdev,
+			  "tx_info doesn't have valid %s. qid %u req_id %u",
+			   is_xdp ? "xdp frame" : "skb", ring->qid, req_id);
 	else
-		netif_err(tx_ring->adapter, tx_done, tx_ring->netdev,
-			  "Invalid req_id: %hu\n", req_id);
+		netif_err(ring->adapter,
+			  tx_done,
+			  ring->netdev,
+			  "Invalid req_id %u in qid %u\n",
+			  req_id, ring->qid);
 
-	u64_stats_update_begin(&tx_ring->syncp);
-	tx_ring->tx_stats.bad_req_id++;
-	u64_stats_update_end(&tx_ring->syncp);
+	ena_increase_stat(&ring->tx_stats.bad_req_id, 1, &ring->syncp);
+	ena_reset_device(ring->adapter, ENA_REGS_RESET_INV_TX_REQ_ID);
 
-	/* Trigger device reset */
-	set_bit(ENA_FLAG_TRIGGER_RESET, &tx_ring->adapter->flags);
 	return -EFAULT;
 }
 
+static int validate_tx_req_id(struct ena_ring *tx_ring, u16 req_id)
+{
+	struct ena_tx_buffer *tx_info;
+
+	tx_info = &tx_ring->tx_buffer_info[req_id];
+	if (likely(tx_info->skb))
+		return 0;
+
+	return handle_invalid_req_id(tx_ring, req_id, tx_info, false);
+}
+
 static int ena_clean_tx_irq(struct ena_ring *tx_ring, u32 budget)
 {
 	struct netdev_queue *txq;
@@ -690,14 +821,16 @@ static int ena_clean_tx_irq(struct ena_ring *tx_ring, u32 budget)
 	while (tx_pkts < budget) {
 		struct ena_tx_buffer *tx_info;
 		struct sk_buff *skb;
-		struct ena_com_buf *ena_buf;
-		int i, nr_frags;
 
 		rc = ena_com_tx_comp_req_id_get(tx_ring->ena_com_io_cq,
 						&req_id);
-		if (rc)
+		if (rc) {
+			if (unlikely(rc == -EINVAL))
+				handle_invalid_req_id(tx_ring, req_id, NULL, false);
 			break;
+		}
 
+		/* validate that the request id points to a valid skb */
 		rc = validate_tx_req_id(tx_ring, req_id);
 		if (rc)
 			break;
@@ -711,42 +844,24 @@ static int ena_clean_tx_irq(struct ena_ring *tx_ring, u32 budget)
 		tx_info->skb = NULL;
 		tx_info->last_jiffies = 0;
 
-		if (likely(tx_info->num_of_bufs != 0)) {
-			ena_buf = tx_info->bufs;
-
-			dma_unmap_single(tx_ring->dev,
-					 dma_unmap_addr(ena_buf, paddr),
-					 dma_unmap_len(ena_buf, len),
-					 DMA_TO_DEVICE);
-
-			/* unmap remaining mapped pages */
-			nr_frags = tx_info->num_of_bufs - 1;
-			for (i = 0; i < nr_frags; i++) {
-				ena_buf++;
-				dma_unmap_page(tx_ring->dev,
-					       dma_unmap_addr(ena_buf, paddr),
-					       dma_unmap_len(ena_buf, len),
-					       DMA_TO_DEVICE);
-			}
-		}
+		ena_unmap_tx_buff(tx_ring, tx_info);
 
 		netif_dbg(tx_ring->adapter, tx_done, tx_ring->netdev,
 			  "tx_poll: q %d skb %p completed\n", tx_ring->qid,
 			  skb);
 
-		tx_bytes += skb->len;
+		tx_bytes += tx_info->total_tx_size;
 		dev_kfree_skb(skb);
 		tx_pkts++;
 		total_done += tx_info->tx_descs;
 
-		tx_ring->free_tx_ids[next_to_clean] = req_id;
+		tx_ring->free_ids[next_to_clean] = req_id;
 		next_to_clean = ENA_TX_RING_IDX_NEXT(next_to_clean,
 						     tx_ring->ring_size);
 	}
 
 	tx_ring->next_to_clean = next_to_clean;
 	ena_com_comp_ack(tx_ring->ena_com_io_sq, total_done);
-	ena_com_update_dev_comp_head(tx_ring->ena_com_io_cq);
 
 	netdev_tx_completed_queue(txq, tx_pkts, tx_bytes);
 
@@ -759,43 +874,94 @@ static int ena_clean_tx_irq(struct ena_ring *tx_ring, u32 budget)
 	 */
 	smp_mb();
 
-	above_thresh = ena_com_sq_empty_space(tx_ring->ena_com_io_sq) >
-		ENA_TX_WAKEUP_THRESH;
+	above_thresh = ena_com_sq_have_enough_space(tx_ring->ena_com_io_sq,
+						    ENA_TX_WAKEUP_THRESH);
 	if (unlikely(netif_tx_queue_stopped(txq) && above_thresh)) {
 		__netif_tx_lock(txq, smp_processor_id());
-		above_thresh = ena_com_sq_empty_space(tx_ring->ena_com_io_sq) >
-			ENA_TX_WAKEUP_THRESH;
-		if (netif_tx_queue_stopped(txq) && above_thresh) {
+		above_thresh =
+			ena_com_sq_have_enough_space(tx_ring->ena_com_io_sq,
+						     ENA_TX_WAKEUP_THRESH);
+		if (netif_tx_queue_stopped(txq) && above_thresh &&
+		    test_bit(ENA_FLAG_DEV_UP, &tx_ring->adapter->flags)) {
 			netif_tx_wake_queue(txq);
-			u64_stats_update_begin(&tx_ring->syncp);
-			tx_ring->tx_stats.queue_wakeup++;
-			u64_stats_update_end(&tx_ring->syncp);
+			ena_increase_stat(&tx_ring->tx_stats.queue_wakeup, 1,
+					  &tx_ring->syncp);
 		}
 		__netif_tx_unlock(txq);
 	}
 
-	tx_ring->per_napi_bytes += tx_bytes;
-	tx_ring->per_napi_packets += tx_pkts;
-
 	return tx_pkts;
 }
 
+static struct sk_buff *ena_alloc_skb(struct ena_ring *rx_ring, void *first_frag, u16 len)
+{
+	struct sk_buff *skb;
+
+	if (!first_frag)
+		skb = napi_alloc_skb(rx_ring->napi, len);
+	else
+		skb = napi_build_skb(first_frag, len);
+
+	if (unlikely(!skb)) {
+		ena_increase_stat(&rx_ring->rx_stats.skb_alloc_fail, 1,
+				  &rx_ring->syncp);
+
+		netif_dbg(rx_ring->adapter, rx_err, rx_ring->netdev,
+			  "Failed to allocate skb. first_frag %s\n",
+			  first_frag ? "provided" : "not provided");
+	}
+
+	return skb;
+}
+
+static bool ena_try_rx_buf_page_reuse(struct ena_rx_buffer *rx_info, u16 buf_len,
+				      u16 len, int pkt_offset)
+{
+	struct ena_com_buf *ena_buf = &rx_info->ena_buf;
+
+	/* More than ENA_MIN_RX_BUF_SIZE left in the reused buffer
+	 * for data + headroom + tailroom.
+	 */
+	if (SKB_DATA_ALIGN(len + pkt_offset) + ENA_MIN_RX_BUF_SIZE <= ena_buf->len) {
+		page_ref_inc(rx_info->page);
+		rx_info->page_offset += buf_len;
+		ena_buf->paddr += buf_len;
+		ena_buf->len -= buf_len;
+		return true;
+	}
+
+	return false;
+}
+
 static struct sk_buff *ena_rx_skb(struct ena_ring *rx_ring,
 				  struct ena_com_rx_buf_info *ena_bufs,
 				  u32 descs,
 				  u16 *next_to_clean)
 {
+	int tailroom = SKB_DATA_ALIGN(sizeof(struct skb_shared_info));
+	bool is_xdp_loaded = ena_xdp_present_ring(rx_ring);
+	struct ena_rx_buffer *rx_info;
+	struct ena_adapter *adapter;
+	int page_offset, pkt_offset;
+	dma_addr_t pre_reuse_paddr;
+	u16 len, req_id, buf = 0;
+	bool reuse_rx_buf_page;
 	struct sk_buff *skb;
-	struct ena_rx_buffer *rx_info =
-		&rx_ring->rx_buffer_info[*next_to_clean];
-	u32 len;
-	u32 buf = 0;
-	void *va;
+	void *buf_addr;
+	int buf_offset;
+	u16 buf_len;
+
+	len = ena_bufs[buf].len;
+	req_id = ena_bufs[buf].req_id;
+
+	rx_info = &rx_ring->rx_buffer_info[req_id];
 
-	len = ena_bufs[0].len;
 	if (unlikely(!rx_info->page)) {
-		netif_err(rx_ring->adapter, rx_err, rx_ring->netdev,
-			  "Page is NULL\n");
+		adapter = rx_ring->adapter;
+		netif_err(adapter, rx_err, rx_ring->netdev,
+			  "Page is NULL. qid %u req_id %u\n", rx_ring->qid, req_id);
+		ena_increase_stat(&rx_ring->rx_stats.bad_req_id, 1, &rx_ring->syncp);
+		ena_reset_device(adapter, ENA_REGS_RESET_INV_RX_REQ_ID);
 		return NULL;
 	}
 
@@ -803,74 +969,93 @@ static struct sk_buff *ena_rx_skb(struct ena_ring *rx_ring,
 		  "rx_info %p page %p\n",
 		  rx_info, rx_info->page);
 
-	/* save virt address of first buffer */
-	va = page_address(rx_info->page) + rx_info->page_offset;
-	prefetch(va + NET_IP_ALIGN);
+	buf_offset = rx_info->buf_offset;
+	pkt_offset = buf_offset - rx_ring->rx_headroom;
+	page_offset = rx_info->page_offset;
+	buf_addr = page_address(rx_info->page) + page_offset;
 
 	if (len <= rx_ring->rx_copybreak) {
-		skb = netdev_alloc_skb_ip_align(rx_ring->netdev,
-						rx_ring->rx_copybreak);
-		if (unlikely(!skb)) {
-			u64_stats_update_begin(&rx_ring->syncp);
-			rx_ring->rx_stats.skb_alloc_fail++;
-			u64_stats_update_end(&rx_ring->syncp);
-			netif_err(rx_ring->adapter, rx_err, rx_ring->netdev,
-				  "Failed to allocate skb\n");
+		skb = ena_alloc_skb(rx_ring, NULL, len);
+		if (unlikely(!skb))
 			return NULL;
-		}
-
-		netif_dbg(rx_ring->adapter, rx_status, rx_ring->netdev,
-			  "rx allocated small packet. len %d. data_len %d\n",
-			  skb->len, skb->data_len);
 
-		/* sync this buffer for CPU use */
-		dma_sync_single_for_cpu(rx_ring->dev,
-					dma_unmap_addr(&rx_info->ena_buf, paddr),
-					len,
-					DMA_FROM_DEVICE);
-		skb_copy_to_linear_data(skb, va, len);
+		skb_copy_to_linear_data(skb, buf_addr + buf_offset, len);
 		dma_sync_single_for_device(rx_ring->dev,
-					   dma_unmap_addr(&rx_info->ena_buf, paddr),
+					   dma_unmap_addr(&rx_info->ena_buf, paddr) + pkt_offset,
 					   len,
 					   DMA_FROM_DEVICE);
 
 		skb_put(skb, len);
+		netif_dbg(rx_ring->adapter, rx_status, rx_ring->netdev,
+			  "RX allocated small packet. len %d.\n", skb->len);
 		skb->protocol = eth_type_trans(skb, rx_ring->netdev);
+		rx_ring->free_ids[*next_to_clean] = req_id;
 		*next_to_clean = ENA_RX_RING_IDX_ADD(*next_to_clean, descs,
 						     rx_ring->ring_size);
 		return skb;
 	}
 
-	skb = napi_get_frags(rx_ring->napi);
-	if (unlikely(!skb)) {
-		netif_dbg(rx_ring->adapter, rx_status, rx_ring->netdev,
-			  "Failed allocating skb\n");
-		u64_stats_update_begin(&rx_ring->syncp);
-		rx_ring->rx_stats.skb_alloc_fail++;
-		u64_stats_update_end(&rx_ring->syncp);
-		return NULL;
-	}
+	buf_len = SKB_DATA_ALIGN(len + buf_offset + tailroom);
 
-	do {
-		dma_unmap_page(rx_ring->dev,
-			       dma_unmap_addr(&rx_info->ena_buf, paddr),
-			       PAGE_SIZE, DMA_FROM_DEVICE);
+	/* If XDP isn't loaded try to reuse part of the RX buffer */
+	reuse_rx_buf_page = !is_xdp_loaded &&
+			    ena_try_rx_buf_page_reuse(rx_info, buf_len, len, pkt_offset);
 
-		skb_add_rx_frag(skb, skb_shinfo(skb)->nr_frags, rx_info->page,
-				rx_info->page_offset, len, PAGE_SIZE);
+	if (!reuse_rx_buf_page)
+		ena_unmap_rx_buff_attrs(rx_ring, rx_info, DMA_ATTR_SKIP_CPU_SYNC);
 
+	skb = ena_alloc_skb(rx_ring, buf_addr, buf_len);
+	if (unlikely(!skb))
+		return NULL;
+
+	/* Populate skb's linear part */
+	skb_reserve(skb, buf_offset);
+	skb_put(skb, len);
+	skb->protocol = eth_type_trans(skb, rx_ring->netdev);
+
+	do {
 		netif_dbg(rx_ring->adapter, rx_status, rx_ring->netdev,
-			  "rx skb updated. len %d. data_len %d\n",
+			  "RX skb updated. len %d. data_len %d\n",
 			  skb->len, skb->data_len);
 
-		rx_info->page = NULL;
+		if (!reuse_rx_buf_page)
+			rx_info->page = NULL;
+
+		rx_ring->free_ids[*next_to_clean] = req_id;
 		*next_to_clean =
 			ENA_RX_RING_IDX_NEXT(*next_to_clean,
 					     rx_ring->ring_size);
 		if (likely(--descs == 0))
 			break;
-		rx_info = &rx_ring->rx_buffer_info[*next_to_clean];
-		len = ena_bufs[++buf].len;
+
+		buf++;
+		len = ena_bufs[buf].len;
+		req_id = ena_bufs[buf].req_id;
+
+		rx_info = &rx_ring->rx_buffer_info[req_id];
+
+		/* rx_info->buf_offset includes rx_ring->rx_headroom */
+		buf_offset = rx_info->buf_offset;
+		pkt_offset = buf_offset - rx_ring->rx_headroom;
+		buf_len = SKB_DATA_ALIGN(len + buf_offset + tailroom);
+		page_offset = rx_info->page_offset;
+
+		pre_reuse_paddr = dma_unmap_addr(&rx_info->ena_buf, paddr);
+
+		reuse_rx_buf_page = !is_xdp_loaded &&
+				    ena_try_rx_buf_page_reuse(rx_info, buf_len, len, pkt_offset);
+
+		dma_sync_single_for_cpu(rx_ring->dev,
+					pre_reuse_paddr + pkt_offset,
+					len,
+					DMA_FROM_DEVICE);
+
+		if (!reuse_rx_buf_page)
+			ena_unmap_rx_buff_attrs(rx_ring, rx_info, DMA_ATTR_SKIP_CPU_SYNC);
+
+		skb_add_rx_frag(skb, skb_shinfo(skb)->nr_frags, rx_info->page,
+				page_offset + buf_offset, len, buf_len);
+
 	} while (1);
 
 	return skb;
@@ -881,7 +1066,7 @@ static struct sk_buff *ena_rx_skb(struct ena_ring *rx_ring,
  * @ena_rx_ctx: received packet context/metadata
  * @skb: skb currently being received and modified
  */
-static inline void ena_rx_checksum(struct ena_ring *rx_ring,
+static void ena_rx_checksum(struct ena_ring *rx_ring,
 				   struct ena_com_rx_ctx *ena_rx_ctx,
 				   struct sk_buff *skb)
 {
@@ -902,10 +1087,9 @@ static inline void ena_rx_checksum(struct ena_ring *rx_ring,
 		     (ena_rx_ctx->l3_csum_err))) {
 		/* ipv4 checksum error */
 		skb->ip_summed = CHECKSUM_NONE;
-		u64_stats_update_begin(&rx_ring->syncp);
-		rx_ring->rx_stats.bad_csum++;
-		u64_stats_update_end(&rx_ring->syncp);
-		netif_err(rx_ring->adapter, rx_err, rx_ring->netdev,
+		ena_increase_stat(&rx_ring->rx_stats.csum_bad, 1,
+				  &rx_ring->syncp);
+		netif_dbg(rx_ring->adapter, rx_err, rx_ring->netdev,
 			  "RX IPv4 header checksum error\n");
 		return;
 	}
@@ -915,17 +1099,28 @@ static inline void ena_rx_checksum(struct ena_ring *rx_ring,
 		   (ena_rx_ctx->l4_proto == ENA_ETH_IO_L4_PROTO_UDP))) {
 		if (unlikely(ena_rx_ctx->l4_csum_err)) {
 			/* TCP/UDP checksum error */
-			u64_stats_update_begin(&rx_ring->syncp);
-			rx_ring->rx_stats.bad_csum++;
-			u64_stats_update_end(&rx_ring->syncp);
-			netif_err(rx_ring->adapter, rx_err, rx_ring->netdev,
+			ena_increase_stat(&rx_ring->rx_stats.csum_bad, 1,
+					  &rx_ring->syncp);
+			netif_dbg(rx_ring->adapter, rx_err, rx_ring->netdev,
 				  "RX L4 checksum error\n");
 			skb->ip_summed = CHECKSUM_NONE;
 			return;
 		}
 
-		skb->ip_summed = CHECKSUM_UNNECESSARY;
+		if (likely(ena_rx_ctx->l4_csum_checked)) {
+			skb->ip_summed = CHECKSUM_UNNECESSARY;
+			ena_increase_stat(&rx_ring->rx_stats.csum_good, 1,
+					  &rx_ring->syncp);
+		} else {
+			ena_increase_stat(&rx_ring->rx_stats.csum_unchecked, 1,
+					  &rx_ring->syncp);
+			skb->ip_summed = CHECKSUM_NONE;
+		}
+	} else {
+		skb->ip_summed = CHECKSUM_NONE;
+		return;
 	}
+
 }
 
 static void ena_set_rx_hash(struct ena_ring *rx_ring,
@@ -950,6 +1145,35 @@ static void ena_set_rx_hash(struct ena_ring *rx_ring,
 	}
 }
 
+static int ena_xdp_handle_buff(struct ena_ring *rx_ring, struct xdp_buff *xdp, u16 num_descs)
+{
+	struct ena_rx_buffer *rx_info;
+	int ret;
+
+	/* XDP multi-buffer packets not supported */
+	if (unlikely(num_descs > 1)) {
+		netdev_err_once(rx_ring->adapter->netdev,
+				"xdp: dropped unsupported multi-buffer packets\n");
+		ena_increase_stat(&rx_ring->rx_stats.xdp_drop, 1, &rx_ring->syncp);
+		return ENA_XDP_DROP;
+	}
+
+	rx_info = &rx_ring->rx_buffer_info[rx_ring->ena_bufs[0].req_id];
+	xdp_prepare_buff(xdp, page_address(rx_info->page),
+			 rx_info->buf_offset,
+			 rx_ring->ena_bufs[0].len, false);
+
+	ret = ena_xdp_execute(rx_ring, xdp);
+
+	/* The xdp program might expand the headers */
+	if (ret == ENA_XDP_PASS) {
+		rx_info->buf_offset = xdp->data - xdp->data_hard_start;
+		rx_ring->ena_bufs[0].len = xdp->data_end - xdp->data;
+	}
+
+	return ret;
+}
+
 /* ena_clean_rx_irq - Cleanup RX irq
  * @rx_ring: RX ring to clean
  * @napi: napi handler
@@ -961,25 +1185,34 @@ static int ena_clean_rx_irq(struct ena_ring *rx_ring, struct napi_struct *napi,
 			    u32 budget)
 {
 	u16 next_to_clean = rx_ring->next_to_clean;
-	u32 res_budget, work_done;
-
 	struct ena_com_rx_ctx ena_rx_ctx;
+	struct ena_rx_buffer *rx_info;
 	struct ena_adapter *adapter;
+	u32 res_budget, work_done;
+	int rx_copybreak_pkt = 0;
+	int refill_threshold;
 	struct sk_buff *skb;
 	int refill_required;
-	int refill_threshold;
-	int rc = 0;
+	struct xdp_buff xdp;
+	int xdp_flags = 0;
 	int total_len = 0;
-	int rx_copybreak_pkt = 0;
+	int xdp_verdict;
+	u8 pkt_offset;
+	int rc = 0;
+	int i;
 
 	netif_dbg(rx_ring->adapter, rx_status, rx_ring->netdev,
 		  "%s qid %d\n", __func__, rx_ring->qid);
 	res_budget = budget;
+	xdp_init_buff(&xdp, ENA_PAGE_SIZE, &rx_ring->xdp_rxq);
 
 	do {
+		xdp_verdict = ENA_XDP_PASS;
+		skb = NULL;
 		ena_rx_ctx.ena_bufs = rx_ring->ena_bufs;
 		ena_rx_ctx.max_bufs = rx_ring->sgl_size;
 		ena_rx_ctx.descs = 0;
+		ena_rx_ctx.pkt_offset = 0;
 		rc = ena_com_rx_pkt(rx_ring->ena_com_io_cq,
 				    rx_ring->ena_com_io_sq,
 				    &ena_rx_ctx);
@@ -989,20 +1222,56 @@ static int ena_clean_rx_irq(struct ena_ring *rx_ring, struct napi_struct *napi,
 		if (unlikely(ena_rx_ctx.descs == 0))
 			break;
 
+		/* First descriptor might have an offset set by the device */
+		rx_info = &rx_ring->rx_buffer_info[rx_ring->ena_bufs[0].req_id];
+		pkt_offset = ena_rx_ctx.pkt_offset;
+		rx_info->buf_offset += pkt_offset;
+
 		netif_dbg(rx_ring->adapter, rx_status, rx_ring->netdev,
 			  "rx_poll: q %d got packet from ena. descs #: %d l3 proto %d l4 proto %d hash: %x\n",
 			  rx_ring->qid, ena_rx_ctx.descs, ena_rx_ctx.l3_proto,
 			  ena_rx_ctx.l4_proto, ena_rx_ctx.hash);
 
+		dma_sync_single_for_cpu(rx_ring->dev,
+					dma_unmap_addr(&rx_info->ena_buf, paddr) + pkt_offset,
+					rx_ring->ena_bufs[0].len,
+					DMA_FROM_DEVICE);
+
+		if (ena_xdp_present_ring(rx_ring))
+			xdp_verdict = ena_xdp_handle_buff(rx_ring, &xdp, ena_rx_ctx.descs);
+
 		/* allocate skb and fill it */
-		skb = ena_rx_skb(rx_ring, rx_ring->ena_bufs, ena_rx_ctx.descs,
-				 &next_to_clean);
+		if (xdp_verdict == ENA_XDP_PASS)
+			skb = ena_rx_skb(rx_ring,
+					 rx_ring->ena_bufs,
+					 ena_rx_ctx.descs,
+					 &next_to_clean);
 
-		/* exit if we failed to retrieve a buffer */
 		if (unlikely(!skb)) {
-			next_to_clean = ENA_RX_RING_IDX_ADD(next_to_clean,
-							    ena_rx_ctx.descs,
-							    rx_ring->ring_size);
+			for (i = 0; i < ena_rx_ctx.descs; i++) {
+				int req_id = rx_ring->ena_bufs[i].req_id;
+
+				rx_ring->free_ids[next_to_clean] = req_id;
+				next_to_clean =
+					ENA_RX_RING_IDX_NEXT(next_to_clean,
+							     rx_ring->ring_size);
+
+				/* Packets was passed for transmission, unmap it
+				 * from RX side.
+				 */
+				if (xdp_verdict & ENA_XDP_FORWARDED) {
+					ena_unmap_rx_buff_attrs(rx_ring,
+								&rx_ring->rx_buffer_info[req_id],
+								DMA_ATTR_SKIP_CPU_SYNC);
+					rx_ring->rx_buffer_info[req_id].page = NULL;
+				}
+			}
+			if (xdp_verdict != ENA_XDP_PASS) {
+				xdp_flags |= xdp_verdict;
+				total_len += ena_rx_ctx.ena_bufs[0].len;
+				res_budget--;
+				continue;
+			}
 			break;
 		}
 
@@ -1012,20 +1281,17 @@ static int ena_clean_rx_irq(struct ena_ring *rx_ring, struct napi_struct *napi,
 
 		skb_record_rx_queue(skb, rx_ring->qid);
 
-		if (rx_ring->ena_bufs[0].len <= rx_ring->rx_copybreak) {
-			total_len += rx_ring->ena_bufs[0].len;
+		if (rx_ring->ena_bufs[0].len <= rx_ring->rx_copybreak)
 			rx_copybreak_pkt++;
-			napi_gro_receive(napi, skb);
-		} else {
-			total_len += skb->len;
-			napi_gro_frags(napi);
-		}
+
+		total_len += skb->len;
+
+		napi_gro_receive(napi, skb);
 
 		res_budget--;
 	} while (likely(res_budget));
 
 	work_done = budget - res_budget;
-	rx_ring->per_napi_bytes += total_len;
 	rx_ring->per_napi_packets += work_done;
 	u64_stats_update_begin(&rx_ring->syncp);
 	rx_ring->rx_stats.bytes += total_len;
@@ -1035,51 +1301,105 @@ static int ena_clean_rx_irq(struct ena_ring *rx_ring, struct napi_struct *napi,
 
 	rx_ring->next_to_clean = next_to_clean;
 
-	refill_required = ena_com_sq_empty_space(rx_ring->ena_com_io_sq);
-	refill_threshold = rx_ring->ring_size / ENA_RX_REFILL_THRESH_DIVIDER;
+	refill_required = ena_com_free_q_entries(rx_ring->ena_com_io_sq);
+	refill_threshold =
+		min_t(int, rx_ring->ring_size / ENA_RX_REFILL_THRESH_DIVIDER,
+		      ENA_RX_REFILL_THRESH_PACKET);
 
 	/* Optimization, try to batch new rx buffers */
-	if (refill_required > refill_threshold) {
-		ena_com_update_dev_comp_head(rx_ring->ena_com_io_cq);
+	if (refill_required > refill_threshold)
 		ena_refill_rx_bufs(rx_ring, refill_required);
-	}
+
+	if (xdp_flags & ENA_XDP_REDIRECT)
+		xdp_do_flush();
 
 	return work_done;
 
 error:
+	if (xdp_flags & ENA_XDP_REDIRECT)
+		xdp_do_flush();
+
 	adapter = netdev_priv(rx_ring->netdev);
 
-	u64_stats_update_begin(&rx_ring->syncp);
-	rx_ring->rx_stats.bad_desc_num++;
-	u64_stats_update_end(&rx_ring->syncp);
+	if (rc == -ENOSPC) {
+		ena_increase_stat(&rx_ring->rx_stats.bad_desc_num, 1, &rx_ring->syncp);
+		ena_reset_device(adapter, ENA_REGS_RESET_TOO_MANY_RX_DESCS);
+	} else if (rc == -EFAULT) {
+		ena_reset_device(adapter, ENA_REGS_RESET_RX_DESCRIPTOR_MALFORMED);
+	} else {
+		ena_increase_stat(&rx_ring->rx_stats.bad_req_id, 1,
+				  &rx_ring->syncp);
+		ena_reset_device(adapter, ENA_REGS_RESET_INV_RX_REQ_ID);
+	}
+	return 0;
+}
 
-	/* Too many desc from the device. Trigger reset */
-	set_bit(ENA_FLAG_TRIGGER_RESET, &adapter->flags);
+static void ena_dim_work(struct work_struct *w)
+{
+	struct dim *dim = container_of(w, struct dim, work);
+	struct dim_cq_moder cur_moder =
+		net_dim_get_rx_moderation(dim->mode, dim->profile_ix);
+	struct ena_napi *ena_napi = container_of(dim, struct ena_napi, dim);
 
-	return 0;
+	ena_napi->rx_ring->smoothed_interval = cur_moder.usec;
+	dim->state = DIM_START_MEASURE;
 }
 
-inline void ena_adjust_intr_moderation(struct ena_ring *rx_ring,
-				       struct ena_ring *tx_ring)
+static void ena_adjust_adaptive_rx_intr_moderation(struct ena_napi *ena_napi)
 {
-	/* We apply adaptive moderation on Rx path only.
-	 * Tx uses static interrupt moderation.
-	 */
-	ena_com_calculate_interrupt_delay(rx_ring->ena_dev,
-					  rx_ring->per_napi_packets,
-					  rx_ring->per_napi_bytes,
-					  &rx_ring->smoothed_interval,
-					  &rx_ring->moder_tbl_idx);
-
-	/* Reset per napi packets/bytes */
-	tx_ring->per_napi_packets = 0;
-	tx_ring->per_napi_bytes = 0;
+	struct dim_sample dim_sample;
+	struct ena_ring *rx_ring = ena_napi->rx_ring;
+
+	if (!rx_ring->per_napi_packets)
+		return;
+
+	rx_ring->non_empty_napi_events++;
+
+	dim_update_sample(rx_ring->non_empty_napi_events,
+			  rx_ring->rx_stats.cnt,
+			  rx_ring->rx_stats.bytes,
+			  &dim_sample);
+
+	net_dim(&ena_napi->dim, &dim_sample);
+
 	rx_ring->per_napi_packets = 0;
-	rx_ring->per_napi_bytes = 0;
 }
 
-static inline void ena_update_ring_numa_node(struct ena_ring *tx_ring,
-					     struct ena_ring *rx_ring)
+void ena_unmask_interrupt(struct ena_ring *tx_ring,
+			  struct ena_ring *rx_ring)
+{
+	u32 rx_interval = tx_ring->smoothed_interval;
+	struct ena_eth_io_intr_reg intr_reg;
+
+	/* Rx ring can be NULL when for XDP tx queues which don't have an
+	 * accompanying rx_ring pair.
+	 */
+	if (rx_ring)
+		rx_interval = ena_com_get_adaptive_moderation_enabled(rx_ring->ena_dev) ?
+			rx_ring->smoothed_interval :
+			ena_com_get_nonadaptive_moderation_interval_rx(rx_ring->ena_dev);
+
+	/* Update intr register: rx intr delay,
+	 * tx intr delay and interrupt unmask
+	 */
+	ena_com_update_intr_reg(&intr_reg,
+				rx_interval,
+				tx_ring->smoothed_interval,
+				true);
+
+	ena_increase_stat(&tx_ring->tx_stats.unmask_interrupt, 1,
+			  &tx_ring->syncp);
+
+	/* It is a shared MSI-X.
+	 * Tx and Rx CQ have pointer to it.
+	 * So we use one of them to reach the intr reg
+	 * The Tx ring is used because the rx_ring is NULL for XDP queues
+	 */
+	ena_com_unmask_intr(tx_ring->ena_com_io_cq, &intr_reg);
+}
+
+void ena_update_ring_numa_node(struct ena_ring *tx_ring,
+			       struct ena_ring *rx_ring)
 {
 	int cpu = get_cpu();
 	int numa_node;
@@ -1088,17 +1408,27 @@ static inline void ena_update_ring_numa_node(struct ena_ring *tx_ring,
 	if (likely(tx_ring->cpu == cpu))
 		goto out;
 
+	tx_ring->cpu = cpu;
+	if (rx_ring)
+		rx_ring->cpu = cpu;
+
 	numa_node = cpu_to_node(cpu);
+
+	if (likely(tx_ring->numa_node == numa_node))
+		goto out;
+
 	put_cpu();
 
 	if (numa_node != NUMA_NO_NODE) {
 		ena_com_update_numa_node(tx_ring->ena_com_io_cq, numa_node);
-		ena_com_update_numa_node(rx_ring->ena_com_io_cq, numa_node);
+		tx_ring->numa_node = numa_node;
+		if (rx_ring) {
+			rx_ring->numa_node = numa_node;
+			ena_com_update_numa_node(rx_ring->ena_com_io_cq,
+						 numa_node);
+		}
 	}
 
-	tx_ring->cpu = cpu;
-	rx_ring->cpu = cpu;
-
 	return;
 out:
 	put_cpu();
@@ -1108,10 +1438,8 @@ static int ena_io_poll(struct napi_struct *napi, int budget)
 {
 	struct ena_napi *ena_napi = container_of(napi, struct ena_napi, napi);
 	struct ena_ring *tx_ring, *rx_ring;
-	struct ena_eth_io_intr_reg intr_reg;
-
-	u32 tx_work_done;
-	u32 rx_work_done;
+	int tx_work_done;
+	int rx_work_done = 0;
 	int tx_budget;
 	int napi_comp_call = 0;
 	int ret;
@@ -1128,7 +1456,11 @@ static int ena_io_poll(struct napi_struct *napi, int budget)
 	}
 
 	tx_work_done = ena_clean_tx_irq(tx_ring, tx_budget);
-	rx_work_done = ena_clean_rx_irq(rx_ring, napi, budget);
+	/* On netpoll the budget is zero and the handler should only clean the
+	 * tx completions.
+	 */
+	if (likely(budget))
+		rx_work_done = ena_clean_rx_irq(rx_ring, napi, budget);
 
 	/* If the device is about to reset or down, avoid unmask
 	 * the interrupt and return 0 so NAPI won't reschedule
@@ -1144,29 +1476,20 @@ static int ena_io_poll(struct napi_struct *napi, int budget)
 		/* Update numa and unmask the interrupt only when schedule
 		 * from the interrupt context (vs from sk_busy_loop)
 		 */
-		if (napi_complete_done(napi, rx_work_done)) {
-			/* Tx and Rx share the same interrupt vector */
+		if (napi_complete_done(napi, rx_work_done) &&
+		    READ_ONCE(ena_napi->interrupts_masked)) {
+			smp_rmb(); /* make sure interrupts_masked is read */
+			WRITE_ONCE(ena_napi->interrupts_masked, false);
+			/* We apply adaptive moderation on Rx path only.
+			 * Tx uses static interrupt moderation.
+			 */
 			if (ena_com_get_adaptive_moderation_enabled(rx_ring->ena_dev))
-				ena_adjust_intr_moderation(rx_ring, tx_ring);
+				ena_adjust_adaptive_rx_intr_moderation(ena_napi);
 
-			/* Update intr register: rx intr delay,
-			 * tx intr delay and interrupt unmask
-			 */
-			ena_com_update_intr_reg(&intr_reg,
-						rx_ring->smoothed_interval,
-						tx_ring->smoothed_interval,
-						true);
-
-			/* It is a shared MSI-X.
-			 * Tx and Rx CQ have pointer to it.
-			 * So we use one of them to reach the intr reg
-			 */
-			ena_com_unmask_intr(rx_ring->ena_com_io_cq, &intr_reg);
+			ena_update_ring_numa_node(tx_ring, rx_ring);
+			ena_unmask_interrupt(tx_ring, rx_ring);
 		}
 
-
-		ena_update_ring_numa_node(tx_ring, rx_ring);
-
 		ret = rx_work_done;
 	} else {
 		ret = budget;
@@ -1177,6 +1500,8 @@ static int ena_io_poll(struct napi_struct *napi, int budget)
 	tx_ring->tx_stats.tx_poll++;
 	u64_stats_update_end(&tx_ring->syncp);
 
+	tx_ring->tx_stats.last_napi_jiffies = jiffies;
+
 	return ret;
 }
 
@@ -1201,40 +1526,59 @@ static irqreturn_t ena_intr_msix_io(int irq, void *data)
 {
 	struct ena_napi *ena_napi = data;
 
-	napi_schedule(&ena_napi->napi);
+	/* Used to check HW health */
+	WRITE_ONCE(ena_napi->first_interrupt, true);
+
+	WRITE_ONCE(ena_napi->interrupts_masked, true);
+	smp_wmb(); /* write interrupts_masked before calling napi */
+
+	napi_schedule_irqoff(&ena_napi->napi);
 
 	return IRQ_HANDLED;
 }
 
-static int ena_enable_msix(struct ena_adapter *adapter, int num_queues)
+/* Reserve a single MSI-X vector for management (admin + aenq).
+ * plus reserve one vector for each potential io queue.
+ * the number of potential io queues is the minimum of what the device
+ * supports and the number of vCPUs.
+ */
+static int ena_enable_msix(struct ena_adapter *adapter)
 {
-	int msix_vecs, rc;
+	int msix_vecs, irq_cnt;
 
-	/* Reserved the max msix vectors we might need */
-	msix_vecs = ENA_MAX_MSIX_VEC(num_queues);
+	if (test_bit(ENA_FLAG_MSIX_ENABLED, &adapter->flags)) {
+		netif_err(adapter, probe, adapter->netdev,
+			  "Error, MSI-X is already enabled\n");
+		return -EPERM;
+	}
 
+	/* Reserved the max msix vectors we might need */
+	msix_vecs = ENA_MAX_MSIX_VEC(adapter->max_num_io_queues);
 	netif_dbg(adapter, probe, adapter->netdev,
-		  "trying to enable MSI-X, vectors %d\n", msix_vecs);
+		  "Trying to enable MSI-X, vectors %d\n", msix_vecs);
+
+	irq_cnt = pci_alloc_irq_vectors(adapter->pdev, ENA_MIN_MSIX_VEC,
+					msix_vecs, PCI_IRQ_MSIX);
 
-	rc = pci_alloc_irq_vectors(adapter->pdev, msix_vecs, msix_vecs,
-			PCI_IRQ_MSIX);
-	if (rc < 0) {
+	if (irq_cnt < 0) {
 		netif_err(adapter, probe, adapter->netdev,
-			  "Failed to enable MSI-X, vectors %d rc %d\n",
-			  msix_vecs, rc);
+			  "Failed to enable MSI-X. irq_cnt %d\n", irq_cnt);
 		return -ENOSPC;
 	}
 
-	netif_dbg(adapter, probe, adapter->netdev, "enable MSI-X, vectors %d\n",
-		  msix_vecs);
-
-	if (msix_vecs >= 1) {
-		if (ena_init_rx_cpu_rmap(adapter))
-			netif_warn(adapter, probe, adapter->netdev,
-				   "Failed to map IRQs to CPUs\n");
+	if (irq_cnt != msix_vecs) {
+		netif_notice(adapter, probe, adapter->netdev,
+			     "Enable only %d MSI-X (out of %d), reduce the number of queues\n",
+			     irq_cnt, msix_vecs);
+		adapter->num_io_queues = irq_cnt - ENA_ADMIN_MSIX_VEC;
 	}
 
-	adapter->msix_vecs = msix_vecs;
+	if (netif_enable_cpu_rmap(adapter->netdev, adapter->num_io_queues))
+		netif_warn(adapter, probe, adapter->netdev,
+			   "Failed to map IRQs to CPUs\n");
+
+	adapter->msix_vecs = irq_cnt;
+	set_bit(ENA_FLAG_MSIX_ENABLED, &adapter->flags);
 
 	return 0;
 }
@@ -1261,10 +1605,12 @@ static void ena_setup_io_intr(struct ena_adapter *adapter)
 {
 	struct net_device *netdev;
 	int irq_idx, i, cpu;
+	int io_queue_count;
 
 	netdev = adapter->netdev;
+	io_queue_count = adapter->num_io_queues + adapter->xdp_num_queues;
 
-	for (i = 0; i < adapter->num_queues; i++) {
+	for (i = 0; i < io_queue_count; i++) {
 		irq_idx = ENA_IO_IRQ_IDX(i);
 		cpu = i % num_online_cpus();
 
@@ -1292,12 +1638,12 @@ static int ena_request_mgmnt_irq(struct ena_adapter *adapter)
 			 irq->data);
 	if (rc) {
 		netif_err(adapter, probe, adapter->netdev,
-			  "failed to request admin irq\n");
+			  "Failed to request admin irq\n");
 		return rc;
 	}
 
 	netif_dbg(adapter, probe, adapter->netdev,
-		  "set affinity hint of mgmnt irq.to 0x%lx (irq vector: %d)\n",
+		  "Set affinity hint of mgmnt irq.to 0x%lx (irq vector: %d)\n",
 		  irq->affinity_hint_mask.bits[0], irq->vector);
 
 	irq_set_affinity_hint(irq->vector, &irq->affinity_hint_mask);
@@ -1307,11 +1653,18 @@ static int ena_request_mgmnt_irq(struct ena_adapter *adapter)
 
 static int ena_request_io_irq(struct ena_adapter *adapter)
 {
+	u32 io_queue_count = adapter->num_io_queues + adapter->xdp_num_queues;
+	int rc = 0, i, k, irq_idx;
 	unsigned long flags = 0;
 	struct ena_irq *irq;
-	int rc = 0, i, k;
 
-	for (i = ENA_IO_IRQ_FIRST_IDX; i < adapter->msix_vecs; i++) {
+	if (!test_bit(ENA_FLAG_MSIX_ENABLED, &adapter->flags)) {
+		netif_err(adapter, ifup, adapter->netdev,
+			  "Failed to request I/O IRQ: MSI-X is not enabled\n");
+		return -EINVAL;
+	}
+
+	for (i = ENA_IO_IRQ_FIRST_IDX; i < ENA_MAX_MSIX_VEC(io_queue_count); i++) {
 		irq = &adapter->irq_tbl[i];
 		rc = request_irq(irq->vector, irq->handler, flags, irq->name,
 				 irq->data);
@@ -1323,12 +1676,22 @@ static int ena_request_io_irq(struct ena_adapter *adapter)
 		}
 
 		netif_dbg(adapter, ifup, adapter->netdev,
-			  "set affinity hint of irq. index %d to 0x%lx (irq vector: %d)\n",
+			  "Set affinity hint of irq. index %d to 0x%lx (irq vector: %d)\n",
 			  i, irq->affinity_hint_mask.bits[0], irq->vector);
 
 		irq_set_affinity_hint(irq->vector, &irq->affinity_hint_mask);
 	}
 
+	/* Now that IO IRQs have been successfully allocated map them to the
+	 * corresponding IO NAPI instance. Note that the mgmnt IRQ does not
+	 * have a NAPI, so care must be taken to correctly map IRQs to NAPIs.
+	 */
+	for (i = 0; i < io_queue_count; i++) {
+		irq_idx = ENA_IO_IRQ_IDX(i);
+		irq = &adapter->irq_tbl[irq_idx];
+		netif_napi_set_irq(&adapter->ena_napi[i].napi, irq->vector);
+	}
+
 	return rc;
 
 err:
@@ -1352,82 +1715,120 @@ static void ena_free_mgmnt_irq(struct ena_adapter *adapter)
 
 static void ena_free_io_irq(struct ena_adapter *adapter)
 {
+	u32 io_queue_count = adapter->num_io_queues + adapter->xdp_num_queues;
 	struct ena_irq *irq;
 	int i;
 
-#ifdef CONFIG_RFS_ACCEL
-	if (adapter->msix_vecs >= 1) {
-		free_irq_cpu_rmap(adapter->netdev->rx_cpu_rmap);
-		adapter->netdev->rx_cpu_rmap = NULL;
-	}
-#endif /* CONFIG_RFS_ACCEL */
+	for (i = ENA_IO_IRQ_FIRST_IDX; i < ENA_MAX_MSIX_VEC(io_queue_count); i++) {
+		struct ena_napi *ena_napi;
 
-	for (i = ENA_IO_IRQ_FIRST_IDX; i < adapter->msix_vecs; i++) {
 		irq = &adapter->irq_tbl[i];
 		irq_set_affinity_hint(irq->vector, NULL);
+		ena_napi = irq->data;
+		netif_napi_set_irq(&ena_napi->napi, -1);
 		free_irq(irq->vector, irq->data);
 	}
 }
 
+static void ena_disable_msix(struct ena_adapter *adapter)
+{
+	if (test_and_clear_bit(ENA_FLAG_MSIX_ENABLED, &adapter->flags))
+		pci_free_irq_vectors(adapter->pdev);
+}
+
 static void ena_disable_io_intr_sync(struct ena_adapter *adapter)
 {
+	u32 io_queue_count = adapter->num_io_queues + adapter->xdp_num_queues;
 	int i;
 
 	if (!netif_running(adapter->netdev))
 		return;
 
-	for (i = ENA_IO_IRQ_FIRST_IDX; i < adapter->msix_vecs; i++)
+	for (i = ENA_IO_IRQ_FIRST_IDX; i < ENA_MAX_MSIX_VEC(io_queue_count); i++)
 		synchronize_irq(adapter->irq_tbl[i].vector);
 }
 
-static void ena_del_napi(struct ena_adapter *adapter)
+static void ena_del_napi_in_range(struct ena_adapter *adapter,
+				  int first_index,
+				  int count)
 {
 	int i;
 
-	for (i = 0; i < adapter->num_queues; i++)
+	for (i = first_index; i < first_index + count; i++) {
 		netif_napi_del(&adapter->ena_napi[i].napi);
+
+		WARN_ON(ENA_IS_XDP_INDEX(adapter, i) &&
+			adapter->ena_napi[i].rx_ring);
+	}
 }
 
-static void ena_init_napi(struct ena_adapter *adapter)
+static void ena_init_napi_in_range(struct ena_adapter *adapter,
+				   int first_index, int count)
 {
-	struct ena_napi *napi;
+	int (*napi_handler)(struct napi_struct *napi, int budget);
 	int i;
 
-	for (i = 0; i < adapter->num_queues; i++) {
-		napi = &adapter->ena_napi[i];
+	for (i = first_index; i < first_index + count; i++) {
+		struct ena_napi *napi = &adapter->ena_napi[i];
+		struct ena_ring *rx_ring, *tx_ring;
 
-		netif_napi_add(adapter->netdev,
-			       &adapter->ena_napi[i].napi,
-			       ena_io_poll,
-			       ENA_NAPI_BUDGET);
-		napi->rx_ring = &adapter->rx_ring[i];
-		napi->tx_ring = &adapter->tx_ring[i];
+		memset(napi, 0, sizeof(*napi));
+
+		rx_ring = &adapter->rx_ring[i];
+		tx_ring = &adapter->tx_ring[i];
+
+		napi_handler = ena_io_poll;
+		if (ENA_IS_XDP_INDEX(adapter, i))
+			napi_handler = ena_xdp_io_poll;
+
+		netif_napi_add_config(adapter->netdev, &napi->napi, napi_handler, i);
+
+		if (!ENA_IS_XDP_INDEX(adapter, i))
+			napi->rx_ring = rx_ring;
+
+		napi->tx_ring = tx_ring;
 		napi->qid = i;
 	}
 }
 
-static void ena_napi_disable_all(struct ena_adapter *adapter)
+static void ena_napi_disable_in_range(struct ena_adapter *adapter,
+				      int first_index,
+				      int count)
 {
+	struct napi_struct *napi;
 	int i;
 
-	for (i = 0; i < adapter->num_queues; i++)
-		napi_disable(&adapter->ena_napi[i].napi);
+	for (i = first_index; i < first_index + count; i++) {
+		napi = &adapter->ena_napi[i].napi;
+		if (!ENA_IS_XDP_INDEX(adapter, i)) {
+			/* This API is supported for non-XDP queues only */
+			netif_queue_set_napi(adapter->netdev, i,
+					     NETDEV_QUEUE_TYPE_TX, NULL);
+			netif_queue_set_napi(adapter->netdev, i,
+					     NETDEV_QUEUE_TYPE_RX, NULL);
+		}
+		napi_disable(napi);
+	}
 }
 
-static void ena_napi_enable_all(struct ena_adapter *adapter)
+static void ena_napi_enable_in_range(struct ena_adapter *adapter,
+				     int first_index,
+				     int count)
 {
+	struct napi_struct *napi;
 	int i;
 
-	for (i = 0; i < adapter->num_queues; i++)
-		napi_enable(&adapter->ena_napi[i].napi);
-}
-
-static void ena_restore_ethtool_params(struct ena_adapter *adapter)
-{
-	adapter->tx_usecs = 0;
-	adapter->rx_usecs = 0;
-	adapter->tx_frames = 1;
-	adapter->rx_frames = 1;
+	for (i = first_index; i < first_index + count; i++) {
+		napi = &adapter->ena_napi[i].napi;
+		napi_enable(napi);
+		if (!ENA_IS_XDP_INDEX(adapter, i)) {
+			/* This API is supported for non-XDP queues only */
+			netif_queue_set_napi(adapter->netdev, i,
+					     NETDEV_QUEUE_TYPE_RX, napi);
+			netif_queue_set_napi(adapter->netdev, i,
+					     NETDEV_QUEUE_TYPE_TX, napi);
+		}
+	}
 }
 
 /* Configure the Rx forwarding */
@@ -1439,26 +1840,25 @@ static int ena_rss_configure(struct ena_adapter *adapter)
 	/* In case the RSS table wasn't initialized by probe */
 	if (!ena_dev->rss.tbl_log_size) {
 		rc = ena_rss_init_default(adapter);
-		if (rc && (rc != -EPERM)) {
-			netif_err(adapter, ifup, adapter->netdev,
-				  "Failed to init RSS rc: %d\n", rc);
+		if (rc && (rc != -EOPNOTSUPP)) {
+			netif_err(adapter, ifup, adapter->netdev, "Failed to init RSS rc: %d\n", rc);
 			return rc;
 		}
 	}
 
 	/* Set indirect table */
 	rc = ena_com_indirect_table_set(ena_dev);
-	if (unlikely(rc && rc != -EPERM))
+	if (unlikely(rc && rc != -EOPNOTSUPP))
 		return rc;
 
 	/* Configure hash function (if supported) */
 	rc = ena_com_set_hash_function(ena_dev);
-	if (unlikely(rc && (rc != -EPERM)))
+	if (unlikely(rc && (rc != -EOPNOTSUPP)))
 		return rc;
 
 	/* Configure hash inputs (if supported) */
 	rc = ena_com_set_hash_ctrl(ena_dev);
-	if (unlikely(rc && (rc != -EPERM)))
+	if (unlikely(rc && (rc != -EOPNOTSUPP)))
 		return rc;
 
 	return 0;
@@ -1466,14 +1866,12 @@ static int ena_rss_configure(struct ena_adapter *adapter)
 
 static int ena_up_complete(struct ena_adapter *adapter)
 {
-	int rc, i;
+	int rc;
 
 	rc = ena_rss_configure(adapter);
 	if (rc)
 		return rc;
 
-	ena_init_napi(adapter);
-
 	ena_change_mtu(adapter->netdev, adapter->netdev->mtu);
 
 	ena_refill_all_rx_bufs(adapter);
@@ -1481,22 +1879,16 @@ static int ena_up_complete(struct ena_adapter *adapter)
 	/* enable transmits */
 	netif_tx_start_all_queues(adapter->netdev);
 
-	ena_restore_ethtool_params(adapter);
-
-	ena_napi_enable_all(adapter);
-
-	/* schedule napi in case we had pending packets
-	 * from the last time we disable napi
-	 */
-	for (i = 0; i < adapter->num_queues; i++)
-		napi_schedule(&adapter->ena_napi[i].napi);
+	ena_napi_enable_in_range(adapter,
+				 0,
+				 adapter->xdp_num_queues + adapter->num_io_queues);
 
 	return 0;
 }
 
 static int ena_create_io_tx_queue(struct ena_adapter *adapter, int qid)
 {
-	struct ena_com_create_io_ctx ctx = { 0 };
+	struct ena_com_create_io_ctx ctx;
 	struct ena_com_dev *ena_dev;
 	struct ena_ring *tx_ring;
 	u32 msix_vector;
@@ -1509,12 +1901,14 @@ static int ena_create_io_tx_queue(struct ena_adapter *adapter, int qid)
 	msix_vector = ENA_IO_IRQ_IDX(qid);
 	ena_qid = ENA_IO_TXQ_IDX(qid);
 
+	memset(&ctx, 0x0, sizeof(ctx));
+
 	ctx.direction = ENA_COM_IO_QUEUE_DIRECTION_TX;
 	ctx.qid = ena_qid;
 	ctx.mem_queue_type = ena_dev->tx_mem_queue_type;
 	ctx.msix_vector = msix_vector;
-	ctx.queue_size = adapter->tx_ring_size;
-	ctx.numa_node = cpu_to_node(tx_ring->cpu);
+	ctx.queue_size = tx_ring->ring_size;
+	ctx.numa_node = tx_ring->numa_node;
 
 	rc = ena_com_create_io_queue(ena_dev, &ctx);
 	if (rc) {
@@ -1532,18 +1926,20 @@ static int ena_create_io_tx_queue(struct ena_adapter *adapter, int qid)
 			  "Failed to get TX queue handlers. TX queue num %d rc: %d\n",
 			  qid, rc);
 		ena_com_destroy_io_queue(ena_dev, ena_qid);
+		return rc;
 	}
 
 	ena_com_update_numa_node(tx_ring->ena_com_io_cq, ctx.numa_node);
 	return rc;
 }
 
-static int ena_create_all_io_tx_queues(struct ena_adapter *adapter)
+int ena_create_io_tx_queues_in_range(struct ena_adapter *adapter,
+				     int first_index, int count)
 {
 	struct ena_com_dev *ena_dev = adapter->ena_dev;
 	int rc, i;
 
-	for (i = 0; i < adapter->num_queues; i++) {
+	for (i = first_index; i < first_index + count; i++) {
 		rc = ena_create_io_tx_queue(adapter, i);
 		if (rc)
 			goto create_err;
@@ -1552,7 +1948,7 @@ static int ena_create_all_io_tx_queues(struct ena_adapter *adapter)
 	return 0;
 
 create_err:
-	while (i--)
+	while (i-- > first_index)
 		ena_com_destroy_io_queue(ena_dev, ENA_IO_TXQ_IDX(i));
 
 	return rc;
@@ -1561,7 +1957,7 @@ create_err:
 static int ena_create_io_rx_queue(struct ena_adapter *adapter, int qid)
 {
 	struct ena_com_dev *ena_dev;
-	struct ena_com_create_io_ctx ctx = { 0 };
+	struct ena_com_create_io_ctx ctx;
 	struct ena_ring *rx_ring;
 	u32 msix_vector;
 	u16 ena_qid;
@@ -1573,12 +1969,14 @@ static int ena_create_io_rx_queue(struct ena_adapter *adapter, int qid)
 	msix_vector = ENA_IO_IRQ_IDX(qid);
 	ena_qid = ENA_IO_RXQ_IDX(qid);
 
+	memset(&ctx, 0x0, sizeof(ctx));
+
 	ctx.qid = ena_qid;
 	ctx.direction = ENA_COM_IO_QUEUE_DIRECTION_RX;
 	ctx.mem_queue_type = ENA_ADMIN_PLACEMENT_POLICY_HOST;
 	ctx.msix_vector = msix_vector;
-	ctx.queue_size = adapter->rx_ring_size;
-	ctx.numa_node = cpu_to_node(rx_ring->cpu);
+	ctx.queue_size = rx_ring->ring_size;
+	ctx.numa_node = rx_ring->numa_node;
 
 	rc = ena_com_create_io_queue(ena_dev, &ctx);
 	if (rc) {
@@ -1595,12 +1993,15 @@ static int ena_create_io_rx_queue(struct ena_adapter *adapter, int qid)
 		netif_err(adapter, ifup, adapter->netdev,
 			  "Failed to get RX queue handlers. RX queue num %d rc: %d\n",
 			  qid, rc);
-		ena_com_destroy_io_queue(ena_dev, ena_qid);
+		goto err;
 	}
 
 	ena_com_update_numa_node(rx_ring->ena_com_io_cq, ctx.numa_node);
 
 	return rc;
+err:
+	ena_com_destroy_io_queue(ena_dev, ena_qid);
+	return rc;
 }
 
 static int ena_create_all_io_rx_queues(struct ena_adapter *adapter)
@@ -1608,52 +2009,173 @@ static int ena_create_all_io_rx_queues(struct ena_adapter *adapter)
 	struct ena_com_dev *ena_dev = adapter->ena_dev;
 	int rc, i;
 
-	for (i = 0; i < adapter->num_queues; i++) {
+	for (i = 0; i < adapter->num_io_queues; i++) {
 		rc = ena_create_io_rx_queue(adapter, i);
 		if (rc)
 			goto create_err;
+		INIT_WORK(&adapter->ena_napi[i].dim.work, ena_dim_work);
+
+		ena_xdp_register_rxq_info(&adapter->rx_ring[i]);
 	}
 
 	return 0;
 
 create_err:
-	while (i--)
+	while (i--) {
+		ena_xdp_unregister_rxq_info(&adapter->rx_ring[i]);
+		cancel_work_sync(&adapter->ena_napi[i].dim.work);
 		ena_com_destroy_io_queue(ena_dev, ENA_IO_RXQ_IDX(i));
+	}
 
 	return rc;
 }
 
-static int ena_up(struct ena_adapter *adapter)
+static void set_io_rings_size(struct ena_adapter *adapter,
+			      int new_tx_size,
+			      int new_rx_size)
 {
-	int rc;
+	int i;
 
-	netdev_dbg(adapter->netdev, "%s\n", __func__);
+	for (i = 0; i < adapter->num_io_queues; i++) {
+		adapter->tx_ring[i].ring_size = new_tx_size;
+		adapter->rx_ring[i].ring_size = new_rx_size;
+	}
+}
 
-	ena_setup_io_intr(adapter);
+/* This function allows queue allocation to backoff when the system is
+ * low on memory. If there is not enough memory to allocate io queues
+ * the driver will try to allocate smaller queues.
+ *
+ * The backoff algorithm is as follows:
+ *  1. Try to allocate TX and RX and if successful.
+ *  1.1. return success
+ *
+ *  2. Divide by 2 the size of the larger of RX and TX queues (or both if their size is the same).
+ *
+ *  3. If TX or RX is smaller than 256
+ *  3.1. return failure.
+ *  4. else
+ *  4.1. go back to 1.
+ */
+static int create_queues_with_size_backoff(struct ena_adapter *adapter)
+{
+	int rc, cur_rx_ring_size, cur_tx_ring_size;
+	int new_rx_ring_size, new_tx_ring_size;
 
-	rc = ena_request_io_irq(adapter);
-	if (rc)
-		goto err_req_irq;
+	/* current queue sizes might be set to smaller than the requested
+	 * ones due to past queue allocation failures.
+	 */
+	set_io_rings_size(adapter, adapter->requested_tx_ring_size,
+			  adapter->requested_rx_ring_size);
 
-	/* allocate transmit descriptors */
-	rc = ena_setup_all_tx_resources(adapter);
-	if (rc)
-		goto err_setup_tx;
+	while (1) {
+		if (ena_xdp_present(adapter)) {
+			rc = ena_setup_and_create_all_xdp_queues(adapter);
 
-	/* allocate receive descriptors */
-	rc = ena_setup_all_rx_resources(adapter);
-	if (rc)
-		goto err_setup_rx;
+			if (rc)
+				goto err_setup_tx;
+		}
+		rc = ena_setup_tx_resources_in_range(adapter,
+						     0,
+						     adapter->num_io_queues);
+		if (rc)
+			goto err_setup_tx;
+
+		rc = ena_create_io_tx_queues_in_range(adapter,
+						      0,
+						      adapter->num_io_queues);
+		if (rc)
+			goto err_create_tx_queues;
+
+		rc = ena_setup_all_rx_resources(adapter);
+		if (rc)
+			goto err_setup_rx;
+
+		rc = ena_create_all_io_rx_queues(adapter);
+		if (rc)
+			goto err_create_rx_queues;
+
+		return 0;
+
+err_create_rx_queues:
+		ena_free_all_io_rx_resources(adapter);
+err_setup_rx:
+		ena_destroy_all_tx_queues(adapter);
+err_create_tx_queues:
+		ena_free_all_io_tx_resources(adapter);
+err_setup_tx:
+		if (rc != -ENOMEM) {
+			netif_err(adapter, ifup, adapter->netdev,
+				  "Queue creation failed with error code %d\n",
+				  rc);
+			return rc;
+		}
+
+		cur_tx_ring_size = adapter->tx_ring[0].ring_size;
+		cur_rx_ring_size = adapter->rx_ring[0].ring_size;
+
+		netif_err(adapter, ifup, adapter->netdev,
+			  "Not enough memory to create queues with sizes TX=%d, RX=%d\n",
+			  cur_tx_ring_size, cur_rx_ring_size);
+
+		new_tx_ring_size = cur_tx_ring_size;
+		new_rx_ring_size = cur_rx_ring_size;
+
+		/* Decrease the size of the larger queue, or
+		 * decrease both if they are the same size.
+		 */
+		if (cur_rx_ring_size <= cur_tx_ring_size)
+			new_tx_ring_size = cur_tx_ring_size / 2;
+		if (cur_rx_ring_size >= cur_tx_ring_size)
+			new_rx_ring_size = cur_rx_ring_size / 2;
+
+		if (new_tx_ring_size < ENA_MIN_RING_SIZE ||
+		    new_rx_ring_size < ENA_MIN_RING_SIZE) {
+			netif_err(adapter, ifup, adapter->netdev,
+				  "Queue creation failed with the smallest possible queue size of %d for both queues. Not retrying with smaller queues\n",
+				  ENA_MIN_RING_SIZE);
+			return rc;
+		}
 
-	/* Create TX queues */
-	rc = ena_create_all_io_tx_queues(adapter);
+		netif_err(adapter, ifup, adapter->netdev,
+			  "Retrying queue creation with sizes TX=%d, RX=%d\n",
+			  new_tx_ring_size,
+			  new_rx_ring_size);
+
+		set_io_rings_size(adapter, new_tx_ring_size,
+				  new_rx_ring_size);
+	}
+}
+
+int ena_up(struct ena_adapter *adapter)
+{
+	int io_queue_count, rc, i;
+
+	netif_dbg(adapter, ifup, adapter->netdev, "%s\n", __func__);
+
+	io_queue_count = adapter->num_io_queues + adapter->xdp_num_queues;
+	ena_setup_io_intr(adapter);
+
+	/* napi poll functions should be initialized before running
+	 * request_irq(), to handle a rare condition where there is a pending
+	 * interrupt, causing the ISR to fire immediately while the poll
+	 * function wasn't set yet, causing a null dereference
+	 */
+	ena_init_napi_in_range(adapter, 0, io_queue_count);
+
+	/* Enabling DIM needs to happen before enabling IRQs since DIM
+	 * is run from napi routine
+	 */
+	if (ena_com_interrupt_moderation_supported(adapter->ena_dev))
+		ena_com_enable_adaptive_moderation(adapter->ena_dev);
+
+	rc = ena_request_io_irq(adapter);
 	if (rc)
-		goto err_create_tx_queues;
+		goto err_req_irq;
 
-	/* Create RX queues */
-	rc = ena_create_all_io_rx_queues(adapter);
+	rc = create_queues_with_size_backoff(adapter);
 	if (rc)
-		goto err_create_rx_queues;
+		goto err_create_queues_with_backoff;
 
 	rc = ena_up_complete(adapter);
 	if (rc)
@@ -1662,60 +2184,70 @@ static int ena_up(struct ena_adapter *adapter)
 	if (test_bit(ENA_FLAG_LINK_UP, &adapter->flags))
 		netif_carrier_on(adapter->netdev);
 
-	u64_stats_update_begin(&adapter->syncp);
-	adapter->dev_stats.interface_up++;
-	u64_stats_update_end(&adapter->syncp);
+	ena_increase_stat(&adapter->dev_stats.interface_up, 1,
+			  &adapter->syncp);
 
 	set_bit(ENA_FLAG_DEV_UP, &adapter->flags);
 
+	/* Enable completion queues interrupt */
+	for (i = 0; i < adapter->num_io_queues; i++)
+		ena_unmask_interrupt(&adapter->tx_ring[i],
+				     &adapter->rx_ring[i]);
+
+	/* schedule napi in case we had pending packets
+	 * from the last time we disable napi
+	 */
+	for (i = 0; i < io_queue_count; i++)
+		napi_schedule(&adapter->ena_napi[i].napi);
+
 	return rc;
 
 err_up:
-	ena_destroy_all_rx_queues(adapter);
-err_create_rx_queues:
 	ena_destroy_all_tx_queues(adapter);
-err_create_tx_queues:
-	ena_free_all_io_rx_resources(adapter);
-err_setup_rx:
 	ena_free_all_io_tx_resources(adapter);
-err_setup_tx:
+	ena_destroy_all_rx_queues(adapter);
+	ena_free_all_io_rx_resources(adapter);
+err_create_queues_with_backoff:
 	ena_free_io_irq(adapter);
 err_req_irq:
+	ena_del_napi_in_range(adapter, 0, io_queue_count);
 
 	return rc;
 }
 
-static void ena_down(struct ena_adapter *adapter)
+void ena_down(struct ena_adapter *adapter)
 {
-	netif_info(adapter, ifdown, adapter->netdev, "%s\n", __func__);
+	int io_queue_count = adapter->num_io_queues + adapter->xdp_num_queues;
+
+	netif_dbg(adapter, ifdown, adapter->netdev, "%s\n", __func__);
 
 	clear_bit(ENA_FLAG_DEV_UP, &adapter->flags);
 
-	u64_stats_update_begin(&adapter->syncp);
-	adapter->dev_stats.interface_down++;
-	u64_stats_update_end(&adapter->syncp);
+	ena_increase_stat(&adapter->dev_stats.interface_down, 1,
+			  &adapter->syncp);
 
 	netif_carrier_off(adapter->netdev);
 	netif_tx_disable(adapter->netdev);
 
 	/* After this point the napi handler won't enable the tx queue */
-	ena_napi_disable_all(adapter);
-
-	/* After destroy the queue there won't be any new interrupts */
+	ena_napi_disable_in_range(adapter, 0, io_queue_count);
 
 	if (test_bit(ENA_FLAG_TRIGGER_RESET, &adapter->flags)) {
 		int rc;
 
-		rc = ena_com_dev_reset(adapter->ena_dev);
+		rc = ena_com_dev_reset(adapter->ena_dev, adapter->reset_reason);
 		if (rc)
-			dev_err(&adapter->pdev->dev, "Device reset failed\n");
+			netif_err(adapter, ifdown, adapter->netdev,
+				  "Device reset failed\n");
+		/* stop submitting admin commands on a device that was reset */
+		ena_com_set_admin_running_state(adapter->ena_dev, false);
 	}
 
 	ena_destroy_all_io_queues(adapter);
 
 	ena_disable_io_intr_sync(adapter);
 	ena_free_io_irq(adapter);
-	ena_del_napi(adapter);
+	ena_del_napi_in_range(adapter, 0, io_queue_count);
 
 	ena_free_all_tx_bufs(adapter);
 	ena_free_all_rx_bufs(adapter);
@@ -1740,13 +2272,13 @@ static int ena_open(struct net_device *netdev)
 	int rc;
 
 	/* Notify the stack of the actual queue counts. */
-	rc = netif_set_real_num_tx_queues(netdev, adapter->num_queues);
+	rc = netif_set_real_num_tx_queues(netdev, adapter->num_io_queues);
 	if (rc) {
 		netif_err(adapter, ifup, netdev, "Can't set num tx queues\n");
 		return rc;
 	}
 
-	rc = netif_set_real_num_rx_queues(netdev, adapter->num_queues);
+	rc = netif_set_real_num_rx_queues(netdev, adapter->num_io_queues);
 	if (rc) {
 		netif_err(adapter, ifup, netdev, "Can't set num rx queues\n");
 		return rc;
@@ -1775,13 +2307,117 @@ static int ena_close(struct net_device *netdev)
 
 	netif_dbg(adapter, ifdown, netdev, "%s\n", __func__);
 
+	if (!test_bit(ENA_FLAG_DEVICE_RUNNING, &adapter->flags))
+		return 0;
+
 	if (test_bit(ENA_FLAG_DEV_UP, &adapter->flags))
 		ena_down(adapter);
 
+	/* Check for device status and issue reset if needed*/
+	check_for_admin_com_state(adapter);
+	if (unlikely(test_bit(ENA_FLAG_TRIGGER_RESET, &adapter->flags))) {
+		netif_err(adapter, ifdown, adapter->netdev,
+			  "Destroy failure, restarting device\n");
+		ena_dump_stats_to_dmesg(adapter);
+		/* rtnl lock already obtained in dev_ioctl() layer */
+		ena_destroy_device(adapter, false);
+		ena_restore_device(adapter);
+	}
+
+	return 0;
+}
+
+int ena_update_queue_params(struct ena_adapter *adapter,
+			    u32 new_tx_size,
+			    u32 new_rx_size,
+			    u32 new_llq_header_len)
+{
+	bool dev_was_up, large_llq_changed = false;
+	int rc = 0;
+
+	dev_was_up = test_bit(ENA_FLAG_DEV_UP, &adapter->flags);
+	ena_close(adapter->netdev);
+	adapter->requested_tx_ring_size = new_tx_size;
+	adapter->requested_rx_ring_size = new_rx_size;
+	ena_init_io_rings(adapter,
+			  0,
+			  adapter->xdp_num_queues +
+			  adapter->num_io_queues);
+
+	large_llq_changed = adapter->ena_dev->tx_mem_queue_type ==
+			    ENA_ADMIN_PLACEMENT_POLICY_DEV;
+	large_llq_changed &=
+		new_llq_header_len != adapter->ena_dev->tx_max_header_size;
+
+	/* a check that the configuration is valid is done by caller */
+	if (large_llq_changed) {
+		adapter->large_llq_header_enabled = !adapter->large_llq_header_enabled;
+
+		ena_destroy_device(adapter, false);
+		rc = ena_restore_device(adapter);
+	}
+
+	return dev_was_up && !rc ? ena_up(adapter) : rc;
+}
+
+int ena_set_rx_copybreak(struct ena_adapter *adapter, u32 rx_copybreak)
+{
+	struct ena_ring *rx_ring;
+	int i;
+
+	if (rx_copybreak > min_t(u16, adapter->netdev->mtu, ENA_PAGE_SIZE))
+		return -EINVAL;
+
+	adapter->rx_copybreak = rx_copybreak;
+
+	for (i = 0; i < adapter->num_io_queues; i++) {
+		rx_ring = &adapter->rx_ring[i];
+		rx_ring->rx_copybreak = rx_copybreak;
+	}
+
 	return 0;
 }
 
-static void ena_tx_csum(struct ena_com_tx_ctx *ena_tx_ctx, struct sk_buff *skb)
+int ena_update_queue_count(struct ena_adapter *adapter, u32 new_channel_count)
+{
+	struct ena_com_dev *ena_dev = adapter->ena_dev;
+	int prev_channel_count;
+	bool dev_was_up;
+
+	dev_was_up = test_bit(ENA_FLAG_DEV_UP, &adapter->flags);
+	ena_close(adapter->netdev);
+	prev_channel_count = adapter->num_io_queues;
+	adapter->num_io_queues = new_channel_count;
+	if (ena_xdp_present(adapter) &&
+	    ena_xdp_allowed(adapter) == ENA_XDP_ALLOWED) {
+		adapter->xdp_first_ring = new_channel_count;
+		adapter->xdp_num_queues = new_channel_count;
+		if (prev_channel_count > new_channel_count)
+			ena_xdp_exchange_program_rx_in_range(adapter,
+							     NULL,
+							     new_channel_count,
+							     prev_channel_count);
+		else
+			ena_xdp_exchange_program_rx_in_range(adapter,
+							     adapter->xdp_bpf_prog,
+							     prev_channel_count,
+							     new_channel_count);
+	}
+
+	/* We need to destroy the rss table so that the indirection
+	 * table will be reinitialized by ena_up()
+	 */
+	ena_com_rss_destroy(ena_dev);
+	ena_init_io_rings(adapter,
+			  0,
+			  adapter->xdp_num_queues +
+			  adapter->num_io_queues);
+	return dev_was_up ? ena_open(adapter->netdev) : 0;
+}
+
+static void ena_tx_csum(struct ena_com_tx_ctx *ena_tx_ctx,
+			struct sk_buff *skb,
+			bool disable_meta_caching)
 {
 	u32 mss = skb_shinfo(skb)->gso_size;
 	struct ena_com_tx_meta *ena_meta = &ena_tx_ctx->ena_meta;
@@ -1825,7 +2461,9 @@ static void ena_tx_csum(struct ena_com_tx_ctx *ena_tx_ctx, struct sk_buff *skb)
 		ena_meta->l3_hdr_len = skb_network_header_len(skb);
 		ena_meta->l3_hdr_offset = skb_network_offset(skb);
 		ena_tx_ctx->meta_valid = 1;
-
+	} else if (disable_meta_caching) {
+		memset(ena_meta, 0, sizeof(*ena_meta));
+		ena_tx_ctx->meta_valid = 1;
 	} else {
 		ena_tx_ctx->meta_valid = 0;
 	}
@@ -1846,87 +2484,80 @@ static int ena_check_and_linearize_skb(struct ena_ring *tx_ring,
 	    (header_len < tx_ring->tx_max_header_size))
 		return 0;
 
-	u64_stats_update_begin(&tx_ring->syncp);
-	tx_ring->tx_stats.linearize++;
-	u64_stats_update_end(&tx_ring->syncp);
+	ena_increase_stat(&tx_ring->tx_stats.linearize, 1, &tx_ring->syncp);
 
 	rc = skb_linearize(skb);
 	if (unlikely(rc)) {
-		u64_stats_update_begin(&tx_ring->syncp);
-		tx_ring->tx_stats.linearize_failed++;
-		u64_stats_update_end(&tx_ring->syncp);
+		ena_increase_stat(&tx_ring->tx_stats.linearize_failed, 1,
+				  &tx_ring->syncp);
 	}
 
 	return rc;
 }
 
-/* Called with netif_tx_lock. */
-static netdev_tx_t ena_start_xmit(struct sk_buff *skb, struct net_device *dev)
+static int ena_tx_map_skb(struct ena_ring *tx_ring,
+			  struct ena_tx_buffer *tx_info,
+			  struct sk_buff *skb,
+			  void **push_hdr,
+			  u16 *header_len)
 {
-	struct ena_adapter *adapter = netdev_priv(dev);
-	struct ena_tx_buffer *tx_info;
-	struct ena_com_tx_ctx ena_tx_ctx;
-	struct ena_ring *tx_ring;
-	struct netdev_queue *txq;
+	struct ena_adapter *adapter = tx_ring->adapter;
 	struct ena_com_buf *ena_buf;
-	void *push_hdr;
-	u32 len, last_frag;
-	u16 next_to_use;
-	u16 req_id;
-	u16 push_len;
-	u16 header_len;
 	dma_addr_t dma;
-	int qid, rc, nb_hw_desc;
-	int i = -1;
-
-	netif_dbg(adapter, tx_queued, dev, "%s skb %p\n", __func__, skb);
-	/*  Determine which tx ring we will be placed on */
-	qid = skb_get_queue_mapping(skb);
-	tx_ring = &adapter->tx_ring[qid];
-	txq = netdev_get_tx_queue(dev, qid);
-
-	rc = ena_check_and_linearize_skb(tx_ring, skb);
-	if (unlikely(rc))
-		goto error_drop_packet;
-
-	skb_tx_timestamp(skb);
-	len = skb_headlen(skb);
+	u32 skb_head_len, frag_len, last_frag;
+	u16 push_len = 0;
+	u16 delta = 0;
+	int i = 0;
 
-	next_to_use = tx_ring->next_to_use;
-	req_id = tx_ring->free_tx_ids[next_to_use];
-	tx_info = &tx_ring->tx_buffer_info[req_id];
-	tx_info->num_of_bufs = 0;
-
-	WARN(tx_info->skb, "SKB isn't NULL req_id %d\n", req_id);
-	ena_buf = tx_info->bufs;
+	skb_head_len = skb_headlen(skb);
 	tx_info->skb = skb;
+	ena_buf = tx_info->bufs;
 
 	if (tx_ring->tx_mem_queue_type == ENA_ADMIN_PLACEMENT_POLICY_DEV) {
-		/* prepared the push buffer */
-		push_len = min_t(u32, len, tx_ring->tx_max_header_size);
-		header_len = push_len;
-		push_hdr = skb->data;
+		/* When the device is LLQ mode, the driver will copy
+		 * the header into the device memory space.
+		 * the ena_com layer assume the header is in a linear
+		 * memory space.
+		 * This assumption might be wrong since part of the header
+		 * can be in the fragmented buffers.
+		 * Use skb_header_pointer to make sure the header is in a
+		 * linear memory space.
+		 */
+
+		push_len = min_t(u32, skb->len, tx_ring->tx_max_header_size);
+		*push_hdr = skb_header_pointer(skb, 0, push_len,
+					       tx_ring->push_buf_intermediate_buf);
+		*header_len = push_len;
+		if (unlikely(skb->data != *push_hdr)) {
+			ena_increase_stat(&tx_ring->tx_stats.llq_buffer_copy, 1,
+					  &tx_ring->syncp);
+
+			delta = push_len - skb_head_len;
+		}
 	} else {
-		push_len = 0;
-		header_len = min_t(u32, len, tx_ring->tx_max_header_size);
-		push_hdr = NULL;
+		*push_hdr = NULL;
+		*header_len = min_t(u32, skb_head_len,
+				    tx_ring->tx_max_header_size);
 	}
 
-	netif_dbg(adapter, tx_queued, dev,
+	netif_dbg(adapter, tx_queued, adapter->netdev,
 		  "skb: %p header_buf->vaddr: %p push_len: %d\n", skb,
-		  push_hdr, push_len);
+		  *push_hdr, push_len);
 
-	if (len > push_len) {
+	if (skb_head_len > push_len) {
 		dma = dma_map_single(tx_ring->dev, skb->data + push_len,
-				     len - push_len, DMA_TO_DEVICE);
-		if (dma_mapping_error(tx_ring->dev, dma))
+				     skb_head_len - push_len, DMA_TO_DEVICE);
+		if (unlikely(dma_mapping_error(tx_ring->dev, dma)))
 			goto error_report_dma_error;
 
 		ena_buf->paddr = dma;
-		ena_buf->len = len - push_len;
+		ena_buf->len = skb_head_len - push_len;
 
 		ena_buf++;
 		tx_info->num_of_bufs++;
+		tx_info->map_linear_data = 1;
+	} else {
+		tx_info->map_linear_data = 0;
 	}
 
 	last_frag = skb_shinfo(skb)->nr_frags;
@@ -1934,18 +2565,72 @@ static netdev_tx_t ena_start_xmit(struct sk_buff *skb, struct net_device *dev)
 	for (i = 0; i < last_frag; i++) {
 		const skb_frag_t *frag = &skb_shinfo(skb)->frags[i];
 
-		len = skb_frag_size(frag);
-		dma = skb_frag_dma_map(tx_ring->dev, frag, 0, len,
-				       DMA_TO_DEVICE);
-		if (dma_mapping_error(tx_ring->dev, dma))
+		frag_len = skb_frag_size(frag);
+
+		if (unlikely(delta >= frag_len)) {
+			delta -= frag_len;
+			continue;
+		}
+
+		dma = skb_frag_dma_map(tx_ring->dev, frag, delta,
+				       frag_len - delta, DMA_TO_DEVICE);
+		if (unlikely(dma_mapping_error(tx_ring->dev, dma)))
 			goto error_report_dma_error;
 
 		ena_buf->paddr = dma;
-		ena_buf->len = len;
+		ena_buf->len = frag_len - delta;
 		ena_buf++;
+		tx_info->num_of_bufs++;
+		delta = 0;
 	}
 
-	tx_info->num_of_bufs += last_frag;
+	return 0;
+
+error_report_dma_error:
+	ena_increase_stat(&tx_ring->tx_stats.dma_mapping_err, 1,
+			  &tx_ring->syncp);
+	netif_warn(adapter, tx_queued, adapter->netdev, "Failed to map skb\n");
+
+	tx_info->skb = NULL;
+
+	tx_info->num_of_bufs += i;
+	ena_unmap_tx_buff(tx_ring, tx_info);
+
+	return -EINVAL;
+}
+
+/* Called with netif_tx_lock. */
+static netdev_tx_t ena_start_xmit(struct sk_buff *skb, struct net_device *dev)
+{
+	struct ena_adapter *adapter = netdev_priv(dev);
+	struct ena_tx_buffer *tx_info;
+	struct ena_com_tx_ctx ena_tx_ctx;
+	struct ena_ring *tx_ring;
+	struct netdev_queue *txq;
+	void *push_hdr;
+	u16 next_to_use, req_id, header_len;
+	int qid, rc;
+
+	netif_dbg(adapter, tx_queued, dev, "%s skb %p\n", __func__, skb);
+	/*  Determine which tx ring we will be placed on */
+	qid = skb_get_queue_mapping(skb);
+	tx_ring = &adapter->tx_ring[qid];
+	txq = netdev_get_tx_queue(dev, qid);
+
+	rc = ena_check_and_linearize_skb(tx_ring, skb);
+	if (unlikely(rc))
+		goto error_drop_packet;
+
+	next_to_use = tx_ring->next_to_use;
+	req_id = tx_ring->free_ids[next_to_use];
+	tx_info = &tx_ring->tx_buffer_info[req_id];
+	tx_info->num_of_bufs = 0;
+
+	WARN(tx_info->skb, "SKB isn't NULL req_id %d\n", req_id);
+
+	rc = ena_tx_map_skb(tx_ring, tx_info, skb, &push_hdr, &header_len);
+	if (unlikely(rc))
+		goto error_drop_packet;
 
 	memset(&ena_tx_ctx, 0x0, sizeof(struct ena_com_tx_ctx));
 	ena_tx_ctx.ena_bufs = tx_info->bufs;
@@ -1955,184 +2640,122 @@ static netdev_tx_t ena_start_xmit(struct sk_buff *skb, struct net_device *dev)
 	ena_tx_ctx.header_len = header_len;
 
 	/* set flags and meta data */
-	ena_tx_csum(&ena_tx_ctx, skb);
-
-	/* prepare the packet's descriptors to dma engine */
-	rc = ena_com_prepare_tx(tx_ring->ena_com_io_sq, &ena_tx_ctx,
-				&nb_hw_desc);
-
-	if (unlikely(rc)) {
-		netif_err(adapter, tx_queued, dev,
-			  "failed to prepare tx bufs\n");
-		u64_stats_update_begin(&tx_ring->syncp);
-		tx_ring->tx_stats.queue_stop++;
-		tx_ring->tx_stats.prepare_ctx_err++;
-		u64_stats_update_end(&tx_ring->syncp);
-		netif_tx_stop_queue(txq);
+	ena_tx_csum(&ena_tx_ctx, skb, tx_ring->disable_meta_caching);
+
+	rc = ena_xmit_common(adapter,
+			     tx_ring,
+			     tx_info,
+			     &ena_tx_ctx,
+			     next_to_use,
+			     skb->len);
+	if (rc)
 		goto error_unmap_dma;
-	}
 
 	netdev_tx_sent_queue(txq, skb->len);
 
-	u64_stats_update_begin(&tx_ring->syncp);
-	tx_ring->tx_stats.cnt++;
-	tx_ring->tx_stats.bytes += skb->len;
-	u64_stats_update_end(&tx_ring->syncp);
-
-	tx_info->tx_descs = nb_hw_desc;
-	tx_info->last_jiffies = jiffies;
-
-	tx_ring->next_to_use = ENA_TX_RING_IDX_NEXT(next_to_use,
-		tx_ring->ring_size);
-
-	/* This WMB is aimed to:
-	 * 1 - perform smp barrier before reading next_to_completion
-	 * 2 - make sure the desc were written before trigger DB
-	 */
-	wmb();
-
 	/* stop the queue when no more space available, the packet can have up
 	 * to sgl_size + 2. one for the meta descriptor and one for header
 	 * (if the header is larger than tx_max_header_size).
 	 */
-	if (unlikely(ena_com_sq_empty_space(tx_ring->ena_com_io_sq) <
-		     (tx_ring->sgl_size + 2))) {
+	if (unlikely(!ena_com_sq_have_enough_space(tx_ring->ena_com_io_sq,
+						   tx_ring->sgl_size + 2))) {
 		netif_dbg(adapter, tx_queued, dev, "%s stop queue %d\n",
 			  __func__, qid);
 
 		netif_tx_stop_queue(txq);
-		u64_stats_update_begin(&tx_ring->syncp);
-		tx_ring->tx_stats.queue_stop++;
-		u64_stats_update_end(&tx_ring->syncp);
+		ena_increase_stat(&tx_ring->tx_stats.queue_stop, 1,
+				  &tx_ring->syncp);
 
 		/* There is a rare condition where this function decide to
 		 * stop the queue but meanwhile clean_tx_irq updates
 		 * next_to_completion and terminates.
 		 * The queue will remain stopped forever.
-		 * To solve this issue this function perform rmb, check
-		 * the wakeup condition and wake up the queue if needed.
+		 * To solve this issue add a mb() to make sure that
+		 * netif_tx_stop_queue() write is vissible before checking if
+		 * there is additional space in the queue.
 		 */
-		smp_rmb();
+		smp_mb();
 
-		if (ena_com_sq_empty_space(tx_ring->ena_com_io_sq)
-				> ENA_TX_WAKEUP_THRESH) {
+		if (ena_com_sq_have_enough_space(tx_ring->ena_com_io_sq,
+						 ENA_TX_WAKEUP_THRESH)) {
 			netif_tx_wake_queue(txq);
-			u64_stats_update_begin(&tx_ring->syncp);
-			tx_ring->tx_stats.queue_wakeup++;
-			u64_stats_update_end(&tx_ring->syncp);
+			ena_increase_stat(&tx_ring->tx_stats.queue_wakeup, 1,
+					  &tx_ring->syncp);
 		}
 	}
 
-	if (netif_xmit_stopped(txq) || !skb->xmit_more) {
-		/* trigger the dma engine */
-		ena_com_write_sq_doorbell(tx_ring->ena_com_io_sq);
-		u64_stats_update_begin(&tx_ring->syncp);
-		tx_ring->tx_stats.doorbells++;
-		u64_stats_update_end(&tx_ring->syncp);
-	}
-
-	return NETDEV_TX_OK;
+	skb_tx_timestamp(skb);
 
-error_report_dma_error:
-	u64_stats_update_begin(&tx_ring->syncp);
-	tx_ring->tx_stats.dma_mapping_err++;
-	u64_stats_update_end(&tx_ring->syncp);
-	netdev_warn(adapter->netdev, "failed to map skb\n");
+	if (netif_xmit_stopped(txq) || !netdev_xmit_more())
+		/* trigger the dma engine. ena_ring_tx_doorbell()
+		 * calls a memory barrier inside it.
+		 */
+		ena_ring_tx_doorbell(tx_ring);
 
-	tx_info->skb = NULL;
+	return NETDEV_TX_OK;
 
 error_unmap_dma:
-	if (i >= 0) {
-		/* save value of frag that failed */
-		last_frag = i;
-
-		/* start back at beginning and unmap skb */
-		tx_info->skb = NULL;
-		ena_buf = tx_info->bufs;
-		dma_unmap_single(tx_ring->dev, dma_unmap_addr(ena_buf, paddr),
-				 dma_unmap_len(ena_buf, len), DMA_TO_DEVICE);
-
-		/* unmap remaining mapped pages */
-		for (i = 0; i < last_frag; i++) {
-			ena_buf++;
-			dma_unmap_page(tx_ring->dev, dma_unmap_addr(ena_buf, paddr),
-				       dma_unmap_len(ena_buf, len), DMA_TO_DEVICE);
-		}
-	}
+	ena_unmap_tx_buff(tx_ring, tx_info);
+	tx_info->skb = NULL;
 
 error_drop_packet:
-
 	dev_kfree_skb(skb);
 	return NETDEV_TX_OK;
 }
 
-#ifdef CONFIG_NET_POLL_CONTROLLER
-static void ena_netpoll(struct net_device *netdev)
-{
-	struct ena_adapter *adapter = netdev_priv(netdev);
-	int i;
-
-	/* Dont schedule NAPI if the driver is in the middle of reset
-	 * or netdev is down.
-	 */
-
-	if (!test_bit(ENA_FLAG_DEV_UP, &adapter->flags) ||
-	    test_bit(ENA_FLAG_TRIGGER_RESET, &adapter->flags))
-		return;
-
-	for (i = 0; i < adapter->num_queues; i++)
-		napi_schedule(&adapter->ena_napi[i].napi);
-}
-#endif /* CONFIG_NET_POLL_CONTROLLER */
-
-static u16 ena_select_queue(struct net_device *dev, struct sk_buff *skb,
-			    void *accel_priv, select_queue_fallback_t fallback)
-{
-	u16 qid;
-	/* we suspect that this is good for in--kernel network services that
-	 * want to loop incoming skb rx to tx in normal user generated traffic,
-	 * most probably we will not get to this
-	 */
-	if (skb_rx_queue_recorded(skb))
-		qid = skb_get_rx_queue(skb);
-	else
-		qid = fallback(dev, skb);
-
-	return qid;
-}
-
-static void ena_config_host_info(struct ena_com_dev *ena_dev)
+static void ena_config_host_info(struct ena_com_dev *ena_dev, struct pci_dev *pdev)
 {
+	struct device *dev = &pdev->dev;
 	struct ena_admin_host_info *host_info;
+	ssize_t ret;
 	int rc;
 
 	/* Allocate only the host info */
 	rc = ena_com_allocate_host_info(ena_dev);
 	if (rc) {
-		pr_err("Cannot allocate host info\n");
+		dev_err(dev, "Cannot allocate host info\n");
 		return;
 	}
 
 	host_info = ena_dev->host_attr.host_info;
 
+	host_info->bdf = pci_dev_id(pdev);
 	host_info->os_type = ENA_ADMIN_OS_LINUX;
 	host_info->kernel_ver = LINUX_VERSION_CODE;
-	strncpy(host_info->kernel_ver_str, utsname()->version,
-		sizeof(host_info->kernel_ver_str) - 1);
+	ret = strscpy(host_info->kernel_ver_str, utsname()->version,
+		      sizeof(host_info->kernel_ver_str));
+	if (ret < 0)
+		dev_dbg(dev,
+			"kernel version string will be truncated, status = %zd\n", ret);
+
 	host_info->os_dist = 0;
-	strncpy(host_info->os_dist_str, utsname()->release,
-		sizeof(host_info->os_dist_str) - 1);
+	ret = strscpy(host_info->os_dist_str, utsname()->release,
+		      sizeof(host_info->os_dist_str));
+	if (ret < 0)
+		dev_dbg(dev,
+			"OS distribution string will be truncated, status = %zd\n", ret);
+
 	host_info->driver_version =
-		(DRV_MODULE_VER_MAJOR) |
-		(DRV_MODULE_VER_MINOR << ENA_ADMIN_HOST_INFO_MINOR_SHIFT) |
-		(DRV_MODULE_VER_SUBMINOR << ENA_ADMIN_HOST_INFO_SUB_MINOR_SHIFT);
+		(DRV_MODULE_GEN_MAJOR) |
+		(DRV_MODULE_GEN_MINOR << ENA_ADMIN_HOST_INFO_MINOR_SHIFT) |
+		(DRV_MODULE_GEN_SUBMINOR << ENA_ADMIN_HOST_INFO_SUB_MINOR_SHIFT) |
+		("K"[0] << ENA_ADMIN_HOST_INFO_MODULE_TYPE_SHIFT);
+	host_info->num_cpus = num_online_cpus();
+
+	host_info->driver_supported_features =
+		ENA_ADMIN_HOST_INFO_RX_OFFSET_MASK |
+		ENA_ADMIN_HOST_INFO_INTERRUPT_MODERATION_MASK |
+		ENA_ADMIN_HOST_INFO_RX_BUF_MIRRORING_MASK |
+		ENA_ADMIN_HOST_INFO_RSS_CONFIGURABLE_FUNCTION_KEY_MASK |
+		ENA_ADMIN_HOST_INFO_RX_PAGE_REUSE_MASK |
+		ENA_ADMIN_HOST_INFO_PHC_MASK;
 
 	rc = ena_com_set_host_attributes(ena_dev);
 	if (rc) {
-		if (rc == -EPERM)
-			pr_warn("Cannot set host attributes\n");
+		if (rc == -EOPNOTSUPP)
+			dev_warn(dev, "Cannot set host attributes\n");
 		else
-			pr_err("Cannot set host attributes\n");
+			dev_err(dev, "Cannot set host attributes\n");
 
 		goto err;
 	}
@@ -2160,15 +2783,15 @@ static void ena_config_debug_area(struct ena_adapter *adapter)
 
 	rc = ena_com_allocate_debug_area(adapter->ena_dev, debug_area_size);
 	if (rc) {
-		pr_err("Cannot allocate debug area\n");
+		netif_err(adapter, drv, adapter->netdev,
+			  "Cannot allocate debug area\n");
 		return;
 	}
 
 	rc = ena_com_set_host_attributes(adapter->ena_dev);
 	if (rc) {
-		if (rc == -EPERM)
-			netif_warn(adapter, drv, adapter->netdev,
-				   "Cannot set host attributes\n");
+		if (rc == -EOPNOTSUPP)
+			netif_warn(adapter, drv, adapter->netdev, "Cannot set host attributes\n");
 		else
 			netif_err(adapter, drv, adapter->netdev,
 				  "Cannot set host attributes\n");
@@ -2185,45 +2808,55 @@ static void ena_get_stats64(struct net_device *netdev,
 {
 	struct ena_adapter *adapter = netdev_priv(netdev);
 	struct ena_ring *rx_ring, *tx_ring;
+	u64 total_xdp_rx_drops = 0;
 	unsigned int start;
 	u64 rx_drops;
+	u64 tx_drops;
 	int i;
 
 	if (!test_bit(ENA_FLAG_DEV_UP, &adapter->flags))
 		return;
 
-	for (i = 0; i < adapter->num_queues; i++) {
-		u64 bytes, packets;
+	for (i = 0; i < adapter->num_io_queues + adapter->xdp_num_queues; i++) {
+		u64 bytes, packets, xdp_rx_drops;
 
 		tx_ring = &adapter->tx_ring[i];
 
 		do {
-			start = u64_stats_fetch_begin_irq(&tx_ring->syncp);
+			start = u64_stats_fetch_begin(&tx_ring->syncp);
 			packets = tx_ring->tx_stats.cnt;
 			bytes = tx_ring->tx_stats.bytes;
-		} while (u64_stats_fetch_retry_irq(&tx_ring->syncp, start));
+		} while (u64_stats_fetch_retry(&tx_ring->syncp, start));
 
 		stats->tx_packets += packets;
 		stats->tx_bytes += bytes;
 
+		/* In XDP there isn't an RX queue counterpart */
+		if (ENA_IS_XDP_INDEX(adapter, i))
+			continue;
+
 		rx_ring = &adapter->rx_ring[i];
 
 		do {
-			start = u64_stats_fetch_begin_irq(&rx_ring->syncp);
+			start = u64_stats_fetch_begin(&rx_ring->syncp);
 			packets = rx_ring->rx_stats.cnt;
 			bytes = rx_ring->rx_stats.bytes;
-		} while (u64_stats_fetch_retry_irq(&rx_ring->syncp, start));
+			xdp_rx_drops = rx_ring->rx_stats.xdp_drop;
+		} while (u64_stats_fetch_retry(&rx_ring->syncp, start));
 
 		stats->rx_packets += packets;
 		stats->rx_bytes += bytes;
+		total_xdp_rx_drops += xdp_rx_drops;
 	}
 
 	do {
-		start = u64_stats_fetch_begin_irq(&adapter->syncp);
+		start = u64_stats_fetch_begin(&adapter->syncp);
 		rx_drops = adapter->dev_stats.rx_drops;
-	} while (u64_stats_fetch_retry_irq(&adapter->syncp, start));
+		tx_drops = adapter->dev_stats.tx_drops;
+	} while (u64_stats_fetch_retry(&adapter->syncp, start));
 
-	stats->rx_dropped = rx_drops;
+	stats->rx_dropped = rx_drops + total_xdp_rx_drops;
+	stats->tx_dropped = tx_drops;
 
 	stats->multicast = 0;
 	stats->collisions = 0;
@@ -2243,47 +2876,118 @@ static const struct net_device_ops ena_netdev_ops = {
 	.ndo_open		= ena_open,
 	.ndo_stop		= ena_close,
 	.ndo_start_xmit		= ena_start_xmit,
-	.ndo_select_queue	= ena_select_queue,
 	.ndo_get_stats64	= ena_get_stats64,
 	.ndo_tx_timeout		= ena_tx_timeout,
 	.ndo_change_mtu		= ena_change_mtu,
-	.ndo_set_mac_address	= NULL,
 	.ndo_validate_addr	= eth_validate_addr,
-#ifdef CONFIG_NET_POLL_CONTROLLER
-	.ndo_poll_controller	= ena_netpoll,
-#endif /* CONFIG_NET_POLL_CONTROLLER */
+	.ndo_bpf		= ena_xdp,
+	.ndo_xdp_xmit		= ena_xdp_xmit,
 };
 
-static void ena_device_io_suspend(struct work_struct *work)
+static int ena_calc_io_queue_size(struct ena_adapter *adapter,
+				  struct ena_com_dev_get_features_ctx *get_feat_ctx)
 {
-	struct ena_adapter *adapter =
-		container_of(work, struct ena_adapter, suspend_io_task);
-	struct net_device *netdev = adapter->netdev;
+	struct ena_admin_feature_llq_desc *llq = &get_feat_ctx->llq;
+	struct ena_com_dev *ena_dev = adapter->ena_dev;
+	u32 tx_queue_size = ENA_DEFAULT_RING_SIZE;
+	u32 rx_queue_size = ENA_DEFAULT_RING_SIZE;
+	u32 max_tx_queue_size;
+	u32 max_rx_queue_size;
 
-	/* ena_napi_disable_all disables only the IO handling.
-	 * We are still subject to AENQ keep alive watchdog.
+	/* If this function is called after driver load, the ring sizes have already
+	 * been configured. Take it into account when recalculating ring size.
 	 */
-	u64_stats_update_begin(&adapter->syncp);
-	adapter->dev_stats.io_suspend++;
-	u64_stats_update_begin(&adapter->syncp);
-	ena_napi_disable_all(adapter);
-	netif_tx_lock(netdev);
-	netif_device_detach(netdev);
-	netif_tx_unlock(netdev);
-}
+	if (adapter->tx_ring->ring_size)
+		tx_queue_size = adapter->tx_ring->ring_size;
+
+	if (adapter->rx_ring->ring_size)
+		rx_queue_size = adapter->rx_ring->ring_size;
+
+	if (ena_dev->supported_features & BIT(ENA_ADMIN_MAX_QUEUES_EXT)) {
+		struct ena_admin_queue_ext_feature_fields *max_queue_ext =
+			&get_feat_ctx->max_queue_ext.max_queue_ext;
+		max_rx_queue_size = min_t(u32, max_queue_ext->max_rx_cq_depth,
+					  max_queue_ext->max_rx_sq_depth);
+		max_tx_queue_size = max_queue_ext->max_tx_cq_depth;
+
+		if (ena_dev->tx_mem_queue_type == ENA_ADMIN_PLACEMENT_POLICY_DEV)
+			max_tx_queue_size = min_t(u32, max_tx_queue_size,
+						  llq->max_llq_depth);
+		else
+			max_tx_queue_size = min_t(u32, max_tx_queue_size,
+						  max_queue_ext->max_tx_sq_depth);
 
-static void ena_device_io_resume(struct work_struct *work)
-{
-	struct ena_adapter *adapter =
-		container_of(work, struct ena_adapter, resume_io_task);
-	struct net_device *netdev = adapter->netdev;
+		adapter->max_tx_sgl_size = min_t(u16, ENA_PKT_MAX_BUFS,
+						 max_queue_ext->max_per_packet_tx_descs);
+		adapter->max_rx_sgl_size = min_t(u16, ENA_PKT_MAX_BUFS,
+						 max_queue_ext->max_per_packet_rx_descs);
+	} else {
+		struct ena_admin_queue_feature_desc *max_queues =
+			&get_feat_ctx->max_queues;
+		max_rx_queue_size = min_t(u32, max_queues->max_cq_depth,
+					  max_queues->max_sq_depth);
+		max_tx_queue_size = max_queues->max_cq_depth;
+
+		if (ena_dev->tx_mem_queue_type == ENA_ADMIN_PLACEMENT_POLICY_DEV)
+			max_tx_queue_size = min_t(u32, max_tx_queue_size,
+						  llq->max_llq_depth);
+		else
+			max_tx_queue_size = min_t(u32, max_tx_queue_size,
+						  max_queues->max_sq_depth);
 
-	u64_stats_update_begin(&adapter->syncp);
-	adapter->dev_stats.io_resume++;
-	u64_stats_update_end(&adapter->syncp);
+		adapter->max_tx_sgl_size = min_t(u16, ENA_PKT_MAX_BUFS,
+						 max_queues->max_packet_tx_descs);
+		adapter->max_rx_sgl_size = min_t(u16, ENA_PKT_MAX_BUFS,
+						 max_queues->max_packet_rx_descs);
+	}
+
+	max_tx_queue_size = rounddown_pow_of_two(max_tx_queue_size);
+	max_rx_queue_size = rounddown_pow_of_two(max_rx_queue_size);
 
-	netif_device_attach(netdev);
-	ena_napi_enable_all(adapter);
+	if (max_tx_queue_size < ENA_MIN_RING_SIZE) {
+		netdev_err(adapter->netdev, "Device max TX queue size: %d < minimum: %d\n",
+			   max_tx_queue_size, ENA_MIN_RING_SIZE);
+		return -EINVAL;
+	}
+
+	if (max_rx_queue_size < ENA_MIN_RING_SIZE) {
+		netdev_err(adapter->netdev, "Device max RX queue size: %d < minimum: %d\n",
+			   max_rx_queue_size, ENA_MIN_RING_SIZE);
+		return -EINVAL;
+	}
+
+	/* When forcing large headers, we multiply the entry size by 2, and therefore divide
+	 * the queue size by 2, leaving the amount of memory used by the queues unchanged.
+	 */
+	if (adapter->large_llq_header_enabled) {
+		if ((llq->entry_size_ctrl_supported & ENA_ADMIN_LIST_ENTRY_SIZE_256B) &&
+		    ena_dev->tx_mem_queue_type == ENA_ADMIN_PLACEMENT_POLICY_DEV) {
+			max_tx_queue_size /= 2;
+			dev_info(&adapter->pdev->dev,
+				 "Forcing large headers and decreasing maximum TX queue size to %d\n",
+				 max_tx_queue_size);
+		} else {
+			dev_err(&adapter->pdev->dev,
+				"Forcing large headers failed: LLQ is disabled or device does not support large headers\n");
+
+			adapter->large_llq_header_enabled = false;
+		}
+	}
+
+	tx_queue_size = clamp_val(tx_queue_size, ENA_MIN_RING_SIZE,
+				  max_tx_queue_size);
+	rx_queue_size = clamp_val(rx_queue_size, ENA_MIN_RING_SIZE,
+				  max_rx_queue_size);
+
+	tx_queue_size = rounddown_pow_of_two(tx_queue_size);
+	rx_queue_size = rounddown_pow_of_two(rx_queue_size);
+
+	adapter->max_tx_ring_size  = max_tx_queue_size;
+	adapter->max_rx_ring_size = max_rx_queue_size;
+	adapter->requested_tx_ring_size = tx_queue_size;
+	adapter->requested_rx_ring_size = rx_queue_size;
+
+	return 0;
 }
 
 static int ena_device_validate_params(struct ena_adapter *adapter,
@@ -2300,13 +3004,6 @@ static int ena_device_validate_params(struct ena_adapter *adapter,
 		return -EINVAL;
 	}
 
-	if ((get_feat_ctx->max_queues.max_cq_num < adapter->num_queues) ||
-	    (get_feat_ctx->max_queues.max_sq_num < adapter->num_queues)) {
-		netif_err(adapter, drv, netdev,
-			  "Error, device doesn't support enough queues\n");
-		return -EINVAL;
-	}
-
 	if (get_feat_ctx->dev_attr.max_mtu < netdev->mtu) {
 		netif_err(adapter, drv, netdev,
 			  "Error, device max mtu is smaller than netdev MTU\n");
@@ -2316,10 +3013,90 @@ static int ena_device_validate_params(struct ena_adapter *adapter,
 	return 0;
 }
 
-static int ena_device_init(struct ena_com_dev *ena_dev, struct pci_dev *pdev,
+static void set_default_llq_configurations(struct ena_adapter *adapter,
+					   struct ena_llq_configurations *llq_config,
+					   struct ena_admin_feature_llq_desc *llq)
+{
+	struct ena_com_dev *ena_dev = adapter->ena_dev;
+
+	llq_config->llq_header_location = ENA_ADMIN_INLINE_HEADER;
+	llq_config->llq_stride_ctrl = ENA_ADMIN_MULTIPLE_DESCS_PER_ENTRY;
+	llq_config->llq_num_decs_before_header = ENA_ADMIN_LLQ_NUM_DESCS_BEFORE_HEADER_2;
+
+	adapter->large_llq_header_supported =
+		!!(ena_dev->supported_features & BIT(ENA_ADMIN_LLQ));
+	adapter->large_llq_header_supported &=
+		!!(llq->entry_size_ctrl_supported &
+			ENA_ADMIN_LIST_ENTRY_SIZE_256B);
+
+	if ((llq->entry_size_ctrl_supported & ENA_ADMIN_LIST_ENTRY_SIZE_256B) &&
+	    adapter->large_llq_header_enabled) {
+		llq_config->llq_ring_entry_size = ENA_ADMIN_LIST_ENTRY_SIZE_256B;
+		llq_config->llq_ring_entry_size_value = 256;
+	} else {
+		llq_config->llq_ring_entry_size = ENA_ADMIN_LIST_ENTRY_SIZE_128B;
+		llq_config->llq_ring_entry_size_value = 128;
+	}
+}
+
+static int ena_set_queues_placement_policy(struct pci_dev *pdev,
+					   struct ena_com_dev *ena_dev,
+					   struct ena_admin_feature_llq_desc *llq,
+					   struct ena_llq_configurations *llq_default_configurations)
+{
+	int rc;
+	u32 llq_feature_mask;
+
+	llq_feature_mask = 1 << ENA_ADMIN_LLQ;
+	if (!(ena_dev->supported_features & llq_feature_mask)) {
+		dev_warn(&pdev->dev,
+			"LLQ is not supported Fallback to host mode policy.\n");
+		ena_dev->tx_mem_queue_type = ENA_ADMIN_PLACEMENT_POLICY_HOST;
+		return 0;
+	}
+
+	if (!ena_dev->mem_bar) {
+		netdev_err(ena_dev->net_device,
+			   "LLQ is advertised as supported but device doesn't expose mem bar\n");
+		ena_dev->tx_mem_queue_type = ENA_ADMIN_PLACEMENT_POLICY_HOST;
+		return 0;
+	}
+
+	rc = ena_com_config_dev_mode(ena_dev, llq, llq_default_configurations);
+	if (unlikely(rc)) {
+		dev_err(&pdev->dev,
+			"Failed to configure the device mode.  Fallback to host mode policy.\n");
+		ena_dev->tx_mem_queue_type = ENA_ADMIN_PLACEMENT_POLICY_HOST;
+	}
+
+	return 0;
+}
+
+static int ena_map_llq_mem_bar(struct pci_dev *pdev, struct ena_com_dev *ena_dev,
+			       int bars)
+{
+	bool has_mem_bar = !!(bars & BIT(ENA_MEM_BAR));
+
+	if (!has_mem_bar)
+		return 0;
+
+	ena_dev->mem_bar = devm_ioremap_wc(&pdev->dev,
+					   pci_resource_start(pdev, ENA_MEM_BAR),
+					   pci_resource_len(pdev, ENA_MEM_BAR));
+
+	if (!ena_dev->mem_bar)
+		return -EFAULT;
+
+	return 0;
+}
+
+static int ena_device_init(struct ena_adapter *adapter, struct pci_dev *pdev,
 			   struct ena_com_dev_get_features_ctx *get_feat_ctx,
 			   bool *wd_state)
 {
+	struct ena_com_dev *ena_dev = adapter->ena_dev;
+	struct net_device *netdev = adapter->netdev;
+	struct ena_llq_configurations llq_config;
 	struct device *dev = &pdev->dev;
 	bool readless_supported;
 	u32 aenq_groups;
@@ -2328,7 +3105,7 @@ static int ena_device_init(struct ena_com_dev *ena_dev, struct pci_dev *pdev,
 
 	rc = ena_com_mmio_reg_read_request_init(ena_dev);
 	if (rc) {
-		dev_err(dev, "failed to init mmio read less\n");
+		dev_err(dev, "Failed to init mmio read less\n");
 		return rc;
 	}
 
@@ -2338,7 +3115,7 @@ static int ena_device_init(struct ena_com_dev *ena_dev, struct pci_dev *pdev,
 	readless_supported = !(pdev->revision & ENA_MMIO_DISABLE_REG_READ);
 	ena_com_set_mmio_read_mode(ena_dev, readless_supported);
 
-	rc = ena_com_dev_reset(ena_dev);
+	rc = ena_com_dev_reset(ena_dev, ENA_REGS_RESET_NORMAL);
 	if (rc) {
 		dev_err(dev, "Can not reset device\n");
 		goto err_mmio_read_less;
@@ -2346,7 +3123,7 @@ static int ena_device_init(struct ena_com_dev *ena_dev, struct pci_dev *pdev,
 
 	rc = ena_com_validate_version(ena_dev);
 	if (rc) {
-		dev_err(dev, "device version is too low\n");
+		dev_err(dev, "Device version is too low\n");
 		goto err_mmio_read_less;
 	}
 
@@ -2357,21 +3134,16 @@ static int ena_device_init(struct ena_com_dev *ena_dev, struct pci_dev *pdev,
 		goto err_mmio_read_less;
 	}
 
-	rc = pci_set_dma_mask(pdev, DMA_BIT_MASK(dma_width));
+	rc = dma_set_mask_and_coherent(dev, DMA_BIT_MASK(dma_width));
 	if (rc) {
-		dev_err(dev, "pci_set_dma_mask failed 0x%x\n", rc);
+		dev_err(dev, "dma_set_mask_and_coherent failed %d\n", rc);
 		goto err_mmio_read_less;
 	}
 
-	rc = pci_set_consistent_dma_mask(pdev, DMA_BIT_MASK(dma_width));
-	if (rc) {
-		dev_err(dev, "err_pci_set_consistent_dma_mask failed 0x%x\n",
-			rc);
-		goto err_mmio_read_less;
-	}
+	ena_devlink_params_get(adapter->devlink);
 
 	/* ENA admin level init */
-	rc = ena_com_admin_init(ena_dev, &aenq_handlers, true);
+	rc = ena_com_admin_init(ena_dev, &aenq_handlers);
 	if (rc) {
 		dev_err(dev,
 			"Can not initialize ena admin queue with device\n");
@@ -2384,7 +3156,7 @@ static int ena_device_init(struct ena_com_dev *ena_dev, struct pci_dev *pdev,
 	 */
 	ena_com_set_admin_polling_mode(ena_dev, true);
 
-	ena_config_host_info(ena_dev);
+	ena_config_host_info(ena_dev, pdev);
 
 	/* Get Device Attributes*/
 	rc = ena_com_get_dev_attr_feat(ena_dev, get_feat_ctx);
@@ -2410,9 +3182,28 @@ static int ena_device_init(struct ena_com_dev *ena_dev, struct pci_dev *pdev,
 
 	*wd_state = !!(aenq_groups & BIT(ENA_ADMIN_KEEP_ALIVE));
 
+	set_default_llq_configurations(adapter, &llq_config, &get_feat_ctx->llq);
+
+	rc = ena_set_queues_placement_policy(pdev, ena_dev, &get_feat_ctx->llq,
+					     &llq_config);
+	if (rc) {
+		netdev_err(netdev, "Cannot set queues placement policy rc= %d\n", rc);
+		goto err_admin_init;
+	}
+
+	rc = ena_calc_io_queue_size(adapter, get_feat_ctx);
+	if (unlikely(rc))
+		goto err_admin_init;
+
+	rc = ena_phc_init(adapter);
+	if (unlikely(rc && (rc != -EOPNOTSUPP)))
+		netdev_err(netdev, "Failed initializing PHC, error: %d\n", rc);
+
 	return 0;
 
 err_admin_init:
+	ena_com_abort_admin_commands(ena_dev);
+	ena_com_wait_for_abort_completion(ena_dev);
 	ena_com_delete_host_info(ena_dev);
 	ena_com_admin_destroy(ena_dev);
 err_mmio_read_less:
@@ -2421,14 +3212,13 @@ err_mmio_read_less:
 	return rc;
 }
 
-static int ena_enable_msix_and_set_admin_interrupts(struct ena_adapter *adapter,
-						    int io_vectors)
+static int ena_enable_msix_and_set_admin_interrupts(struct ena_adapter *adapter)
 {
 	struct ena_com_dev *ena_dev = adapter->ena_dev;
 	struct device *dev = &adapter->pdev->dev;
 	int rc;
 
-	rc = ena_enable_msix(adapter, io_vectors);
+	rc = ena_enable_msix(adapter);
 	if (rc) {
 		dev_err(dev, "Can not reserve msix vectors\n");
 		return rc;
@@ -2449,45 +3239,42 @@ static int ena_enable_msix_and_set_admin_interrupts(struct ena_adapter *adapter,
 	return 0;
 
 err_disable_msix:
-	pci_free_irq_vectors(adapter->pdev);
+	ena_disable_msix(adapter);
+
 	return rc;
 }
 
-static void ena_fw_reset_device(struct work_struct *work)
+int ena_destroy_device(struct ena_adapter *adapter, bool graceful)
 {
-	struct ena_com_dev_get_features_ctx get_feat_ctx;
-	struct ena_adapter *adapter =
-		container_of(work, struct ena_adapter, reset_task);
 	struct net_device *netdev = adapter->netdev;
 	struct ena_com_dev *ena_dev = adapter->ena_dev;
-	struct pci_dev *pdev = adapter->pdev;
-	bool dev_up, wd_state;
-	int rc;
+	bool dev_up;
+	int rc = 0;
 
-	if (unlikely(!test_bit(ENA_FLAG_TRIGGER_RESET, &adapter->flags))) {
-		dev_err(&pdev->dev,
-			"device reset schedule while reset bit is off\n");
-		return;
-	}
+	if (!test_bit(ENA_FLAG_DEVICE_RUNNING, &adapter->flags))
+		return 0;
 
 	netif_carrier_off(netdev);
 
-	del_timer_sync(&adapter->timer_service);
-
-	rtnl_lock();
+	timer_delete_sync(&adapter->timer_service);
 
 	dev_up = test_bit(ENA_FLAG_DEV_UP, &adapter->flags);
-	ena_com_set_admin_running_state(ena_dev, false);
+	adapter->dev_up_before_reset = dev_up;
+	if (!graceful)
+		ena_com_set_admin_running_state(ena_dev, false);
+
+	if (dev_up)
+		ena_down(adapter);
 
-	/* After calling ena_close the tx queues and the napi
-	 * are disabled so no one can interfere or touch the
-	 * data structures
+	/* Stop the device from sending AENQ events (in case reset flag is set
+	 *  and device is up, ena_down() already reset the device.
 	 */
-	ena_close(netdev);
+	if (!(test_bit(ENA_FLAG_TRIGGER_RESET, &adapter->flags) && dev_up))
+		rc = ena_com_dev_reset(adapter->ena_dev, adapter->reset_reason);
 
 	ena_free_mgmnt_irq(adapter);
 
-	pci_free_irq_vectors(adapter->pdev);
+	ena_disable_msix(adapter);
 
 	ena_com_abort_admin_commands(ena_dev);
 
@@ -2495,33 +3282,56 @@ static void ena_fw_reset_device(struct work_struct *work)
 
 	ena_com_admin_destroy(ena_dev);
 
+	ena_phc_destroy(adapter);
+
 	ena_com_mmio_reg_read_request_destroy(ena_dev);
 
+	/* return reset reason to default value */
+	adapter->reset_reason = ENA_REGS_RESET_NORMAL;
+
 	clear_bit(ENA_FLAG_TRIGGER_RESET, &adapter->flags);
+	clear_bit(ENA_FLAG_DEVICE_RUNNING, &adapter->flags);
+
+	return rc;
+}
 
-	/* Finish with the destroy part. Start the init part */
+int ena_restore_device(struct ena_adapter *adapter)
+{
+	struct ena_com_dev_get_features_ctx get_feat_ctx;
+	struct ena_com_dev *ena_dev = adapter->ena_dev;
+	struct pci_dev *pdev = adapter->pdev;
+	struct ena_ring *txr;
+	int rc, count, i;
+	bool wd_state;
 
-	rc = ena_device_init(ena_dev, adapter->pdev, &get_feat_ctx, &wd_state);
+	set_bit(ENA_FLAG_ONGOING_RESET, &adapter->flags);
+	rc = ena_device_init(adapter, adapter->pdev, &get_feat_ctx, &wd_state);
 	if (rc) {
 		dev_err(&pdev->dev, "Can not initialize device\n");
 		goto err;
 	}
 	adapter->wd_state = wd_state;
 
+	count =  adapter->xdp_num_queues + adapter->num_io_queues;
+	for (i = 0 ; i < count; i++) {
+		txr = &adapter->tx_ring[i];
+		txr->tx_mem_queue_type = ena_dev->tx_mem_queue_type;
+		txr->tx_max_header_size = ena_dev->tx_max_header_size;
+	}
+
 	rc = ena_device_validate_params(adapter, &get_feat_ctx);
 	if (rc) {
 		dev_err(&pdev->dev, "Validation of device parameters failed\n");
 		goto err_device_destroy;
 	}
 
-	rc = ena_enable_msix_and_set_admin_interrupts(adapter,
-						      adapter->num_queues);
+	rc = ena_enable_msix_and_set_admin_interrupts(adapter);
 	if (rc) {
 		dev_err(&pdev->dev, "Enable MSI-X failed\n");
 		goto err_device_destroy;
 	}
 	/* If the interface was up before the reset bring it up */
-	if (dev_up) {
+	if (adapter->dev_up_before_reset) {
 		rc = ena_up(adapter);
 		if (rc) {
 			dev_err(&pdev->dev, "Failed to create I/O queues\n");
@@ -2529,34 +3339,178 @@ static void ena_fw_reset_device(struct work_struct *work)
 		}
 	}
 
-	mod_timer(&adapter->timer_service, round_jiffies(jiffies + HZ));
+	set_bit(ENA_FLAG_DEVICE_RUNNING, &adapter->flags);
 
-	rtnl_unlock();
+	clear_bit(ENA_FLAG_ONGOING_RESET, &adapter->flags);
+	if (test_bit(ENA_FLAG_LINK_UP, &adapter->flags))
+		netif_carrier_on(adapter->netdev);
 
-	dev_err(&pdev->dev, "Device reset completed successfully\n");
+	mod_timer(&adapter->timer_service, round_jiffies(jiffies + HZ));
+	adapter->last_keep_alive_jiffies = jiffies;
 
-	return;
+	return rc;
 err_disable_msix:
 	ena_free_mgmnt_irq(adapter);
-	pci_free_irq_vectors(adapter->pdev);
+	ena_disable_msix(adapter);
 err_device_destroy:
+	ena_com_abort_admin_commands(ena_dev);
+	ena_com_wait_for_abort_completion(ena_dev);
 	ena_com_admin_destroy(ena_dev);
+	ena_com_dev_reset(ena_dev, ENA_REGS_RESET_DRIVER_INVALID_STATE);
+	ena_phc_destroy(adapter);
+	ena_com_mmio_reg_read_request_destroy(ena_dev);
 err:
-	rtnl_unlock();
-
 	clear_bit(ENA_FLAG_DEVICE_RUNNING, &adapter->flags);
-
+	clear_bit(ENA_FLAG_ONGOING_RESET, &adapter->flags);
 	dev_err(&pdev->dev,
 		"Reset attempt failed. Can not reset the device\n");
+
+	return rc;
 }
 
-static void check_for_missing_tx_completions(struct ena_adapter *adapter)
+static void ena_fw_reset_device(struct work_struct *work)
 {
+	int rc = 0;
+
+	struct ena_adapter *adapter =
+		container_of(work, struct ena_adapter, reset_task);
+
+	rtnl_lock();
+
+	if (likely(test_bit(ENA_FLAG_TRIGGER_RESET, &adapter->flags))) {
+		rc |= ena_destroy_device(adapter, false);
+		rc |= ena_restore_device(adapter);
+		adapter->dev_stats.reset_fail += !!rc;
+
+		dev_err(&adapter->pdev->dev, "Device reset completed successfully\n");
+	}
+
+	rtnl_unlock();
+}
+
+static int check_for_rx_interrupt_queue(struct ena_adapter *adapter,
+					struct ena_ring *rx_ring)
+{
+	struct ena_napi *ena_napi = container_of(rx_ring->napi, struct ena_napi, napi);
+
+	if (likely(READ_ONCE(ena_napi->first_interrupt)))
+		return 0;
+
+	if (ena_com_cq_empty(rx_ring->ena_com_io_cq))
+		return 0;
+
+	rx_ring->no_interrupt_event_cnt++;
+
+	if (rx_ring->no_interrupt_event_cnt == ENA_MAX_NO_INTERRUPT_ITERATIONS) {
+		netif_err(adapter, rx_err, adapter->netdev,
+			  "Potential MSIX issue on Rx side Queue = %d. Reset the device\n",
+			  rx_ring->qid);
+
+		ena_reset_device(adapter, ENA_REGS_RESET_MISS_INTERRUPT);
+		return -EIO;
+	}
+
+	return 0;
+}
+
+static int check_missing_comp_in_tx_queue(struct ena_adapter *adapter,
+					  struct ena_ring *tx_ring)
+{
+	struct ena_napi *ena_napi = container_of(tx_ring->napi, struct ena_napi, napi);
+	enum ena_regs_reset_reason_types reset_reason = ENA_REGS_RESET_MISS_TX_CMPL;
+	unsigned int time_since_last_napi;
+	unsigned int missing_tx_comp_to;
+	bool is_tx_comp_time_expired;
 	struct ena_tx_buffer *tx_buf;
 	unsigned long last_jiffies;
+	int napi_scheduled;
+	u32 missed_tx = 0;
+	int i, rc = 0;
+
+	missing_tx_comp_to = jiffies_to_msecs(adapter->missing_tx_completion_to);
+
+	for (i = 0; i < tx_ring->ring_size; i++) {
+		tx_buf = &tx_ring->tx_buffer_info[i];
+		last_jiffies = tx_buf->last_jiffies;
+
+		if (last_jiffies == 0)
+			/* no pending Tx at this location */
+			continue;
+
+		is_tx_comp_time_expired = time_is_before_jiffies(last_jiffies +
+			 2 * adapter->missing_tx_completion_to);
+
+		if (unlikely(!READ_ONCE(ena_napi->first_interrupt) && is_tx_comp_time_expired)) {
+			/* If after graceful period interrupt is still not
+			 * received, we schedule a reset
+			 */
+			netif_err(adapter, tx_err, adapter->netdev,
+				  "Potential MSIX issue on Tx side Queue = %d. Reset the device\n",
+				  tx_ring->qid);
+			ena_reset_device(adapter, ENA_REGS_RESET_MISS_INTERRUPT);
+			return -EIO;
+		}
+
+		is_tx_comp_time_expired = time_is_before_jiffies(last_jiffies +
+			adapter->missing_tx_completion_to);
+
+		if (unlikely(is_tx_comp_time_expired)) {
+			time_since_last_napi =
+				jiffies_to_usecs(jiffies - tx_ring->tx_stats.last_napi_jiffies);
+			napi_scheduled = !!(ena_napi->napi.state & NAPIF_STATE_SCHED);
+
+			if (missing_tx_comp_to < time_since_last_napi && napi_scheduled) {
+				/* We suspect napi isn't called because the
+				 * bottom half is not run. Require a bigger
+				 * timeout for these cases
+				 */
+				if (!time_is_before_jiffies(last_jiffies +
+					2 * adapter->missing_tx_completion_to))
+					continue;
+
+				reset_reason = ENA_REGS_RESET_SUSPECTED_POLL_STARVATION;
+			}
+
+			missed_tx++;
+
+			if (tx_buf->print_once)
+				continue;
+
+			netif_notice(adapter, tx_err, adapter->netdev,
+				     "TX hasn't completed, qid %d, index %d. %u usecs from last napi execution, napi scheduled: %d\n",
+				     tx_ring->qid, i, time_since_last_napi, napi_scheduled);
+
+			tx_buf->print_once = 1;
+		}
+	}
+
+	if (unlikely(missed_tx > adapter->missing_tx_completion_threshold)) {
+		netif_err(adapter, tx_err, adapter->netdev,
+			  "Lost TX completions are above the threshold (%d > %d). Completion transmission timeout: %u.\n",
+			  missed_tx,
+			  adapter->missing_tx_completion_threshold,
+			  missing_tx_comp_to);
+		netif_err(adapter, tx_err, adapter->netdev,
+			  "Resetting the device\n");
+
+		ena_reset_device(adapter, reset_reason);
+		rc = -EIO;
+	}
+
+	ena_increase_stat(&tx_ring->tx_stats.missed_tx, missed_tx,
+			  &tx_ring->syncp);
+
+	return rc;
+}
+
+static void check_for_missing_completions(struct ena_adapter *adapter)
+{
 	struct ena_ring *tx_ring;
-	int i, j, budget;
-	u32 missed_tx;
+	struct ena_ring *rx_ring;
+	int qid, budget, rc;
+	int io_queue_count;
+
+	io_queue_count = adapter->xdp_num_queues + adapter->num_io_queues;
 
 	/* Make sure the driver doesn't turn the device in other process */
 	smp_rmb();
@@ -2567,43 +3521,82 @@ static void check_for_missing_tx_completions(struct ena_adapter *adapter)
 	if (test_bit(ENA_FLAG_TRIGGER_RESET, &adapter->flags))
 		return;
 
-	budget = ENA_MONITORED_TX_QUEUES;
+	if (adapter->missing_tx_completion_to == ENA_HW_HINTS_NO_TIMEOUT)
+		return;
 
-	for (i = adapter->last_monitored_tx_qid; i < adapter->num_queues; i++) {
-		tx_ring = &adapter->tx_ring[i];
+	budget = min_t(u32, io_queue_count, ENA_MONITORED_TX_QUEUES);
 
-		for (j = 0; j < tx_ring->ring_size; j++) {
-			tx_buf = &tx_ring->tx_buffer_info[j];
-			last_jiffies = tx_buf->last_jiffies;
-			if (unlikely(last_jiffies && time_is_before_jiffies(last_jiffies + TX_TIMEOUT))) {
-				netif_notice(adapter, tx_err, adapter->netdev,
-					     "Found a Tx that wasn't completed on time, qid %d, index %d.\n",
-					     tx_ring->qid, j);
+	qid = adapter->last_monitored_tx_qid;
 
-				u64_stats_update_begin(&tx_ring->syncp);
-				missed_tx = tx_ring->tx_stats.missing_tx_comp++;
-				u64_stats_update_end(&tx_ring->syncp);
+	while (budget) {
+		qid = (qid + 1) % io_queue_count;
 
-				/* Clear last jiffies so the lost buffer won't
-				 * be counted twice.
-				 */
-				tx_buf->last_jiffies = 0;
+		tx_ring = &adapter->tx_ring[qid];
+		rx_ring = &adapter->rx_ring[qid];
 
-				if (unlikely(missed_tx > MAX_NUM_OF_TIMEOUTED_PACKETS)) {
-					netif_err(adapter, tx_err, adapter->netdev,
-						  "The number of lost tx completion is above the threshold (%d > %d). Reset the device\n",
-						  missed_tx, MAX_NUM_OF_TIMEOUTED_PACKETS);
-					set_bit(ENA_FLAG_TRIGGER_RESET, &adapter->flags);
-				}
-			}
-		}
+		rc = check_missing_comp_in_tx_queue(adapter, tx_ring);
+		if (unlikely(rc))
+			return;
+
+		rc =  !ENA_IS_XDP_INDEX(adapter, qid) ?
+			check_for_rx_interrupt_queue(adapter, rx_ring) : 0;
+		if (unlikely(rc))
+			return;
 
 		budget--;
-		if (!budget)
-			break;
 	}
 
-	adapter->last_monitored_tx_qid = i % adapter->num_queues;
+	adapter->last_monitored_tx_qid = qid;
+}
+
+/* trigger napi schedule after 2 consecutive detections */
+#define EMPTY_RX_REFILL 2
+/* For the rare case where the device runs out of Rx descriptors and the
+ * napi handler failed to refill new Rx descriptors (due to a lack of memory
+ * for example).
+ * This case will lead to a deadlock:
+ * The device won't send interrupts since all the new Rx packets will be dropped
+ * The napi handler won't allocate new Rx descriptors so the device will be
+ * able to send new packets.
+ *
+ * This scenario can happen when the kernel's vm.min_free_kbytes is too small.
+ * It is recommended to have at least 512MB, with a minimum of 128MB for
+ * constrained environment).
+ *
+ * When such a situation is detected - Reschedule napi
+ */
+static void check_for_empty_rx_ring(struct ena_adapter *adapter)
+{
+	struct ena_ring *rx_ring;
+	int i, refill_required;
+
+	if (!test_bit(ENA_FLAG_DEV_UP, &adapter->flags))
+		return;
+
+	if (test_bit(ENA_FLAG_TRIGGER_RESET, &adapter->flags))
+		return;
+
+	for (i = 0; i < adapter->num_io_queues; i++) {
+		rx_ring = &adapter->rx_ring[i];
+
+		refill_required = ena_com_free_q_entries(rx_ring->ena_com_io_sq);
+		if (unlikely(refill_required == (rx_ring->ring_size - 1))) {
+			rx_ring->empty_rx_queue++;
+
+			if (rx_ring->empty_rx_queue >= EMPTY_RX_REFILL) {
+				ena_increase_stat(&rx_ring->rx_stats.empty_rx_ring, 1,
+						  &rx_ring->syncp);
+
+				netif_err(adapter, drv, adapter->netdev,
+					  "Trigger refill for ring %d\n", i);
+
+				napi_schedule(rx_ring->napi);
+				rx_ring->empty_rx_queue = 0;
+			}
+		} else {
+			rx_ring->empty_rx_queue = 0;
+		}
+	}
 }
 
 /* Check for keep alive expiration */
@@ -2614,15 +3607,17 @@ static void check_for_missing_keep_alive(struct ena_adapter *adapter)
 	if (!adapter->wd_state)
 		return;
 
-	keep_alive_expired = round_jiffies(adapter->last_keep_alive_jiffies
-					   + ENA_DEVICE_KALIVE_TIMEOUT);
+	if (adapter->keep_alive_timeout == ENA_HW_HINTS_NO_TIMEOUT)
+		return;
+
+	keep_alive_expired = adapter->last_keep_alive_jiffies +
+			     adapter->keep_alive_timeout;
 	if (unlikely(time_is_before_jiffies(keep_alive_expired))) {
 		netif_err(adapter, drv, adapter->netdev,
 			  "Keep alive watchdog timeout.\n");
-		u64_stats_update_begin(&adapter->syncp);
-		adapter->dev_stats.wd_expired++;
-		u64_stats_update_end(&adapter->syncp);
-		set_bit(ENA_FLAG_TRIGGER_RESET, &adapter->flags);
+		ena_increase_stat(&adapter->dev_stats.wd_expired, 1,
+				  &adapter->syncp);
+		ena_reset_device(adapter, ENA_REGS_RESET_KEEP_ALIVE_TO);
 	}
 }
 
@@ -2631,10 +3626,47 @@ static void check_for_admin_com_state(struct ena_adapter *adapter)
 	if (unlikely(!ena_com_get_admin_running_state(adapter->ena_dev))) {
 		netif_err(adapter, drv, adapter->netdev,
 			  "ENA admin queue is not in running state!\n");
-		u64_stats_update_begin(&adapter->syncp);
-		adapter->dev_stats.admin_q_pause++;
-		u64_stats_update_end(&adapter->syncp);
-		set_bit(ENA_FLAG_TRIGGER_RESET, &adapter->flags);
+		ena_increase_stat(&adapter->dev_stats.admin_q_pause, 1,
+				  &adapter->syncp);
+		ena_reset_device(adapter, ENA_REGS_RESET_ADMIN_TO);
+	}
+}
+
+static void ena_update_hints(struct ena_adapter *adapter,
+			     struct ena_admin_ena_hw_hints *hints)
+{
+	struct net_device *netdev = adapter->netdev;
+
+	if (hints->admin_completion_tx_timeout)
+		adapter->ena_dev->admin_queue.completion_timeout =
+			hints->admin_completion_tx_timeout * 1000;
+
+	if (hints->mmio_read_timeout)
+		/* convert to usec */
+		adapter->ena_dev->mmio_read.reg_read_to =
+			hints->mmio_read_timeout * 1000;
+
+	if (hints->missed_tx_completion_count_threshold_to_reset)
+		adapter->missing_tx_completion_threshold =
+			hints->missed_tx_completion_count_threshold_to_reset;
+
+	if (hints->missing_tx_completion_timeout) {
+		if (hints->missing_tx_completion_timeout == ENA_HW_HINTS_NO_TIMEOUT)
+			adapter->missing_tx_completion_to = ENA_HW_HINTS_NO_TIMEOUT;
+		else
+			adapter->missing_tx_completion_to =
+				msecs_to_jiffies(hints->missing_tx_completion_timeout);
+	}
+
+	if (hints->netdev_wd_timeout)
+		netdev->watchdog_timeo = msecs_to_jiffies(hints->netdev_wd_timeout);
+
+	if (hints->driver_watchdog_timeout) {
+		if (hints->driver_watchdog_timeout == ENA_HW_HINTS_NO_TIMEOUT)
+			adapter->keep_alive_timeout = ENA_HW_HINTS_NO_TIMEOUT;
+		else
+			adapter->keep_alive_timeout =
+				msecs_to_jiffies(hints->driver_watchdog_timeout);
 	}
 }
 
@@ -2647,9 +3679,10 @@ static void ena_update_host_info(struct ena_admin_host_info *host_info,
 		(netdev->features & GENMASK_ULL(63, 32)) >> 32;
 }
 
-static void ena_timer_service(unsigned long data)
+static void ena_timer_service(struct timer_list *t)
 {
-	struct ena_adapter *adapter = (struct ena_adapter *)data;
+	struct ena_adapter *adapter = timer_container_of(adapter, t,
+							 timer_service);
 	u8 *debug_area = adapter->ena_dev->host_attr.debug_area_virt_addr;
 	struct ena_admin_host_info *host_info =
 		adapter->ena_dev->host_attr.host_info;
@@ -2658,7 +3691,9 @@ static void ena_timer_service(unsigned long data)
 
 	check_for_admin_com_state(adapter);
 
-	check_for_missing_tx_completions(adapter);
+	check_for_missing_completions(adapter);
+
+	check_for_empty_rx_ring(adapter);
 
 	if (debug_area)
 		ena_dump_stats_to_buf(adapter, debug_area);
@@ -2675,57 +3710,43 @@ static void ena_timer_service(unsigned long data)
 	}
 
 	/* Reset the timer */
-	mod_timer(&adapter->timer_service, jiffies + HZ);
+	mod_timer(&adapter->timer_service, round_jiffies(jiffies + HZ));
 }
 
-static int ena_calc_io_queue_num(struct pci_dev *pdev,
-				 struct ena_com_dev *ena_dev,
-				 struct ena_com_dev_get_features_ctx *get_feat_ctx)
+static u32 ena_calc_max_io_queue_num(struct pci_dev *pdev,
+				     struct ena_com_dev *ena_dev,
+				     struct ena_com_dev_get_features_ctx *get_feat_ctx)
 {
-	int io_sq_num, io_queue_num;
+	u32 io_tx_sq_num, io_tx_cq_num, io_rx_num, max_num_io_queues;
 
-	/* In case of LLQ use the llq number in the get feature cmd */
-	if (ena_dev->tx_mem_queue_type == ENA_ADMIN_PLACEMENT_POLICY_DEV) {
-		io_sq_num = get_feat_ctx->max_queues.max_llq_num;
+	if (ena_dev->supported_features & BIT(ENA_ADMIN_MAX_QUEUES_EXT)) {
+		struct ena_admin_queue_ext_feature_fields *max_queue_ext =
+			&get_feat_ctx->max_queue_ext.max_queue_ext;
+		io_rx_num = min_t(u32, max_queue_ext->max_rx_sq_num,
+				  max_queue_ext->max_rx_cq_num);
 
-		if (io_sq_num == 0) {
-			dev_err(&pdev->dev,
-				"Trying to use LLQ but llq_num is 0. Fall back into regular queues\n");
-
-			ena_dev->tx_mem_queue_type =
-				ENA_ADMIN_PLACEMENT_POLICY_HOST;
-			io_sq_num = get_feat_ctx->max_queues.max_sq_num;
-		}
+		io_tx_sq_num = max_queue_ext->max_tx_sq_num;
+		io_tx_cq_num = max_queue_ext->max_tx_cq_num;
 	} else {
-		io_sq_num = get_feat_ctx->max_queues.max_sq_num;
-	}
-
-	io_queue_num = min_t(int, num_online_cpus(), ENA_MAX_NUM_IO_QUEUES);
-	io_queue_num = min_t(int, io_queue_num, io_sq_num);
-	io_queue_num = min_t(int, io_queue_num,
-			     get_feat_ctx->max_queues.max_cq_num);
-	/* 1 IRQ for for mgmnt and 1 IRQs for each IO direction */
-	io_queue_num = min_t(int, io_queue_num, pci_msix_vec_count(pdev) - 1);
-	if (unlikely(!io_queue_num)) {
-		dev_err(&pdev->dev, "The device doesn't have io queues\n");
-		return -EFAULT;
+		struct ena_admin_queue_feature_desc *max_queues =
+			&get_feat_ctx->max_queues;
+		io_tx_sq_num = max_queues->max_sq_num;
+		io_tx_cq_num = max_queues->max_cq_num;
+		io_rx_num = min_t(u32, io_tx_sq_num, io_tx_cq_num);
 	}
 
-	return io_queue_num;
-}
-
-static void ena_set_push_mode(struct pci_dev *pdev, struct ena_com_dev *ena_dev,
-			      struct ena_com_dev_get_features_ctx *get_feat_ctx)
-{
-	bool has_mem_bar;
+	/* In case of LLQ use the llq fields for the tx SQ/CQ */
+	if (ena_dev->tx_mem_queue_type == ENA_ADMIN_PLACEMENT_POLICY_DEV)
+		io_tx_sq_num = get_feat_ctx->llq.max_llq_num;
 
-	has_mem_bar = pci_select_bars(pdev, IORESOURCE_MEM) & BIT(ENA_MEM_BAR);
+	max_num_io_queues = min_t(u32, num_online_cpus(), ENA_MAX_NUM_IO_QUEUES);
+	max_num_io_queues = min_t(u32, max_num_io_queues, io_rx_num);
+	max_num_io_queues = min_t(u32, max_num_io_queues, io_tx_sq_num);
+	max_num_io_queues = min_t(u32, max_num_io_queues, io_tx_cq_num);
+	/* 1 IRQ for mgmnt and 1 IRQs for each IO direction */
+	max_num_io_queues = min_t(u32, max_num_io_queues, pci_msix_vec_count(pdev) - 1);
 
-	/* Enable push mode if device supports LLQ */
-	if (has_mem_bar && (get_feat_ctx->max_queues.max_llq_num > 0))
-		ena_dev->tx_mem_queue_type = ENA_ADMIN_PLACEMENT_POLICY_DEV;
-	else
-		ena_dev->tx_mem_queue_type = ENA_ADMIN_PLACEMENT_POLICY_HOST;
+	return max_num_io_queues;
 }
 
 static void ena_set_dev_offloads(struct ena_com_dev_get_features_ctx *feat,
@@ -2780,7 +3801,7 @@ static void ena_set_conf_feat_params(struct ena_adapter *adapter,
 		ether_addr_copy(adapter->mac_addr, netdev->dev_addr);
 	} else {
 		ether_addr_copy(adapter->mac_addr, feat->dev_attr.mac_addr);
-		ether_addr_copy(netdev->dev_addr, adapter->mac_addr);
+		eth_hw_addr_set(netdev, adapter->mac_addr);
 	}
 
 	/* Set offload features */
@@ -2805,24 +3826,24 @@ static int ena_rss_init_default(struct ena_adapter *adapter)
 	}
 
 	for (i = 0; i < ENA_RX_RSS_TABLE_SIZE; i++) {
-		val = ethtool_rxfh_indir_default(i, adapter->num_queues);
+		val = ethtool_rxfh_indir_default(i, adapter->num_io_queues);
 		rc = ena_com_indirect_table_fill_entry(ena_dev, i,
 						       ENA_IO_RXQ_IDX(val));
-		if (unlikely(rc && (rc != -EPERM))) {
+		if (unlikely(rc)) {
 			dev_err(dev, "Cannot fill indirect table\n");
 			goto err_fill_indir;
 		}
 	}
 
-	rc = ena_com_fill_hash_function(ena_dev, ENA_ADMIN_CRC32, NULL,
-					ENA_HASH_KEY_SIZE, 0xFFFFFFFF);
-	if (unlikely(rc && (rc != -EPERM))) {
+	rc = ena_com_fill_hash_function(ena_dev, ENA_ADMIN_TOEPLITZ, NULL, ENA_HASH_KEY_SIZE,
+					0xFFFFFFFF);
+	if (unlikely(rc && (rc != -EOPNOTSUPP))) {
 		dev_err(dev, "Cannot fill hash function\n");
 		goto err_fill_indir;
 	}
 
 	rc = ena_com_set_default_hash_ctrl(ena_dev);
-	if (unlikely(rc && (rc != -EPERM))) {
+	if (unlikely(rc && (rc != -EOPNOTSUPP))) {
 		dev_err(dev, "Cannot fill hash control\n");
 		goto err_fill_indir;
 	}
@@ -2838,44 +3859,11 @@ err_rss_init:
 
 static void ena_release_bars(struct ena_com_dev *ena_dev, struct pci_dev *pdev)
 {
-	int release_bars;
+	int release_bars = pci_select_bars(pdev, IORESOURCE_MEM) & ENA_BAR_MASK;
 
-	release_bars = pci_select_bars(pdev, IORESOURCE_MEM) & ENA_BAR_MASK;
 	pci_release_selected_regions(pdev, release_bars);
 }
 
-static int ena_calc_queue_size(struct pci_dev *pdev,
-			       struct ena_com_dev *ena_dev,
-			       u16 *max_tx_sgl_size,
-			       u16 *max_rx_sgl_size,
-			       struct ena_com_dev_get_features_ctx *get_feat_ctx)
-{
-	u32 queue_size = ENA_DEFAULT_RING_SIZE;
-
-	queue_size = min_t(u32, queue_size,
-			   get_feat_ctx->max_queues.max_cq_depth);
-	queue_size = min_t(u32, queue_size,
-			   get_feat_ctx->max_queues.max_sq_depth);
-
-	if (ena_dev->tx_mem_queue_type == ENA_ADMIN_PLACEMENT_POLICY_DEV)
-		queue_size = min_t(u32, queue_size,
-				   get_feat_ctx->max_queues.max_llq_depth);
-
-	queue_size = rounddown_pow_of_two(queue_size);
-
-	if (unlikely(!queue_size)) {
-		dev_err(&pdev->dev, "Invalid queue size\n");
-		return -EFAULT;
-	}
-
-	*max_tx_sgl_size = min_t(u16, ENA_PKT_MAX_BUFS,
-				 get_feat_ctx->max_queues.max_packet_tx_descs);
-	*max_rx_sgl_size = min_t(u16, ENA_PKT_MAX_BUFS,
-				 get_feat_ctx->max_queues.max_packet_rx_descs);
-
-	return queue_size;
-}
-
 /* ena_probe - Device Initialization Routine
  * @pdev: PCI device information struct
  * @ent: entry in ena_pci_tbl
@@ -2889,28 +3877,29 @@ static int ena_calc_queue_size(struct pci_dev *pdev,
 static int ena_probe(struct pci_dev *pdev, const struct pci_device_id *ent)
 {
 	struct ena_com_dev_get_features_ctx get_feat_ctx;
-	static int version_printed;
-	struct net_device *netdev;
-	struct ena_adapter *adapter;
 	struct ena_com_dev *ena_dev = NULL;
+	struct ena_adapter *adapter;
+	struct net_device *netdev;
 	static int adapters_found;
-	int io_queue_num, bars, rc;
-	int queue_size;
-	u16 tx_sgl_size = 0;
-	u16 rx_sgl_size = 0;
+	struct devlink *devlink;
+	u32 max_num_io_queues;
 	bool wd_state;
+	int bars, rc;
 
 	dev_dbg(&pdev->dev, "%s\n", __func__);
 
-	if (version_printed++ == 0)
-		dev_info(&pdev->dev, "%s", version);
-
 	rc = pci_enable_device_mem(pdev);
 	if (rc) {
 		dev_err(&pdev->dev, "pci_enable_device_mem() failed!\n");
 		return rc;
 	}
 
+	rc = dma_set_mask_and_coherent(&pdev->dev, DMA_BIT_MASK(ENA_MAX_PHYS_ADDR_SIZE_BITS));
+	if (rc) {
+		dev_err(&pdev->dev, "dma_set_mask_and_coherent failed %d\n", rc);
+		goto err_disable_device;
+	}
+
 	pci_set_master(pdev);
 
 	ena_dev = vzalloc(sizeof(*ena_dev));
@@ -2927,81 +3916,100 @@ static int ena_probe(struct pci_dev *pdev, const struct pci_device_id *ent)
 		goto err_free_ena_dev;
 	}
 
-	ena_dev->reg_bar = ioremap(pci_resource_start(pdev, ENA_REG_BAR),
-				   pci_resource_len(pdev, ENA_REG_BAR));
+	ena_dev->reg_bar = devm_ioremap(&pdev->dev,
+					pci_resource_start(pdev, ENA_REG_BAR),
+					pci_resource_len(pdev, ENA_REG_BAR));
 	if (!ena_dev->reg_bar) {
-		dev_err(&pdev->dev, "failed to remap regs bar\n");
+		dev_err(&pdev->dev, "Failed to remap regs bar\n");
 		rc = -EFAULT;
 		goto err_free_region;
 	}
 
+	ena_dev->ena_min_poll_delay_us = ENA_ADMIN_POLL_DELAY_US;
+
 	ena_dev->dmadev = &pdev->dev;
 
-	rc = ena_device_init(ena_dev, pdev, &get_feat_ctx, &wd_state);
-	if (rc) {
-		dev_err(&pdev->dev, "ena device init failed\n");
-		if (rc == -ETIME)
-			rc = -EPROBE_DEFER;
+	netdev = alloc_etherdev_mq(sizeof(struct ena_adapter), ENA_MAX_RINGS);
+	if (!netdev) {
+		dev_err(&pdev->dev, "alloc_etherdev_mq failed\n");
+		rc = -ENOMEM;
 		goto err_free_region;
 	}
 
-	ena_set_push_mode(pdev, ena_dev, &get_feat_ctx);
+	SET_NETDEV_DEV(netdev, &pdev->dev);
+	adapter = netdev_priv(netdev);
+	adapter->ena_dev = ena_dev;
+	adapter->netdev = netdev;
+	adapter->pdev = pdev;
+	adapter->msg_enable = DEFAULT_MSG_ENABLE;
 
-	if (ena_dev->tx_mem_queue_type == ENA_ADMIN_PLACEMENT_POLICY_DEV) {
-		ena_dev->mem_bar = ioremap_wc(pci_resource_start(pdev, ENA_MEM_BAR),
-					      pci_resource_len(pdev, ENA_MEM_BAR));
-		if (!ena_dev->mem_bar) {
-			rc = -EFAULT;
-			goto err_device_destroy;
-		}
+	ena_dev->net_device = netdev;
+
+	pci_set_drvdata(pdev, adapter);
+
+	rc = ena_phc_alloc(adapter);
+	if (rc) {
+		netdev_err(netdev, "ena_phc_alloc failed\n");
+		goto err_netdev_destroy;
 	}
 
-	/* initial Tx interrupt delay, Assumes 1 usec granularity.
-	* Updated during device initialization with the real granularity
-	*/
-	ena_dev->intr_moder_tx_interval = ENA_INTR_INITIAL_TX_INTERVAL_USECS;
-	io_queue_num = ena_calc_io_queue_num(pdev, ena_dev, &get_feat_ctx);
-	queue_size = ena_calc_queue_size(pdev, ena_dev, &tx_sgl_size,
-					 &rx_sgl_size, &get_feat_ctx);
-	if ((queue_size <= 0) || (io_queue_num <= 0)) {
-		rc = -EFAULT;
-		goto err_device_destroy;
+	rc = ena_com_allocate_customer_metrics_buffer(ena_dev);
+	if (rc) {
+		netdev_err(netdev, "ena_com_allocate_customer_metrics_buffer failed\n");
+		goto err_free_phc;
 	}
 
-	dev_info(&pdev->dev, "creating %d io queues. queue size: %d\n",
-		 io_queue_num, queue_size);
+	rc = ena_map_llq_mem_bar(pdev, ena_dev, bars);
+	if (rc) {
+		dev_err(&pdev->dev, "ENA LLQ bar mapping failed\n");
+		goto err_metrics_destroy;
+	}
 
-	/* dev zeroed in init_etherdev */
-	netdev = alloc_etherdev_mq(sizeof(struct ena_adapter), io_queue_num);
-	if (!netdev) {
-		dev_err(&pdev->dev, "alloc_etherdev_mq failed\n");
+	/* Need to do this before ena_device_init */
+	devlink = ena_devlink_alloc(adapter);
+	if (!devlink) {
+		netdev_err(netdev, "ena_devlink_alloc failed\n");
 		rc = -ENOMEM;
-		goto err_device_destroy;
+		goto err_metrics_destroy;
 	}
 
-	SET_NETDEV_DEV(netdev, &pdev->dev);
-
-	adapter = netdev_priv(netdev);
-	pci_set_drvdata(pdev, adapter);
+	rc = ena_device_init(adapter, pdev, &get_feat_ctx, &wd_state);
+	if (rc) {
+		dev_err(&pdev->dev, "ENA device init failed\n");
+		if (rc == -ETIME)
+			rc = -EPROBE_DEFER;
+		goto ena_devlink_destroy;
+	}
 
-	adapter->ena_dev = ena_dev;
-	adapter->netdev = netdev;
-	adapter->pdev = pdev;
+	/* Initial TX and RX interrupt delay. Assumes 1 usec granularity.
+	 * Updated during device initialization with the real granularity
+	 */
+	ena_dev->intr_moder_tx_interval = ENA_INTR_INITIAL_TX_INTERVAL_USECS;
+	ena_dev->intr_moder_rx_interval = ENA_INTR_INITIAL_RX_INTERVAL_USECS;
+	ena_dev->intr_delay_resolution = ENA_DEFAULT_INTR_DELAY_RESOLUTION;
+	max_num_io_queues = ena_calc_max_io_queue_num(pdev, ena_dev, &get_feat_ctx);
+	if (unlikely(!max_num_io_queues)) {
+		rc = -EFAULT;
+		goto err_device_destroy;
+	}
 
 	ena_set_conf_feat_params(adapter, &get_feat_ctx);
 
-	adapter->msg_enable = netif_msg_init(debug, DEFAULT_MSG_ENABLE);
-
-	adapter->tx_ring_size = queue_size;
-	adapter->rx_ring_size = queue_size;
+	adapter->reset_reason = ENA_REGS_RESET_NORMAL;
 
-	adapter->max_tx_sgl_size = tx_sgl_size;
-	adapter->max_rx_sgl_size = rx_sgl_size;
-
-	adapter->num_queues = io_queue_num;
+	adapter->num_io_queues = max_num_io_queues;
+	adapter->max_num_io_queues = max_num_io_queues;
 	adapter->last_monitored_tx_qid = 0;
 
+	adapter->xdp_first_ring = 0;
+	adapter->xdp_num_queues = 0;
+
 	adapter->rx_copybreak = ENA_DEFAULT_RX_COPYBREAK;
+	if (ena_dev->tx_mem_queue_type == ENA_ADMIN_PLACEMENT_POLICY_DEV)
+		adapter->disable_meta_caching =
+			!!(get_feat_ctx.llq.accel_mode.u.get.supported_flags &
+			   BIT(ENA_ADMIN_DISABLE_META_CACHING));
+
 	adapter->wd_state = wd_state;
 
 	snprintf(adapter->name, ENA_NAME_MAX_LEN, "ena_%d", adapters_found);
@@ -3010,9 +4018,13 @@ static int ena_probe(struct pci_dev *pdev, const struct pci_device_id *ent)
 	if (rc) {
 		dev_err(&pdev->dev,
 			"Failed to query interrupt moderation feature\n");
-		goto err_netdev_destroy;
+		goto err_device_destroy;
 	}
-	ena_init_io_rings(adapter);
+
+	ena_init_io_rings(adapter,
+			  0,
+			  adapter->xdp_num_queues +
+			  adapter->num_io_queues);
 
 	netdev->netdev_ops = &ena_netdev_ops;
 	netdev->watchdog_timeo = TX_TIMEOUT;
@@ -3022,20 +4034,24 @@ static int ena_probe(struct pci_dev *pdev, const struct pci_device_id *ent)
 
 	u64_stats_init(&adapter->syncp);
 
-	rc = ena_enable_msix_and_set_admin_interrupts(adapter, io_queue_num);
+	rc = ena_enable_msix_and_set_admin_interrupts(adapter);
 	if (rc) {
 		dev_err(&pdev->dev,
 			"Failed to enable and set the admin interrupts\n");
 		goto err_worker_destroy;
 	}
 	rc = ena_rss_init_default(adapter);
-	if (rc && (rc != -EPERM)) {
+	if (rc && (rc != -EOPNOTSUPP)) {
 		dev_err(&pdev->dev, "Cannot init RSS rc: %d\n", rc);
 		goto err_free_msix;
 	}
 
 	ena_config_debug_area(adapter);
 
+	if (ena_xdp_legal_queue_count(adapter, adapter->num_io_queues))
+		netdev->xdp_features = NETDEV_XDP_ACT_BASIC |
+				       NETDEV_XDP_ACT_REDIRECT;
+
 	memcpy(adapter->netdev->perm_addr, adapter->mac_addr, netdev->addr_len);
 
 	netif_carrier_off(netdev);
@@ -3046,43 +4062,59 @@ static int ena_probe(struct pci_dev *pdev, const struct pci_device_id *ent)
 		goto err_rss;
 	}
 
-	INIT_WORK(&adapter->suspend_io_task, ena_device_io_suspend);
-	INIT_WORK(&adapter->resume_io_task, ena_device_io_resume);
+	ena_debugfs_init(netdev);
+
 	INIT_WORK(&adapter->reset_task, ena_fw_reset_device);
 
 	adapter->last_keep_alive_jiffies = jiffies;
+	adapter->keep_alive_timeout = ENA_DEVICE_KALIVE_TIMEOUT;
+	adapter->missing_tx_completion_to = TX_TIMEOUT;
+	adapter->missing_tx_completion_threshold = MAX_NUM_OF_TIMEOUTED_PACKETS;
+
+	ena_update_hints(adapter, &get_feat_ctx.hw_hints);
 
-	setup_timer(&adapter->timer_service, ena_timer_service,
-		    (unsigned long)adapter);
+	timer_setup(&adapter->timer_service, ena_timer_service, 0);
 	mod_timer(&adapter->timer_service, round_jiffies(jiffies + HZ));
 
-	dev_info(&pdev->dev, "%s found at mem %lx, mac addr %pM Queues %d\n",
+	dev_info(&pdev->dev,
+		 "%s found at mem %lx, mac addr %pM\n",
 		 DEVICE_NAME, (long)pci_resource_start(pdev, 0),
-		 netdev->dev_addr, io_queue_num);
+		 netdev->dev_addr);
 
 	set_bit(ENA_FLAG_DEVICE_RUNNING, &adapter->flags);
 
 	adapters_found++;
 
+	/* From this point, the devlink device is visible to users.
+	 * Perform the registration last to ensure that all the resources
+	 * are available and that the netdevice is registered.
+	 */
+	ena_devlink_register(devlink, &pdev->dev);
+
 	return 0;
 
 err_rss:
 	ena_com_delete_debug_area(ena_dev);
 	ena_com_rss_destroy(ena_dev);
 err_free_msix:
-	ena_com_dev_reset(ena_dev);
+	ena_com_dev_reset(ena_dev, ENA_REGS_RESET_INIT_ERR);
+	/* stop submitting admin commands on a device that was reset */
+	ena_com_set_admin_running_state(ena_dev, false);
 	ena_free_mgmnt_irq(adapter);
-	pci_free_irq_vectors(adapter->pdev);
+	ena_disable_msix(adapter);
 err_worker_destroy:
-	ena_com_destroy_interrupt_moderation(ena_dev);
-	del_timer(&adapter->timer_service);
-	cancel_work_sync(&adapter->suspend_io_task);
-	cancel_work_sync(&adapter->resume_io_task);
-err_netdev_destroy:
-	free_netdev(netdev);
+	timer_delete(&adapter->timer_service);
 err_device_destroy:
 	ena_com_delete_host_info(ena_dev);
 	ena_com_admin_destroy(ena_dev);
+ena_devlink_destroy:
+	ena_devlink_free(devlink);
+err_metrics_destroy:
+	ena_com_delete_customer_metrics_buffer(ena_dev);
+err_free_phc:
+	ena_phc_free(adapter);
+err_netdev_destroy:
+	free_netdev(netdev);
 err_free_region:
 	ena_release_bars(ena_dev, pdev);
 err_free_ena_dev:
@@ -3093,40 +4125,16 @@ err_disable_device:
 }
 
 /*****************************************************************************/
-static int ena_sriov_configure(struct pci_dev *dev, int numvfs)
-{
-	int rc;
-
-	if (numvfs > 0) {
-		rc = pci_enable_sriov(dev, numvfs);
-		if (rc != 0) {
-			dev_err(&dev->dev,
-				"pci_enable_sriov failed to enable: %d vfs with the error: %d\n",
-				numvfs, rc);
-			return rc;
-		}
-
-		return numvfs;
-	}
-
-	if (numvfs == 0) {
-		pci_disable_sriov(dev);
-		return 0;
-	}
-
-	return -EINVAL;
-}
-
-/*****************************************************************************/
-/*****************************************************************************/
 
-/* ena_remove - Device Removal Routine
+/* __ena_shutoff - Helper used in both PCI remove/shutdown routines
  * @pdev: PCI device information struct
+ * @shutdown: Is it a shutdown operation? If false, means it is a removal
  *
- * ena_remove is called by the PCI subsystem to alert the driver
- * that it should release a PCI device.
+ * __ena_shutoff is a helper routine that does the real work on shutdown and
+ * removal paths; the difference between those paths is with regards to whether
+ * dettach or unregister the netdevice.
  */
-static void ena_remove(struct pci_dev *pdev)
+static void __ena_shutoff(struct pci_dev *pdev, bool shutdown)
 {
 	struct ena_adapter *adapter = pci_get_drvdata(pdev);
 	struct ena_com_dev *ena_dev;
@@ -3135,66 +4143,124 @@ static void ena_remove(struct pci_dev *pdev)
 	ena_dev = adapter->ena_dev;
 	netdev = adapter->netdev;
 
-#ifdef CONFIG_RFS_ACCEL
-	if ((adapter->msix_vecs >= 1) && (netdev->rx_cpu_rmap)) {
-		free_irq_cpu_rmap(netdev->rx_cpu_rmap);
-		netdev->rx_cpu_rmap = NULL;
-	}
-#endif /* CONFIG_RFS_ACCEL */
-
-	unregister_netdev(netdev);
-	del_timer_sync(&adapter->timer_service);
+	ena_debugfs_terminate(netdev);
 
+	/* Make sure timer and reset routine won't be called after
+	 * freeing device resources.
+	 */
+	timer_delete_sync(&adapter->timer_service);
 	cancel_work_sync(&adapter->reset_task);
 
-	cancel_work_sync(&adapter->suspend_io_task);
+	rtnl_lock(); /* lock released inside the below if-else block */
+	adapter->reset_reason = ENA_REGS_RESET_SHUTDOWN;
+	ena_destroy_device(adapter, true);
 
-	cancel_work_sync(&adapter->resume_io_task);
+	ena_phc_free(adapter);
 
-	/* Reset the device only if the device is running. */
-	if (test_bit(ENA_FLAG_DEVICE_RUNNING, &adapter->flags))
-		ena_com_dev_reset(ena_dev);
+	ena_devlink_unregister(adapter->devlink);
+	ena_devlink_free(adapter->devlink);
 
-	ena_free_mgmnt_irq(adapter);
+	if (shutdown) {
+		netif_device_detach(netdev);
+		dev_close(netdev);
+		rtnl_unlock();
+	} else {
+		rtnl_unlock();
+		unregister_netdev(netdev);
+		free_netdev(netdev);
+	}
+
+	ena_com_rss_destroy(ena_dev);
 
-	pci_free_irq_vectors(adapter->pdev);
+	ena_com_delete_debug_area(ena_dev);
 
-	free_netdev(netdev);
+	ena_com_delete_host_info(ena_dev);
 
-	ena_com_mmio_reg_read_request_destroy(ena_dev);
+	ena_com_delete_customer_metrics_buffer(ena_dev);
 
-	ena_com_abort_admin_commands(ena_dev);
+	ena_release_bars(ena_dev, pdev);
 
-	ena_com_wait_for_abort_completion(ena_dev);
+	pci_disable_device(pdev);
 
-	ena_com_admin_destroy(ena_dev);
+	vfree(ena_dev);
+}
 
-	ena_com_rss_destroy(ena_dev);
+/* ena_remove - Device Removal Routine
+ * @pdev: PCI device information struct
+ *
+ * ena_remove is called by the PCI subsystem to alert the driver
+ * that it should release a PCI device.
+ */
 
-	ena_com_delete_debug_area(ena_dev);
+static void ena_remove(struct pci_dev *pdev)
+{
+	__ena_shutoff(pdev, false);
+}
 
-	ena_com_delete_host_info(ena_dev);
+/* ena_shutdown - Device Shutdown Routine
+ * @pdev: PCI device information struct
+ *
+ * ena_shutdown is called by the PCI subsystem to alert the driver that
+ * a shutdown/reboot (or kexec) is happening and device must be disabled.
+ */
 
-	ena_release_bars(ena_dev, pdev);
+static void ena_shutdown(struct pci_dev *pdev)
+{
+	__ena_shutoff(pdev, true);
+}
 
-	pci_disable_device(pdev);
+/* ena_suspend - PM suspend callback
+ * @dev_d: Device information struct
+ */
+static int __maybe_unused ena_suspend(struct device *dev_d)
+{
+	struct pci_dev *pdev = to_pci_dev(dev_d);
+	struct ena_adapter *adapter = pci_get_drvdata(pdev);
 
-	ena_com_destroy_interrupt_moderation(ena_dev);
+	ena_increase_stat(&adapter->dev_stats.suspend, 1, &adapter->syncp);
 
-	vfree(ena_dev);
+	rtnl_lock();
+	if (unlikely(test_bit(ENA_FLAG_TRIGGER_RESET, &adapter->flags))) {
+		dev_err(&pdev->dev,
+			"Ignoring device reset request as the device is being suspended\n");
+		clear_bit(ENA_FLAG_TRIGGER_RESET, &adapter->flags);
+	}
+	ena_destroy_device(adapter, true);
+	rtnl_unlock();
+	return 0;
+}
+
+/* ena_resume - PM resume callback
+ * @dev_d: Device information struct
+ */
+static int __maybe_unused ena_resume(struct device *dev_d)
+{
+	struct ena_adapter *adapter = dev_get_drvdata(dev_d);
+	int rc;
+
+	ena_increase_stat(&adapter->dev_stats.resume, 1, &adapter->syncp);
+
+	rtnl_lock();
+	rc = ena_restore_device(adapter);
+	rtnl_unlock();
+	return rc;
 }
 
+static SIMPLE_DEV_PM_OPS(ena_pm_ops, ena_suspend, ena_resume);
+
 static struct pci_driver ena_pci_driver = {
 	.name		= DRV_MODULE_NAME,
 	.id_table	= ena_pci_tbl,
 	.probe		= ena_probe,
 	.remove		= ena_remove,
-	.sriov_configure = ena_sriov_configure,
+	.shutdown	= ena_shutdown,
+	.driver.pm	= &ena_pm_ops,
+	.sriov_configure = pci_sriov_configure_simple,
 };
 
 static int __init ena_init(void)
 {
-	pr_info("%s", version);
+	int ret;
 
 	ena_wq = create_singlethread_workqueue(DRV_MODULE_NAME);
 	if (!ena_wq) {
@@ -3202,7 +4268,11 @@ static int __init ena_init(void)
 		return -ENOMEM;
 	}
 
-	return pci_register_driver(&ena_pci_driver);
+	ret = pci_register_driver(&ena_pci_driver);
+	if (ret)
+		destroy_workqueue(ena_wq);
+
+	return ret;
 }
 
 static void __exit ena_cleanup(void)
@@ -3231,9 +4301,10 @@ static void ena_update_on_link_change(void *adapter_data,
 		ENA_ADMIN_AENQ_LINK_CHANGE_DESC_LINK_STATUS_MASK;
 
 	if (status) {
-		netdev_dbg(adapter->netdev, "%s\n", __func__);
+		netif_dbg(adapter, ifup, adapter->netdev, "%s\n", __func__);
 		set_bit(ENA_FLAG_LINK_UP, &adapter->flags);
-		netif_carrier_on(adapter->netdev);
+		if (!test_bit(ENA_FLAG_ONGOING_RESET, &adapter->flags))
+			netif_carrier_on(adapter->netdev);
 	} else {
 		clear_bit(ENA_FLAG_LINK_UP, &adapter->flags);
 		netif_carrier_off(adapter->netdev);
@@ -3244,35 +4315,46 @@ static void ena_keep_alive_wd(void *adapter_data,
 			      struct ena_admin_aenq_entry *aenq_e)
 {
 	struct ena_adapter *adapter = (struct ena_adapter *)adapter_data;
+	struct ena_admin_aenq_keep_alive_desc *desc;
+	u64 rx_drops;
+	u64 tx_drops;
 
+	desc = (struct ena_admin_aenq_keep_alive_desc *)aenq_e;
 	adapter->last_keep_alive_jiffies = jiffies;
+
+	rx_drops = ((u64)desc->rx_drops_high << 32) | desc->rx_drops_low;
+	tx_drops = ((u64)desc->tx_drops_high << 32) | desc->tx_drops_low;
+
+	u64_stats_update_begin(&adapter->syncp);
+	/* These stats are accumulated by the device, so the counters indicate
+	 * all drops since last reset.
+	 */
+	adapter->dev_stats.rx_drops = rx_drops;
+	adapter->dev_stats.tx_drops = tx_drops;
+	u64_stats_update_end(&adapter->syncp);
 }
 
 static void ena_notification(void *adapter_data,
 			     struct ena_admin_aenq_entry *aenq_e)
 {
 	struct ena_adapter *adapter = (struct ena_adapter *)adapter_data;
+	struct ena_admin_ena_hw_hints *hints;
 
 	WARN(aenq_e->aenq_common_desc.group != ENA_ADMIN_NOTIFICATION,
 	     "Invalid group(%x) expected %x\n",
 	     aenq_e->aenq_common_desc.group,
 	     ENA_ADMIN_NOTIFICATION);
 
-	switch (aenq_e->aenq_common_desc.syndrom) {
-	case ENA_ADMIN_SUSPEND:
-		/* Suspend just the IO queues.
-		 * We deliberately don't suspend admin so the timer and
-		 * the keep_alive events should remain.
-		 */
-		queue_work(ena_wq, &adapter->suspend_io_task);
-		break;
-	case ENA_ADMIN_RESUME:
-		queue_work(ena_wq, &adapter->resume_io_task);
+	switch (aenq_e->aenq_common_desc.syndrome) {
+	case ENA_ADMIN_UPDATE_HINTS:
+		hints = (struct ena_admin_ena_hw_hints *)
+			(&aenq_e->inline_data_w4);
+		ena_update_hints(adapter, hints);
 		break;
 	default:
 		netif_err(adapter, drv, adapter->netdev,
 			  "Invalid aenq notification link state %d\n",
-			  aenq_e->aenq_common_desc.syndrom);
+			  aenq_e->aenq_common_desc.syndrome);
 	}
 }
 
diff --git a/drivers/net/ethernet/amazon/ena/ena_netdev.h b/drivers/net/ethernet/amazon/ena/ena_netdev.h
index 0e22bce6239d..006f9a3acea6 100644
--- a/drivers/net/ethernet/amazon/ena/ena_netdev.h
+++ b/drivers/net/ethernet/amazon/ena/ena_netdev.h
@@ -1,79 +1,64 @@
+/* SPDX-License-Identifier: GPL-2.0 OR Linux-OpenIB */
 /*
- * Copyright 2015 Amazon.com, Inc. or its affiliates.
- *
- * This software is available to you under a choice of one of two
- * licenses.  You may choose to be licensed under the terms of the GNU
- * General Public License (GPL) Version 2, available from the file
- * COPYING in the main directory of this source tree, or the
- * BSD license below:
- *
- *     Redistribution and use in source and binary forms, with or
- *     without modification, are permitted provided that the following
- *     conditions are met:
- *
- *      - Redistributions of source code must retain the above
- *        copyright notice, this list of conditions and the following
- *        disclaimer.
- *
- *      - Redistributions in binary form must reproduce the above
- *        copyright notice, this list of conditions and the following
- *        disclaimer in the documentation and/or other materials
- *        provided with the distribution.
- *
- * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND,
- * EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF
- * MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND
- * NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS
- * BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN
- * ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN
- * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
- * SOFTWARE.
+ * Copyright 2015-2020 Amazon.com, Inc. or its affiliates. All rights reserved.
  */
 
 #ifndef ENA_H
 #define ENA_H
 
 #include <linux/bitops.h>
+#include <linux/dim.h>
 #include <linux/etherdevice.h>
+#include <linux/if_vlan.h>
 #include <linux/inetdevice.h>
 #include <linux/interrupt.h>
 #include <linux/netdevice.h>
 #include <linux/skbuff.h>
+#include <net/xdp.h>
+#include <uapi/linux/bpf.h>
+#include <net/devlink.h>
 
 #include "ena_com.h"
 #include "ena_eth_com.h"
 
-#define DRV_MODULE_VER_MAJOR	1
-#define DRV_MODULE_VER_MINOR	1
-#define DRV_MODULE_VER_SUBMINOR 2
+#define DRV_MODULE_GEN_MAJOR	2
+#define DRV_MODULE_GEN_MINOR	1
+#define DRV_MODULE_GEN_SUBMINOR 0
 
 #define DRV_MODULE_NAME		"ena"
-#ifndef DRV_MODULE_VERSION
-#define DRV_MODULE_VERSION \
-	__stringify(DRV_MODULE_VER_MAJOR) "."	\
-	__stringify(DRV_MODULE_VER_MINOR) "."	\
-	__stringify(DRV_MODULE_VER_SUBMINOR)
-#endif
 
 #define DEVICE_NAME	"Elastic Network Adapter (ENA)"
 
 /* 1 for AENQ + ADMIN */
-#define ENA_MAX_MSIX_VEC(io_queues)	(1 + (io_queues))
+#define ENA_ADMIN_MSIX_VEC		1
+#define ENA_MAX_MSIX_VEC(io_queues)	(ENA_ADMIN_MSIX_VEC + (io_queues))
+
+/* The ENA buffer length fields is 16 bit long. So when PAGE_SIZE == 64kB the
+ * driver passes 0.
+ * Since the max packet size the ENA handles is ~9kB limit the buffer length to
+ * 16kB.
+ */
+#if PAGE_SIZE > SZ_16K
+#define ENA_PAGE_SIZE (_AC(SZ_16K, UL))
+#else
+#define ENA_PAGE_SIZE PAGE_SIZE
+#endif
+
+#define ENA_MIN_MSIX_VEC		2
 
 #define ENA_REG_BAR			0
 #define ENA_MEM_BAR			2
 #define ENA_BAR_MASK (BIT(ENA_REG_BAR) | BIT(ENA_MEM_BAR))
 
 #define ENA_DEFAULT_RING_SIZE	(1024)
+#define ENA_MIN_RING_SIZE	(256)
 
-#define ENA_TX_WAKEUP_THRESH		(MAX_SKB_FRAGS + 2)
-#define ENA_DEFAULT_RX_COPYBREAK	(128 - NET_IP_ALIGN)
+#define ENA_MIN_RX_BUF_SIZE (2048)
 
-/* limit the buffer size to 600 bytes to handle MTU changes from very
- * small to very large, in which case the number of buffers per packet
- * could exceed ENA_PKT_MAX_BUFS
- */
-#define ENA_DEFAULT_MIN_RX_BUFF_ALLOC_SIZE 600
+#define ENA_MIN_NUM_IO_QUEUES	(1)
+
+#define ENA_TX_WAKEUP_THRESH		(MAX_SKB_FRAGS + 2)
+#define ENA_DEFAULT_RX_COPYBREAK	(256 - NET_IP_ALIGN)
 
 #define ENA_MIN_MTU		128
 
@@ -85,17 +70,16 @@
 #define ENA_RX_RSS_TABLE_LOG_SIZE  7
 #define ENA_RX_RSS_TABLE_SIZE	(1 << ENA_RX_RSS_TABLE_LOG_SIZE)
 
-#define ENA_HASH_KEY_SIZE	40
-
 /* The number of tx packet completions that will be handled each NAPI poll
  * cycle is ring_size / ENA_TX_POLL_BUDGET_DIVIDER.
  */
 #define ENA_TX_POLL_BUDGET_DIVIDER	4
 
-/* Refill Rx queue when number of available descriptors is below
- * QUEUE_SIZE / ENA_RX_REFILL_THRESH_DIVIDER
+/* Refill Rx queue when number of required descriptors is above
+ * QUEUE_SIZE / ENA_RX_REFILL_THRESH_DIVIDER or ENA_RX_REFILL_THRESH_PACKET
  */
 #define ENA_RX_REFILL_THRESH_DIVIDER	8
+#define ENA_RX_REFILL_THRESH_PACKET	256
 
 /* Number of queues to check for missing queues per timer service */
 #define ENA_MONITORED_TX_QUEUES	4
@@ -110,18 +94,25 @@
 
 #define ENA_IO_TXQ_IDX(q)	(2 * (q))
 #define ENA_IO_RXQ_IDX(q)	(2 * (q) + 1)
+#define ENA_IO_TXQ_IDX_TO_COMBINED_IDX(q)	((q) / 2)
+#define ENA_IO_RXQ_IDX_TO_COMBINED_IDX(q)	(((q) - 1) / 2)
 
 #define ENA_MGMNT_IRQ_IDX		0
 #define ENA_IO_IRQ_FIRST_IDX		1
 #define ENA_IO_IRQ_IDX(q)		(ENA_IO_IRQ_FIRST_IDX + (q))
 
+#define ENA_ADMIN_POLL_DELAY_US 100
+
 /* ENA device should send keep alive msg every 1 sec.
  * We wait for 6 sec just to be on the safe side.
  */
 #define ENA_DEVICE_KALIVE_TIMEOUT	(6 * HZ)
+#define ENA_MAX_NO_INTERRUPT_ITERATIONS 3
 
 #define ENA_MMIO_DISABLE_REG_READ	BIT(0)
 
+struct ena_phc_info;
+
 struct ena_irq {
 	irq_handler_t handler;
 	void *data;
@@ -132,21 +123,47 @@ struct ena_irq {
 };
 
 struct ena_napi {
-	struct napi_struct napi ____cacheline_aligned;
+	u8 first_interrupt ____cacheline_aligned;
+	u8 interrupts_masked;
+	struct napi_struct napi;
 	struct ena_ring *tx_ring;
 	struct ena_ring *rx_ring;
 	u32 qid;
+	struct dim dim;
 };
 
 struct ena_tx_buffer {
-	struct sk_buff *skb;
+	union {
+		struct sk_buff *skb;
+		/* XDP buffer structure which is used for sending packets in
+		 * the xdp queues
+		 */
+		struct xdp_frame *xdpf;
+	};
 	/* num of ena desc for this specific skb
 	 * (includes data desc and metadata desc)
 	 */
 	u32 tx_descs;
 	/* num of buffers used by this skb */
 	u32 num_of_bufs;
-	/* Save the last jiffies to detect missing tx packets */
+
+	/* Total size of all buffers in bytes */
+	u32 total_tx_size;
+
+	/* Indicate if bufs[0] map the linear data of the skb. */
+	u8 map_linear_data;
+
+	/* Used for detect missing tx packets to limit the number of prints */
+	u8 print_once;
+	/* Save the last jiffies to detect missing tx packets
+	 *
+	 * sets to non zero value on ena_start_xmit and set to zero on
+	 * napi and timer_Service_routine.
+	 *
+	 * while this value is not protected by lock,
+	 * a given packet is not expected to be handled by ena_start_xmit
+	 * and by napi/timer_service at the same time.
+	 */
 	unsigned long last_jiffies;
 	struct ena_com_buf bufs[ENA_PKT_MAX_BUFS];
 } ____cacheline_aligned;
@@ -154,7 +171,9 @@ struct ena_tx_buffer {
 struct ena_rx_buffer {
 	struct sk_buff *skb;
 	struct page *page;
+	dma_addr_t dma_addr;
 	u32 page_offset;
+	u32 buf_offset;
 	struct ena_com_buf ena_buf;
 } ____cacheline_aligned;
 
@@ -170,25 +189,41 @@ struct ena_stats_tx {
 	u64 napi_comp;
 	u64 tx_poll;
 	u64 doorbells;
-	u64 missing_tx_comp;
 	u64 bad_req_id;
+	u64 llq_buffer_copy;
+	u64 missed_tx;
+	u64 unmask_interrupt;
+	u64 last_napi_jiffies;
 };
 
 struct ena_stats_rx {
 	u64 cnt;
 	u64 bytes;
+	u64 rx_copybreak_pkt;
+	u64 csum_good;
 	u64 refil_partial;
-	u64 bad_csum;
+	u64 csum_bad;
 	u64 page_alloc_fail;
 	u64 skb_alloc_fail;
 	u64 dma_mapping_err;
 	u64 bad_desc_num;
-	u64 rx_copybreak_pkt;
+	u64 bad_req_id;
+	u64 empty_rx_ring;
+	u64 csum_unchecked;
+	u64 xdp_aborted;
+	u64 xdp_drop;
+	u64 xdp_pass;
+	u64 xdp_tx;
+	u64 xdp_invalid;
+	u64 xdp_redirect;
 };
 
 struct ena_ring {
-	/* Holds the empty requests for TX out of order completions */
-	u16 *free_tx_ids;
+	/* Holds the empty requests for TX/RX
+	 * out of order completions
+	 */
+	u16 *free_ids;
+
 	union {
 		struct ena_tx_buffer *tx_buffer_info;
 		struct ena_rx_buffer *rx_buffer_info;
@@ -203,10 +238,18 @@ struct ena_ring {
 	struct ena_adapter *adapter;
 	struct ena_com_io_cq *ena_com_io_cq;
 	struct ena_com_io_sq *ena_com_io_sq;
+	struct bpf_prog *xdp_bpf_prog;
+	struct xdp_rxq_info xdp_rxq;
+	spinlock_t xdp_tx_lock;	/* synchronize XDP TX/Redirect traffic */
+	/* Used for rx queues only to point to the xdp tx ring, to
+	 * which traffic should be redirected from this rx ring.
+	 */
+	struct ena_ring *xdp_ring;
 
 	u16 next_to_use;
 	u16 next_to_clean;
 	u16 rx_copybreak;
+	u16 rx_headroom;
 	u16 qid;
 	u16 mtu;
 	u16 sgl_size;
@@ -214,9 +257,14 @@ struct ena_ring {
 	/* The maximum header length the device can handle */
 	u8 tx_max_header_size;
 
-	/* cpu for TPH */
+	bool disable_meta_caching;
+	u16 no_interrupt_event_cnt;
+
+	/* cpu and NUMA for TPH */
 	int cpu;
-	 /* number of tx/rx_buffer_info's entries */
+	int numa_node;
+
+	/* number of tx/rx_buffer_info's entries */
 	int ring_size;
 
 	enum ena_admin_placement_policy_type tx_mem_queue_type;
@@ -224,31 +272,37 @@ struct ena_ring {
 	struct ena_com_rx_buf_info ena_bufs[ENA_PKT_MAX_BUFS];
 	u32  smoothed_interval;
 	u32  per_napi_packets;
-	u32  per_napi_bytes;
-	enum ena_intr_moder_level moder_tbl_idx;
+	u16 non_empty_napi_events;
 	struct u64_stats_sync syncp;
 	union {
 		struct ena_stats_tx tx_stats;
 		struct ena_stats_rx rx_stats;
 	};
+
+	u8 *push_buf_intermediate_buf;
+	int empty_rx_queue;
 } ____cacheline_aligned;
 
 struct ena_stats_dev {
 	u64 tx_timeout;
-	u64 io_suspend;
-	u64 io_resume;
+	u64 suspend;
+	u64 resume;
 	u64 wd_expired;
 	u64 interface_up;
 	u64 interface_down;
 	u64 admin_q_pause;
 	u64 rx_drops;
+	u64 tx_drops;
+	u64 reset_fail;
 };
 
 enum ena_flags_t {
 	ENA_FLAG_DEVICE_RUNNING,
 	ENA_FLAG_DEV_UP,
 	ENA_FLAG_LINK_UP,
-	ENA_FLAG_TRIGGER_RESET
+	ENA_FLAG_MSIX_ENABLED,
+	ENA_FLAG_TRIGGER_RESET,
+	ENA_FLAG_ONGOING_RESET
 };
 
 /* adapter specific private data structure */
@@ -264,25 +318,41 @@ struct ena_adapter {
 	u32 rx_copybreak;
 	u32 max_mtu;
 
-	int num_queues;
+	u32 num_io_queues;
+	u32 max_num_io_queues;
 
 	int msix_vecs;
 
-	u32 tx_usecs, rx_usecs; /* interrupt moderation */
-	u32 tx_frames, rx_frames; /* interrupt moderation */
+	u32 missing_tx_completion_threshold;
+
+	u32 requested_tx_ring_size;
+	u32 requested_rx_ring_size;
 
-	u32 tx_ring_size;
-	u32 rx_ring_size;
+	u32 max_tx_ring_size;
+	u32 max_rx_ring_size;
 
 	u32 msg_enable;
 
+	/* large_llq_header_enabled is used for two purposes:
+	 * 1. Indicates that large LLQ has been requested.
+	 * 2. Indicates whether large LLQ is set or not after device
+	 *    initialization / configuration.
+	 */
+	bool large_llq_header_enabled;
+	bool large_llq_header_supported;
+
 	u16 max_tx_sgl_size;
 	u16 max_rx_sgl_size;
 
 	u8 mac_addr[ETH_ALEN];
 
+	unsigned long keep_alive_timeout;
+	unsigned long missing_tx_completion_to;
+
 	char name[ENA_NAME_MAX_LEN];
 
+	struct ena_phc_info *phc_info;
+
 	unsigned long flags;
 	/* TX */
 	struct ena_ring tx_ring[ENA_MAX_NUM_IO_QUEUES]
@@ -298,18 +368,33 @@ struct ena_adapter {
 
 	/* timer service */
 	struct work_struct reset_task;
-	struct work_struct suspend_io_task;
-	struct work_struct resume_io_task;
 	struct timer_list timer_service;
 
 	bool wd_state;
+	bool dev_up_before_reset;
+	bool disable_meta_caching;
 	unsigned long last_keep_alive_jiffies;
 
 	struct u64_stats_sync syncp;
 	struct ena_stats_dev dev_stats;
+	struct ena_admin_eni_stats eni_stats;
+	struct ena_admin_ena_srd_info ena_srd_info;
 
 	/* last queue index that was checked for uncompleted tx packets */
 	u32 last_monitored_tx_qid;
+
+	enum ena_regs_reset_reason_types reset_reason;
+
+	struct bpf_prog *xdp_bpf_prog;
+	u32 xdp_first_ring;
+	u32 xdp_num_queues;
+
+	struct devlink *devlink;
+	struct devlink_port devlink_port;
+#ifdef CONFIG_DEBUG_FS
+
+	struct dentry *debugfs_base;
+#endif /* CONFIG_DEBUG_FS */
 };
 
 void ena_set_ethtool_ops(struct net_device *netdev);
@@ -318,6 +403,67 @@ void ena_dump_stats_to_dmesg(struct ena_adapter *adapter);
 
 void ena_dump_stats_to_buf(struct ena_adapter *adapter, u8 *buf);
 
+
+int ena_update_queue_params(struct ena_adapter *adapter,
+			    u32 new_tx_size,
+			    u32 new_rx_size,
+			    u32 new_llq_header_len);
+
+int ena_update_queue_count(struct ena_adapter *adapter, u32 new_channel_count);
+
+int ena_set_rx_copybreak(struct ena_adapter *adapter, u32 rx_copybreak);
+
 int ena_get_sset_count(struct net_device *netdev, int sset);
 
+static inline void ena_reset_device(struct ena_adapter *adapter,
+				    enum ena_regs_reset_reason_types reset_reason)
+{
+	adapter->reset_reason = reset_reason;
+	/* Make sure reset reason is set before triggering the reset */
+	smp_mb__before_atomic();
+	set_bit(ENA_FLAG_TRIGGER_RESET, &adapter->flags);
+}
+
+int ena_destroy_device(struct ena_adapter *adapter, bool graceful);
+int ena_restore_device(struct ena_adapter *adapter);
+int handle_invalid_req_id(struct ena_ring *ring, u16 req_id,
+			  struct ena_tx_buffer *tx_info, bool is_xdp);
+
+/* Increase a stat by cnt while holding syncp seqlock on 32bit machines */
+static inline void ena_increase_stat(u64 *statp, u64 cnt,
+				     struct u64_stats_sync *syncp)
+{
+	u64_stats_update_begin(syncp);
+	(*statp) += cnt;
+	u64_stats_update_end(syncp);
+}
+
+static inline void ena_ring_tx_doorbell(struct ena_ring *tx_ring)
+{
+	ena_com_write_sq_doorbell(tx_ring->ena_com_io_sq);
+	ena_increase_stat(&tx_ring->tx_stats.doorbells, 1, &tx_ring->syncp);
+}
+
+int ena_xmit_common(struct ena_adapter *adapter,
+		    struct ena_ring *ring,
+		    struct ena_tx_buffer *tx_info,
+		    struct ena_com_tx_ctx *ena_tx_ctx,
+		    u16 next_to_use,
+		    u32 bytes);
+void ena_unmap_tx_buff(struct ena_ring *tx_ring,
+		       struct ena_tx_buffer *tx_info);
+void ena_init_io_rings(struct ena_adapter *adapter,
+		       int first_index, int count);
+int ena_create_io_tx_queues_in_range(struct ena_adapter *adapter,
+				     int first_index, int count);
+int ena_setup_tx_resources_in_range(struct ena_adapter *adapter,
+				    int first_index, int count);
+void ena_free_all_io_tx_resources_in_range(struct ena_adapter *adapter,
+					   int first_index, int count);
+void ena_free_all_io_tx_resources(struct ena_adapter *adapter);
+void ena_down(struct ena_adapter *adapter);
+int ena_up(struct ena_adapter *adapter);
+void ena_unmask_interrupt(struct ena_ring *tx_ring, struct ena_ring *rx_ring);
+void ena_update_ring_numa_node(struct ena_ring *tx_ring,
+			       struct ena_ring *rx_ring);
 #endif /* !(ENA_H) */
diff --git a/drivers/net/ethernet/amazon/ena/ena_pci_id_tbl.h b/drivers/net/ethernet/amazon/ena/ena_pci_id_tbl.h
index f80d2a47fa94..3ecdf29160ca 100644
--- a/drivers/net/ethernet/amazon/ena/ena_pci_id_tbl.h
+++ b/drivers/net/ethernet/amazon/ena/ena_pci_id_tbl.h
@@ -1,33 +1,6 @@
+/* SPDX-License-Identifier: GPL-2.0 OR Linux-OpenIB */
 /*
- * Copyright 2015 Amazon.com, Inc. or its affiliates.
- *
- * This software is available to you under a choice of one of two
- * licenses.  You may choose to be licensed under the terms of the GNU
- * General Public License (GPL) Version 2, available from the file
- * COPYING in the main directory of this source tree, or the
- * BSD license below:
- *
- *     Redistribution and use in source and binary forms, with or
- *     without modification, are permitted provided that the following
- *     conditions are met:
- *
- *      - Redistributions of source code must retain the above
- *        copyright notice, this list of conditions and the following
- *        disclaimer.
- *
- *      - Redistributions in binary form must reproduce the above
- *        copyright notice, this list of conditions and the following
- *        disclaimer in the documentation and/or other materials
- *        provided with the distribution.
- *
- * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND,
- * EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF
- * MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND
- * NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS
- * BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN
- * ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN
- * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
- * SOFTWARE.
+ * Copyright 2015-2020 Amazon.com, Inc. or its affiliates. All rights reserved.
  */
 
 #ifndef ENA_PCI_ID_TBL_H_
@@ -53,10 +26,15 @@
 #define PCI_DEV_ID_ENA_LLQ_VF	0xec21
 #endif
 
+#ifndef PCI_DEV_ID_ENA_RESRV0
+#define PCI_DEV_ID_ENA_RESRV0	0x0051
+#endif
+
 #define ENA_PCI_ID_TABLE_ENTRY(devid) \
 	{PCI_DEVICE(PCI_VENDOR_ID_AMAZON, devid)},
 
 static const struct pci_device_id ena_pci_tbl[] = {
+	ENA_PCI_ID_TABLE_ENTRY(PCI_DEV_ID_ENA_RESRV0)
 	ENA_PCI_ID_TABLE_ENTRY(PCI_DEV_ID_ENA_PF)
 	ENA_PCI_ID_TABLE_ENTRY(PCI_DEV_ID_ENA_LLQ_PF)
 	ENA_PCI_ID_TABLE_ENTRY(PCI_DEV_ID_ENA_VF)
diff --git a/drivers/net/ethernet/amazon/ena/ena_phc.c b/drivers/net/ethernet/amazon/ena/ena_phc.c
new file mode 100644
index 000000000000..7867e893fd15
--- /dev/null
+++ b/drivers/net/ethernet/amazon/ena/ena_phc.c
@@ -0,0 +1,233 @@
+// SPDX-License-Identifier: GPL-2.0 OR Linux-OpenIB
+/*
+ * Copyright 2015-2022 Amazon.com, Inc. or its affiliates. All rights reserved.
+ */
+
+#include <linux/pci.h>
+#include "ena_netdev.h"
+#include "ena_phc.h"
+#include "ena_devlink.h"
+
+static int ena_phc_adjtime(struct ptp_clock_info *clock_info, s64 delta)
+{
+	return -EOPNOTSUPP;
+}
+
+static int ena_phc_adjfine(struct ptp_clock_info *clock_info, long scaled_ppm)
+{
+	return -EOPNOTSUPP;
+}
+
+static int ena_phc_feature_enable(struct ptp_clock_info *clock_info,
+				  struct ptp_clock_request *rq,
+				  int on)
+{
+	return -EOPNOTSUPP;
+}
+
+static int ena_phc_gettimex64(struct ptp_clock_info *clock_info,
+			      struct timespec64 *ts,
+			      struct ptp_system_timestamp *sts)
+{
+	struct ena_phc_info *phc_info =
+		container_of(clock_info, struct ena_phc_info, clock_info);
+	unsigned long flags;
+	u64 timestamp_nsec;
+	int rc;
+
+	spin_lock_irqsave(&phc_info->lock, flags);
+
+	ptp_read_system_prets(sts);
+
+	rc = ena_com_phc_get_timestamp(phc_info->adapter->ena_dev,
+				       &timestamp_nsec);
+
+	ptp_read_system_postts(sts);
+
+	spin_unlock_irqrestore(&phc_info->lock, flags);
+
+	*ts = ns_to_timespec64(timestamp_nsec);
+
+	return rc;
+}
+
+static int ena_phc_settime64(struct ptp_clock_info *clock_info,
+			     const struct timespec64 *ts)
+{
+	return -EOPNOTSUPP;
+}
+
+static struct ptp_clock_info ena_ptp_clock_info = {
+	.owner		= THIS_MODULE,
+	.n_alarm	= 0,
+	.n_ext_ts	= 0,
+	.n_per_out	= 0,
+	.pps		= 0,
+	.adjtime	= ena_phc_adjtime,
+	.adjfine	= ena_phc_adjfine,
+	.gettimex64	= ena_phc_gettimex64,
+	.settime64	= ena_phc_settime64,
+	.enable		= ena_phc_feature_enable,
+};
+
+/* Enable/Disable PHC by the kernel, affects on the next init flow */
+void ena_phc_enable(struct ena_adapter *adapter, bool enable)
+{
+	struct ena_phc_info *phc_info = adapter->phc_info;
+
+	if (!phc_info) {
+		netdev_err(adapter->netdev, "phc_info is not allocated\n");
+		return;
+	}
+
+	phc_info->enabled = enable;
+}
+
+/* Check if PHC is enabled by the kernel */
+bool ena_phc_is_enabled(struct ena_adapter *adapter)
+{
+	struct ena_phc_info *phc_info = adapter->phc_info;
+
+	return (phc_info && phc_info->enabled);
+}
+
+/* PHC is activated if ptp clock is registered in the kernel */
+bool ena_phc_is_active(struct ena_adapter *adapter)
+{
+	struct ena_phc_info *phc_info = adapter->phc_info;
+
+	return (phc_info && phc_info->clock);
+}
+
+static int ena_phc_register(struct ena_adapter *adapter)
+{
+	struct pci_dev *pdev = adapter->pdev;
+	struct ptp_clock_info *clock_info;
+	struct ena_phc_info *phc_info;
+	int rc = 0;
+
+	phc_info = adapter->phc_info;
+	clock_info = &phc_info->clock_info;
+
+	/* PHC may already be registered in case of a reset */
+	if (ena_phc_is_active(adapter))
+		return 0;
+
+	phc_info->adapter = adapter;
+
+	spin_lock_init(&phc_info->lock);
+
+	/* Fill the ptp_clock_info struct and register PTP clock */
+	*clock_info = ena_ptp_clock_info;
+	snprintf(clock_info->name,
+		 sizeof(clock_info->name),
+		 "ena-ptp-%02x",
+		 PCI_SLOT(pdev->devfn));
+
+	phc_info->clock = ptp_clock_register(clock_info, &pdev->dev);
+	if (IS_ERR(phc_info->clock)) {
+		rc = PTR_ERR(phc_info->clock);
+		netdev_err(adapter->netdev, "Failed registering ptp clock, error: %d\n",
+			   rc);
+		phc_info->clock = NULL;
+	}
+
+	return rc;
+}
+
+static void ena_phc_unregister(struct ena_adapter *adapter)
+{
+	struct ena_phc_info *phc_info = adapter->phc_info;
+
+	/* During reset flow, PHC must stay registered
+	 * to keep kernel's PHC index
+	 */
+	if (ena_phc_is_active(adapter) &&
+	    !test_bit(ENA_FLAG_TRIGGER_RESET, &adapter->flags)) {
+		ptp_clock_unregister(phc_info->clock);
+		phc_info->clock = NULL;
+	}
+}
+
+int ena_phc_alloc(struct ena_adapter *adapter)
+{
+	/* Allocate driver specific PHC info */
+	adapter->phc_info = vzalloc(sizeof(*adapter->phc_info));
+	if (unlikely(!adapter->phc_info)) {
+		netdev_err(adapter->netdev, "Failed to alloc phc_info\n");
+		return -ENOMEM;
+	}
+
+	return 0;
+}
+
+void ena_phc_free(struct ena_adapter *adapter)
+{
+	if (adapter->phc_info) {
+		vfree(adapter->phc_info);
+		adapter->phc_info = NULL;
+	}
+}
+
+int ena_phc_init(struct ena_adapter *adapter)
+{
+	struct ena_com_dev *ena_dev = adapter->ena_dev;
+	struct net_device *netdev = adapter->netdev;
+	int rc = -EOPNOTSUPP;
+
+	/* Validate PHC feature is supported in the device */
+	if (!ena_com_phc_supported(ena_dev)) {
+		netdev_dbg(netdev, "PHC feature is not supported by the device\n");
+		goto err_ena_com_phc_init;
+	}
+
+	/* Validate PHC feature is enabled by the kernel */
+	if (!ena_phc_is_enabled(adapter)) {
+		netdev_dbg(netdev, "PHC feature is not enabled by the kernel\n");
+		goto err_ena_com_phc_init;
+	}
+
+	/* Initialize device specific PHC info */
+	rc = ena_com_phc_init(ena_dev);
+	if (unlikely(rc)) {
+		netdev_err(netdev, "Failed to init phc, error: %d\n", rc);
+		goto err_ena_com_phc_init;
+	}
+
+	/* Configure PHC feature in driver and device */
+	rc = ena_com_phc_config(ena_dev);
+	if (unlikely(rc)) {
+		netdev_err(netdev, "Failed to config phc, error: %d\n", rc);
+		goto err_ena_com_phc_config;
+	}
+
+	/* Register to PTP class driver */
+	rc = ena_phc_register(adapter);
+	if (unlikely(rc)) {
+		netdev_err(netdev, "Failed to register phc, error: %d\n", rc);
+		goto err_ena_com_phc_config;
+	}
+
+	return 0;
+
+err_ena_com_phc_config:
+	ena_com_phc_destroy(ena_dev);
+err_ena_com_phc_init:
+	ena_phc_enable(adapter, false);
+	ena_devlink_disable_phc_param(adapter->devlink);
+	return rc;
+}
+
+void ena_phc_destroy(struct ena_adapter *adapter)
+{
+	ena_phc_unregister(adapter);
+	ena_com_phc_destroy(adapter->ena_dev);
+}
+
+int ena_phc_get_index(struct ena_adapter *adapter)
+{
+	if (ena_phc_is_active(adapter))
+		return ptp_clock_index(adapter->phc_info->clock);
+
+	return -1;
+}
diff --git a/drivers/net/ethernet/amazon/ena/ena_phc.h b/drivers/net/ethernet/amazon/ena/ena_phc.h
new file mode 100644
index 000000000000..7364fe714e44
--- /dev/null
+++ b/drivers/net/ethernet/amazon/ena/ena_phc.h
@@ -0,0 +1,37 @@
+/* SPDX-License-Identifier: GPL-2.0 OR Linux-OpenIB */
+/*
+ * Copyright 2015-2022 Amazon.com, Inc. or its affiliates. All rights reserved.
+ */
+
+#ifndef ENA_PHC_H
+#define ENA_PHC_H
+
+#include <linux/ptp_clock_kernel.h>
+
+struct ena_phc_info {
+	/* PTP hardware capabilities */
+	struct ptp_clock_info clock_info;
+
+	/* Registered PTP clock device */
+	struct ptp_clock *clock;
+
+	/* Adapter specific private data structure */
+	struct ena_adapter *adapter;
+
+	/* PHC lock */
+	spinlock_t lock;
+
+	/* Enabled by kernel */
+	bool enabled;
+};
+
+void ena_phc_enable(struct ena_adapter *adapter, bool enable);
+bool ena_phc_is_enabled(struct ena_adapter *adapter);
+bool ena_phc_is_active(struct ena_adapter *adapter);
+int ena_phc_get_index(struct ena_adapter *adapter);
+int ena_phc_init(struct ena_adapter *adapter);
+void ena_phc_destroy(struct ena_adapter *adapter);
+int ena_phc_alloc(struct ena_adapter *adapter);
+void ena_phc_free(struct ena_adapter *adapter);
+
+#endif /* ENA_PHC_H */
diff --git a/drivers/net/ethernet/amazon/ena/ena_regs_defs.h b/drivers/net/ethernet/amazon/ena/ena_regs_defs.h
index 26097a2b6030..51068dc1cc2a 100644
--- a/drivers/net/ethernet/amazon/ena/ena_regs_defs.h
+++ b/drivers/net/ethernet/amazon/ena/ena_regs_defs.h
@@ -1,133 +1,140 @@
+/* SPDX-License-Identifier: GPL-2.0 OR Linux-OpenIB */
 /*
- * Copyright 2015 - 2016 Amazon.com, Inc. or its affiliates.
- *
- * This software is available to you under a choice of one of two
- * licenses.  You may choose to be licensed under the terms of the GNU
- * General Public License (GPL) Version 2, available from the file
- * COPYING in the main directory of this source tree, or the
- * BSD license below:
- *
- *     Redistribution and use in source and binary forms, with or
- *     without modification, are permitted provided that the following
- *     conditions are met:
- *
- *      - Redistributions of source code must retain the above
- *        copyright notice, this list of conditions and the following
- *        disclaimer.
- *
- *      - Redistributions in binary form must reproduce the above
- *        copyright notice, this list of conditions and the following
- *        disclaimer in the documentation and/or other materials
- *        provided with the distribution.
- *
- * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND,
- * EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF
- * MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND
- * NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS
- * BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN
- * ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN
- * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
- * SOFTWARE.
+ * Copyright 2015-2020 Amazon.com, Inc. or its affiliates. All rights reserved.
  */
 #ifndef _ENA_REGS_H_
 #define _ENA_REGS_H_
 
+enum ena_regs_reset_reason_types {
+	ENA_REGS_RESET_NORMAL                       = 0,
+	ENA_REGS_RESET_KEEP_ALIVE_TO                = 1,
+	ENA_REGS_RESET_ADMIN_TO                     = 2,
+	ENA_REGS_RESET_MISS_TX_CMPL                 = 3,
+	ENA_REGS_RESET_INV_RX_REQ_ID                = 4,
+	ENA_REGS_RESET_INV_TX_REQ_ID                = 5,
+	ENA_REGS_RESET_TOO_MANY_RX_DESCS            = 6,
+	ENA_REGS_RESET_INIT_ERR                     = 7,
+	ENA_REGS_RESET_DRIVER_INVALID_STATE         = 8,
+	ENA_REGS_RESET_OS_TRIGGER                   = 9,
+	ENA_REGS_RESET_OS_NETDEV_WD                 = 10,
+	ENA_REGS_RESET_SHUTDOWN                     = 11,
+	ENA_REGS_RESET_USER_TRIGGER                 = 12,
+	ENA_REGS_RESET_GENERIC                      = 13,
+	ENA_REGS_RESET_MISS_INTERRUPT               = 14,
+	ENA_REGS_RESET_SUSPECTED_POLL_STARVATION    = 15,
+	ENA_REGS_RESET_RX_DESCRIPTOR_MALFORMED	    = 16,
+};
+
 /* ena_registers offsets */
-#define ENA_REGS_VERSION_OFF		0x0
-#define ENA_REGS_CONTROLLER_VERSION_OFF		0x4
-#define ENA_REGS_CAPS_OFF		0x8
-#define ENA_REGS_CAPS_EXT_OFF		0xc
-#define ENA_REGS_AQ_BASE_LO_OFF		0x10
-#define ENA_REGS_AQ_BASE_HI_OFF		0x14
-#define ENA_REGS_AQ_CAPS_OFF		0x18
-#define ENA_REGS_ACQ_BASE_LO_OFF		0x20
-#define ENA_REGS_ACQ_BASE_HI_OFF		0x24
-#define ENA_REGS_ACQ_CAPS_OFF		0x28
-#define ENA_REGS_AQ_DB_OFF		0x2c
-#define ENA_REGS_ACQ_TAIL_OFF		0x30
-#define ENA_REGS_AENQ_CAPS_OFF		0x34
-#define ENA_REGS_AENQ_BASE_LO_OFF		0x38
-#define ENA_REGS_AENQ_BASE_HI_OFF		0x3c
-#define ENA_REGS_AENQ_HEAD_DB_OFF		0x40
-#define ENA_REGS_AENQ_TAIL_OFF		0x44
-#define ENA_REGS_INTR_MASK_OFF		0x4c
-#define ENA_REGS_DEV_CTL_OFF		0x54
-#define ENA_REGS_DEV_STS_OFF		0x58
-#define ENA_REGS_MMIO_REG_READ_OFF		0x5c
-#define ENA_REGS_MMIO_RESP_LO_OFF		0x60
-#define ENA_REGS_MMIO_RESP_HI_OFF		0x64
-#define ENA_REGS_RSS_IND_ENTRY_UPDATE_OFF		0x68
+
+/* 0 base */
+#define ENA_REGS_VERSION_OFF                                0x0
+#define ENA_REGS_CONTROLLER_VERSION_OFF                     0x4
+#define ENA_REGS_CAPS_OFF                                   0x8
+#define ENA_REGS_CAPS_EXT_OFF                               0xc
+#define ENA_REGS_AQ_BASE_LO_OFF                             0x10
+#define ENA_REGS_AQ_BASE_HI_OFF                             0x14
+#define ENA_REGS_AQ_CAPS_OFF                                0x18
+#define ENA_REGS_ACQ_BASE_LO_OFF                            0x20
+#define ENA_REGS_ACQ_BASE_HI_OFF                            0x24
+#define ENA_REGS_ACQ_CAPS_OFF                               0x28
+#define ENA_REGS_AQ_DB_OFF                                  0x2c
+#define ENA_REGS_ACQ_TAIL_OFF                               0x30
+#define ENA_REGS_AENQ_CAPS_OFF                              0x34
+#define ENA_REGS_AENQ_BASE_LO_OFF                           0x38
+#define ENA_REGS_AENQ_BASE_HI_OFF                           0x3c
+#define ENA_REGS_AENQ_HEAD_DB_OFF                           0x40
+#define ENA_REGS_AENQ_TAIL_OFF                              0x44
+#define ENA_REGS_INTR_MASK_OFF                              0x4c
+#define ENA_REGS_DEV_CTL_OFF                                0x54
+#define ENA_REGS_DEV_STS_OFF                                0x58
+#define ENA_REGS_MMIO_REG_READ_OFF                          0x5c
+#define ENA_REGS_MMIO_RESP_LO_OFF                           0x60
+#define ENA_REGS_MMIO_RESP_HI_OFF                           0x64
+#define ENA_REGS_RSS_IND_ENTRY_UPDATE_OFF                   0x68
+
+/* phc_registers offsets */
+
+/* 100 base */
+#define ENA_REGS_PHC_DB_OFF                                 0x100
 
 /* version register */
-#define ENA_REGS_VERSION_MINOR_VERSION_MASK		0xff
-#define ENA_REGS_VERSION_MAJOR_VERSION_SHIFT		8
-#define ENA_REGS_VERSION_MAJOR_VERSION_MASK		0xff00
+#define ENA_REGS_VERSION_MINOR_VERSION_MASK                 0xff
+#define ENA_REGS_VERSION_MAJOR_VERSION_SHIFT                8
+#define ENA_REGS_VERSION_MAJOR_VERSION_MASK                 0xff00
 
 /* controller_version register */
-#define ENA_REGS_CONTROLLER_VERSION_SUBMINOR_VERSION_MASK		0xff
-#define ENA_REGS_CONTROLLER_VERSION_MINOR_VERSION_SHIFT		8
-#define ENA_REGS_CONTROLLER_VERSION_MINOR_VERSION_MASK		0xff00
-#define ENA_REGS_CONTROLLER_VERSION_MAJOR_VERSION_SHIFT		16
-#define ENA_REGS_CONTROLLER_VERSION_MAJOR_VERSION_MASK		0xff0000
-#define ENA_REGS_CONTROLLER_VERSION_IMPL_ID_SHIFT		24
-#define ENA_REGS_CONTROLLER_VERSION_IMPL_ID_MASK		0xff000000
+#define ENA_REGS_CONTROLLER_VERSION_SUBMINOR_VERSION_MASK   0xff
+#define ENA_REGS_CONTROLLER_VERSION_MINOR_VERSION_SHIFT     8
+#define ENA_REGS_CONTROLLER_VERSION_MINOR_VERSION_MASK      0xff00
+#define ENA_REGS_CONTROLLER_VERSION_MAJOR_VERSION_SHIFT     16
+#define ENA_REGS_CONTROLLER_VERSION_MAJOR_VERSION_MASK      0xff0000
+#define ENA_REGS_CONTROLLER_VERSION_IMPL_ID_SHIFT           24
+#define ENA_REGS_CONTROLLER_VERSION_IMPL_ID_MASK            0xff000000
 
 /* caps register */
-#define ENA_REGS_CAPS_CONTIGUOUS_QUEUE_REQUIRED_MASK		0x1
-#define ENA_REGS_CAPS_RESET_TIMEOUT_SHIFT		1
-#define ENA_REGS_CAPS_RESET_TIMEOUT_MASK		0x3e
-#define ENA_REGS_CAPS_DMA_ADDR_WIDTH_SHIFT		8
-#define ENA_REGS_CAPS_DMA_ADDR_WIDTH_MASK		0xff00
+#define ENA_REGS_CAPS_CONTIGUOUS_QUEUE_REQUIRED_MASK        0x1
+#define ENA_REGS_CAPS_RESET_TIMEOUT_SHIFT                   1
+#define ENA_REGS_CAPS_RESET_TIMEOUT_MASK                    0x3e
+#define ENA_REGS_CAPS_DMA_ADDR_WIDTH_SHIFT                  8
+#define ENA_REGS_CAPS_DMA_ADDR_WIDTH_MASK                   0xff00
+#define ENA_REGS_CAPS_ADMIN_CMD_TO_SHIFT                    16
+#define ENA_REGS_CAPS_ADMIN_CMD_TO_MASK                     0xf0000
 
 /* aq_caps register */
-#define ENA_REGS_AQ_CAPS_AQ_DEPTH_MASK		0xffff
-#define ENA_REGS_AQ_CAPS_AQ_ENTRY_SIZE_SHIFT		16
-#define ENA_REGS_AQ_CAPS_AQ_ENTRY_SIZE_MASK		0xffff0000
+#define ENA_REGS_AQ_CAPS_AQ_DEPTH_MASK                      0xffff
+#define ENA_REGS_AQ_CAPS_AQ_ENTRY_SIZE_SHIFT                16
+#define ENA_REGS_AQ_CAPS_AQ_ENTRY_SIZE_MASK                 0xffff0000
 
 /* acq_caps register */
-#define ENA_REGS_ACQ_CAPS_ACQ_DEPTH_MASK		0xffff
-#define ENA_REGS_ACQ_CAPS_ACQ_ENTRY_SIZE_SHIFT		16
-#define ENA_REGS_ACQ_CAPS_ACQ_ENTRY_SIZE_MASK		0xffff0000
+#define ENA_REGS_ACQ_CAPS_ACQ_DEPTH_MASK                    0xffff
+#define ENA_REGS_ACQ_CAPS_ACQ_ENTRY_SIZE_SHIFT              16
+#define ENA_REGS_ACQ_CAPS_ACQ_ENTRY_SIZE_MASK               0xffff0000
 
 /* aenq_caps register */
-#define ENA_REGS_AENQ_CAPS_AENQ_DEPTH_MASK		0xffff
-#define ENA_REGS_AENQ_CAPS_AENQ_ENTRY_SIZE_SHIFT		16
-#define ENA_REGS_AENQ_CAPS_AENQ_ENTRY_SIZE_MASK		0xffff0000
+#define ENA_REGS_AENQ_CAPS_AENQ_DEPTH_MASK                  0xffff
+#define ENA_REGS_AENQ_CAPS_AENQ_ENTRY_SIZE_SHIFT            16
+#define ENA_REGS_AENQ_CAPS_AENQ_ENTRY_SIZE_MASK             0xffff0000
 
 /* dev_ctl register */
-#define ENA_REGS_DEV_CTL_DEV_RESET_MASK		0x1
-#define ENA_REGS_DEV_CTL_AQ_RESTART_SHIFT		1
-#define ENA_REGS_DEV_CTL_AQ_RESTART_MASK		0x2
-#define ENA_REGS_DEV_CTL_QUIESCENT_SHIFT		2
-#define ENA_REGS_DEV_CTL_QUIESCENT_MASK		0x4
-#define ENA_REGS_DEV_CTL_IO_RESUME_SHIFT		3
-#define ENA_REGS_DEV_CTL_IO_RESUME_MASK		0x8
+#define ENA_REGS_DEV_CTL_DEV_RESET_MASK                     0x1
+#define ENA_REGS_DEV_CTL_AQ_RESTART_SHIFT                   1
+#define ENA_REGS_DEV_CTL_AQ_RESTART_MASK                    0x2
+#define ENA_REGS_DEV_CTL_QUIESCENT_SHIFT                    2
+#define ENA_REGS_DEV_CTL_QUIESCENT_MASK                     0x4
+#define ENA_REGS_DEV_CTL_IO_RESUME_SHIFT                    3
+#define ENA_REGS_DEV_CTL_IO_RESUME_MASK                     0x8
+#define ENA_REGS_DEV_CTL_RESET_REASON_SHIFT                 28
+#define ENA_REGS_DEV_CTL_RESET_REASON_MASK                  0xf0000000
 
 /* dev_sts register */
-#define ENA_REGS_DEV_STS_READY_MASK		0x1
-#define ENA_REGS_DEV_STS_AQ_RESTART_IN_PROGRESS_SHIFT		1
-#define ENA_REGS_DEV_STS_AQ_RESTART_IN_PROGRESS_MASK		0x2
-#define ENA_REGS_DEV_STS_AQ_RESTART_FINISHED_SHIFT		2
-#define ENA_REGS_DEV_STS_AQ_RESTART_FINISHED_MASK		0x4
-#define ENA_REGS_DEV_STS_RESET_IN_PROGRESS_SHIFT		3
-#define ENA_REGS_DEV_STS_RESET_IN_PROGRESS_MASK		0x8
-#define ENA_REGS_DEV_STS_RESET_FINISHED_SHIFT		4
-#define ENA_REGS_DEV_STS_RESET_FINISHED_MASK		0x10
-#define ENA_REGS_DEV_STS_FATAL_ERROR_SHIFT		5
-#define ENA_REGS_DEV_STS_FATAL_ERROR_MASK		0x20
-#define ENA_REGS_DEV_STS_QUIESCENT_STATE_IN_PROGRESS_SHIFT		6
-#define ENA_REGS_DEV_STS_QUIESCENT_STATE_IN_PROGRESS_MASK		0x40
-#define ENA_REGS_DEV_STS_QUIESCENT_STATE_ACHIEVED_SHIFT		7
-#define ENA_REGS_DEV_STS_QUIESCENT_STATE_ACHIEVED_MASK		0x80
+#define ENA_REGS_DEV_STS_READY_MASK                         0x1
+#define ENA_REGS_DEV_STS_AQ_RESTART_IN_PROGRESS_SHIFT       1
+#define ENA_REGS_DEV_STS_AQ_RESTART_IN_PROGRESS_MASK        0x2
+#define ENA_REGS_DEV_STS_AQ_RESTART_FINISHED_SHIFT          2
+#define ENA_REGS_DEV_STS_AQ_RESTART_FINISHED_MASK           0x4
+#define ENA_REGS_DEV_STS_RESET_IN_PROGRESS_SHIFT            3
+#define ENA_REGS_DEV_STS_RESET_IN_PROGRESS_MASK             0x8
+#define ENA_REGS_DEV_STS_RESET_FINISHED_SHIFT               4
+#define ENA_REGS_DEV_STS_RESET_FINISHED_MASK                0x10
+#define ENA_REGS_DEV_STS_FATAL_ERROR_SHIFT                  5
+#define ENA_REGS_DEV_STS_FATAL_ERROR_MASK                   0x20
+#define ENA_REGS_DEV_STS_QUIESCENT_STATE_IN_PROGRESS_SHIFT  6
+#define ENA_REGS_DEV_STS_QUIESCENT_STATE_IN_PROGRESS_MASK   0x40
+#define ENA_REGS_DEV_STS_QUIESCENT_STATE_ACHIEVED_SHIFT     7
+#define ENA_REGS_DEV_STS_QUIESCENT_STATE_ACHIEVED_MASK      0x80
 
 /* mmio_reg_read register */
-#define ENA_REGS_MMIO_REG_READ_REQ_ID_MASK		0xffff
-#define ENA_REGS_MMIO_REG_READ_REG_OFF_SHIFT		16
-#define ENA_REGS_MMIO_REG_READ_REG_OFF_MASK		0xffff0000
+#define ENA_REGS_MMIO_REG_READ_REQ_ID_MASK                  0xffff
+#define ENA_REGS_MMIO_REG_READ_REG_OFF_SHIFT                16
+#define ENA_REGS_MMIO_REG_READ_REG_OFF_MASK                 0xffff0000
 
 /* rss_ind_entry_update register */
-#define ENA_REGS_RSS_IND_ENTRY_UPDATE_INDEX_MASK		0xffff
-#define ENA_REGS_RSS_IND_ENTRY_UPDATE_CQ_IDX_SHIFT		16
-#define ENA_REGS_RSS_IND_ENTRY_UPDATE_CQ_IDX_MASK		0xffff0000
+#define ENA_REGS_RSS_IND_ENTRY_UPDATE_INDEX_MASK            0xffff
+#define ENA_REGS_RSS_IND_ENTRY_UPDATE_CQ_IDX_SHIFT          16
+#define ENA_REGS_RSS_IND_ENTRY_UPDATE_CQ_IDX_MASK           0xffff0000
+
+/* phc_db_req_id register */
+#define ENA_REGS_PHC_DB_REQ_ID_MASK                         0xffff
 
-#endif /*_ENA_REGS_H_ */
+#endif /* _ENA_REGS_H_ */
diff --git a/drivers/net/ethernet/amazon/ena/ena_xdp.c b/drivers/net/ethernet/amazon/ena/ena_xdp.c
new file mode 100644
index 000000000000..5b175e7e92a1
--- /dev/null
+++ b/drivers/net/ethernet/amazon/ena/ena_xdp.c
@@ -0,0 +1,469 @@
+// SPDX-License-Identifier: GPL-2.0 OR Linux-OpenIB
+/*
+ * Copyright 2015-2021 Amazon.com, Inc. or its affiliates. All rights reserved.
+ */
+
+#include "ena_xdp.h"
+
+static int validate_xdp_req_id(struct ena_ring *tx_ring, u16 req_id)
+{
+	struct ena_tx_buffer *tx_info;
+
+	tx_info = &tx_ring->tx_buffer_info[req_id];
+	if (likely(tx_info->xdpf))
+		return 0;
+
+	return handle_invalid_req_id(tx_ring, req_id, tx_info, true);
+}
+
+static int ena_xdp_tx_map_frame(struct ena_ring *tx_ring,
+				struct ena_tx_buffer *tx_info,
+				struct xdp_frame *xdpf,
+				struct ena_com_tx_ctx *ena_tx_ctx)
+{
+	struct ena_adapter *adapter = tx_ring->adapter;
+	struct ena_com_buf *ena_buf;
+	int push_len = 0;
+	dma_addr_t dma;
+	void *data;
+	u32 size;
+
+	tx_info->xdpf = xdpf;
+	data = tx_info->xdpf->data;
+	size = tx_info->xdpf->len;
+
+	if (tx_ring->tx_mem_queue_type == ENA_ADMIN_PLACEMENT_POLICY_DEV) {
+		/* Designate part of the packet for LLQ */
+		push_len = min_t(u32, size, tx_ring->tx_max_header_size);
+
+		ena_tx_ctx->push_header = data;
+
+		size -= push_len;
+		data += push_len;
+	}
+
+	ena_tx_ctx->header_len = push_len;
+
+	if (size > 0) {
+		dma = dma_map_single(tx_ring->dev,
+				     data,
+				     size,
+				     DMA_TO_DEVICE);
+		if (unlikely(dma_mapping_error(tx_ring->dev, dma)))
+			goto error_report_dma_error;
+
+		tx_info->map_linear_data = 0;
+
+		ena_buf = tx_info->bufs;
+		ena_buf->paddr = dma;
+		ena_buf->len = size;
+
+		ena_tx_ctx->ena_bufs = ena_buf;
+		ena_tx_ctx->num_bufs = tx_info->num_of_bufs = 1;
+	}
+
+	return 0;
+
+error_report_dma_error:
+	ena_increase_stat(&tx_ring->tx_stats.dma_mapping_err, 1,
+			  &tx_ring->syncp);
+	netif_warn(adapter, tx_queued, adapter->netdev, "Failed to map xdp buff\n");
+
+	return -EINVAL;
+}
+
+int ena_xdp_xmit_frame(struct ena_ring *tx_ring,
+		       struct ena_adapter *adapter,
+		       struct xdp_frame *xdpf,
+		       int flags)
+{
+	struct ena_com_tx_ctx ena_tx_ctx = {};
+	struct ena_tx_buffer *tx_info;
+	u16 next_to_use, req_id;
+	int rc;
+
+	next_to_use = tx_ring->next_to_use;
+	req_id = tx_ring->free_ids[next_to_use];
+	tx_info = &tx_ring->tx_buffer_info[req_id];
+	tx_info->num_of_bufs = 0;
+
+	rc = ena_xdp_tx_map_frame(tx_ring, tx_info, xdpf, &ena_tx_ctx);
+	if (unlikely(rc))
+		goto err;
+
+	ena_tx_ctx.req_id = req_id;
+
+	rc = ena_xmit_common(adapter,
+			     tx_ring,
+			     tx_info,
+			     &ena_tx_ctx,
+			     next_to_use,
+			     xdpf->len);
+	if (rc)
+		goto error_unmap_dma;
+
+	/* trigger the dma engine. ena_ring_tx_doorbell()
+	 * calls a memory barrier inside it.
+	 */
+	if (flags & XDP_XMIT_FLUSH)
+		ena_ring_tx_doorbell(tx_ring);
+
+	return rc;
+
+error_unmap_dma:
+	ena_unmap_tx_buff(tx_ring, tx_info);
+err:
+	tx_info->xdpf = NULL;
+
+	return rc;
+}
+
+int ena_xdp_xmit(struct net_device *dev, int n,
+		 struct xdp_frame **frames, u32 flags)
+{
+	struct ena_adapter *adapter = netdev_priv(dev);
+	struct ena_ring *tx_ring;
+	int qid, i, nxmit = 0;
+
+	if (unlikely(flags & ~XDP_XMIT_FLAGS_MASK))
+		return -EINVAL;
+
+	if (!test_bit(ENA_FLAG_DEV_UP, &adapter->flags))
+		return -ENETDOWN;
+
+	/* We assume that all rings have the same XDP program */
+	if (!READ_ONCE(adapter->rx_ring->xdp_bpf_prog))
+		return -ENXIO;
+
+	qid = smp_processor_id() % adapter->xdp_num_queues;
+	qid += adapter->xdp_first_ring;
+	tx_ring = &adapter->tx_ring[qid];
+
+	/* Other CPU ids might try to send thorugh this queue */
+	spin_lock(&tx_ring->xdp_tx_lock);
+
+	for (i = 0; i < n; i++) {
+		if (ena_xdp_xmit_frame(tx_ring, adapter, frames[i], 0))
+			break;
+		nxmit++;
+	}
+
+	/* Ring doorbell to make device aware of the packets */
+	if (flags & XDP_XMIT_FLUSH)
+		ena_ring_tx_doorbell(tx_ring);
+
+	spin_unlock(&tx_ring->xdp_tx_lock);
+
+	/* Return number of packets sent */
+	return nxmit;
+}
+
+static void ena_init_all_xdp_queues(struct ena_adapter *adapter)
+{
+	adapter->xdp_first_ring = adapter->num_io_queues;
+	adapter->xdp_num_queues = adapter->num_io_queues;
+
+	ena_init_io_rings(adapter,
+			  adapter->xdp_first_ring,
+			  adapter->xdp_num_queues);
+}
+
+int ena_setup_and_create_all_xdp_queues(struct ena_adapter *adapter)
+{
+	u32 xdp_first_ring = adapter->xdp_first_ring;
+	u32 xdp_num_queues = adapter->xdp_num_queues;
+	int rc = 0;
+
+	rc = ena_setup_tx_resources_in_range(adapter, xdp_first_ring, xdp_num_queues);
+	if (rc)
+		goto setup_err;
+
+	rc = ena_create_io_tx_queues_in_range(adapter, xdp_first_ring, xdp_num_queues);
+	if (rc)
+		goto create_err;
+
+	return 0;
+
+create_err:
+	ena_free_all_io_tx_resources_in_range(adapter, xdp_first_ring, xdp_num_queues);
+setup_err:
+	return rc;
+}
+
+/* Provides a way for both kernel and bpf-prog to know
+ * more about the RX-queue a given XDP frame arrived on.
+ */
+int ena_xdp_register_rxq_info(struct ena_ring *rx_ring)
+{
+	int rc;
+
+	rc = xdp_rxq_info_reg(&rx_ring->xdp_rxq, rx_ring->netdev, rx_ring->qid, 0);
+
+	netif_dbg(rx_ring->adapter, ifup, rx_ring->netdev, "Registering RX info for queue %d",
+		  rx_ring->qid);
+	if (rc) {
+		netif_err(rx_ring->adapter, ifup, rx_ring->netdev,
+			  "Failed to register xdp rx queue info. RX queue num %d rc: %d\n",
+			  rx_ring->qid, rc);
+		goto err;
+	}
+
+	rc = xdp_rxq_info_reg_mem_model(&rx_ring->xdp_rxq, MEM_TYPE_PAGE_SHARED, NULL);
+
+	if (rc) {
+		netif_err(rx_ring->adapter, ifup, rx_ring->netdev,
+			  "Failed to register xdp rx queue info memory model. RX queue num %d rc: %d\n",
+			  rx_ring->qid, rc);
+		xdp_rxq_info_unreg(&rx_ring->xdp_rxq);
+	}
+
+err:
+	return rc;
+}
+
+void ena_xdp_unregister_rxq_info(struct ena_ring *rx_ring)
+{
+	netif_dbg(rx_ring->adapter, ifdown, rx_ring->netdev,
+		  "Unregistering RX info for queue %d",
+		  rx_ring->qid);
+	xdp_rxq_info_unreg_mem_model(&rx_ring->xdp_rxq);
+	xdp_rxq_info_unreg(&rx_ring->xdp_rxq);
+}
+
+void ena_xdp_exchange_program_rx_in_range(struct ena_adapter *adapter,
+					  struct bpf_prog *prog,
+					  int first, int count)
+{
+	struct bpf_prog *old_bpf_prog;
+	struct ena_ring *rx_ring;
+	int i = 0;
+
+	for (i = first; i < count; i++) {
+		rx_ring = &adapter->rx_ring[i];
+		old_bpf_prog = xchg(&rx_ring->xdp_bpf_prog, prog);
+
+		if (!old_bpf_prog && prog) {
+			rx_ring->rx_headroom = XDP_PACKET_HEADROOM;
+		} else if (old_bpf_prog && !prog) {
+			rx_ring->rx_headroom = NET_SKB_PAD;
+		}
+	}
+}
+
+static void ena_xdp_exchange_program(struct ena_adapter *adapter,
+				     struct bpf_prog *prog)
+{
+	struct bpf_prog *old_bpf_prog = xchg(&adapter->xdp_bpf_prog, prog);
+
+	ena_xdp_exchange_program_rx_in_range(adapter,
+					     prog,
+					     0,
+					     adapter->num_io_queues);
+
+	if (old_bpf_prog)
+		bpf_prog_put(old_bpf_prog);
+}
+
+static int ena_destroy_and_free_all_xdp_queues(struct ena_adapter *adapter)
+{
+	bool was_up;
+	int rc;
+
+	was_up = test_bit(ENA_FLAG_DEV_UP, &adapter->flags);
+
+	if (was_up)
+		ena_down(adapter);
+
+	adapter->xdp_first_ring = 0;
+	adapter->xdp_num_queues = 0;
+	ena_xdp_exchange_program(adapter, NULL);
+	if (was_up) {
+		rc = ena_up(adapter);
+		if (rc)
+			return rc;
+	}
+	return 0;
+}
+
+static int ena_xdp_set(struct net_device *netdev, struct netdev_bpf *bpf)
+{
+	struct ena_adapter *adapter = netdev_priv(netdev);
+	struct bpf_prog *prog = bpf->prog;
+	struct bpf_prog *old_bpf_prog;
+	int rc, prev_mtu;
+	bool is_up;
+
+	is_up = test_bit(ENA_FLAG_DEV_UP, &adapter->flags);
+	rc = ena_xdp_allowed(adapter);
+	if (rc == ENA_XDP_ALLOWED) {
+		old_bpf_prog = adapter->xdp_bpf_prog;
+		if (prog) {
+			if (!is_up) {
+				ena_init_all_xdp_queues(adapter);
+			} else if (!old_bpf_prog) {
+				ena_down(adapter);
+				ena_init_all_xdp_queues(adapter);
+			}
+			ena_xdp_exchange_program(adapter, prog);
+
+			netif_dbg(adapter, drv, adapter->netdev, "Set a new XDP program\n");
+
+			if (is_up && !old_bpf_prog) {
+				rc = ena_up(adapter);
+				if (rc)
+					return rc;
+			}
+			xdp_features_set_redirect_target(netdev, false);
+		} else if (old_bpf_prog) {
+			xdp_features_clear_redirect_target(netdev);
+			netif_dbg(adapter, drv, adapter->netdev, "Removing XDP program\n");
+
+			rc = ena_destroy_and_free_all_xdp_queues(adapter);
+			if (rc)
+				return rc;
+		}
+
+		prev_mtu = netdev->max_mtu;
+		netdev->max_mtu = prog ? ENA_XDP_MAX_MTU : adapter->max_mtu;
+
+		if (!old_bpf_prog)
+			netif_info(adapter, drv, adapter->netdev,
+				   "XDP program is set, changing the max_mtu from %d to %d",
+				   prev_mtu, netdev->max_mtu);
+
+	} else if (rc == ENA_XDP_CURRENT_MTU_TOO_LARGE) {
+		netif_err(adapter, drv, adapter->netdev,
+			  "Failed to set xdp program, the current MTU (%d) is larger than the maximum allowed MTU (%lu) while xdp is on",
+			  netdev->mtu, ENA_XDP_MAX_MTU);
+		NL_SET_ERR_MSG_MOD(bpf->extack,
+				   "Failed to set xdp program, the current MTU is larger than the maximum allowed MTU. Check the dmesg for more info");
+		return -EINVAL;
+	} else if (rc == ENA_XDP_NO_ENOUGH_QUEUES) {
+		netif_err(adapter, drv, adapter->netdev,
+			  "Failed to set xdp program, the Rx/Tx channel count should be at most half of the maximum allowed channel count. The current queue count (%d), the maximal queue count (%d)\n",
+			  adapter->num_io_queues, adapter->max_num_io_queues);
+		NL_SET_ERR_MSG_MOD(bpf->extack,
+				   "Failed to set xdp program, there is no enough space for allocating XDP queues, Check the dmesg for more info");
+		return -EINVAL;
+	}
+
+	return 0;
+}
+
+/* This is the main xdp callback, it's used by the kernel to set/unset the xdp
+ * program as well as to query the current xdp program id.
+ */
+int ena_xdp(struct net_device *netdev, struct netdev_bpf *bpf)
+{
+	switch (bpf->command) {
+	case XDP_SETUP_PROG:
+		return ena_xdp_set(netdev, bpf);
+	default:
+		return -EINVAL;
+	}
+	return 0;
+}
+
+static int ena_clean_xdp_irq(struct ena_ring *tx_ring, u32 budget)
+{
+	u32 total_done = 0;
+	u16 next_to_clean;
+	int tx_pkts = 0;
+	u16 req_id;
+	int rc;
+
+	if (unlikely(!tx_ring))
+		return 0;
+	next_to_clean = tx_ring->next_to_clean;
+
+	while (tx_pkts < budget) {
+		struct ena_tx_buffer *tx_info;
+		struct xdp_frame *xdpf;
+
+		rc = ena_com_tx_comp_req_id_get(tx_ring->ena_com_io_cq,
+						&req_id);
+		if (rc) {
+			if (unlikely(rc == -EINVAL))
+				handle_invalid_req_id(tx_ring, req_id, NULL, true);
+			break;
+		}
+
+		/* validate that the request id points to a valid xdp_frame */
+		rc = validate_xdp_req_id(tx_ring, req_id);
+		if (rc)
+			break;
+
+		tx_info = &tx_ring->tx_buffer_info[req_id];
+
+		tx_info->last_jiffies = 0;
+
+		xdpf = tx_info->xdpf;
+		tx_info->xdpf = NULL;
+		ena_unmap_tx_buff(tx_ring, tx_info);
+		xdp_return_frame(xdpf);
+
+		tx_pkts++;
+		total_done += tx_info->tx_descs;
+		tx_ring->free_ids[next_to_clean] = req_id;
+		next_to_clean = ENA_TX_RING_IDX_NEXT(next_to_clean,
+						     tx_ring->ring_size);
+
+		netif_dbg(tx_ring->adapter, tx_done, tx_ring->netdev,
+			  "tx_poll: q %d pkt #%d req_id %d\n", tx_ring->qid, tx_pkts, req_id);
+	}
+
+	tx_ring->next_to_clean = next_to_clean;
+	ena_com_comp_ack(tx_ring->ena_com_io_sq, total_done);
+
+	netif_dbg(tx_ring->adapter, tx_done, tx_ring->netdev,
+		  "tx_poll: q %d done. total pkts: %d\n",
+		  tx_ring->qid, tx_pkts);
+
+	return tx_pkts;
+}
+
+/* This is the XDP napi callback. XDP queues use a separate napi callback
+ * than Rx/Tx queues.
+ */
+int ena_xdp_io_poll(struct napi_struct *napi, int budget)
+{
+	struct ena_napi *ena_napi = container_of(napi, struct ena_napi, napi);
+	struct ena_ring *tx_ring;
+	u32 work_done;
+	int ret;
+
+	tx_ring = ena_napi->tx_ring;
+
+	if (!test_bit(ENA_FLAG_DEV_UP, &tx_ring->adapter->flags) ||
+	    test_bit(ENA_FLAG_TRIGGER_RESET, &tx_ring->adapter->flags)) {
+		napi_complete_done(napi, 0);
+		return 0;
+	}
+
+	work_done = ena_clean_xdp_irq(tx_ring, budget);
+
+	/* If the device is about to reset or down, avoid unmask
+	 * the interrupt and return 0 so NAPI won't reschedule
+	 */
+	if (unlikely(!test_bit(ENA_FLAG_DEV_UP, &tx_ring->adapter->flags))) {
+		napi_complete_done(napi, 0);
+		ret = 0;
+	} else if (budget > work_done) {
+		ena_increase_stat(&tx_ring->tx_stats.napi_comp, 1,
+				  &tx_ring->syncp);
+		if (napi_complete_done(napi, work_done))
+			ena_unmask_interrupt(tx_ring, NULL);
+
+		ena_update_ring_numa_node(tx_ring, NULL);
+		ret = work_done;
+	} else {
+		ret = budget;
+	}
+
+	u64_stats_update_begin(&tx_ring->syncp);
+	tx_ring->tx_stats.tx_poll++;
+	u64_stats_update_end(&tx_ring->syncp);
+	tx_ring->tx_stats.last_napi_jiffies = jiffies;
+
+	return ret;
+}
diff --git a/drivers/net/ethernet/amazon/ena/ena_xdp.h b/drivers/net/ethernet/amazon/ena/ena_xdp.h
new file mode 100644
index 000000000000..cfd82728486a
--- /dev/null
+++ b/drivers/net/ethernet/amazon/ena/ena_xdp.h
@@ -0,0 +1,151 @@
+/* SPDX-License-Identifier: GPL-2.0 OR Linux-OpenIB */
+/*
+ * Copyright 2015-2021 Amazon.com, Inc. or its affiliates. All rights reserved.
+ */
+
+#ifndef ENA_XDP_H
+#define ENA_XDP_H
+
+#include "ena_netdev.h"
+#include <linux/bpf_trace.h>
+
+/* The max MTU size is configured to be the ethernet frame size without
+ * the overhead of the ethernet header, which can have a VLAN header, and
+ * a frame check sequence (FCS).
+ * The buffer size we share with the device is defined to be ENA_PAGE_SIZE
+ */
+#define ENA_XDP_MAX_MTU (ENA_PAGE_SIZE - ETH_HLEN - ETH_FCS_LEN -	\
+			 VLAN_HLEN - XDP_PACKET_HEADROOM -		\
+			 SKB_DATA_ALIGN(sizeof(struct skb_shared_info)))
+
+#define ENA_IS_XDP_INDEX(adapter, index) (((index) >= (adapter)->xdp_first_ring) && \
+	((index) < (adapter)->xdp_first_ring + (adapter)->xdp_num_queues))
+
+enum ENA_XDP_ACTIONS {
+	ENA_XDP_PASS		= 0,
+	ENA_XDP_TX		= BIT(0),
+	ENA_XDP_REDIRECT	= BIT(1),
+	ENA_XDP_DROP		= BIT(2)
+};
+
+#define ENA_XDP_FORWARDED (ENA_XDP_TX | ENA_XDP_REDIRECT)
+
+int ena_setup_and_create_all_xdp_queues(struct ena_adapter *adapter);
+void ena_xdp_exchange_program_rx_in_range(struct ena_adapter *adapter,
+					  struct bpf_prog *prog,
+					  int first, int count);
+int ena_xdp_io_poll(struct napi_struct *napi, int budget);
+int ena_xdp_xmit_frame(struct ena_ring *tx_ring,
+		       struct ena_adapter *adapter,
+		       struct xdp_frame *xdpf,
+		       int flags);
+int ena_xdp_xmit(struct net_device *dev, int n,
+		 struct xdp_frame **frames, u32 flags);
+int ena_xdp(struct net_device *netdev, struct netdev_bpf *bpf);
+int ena_xdp_register_rxq_info(struct ena_ring *rx_ring);
+void ena_xdp_unregister_rxq_info(struct ena_ring *rx_ring);
+
+enum ena_xdp_errors_t {
+	ENA_XDP_ALLOWED = 0,
+	ENA_XDP_CURRENT_MTU_TOO_LARGE,
+	ENA_XDP_NO_ENOUGH_QUEUES,
+};
+
+static inline bool ena_xdp_present(struct ena_adapter *adapter)
+{
+	return !!adapter->xdp_bpf_prog;
+}
+
+static inline bool ena_xdp_present_ring(struct ena_ring *ring)
+{
+	return !!ring->xdp_bpf_prog;
+}
+
+static inline bool ena_xdp_legal_queue_count(struct ena_adapter *adapter,
+					     u32 queues)
+{
+	return 2 * queues <= adapter->max_num_io_queues;
+}
+
+static inline enum ena_xdp_errors_t ena_xdp_allowed(struct ena_adapter *adapter)
+{
+	enum ena_xdp_errors_t rc = ENA_XDP_ALLOWED;
+
+	if (adapter->netdev->mtu > ENA_XDP_MAX_MTU)
+		rc = ENA_XDP_CURRENT_MTU_TOO_LARGE;
+	else if (!ena_xdp_legal_queue_count(adapter, adapter->num_io_queues))
+		rc = ENA_XDP_NO_ENOUGH_QUEUES;
+
+	return rc;
+}
+
+static inline int ena_xdp_execute(struct ena_ring *rx_ring, struct xdp_buff *xdp)
+{
+	u32 verdict = ENA_XDP_PASS;
+	struct bpf_prog *xdp_prog;
+	struct ena_ring *xdp_ring;
+	struct xdp_frame *xdpf;
+	u64 *xdp_stat;
+
+	xdp_prog = READ_ONCE(rx_ring->xdp_bpf_prog);
+
+	verdict = bpf_prog_run_xdp(xdp_prog, xdp);
+
+	switch (verdict) {
+	case XDP_TX:
+		xdpf = xdp_convert_buff_to_frame(xdp);
+		if (unlikely(!xdpf)) {
+			trace_xdp_exception(rx_ring->netdev, xdp_prog, verdict);
+			xdp_stat = &rx_ring->rx_stats.xdp_aborted;
+			verdict = ENA_XDP_DROP;
+			break;
+		}
+
+		/* Find xmit queue */
+		xdp_ring = rx_ring->xdp_ring;
+
+		/* The XDP queues are shared between XDP_TX and XDP_REDIRECT */
+		spin_lock(&xdp_ring->xdp_tx_lock);
+
+		if (ena_xdp_xmit_frame(xdp_ring, rx_ring->adapter, xdpf,
+				       XDP_XMIT_FLUSH))
+			xdp_return_frame(xdpf);
+
+		spin_unlock(&xdp_ring->xdp_tx_lock);
+		xdp_stat = &rx_ring->rx_stats.xdp_tx;
+		verdict = ENA_XDP_TX;
+		break;
+	case XDP_REDIRECT:
+		if (likely(!xdp_do_redirect(rx_ring->netdev, xdp, xdp_prog))) {
+			xdp_stat = &rx_ring->rx_stats.xdp_redirect;
+			verdict = ENA_XDP_REDIRECT;
+			break;
+		}
+		trace_xdp_exception(rx_ring->netdev, xdp_prog, verdict);
+		xdp_stat = &rx_ring->rx_stats.xdp_aborted;
+		verdict = ENA_XDP_DROP;
+		break;
+	case XDP_ABORTED:
+		trace_xdp_exception(rx_ring->netdev, xdp_prog, verdict);
+		xdp_stat = &rx_ring->rx_stats.xdp_aborted;
+		verdict = ENA_XDP_DROP;
+		break;
+	case XDP_DROP:
+		xdp_stat = &rx_ring->rx_stats.xdp_drop;
+		verdict = ENA_XDP_DROP;
+		break;
+	case XDP_PASS:
+		xdp_stat = &rx_ring->rx_stats.xdp_pass;
+		verdict = ENA_XDP_PASS;
+		break;
+	default:
+		bpf_warn_invalid_xdp_action(rx_ring->netdev, xdp_prog, verdict);
+		xdp_stat = &rx_ring->rx_stats.xdp_invalid;
+		verdict = ENA_XDP_DROP;
+	}
+
+	ena_increase_stat(xdp_stat, 1, &rx_ring->syncp);
+
+	return verdict;
+}
+#endif /* ENA_XDP_H */
diff --git a/drivers/net/ethernet/amd/7990.c b/drivers/net/ethernet/amd/7990.c
index dc57f2759f44..27792a52b6cf 100644
--- a/drivers/net/ethernet/amd/7990.c
+++ b/drivers/net/ethernet/amd/7990.c
@@ -1,3 +1,4 @@
+// SPDX-License-Identifier: GPL-2.0-only
 /*
  * 7990.c -- LANCE ethernet IC generic routines.
  * This is an attempt to separate out the bits of various ethernet
@@ -27,6 +28,7 @@
 #include <linux/route.h>
 #include <linux/string.h>
 #include <linux/skbuff.h>
+#include <linux/pgtable.h>
 #include <asm/irq.h>
 /* Used for the temporal inet entries and routing */
 #include <linux/socket.h>
@@ -34,7 +36,6 @@
 
 #include <asm/io.h>
 #include <asm/dma.h>
-#include <asm/pgtable.h>
 #ifdef CONFIG_HP300
 #include <asm/blinken.h>
 #endif
@@ -526,7 +527,7 @@ int lance_close(struct net_device *dev)
 }
 EXPORT_SYMBOL_GPL(lance_close);
 
-void lance_tx_timeout(struct net_device *dev)
+void lance_tx_timeout(struct net_device *dev, unsigned int txqueue)
 {
 	printk("lance_tx_timeout\n");
 	lance_reset(dev);
@@ -535,7 +536,7 @@ void lance_tx_timeout(struct net_device *dev)
 }
 EXPORT_SYMBOL_GPL(lance_tx_timeout);
 
-int lance_start_xmit(struct sk_buff *skb, struct net_device *dev)
+netdev_tx_t lance_start_xmit(struct sk_buff *skb, struct net_device *dev)
 {
 	struct lance_private *lp = netdev_priv(dev);
 	volatile struct lance_init_block *ib = lp->init_block;
@@ -666,4 +667,5 @@ void lance_poll(struct net_device *dev)
 EXPORT_SYMBOL_GPL(lance_poll);
 #endif
 
+MODULE_DESCRIPTION("LANCE Ethernet IC generic routines");
 MODULE_LICENSE("GPL");
diff --git a/drivers/net/ethernet/amd/7990.h b/drivers/net/ethernet/amd/7990.h
index e9e0be313804..e53551daeea1 100644
--- a/drivers/net/ethernet/amd/7990.h
+++ b/drivers/net/ethernet/amd/7990.h
@@ -1,3 +1,4 @@
+/* SPDX-License-Identifier: GPL-2.0 */
 /*
  * 7990.h -- LANCE ethernet IC generic routines.
  * This is an attempt to separate out the bits of various ethernet
@@ -240,9 +241,9 @@ struct lance_private {
 /* Now the prototypes we export */
 int lance_open(struct net_device *dev);
 int lance_close(struct net_device *dev);
-int lance_start_xmit(struct sk_buff *skb, struct net_device *dev);
+netdev_tx_t lance_start_xmit(struct sk_buff *skb, struct net_device *dev);
 void lance_set_multicast(struct net_device *dev);
-void lance_tx_timeout(struct net_device *dev);
+void lance_tx_timeout(struct net_device *dev, unsigned int txqueue);
 #ifdef CONFIG_NET_POLL_CONTROLLER
 void lance_poll(struct net_device *dev);
 #endif
diff --git a/drivers/net/ethernet/amd/Kconfig b/drivers/net/ethernet/amd/Kconfig
index d5c15e8bb3de..b39c6f3e1eda 100644
--- a/drivers/net/ethernet/amd/Kconfig
+++ b/drivers/net/ethernet/amd/Kconfig
@@ -1,3 +1,4 @@
+# SPDX-License-Identifier: GPL-2.0-only
 #
 # AMD network device configuration
 #
@@ -7,13 +8,13 @@ config NET_VENDOR_AMD
 	default y
 	depends on DIO || MACH_DECSTATION || MVME147 || ATARI || SUN3 || \
 		   SUN3X || SBUS || PCI || ZORRO || (ISA && ISA_DMA_API) || \
-		   (ARM && ARCH_EBSA110) || ISA || EISA || PCMCIA || ARM64
-	---help---
+		   ISA || EISA || PCMCIA || ARM64
+	help
 	  If you have a network (Ethernet) chipset belonging to this class,
 	  say Y.
 
 	  Note that the answer to this question does not directly affect
-	  the kernel: saying N will just case the configurator to skip all
+	  the kernel: saying N will just cause the configurator to skip all
 	  the questions regarding AMD chipsets. If you say Y, you will be asked
 	  for your specific chipset/driver in the following questions.
 
@@ -23,7 +24,7 @@ config A2065
 	tristate "A2065 support"
 	depends on ZORRO
 	select CRC32
-	---help---
+	help
 	  If you have a Commodore A2065 Ethernet adapter, say Y. Otherwise,
 	  say N.
 
@@ -35,7 +36,7 @@ config AMD8111_ETH
 	depends on PCI
 	select CRC32
 	select MII
-	---help---
+	help
 	  If you have an AMD 8111-based PCI LANCE ethernet card,
 	  answer Y here.
 
@@ -44,8 +45,9 @@ config AMD8111_ETH
 
 config LANCE
 	tristate "AMD LANCE and PCnet (AT1500 and NE2100) support"
-	depends on ISA && ISA_DMA_API && !ARM
-	---help---
+	depends on ISA && ISA_DMA_API && !ARM && !PPC32
+	select NETDEV_LEGACY_INIT
+	help
 	  If you have a network (Ethernet) card of this type, say Y here.
 	  Some LinkSys cards are of this type.
 
@@ -54,10 +56,10 @@ config LANCE
 
 config PCNET32
 	tristate "AMD PCnet32 PCI support"
-	depends on PCI
+	depends on PCI && HAS_IOPORT
 	select CRC32
 	select MII
-	---help---
+	help
 	  If you have a PCnet32 or PCnetPCI based network (Ethernet) card,
 	  answer Y here.
 
@@ -67,25 +69,17 @@ config PCNET32
 config ARIADNE
 	tristate "Ariadne support"
 	depends on ZORRO
-	---help---
+	help
 	  If you have a Village Tronic Ariadne Ethernet adapter, say Y.
 	  Otherwise, say N.
 
 	  To compile this driver as a module, choose M here: the module
 	  will be called ariadne.
 
-config ARM_AM79C961A
-	bool "ARM EBSA110 AM79C961A support"
-	depends on ARM && ARCH_EBSA110
-	select CRC32
-	---help---
-	  If you wish to compile a kernel for the EBSA-110, then you should
-	  always answer Y to this.
-
 config ATARILANCE
 	tristate "Atari LANCE support"
 	depends on ATARI
-	---help---
+	help
 	  Say Y to include support for several Atari Ethernet adapters based
 	  on the AMD LANCE chipset: RieblCard (with or without battery), or
 	  PAMCard VME (also the version by Rhotron, with different addresses).
@@ -94,7 +88,7 @@ config DECLANCE
 	tristate "DEC LANCE ethernet controller support"
 	depends on MACH_DECSTATION
 	select CRC32
-	---help---
+	help
 	  This driver is for the series of Ethernet controllers produced by
 	  DEC (now Compaq) based on the AMD LANCE chipset, including the
 	  DEPCA series.  (This chipset is better known via the NE2100 cards.)
@@ -103,7 +97,7 @@ config HPLANCE
 	tristate "HP on-board LANCE support"
 	depends on DIO
 	select CRC32
-	---help---
+	help
 	  If you want to use the builtin "LANCE" Ethernet controller on an
 	  HP300 machine, say Y here.
 
@@ -112,7 +106,7 @@ config MIPS_AU1X00_ENET
 	depends on MIPS_ALCHEMY
 	select PHYLIB
 	select CRC32
-	---help---
+	help
 	  If you have an Alchemy Semi AU1X00 based system
 	  say Y.  Otherwise, say N.
 
@@ -120,7 +114,7 @@ config MVME147_NET
 	tristate "MVME147 (LANCE) Ethernet support"
 	depends on MVME147
 	select CRC32
-	---help---
+	help
 	  Support for the on-board Ethernet interface on the Motorola MVME147
 	  single-board computer.  Say Y here to include the
 	  driver for this chip in your kernel.
@@ -128,7 +122,7 @@ config MVME147_NET
 
 config PCMCIA_NMCLAN
 	tristate "New Media PCMCIA support"
-	depends on PCMCIA
+	depends on PCMCIA && HAS_IOPORT
 	help
 	  Say Y here if you intend to attach a New Media Ethernet or LiveWire
 	  PCMCIA (PC-card) Ethernet card to your computer.
@@ -136,19 +130,10 @@ config PCMCIA_NMCLAN
 	  To compile this driver as a module, choose M here: the module will be
 	  called nmclan_cs.  If unsure, say N.
 
-config NI65
-	tristate "NI6510 support"
-	depends on ISA && ISA_DMA_API && !ARM
-	---help---
-	  If you have a network (Ethernet) card of this type, say Y here.
-
-	  To compile this driver as a module, choose M here. The module
-	  will be called ni65.
-
 config SUN3LANCE
 	tristate "Sun3/Sun3x on-board LANCE support"
 	depends on (SUN3 || SUN3X)
-	---help---
+	help
 	  Most Sun3 and Sun3x motherboards (including the 3/50, 3/60 and 3/80)
 	  featured an AMD LANCE 10Mbit Ethernet controller on board; say Y
 	  here to compile in the Linux driver for this and enable Ethernet.
@@ -162,7 +147,7 @@ config SUNLANCE
 	tristate "Sun LANCE support"
 	depends on SBUS
 	select CRC32
-	---help---
+	help
 	  This driver supports the "le" interface present on all 32-bit Sparc
 	  systems, on some older Ultra systems and as an Sbus option.  These
 	  cards are based on the AMD LANCE chipset, which is better known
@@ -173,14 +158,14 @@ config SUNLANCE
 
 config AMD_XGBE
 	tristate "AMD 10GbE Ethernet driver"
-	depends on ((OF_NET && OF_ADDRESS) || ACPI || PCI) && HAS_IOMEM && HAS_DMA
+	depends on (OF_ADDRESS || ACPI || PCI) && HAS_IOMEM
 	depends on X86 || ARM64 || COMPILE_TEST
+	depends on PTP_1588_CLOCK_OPTIONAL
 	select BITREVERSE
 	select CRC32
 	select PHYLIB
 	select AMD_XGBE_HAVE_ECC if X86
-	imply PTP_1588_CLOCK
-	---help---
+	help
 	  This driver supports the AMD 10GbE Ethernet device found on an
 	  AMD SoC.
 
@@ -191,7 +176,7 @@ config AMD_XGBE_DCB
 	bool "Data Center Bridging (DCB) support"
 	default n
 	depends on AMD_XGBE && DCB
-	---help---
+	help
 	  Say Y here to enable Data Center Bridging (DCB) support in the
 	  driver.
 
@@ -201,4 +186,18 @@ config AMD_XGBE_HAVE_ECC
 	bool
 	default n
 
+config PDS_CORE
+	tristate "AMD/Pensando Data Systems Core Device Support"
+	depends on 64BIT && PCI
+	select AUXILIARY_BUS
+	select NET_DEVLINK
+	help
+	  This enables the support for the AMD/Pensando Core device family of
+	  adapters.  More specific information on this driver can be
+	  found in
+	  <file:Documentation/networking/device_drivers/ethernet/amd/pds_core.rst>.
+
+	  To compile this driver as a module, choose M here. The module
+	  will be called pds_core.
+
 endif # NET_VENDOR_AMD
diff --git a/drivers/net/ethernet/amd/Makefile b/drivers/net/ethernet/amd/Makefile
index a38a2dce3eb3..2dcfb84731e1 100644
--- a/drivers/net/ethernet/amd/Makefile
+++ b/drivers/net/ethernet/amd/Makefile
@@ -1,10 +1,10 @@
+# SPDX-License-Identifier: GPL-2.0
 #
 # Makefile for the AMD network device drivers.
 #
 
 obj-$(CONFIG_A2065) += a2065.o
 obj-$(CONFIG_AMD8111_ETH) += amd8111e.o
-obj-$(CONFIG_ARM_AM79C961A) += am79c961a.o
 obj-$(CONFIG_ARIADNE) += ariadne.o
 obj-$(CONFIG_ATARILANCE) += atarilance.o
 obj-$(CONFIG_DECLANCE) += declance.o
@@ -13,8 +13,8 @@ obj-$(CONFIG_LANCE) += lance.o
 obj-$(CONFIG_MIPS_AU1X00_ENET) += au1000_eth.o
 obj-$(CONFIG_MVME147_NET) += mvme147.o 7990.o
 obj-$(CONFIG_PCMCIA_NMCLAN) += nmclan_cs.o
-obj-$(CONFIG_NI65) += ni65.o
 obj-$(CONFIG_PCNET32) += pcnet32.o
 obj-$(CONFIG_SUN3LANCE) += sun3lance.o
 obj-$(CONFIG_SUNLANCE) += sunlance.o
 obj-$(CONFIG_AMD_XGBE) += xgbe/
+obj-$(CONFIG_PDS_CORE) += pds_core/
diff --git a/drivers/net/ethernet/amd/a2065.c b/drivers/net/ethernet/amd/a2065.c
index ee4b94e3cda9..ce9445425045 100644
--- a/drivers/net/ethernet/amd/a2065.c
+++ b/drivers/net/ethernet/amd/a2065.c
@@ -118,11 +118,8 @@ struct lance_private {
 	int auto_select;	      /* cable-selection by carrier */
 	unsigned short busmaster_regval;
 
-#ifdef CONFIG_SUNLANCE
-	struct Linux_SBus_DMA *ledma; /* if set this points to ledma and arch=4m */
-	int burst_sizes;	      /* ledma SBus burst sizes */
-#endif
 	struct timer_list         multicast_timer;
+	struct net_device	  *dev;
 };
 
 #define LANCE_ADDR(x) ((int)(x) & ~0xff000000)
@@ -489,7 +486,7 @@ static int lance_close(struct net_device *dev)
 	volatile struct lance_regs *ll = lp->ll;
 
 	netif_stop_queue(dev);
-	del_timer_sync(&lp->multicast_timer);
+	timer_delete_sync(&lp->multicast_timer);
 
 	/* Stop the card */
 	ll->rap = LE_CSR0;
@@ -521,7 +518,7 @@ static inline int lance_reset(struct net_device *dev)
 	return status;
 }
 
-static void lance_tx_timeout(struct net_device *dev)
+static void lance_tx_timeout(struct net_device *dev, unsigned int txqueue)
 {
 	struct lance_private *lp = netdev_priv(dev);
 	volatile struct lance_regs *ll = lp->ll;
@@ -550,11 +547,10 @@ static netdev_tx_t lance_start_xmit(struct sk_buff *skb,
 	if (!lance_tx_buffs_avail(lp))
 		goto out_free;
 
-#ifdef DEBUG
 	/* dump the packet */
-	print_hex_dump(KERN_DEBUG, "skb->data: ", DUMP_PREFIX_NONE,
-		       16, 1, skb->data, 64, true);
-#endif
+	print_hex_dump_debug("skb->data: ", DUMP_PREFIX_NONE, 16, 1, skb->data,
+			     64, true);
+
 	entry = lp->tx_new & lp->tx_ring_mod_mask;
 	ib->btx_ring[entry].length = (-skblen) | 0xf000;
 	ib->btx_ring[entry].misc = 0;
@@ -638,12 +634,19 @@ static void lance_set_multicast(struct net_device *dev)
 	netif_wake_queue(dev);
 }
 
+static void lance_set_multicast_retry(struct timer_list *t)
+{
+	struct lance_private *lp = timer_container_of(lp, t, multicast_timer);
+
+	lance_set_multicast(lp->dev);
+}
+
 static int a2065_init_one(struct zorro_dev *z,
 			  const struct zorro_device_id *ent);
 static void a2065_remove_one(struct zorro_dev *z);
 
 
-static struct zorro_device_id a2065_zorro_tbl[] = {
+static const struct zorro_device_id a2065_zorro_tbl[] = {
 	{ ZORRO_PROD_CBM_A2065_1 },
 	{ ZORRO_PROD_CBM_A2065_2 },
 	{ ZORRO_PROD_AMERISTAR_A2065 },
@@ -677,6 +680,7 @@ static int a2065_init_one(struct zorro_dev *z,
 	unsigned long base_addr = board + A2065_LANCE;
 	unsigned long mem_start = board + A2065_RAM;
 	struct resource *r1, *r2;
+	u8 addr[ETH_ALEN];
 	u32 serial;
 	int err;
 
@@ -691,7 +695,7 @@ static int a2065_init_one(struct zorro_dev *z,
 	}
 
 	dev = alloc_etherdev(sizeof(struct lance_private));
-	if (dev == NULL) {
+	if (!dev) {
 		release_mem_region(base_addr, sizeof(struct lance_regs));
 		release_mem_region(mem_start, A2065_RAM_SIZE);
 		return -ENOMEM;
@@ -703,17 +707,18 @@ static int a2065_init_one(struct zorro_dev *z,
 	r2->name = dev->name;
 
 	serial = be32_to_cpu(z->rom.er_SerialNumber);
-	dev->dev_addr[0] = 0x00;
+	addr[0] = 0x00;
 	if (z->id != ZORRO_PROD_AMERISTAR_A2065) {	/* Commodore */
-		dev->dev_addr[1] = 0x80;
-		dev->dev_addr[2] = 0x10;
+		addr[1] = 0x80;
+		addr[2] = 0x10;
 	} else {					/* Ameristar */
-		dev->dev_addr[1] = 0x00;
-		dev->dev_addr[2] = 0x9f;
+		addr[1] = 0x00;
+		addr[2] = 0x9f;
 	}
-	dev->dev_addr[3] = (serial >> 16) & 0xff;
-	dev->dev_addr[4] = (serial >> 8) & 0xff;
-	dev->dev_addr[5] = serial & 0xff;
+	addr[3] = (serial >> 16) & 0xff;
+	addr[4] = (serial >> 8) & 0xff;
+	addr[5] = serial & 0xff;
+	eth_hw_addr_set(dev, addr);
 	dev->base_addr = (unsigned long)ZTWO_VADDR(base_addr);
 	dev->mem_start = (unsigned long)ZTWO_VADDR(mem_start);
 	dev->mem_end = dev->mem_start + A2065_RAM_SIZE;
@@ -728,15 +733,13 @@ static int a2065_init_one(struct zorro_dev *z,
 	priv->lance_log_tx_bufs = LANCE_LOG_TX_BUFFERS;
 	priv->rx_ring_mod_mask = RX_RING_MOD_MASK;
 	priv->tx_ring_mod_mask = TX_RING_MOD_MASK;
+	priv->dev = dev;
 
 	dev->netdev_ops = &lance_netdev_ops;
 	dev->watchdog_timeo = 5*HZ;
 	dev->dma = 0;
 
-	init_timer(&priv->multicast_timer);
-	priv->multicast_timer.data = (unsigned long) dev;
-	priv->multicast_timer.function =
-		(void (*)(unsigned long))lance_set_multicast;
+	timer_setup(&priv->multicast_timer, lance_set_multicast_retry, 0);
 
 	err = register_netdev(dev);
 	if (err) {
@@ -778,4 +781,5 @@ static void __exit a2065_cleanup_module(void)
 module_init(a2065_init_module);
 module_exit(a2065_cleanup_module);
 
+MODULE_DESCRIPTION("Commodore A2065 Ethernet driver");
 MODULE_LICENSE("GPL");
diff --git a/drivers/net/ethernet/amd/am79c961a.c b/drivers/net/ethernet/amd/am79c961a.c
deleted file mode 100644
index b11e910850f7..000000000000
--- a/drivers/net/ethernet/amd/am79c961a.c
+++ /dev/null
@@ -1,767 +0,0 @@
-/*
- *  linux/drivers/net/ethernet/amd/am79c961a.c
- *
- *  by Russell King <rmk@arm.linux.org.uk> 1995-2001.
- *
- * This program is free software; you can redistribute it and/or modify
- * it under the terms of the GNU General Public License version 2 as
- * published by the Free Software Foundation.
- *
- * Derived from various things including skeleton.c
- *
- * This is a special driver for the am79c961A Lance chip used in the
- * Intel (formally Digital Equipment Corp) EBSA110 platform.  Please
- * note that this can not be built as a module (it doesn't make sense).
- */
-#include <linux/kernel.h>
-#include <linux/types.h>
-#include <linux/interrupt.h>
-#include <linux/ioport.h>
-#include <linux/slab.h>
-#include <linux/string.h>
-#include <linux/errno.h>
-#include <linux/netdevice.h>
-#include <linux/etherdevice.h>
-#include <linux/delay.h>
-#include <linux/init.h>
-#include <linux/crc32.h>
-#include <linux/bitops.h>
-#include <linux/platform_device.h>
-#include <linux/io.h>
-
-#include <mach/hardware.h>
-
-#define TX_BUFFERS 15
-#define RX_BUFFERS 25
-
-#include "am79c961a.h"
-
-static irqreturn_t
-am79c961_interrupt (int irq, void *dev_id);
-
-static unsigned int net_debug = NET_DEBUG;
-
-static const char version[] =
-	"am79c961 ethernet driver (C) 1995-2001 Russell King v0.04\n";
-
-/* --------------------------------------------------------------------------- */
-
-#ifdef __arm__
-static void write_rreg(u_long base, u_int reg, u_int val)
-{
-	asm volatile(
-	"strh	%1, [%2]	@ NET_RAP\n\t"
-	"strh	%0, [%2, #-4]	@ NET_RDP"
-	:
-	: "r" (val), "r" (reg), "r" (ISAIO_BASE + 0x0464));
-}
-
-static inline unsigned short read_rreg(u_long base_addr, u_int reg)
-{
-	unsigned short v;
-	asm volatile(
-	"strh	%1, [%2]	@ NET_RAP\n\t"
-	"ldrh	%0, [%2, #-4]	@ NET_RDP"
-	: "=r" (v)
-	: "r" (reg), "r" (ISAIO_BASE + 0x0464));
-	return v;
-}
-
-static inline void write_ireg(u_long base, u_int reg, u_int val)
-{
-	asm volatile(
-	"strh	%1, [%2]	@ NET_RAP\n\t"
-	"strh	%0, [%2, #8]	@ NET_IDP"
-	:
-	: "r" (val), "r" (reg), "r" (ISAIO_BASE + 0x0464));
-}
-
-static inline unsigned short read_ireg(u_long base_addr, u_int reg)
-{
-	u_short v;
-	asm volatile(
-	"strh	%1, [%2]	@ NAT_RAP\n\t"
-	"ldrh	%0, [%2, #8]	@ NET_IDP\n\t"
-	: "=r" (v)
-	: "r" (reg), "r" (ISAIO_BASE + 0x0464));
-	return v;
-}
-
-#define am_writeword(dev,off,val) __raw_writew(val, ISAMEM_BASE + ((off) << 1))
-#define am_readword(dev,off)      __raw_readw(ISAMEM_BASE + ((off) << 1))
-
-static void
-am_writebuffer(struct net_device *dev, u_int offset, unsigned char *buf, unsigned int length)
-{
-	offset = ISAMEM_BASE + (offset << 1);
-	length = (length + 1) & ~1;
-	if ((int)buf & 2) {
-		asm volatile("strh	%2, [%0], #4"
-		 : "=&r" (offset) : "0" (offset), "r" (buf[0] | (buf[1] << 8)));
-		buf += 2;
-		length -= 2;
-	}
-	while (length > 8) {
-		register unsigned int tmp asm("r2"), tmp2 asm("r3");
-		asm volatile(
-			"ldmia	%0!, {%1, %2}"
-			: "+r" (buf), "=&r" (tmp), "=&r" (tmp2));
-		length -= 8;
-		asm volatile(
-			"strh	%1, [%0], #4\n\t"
-			"mov	%1, %1, lsr #16\n\t"
-			"strh	%1, [%0], #4\n\t"
-			"strh	%2, [%0], #4\n\t"
-			"mov	%2, %2, lsr #16\n\t"
-			"strh	%2, [%0], #4"
-		: "+r" (offset), "=&r" (tmp), "=&r" (tmp2));
-	}
-	while (length > 0) {
-		asm volatile("strh	%2, [%0], #4"
-		 : "=&r" (offset) : "0" (offset), "r" (buf[0] | (buf[1] << 8)));
-		buf += 2;
-		length -= 2;
-	}
-}
-
-static void
-am_readbuffer(struct net_device *dev, u_int offset, unsigned char *buf, unsigned int length)
-{
-	offset = ISAMEM_BASE + (offset << 1);
-	length = (length + 1) & ~1;
-	if ((int)buf & 2) {
-		unsigned int tmp;
-		asm volatile(
-			"ldrh	%2, [%0], #4\n\t"
-			"strb	%2, [%1], #1\n\t"
-			"mov	%2, %2, lsr #8\n\t"
-			"strb	%2, [%1], #1"
-		: "=&r" (offset), "=&r" (buf), "=r" (tmp): "0" (offset), "1" (buf));
-		length -= 2;
-	}
-	while (length > 8) {
-		register unsigned int tmp asm("r2"), tmp2 asm("r3"), tmp3;
-		asm volatile(
-			"ldrh	%2, [%0], #4\n\t"
-			"ldrh	%4, [%0], #4\n\t"
-			"ldrh	%3, [%0], #4\n\t"
-			"orr	%2, %2, %4, lsl #16\n\t"
-			"ldrh	%4, [%0], #4\n\t"
-			"orr	%3, %3, %4, lsl #16\n\t"
-			"stmia	%1!, {%2, %3}"
-		: "=&r" (offset), "=&r" (buf), "=r" (tmp), "=r" (tmp2), "=r" (tmp3)
-		: "0" (offset), "1" (buf));
-		length -= 8;
-	}
-	while (length > 0) {
-		unsigned int tmp;
-		asm volatile(
-			"ldrh	%2, [%0], #4\n\t"
-			"strb	%2, [%1], #1\n\t"
-			"mov	%2, %2, lsr #8\n\t"
-			"strb	%2, [%1], #1"
-		: "=&r" (offset), "=&r" (buf), "=r" (tmp) : "0" (offset), "1" (buf));
-		length -= 2;
-	}
-}
-#else
-#error Not compatible
-#endif
-
-static int
-am79c961_ramtest(struct net_device *dev, unsigned int val)
-{
-	unsigned char *buffer = kmalloc (65536, GFP_KERNEL);
-	int i, error = 0, errorcount = 0;
-
-	if (!buffer)
-		return 0;
-	memset (buffer, val, 65536);
-	am_writebuffer(dev, 0, buffer, 65536);
-	memset (buffer, val ^ 255, 65536);
-	am_readbuffer(dev, 0, buffer, 65536);
-	for (i = 0; i < 65536; i++) {
-		if (buffer[i] != val && !error) {
-			printk ("%s: buffer error (%02X %02X) %05X - ", dev->name, val, buffer[i], i);
-			error = 1;
-			errorcount ++;
-		} else if (error && buffer[i] == val) {
-			printk ("%05X\n", i);
-			error = 0;
-		}
-	}
-	if (error)
-		printk ("10000\n");
-	kfree (buffer);
-	return errorcount;
-}
-
-static void am79c961_mc_hash(char *addr, u16 *hash)
-{
-	int idx, bit;
-	u32 crc;
-
-	crc = ether_crc_le(ETH_ALEN, addr);
-
-	idx = crc >> 30;
-	bit = (crc >> 26) & 15;
-
-	hash[idx] |= 1 << bit;
-}
-
-static unsigned int am79c961_get_rx_mode(struct net_device *dev, u16 *hash)
-{
-	unsigned int mode = MODE_PORT_10BT;
-
-	if (dev->flags & IFF_PROMISC) {
-		mode |= MODE_PROMISC;
-		memset(hash, 0xff, 4 * sizeof(*hash));
-	} else if (dev->flags & IFF_ALLMULTI) {
-		memset(hash, 0xff, 4 * sizeof(*hash));
-	} else {
-		struct netdev_hw_addr *ha;
-
-		memset(hash, 0, 4 * sizeof(*hash));
-
-		netdev_for_each_mc_addr(ha, dev)
-			am79c961_mc_hash(ha->addr, hash);
-	}
-
-	return mode;
-}
-
-static void
-am79c961_init_for_open(struct net_device *dev)
-{
-	struct dev_priv *priv = netdev_priv(dev);
-	unsigned long flags;
-	unsigned char *p;
-	u_int hdr_addr, first_free_addr;
-	u16 multi_hash[4], mode = am79c961_get_rx_mode(dev, multi_hash);
-	int i;
-
-	/*
-	 * Stop the chip.
-	 */
-	spin_lock_irqsave(&priv->chip_lock, flags);
-	write_rreg (dev->base_addr, CSR0, CSR0_BABL|CSR0_CERR|CSR0_MISS|CSR0_MERR|CSR0_TINT|CSR0_RINT|CSR0_STOP);
-	spin_unlock_irqrestore(&priv->chip_lock, flags);
-
-	write_ireg (dev->base_addr, 5, 0x00a0); /* Receive address LED */
-	write_ireg (dev->base_addr, 6, 0x0081); /* Collision LED */
-	write_ireg (dev->base_addr, 7, 0x0090); /* XMIT LED */
-	write_ireg (dev->base_addr, 2, 0x0000); /* MODE register selects media */
-
-	for (i = LADRL; i <= LADRH; i++)
-		write_rreg (dev->base_addr, i, multi_hash[i - LADRL]);
-
-	for (i = PADRL, p = dev->dev_addr; i <= PADRH; i++, p += 2)
-		write_rreg (dev->base_addr, i, p[0] | (p[1] << 8));
-
-	write_rreg (dev->base_addr, MODE, mode);
-	write_rreg (dev->base_addr, POLLINT, 0);
-	write_rreg (dev->base_addr, SIZERXR, -RX_BUFFERS);
-	write_rreg (dev->base_addr, SIZETXR, -TX_BUFFERS);
-
-	first_free_addr = RX_BUFFERS * 8 + TX_BUFFERS * 8 + 16;
-	hdr_addr = 0;
-
-	priv->rxhead = 0;
-	priv->rxtail = 0;
-	priv->rxhdr = hdr_addr;
-
-	for (i = 0; i < RX_BUFFERS; i++) {
-		priv->rxbuffer[i] = first_free_addr;
-		am_writeword (dev, hdr_addr, first_free_addr);
-		am_writeword (dev, hdr_addr + 2, RMD_OWN);
-		am_writeword (dev, hdr_addr + 4, (-1600));
-		am_writeword (dev, hdr_addr + 6, 0);
-		first_free_addr += 1600;
-		hdr_addr += 8;
-	}
-	priv->txhead = 0;
-	priv->txtail = 0;
-	priv->txhdr = hdr_addr;
-	for (i = 0; i < TX_BUFFERS; i++) {
-		priv->txbuffer[i] = first_free_addr;
-		am_writeword (dev, hdr_addr, first_free_addr);
-		am_writeword (dev, hdr_addr + 2, TMD_STP|TMD_ENP);
-		am_writeword (dev, hdr_addr + 4, 0xf000);
-		am_writeword (dev, hdr_addr + 6, 0);
-		first_free_addr += 1600;
-		hdr_addr += 8;
-	}
-
-	write_rreg (dev->base_addr, BASERXL, priv->rxhdr);
-	write_rreg (dev->base_addr, BASERXH, 0);
-	write_rreg (dev->base_addr, BASETXL, priv->txhdr);
-	write_rreg (dev->base_addr, BASERXH, 0);
-	write_rreg (dev->base_addr, CSR0, CSR0_STOP);
-	write_rreg (dev->base_addr, CSR3, CSR3_IDONM|CSR3_BABLM|CSR3_DXSUFLO);
-	write_rreg (dev->base_addr, CSR4, CSR4_APAD_XMIT|CSR4_MFCOM|CSR4_RCVCCOM|CSR4_TXSTRTM|CSR4_JABM);
-	write_rreg (dev->base_addr, CSR0, CSR0_IENA|CSR0_STRT);
-}
-
-static void am79c961_timer(unsigned long data)
-{
-	struct net_device *dev = (struct net_device *)data;
-	struct dev_priv *priv = netdev_priv(dev);
-	unsigned int lnkstat, carrier;
-	unsigned long flags;
-
-	spin_lock_irqsave(&priv->chip_lock, flags);
-	lnkstat = read_ireg(dev->base_addr, ISALED0) & ISALED0_LNKST;
-	spin_unlock_irqrestore(&priv->chip_lock, flags);
-	carrier = netif_carrier_ok(dev);
-
-	if (lnkstat && !carrier) {
-		netif_carrier_on(dev);
-		printk("%s: link up\n", dev->name);
-	} else if (!lnkstat && carrier) {
-		netif_carrier_off(dev);
-		printk("%s: link down\n", dev->name);
-	}
-
-	mod_timer(&priv->timer, jiffies + msecs_to_jiffies(500));
-}
-
-/*
- * Open/initialize the board.
- */
-static int
-am79c961_open(struct net_device *dev)
-{
-	struct dev_priv *priv = netdev_priv(dev);
-	int ret;
-
-	ret = request_irq(dev->irq, am79c961_interrupt, 0, dev->name, dev);
-	if (ret)
-		return ret;
-
-	am79c961_init_for_open(dev);
-
-	netif_carrier_off(dev);
-
-	priv->timer.expires = jiffies;
-	add_timer(&priv->timer);
-
-	netif_start_queue(dev);
-
-	return 0;
-}
-
-/*
- * The inverse routine to am79c961_open().
- */
-static int
-am79c961_close(struct net_device *dev)
-{
-	struct dev_priv *priv = netdev_priv(dev);
-	unsigned long flags;
-
-	del_timer_sync(&priv->timer);
-
-	netif_stop_queue(dev);
-	netif_carrier_off(dev);
-
-	spin_lock_irqsave(&priv->chip_lock, flags);
-	write_rreg (dev->base_addr, CSR0, CSR0_STOP);
-	write_rreg (dev->base_addr, CSR3, CSR3_MASKALL);
-	spin_unlock_irqrestore(&priv->chip_lock, flags);
-
-	free_irq (dev->irq, dev);
-
-	return 0;
-}
-
-/*
- * Set or clear promiscuous/multicast mode filter for this adapter.
- */
-static void am79c961_setmulticastlist (struct net_device *dev)
-{
-	struct dev_priv *priv = netdev_priv(dev);
-	unsigned long flags;
-	u16 multi_hash[4], mode = am79c961_get_rx_mode(dev, multi_hash);
-	int i, stopped;
-
-	spin_lock_irqsave(&priv->chip_lock, flags);
-
-	stopped = read_rreg(dev->base_addr, CSR0) & CSR0_STOP;
-
-	if (!stopped) {
-		/*
-		 * Put the chip into suspend mode
-		 */
-		write_rreg(dev->base_addr, CTRL1, CTRL1_SPND);
-
-		/*
-		 * Spin waiting for chip to report suspend mode
-		 */
-		while ((read_rreg(dev->base_addr, CTRL1) & CTRL1_SPND) == 0) {
-			spin_unlock_irqrestore(&priv->chip_lock, flags);
-			nop();
-			spin_lock_irqsave(&priv->chip_lock, flags);
-		}
-	}
-
-	/*
-	 * Update the multicast hash table
-	 */
-	for (i = 0; i < ARRAY_SIZE(multi_hash); i++)
-		write_rreg(dev->base_addr, i + LADRL, multi_hash[i]);
-
-	/*
-	 * Write the mode register
-	 */
-	write_rreg(dev->base_addr, MODE, mode);
-
-	if (!stopped) {
-		/*
-		 * Put the chip back into running mode
-		 */
-		write_rreg(dev->base_addr, CTRL1, 0);
-	}
-
-	spin_unlock_irqrestore(&priv->chip_lock, flags);
-}
-
-static void am79c961_timeout(struct net_device *dev)
-{
-	printk(KERN_WARNING "%s: transmit timed out, network cable problem?\n",
-		dev->name);
-
-	/*
-	 * ought to do some setup of the tx side here
-	 */
-
-	netif_wake_queue(dev);
-}
-
-/*
- * Transmit a packet
- */
-static int
-am79c961_sendpacket(struct sk_buff *skb, struct net_device *dev)
-{
-	struct dev_priv *priv = netdev_priv(dev);
-	unsigned int hdraddr, bufaddr;
-	unsigned int head;
-	unsigned long flags;
-
-	head = priv->txhead;
-	hdraddr = priv->txhdr + (head << 3);
-	bufaddr = priv->txbuffer[head];
-	head += 1;
-	if (head >= TX_BUFFERS)
-		head = 0;
-
-	am_writebuffer (dev, bufaddr, skb->data, skb->len);
-	am_writeword (dev, hdraddr + 4, -skb->len);
-	am_writeword (dev, hdraddr + 2, TMD_OWN|TMD_STP|TMD_ENP);
-	priv->txhead = head;
-
-	spin_lock_irqsave(&priv->chip_lock, flags);
-	write_rreg (dev->base_addr, CSR0, CSR0_TDMD|CSR0_IENA);
-	spin_unlock_irqrestore(&priv->chip_lock, flags);
-
-	/*
-	 * If the next packet is owned by the ethernet device,
-	 * then the tx ring is full and we can't add another
-	 * packet.
-	 */
-	if (am_readword(dev, priv->txhdr + (priv->txhead << 3) + 2) & TMD_OWN)
-		netif_stop_queue(dev);
-
-	dev_consume_skb_any(skb);
-
-	return NETDEV_TX_OK;
-}
-
-/*
- * If we have a good packet(s), get it/them out of the buffers.
- */
-static void
-am79c961_rx(struct net_device *dev, struct dev_priv *priv)
-{
-	do {
-		struct sk_buff *skb;
-		u_int hdraddr;
-		u_int pktaddr;
-		u_int status;
-		int len;
-
-		hdraddr = priv->rxhdr + (priv->rxtail << 3);
-		pktaddr = priv->rxbuffer[priv->rxtail];
-
-		status = am_readword (dev, hdraddr + 2);
-		if (status & RMD_OWN) /* do we own it? */
-			break;
-
-		priv->rxtail ++;
-		if (priv->rxtail >= RX_BUFFERS)
-			priv->rxtail = 0;
-
-		if ((status & (RMD_ERR|RMD_STP|RMD_ENP)) != (RMD_STP|RMD_ENP)) {
-			am_writeword (dev, hdraddr + 2, RMD_OWN);
-			dev->stats.rx_errors++;
-			if (status & RMD_ERR) {
-				if (status & RMD_FRAM)
-					dev->stats.rx_frame_errors++;
-				if (status & RMD_CRC)
-					dev->stats.rx_crc_errors++;
-			} else if (status & RMD_STP)
-				dev->stats.rx_length_errors++;
-			continue;
-		}
-
-		len = am_readword(dev, hdraddr + 6);
-		skb = netdev_alloc_skb(dev, len + 2);
-
-		if (skb) {
-			skb_reserve(skb, 2);
-
-			am_readbuffer(dev, pktaddr, skb_put(skb, len), len);
-			am_writeword(dev, hdraddr + 2, RMD_OWN);
-			skb->protocol = eth_type_trans(skb, dev);
-			netif_rx(skb);
-			dev->stats.rx_bytes += len;
-			dev->stats.rx_packets++;
-		} else {
-			am_writeword (dev, hdraddr + 2, RMD_OWN);
-			dev->stats.rx_dropped++;
-			break;
-		}
-	} while (1);
-}
-
-/*
- * Update stats for the transmitted packet
- */
-static void
-am79c961_tx(struct net_device *dev, struct dev_priv *priv)
-{
-	do {
-		short len;
-		u_int hdraddr;
-		u_int status;
-
-		hdraddr = priv->txhdr + (priv->txtail << 3);
-		status = am_readword (dev, hdraddr + 2);
-		if (status & TMD_OWN)
-			break;
-
-		priv->txtail ++;
-		if (priv->txtail >= TX_BUFFERS)
-			priv->txtail = 0;
-
-		if (status & TMD_ERR) {
-			u_int status2;
-
-			dev->stats.tx_errors++;
-
-			status2 = am_readword (dev, hdraddr + 6);
-
-			/*
-			 * Clear the error byte
-			 */
-			am_writeword (dev, hdraddr + 6, 0);
-
-			if (status2 & TST_RTRY)
-				dev->stats.collisions += 16;
-			if (status2 & TST_LCOL)
-				dev->stats.tx_window_errors++;
-			if (status2 & TST_LCAR)
-				dev->stats.tx_carrier_errors++;
-			if (status2 & TST_UFLO)
-				dev->stats.tx_fifo_errors++;
-			continue;
-		}
-		dev->stats.tx_packets++;
-		len = am_readword (dev, hdraddr + 4);
-		dev->stats.tx_bytes += -len;
-	} while (priv->txtail != priv->txhead);
-
-	netif_wake_queue(dev);
-}
-
-static irqreturn_t
-am79c961_interrupt(int irq, void *dev_id)
-{
-	struct net_device *dev = (struct net_device *)dev_id;
-	struct dev_priv *priv = netdev_priv(dev);
-	u_int status, n = 100;
-	int handled = 0;
-
-	do {
-		status = read_rreg(dev->base_addr, CSR0);
-		write_rreg(dev->base_addr, CSR0, status &
-			   (CSR0_IENA|CSR0_TINT|CSR0_RINT|
-			    CSR0_MERR|CSR0_MISS|CSR0_CERR|CSR0_BABL));
-
-		if (status & CSR0_RINT) {
-			handled = 1;
-			am79c961_rx(dev, priv);
-		}
-		if (status & CSR0_TINT) {
-			handled = 1;
-			am79c961_tx(dev, priv);
-		}
-		if (status & CSR0_MISS) {
-			handled = 1;
-			dev->stats.rx_dropped++;
-		}
-		if (status & CSR0_CERR) {
-			handled = 1;
-			mod_timer(&priv->timer, jiffies);
-		}
-	} while (--n && status & (CSR0_RINT | CSR0_TINT));
-
-	return IRQ_RETVAL(handled);
-}
-
-#ifdef CONFIG_NET_POLL_CONTROLLER
-static void am79c961_poll_controller(struct net_device *dev)
-{
-	unsigned long flags;
-	local_irq_save(flags);
-	am79c961_interrupt(dev->irq, dev);
-	local_irq_restore(flags);
-}
-#endif
-
-/*
- * Initialise the chip.  Note that we always expect
- * to be entered with interrupts enabled.
- */
-static int
-am79c961_hw_init(struct net_device *dev)
-{
-	struct dev_priv *priv = netdev_priv(dev);
-
-	spin_lock_irq(&priv->chip_lock);
-	write_rreg (dev->base_addr, CSR0, CSR0_STOP);
-	write_rreg (dev->base_addr, CSR3, CSR3_MASKALL);
-	spin_unlock_irq(&priv->chip_lock);
-
-	am79c961_ramtest(dev, 0x66);
-	am79c961_ramtest(dev, 0x99);
-
-	return 0;
-}
-
-static void __init am79c961_banner(void)
-{
-	static unsigned version_printed;
-
-	if (net_debug && version_printed++ == 0)
-		printk(KERN_INFO "%s", version);
-}
-static const struct net_device_ops am79c961_netdev_ops = {
-	.ndo_open		= am79c961_open,
-	.ndo_stop		= am79c961_close,
-	.ndo_start_xmit		= am79c961_sendpacket,
-	.ndo_set_rx_mode	= am79c961_setmulticastlist,
-	.ndo_tx_timeout		= am79c961_timeout,
-	.ndo_validate_addr	= eth_validate_addr,
-	.ndo_set_mac_address	= eth_mac_addr,
-#ifdef CONFIG_NET_POLL_CONTROLLER
-	.ndo_poll_controller	= am79c961_poll_controller,
-#endif
-};
-
-static int am79c961_probe(struct platform_device *pdev)
-{
-	struct resource *res;
-	struct net_device *dev;
-	struct dev_priv *priv;
-	int i, ret;
-
-	res = platform_get_resource(pdev, IORESOURCE_IO, 0);
-	if (!res)
-		return -ENODEV;
-
-	dev = alloc_etherdev(sizeof(struct dev_priv));
-	ret = -ENOMEM;
-	if (!dev)
-		goto out;
-
-	SET_NETDEV_DEV(dev, &pdev->dev);
-
-	priv = netdev_priv(dev);
-
-	/*
-	 * Fixed address and IRQ lines here.
-	 * The PNP initialisation should have been
-	 * done by the ether bootp loader.
-	 */
-	dev->base_addr = res->start;
-	ret = platform_get_irq(pdev, 0);
-
-	if (ret < 0) {
-		ret = -ENODEV;
-		goto nodev;
-	}
-	dev->irq = ret;
-
-	ret = -ENODEV;
-	if (!request_region(dev->base_addr, 0x18, dev->name))
-		goto nodev;
-
-	/*
-	 * Reset the device.
-	 */
-	inb(dev->base_addr + NET_RESET);
-	udelay(5);
-
-	/*
-	 * Check the manufacturer part of the
-	 * ether address.
-	 */
-	if (inb(dev->base_addr) != 0x08 ||
-	    inb(dev->base_addr + 2) != 0x00 ||
-	    inb(dev->base_addr + 4) != 0x2b)
-	    	goto release;
-
-	for (i = 0; i < 6; i++)
-		dev->dev_addr[i] = inb(dev->base_addr + i * 2) & 0xff;
-
-	am79c961_banner();
-
-	spin_lock_init(&priv->chip_lock);
-	init_timer(&priv->timer);
-	priv->timer.data = (unsigned long)dev;
-	priv->timer.function = am79c961_timer;
-
-	if (am79c961_hw_init(dev))
-		goto release;
-
-	dev->netdev_ops = &am79c961_netdev_ops;
-
-	ret = register_netdev(dev);
-	if (ret == 0) {
-		printk(KERN_INFO "%s: ether address %pM\n",
-		       dev->name, dev->dev_addr);
-		return 0;
-	}
-
-release:
-	release_region(dev->base_addr, 0x18);
-nodev:
-	free_netdev(dev);
-out:
-	return ret;
-}
-
-static struct platform_driver am79c961_driver = {
-	.probe		= am79c961_probe,
-	.driver		= {
-		.name	= "am79c961",
-	},
-};
-
-static int __init am79c961_init(void)
-{
-	return platform_driver_register(&am79c961_driver);
-}
-
-__initcall(am79c961_init);
diff --git a/drivers/net/ethernet/amd/am79c961a.h b/drivers/net/ethernet/amd/am79c961a.h
deleted file mode 100644
index 9f384b79507b..000000000000
--- a/drivers/net/ethernet/amd/am79c961a.h
+++ /dev/null
@@ -1,145 +0,0 @@
-/*
- * linux/drivers/net/ethernet/amd/am79c961a.h
- *
- * This program is free software; you can redistribute it and/or modify
- * it under the terms of the GNU General Public License version 2 as
- * published by the Free Software Foundation.
- */
-
-#ifndef _LINUX_am79c961a_H
-#define _LINUX_am79c961a_H
-
-/* use 0 for production, 1 for verification, >2 for debug. debug flags: */
-#define DEBUG_TX	 2
-#define DEBUG_RX	 4
-#define DEBUG_INT	 8
-#define DEBUG_IC	16
-#ifndef NET_DEBUG
-#define NET_DEBUG 	0
-#endif
-
-#define NET_UID		0
-#define NET_RDP		0x10
-#define NET_RAP		0x12
-#define NET_RESET	0x14
-#define NET_IDP		0x16
-
-/*
- * RAP registers
- */
-#define CSR0		0
-#define CSR0_INIT	0x0001
-#define CSR0_STRT	0x0002
-#define CSR0_STOP	0x0004
-#define CSR0_TDMD	0x0008
-#define CSR0_TXON	0x0010
-#define CSR0_RXON	0x0020
-#define CSR0_IENA	0x0040
-#define CSR0_INTR	0x0080
-#define CSR0_IDON	0x0100
-#define CSR0_TINT	0x0200
-#define CSR0_RINT	0x0400
-#define CSR0_MERR	0x0800
-#define CSR0_MISS	0x1000
-#define CSR0_CERR	0x2000
-#define CSR0_BABL	0x4000
-#define CSR0_ERR	0x8000
-
-#define CSR3		3
-#define CSR3_EMBA	0x0008
-#define CSR3_DXMT2PD	0x0010
-#define CSR3_LAPPEN	0x0020
-#define CSR3_DXSUFLO	0x0040
-#define CSR3_IDONM	0x0100
-#define CSR3_TINTM	0x0200
-#define CSR3_RINTM	0x0400
-#define CSR3_MERRM	0x0800
-#define CSR3_MISSM	0x1000
-#define CSR3_BABLM	0x4000
-#define CSR3_MASKALL	0x5F00
-
-#define CSR4		4
-#define CSR4_JABM	0x0001
-#define CSR4_JAB	0x0002
-#define CSR4_TXSTRTM	0x0004
-#define CSR4_TXSTRT	0x0008
-#define CSR4_RCVCCOM	0x0010
-#define CSR4_RCVCCO	0x0020
-#define CSR4_MFCOM	0x0100
-#define CSR4_MFCO	0x0200
-#define CSR4_ASTRP_RCV	0x0400
-#define CSR4_APAD_XMIT	0x0800
-
-#define CTRL1		5
-#define CTRL1_SPND	0x0001
-
-#define LADRL		8
-#define LADRM1		9
-#define LADRM2		10
-#define LADRH		11
-#define PADRL		12
-#define PADRM		13
-#define PADRH		14
-
-#define MODE		15
-#define MODE_DISRX	0x0001
-#define MODE_DISTX	0x0002
-#define MODE_LOOP	0x0004
-#define MODE_DTCRC	0x0008
-#define MODE_COLL	0x0010
-#define MODE_DRETRY	0x0020
-#define MODE_INTLOOP	0x0040
-#define MODE_PORT_AUI	0x0000
-#define MODE_PORT_10BT	0x0080
-#define MODE_DRXPA	0x2000
-#define MODE_DRXBA	0x4000
-#define MODE_PROMISC	0x8000
-
-#define BASERXL		24
-#define BASERXH		25
-#define BASETXL		30
-#define BASETXH		31
-
-#define POLLINT		47
-
-#define SIZERXR		76
-#define SIZETXR		78
-
-#define CSR_MFC		112
-
-#define RMD_ENP		0x0100
-#define RMD_STP		0x0200
-#define RMD_CRC		0x0800
-#define RMD_FRAM	0x2000
-#define RMD_ERR		0x4000
-#define RMD_OWN		0x8000
-
-#define TMD_ENP		0x0100
-#define TMD_STP		0x0200
-#define TMD_MORE	0x1000
-#define TMD_ERR		0x4000
-#define TMD_OWN		0x8000
-
-#define TST_RTRY	0x0400
-#define TST_LCAR	0x0800
-#define TST_LCOL	0x1000
-#define TST_UFLO	0x4000
-#define TST_BUFF	0x8000
-
-#define ISALED0		0x0004
-#define ISALED0_LNKST	0x8000
-
-struct dev_priv {
-    unsigned long	rxbuffer[RX_BUFFERS];
-    unsigned long	txbuffer[TX_BUFFERS];
-    unsigned char	txhead;
-    unsigned char	txtail;
-    unsigned char	rxhead;
-    unsigned char	rxtail;
-    unsigned long	rxhdr;
-    unsigned long	txhdr;
-    spinlock_t		chip_lock;
-    struct timer_list	timer;
-};
-
-#endif
diff --git a/drivers/net/ethernet/amd/amd8111e.c b/drivers/net/ethernet/amd/amd8111e.c
index 7b5df562f30f..76e8c13d5985 100644
--- a/drivers/net/ethernet/amd/amd8111e.c
+++ b/drivers/net/ethernet/amd/amd8111e.c
@@ -1,8 +1,8 @@
+// SPDX-License-Identifier: GPL-2.0-or-later
 
 /* Advanced  Micro Devices Inc. AMD8111E Linux Network Driver
  * Copyright (C) 2004 Advanced Micro Devices
  *
- *
  * Copyright 2001,2002 Jeff Garzik <jgarzik@mandrakesoft.com> [ 8139cp.c,tg3.c ]
  * Copyright (C) 2001, 2002 David S. Miller (davem@redhat.com)[ tg3.c]
  * Copyright 1996-1999 Thomas Bogendoerfer [ pcnet32.c ]
@@ -12,19 +12,6 @@
  * Carsten Langgaard, carstenl@mips.com [ pcnet32.c ]
  * Copyright (C) 2000 MIPS Technologies, Inc.  All rights reserved.
  *
- *
- * This program is free software; you can redistribute it and/or modify
- * it under the terms of the GNU General Public License as published by
- * the Free Software Foundation; either version 2 of the License, or
- * (at your option) any later version.
- *
- * This program is distributed in the hope that it will be useful,
- * but WITHOUT ANY WARRANTY; without even the implied warranty of
- * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
- * GNU General Public License for more details.
- *
- * You should have received a copy of the GNU General Public License
- * along with this program; if not, see <http://www.gnu.org/licenses/>.
 
 Module Name:
 
@@ -32,14 +19,14 @@ Module Name:
 
 Abstract:
 
- 	 AMD8111 based 10/100 Ethernet Controller Driver.
+	 AMD8111 based 10/100 Ethernet Controller Driver.
 
 Environment:
 
 	Kernel Mode
 
 Revision History:
- 	3.0.0
+	3.0.0
 	   Initial Revision.
 	3.0.1
 	 1. Dynamic interrupt coalescing.
@@ -56,7 +43,7 @@ Revision History:
 	3.0.4 12/09/2003
 	 1. Added set_mac_address routine for bonding driver support.
 	 2. Tested the driver for bonding support
-	 3. Bug fix: Fixed mismach in actual receive buffer lenth and lenth
+	 3. Bug fix: Fixed mismach in actual receive buffer length and length
 	    indicated to the h/w.
 	 4. Modified amd8111e_rx() routine to receive all the received packets
 	    in the first interrupt.
@@ -97,9 +84,8 @@ Revision History:
 
 #include "amd8111e.h"
 #define MODULE_NAME	"amd8111e"
-#define MODULE_VERS	"3.0.7"
 MODULE_AUTHOR("Advanced Micro Devices, Inc.");
-MODULE_DESCRIPTION ("AMD8111 based 10/100 Ethernet Controller. Driver Version "MODULE_VERS);
+MODULE_DESCRIPTION("AMD8111 based 10/100 Ethernet Controller.");
 MODULE_LICENSE("GPL");
 module_param_array(speed_duplex, int, NULL, 0);
 MODULE_PARM_DESC(speed_duplex, "Set device speed and duplex modes, 0: Auto Negotiate, 1: 10Mbps Half Duplex, 2: 10Mbps Full Duplex, 3: 100Mbps Half Duplex, 4: 100Mbps Full Duplex");
@@ -114,19 +100,19 @@ static int amd8111e_read_phy(struct amd8111e_priv *lp,
 {
 	void __iomem *mmio = lp->mmio;
 	unsigned int reg_val;
-	unsigned int repeat= REPEAT_CNT;
+	unsigned int repeat = REPEAT_CNT;
 
 	reg_val = readl(mmio + PHY_ACCESS);
 	while (reg_val & PHY_CMD_ACTIVE)
-		reg_val = readl( mmio + PHY_ACCESS );
+		reg_val = readl(mmio + PHY_ACCESS);
 
-	writel( PHY_RD_CMD | ((phy_id & 0x1f) << 21) |
-			   ((reg & 0x1f) << 16),  mmio +PHY_ACCESS);
-	do{
+	writel(PHY_RD_CMD | ((phy_id & 0x1f) << 21) |
+			   ((reg & 0x1f) << 16), mmio + PHY_ACCESS);
+	do {
 		reg_val = readl(mmio + PHY_ACCESS);
 		udelay(30);  /* It takes 30 us to read/write data */
 	} while (--repeat && (reg_val & PHY_CMD_ACTIVE));
-	if(reg_val & PHY_RD_ERR)
+	if (reg_val & PHY_RD_ERR)
 		goto err_phy_read;
 
 	*val = reg_val & 0xffff;
@@ -147,17 +133,17 @@ static int amd8111e_write_phy(struct amd8111e_priv *lp,
 
 	reg_val = readl(mmio + PHY_ACCESS);
 	while (reg_val & PHY_CMD_ACTIVE)
-		reg_val = readl( mmio + PHY_ACCESS );
+		reg_val = readl(mmio + PHY_ACCESS);
 
-	writel( PHY_WR_CMD | ((phy_id & 0x1f) << 21) |
+	writel(PHY_WR_CMD | ((phy_id & 0x1f) << 21) |
 			   ((reg & 0x1f) << 16)|val, mmio + PHY_ACCESS);
 
-	do{
+	do {
 		reg_val = readl(mmio + PHY_ACCESS);
 		udelay(30);  /* It takes 30 us to read/write the data */
 	} while (--repeat && (reg_val & PHY_CMD_ACTIVE));
 
-	if(reg_val & PHY_RD_ERR)
+	if (reg_val & PHY_RD_ERR)
 		goto err_phy_write;
 
 	return 0;
@@ -173,7 +159,7 @@ static int amd8111e_mdio_read(struct net_device *dev, int phy_id, int reg_num)
 	struct amd8111e_priv *lp = netdev_priv(dev);
 	unsigned int reg_val;
 
-	amd8111e_read_phy(lp,phy_id,reg_num,&reg_val);
+	amd8111e_read_phy(lp, phy_id, reg_num, &reg_val);
 	return reg_val;
 
 }
@@ -193,30 +179,29 @@ static void amd8111e_mdio_write(struct net_device *dev,
 static void amd8111e_set_ext_phy(struct net_device *dev)
 {
 	struct amd8111e_priv *lp = netdev_priv(dev);
-	u32 bmcr,advert,tmp;
+	u32 bmcr, advert, tmp;
 
 	/* Determine mii register values to set the speed */
 	advert = amd8111e_mdio_read(dev, lp->ext_phy_addr, MII_ADVERTISE);
 	tmp = advert & ~(ADVERTISE_ALL | ADVERTISE_100BASE4);
-	switch (lp->ext_phy_option){
-
-		default:
-		case SPEED_AUTONEG: /* advertise all values */
-			tmp |= ( ADVERTISE_10HALF|ADVERTISE_10FULL|
-				ADVERTISE_100HALF|ADVERTISE_100FULL) ;
-			break;
-		case SPEED10_HALF:
-			tmp |= ADVERTISE_10HALF;
-			break;
-		case SPEED10_FULL:
-			tmp |= ADVERTISE_10FULL;
-			break;
-		case SPEED100_HALF:
-			tmp |= ADVERTISE_100HALF;
-			break;
-		case SPEED100_FULL:
-			tmp |= ADVERTISE_100FULL;
-			break;
+	switch (lp->ext_phy_option) {
+	default:
+	case SPEED_AUTONEG: /* advertise all values */
+		tmp |= (ADVERTISE_10HALF | ADVERTISE_10FULL |
+			ADVERTISE_100HALF | ADVERTISE_100FULL);
+		break;
+	case SPEED10_HALF:
+		tmp |= ADVERTISE_10HALF;
+		break;
+	case SPEED10_FULL:
+		tmp |= ADVERTISE_10FULL;
+		break;
+	case SPEED100_HALF:
+		tmp |= ADVERTISE_100HALF;
+		break;
+	case SPEED100_FULL:
+		tmp |= ADVERTISE_100FULL;
+		break;
 	}
 
 	if(advert != tmp)
@@ -238,20 +223,23 @@ static int amd8111e_free_skbs(struct net_device *dev)
 	int i;
 
 	/* Freeing transmit skbs */
-	for(i = 0; i < NUM_TX_BUFFERS; i++){
-		if(lp->tx_skbuff[i]){
-			pci_unmap_single(lp->pci_dev,lp->tx_dma_addr[i],					lp->tx_skbuff[i]->len,PCI_DMA_TODEVICE);
-			dev_kfree_skb (lp->tx_skbuff[i]);
+	for (i = 0; i < NUM_TX_BUFFERS; i++) {
+		if (lp->tx_skbuff[i]) {
+			dma_unmap_single(&lp->pci_dev->dev,
+					 lp->tx_dma_addr[i],
+					 lp->tx_skbuff[i]->len, DMA_TO_DEVICE);
+			dev_kfree_skb(lp->tx_skbuff[i]);
 			lp->tx_skbuff[i] = NULL;
 			lp->tx_dma_addr[i] = 0;
 		}
 	}
 	/* Freeing previously allocated receive buffers */
-	for (i = 0; i < NUM_RX_BUFFERS; i++){
+	for (i = 0; i < NUM_RX_BUFFERS; i++) {
 		rx_skbuff = lp->rx_skbuff[i];
-		if(rx_skbuff != NULL){
-			pci_unmap_single(lp->pci_dev,lp->rx_dma_addr[i],
-				  lp->rx_buff_len - 2,PCI_DMA_FROMDEVICE);
+		if (rx_skbuff) {
+			dma_unmap_single(&lp->pci_dev->dev,
+					 lp->rx_dma_addr[i],
+					 lp->rx_buff_len - 2, DMA_FROM_DEVICE);
 			dev_kfree_skb(lp->rx_skbuff[i]);
 			lp->rx_skbuff[i] = NULL;
 			lp->rx_dma_addr[i] = 0;
@@ -269,13 +257,13 @@ static inline void amd8111e_set_rx_buff_len(struct net_device *dev)
 	struct amd8111e_priv *lp = netdev_priv(dev);
 	unsigned int mtu = dev->mtu;
 
-	if (mtu > ETH_DATA_LEN){
+	if (mtu > ETH_DATA_LEN) {
 		/* MTU + ethernet header + FCS
 		 * + optional VLAN tag + skb reserve space 2
 		 */
 		lp->rx_buff_len = mtu + ETH_HLEN + 10;
 		lp->options |= OPTION_JUMBO_ENABLE;
-	} else{
+	} else {
 		lp->rx_buff_len = PKT_BUFF_SZ;
 		lp->options &= ~OPTION_JUMBO_ENABLE;
 	}
@@ -296,25 +284,25 @@ static int amd8111e_init_ring(struct net_device *dev)
 	lp->tx_ring_idx = 0;
 
 
-	if(lp->opened)
+	if (lp->opened)
 		/* Free previously allocated transmit and receive skbs */
 		amd8111e_free_skbs(dev);
 
-	else{
-		 /* allocate the tx and rx descriptors */
-	     	if((lp->tx_ring = pci_alloc_consistent(lp->pci_dev,
-			sizeof(struct amd8111e_tx_dr)*NUM_TX_RING_DR,
-			&lp->tx_ring_dma_addr)) == NULL)
-
+	else {
+		/* allocate the tx and rx descriptors */
+		lp->tx_ring = dma_alloc_coherent(&lp->pci_dev->dev,
+			sizeof(struct amd8111e_tx_dr) * NUM_TX_RING_DR,
+			&lp->tx_ring_dma_addr, GFP_ATOMIC);
+		if (!lp->tx_ring)
 			goto err_no_mem;
 
-	     	if((lp->rx_ring = pci_alloc_consistent(lp->pci_dev,
-			sizeof(struct amd8111e_rx_dr)*NUM_RX_RING_DR,
-			&lp->rx_ring_dma_addr)) == NULL)
-
+		lp->rx_ring = dma_alloc_coherent(&lp->pci_dev->dev,
+			sizeof(struct amd8111e_rx_dr) * NUM_RX_RING_DR,
+			&lp->rx_ring_dma_addr, GFP_ATOMIC);
+		if (!lp->rx_ring)
 			goto err_free_tx_ring;
-
 	}
+
 	/* Set new receive buff size */
 	amd8111e_set_rx_buff_len(dev);
 
@@ -323,17 +311,19 @@ static int amd8111e_init_ring(struct net_device *dev)
 
 		lp->rx_skbuff[i] = netdev_alloc_skb(dev, lp->rx_buff_len);
 		if (!lp->rx_skbuff[i]) {
-				/* Release previos allocated skbs */
-				for(--i; i >= 0 ;i--)
-					dev_kfree_skb(lp->rx_skbuff[i]);
-				goto err_free_rx_ring;
+			/* Release previos allocated skbs */
+			for (--i; i >= 0; i--)
+				dev_kfree_skb(lp->rx_skbuff[i]);
+			goto err_free_rx_ring;
 		}
-		skb_reserve(lp->rx_skbuff[i],2);
+		skb_reserve(lp->rx_skbuff[i], 2);
 	}
         /* Initilaizing receive descriptors */
 	for (i = 0; i < NUM_RX_BUFFERS; i++) {
-		lp->rx_dma_addr[i] = pci_map_single(lp->pci_dev,
-			lp->rx_skbuff[i]->data,lp->rx_buff_len-2, PCI_DMA_FROMDEVICE);
+		lp->rx_dma_addr[i] = dma_map_single(&lp->pci_dev->dev,
+						    lp->rx_skbuff[i]->data,
+						    lp->rx_buff_len - 2,
+						    DMA_FROM_DEVICE);
 
 		lp->rx_ring[i].buff_phy_addr = cpu_to_le32(lp->rx_dma_addr[i]);
 		lp->rx_ring[i].buff_count = cpu_to_le16(lp->rx_buff_len-2);
@@ -352,15 +342,15 @@ static int amd8111e_init_ring(struct net_device *dev)
 
 err_free_rx_ring:
 
-	pci_free_consistent(lp->pci_dev,
-		sizeof(struct amd8111e_rx_dr)*NUM_RX_RING_DR,lp->rx_ring,
-		lp->rx_ring_dma_addr);
+	dma_free_coherent(&lp->pci_dev->dev,
+			  sizeof(struct amd8111e_rx_dr) * NUM_RX_RING_DR,
+			  lp->rx_ring, lp->rx_ring_dma_addr);
 
 err_free_tx_ring:
 
-	pci_free_consistent(lp->pci_dev,
-		 sizeof(struct amd8111e_tx_dr)*NUM_TX_RING_DR,lp->tx_ring,
-		 lp->tx_ring_dma_addr);
+	dma_free_coherent(&lp->pci_dev->dev,
+			  sizeof(struct amd8111e_tx_dr) * NUM_TX_RING_DR,
+			  lp->tx_ring, lp->tx_ring_dma_addr);
 
 err_no_mem:
 	return -ENOMEM;
@@ -384,40 +374,40 @@ static int amd8111e_set_coalesce(struct net_device *dev, enum coal_mode cmod)
 		case RX_INTR_COAL :
 			timeout = coal_conf->rx_timeout;
 			event_count = coal_conf->rx_event_count;
-			if( timeout > MAX_TIMEOUT ||
-					event_count > MAX_EVENT_COUNT )
+			if (timeout > MAX_TIMEOUT ||
+			    event_count > MAX_EVENT_COUNT)
 				return -EINVAL;
 
 			timeout = timeout * DELAY_TIMER_CONV;
 			writel(VAL0|STINTEN, mmio+INTEN0);
-			writel((u32)DLY_INT_A_R0|( event_count<< 16 )|timeout,
-							mmio+DLY_INT_A);
+			writel((u32)DLY_INT_A_R0 | (event_count << 16) |
+				timeout, mmio + DLY_INT_A);
 			break;
 
-		case TX_INTR_COAL :
+		case TX_INTR_COAL:
 			timeout = coal_conf->tx_timeout;
 			event_count = coal_conf->tx_event_count;
-			if( timeout > MAX_TIMEOUT ||
-					event_count > MAX_EVENT_COUNT )
+			if (timeout > MAX_TIMEOUT ||
+			    event_count > MAX_EVENT_COUNT)
 				return -EINVAL;
 
 
 			timeout = timeout * DELAY_TIMER_CONV;
-			writel(VAL0|STINTEN,mmio+INTEN0);
-			writel((u32)DLY_INT_B_T0|( event_count<< 16 )|timeout,
-							 mmio+DLY_INT_B);
+			writel(VAL0 | STINTEN, mmio + INTEN0);
+			writel((u32)DLY_INT_B_T0 | (event_count << 16) |
+				timeout, mmio + DLY_INT_B);
 			break;
 
 		case DISABLE_COAL:
-			writel(0,mmio+STVAL);
-			writel(STINTEN, mmio+INTEN0);
-			writel(0, mmio +DLY_INT_B);
-			writel(0, mmio+DLY_INT_A);
+			writel(0, mmio + STVAL);
+			writel(STINTEN, mmio + INTEN0);
+			writel(0, mmio + DLY_INT_B);
+			writel(0, mmio + DLY_INT_A);
 			break;
 		 case ENABLE_COAL:
 		       /* Start the timer */
-			writel((u32)SOFT_TIMER_FREQ, mmio+STVAL); /*  0.5 sec */
-			writel(VAL0|STINTEN, mmio+INTEN0);
+			writel((u32)SOFT_TIMER_FREQ, mmio + STVAL); /* 0.5 sec */
+			writel(VAL0 | STINTEN, mmio + INTEN0);
 			break;
 		default:
 			break;
@@ -432,67 +422,67 @@ static int amd8111e_restart(struct net_device *dev)
 {
 	struct amd8111e_priv *lp = netdev_priv(dev);
 	void __iomem *mmio = lp->mmio;
-	int i,reg_val;
+	int i, reg_val;
 
 	/* stop the chip */
-	 writel(RUN, mmio + CMD0);
+	writel(RUN, mmio + CMD0);
 
-	if(amd8111e_init_ring(dev))
+	if (amd8111e_init_ring(dev))
 		return -ENOMEM;
 
 	/* enable the port manager and set auto negotiation always */
-	writel((u32) VAL1|EN_PMGR, mmio + CMD3 );
-	writel((u32)XPHYANE|XPHYRST , mmio + CTRL2);
+	writel((u32)VAL1 | EN_PMGR, mmio + CMD3);
+	writel((u32)XPHYANE | XPHYRST, mmio + CTRL2);
 
 	amd8111e_set_ext_phy(dev);
 
 	/* set control registers */
 	reg_val = readl(mmio + CTRL1);
 	reg_val &= ~XMTSP_MASK;
-	writel( reg_val| XMTSP_128 | CACHE_ALIGN, mmio + CTRL1 );
+	writel(reg_val | XMTSP_128 | CACHE_ALIGN, mmio + CTRL1);
 
 	/* enable interrupt */
-	writel( APINT5EN | APINT4EN | APINT3EN | APINT2EN | APINT1EN |
+	writel(APINT5EN | APINT4EN | APINT3EN | APINT2EN | APINT1EN |
 		APINT0EN | MIIPDTINTEN | MCCIINTEN | MCCINTEN | MREINTEN |
 		SPNDINTEN | MPINTEN | SINTEN | STINTEN, mmio + INTEN0);
 
 	writel(VAL3 | LCINTEN | VAL1 | TINTEN0 | VAL0 | RINTEN0, mmio + INTEN0);
 
 	/* initialize tx and rx ring base addresses */
-	writel((u32)lp->tx_ring_dma_addr,mmio + XMT_RING_BASE_ADDR0);
-	writel((u32)lp->rx_ring_dma_addr,mmio+ RCV_RING_BASE_ADDR0);
+	writel((u32)lp->tx_ring_dma_addr, mmio + XMT_RING_BASE_ADDR0);
+	writel((u32)lp->rx_ring_dma_addr, mmio + RCV_RING_BASE_ADDR0);
 
 	writew((u32)NUM_TX_RING_DR, mmio + XMT_RING_LEN0);
 	writew((u16)NUM_RX_RING_DR, mmio + RCV_RING_LEN0);
 
 	/* set default IPG to 96 */
-	writew((u32)DEFAULT_IPG,mmio+IPG);
+	writew((u32)DEFAULT_IPG, mmio + IPG);
 	writew((u32)(DEFAULT_IPG-IFS1_DELTA), mmio + IFS1);
 
-	if(lp->options & OPTION_JUMBO_ENABLE){
+	if (lp->options & OPTION_JUMBO_ENABLE) {
 		writel((u32)VAL2|JUMBO, mmio + CMD3);
 		/* Reset REX_UFLO */
-		writel( REX_UFLO, mmio + CMD2);
+		writel(REX_UFLO, mmio + CMD2);
 		/* Should not set REX_UFLO for jumbo frames */
-		writel( VAL0 | APAD_XMT|REX_RTRY , mmio + CMD2);
-	}else{
-		writel( VAL0 | APAD_XMT | REX_RTRY|REX_UFLO, mmio + CMD2);
+		writel(VAL0 | APAD_XMT | REX_RTRY, mmio + CMD2);
+	} else {
+		writel(VAL0 | APAD_XMT | REX_RTRY | REX_UFLO, mmio + CMD2);
 		writel((u32)JUMBO, mmio + CMD3);
 	}
 
 #if AMD8111E_VLAN_TAG_USED
-	writel((u32) VAL2|VSIZE|VL_TAG_DEL, mmio + CMD3);
+	writel((u32)VAL2 | VSIZE | VL_TAG_DEL, mmio + CMD3);
 #endif
-	writel( VAL0 | APAD_XMT | REX_RTRY, mmio + CMD2 );
+	writel(VAL0 | APAD_XMT | REX_RTRY, mmio + CMD2);
 
 	/* Setting the MAC address to the device */
 	for (i = 0; i < ETH_ALEN; i++)
-		writeb( dev->dev_addr[i], mmio + PADR + i );
+		writeb(dev->dev_addr[i], mmio + PADR + i);
 
 	/* Enable interrupt coalesce */
-	if(lp->options & OPTION_INTR_COAL_ENABLE){
+	if (lp->options & OPTION_INTR_COAL_ENABLE) {
 		netdev_info(dev, "Interrupt Coalescing Enabled.\n");
-		amd8111e_set_coalesce(dev,ENABLE_COAL);
+		amd8111e_set_coalesce(dev, ENABLE_COAL);
 	}
 
 	/* set RUN bit to start the chip */
@@ -508,11 +498,11 @@ static int amd8111e_restart(struct net_device *dev)
 static void amd8111e_init_hw_default(struct amd8111e_priv *lp)
 {
 	unsigned int reg_val;
-	unsigned int logic_filter[2] ={0,};
+	unsigned int logic_filter[2] = {0,};
 	void __iomem *mmio = lp->mmio;
 
 
-        /* stop the chip */
+	/* stop the chip */
 	writel(RUN, mmio + CMD0);
 
 	/* AUTOPOLL0 Register *//*TBD default value is 8100 in FPS */
@@ -528,13 +518,13 @@ static void amd8111e_init_hw_default(struct amd8111e_priv *lp)
 	writel(0, mmio + XMT_RING_BASE_ADDR3);
 
 	/* Clear CMD0  */
-	writel(CMD0_CLEAR,mmio + CMD0);
+	writel(CMD0_CLEAR, mmio + CMD0);
 
 	/* Clear CMD2 */
-	writel(CMD2_CLEAR, mmio +CMD2);
+	writel(CMD2_CLEAR, mmio + CMD2);
 
 	/* Clear CMD7 */
-	writel(CMD7_CLEAR , mmio + CMD7);
+	writel(CMD7_CLEAR, mmio + CMD7);
 
 	/* Clear DLY_INT_A and DLY_INT_B */
 	writel(0x0, mmio + DLY_INT_A);
@@ -551,16 +541,16 @@ static void amd8111e_init_hw_default(struct amd8111e_priv *lp)
 	writel(0x0, mmio + STVAL);
 
 	/* Clear INTEN0 */
-	writel( INTEN0_CLEAR, mmio + INTEN0);
+	writel(INTEN0_CLEAR, mmio + INTEN0);
 
 	/* Clear LADRF */
-	writel(0x0 , mmio + LADRF);
+	writel(0x0, mmio + LADRF);
 
 	/* Set SRAM_SIZE & SRAM_BOUNDARY registers  */
-	writel( 0x80010,mmio + SRAM_SIZE);
+	writel(0x80010, mmio + SRAM_SIZE);
 
 	/* Clear RCV_RING0_LEN */
-	writel(0x0, mmio +  RCV_RING_LEN0);
+	writel(0x0, mmio + RCV_RING_LEN0);
 
 	/* Clear XMT_RING0/1/2/3_LEN */
 	writel(0x0, mmio +  XMT_RING_LEN0);
@@ -580,10 +570,10 @@ static void amd8111e_init_hw_default(struct amd8111e_priv *lp)
 	/* SRAM_SIZE register */
 	reg_val = readl(mmio + SRAM_SIZE);
 
-	if(lp->options & OPTION_JUMBO_ENABLE)
-		writel( VAL2|JUMBO, mmio + CMD3);
+	if (lp->options & OPTION_JUMBO_ENABLE)
+		writel(VAL2 | JUMBO, mmio + CMD3);
 #if AMD8111E_VLAN_TAG_USED
-	writel(VAL2|VSIZE|VL_TAG_DEL, mmio + CMD3 );
+	writel(VAL2 | VSIZE | VL_TAG_DEL, mmio + CMD3);
 #endif
 	/* Set default value to CTRL1 Register */
 	writel(CTRL1_DEFAULT, mmio + CTRL1);
@@ -625,17 +615,17 @@ static void amd8111e_stop_chip(struct amd8111e_priv *lp)
 static void amd8111e_free_ring(struct amd8111e_priv *lp)
 {
 	/* Free transmit and receive descriptor rings */
-	if(lp->rx_ring){
-		pci_free_consistent(lp->pci_dev,
-			sizeof(struct amd8111e_rx_dr)*NUM_RX_RING_DR,
-			lp->rx_ring, lp->rx_ring_dma_addr);
+	if (lp->rx_ring) {
+		dma_free_coherent(&lp->pci_dev->dev,
+				  sizeof(struct amd8111e_rx_dr) * NUM_RX_RING_DR,
+				  lp->rx_ring, lp->rx_ring_dma_addr);
 		lp->rx_ring = NULL;
 	}
 
-	if(lp->tx_ring){
-		pci_free_consistent(lp->pci_dev,
-			sizeof(struct amd8111e_tx_dr)*NUM_TX_RING_DR,
-			lp->tx_ring, lp->tx_ring_dma_addr);
+	if (lp->tx_ring) {
+		dma_free_coherent(&lp->pci_dev->dev,
+				  sizeof(struct amd8111e_tx_dr) * NUM_TX_RING_DR,
+				  lp->tx_ring, lp->tx_ring_dma_addr);
 
 		lp->tx_ring = NULL;
 	}
@@ -649,24 +639,25 @@ static void amd8111e_free_ring(struct amd8111e_priv *lp)
 static int amd8111e_tx(struct net_device *dev)
 {
 	struct amd8111e_priv *lp = netdev_priv(dev);
-	int tx_index = lp->tx_complete_idx & TX_RING_DR_MOD_MASK;
+	int tx_index;
 	int status;
 	/* Complete all the transmit packet */
-	while (lp->tx_complete_idx != lp->tx_idx){
+	while (lp->tx_complete_idx != lp->tx_idx) {
 		tx_index =  lp->tx_complete_idx & TX_RING_DR_MOD_MASK;
 		status = le16_to_cpu(lp->tx_ring[tx_index].tx_flags);
 
-		if(status & OWN_BIT)
+		if (status & OWN_BIT)
 			break;	/* It still hasn't been Txed */
 
 		lp->tx_ring[tx_index].buff_phy_addr = 0;
 
 		/* We must free the original skb */
 		if (lp->tx_skbuff[tx_index]) {
-			pci_unmap_single(lp->pci_dev, lp->tx_dma_addr[tx_index],
-				  	lp->tx_skbuff[tx_index]->len,
-					PCI_DMA_TODEVICE);
-			dev_kfree_skb_irq (lp->tx_skbuff[tx_index]);
+			dma_unmap_single(&lp->pci_dev->dev,
+					 lp->tx_dma_addr[tx_index],
+					 lp->tx_skbuff[tx_index]->len,
+					 DMA_TO_DEVICE);
+			dev_consume_skb_irq(lp->tx_skbuff[tx_index]);
 			lp->tx_skbuff[tx_index] = NULL;
 			lp->tx_dma_addr[tx_index] = 0;
 		}
@@ -677,10 +668,10 @@ static int amd8111e_tx(struct net_device *dev)
 			le16_to_cpu(lp->tx_ring[tx_index].buff_count);
 
 		if (netif_queue_stopped(dev) &&
-			lp->tx_complete_idx > lp->tx_idx - NUM_TX_BUFFERS +2){
+			lp->tx_complete_idx > lp->tx_idx - NUM_TX_BUFFERS + 2) {
 			/* The ring is no longer full, clear tbusy. */
 			/* lp->tx_full = 0; */
-			netif_wake_queue (dev);
+			netif_wake_queue(dev);
 		}
 	}
 	return 0;
@@ -693,7 +684,7 @@ static int amd8111e_rx_poll(struct napi_struct *napi, int budget)
 	struct net_device *dev = lp->amd8111e_net_dev;
 	int rx_index = lp->rx_idx & RX_RING_DR_MOD_MASK;
 	void __iomem *mmio = lp->mmio;
-	struct sk_buff *skb,*new_skb;
+	struct sk_buff *skb, *new_skb;
 	int min_pkt_len, status;
 	int num_rx_pkt = 0;
 	short pkt_len;
@@ -718,7 +709,7 @@ static int amd8111e_rx_poll(struct napi_struct *napi, int budget)
 			goto err_next_pkt;
 		}
 		/* check for STP and ENP */
-		if (!((status & STP_BIT) && (status & ENP_BIT))){
+		if (!((status & STP_BIT) && (status & ENP_BIT))) {
 			/* resetting flags */
 			lp->rx_ring[rx_index].rx_flags &= RESET_RX_FLAGS;
 			goto err_next_pkt;
@@ -751,19 +742,19 @@ static int amd8111e_rx_poll(struct napi_struct *napi, int budget)
 
 		skb_reserve(new_skb, 2);
 		skb = lp->rx_skbuff[rx_index];
-		pci_unmap_single(lp->pci_dev,lp->rx_dma_addr[rx_index],
-				 lp->rx_buff_len-2, PCI_DMA_FROMDEVICE);
+		dma_unmap_single(&lp->pci_dev->dev, lp->rx_dma_addr[rx_index],
+				 lp->rx_buff_len - 2, DMA_FROM_DEVICE);
 		skb_put(skb, pkt_len);
 		lp->rx_skbuff[rx_index] = new_skb;
-		lp->rx_dma_addr[rx_index] = pci_map_single(lp->pci_dev,
+		lp->rx_dma_addr[rx_index] = dma_map_single(&lp->pci_dev->dev,
 							   new_skb->data,
-							   lp->rx_buff_len-2,
-							   PCI_DMA_FROMDEVICE);
+							   lp->rx_buff_len - 2,
+							   DMA_FROM_DEVICE);
 
 		skb->protocol = eth_type_trans(skb, dev);
 
 #if AMD8111E_VLAN_TAG_USED
-		if (vtag == TT_VLAN_TAGGED){
+		if (vtag == TT_VLAN_TAGGED) {
 			u16 vlan_tag = le16_to_cpu(lp->rx_ring[rx_index].tag_ctrl_info);
 			__vlan_hwaccel_put_tag(skb, htons(ETH_P_8021Q), vlan_tag);
 		}
@@ -801,25 +792,25 @@ err_next_pkt:
 static int amd8111e_link_change(struct net_device *dev)
 {
 	struct amd8111e_priv *lp = netdev_priv(dev);
-	int status0,speed;
+	int status0, speed;
 
 	/* read the link change */
-     	status0 = readl(lp->mmio + STAT0);
+	status0 = readl(lp->mmio + STAT0);
 
-	if(status0 & LINK_STATS){
-		if(status0 & AUTONEG_COMPLETE)
+	if (status0 & LINK_STATS) {
+		if (status0 & AUTONEG_COMPLETE)
 			lp->link_config.autoneg = AUTONEG_ENABLE;
 		else
 			lp->link_config.autoneg = AUTONEG_DISABLE;
 
-		if(status0 & FULL_DPLX)
+		if (status0 & FULL_DPLX)
 			lp->link_config.duplex = DUPLEX_FULL;
 		else
 			lp->link_config.duplex = DUPLEX_HALF;
 		speed = (status0 & SPEED_MASK) >> 7;
-		if(speed == PHY_SPEED_10)
+		if (speed == PHY_SPEED_10)
 			lp->link_config.speed = SPEED_10;
-		else if(speed == PHY_SPEED_100)
+		else if (speed == PHY_SPEED_100)
 			lp->link_config.speed = SPEED_100;
 
 		netdev_info(dev, "Link is Up. Speed is %s Mbps %s Duplex\n",
@@ -829,8 +820,7 @@ static int amd8111e_link_change(struct net_device *dev)
 							"Full" : "Half");
 
 		netif_carrier_on(dev);
-	}
-	else{
+	} else {
 		lp->link_config.speed = SPEED_INVALID;
 		lp->link_config.duplex = DUPLEX_INVALID;
 		lp->link_config.autoneg = AUTONEG_INVALID;
@@ -848,7 +838,7 @@ static int amd8111e_read_mib(void __iomem *mmio, u8 MIB_COUNTER)
 	unsigned  int data;
 	unsigned int repeat = REPEAT_CNT;
 
-	writew( MIB_RD_CMD | MIB_COUNTER, mmio + MIB_ADDR);
+	writew(MIB_RD_CMD | MIB_COUNTER, mmio + MIB_ADDR);
 	do {
 		status = readw(mmio + MIB_ADDR);
 		udelay(2);	/* controller takes MAX 2 us to get mib data */
@@ -871,7 +861,7 @@ static struct net_device_stats *amd8111e_get_stats(struct net_device *dev)
 
 	if (!lp->opened)
 		return new_stats;
-	spin_lock_irqsave (&lp->lock, flags);
+	spin_lock_irqsave(&lp->lock, flags);
 
 	/* stats.rx_packets */
 	new_stats->rx_packets = amd8111e_read_mib(mmio, rcv_broadcast_pkts)+
@@ -951,7 +941,7 @@ static struct net_device_stats *amd8111e_get_stats(struct net_device *dev)
 	/* Reset the mibs for collecting new statistics */
 	/* writew(MIB_CLEAR, mmio + MIB_ADDR);*/
 
-	spin_unlock_irqrestore (&lp->lock, flags);
+	spin_unlock_irqrestore(&lp->lock, flags);
 
 	return new_stats;
 }
@@ -982,108 +972,98 @@ static int amd8111e_calc_coalesce(struct net_device *dev)
 	rx_data_rate = coal_conf->rx_bytes - coal_conf->rx_prev_bytes;
 	coal_conf->rx_prev_bytes =  coal_conf->rx_bytes;
 
-	if(rx_pkt_rate < 800){
-		if(coal_conf->rx_coal_type != NO_COALESCE){
+	if (rx_pkt_rate < 800) {
+		if (coal_conf->rx_coal_type != NO_COALESCE) {
 
 			coal_conf->rx_timeout = 0x0;
 			coal_conf->rx_event_count = 0;
-			amd8111e_set_coalesce(dev,RX_INTR_COAL);
+			amd8111e_set_coalesce(dev, RX_INTR_COAL);
 			coal_conf->rx_coal_type = NO_COALESCE;
 		}
-	}
-	else{
+	} else {
 
 		rx_pkt_size = rx_data_rate/rx_pkt_rate;
-		if (rx_pkt_size < 128){
-			if(coal_conf->rx_coal_type != NO_COALESCE){
+		if (rx_pkt_size < 128) {
+			if (coal_conf->rx_coal_type != NO_COALESCE) {
 
 				coal_conf->rx_timeout = 0;
 				coal_conf->rx_event_count = 0;
-				amd8111e_set_coalesce(dev,RX_INTR_COAL);
+				amd8111e_set_coalesce(dev, RX_INTR_COAL);
 				coal_conf->rx_coal_type = NO_COALESCE;
 			}
 
-		}
-		else if ( (rx_pkt_size >= 128) && (rx_pkt_size < 512) ){
+		} else if ((rx_pkt_size >= 128) && (rx_pkt_size < 512)) {
 
-			if(coal_conf->rx_coal_type !=  LOW_COALESCE){
+			if (coal_conf->rx_coal_type !=  LOW_COALESCE) {
 				coal_conf->rx_timeout = 1;
 				coal_conf->rx_event_count = 4;
-				amd8111e_set_coalesce(dev,RX_INTR_COAL);
+				amd8111e_set_coalesce(dev, RX_INTR_COAL);
 				coal_conf->rx_coal_type = LOW_COALESCE;
 			}
-		}
-		else if ((rx_pkt_size >= 512) && (rx_pkt_size < 1024)){
+		} else if ((rx_pkt_size >= 512) && (rx_pkt_size < 1024)) {
 
-			if(coal_conf->rx_coal_type !=  MEDIUM_COALESCE){
+			if (coal_conf->rx_coal_type != MEDIUM_COALESCE) {
 				coal_conf->rx_timeout = 1;
 				coal_conf->rx_event_count = 4;
-				amd8111e_set_coalesce(dev,RX_INTR_COAL);
+				amd8111e_set_coalesce(dev, RX_INTR_COAL);
 				coal_conf->rx_coal_type = MEDIUM_COALESCE;
 			}
 
-		}
-		else if(rx_pkt_size >= 1024){
-			if(coal_conf->rx_coal_type !=  HIGH_COALESCE){
+		} else if (rx_pkt_size >= 1024) {
+
+			if (coal_conf->rx_coal_type !=  HIGH_COALESCE) {
 				coal_conf->rx_timeout = 2;
 				coal_conf->rx_event_count = 3;
-				amd8111e_set_coalesce(dev,RX_INTR_COAL);
+				amd8111e_set_coalesce(dev, RX_INTR_COAL);
 				coal_conf->rx_coal_type = HIGH_COALESCE;
 			}
 		}
 	}
-    	/* NOW FOR TX INTR COALESC */
-	if(tx_pkt_rate < 800){
-		if(coal_conf->tx_coal_type != NO_COALESCE){
+	/* NOW FOR TX INTR COALESC */
+	if (tx_pkt_rate < 800) {
+		if (coal_conf->tx_coal_type != NO_COALESCE) {
 
 			coal_conf->tx_timeout = 0x0;
 			coal_conf->tx_event_count = 0;
-			amd8111e_set_coalesce(dev,TX_INTR_COAL);
+			amd8111e_set_coalesce(dev, TX_INTR_COAL);
 			coal_conf->tx_coal_type = NO_COALESCE;
 		}
-	}
-	else{
+	} else {
 
 		tx_pkt_size = tx_data_rate/tx_pkt_rate;
-		if (tx_pkt_size < 128){
+		if (tx_pkt_size < 128) {
 
-			if(coal_conf->tx_coal_type != NO_COALESCE){
+			if (coal_conf->tx_coal_type != NO_COALESCE) {
 
 				coal_conf->tx_timeout = 0;
 				coal_conf->tx_event_count = 0;
-				amd8111e_set_coalesce(dev,TX_INTR_COAL);
+				amd8111e_set_coalesce(dev, TX_INTR_COAL);
 				coal_conf->tx_coal_type = NO_COALESCE;
 			}
 
-		}
-		else if ( (tx_pkt_size >= 128) && (tx_pkt_size < 512) ){
+		} else if ((tx_pkt_size >= 128) && (tx_pkt_size < 512)) {
 
-			if(coal_conf->tx_coal_type !=  LOW_COALESCE){
+			if (coal_conf->tx_coal_type != LOW_COALESCE) {
 				coal_conf->tx_timeout = 1;
 				coal_conf->tx_event_count = 2;
-				amd8111e_set_coalesce(dev,TX_INTR_COAL);
+				amd8111e_set_coalesce(dev, TX_INTR_COAL);
 				coal_conf->tx_coal_type = LOW_COALESCE;
 
 			}
-		}
-		else if ((tx_pkt_size >= 512) && (tx_pkt_size < 1024)){
+		} else if ((tx_pkt_size >= 512) && (tx_pkt_size < 1024)) {
 
-			if(coal_conf->tx_coal_type !=  MEDIUM_COALESCE){
+			if (coal_conf->tx_coal_type != MEDIUM_COALESCE) {
 				coal_conf->tx_timeout = 2;
 				coal_conf->tx_event_count = 5;
-				amd8111e_set_coalesce(dev,TX_INTR_COAL);
+				amd8111e_set_coalesce(dev, TX_INTR_COAL);
 				coal_conf->tx_coal_type = MEDIUM_COALESCE;
 			}
-
-		}
-		else if(tx_pkt_size >= 1024){
-			if (tx_pkt_size >= 1024){
-				if(coal_conf->tx_coal_type !=  HIGH_COALESCE){
-					coal_conf->tx_timeout = 4;
-					coal_conf->tx_event_count = 8;
-					amd8111e_set_coalesce(dev,TX_INTR_COAL);
-					coal_conf->tx_coal_type = HIGH_COALESCE;
-				}
+		} else if (tx_pkt_size >= 1024) {
+			if (coal_conf->tx_coal_type != HIGH_COALESCE) {
+				coal_conf->tx_timeout = 4;
+				coal_conf->tx_event_count = 8;
+				amd8111e_set_coalesce(dev, TX_INTR_COAL);
+				coal_conf->tx_coal_type = HIGH_COALESCE;
 			}
 		}
 	}
@@ -1103,7 +1083,7 @@ static irqreturn_t amd8111e_interrupt(int irq, void *dev_id)
 	unsigned int intr0, intren0;
 	unsigned int handled = 1;
 
-	if(unlikely(dev == NULL))
+	if (unlikely(!dev))
 		return IRQ_NONE;
 
 	spin_lock(&lp->lock);
@@ -1117,7 +1097,7 @@ static irqreturn_t amd8111e_interrupt(int irq, void *dev_id)
 
 	/* Process all the INT event until INTR bit is clear. */
 
-	if (!(intr0 & INTR)){
+	if (!(intr0 & INTR)) {
 		handled = 0;
 		goto err_no_interrupt;
 	}
@@ -1128,7 +1108,7 @@ static irqreturn_t amd8111e_interrupt(int irq, void *dev_id)
 	/* Check if Receive Interrupt has occurred. */
 	if (intr0 & RINT0) {
 		if (napi_schedule_prep(&lp->napi)) {
-			/* Disable receive interupts */
+			/* Disable receive interrupts */
 			writel(RINTEN0, mmio + INTEN0);
 			/* Schedule a polling routine */
 			__napi_schedule(&lp->napi);
@@ -1152,7 +1132,7 @@ static irqreturn_t amd8111e_interrupt(int irq, void *dev_id)
 		amd8111e_calc_coalesce(dev);
 
 err_no_interrupt:
-	writel( VAL0 | INTREN,mmio + CMD0);
+	writel(VAL0 | INTREN, mmio + CMD0);
 
 	spin_unlock(&lp->lock);
 
@@ -1192,8 +1172,8 @@ static int amd8111e_close(struct net_device *dev)
 	netif_carrier_off(lp->amd8111e_net_dev);
 
 	/* Delete ipg timer */
-	if(lp->options & OPTION_DYN_IPG_ENABLE)
-		del_timer_sync(&lp->ipg_data.ipg_timer);
+	if (lp->options & OPTION_DYN_IPG_ENABLE)
+		timer_delete_sync(&lp->ipg_data.ipg_timer);
 
 	spin_unlock_irq(&lp->lock);
 	free_irq(dev->irq, dev);
@@ -1212,8 +1192,8 @@ static int amd8111e_open(struct net_device *dev)
 {
 	struct amd8111e_priv *lp = netdev_priv(dev);
 
-	if(dev->irq ==0 || request_irq(dev->irq, amd8111e_interrupt, IRQF_SHARED,
-					 dev->name, dev))
+	if (dev->irq == 0 || request_irq(dev->irq, amd8111e_interrupt,
+					 IRQF_SHARED, dev->name, dev))
 		return -EAGAIN;
 
 	napi_enable(&lp->napi);
@@ -1222,7 +1202,7 @@ static int amd8111e_open(struct net_device *dev)
 
 	amd8111e_init_hw_default(lp);
 
-	if(amd8111e_restart(dev)){
+	if (amd8111e_restart(dev)) {
 		spin_unlock_irq(&lp->lock);
 		napi_disable(&lp->napi);
 		if (dev->irq)
@@ -1230,7 +1210,7 @@ static int amd8111e_open(struct net_device *dev)
 		return -ENOMEM;
 	}
 	/* Start ipg timer */
-	if(lp->options & OPTION_DYN_IPG_ENABLE){
+	if (lp->options & OPTION_DYN_IPG_ENABLE) {
 		add_timer(&lp->ipg_data.ipg_timer);
 		netdev_info(dev, "Dynamic IPG Enabled\n");
 	}
@@ -1288,7 +1268,8 @@ static netdev_tx_t amd8111e_start_xmit(struct sk_buff *skb,
 	}
 #endif
 	lp->tx_dma_addr[tx_index] =
-	    pci_map_single(lp->pci_dev, skb->data, skb->len, PCI_DMA_TODEVICE);
+	    dma_map_single(&lp->pci_dev->dev, skb->data, skb->len,
+			   DMA_TO_DEVICE);
 	lp->tx_ring[tx_index].buff_phy_addr =
 	    cpu_to_le32(lp->tx_dma_addr[tx_index]);
 
@@ -1300,10 +1281,10 @@ static netdev_tx_t amd8111e_start_xmit(struct sk_buff *skb,
 	lp->tx_idx++;
 
 	/* Trigger an immediate send poll. */
-	writel( VAL1 | TDMD0, lp->mmio + CMD0);
-	writel( VAL2 | RDMD0,lp->mmio + CMD0);
+	writel(VAL1 | TDMD0, lp->mmio + CMD0);
+	writel(VAL2 | RDMD0, lp->mmio + CMD0);
 
-	if(amd8111e_tx_queue_avail(lp) < 0){
+	if (amd8111e_tx_queue_avail(lp) < 0) {
 		netif_stop_queue(dev);
 	}
 	spin_unlock_irqrestore(&lp->lock, flags);
@@ -1337,15 +1318,15 @@ static void amd8111e_set_multicast_list(struct net_device *dev)
 {
 	struct netdev_hw_addr *ha;
 	struct amd8111e_priv *lp = netdev_priv(dev);
-	u32 mc_filter[2] ;
+	u32 mc_filter[2];
 	int bit_num;
 
-	if(dev->flags & IFF_PROMISC){
-		writel( VAL2 | PROM, lp->mmio + CMD2);
+	if (dev->flags & IFF_PROMISC) {
+		writel(VAL2 | PROM, lp->mmio + CMD2);
 		return;
 	}
 	else
-		writel( PROM, lp->mmio + CMD2);
+		writel(PROM, lp->mmio + CMD2);
 	if (dev->flags & IFF_ALLMULTI ||
 	    netdev_mc_count(dev) > MAX_FILTER_SIZE) {
 		/* get all multicast packet */
@@ -1382,11 +1363,10 @@ static void amd8111e_get_drvinfo(struct net_device *dev,
 {
 	struct amd8111e_priv *lp = netdev_priv(dev);
 	struct pci_dev *pci_dev = lp->pci_dev;
-	strlcpy(info->driver, MODULE_NAME, sizeof(info->driver));
-	strlcpy(info->version, MODULE_VERS, sizeof(info->version));
+	strscpy(info->driver, MODULE_NAME, sizeof(info->driver));
 	snprintf(info->fw_version, sizeof(info->fw_version),
 		"%u", chip_version);
-	strlcpy(info->bus_info, pci_name(pci_dev), sizeof(info->bus_info));
+	strscpy(info->bus_info, pci_name(pci_dev), sizeof(info->bus_info));
 }
 
 static int amd8111e_get_regs_len(struct net_device *dev)
@@ -1451,7 +1431,7 @@ static int amd8111e_set_wol(struct net_device *dev, struct ethtool_wolinfo *wol_
 	if (wol_info->wolopts & WAKE_MAGIC)
 		lp->options |=
 			(OPTION_WOL_ENABLE | OPTION_WAKE_MAGIC_ENABLE);
-	else if(wol_info->wolopts & WAKE_PHY)
+	else if (wol_info->wolopts & WAKE_PHY)
 		lp->options |=
 			(OPTION_WOL_ENABLE | OPTION_WAKE_PHY_ENABLE);
 	else
@@ -1476,18 +1456,18 @@ static const struct ethtool_ops ops = {
  * gets/sets driver speed, gets memory mapped register values, forces
  * auto negotiation, sets/gets WOL options for ethtool application.
  */
-static int amd8111e_ioctl(struct net_device *dev , struct ifreq *ifr, int cmd)
+static int amd8111e_ioctl(struct net_device *dev, struct ifreq *ifr, int cmd)
 {
 	struct mii_ioctl_data *data = if_mii(ifr);
 	struct amd8111e_priv *lp = netdev_priv(dev);
 	int err;
 	u32 mii_regval;
 
-	switch(cmd) {
+	switch (cmd) {
 	case SIOCGMIIPHY:
 		data->phy_id = lp->ext_phy_addr;
 
-	/* fallthru */
+		fallthrough;
 	case SIOCGMIIREG:
 
 		spin_lock_irq(&lp->lock);
@@ -1519,11 +1499,11 @@ static int amd8111e_set_mac_address(struct net_device *dev, void *p)
 	int i;
 	struct sockaddr *addr = p;
 
-	memcpy(dev->dev_addr, addr->sa_data, dev->addr_len);
+	eth_hw_addr_set(dev, addr->sa_data);
 	spin_lock_irq(&lp->lock);
 	/* Setting the MAC address to the device */
 	for (i = 0; i < ETH_ALEN; i++)
-		writeb( dev->dev_addr[i], lp->mmio + PADR + i );
+		writeb(dev->dev_addr[i], lp->mmio + PADR + i);
 
 	spin_unlock_irq(&lp->lock);
 
@@ -1540,30 +1520,30 @@ static int amd8111e_change_mtu(struct net_device *dev, int new_mtu)
 
 	if (!netif_running(dev)) {
 		/* new_mtu will be used
-		 * when device starts netxt time
+		 * when device starts next time
 		 */
-		dev->mtu = new_mtu;
+		WRITE_ONCE(dev->mtu, new_mtu);
 		return 0;
 	}
 
 	spin_lock_irq(&lp->lock);
 
-        /* stop the chip */
+	/* stop the chip */
 	writel(RUN, lp->mmio + CMD0);
 
-	dev->mtu = new_mtu;
+	WRITE_ONCE(dev->mtu, new_mtu);
 
 	err = amd8111e_restart(dev);
 	spin_unlock_irq(&lp->lock);
-	if(!err)
+	if (!err)
 		netif_start_queue(dev);
 	return err;
 }
 
 static int amd8111e_enable_magicpkt(struct amd8111e_priv *lp)
 {
-	writel( VAL1|MPPLBA, lp->mmio + CMD3);
-	writel( VAL0|MPEN_SW, lp->mmio + CMD7);
+	writel(VAL1 | MPPLBA, lp->mmio + CMD3);
+	writel(VAL0 | MPEN_SW, lp->mmio + CMD7);
 
 	/* To eliminate PCI posting bug */
 	readl(lp->mmio + CMD7);
@@ -1573,8 +1553,8 @@ static int amd8111e_enable_magicpkt(struct amd8111e_priv *lp)
 static int amd8111e_enable_link_change(struct amd8111e_priv *lp)
 {
 
-	/* Adapter is already stoped/suspended/interrupt-disabled */
-	writel(VAL0|LCMODE_SW,lp->mmio + CMD7);
+	/* Adapter is already stopped/suspended/interrupt-disabled */
+	writel(VAL0 | LCMODE_SW, lp->mmio + CMD7);
 
 	/* To eliminate PCI posting bug */
 	readl(lp->mmio + CMD7);
@@ -1586,7 +1566,7 @@ static int amd8111e_enable_link_change(struct amd8111e_priv *lp)
  * failed or the interface is locked up. This function will reinitialize
  * the hardware.
  */
-static void amd8111e_tx_timeout(struct net_device *dev)
+static void amd8111e_tx_timeout(struct net_device *dev, unsigned int txqueue)
 {
 	struct amd8111e_priv *lp = netdev_priv(dev);
 	int err;
@@ -1596,12 +1576,13 @@ static void amd8111e_tx_timeout(struct net_device *dev)
 	spin_lock_irq(&lp->lock);
 	err = amd8111e_restart(dev);
 	spin_unlock_irq(&lp->lock);
-	if(!err)
+	if (!err)
 		netif_wake_queue(dev);
 }
-static int amd8111e_suspend(struct pci_dev *pci_dev, pm_message_t state)
+
+static int __maybe_unused amd8111e_suspend(struct device *dev_d)
 {
-	struct net_device *dev = pci_get_drvdata(pci_dev);
+	struct net_device *dev = dev_get_drvdata(dev_d);
 	struct amd8111e_priv *lp = netdev_priv(dev);
 
 	if (!netif_running(dev))
@@ -1616,52 +1597,41 @@ static int amd8111e_suspend(struct pci_dev *pci_dev, pm_message_t state)
 
 	/* stop chip */
 	spin_lock_irq(&lp->lock);
-	if(lp->options & OPTION_DYN_IPG_ENABLE)
-		del_timer_sync(&lp->ipg_data.ipg_timer);
+	if (lp->options & OPTION_DYN_IPG_ENABLE)
+		timer_delete_sync(&lp->ipg_data.ipg_timer);
 	amd8111e_stop_chip(lp);
 	spin_unlock_irq(&lp->lock);
 
-	if(lp->options & OPTION_WOL_ENABLE){
+	if (lp->options & OPTION_WOL_ENABLE) {
 		 /* enable wol */
-		if(lp->options & OPTION_WAKE_MAGIC_ENABLE)
+		if (lp->options & OPTION_WAKE_MAGIC_ENABLE)
 			amd8111e_enable_magicpkt(lp);
-		if(lp->options & OPTION_WAKE_PHY_ENABLE)
+		if (lp->options & OPTION_WAKE_PHY_ENABLE)
 			amd8111e_enable_link_change(lp);
 
-		pci_enable_wake(pci_dev, PCI_D3hot, 1);
-		pci_enable_wake(pci_dev, PCI_D3cold, 1);
+		device_set_wakeup_enable(dev_d, 1);
 
-	}
-	else{
-		pci_enable_wake(pci_dev, PCI_D3hot, 0);
-		pci_enable_wake(pci_dev, PCI_D3cold, 0);
+	} else {
+		device_set_wakeup_enable(dev_d, 0);
 	}
 
-	pci_save_state(pci_dev);
-	pci_set_power_state(pci_dev, PCI_D3hot);
-
 	return 0;
 }
-static int amd8111e_resume(struct pci_dev *pci_dev)
+
+static int __maybe_unused amd8111e_resume(struct device *dev_d)
 {
-	struct net_device *dev = pci_get_drvdata(pci_dev);
+	struct net_device *dev = dev_get_drvdata(dev_d);
 	struct amd8111e_priv *lp = netdev_priv(dev);
 
 	if (!netif_running(dev))
 		return 0;
 
-	pci_set_power_state(pci_dev, PCI_D0);
-	pci_restore_state(pci_dev);
-
-	pci_enable_wake(pci_dev, PCI_D3hot, 0);
-	pci_enable_wake(pci_dev, PCI_D3cold, 0); /* D3 cold */
-
 	netif_device_attach(dev);
 
 	spin_lock_irq(&lp->lock);
 	amd8111e_restart(dev);
 	/* Restart ipg timer */
-	if(lp->options & OPTION_DYN_IPG_ENABLE)
+	if (lp->options & OPTION_DYN_IPG_ENABLE)
 		mod_timer(&lp->ipg_data.ipg_timer,
 				jiffies + IPG_CONVERGE_JIFFIES);
 	spin_unlock_irq(&lp->lock);
@@ -1669,23 +1639,24 @@ static int amd8111e_resume(struct pci_dev *pci_dev)
 	return 0;
 }
 
-static void amd8111e_config_ipg(struct net_device *dev)
+static void amd8111e_config_ipg(struct timer_list *t)
 {
-	struct amd8111e_priv *lp = netdev_priv(dev);
+	struct amd8111e_priv *lp = timer_container_of(lp, t,
+						      ipg_data.ipg_timer);
 	struct ipg_info *ipg_data = &lp->ipg_data;
 	void __iomem *mmio = lp->mmio;
 	unsigned int prev_col_cnt = ipg_data->col_cnt;
 	unsigned int total_col_cnt;
 	unsigned int tmp_ipg;
 
-	if(lp->link_config.duplex == DUPLEX_FULL){
+	if (lp->link_config.duplex == DUPLEX_FULL) {
 		ipg_data->ipg = DEFAULT_IPG;
 		return;
 	}
 
-	if(ipg_data->ipg_state == SSTATE){
+	if (ipg_data->ipg_state == SSTATE) {
 
-		if(ipg_data->timer_tick == IPG_STABLE_TIME){
+		if (ipg_data->timer_tick == IPG_STABLE_TIME) {
 
 			ipg_data->timer_tick = 0;
 			ipg_data->ipg = MIN_IPG - IPG_STEP;
@@ -1697,7 +1668,7 @@ static void amd8111e_config_ipg(struct net_device *dev)
 			ipg_data->timer_tick++;
 	}
 
-	if(ipg_data->ipg_state == CSTATE){
+	if (ipg_data->ipg_state == CSTATE) {
 
 		/* Get the current collision count */
 
@@ -1705,10 +1676,10 @@ static void amd8111e_config_ipg(struct net_device *dev)
 				amd8111e_read_mib(mmio, xmt_collisions);
 
 		if ((total_col_cnt - prev_col_cnt) <
-				(ipg_data->diff_col_cnt)){
+				(ipg_data->diff_col_cnt)) {
 
 			ipg_data->diff_col_cnt =
-				total_col_cnt - prev_col_cnt ;
+				total_col_cnt - prev_col_cnt;
 
 			ipg_data->ipg = ipg_data->current_ipg;
 		}
@@ -1717,14 +1688,14 @@ static void amd8111e_config_ipg(struct net_device *dev)
 
 		if (ipg_data->current_ipg <= MAX_IPG)
 			tmp_ipg = ipg_data->current_ipg;
-		else{
+		else {
 			tmp_ipg = ipg_data->ipg;
 			ipg_data->ipg_state = SSTATE;
 		}
 		writew((u32)tmp_ipg, mmio + IPG);
 		writew((u32)(tmp_ipg - IFS1_DELTA), mmio + IFS1);
 	}
-	 mod_timer(&lp->ipg_data.ipg_timer, jiffies + IPG_CONVERGE_JIFFIES);
+	mod_timer(&lp->ipg_data.ipg_timer, jiffies + IPG_CONVERGE_JIFFIES);
 	return;
 
 }
@@ -1758,7 +1729,7 @@ static const struct net_device_ops amd8111e_netdev_ops = {
 	.ndo_set_rx_mode	= amd8111e_set_multicast_list,
 	.ndo_validate_addr	= eth_validate_addr,
 	.ndo_set_mac_address	= amd8111e_set_mac_address,
-	.ndo_do_ioctl		= amd8111e_ioctl,
+	.ndo_eth_ioctl		= amd8111e_ioctl,
 	.ndo_change_mtu		= amd8111e_change_mtu,
 #ifdef CONFIG_NET_POLL_CONTROLLER
 	.ndo_poll_controller	 = amd8111e_poll,
@@ -1769,24 +1740,25 @@ static int amd8111e_probe_one(struct pci_dev *pdev,
 				  const struct pci_device_id *ent)
 {
 	int err, i;
-	unsigned long reg_addr,reg_len;
+	unsigned long reg_addr, reg_len;
 	struct amd8111e_priv *lp;
 	struct net_device *dev;
+	u8 addr[ETH_ALEN];
 
 	err = pci_enable_device(pdev);
-	if(err){
+	if (err) {
 		dev_err(&pdev->dev, "Cannot enable new PCI device\n");
 		return err;
 	}
 
-	if(!(pci_resource_flags(pdev, 0) & IORESOURCE_MEM)){
+	if (!(pci_resource_flags(pdev, 0) & IORESOURCE_MEM)) {
 		dev_err(&pdev->dev, "Cannot find PCI base address\n");
 		err = -ENODEV;
 		goto err_disable_pdev;
 	}
 
 	err = pci_request_regions(pdev, MODULE_NAME);
-	if(err){
+	if (err) {
 		dev_err(&pdev->dev, "Cannot obtain PCI resources\n");
 		goto err_disable_pdev;
 	}
@@ -1801,7 +1773,7 @@ static int amd8111e_probe_one(struct pci_dev *pdev,
 	}
 
 	/* Initialize DMA */
-	if (pci_set_dma_mask(pdev, DMA_BIT_MASK(32)) < 0) {
+	if (dma_set_mask(&pdev->dev, DMA_BIT_MASK(32)) < 0) {
 		dev_err(&pdev->dev, "DMA not supported\n");
 		err = -ENODEV;
 		goto err_free_reg;
@@ -1819,13 +1791,12 @@ static int amd8111e_probe_one(struct pci_dev *pdev,
 	SET_NETDEV_DEV(dev, &pdev->dev);
 
 #if AMD8111E_VLAN_TAG_USED
-	dev->features |= NETIF_F_HW_VLAN_CTAG_TX | NETIF_F_HW_VLAN_CTAG_RX ;
+	dev->features |= NETIF_F_HW_VLAN_CTAG_TX | NETIF_F_HW_VLAN_CTAG_RX;
 #endif
 
 	lp = netdev_priv(dev);
 	lp->pci_dev = pdev;
 	lp->amd8111e_net_dev = dev;
-	lp->pm_cap = pdev->pm_cap;
 
 	spin_lock_init(&lp->lock);
 
@@ -1838,28 +1809,26 @@ static int amd8111e_probe_one(struct pci_dev *pdev,
 
 	/* Initializing MAC address */
 	for (i = 0; i < ETH_ALEN; i++)
-		dev->dev_addr[i] = readb(lp->mmio + PADR + i);
+		addr[i] = readb(lp->mmio + PADR + i);
+	eth_hw_addr_set(dev, addr);
 
 	/* Setting user defined parametrs */
 	lp->ext_phy_option = speed_duplex[card_idx];
-	if(coalesce[card_idx])
+	if (coalesce[card_idx])
 		lp->options |= OPTION_INTR_COAL_ENABLE;
-	if(dynamic_ipg[card_idx++])
+	if (dynamic_ipg[card_idx++])
 		lp->options |= OPTION_DYN_IPG_ENABLE;
 
 
 	/* Initialize driver entry points */
 	dev->netdev_ops = &amd8111e_netdev_ops;
 	dev->ethtool_ops = &ops;
-	dev->irq =pdev->irq;
+	dev->irq = pdev->irq;
 	dev->watchdog_timeo = AMD8111E_TX_TIMEOUT;
 	dev->min_mtu = AMD8111E_MIN_MTU;
 	dev->max_mtu = AMD8111E_MAX_MTU;
-	netif_napi_add(dev, &lp->napi, amd8111e_rx_poll, 32);
+	netif_napi_add_weight(dev, &lp->napi, amd8111e_rx_poll, 32);
 
-#if AMD8111E_VLAN_TAG_USED
-	dev->features |= NETIF_F_HW_VLAN_CTAG_TX | NETIF_F_HW_VLAN_CTAG_RX;
-#endif
 	/* Probe the external PHY */
 	amd8111e_probe_ext_phy(dev);
 
@@ -1882,10 +1851,8 @@ static int amd8111e_probe_one(struct pci_dev *pdev,
 	pci_set_drvdata(pdev, dev);
 
 	/* Initialize software ipg timer */
-	if(lp->options & OPTION_DYN_IPG_ENABLE){
-		init_timer(&lp->ipg_data.ipg_timer);
-		lp->ipg_data.ipg_timer.data = (unsigned long) dev;
-		lp->ipg_data.ipg_timer.function = (void *)&amd8111e_config_ipg;
+	if (lp->options & OPTION_DYN_IPG_ENABLE) {
+		timer_setup(&lp->ipg_data.ipg_timer, amd8111e_config_ipg, 0);
 		lp->ipg_data.ipg_timer.expires = jiffies +
 						 IPG_CONVERGE_JIFFIES;
 		lp->ipg_data.ipg = DEFAULT_IPG;
@@ -1893,8 +1860,7 @@ static int amd8111e_probe_one(struct pci_dev *pdev,
 	}
 
 	/*  display driver and device information */
-    	chip_version = (readl(lp->mmio + CHIPID) & 0xf0000000)>>28;
-	dev_info(&pdev->dev, "AMD-8111e Driver Version: %s\n", MODULE_VERS);
+	chip_version = (readl(lp->mmio + CHIPID) & 0xf0000000) >> 28;
 	dev_info(&pdev->dev, "[ Rev %x ] PCI 10/100BaseT Ethernet %pM\n",
 		 chip_version, dev->dev_addr);
 	if (lp->ext_phy_id)
@@ -1903,7 +1869,7 @@ static int amd8111e_probe_one(struct pci_dev *pdev,
 	else
 		dev_info(&pdev->dev, "Couldn't detect MII PHY, assuming address 0x01\n");
 
-    	return 0;
+	return 0;
 
 err_free_dev:
 	free_netdev(dev);
@@ -1940,13 +1906,14 @@ static const struct pci_device_id amd8111e_pci_tbl[] = {
 };
 MODULE_DEVICE_TABLE(pci, amd8111e_pci_tbl);
 
+static SIMPLE_DEV_PM_OPS(amd8111e_pm_ops, amd8111e_suspend, amd8111e_resume);
+
 static struct pci_driver amd8111e_driver = {
-	.name   	= MODULE_NAME,
+	.name		= MODULE_NAME,
 	.id_table	= amd8111e_pci_tbl,
 	.probe		= amd8111e_probe_one,
 	.remove		= amd8111e_remove_one,
-	.suspend	= amd8111e_suspend,
-	.resume		= amd8111e_resume
+	.driver.pm	= &amd8111e_pm_ops
 };
 
 module_pci_driver(amd8111e_driver);
diff --git a/drivers/net/ethernet/amd/amd8111e.h b/drivers/net/ethernet/amd/amd8111e.h
index 2a57b46fd6a6..e4ee4c28800c 100644
--- a/drivers/net/ethernet/amd/amd8111e.h
+++ b/drivers/net/ethernet/amd/amd8111e.h
@@ -1,19 +1,8 @@
+/* SPDX-License-Identifier: GPL-2.0-or-later */
 /*
  * Advanced  Micro Devices Inc. AMD8111E Linux Network Driver
  * Copyright (C) 2003 Advanced Micro Devices
  *
- * This program is free software; you can redistribute it and/or modify
- * it under the terms of the GNU General Public License as published by
- * the Free Software Foundation; either version 2 of the License, or
- * (at your option) any later version.
- *
- * This program is distributed in the hope that it will be useful,
- * but WITHOUT ANY WARRANTY; without even the implied warranty of
- * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
- * GNU General Public License for more details.
- *
- * You should have received a copy of the GNU General Public License
- * along with this program; if not, see <http://www.gnu.org/licenses/>.
 
 Module Name:
 
@@ -21,14 +10,14 @@ Module Name:
 
 Abstract:
 
- 	 AMD8111 based 10/100 Ethernet Controller driver definitions.
+	 AMD8111 based 10/100 Ethernet Controller driver definitions.
 
 Environment:
 
 	Kernel Mode
 
 Revision History:
- 	3.0.0
+	3.0.0
 	   Initial Revision.
 	3.0.1
 */
@@ -561,7 +550,6 @@ typedef enum {
 
 /* Driver definitions */
 
-#define	 PCI_VENDOR_ID_AMD		0x1022
 #define  PCI_DEVICE_ID_AMD8111E_7462	0x7462
 
 #define MAX_UNITS			8 /* Maximum number of devices possible */
@@ -611,7 +599,7 @@ typedef enum {
 #define CSTATE  1
 #define SSTATE  2
 
-/* Assume contoller gets data 10 times the maximum processing time */
+/* Assume controller gets data 10 times the maximum processing time */
 #define  REPEAT_CNT			10
 
 /* amd8111e descriptor flag definitions */
@@ -703,7 +691,7 @@ enum coal_type{
 };
 
 enum coal_mode{
-       	RX_INTR_COAL,
+	RX_INTR_COAL,
 	TX_INTR_COAL,
 	DISABLE_COAL,
 	ENABLE_COAL,
@@ -775,7 +763,6 @@ struct amd8111e_priv{
 	u32 ext_phy_id;
 
 	struct amd8111e_link_config link_config;
-	int pm_cap;
 
 	struct net_device *next;
 	int mii;
diff --git a/drivers/net/ethernet/amd/ariadne.c b/drivers/net/ethernet/amd/ariadne.c
index 5fd7b15b0574..fa201da567ed 100644
--- a/drivers/net/ethernet/amd/ariadne.c
+++ b/drivers/net/ethernet/amd/ariadne.c
@@ -193,7 +193,7 @@ static int ariadne_rx(struct net_device *dev)
 			struct sk_buff *skb;
 
 			skb = netdev_alloc_skb(dev, pkt_len + 2);
-			if (skb == NULL) {
+			if (!skb) {
 				for (i = 0; i < RX_RING_SIZE; i++)
 					if (lowb(priv->rx_ring[(entry + i) % RX_RING_SIZE]->RMD1) & RF_OWN)
 						break;
@@ -441,11 +441,11 @@ static int ariadne_open(struct net_device *dev)
 
 	/* Set the Ethernet Hardware Address */
 	lance->RAP = CSR12;		/* Physical Address Register, PADR[15:0] */
-	lance->RDP = ((u_short *)&dev->dev_addr[0])[0];
+	lance->RDP = ((const u_short *)&dev->dev_addr[0])[0];
 	lance->RAP = CSR13;		/* Physical Address Register, PADR[31:16] */
-	lance->RDP = ((u_short *)&dev->dev_addr[0])[1];
+	lance->RDP = ((const u_short *)&dev->dev_addr[0])[1];
 	lance->RAP = CSR14;		/* Physical Address Register, PADR[47:32] */
-	lance->RDP = ((u_short *)&dev->dev_addr[0])[2];
+	lance->RDP = ((const u_short *)&dev->dev_addr[0])[2];
 
 	/* Set the Init Block Mode */
 	lance->RAP = CSR15;		/* Mode Register */
@@ -530,7 +530,7 @@ static inline void ariadne_reset(struct net_device *dev)
 	netif_start_queue(dev);
 }
 
-static void ariadne_tx_timeout(struct net_device *dev)
+static void ariadne_tx_timeout(struct net_device *dev, unsigned int txqueue)
 {
 	volatile struct Am79C960 *lance = (struct Am79C960 *)dev->base_addr;
 
@@ -692,7 +692,7 @@ static void ariadne_remove_one(struct zorro_dev *z)
 	free_netdev(dev);
 }
 
-static struct zorro_device_id ariadne_zorro_tbl[] = {
+static const struct zorro_device_id ariadne_zorro_tbl[] = {
 	{ ZORRO_PROD_VILLAGE_TRONIC_ARIADNE },
 	{ 0 }
 };
@@ -717,6 +717,7 @@ static int ariadne_init_one(struct zorro_dev *z,
 	unsigned long mem_start = board + ARIADNE_RAM;
 	struct resource *r1, *r2;
 	struct net_device *dev;
+	u8 addr[ETH_ALEN];
 	u32 serial;
 	int err;
 
@@ -730,7 +731,7 @@ static int ariadne_init_one(struct zorro_dev *z,
 	}
 
 	dev = alloc_etherdev(sizeof(struct ariadne_private));
-	if (dev == NULL) {
+	if (!dev) {
 		release_mem_region(base_addr, sizeof(struct Am79C960));
 		release_mem_region(mem_start, ARIADNE_RAM_SIZE);
 		return -ENOMEM;
@@ -740,12 +741,13 @@ static int ariadne_init_one(struct zorro_dev *z,
 	r2->name = dev->name;
 
 	serial = be32_to_cpu(z->rom.er_SerialNumber);
-	dev->dev_addr[0] = 0x00;
-	dev->dev_addr[1] = 0x60;
-	dev->dev_addr[2] = 0x30;
-	dev->dev_addr[3] = (serial >> 16) & 0xff;
-	dev->dev_addr[4] = (serial >> 8) & 0xff;
-	dev->dev_addr[5] = serial & 0xff;
+	addr[0] = 0x00;
+	addr[1] = 0x60;
+	addr[2] = 0x30;
+	addr[3] = (serial >> 16) & 0xff;
+	addr[4] = (serial >> 8) & 0xff;
+	addr[5] = serial & 0xff;
+	eth_hw_addr_set(dev, addr);
 	dev->base_addr = (unsigned long)ZTWO_VADDR(base_addr);
 	dev->mem_start = (unsigned long)ZTWO_VADDR(mem_start);
 	dev->mem_end = dev->mem_start + ARIADNE_RAM_SIZE;
@@ -788,4 +790,5 @@ static void __exit ariadne_cleanup_module(void)
 module_init(ariadne_init_module);
 module_exit(ariadne_cleanup_module);
 
+MODULE_DESCRIPTION("Ariadne Ethernet Driver");
 MODULE_LICENSE("GPL");
diff --git a/drivers/net/ethernet/amd/atarilance.c b/drivers/net/ethernet/amd/atarilance.c
index c5b81268c284..8c8cc7d0f42d 100644
--- a/drivers/net/ethernet/amd/atarilance.c
+++ b/drivers/net/ethernet/amd/atarilance.c
@@ -79,6 +79,7 @@ static int lance_debug = 1;
 #endif
 module_param(lance_debug, int, 0);
 MODULE_PARM_DESC(lance_debug, "atarilance debug level (0-3)");
+MODULE_DESCRIPTION("Atari LANCE Ethernet driver");
 MODULE_LICENSE("GPL");
 
 /* Print debug messages on probing? */
@@ -156,7 +157,7 @@ struct lance_memory {
 	struct lance_init_block	init;
 	struct lance_tx_head	tx_head[TX_RING_SIZE];
 	struct lance_rx_head	rx_head[RX_RING_SIZE];
-	char					packet_area[0];	/* packet data follow after the
+	char					packet_area[];	/* packet data follow after the
 											 * init block and the ring
 											 * descriptors and are located
 											 * at runtime */
@@ -339,13 +340,14 @@ static unsigned long lance_probe1( struct net_device *dev, struct lance_addr
                                    *init_rec );
 static int lance_open( struct net_device *dev );
 static void lance_init_ring( struct net_device *dev );
-static int lance_start_xmit( struct sk_buff *skb, struct net_device *dev );
+static netdev_tx_t lance_start_xmit(struct sk_buff *skb,
+				    struct net_device *dev);
 static irqreturn_t lance_interrupt( int irq, void *dev_id );
 static int lance_rx( struct net_device *dev );
 static int lance_close( struct net_device *dev );
 static void set_multicast_list( struct net_device *dev );
 static int lance_set_mac_address( struct net_device *dev, void *addr );
-static void lance_tx_timeout (struct net_device *dev);
+static void lance_tx_timeout (struct net_device *dev, unsigned int txqueue);
 
 /************************* End of Prototypes **************************/
 
@@ -366,7 +368,7 @@ static void *slow_memcpy( void *dst, const void *src, size_t len )
 }
 
 
-struct net_device * __init atarilance_probe(int unit)
+static struct net_device * __init atarilance_probe(void)
 {
 	int i;
 	static int found;
@@ -381,10 +383,6 @@ struct net_device * __init atarilance_probe(int unit)
 	dev = alloc_etherdev(sizeof(struct lance_private));
 	if (!dev)
 		return ERR_PTR(-ENOMEM);
-	if (unit >= 0) {
-		sprintf(dev->name, "eth%d", unit);
-		netdev_boot_setup_check(dev);
-	}
 
 	for( i = 0; i < N_LANCE_ADDR; ++i ) {
 		if (lance_probe1( dev, &lance_addr_list[i] )) {
@@ -474,6 +472,7 @@ static unsigned long __init lance_probe1( struct net_device *dev,
 	int 					i;
 	static int 				did_version;
 	unsigned short			save1, save2;
+	u8 addr[ETH_ALEN];
 
 	PROBE_PRINT(( "Probing for Lance card at mem %#lx io %#lx\n",
 				  (long)memaddr, (long)ioaddr ));
@@ -583,19 +582,21 @@ static unsigned long __init lance_probe1( struct net_device *dev,
 
 	/* Get the ethernet address */
 	switch( lp->cardtype ) {
-	  case OLD_RIEBL:
+	case OLD_RIEBL:
 		/* No ethernet address! (Set some default address) */
-		memcpy(dev->dev_addr, OldRieblDefHwaddr, ETH_ALEN);
+		eth_hw_addr_set(dev, OldRieblDefHwaddr);
 		break;
-	  case NEW_RIEBL:
-		lp->memcpy_f(dev->dev_addr, RIEBL_HWADDR_ADDR, ETH_ALEN);
+	case NEW_RIEBL:
+		lp->memcpy_f(addr, RIEBL_HWADDR_ADDR, ETH_ALEN);
+		eth_hw_addr_set(dev, addr);
 		break;
-	  case PAM_CARD:
+	case PAM_CARD:
 		i = IO->eeprom;
 		for( i = 0; i < 6; ++i )
-			dev->dev_addr[i] =
+			addr[i] =
 				((((unsigned short *)MEM)[i*2] & 0x0f) << 4) |
 				((((unsigned short *)MEM)[i*2+1] & 0x0f));
+		eth_hw_addr_set(dev, addr);
 		i = IO->mem;
 		break;
 	}
@@ -705,7 +706,7 @@ static void lance_init_ring( struct net_device *dev )
 		CHECK_OFFSET(offset);
 		MEM->tx_head[i].base = offset;
 		MEM->tx_head[i].flag = TMD1_OWN_HOST;
- 		MEM->tx_head[i].base_hi = 0;
+		MEM->tx_head[i].base_hi = 0;
 		MEM->tx_head[i].length = 0;
 		MEM->tx_head[i].misc = 0;
 		offset += PKT_BUF_SZ;
@@ -726,7 +727,7 @@ static void lance_init_ring( struct net_device *dev )
 /* XXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXX */
 
 
-static void lance_tx_timeout (struct net_device *dev)
+static void lance_tx_timeout (struct net_device *dev, unsigned int txqueue)
 {
 	struct lance_private *lp = netdev_priv(dev);
 	struct lance_ioreg	 *IO = lp->iobase;
@@ -769,7 +770,8 @@ static void lance_tx_timeout (struct net_device *dev)
 
 /* XXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXX */
 
-static int lance_start_xmit( struct sk_buff *skb, struct net_device *dev )
+static netdev_tx_t
+lance_start_xmit(struct sk_buff *skb, struct net_device *dev)
 {
 	struct lance_private *lp = netdev_priv(dev);
 	struct lance_ioreg	 *IO = lp->iobase;
@@ -823,7 +825,7 @@ static int lance_start_xmit( struct sk_buff *skb, struct net_device *dev )
 	lp->memcpy_f( PKTBUF_ADDR(head), (void *)skb->data, skb->len );
 	head->flag = TMD1_OWN_CHIP | TMD1_ENP | TMD1_STP;
 	dev->stats.tx_bytes += skb->len;
-	dev_kfree_skb( skb );
+	dev_consume_skb_irq(skb);
 	lp->cur_tx++;
 	while( lp->cur_tx >= TX_RING_SIZE && lp->dirty_tx >= TX_RING_SIZE ) {
 		lp->cur_tx -= TX_RING_SIZE;
@@ -853,7 +855,7 @@ static irqreturn_t lance_interrupt( int irq, void *dev_id )
 	int csr0, boguscnt = 10;
 	int handled = 0;
 
-	if (dev == NULL) {
+	if (!dev) {
 		DPRINTK( 1, ( "lance_interrupt(): interrupt for unknown device.\n" ));
 		return IRQ_NONE;
 	}
@@ -994,7 +996,7 @@ static int lance_rx( struct net_device *dev )
 			}
 			else {
 				skb = netdev_alloc_skb(dev, pkt_len + 2);
-				if (skb == NULL) {
+				if (!skb) {
 					for( i = 0; i < RX_RING_SIZE; i++ )
 						if (MEM->rx_head[(entry+i) & RX_RING_MOD_MASK].flag &
 							RMD1_OWN_CHIP)
@@ -1125,7 +1127,7 @@ static int lance_set_mac_address( struct net_device *dev, void *addr )
 		return -EIO;
 	}
 
-	memcpy( dev->dev_addr, saddr->sa_data, dev->addr_len );
+	eth_hw_addr_set(dev, saddr->sa_data);
 	for( i = 0; i < 6; i++ )
 		MEM->init.hwaddr[i] = dev->dev_addr[i^1]; /* <- 16 bit swap! */
 	lp->memcpy_f( RIEBL_HWADDR_ADDR, dev->dev_addr, 6 );
@@ -1135,13 +1137,11 @@ static int lance_set_mac_address( struct net_device *dev, void *addr )
 	return 0;
 }
 
-
-#ifdef MODULE
 static struct net_device *atarilance_dev;
 
 static int __init atarilance_module_init(void)
 {
-	atarilance_dev = atarilance_probe(-1);
+	atarilance_dev = atarilance_probe();
 	return PTR_ERR_OR_ZERO(atarilance_dev);
 }
 
@@ -1153,12 +1153,3 @@ static void __exit atarilance_module_exit(void)
 }
 module_init(atarilance_module_init);
 module_exit(atarilance_module_exit);
-#endif /* MODULE */
-
-
-/*
- * Local variables:
- *  c-indent-level: 4
- *  tab-width: 4
- * End:
- */
diff --git a/drivers/net/ethernet/amd/au1000_eth.c b/drivers/net/ethernet/amd/au1000_eth.c
index a3c90fe5de00..9d35ac348ebe 100644
--- a/drivers/net/ethernet/amd/au1000_eth.c
+++ b/drivers/net/ethernet/amd/au1000_eth.c
@@ -1,3 +1,4 @@
+// SPDX-License-Identifier: GPL-2.0-only
 /*
  *
  * Alchemy Au1x00 ethernet driver
@@ -14,24 +15,6 @@
  *
  * Author: MontaVista Software, Inc.
  *		ppopov@mvista.com or source@mvista.com
- *
- * ########################################################################
- *
- *  This program is free software; you can distribute it and/or modify it
- *  under the terms of the GNU General Public License (Version 2) as
- *  published by the Free Software Foundation.
- *
- *  This program is distributed in the hope it will be useful, but WITHOUT
- *  ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
- *  FITNESS FOR A PARTICULAR PURPOSE.  See the GNU General Public License
- *  for more details.
- *
- *  You should have received a copy of the GNU General Public License along
- *  with this program; if not, see <http://www.gnu.org/licenses/>.
- *
- * ########################################################################
- *
- *
  */
 #define pr_fmt(fmt) KBUILD_MODNAME ": " fmt
 
@@ -80,14 +63,12 @@ static int au1000_debug = 3;
 				NETIF_MSG_LINK)
 
 #define DRV_NAME	"au1000_eth"
-#define DRV_VERSION	"1.7"
 #define DRV_AUTHOR	"Pete Popov <ppopov@embeddedalley.com>"
 #define DRV_DESC	"Au1xxx on-chip Ethernet driver"
 
 MODULE_AUTHOR(DRV_AUTHOR);
 MODULE_DESCRIPTION(DRV_DESC);
 MODULE_LICENSE("GPL");
-MODULE_VERSION(DRV_VERSION);
 
 /* AU1000 MAC registers and bits */
 #define MAC_CONTROL		0x0
@@ -260,7 +241,6 @@ MODULE_VERSION(DRV_VERSION);
  * ps: make sure the used irqs are configured properly in the board
  * specific irq-map
  */
-
 static void au1000_enable_mac(struct net_device *dev, int force_reset)
 {
 	unsigned long flags;
@@ -564,17 +544,7 @@ static int au1000_mii_probe(struct net_device *dev)
 		return PTR_ERR(phydev);
 	}
 
-	/* mask with MAC supported features */
-	phydev->supported &= (SUPPORTED_10baseT_Half
-			      | SUPPORTED_10baseT_Full
-			      | SUPPORTED_100baseT_Half
-			      | SUPPORTED_100baseT_Full
-			      | SUPPORTED_Autoneg
-			      /* | SUPPORTED_Pause | SUPPORTED_Asym_Pause */
-			      | SUPPORTED_MII
-			      | SUPPORTED_TP);
-
-	phydev->advertising = phydev->supported;
+	phy_set_max_speed(phydev, SPEED_100);
 
 	aup->old_link = 0;
 	aup->old_speed = 0;
@@ -585,7 +555,6 @@ static int au1000_mii_probe(struct net_device *dev)
 	return 0;
 }
 
-
 /*
  * Buffer allocation/deallocation routines. The buffer descriptor returned
  * has the virtual and dma address of a buffer suitable for
@@ -602,7 +571,7 @@ static struct db_dest *au1000_GetFreeDB(struct au1000_private *aup)
 	return pDB;
 }
 
-void au1000_ReleaseDB(struct au1000_private *aup, struct db_dest *pDB)
+static void au1000_ReleaseDB(struct au1000_private *aup, struct db_dest *pDB)
 {
 	struct db_dest *pDBfree = aup->pDBfree;
 	if (pDBfree)
@@ -676,14 +645,12 @@ au1000_setup_hw_rings(struct au1000_private *aup, void __iomem *tx_base)
 /*
  * ethtool operations
  */
-
 static void
 au1000_get_drvinfo(struct net_device *dev, struct ethtool_drvinfo *info)
 {
 	struct au1000_private *aup = netdev_priv(dev);
 
-	strlcpy(info->driver, DRV_NAME, sizeof(info->driver));
-	strlcpy(info->version, DRV_VERSION, sizeof(info->version));
+	strscpy(info->driver, DRV_NAME, sizeof(info->driver));
 	snprintf(info->bus_info, sizeof(info->bus_info), "%s %d", DRV_NAME,
 		 aup->mac_id);
 }
@@ -709,7 +676,6 @@ static const struct ethtool_ops au1000_ethtool_ops = {
 	.set_link_ksettings = phy_ethtool_set_link_ksettings,
 };
 
-
 /*
  * Initialize the interface.
  *
@@ -820,7 +786,7 @@ static int au1000_rx(struct net_device *dev)
 			frmlen = (status & RX_FRAME_LEN_MASK);
 			frmlen -= 4; /* Remove FCS */
 			skb = netdev_alloc_skb(dev, frmlen + 2);
-			if (skb == NULL) {
+			if (!skb) {
 				dev->stats.rx_dropped++;
 				continue;
 			}
@@ -854,7 +820,7 @@ static int au1000_rx(struct net_device *dev)
 				pr_cont("\n");
 			}
 		}
-		prxd->buff_stat = (u32)(pDB->dma_addr | RX_DMA_ENABLE);
+		prxd->buff_stat = lower_32_bits(pDB->dma_addr) | RX_DMA_ENABLE;
 		aup->rx_head = (aup->rx_head + 1) & (NUM_RX_DMA - 1);
 		wmb(); /* drain writebuffer */
 
@@ -950,11 +916,8 @@ static int au1000_open(struct net_device *dev)
 		return retval;
 	}
 
-	if (dev->phydev) {
-		/* cause the PHY state machine to schedule a link state check */
-		dev->phydev->state = PHY_CHANGELINK;
+	if (dev->phydev)
 		phy_start(dev->phydev);
-	}
 
 	netif_start_queue(dev);
 
@@ -1033,7 +996,7 @@ static netdev_tx_t au1000_tx(struct sk_buff *skb, struct net_device *dev)
 	ps->tx_packets++;
 	ps->tx_bytes += ptxd->len;
 
-	ptxd->buff_stat = pDB->dma_addr | TX_DMA_ENABLE;
+	ptxd->buff_stat = lower_32_bits(pDB->dma_addr) | TX_DMA_ENABLE;
 	wmb(); /* drain writebuffer */
 	dev_kfree_skb(skb);
 	aup->tx_head = (aup->tx_head + 1) & (NUM_TX_DMA - 1);
@@ -1044,7 +1007,7 @@ static netdev_tx_t au1000_tx(struct sk_buff *skb, struct net_device *dev)
  * The Tx ring has been full longer than the watchdog timeout
  * value. The transmitter must be hung?
  */
-static void au1000_tx_timeout(struct net_device *dev)
+static void au1000_tx_timeout(struct net_device *dev, unsigned int txqueue)
 {
 	netdev_err(dev, "au1000_tx_timeout: dev=%p\n", dev);
 	au1000_reset_mac(dev);
@@ -1083,23 +1046,12 @@ static void au1000_multicast_list(struct net_device *dev)
 	writel(reg, &aup->mac->control);
 }
 
-static int au1000_ioctl(struct net_device *dev, struct ifreq *rq, int cmd)
-{
-	if (!netif_running(dev))
-		return -EINVAL;
-
-	if (!dev->phydev)
-		return -EINVAL; /* PHY not controllable */
-
-	return phy_mii_ioctl(dev->phydev, rq, cmd);
-}
-
 static const struct net_device_ops au1000_netdev_ops = {
 	.ndo_open		= au1000_open,
 	.ndo_stop		= au1000_close,
 	.ndo_start_xmit		= au1000_tx,
 	.ndo_set_rx_mode	= au1000_multicast_list,
-	.ndo_do_ioctl		= au1000_ioctl,
+	.ndo_eth_ioctl		= phy_do_ioctl_running,
 	.ndo_tx_timeout		= au1000_tx_timeout,
 	.ndo_set_mac_address	= eth_mac_addr,
 	.ndo_validate_addr	= eth_validate_addr,
@@ -1130,7 +1082,6 @@ static int au1000_probe(struct platform_device *pdev)
 
 	irq = platform_get_irq(pdev, 0);
 	if (irq < 0) {
-		dev_err(&pdev->dev, "failed to retrieve IRQ\n");
 		err = -ENODEV;
 		goto out;
 	}
@@ -1180,9 +1131,9 @@ static int au1000_probe(struct platform_device *pdev)
 	/* Allocate the data buffers
 	 * Snooping works fine with eth on all au1xxx
 	 */
-	aup->vaddr = (u32)dma_alloc_noncoherent(NULL, MAX_BUF_SIZE *
-						(NUM_TX_BUFFS + NUM_RX_BUFFS),
-						&aup->dma_addr,	0);
+	aup->vaddr = dma_alloc_coherent(&pdev->dev, MAX_BUF_SIZE *
+					(NUM_TX_BUFFS + NUM_RX_BUFFS),
+					&aup->dma_addr, 0);
 	if (!aup->vaddr) {
 		dev_err(&pdev->dev, "failed to allocate data buffers\n");
 		err = -ENOMEM;
@@ -1191,7 +1142,7 @@ static int au1000_probe(struct platform_device *pdev)
 
 	/* aup->mac is the base address of the MAC's registers */
 	aup->mac = (struct mac_reg *)
-			ioremap_nocache(base->start, resource_size(base));
+			ioremap(base->start, resource_size(base));
 	if (!aup->mac) {
 		dev_err(&pdev->dev, "failed to ioremap MAC registers\n");
 		err = -ENXIO;
@@ -1199,7 +1150,7 @@ static int au1000_probe(struct platform_device *pdev)
 	}
 
 	/* Setup some variables for quick register address access */
-	aup->enable = (u32 *)ioremap_nocache(macen->start,
+	aup->enable = (u32 *)ioremap(macen->start,
 						resource_size(macen));
 	if (!aup->enable) {
 		dev_err(&pdev->dev, "failed to ioremap MAC enable register\n");
@@ -1208,7 +1159,7 @@ static int au1000_probe(struct platform_device *pdev)
 	}
 	aup->mac_id = pdev->id;
 
-	aup->macdma = ioremap_nocache(macdma->start, resource_size(macdma));
+	aup->macdma = ioremap(macdma->start, resource_size(macdma));
 	if (!aup->macdma) {
 		dev_err(&pdev->dev, "failed to ioremap MACDMA registers\n");
 		err = -ENXIO;
@@ -1227,7 +1178,7 @@ static int au1000_probe(struct platform_device *pdev)
 		aup->phy1_search_mac0 = 1;
 	} else {
 		if (is_valid_ether_addr(pd->mac)) {
-			memcpy(dev->dev_addr, pd->mac, ETH_ALEN);
+			eth_hw_addr_set(dev, pd->mac);
 		} else {
 			/* Set a random MAC since no valid provided by platform_data. */
 			eth_hw_addr_random(dev);
@@ -1248,7 +1199,7 @@ static int au1000_probe(struct platform_device *pdev)
 	}
 
 	aup->mii_bus = mdiobus_alloc();
-	if (aup->mii_bus == NULL) {
+	if (!aup->mii_bus) {
 		dev_err(&pdev->dev, "failed to allocate mdiobus structure\n");
 		err = -ENOMEM;
 		goto err_mdiobus_alloc;
@@ -1283,8 +1234,8 @@ static int au1000_probe(struct platform_device *pdev)
 	for (i = 0; i < (NUM_TX_BUFFS+NUM_RX_BUFFS); i++) {
 		pDB->pnext = pDBfree;
 		pDBfree = pDB;
-		pDB->vaddr = (u32 *)((unsigned)aup->vaddr + MAX_BUF_SIZE*i);
-		pDB->dma_addr = (dma_addr_t)virt_to_bus(pDB->vaddr);
+		pDB->vaddr = aup->vaddr + MAX_BUF_SIZE * i;
+		pDB->dma_addr = aup->dma_addr + MAX_BUF_SIZE * i;
 		pDB++;
 	}
 	aup->pDBfree = pDBfree;
@@ -1295,17 +1246,16 @@ static int au1000_probe(struct platform_device *pdev)
 		if (!pDB)
 			goto err_out;
 
-		aup->rx_dma_ring[i]->buff_stat = (unsigned)pDB->dma_addr;
+		aup->rx_dma_ring[i]->buff_stat = lower_32_bits(pDB->dma_addr);
 		aup->rx_db_inuse[i] = pDB;
 	}
 
-	err = -ENODEV;
 	for (i = 0; i < NUM_TX_DMA; i++) {
 		pDB = au1000_GetFreeDB(aup);
 		if (!pDB)
 			goto err_out;
 
-		aup->tx_dma_ring[i]->buff_stat = (unsigned)pDB->dma_addr;
+		aup->tx_dma_ring[i]->buff_stat = lower_32_bits(pDB->dma_addr);
 		aup->tx_dma_ring[i]->len = 0;
 		aup->tx_db_inuse[i] = pDB;
 	}
@@ -1331,12 +1281,10 @@ static int au1000_probe(struct platform_device *pdev)
 	netdev_info(dev, "Au1xx0 Ethernet found at 0x%lx, irq %d\n",
 			(unsigned long)base->start, irq);
 
-	pr_info_once("%s version %s %s\n", DRV_NAME, DRV_VERSION, DRV_AUTHOR);
-
 	return 0;
 
 err_out:
-	if (aup->mii_bus != NULL)
+	if (aup->mii_bus)
 		mdiobus_unregister(aup->mii_bus);
 
 	/* here we should have a valid dev plus aup-> register addresses
@@ -1361,8 +1309,8 @@ err_remap3:
 err_remap2:
 	iounmap(aup->mac);
 err_remap1:
-	dma_free_noncoherent(NULL, MAX_BUF_SIZE * (NUM_TX_BUFFS + NUM_RX_BUFFS),
-			     (void *)aup->vaddr, aup->dma_addr);
+	dma_free_coherent(&pdev->dev, MAX_BUF_SIZE * (NUM_TX_BUFFS + NUM_RX_BUFFS),
+			  aup->vaddr, aup->dma_addr);
 err_vaddr:
 	free_netdev(dev);
 err_alloc:
@@ -1375,7 +1323,7 @@ out:
 	return err;
 }
 
-static int au1000_remove(struct platform_device *pdev)
+static void au1000_remove(struct platform_device *pdev)
 {
 	struct net_device *dev = platform_get_drvdata(pdev);
 	struct au1000_private *aup = netdev_priv(dev);
@@ -1394,9 +1342,8 @@ static int au1000_remove(struct platform_device *pdev)
 		if (aup->tx_db_inuse[i])
 			au1000_ReleaseDB(aup, aup->tx_db_inuse[i]);
 
-	dma_free_noncoherent(NULL, MAX_BUF_SIZE *
-			(NUM_TX_BUFFS + NUM_RX_BUFFS),
-			(void *)aup->vaddr, aup->dma_addr);
+	dma_free_coherent(&pdev->dev, MAX_BUF_SIZE * (NUM_TX_BUFFS + NUM_RX_BUFFS),
+			  aup->vaddr, aup->dma_addr);
 
 	iounmap(aup->macdma);
 	iounmap(aup->mac);
@@ -1412,8 +1359,6 @@ static int au1000_remove(struct platform_device *pdev)
 	release_mem_region(macen->start, resource_size(macen));
 
 	free_netdev(dev);
-
-	return 0;
 }
 
 static struct platform_driver au1000_eth_driver = {
diff --git a/drivers/net/ethernet/amd/au1000_eth.h b/drivers/net/ethernet/amd/au1000_eth.h
index 4c47c2377d74..2489c2f4fd8a 100644
--- a/drivers/net/ethernet/amd/au1000_eth.h
+++ b/drivers/net/ethernet/amd/au1000_eth.h
@@ -1,3 +1,4 @@
+/* SPDX-License-Identifier: GPL-2.0-only */
 /*
  *
  * Alchemy Au1x00 ethernet driver include file
@@ -5,24 +6,6 @@
  * Author: Pete Popov <ppopov@mvista.com>
  *
  * Copyright 2001 MontaVista Software Inc.
- *
- * ########################################################################
- *
- *  This program is free software; you can distribute it and/or modify it
- *  under the terms of the GNU General Public License (Version 2) as
- *  published by the Free Software Foundation.
- *
- *  This program is distributed in the hope it will be useful, but WITHOUT
- *  ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
- *  FITNESS FOR A PARTICULAR PURPOSE.  See the GNU General Public License
- *  for more details.
- *
- *  You should have received a copy of the GNU General Public License along
- *  with this program; if not, see <http://www.gnu.org/licenses/>.
- *
- * ########################################################################
- *
- *
  */
 
 
@@ -123,8 +106,8 @@ struct au1000_private {
 	struct mac_reg *mac;  /* mac registers                      */
 	u32 *enable;     /* address of MAC Enable Register     */
 	void __iomem *macdma;	/* base of MAC DMA port */
-	u32 vaddr;                /* virtual address of rx/tx buffers   */
-	dma_addr_t dma_addr;      /* dma address of rx/tx buffers       */
+	void *vaddr;		/* virtual address of rx/tx buffers   */
+	dma_addr_t dma_addr;	/* dma address of rx/tx buffers       */
 
 	spinlock_t lock;       /* Serialise access to device */
 
diff --git a/drivers/net/ethernet/amd/declance.c b/drivers/net/ethernet/amd/declance.c
index 82cc81385033..8d05a0c5f2d5 100644
--- a/drivers/net/ethernet/amd/declance.c
+++ b/drivers/net/ethernet/amd/declance.c
@@ -1,3 +1,4 @@
+// SPDX-License-Identifier: GPL-2.0-only
 /*
  *    Lance ethernet driver for the MIPS processor based
  *      DECstation family
@@ -260,6 +261,7 @@ struct lance_private {
 	unsigned short busmaster_regval;
 
 	struct timer_list       multicast_timer;
+	struct net_device	*dev;
 
 	/* Pointers to the ring buffers as seen from the CPU */
 	char *rx_buf_ptr_cpu[RX_RING_SIZE];
@@ -606,7 +608,7 @@ static int lance_rx(struct net_device *dev)
 			len = (*rds_ptr(rd, mblength, lp->type) & 0xfff) - 4;
 			skb = netdev_alloc_skb(dev, len + 2);
 
-			if (skb == 0) {
+			if (!skb) {
 				dev->stats.rx_dropped++;
 				*rds_ptr(rd, mblength, lp->type) = 0;
 				*rds_ptr(rd, rmd1, lp->type) =
@@ -840,7 +842,7 @@ static int lance_close(struct net_device *dev)
 	volatile struct lance_regs *ll = lp->ll;
 
 	netif_stop_queue(dev);
-	del_timer_sync(&lp->multicast_timer);
+	timer_delete_sync(&lp->multicast_timer);
 
 	/* Stop the card */
 	writereg(&ll->rap, LE_CSR0);
@@ -882,7 +884,7 @@ static inline int lance_reset(struct net_device *dev)
 	return status;
 }
 
-static void lance_tx_timeout(struct net_device *dev)
+static void lance_tx_timeout(struct net_device *dev, unsigned int txqueue)
 {
 	struct lance_private *lp = netdev_priv(dev);
 	volatile struct lance_regs *ll = lp->ll;
@@ -893,7 +895,7 @@ static void lance_tx_timeout(struct net_device *dev)
 	netif_wake_queue(dev);
 }
 
-static int lance_start_xmit(struct sk_buff *skb, struct net_device *dev)
+static netdev_tx_t lance_start_xmit(struct sk_buff *skb, struct net_device *dev)
 {
 	struct lance_private *lp = netdev_priv(dev);
 	volatile struct lance_regs *ll = lp->ll;
@@ -935,7 +937,7 @@ static int lance_start_xmit(struct sk_buff *skb, struct net_device *dev)
 
 	dev_kfree_skb(skb);
 
- 	return NETDEV_TX_OK;
+	return NETDEV_TX_OK;
 }
 
 static void lance_load_multicast(struct net_device *dev)
@@ -1000,9 +1002,10 @@ static void lance_set_multicast(struct net_device *dev)
 	netif_wake_queue(dev);
 }
 
-static void lance_set_multicast_retry(unsigned long _opaque)
+static void lance_set_multicast_retry(struct timer_list *t)
 {
-	struct net_device *dev = (struct net_device *) _opaque;
+	struct lance_private *lp = timer_container_of(lp, t, multicast_timer);
+	struct net_device *dev = lp->dev;
 
 	lance_set_multicast(dev);
 }
@@ -1029,6 +1032,8 @@ static int dec_lance_probe(struct device *bdev, const int type)
 	int i, ret;
 	unsigned long esar_base;
 	unsigned char *esar;
+	u8 addr[ETH_ALEN];
+	const char *desc;
 
 	if (dec_lance_debug && version_printed++ == 0)
 		printk(version);
@@ -1214,19 +1219,21 @@ static int dec_lance_probe(struct device *bdev, const int type)
 	 */
 	switch (type) {
 	case ASIC_LANCE:
-		printk("%s: IOASIC onboard LANCE", name);
+		desc = "IOASIC onboard LANCE";
 		break;
 	case PMAD_LANCE:
-		printk("%s: PMAD-AA", name);
+		desc = "PMAD-AA";
 		break;
 	case PMAX_LANCE:
-		printk("%s: PMAX onboard LANCE", name);
+		desc = "PMAX onboard LANCE";
 		break;
 	}
 	for (i = 0; i < 6; i++)
-		dev->dev_addr[i] = esar[i * 4];
+		addr[i] = esar[i * 4];
+	eth_hw_addr_set(dev, addr);
 
-	printk(", addr = %pM, irq = %d\n", dev->dev_addr, dev->irq);
+	printk("%s: %s, addr = %pM, irq = %d\n",
+	       name, desc, dev->dev_addr, dev->irq);
 
 	dev->netdev_ops = &lance_netdev_ops;
 	dev->watchdog_timeo = 5*HZ;
@@ -1246,9 +1253,9 @@ static int dec_lance_probe(struct device *bdev, const int type)
 	 * can occur from interrupts (ex. IPv6).  So we
 	 * use a timer to try again later when necessary. -DaveM
 	 */
-	init_timer(&lp->multicast_timer);
-	lp->multicast_timer.data = (unsigned long) dev;
-	lp->multicast_timer.function = lance_set_multicast_retry;
+	lp->dev = dev;
+	timer_setup(&lp->multicast_timer, lance_set_multicast_retry, 0);
+
 
 	ret = register_netdev(dev);
 	if (ret) {
diff --git a/drivers/net/ethernet/amd/hplance.c b/drivers/net/ethernet/amd/hplance.c
index c3dbf1c8a269..df42294530cb 100644
--- a/drivers/net/ethernet/amd/hplance.c
+++ b/drivers/net/ethernet/amd/hplance.c
@@ -1,3 +1,4 @@
+// SPDX-License-Identifier: GPL-2.0-only
 /* hplance.c  : the  Linux/hp300/lance ethernet driver
  *
  * Copyright (C) 05/1998 Peter Maydell <pmaydell@chiark.greenend.org.uk>
@@ -14,6 +15,7 @@
 #include <linux/delay.h>
 #include <linux/init.h>
 #include <linux/errno.h>
+#include <linux/pgtable.h>
 /* Used for the temporal inet entries and routing */
 #include <linux/socket.h>
 #include <linux/route.h>
@@ -23,7 +25,6 @@
 #include <linux/skbuff.h>
 
 #include <asm/io.h>
-#include <asm/pgtable.h>
 
 #include "hplance.h"
 
@@ -128,6 +129,7 @@ static void hplance_init(struct net_device *dev, struct dio_dev *d)
 {
 	unsigned long va = (d->resource.start + DIO_VIRADDRBASE);
 	struct hplance_private *lp;
+	u8 addr[ETH_ALEN];
 	int i;
 
 	/* reset the board */
@@ -143,9 +145,10 @@ static void hplance_init(struct net_device *dev, struct dio_dev *d)
 		/* The NVRAM holds our ethernet address, one nibble per byte,
 		 * at bytes NVRAMOFF+1,3,5,7,9...
 		 */
-		dev->dev_addr[i] = ((in_8(va + HPLANCE_NVRAMOFF + i*4 + 1) & 0xF) << 4)
+		addr[i] = ((in_8(va + HPLANCE_NVRAMOFF + i*4 + 1) & 0xF) << 4)
 			| (in_8(va + HPLANCE_NVRAMOFF + i*4 + 3) & 0xF);
 	}
+	eth_hw_addr_set(dev, addr);
 
 	lp = netdev_priv(dev);
 	lp->lance.name = d->name;
@@ -169,6 +172,7 @@ static void hplance_init(struct net_device *dev, struct dio_dev *d)
 static void hplance_writerap(void *priv, unsigned short value)
 {
 	struct lance_private *lp = (struct lance_private *)priv;
+
 	do {
 		out_be16(lp->base + HPLANCE_REGOFF + LANCE_RAP, value);
 	} while ((in_8(lp->base + HPLANCE_STATUS) & LE_ACK) == 0);
@@ -177,6 +181,7 @@ static void hplance_writerap(void *priv, unsigned short value)
 static void hplance_writerdp(void *priv, unsigned short value)
 {
 	struct lance_private *lp = (struct lance_private *)priv;
+
 	do {
 		out_be16(lp->base + HPLANCE_REGOFF + LANCE_RDP, value);
 	} while ((in_8(lp->base + HPLANCE_STATUS) & LE_ACK) == 0);
@@ -186,6 +191,7 @@ static unsigned short hplance_readrdp(void *priv)
 {
 	struct lance_private *lp = (struct lance_private *)priv;
 	__u16 value;
+
 	do {
 		value = in_be16(lp->base + HPLANCE_REGOFF + LANCE_RDP);
 	} while ((in_8(lp->base + HPLANCE_STATUS) & LE_ACK) == 0);
@@ -228,4 +234,5 @@ static void __exit hplance_cleanup_module(void)
 module_init(hplance_init_module);
 module_exit(hplance_cleanup_module);
 
+MODULE_DESCRIPTION("HP300 on-board LANCE Ethernet driver");
 MODULE_LICENSE("GPL");
diff --git a/drivers/net/ethernet/amd/hplance.h b/drivers/net/ethernet/amd/hplance.h
index 04aee9e0376a..bc845a2c60c1 100644
--- a/drivers/net/ethernet/amd/hplance.h
+++ b/drivers/net/ethernet/amd/hplance.h
@@ -1,3 +1,4 @@
+/* SPDX-License-Identifier: GPL-2.0 */
 /* Random defines and structures for the HP Lance driver.
  * Copyright (C) 05/1998 Peter Maydell <pmaydell@chiark.greenend.org.uk>
  * Based on the Sun Lance driver and the NetBSD HP Lance driver
diff --git a/drivers/net/ethernet/amd/lance.c b/drivers/net/ethernet/amd/lance.c
index 12a6a93d221b..b1e6620ad41d 100644
--- a/drivers/net/ethernet/amd/lance.c
+++ b/drivers/net/ethernet/amd/lance.c
@@ -59,6 +59,7 @@ static const char version[] = "lance.c:v1.16 2006/11/09 dplatt@3do.com, becker@c
 #include <linux/skbuff.h>
 #include <linux/mm.h>
 #include <linux/bitops.h>
+#include <net/Space.h>
 
 #include <asm/io.h>
 #include <asm/dma.h>
@@ -306,7 +307,7 @@ static irqreturn_t lance_interrupt(int irq, void *dev_id);
 static int lance_close(struct net_device *dev);
 static struct net_device_stats *lance_get_stats(struct net_device *dev);
 static void set_multicast_list(struct net_device *dev);
-static void lance_tx_timeout (struct net_device *dev);
+static void lance_tx_timeout (struct net_device *dev, unsigned int txqueue);
 
 
 
@@ -327,7 +328,7 @@ MODULE_PARM_DESC(dma, "LANCE/PCnet ISA DMA channel (ignored for some devices)");
 MODULE_PARM_DESC(irq, "LANCE/PCnet IRQ number (ignored for some devices)");
 MODULE_PARM_DESC(lance_debug, "LANCE/PCnet debug level (0-7)");
 
-int __init init_module(void)
+static int __init lance_init_module(void)
 {
 	struct net_device *dev;
 	int this_dev, found = 0;
@@ -356,6 +357,7 @@ int __init init_module(void)
 		return 0;
 	return -ENXIO;
 }
+module_init(lance_init_module);
 
 static void cleanup_card(struct net_device *dev)
 {
@@ -368,7 +370,7 @@ static void cleanup_card(struct net_device *dev)
 	kfree(lp);
 }
 
-void __exit cleanup_module(void)
+static void __exit lance_cleanup_module(void)
 {
 	int this_dev;
 
@@ -381,7 +383,9 @@ void __exit cleanup_module(void)
 		}
 	}
 }
+module_exit(lance_cleanup_module);
 #endif /* MODULE */
+MODULE_DESCRIPTION("AMD LANCE/PCnet Ethernet driver");
 MODULE_LICENSE("GPL");
 
 
@@ -478,6 +482,7 @@ static int __init lance_probe1(struct net_device *dev, int ioaddr, int irq, int
 	unsigned long flags;
 	int err = -ENOMEM;
 	void __iomem *bios;
+	u8 addr[ETH_ALEN];
 
 	/* First we look for special cases.
 	   Check for HP's on-board ethernet by looking for 'HP' in the BIOS.
@@ -539,7 +544,8 @@ static int __init lance_probe1(struct net_device *dev, int ioaddr, int irq, int
 	/* There is a 16 byte station address PROM at the base address.
 	   The first six bytes are the station address. */
 	for (i = 0; i < 6; i++)
-		dev->dev_addr[i] = inb(ioaddr + i);
+		addr[i] = inb(ioaddr + i);
+	eth_hw_addr_set(dev, addr);
 	printk("%pM", dev->dev_addr);
 
 	dev->base_addr = ioaddr;
@@ -551,13 +557,13 @@ static int __init lance_probe1(struct net_device *dev, int ioaddr, int irq, int
 	if (lance_debug > 6) printk(" (#0x%05lx)", (unsigned long)lp);
 	dev->ml_priv = lp;
 	lp->name = chipname;
-	lp->rx_buffs = (unsigned long)kmalloc(PKT_BUF_SZ*RX_RING_SIZE,
-						  GFP_DMA | GFP_KERNEL);
+	lp->rx_buffs = (unsigned long)kmalloc_array(RX_RING_SIZE, PKT_BUF_SZ,
+						    GFP_DMA | GFP_KERNEL);
 	if (!lp->rx_buffs)
 		goto out_lp;
 	if (lance_need_isa_bounce_buffers) {
-		lp->tx_bounce_buffs = kmalloc(PKT_BUF_SZ*TX_RING_SIZE,
-						  GFP_DMA | GFP_KERNEL);
+		lp->tx_bounce_buffs = kmalloc_array(TX_RING_SIZE, PKT_BUF_SZ,
+						    GFP_DMA | GFP_KERNEL);
 		if (!lp->tx_bounce_buffs)
 			goto out_rx;
 	} else
@@ -780,7 +786,7 @@ lance_open(struct net_device *dev)
 		outw(0x0002, ioaddr+LANCE_ADDR);
 		/* Only touch autoselect bit. */
 		outw(inw(ioaddr+LANCE_BUS_IF) | 0x0002, ioaddr+LANCE_BUS_IF);
- 	}
+	}
 
 	if (lance_debug > 1)
 		printk("%s: lance_open() irq %d dma %d tx/rx rings %#x/%#x init %#x.\n",
@@ -812,7 +818,7 @@ lance_open(struct net_device *dev)
 	 * We used to clear the InitDone bit, 0x0100, here but Mark Stockton
 	 * reports that doing so triggers a bug in the '974.
 	 */
- 	outw(0x0042, ioaddr+LANCE_DATA);
+	outw(0x0042, ioaddr+LANCE_DATA);
 
 	if (lance_debug > 2)
 		printk("%s: LANCE open after %d ticks, init block %#x csr0 %4.4x.\n",
@@ -876,7 +882,7 @@ lance_init_ring(struct net_device *dev, gfp_t gfp)
 			rx_buff = skb->data;
 		else
 			rx_buff = kmalloc(PKT_BUF_SZ, GFP_DMA | gfp);
-		if (rx_buff == NULL)
+		if (!rx_buff)
 			lp->rx_ring[i].base = 0;
 		else
 			lp->rx_ring[i].base = (u32)isa_virt_to_bus(rx_buff) | 0x80000000;
@@ -913,7 +919,7 @@ lance_restart(struct net_device *dev, unsigned int csr0_bits, int must_reinit)
 }
 
 
-static void lance_tx_timeout (struct net_device *dev)
+static void lance_tx_timeout (struct net_device *dev, unsigned int txqueue)
 {
 	struct lance_private *lp = (struct lance_private *) dev->ml_priv;
 	int ioaddr = dev->base_addr;
@@ -997,7 +1003,7 @@ static netdev_tx_t lance_start_xmit(struct sk_buff *skb,
 		skb_copy_from_linear_data(skb, &lp->tx_bounce_buffs[entry], skb->len);
 		lp->tx_ring[entry].base =
 			((u32)isa_virt_to_bus((lp->tx_bounce_buffs + entry)) & 0xffffff) | 0x83000000;
-		dev_kfree_skb(skb);
+		dev_consume_skb_irq(skb);
 	} else {
 		lp->tx_skbuff[entry] = skb;
 		lp->tx_ring[entry].base = ((u32)isa_virt_to_bus(skb->data) & 0xffffff) | 0x83000000;
@@ -1084,7 +1090,7 @@ static irqreturn_t lance_interrupt(int irq, void *dev_id)
 				/* We must free the original skb if it's not a data-only copy
 				   in the bounce buffer. */
 				if (lp->tx_skbuff[entry]) {
-					dev_kfree_skb_irq(lp->tx_skbuff[entry]);
+					dev_consume_skb_irq(lp->tx_skbuff[entry]);
 					lp->tx_skbuff[entry] = NULL;
 				}
 				dirty_tx++;
@@ -1182,7 +1188,7 @@ lance_rx(struct net_device *dev)
 			else
 			{
 				skb = dev_alloc_skb(pkt_len+2);
-				if (skb == NULL)
+				if (!skb)
 				{
 					printk("%s: Memory squeeze, deferring packet.\n", dev->name);
 					for (i=0; i < RX_RING_SIZE; i++)
diff --git a/drivers/net/ethernet/amd/mvme147.c b/drivers/net/ethernet/amd/mvme147.c
index 0a920448522f..f19b04b92fa9 100644
--- a/drivers/net/ethernet/amd/mvme147.c
+++ b/drivers/net/ethernet/amd/mvme147.c
@@ -1,3 +1,4 @@
+// SPDX-License-Identifier: GPL-2.0-only
 /* mvme147.c  : the  Linux/mvme147/lance ethernet driver
  *
  * Copyright (C) 05/1998 Peter Maydell <pmaydell@chiark.greenend.org.uk>
@@ -15,6 +16,7 @@
 #include <linux/init.h>
 #include <linux/errno.h>
 #include <linux/gfp.h>
+#include <linux/pgtable.h>
 /* Used for the temporal inet entries and routing */
 #include <linux/socket.h>
 #include <linux/route.h>
@@ -23,7 +25,6 @@
 #include <linux/skbuff.h>
 
 #include <asm/io.h>
-#include <asm/pgtable.h>
 #include <asm/mvme147hw.h>
 
 /* We have 32K of RAM for the init block and buffers. This places
@@ -67,12 +68,13 @@ static const struct net_device_ops lance_netdev_ops = {
 };
 
 /* Initialise the one and only on-board 7990 */
-struct net_device * __init mvme147lance_probe(int unit)
+static struct net_device * __init mvme147lance_probe(void)
 {
 	struct net_device *dev;
 	static int called;
 	static const char name[] = "MVME147 LANCE";
 	struct m147lance_private *lp;
+	u8 macaddr[ETH_ALEN];
 	u_long *addr;
 	u_long address;
 	int err;
@@ -85,9 +87,6 @@ struct net_device * __init mvme147lance_probe(int unit)
 	if (!dev)
 		return ERR_PTR(-ENOMEM);
 
-	if (unit >= 0)
-		sprintf(dev->name, "eth%d", unit);
-
 	/* Fill the dev fields */
 	dev->base_addr = (unsigned long)MVME147_LANCE_BASE;
 	dev->netdev_ops = &lance_netdev_ops;
@@ -95,19 +94,16 @@ struct net_device * __init mvme147lance_probe(int unit)
 
 	addr = (u_long *)ETHERNET_ADDRESS;
 	address = *addr;
-	dev->dev_addr[0] = 0x08;
-	dev->dev_addr[1] = 0x00;
-	dev->dev_addr[2] = 0x3e;
+	macaddr[0] = 0x08;
+	macaddr[1] = 0x00;
+	macaddr[2] = 0x3e;
 	address = address >> 8;
-	dev->dev_addr[5] = address&0xff;
+	macaddr[5] = address&0xff;
 	address = address >> 8;
-	dev->dev_addr[4] = address&0xff;
+	macaddr[4] = address&0xff;
 	address = address >> 8;
-	dev->dev_addr[3] = address&0xff;
-
-	printk("%s: MVME147 at 0x%08lx, irq %d, Hardware Address %pM\n",
-	       dev->name, dev->base_addr, MVME147_LANCE_IRQ,
-	       dev->dev_addr);
+	macaddr[3] = address&0xff;
+	eth_hw_addr_set(dev, macaddr);
 
 	lp = netdev_priv(dev);
 	lp->ram = __get_dma_pages(GFP_ATOMIC, 3);	/* 32K */
@@ -138,6 +134,9 @@ struct net_device * __init mvme147lance_probe(int unit)
 		return ERR_PTR(err);
 	}
 
+	netdev_info(dev, "MVME147 at 0x%08lx, irq %d, Hardware Address %pM\n",
+		    dev->base_addr, MVME147_LANCE_IRQ, dev->dev_addr);
+
 	return dev;
 }
 
@@ -178,22 +177,22 @@ static int m147lance_close(struct net_device *dev)
 	return 0;
 }
 
-#ifdef MODULE
+MODULE_DESCRIPTION("MVME147 LANCE Ethernet driver");
 MODULE_LICENSE("GPL");
 
 static struct net_device *dev_mvme147_lance;
-int __init init_module(void)
+static int __init m147lance_init(void)
 {
-	dev_mvme147_lance = mvme147lance_probe(-1);
+	dev_mvme147_lance = mvme147lance_probe();
 	return PTR_ERR_OR_ZERO(dev_mvme147_lance);
 }
+module_init(m147lance_init);
 
-void __exit cleanup_module(void)
+static void __exit m147lance_exit(void)
 {
 	struct m147lance_private *lp = netdev_priv(dev_mvme147_lance);
 	unregister_netdev(dev_mvme147_lance);
 	free_pages(lp->ram, 3);
 	free_netdev(dev_mvme147_lance);
 }
-
-#endif /* MODULE */
+module_exit(m147lance_exit);
diff --git a/drivers/net/ethernet/amd/ni65.c b/drivers/net/ethernet/amd/ni65.c
deleted file mode 100644
index e248d1ab3e47..000000000000
--- a/drivers/net/ethernet/amd/ni65.c
+++ /dev/null
@@ -1,1251 +0,0 @@
-/*
- * ni6510 (am7990 'lance' chip) driver for Linux-net-3
- * BETAcode v0.71 (96/09/29) for 2.0.0 (or later)
- * copyrights (c) 1994,1995,1996 by M.Hipp
- *
- * This driver can handle the old ni6510 board and the newer ni6510
- * EtherBlaster. (probably it also works with every full NE2100
- * compatible card)
- *
- * driver probes: io: 0x360,0x300,0x320,0x340 / dma: 3,5,6,7
- *
- * This is an extension to the Linux operating system, and is covered by the
- * same GNU General Public License that covers the Linux-kernel.
- *
- * comments/bugs/suggestions can be sent to:
- *   Michael Hipp
- *   email: hippm@informatik.uni-tuebingen.de
- *
- * sources:
- *   some things are from the 'ni6510-packet-driver for dos by Russ Nelson'
- *   and from the original drivers by D.Becker
- *
- * known problems:
- *   - on some PCI boards (including my own) the card/board/ISA-bridge has
- *     problems with bus master DMA. This results in lotsa overruns.
- *     It may help to '#define RCV_PARANOIA_CHECK' or try to #undef
- *     the XMT and RCV_VIA_SKB option .. this reduces driver performance.
- *     Or just play with your BIOS options to optimize ISA-DMA access.
- *     Maybe you also wanna play with the LOW_PERFORAMCE and MID_PERFORMANCE
- *     defines -> please report me your experience then
- *   - Harald reported for ASUS SP3G mainboards, that you should use
- *     the 'optimal settings' from the user's manual on page 3-12!
- *
- * credits:
- *   thanx to Jason Sullivan for sending me a ni6510 card!
- *   lot of debug runs with ASUS SP3G Boards (Intel Saturn) by Harald Koenig
- *
- * simple performance test: (486DX-33/Ni6510-EB receives from 486DX4-100/Ni6510-EB)
- *    average: FTP -> 8384421 bytes received in 8.5 seconds
- *           (no RCV_VIA_SKB,no XMT_VIA_SKB,PARANOIA_CHECK,4 XMIT BUFS, 8 RCV_BUFFS)
- *    peak: FTP -> 8384421 bytes received in 7.5 seconds
- *           (RCV_VIA_SKB,XMT_VIA_SKB,no PARANOIA_CHECK,1(!) XMIT BUF, 16 RCV BUFFS)
- */
-
-/*
- * 99.Jun.8: added support for /proc/net/dev byte count for xosview (HK)
- * 96.Sept.29: virt_to_bus stuff added for new memory modell
- * 96.April.29: Added Harald Koenig's Patches (MH)
- * 96.April.13: enhanced error handling .. more tests (MH)
- * 96.April.5/6: a lot of performance tests. Got it stable now (hopefully) (MH)
- * 96.April.1: (no joke ;) .. added EtherBlaster and Module support (MH)
- * 96.Feb.19: fixed a few bugs .. cleanups .. tested for 1.3.66 (MH)
- *            hopefully no more 16MB limit
- *
- * 95.Nov.18: multicast tweaked (AC).
- *
- * 94.Aug.22: changes in xmit_intr (ack more than one xmitted-packet), ni65_send_packet (p->lock) (MH)
- *
- * 94.July.16: fixed bugs in recv_skb and skb-alloc stuff  (MH)
- */
-
-#include <linux/kernel.h>
-#include <linux/string.h>
-#include <linux/errno.h>
-#include <linux/ioport.h>
-#include <linux/slab.h>
-#include <linux/interrupt.h>
-#include <linux/delay.h>
-#include <linux/init.h>
-#include <linux/netdevice.h>
-#include <linux/etherdevice.h>
-#include <linux/skbuff.h>
-#include <linux/module.h>
-#include <linux/bitops.h>
-
-#include <asm/io.h>
-#include <asm/dma.h>
-
-#include "ni65.h"
-
-/*
- * the current setting allows an acceptable performance
- * for 'RCV_PARANOIA_CHECK' read the 'known problems' part in
- * the header of this file
- * 'invert' the defines for max. performance. This may cause DMA problems
- * on some boards (e.g on my ASUS SP3G)
- */
-#undef XMT_VIA_SKB
-#undef RCV_VIA_SKB
-#define RCV_PARANOIA_CHECK
-
-#define MID_PERFORMANCE
-
-#if   defined( LOW_PERFORMANCE )
- static int isa0=7,isa1=7,csr80=0x0c10;
-#elif defined( MID_PERFORMANCE )
- static int isa0=5,isa1=5,csr80=0x2810;
-#else	/* high performance */
- static int isa0=4,isa1=4,csr80=0x0017;
-#endif
-
-/*
- * a few card/vendor specific defines
- */
-#define NI65_ID0    0x00
-#define NI65_ID1    0x55
-#define NI65_EB_ID0 0x52
-#define NI65_EB_ID1 0x44
-#define NE2100_ID0  0x57
-#define NE2100_ID1  0x57
-
-#define PORT p->cmdr_addr
-
-/*
- * buffer configuration
- */
-#if 1
-#define RMDNUM 16
-#define RMDNUMMASK 0x80000000
-#else
-#define RMDNUM 8
-#define RMDNUMMASK 0x60000000 /* log2(RMDNUM)<<29 */
-#endif
-
-#if 0
-#define TMDNUM 1
-#define TMDNUMMASK 0x00000000
-#else
-#define TMDNUM 4
-#define TMDNUMMASK 0x40000000 /* log2(TMDNUM)<<29 */
-#endif
-
-/* slightly oversized */
-#define R_BUF_SIZE 1544
-#define T_BUF_SIZE 1544
-
-/*
- * lance register defines
- */
-#define L_DATAREG 0x00
-#define L_ADDRREG 0x02
-#define L_RESET   0x04
-#define L_CONFIG  0x05
-#define L_BUSIF   0x06
-
-/*
- * to access the lance/am7990-regs, you have to write
- * reg-number into L_ADDRREG, then you can access it using L_DATAREG
- */
-#define CSR0  0x00
-#define CSR1  0x01
-#define CSR2  0x02
-#define CSR3  0x03
-
-#define INIT_RING_BEFORE_START	0x1
-#define FULL_RESET_ON_ERROR	0x2
-
-#if 0
-#define writereg(val,reg) {outw(reg,PORT+L_ADDRREG);inw(PORT+L_ADDRREG); \
-                           outw(val,PORT+L_DATAREG);inw(PORT+L_DATAREG);}
-#define readreg(reg) (outw(reg,PORT+L_ADDRREG),inw(PORT+L_ADDRREG),\
-                       inw(PORT+L_DATAREG))
-#if 0
-#define writedatareg(val) {outw(val,PORT+L_DATAREG);inw(PORT+L_DATAREG);}
-#else
-#define writedatareg(val) {  writereg(val,CSR0); }
-#endif
-#else
-#define writereg(val,reg) {outw(reg,PORT+L_ADDRREG);outw(val,PORT+L_DATAREG);}
-#define readreg(reg) (outw(reg,PORT+L_ADDRREG),inw(PORT+L_DATAREG))
-#define writedatareg(val) { writereg(val,CSR0); }
-#endif
-
-static unsigned char ni_vendor[] = { 0x02,0x07,0x01 };
-
-static struct card {
-	unsigned char id0,id1;
-	short id_offset;
-	short total_size;
-	short cmd_offset;
-	short addr_offset;
-	unsigned char *vendor_id;
-	char *cardname;
-	unsigned long config;
-} cards[] = {
-	{
-		.id0	     = NI65_ID0,
-		.id1	     = NI65_ID1,
-		.id_offset   = 0x0e,
-		.total_size  = 0x10,
-		.cmd_offset  = 0x0,
-		.addr_offset = 0x8,
-		.vendor_id   = ni_vendor,
-		.cardname    = "ni6510",
-		.config	     = 0x1,
-       	},
-	{
-		.id0	     = NI65_EB_ID0,
-		.id1	     = NI65_EB_ID1,
-		.id_offset   = 0x0e,
-		.total_size  = 0x18,
-		.cmd_offset  = 0x10,
-		.addr_offset = 0x0,
-		.vendor_id   = ni_vendor,
-		.cardname    = "ni6510 EtherBlaster",
-		.config	     = 0x2,
-       	},
-	{
-		.id0	     = NE2100_ID0,
-		.id1	     = NE2100_ID1,
-		.id_offset   = 0x0e,
-		.total_size  = 0x18,
-		.cmd_offset  = 0x10,
-		.addr_offset = 0x0,
-		.vendor_id   = NULL,
-		.cardname    = "generic NE2100",
-		.config	     = 0x0,
-	},
-};
-#define NUM_CARDS 3
-
-struct priv
-{
-	struct rmd rmdhead[RMDNUM];
-	struct tmd tmdhead[TMDNUM];
-	struct init_block ib;
-	int rmdnum;
-	int tmdnum,tmdlast;
-#ifdef RCV_VIA_SKB
-	struct sk_buff *recv_skb[RMDNUM];
-#else
-	void *recvbounce[RMDNUM];
-#endif
-#ifdef XMT_VIA_SKB
-	struct sk_buff *tmd_skb[TMDNUM];
-#endif
-	void *tmdbounce[TMDNUM];
-	int tmdbouncenum;
-	int lock,xmit_queued;
-
-	void *self;
-	int cmdr_addr;
-	int cardno;
-	int features;
-	spinlock_t ring_lock;
-};
-
-static int  ni65_probe1(struct net_device *dev,int);
-static irqreturn_t ni65_interrupt(int irq, void * dev_id);
-static void ni65_recv_intr(struct net_device *dev,int);
-static void ni65_xmit_intr(struct net_device *dev,int);
-static int  ni65_open(struct net_device *dev);
-static int  ni65_lance_reinit(struct net_device *dev);
-static void ni65_init_lance(struct priv *p,unsigned char*,int,int);
-static netdev_tx_t ni65_send_packet(struct sk_buff *skb,
-				    struct net_device *dev);
-static void  ni65_timeout(struct net_device *dev);
-static int  ni65_close(struct net_device *dev);
-static int  ni65_alloc_buffer(struct net_device *dev);
-static void ni65_free_buffer(struct priv *p);
-static void set_multicast_list(struct net_device *dev);
-
-static int irqtab[] __initdata = { 9,12,15,5 }; /* irq config-translate */
-static int dmatab[] __initdata = { 0,3,5,6,7 }; /* dma config-translate and autodetect */
-
-static int debuglevel = 1;
-
-/*
- * set 'performance' registers .. we must STOP lance for that
- */
-static void ni65_set_performance(struct priv *p)
-{
-	writereg(CSR0_STOP | CSR0_CLRALL,CSR0); /* STOP */
-
-	if( !(cards[p->cardno].config & 0x02) )
-		return;
-
-	outw(80,PORT+L_ADDRREG);
-	if(inw(PORT+L_ADDRREG) != 80)
-		return;
-
-	writereg( (csr80 & 0x3fff) ,80); /* FIFO watermarks */
-	outw(0,PORT+L_ADDRREG);
-	outw((short)isa0,PORT+L_BUSIF); /* write ISA 0: DMA_R : isa0 * 50ns */
-	outw(1,PORT+L_ADDRREG);
-	outw((short)isa1,PORT+L_BUSIF); /* write ISA 1: DMA_W : isa1 * 50ns	*/
-
-	outw(CSR0,PORT+L_ADDRREG);	/* switch back to CSR0 */
-}
-
-/*
- * open interface (up)
- */
-static int ni65_open(struct net_device *dev)
-{
-	struct priv *p = dev->ml_priv;
-	int irqval = request_irq(dev->irq, ni65_interrupt,0,
-                        cards[p->cardno].cardname,dev);
-	if (irqval) {
-		printk(KERN_ERR "%s: unable to get IRQ %d (irqval=%d).\n",
-		          dev->name,dev->irq, irqval);
-		return -EAGAIN;
-	}
-
-	if(ni65_lance_reinit(dev))
-	{
-		netif_start_queue(dev);
-		return 0;
-	}
-	else
-	{
-		free_irq(dev->irq,dev);
-		return -EAGAIN;
-	}
-}
-
-/*
- * close interface (down)
- */
-static int ni65_close(struct net_device *dev)
-{
-	struct priv *p = dev->ml_priv;
-
-	netif_stop_queue(dev);
-
-	outw(inw(PORT+L_RESET),PORT+L_RESET); /* that's the hard way */
-
-#ifdef XMT_VIA_SKB
-	{
-		int i;
-		for(i=0;i<TMDNUM;i++)
-		{
-			if(p->tmd_skb[i]) {
-				dev_kfree_skb(p->tmd_skb[i]);
-				p->tmd_skb[i] = NULL;
-			}
-		}
-	}
-#endif
-	free_irq(dev->irq,dev);
-	return 0;
-}
-
-static void cleanup_card(struct net_device *dev)
-{
-	struct priv *p = dev->ml_priv;
-	disable_dma(dev->dma);
-	free_dma(dev->dma);
-	release_region(dev->base_addr, cards[p->cardno].total_size);
-	ni65_free_buffer(p);
-}
-
-/* set: io,irq,dma or set it when calling insmod */
-static int irq;
-static int io;
-static int dma;
-
-/*
- * Probe The Card (not the lance-chip)
- */
-struct net_device * __init ni65_probe(int unit)
-{
-	struct net_device *dev = alloc_etherdev(0);
-	static const int ports[] = { 0x360, 0x300, 0x320, 0x340, 0 };
-	const int *port;
-	int err = 0;
-
-	if (!dev)
-		return ERR_PTR(-ENOMEM);
-
-	if (unit >= 0) {
-		sprintf(dev->name, "eth%d", unit);
-		netdev_boot_setup_check(dev);
-		irq = dev->irq;
-		dma = dev->dma;
-	} else {
-		dev->base_addr = io;
-	}
-
-	if (dev->base_addr > 0x1ff) { /* Check a single specified location. */
-		err = ni65_probe1(dev, dev->base_addr);
-	} else if (dev->base_addr > 0) { /* Don't probe at all. */
-		err = -ENXIO;
-	} else {
-		for (port = ports; *port && ni65_probe1(dev, *port); port++)
-			;
-		if (!*port)
-			err = -ENODEV;
-	}
-	if (err)
-		goto out;
-
-	err = register_netdev(dev);
-	if (err)
-		goto out1;
-	return dev;
-out1:
-	cleanup_card(dev);
-out:
-	free_netdev(dev);
-	return ERR_PTR(err);
-}
-
-static const struct net_device_ops ni65_netdev_ops = {
-	.ndo_open		= ni65_open,
-	.ndo_stop		= ni65_close,
-	.ndo_start_xmit		= ni65_send_packet,
-	.ndo_tx_timeout		= ni65_timeout,
-	.ndo_set_rx_mode	= set_multicast_list,
-	.ndo_set_mac_address 	= eth_mac_addr,
-	.ndo_validate_addr	= eth_validate_addr,
-};
-
-/*
- * this is the real card probe ..
- */
-static int __init ni65_probe1(struct net_device *dev,int ioaddr)
-{
-	int i,j;
-	struct priv *p;
-	unsigned long flags;
-
-	dev->irq = irq;
-	dev->dma = dma;
-
-	for(i=0;i<NUM_CARDS;i++) {
-		if(!request_region(ioaddr, cards[i].total_size, cards[i].cardname))
-			continue;
-		if(cards[i].id_offset >= 0) {
-			if(inb(ioaddr+cards[i].id_offset+0) != cards[i].id0 ||
-				 inb(ioaddr+cards[i].id_offset+1) != cards[i].id1) {
-				 release_region(ioaddr, cards[i].total_size);
-				 continue;
-			}
-		}
-		if(cards[i].vendor_id) {
-			for(j=0;j<3;j++)
-				if(inb(ioaddr+cards[i].addr_offset+j) != cards[i].vendor_id[j]) {
-					release_region(ioaddr, cards[i].total_size);
-					continue;
-			  }
-		}
-		break;
-	}
-	if(i == NUM_CARDS)
-		return -ENODEV;
-
-	for(j=0;j<6;j++)
-		dev->dev_addr[j] = inb(ioaddr+cards[i].addr_offset+j);
-
-	if( (j=ni65_alloc_buffer(dev)) < 0) {
-		release_region(ioaddr, cards[i].total_size);
-		return j;
-	}
-	p = dev->ml_priv;
-	p->cmdr_addr = ioaddr + cards[i].cmd_offset;
-	p->cardno = i;
-	spin_lock_init(&p->ring_lock);
-
-	printk(KERN_INFO "%s: %s found at %#3x, ", dev->name, cards[p->cardno].cardname , ioaddr);
-
-	outw(inw(PORT+L_RESET),PORT+L_RESET); /* first: reset the card */
-	if( (j=readreg(CSR0)) != 0x4) {
-		 printk("failed.\n");
-		 printk(KERN_ERR "%s: Can't RESET card: %04x\n", dev->name, j);
-		 ni65_free_buffer(p);
-		 release_region(ioaddr, cards[p->cardno].total_size);
-		 return -EAGAIN;
-	}
-
-	outw(88,PORT+L_ADDRREG);
-	if(inw(PORT+L_ADDRREG) == 88) {
-		unsigned long v;
-		v = inw(PORT+L_DATAREG);
-		v <<= 16;
-		outw(89,PORT+L_ADDRREG);
-		v |= inw(PORT+L_DATAREG);
-		printk("Version %#08lx, ",v);
-		p->features = INIT_RING_BEFORE_START;
-	}
-	else {
-		printk("ancient LANCE, ");
-		p->features = 0x0;
-	}
-
-	if(test_bit(0,&cards[i].config)) {
-		dev->irq = irqtab[(inw(ioaddr+L_CONFIG)>>2)&3];
-		dev->dma = dmatab[inw(ioaddr+L_CONFIG)&3];
-		printk("IRQ %d (from card), DMA %d (from card).\n",dev->irq,dev->dma);
-	}
-	else {
-		if(dev->dma == 0) {
-		/* 'stuck test' from lance.c */
-			unsigned long dma_channels =
-				((inb(DMA1_STAT_REG) >> 4) & 0x0f)
-				| (inb(DMA2_STAT_REG) & 0xf0);
-			for(i=1;i<5;i++) {
-				int dma = dmatab[i];
-				if(test_bit(dma,&dma_channels) || request_dma(dma,"ni6510"))
-					continue;
-
-				flags=claim_dma_lock();
-				disable_dma(dma);
-				set_dma_mode(dma,DMA_MODE_CASCADE);
-				enable_dma(dma);
-				release_dma_lock(flags);
-
-				ni65_init_lance(p,dev->dev_addr,0,0); /* trigger memory access */
-
-				flags=claim_dma_lock();
-				disable_dma(dma);
-				free_dma(dma);
-				release_dma_lock(flags);
-
-				if(readreg(CSR0) & CSR0_IDON)
-					break;
-			}
-			if(i == 5) {
-				printk("failed.\n");
-				printk(KERN_ERR "%s: Can't detect DMA channel!\n", dev->name);
-				ni65_free_buffer(p);
-				release_region(ioaddr, cards[p->cardno].total_size);
-				return -EAGAIN;
-			}
-			dev->dma = dmatab[i];
-			printk("DMA %d (autodetected), ",dev->dma);
-		}
-		else
-			printk("DMA %d (assigned), ",dev->dma);
-
-		if(dev->irq < 2)
-		{
-			unsigned long irq_mask;
-
-			ni65_init_lance(p,dev->dev_addr,0,0);
-			irq_mask = probe_irq_on();
-			writereg(CSR0_INIT|CSR0_INEA,CSR0); /* trigger interrupt */
-			msleep(20);
-			dev->irq = probe_irq_off(irq_mask);
-			if(!dev->irq)
-			{
-				printk("Failed to detect IRQ line!\n");
-				ni65_free_buffer(p);
-				release_region(ioaddr, cards[p->cardno].total_size);
-				return -EAGAIN;
-			}
-			printk("IRQ %d (autodetected).\n",dev->irq);
-		}
-		else
-			printk("IRQ %d (assigned).\n",dev->irq);
-	}
-
-	if(request_dma(dev->dma, cards[p->cardno].cardname ) != 0)
-	{
-		printk(KERN_ERR "%s: Can't request dma-channel %d\n",dev->name,(int) dev->dma);
-		ni65_free_buffer(p);
-		release_region(ioaddr, cards[p->cardno].total_size);
-		return -EAGAIN;
-	}
-
-	dev->base_addr = ioaddr;
-	dev->netdev_ops = &ni65_netdev_ops;
-	dev->watchdog_timeo	= HZ/2;
-
-	return 0; /* everything is OK */
-}
-
-/*
- * set lance register and trigger init
- */
-static void ni65_init_lance(struct priv *p,unsigned char *daddr,int filter,int mode)
-{
-	int i;
-	u32 pib;
-
-	writereg(CSR0_CLRALL|CSR0_STOP,CSR0);
-
-	for(i=0;i<6;i++)
-		p->ib.eaddr[i] = daddr[i];
-
-	for(i=0;i<8;i++)
-		p->ib.filter[i] = filter;
-	p->ib.mode = mode;
-
-	p->ib.trp = (u32) isa_virt_to_bus(p->tmdhead) | TMDNUMMASK;
-	p->ib.rrp = (u32) isa_virt_to_bus(p->rmdhead) | RMDNUMMASK;
-	writereg(0,CSR3);	/* busmaster/no word-swap */
-	pib = (u32) isa_virt_to_bus(&p->ib);
-	writereg(pib & 0xffff,CSR1);
-	writereg(pib >> 16,CSR2);
-
-	writereg(CSR0_INIT,CSR0); /* this changes L_ADDRREG to CSR0 */
-
-	for(i=0;i<32;i++)
-	{
-		mdelay(4);
-		if(inw(PORT+L_DATAREG) & (CSR0_IDON | CSR0_MERR) )
-			break; /* init ok ? */
-	}
-}
-
-/*
- * allocate memory area and check the 16MB border
- */
-static void *ni65_alloc_mem(struct net_device *dev,char *what,int size,int type)
-{
-	struct sk_buff *skb=NULL;
-	unsigned char *ptr;
-	void *ret;
-
-	if(type) {
-		ret = skb = alloc_skb(2+16+size,GFP_KERNEL|GFP_DMA);
-		if(!skb) {
-			printk(KERN_WARNING "%s: unable to allocate %s memory.\n",dev->name,what);
-			return NULL;
-		}
-		skb_reserve(skb,2+16);
-		skb_put(skb,R_BUF_SIZE);	 /* grab the whole space .. (not necessary) */
-		ptr = skb->data;
-	}
-	else {
-		ret = ptr = kmalloc(T_BUF_SIZE,GFP_KERNEL | GFP_DMA);
-		if(!ret)
-			return NULL;
-	}
-	if( (u32) virt_to_phys(ptr+size) > 0x1000000) {
-		printk(KERN_WARNING "%s: unable to allocate %s memory in lower 16MB!\n",dev->name,what);
-		if(type)
-			kfree_skb(skb);
-		else
-			kfree(ptr);
-		return NULL;
-	}
-	return ret;
-}
-
-/*
- * allocate all memory structures .. send/recv buffers etc ...
- */
-static int ni65_alloc_buffer(struct net_device *dev)
-{
-	unsigned char *ptr;
-	struct priv *p;
-	int i;
-
-	/*
-	 * we need 8-aligned memory ..
-	 */
-	ptr = ni65_alloc_mem(dev,"BUFFER",sizeof(struct priv)+8,0);
-	if(!ptr)
-		return -ENOMEM;
-
-	p = dev->ml_priv = (struct priv *) (((unsigned long) ptr + 7) & ~0x7);
-	memset((char *)p, 0, sizeof(struct priv));
-	p->self = ptr;
-
-	for(i=0;i<TMDNUM;i++)
-	{
-#ifdef XMT_VIA_SKB
-		p->tmd_skb[i] = NULL;
-#endif
-		p->tmdbounce[i] = ni65_alloc_mem(dev,"XMIT",T_BUF_SIZE,0);
-		if(!p->tmdbounce[i]) {
-			ni65_free_buffer(p);
-			return -ENOMEM;
-		}
-	}
-
-	for(i=0;i<RMDNUM;i++)
-	{
-#ifdef RCV_VIA_SKB
-		p->recv_skb[i] = ni65_alloc_mem(dev,"RECV",R_BUF_SIZE,1);
-		if(!p->recv_skb[i]) {
-			ni65_free_buffer(p);
-			return -ENOMEM;
-		}
-#else
-		p->recvbounce[i] = ni65_alloc_mem(dev,"RECV",R_BUF_SIZE,0);
-		if(!p->recvbounce[i]) {
-			ni65_free_buffer(p);
-			return -ENOMEM;
-		}
-#endif
-	}
-
-	return 0; /* everything is OK */
-}
-
-/*
- * free buffers and private struct
- */
-static void ni65_free_buffer(struct priv *p)
-{
-	int i;
-
-	if(!p)
-		return;
-
-	for(i=0;i<TMDNUM;i++) {
-		kfree(p->tmdbounce[i]);
-#ifdef XMT_VIA_SKB
-		if(p->tmd_skb[i])
-			dev_kfree_skb(p->tmd_skb[i]);
-#endif
-	}
-
-	for(i=0;i<RMDNUM;i++)
-	{
-#ifdef RCV_VIA_SKB
-		if(p->recv_skb[i])
-			dev_kfree_skb(p->recv_skb[i]);
-#else
-		kfree(p->recvbounce[i]);
-#endif
-	}
-	kfree(p->self);
-}
-
-
-/*
- * stop and (re)start lance .. e.g after an error
- */
-static void ni65_stop_start(struct net_device *dev,struct priv *p)
-{
-	int csr0 = CSR0_INEA;
-
-	writedatareg(CSR0_STOP);
-
-	if(debuglevel > 1)
-		printk(KERN_DEBUG "ni65_stop_start\n");
-
-	if(p->features & INIT_RING_BEFORE_START) {
-		int i;
-#ifdef XMT_VIA_SKB
-		struct sk_buff *skb_save[TMDNUM];
-#endif
-		unsigned long buffer[TMDNUM];
-		short blen[TMDNUM];
-
-		if(p->xmit_queued) {
-			while(1) {
-				if((p->tmdhead[p->tmdlast].u.s.status & XMIT_OWN))
-					break;
-				p->tmdlast = (p->tmdlast + 1) & (TMDNUM-1);
-				if(p->tmdlast == p->tmdnum)
-					break;
-			}
-		}
-
-		for(i=0;i<TMDNUM;i++) {
-			struct tmd *tmdp = p->tmdhead + i;
-#ifdef XMT_VIA_SKB
-			skb_save[i] = p->tmd_skb[i];
-#endif
-			buffer[i] = (u32) isa_bus_to_virt(tmdp->u.buffer);
-			blen[i] = tmdp->blen;
-			tmdp->u.s.status = 0x0;
-		}
-
-		for(i=0;i<RMDNUM;i++) {
-			struct rmd *rmdp = p->rmdhead + i;
-			rmdp->u.s.status = RCV_OWN;
-		}
-		p->tmdnum = p->xmit_queued = 0;
-		writedatareg(CSR0_STRT | csr0);
-
-		for(i=0;i<TMDNUM;i++) {
-			int num = (i + p->tmdlast) & (TMDNUM-1);
-			p->tmdhead[i].u.buffer = (u32) isa_virt_to_bus((char *)buffer[num]); /* status is part of buffer field */
-			p->tmdhead[i].blen = blen[num];
-			if(p->tmdhead[i].u.s.status & XMIT_OWN) {
-				 p->tmdnum = (p->tmdnum + 1) & (TMDNUM-1);
-				 p->xmit_queued = 1;
-	 writedatareg(CSR0_TDMD | CSR0_INEA | csr0);
-			}
-#ifdef XMT_VIA_SKB
-			p->tmd_skb[i] = skb_save[num];
-#endif
-		}
-		p->rmdnum = p->tmdlast = 0;
-		if(!p->lock)
-			if (p->tmdnum || !p->xmit_queued)
-				netif_wake_queue(dev);
-		netif_trans_update(dev); /* prevent tx timeout */
-	}
-	else
-		writedatareg(CSR0_STRT | csr0);
-}
-
-/*
- * init lance (write init-values .. init-buffers) (open-helper)
- */
-static int ni65_lance_reinit(struct net_device *dev)
-{
-	 int i;
-	 struct priv *p = dev->ml_priv;
-	 unsigned long flags;
-
-	 p->lock = 0;
-	 p->xmit_queued = 0;
-
-	 flags=claim_dma_lock();
-	 disable_dma(dev->dma); /* I've never worked with dma, but we do it like the packetdriver */
-	 set_dma_mode(dev->dma,DMA_MODE_CASCADE);
-	 enable_dma(dev->dma);
-	 release_dma_lock(flags);
-
-	 outw(inw(PORT+L_RESET),PORT+L_RESET); /* first: reset the card */
-	 if( (i=readreg(CSR0) ) != 0x4)
-	 {
-		 printk(KERN_ERR "%s: can't RESET %s card: %04x\n",dev->name,
-							cards[p->cardno].cardname,(int) i);
-		 flags=claim_dma_lock();
-		 disable_dma(dev->dma);
-		 release_dma_lock(flags);
-		 return 0;
-	 }
-
-	 p->rmdnum = p->tmdnum = p->tmdlast = p->tmdbouncenum = 0;
-	 for(i=0;i<TMDNUM;i++)
-	 {
-		 struct tmd *tmdp = p->tmdhead + i;
-#ifdef XMT_VIA_SKB
-		 if(p->tmd_skb[i]) {
-			 dev_kfree_skb(p->tmd_skb[i]);
-			 p->tmd_skb[i] = NULL;
-		 }
-#endif
-		 tmdp->u.buffer = 0x0;
-		 tmdp->u.s.status = XMIT_START | XMIT_END;
-		 tmdp->blen = tmdp->status2 = 0;
-	 }
-
-	 for(i=0;i<RMDNUM;i++)
-	 {
-		 struct rmd *rmdp = p->rmdhead + i;
-#ifdef RCV_VIA_SKB
-		 rmdp->u.buffer = (u32) isa_virt_to_bus(p->recv_skb[i]->data);
-#else
-		 rmdp->u.buffer = (u32) isa_virt_to_bus(p->recvbounce[i]);
-#endif
-		 rmdp->blen = -(R_BUF_SIZE-8);
-		 rmdp->mlen = 0;
-		 rmdp->u.s.status = RCV_OWN;
-	 }
-
-	 if(dev->flags & IFF_PROMISC)
-		 ni65_init_lance(p,dev->dev_addr,0x00,M_PROM);
-	 else if (netdev_mc_count(dev) || dev->flags & IFF_ALLMULTI)
-		 ni65_init_lance(p,dev->dev_addr,0xff,0x0);
-	 else
-		 ni65_init_lance(p,dev->dev_addr,0x00,0x00);
-
-	/*
-	 * ni65_set_lance_mem() sets L_ADDRREG to CSR0
-	 * NOW, WE WILL NEVER CHANGE THE L_ADDRREG, CSR0 IS ALWAYS SELECTED
-	 */
-
-	 if(inw(PORT+L_DATAREG) & CSR0_IDON)	{
-		 ni65_set_performance(p);
-					 /* init OK: start lance , enable interrupts */
-		 writedatareg(CSR0_CLRALL | CSR0_INEA | CSR0_STRT);
-		 return 1; /* ->OK */
-	 }
-	 printk(KERN_ERR "%s: can't init lance, status: %04x\n",dev->name,(int) inw(PORT+L_DATAREG));
-	 flags=claim_dma_lock();
-	 disable_dma(dev->dma);
-	 release_dma_lock(flags);
-	 return 0; /* ->Error */
-}
-
-/*
- * interrupt handler
- */
-static irqreturn_t ni65_interrupt(int irq, void * dev_id)
-{
-	int csr0 = 0;
-	struct net_device *dev = dev_id;
-	struct priv *p;
-	int bcnt = 32;
-
-	p = dev->ml_priv;
-
-	spin_lock(&p->ring_lock);
-
-	while(--bcnt) {
-		csr0 = inw(PORT+L_DATAREG);
-
-#if 0
-		writedatareg( (csr0 & CSR0_CLRALL) ); /* ack interrupts, disable int. */
-#else
-		writedatareg( (csr0 & CSR0_CLRALL) | CSR0_INEA ); /* ack interrupts, interrupts enabled */
-#endif
-
-		if(!(csr0 & (CSR0_ERR | CSR0_RINT | CSR0_TINT)))
-			break;
-
-		if(csr0 & CSR0_RINT) /* RECV-int? */
-			ni65_recv_intr(dev,csr0);
-		if(csr0 & CSR0_TINT) /* XMIT-int? */
-			ni65_xmit_intr(dev,csr0);
-
-		if(csr0 & CSR0_ERR)
-		{
-			if(debuglevel > 1)
-				printk(KERN_ERR "%s: general error: %04x.\n",dev->name,csr0);
-			if(csr0 & CSR0_BABL)
-				dev->stats.tx_errors++;
-			if(csr0 & CSR0_MISS) {
-				int i;
-				for(i=0;i<RMDNUM;i++)
-					printk("%02x ",p->rmdhead[i].u.s.status);
-				printk("\n");
-				dev->stats.rx_errors++;
-			}
-			if(csr0 & CSR0_MERR) {
-				if(debuglevel > 1)
-					printk(KERN_ERR "%s: Ooops .. memory error: %04x.\n",dev->name,csr0);
-				ni65_stop_start(dev,p);
-			}
-		}
-	}
-
-#ifdef RCV_PARANOIA_CHECK
-{
- int j;
- for(j=0;j<RMDNUM;j++)
- {
-	int i, num2;
-	for(i=RMDNUM-1;i>0;i--) {
-		 num2 = (p->rmdnum + i) & (RMDNUM-1);
-		 if(!(p->rmdhead[num2].u.s.status & RCV_OWN))
-				break;
-	}
-
-	if(i) {
-		int k, num1;
-		for(k=0;k<RMDNUM;k++) {
-			num1 = (p->rmdnum + k) & (RMDNUM-1);
-			if(!(p->rmdhead[num1].u.s.status & RCV_OWN))
-				break;
-		}
-		if(!k)
-			break;
-
-		if(debuglevel > 0)
-		{
-			char buf[256],*buf1;
-			buf1 = buf;
-			for(k=0;k<RMDNUM;k++) {
-				sprintf(buf1,"%02x ",(p->rmdhead[k].u.s.status)); /* & RCV_OWN) ); */
-				buf1 += 3;
-			}
-			*buf1 = 0;
-			printk(KERN_ERR "%s: Ooops, receive ring corrupted %2d %2d | %s\n",dev->name,p->rmdnum,i,buf);
-		}
-
-		p->rmdnum = num1;
-		ni65_recv_intr(dev,csr0);
-		if((p->rmdhead[num2].u.s.status & RCV_OWN))
-			break;	/* ok, we are 'in sync' again */
-	}
-	else
-		break;
- }
-}
-#endif
-
-	if( (csr0 & (CSR0_RXON | CSR0_TXON)) != (CSR0_RXON | CSR0_TXON) ) {
-		printk(KERN_DEBUG "%s: RX or TX was offline -> restart\n",dev->name);
-		ni65_stop_start(dev,p);
-	}
-	else
-		writedatareg(CSR0_INEA);
-
-	spin_unlock(&p->ring_lock);
-	return IRQ_HANDLED;
-}
-
-/*
- * We have received an Xmit-Interrupt ..
- * send a new packet if necessary
- */
-static void ni65_xmit_intr(struct net_device *dev,int csr0)
-{
-	struct priv *p = dev->ml_priv;
-
-	while(p->xmit_queued)
-	{
-		struct tmd *tmdp = p->tmdhead + p->tmdlast;
-		int tmdstat = tmdp->u.s.status;
-
-		if(tmdstat & XMIT_OWN)
-			break;
-
-		if(tmdstat & XMIT_ERR)
-		{
-#if 0
-			if(tmdp->status2 & XMIT_TDRMASK && debuglevel > 3)
-				printk(KERN_ERR "%s: tdr-problems (e.g. no resistor)\n",dev->name);
-#endif
-		 /* checking some errors */
-			if(tmdp->status2 & XMIT_RTRY)
-				dev->stats.tx_aborted_errors++;
-			if(tmdp->status2 & XMIT_LCAR)
-				dev->stats.tx_carrier_errors++;
-			if(tmdp->status2 & (XMIT_BUFF | XMIT_UFLO )) {
-		/* this stops the xmitter */
-				dev->stats.tx_fifo_errors++;
-				if(debuglevel > 0)
-					printk(KERN_ERR "%s: Xmit FIFO/BUFF error\n",dev->name);
-				if(p->features & INIT_RING_BEFORE_START) {
-					tmdp->u.s.status = XMIT_OWN | XMIT_START | XMIT_END;	/* test: resend this frame */
-					ni65_stop_start(dev,p);
-					break;	/* no more Xmit processing .. */
-				}
-				else
-				 ni65_stop_start(dev,p);
-			}
-			if(debuglevel > 2)
-				printk(KERN_ERR "%s: xmit-error: %04x %02x-%04x\n",dev->name,csr0,(int) tmdstat,(int) tmdp->status2);
-			if(!(csr0 & CSR0_BABL)) /* don't count errors twice */
-				dev->stats.tx_errors++;
-			tmdp->status2 = 0;
-		}
-		else {
-			dev->stats.tx_bytes -= (short)(tmdp->blen);
-			dev->stats.tx_packets++;
-		}
-
-#ifdef XMT_VIA_SKB
-		if(p->tmd_skb[p->tmdlast]) {
-			 dev_kfree_skb_irq(p->tmd_skb[p->tmdlast]);
-			 p->tmd_skb[p->tmdlast] = NULL;
-		}
-#endif
-
-		p->tmdlast = (p->tmdlast + 1) & (TMDNUM-1);
-		if(p->tmdlast == p->tmdnum)
-			p->xmit_queued = 0;
-	}
-	netif_wake_queue(dev);
-}
-
-/*
- * We have received a packet
- */
-static void ni65_recv_intr(struct net_device *dev,int csr0)
-{
-	struct rmd *rmdp;
-	int rmdstat,len;
-	int cnt=0;
-	struct priv *p = dev->ml_priv;
-
-	rmdp = p->rmdhead + p->rmdnum;
-	while(!( (rmdstat = rmdp->u.s.status) & RCV_OWN))
-	{
-		cnt++;
-		if( (rmdstat & (RCV_START | RCV_END | RCV_ERR)) != (RCV_START | RCV_END) ) /* error or oversized? */
-		{
-			if(!(rmdstat & RCV_ERR)) {
-				if(rmdstat & RCV_START)
-				{
-					dev->stats.rx_length_errors++;
-					printk(KERN_ERR "%s: recv, packet too long: %d\n",dev->name,rmdp->mlen & 0x0fff);
-				}
-			}
-			else {
-				if(debuglevel > 2)
-					printk(KERN_ERR "%s: receive-error: %04x, lance-status: %04x/%04x\n",
-									dev->name,(int) rmdstat,csr0,(int) inw(PORT+L_DATAREG) );
-				if(rmdstat & RCV_FRAM)
-					dev->stats.rx_frame_errors++;
-				if(rmdstat & RCV_OFLO)
-					dev->stats.rx_over_errors++;
-				if(rmdstat & RCV_CRC)
-					dev->stats.rx_crc_errors++;
-				if(rmdstat & RCV_BUF_ERR)
-					dev->stats.rx_fifo_errors++;
-			}
-			if(!(csr0 & CSR0_MISS)) /* don't count errors twice */
-				dev->stats.rx_errors++;
-		}
-		else if( (len = (rmdp->mlen & 0x0fff) - 4) >= 60)
-		{
-#ifdef RCV_VIA_SKB
-			struct sk_buff *skb = alloc_skb(R_BUF_SIZE+2+16,GFP_ATOMIC);
-			if (skb)
-				skb_reserve(skb,16);
-#else
-			struct sk_buff *skb = netdev_alloc_skb(dev, len + 2);
-#endif
-			if(skb)
-			{
-				skb_reserve(skb,2);
-#ifdef RCV_VIA_SKB
-				if( (unsigned long) (skb->data + R_BUF_SIZE) > 0x1000000) {
-					skb_put(skb,len);
-					skb_copy_to_linear_data(skb, (unsigned char *)(p->recv_skb[p->rmdnum]->data),len);
-				}
-				else {
-					struct sk_buff *skb1 = p->recv_skb[p->rmdnum];
-					skb_put(skb,R_BUF_SIZE);
-					p->recv_skb[p->rmdnum] = skb;
-					rmdp->u.buffer = (u32) isa_virt_to_bus(skb->data);
-					skb = skb1;
-					skb_trim(skb,len);
-				}
-#else
-				skb_put(skb,len);
-				skb_copy_to_linear_data(skb, (unsigned char *) p->recvbounce[p->rmdnum],len);
-#endif
-				dev->stats.rx_packets++;
-				dev->stats.rx_bytes += len;
-				skb->protocol=eth_type_trans(skb,dev);
-				netif_rx(skb);
-			}
-			else
-			{
-				printk(KERN_ERR "%s: can't alloc new sk_buff\n",dev->name);
-				dev->stats.rx_dropped++;
-			}
-		}
-		else {
-			printk(KERN_INFO "%s: received runt packet\n",dev->name);
-			dev->stats.rx_errors++;
-		}
-		rmdp->blen = -(R_BUF_SIZE-8);
-		rmdp->mlen = 0;
-		rmdp->u.s.status = RCV_OWN; /* change owner */
-		p->rmdnum = (p->rmdnum + 1) & (RMDNUM-1);
-		rmdp = p->rmdhead + p->rmdnum;
-	}
-}
-
-/*
- * kick xmitter ..
- */
-
-static void ni65_timeout(struct net_device *dev)
-{
-	int i;
-	struct priv *p = dev->ml_priv;
-
-	printk(KERN_ERR "%s: xmitter timed out, try to restart!\n",dev->name);
-	for(i=0;i<TMDNUM;i++)
-		printk("%02x ",p->tmdhead[i].u.s.status);
-	printk("\n");
-	ni65_lance_reinit(dev);
-	netif_trans_update(dev); /* prevent tx timeout */
-	netif_wake_queue(dev);
-}
-
-/*
- *	Send a packet
- */
-
-static netdev_tx_t ni65_send_packet(struct sk_buff *skb,
-				    struct net_device *dev)
-{
-	struct priv *p = dev->ml_priv;
-
-	netif_stop_queue(dev);
-
-	if (test_and_set_bit(0, (void*)&p->lock)) {
-		printk(KERN_ERR "%s: Queue was locked.\n", dev->name);
-		return NETDEV_TX_BUSY;
-	}
-
-	{
-		short len = ETH_ZLEN < skb->len ? skb->len : ETH_ZLEN;
-		struct tmd *tmdp;
-		unsigned long flags;
-
-#ifdef XMT_VIA_SKB
-		if( (unsigned long) (skb->data + skb->len) > 0x1000000) {
-#endif
-
-			skb_copy_from_linear_data(skb, p->tmdbounce[p->tmdbouncenum],
-				      skb->len > T_BUF_SIZE ? T_BUF_SIZE :
-							      skb->len);
-			if (len > skb->len)
-				memset((char *)p->tmdbounce[p->tmdbouncenum]+skb->len, 0, len-skb->len);
-			dev_kfree_skb (skb);
-
-			spin_lock_irqsave(&p->ring_lock, flags);
-			tmdp = p->tmdhead + p->tmdnum;
-			tmdp->u.buffer = (u32) isa_virt_to_bus(p->tmdbounce[p->tmdbouncenum]);
-			p->tmdbouncenum = (p->tmdbouncenum + 1) & (TMDNUM - 1);
-
-#ifdef XMT_VIA_SKB
-		}
-		else {
-			spin_lock_irqsave(&p->ring_lock, flags);
-
-			tmdp = p->tmdhead + p->tmdnum;
-			tmdp->u.buffer = (u32) isa_virt_to_bus(skb->data);
-			p->tmd_skb[p->tmdnum] = skb;
-		}
-#endif
-		tmdp->blen = -len;
-
-		tmdp->u.s.status = XMIT_OWN | XMIT_START | XMIT_END;
-		writedatareg(CSR0_TDMD | CSR0_INEA); /* enable xmit & interrupt */
-
-		p->xmit_queued = 1;
-		p->tmdnum = (p->tmdnum + 1) & (TMDNUM-1);
-
-		if(p->tmdnum != p->tmdlast)
-			netif_wake_queue(dev);
-
-		p->lock = 0;
-
-		spin_unlock_irqrestore(&p->ring_lock, flags);
-	}
-
-	return NETDEV_TX_OK;
-}
-
-static void set_multicast_list(struct net_device *dev)
-{
-	if(!ni65_lance_reinit(dev))
-		printk(KERN_ERR "%s: Can't switch card into MC mode!\n",dev->name);
-	netif_wake_queue(dev);
-}
-
-#ifdef MODULE
-static struct net_device *dev_ni65;
-
-module_param_hw(irq, int, irq, 0);
-module_param_hw(io, int, ioport, 0);
-module_param_hw(dma, int, dma, 0);
-MODULE_PARM_DESC(irq, "ni6510 IRQ number (ignored for some cards)");
-MODULE_PARM_DESC(io, "ni6510 I/O base address");
-MODULE_PARM_DESC(dma, "ni6510 ISA DMA channel (ignored for some cards)");
-
-int __init init_module(void)
-{
- 	dev_ni65 = ni65_probe(-1);
-	return PTR_ERR_OR_ZERO(dev_ni65);
-}
-
-void __exit cleanup_module(void)
-{
- 	unregister_netdev(dev_ni65);
- 	cleanup_card(dev_ni65);
- 	free_netdev(dev_ni65);
-}
-#endif /* MODULE */
-
-MODULE_LICENSE("GPL");
diff --git a/drivers/net/ethernet/amd/ni65.h b/drivers/net/ethernet/amd/ni65.h
deleted file mode 100644
index e6217e35edf0..000000000000
--- a/drivers/net/ethernet/amd/ni65.h
+++ /dev/null
@@ -1,121 +0,0 @@
-/* am7990 (lance) definitions
- *
- * This is an extension to the Linux operating system, and is covered by
- * same GNU General Public License that covers that work.
- *
- * Michael Hipp
- * email: mhipp@student.uni-tuebingen.de
- *
- * sources: (mail me or ask archie if you need them)
- *    crynwr-packet-driver
- */
-
-/*
- * 	Control and Status Register 0 (CSR0) bit definitions
- * (R=Readable) (W=Writeable) (S=Set on write) (C-Clear on write)
- *
- */
-
-#define CSR0_ERR	0x8000	/* Error summary (R) */
-#define CSR0_BABL	0x4000	/* Babble transmitter timeout error (RC) */
-#define CSR0_CERR	0x2000	/* Collision Error (RC) */
-#define CSR0_MISS	0x1000	/* Missed packet (RC) */
-#define CSR0_MERR	0x0800	/* Memory Error (RC) */
-#define CSR0_RINT	0x0400	/* Receiver Interrupt (RC) */
-#define CSR0_TINT       0x0200	/* Transmit Interrupt (RC) */
-#define CSR0_IDON	0x0100	/* Initialization Done (RC) */
-#define CSR0_INTR	0x0080	/* Interrupt Flag (R) */
-#define CSR0_INEA	0x0040	/* Interrupt Enable (RW) */
-#define CSR0_RXON	0x0020	/* Receiver on (R) */
-#define CSR0_TXON	0x0010	/* Transmitter on (R) */
-#define CSR0_TDMD	0x0008	/* Transmit Demand (RS) */
-#define CSR0_STOP	0x0004	/* Stop (RS) */
-#define CSR0_STRT	0x0002	/* Start (RS) */
-#define CSR0_INIT	0x0001	/* Initialize (RS) */
-
-#define CSR0_CLRALL    0x7f00	/* mask for all clearable bits */
-/*
- *	Initialization Block  Mode operation Bit Definitions.
- */
-
-#define M_PROM		0x8000	/* Promiscuous Mode */
-#define M_INTL		0x0040	/* Internal Loopback */
-#define M_DRTY		0x0020	/* Disable Retry */
-#define M_COLL		0x0010	/* Force Collision */
-#define M_DTCR		0x0008	/* Disable Transmit CRC) */
-#define M_LOOP		0x0004	/* Loopback */
-#define M_DTX		0x0002	/* Disable the Transmitter */
-#define M_DRX		0x0001	/* Disable the Receiver */
-
-
-/*
- * 	Receive message descriptor bit definitions.
- */
-
-#define RCV_OWN		0x80	/* owner bit 0 = host, 1 = lance */
-#define RCV_ERR		0x40	/* Error Summary */
-#define RCV_FRAM	0x20	/* Framing Error */
-#define RCV_OFLO	0x10	/* Overflow Error */
-#define RCV_CRC		0x08	/* CRC Error */
-#define RCV_BUF_ERR	0x04	/* Buffer Error */
-#define RCV_START	0x02	/* Start of Packet */
-#define RCV_END		0x01	/* End of Packet */
-
-
-/*
- *	Transmit  message descriptor bit definitions.
- */
-
-#define XMIT_OWN	0x80	/* owner bit 0 = host, 1 = lance */
-#define XMIT_ERR	0x40	/* Error Summary */
-#define XMIT_RETRY	0x10	/* more the 1 retry needed to Xmit */
-#define XMIT_1_RETRY	0x08	/* one retry needed to Xmit */
-#define XMIT_DEF	0x04	/* Deferred */
-#define XMIT_START	0x02	/* Start of Packet */
-#define XMIT_END	0x01	/* End of Packet */
-
-/*
- * transmit status (2) (valid if XMIT_ERR == 1)
- */
-
-#define XMIT_TDRMASK    0x03ff	/* time-domain-reflectometer-value */
-#define XMIT_RTRY 	0x0400	/* Failed after 16 retransmissions  */
-#define XMIT_LCAR 	0x0800	/* Loss of Carrier */
-#define XMIT_LCOL 	0x1000	/* Late collision */
-#define XMIT_RESERV 	0x2000	/* Reserved */
-#define XMIT_UFLO 	0x4000	/* Underflow (late memory) */
-#define XMIT_BUFF 	0x8000	/* Buffering error (no ENP) */
-
-struct init_block {
-	unsigned short mode;
-	unsigned char eaddr[6];
-	unsigned char filter[8];
-	/* bit 29-31: number of rmd's (power of 2) */
-	u32 rrp;		/* receive ring pointer (align 8) */
-	/* bit 29-31: number of tmd's (power of 2) */
-	u32 trp;		/* transmit ring pointer (align 8) */
-};
-
-struct rmd {			/* Receive Message Descriptor */
-	union {
-		volatile u32 buffer;
-		struct {
-			volatile unsigned char dummy[3];
-			volatile unsigned char status;
-		} s;
-	} u;
-	volatile short blen;
-	volatile unsigned short mlen;
-};
-
-struct tmd {
-	union {
-		volatile u32 buffer;
-		struct {
-			volatile unsigned char dummy[3];
-			volatile unsigned char status;
-		} s;
-	} u;
-	volatile unsigned short blen;
-	volatile unsigned short status2;
-};
diff --git a/drivers/net/ethernet/amd/nmclan_cs.c b/drivers/net/ethernet/amd/nmclan_cs.c
index 9c152d85840d..37054a670407 100644
--- a/drivers/net/ethernet/amd/nmclan_cs.c
+++ b/drivers/net/ethernet/amd/nmclan_cs.c
@@ -114,8 +114,6 @@ Log: nmclan_cs.c,v
 #define pr_fmt(fmt) KBUILD_MODNAME ": " fmt
 
 #define DRV_NAME	"nmclan_cs"
-#define DRV_VERSION	"0.16"
-
 
 /* ----------------------------------------------------------------------------
 Conditional Compilation Options
@@ -367,7 +365,7 @@ typedef struct _mace_private {
 
     char tx_free_frames; /* Number of free transmit frame buffers */
     char tx_irq_disabled; /* MACE TX interrupt disabled */
-    
+
     spinlock_t bank_lock; /* Must be held if you step off bank 0 */
 } mace_private;
 
@@ -407,7 +405,7 @@ static int mace_open(struct net_device *dev);
 static int mace_close(struct net_device *dev);
 static netdev_tx_t mace_start_xmit(struct sk_buff *skb,
 					 struct net_device *dev);
-static void mace_tx_timeout(struct net_device *dev);
+static void mace_tx_timeout(struct net_device *dev, unsigned int txqueue);
 static irqreturn_t mace_interrupt(int irq, void *dev_id);
 static struct net_device_stats *mace_get_stats(struct net_device *dev);
 static int mace_rx(struct net_device *dev, unsigned char RxCnt);
@@ -444,7 +442,7 @@ static int nmclan_probe(struct pcmcia_device *link)
     lp = netdev_priv(dev);
     lp->p_dev = link;
     link->priv = dev;
-    
+
     spin_lock_init(&lp->bank_lock);
     link->resource[0]->end = 32;
     link->resource[0]->flags |= IO_DATA_PATH_WIDTH_AUTO;
@@ -487,10 +485,10 @@ static int mace_read(mace_private *lp, unsigned int ioaddr, int reg)
   unsigned long flags;
 
   switch (reg >> 4) {
-    case 0: /* register 0-15 */
+  case 0: /* register 0-15 */
       data = inb(ioaddr + AM2150_MACE_BASE + reg);
       break;
-    case 1: /* register 16-31 */
+  case 1: /* register 16-31 */
       spin_lock_irqsave(&lp->bank_lock, flags);
       MACEBANK(1);
       data = inb(ioaddr + AM2150_MACE_BASE + (reg & 0x0F));
@@ -514,10 +512,10 @@ static void mace_write(mace_private *lp, unsigned int ioaddr, int reg,
   unsigned long flags;
 
   switch (reg >> 4) {
-    case 0: /* register 0-15 */
+  case 0: /* register 0-15 */
       outb(data & 0xFF, ioaddr + AM2150_MACE_BASE + reg);
       break;
-    case 1: /* register 16-31 */
+  case 1: /* register 16-31 */
       spin_lock_irqsave(&lp->bank_lock, flags);
       MACEBANK(1);
       outb(data & 0xFF, ioaddr + AM2150_MACE_BASE + (reg & 0x0F));
@@ -531,7 +529,8 @@ static void mace_write(mace_private *lp, unsigned int ioaddr, int reg,
 mace_init
 	Resets the MACE chip.
 ---------------------------------------------------------------------------- */
-static int mace_init(mace_private *lp, unsigned int ioaddr, char *enet_addr)
+static int mace_init(mace_private *lp, unsigned int ioaddr,
+		     const char *enet_addr)
 {
   int i;
   int ct = 0;
@@ -543,7 +542,7 @@ static int mace_init(mace_private *lp, unsigned int ioaddr, char *enet_addr)
     if(++ct > 500)
     {
 	pr_err("reset failed, card removed?\n");
-    	return -1;
+	return -1;
     }
     udelay(1);
   }
@@ -568,13 +567,13 @@ static int mace_init(mace_private *lp, unsigned int ioaddr, char *enet_addr)
    * Or just set ASEL in PHYCC below!
    */
   switch (if_port) {
-    case 1:
+  case 1:
       mace_write(lp, ioaddr, MACE_PLSCC, 0x02);
       break;
-    case 2:
+  case 2:
       mace_write(lp, ioaddr, MACE_PLSCC, 0x00);
       break;
-    default:
+  default:
       mace_write(lp, ioaddr, MACE_PHYCC, /* ASEL */ 4);
       /* ASEL Auto Select.  When set, the PORTSEL[1-0] bits are overridden,
 	 and the MACE device will automatically select the operating media
@@ -587,11 +586,11 @@ static int mace_init(mace_private *lp, unsigned int ioaddr, char *enet_addr)
   ct = 0;
   while (mace_read(lp, ioaddr, MACE_IAC) & MACE_IAC_ADDRCHG)
   {
-  	if(++ ct > 500)
-  	{
+	if(++ ct > 500)
+	{
 		pr_err("ADDRCHG timeout, card removed?\n");
-  		return -1;
-  	}
+		return -1;
+	}
   }
   /* Set PADR register */
   for (i = 0; i < ETH_ALEN; i++)
@@ -637,7 +636,7 @@ static int nmclan_config(struct pcmcia_device *link)
 	  kfree(buf);
 	  goto failed;
   }
-  memcpy(dev->dev_addr, buf, ETH_ALEN);
+  eth_hw_addr_set(dev, buf);
   kfree(buf);
 
   /* Verify configuration by reading the MACE ID. */
@@ -652,12 +651,12 @@ static int nmclan_config(struct pcmcia_device *link)
     } else {
       pr_notice("mace id not found: %x %x should be 0x40 0x?9\n",
 		sig[0], sig[1]);
-      return -ENODEV;
+      goto failed;
     }
   }
 
   if(mace_init(lp, ioaddr, dev->dev_addr) == -1)
-  	goto failed;
+	goto failed;
 
   /* The if_port symbol can be set when the module is loaded */
   if (if_port <= 2)
@@ -761,7 +760,7 @@ static int mace_config(struct net_device *dev, struct ifmap *map)
 {
   if ((map->port != (u_char)(-1)) && (map->port != dev->if_port)) {
     if (map->port <= 2) {
-      dev->if_port = map->port;
+      WRITE_ONCE(dev->if_port, map->port);
       netdev_info(dev, "switched to %s port\n", if_names[dev->if_port]);
     } else
       return -EINVAL;
@@ -816,8 +815,7 @@ static int mace_close(struct net_device *dev)
 static void netdev_get_drvinfo(struct net_device *dev,
 			       struct ethtool_drvinfo *info)
 {
-	strlcpy(info->driver, DRV_NAME, sizeof(info->driver));
-	strlcpy(info->version, DRV_VERSION, sizeof(info->version));
+	strscpy(info->driver, DRV_NAME, sizeof(info->driver));
 	snprintf(info->bus_info, sizeof(info->bus_info),
 		"PCMCIA 0x%lx", dev->base_addr);
 }
@@ -837,7 +835,7 @@ mace_start_xmit
 	failed, put skb back into a list."
 ---------------------------------------------------------------------------- */
 
-static void mace_tx_timeout(struct net_device *dev)
+static void mace_tx_timeout(struct net_device *dev, unsigned int txqueue)
 {
   mace_private *lp = netdev_priv(dev);
   struct pcmcia_device *link = lp->p_dev;
@@ -920,7 +918,7 @@ static irqreturn_t mace_interrupt(int irq, void *dev_id)
   int status;
   int IntrCnt = MACE_MAX_IR_ITERATIONS;
 
-  if (dev == NULL) {
+  if (!dev) {
     pr_debug("mace_interrupt(): irq 0x%X for unknown device.\n",
 	  irq);
     return IRQ_NONE;
@@ -1104,13 +1102,13 @@ static int mace_rx(struct net_device *dev, unsigned char RxCnt)
 
       skb = netdev_alloc_skb(dev, pkt_len + 2);
 
-      if (skb != NULL) {
+      if (skb) {
 	skb_reserve(skb, 2);
 	insw(ioaddr + AM2150_RCV, skb_put(skb, pkt_len), pkt_len>>1);
 	if (pkt_len & 1)
 	    *(skb_tail_pointer(skb) - 1) = inb(ioaddr + AM2150_RCV);
 	skb->protocol = eth_type_trans(skb, dev);
-	
+
 	netif_rx(skb); /* Send the packet to the upper (protocol) layers. */
 
 	dev->stats.rx_packets++;
diff --git a/drivers/net/ethernet/amd/pcnet32.c b/drivers/net/ethernet/amd/pcnet32.c
index 86369d7c9a0f..9eaefa0f5e80 100644
--- a/drivers/net/ethernet/amd/pcnet32.c
+++ b/drivers/net/ethernet/amd/pcnet32.c
@@ -24,13 +24,9 @@
 #define pr_fmt(fmt) KBUILD_MODNAME ": " fmt
 
 #define DRV_NAME	"pcnet32"
-#define DRV_VERSION	"1.35"
 #define DRV_RELDATE	"21.Apr.2008"
 #define PFX		DRV_NAME ": "
 
-static const char *const version =
-    DRV_NAME ".c:v" DRV_VERSION " " DRV_RELDATE " tsbogend@alpha.franken.de\n";
-
 #include <linux/module.h>
 #include <linux/kernel.h>
 #include <linux/sched.h>
@@ -254,7 +250,7 @@ struct pcnet32_access {
 
 /*
  * The first field of pcnet32_private is read by the ethernet device
- * so the structure should be allocated using pci_alloc_consistent().
+ * so the structure should be allocated using dma_alloc_coherent().
  */
 struct pcnet32_private {
 	struct pcnet32_init_block *init_block;
@@ -262,7 +258,7 @@ struct pcnet32_private {
 	struct pcnet32_rx_head	*rx_ring;
 	struct pcnet32_tx_head	*tx_ring;
 	dma_addr_t		init_dma_addr;/* DMA address of beginning of the init block,
-				   returned by pci_alloc_consistent */
+				   returned by dma_alloc_coherent */
 	struct pci_dev		*pci_dev;
 	const char		*name;
 	/* The saved address of a sent-in-place packet/buffer, for skfree(). */
@@ -314,14 +310,14 @@ static int pcnet32_open(struct net_device *);
 static int pcnet32_init_ring(struct net_device *);
 static netdev_tx_t pcnet32_start_xmit(struct sk_buff *,
 				      struct net_device *);
-static void pcnet32_tx_timeout(struct net_device *dev);
+static void pcnet32_tx_timeout(struct net_device *dev, unsigned int txqueue);
 static irqreturn_t pcnet32_interrupt(int, void *);
 static int pcnet32_close(struct net_device *);
 static struct net_device_stats *pcnet32_get_stats(struct net_device *);
 static void pcnet32_load_multicast(struct net_device *dev);
 static void pcnet32_set_multicast_list(struct net_device *);
 static int pcnet32_ioctl(struct net_device *, struct ifreq *, int);
-static void pcnet32_watchdog(struct net_device *);
+static void pcnet32_watchdog(struct timer_list *);
 static int mdio_read(struct net_device *dev, int phy_id, int reg_num);
 static void mdio_write(struct net_device *dev, int phy_id, int reg_num,
 		       int val);
@@ -466,7 +462,7 @@ static void pcnet32_netif_start(struct net_device *dev)
 	val = lp->a->read_csr(ioaddr, CSR3);
 	val &= 0x00ff;
 	lp->a->write_csr(ioaddr, CSR3, val);
-	napi_enable(&lp->napi);
+	napi_enable_locked(&lp->napi);
 }
 
 /*
@@ -489,10 +485,10 @@ static void pcnet32_realloc_tx_ring(struct net_device *dev,
 	pcnet32_purge_tx_ring(dev);
 
 	new_tx_ring =
-		pci_zalloc_consistent(lp->pci_dev,
-				      sizeof(struct pcnet32_tx_head) * entries,
-				      &new_ring_dma_addr);
-	if (new_tx_ring == NULL)
+		dma_alloc_coherent(&lp->pci_dev->dev,
+				   sizeof(struct pcnet32_tx_head) * entries,
+				   &new_ring_dma_addr, GFP_ATOMIC);
+	if (!new_tx_ring)
 		return;
 
 	new_dma_addr_list = kcalloc(entries, sizeof(dma_addr_t), GFP_ATOMIC);
@@ -505,9 +501,9 @@ static void pcnet32_realloc_tx_ring(struct net_device *dev,
 
 	kfree(lp->tx_skbuff);
 	kfree(lp->tx_dma_addr);
-	pci_free_consistent(lp->pci_dev,
-			    sizeof(struct pcnet32_tx_head) * lp->tx_ring_size,
-			    lp->tx_ring, lp->tx_ring_dma_addr);
+	dma_free_coherent(&lp->pci_dev->dev,
+			  sizeof(struct pcnet32_tx_head) * lp->tx_ring_size,
+			  lp->tx_ring, lp->tx_ring_dma_addr);
 
 	lp->tx_ring_size = entries;
 	lp->tx_mod_mask = lp->tx_ring_size - 1;
@@ -521,10 +517,9 @@ static void pcnet32_realloc_tx_ring(struct net_device *dev,
 free_new_lists:
 	kfree(new_dma_addr_list);
 free_new_tx_ring:
-	pci_free_consistent(lp->pci_dev,
-			    sizeof(struct pcnet32_tx_head) * entries,
-			    new_tx_ring,
-			    new_ring_dma_addr);
+	dma_free_coherent(&lp->pci_dev->dev,
+			  sizeof(struct pcnet32_tx_head) * entries,
+			  new_tx_ring, new_ring_dma_addr);
 }
 
 /*
@@ -549,10 +544,10 @@ static void pcnet32_realloc_rx_ring(struct net_device *dev,
 	unsigned int entries = BIT(size);
 
 	new_rx_ring =
-		pci_zalloc_consistent(lp->pci_dev,
-				      sizeof(struct pcnet32_rx_head) * entries,
-				      &new_ring_dma_addr);
-	if (new_rx_ring == NULL)
+		dma_alloc_coherent(&lp->pci_dev->dev,
+				   sizeof(struct pcnet32_rx_head) * entries,
+				   &new_ring_dma_addr, GFP_ATOMIC);
+	if (!new_rx_ring)
 		return;
 
 	new_dma_addr_list = kcalloc(entries, sizeof(dma_addr_t), GFP_ATOMIC);
@@ -584,10 +579,9 @@ static void pcnet32_realloc_rx_ring(struct net_device *dev,
 		skb_reserve(rx_skbuff, NET_IP_ALIGN);
 
 		new_dma_addr_list[new] =
-			    pci_map_single(lp->pci_dev, rx_skbuff->data,
-					   PKT_BUF_SIZE, PCI_DMA_FROMDEVICE);
-		if (pci_dma_mapping_error(lp->pci_dev,
-					  new_dma_addr_list[new])) {
+			    dma_map_single(&lp->pci_dev->dev, rx_skbuff->data,
+					   PKT_BUF_SIZE, DMA_FROM_DEVICE);
+		if (dma_mapping_error(&lp->pci_dev->dev, new_dma_addr_list[new])) {
 			netif_err(lp, drv, dev, "%s dma mapping failed\n",
 				  __func__);
 			dev_kfree_skb(new_skb_list[new]);
@@ -600,22 +594,20 @@ static void pcnet32_realloc_rx_ring(struct net_device *dev,
 	/* and free any unneeded buffers */
 	for (; new < lp->rx_ring_size; new++) {
 		if (lp->rx_skbuff[new]) {
-			if (!pci_dma_mapping_error(lp->pci_dev,
-						   lp->rx_dma_addr[new]))
-				pci_unmap_single(lp->pci_dev,
+			if (!dma_mapping_error(&lp->pci_dev->dev, lp->rx_dma_addr[new]))
+				dma_unmap_single(&lp->pci_dev->dev,
 						 lp->rx_dma_addr[new],
 						 PKT_BUF_SIZE,
-						 PCI_DMA_FROMDEVICE);
+						 DMA_FROM_DEVICE);
 			dev_kfree_skb(lp->rx_skbuff[new]);
 		}
 	}
 
 	kfree(lp->rx_skbuff);
 	kfree(lp->rx_dma_addr);
-	pci_free_consistent(lp->pci_dev,
-			    sizeof(struct pcnet32_rx_head) *
-			    lp->rx_ring_size, lp->rx_ring,
-			    lp->rx_ring_dma_addr);
+	dma_free_coherent(&lp->pci_dev->dev,
+			  sizeof(struct pcnet32_rx_head) * lp->rx_ring_size,
+			  lp->rx_ring, lp->rx_ring_dma_addr);
 
 	lp->rx_ring_size = entries;
 	lp->rx_mod_mask = lp->rx_ring_size - 1;
@@ -629,12 +621,11 @@ static void pcnet32_realloc_rx_ring(struct net_device *dev,
 free_all_new:
 	while (--new >= lp->rx_ring_size) {
 		if (new_skb_list[new]) {
-			if (!pci_dma_mapping_error(lp->pci_dev,
-						   new_dma_addr_list[new]))
-				pci_unmap_single(lp->pci_dev,
+			if (!dma_mapping_error(&lp->pci_dev->dev, new_dma_addr_list[new]))
+				dma_unmap_single(&lp->pci_dev->dev,
 						 new_dma_addr_list[new],
 						 PKT_BUF_SIZE,
-						 PCI_DMA_FROMDEVICE);
+						 DMA_FROM_DEVICE);
 			dev_kfree_skb(new_skb_list[new]);
 		}
 	}
@@ -642,10 +633,9 @@ free_all_new:
 free_new_lists:
 	kfree(new_dma_addr_list);
 free_new_rx_ring:
-	pci_free_consistent(lp->pci_dev,
-			    sizeof(struct pcnet32_rx_head) * entries,
-			    new_rx_ring,
-			    new_ring_dma_addr);
+	dma_free_coherent(&lp->pci_dev->dev,
+			  sizeof(struct pcnet32_rx_head) * entries,
+			  new_rx_ring, new_ring_dma_addr);
 }
 
 static void pcnet32_purge_rx_ring(struct net_device *dev)
@@ -658,12 +648,11 @@ static void pcnet32_purge_rx_ring(struct net_device *dev)
 		lp->rx_ring[i].status = 0;	/* CPU owns buffer */
 		wmb();		/* Make sure adapter sees owner change */
 		if (lp->rx_skbuff[i]) {
-			if (!pci_dma_mapping_error(lp->pci_dev,
-						   lp->rx_dma_addr[i]))
-				pci_unmap_single(lp->pci_dev,
+			if (!dma_mapping_error(&lp->pci_dev->dev, lp->rx_dma_addr[i]))
+				dma_unmap_single(&lp->pci_dev->dev,
 						 lp->rx_dma_addr[i],
 						 PKT_BUF_SIZE,
-						 PCI_DMA_FROMDEVICE);
+						 DMA_FROM_DEVICE);
 			dev_kfree_skb_any(lp->rx_skbuff[i]);
 		}
 		lp->rx_skbuff[i] = NULL;
@@ -731,12 +720,10 @@ static int pcnet32_get_link_ksettings(struct net_device *dev,
 {
 	struct pcnet32_private *lp = netdev_priv(dev);
 	unsigned long flags;
-	int r = -EOPNOTSUPP;
 
 	spin_lock_irqsave(&lp->lock, flags);
 	if (lp->mii) {
 		mii_ethtool_get_link_ksettings(&lp->mii_if, cmd);
-		r = 0;
 	} else if (lp->chip_version == PCNET32_79C970A) {
 		if (lp->autoneg) {
 			cmd->base.autoneg = AUTONEG_ENABLE;
@@ -753,10 +740,9 @@ static int pcnet32_get_link_ksettings(struct net_device *dev,
 		ethtool_convert_legacy_u32_to_link_mode(
 						cmd->link_modes.supported,
 						SUPPORTED_TP | SUPPORTED_AUI);
-		r = 0;
 	}
 	spin_unlock_irqrestore(&lp->lock, flags);
-	return r;
+	return 0;
 }
 
 static int pcnet32_set_link_ksettings(struct net_device *dev,
@@ -811,10 +797,9 @@ static void pcnet32_get_drvinfo(struct net_device *dev,
 {
 	struct pcnet32_private *lp = netdev_priv(dev);
 
-	strlcpy(info->driver, DRV_NAME, sizeof(info->driver));
-	strlcpy(info->version, DRV_VERSION, sizeof(info->version));
+	strscpy(info->driver, DRV_NAME, sizeof(info->driver));
 	if (lp->pci_dev)
-		strlcpy(info->bus_info, pci_name(lp->pci_dev),
+		strscpy(info->bus_info, pci_name(lp->pci_dev),
 			sizeof(info->bus_info));
 	else
 		snprintf(info->bus_info, sizeof(info->bus_info),
@@ -875,7 +860,9 @@ static int pcnet32_nway_reset(struct net_device *dev)
 }
 
 static void pcnet32_get_ringparam(struct net_device *dev,
-				  struct ethtool_ringparam *ering)
+				  struct ethtool_ringparam *ering,
+				  struct kernel_ethtool_ringparam *kernel_ering,
+				  struct netlink_ext_ack *extack)
 {
 	struct pcnet32_private *lp = netdev_priv(dev);
 
@@ -886,7 +873,9 @@ static void pcnet32_get_ringparam(struct net_device *dev,
 }
 
 static int pcnet32_set_ringparam(struct net_device *dev,
-				 struct ethtool_ringparam *ering)
+				 struct ethtool_ringparam *ering,
+				 struct kernel_ethtool_ringparam *kernel_ering,
+				 struct netlink_ext_ack *extack)
 {
 	struct pcnet32_private *lp = netdev_priv(dev);
 	unsigned long flags;
@@ -900,6 +889,7 @@ static int pcnet32_set_ringparam(struct net_device *dev,
 	if (netif_running(dev))
 		pcnet32_netif_stop(dev);
 
+	netdev_lock(dev);
 	spin_lock_irqsave(&lp->lock, flags);
 	lp->a->write_csr(ioaddr, CSR0, CSR0_STOP);	/* stop the chip */
 
@@ -931,6 +921,7 @@ static int pcnet32_set_ringparam(struct net_device *dev,
 	}
 
 	spin_unlock_irqrestore(&lp->lock, flags);
+	netdev_unlock(dev);
 
 	netif_info(lp, drv, dev, "Ring Param Settings: RX: %d, TX: %d\n",
 		   lp->rx_ring_size, lp->tx_ring_size);
@@ -996,6 +987,7 @@ static int pcnet32_loopback_test(struct net_device *dev, uint64_t * data1)
 	if (netif_running(dev))
 		pcnet32_netif_stop(dev);
 
+	netdev_lock(dev);
 	spin_lock_irqsave(&lp->lock, flags);
 	lp->a->write_csr(ioaddr, CSR0, CSR0_STOP);	/* stop the chip */
 
@@ -1044,9 +1036,9 @@ static int pcnet32_loopback_test(struct net_device *dev, uint64_t * data1)
 			*packet++ = i;
 
 		lp->tx_dma_addr[x] =
-			pci_map_single(lp->pci_dev, skb->data, skb->len,
-				       PCI_DMA_TODEVICE);
-		if (pci_dma_mapping_error(lp->pci_dev, lp->tx_dma_addr[x])) {
+			dma_map_single(&lp->pci_dev->dev, skb->data, skb->len,
+				       DMA_TO_DEVICE);
+		if (dma_mapping_error(&lp->pci_dev->dev, lp->tx_dma_addr[x])) {
 			netif_printk(lp, hw, KERN_DEBUG, dev,
 				     "DMA mapping error at line: %d!\n",
 				     __LINE__);
@@ -1133,6 +1125,7 @@ clean_up:
 		lp->a->write_bcr(ioaddr, 20, 4);	/* return to 16bit mode */
 	}
 	spin_unlock_irqrestore(&lp->lock, flags);
+	netdev_unlock(dev);
 
 	return rc;
 }				/* end pcnet32_loopback_test  */
@@ -1234,21 +1227,21 @@ static void pcnet32_rx_entry(struct net_device *dev,
 		 */
 		if (newskb) {
 			skb_reserve(newskb, NET_IP_ALIGN);
-			new_dma_addr = pci_map_single(lp->pci_dev,
+			new_dma_addr = dma_map_single(&lp->pci_dev->dev,
 						      newskb->data,
 						      PKT_BUF_SIZE,
-						      PCI_DMA_FROMDEVICE);
-			if (pci_dma_mapping_error(lp->pci_dev, new_dma_addr)) {
+						      DMA_FROM_DEVICE);
+			if (dma_mapping_error(&lp->pci_dev->dev, new_dma_addr)) {
 				netif_err(lp, rx_err, dev,
 					  "DMA mapping error.\n");
 				dev_kfree_skb(newskb);
 				skb = NULL;
 			} else {
 				skb = lp->rx_skbuff[entry];
-				pci_unmap_single(lp->pci_dev,
+				dma_unmap_single(&lp->pci_dev->dev,
 						 lp->rx_dma_addr[entry],
 						 PKT_BUF_SIZE,
-						 PCI_DMA_FROMDEVICE);
+						 DMA_FROM_DEVICE);
 				skb_put(skb, pkt_len);
 				lp->rx_skbuff[entry] = newskb;
 				lp->rx_dma_addr[entry] = new_dma_addr;
@@ -1260,24 +1253,22 @@ static void pcnet32_rx_entry(struct net_device *dev,
 	} else
 		skb = netdev_alloc_skb(dev, pkt_len + NET_IP_ALIGN);
 
-	if (skb == NULL) {
+	if (!skb) {
 		dev->stats.rx_dropped++;
 		return;
 	}
 	if (!rx_in_place) {
 		skb_reserve(skb, NET_IP_ALIGN);
 		skb_put(skb, pkt_len);	/* Make room */
-		pci_dma_sync_single_for_cpu(lp->pci_dev,
-					    lp->rx_dma_addr[entry],
-					    pkt_len,
-					    PCI_DMA_FROMDEVICE);
+		dma_sync_single_for_cpu(&lp->pci_dev->dev,
+					lp->rx_dma_addr[entry], pkt_len,
+					DMA_FROM_DEVICE);
 		skb_copy_to_linear_data(skb,
 				 (unsigned char *)(lp->rx_skbuff[entry]->data),
 				 pkt_len);
-		pci_dma_sync_single_for_device(lp->pci_dev,
-					       lp->rx_dma_addr[entry],
-					       pkt_len,
-					       PCI_DMA_FROMDEVICE);
+		dma_sync_single_for_device(&lp->pci_dev->dev,
+					   lp->rx_dma_addr[entry], pkt_len,
+					   DMA_FROM_DEVICE);
 	}
 	dev->stats.rx_bytes += skb->len;
 	skb->protocol = eth_type_trans(skb, dev);
@@ -1366,10 +1357,10 @@ static int pcnet32_tx(struct net_device *dev)
 
 		/* We must free the original skb */
 		if (lp->tx_skbuff[entry]) {
-			pci_unmap_single(lp->pci_dev,
+			dma_unmap_single(&lp->pci_dev->dev,
 					 lp->tx_dma_addr[entry],
-					 lp->tx_skbuff[entry]->
-					 len, PCI_DMA_TODEVICE);
+					 lp->tx_skbuff[entry]->len,
+					 DMA_TO_DEVICE);
 			dev_kfree_skb_any(lp->tx_skbuff[entry]);
 			lp->tx_skbuff[entry] = NULL;
 			lp->tx_dma_addr[entry] = 0;
@@ -1551,26 +1542,31 @@ pcnet32_probe_pci(struct pci_dev *pdev, const struct pci_device_id *ent)
 	}
 	pci_set_master(pdev);
 
-	ioaddr = pci_resource_start(pdev, 0);
-	if (!ioaddr) {
+	if (!pci_resource_len(pdev, 0)) {
 		if (pcnet32_debug & NETIF_MSG_PROBE)
 			pr_err("card has no PCI IO resources, aborting\n");
-		return -ENODEV;
+		err = -ENODEV;
+		goto err_disable_dev;
 	}
 
-	err = pci_set_dma_mask(pdev, PCNET32_DMA_MASK);
+	err = dma_set_mask(&pdev->dev, PCNET32_DMA_MASK);
 	if (err) {
 		if (pcnet32_debug & NETIF_MSG_PROBE)
 			pr_err("architecture does not support 32bit PCI busmaster DMA\n");
-		return err;
+		goto err_disable_dev;
 	}
+
+	ioaddr = pci_resource_start(pdev, 0);
 	if (!request_region(ioaddr, PCNET32_TOTAL_SIZE, "pcnet32_probe_pci")) {
 		if (pcnet32_debug & NETIF_MSG_PROBE)
 			pr_err("io address range already allocated\n");
-		return -EBUSY;
+		err = -EBUSY;
+		goto err_disable_dev;
 	}
 
 	err = pcnet32_probe1(ioaddr, 1, pdev);
+
+err_disable_dev:
 	if (err < 0)
 		pci_disable_device(pdev);
 
@@ -1584,7 +1580,7 @@ static const struct net_device_ops pcnet32_netdev_ops = {
 	.ndo_tx_timeout		= pcnet32_tx_timeout,
 	.ndo_get_stats		= pcnet32_get_stats,
 	.ndo_set_rx_mode	= pcnet32_set_multicast_list,
-	.ndo_do_ioctl		= pcnet32_ioctl,
+	.ndo_eth_ioctl		= pcnet32_ioctl,
 	.ndo_set_mac_address 	= eth_mac_addr,
 	.ndo_validate_addr	= eth_validate_addr,
 #ifdef CONFIG_NET_POLL_CONTROLLER
@@ -1607,6 +1603,7 @@ pcnet32_probe1(unsigned long ioaddr, int shared, struct pci_dev *pdev)
 	struct net_device *dev;
 	const struct pcnet32_access *a = NULL;
 	u8 promaddr[ETH_ALEN];
+	u8 addr[ETH_ALEN];
 	int ret = -ENODEV;
 
 	/* reset the chip */
@@ -1772,9 +1769,10 @@ pcnet32_probe1(unsigned long ioaddr, int shared, struct pci_dev *pdev)
 		unsigned int val;
 		val = a->read_csr(ioaddr, i + 12) & 0x0ffff;
 		/* There may be endianness issues here. */
-		dev->dev_addr[2 * i] = val & 0x0ff;
-		dev->dev_addr[2 * i + 1] = (val >> 8) & 0x0ff;
+		addr[2 * i] = val & 0x0ff;
+		addr[2 * i + 1] = (val >> 8) & 0x0ff;
 	}
+	eth_hw_addr_set(dev, addr);
 
 	/* read PROM address and compare with CSR address */
 	for (i = 0; i < ETH_ALEN; i++)
@@ -1787,13 +1785,16 @@ pcnet32_probe1(unsigned long ioaddr, int shared, struct pci_dev *pdev)
 				pr_cont(" warning: CSR address invalid,\n");
 				pr_info("    using instead PROM address of");
 			}
-			memcpy(dev->dev_addr, promaddr, ETH_ALEN);
+			eth_hw_addr_set(dev, promaddr);
 		}
 	}
 
 	/* if the ethernet address is not valid, force to 00:00:00:00:00:00 */
-	if (!is_valid_ether_addr(dev->dev_addr))
-		eth_zero_addr(dev->dev_addr);
+	if (!is_valid_ether_addr(dev->dev_addr)) {
+		static const u8 zero_addr[ETH_ALEN] = {};
+
+		eth_hw_addr_set(dev, zero_addr);
+	}
 
 	if (pcnet32_debug & NETIF_MSG_PROBE) {
 		pr_cont(" %pM", dev->dev_addr);
@@ -1838,12 +1839,13 @@ pcnet32_probe1(unsigned long ioaddr, int shared, struct pci_dev *pdev)
 
 	dev->base_addr = ioaddr;
 	lp = netdev_priv(dev);
-	/* pci_alloc_consistent returns page-aligned memory, so we do not have to check the alignment */
-	lp->init_block = pci_alloc_consistent(pdev, sizeof(*lp->init_block),
-					      &lp->init_dma_addr);
+	/* dma_alloc_coherent returns page-aligned memory, so we do not have to check the alignment */
+	lp->init_block = dma_alloc_coherent(&pdev->dev,
+					    sizeof(*lp->init_block),
+					    &lp->init_dma_addr, GFP_KERNEL);
 	if (!lp->init_block) {
 		if (pcnet32_debug & NETIF_MSG_PROBE)
-			pr_err("Consistent memory allocation failed\n");
+			pr_err("Coherent memory allocation failed\n");
 		ret = -ENOMEM;
 		goto err_free_netdev;
 	}
@@ -1883,7 +1885,8 @@ pcnet32_probe1(unsigned long ioaddr, int shared, struct pci_dev *pdev)
 	/* napi.weight is used in both the napi and non-napi cases */
 	lp->napi.weight = lp->rx_ring_size / 2;
 
-	netif_napi_add(dev, &lp->napi, pcnet32_poll, lp->rx_ring_size / 2);
+	netif_napi_add_weight(dev, &lp->napi, pcnet32_poll,
+			      lp->rx_ring_size / 2);
 
 	if (fdx && !(lp->options & PCNET32_PORT_ASEL) &&
 	    ((cards_found >= MAX_UNITS) || full_duplex[cards_found]))
@@ -1973,9 +1976,7 @@ pcnet32_probe1(unsigned long ioaddr, int shared, struct pci_dev *pdev)
 			lp->options |= PCNET32_PORT_MII;
 	}
 
-	init_timer(&lp->watchdog_timer);
-	lp->watchdog_timer.data = (unsigned long)dev;
-	lp->watchdog_timer.function = (void *)&pcnet32_watchdog;
+	timer_setup(&lp->watchdog_timer, pcnet32_watchdog, 0);
 
 	/* The PCNET32-specific entries in the device structure. */
 	dev->netdev_ops = &pcnet32_netdev_ops;
@@ -2004,8 +2005,8 @@ pcnet32_probe1(unsigned long ioaddr, int shared, struct pci_dev *pdev)
 
 err_free_ring:
 	pcnet32_free_ring(dev);
-	pci_free_consistent(lp->pci_dev, sizeof(*lp->init_block),
-			    lp->init_block, lp->init_dma_addr);
+	dma_free_coherent(&lp->pci_dev->dev, sizeof(*lp->init_block),
+			  lp->init_block, lp->init_dma_addr);
 err_free_netdev:
 	free_netdev(dev);
 err_release_region:
@@ -2018,41 +2019,39 @@ static int pcnet32_alloc_ring(struct net_device *dev, const char *name)
 {
 	struct pcnet32_private *lp = netdev_priv(dev);
 
-	lp->tx_ring = pci_alloc_consistent(lp->pci_dev,
-					   sizeof(struct pcnet32_tx_head) *
-					   lp->tx_ring_size,
-					   &lp->tx_ring_dma_addr);
-	if (lp->tx_ring == NULL) {
-		netif_err(lp, drv, dev, "Consistent memory allocation failed\n");
+	lp->tx_ring = dma_alloc_coherent(&lp->pci_dev->dev,
+					 sizeof(struct pcnet32_tx_head) * lp->tx_ring_size,
+					 &lp->tx_ring_dma_addr, GFP_KERNEL);
+	if (!lp->tx_ring) {
+		netif_err(lp, drv, dev, "Coherent memory allocation failed\n");
 		return -ENOMEM;
 	}
 
-	lp->rx_ring = pci_alloc_consistent(lp->pci_dev,
-					   sizeof(struct pcnet32_rx_head) *
-					   lp->rx_ring_size,
-					   &lp->rx_ring_dma_addr);
-	if (lp->rx_ring == NULL) {
-		netif_err(lp, drv, dev, "Consistent memory allocation failed\n");
+	lp->rx_ring = dma_alloc_coherent(&lp->pci_dev->dev,
+					 sizeof(struct pcnet32_rx_head) * lp->rx_ring_size,
+					 &lp->rx_ring_dma_addr, GFP_KERNEL);
+	if (!lp->rx_ring) {
+		netif_err(lp, drv, dev, "Coherent memory allocation failed\n");
 		return -ENOMEM;
 	}
 
 	lp->tx_dma_addr = kcalloc(lp->tx_ring_size, sizeof(dma_addr_t),
-				  GFP_ATOMIC);
+				  GFP_KERNEL);
 	if (!lp->tx_dma_addr)
 		return -ENOMEM;
 
 	lp->rx_dma_addr = kcalloc(lp->rx_ring_size, sizeof(dma_addr_t),
-				  GFP_ATOMIC);
+				  GFP_KERNEL);
 	if (!lp->rx_dma_addr)
 		return -ENOMEM;
 
 	lp->tx_skbuff = kcalloc(lp->tx_ring_size, sizeof(struct sk_buff *),
-				GFP_ATOMIC);
+				GFP_KERNEL);
 	if (!lp->tx_skbuff)
 		return -ENOMEM;
 
 	lp->rx_skbuff = kcalloc(lp->rx_ring_size, sizeof(struct sk_buff *),
-				GFP_ATOMIC);
+				GFP_KERNEL);
 	if (!lp->rx_skbuff)
 		return -ENOMEM;
 
@@ -2076,18 +2075,16 @@ static void pcnet32_free_ring(struct net_device *dev)
 	lp->rx_dma_addr = NULL;
 
 	if (lp->tx_ring) {
-		pci_free_consistent(lp->pci_dev,
-				    sizeof(struct pcnet32_tx_head) *
-				    lp->tx_ring_size, lp->tx_ring,
-				    lp->tx_ring_dma_addr);
+		dma_free_coherent(&lp->pci_dev->dev,
+				  sizeof(struct pcnet32_tx_head) * lp->tx_ring_size,
+				  lp->tx_ring, lp->tx_ring_dma_addr);
 		lp->tx_ring = NULL;
 	}
 
 	if (lp->rx_ring) {
-		pci_free_consistent(lp->pci_dev,
-				    sizeof(struct pcnet32_rx_head) *
-				    lp->rx_ring_size, lp->rx_ring,
-				    lp->rx_ring_dma_addr);
+		dma_free_coherent(&lp->pci_dev->dev,
+				  sizeof(struct pcnet32_rx_head) * lp->rx_ring_size,
+				  lp->rx_ring, lp->rx_ring_dma_addr);
 		lp->rx_ring = NULL;
 	}
 }
@@ -2108,6 +2105,7 @@ static int pcnet32_open(struct net_device *dev)
 		return -EAGAIN;
 	}
 
+	netdev_lock(dev);
 	spin_lock_irqsave(&lp->lock, flags);
 	/* Check for a valid station address */
 	if (!is_valid_ether_addr(dev->dev_addr)) {
@@ -2273,7 +2271,7 @@ static int pcnet32_open(struct net_device *dev)
 		goto err_free_ring;
 	}
 
-	napi_enable(&lp->napi);
+	napi_enable_locked(&lp->napi);
 
 	/* Re-initialize the PCNET32, and start it when done. */
 	lp->a->write_csr(ioaddr, 1, (lp->init_dma_addr & 0xffff));
@@ -2307,6 +2305,7 @@ static int pcnet32_open(struct net_device *dev)
 		     lp->a->read_csr(ioaddr, CSR0));
 
 	spin_unlock_irqrestore(&lp->lock, flags);
+	netdev_unlock(dev);
 
 	return 0;		/* Always succeed */
 
@@ -2322,6 +2321,7 @@ err_free_ring:
 
 err_free_irq:
 	spin_unlock_irqrestore(&lp->lock, flags);
+	netdev_unlock(dev);
 	free_irq(dev->irq, dev);
 	return rc;
 }
@@ -2348,12 +2348,11 @@ static void pcnet32_purge_tx_ring(struct net_device *dev)
 		lp->tx_ring[i].status = 0;	/* CPU owns buffer */
 		wmb();		/* Make sure adapter sees owner change */
 		if (lp->tx_skbuff[i]) {
-			if (!pci_dma_mapping_error(lp->pci_dev,
-						   lp->tx_dma_addr[i]))
-				pci_unmap_single(lp->pci_dev,
+			if (!dma_mapping_error(&lp->pci_dev->dev, lp->tx_dma_addr[i]))
+				dma_unmap_single(&lp->pci_dev->dev,
 						 lp->tx_dma_addr[i],
 						 lp->tx_skbuff[i]->len,
-						 PCI_DMA_TODEVICE);
+						 DMA_TO_DEVICE);
 			dev_kfree_skb_any(lp->tx_skbuff[i]);
 		}
 		lp->tx_skbuff[i] = NULL;
@@ -2373,7 +2372,7 @@ static int pcnet32_init_ring(struct net_device *dev)
 
 	for (i = 0; i < lp->rx_ring_size; i++) {
 		struct sk_buff *rx_skbuff = lp->rx_skbuff[i];
-		if (rx_skbuff == NULL) {
+		if (!rx_skbuff) {
 			lp->rx_skbuff[i] = netdev_alloc_skb(dev, PKT_BUF_SKB);
 			rx_skbuff = lp->rx_skbuff[i];
 			if (!rx_skbuff) {
@@ -2388,10 +2387,9 @@ static int pcnet32_init_ring(struct net_device *dev)
 		rmb();
 		if (lp->rx_dma_addr[i] == 0) {
 			lp->rx_dma_addr[i] =
-			    pci_map_single(lp->pci_dev, rx_skbuff->data,
-					   PKT_BUF_SIZE, PCI_DMA_FROMDEVICE);
-			if (pci_dma_mapping_error(lp->pci_dev,
-						  lp->rx_dma_addr[i])) {
+			    dma_map_single(&lp->pci_dev->dev, rx_skbuff->data,
+					   PKT_BUF_SIZE, DMA_FROM_DEVICE);
+			if (dma_mapping_error(&lp->pci_dev->dev, lp->rx_dma_addr[i])) {
 				/* there is not much we can do at this point */
 				netif_err(lp, drv, dev,
 					  "%s pci dma mapping error\n",
@@ -2456,7 +2454,7 @@ static void pcnet32_restart(struct net_device *dev, unsigned int csr0_bits)
 	lp->a->write_csr(ioaddr, CSR0, csr0_bits);
 }
 
-static void pcnet32_tx_timeout(struct net_device *dev)
+static void pcnet32_tx_timeout(struct net_device *dev, unsigned int txqueue)
 {
 	struct pcnet32_private *lp = netdev_priv(dev);
 	unsigned long ioaddr = dev->base_addr, flags;
@@ -2529,8 +2527,9 @@ static netdev_tx_t pcnet32_start_xmit(struct sk_buff *skb,
 	lp->tx_ring[entry].misc = 0x00000000;
 
 	lp->tx_dma_addr[entry] =
-	    pci_map_single(lp->pci_dev, skb->data, skb->len, PCI_DMA_TODEVICE);
-	if (pci_dma_mapping_error(lp->pci_dev, lp->tx_dma_addr[entry])) {
+	    dma_map_single(&lp->pci_dev->dev, skb->data, skb->len,
+			   DMA_TO_DEVICE);
+	if (dma_mapping_error(&lp->pci_dev->dev, lp->tx_dma_addr[entry])) {
 		dev_kfree_skb_any(skb);
 		dev->stats.tx_dropped++;
 		goto drop_packet;
@@ -2631,7 +2630,7 @@ static int pcnet32_close(struct net_device *dev)
 	struct pcnet32_private *lp = netdev_priv(dev);
 	unsigned long flags;
 
-	del_timer_sync(&lp->watchdog_timer);
+	timer_delete_sync(&lp->watchdog_timer);
 
 	netif_stop_queue(dev);
 	napi_disable(&lp->napi);
@@ -2871,8 +2870,7 @@ static void pcnet32_check_media(struct net_device *dev, int verbose)
 			netif_info(lp, link, dev, "link down\n");
 		}
 		if (lp->phycount > 1) {
-			curr_link = pcnet32_check_otherphy(dev);
-			prev_link = 0;
+			pcnet32_check_otherphy(dev);
 		}
 	} else if (verbose || !prev_link) {
 		netif_carrier_on(dev);
@@ -2905,9 +2903,10 @@ static void pcnet32_check_media(struct net_device *dev, int verbose)
  * Could possibly be changed to use mii_check_media instead.
  */
 
-static void pcnet32_watchdog(struct net_device *dev)
+static void pcnet32_watchdog(struct timer_list *t)
 {
-	struct pcnet32_private *lp = netdev_priv(dev);
+	struct pcnet32_private *lp = timer_container_of(lp, t, watchdog_timer);
+	struct net_device *dev = lp->dev;
 	unsigned long flags;
 
 	/* Print the link status if it has changed */
@@ -2918,30 +2917,27 @@ static void pcnet32_watchdog(struct net_device *dev)
 	mod_timer(&lp->watchdog_timer, round_jiffies(PCNET32_WATCHDOG_TIMEOUT));
 }
 
-static int pcnet32_pm_suspend(struct pci_dev *pdev, pm_message_t state)
+static int __maybe_unused pcnet32_pm_suspend(struct device *device_d)
 {
-	struct net_device *dev = pci_get_drvdata(pdev);
+	struct net_device *dev = dev_get_drvdata(device_d);
 
 	if (netif_running(dev)) {
 		netif_device_detach(dev);
 		pcnet32_close(dev);
 	}
-	pci_save_state(pdev);
-	pci_set_power_state(pdev, pci_choose_state(pdev, state));
+
 	return 0;
 }
 
-static int pcnet32_pm_resume(struct pci_dev *pdev)
+static int __maybe_unused pcnet32_pm_resume(struct device *device_d)
 {
-	struct net_device *dev = pci_get_drvdata(pdev);
-
-	pci_set_power_state(pdev, PCI_D0);
-	pci_restore_state(pdev);
+	struct net_device *dev = dev_get_drvdata(device_d);
 
 	if (netif_running(dev)) {
 		pcnet32_open(dev);
 		netif_device_attach(dev);
 	}
+
 	return 0;
 }
 
@@ -2955,20 +2951,23 @@ static void pcnet32_remove_one(struct pci_dev *pdev)
 		unregister_netdev(dev);
 		pcnet32_free_ring(dev);
 		release_region(dev->base_addr, PCNET32_TOTAL_SIZE);
-		pci_free_consistent(lp->pci_dev, sizeof(*lp->init_block),
-				    lp->init_block, lp->init_dma_addr);
+		dma_free_coherent(&lp->pci_dev->dev, sizeof(*lp->init_block),
+				  lp->init_block, lp->init_dma_addr);
 		free_netdev(dev);
 		pci_disable_device(pdev);
 	}
 }
 
+static SIMPLE_DEV_PM_OPS(pcnet32_pm_ops, pcnet32_pm_suspend, pcnet32_pm_resume);
+
 static struct pci_driver pcnet32_driver = {
 	.name = DRV_NAME,
 	.probe = pcnet32_probe_pci,
 	.remove = pcnet32_remove_one,
 	.id_table = pcnet32_pci_tbl,
-	.suspend = pcnet32_pm_suspend,
-	.resume = pcnet32_pm_resume,
+	.driver = {
+		.pm = &pcnet32_pm_ops,
+	},
 };
 
 /* An additional parameter that may be passed in... */
@@ -3006,8 +3005,6 @@ MODULE_LICENSE("GPL");
 
 static int __init pcnet32_init_module(void)
 {
-	pr_info("%s", version);
-
 	pcnet32_debug = netif_msg_init(debug, PCNET32_MSG_DEFAULT);
 
 	if ((tx_start_pt >= 0) && (tx_start_pt <= 3))
@@ -3037,8 +3034,8 @@ static void __exit pcnet32_cleanup_module(void)
 		unregister_netdev(pcnet32_dev);
 		pcnet32_free_ring(pcnet32_dev);
 		release_region(pcnet32_dev->base_addr, PCNET32_TOTAL_SIZE);
-		pci_free_consistent(lp->pci_dev, sizeof(*lp->init_block),
-				    lp->init_block, lp->init_dma_addr);
+		dma_free_coherent(&lp->pci_dev->dev, sizeof(*lp->init_block),
+				  lp->init_block, lp->init_dma_addr);
 		free_netdev(pcnet32_dev);
 		pcnet32_dev = next_dev;
 	}
@@ -3049,10 +3046,3 @@ static void __exit pcnet32_cleanup_module(void)
 
 module_init(pcnet32_init_module);
 module_exit(pcnet32_cleanup_module);
-
-/*
- * Local variables:
- *  c-indent-level: 4
- *  tab-width: 8
- * End:
- */
diff --git a/drivers/net/ethernet/amd/pds_core/Makefile b/drivers/net/ethernet/amd/pds_core/Makefile
new file mode 100644
index 000000000000..8239742e681f
--- /dev/null
+++ b/drivers/net/ethernet/amd/pds_core/Makefile
@@ -0,0 +1,13 @@
+# SPDX-License-Identifier: GPL-2.0
+# Copyright (c) 2023 Advanced Micro Devices, Inc.
+
+obj-$(CONFIG_PDS_CORE) := pds_core.o
+
+pds_core-y := main.o \
+	      devlink.o \
+	      auxbus.o \
+	      dev.o \
+	      adminq.o \
+	      core.o \
+	      debugfs.o \
+	      fw.o
diff --git a/drivers/net/ethernet/amd/pds_core/adminq.c b/drivers/net/ethernet/amd/pds_core/adminq.c
new file mode 100644
index 000000000000..097bb092bdb8
--- /dev/null
+++ b/drivers/net/ethernet/amd/pds_core/adminq.c
@@ -0,0 +1,310 @@
+// SPDX-License-Identifier: GPL-2.0
+/* Copyright(c) 2023 Advanced Micro Devices, Inc */
+
+#include <linux/dynamic_debug.h>
+
+#include "core.h"
+
+static int pdsc_process_notifyq(struct pdsc_qcq *qcq)
+{
+	union pds_core_notifyq_comp *comp;
+	struct pdsc *pdsc = qcq->pdsc;
+	struct pdsc_cq *cq = &qcq->cq;
+	struct pdsc_cq_info *cq_info;
+	int nq_work = 0;
+	u64 eid;
+
+	cq_info = &cq->info[cq->tail_idx];
+	comp = cq_info->comp;
+	eid = le64_to_cpu(comp->event.eid);
+	while (eid > pdsc->last_eid) {
+		u16 ecode = le16_to_cpu(comp->event.ecode);
+
+		switch (ecode) {
+		case PDS_EVENT_LINK_CHANGE:
+			dev_info(pdsc->dev, "NotifyQ LINK_CHANGE ecode %d eid %lld\n",
+				 ecode, eid);
+			pdsc_notify(PDS_EVENT_LINK_CHANGE, comp);
+			break;
+
+		case PDS_EVENT_RESET:
+			dev_info(pdsc->dev, "NotifyQ RESET ecode %d eid %lld\n",
+				 ecode, eid);
+			pdsc_notify(PDS_EVENT_RESET, comp);
+			break;
+
+		case PDS_EVENT_XCVR:
+			dev_info(pdsc->dev, "NotifyQ XCVR ecode %d eid %lld\n",
+				 ecode, eid);
+			break;
+
+		default:
+			dev_info(pdsc->dev, "NotifyQ ecode %d eid %lld\n",
+				 ecode, eid);
+			break;
+		}
+
+		pdsc->last_eid = eid;
+		cq->tail_idx = (cq->tail_idx + 1) & (cq->num_descs - 1);
+		cq_info = &cq->info[cq->tail_idx];
+		comp = cq_info->comp;
+		eid = le64_to_cpu(comp->event.eid);
+
+		nq_work++;
+	}
+
+	qcq->accum_work += nq_work;
+
+	return nq_work;
+}
+
+static bool pdsc_adminq_inc_if_up(struct pdsc *pdsc)
+{
+	if (pdsc->state & BIT_ULL(PDSC_S_STOPPING_DRIVER) ||
+	    pdsc->state & BIT_ULL(PDSC_S_FW_DEAD))
+		return false;
+
+	return refcount_inc_not_zero(&pdsc->adminq_refcnt);
+}
+
+void pdsc_process_adminq(struct pdsc_qcq *qcq)
+{
+	union pds_core_adminq_comp *comp;
+	struct pdsc_queue *q = &qcq->q;
+	struct pdsc *pdsc = qcq->pdsc;
+	struct pdsc_cq *cq = &qcq->cq;
+	struct pdsc_q_info *q_info;
+	unsigned long irqflags;
+	int nq_work = 0;
+	int aq_work = 0;
+
+	/* Don't process AdminQ when it's not up */
+	if (!pdsc_adminq_inc_if_up(pdsc)) {
+		dev_err(pdsc->dev, "%s: called while adminq is unavailable\n",
+			__func__);
+		return;
+	}
+
+	/* Check for NotifyQ event */
+	nq_work = pdsc_process_notifyq(&pdsc->notifyqcq);
+
+	/* Check for empty queue, which can happen if the interrupt was
+	 * for a NotifyQ event and there are no new AdminQ completions.
+	 */
+	if (q->tail_idx == q->head_idx)
+		goto credits;
+
+	/* Find the first completion to clean,
+	 * run the callback in the related q_info,
+	 * and continue while we still match done color
+	 */
+	spin_lock_irqsave(&pdsc->adminq_lock, irqflags);
+	comp = cq->info[cq->tail_idx].comp;
+	while (pdsc_color_match(comp->color, cq->done_color)) {
+		q_info = &q->info[q->tail_idx];
+		q->tail_idx = (q->tail_idx + 1) & (q->num_descs - 1);
+
+		if (!completion_done(&q_info->completion)) {
+			memcpy(q_info->dest, comp, sizeof(*comp));
+			complete(&q_info->completion);
+		}
+
+		if (cq->tail_idx == cq->num_descs - 1)
+			cq->done_color = !cq->done_color;
+		cq->tail_idx = (cq->tail_idx + 1) & (cq->num_descs - 1);
+		comp = cq->info[cq->tail_idx].comp;
+
+		aq_work++;
+	}
+	spin_unlock_irqrestore(&pdsc->adminq_lock, irqflags);
+
+	qcq->accum_work += aq_work;
+
+credits:
+	/* Return the interrupt credits, one for each completion */
+	pds_core_intr_credits(&pdsc->intr_ctrl[qcq->intx],
+			      nq_work + aq_work,
+			      PDS_CORE_INTR_CRED_REARM);
+	refcount_dec(&pdsc->adminq_refcnt);
+}
+
+void pdsc_work_thread(struct work_struct *work)
+{
+	struct pdsc_qcq *qcq = container_of(work, struct pdsc_qcq, work);
+
+	pdsc_process_adminq(qcq);
+}
+
+irqreturn_t pdsc_adminq_isr(int irq, void *data)
+{
+	struct pdsc *pdsc = data;
+	struct pdsc_qcq *qcq;
+
+	/* Don't process AdminQ when it's not up */
+	if (!pdsc_adminq_inc_if_up(pdsc)) {
+		dev_err(pdsc->dev, "%s: called while adminq is unavailable\n",
+			__func__);
+		return IRQ_HANDLED;
+	}
+
+	qcq = &pdsc->adminqcq;
+	queue_work(pdsc->wq, &qcq->work);
+	refcount_dec(&pdsc->adminq_refcnt);
+
+	return IRQ_HANDLED;
+}
+
+static int __pdsc_adminq_post(struct pdsc *pdsc,
+			      struct pdsc_qcq *qcq,
+			      union pds_core_adminq_cmd *cmd,
+			      union pds_core_adminq_comp *comp)
+{
+	struct pdsc_queue *q = &qcq->q;
+	struct pdsc_q_info *q_info;
+	unsigned long irqflags;
+	unsigned int avail;
+	int index;
+	int ret;
+
+	spin_lock_irqsave(&pdsc->adminq_lock, irqflags);
+
+	/* Check for space in the queue */
+	avail = q->tail_idx;
+	if (q->head_idx >= avail)
+		avail += q->num_descs - q->head_idx - 1;
+	else
+		avail -= q->head_idx + 1;
+	if (!avail) {
+		ret = -ENOSPC;
+		goto err_out_unlock;
+	}
+
+	/* Check that the FW is running */
+	if (!pdsc_is_fw_running(pdsc)) {
+		if (pdsc->info_regs) {
+			u8 fw_status =
+				ioread8(&pdsc->info_regs->fw_status);
+
+			dev_info(pdsc->dev, "%s: post failed - fw not running %#02x:\n",
+				 __func__, fw_status);
+		} else {
+			dev_info(pdsc->dev, "%s: post failed - BARs not setup\n",
+				 __func__);
+		}
+		ret = -ENXIO;
+
+		goto err_out_unlock;
+	}
+
+	/* Post the request */
+	index = q->head_idx;
+	q_info = &q->info[index];
+	q_info->dest = comp;
+	memcpy(q_info->desc, cmd, sizeof(*cmd));
+	reinit_completion(&q_info->completion);
+
+	dev_dbg(pdsc->dev, "head_idx %d tail_idx %d\n",
+		q->head_idx, q->tail_idx);
+	dev_dbg(pdsc->dev, "post admin queue command:\n");
+	dynamic_hex_dump("cmd ", DUMP_PREFIX_OFFSET, 16, 1,
+			 cmd, sizeof(*cmd), true);
+
+	q->head_idx = (q->head_idx + 1) & (q->num_descs - 1);
+
+	pds_core_dbell_ring(pdsc->kern_dbpage,
+			    q->hw_type, q->dbval | q->head_idx);
+	ret = index;
+
+err_out_unlock:
+	spin_unlock_irqrestore(&pdsc->adminq_lock, irqflags);
+	return ret;
+}
+
+int pdsc_adminq_post(struct pdsc *pdsc,
+		     union pds_core_adminq_cmd *cmd,
+		     union pds_core_adminq_comp *comp,
+		     bool fast_poll)
+{
+	unsigned long poll_interval = 200;
+	unsigned long poll_jiffies;
+	unsigned long time_limit;
+	unsigned long time_start;
+	unsigned long time_done;
+	unsigned long remaining;
+	struct completion *wc;
+	int err = 0;
+	int index;
+
+	if (!pdsc_adminq_inc_if_up(pdsc)) {
+		dev_dbg(pdsc->dev, "%s: preventing adminq cmd %u\n",
+			__func__, cmd->opcode);
+		return -ENXIO;
+	}
+
+	index = __pdsc_adminq_post(pdsc, &pdsc->adminqcq, cmd, comp);
+	if (index < 0) {
+		err = index;
+		goto err_out;
+	}
+
+	wc = &pdsc->adminqcq.q.info[index].completion;
+	time_start = jiffies;
+	time_limit = time_start + HZ * pdsc->devcmd_timeout;
+	do {
+		/* Timeslice the actual wait to catch IO errors etc early */
+		poll_jiffies = usecs_to_jiffies(poll_interval);
+		remaining = wait_for_completion_timeout(wc, poll_jiffies);
+		if (remaining)
+			break;
+
+		if (!pdsc_is_fw_running(pdsc)) {
+			if (pdsc->info_regs) {
+				u8 fw_status =
+					ioread8(&pdsc->info_regs->fw_status);
+
+				dev_dbg(pdsc->dev, "%s: post wait failed - fw not running %#02x:\n",
+					__func__, fw_status);
+			} else {
+				dev_dbg(pdsc->dev, "%s: post wait failed - BARs not setup\n",
+					__func__);
+			}
+			err = -ENXIO;
+			break;
+		}
+
+		/* When fast_poll is not requested, prevent aggressive polling
+		 * on failures due to timeouts by doing exponential back off.
+		 */
+		if (!fast_poll && poll_interval < PDSC_ADMINQ_MAX_POLL_INTERVAL)
+			poll_interval <<= 1;
+	} while (time_before(jiffies, time_limit));
+	time_done = jiffies;
+	dev_dbg(pdsc->dev, "%s: elapsed %d msecs\n",
+		__func__, jiffies_to_msecs(time_done - time_start));
+
+	/* Check the results and clear an un-completed timeout */
+	if (time_after_eq(time_done, time_limit) && !completion_done(wc)) {
+		err = -ETIMEDOUT;
+		complete(wc);
+	}
+
+	dev_dbg(pdsc->dev, "read admin queue completion idx %d:\n", index);
+	dynamic_hex_dump("comp ", DUMP_PREFIX_OFFSET, 16, 1,
+			 comp, sizeof(*comp), true);
+
+	if (remaining && comp->status)
+		err = pdsc_err_to_errno(comp->status);
+
+err_out:
+	if (err) {
+		dev_dbg(pdsc->dev, "%s: opcode %d status %d err %pe\n",
+			__func__, cmd->opcode, comp->status, ERR_PTR(err));
+		if (err == -ENXIO || err == -ETIMEDOUT)
+			queue_work(pdsc->wq, &pdsc->health_work);
+	}
+
+	refcount_dec(&pdsc->adminq_refcnt);
+
+	return err;
+}
+EXPORT_SYMBOL_GPL(pdsc_adminq_post);
diff --git a/drivers/net/ethernet/amd/pds_core/auxbus.c b/drivers/net/ethernet/amd/pds_core/auxbus.c
new file mode 100644
index 000000000000..92f359f2b449
--- /dev/null
+++ b/drivers/net/ethernet/amd/pds_core/auxbus.c
@@ -0,0 +1,256 @@
+// SPDX-License-Identifier: GPL-2.0
+/* Copyright(c) 2023 Advanced Micro Devices, Inc */
+
+#include <linux/pci.h>
+
+#include "core.h"
+#include <linux/pds/pds_auxbus.h>
+
+/**
+ * pds_client_register - Link the client to the firmware
+ * @pf:		ptr to the PF driver's private data struct
+ * @devname:	name that includes service into, e.g. pds_core.vDPA
+ *
+ * Return: positive client ID (ci) on success, or
+ *         negative for error
+ */
+int pds_client_register(struct pdsc *pf, char *devname)
+{
+	union pds_core_adminq_comp comp = {};
+	union pds_core_adminq_cmd cmd = {};
+	int err;
+	u16 ci;
+
+	cmd.client_reg.opcode = PDS_AQ_CMD_CLIENT_REG;
+	strscpy(cmd.client_reg.devname, devname,
+		sizeof(cmd.client_reg.devname));
+
+	err = pdsc_adminq_post(pf, &cmd, &comp, false);
+	if (err) {
+		dev_info(pf->dev, "register dev_name %s with DSC failed, status %d: %pe\n",
+			 devname, comp.status, ERR_PTR(err));
+		return err;
+	}
+
+	ci = le16_to_cpu(comp.client_reg.client_id);
+	if (!ci) {
+		dev_err(pf->dev, "%s: device returned null client_id\n",
+			__func__);
+		return -EIO;
+	}
+
+	dev_dbg(pf->dev, "%s: device returned client_id %d for %s\n",
+		__func__, ci, devname);
+
+	return ci;
+}
+EXPORT_SYMBOL_GPL(pds_client_register);
+
+/**
+ * pds_client_unregister - Unlink the client from the firmware
+ * @pf:		ptr to the PF driver's private data struct
+ * @client_id:	id returned from pds_client_register()
+ *
+ * Return: 0 on success, or
+ *         negative for error
+ */
+int pds_client_unregister(struct pdsc *pf, u16 client_id)
+{
+	union pds_core_adminq_comp comp = {};
+	union pds_core_adminq_cmd cmd = {};
+	int err;
+
+	cmd.client_unreg.opcode = PDS_AQ_CMD_CLIENT_UNREG;
+	cmd.client_unreg.client_id = cpu_to_le16(client_id);
+
+	err = pdsc_adminq_post(pf, &cmd, &comp, false);
+	if (err)
+		dev_info(pf->dev, "unregister client_id %d failed, status %d: %pe\n",
+			 client_id, comp.status, ERR_PTR(err));
+
+	return err;
+}
+EXPORT_SYMBOL_GPL(pds_client_unregister);
+
+/**
+ * pds_client_adminq_cmd - Process an adminq request for the client
+ * @padev:   ptr to the client device
+ * @req:     ptr to buffer with request
+ * @req_len: length of actual struct used for request
+ * @resp:    ptr to buffer where answer is to be copied
+ * @flags:   optional flags from pds_core_adminq_flags
+ *
+ * Return: 0 on success, or
+ *         negative for error
+ *
+ * Client sends pointers to request and response buffers
+ * Core copies request data into pds_core_client_request_cmd
+ * Core sets other fields as needed
+ * Core posts to AdminQ
+ * Core copies completion data into response buffer
+ */
+int pds_client_adminq_cmd(struct pds_auxiliary_dev *padev,
+			  union pds_core_adminq_cmd *req,
+			  size_t req_len,
+			  union pds_core_adminq_comp *resp,
+			  u64 flags)
+{
+	union pds_core_adminq_cmd cmd = {};
+	struct pci_dev *pf_pdev;
+	struct pdsc *pf;
+	size_t cp_len;
+	int err;
+
+	pf_pdev = pci_physfn(padev->vf_pdev);
+	pf = pci_get_drvdata(pf_pdev);
+
+	dev_dbg(pf->dev, "%s: %s opcode %d\n",
+		__func__, dev_name(&padev->aux_dev.dev), req->opcode);
+
+	/* Wrap the client's request */
+	cmd.client_request.opcode = PDS_AQ_CMD_CLIENT_CMD;
+	cmd.client_request.client_id = cpu_to_le16(padev->client_id);
+	cp_len = min_t(size_t, req_len, sizeof(cmd.client_request.client_cmd));
+	memcpy(cmd.client_request.client_cmd, req, cp_len);
+
+	err = pdsc_adminq_post(pf, &cmd, resp,
+			       !!(flags & PDS_AQ_FLAG_FASTPOLL));
+	if (err && err != -EAGAIN)
+		dev_info(pf->dev, "client admin cmd failed: %pe\n",
+			 ERR_PTR(err));
+
+	return err;
+}
+EXPORT_SYMBOL_GPL(pds_client_adminq_cmd);
+
+static void pdsc_auxbus_dev_release(struct device *dev)
+{
+	struct pds_auxiliary_dev *padev =
+		container_of(dev, struct pds_auxiliary_dev, aux_dev.dev);
+
+	kfree(padev);
+}
+
+static struct pds_auxiliary_dev *pdsc_auxbus_dev_register(struct pdsc *cf,
+							  struct pdsc *pf,
+							  u16 client_id,
+							  char *name)
+{
+	struct auxiliary_device *aux_dev;
+	struct pds_auxiliary_dev *padev;
+	int err;
+
+	padev = kzalloc(sizeof(*padev), GFP_KERNEL);
+	if (!padev)
+		return ERR_PTR(-ENOMEM);
+
+	padev->vf_pdev = cf->pdev;
+	padev->client_id = client_id;
+
+	aux_dev = &padev->aux_dev;
+	aux_dev->name = name;
+	aux_dev->id = cf->uid;
+	aux_dev->dev.parent = cf->dev;
+	aux_dev->dev.release = pdsc_auxbus_dev_release;
+
+	err = auxiliary_device_init(aux_dev);
+	if (err < 0) {
+		dev_warn(cf->dev, "auxiliary_device_init of %s failed: %pe\n",
+			 name, ERR_PTR(err));
+		kfree(padev);
+		return ERR_PTR(err);
+	}
+
+	err = auxiliary_device_add(aux_dev);
+	if (err) {
+		dev_warn(cf->dev, "auxiliary_device_add of %s failed: %pe\n",
+			 name, ERR_PTR(err));
+		auxiliary_device_uninit(aux_dev);
+		return ERR_PTR(err);
+	}
+
+	return padev;
+}
+
+void pdsc_auxbus_dev_del(struct pdsc *cf, struct pdsc *pf,
+			 struct pds_auxiliary_dev **pd_ptr)
+{
+	struct pds_auxiliary_dev *padev;
+
+	if (!*pd_ptr)
+		return;
+
+	mutex_lock(&pf->config_lock);
+
+	padev = *pd_ptr;
+	pds_client_unregister(pf, padev->client_id);
+	auxiliary_device_delete(&padev->aux_dev);
+	auxiliary_device_uninit(&padev->aux_dev);
+	*pd_ptr = NULL;
+
+	mutex_unlock(&pf->config_lock);
+}
+
+int pdsc_auxbus_dev_add(struct pdsc *cf, struct pdsc *pf,
+			enum pds_core_vif_types vt,
+			struct pds_auxiliary_dev **pd_ptr)
+{
+	struct pds_auxiliary_dev *padev;
+	char devname[PDS_DEVNAME_LEN];
+	unsigned long mask;
+	u16 vt_support;
+	int client_id;
+	int err = 0;
+
+	if (!cf)
+		return -ENODEV;
+
+	if (vt >= PDS_DEV_TYPE_MAX)
+		return -EINVAL;
+
+	mutex_lock(&pf->config_lock);
+
+	mask = BIT_ULL(PDSC_S_FW_DEAD) |
+	       BIT_ULL(PDSC_S_STOPPING_DRIVER);
+	if (cf->state & mask) {
+		dev_err(pf->dev, "%s: can't add dev, VF client in bad state %#lx\n",
+			__func__, cf->state);
+		err = -ENXIO;
+		goto out_unlock;
+	}
+
+	/* Verify that the type is supported and enabled.  It is not
+	 * an error if the firmware doesn't support the feature, the
+	 * driver just won't set up an auxiliary_device for it.
+	 */
+	vt_support = !!le16_to_cpu(pf->dev_ident.vif_types[vt]);
+	if (!(vt_support &&
+	      pf->viftype_status[vt].supported &&
+	      pf->viftype_status[vt].enabled))
+		goto out_unlock;
+
+	/* Need to register with FW and get the client_id before
+	 * creating the aux device so that the aux client can run
+	 * adminq commands as part its probe
+	 */
+	snprintf(devname, sizeof(devname), "%s.%s.%d",
+		 PDS_CORE_DRV_NAME, pf->viftype_status[vt].name, cf->uid);
+	client_id = pds_client_register(pf, devname);
+	if (client_id < 0) {
+		err = client_id;
+		goto out_unlock;
+	}
+
+	padev = pdsc_auxbus_dev_register(cf, pf, client_id,
+					 pf->viftype_status[vt].name);
+	if (IS_ERR(padev)) {
+		pds_client_unregister(pf, client_id);
+		err = PTR_ERR(padev);
+		goto out_unlock;
+	}
+	*pd_ptr = padev;
+
+out_unlock:
+	mutex_unlock(&pf->config_lock);
+	return err;
+}
diff --git a/drivers/net/ethernet/amd/pds_core/core.c b/drivers/net/ethernet/amd/pds_core/core.c
new file mode 100644
index 000000000000..076dfe2910c7
--- /dev/null
+++ b/drivers/net/ethernet/amd/pds_core/core.c
@@ -0,0 +1,661 @@
+// SPDX-License-Identifier: GPL-2.0
+/* Copyright(c) 2023 Advanced Micro Devices, Inc */
+
+#include <linux/pci.h>
+#include <linux/vmalloc.h>
+
+#include "core.h"
+
+static BLOCKING_NOTIFIER_HEAD(pds_notify_chain);
+
+int pdsc_register_notify(struct notifier_block *nb)
+{
+	return blocking_notifier_chain_register(&pds_notify_chain, nb);
+}
+EXPORT_SYMBOL_GPL(pdsc_register_notify);
+
+void pdsc_unregister_notify(struct notifier_block *nb)
+{
+	blocking_notifier_chain_unregister(&pds_notify_chain, nb);
+}
+EXPORT_SYMBOL_GPL(pdsc_unregister_notify);
+
+void pdsc_notify(unsigned long event, void *data)
+{
+	blocking_notifier_call_chain(&pds_notify_chain, event, data);
+}
+
+void pdsc_intr_free(struct pdsc *pdsc, int index)
+{
+	struct pdsc_intr_info *intr_info;
+
+	if (index >= pdsc->nintrs || index < 0) {
+		WARN(true, "bad intr index %d\n", index);
+		return;
+	}
+
+	intr_info = &pdsc->intr_info[index];
+	if (!intr_info->vector)
+		return;
+	dev_dbg(pdsc->dev, "%s: idx %d vec %d name %s\n",
+		__func__, index, intr_info->vector, intr_info->name);
+
+	pds_core_intr_mask(&pdsc->intr_ctrl[index], PDS_CORE_INTR_MASK_SET);
+	pds_core_intr_clean(&pdsc->intr_ctrl[index]);
+
+	free_irq(intr_info->vector, intr_info->data);
+
+	memset(intr_info, 0, sizeof(*intr_info));
+}
+
+int pdsc_intr_alloc(struct pdsc *pdsc, char *name,
+		    irq_handler_t handler, void *data)
+{
+	struct pdsc_intr_info *intr_info;
+	unsigned int index;
+	int err;
+
+	/* Find the first available interrupt */
+	for (index = 0; index < pdsc->nintrs; index++)
+		if (!pdsc->intr_info[index].vector)
+			break;
+	if (index >= pdsc->nintrs) {
+		dev_warn(pdsc->dev, "%s: no intr, index=%d nintrs=%d\n",
+			 __func__, index, pdsc->nintrs);
+		return -ENOSPC;
+	}
+
+	pds_core_intr_clean_flags(&pdsc->intr_ctrl[index],
+				  PDS_CORE_INTR_CRED_RESET_COALESCE);
+
+	intr_info = &pdsc->intr_info[index];
+
+	intr_info->index = index;
+	intr_info->data = data;
+	strscpy(intr_info->name, name, sizeof(intr_info->name));
+
+	/* Get the OS vector number for the interrupt */
+	err = pci_irq_vector(pdsc->pdev, index);
+	if (err < 0) {
+		dev_err(pdsc->dev, "failed to get intr vector index %d: %pe\n",
+			index, ERR_PTR(err));
+		goto err_out_free_intr;
+	}
+	intr_info->vector = err;
+
+	/* Init the device's intr mask */
+	pds_core_intr_clean(&pdsc->intr_ctrl[index]);
+	pds_core_intr_mask_assert(&pdsc->intr_ctrl[index], 1);
+	pds_core_intr_mask(&pdsc->intr_ctrl[index], PDS_CORE_INTR_MASK_SET);
+
+	/* Register the isr with a name */
+	err = request_irq(intr_info->vector, handler, 0, intr_info->name, data);
+	if (err) {
+		dev_err(pdsc->dev, "failed to get intr irq vector %d: %pe\n",
+			intr_info->vector, ERR_PTR(err));
+		goto err_out_free_intr;
+	}
+
+	return index;
+
+err_out_free_intr:
+	pdsc_intr_free(pdsc, index);
+	return err;
+}
+
+static void pdsc_qcq_intr_free(struct pdsc *pdsc, struct pdsc_qcq *qcq)
+{
+	if (!(qcq->flags & PDS_CORE_QCQ_F_INTR) ||
+	    qcq->intx == PDS_CORE_INTR_INDEX_NOT_ASSIGNED)
+		return;
+
+	pdsc_intr_free(pdsc, qcq->intx);
+	qcq->intx = PDS_CORE_INTR_INDEX_NOT_ASSIGNED;
+}
+
+static int pdsc_qcq_intr_alloc(struct pdsc *pdsc, struct pdsc_qcq *qcq)
+{
+	char name[PDSC_INTR_NAME_MAX_SZ];
+	int index;
+
+	if (!(qcq->flags & PDS_CORE_QCQ_F_INTR)) {
+		qcq->intx = PDS_CORE_INTR_INDEX_NOT_ASSIGNED;
+		return 0;
+	}
+
+	snprintf(name, sizeof(name), "%s-%d-%s",
+		 PDS_CORE_DRV_NAME, pdsc->pdev->bus->number, qcq->q.name);
+	index = pdsc_intr_alloc(pdsc, name, pdsc_adminq_isr, pdsc);
+	if (index < 0)
+		return index;
+	qcq->intx = index;
+	qcq->cq.bound_intr = &pdsc->intr_info[index];
+
+	return 0;
+}
+
+void pdsc_qcq_free(struct pdsc *pdsc, struct pdsc_qcq *qcq)
+{
+	struct device *dev = pdsc->dev;
+
+	if (!(qcq && qcq->pdsc))
+		return;
+
+	pdsc_debugfs_del_qcq(qcq);
+
+	pdsc_qcq_intr_free(pdsc, qcq);
+
+	if (qcq->q_base)
+		dma_free_coherent(dev, qcq->q_size,
+				  qcq->q_base, qcq->q_base_pa);
+
+	if (qcq->cq_base)
+		dma_free_coherent(dev, qcq->cq_size,
+				  qcq->cq_base, qcq->cq_base_pa);
+
+	vfree(qcq->cq.info);
+	vfree(qcq->q.info);
+
+	memset(qcq, 0, sizeof(*qcq));
+}
+
+static void pdsc_q_map(struct pdsc_queue *q, void *base, dma_addr_t base_pa)
+{
+	struct pdsc_q_info *cur;
+	unsigned int i;
+
+	q->base = base;
+	q->base_pa = base_pa;
+
+	for (i = 0, cur = q->info; i < q->num_descs; i++, cur++) {
+		cur->desc = base + (i * q->desc_size);
+		init_completion(&cur->completion);
+	}
+}
+
+static void pdsc_cq_map(struct pdsc_cq *cq, void *base, dma_addr_t base_pa)
+{
+	struct pdsc_cq_info *cur;
+	unsigned int i;
+
+	cq->base = base;
+	cq->base_pa = base_pa;
+
+	for (i = 0, cur = cq->info; i < cq->num_descs; i++, cur++)
+		cur->comp = base + (i * cq->desc_size);
+}
+
+int pdsc_qcq_alloc(struct pdsc *pdsc, unsigned int type, unsigned int index,
+		   const char *name, unsigned int flags, unsigned int num_descs,
+		   unsigned int desc_size, unsigned int cq_desc_size,
+		   unsigned int pid, struct pdsc_qcq *qcq)
+{
+	struct device *dev = pdsc->dev;
+	void *q_base, *cq_base;
+	dma_addr_t cq_base_pa;
+	dma_addr_t q_base_pa;
+	int err;
+
+	qcq->q.info = vcalloc(num_descs, sizeof(*qcq->q.info));
+	if (!qcq->q.info) {
+		err = -ENOMEM;
+		goto err_out;
+	}
+
+	qcq->pdsc = pdsc;
+	qcq->flags = flags;
+	INIT_WORK(&qcq->work, pdsc_work_thread);
+
+	qcq->q.type = type;
+	qcq->q.index = index;
+	qcq->q.num_descs = num_descs;
+	qcq->q.desc_size = desc_size;
+	qcq->q.tail_idx = 0;
+	qcq->q.head_idx = 0;
+	qcq->q.pid = pid;
+	snprintf(qcq->q.name, sizeof(qcq->q.name), "%s%u", name, index);
+
+	err = pdsc_qcq_intr_alloc(pdsc, qcq);
+	if (err)
+		goto err_out_free_q_info;
+
+	qcq->cq.info = vcalloc(num_descs, sizeof(*qcq->cq.info));
+	if (!qcq->cq.info) {
+		err = -ENOMEM;
+		goto err_out_free_irq;
+	}
+
+	qcq->cq.num_descs = num_descs;
+	qcq->cq.desc_size = cq_desc_size;
+	qcq->cq.tail_idx = 0;
+	qcq->cq.done_color = 1;
+
+	if (flags & PDS_CORE_QCQ_F_NOTIFYQ) {
+		/* q & cq need to be contiguous in case of notifyq */
+		qcq->q_size = PDS_PAGE_SIZE +
+			      ALIGN(num_descs * desc_size, PDS_PAGE_SIZE) +
+			      ALIGN(num_descs * cq_desc_size, PDS_PAGE_SIZE);
+		qcq->q_base = dma_alloc_coherent(dev,
+						 qcq->q_size + qcq->cq_size,
+						 &qcq->q_base_pa,
+						 GFP_KERNEL);
+		if (!qcq->q_base) {
+			err = -ENOMEM;
+			goto err_out_free_cq_info;
+		}
+		q_base = PTR_ALIGN(qcq->q_base, PDS_PAGE_SIZE);
+		q_base_pa = ALIGN(qcq->q_base_pa, PDS_PAGE_SIZE);
+		pdsc_q_map(&qcq->q, q_base, q_base_pa);
+
+		cq_base = PTR_ALIGN(q_base +
+				    ALIGN(num_descs * desc_size, PDS_PAGE_SIZE),
+				    PDS_PAGE_SIZE);
+		cq_base_pa = ALIGN(qcq->q_base_pa +
+				   ALIGN(num_descs * desc_size, PDS_PAGE_SIZE),
+				   PDS_PAGE_SIZE);
+
+	} else {
+		/* q DMA descriptors */
+		qcq->q_size = PDS_PAGE_SIZE + (num_descs * desc_size);
+		qcq->q_base = dma_alloc_coherent(dev, qcq->q_size,
+						 &qcq->q_base_pa,
+						 GFP_KERNEL);
+		if (!qcq->q_base) {
+			err = -ENOMEM;
+			goto err_out_free_cq_info;
+		}
+		q_base = PTR_ALIGN(qcq->q_base, PDS_PAGE_SIZE);
+		q_base_pa = ALIGN(qcq->q_base_pa, PDS_PAGE_SIZE);
+		pdsc_q_map(&qcq->q, q_base, q_base_pa);
+
+		/* cq DMA descriptors */
+		qcq->cq_size = PDS_PAGE_SIZE + (num_descs * cq_desc_size);
+		qcq->cq_base = dma_alloc_coherent(dev, qcq->cq_size,
+						  &qcq->cq_base_pa,
+						  GFP_KERNEL);
+		if (!qcq->cq_base) {
+			err = -ENOMEM;
+			goto err_out_free_q;
+		}
+		cq_base = PTR_ALIGN(qcq->cq_base, PDS_PAGE_SIZE);
+		cq_base_pa = ALIGN(qcq->cq_base_pa, PDS_PAGE_SIZE);
+	}
+
+	pdsc_cq_map(&qcq->cq, cq_base, cq_base_pa);
+	qcq->cq.bound_q = &qcq->q;
+
+	pdsc_debugfs_add_qcq(pdsc, qcq);
+
+	return 0;
+
+err_out_free_q:
+	dma_free_coherent(dev, qcq->q_size, qcq->q_base, qcq->q_base_pa);
+err_out_free_cq_info:
+	vfree(qcq->cq.info);
+err_out_free_irq:
+	pdsc_qcq_intr_free(pdsc, qcq);
+err_out_free_q_info:
+	vfree(qcq->q.info);
+	memset(qcq, 0, sizeof(*qcq));
+err_out:
+	dev_err(dev, "qcq alloc of %s%d failed %d\n", name, index, err);
+	return err;
+}
+
+static void pdsc_core_uninit(struct pdsc *pdsc)
+{
+	pdsc_qcq_free(pdsc, &pdsc->notifyqcq);
+	pdsc_qcq_free(pdsc, &pdsc->adminqcq);
+
+	if (pdsc->kern_dbpage) {
+		iounmap(pdsc->kern_dbpage);
+		pdsc->kern_dbpage = NULL;
+	}
+}
+
+static int pdsc_core_init(struct pdsc *pdsc)
+{
+	union pds_core_dev_comp comp = {};
+	union pds_core_dev_cmd cmd = {
+		.init.opcode = PDS_CORE_CMD_INIT,
+	};
+	struct pds_core_dev_init_data_out cido;
+	struct pds_core_dev_init_data_in cidi;
+	u32 dbid_count;
+	u32 dbpage_num;
+	int numdescs;
+	size_t sz;
+	int err;
+
+	numdescs = PDSC_ADMINQ_MAX_LENGTH;
+	err = pdsc_qcq_alloc(pdsc, PDS_CORE_QTYPE_ADMINQ, 0, "adminq",
+			     PDS_CORE_QCQ_F_CORE | PDS_CORE_QCQ_F_INTR,
+			     numdescs,
+			     sizeof(union pds_core_adminq_cmd),
+			     sizeof(union pds_core_adminq_comp),
+			     0, &pdsc->adminqcq);
+	if (err)
+		return err;
+
+	err = pdsc_qcq_alloc(pdsc, PDS_CORE_QTYPE_NOTIFYQ, 0, "notifyq",
+			     PDS_CORE_QCQ_F_NOTIFYQ,
+			     PDSC_NOTIFYQ_LENGTH,
+			     sizeof(struct pds_core_notifyq_cmd),
+			     sizeof(union pds_core_notifyq_comp),
+			     0, &pdsc->notifyqcq);
+	if (err)
+		goto err_out_uninit;
+
+	cidi.adminq_q_base = cpu_to_le64(pdsc->adminqcq.q_base_pa);
+	cidi.adminq_cq_base = cpu_to_le64(pdsc->adminqcq.cq_base_pa);
+	cidi.notifyq_cq_base = cpu_to_le64(pdsc->notifyqcq.cq.base_pa);
+	cidi.flags = cpu_to_le32(PDS_CORE_QINIT_F_IRQ | PDS_CORE_QINIT_F_ENA);
+	cidi.intr_index = cpu_to_le16(pdsc->adminqcq.intx);
+	cidi.adminq_ring_size = ilog2(pdsc->adminqcq.q.num_descs);
+	cidi.notifyq_ring_size = ilog2(pdsc->notifyqcq.q.num_descs);
+
+	mutex_lock(&pdsc->devcmd_lock);
+
+	sz = min_t(size_t, sizeof(cidi), sizeof(pdsc->cmd_regs->data));
+	memcpy_toio(&pdsc->cmd_regs->data, &cidi, sz);
+
+	err = pdsc_devcmd_locked(pdsc, &cmd, &comp, pdsc->devcmd_timeout);
+	if (!err) {
+		sz = min_t(size_t, sizeof(cido), sizeof(pdsc->cmd_regs->data));
+		memcpy_fromio(&cido, &pdsc->cmd_regs->data, sz);
+	}
+
+	mutex_unlock(&pdsc->devcmd_lock);
+	if (err) {
+		dev_err(pdsc->dev, "Device init command failed: %pe\n",
+			ERR_PTR(err));
+		goto err_out_uninit;
+	}
+
+	pdsc->hw_index = le32_to_cpu(cido.core_hw_index);
+
+	dbid_count = le32_to_cpu(pdsc->dev_ident.ndbpgs_per_lif);
+	dbpage_num = pdsc->hw_index * dbid_count;
+	pdsc->kern_dbpage = pdsc_map_dbpage(pdsc, dbpage_num);
+	if (!pdsc->kern_dbpage) {
+		dev_err(pdsc->dev, "Cannot map dbpage, aborting\n");
+		err = -ENOMEM;
+		goto err_out_uninit;
+	}
+
+	pdsc->adminqcq.q.hw_type = cido.adminq_hw_type;
+	pdsc->adminqcq.q.hw_index = le32_to_cpu(cido.adminq_hw_index);
+	pdsc->adminqcq.q.dbval = PDS_CORE_DBELL_QID(pdsc->adminqcq.q.hw_index);
+
+	pdsc->notifyqcq.q.hw_type = cido.notifyq_hw_type;
+	pdsc->notifyqcq.q.hw_index = le32_to_cpu(cido.notifyq_hw_index);
+	pdsc->notifyqcq.q.dbval = PDS_CORE_DBELL_QID(pdsc->notifyqcq.q.hw_index);
+
+	pdsc->last_eid = 0;
+
+	return 0;
+
+err_out_uninit:
+	pdsc_core_uninit(pdsc);
+	return err;
+}
+
+static struct pdsc_viftype pdsc_viftype_defaults[] = {
+	[PDS_DEV_TYPE_FWCTL] = { .name = PDS_DEV_TYPE_FWCTL_STR,
+				 .enabled = true,
+				 .vif_id = PDS_DEV_TYPE_FWCTL,
+				 .dl_id = -1 },
+	[PDS_DEV_TYPE_VDPA] = { .name = PDS_DEV_TYPE_VDPA_STR,
+				.vif_id = PDS_DEV_TYPE_VDPA,
+				.dl_id = DEVLINK_PARAM_GENERIC_ID_ENABLE_VNET },
+	[PDS_DEV_TYPE_MAX] = {}
+};
+
+static int pdsc_viftypes_init(struct pdsc *pdsc)
+{
+	enum pds_core_vif_types vt;
+
+	pdsc->viftype_status = kcalloc(ARRAY_SIZE(pdsc_viftype_defaults),
+				       sizeof(*pdsc->viftype_status),
+				       GFP_KERNEL);
+	if (!pdsc->viftype_status)
+		return -ENOMEM;
+
+	for (vt = 0; vt < PDS_DEV_TYPE_MAX; vt++) {
+		bool vt_support;
+
+		if (!pdsc_viftype_defaults[vt].name)
+			continue;
+
+		/* Grab the defaults */
+		pdsc->viftype_status[vt] = pdsc_viftype_defaults[vt];
+
+		/* See what the Core device has for support */
+		vt_support = !!le16_to_cpu(pdsc->dev_ident.vif_types[vt]);
+
+		dev_dbg(pdsc->dev, "VIF %s is %ssupported\n",
+			pdsc->viftype_status[vt].name,
+			vt_support ? "" : "not ");
+
+		pdsc->viftype_status[vt].supported = vt_support;
+	}
+
+	return 0;
+}
+
+int pdsc_setup(struct pdsc *pdsc, bool init)
+{
+	int err;
+
+	err = pdsc_dev_init(pdsc);
+	if (err)
+		return err;
+
+	/* Set up the Core with the AdminQ and NotifyQ info */
+	err = pdsc_core_init(pdsc);
+	if (err)
+		goto err_out_teardown;
+
+	/* Set up the VIFs */
+	if (init) {
+		err = pdsc_viftypes_init(pdsc);
+		if (err)
+			goto err_out_teardown;
+
+		pdsc_debugfs_add_viftype(pdsc);
+	}
+
+	refcount_set(&pdsc->adminq_refcnt, 1);
+	clear_bit(PDSC_S_FW_DEAD, &pdsc->state);
+	return 0;
+
+err_out_teardown:
+	pdsc_teardown(pdsc, init);
+	return err;
+}
+
+void pdsc_teardown(struct pdsc *pdsc, bool removing)
+{
+	if (!pdsc->pdev->is_virtfn)
+		pdsc_devcmd_reset(pdsc);
+	if (pdsc->adminqcq.work.func)
+		cancel_work_sync(&pdsc->adminqcq.work);
+
+	pdsc_core_uninit(pdsc);
+
+	if (removing) {
+		kfree(pdsc->viftype_status);
+		pdsc->viftype_status = NULL;
+	}
+
+	pdsc_dev_uninit(pdsc);
+
+	set_bit(PDSC_S_FW_DEAD, &pdsc->state);
+}
+
+int pdsc_start(struct pdsc *pdsc)
+{
+	pds_core_intr_mask(&pdsc->intr_ctrl[pdsc->adminqcq.intx],
+			   PDS_CORE_INTR_MASK_CLEAR);
+
+	return 0;
+}
+
+void pdsc_stop(struct pdsc *pdsc)
+{
+	int i;
+
+	if (!pdsc->intr_info)
+		return;
+
+	/* Mask interrupts that are in use */
+	for (i = 0; i < pdsc->nintrs; i++)
+		if (pdsc->intr_info[i].vector)
+			pds_core_intr_mask(&pdsc->intr_ctrl[i],
+					   PDS_CORE_INTR_MASK_SET);
+}
+
+static void pdsc_adminq_wait_and_dec_once_unused(struct pdsc *pdsc)
+{
+	/* The driver initializes the adminq_refcnt to 1 when the adminq is
+	 * allocated and ready for use. Other users/requesters will increment
+	 * the refcnt while in use. If the refcnt is down to 1 then the adminq
+	 * is not in use and the refcnt can be cleared and adminq freed. Before
+	 * calling this function the driver will set PDSC_S_FW_DEAD, which
+	 * prevent subsequent attempts to use the adminq and increment the
+	 * refcnt to fail. This guarantees that this function will eventually
+	 * exit.
+	 */
+	while (!refcount_dec_if_one(&pdsc->adminq_refcnt)) {
+		dev_dbg_ratelimited(pdsc->dev, "%s: adminq in use\n",
+				    __func__);
+		cpu_relax();
+	}
+}
+
+void pdsc_fw_down(struct pdsc *pdsc)
+{
+	union pds_core_notifyq_comp reset_event = {
+		.reset.ecode = cpu_to_le16(PDS_EVENT_RESET),
+		.reset.state = 0,
+	};
+
+	if (test_and_set_bit(PDSC_S_FW_DEAD, &pdsc->state)) {
+		dev_warn(pdsc->dev, "%s: already happening\n", __func__);
+		return;
+	}
+
+	if (pdsc->pdev->is_virtfn)
+		return;
+
+	pdsc_adminq_wait_and_dec_once_unused(pdsc);
+
+	/* Notify clients of fw_down */
+	if (pdsc->fw_reporter)
+		devlink_health_report(pdsc->fw_reporter, "FW down reported", pdsc);
+	pdsc_notify(PDS_EVENT_RESET, &reset_event);
+
+	pdsc_stop(pdsc);
+	pdsc_teardown(pdsc, PDSC_TEARDOWN_RECOVERY);
+}
+
+void pdsc_fw_up(struct pdsc *pdsc)
+{
+	union pds_core_notifyq_comp reset_event = {
+		.reset.ecode = cpu_to_le16(PDS_EVENT_RESET),
+		.reset.state = 1,
+	};
+	int err;
+
+	if (!test_bit(PDSC_S_FW_DEAD, &pdsc->state)) {
+		dev_err(pdsc->dev, "%s: fw not dead\n", __func__);
+		return;
+	}
+
+	if (pdsc->pdev->is_virtfn) {
+		clear_bit(PDSC_S_FW_DEAD, &pdsc->state);
+		return;
+	}
+
+	err = pdsc_setup(pdsc, PDSC_SETUP_RECOVERY);
+	if (err)
+		goto err_out;
+
+	err = pdsc_start(pdsc);
+	if (err)
+		goto err_out;
+
+	/* Notify clients of fw_up */
+	pdsc->fw_recoveries++;
+	if (pdsc->fw_reporter)
+		devlink_health_reporter_state_update(pdsc->fw_reporter,
+						     DEVLINK_HEALTH_REPORTER_STATE_HEALTHY);
+	pdsc_notify(PDS_EVENT_RESET, &reset_event);
+
+	return;
+
+err_out:
+	pdsc_teardown(pdsc, PDSC_TEARDOWN_RECOVERY);
+}
+
+void pdsc_pci_reset_thread(struct work_struct *work)
+{
+	struct pdsc *pdsc = container_of(work, struct pdsc, pci_reset_work);
+	struct pci_dev *pdev = pdsc->pdev;
+
+	pci_dev_get(pdev);
+	pci_reset_function(pdev);
+	pci_dev_put(pdev);
+}
+
+static void pdsc_check_pci_health(struct pdsc *pdsc)
+{
+	u8 fw_status;
+
+	/* some sort of teardown already in progress */
+	if (!pdsc->info_regs)
+		return;
+
+	fw_status = ioread8(&pdsc->info_regs->fw_status);
+
+	/* is PCI broken? */
+	if (fw_status != PDS_RC_BAD_PCI)
+		return;
+
+	/* prevent deadlock between pdsc_reset_prepare and pdsc_health_thread */
+	queue_work(pdsc->wq, &pdsc->pci_reset_work);
+}
+
+void pdsc_health_thread(struct work_struct *work)
+{
+	struct pdsc *pdsc = container_of(work, struct pdsc, health_work);
+	unsigned long mask;
+	bool healthy;
+
+	mutex_lock(&pdsc->config_lock);
+
+	/* Don't do a check when in a transition state */
+	mask = BIT_ULL(PDSC_S_INITING_DRIVER) |
+	       BIT_ULL(PDSC_S_STOPPING_DRIVER);
+	if (pdsc->state & mask)
+		goto out_unlock;
+
+	healthy = pdsc_is_fw_good(pdsc);
+	dev_dbg(pdsc->dev, "%s: health %d fw_status %#02x fw_heartbeat %d\n",
+		__func__, healthy, pdsc->fw_status, pdsc->last_hb);
+
+	if (test_bit(PDSC_S_FW_DEAD, &pdsc->state)) {
+		if (healthy)
+			pdsc_fw_up(pdsc);
+	} else {
+		if (!healthy)
+			pdsc_fw_down(pdsc);
+	}
+
+	pdsc_check_pci_health(pdsc);
+
+	pdsc->fw_generation = pdsc->fw_status & PDS_CORE_FW_STS_F_GENERATION;
+
+out_unlock:
+	mutex_unlock(&pdsc->config_lock);
+}
diff --git a/drivers/net/ethernet/amd/pds_core/core.h b/drivers/net/ethernet/amd/pds_core/core.h
new file mode 100644
index 000000000000..0b53a1fab46d
--- /dev/null
+++ b/drivers/net/ethernet/amd/pds_core/core.h
@@ -0,0 +1,324 @@
+/* SPDX-License-Identifier: GPL-2.0 */
+/* Copyright(c) 2023 Advanced Micro Devices, Inc */
+
+#ifndef _PDSC_H_
+#define _PDSC_H_
+
+#include <linux/debugfs.h>
+#include <net/devlink.h>
+
+#include <linux/pds/pds_common.h>
+#include <linux/pds/pds_core_if.h>
+#include <linux/pds/pds_adminq.h>
+#include <linux/pds/pds_intr.h>
+
+#define PDSC_DRV_DESCRIPTION	"AMD/Pensando Core Driver"
+
+#define PDSC_WATCHDOG_SECS	5
+#define PDSC_QUEUE_NAME_MAX_SZ  16
+#define PDSC_ADMINQ_MAX_LENGTH	16	/* must be a power of two */
+#define PDSC_NOTIFYQ_LENGTH	64	/* must be a power of two */
+#define PDSC_TEARDOWN_RECOVERY	false
+#define PDSC_TEARDOWN_REMOVING	true
+#define PDSC_SETUP_RECOVERY	false
+#define PDSC_SETUP_INIT		true
+
+struct pdsc_dev_bar {
+	void __iomem *vaddr;
+	phys_addr_t bus_addr;
+	unsigned long len;
+	int res_index;
+};
+
+struct pdsc;
+
+struct pdsc_vf {
+	struct pds_auxiliary_dev *padev;
+	struct pdsc *vf;
+	u16     index;
+	__le16  vif_types[PDS_DEV_TYPE_MAX];
+};
+
+struct pdsc_devinfo {
+	u8 asic_type;
+	u8 asic_rev;
+	char fw_version[PDS_CORE_DEVINFO_FWVERS_BUFLEN + 1];
+	char serial_num[PDS_CORE_DEVINFO_SERIAL_BUFLEN + 1];
+};
+
+struct pdsc_queue {
+	struct pdsc_q_info *info;
+	u64 dbval;
+	u16 head_idx;
+	u16 tail_idx;
+	u8 hw_type;
+	unsigned int index;
+	unsigned int num_descs;
+	u64 dbell_count;
+	u64 features;
+	unsigned int type;
+	unsigned int hw_index;
+	union {
+		void *base;
+		struct pds_core_admin_cmd *adminq;
+	};
+	dma_addr_t base_pa;	/* must be page aligned */
+	unsigned int desc_size;
+	unsigned int pid;
+	char name[PDSC_QUEUE_NAME_MAX_SZ];
+};
+
+#define PDSC_INTR_NAME_MAX_SZ		32
+
+struct pdsc_intr_info {
+	char name[PDSC_INTR_NAME_MAX_SZ];
+	unsigned int index;
+	unsigned int vector;
+	void *data;
+};
+
+struct pdsc_cq_info {
+	void *comp;
+};
+
+struct pdsc_buf_info {
+	struct page *page;
+	dma_addr_t dma_addr;
+	u32 page_offset;
+	u32 len;
+};
+
+struct pdsc_q_info {
+	union {
+		void *desc;
+		struct pdsc_admin_cmd *adminq_desc;
+	};
+	unsigned int bytes;
+	unsigned int nbufs;
+	struct pdsc_buf_info bufs[PDS_CORE_MAX_FRAGS];
+	struct completion completion;
+	void *dest;
+};
+
+struct pdsc_cq {
+	struct pdsc_cq_info *info;
+	struct pdsc_queue *bound_q;
+	struct pdsc_intr_info *bound_intr;
+	u16 tail_idx;
+	bool done_color;
+	unsigned int num_descs;
+	unsigned int desc_size;
+	void *base;
+	dma_addr_t base_pa;	/* must be page aligned */
+} ____cacheline_aligned_in_smp;
+
+struct pdsc_qcq {
+	struct pdsc *pdsc;
+	void *q_base;
+	dma_addr_t q_base_pa;	/* might not be page aligned */
+	void *cq_base;
+	dma_addr_t cq_base_pa;	/* might not be page aligned */
+	u32 q_size;
+	u32 cq_size;
+	bool armed;
+	unsigned int flags;
+
+	struct work_struct work;
+	struct pdsc_queue q;
+	struct pdsc_cq cq;
+	int intx;
+
+	u32 accum_work;
+	struct dentry *dentry;
+};
+
+struct pdsc_viftype {
+	char *name;
+	bool supported;
+	bool enabled;
+	int dl_id;
+	int vif_id;
+	struct pds_auxiliary_dev *padev;
+};
+
+/* No state flags set means we are in a steady running state */
+enum pdsc_state_flags {
+	PDSC_S_FW_DEAD,		    /* stopped, wait on startup or recovery */
+	PDSC_S_INITING_DRIVER,	    /* initial startup from probe */
+	PDSC_S_STOPPING_DRIVER,	    /* driver remove */
+
+	/* leave this as last */
+	PDSC_S_STATE_SIZE
+};
+
+struct pdsc {
+	struct pci_dev *pdev;
+	struct dentry *dentry;
+	struct device *dev;
+	struct pdsc_dev_bar bars[PDS_CORE_BARS_MAX];
+	struct pds_auxiliary_dev *padev;
+	struct pdsc_vf *vfs;
+	int num_vfs;
+	int vf_id;
+	int hw_index;
+	int uid;
+
+	unsigned long state;
+	u8 fw_status;
+	u8 fw_generation;
+	unsigned long last_fw_time;
+	u32 last_hb;
+	struct timer_list wdtimer;
+	unsigned int wdtimer_period;
+	struct work_struct health_work;
+	struct devlink_health_reporter *fw_reporter;
+	u32 fw_recoveries;
+
+	struct pdsc_devinfo dev_info;
+	struct pds_core_dev_identity dev_ident;
+	unsigned int nintrs;
+	struct pdsc_intr_info *intr_info;	/* array of nintrs elements */
+
+	struct workqueue_struct *wq;
+
+	unsigned int devcmd_timeout;
+	struct mutex devcmd_lock;	/* lock for dev_cmd operations */
+	struct mutex config_lock;	/* lock for configuration operations */
+	spinlock_t adminq_lock;		/* lock for adminq operations */
+	refcount_t adminq_refcnt;
+	struct pds_core_dev_info_regs __iomem *info_regs;
+	struct pds_core_dev_cmd_regs __iomem *cmd_regs;
+	struct pds_core_intr __iomem *intr_ctrl;
+	u64 __iomem *intr_status;
+	u64 __iomem *db_pages;
+	dma_addr_t phy_db_pages;
+	u64 __iomem *kern_dbpage;
+
+	struct pdsc_qcq adminqcq;
+	struct pdsc_qcq notifyqcq;
+	u64 last_eid;
+	struct pdsc_viftype *viftype_status;
+	struct work_struct pci_reset_work;
+};
+
+/** enum pds_core_dbell_bits - bitwise composition of dbell values.
+ *
+ * @PDS_CORE_DBELL_QID_MASK:	unshifted mask of valid queue id bits.
+ * @PDS_CORE_DBELL_QID_SHIFT:	queue id shift amount in dbell value.
+ * @PDS_CORE_DBELL_QID:		macro to build QID component of dbell value.
+ *
+ * @PDS_CORE_DBELL_RING_MASK:	unshifted mask of valid ring bits.
+ * @PDS_CORE_DBELL_RING_SHIFT:	ring shift amount in dbell value.
+ * @PDS_CORE_DBELL_RING:	macro to build ring component of dbell value.
+ *
+ * @PDS_CORE_DBELL_RING_0:	ring zero dbell component value.
+ * @PDS_CORE_DBELL_RING_1:	ring one dbell component value.
+ * @PDS_CORE_DBELL_RING_2:	ring two dbell component value.
+ * @PDS_CORE_DBELL_RING_3:	ring three dbell component value.
+ *
+ * @PDS_CORE_DBELL_INDEX_MASK:	bit mask of valid index bits, no shift needed.
+ */
+enum pds_core_dbell_bits {
+	PDS_CORE_DBELL_QID_MASK		= 0xffffff,
+	PDS_CORE_DBELL_QID_SHIFT		= 24,
+
+#define PDS_CORE_DBELL_QID(n) \
+	(((u64)(n) & PDS_CORE_DBELL_QID_MASK) << PDS_CORE_DBELL_QID_SHIFT)
+
+	PDS_CORE_DBELL_RING_MASK		= 0x7,
+	PDS_CORE_DBELL_RING_SHIFT		= 16,
+
+#define PDS_CORE_DBELL_RING(n) \
+	(((u64)(n) & PDS_CORE_DBELL_RING_MASK) << PDS_CORE_DBELL_RING_SHIFT)
+
+	PDS_CORE_DBELL_RING_0		= 0,
+	PDS_CORE_DBELL_RING_1		= PDS_CORE_DBELL_RING(1),
+	PDS_CORE_DBELL_RING_2		= PDS_CORE_DBELL_RING(2),
+	PDS_CORE_DBELL_RING_3		= PDS_CORE_DBELL_RING(3),
+
+	PDS_CORE_DBELL_INDEX_MASK		= 0xffff,
+};
+
+static inline void pds_core_dbell_ring(u64 __iomem *db_page,
+				       enum pds_core_logical_qtype qtype,
+				       u64 val)
+{
+	writeq(val, &db_page[qtype]);
+}
+
+int pdsc_fw_reporter_diagnose(struct devlink_health_reporter *reporter,
+			      struct devlink_fmsg *fmsg,
+			      struct netlink_ext_ack *extack);
+int pdsc_dl_info_get(struct devlink *dl, struct devlink_info_req *req,
+		     struct netlink_ext_ack *extack);
+int pdsc_dl_flash_update(struct devlink *dl,
+			 struct devlink_flash_update_params *params,
+			 struct netlink_ext_ack *extack);
+int pdsc_dl_enable_get(struct devlink *dl, u32 id,
+		       struct devlink_param_gset_ctx *ctx);
+int pdsc_dl_enable_set(struct devlink *dl, u32 id,
+		       struct devlink_param_gset_ctx *ctx,
+		       struct netlink_ext_ack *extack);
+int pdsc_dl_enable_validate(struct devlink *dl, u32 id,
+			    union devlink_param_value val,
+			    struct netlink_ext_ack *extack);
+
+void __iomem *pdsc_map_dbpage(struct pdsc *pdsc, int page_num);
+
+void pdsc_debugfs_create(void);
+void pdsc_debugfs_destroy(void);
+void pdsc_debugfs_add_dev(struct pdsc *pdsc);
+void pdsc_debugfs_del_dev(struct pdsc *pdsc);
+void pdsc_debugfs_add_ident(struct pdsc *pdsc);
+void pdsc_debugfs_add_viftype(struct pdsc *pdsc);
+void pdsc_debugfs_add_irqs(struct pdsc *pdsc);
+void pdsc_debugfs_add_qcq(struct pdsc *pdsc, struct pdsc_qcq *qcq);
+void pdsc_debugfs_del_qcq(struct pdsc_qcq *qcq);
+
+int pdsc_err_to_errno(enum pds_core_status_code code);
+bool pdsc_is_fw_running(struct pdsc *pdsc);
+bool pdsc_is_fw_good(struct pdsc *pdsc);
+int pdsc_devcmd(struct pdsc *pdsc, union pds_core_dev_cmd *cmd,
+		union pds_core_dev_comp *comp, int max_seconds);
+int pdsc_devcmd_locked(struct pdsc *pdsc, union pds_core_dev_cmd *cmd,
+		       union pds_core_dev_comp *comp, int max_seconds);
+int pdsc_devcmd_init(struct pdsc *pdsc);
+int pdsc_devcmd_reset(struct pdsc *pdsc);
+int pdsc_dev_init(struct pdsc *pdsc);
+void pdsc_dev_uninit(struct pdsc *pdsc);
+
+int pdsc_intr_alloc(struct pdsc *pdsc, char *name,
+		    irq_handler_t handler, void *data);
+void pdsc_intr_free(struct pdsc *pdsc, int index);
+void pdsc_qcq_free(struct pdsc *pdsc, struct pdsc_qcq *qcq);
+int pdsc_qcq_alloc(struct pdsc *pdsc, unsigned int type, unsigned int index,
+		   const char *name, unsigned int flags, unsigned int num_descs,
+		   unsigned int desc_size, unsigned int cq_desc_size,
+		   unsigned int pid, struct pdsc_qcq *qcq);
+int pdsc_setup(struct pdsc *pdsc, bool init);
+void pdsc_teardown(struct pdsc *pdsc, bool removing);
+int pdsc_start(struct pdsc *pdsc);
+void pdsc_stop(struct pdsc *pdsc);
+void pdsc_health_thread(struct work_struct *work);
+
+int pdsc_register_notify(struct notifier_block *nb);
+void pdsc_unregister_notify(struct notifier_block *nb);
+void pdsc_notify(unsigned long event, void *data);
+int pdsc_auxbus_dev_add(struct pdsc *cf, struct pdsc *pf,
+			enum pds_core_vif_types vt,
+			struct pds_auxiliary_dev **pd_ptr);
+void pdsc_auxbus_dev_del(struct pdsc *cf, struct pdsc *pf,
+			 struct pds_auxiliary_dev **pd_ptr);
+
+void pdsc_process_adminq(struct pdsc_qcq *qcq);
+void pdsc_work_thread(struct work_struct *work);
+irqreturn_t pdsc_adminq_isr(int irq, void *data);
+
+int pdsc_firmware_update(struct pdsc *pdsc, const struct firmware *fw,
+			 struct netlink_ext_ack *extack);
+
+void pdsc_fw_down(struct pdsc *pdsc);
+void pdsc_fw_up(struct pdsc *pdsc);
+void pdsc_pci_reset_thread(struct work_struct *work);
+
+#endif /* _PDSC_H_ */
diff --git a/drivers/net/ethernet/amd/pds_core/debugfs.c b/drivers/net/ethernet/amd/pds_core/debugfs.c
new file mode 100644
index 000000000000..04c5e3abd8d7
--- /dev/null
+++ b/drivers/net/ethernet/amd/pds_core/debugfs.c
@@ -0,0 +1,175 @@
+// SPDX-License-Identifier: GPL-2.0
+/* Copyright(c) 2023 Advanced Micro Devices, Inc */
+
+#include <linux/pci.h>
+
+#include "core.h"
+
+static struct dentry *pdsc_dir;
+
+void pdsc_debugfs_create(void)
+{
+	pdsc_dir = debugfs_create_dir(PDS_CORE_DRV_NAME, NULL);
+}
+
+void pdsc_debugfs_destroy(void)
+{
+	debugfs_remove_recursive(pdsc_dir);
+}
+
+void pdsc_debugfs_add_dev(struct pdsc *pdsc)
+{
+	pdsc->dentry = debugfs_create_dir(pci_name(pdsc->pdev), pdsc_dir);
+
+	debugfs_create_ulong("state", 0400, pdsc->dentry, &pdsc->state);
+}
+
+void pdsc_debugfs_del_dev(struct pdsc *pdsc)
+{
+	debugfs_remove_recursive(pdsc->dentry);
+	pdsc->dentry = NULL;
+}
+
+static int identity_show(struct seq_file *seq, void *v)
+{
+	struct pds_core_dev_identity *ident;
+	struct pdsc *pdsc = seq->private;
+	int vt;
+
+	ident = &pdsc->dev_ident;
+
+	seq_printf(seq, "fw_heartbeat:     0x%x\n",
+		   ioread32(&pdsc->info_regs->fw_heartbeat));
+
+	seq_printf(seq, "nlifs:            %d\n",
+		   le32_to_cpu(ident->nlifs));
+	seq_printf(seq, "nintrs:           %d\n",
+		   le32_to_cpu(ident->nintrs));
+	seq_printf(seq, "ndbpgs_per_lif:   %d\n",
+		   le32_to_cpu(ident->ndbpgs_per_lif));
+	seq_printf(seq, "intr_coal_mult:   %d\n",
+		   le32_to_cpu(ident->intr_coal_mult));
+	seq_printf(seq, "intr_coal_div:    %d\n",
+		   le32_to_cpu(ident->intr_coal_div));
+
+	seq_puts(seq, "vif_types:        ");
+	for (vt = 0; vt < PDS_DEV_TYPE_MAX; vt++)
+		seq_printf(seq, "%d ",
+			   le16_to_cpu(pdsc->dev_ident.vif_types[vt]));
+	seq_puts(seq, "\n");
+
+	return 0;
+}
+DEFINE_SHOW_ATTRIBUTE(identity);
+
+void pdsc_debugfs_add_ident(struct pdsc *pdsc)
+{
+	/* This file will already exist in the reset flow */
+	if (debugfs_lookup("identity", pdsc->dentry))
+		return;
+
+	debugfs_create_file("identity", 0400, pdsc->dentry,
+			    pdsc, &identity_fops);
+}
+
+static int viftype_show(struct seq_file *seq, void *v)
+{
+	struct pdsc *pdsc = seq->private;
+	int vt;
+
+	for (vt = 0; vt < PDS_DEV_TYPE_MAX; vt++) {
+		if (!pdsc->viftype_status[vt].name)
+			continue;
+
+		seq_printf(seq, "%s\t%d supported %d enabled\n",
+			   pdsc->viftype_status[vt].name,
+			   pdsc->viftype_status[vt].supported,
+			   pdsc->viftype_status[vt].enabled);
+	}
+	return 0;
+}
+DEFINE_SHOW_ATTRIBUTE(viftype);
+
+void pdsc_debugfs_add_viftype(struct pdsc *pdsc)
+{
+	debugfs_create_file("viftypes", 0400, pdsc->dentry,
+			    pdsc, &viftype_fops);
+}
+
+static const struct debugfs_reg32 intr_ctrl_regs[] = {
+	{ .name = "coal_init", .offset = 0, },
+	{ .name = "mask", .offset = 4, },
+	{ .name = "credits", .offset = 8, },
+	{ .name = "mask_on_assert", .offset = 12, },
+	{ .name = "coal_timer", .offset = 16, },
+};
+
+void pdsc_debugfs_add_qcq(struct pdsc *pdsc, struct pdsc_qcq *qcq)
+{
+	struct dentry *qcq_dentry, *q_dentry, *cq_dentry, *intr_dentry;
+	struct debugfs_regset32 *intr_ctrl_regset;
+	struct pdsc_queue *q = &qcq->q;
+	struct pdsc_cq *cq = &qcq->cq;
+
+	qcq_dentry = debugfs_create_dir(q->name, pdsc->dentry);
+	if (IS_ERR(qcq_dentry))
+		return;
+	qcq->dentry = qcq_dentry;
+
+	debugfs_create_x64("q_base_pa", 0400, qcq_dentry, &qcq->q_base_pa);
+	debugfs_create_x32("q_size", 0400, qcq_dentry, &qcq->q_size);
+	debugfs_create_x64("cq_base_pa", 0400, qcq_dentry, &qcq->cq_base_pa);
+	debugfs_create_x32("cq_size", 0400, qcq_dentry, &qcq->cq_size);
+	debugfs_create_x32("accum_work", 0400, qcq_dentry, &qcq->accum_work);
+
+	q_dentry = debugfs_create_dir("q", qcq->dentry);
+	if (IS_ERR(q_dentry))
+		return;
+
+	debugfs_create_u32("index", 0400, q_dentry, &q->index);
+	debugfs_create_u32("num_descs", 0400, q_dentry, &q->num_descs);
+	debugfs_create_u32("desc_size", 0400, q_dentry, &q->desc_size);
+	debugfs_create_u32("pid", 0400, q_dentry, &q->pid);
+
+	debugfs_create_u16("tail", 0400, q_dentry, &q->tail_idx);
+	debugfs_create_u16("head", 0400, q_dentry, &q->head_idx);
+
+	cq_dentry = debugfs_create_dir("cq", qcq->dentry);
+	if (IS_ERR(cq_dentry))
+		return;
+
+	debugfs_create_x64("base_pa", 0400, cq_dentry, &cq->base_pa);
+	debugfs_create_u32("num_descs", 0400, cq_dentry, &cq->num_descs);
+	debugfs_create_u32("desc_size", 0400, cq_dentry, &cq->desc_size);
+	debugfs_create_bool("done_color", 0400, cq_dentry, &cq->done_color);
+	debugfs_create_u16("tail", 0400, cq_dentry, &cq->tail_idx);
+
+	if (qcq->flags & PDS_CORE_QCQ_F_INTR) {
+		struct pdsc_intr_info *intr = &pdsc->intr_info[qcq->intx];
+
+		intr_dentry = debugfs_create_dir("intr", qcq->dentry);
+		if (IS_ERR(intr_dentry))
+			return;
+
+		debugfs_create_u32("index", 0400, intr_dentry, &intr->index);
+		debugfs_create_u32("vector", 0400, intr_dentry, &intr->vector);
+
+		intr_ctrl_regset = devm_kzalloc(pdsc->dev,
+						sizeof(*intr_ctrl_regset),
+						GFP_KERNEL);
+		if (!intr_ctrl_regset)
+			return;
+		intr_ctrl_regset->regs = intr_ctrl_regs;
+		intr_ctrl_regset->nregs = ARRAY_SIZE(intr_ctrl_regs);
+		intr_ctrl_regset->base = &pdsc->intr_ctrl[intr->index];
+
+		debugfs_create_regset32("intr_ctrl", 0400, intr_dentry,
+					intr_ctrl_regset);
+	}
+};
+
+void pdsc_debugfs_del_qcq(struct pdsc_qcq *qcq)
+{
+	debugfs_remove_recursive(qcq->dentry);
+	qcq->dentry = NULL;
+}
diff --git a/drivers/net/ethernet/amd/pds_core/dev.c b/drivers/net/ethernet/amd/pds_core/dev.c
new file mode 100644
index 000000000000..495ef4ef8c10
--- /dev/null
+++ b/drivers/net/ethernet/amd/pds_core/dev.c
@@ -0,0 +1,382 @@
+// SPDX-License-Identifier: GPL-2.0
+/* Copyright(c) 2023 Advanced Micro Devices, Inc */
+
+#include <linux/errno.h>
+#include <linux/pci.h>
+#include <linux/utsname.h>
+
+#include "core.h"
+
+int pdsc_err_to_errno(enum pds_core_status_code code)
+{
+	switch (code) {
+	case PDS_RC_SUCCESS:
+		return 0;
+	case PDS_RC_EVERSION:
+	case PDS_RC_EQTYPE:
+	case PDS_RC_EQID:
+	case PDS_RC_EINVAL:
+	case PDS_RC_ENOSUPP:
+		return -EINVAL;
+	case PDS_RC_EPERM:
+		return -EPERM;
+	case PDS_RC_ENOENT:
+		return -ENOENT;
+	case PDS_RC_EAGAIN:
+		return -EAGAIN;
+	case PDS_RC_ENOMEM:
+		return -ENOMEM;
+	case PDS_RC_EFAULT:
+		return -EFAULT;
+	case PDS_RC_EBUSY:
+		return -EBUSY;
+	case PDS_RC_EEXIST:
+		return -EEXIST;
+	case PDS_RC_EVFID:
+		return -ENODEV;
+	case PDS_RC_ECLIENT:
+		return -ECHILD;
+	case PDS_RC_ENOSPC:
+		return -ENOSPC;
+	case PDS_RC_ERANGE:
+		return -ERANGE;
+	case PDS_RC_BAD_ADDR:
+		return -EFAULT;
+	case PDS_RC_BAD_PCI:
+		return -ENXIO;
+	case PDS_RC_EOPCODE:
+	case PDS_RC_EINTR:
+	case PDS_RC_DEV_CMD:
+	case PDS_RC_ERROR:
+	case PDS_RC_ERDMA:
+	case PDS_RC_EIO:
+	default:
+		return -EIO;
+	}
+}
+
+bool pdsc_is_fw_running(struct pdsc *pdsc)
+{
+	if (!pdsc->info_regs)
+		return false;
+
+	pdsc->fw_status = ioread8(&pdsc->info_regs->fw_status);
+	pdsc->last_fw_time = jiffies;
+	pdsc->last_hb = ioread32(&pdsc->info_regs->fw_heartbeat);
+
+	/* Firmware is useful only if the running bit is set and
+	 * fw_status != 0xff (bad PCI read)
+	 */
+	return (pdsc->fw_status != PDS_RC_BAD_PCI) &&
+		(pdsc->fw_status & PDS_CORE_FW_STS_F_RUNNING);
+}
+
+bool pdsc_is_fw_good(struct pdsc *pdsc)
+{
+	bool fw_running = pdsc_is_fw_running(pdsc);
+	u8 gen;
+
+	/* Make sure to update the cached fw_status by calling
+	 * pdsc_is_fw_running() before getting the generation
+	 */
+	gen = pdsc->fw_status & PDS_CORE_FW_STS_F_GENERATION;
+
+	return fw_running && gen == pdsc->fw_generation;
+}
+
+static u8 pdsc_devcmd_status(struct pdsc *pdsc)
+{
+	return ioread8(&pdsc->cmd_regs->comp.status);
+}
+
+static bool pdsc_devcmd_done(struct pdsc *pdsc)
+{
+	return ioread32(&pdsc->cmd_regs->done) & PDS_CORE_DEV_CMD_DONE;
+}
+
+static void pdsc_devcmd_dbell(struct pdsc *pdsc)
+{
+	iowrite32(0, &pdsc->cmd_regs->done);
+	iowrite32(1, &pdsc->cmd_regs->doorbell);
+}
+
+static void pdsc_devcmd_clean(struct pdsc *pdsc)
+{
+	iowrite32(0, &pdsc->cmd_regs->doorbell);
+	memset_io(&pdsc->cmd_regs->cmd, 0, sizeof(pdsc->cmd_regs->cmd));
+}
+
+static const char *pdsc_devcmd_str(int opcode)
+{
+	switch (opcode) {
+	case PDS_CORE_CMD_NOP:
+		return "PDS_CORE_CMD_NOP";
+	case PDS_CORE_CMD_IDENTIFY:
+		return "PDS_CORE_CMD_IDENTIFY";
+	case PDS_CORE_CMD_RESET:
+		return "PDS_CORE_CMD_RESET";
+	case PDS_CORE_CMD_INIT:
+		return "PDS_CORE_CMD_INIT";
+	case PDS_CORE_CMD_FW_DOWNLOAD:
+		return "PDS_CORE_CMD_FW_DOWNLOAD";
+	case PDS_CORE_CMD_FW_CONTROL:
+		return "PDS_CORE_CMD_FW_CONTROL";
+	default:
+		return "PDS_CORE_CMD_UNKNOWN";
+	}
+}
+
+static int pdsc_devcmd_wait(struct pdsc *pdsc, u8 opcode, int max_seconds)
+{
+	struct device *dev = pdsc->dev;
+	unsigned long start_time;
+	unsigned long max_wait;
+	unsigned long duration;
+	int timeout = 0;
+	bool running;
+	int done = 0;
+	int err = 0;
+	int status;
+
+	start_time = jiffies;
+	max_wait = start_time + (max_seconds * HZ);
+
+	while (!done && !timeout) {
+		running = pdsc_is_fw_running(pdsc);
+		if (!running)
+			break;
+
+		done = pdsc_devcmd_done(pdsc);
+		if (done)
+			break;
+
+		timeout = time_after(jiffies, max_wait);
+		if (timeout)
+			break;
+
+		usleep_range(100, 200);
+	}
+	duration = jiffies - start_time;
+
+	if (done && duration > HZ)
+		dev_dbg(dev, "DEVCMD %d %s after %ld secs\n",
+			opcode, pdsc_devcmd_str(opcode), duration / HZ);
+
+	if ((!done || timeout) && running) {
+		dev_err(dev, "DEVCMD %d %s timeout, done %d timeout %d max_seconds=%d\n",
+			opcode, pdsc_devcmd_str(opcode), done, timeout,
+			max_seconds);
+		err = -ETIMEDOUT;
+		pdsc_devcmd_clean(pdsc);
+	}
+
+	status = pdsc_devcmd_status(pdsc);
+	err = pdsc_err_to_errno(status);
+	if (err && err != -EAGAIN)
+		dev_err(dev, "DEVCMD %d %s failed, status=%d err %d %pe\n",
+			opcode, pdsc_devcmd_str(opcode), status, err,
+			ERR_PTR(err));
+
+	return err;
+}
+
+int pdsc_devcmd_locked(struct pdsc *pdsc, union pds_core_dev_cmd *cmd,
+		       union pds_core_dev_comp *comp, int max_seconds)
+{
+	int err;
+
+	if (!pdsc->cmd_regs)
+		return -ENXIO;
+
+	memcpy_toio(&pdsc->cmd_regs->cmd, cmd, sizeof(*cmd));
+	pdsc_devcmd_dbell(pdsc);
+	err = pdsc_devcmd_wait(pdsc, cmd->opcode, max_seconds);
+
+	if ((err == -ENXIO || err == -ETIMEDOUT) && pdsc->wq)
+		queue_work(pdsc->wq, &pdsc->health_work);
+	else
+		memcpy_fromio(comp, &pdsc->cmd_regs->comp, sizeof(*comp));
+
+	return err;
+}
+
+int pdsc_devcmd(struct pdsc *pdsc, union pds_core_dev_cmd *cmd,
+		union pds_core_dev_comp *comp, int max_seconds)
+{
+	int err;
+
+	mutex_lock(&pdsc->devcmd_lock);
+	err = pdsc_devcmd_locked(pdsc, cmd, comp, max_seconds);
+	mutex_unlock(&pdsc->devcmd_lock);
+
+	return err;
+}
+
+int pdsc_devcmd_init(struct pdsc *pdsc)
+{
+	union pds_core_dev_comp comp = {};
+	union pds_core_dev_cmd cmd = {
+		.opcode = PDS_CORE_CMD_INIT,
+	};
+
+	return pdsc_devcmd(pdsc, &cmd, &comp, pdsc->devcmd_timeout);
+}
+
+int pdsc_devcmd_reset(struct pdsc *pdsc)
+{
+	union pds_core_dev_comp comp = {};
+	union pds_core_dev_cmd cmd = {
+		.reset.opcode = PDS_CORE_CMD_RESET,
+	};
+
+	if (!pdsc_is_fw_running(pdsc))
+		return 0;
+
+	return pdsc_devcmd(pdsc, &cmd, &comp, pdsc->devcmd_timeout);
+}
+
+static int pdsc_devcmd_identify_locked(struct pdsc *pdsc)
+{
+	union pds_core_dev_comp comp = {};
+	union pds_core_dev_cmd cmd = {
+		.identify.opcode = PDS_CORE_CMD_IDENTIFY,
+		.identify.ver = PDS_CORE_IDENTITY_VERSION_1,
+	};
+
+	return pdsc_devcmd_locked(pdsc, &cmd, &comp, pdsc->devcmd_timeout);
+}
+
+static void pdsc_init_devinfo(struct pdsc *pdsc)
+{
+	pdsc->dev_info.asic_type = ioread8(&pdsc->info_regs->asic_type);
+	pdsc->dev_info.asic_rev = ioread8(&pdsc->info_regs->asic_rev);
+	pdsc->fw_generation = PDS_CORE_FW_STS_F_GENERATION &
+			      ioread8(&pdsc->info_regs->fw_status);
+
+	memcpy_fromio(pdsc->dev_info.fw_version,
+		      pdsc->info_regs->fw_version,
+		      PDS_CORE_DEVINFO_FWVERS_BUFLEN);
+	pdsc->dev_info.fw_version[PDS_CORE_DEVINFO_FWVERS_BUFLEN] = 0;
+
+	memcpy_fromio(pdsc->dev_info.serial_num,
+		      pdsc->info_regs->serial_num,
+		      PDS_CORE_DEVINFO_SERIAL_BUFLEN);
+	pdsc->dev_info.serial_num[PDS_CORE_DEVINFO_SERIAL_BUFLEN] = 0;
+
+	dev_dbg(pdsc->dev, "fw_version %s\n", pdsc->dev_info.fw_version);
+}
+
+static int pdsc_identify(struct pdsc *pdsc)
+{
+	struct pds_core_drv_identity drv = {};
+	size_t sz;
+	int err;
+	int n;
+
+	drv.drv_type = cpu_to_le32(PDS_DRIVER_LINUX);
+	/* Catching the return quiets a Wformat-truncation complaint */
+	n = snprintf(drv.driver_ver_str, sizeof(drv.driver_ver_str),
+		     "%s %s", PDS_CORE_DRV_NAME, utsname()->release);
+	if (n > sizeof(drv.driver_ver_str))
+		dev_dbg(pdsc->dev, "release name truncated, don't care\n");
+
+	/* Next let's get some info about the device
+	 * We use the devcmd_lock at this level in order to
+	 * get safe access to the cmd_regs->data before anyone
+	 * else can mess it up
+	 */
+	mutex_lock(&pdsc->devcmd_lock);
+
+	sz = min_t(size_t, sizeof(drv), sizeof(pdsc->cmd_regs->data));
+	memcpy_toio(&pdsc->cmd_regs->data, &drv, sz);
+
+	err = pdsc_devcmd_identify_locked(pdsc);
+	if (!err) {
+		sz = min_t(size_t, sizeof(pdsc->dev_ident),
+			   sizeof(pdsc->cmd_regs->data));
+		memcpy_fromio(&pdsc->dev_ident, &pdsc->cmd_regs->data, sz);
+	}
+	mutex_unlock(&pdsc->devcmd_lock);
+
+	if (err) {
+		dev_err(pdsc->dev, "Cannot identify device: %pe\n",
+			ERR_PTR(err));
+		return err;
+	}
+
+	if (isprint(pdsc->dev_info.fw_version[0]) &&
+	    isascii(pdsc->dev_info.fw_version[0]))
+		dev_info(pdsc->dev, "FW: %.*s\n",
+			 (int)(sizeof(pdsc->dev_info.fw_version) - 1),
+			 pdsc->dev_info.fw_version);
+	else
+		dev_info(pdsc->dev, "FW: (invalid string) 0x%02x 0x%02x 0x%02x 0x%02x ...\n",
+			 (u8)pdsc->dev_info.fw_version[0],
+			 (u8)pdsc->dev_info.fw_version[1],
+			 (u8)pdsc->dev_info.fw_version[2],
+			 (u8)pdsc->dev_info.fw_version[3]);
+
+	return 0;
+}
+
+void pdsc_dev_uninit(struct pdsc *pdsc)
+{
+	if (pdsc->intr_info) {
+		int i;
+
+		for (i = 0; i < pdsc->nintrs; i++)
+			pdsc_intr_free(pdsc, i);
+
+		kfree(pdsc->intr_info);
+		pdsc->intr_info = NULL;
+		pdsc->nintrs = 0;
+	}
+
+	pci_free_irq_vectors(pdsc->pdev);
+}
+
+int pdsc_dev_init(struct pdsc *pdsc)
+{
+	unsigned int nintrs;
+	int err;
+
+	/* Initial init and reset of device */
+	pdsc_init_devinfo(pdsc);
+	pdsc->devcmd_timeout = PDS_CORE_DEVCMD_TIMEOUT;
+
+	err = pdsc_devcmd_reset(pdsc);
+	if (err)
+		return err;
+
+	err = pdsc_identify(pdsc);
+	if (err)
+		return err;
+
+	pdsc_debugfs_add_ident(pdsc);
+
+	/* Now we can reserve interrupts */
+	nintrs = le32_to_cpu(pdsc->dev_ident.nintrs);
+	nintrs = min_t(unsigned int, num_online_cpus(), nintrs);
+
+	/* Get intr_info struct array for tracking */
+	pdsc->intr_info = kcalloc(nintrs, sizeof(*pdsc->intr_info), GFP_KERNEL);
+	if (!pdsc->intr_info)
+		return -ENOMEM;
+
+	err = pci_alloc_irq_vectors(pdsc->pdev, nintrs, nintrs, PCI_IRQ_MSIX);
+	if (err != nintrs) {
+		dev_err(pdsc->dev, "Can't get %d intrs from OS: %pe\n",
+			nintrs, ERR_PTR(err));
+		err = -ENOSPC;
+		goto err_out;
+	}
+	pdsc->nintrs = nintrs;
+
+	return 0;
+
+err_out:
+	kfree(pdsc->intr_info);
+	pdsc->intr_info = NULL;
+
+	return err;
+}
diff --git a/drivers/net/ethernet/amd/pds_core/devlink.c b/drivers/net/ethernet/amd/pds_core/devlink.c
new file mode 100644
index 000000000000..d8dc39da4161
--- /dev/null
+++ b/drivers/net/ethernet/amd/pds_core/devlink.c
@@ -0,0 +1,178 @@
+// SPDX-License-Identifier: GPL-2.0
+/* Copyright(c) 2023 Advanced Micro Devices, Inc */
+
+#include "core.h"
+#include <linux/pds/pds_auxbus.h>
+
+static struct
+pdsc_viftype *pdsc_dl_find_viftype_by_id(struct pdsc *pdsc,
+					 enum devlink_param_type dl_id)
+{
+	int vt;
+
+	if (!pdsc->viftype_status)
+		return NULL;
+
+	for (vt = 0; vt < PDS_DEV_TYPE_MAX; vt++) {
+		if (pdsc->viftype_status[vt].dl_id == dl_id)
+			return &pdsc->viftype_status[vt];
+	}
+
+	return NULL;
+}
+
+int pdsc_dl_enable_get(struct devlink *dl, u32 id,
+		       struct devlink_param_gset_ctx *ctx)
+{
+	struct pdsc *pdsc = devlink_priv(dl);
+	struct pdsc_viftype *vt_entry;
+
+	vt_entry = pdsc_dl_find_viftype_by_id(pdsc, id);
+	if (!vt_entry)
+		return -ENOENT;
+
+	ctx->val.vbool = vt_entry->enabled;
+
+	return 0;
+}
+
+int pdsc_dl_enable_set(struct devlink *dl, u32 id,
+		       struct devlink_param_gset_ctx *ctx,
+		       struct netlink_ext_ack *extack)
+{
+	struct pdsc *pdsc = devlink_priv(dl);
+	struct pdsc_viftype *vt_entry;
+	int err = 0;
+	int vf_id;
+
+	vt_entry = pdsc_dl_find_viftype_by_id(pdsc, id);
+	if (!vt_entry || !vt_entry->supported)
+		return -EOPNOTSUPP;
+
+	if (vt_entry->enabled == ctx->val.vbool)
+		return 0;
+
+	vt_entry->enabled = ctx->val.vbool;
+	for (vf_id = 0; vf_id < pdsc->num_vfs; vf_id++) {
+		struct pdsc *vf = pdsc->vfs[vf_id].vf;
+
+		if (ctx->val.vbool)
+			err = pdsc_auxbus_dev_add(vf, pdsc, vt_entry->vif_id,
+						  &pdsc->vfs[vf_id].padev);
+		else
+			pdsc_auxbus_dev_del(vf, pdsc, &pdsc->vfs[vf_id].padev);
+	}
+
+	return err;
+}
+
+int pdsc_dl_enable_validate(struct devlink *dl, u32 id,
+			    union devlink_param_value val,
+			    struct netlink_ext_ack *extack)
+{
+	struct pdsc *pdsc = devlink_priv(dl);
+	struct pdsc_viftype *vt_entry;
+
+	vt_entry = pdsc_dl_find_viftype_by_id(pdsc, id);
+	if (!vt_entry || !vt_entry->supported)
+		return -EOPNOTSUPP;
+
+	if (!pdsc->viftype_status[vt_entry->vif_id].supported)
+		return -ENODEV;
+
+	return 0;
+}
+
+int pdsc_dl_flash_update(struct devlink *dl,
+			 struct devlink_flash_update_params *params,
+			 struct netlink_ext_ack *extack)
+{
+	struct pdsc *pdsc = devlink_priv(dl);
+
+	return pdsc_firmware_update(pdsc, params->fw, extack);
+}
+
+static char *fw_slotnames[] = {
+	"fw.goldfw",
+	"fw.mainfwa",
+	"fw.mainfwb",
+};
+
+int pdsc_dl_info_get(struct devlink *dl, struct devlink_info_req *req,
+		     struct netlink_ext_ack *extack)
+{
+	union pds_core_dev_cmd cmd = {
+		.fw_control.opcode = PDS_CORE_CMD_FW_CONTROL,
+		.fw_control.oper = PDS_CORE_FW_GET_LIST,
+	};
+	struct pds_core_fw_list_info fw_list = {};
+	struct pdsc *pdsc = devlink_priv(dl);
+	union pds_core_dev_comp comp;
+	char buf[32];
+	int listlen;
+	int err;
+	int i;
+
+	mutex_lock(&pdsc->devcmd_lock);
+	err = pdsc_devcmd_locked(pdsc, &cmd, &comp, pdsc->devcmd_timeout * 2);
+	if (!err)
+		memcpy_fromio(&fw_list, pdsc->cmd_regs->data, sizeof(fw_list));
+	mutex_unlock(&pdsc->devcmd_lock);
+
+	listlen = min(fw_list.num_fw_slots, ARRAY_SIZE(fw_list.fw_names));
+	for (i = 0; i < listlen; i++) {
+		if (i < ARRAY_SIZE(fw_slotnames))
+			strscpy(buf, fw_slotnames[i], sizeof(buf));
+		else
+			snprintf(buf, sizeof(buf), "fw.slot_%d", i);
+		err = devlink_info_version_stored_put(req, buf,
+						      fw_list.fw_names[i].fw_version);
+		if (err)
+			return err;
+	}
+
+	err = devlink_info_version_running_put(req,
+					       DEVLINK_INFO_VERSION_GENERIC_FW,
+					       pdsc->dev_info.fw_version);
+	if (err)
+		return err;
+
+	snprintf(buf, sizeof(buf), "0x%x", pdsc->dev_info.asic_type);
+	err = devlink_info_version_fixed_put(req,
+					     DEVLINK_INFO_VERSION_GENERIC_ASIC_ID,
+					     buf);
+	if (err)
+		return err;
+
+	snprintf(buf, sizeof(buf), "0x%x", pdsc->dev_info.asic_rev);
+	err = devlink_info_version_fixed_put(req,
+					     DEVLINK_INFO_VERSION_GENERIC_ASIC_REV,
+					     buf);
+	if (err)
+		return err;
+
+	return devlink_info_serial_number_put(req, pdsc->dev_info.serial_num);
+}
+
+int pdsc_fw_reporter_diagnose(struct devlink_health_reporter *reporter,
+			      struct devlink_fmsg *fmsg,
+			      struct netlink_ext_ack *extack)
+{
+	struct pdsc *pdsc = devlink_health_reporter_priv(reporter);
+
+	mutex_lock(&pdsc->config_lock);
+	if (test_bit(PDSC_S_FW_DEAD, &pdsc->state))
+		devlink_fmsg_string_pair_put(fmsg, "Status", "dead");
+	else if (!pdsc_is_fw_good(pdsc))
+		devlink_fmsg_string_pair_put(fmsg, "Status", "unhealthy");
+	else
+		devlink_fmsg_string_pair_put(fmsg, "Status", "healthy");
+	mutex_unlock(&pdsc->config_lock);
+
+	devlink_fmsg_u32_pair_put(fmsg, "State",
+				  pdsc->fw_status & ~PDS_CORE_FW_STS_F_GENERATION);
+	devlink_fmsg_u32_pair_put(fmsg, "Generation", pdsc->fw_generation >> 4);
+	devlink_fmsg_u32_pair_put(fmsg, "Recoveries", pdsc->fw_recoveries);
+
+	return 0;
+}
diff --git a/drivers/net/ethernet/amd/pds_core/fw.c b/drivers/net/ethernet/amd/pds_core/fw.c
new file mode 100644
index 000000000000..fa626719e68d
--- /dev/null
+++ b/drivers/net/ethernet/amd/pds_core/fw.c
@@ -0,0 +1,197 @@
+// SPDX-License-Identifier: GPL-2.0
+/* Copyright(c) 2023 Advanced Micro Devices, Inc */
+
+#include "core.h"
+
+/* The worst case wait for the install activity is about 25 minutes when
+ * installing a new CPLD, which is very seldom.  Normal is about 30-35
+ * seconds.  Since the driver can't tell if a CPLD update will happen we
+ * set the timeout for the ugly case.
+ */
+#define PDSC_FW_INSTALL_TIMEOUT	(25 * 60)
+#define PDSC_FW_SELECT_TIMEOUT	30
+
+/* Number of periodic log updates during fw file download */
+#define PDSC_FW_INTERVAL_FRACTION	32
+
+static int pdsc_devcmd_fw_download_locked(struct pdsc *pdsc, u64 addr,
+					  u32 offset, u32 length)
+{
+	union pds_core_dev_cmd cmd = {
+		.fw_download.opcode = PDS_CORE_CMD_FW_DOWNLOAD,
+		.fw_download.offset = cpu_to_le32(offset),
+		.fw_download.addr = cpu_to_le64(addr),
+		.fw_download.length = cpu_to_le32(length),
+	};
+	union pds_core_dev_comp comp;
+
+	return pdsc_devcmd_locked(pdsc, &cmd, &comp, pdsc->devcmd_timeout);
+}
+
+static int pdsc_devcmd_fw_install(struct pdsc *pdsc)
+{
+	union pds_core_dev_cmd cmd = {
+		.fw_control.opcode = PDS_CORE_CMD_FW_CONTROL,
+		.fw_control.oper = PDS_CORE_FW_INSTALL_ASYNC
+	};
+	union pds_core_dev_comp comp;
+	int err;
+
+	err = pdsc_devcmd(pdsc, &cmd, &comp, pdsc->devcmd_timeout);
+	if (err < 0)
+		return err;
+
+	return comp.fw_control.slot;
+}
+
+static int pdsc_devcmd_fw_activate(struct pdsc *pdsc,
+				   enum pds_core_fw_slot slot)
+{
+	union pds_core_dev_cmd cmd = {
+		.fw_control.opcode = PDS_CORE_CMD_FW_CONTROL,
+		.fw_control.oper = PDS_CORE_FW_ACTIVATE_ASYNC,
+		.fw_control.slot = slot
+	};
+	union pds_core_dev_comp comp;
+
+	return pdsc_devcmd(pdsc, &cmd, &comp, pdsc->devcmd_timeout);
+}
+
+static int pdsc_fw_status_long_wait(struct pdsc *pdsc,
+				    const char *label,
+				    unsigned long timeout,
+				    u8 fw_cmd,
+				    struct netlink_ext_ack *extack)
+{
+	union pds_core_dev_cmd cmd = {
+		.fw_control.opcode = PDS_CORE_CMD_FW_CONTROL,
+		.fw_control.oper = fw_cmd,
+	};
+	union pds_core_dev_comp comp;
+	unsigned long start_time;
+	unsigned long end_time;
+	int err;
+
+	/* Ping on the status of the long running async install
+	 * command.  We get EAGAIN while the command is still
+	 * running, else we get the final command status.
+	 */
+	start_time = jiffies;
+	end_time = start_time + (timeout * HZ);
+	do {
+		err = pdsc_devcmd(pdsc, &cmd, &comp, pdsc->devcmd_timeout);
+		msleep(20);
+	} while (time_before(jiffies, end_time) &&
+		 (err == -EAGAIN || err == -ETIMEDOUT));
+
+	if (err == -EAGAIN || err == -ETIMEDOUT) {
+		NL_SET_ERR_MSG_MOD(extack, "Firmware wait timed out");
+		dev_err(pdsc->dev, "DEV_CMD firmware wait %s timed out\n",
+			label);
+	} else if (err) {
+		NL_SET_ERR_MSG_MOD(extack, "Firmware wait failed");
+	}
+
+	return err;
+}
+
+int pdsc_firmware_update(struct pdsc *pdsc, const struct firmware *fw,
+			 struct netlink_ext_ack *extack)
+{
+	u32 buf_sz, copy_sz, offset;
+	struct devlink *dl;
+	int next_interval;
+	u64 data_addr;
+	int err = 0;
+	int fw_slot;
+
+	dev_info(pdsc->dev, "Installing firmware\n");
+
+	if (!pdsc->cmd_regs)
+		return -ENXIO;
+
+	dl = priv_to_devlink(pdsc);
+	devlink_flash_update_status_notify(dl, "Preparing to flash",
+					   NULL, 0, 0);
+
+	buf_sz = sizeof(pdsc->cmd_regs->data);
+
+	dev_dbg(pdsc->dev,
+		"downloading firmware - size %d part_sz %d nparts %lu\n",
+		(int)fw->size, buf_sz, DIV_ROUND_UP(fw->size, buf_sz));
+
+	offset = 0;
+	next_interval = 0;
+	data_addr = offsetof(struct pds_core_dev_cmd_regs, data);
+	while (offset < fw->size) {
+		if (offset >= next_interval) {
+			devlink_flash_update_status_notify(dl, "Downloading",
+							   NULL, offset,
+							   fw->size);
+			next_interval = offset +
+					(fw->size / PDSC_FW_INTERVAL_FRACTION);
+		}
+
+		copy_sz = min_t(unsigned int, buf_sz, fw->size - offset);
+		mutex_lock(&pdsc->devcmd_lock);
+		memcpy_toio(&pdsc->cmd_regs->data, fw->data + offset, copy_sz);
+		err = pdsc_devcmd_fw_download_locked(pdsc, data_addr,
+						     offset, copy_sz);
+		mutex_unlock(&pdsc->devcmd_lock);
+		if (err) {
+			dev_err(pdsc->dev,
+				"download failed offset 0x%x addr 0x%llx len 0x%x: %pe\n",
+				offset, data_addr, copy_sz, ERR_PTR(err));
+			NL_SET_ERR_MSG_MOD(extack, "Segment download failed");
+			goto err_out;
+		}
+		offset += copy_sz;
+	}
+	devlink_flash_update_status_notify(dl, "Downloading", NULL,
+					   fw->size, fw->size);
+
+	devlink_flash_update_timeout_notify(dl, "Installing", NULL,
+					    PDSC_FW_INSTALL_TIMEOUT);
+
+	fw_slot = pdsc_devcmd_fw_install(pdsc);
+	if (fw_slot < 0) {
+		err = fw_slot;
+		dev_err(pdsc->dev, "install failed: %pe\n", ERR_PTR(err));
+		NL_SET_ERR_MSG_MOD(extack, "Failed to start firmware install");
+		goto err_out;
+	}
+
+	err = pdsc_fw_status_long_wait(pdsc, "Installing",
+				       PDSC_FW_INSTALL_TIMEOUT,
+				       PDS_CORE_FW_INSTALL_STATUS,
+				       extack);
+	if (err)
+		goto err_out;
+
+	devlink_flash_update_timeout_notify(dl, "Selecting", NULL,
+					    PDSC_FW_SELECT_TIMEOUT);
+
+	err = pdsc_devcmd_fw_activate(pdsc, fw_slot);
+	if (err) {
+		NL_SET_ERR_MSG_MOD(extack, "Failed to start firmware select");
+		goto err_out;
+	}
+
+	err = pdsc_fw_status_long_wait(pdsc, "Selecting",
+				       PDSC_FW_SELECT_TIMEOUT,
+				       PDS_CORE_FW_ACTIVATE_STATUS,
+				       extack);
+	if (err)
+		goto err_out;
+
+	dev_info(pdsc->dev, "Firmware update completed, slot %d\n", fw_slot);
+
+err_out:
+	if (err)
+		devlink_flash_update_status_notify(dl, "Flash failed",
+						   NULL, 0, 0);
+	else
+		devlink_flash_update_status_notify(dl, "Flash done",
+						   NULL, 0, 0);
+	return err;
+}
diff --git a/drivers/net/ethernet/amd/pds_core/main.c b/drivers/net/ethernet/amd/pds_core/main.c
new file mode 100644
index 000000000000..c7a2eff57632
--- /dev/null
+++ b/drivers/net/ethernet/amd/pds_core/main.c
@@ -0,0 +1,609 @@
+// SPDX-License-Identifier: GPL-2.0
+/* Copyright(c) 2023 Advanced Micro Devices, Inc */
+
+#define pr_fmt(fmt) KBUILD_MODNAME ": " fmt
+
+#include <linux/pci.h>
+
+#include <linux/pds/pds_common.h>
+
+#include "core.h"
+
+MODULE_DESCRIPTION(PDSC_DRV_DESCRIPTION);
+MODULE_AUTHOR("Advanced Micro Devices, Inc");
+MODULE_LICENSE("GPL");
+
+/* Supported devices */
+static const struct pci_device_id pdsc_id_table[] = {
+	{ PCI_VDEVICE(PENSANDO, PCI_DEVICE_ID_PENSANDO_CORE_PF) },
+	{ PCI_VDEVICE(PENSANDO, PCI_DEVICE_ID_PENSANDO_VDPA_VF) },
+	{ 0, }	/* end of table */
+};
+MODULE_DEVICE_TABLE(pci, pdsc_id_table);
+
+static void pdsc_wdtimer_cb(struct timer_list *t)
+{
+	struct pdsc *pdsc = timer_container_of(pdsc, t, wdtimer);
+
+	dev_dbg(pdsc->dev, "%s: jiffies %ld\n", __func__, jiffies);
+	mod_timer(&pdsc->wdtimer,
+		  round_jiffies(jiffies + pdsc->wdtimer_period));
+
+	queue_work(pdsc->wq, &pdsc->health_work);
+}
+
+static void pdsc_unmap_bars(struct pdsc *pdsc)
+{
+	struct pdsc_dev_bar *bars = pdsc->bars;
+	unsigned int i;
+
+	pdsc->info_regs = NULL;
+	pdsc->cmd_regs = NULL;
+	pdsc->intr_status = NULL;
+	pdsc->intr_ctrl = NULL;
+
+	for (i = 0; i < PDS_CORE_BARS_MAX; i++) {
+		if (bars[i].vaddr)
+			pci_iounmap(pdsc->pdev, bars[i].vaddr);
+		bars[i].vaddr = NULL;
+	}
+}
+
+static int pdsc_map_bars(struct pdsc *pdsc)
+{
+	struct pdsc_dev_bar *bar = pdsc->bars;
+	struct pci_dev *pdev = pdsc->pdev;
+	struct device *dev = pdsc->dev;
+	struct pdsc_dev_bar *bars;
+	unsigned int i, j;
+	int num_bars = 0;
+	int err;
+	u32 sig;
+
+	bars = pdsc->bars;
+
+	/* Since the PCI interface in the hardware is configurable,
+	 * we need to poke into all the bars to find the set we're
+	 * expecting.
+	 */
+	for (i = 0, j = 0; i < PDS_CORE_BARS_MAX; i++) {
+		if (!(pci_resource_flags(pdev, i) & IORESOURCE_MEM))
+			continue;
+
+		bars[j].len = pci_resource_len(pdev, i);
+		bars[j].bus_addr = pci_resource_start(pdev, i);
+		bars[j].res_index = i;
+
+		/* only map the whole bar 0 */
+		if (j > 0) {
+			bars[j].vaddr = NULL;
+		} else {
+			bars[j].vaddr = pci_iomap(pdev, i, bars[j].len);
+			if (!bars[j].vaddr) {
+				dev_err(dev, "Cannot map BAR %d, aborting\n", i);
+				return -ENODEV;
+			}
+		}
+
+		j++;
+	}
+	num_bars = j;
+
+	/* BAR0: dev_cmd and interrupts */
+	if (num_bars < 1) {
+		dev_err(dev, "No bars found\n");
+		err = -EFAULT;
+		goto err_out;
+	}
+
+	if (bar->len < PDS_CORE_BAR0_SIZE) {
+		dev_err(dev, "Resource bar size %lu too small\n", bar->len);
+		err = -EFAULT;
+		goto err_out;
+	}
+
+	pdsc->info_regs = bar->vaddr + PDS_CORE_BAR0_DEV_INFO_REGS_OFFSET;
+	pdsc->cmd_regs = bar->vaddr + PDS_CORE_BAR0_DEV_CMD_REGS_OFFSET;
+	pdsc->intr_status = bar->vaddr + PDS_CORE_BAR0_INTR_STATUS_OFFSET;
+	pdsc->intr_ctrl = bar->vaddr + PDS_CORE_BAR0_INTR_CTRL_OFFSET;
+
+	sig = ioread32(&pdsc->info_regs->signature);
+	if (sig != PDS_CORE_DEV_INFO_SIGNATURE) {
+		dev_err(dev, "Incompatible firmware signature %x", sig);
+		err = -EFAULT;
+		goto err_out;
+	}
+
+	/* BAR1: doorbells */
+	bar++;
+	if (num_bars < 2) {
+		dev_err(dev, "Doorbell bar missing\n");
+		err = -EFAULT;
+		goto err_out;
+	}
+
+	pdsc->db_pages = bar->vaddr;
+	pdsc->phy_db_pages = bar->bus_addr;
+
+	return 0;
+
+err_out:
+	pdsc_unmap_bars(pdsc);
+	return err;
+}
+
+void __iomem *pdsc_map_dbpage(struct pdsc *pdsc, int page_num)
+{
+	return pci_iomap_range(pdsc->pdev,
+			       pdsc->bars[PDS_CORE_PCI_BAR_DBELL].res_index,
+			       (u64)page_num << PAGE_SHIFT, PAGE_SIZE);
+}
+
+static int pdsc_sriov_configure(struct pci_dev *pdev, int num_vfs)
+{
+	struct pdsc *pdsc = pci_get_drvdata(pdev);
+	struct device *dev = pdsc->dev;
+	int ret = 0;
+
+	if (num_vfs > 0) {
+		pdsc->vfs = kcalloc(num_vfs, sizeof(struct pdsc_vf),
+				    GFP_KERNEL);
+		if (!pdsc->vfs)
+			return -ENOMEM;
+		pdsc->num_vfs = num_vfs;
+
+		ret = pci_enable_sriov(pdev, num_vfs);
+		if (ret) {
+			dev_err(dev, "Cannot enable SRIOV: %pe\n",
+				ERR_PTR(ret));
+			goto no_vfs;
+		}
+
+		return num_vfs;
+	}
+
+no_vfs:
+	pci_disable_sriov(pdev);
+
+	kfree(pdsc->vfs);
+	pdsc->vfs = NULL;
+	pdsc->num_vfs = 0;
+
+	return ret;
+}
+
+static int pdsc_init_vf(struct pdsc *vf)
+{
+	struct devlink *dl;
+	struct pdsc *pf;
+	int err;
+
+	pf = pdsc_get_pf_struct(vf->pdev);
+	if (IS_ERR_OR_NULL(pf))
+		return PTR_ERR(pf) ?: -1;
+
+	vf->vf_id = pci_iov_vf_id(vf->pdev);
+
+	dl = priv_to_devlink(vf);
+	devl_lock(dl);
+	devl_register(dl);
+	devl_unlock(dl);
+
+	pf->vfs[vf->vf_id].vf = vf;
+	err = pdsc_auxbus_dev_add(vf, pf, PDS_DEV_TYPE_VDPA,
+				  &pf->vfs[vf->vf_id].padev);
+	if (err) {
+		devl_lock(dl);
+		devl_unregister(dl);
+		devl_unlock(dl);
+	}
+
+	return err;
+}
+
+static const struct devlink_health_reporter_ops pdsc_fw_reporter_ops = {
+	.name = "fw",
+	.diagnose = pdsc_fw_reporter_diagnose,
+};
+
+static const struct devlink_param pdsc_dl_params[] = {
+	DEVLINK_PARAM_GENERIC(ENABLE_VNET,
+			      BIT(DEVLINK_PARAM_CMODE_RUNTIME),
+			      pdsc_dl_enable_get,
+			      pdsc_dl_enable_set,
+			      pdsc_dl_enable_validate),
+};
+
+#define PDSC_WQ_NAME_LEN 24
+
+static int pdsc_init_pf(struct pdsc *pdsc)
+{
+	struct devlink_health_reporter *hr;
+	char wq_name[PDSC_WQ_NAME_LEN];
+	struct devlink *dl;
+	int err;
+
+	pcie_print_link_status(pdsc->pdev);
+
+	err = pci_request_regions(pdsc->pdev, PDS_CORE_DRV_NAME);
+	if (err) {
+		dev_err(pdsc->dev, "Cannot request PCI regions: %pe\n",
+			ERR_PTR(err));
+		return err;
+	}
+
+	err = pdsc_map_bars(pdsc);
+	if (err)
+		goto err_out_release_regions;
+
+	/* General workqueue and timer, but don't start timer yet */
+	snprintf(wq_name, sizeof(wq_name), "%s.%d", PDS_CORE_DRV_NAME, pdsc->uid);
+	pdsc->wq = create_singlethread_workqueue(wq_name);
+	INIT_WORK(&pdsc->health_work, pdsc_health_thread);
+	INIT_WORK(&pdsc->pci_reset_work, pdsc_pci_reset_thread);
+	timer_setup(&pdsc->wdtimer, pdsc_wdtimer_cb, 0);
+	pdsc->wdtimer_period = PDSC_WATCHDOG_SECS * HZ;
+
+	mutex_init(&pdsc->devcmd_lock);
+	mutex_init(&pdsc->config_lock);
+	spin_lock_init(&pdsc->adminq_lock);
+
+	mutex_lock(&pdsc->config_lock);
+	set_bit(PDSC_S_FW_DEAD, &pdsc->state);
+
+	err = pdsc_setup(pdsc, PDSC_SETUP_INIT);
+	if (err) {
+		mutex_unlock(&pdsc->config_lock);
+		goto err_out_unmap_bars;
+	}
+
+	err = pdsc_start(pdsc);
+	if (err) {
+		mutex_unlock(&pdsc->config_lock);
+		goto err_out_teardown;
+	}
+
+	mutex_unlock(&pdsc->config_lock);
+
+	err = pdsc_auxbus_dev_add(pdsc, pdsc, PDS_DEV_TYPE_FWCTL, &pdsc->padev);
+	if (err)
+		goto err_out_stop;
+
+	dl = priv_to_devlink(pdsc);
+	devl_lock(dl);
+	err = devl_params_register(dl, pdsc_dl_params,
+				   ARRAY_SIZE(pdsc_dl_params));
+	if (err) {
+		devl_unlock(dl);
+		dev_warn(pdsc->dev, "Failed to register devlink params: %pe\n",
+			 ERR_PTR(err));
+		goto err_out_del_dev;
+	}
+
+	hr = devl_health_reporter_create(dl, &pdsc_fw_reporter_ops, pdsc);
+	if (IS_ERR(hr)) {
+		devl_unlock(dl);
+		dev_warn(pdsc->dev, "Failed to create fw reporter: %pe\n", hr);
+		err = PTR_ERR(hr);
+		goto err_out_unreg_params;
+	}
+	pdsc->fw_reporter = hr;
+
+	devl_register(dl);
+	devl_unlock(dl);
+
+	/* Lastly, start the health check timer */
+	mod_timer(&pdsc->wdtimer, round_jiffies(jiffies + pdsc->wdtimer_period));
+
+	return 0;
+
+err_out_unreg_params:
+	devlink_params_unregister(dl, pdsc_dl_params,
+				  ARRAY_SIZE(pdsc_dl_params));
+err_out_del_dev:
+	pdsc_auxbus_dev_del(pdsc, pdsc, &pdsc->padev);
+err_out_stop:
+	pdsc_stop(pdsc);
+err_out_teardown:
+	pdsc_teardown(pdsc, PDSC_TEARDOWN_REMOVING);
+err_out_unmap_bars:
+	timer_shutdown_sync(&pdsc->wdtimer);
+	if (pdsc->wq)
+		destroy_workqueue(pdsc->wq);
+	mutex_destroy(&pdsc->config_lock);
+	mutex_destroy(&pdsc->devcmd_lock);
+	pci_free_irq_vectors(pdsc->pdev);
+	pdsc_unmap_bars(pdsc);
+err_out_release_regions:
+	pci_release_regions(pdsc->pdev);
+
+	return err;
+}
+
+static const struct devlink_ops pdsc_dl_ops = {
+	.info_get	= pdsc_dl_info_get,
+	.flash_update	= pdsc_dl_flash_update,
+};
+
+static const struct devlink_ops pdsc_dl_vf_ops = {
+};
+
+static DEFINE_IDA(pdsc_ida);
+
+static int pdsc_probe(struct pci_dev *pdev, const struct pci_device_id *ent)
+{
+	struct device *dev = &pdev->dev;
+	const struct devlink_ops *ops;
+	struct devlink *dl;
+	struct pdsc *pdsc;
+	bool is_pf;
+	int err;
+
+	is_pf = !pdev->is_virtfn;
+	ops = is_pf ? &pdsc_dl_ops : &pdsc_dl_vf_ops;
+	dl = devlink_alloc(ops, sizeof(struct pdsc), dev);
+	if (!dl)
+		return -ENOMEM;
+	pdsc = devlink_priv(dl);
+
+	pdsc->pdev = pdev;
+	pdsc->dev = &pdev->dev;
+	set_bit(PDSC_S_INITING_DRIVER, &pdsc->state);
+	pci_set_drvdata(pdev, pdsc);
+	pdsc_debugfs_add_dev(pdsc);
+
+	err = ida_alloc(&pdsc_ida, GFP_KERNEL);
+	if (err < 0) {
+		dev_err(pdsc->dev, "%s: id alloc failed: %pe\n",
+			__func__, ERR_PTR(err));
+		goto err_out_free_devlink;
+	}
+	pdsc->uid = err;
+
+	/* Query system for DMA addressing limitation for the device. */
+	err = dma_set_mask_and_coherent(dev, DMA_BIT_MASK(PDS_CORE_ADDR_LEN));
+	if (err) {
+		dev_err(dev, "Unable to obtain 64-bit DMA for consistent allocations, aborting: %pe\n",
+			ERR_PTR(err));
+		goto err_out_free_ida;
+	}
+
+	err = pci_enable_device(pdev);
+	if (err) {
+		dev_err(dev, "Cannot enable PCI device: %pe\n", ERR_PTR(err));
+		goto err_out_free_ida;
+	}
+	pci_set_master(pdev);
+
+	if (is_pf)
+		err = pdsc_init_pf(pdsc);
+	else
+		err = pdsc_init_vf(pdsc);
+	if (err) {
+		dev_err(dev, "Cannot init device: %pe\n", ERR_PTR(err));
+		goto err_out_disable_device;
+	}
+
+	clear_bit(PDSC_S_INITING_DRIVER, &pdsc->state);
+	return 0;
+
+err_out_disable_device:
+	pci_disable_device(pdev);
+err_out_free_ida:
+	ida_free(&pdsc_ida, pdsc->uid);
+err_out_free_devlink:
+	pdsc_debugfs_del_dev(pdsc);
+	devlink_free(dl);
+
+	return err;
+}
+
+static void pdsc_remove(struct pci_dev *pdev)
+{
+	struct pdsc *pdsc = pci_get_drvdata(pdev);
+	struct devlink *dl;
+
+	/* Unhook the registrations first to be sure there
+	 * are no requests while we're stopping.
+	 */
+	dl = priv_to_devlink(pdsc);
+	devl_lock(dl);
+	devl_unregister(dl);
+	if (!pdev->is_virtfn) {
+		if (pdsc->fw_reporter) {
+			devl_health_reporter_destroy(pdsc->fw_reporter);
+			pdsc->fw_reporter = NULL;
+		}
+		devl_params_unregister(dl, pdsc_dl_params,
+				       ARRAY_SIZE(pdsc_dl_params));
+	}
+	devl_unlock(dl);
+
+	if (pdev->is_virtfn) {
+		struct pdsc *pf;
+
+		pf = pdsc_get_pf_struct(pdsc->pdev);
+		if (!IS_ERR(pf)) {
+			pdsc_auxbus_dev_del(pdsc, pf, &pf->vfs[pdsc->vf_id].padev);
+			pf->vfs[pdsc->vf_id].vf = NULL;
+		}
+	} else {
+		/* Remove the VFs and their aux_bus connections before other
+		 * cleanup so that the clients can use the AdminQ to cleanly
+		 * shut themselves down.
+		 */
+		pdsc_sriov_configure(pdev, 0);
+		pdsc_auxbus_dev_del(pdsc, pdsc, &pdsc->padev);
+
+		timer_shutdown_sync(&pdsc->wdtimer);
+		if (pdsc->wq)
+			destroy_workqueue(pdsc->wq);
+
+		mutex_lock(&pdsc->config_lock);
+		set_bit(PDSC_S_STOPPING_DRIVER, &pdsc->state);
+
+		pdsc_stop(pdsc);
+		pdsc_teardown(pdsc, PDSC_TEARDOWN_REMOVING);
+		mutex_unlock(&pdsc->config_lock);
+		mutex_destroy(&pdsc->config_lock);
+		mutex_destroy(&pdsc->devcmd_lock);
+
+		pdsc_unmap_bars(pdsc);
+		pci_release_regions(pdev);
+	}
+
+	pci_disable_device(pdev);
+
+	ida_free(&pdsc_ida, pdsc->uid);
+	pdsc_debugfs_del_dev(pdsc);
+	devlink_free(dl);
+}
+
+static void pdsc_stop_health_thread(struct pdsc *pdsc)
+{
+	if (pdsc->pdev->is_virtfn)
+		return;
+
+	timer_shutdown_sync(&pdsc->wdtimer);
+	if (pdsc->health_work.func)
+		cancel_work_sync(&pdsc->health_work);
+}
+
+static void pdsc_restart_health_thread(struct pdsc *pdsc)
+{
+	if (pdsc->pdev->is_virtfn)
+		return;
+
+	timer_setup(&pdsc->wdtimer, pdsc_wdtimer_cb, 0);
+	mod_timer(&pdsc->wdtimer, jiffies + 1);
+}
+
+static void pdsc_reset_prepare(struct pci_dev *pdev)
+{
+	struct pdsc *pdsc = pci_get_drvdata(pdev);
+
+	pdsc_stop_health_thread(pdsc);
+	pdsc_fw_down(pdsc);
+
+	if (pdev->is_virtfn) {
+		struct pdsc *pf;
+
+		pf = pdsc_get_pf_struct(pdsc->pdev);
+		if (!IS_ERR(pf))
+			pdsc_auxbus_dev_del(pdsc, pf,
+					    &pf->vfs[pdsc->vf_id].padev);
+	} else {
+		pdsc_auxbus_dev_del(pdsc, pdsc, &pdsc->padev);
+	}
+
+	pdsc_unmap_bars(pdsc);
+	pci_release_regions(pdev);
+	if (pci_is_enabled(pdev))
+		pci_disable_device(pdev);
+}
+
+static void pdsc_reset_done(struct pci_dev *pdev)
+{
+	struct pdsc *pdsc = pci_get_drvdata(pdev);
+	struct device *dev = pdsc->dev;
+	int err;
+
+	err = pci_enable_device(pdev);
+	if (err) {
+		dev_err(dev, "Cannot enable PCI device: %pe\n", ERR_PTR(err));
+		return;
+	}
+	pci_set_master(pdev);
+
+	if (!pdev->is_virtfn) {
+		pcie_print_link_status(pdsc->pdev);
+
+		err = pci_request_regions(pdsc->pdev, PDS_CORE_DRV_NAME);
+		if (err) {
+			dev_err(pdsc->dev, "Cannot request PCI regions: %pe\n",
+				ERR_PTR(err));
+			return;
+		}
+
+		err = pdsc_map_bars(pdsc);
+		if (err)
+			return;
+	}
+
+	pdsc_fw_up(pdsc);
+	pdsc_restart_health_thread(pdsc);
+
+	if (pdev->is_virtfn) {
+		struct pdsc *pf;
+
+		pf = pdsc_get_pf_struct(pdsc->pdev);
+		if (!IS_ERR(pf))
+			pdsc_auxbus_dev_add(pdsc, pf, PDS_DEV_TYPE_VDPA,
+					    &pf->vfs[pdsc->vf_id].padev);
+	} else {
+		pdsc_auxbus_dev_add(pdsc, pdsc, PDS_DEV_TYPE_FWCTL,
+				    &pdsc->padev);
+	}
+}
+
+static pci_ers_result_t pdsc_pci_error_detected(struct pci_dev *pdev,
+						pci_channel_state_t error)
+{
+	if (error == pci_channel_io_frozen) {
+		pdsc_reset_prepare(pdev);
+		return PCI_ERS_RESULT_NEED_RESET;
+	}
+
+	return PCI_ERS_RESULT_NONE;
+}
+
+static void pdsc_pci_error_resume(struct pci_dev *pdev)
+{
+	struct pdsc *pdsc = pci_get_drvdata(pdev);
+
+	if (test_bit(PDSC_S_FW_DEAD, &pdsc->state))
+		pci_reset_function_locked(pdev);
+}
+
+static const struct pci_error_handlers pdsc_err_handler = {
+	/* FLR handling */
+	.reset_prepare      = pdsc_reset_prepare,
+	.reset_done         = pdsc_reset_done,
+
+	/* AER handling */
+	.error_detected     = pdsc_pci_error_detected,
+	.resume             = pdsc_pci_error_resume,
+};
+
+static struct pci_driver pdsc_driver = {
+	.name = PDS_CORE_DRV_NAME,
+	.id_table = pdsc_id_table,
+	.probe = pdsc_probe,
+	.remove = pdsc_remove,
+	.sriov_configure = pdsc_sriov_configure,
+	.err_handler = &pdsc_err_handler,
+};
+
+void *pdsc_get_pf_struct(struct pci_dev *vf_pdev)
+{
+	return pci_iov_get_pf_drvdata(vf_pdev, &pdsc_driver);
+}
+EXPORT_SYMBOL_GPL(pdsc_get_pf_struct);
+
+static int __init pdsc_init_module(void)
+{
+	if (strcmp(KBUILD_MODNAME, PDS_CORE_DRV_NAME))
+		return -EINVAL;
+
+	pdsc_debugfs_create();
+	return pci_register_driver(&pdsc_driver);
+}
+
+static void __exit pdsc_cleanup_module(void)
+{
+	pci_unregister_driver(&pdsc_driver);
+	pdsc_debugfs_destroy();
+}
+
+module_init(pdsc_init_module);
+module_exit(pdsc_cleanup_module);
diff --git a/drivers/net/ethernet/amd/sun3lance.c b/drivers/net/ethernet/amd/sun3lance.c
index 77b1db267730..c60df4a21158 100644
--- a/drivers/net/ethernet/amd/sun3lance.c
+++ b/drivers/net/ethernet/amd/sun3lance.c
@@ -37,12 +37,12 @@ static const char version[] =
 #include <linux/etherdevice.h>
 #include <linux/skbuff.h>
 #include <linux/bitops.h>
+#include <linux/pgtable.h>
 
 #include <asm/cacheflush.h>
 #include <asm/setup.h>
 #include <asm/irq.h>
 #include <asm/io.h>
-#include <asm/pgtable.h>
 #include <asm/dvma.h>
 #include <asm/idprom.h>
 #include <asm/machines.h>
@@ -74,6 +74,7 @@ static int lance_debug = 1;
 #endif
 module_param(lance_debug, int, 0);
 MODULE_PARM_DESC(lance_debug, "SUN3 Lance debug level (0-3)");
+MODULE_DESCRIPTION("Sun3/Sun3x on-board LANCE Ethernet driver");
 MODULE_LICENSE("GPL");
 
 #define	DPRINTK(n,a) \
@@ -150,7 +151,7 @@ struct lance_memory {
 struct lance_private {
 	volatile unsigned short	*iobase;
 	struct lance_memory	*mem;
-     	int new_rx, new_tx;	/* The next free ring entry */
+	int new_rx, new_tx;	/* The next free ring entry */
 	int old_tx, old_rx;     /* ring entry to be processed */
 /* These two must be longs for set_bit() */
 	long	    tx_full;
@@ -236,7 +237,8 @@ struct lance_private {
 static int lance_probe( struct net_device *dev);
 static int lance_open( struct net_device *dev );
 static void lance_init_ring( struct net_device *dev );
-static int lance_start_xmit( struct sk_buff *skb, struct net_device *dev );
+static netdev_tx_t lance_start_xmit(struct sk_buff *skb,
+				    struct net_device *dev);
 static irqreturn_t lance_interrupt( int irq, void *dev_id);
 static int lance_rx( struct net_device *dev );
 static int lance_close( struct net_device *dev );
@@ -244,7 +246,7 @@ static void set_multicast_list( struct net_device *dev );
 
 /************************* End of Prototypes **************************/
 
-struct net_device * __init sun3lance_probe(int unit)
+static struct net_device * __init sun3lance_probe(void)
 {
 	struct net_device *dev;
 	static int found;
@@ -271,10 +273,6 @@ struct net_device * __init sun3lance_probe(int unit)
 	dev = alloc_etherdev(sizeof(struct lance_private));
 	if (!dev)
 		return ERR_PTR(-ENOMEM);
-	if (unit >= 0) {
-		sprintf(dev->name, "eth%d", unit);
-		netdev_boot_setup_check(dev);
-	}
 
 	if (!lance_probe(dev))
 		goto out;
@@ -308,7 +306,6 @@ static int __init lance_probe( struct net_device *dev)
 	unsigned long ioaddr;
 
 	struct lance_private	*lp;
-	int 			i;
 	static int 		did_version;
 	volatile unsigned short *ioaddr_probe;
 	unsigned short tmp1, tmp2;
@@ -345,7 +342,7 @@ static int __init lance_probe( struct net_device *dev)
 
 	/* XXX - leak? */
 	MEM = dvma_malloc_align(sizeof(struct lance_memory), 0x10000);
-	if (MEM == NULL) {
+	if (!MEM) {
 #ifdef CONFIG_SUN3
 		iounmap((void __iomem *)ioaddr);
 #endif
@@ -376,8 +373,7 @@ static int __init lance_probe( struct net_device *dev)
 		   dev->irq);
 
 	/* copy in the ethernet address from the prom */
-	for(i = 0; i < 6 ; i++)
-	     dev->dev_addr[i] = idprom->id_ethaddr[i];
+	eth_hw_addr_set(dev, idprom->id_ethaddr);
 
 	/* tell the card it's ether address, bytes swapped */
 	MEM->init.hwaddr[0] = dev->dev_addr[1];
@@ -464,7 +460,7 @@ static void lance_init_ring( struct net_device *dev )
 	for( i = 0; i < TX_RING_SIZE; i++ ) {
 		MEM->tx_head[i].base = dvma_vtob(MEM->tx_data[i]);
 		MEM->tx_head[i].flag = 0;
- 		MEM->tx_head[i].base_hi =
+		MEM->tx_head[i].base_hi =
 			(dvma_vtob(MEM->tx_data[i])) >>16;
 		MEM->tx_head[i].length = 0;
 		MEM->tx_head[i].misc = 0;
@@ -511,7 +507,8 @@ static void lance_init_ring( struct net_device *dev )
 }
 
 
-static int lance_start_xmit( struct sk_buff *skb, struct net_device *dev )
+static netdev_tx_t
+lance_start_xmit(struct sk_buff *skb, struct net_device *dev)
 {
 	struct lance_private *lp = netdev_priv(dev);
 	int entry, len;
@@ -579,8 +576,8 @@ static int lance_start_xmit( struct sk_buff *skb, struct net_device *dev )
 	}
 
 	AREG = CSR0;
-  	DPRINTK( 2, ( "%s: lance_start_xmit() called, csr0 %4.4x.\n",
-  				  dev->name, DREG ));
+	DPRINTK( 2, ( "%s: lance_start_xmit() called, csr0 %4.4x.\n",
+				  dev->name, DREG ));
 
 #ifdef CONFIG_SUN3X
 	/* this weirdness doesn't appear on sun3... */
@@ -634,8 +631,8 @@ static int lance_start_xmit( struct sk_buff *skb, struct net_device *dev )
 	/* Trigger an immediate send poll. */
 	REGA(CSR0) = CSR0_INEA | CSR0_TDMD | CSR0_STRT;
 	AREG = CSR0;
-  	DPRINTK( 2, ( "%s: lance_start_xmit() exiting, csr0 %4.4x.\n",
-  				  dev->name, DREG ));
+	DPRINTK( 2, ( "%s: lance_start_xmit() exiting, csr0 %4.4x.\n",
+				  dev->name, DREG ));
 	dev_kfree_skb(skb);
 
 	lp->lock = 0;
@@ -655,16 +652,6 @@ static irqreturn_t lance_interrupt( int irq, void *dev_id)
 	struct net_device *dev = dev_id;
 	struct lance_private *lp = netdev_priv(dev);
 	int csr0;
-	static int in_interrupt;
-
-	if (dev == NULL) {
-		DPRINTK( 1, ( "lance_interrupt(): invalid dev_id\n" ));
-		return IRQ_NONE;
-	}
-
-	if (in_interrupt)
-		DPRINTK( 2, ( "%s: Re-entering the interrupt handler.\n", dev->name ));
-	in_interrupt = 1;
 
  still_more:
 	flush_cache_all();
@@ -772,7 +759,6 @@ static irqreturn_t lance_interrupt( int irq, void *dev_id)
 
 	DPRINTK( 2, ( "%s: exiting interrupt, csr0=%#04x.\n",
 				  dev->name, DREG ));
-	in_interrupt = 0;
 	return IRQ_HANDLED;
 }
 
@@ -811,7 +797,7 @@ static int lance_rx( struct net_device *dev )
 			}
 			else {
 				skb = netdev_alloc_skb(dev, pkt_len + 2);
-				if (skb == NULL) {
+				if (!skb) {
 					dev->stats.rx_dropped++;
 					head->msg_length = 0;
 					head->flag |= RMD1_OWN_CHIP;
@@ -933,17 +919,16 @@ static void set_multicast_list( struct net_device *dev )
 }
 
 
-#ifdef MODULE
-
 static struct net_device *sun3lance_dev;
 
-int __init init_module(void)
+static int __init sun3lance_init(void)
 {
-	sun3lance_dev = sun3lance_probe(-1);
+	sun3lance_dev = sun3lance_probe();
 	return PTR_ERR_OR_ZERO(sun3lance_dev);
 }
+module_init(sun3lance_init);
 
-void __exit cleanup_module(void)
+static void __exit sun3lance_cleanup(void)
 {
 	unregister_netdev(sun3lance_dev);
 #ifdef CONFIG_SUN3
@@ -951,6 +936,4 @@ void __exit cleanup_module(void)
 #endif
 	free_netdev(sun3lance_dev);
 }
-
-#endif /* MODULE */
-
+module_exit(sun3lance_cleanup);
diff --git a/drivers/net/ethernet/amd/sunlance.c b/drivers/net/ethernet/amd/sunlance.c
index 291ca5187f12..0b273327f5a6 100644
--- a/drivers/net/ethernet/amd/sunlance.c
+++ b/drivers/net/ethernet/amd/sunlance.c
@@ -1,3 +1,4 @@
+// SPDX-License-Identifier: GPL-2.0-only
 /* $Id: sunlance.c,v 1.112 2002/01/15 06:48:55 davem Exp $
  * lance.c: Linux/Sparc/Lance driver
  *
@@ -91,12 +92,12 @@ static char lancestr[] = "LANCE";
 #include <linux/bitops.h>
 #include <linux/dma-mapping.h>
 #include <linux/of.h>
-#include <linux/of_device.h>
+#include <linux/platform_device.h>
 #include <linux/gfp.h>
+#include <linux/pgtable.h>
 
 #include <asm/io.h>
 #include <asm/dma.h>
-#include <asm/pgtable.h>
 #include <asm/byteorder.h>	/* Used by the checksum routines */
 #include <asm/idprom.h>
 #include <asm/prom.h>
@@ -104,14 +105,9 @@ static char lancestr[] = "LANCE";
 #include <asm/irq.h>
 
 #define DRV_NAME	"sunlance"
-#define DRV_VERSION	"2.02"
 #define DRV_RELDATE	"8/24/03"
 #define DRV_AUTHOR	"Miguel de Icaza (miguel@nuclecu.unam.mx)"
 
-static char version[] =
-	DRV_NAME ".c:v" DRV_VERSION " " DRV_RELDATE " " DRV_AUTHOR "\n";
-
-MODULE_VERSION(DRV_VERSION);
 MODULE_AUTHOR(DRV_AUTHOR);
 MODULE_DESCRIPTION("Sun Lance ethernet driver");
 MODULE_LICENSE("GPL");
@@ -534,7 +530,7 @@ static void lance_rx_dvma(struct net_device *dev)
 			len = (rd->mblength & 0xfff) - 4;
 			skb = netdev_alloc_skb(dev, len + 2);
 
-			if (skb == NULL) {
+			if (!skb) {
 				dev->stats.rx_dropped++;
 				rd->mblength = 0;
 				rd->rmd1_bits = LE_R1_OWN;
@@ -704,7 +700,7 @@ static void lance_rx_pio(struct net_device *dev)
 			len = (sbus_readw(&rd->mblength) & 0xfff) - 4;
 			skb = netdev_alloc_skb(dev, len + 2);
 
-			if (skb == NULL) {
+			if (!skb) {
 				dev->stats.rx_dropped++;
 				sbus_writew(0, &rd->mblength);
 				sbus_writeb(LE_R1_OWN, &rd->rmd1_bits);
@@ -967,7 +963,7 @@ static int lance_close(struct net_device *dev)
 	struct lance_private *lp = netdev_priv(dev);
 
 	netif_stop_queue(dev);
-	del_timer_sync(&lp->multicast_timer);
+	timer_delete_sync(&lp->multicast_timer);
 
 	STOP_LANCE(lp);
 
@@ -1096,7 +1092,7 @@ static void lance_piozero(void __iomem *dest, int len)
 		sbus_writeb(0, piobuf);
 }
 
-static void lance_tx_timeout(struct net_device *dev)
+static void lance_tx_timeout(struct net_device *dev, unsigned int txqueue)
 {
 	struct lance_private *lp = netdev_priv(dev);
 
@@ -1106,7 +1102,7 @@ static void lance_tx_timeout(struct net_device *dev)
 	netif_wake_queue(dev);
 }
 
-static int lance_start_xmit(struct sk_buff *skb, struct net_device *dev)
+static netdev_tx_t lance_start_xmit(struct sk_buff *skb, struct net_device *dev)
 {
 	struct lance_private *lp = netdev_priv(dev);
 	int entry, skblen, len;
@@ -1248,9 +1244,10 @@ static void lance_set_multicast(struct net_device *dev)
 	netif_wake_queue(dev);
 }
 
-static void lance_set_multicast_retry(unsigned long _opaque)
+static void lance_set_multicast_retry(struct timer_list *t)
 {
-	struct net_device *dev = (struct net_device *) _opaque;
+	struct lance_private *lp = timer_container_of(lp, t, multicast_timer);
+	struct net_device *dev = lp->dev;
 
 	lance_set_multicast(dev);
 }
@@ -1279,8 +1276,7 @@ static void lance_free_hwresources(struct lance_private *lp)
 /* Ethtool support... */
 static void sparc_lance_get_drvinfo(struct net_device *dev, struct ethtool_drvinfo *info)
 {
-	strlcpy(info->driver, "sunlance", sizeof(info->driver));
-	strlcpy(info->version, "2.02", sizeof(info->version));
+	strscpy(info->driver, "sunlance", sizeof(info->driver));
 }
 
 static const struct ethtool_ops sparc_lance_ethtool_ops = {
@@ -1303,10 +1299,8 @@ static int sparc_lance_probe_one(struct platform_device *op,
 				 struct platform_device *lebuffer)
 {
 	struct device_node *dp = op->dev.of_node;
-	static unsigned version_printed;
 	struct lance_private *lp;
 	struct net_device *dev;
-	int    i;
 
 	dev = alloc_etherdev(sizeof(struct lance_private) + 8);
 	if (!dev)
@@ -1314,17 +1308,13 @@ static int sparc_lance_probe_one(struct platform_device *op,
 
 	lp = netdev_priv(dev);
 
-	if (sparc_lance_debug && version_printed++ == 0)
-		printk (KERN_INFO "%s", version);
-
 	spin_lock_init(&lp->lock);
 
 	/* Copy the IDPROM ethernet address to the device structure, later we
 	 * will copy the address in the device structure to the lance
 	 * initialization block.
 	 */
-	for (i = 0; i < 6; i++)
-		dev->dev_addr[i] = idprom->id_ethaddr[i];
+	eth_hw_addr_set(dev, idprom->id_ethaddr);
 
 	/* Get the IO region */
 	lp->lregs = of_ioremap(&op->resource[0], 0,
@@ -1418,7 +1408,7 @@ static int sparc_lance_probe_one(struct platform_device *op,
 
 			prop = of_get_property(nd, "tpe-link-test?", NULL);
 			if (!prop)
-				goto no_link_test;
+				goto node_put;
 
 			if (strcmp(prop, "true")) {
 				printk(KERN_NOTICE "SunLance: warning: overriding option "
@@ -1427,6 +1417,8 @@ static int sparc_lance_probe_one(struct platform_device *op,
 				       "to ecd@skynet.be\n");
 				auxio_set_lte(AUXIO_LTE_ON);
 			}
+node_put:
+			of_node_put(nd);
 no_link_test:
 			lp->auto_select = 1;
 			lp->tpe = 0;
@@ -1459,9 +1451,7 @@ no_link_test:
 	 * can occur from interrupts (ex. IPv6).  So we
 	 * use a timer to try again later when necessary. -DaveM
 	 */
-	init_timer(&lp->multicast_timer);
-	lp->multicast_timer.data = (unsigned long) dev;
-	lp->multicast_timer.function = lance_set_multicast_retry;
+	timer_setup(&lp->multicast_timer, lance_set_multicast_retry, 0);
 
 	if (register_netdev(dev)) {
 		printk(KERN_ERR "SunLance: Cannot register device.\n");
@@ -1487,9 +1477,9 @@ static int sunlance_sbus_probe(struct platform_device *op)
 	struct device_node *parent_dp = parent->dev.of_node;
 	int err;
 
-	if (!strcmp(parent_dp->name, "ledma")) {
+	if (of_node_name_eq(parent_dp, "ledma")) {
 		err = sparc_lance_probe_one(op, parent, NULL);
-	} else if (!strcmp(parent_dp->name, "lebuffer")) {
+	} else if (of_node_name_eq(parent_dp, "lebuffer")) {
 		err = sparc_lance_probe_one(op, NULL, parent);
 	} else
 		err = sparc_lance_probe_one(op, NULL, NULL);
@@ -1497,7 +1487,7 @@ static int sunlance_sbus_probe(struct platform_device *op)
 	return err;
 }
 
-static int sunlance_sbus_remove(struct platform_device *op)
+static void sunlance_sbus_remove(struct platform_device *op)
 {
 	struct lance_private *lp = platform_get_drvdata(op);
 	struct net_device *net_dev = lp->dev;
@@ -1507,8 +1497,6 @@ static int sunlance_sbus_remove(struct platform_device *op)
 	lance_free_hwresources(lp);
 
 	free_netdev(net_dev);
-
-	return 0;
 }
 
 static const struct of_device_id sunlance_sbus_match[] = {
diff --git a/drivers/net/ethernet/amd/xgbe/Makefile b/drivers/net/ethernet/amd/xgbe/Makefile
index 0dea8f5da899..980e27652237 100644
--- a/drivers/net/ethernet/amd/xgbe/Makefile
+++ b/drivers/net/ethernet/amd/xgbe/Makefile
@@ -1,8 +1,9 @@
+# SPDX-License-Identifier: GPL-2.0
 obj-$(CONFIG_AMD_XGBE) += amd-xgbe.o
 
 amd-xgbe-objs := xgbe-main.o xgbe-drv.o xgbe-dev.o \
 		 xgbe-desc.o xgbe-ethtool.o xgbe-mdio.o \
-		 xgbe-ptp.o \
+		 xgbe-hwtstamp.o xgbe-ptp.o xgbe-pps.o \
 		 xgbe-i2c.o xgbe-phy-v1.o xgbe-phy-v2.o \
 		 xgbe-platform.o
 
diff --git a/drivers/net/ethernet/amd/xgbe/xgbe-common.h b/drivers/net/ethernet/amd/xgbe/xgbe-common.h
index 127adbeefb10..62b01de93db4 100644
--- a/drivers/net/ethernet/amd/xgbe/xgbe-common.h
+++ b/drivers/net/ethernet/amd/xgbe/xgbe-common.h
@@ -1,117 +1,8 @@
+// SPDX-License-Identifier: (GPL-2.0-or-later OR BSD-3-Clause)
 /*
- * AMD 10Gb Ethernet driver
- *
- * This file is available to you under your choice of the following two
- * licenses:
- *
- * License 1: GPLv2
- *
- * Copyright (c) 2014-2016 Advanced Micro Devices, Inc.
- *
- * This file is free software; you may copy, redistribute and/or modify
- * it under the terms of the GNU General Public License as published by
- * the Free Software Foundation, either version 2 of the License, or (at
- * your option) any later version.
- *
- * This file is distributed in the hope that it will be useful, but
- * WITHOUT ANY WARRANTY; without even the implied warranty of
- * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
- * General Public License for more details.
- *
- * You should have received a copy of the GNU General Public License
- * along with this program.  If not, see <http://www.gnu.org/licenses/>.
- *
- * This file incorporates work covered by the following copyright and
- * permission notice:
- *     The Synopsys DWC ETHER XGMAC Software Driver and documentation
- *     (hereinafter "Software") is an unsupported proprietary work of Synopsys,
- *     Inc. unless otherwise expressly agreed to in writing between Synopsys
- *     and you.
- *
- *     The Software IS NOT an item of Licensed Software or Licensed Product
- *     under any End User Software License Agreement or Agreement for Licensed
- *     Product with Synopsys or any supplement thereto.  Permission is hereby
- *     granted, free of charge, to any person obtaining a copy of this software
- *     annotated with this license and the Software, to deal in the Software
- *     without restriction, including without limitation the rights to use,
- *     copy, modify, merge, publish, distribute, sublicense, and/or sell copies
- *     of the Software, and to permit persons to whom the Software is furnished
- *     to do so, subject to the following conditions:
- *
- *     The above copyright notice and this permission notice shall be included
- *     in all copies or substantial portions of the Software.
- *
- *     THIS SOFTWARE IS BEING DISTRIBUTED BY SYNOPSYS SOLELY ON AN "AS IS"
- *     BASIS AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED
- *     TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A
- *     PARTICULAR PURPOSE ARE HEREBY DISCLAIMED. IN NO EVENT SHALL SYNOPSYS
- *     BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR
- *     CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF
- *     SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS
- *     INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN
- *     CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE)
- *     ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF
- *     THE POSSIBILITY OF SUCH DAMAGE.
- *
- *
- * License 2: Modified BSD
- *
- * Copyright (c) 2014-2016 Advanced Micro Devices, Inc.
- * All rights reserved.
- *
- * Redistribution and use in source and binary forms, with or without
- * modification, are permitted provided that the following conditions are met:
- *     * Redistributions of source code must retain the above copyright
- *       notice, this list of conditions and the following disclaimer.
- *     * Redistributions in binary form must reproduce the above copyright
- *       notice, this list of conditions and the following disclaimer in the
- *       documentation and/or other materials provided with the distribution.
- *     * Neither the name of Advanced Micro Devices, Inc. nor the
- *       names of its contributors may be used to endorse or promote products
- *       derived from this software without specific prior written permission.
- *
- * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS"
- * AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
- * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
- * ARE DISCLAIMED. IN NO EVENT SHALL <COPYRIGHT HOLDER> BE LIABLE FOR ANY
- * DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES
- * (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES;
- * LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND
- * ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
- * (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF
- * THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
- *
- * This file incorporates work covered by the following copyright and
- * permission notice:
- *     The Synopsys DWC ETHER XGMAC Software Driver and documentation
- *     (hereinafter "Software") is an unsupported proprietary work of Synopsys,
- *     Inc. unless otherwise expressly agreed to in writing between Synopsys
- *     and you.
- *
- *     The Software IS NOT an item of Licensed Software or Licensed Product
- *     under any End User Software License Agreement or Agreement for Licensed
- *     Product with Synopsys or any supplement thereto.  Permission is hereby
- *     granted, free of charge, to any person obtaining a copy of this software
- *     annotated with this license and the Software, to deal in the Software
- *     without restriction, including without limitation the rights to use,
- *     copy, modify, merge, publish, distribute, sublicense, and/or sell copies
- *     of the Software, and to permit persons to whom the Software is furnished
- *     to do so, subject to the following conditions:
- *
- *     The above copyright notice and this permission notice shall be included
- *     in all copies or substantial portions of the Software.
- *
- *     THIS SOFTWARE IS BEING DISTRIBUTED BY SYNOPSYS SOLELY ON AN "AS IS"
- *     BASIS AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED
- *     TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A
- *     PARTICULAR PURPOSE ARE HEREBY DISCLAIMED. IN NO EVENT SHALL SYNOPSYS
- *     BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR
- *     CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF
- *     SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS
- *     INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN
- *     CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE)
- *     ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF
- *     THE POSSIBILITY OF SUCH DAMAGE.
+ * Copyright (c) 2014-2025, Advanced Micro Devices, Inc.
+ * Copyright (c) 2014, Synopsys, Inc.
+ * All rights reserved
  */
 
 #ifndef __XGBE_COMMON_H__
@@ -123,38 +14,13 @@
 #define DMA_ISR				0x3008
 #define DMA_AXIARCR			0x3010
 #define DMA_AXIAWCR			0x3018
+#define DMA_AXIAWARCR			0x301c
 #define DMA_DSR0			0x3020
 #define DMA_DSR1			0x3024
+#define DMA_TXEDMACR			0x3040
+#define DMA_RXEDMACR			0x3044
 
 /* DMA register entry bit positions and sizes */
-#define DMA_AXIARCR_DRC_INDEX		0
-#define DMA_AXIARCR_DRC_WIDTH		4
-#define DMA_AXIARCR_DRD_INDEX		4
-#define DMA_AXIARCR_DRD_WIDTH		2
-#define DMA_AXIARCR_TEC_INDEX		8
-#define DMA_AXIARCR_TEC_WIDTH		4
-#define DMA_AXIARCR_TED_INDEX		12
-#define DMA_AXIARCR_TED_WIDTH		2
-#define DMA_AXIARCR_THC_INDEX		16
-#define DMA_AXIARCR_THC_WIDTH		4
-#define DMA_AXIARCR_THD_INDEX		20
-#define DMA_AXIARCR_THD_WIDTH		2
-#define DMA_AXIAWCR_DWC_INDEX		0
-#define DMA_AXIAWCR_DWC_WIDTH		4
-#define DMA_AXIAWCR_DWD_INDEX		4
-#define DMA_AXIAWCR_DWD_WIDTH		2
-#define DMA_AXIAWCR_RPC_INDEX		8
-#define DMA_AXIAWCR_RPC_WIDTH		4
-#define DMA_AXIAWCR_RPD_INDEX		12
-#define DMA_AXIAWCR_RPD_WIDTH		2
-#define DMA_AXIAWCR_RHC_INDEX		16
-#define DMA_AXIAWCR_RHC_WIDTH		4
-#define DMA_AXIAWCR_RHD_INDEX		20
-#define DMA_AXIAWCR_RHD_WIDTH		2
-#define DMA_AXIAWCR_TDC_INDEX		24
-#define DMA_AXIAWCR_TDC_WIDTH		4
-#define DMA_AXIAWCR_TDD_INDEX		28
-#define DMA_AXIAWCR_TDD_WIDTH		2
 #define DMA_ISR_MACIS_INDEX		17
 #define DMA_ISR_MACIS_WIDTH		1
 #define DMA_ISR_MTLIS_INDEX		16
@@ -163,14 +29,31 @@
 #define DMA_MR_INTM_WIDTH		2
 #define DMA_MR_SWR_INDEX		0
 #define DMA_MR_SWR_WIDTH		1
+#define DMA_RXEDMACR_RDPS_INDEX		0
+#define DMA_RXEDMACR_RDPS_WIDTH		3
+#define DMA_SBMR_AAL_INDEX		12
+#define DMA_SBMR_AAL_WIDTH		1
 #define DMA_SBMR_EAME_INDEX		11
 #define DMA_SBMR_EAME_WIDTH		1
-#define DMA_SBMR_BLEN_256_INDEX		7
-#define DMA_SBMR_BLEN_256_WIDTH		1
+#define DMA_SBMR_BLEN_INDEX		1
+#define DMA_SBMR_BLEN_WIDTH		7
+#define DMA_SBMR_RD_OSR_LMT_INDEX	16
+#define DMA_SBMR_RD_OSR_LMT_WIDTH	6
 #define DMA_SBMR_UNDEF_INDEX		0
 #define DMA_SBMR_UNDEF_WIDTH		1
+#define DMA_SBMR_WR_OSR_LMT_INDEX	24
+#define DMA_SBMR_WR_OSR_LMT_WIDTH	6
+#define DMA_TXEDMACR_TDPS_INDEX		0
+#define DMA_TXEDMACR_TDPS_WIDTH		3
 
 /* DMA register values */
+#define DMA_SBMR_BLEN_256		256
+#define DMA_SBMR_BLEN_128		128
+#define DMA_SBMR_BLEN_64		64
+#define DMA_SBMR_BLEN_32		32
+#define DMA_SBMR_BLEN_16		16
+#define DMA_SBMR_BLEN_8			8
+#define DMA_SBMR_BLEN_4			4
 #define DMA_DSR_RPS_WIDTH		4
 #define DMA_DSR_TPS_WIDTH		4
 #define DMA_DSR_Q_WIDTH			(DMA_DSR_RPS_WIDTH + DMA_DSR_TPS_WIDTH)
@@ -218,11 +101,15 @@
 #define DMA_CH_CR_PBLX8_WIDTH		1
 #define DMA_CH_CR_SPH_INDEX		24
 #define DMA_CH_CR_SPH_WIDTH		1
-#define DMA_CH_IER_AIE_INDEX		15
+#define DMA_CH_IER_AIE20_INDEX		15
+#define DMA_CH_IER_AIE20_WIDTH		1
+#define DMA_CH_IER_AIE_INDEX		14
 #define DMA_CH_IER_AIE_WIDTH		1
 #define DMA_CH_IER_FBEE_INDEX		12
 #define DMA_CH_IER_FBEE_WIDTH		1
-#define DMA_CH_IER_NIE_INDEX		16
+#define DMA_CH_IER_NIE20_INDEX		16
+#define DMA_CH_IER_NIE20_WIDTH		1
+#define DMA_CH_IER_NIE_INDEX		15
 #define DMA_CH_IER_NIE_WIDTH		1
 #define DMA_CH_IER_RBUE_INDEX		7
 #define DMA_CH_IER_RBUE_WIDTH		1
@@ -306,6 +193,7 @@
 #define MAC_RWKPFR			0x00c4
 #define MAC_LPICSR			0x00d0
 #define MAC_LPITCR			0x00d4
+#define MAC_TIR				0x00e0
 #define MAC_VR				0x0110
 #define MAC_DR				0x0114
 #define MAC_HWF0R			0x011c
@@ -335,7 +223,15 @@
 #define MAC_TSSR			0x0d20
 #define MAC_TXSNR			0x0d30
 #define MAC_TXSSR			0x0d34
-
+#define MAC_TICNR                       0x0d58
+#define MAC_TICSNR                      0x0d5C
+#define MAC_TECNR                       0x0d60
+#define MAC_TECSNR                      0x0d64
+#define MAC_PPSCR			0x0d70
+#define MAC_PPS0_TTSR			0x0d80
+#define MAC_PPS0_TTNSR			0x0d84
+#define MAC_PPS0_INTERVAL		0x0d88
+#define MAC_PPS0_WIDTH			0x0d8C
 #define MAC_QTFCR_INC			4
 #define MAC_MACA_INC			4
 #define MAC_HTR_INC			4
@@ -343,6 +239,18 @@
 #define MAC_RQC2_INC			4
 #define MAC_RQC2_Q_PER_REG		4
 
+/* PPS helpers */
+#define PPSEN0				BIT(4)
+#define MAC_PPSx_TTSR(x)		((MAC_PPS0_TTSR) + ((x) * 0x10))
+#define MAC_PPSx_TTNSR(x)		((MAC_PPS0_TTNSR) + ((x) * 0x10))
+#define MAC_PPSx_INTERVAL(x)		((MAC_PPS0_INTERVAL) + ((x) * 0x10))
+#define MAC_PPSx_WIDTH(x)		((MAC_PPS0_WIDTH) + ((x) * 0x10))
+#define PPS_MAXIDX(x)			((((x) + 1) * 8) - 1)
+#define PPS_MINIDX(x)			((x) * 8)
+#define XGBE_PPSCMD_STOP		0x5
+#define XGBE_PPSCMD_START		0x2
+#define XGBE_PPSTARGET_PULSE		0x2
+
 /* MAC register entry bit positions and sizes */
 #define MAC_HWF0R_ADDMACADRSEL_INDEX	18
 #define MAC_HWF0R_ADDMACADRSEL_WIDTH	5
@@ -372,6 +280,8 @@
 #define MAC_HWF0R_TXCOESEL_WIDTH	1
 #define MAC_HWF0R_VLHASH_INDEX		4
 #define MAC_HWF0R_VLHASH_WIDTH		1
+#define MAC_HWF0R_VXN_INDEX		29
+#define MAC_HWF0R_VXN_WIDTH		1
 #define MAC_HWF1R_ADDR64_INDEX		14
 #define MAC_HWF1R_ADDR64_WIDTH		2
 #define MAC_HWF1R_ADVTHWORD_INDEX	13
@@ -432,8 +342,6 @@
 #define MAC_MDIOSCAR_PA_WIDTH		5
 #define MAC_MDIOSCAR_RA_INDEX		0
 #define MAC_MDIOSCAR_RA_WIDTH		16
-#define MAC_MDIOSCAR_REG_INDEX		0
-#define MAC_MDIOSCAR_REG_WIDTH		21
 #define MAC_MDIOSCCDR_BUSY_INDEX	22
 #define MAC_MDIOSCCDR_BUSY_WIDTH	1
 #define MAC_MDIOSCCDR_CMD_INDEX		16
@@ -456,6 +364,8 @@
 #define MAC_PFR_PR_WIDTH		1
 #define MAC_PFR_VTFE_INDEX		16
 #define MAC_PFR_VTFE_WIDTH		1
+#define MAC_PFR_VUCC_INDEX		22
+#define MAC_PFR_VUCC_WIDTH		1
 #define MAC_PMTCSR_MGKPKTEN_INDEX	1
 #define MAC_PMTCSR_MGKPKTEN_WIDTH	1
 #define MAC_PMTCSR_PWRDWN_INDEX		0
@@ -474,6 +384,10 @@
 #define MAC_RCR_CST_WIDTH		1
 #define MAC_RCR_DCRCC_INDEX		3
 #define MAC_RCR_DCRCC_WIDTH		1
+#define MAC_RCR_GPSLCE_INDEX		6
+#define MAC_RCR_GPSLCE_WIDTH		1
+#define MAC_RCR_WD_INDEX		7
+#define MAC_RCR_WD_WIDTH		1
 #define MAC_RCR_HDSMS_INDEX		12
 #define MAC_RCR_HDSMS_WIDTH		3
 #define MAC_RCR_IPC_INDEX		9
@@ -484,6 +398,8 @@
 #define MAC_RCR_LM_WIDTH		1
 #define MAC_RCR_RE_INDEX		0
 #define MAC_RCR_RE_WIDTH		1
+#define MAC_RCR_GPSL_INDEX		16
+#define MAC_RCR_GPSL_WIDTH		14
 #define MAC_RFCR_PFCE_INDEX		8
 #define MAC_RFCR_PFCE_WIDTH		1
 #define MAC_RFCR_RFE_INDEX		0
@@ -518,12 +434,22 @@
 #define MAC_TCR_SS_WIDTH		2
 #define MAC_TCR_TE_INDEX		0
 #define MAC_TCR_TE_WIDTH		1
+#define MAC_TCR_VNE_INDEX		24
+#define MAC_TCR_VNE_WIDTH		1
+#define MAC_TCR_VNM_INDEX		25
+#define MAC_TCR_VNM_WIDTH		1
+#define MAC_TCR_JD_INDEX		16
+#define MAC_TCR_JD_WIDTH		1
+#define MAC_TIR_TNID_INDEX		0
+#define MAC_TIR_TNID_WIDTH		16
 #define MAC_TSCR_AV8021ASMEN_INDEX	28
 #define MAC_TSCR_AV8021ASMEN_WIDTH	1
 #define MAC_TSCR_SNAPTYPSEL_INDEX	16
 #define MAC_TSCR_SNAPTYPSEL_WIDTH	2
 #define MAC_TSCR_TSADDREG_INDEX		5
 #define MAC_TSCR_TSADDREG_WIDTH		1
+#define MAC_TSCR_TSUPDT_INDEX		3
+#define MAC_TSCR_TSUPDT_WIDTH		1
 #define MAC_TSCR_TSCFUPDT_INDEX		1
 #define MAC_TSCR_TSCFUPDT_WIDTH		1
 #define MAC_TSCR_TSCTRLSSR_INDEX	9
@@ -552,6 +478,10 @@
 #define MAC_TSSR_TXTSC_WIDTH		1
 #define MAC_TXSNR_TXTSSTSMIS_INDEX	31
 #define MAC_TXSNR_TXTSSTSMIS_WIDTH	1
+#define MAC_TICSNR_TSICSNS_INDEX	8
+#define MAC_TICSNR_TSICSNS_WIDTH	8
+#define MAC_TECSNR_TSECSNS_INDEX	8
+#define MAC_TECSNR_TSECSNS_WIDTH	8
 #define MAC_VLANHTR_VLHT_INDEX		0
 #define MAC_VLANHTR_VLHT_WIDTH		16
 #define MAC_VLANIR_VLTI_INDEX		20
@@ -582,8 +512,10 @@
 #define MAC_VR_SNPSVER_WIDTH		8
 #define MAC_VR_USERVER_INDEX		16
 #define MAC_VR_USERVER_WIDTH		8
+#define MAC_PPSx_TTNSR_TRGTBUSY0_INDEX	31
+#define MAC_PPSx_TTNSR_TRGTBUSY0_WIDTH	1
 
-/* MMC register offsets */
+ /* MMC register offsets */
 #define MMC_CR				0x0800
 #define MMC_RISR			0x0804
 #define MMC_TISR			0x0808
@@ -893,6 +825,13 @@
 #define PCS_V2_WINDOW_SELECT		0x9064
 #define PCS_V2_RV_WINDOW_DEF		0x1060
 #define PCS_V2_RV_WINDOW_SELECT		0x1064
+#define PCS_V2_YC_WINDOW_DEF		0x18060
+#define PCS_V2_YC_WINDOW_SELECT		0x18064
+#define PCS_V3_RN_WINDOW_DEF		0xf8078
+#define PCS_V3_RN_WINDOW_SELECT		0xf807c
+
+#define PCS_RN_SMN_BASE_ADDR		0x11e00000
+#define PCS_RN_PORT_ADDR_SIZE		0x100000
 
 /* PCS register entry bit positions and sizes */
 #define PCS_V2_WINDOW_DEF_OFFSET_INDEX	6
@@ -959,6 +898,7 @@
 #define XP_DRIVER_INT_RO		0x0064
 #define XP_DRIVER_SCRATCH_0		0x0068
 #define XP_DRIVER_SCRATCH_1		0x006c
+#define XP_INT_REISSUE_EN		0x0074
 #define XP_INT_EN			0x0078
 #define XP_I2C_MUTEX			0x0080
 #define XP_MDIO_MUTEX			0x0084
@@ -1024,8 +964,8 @@
 #define XP_PROP_0_PORT_ID_WIDTH			8
 #define XP_PROP_0_PORT_MODE_INDEX		8
 #define XP_PROP_0_PORT_MODE_WIDTH		4
-#define XP_PROP_0_PORT_SPEEDS_INDEX		23
-#define XP_PROP_0_PORT_SPEEDS_WIDTH		4
+#define XP_PROP_0_PORT_SPEEDS_INDEX		22
+#define XP_PROP_0_PORT_SPEEDS_WIDTH		5
 #define XP_PROP_1_MAX_RX_DMA_INDEX		24
 #define XP_PROP_1_MAX_RX_DMA_WIDTH		5
 #define XP_PROP_1_MAX_RX_QUEUES_INDEX		8
@@ -1160,11 +1100,17 @@
 #define RX_PACKET_ATTRIBUTES_RSS_HASH_WIDTH	1
 #define RX_PACKET_ATTRIBUTES_FIRST_INDEX	7
 #define RX_PACKET_ATTRIBUTES_FIRST_WIDTH	1
+#define RX_PACKET_ATTRIBUTES_TNP_INDEX		8
+#define RX_PACKET_ATTRIBUTES_TNP_WIDTH		1
+#define RX_PACKET_ATTRIBUTES_TNPCSUM_DONE_INDEX	9
+#define RX_PACKET_ATTRIBUTES_TNPCSUM_DONE_WIDTH	1
 
 #define RX_NORMAL_DESC0_OVT_INDEX		0
 #define RX_NORMAL_DESC0_OVT_WIDTH		16
 #define RX_NORMAL_DESC2_HL_INDEX		0
 #define RX_NORMAL_DESC2_HL_WIDTH		10
+#define RX_NORMAL_DESC2_TNP_INDEX		11
+#define RX_NORMAL_DESC2_TNP_WIDTH		1
 #define RX_NORMAL_DESC3_CDA_INDEX		27
 #define RX_NORMAL_DESC3_CDA_WIDTH		1
 #define RX_NORMAL_DESC3_CTXT_INDEX		30
@@ -1191,9 +1137,11 @@
 #define RX_DESC3_L34T_IPV4_TCP			1
 #define RX_DESC3_L34T_IPV4_UDP			2
 #define RX_DESC3_L34T_IPV4_ICMP			3
+#define RX_DESC3_L34T_IPV4_UNKNOWN		7
 #define RX_DESC3_L34T_IPV6_TCP			9
 #define RX_DESC3_L34T_IPV6_UDP			10
 #define RX_DESC3_L34T_IPV6_ICMP			11
+#define RX_DESC3_L34T_IPV6_UNKNOWN		15
 
 #define RX_CONTEXT_DESC3_TSA_INDEX		4
 #define RX_CONTEXT_DESC3_TSA_WIDTH		1
@@ -1208,6 +1156,8 @@
 #define TX_PACKET_ATTRIBUTES_VLAN_CTAG_WIDTH	1
 #define TX_PACKET_ATTRIBUTES_PTP_INDEX		3
 #define TX_PACKET_ATTRIBUTES_PTP_WIDTH		1
+#define TX_PACKET_ATTRIBUTES_VXLAN_INDEX	4
+#define TX_PACKET_ATTRIBUTES_VXLAN_WIDTH	1
 
 #define TX_CONTEXT_DESC2_MSS_INDEX		0
 #define TX_CONTEXT_DESC2_MSS_WIDTH		15
@@ -1248,8 +1198,11 @@
 #define TX_NORMAL_DESC3_TCPPL_WIDTH		18
 #define TX_NORMAL_DESC3_TSE_INDEX		18
 #define TX_NORMAL_DESC3_TSE_WIDTH		1
+#define TX_NORMAL_DESC3_VNP_INDEX		23
+#define TX_NORMAL_DESC3_VNP_WIDTH		3
 
 #define TX_NORMAL_DESC2_VLAN_INSERT		0x2
+#define TX_NORMAL_DESC3_VXLAN_PACKET		0x3
 
 /* MDIO undefined or vendor specific registers */
 #ifndef MDIO_PMA_10GBR_PMD_CTRL
@@ -1260,10 +1213,34 @@
 #define MDIO_PMA_10GBR_FECCTRL		0x00ab
 #endif
 
+#ifndef MDIO_PMA_RX_CTRL1
+#define MDIO_PMA_RX_CTRL1		0x8051
+#endif
+
+#ifndef MDIO_PMA_RX_LSTS
+#define MDIO_PMA_RX_LSTS		0x018020
+#endif
+
+#ifndef MDIO_PMA_RX_EQ_CTRL4
+#define MDIO_PMA_RX_EQ_CTRL4		0x0001805C
+#endif
+
+#ifndef MDIO_PMA_MP_MISC_STS
+#define MDIO_PMA_MP_MISC_STS		0x0078
+#endif
+
+#ifndef MDIO_PMA_PHY_RX_EQ_CEU
+#define MDIO_PMA_PHY_RX_EQ_CEU		0x1800E
+#endif
+
 #ifndef MDIO_PCS_DIG_CTRL
 #define MDIO_PCS_DIG_CTRL		0x8000
 #endif
 
+#ifndef MDIO_PCS_DIGITAL_STAT
+#define MDIO_PCS_DIGITAL_STAT		0x8010
+#endif
+
 #ifndef MDIO_AN_XNP
 #define MDIO_AN_XNP			0x0016
 #endif
@@ -1300,6 +1277,14 @@
 #define MDIO_VEND2_AN_STAT		0x8002
 #endif
 
+#ifndef MDIO_VEND2_PMA_CDR_CONTROL
+#define MDIO_VEND2_PMA_CDR_CONTROL	0x8056
+#endif
+
+#ifndef MDIO_VEND2_PMA_MISC_CTRL0
+#define MDIO_VEND2_PMA_MISC_CTRL0	0x8090
+#endif
+
 #ifndef MDIO_CTRL1_SPEED1G
 #define MDIO_CTRL1_SPEED1G		(MDIO_CTRL1_SPEED10G & ~BMCR_SPEED100)
 #endif
@@ -1320,6 +1305,8 @@
 #define MDIO_VEND2_CTRL1_SS13		BIT(13)
 #endif
 
+#define XGBE_VEND2_MAC_AUTO_SW		BIT(9)
+
 /* MDIO mask values */
 #define XGBE_AN_CL73_INT_CMPLT		BIT(0)
 #define XGBE_AN_CL73_INC_LINK		BIT(1)
@@ -1335,6 +1322,8 @@
 #define XGBE_KR_TRAINING_ENABLE		BIT(1)
 
 #define XGBE_PCS_CL37_BP		BIT(12)
+#define XGBE_PCS_PSEQ_STATE_MASK	0x1c
+#define XGBE_PCS_PSEQ_STATE_POWER_GOOD	0x10
 
 #define XGBE_AN_CL37_INT_CMPLT		BIT(0)
 #define XGBE_AN_CL37_INT_MASK		0x01
@@ -1346,6 +1335,41 @@
 #define XGBE_AN_CL37_PCS_MODE_BASEX	0x00
 #define XGBE_AN_CL37_PCS_MODE_SGMII	0x04
 #define XGBE_AN_CL37_TX_CONFIG_MASK	0x08
+#define XGBE_AN_CL37_MII_CTRL_8BIT	0x0100
+
+#define XGBE_PMA_CDR_TRACK_EN_MASK	0x01
+#define XGBE_PMA_CDR_TRACK_EN_OFF	0x00
+#define XGBE_PMA_CDR_TRACK_EN_ON	0x01
+
+#define XGBE_PMA_RX_RST_0_MASK		BIT(4)
+#define XGBE_PMA_RX_RST_0_RESET_ON	0x10
+#define XGBE_PMA_RX_RST_0_RESET_OFF	0x00
+
+#define XGBE_PMA_RX_SIG_DET_0_MASK	BIT(4)
+#define XGBE_PMA_RX_SIG_DET_0_ENABLE	BIT(4)
+#define XGBE_PMA_RX_SIG_DET_0_DISABLE	0x0000
+
+#define XGBE_PMA_RX_VALID_0_MASK	BIT(12)
+#define XGBE_PMA_RX_VALID_0_ENABLE	BIT(12)
+#define XGBE_PMA_RX_VALID_0_DISABLE	0x0000
+
+#define XGBE_PMA_RX_AD_REQ_MASK		BIT(12)
+#define XGBE_PMA_RX_AD_REQ_ENABLE	BIT(12)
+#define XGBE_PMA_RX_AD_REQ_DISABLE	0x0000
+
+#define XGBE_PMA_RX_ADPT_ACK_MASK	BIT(12)
+#define XGBE_PMA_RX_ADPT_ACK		BIT(12)
+
+#define XGBE_PMA_CFF_UPDTM1_VLD		BIT(8)
+#define XGBE_PMA_CFF_UPDT0_VLD		BIT(9)
+#define XGBE_PMA_CFF_UPDT1_VLD		BIT(10)
+#define XGBE_PMA_CFF_UPDT_MASK		(XGBE_PMA_CFF_UPDTM1_VLD |\
+					 XGBE_PMA_CFF_UPDT0_VLD | \
+					 XGBE_PMA_CFF_UPDT1_VLD)
+
+#define XGBE_PMA_PLL_CTRL_MASK		BIT(15)
+#define XGBE_PMA_PLL_CTRL_ENABLE	BIT(15)
+#define XGBE_PMA_PLL_CTRL_DISABLE	0x0000
 
 /* Bit setting and getting macros
  *  The get macro will extract the current bit field value from within
@@ -1647,20 +1671,21 @@ do {									\
 } while (0)
 
 /* Macros for building, reading or writing register values or bits
- * using MDIO.  Different from above because of the use of standardized
- * Linux include values.  No shifting is performed with the bit
- * operations, everything works on mask values.
+ * using MDIO.
  */
+
+#define XGBE_ADDR_C45 BIT(30)
+
 #define XMDIO_READ(_pdata, _mmd, _reg)					\
 	((_pdata)->hw_if.read_mmd_regs((_pdata), 0,			\
-		MII_ADDR_C45 | (_mmd << 16) | ((_reg) & 0xffff)))
+		XGBE_ADDR_C45 | (_mmd << 16) | ((_reg) & 0xffff)))
 
 #define XMDIO_READ_BITS(_pdata, _mmd, _reg, _mask)			\
 	(XMDIO_READ((_pdata), _mmd, _reg) & _mask)
 
 #define XMDIO_WRITE(_pdata, _mmd, _reg, _val)				\
 	((_pdata)->hw_if.write_mmd_regs((_pdata), 0,			\
-		MII_ADDR_C45 | (_mmd << 16) | ((_reg) & 0xffff), (_val)))
+		XGBE_ADDR_C45 | (_mmd << 16) | ((_reg) & 0xffff), (_val)))
 
 #define XMDIO_WRITE_BITS(_pdata, _mmd, _reg, _mask, _val)		\
 do {									\
diff --git a/drivers/net/ethernet/amd/xgbe/xgbe-dcb.c b/drivers/net/ethernet/amd/xgbe/xgbe-dcb.c
index 895d35639129..1474df5544fa 100644
--- a/drivers/net/ethernet/amd/xgbe/xgbe-dcb.c
+++ b/drivers/net/ethernet/amd/xgbe/xgbe-dcb.c
@@ -1,117 +1,8 @@
+// SPDX-License-Identifier: (GPL-2.0-or-later OR BSD-3-Clause)
 /*
- * AMD 10Gb Ethernet driver
- *
- * This file is available to you under your choice of the following two
- * licenses:
- *
- * License 1: GPLv2
- *
- * Copyright (c) 2014-2016 Advanced Micro Devices, Inc.
- *
- * This file is free software; you may copy, redistribute and/or modify
- * it under the terms of the GNU General Public License as published by
- * the Free Software Foundation, either version 2 of the License, or (at
- * your option) any later version.
- *
- * This file is distributed in the hope that it will be useful, but
- * WITHOUT ANY WARRANTY; without even the implied warranty of
- * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
- * General Public License for more details.
- *
- * You should have received a copy of the GNU General Public License
- * along with this program.  If not, see <http://www.gnu.org/licenses/>.
- *
- * This file incorporates work covered by the following copyright and
- * permission notice:
- *     The Synopsys DWC ETHER XGMAC Software Driver and documentation
- *     (hereinafter "Software") is an unsupported proprietary work of Synopsys,
- *     Inc. unless otherwise expressly agreed to in writing between Synopsys
- *     and you.
- *
- *     The Software IS NOT an item of Licensed Software or Licensed Product
- *     under any End User Software License Agreement or Agreement for Licensed
- *     Product with Synopsys or any supplement thereto.  Permission is hereby
- *     granted, free of charge, to any person obtaining a copy of this software
- *     annotated with this license and the Software, to deal in the Software
- *     without restriction, including without limitation the rights to use,
- *     copy, modify, merge, publish, distribute, sublicense, and/or sell copies
- *     of the Software, and to permit persons to whom the Software is furnished
- *     to do so, subject to the following conditions:
- *
- *     The above copyright notice and this permission notice shall be included
- *     in all copies or substantial portions of the Software.
- *
- *     THIS SOFTWARE IS BEING DISTRIBUTED BY SYNOPSYS SOLELY ON AN "AS IS"
- *     BASIS AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED
- *     TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A
- *     PARTICULAR PURPOSE ARE HEREBY DISCLAIMED. IN NO EVENT SHALL SYNOPSYS
- *     BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR
- *     CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF
- *     SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS
- *     INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN
- *     CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE)
- *     ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF
- *     THE POSSIBILITY OF SUCH DAMAGE.
- *
- *
- * License 2: Modified BSD
- *
- * Copyright (c) 2014-2016 Advanced Micro Devices, Inc.
- * All rights reserved.
- *
- * Redistribution and use in source and binary forms, with or without
- * modification, are permitted provided that the following conditions are met:
- *     * Redistributions of source code must retain the above copyright
- *       notice, this list of conditions and the following disclaimer.
- *     * Redistributions in binary form must reproduce the above copyright
- *       notice, this list of conditions and the following disclaimer in the
- *       documentation and/or other materials provided with the distribution.
- *     * Neither the name of Advanced Micro Devices, Inc. nor the
- *       names of its contributors may be used to endorse or promote products
- *       derived from this software without specific prior written permission.
- *
- * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS"
- * AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
- * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
- * ARE DISCLAIMED. IN NO EVENT SHALL <COPYRIGHT HOLDER> BE LIABLE FOR ANY
- * DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES
- * (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES;
- * LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND
- * ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
- * (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF
- * THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
- *
- * This file incorporates work covered by the following copyright and
- * permission notice:
- *     The Synopsys DWC ETHER XGMAC Software Driver and documentation
- *     (hereinafter "Software") is an unsupported proprietary work of Synopsys,
- *     Inc. unless otherwise expressly agreed to in writing between Synopsys
- *     and you.
- *
- *     The Software IS NOT an item of Licensed Software or Licensed Product
- *     under any End User Software License Agreement or Agreement for Licensed
- *     Product with Synopsys or any supplement thereto.  Permission is hereby
- *     granted, free of charge, to any person obtaining a copy of this software
- *     annotated with this license and the Software, to deal in the Software
- *     without restriction, including without limitation the rights to use,
- *     copy, modify, merge, publish, distribute, sublicense, and/or sell copies
- *     of the Software, and to permit persons to whom the Software is furnished
- *     to do so, subject to the following conditions:
- *
- *     The above copyright notice and this permission notice shall be included
- *     in all copies or substantial portions of the Software.
- *
- *     THIS SOFTWARE IS BEING DISTRIBUTED BY SYNOPSYS SOLELY ON AN "AS IS"
- *     BASIS AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED
- *     TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A
- *     PARTICULAR PURPOSE ARE HEREBY DISCLAIMED. IN NO EVENT SHALL SYNOPSYS
- *     BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR
- *     CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF
- *     SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS
- *     INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN
- *     CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE)
- *     ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF
- *     THE POSSIBILITY OF SUCH DAMAGE.
+ * Copyright (c) 2014-2025, Advanced Micro Devices, Inc.
+ * Copyright (c) 2014, Synopsys, Inc.
+ * All rights reserved
  */
 
 #include <linux/netdevice.h>
@@ -230,7 +121,7 @@ static int xgbe_dcb_ieee_setpfc(struct net_device *netdev,
 	struct xgbe_prv_data *pdata = netdev_priv(netdev);
 
 	netif_dbg(pdata, drv, netdev,
-		  "cap=%hhu, en=%#hhx, mbc=%hhu, delay=%hhu\n",
+		  "cap=%d, en=%#x, mbc=%d, delay=%d\n",
 		  pfc->pfc_cap, pfc->pfc_en, pfc->mbc, pfc->delay);
 
 	/* Check PFC for supported number of traffic classes */
diff --git a/drivers/net/ethernet/amd/xgbe/xgbe-debugfs.c b/drivers/net/ethernet/amd/xgbe/xgbe-debugfs.c
index 7546b660d6b5..d9157c4acde9 100644
--- a/drivers/net/ethernet/amd/xgbe/xgbe-debugfs.c
+++ b/drivers/net/ethernet/amd/xgbe/xgbe-debugfs.c
@@ -1,117 +1,8 @@
+// SPDX-License-Identifier: (GPL-2.0-or-later OR BSD-3-Clause)
 /*
- * AMD 10Gb Ethernet driver
- *
- * This file is available to you under your choice of the following two
- * licenses:
- *
- * License 1: GPLv2
- *
- * Copyright (c) 2014 Advanced Micro Devices, Inc.
- *
- * This file is free software; you may copy, redistribute and/or modify
- * it under the terms of the GNU General Public License as published by
- * the Free Software Foundation, either version 2 of the License, or (at
- * your option) any later version.
- *
- * This file is distributed in the hope that it will be useful, but
- * WITHOUT ANY WARRANTY; without even the implied warranty of
- * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
- * General Public License for more details.
- *
- * You should have received a copy of the GNU General Public License
- * along with this program.  If not, see <http://www.gnu.org/licenses/>.
- *
- * This file incorporates work covered by the following copyright and
- * permission notice:
- *     The Synopsys DWC ETHER XGMAC Software Driver and documentation
- *     (hereinafter "Software") is an unsupported proprietary work of Synopsys,
- *     Inc. unless otherwise expressly agreed to in writing between Synopsys
- *     and you.
- *
- *     The Software IS NOT an item of Licensed Software or Licensed Product
- *     under any End User Software License Agreement or Agreement for Licensed
- *     Product with Synopsys or any supplement thereto.  Permission is hereby
- *     granted, free of charge, to any person obtaining a copy of this software
- *     annotated with this license and the Software, to deal in the Software
- *     without restriction, including without limitation the rights to use,
- *     copy, modify, merge, publish, distribute, sublicense, and/or sell copies
- *     of the Software, and to permit persons to whom the Software is furnished
- *     to do so, subject to the following conditions:
- *
- *     The above copyright notice and this permission notice shall be included
- *     in all copies or substantial portions of the Software.
- *
- *     THIS SOFTWARE IS BEING DISTRIBUTED BY SYNOPSYS SOLELY ON AN "AS IS"
- *     BASIS AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED
- *     TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A
- *     PARTICULAR PURPOSE ARE HEREBY DISCLAIMED. IN NO EVENT SHALL SYNOPSYS
- *     BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR
- *     CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF
- *     SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS
- *     INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN
- *     CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE)
- *     ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF
- *     THE POSSIBILITY OF SUCH DAMAGE.
- *
- *
- * License 2: Modified BSD
- *
- * Copyright (c) 2014 Advanced Micro Devices, Inc.
- * All rights reserved.
- *
- * Redistribution and use in source and binary forms, with or without
- * modification, are permitted provided that the following conditions are met:
- *     * Redistributions of source code must retain the above copyright
- *       notice, this list of conditions and the following disclaimer.
- *     * Redistributions in binary form must reproduce the above copyright
- *       notice, this list of conditions and the following disclaimer in the
- *       documentation and/or other materials provided with the distribution.
- *     * Neither the name of Advanced Micro Devices, Inc. nor the
- *       names of its contributors may be used to endorse or promote products
- *       derived from this software without specific prior written permission.
- *
- * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS"
- * AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
- * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
- * ARE DISCLAIMED. IN NO EVENT SHALL <COPYRIGHT HOLDER> BE LIABLE FOR ANY
- * DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES
- * (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES;
- * LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND
- * ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
- * (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF
- * THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
- *
- * This file incorporates work covered by the following copyright and
- * permission notice:
- *     The Synopsys DWC ETHER XGMAC Software Driver and documentation
- *     (hereinafter "Software") is an unsupported proprietary work of Synopsys,
- *     Inc. unless otherwise expressly agreed to in writing between Synopsys
- *     and you.
- *
- *     The Software IS NOT an item of Licensed Software or Licensed Product
- *     under any End User Software License Agreement or Agreement for Licensed
- *     Product with Synopsys or any supplement thereto.  Permission is hereby
- *     granted, free of charge, to any person obtaining a copy of this software
- *     annotated with this license and the Software, to deal in the Software
- *     without restriction, including without limitation the rights to use,
- *     copy, modify, merge, publish, distribute, sublicense, and/or sell copies
- *     of the Software, and to permit persons to whom the Software is furnished
- *     to do so, subject to the following conditions:
- *
- *     The above copyright notice and this permission notice shall be included
- *     in all copies or substantial portions of the Software.
- *
- *     THIS SOFTWARE IS BEING DISTRIBUTED BY SYNOPSYS SOLELY ON AN "AS IS"
- *     BASIS AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED
- *     TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A
- *     PARTICULAR PURPOSE ARE HEREBY DISCLAIMED. IN NO EVENT SHALL SYNOPSYS
- *     BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR
- *     CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF
- *     SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS
- *     INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN
- *     CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE)
- *     ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF
- *     THE POSSIBILITY OF SUCH DAMAGE.
+ * Copyright (c) 2014-2025, Advanced Micro Devices, Inc.
+ * Copyright (c) 2014, Synopsys, Inc.
+ * All rights reserved
  */
 
 #include <linux/debugfs.h>
@@ -438,7 +329,6 @@ static const struct file_operations xi2c_reg_value_fops = {
 
 void xgbe_debugfs_init(struct xgbe_prv_data *pdata)
 {
-	struct dentry *pfile;
 	char *buf;
 
 	/* Set defaults */
@@ -451,72 +341,48 @@ void xgbe_debugfs_init(struct xgbe_prv_data *pdata)
 		return;
 
 	pdata->xgbe_debugfs = debugfs_create_dir(buf, NULL);
-	if (!pdata->xgbe_debugfs) {
-		netdev_err(pdata->netdev, "debugfs_create_dir failed\n");
-		kfree(buf);
-		return;
-	}
 
-	pfile = debugfs_create_file("xgmac_register", 0600,
-				    pdata->xgbe_debugfs, pdata,
-				    &xgmac_reg_addr_fops);
-	if (!pfile)
-		netdev_err(pdata->netdev, "debugfs_create_file failed\n");
+	debugfs_create_file("xgmac_register", 0600, pdata->xgbe_debugfs, pdata,
+			    &xgmac_reg_addr_fops);
 
-	pfile = debugfs_create_file("xgmac_register_value", 0600,
-				    pdata->xgbe_debugfs, pdata,
-				    &xgmac_reg_value_fops);
-	if (!pfile)
-		netdev_err(pdata->netdev, "debugfs_create_file failed\n");
+	debugfs_create_file("xgmac_register_value", 0600, pdata->xgbe_debugfs,
+			    pdata, &xgmac_reg_value_fops);
 
-	pfile = debugfs_create_file("xpcs_mmd", 0600,
-				    pdata->xgbe_debugfs, pdata,
-				    &xpcs_mmd_fops);
-	if (!pfile)
-		netdev_err(pdata->netdev, "debugfs_create_file failed\n");
+	debugfs_create_file("xpcs_mmd", 0600, pdata->xgbe_debugfs, pdata,
+			    &xpcs_mmd_fops);
 
-	pfile = debugfs_create_file("xpcs_register", 0600,
-				    pdata->xgbe_debugfs, pdata,
-				    &xpcs_reg_addr_fops);
-	if (!pfile)
-		netdev_err(pdata->netdev, "debugfs_create_file failed\n");
+	debugfs_create_file("xpcs_register", 0600, pdata->xgbe_debugfs, pdata,
+			    &xpcs_reg_addr_fops);
 
-	pfile = debugfs_create_file("xpcs_register_value", 0600,
-				    pdata->xgbe_debugfs, pdata,
-				    &xpcs_reg_value_fops);
-	if (!pfile)
-		netdev_err(pdata->netdev, "debugfs_create_file failed\n");
+	debugfs_create_file("xpcs_register_value", 0600, pdata->xgbe_debugfs,
+			    pdata, &xpcs_reg_value_fops);
 
 	if (pdata->xprop_regs) {
-		pfile = debugfs_create_file("xprop_register", 0600,
-					    pdata->xgbe_debugfs, pdata,
-					    &xprop_reg_addr_fops);
-		if (!pfile)
-			netdev_err(pdata->netdev,
-				   "debugfs_create_file failed\n");
-
-		pfile = debugfs_create_file("xprop_register_value", 0600,
-					    pdata->xgbe_debugfs, pdata,
-					    &xprop_reg_value_fops);
-		if (!pfile)
-			netdev_err(pdata->netdev,
-				   "debugfs_create_file failed\n");
+		debugfs_create_file("xprop_register", 0600, pdata->xgbe_debugfs,
+				    pdata, &xprop_reg_addr_fops);
+
+		debugfs_create_file("xprop_register_value", 0600,
+				    pdata->xgbe_debugfs, pdata,
+				    &xprop_reg_value_fops);
 	}
 
 	if (pdata->xi2c_regs) {
-		pfile = debugfs_create_file("xi2c_register", 0600,
-					    pdata->xgbe_debugfs, pdata,
-					    &xi2c_reg_addr_fops);
-		if (!pfile)
-			netdev_err(pdata->netdev,
-				   "debugfs_create_file failed\n");
-
-		pfile = debugfs_create_file("xi2c_register_value", 0600,
-					    pdata->xgbe_debugfs, pdata,
-					    &xi2c_reg_value_fops);
-		if (!pfile)
-			netdev_err(pdata->netdev,
-				   "debugfs_create_file failed\n");
+		debugfs_create_file("xi2c_register", 0600, pdata->xgbe_debugfs,
+				    pdata, &xi2c_reg_addr_fops);
+
+		debugfs_create_file("xi2c_register_value", 0600,
+				    pdata->xgbe_debugfs, pdata,
+				    &xi2c_reg_value_fops);
+	}
+
+	if (pdata->vdata->an_cdr_workaround) {
+		debugfs_create_bool("an_cdr_workaround", 0600,
+				    pdata->xgbe_debugfs,
+				    &pdata->debugfs_an_cdr_workaround);
+
+		debugfs_create_bool("an_cdr_track_early", 0600,
+				    pdata->xgbe_debugfs,
+				    &pdata->debugfs_an_cdr_track_early);
 	}
 
 	kfree(buf);
@@ -527,3 +393,9 @@ void xgbe_debugfs_exit(struct xgbe_prv_data *pdata)
 	debugfs_remove_recursive(pdata->xgbe_debugfs);
 	pdata->xgbe_debugfs = NULL;
 }
+
+void xgbe_debugfs_rename(struct xgbe_prv_data *pdata)
+{
+	debugfs_change_name(pdata->xgbe_debugfs,
+			    "amd-xgbe-%s", pdata->netdev->name);
+}
diff --git a/drivers/net/ethernet/amd/xgbe/xgbe-desc.c b/drivers/net/ethernet/amd/xgbe/xgbe-desc.c
index b3bc87fe3764..7c8a19988a52 100644
--- a/drivers/net/ethernet/amd/xgbe/xgbe-desc.c
+++ b/drivers/net/ethernet/amd/xgbe/xgbe-desc.c
@@ -1,117 +1,8 @@
+// SPDX-License-Identifier: (GPL-2.0-or-later OR BSD-3-Clause)
 /*
- * AMD 10Gb Ethernet driver
- *
- * This file is available to you under your choice of the following two
- * licenses:
- *
- * License 1: GPLv2
- *
- * Copyright (c) 2014 Advanced Micro Devices, Inc.
- *
- * This file is free software; you may copy, redistribute and/or modify
- * it under the terms of the GNU General Public License as published by
- * the Free Software Foundation, either version 2 of the License, or (at
- * your option) any later version.
- *
- * This file is distributed in the hope that it will be useful, but
- * WITHOUT ANY WARRANTY; without even the implied warranty of
- * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
- * General Public License for more details.
- *
- * You should have received a copy of the GNU General Public License
- * along with this program.  If not, see <http://www.gnu.org/licenses/>.
- *
- * This file incorporates work covered by the following copyright and
- * permission notice:
- *     The Synopsys DWC ETHER XGMAC Software Driver and documentation
- *     (hereinafter "Software") is an unsupported proprietary work of Synopsys,
- *     Inc. unless otherwise expressly agreed to in writing between Synopsys
- *     and you.
- *
- *     The Software IS NOT an item of Licensed Software or Licensed Product
- *     under any End User Software License Agreement or Agreement for Licensed
- *     Product with Synopsys or any supplement thereto.  Permission is hereby
- *     granted, free of charge, to any person obtaining a copy of this software
- *     annotated with this license and the Software, to deal in the Software
- *     without restriction, including without limitation the rights to use,
- *     copy, modify, merge, publish, distribute, sublicense, and/or sell copies
- *     of the Software, and to permit persons to whom the Software is furnished
- *     to do so, subject to the following conditions:
- *
- *     The above copyright notice and this permission notice shall be included
- *     in all copies or substantial portions of the Software.
- *
- *     THIS SOFTWARE IS BEING DISTRIBUTED BY SYNOPSYS SOLELY ON AN "AS IS"
- *     BASIS AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED
- *     TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A
- *     PARTICULAR PURPOSE ARE HEREBY DISCLAIMED. IN NO EVENT SHALL SYNOPSYS
- *     BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR
- *     CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF
- *     SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS
- *     INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN
- *     CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE)
- *     ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF
- *     THE POSSIBILITY OF SUCH DAMAGE.
- *
- *
- * License 2: Modified BSD
- *
- * Copyright (c) 2014 Advanced Micro Devices, Inc.
- * All rights reserved.
- *
- * Redistribution and use in source and binary forms, with or without
- * modification, are permitted provided that the following conditions are met:
- *     * Redistributions of source code must retain the above copyright
- *       notice, this list of conditions and the following disclaimer.
- *     * Redistributions in binary form must reproduce the above copyright
- *       notice, this list of conditions and the following disclaimer in the
- *       documentation and/or other materials provided with the distribution.
- *     * Neither the name of Advanced Micro Devices, Inc. nor the
- *       names of its contributors may be used to endorse or promote products
- *       derived from this software without specific prior written permission.
- *
- * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS"
- * AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
- * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
- * ARE DISCLAIMED. IN NO EVENT SHALL <COPYRIGHT HOLDER> BE LIABLE FOR ANY
- * DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES
- * (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES;
- * LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND
- * ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
- * (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF
- * THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
- *
- * This file incorporates work covered by the following copyright and
- * permission notice:
- *     The Synopsys DWC ETHER XGMAC Software Driver and documentation
- *     (hereinafter "Software") is an unsupported proprietary work of Synopsys,
- *     Inc. unless otherwise expressly agreed to in writing between Synopsys
- *     and you.
- *
- *     The Software IS NOT an item of Licensed Software or Licensed Product
- *     under any End User Software License Agreement or Agreement for Licensed
- *     Product with Synopsys or any supplement thereto.  Permission is hereby
- *     granted, free of charge, to any person obtaining a copy of this software
- *     annotated with this license and the Software, to deal in the Software
- *     without restriction, including without limitation the rights to use,
- *     copy, modify, merge, publish, distribute, sublicense, and/or sell copies
- *     of the Software, and to permit persons to whom the Software is furnished
- *     to do so, subject to the following conditions:
- *
- *     The above copyright notice and this permission notice shall be included
- *     in all copies or substantial portions of the Software.
- *
- *     THIS SOFTWARE IS BEING DISTRIBUTED BY SYNOPSYS SOLELY ON AN "AS IS"
- *     BASIS AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED
- *     TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A
- *     PARTICULAR PURPOSE ARE HEREBY DISCLAIMED. IN NO EVENT SHALL SYNOPSYS
- *     BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR
- *     CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF
- *     SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS
- *     INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN
- *     CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE)
- *     ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF
- *     THE POSSIBILITY OF SUCH DAMAGE.
+ * Copyright (c) 2014-2025, Advanced Micro Devices, Inc.
+ * Copyright (c) 2014, Synopsys, Inc.
+ * All rights reserved
  */
 
 #include "xgbe.h"
@@ -176,8 +67,8 @@ static void xgbe_free_ring_resources(struct xgbe_prv_data *pdata)
 
 	DBGPR("-->xgbe_free_ring_resources\n");
 
-	channel = pdata->channel;
-	for (i = 0; i < pdata->channel_count; i++, channel++) {
+	for (i = 0; i < pdata->channel_count; i++) {
+		channel = pdata->channel[i];
 		xgbe_free_ring(pdata, channel->tx_ring);
 		xgbe_free_ring(pdata, channel->rx_ring);
 	}
@@ -185,34 +76,60 @@ static void xgbe_free_ring_resources(struct xgbe_prv_data *pdata)
 	DBGPR("<--xgbe_free_ring_resources\n");
 }
 
+static void *xgbe_alloc_node(size_t size, int node)
+{
+	void *mem;
+
+	mem = kzalloc_node(size, GFP_KERNEL, node);
+	if (!mem)
+		mem = kzalloc(size, GFP_KERNEL);
+
+	return mem;
+}
+
+static void *xgbe_dma_alloc_node(struct device *dev, size_t size,
+				 dma_addr_t *dma, int node)
+{
+	void *mem;
+	int cur_node = dev_to_node(dev);
+
+	set_dev_node(dev, node);
+	mem = dma_alloc_coherent(dev, size, dma, GFP_KERNEL);
+	set_dev_node(dev, cur_node);
+
+	if (!mem)
+		mem = dma_alloc_coherent(dev, size, dma, GFP_KERNEL);
+
+	return mem;
+}
+
 static int xgbe_init_ring(struct xgbe_prv_data *pdata,
 			  struct xgbe_ring *ring, unsigned int rdesc_count)
 {
-	DBGPR("-->xgbe_init_ring\n");
+	size_t size;
 
 	if (!ring)
 		return 0;
 
 	/* Descriptors */
+	size = rdesc_count * sizeof(struct xgbe_ring_desc);
+
 	ring->rdesc_count = rdesc_count;
-	ring->rdesc = dma_alloc_coherent(pdata->dev,
-					 (sizeof(struct xgbe_ring_desc) *
-					  rdesc_count), &ring->rdesc_dma,
-					 GFP_KERNEL);
+	ring->rdesc = xgbe_dma_alloc_node(pdata->dev, size, &ring->rdesc_dma,
+					  ring->node);
 	if (!ring->rdesc)
 		return -ENOMEM;
 
 	/* Descriptor information */
-	ring->rdata = kcalloc(rdesc_count, sizeof(struct xgbe_ring_data),
-			      GFP_KERNEL);
+	size = rdesc_count * sizeof(struct xgbe_ring_data);
+
+	ring->rdata = xgbe_alloc_node(size, ring->node);
 	if (!ring->rdata)
 		return -ENOMEM;
 
 	netif_dbg(pdata, drv, pdata->netdev,
-		  "rdesc=%p, rdesc_dma=%pad, rdata=%p\n",
-		  ring->rdesc, &ring->rdesc_dma, ring->rdata);
-
-	DBGPR("<--xgbe_init_ring\n");
+		  "rdesc=%p, rdesc_dma=%pad, rdata=%p, node=%d\n",
+		  ring->rdesc, &ring->rdesc_dma, ring->rdata, ring->node);
 
 	return 0;
 }
@@ -223,10 +140,8 @@ static int xgbe_alloc_ring_resources(struct xgbe_prv_data *pdata)
 	unsigned int i;
 	int ret;
 
-	DBGPR("-->xgbe_alloc_ring_resources\n");
-
-	channel = pdata->channel;
-	for (i = 0; i < pdata->channel_count; i++, channel++) {
+	for (i = 0; i < pdata->channel_count; i++) {
+		channel = pdata->channel[i];
 		netif_dbg(pdata, drv, pdata->netdev, "%s - Tx ring:\n",
 			  channel->name);
 
@@ -250,8 +165,6 @@ static int xgbe_alloc_ring_resources(struct xgbe_prv_data *pdata)
 		}
 	}
 
-	DBGPR("<--xgbe_alloc_ring_resources\n");
-
 	return 0;
 
 err_ring:
@@ -261,31 +174,42 @@ err_ring:
 }
 
 static int xgbe_alloc_pages(struct xgbe_prv_data *pdata,
-			    struct xgbe_page_alloc *pa, gfp_t gfp, int order)
+			    struct xgbe_page_alloc *pa, int alloc_order,
+			    int node)
 {
 	struct page *pages = NULL;
 	dma_addr_t pages_dma;
-	int ret;
+	gfp_t gfp;
+	int order;
+
+again:
+	order = alloc_order;
 
 	/* Try to obtain pages, decreasing order if necessary */
-	gfp |= __GFP_COLD | __GFP_COMP | __GFP_NOWARN;
+	gfp = GFP_ATOMIC | __GFP_COMP | __GFP_NOWARN;
 	while (order >= 0) {
-		pages = alloc_pages(gfp, order);
+		pages = alloc_pages_node(node, gfp, order);
 		if (pages)
 			break;
 
 		order--;
 	}
+
+	/* If we couldn't get local pages, try getting from anywhere */
+	if (!pages && (node != NUMA_NO_NODE)) {
+		node = NUMA_NO_NODE;
+		goto again;
+	}
+
 	if (!pages)
 		return -ENOMEM;
 
 	/* Map the pages */
 	pages_dma = dma_map_page(pdata->dev, pages, 0,
 				 PAGE_SIZE << order, DMA_FROM_DEVICE);
-	ret = dma_mapping_error(pdata->dev, pages_dma);
-	if (ret) {
+	if (dma_mapping_error(pdata->dev, pages_dma)) {
 		put_page(pages);
-		return ret;
+		return -ENOMEM;
 	}
 
 	pa->pages = pages;
@@ -324,25 +248,29 @@ static int xgbe_map_rx_buffer(struct xgbe_prv_data *pdata,
 			      struct xgbe_ring *ring,
 			      struct xgbe_ring_data *rdata)
 {
-	int order, ret;
+	int ret;
 
 	if (!ring->rx_hdr_pa.pages) {
-		ret = xgbe_alloc_pages(pdata, &ring->rx_hdr_pa, GFP_ATOMIC, 0);
+		ret = xgbe_alloc_pages(pdata, &ring->rx_hdr_pa, 0, ring->node);
 		if (ret)
 			return ret;
 	}
 
 	if (!ring->rx_buf_pa.pages) {
-		order = max_t(int, PAGE_ALLOC_COSTLY_ORDER - 1, 0);
-		ret = xgbe_alloc_pages(pdata, &ring->rx_buf_pa, GFP_ATOMIC,
-				       order);
+		ret = xgbe_alloc_pages(pdata, &ring->rx_buf_pa,
+				       PAGE_ALLOC_COSTLY_ORDER, ring->node);
 		if (ret)
 			return ret;
 	}
 
 	/* Set up the header page info */
-	xgbe_set_buffer_data(&rdata->rx.hdr, &ring->rx_hdr_pa,
-			     XGBE_SKB_ALLOC_SIZE);
+	if (pdata->netdev->features & NETIF_F_RXCSUM) {
+		xgbe_set_buffer_data(&rdata->rx.hdr, &ring->rx_hdr_pa,
+				     XGBE_SKB_ALLOC_SIZE);
+	} else {
+		xgbe_set_buffer_data(&rdata->rx.hdr, &ring->rx_hdr_pa,
+				     pdata->rx_buf_size);
+	}
 
 	/* Set up the buffer page info */
 	xgbe_set_buffer_data(&rdata->rx.buf, &ring->rx_buf_pa,
@@ -363,8 +291,8 @@ static void xgbe_wrapper_tx_descriptor_init(struct xgbe_prv_data *pdata)
 
 	DBGPR("-->xgbe_wrapper_tx_descriptor_init\n");
 
-	channel = pdata->channel;
-	for (i = 0; i < pdata->channel_count; i++, channel++) {
+	for (i = 0; i < pdata->channel_count; i++) {
+		channel = pdata->channel[i];
 		ring = channel->tx_ring;
 		if (!ring)
 			break;
@@ -404,8 +332,8 @@ static void xgbe_wrapper_rx_descriptor_init(struct xgbe_prv_data *pdata)
 
 	DBGPR("-->xgbe_wrapper_rx_descriptor_init\n");
 
-	channel = pdata->channel;
-	for (i = 0; i < pdata->channel_count; i++, channel++) {
+	for (i = 0; i < pdata->channel_count; i++) {
+		channel = pdata->channel[i];
 		ring = channel->rx_ring;
 		if (!ring)
 			break;
@@ -494,7 +422,7 @@ static int xgbe_map_tx_skb(struct xgbe_channel *channel, struct sk_buff *skb)
 	struct xgbe_ring *ring = channel->tx_ring;
 	struct xgbe_ring_data *rdata;
 	struct xgbe_packet_data *packet;
-	struct skb_frag_struct *frag;
+	skb_frag_t *frag;
 	dma_addr_t skb_dma;
 	unsigned int start_index, cur_index;
 	unsigned int offset, tso, vlan, datalen, len;
diff --git a/drivers/net/ethernet/amd/xgbe/xgbe-dev.c b/drivers/net/ethernet/amd/xgbe/xgbe-dev.c
index 24a687ce4388..e5391a2eca51 100644
--- a/drivers/net/ethernet/amd/xgbe/xgbe-dev.c
+++ b/drivers/net/ethernet/amd/xgbe/xgbe-dev.c
@@ -1,117 +1,8 @@
+// SPDX-License-Identifier: (GPL-2.0-or-later OR BSD-3-Clause)
 /*
- * AMD 10Gb Ethernet driver
- *
- * This file is available to you under your choice of the following two
- * licenses:
- *
- * License 1: GPLv2
- *
- * Copyright (c) 2014-2016 Advanced Micro Devices, Inc.
- *
- * This file is free software; you may copy, redistribute and/or modify
- * it under the terms of the GNU General Public License as published by
- * the Free Software Foundation, either version 2 of the License, or (at
- * your option) any later version.
- *
- * This file is distributed in the hope that it will be useful, but
- * WITHOUT ANY WARRANTY; without even the implied warranty of
- * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
- * General Public License for more details.
- *
- * You should have received a copy of the GNU General Public License
- * along with this program.  If not, see <http://www.gnu.org/licenses/>.
- *
- * This file incorporates work covered by the following copyright and
- * permission notice:
- *     The Synopsys DWC ETHER XGMAC Software Driver and documentation
- *     (hereinafter "Software") is an unsupported proprietary work of Synopsys,
- *     Inc. unless otherwise expressly agreed to in writing between Synopsys
- *     and you.
- *
- *     The Software IS NOT an item of Licensed Software or Licensed Product
- *     under any End User Software License Agreement or Agreement for Licensed
- *     Product with Synopsys or any supplement thereto.  Permission is hereby
- *     granted, free of charge, to any person obtaining a copy of this software
- *     annotated with this license and the Software, to deal in the Software
- *     without restriction, including without limitation the rights to use,
- *     copy, modify, merge, publish, distribute, sublicense, and/or sell copies
- *     of the Software, and to permit persons to whom the Software is furnished
- *     to do so, subject to the following conditions:
- *
- *     The above copyright notice and this permission notice shall be included
- *     in all copies or substantial portions of the Software.
- *
- *     THIS SOFTWARE IS BEING DISTRIBUTED BY SYNOPSYS SOLELY ON AN "AS IS"
- *     BASIS AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED
- *     TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A
- *     PARTICULAR PURPOSE ARE HEREBY DISCLAIMED. IN NO EVENT SHALL SYNOPSYS
- *     BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR
- *     CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF
- *     SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS
- *     INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN
- *     CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE)
- *     ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF
- *     THE POSSIBILITY OF SUCH DAMAGE.
- *
- *
- * License 2: Modified BSD
- *
- * Copyright (c) 2014-2016 Advanced Micro Devices, Inc.
- * All rights reserved.
- *
- * Redistribution and use in source and binary forms, with or without
- * modification, are permitted provided that the following conditions are met:
- *     * Redistributions of source code must retain the above copyright
- *       notice, this list of conditions and the following disclaimer.
- *     * Redistributions in binary form must reproduce the above copyright
- *       notice, this list of conditions and the following disclaimer in the
- *       documentation and/or other materials provided with the distribution.
- *     * Neither the name of Advanced Micro Devices, Inc. nor the
- *       names of its contributors may be used to endorse or promote products
- *       derived from this software without specific prior written permission.
- *
- * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS"
- * AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
- * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
- * ARE DISCLAIMED. IN NO EVENT SHALL <COPYRIGHT HOLDER> BE LIABLE FOR ANY
- * DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES
- * (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES;
- * LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND
- * ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
- * (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF
- * THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
- *
- * This file incorporates work covered by the following copyright and
- * permission notice:
- *     The Synopsys DWC ETHER XGMAC Software Driver and documentation
- *     (hereinafter "Software") is an unsupported proprietary work of Synopsys,
- *     Inc. unless otherwise expressly agreed to in writing between Synopsys
- *     and you.
- *
- *     The Software IS NOT an item of Licensed Software or Licensed Product
- *     under any End User Software License Agreement or Agreement for Licensed
- *     Product with Synopsys or any supplement thereto.  Permission is hereby
- *     granted, free of charge, to any person obtaining a copy of this software
- *     annotated with this license and the Software, to deal in the Software
- *     without restriction, including without limitation the rights to use,
- *     copy, modify, merge, publish, distribute, sublicense, and/or sell copies
- *     of the Software, and to permit persons to whom the Software is furnished
- *     to do so, subject to the following conditions:
- *
- *     The above copyright notice and this permission notice shall be included
- *     in all copies or substantial portions of the Software.
- *
- *     THIS SOFTWARE IS BEING DISTRIBUTED BY SYNOPSYS SOLELY ON AN "AS IS"
- *     BASIS AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED
- *     TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A
- *     PARTICULAR PURPOSE ARE HEREBY DISCLAIMED. IN NO EVENT SHALL SYNOPSYS
- *     BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR
- *     CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF
- *     SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS
- *     INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN
- *     CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE)
- *     ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF
- *     THE POSSIBILITY OF SUCH DAMAGE.
+ * Copyright (c) 2014-2025, Advanced Micro Devices, Inc.
+ * Copyright (c) 2014, Synopsys, Inc.
+ * All rights reserved
  */
 
 #include <linux/phy.h>
@@ -119,9 +10,12 @@
 #include <linux/clk.h>
 #include <linux/bitrev.h>
 #include <linux/crc32.h>
+#include <linux/crc32poly.h>
+#include <linux/pci.h>
 
 #include "xgbe.h"
 #include "xgbe-common.h"
+#include "xgbe-smn.h"
 
 static inline unsigned int xgbe_get_max_frame(struct xgbe_prv_data *pdata)
 {
@@ -174,58 +68,30 @@ static unsigned int xgbe_riwt_to_usec(struct xgbe_prv_data *pdata,
 	return ret;
 }
 
-static int xgbe_config_pblx8(struct xgbe_prv_data *pdata)
+static int xgbe_config_pbl_val(struct xgbe_prv_data *pdata)
 {
-	struct xgbe_channel *channel;
-	unsigned int i;
-
-	channel = pdata->channel;
-	for (i = 0; i < pdata->channel_count; i++, channel++)
-		XGMAC_DMA_IOWRITE_BITS(channel, DMA_CH_CR, PBLX8,
-				       pdata->pblx8);
-
-	return 0;
-}
-
-static int xgbe_get_tx_pbl_val(struct xgbe_prv_data *pdata)
-{
-	return XGMAC_DMA_IOREAD_BITS(pdata->channel, DMA_CH_TCR, PBL);
-}
-
-static int xgbe_config_tx_pbl_val(struct xgbe_prv_data *pdata)
-{
-	struct xgbe_channel *channel;
+	unsigned int pblx8, pbl;
 	unsigned int i;
 
-	channel = pdata->channel;
-	for (i = 0; i < pdata->channel_count; i++, channel++) {
-		if (!channel->tx_ring)
-			break;
+	pblx8 = DMA_PBL_X8_DISABLE;
+	pbl = pdata->pbl;
 
-		XGMAC_DMA_IOWRITE_BITS(channel, DMA_CH_TCR, PBL,
-				       pdata->tx_pbl);
+	if (pdata->pbl > 32) {
+		pblx8 = DMA_PBL_X8_ENABLE;
+		pbl >>= 3;
 	}
 
-	return 0;
-}
+	for (i = 0; i < pdata->channel_count; i++) {
+		XGMAC_DMA_IOWRITE_BITS(pdata->channel[i], DMA_CH_CR, PBLX8,
+				       pblx8);
 
-static int xgbe_get_rx_pbl_val(struct xgbe_prv_data *pdata)
-{
-	return XGMAC_DMA_IOREAD_BITS(pdata->channel, DMA_CH_RCR, PBL);
-}
+		if (pdata->channel[i]->tx_ring)
+			XGMAC_DMA_IOWRITE_BITS(pdata->channel[i], DMA_CH_TCR,
+					       PBL, pbl);
 
-static int xgbe_config_rx_pbl_val(struct xgbe_prv_data *pdata)
-{
-	struct xgbe_channel *channel;
-	unsigned int i;
-
-	channel = pdata->channel;
-	for (i = 0; i < pdata->channel_count; i++, channel++) {
-		if (!channel->rx_ring)
-			break;
-
-		XGMAC_DMA_IOWRITE_BITS(channel, DMA_CH_RCR, PBL,
-				       pdata->rx_pbl);
+		if (pdata->channel[i]->rx_ring)
+			XGMAC_DMA_IOWRITE_BITS(pdata->channel[i], DMA_CH_RCR,
+					       PBL, pbl);
 	}
 
 	return 0;
@@ -233,15 +99,13 @@ static int xgbe_config_rx_pbl_val(struct xgbe_prv_data *pdata)
 
 static int xgbe_config_osp_mode(struct xgbe_prv_data *pdata)
 {
-	struct xgbe_channel *channel;
 	unsigned int i;
 
-	channel = pdata->channel;
-	for (i = 0; i < pdata->channel_count; i++, channel++) {
-		if (!channel->tx_ring)
+	for (i = 0; i < pdata->channel_count; i++) {
+		if (!pdata->channel[i]->tx_ring)
 			break;
 
-		XGMAC_DMA_IOWRITE_BITS(channel, DMA_CH_TCR, OSP,
+		XGMAC_DMA_IOWRITE_BITS(pdata->channel[i], DMA_CH_TCR, OSP,
 				       pdata->tx_osp_mode);
 	}
 
@@ -292,15 +156,13 @@ static int xgbe_config_tx_threshold(struct xgbe_prv_data *pdata,
 
 static int xgbe_config_rx_coalesce(struct xgbe_prv_data *pdata)
 {
-	struct xgbe_channel *channel;
 	unsigned int i;
 
-	channel = pdata->channel;
-	for (i = 0; i < pdata->channel_count; i++, channel++) {
-		if (!channel->rx_ring)
+	for (i = 0; i < pdata->channel_count; i++) {
+		if (!pdata->channel[i]->rx_ring)
 			break;
 
-		XGMAC_DMA_IOWRITE_BITS(channel, DMA_CH_RIWT, RWT,
+		XGMAC_DMA_IOWRITE_BITS(pdata->channel[i], DMA_CH_RIWT, RWT,
 				       pdata->rx_riwt);
 	}
 
@@ -314,49 +176,55 @@ static int xgbe_config_tx_coalesce(struct xgbe_prv_data *pdata)
 
 static void xgbe_config_rx_buffer_size(struct xgbe_prv_data *pdata)
 {
-	struct xgbe_channel *channel;
 	unsigned int i;
 
-	channel = pdata->channel;
-	for (i = 0; i < pdata->channel_count; i++, channel++) {
-		if (!channel->rx_ring)
+	for (i = 0; i < pdata->channel_count; i++) {
+		if (!pdata->channel[i]->rx_ring)
 			break;
 
-		XGMAC_DMA_IOWRITE_BITS(channel, DMA_CH_RCR, RBSZ,
+		XGMAC_DMA_IOWRITE_BITS(pdata->channel[i], DMA_CH_RCR, RBSZ,
 				       pdata->rx_buf_size);
 	}
 }
 
 static void xgbe_config_tso_mode(struct xgbe_prv_data *pdata)
 {
-	struct xgbe_channel *channel;
 	unsigned int i;
 
-	channel = pdata->channel;
-	for (i = 0; i < pdata->channel_count; i++, channel++) {
-		if (!channel->tx_ring)
+	for (i = 0; i < pdata->channel_count; i++) {
+		if (!pdata->channel[i]->tx_ring)
 			break;
 
-		XGMAC_DMA_IOWRITE_BITS(channel, DMA_CH_TCR, TSE, 1);
+		XGMAC_DMA_IOWRITE_BITS(pdata->channel[i], DMA_CH_TCR, TSE, 1);
 	}
 }
 
 static void xgbe_config_sph_mode(struct xgbe_prv_data *pdata)
 {
-	struct xgbe_channel *channel;
 	unsigned int i;
 
-	channel = pdata->channel;
-	for (i = 0; i < pdata->channel_count; i++, channel++) {
-		if (!channel->rx_ring)
+	for (i = 0; i < pdata->channel_count; i++) {
+		if (!pdata->channel[i]->rx_ring)
 			break;
 
-		XGMAC_DMA_IOWRITE_BITS(channel, DMA_CH_CR, SPH, 1);
+		XGMAC_DMA_IOWRITE_BITS(pdata->channel[i], DMA_CH_CR, SPH, 1);
 	}
 
 	XGMAC_IOWRITE_BITS(pdata, MAC_RCR, HDSMS, XGBE_SPH_HDSMS_SIZE);
 }
 
+static void xgbe_disable_sph_mode(struct xgbe_prv_data *pdata)
+{
+	unsigned int i;
+
+	for (i = 0; i < pdata->channel_count; i++) {
+		if (!pdata->channel[i]->rx_ring)
+			break;
+
+		XGMAC_DMA_IOWRITE_BITS(pdata->channel[i], DMA_CH_CR, SPH, 0);
+	}
+}
+
 static int xgbe_write_rss_reg(struct xgbe_prv_data *pdata, unsigned int type,
 			      unsigned int index, unsigned int val)
 {
@@ -517,19 +385,72 @@ static bool xgbe_is_pfc_queue(struct xgbe_prv_data *pdata,
 	return false;
 }
 
+static void xgbe_set_vxlan_id(struct xgbe_prv_data *pdata)
+{
+	/* Program the VXLAN port */
+	XGMAC_IOWRITE_BITS(pdata, MAC_TIR, TNID, pdata->vxlan_port);
+
+	netif_dbg(pdata, drv, pdata->netdev, "VXLAN tunnel id set to %hx\n",
+		  pdata->vxlan_port);
+}
+
+static void xgbe_enable_vxlan(struct xgbe_prv_data *pdata)
+{
+	if (!pdata->hw_feat.vxn)
+		return;
+
+	/* Program the VXLAN port */
+	xgbe_set_vxlan_id(pdata);
+
+	/* Allow for IPv6/UDP zero-checksum VXLAN packets */
+	XGMAC_IOWRITE_BITS(pdata, MAC_PFR, VUCC, 1);
+
+	/* Enable VXLAN tunneling mode */
+	XGMAC_IOWRITE_BITS(pdata, MAC_TCR, VNM, 0);
+	XGMAC_IOWRITE_BITS(pdata, MAC_TCR, VNE, 1);
+
+	netif_dbg(pdata, drv, pdata->netdev, "VXLAN acceleration enabled\n");
+}
+
+static void xgbe_disable_vxlan(struct xgbe_prv_data *pdata)
+{
+	if (!pdata->hw_feat.vxn)
+		return;
+
+	/* Disable tunneling mode */
+	XGMAC_IOWRITE_BITS(pdata, MAC_TCR, VNE, 0);
+
+	/* Clear IPv6/UDP zero-checksum VXLAN packets setting */
+	XGMAC_IOWRITE_BITS(pdata, MAC_PFR, VUCC, 0);
+
+	/* Clear the VXLAN port */
+	XGMAC_IOWRITE_BITS(pdata, MAC_TIR, TNID, 0);
+
+	netif_dbg(pdata, drv, pdata->netdev, "VXLAN acceleration disabled\n");
+}
+
+static unsigned int xgbe_get_fc_queue_count(struct xgbe_prv_data *pdata)
+{
+	unsigned int max_q_count = XGMAC_MAX_FLOW_CONTROL_QUEUES;
+
+	/* From MAC ver 30H the TFCR is per priority, instead of per queue */
+	if (XGMAC_GET_BITS(pdata->hw_feat.version, MAC_VR, SNPSVER) >= 0x30)
+		return max_q_count;
+	else
+		return min_t(unsigned int, pdata->tx_q_count, max_q_count);
+}
+
 static int xgbe_disable_tx_flow_control(struct xgbe_prv_data *pdata)
 {
-	unsigned int max_q_count, q_count;
 	unsigned int reg, reg_val;
-	unsigned int i;
+	unsigned int i, q_count;
 
 	/* Clear MTL flow control */
 	for (i = 0; i < pdata->rx_q_count; i++)
 		XGMAC_MTL_IOWRITE_BITS(pdata, i, MTL_Q_RQOMR, EHFC, 0);
 
 	/* Clear MAC flow control */
-	max_q_count = XGMAC_MAX_FLOW_CONTROL_QUEUES;
-	q_count = min_t(unsigned int, pdata->tx_q_count, max_q_count);
+	q_count = xgbe_get_fc_queue_count(pdata);
 	reg = MAC_Q0TFCR;
 	for (i = 0; i < q_count; i++) {
 		reg_val = XGMAC_IOREAD(pdata, reg);
@@ -546,9 +467,8 @@ static int xgbe_enable_tx_flow_control(struct xgbe_prv_data *pdata)
 {
 	struct ieee_pfc *pfc = pdata->pfc;
 	struct ieee_ets *ets = pdata->ets;
-	unsigned int max_q_count, q_count;
 	unsigned int reg, reg_val;
-	unsigned int i;
+	unsigned int i, q_count;
 
 	/* Set MTL flow control */
 	for (i = 0; i < pdata->rx_q_count; i++) {
@@ -572,8 +492,7 @@ static int xgbe_enable_tx_flow_control(struct xgbe_prv_data *pdata)
 	}
 
 	/* Set MAC flow control */
-	max_q_count = XGMAC_MAX_FLOW_CONTROL_QUEUES;
-	q_count = min_t(unsigned int, pdata->tx_q_count, max_q_count);
+	q_count = xgbe_get_fc_queue_count(pdata);
 	reg = MAC_Q0TFCR;
 	for (i = 0; i < q_count; i++) {
 		reg_val = XGMAC_IOREAD(pdata, reg);
@@ -643,31 +562,38 @@ static void xgbe_config_flow_control(struct xgbe_prv_data *pdata)
 static void xgbe_enable_dma_interrupts(struct xgbe_prv_data *pdata)
 {
 	struct xgbe_channel *channel;
-	unsigned int dma_ch_isr, dma_ch_ier;
-	unsigned int i;
+	unsigned int i, ver;
 
 	/* Set the interrupt mode if supported */
 	if (pdata->channel_irq_mode)
 		XGMAC_IOWRITE_BITS(pdata, DMA_MR, INTM,
 				   pdata->channel_irq_mode);
 
-	channel = pdata->channel;
-	for (i = 0; i < pdata->channel_count; i++, channel++) {
+	ver = XGMAC_GET_BITS(pdata->hw_feat.version, MAC_VR, SNPSVER);
+
+	for (i = 0; i < pdata->channel_count; i++) {
+		channel = pdata->channel[i];
+
 		/* Clear all the interrupts which are set */
-		dma_ch_isr = XGMAC_DMA_IOREAD(channel, DMA_CH_SR);
-		XGMAC_DMA_IOWRITE(channel, DMA_CH_SR, dma_ch_isr);
+		XGMAC_DMA_IOWRITE(channel, DMA_CH_SR,
+				  XGMAC_DMA_IOREAD(channel, DMA_CH_SR));
 
 		/* Clear all interrupt enable bits */
-		dma_ch_ier = 0;
+		channel->curr_ier = 0;
 
 		/* Enable following interrupts
 		 *   NIE  - Normal Interrupt Summary Enable
 		 *   AIE  - Abnormal Interrupt Summary Enable
 		 *   FBEE - Fatal Bus Error Enable
 		 */
-		XGMAC_SET_BITS(dma_ch_ier, DMA_CH_IER, NIE, 1);
-		XGMAC_SET_BITS(dma_ch_ier, DMA_CH_IER, AIE, 1);
-		XGMAC_SET_BITS(dma_ch_ier, DMA_CH_IER, FBEE, 1);
+		if (ver < 0x21) {
+			XGMAC_SET_BITS(channel->curr_ier, DMA_CH_IER, NIE20, 1);
+			XGMAC_SET_BITS(channel->curr_ier, DMA_CH_IER, AIE20, 1);
+		} else {
+			XGMAC_SET_BITS(channel->curr_ier, DMA_CH_IER, NIE, 1);
+			XGMAC_SET_BITS(channel->curr_ier, DMA_CH_IER, AIE, 1);
+		}
+		XGMAC_SET_BITS(channel->curr_ier, DMA_CH_IER, FBEE, 1);
 
 		if (channel->tx_ring) {
 			/* Enable the following Tx interrupts
@@ -676,7 +602,8 @@ static void xgbe_enable_dma_interrupts(struct xgbe_prv_data *pdata)
 			 *          mode)
 			 */
 			if (!pdata->per_channel_irq || pdata->channel_irq_mode)
-				XGMAC_SET_BITS(dma_ch_ier, DMA_CH_IER, TIE, 1);
+				XGMAC_SET_BITS(channel->curr_ier,
+					       DMA_CH_IER, TIE, 1);
 		}
 		if (channel->rx_ring) {
 			/* Enable following Rx interrupts
@@ -685,12 +612,13 @@ static void xgbe_enable_dma_interrupts(struct xgbe_prv_data *pdata)
 			 *          per channel interrupts in edge triggered
 			 *          mode)
 			 */
-			XGMAC_SET_BITS(dma_ch_ier, DMA_CH_IER, RBUE, 1);
+			XGMAC_SET_BITS(channel->curr_ier, DMA_CH_IER, RBUE, 1);
 			if (!pdata->per_channel_irq || pdata->channel_irq_mode)
-				XGMAC_SET_BITS(dma_ch_ier, DMA_CH_IER, RIE, 1);
+				XGMAC_SET_BITS(channel->curr_ier,
+					       DMA_CH_IER, RIE, 1);
 		}
 
-		XGMAC_DMA_IOWRITE(channel, DMA_CH_IER, dma_ch_ier);
+		XGMAC_DMA_IOWRITE(channel, DMA_CH_IER, channel->curr_ier);
 	}
 }
 
@@ -791,6 +719,9 @@ static int xgbe_set_speed(struct xgbe_prv_data *pdata, int speed)
 	unsigned int ss;
 
 	switch (speed) {
+	case SPEED_10:
+		ss = 0x07;
+		break;
 	case SPEED_1000:
 		ss = 0x03;
 		break;
@@ -872,7 +803,6 @@ static int xgbe_disable_rx_vlan_filtering(struct xgbe_prv_data *pdata)
 
 static u32 xgbe_vid_crc32_le(__le16 vid_le)
 {
-	u32 poly = 0xedb88320;	/* CRCPOLY_LE */
 	u32 crc = ~0;
 	u32 temp = 0;
 	unsigned char *data = (unsigned char *)&vid_le;
@@ -889,7 +819,7 @@ static u32 xgbe_vid_crc32_le(__le16 vid_le)
 		data_byte >>= 1;
 
 		if (temp)
-			crc ^= poly;
+			crc ^= CRC32_POLY_LE;
 	}
 
 	return crc;
@@ -1065,7 +995,7 @@ static int xgbe_add_mac_addresses(struct xgbe_prv_data *pdata)
 	return 0;
 }
 
-static int xgbe_set_mac_address(struct xgbe_prv_data *pdata, u8 *addr)
+static int xgbe_set_mac_address(struct xgbe_prv_data *pdata, const u8 *addr)
 {
 	unsigned int mac_addr_hi, mac_addr_lo;
 
@@ -1125,18 +1055,19 @@ static int xgbe_set_gpio(struct xgbe_prv_data *pdata, unsigned int gpio)
 	return 0;
 }
 
-static int xgbe_read_mmd_regs_v2(struct xgbe_prv_data *pdata, int prtad,
-				 int mmd_reg)
+static unsigned int xgbe_get_mmd_address(struct xgbe_prv_data *pdata,
+					 int mmd_reg)
 {
-	unsigned long flags;
-	unsigned int mmd_address, index, offset;
-	int mmd_data;
-
-	if (mmd_reg & MII_ADDR_C45)
-		mmd_address = mmd_reg & ~MII_ADDR_C45;
-	else
-		mmd_address = (pdata->mdio_mmd << 16) | (mmd_reg & 0xffff);
+	return (mmd_reg & XGBE_ADDR_C45) ?
+		mmd_reg & ~XGBE_ADDR_C45 :
+		(pdata->mdio_mmd << 16) | (mmd_reg & 0xffff);
+}
 
+static void xgbe_get_pcs_index_and_offset(struct xgbe_prv_data *pdata,
+					  unsigned int mmd_address,
+					  unsigned int *index,
+					  unsigned int *offset)
+{
 	/* The PCS registers are accessed using mmio. The underlying
 	 * management interface uses indirect addressing to access the MMD
 	 * register sets. This requires accessing of the PCS register in two
@@ -1147,8 +1078,98 @@ static int xgbe_read_mmd_regs_v2(struct xgbe_prv_data *pdata, int prtad,
 	 * offset 1 bit and reading 16 bits of data.
 	 */
 	mmd_address <<= 1;
-	index = mmd_address & ~pdata->xpcs_window_mask;
-	offset = pdata->xpcs_window + (mmd_address & pdata->xpcs_window_mask);
+	*index = mmd_address & ~pdata->xpcs_window_mask;
+	*offset = pdata->xpcs_window + (mmd_address & pdata->xpcs_window_mask);
+}
+
+static int xgbe_read_mmd_regs_v3(struct xgbe_prv_data *pdata, int prtad,
+				 int mmd_reg)
+{
+	unsigned int mmd_address, index, offset;
+	u32 smn_address;
+	int mmd_data;
+	int ret;
+
+	mmd_address = xgbe_get_mmd_address(pdata, mmd_reg);
+
+	xgbe_get_pcs_index_and_offset(pdata, mmd_address, &index, &offset);
+
+	smn_address = pdata->smn_base + pdata->xpcs_window_sel_reg;
+	ret = amd_smn_write(0, smn_address, index);
+	if (ret)
+		return ret;
+
+	ret = amd_smn_read(0, pdata->smn_base + offset, &mmd_data);
+	if (ret)
+		return ret;
+
+	mmd_data = (offset % 4) ? FIELD_GET(XGBE_GEN_HI_MASK, mmd_data) :
+				  FIELD_GET(XGBE_GEN_LO_MASK, mmd_data);
+
+	return mmd_data;
+}
+
+static void xgbe_write_mmd_regs_v3(struct xgbe_prv_data *pdata, int prtad,
+				   int mmd_reg, int mmd_data)
+{
+	unsigned int pci_mmd_data, hi_mask, lo_mask;
+	unsigned int mmd_address, index, offset;
+	struct pci_dev *dev;
+	u32 smn_address;
+	int ret;
+
+	dev = pdata->pcidev;
+	mmd_address = xgbe_get_mmd_address(pdata, mmd_reg);
+
+	xgbe_get_pcs_index_and_offset(pdata, mmd_address, &index, &offset);
+
+	smn_address = pdata->smn_base + pdata->xpcs_window_sel_reg;
+	ret = amd_smn_write(0, smn_address, index);
+	if (ret) {
+		pci_err(dev, "Failed to write data 0x%x\n", index);
+		return;
+	}
+
+	ret = amd_smn_read(0, pdata->smn_base + offset, &pci_mmd_data);
+	if (ret) {
+		pci_err(dev, "Failed to read data\n");
+		return;
+	}
+
+	if (offset % 4) {
+		hi_mask = FIELD_PREP(XGBE_GEN_HI_MASK, mmd_data);
+		lo_mask = FIELD_GET(XGBE_GEN_LO_MASK, pci_mmd_data);
+	} else {
+		hi_mask = FIELD_PREP(XGBE_GEN_HI_MASK,
+				     FIELD_GET(XGBE_GEN_HI_MASK, pci_mmd_data));
+		lo_mask = FIELD_GET(XGBE_GEN_LO_MASK, mmd_data);
+	}
+
+	pci_mmd_data = hi_mask | lo_mask;
+
+	ret = amd_smn_write(0, smn_address, index);
+	if (ret) {
+		pci_err(dev, "Failed to write data 0x%x\n", index);
+		return;
+	}
+
+	ret = amd_smn_write(0, (pdata->smn_base + offset), pci_mmd_data);
+	if (ret) {
+		pci_err(dev, "Failed to write data 0x%x\n", pci_mmd_data);
+		return;
+	}
+}
+
+static int xgbe_read_mmd_regs_v2(struct xgbe_prv_data *pdata, int prtad,
+				 int mmd_reg)
+{
+	unsigned int mmd_address, index, offset;
+	unsigned long flags;
+	int mmd_data;
+
+	mmd_address = xgbe_get_mmd_address(pdata, mmd_reg);
+
+	xgbe_get_pcs_index_and_offset(pdata, mmd_address, &index, &offset);
 
 	spin_lock_irqsave(&pdata->xpcs_lock, flags);
 	XPCS32_IOWRITE(pdata, pdata->xpcs_window_sel_reg, index);
@@ -1164,23 +1185,9 @@ static void xgbe_write_mmd_regs_v2(struct xgbe_prv_data *pdata, int prtad,
 	unsigned long flags;
 	unsigned int mmd_address, index, offset;
 
-	if (mmd_reg & MII_ADDR_C45)
-		mmd_address = mmd_reg & ~MII_ADDR_C45;
-	else
-		mmd_address = (pdata->mdio_mmd << 16) | (mmd_reg & 0xffff);
+	mmd_address = xgbe_get_mmd_address(pdata, mmd_reg);
 
-	/* The PCS registers are accessed using mmio. The underlying
-	 * management interface uses indirect addressing to access the MMD
-	 * register sets. This requires accessing of the PCS register in two
-	 * phases, an address phase and a data phase.
-	 *
-	 * The mmio interface is based on 16-bit offsets and values. All
-	 * register offsets must therefore be adjusted by left shifting the
-	 * offset 1 bit and writing 16 bits of data.
-	 */
-	mmd_address <<= 1;
-	index = mmd_address & ~pdata->xpcs_window_mask;
-	offset = pdata->xpcs_window + (mmd_address & pdata->xpcs_window_mask);
+	xgbe_get_pcs_index_and_offset(pdata, mmd_address, &index, &offset);
 
 	spin_lock_irqsave(&pdata->xpcs_lock, flags);
 	XPCS32_IOWRITE(pdata, pdata->xpcs_window_sel_reg, index);
@@ -1195,10 +1202,7 @@ static int xgbe_read_mmd_regs_v1(struct xgbe_prv_data *pdata, int prtad,
 	unsigned int mmd_address;
 	int mmd_data;
 
-	if (mmd_reg & MII_ADDR_C45)
-		mmd_address = mmd_reg & ~MII_ADDR_C45;
-	else
-		mmd_address = (pdata->mdio_mmd << 16) | (mmd_reg & 0xffff);
+	mmd_address = xgbe_get_mmd_address(pdata, mmd_reg);
 
 	/* The PCS registers are accessed using mmio. The underlying APB3
 	 * management interface uses indirect addressing to access the MMD
@@ -1223,10 +1227,7 @@ static void xgbe_write_mmd_regs_v1(struct xgbe_prv_data *pdata, int prtad,
 	unsigned int mmd_address;
 	unsigned long flags;
 
-	if (mmd_reg & MII_ADDR_C45)
-		mmd_address = mmd_reg & ~MII_ADDR_C45;
-	else
-		mmd_address = (pdata->mdio_mmd << 16) | (mmd_reg & 0xffff);
+	mmd_address = xgbe_get_mmd_address(pdata, mmd_reg);
 
 	/* The PCS registers are accessed using mmio. The underlying APB3
 	 * management interface uses indirect addressing to access the MMD
@@ -1253,6 +1254,9 @@ static int xgbe_read_mmd_regs(struct xgbe_prv_data *pdata, int prtad,
 	case XGBE_XPCS_ACCESS_V2:
 	default:
 		return xgbe_read_mmd_regs_v2(pdata, prtad, mmd_reg);
+
+	case XGBE_XPCS_ACCESS_V3:
+		return xgbe_read_mmd_regs_v3(pdata, prtad, mmd_reg);
 	}
 }
 
@@ -1263,22 +1267,45 @@ static void xgbe_write_mmd_regs(struct xgbe_prv_data *pdata, int prtad,
 	case XGBE_XPCS_ACCESS_V1:
 		return xgbe_write_mmd_regs_v1(pdata, prtad, mmd_reg, mmd_data);
 
+	case XGBE_XPCS_ACCESS_V3:
+		return xgbe_write_mmd_regs_v3(pdata, prtad, mmd_reg, mmd_data);
+
 	case XGBE_XPCS_ACCESS_V2:
 	default:
 		return xgbe_write_mmd_regs_v2(pdata, prtad, mmd_reg, mmd_data);
 	}
 }
 
-static int xgbe_write_ext_mii_regs(struct xgbe_prv_data *pdata, int addr,
-				   int reg, u16 val)
+static unsigned int xgbe_create_mdio_sca_c22(int port, int reg)
 {
-	unsigned int mdio_sca, mdio_sccd;
+	unsigned int mdio_sca;
 
-	reinit_completion(&pdata->mdio_complete);
+	mdio_sca = 0;
+	XGMAC_SET_BITS(mdio_sca, MAC_MDIOSCAR, RA, reg);
+	XGMAC_SET_BITS(mdio_sca, MAC_MDIOSCAR, PA, port);
+
+	return mdio_sca;
+}
+
+static unsigned int xgbe_create_mdio_sca_c45(int port, unsigned int da, int reg)
+{
+	unsigned int mdio_sca;
 
 	mdio_sca = 0;
-	XGMAC_SET_BITS(mdio_sca, MAC_MDIOSCAR, REG, reg);
-	XGMAC_SET_BITS(mdio_sca, MAC_MDIOSCAR, DA, addr);
+	XGMAC_SET_BITS(mdio_sca, MAC_MDIOSCAR, RA, reg);
+	XGMAC_SET_BITS(mdio_sca, MAC_MDIOSCAR, PA, port);
+	XGMAC_SET_BITS(mdio_sca, MAC_MDIOSCAR, DA, da);
+
+	return mdio_sca;
+}
+
+static int xgbe_write_ext_mii_regs(struct xgbe_prv_data *pdata,
+				   unsigned int mdio_sca, u16 val)
+{
+	unsigned int mdio_sccd;
+
+	reinit_completion(&pdata->mdio_complete);
+
 	XGMAC_IOWRITE(pdata, MAC_MDIOSCAR, mdio_sca);
 
 	mdio_sccd = 0;
@@ -1295,16 +1322,33 @@ static int xgbe_write_ext_mii_regs(struct xgbe_prv_data *pdata, int addr,
 	return 0;
 }
 
-static int xgbe_read_ext_mii_regs(struct xgbe_prv_data *pdata, int addr,
-				  int reg)
+static int xgbe_write_ext_mii_regs_c22(struct xgbe_prv_data *pdata, int addr,
+				       int reg, u16 val)
 {
-	unsigned int mdio_sca, mdio_sccd;
+	unsigned int mdio_sca;
+
+	mdio_sca = xgbe_create_mdio_sca_c22(addr, reg);
+
+	return xgbe_write_ext_mii_regs(pdata, mdio_sca, val);
+}
+
+static int xgbe_write_ext_mii_regs_c45(struct xgbe_prv_data *pdata, int addr,
+				       int devad, int reg, u16 val)
+{
+	unsigned int mdio_sca;
+
+	mdio_sca = xgbe_create_mdio_sca_c45(addr, devad, reg);
+
+	return xgbe_write_ext_mii_regs(pdata, mdio_sca, val);
+}
+
+static int xgbe_read_ext_mii_regs(struct xgbe_prv_data *pdata,
+				  unsigned int mdio_sca)
+{
+	unsigned int mdio_sccd;
 
 	reinit_completion(&pdata->mdio_complete);
 
-	mdio_sca = 0;
-	XGMAC_SET_BITS(mdio_sca, MAC_MDIOSCAR, REG, reg);
-	XGMAC_SET_BITS(mdio_sca, MAC_MDIOSCAR, DA, addr);
 	XGMAC_IOWRITE(pdata, MAC_MDIOSCAR, mdio_sca);
 
 	mdio_sccd = 0;
@@ -1320,6 +1364,26 @@ static int xgbe_read_ext_mii_regs(struct xgbe_prv_data *pdata, int addr,
 	return XGMAC_IOREAD_BITS(pdata, MAC_MDIOSCCDR, DATA);
 }
 
+static int xgbe_read_ext_mii_regs_c22(struct xgbe_prv_data *pdata, int addr,
+				      int reg)
+{
+	unsigned int mdio_sca;
+
+	mdio_sca = xgbe_create_mdio_sca_c22(addr, reg);
+
+	return xgbe_read_ext_mii_regs(pdata, mdio_sca);
+}
+
+static int xgbe_read_ext_mii_regs_c45(struct xgbe_prv_data *pdata, int addr,
+				      int devad, int reg)
+{
+	unsigned int mdio_sca;
+
+	mdio_sca = xgbe_create_mdio_sca_c45(addr, devad, reg);
+
+	return xgbe_read_ext_mii_regs(pdata, mdio_sca);
+}
+
 static int xgbe_set_ext_mii_mode(struct xgbe_prv_data *pdata, unsigned int port,
 				 enum xgbe_mdio_mode mode)
 {
@@ -1494,114 +1558,6 @@ static void xgbe_rx_desc_init(struct xgbe_channel *channel)
 	DBGPR("<--rx_desc_init\n");
 }
 
-static void xgbe_update_tstamp_addend(struct xgbe_prv_data *pdata,
-				      unsigned int addend)
-{
-	/* Set the addend register value and tell the device */
-	XGMAC_IOWRITE(pdata, MAC_TSAR, addend);
-	XGMAC_IOWRITE_BITS(pdata, MAC_TSCR, TSADDREG, 1);
-
-	/* Wait for addend update to complete */
-	while (XGMAC_IOREAD_BITS(pdata, MAC_TSCR, TSADDREG))
-		udelay(5);
-}
-
-static void xgbe_set_tstamp_time(struct xgbe_prv_data *pdata, unsigned int sec,
-				 unsigned int nsec)
-{
-	/* Set the time values and tell the device */
-	XGMAC_IOWRITE(pdata, MAC_STSUR, sec);
-	XGMAC_IOWRITE(pdata, MAC_STNUR, nsec);
-	XGMAC_IOWRITE_BITS(pdata, MAC_TSCR, TSINIT, 1);
-
-	/* Wait for time update to complete */
-	while (XGMAC_IOREAD_BITS(pdata, MAC_TSCR, TSINIT))
-		udelay(5);
-}
-
-static u64 xgbe_get_tstamp_time(struct xgbe_prv_data *pdata)
-{
-	u64 nsec;
-
-	nsec = XGMAC_IOREAD(pdata, MAC_STSR);
-	nsec *= NSEC_PER_SEC;
-	nsec += XGMAC_IOREAD(pdata, MAC_STNR);
-
-	return nsec;
-}
-
-static u64 xgbe_get_tx_tstamp(struct xgbe_prv_data *pdata)
-{
-	unsigned int tx_snr, tx_ssr;
-	u64 nsec;
-
-	if (pdata->vdata->tx_tstamp_workaround) {
-		tx_snr = XGMAC_IOREAD(pdata, MAC_TXSNR);
-		tx_ssr = XGMAC_IOREAD(pdata, MAC_TXSSR);
-	} else {
-		tx_ssr = XGMAC_IOREAD(pdata, MAC_TXSSR);
-		tx_snr = XGMAC_IOREAD(pdata, MAC_TXSNR);
-	}
-
-	if (XGMAC_GET_BITS(tx_snr, MAC_TXSNR, TXTSSTSMIS))
-		return 0;
-
-	nsec = tx_ssr;
-	nsec *= NSEC_PER_SEC;
-	nsec += tx_snr;
-
-	return nsec;
-}
-
-static void xgbe_get_rx_tstamp(struct xgbe_packet_data *packet,
-			       struct xgbe_ring_desc *rdesc)
-{
-	u64 nsec;
-
-	if (XGMAC_GET_BITS_LE(rdesc->desc3, RX_CONTEXT_DESC3, TSA) &&
-	    !XGMAC_GET_BITS_LE(rdesc->desc3, RX_CONTEXT_DESC3, TSD)) {
-		nsec = le32_to_cpu(rdesc->desc1);
-		nsec <<= 32;
-		nsec |= le32_to_cpu(rdesc->desc0);
-		if (nsec != 0xffffffffffffffffULL) {
-			packet->rx_tstamp = nsec;
-			XGMAC_SET_BITS(packet->attributes, RX_PACKET_ATTRIBUTES,
-				       RX_TSTAMP, 1);
-		}
-	}
-}
-
-static int xgbe_config_tstamp(struct xgbe_prv_data *pdata,
-			      unsigned int mac_tscr)
-{
-	/* Set one nano-second accuracy */
-	XGMAC_SET_BITS(mac_tscr, MAC_TSCR, TSCTRLSSR, 1);
-
-	/* Set fine timestamp update */
-	XGMAC_SET_BITS(mac_tscr, MAC_TSCR, TSCFUPDT, 1);
-
-	/* Overwrite earlier timestamps */
-	XGMAC_SET_BITS(mac_tscr, MAC_TSCR, TXTSSTSM, 1);
-
-	XGMAC_IOWRITE(pdata, MAC_TSCR, mac_tscr);
-
-	/* Exit if timestamping is not enabled */
-	if (!XGMAC_GET_BITS(mac_tscr, MAC_TSCR, TSENA))
-		return 0;
-
-	/* Initialize time registers */
-	XGMAC_IOWRITE_BITS(pdata, MAC_SSIR, SSINC, XGBE_TSTAMP_SSINC);
-	XGMAC_IOWRITE_BITS(pdata, MAC_SSIR, SNSINC, XGBE_TSTAMP_SNSINC);
-	xgbe_update_tstamp_addend(pdata, pdata->tstamp_addend);
-	xgbe_set_tstamp_time(pdata, 0, 0);
-
-	/* Initialize the timecounter */
-	timecounter_init(&pdata->tstamp_tc, &pdata->tstamp_cc,
-			 ktime_to_ns(ktime_get_real()));
-
-	return 0;
-}
-
 static void xgbe_tx_start_xmit(struct xgbe_channel *channel,
 			       struct xgbe_ring *ring)
 {
@@ -1634,7 +1590,8 @@ static void xgbe_dev_xmit(struct xgbe_channel *channel)
 	struct xgbe_ring_data *rdata;
 	struct xgbe_ring_desc *rdesc;
 	struct xgbe_packet_data *packet = &ring->packet_data;
-	unsigned int csum, tso, vlan;
+	unsigned int tx_packets, tx_bytes;
+	unsigned int csum, tso, vlan, vxlan;
 	unsigned int tso_context, vlan_context;
 	unsigned int tx_set_ic;
 	int start_index = ring->cur;
@@ -1643,12 +1600,17 @@ static void xgbe_dev_xmit(struct xgbe_channel *channel)
 
 	DBGPR("-->xgbe_dev_xmit\n");
 
+	tx_packets = packet->tx_packets;
+	tx_bytes = packet->tx_bytes;
+
 	csum = XGMAC_GET_BITS(packet->attributes, TX_PACKET_ATTRIBUTES,
 			      CSUM_ENABLE);
 	tso = XGMAC_GET_BITS(packet->attributes, TX_PACKET_ATTRIBUTES,
 			     TSO_ENABLE);
 	vlan = XGMAC_GET_BITS(packet->attributes, TX_PACKET_ATTRIBUTES,
 			      VLAN_CTAG);
+	vxlan = XGMAC_GET_BITS(packet->attributes, TX_PACKET_ATTRIBUTES,
+			       VXLAN);
 
 	if (tso && (packet->mss != ring->tx.cur_mss))
 		tso_context = 1;
@@ -1670,13 +1632,12 @@ static void xgbe_dev_xmit(struct xgbe_channel *channel)
 	 *     - Addition of Tx frame count to the frame count since the
 	 *       last interrupt was set does not exceed the frame count setting
 	 */
-	ring->coalesce_count += packet->tx_packets;
+	ring->coalesce_count += tx_packets;
 	if (!pdata->tx_frames)
 		tx_set_ic = 0;
-	else if (packet->tx_packets > pdata->tx_frames)
+	else if (tx_packets > pdata->tx_frames)
 		tx_set_ic = 1;
-	else if ((ring->coalesce_count % pdata->tx_frames) <
-		 packet->tx_packets)
+	else if ((ring->coalesce_count % pdata->tx_frames) < tx_packets)
 		tx_set_ic = 1;
 	else
 		tx_set_ic = 0;
@@ -1766,7 +1727,7 @@ static void xgbe_dev_xmit(struct xgbe_channel *channel)
 		XGMAC_SET_BITS_LE(rdesc->desc3, TX_NORMAL_DESC3, TCPHDRLEN,
 				  packet->tcp_header_len / 4);
 
-		pdata->ext_stats.tx_tso_packets++;
+		pdata->ext_stats.tx_tso_packets += tx_packets;
 	} else {
 		/* Enable CRC and Pad Insertion */
 		XGMAC_SET_BITS_LE(rdesc->desc3, TX_NORMAL_DESC3, CPC, 0);
@@ -1781,6 +1742,13 @@ static void xgbe_dev_xmit(struct xgbe_channel *channel)
 				  packet->length);
 	}
 
+	if (vxlan) {
+		XGMAC_SET_BITS_LE(rdesc->desc3, TX_NORMAL_DESC3, VNP,
+				  TX_NORMAL_DESC3_VXLAN_PACKET);
+
+		pdata->ext_stats.tx_vxlan_packets += packet->tx_packets;
+	}
+
 	for (i = cur_index - start_index + 1; i < packet->rdesc_count; i++) {
 		cur_index++;
 		rdata = XGBE_GET_DESC_DATA(ring, cur_index);
@@ -1814,8 +1782,11 @@ static void xgbe_dev_xmit(struct xgbe_channel *channel)
 		XGMAC_SET_BITS_LE(rdesc->desc2, TX_NORMAL_DESC2, IC, 1);
 
 	/* Save the Tx info to report back during cleanup */
-	rdata->tx.packets = packet->tx_packets;
-	rdata->tx.bytes = packet->tx_bytes;
+	rdata->tx.packets = tx_packets;
+	rdata->tx.bytes = tx_bytes;
+
+	pdata->ext_stats.txq_packets[channel->queue_index] += tx_packets;
+	pdata->ext_stats.txq_bytes[channel->queue_index] += tx_bytes;
 
 	/* In case the Tx DMA engine is running, make sure everything
 	 * is written to the descriptor(s) before setting the OWN bit
@@ -1836,7 +1807,7 @@ static void xgbe_dev_xmit(struct xgbe_channel *channel)
 	smp_wmb();
 
 	ring->cur = cur_index + 1;
-	if (!packet->skb->xmit_more ||
+	if (!netdev_xmit_more() ||
 	    netif_xmit_stopped(netdev_get_tx_queue(pdata->netdev,
 						   channel->queue_index)))
 		xgbe_tx_start_xmit(channel, ring);
@@ -1939,9 +1910,28 @@ static int xgbe_dev_read(struct xgbe_channel *channel)
 	rdata->rx.len = XGMAC_GET_BITS_LE(rdesc->desc3, RX_NORMAL_DESC3, PL);
 
 	/* Set checksum done indicator as appropriate */
-	if (netdev->features & NETIF_F_RXCSUM)
+	if (netdev->features & NETIF_F_RXCSUM) {
 		XGMAC_SET_BITS(packet->attributes, RX_PACKET_ATTRIBUTES,
 			       CSUM_DONE, 1);
+		XGMAC_SET_BITS(packet->attributes, RX_PACKET_ATTRIBUTES,
+			       TNPCSUM_DONE, 1);
+	}
+
+	/* Set the tunneled packet indicator */
+	if (XGMAC_GET_BITS_LE(rdesc->desc2, RX_NORMAL_DESC2, TNP)) {
+		XGMAC_SET_BITS(packet->attributes, RX_PACKET_ATTRIBUTES,
+			       TNP, 1);
+		pdata->ext_stats.rx_vxlan_packets++;
+
+		l34t = XGMAC_GET_BITS_LE(rdesc->desc3, RX_NORMAL_DESC3, L34T);
+		switch (l34t) {
+		case RX_DESC3_L34T_IPV4_UNKNOWN:
+		case RX_DESC3_L34T_IPV6_UNKNOWN:
+			XGMAC_SET_BITS(packet->attributes, RX_PACKET_ATTRIBUTES,
+				       TNPCSUM_DONE, 0);
+			break;
+		}
+	}
 
 	/* Check for errors (only valid in last descriptor) */
 	err = XGMAC_GET_BITS_LE(rdesc->desc3, RX_NORMAL_DESC3, ES);
@@ -1961,14 +1951,30 @@ static int xgbe_dev_read(struct xgbe_channel *channel)
 				  packet->vlan_ctag);
 		}
 	} else {
-		if ((etlt == 0x05) || (etlt == 0x06))
+		unsigned int tnp = XGMAC_GET_BITS(packet->attributes,
+						  RX_PACKET_ATTRIBUTES, TNP);
+
+		if ((etlt == 0x05) || (etlt == 0x06)) {
 			XGMAC_SET_BITS(packet->attributes, RX_PACKET_ATTRIBUTES,
 				       CSUM_DONE, 0);
-		else
+			XGMAC_SET_BITS(packet->attributes, RX_PACKET_ATTRIBUTES,
+				       TNPCSUM_DONE, 0);
+			pdata->ext_stats.rx_csum_errors++;
+		} else if (tnp && ((etlt == 0x09) || (etlt == 0x0a))) {
+			XGMAC_SET_BITS(packet->attributes, RX_PACKET_ATTRIBUTES,
+				       CSUM_DONE, 0);
+			XGMAC_SET_BITS(packet->attributes, RX_PACKET_ATTRIBUTES,
+				       TNPCSUM_DONE, 0);
+			pdata->ext_stats.rx_vxlan_csum_errors++;
+		} else {
 			XGMAC_SET_BITS(packet->errors, RX_PACKET_ERRORS,
 				       FRAME, 1);
+		}
 	}
 
+	pdata->ext_stats.rxq_packets[channel->queue_index]++;
+	pdata->ext_stats.rxq_bytes[channel->queue_index] += rdata->rx.len;
+
 	DBGPR("<--xgbe_dev_read: %s - descriptor=%u (cur=%d)\n", channel->name,
 	      ring->cur & (ring->rdesc_count - 1), ring->cur);
 
@@ -1990,44 +1996,40 @@ static int xgbe_is_last_desc(struct xgbe_ring_desc *rdesc)
 static int xgbe_enable_int(struct xgbe_channel *channel,
 			   enum xgbe_int int_id)
 {
-	unsigned int dma_ch_ier;
-
-	dma_ch_ier = XGMAC_DMA_IOREAD(channel, DMA_CH_IER);
-
 	switch (int_id) {
 	case XGMAC_INT_DMA_CH_SR_TI:
-		XGMAC_SET_BITS(dma_ch_ier, DMA_CH_IER, TIE, 1);
+		XGMAC_SET_BITS(channel->curr_ier, DMA_CH_IER, TIE, 1);
 		break;
 	case XGMAC_INT_DMA_CH_SR_TPS:
-		XGMAC_SET_BITS(dma_ch_ier, DMA_CH_IER, TXSE, 1);
+		XGMAC_SET_BITS(channel->curr_ier, DMA_CH_IER, TXSE, 1);
 		break;
 	case XGMAC_INT_DMA_CH_SR_TBU:
-		XGMAC_SET_BITS(dma_ch_ier, DMA_CH_IER, TBUE, 1);
+		XGMAC_SET_BITS(channel->curr_ier, DMA_CH_IER, TBUE, 1);
 		break;
 	case XGMAC_INT_DMA_CH_SR_RI:
-		XGMAC_SET_BITS(dma_ch_ier, DMA_CH_IER, RIE, 1);
+		XGMAC_SET_BITS(channel->curr_ier, DMA_CH_IER, RIE, 1);
 		break;
 	case XGMAC_INT_DMA_CH_SR_RBU:
-		XGMAC_SET_BITS(dma_ch_ier, DMA_CH_IER, RBUE, 1);
+		XGMAC_SET_BITS(channel->curr_ier, DMA_CH_IER, RBUE, 1);
 		break;
 	case XGMAC_INT_DMA_CH_SR_RPS:
-		XGMAC_SET_BITS(dma_ch_ier, DMA_CH_IER, RSE, 1);
+		XGMAC_SET_BITS(channel->curr_ier, DMA_CH_IER, RSE, 1);
 		break;
 	case XGMAC_INT_DMA_CH_SR_TI_RI:
-		XGMAC_SET_BITS(dma_ch_ier, DMA_CH_IER, TIE, 1);
-		XGMAC_SET_BITS(dma_ch_ier, DMA_CH_IER, RIE, 1);
+		XGMAC_SET_BITS(channel->curr_ier, DMA_CH_IER, TIE, 1);
+		XGMAC_SET_BITS(channel->curr_ier, DMA_CH_IER, RIE, 1);
 		break;
 	case XGMAC_INT_DMA_CH_SR_FBE:
-		XGMAC_SET_BITS(dma_ch_ier, DMA_CH_IER, FBEE, 1);
+		XGMAC_SET_BITS(channel->curr_ier, DMA_CH_IER, FBEE, 1);
 		break;
 	case XGMAC_INT_DMA_ALL:
-		dma_ch_ier |= channel->saved_ier;
+		channel->curr_ier |= channel->saved_ier;
 		break;
 	default:
 		return -1;
 	}
 
-	XGMAC_DMA_IOWRITE(channel, DMA_CH_IER, dma_ch_ier);
+	XGMAC_DMA_IOWRITE(channel, DMA_CH_IER, channel->curr_ier);
 
 	return 0;
 }
@@ -2035,45 +2037,41 @@ static int xgbe_enable_int(struct xgbe_channel *channel,
 static int xgbe_disable_int(struct xgbe_channel *channel,
 			    enum xgbe_int int_id)
 {
-	unsigned int dma_ch_ier;
-
-	dma_ch_ier = XGMAC_DMA_IOREAD(channel, DMA_CH_IER);
-
 	switch (int_id) {
 	case XGMAC_INT_DMA_CH_SR_TI:
-		XGMAC_SET_BITS(dma_ch_ier, DMA_CH_IER, TIE, 0);
+		XGMAC_SET_BITS(channel->curr_ier, DMA_CH_IER, TIE, 0);
 		break;
 	case XGMAC_INT_DMA_CH_SR_TPS:
-		XGMAC_SET_BITS(dma_ch_ier, DMA_CH_IER, TXSE, 0);
+		XGMAC_SET_BITS(channel->curr_ier, DMA_CH_IER, TXSE, 0);
 		break;
 	case XGMAC_INT_DMA_CH_SR_TBU:
-		XGMAC_SET_BITS(dma_ch_ier, DMA_CH_IER, TBUE, 0);
+		XGMAC_SET_BITS(channel->curr_ier, DMA_CH_IER, TBUE, 0);
 		break;
 	case XGMAC_INT_DMA_CH_SR_RI:
-		XGMAC_SET_BITS(dma_ch_ier, DMA_CH_IER, RIE, 0);
+		XGMAC_SET_BITS(channel->curr_ier, DMA_CH_IER, RIE, 0);
 		break;
 	case XGMAC_INT_DMA_CH_SR_RBU:
-		XGMAC_SET_BITS(dma_ch_ier, DMA_CH_IER, RBUE, 0);
+		XGMAC_SET_BITS(channel->curr_ier, DMA_CH_IER, RBUE, 0);
 		break;
 	case XGMAC_INT_DMA_CH_SR_RPS:
-		XGMAC_SET_BITS(dma_ch_ier, DMA_CH_IER, RSE, 0);
+		XGMAC_SET_BITS(channel->curr_ier, DMA_CH_IER, RSE, 0);
 		break;
 	case XGMAC_INT_DMA_CH_SR_TI_RI:
-		XGMAC_SET_BITS(dma_ch_ier, DMA_CH_IER, TIE, 0);
-		XGMAC_SET_BITS(dma_ch_ier, DMA_CH_IER, RIE, 0);
+		XGMAC_SET_BITS(channel->curr_ier, DMA_CH_IER, TIE, 0);
+		XGMAC_SET_BITS(channel->curr_ier, DMA_CH_IER, RIE, 0);
 		break;
 	case XGMAC_INT_DMA_CH_SR_FBE:
-		XGMAC_SET_BITS(dma_ch_ier, DMA_CH_IER, FBEE, 0);
+		XGMAC_SET_BITS(channel->curr_ier, DMA_CH_IER, FBEE, 0);
 		break;
 	case XGMAC_INT_DMA_ALL:
-		channel->saved_ier = dma_ch_ier & XGBE_DMA_INTERRUPT_MASK;
-		dma_ch_ier &= ~XGBE_DMA_INTERRUPT_MASK;
+		channel->saved_ier = channel->curr_ier;
+		channel->curr_ier = 0;
 		break;
 	default:
 		return -1;
 	}
 
-	XGMAC_DMA_IOWRITE(channel, DMA_CH_IER, dma_ch_ier);
+	XGMAC_DMA_IOWRITE(channel, DMA_CH_IER, channel->curr_ier);
 
 	return 0;
 }
@@ -2140,37 +2138,38 @@ static int xgbe_flush_tx_queues(struct xgbe_prv_data *pdata)
 
 static void xgbe_config_dma_bus(struct xgbe_prv_data *pdata)
 {
+	unsigned int sbmr;
+
+	sbmr = XGMAC_IOREAD(pdata, DMA_SBMR);
+
 	/* Set enhanced addressing mode */
-	XGMAC_IOWRITE_BITS(pdata, DMA_SBMR, EAME, 1);
+	XGMAC_SET_BITS(sbmr, DMA_SBMR, EAME, 1);
 
 	/* Set the System Bus mode */
-	XGMAC_IOWRITE_BITS(pdata, DMA_SBMR, UNDEF, 1);
-	XGMAC_IOWRITE_BITS(pdata, DMA_SBMR, BLEN_256, 1);
+	XGMAC_SET_BITS(sbmr, DMA_SBMR, UNDEF, 1);
+	XGMAC_SET_BITS(sbmr, DMA_SBMR, BLEN, pdata->blen >> 2);
+	XGMAC_SET_BITS(sbmr, DMA_SBMR, AAL, pdata->aal);
+	XGMAC_SET_BITS(sbmr, DMA_SBMR, RD_OSR_LMT, pdata->rd_osr_limit - 1);
+	XGMAC_SET_BITS(sbmr, DMA_SBMR, WR_OSR_LMT, pdata->wr_osr_limit - 1);
+
+	XGMAC_IOWRITE(pdata, DMA_SBMR, sbmr);
+
+	/* Set descriptor fetching threshold */
+	if (pdata->vdata->tx_desc_prefetch)
+		XGMAC_IOWRITE_BITS(pdata, DMA_TXEDMACR, TDPS,
+				   pdata->vdata->tx_desc_prefetch);
+
+	if (pdata->vdata->rx_desc_prefetch)
+		XGMAC_IOWRITE_BITS(pdata, DMA_RXEDMACR, RDPS,
+				   pdata->vdata->rx_desc_prefetch);
 }
 
 static void xgbe_config_dma_cache(struct xgbe_prv_data *pdata)
 {
-	unsigned int arcache, awcache;
-
-	arcache = 0;
-	XGMAC_SET_BITS(arcache, DMA_AXIARCR, DRC, pdata->arcache);
-	XGMAC_SET_BITS(arcache, DMA_AXIARCR, DRD, pdata->axdomain);
-	XGMAC_SET_BITS(arcache, DMA_AXIARCR, TEC, pdata->arcache);
-	XGMAC_SET_BITS(arcache, DMA_AXIARCR, TED, pdata->axdomain);
-	XGMAC_SET_BITS(arcache, DMA_AXIARCR, THC, pdata->arcache);
-	XGMAC_SET_BITS(arcache, DMA_AXIARCR, THD, pdata->axdomain);
-	XGMAC_IOWRITE(pdata, DMA_AXIARCR, arcache);
-
-	awcache = 0;
-	XGMAC_SET_BITS(awcache, DMA_AXIAWCR, DWC, pdata->awcache);
-	XGMAC_SET_BITS(awcache, DMA_AXIAWCR, DWD, pdata->axdomain);
-	XGMAC_SET_BITS(awcache, DMA_AXIAWCR, RPC, pdata->awcache);
-	XGMAC_SET_BITS(awcache, DMA_AXIAWCR, RPD, pdata->axdomain);
-	XGMAC_SET_BITS(awcache, DMA_AXIAWCR, RHC, pdata->awcache);
-	XGMAC_SET_BITS(awcache, DMA_AXIAWCR, RHD, pdata->axdomain);
-	XGMAC_SET_BITS(awcache, DMA_AXIAWCR, TDC, pdata->awcache);
-	XGMAC_SET_BITS(awcache, DMA_AXIAWCR, TDD, pdata->axdomain);
-	XGMAC_IOWRITE(pdata, DMA_AXIAWCR, awcache);
+	XGMAC_IOWRITE(pdata, DMA_AXIARCR, pdata->arcr);
+	XGMAC_IOWRITE(pdata, DMA_AXIAWCR, pdata->awcr);
+	if (pdata->awarcr)
+		XGMAC_IOWRITE(pdata, DMA_AXIAWARCR, pdata->awarcr);
 }
 
 static void xgbe_config_mtl_mode(struct xgbe_prv_data *pdata)
@@ -2732,9 +2731,19 @@ static void xgbe_config_jumbo_enable(struct xgbe_prv_data *pdata)
 {
 	unsigned int val;
 
-	val = (pdata->netdev->mtu > XGMAC_STD_PACKET_MTU) ? 1 : 0;
-
-	XGMAC_IOWRITE_BITS(pdata, MAC_RCR, JE, val);
+	if (pdata->netdev->mtu > XGMAC_JUMBO_PACKET_MTU) {
+		XGMAC_IOWRITE_BITS(pdata, MAC_RCR, GPSL,
+				   XGMAC_GIANT_PACKET_MTU);
+		XGMAC_IOWRITE_BITS(pdata, MAC_RCR, WD, 1);
+		XGMAC_IOWRITE_BITS(pdata, MAC_TCR, JD, 1);
+		XGMAC_IOWRITE_BITS(pdata, MAC_RCR, GPSLCE, 1);
+	} else {
+		val = pdata->netdev->mtu > XGMAC_STD_PACKET_MTU ? 1 : 0;
+		XGMAC_IOWRITE_BITS(pdata, MAC_RCR, GPSLCE, 0);
+		XGMAC_IOWRITE_BITS(pdata, MAC_RCR, WD, 0);
+		XGMAC_IOWRITE_BITS(pdata, MAC_TCR, JD, 0);
+		XGMAC_IOWRITE_BITS(pdata, MAC_RCR, JE, val);
+	}
 }
 
 static void xgbe_config_mac_speed(struct xgbe_prv_data *pdata)
@@ -3202,16 +3211,14 @@ static void xgbe_prepare_tx_stop(struct xgbe_prv_data *pdata,
 
 static void xgbe_enable_tx(struct xgbe_prv_data *pdata)
 {
-	struct xgbe_channel *channel;
 	unsigned int i;
 
 	/* Enable each Tx DMA channel */
-	channel = pdata->channel;
-	for (i = 0; i < pdata->channel_count; i++, channel++) {
-		if (!channel->tx_ring)
+	for (i = 0; i < pdata->channel_count; i++) {
+		if (!pdata->channel[i]->tx_ring)
 			break;
 
-		XGMAC_DMA_IOWRITE_BITS(channel, DMA_CH_TCR, ST, 1);
+		XGMAC_DMA_IOWRITE_BITS(pdata->channel[i], DMA_CH_TCR, ST, 1);
 	}
 
 	/* Enable each Tx queue */
@@ -3225,7 +3232,6 @@ static void xgbe_enable_tx(struct xgbe_prv_data *pdata)
 
 static void xgbe_disable_tx(struct xgbe_prv_data *pdata)
 {
-	struct xgbe_channel *channel;
 	unsigned int i;
 
 	/* Prepare for Tx DMA channel stop */
@@ -3240,12 +3246,11 @@ static void xgbe_disable_tx(struct xgbe_prv_data *pdata)
 		XGMAC_MTL_IOWRITE_BITS(pdata, i, MTL_Q_TQOMR, TXQEN, 0);
 
 	/* Disable each Tx DMA channel */
-	channel = pdata->channel;
-	for (i = 0; i < pdata->channel_count; i++, channel++) {
-		if (!channel->tx_ring)
+	for (i = 0; i < pdata->channel_count; i++) {
+		if (!pdata->channel[i]->tx_ring)
 			break;
 
-		XGMAC_DMA_IOWRITE_BITS(channel, DMA_CH_TCR, ST, 0);
+		XGMAC_DMA_IOWRITE_BITS(pdata->channel[i], DMA_CH_TCR, ST, 0);
 	}
 }
 
@@ -3277,16 +3282,14 @@ static void xgbe_prepare_rx_stop(struct xgbe_prv_data *pdata,
 
 static void xgbe_enable_rx(struct xgbe_prv_data *pdata)
 {
-	struct xgbe_channel *channel;
 	unsigned int reg_val, i;
 
 	/* Enable each Rx DMA channel */
-	channel = pdata->channel;
-	for (i = 0; i < pdata->channel_count; i++, channel++) {
-		if (!channel->rx_ring)
+	for (i = 0; i < pdata->channel_count; i++) {
+		if (!pdata->channel[i]->rx_ring)
 			break;
 
-		XGMAC_DMA_IOWRITE_BITS(channel, DMA_CH_RCR, SR, 1);
+		XGMAC_DMA_IOWRITE_BITS(pdata->channel[i], DMA_CH_RCR, SR, 1);
 	}
 
 	/* Enable each Rx queue */
@@ -3304,7 +3307,6 @@ static void xgbe_enable_rx(struct xgbe_prv_data *pdata)
 
 static void xgbe_disable_rx(struct xgbe_prv_data *pdata)
 {
-	struct xgbe_channel *channel;
 	unsigned int i;
 
 	/* Disable MAC Rx */
@@ -3321,27 +3323,24 @@ static void xgbe_disable_rx(struct xgbe_prv_data *pdata)
 	XGMAC_IOWRITE(pdata, MAC_RQC0R, 0);
 
 	/* Disable each Rx DMA channel */
-	channel = pdata->channel;
-	for (i = 0; i < pdata->channel_count; i++, channel++) {
-		if (!channel->rx_ring)
+	for (i = 0; i < pdata->channel_count; i++) {
+		if (!pdata->channel[i]->rx_ring)
 			break;
 
-		XGMAC_DMA_IOWRITE_BITS(channel, DMA_CH_RCR, SR, 0);
+		XGMAC_DMA_IOWRITE_BITS(pdata->channel[i], DMA_CH_RCR, SR, 0);
 	}
 }
 
 static void xgbe_powerup_tx(struct xgbe_prv_data *pdata)
 {
-	struct xgbe_channel *channel;
 	unsigned int i;
 
 	/* Enable each Tx DMA channel */
-	channel = pdata->channel;
-	for (i = 0; i < pdata->channel_count; i++, channel++) {
-		if (!channel->tx_ring)
+	for (i = 0; i < pdata->channel_count; i++) {
+		if (!pdata->channel[i]->tx_ring)
 			break;
 
-		XGMAC_DMA_IOWRITE_BITS(channel, DMA_CH_TCR, ST, 1);
+		XGMAC_DMA_IOWRITE_BITS(pdata->channel[i], DMA_CH_TCR, ST, 1);
 	}
 
 	/* Enable MAC Tx */
@@ -3350,7 +3349,6 @@ static void xgbe_powerup_tx(struct xgbe_prv_data *pdata)
 
 static void xgbe_powerdown_tx(struct xgbe_prv_data *pdata)
 {
-	struct xgbe_channel *channel;
 	unsigned int i;
 
 	/* Prepare for Tx DMA channel stop */
@@ -3361,42 +3359,37 @@ static void xgbe_powerdown_tx(struct xgbe_prv_data *pdata)
 	XGMAC_IOWRITE_BITS(pdata, MAC_TCR, TE, 0);
 
 	/* Disable each Tx DMA channel */
-	channel = pdata->channel;
-	for (i = 0; i < pdata->channel_count; i++, channel++) {
-		if (!channel->tx_ring)
+	for (i = 0; i < pdata->channel_count; i++) {
+		if (!pdata->channel[i]->tx_ring)
 			break;
 
-		XGMAC_DMA_IOWRITE_BITS(channel, DMA_CH_TCR, ST, 0);
+		XGMAC_DMA_IOWRITE_BITS(pdata->channel[i], DMA_CH_TCR, ST, 0);
 	}
 }
 
 static void xgbe_powerup_rx(struct xgbe_prv_data *pdata)
 {
-	struct xgbe_channel *channel;
 	unsigned int i;
 
 	/* Enable each Rx DMA channel */
-	channel = pdata->channel;
-	for (i = 0; i < pdata->channel_count; i++, channel++) {
-		if (!channel->rx_ring)
+	for (i = 0; i < pdata->channel_count; i++) {
+		if (!pdata->channel[i]->rx_ring)
 			break;
 
-		XGMAC_DMA_IOWRITE_BITS(channel, DMA_CH_RCR, SR, 1);
+		XGMAC_DMA_IOWRITE_BITS(pdata->channel[i], DMA_CH_RCR, SR, 1);
 	}
 }
 
 static void xgbe_powerdown_rx(struct xgbe_prv_data *pdata)
 {
-	struct xgbe_channel *channel;
 	unsigned int i;
 
 	/* Disable each Rx DMA channel */
-	channel = pdata->channel;
-	for (i = 0; i < pdata->channel_count; i++, channel++) {
-		if (!channel->rx_ring)
+	for (i = 0; i < pdata->channel_count; i++) {
+		if (!pdata->channel[i]->rx_ring)
 			break;
 
-		XGMAC_DMA_IOWRITE_BITS(channel, DMA_CH_RCR, SR, 0);
+		XGMAC_DMA_IOWRITE_BITS(pdata->channel[i], DMA_CH_RCR, SR, 0);
 	}
 }
 
@@ -3420,15 +3413,17 @@ static int xgbe_init(struct xgbe_prv_data *pdata)
 	xgbe_config_dma_bus(pdata);
 	xgbe_config_dma_cache(pdata);
 	xgbe_config_osp_mode(pdata);
-	xgbe_config_pblx8(pdata);
-	xgbe_config_tx_pbl_val(pdata);
-	xgbe_config_rx_pbl_val(pdata);
+	xgbe_config_pbl_val(pdata);
 	xgbe_config_rx_coalesce(pdata);
 	xgbe_config_tx_coalesce(pdata);
 	xgbe_config_rx_buffer_size(pdata);
 	xgbe_config_tso_mode(pdata);
-	xgbe_config_sph_mode(pdata);
-	xgbe_config_rss(pdata);
+
+	if (pdata->netdev->features & NETIF_F_RXCSUM) {
+		xgbe_config_sph_mode(pdata);
+		xgbe_config_rss(pdata);
+	}
+
 	desc_if->wrapper_tx_desc_init(pdata);
 	desc_if->wrapper_rx_desc_init(pdata);
 	xgbe_enable_dma_interrupts(pdata);
@@ -3497,8 +3492,10 @@ void xgbe_init_function_ptrs_dev(struct xgbe_hw_if *hw_if)
 	hw_if->set_speed = xgbe_set_speed;
 
 	hw_if->set_ext_mii_mode = xgbe_set_ext_mii_mode;
-	hw_if->read_ext_mii_regs = xgbe_read_ext_mii_regs;
-	hw_if->write_ext_mii_regs = xgbe_write_ext_mii_regs;
+	hw_if->read_ext_mii_regs_c22 = xgbe_read_ext_mii_regs_c22;
+	hw_if->write_ext_mii_regs_c22 = xgbe_write_ext_mii_regs_c22;
+	hw_if->read_ext_mii_regs_c45 = xgbe_read_ext_mii_regs_c45;
+	hw_if->write_ext_mii_regs_c45 = xgbe_write_ext_mii_regs_c45;
 
 	hw_if->set_gpio = xgbe_set_gpio;
 	hw_if->clr_gpio = xgbe_clr_gpio;
@@ -3550,25 +3547,11 @@ void xgbe_init_function_ptrs_dev(struct xgbe_hw_if *hw_if)
 	/* For TX DMA Operating on Second Frame config */
 	hw_if->config_osp_mode = xgbe_config_osp_mode;
 
-	/* For RX and TX PBL config */
-	hw_if->config_rx_pbl_val = xgbe_config_rx_pbl_val;
-	hw_if->get_rx_pbl_val = xgbe_get_rx_pbl_val;
-	hw_if->config_tx_pbl_val = xgbe_config_tx_pbl_val;
-	hw_if->get_tx_pbl_val = xgbe_get_tx_pbl_val;
-	hw_if->config_pblx8 = xgbe_config_pblx8;
-
 	/* For MMC statistics support */
 	hw_if->tx_mmc_int = xgbe_tx_mmc_int;
 	hw_if->rx_mmc_int = xgbe_rx_mmc_int;
 	hw_if->read_mmc_stats = xgbe_read_mmc_stats;
 
-	/* For PTP config */
-	hw_if->config_tstamp = xgbe_config_tstamp;
-	hw_if->update_tstamp_addend = xgbe_update_tstamp_addend;
-	hw_if->set_tstamp_time = xgbe_set_tstamp_time;
-	hw_if->get_tstamp_time = xgbe_get_tstamp_time;
-	hw_if->get_tx_tstamp = xgbe_get_tx_tstamp;
-
 	/* For Data Center Bridging config */
 	hw_if->config_tc = xgbe_config_tc;
 	hw_if->config_dcb_tc = xgbe_config_dcb_tc;
@@ -3584,5 +3567,14 @@ void xgbe_init_function_ptrs_dev(struct xgbe_hw_if *hw_if)
 	hw_if->disable_ecc_ded = xgbe_disable_ecc_ded;
 	hw_if->disable_ecc_sec = xgbe_disable_ecc_sec;
 
+	/* For VXLAN */
+	hw_if->enable_vxlan = xgbe_enable_vxlan;
+	hw_if->disable_vxlan = xgbe_disable_vxlan;
+	hw_if->set_vxlan_id = xgbe_set_vxlan_id;
+
+	/* For Split Header*/
+	hw_if->enable_sph = xgbe_config_sph_mode;
+	hw_if->disable_sph = xgbe_disable_sph_mode;
+
 	DBGPR("<--xgbe_init_function_ptrs\n");
 }
diff --git a/drivers/net/ethernet/amd/xgbe/xgbe-drv.c b/drivers/net/ethernet/amd/xgbe/xgbe-drv.c
index 5a2ad9c5faab..f0989aa01855 100644
--- a/drivers/net/ethernet/amd/xgbe/xgbe-drv.c
+++ b/drivers/net/ethernet/amd/xgbe/xgbe-drv.c
@@ -1,117 +1,8 @@
+// SPDX-License-Identifier: (GPL-2.0-or-later OR BSD-3-Clause)
 /*
- * AMD 10Gb Ethernet driver
- *
- * This file is available to you under your choice of the following two
- * licenses:
- *
- * License 1: GPLv2
- *
- * Copyright (c) 2014-2016 Advanced Micro Devices, Inc.
- *
- * This file is free software; you may copy, redistribute and/or modify
- * it under the terms of the GNU General Public License as published by
- * the Free Software Foundation, either version 2 of the License, or (at
- * your option) any later version.
- *
- * This file is distributed in the hope that it will be useful, but
- * WITHOUT ANY WARRANTY; without even the implied warranty of
- * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
- * General Public License for more details.
- *
- * You should have received a copy of the GNU General Public License
- * along with this program.  If not, see <http://www.gnu.org/licenses/>.
- *
- * This file incorporates work covered by the following copyright and
- * permission notice:
- *     The Synopsys DWC ETHER XGMAC Software Driver and documentation
- *     (hereinafter "Software") is an unsupported proprietary work of Synopsys,
- *     Inc. unless otherwise expressly agreed to in writing between Synopsys
- *     and you.
- *
- *     The Software IS NOT an item of Licensed Software or Licensed Product
- *     under any End User Software License Agreement or Agreement for Licensed
- *     Product with Synopsys or any supplement thereto.  Permission is hereby
- *     granted, free of charge, to any person obtaining a copy of this software
- *     annotated with this license and the Software, to deal in the Software
- *     without restriction, including without limitation the rights to use,
- *     copy, modify, merge, publish, distribute, sublicense, and/or sell copies
- *     of the Software, and to permit persons to whom the Software is furnished
- *     to do so, subject to the following conditions:
- *
- *     The above copyright notice and this permission notice shall be included
- *     in all copies or substantial portions of the Software.
- *
- *     THIS SOFTWARE IS BEING DISTRIBUTED BY SYNOPSYS SOLELY ON AN "AS IS"
- *     BASIS AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED
- *     TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A
- *     PARTICULAR PURPOSE ARE HEREBY DISCLAIMED. IN NO EVENT SHALL SYNOPSYS
- *     BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR
- *     CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF
- *     SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS
- *     INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN
- *     CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE)
- *     ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF
- *     THE POSSIBILITY OF SUCH DAMAGE.
- *
- *
- * License 2: Modified BSD
- *
- * Copyright (c) 2014-2016 Advanced Micro Devices, Inc.
- * All rights reserved.
- *
- * Redistribution and use in source and binary forms, with or without
- * modification, are permitted provided that the following conditions are met:
- *     * Redistributions of source code must retain the above copyright
- *       notice, this list of conditions and the following disclaimer.
- *     * Redistributions in binary form must reproduce the above copyright
- *       notice, this list of conditions and the following disclaimer in the
- *       documentation and/or other materials provided with the distribution.
- *     * Neither the name of Advanced Micro Devices, Inc. nor the
- *       names of its contributors may be used to endorse or promote products
- *       derived from this software without specific prior written permission.
- *
- * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS"
- * AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
- * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
- * ARE DISCLAIMED. IN NO EVENT SHALL <COPYRIGHT HOLDER> BE LIABLE FOR ANY
- * DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES
- * (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES;
- * LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND
- * ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
- * (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF
- * THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
- *
- * This file incorporates work covered by the following copyright and
- * permission notice:
- *     The Synopsys DWC ETHER XGMAC Software Driver and documentation
- *     (hereinafter "Software") is an unsupported proprietary work of Synopsys,
- *     Inc. unless otherwise expressly agreed to in writing between Synopsys
- *     and you.
- *
- *     The Software IS NOT an item of Licensed Software or Licensed Product
- *     under any End User Software License Agreement or Agreement for Licensed
- *     Product with Synopsys or any supplement thereto.  Permission is hereby
- *     granted, free of charge, to any person obtaining a copy of this software
- *     annotated with this license and the Software, to deal in the Software
- *     without restriction, including without limitation the rights to use,
- *     copy, modify, merge, publish, distribute, sublicense, and/or sell copies
- *     of the Software, and to permit persons to whom the Software is furnished
- *     to do so, subject to the following conditions:
- *
- *     The above copyright notice and this permission notice shall be included
- *     in all copies or substantial portions of the Software.
- *
- *     THIS SOFTWARE IS BEING DISTRIBUTED BY SYNOPSYS SOLELY ON AN "AS IS"
- *     BASIS AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED
- *     TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A
- *     PARTICULAR PURPOSE ARE HEREBY DISCLAIMED. IN NO EVENT SHALL SYNOPSYS
- *     BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR
- *     CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF
- *     SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS
- *     INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN
- *     CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE)
- *     ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF
- *     THE POSSIBILITY OF SUCH DAMAGE.
+ * Copyright (c) 2014-2025, Advanced Micro Devices, Inc.
+ * Copyright (c) 2014, Synopsys, Inc.
+ * All rights reserved
  */
 
 #include <linux/module.h>
@@ -119,11 +10,11 @@
 #include <linux/tcp.h>
 #include <linux/if_vlan.h>
 #include <linux/interrupt.h>
-#include <net/busy_poll.h>
 #include <linux/clk.h>
 #include <linux/if_ether.h>
 #include <linux/net_tstamp.h>
 #include <linux/phy.h>
+#include <net/vxlan.h>
 
 #include "xgbe.h"
 #include "xgbe-common.h"
@@ -136,21 +27,21 @@ static unsigned int ecc_ded_period = 600;
 
 #ifdef CONFIG_AMD_XGBE_HAVE_ECC
 /* Only expose the ECC parameters if supported */
-module_param(ecc_sec_info_threshold, uint, S_IWUSR | S_IRUGO);
+module_param(ecc_sec_info_threshold, uint, 0644);
 MODULE_PARM_DESC(ecc_sec_info_threshold,
 		 " ECC corrected error informational threshold setting");
 
-module_param(ecc_sec_warn_threshold, uint, S_IWUSR | S_IRUGO);
+module_param(ecc_sec_warn_threshold, uint, 0644);
 MODULE_PARM_DESC(ecc_sec_warn_threshold,
 		 " ECC corrected error warning threshold setting");
 
-module_param(ecc_sec_period, uint, S_IWUSR | S_IRUGO);
+module_param(ecc_sec_period, uint, 0644);
 MODULE_PARM_DESC(ecc_sec_period, " ECC corrected error period (in seconds)");
 
-module_param(ecc_ded_threshold, uint, S_IWUSR | S_IRUGO);
+module_param(ecc_ded_threshold, uint, 0644);
 MODULE_PARM_DESC(ecc_ded_threshold, " ECC detected error threshold setting");
 
-module_param(ecc_ded_period, uint, S_IWUSR | S_IRUGO);
+module_param(ecc_ded_period, uint, 0644);
 MODULE_PARM_DESC(ecc_ded_period, " ECC detected error period (in seconds)");
 #endif
 
@@ -158,81 +49,106 @@ static int xgbe_one_poll(struct napi_struct *, int);
 static int xgbe_all_poll(struct napi_struct *, int);
 static void xgbe_stop(struct xgbe_prv_data *);
 
+static void *xgbe_alloc_node(size_t size, int node)
+{
+	void *mem;
+
+	mem = kzalloc_node(size, GFP_KERNEL, node);
+	if (!mem)
+		mem = kzalloc(size, GFP_KERNEL);
+
+	return mem;
+}
+
+static void xgbe_free_channels(struct xgbe_prv_data *pdata)
+{
+	unsigned int i;
+
+	for (i = 0; i < ARRAY_SIZE(pdata->channel); i++) {
+		if (!pdata->channel[i])
+			continue;
+
+		kfree(pdata->channel[i]->rx_ring);
+		kfree(pdata->channel[i]->tx_ring);
+		kfree(pdata->channel[i]);
+
+		pdata->channel[i] = NULL;
+	}
+
+	pdata->channel_count = 0;
+}
+
 static int xgbe_alloc_channels(struct xgbe_prv_data *pdata)
 {
-	struct xgbe_channel *channel_mem, *channel;
-	struct xgbe_ring *tx_ring, *rx_ring;
+	struct xgbe_channel *channel;
+	struct xgbe_ring *ring;
 	unsigned int count, i;
-	int ret = -ENOMEM;
+	unsigned int cpu;
+	int node;
 
 	count = max_t(unsigned int, pdata->tx_ring_count, pdata->rx_ring_count);
+	for (i = 0; i < count; i++) {
+		/* Attempt to use a CPU on the node the device is on */
+		cpu = cpumask_local_spread(i, dev_to_node(pdata->dev));
 
-	channel_mem = kcalloc(count, sizeof(struct xgbe_channel), GFP_KERNEL);
-	if (!channel_mem)
-		goto err_channel;
-
-	tx_ring = kcalloc(pdata->tx_ring_count, sizeof(struct xgbe_ring),
-			  GFP_KERNEL);
-	if (!tx_ring)
-		goto err_tx_ring;
+		/* Set the allocation node based on the returned CPU */
+		node = cpu_to_node(cpu);
 
-	rx_ring = kcalloc(pdata->rx_ring_count, sizeof(struct xgbe_ring),
-			  GFP_KERNEL);
-	if (!rx_ring)
-		goto err_rx_ring;
+		channel = xgbe_alloc_node(sizeof(*channel), node);
+		if (!channel)
+			goto err_mem;
+		pdata->channel[i] = channel;
 
-	for (i = 0, channel = channel_mem; i < count; i++, channel++) {
 		snprintf(channel->name, sizeof(channel->name), "channel-%u", i);
 		channel->pdata = pdata;
 		channel->queue_index = i;
 		channel->dma_regs = pdata->xgmac_regs + DMA_CH_BASE +
 				    (DMA_CH_INC * i);
+		channel->node = node;
+		cpumask_set_cpu(cpu, &channel->affinity_mask);
 
 		if (pdata->per_channel_irq)
 			channel->dma_irq = pdata->channel_irq[i];
 
 		if (i < pdata->tx_ring_count) {
-			spin_lock_init(&tx_ring->lock);
-			channel->tx_ring = tx_ring++;
+			ring = xgbe_alloc_node(sizeof(*ring), node);
+			if (!ring)
+				goto err_mem;
+
+			spin_lock_init(&ring->lock);
+			ring->node = node;
+
+			channel->tx_ring = ring;
 		}
 
 		if (i < pdata->rx_ring_count) {
-			spin_lock_init(&rx_ring->lock);
-			channel->rx_ring = rx_ring++;
+			ring = xgbe_alloc_node(sizeof(*ring), node);
+			if (!ring)
+				goto err_mem;
+
+			spin_lock_init(&ring->lock);
+			ring->node = node;
+
+			channel->rx_ring = ring;
 		}
 
 		netif_dbg(pdata, drv, pdata->netdev,
+			  "%s: cpu=%u, node=%d\n", channel->name, cpu, node);
+
+		netif_dbg(pdata, drv, pdata->netdev,
 			  "%s: dma_regs=%p, dma_irq=%d, tx=%p, rx=%p\n",
 			  channel->name, channel->dma_regs, channel->dma_irq,
 			  channel->tx_ring, channel->rx_ring);
 	}
 
-	pdata->channel = channel_mem;
 	pdata->channel_count = count;
 
 	return 0;
 
-err_rx_ring:
-	kfree(tx_ring);
-
-err_tx_ring:
-	kfree(channel_mem);
-
-err_channel:
-	return ret;
-}
-
-static void xgbe_free_channels(struct xgbe_prv_data *pdata)
-{
-	if (!pdata->channel)
-		return;
-
-	kfree(pdata->channel->rx_ring);
-	kfree(pdata->channel->tx_ring);
-	kfree(pdata->channel);
+err_mem:
+	xgbe_free_channels(pdata);
 
-	pdata->channel = NULL;
-	pdata->channel_count = 0;
+	return -ENOMEM;
 }
 
 static inline unsigned int xgbe_tx_avail_desc(struct xgbe_ring *ring)
@@ -301,12 +217,10 @@ static void xgbe_enable_rx_tx_int(struct xgbe_prv_data *pdata,
 
 static void xgbe_enable_rx_tx_ints(struct xgbe_prv_data *pdata)
 {
-	struct xgbe_channel *channel;
 	unsigned int i;
 
-	channel = pdata->channel;
-	for (i = 0; i < pdata->channel_count; i++, channel++)
-		xgbe_enable_rx_tx_int(pdata, channel);
+	for (i = 0; i < pdata->channel_count; i++)
+		xgbe_enable_rx_tx_int(pdata, pdata->channel[i]);
 }
 
 static void xgbe_disable_rx_tx_int(struct xgbe_prv_data *pdata,
@@ -329,12 +243,10 @@ static void xgbe_disable_rx_tx_int(struct xgbe_prv_data *pdata,
 
 static void xgbe_disable_rx_tx_ints(struct xgbe_prv_data *pdata)
 {
-	struct xgbe_channel *channel;
 	unsigned int i;
 
-	channel = pdata->channel;
-	for (i = 0; i < pdata->channel_count; i++, channel++)
-		xgbe_disable_rx_tx_int(pdata, channel);
+	for (i = 0; i < pdata->channel_count; i++)
+		xgbe_disable_rx_tx_int(pdata, pdata->channel[i]);
 }
 
 static bool xgbe_ecc_sec(struct xgbe_prv_data *pdata, unsigned long *period,
@@ -382,9 +294,9 @@ static bool xgbe_ecc_ded(struct xgbe_prv_data *pdata, unsigned long *period,
 	return false;
 }
 
-static irqreturn_t xgbe_ecc_isr(int irq, void *data)
+static void xgbe_ecc_isr_bh_work(struct work_struct *work)
 {
-	struct xgbe_prv_data *pdata = data;
+	struct xgbe_prv_data *pdata = from_work(pdata, work, ecc_bh_work);
 	unsigned int ecc_isr;
 	bool stop = false;
 
@@ -435,12 +347,26 @@ out:
 	/* Clear all ECC interrupts */
 	XP_IOWRITE(pdata, XP_ECC_ISR, ecc_isr);
 
-	return IRQ_HANDLED;
+	/* Reissue interrupt if status is not clear */
+	if (pdata->vdata->irq_reissue_support)
+		XP_IOWRITE(pdata, XP_INT_REISSUE_EN, 1 << 1);
 }
 
-static irqreturn_t xgbe_isr(int irq, void *data)
+static irqreturn_t xgbe_ecc_isr(int irq, void *data)
 {
 	struct xgbe_prv_data *pdata = data;
+
+	if (pdata->isr_as_bh_work)
+		queue_work(system_bh_wq, &pdata->ecc_bh_work);
+	else
+		xgbe_ecc_isr_bh_work(&pdata->ecc_bh_work);
+
+	return IRQ_HANDLED;
+}
+
+static void xgbe_isr_bh_work(struct work_struct *work)
+{
+	struct xgbe_prv_data *pdata = from_work(pdata, work, dev_bh_work);
 	struct xgbe_hw_if *hw_if = &pdata->hw_if;
 	struct xgbe_channel *channel;
 	unsigned int dma_isr, dma_ch_isr;
@@ -461,7 +387,7 @@ static irqreturn_t xgbe_isr(int irq, void *data)
 		if (!(dma_isr & (1 << i)))
 			continue;
 
-		channel = pdata->channel + i;
+		channel = pdata->channel[i];
 
 		dma_ch_isr = XGMAC_DMA_IOREAD(channel, DMA_CH_SR);
 		netif_dbg(pdata, intr, pdata->netdev, "DMA_CH%u_ISR=%#010x\n",
@@ -479,7 +405,7 @@ static irqreturn_t xgbe_isr(int irq, void *data)
 				xgbe_disable_rx_tx_ints(pdata);
 
 				/* Turn on polling */
-				__napi_schedule_irqoff(&pdata->napi);
+				__napi_schedule(&pdata->napi);
 			}
 		} else {
 			/* Don't clear Rx/Tx status if doing per channel DMA
@@ -522,7 +448,7 @@ static irqreturn_t xgbe_isr(int irq, void *data)
 			if (XGMAC_GET_BITS(mac_tssr, MAC_TSSR, TXTSC)) {
 				/* Read Tx Timestamp to clear interrupt */
 				pdata->tx_tstamp =
-					hw_if->get_tx_tstamp(pdata);
+					xgbe_get_tx_tstamp(pdata);
 				queue_work(pdata->dev_workqueue,
 					   &pdata->tx_tstamp_work);
 			}
@@ -543,15 +469,36 @@ static irqreturn_t xgbe_isr(int irq, void *data)
 isr_done:
 	/* If there is not a separate AN irq, handle it here */
 	if (pdata->dev_irq == pdata->an_irq)
-		pdata->phy_if.an_isr(irq, pdata);
+		pdata->phy_if.an_isr(pdata);
 
 	/* If there is not a separate ECC irq, handle it here */
 	if (pdata->vdata->ecc_support && (pdata->dev_irq == pdata->ecc_irq))
-		xgbe_ecc_isr(irq, pdata);
+		xgbe_ecc_isr_bh_work(&pdata->ecc_bh_work);
 
 	/* If there is not a separate I2C irq, handle it here */
 	if (pdata->vdata->i2c_support && (pdata->dev_irq == pdata->i2c_irq))
-		pdata->i2c_if.i2c_isr(irq, pdata);
+		pdata->i2c_if.i2c_isr(pdata);
+
+	/* Reissue interrupt if status is not clear */
+	if (pdata->vdata->irq_reissue_support) {
+		unsigned int reissue_mask;
+
+		reissue_mask = 1 << 0;
+		if (!pdata->per_channel_irq)
+			reissue_mask |= 0xffff << 4;
+
+		XP_IOWRITE(pdata, XP_INT_REISSUE_EN, reissue_mask);
+	}
+}
+
+static irqreturn_t xgbe_isr(int irq, void *data)
+{
+	struct xgbe_prv_data *pdata = data;
+
+	if (pdata->isr_as_bh_work)
+		queue_work(system_bh_wq, &pdata->dev_bh_work);
+	else
+		xgbe_isr_bh_work(&pdata->dev_bh_work);
 
 	return IRQ_HANDLED;
 }
@@ -585,9 +532,10 @@ static irqreturn_t xgbe_dma_isr(int irq, void *data)
 	return IRQ_HANDLED;
 }
 
-static void xgbe_tx_timer(unsigned long data)
+static void xgbe_tx_timer(struct timer_list *t)
 {
-	struct xgbe_channel *channel = (struct xgbe_channel *)data;
+	struct xgbe_channel *channel = timer_container_of(channel, t,
+							  tx_timer);
 	struct xgbe_prv_data *pdata = channel->pdata;
 	struct napi_struct *napi;
 
@@ -623,13 +571,28 @@ static void xgbe_service(struct work_struct *work)
 	pdata->phy_if.phy_status(pdata);
 }
 
-static void xgbe_service_timer(unsigned long data)
+static void xgbe_service_timer(struct timer_list *t)
 {
-	struct xgbe_prv_data *pdata = (struct xgbe_prv_data *)data;
+	struct xgbe_prv_data *pdata = timer_container_of(pdata, t,
+							 service_timer);
+	struct xgbe_channel *channel;
+	unsigned int i;
 
 	queue_work(pdata->dev_workqueue, &pdata->service_work);
 
 	mod_timer(&pdata->service_timer, jiffies + HZ);
+
+	if (!pdata->tx_usecs)
+		return;
+
+	for (i = 0; i < pdata->channel_count; i++) {
+		channel = pdata->channel[i];
+		if (!channel->tx_ring || channel->tx_timer_active)
+			break;
+		channel->tx_timer_active = 1;
+		mod_timer(&channel->tx_timer,
+			  jiffies + usecs_to_jiffies(pdata->tx_usecs));
+	}
 }
 
 static void xgbe_init_timers(struct xgbe_prv_data *pdata)
@@ -637,16 +600,14 @@ static void xgbe_init_timers(struct xgbe_prv_data *pdata)
 	struct xgbe_channel *channel;
 	unsigned int i;
 
-	setup_timer(&pdata->service_timer, xgbe_service_timer,
-		    (unsigned long)pdata);
+	timer_setup(&pdata->service_timer, xgbe_service_timer, 0);
 
-	channel = pdata->channel;
-	for (i = 0; i < pdata->channel_count; i++, channel++) {
+	for (i = 0; i < pdata->channel_count; i++) {
+		channel = pdata->channel[i];
 		if (!channel->tx_ring)
 			break;
 
-		setup_timer(&channel->tx_timer, xgbe_tx_timer,
-			    (unsigned long)channel);
+		timer_setup(&channel->tx_timer, xgbe_tx_timer, 0);
 	}
 }
 
@@ -660,14 +621,16 @@ static void xgbe_stop_timers(struct xgbe_prv_data *pdata)
 	struct xgbe_channel *channel;
 	unsigned int i;
 
-	del_timer_sync(&pdata->service_timer);
+	timer_delete_sync(&pdata->service_timer);
 
-	channel = pdata->channel;
-	for (i = 0; i < pdata->channel_count; i++, channel++) {
+	for (i = 0; i < pdata->channel_count; i++) {
+		channel = pdata->channel[i];
 		if (!channel->tx_ring)
 			break;
 
-		del_timer_sync(&channel->tx_timer);
+		/* Deactivate the Tx timer */
+		timer_delete_sync(&channel->tx_timer);
+		channel->tx_timer_active = 0;
 	}
 }
 
@@ -676,8 +639,6 @@ void xgbe_get_all_hw_features(struct xgbe_prv_data *pdata)
 	unsigned int mac_hfr0, mac_hfr1, mac_hfr2;
 	struct xgbe_hw_features *hw_feat = &pdata->hw_feat;
 
-	DBGPR("-->xgbe_get_all_hw_features\n");
-
 	mac_hfr0 = XGMAC_IOREAD(pdata, MAC_HWF0R);
 	mac_hfr1 = XGMAC_IOREAD(pdata, MAC_HWF1R);
 	mac_hfr2 = XGMAC_IOREAD(pdata, MAC_HWF2R);
@@ -702,6 +663,7 @@ void xgbe_get_all_hw_features(struct xgbe_prv_data *pdata)
 					      ADDMACADRSEL);
 	hw_feat->ts_src      = XGMAC_GET_BITS(mac_hfr0, MAC_HWF0R, TSSTSSEL);
 	hw_feat->sa_vlan_ins = XGMAC_GET_BITS(mac_hfr0, MAC_HWF0R, SAVLANINS);
+	hw_feat->vxn         = XGMAC_GET_BITS(mac_hfr0, MAC_HWF0R, VXN);
 
 	/* Hardware feature register 1 */
 	hw_feat->rx_fifo_size  = XGMAC_GET_BITS(mac_hfr1, MAC_HWF1R,
@@ -729,6 +691,21 @@ void xgbe_get_all_hw_features(struct xgbe_prv_data *pdata)
 	hw_feat->pps_out_num  = XGMAC_GET_BITS(mac_hfr2, MAC_HWF2R, PPSOUTNUM);
 	hw_feat->aux_snap_num = XGMAC_GET_BITS(mac_hfr2, MAC_HWF2R, AUXSNAPNUM);
 
+	/* Sanity check and warn if hardware reports more than supported */
+	if (hw_feat->pps_out_num > XGBE_MAX_PPS_OUT) {
+		dev_warn(pdata->dev,
+			 "Hardware reports %u PPS outputs, limiting to %u\n",
+			 hw_feat->pps_out_num, XGBE_MAX_PPS_OUT);
+		hw_feat->pps_out_num = XGBE_MAX_PPS_OUT;
+	}
+
+	if (hw_feat->aux_snap_num > XGBE_MAX_AUX_SNAP) {
+		dev_warn(pdata->dev,
+			 "Hardware reports %u aux snapshot inputs, limiting to %u\n",
+			 hw_feat->aux_snap_num, XGBE_MAX_AUX_SNAP);
+		hw_feat->aux_snap_num = XGBE_MAX_AUX_SNAP;
+	}
+
 	/* Translate the Hash Table size into actual number */
 	switch (hw_feat->hash_table_size) {
 	case 0:
@@ -772,7 +749,119 @@ void xgbe_get_all_hw_features(struct xgbe_prv_data *pdata)
 	hw_feat->rx_fifo_size = 1 << (hw_feat->rx_fifo_size + 7);
 	hw_feat->tx_fifo_size = 1 << (hw_feat->tx_fifo_size + 7);
 
-	DBGPR("<--xgbe_get_all_hw_features\n");
+	if (netif_msg_probe(pdata)) {
+		dev_dbg(pdata->dev, "Hardware features:\n");
+
+		/* Hardware feature register 0 */
+		dev_dbg(pdata->dev, "  1GbE support              : %s\n",
+			hw_feat->gmii ? "yes" : "no");
+		dev_dbg(pdata->dev, "  VLAN hash filter          : %s\n",
+			hw_feat->vlhash ? "yes" : "no");
+		dev_dbg(pdata->dev, "  MDIO interface            : %s\n",
+			hw_feat->sma ? "yes" : "no");
+		dev_dbg(pdata->dev, "  Wake-up packet support    : %s\n",
+			hw_feat->rwk ? "yes" : "no");
+		dev_dbg(pdata->dev, "  Magic packet support      : %s\n",
+			hw_feat->mgk ? "yes" : "no");
+		dev_dbg(pdata->dev, "  Management counters       : %s\n",
+			hw_feat->mmc ? "yes" : "no");
+		dev_dbg(pdata->dev, "  ARP offload               : %s\n",
+			hw_feat->aoe ? "yes" : "no");
+		dev_dbg(pdata->dev, "  IEEE 1588-2008 Timestamp  : %s\n",
+			hw_feat->ts ? "yes" : "no");
+		dev_dbg(pdata->dev, "  Energy Efficient Ethernet : %s\n",
+			hw_feat->eee ? "yes" : "no");
+		dev_dbg(pdata->dev, "  TX checksum offload       : %s\n",
+			hw_feat->tx_coe ? "yes" : "no");
+		dev_dbg(pdata->dev, "  RX checksum offload       : %s\n",
+			hw_feat->rx_coe ? "yes" : "no");
+		dev_dbg(pdata->dev, "  Additional MAC addresses  : %u\n",
+			hw_feat->addn_mac);
+		dev_dbg(pdata->dev, "  Timestamp source          : %s\n",
+			(hw_feat->ts_src == 1) ? "internal" :
+			(hw_feat->ts_src == 2) ? "external" :
+			(hw_feat->ts_src == 3) ? "internal/external" : "n/a");
+		dev_dbg(pdata->dev, "  SA/VLAN insertion         : %s\n",
+			hw_feat->sa_vlan_ins ? "yes" : "no");
+		dev_dbg(pdata->dev, "  VXLAN/NVGRE support       : %s\n",
+			hw_feat->vxn ? "yes" : "no");
+
+		/* Hardware feature register 1 */
+		dev_dbg(pdata->dev, "  RX fifo size              : %u\n",
+			hw_feat->rx_fifo_size);
+		dev_dbg(pdata->dev, "  TX fifo size              : %u\n",
+			hw_feat->tx_fifo_size);
+		dev_dbg(pdata->dev, "  IEEE 1588 high word       : %s\n",
+			hw_feat->adv_ts_hi ? "yes" : "no");
+		dev_dbg(pdata->dev, "  DMA width                 : %u\n",
+			hw_feat->dma_width);
+		dev_dbg(pdata->dev, "  Data Center Bridging      : %s\n",
+			hw_feat->dcb ? "yes" : "no");
+		dev_dbg(pdata->dev, "  Split header              : %s\n",
+			hw_feat->sph ? "yes" : "no");
+		dev_dbg(pdata->dev, "  TCP Segmentation Offload  : %s\n",
+			hw_feat->tso ? "yes" : "no");
+		dev_dbg(pdata->dev, "  Debug memory interface    : %s\n",
+			hw_feat->dma_debug ? "yes" : "no");
+		dev_dbg(pdata->dev, "  Receive Side Scaling      : %s\n",
+			hw_feat->rss ? "yes" : "no");
+		dev_dbg(pdata->dev, "  Traffic Class count       : %u\n",
+			hw_feat->tc_cnt);
+		dev_dbg(pdata->dev, "  Hash table size           : %u\n",
+			hw_feat->hash_table_size);
+		dev_dbg(pdata->dev, "  L3/L4 Filters             : %u\n",
+			hw_feat->l3l4_filter_num);
+
+		/* Hardware feature register 2 */
+		dev_dbg(pdata->dev, "  RX queue count            : %u\n",
+			hw_feat->rx_q_cnt);
+		dev_dbg(pdata->dev, "  TX queue count            : %u\n",
+			hw_feat->tx_q_cnt);
+		dev_dbg(pdata->dev, "  RX DMA channel count      : %u\n",
+			hw_feat->rx_ch_cnt);
+		dev_dbg(pdata->dev, "  TX DMA channel count      : %u\n",
+			hw_feat->rx_ch_cnt);
+		dev_dbg(pdata->dev, "  PPS outputs               : %u\n",
+			hw_feat->pps_out_num);
+		dev_dbg(pdata->dev, "  Auxiliary snapshot inputs : %u\n",
+			hw_feat->aux_snap_num);
+	}
+}
+
+static int xgbe_vxlan_set_port(struct net_device *netdev, unsigned int table,
+			       unsigned int entry, struct udp_tunnel_info *ti)
+{
+	struct xgbe_prv_data *pdata = netdev_priv(netdev);
+
+	pdata->vxlan_port = be16_to_cpu(ti->port);
+	pdata->hw_if.enable_vxlan(pdata);
+
+	return 0;
+}
+
+static int xgbe_vxlan_unset_port(struct net_device *netdev, unsigned int table,
+				 unsigned int entry, struct udp_tunnel_info *ti)
+{
+	struct xgbe_prv_data *pdata = netdev_priv(netdev);
+
+	pdata->hw_if.disable_vxlan(pdata);
+	pdata->vxlan_port = 0;
+
+	return 0;
+}
+
+static const struct udp_tunnel_nic_info xgbe_udp_tunnels = {
+	.set_port	= xgbe_vxlan_set_port,
+	.unset_port	= xgbe_vxlan_unset_port,
+	.flags		= UDP_TUNNEL_NIC_INFO_OPEN_ONLY,
+	.tables		= {
+		{ .n_entries = 1, .tunnel_types = UDP_TUNNEL_TYPE_VXLAN, },
+	},
+};
+
+const struct udp_tunnel_nic_info *xgbe_get_udp_tunnel_info(void)
+{
+	return &xgbe_udp_tunnels;
 }
 
 static void xgbe_napi_enable(struct xgbe_prv_data *pdata, unsigned int add)
@@ -781,18 +870,18 @@ static void xgbe_napi_enable(struct xgbe_prv_data *pdata, unsigned int add)
 	unsigned int i;
 
 	if (pdata->per_channel_irq) {
-		channel = pdata->channel;
-		for (i = 0; i < pdata->channel_count; i++, channel++) {
+		for (i = 0; i < pdata->channel_count; i++) {
+			channel = pdata->channel[i];
 			if (add)
 				netif_napi_add(pdata->netdev, &channel->napi,
-					       xgbe_one_poll, NAPI_POLL_WEIGHT);
+					       xgbe_one_poll);
 
 			napi_enable(&channel->napi);
 		}
 	} else {
 		if (add)
 			netif_napi_add(pdata->netdev, &pdata->napi,
-				       xgbe_all_poll, NAPI_POLL_WEIGHT);
+				       xgbe_all_poll);
 
 		napi_enable(&pdata->napi);
 	}
@@ -804,8 +893,8 @@ static void xgbe_napi_disable(struct xgbe_prv_data *pdata, unsigned int del)
 	unsigned int i;
 
 	if (pdata->per_channel_irq) {
-		channel = pdata->channel;
-		for (i = 0; i < pdata->channel_count; i++, channel++) {
+		for (i = 0; i < pdata->channel_count; i++) {
+			channel = pdata->channel[i];
 			napi_disable(&channel->napi);
 
 			if (del)
@@ -826,8 +915,11 @@ static int xgbe_request_irqs(struct xgbe_prv_data *pdata)
 	unsigned int i;
 	int ret;
 
+	INIT_WORK(&pdata->dev_bh_work, xgbe_isr_bh_work);
+	INIT_WORK(&pdata->ecc_bh_work, xgbe_ecc_isr_bh_work);
+
 	ret = devm_request_irq(pdata->dev, pdata->dev_irq, xgbe_isr, 0,
-			       netdev->name, pdata);
+			       netdev_name(netdev), pdata);
 	if (ret) {
 		netdev_alert(netdev, "error requesting irq %d\n",
 			     pdata->dev_irq);
@@ -847,8 +939,8 @@ static int xgbe_request_irqs(struct xgbe_prv_data *pdata)
 	if (!pdata->per_channel_irq)
 		return 0;
 
-	channel = pdata->channel;
-	for (i = 0; i < pdata->channel_count; i++, channel++) {
+	for (i = 0; i < pdata->channel_count; i++) {
+		channel = pdata->channel[i];
 		snprintf(channel->dma_irq_name,
 			 sizeof(channel->dma_irq_name) - 1,
 			 "%s-TxRx-%u", netdev_name(netdev),
@@ -862,14 +954,21 @@ static int xgbe_request_irqs(struct xgbe_prv_data *pdata)
 				     channel->dma_irq);
 			goto err_dma_irq;
 		}
+
+		irq_set_affinity_hint(channel->dma_irq,
+				      &channel->affinity_mask);
 	}
 
 	return 0;
 
 err_dma_irq:
 	/* Using an unsigned int, 'i' will go to UINT_MAX and exit */
-	for (i--, channel--; i < pdata->channel_count; i--, channel--)
+	for (i--; i < pdata->channel_count; i--) {
+		channel = pdata->channel[i];
+
+		irq_set_affinity_hint(channel->dma_irq, NULL);
 		devm_free_irq(pdata->dev, channel->dma_irq, channel);
+	}
 
 	if (pdata->vdata->ecc_support && (pdata->dev_irq != pdata->ecc_irq))
 		devm_free_irq(pdata->dev, pdata->ecc_irq, pdata);
@@ -887,15 +986,21 @@ static void xgbe_free_irqs(struct xgbe_prv_data *pdata)
 
 	devm_free_irq(pdata->dev, pdata->dev_irq, pdata);
 
+	cancel_work_sync(&pdata->dev_bh_work);
+	cancel_work_sync(&pdata->ecc_bh_work);
+
 	if (pdata->vdata->ecc_support && (pdata->dev_irq != pdata->ecc_irq))
 		devm_free_irq(pdata->dev, pdata->ecc_irq, pdata);
 
 	if (!pdata->per_channel_irq)
 		return;
 
-	channel = pdata->channel;
-	for (i = 0; i < pdata->channel_count; i++, channel++)
+	for (i = 0; i < pdata->channel_count; i++) {
+		channel = pdata->channel[i];
+
+		irq_set_affinity_hint(channel->dma_irq, NULL);
 		devm_free_irq(pdata->dev, channel->dma_irq, channel);
+	}
 }
 
 void xgbe_init_tx_coalesce(struct xgbe_prv_data *pdata)
@@ -930,16 +1035,14 @@ void xgbe_init_rx_coalesce(struct xgbe_prv_data *pdata)
 static void xgbe_free_tx_data(struct xgbe_prv_data *pdata)
 {
 	struct xgbe_desc_if *desc_if = &pdata->desc_if;
-	struct xgbe_channel *channel;
 	struct xgbe_ring *ring;
 	struct xgbe_ring_data *rdata;
 	unsigned int i, j;
 
 	DBGPR("-->xgbe_free_tx_data\n");
 
-	channel = pdata->channel;
-	for (i = 0; i < pdata->channel_count; i++, channel++) {
-		ring = channel->tx_ring;
+	for (i = 0; i < pdata->channel_count; i++) {
+		ring = pdata->channel[i]->tx_ring;
 		if (!ring)
 			break;
 
@@ -955,16 +1058,14 @@ static void xgbe_free_tx_data(struct xgbe_prv_data *pdata)
 static void xgbe_free_rx_data(struct xgbe_prv_data *pdata)
 {
 	struct xgbe_desc_if *desc_if = &pdata->desc_if;
-	struct xgbe_channel *channel;
 	struct xgbe_ring *ring;
 	struct xgbe_ring_data *rdata;
 	unsigned int i, j;
 
 	DBGPR("-->xgbe_free_rx_data\n");
 
-	channel = pdata->channel;
-	for (i = 0; i < pdata->channel_count; i++, channel++) {
-		ring = channel->rx_ring;
+	for (i = 0; i < pdata->channel_count; i++) {
+		ring = pdata->channel[i]->rx_ring;
 		if (!ring)
 			break;
 
@@ -1062,14 +1163,83 @@ int xgbe_powerup(struct net_device *netdev, unsigned int caller)
 	return 0;
 }
 
+static void xgbe_free_memory(struct xgbe_prv_data *pdata)
+{
+	struct xgbe_desc_if *desc_if = &pdata->desc_if;
+
+	/* Free the ring descriptors and buffers */
+	desc_if->free_ring_resources(pdata);
+
+	/* Free the channel and ring structures */
+	xgbe_free_channels(pdata);
+}
+
+static int xgbe_alloc_memory(struct xgbe_prv_data *pdata)
+{
+	struct xgbe_desc_if *desc_if = &pdata->desc_if;
+	struct net_device *netdev = pdata->netdev;
+	int ret;
+
+	if (pdata->new_tx_ring_count) {
+		pdata->tx_ring_count = pdata->new_tx_ring_count;
+		pdata->tx_q_count = pdata->tx_ring_count;
+		pdata->new_tx_ring_count = 0;
+	}
+
+	if (pdata->new_rx_ring_count) {
+		pdata->rx_ring_count = pdata->new_rx_ring_count;
+		pdata->new_rx_ring_count = 0;
+	}
+
+	/* Calculate the Rx buffer size before allocating rings */
+	pdata->rx_buf_size = xgbe_calc_rx_buf_size(netdev, netdev->mtu);
+
+	/* Allocate the channel and ring structures */
+	ret = xgbe_alloc_channels(pdata);
+	if (ret)
+		return ret;
+
+	/* Allocate the ring descriptors and buffers */
+	ret = desc_if->alloc_ring_resources(pdata);
+	if (ret)
+		goto err_channels;
+
+	/* Initialize the service and Tx timers */
+	xgbe_init_timers(pdata);
+
+	return 0;
+
+err_channels:
+	xgbe_free_memory(pdata);
+
+	return ret;
+}
+
 static int xgbe_start(struct xgbe_prv_data *pdata)
 {
 	struct xgbe_hw_if *hw_if = &pdata->hw_if;
 	struct xgbe_phy_if *phy_if = &pdata->phy_if;
 	struct net_device *netdev = pdata->netdev;
+	unsigned int i;
 	int ret;
 
-	DBGPR("-->xgbe_start\n");
+	/* Set the number of queues */
+	ret = netif_set_real_num_tx_queues(netdev, pdata->tx_ring_count);
+	if (ret) {
+		netdev_err(netdev, "error setting real tx queue count\n");
+		return ret;
+	}
+
+	ret = netif_set_real_num_rx_queues(netdev, pdata->rx_ring_count);
+	if (ret) {
+		netdev_err(netdev, "error setting real rx queue count\n");
+		return ret;
+	}
+
+	/* Set RSS lookup table data for programming */
+	for (i = 0; i < XGBE_RSS_MAX_TABLE_SIZE; i++)
+		XGMAC_SET_BITS(pdata->rss_table[i], MAC_RSSDR, DMCH,
+			       i % pdata->rx_ring_count);
 
 	ret = hw_if->init(pdata);
 	if (ret)
@@ -1088,6 +1258,8 @@ static int xgbe_start(struct xgbe_prv_data *pdata)
 	hw_if->enable_tx(pdata);
 	hw_if->enable_rx(pdata);
 
+	udp_tunnel_nic_reset_ntf(netdev);
+
 	netif_tx_start_all_queues(netdev);
 
 	xgbe_start_timers(pdata);
@@ -1095,8 +1267,6 @@ static int xgbe_start(struct xgbe_prv_data *pdata)
 
 	clear_bit(XGBE_STOPPED, &pdata->dev_state);
 
-	DBGPR("<--xgbe_start\n");
-
 	return 0;
 
 err_irqs:
@@ -1125,10 +1295,13 @@ static void xgbe_stop(struct xgbe_prv_data *pdata)
 		return;
 
 	netif_tx_stop_all_queues(netdev);
+	netif_carrier_off(pdata->netdev);
 
 	xgbe_stop_timers(pdata);
 	flush_workqueue(pdata->dev_workqueue);
 
+	xgbe_vxlan_unset_port(netdev, 0, 0, NULL);
+
 	hw_if->disable_tx(pdata);
 	hw_if->disable_rx(pdata);
 
@@ -1140,8 +1313,8 @@ static void xgbe_stop(struct xgbe_prv_data *pdata)
 
 	hw_if->exit(pdata);
 
-	channel = pdata->channel;
-	for (i = 0; i < pdata->channel_count; i++, channel++) {
+	for (i = 0; i < pdata->channel_count; i++) {
+		channel = pdata->channel[i];
 		if (!channel->tx_ring)
 			continue;
 
@@ -1172,10 +1345,22 @@ static void xgbe_stopdev(struct work_struct *work)
 	netdev_alert(pdata->netdev, "device stopped\n");
 }
 
-static void xgbe_restart_dev(struct xgbe_prv_data *pdata)
+void xgbe_full_restart_dev(struct xgbe_prv_data *pdata)
 {
-	DBGPR("-->xgbe_restart_dev\n");
+	/* If not running, "restart" will happen on open */
+	if (!netif_running(pdata->netdev))
+		return;
+
+	xgbe_stop(pdata);
 
+	xgbe_free_memory(pdata);
+	xgbe_alloc_memory(pdata);
+
+	xgbe_start(pdata);
+}
+
+void xgbe_restart_dev(struct xgbe_prv_data *pdata)
+{
 	/* If not running, "restart" will happen on open */
 	if (!netif_running(pdata->netdev))
 		return;
@@ -1186,8 +1371,6 @@ static void xgbe_restart_dev(struct xgbe_prv_data *pdata)
 	xgbe_free_rx_data(pdata);
 
 	xgbe_start(pdata);
-
-	DBGPR("<--xgbe_restart_dev\n");
 }
 
 static void xgbe_restart(struct work_struct *work)
@@ -1203,197 +1386,6 @@ static void xgbe_restart(struct work_struct *work)
 	rtnl_unlock();
 }
 
-static void xgbe_tx_tstamp(struct work_struct *work)
-{
-	struct xgbe_prv_data *pdata = container_of(work,
-						   struct xgbe_prv_data,
-						   tx_tstamp_work);
-	struct skb_shared_hwtstamps hwtstamps;
-	u64 nsec;
-	unsigned long flags;
-
-	if (pdata->tx_tstamp) {
-		nsec = timecounter_cyc2time(&pdata->tstamp_tc,
-					    pdata->tx_tstamp);
-
-		memset(&hwtstamps, 0, sizeof(hwtstamps));
-		hwtstamps.hwtstamp = ns_to_ktime(nsec);
-		skb_tstamp_tx(pdata->tx_tstamp_skb, &hwtstamps);
-	}
-
-	dev_kfree_skb_any(pdata->tx_tstamp_skb);
-
-	spin_lock_irqsave(&pdata->tstamp_lock, flags);
-	pdata->tx_tstamp_skb = NULL;
-	spin_unlock_irqrestore(&pdata->tstamp_lock, flags);
-}
-
-static int xgbe_get_hwtstamp_settings(struct xgbe_prv_data *pdata,
-				      struct ifreq *ifreq)
-{
-	if (copy_to_user(ifreq->ifr_data, &pdata->tstamp_config,
-			 sizeof(pdata->tstamp_config)))
-		return -EFAULT;
-
-	return 0;
-}
-
-static int xgbe_set_hwtstamp_settings(struct xgbe_prv_data *pdata,
-				      struct ifreq *ifreq)
-{
-	struct hwtstamp_config config;
-	unsigned int mac_tscr;
-
-	if (copy_from_user(&config, ifreq->ifr_data, sizeof(config)))
-		return -EFAULT;
-
-	if (config.flags)
-		return -EINVAL;
-
-	mac_tscr = 0;
-
-	switch (config.tx_type) {
-	case HWTSTAMP_TX_OFF:
-		break;
-
-	case HWTSTAMP_TX_ON:
-		XGMAC_SET_BITS(mac_tscr, MAC_TSCR, TSENA, 1);
-		break;
-
-	default:
-		return -ERANGE;
-	}
-
-	switch (config.rx_filter) {
-	case HWTSTAMP_FILTER_NONE:
-		break;
-
-	case HWTSTAMP_FILTER_NTP_ALL:
-	case HWTSTAMP_FILTER_ALL:
-		XGMAC_SET_BITS(mac_tscr, MAC_TSCR, TSENALL, 1);
-		XGMAC_SET_BITS(mac_tscr, MAC_TSCR, TSENA, 1);
-		break;
-
-	/* PTP v2, UDP, any kind of event packet */
-	case HWTSTAMP_FILTER_PTP_V2_L4_EVENT:
-		XGMAC_SET_BITS(mac_tscr, MAC_TSCR, TSVER2ENA, 1);
-	/* PTP v1, UDP, any kind of event packet */
-	case HWTSTAMP_FILTER_PTP_V1_L4_EVENT:
-		XGMAC_SET_BITS(mac_tscr, MAC_TSCR, TSIPV4ENA, 1);
-		XGMAC_SET_BITS(mac_tscr, MAC_TSCR, TSIPV6ENA, 1);
-		XGMAC_SET_BITS(mac_tscr, MAC_TSCR, SNAPTYPSEL, 1);
-		XGMAC_SET_BITS(mac_tscr, MAC_TSCR, TSENA, 1);
-		break;
-
-	/* PTP v2, UDP, Sync packet */
-	case HWTSTAMP_FILTER_PTP_V2_L4_SYNC:
-		XGMAC_SET_BITS(mac_tscr, MAC_TSCR, TSVER2ENA, 1);
-	/* PTP v1, UDP, Sync packet */
-	case HWTSTAMP_FILTER_PTP_V1_L4_SYNC:
-		XGMAC_SET_BITS(mac_tscr, MAC_TSCR, TSIPV4ENA, 1);
-		XGMAC_SET_BITS(mac_tscr, MAC_TSCR, TSIPV6ENA, 1);
-		XGMAC_SET_BITS(mac_tscr, MAC_TSCR, TSEVNTENA, 1);
-		XGMAC_SET_BITS(mac_tscr, MAC_TSCR, TSENA, 1);
-		break;
-
-	/* PTP v2, UDP, Delay_req packet */
-	case HWTSTAMP_FILTER_PTP_V2_L4_DELAY_REQ:
-		XGMAC_SET_BITS(mac_tscr, MAC_TSCR, TSVER2ENA, 1);
-	/* PTP v1, UDP, Delay_req packet */
-	case HWTSTAMP_FILTER_PTP_V1_L4_DELAY_REQ:
-		XGMAC_SET_BITS(mac_tscr, MAC_TSCR, TSIPV4ENA, 1);
-		XGMAC_SET_BITS(mac_tscr, MAC_TSCR, TSIPV6ENA, 1);
-		XGMAC_SET_BITS(mac_tscr, MAC_TSCR, TSEVNTENA, 1);
-		XGMAC_SET_BITS(mac_tscr, MAC_TSCR, TSMSTRENA, 1);
-		XGMAC_SET_BITS(mac_tscr, MAC_TSCR, TSENA, 1);
-		break;
-
-	/* 802.AS1, Ethernet, any kind of event packet */
-	case HWTSTAMP_FILTER_PTP_V2_L2_EVENT:
-		XGMAC_SET_BITS(mac_tscr, MAC_TSCR, AV8021ASMEN, 1);
-		XGMAC_SET_BITS(mac_tscr, MAC_TSCR, SNAPTYPSEL, 1);
-		XGMAC_SET_BITS(mac_tscr, MAC_TSCR, TSENA, 1);
-		break;
-
-	/* 802.AS1, Ethernet, Sync packet */
-	case HWTSTAMP_FILTER_PTP_V2_L2_SYNC:
-		XGMAC_SET_BITS(mac_tscr, MAC_TSCR, AV8021ASMEN, 1);
-		XGMAC_SET_BITS(mac_tscr, MAC_TSCR, TSEVNTENA, 1);
-		XGMAC_SET_BITS(mac_tscr, MAC_TSCR, TSENA, 1);
-		break;
-
-	/* 802.AS1, Ethernet, Delay_req packet */
-	case HWTSTAMP_FILTER_PTP_V2_L2_DELAY_REQ:
-		XGMAC_SET_BITS(mac_tscr, MAC_TSCR, AV8021ASMEN, 1);
-		XGMAC_SET_BITS(mac_tscr, MAC_TSCR, TSMSTRENA, 1);
-		XGMAC_SET_BITS(mac_tscr, MAC_TSCR, TSEVNTENA, 1);
-		XGMAC_SET_BITS(mac_tscr, MAC_TSCR, TSENA, 1);
-		break;
-
-	/* PTP v2/802.AS1, any layer, any kind of event packet */
-	case HWTSTAMP_FILTER_PTP_V2_EVENT:
-		XGMAC_SET_BITS(mac_tscr, MAC_TSCR, TSVER2ENA, 1);
-		XGMAC_SET_BITS(mac_tscr, MAC_TSCR, TSIPENA, 1);
-		XGMAC_SET_BITS(mac_tscr, MAC_TSCR, TSIPV4ENA, 1);
-		XGMAC_SET_BITS(mac_tscr, MAC_TSCR, TSIPV6ENA, 1);
-		XGMAC_SET_BITS(mac_tscr, MAC_TSCR, SNAPTYPSEL, 1);
-		XGMAC_SET_BITS(mac_tscr, MAC_TSCR, TSENA, 1);
-		break;
-
-	/* PTP v2/802.AS1, any layer, Sync packet */
-	case HWTSTAMP_FILTER_PTP_V2_SYNC:
-		XGMAC_SET_BITS(mac_tscr, MAC_TSCR, TSVER2ENA, 1);
-		XGMAC_SET_BITS(mac_tscr, MAC_TSCR, TSIPENA, 1);
-		XGMAC_SET_BITS(mac_tscr, MAC_TSCR, TSIPV4ENA, 1);
-		XGMAC_SET_BITS(mac_tscr, MAC_TSCR, TSIPV6ENA, 1);
-		XGMAC_SET_BITS(mac_tscr, MAC_TSCR, TSEVNTENA, 1);
-		XGMAC_SET_BITS(mac_tscr, MAC_TSCR, TSENA, 1);
-		break;
-
-	/* PTP v2/802.AS1, any layer, Delay_req packet */
-	case HWTSTAMP_FILTER_PTP_V2_DELAY_REQ:
-		XGMAC_SET_BITS(mac_tscr, MAC_TSCR, TSVER2ENA, 1);
-		XGMAC_SET_BITS(mac_tscr, MAC_TSCR, TSIPENA, 1);
-		XGMAC_SET_BITS(mac_tscr, MAC_TSCR, TSIPV4ENA, 1);
-		XGMAC_SET_BITS(mac_tscr, MAC_TSCR, TSIPV6ENA, 1);
-		XGMAC_SET_BITS(mac_tscr, MAC_TSCR, TSMSTRENA, 1);
-		XGMAC_SET_BITS(mac_tscr, MAC_TSCR, TSEVNTENA, 1);
-		XGMAC_SET_BITS(mac_tscr, MAC_TSCR, TSENA, 1);
-		break;
-
-	default:
-		return -ERANGE;
-	}
-
-	pdata->hw_if.config_tstamp(pdata, mac_tscr);
-
-	memcpy(&pdata->tstamp_config, &config, sizeof(config));
-
-	return 0;
-}
-
-static void xgbe_prep_tx_tstamp(struct xgbe_prv_data *pdata,
-				struct sk_buff *skb,
-				struct xgbe_packet_data *packet)
-{
-	unsigned long flags;
-
-	if (XGMAC_GET_BITS(packet->attributes, TX_PACKET_ATTRIBUTES, PTP)) {
-		spin_lock_irqsave(&pdata->tstamp_lock, flags);
-		if (pdata->tx_tstamp_skb) {
-			/* Another timestamp in progress, ignore this one */
-			XGMAC_SET_BITS(packet->attributes,
-				       TX_PACKET_ATTRIBUTES, PTP, 0);
-		} else {
-			pdata->tx_tstamp_skb = skb_get(skb);
-			skb_shinfo(skb)->tx_flags |= SKBTX_IN_PROGRESS;
-		}
-		spin_unlock_irqrestore(&pdata->tstamp_lock, flags);
-	}
-
-	skb_tx_timestamp(skb);
-}
-
 static void xgbe_prep_vlan(struct sk_buff *skb, struct xgbe_packet_data *packet)
 {
 	if (skb_vlan_tag_present(skb))
@@ -1412,10 +1404,16 @@ static int xgbe_prep_tso(struct sk_buff *skb, struct xgbe_packet_data *packet)
 	if (ret)
 		return ret;
 
-	packet->header_len = skb_transport_offset(skb) + tcp_hdrlen(skb);
-	packet->tcp_header_len = tcp_hdrlen(skb);
+	if (XGMAC_GET_BITS(packet->attributes, TX_PACKET_ATTRIBUTES, VXLAN)) {
+		packet->header_len = skb_inner_tcp_all_headers(skb);
+		packet->tcp_header_len = inner_tcp_hdrlen(skb);
+	} else {
+		packet->header_len = skb_tcp_all_headers(skb);
+		packet->tcp_header_len = tcp_hdrlen(skb);
+	}
 	packet->tcp_payload_len = skb->len - packet->header_len;
 	packet->mss = skb_shinfo(skb)->gso_size;
+
 	DBGPR("  packet->header_len=%u\n", packet->header_len);
 	DBGPR("  packet->tcp_header_len=%u, packet->tcp_payload_len=%u\n",
 	      packet->tcp_header_len, packet->tcp_payload_len);
@@ -1430,6 +1428,38 @@ static int xgbe_prep_tso(struct sk_buff *skb, struct xgbe_packet_data *packet)
 	return 0;
 }
 
+static bool xgbe_is_vxlan(struct sk_buff *skb)
+{
+	if (!skb->encapsulation)
+		return false;
+
+	if (skb->ip_summed != CHECKSUM_PARTIAL)
+		return false;
+
+	switch (skb->protocol) {
+	case htons(ETH_P_IP):
+		if (ip_hdr(skb)->protocol != IPPROTO_UDP)
+			return false;
+		break;
+
+	case htons(ETH_P_IPV6):
+		if (ipv6_hdr(skb)->nexthdr != IPPROTO_UDP)
+			return false;
+		break;
+
+	default:
+		return false;
+	}
+
+	if (skb->inner_protocol_type != ENCAP_TYPE_ETHER ||
+	    skb->inner_protocol != htons(ETH_P_TEB) ||
+	    (skb_inner_mac_header(skb) - skb_transport_header(skb) !=
+	     sizeof(struct udphdr) + sizeof(struct vxlanhdr)))
+		return false;
+
+	return true;
+}
+
 static int xgbe_is_tso(struct sk_buff *skb)
 {
 	if (skb->ip_summed != CHECKSUM_PARTIAL)
@@ -1447,7 +1477,7 @@ static void xgbe_packet_info(struct xgbe_prv_data *pdata,
 			     struct xgbe_ring *ring, struct sk_buff *skb,
 			     struct xgbe_packet_data *packet)
 {
-	struct skb_frag_struct *frag;
+	skb_frag_t *frag;
 	unsigned int context_desc;
 	unsigned int len;
 	unsigned int i;
@@ -1478,6 +1508,10 @@ static void xgbe_packet_info(struct xgbe_prv_data *pdata,
 		XGMAC_SET_BITS(packet->attributes, TX_PACKET_ATTRIBUTES,
 			       CSUM_ENABLE, 1);
 
+	if (xgbe_is_vxlan(skb))
+		XGMAC_SET_BITS(packet->attributes, TX_PACKET_ATTRIBUTES,
+			       VXLAN, 1);
+
 	if (skb_vlan_tag_present(skb)) {
 		/* VLAN requires an extra descriptor if tag is different */
 		if (skb_vlan_tag_get(skb) != ring->tx.cur_vlan_ctag)
@@ -1513,21 +1547,44 @@ static void xgbe_packet_info(struct xgbe_prv_data *pdata,
 static int xgbe_open(struct net_device *netdev)
 {
 	struct xgbe_prv_data *pdata = netdev_priv(netdev);
-	struct xgbe_desc_if *desc_if = &pdata->desc_if;
 	int ret;
 
-	DBGPR("-->xgbe_open\n");
+	/* Create the various names based on netdev name */
+	snprintf(pdata->an_name, sizeof(pdata->an_name) - 1, "%s-pcs",
+		 netdev_name(netdev));
+
+	snprintf(pdata->ecc_name, sizeof(pdata->ecc_name) - 1, "%s-ecc",
+		 netdev_name(netdev));
+
+	snprintf(pdata->i2c_name, sizeof(pdata->i2c_name) - 1, "%s-i2c",
+		 netdev_name(netdev));
+
+	/* Create workqueues */
+	pdata->dev_workqueue =
+		create_singlethread_workqueue(netdev_name(netdev));
+	if (!pdata->dev_workqueue) {
+		netdev_err(netdev, "device workqueue creation failed\n");
+		return -ENOMEM;
+	}
+
+	pdata->an_workqueue =
+		create_singlethread_workqueue(pdata->an_name);
+	if (!pdata->an_workqueue) {
+		netdev_err(netdev, "phy workqueue creation failed\n");
+		ret = -ENOMEM;
+		goto err_dev_wq;
+	}
 
 	/* Reset the phy settings */
 	ret = xgbe_phy_reset(pdata);
 	if (ret)
-		return ret;
+		goto err_an_wq;
 
 	/* Enable the clocks */
 	ret = clk_prepare_enable(pdata->sysclk);
 	if (ret) {
 		netdev_alert(netdev, "dma clk_prepare_enable failed\n");
-		return ret;
+		goto err_an_wq;
 	}
 
 	ret = clk_prepare_enable(pdata->ptpclk);
@@ -1536,43 +1593,28 @@ static int xgbe_open(struct net_device *netdev)
 		goto err_sysclk;
 	}
 
-	/* Calculate the Rx buffer size before allocating rings */
-	ret = xgbe_calc_rx_buf_size(netdev, netdev->mtu);
-	if (ret < 0)
-		goto err_ptpclk;
-	pdata->rx_buf_size = ret;
-
-	/* Allocate the channel and ring structures */
-	ret = xgbe_alloc_channels(pdata);
-	if (ret)
-		goto err_ptpclk;
-
-	/* Allocate the ring descriptors and buffers */
-	ret = desc_if->alloc_ring_resources(pdata);
-	if (ret)
-		goto err_channels;
-
 	INIT_WORK(&pdata->service_work, xgbe_service);
 	INIT_WORK(&pdata->restart_work, xgbe_restart);
 	INIT_WORK(&pdata->stopdev_work, xgbe_stopdev);
 	INIT_WORK(&pdata->tx_tstamp_work, xgbe_tx_tstamp);
-	xgbe_init_timers(pdata);
+
+	/* Initialize PTP timestamping and clock. */
+	xgbe_init_ptp(pdata);
+
+	ret = xgbe_alloc_memory(pdata);
+	if (ret)
+		goto err_ptpclk;
 
 	ret = xgbe_start(pdata);
 	if (ret)
-		goto err_rings;
+		goto err_mem;
 
 	clear_bit(XGBE_DOWN, &pdata->dev_state);
 
-	DBGPR("<--xgbe_open\n");
-
 	return 0;
 
-err_rings:
-	desc_if->free_ring_resources(pdata);
-
-err_channels:
-	xgbe_free_channels(pdata);
+err_mem:
+	xgbe_free_memory(pdata);
 
 err_ptpclk:
 	clk_disable_unprepare(pdata->ptpclk);
@@ -1580,37 +1622,38 @@ err_ptpclk:
 err_sysclk:
 	clk_disable_unprepare(pdata->sysclk);
 
+err_an_wq:
+	destroy_workqueue(pdata->an_workqueue);
+
+err_dev_wq:
+	destroy_workqueue(pdata->dev_workqueue);
+
 	return ret;
 }
 
 static int xgbe_close(struct net_device *netdev)
 {
 	struct xgbe_prv_data *pdata = netdev_priv(netdev);
-	struct xgbe_desc_if *desc_if = &pdata->desc_if;
-
-	DBGPR("-->xgbe_close\n");
 
 	/* Stop the device */
 	xgbe_stop(pdata);
 
-	/* Free the ring descriptors and buffers */
-	desc_if->free_ring_resources(pdata);
-
-	/* Free the channel and ring structures */
-	xgbe_free_channels(pdata);
+	xgbe_free_memory(pdata);
 
 	/* Disable the clocks */
 	clk_disable_unprepare(pdata->ptpclk);
 	clk_disable_unprepare(pdata->sysclk);
 
-	set_bit(XGBE_DOWN, &pdata->dev_state);
+	destroy_workqueue(pdata->an_workqueue);
+
+	destroy_workqueue(pdata->dev_workqueue);
 
-	DBGPR("<--xgbe_close\n");
+	set_bit(XGBE_DOWN, &pdata->dev_state);
 
 	return 0;
 }
 
-static int xgbe_xmit(struct sk_buff *skb, struct net_device *netdev)
+static netdev_tx_t xgbe_xmit(struct sk_buff *skb, struct net_device *netdev)
 {
 	struct xgbe_prv_data *pdata = netdev_priv(netdev);
 	struct xgbe_hw_if *hw_if = &pdata->hw_if;
@@ -1619,11 +1662,11 @@ static int xgbe_xmit(struct sk_buff *skb, struct net_device *netdev)
 	struct xgbe_ring *ring;
 	struct xgbe_packet_data *packet;
 	struct netdev_queue *txq;
-	int ret;
+	netdev_tx_t ret;
 
 	DBGPR("-->xgbe_xmit: skb->len = %d\n", skb->len);
 
-	channel = pdata->channel + skb->queue_mapping;
+	channel = pdata->channel[skb->queue_mapping];
 	txq = netdev_get_tx_queue(netdev, channel->queue_index);
 	ring = channel->tx_ring;
 	packet = &ring->packet_data;
@@ -1703,7 +1746,7 @@ static int xgbe_set_mac_address(struct net_device *netdev, void *addr)
 	if (!is_valid_ether_addr(saddr->sa_data))
 		return -EADDRNOTAVAIL;
 
-	memcpy(netdev->dev_addr, saddr->sa_data, netdev->addr_len);
+	eth_hw_addr_set(netdev, saddr->sa_data);
 
 	hw_if->set_mac_address(pdata, netdev->dev_addr);
 
@@ -1745,7 +1788,7 @@ static int xgbe_change_mtu(struct net_device *netdev, int mtu)
 		return ret;
 
 	pdata->rx_buf_size = ret;
-	netdev->mtu = mtu;
+	WRITE_ONCE(netdev->mtu, mtu);
 
 	xgbe_restart_dev(pdata);
 
@@ -1754,7 +1797,7 @@ static int xgbe_change_mtu(struct net_device *netdev, int mtu)
 	return 0;
 }
 
-static void xgbe_tx_timeout(struct net_device *netdev)
+static void xgbe_tx_timeout(struct net_device *netdev, unsigned int txqueue)
 {
 	struct xgbe_prv_data *pdata = netdev_priv(netdev);
 
@@ -1833,9 +1876,10 @@ static void xgbe_poll_controller(struct net_device *netdev)
 	DBGPR("-->xgbe_poll_controller\n");
 
 	if (pdata->per_channel_irq) {
-		channel = pdata->channel;
-		for (i = 0; i < pdata->channel_count; i++, channel++)
+		for (i = 0; i < pdata->channel_count; i++) {
+			channel = pdata->channel[i];
 			xgbe_dma_isr(channel->dma_irq, channel);
+		}
 	} else {
 		disable_irq(pdata->dev_irq);
 		xgbe_isr(pdata->dev_irq, pdata);
@@ -1846,17 +1890,18 @@ static void xgbe_poll_controller(struct net_device *netdev)
 }
 #endif /* End CONFIG_NET_POLL_CONTROLLER */
 
-static int xgbe_setup_tc(struct net_device *netdev, u32 handle, __be16 proto,
-			 struct tc_to_netdev *tc_to_netdev)
+static int xgbe_setup_tc(struct net_device *netdev, enum tc_setup_type type,
+			 void *type_data)
 {
 	struct xgbe_prv_data *pdata = netdev_priv(netdev);
+	struct tc_mqprio_qopt *mqprio = type_data;
 	u8 tc;
 
-	if (tc_to_netdev->type != TC_SETUP_MQPRIO)
-		return -EINVAL;
+	if (type != TC_SETUP_QDISC_MQPRIO)
+		return -EOPNOTSUPP;
 
-	tc_to_netdev->mqprio->hw = TC_MQPRIO_HW_OFFLOAD_TCS;
-	tc = tc_to_netdev->mqprio->num_tc;
+	mqprio->hw = TC_MQPRIO_HW_OFFLOAD_TCS;
+	tc = mqprio->num_tc;
 
 	if (tc > pdata->hw_feat.tc_cnt)
 		return -EINVAL;
@@ -1867,6 +1912,49 @@ static int xgbe_setup_tc(struct net_device *netdev, u32 handle, __be16 proto,
 	return 0;
 }
 
+static netdev_features_t xgbe_fix_features(struct net_device *netdev,
+					   netdev_features_t features)
+{
+	struct xgbe_prv_data *pdata = netdev_priv(netdev);
+	netdev_features_t vxlan_base;
+
+	vxlan_base = NETIF_F_GSO_UDP_TUNNEL | NETIF_F_RX_UDP_TUNNEL_PORT;
+
+	if (!pdata->hw_feat.vxn)
+		return features;
+
+	/* VXLAN CSUM requires VXLAN base */
+	if ((features & NETIF_F_GSO_UDP_TUNNEL_CSUM) &&
+	    !(features & NETIF_F_GSO_UDP_TUNNEL)) {
+		netdev_notice(netdev,
+			      "forcing tx udp tunnel support\n");
+		features |= NETIF_F_GSO_UDP_TUNNEL;
+	}
+
+	/* Can't do one without doing the other */
+	if ((features & vxlan_base) != vxlan_base) {
+		netdev_notice(netdev,
+			      "forcing both tx and rx udp tunnel support\n");
+		features |= vxlan_base;
+	}
+
+	if (features & (NETIF_F_IP_CSUM | NETIF_F_IPV6_CSUM)) {
+		if (!(features & NETIF_F_GSO_UDP_TUNNEL_CSUM)) {
+			netdev_notice(netdev,
+				      "forcing tx udp tunnel checksumming on\n");
+			features |= NETIF_F_GSO_UDP_TUNNEL_CSUM;
+		}
+	} else {
+		if (features & NETIF_F_GSO_UDP_TUNNEL_CSUM) {
+			netdev_notice(netdev,
+				      "forcing tx udp tunnel checksumming off\n");
+			features &= ~NETIF_F_GSO_UDP_TUNNEL_CSUM;
+		}
+	}
+
+	return features;
+}
+
 static int xgbe_set_features(struct net_device *netdev,
 			     netdev_features_t features)
 {
@@ -1887,10 +1975,17 @@ static int xgbe_set_features(struct net_device *netdev,
 	if (ret)
 		return ret;
 
-	if ((features & NETIF_F_RXCSUM) && !rxcsum)
+	if ((features & NETIF_F_RXCSUM) && !rxcsum) {
+		hw_if->enable_sph(pdata);
+		hw_if->enable_vxlan(pdata);
 		hw_if->enable_rx_csum(pdata);
-	else if (!(features & NETIF_F_RXCSUM) && rxcsum)
+		schedule_work(&pdata->restart_work);
+	} else if (!(features & NETIF_F_RXCSUM) && rxcsum) {
+		hw_if->disable_sph(pdata);
+		hw_if->disable_vxlan(pdata);
 		hw_if->disable_rx_csum(pdata);
+		schedule_work(&pdata->restart_work);
+	}
 
 	if ((features & NETIF_F_HW_VLAN_CTAG_RX) && !rxvlan)
 		hw_if->enable_rx_vlan_stripping(pdata);
@@ -1909,6 +2004,16 @@ static int xgbe_set_features(struct net_device *netdev,
 	return 0;
 }
 
+static netdev_features_t xgbe_features_check(struct sk_buff *skb,
+					     struct net_device *netdev,
+					     netdev_features_t features)
+{
+	features = vlan_features_check(skb, features);
+	features = vxlan_features_check(skb, features);
+
+	return features;
+}
+
 static const struct net_device_ops xgbe_netdev_ops = {
 	.ndo_open		= xgbe_open,
 	.ndo_stop		= xgbe_close,
@@ -1916,7 +2021,7 @@ static const struct net_device_ops xgbe_netdev_ops = {
 	.ndo_set_rx_mode	= xgbe_set_rx_mode,
 	.ndo_set_mac_address	= xgbe_set_mac_address,
 	.ndo_validate_addr	= eth_validate_addr,
-	.ndo_do_ioctl		= xgbe_ioctl,
+	.ndo_eth_ioctl		= xgbe_ioctl,
 	.ndo_change_mtu		= xgbe_change_mtu,
 	.ndo_tx_timeout		= xgbe_tx_timeout,
 	.ndo_get_stats64	= xgbe_get_stats64,
@@ -1926,7 +2031,9 @@ static const struct net_device_ops xgbe_netdev_ops = {
 	.ndo_poll_controller	= xgbe_poll_controller,
 #endif
 	.ndo_setup_tc		= xgbe_setup_tc,
+	.ndo_fix_features	= xgbe_fix_features,
 	.ndo_set_features	= xgbe_set_features,
+	.ndo_features_check	= xgbe_features_check,
 };
 
 const struct net_device_ops *xgbe_get_netdev_ops(void)
@@ -2185,6 +2292,14 @@ read_again:
 			buf2_len = xgbe_rx_buf2_len(rdata, packet, len);
 			len += buf2_len;
 
+			if (buf2_len > rdata->rx.buf.dma_len) {
+				/* Hardware inconsistency within the descriptors
+				 * that has resulted in a length underflow.
+				 */
+				error = 1;
+				goto skip_data;
+			}
+
 			if (!skb) {
 				skb = xgbe_create_skb(pdata, napi, rdata,
 						      buf1_len);
@@ -2214,8 +2329,10 @@ skip_data:
 		if (!last || context_next)
 			goto read_again;
 
-		if (!skb)
+		if (!skb || error) {
+			dev_kfree_skb(skb);
 			goto next_packet;
+		}
 
 		/* Be sure we don't exceed the configured MTU */
 		max_len = netdev->mtu + ETH_HLEN;
@@ -2239,18 +2356,23 @@ skip_data:
 			skb->ip_summed = CHECKSUM_UNNECESSARY;
 
 		if (XGMAC_GET_BITS(packet->attributes,
+				   RX_PACKET_ATTRIBUTES, TNP)) {
+			skb->encapsulation = 1;
+
+			if (XGMAC_GET_BITS(packet->attributes,
+					   RX_PACKET_ATTRIBUTES, TNPCSUM_DONE))
+				skb->csum_level = 1;
+		}
+
+		if (XGMAC_GET_BITS(packet->attributes,
 				   RX_PACKET_ATTRIBUTES, VLAN_CTAG))
 			__vlan_hwaccel_put_tag(skb, htons(ETH_P_8021Q),
 					       packet->vlan_ctag);
 
 		if (XGMAC_GET_BITS(packet->attributes,
 				   RX_PACKET_ATTRIBUTES, RX_TSTAMP)) {
-			u64 nsec;
-
-			nsec = timecounter_cyc2time(&pdata->tstamp_tc,
-						    packet->rx_tstamp);
 			hwtstamps = skb_hwtstamps(skb);
-			hwtstamps->hwtstamp = ns_to_ktime(nsec);
+			hwtstamps->hwtstamp = ns_to_ktime(packet->rx_tstamp);
 		}
 
 		if (XGMAC_GET_BITS(packet->attributes,
@@ -2327,8 +2449,9 @@ static int xgbe_all_poll(struct napi_struct *napi, int budget)
 	do {
 		last_processed = processed;
 
-		channel = pdata->channel;
-		for (i = 0; i < pdata->channel_count; i++, channel++) {
+		for (i = 0; i < pdata->channel_count; i++) {
+			channel = pdata->channel[i];
+
 			/* Cleanup Tx ring first */
 			xgbe_tx_poll(channel);
 
@@ -2387,9 +2510,8 @@ void xgbe_dump_rx_desc(struct xgbe_prv_data *pdata, struct xgbe_ring *ring,
 void xgbe_print_pkt(struct net_device *netdev, struct sk_buff *skb, bool tx_rx)
 {
 	struct ethhdr *eth = (struct ethhdr *)skb->data;
-	unsigned char *buf = skb->data;
 	unsigned char buffer[128];
-	unsigned int i, j;
+	unsigned int i;
 
 	netdev_dbg(netdev, "\n************** SKB dump ****************\n");
 
@@ -2398,24 +2520,15 @@ void xgbe_print_pkt(struct net_device *netdev, struct sk_buff *skb, bool tx_rx)
 
 	netdev_dbg(netdev, "Dst MAC addr: %pM\n", eth->h_dest);
 	netdev_dbg(netdev, "Src MAC addr: %pM\n", eth->h_source);
-	netdev_dbg(netdev, "Protocol: %#06hx\n", ntohs(eth->h_proto));
-
-	for (i = 0, j = 0; i < skb->len;) {
-		j += snprintf(buffer + j, sizeof(buffer) - j, "%02hhx",
-			      buf[i++]);
-
-		if ((i % 32) == 0) {
-			netdev_dbg(netdev, "  %#06x: %s\n", i - 32, buffer);
-			j = 0;
-		} else if ((i % 16) == 0) {
-			buffer[j++] = ' ';
-			buffer[j++] = ' ';
-		} else if ((i % 4) == 0) {
-			buffer[j++] = ' ';
-		}
+	netdev_dbg(netdev, "Protocol: %#06x\n", ntohs(eth->h_proto));
+
+	for (i = 0; i < skb->len; i += 32) {
+		unsigned int len = min(skb->len - i, 32U);
+
+		hex_dump_to_buffer(&skb->data[i], len, 32, 1,
+				   buffer, sizeof(buffer), false);
+		netdev_dbg(netdev, "  %#06x: %s\n", i, buffer);
 	}
-	if (i % 32)
-		netdev_dbg(netdev, "  %#06x: %s\n", i - (i % 32), buffer);
 
 	netdev_dbg(netdev, "\n************** SKB dump ****************\n");
 }
diff --git a/drivers/net/ethernet/amd/xgbe/xgbe-ethtool.c b/drivers/net/ethernet/amd/xgbe/xgbe-ethtool.c
index 920566a3a599..b6e1b67a2d0e 100644
--- a/drivers/net/ethernet/amd/xgbe/xgbe-ethtool.c
+++ b/drivers/net/ethernet/amd/xgbe/xgbe-ethtool.c
@@ -1,117 +1,8 @@
+// SPDX-License-Identifier: (GPL-2.0-or-later OR BSD-3-Clause)
 /*
- * AMD 10Gb Ethernet driver
- *
- * This file is available to you under your choice of the following two
- * licenses:
- *
- * License 1: GPLv2
- *
- * Copyright (c) 2014-2016 Advanced Micro Devices, Inc.
- *
- * This file is free software; you may copy, redistribute and/or modify
- * it under the terms of the GNU General Public License as published by
- * the Free Software Foundation, either version 2 of the License, or (at
- * your option) any later version.
- *
- * This file is distributed in the hope that it will be useful, but
- * WITHOUT ANY WARRANTY; without even the implied warranty of
- * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
- * General Public License for more details.
- *
- * You should have received a copy of the GNU General Public License
- * along with this program.  If not, see <http://www.gnu.org/licenses/>.
- *
- * This file incorporates work covered by the following copyright and
- * permission notice:
- *     The Synopsys DWC ETHER XGMAC Software Driver and documentation
- *     (hereinafter "Software") is an unsupported proprietary work of Synopsys,
- *     Inc. unless otherwise expressly agreed to in writing between Synopsys
- *     and you.
- *
- *     The Software IS NOT an item of Licensed Software or Licensed Product
- *     under any End User Software License Agreement or Agreement for Licensed
- *     Product with Synopsys or any supplement thereto.  Permission is hereby
- *     granted, free of charge, to any person obtaining a copy of this software
- *     annotated with this license and the Software, to deal in the Software
- *     without restriction, including without limitation the rights to use,
- *     copy, modify, merge, publish, distribute, sublicense, and/or sell copies
- *     of the Software, and to permit persons to whom the Software is furnished
- *     to do so, subject to the following conditions:
- *
- *     The above copyright notice and this permission notice shall be included
- *     in all copies or substantial portions of the Software.
- *
- *     THIS SOFTWARE IS BEING DISTRIBUTED BY SYNOPSYS SOLELY ON AN "AS IS"
- *     BASIS AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED
- *     TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A
- *     PARTICULAR PURPOSE ARE HEREBY DISCLAIMED. IN NO EVENT SHALL SYNOPSYS
- *     BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR
- *     CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF
- *     SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS
- *     INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN
- *     CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE)
- *     ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF
- *     THE POSSIBILITY OF SUCH DAMAGE.
- *
- *
- * License 2: Modified BSD
- *
- * Copyright (c) 2014-2016 Advanced Micro Devices, Inc.
- * All rights reserved.
- *
- * Redistribution and use in source and binary forms, with or without
- * modification, are permitted provided that the following conditions are met:
- *     * Redistributions of source code must retain the above copyright
- *       notice, this list of conditions and the following disclaimer.
- *     * Redistributions in binary form must reproduce the above copyright
- *       notice, this list of conditions and the following disclaimer in the
- *       documentation and/or other materials provided with the distribution.
- *     * Neither the name of Advanced Micro Devices, Inc. nor the
- *       names of its contributors may be used to endorse or promote products
- *       derived from this software without specific prior written permission.
- *
- * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS"
- * AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
- * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
- * ARE DISCLAIMED. IN NO EVENT SHALL <COPYRIGHT HOLDER> BE LIABLE FOR ANY
- * DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES
- * (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES;
- * LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND
- * ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
- * (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF
- * THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
- *
- * This file incorporates work covered by the following copyright and
- * permission notice:
- *     The Synopsys DWC ETHER XGMAC Software Driver and documentation
- *     (hereinafter "Software") is an unsupported proprietary work of Synopsys,
- *     Inc. unless otherwise expressly agreed to in writing between Synopsys
- *     and you.
- *
- *     The Software IS NOT an item of Licensed Software or Licensed Product
- *     under any End User Software License Agreement or Agreement for Licensed
- *     Product with Synopsys or any supplement thereto.  Permission is hereby
- *     granted, free of charge, to any person obtaining a copy of this software
- *     annotated with this license and the Software, to deal in the Software
- *     without restriction, including without limitation the rights to use,
- *     copy, modify, merge, publish, distribute, sublicense, and/or sell copies
- *     of the Software, and to permit persons to whom the Software is furnished
- *     to do so, subject to the following conditions:
- *
- *     The above copyright notice and this permission notice shall be included
- *     in all copies or substantial portions of the Software.
- *
- *     THIS SOFTWARE IS BEING DISTRIBUTED BY SYNOPSYS SOLELY ON AN "AS IS"
- *     BASIS AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED
- *     TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A
- *     PARTICULAR PURPOSE ARE HEREBY DISCLAIMED. IN NO EVENT SHALL SYNOPSYS
- *     BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR
- *     CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF
- *     SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS
- *     INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN
- *     CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE)
- *     ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF
- *     THE POSSIBILITY OF SUCH DAMAGE.
+ * Copyright (c) 2014-2025, Advanced Micro Devices, Inc.
+ * Copyright (c) 2014, Synopsys, Inc.
+ * All rights reserved
  */
 
 #include <linux/spinlock.h>
@@ -129,13 +20,13 @@ struct xgbe_stats {
 
 #define XGMAC_MMC_STAT(_string, _var)				\
 	{ _string,						\
-	  FIELD_SIZEOF(struct xgbe_mmc_stats, _var),		\
+	  sizeof_field(struct xgbe_mmc_stats, _var),		\
 	  offsetof(struct xgbe_prv_data, mmc_stats._var),	\
 	}
 
 #define XGMAC_EXT_STAT(_string, _var)				\
 	{ _string,						\
-	  FIELD_SIZEOF(struct xgbe_ext_stats, _var),		\
+	  sizeof_field(struct xgbe_ext_stats, _var),		\
 	  offsetof(struct xgbe_prv_data, ext_stats._var),	\
 	}
 
@@ -146,6 +37,7 @@ static const struct xgbe_stats xgbe_gstring_stats[] = {
 	XGMAC_MMC_STAT("tx_broadcast_packets", txbroadcastframes_gb),
 	XGMAC_MMC_STAT("tx_multicast_packets", txmulticastframes_gb),
 	XGMAC_MMC_STAT("tx_vlan_packets", txvlanframes_g),
+	XGMAC_EXT_STAT("tx_vxlan_packets", tx_vxlan_packets),
 	XGMAC_EXT_STAT("tx_tso_packets", tx_tso_packets),
 	XGMAC_MMC_STAT("tx_64_byte_packets", tx64octets_gb),
 	XGMAC_MMC_STAT("tx_65_to_127_byte_packets", tx65to127octets_gb),
@@ -162,6 +54,7 @@ static const struct xgbe_stats xgbe_gstring_stats[] = {
 	XGMAC_MMC_STAT("rx_broadcast_packets", rxbroadcastframes_g),
 	XGMAC_MMC_STAT("rx_multicast_packets", rxmulticastframes_g),
 	XGMAC_MMC_STAT("rx_vlan_packets", rxvlanframes_gb),
+	XGMAC_EXT_STAT("rx_vxlan_packets", rx_vxlan_packets),
 	XGMAC_MMC_STAT("rx_64_byte_packets", rx64octets_gb),
 	XGMAC_MMC_STAT("rx_65_to_127_byte_packets", rx65to127octets_gb),
 	XGMAC_MMC_STAT("rx_128_to_255_byte_packets", rx128to255octets_gb),
@@ -177,6 +70,8 @@ static const struct xgbe_stats xgbe_gstring_stats[] = {
 	XGMAC_MMC_STAT("rx_out_of_range_errors", rxoutofrangetype),
 	XGMAC_MMC_STAT("rx_fifo_overflow_errors", rxfifooverflow),
 	XGMAC_MMC_STAT("rx_watchdog_errors", rxwatchdogerror),
+	XGMAC_EXT_STAT("rx_csum_errors", rx_csum_errors),
+	XGMAC_EXT_STAT("rx_vxlan_csum_errors", rx_vxlan_csum_errors),
 	XGMAC_MMC_STAT("rx_pause_frames", rxpauseframes),
 	XGMAC_EXT_STAT("rx_split_header_packets", rx_split_header_packets),
 	XGMAC_EXT_STAT("rx_buffer_unavailable", rx_buffer_unavailable),
@@ -186,15 +81,24 @@ static const struct xgbe_stats xgbe_gstring_stats[] = {
 
 static void xgbe_get_strings(struct net_device *netdev, u32 stringset, u8 *data)
 {
+	struct xgbe_prv_data *pdata = netdev_priv(netdev);
 	int i;
 
 	switch (stringset) {
 	case ETH_SS_STATS:
-		for (i = 0; i < XGBE_STATS_COUNT; i++) {
-			memcpy(data, xgbe_gstring_stats[i].stat_string,
-			       ETH_GSTRING_LEN);
-			data += ETH_GSTRING_LEN;
+		for (i = 0; i < XGBE_STATS_COUNT; i++)
+			ethtool_puts(&data, xgbe_gstring_stats[i].stat_string);
+
+		for (i = 0; i < pdata->tx_ring_count; i++) {
+			ethtool_sprintf(&data, "txq_%u_packets", i);
+			ethtool_sprintf(&data, "txq_%u_bytes", i);
+		}
+
+		for (i = 0; i < pdata->rx_ring_count; i++) {
+			ethtool_sprintf(&data, "rxq_%u_packets", i);
+			ethtool_sprintf(&data, "rxq_%u_bytes", i);
 		}
+
 		break;
 	}
 }
@@ -211,15 +115,26 @@ static void xgbe_get_ethtool_stats(struct net_device *netdev,
 		stat = (u8 *)pdata + xgbe_gstring_stats[i].stat_offset;
 		*data++ = *(u64 *)stat;
 	}
+	for (i = 0; i < pdata->tx_ring_count; i++) {
+		*data++ = pdata->ext_stats.txq_packets[i];
+		*data++ = pdata->ext_stats.txq_bytes[i];
+	}
+	for (i = 0; i < pdata->rx_ring_count; i++) {
+		*data++ = pdata->ext_stats.rxq_packets[i];
+		*data++ = pdata->ext_stats.rxq_bytes[i];
+	}
 }
 
 static int xgbe_get_sset_count(struct net_device *netdev, int stringset)
 {
+	struct xgbe_prv_data *pdata = netdev_priv(netdev);
 	int ret;
 
 	switch (stringset) {
 	case ETH_SS_STATS:
-		ret = XGBE_STATS_COUNT;
+		ret = XGBE_STATS_COUNT +
+		      (pdata->tx_ring_count * 2) +
+		      (pdata->rx_ring_count * 2);
 		break;
 
 	default:
@@ -243,11 +158,12 @@ static int xgbe_set_pauseparam(struct net_device *netdev,
 			       struct ethtool_pauseparam *pause)
 {
 	struct xgbe_prv_data *pdata = netdev_priv(netdev);
+	struct ethtool_link_ksettings *lks = &pdata->phy.lks;
 	int ret = 0;
 
 	if (pause->autoneg && (pdata->phy.autoneg != AUTONEG_ENABLE)) {
 		netdev_err(netdev,
-			   "autoneg disabled, pause autoneg not avialable\n");
+			   "autoneg disabled, pause autoneg not available\n");
 		return -EINVAL;
 	}
 
@@ -255,16 +171,21 @@ static int xgbe_set_pauseparam(struct net_device *netdev,
 	pdata->phy.tx_pause = pause->tx_pause;
 	pdata->phy.rx_pause = pause->rx_pause;
 
-	pdata->phy.advertising &= ~ADVERTISED_Pause;
-	pdata->phy.advertising &= ~ADVERTISED_Asym_Pause;
+	XGBE_CLR_ADV(lks, Pause);
+	XGBE_CLR_ADV(lks, Asym_Pause);
 
 	if (pause->rx_pause) {
-		pdata->phy.advertising |= ADVERTISED_Pause;
-		pdata->phy.advertising |= ADVERTISED_Asym_Pause;
+		XGBE_SET_ADV(lks, Pause);
+		XGBE_SET_ADV(lks, Asym_Pause);
 	}
 
-	if (pause->tx_pause)
-		pdata->phy.advertising ^= ADVERTISED_Asym_Pause;
+	if (pause->tx_pause) {
+		/* Equivalent to XOR of Asym_Pause */
+		if (XGBE_ADV(lks, Asym_Pause))
+			XGBE_CLR_ADV(lks, Asym_Pause);
+		else
+			XGBE_SET_ADV(lks, Asym_Pause);
+	}
 
 	if (netif_running(netdev))
 		ret = pdata->phy_if.phy_config_aneg(pdata);
@@ -276,22 +197,25 @@ static int xgbe_get_link_ksettings(struct net_device *netdev,
 				   struct ethtool_link_ksettings *cmd)
 {
 	struct xgbe_prv_data *pdata = netdev_priv(netdev);
+	struct ethtool_link_ksettings *lks = &pdata->phy.lks;
 
 	cmd->base.phy_address = pdata->phy.address;
 
-	ethtool_convert_legacy_u32_to_link_mode(cmd->link_modes.supported,
-						pdata->phy.supported);
-	ethtool_convert_legacy_u32_to_link_mode(cmd->link_modes.advertising,
-						pdata->phy.advertising);
-	ethtool_convert_legacy_u32_to_link_mode(cmd->link_modes.lp_advertising,
-						pdata->phy.lp_advertising);
+	if (netif_carrier_ok(netdev)) {
+		cmd->base.speed = pdata->phy.speed;
+		cmd->base.duplex = pdata->phy.duplex;
+	} else {
+		cmd->base.speed = SPEED_UNKNOWN;
+		cmd->base.duplex = DUPLEX_UNKNOWN;
+	}
 
 	cmd->base.autoneg = pdata->phy.autoneg;
-	cmd->base.speed = pdata->phy.speed;
-	cmd->base.duplex = pdata->phy.duplex;
-
 	cmd->base.port = PORT_NONE;
 
+	XGBE_LM_COPY(cmd, supported, lks, supported);
+	XGBE_LM_COPY(cmd, advertising, lks, advertising);
+	XGBE_LM_COPY(cmd, lp_advertising, lks, lp_advertising);
+
 	return 0;
 }
 
@@ -299,7 +223,8 @@ static int xgbe_set_link_ksettings(struct net_device *netdev,
 				   const struct ethtool_link_ksettings *cmd)
 {
 	struct xgbe_prv_data *pdata = netdev_priv(netdev);
-	u32 advertising;
+	struct ethtool_link_ksettings *lks = &pdata->phy.lks;
+	__ETHTOOL_DECLARE_LINK_MODE_MASK(advertising);
 	u32 speed;
 	int ret;
 
@@ -331,15 +256,16 @@ static int xgbe_set_link_ksettings(struct net_device *netdev,
 		}
 	}
 
-	ethtool_convert_link_mode_to_legacy_u32(&advertising,
-						cmd->link_modes.advertising);
-
 	netif_dbg(pdata, link, netdev,
-		  "requested advertisement %#x, phy supported %#x\n",
-		  advertising, pdata->phy.supported);
+		  "requested advertisement 0x%*pb, phy supported 0x%*pb\n",
+		  __ETHTOOL_LINK_MODE_MASK_NBITS, cmd->link_modes.advertising,
+		  __ETHTOOL_LINK_MODE_MASK_NBITS, lks->link_modes.supported);
 
-	advertising &= pdata->phy.supported;
-	if ((cmd->base.autoneg == AUTONEG_ENABLE) && !advertising) {
+	linkmode_and(advertising, cmd->link_modes.advertising,
+		     lks->link_modes.supported);
+
+	if ((cmd->base.autoneg == AUTONEG_ENABLE) &&
+	    bitmap_empty(advertising, __ETHTOOL_LINK_MODE_MASK_NBITS)) {
 		netdev_err(netdev,
 			   "unsupported requested advertisement\n");
 		return -EINVAL;
@@ -349,12 +275,12 @@ static int xgbe_set_link_ksettings(struct net_device *netdev,
 	pdata->phy.autoneg = cmd->base.autoneg;
 	pdata->phy.speed = speed;
 	pdata->phy.duplex = cmd->base.duplex;
-	pdata->phy.advertising = advertising;
+	linkmode_copy(lks->link_modes.advertising, advertising);
 
 	if (cmd->base.autoneg == AUTONEG_ENABLE)
-		pdata->phy.advertising |= ADVERTISED_Autoneg;
+		XGBE_SET_ADV(lks, Autoneg);
 	else
-		pdata->phy.advertising &= ~ADVERTISED_Autoneg;
+		XGBE_CLR_ADV(lks, Autoneg);
 
 	if (netif_running(netdev))
 		ret = pdata->phy_if.phy_config_aneg(pdata);
@@ -368,9 +294,8 @@ static void xgbe_get_drvinfo(struct net_device *netdev,
 	struct xgbe_prv_data *pdata = netdev_priv(netdev);
 	struct xgbe_hw_features *hw_feat = &pdata->hw_feat;
 
-	strlcpy(drvinfo->driver, XGBE_DRV_NAME, sizeof(drvinfo->driver));
-	strlcpy(drvinfo->version, XGBE_DRV_VERSION, sizeof(drvinfo->version));
-	strlcpy(drvinfo->bus_info, dev_name(pdata->dev),
+	strscpy(drvinfo->driver, XGBE_DRV_NAME, sizeof(drvinfo->driver));
+	strscpy(drvinfo->bus_info, dev_name(pdata->dev),
 		sizeof(drvinfo->bus_info));
 	snprintf(drvinfo->fw_version, sizeof(drvinfo->fw_version), "%d.%d.%d",
 		 XGMAC_GET_BITS(hw_feat->version, MAC_VR, USERVER),
@@ -393,7 +318,9 @@ static void xgbe_set_msglevel(struct net_device *netdev, u32 msglevel)
 }
 
 static int xgbe_get_coalesce(struct net_device *netdev,
-			     struct ethtool_coalesce *ec)
+			     struct ethtool_coalesce *ec,
+			     struct kernel_ethtool_coalesce *kernel_coal,
+			     struct netlink_ext_ack *extack)
 {
 	struct xgbe_prv_data *pdata = netdev_priv(netdev);
 
@@ -402,42 +329,22 @@ static int xgbe_get_coalesce(struct net_device *netdev,
 	ec->rx_coalesce_usecs = pdata->rx_usecs;
 	ec->rx_max_coalesced_frames = pdata->rx_frames;
 
+	ec->tx_coalesce_usecs = pdata->tx_usecs;
 	ec->tx_max_coalesced_frames = pdata->tx_frames;
 
 	return 0;
 }
 
 static int xgbe_set_coalesce(struct net_device *netdev,
-			     struct ethtool_coalesce *ec)
+			     struct ethtool_coalesce *ec,
+			     struct kernel_ethtool_coalesce *kernel_coal,
+			     struct netlink_ext_ack *extack)
 {
 	struct xgbe_prv_data *pdata = netdev_priv(netdev);
 	struct xgbe_hw_if *hw_if = &pdata->hw_if;
 	unsigned int rx_frames, rx_riwt, rx_usecs;
-	unsigned int tx_frames;
-
-	/* Check for not supported parameters  */
-	if ((ec->rx_coalesce_usecs_irq) ||
-	    (ec->rx_max_coalesced_frames_irq) ||
-	    (ec->tx_coalesce_usecs) ||
-	    (ec->tx_coalesce_usecs_irq) ||
-	    (ec->tx_max_coalesced_frames_irq) ||
-	    (ec->stats_block_coalesce_usecs) ||
-	    (ec->use_adaptive_rx_coalesce) ||
-	    (ec->use_adaptive_tx_coalesce) ||
-	    (ec->pkt_rate_low) ||
-	    (ec->rx_coalesce_usecs_low) ||
-	    (ec->rx_max_coalesced_frames_low) ||
-	    (ec->tx_coalesce_usecs_low) ||
-	    (ec->tx_max_coalesced_frames_low) ||
-	    (ec->pkt_rate_high) ||
-	    (ec->rx_coalesce_usecs_high) ||
-	    (ec->rx_max_coalesced_frames_high) ||
-	    (ec->tx_coalesce_usecs_high) ||
-	    (ec->tx_max_coalesced_frames_high) ||
-	    (ec->rate_sample_interval)) {
-		netdev_err(netdev, "unsupported coalescing parameter\n");
-		return -EOPNOTSUPP;
-	}
+	unsigned int tx_frames, tx_usecs;
+	unsigned int jiffy_us = jiffies_to_usecs(1);
 
 	rx_riwt = hw_if->usec_to_riwt(pdata, ec->rx_coalesce_usecs);
 	rx_usecs = ec->rx_coalesce_usecs;
@@ -459,20 +366,42 @@ static int xgbe_set_coalesce(struct net_device *netdev,
 		return -EINVAL;
 	}
 
+	tx_usecs = ec->tx_coalesce_usecs;
 	tx_frames = ec->tx_max_coalesced_frames;
 
 	/* Check the bounds of values for Tx */
+	if (!tx_usecs) {
+		NL_SET_ERR_MSG_FMT_MOD(extack,
+				       "tx-usecs must not be 0");
+		return -EINVAL;
+	}
+	if (tx_usecs > XGMAC_MAX_COAL_TX_TICK) {
+		NL_SET_ERR_MSG_FMT_MOD(extack, "tx-usecs is limited to %d usec",
+				       XGMAC_MAX_COAL_TX_TICK);
+		return -EINVAL;
+	}
 	if (tx_frames > pdata->tx_desc_count) {
 		netdev_err(netdev, "tx-frames is limited to %d frames\n",
 			   pdata->tx_desc_count);
 		return -EINVAL;
 	}
 
+	/* Round tx-usecs to nearest multiple of jiffy granularity */
+	if (tx_usecs % jiffy_us) {
+		tx_usecs = rounddown(tx_usecs, jiffy_us);
+		if (!tx_usecs)
+			tx_usecs = jiffy_us;
+		NL_SET_ERR_MSG_FMT_MOD(extack,
+				       "tx-usecs rounded to %u usec due to jiffy granularity (%u usec)",
+				       tx_usecs, jiffy_us);
+	}
+
 	pdata->rx_riwt = rx_riwt;
 	pdata->rx_usecs = rx_usecs;
 	pdata->rx_frames = rx_frames;
 	hw_if->config_rx_coalesce(pdata);
 
+	pdata->tx_usecs = tx_usecs;
 	pdata->tx_frames = tx_frames;
 	hw_if->config_tx_coalesce(pdata);
 
@@ -509,47 +438,48 @@ static u32 xgbe_get_rxfh_indir_size(struct net_device *netdev)
 	return ARRAY_SIZE(pdata->rss_table);
 }
 
-static int xgbe_get_rxfh(struct net_device *netdev, u32 *indir, u8 *key,
-			 u8 *hfunc)
+static int xgbe_get_rxfh(struct net_device *netdev,
+			 struct ethtool_rxfh_param *rxfh)
 {
 	struct xgbe_prv_data *pdata = netdev_priv(netdev);
 	unsigned int i;
 
-	if (indir) {
+	if (rxfh->indir) {
 		for (i = 0; i < ARRAY_SIZE(pdata->rss_table); i++)
-			indir[i] = XGMAC_GET_BITS(pdata->rss_table[i],
-						  MAC_RSSDR, DMCH);
+			rxfh->indir[i] = XGMAC_GET_BITS(pdata->rss_table[i],
+							MAC_RSSDR, DMCH);
 	}
 
-	if (key)
-		memcpy(key, pdata->rss_key, sizeof(pdata->rss_key));
+	if (rxfh->key)
+		memcpy(rxfh->key, pdata->rss_key, sizeof(pdata->rss_key));
 
-	if (hfunc)
-		*hfunc = ETH_RSS_HASH_TOP;
+	rxfh->hfunc = ETH_RSS_HASH_TOP;
 
 	return 0;
 }
 
-static int xgbe_set_rxfh(struct net_device *netdev, const u32 *indir,
-			 const u8 *key, const u8 hfunc)
+static int xgbe_set_rxfh(struct net_device *netdev,
+			 struct ethtool_rxfh_param *rxfh,
+			 struct netlink_ext_ack *extack)
 {
 	struct xgbe_prv_data *pdata = netdev_priv(netdev);
 	struct xgbe_hw_if *hw_if = &pdata->hw_if;
-	unsigned int ret;
+	int ret;
 
-	if (hfunc != ETH_RSS_HASH_NO_CHANGE && hfunc != ETH_RSS_HASH_TOP) {
+	if (rxfh->hfunc != ETH_RSS_HASH_NO_CHANGE &&
+	    rxfh->hfunc != ETH_RSS_HASH_TOP) {
 		netdev_err(netdev, "unsupported hash function\n");
 		return -EOPNOTSUPP;
 	}
 
-	if (indir) {
-		ret = hw_if->set_rss_lookup_table(pdata, indir);
+	if (rxfh->indir) {
+		ret = hw_if->set_rss_lookup_table(pdata, rxfh->indir);
 		if (ret)
 			return ret;
 	}
 
-	if (key) {
-		ret = hw_if->set_rss_hash_key(pdata, key);
+	if (rxfh->key) {
+		ret = hw_if->set_rss_hash_key(pdata, rxfh->key);
 		if (ret)
 			return ret;
 	}
@@ -558,21 +488,17 @@ static int xgbe_set_rxfh(struct net_device *netdev, const u32 *indir,
 }
 
 static int xgbe_get_ts_info(struct net_device *netdev,
-			    struct ethtool_ts_info *ts_info)
+			    struct kernel_ethtool_ts_info *ts_info)
 {
 	struct xgbe_prv_data *pdata = netdev_priv(netdev);
 
 	ts_info->so_timestamping = SOF_TIMESTAMPING_TX_SOFTWARE |
-				   SOF_TIMESTAMPING_RX_SOFTWARE |
-				   SOF_TIMESTAMPING_SOFTWARE |
 				   SOF_TIMESTAMPING_TX_HARDWARE |
 				   SOF_TIMESTAMPING_RX_HARDWARE |
 				   SOF_TIMESTAMPING_RAW_HARDWARE;
 
 	if (pdata->ptp_clock)
 		ts_info->phc_index = ptp_clock_index(pdata->ptp_clock);
-	else
-		ts_info->phc_index = -1;
 
 	ts_info->tx_types = (1 << HWTSTAMP_TX_OFF) | (1 << HWTSTAMP_TX_ON);
 	ts_info->rx_filters = (1 << HWTSTAMP_FILTER_NONE) |
@@ -590,7 +516,225 @@ static int xgbe_get_ts_info(struct net_device *netdev,
 	return 0;
 }
 
+static int xgbe_get_module_info(struct net_device *netdev,
+				struct ethtool_modinfo *modinfo)
+{
+	struct xgbe_prv_data *pdata = netdev_priv(netdev);
+
+	return pdata->phy_if.module_info(pdata, modinfo);
+}
+
+static int xgbe_get_module_eeprom(struct net_device *netdev,
+				  struct ethtool_eeprom *eeprom, u8 *data)
+{
+	struct xgbe_prv_data *pdata = netdev_priv(netdev);
+
+	return pdata->phy_if.module_eeprom(pdata, eeprom, data);
+}
+
+static void
+xgbe_get_ringparam(struct net_device *netdev,
+		   struct ethtool_ringparam *ringparam,
+		   struct kernel_ethtool_ringparam *kernel_ringparam,
+		   struct netlink_ext_ack *extack)
+{
+	struct xgbe_prv_data *pdata = netdev_priv(netdev);
+
+	ringparam->rx_max_pending = XGBE_RX_DESC_CNT_MAX;
+	ringparam->tx_max_pending = XGBE_TX_DESC_CNT_MAX;
+	ringparam->rx_pending = pdata->rx_desc_count;
+	ringparam->tx_pending = pdata->tx_desc_count;
+}
+
+static int xgbe_set_ringparam(struct net_device *netdev,
+			      struct ethtool_ringparam *ringparam,
+			      struct kernel_ethtool_ringparam *kernel_ringparam,
+			      struct netlink_ext_ack *extack)
+{
+	struct xgbe_prv_data *pdata = netdev_priv(netdev);
+	unsigned int rx, tx;
+
+	if (ringparam->rx_mini_pending || ringparam->rx_jumbo_pending) {
+		netdev_err(netdev, "unsupported ring parameter\n");
+		return -EINVAL;
+	}
+
+	if ((ringparam->rx_pending < XGBE_RX_DESC_CNT_MIN) ||
+	    (ringparam->rx_pending > XGBE_RX_DESC_CNT_MAX)) {
+		netdev_err(netdev,
+			   "rx ring parameter must be between %u and %u\n",
+			   XGBE_RX_DESC_CNT_MIN, XGBE_RX_DESC_CNT_MAX);
+		return -EINVAL;
+	}
+
+	if ((ringparam->tx_pending < XGBE_TX_DESC_CNT_MIN) ||
+	    (ringparam->tx_pending > XGBE_TX_DESC_CNT_MAX)) {
+		netdev_err(netdev,
+			   "tx ring parameter must be between %u and %u\n",
+			   XGBE_TX_DESC_CNT_MIN, XGBE_TX_DESC_CNT_MAX);
+		return -EINVAL;
+	}
+
+	rx = __rounddown_pow_of_two(ringparam->rx_pending);
+	if (rx != ringparam->rx_pending)
+		netdev_notice(netdev,
+			      "rx ring parameter rounded to power of two: %u\n",
+			      rx);
+
+	tx = __rounddown_pow_of_two(ringparam->tx_pending);
+	if (tx != ringparam->tx_pending)
+		netdev_notice(netdev,
+			      "tx ring parameter rounded to power of two: %u\n",
+			      tx);
+
+	if ((rx == pdata->rx_desc_count) &&
+	    (tx == pdata->tx_desc_count))
+		goto out;
+
+	pdata->rx_desc_count = rx;
+	pdata->tx_desc_count = tx;
+
+	xgbe_restart_dev(pdata);
+
+out:
+	return 0;
+}
+
+static void xgbe_get_channels(struct net_device *netdev,
+			      struct ethtool_channels *channels)
+{
+	struct xgbe_prv_data *pdata = netdev_priv(netdev);
+	unsigned int rx, tx, combined;
+
+	/* Calculate maximums allowed:
+	 *   - Take into account the number of available IRQs
+	 *   - Do not take into account the number of online CPUs so that
+	 *     the user can over-subscribe if desired
+	 *   - Tx is additionally limited by the number of hardware queues
+	 */
+	rx = min(pdata->hw_feat.rx_ch_cnt, pdata->rx_max_channel_count);
+	rx = min(rx, pdata->channel_irq_count);
+	tx = min(pdata->hw_feat.tx_ch_cnt, pdata->tx_max_channel_count);
+	tx = min(tx, pdata->channel_irq_count);
+	tx = min(tx, pdata->tx_max_q_count);
+
+	combined = min(rx, tx);
+
+	channels->max_combined = combined;
+	channels->max_rx = rx ? rx - 1 : 0;
+	channels->max_tx = tx ? tx - 1 : 0;
+
+	/* Get current settings based on device state */
+	rx = pdata->new_rx_ring_count ? : pdata->rx_ring_count;
+	tx = pdata->new_tx_ring_count ? : pdata->tx_ring_count;
+
+	combined = min(rx, tx);
+	rx -= combined;
+	tx -= combined;
+
+	channels->combined_count = combined;
+	channels->rx_count = rx;
+	channels->tx_count = tx;
+}
+
+static void xgbe_print_set_channels_input(struct net_device *netdev,
+					  struct ethtool_channels *channels)
+{
+	netdev_err(netdev, "channel inputs: combined=%u, rx-only=%u, tx-only=%u\n",
+		   channels->combined_count, channels->rx_count,
+		   channels->tx_count);
+}
+
+static int xgbe_set_channels(struct net_device *netdev,
+			     struct ethtool_channels *channels)
+{
+	struct xgbe_prv_data *pdata = netdev_priv(netdev);
+	unsigned int rx, rx_curr, tx, tx_curr, combined;
+
+	/* Calculate maximums allowed:
+	 *   - Take into account the number of available IRQs
+	 *   - Do not take into account the number of online CPUs so that
+	 *     the user can over-subscribe if desired
+	 *   - Tx is additionally limited by the number of hardware queues
+	 */
+	rx = min(pdata->hw_feat.rx_ch_cnt, pdata->rx_max_channel_count);
+	rx = min(rx, pdata->channel_irq_count);
+	tx = min(pdata->hw_feat.tx_ch_cnt, pdata->tx_max_channel_count);
+	tx = min(tx, pdata->tx_max_q_count);
+	tx = min(tx, pdata->channel_irq_count);
+
+	combined = min(rx, tx);
+
+	/* Should not be setting other count */
+	if (channels->other_count) {
+		netdev_err(netdev,
+			   "other channel count must be zero\n");
+		return -EINVAL;
+	}
+
+	/* Require at least one Combined (Rx and Tx) channel */
+	if (!channels->combined_count) {
+		netdev_err(netdev,
+			   "at least one combined Rx/Tx channel is required\n");
+		xgbe_print_set_channels_input(netdev, channels);
+		return -EINVAL;
+	}
+
+	/* Check combined channels */
+	if (channels->combined_count > combined) {
+		netdev_err(netdev,
+			   "combined channel count cannot exceed %u\n",
+			   combined);
+		xgbe_print_set_channels_input(netdev, channels);
+		return -EINVAL;
+	}
+
+	/* Can have some Rx-only or Tx-only channels, but not both */
+	if (channels->rx_count && channels->tx_count) {
+		netdev_err(netdev,
+			   "cannot specify both Rx-only and Tx-only channels\n");
+		xgbe_print_set_channels_input(netdev, channels);
+		return -EINVAL;
+	}
+
+	/* Check that we don't exceed the maximum number of channels */
+	if ((channels->combined_count + channels->rx_count) > rx) {
+		netdev_err(netdev,
+			   "total Rx channels (%u) requested exceeds maximum available (%u)\n",
+			   channels->combined_count + channels->rx_count, rx);
+		xgbe_print_set_channels_input(netdev, channels);
+		return -EINVAL;
+	}
+
+	if ((channels->combined_count + channels->tx_count) > tx) {
+		netdev_err(netdev,
+			   "total Tx channels (%u) requested exceeds maximum available (%u)\n",
+			   channels->combined_count + channels->tx_count, tx);
+		xgbe_print_set_channels_input(netdev, channels);
+		return -EINVAL;
+	}
+
+	rx = channels->combined_count + channels->rx_count;
+	tx = channels->combined_count + channels->tx_count;
+
+	rx_curr = pdata->new_rx_ring_count ? : pdata->rx_ring_count;
+	tx_curr = pdata->new_tx_ring_count ? : pdata->tx_ring_count;
+
+	if ((rx == rx_curr) && (tx == tx_curr))
+		goto out;
+
+	pdata->new_rx_ring_count = rx;
+	pdata->new_tx_ring_count = tx;
+
+	xgbe_full_restart_dev(pdata);
+
+out:
+	return 0;
+}
+
 static const struct ethtool_ops xgbe_ethtool_ops = {
+	.supported_coalesce_params = ETHTOOL_COALESCE_USECS |
+				     ETHTOOL_COALESCE_MAX_FRAMES,
 	.get_drvinfo = xgbe_get_drvinfo,
 	.get_msglevel = xgbe_get_msglevel,
 	.set_msglevel = xgbe_set_msglevel,
@@ -610,6 +754,12 @@ static const struct ethtool_ops xgbe_ethtool_ops = {
 	.get_ts_info = xgbe_get_ts_info,
 	.get_link_ksettings = xgbe_get_link_ksettings,
 	.set_link_ksettings = xgbe_set_link_ksettings,
+	.get_module_info = xgbe_get_module_info,
+	.get_module_eeprom = xgbe_get_module_eeprom,
+	.get_ringparam = xgbe_get_ringparam,
+	.set_ringparam = xgbe_set_ringparam,
+	.get_channels = xgbe_get_channels,
+	.set_channels = xgbe_set_channels,
 };
 
 const struct ethtool_ops *xgbe_get_ethtool_ops(void)
diff --git a/drivers/net/ethernet/amd/xgbe/xgbe-hwtstamp.c b/drivers/net/ethernet/amd/xgbe/xgbe-hwtstamp.c
new file mode 100644
index 000000000000..bc52e5ec6420
--- /dev/null
+++ b/drivers/net/ethernet/amd/xgbe/xgbe-hwtstamp.c
@@ -0,0 +1,401 @@
+// SPDX-License-Identifier: (GPL-2.0-or-later OR BSD-3-Clause)
+/*
+ * Copyright (c) 2014-2025, Advanced Micro Devices, Inc.
+ * Copyright (c) 2014, Synopsys, Inc.
+ * All rights reserved
+ *
+ * Author: Raju Rangoju <Raju.Rangoju@amd.com>
+ */
+
+#include "xgbe.h"
+#include "xgbe-common.h"
+
+void xgbe_update_tstamp_time(struct xgbe_prv_data *pdata,
+			     unsigned int sec, unsigned int nsec)
+{
+	int count;
+
+	/* Set the time values and tell the device */
+	XGMAC_IOWRITE(pdata, MAC_STSUR, sec);
+	XGMAC_IOWRITE(pdata, MAC_STNUR, nsec);
+
+	/* issue command to update the system time value */
+	XGMAC_IOWRITE(pdata, MAC_TSCR,
+		      XGMAC_IOREAD(pdata, MAC_TSCR) |
+		      (1 << MAC_TSCR_TSUPDT_INDEX));
+
+	/* Wait for the time adjust/update to complete */
+	count = 10000;
+	while (--count && XGMAC_IOREAD_BITS(pdata, MAC_TSCR, TSUPDT))
+		udelay(5);
+
+	if (count < 0)
+		netdev_err(pdata->netdev,
+			   "timed out updating system timestamp\n");
+}
+
+void xgbe_update_tstamp_addend(struct xgbe_prv_data *pdata,
+			       unsigned int addend)
+{
+	unsigned int count = 10000;
+
+	/* Set the addend register value and tell the device */
+	XGMAC_IOWRITE(pdata, MAC_TSAR, addend);
+	XGMAC_IOWRITE_BITS(pdata, MAC_TSCR, TSADDREG, 1);
+
+	/* Wait for addend update to complete */
+	while (--count && XGMAC_IOREAD_BITS(pdata, MAC_TSCR, TSADDREG))
+		udelay(5);
+
+	if (!count)
+		netdev_err(pdata->netdev,
+			   "timed out updating timestamp addend register\n");
+}
+
+void xgbe_set_tstamp_time(struct xgbe_prv_data *pdata, unsigned int sec,
+			  unsigned int nsec)
+{
+	unsigned int count = 10000;
+
+	/* Set the time values and tell the device */
+	XGMAC_IOWRITE(pdata, MAC_STSUR, sec);
+	XGMAC_IOWRITE(pdata, MAC_STNUR, nsec);
+	XGMAC_IOWRITE_BITS(pdata, MAC_TSCR, TSINIT, 1);
+
+	/* Wait for time update to complete */
+	while (--count && XGMAC_IOREAD_BITS(pdata, MAC_TSCR, TSINIT))
+		udelay(5);
+
+	if (!count)
+		netdev_err(pdata->netdev, "timed out initializing timestamp\n");
+}
+
+u64 xgbe_get_tstamp_time(struct xgbe_prv_data *pdata)
+{
+	u64 nsec;
+
+	nsec = XGMAC_IOREAD(pdata, MAC_STSR);
+	nsec *= NSEC_PER_SEC;
+	nsec += XGMAC_IOREAD(pdata, MAC_STNR);
+
+	return nsec;
+}
+
+u64 xgbe_get_tx_tstamp(struct xgbe_prv_data *pdata)
+{
+	unsigned int tx_snr, tx_ssr;
+	u64 nsec;
+
+	if (pdata->vdata->tx_tstamp_workaround) {
+		tx_snr = XGMAC_IOREAD(pdata, MAC_TXSNR);
+		tx_ssr = XGMAC_IOREAD(pdata, MAC_TXSSR);
+	} else {
+		tx_ssr = XGMAC_IOREAD(pdata, MAC_TXSSR);
+		tx_snr = XGMAC_IOREAD(pdata, MAC_TXSNR);
+	}
+
+	if (XGMAC_GET_BITS(tx_snr, MAC_TXSNR, TXTSSTSMIS))
+		return 0;
+
+	nsec = tx_ssr;
+	nsec *= NSEC_PER_SEC;
+	nsec += tx_snr;
+
+	return nsec;
+}
+
+void xgbe_get_rx_tstamp(struct xgbe_packet_data *packet,
+			struct xgbe_ring_desc *rdesc)
+{
+	u64 nsec;
+
+	if (XGMAC_GET_BITS_LE(rdesc->desc3, RX_CONTEXT_DESC3, TSA) &&
+	    !XGMAC_GET_BITS_LE(rdesc->desc3, RX_CONTEXT_DESC3, TSD)) {
+		nsec = le32_to_cpu(rdesc->desc1);
+		nsec *= NSEC_PER_SEC;
+		nsec += le32_to_cpu(rdesc->desc0);
+		if (nsec != 0xffffffffffffffffULL) {
+			packet->rx_tstamp = nsec;
+			XGMAC_SET_BITS(packet->attributes, RX_PACKET_ATTRIBUTES,
+				       RX_TSTAMP, 1);
+		}
+	}
+}
+
+void xgbe_config_tstamp(struct xgbe_prv_data *pdata, unsigned int mac_tscr)
+{
+	unsigned int value = 0;
+
+	value = XGMAC_IOREAD(pdata, MAC_TSCR);
+	value |= mac_tscr;
+	XGMAC_IOWRITE(pdata, MAC_TSCR, value);
+}
+
+void xgbe_tx_tstamp(struct work_struct *work)
+{
+	struct xgbe_prv_data *pdata = container_of(work,
+						   struct xgbe_prv_data,
+						   tx_tstamp_work);
+	struct skb_shared_hwtstamps hwtstamps;
+	unsigned long flags;
+
+	spin_lock_irqsave(&pdata->tstamp_lock, flags);
+	if (!pdata->tx_tstamp_skb)
+		goto unlock;
+
+	if (pdata->tx_tstamp) {
+		memset(&hwtstamps, 0, sizeof(hwtstamps));
+		hwtstamps.hwtstamp = ns_to_ktime(pdata->tx_tstamp);
+		skb_tstamp_tx(pdata->tx_tstamp_skb, &hwtstamps);
+	}
+
+	dev_kfree_skb_any(pdata->tx_tstamp_skb);
+
+	pdata->tx_tstamp_skb = NULL;
+
+unlock:
+	spin_unlock_irqrestore(&pdata->tstamp_lock, flags);
+}
+
+int xgbe_get_hwtstamp_settings(struct xgbe_prv_data *pdata, struct ifreq *ifreq)
+{
+	if (copy_to_user(ifreq->ifr_data, &pdata->tstamp_config,
+			 sizeof(pdata->tstamp_config)))
+		return -EFAULT;
+
+	return 0;
+}
+
+int xgbe_set_hwtstamp_settings(struct xgbe_prv_data *pdata, struct ifreq *ifreq)
+{
+	struct hwtstamp_config config;
+	unsigned int mac_tscr;
+
+	if (copy_from_user(&config, ifreq->ifr_data, sizeof(config)))
+		return -EFAULT;
+
+	mac_tscr = 0;
+
+	switch (config.tx_type) {
+	case HWTSTAMP_TX_OFF:
+		break;
+
+	case HWTSTAMP_TX_ON:
+		XGMAC_SET_BITS(mac_tscr, MAC_TSCR, TSENA, 1);
+		break;
+
+	default:
+		return -ERANGE;
+	}
+
+	switch (config.rx_filter) {
+	case HWTSTAMP_FILTER_NONE:
+		break;
+
+	case HWTSTAMP_FILTER_NTP_ALL:
+	case HWTSTAMP_FILTER_ALL:
+		XGMAC_SET_BITS(mac_tscr, MAC_TSCR, TSENALL, 1);
+		XGMAC_SET_BITS(mac_tscr, MAC_TSCR, TSENA, 1);
+		break;
+
+		/* PTP v2, UDP, any kind of event packet */
+	case HWTSTAMP_FILTER_PTP_V2_L4_EVENT:
+		XGMAC_SET_BITS(mac_tscr, MAC_TSCR, TSVER2ENA, 1);
+		fallthrough;    /* to PTP v1, UDP, any kind of event packet */
+	case HWTSTAMP_FILTER_PTP_V1_L4_EVENT:
+		XGMAC_SET_BITS(mac_tscr, MAC_TSCR, TSIPV4ENA, 1);
+		XGMAC_SET_BITS(mac_tscr, MAC_TSCR, TSIPV6ENA, 1);
+		XGMAC_SET_BITS(mac_tscr, MAC_TSCR, SNAPTYPSEL, 1);
+		XGMAC_SET_BITS(mac_tscr, MAC_TSCR, TSENA, 1);
+		break;
+		/* PTP v2, UDP, Sync packet */
+	case HWTSTAMP_FILTER_PTP_V2_L4_SYNC:
+		XGMAC_SET_BITS(mac_tscr, MAC_TSCR, TSVER2ENA, 1);
+		fallthrough;    /* to PTP v1, UDP, Sync packet */
+	case HWTSTAMP_FILTER_PTP_V1_L4_SYNC:
+		XGMAC_SET_BITS(mac_tscr, MAC_TSCR, TSIPV4ENA, 1);
+		XGMAC_SET_BITS(mac_tscr, MAC_TSCR, TSIPV6ENA, 1);
+		XGMAC_SET_BITS(mac_tscr, MAC_TSCR, TSEVNTENA, 1);
+		XGMAC_SET_BITS(mac_tscr, MAC_TSCR, TSENA, 1);
+		break;
+
+		/* PTP v2, UDP, Delay_req packet */
+	case HWTSTAMP_FILTER_PTP_V2_L4_DELAY_REQ:
+		XGMAC_SET_BITS(mac_tscr, MAC_TSCR, TSVER2ENA, 1);
+		fallthrough;    /* to PTP v1, UDP, Delay_req packet */
+	case HWTSTAMP_FILTER_PTP_V1_L4_DELAY_REQ:
+		XGMAC_SET_BITS(mac_tscr, MAC_TSCR, TSIPV4ENA, 1);
+		XGMAC_SET_BITS(mac_tscr, MAC_TSCR, TSIPV6ENA, 1);
+		XGMAC_SET_BITS(mac_tscr, MAC_TSCR, TSEVNTENA, 1);
+		XGMAC_SET_BITS(mac_tscr, MAC_TSCR, TSMSTRENA, 1);
+		XGMAC_SET_BITS(mac_tscr, MAC_TSCR, TSENA, 1);
+		break;
+
+		/* 802.AS1, Ethernet, any kind of event packet */
+	case HWTSTAMP_FILTER_PTP_V2_L2_EVENT:
+		XGMAC_SET_BITS(mac_tscr, MAC_TSCR, AV8021ASMEN, 1);
+		XGMAC_SET_BITS(mac_tscr, MAC_TSCR, SNAPTYPSEL, 1);
+		XGMAC_SET_BITS(mac_tscr, MAC_TSCR, TSENA, 1);
+		break;
+
+		/* 802.AS1, Ethernet, Sync packet */
+	case HWTSTAMP_FILTER_PTP_V2_L2_SYNC:
+		XGMAC_SET_BITS(mac_tscr, MAC_TSCR, AV8021ASMEN, 1);
+		XGMAC_SET_BITS(mac_tscr, MAC_TSCR, TSEVNTENA, 1);
+		XGMAC_SET_BITS(mac_tscr, MAC_TSCR, TSENA, 1);
+		break;
+
+		/* 802.AS1, Ethernet, Delay_req packet */
+	case HWTSTAMP_FILTER_PTP_V2_L2_DELAY_REQ:
+		XGMAC_SET_BITS(mac_tscr, MAC_TSCR, AV8021ASMEN, 1);
+		XGMAC_SET_BITS(mac_tscr, MAC_TSCR, TSMSTRENA, 1);
+		XGMAC_SET_BITS(mac_tscr, MAC_TSCR, TSEVNTENA, 1);
+		XGMAC_SET_BITS(mac_tscr, MAC_TSCR, TSENA, 1);
+		break;
+
+		/* PTP v2/802.AS1, any layer, any kind of event packet */
+	case HWTSTAMP_FILTER_PTP_V2_EVENT:
+		XGMAC_SET_BITS(mac_tscr, MAC_TSCR, TSVER2ENA, 1);
+		XGMAC_SET_BITS(mac_tscr, MAC_TSCR, TSIPENA, 1);
+		XGMAC_SET_BITS(mac_tscr, MAC_TSCR, TSIPV4ENA, 1);
+		XGMAC_SET_BITS(mac_tscr, MAC_TSCR, TSIPV6ENA, 1);
+		XGMAC_SET_BITS(mac_tscr, MAC_TSCR, SNAPTYPSEL, 1);
+		XGMAC_SET_BITS(mac_tscr, MAC_TSCR, TSENA, 1);
+		break;
+
+		/* PTP v2/802.AS1, any layer, Sync packet */
+	case HWTSTAMP_FILTER_PTP_V2_SYNC:
+		XGMAC_SET_BITS(mac_tscr, MAC_TSCR, TSVER2ENA, 1);
+		XGMAC_SET_BITS(mac_tscr, MAC_TSCR, TSIPENA, 1);
+		XGMAC_SET_BITS(mac_tscr, MAC_TSCR, TSIPV4ENA, 1);
+		XGMAC_SET_BITS(mac_tscr, MAC_TSCR, TSIPV6ENA, 1);
+		XGMAC_SET_BITS(mac_tscr, MAC_TSCR, TSEVNTENA, 1);
+		XGMAC_SET_BITS(mac_tscr, MAC_TSCR, TSENA, 1);
+		break;
+
+		/* PTP v2/802.AS1, any layer, Delay_req packet */
+	case HWTSTAMP_FILTER_PTP_V2_DELAY_REQ:
+		XGMAC_SET_BITS(mac_tscr, MAC_TSCR, TSVER2ENA, 1);
+		XGMAC_SET_BITS(mac_tscr, MAC_TSCR, TSIPENA, 1);
+		XGMAC_SET_BITS(mac_tscr, MAC_TSCR, TSIPV4ENA, 1);
+		XGMAC_SET_BITS(mac_tscr, MAC_TSCR, TSIPV6ENA, 1);
+		XGMAC_SET_BITS(mac_tscr, MAC_TSCR, TSMSTRENA, 1);
+		XGMAC_SET_BITS(mac_tscr, MAC_TSCR, TSEVNTENA, 1);
+		XGMAC_SET_BITS(mac_tscr, MAC_TSCR, TSENA, 1);
+		break;
+
+	default:
+		return -ERANGE;
+	}
+
+	xgbe_config_tstamp(pdata, mac_tscr);
+
+	memcpy(&pdata->tstamp_config, &config, sizeof(config));
+
+	return 0;
+}
+
+void xgbe_prep_tx_tstamp(struct xgbe_prv_data *pdata,
+			 struct sk_buff *skb,
+			 struct xgbe_packet_data *packet)
+{
+	unsigned long flags;
+
+	if (XGMAC_GET_BITS(packet->attributes, TX_PACKET_ATTRIBUTES, PTP)) {
+		spin_lock_irqsave(&pdata->tstamp_lock, flags);
+		if (pdata->tx_tstamp_skb) {
+			/* Another timestamp in progress, ignore this one */
+			XGMAC_SET_BITS(packet->attributes,
+				       TX_PACKET_ATTRIBUTES, PTP, 0);
+		} else {
+			pdata->tx_tstamp_skb = skb_get(skb);
+			skb_shinfo(skb)->tx_flags |= SKBTX_IN_PROGRESS;
+		}
+		spin_unlock_irqrestore(&pdata->tstamp_lock, flags);
+	}
+
+	skb_tx_timestamp(skb);
+}
+
+int xgbe_init_ptp(struct xgbe_prv_data *pdata)
+{
+	unsigned int mac_tscr = 0;
+	struct timespec64 now;
+	u64 dividend;
+
+	/* Register Settings to be done based on the link speed. */
+	switch (pdata->phy.speed) {
+	case SPEED_1000:
+		XGMAC_IOWRITE(pdata, MAC_TICNR, MAC_TICNR_1G_INITVAL);
+		XGMAC_IOWRITE(pdata, MAC_TECNR, MAC_TECNR_1G_INITVAL);
+		break;
+	case SPEED_2500:
+	case SPEED_10000:
+		XGMAC_IOWRITE_BITS(pdata, MAC_TICSNR, TSICSNS,
+				   MAC_TICSNR_10G_INITVAL);
+		XGMAC_IOWRITE(pdata, MAC_TECNR, MAC_TECNR_10G_INITVAL);
+		XGMAC_IOWRITE_BITS(pdata, MAC_TECSNR, TSECSNS,
+				   MAC_TECSNR_10G_INITVAL);
+		break;
+	case SPEED_UNKNOWN:
+	default:
+		break;
+	}
+
+	/* Enable IEEE1588 PTP clock. */
+	XGMAC_SET_BITS(mac_tscr, MAC_TSCR, TSENA, 1);
+
+	/* Overwrite earlier timestamps */
+	XGMAC_SET_BITS(mac_tscr, MAC_TSCR, TXTSSTSM, 1);
+
+	/* Set one nano-second accuracy */
+	XGMAC_SET_BITS(mac_tscr, MAC_TSCR, TSCTRLSSR, 1);
+
+	/* Set fine timestamp update */
+	XGMAC_SET_BITS(mac_tscr, MAC_TSCR, TSCFUPDT, 1);
+
+	xgbe_config_tstamp(pdata, mac_tscr);
+
+	/* Exit if timestamping is not enabled */
+	if (!XGMAC_GET_BITS(mac_tscr, MAC_TSCR, TSENA))
+		return -EOPNOTSUPP;
+
+	if (pdata->vdata->tstamp_ptp_clock_freq) {
+		/* Initialize time registers based on
+		 * 125MHz PTP Clock Frequency
+		 */
+		XGMAC_IOWRITE_BITS(pdata, MAC_SSIR, SSINC,
+				   XGBE_V2_TSTAMP_SSINC);
+		XGMAC_IOWRITE_BITS(pdata, MAC_SSIR, SNSINC,
+				   XGBE_V2_TSTAMP_SNSINC);
+	} else {
+		/* Initialize time registers based on
+		 * 50MHz PTP Clock Frequency
+		 */
+		XGMAC_IOWRITE_BITS(pdata, MAC_SSIR, SSINC, XGBE_TSTAMP_SSINC);
+		XGMAC_IOWRITE_BITS(pdata, MAC_SSIR, SNSINC, XGBE_TSTAMP_SNSINC);
+	}
+
+	/* Calculate the addend:
+	 *   addend = 2^32 / (PTP ref clock / (PTP clock based on SSINC))
+	 *          = (2^32 * (PTP clock based on SSINC)) / PTP ref clock
+	 */
+	if (pdata->vdata->tstamp_ptp_clock_freq)
+		dividend = XGBE_V2_PTP_ACT_CLK_FREQ;
+	else
+		dividend = XGBE_PTP_ACT_CLK_FREQ;
+
+	dividend = (u64)(dividend << 32);
+	pdata->tstamp_addend = div_u64(dividend, pdata->ptpclk_rate);
+
+	xgbe_update_tstamp_addend(pdata, pdata->tstamp_addend);
+
+	dma_wmb();
+	/* initialize system time */
+	ktime_get_real_ts64(&now);
+
+	/* lower 32 bits of tv_sec are safe until y2106 */
+	xgbe_set_tstamp_time(pdata, (u32)now.tv_sec, now.tv_nsec);
+
+	return 0;
+}
diff --git a/drivers/net/ethernet/amd/xgbe/xgbe-i2c.c b/drivers/net/ethernet/amd/xgbe/xgbe-i2c.c
index 417bdb5982a9..65eb7b577b65 100644
--- a/drivers/net/ethernet/amd/xgbe/xgbe-i2c.c
+++ b/drivers/net/ethernet/amd/xgbe/xgbe-i2c.c
@@ -1,117 +1,8 @@
+// SPDX-License-Identifier: (GPL-2.0-or-later OR BSD-3-Clause)
 /*
- * AMD 10Gb Ethernet driver
- *
- * This file is available to you under your choice of the following two
- * licenses:
- *
- * License 1: GPLv2
- *
- * Copyright (c) 2016 Advanced Micro Devices, Inc.
- *
- * This file is free software; you may copy, redistribute and/or modify
- * it under the terms of the GNU General Public License as published by
- * the Free Software Foundation, either version 2 of the License, or (at
- * your option) any later version.
- *
- * This file is distributed in the hope that it will be useful, but
- * WITHOUT ANY WARRANTY; without even the implied warranty of
- * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
- * General Public License for more details.
- *
- * You should have received a copy of the GNU General Public License
- * along with this program.  If not, see <http://www.gnu.org/licenses/>.
- *
- * This file incorporates work covered by the following copyright and
- * permission notice:
- *     The Synopsys DWC ETHER XGMAC Software Driver and documentation
- *     (hereinafter "Software") is an unsupported proprietary work of Synopsys,
- *     Inc. unless otherwise expressly agreed to in writing between Synopsys
- *     and you.
- *
- *     The Software IS NOT an item of Licensed Software or Licensed Product
- *     under any End User Software License Agreement or Agreement for Licensed
- *     Product with Synopsys or any supplement thereto.  Permission is hereby
- *     granted, free of charge, to any person obtaining a copy of this software
- *     annotated with this license and the Software, to deal in the Software
- *     without restriction, including without limitation the rights to use,
- *     copy, modify, merge, publish, distribute, sublicense, and/or sell copies
- *     of the Software, and to permit persons to whom the Software is furnished
- *     to do so, subject to the following conditions:
- *
- *     The above copyright notice and this permission notice shall be included
- *     in all copies or substantial portions of the Software.
- *
- *     THIS SOFTWARE IS BEING DISTRIBUTED BY SYNOPSYS SOLELY ON AN "AS IS"
- *     BASIS AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED
- *     TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A
- *     PARTICULAR PURPOSE ARE HEREBY DISCLAIMED. IN NO EVENT SHALL SYNOPSYS
- *     BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR
- *     CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF
- *     SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS
- *     INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN
- *     CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE)
- *     ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF
- *     THE POSSIBILITY OF SUCH DAMAGE.
- *
- *
- * License 2: Modified BSD
- *
- * Copyright (c) 2016 Advanced Micro Devices, Inc.
- * All rights reserved.
- *
- * Redistribution and use in source and binary forms, with or without
- * modification, are permitted provided that the following conditions are met:
- *     * Redistributions of source code must retain the above copyright
- *       notice, this list of conditions and the following disclaimer.
- *     * Redistributions in binary form must reproduce the above copyright
- *       notice, this list of conditions and the following disclaimer in the
- *       documentation and/or other materials provided with the distribution.
- *     * Neither the name of Advanced Micro Devices, Inc. nor the
- *       names of its contributors may be used to endorse or promote products
- *       derived from this software without specific prior written permission.
- *
- * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS"
- * AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
- * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
- * ARE DISCLAIMED. IN NO EVENT SHALL <COPYRIGHT HOLDER> BE LIABLE FOR ANY
- * DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES
- * (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES;
- * LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND
- * ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
- * (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF
- * THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
- *
- * This file incorporates work covered by the following copyright and
- * permission notice:
- *     The Synopsys DWC ETHER XGMAC Software Driver and documentation
- *     (hereinafter "Software") is an unsupported proprietary work of Synopsys,
- *     Inc. unless otherwise expressly agreed to in writing between Synopsys
- *     and you.
- *
- *     The Software IS NOT an item of Licensed Software or Licensed Product
- *     under any End User Software License Agreement or Agreement for Licensed
- *     Product with Synopsys or any supplement thereto.  Permission is hereby
- *     granted, free of charge, to any person obtaining a copy of this software
- *     annotated with this license and the Software, to deal in the Software
- *     without restriction, including without limitation the rights to use,
- *     copy, modify, merge, publish, distribute, sublicense, and/or sell copies
- *     of the Software, and to permit persons to whom the Software is furnished
- *     to do so, subject to the following conditions:
- *
- *     The above copyright notice and this permission notice shall be included
- *     in all copies or substantial portions of the Software.
- *
- *     THIS SOFTWARE IS BEING DISTRIBUTED BY SYNOPSYS SOLELY ON AN "AS IS"
- *     BASIS AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED
- *     TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A
- *     PARTICULAR PURPOSE ARE HEREBY DISCLAIMED. IN NO EVENT SHALL SYNOPSYS
- *     BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR
- *     CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF
- *     SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS
- *     INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN
- *     CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE)
- *     ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF
- *     THE POSSIBILITY OF SUCH DAMAGE.
+ * Copyright (c) 2014-2025, Advanced Micro Devices, Inc.
+ * Copyright (c) 2014, Synopsys, Inc.
+ * All rights reserved
  */
 
 #include <linux/module.h>
@@ -179,7 +70,7 @@ static int xgbe_i2c_set_enable(struct xgbe_prv_data *pdata, bool enable)
 
 static int xgbe_i2c_disable(struct xgbe_prv_data *pdata)
 {
-	unsigned int ret;
+	int ret;
 
 	ret = xgbe_i2c_set_enable(pdata, false);
 	if (ret) {
@@ -274,13 +165,16 @@ static void xgbe_i2c_clear_isr_interrupts(struct xgbe_prv_data *pdata,
 		XI2C_IOREAD(pdata, IC_CLR_STOP_DET);
 }
 
-static irqreturn_t xgbe_i2c_isr(int irq, void *data)
+static void xgbe_i2c_isr_bh_work(struct work_struct *work)
 {
-	struct xgbe_prv_data *pdata = (struct xgbe_prv_data *)data;
+	struct xgbe_prv_data *pdata = from_work(pdata, work, i2c_bh_work);
 	struct xgbe_i2c_op_state *state = &pdata->i2c.op_state;
 	unsigned int isr;
 
 	isr = XI2C_IOREAD(pdata, IC_RAW_INTR_STAT);
+	if (!isr)
+		goto reissue_check;
+
 	netif_dbg(pdata, intr, pdata->netdev,
 		  "I2C interrupt received: status=%#010x\n", isr);
 
@@ -308,6 +202,21 @@ out:
 	if (state->ret || XI2C_GET_BITS(isr, IC_RAW_INTR_STAT, STOP_DET))
 		complete(&pdata->i2c_complete);
 
+reissue_check:
+	/* Reissue interrupt if status is not clear */
+	if (pdata->vdata->irq_reissue_support)
+		XP_IOWRITE(pdata, XP_INT_REISSUE_EN, 1 << 2);
+}
+
+static irqreturn_t xgbe_i2c_isr(int irq, void *data)
+{
+	struct xgbe_prv_data *pdata = (struct xgbe_prv_data *)data;
+
+	if (pdata->isr_as_bh_work)
+		queue_work(system_bh_wq, &pdata->i2c_bh_work);
+	else
+		xgbe_i2c_isr_bh_work(&pdata->i2c_bh_work);
+
 	return IRQ_HANDLED;
 }
 
@@ -349,12 +258,11 @@ static void xgbe_i2c_set_target(struct xgbe_prv_data *pdata, unsigned int addr)
 	XI2C_IOWRITE(pdata, IC_TAR, addr);
 }
 
-static irqreturn_t xgbe_i2c_combined_isr(int irq, struct xgbe_prv_data *pdata)
+static irqreturn_t xgbe_i2c_combined_isr(struct xgbe_prv_data *pdata)
 {
-	if (!XI2C_IOREAD(pdata, IC_RAW_INTR_STAT))
-		return IRQ_HANDLED;
+	xgbe_i2c_isr_bh_work(&pdata->i2c_bh_work);
 
-	return xgbe_i2c_isr(irq, pdata);
+	return IRQ_HANDLED;
 }
 
 static int xgbe_i2c_xfer(struct xgbe_prv_data *pdata, struct xgbe_i2c_op *op)
@@ -430,8 +338,10 @@ static void xgbe_i2c_stop(struct xgbe_prv_data *pdata)
 	xgbe_i2c_disable(pdata);
 	xgbe_i2c_clear_all_interrupts(pdata);
 
-	if (pdata->dev_irq != pdata->i2c_irq)
+	if (pdata->dev_irq != pdata->i2c_irq) {
 		devm_free_irq(pdata->dev, pdata->i2c_irq, pdata);
+		cancel_work_sync(&pdata->i2c_bh_work);
+	}
 }
 
 static int xgbe_i2c_start(struct xgbe_prv_data *pdata)
@@ -445,6 +355,8 @@ static int xgbe_i2c_start(struct xgbe_prv_data *pdata)
 
 	/* If we have a separate I2C irq, enable it */
 	if (pdata->dev_irq != pdata->i2c_irq) {
+		INIT_WORK(&pdata->i2c_bh_work, xgbe_i2c_isr_bh_work);
+
 		ret = devm_request_irq(pdata->dev, pdata->i2c_irq,
 				       xgbe_i2c_isr, 0, pdata->i2c_name,
 				       pdata);
diff --git a/drivers/net/ethernet/amd/xgbe/xgbe-main.c b/drivers/net/ethernet/amd/xgbe/xgbe-main.c
index 17ac8f9a51a0..d1f0419edb23 100644
--- a/drivers/net/ethernet/amd/xgbe/xgbe-main.c
+++ b/drivers/net/ethernet/amd/xgbe/xgbe-main.c
@@ -1,117 +1,8 @@
+// SPDX-License-Identifier: (GPL-2.0-or-later OR BSD-3-Clause)
 /*
- * AMD 10Gb Ethernet driver
- *
- * This file is available to you under your choice of the following two
- * licenses:
- *
- * License 1: GPLv2
- *
- * Copyright (c) 2014-2016 Advanced Micro Devices, Inc.
- *
- * This file is free software; you may copy, redistribute and/or modify
- * it under the terms of the GNU General Public License as published by
- * the Free Software Foundation, either version 2 of the License, or (at
- * your option) any later version.
- *
- * This file is distributed in the hope that it will be useful, but
- * WITHOUT ANY WARRANTY; without even the implied warranty of
- * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
- * General Public License for more details.
- *
- * You should have received a copy of the GNU General Public License
- * along with this program.  If not, see <http://www.gnu.org/licenses/>.
- *
- * This file incorporates work covered by the following copyright and
- * permission notice:
- *     The Synopsys DWC ETHER XGMAC Software Driver and documentation
- *     (hereinafter "Software") is an unsupported proprietary work of Synopsys,
- *     Inc. unless otherwise expressly agreed to in writing between Synopsys
- *     and you.
- *
- *     The Software IS NOT an item of Licensed Software or Licensed Product
- *     under any End User Software License Agreement or Agreement for Licensed
- *     Product with Synopsys or any supplement thereto.  Permission is hereby
- *     granted, free of charge, to any person obtaining a copy of this software
- *     annotated with this license and the Software, to deal in the Software
- *     without restriction, including without limitation the rights to use,
- *     copy, modify, merge, publish, distribute, sublicense, and/or sell copies
- *     of the Software, and to permit persons to whom the Software is furnished
- *     to do so, subject to the following conditions:
- *
- *     The above copyright notice and this permission notice shall be included
- *     in all copies or substantial portions of the Software.
- *
- *     THIS SOFTWARE IS BEING DISTRIBUTED BY SYNOPSYS SOLELY ON AN "AS IS"
- *     BASIS AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED
- *     TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A
- *     PARTICULAR PURPOSE ARE HEREBY DISCLAIMED. IN NO EVENT SHALL SYNOPSYS
- *     BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR
- *     CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF
- *     SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS
- *     INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN
- *     CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE)
- *     ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF
- *     THE POSSIBILITY OF SUCH DAMAGE.
- *
- *
- * License 2: Modified BSD
- *
- * Copyright (c) 2014-2016 Advanced Micro Devices, Inc.
- * All rights reserved.
- *
- * Redistribution and use in source and binary forms, with or without
- * modification, are permitted provided that the following conditions are met:
- *     * Redistributions of source code must retain the above copyright
- *       notice, this list of conditions and the following disclaimer.
- *     * Redistributions in binary form must reproduce the above copyright
- *       notice, this list of conditions and the following disclaimer in the
- *       documentation and/or other materials provided with the distribution.
- *     * Neither the name of Advanced Micro Devices, Inc. nor the
- *       names of its contributors may be used to endorse or promote products
- *       derived from this software without specific prior written permission.
- *
- * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS"
- * AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
- * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
- * ARE DISCLAIMED. IN NO EVENT SHALL <COPYRIGHT HOLDER> BE LIABLE FOR ANY
- * DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES
- * (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES;
- * LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND
- * ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
- * (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF
- * THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
- *
- * This file incorporates work covered by the following copyright and
- * permission notice:
- *     The Synopsys DWC ETHER XGMAC Software Driver and documentation
- *     (hereinafter "Software") is an unsupported proprietary work of Synopsys,
- *     Inc. unless otherwise expressly agreed to in writing between Synopsys
- *     and you.
- *
- *     The Software IS NOT an item of Licensed Software or Licensed Product
- *     under any End User Software License Agreement or Agreement for Licensed
- *     Product with Synopsys or any supplement thereto.  Permission is hereby
- *     granted, free of charge, to any person obtaining a copy of this software
- *     annotated with this license and the Software, to deal in the Software
- *     without restriction, including without limitation the rights to use,
- *     copy, modify, merge, publish, distribute, sublicense, and/or sell copies
- *     of the Software, and to permit persons to whom the Software is furnished
- *     to do so, subject to the following conditions:
- *
- *     The above copyright notice and this permission notice shall be included
- *     in all copies or substantial portions of the Software.
- *
- *     THIS SOFTWARE IS BEING DISTRIBUTED BY SYNOPSYS SOLELY ON AN "AS IS"
- *     BASIS AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED
- *     TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A
- *     PARTICULAR PURPOSE ARE HEREBY DISCLAIMED. IN NO EVENT SHALL SYNOPSYS
- *     BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR
- *     CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF
- *     SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS
- *     INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN
- *     CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE)
- *     ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF
- *     THE POSSIBILITY OF SUCH DAMAGE.
+ * Copyright (c) 2014-2025, Advanced Micro Devices, Inc.
+ * Copyright (c) 2014, Synopsys, Inc.
+ * All rights reserved
  */
 
 #include <linux/module.h>
@@ -120,17 +11,17 @@
 #include <linux/netdevice.h>
 #include <linux/etherdevice.h>
 #include <linux/io.h>
+#include <linux/notifier.h>
 
 #include "xgbe.h"
 #include "xgbe-common.h"
 
 MODULE_AUTHOR("Tom Lendacky <thomas.lendacky@amd.com>");
 MODULE_LICENSE("Dual BSD/GPL");
-MODULE_VERSION(XGBE_DRV_VERSION);
 MODULE_DESCRIPTION(XGBE_DRV_DESC);
 
 static int debug = -1;
-module_param(debug, int, S_IWUSR | S_IRUGO);
+module_param(debug, int, 0644);
 MODULE_PARM_DESC(debug, " Network interface message level setting");
 
 static const u32 default_msg_level = (NETIF_MSG_LINK | NETIF_MSG_IFDOWN |
@@ -140,14 +31,16 @@ static void xgbe_default_config(struct xgbe_prv_data *pdata)
 {
 	DBGPR("-->xgbe_default_config\n");
 
-	pdata->pblx8 = DMA_PBL_X8_ENABLE;
+	pdata->blen = DMA_SBMR_BLEN_64;
+	pdata->pbl = DMA_PBL_128;
+	pdata->aal = 1;
+	pdata->rd_osr_limit = 8;
+	pdata->wr_osr_limit = 8;
 	pdata->tx_sf_mode = MTL_TSF_ENABLE;
 	pdata->tx_threshold = MTL_TX_THRESHOLD_64;
-	pdata->tx_pbl = DMA_PBL_16;
 	pdata->tx_osp_mode = DMA_OSP_ENABLE;
 	pdata->rx_sf_mode = MTL_RSF_DISABLE;
 	pdata->rx_threshold = MTL_RX_THRESHOLD_64;
-	pdata->rx_pbl = DMA_PBL_16;
 	pdata->pause_autoneg = 1;
 	pdata->tx_pause = 1;
 	pdata->rx_pause = 1;
@@ -261,12 +154,11 @@ int xgbe_config_netdev(struct xgbe_prv_data *pdata)
 {
 	struct net_device *netdev = pdata->netdev;
 	struct device *dev = pdata->dev;
-	unsigned int i;
 	int ret;
 
 	netdev->irq = pdata->dev_irq;
 	netdev->base_addr = (unsigned long)pdata->xgmac_regs;
-	memcpy(netdev->dev_addr, pdata->mac_addr, netdev->addr_len);
+	eth_hw_addr_set(netdev, pdata->mac_addr);
 
 	/* Initialize ECC timestamps */
 	pdata->tx_sec_period = jiffies;
@@ -277,7 +169,11 @@ int xgbe_config_netdev(struct xgbe_prv_data *pdata)
 	pdata->desc_ded_period = jiffies;
 
 	/* Issue software reset to device */
-	pdata->hw_if.exit(pdata);
+	ret = pdata->hw_if.exit(pdata);
+	if (ret) {
+		dev_err(dev, "software reset failed\n");
+		return ret;
+	}
 
 	/* Set default configuration data */
 	xgbe_default_config(pdata);
@@ -316,31 +212,15 @@ int xgbe_config_netdev(struct xgbe_prv_data *pdata)
 				pdata->tx_ring_count, pdata->rx_ring_count);
 	}
 
-	/* Set the number of queues */
-	ret = netif_set_real_num_tx_queues(netdev, pdata->tx_ring_count);
-	if (ret) {
-		dev_err(dev, "error setting real tx queue count\n");
-		return ret;
-	}
-
-	ret = netif_set_real_num_rx_queues(netdev, pdata->rx_ring_count);
-	if (ret) {
-		dev_err(dev, "error setting real rx queue count\n");
-		return ret;
-	}
-
-	/* Initialize RSS hash key and lookup table */
+	/* Initialize RSS hash key */
 	netdev_rss_key_fill(pdata->rss_key, sizeof(pdata->rss_key));
 
-	for (i = 0; i < XGBE_RSS_MAX_TABLE_SIZE; i++)
-		XGMAC_SET_BITS(pdata->rss_table[i], MAC_RSSDR, DMCH,
-			       i % pdata->rx_ring_count);
-
 	XGMAC_SET_BITS(pdata->rss_options, MAC_RSSCR, IP2TE, 1);
 	XGMAC_SET_BITS(pdata->rss_options, MAC_RSSCR, TCP4TE, 1);
 	XGMAC_SET_BITS(pdata->rss_options, MAC_RSSCR, UDP4TE, 1);
 
 	/* Call MDIO/PHY initialization routine */
+	pdata->debugfs_an_cdr_workaround = pdata->vdata->an_cdr_workaround;
 	ret = pdata->phy_if.phy_init(pdata);
 	if (ret)
 		return ret;
@@ -367,6 +247,23 @@ int xgbe_config_netdev(struct xgbe_prv_data *pdata)
 	if (pdata->hw_feat.rss)
 		netdev->hw_features |= NETIF_F_RXHASH;
 
+	if (pdata->hw_feat.vxn) {
+		netdev->hw_enc_features = NETIF_F_SG |
+					  NETIF_F_IP_CSUM |
+					  NETIF_F_IPV6_CSUM |
+					  NETIF_F_RXCSUM |
+					  NETIF_F_TSO |
+					  NETIF_F_TSO6 |
+					  NETIF_F_GRO |
+					  NETIF_F_GSO_UDP_TUNNEL |
+					  NETIF_F_GSO_UDP_TUNNEL_CSUM;
+
+		netdev->hw_features |= NETIF_F_GSO_UDP_TUNNEL |
+				       NETIF_F_GSO_UDP_TUNNEL_CSUM;
+
+		netdev->udp_tunnel_nic_info = xgbe_get_udp_tunnel_info();
+	}
+
 	netdev->vlan_features |= NETIF_F_SG |
 				 NETIF_F_IP_CSUM |
 				 NETIF_F_IPV6_CSUM |
@@ -378,7 +275,7 @@ int xgbe_config_netdev(struct xgbe_prv_data *pdata)
 
 	netdev->priv_flags |= IFF_UNICAST_FLT;
 	netdev->min_mtu = 0;
-	netdev->max_mtu = XGMAC_JUMBO_PACKET_MTU;
+	netdev->max_mtu = XGMAC_GIANT_PACKET_MTU - XGBE_ETH_FRAME_HDR;
 
 	/* Use default watchdog timeout */
 	netdev->watchdog_timeo = 0;
@@ -393,35 +290,6 @@ int xgbe_config_netdev(struct xgbe_prv_data *pdata)
 		return ret;
 	}
 
-	/* Create the PHY/ANEG name based on netdev name */
-	snprintf(pdata->an_name, sizeof(pdata->an_name) - 1, "%s-pcs",
-		 netdev_name(netdev));
-
-	/* Create the ECC name based on netdev name */
-	snprintf(pdata->ecc_name, sizeof(pdata->ecc_name) - 1, "%s-ecc",
-		 netdev_name(netdev));
-
-	/* Create the I2C name based on netdev name */
-	snprintf(pdata->i2c_name, sizeof(pdata->i2c_name) - 1, "%s-i2c",
-		 netdev_name(netdev));
-
-	/* Create workqueues */
-	pdata->dev_workqueue =
-		create_singlethread_workqueue(netdev_name(netdev));
-	if (!pdata->dev_workqueue) {
-		netdev_err(netdev, "device workqueue creation failed\n");
-		ret = -ENOMEM;
-		goto err_netdev;
-	}
-
-	pdata->an_workqueue =
-		create_singlethread_workqueue(pdata->an_name);
-	if (!pdata->an_workqueue) {
-		netdev_err(netdev, "phy workqueue creation failed\n");
-		ret = -ENOMEM;
-		goto err_wq;
-	}
-
 	if (IS_REACHABLE(CONFIG_PTP_1588_CLOCK))
 		xgbe_ptp_register(pdata);
 
@@ -433,14 +301,6 @@ int xgbe_config_netdev(struct xgbe_prv_data *pdata)
 		  pdata->rx_ring_count);
 
 	return 0;
-
-err_wq:
-	destroy_workqueue(pdata->dev_workqueue);
-
-err_netdev:
-	unregister_netdev(netdev);
-
-	return ret;
 }
 
 void xgbe_deconfig_netdev(struct xgbe_prv_data *pdata)
@@ -452,30 +312,60 @@ void xgbe_deconfig_netdev(struct xgbe_prv_data *pdata)
 	if (IS_REACHABLE(CONFIG_PTP_1588_CLOCK))
 		xgbe_ptp_unregister(pdata);
 
+	unregister_netdev(netdev);
+
 	pdata->phy_if.phy_exit(pdata);
+}
 
-	flush_workqueue(pdata->an_workqueue);
-	destroy_workqueue(pdata->an_workqueue);
+static int xgbe_netdev_event(struct notifier_block *nb, unsigned long event,
+			     void *data)
+{
+	struct net_device *netdev = netdev_notifier_info_to_dev(data);
+	struct xgbe_prv_data *pdata = netdev_priv(netdev);
 
-	flush_workqueue(pdata->dev_workqueue);
-	destroy_workqueue(pdata->dev_workqueue);
+	if (netdev->netdev_ops != xgbe_get_netdev_ops())
+		goto out;
 
-	unregister_netdev(netdev);
+	switch (event) {
+	case NETDEV_CHANGENAME:
+		xgbe_debugfs_rename(pdata);
+		break;
+
+	default:
+		break;
+	}
+
+out:
+	return NOTIFY_DONE;
 }
 
+static struct notifier_block xgbe_netdev_notifier = {
+	.notifier_call = xgbe_netdev_event,
+};
+
 static int __init xgbe_mod_init(void)
 {
 	int ret;
 
-	ret = xgbe_platform_init();
+	ret = register_netdevice_notifier(&xgbe_netdev_notifier);
 	if (ret)
 		return ret;
 
+	ret = xgbe_platform_init();
+	if (ret)
+		goto err_platform_init;
+
 	ret = xgbe_pci_init();
 	if (ret)
-		return ret;
+		goto err_pci_init;
 
 	return 0;
+
+err_pci_init:
+	xgbe_platform_exit();
+err_platform_init:
+	unregister_netdevice_notifier(&xgbe_netdev_notifier);
+	return ret;
 }
 
 static void __exit xgbe_mod_exit(void)
@@ -483,6 +373,8 @@ static void __exit xgbe_mod_exit(void)
 	xgbe_pci_exit();
 
 	xgbe_platform_exit();
+
+	unregister_netdevice_notifier(&xgbe_netdev_notifier);
 }
 
 module_init(xgbe_mod_init);
diff --git a/drivers/net/ethernet/amd/xgbe/xgbe-mdio.c b/drivers/net/ethernet/amd/xgbe/xgbe-mdio.c
index b672d9249539..1a37ec45e650 100644
--- a/drivers/net/ethernet/amd/xgbe/xgbe-mdio.c
+++ b/drivers/net/ethernet/amd/xgbe/xgbe-mdio.c
@@ -1,117 +1,8 @@
+// SPDX-License-Identifier: (GPL-2.0-or-later OR BSD-3-Clause)
 /*
- * AMD 10Gb Ethernet driver
- *
- * This file is available to you under your choice of the following two
- * licenses:
- *
- * License 1: GPLv2
- *
- * Copyright (c) 2014-2016 Advanced Micro Devices, Inc.
- *
- * This file is free software; you may copy, redistribute and/or modify
- * it under the terms of the GNU General Public License as published by
- * the Free Software Foundation, either version 2 of the License, or (at
- * your option) any later version.
- *
- * This file is distributed in the hope that it will be useful, but
- * WITHOUT ANY WARRANTY; without even the implied warranty of
- * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
- * General Public License for more details.
- *
- * You should have received a copy of the GNU General Public License
- * along with this program.  If not, see <http://www.gnu.org/licenses/>.
- *
- * This file incorporates work covered by the following copyright and
- * permission notice:
- *     The Synopsys DWC ETHER XGMAC Software Driver and documentation
- *     (hereinafter "Software") is an unsupported proprietary work of Synopsys,
- *     Inc. unless otherwise expressly agreed to in writing between Synopsys
- *     and you.
- *
- *     The Software IS NOT an item of Licensed Software or Licensed Product
- *     under any End User Software License Agreement or Agreement for Licensed
- *     Product with Synopsys or any supplement thereto.  Permission is hereby
- *     granted, free of charge, to any person obtaining a copy of this software
- *     annotated with this license and the Software, to deal in the Software
- *     without restriction, including without limitation the rights to use,
- *     copy, modify, merge, publish, distribute, sublicense, and/or sell copies
- *     of the Software, and to permit persons to whom the Software is furnished
- *     to do so, subject to the following conditions:
- *
- *     The above copyright notice and this permission notice shall be included
- *     in all copies or substantial portions of the Software.
- *
- *     THIS SOFTWARE IS BEING DISTRIBUTED BY SYNOPSYS SOLELY ON AN "AS IS"
- *     BASIS AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED
- *     TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A
- *     PARTICULAR PURPOSE ARE HEREBY DISCLAIMED. IN NO EVENT SHALL SYNOPSYS
- *     BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR
- *     CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF
- *     SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS
- *     INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN
- *     CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE)
- *     ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF
- *     THE POSSIBILITY OF SUCH DAMAGE.
- *
- *
- * License 2: Modified BSD
- *
- * Copyright (c) 2014-2016 Advanced Micro Devices, Inc.
- * All rights reserved.
- *
- * Redistribution and use in source and binary forms, with or without
- * modification, are permitted provided that the following conditions are met:
- *     * Redistributions of source code must retain the above copyright
- *       notice, this list of conditions and the following disclaimer.
- *     * Redistributions in binary form must reproduce the above copyright
- *       notice, this list of conditions and the following disclaimer in the
- *       documentation and/or other materials provided with the distribution.
- *     * Neither the name of Advanced Micro Devices, Inc. nor the
- *       names of its contributors may be used to endorse or promote products
- *       derived from this software without specific prior written permission.
- *
- * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS"
- * AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
- * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
- * ARE DISCLAIMED. IN NO EVENT SHALL <COPYRIGHT HOLDER> BE LIABLE FOR ANY
- * DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES
- * (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES;
- * LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND
- * ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
- * (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF
- * THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
- *
- * This file incorporates work covered by the following copyright and
- * permission notice:
- *     The Synopsys DWC ETHER XGMAC Software Driver and documentation
- *     (hereinafter "Software") is an unsupported proprietary work of Synopsys,
- *     Inc. unless otherwise expressly agreed to in writing between Synopsys
- *     and you.
- *
- *     The Software IS NOT an item of Licensed Software or Licensed Product
- *     under any End User Software License Agreement or Agreement for Licensed
- *     Product with Synopsys or any supplement thereto.  Permission is hereby
- *     granted, free of charge, to any person obtaining a copy of this software
- *     annotated with this license and the Software, to deal in the Software
- *     without restriction, including without limitation the rights to use,
- *     copy, modify, merge, publish, distribute, sublicense, and/or sell copies
- *     of the Software, and to permit persons to whom the Software is furnished
- *     to do so, subject to the following conditions:
- *
- *     The above copyright notice and this permission notice shall be included
- *     in all copies or substantial portions of the Software.
- *
- *     THIS SOFTWARE IS BEING DISTRIBUTED BY SYNOPSYS SOLELY ON AN "AS IS"
- *     BASIS AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED
- *     TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A
- *     PARTICULAR PURPOSE ARE HEREBY DISCLAIMED. IN NO EVENT SHALL SYNOPSYS
- *     BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR
- *     CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF
- *     SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS
- *     INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN
- *     CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE)
- *     ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF
- *     THE POSSIBILITY OF SUCH DAMAGE.
+ * Copyright (c) 2014-2025, Advanced Micro Devices, Inc.
+ * Copyright (c) 2014, Synopsys, Inc.
+ * All rights reserved
  */
 
 #include <linux/interrupt.h>
@@ -126,6 +17,24 @@
 #include "xgbe.h"
 #include "xgbe-common.h"
 
+static int xgbe_phy_module_eeprom(struct xgbe_prv_data *pdata,
+				  struct ethtool_eeprom *eeprom, u8 *data)
+{
+	if (!pdata->phy_if.phy_impl.module_eeprom)
+		return -ENXIO;
+
+	return pdata->phy_if.phy_impl.module_eeprom(pdata, eeprom, data);
+}
+
+static int xgbe_phy_module_info(struct xgbe_prv_data *pdata,
+				struct ethtool_modinfo *modinfo)
+{
+	if (!pdata->phy_if.phy_impl.module_info)
+		return -ENXIO;
+
+	return pdata->phy_if.phy_impl.module_info(pdata, modinfo);
+}
+
 static void xgbe_an37_clear_interrupts(struct xgbe_prv_data *pdata)
 {
 	int reg;
@@ -198,31 +107,8 @@ static void xgbe_an_clear_interrupts_all(struct xgbe_prv_data *pdata)
 	xgbe_an37_clear_interrupts(pdata);
 }
 
-static void xgbe_an73_enable_kr_training(struct xgbe_prv_data *pdata)
-{
-	unsigned int reg;
-
-	reg = XMDIO_READ(pdata, MDIO_MMD_PMAPMD, MDIO_PMA_10GBR_PMD_CTRL);
-
-	reg |= XGBE_KR_TRAINING_ENABLE;
-	XMDIO_WRITE(pdata, MDIO_MMD_PMAPMD, MDIO_PMA_10GBR_PMD_CTRL, reg);
-}
-
-static void xgbe_an73_disable_kr_training(struct xgbe_prv_data *pdata)
-{
-	unsigned int reg;
-
-	reg = XMDIO_READ(pdata, MDIO_MMD_PMAPMD, MDIO_PMA_10GBR_PMD_CTRL);
-
-	reg &= ~XGBE_KR_TRAINING_ENABLE;
-	XMDIO_WRITE(pdata, MDIO_MMD_PMAPMD, MDIO_PMA_10GBR_PMD_CTRL, reg);
-}
-
 static void xgbe_kr_mode(struct xgbe_prv_data *pdata)
 {
-	/* Enable KR training */
-	xgbe_an73_enable_kr_training(pdata);
-
 	/* Set MAC to 10G speed */
 	pdata->hw_if.set_speed(pdata, SPEED_10000);
 
@@ -232,9 +118,6 @@ static void xgbe_kr_mode(struct xgbe_prv_data *pdata)
 
 static void xgbe_kx_2500_mode(struct xgbe_prv_data *pdata)
 {
-	/* Disable KR training */
-	xgbe_an73_disable_kr_training(pdata);
-
 	/* Set MAC to 2.5G speed */
 	pdata->hw_if.set_speed(pdata, SPEED_2500);
 
@@ -244,9 +127,6 @@ static void xgbe_kx_2500_mode(struct xgbe_prv_data *pdata)
 
 static void xgbe_kx_1000_mode(struct xgbe_prv_data *pdata)
 {
-	/* Disable KR training */
-	xgbe_an73_disable_kr_training(pdata);
-
 	/* Set MAC to 1G speed */
 	pdata->hw_if.set_speed(pdata, SPEED_1000);
 
@@ -260,9 +140,6 @@ static void xgbe_sfi_mode(struct xgbe_prv_data *pdata)
 	if (pdata->kr_redrv)
 		return xgbe_kr_mode(pdata);
 
-	/* Disable KR training */
-	xgbe_an73_disable_kr_training(pdata);
-
 	/* Set MAC to 10G speed */
 	pdata->hw_if.set_speed(pdata, SPEED_10000);
 
@@ -272,9 +149,6 @@ static void xgbe_sfi_mode(struct xgbe_prv_data *pdata)
 
 static void xgbe_x_mode(struct xgbe_prv_data *pdata)
 {
-	/* Disable KR training */
-	xgbe_an73_disable_kr_training(pdata);
-
 	/* Set MAC to 1G speed */
 	pdata->hw_if.set_speed(pdata, SPEED_1000);
 
@@ -284,9 +158,6 @@ static void xgbe_x_mode(struct xgbe_prv_data *pdata)
 
 static void xgbe_sgmii_1000_mode(struct xgbe_prv_data *pdata)
 {
-	/* Disable KR training */
-	xgbe_an73_disable_kr_training(pdata);
-
 	/* Set MAC to 1G speed */
 	pdata->hw_if.set_speed(pdata, SPEED_1000);
 
@@ -294,11 +165,17 @@ static void xgbe_sgmii_1000_mode(struct xgbe_prv_data *pdata)
 	pdata->phy_if.phy_impl.set_mode(pdata, XGBE_MODE_SGMII_1000);
 }
 
-static void xgbe_sgmii_100_mode(struct xgbe_prv_data *pdata)
+static void xgbe_sgmii_10_mode(struct xgbe_prv_data *pdata)
 {
-	/* Disable KR training */
-	xgbe_an73_disable_kr_training(pdata);
+	/* Set MAC to 10M speed */
+	pdata->hw_if.set_speed(pdata, SPEED_10);
+
+	/* Call PHY implementation support to complete rate change */
+	pdata->phy_if.phy_impl.set_mode(pdata, XGBE_MODE_SGMII_10);
+}
 
+static void xgbe_sgmii_100_mode(struct xgbe_prv_data *pdata)
+{
 	/* Set MAC to 1G speed */
 	pdata->hw_if.set_speed(pdata, SPEED_1000);
 
@@ -329,6 +206,9 @@ static void xgbe_change_mode(struct xgbe_prv_data *pdata,
 	case XGBE_MODE_KR:
 		xgbe_kr_mode(pdata);
 		break;
+	case XGBE_MODE_SGMII_10:
+		xgbe_sgmii_10_mode(pdata);
+		break;
 	case XGBE_MODE_SGMII_100:
 		xgbe_sgmii_100_mode(pdata);
 		break;
@@ -354,13 +234,15 @@ static void xgbe_switch_mode(struct xgbe_prv_data *pdata)
 	xgbe_change_mode(pdata, pdata->phy_if.phy_impl.switch_mode(pdata));
 }
 
-static void xgbe_set_mode(struct xgbe_prv_data *pdata,
+static bool xgbe_set_mode(struct xgbe_prv_data *pdata,
 			  enum xgbe_mode mode)
 {
 	if (mode == xgbe_cur_mode(pdata))
-		return;
+		return false;
 
 	xgbe_change_mode(pdata, mode);
+
+	return true;
 }
 
 static bool xgbe_use_mode(struct xgbe_prv_data *pdata,
@@ -384,6 +266,10 @@ static void xgbe_an37_set(struct xgbe_prv_data *pdata, bool enable,
 		reg |= MDIO_VEND2_CTRL1_AN_RESTART;
 
 	XMDIO_WRITE(pdata, MDIO_MMD_VEND2, MDIO_CTRL1, reg);
+
+	reg = XMDIO_READ(pdata, MDIO_MMD_VEND2, MDIO_PCS_DIG_CTRL);
+	reg |= XGBE_VEND2_MAC_AUTO_SW;
+	XMDIO_WRITE(pdata, MDIO_MMD_VEND2, MDIO_PCS_DIG_CTRL, reg);
 }
 
 static void xgbe_an37_restart(struct xgbe_prv_data *pdata)
@@ -407,6 +293,12 @@ static void xgbe_an73_set(struct xgbe_prv_data *pdata, bool enable,
 {
 	unsigned int reg;
 
+	/* Disable KR training for now */
+	reg = XMDIO_READ(pdata, MDIO_MMD_PMAPMD, MDIO_PMA_10GBR_PMD_CTRL);
+	reg &= ~XGBE_KR_TRAINING_ENABLE;
+	XMDIO_WRITE(pdata, MDIO_MMD_PMAPMD, MDIO_PMA_10GBR_PMD_CTRL, reg);
+
+	/* Update AN settings */
 	reg = XMDIO_READ(pdata, MDIO_MMD_AN, MDIO_CTRL1);
 	reg &= ~MDIO_AN_CTRL1_ENABLE;
 
@@ -432,11 +324,16 @@ static void xgbe_an73_disable(struct xgbe_prv_data *pdata)
 	xgbe_an73_set(pdata, false, false);
 	xgbe_an73_disable_interrupts(pdata);
 
+	pdata->an_start = 0;
+
 	netif_dbg(pdata, link, pdata->netdev, "CL73 AN disabled\n");
 }
 
 static void xgbe_an_restart(struct xgbe_prv_data *pdata)
 {
+	if (pdata->phy_if.phy_impl.an_pre)
+		pdata->phy_if.phy_impl.an_pre(pdata);
+
 	switch (pdata->an_mode) {
 	case XGBE_AN_MODE_CL73:
 	case XGBE_AN_MODE_CL73_REDRV:
@@ -453,6 +350,9 @@ static void xgbe_an_restart(struct xgbe_prv_data *pdata)
 
 static void xgbe_an_disable(struct xgbe_prv_data *pdata)
 {
+	if (pdata->phy_if.phy_impl.an_post)
+		pdata->phy_if.phy_impl.an_post(pdata);
+
 	switch (pdata->an_mode) {
 	case XGBE_AN_MODE_CL73:
 	case XGBE_AN_MODE_CL73_REDRV:
@@ -496,21 +396,20 @@ static enum xgbe_an xgbe_an73_tx_training(struct xgbe_prv_data *pdata,
 	XMDIO_WRITE(pdata, MDIO_MMD_PMAPMD, MDIO_PMA_10GBR_FECCTRL, reg);
 
 	/* Start KR training */
+	if (pdata->phy_if.phy_impl.kr_training_pre)
+		pdata->phy_if.phy_impl.kr_training_pre(pdata);
+
 	reg = XMDIO_READ(pdata, MDIO_MMD_PMAPMD, MDIO_PMA_10GBR_PMD_CTRL);
-	if (reg & XGBE_KR_TRAINING_ENABLE) {
-		if (pdata->phy_if.phy_impl.kr_training_pre)
-			pdata->phy_if.phy_impl.kr_training_pre(pdata);
+	reg |= XGBE_KR_TRAINING_ENABLE;
+	reg |= XGBE_KR_TRAINING_START;
+	XMDIO_WRITE(pdata, MDIO_MMD_PMAPMD, MDIO_PMA_10GBR_PMD_CTRL, reg);
+	pdata->kr_start_time = jiffies;
 
-		reg |= XGBE_KR_TRAINING_START;
-		XMDIO_WRITE(pdata, MDIO_MMD_PMAPMD, MDIO_PMA_10GBR_PMD_CTRL,
-			    reg);
+	netif_dbg(pdata, link, pdata->netdev,
+		  "KR training initiated\n");
 
-		if (pdata->phy_if.phy_impl.kr_training_post)
-			pdata->phy_if.phy_impl.kr_training_post(pdata);
-
-		netif_dbg(pdata, link, pdata->netdev,
-			  "KR training initiated\n");
-	}
+	if (pdata->phy_if.phy_impl.kr_training_post)
+		pdata->phy_if.phy_impl.kr_training_post(pdata);
 
 	return XGBE_AN_PAGE_RECEIVED;
 }
@@ -615,12 +514,14 @@ static enum xgbe_an xgbe_an73_page_received(struct xgbe_prv_data *pdata)
 
 static enum xgbe_an xgbe_an73_incompat_link(struct xgbe_prv_data *pdata)
 {
+	struct ethtool_link_ksettings *lks = &pdata->phy.lks;
+
 	/* Be sure we aren't looping trying to negotiate */
 	if (xgbe_in_kr_mode(pdata)) {
 		pdata->kr_state = XGBE_RX_ERROR;
 
-		if (!(pdata->phy.advertising & ADVERTISED_1000baseKX_Full) &&
-		    !(pdata->phy.advertising & ADVERTISED_2500baseX_Full))
+		if (!XGBE_ADV(lks, 1000baseKX_Full) &&
+		    !XGBE_ADV(lks, 2500baseX_Full))
 			return XGBE_AN_NO_LINK;
 
 		if (pdata->kx_state != XGBE_RX_BPA)
@@ -628,18 +529,20 @@ static enum xgbe_an xgbe_an73_incompat_link(struct xgbe_prv_data *pdata)
 	} else {
 		pdata->kx_state = XGBE_RX_ERROR;
 
-		if (!(pdata->phy.advertising & ADVERTISED_10000baseKR_Full))
+		if (!XGBE_ADV(lks, 10000baseKR_Full))
 			return XGBE_AN_NO_LINK;
 
 		if (pdata->kr_state != XGBE_RX_BPA)
 			return XGBE_AN_NO_LINK;
 	}
 
-	xgbe_an73_disable(pdata);
+	xgbe_an_disable(pdata);
 
 	xgbe_switch_mode(pdata);
 
-	xgbe_an73_restart(pdata);
+	pdata->an_result = XGBE_AN_READY;
+
+	xgbe_an_restart(pdata);
 
 	return XGBE_AN_INCOMPAT_LINK;
 }
@@ -665,6 +568,10 @@ static void xgbe_an37_isr(struct xgbe_prv_data *pdata)
 	} else {
 		/* Enable AN interrupts */
 		xgbe_an37_enable_interrupts(pdata);
+
+		/* Reissue interrupt if status is not clear */
+		if (pdata->vdata->irq_reissue_support)
+			XP_IOWRITE(pdata, XP_INT_REISSUE_EN, 1 << 3);
 	}
 }
 
@@ -684,12 +591,16 @@ static void xgbe_an73_isr(struct xgbe_prv_data *pdata)
 	} else {
 		/* Enable AN interrupts */
 		xgbe_an73_enable_interrupts(pdata);
+
+		/* Reissue interrupt if status is not clear */
+		if (pdata->vdata->irq_reissue_support)
+			XP_IOWRITE(pdata, XP_INT_REISSUE_EN, 1 << 3);
 	}
 }
 
-static irqreturn_t xgbe_an_isr(int irq, void *data)
+static void xgbe_an_isr_bh_work(struct work_struct *work)
 {
-	struct xgbe_prv_data *pdata = (struct xgbe_prv_data *)data;
+	struct xgbe_prv_data *pdata = from_work(pdata, work, an_bh_work);
 
 	netif_dbg(pdata, intr, pdata->netdev, "AN interrupt received\n");
 
@@ -705,13 +616,25 @@ static irqreturn_t xgbe_an_isr(int irq, void *data)
 	default:
 		break;
 	}
+}
+
+static irqreturn_t xgbe_an_isr(int irq, void *data)
+{
+	struct xgbe_prv_data *pdata = (struct xgbe_prv_data *)data;
+
+	if (pdata->isr_as_bh_work)
+		queue_work(system_bh_wq, &pdata->an_bh_work);
+	else
+		xgbe_an_isr_bh_work(&pdata->an_bh_work);
 
 	return IRQ_HANDLED;
 }
 
-static irqreturn_t xgbe_an_combined_isr(int irq, struct xgbe_prv_data *pdata)
+static irqreturn_t xgbe_an_combined_isr(struct xgbe_prv_data *pdata)
 {
-	return xgbe_an_isr(irq, pdata);
+	xgbe_an_isr_bh_work(&pdata->an_bh_work);
+
+	return IRQ_HANDLED;
 }
 
 static void xgbe_an_irq_work(struct work_struct *work)
@@ -798,6 +721,9 @@ static void xgbe_an37_state_machine(struct xgbe_prv_data *pdata)
 		pdata->an_result = pdata->an_state;
 		pdata->an_state = XGBE_AN_READY;
 
+		if (pdata->phy_if.phy_impl.an_post)
+			pdata->phy_if.phy_impl.an_post(pdata);
+
 		netif_dbg(pdata, link, pdata->netdev, "CL37 AN result: %s\n",
 			  xgbe_state_as_string(pdata->an_result));
 	}
@@ -881,6 +807,9 @@ again:
 		pdata->kx_state = XGBE_RX_BPA;
 		pdata->an_start = 0;
 
+		if (pdata->phy_if.phy_impl.an_post)
+			pdata->phy_if.phy_impl.an_post(pdata);
+
 		netif_dbg(pdata, link, pdata->netdev, "CL73 AN result: %s\n",
 			  xgbe_state_as_string(pdata->an_result));
 	}
@@ -915,23 +844,28 @@ static void xgbe_an_state_machine(struct work_struct *work)
 		break;
 	}
 
+	/* Reissue interrupt if status is not clear */
+	if (pdata->vdata->irq_reissue_support)
+		XP_IOWRITE(pdata, XP_INT_REISSUE_EN, 1 << 3);
+
 	mutex_unlock(&pdata->an_mutex);
 }
 
 static void xgbe_an37_init(struct xgbe_prv_data *pdata)
 {
-	unsigned int advertising, reg;
+	struct ethtool_link_ksettings lks;
+	unsigned int reg;
 
-	advertising = pdata->phy_if.phy_impl.an_advertising(pdata);
+	pdata->phy_if.phy_impl.an_advertising(pdata, &lks);
 
 	/* Set up Advertisement register */
 	reg = XMDIO_READ(pdata, MDIO_MMD_VEND2, MDIO_VEND2_AN_ADVERTISE);
-	if (advertising & ADVERTISED_Pause)
+	if (XGBE_ADV(&lks, Pause))
 		reg |= 0x100;
 	else
 		reg &= ~0x100;
 
-	if (advertising & ADVERTISED_Asym_Pause)
+	if (XGBE_ADV(&lks, Asym_Pause))
 		reg |= 0x80;
 	else
 		reg &= ~0x80;
@@ -958,21 +892,29 @@ static void xgbe_an37_init(struct xgbe_prv_data *pdata)
 		break;
 	}
 
+	reg |= XGBE_AN_CL37_MII_CTRL_8BIT;
+
 	XMDIO_WRITE(pdata, MDIO_MMD_VEND2, MDIO_VEND2_AN_CTRL, reg);
 
 	netif_dbg(pdata, link, pdata->netdev, "CL37 AN (%s) initialized\n",
 		  (pdata->an_mode == XGBE_AN_MODE_CL37) ? "BaseX" : "SGMII");
+
+	reg = XMDIO_READ(pdata, MDIO_MMD_AN, MDIO_CTRL1);
+	reg &= ~MDIO_AN_CTRL1_ENABLE;
+	XMDIO_WRITE(pdata, MDIO_MMD_AN, MDIO_CTRL1, reg);
+
 }
 
 static void xgbe_an73_init(struct xgbe_prv_data *pdata)
 {
-	unsigned int advertising, reg;
+	struct ethtool_link_ksettings lks;
+	unsigned int reg;
 
-	advertising = pdata->phy_if.phy_impl.an_advertising(pdata);
+	pdata->phy_if.phy_impl.an_advertising(pdata, &lks);
 
 	/* Set up Advertisement register 3 first */
 	reg = XMDIO_READ(pdata, MDIO_MMD_AN, MDIO_AN_ADVERTISE + 2);
-	if (advertising & ADVERTISED_10000baseR_FEC)
+	if (XGBE_ADV(&lks, 10000baseR_FEC))
 		reg |= 0xc000;
 	else
 		reg &= ~0xc000;
@@ -981,13 +923,13 @@ static void xgbe_an73_init(struct xgbe_prv_data *pdata)
 
 	/* Set up Advertisement register 2 next */
 	reg = XMDIO_READ(pdata, MDIO_MMD_AN, MDIO_AN_ADVERTISE + 1);
-	if (advertising & ADVERTISED_10000baseKR_Full)
+	if (XGBE_ADV(&lks, 10000baseKR_Full))
 		reg |= 0x80;
 	else
 		reg &= ~0x80;
 
-	if ((advertising & ADVERTISED_1000baseKX_Full) ||
-	    (advertising & ADVERTISED_2500baseX_Full))
+	if (XGBE_ADV(&lks, 1000baseKX_Full) ||
+	    XGBE_ADV(&lks, 2500baseX_Full))
 		reg |= 0x20;
 	else
 		reg &= ~0x20;
@@ -996,12 +938,12 @@ static void xgbe_an73_init(struct xgbe_prv_data *pdata)
 
 	/* Set up Advertisement register 1 last */
 	reg = XMDIO_READ(pdata, MDIO_MMD_AN, MDIO_AN_ADVERTISE);
-	if (advertising & ADVERTISED_Pause)
+	if (XGBE_ADV(&lks, Pause))
 		reg |= 0x400;
 	else
 		reg &= ~0x400;
 
-	if (advertising & ADVERTISED_Asym_Pause)
+	if (XGBE_ADV(&lks, Asym_Pause))
 		reg |= 0x800;
 	else
 		reg &= ~0x800;
@@ -1047,6 +989,8 @@ static const char *xgbe_phy_fc_string(struct xgbe_prv_data *pdata)
 static const char *xgbe_phy_speed_string(int speed)
 {
 	switch (speed) {
+	case SPEED_10:
+		return "10Mbps";
 	case SPEED_100:
 		return "100Mbps";
 	case SPEED_1000:
@@ -1084,14 +1028,14 @@ static void xgbe_phy_adjust_link(struct xgbe_prv_data *pdata)
 
 		if (pdata->tx_pause != pdata->phy.tx_pause) {
 			new_state = 1;
-			pdata->hw_if.config_tx_flow_control(pdata);
 			pdata->tx_pause = pdata->phy.tx_pause;
+			pdata->hw_if.config_tx_flow_control(pdata);
 		}
 
 		if (pdata->rx_pause != pdata->phy.rx_pause) {
 			new_state = 1;
-			pdata->hw_if.config_rx_flow_control(pdata);
 			pdata->rx_pause = pdata->phy.rx_pause;
+			pdata->hw_if.config_rx_flow_control(pdata);
 		}
 
 		/* Speed support */
@@ -1134,6 +1078,7 @@ static int xgbe_phy_config_fixed(struct xgbe_prv_data *pdata)
 	case XGBE_MODE_KX_1000:
 	case XGBE_MODE_KX_2500:
 	case XGBE_MODE_KR:
+	case XGBE_MODE_SGMII_10:
 	case XGBE_MODE_SGMII_100:
 	case XGBE_MODE_SGMII_1000:
 	case XGBE_MODE_X:
@@ -1148,26 +1093,40 @@ static int xgbe_phy_config_fixed(struct xgbe_prv_data *pdata)
 	if (pdata->phy.duplex != DUPLEX_FULL)
 		return -EINVAL;
 
-	xgbe_set_mode(pdata, mode);
+	/* Force the mode change for SFI in Fixed PHY config.
+	 * Fixed PHY configs needs PLL to be enabled while doing mode set.
+	 * When the SFP module isn't connected during boot, driver assumes
+	 * AN is ON and attempts autonegotiation. However, if the connected
+	 * SFP comes up in Fixed PHY config, the link will not come up as
+	 * PLL isn't enabled while the initial mode set command is issued.
+	 * So, force the mode change for SFI in Fixed PHY configuration to
+	 * fix link issues.
+	 */
+	if (mode == XGBE_MODE_SFI)
+		xgbe_change_mode(pdata, mode);
+	else
+		xgbe_set_mode(pdata, mode);
 
 	return 0;
 }
 
-static int __xgbe_phy_config_aneg(struct xgbe_prv_data *pdata)
+static int __xgbe_phy_config_aneg(struct xgbe_prv_data *pdata, bool set_mode)
 {
 	int ret;
 
+	mutex_lock(&pdata->an_mutex);
+
 	set_bit(XGBE_LINK_INIT, &pdata->dev_state);
 	pdata->link_check = jiffies;
 
 	ret = pdata->phy_if.phy_impl.an_config(pdata);
 	if (ret)
-		return ret;
+		goto out;
 
 	if (pdata->phy.autoneg != AUTONEG_ENABLE) {
 		ret = xgbe_phy_config_fixed(pdata);
 		if (ret || !pdata->kr_redrv)
-			return ret;
+			goto out;
 
 		netif_dbg(pdata, link, pdata->netdev, "AN redriver support\n");
 	} else {
@@ -1177,24 +1136,29 @@ static int __xgbe_phy_config_aneg(struct xgbe_prv_data *pdata)
 	/* Disable auto-negotiation interrupt */
 	disable_irq(pdata->an_irq);
 
-	/* Start auto-negotiation in a supported mode */
-	if (xgbe_use_mode(pdata, XGBE_MODE_KR)) {
-		xgbe_set_mode(pdata, XGBE_MODE_KR);
-	} else if (xgbe_use_mode(pdata, XGBE_MODE_KX_2500)) {
-		xgbe_set_mode(pdata, XGBE_MODE_KX_2500);
-	} else if (xgbe_use_mode(pdata, XGBE_MODE_KX_1000)) {
-		xgbe_set_mode(pdata, XGBE_MODE_KX_1000);
-	} else if (xgbe_use_mode(pdata, XGBE_MODE_SFI)) {
-		xgbe_set_mode(pdata, XGBE_MODE_SFI);
-	} else if (xgbe_use_mode(pdata, XGBE_MODE_X)) {
-		xgbe_set_mode(pdata, XGBE_MODE_X);
-	} else if (xgbe_use_mode(pdata, XGBE_MODE_SGMII_1000)) {
-		xgbe_set_mode(pdata, XGBE_MODE_SGMII_1000);
-	} else if (xgbe_use_mode(pdata, XGBE_MODE_SGMII_100)) {
-		xgbe_set_mode(pdata, XGBE_MODE_SGMII_100);
-	} else {
-		enable_irq(pdata->an_irq);
-		return -EINVAL;
+	if (set_mode) {
+		/* Start auto-negotiation in a supported mode */
+		if (xgbe_use_mode(pdata, XGBE_MODE_KR)) {
+			xgbe_set_mode(pdata, XGBE_MODE_KR);
+		} else if (xgbe_use_mode(pdata, XGBE_MODE_KX_2500)) {
+			xgbe_set_mode(pdata, XGBE_MODE_KX_2500);
+		} else if (xgbe_use_mode(pdata, XGBE_MODE_KX_1000)) {
+			xgbe_set_mode(pdata, XGBE_MODE_KX_1000);
+		} else if (xgbe_use_mode(pdata, XGBE_MODE_SFI)) {
+			xgbe_set_mode(pdata, XGBE_MODE_SFI);
+		} else if (xgbe_use_mode(pdata, XGBE_MODE_X)) {
+			xgbe_set_mode(pdata, XGBE_MODE_X);
+		} else if (xgbe_use_mode(pdata, XGBE_MODE_SGMII_1000)) {
+			xgbe_set_mode(pdata, XGBE_MODE_SGMII_1000);
+		} else if (xgbe_use_mode(pdata, XGBE_MODE_SGMII_100)) {
+			xgbe_set_mode(pdata, XGBE_MODE_SGMII_100);
+		} else if (xgbe_use_mode(pdata, XGBE_MODE_SGMII_10)) {
+			xgbe_set_mode(pdata, XGBE_MODE_SGMII_10);
+		} else {
+			enable_irq(pdata->an_irq);
+			ret = -EINVAL;
+			goto out;
+		}
 	}
 
 	/* Disable and stop any in progress auto-negotiation */
@@ -1214,16 +1178,7 @@ static int __xgbe_phy_config_aneg(struct xgbe_prv_data *pdata)
 	xgbe_an_init(pdata);
 	xgbe_an_restart(pdata);
 
-	return 0;
-}
-
-static int xgbe_phy_config_aneg(struct xgbe_prv_data *pdata)
-{
-	int ret;
-
-	mutex_lock(&pdata->an_mutex);
-
-	ret = __xgbe_phy_config_aneg(pdata);
+out:
 	if (ret)
 		set_bit(XGBE_LINK_ERR, &pdata->dev_state);
 	else
@@ -1234,6 +1189,16 @@ static int xgbe_phy_config_aneg(struct xgbe_prv_data *pdata)
 	return ret;
 }
 
+static int xgbe_phy_config_aneg(struct xgbe_prv_data *pdata)
+{
+	return __xgbe_phy_config_aneg(pdata, true);
+}
+
+static int xgbe_phy_reconfig_aneg(struct xgbe_prv_data *pdata)
+{
+	return __xgbe_phy_config_aneg(pdata, false);
+}
+
 static bool xgbe_phy_aneg_done(struct xgbe_prv_data *pdata)
 {
 	return (pdata->an_result == XGBE_AN_COMPLETE);
@@ -1242,9 +1207,30 @@ static bool xgbe_phy_aneg_done(struct xgbe_prv_data *pdata)
 static void xgbe_check_link_timeout(struct xgbe_prv_data *pdata)
 {
 	unsigned long link_timeout;
+	unsigned long kr_time;
+	int wait;
 
 	link_timeout = pdata->link_check + (XGBE_LINK_TIMEOUT * HZ);
 	if (time_after(jiffies, link_timeout)) {
+		if ((xgbe_cur_mode(pdata) == XGBE_MODE_KR) &&
+		    pdata->phy.autoneg == AUTONEG_ENABLE) {
+			/* AN restart should not happen while KR training is in progress.
+			 * The while loop ensures no AN restart during KR training,
+			 * waits up to 500ms and AN restart is triggered only if KR
+			 * training is failed.
+			 */
+			wait = XGBE_KR_TRAINING_WAIT_ITER;
+			while (wait--) {
+				kr_time = pdata->kr_start_time +
+					  msecs_to_jiffies(XGBE_AN_MS_TIMEOUT);
+				if (time_after(jiffies, kr_time))
+					break;
+				/* AN restart is not required, if AN result is COMPLETE */
+				if (pdata->an_result == XGBE_AN_COMPLETE)
+					return;
+				usleep_range(10000, 11000);
+			}
+		}
 		netif_dbg(pdata, link, pdata->netdev, "AN link timeout\n");
 		xgbe_phy_config_aneg(pdata);
 	}
@@ -1255,11 +1241,12 @@ static enum xgbe_mode xgbe_phy_status_aneg(struct xgbe_prv_data *pdata)
 	return pdata->phy_if.phy_impl.an_outcome(pdata);
 }
 
-static void xgbe_phy_status_result(struct xgbe_prv_data *pdata)
+static bool xgbe_phy_status_result(struct xgbe_prv_data *pdata)
 {
+	struct ethtool_link_ksettings *lks = &pdata->phy.lks;
 	enum xgbe_mode mode;
 
-	pdata->phy.lp_advertising = 0;
+	XGBE_ZERO_LP_ADV(lks);
 
 	if ((pdata->phy.autoneg != AUTONEG_ENABLE) || pdata->parallel_detect)
 		mode = xgbe_cur_mode(pdata);
@@ -1267,6 +1254,9 @@ static void xgbe_phy_status_result(struct xgbe_prv_data *pdata)
 		mode = xgbe_phy_status_aneg(pdata);
 
 	switch (mode) {
+	case XGBE_MODE_SGMII_10:
+		pdata->phy.speed = SPEED_10;
+		break;
 	case XGBE_MODE_SGMII_100:
 		pdata->phy.speed = SPEED_100;
 		break;
@@ -1289,7 +1279,13 @@ static void xgbe_phy_status_result(struct xgbe_prv_data *pdata)
 
 	pdata->phy.duplex = DUPLEX_FULL;
 
-	xgbe_set_mode(pdata, mode);
+	if (!xgbe_set_mode(pdata, mode))
+		return false;
+
+	if (pdata->an_again)
+		xgbe_phy_reconfig_aneg(pdata);
+
+	return true;
 }
 
 static void xgbe_phy_status(struct xgbe_prv_data *pdata)
@@ -1308,9 +1304,13 @@ static void xgbe_phy_status(struct xgbe_prv_data *pdata)
 
 	pdata->phy.link = pdata->phy_if.phy_impl.link_status(pdata,
 							     &an_restart);
+	/* bail out if the link status register read fails */
+	if (pdata->phy.link < 0)
+		return;
+
 	if (an_restart) {
 		xgbe_phy_config_aneg(pdata);
-		return;
+		goto adjust_link;
 	}
 
 	if (pdata->phy.link) {
@@ -1319,7 +1319,8 @@ static void xgbe_phy_status(struct xgbe_prv_data *pdata)
 			return;
 		}
 
-		xgbe_phy_status_result(pdata);
+		if (xgbe_phy_status_result(pdata))
+			return;
 
 		if (test_bit(XGBE_LINK_INIT, &pdata->dev_state))
 			clear_bit(XGBE_LINK_INIT, &pdata->dev_state);
@@ -1355,13 +1356,14 @@ static void xgbe_phy_stop(struct xgbe_prv_data *pdata)
 	/* Disable auto-negotiation */
 	xgbe_an_disable_all(pdata);
 
-	if (pdata->dev_irq != pdata->an_irq)
+	if (pdata->dev_irq != pdata->an_irq) {
 		devm_free_irq(pdata->dev, pdata->an_irq, pdata);
+		cancel_work_sync(&pdata->an_bh_work);
+	}
 
 	pdata->phy_if.phy_impl.stop(pdata);
 
 	pdata->phy.link = 0;
-	netif_carrier_off(pdata->netdev);
 
 	xgbe_phy_adjust_link(pdata);
 }
@@ -1379,6 +1381,8 @@ static int xgbe_phy_start(struct xgbe_prv_data *pdata)
 
 	/* If we have a separate AN irq, enable it */
 	if (pdata->dev_irq != pdata->an_irq) {
+		INIT_WORK(&pdata->an_bh_work, xgbe_an_isr_bh_work);
+
 		ret = devm_request_irq(pdata->dev, pdata->an_irq,
 				       xgbe_an_isr, 0, pdata->an_name,
 				       pdata);
@@ -1405,6 +1409,8 @@ static int xgbe_phy_start(struct xgbe_prv_data *pdata)
 		xgbe_sgmii_1000_mode(pdata);
 	} else if (xgbe_use_mode(pdata, XGBE_MODE_SGMII_100)) {
 		xgbe_sgmii_100_mode(pdata);
+	} else if (xgbe_use_mode(pdata, XGBE_MODE_SGMII_10)) {
+		xgbe_sgmii_10_mode(pdata);
 	} else {
 		ret = -EINVAL;
 		goto err_irq;
@@ -1486,31 +1492,36 @@ static void xgbe_dump_phy_registers(struct xgbe_prv_data *pdata)
 
 static int xgbe_phy_best_advertised_speed(struct xgbe_prv_data *pdata)
 {
-	if (pdata->phy.advertising & ADVERTISED_10000baseKR_Full)
+	struct ethtool_link_ksettings *lks = &pdata->phy.lks;
+
+	if (XGBE_ADV(lks, 10000baseKR_Full))
 		return SPEED_10000;
-	else if (pdata->phy.advertising & ADVERTISED_10000baseT_Full)
+	else if (XGBE_ADV(lks, 10000baseT_Full))
 		return SPEED_10000;
-	else if (pdata->phy.advertising & ADVERTISED_2500baseX_Full)
+	else if (XGBE_ADV(lks, 2500baseX_Full))
+		return SPEED_2500;
+	else if (XGBE_ADV(lks, 2500baseT_Full))
 		return SPEED_2500;
-	else if (pdata->phy.advertising & ADVERTISED_1000baseKX_Full)
+	else if (XGBE_ADV(lks, 1000baseKX_Full))
 		return SPEED_1000;
-	else if (pdata->phy.advertising & ADVERTISED_1000baseT_Full)
+	else if (XGBE_ADV(lks, 1000baseT_Full))
 		return SPEED_1000;
-	else if (pdata->phy.advertising & ADVERTISED_100baseT_Full)
+	else if (XGBE_ADV(lks, 100baseT_Full))
 		return SPEED_100;
+	else if (XGBE_ADV(lks, 10baseT_Full))
+		return SPEED_10;
 
 	return SPEED_UNKNOWN;
 }
 
 static void xgbe_phy_exit(struct xgbe_prv_data *pdata)
 {
-	xgbe_phy_stop(pdata);
-
 	pdata->phy_if.phy_impl.exit(pdata);
 }
 
 static int xgbe_phy_init(struct xgbe_prv_data *pdata)
 {
+	struct ethtool_link_ksettings *lks = &pdata->phy.lks;
 	int ret;
 
 	mutex_init(&pdata->an_mutex);
@@ -1528,11 +1539,13 @@ static int xgbe_phy_init(struct xgbe_prv_data *pdata)
 	ret = pdata->phy_if.phy_impl.init(pdata);
 	if (ret)
 		return ret;
-	pdata->phy.advertising = pdata->phy.supported;
+
+	/* Copy supported link modes to advertising link modes */
+	XGBE_LM_COPY(lks, advertising, lks, supported);
 
 	pdata->phy.address = 0;
 
-	if (pdata->phy.advertising & ADVERTISED_Autoneg) {
+	if (XGBE_ADV(lks, Autoneg)) {
 		pdata->phy.autoneg = AUTONEG_ENABLE;
 		pdata->phy.speed = SPEED_UNKNOWN;
 		pdata->phy.duplex = DUPLEX_UNKNOWN;
@@ -1549,16 +1562,21 @@ static int xgbe_phy_init(struct xgbe_prv_data *pdata)
 	pdata->phy.rx_pause = pdata->rx_pause;
 
 	/* Fix up Flow Control advertising */
-	pdata->phy.advertising &= ~ADVERTISED_Pause;
-	pdata->phy.advertising &= ~ADVERTISED_Asym_Pause;
+	XGBE_CLR_ADV(lks, Pause);
+	XGBE_CLR_ADV(lks, Asym_Pause);
 
 	if (pdata->rx_pause) {
-		pdata->phy.advertising |= ADVERTISED_Pause;
-		pdata->phy.advertising |= ADVERTISED_Asym_Pause;
+		XGBE_SET_ADV(lks, Pause);
+		XGBE_SET_ADV(lks, Asym_Pause);
 	}
 
-	if (pdata->tx_pause)
-		pdata->phy.advertising ^= ADVERTISED_Asym_Pause;
+	if (pdata->tx_pause) {
+		/* Equivalent to XOR of Asym_Pause */
+		if (XGBE_ADV(lks, Asym_Pause))
+			XGBE_CLR_ADV(lks, Asym_Pause);
+		else
+			XGBE_SET_ADV(lks, Asym_Pause);
+	}
 
 	if (netif_msg_drv(pdata))
 		xgbe_dump_phy_registers(pdata);
@@ -1581,4 +1599,7 @@ void xgbe_init_function_ptrs_phy(struct xgbe_phy_if *phy_if)
 	phy_if->phy_valid_speed = xgbe_phy_valid_speed;
 
 	phy_if->an_isr          = xgbe_an_combined_isr;
+
+	phy_if->module_info     = xgbe_phy_module_info;
+	phy_if->module_eeprom   = xgbe_phy_module_eeprom;
 }
diff --git a/drivers/net/ethernet/amd/xgbe/xgbe-pci.c b/drivers/net/ethernet/amd/xgbe/xgbe-pci.c
index 38392a520725..e3e1dca9856a 100644
--- a/drivers/net/ethernet/amd/xgbe/xgbe-pci.c
+++ b/drivers/net/ethernet/amd/xgbe/xgbe-pci.c
@@ -1,123 +1,15 @@
+// SPDX-License-Identifier: (GPL-2.0-or-later OR BSD-3-Clause)
 /*
- * AMD 10Gb Ethernet driver
- *
- * This file is available to you under your choice of the following two
- * licenses:
- *
- * License 1: GPLv2
- *
- * Copyright (c) 2016 Advanced Micro Devices, Inc.
- *
- * This file is free software; you may copy, redistribute and/or modify
- * it under the terms of the GNU General Public License as published by
- * the Free Software Foundation, either version 2 of the License, or (at
- * your option) any later version.
- *
- * This file is distributed in the hope that it will be useful, but
- * WITHOUT ANY WARRANTY; without even the implied warranty of
- * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
- * General Public License for more details.
- *
- * You should have received a copy of the GNU General Public License
- * along with this program.  If not, see <http://www.gnu.org/licenses/>.
- *
- * This file incorporates work covered by the following copyright and
- * permission notice:
- *     The Synopsys DWC ETHER XGMAC Software Driver and documentation
- *     (hereinafter "Software") is an unsupported proprietary work of Synopsys,
- *     Inc. unless otherwise expressly agreed to in writing between Synopsys
- *     and you.
- *
- *     The Software IS NOT an item of Licensed Software or Licensed Product
- *     under any End User Software License Agreement or Agreement for Licensed
- *     Product with Synopsys or any supplement thereto.  Permission is hereby
- *     granted, free of charge, to any person obtaining a copy of this software
- *     annotated with this license and the Software, to deal in the Software
- *     without restriction, including without limitation the rights to use,
- *     copy, modify, merge, publish, distribute, sublicense, and/or sell copies
- *     of the Software, and to permit persons to whom the Software is furnished
- *     to do so, subject to the following conditions:
- *
- *     The above copyright notice and this permission notice shall be included
- *     in all copies or substantial portions of the Software.
- *
- *     THIS SOFTWARE IS BEING DISTRIBUTED BY SYNOPSYS SOLELY ON AN "AS IS"
- *     BASIS AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED
- *     TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A
- *     PARTICULAR PURPOSE ARE HEREBY DISCLAIMED. IN NO EVENT SHALL SYNOPSYS
- *     BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR
- *     CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF
- *     SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS
- *     INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN
- *     CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE)
- *     ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF
- *     THE POSSIBILITY OF SUCH DAMAGE.
- *
- *
- * License 2: Modified BSD
- *
- * Copyright (c) 2016 Advanced Micro Devices, Inc.
- * All rights reserved.
- *
- * Redistribution and use in source and binary forms, with or without
- * modification, are permitted provided that the following conditions are met:
- *     * Redistributions of source code must retain the above copyright
- *       notice, this list of conditions and the following disclaimer.
- *     * Redistributions in binary form must reproduce the above copyright
- *       notice, this list of conditions and the following disclaimer in the
- *       documentation and/or other materials provided with the distribution.
- *     * Neither the name of Advanced Micro Devices, Inc. nor the
- *       names of its contributors may be used to endorse or promote products
- *       derived from this software without specific prior written permission.
- *
- * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS"
- * AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
- * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
- * ARE DISCLAIMED. IN NO EVENT SHALL <COPYRIGHT HOLDER> BE LIABLE FOR ANY
- * DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES
- * (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES;
- * LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND
- * ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
- * (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF
- * THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
- *
- * This file incorporates work covered by the following copyright and
- * permission notice:
- *     The Synopsys DWC ETHER XGMAC Software Driver and documentation
- *     (hereinafter "Software") is an unsupported proprietary work of Synopsys,
- *     Inc. unless otherwise expressly agreed to in writing between Synopsys
- *     and you.
- *
- *     The Software IS NOT an item of Licensed Software or Licensed Product
- *     under any End User Software License Agreement or Agreement for Licensed
- *     Product with Synopsys or any supplement thereto.  Permission is hereby
- *     granted, free of charge, to any person obtaining a copy of this software
- *     annotated with this license and the Software, to deal in the Software
- *     without restriction, including without limitation the rights to use,
- *     copy, modify, merge, publish, distribute, sublicense, and/or sell copies
- *     of the Software, and to permit persons to whom the Software is furnished
- *     to do so, subject to the following conditions:
- *
- *     The above copyright notice and this permission notice shall be included
- *     in all copies or substantial portions of the Software.
- *
- *     THIS SOFTWARE IS BEING DISTRIBUTED BY SYNOPSYS SOLELY ON AN "AS IS"
- *     BASIS AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED
- *     TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A
- *     PARTICULAR PURPOSE ARE HEREBY DISCLAIMED. IN NO EVENT SHALL SYNOPSYS
- *     BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR
- *     CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF
- *     SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS
- *     INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN
- *     CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE)
- *     ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF
- *     THE POSSIBILITY OF SUCH DAMAGE.
+ * Copyright (c) 2014-2025, Advanced Micro Devices, Inc.
+ * Copyright (c) 2014, Synopsys, Inc.
+ * All rights reserved
  */
 
 #include <linux/module.h>
 #include <linux/device.h>
 #include <linux/pci.h>
 #include <linux/log2.h>
+#include "xgbe-smn.h"
 
 #include "xgbe.h"
 #include "xgbe-common.h"
@@ -139,6 +31,7 @@ static int xgbe_config_multi_msi(struct xgbe_prv_data *pdata)
 		return ret;
 	}
 
+	pdata->isr_as_bh_work = 1;
 	pdata->irq_count = ret;
 
 	pdata->dev_irq = pci_irq_vector(pdata->pcidev, 0);
@@ -169,12 +62,13 @@ static int xgbe_config_irqs(struct xgbe_prv_data *pdata)
 		goto out;
 
 	ret = pci_alloc_irq_vectors(pdata->pcidev, 1, 1,
-				    PCI_IRQ_LEGACY | PCI_IRQ_MSI);
+				    PCI_IRQ_INTX | PCI_IRQ_MSI);
 	if (ret < 0) {
 		dev_info(pdata->dev, "single IRQ enablement failed\n");
 		return ret;
 	}
 
+	pdata->isr_as_bh_work = pdata->pcidev->msi_enabled ? 1 : 0;
 	pdata->irq_count = 1;
 	pdata->channel_irq_count = 1;
 
@@ -205,14 +99,14 @@ out:
 
 static int xgbe_pci_probe(struct pci_dev *pdev, const struct pci_device_id *id)
 {
-	struct xgbe_prv_data *pdata;
-	struct device *dev = &pdev->dev;
 	void __iomem * const *iomap_table;
-	struct pci_dev *rdev;
+	unsigned int port_addr_size, reg;
+	struct device *dev = &pdev->dev;
+	struct xgbe_prv_data *pdata;
 	unsigned int ma_lo, ma_hi;
-	unsigned int reg;
-	int bar_mask;
-	int ret;
+	struct pci_dev *rdev;
+	int bar_mask, ret;
+	u32 address;
 
 	pdata = xgbe_alloc_pdata(dev);
 	if (IS_ERR(pdata)) {
@@ -272,10 +166,31 @@ static int xgbe_pci_probe(struct pci_dev *pdev, const struct pci_device_id *id)
 
 	/* Set the PCS indirect addressing definition registers */
 	rdev = pci_get_domain_bus_and_slot(0, 0, PCI_DEVFN(0, 0));
-	if (rdev &&
-	    (rdev->vendor == PCI_VENDOR_ID_AMD) && (rdev->device == 0x15d0)) {
-		pdata->xpcs_window_def_reg = PCS_V2_RV_WINDOW_DEF;
-		pdata->xpcs_window_sel_reg = PCS_V2_RV_WINDOW_SELECT;
+	if (rdev && rdev->vendor == PCI_VENDOR_ID_AMD) {
+		switch (rdev->device) {
+		case XGBE_RV_PCI_DEVICE_ID:
+			pdata->xpcs_window_def_reg = PCS_V2_RV_WINDOW_DEF;
+			pdata->xpcs_window_sel_reg = PCS_V2_RV_WINDOW_SELECT;
+			break;
+		case XGBE_YC_PCI_DEVICE_ID:
+			pdata->xpcs_window_def_reg = PCS_V2_YC_WINDOW_DEF;
+			pdata->xpcs_window_sel_reg = PCS_V2_YC_WINDOW_SELECT;
+
+			/* Yellow Carp devices do not need cdr workaround */
+			pdata->vdata->an_cdr_workaround = 0;
+
+			/* Yellow Carp devices do not need rrc */
+			pdata->vdata->enable_rrc = 0;
+			break;
+		case XGBE_RN_PCI_DEVICE_ID:
+			pdata->xpcs_window_def_reg = PCS_V3_RN_WINDOW_DEF;
+			pdata->xpcs_window_sel_reg = PCS_V3_RN_WINDOW_SELECT;
+			break;
+		default:
+			pdata->xpcs_window_def_reg = PCS_V2_WINDOW_DEF;
+			pdata->xpcs_window_sel_reg = PCS_V2_WINDOW_SELECT;
+			break;
+		}
 	} else {
 		pdata->xpcs_window_def_reg = PCS_V2_WINDOW_DEF;
 		pdata->xpcs_window_sel_reg = PCS_V2_WINDOW_SELECT;
@@ -283,13 +198,32 @@ static int xgbe_pci_probe(struct pci_dev *pdev, const struct pci_device_id *id)
 	pci_dev_put(rdev);
 
 	/* Configure the PCS indirect addressing support */
-	reg = XPCS32_IOREAD(pdata, pdata->xpcs_window_def_reg);
+	if (pdata->vdata->xpcs_access == XGBE_XPCS_ACCESS_V3) {
+		reg = XP_IOREAD(pdata, XP_PROP_0);
+		port_addr_size = PCS_RN_PORT_ADDR_SIZE *
+				 XP_GET_BITS(reg, XP_PROP_0, PORT_ID);
+		pdata->smn_base = PCS_RN_SMN_BASE_ADDR + port_addr_size;
+
+		address = pdata->smn_base + (pdata->xpcs_window_def_reg);
+		ret = amd_smn_read(0, address, &reg);
+		if (ret) {
+			pci_err(pdata->pcidev, "Failed to read data\n");
+			goto err_pci_enable;
+		}
+	} else {
+		reg = XPCS32_IOREAD(pdata, pdata->xpcs_window_def_reg);
+	}
+
 	pdata->xpcs_window = XPCS_GET_BITS(reg, PCS_V2_WINDOW_DEF, OFFSET);
 	pdata->xpcs_window <<= 6;
 	pdata->xpcs_window_size = XPCS_GET_BITS(reg, PCS_V2_WINDOW_DEF, SIZE);
 	pdata->xpcs_window_size = 1 << (pdata->xpcs_window_size + 7);
 	pdata->xpcs_window_mask = pdata->xpcs_window_size - 1;
 	if (netif_msg_probe(pdata)) {
+		dev_dbg(dev, "xpcs window def  = %#010x\n",
+			pdata->xpcs_window_def_reg);
+		dev_dbg(dev, "xpcs window sel  = %#010x\n",
+			pdata->xpcs_window_sel_reg);
 		dev_dbg(dev, "xpcs window      = %#010x\n",
 			pdata->xpcs_window);
 		dev_dbg(dev, "xpcs window size = %#010x\n",
@@ -325,20 +259,37 @@ static int xgbe_pci_probe(struct pci_dev *pdev, const struct pci_device_id *id)
 
 	/* Set the DMA coherency values */
 	pdata->coherent = 1;
-	pdata->axdomain = XGBE_DMA_OS_AXDOMAIN;
-	pdata->arcache = XGBE_DMA_OS_ARCACHE;
-	pdata->awcache = XGBE_DMA_OS_AWCACHE;
+	pdata->arcr = XGBE_DMA_PCI_ARCR;
+	pdata->awcr = XGBE_DMA_PCI_AWCR;
+	pdata->awarcr = XGBE_DMA_PCI_AWARCR;
+
+	/* Read the port property registers */
+	pdata->pp0 = XP_IOREAD(pdata, XP_PROP_0);
+	pdata->pp1 = XP_IOREAD(pdata, XP_PROP_1);
+	pdata->pp2 = XP_IOREAD(pdata, XP_PROP_2);
+	pdata->pp3 = XP_IOREAD(pdata, XP_PROP_3);
+	pdata->pp4 = XP_IOREAD(pdata, XP_PROP_4);
+	if (netif_msg_probe(pdata)) {
+		dev_dbg(dev, "port property 0 = %#010x\n", pdata->pp0);
+		dev_dbg(dev, "port property 1 = %#010x\n", pdata->pp1);
+		dev_dbg(dev, "port property 2 = %#010x\n", pdata->pp2);
+		dev_dbg(dev, "port property 3 = %#010x\n", pdata->pp3);
+		dev_dbg(dev, "port property 4 = %#010x\n", pdata->pp4);
+	}
 
 	/* Set the maximum channels and queues */
-	reg = XP_IOREAD(pdata, XP_PROP_1);
-	pdata->tx_max_channel_count = XP_GET_BITS(reg, XP_PROP_1, MAX_TX_DMA);
-	pdata->rx_max_channel_count = XP_GET_BITS(reg, XP_PROP_1, MAX_RX_DMA);
-	pdata->tx_max_q_count = XP_GET_BITS(reg, XP_PROP_1, MAX_TX_QUEUES);
-	pdata->rx_max_q_count = XP_GET_BITS(reg, XP_PROP_1, MAX_RX_QUEUES);
+	pdata->tx_max_channel_count = XP_GET_BITS(pdata->pp1, XP_PROP_1,
+						  MAX_TX_DMA);
+	pdata->rx_max_channel_count = XP_GET_BITS(pdata->pp1, XP_PROP_1,
+						  MAX_RX_DMA);
+	pdata->tx_max_q_count = XP_GET_BITS(pdata->pp1, XP_PROP_1,
+					    MAX_TX_QUEUES);
+	pdata->rx_max_q_count = XP_GET_BITS(pdata->pp1, XP_PROP_1,
+					    MAX_RX_QUEUES);
 	if (netif_msg_probe(pdata)) {
 		dev_dbg(dev, "max tx/rx channel count = %u/%u\n",
 			pdata->tx_max_channel_count,
-			pdata->tx_max_channel_count);
+			pdata->rx_max_channel_count);
 		dev_dbg(dev, "max tx/rx hw queue count = %u/%u\n",
 			pdata->tx_max_q_count, pdata->rx_max_q_count);
 	}
@@ -347,12 +298,13 @@ static int xgbe_pci_probe(struct pci_dev *pdev, const struct pci_device_id *id)
 	xgbe_set_counts(pdata);
 
 	/* Set the maximum fifo amounts */
-	reg = XP_IOREAD(pdata, XP_PROP_2);
-	pdata->tx_max_fifo_size = XP_GET_BITS(reg, XP_PROP_2, TX_FIFO_SIZE);
+	pdata->tx_max_fifo_size = XP_GET_BITS(pdata->pp2, XP_PROP_2,
+					      TX_FIFO_SIZE);
 	pdata->tx_max_fifo_size *= 16384;
 	pdata->tx_max_fifo_size = min(pdata->tx_max_fifo_size,
 				      pdata->vdata->tx_max_fifo_size);
-	pdata->rx_max_fifo_size = XP_GET_BITS(reg, XP_PROP_2, RX_FIFO_SIZE);
+	pdata->rx_max_fifo_size = XP_GET_BITS(pdata->pp2, XP_PROP_2,
+					      RX_FIFO_SIZE);
 	pdata->rx_max_fifo_size *= 16384;
 	pdata->rx_max_fifo_size = min(pdata->rx_max_fifo_size,
 				      pdata->vdata->rx_max_fifo_size);
@@ -394,13 +346,15 @@ static void xgbe_pci_remove(struct pci_dev *pdev)
 
 	pci_free_irq_vectors(pdata->pcidev);
 
+	/* Disable all interrupts in the hardware */
+	XP_IOWRITE(pdata, XP_INT_EN, 0x0);
+
 	xgbe_free_pdata(pdata);
 }
 
-#ifdef CONFIG_PM
-static int xgbe_pci_suspend(struct pci_dev *pdev, pm_message_t state)
+static int __maybe_unused xgbe_pci_suspend(struct device *dev)
 {
-	struct xgbe_prv_data *pdata = pci_get_drvdata(pdev);
+	struct xgbe_prv_data *pdata = dev_get_drvdata(dev);
 	struct net_device *netdev = pdata->netdev;
 	int ret = 0;
 
@@ -414,12 +368,14 @@ static int xgbe_pci_suspend(struct pci_dev *pdev, pm_message_t state)
 	return ret;
 }
 
-static int xgbe_pci_resume(struct pci_dev *pdev)
+static int __maybe_unused xgbe_pci_resume(struct device *dev)
 {
-	struct xgbe_prv_data *pdata = pci_get_drvdata(pdev);
+	struct xgbe_prv_data *pdata = dev_get_drvdata(dev);
 	struct net_device *netdev = pdata->netdev;
 	int ret = 0;
 
+	XP_IOWRITE(pdata, XP_INT_EN, 0x1fffff);
+
 	pdata->lpm_ctrl &= ~MDIO_CTRL1_LPOWER;
 	XMDIO_WRITE(pdata, MDIO_MMD_PCS, MDIO_CTRL1, pdata->lpm_ctrl);
 
@@ -434,28 +390,55 @@ static int xgbe_pci_resume(struct pci_dev *pdev)
 
 	return ret;
 }
-#endif /* CONFIG_PM */
 
-static const struct xgbe_version_data xgbe_v2a = {
+static struct xgbe_version_data xgbe_v3 = {
+	.init_function_ptrs_phy_impl	= xgbe_init_function_ptrs_phy_v2,
+	.xpcs_access			= XGBE_XPCS_ACCESS_V3,
+	.mmc_64bit			= 1,
+	.tx_max_fifo_size		= 65536,
+	.rx_max_fifo_size		= 65536,
+	.tx_tstamp_workaround		= 1,
+	.ecc_support			= 1,
+	.i2c_support			= 1,
+	.irq_reissue_support		= 1,
+	.tx_desc_prefetch		= 5,
+	.rx_desc_prefetch		= 5,
+	.an_cdr_workaround		= 0,
+	.enable_rrc			= 0,
+};
+
+static struct xgbe_version_data xgbe_v2a = {
 	.init_function_ptrs_phy_impl	= xgbe_init_function_ptrs_phy_v2,
 	.xpcs_access			= XGBE_XPCS_ACCESS_V2,
 	.mmc_64bit			= 1,
 	.tx_max_fifo_size		= 229376,
 	.rx_max_fifo_size		= 229376,
 	.tx_tstamp_workaround		= 1,
+	.tstamp_ptp_clock_freq		= 1,
 	.ecc_support			= 1,
 	.i2c_support			= 1,
+	.irq_reissue_support		= 1,
+	.tx_desc_prefetch		= 5,
+	.rx_desc_prefetch		= 5,
+	.an_cdr_workaround		= 1,
+	.enable_rrc			= 1,
 };
 
-static const struct xgbe_version_data xgbe_v2b = {
+static struct xgbe_version_data xgbe_v2b = {
 	.init_function_ptrs_phy_impl	= xgbe_init_function_ptrs_phy_v2,
 	.xpcs_access			= XGBE_XPCS_ACCESS_V2,
 	.mmc_64bit			= 1,
 	.tx_max_fifo_size		= 65536,
 	.rx_max_fifo_size		= 65536,
 	.tx_tstamp_workaround		= 1,
+	.tstamp_ptp_clock_freq		= 1,
 	.ecc_support			= 1,
 	.i2c_support			= 1,
+	.irq_reissue_support		= 1,
+	.tx_desc_prefetch		= 5,
+	.rx_desc_prefetch		= 5,
+	.an_cdr_workaround		= 1,
+	.enable_rrc			= 1,
 };
 
 static const struct pci_device_id xgbe_pci_table[] = {
@@ -463,20 +446,23 @@ static const struct pci_device_id xgbe_pci_table[] = {
 	  .driver_data = (kernel_ulong_t)&xgbe_v2a },
 	{ PCI_VDEVICE(AMD, 0x1459),
 	  .driver_data = (kernel_ulong_t)&xgbe_v2b },
+	{ PCI_VDEVICE(AMD, 0x1641),
+	  .driver_data = (kernel_ulong_t)&xgbe_v3 },
 	/* Last entry must be zero */
 	{ 0, }
 };
 MODULE_DEVICE_TABLE(pci, xgbe_pci_table);
 
+static SIMPLE_DEV_PM_OPS(xgbe_pci_pm_ops, xgbe_pci_suspend, xgbe_pci_resume);
+
 static struct pci_driver xgbe_driver = {
 	.name = XGBE_DRV_NAME,
 	.id_table = xgbe_pci_table,
 	.probe = xgbe_pci_probe,
 	.remove = xgbe_pci_remove,
-#ifdef CONFIG_PM
-	.suspend = xgbe_pci_suspend,
-	.resume = xgbe_pci_resume,
-#endif
+	.driver = {
+		.pm = &xgbe_pci_pm_ops,
+	}
 };
 
 int xgbe_pci_init(void)
diff --git a/drivers/net/ethernet/amd/xgbe/xgbe-phy-v1.c b/drivers/net/ethernet/amd/xgbe/xgbe-phy-v1.c
index c75edcac5e0a..2e6b8ffe785c 100644
--- a/drivers/net/ethernet/amd/xgbe/xgbe-phy-v1.c
+++ b/drivers/net/ethernet/amd/xgbe/xgbe-phy-v1.c
@@ -1,117 +1,8 @@
+// SPDX-License-Identifier: (GPL-2.0-or-later OR BSD-3-Clause)
 /*
- * AMD 10Gb Ethernet driver
- *
- * This file is available to you under your choice of the following two
- * licenses:
- *
- * License 1: GPLv2
- *
- * Copyright (c) 2016 Advanced Micro Devices, Inc.
- *
- * This file is free software; you may copy, redistribute and/or modify
- * it under the terms of the GNU General Public License as published by
- * the Free Software Foundation, either version 2 of the License, or (at
- * your option) any later version.
- *
- * This file is distributed in the hope that it will be useful, but
- * WITHOUT ANY WARRANTY; without even the implied warranty of
- * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
- * General Public License for more details.
- *
- * You should have received a copy of the GNU General Public License
- * along with this program.  If not, see <http://www.gnu.org/licenses/>.
- *
- * This file incorporates work covered by the following copyright and
- * permission notice:
- *     The Synopsys DWC ETHER XGMAC Software Driver and documentation
- *     (hereinafter "Software") is an unsupported proprietary work of Synopsys,
- *     Inc. unless otherwise expressly agreed to in writing between Synopsys
- *     and you.
- *
- *     The Software IS NOT an item of Licensed Software or Licensed Product
- *     under any End User Software License Agreement or Agreement for Licensed
- *     Product with Synopsys or any supplement thereto.  Permission is hereby
- *     granted, free of charge, to any person obtaining a copy of this software
- *     annotated with this license and the Software, to deal in the Software
- *     without restriction, including without limitation the rights to use,
- *     copy, modify, merge, publish, distribute, sublicense, and/or sell copies
- *     of the Software, and to permit persons to whom the Software is furnished
- *     to do so, subject to the following conditions:
- *
- *     The above copyright notice and this permission notice shall be included
- *     in all copies or substantial portions of the Software.
- *
- *     THIS SOFTWARE IS BEING DISTRIBUTED BY SYNOPSYS SOLELY ON AN "AS IS"
- *     BASIS AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED
- *     TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A
- *     PARTICULAR PURPOSE ARE HEREBY DISCLAIMED. IN NO EVENT SHALL SYNOPSYS
- *     BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR
- *     CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF
- *     SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS
- *     INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN
- *     CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE)
- *     ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF
- *     THE POSSIBILITY OF SUCH DAMAGE.
- *
- *
- * License 2: Modified BSD
- *
- * Copyright (c) 2016 Advanced Micro Devices, Inc.
- * All rights reserved.
- *
- * Redistribution and use in source and binary forms, with or without
- * modification, are permitted provided that the following conditions are met:
- *     * Redistributions of source code must retain the above copyright
- *       notice, this list of conditions and the following disclaimer.
- *     * Redistributions in binary form must reproduce the above copyright
- *       notice, this list of conditions and the following disclaimer in the
- *       documentation and/or other materials provided with the distribution.
- *     * Neither the name of Advanced Micro Devices, Inc. nor the
- *       names of its contributors may be used to endorse or promote products
- *       derived from this software without specific prior written permission.
- *
- * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS"
- * AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
- * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
- * ARE DISCLAIMED. IN NO EVENT SHALL <COPYRIGHT HOLDER> BE LIABLE FOR ANY
- * DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES
- * (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES;
- * LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND
- * ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
- * (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF
- * THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
- *
- * This file incorporates work covered by the following copyright and
- * permission notice:
- *     The Synopsys DWC ETHER XGMAC Software Driver and documentation
- *     (hereinafter "Software") is an unsupported proprietary work of Synopsys,
- *     Inc. unless otherwise expressly agreed to in writing between Synopsys
- *     and you.
- *
- *     The Software IS NOT an item of Licensed Software or Licensed Product
- *     under any End User Software License Agreement or Agreement for Licensed
- *     Product with Synopsys or any supplement thereto.  Permission is hereby
- *     granted, free of charge, to any person obtaining a copy of this software
- *     annotated with this license and the Software, to deal in the Software
- *     without restriction, including without limitation the rights to use,
- *     copy, modify, merge, publish, distribute, sublicense, and/or sell copies
- *     of the Software, and to permit persons to whom the Software is furnished
- *     to do so, subject to the following conditions:
- *
- *     The above copyright notice and this permission notice shall be included
- *     in all copies or substantial portions of the Software.
- *
- *     THIS SOFTWARE IS BEING DISTRIBUTED BY SYNOPSYS SOLELY ON AN "AS IS"
- *     BASIS AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED
- *     TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A
- *     PARTICULAR PURPOSE ARE HEREBY DISCLAIMED. IN NO EVENT SHALL SYNOPSYS
- *     BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR
- *     CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF
- *     SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS
- *     INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN
- *     CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE)
- *     ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF
- *     THE POSSIBILITY OF SUCH DAMAGE.
+ * Copyright (c) 2014-2025, Advanced Micro Devices, Inc.
+ * Copyright (c) 2014, Synopsys, Inc.
+ * All rights reserved
  */
 
 #include <linux/module.h>
@@ -231,20 +122,21 @@ static void xgbe_phy_kr_training_post(struct xgbe_prv_data *pdata)
 
 static enum xgbe_mode xgbe_phy_an_outcome(struct xgbe_prv_data *pdata)
 {
+	struct ethtool_link_ksettings *lks = &pdata->phy.lks;
 	struct xgbe_phy_data *phy_data = pdata->phy_data;
 	enum xgbe_mode mode;
 	unsigned int ad_reg, lp_reg;
 
-	pdata->phy.lp_advertising |= ADVERTISED_Autoneg;
-	pdata->phy.lp_advertising |= ADVERTISED_Backplane;
+	XGBE_SET_LP_ADV(lks, Autoneg);
+	XGBE_SET_LP_ADV(lks, Backplane);
 
 	/* Compare Advertisement and Link Partner register 1 */
 	ad_reg = XMDIO_READ(pdata, MDIO_MMD_AN, MDIO_AN_ADVERTISE);
 	lp_reg = XMDIO_READ(pdata, MDIO_MMD_AN, MDIO_AN_LPA);
 	if (lp_reg & 0x400)
-		pdata->phy.lp_advertising |= ADVERTISED_Pause;
+		XGBE_SET_LP_ADV(lks, Pause);
 	if (lp_reg & 0x800)
-		pdata->phy.lp_advertising |= ADVERTISED_Asym_Pause;
+		XGBE_SET_LP_ADV(lks, Asym_Pause);
 
 	if (pdata->phy.pause_autoneg) {
 		/* Set flow control based on auto-negotiation result */
@@ -266,12 +158,12 @@ static enum xgbe_mode xgbe_phy_an_outcome(struct xgbe_prv_data *pdata)
 	ad_reg = XMDIO_READ(pdata, MDIO_MMD_AN, MDIO_AN_ADVERTISE + 1);
 	lp_reg = XMDIO_READ(pdata, MDIO_MMD_AN, MDIO_AN_LPA + 1);
 	if (lp_reg & 0x80)
-		pdata->phy.lp_advertising |= ADVERTISED_10000baseKR_Full;
+		XGBE_SET_LP_ADV(lks, 10000baseKR_Full);
 	if (lp_reg & 0x20) {
 		if (phy_data->speed_set == XGBE_SPEEDSET_2500_10000)
-			pdata->phy.lp_advertising |= ADVERTISED_2500baseX_Full;
+			XGBE_SET_LP_ADV(lks, 2500baseX_Full);
 		else
-			pdata->phy.lp_advertising |= ADVERTISED_1000baseKX_Full;
+			XGBE_SET_LP_ADV(lks, 1000baseKX_Full);
 	}
 
 	ad_reg &= lp_reg;
@@ -290,14 +182,17 @@ static enum xgbe_mode xgbe_phy_an_outcome(struct xgbe_prv_data *pdata)
 	ad_reg = XMDIO_READ(pdata, MDIO_MMD_AN, MDIO_AN_ADVERTISE + 2);
 	lp_reg = XMDIO_READ(pdata, MDIO_MMD_AN, MDIO_AN_LPA + 2);
 	if (lp_reg & 0xc000)
-		pdata->phy.lp_advertising |= ADVERTISED_10000baseR_FEC;
+		XGBE_SET_LP_ADV(lks, 10000baseR_FEC);
 
 	return mode;
 }
 
-static unsigned int xgbe_phy_an_advertising(struct xgbe_prv_data *pdata)
+static void xgbe_phy_an_advertising(struct xgbe_prv_data *pdata,
+				    struct ethtool_link_ksettings *dlks)
 {
-	return pdata->phy.advertising;
+	struct ethtool_link_ksettings *slks = &pdata->phy.lks;
+
+	XGBE_LM_COPY(dlks, advertising, slks, advertising);
 }
 
 static int xgbe_phy_an_config(struct xgbe_prv_data *pdata)
@@ -565,11 +460,10 @@ static void xgbe_phy_set_mode(struct xgbe_prv_data *pdata, enum xgbe_mode mode)
 }
 
 static bool xgbe_phy_check_mode(struct xgbe_prv_data *pdata,
-				enum xgbe_mode mode, u32 advert)
+				enum xgbe_mode mode, bool advert)
 {
 	if (pdata->phy.autoneg == AUTONEG_ENABLE) {
-		if (pdata->phy.advertising & advert)
-			return true;
+		return advert;
 	} else {
 		enum xgbe_mode cur_mode;
 
@@ -583,16 +477,18 @@ static bool xgbe_phy_check_mode(struct xgbe_prv_data *pdata,
 
 static bool xgbe_phy_use_mode(struct xgbe_prv_data *pdata, enum xgbe_mode mode)
 {
+	struct ethtool_link_ksettings *lks = &pdata->phy.lks;
+
 	switch (mode) {
 	case XGBE_MODE_KX_1000:
 		return xgbe_phy_check_mode(pdata, mode,
-					   ADVERTISED_1000baseKX_Full);
+					   XGBE_ADV(lks, 1000baseKX_Full));
 	case XGBE_MODE_KX_2500:
 		return xgbe_phy_check_mode(pdata, mode,
-					   ADVERTISED_2500baseX_Full);
+					   XGBE_ADV(lks, 2500baseX_Full));
 	case XGBE_MODE_KR:
 		return xgbe_phy_check_mode(pdata, mode,
-					   ADVERTISED_10000baseKR_Full);
+					   XGBE_ADV(lks, 10000baseKR_Full));
 	default:
 		return false;
 	}
@@ -672,6 +568,7 @@ static void xgbe_phy_exit(struct xgbe_prv_data *pdata)
 
 static int xgbe_phy_init(struct xgbe_prv_data *pdata)
 {
+	struct ethtool_link_ksettings *lks = &pdata->phy.lks;
 	struct xgbe_phy_data *phy_data;
 	int ret;
 
@@ -790,21 +687,23 @@ static int xgbe_phy_init(struct xgbe_prv_data *pdata)
 	}
 
 	/* Initialize supported features */
-	pdata->phy.supported = SUPPORTED_Autoneg;
-	pdata->phy.supported |= SUPPORTED_Pause | SUPPORTED_Asym_Pause;
-	pdata->phy.supported |= SUPPORTED_Backplane;
-	pdata->phy.supported |= SUPPORTED_10000baseKR_Full;
+	XGBE_ZERO_SUP(lks);
+	XGBE_SET_SUP(lks, Autoneg);
+	XGBE_SET_SUP(lks, Pause);
+	XGBE_SET_SUP(lks, Asym_Pause);
+	XGBE_SET_SUP(lks, Backplane);
+	XGBE_SET_SUP(lks, 10000baseKR_Full);
 	switch (phy_data->speed_set) {
 	case XGBE_SPEEDSET_1000_10000:
-		pdata->phy.supported |= SUPPORTED_1000baseKX_Full;
+		XGBE_SET_SUP(lks, 1000baseKX_Full);
 		break;
 	case XGBE_SPEEDSET_2500_10000:
-		pdata->phy.supported |= SUPPORTED_2500baseX_Full;
+		XGBE_SET_SUP(lks, 2500baseX_Full);
 		break;
 	}
 
 	if (pdata->fec_ability & MDIO_PMA_10GBR_FECABLE_ABLE)
-		pdata->phy.supported |= SUPPORTED_10000baseR_FEC;
+		XGBE_SET_SUP(lks, 10000baseR_FEC);
 
 	pdata->phy_data = phy_data;
 
diff --git a/drivers/net/ethernet/amd/xgbe/xgbe-phy-v2.c b/drivers/net/ethernet/amd/xgbe/xgbe-phy-v2.c
index e707c49cc55a..a56efc1bee33 100644
--- a/drivers/net/ethernet/amd/xgbe/xgbe-phy-v2.c
+++ b/drivers/net/ethernet/amd/xgbe/xgbe-phy-v2.c
@@ -1,117 +1,8 @@
+// SPDX-License-Identifier: (GPL-2.0-or-later OR BSD-3-Clause)
 /*
- * AMD 10Gb Ethernet driver
- *
- * This file is available to you under your choice of the following two
- * licenses:
- *
- * License 1: GPLv2
- *
- * Copyright (c) 2016 Advanced Micro Devices, Inc.
- *
- * This file is free software; you may copy, redistribute and/or modify
- * it under the terms of the GNU General Public License as published by
- * the Free Software Foundation, either version 2 of the License, or (at
- * your option) any later version.
- *
- * This file is distributed in the hope that it will be useful, but
- * WITHOUT ANY WARRANTY; without even the implied warranty of
- * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
- * General Public License for more details.
- *
- * You should have received a copy of the GNU General Public License
- * along with this program.  If not, see <http://www.gnu.org/licenses/>.
- *
- * This file incorporates work covered by the following copyright and
- * permission notice:
- *     The Synopsys DWC ETHER XGMAC Software Driver and documentation
- *     (hereinafter "Software") is an unsupported proprietary work of Synopsys,
- *     Inc. unless otherwise expressly agreed to in writing between Synopsys
- *     and you.
- *
- *     The Software IS NOT an item of Licensed Software or Licensed Product
- *     under any End User Software License Agreement or Agreement for Licensed
- *     Product with Synopsys or any supplement thereto.  Permission is hereby
- *     granted, free of charge, to any person obtaining a copy of this software
- *     annotated with this license and the Software, to deal in the Software
- *     without restriction, including without limitation the rights to use,
- *     copy, modify, merge, publish, distribute, sublicense, and/or sell copies
- *     of the Software, and to permit persons to whom the Software is furnished
- *     to do so, subject to the following conditions:
- *
- *     The above copyright notice and this permission notice shall be included
- *     in all copies or substantial portions of the Software.
- *
- *     THIS SOFTWARE IS BEING DISTRIBUTED BY SYNOPSYS SOLELY ON AN "AS IS"
- *     BASIS AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED
- *     TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A
- *     PARTICULAR PURPOSE ARE HEREBY DISCLAIMED. IN NO EVENT SHALL SYNOPSYS
- *     BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR
- *     CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF
- *     SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS
- *     INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN
- *     CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE)
- *     ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF
- *     THE POSSIBILITY OF SUCH DAMAGE.
- *
- *
- * License 2: Modified BSD
- *
- * Copyright (c) 2016 Advanced Micro Devices, Inc.
- * All rights reserved.
- *
- * Redistribution and use in source and binary forms, with or without
- * modification, are permitted provided that the following conditions are met:
- *     * Redistributions of source code must retain the above copyright
- *       notice, this list of conditions and the following disclaimer.
- *     * Redistributions in binary form must reproduce the above copyright
- *       notice, this list of conditions and the following disclaimer in the
- *       documentation and/or other materials provided with the distribution.
- *     * Neither the name of Advanced Micro Devices, Inc. nor the
- *       names of its contributors may be used to endorse or promote products
- *       derived from this software without specific prior written permission.
- *
- * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS"
- * AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
- * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
- * ARE DISCLAIMED. IN NO EVENT SHALL <COPYRIGHT HOLDER> BE LIABLE FOR ANY
- * DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES
- * (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES;
- * LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND
- * ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
- * (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF
- * THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
- *
- * This file incorporates work covered by the following copyright and
- * permission notice:
- *     The Synopsys DWC ETHER XGMAC Software Driver and documentation
- *     (hereinafter "Software") is an unsupported proprietary work of Synopsys,
- *     Inc. unless otherwise expressly agreed to in writing between Synopsys
- *     and you.
- *
- *     The Software IS NOT an item of Licensed Software or Licensed Product
- *     under any End User Software License Agreement or Agreement for Licensed
- *     Product with Synopsys or any supplement thereto.  Permission is hereby
- *     granted, free of charge, to any person obtaining a copy of this software
- *     annotated with this license and the Software, to deal in the Software
- *     without restriction, including without limitation the rights to use,
- *     copy, modify, merge, publish, distribute, sublicense, and/or sell copies
- *     of the Software, and to permit persons to whom the Software is furnished
- *     to do so, subject to the following conditions:
- *
- *     The above copyright notice and this permission notice shall be included
- *     in all copies or substantial portions of the Software.
- *
- *     THIS SOFTWARE IS BEING DISTRIBUTED BY SYNOPSYS SOLELY ON AN "AS IS"
- *     BASIS AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED
- *     TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A
- *     PARTICULAR PURPOSE ARE HEREBY DISCLAIMED. IN NO EVENT SHALL SYNOPSYS
- *     BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR
- *     CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF
- *     SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS
- *     INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN
- *     CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE)
- *     ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF
- *     THE POSSIBILITY OF SUCH DAMAGE.
+ * Copyright (c) 2014-2025, Advanced Micro Devices, Inc.
+ * Copyright (c) 2014, Synopsys, Inc.
+ * All rights reserved
  */
 
 #include <linux/module.h>
@@ -119,14 +10,16 @@
 #include <linux/kmod.h>
 #include <linux/mdio.h>
 #include <linux/phy.h>
+#include <linux/ethtool.h>
 
 #include "xgbe.h"
 #include "xgbe-common.h"
 
-#define XGBE_PHY_PORT_SPEED_100		BIT(0)
-#define XGBE_PHY_PORT_SPEED_1000	BIT(1)
-#define XGBE_PHY_PORT_SPEED_2500	BIT(2)
-#define XGBE_PHY_PORT_SPEED_10000	BIT(3)
+#define XGBE_PHY_PORT_SPEED_10		BIT(0)
+#define XGBE_PHY_PORT_SPEED_100		BIT(1)
+#define XGBE_PHY_PORT_SPEED_1000	BIT(2)
+#define XGBE_PHY_PORT_SPEED_2500	BIT(3)
+#define XGBE_PHY_PORT_SPEED_10000	BIT(4)
 
 #define XGBE_MUTEX_RELEASE		0x80000000
 
@@ -147,6 +40,14 @@
 /* Rate-change complete wait/retry count */
 #define XGBE_RATECHANGE_COUNT		500
 
+/* CDR delay values for KR support (in usec) */
+#define XGBE_CDR_DELAY_INIT		10000
+#define XGBE_CDR_DELAY_INC		10000
+#define XGBE_CDR_DELAY_MAX		100000
+
+/* RRC frequency during link status check */
+#define XGBE_RRC_FREQUENCY		10
+
 enum xgbe_port_mode {
 	XGBE_PORT_MODE_RSVD = 0,
 	XGBE_PORT_MODE_BACKPLANE,
@@ -157,6 +58,7 @@ enum xgbe_port_mode {
 	XGBE_PORT_MODE_10GBASE_T,
 	XGBE_PORT_MODE_10GBASE_R,
 	XGBE_PORT_MODE_SFP,
+	XGBE_PORT_MODE_BACKPLANE_NO_AUTONEG,
 	XGBE_PORT_MODE_MAX,
 };
 
@@ -179,6 +81,7 @@ enum xgbe_sfp_cable {
 	XGBE_SFP_CABLE_UNKNOWN = 0,
 	XGBE_SFP_CABLE_ACTIVE,
 	XGBE_SFP_CABLE_PASSIVE,
+	XGBE_SFP_CABLE_FIBER,
 };
 
 enum xgbe_sfp_base {
@@ -226,9 +129,7 @@ enum xgbe_sfp_speed {
 
 #define XGBE_SFP_BASE_BR			12
 #define XGBE_SFP_BASE_BR_1GBE_MIN		0x0a
-#define XGBE_SFP_BASE_BR_1GBE_MAX		0x0d
 #define XGBE_SFP_BASE_BR_10GBE_MIN		0x64
-#define XGBE_SFP_BASE_BR_10GBE_MAX		0x68
 
 #define XGBE_SFP_BASE_CU_CABLE_LEN		18
 
@@ -245,6 +146,10 @@ enum xgbe_sfp_speed {
 #define XGBE_SFP_BASE_VENDOR_SN			4
 #define XGBE_SFP_BASE_VENDOR_SN_LEN		16
 
+#define XGBE_SFP_EXTD_OPT1			1
+#define XGBE_SFP_EXTD_OPT1_RX_LOS		BIT(1)
+#define XGBE_SFP_EXTD_OPT1_TX_FAULT		BIT(3)
+
 #define XGBE_SFP_EXTD_DIAG			28
 #define XGBE_SFP_EXTD_DIAG_ADDR_CHANGE		BIT(2)
 
@@ -258,9 +163,20 @@ struct xgbe_sfp_eeprom {
 	u8 vendor[32];
 };
 
+#define XGBE_SFP_DIAGS_SUPPORTED(_x)			\
+	((_x)->extd[XGBE_SFP_EXTD_SFF_8472] &&		\
+	 !((_x)->extd[XGBE_SFP_EXTD_DIAG] & XGBE_SFP_EXTD_DIAG_ADDR_CHANGE))
+
+#define XGBE_SFP_EEPROM_BASE_LEN	256
+#define XGBE_SFP_EEPROM_DIAG_LEN	256
+#define XGBE_SFP_EEPROM_MAX		(XGBE_SFP_EEPROM_BASE_LEN +	\
+					 XGBE_SFP_EEPROM_DIAG_LEN)
+
 #define XGBE_BEL_FUSE_VENDOR	"BEL-FUSE        "
 #define XGBE_BEL_FUSE_PARTNO	"1GBT-SFP06      "
 
+#define XGBE_MOLEX_VENDOR	"Molex Inc.      "
+
 struct xgbe_sfp_ascii {
 	union {
 		char vendor[XGBE_SFP_BASE_VENDOR_NAME_LEN + 1];
@@ -315,8 +231,6 @@ struct xgbe_phy_data {
 
 	unsigned int mdio_addr;
 
-	unsigned int comm_owned;
-
 	/* SFP Support */
 	enum xgbe_sfp_comm sfp_comm;
 	unsigned int sfp_mux_address;
@@ -324,6 +238,7 @@ struct xgbe_phy_data {
 
 	unsigned int sfp_gpio_address;
 	unsigned int sfp_gpio_mask;
+	unsigned int sfp_gpio_inputs;
 	unsigned int sfp_gpio_rx_los;
 	unsigned int sfp_gpio_tx_fault;
 	unsigned int sfp_gpio_mod_absent;
@@ -332,7 +247,6 @@ struct xgbe_phy_data {
 	unsigned int sfp_rx_los;
 	unsigned int sfp_tx_fault;
 	unsigned int sfp_mod_absent;
-	unsigned int sfp_diags;
 	unsigned int sfp_changed;
 	unsigned int sfp_phy_avail;
 	unsigned int sfp_cable_len;
@@ -355,22 +269,24 @@ struct xgbe_phy_data {
 	unsigned int redrv_addr;
 	unsigned int redrv_lane;
 	unsigned int redrv_model;
+
+	/* KR AN support */
+	unsigned int phy_cdr_notrack;
+	unsigned int phy_cdr_delay;
 };
 
 /* I2C, MDIO and GPIO lines are muxed, so only one device at a time */
 static DEFINE_MUTEX(xgbe_phy_comm_lock);
 
 static enum xgbe_an_mode xgbe_phy_an_mode(struct xgbe_prv_data *pdata);
+static void xgbe_phy_rrc(struct xgbe_prv_data *pdata);
+static void xgbe_phy_perform_ratechange(struct xgbe_prv_data *pdata,
+					enum xgbe_mb_cmd cmd,
+					enum xgbe_mb_subcmd sub_cmd);
 
 static int xgbe_phy_i2c_xfer(struct xgbe_prv_data *pdata,
 			     struct xgbe_i2c_op *i2c_op)
 {
-	struct xgbe_phy_data *phy_data = pdata->phy_data;
-
-	/* Be sure we own the bus */
-	if (WARN_ON(!phy_data->comm_owned))
-		return -EIO;
-
 	return pdata->i2c_if.i2c_xfer(pdata, i2c_op);
 }
 
@@ -532,10 +448,6 @@ static int xgbe_phy_sfp_get_mux(struct xgbe_prv_data *pdata)
 
 static void xgbe_phy_put_comm_ownership(struct xgbe_prv_data *pdata)
 {
-	struct xgbe_phy_data *phy_data = pdata->phy_data;
-
-	phy_data->comm_owned = 0;
-
 	mutex_unlock(&xgbe_phy_comm_lock);
 }
 
@@ -545,9 +457,6 @@ static int xgbe_phy_get_comm_ownership(struct xgbe_prv_data *pdata)
 	unsigned long timeout;
 	unsigned int mutex_id;
 
-	if (phy_data->comm_owned)
-		return 0;
-
 	/* The I2C and MDIO/GPIO bus is multiplexed between multiple devices,
 	 * the driver needs to take the software mutex and then the hardware
 	 * mutexes before being able to use the busses.
@@ -576,7 +485,6 @@ static int xgbe_phy_get_comm_ownership(struct xgbe_prv_data *pdata)
 		XP_IOWRITE(pdata, XP_I2C_MUTEX, mutex_id);
 		XP_IOWRITE(pdata, XP_MDIO_MUTEX, mutex_id);
 
-		phy_data->comm_owned = 1;
 		return 0;
 	}
 
@@ -587,20 +495,27 @@ static int xgbe_phy_get_comm_ownership(struct xgbe_prv_data *pdata)
 	return -ETIMEDOUT;
 }
 
-static int xgbe_phy_mdio_mii_write(struct xgbe_prv_data *pdata, int addr,
-				   int reg, u16 val)
+static int xgbe_phy_mdio_mii_write_c22(struct xgbe_prv_data *pdata, int addr,
+				       int reg, u16 val)
 {
 	struct xgbe_phy_data *phy_data = pdata->phy_data;
 
-	if (reg & MII_ADDR_C45) {
-		if (phy_data->phydev_mode != XGBE_MDIO_MODE_CL45)
-			return -ENOTSUPP;
-	} else {
-		if (phy_data->phydev_mode != XGBE_MDIO_MODE_CL22)
-			return -ENOTSUPP;
-	}
+	if (phy_data->phydev_mode != XGBE_MDIO_MODE_CL22)
+		return -EOPNOTSUPP;
+
+	return pdata->hw_if.write_ext_mii_regs_c22(pdata, addr, reg, val);
+}
+
+static int xgbe_phy_mdio_mii_write_c45(struct xgbe_prv_data *pdata, int addr,
+				       int devad, int reg, u16 val)
+{
+	struct xgbe_phy_data *phy_data = pdata->phy_data;
+
+	if (phy_data->phydev_mode != XGBE_MDIO_MODE_CL45)
+		return -EOPNOTSUPP;
 
-	return pdata->hw_if.write_ext_mii_regs(pdata, addr, reg, val);
+	return pdata->hw_if.write_ext_mii_regs_c45(pdata, addr, devad,
+						   reg, val);
 }
 
 static int xgbe_phy_i2c_mii_write(struct xgbe_prv_data *pdata, int reg, u16 val)
@@ -625,7 +540,8 @@ static int xgbe_phy_i2c_mii_write(struct xgbe_prv_data *pdata, int reg, u16 val)
 	return ret;
 }
 
-static int xgbe_phy_mii_write(struct mii_bus *mii, int addr, int reg, u16 val)
+static int xgbe_phy_mii_write_c22(struct mii_bus *mii, int addr, int reg,
+				  u16 val)
 {
 	struct xgbe_prv_data *pdata = mii->priv;
 	struct xgbe_phy_data *phy_data = pdata->phy_data;
@@ -638,29 +554,58 @@ static int xgbe_phy_mii_write(struct mii_bus *mii, int addr, int reg, u16 val)
 	if (phy_data->conn_type == XGBE_CONN_TYPE_SFP)
 		ret = xgbe_phy_i2c_mii_write(pdata, reg, val);
 	else if (phy_data->conn_type & XGBE_CONN_TYPE_MDIO)
-		ret = xgbe_phy_mdio_mii_write(pdata, addr, reg, val);
+		ret = xgbe_phy_mdio_mii_write_c22(pdata, addr, reg, val);
 	else
-		ret = -ENOTSUPP;
+		ret = -EOPNOTSUPP;
 
 	xgbe_phy_put_comm_ownership(pdata);
 
 	return ret;
 }
 
-static int xgbe_phy_mdio_mii_read(struct xgbe_prv_data *pdata, int addr,
-				  int reg)
+static int xgbe_phy_mii_write_c45(struct mii_bus *mii, int addr, int devad,
+				  int reg, u16 val)
 {
+	struct xgbe_prv_data *pdata = mii->priv;
 	struct xgbe_phy_data *phy_data = pdata->phy_data;
+	int ret;
 
-	if (reg & MII_ADDR_C45) {
-		if (phy_data->phydev_mode != XGBE_MDIO_MODE_CL45)
-			return -ENOTSUPP;
-	} else {
-		if (phy_data->phydev_mode != XGBE_MDIO_MODE_CL22)
-			return -ENOTSUPP;
-	}
+	ret = xgbe_phy_get_comm_ownership(pdata);
+	if (ret)
+		return ret;
 
-	return pdata->hw_if.read_ext_mii_regs(pdata, addr, reg);
+	if (phy_data->conn_type == XGBE_CONN_TYPE_SFP)
+		ret = -EOPNOTSUPP;
+	else if (phy_data->conn_type & XGBE_CONN_TYPE_MDIO)
+		ret = xgbe_phy_mdio_mii_write_c45(pdata, addr, devad, reg, val);
+	else
+		ret = -EOPNOTSUPP;
+
+	xgbe_phy_put_comm_ownership(pdata);
+
+	return ret;
+}
+
+static int xgbe_phy_mdio_mii_read_c22(struct xgbe_prv_data *pdata, int addr,
+				      int reg)
+{
+	struct xgbe_phy_data *phy_data = pdata->phy_data;
+
+	if (phy_data->phydev_mode != XGBE_MDIO_MODE_CL22)
+		return -EOPNOTSUPP;
+
+	return pdata->hw_if.read_ext_mii_regs_c22(pdata, addr, reg);
+}
+
+static int xgbe_phy_mdio_mii_read_c45(struct xgbe_prv_data *pdata, int addr,
+				      int devad, int reg)
+{
+	struct xgbe_phy_data *phy_data = pdata->phy_data;
+
+	if (phy_data->phydev_mode != XGBE_MDIO_MODE_CL45)
+		return -EOPNOTSUPP;
+
+	return pdata->hw_if.read_ext_mii_regs_c45(pdata, addr, devad, reg);
 }
 
 static int xgbe_phy_i2c_mii_read(struct xgbe_prv_data *pdata, int reg)
@@ -685,7 +630,7 @@ static int xgbe_phy_i2c_mii_read(struct xgbe_prv_data *pdata, int reg)
 	return ret;
 }
 
-static int xgbe_phy_mii_read(struct mii_bus *mii, int addr, int reg)
+static int xgbe_phy_mii_read_c22(struct mii_bus *mii, int addr, int reg)
 {
 	struct xgbe_prv_data *pdata = mii->priv;
 	struct xgbe_phy_data *phy_data = pdata->phy_data;
@@ -698,7 +643,30 @@ static int xgbe_phy_mii_read(struct mii_bus *mii, int addr, int reg)
 	if (phy_data->conn_type == XGBE_CONN_TYPE_SFP)
 		ret = xgbe_phy_i2c_mii_read(pdata, reg);
 	else if (phy_data->conn_type & XGBE_CONN_TYPE_MDIO)
-		ret = xgbe_phy_mdio_mii_read(pdata, addr, reg);
+		ret = xgbe_phy_mdio_mii_read_c22(pdata, addr, reg);
+	else
+		ret = -EOPNOTSUPP;
+
+	xgbe_phy_put_comm_ownership(pdata);
+
+	return ret;
+}
+
+static int xgbe_phy_mii_read_c45(struct mii_bus *mii, int addr, int devad,
+				 int reg)
+{
+	struct xgbe_prv_data *pdata = mii->priv;
+	struct xgbe_phy_data *phy_data = pdata->phy_data;
+	int ret;
+
+	ret = xgbe_phy_get_comm_ownership(pdata);
+	if (ret)
+		return ret;
+
+	if (phy_data->conn_type == XGBE_CONN_TYPE_SFP)
+		ret = -EOPNOTSUPP;
+	else if (phy_data->conn_type & XGBE_CONN_TYPE_MDIO)
+		ret = xgbe_phy_mdio_mii_read_c45(pdata, addr, devad, reg);
 	else
 		ret = -ENOTSUPP;
 
@@ -709,25 +677,31 @@ static int xgbe_phy_mii_read(struct mii_bus *mii, int addr, int reg)
 
 static void xgbe_phy_sfp_phy_settings(struct xgbe_prv_data *pdata)
 {
+	struct ethtool_link_ksettings *lks = &pdata->phy.lks;
 	struct xgbe_phy_data *phy_data = pdata->phy_data;
 
+	if (!phy_data->sfp_mod_absent && !phy_data->sfp_changed)
+		return;
+
+	XGBE_ZERO_SUP(lks);
+
 	if (phy_data->sfp_mod_absent) {
 		pdata->phy.speed = SPEED_UNKNOWN;
 		pdata->phy.duplex = DUPLEX_UNKNOWN;
 		pdata->phy.autoneg = AUTONEG_ENABLE;
-		pdata->phy.advertising = pdata->phy.supported;
+		pdata->phy.pause_autoneg = AUTONEG_ENABLE;
+
+		XGBE_SET_SUP(lks, Autoneg);
+		XGBE_SET_SUP(lks, Pause);
+		XGBE_SET_SUP(lks, Asym_Pause);
+		XGBE_SET_SUP(lks, TP);
+		XGBE_SET_SUP(lks, FIBRE);
+
+		XGBE_LM_COPY(lks, advertising, lks, supported);
 
 		return;
 	}
 
-	pdata->phy.advertising &= ~ADVERTISED_Autoneg;
-	pdata->phy.advertising &= ~ADVERTISED_TP;
-	pdata->phy.advertising &= ~ADVERTISED_FIBRE;
-	pdata->phy.advertising &= ~ADVERTISED_100baseT_Full;
-	pdata->phy.advertising &= ~ADVERTISED_1000baseT_Full;
-	pdata->phy.advertising &= ~ADVERTISED_10000baseT_Full;
-	pdata->phy.advertising &= ~ADVERTISED_10000baseR_FEC;
-
 	switch (phy_data->sfp_base) {
 	case XGBE_SFP_BASE_1000_T:
 	case XGBE_SFP_BASE_1000_SX:
@@ -736,17 +710,58 @@ static void xgbe_phy_sfp_phy_settings(struct xgbe_prv_data *pdata)
 		pdata->phy.speed = SPEED_UNKNOWN;
 		pdata->phy.duplex = DUPLEX_UNKNOWN;
 		pdata->phy.autoneg = AUTONEG_ENABLE;
-		pdata->phy.advertising |= ADVERTISED_Autoneg;
+		pdata->phy.pause_autoneg = AUTONEG_ENABLE;
+		XGBE_SET_SUP(lks, Autoneg);
+		XGBE_SET_SUP(lks, Pause);
+		XGBE_SET_SUP(lks, Asym_Pause);
+		if (phy_data->sfp_base == XGBE_SFP_BASE_1000_T) {
+			if (phy_data->port_speeds & XGBE_PHY_PORT_SPEED_10)
+				XGBE_SET_SUP(lks, 10baseT_Full);
+			if (phy_data->port_speeds & XGBE_PHY_PORT_SPEED_100)
+				XGBE_SET_SUP(lks, 100baseT_Full);
+			if (phy_data->port_speeds & XGBE_PHY_PORT_SPEED_1000)
+				XGBE_SET_SUP(lks, 1000baseT_Full);
+		} else {
+			if (phy_data->port_speeds & XGBE_PHY_PORT_SPEED_1000)
+				XGBE_SET_SUP(lks, 1000baseX_Full);
+		}
 		break;
 	case XGBE_SFP_BASE_10000_SR:
 	case XGBE_SFP_BASE_10000_LR:
 	case XGBE_SFP_BASE_10000_LRM:
 	case XGBE_SFP_BASE_10000_ER:
 	case XGBE_SFP_BASE_10000_CR:
-	default:
 		pdata->phy.speed = SPEED_10000;
 		pdata->phy.duplex = DUPLEX_FULL;
 		pdata->phy.autoneg = AUTONEG_DISABLE;
+		pdata->phy.pause_autoneg = AUTONEG_DISABLE;
+		if (phy_data->port_speeds & XGBE_PHY_PORT_SPEED_10000) {
+			switch (phy_data->sfp_base) {
+			case XGBE_SFP_BASE_10000_SR:
+				XGBE_SET_SUP(lks, 10000baseSR_Full);
+				break;
+			case XGBE_SFP_BASE_10000_LR:
+				XGBE_SET_SUP(lks, 10000baseLR_Full);
+				break;
+			case XGBE_SFP_BASE_10000_LRM:
+				XGBE_SET_SUP(lks, 10000baseLRM_Full);
+				break;
+			case XGBE_SFP_BASE_10000_ER:
+				XGBE_SET_SUP(lks, 10000baseER_Full);
+				break;
+			case XGBE_SFP_BASE_10000_CR:
+				XGBE_SET_SUP(lks, 10000baseCR_Full);
+				break;
+			default:
+				break;
+			}
+		}
+		break;
+	default:
+		pdata->phy.speed = SPEED_UNKNOWN;
+		pdata->phy.duplex = DUPLEX_UNKNOWN;
+		pdata->phy.autoneg = AUTONEG_DISABLE;
+		pdata->phy.pause_autoneg = AUTONEG_DISABLE;
 		break;
 	}
 
@@ -754,60 +769,35 @@ static void xgbe_phy_sfp_phy_settings(struct xgbe_prv_data *pdata)
 	case XGBE_SFP_BASE_1000_T:
 	case XGBE_SFP_BASE_1000_CX:
 	case XGBE_SFP_BASE_10000_CR:
-		pdata->phy.advertising |= ADVERTISED_TP;
+		XGBE_SET_SUP(lks, TP);
 		break;
 	default:
-		pdata->phy.advertising |= ADVERTISED_FIBRE;
-	}
-
-	switch (phy_data->sfp_speed) {
-	case XGBE_SFP_SPEED_100_1000:
-		if (phy_data->port_speeds & XGBE_PHY_PORT_SPEED_100)
-			pdata->phy.advertising |= ADVERTISED_100baseT_Full;
-		if (phy_data->port_speeds & XGBE_PHY_PORT_SPEED_1000)
-			pdata->phy.advertising |= ADVERTISED_1000baseT_Full;
+		XGBE_SET_SUP(lks, FIBRE);
 		break;
-	case XGBE_SFP_SPEED_1000:
-		if (phy_data->port_speeds & XGBE_PHY_PORT_SPEED_1000)
-			pdata->phy.advertising |= ADVERTISED_1000baseT_Full;
-		break;
-	case XGBE_SFP_SPEED_10000:
-		if (phy_data->port_speeds & XGBE_PHY_PORT_SPEED_10000)
-			pdata->phy.advertising |= ADVERTISED_10000baseT_Full;
-		break;
-	default:
-		/* Choose the fastest supported speed */
-		if (phy_data->port_speeds & XGBE_PHY_PORT_SPEED_10000)
-			pdata->phy.advertising |= ADVERTISED_10000baseT_Full;
-		else if (phy_data->port_speeds & XGBE_PHY_PORT_SPEED_1000)
-			pdata->phy.advertising |= ADVERTISED_1000baseT_Full;
-		else if (phy_data->port_speeds & XGBE_PHY_PORT_SPEED_100)
-			pdata->phy.advertising |= ADVERTISED_100baseT_Full;
 	}
+
+	XGBE_LM_COPY(lks, advertising, lks, supported);
 }
 
 static bool xgbe_phy_sfp_bit_rate(struct xgbe_sfp_eeprom *sfp_eeprom,
 				  enum xgbe_sfp_speed sfp_speed)
 {
-	u8 *sfp_base, min, max;
+	u8 *sfp_base, min;
 
 	sfp_base = sfp_eeprom->base;
 
 	switch (sfp_speed) {
 	case XGBE_SFP_SPEED_1000:
 		min = XGBE_SFP_BASE_BR_1GBE_MIN;
-		max = XGBE_SFP_BASE_BR_1GBE_MAX;
 		break;
 	case XGBE_SFP_SPEED_10000:
 		min = XGBE_SFP_BASE_BR_10GBE_MIN;
-		max = XGBE_SFP_BASE_BR_10GBE_MAX;
 		break;
 	default:
 		return false;
 	}
 
-	return ((sfp_base[XGBE_SFP_BASE_BR] >= min) &&
-		(sfp_base[XGBE_SFP_BASE_BR] <= max));
+	return sfp_base[XGBE_SFP_BASE_BR] >= min;
 }
 
 static void xgbe_phy_free_phy_device(struct xgbe_prv_data *pdata)
@@ -827,6 +817,9 @@ static bool xgbe_phy_finisar_phy_quirks(struct xgbe_prv_data *pdata)
 	struct xgbe_phy_data *phy_data = pdata->phy_data;
 	unsigned int phy_id = phy_data->phydev->phy_id;
 
+	if (phy_data->port_mode != XGBE_PORT_MODE_SFP)
+		return false;
+
 	if ((phy_id & 0xfffffff0) != 0x01ff0cc0)
 		return false;
 
@@ -842,9 +835,9 @@ static bool xgbe_phy_finisar_phy_quirks(struct xgbe_prv_data *pdata)
 	phy_write(phy_data->phydev, 0x04, 0x0d01);
 	phy_write(phy_data->phydev, 0x00, 0x9140);
 
-	phy_data->phydev->supported = PHY_GBIT_FEATURES;
-	phy_data->phydev->supported |= SUPPORTED_Pause | SUPPORTED_Asym_Pause;
-	phy_data->phydev->advertising = phy_data->phydev->supported;
+	linkmode_copy(phy_data->phydev->supported, PHY_GBIT_FEATURES);
+
+	phy_support_asym_pause(phy_data->phydev);
 
 	netif_dbg(pdata, drv, pdata->netdev,
 		  "Finisar PHY quirk in place\n");
@@ -852,14 +845,92 @@ static bool xgbe_phy_finisar_phy_quirks(struct xgbe_prv_data *pdata)
 	return true;
 }
 
+static bool xgbe_phy_belfuse_phy_quirks(struct xgbe_prv_data *pdata)
+{
+	struct xgbe_phy_data *phy_data = pdata->phy_data;
+	struct xgbe_sfp_eeprom *sfp_eeprom = &phy_data->sfp_eeprom;
+	unsigned int phy_id = phy_data->phydev->phy_id;
+	int reg;
+
+	if (phy_data->port_mode != XGBE_PORT_MODE_SFP)
+		return false;
+
+	if (memcmp(&sfp_eeprom->base[XGBE_SFP_BASE_VENDOR_NAME],
+		   XGBE_BEL_FUSE_VENDOR, XGBE_SFP_BASE_VENDOR_NAME_LEN))
+		return false;
+
+	/* For Bel-Fuse, use the extra AN flag */
+	pdata->an_again = 1;
+
+	if (memcmp(&sfp_eeprom->base[XGBE_SFP_BASE_VENDOR_PN],
+		   XGBE_BEL_FUSE_PARTNO, XGBE_SFP_BASE_VENDOR_PN_LEN))
+		return false;
+
+	if ((phy_id & 0xfffffff0) != 0x03625d10)
+		return false;
+
+	/* Reset PHY - wait for self-clearing reset bit to clear */
+	genphy_soft_reset(phy_data->phydev);
+
+	/* Disable RGMII mode */
+	phy_write(phy_data->phydev, 0x18, 0x7007);
+	reg = phy_read(phy_data->phydev, 0x18);
+	phy_write(phy_data->phydev, 0x18, reg & ~0x0080);
+
+	/* Enable fiber register bank */
+	phy_write(phy_data->phydev, 0x1c, 0x7c00);
+	reg = phy_read(phy_data->phydev, 0x1c);
+	reg &= 0x03ff;
+	reg &= ~0x0001;
+	phy_write(phy_data->phydev, 0x1c, 0x8000 | 0x7c00 | reg | 0x0001);
+
+	/* Power down SerDes */
+	reg = phy_read(phy_data->phydev, 0x00);
+	phy_write(phy_data->phydev, 0x00, reg | 0x00800);
+
+	/* Configure SGMII-to-Copper mode */
+	phy_write(phy_data->phydev, 0x1c, 0x7c00);
+	reg = phy_read(phy_data->phydev, 0x1c);
+	reg &= 0x03ff;
+	reg &= ~0x0006;
+	phy_write(phy_data->phydev, 0x1c, 0x8000 | 0x7c00 | reg | 0x0004);
+
+	/* Power up SerDes */
+	reg = phy_read(phy_data->phydev, 0x00);
+	phy_write(phy_data->phydev, 0x00, reg & ~0x00800);
+
+	/* Enable copper register bank */
+	phy_write(phy_data->phydev, 0x1c, 0x7c00);
+	reg = phy_read(phy_data->phydev, 0x1c);
+	reg &= 0x03ff;
+	reg &= ~0x0001;
+	phy_write(phy_data->phydev, 0x1c, 0x8000 | 0x7c00 | reg);
+
+	/* Power up SerDes */
+	reg = phy_read(phy_data->phydev, 0x00);
+	phy_write(phy_data->phydev, 0x00, reg & ~0x00800);
+
+	linkmode_copy(phy_data->phydev->supported, PHY_GBIT_FEATURES);
+	phy_support_asym_pause(phy_data->phydev);
+
+	netif_dbg(pdata, drv, pdata->netdev,
+		  "BelFuse PHY quirk in place\n");
+
+	return true;
+}
+
 static void xgbe_phy_external_phy_quirks(struct xgbe_prv_data *pdata)
 {
+	if (xgbe_phy_belfuse_phy_quirks(pdata))
+		return;
+
 	if (xgbe_phy_finisar_phy_quirks(pdata))
 		return;
 }
 
 static int xgbe_phy_find_phy_device(struct xgbe_prv_data *pdata)
 {
+	struct ethtool_link_ksettings *lks = &pdata->phy.lks;
 	struct xgbe_phy_data *phy_data = pdata->phy_data;
 	struct phy_device *phydev;
 	int ret;
@@ -868,6 +939,9 @@ static int xgbe_phy_find_phy_device(struct xgbe_prv_data *pdata)
 	if (phy_data->phydev)
 		return 0;
 
+	/* Clear the extra AN flag */
+	pdata->an_again = 0;
+
 	/* Check for the use of an external PHY */
 	if (phy_data->phydev_mode == XGBE_MDIO_MODE_NONE)
 		return 0;
@@ -917,7 +991,9 @@ static int xgbe_phy_find_phy_device(struct xgbe_prv_data *pdata)
 	phy_data->phydev = phydev;
 
 	xgbe_phy_external_phy_quirks(pdata);
-	phydev->advertising &= pdata->phy.advertising;
+
+	linkmode_and(phydev->advertising, phydev->advertising,
+		     lks->link_modes.advertising);
 
 	phy_start_aneg(phy_data->phydev);
 
@@ -946,32 +1022,44 @@ static void xgbe_phy_sfp_external_phy(struct xgbe_prv_data *pdata)
 	phy_data->sfp_phy_avail = 1;
 }
 
-static bool xgbe_phy_belfuse_parse_quirks(struct xgbe_prv_data *pdata)
+static bool xgbe_phy_check_sfp_rx_los(struct xgbe_phy_data *phy_data)
 {
-	struct xgbe_phy_data *phy_data = pdata->phy_data;
-	struct xgbe_sfp_eeprom *sfp_eeprom = &phy_data->sfp_eeprom;
+	u8 *sfp_extd = phy_data->sfp_eeprom.extd;
 
-	if (memcmp(&sfp_eeprom->base[XGBE_SFP_BASE_VENDOR_NAME],
-		   XGBE_BEL_FUSE_VENDOR, XGBE_SFP_BASE_VENDOR_NAME_LEN))
+	if (!(sfp_extd[XGBE_SFP_EXTD_OPT1] & XGBE_SFP_EXTD_OPT1_RX_LOS))
 		return false;
 
-	if (!memcmp(&sfp_eeprom->base[XGBE_SFP_BASE_VENDOR_PN],
-		    XGBE_BEL_FUSE_PARTNO, XGBE_SFP_BASE_VENDOR_PN_LEN)) {
-		phy_data->sfp_base = XGBE_SFP_BASE_1000_SX;
-		phy_data->sfp_cable = XGBE_SFP_CABLE_ACTIVE;
-		phy_data->sfp_speed = XGBE_SFP_SPEED_1000;
-		if (phy_data->sfp_changed)
-			netif_dbg(pdata, drv, pdata->netdev,
-				  "Bel-Fuse SFP quirk in place\n");
+	if (phy_data->sfp_gpio_mask & XGBE_GPIO_NO_RX_LOS)
+		return false;
+
+	if (phy_data->sfp_gpio_inputs & (1 << phy_data->sfp_gpio_rx_los))
 		return true;
-	}
 
 	return false;
 }
 
-static bool xgbe_phy_sfp_parse_quirks(struct xgbe_prv_data *pdata)
+static bool xgbe_phy_check_sfp_tx_fault(struct xgbe_phy_data *phy_data)
 {
-	if (xgbe_phy_belfuse_parse_quirks(pdata))
+	u8 *sfp_extd = phy_data->sfp_eeprom.extd;
+
+	if (!(sfp_extd[XGBE_SFP_EXTD_OPT1] & XGBE_SFP_EXTD_OPT1_TX_FAULT))
+		return false;
+
+	if (phy_data->sfp_gpio_mask & XGBE_GPIO_NO_TX_FAULT)
+		return false;
+
+	if (phy_data->sfp_gpio_inputs & (1 << phy_data->sfp_gpio_tx_fault))
+		return true;
+
+	return false;
+}
+
+static bool xgbe_phy_check_sfp_mod_absent(struct xgbe_phy_data *phy_data)
+{
+	if (phy_data->sfp_gpio_mask & XGBE_GPIO_NO_MOD_ABSENT)
+		return false;
+
+	if (phy_data->sfp_gpio_inputs & (1 << phy_data->sfp_gpio_mod_absent))
 		return true;
 
 	return false;
@@ -991,19 +1079,25 @@ static void xgbe_phy_sfp_parse_eeprom(struct xgbe_prv_data *pdata)
 	if (sfp_base[XGBE_SFP_BASE_EXT_ID] != XGBE_SFP_EXT_ID_SFP)
 		return;
 
-	if (xgbe_phy_sfp_parse_quirks(pdata))
-		return;
+	/* Update transceiver signals (eeprom extd/options) */
+	phy_data->sfp_tx_fault = xgbe_phy_check_sfp_tx_fault(phy_data);
+	phy_data->sfp_rx_los = xgbe_phy_check_sfp_rx_los(phy_data);
 
-	/* Assume ACTIVE cable unless told it is PASSIVE */
+	/* Assume FIBER cable unless told otherwise */
 	if (sfp_base[XGBE_SFP_BASE_CABLE] & XGBE_SFP_BASE_CABLE_PASSIVE) {
 		phy_data->sfp_cable = XGBE_SFP_CABLE_PASSIVE;
 		phy_data->sfp_cable_len = sfp_base[XGBE_SFP_BASE_CU_CABLE_LEN];
-	} else {
+	} else if (sfp_base[XGBE_SFP_BASE_CABLE] & XGBE_SFP_BASE_CABLE_ACTIVE) {
 		phy_data->sfp_cable = XGBE_SFP_CABLE_ACTIVE;
+	} else {
+		phy_data->sfp_cable = XGBE_SFP_CABLE_FIBER;
 	}
 
 	/* Determine the type of SFP */
-	if (sfp_base[XGBE_SFP_BASE_10GBE_CC] & XGBE_SFP_BASE_10GBE_CC_SR)
+	if (phy_data->sfp_cable != XGBE_SFP_CABLE_FIBER &&
+	    xgbe_phy_sfp_bit_rate(sfp_eeprom, XGBE_SFP_SPEED_10000))
+		phy_data->sfp_base = XGBE_SFP_BASE_10000_CR;
+	else if (sfp_base[XGBE_SFP_BASE_10GBE_CC] & XGBE_SFP_BASE_10GBE_CC_SR)
 		phy_data->sfp_base = XGBE_SFP_BASE_10000_SR;
 	else if (sfp_base[XGBE_SFP_BASE_10GBE_CC] & XGBE_SFP_BASE_10GBE_CC_LR)
 		phy_data->sfp_base = XGBE_SFP_BASE_10000_LR;
@@ -1019,9 +1113,6 @@ static void xgbe_phy_sfp_parse_eeprom(struct xgbe_prv_data *pdata)
 		phy_data->sfp_base = XGBE_SFP_BASE_1000_CX;
 	else if (sfp_base[XGBE_SFP_BASE_1GBE_CC] & XGBE_SFP_BASE_1GBE_CC_T)
 		phy_data->sfp_base = XGBE_SFP_BASE_1000_T;
-	else if ((phy_data->sfp_cable == XGBE_SFP_CABLE_PASSIVE) &&
-		 xgbe_phy_sfp_bit_rate(sfp_eeprom, XGBE_SFP_SPEED_10000))
-		phy_data->sfp_base = XGBE_SFP_BASE_10000_CR;
 
 	switch (phy_data->sfp_base) {
 	case XGBE_SFP_BASE_1000_T:
@@ -1083,7 +1174,7 @@ static bool xgbe_phy_sfp_verify_eeprom(u8 cc_in, u8 *buf, unsigned int len)
 	for (cc = 0; len; buf++, len--)
 		cc += *buf;
 
-	return (cc == cc_in) ? true : false;
+	return cc == cc_in;
 }
 
 static int xgbe_phy_sfp_read_eeprom(struct xgbe_prv_data *pdata)
@@ -1095,7 +1186,8 @@ static int xgbe_phy_sfp_read_eeprom(struct xgbe_prv_data *pdata)
 
 	ret = xgbe_phy_sfp_get_mux(pdata);
 	if (ret) {
-		netdev_err(pdata->netdev, "I2C error setting SFP MUX\n");
+		dev_err_once(pdata->dev, "%s: I2C error setting SFP MUX\n",
+			     netdev_name(pdata->netdev));
 		return ret;
 	}
 
@@ -1105,7 +1197,8 @@ static int xgbe_phy_sfp_read_eeprom(struct xgbe_prv_data *pdata)
 				&eeprom_addr, sizeof(eeprom_addr),
 				&sfp_eeprom, sizeof(sfp_eeprom));
 	if (ret) {
-		netdev_err(pdata->netdev, "I2C error reading SFP EEPROM\n");
+		dev_err_once(pdata->dev, "%s: I2C error reading SFP EEPROM\n",
+			     netdev_name(pdata->netdev));
 		goto put;
 	}
 
@@ -1133,13 +1226,6 @@ static int xgbe_phy_sfp_read_eeprom(struct xgbe_prv_data *pdata)
 
 		memcpy(&phy_data->sfp_eeprom, &sfp_eeprom, sizeof(sfp_eeprom));
 
-		if (sfp_eeprom.extd[XGBE_SFP_EXTD_SFF_8472]) {
-			u8 diag_type = sfp_eeprom.extd[XGBE_SFP_EXTD_DIAG];
-
-			if (!(diag_type & XGBE_SFP_EXTD_DIAG_ADDR_CHANGE))
-				phy_data->sfp_diags = 1;
-		}
-
 		xgbe_phy_free_phy_device(pdata);
 	} else {
 		phy_data->sfp_changed = 0;
@@ -1154,7 +1240,6 @@ put:
 static void xgbe_phy_sfp_signals(struct xgbe_prv_data *pdata)
 {
 	struct xgbe_phy_data *phy_data = pdata->phy_data;
-	unsigned int gpio_input;
 	u8 gpio_reg, gpio_ports[2];
 	int ret;
 
@@ -1164,27 +1249,14 @@ static void xgbe_phy_sfp_signals(struct xgbe_prv_data *pdata)
 				&gpio_reg, sizeof(gpio_reg),
 				gpio_ports, sizeof(gpio_ports));
 	if (ret) {
-		netdev_err(pdata->netdev, "I2C error reading SFP GPIOs\n");
+		dev_err_once(pdata->dev, "%s: I2C error reading SFP GPIOs\n",
+			     netdev_name(pdata->netdev));
 		return;
 	}
 
-	gpio_input = (gpio_ports[1] << 8) | gpio_ports[0];
+	phy_data->sfp_gpio_inputs = (gpio_ports[1] << 8) | gpio_ports[0];
 
-	if (phy_data->sfp_gpio_mask & XGBE_GPIO_NO_MOD_ABSENT) {
-		/* No GPIO, just assume the module is present for now */
-		phy_data->sfp_mod_absent = 0;
-	} else {
-		if (!(gpio_input & (1 << phy_data->sfp_gpio_mod_absent)))
-			phy_data->sfp_mod_absent = 0;
-	}
-
-	if (!(phy_data->sfp_gpio_mask & XGBE_GPIO_NO_RX_LOS) &&
-	    (gpio_input & (1 << phy_data->sfp_gpio_rx_los)))
-		phy_data->sfp_rx_los = 1;
-
-	if (!(phy_data->sfp_gpio_mask & XGBE_GPIO_NO_TX_FAULT) &&
-	    (gpio_input & (1 << phy_data->sfp_gpio_tx_fault)))
-		phy_data->sfp_tx_fault = 1;
+	phy_data->sfp_mod_absent = xgbe_phy_check_sfp_mod_absent(phy_data);
 }
 
 static void xgbe_phy_sfp_mod_absent(struct xgbe_prv_data *pdata)
@@ -1203,7 +1275,6 @@ static void xgbe_phy_sfp_reset(struct xgbe_phy_data *phy_data)
 	phy_data->sfp_rx_los = 0;
 	phy_data->sfp_tx_fault = 0;
 	phy_data->sfp_mod_absent = 1;
-	phy_data->sfp_diags = 0;
 	phy_data->sfp_base = XGBE_SFP_BASE_UNKNOWN;
 	phy_data->sfp_cable = XGBE_SFP_CABLE_UNKNOWN;
 	phy_data->sfp_speed = XGBE_SFP_SPEED_UNKNOWN;
@@ -1246,8 +1317,133 @@ put:
 	xgbe_phy_put_comm_ownership(pdata);
 }
 
+static int xgbe_phy_module_eeprom(struct xgbe_prv_data *pdata,
+				  struct ethtool_eeprom *eeprom, u8 *data)
+{
+	struct xgbe_phy_data *phy_data = pdata->phy_data;
+	u8 eeprom_addr, eeprom_data[XGBE_SFP_EEPROM_MAX];
+	struct xgbe_sfp_eeprom *sfp_eeprom;
+	unsigned int i, j, rem;
+	int ret;
+
+	rem = eeprom->len;
+
+	if (!eeprom->len) {
+		ret = -EINVAL;
+		goto done;
+	}
+
+	if ((eeprom->offset + eeprom->len) > XGBE_SFP_EEPROM_MAX) {
+		ret = -EINVAL;
+		goto done;
+	}
+
+	if (phy_data->port_mode != XGBE_PORT_MODE_SFP) {
+		ret = -ENXIO;
+		goto done;
+	}
+
+	if (!netif_running(pdata->netdev)) {
+		ret = -EIO;
+		goto done;
+	}
+
+	if (phy_data->sfp_mod_absent) {
+		ret = -EIO;
+		goto done;
+	}
+
+	ret = xgbe_phy_get_comm_ownership(pdata);
+	if (ret) {
+		ret = -EIO;
+		goto done;
+	}
+
+	ret = xgbe_phy_sfp_get_mux(pdata);
+	if (ret) {
+		netdev_err(pdata->netdev, "I2C error setting SFP MUX\n");
+		ret = -EIO;
+		goto put_own;
+	}
+
+	/* Read the SFP serial ID eeprom */
+	eeprom_addr = 0;
+	ret = xgbe_phy_i2c_read(pdata, XGBE_SFP_SERIAL_ID_ADDRESS,
+				&eeprom_addr, sizeof(eeprom_addr),
+				eeprom_data, XGBE_SFP_EEPROM_BASE_LEN);
+	if (ret) {
+		netdev_err(pdata->netdev,
+			   "I2C error reading SFP EEPROM\n");
+		ret = -EIO;
+		goto put_mux;
+	}
+
+	sfp_eeprom = (struct xgbe_sfp_eeprom *)eeprom_data;
+
+	if (XGBE_SFP_DIAGS_SUPPORTED(sfp_eeprom)) {
+		/* Read the SFP diagnostic eeprom */
+		eeprom_addr = 0;
+		ret = xgbe_phy_i2c_read(pdata, XGBE_SFP_DIAG_INFO_ADDRESS,
+					&eeprom_addr, sizeof(eeprom_addr),
+					eeprom_data + XGBE_SFP_EEPROM_BASE_LEN,
+					XGBE_SFP_EEPROM_DIAG_LEN);
+		if (ret) {
+			netdev_err(pdata->netdev,
+				   "I2C error reading SFP DIAGS\n");
+			ret = -EIO;
+			goto put_mux;
+		}
+	}
+
+	for (i = 0, j = eeprom->offset; i < eeprom->len; i++, j++) {
+		if ((j >= XGBE_SFP_EEPROM_BASE_LEN) &&
+		    !XGBE_SFP_DIAGS_SUPPORTED(sfp_eeprom))
+			break;
+
+		data[i] = eeprom_data[j];
+		rem--;
+	}
+
+put_mux:
+	xgbe_phy_sfp_put_mux(pdata);
+
+put_own:
+	xgbe_phy_put_comm_ownership(pdata);
+
+done:
+	eeprom->len -= rem;
+
+	return ret;
+}
+
+static int xgbe_phy_module_info(struct xgbe_prv_data *pdata,
+				struct ethtool_modinfo *modinfo)
+{
+	struct xgbe_phy_data *phy_data = pdata->phy_data;
+
+	if (phy_data->port_mode != XGBE_PORT_MODE_SFP)
+		return -ENXIO;
+
+	if (!netif_running(pdata->netdev))
+		return -EIO;
+
+	if (phy_data->sfp_mod_absent)
+		return -EIO;
+
+	if (XGBE_SFP_DIAGS_SUPPORTED(&phy_data->sfp_eeprom)) {
+		modinfo->type = ETH_MODULE_SFF_8472;
+		modinfo->eeprom_len = ETH_MODULE_SFF_8472_LEN;
+	} else {
+		modinfo->type = ETH_MODULE_SFF_8079;
+		modinfo->eeprom_len = ETH_MODULE_SFF_8079_LEN;
+	}
+
+	return 0;
+}
+
 static void xgbe_phy_phydev_flowctrl(struct xgbe_prv_data *pdata)
 {
+	struct ethtool_link_ksettings *lks = &pdata->phy.lks;
 	struct xgbe_phy_data *phy_data = pdata->phy_data;
 	u16 lcl_adv = 0, rmt_adv = 0;
 	u8 fc;
@@ -1258,17 +1454,14 @@ static void xgbe_phy_phydev_flowctrl(struct xgbe_prv_data *pdata)
 	if (!phy_data->phydev)
 		return;
 
-	if (phy_data->phydev->advertising & ADVERTISED_Pause)
-		lcl_adv |= ADVERTISE_PAUSE_CAP;
-	if (phy_data->phydev->advertising & ADVERTISED_Asym_Pause)
-		lcl_adv |= ADVERTISE_PAUSE_ASYM;
+	lcl_adv = linkmode_adv_to_lcl_adv_t(phy_data->phydev->advertising);
 
 	if (phy_data->phydev->pause) {
-		pdata->phy.lp_advertising |= ADVERTISED_Pause;
+		XGBE_SET_LP_ADV(lks, Pause);
 		rmt_adv |= LPA_PAUSE_CAP;
 	}
 	if (phy_data->phydev->asym_pause) {
-		pdata->phy.lp_advertising |= ADVERTISED_Asym_Pause;
+		XGBE_SET_LP_ADV(lks, Asym_Pause);
 		rmt_adv |= LPA_PAUSE_ASYM;
 	}
 
@@ -1281,33 +1474,44 @@ static void xgbe_phy_phydev_flowctrl(struct xgbe_prv_data *pdata)
 
 static enum xgbe_mode xgbe_phy_an37_sgmii_outcome(struct xgbe_prv_data *pdata)
 {
+	struct ethtool_link_ksettings *lks = &pdata->phy.lks;
 	enum xgbe_mode mode;
 
-	pdata->phy.lp_advertising |= ADVERTISED_Autoneg;
-	pdata->phy.lp_advertising |= ADVERTISED_TP;
+	XGBE_SET_LP_ADV(lks, Autoneg);
+	XGBE_SET_LP_ADV(lks, TP);
 
 	/* Use external PHY to determine flow control */
 	if (pdata->phy.pause_autoneg)
 		xgbe_phy_phydev_flowctrl(pdata);
 
 	switch (pdata->an_status & XGBE_SGMII_AN_LINK_SPEED) {
+	case XGBE_SGMII_AN_LINK_SPEED_10:
+		if (pdata->an_status & XGBE_SGMII_AN_LINK_DUPLEX) {
+			XGBE_SET_LP_ADV(lks, 10baseT_Full);
+			mode = XGBE_MODE_SGMII_10;
+		} else {
+			/* Half-duplex not supported */
+			XGBE_SET_LP_ADV(lks, 10baseT_Half);
+			mode = XGBE_MODE_UNKNOWN;
+		}
+		break;
 	case XGBE_SGMII_AN_LINK_SPEED_100:
 		if (pdata->an_status & XGBE_SGMII_AN_LINK_DUPLEX) {
-			pdata->phy.lp_advertising |= ADVERTISED_100baseT_Full;
+			XGBE_SET_LP_ADV(lks, 100baseT_Full);
 			mode = XGBE_MODE_SGMII_100;
 		} else {
 			/* Half-duplex not supported */
-			pdata->phy.lp_advertising |= ADVERTISED_100baseT_Half;
+			XGBE_SET_LP_ADV(lks, 100baseT_Half);
 			mode = XGBE_MODE_UNKNOWN;
 		}
 		break;
 	case XGBE_SGMII_AN_LINK_SPEED_1000:
 		if (pdata->an_status & XGBE_SGMII_AN_LINK_DUPLEX) {
-			pdata->phy.lp_advertising |= ADVERTISED_1000baseT_Full;
+			XGBE_SET_LP_ADV(lks, 1000baseT_Full);
 			mode = XGBE_MODE_SGMII_1000;
 		} else {
 			/* Half-duplex not supported */
-			pdata->phy.lp_advertising |= ADVERTISED_1000baseT_Half;
+			XGBE_SET_LP_ADV(lks, 1000baseT_Half);
 			mode = XGBE_MODE_UNKNOWN;
 		}
 		break;
@@ -1320,19 +1524,20 @@ static enum xgbe_mode xgbe_phy_an37_sgmii_outcome(struct xgbe_prv_data *pdata)
 
 static enum xgbe_mode xgbe_phy_an37_outcome(struct xgbe_prv_data *pdata)
 {
+	struct ethtool_link_ksettings *lks = &pdata->phy.lks;
 	enum xgbe_mode mode;
 	unsigned int ad_reg, lp_reg;
 
-	pdata->phy.lp_advertising |= ADVERTISED_Autoneg;
-	pdata->phy.lp_advertising |= ADVERTISED_FIBRE;
+	XGBE_SET_LP_ADV(lks, Autoneg);
+	XGBE_SET_LP_ADV(lks, FIBRE);
 
 	/* Compare Advertisement and Link Partner register */
 	ad_reg = XMDIO_READ(pdata, MDIO_MMD_VEND2, MDIO_VEND2_AN_ADVERTISE);
 	lp_reg = XMDIO_READ(pdata, MDIO_MMD_VEND2, MDIO_VEND2_AN_LP_ABILITY);
 	if (lp_reg & 0x100)
-		pdata->phy.lp_advertising |= ADVERTISED_Pause;
+		XGBE_SET_LP_ADV(lks, Pause);
 	if (lp_reg & 0x80)
-		pdata->phy.lp_advertising |= ADVERTISED_Asym_Pause;
+		XGBE_SET_LP_ADV(lks, Asym_Pause);
 
 	if (pdata->phy.pause_autoneg) {
 		/* Set flow control based on auto-negotiation result */
@@ -1350,10 +1555,8 @@ static enum xgbe_mode xgbe_phy_an37_outcome(struct xgbe_prv_data *pdata)
 		}
 	}
 
-	if (lp_reg & 0x40)
-		pdata->phy.lp_advertising |= ADVERTISED_1000baseT_Half;
 	if (lp_reg & 0x20)
-		pdata->phy.lp_advertising |= ADVERTISED_1000baseT_Full;
+		XGBE_SET_LP_ADV(lks, 1000baseX_Full);
 
 	/* Half duplex is not supported */
 	ad_reg &= lp_reg;
@@ -1364,12 +1567,13 @@ static enum xgbe_mode xgbe_phy_an37_outcome(struct xgbe_prv_data *pdata)
 
 static enum xgbe_mode xgbe_phy_an73_redrv_outcome(struct xgbe_prv_data *pdata)
 {
+	struct ethtool_link_ksettings *lks = &pdata->phy.lks;
 	struct xgbe_phy_data *phy_data = pdata->phy_data;
 	enum xgbe_mode mode;
 	unsigned int ad_reg, lp_reg;
 
-	pdata->phy.lp_advertising |= ADVERTISED_Autoneg;
-	pdata->phy.lp_advertising |= ADVERTISED_Backplane;
+	XGBE_SET_LP_ADV(lks, Autoneg);
+	XGBE_SET_LP_ADV(lks, Backplane);
 
 	/* Use external PHY to determine flow control */
 	if (pdata->phy.pause_autoneg)
@@ -1379,14 +1583,15 @@ static enum xgbe_mode xgbe_phy_an73_redrv_outcome(struct xgbe_prv_data *pdata)
 	ad_reg = XMDIO_READ(pdata, MDIO_MMD_AN, MDIO_AN_ADVERTISE + 1);
 	lp_reg = XMDIO_READ(pdata, MDIO_MMD_AN, MDIO_AN_LPA + 1);
 	if (lp_reg & 0x80)
-		pdata->phy.lp_advertising |= ADVERTISED_10000baseKR_Full;
+		XGBE_SET_LP_ADV(lks, 10000baseKR_Full);
 	if (lp_reg & 0x20)
-		pdata->phy.lp_advertising |= ADVERTISED_1000baseKX_Full;
+		XGBE_SET_LP_ADV(lks, 1000baseKX_Full);
 
 	ad_reg &= lp_reg;
 	if (ad_reg & 0x80) {
 		switch (phy_data->port_mode) {
 		case XGBE_PORT_MODE_BACKPLANE:
+		case XGBE_PORT_MODE_BACKPLANE_NO_AUTONEG:
 			mode = XGBE_MODE_KR;
 			break;
 		default:
@@ -1396,6 +1601,7 @@ static enum xgbe_mode xgbe_phy_an73_redrv_outcome(struct xgbe_prv_data *pdata)
 	} else if (ad_reg & 0x20) {
 		switch (phy_data->port_mode) {
 		case XGBE_PORT_MODE_BACKPLANE:
+		case XGBE_PORT_MODE_BACKPLANE_NO_AUTONEG:
 			mode = XGBE_MODE_KX_1000;
 			break;
 		case XGBE_PORT_MODE_1000BASE_X:
@@ -1405,7 +1611,10 @@ static enum xgbe_mode xgbe_phy_an73_redrv_outcome(struct xgbe_prv_data *pdata)
 			switch (phy_data->sfp_base) {
 			case XGBE_SFP_BASE_1000_T:
 				if (phy_data->phydev &&
-				    (phy_data->phydev->speed == SPEED_100))
+				    (phy_data->phydev->speed == SPEED_10))
+					mode = XGBE_MODE_SGMII_10;
+				else if (phy_data->phydev &&
+					 (phy_data->phydev->speed == SPEED_100))
 					mode = XGBE_MODE_SGMII_100;
 				else
 					mode = XGBE_MODE_SGMII_1000;
@@ -1420,7 +1629,10 @@ static enum xgbe_mode xgbe_phy_an73_redrv_outcome(struct xgbe_prv_data *pdata)
 			break;
 		default:
 			if (phy_data->phydev &&
-			    (phy_data->phydev->speed == SPEED_100))
+			    (phy_data->phydev->speed == SPEED_10))
+				mode = XGBE_MODE_SGMII_10;
+			else if (phy_data->phydev &&
+				 (phy_data->phydev->speed == SPEED_100))
 				mode = XGBE_MODE_SGMII_100;
 			else
 				mode = XGBE_MODE_SGMII_1000;
@@ -1434,26 +1646,27 @@ static enum xgbe_mode xgbe_phy_an73_redrv_outcome(struct xgbe_prv_data *pdata)
 	ad_reg = XMDIO_READ(pdata, MDIO_MMD_AN, MDIO_AN_ADVERTISE + 2);
 	lp_reg = XMDIO_READ(pdata, MDIO_MMD_AN, MDIO_AN_LPA + 2);
 	if (lp_reg & 0xc000)
-		pdata->phy.lp_advertising |= ADVERTISED_10000baseR_FEC;
+		XGBE_SET_LP_ADV(lks, 10000baseR_FEC);
 
 	return mode;
 }
 
 static enum xgbe_mode xgbe_phy_an73_outcome(struct xgbe_prv_data *pdata)
 {
+	struct ethtool_link_ksettings *lks = &pdata->phy.lks;
 	enum xgbe_mode mode;
 	unsigned int ad_reg, lp_reg;
 
-	pdata->phy.lp_advertising |= ADVERTISED_Autoneg;
-	pdata->phy.lp_advertising |= ADVERTISED_Backplane;
+	XGBE_SET_LP_ADV(lks, Autoneg);
+	XGBE_SET_LP_ADV(lks, Backplane);
 
 	/* Compare Advertisement and Link Partner register 1 */
 	ad_reg = XMDIO_READ(pdata, MDIO_MMD_AN, MDIO_AN_ADVERTISE);
 	lp_reg = XMDIO_READ(pdata, MDIO_MMD_AN, MDIO_AN_LPA);
 	if (lp_reg & 0x400)
-		pdata->phy.lp_advertising |= ADVERTISED_Pause;
+		XGBE_SET_LP_ADV(lks, Pause);
 	if (lp_reg & 0x800)
-		pdata->phy.lp_advertising |= ADVERTISED_Asym_Pause;
+		XGBE_SET_LP_ADV(lks, Asym_Pause);
 
 	if (pdata->phy.pause_autoneg) {
 		/* Set flow control based on auto-negotiation result */
@@ -1475,9 +1688,9 @@ static enum xgbe_mode xgbe_phy_an73_outcome(struct xgbe_prv_data *pdata)
 	ad_reg = XMDIO_READ(pdata, MDIO_MMD_AN, MDIO_AN_ADVERTISE + 1);
 	lp_reg = XMDIO_READ(pdata, MDIO_MMD_AN, MDIO_AN_LPA + 1);
 	if (lp_reg & 0x80)
-		pdata->phy.lp_advertising |= ADVERTISED_10000baseKR_Full;
+		XGBE_SET_LP_ADV(lks, 10000baseKR_Full);
 	if (lp_reg & 0x20)
-		pdata->phy.lp_advertising |= ADVERTISED_1000baseKX_Full;
+		XGBE_SET_LP_ADV(lks, 1000baseKX_Full);
 
 	ad_reg &= lp_reg;
 	if (ad_reg & 0x80)
@@ -1491,7 +1704,7 @@ static enum xgbe_mode xgbe_phy_an73_outcome(struct xgbe_prv_data *pdata)
 	ad_reg = XMDIO_READ(pdata, MDIO_MMD_AN, MDIO_AN_ADVERTISE + 2);
 	lp_reg = XMDIO_READ(pdata, MDIO_MMD_AN, MDIO_AN_LPA + 2);
 	if (lp_reg & 0xc000)
-		pdata->phy.lp_advertising |= ADVERTISED_10000baseR_FEC;
+		XGBE_SET_LP_ADV(lks, 10000baseR_FEC);
 
 	return mode;
 }
@@ -1512,41 +1725,51 @@ static enum xgbe_mode xgbe_phy_an_outcome(struct xgbe_prv_data *pdata)
 	}
 }
 
-static unsigned int xgbe_phy_an_advertising(struct xgbe_prv_data *pdata)
+static void xgbe_phy_an_advertising(struct xgbe_prv_data *pdata,
+				    struct ethtool_link_ksettings *dlks)
 {
+	struct ethtool_link_ksettings *slks = &pdata->phy.lks;
 	struct xgbe_phy_data *phy_data = pdata->phy_data;
-	unsigned int advertising;
+
+	XGBE_LM_COPY(dlks, advertising, slks, advertising);
 
 	/* Without a re-driver, just return current advertising */
 	if (!phy_data->redrv)
-		return pdata->phy.advertising;
+		return;
 
 	/* With the KR re-driver we need to advertise a single speed */
-	advertising = pdata->phy.advertising;
-	advertising &= ~ADVERTISED_1000baseKX_Full;
-	advertising &= ~ADVERTISED_10000baseKR_Full;
+	XGBE_CLR_ADV(dlks, 1000baseKX_Full);
+	XGBE_CLR_ADV(dlks, 10000baseKR_Full);
+
+	/* Advertise FEC support is present */
+	if (pdata->fec_ability & MDIO_PMA_10GBR_FECABLE_ABLE)
+		XGBE_SET_ADV(dlks, 10000baseR_FEC);
 
 	switch (phy_data->port_mode) {
 	case XGBE_PORT_MODE_BACKPLANE:
-		advertising |= ADVERTISED_10000baseKR_Full;
+	case XGBE_PORT_MODE_BACKPLANE_NO_AUTONEG:
+		XGBE_SET_ADV(dlks, 10000baseKR_Full);
 		break;
 	case XGBE_PORT_MODE_BACKPLANE_2500:
-		advertising |= ADVERTISED_1000baseKX_Full;
+		XGBE_SET_ADV(dlks, 1000baseKX_Full);
 		break;
 	case XGBE_PORT_MODE_1000BASE_T:
 	case XGBE_PORT_MODE_1000BASE_X:
 	case XGBE_PORT_MODE_NBASE_T:
-		advertising |= ADVERTISED_1000baseKX_Full;
+		XGBE_SET_ADV(dlks, 1000baseKX_Full);
 		break;
 	case XGBE_PORT_MODE_10GBASE_T:
 		if (phy_data->phydev &&
 		    (phy_data->phydev->speed == SPEED_10000))
-			advertising |= ADVERTISED_10000baseKR_Full;
+			XGBE_SET_ADV(dlks, 10000baseKR_Full);
+		else if (phy_data->phydev &&
+			 (phy_data->phydev->speed == SPEED_2500))
+			XGBE_SET_ADV(dlks, 2500baseX_Full);
 		else
-			advertising |= ADVERTISED_1000baseKX_Full;
+			XGBE_SET_ADV(dlks, 1000baseKX_Full);
 		break;
 	case XGBE_PORT_MODE_10GBASE_R:
-		advertising |= ADVERTISED_10000baseKR_Full;
+		XGBE_SET_ADV(dlks, 10000baseKR_Full);
 		break;
 	case XGBE_PORT_MODE_SFP:
 		switch (phy_data->sfp_base) {
@@ -1554,23 +1777,22 @@ static unsigned int xgbe_phy_an_advertising(struct xgbe_prv_data *pdata)
 		case XGBE_SFP_BASE_1000_SX:
 		case XGBE_SFP_BASE_1000_LX:
 		case XGBE_SFP_BASE_1000_CX:
-			advertising |= ADVERTISED_1000baseKX_Full;
+			XGBE_SET_ADV(dlks, 1000baseKX_Full);
 			break;
 		default:
-			advertising |= ADVERTISED_10000baseKR_Full;
+			XGBE_SET_ADV(dlks, 10000baseKR_Full);
 			break;
 		}
 		break;
 	default:
-		advertising |= ADVERTISED_10000baseKR_Full;
+		XGBE_SET_ADV(dlks, 10000baseKR_Full);
 		break;
 	}
-
-	return advertising;
 }
 
 static int xgbe_phy_an_config(struct xgbe_prv_data *pdata)
 {
+	struct ethtool_link_ksettings *lks = &pdata->phy.lks;
 	struct xgbe_phy_data *phy_data = pdata->phy_data;
 	int ret;
 
@@ -1582,8 +1804,9 @@ static int xgbe_phy_an_config(struct xgbe_prv_data *pdata)
 		return 0;
 
 	phy_data->phydev->autoneg = pdata->phy.autoneg;
-	phy_data->phydev->advertising = phy_data->phydev->supported &
-					pdata->phy.advertising;
+	linkmode_and(phy_data->phydev->advertising,
+		     phy_data->phydev->supported,
+		     lks->link_modes.advertising);
 
 	if (pdata->phy.autoneg != AUTONEG_ENABLE) {
 		phy_data->phydev->speed = pdata->phy.speed;
@@ -1620,6 +1843,7 @@ static enum xgbe_an_mode xgbe_phy_an_mode(struct xgbe_prv_data *pdata)
 	switch (phy_data->port_mode) {
 	case XGBE_PORT_MODE_BACKPLANE:
 		return XGBE_AN_MODE_CL73;
+	case XGBE_PORT_MODE_BACKPLANE_NO_AUTONEG:
 	case XGBE_PORT_MODE_BACKPLANE_2500:
 		return XGBE_AN_MODE_NONE;
 	case XGBE_PORT_MODE_1000BASE_T:
@@ -1648,8 +1872,8 @@ static int xgbe_phy_set_redrv_mode_mdio(struct xgbe_prv_data *pdata,
 	redrv_reg = XGBE_PHY_REDRV_MODE_REG + (phy_data->redrv_lane * 0x1000);
 	redrv_val = (u16)mode;
 
-	return pdata->hw_if.write_ext_mii_regs(pdata, phy_data->redrv_addr,
-					       redrv_reg, redrv_val);
+	return pdata->hw_if.write_ext_mii_regs_c22(pdata, phy_data->redrv_addr,
+						   redrv_reg, redrv_val);
 }
 
 static int xgbe_phy_set_redrv_mode_i2c(struct xgbe_prv_data *pdata,
@@ -1694,50 +1918,194 @@ static void xgbe_phy_set_redrv_mode(struct xgbe_prv_data *pdata)
 	xgbe_phy_put_comm_ownership(pdata);
 }
 
-static void xgbe_phy_start_ratechange(struct xgbe_prv_data *pdata)
+#define MAX_RX_ADAPT_RETRIES		1
+#define XGBE_PMA_RX_VAL_SIG_MASK	(XGBE_PMA_RX_SIG_DET_0_MASK | \
+					 XGBE_PMA_RX_VALID_0_MASK)
+
+static void xgbe_set_rx_adap_mode(struct xgbe_prv_data *pdata,
+				  enum xgbe_mode mode)
 {
-	if (!XP_IOREAD_BITS(pdata, XP_DRIVER_INT_RO, STATUS))
+	if (pdata->rx_adapt_retries++ >= MAX_RX_ADAPT_RETRIES) {
+		pdata->rx_adapt_retries = 0;
 		return;
+	}
 
-	/* Log if a previous command did not complete */
-	netif_dbg(pdata, link, pdata->netdev,
-		  "firmware mailbox not ready for command\n");
+	xgbe_phy_perform_ratechange(pdata,
+				    mode == XGBE_MODE_KR ?
+				    XGBE_MB_CMD_SET_10G_KR :
+				    XGBE_MB_CMD_SET_10G_SFI,
+				    XGBE_MB_SUBCMD_RX_ADAP);
+}
+
+static void xgbe_rx_adaptation(struct xgbe_prv_data *pdata)
+{
+	struct xgbe_phy_data *phy_data = pdata->phy_data;
+	unsigned int reg;
+
+	/* step 2: force PCS to send RX_ADAPT Req to PHY */
+	XMDIO_WRITE_BITS(pdata, MDIO_MMD_PMAPMD, MDIO_PMA_RX_EQ_CTRL4,
+			 XGBE_PMA_RX_AD_REQ_MASK, XGBE_PMA_RX_AD_REQ_ENABLE);
+
+	/* Step 3: Wait for RX_ADAPT ACK from the PHY */
+	msleep(200);
+
+	/* Software polls for coefficient update command (given by local PHY) */
+	reg = XMDIO_READ(pdata, MDIO_MMD_PMAPMD, MDIO_PMA_PHY_RX_EQ_CEU);
+
+	/* Clear the RX_AD_REQ bit */
+	XMDIO_WRITE_BITS(pdata, MDIO_MMD_PMAPMD, MDIO_PMA_RX_EQ_CTRL4,
+			 XGBE_PMA_RX_AD_REQ_MASK, XGBE_PMA_RX_AD_REQ_DISABLE);
+
+	/* Check if coefficient update command is set */
+	if ((reg & XGBE_PMA_CFF_UPDT_MASK) != XGBE_PMA_CFF_UPDT_MASK)
+		goto set_mode;
+
+	/* Step 4: Check for Block lock */
+
+	/* Link status is latched low, so read once to clear
+	 * and then read again to get current state
+	 */
+	reg = XMDIO_READ(pdata, MDIO_MMD_PCS, MDIO_STAT1);
+	reg = XMDIO_READ(pdata, MDIO_MMD_PCS, MDIO_STAT1);
+	if (reg & MDIO_STAT1_LSTATUS) {
+		/* If the block lock is found, update the helpers
+		 * and declare the link up
+		 */
+		netif_dbg(pdata, link, pdata->netdev, "Block_lock done");
+		pdata->rx_adapt_done = true;
+		pdata->mode_set = false;
+		return;
+	}
+
+set_mode:
+	xgbe_set_rx_adap_mode(pdata, phy_data->cur_mode);
+}
+
+static void xgbe_phy_rx_adaptation(struct xgbe_prv_data *pdata)
+{
+	unsigned int reg;
+
+rx_adapt_reinit:
+	reg = XMDIO_READ_BITS(pdata, MDIO_MMD_PMAPMD, MDIO_PMA_RX_LSTS,
+			      XGBE_PMA_RX_VAL_SIG_MASK);
+
+	/* step 1: Check for RX_VALID && LF_SIGDET */
+	if ((reg & XGBE_PMA_RX_VAL_SIG_MASK) != XGBE_PMA_RX_VAL_SIG_MASK) {
+		netif_dbg(pdata, link, pdata->netdev,
+			  "RX_VALID or LF_SIGDET is unset, issue rrc");
+		xgbe_phy_rrc(pdata);
+		if (pdata->rx_adapt_retries++ >= MAX_RX_ADAPT_RETRIES) {
+			pdata->rx_adapt_retries = 0;
+			return;
+		}
+		goto rx_adapt_reinit;
+	}
+
+	/* perform rx adaptation */
+	xgbe_rx_adaptation(pdata);
+}
+
+static void xgbe_phy_rx_reset(struct xgbe_prv_data *pdata)
+{
+	int reg;
+
+	reg = XMDIO_READ_BITS(pdata, MDIO_MMD_PCS, MDIO_PCS_DIGITAL_STAT,
+			      XGBE_PCS_PSEQ_STATE_MASK);
+	if (reg == XGBE_PCS_PSEQ_STATE_POWER_GOOD) {
+		/* Mailbox command timed out, reset of RX block is required.
+		 * This can be done by asseting the reset bit and wait for
+		 * its compeletion.
+		 */
+		XMDIO_WRITE_BITS(pdata, MDIO_MMD_PMAPMD, MDIO_PMA_RX_CTRL1,
+				 XGBE_PMA_RX_RST_0_MASK, XGBE_PMA_RX_RST_0_RESET_ON);
+		ndelay(20);
+		XMDIO_WRITE_BITS(pdata, MDIO_MMD_PMAPMD, MDIO_PMA_RX_CTRL1,
+				 XGBE_PMA_RX_RST_0_MASK, XGBE_PMA_RX_RST_0_RESET_OFF);
+		usleep_range(40, 50);
+		netif_err(pdata, link, pdata->netdev, "firmware mailbox reset performed\n");
+	}
+}
+
+static void xgbe_phy_pll_ctrl(struct xgbe_prv_data *pdata, bool enable)
+{
+	/* PLL_CTRL feature needs to be enabled for fixed PHY modes (Non-Autoneg) only */
+	if (pdata->phy.autoneg != AUTONEG_DISABLE)
+		return;
+
+	XMDIO_WRITE_BITS(pdata, MDIO_MMD_PMAPMD, MDIO_VEND2_PMA_MISC_CTRL0,
+			 XGBE_PMA_PLL_CTRL_MASK,
+			 enable ? XGBE_PMA_PLL_CTRL_ENABLE
+				: XGBE_PMA_PLL_CTRL_DISABLE);
+
+	/* Wait for command to complete */
+	usleep_range(100, 200);
 }
 
-static void xgbe_phy_complete_ratechange(struct xgbe_prv_data *pdata)
+static void xgbe_phy_perform_ratechange(struct xgbe_prv_data *pdata,
+					enum xgbe_mb_cmd cmd, enum xgbe_mb_subcmd sub_cmd)
 {
+	unsigned int s0 = 0;
 	unsigned int wait;
 
+	/* Disable PLL re-initialization during FW command processing */
+	xgbe_phy_pll_ctrl(pdata, false);
+
+	/* Log if a previous command did not complete */
+	if (XP_IOREAD_BITS(pdata, XP_DRIVER_INT_RO, STATUS)) {
+		netif_dbg(pdata, link, pdata->netdev,
+			  "firmware mailbox not ready for command\n");
+		xgbe_phy_rx_reset(pdata);
+	}
+
+	/* Construct the command */
+	XP_SET_BITS(s0, XP_DRIVER_SCRATCH_0, COMMAND, cmd);
+	XP_SET_BITS(s0, XP_DRIVER_SCRATCH_0, SUB_COMMAND, sub_cmd);
+
+	/* Issue the command */
+	XP_IOWRITE(pdata, XP_DRIVER_SCRATCH_0, s0);
+	XP_IOWRITE(pdata, XP_DRIVER_SCRATCH_1, 0);
+	XP_IOWRITE_BITS(pdata, XP_DRIVER_INT_REQ, REQUEST, 1);
+
 	/* Wait for command to complete */
 	wait = XGBE_RATECHANGE_COUNT;
 	while (wait--) {
 		if (!XP_IOREAD_BITS(pdata, XP_DRIVER_INT_RO, STATUS))
-			return;
+			goto do_rx_adaptation;
 
 		usleep_range(1000, 2000);
 	}
 
 	netif_dbg(pdata, link, pdata->netdev,
 		  "firmware mailbox command did not complete\n");
+
+	/* Reset on error */
+	xgbe_phy_rx_reset(pdata);
+	goto reenable_pll;
+
+do_rx_adaptation:
+	if (pdata->en_rx_adap && sub_cmd == XGBE_MB_SUBCMD_RX_ADAP &&
+	    (cmd == XGBE_MB_CMD_SET_10G_KR || cmd == XGBE_MB_CMD_SET_10G_SFI)) {
+		netif_dbg(pdata, link, pdata->netdev,
+			  "Enabling RX adaptation\n");
+		pdata->mode_set = true;
+		xgbe_phy_rx_adaptation(pdata);
+		/* return from here to avoid enabling PLL ctrl
+		 * during adaptation phase
+		 */
+		return;
+	}
+
+reenable_pll:
+	/* Enable PLL re-initialization, not needed for PHY Power Off and RRC cmds */
+	if (cmd != XGBE_MB_CMD_POWER_OFF &&
+	    cmd != XGBE_MB_CMD_RRC)
+		xgbe_phy_pll_ctrl(pdata, true);
 }
 
 static void xgbe_phy_rrc(struct xgbe_prv_data *pdata)
 {
-	unsigned int s0;
-
-	xgbe_phy_start_ratechange(pdata);
-
 	/* Receiver Reset Cycle */
-	s0 = 0;
-	XP_SET_BITS(s0, XP_DRIVER_SCRATCH_0, COMMAND, 5);
-	XP_SET_BITS(s0, XP_DRIVER_SCRATCH_0, SUB_COMMAND, 0);
-
-	/* Call FW to make the change */
-	XP_IOWRITE(pdata, XP_DRIVER_SCRATCH_0, s0);
-	XP_IOWRITE(pdata, XP_DRIVER_SCRATCH_1, 0);
-	XP_IOWRITE_BITS(pdata, XP_DRIVER_INT_REQ, REQUEST, 1);
-
-	xgbe_phy_complete_ratechange(pdata);
+	xgbe_phy_perform_ratechange(pdata, XGBE_MB_CMD_RRC, XGBE_MB_SUBCMD_NONE);
 
 	netif_dbg(pdata, link, pdata->netdev, "receiver reset complete\n");
 }
@@ -1746,50 +2114,65 @@ static void xgbe_phy_power_off(struct xgbe_prv_data *pdata)
 {
 	struct xgbe_phy_data *phy_data = pdata->phy_data;
 
-	xgbe_phy_start_ratechange(pdata);
-
-	/* Call FW to make the change */
-	XP_IOWRITE(pdata, XP_DRIVER_SCRATCH_0, 0);
-	XP_IOWRITE(pdata, XP_DRIVER_SCRATCH_1, 0);
-	XP_IOWRITE_BITS(pdata, XP_DRIVER_INT_REQ, REQUEST, 1);
-
-	xgbe_phy_complete_ratechange(pdata);
+	/* Power off */
+	xgbe_phy_perform_ratechange(pdata, XGBE_MB_CMD_POWER_OFF, XGBE_MB_SUBCMD_NONE);
 
 	phy_data->cur_mode = XGBE_MODE_UNKNOWN;
 
 	netif_dbg(pdata, link, pdata->netdev, "phy powered off\n");
 }
 
+static bool enable_rx_adap(struct xgbe_prv_data *pdata, enum xgbe_mode mode)
+{
+	struct xgbe_phy_data *phy_data = pdata->phy_data;
+	unsigned int ver;
+
+	/* Rx-Adaptation is not supported on older platforms(< 0x30H) */
+	ver = XGMAC_GET_BITS(pdata->hw_feat.version, MAC_VR, SNPSVER);
+	if (ver < 0x30)
+		return false;
+
+	/* Re-driver models 4223 && 4227 do not support Rx-Adaptation */
+	if (phy_data->redrv &&
+	    (phy_data->redrv_model == XGBE_PHY_REDRV_MODEL_4223 ||
+	     phy_data->redrv_model == XGBE_PHY_REDRV_MODEL_4227))
+		return false;
+
+	/* 10G KR mode with AN does not support Rx-Adaptation */
+	if (mode == XGBE_MODE_KR &&
+	    phy_data->port_mode != XGBE_PORT_MODE_BACKPLANE_NO_AUTONEG)
+		return false;
+
+	pdata->en_rx_adap = 1;
+	return true;
+}
+
 static void xgbe_phy_sfi_mode(struct xgbe_prv_data *pdata)
 {
 	struct xgbe_phy_data *phy_data = pdata->phy_data;
-	unsigned int s0;
 
 	xgbe_phy_set_redrv_mode(pdata);
 
-	xgbe_phy_start_ratechange(pdata);
-
 	/* 10G/SFI */
-	s0 = 0;
-	XP_SET_BITS(s0, XP_DRIVER_SCRATCH_0, COMMAND, 3);
 	if (phy_data->sfp_cable != XGBE_SFP_CABLE_PASSIVE) {
-		XP_SET_BITS(s0, XP_DRIVER_SCRATCH_0, SUB_COMMAND, 0);
+		pdata->en_rx_adap = 0;
+		xgbe_phy_perform_ratechange(pdata, XGBE_MB_CMD_SET_10G_SFI, XGBE_MB_SUBCMD_ACTIVE);
+	} else if ((phy_data->sfp_cable == XGBE_SFP_CABLE_PASSIVE) &&
+		   (enable_rx_adap(pdata, XGBE_MODE_SFI))) {
+		xgbe_phy_perform_ratechange(pdata, XGBE_MB_CMD_SET_10G_SFI,
+					    XGBE_MB_SUBCMD_RX_ADAP);
 	} else {
 		if (phy_data->sfp_cable_len <= 1)
-			XP_SET_BITS(s0, XP_DRIVER_SCRATCH_0, SUB_COMMAND, 1);
+			xgbe_phy_perform_ratechange(pdata, XGBE_MB_CMD_SET_10G_SFI,
+						    XGBE_MB_SUBCMD_PASSIVE_1M);
 		else if (phy_data->sfp_cable_len <= 3)
-			XP_SET_BITS(s0, XP_DRIVER_SCRATCH_0, SUB_COMMAND, 2);
+			xgbe_phy_perform_ratechange(pdata, XGBE_MB_CMD_SET_10G_SFI,
+						    XGBE_MB_SUBCMD_PASSIVE_3M);
 		else
-			XP_SET_BITS(s0, XP_DRIVER_SCRATCH_0, SUB_COMMAND, 3);
+			xgbe_phy_perform_ratechange(pdata, XGBE_MB_CMD_SET_10G_SFI,
+						    XGBE_MB_SUBCMD_PASSIVE_OTHER);
 	}
 
-	/* Call FW to make the change */
-	XP_IOWRITE(pdata, XP_DRIVER_SCRATCH_0, s0);
-	XP_IOWRITE(pdata, XP_DRIVER_SCRATCH_1, 0);
-	XP_IOWRITE_BITS(pdata, XP_DRIVER_INT_REQ, REQUEST, 1);
-
-	xgbe_phy_complete_ratechange(pdata);
-
 	phy_data->cur_mode = XGBE_MODE_SFI;
 
 	netif_dbg(pdata, link, pdata->netdev, "10GbE SFI mode set\n");
@@ -1798,23 +2181,11 @@ static void xgbe_phy_sfi_mode(struct xgbe_prv_data *pdata)
 static void xgbe_phy_x_mode(struct xgbe_prv_data *pdata)
 {
 	struct xgbe_phy_data *phy_data = pdata->phy_data;
-	unsigned int s0;
 
 	xgbe_phy_set_redrv_mode(pdata);
 
-	xgbe_phy_start_ratechange(pdata);
-
 	/* 1G/X */
-	s0 = 0;
-	XP_SET_BITS(s0, XP_DRIVER_SCRATCH_0, COMMAND, 1);
-	XP_SET_BITS(s0, XP_DRIVER_SCRATCH_0, SUB_COMMAND, 3);
-
-	/* Call FW to make the change */
-	XP_IOWRITE(pdata, XP_DRIVER_SCRATCH_0, s0);
-	XP_IOWRITE(pdata, XP_DRIVER_SCRATCH_1, 0);
-	XP_IOWRITE_BITS(pdata, XP_DRIVER_INT_REQ, REQUEST, 1);
-
-	xgbe_phy_complete_ratechange(pdata);
+	xgbe_phy_perform_ratechange(pdata, XGBE_MB_CMD_SET_1G, XGBE_MB_SUBCMD_1G_KX);
 
 	phy_data->cur_mode = XGBE_MODE_X;
 
@@ -1824,23 +2195,11 @@ static void xgbe_phy_x_mode(struct xgbe_prv_data *pdata)
 static void xgbe_phy_sgmii_1000_mode(struct xgbe_prv_data *pdata)
 {
 	struct xgbe_phy_data *phy_data = pdata->phy_data;
-	unsigned int s0;
 
 	xgbe_phy_set_redrv_mode(pdata);
 
-	xgbe_phy_start_ratechange(pdata);
-
 	/* 1G/SGMII */
-	s0 = 0;
-	XP_SET_BITS(s0, XP_DRIVER_SCRATCH_0, COMMAND, 1);
-	XP_SET_BITS(s0, XP_DRIVER_SCRATCH_0, SUB_COMMAND, 2);
-
-	/* Call FW to make the change */
-	XP_IOWRITE(pdata, XP_DRIVER_SCRATCH_0, s0);
-	XP_IOWRITE(pdata, XP_DRIVER_SCRATCH_1, 0);
-	XP_IOWRITE_BITS(pdata, XP_DRIVER_INT_REQ, REQUEST, 1);
-
-	xgbe_phy_complete_ratechange(pdata);
+	xgbe_phy_perform_ratechange(pdata, XGBE_MB_CMD_SET_1G, XGBE_MB_SUBCMD_1G_SGMII);
 
 	phy_data->cur_mode = XGBE_MODE_SGMII_1000;
 
@@ -1850,49 +2209,44 @@ static void xgbe_phy_sgmii_1000_mode(struct xgbe_prv_data *pdata)
 static void xgbe_phy_sgmii_100_mode(struct xgbe_prv_data *pdata)
 {
 	struct xgbe_phy_data *phy_data = pdata->phy_data;
-	unsigned int s0;
 
 	xgbe_phy_set_redrv_mode(pdata);
 
-	xgbe_phy_start_ratechange(pdata);
-
-	/* 1G/SGMII */
-	s0 = 0;
-	XP_SET_BITS(s0, XP_DRIVER_SCRATCH_0, COMMAND, 1);
-	XP_SET_BITS(s0, XP_DRIVER_SCRATCH_0, SUB_COMMAND, 1);
-
-	/* Call FW to make the change */
-	XP_IOWRITE(pdata, XP_DRIVER_SCRATCH_0, s0);
-	XP_IOWRITE(pdata, XP_DRIVER_SCRATCH_1, 0);
-	XP_IOWRITE_BITS(pdata, XP_DRIVER_INT_REQ, REQUEST, 1);
-
-	xgbe_phy_complete_ratechange(pdata);
+	/* 100M/SGMII */
+	xgbe_phy_perform_ratechange(pdata, XGBE_MB_CMD_SET_1G, XGBE_MB_SUBCMD_100MBITS);
 
 	phy_data->cur_mode = XGBE_MODE_SGMII_100;
 
 	netif_dbg(pdata, link, pdata->netdev, "100MbE SGMII mode set\n");
 }
 
-static void xgbe_phy_kr_mode(struct xgbe_prv_data *pdata)
+static void xgbe_phy_sgmii_10_mode(struct xgbe_prv_data *pdata)
 {
 	struct xgbe_phy_data *phy_data = pdata->phy_data;
-	unsigned int s0;
 
 	xgbe_phy_set_redrv_mode(pdata);
 
-	xgbe_phy_start_ratechange(pdata);
+	/* 10M/SGMII */
+	xgbe_phy_perform_ratechange(pdata, XGBE_MB_CMD_SET_1G, XGBE_MB_SUBCMD_10MBITS);
 
-	/* 10G/KR */
-	s0 = 0;
-	XP_SET_BITS(s0, XP_DRIVER_SCRATCH_0, COMMAND, 4);
-	XP_SET_BITS(s0, XP_DRIVER_SCRATCH_0, SUB_COMMAND, 0);
+	phy_data->cur_mode = XGBE_MODE_SGMII_10;
 
-	/* Call FW to make the change */
-	XP_IOWRITE(pdata, XP_DRIVER_SCRATCH_0, s0);
-	XP_IOWRITE(pdata, XP_DRIVER_SCRATCH_1, 0);
-	XP_IOWRITE_BITS(pdata, XP_DRIVER_INT_REQ, REQUEST, 1);
+	netif_dbg(pdata, link, pdata->netdev, "10MbE SGMII mode set\n");
+}
 
-	xgbe_phy_complete_ratechange(pdata);
+static void xgbe_phy_kr_mode(struct xgbe_prv_data *pdata)
+{
+	struct xgbe_phy_data *phy_data = pdata->phy_data;
+
+	xgbe_phy_set_redrv_mode(pdata);
+
+	/* 10G/KR */
+	if (enable_rx_adap(pdata, XGBE_MODE_KR))
+		xgbe_phy_perform_ratechange(pdata, XGBE_MB_CMD_SET_10G_KR,
+					    XGBE_MB_SUBCMD_RX_ADAP);
+	else
+		xgbe_phy_perform_ratechange(pdata, XGBE_MB_CMD_SET_10G_KR,
+					    XGBE_MB_SUBCMD_NONE);
 
 	phy_data->cur_mode = XGBE_MODE_KR;
 
@@ -1902,23 +2256,11 @@ static void xgbe_phy_kr_mode(struct xgbe_prv_data *pdata)
 static void xgbe_phy_kx_2500_mode(struct xgbe_prv_data *pdata)
 {
 	struct xgbe_phy_data *phy_data = pdata->phy_data;
-	unsigned int s0;
 
 	xgbe_phy_set_redrv_mode(pdata);
 
-	xgbe_phy_start_ratechange(pdata);
-
 	/* 2.5G/KX */
-	s0 = 0;
-	XP_SET_BITS(s0, XP_DRIVER_SCRATCH_0, COMMAND, 2);
-	XP_SET_BITS(s0, XP_DRIVER_SCRATCH_0, SUB_COMMAND, 0);
-
-	/* Call FW to make the change */
-	XP_IOWRITE(pdata, XP_DRIVER_SCRATCH_0, s0);
-	XP_IOWRITE(pdata, XP_DRIVER_SCRATCH_1, 0);
-	XP_IOWRITE_BITS(pdata, XP_DRIVER_INT_REQ, REQUEST, 1);
-
-	xgbe_phy_complete_ratechange(pdata);
+	xgbe_phy_perform_ratechange(pdata, XGBE_MB_CMD_SET_2_5G, XGBE_MB_SUBCMD_NONE);
 
 	phy_data->cur_mode = XGBE_MODE_KX_2500;
 
@@ -1928,23 +2270,11 @@ static void xgbe_phy_kx_2500_mode(struct xgbe_prv_data *pdata)
 static void xgbe_phy_kx_1000_mode(struct xgbe_prv_data *pdata)
 {
 	struct xgbe_phy_data *phy_data = pdata->phy_data;
-	unsigned int s0;
 
 	xgbe_phy_set_redrv_mode(pdata);
 
-	xgbe_phy_start_ratechange(pdata);
-
 	/* 1G/KX */
-	s0 = 0;
-	XP_SET_BITS(s0, XP_DRIVER_SCRATCH_0, COMMAND, 1);
-	XP_SET_BITS(s0, XP_DRIVER_SCRATCH_0, SUB_COMMAND, 3);
-
-	/* Call FW to make the change */
-	XP_IOWRITE(pdata, XP_DRIVER_SCRATCH_0, s0);
-	XP_IOWRITE(pdata, XP_DRIVER_SCRATCH_1, 0);
-	XP_IOWRITE_BITS(pdata, XP_DRIVER_INT_REQ, REQUEST, 1);
-
-	xgbe_phy_complete_ratechange(pdata);
+	xgbe_phy_perform_ratechange(pdata, XGBE_MB_CMD_SET_1G, XGBE_MB_SUBCMD_1G_KX);
 
 	phy_data->cur_mode = XGBE_MODE_KX_1000;
 
@@ -1967,12 +2297,15 @@ static enum xgbe_mode xgbe_phy_switch_baset_mode(struct xgbe_prv_data *pdata)
 		return xgbe_phy_cur_mode(pdata);
 
 	switch (xgbe_phy_cur_mode(pdata)) {
+	case XGBE_MODE_SGMII_10:
 	case XGBE_MODE_SGMII_100:
 	case XGBE_MODE_SGMII_1000:
 		return XGBE_MODE_KR;
+	case XGBE_MODE_KX_2500:
+		return XGBE_MODE_SGMII_1000;
 	case XGBE_MODE_KR:
 	default:
-		return XGBE_MODE_SGMII_1000;
+		return XGBE_MODE_KX_2500;
 	}
 }
 
@@ -1999,6 +2332,7 @@ static enum xgbe_mode xgbe_phy_switch_mode(struct xgbe_prv_data *pdata)
 
 	switch (phy_data->port_mode) {
 	case XGBE_PORT_MODE_BACKPLANE:
+	case XGBE_PORT_MODE_BACKPLANE_NO_AUTONEG:
 		return xgbe_phy_switch_bp_mode(pdata);
 	case XGBE_PORT_MODE_BACKPLANE_2500:
 		return xgbe_phy_switch_bp_2500_mode(pdata);
@@ -2033,10 +2367,14 @@ static enum xgbe_mode xgbe_phy_get_baset_mode(struct xgbe_phy_data *phy_data,
 					      int speed)
 {
 	switch (speed) {
+	case SPEED_10:
+		return XGBE_MODE_SGMII_10;
 	case SPEED_100:
 		return XGBE_MODE_SGMII_100;
 	case SPEED_1000:
 		return XGBE_MODE_SGMII_1000;
+	case SPEED_2500:
+		return XGBE_MODE_KX_2500;
 	case SPEED_10000:
 		return XGBE_MODE_KR;
 	default:
@@ -2048,6 +2386,8 @@ static enum xgbe_mode xgbe_phy_get_sfp_mode(struct xgbe_phy_data *phy_data,
 					    int speed)
 {
 	switch (speed) {
+	case SPEED_10:
+		return XGBE_MODE_SGMII_10;
 	case SPEED_100:
 		return XGBE_MODE_SGMII_100;
 	case SPEED_1000:
@@ -2092,6 +2432,7 @@ static enum xgbe_mode xgbe_phy_get_mode(struct xgbe_prv_data *pdata,
 
 	switch (phy_data->port_mode) {
 	case XGBE_PORT_MODE_BACKPLANE:
+	case XGBE_PORT_MODE_BACKPLANE_NO_AUTONEG:
 		return xgbe_phy_get_bp_mode(speed);
 	case XGBE_PORT_MODE_BACKPLANE_2500:
 		return xgbe_phy_get_bp_2500_mode(speed);
@@ -2121,6 +2462,9 @@ static void xgbe_phy_set_mode(struct xgbe_prv_data *pdata, enum xgbe_mode mode)
 	case XGBE_MODE_KR:
 		xgbe_phy_kr_mode(pdata);
 		break;
+	case XGBE_MODE_SGMII_10:
+		xgbe_phy_sgmii_10_mode(pdata);
+		break;
 	case XGBE_MODE_SGMII_100:
 		xgbe_phy_sgmii_100_mode(pdata);
 		break;
@@ -2139,11 +2483,10 @@ static void xgbe_phy_set_mode(struct xgbe_prv_data *pdata, enum xgbe_mode mode)
 }
 
 static bool xgbe_phy_check_mode(struct xgbe_prv_data *pdata,
-				enum xgbe_mode mode, u32 advert)
+				enum xgbe_mode mode, bool advert)
 {
 	if (pdata->phy.autoneg == AUTONEG_ENABLE) {
-		if (pdata->phy.advertising & advert)
-			return true;
+		return advert;
 	} else {
 		enum xgbe_mode cur_mode;
 
@@ -2158,13 +2501,15 @@ static bool xgbe_phy_check_mode(struct xgbe_prv_data *pdata,
 static bool xgbe_phy_use_basex_mode(struct xgbe_prv_data *pdata,
 				    enum xgbe_mode mode)
 {
+	struct ethtool_link_ksettings *lks = &pdata->phy.lks;
+
 	switch (mode) {
 	case XGBE_MODE_X:
 		return xgbe_phy_check_mode(pdata, mode,
-					   ADVERTISED_1000baseT_Full);
+					   XGBE_ADV(lks, 1000baseX_Full));
 	case XGBE_MODE_KR:
 		return xgbe_phy_check_mode(pdata, mode,
-					   ADVERTISED_10000baseT_Full);
+					   XGBE_ADV(lks, 10000baseKR_Full));
 	default:
 		return false;
 	}
@@ -2173,16 +2518,24 @@ static bool xgbe_phy_use_basex_mode(struct xgbe_prv_data *pdata,
 static bool xgbe_phy_use_baset_mode(struct xgbe_prv_data *pdata,
 				    enum xgbe_mode mode)
 {
+	struct ethtool_link_ksettings *lks = &pdata->phy.lks;
+
 	switch (mode) {
+	case XGBE_MODE_SGMII_10:
+		return xgbe_phy_check_mode(pdata, mode,
+					   XGBE_ADV(lks, 10baseT_Full));
 	case XGBE_MODE_SGMII_100:
 		return xgbe_phy_check_mode(pdata, mode,
-					   ADVERTISED_100baseT_Full);
+					   XGBE_ADV(lks, 100baseT_Full));
 	case XGBE_MODE_SGMII_1000:
 		return xgbe_phy_check_mode(pdata, mode,
-					   ADVERTISED_1000baseT_Full);
+					   XGBE_ADV(lks, 1000baseT_Full));
+	case XGBE_MODE_KX_2500:
+		return xgbe_phy_check_mode(pdata, mode,
+					   XGBE_ADV(lks, 2500baseT_Full));
 	case XGBE_MODE_KR:
 		return xgbe_phy_check_mode(pdata, mode,
-					   ADVERTISED_10000baseT_Full);
+					   XGBE_ADV(lks, 10000baseT_Full));
 	default:
 		return false;
 	}
@@ -2191,6 +2544,7 @@ static bool xgbe_phy_use_baset_mode(struct xgbe_prv_data *pdata,
 static bool xgbe_phy_use_sfp_mode(struct xgbe_prv_data *pdata,
 				  enum xgbe_mode mode)
 {
+	struct ethtool_link_ksettings *lks = &pdata->phy.lks;
 	struct xgbe_phy_data *phy_data = pdata->phy_data;
 
 	switch (mode) {
@@ -2198,20 +2552,31 @@ static bool xgbe_phy_use_sfp_mode(struct xgbe_prv_data *pdata,
 		if (phy_data->sfp_base == XGBE_SFP_BASE_1000_T)
 			return false;
 		return xgbe_phy_check_mode(pdata, mode,
-					   ADVERTISED_1000baseT_Full);
+					   XGBE_ADV(lks, 1000baseX_Full));
+	case XGBE_MODE_SGMII_10:
+		if (phy_data->sfp_base != XGBE_SFP_BASE_1000_T)
+			return false;
+		return xgbe_phy_check_mode(pdata, mode,
+					   XGBE_ADV(lks, 10baseT_Full));
 	case XGBE_MODE_SGMII_100:
 		if (phy_data->sfp_base != XGBE_SFP_BASE_1000_T)
 			return false;
 		return xgbe_phy_check_mode(pdata, mode,
-					   ADVERTISED_100baseT_Full);
+					   XGBE_ADV(lks, 100baseT_Full));
 	case XGBE_MODE_SGMII_1000:
 		if (phy_data->sfp_base != XGBE_SFP_BASE_1000_T)
 			return false;
 		return xgbe_phy_check_mode(pdata, mode,
-					   ADVERTISED_1000baseT_Full);
+					   XGBE_ADV(lks, 1000baseT_Full));
 	case XGBE_MODE_SFI:
+		if (phy_data->sfp_mod_absent)
+			return true;
 		return xgbe_phy_check_mode(pdata, mode,
-					   ADVERTISED_10000baseT_Full);
+					   XGBE_ADV(lks, 10000baseSR_Full)  ||
+					   XGBE_ADV(lks, 10000baseLR_Full)  ||
+					   XGBE_ADV(lks, 10000baseLRM_Full) ||
+					   XGBE_ADV(lks, 10000baseER_Full)  ||
+					   XGBE_ADV(lks, 10000baseCR_Full));
 	default:
 		return false;
 	}
@@ -2220,10 +2585,12 @@ static bool xgbe_phy_use_sfp_mode(struct xgbe_prv_data *pdata,
 static bool xgbe_phy_use_bp_2500_mode(struct xgbe_prv_data *pdata,
 				      enum xgbe_mode mode)
 {
+	struct ethtool_link_ksettings *lks = &pdata->phy.lks;
+
 	switch (mode) {
 	case XGBE_MODE_KX_2500:
 		return xgbe_phy_check_mode(pdata, mode,
-					   ADVERTISED_2500baseX_Full);
+					   XGBE_ADV(lks, 2500baseX_Full));
 	default:
 		return false;
 	}
@@ -2232,13 +2599,15 @@ static bool xgbe_phy_use_bp_2500_mode(struct xgbe_prv_data *pdata,
 static bool xgbe_phy_use_bp_mode(struct xgbe_prv_data *pdata,
 				 enum xgbe_mode mode)
 {
+	struct ethtool_link_ksettings *lks = &pdata->phy.lks;
+
 	switch (mode) {
 	case XGBE_MODE_KX_1000:
 		return xgbe_phy_check_mode(pdata, mode,
-					   ADVERTISED_1000baseKX_Full);
+					   XGBE_ADV(lks, 1000baseKX_Full));
 	case XGBE_MODE_KR:
 		return xgbe_phy_check_mode(pdata, mode,
-					   ADVERTISED_10000baseKR_Full);
+					   XGBE_ADV(lks, 10000baseKR_Full));
 	default:
 		return false;
 	}
@@ -2250,6 +2619,7 @@ static bool xgbe_phy_use_mode(struct xgbe_prv_data *pdata, enum xgbe_mode mode)
 
 	switch (phy_data->port_mode) {
 	case XGBE_PORT_MODE_BACKPLANE:
+	case XGBE_PORT_MODE_BACKPLANE_NO_AUTONEG:
 		return xgbe_phy_use_bp_mode(pdata, mode);
 	case XGBE_PORT_MODE_BACKPLANE_2500:
 		return xgbe_phy_use_bp_2500_mode(pdata, mode);
@@ -2280,13 +2650,23 @@ static bool xgbe_phy_valid_speed_basex_mode(struct xgbe_phy_data *phy_data,
 	}
 }
 
-static bool xgbe_phy_valid_speed_baset_mode(struct xgbe_phy_data *phy_data,
+static bool xgbe_phy_valid_speed_baset_mode(struct xgbe_prv_data *pdata,
 					    int speed)
 {
+	struct xgbe_phy_data *phy_data = pdata->phy_data;
+	unsigned int ver;
+
 	switch (speed) {
+	case SPEED_10:
+		/* Supported in ver 21H and ver >= 30H */
+		ver = XGMAC_GET_BITS(pdata->hw_feat.version, MAC_VR, SNPSVER);
+		return (ver == 0x21 || ver >= 0x30);
 	case SPEED_100:
 	case SPEED_1000:
 		return true;
+	case SPEED_2500:
+		return ((phy_data->port_mode == XGBE_PORT_MODE_10GBASE_T) ||
+			(phy_data->port_mode == XGBE_PORT_MODE_NBASE_T));
 	case SPEED_10000:
 		return (phy_data->port_mode == XGBE_PORT_MODE_10GBASE_T);
 	default:
@@ -2294,10 +2674,18 @@ static bool xgbe_phy_valid_speed_baset_mode(struct xgbe_phy_data *phy_data,
 	}
 }
 
-static bool xgbe_phy_valid_speed_sfp_mode(struct xgbe_phy_data *phy_data,
+static bool xgbe_phy_valid_speed_sfp_mode(struct xgbe_prv_data *pdata,
 					  int speed)
 {
+	struct xgbe_phy_data *phy_data = pdata->phy_data;
+	unsigned int ver;
+
 	switch (speed) {
+	case SPEED_10:
+		/* Supported in ver 21H and ver >= 30H */
+		ver = XGMAC_GET_BITS(pdata->hw_feat.version, MAC_VR, SNPSVER);
+		return ((ver == 0x21 || ver >= 0x30) &&
+			(phy_data->sfp_speed == XGBE_SFP_SPEED_100_1000));
 	case SPEED_100:
 		return (phy_data->sfp_speed == XGBE_SFP_SPEED_100_1000);
 	case SPEED_1000:
@@ -2337,18 +2725,19 @@ static bool xgbe_phy_valid_speed(struct xgbe_prv_data *pdata, int speed)
 
 	switch (phy_data->port_mode) {
 	case XGBE_PORT_MODE_BACKPLANE:
+	case XGBE_PORT_MODE_BACKPLANE_NO_AUTONEG:
 		return xgbe_phy_valid_speed_bp_mode(speed);
 	case XGBE_PORT_MODE_BACKPLANE_2500:
 		return xgbe_phy_valid_speed_bp_2500_mode(speed);
 	case XGBE_PORT_MODE_1000BASE_T:
 	case XGBE_PORT_MODE_NBASE_T:
 	case XGBE_PORT_MODE_10GBASE_T:
-		return xgbe_phy_valid_speed_baset_mode(phy_data, speed);
+		return xgbe_phy_valid_speed_baset_mode(pdata, speed);
 	case XGBE_PORT_MODE_1000BASE_X:
 	case XGBE_PORT_MODE_10GBASE_R:
 		return xgbe_phy_valid_speed_basex_mode(phy_data, speed);
 	case XGBE_PORT_MODE_SFP:
-		return xgbe_phy_valid_speed_sfp_mode(phy_data, speed);
+		return xgbe_phy_valid_speed_sfp_mode(pdata, speed);
 	default:
 		return false;
 	}
@@ -2357,8 +2746,7 @@ static bool xgbe_phy_valid_speed(struct xgbe_prv_data *pdata, int speed)
 static int xgbe_phy_link_status(struct xgbe_prv_data *pdata, int *an_restart)
 {
 	struct xgbe_phy_data *phy_data = pdata->phy_data;
-	unsigned int reg;
-	int ret;
+	int reg, ret;
 
 	*an_restart = 0;
 
@@ -2371,8 +2759,11 @@ static int xgbe_phy_link_status(struct xgbe_prv_data *pdata, int *an_restart)
 			return 0;
 		}
 
-		if (phy_data->sfp_mod_absent || phy_data->sfp_rx_los)
+		if (phy_data->sfp_mod_absent || phy_data->sfp_rx_los) {
+			if (pdata->en_rx_adap)
+				pdata->rx_adapt_done = false;
 			return 0;
+		}
 	}
 
 	if (phy_data->phydev) {
@@ -2389,16 +2780,53 @@ static int xgbe_phy_link_status(struct xgbe_prv_data *pdata, int *an_restart)
 			return 0;
 	}
 
-	/* Link status is latched low, so read once to clear
-	 * and then read again to get current state
-	 */
 	reg = XMDIO_READ(pdata, MDIO_MMD_PCS, MDIO_STAT1);
-	reg = XMDIO_READ(pdata, MDIO_MMD_PCS, MDIO_STAT1);
-	if (reg & MDIO_STAT1_LSTATUS)
+	if (reg < 0)
+		return reg;
+
+	/* Link status is latched low so that momentary link drops
+	 * can be detected. If link was already down read again
+	 * to get the latest state.
+	 */
+
+	if (!pdata->phy.link && !(reg & MDIO_STAT1_LSTATUS)) {
+		reg = XMDIO_READ(pdata, MDIO_MMD_PCS, MDIO_STAT1);
+		if (reg < 0)
+			return reg;
+	}
+
+	if (pdata->en_rx_adap) {
+		/* if the link is available and adaptation is done,
+		 * declare link up
+		 */
+		if ((reg & MDIO_STAT1_LSTATUS) && pdata->rx_adapt_done)
+			return 1;
+		/* If either link is not available or adaptation is not done,
+		 * retrigger the adaptation logic. (if the mode is not set,
+		 * then issue mailbox command first)
+		 */
+		if (pdata->mode_set) {
+			xgbe_phy_rx_adaptation(pdata);
+		} else {
+			pdata->rx_adapt_done = false;
+			xgbe_phy_set_mode(pdata, phy_data->cur_mode);
+		}
+
+		if (pdata->rx_adapt_done)
+			return 1;
+	} else if (reg & MDIO_STAT1_LSTATUS)
 		return 1;
 
+	if (pdata->phy.autoneg == AUTONEG_ENABLE &&
+	    phy_data->port_mode == XGBE_PORT_MODE_BACKPLANE) {
+		if (!test_bit(XGBE_LINK_INIT, &pdata->dev_state)) {
+			netif_carrier_off(pdata->netdev);
+			*an_restart = 1;
+		}
+	}
+
 	/* No link, attempt a receiver reset cycle */
-	if (phy_data->rrc_count++) {
+	if (pdata->vdata->enable_rrc && phy_data->rrc_count++ > XGBE_RRC_FREQUENCY) {
 		phy_data->rrc_count = 0;
 		xgbe_phy_rrc(pdata);
 	}
@@ -2409,22 +2837,21 @@ static int xgbe_phy_link_status(struct xgbe_prv_data *pdata, int *an_restart)
 static void xgbe_phy_sfp_gpio_setup(struct xgbe_prv_data *pdata)
 {
 	struct xgbe_phy_data *phy_data = pdata->phy_data;
-	unsigned int reg;
-
-	reg = XP_IOREAD(pdata, XP_PROP_3);
 
 	phy_data->sfp_gpio_address = XGBE_GPIO_ADDRESS_PCA9555 +
-				     XP_GET_BITS(reg, XP_PROP_3, GPIO_ADDR);
+				     XP_GET_BITS(pdata->pp3, XP_PROP_3,
+						 GPIO_ADDR);
 
-	phy_data->sfp_gpio_mask = XP_GET_BITS(reg, XP_PROP_3, GPIO_MASK);
+	phy_data->sfp_gpio_mask = XP_GET_BITS(pdata->pp3, XP_PROP_3,
+					      GPIO_MASK);
 
-	phy_data->sfp_gpio_rx_los = XP_GET_BITS(reg, XP_PROP_3,
+	phy_data->sfp_gpio_rx_los = XP_GET_BITS(pdata->pp3, XP_PROP_3,
 						GPIO_RX_LOS);
-	phy_data->sfp_gpio_tx_fault = XP_GET_BITS(reg, XP_PROP_3,
+	phy_data->sfp_gpio_tx_fault = XP_GET_BITS(pdata->pp3, XP_PROP_3,
 						  GPIO_TX_FAULT);
-	phy_data->sfp_gpio_mod_absent = XP_GET_BITS(reg, XP_PROP_3,
+	phy_data->sfp_gpio_mod_absent = XP_GET_BITS(pdata->pp3, XP_PROP_3,
 						    GPIO_MOD_ABS);
-	phy_data->sfp_gpio_rate_select = XP_GET_BITS(reg, XP_PROP_3,
+	phy_data->sfp_gpio_rate_select = XP_GET_BITS(pdata->pp3, XP_PROP_3,
 						     GPIO_RATE_SELECT);
 
 	if (netif_msg_probe(pdata)) {
@@ -2446,18 +2873,17 @@ static void xgbe_phy_sfp_gpio_setup(struct xgbe_prv_data *pdata)
 static void xgbe_phy_sfp_comm_setup(struct xgbe_prv_data *pdata)
 {
 	struct xgbe_phy_data *phy_data = pdata->phy_data;
-	unsigned int reg, mux_addr_hi, mux_addr_lo;
+	unsigned int mux_addr_hi, mux_addr_lo;
 
-	reg = XP_IOREAD(pdata, XP_PROP_4);
-
-	mux_addr_hi = XP_GET_BITS(reg, XP_PROP_4, MUX_ADDR_HI);
-	mux_addr_lo = XP_GET_BITS(reg, XP_PROP_4, MUX_ADDR_LO);
+	mux_addr_hi = XP_GET_BITS(pdata->pp4, XP_PROP_4, MUX_ADDR_HI);
+	mux_addr_lo = XP_GET_BITS(pdata->pp4, XP_PROP_4, MUX_ADDR_LO);
 	if (mux_addr_lo == XGBE_SFP_DIRECT)
 		return;
 
 	phy_data->sfp_comm = XGBE_SFP_COMM_PCA9545;
 	phy_data->sfp_mux_address = (mux_addr_hi << 2) + mux_addr_lo;
-	phy_data->sfp_mux_channel = XP_GET_BITS(reg, XP_PROP_4, MUX_CHAN);
+	phy_data->sfp_mux_channel = XP_GET_BITS(pdata->pp4, XP_PROP_4,
+						MUX_CHAN);
 
 	if (netif_msg_probe(pdata)) {
 		dev_dbg(pdata->dev, "SFP: mux_address=%#x\n",
@@ -2476,7 +2902,7 @@ static void xgbe_phy_sfp_setup(struct xgbe_prv_data *pdata)
 static int xgbe_phy_int_mdio_reset(struct xgbe_prv_data *pdata)
 {
 	struct xgbe_phy_data *phy_data = pdata->phy_data;
-	unsigned int ret;
+	int ret;
 
 	ret = pdata->hw_if.set_gpio(pdata, phy_data->mdio_reset_gpio);
 	if (ret)
@@ -2580,13 +3006,11 @@ static bool xgbe_phy_redrv_error(struct xgbe_phy_data *phy_data)
 static int xgbe_phy_mdio_reset_setup(struct xgbe_prv_data *pdata)
 {
 	struct xgbe_phy_data *phy_data = pdata->phy_data;
-	unsigned int reg;
 
 	if (phy_data->conn_type != XGBE_CONN_TYPE_MDIO)
 		return 0;
 
-	reg = XP_IOREAD(pdata, XP_PROP_3);
-	phy_data->mdio_reset = XP_GET_BITS(reg, XP_PROP_3, MDIO_RESET);
+	phy_data->mdio_reset = XP_GET_BITS(pdata->pp3, XP_PROP_3, MDIO_RESET);
 	switch (phy_data->mdio_reset) {
 	case XGBE_MDIO_RESET_NONE:
 	case XGBE_MDIO_RESET_I2C_GPIO:
@@ -2600,12 +3024,12 @@ static int xgbe_phy_mdio_reset_setup(struct xgbe_prv_data *pdata)
 
 	if (phy_data->mdio_reset == XGBE_MDIO_RESET_I2C_GPIO) {
 		phy_data->mdio_reset_addr = XGBE_GPIO_ADDRESS_PCA9555 +
-					    XP_GET_BITS(reg, XP_PROP_3,
+					    XP_GET_BITS(pdata->pp3, XP_PROP_3,
 							MDIO_RESET_I2C_ADDR);
-		phy_data->mdio_reset_gpio = XP_GET_BITS(reg, XP_PROP_3,
+		phy_data->mdio_reset_gpio = XP_GET_BITS(pdata->pp3, XP_PROP_3,
 							MDIO_RESET_I2C_GPIO);
 	} else if (phy_data->mdio_reset == XGBE_MDIO_RESET_INT_GPIO) {
-		phy_data->mdio_reset_gpio = XP_GET_BITS(reg, XP_PROP_3,
+		phy_data->mdio_reset_gpio = XP_GET_BITS(pdata->pp3, XP_PROP_3,
 							MDIO_RESET_INT_GPIO);
 	}
 
@@ -2615,9 +3039,16 @@ static int xgbe_phy_mdio_reset_setup(struct xgbe_prv_data *pdata)
 static bool xgbe_phy_port_mode_mismatch(struct xgbe_prv_data *pdata)
 {
 	struct xgbe_phy_data *phy_data = pdata->phy_data;
+	unsigned int ver;
+
+	/* 10 Mbps speed is supported in ver 21H and ver >= 30H */
+	ver = XGMAC_GET_BITS(pdata->hw_feat.version, MAC_VR, SNPSVER);
+	if ((ver < 0x30 && ver != 0x21) && (phy_data->port_speeds & XGBE_PHY_PORT_SPEED_10))
+		return true;
 
 	switch (phy_data->port_mode) {
 	case XGBE_PORT_MODE_BACKPLANE:
+	case XGBE_PORT_MODE_BACKPLANE_NO_AUTONEG:
 		if ((phy_data->port_speeds & XGBE_PHY_PORT_SPEED_1000) ||
 		    (phy_data->port_speeds & XGBE_PHY_PORT_SPEED_10000))
 			return false;
@@ -2627,7 +3058,8 @@ static bool xgbe_phy_port_mode_mismatch(struct xgbe_prv_data *pdata)
 			return false;
 		break;
 	case XGBE_PORT_MODE_1000BASE_T:
-		if ((phy_data->port_speeds & XGBE_PHY_PORT_SPEED_100) ||
+		if ((phy_data->port_speeds & XGBE_PHY_PORT_SPEED_10) ||
+		    (phy_data->port_speeds & XGBE_PHY_PORT_SPEED_100) ||
 		    (phy_data->port_speeds & XGBE_PHY_PORT_SPEED_1000))
 			return false;
 		break;
@@ -2636,14 +3068,17 @@ static bool xgbe_phy_port_mode_mismatch(struct xgbe_prv_data *pdata)
 			return false;
 		break;
 	case XGBE_PORT_MODE_NBASE_T:
-		if ((phy_data->port_speeds & XGBE_PHY_PORT_SPEED_100) ||
+		if ((phy_data->port_speeds & XGBE_PHY_PORT_SPEED_10) ||
+		    (phy_data->port_speeds & XGBE_PHY_PORT_SPEED_100) ||
 		    (phy_data->port_speeds & XGBE_PHY_PORT_SPEED_1000) ||
 		    (phy_data->port_speeds & XGBE_PHY_PORT_SPEED_2500))
 			return false;
 		break;
 	case XGBE_PORT_MODE_10GBASE_T:
-		if ((phy_data->port_speeds & XGBE_PHY_PORT_SPEED_100) ||
+		if ((phy_data->port_speeds & XGBE_PHY_PORT_SPEED_10) ||
+		    (phy_data->port_speeds & XGBE_PHY_PORT_SPEED_100) ||
 		    (phy_data->port_speeds & XGBE_PHY_PORT_SPEED_1000) ||
+		    (phy_data->port_speeds & XGBE_PHY_PORT_SPEED_2500) ||
 		    (phy_data->port_speeds & XGBE_PHY_PORT_SPEED_10000))
 			return false;
 		break;
@@ -2652,7 +3087,8 @@ static bool xgbe_phy_port_mode_mismatch(struct xgbe_prv_data *pdata)
 			return false;
 		break;
 	case XGBE_PORT_MODE_SFP:
-		if ((phy_data->port_speeds & XGBE_PHY_PORT_SPEED_100) ||
+		if ((phy_data->port_speeds & XGBE_PHY_PORT_SPEED_10) ||
+		    (phy_data->port_speeds & XGBE_PHY_PORT_SPEED_100) ||
 		    (phy_data->port_speeds & XGBE_PHY_PORT_SPEED_1000) ||
 		    (phy_data->port_speeds & XGBE_PHY_PORT_SPEED_10000))
 			return false;
@@ -2670,6 +3106,7 @@ static bool xgbe_phy_conn_type_mismatch(struct xgbe_prv_data *pdata)
 
 	switch (phy_data->port_mode) {
 	case XGBE_PORT_MODE_BACKPLANE:
+	case XGBE_PORT_MODE_BACKPLANE_NO_AUTONEG:
 	case XGBE_PORT_MODE_BACKPLANE_2500:
 		if (phy_data->conn_type == XGBE_CONN_TYPE_BACKPLANE)
 			return false;
@@ -2695,17 +3132,111 @@ static bool xgbe_phy_conn_type_mismatch(struct xgbe_prv_data *pdata)
 
 static bool xgbe_phy_port_enabled(struct xgbe_prv_data *pdata)
 {
-	unsigned int reg;
-
-	reg = XP_IOREAD(pdata, XP_PROP_0);
-	if (!XP_GET_BITS(reg, XP_PROP_0, PORT_SPEEDS))
+	if (!XP_GET_BITS(pdata->pp0, XP_PROP_0, PORT_SPEEDS))
 		return false;
-	if (!XP_GET_BITS(reg, XP_PROP_0, CONN_TYPE))
+	if (!XP_GET_BITS(pdata->pp0, XP_PROP_0, CONN_TYPE))
 		return false;
 
 	return true;
 }
 
+static void xgbe_phy_cdr_track(struct xgbe_prv_data *pdata)
+{
+	struct xgbe_phy_data *phy_data = pdata->phy_data;
+
+	if (!pdata->debugfs_an_cdr_workaround)
+		return;
+
+	if (!phy_data->phy_cdr_notrack)
+		return;
+
+	usleep_range(phy_data->phy_cdr_delay,
+		     phy_data->phy_cdr_delay + 500);
+
+	XMDIO_WRITE_BITS(pdata, MDIO_MMD_PMAPMD, MDIO_VEND2_PMA_CDR_CONTROL,
+			 XGBE_PMA_CDR_TRACK_EN_MASK,
+			 XGBE_PMA_CDR_TRACK_EN_ON);
+
+	phy_data->phy_cdr_notrack = 0;
+}
+
+static void xgbe_phy_cdr_notrack(struct xgbe_prv_data *pdata)
+{
+	struct xgbe_phy_data *phy_data = pdata->phy_data;
+
+	if (!pdata->debugfs_an_cdr_workaround)
+		return;
+
+	if (phy_data->phy_cdr_notrack)
+		return;
+
+	XMDIO_WRITE_BITS(pdata, MDIO_MMD_PMAPMD, MDIO_VEND2_PMA_CDR_CONTROL,
+			 XGBE_PMA_CDR_TRACK_EN_MASK,
+			 XGBE_PMA_CDR_TRACK_EN_OFF);
+
+	xgbe_phy_rrc(pdata);
+
+	phy_data->phy_cdr_notrack = 1;
+}
+
+static void xgbe_phy_kr_training_post(struct xgbe_prv_data *pdata)
+{
+	if (!pdata->debugfs_an_cdr_track_early)
+		xgbe_phy_cdr_track(pdata);
+}
+
+static void xgbe_phy_kr_training_pre(struct xgbe_prv_data *pdata)
+{
+	if (pdata->debugfs_an_cdr_track_early)
+		xgbe_phy_cdr_track(pdata);
+}
+
+static void xgbe_phy_an_post(struct xgbe_prv_data *pdata)
+{
+	struct xgbe_phy_data *phy_data = pdata->phy_data;
+
+	switch (pdata->an_mode) {
+	case XGBE_AN_MODE_CL73:
+	case XGBE_AN_MODE_CL73_REDRV:
+		if (phy_data->cur_mode != XGBE_MODE_KR)
+			break;
+
+		xgbe_phy_cdr_track(pdata);
+
+		switch (pdata->an_result) {
+		case XGBE_AN_READY:
+		case XGBE_AN_COMPLETE:
+			break;
+		default:
+			if (phy_data->phy_cdr_delay < XGBE_CDR_DELAY_MAX)
+				phy_data->phy_cdr_delay += XGBE_CDR_DELAY_INC;
+			else
+				phy_data->phy_cdr_delay = XGBE_CDR_DELAY_INIT;
+			break;
+		}
+		break;
+	default:
+		break;
+	}
+}
+
+static void xgbe_phy_an_pre(struct xgbe_prv_data *pdata)
+{
+	struct xgbe_phy_data *phy_data = pdata->phy_data;
+
+	switch (pdata->an_mode) {
+	case XGBE_AN_MODE_CL73:
+	case XGBE_AN_MODE_CL73_REDRV:
+		if (phy_data->cur_mode != XGBE_MODE_KR)
+			break;
+
+		xgbe_phy_cdr_notrack(pdata);
+		break;
+	default:
+		break;
+	}
+}
+
 static void xgbe_phy_stop(struct xgbe_prv_data *pdata)
 {
 	struct xgbe_phy_data *phy_data = pdata->phy_data;
@@ -2717,6 +3248,9 @@ static void xgbe_phy_stop(struct xgbe_prv_data *pdata)
 	xgbe_phy_sfp_reset(phy_data);
 	xgbe_phy_sfp_mod_absent(pdata);
 
+	/* Reset CDR support */
+	xgbe_phy_cdr_track(pdata);
+
 	/* Power off the PHY */
 	xgbe_phy_power_off(pdata);
 
@@ -2749,6 +3283,9 @@ static int xgbe_phy_start(struct xgbe_prv_data *pdata)
 	/* Start in highest supported mode */
 	xgbe_phy_set_mode(pdata, phy_data->start_mode);
 
+	/* Reset CDR support */
+	xgbe_phy_cdr_track(pdata);
+
 	/* After starting the I2C controller, we can check for an SFP */
 	switch (phy_data->port_mode) {
 	case XGBE_PORT_MODE_SFP:
@@ -2803,9 +3340,9 @@ static void xgbe_phy_exit(struct xgbe_prv_data *pdata)
 
 static int xgbe_phy_init(struct xgbe_prv_data *pdata)
 {
+	struct ethtool_link_ksettings *lks = &pdata->phy.lks;
 	struct xgbe_phy_data *phy_data;
 	struct mii_bus *mii;
-	unsigned int reg;
 	int ret;
 
 	/* Check if enabled */
@@ -2824,12 +3361,11 @@ static int xgbe_phy_init(struct xgbe_prv_data *pdata)
 		return -ENOMEM;
 	pdata->phy_data = phy_data;
 
-	reg = XP_IOREAD(pdata, XP_PROP_0);
-	phy_data->port_mode = XP_GET_BITS(reg, XP_PROP_0, PORT_MODE);
-	phy_data->port_id = XP_GET_BITS(reg, XP_PROP_0, PORT_ID);
-	phy_data->port_speeds = XP_GET_BITS(reg, XP_PROP_0, PORT_SPEEDS);
-	phy_data->conn_type = XP_GET_BITS(reg, XP_PROP_0, CONN_TYPE);
-	phy_data->mdio_addr = XP_GET_BITS(reg, XP_PROP_0, MDIO_ADDR);
+	phy_data->port_mode = XP_GET_BITS(pdata->pp0, XP_PROP_0, PORT_MODE);
+	phy_data->port_id = XP_GET_BITS(pdata->pp0, XP_PROP_0, PORT_ID);
+	phy_data->port_speeds = XP_GET_BITS(pdata->pp0, XP_PROP_0, PORT_SPEEDS);
+	phy_data->conn_type = XP_GET_BITS(pdata->pp0, XP_PROP_0, CONN_TYPE);
+	phy_data->mdio_addr = XP_GET_BITS(pdata->pp0, XP_PROP_0, MDIO_ADDR);
 	if (netif_msg_probe(pdata)) {
 		dev_dbg(pdata->dev, "port mode=%u\n", phy_data->port_mode);
 		dev_dbg(pdata->dev, "port id=%u\n", phy_data->port_id);
@@ -2838,12 +3374,11 @@ static int xgbe_phy_init(struct xgbe_prv_data *pdata)
 		dev_dbg(pdata->dev, "mdio addr=%u\n", phy_data->mdio_addr);
 	}
 
-	reg = XP_IOREAD(pdata, XP_PROP_4);
-	phy_data->redrv = XP_GET_BITS(reg, XP_PROP_4, REDRV_PRESENT);
-	phy_data->redrv_if = XP_GET_BITS(reg, XP_PROP_4, REDRV_IF);
-	phy_data->redrv_addr = XP_GET_BITS(reg, XP_PROP_4, REDRV_ADDR);
-	phy_data->redrv_lane = XP_GET_BITS(reg, XP_PROP_4, REDRV_LANE);
-	phy_data->redrv_model = XP_GET_BITS(reg, XP_PROP_4, REDRV_MODEL);
+	phy_data->redrv = XP_GET_BITS(pdata->pp4, XP_PROP_4, REDRV_PRESENT);
+	phy_data->redrv_if = XP_GET_BITS(pdata->pp4, XP_PROP_4, REDRV_IF);
+	phy_data->redrv_addr = XP_GET_BITS(pdata->pp4, XP_PROP_4, REDRV_ADDR);
+	phy_data->redrv_lane = XP_GET_BITS(pdata->pp4, XP_PROP_4, REDRV_LANE);
+	phy_data->redrv_model = XP_GET_BITS(pdata->pp4, XP_PROP_4, REDRV_MODEL);
 	if (phy_data->redrv && netif_msg_probe(pdata)) {
 		dev_dbg(pdata->dev, "redrv present\n");
 		dev_dbg(pdata->dev, "redrv i/f=%u\n", phy_data->redrv_if);
@@ -2882,32 +3417,35 @@ static int xgbe_phy_init(struct xgbe_prv_data *pdata)
 	phy_data->cur_mode = XGBE_MODE_UNKNOWN;
 
 	/* Initialize supported features */
-	pdata->phy.supported = 0;
+	XGBE_ZERO_SUP(lks);
 
 	switch (phy_data->port_mode) {
 	/* Backplane support */
 	case XGBE_PORT_MODE_BACKPLANE:
-		pdata->phy.supported |= SUPPORTED_Autoneg;
-		pdata->phy.supported |= SUPPORTED_Pause | SUPPORTED_Asym_Pause;
-		pdata->phy.supported |= SUPPORTED_Backplane;
+		XGBE_SET_SUP(lks, Autoneg);
+		fallthrough;
+	case XGBE_PORT_MODE_BACKPLANE_NO_AUTONEG:
+		XGBE_SET_SUP(lks, Pause);
+		XGBE_SET_SUP(lks, Asym_Pause);
+		XGBE_SET_SUP(lks, Backplane);
 		if (phy_data->port_speeds & XGBE_PHY_PORT_SPEED_1000) {
-			pdata->phy.supported |= SUPPORTED_1000baseKX_Full;
+			XGBE_SET_SUP(lks, 1000baseKX_Full);
 			phy_data->start_mode = XGBE_MODE_KX_1000;
 		}
 		if (phy_data->port_speeds & XGBE_PHY_PORT_SPEED_10000) {
-			pdata->phy.supported |= SUPPORTED_10000baseKR_Full;
+			XGBE_SET_SUP(lks, 10000baseKR_Full);
 			if (pdata->fec_ability & MDIO_PMA_10GBR_FECABLE_ABLE)
-				pdata->phy.supported |=
-					SUPPORTED_10000baseR_FEC;
+				XGBE_SET_SUP(lks, 10000baseR_FEC);
 			phy_data->start_mode = XGBE_MODE_KR;
 		}
 
 		phy_data->phydev_mode = XGBE_MDIO_MODE_NONE;
 		break;
 	case XGBE_PORT_MODE_BACKPLANE_2500:
-		pdata->phy.supported |= SUPPORTED_Pause | SUPPORTED_Asym_Pause;
-		pdata->phy.supported |= SUPPORTED_Backplane;
-		pdata->phy.supported |= SUPPORTED_2500baseX_Full;
+		XGBE_SET_SUP(lks, Pause);
+		XGBE_SET_SUP(lks, Asym_Pause);
+		XGBE_SET_SUP(lks, Backplane);
+		XGBE_SET_SUP(lks, 2500baseX_Full);
 		phy_data->start_mode = XGBE_MODE_KX_2500;
 
 		phy_data->phydev_mode = XGBE_MDIO_MODE_NONE;
@@ -2915,15 +3453,20 @@ static int xgbe_phy_init(struct xgbe_prv_data *pdata)
 
 	/* MDIO 1GBase-T support */
 	case XGBE_PORT_MODE_1000BASE_T:
-		pdata->phy.supported |= SUPPORTED_Autoneg;
-		pdata->phy.supported |= SUPPORTED_Pause | SUPPORTED_Asym_Pause;
-		pdata->phy.supported |= SUPPORTED_TP;
+		XGBE_SET_SUP(lks, Autoneg);
+		XGBE_SET_SUP(lks, Pause);
+		XGBE_SET_SUP(lks, Asym_Pause);
+		XGBE_SET_SUP(lks, TP);
+		if (phy_data->port_speeds & XGBE_PHY_PORT_SPEED_10) {
+			XGBE_SET_SUP(lks, 10baseT_Full);
+			phy_data->start_mode = XGBE_MODE_SGMII_10;
+		}
 		if (phy_data->port_speeds & XGBE_PHY_PORT_SPEED_100) {
-			pdata->phy.supported |= SUPPORTED_100baseT_Full;
+			XGBE_SET_SUP(lks, 100baseT_Full);
 			phy_data->start_mode = XGBE_MODE_SGMII_100;
 		}
 		if (phy_data->port_speeds & XGBE_PHY_PORT_SPEED_1000) {
-			pdata->phy.supported |= SUPPORTED_1000baseT_Full;
+			XGBE_SET_SUP(lks, 1000baseT_Full);
 			phy_data->start_mode = XGBE_MODE_SGMII_1000;
 		}
 
@@ -2932,10 +3475,11 @@ static int xgbe_phy_init(struct xgbe_prv_data *pdata)
 
 	/* MDIO Base-X support */
 	case XGBE_PORT_MODE_1000BASE_X:
-		pdata->phy.supported |= SUPPORTED_Autoneg;
-		pdata->phy.supported |= SUPPORTED_Pause | SUPPORTED_Asym_Pause;
-		pdata->phy.supported |= SUPPORTED_FIBRE;
-		pdata->phy.supported |= SUPPORTED_1000baseT_Full;
+		XGBE_SET_SUP(lks, Autoneg);
+		XGBE_SET_SUP(lks, Pause);
+		XGBE_SET_SUP(lks, Asym_Pause);
+		XGBE_SET_SUP(lks, FIBRE);
+		XGBE_SET_SUP(lks, 1000baseX_Full);
 		phy_data->start_mode = XGBE_MODE_X;
 
 		phy_data->phydev_mode = XGBE_MDIO_MODE_CL22;
@@ -2943,19 +3487,24 @@ static int xgbe_phy_init(struct xgbe_prv_data *pdata)
 
 	/* MDIO NBase-T support */
 	case XGBE_PORT_MODE_NBASE_T:
-		pdata->phy.supported |= SUPPORTED_Autoneg;
-		pdata->phy.supported |= SUPPORTED_Pause | SUPPORTED_Asym_Pause;
-		pdata->phy.supported |= SUPPORTED_TP;
+		XGBE_SET_SUP(lks, Autoneg);
+		XGBE_SET_SUP(lks, Pause);
+		XGBE_SET_SUP(lks, Asym_Pause);
+		XGBE_SET_SUP(lks, TP);
+		if (phy_data->port_speeds & XGBE_PHY_PORT_SPEED_10) {
+			XGBE_SET_SUP(lks, 10baseT_Full);
+			phy_data->start_mode = XGBE_MODE_SGMII_10;
+		}
 		if (phy_data->port_speeds & XGBE_PHY_PORT_SPEED_100) {
-			pdata->phy.supported |= SUPPORTED_100baseT_Full;
+			XGBE_SET_SUP(lks, 100baseT_Full);
 			phy_data->start_mode = XGBE_MODE_SGMII_100;
 		}
 		if (phy_data->port_speeds & XGBE_PHY_PORT_SPEED_1000) {
-			pdata->phy.supported |= SUPPORTED_1000baseT_Full;
+			XGBE_SET_SUP(lks, 1000baseT_Full);
 			phy_data->start_mode = XGBE_MODE_SGMII_1000;
 		}
 		if (phy_data->port_speeds & XGBE_PHY_PORT_SPEED_2500) {
-			pdata->phy.supported |= SUPPORTED_2500baseX_Full;
+			XGBE_SET_SUP(lks, 2500baseT_Full);
 			phy_data->start_mode = XGBE_MODE_KX_2500;
 		}
 
@@ -2964,33 +3513,46 @@ static int xgbe_phy_init(struct xgbe_prv_data *pdata)
 
 	/* 10GBase-T support */
 	case XGBE_PORT_MODE_10GBASE_T:
-		pdata->phy.supported |= SUPPORTED_Autoneg;
-		pdata->phy.supported |= SUPPORTED_Pause | SUPPORTED_Asym_Pause;
-		pdata->phy.supported |= SUPPORTED_TP;
+		XGBE_SET_SUP(lks, Autoneg);
+		XGBE_SET_SUP(lks, Pause);
+		XGBE_SET_SUP(lks, Asym_Pause);
+		XGBE_SET_SUP(lks, TP);
+		if (phy_data->port_speeds & XGBE_PHY_PORT_SPEED_10) {
+			XGBE_SET_SUP(lks, 10baseT_Full);
+			phy_data->start_mode = XGBE_MODE_SGMII_10;
+		}
 		if (phy_data->port_speeds & XGBE_PHY_PORT_SPEED_100) {
-			pdata->phy.supported |= SUPPORTED_100baseT_Full;
+			XGBE_SET_SUP(lks, 100baseT_Full);
 			phy_data->start_mode = XGBE_MODE_SGMII_100;
 		}
 		if (phy_data->port_speeds & XGBE_PHY_PORT_SPEED_1000) {
-			pdata->phy.supported |= SUPPORTED_1000baseT_Full;
+			XGBE_SET_SUP(lks, 1000baseT_Full);
 			phy_data->start_mode = XGBE_MODE_SGMII_1000;
 		}
+		if (phy_data->port_speeds & XGBE_PHY_PORT_SPEED_2500) {
+			XGBE_SET_SUP(lks, 2500baseT_Full);
+			phy_data->start_mode = XGBE_MODE_KX_2500;
+		}
 		if (phy_data->port_speeds & XGBE_PHY_PORT_SPEED_10000) {
-			pdata->phy.supported |= SUPPORTED_10000baseT_Full;
+			XGBE_SET_SUP(lks, 10000baseT_Full);
 			phy_data->start_mode = XGBE_MODE_KR;
 		}
 
-		phy_data->phydev_mode = XGBE_MDIO_MODE_NONE;
+		phy_data->phydev_mode = XGBE_MDIO_MODE_CL45;
 		break;
 
 	/* 10GBase-R support */
 	case XGBE_PORT_MODE_10GBASE_R:
-		pdata->phy.supported |= SUPPORTED_Autoneg;
-		pdata->phy.supported |= SUPPORTED_Pause | SUPPORTED_Asym_Pause;
-		pdata->phy.supported |= SUPPORTED_TP;
-		pdata->phy.supported |= SUPPORTED_10000baseT_Full;
+		XGBE_SET_SUP(lks, Autoneg);
+		XGBE_SET_SUP(lks, Pause);
+		XGBE_SET_SUP(lks, Asym_Pause);
+		XGBE_SET_SUP(lks, FIBRE);
+		XGBE_SET_SUP(lks, 10000baseSR_Full);
+		XGBE_SET_SUP(lks, 10000baseLR_Full);
+		XGBE_SET_SUP(lks, 10000baseLRM_Full);
+		XGBE_SET_SUP(lks, 10000baseER_Full);
 		if (pdata->fec_ability & MDIO_PMA_10GBR_FECABLE_ABLE)
-			pdata->phy.supported |= SUPPORTED_10000baseR_FEC;
+			XGBE_SET_SUP(lks, 10000baseR_FEC);
 		phy_data->start_mode = XGBE_MODE_SFI;
 
 		phy_data->phydev_mode = XGBE_MDIO_MODE_NONE;
@@ -2998,25 +3560,19 @@ static int xgbe_phy_init(struct xgbe_prv_data *pdata)
 
 	/* SFP support */
 	case XGBE_PORT_MODE_SFP:
-		pdata->phy.supported |= SUPPORTED_Autoneg;
-		pdata->phy.supported |= SUPPORTED_Pause | SUPPORTED_Asym_Pause;
-		pdata->phy.supported |= SUPPORTED_TP;
-		pdata->phy.supported |= SUPPORTED_FIBRE;
-		if (phy_data->port_speeds & XGBE_PHY_PORT_SPEED_100) {
-			pdata->phy.supported |= SUPPORTED_100baseT_Full;
+		XGBE_SET_SUP(lks, Autoneg);
+		XGBE_SET_SUP(lks, Pause);
+		XGBE_SET_SUP(lks, Asym_Pause);
+		XGBE_SET_SUP(lks, TP);
+		XGBE_SET_SUP(lks, FIBRE);
+		if (phy_data->port_speeds & XGBE_PHY_PORT_SPEED_10)
+			phy_data->start_mode = XGBE_MODE_SGMII_10;
+		if (phy_data->port_speeds & XGBE_PHY_PORT_SPEED_100)
 			phy_data->start_mode = XGBE_MODE_SGMII_100;
-		}
-		if (phy_data->port_speeds & XGBE_PHY_PORT_SPEED_1000) {
-			pdata->phy.supported |= SUPPORTED_1000baseT_Full;
+		if (phy_data->port_speeds & XGBE_PHY_PORT_SPEED_1000)
 			phy_data->start_mode = XGBE_MODE_SGMII_1000;
-		}
-		if (phy_data->port_speeds & XGBE_PHY_PORT_SPEED_10000) {
-			pdata->phy.supported |= SUPPORTED_10000baseT_Full;
+		if (phy_data->port_speeds & XGBE_PHY_PORT_SPEED_10000)
 			phy_data->start_mode = XGBE_MODE_SFI;
-			if (pdata->fec_ability & MDIO_PMA_10GBR_FECABLE_ABLE)
-				pdata->phy.supported |=
-					SUPPORTED_10000baseR_FEC;
-		}
 
 		phy_data->phydev_mode = XGBE_MDIO_MODE_CL22;
 
@@ -3027,8 +3583,9 @@ static int xgbe_phy_init(struct xgbe_prv_data *pdata)
 	}
 
 	if (netif_msg_probe(pdata))
-		dev_dbg(pdata->dev, "phy supported=%#x\n",
-			pdata->phy.supported);
+		dev_dbg(pdata->dev, "phy supported=0x%*pb\n",
+			__ETHTOOL_LINK_MODE_MASK_NBITS,
+			lks->link_modes.supported);
 
 	if ((phy_data->conn_type & XGBE_CONN_TYPE_MDIO) &&
 	    (phy_data->phydev_mode != XGBE_MDIO_MODE_NONE)) {
@@ -3053,6 +3610,8 @@ static int xgbe_phy_init(struct xgbe_prv_data *pdata)
 		}
 	}
 
+	phy_data->phy_cdr_delay = XGBE_CDR_DELAY_INIT;
+
 	/* Register for driving external PHYs */
 	mii = devm_mdiobus_alloc(pdata->dev);
 	if (!mii) {
@@ -3062,8 +3621,10 @@ static int xgbe_phy_init(struct xgbe_prv_data *pdata)
 
 	mii->priv = pdata;
 	mii->name = "amd-xgbe-mii";
-	mii->read = xgbe_phy_mii_read;
-	mii->write = xgbe_phy_mii_write;
+	mii->read = xgbe_phy_mii_read_c22;
+	mii->write = xgbe_phy_mii_write_c22;
+	mii->read_c45 = xgbe_phy_mii_read_c45;
+	mii->write_c45 = xgbe_phy_mii_write_c45;
 	mii->parent = pdata->dev;
 	mii->phy_mask = ~0;
 	snprintf(mii->id, sizeof(mii->id), "%s", dev_name(pdata->dev));
@@ -3105,4 +3666,13 @@ void xgbe_init_function_ptrs_phy_v2(struct xgbe_phy_if *phy_if)
 	phy_impl->an_advertising	= xgbe_phy_an_advertising;
 
 	phy_impl->an_outcome		= xgbe_phy_an_outcome;
+
+	phy_impl->an_pre		= xgbe_phy_an_pre;
+	phy_impl->an_post		= xgbe_phy_an_post;
+
+	phy_impl->kr_training_pre	= xgbe_phy_kr_training_pre;
+	phy_impl->kr_training_post	= xgbe_phy_kr_training_post;
+
+	phy_impl->module_info		= xgbe_phy_module_info;
+	phy_impl->module_eeprom		= xgbe_phy_module_eeprom;
 }
diff --git a/drivers/net/ethernet/amd/xgbe/xgbe-platform.c b/drivers/net/ethernet/amd/xgbe/xgbe-platform.c
index 84d4c51cab8c..47d53e59ccf6 100644
--- a/drivers/net/ethernet/amd/xgbe/xgbe-platform.c
+++ b/drivers/net/ethernet/amd/xgbe/xgbe-platform.c
@@ -1,117 +1,8 @@
+// SPDX-License-Identifier: (GPL-2.0-or-later OR BSD-3-Clause)
 /*
- * AMD 10Gb Ethernet driver
- *
- * This file is available to you under your choice of the following two
- * licenses:
- *
- * License 1: GPLv2
- *
- * Copyright (c) 2014-2016 Advanced Micro Devices, Inc.
- *
- * This file is free software; you may copy, redistribute and/or modify
- * it under the terms of the GNU General Public License as published by
- * the Free Software Foundation, either version 2 of the License, or (at
- * your option) any later version.
- *
- * This file is distributed in the hope that it will be useful, but
- * WITHOUT ANY WARRANTY; without even the implied warranty of
- * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
- * General Public License for more details.
- *
- * You should have received a copy of the GNU General Public License
- * along with this program.  If not, see <http://www.gnu.org/licenses/>.
- *
- * This file incorporates work covered by the following copyright and
- * permission notice:
- *     The Synopsys DWC ETHER XGMAC Software Driver and documentation
- *     (hereinafter "Software") is an unsupported proprietary work of Synopsys,
- *     Inc. unless otherwise expressly agreed to in writing between Synopsys
- *     and you.
- *
- *     The Software IS NOT an item of Licensed Software or Licensed Product
- *     under any End User Software License Agreement or Agreement for Licensed
- *     Product with Synopsys or any supplement thereto.  Permission is hereby
- *     granted, free of charge, to any person obtaining a copy of this software
- *     annotated with this license and the Software, to deal in the Software
- *     without restriction, including without limitation the rights to use,
- *     copy, modify, merge, publish, distribute, sublicense, and/or sell copies
- *     of the Software, and to permit persons to whom the Software is furnished
- *     to do so, subject to the following conditions:
- *
- *     The above copyright notice and this permission notice shall be included
- *     in all copies or substantial portions of the Software.
- *
- *     THIS SOFTWARE IS BEING DISTRIBUTED BY SYNOPSYS SOLELY ON AN "AS IS"
- *     BASIS AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED
- *     TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A
- *     PARTICULAR PURPOSE ARE HEREBY DISCLAIMED. IN NO EVENT SHALL SYNOPSYS
- *     BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR
- *     CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF
- *     SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS
- *     INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN
- *     CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE)
- *     ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF
- *     THE POSSIBILITY OF SUCH DAMAGE.
- *
- *
- * License 2: Modified BSD
- *
- * Copyright (c) 2014-2016 Advanced Micro Devices, Inc.
- * All rights reserved.
- *
- * Redistribution and use in source and binary forms, with or without
- * modification, are permitted provided that the following conditions are met:
- *     * Redistributions of source code must retain the above copyright
- *       notice, this list of conditions and the following disclaimer.
- *     * Redistributions in binary form must reproduce the above copyright
- *       notice, this list of conditions and the following disclaimer in the
- *       documentation and/or other materials provided with the distribution.
- *     * Neither the name of Advanced Micro Devices, Inc. nor the
- *       names of its contributors may be used to endorse or promote products
- *       derived from this software without specific prior written permission.
- *
- * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS"
- * AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
- * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
- * ARE DISCLAIMED. IN NO EVENT SHALL <COPYRIGHT HOLDER> BE LIABLE FOR ANY
- * DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES
- * (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES;
- * LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND
- * ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
- * (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF
- * THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
- *
- * This file incorporates work covered by the following copyright and
- * permission notice:
- *     The Synopsys DWC ETHER XGMAC Software Driver and documentation
- *     (hereinafter "Software") is an unsupported proprietary work of Synopsys,
- *     Inc. unless otherwise expressly agreed to in writing between Synopsys
- *     and you.
- *
- *     The Software IS NOT an item of Licensed Software or Licensed Product
- *     under any End User Software License Agreement or Agreement for Licensed
- *     Product with Synopsys or any supplement thereto.  Permission is hereby
- *     granted, free of charge, to any person obtaining a copy of this software
- *     annotated with this license and the Software, to deal in the Software
- *     without restriction, including without limitation the rights to use,
- *     copy, modify, merge, publish, distribute, sublicense, and/or sell copies
- *     of the Software, and to permit persons to whom the Software is furnished
- *     to do so, subject to the following conditions:
- *
- *     The above copyright notice and this permission notice shall be included
- *     in all copies or substantial portions of the Software.
- *
- *     THIS SOFTWARE IS BEING DISTRIBUTED BY SYNOPSYS SOLELY ON AN "AS IS"
- *     BASIS AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED
- *     TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A
- *     PARTICULAR PURPOSE ARE HEREBY DISCLAIMED. IN NO EVENT SHALL SYNOPSYS
- *     BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR
- *     CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF
- *     SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS
- *     INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN
- *     CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE)
- *     ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF
- *     THE POSSIBILITY OF SUCH DAMAGE.
+ * Copyright (c) 2014-2025, Advanced Micro Devices, Inc.
+ * Copyright (c) 2014, Synopsys, Inc.
+ * All rights reserved
  */
 
 #include <linux/module.h>
@@ -123,9 +14,7 @@
 #include <linux/io.h>
 #include <linux/of.h>
 #include <linux/of_net.h>
-#include <linux/of_address.h>
 #include <linux/of_platform.h>
-#include <linux/of_device.h>
 #include <linux/clk.h>
 #include <linux/property.h>
 #include <linux/acpi.h>
@@ -135,17 +24,6 @@
 #include "xgbe-common.h"
 
 #ifdef CONFIG_ACPI
-static const struct acpi_device_id xgbe_acpi_match[];
-
-static struct xgbe_version_data *xgbe_acpi_vdata(struct xgbe_prv_data *pdata)
-{
-	const struct acpi_device_id *id;
-
-	id = acpi_match_device(xgbe_acpi_match, pdata->dev);
-
-	return id ? (struct xgbe_version_data *)id->driver_data : NULL;
-}
-
 static int xgbe_acpi_support(struct xgbe_prv_data *pdata)
 {
 	struct device *dev = pdata->dev;
@@ -173,11 +51,6 @@ static int xgbe_acpi_support(struct xgbe_prv_data *pdata)
 	return 0;
 }
 #else   /* CONFIG_ACPI */
-static struct xgbe_version_data *xgbe_acpi_vdata(struct xgbe_prv_data *pdata)
-{
-	return NULL;
-}
-
 static int xgbe_acpi_support(struct xgbe_prv_data *pdata)
 {
 	return -EINVAL;
@@ -185,17 +58,6 @@ static int xgbe_acpi_support(struct xgbe_prv_data *pdata)
 #endif  /* CONFIG_ACPI */
 
 #ifdef CONFIG_OF
-static const struct of_device_id xgbe_of_match[];
-
-static struct xgbe_version_data *xgbe_of_vdata(struct xgbe_prv_data *pdata)
-{
-	const struct of_device_id *id;
-
-	id = of_match_device(xgbe_of_match, pdata->dev);
-
-	return id ? (struct xgbe_version_data *)id->data : NULL;
-}
-
 static int xgbe_of_support(struct xgbe_prv_data *pdata)
 {
 	struct device *dev = pdata->dev;
@@ -244,11 +106,6 @@ static struct platform_device *xgbe_of_get_phy_pdev(struct xgbe_prv_data *pdata)
 	return phy_pdev;
 }
 #else   /* CONFIG_OF */
-static struct xgbe_version_data *xgbe_of_vdata(struct xgbe_prv_data *pdata)
-{
-	return NULL;
-}
-
 static int xgbe_of_support(struct xgbe_prv_data *pdata)
 {
 	return -EINVAL;
@@ -290,18 +147,11 @@ static struct platform_device *xgbe_get_phy_pdev(struct xgbe_prv_data *pdata)
 	return phy_pdev;
 }
 
-static struct xgbe_version_data *xgbe_get_vdata(struct xgbe_prv_data *pdata)
-{
-	return pdata->use_acpi ? xgbe_acpi_vdata(pdata)
-			       : xgbe_of_vdata(pdata);
-}
-
 static int xgbe_platform_probe(struct platform_device *pdev)
 {
 	struct xgbe_prv_data *pdata;
 	struct device *dev = &pdev->dev;
 	struct platform_device *phy_pdev;
-	struct resource *res;
 	const char *phy_mode;
 	unsigned int phy_memnum, phy_irqnum;
 	unsigned int dma_irqnum, dma_irqend;
@@ -322,7 +172,7 @@ static int xgbe_platform_probe(struct platform_device *pdev)
 	pdata->use_acpi = dev->of_node ? 0 : 1;
 
 	/* Get the version data */
-	pdata->vdata = xgbe_get_vdata(pdata);
+	pdata->vdata = (struct xgbe_version_data *)device_get_match_data(dev);
 
 	phy_pdev = xgbe_get_phy_pdev(pdata);
 	if (!phy_pdev) {
@@ -339,7 +189,7 @@ static int xgbe_platform_probe(struct platform_device *pdev)
 		 *   the PHY resources listed last
 		 */
 		phy_memnum = xgbe_resource_count(pdev, IORESOURCE_MEM) - 3;
-		phy_irqnum = xgbe_resource_count(pdev, IORESOURCE_IRQ) - 1;
+		phy_irqnum = platform_irq_count(pdev) - 1;
 		dma_irqnum = 1;
 		dma_irqend = phy_irqnum;
 	} else {
@@ -349,12 +199,11 @@ static int xgbe_platform_probe(struct platform_device *pdev)
 		phy_memnum = 0;
 		phy_irqnum = 0;
 		dma_irqnum = 1;
-		dma_irqend = xgbe_resource_count(pdev, IORESOURCE_IRQ);
+		dma_irqend = platform_irq_count(pdev);
 	}
 
 	/* Obtain the mmio areas for the device */
-	res = platform_get_resource(pdev, IORESOURCE_MEM, 0);
-	pdata->xgmac_regs = devm_ioremap_resource(dev, res);
+	pdata->xgmac_regs = devm_platform_ioremap_resource(pdev, 0);
 	if (IS_ERR(pdata->xgmac_regs)) {
 		dev_err(dev, "xgmac ioremap failed\n");
 		ret = PTR_ERR(pdata->xgmac_regs);
@@ -363,8 +212,7 @@ static int xgbe_platform_probe(struct platform_device *pdev)
 	if (netif_msg_probe(pdata))
 		dev_dbg(dev, "xgmac_regs = %p\n", pdata->xgmac_regs);
 
-	res = platform_get_resource(pdev, IORESOURCE_MEM, 1);
-	pdata->xpcs_regs = devm_ioremap_resource(dev, res);
+	pdata->xpcs_regs = devm_platform_ioremap_resource(pdev, 1);
 	if (IS_ERR(pdata->xpcs_regs)) {
 		dev_err(dev, "xpcs ioremap failed\n");
 		ret = PTR_ERR(pdata->xpcs_regs);
@@ -373,8 +221,8 @@ static int xgbe_platform_probe(struct platform_device *pdev)
 	if (netif_msg_probe(pdata))
 		dev_dbg(dev, "xpcs_regs  = %p\n", pdata->xpcs_regs);
 
-	res = platform_get_resource(phy_pdev, IORESOURCE_MEM, phy_memnum++);
-	pdata->rxtx_regs = devm_ioremap_resource(dev, res);
+	pdata->rxtx_regs = devm_platform_ioremap_resource(phy_pdev,
+							  phy_memnum++);
 	if (IS_ERR(pdata->rxtx_regs)) {
 		dev_err(dev, "rxtx ioremap failed\n");
 		ret = PTR_ERR(pdata->rxtx_regs);
@@ -383,8 +231,8 @@ static int xgbe_platform_probe(struct platform_device *pdev)
 	if (netif_msg_probe(pdata))
 		dev_dbg(dev, "rxtx_regs  = %p\n", pdata->rxtx_regs);
 
-	res = platform_get_resource(phy_pdev, IORESOURCE_MEM, phy_memnum++);
-	pdata->sir0_regs = devm_ioremap_resource(dev, res);
+	pdata->sir0_regs = devm_platform_ioremap_resource(phy_pdev,
+							  phy_memnum++);
 	if (IS_ERR(pdata->sir0_regs)) {
 		dev_err(dev, "sir0 ioremap failed\n");
 		ret = PTR_ERR(pdata->sir0_regs);
@@ -393,8 +241,8 @@ static int xgbe_platform_probe(struct platform_device *pdev)
 	if (netif_msg_probe(pdata))
 		dev_dbg(dev, "sir0_regs  = %p\n", pdata->sir0_regs);
 
-	res = platform_get_resource(phy_pdev, IORESOURCE_MEM, phy_memnum++);
-	pdata->sir1_regs = devm_ioremap_resource(dev, res);
+	pdata->sir1_regs = devm_platform_ioremap_resource(phy_pdev,
+							  phy_memnum++);
 	if (IS_ERR(pdata->sir1_regs)) {
 		dev_err(dev, "sir1 ioremap failed\n");
 		ret = PTR_ERR(pdata->sir1_regs);
@@ -448,13 +296,11 @@ static int xgbe_platform_probe(struct platform_device *pdev)
 	}
 	pdata->coherent = (attr == DEV_DMA_COHERENT);
 	if (pdata->coherent) {
-		pdata->axdomain = XGBE_DMA_OS_AXDOMAIN;
-		pdata->arcache = XGBE_DMA_OS_ARCACHE;
-		pdata->awcache = XGBE_DMA_OS_AWCACHE;
+		pdata->arcr = XGBE_DMA_OS_ARCR;
+		pdata->awcr = XGBE_DMA_OS_AWCR;
 	} else {
-		pdata->axdomain = XGBE_DMA_SYS_AXDOMAIN;
-		pdata->arcache = XGBE_DMA_SYS_ARCACHE;
-		pdata->awcache = XGBE_DMA_SYS_AWCACHE;
+		pdata->arcr = XGBE_DMA_SYS_ARCR;
+		pdata->awcr = XGBE_DMA_SYS_AWCR;
 	}
 
 	/* Set the maximum fifo amounts */
@@ -469,10 +315,8 @@ static int xgbe_platform_probe(struct platform_device *pdev)
 
 	/* Get the device interrupt */
 	ret = platform_get_irq(pdev, 0);
-	if (ret < 0) {
-		dev_err(dev, "platform_get_irq 0 failed\n");
+	if (ret < 0)
 		goto err_io;
-	}
 	pdata->dev_irq = ret;
 
 	/* Get the per channel DMA interrupts */
@@ -481,12 +325,8 @@ static int xgbe_platform_probe(struct platform_device *pdev)
 
 		for (i = 0; (i < max) && (dma_irqnum < dma_irqend); i++) {
 			ret = platform_get_irq(pdata->platdev, dma_irqnum++);
-			if (ret < 0) {
-				netdev_err(pdata->netdev,
-					   "platform_get_irq %u failed\n",
-					   dma_irqnum - 1);
+			if (ret < 0)
 				goto err_io;
-			}
 
 			pdata->channel_irq[i] = ret;
 		}
@@ -498,10 +338,8 @@ static int xgbe_platform_probe(struct platform_device *pdev)
 
 	/* Get the auto-negotiation interrupt */
 	ret = platform_get_irq(phy_pdev, phy_irqnum++);
-	if (ret < 0) {
-		dev_err(dev, "platform_get_irq phy 0 failed\n");
+	if (ret < 0)
 		goto err_io;
-	}
 	pdata->an_irq = ret;
 
 	/* Configure the netdev resource */
@@ -525,7 +363,7 @@ err_alloc:
 	return ret;
 }
 
-static int xgbe_platform_remove(struct platform_device *pdev)
+static void xgbe_platform_remove(struct platform_device *pdev)
 {
 	struct xgbe_prv_data *pdata = platform_get_drvdata(pdev);
 
@@ -534,8 +372,6 @@ static int xgbe_platform_remove(struct platform_device *pdev)
 	platform_device_put(pdata->phy_platdev);
 
 	xgbe_free_pdata(pdata);
-
-	return 0;
 }
 
 #ifdef CONFIG_PM_SLEEP
@@ -593,7 +429,6 @@ static const struct xgbe_version_data xgbe_v1 = {
 	.tx_tstamp_workaround		= 1,
 };
 
-#ifdef CONFIG_ACPI
 static const struct acpi_device_id xgbe_acpi_match[] = {
 	{ .id = "AMDI8001",
 	  .driver_data = (kernel_ulong_t)&xgbe_v1 },
@@ -601,9 +436,7 @@ static const struct acpi_device_id xgbe_acpi_match[] = {
 };
 
 MODULE_DEVICE_TABLE(acpi, xgbe_acpi_match);
-#endif
 
-#ifdef CONFIG_OF
 static const struct of_device_id xgbe_of_match[] = {
 	{ .compatible = "amd,xgbe-seattle-v1a",
 	  .data = &xgbe_v1 },
@@ -611,7 +444,6 @@ static const struct of_device_id xgbe_of_match[] = {
 };
 
 MODULE_DEVICE_TABLE(of, xgbe_of_match);
-#endif
 
 static SIMPLE_DEV_PM_OPS(xgbe_platform_pm_ops,
 			 xgbe_platform_suspend, xgbe_platform_resume);
@@ -619,12 +451,8 @@ static SIMPLE_DEV_PM_OPS(xgbe_platform_pm_ops,
 static struct platform_driver xgbe_driver = {
 	.driver = {
 		.name = XGBE_DRV_NAME,
-#ifdef CONFIG_ACPI
 		.acpi_match_table = xgbe_acpi_match,
-#endif
-#ifdef CONFIG_OF
 		.of_match_table = xgbe_of_match,
-#endif
 		.pm = &xgbe_platform_pm_ops,
 	},
 	.probe = xgbe_platform_probe,
diff --git a/drivers/net/ethernet/amd/xgbe/xgbe-pps.c b/drivers/net/ethernet/amd/xgbe/xgbe-pps.c
new file mode 100644
index 000000000000..6d03ae7ab36f
--- /dev/null
+++ b/drivers/net/ethernet/amd/xgbe/xgbe-pps.c
@@ -0,0 +1,74 @@
+// SPDX-License-Identifier: (GPL-2.0-or-later OR BSD-3-Clause)
+/*
+ * Copyright (c) 2014-2025, Advanced Micro Devices, Inc.
+ * Copyright (c) 2014, Synopsys, Inc.
+ * All rights reserved
+ *
+ * Author: Raju Rangoju <Raju.Rangoju@amd.com>
+ */
+
+#include "xgbe.h"
+#include "xgbe-common.h"
+
+static u32 get_pps_mask(unsigned int x)
+{
+	return GENMASK(PPS_MAXIDX(x), PPS_MINIDX(x));
+}
+
+static u32 get_pps_cmd(unsigned int x, u32 val)
+{
+	return (val & GENMASK(3, 0)) << PPS_MINIDX(x);
+}
+
+static u32 get_target_mode_sel(unsigned int x, u32 val)
+{
+	return (val & GENMASK(1, 0)) << (PPS_MAXIDX(x) - 2);
+}
+
+int xgbe_pps_config(struct xgbe_prv_data *pdata,
+		    struct xgbe_pps_config *cfg, int index, bool on)
+{
+	unsigned int ppscr = 0;
+	unsigned int tnsec;
+	u64 period;
+
+	/* Check if target time register is busy */
+	tnsec = XGMAC_IOREAD(pdata, MAC_PPSx_TTNSR(index));
+	if (XGMAC_GET_BITS(tnsec, MAC_PPSx_TTNSR, TRGTBUSY0))
+		return -EBUSY;
+
+	ppscr = XGMAC_IOREAD(pdata, MAC_PPSCR);
+	ppscr &= ~get_pps_mask(index);
+
+	if (!on) {
+		/* Disable PPS output */
+		ppscr |= get_pps_cmd(index, XGBE_PPSCMD_STOP);
+		ppscr |= PPSEN0;
+		XGMAC_IOWRITE(pdata, MAC_PPSCR, ppscr);
+
+		return 0;
+	}
+
+	/* Configure start time */
+	XGMAC_IOWRITE(pdata, MAC_PPSx_TTSR(index), cfg->start.tv_sec);
+	XGMAC_IOWRITE(pdata, MAC_PPSx_TTNSR(index), cfg->start.tv_nsec);
+
+	period = cfg->period.tv_sec * NSEC_PER_SEC + cfg->period.tv_nsec;
+	period = div_u64(period, XGBE_V2_TSTAMP_SSINC);
+
+	if (period < 4)
+		return -EINVAL;
+
+	/* Configure interval and pulse width (50% duty cycle) */
+	XGMAC_IOWRITE(pdata, MAC_PPSx_INTERVAL(index), period - 1);
+	XGMAC_IOWRITE(pdata, MAC_PPSx_WIDTH(index), (period >> 1) - 1);
+
+	/* Enable PPS with pulse train mode */
+	ppscr |= get_pps_cmd(index, XGBE_PPSCMD_START);
+	ppscr |= get_target_mode_sel(index, XGBE_PPSTARGET_PULSE);
+	ppscr |= PPSEN0;
+
+	XGMAC_IOWRITE(pdata, MAC_PPSCR, ppscr);
+
+	return 0;
+}
diff --git a/drivers/net/ethernet/amd/xgbe/xgbe-ptp.c b/drivers/net/ethernet/amd/xgbe/xgbe-ptp.c
index a533a6cc2d53..0e0b8ec3b504 100644
--- a/drivers/net/ethernet/amd/xgbe/xgbe-ptp.c
+++ b/drivers/net/ethernet/amd/xgbe/xgbe-ptp.c
@@ -1,117 +1,8 @@
+// SPDX-License-Identifier: (GPL-2.0-or-later OR BSD-3-Clause)
 /*
- * AMD 10Gb Ethernet driver
- *
- * This file is available to you under your choice of the following two
- * licenses:
- *
- * License 1: GPLv2
- *
- * Copyright (c) 2014 Advanced Micro Devices, Inc.
- *
- * This file is free software; you may copy, redistribute and/or modify
- * it under the terms of the GNU General Public License as published by
- * the Free Software Foundation, either version 2 of the License, or (at
- * your option) any later version.
- *
- * This file is distributed in the hope that it will be useful, but
- * WITHOUT ANY WARRANTY; without even the implied warranty of
- * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
- * General Public License for more details.
- *
- * You should have received a copy of the GNU General Public License
- * along with this program.  If not, see <http://www.gnu.org/licenses/>.
- *
- * This file incorporates work covered by the following copyright and
- * permission notice:
- *     The Synopsys DWC ETHER XGMAC Software Driver and documentation
- *     (hereinafter "Software") is an unsupported proprietary work of Synopsys,
- *     Inc. unless otherwise expressly agreed to in writing between Synopsys
- *     and you.
- *
- *     The Software IS NOT an item of Licensed Software or Licensed Product
- *     under any End User Software License Agreement or Agreement for Licensed
- *     Product with Synopsys or any supplement thereto.  Permission is hereby
- *     granted, free of charge, to any person obtaining a copy of this software
- *     annotated with this license and the Software, to deal in the Software
- *     without restriction, including without limitation the rights to use,
- *     copy, modify, merge, publish, distribute, sublicense, and/or sell copies
- *     of the Software, and to permit persons to whom the Software is furnished
- *     to do so, subject to the following conditions:
- *
- *     The above copyright notice and this permission notice shall be included
- *     in all copies or substantial portions of the Software.
- *
- *     THIS SOFTWARE IS BEING DISTRIBUTED BY SYNOPSYS SOLELY ON AN "AS IS"
- *     BASIS AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED
- *     TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A
- *     PARTICULAR PURPOSE ARE HEREBY DISCLAIMED. IN NO EVENT SHALL SYNOPSYS
- *     BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR
- *     CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF
- *     SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS
- *     INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN
- *     CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE)
- *     ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF
- *     THE POSSIBILITY OF SUCH DAMAGE.
- *
- *
- * License 2: Modified BSD
- *
- * Copyright (c) 2014 Advanced Micro Devices, Inc.
- * All rights reserved.
- *
- * Redistribution and use in source and binary forms, with or without
- * modification, are permitted provided that the following conditions are met:
- *     * Redistributions of source code must retain the above copyright
- *       notice, this list of conditions and the following disclaimer.
- *     * Redistributions in binary form must reproduce the above copyright
- *       notice, this list of conditions and the following disclaimer in the
- *       documentation and/or other materials provided with the distribution.
- *     * Neither the name of Advanced Micro Devices, Inc. nor the
- *       names of its contributors may be used to endorse or promote products
- *       derived from this software without specific prior written permission.
- *
- * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS"
- * AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
- * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
- * ARE DISCLAIMED. IN NO EVENT SHALL <COPYRIGHT HOLDER> BE LIABLE FOR ANY
- * DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES
- * (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES;
- * LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND
- * ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
- * (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF
- * THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
- *
- * This file incorporates work covered by the following copyright and
- * permission notice:
- *     The Synopsys DWC ETHER XGMAC Software Driver and documentation
- *     (hereinafter "Software") is an unsupported proprietary work of Synopsys,
- *     Inc. unless otherwise expressly agreed to in writing between Synopsys
- *     and you.
- *
- *     The Software IS NOT an item of Licensed Software or Licensed Product
- *     under any End User Software License Agreement or Agreement for Licensed
- *     Product with Synopsys or any supplement thereto.  Permission is hereby
- *     granted, free of charge, to any person obtaining a copy of this software
- *     annotated with this license and the Software, to deal in the Software
- *     without restriction, including without limitation the rights to use,
- *     copy, modify, merge, publish, distribute, sublicense, and/or sell copies
- *     of the Software, and to permit persons to whom the Software is furnished
- *     to do so, subject to the following conditions:
- *
- *     The above copyright notice and this permission notice shall be included
- *     in all copies or substantial portions of the Software.
- *
- *     THIS SOFTWARE IS BEING DISTRIBUTED BY SYNOPSYS SOLELY ON AN "AS IS"
- *     BASIS AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED
- *     TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A
- *     PARTICULAR PURPOSE ARE HEREBY DISCLAIMED. IN NO EVENT SHALL SYNOPSYS
- *     BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR
- *     CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF
- *     SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS
- *     INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN
- *     CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE)
- *     ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF
- *     THE POSSIBILITY OF SUCH DAMAGE.
+ * Copyright (c) 2014-2025, Advanced Micro Devices, Inc.
+ * Copyright (c) 2014, Synopsys, Inc.
+ * All rights reserved
  */
 
 #include <linux/clk.h>
@@ -122,43 +13,19 @@
 #include "xgbe.h"
 #include "xgbe-common.h"
 
-static u64 xgbe_cc_read(const struct cyclecounter *cc)
-{
-	struct xgbe_prv_data *pdata = container_of(cc,
-						   struct xgbe_prv_data,
-						   tstamp_cc);
-	u64 nsec;
-
-	nsec = pdata->hw_if.get_tstamp_time(pdata);
-
-	return nsec;
-}
-
-static int xgbe_adjfreq(struct ptp_clock_info *info, s32 delta)
+static int xgbe_adjfine(struct ptp_clock_info *info, long scaled_ppm)
 {
 	struct xgbe_prv_data *pdata = container_of(info,
 						   struct xgbe_prv_data,
 						   ptp_clock_info);
 	unsigned long flags;
-	u64 adjust;
-	u32 addend, diff;
-	unsigned int neg_adjust = 0;
+	u64 addend;
 
-	if (delta < 0) {
-		neg_adjust = 1;
-		delta = -delta;
-	}
-
-	adjust = pdata->tstamp_addend;
-	adjust *= delta;
-	diff = div_u64(adjust, 1000000000UL);
-
-	addend = (neg_adjust) ? pdata->tstamp_addend - diff :
-				pdata->tstamp_addend + diff;
+	addend = adjust_by_scaled_ppm(pdata->tstamp_addend, scaled_ppm);
 
 	spin_lock_irqsave(&pdata->tstamp_lock, flags);
 
-	pdata->hw_if.update_tstamp_addend(pdata, addend);
+	xgbe_update_tstamp_addend(pdata, addend);
 
 	spin_unlock_irqrestore(&pdata->tstamp_lock, flags);
 
@@ -170,16 +37,39 @@ static int xgbe_adjtime(struct ptp_clock_info *info, s64 delta)
 	struct xgbe_prv_data *pdata = container_of(info,
 						   struct xgbe_prv_data,
 						   ptp_clock_info);
+	unsigned int neg_adjust = 0;
+	unsigned int sec, nsec;
+	u32 quotient, reminder;
 	unsigned long flags;
 
+	if (delta < 0) {
+		neg_adjust = 1;
+		delta = -delta;
+	}
+
+	quotient = div_u64_rem(delta, 1000000000ULL, &reminder);
+	sec = quotient;
+	nsec = reminder;
+
+	/* Negative adjustment for Hw timer register. */
+	if (neg_adjust) {
+		sec = -sec;
+		if (XGMAC_IOREAD_BITS(pdata, MAC_TSCR, TSCTRLSSR))
+			nsec = (1000000000UL - nsec);
+		else
+			nsec = (0x80000000UL - nsec);
+	}
+	nsec = (neg_adjust << 31) | nsec;
+
 	spin_lock_irqsave(&pdata->tstamp_lock, flags);
-	timecounter_adjtime(&pdata->tstamp_tc, delta);
+	xgbe_update_tstamp_time(pdata, sec, nsec);
 	spin_unlock_irqrestore(&pdata->tstamp_lock, flags);
 
 	return 0;
 }
 
-static int xgbe_gettime(struct ptp_clock_info *info, struct timespec64 *ts)
+static int xgbe_gettimex(struct ptp_clock_info *info, struct timespec64 *ts,
+			 struct ptp_system_timestamp *sts)
 {
 	struct xgbe_prv_data *pdata = container_of(info,
 						   struct xgbe_prv_data,
@@ -188,9 +78,9 @@ static int xgbe_gettime(struct ptp_clock_info *info, struct timespec64 *ts)
 	u64 nsec;
 
 	spin_lock_irqsave(&pdata->tstamp_lock, flags);
-
-	nsec = timecounter_read(&pdata->tstamp_tc);
-
+	ptp_read_system_prets(sts);
+	nsec = xgbe_get_tstamp_time(pdata);
+	ptp_read_system_postts(sts);
 	spin_unlock_irqrestore(&pdata->tstamp_lock, flags);
 
 	*ts = ns_to_timespec64(nsec);
@@ -205,14 +95,9 @@ static int xgbe_settime(struct ptp_clock_info *info,
 						   struct xgbe_prv_data,
 						   ptp_clock_info);
 	unsigned long flags;
-	u64 nsec;
-
-	nsec = timespec64_to_ns(ts);
 
 	spin_lock_irqsave(&pdata->tstamp_lock, flags);
-
-	timecounter_init(&pdata->tstamp_tc, &pdata->tstamp_cc, nsec);
-
+	xgbe_set_tstamp_time(pdata, ts->tv_sec, ts->tv_nsec);
 	spin_unlock_irqrestore(&pdata->tstamp_lock, flags);
 
 	return 0;
@@ -221,24 +106,46 @@ static int xgbe_settime(struct ptp_clock_info *info,
 static int xgbe_enable(struct ptp_clock_info *info,
 		       struct ptp_clock_request *request, int on)
 {
-	return -EOPNOTSUPP;
+	struct xgbe_prv_data *pdata = container_of(info, struct xgbe_prv_data,
+						   ptp_clock_info);
+	struct xgbe_pps_config *pps_cfg;
+	unsigned long flags;
+	int ret;
+
+	dev_dbg(pdata->dev, "rq->type %d on %d\n", request->type, on);
+
+	if (request->type != PTP_CLK_REQ_PEROUT)
+		return -EOPNOTSUPP;
+
+	pps_cfg = &pdata->pps[request->perout.index];
+
+	pps_cfg->start.tv_sec = request->perout.start.sec;
+	pps_cfg->start.tv_nsec = request->perout.start.nsec;
+	pps_cfg->period.tv_sec = request->perout.period.sec;
+	pps_cfg->period.tv_nsec = request->perout.period.nsec;
+
+	spin_lock_irqsave(&pdata->tstamp_lock, flags);
+	ret = xgbe_pps_config(pdata, pps_cfg, request->perout.index, on);
+	spin_unlock_irqrestore(&pdata->tstamp_lock, flags);
+
+	return ret;
 }
 
 void xgbe_ptp_register(struct xgbe_prv_data *pdata)
 {
 	struct ptp_clock_info *info = &pdata->ptp_clock_info;
 	struct ptp_clock *clock;
-	struct cyclecounter *cc = &pdata->tstamp_cc;
-	u64 dividend;
 
 	snprintf(info->name, sizeof(info->name), "%s",
 		 netdev_name(pdata->netdev));
 	info->owner = THIS_MODULE;
 	info->max_adj = pdata->ptpclk_rate;
-	info->adjfreq = xgbe_adjfreq;
+	info->adjfine = xgbe_adjfine;
 	info->adjtime = xgbe_adjtime;
-	info->gettime64 = xgbe_gettime;
+	info->gettimex64 = xgbe_gettimex;
 	info->settime64 = xgbe_settime;
+	info->n_per_out = pdata->hw_feat.pps_out_num;
+	info->n_ext_ts = pdata->hw_feat.aux_snap_num;
 	info->enable = xgbe_enable;
 
 	clock = ptp_clock_register(info, pdata->dev);
@@ -249,25 +156,8 @@ void xgbe_ptp_register(struct xgbe_prv_data *pdata)
 
 	pdata->ptp_clock = clock;
 
-	/* Calculate the addend:
-	 *   addend = 2^32 / (PTP ref clock / 50Mhz)
-	 *          = (2^32 * 50Mhz) / PTP ref clock
-	 */
-	dividend = 50000000;
-	dividend <<= 32;
-	pdata->tstamp_addend = div_u64(dividend, pdata->ptpclk_rate);
-
-	/* Setup the timecounter */
-	cc->read = xgbe_cc_read;
-	cc->mask = CLOCKSOURCE_MASK(64);
-	cc->mult = 1;
-	cc->shift = 0;
-
-	timecounter_init(&pdata->tstamp_tc, &pdata->tstamp_cc,
-			 ktime_to_ns(ktime_get_real()));
-
 	/* Disable all timestamping to start */
-	XGMAC_IOWRITE(pdata, MAC_TCR, 0);
+	XGMAC_IOWRITE(pdata, MAC_TSCR, 0);
 	pdata->tstamp_config.tx_type = HWTSTAMP_TX_OFF;
 	pdata->tstamp_config.rx_filter = HWTSTAMP_FILTER_NONE;
 }
diff --git a/drivers/net/ethernet/amd/xgbe/xgbe-smn.h b/drivers/net/ethernet/amd/xgbe/xgbe-smn.h
new file mode 100644
index 000000000000..c6ae127ced03
--- /dev/null
+++ b/drivers/net/ethernet/amd/xgbe/xgbe-smn.h
@@ -0,0 +1,30 @@
+/* SPDX-License-Identifier: (GPL-2.0-or-later OR BSD-3-Clause) */
+/*
+ * Copyright (c) 2014-2025, Advanced Micro Devices, Inc.
+ * Copyright (c) 2014, Synopsys, Inc.
+ * All rights reserved
+ *
+ * Author: Raju Rangoju <Raju.Rangoju@amd.com>
+ */
+
+#ifndef __SMN_H__
+#define __SMN_H__
+
+#ifdef CONFIG_AMD_NB
+
+#include <asm/amd/nb.h>
+
+#else
+
+static inline int amd_smn_write(u16 node, u32 address, u32 value)
+{
+	return -ENODEV;
+}
+
+static inline int amd_smn_read(u16 node, u32 address, u32 *value)
+{
+	return -ENODEV;
+}
+
+#endif
+#endif
diff --git a/drivers/net/ethernet/amd/xgbe/xgbe.h b/drivers/net/ethernet/amd/xgbe/xgbe.h
index f9a24639f574..e8bbb6805901 100644
--- a/drivers/net/ethernet/amd/xgbe/xgbe.h
+++ b/drivers/net/ethernet/amd/xgbe/xgbe.h
@@ -1,117 +1,8 @@
+// SPDX-License-Identifier: (GPL-2.0-or-later OR BSD-3-Clause)
 /*
- * AMD 10Gb Ethernet driver
- *
- * This file is available to you under your choice of the following two
- * licenses:
- *
- * License 1: GPLv2
- *
- * Copyright (c) 2014-2016 Advanced Micro Devices, Inc.
- *
- * This file is free software; you may copy, redistribute and/or modify
- * it under the terms of the GNU General Public License as published by
- * the Free Software Foundation, either version 2 of the License, or (at
- * your option) any later version.
- *
- * This file is distributed in the hope that it will be useful, but
- * WITHOUT ANY WARRANTY; without even the implied warranty of
- * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
- * General Public License for more details.
- *
- * You should have received a copy of the GNU General Public License
- * along with this program.  If not, see <http://www.gnu.org/licenses/>.
- *
- * This file incorporates work covered by the following copyright and
- * permission notice:
- *     The Synopsys DWC ETHER XGMAC Software Driver and documentation
- *     (hereinafter "Software") is an unsupported proprietary work of Synopsys,
- *     Inc. unless otherwise expressly agreed to in writing between Synopsys
- *     and you.
- *
- *     The Software IS NOT an item of Licensed Software or Licensed Product
- *     under any End User Software License Agreement or Agreement for Licensed
- *     Product with Synopsys or any supplement thereto.  Permission is hereby
- *     granted, free of charge, to any person obtaining a copy of this software
- *     annotated with this license and the Software, to deal in the Software
- *     without restriction, including without limitation the rights to use,
- *     copy, modify, merge, publish, distribute, sublicense, and/or sell copies
- *     of the Software, and to permit persons to whom the Software is furnished
- *     to do so, subject to the following conditions:
- *
- *     The above copyright notice and this permission notice shall be included
- *     in all copies or substantial portions of the Software.
- *
- *     THIS SOFTWARE IS BEING DISTRIBUTED BY SYNOPSYS SOLELY ON AN "AS IS"
- *     BASIS AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED
- *     TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A
- *     PARTICULAR PURPOSE ARE HEREBY DISCLAIMED. IN NO EVENT SHALL SYNOPSYS
- *     BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR
- *     CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF
- *     SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS
- *     INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN
- *     CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE)
- *     ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF
- *     THE POSSIBILITY OF SUCH DAMAGE.
- *
- *
- * License 2: Modified BSD
- *
- * Copyright (c) 2014-2016 Advanced Micro Devices, Inc.
- * All rights reserved.
- *
- * Redistribution and use in source and binary forms, with or without
- * modification, are permitted provided that the following conditions are met:
- *     * Redistributions of source code must retain the above copyright
- *       notice, this list of conditions and the following disclaimer.
- *     * Redistributions in binary form must reproduce the above copyright
- *       notice, this list of conditions and the following disclaimer in the
- *       documentation and/or other materials provided with the distribution.
- *     * Neither the name of Advanced Micro Devices, Inc. nor the
- *       names of its contributors may be used to endorse or promote products
- *       derived from this software without specific prior written permission.
- *
- * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS"
- * AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
- * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
- * ARE DISCLAIMED. IN NO EVENT SHALL <COPYRIGHT HOLDER> BE LIABLE FOR ANY
- * DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES
- * (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES;
- * LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND
- * ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
- * (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF
- * THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
- *
- * This file incorporates work covered by the following copyright and
- * permission notice:
- *     The Synopsys DWC ETHER XGMAC Software Driver and documentation
- *     (hereinafter "Software") is an unsupported proprietary work of Synopsys,
- *     Inc. unless otherwise expressly agreed to in writing between Synopsys
- *     and you.
- *
- *     The Software IS NOT an item of Licensed Software or Licensed Product
- *     under any End User Software License Agreement or Agreement for Licensed
- *     Product with Synopsys or any supplement thereto.  Permission is hereby
- *     granted, free of charge, to any person obtaining a copy of this software
- *     annotated with this license and the Software, to deal in the Software
- *     without restriction, including without limitation the rights to use,
- *     copy, modify, merge, publish, distribute, sublicense, and/or sell copies
- *     of the Software, and to permit persons to whom the Software is furnished
- *     to do so, subject to the following conditions:
- *
- *     The above copyright notice and this permission notice shall be included
- *     in all copies or substantial portions of the Software.
- *
- *     THIS SOFTWARE IS BEING DISTRIBUTED BY SYNOPSYS SOLELY ON AN "AS IS"
- *     BASIS AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED
- *     TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A
- *     PARTICULAR PURPOSE ARE HEREBY DISCLAIMED. IN NO EVENT SHALL SYNOPSYS
- *     BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR
- *     CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF
- *     SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS
- *     INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN
- *     CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE)
- *     ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF
- *     THE POSSIBILITY OF SUCH DAMAGE.
+ * Copyright (c) 2014-2025, Advanced Micro Devices, Inc.
+ * Copyright (c) 2014, Synopsys, Inc.
+ * All rights reserved
  */
 
 #ifndef __XGBE_H__
@@ -128,9 +19,13 @@
 #include <linux/net_tstamp.h>
 #include <net/dcbnl.h>
 #include <linux/completion.h>
+#include <linux/cpumask.h>
+#include <linux/interrupt.h>
+#include <linux/dcache.h>
+#include <linux/ethtool.h>
+#include <linux/list.h>
 
 #define XGBE_DRV_NAME		"amd-xgbe"
-#define XGBE_DRV_VERSION	"1.0.3"
 #define XGBE_DRV_DESC		"AMD 10 Gigabit Ethernet Driver"
 
 /* Descriptor related defines */
@@ -139,10 +34,16 @@
 #define XGBE_TX_DESC_MAX_PROC	(XGBE_TX_DESC_CNT >> 1)
 #define XGBE_RX_DESC_CNT	512
 
+#define XGBE_TX_DESC_CNT_MIN	64
+#define XGBE_TX_DESC_CNT_MAX	4096
+#define XGBE_RX_DESC_CNT_MIN	64
+#define XGBE_RX_DESC_CNT_MAX	4096
+
 #define XGBE_TX_MAX_BUF_SIZE	(0x3fff & ~(64 - 1))
 
 /* Descriptors required for maximum contiguous TSO/GSO packet */
-#define XGBE_TX_MAX_SPLIT	((GSO_MAX_SIZE / XGBE_TX_MAX_BUF_SIZE) + 1)
+#define XGBE_TX_MAX_SPLIT	\
+	((GSO_LEGACY_MAX_SIZE / XGBE_TX_MAX_BUF_SIZE) + 1)
 
 /* Maximum possible descriptors needed for an SKB:
  * - Maximum number of SKB frags
@@ -163,26 +64,29 @@
 #define XGBE_DMA_STOP_TIMEOUT	1
 
 /* DMA cache settings - Outer sharable, write-back, write-allocate */
-#define XGBE_DMA_OS_AXDOMAIN	0x2
-#define XGBE_DMA_OS_ARCACHE	0xb
-#define XGBE_DMA_OS_AWCACHE	0xf
+#define XGBE_DMA_OS_ARCR	0x002b2b2b
+#define XGBE_DMA_OS_AWCR	0x2f2f2f2f
 
 /* DMA cache settings - System, no caches used */
-#define XGBE_DMA_SYS_AXDOMAIN	0x3
-#define XGBE_DMA_SYS_ARCACHE	0x0
-#define XGBE_DMA_SYS_AWCACHE	0x0
+#define XGBE_DMA_SYS_ARCR	0x00303030
+#define XGBE_DMA_SYS_AWCR	0x30303030
+
+/* DMA cache settings - PCI device */
+#define XGBE_DMA_PCI_ARCR	0x000f0f0f
+#define XGBE_DMA_PCI_AWCR	0x0f0f0f0f
+#define XGBE_DMA_PCI_AWARCR	0x00000f0f
 
 /* DMA channel interrupt modes */
 #define XGBE_IRQ_MODE_EDGE	0
 #define XGBE_IRQ_MODE_LEVEL	1
 
-#define XGBE_DMA_INTERRUPT_MASK	0x31c7
-
+#define XGBE_ETH_FRAME_HDR	(ETH_HLEN + ETH_FCS_LEN + VLAN_HLEN)
 #define XGMAC_MIN_PACKET	60
 #define XGMAC_STD_PACKET_MTU	1500
 #define XGMAC_MAX_STD_PACKET	1518
 #define XGMAC_JUMBO_PACKET_MTU	9000
 #define XGMAC_MAX_JUMBO_PACKET	9018
+#define XGMAC_GIANT_PACKET_MTU	16368
 #define XGMAC_ETH_PREAMBLE	(12 + 8)	/* Inter-frame gap + preamble */
 
 #define XGMAC_PFC_DATA_LEN	46
@@ -215,6 +119,14 @@
 #define XGBE_MSI_BASE_COUNT	4
 #define XGBE_MSI_MIN_COUNT	(XGBE_MSI_BASE_COUNT + 1)
 
+/* Initial PTP register values based on Link Speed. */
+#define MAC_TICNR_1G_INITVAL	0x10
+#define MAC_TECNR_1G_INITVAL	0x28
+
+#define MAC_TICSNR_10G_INITVAL	0x33
+#define MAC_TECNR_10G_INITVAL	0x14
+#define MAC_TECSNR_10G_INITVAL	0xCC
+
 /* PCI clock frequencies */
 #define XGBE_V2_DMA_CLOCK_FREQ	500000000	/* 500 MHz */
 #define XGBE_V2_PTP_CLOCK_FREQ	125000000	/* 125 MHz */
@@ -224,6 +136,15 @@
  */
 #define XGBE_TSTAMP_SSINC	20
 #define XGBE_TSTAMP_SNSINC	0
+#define XGBE_PTP_ACT_CLK_FREQ	500000000
+
+#define XGBE_V2_TSTAMP_SSINC	0xA
+#define XGBE_V2_TSTAMP_SNSINC	0
+#define XGBE_V2_PTP_ACT_CLK_FREQ	1000000000
+
+/* Define maximum supported values */
+#define XGBE_MAX_PPS_OUT	4
+#define XGBE_MAX_AUX_SNAP	4
 
 /* Driver PMT macros */
 #define XGMAC_DRIVER_CONTEXT	1
@@ -251,6 +172,7 @@
 /* Default coalescing parameters */
 #define XGMAC_INIT_DMA_TX_USECS		1000
 #define XGMAC_INIT_DMA_TX_FRAMES	25
+#define XGMAC_MAX_COAL_TX_TICK		100000
 
 #define XGMAC_MAX_DMA_RIWT		0xff
 #define XGMAC_INIT_DMA_RX_USECS		30
@@ -279,12 +201,14 @@
 /* Auto-negotiation */
 #define XGBE_AN_MS_TIMEOUT		500
 #define XGBE_LINK_TIMEOUT		5
+#define XGBE_KR_TRAINING_WAIT_ITER	50
 
-#define XGBE_SGMII_AN_LINK_STATUS	BIT(1)
+#define XGBE_SGMII_AN_LINK_DUPLEX	BIT(1)
 #define XGBE_SGMII_AN_LINK_SPEED	(BIT(2) | BIT(3))
+#define XGBE_SGMII_AN_LINK_SPEED_10	0x00
 #define XGBE_SGMII_AN_LINK_SPEED_100	0x04
 #define XGBE_SGMII_AN_LINK_SPEED_1000	0x08
-#define XGBE_SGMII_AN_LINK_DUPLEX	BIT(4)
+#define XGBE_SGMII_AN_LINK_STATUS	BIT(4)
 
 /* ECC correctable error notification window (seconds) */
 #define XGBE_ECC_LIMIT			60
@@ -292,6 +216,57 @@
 /* MDIO port types */
 #define XGMAC_MAX_C22_PORT		3
 
+/* Link mode bit operations */
+#define XGBE_ZERO_SUP(_ls)		\
+	ethtool_link_ksettings_zero_link_mode((_ls), supported)
+
+#define XGBE_SET_SUP(_ls, _mode)	\
+	ethtool_link_ksettings_add_link_mode((_ls), supported, _mode)
+
+#define XGBE_CLR_SUP(_ls, _mode)	\
+	ethtool_link_ksettings_del_link_mode((_ls), supported, _mode)
+
+#define XGBE_IS_SUP(_ls, _mode)	\
+	ethtool_link_ksettings_test_link_mode((_ls), supported, _mode)
+
+#define XGBE_ZERO_ADV(_ls)		\
+	ethtool_link_ksettings_zero_link_mode((_ls), advertising)
+
+#define XGBE_SET_ADV(_ls, _mode)	\
+	ethtool_link_ksettings_add_link_mode((_ls), advertising, _mode)
+
+#define XGBE_CLR_ADV(_ls, _mode)	\
+	ethtool_link_ksettings_del_link_mode((_ls), advertising, _mode)
+
+#define XGBE_ADV(_ls, _mode)		\
+	ethtool_link_ksettings_test_link_mode((_ls), advertising, _mode)
+
+#define XGBE_ZERO_LP_ADV(_ls)		\
+	ethtool_link_ksettings_zero_link_mode((_ls), lp_advertising)
+
+#define XGBE_SET_LP_ADV(_ls, _mode)	\
+	ethtool_link_ksettings_add_link_mode((_ls), lp_advertising, _mode)
+
+#define XGBE_CLR_LP_ADV(_ls, _mode)	\
+	ethtool_link_ksettings_del_link_mode((_ls), lp_advertising, _mode)
+
+#define XGBE_LP_ADV(_ls, _mode)		\
+	ethtool_link_ksettings_test_link_mode((_ls), lp_advertising, _mode)
+
+#define XGBE_LM_COPY(_dst, _dname, _src, _sname)	\
+	bitmap_copy((_dst)->link_modes._dname,		\
+		    (_src)->link_modes._sname,		\
+		    __ETHTOOL_LINK_MODE_MASK_NBITS)
+
+/* XGBE PCI device id */
+#define XGBE_RV_PCI_DEVICE_ID	0x15d0
+#define XGBE_YC_PCI_DEVICE_ID	0x14b5
+#define XGBE_RN_PCI_DEVICE_ID	0x1630
+
+ /* Generic low and high masks */
+#define XGBE_GEN_HI_MASK	GENMASK(31, 16)
+#define XGBE_GEN_LO_MASK	GENMASK(15, 0)
+
 struct xgbe_prv_data;
 
 struct xgbe_packet_data {
@@ -364,7 +339,7 @@ struct xgbe_rx_ring_data {
 
 /* Structure used to hold information related to the descriptor
  * and the packet associated with the descriptor (always use
- * use the XGBE_GET_DESC_DATA macro to access this data from the ring)
+ * the XGBE_GET_DESC_DATA macro to access this data from the ring)
  */
 struct xgbe_ring_data {
 	struct xgbe_ring_desc *rdesc;	/* Virtual address of descriptor */
@@ -412,6 +387,7 @@ struct xgbe_ring {
 	/* Page allocation for RX buffers */
 	struct xgbe_page_alloc rx_hdr_pa;
 	struct xgbe_page_alloc rx_buf_pa;
+	int node;
 
 	/* Ring index values
 	 *  cur   - Tx: index of descriptor to be used for current transfer
@@ -439,7 +415,7 @@ struct xgbe_ring {
  * a DMA channel.
  */
 struct xgbe_channel {
-	char name[16];
+	char name[20];
 
 	/* Address of private data area for device */
 	struct xgbe_prv_data *pdata;
@@ -455,6 +431,8 @@ struct xgbe_channel {
 	/* Netdev related settings */
 	struct napi_struct napi;
 
+	/* Per channel interrupt enablement tracker */
+	unsigned int curr_ier;
 	unsigned int saved_ier;
 
 	unsigned int tx_timer_active;
@@ -462,6 +440,9 @@ struct xgbe_channel {
 
 	struct xgbe_ring *tx_ring;
 	struct xgbe_ring *rx_ring;
+
+	int node;
+	cpumask_t affinity_mask;
 } ____cacheline_aligned;
 
 enum xgbe_state {
@@ -504,6 +485,7 @@ enum xgbe_speed {
 enum xgbe_xpcs_access {
 	XGBE_XPCS_ACCESS_V1 = 0,
 	XGBE_XPCS_ACCESS_V2,
+	XGBE_XPCS_ACCESS_V3,
 };
 
 enum xgbe_an_mode {
@@ -535,6 +517,7 @@ enum xgbe_mode {
 	XGBE_MODE_KX_2500,
 	XGBE_MODE_KR,
 	XGBE_MODE_X,
+	XGBE_MODE_SGMII_10,
 	XGBE_MODE_SGMII_100,
 	XGBE_MODE_SGMII_1000,
 	XGBE_MODE_SFI,
@@ -552,10 +535,34 @@ enum xgbe_mdio_mode {
 	XGBE_MDIO_MODE_CL45,
 };
 
+enum xgbe_mb_cmd {
+	XGBE_MB_CMD_POWER_OFF = 0,
+	XGBE_MB_CMD_SET_1G,
+	XGBE_MB_CMD_SET_2_5G,
+	XGBE_MB_CMD_SET_10G_SFI,
+	XGBE_MB_CMD_SET_10G_KR,
+	XGBE_MB_CMD_RRC
+};
+
+enum xgbe_mb_subcmd {
+	XGBE_MB_SUBCMD_NONE = 0,
+	XGBE_MB_SUBCMD_RX_ADAP,
+
+	/* 10GbE SFP subcommands */
+	XGBE_MB_SUBCMD_ACTIVE = 0,
+	XGBE_MB_SUBCMD_PASSIVE_1M,
+	XGBE_MB_SUBCMD_PASSIVE_3M,
+	XGBE_MB_SUBCMD_PASSIVE_OTHER,
+
+	/* 1GbE Mode subcommands */
+	XGBE_MB_SUBCMD_10MBITS = 0,
+	XGBE_MB_SUBCMD_100MBITS,
+	XGBE_MB_SUBCMD_1G_SGMII,
+	XGBE_MB_SUBCMD_1G_KX
+};
+
 struct xgbe_phy {
-	u32 supported;
-	u32 advertising;
-	u32 lp_advertising;
+	struct ethtool_link_ksettings lks;
 
 	int address;
 
@@ -658,12 +665,27 @@ struct xgbe_ext_stats {
 	u64 tx_tso_packets;
 	u64 rx_split_header_packets;
 	u64 rx_buffer_unavailable;
+
+	u64 txq_packets[XGBE_MAX_DMA_CHANNELS];
+	u64 txq_bytes[XGBE_MAX_DMA_CHANNELS];
+	u64 rxq_packets[XGBE_MAX_DMA_CHANNELS];
+	u64 rxq_bytes[XGBE_MAX_DMA_CHANNELS];
+
+	u64 tx_vxlan_packets;
+	u64 rx_vxlan_packets;
+	u64 rx_csum_errors;
+	u64 rx_vxlan_csum_errors;
+};
+
+struct xgbe_pps_config {
+	struct timespec64 start;
+	struct timespec64 period;
 };
 
 struct xgbe_hw_if {
 	int (*tx_complete)(struct xgbe_ring_desc *);
 
-	int (*set_mac_address)(struct xgbe_prv_data *, u8 *addr);
+	int (*set_mac_address)(struct xgbe_prv_data *, const u8 *addr);
 	int (*config_rx_mode)(struct xgbe_prv_data *);
 
 	int (*enable_rx_csum)(struct xgbe_prv_data *);
@@ -681,8 +703,11 @@ struct xgbe_hw_if {
 
 	int (*set_ext_mii_mode)(struct xgbe_prv_data *, unsigned int,
 				enum xgbe_mdio_mode);
-	int (*read_ext_mii_regs)(struct xgbe_prv_data *, int, int);
-	int (*write_ext_mii_regs)(struct xgbe_prv_data *, int, int, u16);
+	int (*read_ext_mii_regs_c22)(struct xgbe_prv_data *, int, int);
+	int (*write_ext_mii_regs_c22)(struct xgbe_prv_data *, int, int, u16);
+	int (*read_ext_mii_regs_c45)(struct xgbe_prv_data *, int, int, int);
+	int (*write_ext_mii_regs_c45)(struct xgbe_prv_data *, int, int, int,
+				      u16);
 
 	int (*set_gpio)(struct xgbe_prv_data *, unsigned int);
 	int (*clr_gpio)(struct xgbe_prv_data *, unsigned int);
@@ -734,26 +759,11 @@ struct xgbe_hw_if {
 	/* For TX DMA Operate on Second Frame config */
 	int (*config_osp_mode)(struct xgbe_prv_data *);
 
-	/* For RX and TX PBL config */
-	int (*config_rx_pbl_val)(struct xgbe_prv_data *);
-	int (*get_rx_pbl_val)(struct xgbe_prv_data *);
-	int (*config_tx_pbl_val)(struct xgbe_prv_data *);
-	int (*get_tx_pbl_val)(struct xgbe_prv_data *);
-	int (*config_pblx8)(struct xgbe_prv_data *);
-
 	/* For MMC statistics */
 	void (*rx_mmc_int)(struct xgbe_prv_data *);
 	void (*tx_mmc_int)(struct xgbe_prv_data *);
 	void (*read_mmc_stats)(struct xgbe_prv_data *);
 
-	/* For Timestamp config */
-	int (*config_tstamp)(struct xgbe_prv_data *, unsigned int);
-	void (*update_tstamp_addend)(struct xgbe_prv_data *, unsigned int);
-	void (*set_tstamp_time)(struct xgbe_prv_data *, unsigned int sec,
-				unsigned int nsec);
-	u64 (*get_tstamp_time)(struct xgbe_prv_data *);
-	u64 (*get_tx_tstamp)(struct xgbe_prv_data *);
-
 	/* For Data Center Bridging config */
 	void (*config_tc)(struct xgbe_prv_data *);
 	void (*config_dcb_tc)(struct xgbe_prv_data *);
@@ -768,12 +778,23 @@ struct xgbe_hw_if {
 	/* For ECC */
 	void (*disable_ecc_ded)(struct xgbe_prv_data *);
 	void (*disable_ecc_sec)(struct xgbe_prv_data *, enum xgbe_ecc_sec);
+
+	/* For VXLAN */
+	void (*enable_vxlan)(struct xgbe_prv_data *);
+	void (*disable_vxlan)(struct xgbe_prv_data *);
+	void (*set_vxlan_id)(struct xgbe_prv_data *);
+
+	/* For Split Header */
+	void (*enable_sph)(struct xgbe_prv_data *pdata);
+	void (*disable_sph)(struct xgbe_prv_data *pdata);
 };
 
 /* This structure represents implementation specific routines for an
  * implementation of a PHY. All routines are required unless noted below.
  *   Optional routines:
+ *     an_pre, an_post
  *     kr_training_pre, kr_training_post
+ *     module_info, module_eeprom
  */
 struct xgbe_phy_impl_if {
 	/* Perform Setup/teardown actions */
@@ -809,14 +830,25 @@ struct xgbe_phy_impl_if {
 	int (*an_config)(struct xgbe_prv_data *);
 
 	/* Set/override auto-negotiation advertisement settings */
-	unsigned int (*an_advertising)(struct xgbe_prv_data *);
+	void (*an_advertising)(struct xgbe_prv_data *,
+			       struct ethtool_link_ksettings *);
 
 	/* Process results of auto-negotiation */
 	enum xgbe_mode (*an_outcome)(struct xgbe_prv_data *);
 
+	/* Pre/Post auto-negotiation support */
+	void (*an_pre)(struct xgbe_prv_data *);
+	void (*an_post)(struct xgbe_prv_data *);
+
 	/* Pre/Post KR training enablement support */
 	void (*kr_training_pre)(struct xgbe_prv_data *);
 	void (*kr_training_post)(struct xgbe_prv_data *);
+
+	/* SFP module related info */
+	int (*module_info)(struct xgbe_prv_data *pdata,
+			   struct ethtool_modinfo *modinfo);
+	int (*module_eeprom)(struct xgbe_prv_data *pdata,
+			     struct ethtool_eeprom *eeprom, u8 *data);
 };
 
 struct xgbe_phy_if {
@@ -837,7 +869,13 @@ struct xgbe_phy_if {
 	bool (*phy_valid_speed)(struct xgbe_prv_data *, int);
 
 	/* For single interrupt support */
-	irqreturn_t (*an_isr)(int, struct xgbe_prv_data *);
+	irqreturn_t (*an_isr)(struct xgbe_prv_data *);
+
+	/* For ethtool PHY support */
+	int (*module_info)(struct xgbe_prv_data *pdata,
+			   struct ethtool_modinfo *modinfo);
+	int (*module_eeprom)(struct xgbe_prv_data *pdata,
+			     struct ethtool_eeprom *eeprom, u8 *data);
 
 	/* PHY implementation specific services */
 	struct xgbe_phy_impl_if phy_impl;
@@ -855,7 +893,7 @@ struct xgbe_i2c_if {
 	int (*i2c_xfer)(struct xgbe_prv_data *, struct xgbe_i2c_op *);
 
 	/* For single interrupt support */
-	irqreturn_t (*i2c_isr)(int, struct xgbe_prv_data *);
+	irqreturn_t (*i2c_isr)(struct xgbe_prv_data *);
 };
 
 struct xgbe_desc_if {
@@ -891,6 +929,7 @@ struct xgbe_hw_features {
 	unsigned int addn_mac;		/* Additional MAC Addresses */
 	unsigned int ts_src;		/* Timestamp Source */
 	unsigned int sa_vlan_ins;	/* Source Address or VLAN Insertion */
+	unsigned int vxn;		/* VXLAN/NVGRE */
 
 	/* HW Feature Register1 */
 	unsigned int rx_fifo_size;	/* MTL Receive FIFO Size */
@@ -922,8 +961,14 @@ struct xgbe_version_data {
 	unsigned int tx_max_fifo_size;
 	unsigned int rx_max_fifo_size;
 	unsigned int tx_tstamp_workaround;
+	unsigned int tstamp_ptp_clock_freq;
 	unsigned int ecc_support;
 	unsigned int i2c_support;
+	unsigned int irq_reissue_support;
+	unsigned int tx_desc_prefetch;
+	unsigned int rx_desc_prefetch;
+	unsigned int an_cdr_workaround;
+	unsigned int enable_rrc;
 };
 
 struct xgbe_prv_data {
@@ -934,6 +979,7 @@ struct xgbe_prv_data {
 	struct device *dev;
 	struct platform_device *phy_platdev;
 	struct device *phy_dev;
+	unsigned int smn_base;
 
 	/* Version related data */
 	struct xgbe_version_data *vdata;
@@ -950,6 +996,13 @@ struct xgbe_prv_data {
 	void __iomem *xprop_regs;	/* XGBE property registers */
 	void __iomem *xi2c_regs;	/* XGBE I2C CSRs */
 
+	/* Port property registers */
+	unsigned int pp0;
+	unsigned int pp1;
+	unsigned int pp2;
+	unsigned int pp3;
+	unsigned int pp4;
+
 	/* Overall device lock */
 	spinlock_t lock;
 
@@ -1001,9 +1054,9 @@ struct xgbe_prv_data {
 
 	/* AXI DMA settings */
 	unsigned int coherent;
-	unsigned int axdomain;
-	unsigned int arcache;
-	unsigned int awcache;
+	unsigned int arcr;
+	unsigned int awcr;
+	unsigned int awarcr;
 
 	/* Service routine support */
 	struct workqueue_struct *dev_workqueue;
@@ -1011,7 +1064,7 @@ struct xgbe_prv_data {
 	struct timer_list service_timer;
 
 	/* Rings for Tx/Rx on a DMA channel */
-	struct xgbe_channel *channel;
+	struct xgbe_channel *channel[XGBE_MAX_DMA_CHANNELS];
 	unsigned int tx_max_channel_count;
 	unsigned int rx_max_channel_count;
 	unsigned int channel_count;
@@ -1020,25 +1073,30 @@ struct xgbe_prv_data {
 	unsigned int rx_ring_count;
 	unsigned int rx_desc_count;
 
+	unsigned int new_tx_ring_count;
+	unsigned int new_rx_ring_count;
+
 	unsigned int tx_max_q_count;
 	unsigned int rx_max_q_count;
 	unsigned int tx_q_count;
 	unsigned int rx_q_count;
 
 	/* Tx/Rx common settings */
-	unsigned int pblx8;
+	unsigned int blen;
+	unsigned int pbl;
+	unsigned int aal;
+	unsigned int rd_osr_limit;
+	unsigned int wr_osr_limit;
 
 	/* Tx settings */
 	unsigned int tx_sf_mode;
 	unsigned int tx_threshold;
-	unsigned int tx_pbl;
 	unsigned int tx_osp_mode;
 	unsigned int tx_max_fifo_size;
 
 	/* Rx settings */
 	unsigned int rx_sf_mode;
 	unsigned int rx_threshold;
-	unsigned int rx_pbl;
 	unsigned int rx_max_fifo_size;
 
 	/* Tx coalescing settings */
@@ -1065,6 +1123,9 @@ struct xgbe_prv_data {
 	u32 rss_table[XGBE_RSS_MAX_TABLE_SIZE];
 	u32 rss_options;
 
+	/* VXLAN settings */
+	u16 vxlan_port;
+
 	/* Netdev related settings */
 	unsigned char mac_addr[ETH_ALEN];
 	netdev_features_t netdev_features;
@@ -1086,13 +1147,14 @@ struct xgbe_prv_data {
 	struct ptp_clock_info ptp_clock_info;
 	struct ptp_clock *ptp_clock;
 	struct hwtstamp_config tstamp_config;
-	struct cyclecounter tstamp_cc;
-	struct timecounter tstamp_tc;
 	unsigned int tstamp_addend;
 	struct work_struct tx_tstamp_work;
 	struct sk_buff *tx_tstamp_skb;
 	u64 tx_tstamp;
 
+	/* Pulse Per Second output */
+	struct xgbe_pps_config pps[XGBE_MAX_PPS_OUT];
+
 	/* DCB support */
 	struct ieee_ets *ets;
 	struct ieee_pfc *pfc;
@@ -1145,10 +1207,12 @@ struct xgbe_prv_data {
 	enum xgbe_rx kr_state;
 	enum xgbe_rx kx_state;
 	struct work_struct an_work;
+	unsigned int an_again;
 	unsigned int an_supported;
 	unsigned int parallel_detect;
 	unsigned int fec_ability;
 	unsigned long an_start;
+	unsigned long kr_start_time;
 	enum xgbe_an_mode an_mode;
 
 	/* I2C support */
@@ -1159,7 +1223,12 @@ struct xgbe_prv_data {
 
 	unsigned int lpm_ctrl;		/* CTRL1 for resume */
 
-#ifdef CONFIG_DEBUG_FS
+	unsigned int isr_as_bh_work;
+	struct work_struct dev_bh_work;
+	struct work_struct ecc_bh_work;
+	struct work_struct i2c_bh_work;
+	struct work_struct an_bh_work;
+
 	struct dentry *xgbe_debugfs;
 
 	unsigned int debugfs_xgmac_reg;
@@ -1170,7 +1239,13 @@ struct xgbe_prv_data {
 	unsigned int debugfs_xprop_reg;
 
 	unsigned int debugfs_xi2c_reg;
-#endif
+
+	bool debugfs_an_cdr_workaround;
+	bool debugfs_an_cdr_track_early;
+	bool en_rx_adap;
+	int rx_adapt_retries;
+	bool rx_adapt_done;
+	bool mode_set;
 };
 
 /* Function prototypes*/
@@ -1198,6 +1273,7 @@ void xgbe_init_function_ptrs_desc(struct xgbe_desc_if *);
 void xgbe_init_function_ptrs_i2c(struct xgbe_i2c_if *);
 const struct net_device_ops *xgbe_get_netdev_ops(void);
 const struct ethtool_ops *xgbe_get_ethtool_ops(void);
+const struct udp_tunnel_nic_info *xgbe_get_udp_tunnel_info(void);
 
 #ifdef CONFIG_AMD_XGBE_DCB
 const struct dcbnl_rtnl_ops *xgbe_get_dcbnl_ops(void);
@@ -1215,13 +1291,44 @@ int xgbe_powerup(struct net_device *, unsigned int);
 int xgbe_powerdown(struct net_device *, unsigned int);
 void xgbe_init_rx_coalesce(struct xgbe_prv_data *);
 void xgbe_init_tx_coalesce(struct xgbe_prv_data *);
+void xgbe_restart_dev(struct xgbe_prv_data *pdata);
+void xgbe_full_restart_dev(struct xgbe_prv_data *pdata);
+
+/* For Timestamp config */
+void xgbe_config_tstamp(struct xgbe_prv_data *pdata, unsigned int mac_tscr);
+u64 xgbe_get_tstamp_time(struct xgbe_prv_data *pdata);
+u64 xgbe_get_tx_tstamp(struct xgbe_prv_data *pdata);
+void xgbe_get_rx_tstamp(struct xgbe_packet_data *packet,
+			struct xgbe_ring_desc *rdesc);
+void xgbe_get_rx_tstamp(struct xgbe_packet_data *packet,
+			struct xgbe_ring_desc *rdesc);
+void xgbe_update_tstamp_addend(struct xgbe_prv_data *pdata,
+			       unsigned int addend);
+void xgbe_set_tstamp_time(struct xgbe_prv_data *pdata, unsigned int sec,
+			  unsigned int nsec);
+void xgbe_tx_tstamp(struct work_struct *work);
+int xgbe_get_hwtstamp_settings(struct xgbe_prv_data *pdata,
+			       struct ifreq *ifreq);
+int xgbe_set_hwtstamp_settings(struct xgbe_prv_data *pdata,
+			       struct ifreq *ifreq);
+void xgbe_prep_tx_tstamp(struct xgbe_prv_data *pdata,
+			 struct sk_buff *skb,
+			 struct xgbe_packet_data *packet);
+int xgbe_init_ptp(struct xgbe_prv_data *pdata);
+void xgbe_update_tstamp_time(struct xgbe_prv_data *pdata, unsigned int sec,
+			     unsigned int nsec);
+
+int xgbe_pps_config(struct xgbe_prv_data *pdata, struct xgbe_pps_config *cfg,
+		    int index, bool on);
 
 #ifdef CONFIG_DEBUG_FS
 void xgbe_debugfs_init(struct xgbe_prv_data *);
 void xgbe_debugfs_exit(struct xgbe_prv_data *);
+void xgbe_debugfs_rename(struct xgbe_prv_data *pdata);
 #else
 static inline void xgbe_debugfs_init(struct xgbe_prv_data *pdata) {}
 static inline void xgbe_debugfs_exit(struct xgbe_prv_data *pdata) {}
+static inline void xgbe_debugfs_rename(struct xgbe_prv_data *pdata) {}
 #endif /* CONFIG_DEBUG_FS */
 
 /* NOTE: Uncomment for function trace log messages in KERNEL LOG */
diff --git a/drivers/net/ethernet/apm/Kconfig b/drivers/net/ethernet/apm/Kconfig
index 59efe5b145dd..a893ef0e9c49 100644
--- a/drivers/net/ethernet/apm/Kconfig
+++ b/drivers/net/ethernet/apm/Kconfig
@@ -1,2 +1,3 @@
+# SPDX-License-Identifier: GPL-2.0-only
 source "drivers/net/ethernet/apm/xgene/Kconfig"
 source "drivers/net/ethernet/apm/xgene-v2/Kconfig"
diff --git a/drivers/net/ethernet/apm/Makefile b/drivers/net/ethernet/apm/Makefile
index 946b2a4c882d..cc8af97241fb 100644
--- a/drivers/net/ethernet/apm/Makefile
+++ b/drivers/net/ethernet/apm/Makefile
@@ -1,3 +1,4 @@
+# SPDX-License-Identifier: GPL-2.0-only
 #
 # Makefile for APM X-GENE Ethernet driver.
 #
diff --git a/drivers/net/ethernet/apm/xgene-v2/Kconfig b/drivers/net/ethernet/apm/xgene-v2/Kconfig
index 1205861b6318..2274af912fb3 100644
--- a/drivers/net/ethernet/apm/xgene-v2/Kconfig
+++ b/drivers/net/ethernet/apm/xgene-v2/Kconfig
@@ -1,6 +1,6 @@
+# SPDX-License-Identifier: GPL-2.0-only
 config NET_XGENE_V2
 	tristate "APM X-Gene SoC Ethernet-v2 Driver"
-	depends on HAS_DMA
 	depends on ARCH_XGENE || COMPILE_TEST
 	help
 	  This is the Ethernet driver for the on-chip ethernet interface
diff --git a/drivers/net/ethernet/apm/xgene-v2/Makefile b/drivers/net/ethernet/apm/xgene-v2/Makefile
index f16a2b3dde8b..fdde3b668acd 100644
--- a/drivers/net/ethernet/apm/xgene-v2/Makefile
+++ b/drivers/net/ethernet/apm/xgene-v2/Makefile
@@ -1,3 +1,4 @@
+# SPDX-License-Identifier: GPL-2.0-only
 #
 # Makefile for APM X-Gene Ethernet v2 driver
 #
diff --git a/drivers/net/ethernet/apm/xgene-v2/enet.c b/drivers/net/ethernet/apm/xgene-v2/enet.c
index 5998da014923..a8c6b379df82 100644
--- a/drivers/net/ethernet/apm/xgene-v2/enet.c
+++ b/drivers/net/ethernet/apm/xgene-v2/enet.c
@@ -1,22 +1,10 @@
+// SPDX-License-Identifier: GPL-2.0-or-later
 /*
  * Applied Micro X-Gene SoC Ethernet v2 Driver
  *
  * Copyright (c) 2017, Applied Micro Circuits Corporation
  * Author(s): Iyappan Subramanian <isubramanian@apm.com>
  *	      Keyur Chudgar <kchudgar@apm.com>
- *
- * This program is free software; you can redistribute  it and/or modify it
- * under  the terms of  the GNU General  Public License as published by the
- * Free Software Foundation;  either version 2 of the  License, or (at your
- * option) any later version.
- *
- * This program is distributed in the hope that it will be useful,
- * but WITHOUT ANY WARRANTY; without even the implied warranty of
- * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
- * GNU General Public License for more details.
- *
- * You should have received a copy of the GNU General Public License
- * along with this program.  If not, see <http://www.gnu.org/licenses/>.
  */
 
 #include "main.h"
diff --git a/drivers/net/ethernet/apm/xgene-v2/enet.h b/drivers/net/ethernet/apm/xgene-v2/enet.h
index 3fd36dc66a23..15cbd0ca1e9a 100644
--- a/drivers/net/ethernet/apm/xgene-v2/enet.h
+++ b/drivers/net/ethernet/apm/xgene-v2/enet.h
@@ -1,22 +1,10 @@
+/* SPDX-License-Identifier: GPL-2.0-or-later */
 /*
  * Applied Micro X-Gene SoC Ethernet v2 Driver
  *
  * Copyright (c) 2017, Applied Micro Circuits Corporation
  * Author(s): Iyappan Subramanian <isubramanian@apm.com>
  *	      Keyur Chudgar <kchudgar@apm.com>
- *
- * This program is free software; you can redistribute  it and/or modify it
- * under  the terms of  the GNU General  Public License as published by the
- * Free Software Foundation;  either version 2 of the  License, or (at your
- * option) any later version.
- *
- * This program is distributed in the hope that it will be useful,
- * but WITHOUT ANY WARRANTY; without even the implied warranty of
- * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
- * GNU General Public License for more details.
- *
- * You should have received a copy of the GNU General Public License
- * along with this program.  If not, see <http://www.gnu.org/licenses/>.
  */
 
 #ifndef __XGENE_ENET_V2_ENET_H__
diff --git a/drivers/net/ethernet/apm/xgene-v2/ethtool.c b/drivers/net/ethernet/apm/xgene-v2/ethtool.c
index b6666e418e79..b78d1a99fe81 100644
--- a/drivers/net/ethernet/apm/xgene-v2/ethtool.c
+++ b/drivers/net/ethernet/apm/xgene-v2/ethtool.c
@@ -1,22 +1,10 @@
+// SPDX-License-Identifier: GPL-2.0-or-later
 /*
  * Applied Micro X-Gene SoC Ethernet v2 Driver
  *
  * Copyright (c) 2017, Applied Micro Circuits Corporation
  * Author(s): Iyappan Subramanian <isubramanian@apm.com>
  *	      Keyur Chudgar <kchudgar@apm.com>
- *
- * This program is free software; you can redistribute  it and/or modify it
- * under  the terms of  the GNU General  Public License as published by the
- * Free Software Foundation;  either version 2 of the  License, or (at your
- * option) any later version.
- *
- * This program is distributed in the hope that it will be useful,
- * but WITHOUT ANY WARRANTY; without even the implied warranty of
- * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
- * GNU General Public License for more details.
- *
- * You should have received a copy of the GNU General Public License
- * along with this program.  If not, see <http://www.gnu.org/licenses/>.
  */
 
 #include "main.h"
@@ -101,8 +89,6 @@ static void xge_get_drvinfo(struct net_device *ndev,
 	struct platform_device *pdev = pdata->pdev;
 
 	strcpy(info->driver, "xgene-enet-v2");
-	strcpy(info->version, XGENE_ENET_V2_VERSION);
-	snprintf(info->fw_version, ETHTOOL_FWVERS_LEN, "N/A");
 	sprintf(info->bus_info, "%s", pdev->name);
 }
 
@@ -157,7 +143,9 @@ static int xge_get_link_ksettings(struct net_device *ndev,
 	if (!phydev)
 		return -ENODEV;
 
-	return phy_ethtool_ksettings_get(phydev, cmd);
+	phy_ethtool_ksettings_get(phydev, cmd);
+
+	return 0;
 }
 
 static int xge_set_link_ksettings(struct net_device *ndev,
diff --git a/drivers/net/ethernet/apm/xgene-v2/ethtool.h b/drivers/net/ethernet/apm/xgene-v2/ethtool.h
index 54b48d5561b8..8263b4aca6fe 100644
--- a/drivers/net/ethernet/apm/xgene-v2/ethtool.h
+++ b/drivers/net/ethernet/apm/xgene-v2/ethtool.h
@@ -1,21 +1,9 @@
+/* SPDX-License-Identifier: GPL-2.0-or-later */
 /*
  * Applied Micro X-Gene SoC Ethernet v2 Driver
  *
  * Copyright (c) 2017, Applied Micro Circuits Corporation
  * Author(s): Iyappan Subramanian <isubramanian@apm.com>
- *
- * This program is free software; you can redistribute  it and/or modify it
- * under  the terms of  the GNU General  Public License as published by the
- * Free Software Foundation;  either version 2 of the  License, or (at your
- * option) any later version.
- *
- * This program is distributed in the hope that it will be useful,
- * but WITHOUT ANY WARRANTY; without even the implied warranty of
- * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
- * GNU General Public License for more details.
- *
- * You should have received a copy of the GNU General Public License
- * along with this program.  If not, see <http://www.gnu.org/licenses/>.
  */
 
 #ifndef __XGENE_ENET_V2_ETHTOOL_H__
diff --git a/drivers/net/ethernet/apm/xgene-v2/mac.c b/drivers/net/ethernet/apm/xgene-v2/mac.c
index ee431e397e57..6423e22e05b2 100644
--- a/drivers/net/ethernet/apm/xgene-v2/mac.c
+++ b/drivers/net/ethernet/apm/xgene-v2/mac.c
@@ -1,22 +1,10 @@
+// SPDX-License-Identifier: GPL-2.0-or-later
 /*
  * Applied Micro X-Gene SoC Ethernet v2 Driver
  *
  * Copyright (c) 2017, Applied Micro Circuits Corporation
  * Author(s): Iyappan Subramanian <isubramanian@apm.com>
  *	      Keyur Chudgar <kchudgar@apm.com>
- *
- * This program is free software; you can redistribute  it and/or modify it
- * under  the terms of  the GNU General  Public License as published by the
- * Free Software Foundation;  either version 2 of the  License, or (at your
- * option) any later version.
- *
- * This program is distributed in the hope that it will be useful,
- * but WITHOUT ANY WARRANTY; without even the implied warranty of
- * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
- * GNU General Public License for more details.
- *
- * You should have received a copy of the GNU General Public License
- * along with this program.  If not, see <http://www.gnu.org/licenses/>.
  */
 
 #include "main.h"
@@ -77,7 +65,7 @@ void xge_mac_set_speed(struct xge_pdata *pdata)
 
 void xge_mac_set_station_addr(struct xge_pdata *pdata)
 {
-	u8 *dev_addr = pdata->ndev->dev_addr;
+	const u8 *dev_addr = pdata->ndev->dev_addr;
 	u32 addr0, addr1;
 
 	addr0 = (dev_addr[3] << 24) | (dev_addr[2] << 16) |
diff --git a/drivers/net/ethernet/apm/xgene-v2/mac.h b/drivers/net/ethernet/apm/xgene-v2/mac.h
index 3c83fa617356..7392f606687b 100644
--- a/drivers/net/ethernet/apm/xgene-v2/mac.h
+++ b/drivers/net/ethernet/apm/xgene-v2/mac.h
@@ -1,22 +1,10 @@
+/* SPDX-License-Identifier: GPL-2.0-or-later */
 /*
  * Applied Micro X-Gene SoC Ethernet v2 Driver
  *
  * Copyright (c) 2017, Applied Micro Circuits Corporation
  * Author(s): Iyappan Subramanian <isubramanian@apm.com>
  *	      Keyur Chudgar <kchudgar@apm.com>
- *
- * This program is free software; you can redistribute  it and/or modify it
- * under  the terms of  the GNU General  Public License as published by the
- * Free Software Foundation;  either version 2 of the  License, or (at your
- * option) any later version.
- *
- * This program is distributed in the hope that it will be useful,
- * but WITHOUT ANY WARRANTY; without even the implied warranty of
- * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
- * GNU General Public License for more details.
- *
- * You should have received a copy of the GNU General Public License
- * along with this program.  If not, see <http://www.gnu.org/licenses/>.
  */
 
 #ifndef __XGENE_ENET_V2_MAC_H__
diff --git a/drivers/net/ethernet/apm/xgene-v2/main.c b/drivers/net/ethernet/apm/xgene-v2/main.c
index 0f2ad50f3bd7..d7ca847d44c7 100644
--- a/drivers/net/ethernet/apm/xgene-v2/main.c
+++ b/drivers/net/ethernet/apm/xgene-v2/main.c
@@ -1,28 +1,14 @@
+// SPDX-License-Identifier: GPL-2.0-or-later
 /*
  * Applied Micro X-Gene SoC Ethernet v2 Driver
  *
  * Copyright (c) 2017, Applied Micro Circuits Corporation
  * Author(s): Iyappan Subramanian <isubramanian@apm.com>
  *	      Keyur Chudgar <kchudgar@apm.com>
- *
- * This program is free software; you can redistribute  it and/or modify it
- * under  the terms of  the GNU General  Public License as published by the
- * Free Software Foundation;  either version 2 of the  License, or (at your
- * option) any later version.
- *
- * This program is distributed in the hope that it will be useful,
- * but WITHOUT ANY WARRANTY; without even the implied warranty of
- * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
- * GNU General Public License for more details.
- *
- * You should have received a copy of the GNU General Public License
- * along with this program.  If not, see <http://www.gnu.org/licenses/>.
  */
 
 #include "main.h"
 
-static const struct acpi_device_id xge_acpi_match[];
-
 static int xge_get_resources(struct xge_pdata *pdata)
 {
 	struct platform_device *pdev;
@@ -48,7 +34,7 @@ static int xge_get_resources(struct xge_pdata *pdata)
 		return -ENOMEM;
 	}
 
-	if (!device_get_mac_address(dev, ndev->dev_addr, ETH_ALEN))
+	if (device_get_ethdev_address(dev, ndev))
 		eth_hw_addr_random(ndev);
 
 	memcpy(ndev->perm_addr, ndev->dev_addr, ndev->addr_len);
@@ -66,10 +52,8 @@ static int xge_get_resources(struct xge_pdata *pdata)
 	}
 
 	ret = platform_get_irq(pdev, 0);
-	if (ret < 0) {
-		dev_err(dev, "Unable to get irq\n");
+	if (ret < 0)
 		return ret;
-	}
 	pdata->resources.irq = ret;
 
 	return 0;
@@ -206,8 +190,8 @@ static netdev_tx_t xge_start_xmit(struct sk_buff *skb, struct net_device *ndev)
 	}
 
 	/* Packet buffers should be 64B aligned */
-	pkt_buf = dma_zalloc_coherent(dev, XGENE_ENET_STD_MTU, &dma_addr,
-				      GFP_ATOMIC);
+	pkt_buf = dma_alloc_coherent(dev, XGENE_ENET_STD_MTU, &dma_addr,
+				     GFP_ATOMIC);
 	if (unlikely(!pkt_buf)) {
 		dev_kfree_skb_any(skb);
 		return NETDEV_TX_OK;
@@ -428,8 +412,8 @@ static struct xge_desc_ring *xge_create_desc_ring(struct net_device *ndev)
 	ring->ndev = ndev;
 
 	size = XGENE_ENET_DESC_SIZE * XGENE_ENET_NUM_DESC;
-	ring->desc_addr = dma_zalloc_coherent(dev, size, &ring->dma_addr,
-					      GFP_KERNEL);
+	ring->desc_addr = dma_alloc_coherent(dev, size, &ring->dma_addr,
+					     GFP_KERNEL);
 	if (!ring->desc_addr)
 		goto err;
 
@@ -589,7 +573,7 @@ static void xge_free_pending_skb(struct net_device *ndev)
 	}
 }
 
-static void xge_timeout(struct net_device *ndev)
+static void xge_timeout(struct net_device *ndev, unsigned int txqueue)
 {
 	struct xge_pdata *pdata = netdev_priv(ndev);
 
@@ -686,23 +670,25 @@ static int xge_probe(struct platform_device *pdev)
 	if (ret)
 		goto err;
 
-	netif_napi_add(ndev, &pdata->napi, xge_napi, NAPI_POLL_WEIGHT);
+	netif_napi_add(ndev, &pdata->napi, xge_napi);
 
 	ret = register_netdev(ndev);
 	if (ret) {
 		netdev_err(ndev, "Failed to register netdev\n");
-		goto err;
+		goto err_mdio_remove;
 	}
 
 	return 0;
 
+err_mdio_remove:
+	xge_mdio_remove(ndev);
 err:
 	free_netdev(ndev);
 
 	return ret;
 }
 
-static int xge_remove(struct platform_device *pdev)
+static void xge_remove(struct platform_device *pdev)
 {
 	struct xge_pdata *pdata;
 	struct net_device *ndev;
@@ -718,8 +704,6 @@ static int xge_remove(struct platform_device *pdev)
 	xge_mdio_remove(ndev);
 	unregister_netdev(ndev);
 	free_netdev(ndev);
-
-	return 0;
 }
 
 static void xge_shutdown(struct platform_device *pdev)
@@ -745,7 +729,7 @@ MODULE_DEVICE_TABLE(acpi, xge_acpi_match);
 static struct platform_driver xge_driver = {
 	.driver = {
 		   .name = "xgene-enet-v2",
-		   .acpi_match_table = ACPI_PTR(xge_acpi_match),
+		   .acpi_match_table = xge_acpi_match,
 	},
 	.probe = xge_probe,
 	.remove = xge_remove,
@@ -755,5 +739,4 @@ module_platform_driver(xge_driver);
 
 MODULE_DESCRIPTION("APM X-Gene SoC Ethernet v2 driver");
 MODULE_AUTHOR("Iyappan Subramanian <isubramanian@apm.com>");
-MODULE_VERSION(XGENE_ENET_V2_VERSION);
 MODULE_LICENSE("GPL");
diff --git a/drivers/net/ethernet/apm/xgene-v2/main.h b/drivers/net/ethernet/apm/xgene-v2/main.h
index 969b258cb7de..7be6f83e22fe 100644
--- a/drivers/net/ethernet/apm/xgene-v2/main.h
+++ b/drivers/net/ethernet/apm/xgene-v2/main.h
@@ -1,22 +1,10 @@
+/* SPDX-License-Identifier: GPL-2.0-or-later */
 /*
  * Applied Micro X-Gene SoC Ethernet v2 Driver
  *
  * Copyright (c) 2017, Applied Micro Circuits Corporation
  * Author(s): Iyappan Subramanian <isubramanian@apm.com>
  *	      Keyur Chudgar <kchudgar@apm.com>
- *
- * This program is free software; you can redistribute  it and/or modify it
- * under  the terms of  the GNU General  Public License as published by the
- * Free Software Foundation;  either version 2 of the  License, or (at your
- * option) any later version.
- *
- * This program is distributed in the hope that it will be useful,
- * but WITHOUT ANY WARRANTY; without even the implied warranty of
- * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
- * GNU General Public License for more details.
- *
- * You should have received a copy of the GNU General Public License
- * along with this program.  If not, see <http://www.gnu.org/licenses/>.
  */
 
 #ifndef __XGENE_ENET_V2_MAIN_H__
@@ -34,13 +22,13 @@
 #include <linux/of_mdio.h>
 #include <linux/prefetch.h>
 #include <linux/phy.h>
+#include <linux/platform_device.h>
 #include <net/ip.h>
 #include "mac.h"
 #include "enet.h"
 #include "ring.h"
 #include "ethtool.h"
 
-#define XGENE_ENET_V2_VERSION	"v1.0"
 #define XGENE_ENET_STD_MTU	1536
 #define XGENE_ENET_MIN_FRAME	60
 #define IRQ_ID_SIZE             16
diff --git a/drivers/net/ethernet/apm/xgene-v2/mdio.c b/drivers/net/ethernet/apm/xgene-v2/mdio.c
index f5fe3bb2e59d..6a17045a5f62 100644
--- a/drivers/net/ethernet/apm/xgene-v2/mdio.c
+++ b/drivers/net/ethernet/apm/xgene-v2/mdio.c
@@ -1,22 +1,10 @@
+// SPDX-License-Identifier: GPL-2.0-or-later
 /*
  * Applied Micro X-Gene SoC Ethernet v2 Driver
  *
  * Copyright (c) 2017, Applied Micro Circuits Corporation
  * Author(s): Iyappan Subramanian <isubramanian@apm.com>
  *	      Keyur Chudgar <kchudgar@apm.com>
- *
- * This program is free software; you can redistribute  it and/or modify it
- * under  the terms of  the GNU General  Public License as published by the
- * Free Software Foundation;  either version 2 of the  License, or (at your
- * option) any later version.
- *
- * This program is distributed in the hope that it will be useful,
- * but WITHOUT ANY WARRANTY; without even the implied warranty of
- * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
- * GNU General Public License for more details.
- *
- * You should have received a copy of the GNU General Public License
- * along with this program.  If not, see <http://www.gnu.org/licenses/>.
  */
 
 #include "main.h"
@@ -148,16 +136,12 @@ int xge_mdio_config(struct net_device *ndev)
 		goto err;
 	}
 
-	phydev->supported &= ~(SUPPORTED_10baseT_Half |
-			       SUPPORTED_10baseT_Full |
-			       SUPPORTED_100baseT_Half |
-			       SUPPORTED_100baseT_Full |
-			       SUPPORTED_1000baseT_Half |
-			       SUPPORTED_AUI |
-			       SUPPORTED_MII |
-			       SUPPORTED_FIBRE |
-			       SUPPORTED_BNC);
-	phydev->advertising = phydev->supported;
+	phy_remove_link_mode(phydev, ETHTOOL_LINK_MODE_10baseT_Half_BIT);
+	phy_remove_link_mode(phydev, ETHTOOL_LINK_MODE_10baseT_Full_BIT);
+	phy_remove_link_mode(phydev, ETHTOOL_LINK_MODE_100baseT_Half_BIT);
+	phy_remove_link_mode(phydev, ETHTOOL_LINK_MODE_100baseT_Full_BIT);
+	phy_remove_link_mode(phydev, ETHTOOL_LINK_MODE_1000baseT_Half_BIT);
+
 	pdata->phy_speed = SPEED_UNKNOWN;
 
 	return 0;
diff --git a/drivers/net/ethernet/apm/xgene-v2/ring.c b/drivers/net/ethernet/apm/xgene-v2/ring.c
index 38810828f8f0..fbea4bc438a9 100644
--- a/drivers/net/ethernet/apm/xgene-v2/ring.c
+++ b/drivers/net/ethernet/apm/xgene-v2/ring.c
@@ -1,22 +1,10 @@
+// SPDX-License-Identifier: GPL-2.0-or-later
 /*
  * Applied Micro X-Gene SoC Ethernet v2 Driver
  *
  * Copyright (c) 2017, Applied Micro Circuits Corporation
  * Author(s): Iyappan Subramanian <isubramanian@apm.com>
  *	      Keyur Chudgar <kchudgar@apm.com>
- *
- * This program is free software; you can redistribute  it and/or modify it
- * under  the terms of  the GNU General  Public License as published by the
- * Free Software Foundation;  either version 2 of the  License, or (at your
- * option) any later version.
- *
- * This program is distributed in the hope that it will be useful,
- * but WITHOUT ANY WARRANTY; without even the implied warranty of
- * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
- * GNU General Public License for more details.
- *
- * You should have received a copy of the GNU General Public License
- * along with this program.  If not, see <http://www.gnu.org/licenses/>.
  */
 
 #include "main.h"
diff --git a/drivers/net/ethernet/apm/xgene-v2/ring.h b/drivers/net/ethernet/apm/xgene-v2/ring.h
index abc8c9a84954..2fd25553d5d3 100644
--- a/drivers/net/ethernet/apm/xgene-v2/ring.h
+++ b/drivers/net/ethernet/apm/xgene-v2/ring.h
@@ -1,22 +1,10 @@
+/* SPDX-License-Identifier: GPL-2.0-or-later */
 /*
  * Applied Micro X-Gene SoC Ethernet v2 Driver
  *
  * Copyright (c) 2017, Applied Micro Circuits Corporation
  * Author(s): Iyappan Subramanian <isubramanian@apm.com>
  *	      Keyur Chudgar <kchudgar@apm.com>
- *
- * This program is free software; you can redistribute  it and/or modify it
- * under  the terms of  the GNU General  Public License as published by the
- * Free Software Foundation;  either version 2 of the  License, or (at your
- * option) any later version.
- *
- * This program is distributed in the hope that it will be useful,
- * but WITHOUT ANY WARRANTY; without even the implied warranty of
- * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
- * GNU General Public License for more details.
- *
- * You should have received a copy of the GNU General Public License
- * along with this program.  If not, see <http://www.gnu.org/licenses/>.
  */
 
 #ifndef __XGENE_ENET_V2_RING_H__
diff --git a/drivers/net/ethernet/apm/xgene/Kconfig b/drivers/net/ethernet/apm/xgene/Kconfig
index afccb033177b..7bdfe78427df 100644
--- a/drivers/net/ethernet/apm/xgene/Kconfig
+++ b/drivers/net/ethernet/apm/xgene/Kconfig
@@ -1,6 +1,6 @@
+# SPDX-License-Identifier: GPL-2.0-only
 config NET_XGENE
 	tristate "APM X-Gene SoC Ethernet Driver"
-	depends on HAS_DMA
 	depends on ARCH_XGENE || COMPILE_TEST
 	select PHYLIB
 	select MDIO_XGENE
diff --git a/drivers/net/ethernet/apm/xgene/Makefile b/drivers/net/ethernet/apm/xgene/Makefile
index f46321f68315..6d1314757d3c 100644
--- a/drivers/net/ethernet/apm/xgene/Makefile
+++ b/drivers/net/ethernet/apm/xgene/Makefile
@@ -1,3 +1,4 @@
+# SPDX-License-Identifier: GPL-2.0-only
 #
 # Makefile for APM X-Gene Ethernet Driver.
 #
diff --git a/drivers/net/ethernet/apm/xgene/xgene_enet_cle.c b/drivers/net/ethernet/apm/xgene/xgene_enet_cle.c
index e1a51d8892fc..de5464322311 100644
--- a/drivers/net/ethernet/apm/xgene/xgene_enet_cle.c
+++ b/drivers/net/ethernet/apm/xgene/xgene_enet_cle.c
@@ -1,22 +1,10 @@
+// SPDX-License-Identifier: GPL-2.0-or-later
 /* Applied Micro X-Gene SoC Ethernet Classifier structures
  *
  * Copyright (c) 2016, Applied Micro Circuits Corporation
  * Authors: Khuong Dinh <kdinh@apm.com>
  *          Tanmay Inamdar <tinamdar@apm.com>
  *          Iyappan Subramanian <isubramanian@apm.com>
- *
- * This program is free software; you can redistribute  it and/or modify it
- * under  the terms of  the GNU General  Public License as published by the
- * Free Software Foundation;  either version 2 of the  License, or (at your
- * option) any later version.
- *
- * This program is distributed in the hope that it will be useful,
- * but WITHOUT ANY WARRANTY; without even the implied warranty of
- * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
- * GNU General Public License for more details.
- *
- * You should have received a copy of the GNU General Public License
- * along with this program.  If not, see <http://www.gnu.org/licenses/>.
  */
 
 #include "xgene_enet_main.h"
diff --git a/drivers/net/ethernet/apm/xgene/xgene_enet_cle.h b/drivers/net/ethernet/apm/xgene/xgene_enet_cle.h
index 18fe8d56082c..bc05cbcf4403 100644
--- a/drivers/net/ethernet/apm/xgene/xgene_enet_cle.h
+++ b/drivers/net/ethernet/apm/xgene/xgene_enet_cle.h
@@ -1,22 +1,10 @@
+/* SPDX-License-Identifier: GPL-2.0-or-later */
 /* Applied Micro X-Gene SoC Ethernet Classifier structures
  *
  * Copyright (c) 2016, Applied Micro Circuits Corporation
  * Authors: Khuong Dinh <kdinh@apm.com>
  *          Tanmay Inamdar <tinamdar@apm.com>
  *          Iyappan Subramanian <isubramanian@apm.com>
- *
- * This program is free software; you can redistribute  it and/or modify it
- * under  the terms of  the GNU General  Public License as published by the
- * Free Software Foundation;  either version 2 of the  License, or (at your
- * option) any later version.
- *
- * This program is distributed in the hope that it will be useful,
- * but WITHOUT ANY WARRANTY; without even the implied warranty of
- * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
- * GNU General Public License for more details.
- *
- * You should have received a copy of the GNU General Public License
- * along with this program.  If not, see <http://www.gnu.org/licenses/>.
  */
 
 #ifndef __XGENE_ENET_CLE_H__
diff --git a/drivers/net/ethernet/apm/xgene/xgene_enet_ethtool.c b/drivers/net/ethernet/apm/xgene/xgene_enet_ethtool.c
index 559963b1aa32..ada70425b48c 100644
--- a/drivers/net/ethernet/apm/xgene/xgene_enet_ethtool.c
+++ b/drivers/net/ethernet/apm/xgene/xgene_enet_ethtool.c
@@ -1,20 +1,8 @@
+// SPDX-License-Identifier: GPL-2.0-or-later
 /* Applied Micro X-Gene SoC Ethernet Driver
  *
  * Copyright (c) 2014, Applied Micro Circuits Corporation
  * Authors: Iyappan Subramanian <isubramanian@apm.com>
- *
- * This program is free software; you can redistribute  it and/or modify it
- * under  the terms of  the GNU General  Public License as published by the
- * Free Software Foundation;  either version 2 of the  License, or (at your
- * option) any later version.
- *
- * This program is distributed in the hope that it will be useful,
- * but WITHOUT ANY WARRANTY; without even the implied warranty of
- * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
- * GNU General Public License for more details.
- *
- * You should have received a copy of the GNU General Public License
- * along with this program.  If not, see <http://www.gnu.org/licenses/>.
  */
 
 #include <linux/ethtool.h>
@@ -115,8 +103,6 @@ static void xgene_get_drvinfo(struct net_device *ndev,
 	struct platform_device *pdev = pdata->pdev;
 
 	strcpy(info->driver, "xgene_enet");
-	strcpy(info->version, XGENE_DRV_VERSION);
-	snprintf(info->fw_version, ETHTOOL_FWVERS_LEN, "N/A");
 	sprintf(info->bus_info, "%s", pdev->name);
 }
 
@@ -131,13 +117,17 @@ static int xgene_get_link_ksettings(struct net_device *ndev,
 		if (phydev == NULL)
 			return -ENODEV;
 
-		return phy_ethtool_ksettings_get(phydev, cmd);
+		phy_ethtool_ksettings_get(phydev, cmd);
+
+		return 0;
 	} else if (pdata->phy_mode == PHY_INTERFACE_MODE_SGMII) {
 		if (pdata->mdio_driver) {
 			if (!phydev)
 				return -ENODEV;
 
-			return phy_ethtool_ksettings_get(phydev, cmd);
+			phy_ethtool_ksettings_get(phydev, cmd);
+
+			return 0;
 		}
 
 		supported = SUPPORTED_1000baseT_Full | SUPPORTED_Autoneg |
@@ -302,45 +292,25 @@ static int xgene_set_pauseparam(struct net_device *ndev,
 {
 	struct xgene_enet_pdata *pdata = netdev_priv(ndev);
 	struct phy_device *phydev = ndev->phydev;
-	u32 oldadv, newadv;
 
 	if (phy_interface_mode_is_rgmii(pdata->phy_mode) ||
 	    pdata->phy_mode == PHY_INTERFACE_MODE_SGMII) {
 		if (!phydev)
 			return -EINVAL;
 
-		if (!(phydev->supported & SUPPORTED_Pause) ||
-		    (!(phydev->supported & SUPPORTED_Asym_Pause) &&
-		     pp->rx_pause != pp->tx_pause))
+		if (!phy_validate_pause(phydev, pp))
 			return -EINVAL;
 
 		pdata->pause_autoneg = pp->autoneg;
 		pdata->tx_pause = pp->tx_pause;
 		pdata->rx_pause = pp->rx_pause;
 
-		oldadv = phydev->advertising;
-		newadv = oldadv & ~(ADVERTISED_Pause | ADVERTISED_Asym_Pause);
+		phy_set_asym_pause(phydev, pp->rx_pause,  pp->tx_pause);
 
-		if (pp->rx_pause)
-			newadv |= ADVERTISED_Pause | ADVERTISED_Asym_Pause;
-
-		if (pp->tx_pause)
-			newadv ^= ADVERTISED_Asym_Pause;
-
-		if (oldadv ^ newadv) {
-			phydev->advertising = newadv;
-
-			if (phydev->autoneg)
-				return phy_start_aneg(phydev);
-
-			if (!pp->autoneg) {
-				pdata->mac_ops->flowctl_tx(pdata,
-							   pdata->tx_pause);
-				pdata->mac_ops->flowctl_rx(pdata,
-							   pdata->rx_pause);
-			}
+		if (!pp->autoneg) {
+			pdata->mac_ops->flowctl_tx(pdata, pdata->tx_pause);
+			pdata->mac_ops->flowctl_rx(pdata, pdata->rx_pause);
 		}
-
 	} else {
 		if (pp->autoneg)
 			return -EINVAL;
diff --git a/drivers/net/ethernet/apm/xgene/xgene_enet_hw.c b/drivers/net/ethernet/apm/xgene/xgene_enet_hw.c
index e45b587c2994..b854b6b42d77 100644
--- a/drivers/net/ethernet/apm/xgene/xgene_enet_hw.c
+++ b/drivers/net/ethernet/apm/xgene/xgene_enet_hw.c
@@ -1,22 +1,10 @@
+// SPDX-License-Identifier: GPL-2.0-or-later
 /* Applied Micro X-Gene SoC Ethernet Driver
  *
  * Copyright (c) 2014, Applied Micro Circuits Corporation
  * Authors: Iyappan Subramanian <isubramanian@apm.com>
  *	    Ravi Patel <rapatel@apm.com>
  *	    Keyur Chudgar <kchudgar@apm.com>
- *
- * This program is free software; you can redistribute  it and/or modify it
- * under  the terms of  the GNU General  Public License as published by the
- * Free Software Foundation;  either version 2 of the  License, or (at your
- * option) any later version.
- *
- * This program is distributed in the hope that it will be useful,
- * but WITHOUT ANY WARRANTY; without even the implied warranty of
- * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
- * GNU General Public License for more details.
- *
- * You should have received a copy of the GNU General Public License
- * along with this program.  If not, see <http://www.gnu.org/licenses/>.
  */
 
 #include "xgene_enet_main.h"
@@ -390,8 +378,8 @@ u32 xgene_enet_rd_stat(struct xgene_enet_pdata *pdata, u32 rd_addr)
 
 static void xgene_gmac_set_mac_addr(struct xgene_enet_pdata *pdata)
 {
+	const u8 *dev_addr = pdata->ndev->dev_addr;
 	u32 addr0, addr1;
-	u8 *dev_addr = pdata->ndev->dev_addr;
 
 	addr0 = (dev_addr[3] << 24) | (dev_addr[2] << 16) |
 		(dev_addr[1] << 8) | dev_addr[0];
@@ -433,18 +421,12 @@ static void xgene_enet_configure_clock(struct xgene_enet_pdata *pdata)
 
 	if (dev->of_node) {
 		struct clk *parent = clk_get_parent(pdata->clk);
+		long rate = rgmii_clock(pdata->phy_speed);
 
-		switch (pdata->phy_speed) {
-		case SPEED_10:
-			clk_set_rate(parent, 2500000);
-			break;
-		case SPEED_100:
-			clk_set_rate(parent, 25000000);
-			break;
-		default:
-			clk_set_rate(parent, 125000000);
-			break;
-		}
+		if (rate < 0)
+			rate = 125000000;
+
+		clk_set_rate(parent, rate);
 	}
 #ifdef CONFIG_ACPI
 	else {
@@ -468,7 +450,6 @@ static void xgene_enet_configure_clock(struct xgene_enet_pdata *pdata)
 
 static void xgene_gmac_set_speed(struct xgene_enet_pdata *pdata)
 {
-	struct device *dev = &pdata->pdev->dev;
 	u32 icm0, icm2, mc2;
 	u32 intf_ctl, rgmii, value;
 
@@ -500,10 +481,8 @@ static void xgene_gmac_set_speed(struct xgene_enet_pdata *pdata)
 		intf_ctl |= ENET_GHD_MODE;
 		CFG_MACMODE_SET(&icm0, 2);
 		CFG_WAITASYNCRD_SET(&icm2, 0);
-		if (dev->of_node) {
-			CFG_TXCLK_MUXSEL0_SET(&rgmii, pdata->tx_delay);
-			CFG_RXCLK_MUXSEL0_SET(&rgmii, pdata->rx_delay);
-		}
+		CFG_TXCLK_MUXSEL0_SET(&rgmii, pdata->tx_delay);
+		CFG_RXCLK_MUXSEL0_SET(&rgmii, pdata->rx_delay);
 		rgmii |= CFG_SPEED_1250;
 
 		xgene_enet_rd_csr(pdata, DEBUG_REG_ADDR, &value);
@@ -727,11 +706,11 @@ static int xgene_enet_reset(struct xgene_enet_pdata *pdata)
 		udelay(5);
 	} else {
 #ifdef CONFIG_ACPI
-		if (acpi_has_method(ACPI_HANDLE(&pdata->pdev->dev), "_RST")) {
-			acpi_evaluate_object(ACPI_HANDLE(&pdata->pdev->dev),
-					     "_RST", NULL, NULL);
-		} else if (acpi_has_method(ACPI_HANDLE(&pdata->pdev->dev),
-					 "_INI")) {
+		acpi_status status;
+
+		status = acpi_evaluate_object(ACPI_HANDLE(&pdata->pdev->dev),
+					      "_RST", NULL, NULL);
+		if (ACPI_FAILURE(status)) {
 			acpi_evaluate_object(ACPI_HANDLE(&pdata->pdev->dev),
 					     "_INI", NULL, NULL);
 		}
@@ -839,19 +818,19 @@ static void xgene_enet_adjust_link(struct net_device *ndev)
 #ifdef CONFIG_ACPI
 static struct acpi_device *acpi_phy_find_device(struct device *dev)
 {
-	struct acpi_reference_args args;
+	struct fwnode_reference_args args;
 	struct fwnode_handle *fw_node;
 	int status;
 
 	fw_node = acpi_fwnode_handle(ACPI_COMPANION(dev));
 	status = acpi_node_get_property_reference(fw_node, "phy-handle", 0,
 						  &args);
-	if (ACPI_FAILURE(status)) {
+	if (ACPI_FAILURE(status) || !is_acpi_device_node(args.fwnode)) {
 		dev_dbg(dev, "No matching phy in ACPI table\n");
 		return NULL;
 	}
 
-	return args.adev;
+	return to_acpi_device_node(args.fwnode);
 }
 #endif
 
@@ -898,12 +877,10 @@ int xgene_enet_phy_connect(struct net_device *ndev)
 	}
 
 	pdata->phy_speed = SPEED_UNKNOWN;
-	phy_dev->supported &= ~SUPPORTED_10baseT_Half &
-			      ~SUPPORTED_100baseT_Half &
-			      ~SUPPORTED_1000baseT_Half;
-	phy_dev->supported |= SUPPORTED_Pause |
-			      SUPPORTED_Asym_Pause;
-	phy_dev->advertising = phy_dev->supported;
+	phy_remove_link_mode(phy_dev, ETHTOOL_LINK_MODE_10baseT_Half_BIT);
+	phy_remove_link_mode(phy_dev, ETHTOOL_LINK_MODE_100baseT_Half_BIT);
+	phy_remove_link_mode(phy_dev, ETHTOOL_LINK_MODE_1000baseT_Half_BIT);
+	phy_support_asym_pause(phy_dev);
 
 	return 0;
 }
diff --git a/drivers/net/ethernet/apm/xgene/xgene_enet_hw.h b/drivers/net/ethernet/apm/xgene/xgene_enet_hw.h
index 5d3e18d3c94c..2f534f9d4416 100644
--- a/drivers/net/ethernet/apm/xgene/xgene_enet_hw.h
+++ b/drivers/net/ethernet/apm/xgene/xgene_enet_hw.h
@@ -1,22 +1,10 @@
+/* SPDX-License-Identifier: GPL-2.0-or-later */
 /* Applied Micro X-Gene SoC Ethernet Driver
  *
  * Copyright (c) 2014, Applied Micro Circuits Corporation
  * Authors: Iyappan Subramanian <isubramanian@apm.com>
  *	    Ravi Patel <rapatel@apm.com>
  *	    Keyur Chudgar <kchudgar@apm.com>
- *
- * This program is free software; you can redistribute  it and/or modify it
- * under  the terms of  the GNU General  Public License as published by the
- * Free Software Foundation;  either version 2 of the  License, or (at your
- * option) any later version.
- *
- * This program is distributed in the hope that it will be useful,
- * but WITHOUT ANY WARRANTY; without even the implied warranty of
- * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
- * GNU General Public License for more details.
- *
- * You should have received a copy of the GNU General Public License
- * along with this program.  If not, see <http://www.gnu.org/licenses/>.
  */
 
 #ifndef __XGENE_ENET_HW_H__
diff --git a/drivers/net/ethernet/apm/xgene/xgene_enet_main.c b/drivers/net/ethernet/apm/xgene/xgene_enet_main.c
index d3906f6b01bd..3b2951030a38 100644
--- a/drivers/net/ethernet/apm/xgene/xgene_enet_main.c
+++ b/drivers/net/ethernet/apm/xgene/xgene_enet_main.c
@@ -1,22 +1,10 @@
+// SPDX-License-Identifier: GPL-2.0-or-later
 /* Applied Micro X-Gene SoC Ethernet Driver
  *
  * Copyright (c) 2014, Applied Micro Circuits Corporation
  * Authors: Iyappan Subramanian <isubramanian@apm.com>
  *	    Ravi Patel <rapatel@apm.com>
  *	    Keyur Chudgar <kchudgar@apm.com>
- *
- * This program is free software; you can redistribute  it and/or modify it
- * under  the terms of  the GNU General  Public License as published by the
- * Free Software Foundation;  either version 2 of the  License, or (at your
- * option) any later version.
- *
- * This program is distributed in the hope that it will be useful,
- * but WITHOUT ANY WARRANTY; without even the implied warranty of
- * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
- * GNU General Public License for more details.
- *
- * You should have received a copy of the GNU General Public License
- * along with this program.  If not, see <http://www.gnu.org/licenses/>.
  */
 
 #include <linux/gpio.h>
@@ -29,9 +17,6 @@
 #define RES_RING_CSR	1
 #define RES_RING_CMD	2
 
-static const struct of_device_id xgene_enet_of_match[];
-static const struct acpi_device_id xgene_enet_acpi_match[];
-
 static void xgene_enet_init_bufpool(struct xgene_enet_desc_ring *buf_pool)
 {
 	struct xgene_enet_raw_desc16 *raw_desc;
@@ -355,7 +340,8 @@ static int xgene_enet_work_msg(struct sk_buff *skb, u64 *hopinfo)
 				nr_frags = skb_shinfo(skb)->nr_frags;
 
 				for (i = 0; i < 2 && i < nr_frags; i++)
-					len += skb_shinfo(skb)->frags[i].size;
+					len += skb_frag_size(
+						&skb_shinfo(skb)->frags[i]);
 
 				/* HW requires header must reside in 3 buffer */
 				if (unlikely(hdr_len > len)) {
@@ -710,6 +696,12 @@ static int xgene_enet_rx_frame(struct xgene_enet_desc_ring *rx_ring,
 	buf_pool->rx_skb[skb_index] = NULL;
 
 	datalen = xgene_enet_get_data_len(le64_to_cpu(raw_desc->m1));
+
+	/* strip off CRC as HW isn't doing this */
+	nv = GET_VAL(NV, le64_to_cpu(raw_desc->m0));
+	if (!nv)
+		datalen -= 4;
+
 	skb_put(skb, datalen);
 	prefetch(skb->data - NET_IP_ALIGN);
 	skb->protocol = eth_type_trans(skb, ndev);
@@ -731,12 +723,8 @@ static int xgene_enet_rx_frame(struct xgene_enet_desc_ring *rx_ring,
 		}
 	}
 
-	nv = GET_VAL(NV, le64_to_cpu(raw_desc->m0));
-	if (!nv) {
-		/* strip off CRC as HW isn't doing this */
-		datalen -= 4;
+	if (!nv)
 		goto skip_jumbo;
-	}
 
 	slots = page_pool->slots - 1;
 	head = page_pool->head;
@@ -873,7 +861,7 @@ static int xgene_enet_napi(struct napi_struct *napi, const int budget)
 	return processed;
 }
 
-static void xgene_enet_timeout(struct net_device *ndev)
+static void xgene_enet_timeout(struct net_device *ndev, unsigned int txqueue)
 {
 	struct xgene_enet_pdata *pdata = netdev_priv(ndev);
 	struct netdev_queue *txq;
@@ -883,7 +871,7 @@ static void xgene_enet_timeout(struct net_device *ndev)
 
 	for (i = 0; i < pdata->txq_cnt; i++) {
 		txq = netdev_get_tx_queue(ndev, i);
-		txq->trans_start = jiffies;
+		txq_trans_cond_update(txq);
 		netif_tx_start_queue(txq);
 	}
 }
@@ -1016,8 +1004,10 @@ static int xgene_enet_open(struct net_device *ndev)
 
 	xgene_enet_napi_enable(pdata);
 	ret = xgene_enet_register_irq(ndev);
-	if (ret)
+	if (ret) {
+		xgene_enet_napi_disable(pdata);
 		return ret;
+	}
 
 	if (ndev->phydev) {
 		phy_start(ndev->phydev);
@@ -1540,7 +1530,7 @@ static int xgene_change_mtu(struct net_device *ndev, int new_mtu)
 	frame_size = (new_mtu > ETH_DATA_LEN) ? (new_mtu + 18) : 0x600;
 
 	xgene_enet_close(ndev);
-	ndev->mtu = new_mtu;
+	WRITE_ONCE(ndev->mtu, new_mtu);
 	pdata->mac_ops->set_framesize(pdata, frame_size);
 	xgene_enet_open(ndev);
 
@@ -1591,7 +1581,7 @@ static int xgene_get_tx_delay(struct xgene_enet_pdata *pdata)
 	struct device *dev = &pdata->pdev->dev;
 	int delay, ret;
 
-	ret = of_property_read_u32(dev->of_node, "tx-delay", &delay);
+	ret = device_property_read_u32(dev, "tx-delay", &delay);
 	if (ret) {
 		pdata->tx_delay = 4;
 		return 0;
@@ -1612,7 +1602,7 @@ static int xgene_get_rx_delay(struct xgene_enet_pdata *pdata)
 	struct device *dev = &pdata->pdev->dev;
 	int delay, ret;
 
-	ret = of_property_read_u32(dev->of_node, "rx-delay", &delay);
+	ret = device_property_read_u32(dev, "rx-delay", &delay);
 	if (ret) {
 		pdata->rx_delay = 2;
 		return 0;
@@ -1631,7 +1621,6 @@ static int xgene_get_rx_delay(struct xgene_enet_pdata *pdata)
 static int xgene_enet_get_irqs(struct xgene_enet_pdata *pdata)
 {
 	struct platform_device *pdev = pdata->pdev;
-	struct device *dev = &pdev->dev;
 	int i, ret, max_irqs;
 
 	if (phy_interface_mode_is_rgmii(pdata->phy_mode))
@@ -1643,7 +1632,7 @@ static int xgene_enet_get_irqs(struct xgene_enet_pdata *pdata)
 
 	for (i = 0; i < max_irqs; i++) {
 		ret = platform_get_irq(pdev, i);
-		if (ret <= 0) {
+		if (ret < 0) {
 			if (pdata->phy_mode == PHY_INTERFACE_MODE_XGMII) {
 				max_irqs = i;
 				pdata->rxq_cnt = max_irqs / 2;
@@ -1651,8 +1640,6 @@ static int xgene_enet_get_irqs(struct xgene_enet_pdata *pdata)
 				pdata->cq_cnt = max_irqs / 2;
 				break;
 			}
-			dev_err(dev, "Unable to get ENET IRQ\n");
-			ret = ret ? : -ENXIO;
 			return ret;
 		}
 		pdata->irqs[i] = ret;
@@ -1661,21 +1648,19 @@ static int xgene_enet_get_irqs(struct xgene_enet_pdata *pdata)
 	return 0;
 }
 
-static int xgene_enet_check_phy_handle(struct xgene_enet_pdata *pdata)
+static void xgene_enet_check_phy_handle(struct xgene_enet_pdata *pdata)
 {
 	int ret;
 
 	if (pdata->phy_mode == PHY_INTERFACE_MODE_XGMII)
-		return 0;
+		return;
 
 	if (!IS_ENABLED(CONFIG_MDIO_XGENE))
-		return 0;
+		return;
 
 	ret = xgene_enet_phy_connect(pdata->ndev);
 	if (!ret)
 		pdata->mdio_driver = true;
-
-	return 0;
 }
 
 static void xgene_enet_gpiod_get(struct xgene_enet_pdata *pdata)
@@ -1750,7 +1735,7 @@ static int xgene_enet_get_resources(struct xgene_enet_pdata *pdata)
 		xgene_get_port_id_acpi(dev, pdata);
 #endif
 
-	if (!device_get_mac_address(dev, ndev->dev_addr, ETH_ALEN))
+	if (device_get_ethdev_address(dev, ndev))
 		eth_hw_addr_random(ndev);
 
 	memcpy(ndev->perm_addr, ndev->dev_addr, ndev->addr_len);
@@ -1779,22 +1764,20 @@ static int xgene_enet_get_resources(struct xgene_enet_pdata *pdata)
 	if (ret)
 		return ret;
 
-	ret = xgene_enet_check_phy_handle(pdata);
-	if (ret)
-		return ret;
-
 	xgene_enet_gpiod_get(pdata);
 
 	pdata->clk = devm_clk_get(&pdev->dev, NULL);
 	if (IS_ERR(pdata->clk)) {
-		/* Abort if the clock is defined but couldn't be retrived.
-		 * Always abort if the clock is missing on DT system as
-		 * the driver can't cope with this case.
-		 */
-		if (PTR_ERR(pdata->clk) != -ENOENT || dev->of_node)
-			return PTR_ERR(pdata->clk);
-		/* Firmware may have set up the clock already. */
-		dev_info(dev, "clocks have been setup already\n");
+		if (pdata->phy_mode != PHY_INTERFACE_MODE_SGMII) {
+			/* Abort if the clock is defined but couldn't be
+			 * retrived. Always abort if the clock is missing on
+			 * DT system as the driver can't cope with this case.
+			 */
+			if (PTR_ERR(pdata->clk) != -ENOENT || dev->of_node)
+				return PTR_ERR(pdata->clk);
+			/* Firmware may have set up the clock already. */
+			dev_info(dev, "clocks have been setup already\n");
+		}
 	}
 
 	if (pdata->phy_mode != PHY_INTERFACE_MODE_XGMII)
@@ -1996,14 +1979,12 @@ static void xgene_enet_napi_add(struct xgene_enet_pdata *pdata)
 
 	for (i = 0; i < pdata->rxq_cnt; i++) {
 		napi = &pdata->rx_ring[i]->napi;
-		netif_napi_add(pdata->ndev, napi, xgene_enet_napi,
-			       NAPI_POLL_WEIGHT);
+		netif_napi_add(pdata->ndev, napi, xgene_enet_napi);
 	}
 
 	for (i = 0; i < pdata->cq_cnt; i++) {
 		napi = &pdata->tx_ring[i]->cp_ring->napi;
-		netif_napi_add(pdata->ndev, napi, xgene_enet_napi,
-			       NAPI_POLL_WEIGHT);
+		netif_napi_add(pdata->ndev, napi, xgene_enet_napi);
 	}
 }
 
@@ -2037,11 +2018,10 @@ static int xgene_enet_probe(struct platform_device *pdev)
 	struct xgene_enet_pdata *pdata;
 	struct device *dev = &pdev->dev;
 	void (*link_state)(struct work_struct *);
-	const struct of_device_id *of_id;
 	int ret;
 
 	ndev = alloc_etherdev_mqs(sizeof(struct xgene_enet_pdata),
-				  XGENE_NUM_RX_RING, XGENE_NUM_TX_RING);
+				  XGENE_NUM_TX_RING, XGENE_NUM_RX_RING);
 	if (!ndev)
 		return -ENOMEM;
 
@@ -2058,19 +2038,7 @@ static int xgene_enet_probe(struct platform_device *pdev)
 			  NETIF_F_GRO |
 			  NETIF_F_SG;
 
-	of_id = of_match_device(xgene_enet_of_match, &pdev->dev);
-	if (of_id) {
-		pdata->enet_id = (enum xgene_enet_id)of_id->data;
-	}
-#ifdef CONFIG_ACPI
-	else {
-		const struct acpi_device_id *acpi_id;
-
-		acpi_id = acpi_match_device(xgene_enet_acpi_match, &pdev->dev);
-		if (acpi_id)
-			pdata->enet_id = (enum xgene_enet_id) acpi_id->driver_data;
-	}
-#endif
+	pdata->enet_id = (enum xgene_enet_id)device_get_match_data(&pdev->dev);
 	if (!pdata->enet_id) {
 		ret = -ENODEV;
 		goto err;
@@ -2095,9 +2063,11 @@ static int xgene_enet_probe(struct platform_device *pdev)
 		goto err;
 	}
 
+	xgene_enet_check_phy_handle(pdata);
+
 	ret = xgene_enet_init_hw(pdata);
 	if (ret)
-		goto err;
+		goto err2;
 
 	link_state = pdata->mac_ops->link_state;
 	if (pdata->phy_mode == PHY_INTERFACE_MODE_XGMII) {
@@ -2115,35 +2085,36 @@ static int xgene_enet_probe(struct platform_device *pdev)
 	spin_lock_init(&pdata->stats_lock);
 	ret = xgene_extd_stats_init(pdata);
 	if (ret)
-		goto err2;
+		goto err1;
 
 	xgene_enet_napi_add(pdata);
 	ret = register_netdev(ndev);
 	if (ret) {
 		netdev_err(ndev, "Failed to register netdev\n");
-		goto err2;
+		goto err1;
 	}
 
 	return 0;
 
-err2:
+err1:
 	/*
 	 * If necessary, free_netdev() will call netif_napi_del() and undo
 	 * the effects of xgene_enet_napi_add()'s calls to netif_napi_add().
 	 */
 
+	xgene_enet_delete_desc_rings(pdata);
+
+err2:
 	if (pdata->mdio_driver)
 		xgene_enet_phy_disconnect(pdata);
 	else if (phy_interface_mode_is_rgmii(pdata->phy_mode))
 		xgene_enet_mdio_remove(pdata);
-err1:
-	xgene_enet_delete_desc_rings(pdata);
 err:
 	free_netdev(ndev);
 	return ret;
 }
 
-static int xgene_enet_remove(struct platform_device *pdev)
+static void xgene_enet_remove(struct platform_device *pdev)
 {
 	struct xgene_enet_pdata *pdata;
 	struct net_device *ndev;
@@ -2165,8 +2136,6 @@ static int xgene_enet_remove(struct platform_device *pdev)
 	xgene_enet_delete_desc_rings(pdata);
 	pdata->port_ops->shutdown(pdata);
 	free_netdev(ndev);
-
-	return 0;
 }
 
 static void xgene_enet_shutdown(struct platform_device *pdev)
@@ -2186,7 +2155,7 @@ static void xgene_enet_shutdown(struct platform_device *pdev)
 static struct platform_driver xgene_enet_driver = {
 	.driver = {
 		   .name = "xgene-enet",
-		   .of_match_table = of_match_ptr(xgene_enet_of_match),
+		   .of_match_table = xgene_enet_of_match,
 		   .acpi_match_table = ACPI_PTR(xgene_enet_acpi_match),
 	},
 	.probe = xgene_enet_probe,
@@ -2197,7 +2166,6 @@ static struct platform_driver xgene_enet_driver = {
 module_platform_driver(xgene_enet_driver);
 
 MODULE_DESCRIPTION("APM X-Gene SoC Ethernet driver");
-MODULE_VERSION(XGENE_DRV_VERSION);
 MODULE_AUTHOR("Iyappan Subramanian <isubramanian@apm.com>");
 MODULE_AUTHOR("Keyur Chudgar <kchudgar@apm.com>");
 MODULE_LICENSE("GPL");
diff --git a/drivers/net/ethernet/apm/xgene/xgene_enet_main.h b/drivers/net/ethernet/apm/xgene/xgene_enet_main.h
index 985768596900..bce2c19e3f22 100644
--- a/drivers/net/ethernet/apm/xgene/xgene_enet_main.h
+++ b/drivers/net/ethernet/apm/xgene/xgene_enet_main.h
@@ -1,22 +1,10 @@
+/* SPDX-License-Identifier: GPL-2.0-or-later */
 /* Applied Micro X-Gene SoC Ethernet Driver
  *
  * Copyright (c) 2014, Applied Micro Circuits Corporation
  * Authors: Iyappan Subramanian <isubramanian@apm.com>
  *	    Ravi Patel <rapatel@apm.com>
  *	    Keyur Chudgar <kchudgar@apm.com>
- *
- * This program is free software; you can redistribute  it and/or modify it
- * under  the terms of  the GNU General  Public License as published by the
- * Free Software Foundation;  either version 2 of the  License, or (at your
- * option) any later version.
- *
- * This program is distributed in the hope that it will be useful,
- * but WITHOUT ANY WARRANTY; without even the implied warranty of
- * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
- * GNU General Public License for more details.
- *
- * You should have received a copy of the GNU General Public License
- * along with this program.  If not, see <http://www.gnu.org/licenses/>.
  */
 
 #ifndef __XGENE_ENET_MAIN_H__
@@ -27,9 +15,11 @@
 #include <linux/efi.h>
 #include <linux/irq.h>
 #include <linux/io.h>
-#include <linux/of_platform.h>
+#include <linux/of.h>
 #include <linux/of_net.h>
 #include <linux/of_mdio.h>
+#include <linux/platform_device.h>
+#include <linux/mdio/mdio-xgene.h>
 #include <linux/module.h>
 #include <net/ip.h>
 #include <linux/prefetch.h>
@@ -38,9 +28,7 @@
 #include "xgene_enet_hw.h"
 #include "xgene_enet_cle.h"
 #include "xgene_enet_ring2.h"
-#include "../../../phy/mdio-xgene.h"
 
-#define XGENE_DRV_VERSION	"v1.0"
 #define ETHER_MIN_PACKET	64
 #define ETHER_STD_PACKET	1518
 #define XGENE_ENET_STD_MTU	1536
diff --git a/drivers/net/ethernet/apm/xgene/xgene_enet_ring2.c b/drivers/net/ethernet/apm/xgene/xgene_enet_ring2.c
index 4ff40559f970..02892efdc4dc 100644
--- a/drivers/net/ethernet/apm/xgene/xgene_enet_ring2.c
+++ b/drivers/net/ethernet/apm/xgene/xgene_enet_ring2.c
@@ -1,20 +1,8 @@
+// SPDX-License-Identifier: GPL-2.0-or-later
 /* Applied Micro X-Gene SoC Ethernet Driver
  *
  * Copyright (c) 2015, Applied Micro Circuits Corporation
  * Author: Iyappan Subramanian <isubramanian@apm.com>
- *
- * This program is free software; you can redistribute  it and/or modify it
- * under  the terms of  the GNU General  Public License as published by the
- * Free Software Foundation;  either version 2 of the  License, or (at your
- * option) any later version.
- *
- * This program is distributed in the hope that it will be useful,
- * but WITHOUT ANY WARRANTY; without even the implied warranty of
- * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
- * GNU General Public License for more details.
- *
- * You should have received a copy of the GNU General Public License
- * along with this program.  If not, see <http://www.gnu.org/licenses/>.
  */
 
 #include "xgene_enet_main.h"
diff --git a/drivers/net/ethernet/apm/xgene/xgene_enet_ring2.h b/drivers/net/ethernet/apm/xgene/xgene_enet_ring2.h
index 8b235db23c42..4e2edeea54ab 100644
--- a/drivers/net/ethernet/apm/xgene/xgene_enet_ring2.h
+++ b/drivers/net/ethernet/apm/xgene/xgene_enet_ring2.h
@@ -1,20 +1,8 @@
+/* SPDX-License-Identifier: GPL-2.0-or-later */
 /* Applied Micro X-Gene SoC Ethernet Driver
  *
  * Copyright (c) 2015, Applied Micro Circuits Corporation
  * Author: Iyappan Subramanian <isubramanian@apm.com>
- *
- * This program is free software; you can redistribute  it and/or modify it
- * under  the terms of  the GNU General  Public License as published by the
- * Free Software Foundation;  either version 2 of the  License, or (at your
- * option) any later version.
- *
- * This program is distributed in the hope that it will be useful,
- * but WITHOUT ANY WARRANTY; without even the implied warranty of
- * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
- * GNU General Public License for more details.
- *
- * You should have received a copy of the GNU General Public License
- * along with this program.  If not, see <http://www.gnu.org/licenses/>.
  */
 
 #ifndef __XGENE_ENET_RING2_H__
diff --git a/drivers/net/ethernet/apm/xgene/xgene_enet_sgmac.c b/drivers/net/ethernet/apm/xgene/xgene_enet_sgmac.c
index b1a83fdbefb8..72b5e8eb0ec7 100644
--- a/drivers/net/ethernet/apm/xgene/xgene_enet_sgmac.c
+++ b/drivers/net/ethernet/apm/xgene/xgene_enet_sgmac.c
@@ -1,21 +1,9 @@
+// SPDX-License-Identifier: GPL-2.0-or-later
 /* Applied Micro X-Gene SoC Ethernet Driver
  *
  * Copyright (c) 2014, Applied Micro Circuits Corporation
  * Authors: Iyappan Subramanian <isubramanian@apm.com>
  *	    Keyur Chudgar <kchudgar@apm.com>
- *
- * This program is free software; you can redistribute  it and/or modify it
- * under  the terms of  the GNU General  Public License as published by the
- * Free Software Foundation;  either version 2 of the  License, or (at your
- * option) any later version.
- *
- * This program is distributed in the hope that it will be useful,
- * but WITHOUT ANY WARRANTY; without even the implied warranty of
- * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
- * GNU General Public License for more details.
- *
- * You should have received a copy of the GNU General Public License
- * along with this program.  If not, see <http://www.gnu.org/licenses/>.
  */
 
 #include "xgene_enet_main.h"
@@ -177,8 +165,8 @@ static void xgene_sgmac_reset(struct xgene_enet_pdata *p)
 
 static void xgene_sgmac_set_mac_addr(struct xgene_enet_pdata *p)
 {
+	const u8 *dev_addr = p->ndev->dev_addr;
 	u32 addr0, addr1;
-	u8 *dev_addr = p->ndev->dev_addr;
 
 	addr0 = (dev_addr[3] << 24) | (dev_addr[2] << 16) |
 		(dev_addr[1] << 8) | dev_addr[0];
@@ -472,12 +460,14 @@ static int xgene_enet_reset(struct xgene_enet_pdata *p)
 		}
 	} else {
 #ifdef CONFIG_ACPI
-		if (acpi_has_method(ACPI_HANDLE(&p->pdev->dev), "_RST"))
-			acpi_evaluate_object(ACPI_HANDLE(&p->pdev->dev),
-					     "_RST", NULL, NULL);
-		else if (acpi_has_method(ACPI_HANDLE(&p->pdev->dev), "_INI"))
+		acpi_status status;
+
+		status = acpi_evaluate_object(ACPI_HANDLE(&p->pdev->dev),
+					      "_RST", NULL, NULL);
+		if (ACPI_FAILURE(status)) {
 			acpi_evaluate_object(ACPI_HANDLE(&p->pdev->dev),
 					     "_INI", NULL, NULL);
+		}
 #endif
 	}
 
diff --git a/drivers/net/ethernet/apm/xgene/xgene_enet_sgmac.h b/drivers/net/ethernet/apm/xgene/xgene_enet_sgmac.h
index 3d0ba374491b..3bba0ce34bb4 100644
--- a/drivers/net/ethernet/apm/xgene/xgene_enet_sgmac.h
+++ b/drivers/net/ethernet/apm/xgene/xgene_enet_sgmac.h
@@ -1,21 +1,9 @@
+/* SPDX-License-Identifier: GPL-2.0-or-later */
 /* Applied Micro X-Gene SoC Ethernet Driver
  *
  * Copyright (c) 2014, Applied Micro Circuits Corporation
  * Authors: Iyappan Subramanian <isubramanian@apm.com>
  *	    Keyur Chudgar <kchudgar@apm.com>
- *
- * This program is free software; you can redistribute  it and/or modify it
- * under  the terms of  the GNU General  Public License as published by the
- * Free Software Foundation;  either version 2 of the  License, or (at your
- * option) any later version.
- *
- * This program is distributed in the hope that it will be useful,
- * but WITHOUT ANY WARRANTY; without even the implied warranty of
- * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
- * GNU General Public License for more details.
- *
- * You should have received a copy of the GNU General Public License
- * along with this program.  If not, see <http://www.gnu.org/licenses/>.
  */
 
 #ifndef __XGENE_ENET_SGMAC_H__
diff --git a/drivers/net/ethernet/apm/xgene/xgene_enet_xgmac.c b/drivers/net/ethernet/apm/xgene/xgene_enet_xgmac.c
index b7d75d067c7a..cc3b1631c905 100644
--- a/drivers/net/ethernet/apm/xgene/xgene_enet_xgmac.c
+++ b/drivers/net/ethernet/apm/xgene/xgene_enet_xgmac.c
@@ -1,25 +1,19 @@
+// SPDX-License-Identifier: GPL-2.0-or-later
 /* Applied Micro X-Gene SoC Ethernet Driver
  *
  * Copyright (c) 2014, Applied Micro Circuits Corporation
  * Authors: Iyappan Subramanian <isubramanian@apm.com>
  *	    Keyur Chudgar <kchudgar@apm.com>
- *
- * This program is free software; you can redistribute  it and/or modify it
- * under  the terms of  the GNU General  Public License as published by the
- * Free Software Foundation;  either version 2 of the  License, or (at your
- * option) any later version.
- *
- * This program is distributed in the hope that it will be useful,
- * but WITHOUT ANY WARRANTY; without even the implied warranty of
- * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
- * GNU General Public License for more details.
- *
- * You should have received a copy of the GNU General Public License
- * along with this program.  If not, see <http://www.gnu.org/licenses/>.
  */
 
-#include <linux/of_gpio.h>
-#include <linux/gpio.h>
+#include <linux/acpi.h>
+#include <linux/clk.h>
+#include <linux/err.h>
+#include <linux/gpio/consumer.h>
+#include <linux/io.h>
+#include <linux/types.h>
+#include <linux/workqueue.h>
+
 #include "xgene_enet_main.h"
 #include "xgene_enet_hw.h"
 #include "xgene_enet_xgmac.h"
@@ -219,8 +213,8 @@ static void xgene_pcs_reset(struct xgene_enet_pdata *pdata)
 
 static void xgene_xgmac_set_mac_addr(struct xgene_enet_pdata *pdata)
 {
+	const u8 *dev_addr = pdata->ndev->dev_addr;
 	u32 addr0, addr1;
-	u8 *dev_addr = pdata->ndev->dev_addr;
 
 	addr0 = (dev_addr[3] << 24) | (dev_addr[2] << 16) |
 		(dev_addr[1] << 8) | dev_addr[0];
@@ -405,11 +399,11 @@ static int xgene_enet_reset(struct xgene_enet_pdata *pdata)
 		udelay(5);
 	} else {
 #ifdef CONFIG_ACPI
-		if (acpi_has_method(ACPI_HANDLE(&pdata->pdev->dev), "_RST")) {
-			acpi_evaluate_object(ACPI_HANDLE(&pdata->pdev->dev),
-					     "_RST", NULL, NULL);
-		} else if (acpi_has_method(ACPI_HANDLE(&pdata->pdev->dev),
-					   "_INI")) {
+		acpi_status status;
+
+		status = acpi_evaluate_object(ACPI_HANDLE(&pdata->pdev->dev),
+					      "_RST", NULL, NULL);
+		if (ACPI_FAILURE(status)) {
 			acpi_evaluate_object(ACPI_HANDLE(&pdata->pdev->dev),
 					     "_INI", NULL, NULL);
 		}
diff --git a/drivers/net/ethernet/apm/xgene/xgene_enet_xgmac.h b/drivers/net/ethernet/apm/xgene/xgene_enet_xgmac.h
index a3b45517df45..98622dcf6c53 100644
--- a/drivers/net/ethernet/apm/xgene/xgene_enet_xgmac.h
+++ b/drivers/net/ethernet/apm/xgene/xgene_enet_xgmac.h
@@ -1,21 +1,9 @@
+/* SPDX-License-Identifier: GPL-2.0-or-later */
 /* Applied Micro X-Gene SoC Ethernet Driver
  *
  * Copyright (c) 2014, Applied Micro Circuits Corporation
  * Authors: Iyappan Subramanian <isubramanian@apm.com>
  *	    Keyur Chudgar <kchudgar@apm.com>
- *
- * This program is free software; you can redistribute  it and/or modify it
- * under  the terms of  the GNU General  Public License as published by the
- * Free Software Foundation;  either version 2 of the  License, or (at your
- * option) any later version.
- *
- * This program is distributed in the hope that it will be useful,
- * but WITHOUT ANY WARRANTY; without even the implied warranty of
- * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
- * GNU General Public License for more details.
- *
- * You should have received a copy of the GNU General Public License
- * along with this program.  If not, see <http://www.gnu.org/licenses/>.
  */
 
 #ifndef __XGENE_ENET_XGMAC_H__
diff --git a/drivers/net/ethernet/apple/Kconfig b/drivers/net/ethernet/apple/Kconfig
index 31071297896c..a4176d2ecec6 100644
--- a/drivers/net/ethernet/apple/Kconfig
+++ b/drivers/net/ethernet/apple/Kconfig
@@ -1,3 +1,4 @@
+# SPDX-License-Identifier: GPL-2.0-only
 #
 # Apple device configuration
 #
@@ -6,12 +7,12 @@ config NET_VENDOR_APPLE
 	bool "Apple devices"
 	default y
 	depends on (PPC_PMAC && PPC32) || MAC
-	---help---
+	help
 	  If you have a network (Ethernet) card belonging to this class, say Y.
 
 	  Note that the answer to this question doesn't directly affect the
-	  kernel: saying N will just cause the configurator to skip all
-	  the questions about IBM devices. If you say Y, you will be asked for
+	  kernel: saying N will just cause the configurator to skip all the
+	  questions about Apple devices. If you say Y, you will be asked for
 	  your specific card in the following questions.
 
 if NET_VENDOR_APPLE
@@ -20,7 +21,7 @@ config MACE
 	tristate "MACE (Power Mac ethernet) support"
 	depends on PPC_PMAC && PPC32
 	select CRC32
-	---help---
+	help
 	  Power Macintoshes and clones with Ethernet built-in on the
 	  motherboard will usually use a MACE (Medium Access Control for
 	  Ethernet) interface. Say Y to include support for the MACE chip.
@@ -31,7 +32,7 @@ config MACE
 config MACE_AAUI_PORT
 	bool "Use AAUI port instead of TP by default"
 	depends on MACE
-	---help---
+	help
 	  Some Apple machines (notably the Apple Network Server) which use the
 	  MACE ethernet chip have an Apple AUI port (small 15-pin connector),
 	  instead of an 8-pin RJ45 connector for twisted-pair ethernet.  Say
@@ -43,7 +44,7 @@ config BMAC
 	tristate "BMAC (G3 ethernet) support"
 	depends on PPC_PMAC && PPC32
 	select CRC32
-	---help---
+	help
 	  Say Y for support of BMAC Ethernet interfaces. These are used on G3
 	  computers.
 
@@ -54,7 +55,7 @@ config MACMACE
 	tristate "Macintosh (AV) onboard MACE ethernet"
 	depends on MAC
 	select CRC32
-	---help---
+	help
 	  Support for the onboard AMD 79C940 MACE Ethernet controller used in
 	  the 660AV and 840AV Macintosh.  If you have one of these Macintoshes
 	  say Y here.
diff --git a/drivers/net/ethernet/apple/Makefile b/drivers/net/ethernet/apple/Makefile
index 86eaa17af0f4..322457027546 100644
--- a/drivers/net/ethernet/apple/Makefile
+++ b/drivers/net/ethernet/apple/Makefile
@@ -1,3 +1,4 @@
+# SPDX-License-Identifier: GPL-2.0-only
 #
 # Makefile for the Apple network device drivers.
 #
diff --git a/drivers/net/ethernet/apple/bmac.c b/drivers/net/ethernet/apple/bmac.c
index 2b2d87089987..b3bf8d6f88e8 100644
--- a/drivers/net/ethernet/apple/bmac.c
+++ b/drivers/net/ethernet/apple/bmac.c
@@ -1,3 +1,4 @@
+// SPDX-License-Identifier: GPL-2.0-only
 /*
  * Network device driver for the BMAC ethernet controller on
  * Apple Powermacs.  Assumes it's under a DBDMA controller.
@@ -22,11 +23,10 @@
 #include <linux/bitrev.h>
 #include <linux/ethtool.h>
 #include <linux/slab.h>
-#include <asm/prom.h>
+#include <linux/pgtable.h>
 #include <asm/dbdma.h>
 #include <asm/io.h>
 #include <asm/page.h>
-#include <asm/pgtable.h>
 #include <asm/machdep.h>
 #include <asm/pmac_feature.h>
 #include <asm/macio.h>
@@ -37,11 +37,6 @@
 #define trunc_page(x)	((void *)(((unsigned long)(x)) & ~((unsigned long)(PAGE_SIZE - 1))))
 #define round_page(x)	trunc_page(((unsigned long)(x)) + ((unsigned long)(PAGE_SIZE - 1)))
 
-/*
- * CRC polynomial - used in working out multicast filter bits.
- */
-#define ENET_CRCPOLY 0x04c11db7
-
 /* switch to use multicast code lifted from sunhme driver */
 #define SUNHME_MULTICAST
 
@@ -157,8 +152,8 @@ static irqreturn_t bmac_misc_intr(int irq, void *dev_id);
 static irqreturn_t bmac_txdma_intr(int irq, void *dev_id);
 static irqreturn_t bmac_rxdma_intr(int irq, void *dev_id);
 static void bmac_set_timeout(struct net_device *dev);
-static void bmac_tx_timeout(unsigned long data);
-static int bmac_output(struct sk_buff *skb, struct net_device *dev);
+static void bmac_tx_timeout(struct timer_list *t);
+static netdev_tx_t bmac_output(struct sk_buff *skb, struct net_device *dev);
 static void bmac_start(struct net_device *dev);
 
 #define	DBDMA_SET(x)	( ((x) | (x) << 16) )
@@ -311,7 +306,7 @@ bmac_init_registers(struct net_device *dev)
 {
 	struct bmac_data *bp = netdev_priv(dev);
 	volatile unsigned short regValue;
-	unsigned short *pWord16;
+	const unsigned short *pWord16;
 	int i;
 
 	/* XXDEBUG(("bmac: enter init_registers\n")); */
@@ -374,7 +369,7 @@ bmac_init_registers(struct net_device *dev)
 	bmwrite(dev, BHASH1, bp->hash_table_mask[2]); 	/* bits 47 - 32 */
 	bmwrite(dev, BHASH0, bp->hash_table_mask[3]); 	/* bits 63 - 48 */
 
-	pWord16 = (unsigned short *)dev->dev_addr;
+	pWord16 = (const unsigned short *)dev->dev_addr;
 	bmwrite(dev, MADD0, *pWord16++);
 	bmwrite(dev, MADD1, *pWord16++);
 	bmwrite(dev, MADD2, *pWord16);
@@ -465,7 +460,7 @@ static int bmac_suspend(struct macio_dev *mdev, pm_message_t state)
 	/* prolly should wait for dma to finish & turn off the chip */
 	spin_lock_irqsave(&bp->lock, flags);
 	if (bp->timeout_active) {
-		del_timer(&bp->tx_timeout);
+		timer_delete(&bp->tx_timeout);
 		bp->timeout_active = 0;
 	}
 	disable_irq(dev->irq);
@@ -480,26 +475,26 @@ static int bmac_suspend(struct macio_dev *mdev, pm_message_t state)
 		config = bmread(dev, RXCFG);
 		bmwrite(dev, RXCFG, (config & ~RxMACEnable));
 		config = bmread(dev, TXCFG);
-       		bmwrite(dev, TXCFG, (config & ~TxMACEnable));
+		bmwrite(dev, TXCFG, (config & ~TxMACEnable));
 		bmwrite(dev, INTDISABLE, DisableAll); /* disable all intrs */
-       		/* disable rx and tx dma */
+		/* disable rx and tx dma */
 		rd->control = cpu_to_le32(DBDMA_CLEAR(RUN|PAUSE|FLUSH|WAKE));	/* clear run bit */
 		td->control = cpu_to_le32(DBDMA_CLEAR(RUN|PAUSE|FLUSH|WAKE));	/* clear run bit */
-       		/* free some skb's */
-       		for (i=0; i<N_RX_RING; i++) {
-       			if (bp->rx_bufs[i] != NULL) {
-       				dev_kfree_skb(bp->rx_bufs[i]);
-       				bp->rx_bufs[i] = NULL;
-       			}
-       		}
-       		for (i = 0; i<N_TX_RING; i++) {
+		/* free some skb's */
+		for (i=0; i<N_RX_RING; i++) {
+			if (bp->rx_bufs[i] != NULL) {
+				dev_kfree_skb(bp->rx_bufs[i]);
+				bp->rx_bufs[i] = NULL;
+			}
+		}
+		for (i = 0; i<N_TX_RING; i++) {
 			if (bp->tx_bufs[i] != NULL) {
 		       		dev_kfree_skb(bp->tx_bufs[i]);
 	       			bp->tx_bufs[i] = NULL;
 		       	}
 		}
 	}
-       	pmac_call_feature(PMAC_FTR_BMAC_ENABLE, macio_get_of_node(bp->mdev), 0, 0);
+	pmac_call_feature(PMAC_FTR_BMAC_ENABLE, macio_get_of_node(bp->mdev), 0, 0);
 	return 0;
 }
 
@@ -513,9 +508,9 @@ static int bmac_resume(struct macio_dev *mdev)
 		bmac_reset_and_enable(dev);
 
 	enable_irq(dev->irq);
-       	enable_irq(bp->tx_dma_intr);
-       	enable_irq(bp->rx_dma_intr);
-       	netif_device_attach(dev);
+	enable_irq(bp->tx_dma_intr);
+	enable_irq(bp->rx_dma_intr);
+	netif_device_attach(dev);
 
 	return 0;
 }
@@ -524,19 +519,16 @@ static int bmac_resume(struct macio_dev *mdev)
 static int bmac_set_address(struct net_device *dev, void *addr)
 {
 	struct bmac_data *bp = netdev_priv(dev);
-	unsigned char *p = addr;
-	unsigned short *pWord16;
+	const unsigned short *pWord16;
 	unsigned long flags;
-	int i;
 
 	XXDEBUG(("bmac: enter set_address\n"));
 	spin_lock_irqsave(&bp->lock, flags);
 
-	for (i = 0; i < 6; ++i) {
-		dev->dev_addr[i] = p[i];
-	}
+	eth_hw_addr_set(dev, addr);
+
 	/* load up the hardware address */
-	pWord16  = (unsigned short *)dev->dev_addr;
+	pWord16  = (const unsigned short *)dev->dev_addr;
 	bmwrite(dev, MADD0, *pWord16++);
 	bmwrite(dev, MADD1, *pWord16++);
 	bmwrite(dev, MADD2, *pWord16);
@@ -553,10 +545,8 @@ static inline void bmac_set_timeout(struct net_device *dev)
 
 	spin_lock_irqsave(&bp->lock, flags);
 	if (bp->timeout_active)
-		del_timer(&bp->tx_timeout);
+		timer_delete(&bp->tx_timeout);
 	bp->tx_timeout.expires = jiffies + TX_TIMEOUT;
-	bp->tx_timeout.function = bmac_tx_timeout;
-	bp->tx_timeout.data = (unsigned long) dev;
 	add_timer(&bp->tx_timeout);
 	bp->timeout_active = 1;
 	spin_unlock_irqrestore(&bp->lock, flags);
@@ -764,7 +754,7 @@ static irqreturn_t bmac_txdma_intr(int irq, void *dev_id)
 		XXDEBUG(("bmac_txdma_intr\n"));
 	}
 
-	/*     del_timer(&bp->tx_timeout); */
+	/*     timer_delete(&bp->tx_timeout); */
 	/*     bp->timeout_active = 0; */
 
 	while (1) {
@@ -783,7 +773,7 @@ static irqreturn_t bmac_txdma_intr(int irq, void *dev_id)
 
 		if (bp->tx_bufs[bp->tx_empty]) {
 			++dev->stats.tx_packets;
-			dev_kfree_skb_irq(bp->tx_bufs[bp->tx_empty]);
+			dev_consume_skb_irq(bp->tx_bufs[bp->tx_empty]);
 		}
 		bp->tx_bufs[bp->tx_empty] = NULL;
 		bp->tx_fullup = 0;
@@ -805,59 +795,6 @@ static irqreturn_t bmac_txdma_intr(int irq, void *dev_id)
 }
 
 #ifndef SUNHME_MULTICAST
-/* Real fast bit-reversal algorithm, 6-bit values */
-static int reverse6[64] = {
-	0x0,0x20,0x10,0x30,0x8,0x28,0x18,0x38,
-	0x4,0x24,0x14,0x34,0xc,0x2c,0x1c,0x3c,
-	0x2,0x22,0x12,0x32,0xa,0x2a,0x1a,0x3a,
-	0x6,0x26,0x16,0x36,0xe,0x2e,0x1e,0x3e,
-	0x1,0x21,0x11,0x31,0x9,0x29,0x19,0x39,
-	0x5,0x25,0x15,0x35,0xd,0x2d,0x1d,0x3d,
-	0x3,0x23,0x13,0x33,0xb,0x2b,0x1b,0x3b,
-	0x7,0x27,0x17,0x37,0xf,0x2f,0x1f,0x3f
-};
-
-static unsigned int
-crc416(unsigned int curval, unsigned short nxtval)
-{
-	register unsigned int counter, cur = curval, next = nxtval;
-	register int high_crc_set, low_data_set;
-
-	/* Swap bytes */
-	next = ((next & 0x00FF) << 8) | (next >> 8);
-
-	/* Compute bit-by-bit */
-	for (counter = 0; counter < 16; ++counter) {
-		/* is high CRC bit set? */
-		if ((cur & 0x80000000) == 0) high_crc_set = 0;
-		else high_crc_set = 1;
-
-		cur = cur << 1;
-
-		if ((next & 0x0001) == 0) low_data_set = 0;
-		else low_data_set = 1;
-
-		next = next >> 1;
-
-		/* do the XOR */
-		if (high_crc_set ^ low_data_set) cur = cur ^ ENET_CRCPOLY;
-	}
-	return cur;
-}
-
-static unsigned int
-bmac_crc(unsigned short *address)
-{
-	unsigned int newcrc;
-
-	XXDEBUG(("bmac_crc: addr=%#04x, %#04x, %#04x\n", *address, address[1], address[2]));
-	newcrc = crc416(0xffffffff, *address);	/* address bits 47 - 32 */
-	newcrc = crc416(newcrc, address[1]);	/* address bits 31 - 16 */
-	newcrc = crc416(newcrc, address[2]);	/* address bits 15 - 0  */
-
-	return(newcrc);
-}
-
 /*
  * Add requested mcast addr to BMac's hash table filter.
  *
@@ -870,8 +807,7 @@ bmac_addhash(struct bmac_data *bp, unsigned char *addr)
 	unsigned short	 mask;
 
 	if (!(*addr)) return;
-	crc = bmac_crc((unsigned short *)addr) & 0x3f; /* Big-endian alert! */
-	crc = reverse6[crc];	/* Hyperfast bit-reversing algorithm */
+	crc = crc32(~0, addr, ETH_ALEN) >> 26;
 	if (bp->hash_use_count[crc]++) return; /* This bit is already set */
 	mask = crc % 16;
 	mask = (unsigned char)1 << mask;
@@ -885,8 +821,7 @@ bmac_removehash(struct bmac_data *bp, unsigned char *addr)
 	unsigned char mask;
 
 	/* Now, delete the address from the filter copy, as indicated */
-	crc = bmac_crc((unsigned short *)addr) & 0x3f; /* Big-endian alert! */
-	crc = reverse6[crc];	/* Hyperfast bit-reversing algorithm */
+	crc = crc32(~0, addr, ETH_ALEN) >> 26;
 	if (bp->hash_use_count[crc] == 0) return; /* That bit wasn't in use! */
 	if (--bp->hash_use_count[crc]) return; /* That bit is still in use */
 	mask = crc % 16;
@@ -1187,7 +1122,7 @@ bmac_get_station_address(struct net_device *dev, unsigned char *ea)
 	int i;
 	unsigned short data;
 
-	for (i = 0; i < 6; i++)
+	for (i = 0; i < 3; i++)
 		{
 			reset_and_select_srom(dev);
 			data = read_srom(dev, i + EnetAddressOffset/2, SROMAddressBits);
@@ -1218,8 +1153,7 @@ static void bmac_reset_and_enable(struct net_device *dev)
 	 */
 	skb = netdev_alloc_skb(dev, ETHERMINPACKET);
 	if (skb != NULL) {
-		data = skb_put(skb, ETHERMINPACKET);
-		memset(data, 0, ETHERMINPACKET);
+		data = skb_put_zero(skb, ETHERMINPACKET);
 		memcpy(data, dev->dev_addr, ETH_ALEN);
 		memcpy(data + ETH_ALEN, dev->dev_addr, ETH_ALEN);
 		bmac_transmit_packet(skb, dev);
@@ -1246,6 +1180,7 @@ static int bmac_probe(struct macio_dev *mdev, const struct of_device_id *match)
 	struct bmac_data *bp;
 	const unsigned char *prop_addr;
 	unsigned char addr[6];
+	u8 macaddr[6];
 	struct net_device *dev;
 	int is_bmac_plus = ((int)match->data) != 0;
 
@@ -1293,7 +1228,9 @@ static int bmac_probe(struct macio_dev *mdev, const struct of_device_id *match)
 
 	rev = addr[0] == 0 && addr[1] == 0xA0;
 	for (j = 0; j < 6; ++j)
-		dev->dev_addr[j] = rev ? bitrev8(addr[j]): addr[j];
+		macaddr[j] = rev ? bitrev8(addr[j]): addr[j];
+
+	eth_hw_addr_set(dev, macaddr);
 
 	/* Enable chip without interrupts for now */
 	bmac_enable_and_reset_chip(dev);
@@ -1322,9 +1259,9 @@ static int bmac_probe(struct macio_dev *mdev, const struct of_device_id *match)
 	bp->queue = (struct sk_buff_head *)(bp->rx_cmds + N_RX_RING + 1);
 	skb_queue_head_init(bp->queue);
 
-	init_timer(&bp->tx_timeout);
+	timer_setup(&bp->tx_timeout, bmac_tx_timeout, 0);
 
-	ret = request_irq(dev->irq, bmac_misc_intr, 0, "BMAC-misc", dev);
+	ret = request_irq(dev->irq, bmac_misc_intr, IRQF_NO_AUTOEN, "BMAC-misc", dev);
 	if (ret) {
 		printk(KERN_ERR "BMAC: can't get irq %d\n", dev->irq);
 		goto err_out_iounmap_rx;
@@ -1343,7 +1280,6 @@ static int bmac_probe(struct macio_dev *mdev, const struct of_device_id *match)
 	/* Mask chip interrupts and disable chip, will be
 	 * re-enabled on open()
 	 */
-	disable_irq(dev->irq);
 	pmac_call_feature(PMAC_FTR_BMAC_ENABLE, macio_get_of_node(bp->mdev), 0, 0);
 
 	if (register_netdev(dev) != 0) {
@@ -1463,7 +1399,7 @@ bmac_start(struct net_device *dev)
 	spin_unlock_irqrestore(&bp->lock, flags);
 }
 
-static int
+static netdev_tx_t
 bmac_output(struct sk_buff *skb, struct net_device *dev)
 {
 	struct bmac_data *bp = netdev_priv(dev);
@@ -1472,10 +1408,10 @@ bmac_output(struct sk_buff *skb, struct net_device *dev)
 	return NETDEV_TX_OK;
 }
 
-static void bmac_tx_timeout(unsigned long data)
+static void bmac_tx_timeout(struct timer_list *t)
 {
-	struct net_device *dev = (struct net_device *) data;
-	struct bmac_data *bp = netdev_priv(dev);
+	struct bmac_data *bp = timer_container_of(bp, t, tx_timeout);
+	struct net_device *dev = macio_get_drvdata(bp->mdev);
 	volatile struct dbdma_regs __iomem *td = bp->tx_dma;
 	volatile struct dbdma_regs __iomem *rd = bp->rx_dma;
 	volatile struct dbdma_cmd *cp;
@@ -1517,7 +1453,7 @@ static void bmac_tx_timeout(unsigned long data)
 	i = bp->tx_empty;
 	++dev->stats.tx_errors;
 	if (i != bp->tx_fill) {
-		dev_kfree_skb(bp->tx_bufs[i]);
+		dev_kfree_skb_irq(bp->tx_bufs[i]);
 		bp->tx_bufs[i] = NULL;
 		if (++i >= N_TX_RING) i = 0;
 		bp->tx_empty = i;
@@ -1598,14 +1534,14 @@ bmac_proc_info(char *buffer, char **start, off_t offset, int length)
 }
 #endif
 
-static int bmac_remove(struct macio_dev *mdev)
+static void bmac_remove(struct macio_dev *mdev)
 {
 	struct net_device *dev = macio_get_drvdata(mdev);
 	struct bmac_data *bp = netdev_priv(dev);
 
 	unregister_netdev(dev);
 
-       	free_irq(dev->irq, dev);
+	free_irq(dev->irq, dev);
 	free_irq(bp->tx_dma_intr, dev);
 	free_irq(bp->rx_dma_intr, dev);
 
@@ -1616,8 +1552,6 @@ static int bmac_remove(struct macio_dev *mdev)
 	macio_release_resources(mdev);
 
 	free_netdev(dev);
-
-	return 0;
 }
 
 static const struct of_device_id bmac_match[] =
diff --git a/drivers/net/ethernet/apple/bmac.h b/drivers/net/ethernet/apple/bmac.h
index a1d19d867ba5..f8826c4ce792 100644
--- a/drivers/net/ethernet/apple/bmac.h
+++ b/drivers/net/ethernet/apple/bmac.h
@@ -1,13 +1,9 @@
+/* SPDX-License-Identifier: GPL-2.0-or-later */
 /*
  * mace.h - definitions for the registers in the "Big Mac"
  *  Ethernet controller found in PowerMac G3 models.
  *
  * Copyright (C) 1998 Randy Gobbel.
- *
- * This program is free software; you can redistribute it and/or
- * modify it under the terms of the GNU General Public License
- * as published by the Free Software Foundation; either version
- * 2 of the License, or (at your option) any later version.
  */
 
 /* The "Big MAC" appears to have some parts in common with the Sun "Happy Meal"
diff --git a/drivers/net/ethernet/apple/mace.c b/drivers/net/ethernet/apple/mace.c
index 96dd5300e0e5..af26905e44e3 100644
--- a/drivers/net/ethernet/apple/mace.c
+++ b/drivers/net/ethernet/apple/mace.c
@@ -1,3 +1,4 @@
+// SPDX-License-Identifier: GPL-2.0-only
 /*
  * Network device driver for the MACE ethernet controller on
  * Apple Powermacs.  Assumes it's under a DBDMA controller.
@@ -18,10 +19,9 @@
 #include <linux/spinlock.h>
 #include <linux/bitrev.h>
 #include <linux/slab.h>
-#include <asm/prom.h>
+#include <linux/pgtable.h>
 #include <asm/dbdma.h>
 #include <asm/io.h>
-#include <asm/pgtable.h>
 #include <asm/macio.h>
 
 #include "mace.h"
@@ -78,7 +78,7 @@ struct mace_data {
 
 static int mace_open(struct net_device *dev);
 static int mace_close(struct net_device *dev);
-static int mace_xmit_start(struct sk_buff *skb, struct net_device *dev);
+static netdev_tx_t mace_xmit_start(struct sk_buff *skb, struct net_device *dev);
 static void mace_set_multicast(struct net_device *dev);
 static void mace_reset(struct net_device *dev);
 static int mace_set_address(struct net_device *dev, void *addr);
@@ -86,10 +86,10 @@ static irqreturn_t mace_interrupt(int irq, void *dev_id);
 static irqreturn_t mace_txdma_intr(int irq, void *dev_id);
 static irqreturn_t mace_rxdma_intr(int irq, void *dev_id);
 static void mace_set_timeout(struct net_device *dev);
-static void mace_tx_timeout(unsigned long data);
+static void mace_tx_timeout(struct timer_list *t);
 static inline void dbdma_reset(volatile struct dbdma_regs __iomem *dma);
 static inline void mace_clean_rings(struct mace_data *mp);
-static void __mace_set_address(struct net_device *dev, void *addr);
+static void __mace_set_address(struct net_device *dev, const void *addr);
 
 /*
  * If we can't get a skbuff when we need it, we use this area for DMA.
@@ -111,11 +111,12 @@ static int mace_probe(struct macio_dev *mdev, const struct of_device_id *match)
 	struct net_device *dev;
 	struct mace_data *mp;
 	const unsigned char *addr;
+	u8 macaddr[ETH_ALEN];
 	int j, rev, rc = -EBUSY;
 
 	if (macio_resource_count(mdev) != 3 || macio_irq_count(mdev) != 3) {
-		printk(KERN_ERR "can't use MACE %s: need 3 addrs and 3 irqs\n",
-		       mace->full_name);
+		printk(KERN_ERR "can't use MACE %pOF: need 3 addrs and 3 irqs\n",
+		       mace);
 		return -ENODEV;
 	}
 
@@ -123,8 +124,8 @@ static int mace_probe(struct macio_dev *mdev, const struct of_device_id *match)
 	if (addr == NULL) {
 		addr = of_get_property(mace, "local-mac-address", NULL);
 		if (addr == NULL) {
-			printk(KERN_ERR "Can't get mac-address for MACE %s\n",
-			       mace->full_name);
+			printk(KERN_ERR "Can't get mac-address for MACE %pOF\n",
+			       mace);
 			return -ENODEV;
 		}
 	}
@@ -166,8 +167,9 @@ static int mace_probe(struct macio_dev *mdev, const struct of_device_id *match)
 
 	rev = addr[0] == 0 && addr[1] == 0xA0;
 	for (j = 0; j < 6; ++j) {
-		dev->dev_addr[j] = rev ? bitrev8(addr[j]): addr[j];
+		macaddr[j] = rev ? bitrev8(addr[j]): addr[j];
 	}
+	eth_hw_addr_set(dev, macaddr);
 	mp->chipid = (in_8(&mp->mace->chipid_hi) << 8) |
 			in_8(&mp->mace->chipid_lo);
 
@@ -196,7 +198,7 @@ static int mace_probe(struct macio_dev *mdev, const struct of_device_id *match)
 
 	memset((char *) mp->tx_cmds, 0,
 	       (NCMDS_TX*N_TX_RING + N_RX_RING + 2) * sizeof(struct dbdma_cmd));
-	init_timer(&mp->tx_timeout);
+	timer_setup(&mp->tx_timeout, mace_tx_timeout, 0);
 	spin_lock_init(&mp->lock);
 	mp->timeout_active = 0;
 
@@ -270,7 +272,7 @@ static int mace_probe(struct macio_dev *mdev, const struct of_device_id *match)
 	return rc;
 }
 
-static int mace_remove(struct macio_dev *mdev)
+static void mace_remove(struct macio_dev *mdev)
 {
 	struct net_device *dev = macio_get_drvdata(mdev);
 	struct mace_data *mp;
@@ -294,8 +296,6 @@ static int mace_remove(struct macio_dev *mdev)
 	free_netdev(dev);
 
 	macio_release_resources(mdev);
-
-	return 0;
 }
 
 static void dbdma_reset(volatile struct dbdma_regs __iomem *dma)
@@ -363,28 +363,32 @@ static void mace_reset(struct net_device *dev)
 	out_8(&mb->iac, 0);
 
     if (mp->port_aaui)
-    	out_8(&mb->plscc, PORTSEL_AUI + ENPLSIO);
+	out_8(&mb->plscc, PORTSEL_AUI + ENPLSIO);
     else
-    	out_8(&mb->plscc, PORTSEL_GPSI + ENPLSIO);
+	out_8(&mb->plscc, PORTSEL_GPSI + ENPLSIO);
 }
 
-static void __mace_set_address(struct net_device *dev, void *addr)
+static void __mace_set_address(struct net_device *dev, const void *addr)
 {
     struct mace_data *mp = netdev_priv(dev);
     volatile struct mace __iomem *mb = mp->mace;
-    unsigned char *p = addr;
+    const unsigned char *p = addr;
+    u8 macaddr[ETH_ALEN];
     int i;
 
     /* load up the hardware address */
     if (mp->chipid == BROKEN_ADDRCHG_REV)
-    	out_8(&mb->iac, PHYADDR);
+	out_8(&mb->iac, PHYADDR);
     else {
-    	out_8(&mb->iac, ADDRCHG | PHYADDR);
+	out_8(&mb->iac, ADDRCHG | PHYADDR);
 	while ((in_8(&mb->iac) & ADDRCHG) != 0)
 	    ;
     }
     for (i = 0; i < 6; ++i)
-	out_8(&mb->padr, dev->dev_addr[i] = p[i]);
+        out_8(&mb->padr, macaddr[i] = p[i]);
+
+    eth_hw_addr_set(dev, macaddr);
+
     if (mp->chipid != BROKEN_ADDRCHG_REV)
         out_8(&mb->iac, 0);
 }
@@ -519,15 +523,13 @@ static inline void mace_set_timeout(struct net_device *dev)
     struct mace_data *mp = netdev_priv(dev);
 
     if (mp->timeout_active)
-	del_timer(&mp->tx_timeout);
+	timer_delete(&mp->tx_timeout);
     mp->tx_timeout.expires = jiffies + TX_TIMEOUT;
-    mp->tx_timeout.function = mace_tx_timeout;
-    mp->tx_timeout.data = (unsigned long) dev;
     add_timer(&mp->tx_timeout);
     mp->timeout_active = 1;
 }
 
-static int mace_xmit_start(struct sk_buff *skb, struct net_device *dev)
+static netdev_tx_t mace_xmit_start(struct sk_buff *skb, struct net_device *dev)
 {
     struct mace_data *mp = netdev_priv(dev);
     volatile struct dbdma_regs __iomem *td = mp->tx_dma;
@@ -674,7 +676,7 @@ static irqreturn_t mace_interrupt(int irq, void *dev_id)
 
     i = mp->tx_empty;
     while (in_8(&mb->pr) & XMTSV) {
-	del_timer(&mp->tx_timeout);
+	timer_delete(&mp->tx_timeout);
 	mp->timeout_active = 0;
 	/*
 	 * Clear any interrupt indication associated with this status
@@ -766,7 +768,7 @@ static irqreturn_t mace_interrupt(int irq, void *dev_id)
 	    dev->stats.tx_bytes += mp->tx_bufs[i]->len;
 	    ++dev->stats.tx_packets;
 	}
-	dev_kfree_skb_irq(mp->tx_bufs[i]);
+	dev_consume_skb_irq(mp->tx_bufs[i]);
 	--mp->tx_active;
 	if (++i >= N_TX_RING)
 	    i = 0;
@@ -801,10 +803,10 @@ static irqreturn_t mace_interrupt(int irq, void *dev_id)
     return IRQ_HANDLED;
 }
 
-static void mace_tx_timeout(unsigned long data)
+static void mace_tx_timeout(struct timer_list *t)
 {
-    struct net_device *dev = (struct net_device *) data;
-    struct mace_data *mp = netdev_priv(dev);
+    struct mace_data *mp = timer_container_of(mp, t, tx_timeout);
+    struct net_device *dev = macio_get_drvdata(mp->mdev);
     volatile struct mace __iomem *mb = mp->mace;
     volatile struct dbdma_regs __iomem *td = mp->tx_dma;
     volatile struct dbdma_regs __iomem *rd = mp->rx_dma;
@@ -842,7 +844,7 @@ static void mace_tx_timeout(unsigned long data)
     if (mp->tx_bad_runt) {
 	mp->tx_bad_runt = 0;
     } else if (i != mp->tx_fill) {
-	dev_kfree_skb(mp->tx_bufs[i]);
+	dev_kfree_skb_irq(mp->tx_bufs[i]);
 	if (++i >= N_TX_RING)
 	    i = 0;
 	mp->tx_empty = i;
diff --git a/drivers/net/ethernet/apple/mace.h b/drivers/net/ethernet/apple/mace.h
index 30b7ec0cedb5..697f71cf1937 100644
--- a/drivers/net/ethernet/apple/mace.h
+++ b/drivers/net/ethernet/apple/mace.h
@@ -1,13 +1,9 @@
+/* SPDX-License-Identifier: GPL-2.0-or-later */
 /*
  * mace.h - definitions for the registers in the Am79C940 MACE
  * (Medium Access Control for Ethernet) controller.
  *
  * Copyright (C) 1996 Paul Mackerras.
- *
- * This program is free software; you can redistribute it and/or
- * modify it under the terms of the GNU General Public License
- * as published by the Free Software Foundation; either version
- * 2 of the License, or (at your option) any later version.
  */
 
 #define REG(x)	volatile unsigned char x; char x ## _pad[15]
diff --git a/drivers/net/ethernet/apple/macmace.c b/drivers/net/ethernet/apple/macmace.c
index 857df9c45f04..8989506e6248 100644
--- a/drivers/net/ethernet/apple/macmace.c
+++ b/drivers/net/ethernet/apple/macmace.c
@@ -1,13 +1,9 @@
+// SPDX-License-Identifier: GPL-2.0-or-later
 /*
  *	Driver for the Macintosh 68K onboard MACE controller with PSC
  *	driven DMA. The MACE driver code is derived from mace.c. The
  *	Mac68k theory of operation is courtesy of the MacBSD wizards.
  *
- *	This program is free software; you can redistribute it and/or
- *	modify it under the terms of the GNU General Public License
- *	as published by the Free Software Foundation; either version
- *	2 of the License, or (at your option) any later version.
- *
  *	Copyright (C) 1996 Paul Mackerras.
  *	Copyright (C) 1998 Alan Cox <alan@lxorguk.ukuu.org.uk>
  *
@@ -81,7 +77,7 @@ struct mace_frame {
 	u8	pad4;
 	u32	pad5;
 	u32	pad6;
-	u8	data[1];
+	DECLARE_FLEX_ARRAY(u8, data);
 	/* And frame continues.. */
 };
 
@@ -89,14 +85,14 @@ struct mace_frame {
 
 static int mace_open(struct net_device *dev);
 static int mace_close(struct net_device *dev);
-static int mace_xmit_start(struct sk_buff *skb, struct net_device *dev);
+static netdev_tx_t mace_xmit_start(struct sk_buff *skb, struct net_device *dev);
 static void mace_set_multicast(struct net_device *dev);
 static int mace_set_address(struct net_device *dev, void *addr);
 static void mace_reset(struct net_device *dev);
 static irqreturn_t mace_interrupt(int irq, void *dev_id);
 static irqreturn_t mace_dma_intr(int irq, void *dev_id);
-static void mace_tx_timeout(struct net_device *dev);
-static void __mace_set_address(struct net_device *dev, void *addr);
+static void mace_tx_timeout(struct net_device *dev, unsigned int txqueue);
+static void __mace_set_address(struct net_device *dev, const void *addr);
 
 /*
  * Load a receive DMA channel with a base address and ring length
@@ -201,6 +197,7 @@ static int mace_probe(struct platform_device *pdev)
 	unsigned char *addr;
 	struct net_device *dev;
 	unsigned char checksum = 0;
+	u8 macaddr[ETH_ALEN];
 	int err;
 
 	dev = alloc_etherdev(PRIV_BYTES);
@@ -233,8 +230,9 @@ static int mace_probe(struct platform_device *pdev)
 	for (j = 0; j < 6; ++j) {
 		u8 v = bitrev8(addr[j<<4]);
 		checksum ^= v;
-		dev->dev_addr[j] = v;
+		macaddr[j] = v;
 	}
+	eth_hw_addr_set(dev, macaddr);
 	for (; j < 8; ++j) {
 		checksum ^= bitrev8(addr[j<<4]);
 	}
@@ -247,8 +245,8 @@ static int mace_probe(struct platform_device *pdev)
 	dev->netdev_ops		= &mace_netdev_ops;
 	dev->watchdog_timeo	= TX_TIMEOUT;
 
-	printk(KERN_INFO "%s: 68K MACE, hardware address %pM\n",
-	       dev->name, dev->dev_addr);
+	pr_info("Onboard MACE, hardware address %pM, chip revision 0x%04X\n",
+		dev->dev_addr, mp->chipid);
 
 	err = register_netdev(dev);
 	if (!err)
@@ -319,11 +317,12 @@ static void mace_reset(struct net_device *dev)
  * Load the address on a mace controller.
  */
 
-static void __mace_set_address(struct net_device *dev, void *addr)
+static void __mace_set_address(struct net_device *dev, const void *addr)
 {
 	struct mace_data *mp = netdev_priv(dev);
 	volatile struct mace *mb = mp->mace;
-	unsigned char *p = addr;
+	const unsigned char *p = addr;
+	u8 macaddr[ETH_ALEN];
 	int i;
 
 	/* load up the hardware address */
@@ -335,7 +334,8 @@ static void __mace_set_address(struct net_device *dev, void *addr)
 			;
 	}
 	for (i = 0; i < 6; ++i)
-		mb->padr = dev->dev_addr[i] = p[i];
+		mb->padr = macaddr[i] = p[i];
+	eth_hw_addr_set(dev, macaddr);
 	if (mp->chipid != BROKEN_ADDRCHG_REV)
 		mb->iac = 0;
 }
@@ -444,7 +444,7 @@ static int mace_close(struct net_device *dev)
  * Transmit a frame
  */
 
-static int mace_xmit_start(struct sk_buff *skb, struct net_device *dev)
+static netdev_tx_t mace_xmit_start(struct sk_buff *skb, struct net_device *dev)
 {
 	struct mace_data *mp = netdev_priv(dev);
 	unsigned long flags;
@@ -589,7 +589,6 @@ static irqreturn_t mace_interrupt(int irq, void *dev_id)
 			else if (fs & (UFLO|LCOL|RTRY)) {
 				++dev->stats.tx_aborted_errors;
 				if (mb->xmtfs & UFLO) {
-					printk(KERN_ERR "%s: DMA underrun.\n", dev->name);
 					dev->stats.tx_fifo_errors++;
 					mace_txdma_reset(dev);
 				}
@@ -605,7 +604,7 @@ static irqreturn_t mace_interrupt(int irq, void *dev_id)
 	return IRQ_HANDLED;
 }
 
-static void mace_tx_timeout(struct net_device *dev)
+static void mace_tx_timeout(struct net_device *dev, unsigned int txqueue)
 {
 	struct mace_data *mp = netdev_priv(dev);
 	volatile struct mace *mb = mp->mace;
@@ -644,10 +643,8 @@ static void mace_dma_rx_frame(struct net_device *dev, struct mace_frame *mf)
 
 	if (frame_status & (RS_OFLO | RS_CLSN | RS_FRAMERR | RS_FCSERR)) {
 		dev->stats.rx_errors++;
-		if (frame_status & RS_OFLO) {
-			printk(KERN_DEBUG "%s: fifo overflow.\n", dev->name);
+		if (frame_status & RS_OFLO)
 			dev->stats.rx_fifo_errors++;
-		}
 		if (frame_status & RS_CLSN)
 			dev->stats.collisions++;
 		if (frame_status & RS_FRAMERR)
@@ -663,7 +660,7 @@ static void mace_dma_rx_frame(struct net_device *dev, struct mace_frame *mf)
 			return;
 		}
 		skb_reserve(skb, 2);
-		memcpy(skb_put(skb, frame_length), mf->data, frame_length);
+		skb_put_data(skb, mf->data, frame_length);
 
 		skb->protocol = eth_type_trans(skb, dev);
 		netif_rx(skb);
@@ -742,7 +739,7 @@ MODULE_LICENSE("GPL");
 MODULE_DESCRIPTION("Macintosh MACE ethernet driver");
 MODULE_ALIAS("platform:macmace");
 
-static int mac_mace_device_remove(struct platform_device *pdev)
+static void mac_mace_device_remove(struct platform_device *pdev)
 {
 	struct net_device *dev = platform_get_drvdata(pdev);
 	struct mace_data *mp = netdev_priv(dev);
@@ -758,8 +755,6 @@ static int mac_mace_device_remove(struct platform_device *pdev)
 	                  mp->tx_ring, mp->tx_ring_phys);
 
 	free_netdev(dev);
-
-	return 0;
 }
 
 static struct platform_driver mac_mace_driver = {
@@ -770,18 +765,4 @@ static struct platform_driver mac_mace_driver = {
 	},
 };
 
-static int __init mac_mace_init_module(void)
-{
-	if (!MACH_IS_MAC)
-		return -ENODEV;
-
-	return platform_driver_register(&mac_mace_driver);
-}
-
-static void __exit mac_mace_cleanup_module(void)
-{
-	platform_driver_unregister(&mac_mace_driver);
-}
-
-module_init(mac_mace_init_module);
-module_exit(mac_mace_cleanup_module);
+module_platform_driver(mac_mace_driver);
diff --git a/drivers/net/ethernet/aquantia/Kconfig b/drivers/net/ethernet/aquantia/Kconfig
index cdf78e069a39..cec2018c84a9 100644
--- a/drivers/net/ethernet/aquantia/Kconfig
+++ b/drivers/net/ethernet/aquantia/Kconfig
@@ -1,3 +1,4 @@
+# SPDX-License-Identifier: GPL-2.0-only
 #
 # aQuantia device configuration
 #
@@ -5,11 +6,11 @@
 config NET_VENDOR_AQUANTIA
 	bool "aQuantia devices"
 	default y
-	---help---
+	help
 	  Set this to y if you have an Ethernet network cards that uses the aQuantia
 	  AQC107/AQC108 chipset.
 
-	  This option does not build any drivers; it casues the aQuantia
+	  This option does not build any drivers; it causes the aQuantia
 	  drivers that can be built to appear in the list of Ethernet drivers.
 
 
@@ -17,8 +18,9 @@ if NET_VENDOR_AQUANTIA
 
 config AQTION
 	tristate "aQuantia AQtion(tm) Support"
-	depends on PCI && X86_64
-	---help---
+	depends on PCI
+	depends on MACSEC || MACSEC=n
+	help
 	  This enables the support for the aQuantia AQtion(tm) Ethernet card.
 
 endif # NET_VENDOR_AQUANTIA
diff --git a/drivers/net/ethernet/aquantia/Makefile b/drivers/net/ethernet/aquantia/Makefile
index 4f4897b689b2..c4e7d01ea650 100644
--- a/drivers/net/ethernet/aquantia/Makefile
+++ b/drivers/net/ethernet/aquantia/Makefile
@@ -1,3 +1,4 @@
+# SPDX-License-Identifier: GPL-2.0-only
 #
 # Makefile for the aQuantia device drivers.
 #
diff --git a/drivers/net/ethernet/aquantia/atlantic/Makefile b/drivers/net/ethernet/aquantia/atlantic/Makefile
index e4ae696920ef..f6a96931c89a 100644
--- a/drivers/net/ethernet/aquantia/atlantic/Makefile
+++ b/drivers/net/ethernet/aquantia/atlantic/Makefile
@@ -1,34 +1,15 @@
+# SPDX-License-Identifier: GPL-2.0-only
 ################################################################################
 #
 # aQuantia Ethernet Controller AQtion Linux Driver
 # Copyright(c) 2014-2017 aQuantia Corporation.
 #
-# This program is free software; you can redistribute it and/or modify it
-# under the terms and conditions of the GNU General Public License,
-# version 2, as published by the Free Software Foundation.
-#
-# This program is distributed in the hope it will be useful, but WITHOUT
-# ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
-# FITNESS FOR A PARTICULAR PURPOSE.  See the GNU General Public License for
-# more details.
-#
-# You should have received a copy of the GNU General Public License along
-# with this program. If not, see <http://www.gnu.org/licenses/>.
-#
-# The full GNU General Public License is included in this distribution in
-# the file called "COPYING".
-#
-# Contact Information: <rdc-drv@aquantia.com>
-# aQuantia Corporation, 105 E. Tasman Dr. San Jose, CA 95134, USA
-#
 ################################################################################
 
-#
-# Makefile for the AQtion(tm) Ethernet driver
-#
-
 obj-$(CONFIG_AQTION) += atlantic.o
 
+ccflags-y += -I$(src)
+
 atlantic-objs := aq_main.o \
 	aq_nic.o \
 	aq_pci_func.o \
@@ -36,7 +17,20 @@ atlantic-objs := aq_main.o \
 	aq_ring.o \
 	aq_hw_utils.o \
 	aq_ethtool.o \
+	aq_drvinfo.o \
+	aq_filters.o \
+	aq_phy.o \
 	hw_atl/hw_atl_a0.o \
 	hw_atl/hw_atl_b0.o \
 	hw_atl/hw_atl_utils.o \
-	hw_atl/hw_atl_llh.o
+	hw_atl/hw_atl_utils_fw2x.o \
+	hw_atl/hw_atl_llh.o \
+	hw_atl2/hw_atl2.o \
+	hw_atl2/hw_atl2_utils.o \
+	hw_atl2/hw_atl2_utils_fw.o \
+	hw_atl2/hw_atl2_llh.o \
+	macsec/macsec_api.o
+
+atlantic-$(CONFIG_MACSEC) += aq_macsec.o
+
+atlantic-$(CONFIG_PTP_1588_CLOCK) += aq_ptp.o
diff --git a/drivers/net/ethernet/aquantia/atlantic/aq_cfg.h b/drivers/net/ethernet/aquantia/atlantic/aq_cfg.h
index 214986436ece..fc2b325f34e7 100644
--- a/drivers/net/ethernet/aquantia/atlantic/aq_cfg.h
+++ b/drivers/net/ethernet/aquantia/atlantic/aq_cfg.h
@@ -1,10 +1,7 @@
+/* SPDX-License-Identifier: GPL-2.0-only */
 /*
  * aQuantia Corporation Network Driver
- * Copyright (C) 2014-2017 aQuantia Corporation. All rights reserved
- *
- * This program is free software; you can redistribute it and/or modify it
- * under the terms and conditions of the GNU General Public License,
- * version 2, as published by the Free Software Foundation.
+ * Copyright (C) 2014-2019 aQuantia Corporation. All rights reserved
  */
 
 /* File aq_cfg.h: Definition of configuration parameters and constants. */
@@ -12,32 +9,45 @@
 #ifndef AQ_CFG_H
 #define AQ_CFG_H
 
-#define AQ_CFG_VECS_DEF   4U
+#define AQ_CFG_VECS_DEF   8U
 #define AQ_CFG_TCS_DEF    1U
 
 #define AQ_CFG_TXDS_DEF    4096U
-#define AQ_CFG_RXDS_DEF    1024U
+#define AQ_CFG_RXDS_DEF    2048U
 
 #define AQ_CFG_IS_POLLING_DEF 0U
 
-#define AQ_CFG_FORCE_LEGACY_INT 0U
+#define AQ_CFG_FORCE_INTX	0U
 
-#define AQ_CFG_IS_INTERRUPT_MODERATION_DEF   1U
-#define AQ_CFG_INTERRUPT_MODERATION_RATE_DEF 0xFFFFU
-#define AQ_CFG_IRQ_MASK                      0x1FFU
+#define AQ_CFG_INTERRUPT_MODERATION_OFF		0
+#define AQ_CFG_INTERRUPT_MODERATION_ON		1
+#define AQ_CFG_INTERRUPT_MODERATION_AUTO	0xFFFFU
+
+#define AQ_CFG_INTERRUPT_MODERATION_USEC_MAX (0x1FF * 2)
+
+#define AQ_CFG_IRQ_MASK                      0x3FFU
 
 #define AQ_CFG_VECS_MAX   8U
 #define AQ_CFG_TCS_MAX    8U
 
 #define AQ_CFG_TX_FRAME_MAX  (16U * 1024U)
-#define AQ_CFG_RX_FRAME_MAX  (4U * 1024U)
+#define AQ_CFG_RX_FRAME_MAX  (2U * 1024U)
+
+#define AQ_CFG_TX_CLEAN_BUDGET 256U
+
+#define AQ_CFG_RX_REFILL_THRES 32U
+
+#define AQ_CFG_RX_HDR_SIZE 256U
+
+#define AQ_CFG_RX_PAGEORDER 0U
+#define AQ_CFG_XDP_PAGEORDER 2U
 
 /* LRO */
 #define AQ_CFG_IS_LRO_DEF           1U
 
 /* RSS */
-#define AQ_CFG_RSS_INDIRECTION_TABLE_MAX  128U
-#define AQ_CFG_RSS_HASHKEY_SIZE           320U
+#define AQ_CFG_RSS_INDIRECTION_TABLE_MAX  64U
+#define AQ_CFG_RSS_HASHKEY_SIZE           40U
 
 #define AQ_CFG_IS_RSS_DEF           1U
 #define AQ_CFG_NUM_RSS_QUEUES_DEF   AQ_CFG_VECS_DEF
@@ -46,18 +56,21 @@
 #define AQ_CFG_PCI_FUNC_MSIX_IRQS   9U
 #define AQ_CFG_PCI_FUNC_PORTS       2U
 
-#define AQ_CFG_SERVICE_TIMER_INTERVAL    (2 * HZ)
+#define AQ_CFG_SERVICE_TIMER_INTERVAL    (1 * HZ)
 #define AQ_CFG_POLLING_TIMER_INTERVAL   ((unsigned int)(2 * HZ))
 
 #define AQ_CFG_SKB_FRAGS_MAX   32U
 
-#define AQ_CFG_NAPI_WEIGHT     64U
-
-#define AQ_CFG_MULTICAST_ADDRESS_MAX     32U
+/* Number of descriptors available in one ring to resume this ring queue
+ */
+#define AQ_CFG_RESTART_DESC_THRES   (AQ_CFG_SKB_FRAGS_MAX * 2)
 
 /*#define AQ_CFG_MAC_ADDR_PERMANENT {0x30, 0x0E, 0xE3, 0x12, 0x34, 0x56}*/
 
-#define AQ_CFG_FC_MODE 3U
+#define AQ_CFG_FC_MODE AQ_NIC_FC_FULL
+
+/* Default WOL modes used on initialization */
+#define AQ_CFG_WOL_MODES WAKE_MAGIC
 
 #define AQ_CFG_SPEED_MSK  0xFFFFU	/* 0xFFFFU==auto_neg */
 
@@ -66,12 +79,8 @@
 
 #define AQ_CFG_LOCK_TRYS   100U
 
-#define AQ_CFG_DRV_AUTHOR      "aQuantia"
-#define AQ_CFG_DRV_DESC        "aQuantia Corporation(R) Network Driver"
+#define AQ_CFG_DRV_AUTHOR      "Marvell"
+#define AQ_CFG_DRV_DESC        "Marvell (Aquantia) Corporation(R) Network Driver"
 #define AQ_CFG_DRV_NAME        "atlantic"
-#define AQ_CFG_DRV_VERSION	__stringify(NIC_MAJOR_DRIVER_VERSION)"."\
-				__stringify(NIC_MINOR_DRIVER_VERSION)"."\
-				__stringify(NIC_BUILD_DRIVER_VERSION)"."\
-				__stringify(NIC_REVISION_DRIVER_VERSION)
 
 #endif /* AQ_CFG_H */
diff --git a/drivers/net/ethernet/aquantia/atlantic/aq_common.h b/drivers/net/ethernet/aquantia/atlantic/aq_common.h
index 9eb5e222a234..ace691d7cd75 100644
--- a/drivers/net/ethernet/aquantia/atlantic/aq_common.h
+++ b/drivers/net/ethernet/aquantia/atlantic/aq_common.h
@@ -1,10 +1,8 @@
-/*
- * aQuantia Corporation Network Driver
- * Copyright (C) 2014-2017 aQuantia Corporation. All rights reserved
+/* SPDX-License-Identifier: GPL-2.0-only */
+/* Atlantic Network Driver
  *
- * This program is free software; you can redistribute it and/or modify it
- * under the terms and conditions of the GNU General Public License,
- * version 2, as published by the Free Software Foundation.
+ * Copyright (C) 2014-2019 aQuantia Corporation
+ * Copyright (C) 2019-2020 Marvell International Ltd.
  */
 
 /* File aq_common.h: Basic includes for all files in project. */
@@ -14,10 +12,66 @@
 
 #include <linux/etherdevice.h>
 #include <linux/pci.h>
-
-#include "ver.h"
-#include "aq_nic.h"
+#include <linux/if_vlan.h>
 #include "aq_cfg.h"
 #include "aq_utils.h"
 
+#define PCI_VENDOR_ID_AQUANTIA  0x1D6A
+
+#define AQ_DEVICE_ID_0001	0x0001
+#define AQ_DEVICE_ID_D100	0xD100
+#define AQ_DEVICE_ID_D107	0xD107
+#define AQ_DEVICE_ID_D108	0xD108
+#define AQ_DEVICE_ID_D109	0xD109
+
+#define AQ_DEVICE_ID_AQC100	0x00B1
+#define AQ_DEVICE_ID_AQC107	0x07B1
+#define AQ_DEVICE_ID_AQC108	0x08B1
+#define AQ_DEVICE_ID_AQC109	0x09B1
+#define AQ_DEVICE_ID_AQC111	0x11B1
+#define AQ_DEVICE_ID_AQC112	0x12B1
+
+#define AQ_DEVICE_ID_AQC100S	0x80B1
+#define AQ_DEVICE_ID_AQC107S	0x87B1
+#define AQ_DEVICE_ID_AQC108S	0x88B1
+#define AQ_DEVICE_ID_AQC109S	0x89B1
+#define AQ_DEVICE_ID_AQC111S	0x91B1
+#define AQ_DEVICE_ID_AQC112S	0x92B1
+
+#define AQ_DEVICE_ID_AQC113DEV	0x00C0
+#define AQ_DEVICE_ID_AQC113CS	0x94C0
+#define AQ_DEVICE_ID_AQC113CA	0x34C0
+#define AQ_DEVICE_ID_AQC114CS	0x93C0
+#define AQ_DEVICE_ID_AQC113	0x04C0
+#define AQ_DEVICE_ID_AQC113C	0x14C0
+#define AQ_DEVICE_ID_AQC115C	0x12C0
+#define AQ_DEVICE_ID_AQC116C	0x11C0
+
+#define HW_ATL_NIC_NAME "Marvell (aQuantia) AQtion 10Gbit Network Adapter"
+
+#define AQ_HWREV_ANY	0
+#define AQ_HWREV_1	1
+#define AQ_HWREV_2	2
+
+#define AQ_NIC_RATE_10G		BIT(0)
+#define AQ_NIC_RATE_5G		BIT(1)
+#define AQ_NIC_RATE_2G5		BIT(2)
+#define AQ_NIC_RATE_1G		BIT(3)
+#define AQ_NIC_RATE_100M	BIT(4)
+#define AQ_NIC_RATE_10M		BIT(5)
+#define AQ_NIC_RATE_1G_HALF	BIT(6)
+#define AQ_NIC_RATE_100M_HALF	BIT(7)
+#define AQ_NIC_RATE_10M_HALF	BIT(8)
+
+#define AQ_NIC_RATE_EEE_10G	BIT(9)
+#define AQ_NIC_RATE_EEE_5G	BIT(10)
+#define AQ_NIC_RATE_EEE_2G5	BIT(11)
+#define AQ_NIC_RATE_EEE_1G	BIT(12)
+#define AQ_NIC_RATE_EEE_100M	BIT(13)
+#define AQ_NIC_RATE_EEE_MSK     (AQ_NIC_RATE_EEE_10G |\
+				 AQ_NIC_RATE_EEE_5G |\
+				 AQ_NIC_RATE_EEE_2G5 |\
+				 AQ_NIC_RATE_EEE_1G |\
+				 AQ_NIC_RATE_EEE_100M)
+
 #endif /* AQ_COMMON_H */
diff --git a/drivers/net/ethernet/aquantia/atlantic/aq_drvinfo.c b/drivers/net/ethernet/aquantia/atlantic/aq_drvinfo.c
new file mode 100644
index 000000000000..787ea91802e7
--- /dev/null
+++ b/drivers/net/ethernet/aquantia/atlantic/aq_drvinfo.c
@@ -0,0 +1,148 @@
+// SPDX-License-Identifier: GPL-2.0-only
+/* Atlantic Network Driver
+ *
+ * Copyright (C) 2014-2019 aQuantia Corporation
+ * Copyright (C) 2019-2020 Marvell International Ltd.
+ */
+
+/* File aq_drvinfo.c: Definition of common code for firmware info in sys.*/
+
+#include <linux/init.h>
+#include <linux/kobject.h>
+#include <linux/module.h>
+#include <linux/stat.h>
+#include <linux/string.h>
+#include <linux/hwmon.h>
+#include <linux/uaccess.h>
+
+#include "aq_drvinfo.h"
+#include "aq_nic.h"
+
+#if IS_REACHABLE(CONFIG_HWMON)
+static const char * const atl_temp_label[] = {
+	"PHY Temperature",
+	"MAC Temperature",
+};
+
+static int aq_hwmon_read(struct device *dev, enum hwmon_sensor_types type,
+			 u32 attr, int channel, long *value)
+{
+	struct aq_nic_s *aq_nic = dev_get_drvdata(dev);
+	int err = 0;
+	int temp;
+
+	if (!aq_nic)
+		return -EIO;
+
+	if (type != hwmon_temp || attr != hwmon_temp_input)
+		return -EOPNOTSUPP;
+
+	switch (channel) {
+	case 0:
+		if (!aq_nic->aq_fw_ops->get_phy_temp)
+			return -EOPNOTSUPP;
+
+		err = aq_nic->aq_fw_ops->get_phy_temp(aq_nic->aq_hw, &temp);
+		*value = temp;
+		break;
+	case 1:
+		if (!aq_nic->aq_fw_ops->get_mac_temp &&
+		    !aq_nic->aq_hw_ops->hw_get_mac_temp)
+			return -EOPNOTSUPP;
+
+		if (aq_nic->aq_fw_ops->get_mac_temp)
+			err = aq_nic->aq_fw_ops->get_mac_temp(aq_nic->aq_hw, &temp);
+		else
+			err = aq_nic->aq_hw_ops->hw_get_mac_temp(aq_nic->aq_hw, &temp);
+		*value = temp;
+		break;
+	default:
+		return -EOPNOTSUPP;
+	}
+
+	return err;
+}
+
+static int aq_hwmon_read_string(struct device *dev,
+				enum hwmon_sensor_types type,
+				u32 attr, int channel, const char **str)
+{
+	struct aq_nic_s *aq_nic = dev_get_drvdata(dev);
+
+	if (!aq_nic)
+		return -EIO;
+
+	if (type != hwmon_temp || attr != hwmon_temp_label)
+		return -EOPNOTSUPP;
+
+	if (channel < ARRAY_SIZE(atl_temp_label))
+		*str = atl_temp_label[channel];
+	else
+		return -EOPNOTSUPP;
+
+	return 0;
+}
+
+static umode_t aq_hwmon_is_visible(const void *data,
+				   enum hwmon_sensor_types type,
+				   u32 attr, int channel)
+{
+	const struct aq_nic_s *nic = data;
+
+	if (type != hwmon_temp)
+		return 0;
+
+	if (channel == 0 && !nic->aq_fw_ops->get_phy_temp)
+		return 0;
+	else if (channel == 1 && !nic->aq_fw_ops->get_mac_temp &&
+		 !nic->aq_hw_ops->hw_get_mac_temp)
+		return 0;
+
+	switch (attr) {
+	case hwmon_temp_input:
+	case hwmon_temp_label:
+		return 0444;
+	default:
+		return 0;
+	}
+}
+
+static const struct hwmon_ops aq_hwmon_ops = {
+	.is_visible = aq_hwmon_is_visible,
+	.read = aq_hwmon_read,
+	.read_string = aq_hwmon_read_string,
+};
+
+static const struct hwmon_channel_info * const aq_hwmon_info[] = {
+	HWMON_CHANNEL_INFO(temp, HWMON_T_INPUT | HWMON_T_LABEL,
+			   HWMON_T_INPUT | HWMON_T_LABEL),
+	NULL,
+};
+
+static const struct hwmon_chip_info aq_hwmon_chip_info = {
+	.ops = &aq_hwmon_ops,
+	.info = aq_hwmon_info,
+};
+
+int aq_drvinfo_init(struct net_device *ndev)
+{
+	struct aq_nic_s *aq_nic = netdev_priv(ndev);
+	struct device *dev = &aq_nic->pdev->dev;
+	struct device *hwmon_dev;
+	int err = 0;
+
+	hwmon_dev = devm_hwmon_device_register_with_info(dev,
+							 ndev->name,
+							 aq_nic,
+							 &aq_hwmon_chip_info,
+							 NULL);
+
+	if (IS_ERR(hwmon_dev))
+		err = PTR_ERR(hwmon_dev);
+
+	return err;
+}
+
+#else
+int aq_drvinfo_init(struct net_device *ndev) { return 0; }
+#endif
diff --git a/drivers/net/ethernet/aquantia/atlantic/aq_drvinfo.h b/drivers/net/ethernet/aquantia/atlantic/aq_drvinfo.h
new file mode 100644
index 000000000000..59113a20622a
--- /dev/null
+++ b/drivers/net/ethernet/aquantia/atlantic/aq_drvinfo.h
@@ -0,0 +1,17 @@
+/* SPDX-License-Identifier: GPL-2.0-only */
+/* Atlantic Network Driver
+ *
+ * Copyright (C) 2014-2019 aQuantia Corporation
+ * Copyright (C) 2019-2020 Marvell International Ltd.
+ */
+
+/* File aq_drvinfo.h: Declaration of common code for firmware info in sys.*/
+
+#ifndef AQ_DRVINFO_H
+#define AQ_DRVINFO_H
+
+struct net_device;
+
+int aq_drvinfo_init(struct net_device *ndev);
+
+#endif /* AQ_DRVINFO_H */
diff --git a/drivers/net/ethernet/aquantia/atlantic/aq_ethtool.c b/drivers/net/ethernet/aquantia/atlantic/aq_ethtool.c
index a761e91471df..6fef47ba0a59 100644
--- a/drivers/net/ethernet/aquantia/atlantic/aq_ethtool.c
+++ b/drivers/net/ethernet/aquantia/atlantic/aq_ethtool.c
@@ -1,22 +1,31 @@
-/*
- * aQuantia Corporation Network Driver
- * Copyright (C) 2014-2017 aQuantia Corporation. All rights reserved
+// SPDX-License-Identifier: GPL-2.0-only
+/* Atlantic Network Driver
  *
- * This program is free software; you can redistribute it and/or modify it
- * under the terms and conditions of the GNU General Public License,
- * version 2, as published by the Free Software Foundation.
+ * Copyright (C) 2014-2019 aQuantia Corporation
+ * Copyright (C) 2019-2020 Marvell International Ltd.
  */
 
 /* File aq_ethtool.c: Definition of ethertool related functions. */
 
 #include "aq_ethtool.h"
 #include "aq_nic.h"
+#include "aq_vec.h"
+#include "aq_ptp.h"
+#include "aq_filters.h"
+#include "aq_macsec.h"
+#include "aq_main.h"
+
+#include <linux/ethtool.h>
+#include <linux/linkmode.h>
+#include <linux/ptp_clock_kernel.h>
 
 static void aq_ethtool_get_regs(struct net_device *ndev,
 				struct ethtool_regs *regs, void *p)
 {
 	struct aq_nic_s *aq_nic = netdev_priv(ndev);
-	u32 regs_count = aq_nic_get_regs_count(aq_nic);
+	u32 regs_count;
+
+	regs_count = aq_nic_get_regs_count(aq_nic);
 
 	memset(p, 0, regs_count * sizeof(u32));
 	aq_nic_get_regs(aq_nic, regs, p);
@@ -25,7 +34,9 @@ static void aq_ethtool_get_regs(struct net_device *ndev,
 static int aq_ethtool_get_regs_len(struct net_device *ndev)
 {
 	struct aq_nic_s *aq_nic = netdev_priv(ndev);
-	u32 regs_count = aq_nic_get_regs_count(aq_nic);
+	u32 regs_count;
+
+	regs_count = aq_nic_get_regs_count(aq_nic);
 
 	return regs_count * sizeof(u32);
 }
@@ -56,10 +67,6 @@ aq_ethtool_set_link_ksettings(struct net_device *ndev,
 	return aq_nic_set_link_ksettings(aq_nic, cmd);
 }
 
-/* there "5U" is number of queue[#] stats lines (InPackets+...+InErrors) */
-static const unsigned int aq_ethtool_stat_queue_lines = 5U;
-static const unsigned int aq_ethtool_stat_queue_chars =
-	5U * ETH_GSTRING_LEN;
 static const char aq_ethtool_stat_names[][ETH_GSTRING_LEN] = {
 	"InPackets",
 	"InUCast",
@@ -70,92 +77,173 @@ static const char aq_ethtool_stat_names[][ETH_GSTRING_LEN] = {
 	"OutUCast",
 	"OutMCast",
 	"OutBCast",
-	"InUCastOctects",
-	"OutUCastOctects",
-	"InMCastOctects",
-	"OutMCastOctects",
-	"InBCastOctects",
-	"OutBCastOctects",
-	"InOctects",
-	"OutOctects",
+	"InUCastOctets",
+	"OutUCastOctets",
+	"InMCastOctets",
+	"OutMCastOctets",
+	"InBCastOctets",
+	"OutBCastOctets",
+	"InOctets",
+	"OutOctets",
 	"InPacketsDma",
 	"OutPacketsDma",
 	"InOctetsDma",
 	"OutOctetsDma",
 	"InDroppedDma",
-	"Queue[0] InPackets",
-	"Queue[0] OutPackets",
-	"Queue[0] InJumboPackets",
-	"Queue[0] InLroPackets",
-	"Queue[0] InErrors",
-	"Queue[1] InPackets",
-	"Queue[1] OutPackets",
-	"Queue[1] InJumboPackets",
-	"Queue[1] InLroPackets",
-	"Queue[1] InErrors",
-	"Queue[2] InPackets",
-	"Queue[2] OutPackets",
-	"Queue[2] InJumboPackets",
-	"Queue[2] InLroPackets",
-	"Queue[2] InErrors",
-	"Queue[3] InPackets",
-	"Queue[3] OutPackets",
-	"Queue[3] InJumboPackets",
-	"Queue[3] InLroPackets",
-	"Queue[3] InErrors",
-	"Queue[4] InPackets",
-	"Queue[4] OutPackets",
-	"Queue[4] InJumboPackets",
-	"Queue[4] InLroPackets",
-	"Queue[4] InErrors",
-	"Queue[5] InPackets",
-	"Queue[5] OutPackets",
-	"Queue[5] InJumboPackets",
-	"Queue[5] InLroPackets",
-	"Queue[5] InErrors",
-	"Queue[6] InPackets",
-	"Queue[6] OutPackets",
-	"Queue[6] InJumboPackets",
-	"Queue[6] InLroPackets",
-	"Queue[6] InErrors",
-	"Queue[7] InPackets",
-	"Queue[7] OutPackets",
-	"Queue[7] InJumboPackets",
-	"Queue[7] InLroPackets",
-	"Queue[7] InErrors",
 };
 
+static const char * const aq_ethtool_queue_rx_stat_names[] = {
+	"%sQueue[%d] InPackets",
+	"%sQueue[%d] InJumboPackets",
+	"%sQueue[%d] InLroPackets",
+	"%sQueue[%d] InErrors",
+	"%sQueue[%d] AllocFails",
+	"%sQueue[%d] SkbAllocFails",
+	"%sQueue[%d] Polls",
+	"%sQueue[%d] PageFlips",
+	"%sQueue[%d] PageReuses",
+	"%sQueue[%d] PageFrees",
+	"%sQueue[%d] XdpAbort",
+	"%sQueue[%d] XdpDrop",
+	"%sQueue[%d] XdpPass",
+	"%sQueue[%d] XdpTx",
+	"%sQueue[%d] XdpInvalid",
+	"%sQueue[%d] XdpRedirect",
+};
+
+static const char * const aq_ethtool_queue_tx_stat_names[] = {
+	"%sQueue[%d] OutPackets",
+	"%sQueue[%d] Restarts",
+};
+
+#if IS_ENABLED(CONFIG_MACSEC)
+static const char aq_macsec_stat_names[][ETH_GSTRING_LEN] = {
+	"MACSec InCtlPackets",
+	"MACSec InTaggedMissPackets",
+	"MACSec InUntaggedMissPackets",
+	"MACSec InNotagPackets",
+	"MACSec InUntaggedPackets",
+	"MACSec InBadTagPackets",
+	"MACSec InNoSciPackets",
+	"MACSec InUnknownSciPackets",
+	"MACSec InCtrlPortPassPackets",
+	"MACSec InUnctrlPortPassPackets",
+	"MACSec InCtrlPortFailPackets",
+	"MACSec InUnctrlPortFailPackets",
+	"MACSec InTooLongPackets",
+	"MACSec InIgpocCtlPackets",
+	"MACSec InEccErrorPackets",
+	"MACSec InUnctrlHitDropRedir",
+	"MACSec OutCtlPackets",
+	"MACSec OutUnknownSaPackets",
+	"MACSec OutUntaggedPackets",
+	"MACSec OutTooLong",
+	"MACSec OutEccErrorPackets",
+	"MACSec OutUnctrlHitDropRedir",
+};
+
+static const char * const aq_macsec_txsc_stat_names[] = {
+	"MACSecTXSC%d ProtectedPkts",
+	"MACSecTXSC%d EncryptedPkts",
+	"MACSecTXSC%d ProtectedOctets",
+	"MACSecTXSC%d EncryptedOctets",
+};
+
+static const char * const aq_macsec_txsa_stat_names[] = {
+	"MACSecTXSC%dSA%d HitDropRedirect",
+	"MACSecTXSC%dSA%d Protected2Pkts",
+	"MACSecTXSC%dSA%d ProtectedPkts",
+	"MACSecTXSC%dSA%d EncryptedPkts",
+};
+
+static const char * const aq_macsec_rxsa_stat_names[] = {
+	"MACSecRXSC%dSA%d UntaggedHitPkts",
+	"MACSecRXSC%dSA%d CtrlHitDrpRedir",
+	"MACSecRXSC%dSA%d NotUsingSa",
+	"MACSecRXSC%dSA%d UnusedSa",
+	"MACSecRXSC%dSA%d NotValidPkts",
+	"MACSecRXSC%dSA%d InvalidPkts",
+	"MACSecRXSC%dSA%d OkPkts",
+	"MACSecRXSC%dSA%d LatePkts",
+	"MACSecRXSC%dSA%d DelayedPkts",
+	"MACSecRXSC%dSA%d UncheckedPkts",
+	"MACSecRXSC%dSA%d ValidatedOctets",
+	"MACSecRXSC%dSA%d DecryptedOctets",
+};
+#endif
+
+static const char aq_ethtool_priv_flag_names[][ETH_GSTRING_LEN] = {
+	"DMASystemLoopback",
+	"PKTSystemLoopback",
+	"DMANetworkLoopback",
+	"PHYInternalLoopback",
+	"PHYExternalLoopback",
+};
+
+static u32 aq_ethtool_n_stats(struct net_device *ndev)
+{
+	const int rx_stat_cnt = ARRAY_SIZE(aq_ethtool_queue_rx_stat_names);
+	const int tx_stat_cnt = ARRAY_SIZE(aq_ethtool_queue_tx_stat_names);
+	struct aq_nic_s *nic = netdev_priv(ndev);
+	struct aq_nic_cfg_s *cfg = aq_nic_get_cfg(nic);
+	u32 n_stats = ARRAY_SIZE(aq_ethtool_stat_names) +
+		      (rx_stat_cnt + tx_stat_cnt) * cfg->vecs * cfg->tcs;
+
+#if IS_REACHABLE(CONFIG_PTP_1588_CLOCK)
+	n_stats += rx_stat_cnt * aq_ptp_get_ring_cnt(nic, ATL_RING_RX) +
+		   tx_stat_cnt * aq_ptp_get_ring_cnt(nic, ATL_RING_TX);
+#endif
+
+#if IS_ENABLED(CONFIG_MACSEC)
+	if (nic->macsec_cfg) {
+		n_stats += ARRAY_SIZE(aq_macsec_stat_names) +
+			   ARRAY_SIZE(aq_macsec_txsc_stat_names) *
+				   aq_macsec_tx_sc_cnt(nic) +
+			   ARRAY_SIZE(aq_macsec_txsa_stat_names) *
+				   aq_macsec_tx_sa_cnt(nic) +
+			   ARRAY_SIZE(aq_macsec_rxsa_stat_names) *
+				   aq_macsec_rx_sa_cnt(nic);
+	}
+#endif
+
+	return n_stats;
+}
+
 static void aq_ethtool_stats(struct net_device *ndev,
 			     struct ethtool_stats *stats, u64 *data)
 {
 	struct aq_nic_s *aq_nic = netdev_priv(ndev);
 
-/* ASSERT: Need add lines to aq_ethtool_stat_names if AQ_CFG_VECS_MAX > 8 */
-	BUILD_BUG_ON(AQ_CFG_VECS_MAX > 8);
-	memset(data, 0, ARRAY_SIZE(aq_ethtool_stat_names) * sizeof(u64));
-	aq_nic_get_stats(aq_nic, data);
+	memset(data, 0, aq_ethtool_n_stats(ndev) * sizeof(u64));
+	data = aq_nic_get_stats(aq_nic, data);
+#if IS_REACHABLE(CONFIG_PTP_1588_CLOCK)
+	data = aq_ptp_get_stats(aq_nic, data);
+#endif
+#if IS_ENABLED(CONFIG_MACSEC)
+	data = aq_macsec_get_stats(aq_nic, data);
+#endif
 }
 
 static void aq_ethtool_get_drvinfo(struct net_device *ndev,
 				   struct ethtool_drvinfo *drvinfo)
 {
-	struct aq_nic_s *aq_nic = netdev_priv(ndev);
-	struct aq_nic_cfg_s *cfg = aq_nic_get_cfg(aq_nic);
 	struct pci_dev *pdev = to_pci_dev(ndev->dev.parent);
-	u32 firmware_version = aq_nic_get_fw_version(aq_nic);
-	u32 regs_count = aq_nic_get_regs_count(aq_nic);
+	struct aq_nic_s *aq_nic = netdev_priv(ndev);
+	u32 firmware_version;
+	u32 regs_count;
+
+	firmware_version = aq_nic_get_fw_version(aq_nic);
+	regs_count = aq_nic_get_regs_count(aq_nic);
 
 	strlcat(drvinfo->driver, AQ_CFG_DRV_NAME, sizeof(drvinfo->driver));
-	strlcat(drvinfo->version, AQ_CFG_DRV_VERSION, sizeof(drvinfo->version));
 
 	snprintf(drvinfo->fw_version, sizeof(drvinfo->fw_version),
 		 "%u.%u.%u", firmware_version >> 24,
 		 (firmware_version >> 16) & 0xFFU, firmware_version & 0xFFFFU);
 
-	strlcpy(drvinfo->bus_info, pdev ? pci_name(pdev) : "",
+	strscpy(drvinfo->bus_info, pdev ? pci_name(pdev) : "",
 		sizeof(drvinfo->bus_info));
-	drvinfo->n_stats = ARRAY_SIZE(aq_ethtool_stat_names) -
-		(AQ_CFG_VECS_MAX - cfg->vecs) * aq_ethtool_stat_queue_lines;
+	drvinfo->n_stats = aq_ethtool_n_stats(ndev);
 	drvinfo->testinfo_len = 0;
 	drvinfo->regdump_len = regs_count;
 	drvinfo->eedump_len = 0;
@@ -164,31 +252,178 @@ static void aq_ethtool_get_drvinfo(struct net_device *ndev,
 static void aq_ethtool_get_strings(struct net_device *ndev,
 				   u32 stringset, u8 *data)
 {
+	struct aq_nic_s *nic = netdev_priv(ndev);
+	struct aq_nic_cfg_s *cfg;
+	u8 *p = data;
+	int i, si;
+#if IS_ENABLED(CONFIG_MACSEC)
+	int sa;
+#endif
+
+	cfg = aq_nic_get_cfg(nic);
+
+	switch (stringset) {
+	case ETH_SS_STATS: {
+		const int rx_stat_cnt = ARRAY_SIZE(aq_ethtool_queue_rx_stat_names);
+		const int tx_stat_cnt = ARRAY_SIZE(aq_ethtool_queue_tx_stat_names);
+		char tc_string[8];
+		unsigned int tc;
+
+		memset(tc_string, 0, sizeof(tc_string));
+		memcpy(p, aq_ethtool_stat_names,
+		       sizeof(aq_ethtool_stat_names));
+		p = p + sizeof(aq_ethtool_stat_names);
+
+		for (tc = 0; tc < cfg->tcs; tc++) {
+			if (cfg->is_qos)
+				snprintf(tc_string, 8, "TC%u ", tc);
+
+			for (i = 0; i < cfg->vecs; i++) {
+				for (si = 0; si < rx_stat_cnt; si++) {
+					ethtool_sprintf(&p,
+					     aq_ethtool_queue_rx_stat_names[si],
+					     tc_string,
+					     AQ_NIC_CFG_TCVEC2RING(cfg, tc, i));
+				}
+				for (si = 0; si < tx_stat_cnt; si++) {
+					ethtool_sprintf(&p,
+					     aq_ethtool_queue_tx_stat_names[si],
+					     tc_string,
+					     AQ_NIC_CFG_TCVEC2RING(cfg, tc, i));
+				}
+			}
+		}
+#if IS_REACHABLE(CONFIG_PTP_1588_CLOCK)
+		if (nic->aq_ptp) {
+			const int rx_ring_cnt = aq_ptp_get_ring_cnt(nic, ATL_RING_RX);
+			const int tx_ring_cnt = aq_ptp_get_ring_cnt(nic, ATL_RING_TX);
+			unsigned int ptp_ring_idx =
+				aq_ptp_ring_idx(nic->aq_nic_cfg.tc_mode);
+
+			snprintf(tc_string, 8, "PTP ");
+
+			for (i = 0; i < max(rx_ring_cnt, tx_ring_cnt); i++) {
+				for (si = 0; si < rx_stat_cnt; si++) {
+					ethtool_sprintf(&p,
+						 aq_ethtool_queue_rx_stat_names[si],
+						 tc_string,
+						 i ? PTP_HWST_RING_IDX : ptp_ring_idx);
+				}
+				if (i >= tx_ring_cnt)
+					continue;
+				for (si = 0; si < tx_stat_cnt; si++) {
+					ethtool_sprintf(&p,
+						 aq_ethtool_queue_tx_stat_names[si],
+						 tc_string,
+						 i ? PTP_HWST_RING_IDX : ptp_ring_idx);
+				}
+			}
+		}
+#endif
+#if IS_ENABLED(CONFIG_MACSEC)
+		if (!nic->macsec_cfg)
+			break;
+
+		memcpy(p, aq_macsec_stat_names, sizeof(aq_macsec_stat_names));
+		p = p + sizeof(aq_macsec_stat_names);
+		for (i = 0; i < AQ_MACSEC_MAX_SC; i++) {
+			struct aq_macsec_txsc *aq_txsc;
+
+			if (!(test_bit(i, &nic->macsec_cfg->txsc_idx_busy)))
+				continue;
+
+			for (si = 0;
+				si < ARRAY_SIZE(aq_macsec_txsc_stat_names);
+				si++) {
+				ethtool_sprintf(&p,
+					 aq_macsec_txsc_stat_names[si], i);
+			}
+			aq_txsc = &nic->macsec_cfg->aq_txsc[i];
+			for (sa = 0; sa < MACSEC_NUM_AN; sa++) {
+				if (!(test_bit(sa, &aq_txsc->tx_sa_idx_busy)))
+					continue;
+				for (si = 0;
+				     si < ARRAY_SIZE(aq_macsec_txsa_stat_names);
+				     si++) {
+					ethtool_sprintf(&p,
+						 aq_macsec_txsa_stat_names[si],
+						 i, sa);
+				}
+			}
+		}
+		for (i = 0; i < AQ_MACSEC_MAX_SC; i++) {
+			struct aq_macsec_rxsc *aq_rxsc;
+
+			if (!(test_bit(i, &nic->macsec_cfg->rxsc_idx_busy)))
+				continue;
+
+			aq_rxsc = &nic->macsec_cfg->aq_rxsc[i];
+			for (sa = 0; sa < MACSEC_NUM_AN; sa++) {
+				if (!(test_bit(sa, &aq_rxsc->rx_sa_idx_busy)))
+					continue;
+				for (si = 0;
+				     si < ARRAY_SIZE(aq_macsec_rxsa_stat_names);
+				     si++) {
+					ethtool_sprintf(&p,
+						 aq_macsec_rxsa_stat_names[si],
+						 i, sa);
+				}
+			}
+		}
+#endif
+		break;
+	}
+	case ETH_SS_PRIV_FLAGS:
+		memcpy(p, aq_ethtool_priv_flag_names,
+		       sizeof(aq_ethtool_priv_flag_names));
+		break;
+	}
+}
+
+static int aq_ethtool_set_phys_id(struct net_device *ndev,
+				  enum ethtool_phys_id_state state)
+{
 	struct aq_nic_s *aq_nic = netdev_priv(ndev);
-	struct aq_nic_cfg_s *cfg = aq_nic_get_cfg(aq_nic);
+	struct aq_hw_s *hw = aq_nic->aq_hw;
+	int ret = 0;
+
+	if (!aq_nic->aq_fw_ops->led_control)
+		return -EOPNOTSUPP;
+
+	mutex_lock(&aq_nic->fwreq_mutex);
 
-	if (stringset == ETH_SS_STATS)
-		memcpy(data, *aq_ethtool_stat_names,
-		       sizeof(aq_ethtool_stat_names) -
-		       (AQ_CFG_VECS_MAX - cfg->vecs) *
-		       aq_ethtool_stat_queue_chars);
+	switch (state) {
+	case ETHTOOL_ID_ACTIVE:
+		ret = aq_nic->aq_fw_ops->led_control(hw, AQ_HW_LED_BLINK |
+				 AQ_HW_LED_BLINK << 2 | AQ_HW_LED_BLINK << 4);
+		break;
+	case ETHTOOL_ID_INACTIVE:
+		ret = aq_nic->aq_fw_ops->led_control(hw, AQ_HW_LED_DEFAULT);
+		break;
+	default:
+		break;
+	}
+
+	mutex_unlock(&aq_nic->fwreq_mutex);
+
+	return ret;
 }
 
 static int aq_ethtool_get_sset_count(struct net_device *ndev, int stringset)
 {
 	int ret = 0;
-	struct aq_nic_s *aq_nic = netdev_priv(ndev);
-	struct aq_nic_cfg_s *cfg = aq_nic_get_cfg(aq_nic);
 
 	switch (stringset) {
 	case ETH_SS_STATS:
-		ret = ARRAY_SIZE(aq_ethtool_stat_names) -
-			(AQ_CFG_VECS_MAX - cfg->vecs) *
-			aq_ethtool_stat_queue_lines;
+		ret = aq_ethtool_n_stats(ndev);
+		break;
+	case ETH_SS_PRIV_FLAGS:
+		ret = ARRAY_SIZE(aq_ethtool_priv_flag_names);
 		break;
 	default:
 		ret = -EOPNOTSUPP;
 	}
+
 	return ret;
 }
 
@@ -200,43 +435,115 @@ static u32 aq_ethtool_get_rss_indir_size(struct net_device *ndev)
 static u32 aq_ethtool_get_rss_key_size(struct net_device *ndev)
 {
 	struct aq_nic_s *aq_nic = netdev_priv(ndev);
-	struct aq_nic_cfg_s *cfg = aq_nic_get_cfg(aq_nic);
+	struct aq_nic_cfg_s *cfg;
+
+	cfg = aq_nic_get_cfg(aq_nic);
 
 	return sizeof(cfg->aq_rss.hash_secret_key);
 }
 
-static int aq_ethtool_get_rss(struct net_device *ndev, u32 *indir, u8 *key,
-			      u8 *hfunc)
+static int aq_ethtool_get_rss(struct net_device *ndev,
+			      struct ethtool_rxfh_param *rxfh)
 {
 	struct aq_nic_s *aq_nic = netdev_priv(ndev);
-	struct aq_nic_cfg_s *cfg = aq_nic_get_cfg(aq_nic);
+	struct aq_nic_cfg_s *cfg;
 	unsigned int i = 0U;
 
-	if (hfunc)
-		*hfunc = ETH_RSS_HASH_TOP; /* Toeplitz */
-	if (indir) {
+	cfg = aq_nic_get_cfg(aq_nic);
+
+	rxfh->hfunc = ETH_RSS_HASH_TOP; /* Toeplitz */
+	if (rxfh->indir) {
 		for (i = 0; i < AQ_CFG_RSS_INDIRECTION_TABLE_MAX; i++)
-			indir[i] = cfg->aq_rss.indirection_table[i];
+			rxfh->indir[i] = cfg->aq_rss.indirection_table[i];
 	}
-	if (key)
-		memcpy(key, cfg->aq_rss.hash_secret_key,
+	if (rxfh->key)
+		memcpy(rxfh->key, cfg->aq_rss.hash_secret_key,
 		       sizeof(cfg->aq_rss.hash_secret_key));
+
 	return 0;
 }
 
+static int aq_ethtool_set_rss(struct net_device *netdev,
+			      struct ethtool_rxfh_param *rxfh,
+			      struct netlink_ext_ack *extack)
+{
+	struct aq_nic_s *aq_nic = netdev_priv(netdev);
+	struct aq_nic_cfg_s *cfg;
+	unsigned int i = 0U;
+	u32 rss_entries;
+	int err = 0;
+
+	cfg = aq_nic_get_cfg(aq_nic);
+	rss_entries = cfg->aq_rss.indirection_table_size;
+
+	/* We do not allow change in unsupported parameters */
+	if (rxfh->hfunc != ETH_RSS_HASH_NO_CHANGE &&
+	    rxfh->hfunc != ETH_RSS_HASH_TOP)
+		return -EOPNOTSUPP;
+	/* Fill out the redirection table */
+	if (rxfh->indir)
+		for (i = 0; i < rss_entries; i++)
+			cfg->aq_rss.indirection_table[i] = rxfh->indir[i];
+
+	/* Fill out the rss hash key */
+	if (rxfh->key) {
+		memcpy(cfg->aq_rss.hash_secret_key, rxfh->key,
+		       sizeof(cfg->aq_rss.hash_secret_key));
+		err = aq_nic->aq_hw_ops->hw_rss_hash_set(aq_nic->aq_hw,
+			&cfg->aq_rss);
+		if (err)
+			return err;
+	}
+
+	err = aq_nic->aq_hw_ops->hw_rss_set(aq_nic->aq_hw, &cfg->aq_rss);
+
+	return err;
+}
+
 static int aq_ethtool_get_rxnfc(struct net_device *ndev,
 				struct ethtool_rxnfc *cmd,
 				u32 *rule_locs)
 {
 	struct aq_nic_s *aq_nic = netdev_priv(ndev);
-	struct aq_nic_cfg_s *cfg = aq_nic_get_cfg(aq_nic);
+	struct aq_nic_cfg_s *cfg;
 	int err = 0;
 
+	cfg = aq_nic_get_cfg(aq_nic);
+
 	switch (cmd->cmd) {
 	case ETHTOOL_GRXRINGS:
 		cmd->data = cfg->vecs;
 		break;
+	case ETHTOOL_GRXCLSRLCNT:
+		cmd->rule_cnt = aq_get_rxnfc_count_all_rules(aq_nic);
+		break;
+	case ETHTOOL_GRXCLSRULE:
+		err = aq_get_rxnfc_rule(aq_nic, cmd);
+		break;
+	case ETHTOOL_GRXCLSRLALL:
+		err = aq_get_rxnfc_all_rules(aq_nic, cmd, rule_locs);
+		break;
+	default:
+		err = -EOPNOTSUPP;
+		break;
+	}
+
+	return err;
+}
+
+static int aq_ethtool_set_rxnfc(struct net_device *ndev,
+				struct ethtool_rxnfc *cmd)
+{
+	struct aq_nic_s *aq_nic = netdev_priv(ndev);
+	int err = 0;
 
+	switch (cmd->cmd) {
+	case ETHTOOL_SRXCLSRLINS:
+		err = aq_add_rxnfc_rule(aq_nic, cmd);
+		break;
+	case ETHTOOL_SRXCLSRLDEL:
+		err = aq_del_rxnfc_rule(aq_nic, cmd);
+		break;
 	default:
 		err = -EOPNOTSUPP;
 		break;
@@ -245,18 +552,539 @@ static int aq_ethtool_get_rxnfc(struct net_device *ndev,
 	return err;
 }
 
+static int aq_ethtool_get_coalesce(struct net_device *ndev,
+				   struct ethtool_coalesce *coal,
+				   struct kernel_ethtool_coalesce *kernel_coal,
+				   struct netlink_ext_ack *extack)
+{
+	struct aq_nic_s *aq_nic = netdev_priv(ndev);
+	struct aq_nic_cfg_s *cfg;
+
+	cfg = aq_nic_get_cfg(aq_nic);
+
+	if (cfg->itr == AQ_CFG_INTERRUPT_MODERATION_ON ||
+	    cfg->itr == AQ_CFG_INTERRUPT_MODERATION_AUTO) {
+		coal->rx_coalesce_usecs = cfg->rx_itr;
+		coal->tx_coalesce_usecs = cfg->tx_itr;
+		coal->rx_max_coalesced_frames = 0;
+		coal->tx_max_coalesced_frames = 0;
+	} else {
+		coal->rx_coalesce_usecs = 0;
+		coal->tx_coalesce_usecs = 0;
+		coal->rx_max_coalesced_frames = 1;
+		coal->tx_max_coalesced_frames = 1;
+	}
+
+	return 0;
+}
+
+static int aq_ethtool_set_coalesce(struct net_device *ndev,
+				   struct ethtool_coalesce *coal,
+				   struct kernel_ethtool_coalesce *kernel_coal,
+				   struct netlink_ext_ack *extack)
+{
+	struct aq_nic_s *aq_nic = netdev_priv(ndev);
+	struct aq_nic_cfg_s *cfg;
+
+	cfg = aq_nic_get_cfg(aq_nic);
+
+	/* Atlantic only supports timing based coalescing
+	 */
+	if (coal->rx_max_coalesced_frames > 1 ||
+	    coal->tx_max_coalesced_frames > 1)
+		return -EOPNOTSUPP;
+
+	/* We do not support frame counting. Check this
+	 */
+	if (!(coal->rx_max_coalesced_frames == !coal->rx_coalesce_usecs))
+		return -EOPNOTSUPP;
+	if (!(coal->tx_max_coalesced_frames == !coal->tx_coalesce_usecs))
+		return -EOPNOTSUPP;
+
+	if (coal->rx_coalesce_usecs > AQ_CFG_INTERRUPT_MODERATION_USEC_MAX ||
+	    coal->tx_coalesce_usecs > AQ_CFG_INTERRUPT_MODERATION_USEC_MAX)
+		return -EINVAL;
+
+	cfg->itr = AQ_CFG_INTERRUPT_MODERATION_ON;
+
+	cfg->rx_itr = coal->rx_coalesce_usecs;
+	cfg->tx_itr = coal->tx_coalesce_usecs;
+
+	return aq_nic_update_interrupt_moderation_settings(aq_nic);
+}
+
+static void aq_ethtool_get_wol(struct net_device *ndev,
+			       struct ethtool_wolinfo *wol)
+{
+	struct aq_nic_s *aq_nic = netdev_priv(ndev);
+	struct aq_nic_cfg_s *cfg;
+
+	cfg = aq_nic_get_cfg(aq_nic);
+
+	wol->supported = AQ_NIC_WOL_MODES;
+	wol->wolopts = cfg->wol;
+}
+
+static int aq_ethtool_set_wol(struct net_device *ndev,
+			      struct ethtool_wolinfo *wol)
+{
+	struct pci_dev *pdev = to_pci_dev(ndev->dev.parent);
+	struct aq_nic_s *aq_nic = netdev_priv(ndev);
+	struct aq_nic_cfg_s *cfg;
+	int err = 0;
+
+	cfg = aq_nic_get_cfg(aq_nic);
+
+	if (wol->wolopts & ~AQ_NIC_WOL_MODES)
+		return -EOPNOTSUPP;
+
+	cfg->wol = wol->wolopts;
+
+	err = device_set_wakeup_enable(&pdev->dev, !!cfg->wol);
+
+	return err;
+}
+
+static int aq_ethtool_get_ts_info(struct net_device *ndev,
+				  struct kernel_ethtool_ts_info *info)
+{
+	struct aq_nic_s *aq_nic = netdev_priv(ndev);
+
+	ethtool_op_get_ts_info(ndev, info);
+
+	if (!aq_nic->aq_ptp)
+		return 0;
+
+	info->so_timestamping |=
+		SOF_TIMESTAMPING_TX_HARDWARE |
+		SOF_TIMESTAMPING_RX_HARDWARE |
+		SOF_TIMESTAMPING_RAW_HARDWARE;
+
+	info->tx_types = BIT(HWTSTAMP_TX_OFF) |
+			 BIT(HWTSTAMP_TX_ON);
+
+	info->rx_filters = BIT(HWTSTAMP_FILTER_NONE);
+
+	info->rx_filters |= BIT(HWTSTAMP_FILTER_PTP_V2_L4_EVENT) |
+			    BIT(HWTSTAMP_FILTER_PTP_V2_L2_EVENT) |
+			    BIT(HWTSTAMP_FILTER_PTP_V2_EVENT);
+
+#if IS_REACHABLE(CONFIG_PTP_1588_CLOCK)
+	info->phc_index = ptp_clock_index(aq_ptp_get_ptp_clock(aq_nic->aq_ptp));
+#endif
+
+	return 0;
+}
+
+static void eee_mask_to_ethtool_mask(unsigned long *mode, u32 speed)
+{
+	if (speed & AQ_NIC_RATE_EEE_10G)
+		linkmode_set_bit(ETHTOOL_LINK_MODE_10000baseT_Full_BIT, mode);
+
+	if (speed & AQ_NIC_RATE_EEE_1G)
+		linkmode_set_bit(ETHTOOL_LINK_MODE_1000baseT_Full_BIT, mode);
+
+	if (speed & AQ_NIC_RATE_EEE_100M)
+		linkmode_set_bit(ETHTOOL_LINK_MODE_100baseT_Full_BIT, mode);
+}
+
+static int aq_ethtool_get_eee(struct net_device *ndev, struct ethtool_keee *eee)
+{
+	struct aq_nic_s *aq_nic = netdev_priv(ndev);
+	u32 rate, supported_rates;
+	int err = 0;
+
+	if (!aq_nic->aq_fw_ops->get_eee_rate)
+		return -EOPNOTSUPP;
+
+	mutex_lock(&aq_nic->fwreq_mutex);
+	err = aq_nic->aq_fw_ops->get_eee_rate(aq_nic->aq_hw, &rate,
+					      &supported_rates);
+	mutex_unlock(&aq_nic->fwreq_mutex);
+	if (err < 0)
+		return err;
+
+	eee_mask_to_ethtool_mask(eee->supported, supported_rates);
+
+	if (aq_nic->aq_nic_cfg.eee_speeds)
+		linkmode_copy(eee->advertised, eee->supported);
+
+	eee_mask_to_ethtool_mask(eee->lp_advertised, rate);
+
+	eee->eee_enabled = !linkmode_empty(eee->advertised);
+
+	eee->tx_lpi_enabled = eee->eee_enabled;
+	if ((supported_rates & rate) & AQ_NIC_RATE_EEE_MSK)
+		eee->eee_active = true;
+
+	return 0;
+}
+
+static int aq_ethtool_set_eee(struct net_device *ndev, struct ethtool_keee *eee)
+{
+	struct aq_nic_s *aq_nic = netdev_priv(ndev);
+	u32 rate, supported_rates;
+	struct aq_nic_cfg_s *cfg;
+	int err = 0;
+
+	cfg = aq_nic_get_cfg(aq_nic);
+
+	if (unlikely(!aq_nic->aq_fw_ops->get_eee_rate ||
+		     !aq_nic->aq_fw_ops->set_eee_rate))
+		return -EOPNOTSUPP;
+
+	mutex_lock(&aq_nic->fwreq_mutex);
+	err = aq_nic->aq_fw_ops->get_eee_rate(aq_nic->aq_hw, &rate,
+					      &supported_rates);
+	mutex_unlock(&aq_nic->fwreq_mutex);
+	if (err < 0)
+		return err;
+
+	if (eee->eee_enabled) {
+		rate = supported_rates;
+		cfg->eee_speeds = rate;
+	} else {
+		rate = 0;
+		cfg->eee_speeds = 0;
+	}
+
+	mutex_lock(&aq_nic->fwreq_mutex);
+	err = aq_nic->aq_fw_ops->set_eee_rate(aq_nic->aq_hw, rate);
+	mutex_unlock(&aq_nic->fwreq_mutex);
+
+	return err;
+}
+
+static int aq_ethtool_nway_reset(struct net_device *ndev)
+{
+	struct aq_nic_s *aq_nic = netdev_priv(ndev);
+	int err = 0;
+
+	if (unlikely(!aq_nic->aq_fw_ops->renegotiate))
+		return -EOPNOTSUPP;
+
+	if (netif_running(ndev)) {
+		mutex_lock(&aq_nic->fwreq_mutex);
+		err = aq_nic->aq_fw_ops->renegotiate(aq_nic->aq_hw);
+		mutex_unlock(&aq_nic->fwreq_mutex);
+	}
+
+	return err;
+}
+
+static void aq_ethtool_get_pauseparam(struct net_device *ndev,
+				      struct ethtool_pauseparam *pause)
+{
+	struct aq_nic_s *aq_nic = netdev_priv(ndev);
+	int fc = aq_nic->aq_nic_cfg.fc.req;
+
+	pause->autoneg = 0;
+
+	pause->rx_pause = !!(fc & AQ_NIC_FC_RX);
+	pause->tx_pause = !!(fc & AQ_NIC_FC_TX);
+}
+
+static int aq_ethtool_set_pauseparam(struct net_device *ndev,
+				     struct ethtool_pauseparam *pause)
+{
+	struct aq_nic_s *aq_nic = netdev_priv(ndev);
+	int err = 0;
+
+	if (!aq_nic->aq_fw_ops->set_flow_control)
+		return -EOPNOTSUPP;
+
+	if (pause->autoneg == AUTONEG_ENABLE)
+		return -EOPNOTSUPP;
+
+	if (pause->rx_pause)
+		aq_nic->aq_hw->aq_nic_cfg->fc.req |= AQ_NIC_FC_RX;
+	else
+		aq_nic->aq_hw->aq_nic_cfg->fc.req &= ~AQ_NIC_FC_RX;
+
+	if (pause->tx_pause)
+		aq_nic->aq_hw->aq_nic_cfg->fc.req |= AQ_NIC_FC_TX;
+	else
+		aq_nic->aq_hw->aq_nic_cfg->fc.req &= ~AQ_NIC_FC_TX;
+
+	mutex_lock(&aq_nic->fwreq_mutex);
+	err = aq_nic->aq_fw_ops->set_flow_control(aq_nic->aq_hw);
+	mutex_unlock(&aq_nic->fwreq_mutex);
+
+	return err;
+}
+
+static void aq_get_ringparam(struct net_device *ndev,
+			     struct ethtool_ringparam *ring,
+			     struct kernel_ethtool_ringparam *kernel_ring,
+			     struct netlink_ext_ack *extack)
+{
+	struct aq_nic_s *aq_nic = netdev_priv(ndev);
+	struct aq_nic_cfg_s *cfg;
+
+	cfg = aq_nic_get_cfg(aq_nic);
+
+	ring->rx_pending = cfg->rxds;
+	ring->tx_pending = cfg->txds;
+
+	ring->rx_max_pending = cfg->aq_hw_caps->rxds_max;
+	ring->tx_max_pending = cfg->aq_hw_caps->txds_max;
+}
+
+static int aq_set_ringparam(struct net_device *ndev,
+			    struct ethtool_ringparam *ring,
+			    struct kernel_ethtool_ringparam *kernel_ring,
+			    struct netlink_ext_ack *extack)
+{
+	struct aq_nic_s *aq_nic = netdev_priv(ndev);
+	const struct aq_hw_caps_s *hw_caps;
+	bool ndev_running = false;
+	struct aq_nic_cfg_s *cfg;
+	int err = 0;
+
+	cfg = aq_nic_get_cfg(aq_nic);
+	hw_caps = cfg->aq_hw_caps;
+
+	if (ring->rx_mini_pending || ring->rx_jumbo_pending) {
+		err = -EOPNOTSUPP;
+		goto err_exit;
+	}
+
+	if (netif_running(ndev)) {
+		ndev_running = true;
+		aq_ndev_close(ndev);
+	}
+
+	cfg->rxds = max(ring->rx_pending, hw_caps->rxds_min);
+	cfg->rxds = min(cfg->rxds, hw_caps->rxds_max);
+	cfg->rxds = ALIGN(cfg->rxds, AQ_HW_RXD_MULTIPLE);
+
+	cfg->txds = max(ring->tx_pending, hw_caps->txds_min);
+	cfg->txds = min(cfg->txds, hw_caps->txds_max);
+	cfg->txds = ALIGN(cfg->txds, AQ_HW_TXD_MULTIPLE);
+
+	err = aq_nic_realloc_vectors(aq_nic);
+	if (err)
+		goto err_exit;
+
+	if (ndev_running)
+		err = aq_ndev_open(ndev);
+
+err_exit:
+	return err;
+}
+
+static u32 aq_get_msg_level(struct net_device *ndev)
+{
+	struct aq_nic_s *aq_nic = netdev_priv(ndev);
+
+	return aq_nic->msg_enable;
+}
+
+static void aq_set_msg_level(struct net_device *ndev, u32 data)
+{
+	struct aq_nic_s *aq_nic = netdev_priv(ndev);
+
+	aq_nic->msg_enable = data;
+}
+
+static u32 aq_ethtool_get_priv_flags(struct net_device *ndev)
+{
+	struct aq_nic_s *aq_nic = netdev_priv(ndev);
+
+	return aq_nic->aq_nic_cfg.priv_flags;
+}
+
+static int aq_ethtool_set_priv_flags(struct net_device *ndev, u32 flags)
+{
+	struct aq_nic_s *aq_nic = netdev_priv(ndev);
+	struct aq_nic_cfg_s *cfg;
+	u32 priv_flags;
+	int ret = 0;
+
+	cfg = aq_nic_get_cfg(aq_nic);
+	priv_flags = cfg->priv_flags;
+
+	if (flags & ~AQ_PRIV_FLAGS_MASK)
+		return -EOPNOTSUPP;
+
+	if (hweight32((flags | priv_flags) & AQ_HW_LOOPBACK_MASK) > 1) {
+		netdev_info(ndev, "Can't enable more than one loopback simultaneously\n");
+		return -EINVAL;
+	}
+
+	cfg->priv_flags = flags;
+
+	if ((priv_flags ^ flags) & BIT(AQ_HW_LOOPBACK_DMA_NET)) {
+		if (netif_running(ndev)) {
+			dev_close(ndev);
+
+			dev_open(ndev, NULL);
+		}
+	} else if ((priv_flags ^ flags) & AQ_HW_LOOPBACK_MASK) {
+		ret = aq_nic_set_loopback(aq_nic);
+	}
+
+	return ret;
+}
+
+static int aq_ethtool_get_phy_tunable(struct net_device *ndev,
+				      const struct ethtool_tunable *tuna, void *data)
+{
+	struct aq_nic_s *aq_nic = netdev_priv(ndev);
+
+	switch (tuna->id) {
+	case ETHTOOL_PHY_EDPD: {
+		u16 *val = data;
+
+		*val = aq_nic->aq_nic_cfg.is_media_detect ? AQ_HW_MEDIA_DETECT_CNT : 0;
+		break;
+	}
+	case ETHTOOL_PHY_DOWNSHIFT: {
+		u8 *val = data;
+
+		*val = (u8)aq_nic->aq_nic_cfg.downshift_counter;
+		break;
+	}
+	default:
+		return -EOPNOTSUPP;
+	}
+
+	return 0;
+}
+
+static int aq_ethtool_set_phy_tunable(struct net_device *ndev,
+				      const struct ethtool_tunable *tuna, const void *data)
+{
+	int err = -EOPNOTSUPP;
+	struct aq_nic_s *aq_nic = netdev_priv(ndev);
+
+	switch (tuna->id) {
+	case ETHTOOL_PHY_EDPD: {
+		const u16 *val = data;
+
+		err = aq_nic_set_media_detect(aq_nic, *val);
+		break;
+	}
+	case ETHTOOL_PHY_DOWNSHIFT: {
+		const u8 *val = data;
+
+		err = aq_nic_set_downshift(aq_nic, *val);
+		break;
+	}
+	default:
+		break;
+	}
+
+	return err;
+}
+
+static int aq_ethtool_get_module_info(struct net_device *ndev,
+				      struct ethtool_modinfo *modinfo)
+{
+	struct aq_nic_s *aq_nic = netdev_priv(ndev);
+	u8 compliance_val, dom_type;
+	int err;
+
+	/* Module EEPROM is only supported for controllers with external PHY */
+	if (aq_nic->aq_nic_cfg.aq_hw_caps->media_type != AQ_HW_MEDIA_TYPE_FIBRE ||
+	    !aq_nic->aq_hw_ops->hw_read_module_eeprom)
+		return -EOPNOTSUPP;
+
+	err = aq_nic->aq_hw_ops->hw_read_module_eeprom(aq_nic->aq_hw,
+		SFF_8472_ID_ADDR, SFF_8472_COMP_ADDR, 1, &compliance_val);
+	if (err)
+		return err;
+
+	err = aq_nic->aq_hw_ops->hw_read_module_eeprom(aq_nic->aq_hw,
+		SFF_8472_ID_ADDR, SFF_8472_DOM_TYPE_ADDR, 1, &dom_type);
+	if (err)
+		return err;
+
+	if (dom_type & SFF_8472_ADDRESS_CHANGE_REQ_MASK || compliance_val == 0x00) {
+		modinfo->type = ETH_MODULE_SFF_8079;
+		modinfo->eeprom_len = ETH_MODULE_SFF_8079_LEN;
+	} else {
+		modinfo->type = ETH_MODULE_SFF_8472;
+		modinfo->eeprom_len = ETH_MODULE_SFF_8472_LEN;
+	}
+	return 0;
+}
+
+static int aq_ethtool_get_module_eeprom(struct net_device *ndev,
+					struct ethtool_eeprom *ee, unsigned char *data)
+{
+	struct aq_nic_s *aq_nic = netdev_priv(ndev);
+	unsigned int first, last, len;
+	int err;
+
+	if (!aq_nic->aq_hw_ops->hw_read_module_eeprom)
+		return -EOPNOTSUPP;
+
+	first = ee->offset;
+	last = ee->offset + ee->len;
+
+	if (first < ETH_MODULE_SFF_8079_LEN) {
+		len = min(last, ETH_MODULE_SFF_8079_LEN);
+		len -= first;
+
+		err = aq_nic->aq_hw_ops->hw_read_module_eeprom(aq_nic->aq_hw,
+			SFF_8472_ID_ADDR, first, len, data);
+		if (err)
+			return err;
+
+		first += len;
+		data += len;
+	}
+	if (first < ETH_MODULE_SFF_8472_LEN && last > ETH_MODULE_SFF_8079_LEN) {
+		len = min(last, ETH_MODULE_SFF_8472_LEN);
+		len -= first;
+		first -= ETH_MODULE_SFF_8079_LEN;
+
+		err = aq_nic->aq_hw_ops->hw_read_module_eeprom(aq_nic->aq_hw,
+			SFF_8472_DIAGNOSTICS_ADDR, first, len, data);
+		if (err)
+			return err;
+	}
+	return 0;
+}
+
 const struct ethtool_ops aq_ethtool_ops = {
+	.supported_coalesce_params = ETHTOOL_COALESCE_USECS |
+				     ETHTOOL_COALESCE_MAX_FRAMES,
 	.get_link            = aq_ethtool_get_link,
 	.get_regs_len        = aq_ethtool_get_regs_len,
 	.get_regs            = aq_ethtool_get_regs,
 	.get_drvinfo         = aq_ethtool_get_drvinfo,
 	.get_strings         = aq_ethtool_get_strings,
+	.set_phys_id         = aq_ethtool_set_phys_id,
 	.get_rxfh_indir_size = aq_ethtool_get_rss_indir_size,
+	.get_wol             = aq_ethtool_get_wol,
+	.set_wol             = aq_ethtool_set_wol,
+	.nway_reset          = aq_ethtool_nway_reset,
+	.get_ringparam       = aq_get_ringparam,
+	.set_ringparam       = aq_set_ringparam,
+	.get_eee             = aq_ethtool_get_eee,
+	.set_eee             = aq_ethtool_set_eee,
+	.get_pauseparam      = aq_ethtool_get_pauseparam,
+	.set_pauseparam      = aq_ethtool_set_pauseparam,
 	.get_rxfh_key_size   = aq_ethtool_get_rss_key_size,
 	.get_rxfh            = aq_ethtool_get_rss,
+	.set_rxfh            = aq_ethtool_set_rss,
 	.get_rxnfc           = aq_ethtool_get_rxnfc,
+	.set_rxnfc           = aq_ethtool_set_rxnfc,
+	.get_msglevel        = aq_get_msg_level,
+	.set_msglevel        = aq_set_msg_level,
 	.get_sset_count      = aq_ethtool_get_sset_count,
 	.get_ethtool_stats   = aq_ethtool_stats,
+	.get_priv_flags      = aq_ethtool_get_priv_flags,
+	.set_priv_flags      = aq_ethtool_set_priv_flags,
 	.get_link_ksettings  = aq_ethtool_get_link_ksettings,
 	.set_link_ksettings  = aq_ethtool_set_link_ksettings,
+	.get_coalesce	     = aq_ethtool_get_coalesce,
+	.set_coalesce	     = aq_ethtool_set_coalesce,
+	.get_ts_info         = aq_ethtool_get_ts_info,
+	.get_phy_tunable     = aq_ethtool_get_phy_tunable,
+	.set_phy_tunable     = aq_ethtool_set_phy_tunable,
+	.get_module_info     = aq_ethtool_get_module_info,
+	.get_module_eeprom   = aq_ethtool_get_module_eeprom,
 };
diff --git a/drivers/net/ethernet/aquantia/atlantic/aq_ethtool.h b/drivers/net/ethernet/aquantia/atlantic/aq_ethtool.h
index 21c126eeb5eb..f26fe1a75539 100644
--- a/drivers/net/ethernet/aquantia/atlantic/aq_ethtool.h
+++ b/drivers/net/ethernet/aquantia/atlantic/aq_ethtool.h
@@ -1,10 +1,7 @@
+/* SPDX-License-Identifier: GPL-2.0-only */
 /*
  * aQuantia Corporation Network Driver
  * Copyright (C) 2014-2017 aQuantia Corporation. All rights reserved
- *
- * This program is free software; you can redistribute it and/or modify it
- * under the terms and conditions of the GNU General Public License,
- * version 2, as published by the Free Software Foundation.
  */
 
 /* File aq_ethtool.h: Declaration of ethertool related functions. */
@@ -15,5 +12,14 @@
 #include "aq_common.h"
 
 extern const struct ethtool_ops aq_ethtool_ops;
+#define AQ_PRIV_FLAGS_MASK   (AQ_HW_LOOPBACK_MASK)
+
+#define SFF_8472_ID_ADDR 0x50
+#define SFF_8472_DIAGNOSTICS_ADDR 0x51
+
+#define SFF_8472_COMP_ADDR	0x5e
+#define SFF_8472_DOM_TYPE_ADDR	0x5c
+
+#define SFF_8472_ADDRESS_CHANGE_REQ_MASK 0x4
 
 #endif /* AQ_ETHTOOL_H */
diff --git a/drivers/net/ethernet/aquantia/atlantic/aq_filters.c b/drivers/net/ethernet/aquantia/atlantic/aq_filters.c
new file mode 100644
index 000000000000..30a573db02bb
--- /dev/null
+++ b/drivers/net/ethernet/aquantia/atlantic/aq_filters.c
@@ -0,0 +1,889 @@
+// SPDX-License-Identifier: GPL-2.0-only
+/* Copyright (C) 2014-2019 aQuantia Corporation. */
+
+/* File aq_filters.c: RX filters related functions. */
+
+#include "aq_filters.h"
+
+static bool __must_check
+aq_rule_is_approve(struct ethtool_rx_flow_spec *fsp)
+{
+	if (fsp->flow_type & FLOW_MAC_EXT)
+		return false;
+
+	switch (fsp->flow_type & ~FLOW_EXT) {
+	case ETHER_FLOW:
+	case TCP_V4_FLOW:
+	case UDP_V4_FLOW:
+	case SCTP_V4_FLOW:
+	case TCP_V6_FLOW:
+	case UDP_V6_FLOW:
+	case SCTP_V6_FLOW:
+	case IPV4_FLOW:
+	case IPV6_FLOW:
+		return true;
+	case IP_USER_FLOW:
+		switch (fsp->h_u.usr_ip4_spec.proto) {
+		case IPPROTO_TCP:
+		case IPPROTO_UDP:
+		case IPPROTO_SCTP:
+		case IPPROTO_IP:
+			return true;
+		default:
+			return false;
+			}
+	case IPV6_USER_FLOW:
+		switch (fsp->h_u.usr_ip6_spec.l4_proto) {
+		case IPPROTO_TCP:
+		case IPPROTO_UDP:
+		case IPPROTO_SCTP:
+		case IPPROTO_IP:
+			return true;
+		default:
+			return false;
+			}
+	default:
+		return false;
+	}
+
+	return false;
+}
+
+static bool __must_check
+aq_match_filter(struct ethtool_rx_flow_spec *fsp1,
+		struct ethtool_rx_flow_spec *fsp2)
+{
+	if (fsp1->flow_type != fsp2->flow_type ||
+	    memcmp(&fsp1->h_u, &fsp2->h_u, sizeof(fsp2->h_u)) ||
+	    memcmp(&fsp1->h_ext, &fsp2->h_ext, sizeof(fsp2->h_ext)) ||
+	    memcmp(&fsp1->m_u, &fsp2->m_u, sizeof(fsp2->m_u)) ||
+	    memcmp(&fsp1->m_ext, &fsp2->m_ext, sizeof(fsp2->m_ext)))
+		return false;
+
+	return true;
+}
+
+static bool __must_check
+aq_rule_already_exists(struct aq_nic_s *aq_nic,
+		       struct ethtool_rx_flow_spec *fsp)
+{
+	struct aq_rx_filter *rule;
+	struct hlist_node *aq_node2;
+	struct aq_hw_rx_fltrs_s *rx_fltrs = aq_get_hw_rx_fltrs(aq_nic);
+
+	hlist_for_each_entry_safe(rule, aq_node2,
+				  &rx_fltrs->filter_list, aq_node) {
+		if (rule->aq_fsp.location == fsp->location)
+			continue;
+		if (aq_match_filter(&rule->aq_fsp, fsp)) {
+			netdev_err(aq_nic->ndev,
+				   "ethtool: This filter is already set\n");
+			return true;
+		}
+	}
+
+	return false;
+}
+
+static int aq_check_approve_fl3l4(struct aq_nic_s *aq_nic,
+				  struct aq_hw_rx_fltrs_s *rx_fltrs,
+				  struct ethtool_rx_flow_spec *fsp)
+{
+	u32 last_location = AQ_RX_LAST_LOC_FL3L4 -
+			    aq_nic->aq_hw_rx_fltrs.fl3l4.reserved_count;
+
+	if (fsp->location < AQ_RX_FIRST_LOC_FL3L4 ||
+	    fsp->location > last_location) {
+		netdev_err(aq_nic->ndev,
+			   "ethtool: location must be in range [%d, %d]",
+			   AQ_RX_FIRST_LOC_FL3L4, last_location);
+		return -EINVAL;
+	}
+	if (rx_fltrs->fl3l4.is_ipv6 && rx_fltrs->fl3l4.active_ipv4) {
+		rx_fltrs->fl3l4.is_ipv6 = false;
+		netdev_err(aq_nic->ndev,
+			   "ethtool: mixing ipv4 and ipv6 is not allowed");
+		return -EINVAL;
+	} else if (!rx_fltrs->fl3l4.is_ipv6 && rx_fltrs->fl3l4.active_ipv6) {
+		rx_fltrs->fl3l4.is_ipv6 = true;
+		netdev_err(aq_nic->ndev,
+			   "ethtool: mixing ipv4 and ipv6 is not allowed");
+		return -EINVAL;
+	} else if (rx_fltrs->fl3l4.is_ipv6		      &&
+		   fsp->location != AQ_RX_FIRST_LOC_FL3L4 + 4 &&
+		   fsp->location != AQ_RX_FIRST_LOC_FL3L4) {
+		netdev_err(aq_nic->ndev,
+			   "ethtool: The specified location for ipv6 must be %d or %d",
+			   AQ_RX_FIRST_LOC_FL3L4, AQ_RX_FIRST_LOC_FL3L4 + 4);
+		return -EINVAL;
+	}
+
+	return 0;
+}
+
+static int __must_check
+aq_check_approve_fl2(struct aq_nic_s *aq_nic,
+		     struct aq_hw_rx_fltrs_s *rx_fltrs,
+		     struct ethtool_rx_flow_spec *fsp)
+{
+	u32 last_location = AQ_RX_LAST_LOC_FETHERT -
+			    aq_nic->aq_hw_rx_fltrs.fet_reserved_count;
+
+	if (fsp->location < AQ_RX_FIRST_LOC_FETHERT ||
+	    fsp->location > last_location) {
+		netdev_err(aq_nic->ndev,
+			   "ethtool: location must be in range [%d, %d]",
+			   AQ_RX_FIRST_LOC_FETHERT,
+			   last_location);
+		return -EINVAL;
+	}
+
+	if (be16_to_cpu(fsp->m_ext.vlan_tci) == VLAN_PRIO_MASK &&
+	    fsp->m_u.ether_spec.h_proto == 0U) {
+		netdev_err(aq_nic->ndev,
+			   "ethtool: proto (ether_type) parameter must be specified");
+		return -EINVAL;
+	}
+
+	return 0;
+}
+
+static int __must_check
+aq_check_approve_fvlan(struct aq_nic_s *aq_nic,
+		       struct aq_hw_rx_fltrs_s *rx_fltrs,
+		       struct ethtool_rx_flow_spec *fsp)
+{
+	struct aq_nic_cfg_s *cfg = &aq_nic->aq_nic_cfg;
+
+	if (fsp->location < AQ_RX_FIRST_LOC_FVLANID ||
+	    fsp->location > AQ_RX_LAST_LOC_FVLANID) {
+		netdev_err(aq_nic->ndev,
+			   "ethtool: location must be in range [%d, %d]",
+			   AQ_RX_FIRST_LOC_FVLANID,
+			   AQ_RX_LAST_LOC_FVLANID);
+		return -EINVAL;
+	}
+
+	if ((aq_nic->ndev->features & NETIF_F_HW_VLAN_CTAG_FILTER) &&
+	    (!test_bit(be16_to_cpu(fsp->h_ext.vlan_tci) & VLAN_VID_MASK,
+		       aq_nic->active_vlans))) {
+		netdev_err(aq_nic->ndev,
+			   "ethtool: unknown vlan-id specified");
+		return -EINVAL;
+	}
+
+	if (fsp->ring_cookie > cfg->num_rss_queues * cfg->tcs) {
+		netdev_err(aq_nic->ndev,
+			   "ethtool: queue number must be in range [0, %d]",
+			   cfg->num_rss_queues * cfg->tcs - 1);
+		return -EINVAL;
+	}
+	return 0;
+}
+
+static int __must_check
+aq_check_filter(struct aq_nic_s *aq_nic,
+		struct ethtool_rx_flow_spec *fsp)
+{
+	int err = 0;
+	struct aq_hw_rx_fltrs_s *rx_fltrs = aq_get_hw_rx_fltrs(aq_nic);
+
+	if (fsp->flow_type & FLOW_EXT) {
+		if (be16_to_cpu(fsp->m_ext.vlan_tci) == VLAN_VID_MASK) {
+			err = aq_check_approve_fvlan(aq_nic, rx_fltrs, fsp);
+		} else if (be16_to_cpu(fsp->m_ext.vlan_tci) == VLAN_PRIO_MASK) {
+			err = aq_check_approve_fl2(aq_nic, rx_fltrs, fsp);
+		} else {
+			netdev_err(aq_nic->ndev,
+				   "ethtool: invalid vlan mask 0x%x specified",
+				   be16_to_cpu(fsp->m_ext.vlan_tci));
+			err = -EINVAL;
+		}
+	} else {
+		switch (fsp->flow_type & ~FLOW_EXT) {
+		case ETHER_FLOW:
+			err = aq_check_approve_fl2(aq_nic, rx_fltrs, fsp);
+			break;
+		case TCP_V4_FLOW:
+		case UDP_V4_FLOW:
+		case SCTP_V4_FLOW:
+		case IPV4_FLOW:
+		case IP_USER_FLOW:
+			rx_fltrs->fl3l4.is_ipv6 = false;
+			err = aq_check_approve_fl3l4(aq_nic, rx_fltrs, fsp);
+			break;
+		case TCP_V6_FLOW:
+		case UDP_V6_FLOW:
+		case SCTP_V6_FLOW:
+		case IPV6_FLOW:
+		case IPV6_USER_FLOW:
+			rx_fltrs->fl3l4.is_ipv6 = true;
+			err = aq_check_approve_fl3l4(aq_nic, rx_fltrs, fsp);
+			break;
+		default:
+			netdev_err(aq_nic->ndev,
+				   "ethtool: unknown flow-type specified");
+			err = -EINVAL;
+		}
+	}
+
+	return err;
+}
+
+static bool __must_check
+aq_rule_is_not_support(struct aq_nic_s *aq_nic,
+		       struct ethtool_rx_flow_spec *fsp)
+{
+	bool rule_is_not_support = false;
+
+	if (!(aq_nic->ndev->features & NETIF_F_NTUPLE)) {
+		netdev_err(aq_nic->ndev,
+			   "ethtool: Please, to enable the RX flow control:\n"
+			   "ethtool -K %s ntuple on\n", aq_nic->ndev->name);
+		rule_is_not_support = true;
+	} else if (!aq_rule_is_approve(fsp)) {
+		netdev_err(aq_nic->ndev,
+			   "ethtool: The specified flow type is not supported\n");
+		rule_is_not_support = true;
+	} else if ((fsp->flow_type & ~FLOW_EXT) != ETHER_FLOW &&
+		   (fsp->h_u.tcp_ip4_spec.tos ||
+		    fsp->h_u.tcp_ip6_spec.tclass)) {
+		netdev_err(aq_nic->ndev,
+			   "ethtool: The specified tos tclass are not supported\n");
+		rule_is_not_support = true;
+	} else if (fsp->flow_type & FLOW_MAC_EXT) {
+		netdev_err(aq_nic->ndev,
+			   "ethtool: MAC_EXT is not supported");
+		rule_is_not_support = true;
+	}
+
+	return rule_is_not_support;
+}
+
+static bool __must_check
+aq_rule_is_not_correct(struct aq_nic_s *aq_nic,
+		       struct ethtool_rx_flow_spec *fsp)
+{
+	struct aq_nic_cfg_s *cfg = &aq_nic->aq_nic_cfg;
+	bool rule_is_not_correct = false;
+
+	if (!aq_nic) {
+		rule_is_not_correct = true;
+	} else if (fsp->location > AQ_RX_MAX_RXNFC_LOC) {
+		netdev_err(aq_nic->ndev,
+			   "ethtool: The specified number %u rule is invalid\n",
+			   fsp->location);
+		rule_is_not_correct = true;
+	} else if (aq_check_filter(aq_nic, fsp)) {
+		rule_is_not_correct = true;
+	} else if (fsp->ring_cookie != RX_CLS_FLOW_DISC) {
+		if (fsp->ring_cookie >= cfg->num_rss_queues * cfg->tcs) {
+			netdev_err(aq_nic->ndev,
+				   "ethtool: The specified action is invalid.\n"
+				   "Maximum allowable value action is %u.\n",
+				   cfg->num_rss_queues * cfg->tcs - 1);
+			rule_is_not_correct = true;
+		}
+	}
+
+	return rule_is_not_correct;
+}
+
+static int __must_check
+aq_check_rule(struct aq_nic_s *aq_nic,
+	      struct ethtool_rx_flow_spec *fsp)
+{
+	int err = 0;
+
+	if (aq_rule_is_not_correct(aq_nic, fsp))
+		err = -EINVAL;
+	else if (aq_rule_is_not_support(aq_nic, fsp))
+		err = -EOPNOTSUPP;
+	else if (aq_rule_already_exists(aq_nic, fsp))
+		err = -EEXIST;
+
+	return err;
+}
+
+static void aq_set_data_fl2(struct aq_nic_s *aq_nic,
+			    struct aq_rx_filter *aq_rx_fltr,
+			    struct aq_rx_filter_l2 *data, bool add)
+{
+	const struct ethtool_rx_flow_spec *fsp = &aq_rx_fltr->aq_fsp;
+
+	memset(data, 0, sizeof(*data));
+
+	data->location = fsp->location - AQ_RX_FIRST_LOC_FETHERT;
+
+	if (fsp->ring_cookie != RX_CLS_FLOW_DISC)
+		data->queue = fsp->ring_cookie;
+	else
+		data->queue = -1;
+
+	data->ethertype = be16_to_cpu(fsp->h_u.ether_spec.h_proto);
+	data->user_priority_en = be16_to_cpu(fsp->m_ext.vlan_tci)
+				 == VLAN_PRIO_MASK;
+	data->user_priority = (be16_to_cpu(fsp->h_ext.vlan_tci)
+			       & VLAN_PRIO_MASK) >> VLAN_PRIO_SHIFT;
+}
+
+static int aq_add_del_fether(struct aq_nic_s *aq_nic,
+			     struct aq_rx_filter *aq_rx_fltr, bool add)
+{
+	struct aq_rx_filter_l2 data;
+	struct aq_hw_s *aq_hw = aq_nic->aq_hw;
+	const struct aq_hw_ops *aq_hw_ops = aq_nic->aq_hw_ops;
+
+	aq_set_data_fl2(aq_nic, aq_rx_fltr, &data, add);
+
+	if (unlikely(!aq_hw_ops->hw_filter_l2_set))
+		return -EOPNOTSUPP;
+	if (unlikely(!aq_hw_ops->hw_filter_l2_clear))
+		return -EOPNOTSUPP;
+
+	if (add)
+		return aq_hw_ops->hw_filter_l2_set(aq_hw, &data);
+	else
+		return aq_hw_ops->hw_filter_l2_clear(aq_hw, &data);
+}
+
+static bool aq_fvlan_is_busy(struct aq_rx_filter_vlan *aq_vlans, int vlan)
+{
+	int i;
+
+	for (i = 0; i < AQ_VLAN_MAX_FILTERS; ++i) {
+		if (aq_vlans[i].enable &&
+		    aq_vlans[i].queue != AQ_RX_QUEUE_NOT_ASSIGNED &&
+		    aq_vlans[i].vlan_id == vlan) {
+			return true;
+		}
+	}
+
+	return false;
+}
+
+/* Function rebuilds array of vlan filters so that filters with assigned
+ * queue have a precedence over just vlans on the interface.
+ */
+static void aq_fvlan_rebuild(struct aq_nic_s *aq_nic,
+			     unsigned long *active_vlans,
+			     struct aq_rx_filter_vlan *aq_vlans)
+{
+	bool vlan_busy = false;
+	int vlan = -1;
+	int i;
+
+	for (i = 0; i < AQ_VLAN_MAX_FILTERS; ++i) {
+		if (aq_vlans[i].enable &&
+		    aq_vlans[i].queue != AQ_RX_QUEUE_NOT_ASSIGNED)
+			continue;
+		do {
+			vlan = find_next_bit(active_vlans,
+					     VLAN_N_VID,
+					     vlan + 1);
+			if (vlan == VLAN_N_VID) {
+				aq_vlans[i].enable = 0U;
+				aq_vlans[i].queue = AQ_RX_QUEUE_NOT_ASSIGNED;
+				aq_vlans[i].vlan_id = 0;
+				continue;
+			}
+
+			vlan_busy = aq_fvlan_is_busy(aq_vlans, vlan);
+			if (!vlan_busy) {
+				aq_vlans[i].enable = 1U;
+				aq_vlans[i].queue = AQ_RX_QUEUE_NOT_ASSIGNED;
+				aq_vlans[i].vlan_id = vlan;
+			}
+		} while (vlan_busy && vlan != VLAN_N_VID);
+	}
+}
+
+static int aq_set_data_fvlan(struct aq_nic_s *aq_nic,
+			     struct aq_rx_filter *aq_rx_fltr,
+			     struct aq_rx_filter_vlan *aq_vlans, bool add)
+{
+	const struct ethtool_rx_flow_spec *fsp = &aq_rx_fltr->aq_fsp;
+	int location = fsp->location - AQ_RX_FIRST_LOC_FVLANID;
+	int i;
+
+	memset(&aq_vlans[location], 0, sizeof(aq_vlans[location]));
+
+	if (!add)
+		return 0;
+
+	/* remove vlan if it was in table without queue assignment */
+	for (i = 0; i < AQ_VLAN_MAX_FILTERS; ++i) {
+		if (aq_vlans[i].vlan_id ==
+		   (be16_to_cpu(fsp->h_ext.vlan_tci) & VLAN_VID_MASK)) {
+			aq_vlans[i].enable = false;
+		}
+	}
+
+	aq_vlans[location].location = location;
+	aq_vlans[location].vlan_id = be16_to_cpu(fsp->h_ext.vlan_tci)
+				     & VLAN_VID_MASK;
+	aq_vlans[location].queue = fsp->ring_cookie & 0x1FU;
+	aq_vlans[location].enable = 1U;
+
+	return 0;
+}
+
+int aq_del_fvlan_by_vlan(struct aq_nic_s *aq_nic, u16 vlan_id)
+{
+	struct aq_hw_rx_fltrs_s *rx_fltrs = aq_get_hw_rx_fltrs(aq_nic);
+	struct aq_rx_filter *rule = NULL;
+	struct hlist_node *aq_node2;
+
+	hlist_for_each_entry_safe(rule, aq_node2,
+				  &rx_fltrs->filter_list, aq_node) {
+		if (be16_to_cpu(rule->aq_fsp.h_ext.vlan_tci) == vlan_id)
+			break;
+	}
+	if (rule && rule->type == aq_rx_filter_vlan &&
+	    be16_to_cpu(rule->aq_fsp.h_ext.vlan_tci) == vlan_id) {
+		struct ethtool_rxnfc cmd;
+
+		cmd.fs.location = rule->aq_fsp.location;
+		return aq_del_rxnfc_rule(aq_nic, &cmd);
+	}
+
+	return -ENOENT;
+}
+
+static int aq_add_del_fvlan(struct aq_nic_s *aq_nic,
+			    struct aq_rx_filter *aq_rx_fltr, bool add)
+{
+	const struct aq_hw_ops *aq_hw_ops = aq_nic->aq_hw_ops;
+
+	if (unlikely(!aq_hw_ops->hw_filter_vlan_set))
+		return -EOPNOTSUPP;
+
+	aq_set_data_fvlan(aq_nic,
+			  aq_rx_fltr,
+			  aq_nic->aq_hw_rx_fltrs.fl2.aq_vlans,
+			  add);
+
+	return aq_filters_vlans_update(aq_nic);
+}
+
+static int aq_set_data_fl3l4(struct aq_nic_s *aq_nic,
+			     struct aq_rx_filter *aq_rx_fltr,
+			     struct aq_rx_filter_l3l4 *data, bool add)
+{
+	struct aq_hw_rx_fltrs_s *rx_fltrs = aq_get_hw_rx_fltrs(aq_nic);
+	const struct ethtool_rx_flow_spec *fsp = &aq_rx_fltr->aq_fsp;
+
+	memset(data, 0, sizeof(*data));
+
+	data->is_ipv6 = rx_fltrs->fl3l4.is_ipv6;
+	data->location = HW_ATL_GET_REG_LOCATION_FL3L4(fsp->location);
+
+	if (!add) {
+		if (!data->is_ipv6)
+			rx_fltrs->fl3l4.active_ipv4 &= ~BIT(data->location);
+		else
+			rx_fltrs->fl3l4.active_ipv6 &=
+				~BIT((data->location) / 4);
+
+		return 0;
+	}
+
+	data->cmd |= HW_ATL_RX_ENABLE_FLTR_L3L4;
+
+	switch (fsp->flow_type) {
+	case TCP_V4_FLOW:
+	case TCP_V6_FLOW:
+		data->cmd |= HW_ATL_RX_ENABLE_CMP_PROT_L4;
+		break;
+	case UDP_V4_FLOW:
+	case UDP_V6_FLOW:
+		data->cmd |= HW_ATL_RX_UDP;
+		data->cmd |= HW_ATL_RX_ENABLE_CMP_PROT_L4;
+		break;
+	case SCTP_V4_FLOW:
+	case SCTP_V6_FLOW:
+		data->cmd |= HW_ATL_RX_SCTP;
+		data->cmd |= HW_ATL_RX_ENABLE_CMP_PROT_L4;
+		break;
+	default:
+		break;
+	}
+
+	if (!data->is_ipv6) {
+		data->ip_src[0] =
+			ntohl(fsp->h_u.tcp_ip4_spec.ip4src);
+		data->ip_dst[0] =
+			ntohl(fsp->h_u.tcp_ip4_spec.ip4dst);
+		rx_fltrs->fl3l4.active_ipv4 |= BIT(data->location);
+	} else {
+		int i;
+
+		rx_fltrs->fl3l4.active_ipv6 |= BIT((data->location) / 4);
+		for (i = 0; i < HW_ATL_RX_CNT_REG_ADDR_IPV6; ++i) {
+			data->ip_dst[i] =
+				ntohl(fsp->h_u.tcp_ip6_spec.ip6dst[i]);
+			data->ip_src[i] =
+				ntohl(fsp->h_u.tcp_ip6_spec.ip6src[i]);
+		}
+		data->cmd |= HW_ATL_RX_ENABLE_L3_IPV6;
+	}
+	if (fsp->flow_type != IP_USER_FLOW &&
+	    fsp->flow_type != IPV6_USER_FLOW) {
+		if (!data->is_ipv6) {
+			data->p_dst =
+				ntohs(fsp->h_u.tcp_ip4_spec.pdst);
+			data->p_src =
+				ntohs(fsp->h_u.tcp_ip4_spec.psrc);
+		} else {
+			data->p_dst =
+				ntohs(fsp->h_u.tcp_ip6_spec.pdst);
+			data->p_src =
+				ntohs(fsp->h_u.tcp_ip6_spec.psrc);
+		}
+	}
+	if (data->ip_src[0] && !data->is_ipv6)
+		data->cmd |= HW_ATL_RX_ENABLE_CMP_SRC_ADDR_L3;
+	if (data->ip_dst[0] && !data->is_ipv6)
+		data->cmd |= HW_ATL_RX_ENABLE_CMP_DEST_ADDR_L3;
+	if (data->p_dst)
+		data->cmd |= HW_ATL_RX_ENABLE_CMP_DEST_PORT_L4;
+	if (data->p_src)
+		data->cmd |= HW_ATL_RX_ENABLE_CMP_SRC_PORT_L4;
+	if (fsp->ring_cookie != RX_CLS_FLOW_DISC) {
+		data->cmd |= HW_ATL_RX_HOST << HW_ATL_RX_ACTION_FL3F4_SHIFT;
+		data->cmd |= fsp->ring_cookie << HW_ATL_RX_QUEUE_FL3L4_SHIFT;
+		data->cmd |= HW_ATL_RX_ENABLE_QUEUE_L3L4;
+	} else {
+		data->cmd |= HW_ATL_RX_DISCARD << HW_ATL_RX_ACTION_FL3F4_SHIFT;
+	}
+
+	return 0;
+}
+
+static int aq_set_fl3l4(struct aq_hw_s *aq_hw,
+			const struct aq_hw_ops *aq_hw_ops,
+			struct aq_rx_filter_l3l4 *data)
+{
+	if (unlikely(!aq_hw_ops->hw_filter_l3l4_set))
+		return -EOPNOTSUPP;
+
+	return aq_hw_ops->hw_filter_l3l4_set(aq_hw, data);
+}
+
+static int aq_add_del_fl3l4(struct aq_nic_s *aq_nic,
+			    struct aq_rx_filter *aq_rx_fltr, bool add)
+{
+	const struct aq_hw_ops *aq_hw_ops = aq_nic->aq_hw_ops;
+	struct aq_hw_s *aq_hw = aq_nic->aq_hw;
+	struct aq_rx_filter_l3l4 data;
+
+	if (unlikely(aq_rx_fltr->aq_fsp.location < AQ_RX_FIRST_LOC_FL3L4 ||
+		     aq_rx_fltr->aq_fsp.location > AQ_RX_LAST_LOC_FL3L4  ||
+		     aq_set_data_fl3l4(aq_nic, aq_rx_fltr, &data, add)))
+		return -EINVAL;
+
+	return aq_set_fl3l4(aq_hw, aq_hw_ops, &data);
+}
+
+static int aq_add_del_rule(struct aq_nic_s *aq_nic,
+			   struct aq_rx_filter *aq_rx_fltr, bool add)
+{
+	int err = -EINVAL;
+
+	if (aq_rx_fltr->aq_fsp.flow_type & FLOW_EXT) {
+		if (be16_to_cpu(aq_rx_fltr->aq_fsp.m_ext.vlan_tci)
+		    == VLAN_VID_MASK) {
+			aq_rx_fltr->type = aq_rx_filter_vlan;
+			err = aq_add_del_fvlan(aq_nic, aq_rx_fltr, add);
+		} else if (be16_to_cpu(aq_rx_fltr->aq_fsp.m_ext.vlan_tci)
+			== VLAN_PRIO_MASK) {
+			aq_rx_fltr->type = aq_rx_filter_ethertype;
+			err = aq_add_del_fether(aq_nic, aq_rx_fltr, add);
+		}
+	} else {
+		switch (aq_rx_fltr->aq_fsp.flow_type & ~FLOW_EXT) {
+		case ETHER_FLOW:
+			aq_rx_fltr->type = aq_rx_filter_ethertype;
+			err = aq_add_del_fether(aq_nic, aq_rx_fltr, add);
+			break;
+		case TCP_V4_FLOW:
+		case UDP_V4_FLOW:
+		case SCTP_V4_FLOW:
+		case IP_USER_FLOW:
+		case TCP_V6_FLOW:
+		case UDP_V6_FLOW:
+		case SCTP_V6_FLOW:
+		case IPV6_USER_FLOW:
+			aq_rx_fltr->type = aq_rx_filter_l3l4;
+			err = aq_add_del_fl3l4(aq_nic, aq_rx_fltr, add);
+			break;
+		default:
+			err = -EINVAL;
+			break;
+		}
+	}
+
+	return err;
+}
+
+static int aq_update_table_filters(struct aq_nic_s *aq_nic,
+				   struct aq_rx_filter *aq_rx_fltr, u16 index,
+				   struct ethtool_rxnfc *cmd)
+{
+	struct aq_hw_rx_fltrs_s *rx_fltrs = aq_get_hw_rx_fltrs(aq_nic);
+	struct aq_rx_filter *rule = NULL, *parent = NULL;
+	struct hlist_node *aq_node2;
+	int err = -EINVAL;
+
+	hlist_for_each_entry_safe(rule, aq_node2,
+				  &rx_fltrs->filter_list, aq_node) {
+		if (rule->aq_fsp.location >= index)
+			break;
+		parent = rule;
+	}
+
+	if (rule && rule->aq_fsp.location == index) {
+		err = aq_add_del_rule(aq_nic, rule, false);
+		hlist_del(&rule->aq_node);
+		kfree(rule);
+		--rx_fltrs->active_filters;
+	}
+
+	if (unlikely(!aq_rx_fltr))
+		return err;
+
+	INIT_HLIST_NODE(&aq_rx_fltr->aq_node);
+
+	if (parent)
+		hlist_add_behind(&aq_rx_fltr->aq_node, &parent->aq_node);
+	else
+		hlist_add_head(&aq_rx_fltr->aq_node, &rx_fltrs->filter_list);
+
+	++rx_fltrs->active_filters;
+
+	return 0;
+}
+
+u16 aq_get_rxnfc_count_all_rules(struct aq_nic_s *aq_nic)
+{
+	struct aq_hw_rx_fltrs_s *rx_fltrs = aq_get_hw_rx_fltrs(aq_nic);
+
+	return rx_fltrs->active_filters;
+}
+
+struct aq_hw_rx_fltrs_s *aq_get_hw_rx_fltrs(struct aq_nic_s *aq_nic)
+{
+	return &aq_nic->aq_hw_rx_fltrs;
+}
+
+int aq_add_rxnfc_rule(struct aq_nic_s *aq_nic, const struct ethtool_rxnfc *cmd)
+{
+	struct aq_hw_rx_fltrs_s *rx_fltrs = aq_get_hw_rx_fltrs(aq_nic);
+	struct ethtool_rx_flow_spec *fsp =
+		(struct ethtool_rx_flow_spec *)&cmd->fs;
+	struct aq_rx_filter *aq_rx_fltr;
+	int err = 0;
+
+	err = aq_check_rule(aq_nic, fsp);
+	if (err)
+		goto err_exit;
+
+	aq_rx_fltr = kzalloc(sizeof(*aq_rx_fltr), GFP_KERNEL);
+	if (unlikely(!aq_rx_fltr)) {
+		err = -ENOMEM;
+		goto err_exit;
+	}
+
+	memcpy(&aq_rx_fltr->aq_fsp, fsp, sizeof(*fsp));
+
+	err = aq_update_table_filters(aq_nic, aq_rx_fltr, fsp->location, NULL);
+	if (unlikely(err))
+		goto err_free;
+
+	err = aq_add_del_rule(aq_nic, aq_rx_fltr, true);
+	if (unlikely(err)) {
+		hlist_del(&aq_rx_fltr->aq_node);
+		--rx_fltrs->active_filters;
+		goto err_free;
+	}
+
+	return 0;
+
+err_free:
+	kfree(aq_rx_fltr);
+err_exit:
+	return err;
+}
+
+int aq_del_rxnfc_rule(struct aq_nic_s *aq_nic, const struct ethtool_rxnfc *cmd)
+{
+	struct aq_hw_rx_fltrs_s *rx_fltrs = aq_get_hw_rx_fltrs(aq_nic);
+	struct aq_rx_filter *rule = NULL;
+	struct hlist_node *aq_node2;
+	int err = -EINVAL;
+
+	hlist_for_each_entry_safe(rule, aq_node2,
+				  &rx_fltrs->filter_list, aq_node) {
+		if (rule->aq_fsp.location == cmd->fs.location)
+			break;
+	}
+
+	if (rule && rule->aq_fsp.location == cmd->fs.location) {
+		err = aq_add_del_rule(aq_nic, rule, false);
+		hlist_del(&rule->aq_node);
+		kfree(rule);
+		--rx_fltrs->active_filters;
+	}
+	return err;
+}
+
+int aq_get_rxnfc_rule(struct aq_nic_s *aq_nic, struct ethtool_rxnfc *cmd)
+{
+	struct aq_hw_rx_fltrs_s *rx_fltrs = aq_get_hw_rx_fltrs(aq_nic);
+	struct ethtool_rx_flow_spec *fsp =
+			(struct ethtool_rx_flow_spec *)&cmd->fs;
+	struct aq_rx_filter *rule = NULL;
+	struct hlist_node *aq_node2;
+
+	hlist_for_each_entry_safe(rule, aq_node2,
+				  &rx_fltrs->filter_list, aq_node)
+		if (fsp->location <= rule->aq_fsp.location)
+			break;
+
+	if (unlikely(!rule || fsp->location != rule->aq_fsp.location))
+		return -EINVAL;
+
+	memcpy(fsp, &rule->aq_fsp, sizeof(*fsp));
+
+	return 0;
+}
+
+int aq_get_rxnfc_all_rules(struct aq_nic_s *aq_nic, struct ethtool_rxnfc *cmd,
+			   u32 *rule_locs)
+{
+	struct aq_hw_rx_fltrs_s *rx_fltrs = aq_get_hw_rx_fltrs(aq_nic);
+	struct hlist_node *aq_node2;
+	struct aq_rx_filter *rule;
+	int count = 0;
+
+	cmd->data = aq_get_rxnfc_count_all_rules(aq_nic);
+
+	hlist_for_each_entry_safe(rule, aq_node2,
+				  &rx_fltrs->filter_list, aq_node) {
+		if (unlikely(count == cmd->rule_cnt))
+			return -EMSGSIZE;
+
+		rule_locs[count++] = rule->aq_fsp.location;
+	}
+
+	cmd->rule_cnt = count;
+
+	return 0;
+}
+
+int aq_clear_rxnfc_all_rules(struct aq_nic_s *aq_nic)
+{
+	struct aq_hw_rx_fltrs_s *rx_fltrs = aq_get_hw_rx_fltrs(aq_nic);
+	struct hlist_node *aq_node2;
+	struct aq_rx_filter *rule;
+	int err = 0;
+
+	hlist_for_each_entry_safe(rule, aq_node2,
+				  &rx_fltrs->filter_list, aq_node) {
+		err = aq_add_del_rule(aq_nic, rule, false);
+		if (err)
+			goto err_exit;
+		hlist_del(&rule->aq_node);
+		kfree(rule);
+		--rx_fltrs->active_filters;
+	}
+
+err_exit:
+	return err;
+}
+
+int aq_reapply_rxnfc_all_rules(struct aq_nic_s *aq_nic)
+{
+	struct aq_hw_rx_fltrs_s *rx_fltrs = aq_get_hw_rx_fltrs(aq_nic);
+	struct hlist_node *aq_node2;
+	struct aq_rx_filter *rule;
+	int err = 0;
+
+	hlist_for_each_entry_safe(rule, aq_node2,
+				  &rx_fltrs->filter_list, aq_node) {
+		err = aq_add_del_rule(aq_nic, rule, true);
+		if (err)
+			goto err_exit;
+	}
+
+err_exit:
+	return err;
+}
+
+int aq_filters_vlans_update(struct aq_nic_s *aq_nic)
+{
+	const struct aq_hw_ops *aq_hw_ops = aq_nic->aq_hw_ops;
+	struct aq_hw_s *aq_hw = aq_nic->aq_hw;
+	int hweight = 0;
+	int err = 0;
+
+	if (unlikely(!aq_hw_ops->hw_filter_vlan_set))
+		return -EOPNOTSUPP;
+	if (unlikely(!aq_hw_ops->hw_filter_vlan_ctrl))
+		return -EOPNOTSUPP;
+
+	aq_fvlan_rebuild(aq_nic, aq_nic->active_vlans,
+			 aq_nic->aq_hw_rx_fltrs.fl2.aq_vlans);
+
+	if (aq_nic->ndev->features & NETIF_F_HW_VLAN_CTAG_FILTER) {
+		hweight = bitmap_weight(aq_nic->active_vlans, VLAN_N_VID);
+
+		err = aq_hw_ops->hw_filter_vlan_ctrl(aq_hw, false);
+		if (err)
+			return err;
+	}
+
+	err = aq_hw_ops->hw_filter_vlan_set(aq_hw,
+					    aq_nic->aq_hw_rx_fltrs.fl2.aq_vlans
+					   );
+	if (err)
+		return err;
+
+	if (aq_nic->ndev->features & NETIF_F_HW_VLAN_CTAG_FILTER) {
+		if (hweight <= AQ_VLAN_MAX_FILTERS && hweight > 0) {
+			err = aq_hw_ops->hw_filter_vlan_ctrl(aq_hw,
+				!(aq_nic->packet_filter & IFF_PROMISC));
+			aq_nic->aq_nic_cfg.is_vlan_force_promisc = false;
+		} else {
+		/* otherwise left in promiscue mode */
+			aq_nic->aq_nic_cfg.is_vlan_force_promisc = true;
+		}
+	}
+
+	return err;
+}
+
+int aq_filters_vlan_offload_off(struct aq_nic_s *aq_nic)
+{
+	const struct aq_hw_ops *aq_hw_ops = aq_nic->aq_hw_ops;
+	struct aq_hw_s *aq_hw = aq_nic->aq_hw;
+	int err = 0;
+
+	bitmap_zero(aq_nic->active_vlans, VLAN_N_VID);
+	aq_fvlan_rebuild(aq_nic, aq_nic->active_vlans,
+			 aq_nic->aq_hw_rx_fltrs.fl2.aq_vlans);
+
+	if (unlikely(!aq_hw_ops->hw_filter_vlan_set))
+		return -EOPNOTSUPP;
+	if (unlikely(!aq_hw_ops->hw_filter_vlan_ctrl))
+		return -EOPNOTSUPP;
+
+	aq_nic->aq_nic_cfg.is_vlan_force_promisc = true;
+	err = aq_hw_ops->hw_filter_vlan_ctrl(aq_hw, false);
+	if (err)
+		return err;
+	err = aq_hw_ops->hw_filter_vlan_set(aq_hw,
+					    aq_nic->aq_hw_rx_fltrs.fl2.aq_vlans
+					   );
+	return err;
+}
diff --git a/drivers/net/ethernet/aquantia/atlantic/aq_filters.h b/drivers/net/ethernet/aquantia/atlantic/aq_filters.h
new file mode 100644
index 000000000000..122e06c88a33
--- /dev/null
+++ b/drivers/net/ethernet/aquantia/atlantic/aq_filters.h
@@ -0,0 +1,36 @@
+/* SPDX-License-Identifier: GPL-2.0-only */
+/* Copyright (C) 2014-2017 aQuantia Corporation. */
+
+/* File aq_filters.h: RX filters related functions. */
+
+#ifndef AQ_FILTERS_H
+#define AQ_FILTERS_H
+
+#include "aq_nic.h"
+
+enum aq_rx_filter_type {
+	aq_rx_filter_ethertype,
+	aq_rx_filter_vlan,
+	aq_rx_filter_l3l4
+};
+
+struct aq_rx_filter {
+	struct hlist_node aq_node;
+	enum aq_rx_filter_type type;
+	struct ethtool_rx_flow_spec aq_fsp;
+};
+
+u16 aq_get_rxnfc_count_all_rules(struct aq_nic_s *aq_nic);
+struct aq_hw_rx_fltrs_s *aq_get_hw_rx_fltrs(struct aq_nic_s *aq_nic);
+int aq_add_rxnfc_rule(struct aq_nic_s *aq_nic, const struct ethtool_rxnfc *cmd);
+int aq_del_rxnfc_rule(struct aq_nic_s *aq_nic, const struct ethtool_rxnfc *cmd);
+int aq_get_rxnfc_rule(struct aq_nic_s *aq_nic, struct ethtool_rxnfc *cmd);
+int aq_get_rxnfc_all_rules(struct aq_nic_s *aq_nic, struct ethtool_rxnfc *cmd,
+			   u32 *rule_locs);
+int aq_del_fvlan_by_vlan(struct aq_nic_s *aq_nic, u16 vlan_id);
+int aq_clear_rxnfc_all_rules(struct aq_nic_s *aq_nic);
+int aq_reapply_rxnfc_all_rules(struct aq_nic_s *aq_nic);
+int aq_filters_vlans_update(struct aq_nic_s *aq_nic);
+int aq_filters_vlan_offload_off(struct aq_nic_s *aq_nic);
+
+#endif /* AQ_FILTERS_H */
diff --git a/drivers/net/ethernet/aquantia/atlantic/aq_hw.h b/drivers/net/ethernet/aquantia/atlantic/aq_hw.h
index fce0fd3f23ff..4e66fd9b2ab1 100644
--- a/drivers/net/ethernet/aquantia/atlantic/aq_hw.h
+++ b/drivers/net/ethernet/aquantia/atlantic/aq_hw.h
@@ -1,13 +1,11 @@
-/*
- * aQuantia Corporation Network Driver
- * Copyright (C) 2014-2017 aQuantia Corporation. All rights reserved
+/* SPDX-License-Identifier: GPL-2.0-only */
+/* Atlantic Network Driver
  *
- * This program is free software; you can redistribute it and/or modify it
- * under the terms and conditions of the GNU General Public License,
- * version 2, as published by the Free Software Foundation.
+ * Copyright (C) 2014-2019 aQuantia Corporation
+ * Copyright (C) 2019-2020 Marvell International Ltd.
  */
 
-/* File aq_hw.h: Declaraion of abstract interface for NIC hardware specific
+/* File aq_hw.h: Declaration of abstract interface for NIC hardware specific
  * functions.
  */
 
@@ -15,22 +13,52 @@
 #define AQ_HW_H
 
 #include "aq_common.h"
+#include "aq_rss.h"
+#include "hw_atl/hw_atl_utils.h"
+
+#define AQ_HW_MAC_COUNTER_HZ   312500000ll
+#define AQ_HW_PHY_COUNTER_HZ   160000000ll
+
+enum aq_tc_mode {
+	AQ_TC_MODE_INVALID = -1,
+	AQ_TC_MODE_8TCS,
+	AQ_TC_MODE_4TCS,
+};
+
+#define AQ_RX_FIRST_LOC_FVLANID     0U
+#define AQ_RX_LAST_LOC_FVLANID	   15U
+#define AQ_RX_FIRST_LOC_FETHERT    16U
+#define AQ_RX_LAST_LOC_FETHERT	   31U
+#define AQ_RX_FIRST_LOC_FL3L4	   32U
+#define AQ_RX_LAST_LOC_FL3L4	   39U
+#define AQ_RX_MAX_RXNFC_LOC	   AQ_RX_LAST_LOC_FL3L4
+#define AQ_VLAN_MAX_FILTERS   \
+			(AQ_RX_LAST_LOC_FVLANID - AQ_RX_FIRST_LOC_FVLANID + 1U)
+#define AQ_RX_QUEUE_NOT_ASSIGNED   0xFFU
+
+#define AQ_FRAC_PER_NS 0x100000000LL
+
+/* Used for rate to Mbps conversion */
+#define AQ_MBPS_DIVISOR         125000 /* 1000000 / 8 */
 
 /* NIC H/W capabilities */
 struct aq_hw_caps_s {
 	u64 hw_features;
 	u64 link_speed_msk;
 	unsigned int hw_priv_flags;
-	u32 rxds;
-	u32 txds;
+	u32 media_type;
+	u32 rxds_max;
+	u32 txds_max;
+	u32 rxds_min;
+	u32 txds_min;
 	u32 txhwb_alignment;
 	u32 irq_mask;
 	u32 vecs;
 	u32 mtu;
 	u32 mac_regs_count;
-	u8 ports;
+	u32 hw_alive_check_addr;
 	u8 msix_irqs;
-	u8 tcs;
+	u8 tcs_max;
 	u8 rxd_alignment;
 	u8 rxd_size;
 	u8 txd_alignment;
@@ -39,53 +67,147 @@ struct aq_hw_caps_s {
 	u8 rx_rings;
 	bool flow_control;
 	bool is_64_dma;
-	u32 fw_ver_expected;
+	bool op64bit;
+	u32 quirks;
+	u32 priv_data_len;
 };
 
 struct aq_hw_link_status_s {
 	unsigned int mbps;
+	bool full_duplex;
+	u32 lp_link_speed_msk;
+	u32 lp_flow_control;
+};
+
+struct aq_stats_s {
+	u64 brc;
+	u64 btc;
+	u64 uprc;
+	u64 mprc;
+	u64 bprc;
+	u64 erpt;
+	u64 uptc;
+	u64 mptc;
+	u64 bptc;
+	u64 erpr;
+	u64 mbtc;
+	u64 bbtc;
+	u64 mbrc;
+	u64 bbrc;
+	u64 ubrc;
+	u64 ubtc;
+	u64 dpc;
+	u64 dma_pkt_rc;
+	u64 dma_pkt_tc;
+	u64 dma_oct_rc;
+	u64 dma_oct_tc;
 };
 
 #define AQ_HW_IRQ_INVALID 0U
-#define AQ_HW_IRQ_LEGACY  1U
+#define AQ_HW_IRQ_INTX	  1U
 #define AQ_HW_IRQ_MSI     2U
 #define AQ_HW_IRQ_MSIX    3U
 
+#define AQ_HW_SERVICE_IRQS   1U
+
 #define AQ_HW_POWER_STATE_D0   0U
 #define AQ_HW_POWER_STATE_D3   3U
 
+#define	AQ_FW_WAKE_ON_LINK_RTPM BIT(10)
+
 #define AQ_HW_FLAG_STARTED     0x00000004U
 #define AQ_HW_FLAG_STOPPING    0x00000008U
 #define AQ_HW_FLAG_RESETTING   0x00000010U
 #define AQ_HW_FLAG_CLOSING     0x00000020U
+#define AQ_HW_PTP_AVAILABLE    0x01000000U
 #define AQ_HW_LINK_DOWN        0x04000000U
 #define AQ_HW_FLAG_ERR_UNPLUG  0x40000000U
 #define AQ_HW_FLAG_ERR_HW      0x80000000U
 
 #define AQ_HW_FLAG_ERRORS      (AQ_HW_FLAG_ERR_HW | AQ_HW_FLAG_ERR_UNPLUG)
 
+#define AQ_NIC_FLAGS_IS_NOT_READY (AQ_NIC_FLAG_STOPPING | \
+			AQ_NIC_FLAG_RESETTING | AQ_NIC_FLAG_CLOSING | \
+			AQ_NIC_FLAG_ERR_UNPLUG | AQ_NIC_FLAG_ERR_HW)
+
+#define AQ_NIC_FLAGS_IS_NOT_TX_READY (AQ_NIC_FLAGS_IS_NOT_READY | \
+					AQ_NIC_LINK_DOWN)
+
+#define AQ_HW_MEDIA_TYPE_TP    1U
+#define AQ_HW_MEDIA_TYPE_FIBRE 2U
+
+#define AQ_HW_TXD_MULTIPLE 8U
+#define AQ_HW_RXD_MULTIPLE 8U
+
+#define AQ_HW_QUEUES_MAX                32U
+#define AQ_HW_MULTICAST_ADDRESS_MAX     32U
+
+#define AQ_HW_PTP_TC                    2U
+
+#define AQ_HW_LED_BLINK    0x2U
+#define AQ_HW_LED_DEFAULT  0x0U
+
+#define AQ_HW_MEDIA_DETECT_CNT 6000
+
+enum aq_priv_flags {
+	AQ_HW_LOOPBACK_DMA_SYS,
+	AQ_HW_LOOPBACK_PKT_SYS,
+	AQ_HW_LOOPBACK_DMA_NET,
+	AQ_HW_LOOPBACK_PHYINT_SYS,
+	AQ_HW_LOOPBACK_PHYEXT_SYS,
+};
+
+#define AQ_HW_LOOPBACK_MASK	(BIT(AQ_HW_LOOPBACK_DMA_SYS) |\
+				 BIT(AQ_HW_LOOPBACK_PKT_SYS) |\
+				 BIT(AQ_HW_LOOPBACK_DMA_NET) |\
+				 BIT(AQ_HW_LOOPBACK_PHYINT_SYS) |\
+				 BIT(AQ_HW_LOOPBACK_PHYEXT_SYS))
+
+#define ATL_HW_CHIP_MIPS         0x00000001U
+#define ATL_HW_CHIP_TPO2         0x00000002U
+#define ATL_HW_CHIP_RPF2         0x00000004U
+#define ATL_HW_CHIP_MPI_AQ       0x00000010U
+#define ATL_HW_CHIP_ATLANTIC     0x00800000U
+#define ATL_HW_CHIP_REVISION_A0  0x01000000U
+#define ATL_HW_CHIP_REVISION_B0  0x02000000U
+#define ATL_HW_CHIP_REVISION_B1  0x04000000U
+#define ATL_HW_CHIP_ANTIGUA      0x08000000U
+
+#define ATL_HW_IS_CHIP_FEATURE(_HW_, _F_) (!!(ATL_HW_CHIP_##_F_ & \
+	(_HW_)->chip_features))
+
 struct aq_hw_s {
-	struct aq_obj_s header;
+	atomic_t flags;
+	u8 rbl_enabled:1;
 	struct aq_nic_cfg_s *aq_nic_cfg;
-	struct aq_pci_func_s *aq_pci_func;
+	const struct aq_fw_ops *aq_fw_ops;
 	void __iomem *mmio;
-	unsigned int not_ff_addr;
 	struct aq_hw_link_status_s aq_link_status;
+	struct hw_atl_utils_mbox mbox;
+	struct hw_atl_stats_s last_stats;
+	struct aq_stats_s curr_stats;
+	u64 speed;
+	u32 itr_tx;
+	u32 itr_rx;
+	unsigned int chip_features;
+	u32 fw_ver_actual;
+	atomic_t dpc;
+	u32 mbox_addr;
+	u32 rpc_addr;
+	u32 settings_addr;
+	u32 rpc_tid;
+	struct hw_atl_utils_fw_rpc rpc;
+	s64 ptp_clk_offset;
+	u16 phy_id;
+	void *priv;
 };
 
 struct aq_ring_s;
 struct aq_ring_param_s;
-struct aq_nic_cfg_s;
 struct sk_buff;
+struct aq_rx_filter_l3l4;
 
 struct aq_hw_ops {
-	struct aq_hw_s *(*create)(struct aq_pci_func_s *aq_pci_func,
-				  unsigned int port, struct aq_hw_ops *ops);
-
-	void (*destroy)(struct aq_hw_s *self);
-
-	int (*get_hw_caps)(struct aq_hw_s *self,
-			   struct aq_hw_caps_s *aq_hw_caps);
 
 	int (*hw_ring_tx_xmit)(struct aq_hw_s *self, struct aq_ring_s *aq_ring,
 			       unsigned int frags);
@@ -99,21 +221,16 @@ struct aq_hw_ops {
 	int (*hw_ring_tx_head_update)(struct aq_hw_s *self,
 				      struct aq_ring_s *aq_ring);
 
-	int (*hw_get_mac_permanent)(struct aq_hw_s *self,
-				    struct aq_hw_caps_s *aq_hw_caps,
-				    u8 *mac);
+	int (*hw_set_mac_address)(struct aq_hw_s *self, const u8 *mac_addr);
 
-	int (*hw_set_mac_address)(struct aq_hw_s *self, u8 *mac_addr);
+	int (*hw_soft_reset)(struct aq_hw_s *self);
 
-	int (*hw_get_link_status)(struct aq_hw_s *self,
-				  struct aq_hw_link_status_s *link_status);
-
-	int (*hw_set_link_speed)(struct aq_hw_s *self, u32 speed);
+	int (*hw_prepare)(struct aq_hw_s *self,
+			  const struct aq_fw_ops **fw_ops);
 
 	int (*hw_reset)(struct aq_hw_s *self);
 
-	int (*hw_init)(struct aq_hw_s *self, struct aq_nic_cfg_s *aq_nic_cfg,
-		       u8 *mac_addr);
+	int (*hw_init)(struct aq_hw_s *self, const u8 *mac_addr);
 
 	int (*hw_start)(struct aq_hw_s *self);
 
@@ -147,13 +264,29 @@ struct aq_hw_ops {
 	int (*hw_packet_filter_set)(struct aq_hw_s *self,
 				    unsigned int packet_filter);
 
+	int (*hw_filter_l3l4_set)(struct aq_hw_s *self,
+				  struct aq_rx_filter_l3l4 *data);
+
+	int (*hw_filter_l3l4_clear)(struct aq_hw_s *self,
+				    struct aq_rx_filter_l3l4 *data);
+
+	int (*hw_filter_l2_set)(struct aq_hw_s *self,
+				struct aq_rx_filter_l2 *data);
+
+	int (*hw_filter_l2_clear)(struct aq_hw_s *self,
+				  struct aq_rx_filter_l2 *data);
+
+	int (*hw_filter_vlan_set)(struct aq_hw_s *self,
+				  struct aq_rx_filter_vlan *aq_vlans);
+
+	int (*hw_filter_vlan_ctrl)(struct aq_hw_s *self, bool enable);
+
 	int (*hw_multicast_list_set)(struct aq_hw_s *self,
-				     u8 ar_mac[AQ_CFG_MULTICAST_ADDRESS_MAX]
+				     u8 ar_mac[AQ_HW_MULTICAST_ADDRESS_MAX]
 				     [ETH_ALEN],
 				     u32 count);
 
-	int (*hw_interrupt_moderation_set)(struct aq_hw_s *self,
-					   bool itr_enabled);
+	int (*hw_interrupt_moderation_set)(struct aq_hw_s *self);
 
 	int (*hw_rss_set)(struct aq_hw_s *self,
 			  struct aq_rss_parameters *rss_params);
@@ -161,17 +294,116 @@ struct aq_hw_ops {
 	int (*hw_rss_hash_set)(struct aq_hw_s *self,
 			       struct aq_rss_parameters *rss_params);
 
+	int (*hw_tc_rate_limit_set)(struct aq_hw_s *self);
+
 	int (*hw_get_regs)(struct aq_hw_s *self,
-			   struct aq_hw_caps_s *aq_hw_caps, u32 *regs_buff);
+			   const struct aq_hw_caps_s *aq_hw_caps,
+			   u32 *regs_buff);
+
+	struct aq_stats_s *(*hw_get_hw_stats)(struct aq_hw_s *self);
+
+	u32 (*hw_get_fw_version)(struct aq_hw_s *self);
+
+	int (*hw_set_offload)(struct aq_hw_s *self,
+			      struct aq_nic_cfg_s *aq_nic_cfg);
+
+	int (*hw_ring_hwts_rx_fill)(struct aq_hw_s *self,
+				    struct aq_ring_s *aq_ring);
+
+	int (*hw_ring_hwts_rx_receive)(struct aq_hw_s *self,
+				       struct aq_ring_s *ring);
+
+	void (*hw_get_ptp_ts)(struct aq_hw_s *self, u64 *stamp);
+
+	int (*hw_adj_clock_freq)(struct aq_hw_s *self, s32 delta);
+
+	int (*hw_adj_sys_clock)(struct aq_hw_s *self, s64 delta);
+
+	int (*hw_set_sys_clock)(struct aq_hw_s *self, u64 time, u64 ts);
+
+	int (*hw_ts_to_sys_clock)(struct aq_hw_s *self, u64 ts, u64 *time);
+
+	int (*hw_gpio_pulse)(struct aq_hw_s *self, u32 index, u64 start,
+			     u32 period);
+
+	int (*hw_extts_gpio_enable)(struct aq_hw_s *self, u32 index,
+				    u32 enable);
+
+	int (*hw_get_sync_ts)(struct aq_hw_s *self, u64 *ts);
+
+	u16 (*rx_extract_ts)(struct aq_hw_s *self, u8 *p, unsigned int len,
+			     u64 *timestamp);
+
+	int (*extract_hwts)(struct aq_hw_s *self, u8 *p, unsigned int len,
+			    u64 *timestamp);
+
+	int (*hw_set_fc)(struct aq_hw_s *self, u32 fc, u32 tc);
+
+	int (*hw_set_loopback)(struct aq_hw_s *self, u32 mode, bool enable);
+
+	int (*hw_get_mac_temp)(struct aq_hw_s *self, u32 *temp);
+
+	int (*hw_read_module_eeprom)(struct aq_hw_s *self, u8 dev_addr,
+				     u8 reg_start_addr, int len, u8 *data);
+};
+
+struct aq_fw_ops {
+	int (*init)(struct aq_hw_s *self);
+
+	int (*deinit)(struct aq_hw_s *self);
+
+	int (*reset)(struct aq_hw_s *self);
+
+	int (*renegotiate)(struct aq_hw_s *self);
+
+	int (*get_mac_permanent)(struct aq_hw_s *self, u8 *mac);
+
+	int (*set_link_speed)(struct aq_hw_s *self, u32 speed);
+
+	int (*set_state)(struct aq_hw_s *self,
+			 enum hal_atl_utils_fw_state_e state);
+
+	int (*update_link_status)(struct aq_hw_s *self);
+
+	int (*update_stats)(struct aq_hw_s *self);
+
+	int (*get_mac_temp)(struct aq_hw_s *self, int *temp);
+
+	int (*get_phy_temp)(struct aq_hw_s *self, int *temp);
+
+	u32 (*get_flow_control)(struct aq_hw_s *self, u32 *fcmode);
+
+	int (*set_flow_control)(struct aq_hw_s *self);
+
+	int (*led_control)(struct aq_hw_s *self, u32 mode);
+
+	int (*set_phyloopback)(struct aq_hw_s *self, u32 mode, bool enable);
+
+	int (*set_power)(struct aq_hw_s *self, unsigned int power_state,
+			 const u8 *mac);
+
+	int (*send_fw_request)(struct aq_hw_s *self,
+			       const struct hw_fw_request_iface *fw_req,
+			       size_t size);
+
+	void (*enable_ptp)(struct aq_hw_s *self, int enable);
+
+	void (*adjust_ptp)(struct aq_hw_s *self, uint64_t adj);
+
+	int (*set_eee_rate)(struct aq_hw_s *self, u32 speed);
+
+	int (*get_eee_rate)(struct aq_hw_s *self, u32 *rate,
+			    u32 *supported_rates);
 
-	int (*hw_get_hw_stats)(struct aq_hw_s *self, u64 *data,
-			       unsigned int *p_count);
+	int (*set_downshift)(struct aq_hw_s *self, u32 counter);
 
-	int (*hw_get_fw_version)(struct aq_hw_s *self, u32 *fw_version);
+	int (*set_media_detect)(struct aq_hw_s *self, bool enable);
 
-	int (*hw_deinit)(struct aq_hw_s *self);
+	u32 (*get_link_capabilities)(struct aq_hw_s *self);
 
-	int (*hw_set_power)(struct aq_hw_s *self, unsigned int power_state);
+	int (*send_macsec_req)(struct aq_hw_s *self,
+			       struct macsec_msg_fw_request *msg,
+			       struct macsec_msg_fw_response *resp);
 };
 
 #endif /* AQ_HW_H */
diff --git a/drivers/net/ethernet/aquantia/atlantic/aq_hw_utils.c b/drivers/net/ethernet/aquantia/atlantic/aq_hw_utils.c
index 5f13465995f6..1921741f7311 100644
--- a/drivers/net/ethernet/aquantia/atlantic/aq_hw_utils.c
+++ b/drivers/net/ethernet/aquantia/atlantic/aq_hw_utils.c
@@ -1,10 +1,8 @@
-/*
- * aQuantia Corporation Network Driver
- * Copyright (C) 2014-2017 aQuantia Corporation. All rights reserved
+// SPDX-License-Identifier: GPL-2.0-only
+/* Atlantic Network Driver
  *
- * This program is free software; you can redistribute it and/or modify it
- * under the terms and conditions of the GNU General Public License,
- * version 2, as published by the Free Software Foundation.
+ * Copyright (C) 2014-2019 aQuantia Corporation
+ * Copyright (C) 2019-2020 Marvell International Ltd.
  */
 
 /* File aq_hw_utils.c: Definitions of helper functions used across
@@ -12,7 +10,11 @@
  */
 
 #include "aq_hw_utils.h"
+
+#include <linux/io-64-nonatomic-lo-hi.h>
+
 #include "aq_hw.h"
+#include "aq_nic.h"
 
 void aq_hw_write_reg_bit(struct aq_hw_s *aq_hw, u32 addr, u32 msk,
 			 u32 shift, u32 val)
@@ -39,8 +41,9 @@ u32 aq_hw_read_reg(struct aq_hw_s *hw, u32 reg)
 {
 	u32 value = readl(hw->mmio + reg);
 
-	if ((~0U) == value && (~0U) == readl(hw->mmio + hw->not_ff_addr))
-		aq_utils_obj_set(&hw->header.flags, AQ_HW_FLAG_ERR_UNPLUG);
+	if (value == U32_MAX &&
+	    readl(hw->mmio + hw->aq_nic_cfg->aq_hw_caps->hw_alive_check_addr) == U32_MAX)
+		aq_utils_obj_set(&hw->flags, AQ_HW_FLAG_ERR_UNPLUG);
 
 	return value;
 }
@@ -50,15 +53,43 @@ void aq_hw_write_reg(struct aq_hw_s *hw, u32 reg, u32 value)
 	writel(value, hw->mmio + reg);
 }
 
+/* Most of 64-bit registers are in LSW, MSW form.
+   Counters are normally implemented by HW as latched pairs:
+   reading LSW first locks MSW, to overcome LSW overflow
+ */
+u64 aq_hw_read_reg64(struct aq_hw_s *hw, u32 reg)
+{
+	u64 value = U64_MAX;
+
+	if (hw->aq_nic_cfg->aq_hw_caps->op64bit)
+		value = readq(hw->mmio + reg);
+	else
+		value = lo_hi_readq(hw->mmio + reg);
+
+	if (value == U64_MAX &&
+	    readl(hw->mmio + hw->aq_nic_cfg->aq_hw_caps->hw_alive_check_addr) == U32_MAX)
+		aq_utils_obj_set(&hw->flags, AQ_HW_FLAG_ERR_UNPLUG);
+
+	return value;
+}
+
+void aq_hw_write_reg64(struct aq_hw_s *hw, u32 reg, u64 value)
+{
+	if (hw->aq_nic_cfg->aq_hw_caps->op64bit)
+		writeq(value, hw->mmio + reg);
+	else
+		lo_hi_writeq(value, hw->mmio + reg);
+}
+
 int aq_hw_err_from_flags(struct aq_hw_s *hw)
 {
 	int err = 0;
 
-	if (aq_utils_obj_test(&hw->header.flags, AQ_HW_FLAG_ERR_UNPLUG)) {
+	if (aq_utils_obj_test(&hw->flags, AQ_HW_FLAG_ERR_UNPLUG)) {
 		err = -ENXIO;
 		goto err_exit;
 	}
-	if (aq_utils_obj_test(&hw->header.flags, AQ_HW_FLAG_ERR_HW)) {
+	if (aq_utils_obj_test(&hw->flags, AQ_HW_FLAG_ERR_HW)) {
 		err = -EIO;
 		goto err_exit;
 	}
@@ -66,3 +97,29 @@ int aq_hw_err_from_flags(struct aq_hw_s *hw)
 err_exit:
 	return err;
 }
+
+int aq_hw_num_tcs(struct aq_hw_s *hw)
+{
+	switch (hw->aq_nic_cfg->tc_mode) {
+	case AQ_TC_MODE_8TCS:
+		return 8;
+	case AQ_TC_MODE_4TCS:
+		return 4;
+	default:
+		break;
+	}
+
+	return 1;
+}
+
+int aq_hw_q_per_tc(struct aq_hw_s *hw)
+{
+	switch (hw->aq_nic_cfg->tc_mode) {
+	case AQ_TC_MODE_8TCS:
+		return 4;
+	case AQ_TC_MODE_4TCS:
+		return 8;
+	default:
+		return 4;
+	}
+}
diff --git a/drivers/net/ethernet/aquantia/atlantic/aq_hw_utils.h b/drivers/net/ethernet/aquantia/atlantic/aq_hw_utils.h
index 03b72ddbffb9..ffa6e4067c21 100644
--- a/drivers/net/ethernet/aquantia/atlantic/aq_hw_utils.h
+++ b/drivers/net/ethernet/aquantia/atlantic/aq_hw_utils.h
@@ -1,10 +1,8 @@
-/*
- * aQuantia Corporation Network Driver
- * Copyright (C) 2014-2017 aQuantia Corporation. All rights reserved
+/* SPDX-License-Identifier: GPL-2.0-only */
+/* Atlantic Network Driver
  *
- * This program is free software; you can redistribute it and/or modify it
- * under the terms and conditions of the GNU General Public License,
- * version 2, as published by the Free Software Foundation.
+ * Copyright (C) 2014-2019 aQuantia Corporation
+ * Copyright (C) 2019-2020 Marvell International Ltd.
  */
 
 /* File aq_hw_utils.h: Declaration of helper functions used across hardware
@@ -14,6 +12,8 @@
 #ifndef AQ_HW_UTILS_H
 #define AQ_HW_UTILS_H
 
+#include <linux/iopoll.h>
+
 #include "aq_common.h"
 
 #ifndef HIDWORD
@@ -23,17 +23,8 @@
 
 #define AQ_HW_SLEEP(_US_) mdelay(_US_)
 
-#define AQ_HW_WAIT_FOR(_B_, _US_, _N_) \
-do { \
-	unsigned int AQ_HW_WAIT_FOR_i; \
-	for (AQ_HW_WAIT_FOR_i = _N_; (!(_B_)) && (AQ_HW_WAIT_FOR_i);\
-	--AQ_HW_WAIT_FOR_i) {\
-		udelay(_US_); \
-	} \
-	if (!AQ_HW_WAIT_FOR_i) {\
-		err = -ETIME; \
-	} \
-} while (0)
+#define aq_pr_err(...) pr_err(AQ_CFG_DRV_NAME ": " __VA_ARGS__)
+#define aq_pr_trace(...) pr_info(AQ_CFG_DRV_NAME ": " __VA_ARGS__)
 
 struct aq_hw_s;
 
@@ -42,6 +33,10 @@ void aq_hw_write_reg_bit(struct aq_hw_s *aq_hw, u32 addr, u32 msk,
 u32 aq_hw_read_reg_bit(struct aq_hw_s *aq_hw, u32 addr, u32 msk, u32 shift);
 u32 aq_hw_read_reg(struct aq_hw_s *hw, u32 reg);
 void aq_hw_write_reg(struct aq_hw_s *hw, u32 reg, u32 value);
+u64 aq_hw_read_reg64(struct aq_hw_s *hw, u32 reg);
+void aq_hw_write_reg64(struct aq_hw_s *hw, u32 reg, u64 value);
 int aq_hw_err_from_flags(struct aq_hw_s *hw);
+int aq_hw_num_tcs(struct aq_hw_s *hw);
+int aq_hw_q_per_tc(struct aq_hw_s *hw);
 
 #endif /* AQ_HW_UTILS_H */
diff --git a/drivers/net/ethernet/aquantia/atlantic/aq_macsec.c b/drivers/net/ethernet/aquantia/atlantic/aq_macsec.c
new file mode 100644
index 000000000000..6afff8af5e86
--- /dev/null
+++ b/drivers/net/ethernet/aquantia/atlantic/aq_macsec.c
@@ -0,0 +1,1770 @@
+// SPDX-License-Identifier: GPL-2.0-only
+/* Atlantic Network Driver
+ * Copyright (C) 2020 Marvell International Ltd.
+ */
+
+#include "aq_macsec.h"
+#include "aq_nic.h"
+#include <linux/rtnetlink.h>
+
+#include "macsec/macsec_api.h"
+#define AQ_MACSEC_KEY_LEN_128_BIT 16
+#define AQ_MACSEC_KEY_LEN_192_BIT 24
+#define AQ_MACSEC_KEY_LEN_256_BIT 32
+
+enum aq_clear_type {
+	/* update HW configuration */
+	AQ_CLEAR_HW = BIT(0),
+	/* update SW configuration (busy bits, pointers) */
+	AQ_CLEAR_SW = BIT(1),
+	/* update both HW and SW configuration */
+	AQ_CLEAR_ALL = AQ_CLEAR_HW | AQ_CLEAR_SW,
+};
+
+static int aq_clear_txsc(struct aq_nic_s *nic, const int txsc_idx,
+			 enum aq_clear_type clear_type);
+static int aq_clear_txsa(struct aq_nic_s *nic, struct aq_macsec_txsc *aq_txsc,
+			 const int sa_num, enum aq_clear_type clear_type);
+static int aq_clear_rxsc(struct aq_nic_s *nic, const int rxsc_idx,
+			 enum aq_clear_type clear_type);
+static int aq_clear_rxsa(struct aq_nic_s *nic, struct aq_macsec_rxsc *aq_rxsc,
+			 const int sa_num, enum aq_clear_type clear_type);
+static int aq_clear_secy(struct aq_nic_s *nic, const struct macsec_secy *secy,
+			 enum aq_clear_type clear_type);
+static int aq_apply_macsec_cfg(struct aq_nic_s *nic);
+static int aq_apply_secy_cfg(struct aq_nic_s *nic,
+			     const struct macsec_secy *secy);
+
+static void aq_ether_addr_to_mac(u32 mac[2], const unsigned char *emac)
+{
+	u32 tmp[2] = { 0 };
+
+	memcpy(((u8 *)tmp) + 2, emac, ETH_ALEN);
+
+	mac[0] = swab32(tmp[1]);
+	mac[1] = swab32(tmp[0]);
+}
+
+/* There's a 1:1 mapping between SecY and TX SC */
+static int aq_get_txsc_idx_from_secy(struct aq_macsec_cfg *macsec_cfg,
+				     const struct macsec_secy *secy)
+{
+	int i;
+
+	if (unlikely(!secy))
+		return -1;
+
+	for (i = 0; i < AQ_MACSEC_MAX_SC; i++) {
+		if (macsec_cfg->aq_txsc[i].sw_secy == secy)
+			return i;
+	}
+	return -1;
+}
+
+static int aq_get_rxsc_idx_from_rxsc(struct aq_macsec_cfg *macsec_cfg,
+				     const struct macsec_rx_sc *rxsc)
+{
+	int i;
+
+	if (unlikely(!rxsc))
+		return -1;
+
+	for (i = 0; i < AQ_MACSEC_MAX_SC; i++) {
+		if (macsec_cfg->aq_rxsc[i].sw_rxsc == rxsc)
+			return i;
+	}
+
+	return -1;
+}
+
+static int aq_get_txsc_idx_from_sc_idx(const enum aq_macsec_sc_sa sc_sa,
+				       const int sc_idx)
+{
+	switch (sc_sa) {
+	case aq_macsec_sa_sc_4sa_8sc:
+		return sc_idx >> 2;
+	case aq_macsec_sa_sc_2sa_16sc:
+		return sc_idx >> 1;
+	case aq_macsec_sa_sc_1sa_32sc:
+		return sc_idx;
+	default:
+		WARN_ONCE(true, "Invalid sc_sa");
+	}
+	return -1;
+}
+
+/* Rotate keys u32[8] */
+static void aq_rotate_keys(u32 (*key)[8], const int key_len)
+{
+	u32 tmp[8] = { 0 };
+
+	memcpy(&tmp, key, sizeof(tmp));
+	memset(*key, 0, sizeof(*key));
+
+	if (key_len == AQ_MACSEC_KEY_LEN_128_BIT) {
+		(*key)[0] = swab32(tmp[3]);
+		(*key)[1] = swab32(tmp[2]);
+		(*key)[2] = swab32(tmp[1]);
+		(*key)[3] = swab32(tmp[0]);
+	} else if (key_len == AQ_MACSEC_KEY_LEN_192_BIT) {
+		(*key)[0] = swab32(tmp[5]);
+		(*key)[1] = swab32(tmp[4]);
+		(*key)[2] = swab32(tmp[3]);
+		(*key)[3] = swab32(tmp[2]);
+		(*key)[4] = swab32(tmp[1]);
+		(*key)[5] = swab32(tmp[0]);
+	} else if (key_len == AQ_MACSEC_KEY_LEN_256_BIT) {
+		(*key)[0] = swab32(tmp[7]);
+		(*key)[1] = swab32(tmp[6]);
+		(*key)[2] = swab32(tmp[5]);
+		(*key)[3] = swab32(tmp[4]);
+		(*key)[4] = swab32(tmp[3]);
+		(*key)[5] = swab32(tmp[2]);
+		(*key)[6] = swab32(tmp[1]);
+		(*key)[7] = swab32(tmp[0]);
+	} else {
+		pr_warn("Rotate_keys: invalid key_len\n");
+	}
+}
+
+#define STATS_2x32_TO_64(stat_field)                                           \
+	(((u64)stat_field[1] << 32) | stat_field[0])
+
+static int aq_get_macsec_common_stats(struct aq_hw_s *hw,
+				      struct aq_macsec_common_stats *stats)
+{
+	struct aq_mss_ingress_common_counters ingress_counters;
+	struct aq_mss_egress_common_counters egress_counters;
+	int ret;
+
+	/* MACSEC counters */
+	ret = aq_mss_get_ingress_common_counters(hw, &ingress_counters);
+	if (unlikely(ret))
+		return ret;
+
+	stats->in.ctl_pkts = STATS_2x32_TO_64(ingress_counters.ctl_pkts);
+	stats->in.tagged_miss_pkts =
+		STATS_2x32_TO_64(ingress_counters.tagged_miss_pkts);
+	stats->in.untagged_miss_pkts =
+		STATS_2x32_TO_64(ingress_counters.untagged_miss_pkts);
+	stats->in.notag_pkts = STATS_2x32_TO_64(ingress_counters.notag_pkts);
+	stats->in.untagged_pkts =
+		STATS_2x32_TO_64(ingress_counters.untagged_pkts);
+	stats->in.bad_tag_pkts =
+		STATS_2x32_TO_64(ingress_counters.bad_tag_pkts);
+	stats->in.no_sci_pkts = STATS_2x32_TO_64(ingress_counters.no_sci_pkts);
+	stats->in.unknown_sci_pkts =
+		STATS_2x32_TO_64(ingress_counters.unknown_sci_pkts);
+	stats->in.ctrl_prt_pass_pkts =
+		STATS_2x32_TO_64(ingress_counters.ctrl_prt_pass_pkts);
+	stats->in.unctrl_prt_pass_pkts =
+		STATS_2x32_TO_64(ingress_counters.unctrl_prt_pass_pkts);
+	stats->in.ctrl_prt_fail_pkts =
+		STATS_2x32_TO_64(ingress_counters.ctrl_prt_fail_pkts);
+	stats->in.unctrl_prt_fail_pkts =
+		STATS_2x32_TO_64(ingress_counters.unctrl_prt_fail_pkts);
+	stats->in.too_long_pkts =
+		STATS_2x32_TO_64(ingress_counters.too_long_pkts);
+	stats->in.igpoc_ctl_pkts =
+		STATS_2x32_TO_64(ingress_counters.igpoc_ctl_pkts);
+	stats->in.ecc_error_pkts =
+		STATS_2x32_TO_64(ingress_counters.ecc_error_pkts);
+	stats->in.unctrl_hit_drop_redir =
+		STATS_2x32_TO_64(ingress_counters.unctrl_hit_drop_redir);
+
+	ret = aq_mss_get_egress_common_counters(hw, &egress_counters);
+	if (unlikely(ret))
+		return ret;
+	stats->out.ctl_pkts = STATS_2x32_TO_64(egress_counters.ctl_pkt);
+	stats->out.unknown_sa_pkts =
+		STATS_2x32_TO_64(egress_counters.unknown_sa_pkts);
+	stats->out.untagged_pkts =
+		STATS_2x32_TO_64(egress_counters.untagged_pkts);
+	stats->out.too_long = STATS_2x32_TO_64(egress_counters.too_long);
+	stats->out.ecc_error_pkts =
+		STATS_2x32_TO_64(egress_counters.ecc_error_pkts);
+	stats->out.unctrl_hit_drop_redir =
+		STATS_2x32_TO_64(egress_counters.unctrl_hit_drop_redir);
+
+	return 0;
+}
+
+static int aq_get_rxsa_stats(struct aq_hw_s *hw, const int sa_idx,
+			     struct aq_macsec_rx_sa_stats *stats)
+{
+	struct aq_mss_ingress_sa_counters i_sa_counters;
+	int ret;
+
+	ret = aq_mss_get_ingress_sa_counters(hw, &i_sa_counters, sa_idx);
+	if (unlikely(ret))
+		return ret;
+
+	stats->untagged_hit_pkts =
+		STATS_2x32_TO_64(i_sa_counters.untagged_hit_pkts);
+	stats->ctrl_hit_drop_redir_pkts =
+		STATS_2x32_TO_64(i_sa_counters.ctrl_hit_drop_redir_pkts);
+	stats->not_using_sa = STATS_2x32_TO_64(i_sa_counters.not_using_sa);
+	stats->unused_sa = STATS_2x32_TO_64(i_sa_counters.unused_sa);
+	stats->not_valid_pkts = STATS_2x32_TO_64(i_sa_counters.not_valid_pkts);
+	stats->invalid_pkts = STATS_2x32_TO_64(i_sa_counters.invalid_pkts);
+	stats->ok_pkts = STATS_2x32_TO_64(i_sa_counters.ok_pkts);
+	stats->late_pkts = STATS_2x32_TO_64(i_sa_counters.late_pkts);
+	stats->delayed_pkts = STATS_2x32_TO_64(i_sa_counters.delayed_pkts);
+	stats->unchecked_pkts = STATS_2x32_TO_64(i_sa_counters.unchecked_pkts);
+	stats->validated_octets =
+		STATS_2x32_TO_64(i_sa_counters.validated_octets);
+	stats->decrypted_octets =
+		STATS_2x32_TO_64(i_sa_counters.decrypted_octets);
+
+	return 0;
+}
+
+static int aq_get_txsa_stats(struct aq_hw_s *hw, const int sa_idx,
+			     struct aq_macsec_tx_sa_stats *stats)
+{
+	struct aq_mss_egress_sa_counters e_sa_counters;
+	int ret;
+
+	ret = aq_mss_get_egress_sa_counters(hw, &e_sa_counters, sa_idx);
+	if (unlikely(ret))
+		return ret;
+
+	stats->sa_hit_drop_redirect =
+		STATS_2x32_TO_64(e_sa_counters.sa_hit_drop_redirect);
+	stats->sa_protected2_pkts =
+		STATS_2x32_TO_64(e_sa_counters.sa_protected2_pkts);
+	stats->sa_protected_pkts =
+		STATS_2x32_TO_64(e_sa_counters.sa_protected_pkts);
+	stats->sa_encrypted_pkts =
+		STATS_2x32_TO_64(e_sa_counters.sa_encrypted_pkts);
+
+	return 0;
+}
+
+static int aq_get_txsa_next_pn(struct aq_hw_s *hw, const int sa_idx, u32 *pn)
+{
+	struct aq_mss_egress_sa_record sa_rec;
+	int ret;
+
+	ret = aq_mss_get_egress_sa_record(hw, &sa_rec, sa_idx);
+	if (likely(!ret))
+		*pn = sa_rec.next_pn;
+
+	return ret;
+}
+
+static int aq_get_rxsa_next_pn(struct aq_hw_s *hw, const int sa_idx, u32 *pn)
+{
+	struct aq_mss_ingress_sa_record sa_rec;
+	int ret;
+
+	ret = aq_mss_get_ingress_sa_record(hw, &sa_rec, sa_idx);
+	if (likely(!ret))
+		*pn = (!sa_rec.sat_nextpn) ? sa_rec.next_pn : 0;
+
+	return ret;
+}
+
+static int aq_get_txsc_stats(struct aq_hw_s *hw, const int sc_idx,
+			     struct aq_macsec_tx_sc_stats *stats)
+{
+	struct aq_mss_egress_sc_counters e_sc_counters;
+	int ret;
+
+	ret = aq_mss_get_egress_sc_counters(hw, &e_sc_counters, sc_idx);
+	if (unlikely(ret))
+		return ret;
+
+	stats->sc_protected_pkts =
+		STATS_2x32_TO_64(e_sc_counters.sc_protected_pkts);
+	stats->sc_encrypted_pkts =
+		STATS_2x32_TO_64(e_sc_counters.sc_encrypted_pkts);
+	stats->sc_protected_octets =
+		STATS_2x32_TO_64(e_sc_counters.sc_protected_octets);
+	stats->sc_encrypted_octets =
+		STATS_2x32_TO_64(e_sc_counters.sc_encrypted_octets);
+
+	return 0;
+}
+
+static int aq_mdo_dev_open(struct macsec_context *ctx)
+{
+	struct aq_nic_s *nic = macsec_netdev_priv(ctx->netdev);
+	int ret = 0;
+
+	if (netif_carrier_ok(nic->ndev))
+		ret = aq_apply_secy_cfg(nic, ctx->secy);
+
+	return ret;
+}
+
+static int aq_mdo_dev_stop(struct macsec_context *ctx)
+{
+	struct aq_nic_s *nic = macsec_netdev_priv(ctx->netdev);
+	int i;
+
+	for (i = 0; i < AQ_MACSEC_MAX_SC; i++) {
+		if (nic->macsec_cfg->txsc_idx_busy & BIT(i))
+			aq_clear_secy(nic, nic->macsec_cfg->aq_txsc[i].sw_secy,
+				      AQ_CLEAR_HW);
+	}
+
+	return 0;
+}
+
+static int aq_set_txsc(struct aq_nic_s *nic, const int txsc_idx)
+{
+	struct aq_macsec_txsc *aq_txsc = &nic->macsec_cfg->aq_txsc[txsc_idx];
+	struct aq_mss_egress_class_record tx_class_rec = { 0 };
+	const struct macsec_secy *secy = aq_txsc->sw_secy;
+	struct aq_mss_egress_sc_record sc_rec = { 0 };
+	unsigned int sc_idx = aq_txsc->hw_sc_idx;
+	struct aq_hw_s *hw = nic->aq_hw;
+	int ret = 0;
+
+	aq_ether_addr_to_mac(tx_class_rec.mac_sa, secy->netdev->dev_addr);
+
+	put_unaligned_be64((__force u64)secy->sci, tx_class_rec.sci);
+	tx_class_rec.sci_mask = 0;
+
+	tx_class_rec.sa_mask = 0x3f;
+
+	tx_class_rec.action = 0; /* forward to SA/SC table */
+	tx_class_rec.valid = 1;
+
+	tx_class_rec.sc_idx = sc_idx;
+
+	tx_class_rec.sc_sa = nic->macsec_cfg->sc_sa;
+
+	ret = aq_mss_set_egress_class_record(hw, &tx_class_rec, txsc_idx);
+	if (ret)
+		return ret;
+
+	sc_rec.protect = secy->protect_frames;
+	if (secy->tx_sc.encrypt)
+		sc_rec.tci |= BIT(1);
+	if (secy->tx_sc.scb)
+		sc_rec.tci |= BIT(2);
+	if (secy->tx_sc.send_sci)
+		sc_rec.tci |= BIT(3);
+	if (secy->tx_sc.end_station)
+		sc_rec.tci |= BIT(4);
+	/* The C bit is clear if and only if the Secure Data is
+	 * exactly the same as the User Data and the ICV is 16 octets long.
+	 */
+	if (!(secy->icv_len == 16 && !secy->tx_sc.encrypt))
+		sc_rec.tci |= BIT(0);
+
+	sc_rec.an_roll = 0;
+
+	switch (secy->key_len) {
+	case AQ_MACSEC_KEY_LEN_128_BIT:
+		sc_rec.sak_len = 0;
+		break;
+	case AQ_MACSEC_KEY_LEN_192_BIT:
+		sc_rec.sak_len = 1;
+		break;
+	case AQ_MACSEC_KEY_LEN_256_BIT:
+		sc_rec.sak_len = 2;
+		break;
+	default:
+		WARN_ONCE(true, "Invalid sc_sa");
+		return -EINVAL;
+	}
+
+	sc_rec.curr_an = secy->tx_sc.encoding_sa;
+	sc_rec.valid = 1;
+	sc_rec.fresh = 1;
+
+	return aq_mss_set_egress_sc_record(hw, &sc_rec, sc_idx);
+}
+
+static u32 aq_sc_idx_max(const enum aq_macsec_sc_sa sc_sa)
+{
+	u32 result = 0;
+
+	switch (sc_sa) {
+	case aq_macsec_sa_sc_4sa_8sc:
+		result = 8;
+		break;
+	case aq_macsec_sa_sc_2sa_16sc:
+		result = 16;
+		break;
+	case aq_macsec_sa_sc_1sa_32sc:
+		result = 32;
+		break;
+	default:
+		break;
+	}
+
+	return result;
+}
+
+static u32 aq_to_hw_sc_idx(const u32 sc_idx, const enum aq_macsec_sc_sa sc_sa)
+{
+	switch (sc_sa) {
+	case aq_macsec_sa_sc_4sa_8sc:
+		return sc_idx << 2;
+	case aq_macsec_sa_sc_2sa_16sc:
+		return sc_idx << 1;
+	case aq_macsec_sa_sc_1sa_32sc:
+		return sc_idx;
+	default:
+		WARN_ONCE(true, "Invalid sc_sa");
+	}
+
+	return sc_idx;
+}
+
+static enum aq_macsec_sc_sa sc_sa_from_num_an(const int num_an)
+{
+	enum aq_macsec_sc_sa sc_sa = aq_macsec_sa_sc_not_used;
+
+	switch (num_an) {
+	case 4:
+		sc_sa = aq_macsec_sa_sc_4sa_8sc;
+		break;
+	case 2:
+		sc_sa = aq_macsec_sa_sc_2sa_16sc;
+		break;
+	case 1:
+		sc_sa = aq_macsec_sa_sc_1sa_32sc;
+		break;
+	default:
+		break;
+	}
+
+	return sc_sa;
+}
+
+static int aq_mdo_add_secy(struct macsec_context *ctx)
+{
+	struct aq_nic_s *nic = macsec_netdev_priv(ctx->netdev);
+	struct aq_macsec_cfg *cfg = nic->macsec_cfg;
+	const struct macsec_secy *secy = ctx->secy;
+	enum aq_macsec_sc_sa sc_sa;
+	u32 txsc_idx;
+	int ret = 0;
+
+	if (secy->xpn)
+		return -EOPNOTSUPP;
+
+	sc_sa = sc_sa_from_num_an(MACSEC_NUM_AN);
+	if (sc_sa == aq_macsec_sa_sc_not_used)
+		return -EINVAL;
+
+	if (hweight32(cfg->txsc_idx_busy) >= aq_sc_idx_max(sc_sa))
+		return -ENOSPC;
+
+	txsc_idx = ffz(cfg->txsc_idx_busy);
+	if (txsc_idx == AQ_MACSEC_MAX_SC)
+		return -ENOSPC;
+
+	cfg->sc_sa = sc_sa;
+	cfg->aq_txsc[txsc_idx].hw_sc_idx = aq_to_hw_sc_idx(txsc_idx, sc_sa);
+	cfg->aq_txsc[txsc_idx].sw_secy = secy;
+
+	if (netif_carrier_ok(nic->ndev) && netif_running(secy->netdev))
+		ret = aq_set_txsc(nic, txsc_idx);
+
+	set_bit(txsc_idx, &cfg->txsc_idx_busy);
+
+	return ret;
+}
+
+static int aq_mdo_upd_secy(struct macsec_context *ctx)
+{
+	struct aq_nic_s *nic = macsec_netdev_priv(ctx->netdev);
+	const struct macsec_secy *secy = ctx->secy;
+	int txsc_idx;
+	int ret = 0;
+
+	txsc_idx = aq_get_txsc_idx_from_secy(nic->macsec_cfg, secy);
+	if (txsc_idx < 0)
+		return -ENOENT;
+
+	if (netif_carrier_ok(nic->ndev) && netif_running(secy->netdev))
+		ret = aq_set_txsc(nic, txsc_idx);
+
+	return ret;
+}
+
+static int aq_clear_txsc(struct aq_nic_s *nic, const int txsc_idx,
+			 enum aq_clear_type clear_type)
+{
+	struct aq_macsec_txsc *tx_sc = &nic->macsec_cfg->aq_txsc[txsc_idx];
+	struct aq_mss_egress_class_record tx_class_rec = { 0 };
+	struct aq_mss_egress_sc_record sc_rec = { 0 };
+	struct aq_hw_s *hw = nic->aq_hw;
+	int ret = 0;
+	int sa_num;
+
+	for_each_set_bit (sa_num, &tx_sc->tx_sa_idx_busy, AQ_MACSEC_MAX_SA) {
+		ret = aq_clear_txsa(nic, tx_sc, sa_num, clear_type);
+		if (ret)
+			return ret;
+	}
+
+	if (clear_type & AQ_CLEAR_HW) {
+		ret = aq_mss_set_egress_class_record(hw, &tx_class_rec,
+						     txsc_idx);
+		if (ret)
+			return ret;
+
+		sc_rec.fresh = 1;
+		ret = aq_mss_set_egress_sc_record(hw, &sc_rec,
+						  tx_sc->hw_sc_idx);
+		if (ret)
+			return ret;
+	}
+
+	if (clear_type & AQ_CLEAR_SW) {
+		clear_bit(txsc_idx, &nic->macsec_cfg->txsc_idx_busy);
+		nic->macsec_cfg->aq_txsc[txsc_idx].sw_secy = NULL;
+	}
+
+	return ret;
+}
+
+static int aq_mdo_del_secy(struct macsec_context *ctx)
+{
+	struct aq_nic_s *nic = macsec_netdev_priv(ctx->netdev);
+	int ret = 0;
+
+	if (!nic->macsec_cfg)
+		return 0;
+
+	ret = aq_clear_secy(nic, ctx->secy, AQ_CLEAR_ALL);
+
+	return ret;
+}
+
+static int aq_update_txsa(struct aq_nic_s *nic, const unsigned int sc_idx,
+			  const struct macsec_secy *secy,
+			  const struct macsec_tx_sa *tx_sa,
+			  const unsigned char *key, const unsigned char an)
+{
+	const u32 next_pn = tx_sa->next_pn_halves.lower;
+	struct aq_mss_egress_sakey_record key_rec;
+	const unsigned int sa_idx = sc_idx | an;
+	struct aq_mss_egress_sa_record sa_rec;
+	struct aq_hw_s *hw = nic->aq_hw;
+	int ret = 0;
+
+	memset(&sa_rec, 0, sizeof(sa_rec));
+	sa_rec.valid = tx_sa->active;
+	sa_rec.fresh = 1;
+	sa_rec.next_pn = next_pn;
+
+	ret = aq_mss_set_egress_sa_record(hw, &sa_rec, sa_idx);
+	if (ret)
+		return ret;
+
+	if (!key)
+		return ret;
+
+	memset(&key_rec, 0, sizeof(key_rec));
+	memcpy(&key_rec.key, key, secy->key_len);
+
+	aq_rotate_keys(&key_rec.key, secy->key_len);
+
+	ret = aq_mss_set_egress_sakey_record(hw, &key_rec, sa_idx);
+
+	memzero_explicit(&key_rec, sizeof(key_rec));
+	return ret;
+}
+
+static int aq_mdo_add_txsa(struct macsec_context *ctx)
+{
+	struct aq_nic_s *nic = macsec_netdev_priv(ctx->netdev);
+	struct aq_macsec_cfg *cfg = nic->macsec_cfg;
+	const struct macsec_secy *secy = ctx->secy;
+	struct aq_macsec_txsc *aq_txsc;
+	int txsc_idx;
+	int ret = 0;
+
+	txsc_idx = aq_get_txsc_idx_from_secy(cfg, secy);
+	if (txsc_idx < 0)
+		return -EINVAL;
+
+	aq_txsc = &cfg->aq_txsc[txsc_idx];
+	set_bit(ctx->sa.assoc_num, &aq_txsc->tx_sa_idx_busy);
+
+	memcpy(aq_txsc->tx_sa_key[ctx->sa.assoc_num], ctx->sa.key,
+	       secy->key_len);
+
+	if (netif_carrier_ok(nic->ndev) && netif_running(secy->netdev))
+		ret = aq_update_txsa(nic, aq_txsc->hw_sc_idx, secy,
+				     ctx->sa.tx_sa, ctx->sa.key,
+				     ctx->sa.assoc_num);
+
+	return ret;
+}
+
+static int aq_mdo_upd_txsa(struct macsec_context *ctx)
+{
+	struct aq_nic_s *nic = macsec_netdev_priv(ctx->netdev);
+	struct aq_macsec_cfg *cfg = nic->macsec_cfg;
+	const struct macsec_secy *secy = ctx->secy;
+	struct aq_macsec_txsc *aq_txsc;
+	int txsc_idx;
+	int ret = 0;
+
+	txsc_idx = aq_get_txsc_idx_from_secy(cfg, secy);
+	if (txsc_idx < 0)
+		return -EINVAL;
+
+	aq_txsc = &cfg->aq_txsc[txsc_idx];
+	if (netif_carrier_ok(nic->ndev) && netif_running(secy->netdev))
+		ret = aq_update_txsa(nic, aq_txsc->hw_sc_idx, secy,
+				     ctx->sa.tx_sa, NULL, ctx->sa.assoc_num);
+
+	return ret;
+}
+
+static int aq_clear_txsa(struct aq_nic_s *nic, struct aq_macsec_txsc *aq_txsc,
+			 const int sa_num, enum aq_clear_type clear_type)
+{
+	const int sa_idx = aq_txsc->hw_sc_idx | sa_num;
+	struct aq_hw_s *hw = nic->aq_hw;
+	int ret = 0;
+
+	if (clear_type & AQ_CLEAR_SW)
+		clear_bit(sa_num, &aq_txsc->tx_sa_idx_busy);
+
+	if ((clear_type & AQ_CLEAR_HW) && netif_carrier_ok(nic->ndev)) {
+		struct aq_mss_egress_sakey_record key_rec;
+		struct aq_mss_egress_sa_record sa_rec;
+
+		memset(&sa_rec, 0, sizeof(sa_rec));
+		sa_rec.fresh = 1;
+
+		ret = aq_mss_set_egress_sa_record(hw, &sa_rec, sa_idx);
+		if (ret)
+			return ret;
+
+		memset(&key_rec, 0, sizeof(key_rec));
+		return aq_mss_set_egress_sakey_record(hw, &key_rec, sa_idx);
+	}
+
+	return 0;
+}
+
+static int aq_mdo_del_txsa(struct macsec_context *ctx)
+{
+	struct aq_nic_s *nic = macsec_netdev_priv(ctx->netdev);
+	struct aq_macsec_cfg *cfg = nic->macsec_cfg;
+	int txsc_idx;
+	int ret = 0;
+
+	txsc_idx = aq_get_txsc_idx_from_secy(cfg, ctx->secy);
+	if (txsc_idx < 0)
+		return -EINVAL;
+
+	ret = aq_clear_txsa(nic, &cfg->aq_txsc[txsc_idx], ctx->sa.assoc_num,
+			    AQ_CLEAR_ALL);
+
+	return ret;
+}
+
+static int aq_rxsc_validate_frames(const enum macsec_validation_type validate)
+{
+	switch (validate) {
+	case MACSEC_VALIDATE_DISABLED:
+		return 2;
+	case MACSEC_VALIDATE_CHECK:
+		return 1;
+	case MACSEC_VALIDATE_STRICT:
+		return 0;
+	default:
+		WARN_ONCE(true, "Invalid validation type");
+	}
+
+	return 0;
+}
+
+static int aq_set_rxsc(struct aq_nic_s *nic, const u32 rxsc_idx)
+{
+	const struct aq_macsec_rxsc *aq_rxsc =
+		&nic->macsec_cfg->aq_rxsc[rxsc_idx];
+	struct aq_mss_ingress_preclass_record pre_class_record;
+	const struct macsec_rx_sc *rx_sc = aq_rxsc->sw_rxsc;
+	const struct macsec_secy *secy = aq_rxsc->sw_secy;
+	const u32 hw_sc_idx = aq_rxsc->hw_sc_idx;
+	struct aq_mss_ingress_sc_record sc_record;
+	struct aq_hw_s *hw = nic->aq_hw;
+	int ret = 0;
+
+	memset(&pre_class_record, 0, sizeof(pre_class_record));
+	put_unaligned_be64((__force u64)rx_sc->sci, pre_class_record.sci);
+	pre_class_record.sci_mask = 0xff;
+	/* match all MACSEC ethertype packets */
+	pre_class_record.eth_type = ETH_P_MACSEC;
+	pre_class_record.eth_type_mask = 0x3;
+
+	aq_ether_addr_to_mac(pre_class_record.mac_sa, (char *)&rx_sc->sci);
+	pre_class_record.sa_mask = 0x3f;
+
+	pre_class_record.an_mask = nic->macsec_cfg->sc_sa;
+	pre_class_record.sc_idx = hw_sc_idx;
+	/* strip SecTAG & forward for decryption */
+	pre_class_record.action = 0x0;
+	pre_class_record.valid = 1;
+
+	ret = aq_mss_set_ingress_preclass_record(hw, &pre_class_record,
+						 2 * rxsc_idx + 1);
+	if (ret)
+		return ret;
+
+	/* If SCI is absent, then match by SA alone */
+	pre_class_record.sci_mask = 0;
+	pre_class_record.sci_from_table = 1;
+
+	ret = aq_mss_set_ingress_preclass_record(hw, &pre_class_record,
+						 2 * rxsc_idx);
+	if (ret)
+		return ret;
+
+	memset(&sc_record, 0, sizeof(sc_record));
+	sc_record.validate_frames =
+		aq_rxsc_validate_frames(secy->validate_frames);
+	if (secy->replay_protect) {
+		sc_record.replay_protect = 1;
+		sc_record.anti_replay_window = secy->replay_window;
+	}
+	sc_record.valid = 1;
+	sc_record.fresh = 1;
+
+	ret = aq_mss_set_ingress_sc_record(hw, &sc_record, hw_sc_idx);
+	if (ret)
+		return ret;
+
+	return ret;
+}
+
+static int aq_mdo_add_rxsc(struct macsec_context *ctx)
+{
+	struct aq_nic_s *nic = macsec_netdev_priv(ctx->netdev);
+	struct aq_macsec_cfg *cfg = nic->macsec_cfg;
+	const u32 rxsc_idx_max = aq_sc_idx_max(cfg->sc_sa);
+	u32 rxsc_idx;
+	int ret = 0;
+
+	if (hweight32(cfg->rxsc_idx_busy) >= rxsc_idx_max)
+		return -ENOSPC;
+
+	rxsc_idx = ffz(cfg->rxsc_idx_busy);
+	if (rxsc_idx >= rxsc_idx_max)
+		return -ENOSPC;
+
+	cfg->aq_rxsc[rxsc_idx].hw_sc_idx = aq_to_hw_sc_idx(rxsc_idx,
+							   cfg->sc_sa);
+	cfg->aq_rxsc[rxsc_idx].sw_secy = ctx->secy;
+	cfg->aq_rxsc[rxsc_idx].sw_rxsc = ctx->rx_sc;
+
+	if (netif_carrier_ok(nic->ndev) && netif_running(ctx->secy->netdev))
+		ret = aq_set_rxsc(nic, rxsc_idx);
+
+	if (ret < 0)
+		return ret;
+
+	set_bit(rxsc_idx, &cfg->rxsc_idx_busy);
+
+	return 0;
+}
+
+static int aq_mdo_upd_rxsc(struct macsec_context *ctx)
+{
+	struct aq_nic_s *nic = macsec_netdev_priv(ctx->netdev);
+	int rxsc_idx;
+	int ret = 0;
+
+	rxsc_idx = aq_get_rxsc_idx_from_rxsc(nic->macsec_cfg, ctx->rx_sc);
+	if (rxsc_idx < 0)
+		return -ENOENT;
+
+	if (netif_carrier_ok(nic->ndev) && netif_running(ctx->secy->netdev))
+		ret = aq_set_rxsc(nic, rxsc_idx);
+
+	return ret;
+}
+
+static int aq_clear_rxsc(struct aq_nic_s *nic, const int rxsc_idx,
+			 enum aq_clear_type clear_type)
+{
+	struct aq_macsec_rxsc *rx_sc = &nic->macsec_cfg->aq_rxsc[rxsc_idx];
+	struct aq_hw_s *hw = nic->aq_hw;
+	int ret = 0;
+	int sa_num;
+
+	for_each_set_bit (sa_num, &rx_sc->rx_sa_idx_busy, AQ_MACSEC_MAX_SA) {
+		ret = aq_clear_rxsa(nic, rx_sc, sa_num, clear_type);
+		if (ret)
+			return ret;
+	}
+
+	if (clear_type & AQ_CLEAR_HW) {
+		struct aq_mss_ingress_preclass_record pre_class_record;
+		struct aq_mss_ingress_sc_record sc_record;
+
+		memset(&pre_class_record, 0, sizeof(pre_class_record));
+		memset(&sc_record, 0, sizeof(sc_record));
+
+		ret = aq_mss_set_ingress_preclass_record(hw, &pre_class_record,
+							 2 * rxsc_idx);
+		if (ret)
+			return ret;
+
+		ret = aq_mss_set_ingress_preclass_record(hw, &pre_class_record,
+							 2 * rxsc_idx + 1);
+		if (ret)
+			return ret;
+
+		sc_record.fresh = 1;
+		ret = aq_mss_set_ingress_sc_record(hw, &sc_record,
+						   rx_sc->hw_sc_idx);
+		if (ret)
+			return ret;
+	}
+
+	if (clear_type & AQ_CLEAR_SW) {
+		clear_bit(rxsc_idx, &nic->macsec_cfg->rxsc_idx_busy);
+		rx_sc->sw_secy = NULL;
+		rx_sc->sw_rxsc = NULL;
+	}
+
+	return ret;
+}
+
+static int aq_mdo_del_rxsc(struct macsec_context *ctx)
+{
+	struct aq_nic_s *nic = macsec_netdev_priv(ctx->netdev);
+	enum aq_clear_type clear_type = AQ_CLEAR_SW;
+	int rxsc_idx;
+	int ret = 0;
+
+	rxsc_idx = aq_get_rxsc_idx_from_rxsc(nic->macsec_cfg, ctx->rx_sc);
+	if (rxsc_idx < 0)
+		return -ENOENT;
+
+	if (netif_carrier_ok(nic->ndev))
+		clear_type = AQ_CLEAR_ALL;
+
+	ret = aq_clear_rxsc(nic, rxsc_idx, clear_type);
+
+	return ret;
+}
+
+static int aq_update_rxsa(struct aq_nic_s *nic, const unsigned int sc_idx,
+			  const struct macsec_secy *secy,
+			  const struct macsec_rx_sa *rx_sa,
+			  const unsigned char *key, const unsigned char an)
+{
+	struct aq_mss_ingress_sakey_record sa_key_record;
+	const u32 next_pn = rx_sa->next_pn_halves.lower;
+	struct aq_mss_ingress_sa_record sa_record;
+	struct aq_hw_s *hw = nic->aq_hw;
+	const int sa_idx = sc_idx | an;
+	int ret = 0;
+
+	memset(&sa_record, 0, sizeof(sa_record));
+	sa_record.valid = rx_sa->active;
+	sa_record.fresh = 1;
+	sa_record.next_pn = next_pn;
+
+	ret = aq_mss_set_ingress_sa_record(hw, &sa_record, sa_idx);
+	if (ret)
+		return ret;
+
+	if (!key)
+		return ret;
+
+	memset(&sa_key_record, 0, sizeof(sa_key_record));
+	memcpy(&sa_key_record.key, key, secy->key_len);
+
+	switch (secy->key_len) {
+	case AQ_MACSEC_KEY_LEN_128_BIT:
+		sa_key_record.key_len = 0;
+		break;
+	case AQ_MACSEC_KEY_LEN_192_BIT:
+		sa_key_record.key_len = 1;
+		break;
+	case AQ_MACSEC_KEY_LEN_256_BIT:
+		sa_key_record.key_len = 2;
+		break;
+	default:
+		return -1;
+	}
+
+	aq_rotate_keys(&sa_key_record.key, secy->key_len);
+
+	ret = aq_mss_set_ingress_sakey_record(hw, &sa_key_record, sa_idx);
+
+	memzero_explicit(&sa_key_record, sizeof(sa_key_record));
+	return ret;
+}
+
+static int aq_mdo_add_rxsa(struct macsec_context *ctx)
+{
+	struct aq_nic_s *nic = macsec_netdev_priv(ctx->netdev);
+	const struct macsec_rx_sc *rx_sc = ctx->sa.rx_sa->sc;
+	const struct macsec_secy *secy = ctx->secy;
+	struct aq_macsec_rxsc *aq_rxsc;
+	int rxsc_idx;
+	int ret = 0;
+
+	rxsc_idx = aq_get_rxsc_idx_from_rxsc(nic->macsec_cfg, rx_sc);
+	if (rxsc_idx < 0)
+		return -EINVAL;
+
+	aq_rxsc = &nic->macsec_cfg->aq_rxsc[rxsc_idx];
+	set_bit(ctx->sa.assoc_num, &aq_rxsc->rx_sa_idx_busy);
+
+	memcpy(aq_rxsc->rx_sa_key[ctx->sa.assoc_num], ctx->sa.key,
+	       secy->key_len);
+
+	if (netif_carrier_ok(nic->ndev) && netif_running(secy->netdev))
+		ret = aq_update_rxsa(nic, aq_rxsc->hw_sc_idx, secy,
+				     ctx->sa.rx_sa, ctx->sa.key,
+				     ctx->sa.assoc_num);
+
+	return ret;
+}
+
+static int aq_mdo_upd_rxsa(struct macsec_context *ctx)
+{
+	struct aq_nic_s *nic = macsec_netdev_priv(ctx->netdev);
+	const struct macsec_rx_sc *rx_sc = ctx->sa.rx_sa->sc;
+	struct aq_macsec_cfg *cfg = nic->macsec_cfg;
+	const struct macsec_secy *secy = ctx->secy;
+	int rxsc_idx;
+	int ret = 0;
+
+	rxsc_idx = aq_get_rxsc_idx_from_rxsc(cfg, rx_sc);
+	if (rxsc_idx < 0)
+		return -EINVAL;
+
+	if (netif_carrier_ok(nic->ndev) && netif_running(secy->netdev))
+		ret = aq_update_rxsa(nic, cfg->aq_rxsc[rxsc_idx].hw_sc_idx,
+				     secy, ctx->sa.rx_sa, NULL,
+				     ctx->sa.assoc_num);
+
+	return ret;
+}
+
+static int aq_clear_rxsa(struct aq_nic_s *nic, struct aq_macsec_rxsc *aq_rxsc,
+			 const int sa_num, enum aq_clear_type clear_type)
+{
+	int sa_idx = aq_rxsc->hw_sc_idx | sa_num;
+	struct aq_hw_s *hw = nic->aq_hw;
+	int ret = 0;
+
+	if (clear_type & AQ_CLEAR_SW)
+		clear_bit(sa_num, &aq_rxsc->rx_sa_idx_busy);
+
+	if ((clear_type & AQ_CLEAR_HW) && netif_carrier_ok(nic->ndev)) {
+		struct aq_mss_ingress_sakey_record sa_key_record;
+		struct aq_mss_ingress_sa_record sa_record;
+
+		memset(&sa_key_record, 0, sizeof(sa_key_record));
+		memset(&sa_record, 0, sizeof(sa_record));
+		sa_record.fresh = 1;
+		ret = aq_mss_set_ingress_sa_record(hw, &sa_record, sa_idx);
+		if (ret)
+			return ret;
+
+		return aq_mss_set_ingress_sakey_record(hw, &sa_key_record,
+						       sa_idx);
+	}
+
+	return ret;
+}
+
+static int aq_mdo_del_rxsa(struct macsec_context *ctx)
+{
+	struct aq_nic_s *nic = macsec_netdev_priv(ctx->netdev);
+	const struct macsec_rx_sc *rx_sc = ctx->sa.rx_sa->sc;
+	struct aq_macsec_cfg *cfg = nic->macsec_cfg;
+	int rxsc_idx;
+	int ret = 0;
+
+	rxsc_idx = aq_get_rxsc_idx_from_rxsc(cfg, rx_sc);
+	if (rxsc_idx < 0)
+		return -EINVAL;
+
+	ret = aq_clear_rxsa(nic, &cfg->aq_rxsc[rxsc_idx], ctx->sa.assoc_num,
+			    AQ_CLEAR_ALL);
+
+	return ret;
+}
+
+static int aq_mdo_get_dev_stats(struct macsec_context *ctx)
+{
+	struct aq_nic_s *nic = macsec_netdev_priv(ctx->netdev);
+	struct aq_macsec_common_stats *stats = &nic->macsec_cfg->stats;
+	struct aq_hw_s *hw = nic->aq_hw;
+
+	aq_get_macsec_common_stats(hw, stats);
+
+	ctx->stats.dev_stats->OutPktsUntagged = stats->out.untagged_pkts;
+	ctx->stats.dev_stats->InPktsUntagged = stats->in.untagged_pkts;
+	ctx->stats.dev_stats->OutPktsTooLong = stats->out.too_long;
+	ctx->stats.dev_stats->InPktsNoTag = stats->in.notag_pkts;
+	ctx->stats.dev_stats->InPktsBadTag = stats->in.bad_tag_pkts;
+	ctx->stats.dev_stats->InPktsUnknownSCI = stats->in.unknown_sci_pkts;
+	ctx->stats.dev_stats->InPktsNoSCI = stats->in.no_sci_pkts;
+	ctx->stats.dev_stats->InPktsOverrun = 0;
+
+	return 0;
+}
+
+static int aq_mdo_get_tx_sc_stats(struct macsec_context *ctx)
+{
+	struct aq_nic_s *nic = macsec_netdev_priv(ctx->netdev);
+	struct aq_macsec_tx_sc_stats *stats;
+	struct aq_hw_s *hw = nic->aq_hw;
+	struct aq_macsec_txsc *aq_txsc;
+	int txsc_idx;
+
+	txsc_idx = aq_get_txsc_idx_from_secy(nic->macsec_cfg, ctx->secy);
+	if (txsc_idx < 0)
+		return -ENOENT;
+
+	aq_txsc = &nic->macsec_cfg->aq_txsc[txsc_idx];
+	stats = &aq_txsc->stats;
+	aq_get_txsc_stats(hw, aq_txsc->hw_sc_idx, stats);
+
+	ctx->stats.tx_sc_stats->OutPktsProtected = stats->sc_protected_pkts;
+	ctx->stats.tx_sc_stats->OutPktsEncrypted = stats->sc_encrypted_pkts;
+	ctx->stats.tx_sc_stats->OutOctetsProtected = stats->sc_protected_octets;
+	ctx->stats.tx_sc_stats->OutOctetsEncrypted = stats->sc_encrypted_octets;
+
+	return 0;
+}
+
+static int aq_mdo_get_tx_sa_stats(struct macsec_context *ctx)
+{
+	struct aq_nic_s *nic = macsec_netdev_priv(ctx->netdev);
+	struct aq_macsec_cfg *cfg = nic->macsec_cfg;
+	struct aq_macsec_tx_sa_stats *stats;
+	struct aq_hw_s *hw = nic->aq_hw;
+	const struct macsec_secy *secy;
+	struct aq_macsec_txsc *aq_txsc;
+	struct macsec_tx_sa *tx_sa;
+	unsigned int sa_idx;
+	int txsc_idx;
+	u32 next_pn;
+	int ret;
+
+	txsc_idx = aq_get_txsc_idx_from_secy(cfg, ctx->secy);
+	if (txsc_idx < 0)
+		return -EINVAL;
+
+	aq_txsc = &cfg->aq_txsc[txsc_idx];
+	sa_idx = aq_txsc->hw_sc_idx | ctx->sa.assoc_num;
+	stats = &aq_txsc->tx_sa_stats[ctx->sa.assoc_num];
+	ret = aq_get_txsa_stats(hw, sa_idx, stats);
+	if (ret)
+		return ret;
+
+	ctx->stats.tx_sa_stats->OutPktsProtected = stats->sa_protected_pkts;
+	ctx->stats.tx_sa_stats->OutPktsEncrypted = stats->sa_encrypted_pkts;
+
+	secy = aq_txsc->sw_secy;
+	tx_sa = rcu_dereference_bh(secy->tx_sc.sa[ctx->sa.assoc_num]);
+	ret = aq_get_txsa_next_pn(hw, sa_idx, &next_pn);
+	if (ret == 0) {
+		spin_lock_bh(&tx_sa->lock);
+		tx_sa->next_pn = next_pn;
+		spin_unlock_bh(&tx_sa->lock);
+	}
+
+	return ret;
+}
+
+static int aq_mdo_get_rx_sc_stats(struct macsec_context *ctx)
+{
+	struct aq_nic_s *nic = macsec_netdev_priv(ctx->netdev);
+	struct aq_macsec_cfg *cfg = nic->macsec_cfg;
+	struct aq_macsec_rx_sa_stats *stats;
+	struct aq_hw_s *hw = nic->aq_hw;
+	struct aq_macsec_rxsc *aq_rxsc;
+	unsigned int sa_idx;
+	int rxsc_idx;
+	int ret = 0;
+	int i;
+
+	rxsc_idx = aq_get_rxsc_idx_from_rxsc(cfg, ctx->rx_sc);
+	if (rxsc_idx < 0)
+		return -ENOENT;
+
+	aq_rxsc = &cfg->aq_rxsc[rxsc_idx];
+	for (i = 0; i < MACSEC_NUM_AN; i++) {
+		if (!test_bit(i, &aq_rxsc->rx_sa_idx_busy))
+			continue;
+
+		stats = &aq_rxsc->rx_sa_stats[i];
+		sa_idx = aq_rxsc->hw_sc_idx | i;
+		ret = aq_get_rxsa_stats(hw, sa_idx, stats);
+		if (ret)
+			break;
+
+		ctx->stats.rx_sc_stats->InOctetsValidated +=
+			stats->validated_octets;
+		ctx->stats.rx_sc_stats->InOctetsDecrypted +=
+			stats->decrypted_octets;
+		ctx->stats.rx_sc_stats->InPktsUnchecked +=
+			stats->unchecked_pkts;
+		ctx->stats.rx_sc_stats->InPktsDelayed += stats->delayed_pkts;
+		ctx->stats.rx_sc_stats->InPktsOK += stats->ok_pkts;
+		ctx->stats.rx_sc_stats->InPktsInvalid += stats->invalid_pkts;
+		ctx->stats.rx_sc_stats->InPktsLate += stats->late_pkts;
+		ctx->stats.rx_sc_stats->InPktsNotValid += stats->not_valid_pkts;
+		ctx->stats.rx_sc_stats->InPktsNotUsingSA += stats->not_using_sa;
+		ctx->stats.rx_sc_stats->InPktsUnusedSA += stats->unused_sa;
+	}
+
+	return ret;
+}
+
+static int aq_mdo_get_rx_sa_stats(struct macsec_context *ctx)
+{
+	struct aq_nic_s *nic = macsec_netdev_priv(ctx->netdev);
+	struct aq_macsec_cfg *cfg = nic->macsec_cfg;
+	struct aq_macsec_rx_sa_stats *stats;
+	struct aq_hw_s *hw = nic->aq_hw;
+	struct aq_macsec_rxsc *aq_rxsc;
+	struct macsec_rx_sa *rx_sa;
+	unsigned int sa_idx;
+	int rxsc_idx;
+	u32 next_pn;
+	int ret;
+
+	rxsc_idx = aq_get_rxsc_idx_from_rxsc(cfg, ctx->rx_sc);
+	if (rxsc_idx < 0)
+		return -EINVAL;
+
+	aq_rxsc = &cfg->aq_rxsc[rxsc_idx];
+	stats = &aq_rxsc->rx_sa_stats[ctx->sa.assoc_num];
+	sa_idx = aq_rxsc->hw_sc_idx | ctx->sa.assoc_num;
+	ret = aq_get_rxsa_stats(hw, sa_idx, stats);
+	if (ret)
+		return ret;
+
+	ctx->stats.rx_sa_stats->InPktsOK = stats->ok_pkts;
+	ctx->stats.rx_sa_stats->InPktsInvalid = stats->invalid_pkts;
+	ctx->stats.rx_sa_stats->InPktsNotValid = stats->not_valid_pkts;
+	ctx->stats.rx_sa_stats->InPktsNotUsingSA = stats->not_using_sa;
+	ctx->stats.rx_sa_stats->InPktsUnusedSA = stats->unused_sa;
+
+	rx_sa = rcu_dereference_bh(aq_rxsc->sw_rxsc->sa[ctx->sa.assoc_num]);
+	ret = aq_get_rxsa_next_pn(hw, sa_idx, &next_pn);
+	if (ret == 0) {
+		spin_lock_bh(&rx_sa->lock);
+		rx_sa->next_pn = next_pn;
+		spin_unlock_bh(&rx_sa->lock);
+	}
+
+	return ret;
+}
+
+static int apply_txsc_cfg(struct aq_nic_s *nic, const int txsc_idx)
+{
+	struct aq_macsec_txsc *aq_txsc = &nic->macsec_cfg->aq_txsc[txsc_idx];
+	const struct macsec_secy *secy = aq_txsc->sw_secy;
+	struct macsec_tx_sa *tx_sa;
+	int ret = 0;
+	int i;
+
+	if (!netif_running(secy->netdev))
+		return ret;
+
+	ret = aq_set_txsc(nic, txsc_idx);
+	if (ret)
+		return ret;
+
+	for (i = 0; i < MACSEC_NUM_AN; i++) {
+		tx_sa = rcu_dereference_bh(secy->tx_sc.sa[i]);
+		if (tx_sa) {
+			ret = aq_update_txsa(nic, aq_txsc->hw_sc_idx, secy,
+					     tx_sa, aq_txsc->tx_sa_key[i], i);
+			if (ret)
+				return ret;
+		}
+	}
+
+	return ret;
+}
+
+static int apply_rxsc_cfg(struct aq_nic_s *nic, const int rxsc_idx)
+{
+	struct aq_macsec_rxsc *aq_rxsc = &nic->macsec_cfg->aq_rxsc[rxsc_idx];
+	const struct macsec_secy *secy = aq_rxsc->sw_secy;
+	struct macsec_rx_sa *rx_sa;
+	int ret = 0;
+	int i;
+
+	if (!netif_running(secy->netdev))
+		return ret;
+
+	ret = aq_set_rxsc(nic, rxsc_idx);
+	if (ret)
+		return ret;
+
+	for (i = 0; i < MACSEC_NUM_AN; i++) {
+		rx_sa = rcu_dereference_bh(aq_rxsc->sw_rxsc->sa[i]);
+		if (rx_sa) {
+			ret = aq_update_rxsa(nic, aq_rxsc->hw_sc_idx, secy,
+					     rx_sa, aq_rxsc->rx_sa_key[i], i);
+			if (ret)
+				return ret;
+		}
+	}
+
+	return ret;
+}
+
+static int aq_clear_secy(struct aq_nic_s *nic, const struct macsec_secy *secy,
+			 enum aq_clear_type clear_type)
+{
+	struct macsec_rx_sc *rx_sc;
+	int txsc_idx;
+	int rxsc_idx;
+	int ret = 0;
+
+	txsc_idx = aq_get_txsc_idx_from_secy(nic->macsec_cfg, secy);
+	if (txsc_idx >= 0) {
+		ret = aq_clear_txsc(nic, txsc_idx, clear_type);
+		if (ret)
+			return ret;
+	}
+
+	for (rx_sc = rcu_dereference_bh(secy->rx_sc); rx_sc;
+	     rx_sc = rcu_dereference_bh(rx_sc->next)) {
+		rxsc_idx = aq_get_rxsc_idx_from_rxsc(nic->macsec_cfg, rx_sc);
+		if (rxsc_idx < 0)
+			continue;
+
+		ret = aq_clear_rxsc(nic, rxsc_idx, clear_type);
+		if (ret)
+			return ret;
+	}
+
+	return ret;
+}
+
+static int aq_apply_secy_cfg(struct aq_nic_s *nic,
+			     const struct macsec_secy *secy)
+{
+	struct macsec_rx_sc *rx_sc;
+	int txsc_idx;
+	int rxsc_idx;
+	int ret = 0;
+
+	txsc_idx = aq_get_txsc_idx_from_secy(nic->macsec_cfg, secy);
+	if (txsc_idx >= 0)
+		apply_txsc_cfg(nic, txsc_idx);
+
+	for (rx_sc = rcu_dereference_bh(secy->rx_sc); rx_sc && rx_sc->active;
+	     rx_sc = rcu_dereference_bh(rx_sc->next)) {
+		rxsc_idx = aq_get_rxsc_idx_from_rxsc(nic->macsec_cfg, rx_sc);
+		if (unlikely(rxsc_idx < 0))
+			continue;
+
+		ret = apply_rxsc_cfg(nic, rxsc_idx);
+		if (ret)
+			return ret;
+	}
+
+	return ret;
+}
+
+static int aq_apply_macsec_cfg(struct aq_nic_s *nic)
+{
+	int ret = 0;
+	int i;
+
+	for (i = 0; i < AQ_MACSEC_MAX_SC; i++) {
+		if (nic->macsec_cfg->txsc_idx_busy & BIT(i)) {
+			ret = apply_txsc_cfg(nic, i);
+			if (ret)
+				return ret;
+		}
+	}
+
+	for (i = 0; i < AQ_MACSEC_MAX_SC; i++) {
+		if (nic->macsec_cfg->rxsc_idx_busy & BIT(i)) {
+			ret = apply_rxsc_cfg(nic, i);
+			if (ret)
+				return ret;
+		}
+	}
+
+	return ret;
+}
+
+static int aq_sa_from_sa_idx(const enum aq_macsec_sc_sa sc_sa, const int sa_idx)
+{
+	switch (sc_sa) {
+	case aq_macsec_sa_sc_4sa_8sc:
+		return sa_idx & 3;
+	case aq_macsec_sa_sc_2sa_16sc:
+		return sa_idx & 1;
+	case aq_macsec_sa_sc_1sa_32sc:
+		return 0;
+	default:
+		WARN_ONCE(true, "Invalid sc_sa");
+	}
+	return -EINVAL;
+}
+
+static int aq_sc_idx_from_sa_idx(const enum aq_macsec_sc_sa sc_sa,
+				 const int sa_idx)
+{
+	switch (sc_sa) {
+	case aq_macsec_sa_sc_4sa_8sc:
+		return sa_idx & ~3;
+	case aq_macsec_sa_sc_2sa_16sc:
+		return sa_idx & ~1;
+	case aq_macsec_sa_sc_1sa_32sc:
+		return sa_idx;
+	default:
+		WARN_ONCE(true, "Invalid sc_sa");
+	}
+	return -EINVAL;
+}
+
+static void aq_check_txsa_expiration(struct aq_nic_s *nic)
+{
+	u32 egress_sa_expired, egress_sa_threshold_expired;
+	struct aq_macsec_cfg *cfg = nic->macsec_cfg;
+	struct aq_hw_s *hw = nic->aq_hw;
+	struct aq_macsec_txsc *aq_txsc;
+	const struct macsec_secy *secy;
+	int sc_idx = 0, txsc_idx = 0;
+	enum aq_macsec_sc_sa sc_sa;
+	struct macsec_tx_sa *tx_sa;
+	unsigned char an = 0;
+	int ret;
+	int i;
+
+	sc_sa = cfg->sc_sa;
+
+	ret = aq_mss_get_egress_sa_expired(hw, &egress_sa_expired);
+	if (unlikely(ret))
+		return;
+
+	ret = aq_mss_get_egress_sa_threshold_expired(hw,
+		&egress_sa_threshold_expired);
+
+	for (i = 0; i < AQ_MACSEC_MAX_SA; i++) {
+		if (egress_sa_expired & BIT(i)) {
+			an = aq_sa_from_sa_idx(sc_sa, i);
+			sc_idx = aq_sc_idx_from_sa_idx(sc_sa, i);
+			txsc_idx = aq_get_txsc_idx_from_sc_idx(sc_sa, sc_idx);
+			if (txsc_idx < 0)
+				continue;
+
+			aq_txsc = &cfg->aq_txsc[txsc_idx];
+			if (!(cfg->txsc_idx_busy & BIT(txsc_idx))) {
+				netdev_warn(nic->ndev,
+					"PN threshold expired on invalid TX SC");
+				continue;
+			}
+
+			secy = aq_txsc->sw_secy;
+			if (!netif_running(secy->netdev)) {
+				netdev_warn(nic->ndev,
+					"PN threshold expired on down TX SC");
+				continue;
+			}
+
+			if (unlikely(!(aq_txsc->tx_sa_idx_busy & BIT(an)))) {
+				netdev_warn(nic->ndev,
+					"PN threshold expired on invalid TX SA");
+				continue;
+			}
+
+			tx_sa = rcu_dereference_bh(secy->tx_sc.sa[an]);
+			macsec_pn_wrapped((struct macsec_secy *)secy, tx_sa);
+		}
+	}
+
+	aq_mss_set_egress_sa_expired(hw, egress_sa_expired);
+	if (likely(!ret))
+		aq_mss_set_egress_sa_threshold_expired(hw,
+			egress_sa_threshold_expired);
+}
+
+#define AQ_LOCKED_MDO_DEF(mdo)						\
+static int aq_locked_mdo_##mdo(struct macsec_context *ctx)		\
+{									\
+	struct aq_nic_s *nic = macsec_netdev_priv(ctx->netdev);		\
+	int ret;							\
+	mutex_lock(&nic->macsec_mutex);					\
+	ret = aq_mdo_##mdo(ctx);					\
+	mutex_unlock(&nic->macsec_mutex);				\
+	return ret;							\
+}
+
+AQ_LOCKED_MDO_DEF(dev_open)
+AQ_LOCKED_MDO_DEF(dev_stop)
+AQ_LOCKED_MDO_DEF(add_secy)
+AQ_LOCKED_MDO_DEF(upd_secy)
+AQ_LOCKED_MDO_DEF(del_secy)
+AQ_LOCKED_MDO_DEF(add_rxsc)
+AQ_LOCKED_MDO_DEF(upd_rxsc)
+AQ_LOCKED_MDO_DEF(del_rxsc)
+AQ_LOCKED_MDO_DEF(add_rxsa)
+AQ_LOCKED_MDO_DEF(upd_rxsa)
+AQ_LOCKED_MDO_DEF(del_rxsa)
+AQ_LOCKED_MDO_DEF(add_txsa)
+AQ_LOCKED_MDO_DEF(upd_txsa)
+AQ_LOCKED_MDO_DEF(del_txsa)
+AQ_LOCKED_MDO_DEF(get_dev_stats)
+AQ_LOCKED_MDO_DEF(get_tx_sc_stats)
+AQ_LOCKED_MDO_DEF(get_tx_sa_stats)
+AQ_LOCKED_MDO_DEF(get_rx_sc_stats)
+AQ_LOCKED_MDO_DEF(get_rx_sa_stats)
+
+const struct macsec_ops aq_macsec_ops = {
+	.mdo_dev_open = aq_locked_mdo_dev_open,
+	.mdo_dev_stop = aq_locked_mdo_dev_stop,
+	.mdo_add_secy = aq_locked_mdo_add_secy,
+	.mdo_upd_secy = aq_locked_mdo_upd_secy,
+	.mdo_del_secy = aq_locked_mdo_del_secy,
+	.mdo_add_rxsc = aq_locked_mdo_add_rxsc,
+	.mdo_upd_rxsc = aq_locked_mdo_upd_rxsc,
+	.mdo_del_rxsc = aq_locked_mdo_del_rxsc,
+	.mdo_add_rxsa = aq_locked_mdo_add_rxsa,
+	.mdo_upd_rxsa = aq_locked_mdo_upd_rxsa,
+	.mdo_del_rxsa = aq_locked_mdo_del_rxsa,
+	.mdo_add_txsa = aq_locked_mdo_add_txsa,
+	.mdo_upd_txsa = aq_locked_mdo_upd_txsa,
+	.mdo_del_txsa = aq_locked_mdo_del_txsa,
+	.mdo_get_dev_stats = aq_locked_mdo_get_dev_stats,
+	.mdo_get_tx_sc_stats = aq_locked_mdo_get_tx_sc_stats,
+	.mdo_get_tx_sa_stats = aq_locked_mdo_get_tx_sa_stats,
+	.mdo_get_rx_sc_stats = aq_locked_mdo_get_rx_sc_stats,
+	.mdo_get_rx_sa_stats = aq_locked_mdo_get_rx_sa_stats,
+};
+
+int aq_macsec_init(struct aq_nic_s *nic)
+{
+	struct aq_macsec_cfg *cfg;
+	u32 caps_lo;
+
+	if (!nic->aq_fw_ops->get_link_capabilities)
+		return 0;
+
+	caps_lo = nic->aq_fw_ops->get_link_capabilities(nic->aq_hw);
+
+	if (!(caps_lo & BIT(CAPS_LO_MACSEC)))
+		return 0;
+
+	nic->macsec_cfg = kzalloc(sizeof(*cfg), GFP_KERNEL);
+	if (!nic->macsec_cfg)
+		return -ENOMEM;
+
+	nic->ndev->features |= NETIF_F_HW_MACSEC;
+	nic->ndev->macsec_ops = &aq_macsec_ops;
+	mutex_init(&nic->macsec_mutex);
+
+	return 0;
+}
+
+void aq_macsec_free(struct aq_nic_s *nic)
+{
+	kfree(nic->macsec_cfg);
+	nic->macsec_cfg = NULL;
+}
+
+int aq_macsec_enable(struct aq_nic_s *nic)
+{
+	u32 ctl_ether_types[1] = { ETH_P_PAE };
+	struct macsec_msg_fw_response resp = { 0 };
+	struct macsec_msg_fw_request msg = { 0 };
+	struct aq_hw_s *hw = nic->aq_hw;
+	int num_ctl_ether_types = 0;
+	int index = 0, tbl_idx;
+	int ret;
+
+	if (!nic->macsec_cfg)
+		return 0;
+
+	mutex_lock(&nic->macsec_mutex);
+
+	if (nic->aq_fw_ops->send_macsec_req) {
+		struct macsec_cfg_request cfg = { 0 };
+
+		cfg.enabled = 1;
+		cfg.egress_threshold = 0xffffffff;
+		cfg.ingress_threshold = 0xffffffff;
+		cfg.interrupts_enabled = 1;
+
+		msg.msg_type = macsec_cfg_msg;
+		msg.cfg = cfg;
+
+		ret = nic->aq_fw_ops->send_macsec_req(hw, &msg, &resp);
+		if (ret)
+			goto unlock;
+	}
+
+	/* Init Ethertype bypass filters */
+	for (index = 0; index < ARRAY_SIZE(ctl_ether_types); index++) {
+		struct aq_mss_ingress_prectlf_record rx_prectlf_rec;
+		struct aq_mss_egress_ctlf_record tx_ctlf_rec;
+
+		if (ctl_ether_types[index] == 0)
+			continue;
+
+		memset(&tx_ctlf_rec, 0, sizeof(tx_ctlf_rec));
+		tx_ctlf_rec.eth_type = ctl_ether_types[index];
+		tx_ctlf_rec.match_type = 4; /* Match eth_type only */
+		tx_ctlf_rec.match_mask = 0xf; /* match for eth_type */
+		tx_ctlf_rec.action = 0; /* Bypass MACSEC modules */
+		tbl_idx = NUMROWS_EGRESSCTLFRECORD - num_ctl_ether_types - 1;
+		aq_mss_set_egress_ctlf_record(hw, &tx_ctlf_rec, tbl_idx);
+
+		memset(&rx_prectlf_rec, 0, sizeof(rx_prectlf_rec));
+		rx_prectlf_rec.eth_type = ctl_ether_types[index];
+		rx_prectlf_rec.match_type = 4; /* Match eth_type only */
+		rx_prectlf_rec.match_mask = 0xf; /* match for eth_type */
+		rx_prectlf_rec.action = 0; /* Bypass MACSEC modules */
+		tbl_idx =
+			NUMROWS_INGRESSPRECTLFRECORD - num_ctl_ether_types - 1;
+		aq_mss_set_ingress_prectlf_record(hw, &rx_prectlf_rec, tbl_idx);
+
+		num_ctl_ether_types++;
+	}
+
+	ret = aq_apply_macsec_cfg(nic);
+
+unlock:
+	mutex_unlock(&nic->macsec_mutex);
+	return ret;
+}
+
+void aq_macsec_work(struct aq_nic_s *nic)
+{
+	if (!nic->macsec_cfg)
+		return;
+
+	if (!netif_carrier_ok(nic->ndev))
+		return;
+
+	mutex_lock(&nic->macsec_mutex);
+	aq_check_txsa_expiration(nic);
+	mutex_unlock(&nic->macsec_mutex);
+}
+
+int aq_macsec_rx_sa_cnt(struct aq_nic_s *nic)
+{
+	struct aq_macsec_cfg *cfg = nic->macsec_cfg;
+	int i, cnt = 0;
+
+	if (!cfg)
+		return 0;
+
+	mutex_lock(&nic->macsec_mutex);
+
+	for (i = 0; i < AQ_MACSEC_MAX_SC; i++) {
+		if (!test_bit(i, &cfg->rxsc_idx_busy))
+			continue;
+		cnt += hweight_long(cfg->aq_rxsc[i].rx_sa_idx_busy);
+	}
+
+	mutex_unlock(&nic->macsec_mutex);
+	return cnt;
+}
+
+int aq_macsec_tx_sc_cnt(struct aq_nic_s *nic)
+{
+	int cnt;
+
+	if (!nic->macsec_cfg)
+		return 0;
+
+	mutex_lock(&nic->macsec_mutex);
+	cnt = hweight_long(nic->macsec_cfg->txsc_idx_busy);
+	mutex_unlock(&nic->macsec_mutex);
+
+	return cnt;
+}
+
+int aq_macsec_tx_sa_cnt(struct aq_nic_s *nic)
+{
+	struct aq_macsec_cfg *cfg = nic->macsec_cfg;
+	int i, cnt = 0;
+
+	if (!cfg)
+		return 0;
+
+	mutex_lock(&nic->macsec_mutex);
+
+	for (i = 0; i < AQ_MACSEC_MAX_SC; i++) {
+		if (!test_bit(i, &cfg->txsc_idx_busy))
+			continue;
+		cnt += hweight_long(cfg->aq_txsc[i].tx_sa_idx_busy);
+	}
+
+	mutex_unlock(&nic->macsec_mutex);
+	return cnt;
+}
+
+static int aq_macsec_update_stats(struct aq_nic_s *nic)
+{
+	struct aq_macsec_cfg *cfg = nic->macsec_cfg;
+	struct aq_hw_s *hw = nic->aq_hw;
+	struct aq_macsec_txsc *aq_txsc;
+	struct aq_macsec_rxsc *aq_rxsc;
+	int i, sa_idx, assoc_num;
+	int ret = 0;
+
+	aq_get_macsec_common_stats(hw, &cfg->stats);
+
+	for (i = 0; i < AQ_MACSEC_MAX_SC; i++) {
+		if (!(cfg->txsc_idx_busy & BIT(i)))
+			continue;
+		aq_txsc = &cfg->aq_txsc[i];
+
+		ret = aq_get_txsc_stats(hw, aq_txsc->hw_sc_idx,
+					&aq_txsc->stats);
+		if (ret)
+			return ret;
+
+		for (assoc_num = 0; assoc_num < MACSEC_NUM_AN; assoc_num++) {
+			if (!test_bit(assoc_num, &aq_txsc->tx_sa_idx_busy))
+				continue;
+			sa_idx = aq_txsc->hw_sc_idx | assoc_num;
+			ret = aq_get_txsa_stats(hw, sa_idx,
+					      &aq_txsc->tx_sa_stats[assoc_num]);
+			if (ret)
+				return ret;
+		}
+	}
+
+	for (i = 0; i < AQ_MACSEC_MAX_SC; i++) {
+		if (!(test_bit(i, &cfg->rxsc_idx_busy)))
+			continue;
+		aq_rxsc = &cfg->aq_rxsc[i];
+
+		for (assoc_num = 0; assoc_num < MACSEC_NUM_AN; assoc_num++) {
+			if (!test_bit(assoc_num, &aq_rxsc->rx_sa_idx_busy))
+				continue;
+			sa_idx = aq_rxsc->hw_sc_idx | assoc_num;
+
+			ret = aq_get_rxsa_stats(hw, sa_idx,
+					      &aq_rxsc->rx_sa_stats[assoc_num]);
+			if (ret)
+				return ret;
+		}
+	}
+
+	return ret;
+}
+
+u64 *aq_macsec_get_stats(struct aq_nic_s *nic, u64 *data)
+{
+	struct aq_macsec_cfg *cfg = nic->macsec_cfg;
+	struct aq_macsec_common_stats *common_stats;
+	struct aq_macsec_tx_sc_stats *txsc_stats;
+	struct aq_macsec_tx_sa_stats *txsa_stats;
+	struct aq_macsec_rx_sa_stats *rxsa_stats;
+	struct aq_macsec_txsc *aq_txsc;
+	struct aq_macsec_rxsc *aq_rxsc;
+	unsigned int assoc_num;
+	unsigned int sc_num;
+	unsigned int i = 0U;
+
+	if (!cfg)
+		return data;
+
+	mutex_lock(&nic->macsec_mutex);
+
+	aq_macsec_update_stats(nic);
+
+	common_stats = &cfg->stats;
+	data[i] = common_stats->in.ctl_pkts;
+	data[++i] = common_stats->in.tagged_miss_pkts;
+	data[++i] = common_stats->in.untagged_miss_pkts;
+	data[++i] = common_stats->in.notag_pkts;
+	data[++i] = common_stats->in.untagged_pkts;
+	data[++i] = common_stats->in.bad_tag_pkts;
+	data[++i] = common_stats->in.no_sci_pkts;
+	data[++i] = common_stats->in.unknown_sci_pkts;
+	data[++i] = common_stats->in.ctrl_prt_pass_pkts;
+	data[++i] = common_stats->in.unctrl_prt_pass_pkts;
+	data[++i] = common_stats->in.ctrl_prt_fail_pkts;
+	data[++i] = common_stats->in.unctrl_prt_fail_pkts;
+	data[++i] = common_stats->in.too_long_pkts;
+	data[++i] = common_stats->in.igpoc_ctl_pkts;
+	data[++i] = common_stats->in.ecc_error_pkts;
+	data[++i] = common_stats->in.unctrl_hit_drop_redir;
+	data[++i] = common_stats->out.ctl_pkts;
+	data[++i] = common_stats->out.unknown_sa_pkts;
+	data[++i] = common_stats->out.untagged_pkts;
+	data[++i] = common_stats->out.too_long;
+	data[++i] = common_stats->out.ecc_error_pkts;
+	data[++i] = common_stats->out.unctrl_hit_drop_redir;
+
+	for (sc_num = 0; sc_num < AQ_MACSEC_MAX_SC; sc_num++) {
+		if (!(test_bit(sc_num, &cfg->txsc_idx_busy)))
+			continue;
+
+		aq_txsc = &cfg->aq_txsc[sc_num];
+		txsc_stats = &aq_txsc->stats;
+
+		data[++i] = txsc_stats->sc_protected_pkts;
+		data[++i] = txsc_stats->sc_encrypted_pkts;
+		data[++i] = txsc_stats->sc_protected_octets;
+		data[++i] = txsc_stats->sc_encrypted_octets;
+
+		for (assoc_num = 0; assoc_num < MACSEC_NUM_AN; assoc_num++) {
+			if (!test_bit(assoc_num, &aq_txsc->tx_sa_idx_busy))
+				continue;
+
+			txsa_stats = &aq_txsc->tx_sa_stats[assoc_num];
+
+			data[++i] = txsa_stats->sa_hit_drop_redirect;
+			data[++i] = txsa_stats->sa_protected2_pkts;
+			data[++i] = txsa_stats->sa_protected_pkts;
+			data[++i] = txsa_stats->sa_encrypted_pkts;
+		}
+	}
+
+	for (sc_num = 0; sc_num < AQ_MACSEC_MAX_SC; sc_num++) {
+		if (!(test_bit(sc_num, &cfg->rxsc_idx_busy)))
+			continue;
+
+		aq_rxsc = &cfg->aq_rxsc[sc_num];
+
+		for (assoc_num = 0; assoc_num < MACSEC_NUM_AN; assoc_num++) {
+			if (!test_bit(assoc_num, &aq_rxsc->rx_sa_idx_busy))
+				continue;
+
+			rxsa_stats = &aq_rxsc->rx_sa_stats[assoc_num];
+
+			data[++i] = rxsa_stats->untagged_hit_pkts;
+			data[++i] = rxsa_stats->ctrl_hit_drop_redir_pkts;
+			data[++i] = rxsa_stats->not_using_sa;
+			data[++i] = rxsa_stats->unused_sa;
+			data[++i] = rxsa_stats->not_valid_pkts;
+			data[++i] = rxsa_stats->invalid_pkts;
+			data[++i] = rxsa_stats->ok_pkts;
+			data[++i] = rxsa_stats->late_pkts;
+			data[++i] = rxsa_stats->delayed_pkts;
+			data[++i] = rxsa_stats->unchecked_pkts;
+			data[++i] = rxsa_stats->validated_octets;
+			data[++i] = rxsa_stats->decrypted_octets;
+		}
+	}
+
+	i++;
+
+	data += i;
+
+	mutex_unlock(&nic->macsec_mutex);
+
+	return data;
+}
diff --git a/drivers/net/ethernet/aquantia/atlantic/aq_macsec.h b/drivers/net/ethernet/aquantia/atlantic/aq_macsec.h
new file mode 100644
index 000000000000..a47e2710487e
--- /dev/null
+++ b/drivers/net/ethernet/aquantia/atlantic/aq_macsec.h
@@ -0,0 +1,133 @@
+/* SPDX-License-Identifier: GPL-2.0-only */
+/* Atlantic Network Driver
+ * Copyright (C) 2020 Marvell International Ltd.
+ */
+
+#ifndef AQ_MACSEC_H
+#define AQ_MACSEC_H
+
+#include <linux/netdevice.h>
+#if IS_ENABLED(CONFIG_MACSEC)
+
+#include "net/macsec.h"
+
+struct aq_nic_s;
+
+#define AQ_MACSEC_MAX_SC 32
+#define AQ_MACSEC_MAX_SA 32
+
+enum aq_macsec_sc_sa {
+	aq_macsec_sa_sc_4sa_8sc,
+	aq_macsec_sa_sc_not_used,
+	aq_macsec_sa_sc_2sa_16sc,
+	aq_macsec_sa_sc_1sa_32sc,
+};
+
+struct aq_macsec_common_stats {
+	/* Ingress Common Counters */
+	struct {
+		u64 ctl_pkts;
+		u64 tagged_miss_pkts;
+		u64 untagged_miss_pkts;
+		u64 notag_pkts;
+		u64 untagged_pkts;
+		u64 bad_tag_pkts;
+		u64 no_sci_pkts;
+		u64 unknown_sci_pkts;
+		u64 ctrl_prt_pass_pkts;
+		u64 unctrl_prt_pass_pkts;
+		u64 ctrl_prt_fail_pkts;
+		u64 unctrl_prt_fail_pkts;
+		u64 too_long_pkts;
+		u64 igpoc_ctl_pkts;
+		u64 ecc_error_pkts;
+		u64 unctrl_hit_drop_redir;
+	} in;
+
+	/* Egress Common Counters */
+	struct {
+		u64 ctl_pkts;
+		u64 unknown_sa_pkts;
+		u64 untagged_pkts;
+		u64 too_long;
+		u64 ecc_error_pkts;
+		u64 unctrl_hit_drop_redir;
+	} out;
+};
+
+/* Ingress SA Counters */
+struct aq_macsec_rx_sa_stats {
+	u64 untagged_hit_pkts;
+	u64 ctrl_hit_drop_redir_pkts;
+	u64 not_using_sa;
+	u64 unused_sa;
+	u64 not_valid_pkts;
+	u64 invalid_pkts;
+	u64 ok_pkts;
+	u64 late_pkts;
+	u64 delayed_pkts;
+	u64 unchecked_pkts;
+	u64 validated_octets;
+	u64 decrypted_octets;
+};
+
+/* Egress SA Counters */
+struct aq_macsec_tx_sa_stats {
+	u64 sa_hit_drop_redirect;
+	u64 sa_protected2_pkts;
+	u64 sa_protected_pkts;
+	u64 sa_encrypted_pkts;
+};
+
+/* Egress SC Counters */
+struct aq_macsec_tx_sc_stats {
+	u64 sc_protected_pkts;
+	u64 sc_encrypted_pkts;
+	u64 sc_protected_octets;
+	u64 sc_encrypted_octets;
+};
+
+struct aq_macsec_txsc {
+	u32 hw_sc_idx;
+	unsigned long tx_sa_idx_busy;
+	const struct macsec_secy *sw_secy;
+	u8 tx_sa_key[MACSEC_NUM_AN][MACSEC_MAX_KEY_LEN];
+	struct aq_macsec_tx_sc_stats stats;
+	struct aq_macsec_tx_sa_stats tx_sa_stats[MACSEC_NUM_AN];
+};
+
+struct aq_macsec_rxsc {
+	u32 hw_sc_idx;
+	unsigned long rx_sa_idx_busy;
+	const struct macsec_secy *sw_secy;
+	const struct macsec_rx_sc *sw_rxsc;
+	u8 rx_sa_key[MACSEC_NUM_AN][MACSEC_MAX_KEY_LEN];
+	struct aq_macsec_rx_sa_stats rx_sa_stats[MACSEC_NUM_AN];
+};
+
+struct aq_macsec_cfg {
+	enum aq_macsec_sc_sa sc_sa;
+	/* Egress channel configuration */
+	unsigned long txsc_idx_busy;
+	struct aq_macsec_txsc aq_txsc[AQ_MACSEC_MAX_SC];
+	/* Ingress channel configuration */
+	unsigned long rxsc_idx_busy;
+	struct aq_macsec_rxsc aq_rxsc[AQ_MACSEC_MAX_SC];
+	/* Statistics / counters */
+	struct aq_macsec_common_stats stats;
+};
+
+extern const struct macsec_ops aq_macsec_ops;
+
+int aq_macsec_init(struct aq_nic_s *nic);
+void aq_macsec_free(struct aq_nic_s *nic);
+int aq_macsec_enable(struct aq_nic_s *nic);
+void aq_macsec_work(struct aq_nic_s *nic);
+u64 *aq_macsec_get_stats(struct aq_nic_s *nic, u64 *data);
+int aq_macsec_rx_sa_cnt(struct aq_nic_s *nic);
+int aq_macsec_tx_sc_cnt(struct aq_nic_s *nic);
+int aq_macsec_tx_sa_cnt(struct aq_nic_s *nic);
+
+#endif
+
+#endif /* AQ_MACSEC_H */
diff --git a/drivers/net/ethernet/aquantia/atlantic/aq_main.c b/drivers/net/ethernet/aquantia/atlantic/aq_main.c
index 5d6c40d86775..b565189e5913 100644
--- a/drivers/net/ethernet/aquantia/atlantic/aq_main.c
+++ b/drivers/net/ethernet/aquantia/atlantic/aq_main.c
@@ -1,10 +1,8 @@
-/*
- * aQuantia Corporation Network Driver
- * Copyright (C) 2014-2017 aQuantia Corporation. All rights reserved
+// SPDX-License-Identifier: GPL-2.0-only
+/* Atlantic Network Driver
  *
- * This program is free software; you can redistribute it and/or modify it
- * under the terms and conditions of the GNU General Public License,
- * version 2, as published by the Free Software Foundation.
+ * Copyright (C) 2014-2019 aQuantia Corporation
+ * Copyright (C) 2019-2020 Marvell International Ltd.
  */
 
 /* File aq_main.c: Main file for aQuantia Linux driver. */
@@ -13,92 +11,141 @@
 #include "aq_nic.h"
 #include "aq_pci_func.h"
 #include "aq_ethtool.h"
-#include "hw_atl/hw_atl_a0.h"
-#include "hw_atl/hw_atl_b0.h"
+#include "aq_ptp.h"
+#include "aq_filters.h"
+#include "aq_hw_utils.h"
+#include "aq_vec.h"
 
 #include <linux/netdevice.h>
 #include <linux/module.h>
-
-static const struct pci_device_id aq_pci_tbl[] = {
-	{ PCI_VDEVICE(AQUANTIA, HW_ATL_DEVICE_ID_0001), },
-	{ PCI_VDEVICE(AQUANTIA, HW_ATL_DEVICE_ID_D100), },
-	{ PCI_VDEVICE(AQUANTIA, HW_ATL_DEVICE_ID_D107), },
-	{ PCI_VDEVICE(AQUANTIA, HW_ATL_DEVICE_ID_D108), },
-	{ PCI_VDEVICE(AQUANTIA, HW_ATL_DEVICE_ID_D109), },
-	{}
-};
-
-MODULE_DEVICE_TABLE(pci, aq_pci_tbl);
+#include <linux/ip.h>
+#include <linux/udp.h>
+#include <net/pkt_cls.h>
+#include <net/pkt_sched.h>
+#include <linux/filter.h>
 
 MODULE_LICENSE("GPL v2");
-MODULE_VERSION(AQ_CFG_DRV_VERSION);
 MODULE_AUTHOR(AQ_CFG_DRV_AUTHOR);
 MODULE_DESCRIPTION(AQ_CFG_DRV_DESC);
 
-static struct aq_hw_ops *aq_pci_probe_get_hw_ops_by_id(struct pci_dev *pdev)
-{
-	struct aq_hw_ops *ops = NULL;
+DEFINE_STATIC_KEY_FALSE(aq_xdp_locking_key);
+EXPORT_SYMBOL(aq_xdp_locking_key);
+
+static const char aq_ndev_driver_name[] = AQ_CFG_DRV_NAME;
+
+static const struct net_device_ops aq_ndev_ops;
 
-	ops = hw_atl_a0_get_ops_by_id(pdev);
-	if (!ops)
-		ops = hw_atl_b0_get_ops_by_id(pdev);
+static struct workqueue_struct *aq_ndev_wq;
 
-	return ops;
+void aq_ndev_schedule_work(struct work_struct *work)
+{
+	queue_work(aq_ndev_wq, work);
 }
 
-static int aq_ndev_open(struct net_device *ndev)
+struct net_device *aq_ndev_alloc(void)
 {
+	struct net_device *ndev = NULL;
 	struct aq_nic_s *aq_nic = NULL;
+
+	ndev = alloc_etherdev_mq(sizeof(struct aq_nic_s), AQ_HW_QUEUES_MAX);
+	if (!ndev)
+		return NULL;
+
+	aq_nic = netdev_priv(ndev);
+	aq_nic->ndev = ndev;
+	ndev->netdev_ops = &aq_ndev_ops;
+	ndev->ethtool_ops = &aq_ethtool_ops;
+
+	return ndev;
+}
+
+int aq_ndev_open(struct net_device *ndev)
+{
+	struct aq_nic_s *aq_nic = netdev_priv(ndev);
 	int err = 0;
 
-	aq_nic = aq_nic_alloc_hot(ndev);
-	if (!aq_nic) {
-		err = -ENOMEM;
-		goto err_exit;
-	}
 	err = aq_nic_init(aq_nic);
 	if (err < 0)
 		goto err_exit;
-	err = aq_nic_start(aq_nic);
+
+	err = aq_reapply_rxnfc_all_rules(aq_nic);
+	if (err < 0)
+		goto err_exit;
+
+	err = aq_filters_vlans_update(aq_nic);
 	if (err < 0)
 		goto err_exit;
 
+	err = aq_nic_start(aq_nic);
+	if (err < 0) {
+		aq_nic_stop(aq_nic);
+		goto err_exit;
+	}
+
 err_exit:
 	if (err < 0)
-		aq_nic_deinit(aq_nic);
+		aq_nic_deinit(aq_nic, true);
+
 	return err;
 }
 
-static int aq_ndev_close(struct net_device *ndev)
+int aq_ndev_close(struct net_device *ndev)
 {
-	int err = 0;
 	struct aq_nic_s *aq_nic = netdev_priv(ndev);
+	int err = 0;
 
 	err = aq_nic_stop(aq_nic);
-	if (err < 0)
-		goto err_exit;
-	aq_nic_deinit(aq_nic);
-	aq_nic_free_hot_resources(aq_nic);
+	aq_nic_deinit(aq_nic, true);
 
-err_exit:
 	return err;
 }
 
-static int aq_ndev_start_xmit(struct sk_buff *skb, struct net_device *ndev)
+static netdev_tx_t aq_ndev_start_xmit(struct sk_buff *skb, struct net_device *ndev)
 {
 	struct aq_nic_s *aq_nic = netdev_priv(ndev);
 
+#if IS_REACHABLE(CONFIG_PTP_1588_CLOCK)
+	if (unlikely(aq_utils_obj_test(&aq_nic->flags, AQ_NIC_PTP_DPATH_UP))) {
+		/* Hardware adds the Timestamp for PTPv2 802.AS1
+		 * and PTPv2 IPv4 UDP.
+		 * We have to push even general 320 port messages to the ptp
+		 * queue explicitly. This is a limitation of current firmware
+		 * and hardware PTP design of the chip. Otherwise ptp stream
+		 * will fail to sync
+		 */
+		if (unlikely(skb_shinfo(skb)->tx_flags & SKBTX_HW_TSTAMP) ||
+		    unlikely((ip_hdr(skb)->version == 4) &&
+			     (ip_hdr(skb)->protocol == IPPROTO_UDP) &&
+			     ((udp_hdr(skb)->dest == htons(319)) ||
+			      (udp_hdr(skb)->dest == htons(320)))) ||
+		    unlikely(eth_hdr(skb)->h_proto == htons(ETH_P_1588)))
+			return aq_ptp_xmit(aq_nic, skb);
+	}
+#endif
+
 	return aq_nic_xmit(aq_nic, skb);
 }
 
 static int aq_ndev_change_mtu(struct net_device *ndev, int new_mtu)
 {
+	int new_frame_size = new_mtu + ETH_HLEN + ETH_FCS_LEN;
 	struct aq_nic_s *aq_nic = netdev_priv(ndev);
-	int err = aq_nic_set_mtu(aq_nic, new_mtu + ETH_HLEN);
+	struct bpf_prog *prog;
+	int err;
+
+	prog = READ_ONCE(aq_nic->xdp_prog);
+	if (prog && !prog->aux->xdp_has_frags &&
+	    new_frame_size > AQ_CFG_RX_FRAME_MAX) {
+		netdev_err(ndev, "Illegal MTU %d for XDP prog without frags\n",
+			   ndev->mtu);
+		return -EOPNOTSUPP;
+	}
+
+	err = aq_nic_set_mtu(aq_nic, new_mtu + ETH_HLEN);
 
 	if (err < 0)
 		goto err_exit;
-	ndev->mtu = new_mtu;
+	WRITE_ONCE(ndev->mtu, new_mtu);
 
 err_exit:
 	return err;
@@ -107,24 +154,85 @@ err_exit:
 static int aq_ndev_set_features(struct net_device *ndev,
 				netdev_features_t features)
 {
+	bool is_vlan_tx_insert = !!(features & NETIF_F_HW_VLAN_CTAG_TX);
+	bool is_vlan_rx_strip = !!(features & NETIF_F_HW_VLAN_CTAG_RX);
 	struct aq_nic_s *aq_nic = netdev_priv(ndev);
-	struct aq_nic_cfg_s *aq_cfg = aq_nic_get_cfg(aq_nic);
+	bool need_ndev_restart = false;
+	struct aq_nic_cfg_s *aq_cfg;
 	bool is_lro = false;
+	int err = 0;
+
+	aq_cfg = aq_nic_get_cfg(aq_nic);
+
+	if (!(features & NETIF_F_NTUPLE)) {
+		if (aq_nic->ndev->features & NETIF_F_NTUPLE) {
+			err = aq_clear_rxnfc_all_rules(aq_nic);
+			if (unlikely(err))
+				goto err_exit;
+		}
+	}
+	if (!(features & NETIF_F_HW_VLAN_CTAG_FILTER)) {
+		if (aq_nic->ndev->features & NETIF_F_HW_VLAN_CTAG_FILTER) {
+			err = aq_filters_vlan_offload_off(aq_nic);
+			if (unlikely(err))
+				goto err_exit;
+		}
+	}
+
+	aq_cfg->features = features;
 
-	if (aq_cfg->hw_features & NETIF_F_LRO) {
+	if (aq_cfg->aq_hw_caps->hw_features & NETIF_F_LRO) {
 		is_lro = features & NETIF_F_LRO;
 
 		if (aq_cfg->is_lro != is_lro) {
 			aq_cfg->is_lro = is_lro;
-
-			if (netif_running(ndev)) {
-				aq_ndev_close(ndev);
-				aq_ndev_open(ndev);
-			}
+			need_ndev_restart = true;
 		}
 	}
 
-	return 0;
+	if ((aq_nic->ndev->features ^ features) & NETIF_F_RXCSUM) {
+		err = aq_nic->aq_hw_ops->hw_set_offload(aq_nic->aq_hw,
+							aq_cfg);
+
+		if (unlikely(err))
+			goto err_exit;
+	}
+
+	if (aq_cfg->is_vlan_rx_strip != is_vlan_rx_strip) {
+		aq_cfg->is_vlan_rx_strip = is_vlan_rx_strip;
+		need_ndev_restart = true;
+	}
+	if (aq_cfg->is_vlan_tx_insert != is_vlan_tx_insert) {
+		aq_cfg->is_vlan_tx_insert = is_vlan_tx_insert;
+		need_ndev_restart = true;
+	}
+
+	if (need_ndev_restart && netif_running(ndev)) {
+		aq_ndev_close(ndev);
+		aq_ndev_open(ndev);
+	}
+
+err_exit:
+	return err;
+}
+
+static netdev_features_t aq_ndev_fix_features(struct net_device *ndev,
+					      netdev_features_t features)
+{
+	struct aq_nic_s *aq_nic = netdev_priv(ndev);
+	struct bpf_prog *prog;
+
+	if (!(features & NETIF_F_RXCSUM))
+		features &= ~NETIF_F_LRO;
+
+	prog = READ_ONCE(aq_nic->xdp_prog);
+	if (prog && !prog->aux->xdp_has_frags &&
+	    aq_nic->xdp_prog && features & NETIF_F_LRO) {
+		netdev_err(ndev, "LRO is not supported with single buffer XDP, disabling\n");
+		features &= ~NETIF_F_LRO;
+	}
+
+	return features;
 }
 
 static int aq_ndev_set_mac_address(struct net_device *ndev, void *addr)
@@ -146,90 +254,288 @@ err_exit:
 static void aq_ndev_set_multicast_settings(struct net_device *ndev)
 {
 	struct aq_nic_s *aq_nic = netdev_priv(ndev);
-	int err = 0;
 
-	err = aq_nic_set_packet_filter(aq_nic, ndev->flags);
-	if (err < 0)
-		goto err_exit;
+	(void)aq_nic_set_multicast_list(aq_nic, ndev);
+}
 
-	if (netdev_mc_count(ndev)) {
-		err = aq_nic_set_multicast_list(aq_nic, ndev);
-		if (err < 0)
-			goto err_exit;
+#if IS_REACHABLE(CONFIG_PTP_1588_CLOCK)
+static int aq_ndev_config_hwtstamp(struct aq_nic_s *aq_nic,
+				   struct hwtstamp_config *config)
+{
+	switch (config->tx_type) {
+	case HWTSTAMP_TX_OFF:
+	case HWTSTAMP_TX_ON:
+		break;
+	default:
+		return -ERANGE;
 	}
 
-err_exit:;
+	switch (config->rx_filter) {
+	case HWTSTAMP_FILTER_PTP_V2_L4_EVENT:
+	case HWTSTAMP_FILTER_PTP_V2_L4_SYNC:
+	case HWTSTAMP_FILTER_PTP_V2_L4_DELAY_REQ:
+	case HWTSTAMP_FILTER_PTP_V2_L2_EVENT:
+	case HWTSTAMP_FILTER_PTP_V2_L2_SYNC:
+	case HWTSTAMP_FILTER_PTP_V2_L2_DELAY_REQ:
+	case HWTSTAMP_FILTER_PTP_V2_SYNC:
+	case HWTSTAMP_FILTER_PTP_V2_DELAY_REQ:
+		config->rx_filter = HWTSTAMP_FILTER_PTP_V2_EVENT;
+		break;
+	case HWTSTAMP_FILTER_PTP_V2_EVENT:
+	case HWTSTAMP_FILTER_NONE:
+		break;
+	default:
+		return -ERANGE;
+	}
+
+	return aq_ptp_hwtstamp_config_set(aq_nic->aq_ptp, config);
 }
+#endif
 
-static const struct net_device_ops aq_ndev_ops = {
-	.ndo_open = aq_ndev_open,
-	.ndo_stop = aq_ndev_close,
-	.ndo_start_xmit = aq_ndev_start_xmit,
-	.ndo_set_rx_mode = aq_ndev_set_multicast_settings,
-	.ndo_change_mtu = aq_ndev_change_mtu,
-	.ndo_set_mac_address = aq_ndev_set_mac_address,
-	.ndo_set_features = aq_ndev_set_features
-};
+static int aq_ndev_hwtstamp_set(struct aq_nic_s *aq_nic, struct ifreq *ifr)
+{
+	struct hwtstamp_config config;
+#if IS_REACHABLE(CONFIG_PTP_1588_CLOCK)
+	int ret_val;
+#endif
+
+	if (!aq_nic->aq_ptp)
+		return -EOPNOTSUPP;
+
+	if (copy_from_user(&config, ifr->ifr_data, sizeof(config)))
+		return -EFAULT;
+#if IS_REACHABLE(CONFIG_PTP_1588_CLOCK)
+	ret_val = aq_ndev_config_hwtstamp(aq_nic, &config);
+	if (ret_val)
+		return ret_val;
+#endif
+
+	return copy_to_user(ifr->ifr_data, &config, sizeof(config)) ?
+	       -EFAULT : 0;
+}
 
-static int aq_pci_probe(struct pci_dev *pdev,
-			const struct pci_device_id *pci_id)
+#if IS_REACHABLE(CONFIG_PTP_1588_CLOCK)
+static int aq_ndev_hwtstamp_get(struct aq_nic_s *aq_nic, struct ifreq *ifr)
 {
-	struct aq_hw_ops *aq_hw_ops = NULL;
-	struct aq_pci_func_s *aq_pci_func = NULL;
-	int err = 0;
+	struct hwtstamp_config config;
 
-	err = pci_enable_device(pdev);
-	if (err < 0)
-		goto err_exit;
-	aq_hw_ops = aq_pci_probe_get_hw_ops_by_id(pdev);
-	aq_pci_func = aq_pci_func_alloc(aq_hw_ops, pdev,
-					&aq_ndev_ops, &aq_ethtool_ops);
-	if (!aq_pci_func) {
-		err = -ENOMEM;
-		goto err_exit;
-	}
-	err = aq_pci_func_init(aq_pci_func);
-	if (err < 0)
-		goto err_exit;
+	if (!aq_nic->aq_ptp)
+		return -EOPNOTSUPP;
 
-err_exit:
-	if (err < 0) {
-		if (aq_pci_func)
-			aq_pci_func_free(aq_pci_func);
+	aq_ptp_hwtstamp_config_get(aq_nic->aq_ptp, &config);
+	return copy_to_user(ifr->ifr_data, &config, sizeof(config)) ?
+	       -EFAULT : 0;
+}
+#endif
+
+static int aq_ndev_ioctl(struct net_device *netdev, struct ifreq *ifr, int cmd)
+{
+	struct aq_nic_s *aq_nic = netdev_priv(netdev);
+
+	switch (cmd) {
+	case SIOCSHWTSTAMP:
+		return aq_ndev_hwtstamp_set(aq_nic, ifr);
+
+#if IS_REACHABLE(CONFIG_PTP_1588_CLOCK)
+	case SIOCGHWTSTAMP:
+		return aq_ndev_hwtstamp_get(aq_nic, ifr);
+#endif
 	}
-	return err;
+
+	return -EOPNOTSUPP;
 }
 
-static void aq_pci_remove(struct pci_dev *pdev)
+static int aq_ndo_vlan_rx_add_vid(struct net_device *ndev, __be16 proto,
+				  u16 vid)
 {
-	struct aq_pci_func_s *aq_pci_func = pci_get_drvdata(pdev);
+	struct aq_nic_s *aq_nic = netdev_priv(ndev);
 
-	aq_pci_func_deinit(aq_pci_func);
-	aq_pci_func_free(aq_pci_func);
+	if (!aq_nic->aq_hw_ops->hw_filter_vlan_set)
+		return -EOPNOTSUPP;
+
+	set_bit(vid, aq_nic->active_vlans);
+
+	return aq_filters_vlans_update(aq_nic);
 }
 
-static int aq_pci_suspend(struct pci_dev *pdev, pm_message_t pm_msg)
+static int aq_ndo_vlan_rx_kill_vid(struct net_device *ndev, __be16 proto,
+				   u16 vid)
 {
-	struct aq_pci_func_s *aq_pci_func = pci_get_drvdata(pdev);
+	struct aq_nic_s *aq_nic = netdev_priv(ndev);
+
+	if (!aq_nic->aq_hw_ops->hw_filter_vlan_set)
+		return -EOPNOTSUPP;
+
+	clear_bit(vid, aq_nic->active_vlans);
 
-	return aq_pci_func_change_pm_state(aq_pci_func, &pm_msg);
+	if (-ENOENT == aq_del_fvlan_by_vlan(aq_nic, vid))
+		return aq_filters_vlans_update(aq_nic);
+
+	return 0;
 }
 
-static int aq_pci_resume(struct pci_dev *pdev)
+static int aq_validate_mqprio_opt(struct aq_nic_s *self,
+				  struct tc_mqprio_qopt_offload *mqprio,
+				  const unsigned int num_tc)
 {
-	struct aq_pci_func_s *aq_pci_func = pci_get_drvdata(pdev);
-	pm_message_t pm_msg = PMSG_RESTORE;
+	const bool has_min_rate = !!(mqprio->flags & TC_MQPRIO_F_MIN_RATE);
+	struct aq_nic_cfg_s *aq_nic_cfg = aq_nic_get_cfg(self);
+	const unsigned int tcs_max = min_t(u8, aq_nic_cfg->aq_hw_caps->tcs_max,
+					   AQ_CFG_TCS_MAX);
+
+	if (num_tc > tcs_max) {
+		netdev_err(self->ndev, "Too many TCs requested\n");
+		return -EOPNOTSUPP;
+	}
 
-	return aq_pci_func_change_pm_state(aq_pci_func, &pm_msg);
+	if (num_tc != 0 && !is_power_of_2(num_tc)) {
+		netdev_err(self->ndev, "TC count should be power of 2\n");
+		return -EOPNOTSUPP;
+	}
+
+	if (has_min_rate && !ATL_HW_IS_CHIP_FEATURE(self->aq_hw, ANTIGUA)) {
+		netdev_err(self->ndev, "Min tx rate is not supported\n");
+		return -EOPNOTSUPP;
+	}
+
+	return 0;
 }
 
-static struct pci_driver aq_pci_ops = {
-	.name = AQ_CFG_DRV_NAME,
-	.id_table = aq_pci_tbl,
-	.probe = aq_pci_probe,
-	.remove = aq_pci_remove,
-	.suspend = aq_pci_suspend,
-	.resume = aq_pci_resume,
+static int aq_ndo_setup_tc(struct net_device *dev, enum tc_setup_type type,
+			   void *type_data)
+{
+	struct tc_mqprio_qopt_offload *mqprio = type_data;
+	struct aq_nic_s *aq_nic = netdev_priv(dev);
+	bool has_min_rate;
+	bool has_max_rate;
+	int err;
+	int i;
+
+	if (type != TC_SETUP_QDISC_MQPRIO)
+		return -EOPNOTSUPP;
+
+	has_min_rate = !!(mqprio->flags & TC_MQPRIO_F_MIN_RATE);
+	has_max_rate = !!(mqprio->flags & TC_MQPRIO_F_MAX_RATE);
+
+	err = aq_validate_mqprio_opt(aq_nic, mqprio, mqprio->qopt.num_tc);
+	if (err)
+		return err;
+
+	for (i = 0; i < mqprio->qopt.num_tc; i++) {
+		if (has_max_rate) {
+			u64 max_rate = mqprio->max_rate[i];
+
+			do_div(max_rate, AQ_MBPS_DIVISOR);
+			aq_nic_setup_tc_max_rate(aq_nic, i, (u32)max_rate);
+		}
+
+		if (has_min_rate) {
+			u64 min_rate = mqprio->min_rate[i];
+
+			do_div(min_rate, AQ_MBPS_DIVISOR);
+			aq_nic_setup_tc_min_rate(aq_nic, i, (u32)min_rate);
+		}
+	}
+
+	return aq_nic_setup_tc_mqprio(aq_nic, mqprio->qopt.num_tc,
+				      mqprio->qopt.prio_tc_map);
+}
+
+static int aq_xdp_setup(struct net_device *ndev, struct bpf_prog *prog,
+			struct netlink_ext_ack *extack)
+{
+	bool need_update, running = netif_running(ndev);
+	struct aq_nic_s *aq_nic = netdev_priv(ndev);
+	struct bpf_prog *old_prog;
+
+	if (prog && !prog->aux->xdp_has_frags) {
+		if (ndev->mtu > AQ_CFG_RX_FRAME_MAX) {
+			NL_SET_ERR_MSG_MOD(extack,
+					   "prog does not support XDP frags");
+			return -EOPNOTSUPP;
+		}
+
+		if (prog && ndev->features & NETIF_F_LRO) {
+			netdev_err(ndev,
+				   "LRO is not supported with single buffer XDP, disabling\n");
+			ndev->features &= ~NETIF_F_LRO;
+		}
+	}
+
+	need_update = !!aq_nic->xdp_prog != !!prog;
+	if (running && need_update)
+		aq_ndev_close(ndev);
+
+	old_prog = xchg(&aq_nic->xdp_prog, prog);
+	if (old_prog)
+		bpf_prog_put(old_prog);
+
+	if (!old_prog && prog)
+		static_branch_inc(&aq_xdp_locking_key);
+	else if (old_prog && !prog)
+		static_branch_dec(&aq_xdp_locking_key);
+
+	if (running && need_update)
+		return aq_ndev_open(ndev);
+
+	return 0;
+}
+
+static int aq_xdp(struct net_device *dev, struct netdev_bpf *xdp)
+{
+	switch (xdp->command) {
+	case XDP_SETUP_PROG:
+		return aq_xdp_setup(dev, xdp->prog, xdp->extack);
+	default:
+		return -EINVAL;
+	}
+}
+
+static const struct net_device_ops aq_ndev_ops = {
+	.ndo_open = aq_ndev_open,
+	.ndo_stop = aq_ndev_close,
+	.ndo_start_xmit = aq_ndev_start_xmit,
+	.ndo_set_rx_mode = aq_ndev_set_multicast_settings,
+	.ndo_change_mtu = aq_ndev_change_mtu,
+	.ndo_set_mac_address = aq_ndev_set_mac_address,
+	.ndo_set_features = aq_ndev_set_features,
+	.ndo_fix_features = aq_ndev_fix_features,
+	.ndo_eth_ioctl = aq_ndev_ioctl,
+	.ndo_vlan_rx_add_vid = aq_ndo_vlan_rx_add_vid,
+	.ndo_vlan_rx_kill_vid = aq_ndo_vlan_rx_kill_vid,
+	.ndo_setup_tc = aq_ndo_setup_tc,
+	.ndo_bpf = aq_xdp,
+	.ndo_xdp_xmit = aq_xdp_xmit,
 };
 
-module_pci_driver(aq_pci_ops);
+static int __init aq_ndev_init_module(void)
+{
+	int ret;
+
+	aq_ndev_wq = create_singlethread_workqueue(aq_ndev_driver_name);
+	if (!aq_ndev_wq) {
+		pr_err("Failed to create workqueue\n");
+		return -ENOMEM;
+	}
+
+	ret = aq_pci_func_register_driver();
+	if (ret) {
+		destroy_workqueue(aq_ndev_wq);
+		return ret;
+	}
+
+	return 0;
+}
+
+static void __exit aq_ndev_exit_module(void)
+{
+	aq_pci_func_unregister_driver();
+
+	if (aq_ndev_wq) {
+		destroy_workqueue(aq_ndev_wq);
+		aq_ndev_wq = NULL;
+	}
+}
+
+module_init(aq_ndev_init_module);
+module_exit(aq_ndev_exit_module);
diff --git a/drivers/net/ethernet/aquantia/atlantic/aq_main.h b/drivers/net/ethernet/aquantia/atlantic/aq_main.h
index 9748e7e575e0..a78c1a168d8e 100644
--- a/drivers/net/ethernet/aquantia/atlantic/aq_main.h
+++ b/drivers/net/ethernet/aquantia/atlantic/aq_main.h
@@ -1,10 +1,7 @@
+/* SPDX-License-Identifier: GPL-2.0-only */
 /*
  * aQuantia Corporation Network Driver
  * Copyright (C) 2014-2017 aQuantia Corporation. All rights reserved
- *
- * This program is free software; you can redistribute it and/or modify it
- * under the terms and conditions of the GNU General Public License,
- * version 2, as published by the Free Software Foundation.
  */
 
 /* File aq_main.h: Main file for aQuantia Linux driver. */
@@ -13,5 +10,13 @@
 #define AQ_MAIN_H
 
 #include "aq_common.h"
+#include "aq_nic.h"
+
+DECLARE_STATIC_KEY_FALSE(aq_xdp_locking_key);
+
+void aq_ndev_schedule_work(struct work_struct *work);
+struct net_device *aq_ndev_alloc(void);
+int aq_ndev_open(struct net_device *ndev);
+int aq_ndev_close(struct net_device *ndev);
 
 #endif /* AQ_MAIN_H */
diff --git a/drivers/net/ethernet/aquantia/atlantic/aq_nic.c b/drivers/net/ethernet/aquantia/atlantic/aq_nic.c
index 9ee1c5016784..b24eaa5283fa 100644
--- a/drivers/net/ethernet/aquantia/atlantic/aq_nic.c
+++ b/drivers/net/ethernet/aquantia/atlantic/aq_nic.c
@@ -1,10 +1,8 @@
-/*
- * aQuantia Corporation Network Driver
- * Copyright (C) 2014-2017 aQuantia Corporation. All rights reserved
+// SPDX-License-Identifier: GPL-2.0-only
+/* Atlantic Network Driver
  *
- * This program is free software; you can redistribute it and/or modify it
- * under the terms and conditions of the GNU General Public License,
- * version 2, as published by the Free Software Foundation.
+ * Copyright (C) 2014-2019 aQuantia Corporation
+ * Copyright (C) 2019-2020 Marvell International Ltd.
  */
 
 /* File aq_nic.c: Definition of common code for NIC. */
@@ -14,8 +12,13 @@
 #include "aq_vec.h"
 #include "aq_hw.h"
 #include "aq_pci_func.h"
-#include "aq_nic_internal.h"
+#include "aq_macsec.h"
+#include "aq_main.h"
+#include "aq_phy.h"
+#include "aq_ptp.h"
+#include "aq_filters.h"
 
+#include <linux/moduleparam.h>
 #include <linux/netdevice.h>
 #include <linux/etherdevice.h>
 #include <linux/timer.h>
@@ -23,20 +26,36 @@
 #include <linux/ip.h>
 #include <linux/tcp.h>
 #include <net/ip.h>
+#include <net/pkt_cls.h>
+
+static unsigned int aq_itr = AQ_CFG_INTERRUPT_MODERATION_AUTO;
+module_param_named(aq_itr, aq_itr, uint, 0644);
+MODULE_PARM_DESC(aq_itr, "Interrupt throttling mode");
+
+static unsigned int aq_itr_tx;
+module_param_named(aq_itr_tx, aq_itr_tx, uint, 0644);
+MODULE_PARM_DESC(aq_itr_tx, "TX interrupt throttle rate");
+
+static unsigned int aq_itr_rx;
+module_param_named(aq_itr_rx, aq_itr_rx, uint, 0644);
+MODULE_PARM_DESC(aq_itr_rx, "RX interrupt throttle rate");
+
+static void aq_nic_update_ndev_stats(struct aq_nic_s *self);
 
 static void aq_nic_rss_init(struct aq_nic_s *self, unsigned int num_rss_queues)
 {
-	struct aq_nic_cfg_s *cfg = &self->aq_nic_cfg;
-	struct aq_rss_parameters *rss_params = &cfg->aq_rss;
-	int i = 0;
-
-	static u8 rss_key[40] = {
+	static u8 rss_key[AQ_CFG_RSS_HASHKEY_SIZE] = {
 		0x1e, 0xad, 0x71, 0x87, 0x65, 0xfc, 0x26, 0x7d,
 		0x0d, 0x45, 0x67, 0x74, 0xcd, 0x06, 0x1a, 0x18,
 		0xb6, 0xc1, 0xf0, 0xc7, 0xbb, 0x18, 0xbe, 0xf8,
 		0x19, 0x13, 0x4b, 0xa9, 0xd0, 0x3e, 0xfe, 0x70,
 		0x25, 0x03, 0xab, 0x50, 0x6a, 0x8b, 0x82, 0x0c
 	};
+	struct aq_nic_cfg_s *cfg = &self->aq_nic_cfg;
+	struct aq_rss_parameters *rss_params;
+	int i = 0;
+
+	rss_params = &cfg->aq_rss;
 
 	rss_params->hash_secret_key_size = sizeof(rss_key);
 	memcpy(rss_params->hash_secret_key, rss_key, sizeof(rss_key));
@@ -46,302 +65,330 @@ static void aq_nic_rss_init(struct aq_nic_s *self, unsigned int num_rss_queues)
 		rss_params->indirection_table[i] = i & (num_rss_queues - 1);
 }
 
-/* Fills aq_nic_cfg with valid defaults */
-static void aq_nic_cfg_init_defaults(struct aq_nic_s *self)
+/* Recalculate the number of vectors */
+static void aq_nic_cfg_update_num_vecs(struct aq_nic_s *self)
 {
 	struct aq_nic_cfg_s *cfg = &self->aq_nic_cfg;
 
-	cfg->aq_hw_caps = &self->aq_hw_caps;
+	cfg->vecs = min(cfg->aq_hw_caps->vecs, AQ_CFG_VECS_DEF);
+	cfg->vecs = min(cfg->vecs, num_online_cpus());
+	if (self->irqvecs > AQ_HW_SERVICE_IRQS)
+		cfg->vecs = min(cfg->vecs, self->irqvecs - AQ_HW_SERVICE_IRQS);
+	/* cfg->vecs should be power of 2 for RSS */
+	cfg->vecs = rounddown_pow_of_two(cfg->vecs);
 
-	cfg->vecs = AQ_CFG_VECS_DEF;
-	cfg->tcs = AQ_CFG_TCS_DEF;
+	if (ATL_HW_IS_CHIP_FEATURE(self->aq_hw, ANTIGUA)) {
+		if (cfg->tcs > 2)
+			cfg->vecs = min(cfg->vecs, 4U);
+	}
 
-	cfg->rxds = AQ_CFG_RXDS_DEF;
-	cfg->txds = AQ_CFG_TXDS_DEF;
+	if (cfg->vecs <= 4)
+		cfg->tc_mode = AQ_TC_MODE_8TCS;
+	else
+		cfg->tc_mode = AQ_TC_MODE_4TCS;
+
+	/*rss rings */
+	cfg->num_rss_queues = min(cfg->vecs, AQ_CFG_NUM_RSS_QUEUES_DEF);
+	aq_nic_rss_init(self, cfg->num_rss_queues);
+}
+
+/* Checks hw_caps and 'corrects' aq_nic_cfg in runtime */
+void aq_nic_cfg_start(struct aq_nic_s *self)
+{
+	struct aq_nic_cfg_s *cfg = &self->aq_nic_cfg;
+	int i;
+
+	cfg->tcs = AQ_CFG_TCS_DEF;
 
 	cfg->is_polling = AQ_CFG_IS_POLLING_DEF;
 
-	cfg->is_interrupt_moderation = AQ_CFG_IS_INTERRUPT_MODERATION_DEF;
-	cfg->itr = cfg->is_interrupt_moderation ?
-		AQ_CFG_INTERRUPT_MODERATION_RATE_DEF : 0U;
+	cfg->itr = aq_itr;
+	cfg->tx_itr = aq_itr_tx;
+	cfg->rx_itr = aq_itr_rx;
 
+	cfg->rxpageorder = AQ_CFG_RX_PAGEORDER;
 	cfg->is_rss = AQ_CFG_IS_RSS_DEF;
-	cfg->num_rss_queues = AQ_CFG_NUM_RSS_QUEUES_DEF;
 	cfg->aq_rss.base_cpu_number = AQ_CFG_RSS_BASE_CPU_NUM_DEF;
-	cfg->flow_control = AQ_CFG_FC_MODE;
+	cfg->fc.req = AQ_CFG_FC_MODE;
+	cfg->wol = AQ_CFG_WOL_MODES;
 
 	cfg->mtu = AQ_CFG_MTU_DEF;
 	cfg->link_speed_msk = AQ_CFG_SPEED_MSK;
 	cfg->is_autoneg = AQ_CFG_IS_AUTONEG_DEF;
 
 	cfg->is_lro = AQ_CFG_IS_LRO_DEF;
-
-	cfg->vlan_id = 0U;
-
-	aq_nic_rss_init(self, cfg->num_rss_queues);
-}
-
-/* Checks hw_caps and 'corrects' aq_nic_cfg in runtime */
-int aq_nic_cfg_start(struct aq_nic_s *self)
-{
-	struct aq_nic_cfg_s *cfg = &self->aq_nic_cfg;
+	cfg->is_ptp = true;
 
 	/*descriptors */
-	cfg->rxds = min(cfg->rxds, cfg->aq_hw_caps->rxds);
-	cfg->txds = min(cfg->txds, cfg->aq_hw_caps->txds);
+	cfg->rxds = min(cfg->aq_hw_caps->rxds_max, AQ_CFG_RXDS_DEF);
+	cfg->txds = min(cfg->aq_hw_caps->txds_max, AQ_CFG_TXDS_DEF);
 
-	/*rss rings */
-	cfg->vecs = min(cfg->vecs, cfg->aq_hw_caps->vecs);
-	cfg->vecs = min(cfg->vecs, num_online_cpus());
-	/* cfg->vecs should be power of 2 for RSS */
-	if (cfg->vecs >= 8U)
-		cfg->vecs = 8U;
-	else if (cfg->vecs >= 4U)
-		cfg->vecs = 4U;
-	else if (cfg->vecs >= 2U)
-		cfg->vecs = 2U;
-	else
-		cfg->vecs = 1U;
+	aq_nic_cfg_update_num_vecs(self);
 
-	cfg->irq_type = aq_pci_func_get_irq_type(self->aq_pci_func);
+	cfg->irq_type = aq_pci_func_get_irq_type(self);
 
-	if ((cfg->irq_type == AQ_HW_IRQ_LEGACY) ||
-	    (self->aq_hw_caps.vecs == 1U) ||
+	if ((cfg->irq_type == AQ_HW_IRQ_INTX) ||
+	    (cfg->aq_hw_caps->vecs == 1U) ||
 	    (cfg->vecs == 1U)) {
 		cfg->is_rss = 0U;
 		cfg->vecs = 1U;
 	}
 
-	cfg->link_speed_msk &= self->aq_hw_caps.link_speed_msk;
-	cfg->hw_features = self->aq_hw_caps.hw_features;
-	return 0;
-}
-
-static void aq_nic_service_timer_cb(unsigned long param)
-{
-	struct aq_nic_s *self = (struct aq_nic_s *)param;
-	struct net_device *ndev = aq_nic_get_ndev(self);
-	int err = 0;
-	unsigned int i = 0U;
-	struct aq_hw_link_status_s link_status;
-	struct aq_ring_stats_rx_s stats_rx;
-	struct aq_ring_stats_tx_s stats_tx;
+	/* Check if we have enough vectors allocated for
+	 * link status IRQ. If no - we'll know link state from
+	 * slower service task.
+	 */
+	if (AQ_HW_SERVICE_IRQS > 0 && cfg->vecs + 1 <= self->irqvecs)
+		cfg->link_irq_vec = cfg->vecs;
+	else
+		cfg->link_irq_vec = 0;
 
-	if (aq_utils_obj_test(&self->header.flags, AQ_NIC_FLAGS_IS_NOT_READY))
-		goto err_exit;
+	cfg->link_speed_msk &= cfg->aq_hw_caps->link_speed_msk;
+	cfg->features = cfg->aq_hw_caps->hw_features;
+	cfg->is_vlan_rx_strip = !!(cfg->features & NETIF_F_HW_VLAN_CTAG_RX);
+	cfg->is_vlan_tx_insert = !!(cfg->features & NETIF_F_HW_VLAN_CTAG_TX);
+	cfg->is_vlan_force_promisc = true;
 
-	err = self->aq_hw_ops.hw_get_link_status(self->aq_hw, &link_status);
-	if (err < 0)
-		goto err_exit;
-
-	self->aq_hw_ops.hw_interrupt_moderation_set(self->aq_hw,
-			    self->aq_nic_cfg.is_interrupt_moderation);
+	for (i = 0; i < sizeof(cfg->prio_tc_map); i++)
+		cfg->prio_tc_map[i] = cfg->tcs * i / 8;
+}
 
-	if (memcmp(&link_status, &self->link_status, sizeof(link_status))) {
-		if (link_status.mbps) {
-			aq_utils_obj_set(&self->header.flags,
-					 AQ_NIC_FLAG_STARTED);
-			aq_utils_obj_clear(&self->header.flags,
-					   AQ_NIC_LINK_DOWN);
-			netif_carrier_on(self->ndev);
-		} else {
-			netif_carrier_off(self->ndev);
-			aq_utils_obj_set(&self->header.flags, AQ_NIC_LINK_DOWN);
+static int aq_nic_update_link_status(struct aq_nic_s *self)
+{
+	int err = self->aq_fw_ops->update_link_status(self->aq_hw);
+	u32 fc = 0;
+
+	if (err)
+		return err;
+
+	if (self->aq_fw_ops->get_flow_control)
+		self->aq_fw_ops->get_flow_control(self->aq_hw, &fc);
+	self->aq_nic_cfg.fc.cur = fc;
+
+	if (self->link_status.mbps != self->aq_hw->aq_link_status.mbps) {
+		netdev_info(self->ndev, "%s: link change old %d new %d\n",
+			    AQ_CFG_DRV_NAME, self->link_status.mbps,
+			    self->aq_hw->aq_link_status.mbps);
+		aq_nic_update_interrupt_moderation_settings(self);
+
+		if (self->aq_ptp) {
+			aq_ptp_clock_init(self);
+			aq_ptp_tm_offset_set(self,
+					     self->aq_hw->aq_link_status.mbps);
+			aq_ptp_link_change(self);
 		}
 
-		self->link_status = link_status;
+		/* Driver has to update flow control settings on RX block
+		 * on any link event.
+		 * We should query FW whether it negotiated FC.
+		 */
+		if (self->aq_hw_ops->hw_set_fc)
+			self->aq_hw_ops->hw_set_fc(self->aq_hw, fc, 0);
 	}
 
-	memset(&stats_rx, 0U, sizeof(struct aq_ring_stats_rx_s));
-	memset(&stats_tx, 0U, sizeof(struct aq_ring_stats_tx_s));
-	for (i = AQ_DIMOF(self->aq_vec); i--;) {
-		if (self->aq_vec[i])
-			aq_vec_add_stats(self->aq_vec[i], &stats_rx, &stats_tx);
-	}
+	self->link_status = self->aq_hw->aq_link_status;
+	if (!netif_carrier_ok(self->ndev) && self->link_status.mbps) {
+		aq_utils_obj_set(&self->flags,
+				 AQ_NIC_FLAG_STARTED);
+		aq_utils_obj_clear(&self->flags,
+				   AQ_NIC_LINK_DOWN);
+		netif_carrier_on(self->ndev);
+#if IS_ENABLED(CONFIG_MACSEC)
+		aq_macsec_enable(self);
+#endif
+		if (self->aq_hw_ops->hw_tc_rate_limit_set)
+			self->aq_hw_ops->hw_tc_rate_limit_set(self->aq_hw);
 
-	ndev->stats.rx_packets = stats_rx.packets;
-	ndev->stats.rx_bytes = stats_rx.bytes;
-	ndev->stats.rx_errors = stats_rx.errors;
-	ndev->stats.tx_packets = stats_tx.packets;
-	ndev->stats.tx_bytes = stats_tx.bytes;
-	ndev->stats.tx_errors = stats_tx.errors;
+		netif_tx_wake_all_queues(self->ndev);
+	}
+	if (netif_carrier_ok(self->ndev) && !self->link_status.mbps) {
+		netif_carrier_off(self->ndev);
+		netif_tx_disable(self->ndev);
+		aq_utils_obj_set(&self->flags, AQ_NIC_LINK_DOWN);
+	}
 
-err_exit:
-	mod_timer(&self->service_timer,
-		  jiffies + AQ_CFG_SERVICE_TIMER_INTERVAL);
+	return 0;
 }
 
-static void aq_nic_polling_timer_cb(unsigned long param)
+static irqreturn_t aq_linkstate_threaded_isr(int irq, void *private)
 {
-	struct aq_nic_s *self = (struct aq_nic_s *)param;
-	struct aq_vec_s *aq_vec = NULL;
-	unsigned int i = 0U;
+	struct aq_nic_s *self = private;
 
-	for (i = 0U, aq_vec = self->aq_vec[0];
-		self->aq_vecs > i; ++i, aq_vec = self->aq_vec[i])
-		aq_vec_isr(i, (void *)aq_vec);
+	if (!self)
+		return IRQ_NONE;
 
-	mod_timer(&self->polling_timer, jiffies +
-		AQ_CFG_POLLING_TIMER_INTERVAL);
+	aq_nic_update_link_status(self);
+
+	self->aq_hw_ops->hw_irq_enable(self->aq_hw,
+				       BIT(self->aq_nic_cfg.link_irq_vec));
+
+	return IRQ_HANDLED;
 }
 
-static struct net_device *aq_nic_ndev_alloc(void)
+static void aq_nic_service_task(struct work_struct *work)
 {
-	return alloc_etherdev_mq(sizeof(struct aq_nic_s), AQ_CFG_VECS_MAX);
+	struct aq_nic_s *self = container_of(work, struct aq_nic_s,
+					     service_task);
+	int err;
+
+	aq_ptp_service_task(self);
+
+	if (aq_utils_obj_test(&self->flags, AQ_NIC_FLAGS_IS_NOT_READY))
+		return;
+
+	err = aq_nic_update_link_status(self);
+	if (err)
+		return;
+
+#if IS_ENABLED(CONFIG_MACSEC)
+	aq_macsec_work(self);
+#endif
+
+	mutex_lock(&self->fwreq_mutex);
+	if (self->aq_fw_ops->update_stats)
+		self->aq_fw_ops->update_stats(self->aq_hw);
+	mutex_unlock(&self->fwreq_mutex);
+
+	aq_nic_update_ndev_stats(self);
 }
 
-struct aq_nic_s *aq_nic_alloc_cold(const struct net_device_ops *ndev_ops,
-				   const struct ethtool_ops *et_ops,
-				   struct device *dev,
-				   struct aq_pci_func_s *aq_pci_func,
-				   unsigned int port,
-				   const struct aq_hw_ops *aq_hw_ops)
+static void aq_nic_service_timer_cb(struct timer_list *t)
 {
-	struct net_device *ndev = NULL;
-	struct aq_nic_s *self = NULL;
-	int err = 0;
+	struct aq_nic_s *self = timer_container_of(self, t, service_timer);
 
-	ndev = aq_nic_ndev_alloc();
-	if (!ndev) {
-		err = -ENOMEM;
-		goto err_exit;
-	}
+	mod_timer(&self->service_timer,
+		  jiffies + AQ_CFG_SERVICE_TIMER_INTERVAL);
 
-	self = netdev_priv(ndev);
+	aq_ndev_schedule_work(&self->service_task);
+}
 
-	ndev->netdev_ops = ndev_ops;
-	ndev->ethtool_ops = et_ops;
+static void aq_nic_polling_timer_cb(struct timer_list *t)
+{
+	struct aq_nic_s *self = timer_container_of(self, t, polling_timer);
+	unsigned int i = 0U;
 
-	SET_NETDEV_DEV(ndev, dev);
+	for (i = 0U; self->aq_vecs > i; ++i)
+		aq_vec_isr(i, (void *)self->aq_vec[i]);
 
-	ndev->if_port = port;
-	ndev->min_mtu = ETH_MIN_MTU;
-	self->ndev = ndev;
+	mod_timer(&self->polling_timer, jiffies +
+		  AQ_CFG_POLLING_TIMER_INTERVAL);
+}
 
-	self->aq_pci_func = aq_pci_func;
+static int aq_nic_hw_prepare(struct aq_nic_s *self)
+{
+	int err = 0;
 
-	self->aq_hw_ops = *aq_hw_ops;
-	self->port = (u8)port;
+	err = self->aq_hw_ops->hw_soft_reset(self->aq_hw);
+	if (err)
+		goto exit;
 
-	self->aq_hw = self->aq_hw_ops.create(aq_pci_func, self->port,
-						&self->aq_hw_ops);
-	err = self->aq_hw_ops.get_hw_caps(self->aq_hw, &self->aq_hw_caps);
-	if (err < 0)
-		goto err_exit;
+	err = self->aq_hw_ops->hw_prepare(self->aq_hw, &self->aq_fw_ops);
 
-	aq_nic_cfg_init_defaults(self);
+exit:
+	return err;
+}
 
-err_exit:
-	if (err < 0) {
-		aq_nic_free_hot_resources(self);
-		self = NULL;
-	}
-	return self;
+static bool aq_nic_is_valid_ether_addr(const u8 *addr)
+{
+	/* Some engineering samples of Aquantia NICs are provisioned with a
+	 * partially populated MAC, which is still invalid.
+	 */
+	return !(addr[0] == 0 && addr[1] == 0 && addr[2] == 0);
 }
 
 int aq_nic_ndev_register(struct aq_nic_s *self)
 {
+	u8 addr[ETH_ALEN];
 	int err = 0;
-	unsigned int i = 0U;
 
 	if (!self->ndev) {
 		err = -EINVAL;
 		goto err_exit;
 	}
-	err = self->aq_hw_ops.hw_get_mac_permanent(self->aq_hw,
-			    self->aq_nic_cfg.aq_hw_caps,
-			    self->ndev->dev_addr);
-	if (err < 0)
+
+	err = aq_nic_hw_prepare(self);
+	if (err)
 		goto err_exit;
 
+#if IS_ENABLED(CONFIG_MACSEC)
+	aq_macsec_init(self);
+#endif
+
+	if (platform_get_ethdev_address(&self->pdev->dev, self->ndev) != 0) {
+		// If DT has none or an invalid one, ask device for MAC address
+		mutex_lock(&self->fwreq_mutex);
+		err = self->aq_fw_ops->get_mac_permanent(self->aq_hw, addr);
+		mutex_unlock(&self->fwreq_mutex);
+
+		if (err)
+			goto err_exit;
+
+		if (is_valid_ether_addr(addr) &&
+		    aq_nic_is_valid_ether_addr(addr)) {
+			eth_hw_addr_set(self->ndev, addr);
+		} else {
+			netdev_warn(self->ndev, "MAC is invalid, will use random.");
+			eth_hw_addr_random(self->ndev);
+		}
+	}
+
 #if defined(AQ_CFG_MAC_ADDR_PERMANENT)
 	{
 		static u8 mac_addr_permanent[] = AQ_CFG_MAC_ADDR_PERMANENT;
 
-		ether_addr_copy(self->ndev->dev_addr, mac_addr_permanent);
+		eth_hw_addr_set(self->ndev, mac_addr_permanent);
 	}
 #endif
 
+	for (self->aq_vecs = 0; self->aq_vecs < aq_nic_get_cfg(self)->vecs;
+	     self->aq_vecs++) {
+		self->aq_vec[self->aq_vecs] =
+		    aq_vec_alloc(self, self->aq_vecs, aq_nic_get_cfg(self));
+		if (!self->aq_vec[self->aq_vecs]) {
+			err = -ENOMEM;
+			goto err_exit;
+		}
+	}
+
 	netif_carrier_off(self->ndev);
 
-	for (i = AQ_CFG_VECS_MAX; i--;)
-		aq_nic_ndev_queue_stop(self, i);
+	netif_tx_disable(self->ndev);
 
 	err = register_netdev(self->ndev);
-	if (err < 0)
+	if (err)
 		goto err_exit;
 
 err_exit:
+#if IS_ENABLED(CONFIG_MACSEC)
+	if (err)
+		aq_macsec_free(self);
+#endif
 	return err;
 }
 
-int aq_nic_ndev_init(struct aq_nic_s *self)
+void aq_nic_ndev_init(struct aq_nic_s *self)
 {
-	struct aq_hw_caps_s *aq_hw_caps = self->aq_nic_cfg.aq_hw_caps;
+	const struct aq_hw_caps_s *aq_hw_caps = self->aq_nic_cfg.aq_hw_caps;
 	struct aq_nic_cfg_s *aq_nic_cfg = &self->aq_nic_cfg;
 
 	self->ndev->hw_features |= aq_hw_caps->hw_features;
 	self->ndev->features = aq_hw_caps->hw_features;
+	self->ndev->vlan_features |= NETIF_F_HW_CSUM | NETIF_F_RXCSUM |
+				     NETIF_F_RXHASH | NETIF_F_SG |
+				     NETIF_F_LRO | NETIF_F_TSO | NETIF_F_TSO6;
+	self->ndev->gso_partial_features = NETIF_F_GSO_UDP_L4;
 	self->ndev->priv_flags = aq_hw_caps->hw_priv_flags;
-	self->ndev->mtu = aq_nic_cfg->mtu - ETH_HLEN;
-
-	return 0;
-}
-
-void aq_nic_ndev_free(struct aq_nic_s *self)
-{
-	if (!self->ndev)
-		goto err_exit;
-
-	if (self->ndev->reg_state == NETREG_REGISTERED)
-		unregister_netdev(self->ndev);
+	self->ndev->priv_flags |= IFF_LIVE_ADDR_CHANGE;
 
-	if (self->aq_hw)
-		self->aq_hw_ops.destroy(self->aq_hw);
-
-	free_netdev(self->ndev);
-
-err_exit:;
-}
-
-struct aq_nic_s *aq_nic_alloc_hot(struct net_device *ndev)
-{
-	struct aq_nic_s *self = NULL;
-	int err = 0;
-
-	if (!ndev) {
-		err = -EINVAL;
-		goto err_exit;
-	}
-	self = netdev_priv(ndev);
-
-	if (!self) {
-		err = -EINVAL;
-		goto err_exit;
-	}
-	if (netif_running(ndev)) {
-		unsigned int i;
-
-		for (i = AQ_CFG_VECS_MAX; i--;)
-			netif_stop_subqueue(ndev, i);
-	}
-
-	for (self->aq_vecs = 0; self->aq_vecs < self->aq_nic_cfg.vecs;
-		self->aq_vecs++) {
-		self->aq_vec[self->aq_vecs] =
-		    aq_vec_alloc(self, self->aq_vecs, &self->aq_nic_cfg);
-		if (!self->aq_vec[self->aq_vecs]) {
-			err = -ENOMEM;
-			goto err_exit;
-		}
-	}
+	self->msg_enable = NETIF_MSG_DRV | NETIF_MSG_LINK;
+	self->ndev->mtu = aq_nic_cfg->mtu - ETH_HLEN;
+	self->ndev->max_mtu = aq_hw_caps->mtu - ETH_FCS_LEN - ETH_HLEN;
 
-err_exit:
-	if (err < 0) {
-		aq_nic_free_hot_resources(self);
-		self = NULL;
-	}
-	return self;
+	self->ndev->xdp_features = NETDEV_XDP_ACT_BASIC |
+				   NETDEV_XDP_ACT_REDIRECT |
+				   NETDEV_XDP_ACT_NDO_XMIT |
+				   NETDEV_XDP_ACT_RX_SG |
+				   NETDEV_XDP_ACT_NDO_XMIT_SG;
 }
 
 void aq_nic_set_tx_ring(struct aq_nic_s *self, unsigned int idx,
@@ -350,11 +397,6 @@ void aq_nic_set_tx_ring(struct aq_nic_s *self, unsigned int idx,
 	self->aq_ring_tx[idx] = ring;
 }
 
-struct device *aq_nic_get_dev(struct aq_nic_s *self)
-{
-	return self->ndev->dev.parent;
-}
-
 struct net_device *aq_nic_get_ndev(struct aq_nic_s *self)
 {
 	return self->ndev;
@@ -363,212 +405,399 @@ struct net_device *aq_nic_get_ndev(struct aq_nic_s *self)
 int aq_nic_init(struct aq_nic_s *self)
 {
 	struct aq_vec_s *aq_vec = NULL;
-	int err = 0;
 	unsigned int i = 0U;
+	int err = 0;
 
 	self->power_state = AQ_HW_POWER_STATE_D0;
-	err = self->aq_hw_ops.hw_reset(self->aq_hw);
+	mutex_lock(&self->fwreq_mutex);
+	err = self->aq_hw_ops->hw_reset(self->aq_hw);
+	mutex_unlock(&self->fwreq_mutex);
 	if (err < 0)
 		goto err_exit;
+	/* Restore default settings */
+	aq_nic_set_downshift(self, self->aq_nic_cfg.downshift_counter);
+	aq_nic_set_media_detect(self, self->aq_nic_cfg.is_media_detect ?
+				AQ_HW_MEDIA_DETECT_CNT : 0);
 
-	err = self->aq_hw_ops.hw_init(self->aq_hw, &self->aq_nic_cfg,
-			    aq_nic_get_ndev(self)->dev_addr);
+	err = self->aq_hw_ops->hw_init(self->aq_hw,
+				       aq_nic_get_ndev(self)->dev_addr);
 	if (err < 0)
 		goto err_exit;
 
-	for (i = 0U, aq_vec = self->aq_vec[0];
-		self->aq_vecs > i; ++i, aq_vec = self->aq_vec[i])
-		aq_vec_init(aq_vec, &self->aq_hw_ops, self->aq_hw);
+	if (ATL_HW_IS_CHIP_FEATURE(self->aq_hw, ATLANTIC) &&
+	    self->aq_nic_cfg.aq_hw_caps->media_type == AQ_HW_MEDIA_TYPE_TP) {
+		self->aq_hw->phy_id = HW_ATL_PHY_ID_MAX;
+		err = aq_phy_init(self->aq_hw);
+
+		/* Disable the PTP on NICs where it's known to cause datapath
+		 * problems.
+		 * Ideally this should have been done by PHY provisioning, but
+		 * many units have been shipped with enabled PTP block already.
+		 */
+		if (self->aq_nic_cfg.aq_hw_caps->quirks & AQ_NIC_QUIRK_BAD_PTP)
+			if (self->aq_hw->phy_id != HW_ATL_PHY_ID_MAX)
+				aq_phy_disable_ptp(self->aq_hw);
+	}
 
-err_exit:
-	return err;
-}
+	for (i = 0U; i < self->aq_vecs; i++) {
+		aq_vec = self->aq_vec[i];
+		err = aq_vec_ring_alloc(aq_vec, self, i,
+					aq_nic_get_cfg(self));
+		if (err)
+			goto err_exit;
 
-void aq_nic_ndev_queue_start(struct aq_nic_s *self, unsigned int idx)
-{
-	netif_start_subqueue(self->ndev, idx);
-}
+		aq_vec_init(aq_vec, self->aq_hw_ops, self->aq_hw);
+	}
 
-void aq_nic_ndev_queue_stop(struct aq_nic_s *self, unsigned int idx)
-{
-	netif_stop_subqueue(self->ndev, idx);
+	if (aq_nic_get_cfg(self)->is_ptp) {
+		err = aq_ptp_init(self, self->irqvecs - 1);
+		if (err < 0)
+			goto err_exit;
+
+		err = aq_ptp_ring_alloc(self);
+		if (err < 0)
+			goto err_exit;
+
+		err = aq_ptp_ring_init(self);
+		if (err < 0)
+			goto err_exit;
+	}
+
+	netif_carrier_off(self->ndev);
+
+err_exit:
+	return err;
 }
 
 int aq_nic_start(struct aq_nic_s *self)
 {
 	struct aq_vec_s *aq_vec = NULL;
-	int err = 0;
+	struct aq_nic_cfg_s *cfg;
 	unsigned int i = 0U;
+	int err = 0;
 
-	err = self->aq_hw_ops.hw_multicast_list_set(self->aq_hw,
-						    self->mc_list.ar,
-						    self->mc_list.count);
+	cfg = aq_nic_get_cfg(self);
+
+	err = self->aq_hw_ops->hw_multicast_list_set(self->aq_hw,
+						     self->mc_list.ar,
+						     self->mc_list.count);
 	if (err < 0)
 		goto err_exit;
 
-	err = self->aq_hw_ops.hw_packet_filter_set(self->aq_hw,
-						   self->packet_filter);
+	err = self->aq_hw_ops->hw_packet_filter_set(self->aq_hw,
+						    self->packet_filter);
 	if (err < 0)
 		goto err_exit;
 
-	for (i = 0U, aq_vec = self->aq_vec[0];
-		self->aq_vecs > i; ++i, aq_vec = self->aq_vec[i]) {
+	for (i = 0U; self->aq_vecs > i; ++i) {
+		aq_vec = self->aq_vec[i];
 		err = aq_vec_start(aq_vec);
 		if (err < 0)
 			goto err_exit;
 	}
 
-	err = self->aq_hw_ops.hw_start(self->aq_hw);
+	err = aq_ptp_ring_start(self);
 	if (err < 0)
 		goto err_exit;
 
-	err = self->aq_hw_ops.hw_interrupt_moderation_set(self->aq_hw,
-			    self->aq_nic_cfg.is_interrupt_moderation);
+	aq_nic_set_loopback(self);
+
+	err = self->aq_hw_ops->hw_start(self->aq_hw);
 	if (err < 0)
 		goto err_exit;
-	setup_timer(&self->service_timer, &aq_nic_service_timer_cb,
-		    (unsigned long)self);
-	mod_timer(&self->service_timer, jiffies +
-			AQ_CFG_SERVICE_TIMER_INTERVAL);
-
-	if (self->aq_nic_cfg.is_polling) {
-		setup_timer(&self->polling_timer, &aq_nic_polling_timer_cb,
-			    (unsigned long)self);
+
+	err = aq_nic_update_interrupt_moderation_settings(self);
+	if (err)
+		goto err_exit;
+
+	INIT_WORK(&self->service_task, aq_nic_service_task);
+
+	timer_setup(&self->service_timer, aq_nic_service_timer_cb, 0);
+	aq_nic_service_timer_cb(&self->service_timer);
+
+	if (cfg->is_polling) {
+		timer_setup(&self->polling_timer, aq_nic_polling_timer_cb, 0);
 		mod_timer(&self->polling_timer, jiffies +
 			  AQ_CFG_POLLING_TIMER_INTERVAL);
 	} else {
-		for (i = 0U, aq_vec = self->aq_vec[0];
-			self->aq_vecs > i; ++i, aq_vec = self->aq_vec[i]) {
-			err = aq_pci_func_alloc_irq(self->aq_pci_func, i,
-						    self->ndev->name, aq_vec,
-					aq_vec_get_affinity_mask(aq_vec));
+		for (i = 0U; self->aq_vecs > i; ++i) {
+			aq_vec = self->aq_vec[i];
+			err = aq_pci_func_alloc_irq(self, i, self->ndev->name,
+						    aq_vec_isr, aq_vec,
+						    aq_vec_get_affinity_mask(aq_vec));
 			if (err < 0)
 				goto err_exit;
 		}
 
-		err = self->aq_hw_ops.hw_irq_enable(self->aq_hw,
-				    AQ_CFG_IRQ_MASK);
+		err = aq_ptp_irq_alloc(self);
 		if (err < 0)
 			goto err_exit;
-	}
 
-	for (i = 0U, aq_vec = self->aq_vec[0];
-		self->aq_vecs > i; ++i, aq_vec = self->aq_vec[i])
-		aq_nic_ndev_queue_start(self, i);
+		if (cfg->link_irq_vec) {
+			int irqvec = pci_irq_vector(self->pdev,
+						    cfg->link_irq_vec);
+			err = request_threaded_irq(irqvec, NULL,
+						   aq_linkstate_threaded_isr,
+						   IRQF_SHARED | IRQF_ONESHOT,
+						   self->ndev->name, self);
+			if (err < 0)
+				goto err_exit;
+			self->msix_entry_mask |= (1 << cfg->link_irq_vec);
+		}
+
+		err = self->aq_hw_ops->hw_irq_enable(self->aq_hw,
+						     AQ_CFG_IRQ_MASK);
+		if (err < 0)
+			goto err_exit;
+	}
 
-	err = netif_set_real_num_tx_queues(self->ndev, self->aq_vecs);
+	err = netif_set_real_num_tx_queues(self->ndev,
+					   self->aq_vecs * cfg->tcs);
 	if (err < 0)
 		goto err_exit;
 
-	err = netif_set_real_num_rx_queues(self->ndev, self->aq_vecs);
+	err = netif_set_real_num_rx_queues(self->ndev,
+					   self->aq_vecs * cfg->tcs);
 	if (err < 0)
 		goto err_exit;
 
+	for (i = 0; i < cfg->tcs; i++) {
+		u16 offset = self->aq_vecs * i;
+
+		netdev_set_tc_queue(self->ndev, i, self->aq_vecs, offset);
+	}
+	netif_tx_start_all_queues(self->ndev);
+
 err_exit:
 	return err;
 }
 
-static unsigned int aq_nic_map_skb(struct aq_nic_s *self,
-				   struct sk_buff *skb,
+static unsigned int aq_nic_map_xdp(struct aq_nic_s *self,
+				   struct xdp_frame *xdpf,
 				   struct aq_ring_s *ring)
 {
+	struct device *dev = aq_nic_get_dev(self);
+	struct aq_ring_buff_s *first = NULL;
+	unsigned int dx = ring->sw_tail;
+	struct aq_ring_buff_s *dx_buff;
+	struct skb_shared_info *sinfo;
+	unsigned int frag_count = 0U;
+	unsigned int nr_frags = 0U;
 	unsigned int ret = 0U;
+	u16 total_len;
+
+	dx_buff = &ring->buff_ring[dx];
+	dx_buff->flags = 0U;
+
+	sinfo = xdp_get_shared_info_from_frame(xdpf);
+	total_len = xdpf->len;
+	dx_buff->len = total_len;
+	if (xdp_frame_has_frags(xdpf)) {
+		nr_frags = sinfo->nr_frags;
+		total_len += sinfo->xdp_frags_size;
+	}
+	dx_buff->pa = dma_map_single(dev, xdpf->data, dx_buff->len,
+				     DMA_TO_DEVICE);
+
+	if (unlikely(dma_mapping_error(dev, dx_buff->pa)))
+		goto exit;
+
+	first = dx_buff;
+	dx_buff->len_pkt = total_len;
+	dx_buff->is_sop = 1U;
+	dx_buff->is_mapped = 1U;
+	++ret;
+
+	for (; nr_frags--; ++frag_count) {
+		skb_frag_t *frag = &sinfo->frags[frag_count];
+		unsigned int frag_len = skb_frag_size(frag);
+		unsigned int buff_offset = 0U;
+		unsigned int buff_size = 0U;
+		dma_addr_t frag_pa;
+
+		while (frag_len) {
+			if (frag_len > AQ_CFG_TX_FRAME_MAX)
+				buff_size = AQ_CFG_TX_FRAME_MAX;
+			else
+				buff_size = frag_len;
+
+			frag_pa = skb_frag_dma_map(dev, frag, buff_offset,
+						   buff_size, DMA_TO_DEVICE);
+
+			if (unlikely(dma_mapping_error(dev, frag_pa)))
+				goto mapping_error;
+
+			dx = aq_ring_next_dx(ring, dx);
+			dx_buff = &ring->buff_ring[dx];
+
+			dx_buff->flags = 0U;
+			dx_buff->len = buff_size;
+			dx_buff->pa = frag_pa;
+			dx_buff->is_mapped = 1U;
+			dx_buff->eop_index = 0xffffU;
+
+			frag_len -= buff_size;
+			buff_offset += buff_size;
+
+			++ret;
+		}
+	}
+
+	first->eop_index = dx;
+	dx_buff->is_eop = 1U;
+	dx_buff->skb = NULL;
+	dx_buff->xdpf = xdpf;
+	goto exit;
+
+mapping_error:
+	for (dx = ring->sw_tail;
+	     ret > 0;
+	     --ret, dx = aq_ring_next_dx(ring, dx)) {
+		dx_buff = &ring->buff_ring[dx];
+
+		if (!dx_buff->pa)
+			continue;
+		if (unlikely(dx_buff->is_sop))
+			dma_unmap_single(dev, dx_buff->pa, dx_buff->len,
+					 DMA_TO_DEVICE);
+		else
+			dma_unmap_page(dev, dx_buff->pa, dx_buff->len,
+				       DMA_TO_DEVICE);
+	}
+
+exit:
+	return ret;
+}
+
+unsigned int aq_nic_map_skb(struct aq_nic_s *self, struct sk_buff *skb,
+			    struct aq_ring_s *ring)
+{
 	unsigned int nr_frags = skb_shinfo(skb)->nr_frags;
+	struct aq_nic_cfg_s *cfg = aq_nic_get_cfg(self);
+	struct device *dev = aq_nic_get_dev(self);
+	struct aq_ring_buff_s *first = NULL;
+	u8 ipver = ip_hdr(skb)->version;
+	struct aq_ring_buff_s *dx_buff;
+	bool need_context_tag = false;
 	unsigned int frag_count = 0U;
-	unsigned int dx = ring->sw_tail;
-	struct aq_ring_buff_s *dx_buff = &ring->buff_ring[dx];
+	unsigned int ret = 0U;
+	unsigned int dx;
+	u8 l4proto = 0;
+
+	if (ipver == 4)
+		l4proto = ip_hdr(skb)->protocol;
+	else if (ipver == 6)
+		l4proto = ipv6_hdr(skb)->nexthdr;
+
+	dx = ring->sw_tail;
+	dx_buff = &ring->buff_ring[dx];
+	dx_buff->flags = 0U;
 
 	if (unlikely(skb_is_gso(skb))) {
-		dx_buff->flags = 0U;
+		dx_buff->mss = skb_shinfo(skb)->gso_size;
+		if (l4proto == IPPROTO_TCP) {
+			dx_buff->is_gso_tcp = 1U;
+			dx_buff->len_l4 = tcp_hdrlen(skb);
+		} else if (l4proto == IPPROTO_UDP) {
+			dx_buff->is_gso_udp = 1U;
+			dx_buff->len_l4 = sizeof(struct udphdr);
+			/* UDP GSO Hardware does not replace packet length. */
+			udp_hdr(skb)->len = htons(dx_buff->mss +
+						  dx_buff->len_l4);
+		} else {
+			WARN_ONCE(true, "Bad GSO mode");
+			goto exit;
+		}
 		dx_buff->len_pkt = skb->len;
 		dx_buff->len_l2 = ETH_HLEN;
-		dx_buff->len_l3 = ip_hdrlen(skb);
-		dx_buff->len_l4 = tcp_hdrlen(skb);
-		dx_buff->mss = skb_shinfo(skb)->gso_size;
-		dx_buff->is_txc = 1U;
+		dx_buff->len_l3 = skb_network_header_len(skb);
+		dx_buff->eop_index = 0xffffU;
+		dx_buff->is_ipv6 = (ipver == 6);
+		need_context_tag = true;
+	}
 
-		dx_buff->is_ipv6 =
-			(ip_hdr(skb)->version == 6) ? 1U : 0U;
+	if (cfg->is_vlan_tx_insert && skb_vlan_tag_present(skb)) {
+		dx_buff->vlan_tx_tag = skb_vlan_tag_get(skb);
+		dx_buff->len_pkt = skb->len;
+		dx_buff->is_vlan = 1U;
+		need_context_tag = true;
+	}
 
+	if (need_context_tag) {
 		dx = aq_ring_next_dx(ring, dx);
 		dx_buff = &ring->buff_ring[dx];
+		dx_buff->flags = 0U;
 		++ret;
 	}
 
-	dx_buff->flags = 0U;
 	dx_buff->len = skb_headlen(skb);
-	dx_buff->pa = dma_map_single(aq_nic_get_dev(self),
+	dx_buff->pa = dma_map_single(dev,
 				     skb->data,
 				     dx_buff->len,
 				     DMA_TO_DEVICE);
 
-	if (unlikely(dma_mapping_error(aq_nic_get_dev(self), dx_buff->pa)))
+	if (unlikely(dma_mapping_error(dev, dx_buff->pa))) {
+		ret = 0;
 		goto exit;
+	}
 
+	first = dx_buff;
 	dx_buff->len_pkt = skb->len;
 	dx_buff->is_sop = 1U;
 	dx_buff->is_mapped = 1U;
 	++ret;
 
 	if (skb->ip_summed == CHECKSUM_PARTIAL) {
-		dx_buff->is_ip_cso = (htons(ETH_P_IP) == skb->protocol) ?
-			1U : 0U;
-
-		if (ip_hdr(skb)->version == 4) {
-			dx_buff->is_tcp_cso =
-				(ip_hdr(skb)->protocol == IPPROTO_TCP) ?
-					1U : 0U;
-			dx_buff->is_udp_cso =
-				(ip_hdr(skb)->protocol == IPPROTO_UDP) ?
-					1U : 0U;
-		} else if (ip_hdr(skb)->version == 6) {
-			dx_buff->is_tcp_cso =
-				(ipv6_hdr(skb)->nexthdr == NEXTHDR_TCP) ?
-					1U : 0U;
-			dx_buff->is_udp_cso =
-				(ipv6_hdr(skb)->nexthdr == NEXTHDR_UDP) ?
-					1U : 0U;
-		}
+		dx_buff->is_ip_cso = (htons(ETH_P_IP) == skb->protocol);
+		dx_buff->is_tcp_cso = (l4proto == IPPROTO_TCP);
+		dx_buff->is_udp_cso = (l4proto == IPPROTO_UDP);
 	}
 
 	for (; nr_frags--; ++frag_count) {
 		unsigned int frag_len = 0U;
+		unsigned int buff_offset = 0U;
+		unsigned int buff_size = 0U;
 		dma_addr_t frag_pa;
 		skb_frag_t *frag = &skb_shinfo(skb)->frags[frag_count];
 
 		frag_len = skb_frag_size(frag);
-		frag_pa = skb_frag_dma_map(aq_nic_get_dev(self), frag, 0,
-					   frag_len, DMA_TO_DEVICE);
 
-		if (unlikely(dma_mapping_error(aq_nic_get_dev(self), frag_pa)))
-			goto mapping_error;
+		while (frag_len) {
+			if (frag_len > AQ_CFG_TX_FRAME_MAX)
+				buff_size = AQ_CFG_TX_FRAME_MAX;
+			else
+				buff_size = frag_len;
+
+			frag_pa = skb_frag_dma_map(dev,
+						   frag,
+						   buff_offset,
+						   buff_size,
+						   DMA_TO_DEVICE);
+
+			if (unlikely(dma_mapping_error(dev,
+						       frag_pa)))
+				goto mapping_error;
 
-		while (frag_len > AQ_CFG_TX_FRAME_MAX) {
 			dx = aq_ring_next_dx(ring, dx);
 			dx_buff = &ring->buff_ring[dx];
 
 			dx_buff->flags = 0U;
-			dx_buff->len = AQ_CFG_TX_FRAME_MAX;
+			dx_buff->len = buff_size;
 			dx_buff->pa = frag_pa;
 			dx_buff->is_mapped = 1U;
+			dx_buff->eop_index = 0xffffU;
+
+			frag_len -= buff_size;
+			buff_offset += buff_size;
 
-			frag_len -= AQ_CFG_TX_FRAME_MAX;
-			frag_pa += AQ_CFG_TX_FRAME_MAX;
 			++ret;
 		}
-
-		dx = aq_ring_next_dx(ring, dx);
-		dx_buff = &ring->buff_ring[dx];
-
-		dx_buff->flags = 0U;
-		dx_buff->len = frag_len;
-		dx_buff->pa = frag_pa;
-		dx_buff->is_mapped = 1U;
-		++ret;
 	}
 
+	first->eop_index = dx;
 	dx_buff->is_eop = 1U;
 	dx_buff->skb = skb;
+	dx_buff->xdpf = NULL;
 	goto exit;
 
 mapping_error:
@@ -577,14 +806,15 @@ mapping_error:
 	     --ret, dx = aq_ring_next_dx(ring, dx)) {
 		dx_buff = &ring->buff_ring[dx];
 
-		if (!dx_buff->is_txc && dx_buff->pa) {
+		if (!(dx_buff->is_gso_tcp || dx_buff->is_gso_udp) &&
+		    !dx_buff->is_vlan && dx_buff->pa) {
 			if (unlikely(dx_buff->is_sop)) {
-				dma_unmap_single(aq_nic_get_dev(self),
+				dma_unmap_single(dev,
 						 dx_buff->pa,
 						 dx_buff->len,
 						 DMA_TO_DEVICE);
 			} else {
-				dma_unmap_page(aq_nic_get_dev(self),
+				dma_unmap_page(dev,
 					       dx_buff->pa,
 					       dx_buff->len,
 					       DMA_TO_DEVICE);
@@ -596,79 +826,101 @@ exit:
 	return ret;
 }
 
+int aq_nic_xmit_xdpf(struct aq_nic_s *aq_nic, struct aq_ring_s *tx_ring,
+		     struct xdp_frame *xdpf)
+{
+	u16 queue_index = AQ_NIC_RING2QMAP(aq_nic, tx_ring->idx);
+	struct net_device *ndev = aq_nic_get_ndev(aq_nic);
+	struct skb_shared_info *sinfo;
+	int cpu = smp_processor_id();
+	int err = NETDEV_TX_BUSY;
+	struct netdev_queue *nq;
+	unsigned int frags = 1;
+
+	if (xdp_frame_has_frags(xdpf)) {
+		sinfo = xdp_get_shared_info_from_frame(xdpf);
+		frags += sinfo->nr_frags;
+	}
+
+	if (frags > AQ_CFG_SKB_FRAGS_MAX)
+		return err;
+
+	nq = netdev_get_tx_queue(ndev, tx_ring->idx);
+	__netif_tx_lock(nq, cpu);
+
+	aq_ring_update_queue_state(tx_ring);
+
+	/* Above status update may stop the queue. Check this. */
+	if (__netif_subqueue_stopped(aq_nic_get_ndev(aq_nic), queue_index))
+		goto out;
+
+	frags = aq_nic_map_xdp(aq_nic, xdpf, tx_ring);
+	if (likely(frags))
+		err = aq_nic->aq_hw_ops->hw_ring_tx_xmit(aq_nic->aq_hw, tx_ring,
+							 frags);
+out:
+	__netif_tx_unlock(nq);
+
+	return err;
+}
+
 int aq_nic_xmit(struct aq_nic_s *self, struct sk_buff *skb)
-__releases(&ring->lock)
-__acquires(&ring->lock)
 {
+	struct aq_nic_cfg_s *cfg = aq_nic_get_cfg(self);
+	unsigned int vec = skb->queue_mapping % cfg->vecs;
+	unsigned int tc = skb->queue_mapping / cfg->vecs;
 	struct aq_ring_s *ring = NULL;
 	unsigned int frags = 0U;
-	unsigned int vec = skb->queue_mapping % self->aq_nic_cfg.vecs;
-	unsigned int tc = 0U;
-	unsigned int trys = AQ_CFG_LOCK_TRYS;
 	int err = NETDEV_TX_OK;
-	bool is_nic_in_bad_state;
 
 	frags = skb_shinfo(skb)->nr_frags + 1;
 
-	ring = self->aq_ring_tx[AQ_NIC_TCVEC2RING(self, tc, vec)];
+	ring = self->aq_ring_tx[AQ_NIC_CFG_TCVEC2RING(cfg, tc, vec)];
 
 	if (frags > AQ_CFG_SKB_FRAGS_MAX) {
 		dev_kfree_skb_any(skb);
 		goto err_exit;
 	}
 
-	is_nic_in_bad_state = aq_utils_obj_test(&self->header.flags,
-						AQ_NIC_FLAGS_IS_NOT_TX_READY) ||
-						(aq_ring_avail_dx(ring) <
-						AQ_CFG_SKB_FRAGS_MAX);
+	aq_ring_update_queue_state(ring);
 
-	if (is_nic_in_bad_state) {
-		aq_nic_ndev_queue_stop(self, ring->idx);
+	if (cfg->priv_flags & BIT(AQ_HW_LOOPBACK_DMA_NET)) {
 		err = NETDEV_TX_BUSY;
 		goto err_exit;
 	}
 
-	do {
-		if (spin_trylock(&ring->header.lock)) {
-			frags = aq_nic_map_skb(self, skb, ring);
-
-			if (likely(frags)) {
-				err = self->aq_hw_ops.hw_ring_tx_xmit(
-								self->aq_hw,
-								ring, frags);
-				if (err >= 0) {
-					if (aq_ring_avail_dx(ring) <
-					    AQ_CFG_SKB_FRAGS_MAX + 1)
-						aq_nic_ndev_queue_stop(
-								self,
-								ring->idx);
-
-					++ring->stats.tx.packets;
-					ring->stats.tx.bytes += skb->len;
-				}
-			} else {
-				err = NETDEV_TX_BUSY;
-			}
+	/* Above status update may stop the queue. Check this. */
+	if (__netif_subqueue_stopped(self->ndev,
+				     AQ_NIC_RING2QMAP(self, ring->idx))) {
+		err = NETDEV_TX_BUSY;
+		goto err_exit;
+	}
 
-			spin_unlock(&ring->header.lock);
-			break;
-		}
-	} while (--trys);
+	frags = aq_nic_map_skb(self, skb, ring);
 
-	if (!trys) {
+	skb_tx_timestamp(skb);
+
+	if (likely(frags)) {
+		err = self->aq_hw_ops->hw_ring_tx_xmit(self->aq_hw,
+						       ring, frags);
+	} else {
 		err = NETDEV_TX_BUSY;
-		goto err_exit;
 	}
 
 err_exit:
 	return err;
 }
 
+int aq_nic_update_interrupt_moderation_settings(struct aq_nic_s *self)
+{
+	return self->aq_hw_ops->hw_interrupt_moderation_set(self->aq_hw);
+}
+
 int aq_nic_set_packet_filter(struct aq_nic_s *self, unsigned int flags)
 {
 	int err = 0;
 
-	err = self->aq_hw_ops.hw_packet_filter_set(self->aq_hw, flags);
+	err = self->aq_hw_ops->hw_packet_filter_set(self->aq_hw, flags);
 	if (err < 0)
 		goto err_exit;
 
@@ -680,38 +932,56 @@ err_exit:
 
 int aq_nic_set_multicast_list(struct aq_nic_s *self, struct net_device *ndev)
 {
+	const struct aq_hw_ops *hw_ops = self->aq_hw_ops;
+	struct aq_nic_cfg_s *cfg = &self->aq_nic_cfg;
+	unsigned int packet_filter = ndev->flags;
 	struct netdev_hw_addr *ha = NULL;
 	unsigned int i = 0U;
+	int err = 0;
 
-	self->mc_list.count = 0U;
+	self->mc_list.count = 0;
+	if (netdev_uc_count(ndev) > AQ_HW_MULTICAST_ADDRESS_MAX) {
+		packet_filter |= IFF_PROMISC;
+	} else {
+		netdev_for_each_uc_addr(ha, ndev) {
+			ether_addr_copy(self->mc_list.ar[i++], ha->addr);
+		}
+	}
 
-	netdev_for_each_mc_addr(ha, ndev) {
-		ether_addr_copy(self->mc_list.ar[i++], ha->addr);
-		++self->mc_list.count;
+	cfg->is_mc_list_enabled = !!(packet_filter & IFF_MULTICAST);
+	if (cfg->is_mc_list_enabled) {
+		if (i + netdev_mc_count(ndev) > AQ_HW_MULTICAST_ADDRESS_MAX) {
+			packet_filter |= IFF_ALLMULTI;
+		} else {
+			netdev_for_each_mc_addr(ha, ndev) {
+				ether_addr_copy(self->mc_list.ar[i++],
+						ha->addr);
+			}
+		}
 	}
 
-	return self->aq_hw_ops.hw_multicast_list_set(self->aq_hw,
+	if (i > 0 && i <= AQ_HW_MULTICAST_ADDRESS_MAX) {
+		self->mc_list.count = i;
+		err = hw_ops->hw_multicast_list_set(self->aq_hw,
 						    self->mc_list.ar,
 						    self->mc_list.count);
+		if (err < 0)
+			return err;
+	}
+
+	return aq_nic_set_packet_filter(self, packet_filter);
 }
 
 int aq_nic_set_mtu(struct aq_nic_s *self, int new_mtu)
 {
-	int err = 0;
-
-	if (new_mtu > self->aq_hw_caps.mtu) {
-		err = -EINVAL;
-		goto err_exit;
-	}
 	self->aq_nic_cfg.mtu = new_mtu;
 
-err_exit:
-	return err;
+	return 0;
 }
 
 int aq_nic_set_mac(struct aq_nic_s *self, struct net_device *ndev)
 {
-	return self->aq_hw_ops.hw_set_mac_address(self->aq_hw, ndev->dev_addr);
+	return self->aq_hw_ops->hw_set_mac_address(self->aq_hw, ndev->dev_addr);
 }
 
 unsigned int aq_nic_get_link_speed(struct aq_nic_s *self)
@@ -724,10 +994,14 @@ int aq_nic_get_regs(struct aq_nic_s *self, struct ethtool_regs *regs, void *p)
 	u32 *regs_buff = p;
 	int err = 0;
 
+	if (unlikely(!self->aq_hw_ops->hw_get_regs))
+		return -EOPNOTSUPP;
+
 	regs->version = 1;
 
-	err = self->aq_hw_ops.hw_get_regs(self->aq_hw,
-					  &self->aq_hw_caps, regs_buff);
+	err = self->aq_hw_ops->hw_get_regs(self->aq_hw,
+					   self->aq_nic_cfg.aq_hw_caps,
+					   regs_buff);
 	if (err < 0)
 		goto err_exit;
 
@@ -737,126 +1011,297 @@ err_exit:
 
 int aq_nic_get_regs_count(struct aq_nic_s *self)
 {
-	return self->aq_hw_caps.mac_regs_count;
+	if (unlikely(!self->aq_hw_ops->hw_get_regs))
+		return 0;
+
+	return self->aq_nic_cfg.aq_hw_caps->mac_regs_count;
 }
 
-void aq_nic_get_stats(struct aq_nic_s *self, u64 *data)
+u64 *aq_nic_get_stats(struct aq_nic_s *self, u64 *data)
 {
-	struct aq_vec_s *aq_vec = NULL;
-	unsigned int i = 0U;
+	struct aq_stats_s *stats;
 	unsigned int count = 0U;
-	int err = 0;
+	unsigned int i = 0U;
+	unsigned int tc;
 
-	err = self->aq_hw_ops.hw_get_hw_stats(self->aq_hw, data, &count);
-	if (err < 0)
+	if (self->aq_fw_ops->update_stats) {
+		mutex_lock(&self->fwreq_mutex);
+		self->aq_fw_ops->update_stats(self->aq_hw);
+		mutex_unlock(&self->fwreq_mutex);
+	}
+	stats = self->aq_hw_ops->hw_get_hw_stats(self->aq_hw);
+
+	if (!stats)
 		goto err_exit;
 
+	data[i] = stats->uprc + stats->mprc + stats->bprc;
+	data[++i] = stats->uprc;
+	data[++i] = stats->mprc;
+	data[++i] = stats->bprc;
+	data[++i] = stats->erpt;
+	data[++i] = stats->uptc + stats->mptc + stats->bptc;
+	data[++i] = stats->uptc;
+	data[++i] = stats->mptc;
+	data[++i] = stats->bptc;
+	data[++i] = stats->ubrc;
+	data[++i] = stats->ubtc;
+	data[++i] = stats->mbrc;
+	data[++i] = stats->mbtc;
+	data[++i] = stats->bbrc;
+	data[++i] = stats->bbtc;
+	if (stats->brc)
+		data[++i] = stats->brc;
+	else
+		data[++i] = stats->ubrc + stats->mbrc + stats->bbrc;
+	if (stats->btc)
+		data[++i] = stats->btc;
+	else
+		data[++i] = stats->ubtc + stats->mbtc + stats->bbtc;
+	data[++i] = stats->dma_pkt_rc;
+	data[++i] = stats->dma_pkt_tc;
+	data[++i] = stats->dma_oct_rc;
+	data[++i] = stats->dma_oct_tc;
+	data[++i] = stats->dpc;
+
+	i++;
+
+	data += i;
+
+	for (tc = 0U; tc < self->aq_nic_cfg.tcs; tc++) {
+		for (i = 0U; self->aq_vecs > i; ++i) {
+			if (!self->aq_vec[i])
+				break;
+			data += count;
+			count = aq_vec_get_sw_stats(self->aq_vec[i], tc, data);
+		}
+	}
+
 	data += count;
-	count = 0U;
 
-	for (i = 0U, aq_vec = self->aq_vec[0];
-		aq_vec && self->aq_vecs > i; ++i, aq_vec = self->aq_vec[i]) {
-		data += count;
-		aq_vec_get_sw_stats(aq_vec, data, &count);
-	}
+err_exit:
+	return data;
+}
 
-err_exit:;
-	(void)err;
+static void aq_nic_update_ndev_stats(struct aq_nic_s *self)
+{
+	struct aq_stats_s *stats = self->aq_hw_ops->hw_get_hw_stats(self->aq_hw);
+	struct net_device *ndev = self->ndev;
+
+	ndev->stats.rx_packets = stats->dma_pkt_rc;
+	ndev->stats.rx_bytes = stats->dma_oct_rc;
+	ndev->stats.rx_errors = stats->erpr;
+	ndev->stats.rx_dropped = stats->dpc;
+	ndev->stats.tx_packets = stats->dma_pkt_tc;
+	ndev->stats.tx_bytes = stats->dma_oct_tc;
+	ndev->stats.tx_errors = stats->erpt;
+	ndev->stats.multicast = stats->mprc;
 }
 
 void aq_nic_get_link_ksettings(struct aq_nic_s *self,
 			       struct ethtool_link_ksettings *cmd)
 {
-	cmd->base.port = PORT_TP;
-	/* This driver supports only 10G capable adapters, so DUPLEX_FULL */
-	cmd->base.duplex = DUPLEX_FULL;
+	u32 lp_link_speed_msk;
+
+	if (self->aq_nic_cfg.aq_hw_caps->media_type == AQ_HW_MEDIA_TYPE_FIBRE)
+		cmd->base.port = PORT_FIBRE;
+	else
+		cmd->base.port = PORT_TP;
+
+	cmd->base.duplex = DUPLEX_UNKNOWN;
+	if (self->link_status.mbps)
+		cmd->base.duplex = self->link_status.full_duplex ?
+				   DUPLEX_FULL : DUPLEX_HALF;
 	cmd->base.autoneg = self->aq_nic_cfg.is_autoneg;
 
 	ethtool_link_ksettings_zero_link_mode(cmd, supported);
 
-	if (self->aq_hw_caps.link_speed_msk & AQ_NIC_RATE_10G)
+	if (self->aq_nic_cfg.aq_hw_caps->link_speed_msk & AQ_NIC_RATE_10G)
 		ethtool_link_ksettings_add_link_mode(cmd, supported,
 						     10000baseT_Full);
 
-	if (self->aq_hw_caps.link_speed_msk & AQ_NIC_RATE_5G)
+	if (self->aq_nic_cfg.aq_hw_caps->link_speed_msk & AQ_NIC_RATE_5G)
 		ethtool_link_ksettings_add_link_mode(cmd, supported,
 						     5000baseT_Full);
 
-	if (self->aq_hw_caps.link_speed_msk & AQ_NIC_RATE_2GS)
+	if (self->aq_nic_cfg.aq_hw_caps->link_speed_msk & AQ_NIC_RATE_2G5)
 		ethtool_link_ksettings_add_link_mode(cmd, supported,
 						     2500baseT_Full);
 
-	if (self->aq_hw_caps.link_speed_msk & AQ_NIC_RATE_1G)
+	if (self->aq_nic_cfg.aq_hw_caps->link_speed_msk & AQ_NIC_RATE_1G)
 		ethtool_link_ksettings_add_link_mode(cmd, supported,
 						     1000baseT_Full);
 
-	if (self->aq_hw_caps.link_speed_msk & AQ_NIC_RATE_100M)
+	if (self->aq_nic_cfg.aq_hw_caps->link_speed_msk & AQ_NIC_RATE_1G_HALF)
+		ethtool_link_ksettings_add_link_mode(cmd, supported,
+						     1000baseT_Half);
+
+	if (self->aq_nic_cfg.aq_hw_caps->link_speed_msk & AQ_NIC_RATE_100M)
 		ethtool_link_ksettings_add_link_mode(cmd, supported,
 						     100baseT_Full);
 
-	if (self->aq_hw_caps.flow_control)
+	if (self->aq_nic_cfg.aq_hw_caps->link_speed_msk & AQ_NIC_RATE_100M_HALF)
+		ethtool_link_ksettings_add_link_mode(cmd, supported,
+						     100baseT_Half);
+
+	if (self->aq_nic_cfg.aq_hw_caps->link_speed_msk & AQ_NIC_RATE_10M)
+		ethtool_link_ksettings_add_link_mode(cmd, supported,
+						     10baseT_Full);
+
+	if (self->aq_nic_cfg.aq_hw_caps->link_speed_msk & AQ_NIC_RATE_10M_HALF)
+		ethtool_link_ksettings_add_link_mode(cmd, supported,
+						     10baseT_Half);
+
+	if (self->aq_nic_cfg.aq_hw_caps->flow_control) {
 		ethtool_link_ksettings_add_link_mode(cmd, supported,
 						     Pause);
+		ethtool_link_ksettings_add_link_mode(cmd, supported,
+						     Asym_Pause);
+	}
 
 	ethtool_link_ksettings_add_link_mode(cmd, supported, Autoneg);
-	ethtool_link_ksettings_add_link_mode(cmd, supported, TP);
+
+	if (self->aq_nic_cfg.aq_hw_caps->media_type == AQ_HW_MEDIA_TYPE_FIBRE)
+		ethtool_link_ksettings_add_link_mode(cmd, supported, FIBRE);
+	else
+		ethtool_link_ksettings_add_link_mode(cmd, supported, TP);
 
 	ethtool_link_ksettings_zero_link_mode(cmd, advertising);
 
 	if (self->aq_nic_cfg.is_autoneg)
 		ethtool_link_ksettings_add_link_mode(cmd, advertising, Autoneg);
 
-	if (self->aq_nic_cfg.link_speed_msk  & AQ_NIC_RATE_10G)
+	if (self->aq_nic_cfg.link_speed_msk & AQ_NIC_RATE_10G)
 		ethtool_link_ksettings_add_link_mode(cmd, advertising,
 						     10000baseT_Full);
 
-	if (self->aq_nic_cfg.link_speed_msk  & AQ_NIC_RATE_5G)
+	if (self->aq_nic_cfg.link_speed_msk & AQ_NIC_RATE_5G)
 		ethtool_link_ksettings_add_link_mode(cmd, advertising,
 						     5000baseT_Full);
 
-	if (self->aq_nic_cfg.link_speed_msk  & AQ_NIC_RATE_2GS)
+	if (self->aq_nic_cfg.link_speed_msk & AQ_NIC_RATE_2G5)
 		ethtool_link_ksettings_add_link_mode(cmd, advertising,
 						     2500baseT_Full);
 
-	if (self->aq_nic_cfg.link_speed_msk  & AQ_NIC_RATE_1G)
+	if (self->aq_nic_cfg.link_speed_msk & AQ_NIC_RATE_1G)
 		ethtool_link_ksettings_add_link_mode(cmd, advertising,
 						     1000baseT_Full);
 
-	if (self->aq_nic_cfg.link_speed_msk  & AQ_NIC_RATE_100M)
+	if (self->aq_nic_cfg.link_speed_msk & AQ_NIC_RATE_1G_HALF)
+		ethtool_link_ksettings_add_link_mode(cmd, advertising,
+						     1000baseT_Half);
+
+	if (self->aq_nic_cfg.link_speed_msk & AQ_NIC_RATE_100M)
 		ethtool_link_ksettings_add_link_mode(cmd, advertising,
 						     100baseT_Full);
 
-	if (self->aq_nic_cfg.flow_control)
+	if (self->aq_nic_cfg.link_speed_msk & AQ_NIC_RATE_100M_HALF)
+		ethtool_link_ksettings_add_link_mode(cmd, advertising,
+						     100baseT_Half);
+
+	if (self->aq_nic_cfg.link_speed_msk & AQ_NIC_RATE_10M)
+		ethtool_link_ksettings_add_link_mode(cmd, advertising,
+						     10baseT_Full);
+
+	if (self->aq_nic_cfg.link_speed_msk & AQ_NIC_RATE_10M_HALF)
+		ethtool_link_ksettings_add_link_mode(cmd, advertising,
+						     10baseT_Half);
+
+	if (self->aq_nic_cfg.fc.cur & AQ_NIC_FC_RX)
 		ethtool_link_ksettings_add_link_mode(cmd, advertising,
 						     Pause);
 
-	ethtool_link_ksettings_add_link_mode(cmd, advertising, TP);
+	/* Asym is when either RX or TX, but not both */
+	if (!!(self->aq_nic_cfg.fc.cur & AQ_NIC_FC_TX) ^
+	    !!(self->aq_nic_cfg.fc.cur & AQ_NIC_FC_RX))
+		ethtool_link_ksettings_add_link_mode(cmd, advertising,
+						     Asym_Pause);
+
+	if (self->aq_nic_cfg.aq_hw_caps->media_type == AQ_HW_MEDIA_TYPE_FIBRE)
+		ethtool_link_ksettings_add_link_mode(cmd, advertising, FIBRE);
+	else
+		ethtool_link_ksettings_add_link_mode(cmd, advertising, TP);
+
+	ethtool_link_ksettings_zero_link_mode(cmd, lp_advertising);
+	lp_link_speed_msk = self->aq_hw->aq_link_status.lp_link_speed_msk;
+
+	if (lp_link_speed_msk & AQ_NIC_RATE_10G)
+		ethtool_link_ksettings_add_link_mode(cmd, lp_advertising,
+						     10000baseT_Full);
+
+	if (lp_link_speed_msk & AQ_NIC_RATE_5G)
+		ethtool_link_ksettings_add_link_mode(cmd, lp_advertising,
+						     5000baseT_Full);
+
+	if (lp_link_speed_msk & AQ_NIC_RATE_2G5)
+		ethtool_link_ksettings_add_link_mode(cmd, lp_advertising,
+						     2500baseT_Full);
+
+	if (lp_link_speed_msk & AQ_NIC_RATE_1G)
+		ethtool_link_ksettings_add_link_mode(cmd, lp_advertising,
+						     1000baseT_Full);
+
+	if (lp_link_speed_msk & AQ_NIC_RATE_1G_HALF)
+		ethtool_link_ksettings_add_link_mode(cmd, lp_advertising,
+						     1000baseT_Half);
+
+	if (lp_link_speed_msk & AQ_NIC_RATE_100M)
+		ethtool_link_ksettings_add_link_mode(cmd, lp_advertising,
+						     100baseT_Full);
+
+	if (lp_link_speed_msk & AQ_NIC_RATE_100M_HALF)
+		ethtool_link_ksettings_add_link_mode(cmd, lp_advertising,
+						     100baseT_Half);
+
+	if (lp_link_speed_msk & AQ_NIC_RATE_10M)
+		ethtool_link_ksettings_add_link_mode(cmd, lp_advertising,
+						     10baseT_Full);
+
+	if (lp_link_speed_msk & AQ_NIC_RATE_10M_HALF)
+		ethtool_link_ksettings_add_link_mode(cmd, lp_advertising,
+						     10baseT_Half);
+
+	if (self->aq_hw->aq_link_status.lp_flow_control & AQ_NIC_FC_RX)
+		ethtool_link_ksettings_add_link_mode(cmd, lp_advertising,
+						     Pause);
+	if (!!(self->aq_hw->aq_link_status.lp_flow_control & AQ_NIC_FC_TX) ^
+	    !!(self->aq_hw->aq_link_status.lp_flow_control & AQ_NIC_FC_RX))
+		ethtool_link_ksettings_add_link_mode(cmd, lp_advertising,
+						     Asym_Pause);
 }
 
 int aq_nic_set_link_ksettings(struct aq_nic_s *self,
 			      const struct ethtool_link_ksettings *cmd)
 {
-	u32 speed = 0U;
+	int fduplex = (cmd->base.duplex == DUPLEX_FULL);
+	u32 speed = cmd->base.speed;
 	u32 rate = 0U;
 	int err = 0;
 
+	if (!fduplex && speed > SPEED_1000) {
+		err = -EINVAL;
+		goto err_exit;
+	}
+
 	if (cmd->base.autoneg == AUTONEG_ENABLE) {
-		rate = self->aq_hw_caps.link_speed_msk;
+		rate = self->aq_nic_cfg.aq_hw_caps->link_speed_msk;
 		self->aq_nic_cfg.is_autoneg = true;
 	} else {
-		speed = cmd->base.speed;
-
 		switch (speed) {
+		case SPEED_10:
+			rate = fduplex ? AQ_NIC_RATE_10M : AQ_NIC_RATE_10M_HALF;
+			break;
+
 		case SPEED_100:
-			rate = AQ_NIC_RATE_100M;
+			rate = fduplex ? AQ_NIC_RATE_100M
+				       : AQ_NIC_RATE_100M_HALF;
 			break;
 
 		case SPEED_1000:
-			rate = AQ_NIC_RATE_1G;
+			rate = fduplex ? AQ_NIC_RATE_1G : AQ_NIC_RATE_1G_HALF;
 			break;
 
 		case SPEED_2500:
-			rate = AQ_NIC_RATE_2GS;
+			rate = AQ_NIC_RATE_2G5;
 			break;
 
 		case SPEED_5000:
@@ -870,9 +1315,8 @@ int aq_nic_set_link_ksettings(struct aq_nic_s *self,
 		default:
 			err = -1;
 			goto err_exit;
-		break;
 		}
-		if (!(self->aq_hw_caps.link_speed_msk & rate)) {
+		if (!(self->aq_nic_cfg.aq_hw_caps->link_speed_msk & rate)) {
 			err = -1;
 			goto err_exit;
 		}
@@ -880,7 +1324,9 @@ int aq_nic_set_link_ksettings(struct aq_nic_s *self,
 		self->aq_nic_cfg.is_autoneg = false;
 	}
 
-	err = self->aq_hw_ops.hw_set_link_speed(self->aq_hw, rate);
+	mutex_lock(&self->fwreq_mutex);
+	err = self->aq_fw_ops->set_link_speed(self->aq_hw, rate);
+	mutex_unlock(&self->fwreq_mutex);
 	if (err < 0)
 		goto err_exit;
 
@@ -897,39 +1343,88 @@ struct aq_nic_cfg_s *aq_nic_get_cfg(struct aq_nic_s *self)
 
 u32 aq_nic_get_fw_version(struct aq_nic_s *self)
 {
-	u32 fw_version = 0U;
+	return self->aq_hw_ops->hw_get_fw_version(self->aq_hw);
+}
+
+int aq_nic_set_loopback(struct aq_nic_s *self)
+{
+	struct aq_nic_cfg_s *cfg = &self->aq_nic_cfg;
 
-	self->aq_hw_ops.hw_get_fw_version(self->aq_hw, &fw_version);
+	if (!self->aq_hw_ops->hw_set_loopback ||
+	    !self->aq_fw_ops->set_phyloopback)
+		return -EOPNOTSUPP;
+
+	mutex_lock(&self->fwreq_mutex);
+	self->aq_hw_ops->hw_set_loopback(self->aq_hw,
+					 AQ_HW_LOOPBACK_DMA_SYS,
+					 !!(cfg->priv_flags &
+					    BIT(AQ_HW_LOOPBACK_DMA_SYS)));
+
+	self->aq_hw_ops->hw_set_loopback(self->aq_hw,
+					 AQ_HW_LOOPBACK_PKT_SYS,
+					 !!(cfg->priv_flags &
+					    BIT(AQ_HW_LOOPBACK_PKT_SYS)));
+
+	self->aq_hw_ops->hw_set_loopback(self->aq_hw,
+					 AQ_HW_LOOPBACK_DMA_NET,
+					 !!(cfg->priv_flags &
+					    BIT(AQ_HW_LOOPBACK_DMA_NET)));
+
+	self->aq_fw_ops->set_phyloopback(self->aq_hw,
+					 AQ_HW_LOOPBACK_PHYINT_SYS,
+					 !!(cfg->priv_flags &
+					    BIT(AQ_HW_LOOPBACK_PHYINT_SYS)));
+
+	self->aq_fw_ops->set_phyloopback(self->aq_hw,
+					 AQ_HW_LOOPBACK_PHYEXT_SYS,
+					 !!(cfg->priv_flags &
+					    BIT(AQ_HW_LOOPBACK_PHYEXT_SYS)));
+	mutex_unlock(&self->fwreq_mutex);
 
-	return fw_version;
+	return 0;
 }
 
 int aq_nic_stop(struct aq_nic_s *self)
 {
-	struct aq_vec_s *aq_vec = NULL;
 	unsigned int i = 0U;
 
-	for (i = 0U, aq_vec = self->aq_vec[0];
-		self->aq_vecs > i; ++i, aq_vec = self->aq_vec[i])
-		aq_nic_ndev_queue_stop(self, i);
+	netif_tx_disable(self->ndev);
+	netif_carrier_off(self->ndev);
 
-	del_timer_sync(&self->service_timer);
+	timer_delete_sync(&self->service_timer);
+	cancel_work_sync(&self->service_task);
 
-	self->aq_hw_ops.hw_irq_disable(self->aq_hw, AQ_CFG_IRQ_MASK);
+	self->aq_hw_ops->hw_irq_disable(self->aq_hw, AQ_CFG_IRQ_MASK);
 
 	if (self->aq_nic_cfg.is_polling)
-		del_timer_sync(&self->polling_timer);
+		timer_delete_sync(&self->polling_timer);
 	else
-		aq_pci_func_free_irqs(self->aq_pci_func);
+		aq_pci_func_free_irqs(self);
+
+	aq_ptp_irq_free(self);
 
-	for (i = 0U, aq_vec = self->aq_vec[0];
-		self->aq_vecs > i; ++i, aq_vec = self->aq_vec[i])
-		aq_vec_stop(aq_vec);
+	for (i = 0U; self->aq_vecs > i; ++i)
+		aq_vec_stop(self->aq_vec[i]);
 
-	return self->aq_hw_ops.hw_stop(self->aq_hw);
+	aq_ptp_ring_stop(self);
+
+	return self->aq_hw_ops->hw_stop(self->aq_hw);
+}
+
+void aq_nic_set_power(struct aq_nic_s *self)
+{
+	if (self->power_state != AQ_HW_POWER_STATE_D0 ||
+	    self->aq_hw->aq_nic_cfg->wol)
+		if (likely(self->aq_fw_ops->set_power)) {
+			mutex_lock(&self->fwreq_mutex);
+			self->aq_fw_ops->set_power(self->aq_hw,
+						   self->power_state,
+						   self->ndev->dev_addr);
+			mutex_unlock(&self->fwreq_mutex);
+		}
 }
 
-void aq_nic_deinit(struct aq_nic_s *self)
+void aq_nic_deinit(struct aq_nic_s *self, bool link_down)
 {
 	struct aq_vec_s *aq_vec = NULL;
 	unsigned int i = 0U;
@@ -937,28 +1432,36 @@ void aq_nic_deinit(struct aq_nic_s *self)
 	if (!self)
 		goto err_exit;
 
-	for (i = 0U, aq_vec = self->aq_vec[0];
-		self->aq_vecs > i; ++i, aq_vec = self->aq_vec[i])
+	for (i = 0U; i < self->aq_vecs; i++) {
+		aq_vec = self->aq_vec[i];
 		aq_vec_deinit(aq_vec);
+		aq_vec_ring_free(aq_vec);
+	}
 
-	if (self->power_state == AQ_HW_POWER_STATE_D0) {
-		(void)self->aq_hw_ops.hw_deinit(self->aq_hw);
-	} else {
-		(void)self->aq_hw_ops.hw_set_power(self->aq_hw,
-						   self->power_state);
+	aq_ptp_unregister(self);
+	aq_ptp_ring_deinit(self);
+	aq_ptp_ring_free(self);
+	aq_ptp_free(self);
+
+	/* May be invoked during hot unplug. */
+	if (pci_device_is_present(self->pdev) &&
+	    likely(self->aq_fw_ops->deinit) && link_down) {
+		mutex_lock(&self->fwreq_mutex);
+		self->aq_fw_ops->deinit(self->aq_hw);
+		mutex_unlock(&self->fwreq_mutex);
 	}
 
 err_exit:;
 }
 
-void aq_nic_free_hot_resources(struct aq_nic_s *self)
+void aq_nic_free_vectors(struct aq_nic_s *self)
 {
 	unsigned int i = 0U;
 
 	if (!self)
 		goto err_exit;
 
-	for (i = AQ_DIMOF(self->aq_vec); i--;) {
+	for (i = ARRAY_SIZE(self->aq_vec); i--;) {
 		if (self->aq_vec[i]) {
 			aq_vec_free(self->aq_vec[i]);
 			self->aq_vec[i] = NULL;
@@ -968,40 +1471,225 @@ void aq_nic_free_hot_resources(struct aq_nic_s *self)
 err_exit:;
 }
 
-int aq_nic_change_pm_state(struct aq_nic_s *self, pm_message_t *pm_msg)
+int aq_nic_realloc_vectors(struct aq_nic_s *self)
 {
-	int err = 0;
+	struct aq_nic_cfg_s *cfg = aq_nic_get_cfg(self);
 
-	if (!netif_running(self->ndev)) {
-		err = 0;
-		goto out;
+	aq_nic_free_vectors(self);
+
+	for (self->aq_vecs = 0; self->aq_vecs < cfg->vecs; self->aq_vecs++) {
+		self->aq_vec[self->aq_vecs] = aq_vec_alloc(self, self->aq_vecs,
+							   cfg);
+		if (unlikely(!self->aq_vec[self->aq_vecs]))
+			return -ENOMEM;
 	}
+
+	return 0;
+}
+
+void aq_nic_shutdown(struct aq_nic_s *self)
+{
+	int err = 0;
+
+	if (!self->ndev)
+		return;
+
 	rtnl_lock();
-	if (pm_msg->event & PM_EVENT_SLEEP || pm_msg->event & PM_EVENT_FREEZE) {
-		self->power_state = AQ_HW_POWER_STATE_D3;
-		netif_device_detach(self->ndev);
-		netif_tx_stop_all_queues(self->ndev);
 
+	netif_device_detach(self->ndev);
+
+	if (netif_running(self->ndev)) {
 		err = aq_nic_stop(self);
 		if (err < 0)
 			goto err_exit;
+	}
+	aq_nic_deinit(self, !self->aq_hw->aq_nic_cfg->wol);
+	aq_nic_set_power(self);
 
-		aq_nic_deinit(self);
-	} else {
-		err = aq_nic_init(self);
-		if (err < 0)
-			goto err_exit;
+err_exit:
+	rtnl_unlock();
+}
 
-		err = aq_nic_start(self);
-		if (err < 0)
-			goto err_exit;
+u8 aq_nic_reserve_filter(struct aq_nic_s *self, enum aq_rx_filter_type type)
+{
+	u8 location = 0xFF;
+	u32 fltr_cnt;
+	u32 n_bit;
+
+	switch (type) {
+	case aq_rx_filter_ethertype:
+		location = AQ_RX_LAST_LOC_FETHERT - AQ_RX_FIRST_LOC_FETHERT -
+			   self->aq_hw_rx_fltrs.fet_reserved_count;
+		self->aq_hw_rx_fltrs.fet_reserved_count++;
+		break;
+	case aq_rx_filter_l3l4:
+		fltr_cnt = AQ_RX_LAST_LOC_FL3L4 - AQ_RX_FIRST_LOC_FL3L4;
+		n_bit = fltr_cnt - self->aq_hw_rx_fltrs.fl3l4.reserved_count;
 
-		netif_device_attach(self->ndev);
-		netif_tx_start_all_queues(self->ndev);
+		self->aq_hw_rx_fltrs.fl3l4.active_ipv4 |= BIT(n_bit);
+		self->aq_hw_rx_fltrs.fl3l4.reserved_count++;
+		location = n_bit;
+		break;
+	default:
+		break;
 	}
 
+	return location;
+}
+
+void aq_nic_release_filter(struct aq_nic_s *self, enum aq_rx_filter_type type,
+			   u32 location)
+{
+	switch (type) {
+	case aq_rx_filter_ethertype:
+		self->aq_hw_rx_fltrs.fet_reserved_count--;
+		break;
+	case aq_rx_filter_l3l4:
+		self->aq_hw_rx_fltrs.fl3l4.reserved_count--;
+		self->aq_hw_rx_fltrs.fl3l4.active_ipv4 &= ~BIT(location);
+		break;
+	default:
+		break;
+	}
+}
+
+int aq_nic_set_downshift(struct aq_nic_s *self, int val)
+{
+	int err = 0;
+	struct aq_nic_cfg_s *cfg = &self->aq_nic_cfg;
+
+	if (!self->aq_fw_ops->set_downshift)
+		return -EOPNOTSUPP;
+
+	if (val > 15) {
+		netdev_err(self->ndev, "downshift counter should be <= 15\n");
+		return -EINVAL;
+	}
+	cfg->downshift_counter = val;
+
+	mutex_lock(&self->fwreq_mutex);
+	err = self->aq_fw_ops->set_downshift(self->aq_hw, cfg->downshift_counter);
+	mutex_unlock(&self->fwreq_mutex);
+
+	return err;
+}
+
+int aq_nic_set_media_detect(struct aq_nic_s *self, int val)
+{
+	struct aq_nic_cfg_s *cfg = &self->aq_nic_cfg;
+	int err = 0;
+
+	if (!self->aq_fw_ops->set_media_detect)
+		return -EOPNOTSUPP;
+
+	if (val > 0 && val != AQ_HW_MEDIA_DETECT_CNT) {
+		netdev_err(self->ndev, "EDPD on this device could have only fixed value of %d\n",
+			   AQ_HW_MEDIA_DETECT_CNT);
+		return -EINVAL;
+	}
+
+	mutex_lock(&self->fwreq_mutex);
+	err = self->aq_fw_ops->set_media_detect(self->aq_hw, !!val);
+	mutex_unlock(&self->fwreq_mutex);
+
+	/* msecs plays no role - configuration is always fixed in PHY */
+	if (!err)
+		cfg->is_media_detect = !!val;
+
+	return err;
+}
+
+int aq_nic_setup_tc_mqprio(struct aq_nic_s *self, u32 tcs, u8 *prio_tc_map)
+{
+	struct aq_nic_cfg_s *cfg = &self->aq_nic_cfg;
+	const unsigned int prev_vecs = cfg->vecs;
+	bool ndev_running;
+	int err = 0;
+	int i;
+
+	/* if already the same configuration or
+	 * disable request (tcs is 0) and we already is disabled
+	 */
+	if (tcs == cfg->tcs || (tcs == 0 && !cfg->is_qos))
+		return 0;
+
+	ndev_running = netif_running(self->ndev);
+	if (ndev_running)
+		dev_close(self->ndev);
+
+	cfg->tcs = tcs;
+	if (cfg->tcs == 0)
+		cfg->tcs = 1;
+	if (prio_tc_map)
+		memcpy(cfg->prio_tc_map, prio_tc_map, sizeof(cfg->prio_tc_map));
+	else
+		for (i = 0; i < sizeof(cfg->prio_tc_map); i++)
+			cfg->prio_tc_map[i] = cfg->tcs * i / 8;
+
+	cfg->is_qos = !!tcs;
+	cfg->is_ptp = (cfg->tcs <= AQ_HW_PTP_TC);
+	if (!cfg->is_ptp)
+		netdev_warn(self->ndev, "%s\n",
+			    "PTP is auto disabled due to requested TC count.");
+
+	netdev_set_num_tc(self->ndev, cfg->tcs);
+
+	/* Changing the number of TCs might change the number of vectors */
+	aq_nic_cfg_update_num_vecs(self);
+	if (prev_vecs != cfg->vecs) {
+		err = aq_nic_realloc_vectors(self);
+		if (err)
+			goto err_exit;
+	}
+
+	if (ndev_running)
+		err = dev_open(self->ndev, NULL);
+
 err_exit:
-	rtnl_unlock();
-out:
 	return err;
 }
+
+int aq_nic_setup_tc_max_rate(struct aq_nic_s *self, const unsigned int tc,
+			     const u32 max_rate)
+{
+	struct aq_nic_cfg_s *cfg = &self->aq_nic_cfg;
+
+	if (tc >= AQ_CFG_TCS_MAX)
+		return -EINVAL;
+
+	if (max_rate && max_rate < 10) {
+		netdev_warn(self->ndev,
+			"Setting %s to the minimum usable value of %dMbps.\n",
+			"max rate", 10);
+		cfg->tc_max_rate[tc] = 10;
+	} else {
+		cfg->tc_max_rate[tc] = max_rate;
+	}
+
+	return 0;
+}
+
+int aq_nic_setup_tc_min_rate(struct aq_nic_s *self, const unsigned int tc,
+			     const u32 min_rate)
+{
+	struct aq_nic_cfg_s *cfg = &self->aq_nic_cfg;
+
+	if (tc >= AQ_CFG_TCS_MAX)
+		return -EINVAL;
+
+	if (min_rate)
+		set_bit(tc, &cfg->tc_min_rate_msk);
+	else
+		clear_bit(tc, &cfg->tc_min_rate_msk);
+
+	if (min_rate && min_rate < 20) {
+		netdev_warn(self->ndev,
+			"Setting %s to the minimum usable value of %dMbps.\n",
+			"min rate", 20);
+		cfg->tc_min_rate[tc] = 20;
+	} else {
+		cfg->tc_min_rate[tc] = min_rate;
+	}
+
+	return 0;
+}
diff --git a/drivers/net/ethernet/aquantia/atlantic/aq_nic.h b/drivers/net/ethernet/aquantia/atlantic/aq_nic.h
index 7fc2a5ecb2b7..ad33f8586532 100644
--- a/drivers/net/ethernet/aquantia/atlantic/aq_nic.h
+++ b/drivers/net/ethernet/aquantia/atlantic/aq_nic.h
@@ -1,10 +1,8 @@
-/*
- * aQuantia Corporation Network Driver
- * Copyright (C) 2014-2017 aQuantia Corporation. All rights reserved
+/* SPDX-License-Identifier: GPL-2.0-only */
+/* Atlantic Network Driver
  *
- * This program is free software; you can redistribute it and/or modify it
- * under the terms and conditions of the GNU General Public License,
- * version 2, as published by the Free Software Foundation.
+ * Copyright (C) 2014-2019 aQuantia Corporation
+ * Copyright (C) 2019-2020 Marvell International Ltd.
  */
 
 /* File aq_nic.h: Declaration of common code for NIC. */
@@ -12,88 +10,190 @@
 #ifndef AQ_NIC_H
 #define AQ_NIC_H
 
+#include <linux/ethtool.h>
+#include <net/xdp.h>
+#include <linux/bpf.h>
+
 #include "aq_common.h"
 #include "aq_rss.h"
+#include "aq_hw.h"
 
 struct aq_ring_s;
-struct aq_pci_func_s;
 struct aq_hw_ops;
+struct aq_fw_s;
+struct aq_vec_s;
+struct aq_macsec_cfg;
+struct aq_ptp_s;
+enum aq_rx_filter_type;
 
-#define AQ_NIC_FC_OFF    0U
-#define AQ_NIC_FC_TX     1U
-#define AQ_NIC_FC_RX     2U
-#define AQ_NIC_FC_FULL   3U
-#define AQ_NIC_FC_AUTO   4U
+enum aq_fc_mode {
+	AQ_NIC_FC_OFF = 0,
+	AQ_NIC_FC_TX,
+	AQ_NIC_FC_RX,
+	AQ_NIC_FC_FULL,
+};
 
-#define AQ_NIC_RATE_10G        BIT(0)
-#define AQ_NIC_RATE_5G         BIT(1)
-#define AQ_NIC_RATE_5GSR       BIT(2)
-#define AQ_NIC_RATE_2GS        BIT(3)
-#define AQ_NIC_RATE_1G         BIT(4)
-#define AQ_NIC_RATE_100M       BIT(5)
+struct aq_fc_info {
+	enum aq_fc_mode req;
+	enum aq_fc_mode cur;
+};
 
 struct aq_nic_cfg_s {
-	struct aq_hw_caps_s *aq_hw_caps;
-	u64 hw_features;
+	const struct aq_hw_caps_s *aq_hw_caps;
+	u64 features;
 	u32 rxds;		/* rx ring size, descriptors # */
 	u32 txds;		/* tx ring size, descriptors # */
-	u32 vecs;		/* vecs==allocated irqs */
+	u32 vecs;		/* allocated rx/tx vectors */
+	u32 link_irq_vec;
 	u32 irq_type;
 	u32 itr;
+	u16 rx_itr;
+	u16 tx_itr;
+	u32 rxpageorder;
 	u32 num_rss_queues;
 	u32 mtu;
-	u32 ucp_0x364;
-	u32 flow_control;
+	struct aq_fc_info fc;
 	u32 link_speed_msk;
-	u32 vlan_id;
+	u32 wol;
+	u8 is_vlan_rx_strip;
+	u8 is_vlan_tx_insert;
+	bool is_vlan_force_promisc;
 	u16 is_mc_list_enabled;
 	u16 mc_list_count;
 	bool is_autoneg;
-	bool is_interrupt_moderation;
 	bool is_polling;
 	bool is_rss;
 	bool is_lro;
+	bool is_qos;
+	bool is_ptp;
+	bool is_media_detect;
+	int downshift_counter;
+	enum aq_tc_mode tc_mode;
+	u32 priv_flags;
 	u8  tcs;
+	u8 prio_tc_map[8];
+	u32 tc_max_rate[AQ_CFG_TCS_MAX];
+	unsigned long tc_min_rate_msk;
+	u32 tc_min_rate[AQ_CFG_TCS_MAX];
 	struct aq_rss_parameters aq_rss;
+	u32 eee_speeds;
 };
 
 #define AQ_NIC_FLAG_STARTED     0x00000004U
 #define AQ_NIC_FLAG_STOPPING    0x00000008U
 #define AQ_NIC_FLAG_RESETTING   0x00000010U
 #define AQ_NIC_FLAG_CLOSING     0x00000020U
+#define AQ_NIC_PTP_DPATH_UP     0x02000000U
 #define AQ_NIC_LINK_DOWN        0x04000000U
 #define AQ_NIC_FLAG_ERR_UNPLUG  0x40000000U
 #define AQ_NIC_FLAG_ERR_HW      0x80000000U
 
-#define AQ_NIC_TCVEC2RING(_NIC_, _TC_, _VEC_) \
-	((_TC_) * AQ_CFG_TCS_MAX + (_VEC_))
+#define AQ_NIC_QUIRK_BAD_PTP    BIT(0)
+
+#define AQ_NIC_WOL_MODES        (WAKE_MAGIC |\
+				 WAKE_PHY)
+
+#define AQ_NIC_CFG_RING_PER_TC(_NIC_CFG_) \
+	(((_NIC_CFG_)->tc_mode == AQ_TC_MODE_4TCS) ? 8 : 4)
+
+#define AQ_NIC_CFG_TCVEC2RING(_NIC_CFG_, _TC_, _VEC_) \
+	((_TC_) * AQ_NIC_CFG_RING_PER_TC(_NIC_CFG_) + (_VEC_))
+
+#define AQ_NIC_RING2QMAP(_NIC_, _ID_) \
+	((_ID_) / AQ_NIC_CFG_RING_PER_TC(&(_NIC_)->aq_nic_cfg) * \
+		(_NIC_)->aq_vecs + \
+	((_ID_) % AQ_NIC_CFG_RING_PER_TC(&(_NIC_)->aq_nic_cfg)))
+
+struct aq_hw_rx_fl2 {
+	struct aq_rx_filter_vlan aq_vlans[AQ_VLAN_MAX_FILTERS];
+};
+
+struct aq_hw_rx_fl3l4 {
+	u8 active_ipv4;
+	u8 active_ipv6:2;
+	u8 is_ipv6;
+	u8 reserved_count;
+};
+
+struct aq_hw_rx_fltrs_s {
+	struct hlist_head     filter_list;
+	u16                   active_filters;
+	struct aq_hw_rx_fl2   fl2;
+	struct aq_hw_rx_fl3l4 fl3l4;
+	/* filter ether type */
+	u8 fet_reserved_count;
+};
+
+struct aq_nic_s {
+	atomic_t flags;
+	u32 msg_enable;
+	struct aq_vec_s *aq_vec[AQ_CFG_VECS_MAX];
+	struct aq_ring_s *aq_ring_tx[AQ_HW_QUEUES_MAX];
+	struct aq_hw_s *aq_hw;
+	struct bpf_prog *xdp_prog;
+	struct net_device *ndev;
+	unsigned int aq_vecs;
+	unsigned int packet_filter;
+	unsigned int power_state;
+	u8 port;
+	const struct aq_hw_ops *aq_hw_ops;
+	const struct aq_fw_ops *aq_fw_ops;
+	struct aq_nic_cfg_s aq_nic_cfg;
+	struct timer_list service_timer;
+	struct work_struct service_task;
+	struct timer_list polling_timer;
+	struct aq_hw_link_status_s link_status;
+	struct {
+		u32 count;
+		u8 ar[AQ_HW_MULTICAST_ADDRESS_MAX][ETH_ALEN];
+	} mc_list;
+	/* Bitmask of currently assigned vlans from linux */
+	unsigned long active_vlans[BITS_TO_LONGS(VLAN_N_VID)];
 
-struct aq_nic_s *aq_nic_alloc_cold(const struct net_device_ops *ndev_ops,
-				   const struct ethtool_ops *et_ops,
-				   struct device *dev,
-				   struct aq_pci_func_s *aq_pci_func,
-				   unsigned int port,
-				   const struct aq_hw_ops *aq_hw_ops);
-int aq_nic_ndev_init(struct aq_nic_s *self);
+	struct pci_dev *pdev;
+	unsigned int msix_entry_mask;
+	u32 irqvecs;
+	/* mutex to serialize FW interface access operations */
+	struct mutex fwreq_mutex;
+#if IS_ENABLED(CONFIG_MACSEC)
+	struct aq_macsec_cfg *macsec_cfg;
+	/* mutex to protect data in macsec_cfg */
+	struct mutex macsec_mutex;
+#endif
+	/* PTP support */
+	struct aq_ptp_s *aq_ptp;
+	struct aq_hw_rx_fltrs_s aq_hw_rx_fltrs;
+};
+
+static inline struct device *aq_nic_get_dev(struct aq_nic_s *self)
+{
+	return self->ndev->dev.parent;
+}
+
+void aq_nic_ndev_init(struct aq_nic_s *self);
 struct aq_nic_s *aq_nic_alloc_hot(struct net_device *ndev);
 void aq_nic_set_tx_ring(struct aq_nic_s *self, unsigned int idx,
 			struct aq_ring_s *ring);
-struct device *aq_nic_get_dev(struct aq_nic_s *self);
 struct net_device *aq_nic_get_ndev(struct aq_nic_s *self);
 int aq_nic_init(struct aq_nic_s *self);
-int aq_nic_cfg_start(struct aq_nic_s *self);
+void aq_nic_cfg_start(struct aq_nic_s *self);
 int aq_nic_ndev_register(struct aq_nic_s *self);
-void aq_nic_ndev_queue_start(struct aq_nic_s *self, unsigned int idx);
-void aq_nic_ndev_queue_stop(struct aq_nic_s *self, unsigned int idx);
 void aq_nic_ndev_free(struct aq_nic_s *self);
 int aq_nic_start(struct aq_nic_s *self);
+unsigned int aq_nic_map_skb(struct aq_nic_s *self, struct sk_buff *skb,
+			    struct aq_ring_s *ring);
+int aq_nic_xmit_xdpf(struct aq_nic_s *aq_nic, struct aq_ring_s *tx_ring,
+		     struct xdp_frame *xdpf);
 int aq_nic_xmit(struct aq_nic_s *self, struct sk_buff *skb);
 int aq_nic_get_regs(struct aq_nic_s *self, struct ethtool_regs *regs, void *p);
 int aq_nic_get_regs_count(struct aq_nic_s *self);
-void aq_nic_get_stats(struct aq_nic_s *self, u64 *data);
+u64 *aq_nic_get_stats(struct aq_nic_s *self, u64 *data);
 int aq_nic_stop(struct aq_nic_s *self);
-void aq_nic_deinit(struct aq_nic_s *self);
+void aq_nic_deinit(struct aq_nic_s *self, bool link_down);
+void aq_nic_set_power(struct aq_nic_s *self);
 void aq_nic_free_hot_resources(struct aq_nic_s *self);
+void aq_nic_free_vectors(struct aq_nic_s *self);
+int aq_nic_realloc_vectors(struct aq_nic_s *self);
 int aq_nic_set_mtu(struct aq_nic_s *self, int new_mtu);
 int aq_nic_set_mac(struct aq_nic_s *self, struct net_device *ndev);
 int aq_nic_set_packet_filter(struct aq_nic_s *self, unsigned int flags);
@@ -105,6 +205,17 @@ int aq_nic_set_link_ksettings(struct aq_nic_s *self,
 			      const struct ethtool_link_ksettings *cmd);
 struct aq_nic_cfg_s *aq_nic_get_cfg(struct aq_nic_s *self);
 u32 aq_nic_get_fw_version(struct aq_nic_s *self);
-int aq_nic_change_pm_state(struct aq_nic_s *self, pm_message_t *pm_msg);
-
+int aq_nic_set_loopback(struct aq_nic_s *self);
+int aq_nic_set_downshift(struct aq_nic_s *self, int val);
+int aq_nic_set_media_detect(struct aq_nic_s *self, int val);
+int aq_nic_update_interrupt_moderation_settings(struct aq_nic_s *self);
+void aq_nic_shutdown(struct aq_nic_s *self);
+u8 aq_nic_reserve_filter(struct aq_nic_s *self, enum aq_rx_filter_type type);
+void aq_nic_release_filter(struct aq_nic_s *self, enum aq_rx_filter_type type,
+			   u32 location);
+int aq_nic_setup_tc_mqprio(struct aq_nic_s *self, u32 tcs, u8 *prio_tc_map);
+int aq_nic_setup_tc_max_rate(struct aq_nic_s *self, const unsigned int tc,
+			     const u32 max_rate);
+int aq_nic_setup_tc_min_rate(struct aq_nic_s *self, const unsigned int tc,
+			     const u32 min_rate);
 #endif /* AQ_NIC_H */
diff --git a/drivers/net/ethernet/aquantia/atlantic/aq_nic_internal.h b/drivers/net/ethernet/aquantia/atlantic/aq_nic_internal.h
deleted file mode 100644
index e7d2711dc165..000000000000
--- a/drivers/net/ethernet/aquantia/atlantic/aq_nic_internal.h
+++ /dev/null
@@ -1,45 +0,0 @@
-/*
- * aQuantia Corporation Network Driver
- * Copyright (C) 2014-2017 aQuantia Corporation. All rights reserved
- *
- * This program is free software; you can redistribute it and/or modify it
- * under the terms and conditions of the GNU General Public License,
- * version 2, as published by the Free Software Foundation.
- */
-
-/* File aq_nic_internal.h: Definition of private object structure. */
-
-#ifndef AQ_NIC_INTERNAL_H
-#define AQ_NIC_INTERNAL_H
-
-struct aq_nic_s {
-	struct aq_obj_s header;
-	struct aq_vec_s *aq_vec[AQ_CFG_VECS_MAX];
-	struct aq_ring_s *aq_ring_tx[AQ_CFG_VECS_MAX * AQ_CFG_TCS_MAX];
-	struct aq_hw_s *aq_hw;
-	struct net_device *ndev;
-	struct aq_pci_func_s *aq_pci_func;
-	unsigned int aq_vecs;
-	unsigned int packet_filter;
-	unsigned int power_state;
-	u8 port;
-	struct aq_hw_ops aq_hw_ops;
-	struct aq_hw_caps_s aq_hw_caps;
-	struct aq_nic_cfg_s aq_nic_cfg;
-	struct timer_list service_timer;
-	struct timer_list polling_timer;
-	struct aq_hw_link_status_s link_status;
-	struct {
-		u32 count;
-		u8 ar[AQ_CFG_MULTICAST_ADDRESS_MAX][ETH_ALEN];
-	} mc_list;
-};
-
-#define AQ_NIC_FLAGS_IS_NOT_READY (AQ_NIC_FLAG_STOPPING | \
-			AQ_NIC_FLAG_RESETTING | AQ_NIC_FLAG_CLOSING | \
-			AQ_NIC_FLAG_ERR_UNPLUG | AQ_NIC_FLAG_ERR_HW)
-
-#define AQ_NIC_FLAGS_IS_NOT_TX_READY (AQ_NIC_FLAGS_IS_NOT_READY | \
-					AQ_NIC_LINK_DOWN)
-
-#endif /* AQ_NIC_INTERNAL_H */
diff --git a/drivers/net/ethernet/aquantia/atlantic/aq_pci_func.c b/drivers/net/ethernet/aquantia/atlantic/aq_pci_func.c
index 4c6c882c6a1c..ed5231dece3f 100644
--- a/drivers/net/ethernet/aquantia/atlantic/aq_pci_func.c
+++ b/drivers/net/ethernet/aquantia/atlantic/aq_pci_func.c
@@ -1,292 +1,486 @@
-/*
- * aQuantia Corporation Network Driver
- * Copyright (C) 2014-2017 aQuantia Corporation. All rights reserved
+// SPDX-License-Identifier: GPL-2.0-only
+/* Atlantic Network Driver
  *
- * This program is free software; you can redistribute it and/or modify it
- * under the terms and conditions of the GNU General Public License,
- * version 2, as published by the Free Software Foundation.
+ * Copyright (C) 2014-2019 aQuantia Corporation
+ * Copyright (C) 2019-2020 Marvell International Ltd.
  */
 
 /* File aq_pci_func.c: Definition of PCI functions. */
 
-#include "aq_pci_func.h"
+#include <linux/interrupt.h>
+#include <linux/module.h>
+
+#include "aq_main.h"
 #include "aq_nic.h"
 #include "aq_vec.h"
 #include "aq_hw.h"
-#include <linux/interrupt.h>
+#include "aq_pci_func.h"
+#include "hw_atl/hw_atl_a0.h"
+#include "hw_atl/hw_atl_b0.h"
+#include "hw_atl2/hw_atl2.h"
+#include "aq_filters.h"
+#include "aq_drvinfo.h"
+#include "aq_macsec.h"
+
+static const struct pci_device_id aq_pci_tbl[] = {
+	{ PCI_VDEVICE(AQUANTIA, AQ_DEVICE_ID_0001), },
+	{ PCI_VDEVICE(AQUANTIA, AQ_DEVICE_ID_D100), },
+	{ PCI_VDEVICE(AQUANTIA, AQ_DEVICE_ID_D107), },
+	{ PCI_VDEVICE(AQUANTIA, AQ_DEVICE_ID_D108), },
+	{ PCI_VDEVICE(AQUANTIA, AQ_DEVICE_ID_D109), },
+
+	{ PCI_VDEVICE(AQUANTIA, AQ_DEVICE_ID_AQC100), },
+	{ PCI_VDEVICE(AQUANTIA, AQ_DEVICE_ID_AQC107), },
+	{ PCI_VDEVICE(AQUANTIA, AQ_DEVICE_ID_AQC108), },
+	{ PCI_VDEVICE(AQUANTIA, AQ_DEVICE_ID_AQC109), },
+	{ PCI_VDEVICE(AQUANTIA, AQ_DEVICE_ID_AQC111), },
+	{ PCI_VDEVICE(AQUANTIA, AQ_DEVICE_ID_AQC112), },
+
+	{ PCI_VDEVICE(AQUANTIA, AQ_DEVICE_ID_AQC100S), },
+	{ PCI_VDEVICE(AQUANTIA, AQ_DEVICE_ID_AQC107S), },
+	{ PCI_VDEVICE(AQUANTIA, AQ_DEVICE_ID_AQC108S), },
+	{ PCI_VDEVICE(AQUANTIA, AQ_DEVICE_ID_AQC109S), },
+	{ PCI_VDEVICE(AQUANTIA, AQ_DEVICE_ID_AQC111S), },
+	{ PCI_VDEVICE(AQUANTIA, AQ_DEVICE_ID_AQC112S), },
+
+	{ PCI_VDEVICE(AQUANTIA, AQ_DEVICE_ID_AQC113DEV), },
+	{ PCI_VDEVICE(AQUANTIA, AQ_DEVICE_ID_AQC113CS), },
+	{ PCI_VDEVICE(AQUANTIA, AQ_DEVICE_ID_AQC114CS), },
+	{ PCI_VDEVICE(AQUANTIA, AQ_DEVICE_ID_AQC113), },
+	{ PCI_VDEVICE(AQUANTIA, AQ_DEVICE_ID_AQC113C), },
+	{ PCI_VDEVICE(AQUANTIA, AQ_DEVICE_ID_AQC115C), },
+	{ PCI_VDEVICE(AQUANTIA, AQ_DEVICE_ID_AQC113CA), },
+	{ PCI_VDEVICE(AQUANTIA, AQ_DEVICE_ID_AQC116C), },
+
+	{}
+};
+
+static const struct aq_board_revision_s hw_atl_boards[] = {
+	{ AQ_DEVICE_ID_0001,	AQ_HWREV_1,	&hw_atl_ops_a0, &hw_atl_a0_caps_aqc107, },
+	{ AQ_DEVICE_ID_D100,	AQ_HWREV_1,	&hw_atl_ops_a0, &hw_atl_a0_caps_aqc100, },
+	{ AQ_DEVICE_ID_D107,	AQ_HWREV_1,	&hw_atl_ops_a0, &hw_atl_a0_caps_aqc107, },
+	{ AQ_DEVICE_ID_D108,	AQ_HWREV_1,	&hw_atl_ops_a0, &hw_atl_a0_caps_aqc108, },
+	{ AQ_DEVICE_ID_D109,	AQ_HWREV_1,	&hw_atl_ops_a0, &hw_atl_a0_caps_aqc109, },
+
+	{ AQ_DEVICE_ID_0001,	AQ_HWREV_2,	&hw_atl_ops_b0, &hw_atl_b0_caps_aqc107, },
+	{ AQ_DEVICE_ID_D100,	AQ_HWREV_2,	&hw_atl_ops_b0, &hw_atl_b0_caps_aqc100, },
+	{ AQ_DEVICE_ID_D107,	AQ_HWREV_2,	&hw_atl_ops_b0, &hw_atl_b0_caps_aqc107, },
+	{ AQ_DEVICE_ID_D108,	AQ_HWREV_2,	&hw_atl_ops_b0, &hw_atl_b0_caps_aqc108, },
+	{ AQ_DEVICE_ID_D109,	AQ_HWREV_2,	&hw_atl_ops_b0, &hw_atl_b0_caps_aqc109, },
+
+	{ AQ_DEVICE_ID_AQC100,	AQ_HWREV_ANY,	&hw_atl_ops_b1, &hw_atl_b0_caps_aqc100, },
+	{ AQ_DEVICE_ID_AQC107,	AQ_HWREV_ANY,	&hw_atl_ops_b1, &hw_atl_b0_caps_aqc107, },
+	{ AQ_DEVICE_ID_AQC108,	AQ_HWREV_ANY,	&hw_atl_ops_b1, &hw_atl_b0_caps_aqc108, },
+	{ AQ_DEVICE_ID_AQC109,	AQ_HWREV_ANY,	&hw_atl_ops_b1, &hw_atl_b0_caps_aqc109, },
+	{ AQ_DEVICE_ID_AQC111,	AQ_HWREV_ANY,	&hw_atl_ops_b1, &hw_atl_b0_caps_aqc111, },
+	{ AQ_DEVICE_ID_AQC112,	AQ_HWREV_ANY,	&hw_atl_ops_b1, &hw_atl_b0_caps_aqc112, },
+
+	{ AQ_DEVICE_ID_AQC100S,	AQ_HWREV_ANY,	&hw_atl_ops_b1, &hw_atl_b0_caps_aqc100s, },
+	{ AQ_DEVICE_ID_AQC107S,	AQ_HWREV_ANY,	&hw_atl_ops_b1, &hw_atl_b0_caps_aqc107s, },
+	{ AQ_DEVICE_ID_AQC108S,	AQ_HWREV_ANY,	&hw_atl_ops_b1, &hw_atl_b0_caps_aqc108s, },
+	{ AQ_DEVICE_ID_AQC109S,	AQ_HWREV_ANY,	&hw_atl_ops_b1, &hw_atl_b0_caps_aqc109s, },
+	{ AQ_DEVICE_ID_AQC111S,	AQ_HWREV_ANY,	&hw_atl_ops_b1, &hw_atl_b0_caps_aqc111s, },
+	{ AQ_DEVICE_ID_AQC112S,	AQ_HWREV_ANY,	&hw_atl_ops_b1, &hw_atl_b0_caps_aqc112s, },
+
+	{ AQ_DEVICE_ID_AQC113DEV,	AQ_HWREV_ANY,	&hw_atl2_ops, &hw_atl2_caps_aqc113, },
+	{ AQ_DEVICE_ID_AQC113,		AQ_HWREV_ANY,	&hw_atl2_ops, &hw_atl2_caps_aqc113, },
+	{ AQ_DEVICE_ID_AQC113CS,	AQ_HWREV_ANY,	&hw_atl2_ops, &hw_atl2_caps_aqc113, },
+	{ AQ_DEVICE_ID_AQC114CS,	AQ_HWREV_ANY,	&hw_atl2_ops, &hw_atl2_caps_aqc113, },
+	{ AQ_DEVICE_ID_AQC113C,		AQ_HWREV_ANY,	&hw_atl2_ops, &hw_atl2_caps_aqc113, },
+	{ AQ_DEVICE_ID_AQC115C,		AQ_HWREV_ANY,	&hw_atl2_ops, &hw_atl2_caps_aqc115c, },
+	{ AQ_DEVICE_ID_AQC113CA,	AQ_HWREV_ANY,	&hw_atl2_ops, &hw_atl2_caps_aqc113, },
+	{ AQ_DEVICE_ID_AQC116C,		AQ_HWREV_ANY,	&hw_atl2_ops, &hw_atl2_caps_aqc116c, },
 
-struct aq_pci_func_s {
-	struct pci_dev *pdev;
-	struct aq_nic_s *port[AQ_CFG_PCI_FUNC_PORTS];
-	void __iomem *mmio;
-	void *aq_vec[AQ_CFG_PCI_FUNC_MSIX_IRQS];
-	resource_size_t mmio_pa;
-	unsigned int msix_entry_mask;
-	unsigned int ports;
-	bool is_pci_enabled;
-	bool is_regions;
-	bool is_pci_using_dac;
-	struct aq_hw_caps_s aq_hw_caps;
 };
 
-struct aq_pci_func_s *aq_pci_func_alloc(struct aq_hw_ops *aq_hw_ops,
-					struct pci_dev *pdev,
-					const struct net_device_ops *ndev_ops,
-					const struct ethtool_ops *eth_ops)
+MODULE_DEVICE_TABLE(pci, aq_pci_tbl);
+
+static int aq_pci_probe_get_hw_by_id(struct pci_dev *pdev,
+				     const struct aq_hw_ops **ops,
+				     const struct aq_hw_caps_s **caps)
 {
-	struct aq_pci_func_s *self = NULL;
-	int err = 0;
-	unsigned int port = 0U;
+	int i;
 
-	if (!aq_hw_ops) {
-		err = -EFAULT;
-		goto err_exit;
+	if (pdev->vendor != PCI_VENDOR_ID_AQUANTIA)
+		return -EINVAL;
+
+	for (i = 0; i < ARRAY_SIZE(hw_atl_boards); i++) {
+		if (hw_atl_boards[i].devid == pdev->device &&
+		    (hw_atl_boards[i].revision == AQ_HWREV_ANY ||
+		     hw_atl_boards[i].revision == pdev->revision)) {
+			*ops = hw_atl_boards[i].ops;
+			*caps = hw_atl_boards[i].caps;
+			break;
+		}
 	}
-	self = kzalloc(sizeof(*self), GFP_KERNEL);
-	if (!self) {
-		err = -ENOMEM;
+
+	if (i == ARRAY_SIZE(hw_atl_boards))
+		return -EINVAL;
+
+	return 0;
+}
+
+static int aq_pci_func_init(struct pci_dev *pdev)
+{
+	int err;
+
+	err = dma_set_mask_and_coherent(&pdev->dev, DMA_BIT_MASK(64));
+	if (err)
+		err = dma_set_mask_and_coherent(&pdev->dev, DMA_BIT_MASK(32));
+	if (err) {
+		err = -ENOSR;
 		goto err_exit;
 	}
 
-	pci_set_drvdata(pdev, self);
-	self->pdev = pdev;
-
-	err = aq_hw_ops->get_hw_caps(NULL, &self->aq_hw_caps);
+	err = pci_request_regions(pdev, AQ_CFG_DRV_NAME "_mmio");
 	if (err < 0)
 		goto err_exit;
 
-	self->ports = self->aq_hw_caps.ports;
+	pci_set_master(pdev);
 
-	for (port = 0; port < self->ports; ++port) {
-		struct aq_nic_s *aq_nic = aq_nic_alloc_cold(ndev_ops, eth_ops,
-							    &pdev->dev, self,
-							    port, aq_hw_ops);
+	return 0;
 
-		if (!aq_nic) {
-			err = -ENOMEM;
-			goto err_exit;
-		}
-		self->port[port] = aq_nic;
+err_exit:
+	return err;
+}
+
+int aq_pci_func_alloc_irq(struct aq_nic_s *self, unsigned int i,
+			  char *name, irq_handler_t irq_handler,
+			  void *irq_arg, cpumask_t *affinity_mask)
+{
+	struct pci_dev *pdev = self->pdev;
+	int err;
+
+	if (pdev->msix_enabled || pdev->msi_enabled)
+		err = request_irq(pci_irq_vector(pdev, i), irq_handler, 0,
+				  name, irq_arg);
+	else
+		err = request_irq(pci_irq_vector(pdev, i), aq_vec_isr_legacy,
+				  IRQF_SHARED, name, irq_arg);
+
+	if (err >= 0) {
+		self->msix_entry_mask |= (1 << i);
+
+		if (pdev->msix_enabled && affinity_mask)
+			irq_update_affinity_hint(pci_irq_vector(pdev, i),
+						 affinity_mask);
 	}
 
-err_exit:
-	if (err < 0) {
-		if (self)
-			aq_pci_func_free(self);
-		self = NULL;
+	return err;
+}
+
+void aq_pci_func_free_irqs(struct aq_nic_s *self)
+{
+	struct pci_dev *pdev = self->pdev;
+	unsigned int i;
+	void *irq_data;
+
+	for (i = 32U; i--;) {
+		if (!((1U << i) & self->msix_entry_mask))
+			continue;
+		if (self->aq_nic_cfg.link_irq_vec &&
+		    i == self->aq_nic_cfg.link_irq_vec)
+			irq_data = self;
+		else if (i < AQ_CFG_VECS_MAX)
+			irq_data = self->aq_vec[i];
+		else
+			continue;
+
+		if (pdev->msix_enabled)
+			irq_update_affinity_hint(pci_irq_vector(pdev, i), NULL);
+		free_irq(pci_irq_vector(pdev, i), irq_data);
+		self->msix_entry_mask &= ~(1U << i);
 	}
+}
+
+unsigned int aq_pci_func_get_irq_type(struct aq_nic_s *self)
+{
+	if (self->pdev->msix_enabled)
+		return AQ_HW_IRQ_MSIX;
+	if (self->pdev->msi_enabled)
+		return AQ_HW_IRQ_MSI;
 
-	(void)err;
-	return self;
+	return AQ_HW_IRQ_INTX;
 }
 
-int aq_pci_func_init(struct aq_pci_func_s *self)
+static void aq_pci_free_irq_vectors(struct aq_nic_s *self)
 {
-	int err = 0;
-	unsigned int bar = 0U;
-	unsigned int port = 0U;
+	pci_free_irq_vectors(self->pdev);
+}
 
-	err = pci_enable_device(self->pdev);
-	if (err < 0)
-		goto err_exit;
+static int aq_pci_probe(struct pci_dev *pdev,
+			const struct pci_device_id *pci_id)
+{
+	struct net_device *ndev;
+	resource_size_t mmio_pa;
+	struct aq_nic_s *self;
+	u32 numvecs;
+	u32 bar;
+	int err;
 
-	self->is_pci_enabled = true;
+	err = pci_enable_device(pdev);
+	if (err)
+		return err;
 
-	err = pci_set_dma_mask(self->pdev, DMA_BIT_MASK(64));
-	if (!err) {
-		err = pci_set_consistent_dma_mask(self->pdev, DMA_BIT_MASK(64));
-		self->is_pci_using_dac = 1;
-	}
-	if (err) {
-		err = pci_set_dma_mask(self->pdev, DMA_BIT_MASK(32));
-		if (!err)
-			err = pci_set_consistent_dma_mask(self->pdev,
-							  DMA_BIT_MASK(32));
-		self->is_pci_using_dac = 0;
-	}
-	if (err != 0) {
-		err = -ENOSR;
-		goto err_exit;
+	err = aq_pci_func_init(pdev);
+	if (err)
+		goto err_pci_func;
+
+	ndev = aq_ndev_alloc();
+	if (!ndev) {
+		err = -ENOMEM;
+		goto err_ndev;
 	}
 
-	err = pci_request_regions(self->pdev, AQ_CFG_DRV_NAME "_mmio");
-	if (err < 0)
-		goto err_exit;
+	self = netdev_priv(ndev);
+	self->pdev = pdev;
+	SET_NETDEV_DEV(ndev, &pdev->dev);
+	pci_set_drvdata(pdev, self);
+
+	mutex_init(&self->fwreq_mutex);
+
+	err = aq_pci_probe_get_hw_by_id(pdev, &self->aq_hw_ops,
+					&aq_nic_get_cfg(self)->aq_hw_caps);
+	if (err)
+		goto err_ioremap;
 
-	self->is_regions = true;
+	self->aq_hw = kzalloc(sizeof(*self->aq_hw), GFP_KERNEL);
+	if (!self->aq_hw) {
+		err = -ENOMEM;
+		goto err_ioremap;
+	}
+	self->aq_hw->aq_nic_cfg = aq_nic_get_cfg(self);
+	if (self->aq_hw->aq_nic_cfg->aq_hw_caps->priv_data_len) {
+		int len = self->aq_hw->aq_nic_cfg->aq_hw_caps->priv_data_len;
 
-	pci_set_master(self->pdev);
+		self->aq_hw->priv = kzalloc(len, GFP_KERNEL);
+		if (!self->aq_hw->priv) {
+			err = -ENOMEM;
+			goto err_free_aq_hw;
+		}
+	}
 
 	for (bar = 0; bar < 4; ++bar) {
-		if (IORESOURCE_MEM & pci_resource_flags(self->pdev, bar)) {
+		if (IORESOURCE_MEM & pci_resource_flags(pdev, bar)) {
 			resource_size_t reg_sz;
 
-			self->mmio_pa = pci_resource_start(self->pdev, bar);
-			if (self->mmio_pa == 0U) {
+			mmio_pa = pci_resource_start(pdev, bar);
+			if (mmio_pa == 0U) {
 				err = -EIO;
-				goto err_exit;
+				goto err_free_aq_hw_priv;
 			}
 
-			reg_sz = pci_resource_len(self->pdev, bar);
+			reg_sz = pci_resource_len(pdev, bar);
 			if ((reg_sz <= 24 /*ATL_REGS_SIZE*/)) {
 				err = -EIO;
-				goto err_exit;
+				goto err_free_aq_hw_priv;
 			}
 
-			self->mmio = ioremap_nocache(self->mmio_pa, reg_sz);
-			if (!self->mmio) {
+			self->aq_hw->mmio = ioremap(mmio_pa, reg_sz);
+			if (!self->aq_hw->mmio) {
 				err = -EIO;
-				goto err_exit;
+				goto err_free_aq_hw_priv;
 			}
 			break;
 		}
 	}
 
-	/*enable interrupts */
-#if !AQ_CFG_FORCE_LEGACY_INT
-	err = pci_alloc_irq_vectors(self->pdev, self->aq_hw_caps.msix_irqs,
-			      self->aq_hw_caps.msix_irqs, PCI_IRQ_MSIX);
-
-	if (err < 0) {
-		err = pci_alloc_irq_vectors(self->pdev, 1, 1,
-				PCI_IRQ_MSI | PCI_IRQ_LEGACY);
-		if (err < 0)
-			goto err_exit;
+	if (bar == 4) {
+		err = -EIO;
+		goto err_free_aq_hw_priv;
 	}
-#endif
 
-	/* net device init */
-	for (port = 0; port < self->ports; ++port) {
-		if (!self->port[port])
-			continue;
+	numvecs = min((u8)AQ_CFG_VECS_DEF,
+		      aq_nic_get_cfg(self)->aq_hw_caps->msix_irqs);
+	numvecs = min(numvecs, num_online_cpus());
+	/* Request IRQ vector for PTP */
+	numvecs += 1;
 
-		err = aq_nic_cfg_start(self->port[port]);
-		if (err < 0)
-			goto err_exit;
+	numvecs += AQ_HW_SERVICE_IRQS;
+	/*enable interrupts */
+#if !AQ_CFG_FORCE_INTX
+	err = pci_alloc_irq_vectors(self->pdev, 1, numvecs, PCI_IRQ_ALL_TYPES);
+	if (err < 0)
+		goto err_hwinit;
+	numvecs = err;
+#endif
+	self->irqvecs = numvecs;
 
-		err = aq_nic_ndev_init(self->port[port]);
-		if (err < 0)
-			goto err_exit;
+	/* net device init */
+	aq_nic_cfg_start(self);
 
-		err = aq_nic_ndev_register(self->port[port]);
-		if (err < 0)
-			goto err_exit;
-	}
+	aq_nic_ndev_init(self);
 
-err_exit:
+	err = aq_nic_ndev_register(self);
 	if (err < 0)
-		aq_pci_func_deinit(self);
+		goto err_register;
+
+	aq_drvinfo_init(ndev);
+
+	return 0;
+
+err_register:
+	aq_nic_free_vectors(self);
+	aq_pci_free_irq_vectors(self);
+err_hwinit:
+	iounmap(self->aq_hw->mmio);
+err_free_aq_hw_priv:
+	kfree(self->aq_hw->priv);
+err_free_aq_hw:
+	kfree(self->aq_hw);
+err_ioremap:
+	free_netdev(ndev);
+err_ndev:
+	pci_release_regions(pdev);
+err_pci_func:
+	pci_disable_device(pdev);
+
 	return err;
 }
 
-int aq_pci_func_alloc_irq(struct aq_pci_func_s *self, unsigned int i,
-			  char *name, void *aq_vec, cpumask_t *affinity_mask)
+static void aq_pci_remove(struct pci_dev *pdev)
 {
-	struct pci_dev *pdev = self->pdev;
-	int err = 0;
+	struct aq_nic_s *self = pci_get_drvdata(pdev);
 
-	if (pdev->msix_enabled || pdev->msi_enabled)
-		err = request_irq(pci_irq_vector(pdev, i), aq_vec_isr, 0,
-				  name, aq_vec);
-	else
-		err = request_irq(pci_irq_vector(pdev, i), aq_vec_isr_legacy,
-				  IRQF_SHARED, name, aq_vec);
-
-	if (err >= 0) {
-		self->msix_entry_mask |= (1 << i);
-		self->aq_vec[i] = aq_vec;
+	if (self->ndev) {
+		aq_clear_rxnfc_all_rules(self);
+		if (self->ndev->reg_state == NETREG_REGISTERED)
+			unregister_netdev(self->ndev);
 
-		if (pdev->msix_enabled)
-			irq_set_affinity_hint(pci_irq_vector(pdev, i),
-					      affinity_mask);
+#if IS_ENABLED(CONFIG_MACSEC)
+		aq_macsec_free(self);
+#endif
+		aq_nic_free_vectors(self);
+		aq_pci_free_irq_vectors(self);
+		iounmap(self->aq_hw->mmio);
+		kfree(self->aq_hw->priv);
+		kfree(self->aq_hw);
+		pci_release_regions(pdev);
+		free_netdev(self->ndev);
 	}
 
-	return err;
+	pci_disable_device(pdev);
 }
 
-void aq_pci_func_free_irqs(struct aq_pci_func_s *self)
+static void aq_pci_shutdown(struct pci_dev *pdev)
 {
-	struct pci_dev *pdev = self->pdev;
-	unsigned int i = 0U;
+	struct aq_nic_s *self = pci_get_drvdata(pdev);
 
-	for (i = 32U; i--;) {
-		if (!((1U << i) & self->msix_entry_mask))
-			continue;
+	aq_nic_shutdown(self);
 
-		if (pdev->msix_enabled)
-			irq_set_affinity_hint(pci_irq_vector(pdev, i), NULL);
-		free_irq(pci_irq_vector(pdev, i), self->aq_vec[i]);
-		self->msix_entry_mask &= ~(1U << i);
+	pci_disable_device(pdev);
+
+	if (system_state == SYSTEM_POWER_OFF) {
+		pci_wake_from_d3(pdev, false);
+		pci_set_power_state(pdev, PCI_D3hot);
 	}
 }
 
-void __iomem *aq_pci_func_get_mmio(struct aq_pci_func_s *self)
+#ifdef CONFIG_PM
+static int aq_suspend_common(struct device *dev)
 {
-	return self->mmio;
-}
+	struct aq_nic_s *nic = pci_get_drvdata(to_pci_dev(dev));
 
-unsigned int aq_pci_func_get_irq_type(struct aq_pci_func_s *self)
-{
-	if (self->pdev->msix_enabled)
-		return AQ_HW_IRQ_MSIX;
-	if (self->pdev->msi_enabled)
-		return AQ_HW_IRQ_MSIX;
-	return AQ_HW_IRQ_LEGACY;
-}
+	rtnl_lock();
 
-void aq_pci_func_deinit(struct aq_pci_func_s *self)
-{
-	if (!self)
-		goto err_exit;
+	nic->power_state = AQ_HW_POWER_STATE_D3;
+	netif_device_detach(nic->ndev);
+	netif_tx_stop_all_queues(nic->ndev);
 
-	aq_pci_func_free_irqs(self);
-	pci_free_irq_vectors(self->pdev);
+	if (netif_running(nic->ndev))
+		aq_nic_stop(nic);
 
-	if (self->is_regions)
-		pci_release_regions(self->pdev);
+	aq_nic_deinit(nic, !nic->aq_hw->aq_nic_cfg->wol);
+	aq_nic_set_power(nic);
 
-	if (self->is_pci_enabled)
-		pci_disable_device(self->pdev);
+	rtnl_unlock();
 
-err_exit:;
+	return 0;
 }
 
-void aq_pci_func_free(struct aq_pci_func_s *self)
+static int atl_resume_common(struct device *dev)
 {
-	unsigned int port = 0U;
+	struct pci_dev *pdev = to_pci_dev(dev);
+	struct aq_nic_s *nic;
+	int ret = 0;
 
-	if (!self)
-		goto err_exit;
+	nic = pci_get_drvdata(pdev);
 
-	for (port = 0; port < self->ports; ++port) {
-		if (!self->port[port])
-			continue;
+	rtnl_lock();
+
+	pci_set_power_state(pdev, PCI_D0);
+	pci_restore_state(pdev);
+
+	if (netif_running(nic->ndev)) {
+		ret = aq_nic_init(nic);
+		if (ret)
+			goto err_exit;
 
-		aq_nic_ndev_free(self->port[port]);
+		ret = aq_nic_start(nic);
+		if (ret)
+			goto err_exit;
 	}
 
-	kfree(self);
+	netif_device_attach(nic->ndev);
+	netif_tx_start_all_queues(nic->ndev);
+
+err_exit:
+	if (ret < 0)
+		aq_nic_deinit(nic, true);
+
+	rtnl_unlock();
 
-err_exit:;
+	return ret;
 }
 
-int aq_pci_func_change_pm_state(struct aq_pci_func_s *self,
-				pm_message_t *pm_msg)
+static int aq_pm_freeze(struct device *dev)
 {
-	int err = 0;
-	unsigned int port = 0U;
+	return aq_suspend_common(dev);
+}
 
-	if (!self) {
-		err = -EFAULT;
-		goto err_exit;
-	}
-	for (port = 0; port < self->ports; ++port) {
-		if (!self->port[port])
-			continue;
+static int aq_pm_suspend_poweroff(struct device *dev)
+{
+	return aq_suspend_common(dev);
+}
 
-		(void)aq_nic_change_pm_state(self->port[port], pm_msg);
-	}
+static int aq_pm_thaw(struct device *dev)
+{
+	return atl_resume_common(dev);
+}
 
-err_exit:
-	return err;
+static int aq_pm_resume_restore(struct device *dev)
+{
+	return atl_resume_common(dev);
+}
+
+static const struct dev_pm_ops aq_pm_ops = {
+	.suspend = aq_pm_suspend_poweroff,
+	.poweroff = aq_pm_suspend_poweroff,
+	.freeze = aq_pm_freeze,
+	.resume = aq_pm_resume_restore,
+	.restore = aq_pm_resume_restore,
+	.thaw = aq_pm_thaw,
+};
+#endif
+
+static struct pci_driver aq_pci_driver = {
+	.name = AQ_CFG_DRV_NAME,
+	.id_table = aq_pci_tbl,
+	.probe = aq_pci_probe,
+	.remove = aq_pci_remove,
+	.shutdown = aq_pci_shutdown,
+#ifdef CONFIG_PM
+	.driver.pm = &aq_pm_ops,
+#endif
+};
+
+int aq_pci_func_register_driver(void)
+{
+	return pci_register_driver(&aq_pci_driver);
 }
+
+void aq_pci_func_unregister_driver(void)
+{
+	pci_unregister_driver(&aq_pci_driver);
+}
+
diff --git a/drivers/net/ethernet/aquantia/atlantic/aq_pci_func.h b/drivers/net/ethernet/aquantia/atlantic/aq_pci_func.h
index ecb033791203..3fa5f7a73680 100644
--- a/drivers/net/ethernet/aquantia/atlantic/aq_pci_func.h
+++ b/drivers/net/ethernet/aquantia/atlantic/aq_pci_func.h
@@ -1,10 +1,8 @@
-/*
- * aQuantia Corporation Network Driver
- * Copyright (C) 2014-2017 aQuantia Corporation. All rights reserved
+/* SPDX-License-Identifier: GPL-2.0-only */
+/* Atlantic Network Driver
  *
- * This program is free software; you can redistribute it and/or modify it
- * under the terms and conditions of the GNU General Public License,
- * version 2, as published by the Free Software Foundation.
+ * Copyright (C) 2014-2019 aQuantia Corporation
+ * Copyright (C) 2019-2020 Marvell International Ltd.
  */
 
 /* File aq_pci_func.h: Declaration of PCI functions. */
@@ -13,22 +11,22 @@
 #define AQ_PCI_FUNC_H
 
 #include "aq_common.h"
+#include "aq_nic.h"
 
-struct aq_pci_func_s *aq_pci_func_alloc(struct aq_hw_ops *hw_ops,
-					struct pci_dev *pdev,
-					const struct net_device_ops *ndev_ops,
-					const struct ethtool_ops *eth_ops);
-int aq_pci_func_init(struct aq_pci_func_s *self);
-int aq_pci_func_alloc_irq(struct aq_pci_func_s *self, unsigned int i,
-			  char *name, void *aq_vec,
-			  cpumask_t *affinity_mask);
-void aq_pci_func_free_irqs(struct aq_pci_func_s *self);
-int aq_pci_func_start(struct aq_pci_func_s *self);
-void __iomem *aq_pci_func_get_mmio(struct aq_pci_func_s *self);
-unsigned int aq_pci_func_get_irq_type(struct aq_pci_func_s *self);
-void aq_pci_func_deinit(struct aq_pci_func_s *self);
-void aq_pci_func_free(struct aq_pci_func_s *self);
-int aq_pci_func_change_pm_state(struct aq_pci_func_s *self,
-				pm_message_t *pm_msg);
+struct aq_board_revision_s {
+	unsigned short devid;
+	unsigned short revision;
+	const struct aq_hw_ops *ops;
+	const struct aq_hw_caps_s *caps;
+};
+
+int aq_pci_func_alloc_irq(struct aq_nic_s *self, unsigned int i,
+			  char *name, irq_handler_t irq_handler,
+			  void *irq_arg, cpumask_t *affinity_mask);
+void aq_pci_func_free_irqs(struct aq_nic_s *self);
+unsigned int aq_pci_func_get_irq_type(struct aq_nic_s *self);
+
+int aq_pci_func_register_driver(void);
+void aq_pci_func_unregister_driver(void);
 
 #endif /* AQ_PCI_FUNC_H */
diff --git a/drivers/net/ethernet/aquantia/atlantic/aq_phy.c b/drivers/net/ethernet/aquantia/atlantic/aq_phy.c
new file mode 100644
index 000000000000..949ac2351701
--- /dev/null
+++ b/drivers/net/ethernet/aquantia/atlantic/aq_phy.c
@@ -0,0 +1,172 @@
+// SPDX-License-Identifier: GPL-2.0-only
+/* Atlantic Network Driver
+ *
+ * Copyright (C) 2018-2019 aQuantia Corporation
+ * Copyright (C) 2019-2020 Marvell International Ltd.
+ */
+
+#include "aq_phy.h"
+
+#define HW_ATL_PTP_DISABLE_MSK	BIT(10)
+
+bool aq_mdio_busy_wait(struct aq_hw_s *aq_hw)
+{
+	int err = 0;
+	u32 val;
+
+	err = readx_poll_timeout_atomic(hw_atl_mdio_busy_get, aq_hw,
+					val, val == 0U, 10U, 100000U);
+
+	if (err < 0)
+		return false;
+
+	return true;
+}
+
+u16 aq_mdio_read_word(struct aq_hw_s *aq_hw, u16 mmd, u16 addr)
+{
+	u16 phy_addr = aq_hw->phy_id << 5 | mmd;
+
+	/* Set Address register. */
+	hw_atl_glb_mdio_iface4_set(aq_hw, (addr & HW_ATL_MDIO_ADDRESS_MSK) <<
+				   HW_ATL_MDIO_ADDRESS_SHIFT);
+	/* Send Address command. */
+	hw_atl_glb_mdio_iface2_set(aq_hw, HW_ATL_MDIO_EXECUTE_OPERATION_MSK |
+				   (3 << HW_ATL_MDIO_OP_MODE_SHIFT) |
+				   ((phy_addr & HW_ATL_MDIO_PHY_ADDRESS_MSK) <<
+				    HW_ATL_MDIO_PHY_ADDRESS_SHIFT));
+
+	aq_mdio_busy_wait(aq_hw);
+
+	/* Send Read command. */
+	hw_atl_glb_mdio_iface2_set(aq_hw, HW_ATL_MDIO_EXECUTE_OPERATION_MSK |
+				   (1 << HW_ATL_MDIO_OP_MODE_SHIFT) |
+				   ((phy_addr & HW_ATL_MDIO_PHY_ADDRESS_MSK) <<
+				    HW_ATL_MDIO_PHY_ADDRESS_SHIFT));
+	/* Read result. */
+	aq_mdio_busy_wait(aq_hw);
+
+	return (u16)hw_atl_glb_mdio_iface5_get(aq_hw);
+}
+
+void aq_mdio_write_word(struct aq_hw_s *aq_hw, u16 mmd, u16 addr, u16 data)
+{
+	u16 phy_addr = aq_hw->phy_id << 5 | mmd;
+
+	/* Set Address register. */
+	hw_atl_glb_mdio_iface4_set(aq_hw, (addr & HW_ATL_MDIO_ADDRESS_MSK) <<
+				   HW_ATL_MDIO_ADDRESS_SHIFT);
+	/* Send Address command. */
+	hw_atl_glb_mdio_iface2_set(aq_hw, HW_ATL_MDIO_EXECUTE_OPERATION_MSK |
+				   (3 << HW_ATL_MDIO_OP_MODE_SHIFT) |
+				   ((phy_addr & HW_ATL_MDIO_PHY_ADDRESS_MSK) <<
+				    HW_ATL_MDIO_PHY_ADDRESS_SHIFT));
+
+	aq_mdio_busy_wait(aq_hw);
+
+	hw_atl_glb_mdio_iface3_set(aq_hw, (data & HW_ATL_MDIO_WRITE_DATA_MSK) <<
+				   HW_ATL_MDIO_WRITE_DATA_SHIFT);
+	/* Send Write command. */
+	hw_atl_glb_mdio_iface2_set(aq_hw, HW_ATL_MDIO_EXECUTE_OPERATION_MSK |
+				   (2 << HW_ATL_MDIO_OP_MODE_SHIFT) |
+				   ((phy_addr & HW_ATL_MDIO_PHY_ADDRESS_MSK) <<
+				    HW_ATL_MDIO_PHY_ADDRESS_SHIFT));
+
+	aq_mdio_busy_wait(aq_hw);
+}
+
+u16 aq_phy_read_reg(struct aq_hw_s *aq_hw, u16 mmd, u16 address)
+{
+	int err = 0;
+	u32 val;
+
+	err = readx_poll_timeout_atomic(hw_atl_sem_mdio_get, aq_hw,
+					val, val == 1U, 10U, 100000U);
+
+	if (err < 0) {
+		err = 0xffff;
+		goto err_exit;
+	}
+
+	err = aq_mdio_read_word(aq_hw, mmd, address);
+
+	hw_atl_reg_glb_cpu_sem_set(aq_hw, 1U, HW_ATL_FW_SM_MDIO);
+
+err_exit:
+	return err;
+}
+
+void aq_phy_write_reg(struct aq_hw_s *aq_hw, u16 mmd, u16 address, u16 data)
+{
+	int err = 0;
+	u32 val;
+
+	err = readx_poll_timeout_atomic(hw_atl_sem_mdio_get, aq_hw,
+					val, val == 1U, 10U, 100000U);
+	if (err < 0)
+		return;
+
+	aq_mdio_write_word(aq_hw, mmd, address, data);
+	hw_atl_reg_glb_cpu_sem_set(aq_hw, 1U, HW_ATL_FW_SM_MDIO);
+}
+
+bool aq_phy_init_phy_id(struct aq_hw_s *aq_hw)
+{
+	u16 val;
+
+	for (aq_hw->phy_id = 0; aq_hw->phy_id < HW_ATL_PHY_ID_MAX;
+	     ++aq_hw->phy_id) {
+		/* PMA Standard Device Identifier 2: Address 1.3 */
+		val = aq_phy_read_reg(aq_hw, MDIO_MMD_PMAPMD, 3);
+
+		if (val != 0xffff)
+			return true;
+	}
+
+	return false;
+}
+
+bool aq_phy_init(struct aq_hw_s *aq_hw)
+{
+	u32 dev_id;
+
+	if (aq_hw->phy_id == HW_ATL_PHY_ID_MAX)
+		if (!aq_phy_init_phy_id(aq_hw))
+			return false;
+
+	/* PMA Standard Device Identifier:
+	 * Address 1.2 = MSW,
+	 * Address 1.3 = LSW
+	 */
+	dev_id = aq_phy_read_reg(aq_hw, MDIO_MMD_PMAPMD, 2);
+	dev_id <<= 16;
+	dev_id |= aq_phy_read_reg(aq_hw, MDIO_MMD_PMAPMD, 3);
+
+	if (dev_id == 0xffffffff) {
+		aq_hw->phy_id = HW_ATL_PHY_ID_MAX;
+		return false;
+	}
+
+	return true;
+}
+
+void aq_phy_disable_ptp(struct aq_hw_s *aq_hw)
+{
+	static const u16 ptp_registers[] = {
+		0x031e,
+		0x031d,
+		0x031c,
+		0x031b,
+	};
+	u16 val;
+	int i;
+
+	for (i = 0; i < ARRAY_SIZE(ptp_registers); i++) {
+		val = aq_phy_read_reg(aq_hw, MDIO_MMD_VEND1,
+				      ptp_registers[i]);
+
+		aq_phy_write_reg(aq_hw, MDIO_MMD_VEND1,
+				 ptp_registers[i],
+				 val & ~HW_ATL_PTP_DISABLE_MSK);
+	}
+}
diff --git a/drivers/net/ethernet/aquantia/atlantic/aq_phy.h b/drivers/net/ethernet/aquantia/atlantic/aq_phy.h
new file mode 100644
index 000000000000..86cc1ee836e2
--- /dev/null
+++ b/drivers/net/ethernet/aquantia/atlantic/aq_phy.h
@@ -0,0 +1,36 @@
+/* SPDX-License-Identifier: GPL-2.0-only */
+/* Atlantic Network Driver
+ *
+ * Copyright (C) 2018-2019 aQuantia Corporation
+ * Copyright (C) 2019-2020 Marvell International Ltd.
+ */
+
+#ifndef AQ_PHY_H
+#define AQ_PHY_H
+
+#include <linux/mdio.h>
+
+#include "hw_atl/hw_atl_llh.h"
+#include "hw_atl/hw_atl_llh_internal.h"
+#include "aq_hw_utils.h"
+#include "aq_hw.h"
+
+#define HW_ATL_PHY_ID_MAX 32U
+
+bool aq_mdio_busy_wait(struct aq_hw_s *aq_hw);
+
+u16 aq_mdio_read_word(struct aq_hw_s *aq_hw, u16 mmd, u16 addr);
+
+void aq_mdio_write_word(struct aq_hw_s *aq_hw, u16 mmd, u16 addr, u16 data);
+
+u16 aq_phy_read_reg(struct aq_hw_s *aq_hw, u16 mmd, u16 address);
+
+void aq_phy_write_reg(struct aq_hw_s *aq_hw, u16 mmd, u16 address, u16 data);
+
+bool aq_phy_init_phy_id(struct aq_hw_s *aq_hw);
+
+bool aq_phy_init(struct aq_hw_s *aq_hw);
+
+void aq_phy_disable_ptp(struct aq_hw_s *aq_hw);
+
+#endif /* AQ_PHY_H */
diff --git a/drivers/net/ethernet/aquantia/atlantic/aq_ptp.c b/drivers/net/ethernet/aquantia/atlantic/aq_ptp.c
new file mode 100644
index 000000000000..5acb3e16b567
--- /dev/null
+++ b/drivers/net/ethernet/aquantia/atlantic/aq_ptp.c
@@ -0,0 +1,1415 @@
+// SPDX-License-Identifier: GPL-2.0-only
+/* Atlantic Network Driver
+ *
+ * Copyright (C) 2014-2019 aQuantia Corporation
+ * Copyright (C) 2019-2020 Marvell International Ltd.
+ */
+
+/* File aq_ptp.c:
+ * Definition of functions for Linux PTP support.
+ */
+
+#include <linux/ptp_clock_kernel.h>
+#include <linux/ptp_classify.h>
+#include <linux/interrupt.h>
+#include <linux/clocksource.h>
+
+#include "aq_nic.h"
+#include "aq_ptp.h"
+#include "aq_ring.h"
+#include "aq_phy.h"
+#include "aq_filters.h"
+
+#if IS_REACHABLE(CONFIG_PTP_1588_CLOCK)
+
+#define AQ_PTP_TX_TIMEOUT        (HZ *  10)
+
+#define POLL_SYNC_TIMER_MS 15
+
+enum ptp_speed_offsets {
+	ptp_offset_idx_10 = 0,
+	ptp_offset_idx_100,
+	ptp_offset_idx_1000,
+	ptp_offset_idx_2500,
+	ptp_offset_idx_5000,
+	ptp_offset_idx_10000,
+};
+
+struct ptp_skb_ring {
+	struct sk_buff **buff;
+	spinlock_t lock;
+	unsigned int size;
+	unsigned int head;
+	unsigned int tail;
+};
+
+struct ptp_tx_timeout {
+	spinlock_t lock;
+	bool active;
+	unsigned long tx_start;
+};
+
+struct aq_ptp_s {
+	struct aq_nic_s *aq_nic;
+	struct hwtstamp_config hwtstamp_config;
+	spinlock_t ptp_lock;
+	spinlock_t ptp_ring_lock;
+	struct ptp_clock *ptp_clock;
+	struct ptp_clock_info ptp_info;
+
+	atomic_t offset_egress;
+	atomic_t offset_ingress;
+
+	struct aq_ring_param_s ptp_ring_param;
+
+	struct ptp_tx_timeout ptp_tx_timeout;
+
+	unsigned int idx_vector;
+	struct napi_struct napi;
+
+	struct aq_ring_s ptp_tx;
+	struct aq_ring_s ptp_rx;
+	struct aq_ring_s hwts_rx;
+
+	struct ptp_skb_ring skb_ring;
+
+	struct aq_rx_filter_l3l4 udp_filter;
+	struct aq_rx_filter_l2 eth_type_filter;
+
+	struct delayed_work poll_sync;
+	u32 poll_timeout_ms;
+
+	bool extts_pin_enabled;
+	u64 last_sync1588_ts;
+
+	bool a1_ptp;
+};
+
+struct ptp_tm_offset {
+	unsigned int mbps;
+	int egress;
+	int ingress;
+};
+
+static struct ptp_tm_offset ptp_offset[6];
+
+void aq_ptp_tm_offset_set(struct aq_nic_s *aq_nic, unsigned int mbps)
+{
+	struct aq_ptp_s *aq_ptp = aq_nic->aq_ptp;
+	int i, egress, ingress;
+
+	if (!aq_ptp)
+		return;
+
+	egress = 0;
+	ingress = 0;
+
+	for (i = 0; i < ARRAY_SIZE(ptp_offset); i++) {
+		if (mbps == ptp_offset[i].mbps) {
+			egress = ptp_offset[i].egress;
+			ingress = ptp_offset[i].ingress;
+			break;
+		}
+	}
+
+	atomic_set(&aq_ptp->offset_egress, egress);
+	atomic_set(&aq_ptp->offset_ingress, ingress);
+}
+
+static int __aq_ptp_skb_put(struct ptp_skb_ring *ring, struct sk_buff *skb)
+{
+	unsigned int next_head = (ring->head + 1) % ring->size;
+
+	if (next_head == ring->tail)
+		return -ENOMEM;
+
+	ring->buff[ring->head] = skb_get(skb);
+	ring->head = next_head;
+
+	return 0;
+}
+
+static int aq_ptp_skb_put(struct ptp_skb_ring *ring, struct sk_buff *skb)
+{
+	unsigned long flags;
+	int ret;
+
+	spin_lock_irqsave(&ring->lock, flags);
+	ret = __aq_ptp_skb_put(ring, skb);
+	spin_unlock_irqrestore(&ring->lock, flags);
+
+	return ret;
+}
+
+static struct sk_buff *__aq_ptp_skb_get(struct ptp_skb_ring *ring)
+{
+	struct sk_buff *skb;
+
+	if (ring->tail == ring->head)
+		return NULL;
+
+	skb = ring->buff[ring->tail];
+	ring->tail = (ring->tail + 1) % ring->size;
+
+	return skb;
+}
+
+static struct sk_buff *aq_ptp_skb_get(struct ptp_skb_ring *ring)
+{
+	unsigned long flags;
+	struct sk_buff *skb;
+
+	spin_lock_irqsave(&ring->lock, flags);
+	skb = __aq_ptp_skb_get(ring);
+	spin_unlock_irqrestore(&ring->lock, flags);
+
+	return skb;
+}
+
+static unsigned int aq_ptp_skb_buf_len(struct ptp_skb_ring *ring)
+{
+	unsigned long flags;
+	unsigned int len;
+
+	spin_lock_irqsave(&ring->lock, flags);
+	len = (ring->head >= ring->tail) ?
+	ring->head - ring->tail :
+	ring->size - ring->tail + ring->head;
+	spin_unlock_irqrestore(&ring->lock, flags);
+
+	return len;
+}
+
+static int aq_ptp_skb_ring_init(struct ptp_skb_ring *ring, unsigned int size)
+{
+	struct sk_buff **buff = kmalloc(sizeof(*buff) * size, GFP_KERNEL);
+
+	if (!buff)
+		return -ENOMEM;
+
+	spin_lock_init(&ring->lock);
+
+	ring->buff = buff;
+	ring->size = size;
+	ring->head = 0;
+	ring->tail = 0;
+
+	return 0;
+}
+
+static void aq_ptp_skb_ring_clean(struct ptp_skb_ring *ring)
+{
+	struct sk_buff *skb;
+
+	while ((skb = aq_ptp_skb_get(ring)) != NULL)
+		dev_kfree_skb_any(skb);
+}
+
+static void aq_ptp_skb_ring_release(struct ptp_skb_ring *ring)
+{
+	if (ring->buff) {
+		aq_ptp_skb_ring_clean(ring);
+		kfree(ring->buff);
+		ring->buff = NULL;
+	}
+}
+
+static void aq_ptp_tx_timeout_init(struct ptp_tx_timeout *timeout)
+{
+	spin_lock_init(&timeout->lock);
+	timeout->active = false;
+}
+
+static void aq_ptp_tx_timeout_start(struct aq_ptp_s *aq_ptp)
+{
+	struct ptp_tx_timeout *timeout = &aq_ptp->ptp_tx_timeout;
+	unsigned long flags;
+
+	spin_lock_irqsave(&timeout->lock, flags);
+	timeout->active = true;
+	timeout->tx_start = jiffies;
+	spin_unlock_irqrestore(&timeout->lock, flags);
+}
+
+static void aq_ptp_tx_timeout_update(struct aq_ptp_s *aq_ptp)
+{
+	if (!aq_ptp_skb_buf_len(&aq_ptp->skb_ring)) {
+		struct ptp_tx_timeout *timeout = &aq_ptp->ptp_tx_timeout;
+		unsigned long flags;
+
+		spin_lock_irqsave(&timeout->lock, flags);
+		timeout->active = false;
+		spin_unlock_irqrestore(&timeout->lock, flags);
+	}
+}
+
+static void aq_ptp_tx_timeout_check(struct aq_ptp_s *aq_ptp)
+{
+	struct ptp_tx_timeout *timeout = &aq_ptp->ptp_tx_timeout;
+	unsigned long flags;
+	bool timeout_flag;
+
+	timeout_flag = false;
+
+	spin_lock_irqsave(&timeout->lock, flags);
+	if (timeout->active) {
+		timeout_flag = time_is_before_jiffies(timeout->tx_start +
+						      AQ_PTP_TX_TIMEOUT);
+		/* reset active flag if timeout detected */
+		if (timeout_flag)
+			timeout->active = false;
+	}
+	spin_unlock_irqrestore(&timeout->lock, flags);
+
+	if (timeout_flag) {
+		aq_ptp_skb_ring_clean(&aq_ptp->skb_ring);
+		netdev_err(aq_ptp->aq_nic->ndev,
+			   "PTP Timeout. Clearing Tx Timestamp SKBs\n");
+	}
+}
+
+/* aq_ptp_adjfine
+ * @ptp: the ptp clock structure
+ * @ppb: parts per billion adjustment from base
+ *
+ * adjust the frequency of the ptp cycle counter by the
+ * indicated ppb from the base frequency.
+ */
+static int aq_ptp_adjfine(struct ptp_clock_info *ptp, long scaled_ppm)
+{
+	struct aq_ptp_s *aq_ptp = container_of(ptp, struct aq_ptp_s, ptp_info);
+	struct aq_nic_s *aq_nic = aq_ptp->aq_nic;
+
+	mutex_lock(&aq_nic->fwreq_mutex);
+	aq_nic->aq_hw_ops->hw_adj_clock_freq(aq_nic->aq_hw,
+					     scaled_ppm_to_ppb(scaled_ppm));
+	mutex_unlock(&aq_nic->fwreq_mutex);
+
+	return 0;
+}
+
+/* aq_ptp_adjtime
+ * @ptp: the ptp clock structure
+ * @delta: offset to adjust the cycle counter by
+ *
+ * adjust the timer by resetting the timecounter structure.
+ */
+static int aq_ptp_adjtime(struct ptp_clock_info *ptp, s64 delta)
+{
+	struct aq_ptp_s *aq_ptp = container_of(ptp, struct aq_ptp_s, ptp_info);
+	struct aq_nic_s *aq_nic = aq_ptp->aq_nic;
+	unsigned long flags;
+
+	spin_lock_irqsave(&aq_ptp->ptp_lock, flags);
+	aq_nic->aq_hw_ops->hw_adj_sys_clock(aq_nic->aq_hw, delta);
+	spin_unlock_irqrestore(&aq_ptp->ptp_lock, flags);
+
+	return 0;
+}
+
+/* aq_ptp_gettime
+ * @ptp: the ptp clock structure
+ * @ts: timespec structure to hold the current time value
+ *
+ * read the timecounter and return the correct value on ns,
+ * after converting it into a struct timespec.
+ */
+static int aq_ptp_gettime(struct ptp_clock_info *ptp, struct timespec64 *ts)
+{
+	struct aq_ptp_s *aq_ptp = container_of(ptp, struct aq_ptp_s, ptp_info);
+	struct aq_nic_s *aq_nic = aq_ptp->aq_nic;
+	unsigned long flags;
+	u64 ns;
+
+	spin_lock_irqsave(&aq_ptp->ptp_lock, flags);
+	aq_nic->aq_hw_ops->hw_get_ptp_ts(aq_nic->aq_hw, &ns);
+	spin_unlock_irqrestore(&aq_ptp->ptp_lock, flags);
+
+	*ts = ns_to_timespec64(ns);
+
+	return 0;
+}
+
+/* aq_ptp_settime
+ * @ptp: the ptp clock structure
+ * @ts: the timespec containing the new time for the cycle counter
+ *
+ * reset the timecounter to use a new base value instead of the kernel
+ * wall timer value.
+ */
+static int aq_ptp_settime(struct ptp_clock_info *ptp,
+			  const struct timespec64 *ts)
+{
+	struct aq_ptp_s *aq_ptp = container_of(ptp, struct aq_ptp_s, ptp_info);
+	struct aq_nic_s *aq_nic = aq_ptp->aq_nic;
+	unsigned long flags;
+	u64 ns = timespec64_to_ns(ts);
+	u64 now;
+
+	spin_lock_irqsave(&aq_ptp->ptp_lock, flags);
+	aq_nic->aq_hw_ops->hw_get_ptp_ts(aq_nic->aq_hw, &now);
+	aq_nic->aq_hw_ops->hw_adj_sys_clock(aq_nic->aq_hw, (s64)ns - (s64)now);
+
+	spin_unlock_irqrestore(&aq_ptp->ptp_lock, flags);
+
+	return 0;
+}
+
+static void aq_ptp_convert_to_hwtstamp(struct aq_ptp_s *aq_ptp,
+				       struct skb_shared_hwtstamps *hwtstamp,
+				       u64 timestamp)
+{
+	memset(hwtstamp, 0, sizeof(*hwtstamp));
+	hwtstamp->hwtstamp = ns_to_ktime(timestamp);
+}
+
+static int aq_ptp_hw_pin_conf(struct aq_nic_s *aq_nic, u32 pin_index, u64 start,
+			      u64 period)
+{
+	if (period)
+		netdev_dbg(aq_nic->ndev,
+			   "Enable GPIO %d pulsing, start time %llu, period %u\n",
+			   pin_index, start, (u32)period);
+	else
+		netdev_dbg(aq_nic->ndev,
+			   "Disable GPIO %d pulsing, start time %llu, period %u\n",
+			   pin_index, start, (u32)period);
+
+	/* Notify hardware of request to being sending pulses.
+	 * If period is ZERO then pulsen is disabled.
+	 */
+	mutex_lock(&aq_nic->fwreq_mutex);
+	aq_nic->aq_hw_ops->hw_gpio_pulse(aq_nic->aq_hw, pin_index,
+					 start, (u32)period);
+	mutex_unlock(&aq_nic->fwreq_mutex);
+
+	return 0;
+}
+
+static int aq_ptp_perout_pin_configure(struct ptp_clock_info *ptp,
+				       struct ptp_clock_request *rq, int on)
+{
+	struct aq_ptp_s *aq_ptp = container_of(ptp, struct aq_ptp_s, ptp_info);
+	struct ptp_clock_time *t = &rq->perout.period;
+	struct ptp_clock_time *s = &rq->perout.start;
+	struct aq_nic_s *aq_nic = aq_ptp->aq_nic;
+	u64 start, period;
+	u32 pin_index = rq->perout.index;
+
+	/* verify the request channel is there */
+	if (pin_index >= ptp->n_per_out)
+		return -EINVAL;
+
+	/* we cannot support periods greater
+	 * than 4 seconds due to reg limit
+	 */
+	if (t->sec > 4 || t->sec < 0)
+		return -ERANGE;
+
+	/* convert to unsigned 64b ns,
+	 * verify we can put it in a 32b register
+	 */
+	period = on ? t->sec * NSEC_PER_SEC + t->nsec : 0;
+
+	/* verify the value is in range supported by hardware */
+	if (period > U32_MAX)
+		return -ERANGE;
+	/* convert to unsigned 64b ns */
+	/* TODO convert to AQ time */
+	start = on ? s->sec * NSEC_PER_SEC + s->nsec : 0;
+
+	aq_ptp_hw_pin_conf(aq_nic, pin_index, start, period);
+
+	return 0;
+}
+
+static int aq_ptp_pps_pin_configure(struct ptp_clock_info *ptp,
+				    struct ptp_clock_request *rq, int on)
+{
+	struct aq_ptp_s *aq_ptp = container_of(ptp, struct aq_ptp_s, ptp_info);
+	struct aq_nic_s *aq_nic = aq_ptp->aq_nic;
+	u64 start, period;
+	u32 pin_index = 0;
+	u32 rest = 0;
+
+	/* verify the request channel is there */
+	if (pin_index >= ptp->n_per_out)
+		return -EINVAL;
+
+	aq_nic->aq_hw_ops->hw_get_ptp_ts(aq_nic->aq_hw, &start);
+	div_u64_rem(start, NSEC_PER_SEC, &rest);
+	period = on ? NSEC_PER_SEC : 0; /* PPS - pulse per second */
+	start = on ? start - rest + NSEC_PER_SEC *
+		(rest > 990000000LL ? 2 : 1) : 0;
+
+	aq_ptp_hw_pin_conf(aq_nic, pin_index, start, period);
+
+	return 0;
+}
+
+static void aq_ptp_extts_pin_ctrl(struct aq_ptp_s *aq_ptp)
+{
+	struct aq_nic_s *aq_nic = aq_ptp->aq_nic;
+	u32 enable = aq_ptp->extts_pin_enabled;
+
+	if (aq_nic->aq_hw_ops->hw_extts_gpio_enable)
+		aq_nic->aq_hw_ops->hw_extts_gpio_enable(aq_nic->aq_hw, 0,
+							enable);
+}
+
+static int aq_ptp_extts_pin_configure(struct ptp_clock_info *ptp,
+				      struct ptp_clock_request *rq, int on)
+{
+	struct aq_ptp_s *aq_ptp = container_of(ptp, struct aq_ptp_s, ptp_info);
+
+	u32 pin_index = rq->extts.index;
+
+	if (pin_index >= ptp->n_ext_ts)
+		return -EINVAL;
+
+	aq_ptp->extts_pin_enabled = !!on;
+	if (on) {
+		aq_ptp->poll_timeout_ms = POLL_SYNC_TIMER_MS;
+		cancel_delayed_work_sync(&aq_ptp->poll_sync);
+		schedule_delayed_work(&aq_ptp->poll_sync,
+				      msecs_to_jiffies(aq_ptp->poll_timeout_ms));
+	}
+
+	aq_ptp_extts_pin_ctrl(aq_ptp);
+	return 0;
+}
+
+/* aq_ptp_gpio_feature_enable
+ * @ptp: the ptp clock structure
+ * @rq: the requested feature to change
+ * @on: whether to enable or disable the feature
+ */
+static int aq_ptp_gpio_feature_enable(struct ptp_clock_info *ptp,
+				      struct ptp_clock_request *rq, int on)
+{
+	switch (rq->type) {
+	case PTP_CLK_REQ_EXTTS:
+		return aq_ptp_extts_pin_configure(ptp, rq, on);
+	case PTP_CLK_REQ_PEROUT:
+		return aq_ptp_perout_pin_configure(ptp, rq, on);
+	case PTP_CLK_REQ_PPS:
+		return aq_ptp_pps_pin_configure(ptp, rq, on);
+	default:
+		return -EOPNOTSUPP;
+	}
+
+	return 0;
+}
+
+/* aq_ptp_verify
+ * @ptp: the ptp clock structure
+ * @pin: index of the pin in question
+ * @func: the desired function to use
+ * @chan: the function channel index to use
+ */
+static int aq_ptp_verify(struct ptp_clock_info *ptp, unsigned int pin,
+			 enum ptp_pin_function func, unsigned int chan)
+{
+	/* verify the requested pin is there */
+	if (!ptp->pin_config || pin >= ptp->n_pins)
+		return -EINVAL;
+
+	/* enforce locked channels, no changing them */
+	if (chan != ptp->pin_config[pin].chan)
+		return -EINVAL;
+
+	/* we want to keep the functions locked as well */
+	if (func != ptp->pin_config[pin].func)
+		return -EINVAL;
+
+	return 0;
+}
+
+/* aq_ptp_tx_hwtstamp - utility function which checks for TX time stamp
+ * @adapter: the private adapter struct
+ *
+ * if the timestamp is valid, we convert it into the timecounter ns
+ * value, then store that result into the hwtstamps structure which
+ * is passed up the network stack
+ */
+void aq_ptp_tx_hwtstamp(struct aq_nic_s *aq_nic, u64 timestamp)
+{
+	struct aq_ptp_s *aq_ptp = aq_nic->aq_ptp;
+	struct sk_buff *skb = aq_ptp_skb_get(&aq_ptp->skb_ring);
+	struct skb_shared_hwtstamps hwtstamp;
+
+	if (!skb) {
+		netdev_err(aq_nic->ndev, "have timestamp but tx_queues empty\n");
+		return;
+	}
+
+	timestamp += atomic_read(&aq_ptp->offset_egress);
+	aq_ptp_convert_to_hwtstamp(aq_ptp, &hwtstamp, timestamp);
+	skb_tstamp_tx(skb, &hwtstamp);
+	dev_kfree_skb_any(skb);
+
+	aq_ptp_tx_timeout_update(aq_ptp);
+}
+
+/* aq_ptp_rx_hwtstamp - utility function which checks for RX time stamp
+ * @adapter: pointer to adapter struct
+ * @shhwtstamps: particular skb_shared_hwtstamps to save timestamp
+ *
+ * if the timestamp is valid, we convert it into the timecounter ns
+ * value, then store that result into the hwtstamps structure which
+ * is passed up the network stack
+ */
+static void aq_ptp_rx_hwtstamp(struct aq_ptp_s *aq_ptp, struct skb_shared_hwtstamps *shhwtstamps,
+			       u64 timestamp)
+{
+	timestamp -= atomic_read(&aq_ptp->offset_ingress);
+	aq_ptp_convert_to_hwtstamp(aq_ptp, shhwtstamps, timestamp);
+}
+
+void aq_ptp_hwtstamp_config_get(struct aq_ptp_s *aq_ptp,
+				struct hwtstamp_config *config)
+{
+	*config = aq_ptp->hwtstamp_config;
+}
+
+static void aq_ptp_prepare_filters(struct aq_ptp_s *aq_ptp)
+{
+	aq_ptp->udp_filter.cmd = HW_ATL_RX_ENABLE_FLTR_L3L4 |
+			       HW_ATL_RX_ENABLE_CMP_PROT_L4 |
+			       HW_ATL_RX_UDP |
+			       HW_ATL_RX_ENABLE_CMP_DEST_PORT_L4 |
+			       HW_ATL_RX_HOST << HW_ATL_RX_ACTION_FL3F4_SHIFT |
+			       HW_ATL_RX_ENABLE_QUEUE_L3L4 |
+			       aq_ptp->ptp_rx.idx << HW_ATL_RX_QUEUE_FL3L4_SHIFT;
+	aq_ptp->udp_filter.p_dst = PTP_EV_PORT;
+
+	aq_ptp->eth_type_filter.ethertype = ETH_P_1588;
+	aq_ptp->eth_type_filter.queue = aq_ptp->ptp_rx.idx;
+}
+
+int aq_ptp_hwtstamp_config_set(struct aq_ptp_s *aq_ptp,
+			       struct hwtstamp_config *config)
+{
+	struct aq_nic_s *aq_nic = aq_ptp->aq_nic;
+	const struct aq_hw_ops *hw_ops;
+	int err = 0;
+
+	hw_ops = aq_nic->aq_hw_ops;
+	if (config->tx_type == HWTSTAMP_TX_ON ||
+	    config->rx_filter == HWTSTAMP_FILTER_PTP_V2_EVENT) {
+		aq_ptp_prepare_filters(aq_ptp);
+		if (hw_ops->hw_filter_l3l4_set) {
+			err = hw_ops->hw_filter_l3l4_set(aq_nic->aq_hw,
+							 &aq_ptp->udp_filter);
+		}
+		if (!err && hw_ops->hw_filter_l2_set) {
+			err = hw_ops->hw_filter_l2_set(aq_nic->aq_hw,
+						       &aq_ptp->eth_type_filter);
+		}
+		aq_utils_obj_set(&aq_nic->flags, AQ_NIC_PTP_DPATH_UP);
+	} else {
+		aq_ptp->udp_filter.cmd &= ~HW_ATL_RX_ENABLE_FLTR_L3L4;
+		if (hw_ops->hw_filter_l3l4_set) {
+			err = hw_ops->hw_filter_l3l4_set(aq_nic->aq_hw,
+							 &aq_ptp->udp_filter);
+		}
+		if (!err && hw_ops->hw_filter_l2_clear) {
+			err = hw_ops->hw_filter_l2_clear(aq_nic->aq_hw,
+							&aq_ptp->eth_type_filter);
+		}
+		aq_utils_obj_clear(&aq_nic->flags, AQ_NIC_PTP_DPATH_UP);
+	}
+
+	if (err)
+		return -EREMOTEIO;
+
+	aq_ptp->hwtstamp_config = *config;
+
+	return 0;
+}
+
+bool aq_ptp_ring(struct aq_nic_s *aq_nic, struct aq_ring_s *ring)
+{
+	struct aq_ptp_s *aq_ptp = aq_nic->aq_ptp;
+
+	if (!aq_ptp)
+		return false;
+
+	return &aq_ptp->ptp_tx == ring ||
+	       &aq_ptp->ptp_rx == ring || &aq_ptp->hwts_rx == ring;
+}
+
+u16 aq_ptp_extract_ts(struct aq_nic_s *aq_nic, struct skb_shared_hwtstamps *shhwtstamps, u8 *p,
+		      unsigned int len)
+{
+	struct aq_ptp_s *aq_ptp = aq_nic->aq_ptp;
+	u64 timestamp = 0;
+	u16 ret = aq_nic->aq_hw_ops->rx_extract_ts(aq_nic->aq_hw,
+						   p, len, &timestamp);
+
+	if (ret > 0)
+		aq_ptp_rx_hwtstamp(aq_ptp, shhwtstamps, timestamp);
+
+	return ret;
+}
+
+static int aq_ptp_poll(struct napi_struct *napi, int budget)
+{
+	struct aq_ptp_s *aq_ptp = container_of(napi, struct aq_ptp_s, napi);
+	struct aq_nic_s *aq_nic = aq_ptp->aq_nic;
+	bool was_cleaned = false;
+	int work_done = 0;
+	int err;
+
+	/* Processing PTP TX traffic */
+	err = aq_nic->aq_hw_ops->hw_ring_tx_head_update(aq_nic->aq_hw,
+							&aq_ptp->ptp_tx);
+	if (err < 0)
+		goto err_exit;
+
+	if (aq_ptp->ptp_tx.sw_head != aq_ptp->ptp_tx.hw_head) {
+		aq_ring_tx_clean(&aq_ptp->ptp_tx);
+
+		was_cleaned = true;
+	}
+
+	/* Processing HW_TIMESTAMP RX traffic */
+	err = aq_nic->aq_hw_ops->hw_ring_hwts_rx_receive(aq_nic->aq_hw,
+							 &aq_ptp->hwts_rx);
+	if (err < 0)
+		goto err_exit;
+
+	if (aq_ptp->hwts_rx.sw_head != aq_ptp->hwts_rx.hw_head) {
+		aq_ring_hwts_rx_clean(&aq_ptp->hwts_rx, aq_nic);
+
+		err = aq_nic->aq_hw_ops->hw_ring_hwts_rx_fill(aq_nic->aq_hw,
+							      &aq_ptp->hwts_rx);
+		if (err < 0)
+			goto err_exit;
+
+		was_cleaned = true;
+	}
+
+	/* Processing PTP RX traffic */
+	err = aq_nic->aq_hw_ops->hw_ring_rx_receive(aq_nic->aq_hw,
+						    &aq_ptp->ptp_rx);
+	if (err < 0)
+		goto err_exit;
+
+	if (aq_ptp->ptp_rx.sw_head != aq_ptp->ptp_rx.hw_head) {
+		unsigned int sw_tail_old;
+
+		err = aq_ring_rx_clean(&aq_ptp->ptp_rx, napi, &work_done, budget);
+		if (err < 0)
+			goto err_exit;
+
+		sw_tail_old = aq_ptp->ptp_rx.sw_tail;
+		err = aq_ring_rx_fill(&aq_ptp->ptp_rx);
+		if (err < 0)
+			goto err_exit;
+
+		err = aq_nic->aq_hw_ops->hw_ring_rx_fill(aq_nic->aq_hw,
+							 &aq_ptp->ptp_rx,
+							 sw_tail_old);
+		if (err < 0)
+			goto err_exit;
+	}
+
+	if (was_cleaned)
+		work_done = budget;
+
+	if (work_done < budget) {
+		napi_complete_done(napi, work_done);
+		aq_nic->aq_hw_ops->hw_irq_enable(aq_nic->aq_hw,
+					BIT_ULL(aq_ptp->ptp_ring_param.vec_idx));
+	}
+
+err_exit:
+	return work_done;
+}
+
+static irqreturn_t aq_ptp_isr(int irq, void *private)
+{
+	struct aq_ptp_s *aq_ptp = private;
+	int err = 0;
+
+	if (!aq_ptp) {
+		err = -EINVAL;
+		goto err_exit;
+	}
+	napi_schedule(&aq_ptp->napi);
+
+err_exit:
+	return err >= 0 ? IRQ_HANDLED : IRQ_NONE;
+}
+
+int aq_ptp_xmit(struct aq_nic_s *aq_nic, struct sk_buff *skb)
+{
+	struct aq_ptp_s *aq_ptp = aq_nic->aq_ptp;
+	struct aq_ring_s *ring = &aq_ptp->ptp_tx;
+	unsigned long irq_flags;
+	int err = NETDEV_TX_OK;
+	unsigned int frags;
+
+	if (skb->len <= 0) {
+		dev_kfree_skb_any(skb);
+		goto err_exit;
+	}
+
+	frags = skb_shinfo(skb)->nr_frags + 1;
+	/* Frags cannot be bigger 16KB
+	 * because PTP usually works
+	 * without Jumbo even in a background
+	 */
+	if (frags > AQ_CFG_SKB_FRAGS_MAX || frags > aq_ring_avail_dx(ring)) {
+		/* Drop packet because it doesn't make sence to delay it */
+		dev_kfree_skb_any(skb);
+		goto err_exit;
+	}
+
+	err = aq_ptp_skb_put(&aq_ptp->skb_ring, skb);
+	if (err) {
+		netdev_err(aq_nic->ndev, "SKB Ring is overflow (%u)!\n",
+			   ring->size);
+		return NETDEV_TX_BUSY;
+	}
+	skb_shinfo(skb)->tx_flags |= SKBTX_IN_PROGRESS;
+	aq_ptp_tx_timeout_start(aq_ptp);
+	skb_tx_timestamp(skb);
+
+	spin_lock_irqsave(&aq_nic->aq_ptp->ptp_ring_lock, irq_flags);
+	frags = aq_nic_map_skb(aq_nic, skb, ring);
+
+	if (likely(frags)) {
+		err = aq_nic->aq_hw_ops->hw_ring_tx_xmit(aq_nic->aq_hw,
+						       ring, frags);
+		if (err >= 0) {
+			u64_stats_update_begin(&ring->stats.tx.syncp);
+			++ring->stats.tx.packets;
+			ring->stats.tx.bytes += skb->len;
+			u64_stats_update_end(&ring->stats.tx.syncp);
+		}
+	} else {
+		err = NETDEV_TX_BUSY;
+	}
+	spin_unlock_irqrestore(&aq_nic->aq_ptp->ptp_ring_lock, irq_flags);
+
+err_exit:
+	return err;
+}
+
+void aq_ptp_service_task(struct aq_nic_s *aq_nic)
+{
+	struct aq_ptp_s *aq_ptp = aq_nic->aq_ptp;
+
+	if (!aq_ptp)
+		return;
+
+	aq_ptp_tx_timeout_check(aq_ptp);
+}
+
+int aq_ptp_irq_alloc(struct aq_nic_s *aq_nic)
+{
+	struct pci_dev *pdev = aq_nic->pdev;
+	struct aq_ptp_s *aq_ptp = aq_nic->aq_ptp;
+	int err = 0;
+
+	if (!aq_ptp)
+		return 0;
+
+	if (pdev->msix_enabled || pdev->msi_enabled) {
+		err = request_irq(pci_irq_vector(pdev, aq_ptp->idx_vector),
+				  aq_ptp_isr, 0, aq_nic->ndev->name, aq_ptp);
+	} else {
+		err = -EINVAL;
+		goto err_exit;
+	}
+
+err_exit:
+	return err;
+}
+
+void aq_ptp_irq_free(struct aq_nic_s *aq_nic)
+{
+	struct aq_ptp_s *aq_ptp = aq_nic->aq_ptp;
+	struct pci_dev *pdev = aq_nic->pdev;
+
+	if (!aq_ptp)
+		return;
+
+	free_irq(pci_irq_vector(pdev, aq_ptp->idx_vector), aq_ptp);
+}
+
+int aq_ptp_ring_init(struct aq_nic_s *aq_nic)
+{
+	struct aq_ptp_s *aq_ptp = aq_nic->aq_ptp;
+	int err = 0;
+
+	if (!aq_ptp)
+		return 0;
+
+	err = aq_ring_init(&aq_ptp->ptp_tx, ATL_RING_TX);
+	if (err < 0)
+		goto err_exit;
+	err = aq_nic->aq_hw_ops->hw_ring_tx_init(aq_nic->aq_hw,
+						 &aq_ptp->ptp_tx,
+						 &aq_ptp->ptp_ring_param);
+	if (err < 0)
+		goto err_exit;
+
+	err = aq_ring_init(&aq_ptp->ptp_rx, ATL_RING_RX);
+	if (err < 0)
+		goto err_exit;
+	err = aq_nic->aq_hw_ops->hw_ring_rx_init(aq_nic->aq_hw,
+						 &aq_ptp->ptp_rx,
+						 &aq_ptp->ptp_ring_param);
+	if (err < 0)
+		goto err_exit;
+
+	err = aq_ring_rx_fill(&aq_ptp->ptp_rx);
+	if (err < 0)
+		goto err_rx_free;
+	err = aq_nic->aq_hw_ops->hw_ring_rx_fill(aq_nic->aq_hw,
+						 &aq_ptp->ptp_rx,
+						 0U);
+	if (err < 0)
+		goto err_rx_free;
+
+	err = aq_ring_init(&aq_ptp->hwts_rx, ATL_RING_RX);
+	if (err < 0)
+		goto err_rx_free;
+	err = aq_nic->aq_hw_ops->hw_ring_rx_init(aq_nic->aq_hw,
+						 &aq_ptp->hwts_rx,
+						 &aq_ptp->ptp_ring_param);
+	if (err < 0)
+		goto err_exit;
+	err = aq_nic->aq_hw_ops->hw_ring_hwts_rx_fill(aq_nic->aq_hw,
+						      &aq_ptp->hwts_rx);
+	if (err < 0)
+		goto err_exit;
+
+	return err;
+
+err_rx_free:
+	aq_ring_rx_deinit(&aq_ptp->ptp_rx);
+err_exit:
+	return err;
+}
+
+int aq_ptp_ring_start(struct aq_nic_s *aq_nic)
+{
+	struct aq_ptp_s *aq_ptp = aq_nic->aq_ptp;
+	int err = 0;
+
+	if (!aq_ptp)
+		return 0;
+
+	err = aq_nic->aq_hw_ops->hw_ring_tx_start(aq_nic->aq_hw, &aq_ptp->ptp_tx);
+	if (err < 0)
+		goto err_exit;
+
+	err = aq_nic->aq_hw_ops->hw_ring_rx_start(aq_nic->aq_hw, &aq_ptp->ptp_rx);
+	if (err < 0)
+		goto err_exit;
+
+	err = aq_nic->aq_hw_ops->hw_ring_rx_start(aq_nic->aq_hw,
+						  &aq_ptp->hwts_rx);
+	if (err < 0)
+		goto err_exit;
+
+	napi_enable(&aq_ptp->napi);
+
+err_exit:
+	return err;
+}
+
+void aq_ptp_ring_stop(struct aq_nic_s *aq_nic)
+{
+	struct aq_ptp_s *aq_ptp = aq_nic->aq_ptp;
+
+	if (!aq_ptp)
+		return;
+
+	aq_nic->aq_hw_ops->hw_ring_tx_stop(aq_nic->aq_hw, &aq_ptp->ptp_tx);
+	aq_nic->aq_hw_ops->hw_ring_rx_stop(aq_nic->aq_hw, &aq_ptp->ptp_rx);
+
+	aq_nic->aq_hw_ops->hw_ring_rx_stop(aq_nic->aq_hw, &aq_ptp->hwts_rx);
+
+	napi_disable(&aq_ptp->napi);
+}
+
+void aq_ptp_ring_deinit(struct aq_nic_s *aq_nic)
+{
+	struct aq_ptp_s *aq_ptp = aq_nic->aq_ptp;
+
+	if (!aq_ptp || !aq_ptp->ptp_tx.aq_nic || !aq_ptp->ptp_rx.aq_nic)
+		return;
+
+	aq_ring_tx_clean(&aq_ptp->ptp_tx);
+	aq_ring_rx_deinit(&aq_ptp->ptp_rx);
+}
+
+int aq_ptp_ring_alloc(struct aq_nic_s *aq_nic)
+{
+	struct aq_ptp_s *aq_ptp = aq_nic->aq_ptp;
+	unsigned int tx_ring_idx, rx_ring_idx;
+	int err;
+
+	if (!aq_ptp)
+		return 0;
+
+	tx_ring_idx = aq_ptp_ring_idx(aq_nic->aq_nic_cfg.tc_mode);
+
+	err = aq_ring_tx_alloc(&aq_ptp->ptp_tx, aq_nic,
+			       tx_ring_idx, &aq_nic->aq_nic_cfg);
+	if (err)
+		goto err_exit;
+
+	rx_ring_idx = aq_ptp_ring_idx(aq_nic->aq_nic_cfg.tc_mode);
+
+	err = aq_ring_rx_alloc(&aq_ptp->ptp_rx, aq_nic,
+			       rx_ring_idx, &aq_nic->aq_nic_cfg);
+	if (err)
+		goto err_exit_ptp_tx;
+
+	err = aq_ring_hwts_rx_alloc(&aq_ptp->hwts_rx, aq_nic, PTP_HWST_RING_IDX,
+				    aq_nic->aq_nic_cfg.rxds,
+				    aq_nic->aq_nic_cfg.aq_hw_caps->rxd_size);
+	if (err)
+		goto err_exit_ptp_rx;
+
+	err = aq_ptp_skb_ring_init(&aq_ptp->skb_ring, aq_nic->aq_nic_cfg.rxds);
+	if (err != 0) {
+		err = -ENOMEM;
+		goto err_exit_hwts_rx;
+	}
+
+	aq_ptp->ptp_ring_param.vec_idx = aq_ptp->idx_vector;
+	aq_ptp->ptp_ring_param.cpu = aq_ptp->ptp_ring_param.vec_idx +
+			aq_nic_get_cfg(aq_nic)->aq_rss.base_cpu_number;
+	cpumask_set_cpu(aq_ptp->ptp_ring_param.cpu,
+			&aq_ptp->ptp_ring_param.affinity_mask);
+
+	return 0;
+
+err_exit_hwts_rx:
+	aq_ring_hwts_rx_free(&aq_ptp->hwts_rx);
+err_exit_ptp_rx:
+	aq_ring_free(&aq_ptp->ptp_rx);
+err_exit_ptp_tx:
+	aq_ring_free(&aq_ptp->ptp_tx);
+err_exit:
+	return err;
+}
+
+void aq_ptp_ring_free(struct aq_nic_s *aq_nic)
+{
+	struct aq_ptp_s *aq_ptp = aq_nic->aq_ptp;
+
+	if (!aq_ptp)
+		return;
+
+	aq_ring_free(&aq_ptp->ptp_tx);
+	aq_ring_free(&aq_ptp->ptp_rx);
+	aq_ring_hwts_rx_free(&aq_ptp->hwts_rx);
+
+	aq_ptp_skb_ring_release(&aq_ptp->skb_ring);
+}
+
+#define MAX_PTP_GPIO_COUNT 4
+
+static struct ptp_clock_info aq_ptp_clock = {
+	.owner		= THIS_MODULE,
+	.name		= "atlantic ptp",
+	.max_adj	= 999999999,
+	.n_ext_ts	= 0,
+	.pps		= 0,
+	.adjfine	= aq_ptp_adjfine,
+	.adjtime	= aq_ptp_adjtime,
+	.gettime64	= aq_ptp_gettime,
+	.settime64	= aq_ptp_settime,
+	.n_per_out	= 0,
+	.enable		= aq_ptp_gpio_feature_enable,
+	.n_pins		= 0,
+	.verify		= aq_ptp_verify,
+	.pin_config	= NULL,
+};
+
+#define ptp_offset_init(__idx, __mbps, __egress, __ingress)   do { \
+		ptp_offset[__idx].mbps = (__mbps); \
+		ptp_offset[__idx].egress = (__egress); \
+		ptp_offset[__idx].ingress = (__ingress); } \
+		while (0)
+
+static void aq_ptp_offset_init_from_fw(const struct hw_atl_ptp_offset *offsets)
+{
+	int i;
+
+	/* Load offsets for PTP */
+	for (i = 0; i < ARRAY_SIZE(ptp_offset); i++) {
+		switch (i) {
+		/* 100M */
+		case ptp_offset_idx_100:
+			ptp_offset_init(i, 100,
+					offsets->egress_100,
+					offsets->ingress_100);
+			break;
+		/* 1G */
+		case ptp_offset_idx_1000:
+			ptp_offset_init(i, 1000,
+					offsets->egress_1000,
+					offsets->ingress_1000);
+			break;
+		/* 2.5G */
+		case ptp_offset_idx_2500:
+			ptp_offset_init(i, 2500,
+					offsets->egress_2500,
+					offsets->ingress_2500);
+			break;
+		/* 5G */
+		case ptp_offset_idx_5000:
+			ptp_offset_init(i, 5000,
+					offsets->egress_5000,
+					offsets->ingress_5000);
+			break;
+		/* 10G */
+		case ptp_offset_idx_10000:
+			ptp_offset_init(i, 10000,
+					offsets->egress_10000,
+					offsets->ingress_10000);
+			break;
+		}
+	}
+}
+
+static void aq_ptp_offset_init(const struct hw_atl_ptp_offset *offsets)
+{
+	memset(ptp_offset, 0, sizeof(ptp_offset));
+
+	aq_ptp_offset_init_from_fw(offsets);
+}
+
+static void aq_ptp_gpio_init(struct ptp_clock_info *info,
+			     struct hw_atl_info *hw_info)
+{
+	struct ptp_pin_desc pin_desc[MAX_PTP_GPIO_COUNT];
+	u32 extts_pin_cnt = 0;
+	u32 out_pin_cnt = 0;
+	u32 i;
+
+	memset(pin_desc, 0, sizeof(pin_desc));
+
+	for (i = 0; i < MAX_PTP_GPIO_COUNT - 1; i++) {
+		if (hw_info->gpio_pin[i] ==
+		    (GPIO_PIN_FUNCTION_PTP0 + out_pin_cnt)) {
+			snprintf(pin_desc[out_pin_cnt].name,
+				 sizeof(pin_desc[out_pin_cnt].name),
+				 "AQ_GPIO%d", i);
+			pin_desc[out_pin_cnt].index = out_pin_cnt;
+			pin_desc[out_pin_cnt].chan = out_pin_cnt;
+			pin_desc[out_pin_cnt++].func = PTP_PF_PEROUT;
+		}
+	}
+
+	info->n_per_out = out_pin_cnt;
+
+	if (hw_info->caps_ex & BIT(CAPS_EX_PHY_CTRL_TS_PIN)) {
+		extts_pin_cnt += 1;
+
+		snprintf(pin_desc[out_pin_cnt].name,
+			 sizeof(pin_desc[out_pin_cnt].name),
+			  "AQ_GPIO%d", out_pin_cnt);
+		pin_desc[out_pin_cnt].index = out_pin_cnt;
+		pin_desc[out_pin_cnt].chan = 0;
+		pin_desc[out_pin_cnt].func = PTP_PF_EXTTS;
+	}
+
+	info->n_pins = out_pin_cnt + extts_pin_cnt;
+	info->n_ext_ts = extts_pin_cnt;
+
+	if (!info->n_pins)
+		return;
+
+	info->pin_config = kcalloc(info->n_pins, sizeof(struct ptp_pin_desc),
+				   GFP_KERNEL);
+
+	if (!info->pin_config)
+		return;
+
+	memcpy(info->pin_config, &pin_desc,
+	       sizeof(struct ptp_pin_desc) * info->n_pins);
+}
+
+void aq_ptp_clock_init(struct aq_nic_s *aq_nic)
+{
+	struct aq_ptp_s *aq_ptp = aq_nic->aq_ptp;
+	struct timespec64 ts;
+
+	ktime_get_real_ts64(&ts);
+	aq_ptp_settime(&aq_ptp->ptp_info, &ts);
+}
+
+static void aq_ptp_poll_sync_work_cb(struct work_struct *w);
+
+int aq_ptp_init(struct aq_nic_s *aq_nic, unsigned int idx_vec)
+{
+	bool a1_ptp = ATL_HW_IS_CHIP_FEATURE(aq_nic->aq_hw, ATLANTIC);
+	struct hw_atl_utils_mbox mbox;
+	struct ptp_clock *clock;
+	struct aq_ptp_s *aq_ptp;
+	int err = 0;
+
+	if (!a1_ptp) {
+		aq_nic->aq_ptp = NULL;
+		return 0;
+	}
+
+	if (!aq_nic->aq_hw_ops->hw_get_ptp_ts) {
+		aq_nic->aq_ptp = NULL;
+		return 0;
+	}
+
+	if (!aq_nic->aq_fw_ops->enable_ptp) {
+		aq_nic->aq_ptp = NULL;
+		return 0;
+	}
+
+	hw_atl_utils_mpi_read_stats(aq_nic->aq_hw, &mbox);
+
+	if (!(mbox.info.caps_ex & BIT(CAPS_EX_PHY_PTP_EN))) {
+		aq_nic->aq_ptp = NULL;
+		return 0;
+	}
+
+	aq_ptp_offset_init(&mbox.info.ptp_offset);
+
+	aq_ptp = kzalloc(sizeof(*aq_ptp), GFP_KERNEL);
+	if (!aq_ptp) {
+		err = -ENOMEM;
+		goto err_exit;
+	}
+
+	aq_ptp->aq_nic = aq_nic;
+	aq_ptp->a1_ptp = a1_ptp;
+
+	spin_lock_init(&aq_ptp->ptp_lock);
+	spin_lock_init(&aq_ptp->ptp_ring_lock);
+
+	aq_ptp->ptp_info = aq_ptp_clock;
+	aq_ptp_gpio_init(&aq_ptp->ptp_info, &mbox.info);
+	clock = ptp_clock_register(&aq_ptp->ptp_info, &aq_nic->ndev->dev);
+	if (IS_ERR(clock)) {
+		netdev_err(aq_nic->ndev, "ptp_clock_register failed\n");
+		err = PTR_ERR(clock);
+		goto err_exit;
+	}
+	aq_ptp->ptp_clock = clock;
+	aq_ptp_tx_timeout_init(&aq_ptp->ptp_tx_timeout);
+
+	atomic_set(&aq_ptp->offset_egress, 0);
+	atomic_set(&aq_ptp->offset_ingress, 0);
+
+	netif_napi_add(aq_nic_get_ndev(aq_nic), &aq_ptp->napi, aq_ptp_poll);
+
+	aq_ptp->idx_vector = idx_vec;
+
+	aq_nic->aq_ptp = aq_ptp;
+
+	/* enable ptp counter */
+	aq_utils_obj_set(&aq_nic->aq_hw->flags, AQ_HW_PTP_AVAILABLE);
+	mutex_lock(&aq_nic->fwreq_mutex);
+	aq_nic->aq_fw_ops->enable_ptp(aq_nic->aq_hw, 1);
+	aq_ptp_clock_init(aq_nic);
+	mutex_unlock(&aq_nic->fwreq_mutex);
+
+	INIT_DELAYED_WORK(&aq_ptp->poll_sync, &aq_ptp_poll_sync_work_cb);
+	aq_ptp->eth_type_filter.location =
+			aq_nic_reserve_filter(aq_nic, aq_rx_filter_ethertype);
+	aq_ptp->udp_filter.location =
+			aq_nic_reserve_filter(aq_nic, aq_rx_filter_l3l4);
+
+	return 0;
+
+err_exit:
+	if (aq_ptp)
+		kfree(aq_ptp->ptp_info.pin_config);
+	kfree(aq_ptp);
+	aq_nic->aq_ptp = NULL;
+	return err;
+}
+
+void aq_ptp_unregister(struct aq_nic_s *aq_nic)
+{
+	struct aq_ptp_s *aq_ptp = aq_nic->aq_ptp;
+
+	if (!aq_ptp)
+		return;
+
+	ptp_clock_unregister(aq_ptp->ptp_clock);
+}
+
+void aq_ptp_free(struct aq_nic_s *aq_nic)
+{
+	struct aq_ptp_s *aq_ptp = aq_nic->aq_ptp;
+
+	if (!aq_ptp)
+		return;
+
+	aq_nic_release_filter(aq_nic, aq_rx_filter_ethertype,
+			      aq_ptp->eth_type_filter.location);
+	aq_nic_release_filter(aq_nic, aq_rx_filter_l3l4,
+			      aq_ptp->udp_filter.location);
+	cancel_delayed_work_sync(&aq_ptp->poll_sync);
+	/* disable ptp */
+	mutex_lock(&aq_nic->fwreq_mutex);
+	aq_nic->aq_fw_ops->enable_ptp(aq_nic->aq_hw, 0);
+	mutex_unlock(&aq_nic->fwreq_mutex);
+
+	kfree(aq_ptp->ptp_info.pin_config);
+
+	netif_napi_del(&aq_ptp->napi);
+	kfree(aq_ptp);
+	aq_nic->aq_ptp = NULL;
+}
+
+struct ptp_clock *aq_ptp_get_ptp_clock(struct aq_ptp_s *aq_ptp)
+{
+	return aq_ptp->ptp_clock;
+}
+
+/* PTP external GPIO nanoseconds count */
+static uint64_t aq_ptp_get_sync1588_ts(struct aq_nic_s *aq_nic)
+{
+	u64 ts = 0;
+
+	if (aq_nic->aq_hw_ops->hw_get_sync_ts)
+		aq_nic->aq_hw_ops->hw_get_sync_ts(aq_nic->aq_hw, &ts);
+
+	return ts;
+}
+
+static void aq_ptp_start_work(struct aq_ptp_s *aq_ptp)
+{
+	if (aq_ptp->extts_pin_enabled) {
+		aq_ptp->poll_timeout_ms = POLL_SYNC_TIMER_MS;
+		aq_ptp->last_sync1588_ts =
+				aq_ptp_get_sync1588_ts(aq_ptp->aq_nic);
+		schedule_delayed_work(&aq_ptp->poll_sync,
+				      msecs_to_jiffies(aq_ptp->poll_timeout_ms));
+	}
+}
+
+int aq_ptp_link_change(struct aq_nic_s *aq_nic)
+{
+	struct aq_ptp_s *aq_ptp = aq_nic->aq_ptp;
+
+	if (!aq_ptp)
+		return 0;
+
+	if (aq_nic->aq_hw->aq_link_status.mbps)
+		aq_ptp_start_work(aq_ptp);
+	else
+		cancel_delayed_work_sync(&aq_ptp->poll_sync);
+
+	return 0;
+}
+
+static bool aq_ptp_sync_ts_updated(struct aq_ptp_s *aq_ptp, u64 *new_ts)
+{
+	struct aq_nic_s *aq_nic = aq_ptp->aq_nic;
+	u64 sync_ts2;
+	u64 sync_ts;
+
+	sync_ts = aq_ptp_get_sync1588_ts(aq_nic);
+
+	if (sync_ts != aq_ptp->last_sync1588_ts) {
+		sync_ts2 = aq_ptp_get_sync1588_ts(aq_nic);
+		if (sync_ts != sync_ts2) {
+			sync_ts = sync_ts2;
+			sync_ts2 = aq_ptp_get_sync1588_ts(aq_nic);
+			if (sync_ts != sync_ts2) {
+				netdev_err(aq_nic->ndev,
+					   "%s: Unable to get correct GPIO TS",
+					   __func__);
+				sync_ts = 0;
+			}
+		}
+
+		*new_ts = sync_ts;
+		return true;
+	}
+	return false;
+}
+
+static int aq_ptp_check_sync1588(struct aq_ptp_s *aq_ptp)
+{
+	struct aq_nic_s *aq_nic = aq_ptp->aq_nic;
+	u64 sync_ts;
+
+	 /* Sync1588 pin was triggered */
+	if (aq_ptp_sync_ts_updated(aq_ptp, &sync_ts)) {
+		if (aq_ptp->extts_pin_enabled) {
+			struct ptp_clock_event ptp_event;
+			u64 time = 0;
+
+			aq_nic->aq_hw_ops->hw_ts_to_sys_clock(aq_nic->aq_hw,
+							      sync_ts, &time);
+			ptp_event.index = aq_ptp->ptp_info.n_pins - 1;
+			ptp_event.timestamp = time;
+
+			ptp_event.type = PTP_CLOCK_EXTTS;
+			ptp_clock_event(aq_ptp->ptp_clock, &ptp_event);
+		}
+
+		aq_ptp->last_sync1588_ts = sync_ts;
+	}
+
+	return 0;
+}
+
+static void aq_ptp_poll_sync_work_cb(struct work_struct *w)
+{
+	struct delayed_work *dw = to_delayed_work(w);
+	struct aq_ptp_s *aq_ptp = container_of(dw, struct aq_ptp_s, poll_sync);
+
+	aq_ptp_check_sync1588(aq_ptp);
+
+	if (aq_ptp->extts_pin_enabled) {
+		unsigned long timeout = msecs_to_jiffies(aq_ptp->poll_timeout_ms);
+
+		schedule_delayed_work(&aq_ptp->poll_sync, timeout);
+	}
+}
+
+int aq_ptp_get_ring_cnt(struct aq_nic_s *aq_nic, const enum atl_ring_type ring_type)
+{
+	if (!aq_nic->aq_ptp)
+		return 0;
+
+	/* Additional RX ring is allocated for PTP HWTS on A1 */
+	return (aq_nic->aq_ptp->a1_ptp && ring_type == ATL_RING_RX) ? 2 : 1;
+}
+
+u64 *aq_ptp_get_stats(struct aq_nic_s *aq_nic, u64 *data)
+{
+	struct aq_ptp_s *aq_ptp = aq_nic->aq_ptp;
+	unsigned int count = 0U;
+
+	if (!aq_ptp)
+		return data;
+
+	count = aq_ring_fill_stats_data(&aq_ptp->ptp_rx, data);
+	data += count;
+	count = aq_ring_fill_stats_data(&aq_ptp->ptp_tx, data);
+	data += count;
+
+	if (aq_ptp->a1_ptp) {
+		/* Only Receive ring for HWTS */
+		count = aq_ring_fill_stats_data(&aq_ptp->hwts_rx, data);
+		data += count;
+	}
+
+	return data;
+}
+
+#endif
diff --git a/drivers/net/ethernet/aquantia/atlantic/aq_ptp.h b/drivers/net/ethernet/aquantia/atlantic/aq_ptp.h
new file mode 100644
index 000000000000..210b723f2207
--- /dev/null
+++ b/drivers/net/ethernet/aquantia/atlantic/aq_ptp.h
@@ -0,0 +1,163 @@
+/* SPDX-License-Identifier: GPL-2.0-only */
+/* Atlantic Network Driver
+ *
+ * Copyright (C) 2014-2019 aQuantia Corporation
+ * Copyright (C) 2019-2020 Marvell International Ltd.
+ */
+
+/* File aq_ptp.h: Declaration of PTP functions.
+ */
+#ifndef AQ_PTP_H
+#define AQ_PTP_H
+
+#include <linux/net_tstamp.h>
+
+#include "aq_ring.h"
+
+#define PTP_8TC_RING_IDX             8
+#define PTP_4TC_RING_IDX            16
+#define PTP_HWST_RING_IDX           31
+
+/* Index must to be 8 (8 TCs) or 16 (4 TCs).
+ * It depends from Traffic Class mode.
+ */
+static inline unsigned int aq_ptp_ring_idx(const enum aq_tc_mode tc_mode)
+{
+	if (tc_mode == AQ_TC_MODE_8TCS)
+		return PTP_8TC_RING_IDX;
+
+	return PTP_4TC_RING_IDX;
+}
+
+#if IS_REACHABLE(CONFIG_PTP_1588_CLOCK)
+
+/* Common functions */
+int aq_ptp_init(struct aq_nic_s *aq_nic, unsigned int idx_vec);
+
+void aq_ptp_unregister(struct aq_nic_s *aq_nic);
+void aq_ptp_free(struct aq_nic_s *aq_nic);
+
+int aq_ptp_irq_alloc(struct aq_nic_s *aq_nic);
+void aq_ptp_irq_free(struct aq_nic_s *aq_nic);
+
+int aq_ptp_ring_alloc(struct aq_nic_s *aq_nic);
+void aq_ptp_ring_free(struct aq_nic_s *aq_nic);
+
+int aq_ptp_ring_init(struct aq_nic_s *aq_nic);
+int aq_ptp_ring_start(struct aq_nic_s *aq_nic);
+void aq_ptp_ring_stop(struct aq_nic_s *aq_nic);
+void aq_ptp_ring_deinit(struct aq_nic_s *aq_nic);
+
+void aq_ptp_service_task(struct aq_nic_s *aq_nic);
+
+void aq_ptp_tm_offset_set(struct aq_nic_s *aq_nic, unsigned int mbps);
+
+void aq_ptp_clock_init(struct aq_nic_s *aq_nic);
+
+/* Traffic processing functions */
+int aq_ptp_xmit(struct aq_nic_s *aq_nic, struct sk_buff *skb);
+void aq_ptp_tx_hwtstamp(struct aq_nic_s *aq_nic, u64 timestamp);
+
+/* Must be to check available of PTP before call */
+void aq_ptp_hwtstamp_config_get(struct aq_ptp_s *aq_ptp,
+				struct hwtstamp_config *config);
+int aq_ptp_hwtstamp_config_set(struct aq_ptp_s *aq_ptp,
+			       struct hwtstamp_config *config);
+
+/* Return either ring is belong to PTP or not*/
+bool aq_ptp_ring(struct aq_nic_s *aq_nic, struct aq_ring_s *ring);
+
+u16 aq_ptp_extract_ts(struct aq_nic_s *aq_nic, struct skb_shared_hwtstamps *shhwtstamps, u8 *p,
+		      unsigned int len);
+
+struct ptp_clock *aq_ptp_get_ptp_clock(struct aq_ptp_s *aq_ptp);
+
+int aq_ptp_link_change(struct aq_nic_s *aq_nic);
+
+/* PTP ring statistics */
+int aq_ptp_get_ring_cnt(struct aq_nic_s *aq_nic, const enum atl_ring_type ring_type);
+u64 *aq_ptp_get_stats(struct aq_nic_s *aq_nic, u64 *data);
+
+#else
+
+static inline int aq_ptp_init(struct aq_nic_s *aq_nic, unsigned int idx_vec)
+{
+	return 0;
+}
+
+static inline void aq_ptp_unregister(struct aq_nic_s *aq_nic) {}
+
+static inline void aq_ptp_free(struct aq_nic_s *aq_nic)
+{
+}
+
+static inline int aq_ptp_irq_alloc(struct aq_nic_s *aq_nic)
+{
+	return 0;
+}
+
+static inline void aq_ptp_irq_free(struct aq_nic_s *aq_nic)
+{
+}
+
+static inline int aq_ptp_ring_alloc(struct aq_nic_s *aq_nic)
+{
+	return 0;
+}
+
+static inline void aq_ptp_ring_free(struct aq_nic_s *aq_nic) {}
+
+static inline int aq_ptp_ring_init(struct aq_nic_s *aq_nic)
+{
+	return 0;
+}
+
+static inline int aq_ptp_ring_start(struct aq_nic_s *aq_nic)
+{
+	return 0;
+}
+
+static inline void aq_ptp_ring_stop(struct aq_nic_s *aq_nic) {}
+static inline void aq_ptp_ring_deinit(struct aq_nic_s *aq_nic) {}
+static inline void aq_ptp_service_task(struct aq_nic_s *aq_nic) {}
+static inline void aq_ptp_tm_offset_set(struct aq_nic_s *aq_nic,
+					unsigned int mbps) {}
+static inline void aq_ptp_clock_init(struct aq_nic_s *aq_nic) {}
+static inline int aq_ptp_xmit(struct aq_nic_s *aq_nic, struct sk_buff *skb)
+{
+	return -EOPNOTSUPP;
+}
+
+static inline void aq_ptp_tx_hwtstamp(struct aq_nic_s *aq_nic, u64 timestamp) {}
+static inline void aq_ptp_hwtstamp_config_get(struct aq_ptp_s *aq_ptp,
+					      struct hwtstamp_config *config) {}
+static inline int aq_ptp_hwtstamp_config_set(struct aq_ptp_s *aq_ptp,
+					     struct hwtstamp_config *config)
+{
+	return 0;
+}
+
+static inline bool aq_ptp_ring(struct aq_nic_s *aq_nic, struct aq_ring_s *ring)
+{
+	return false;
+}
+
+static inline u16 aq_ptp_extract_ts(struct aq_nic_s *aq_nic,
+				    struct skb_shared_hwtstamps *shhwtstamps, u8 *p,
+				    unsigned int len)
+{
+	return 0;
+}
+
+static inline struct ptp_clock *aq_ptp_get_ptp_clock(struct aq_ptp_s *aq_ptp)
+{
+	return NULL;
+}
+
+static inline int aq_ptp_link_change(struct aq_nic_s *aq_nic)
+{
+	return 0;
+}
+#endif
+
+#endif /* AQ_PTP_H */
diff --git a/drivers/net/ethernet/aquantia/atlantic/aq_ring.c b/drivers/net/ethernet/aquantia/atlantic/aq_ring.c
index 9a0817938eca..f21de0c21e52 100644
--- a/drivers/net/ethernet/aquantia/atlantic/aq_ring.c
+++ b/drivers/net/ethernet/aquantia/atlantic/aq_ring.c
@@ -1,23 +1,139 @@
-/*
- * aQuantia Corporation Network Driver
- * Copyright (C) 2014-2017 aQuantia Corporation. All rights reserved
+// SPDX-License-Identifier: GPL-2.0-only
+/* Atlantic Network Driver
  *
- * This program is free software; you can redistribute it and/or modify it
- * under the terms and conditions of the GNU General Public License,
- * version 2, as published by the Free Software Foundation.
+ * Copyright (C) 2014-2019 aQuantia Corporation
+ * Copyright (C) 2019-2020 Marvell International Ltd.
  */
 
 /* File aq_ring.c: Definition of functions for Rx/Tx rings. */
 
-#include "aq_ring.h"
 #include "aq_nic.h"
 #include "aq_hw.h"
-
+#include "aq_hw_utils.h"
+#include "aq_ptp.h"
+#include "aq_vec.h"
+#include "aq_main.h"
+
+#include <net/xdp.h>
+#include <linux/filter.h>
+#include <linux/bpf_trace.h>
 #include <linux/netdevice.h>
 #include <linux/etherdevice.h>
 
-static struct aq_ring_s *aq_ring_alloc(struct aq_ring_s *self,
-				       struct aq_nic_s *aq_nic)
+static void aq_get_rxpages_xdp(struct aq_ring_buff_s *buff,
+			       struct xdp_buff *xdp)
+{
+	struct skb_shared_info *sinfo;
+	int i;
+
+	if (xdp_buff_has_frags(xdp)) {
+		sinfo = xdp_get_shared_info_from_buff(xdp);
+
+		for (i = 0; i < sinfo->nr_frags; i++) {
+			skb_frag_t *frag = &sinfo->frags[i];
+
+			page_ref_inc(skb_frag_page(frag));
+		}
+	}
+	page_ref_inc(buff->rxdata.page);
+}
+
+static inline void aq_free_rxpage(struct aq_rxpage *rxpage, struct device *dev)
+{
+	unsigned int len = PAGE_SIZE << rxpage->order;
+
+	dma_unmap_page(dev, rxpage->daddr, len, DMA_FROM_DEVICE);
+
+	/* Drop the ref for being in the ring. */
+	__free_pages(rxpage->page, rxpage->order);
+	rxpage->page = NULL;
+}
+
+static int aq_alloc_rxpages(struct aq_rxpage *rxpage, struct aq_ring_s *rx_ring)
+{
+	struct device *dev = aq_nic_get_dev(rx_ring->aq_nic);
+	unsigned int order = rx_ring->page_order;
+	struct page *page;
+	int ret = -ENOMEM;
+	dma_addr_t daddr;
+
+	page = dev_alloc_pages(order);
+	if (unlikely(!page))
+		goto err_exit;
+
+	daddr = dma_map_page(dev, page, 0, PAGE_SIZE << order,
+			     DMA_FROM_DEVICE);
+
+	if (unlikely(dma_mapping_error(dev, daddr)))
+		goto free_page;
+
+	rxpage->page = page;
+	rxpage->daddr = daddr;
+	rxpage->order = order;
+	rxpage->pg_off = rx_ring->page_offset;
+
+	return 0;
+
+free_page:
+	__free_pages(page, order);
+
+err_exit:
+	return ret;
+}
+
+static int aq_get_rxpages(struct aq_ring_s *self, struct aq_ring_buff_s *rxbuf)
+{
+	unsigned int order = self->page_order;
+	u16 page_offset = self->page_offset;
+	u16 frame_max = self->frame_max;
+	u16 tail_size = self->tail_size;
+	int ret;
+
+	if (rxbuf->rxdata.page) {
+		/* One means ring is the only user and can reuse */
+		if (page_ref_count(rxbuf->rxdata.page) > 1) {
+			/* Try reuse buffer */
+			rxbuf->rxdata.pg_off += frame_max + page_offset +
+						tail_size;
+			if (rxbuf->rxdata.pg_off + frame_max + tail_size <=
+			    (PAGE_SIZE << order)) {
+				u64_stats_update_begin(&self->stats.rx.syncp);
+				self->stats.rx.pg_flips++;
+				u64_stats_update_end(&self->stats.rx.syncp);
+
+			} else {
+				/* Buffer exhausted. We have other users and
+				 * should release this page and realloc
+				 */
+				aq_free_rxpage(&rxbuf->rxdata,
+					       aq_nic_get_dev(self->aq_nic));
+				u64_stats_update_begin(&self->stats.rx.syncp);
+				self->stats.rx.pg_losts++;
+				u64_stats_update_end(&self->stats.rx.syncp);
+			}
+		} else {
+			rxbuf->rxdata.pg_off = page_offset;
+			u64_stats_update_begin(&self->stats.rx.syncp);
+			self->stats.rx.pg_reuses++;
+			u64_stats_update_end(&self->stats.rx.syncp);
+		}
+	}
+
+	if (!rxbuf->rxdata.page) {
+		ret = aq_alloc_rxpages(&rxbuf->rxdata, self);
+		if (ret) {
+			u64_stats_update_begin(&self->stats.rx.syncp);
+			self->stats.rx.alloc_fails++;
+			u64_stats_update_end(&self->stats.rx.syncp);
+		}
+		return ret;
+	}
+
+	return 0;
+}
+
+static int aq_ring_alloc(struct aq_ring_s *self,
+			 struct aq_nic_s *aq_nic)
 {
 	int err = 0;
 
@@ -28,9 +144,10 @@ static struct aq_ring_s *aq_ring_alloc(struct aq_ring_s *self,
 		err = -ENOMEM;
 		goto err_exit;
 	}
+
 	self->dx_ring = dma_alloc_coherent(aq_nic_get_dev(aq_nic),
-						self->size * self->dx_size,
-						&self->dx_ring_pa, GFP_KERNEL);
+					   self->size * self->dx_size,
+					   &self->dx_ring_pa, GFP_KERNEL);
 	if (!self->dx_ring) {
 		err = -ENOMEM;
 		goto err_exit;
@@ -39,289 +156,818 @@ static struct aq_ring_s *aq_ring_alloc(struct aq_ring_s *self,
 err_exit:
 	if (err < 0) {
 		aq_ring_free(self);
-		self = NULL;
 	}
-	return self;
+
+	return err;
 }
 
-struct aq_ring_s *aq_ring_tx_alloc(struct aq_ring_s *self,
-				   struct aq_nic_s *aq_nic,
-				   unsigned int idx,
-				   struct aq_nic_cfg_s *aq_nic_cfg)
+int aq_ring_tx_alloc(struct aq_ring_s *self,
+		     struct aq_nic_s *aq_nic,
+		     unsigned int idx,
+		     struct aq_nic_cfg_s *aq_nic_cfg)
 {
-	int err = 0;
-
 	self->aq_nic = aq_nic;
 	self->idx = idx;
 	self->size = aq_nic_cfg->txds;
 	self->dx_size = aq_nic_cfg->aq_hw_caps->txd_size;
 
-	self = aq_ring_alloc(self, aq_nic);
-	if (!self) {
-		err = -ENOMEM;
-		goto err_exit;
-	}
-
-err_exit:
-	if (err < 0) {
-		aq_ring_free(self);
-		self = NULL;
-	}
-	return self;
+	return aq_ring_alloc(self, aq_nic);
 }
 
-struct aq_ring_s *aq_ring_rx_alloc(struct aq_ring_s *self,
-				   struct aq_nic_s *aq_nic,
-				   unsigned int idx,
-				   struct aq_nic_cfg_s *aq_nic_cfg)
+int aq_ring_rx_alloc(struct aq_ring_s *self,
+		     struct aq_nic_s *aq_nic,
+		     unsigned int idx,
+		     struct aq_nic_cfg_s *aq_nic_cfg)
 {
-	int err = 0;
-
 	self->aq_nic = aq_nic;
 	self->idx = idx;
 	self->size = aq_nic_cfg->rxds;
 	self->dx_size = aq_nic_cfg->aq_hw_caps->rxd_size;
-
-	self = aq_ring_alloc(self, aq_nic);
-	if (!self) {
-		err = -ENOMEM;
-		goto err_exit;
+	self->xdp_prog = aq_nic->xdp_prog;
+	self->frame_max = AQ_CFG_RX_FRAME_MAX;
+
+	/* Only order-2 is allowed if XDP is enabled */
+	if (READ_ONCE(self->xdp_prog)) {
+		self->page_offset = AQ_XDP_HEADROOM;
+		self->page_order = AQ_CFG_XDP_PAGEORDER;
+		self->tail_size = AQ_XDP_TAILROOM;
+	} else {
+		self->page_offset = 0;
+		self->page_order = fls(self->frame_max / PAGE_SIZE +
+				       (self->frame_max % PAGE_SIZE ? 1 : 0)) - 1;
+		if (aq_nic_cfg->rxpageorder > self->page_order)
+			self->page_order = aq_nic_cfg->rxpageorder;
+		self->tail_size = 0;
 	}
 
-err_exit:
-	if (err < 0) {
+	return aq_ring_alloc(self, aq_nic);
+}
+
+int
+aq_ring_hwts_rx_alloc(struct aq_ring_s *self, struct aq_nic_s *aq_nic,
+		      unsigned int idx, unsigned int size, unsigned int dx_size)
+{
+	struct device *dev = aq_nic_get_dev(aq_nic);
+	size_t sz = size * dx_size + AQ_CFG_RXDS_DEF;
+
+	memset(self, 0, sizeof(*self));
+
+	self->aq_nic = aq_nic;
+	self->idx = idx;
+	self->size = size;
+	self->dx_size = dx_size;
+
+	self->dx_ring = dma_alloc_coherent(dev, sz, &self->dx_ring_pa,
+					   GFP_KERNEL);
+	if (!self->dx_ring) {
 		aq_ring_free(self);
-		self = NULL;
+		return -ENOMEM;
 	}
-	return self;
+
+	return 0;
 }
 
-int aq_ring_init(struct aq_ring_s *self)
+int aq_ring_init(struct aq_ring_s *self, const enum atl_ring_type ring_type)
 {
 	self->hw_head = 0;
 	self->sw_head = 0;
 	self->sw_tail = 0;
-	spin_lock_init(&self->header.lock);
+	self->ring_type = ring_type;
+
+	if (self->ring_type == ATL_RING_RX)
+		u64_stats_init(&self->stats.rx.syncp);
+	else
+		u64_stats_init(&self->stats.tx.syncp);
+
 	return 0;
 }
 
-void aq_ring_tx_clean(struct aq_ring_s *self)
+static inline bool aq_ring_dx_in_range(unsigned int h, unsigned int i,
+				       unsigned int t)
+{
+	return (h < t) ? ((h < i) && (i < t)) : ((h < i) || (i < t));
+}
+
+void aq_ring_update_queue_state(struct aq_ring_s *ring)
+{
+	if (aq_ring_avail_dx(ring) <= AQ_CFG_SKB_FRAGS_MAX)
+		aq_ring_queue_stop(ring);
+	else if (aq_ring_avail_dx(ring) > AQ_CFG_RESTART_DESC_THRES)
+		aq_ring_queue_wake(ring);
+}
+
+void aq_ring_queue_wake(struct aq_ring_s *ring)
+{
+	struct net_device *ndev = aq_nic_get_ndev(ring->aq_nic);
+
+	if (__netif_subqueue_stopped(ndev,
+				     AQ_NIC_RING2QMAP(ring->aq_nic,
+						      ring->idx))) {
+		netif_wake_subqueue(ndev,
+				    AQ_NIC_RING2QMAP(ring->aq_nic, ring->idx));
+		u64_stats_update_begin(&ring->stats.tx.syncp);
+		ring->stats.tx.queue_restarts++;
+		u64_stats_update_end(&ring->stats.tx.syncp);
+	}
+}
+
+void aq_ring_queue_stop(struct aq_ring_s *ring)
+{
+	struct net_device *ndev = aq_nic_get_ndev(ring->aq_nic);
+
+	if (!__netif_subqueue_stopped(ndev,
+				      AQ_NIC_RING2QMAP(ring->aq_nic,
+						       ring->idx)))
+		netif_stop_subqueue(ndev,
+				    AQ_NIC_RING2QMAP(ring->aq_nic, ring->idx));
+}
+
+bool aq_ring_tx_clean(struct aq_ring_s *self)
 {
 	struct device *dev = aq_nic_get_dev(self->aq_nic);
+	unsigned int budget;
 
-	for (; self->sw_head != self->hw_head;
-		self->sw_head = aq_ring_next_dx(self, self->sw_head)) {
+	for (budget = AQ_CFG_TX_CLEAN_BUDGET;
+	     budget && self->sw_head != self->hw_head; budget--) {
 		struct aq_ring_buff_s *buff = &self->buff_ring[self->sw_head];
 
 		if (likely(buff->is_mapped)) {
-			if (unlikely(buff->is_sop))
+			if (unlikely(buff->is_sop)) {
+				if (!buff->is_eop &&
+				    buff->eop_index != 0xffffU &&
+				    (!aq_ring_dx_in_range(self->sw_head,
+						buff->eop_index,
+						self->hw_head)))
+					break;
+
 				dma_unmap_single(dev, buff->pa, buff->len,
 						 DMA_TO_DEVICE);
-			else
+			} else {
 				dma_unmap_page(dev, buff->pa, buff->len,
 					       DMA_TO_DEVICE);
+			}
 		}
 
-		if (unlikely(buff->is_eop))
+		if (likely(!buff->is_eop))
+			goto out;
+
+		if (buff->skb) {
+			u64_stats_update_begin(&self->stats.tx.syncp);
+			++self->stats.tx.packets;
+			self->stats.tx.bytes += buff->skb->len;
+			u64_stats_update_end(&self->stats.tx.syncp);
 			dev_kfree_skb_any(buff->skb);
+		} else if (buff->xdpf) {
+			u64_stats_update_begin(&self->stats.tx.syncp);
+			++self->stats.tx.packets;
+			self->stats.tx.bytes += xdp_get_frame_len(buff->xdpf);
+			u64_stats_update_end(&self->stats.tx.syncp);
+			xdp_return_frame_rx_napi(buff->xdpf);
+		}
+
+out:
+		buff->skb = NULL;
+		buff->xdpf = NULL;
+		buff->pa = 0U;
+		buff->eop_index = 0xffffU;
+		self->sw_head = aq_ring_next_dx(self, self->sw_head);
 	}
+
+	return !!budget;
 }
 
-static inline unsigned int aq_ring_dx_in_range(unsigned int h, unsigned int i,
-					       unsigned int t)
+static void aq_rx_checksum(struct aq_ring_s *self,
+			   struct aq_ring_buff_s *buff,
+			   struct sk_buff *skb)
 {
-	return (h < t) ? ((h < i) && (i < t)) : ((h < i) || (i < t));
+	if (!(self->aq_nic->ndev->features & NETIF_F_RXCSUM))
+		return;
+
+	if (unlikely(buff->is_cso_err)) {
+		u64_stats_update_begin(&self->stats.rx.syncp);
+		++self->stats.rx.errors;
+		u64_stats_update_end(&self->stats.rx.syncp);
+		skb->ip_summed = CHECKSUM_NONE;
+		return;
+	}
+	if (buff->is_ip_cso) {
+		__skb_incr_checksum_unnecessary(skb);
+	} else {
+		skb->ip_summed = CHECKSUM_NONE;
+	}
+
+	if (buff->is_udp_cso || buff->is_tcp_cso)
+		__skb_incr_checksum_unnecessary(skb);
+}
+
+int aq_xdp_xmit(struct net_device *dev, int num_frames,
+		struct xdp_frame **frames, u32 flags)
+{
+	struct aq_nic_s *aq_nic = netdev_priv(dev);
+	unsigned int vec, i, drop = 0;
+	int cpu = smp_processor_id();
+	struct aq_nic_cfg_s *aq_cfg;
+	struct aq_ring_s *ring;
+
+	aq_cfg = aq_nic_get_cfg(aq_nic);
+	vec = cpu % aq_cfg->vecs;
+	ring = aq_nic->aq_ring_tx[AQ_NIC_CFG_TCVEC2RING(aq_cfg, 0, vec)];
+
+	for (i = 0; i < num_frames; i++) {
+		struct xdp_frame *xdpf = frames[i];
+
+		if (aq_nic_xmit_xdpf(aq_nic, ring, xdpf) == NETDEV_TX_BUSY)
+			drop++;
+	}
+
+	return num_frames - drop;
+}
+
+static struct sk_buff *aq_xdp_build_skb(struct xdp_buff *xdp,
+					struct net_device *dev,
+					struct aq_ring_buff_s *buff)
+{
+	struct xdp_frame *xdpf;
+	struct sk_buff *skb;
+
+	xdpf = xdp_convert_buff_to_frame(xdp);
+	if (unlikely(!xdpf))
+		return NULL;
+
+	skb = xdp_build_skb_from_frame(xdpf, dev);
+	if (!skb)
+		return NULL;
+
+	aq_get_rxpages_xdp(buff, xdp);
+	return skb;
 }
 
-#define AQ_SKB_ALIGN SKB_DATA_ALIGN(sizeof(struct skb_shared_info))
-int aq_ring_rx_clean(struct aq_ring_s *self, int *work_done, int budget)
+static struct sk_buff *aq_xdp_run_prog(struct aq_nic_s *aq_nic,
+				       struct xdp_buff *xdp,
+				       struct aq_ring_s *rx_ring,
+				       struct aq_ring_buff_s *buff)
+{
+	int result = NETDEV_TX_BUSY;
+	struct aq_ring_s *tx_ring;
+	struct xdp_frame *xdpf;
+	struct bpf_prog *prog;
+	u32 act = XDP_ABORTED;
+	struct sk_buff *skb;
+
+	u64_stats_update_begin(&rx_ring->stats.rx.syncp);
+	++rx_ring->stats.rx.packets;
+	rx_ring->stats.rx.bytes += xdp_get_buff_len(xdp);
+	u64_stats_update_end(&rx_ring->stats.rx.syncp);
+
+	prog = READ_ONCE(rx_ring->xdp_prog);
+	if (!prog)
+		return aq_xdp_build_skb(xdp, aq_nic->ndev, buff);
+
+	prefetchw(xdp->data_hard_start); /* xdp_frame write */
+
+	/* single buffer XDP program, but packet is multi buffer, aborted */
+	if (xdp_buff_has_frags(xdp) && !prog->aux->xdp_has_frags)
+		goto out_aborted;
+
+	act = bpf_prog_run_xdp(prog, xdp);
+	switch (act) {
+	case XDP_PASS:
+		skb = aq_xdp_build_skb(xdp, aq_nic->ndev, buff);
+		if (!skb)
+			goto out_aborted;
+		u64_stats_update_begin(&rx_ring->stats.rx.syncp);
+		++rx_ring->stats.rx.xdp_pass;
+		u64_stats_update_end(&rx_ring->stats.rx.syncp);
+		return skb;
+	case XDP_TX:
+		xdpf = xdp_convert_buff_to_frame(xdp);
+		if (unlikely(!xdpf))
+			goto out_aborted;
+		tx_ring = aq_nic->aq_ring_tx[rx_ring->idx];
+		result = aq_nic_xmit_xdpf(aq_nic, tx_ring, xdpf);
+		if (result == NETDEV_TX_BUSY)
+			goto out_aborted;
+		u64_stats_update_begin(&rx_ring->stats.rx.syncp);
+		++rx_ring->stats.rx.xdp_tx;
+		u64_stats_update_end(&rx_ring->stats.rx.syncp);
+		aq_get_rxpages_xdp(buff, xdp);
+		break;
+	case XDP_REDIRECT:
+		if (xdp_do_redirect(aq_nic->ndev, xdp, prog) < 0)
+			goto out_aborted;
+		xdp_do_flush();
+		u64_stats_update_begin(&rx_ring->stats.rx.syncp);
+		++rx_ring->stats.rx.xdp_redirect;
+		u64_stats_update_end(&rx_ring->stats.rx.syncp);
+		aq_get_rxpages_xdp(buff, xdp);
+		break;
+	default:
+		fallthrough;
+	case XDP_ABORTED:
+out_aborted:
+		u64_stats_update_begin(&rx_ring->stats.rx.syncp);
+		++rx_ring->stats.rx.xdp_aborted;
+		u64_stats_update_end(&rx_ring->stats.rx.syncp);
+		trace_xdp_exception(aq_nic->ndev, prog, act);
+		bpf_warn_invalid_xdp_action(aq_nic->ndev, prog, act);
+		break;
+	case XDP_DROP:
+		u64_stats_update_begin(&rx_ring->stats.rx.syncp);
+		++rx_ring->stats.rx.xdp_drop;
+		u64_stats_update_end(&rx_ring->stats.rx.syncp);
+		break;
+	}
+
+	return ERR_PTR(-result);
+}
+
+static bool aq_add_rx_fragment(struct device *dev,
+			       struct aq_ring_s *ring,
+			       struct aq_ring_buff_s *buff,
+			       struct xdp_buff *xdp)
+{
+	struct skb_shared_info *sinfo = xdp_get_shared_info_from_buff(xdp);
+	struct aq_ring_buff_s *buff_ = buff;
+
+	memset(sinfo, 0, sizeof(*sinfo));
+	do {
+		skb_frag_t *frag;
+
+		if (unlikely(sinfo->nr_frags >= MAX_SKB_FRAGS))
+			return true;
+
+		frag = &sinfo->frags[sinfo->nr_frags++];
+		buff_ = &ring->buff_ring[buff_->next];
+		dma_sync_single_range_for_cpu(dev,
+					      buff_->rxdata.daddr,
+					      buff_->rxdata.pg_off,
+					      buff_->len,
+					      DMA_FROM_DEVICE);
+		sinfo->xdp_frags_size += buff_->len;
+		skb_frag_fill_page_desc(frag, buff_->rxdata.page,
+					buff_->rxdata.pg_off,
+					buff_->len);
+
+		buff_->is_cleaned = 1;
+
+		buff->is_ip_cso &= buff_->is_ip_cso;
+		buff->is_udp_cso &= buff_->is_udp_cso;
+		buff->is_tcp_cso &= buff_->is_tcp_cso;
+		buff->is_cso_err |= buff_->is_cso_err;
+
+		if (page_is_pfmemalloc(buff_->rxdata.page))
+			xdp_buff_set_frag_pfmemalloc(xdp);
+
+	} while (!buff_->is_eop);
+
+	xdp_buff_set_frags_flag(xdp);
+
+	return false;
+}
+
+static int __aq_ring_rx_clean(struct aq_ring_s *self, struct napi_struct *napi,
+			      int *work_done, int budget)
 {
 	struct net_device *ndev = aq_nic_get_ndev(self->aq_nic);
 	int err = 0;
-	bool is_rsc_completed = true;
 
 	for (; (self->sw_head != self->hw_head) && budget;
 		self->sw_head = aq_ring_next_dx(self, self->sw_head),
 		--budget, ++(*work_done)) {
 		struct aq_ring_buff_s *buff = &self->buff_ring[self->sw_head];
+		bool is_ptp_ring = aq_ptp_ring(self->aq_nic, self);
+		struct aq_ring_buff_s *buff_ = NULL;
 		struct sk_buff *skb = NULL;
 		unsigned int next_ = 0U;
 		unsigned int i = 0U;
-		struct aq_ring_buff_s *buff_ = NULL;
-
-		if (buff->is_error) {
-			__free_pages(buff->page, 0);
-			continue;
-		}
+		u16 hdr_len;
 
 		if (buff->is_cleaned)
 			continue;
 
 		if (!buff->is_eop) {
-			for (next_ = buff->next,
-			     buff_ = &self->buff_ring[next_]; true;
-			     next_ = buff_->next,
-			     buff_ = &self->buff_ring[next_]) {
+			unsigned int frag_cnt = 0U;
+			buff_ = buff;
+			do {
+				bool is_rsc_completed = true;
+
+				if (buff_->next >= self->size) {
+					err = -EIO;
+					goto err_exit;
+				}
+
+				frag_cnt++;
+				next_ = buff_->next;
+				buff_ = &self->buff_ring[next_];
 				is_rsc_completed =
 					aq_ring_dx_in_range(self->sw_head,
 							    next_,
 							    self->hw_head);
 
-				if (unlikely(!is_rsc_completed)) {
-					is_rsc_completed = false;
-					break;
+				if (unlikely(!is_rsc_completed) ||
+						frag_cnt > MAX_SKB_FRAGS) {
+					err = 0;
+					goto err_exit;
 				}
 
-				if (buff_->is_eop)
-					break;
+				buff->is_error |= buff_->is_error;
+				buff->is_cso_err |= buff_->is_cso_err;
+
+			} while (!buff_->is_eop);
+
+			if (buff->is_error ||
+			    (buff->is_lro && buff->is_cso_err)) {
+				buff_ = buff;
+				do {
+					if (buff_->next >= self->size) {
+						err = -EIO;
+						goto err_exit;
+					}
+					next_ = buff_->next;
+					buff_ = &self->buff_ring[next_];
+
+					buff_->is_cleaned = true;
+				} while (!buff_->is_eop);
+
+				u64_stats_update_begin(&self->stats.rx.syncp);
+				++self->stats.rx.errors;
+				u64_stats_update_end(&self->stats.rx.syncp);
+				continue;
 			}
+		}
 
-			if (!is_rsc_completed) {
-				err = 0;
-				goto err_exit;
-			}
+		if (buff->is_error) {
+			u64_stats_update_begin(&self->stats.rx.syncp);
+			++self->stats.rx.errors;
+			u64_stats_update_end(&self->stats.rx.syncp);
+			continue;
 		}
 
-		/* for single fragment packets use build_skb() */
-		if (buff->is_eop) {
-			skb = build_skb(page_address(buff->page),
-					buff->len + AQ_SKB_ALIGN);
-			if (unlikely(!skb)) {
-				err = -ENOMEM;
-				goto err_exit;
-			}
+		dma_sync_single_range_for_cpu(aq_nic_get_dev(self->aq_nic),
+					      buff->rxdata.daddr,
+					      buff->rxdata.pg_off,
+					      buff->len, DMA_FROM_DEVICE);
 
-			skb_put(skb, buff->len);
-		} else {
-			skb = netdev_alloc_skb(ndev, ETH_HLEN);
-			if (unlikely(!skb)) {
-				err = -ENOMEM;
-				goto err_exit;
-			}
-			skb_put(skb, ETH_HLEN);
-			memcpy(skb->data, page_address(buff->page), ETH_HLEN);
-
-			skb_add_rx_frag(skb, 0, buff->page, ETH_HLEN,
-					buff->len - ETH_HLEN,
-					SKB_TRUESIZE(buff->len - ETH_HLEN));
-
-			for (i = 1U, next_ = buff->next,
-			     buff_ = &self->buff_ring[next_]; true;
-			     next_ = buff_->next,
-			     buff_ = &self->buff_ring[next_], ++i) {
-				skb_add_rx_frag(skb, i, buff_->page, 0,
+		skb = napi_alloc_skb(napi, AQ_CFG_RX_HDR_SIZE);
+		if (unlikely(!skb)) {
+			u64_stats_update_begin(&self->stats.rx.syncp);
+			self->stats.rx.skb_alloc_fails++;
+			u64_stats_update_end(&self->stats.rx.syncp);
+			err = -ENOMEM;
+			goto err_exit;
+		}
+		if (is_ptp_ring)
+			buff->len -=
+				aq_ptp_extract_ts(self->aq_nic, skb_hwtstamps(skb),
+						  aq_buf_vaddr(&buff->rxdata),
+						  buff->len);
+
+		hdr_len = buff->len;
+		if (hdr_len > AQ_CFG_RX_HDR_SIZE)
+			hdr_len = eth_get_headlen(skb->dev,
+						  aq_buf_vaddr(&buff->rxdata),
+						  AQ_CFG_RX_HDR_SIZE);
+
+		memcpy(__skb_put(skb, hdr_len), aq_buf_vaddr(&buff->rxdata),
+		       ALIGN(hdr_len, sizeof(long)));
+
+		if (buff->len - hdr_len > 0) {
+			skb_add_rx_frag(skb, i++, buff->rxdata.page,
+					buff->rxdata.pg_off + hdr_len,
+					buff->len - hdr_len,
+					self->frame_max);
+			page_ref_inc(buff->rxdata.page);
+		}
+
+		if (!buff->is_eop) {
+			buff_ = buff;
+			do {
+				next_ = buff_->next;
+				buff_ = &self->buff_ring[next_];
+
+				dma_sync_single_range_for_cpu(aq_nic_get_dev(self->aq_nic),
+							      buff_->rxdata.daddr,
+							      buff_->rxdata.pg_off,
+							      buff_->len,
+							      DMA_FROM_DEVICE);
+				skb_add_rx_frag(skb, i++,
+						buff_->rxdata.page,
+						buff_->rxdata.pg_off,
 						buff_->len,
-						SKB_TRUESIZE(buff->len -
-						ETH_HLEN));
+						self->frame_max);
+				page_ref_inc(buff_->rxdata.page);
 				buff_->is_cleaned = 1;
 
-				if (buff_->is_eop)
+				buff->is_ip_cso &= buff_->is_ip_cso;
+				buff->is_udp_cso &= buff_->is_udp_cso;
+				buff->is_tcp_cso &= buff_->is_tcp_cso;
+				buff->is_cso_err |= buff_->is_cso_err;
+
+			} while (!buff_->is_eop);
+		}
+
+		if (buff->is_vlan)
+			__vlan_hwaccel_put_tag(skb, htons(ETH_P_8021Q),
+					       buff->vlan_rx_tag);
+
+		skb->protocol = eth_type_trans(skb, ndev);
+
+		aq_rx_checksum(self, buff, skb);
+
+		skb_set_hash(skb, buff->rss_hash,
+			     buff->is_hash_l4 ? PKT_HASH_TYPE_L4 :
+			     PKT_HASH_TYPE_NONE);
+		/* Send all PTP traffic to 0 queue */
+		skb_record_rx_queue(skb,
+				    is_ptp_ring ? 0
+						: AQ_NIC_RING2QMAP(self->aq_nic,
+								   self->idx));
+
+		u64_stats_update_begin(&self->stats.rx.syncp);
+		++self->stats.rx.packets;
+		self->stats.rx.bytes += skb->len;
+		u64_stats_update_end(&self->stats.rx.syncp);
+
+		napi_gro_receive(napi, skb);
+	}
+
+err_exit:
+	return err;
+}
+
+static int __aq_ring_xdp_clean(struct aq_ring_s *rx_ring,
+			       struct napi_struct *napi, int *work_done,
+			       int budget)
+{
+	int frame_sz = rx_ring->page_offset + rx_ring->frame_max +
+		       rx_ring->tail_size;
+	struct aq_nic_s *aq_nic = rx_ring->aq_nic;
+	bool is_rsc_completed = true;
+	struct device *dev;
+	int err = 0;
+
+	dev = aq_nic_get_dev(aq_nic);
+	for (; (rx_ring->sw_head != rx_ring->hw_head) && budget;
+		rx_ring->sw_head = aq_ring_next_dx(rx_ring, rx_ring->sw_head),
+		--budget, ++(*work_done)) {
+		struct aq_ring_buff_s *buff = &rx_ring->buff_ring[rx_ring->sw_head];
+		bool is_ptp_ring = aq_ptp_ring(rx_ring->aq_nic, rx_ring);
+		struct aq_ring_buff_s *buff_ = NULL;
+		u16 ptp_hwtstamp_len = 0;
+		struct skb_shared_hwtstamps shhwtstamps;
+		struct sk_buff *skb = NULL;
+		unsigned int next_ = 0U;
+		struct xdp_buff xdp;
+		void *hard_start;
+
+		if (buff->is_cleaned)
+			continue;
+
+		if (!buff->is_eop) {
+			buff_ = buff;
+			do {
+				if (buff_->next >= rx_ring->size) {
+					err = -EIO;
+					goto err_exit;
+				}
+				next_ = buff_->next;
+				buff_ = &rx_ring->buff_ring[next_];
+				is_rsc_completed =
+					aq_ring_dx_in_range(rx_ring->sw_head,
+							    next_,
+							    rx_ring->hw_head);
+
+				if (unlikely(!is_rsc_completed))
 					break;
+
+				buff->is_error |= buff_->is_error;
+				buff->is_cso_err |= buff_->is_cso_err;
+			} while (!buff_->is_eop);
+
+			if (!is_rsc_completed) {
+				err = 0;
+				goto err_exit;
+			}
+			if (buff->is_error ||
+			    (buff->is_lro && buff->is_cso_err)) {
+				buff_ = buff;
+				do {
+					if (buff_->next >= rx_ring->size) {
+						err = -EIO;
+						goto err_exit;
+					}
+					next_ = buff_->next;
+					buff_ = &rx_ring->buff_ring[next_];
+
+					buff_->is_cleaned = true;
+				} while (!buff_->is_eop);
+
+				u64_stats_update_begin(&rx_ring->stats.rx.syncp);
+				++rx_ring->stats.rx.errors;
+				u64_stats_update_end(&rx_ring->stats.rx.syncp);
+				continue;
 			}
 		}
 
-		skb->protocol = eth_type_trans(skb, ndev);
-		if (unlikely(buff->is_cso_err)) {
-			++self->stats.rx.errors;
-			skb->ip_summed = CHECKSUM_NONE;
-		} else {
-			if (buff->is_ip_cso) {
-				__skb_incr_checksum_unnecessary(skb);
-				if (buff->is_udp_cso || buff->is_tcp_cso)
-					__skb_incr_checksum_unnecessary(skb);
-			} else {
-				skb->ip_summed = CHECKSUM_NONE;
+		if (buff->is_error) {
+			u64_stats_update_begin(&rx_ring->stats.rx.syncp);
+			++rx_ring->stats.rx.errors;
+			u64_stats_update_end(&rx_ring->stats.rx.syncp);
+			continue;
+		}
+
+		dma_sync_single_range_for_cpu(dev,
+					      buff->rxdata.daddr,
+					      buff->rxdata.pg_off,
+					      buff->len, DMA_FROM_DEVICE);
+		hard_start = page_address(buff->rxdata.page) +
+			     buff->rxdata.pg_off - rx_ring->page_offset;
+
+		if (is_ptp_ring) {
+			ptp_hwtstamp_len = aq_ptp_extract_ts(rx_ring->aq_nic, &shhwtstamps,
+							     aq_buf_vaddr(&buff->rxdata),
+							     buff->len);
+			buff->len -= ptp_hwtstamp_len;
+		}
+
+		xdp_init_buff(&xdp, frame_sz, &rx_ring->xdp_rxq);
+		xdp_prepare_buff(&xdp, hard_start, rx_ring->page_offset,
+				 buff->len, false);
+		if (!buff->is_eop) {
+			if (aq_add_rx_fragment(dev, rx_ring, buff, &xdp)) {
+				u64_stats_update_begin(&rx_ring->stats.rx.syncp);
+				++rx_ring->stats.rx.packets;
+				rx_ring->stats.rx.bytes += xdp_get_buff_len(&xdp);
+				++rx_ring->stats.rx.xdp_aborted;
+				u64_stats_update_end(&rx_ring->stats.rx.syncp);
+				continue;
 			}
 		}
 
+		skb = aq_xdp_run_prog(aq_nic, &xdp, rx_ring, buff);
+		if (IS_ERR(skb) || !skb)
+			continue;
+
+		if (ptp_hwtstamp_len > 0)
+			*skb_hwtstamps(skb) = shhwtstamps;
+
+		if (buff->is_vlan)
+			__vlan_hwaccel_put_tag(skb, htons(ETH_P_8021Q),
+					       buff->vlan_rx_tag);
+
+		aq_rx_checksum(rx_ring, buff, skb);
+
 		skb_set_hash(skb, buff->rss_hash,
 			     buff->is_hash_l4 ? PKT_HASH_TYPE_L4 :
 			     PKT_HASH_TYPE_NONE);
+		/* Send all PTP traffic to 0 queue */
+		skb_record_rx_queue(skb,
+				    is_ptp_ring ? 0
+						: AQ_NIC_RING2QMAP(rx_ring->aq_nic,
+								   rx_ring->idx));
 
-		skb_record_rx_queue(skb, self->idx);
-
-		netif_receive_skb(skb);
-
-		++self->stats.rx.packets;
-		self->stats.rx.bytes += skb->len;
+		napi_gro_receive(napi, skb);
 	}
 
 err_exit:
 	return err;
 }
 
+int aq_ring_rx_clean(struct aq_ring_s *self,
+		     struct napi_struct *napi,
+		     int *work_done,
+		     int budget)
+{
+	if (static_branch_unlikely(&aq_xdp_locking_key))
+		return __aq_ring_xdp_clean(self, napi, work_done, budget);
+	else
+		return __aq_ring_rx_clean(self, napi, work_done, budget);
+}
+
+void aq_ring_hwts_rx_clean(struct aq_ring_s *self, struct aq_nic_s *aq_nic)
+{
+#if IS_REACHABLE(CONFIG_PTP_1588_CLOCK)
+	while (self->sw_head != self->hw_head) {
+		u64 ns;
+
+		aq_nic->aq_hw_ops->extract_hwts(aq_nic->aq_hw,
+						self->dx_ring +
+						(self->sw_head * self->dx_size),
+						self->dx_size, &ns);
+		aq_ptp_tx_hwtstamp(aq_nic, ns);
+
+		self->sw_head = aq_ring_next_dx(self, self->sw_head);
+	}
+#endif
+}
+
 int aq_ring_rx_fill(struct aq_ring_s *self)
 {
-	unsigned int pages_order = fls(AQ_CFG_RX_FRAME_MAX / PAGE_SIZE +
-		(AQ_CFG_RX_FRAME_MAX % PAGE_SIZE ? 1 : 0)) - 1;
 	struct aq_ring_buff_s *buff = NULL;
 	int err = 0;
 	int i = 0;
 
+	if (aq_ring_avail_dx(self) < min_t(unsigned int, AQ_CFG_RX_REFILL_THRES,
+					   self->size / 2))
+		return err;
+
 	for (i = aq_ring_avail_dx(self); i--;
 		self->sw_tail = aq_ring_next_dx(self, self->sw_tail)) {
 		buff = &self->buff_ring[self->sw_tail];
 
 		buff->flags = 0U;
-		buff->len = AQ_CFG_RX_FRAME_MAX;
-
-		buff->page = alloc_pages(GFP_ATOMIC | __GFP_COLD |
-					 __GFP_COMP, pages_order);
-		if (!buff->page) {
-			err = -ENOMEM;
-			goto err_exit;
-		}
+		buff->len = self->frame_max;
 
-		buff->pa = dma_map_page(aq_nic_get_dev(self->aq_nic),
-					buff->page, 0,
-					AQ_CFG_RX_FRAME_MAX, DMA_FROM_DEVICE);
-
-		if (dma_mapping_error(aq_nic_get_dev(self->aq_nic), buff->pa)) {
-			err = -ENOMEM;
+		err = aq_get_rxpages(self, buff);
+		if (err)
 			goto err_exit;
-		}
 
+		buff->pa = aq_buf_daddr(&buff->rxdata);
 		buff = NULL;
 	}
 
 err_exit:
-	if (err < 0) {
-		if (buff && buff->page)
-			__free_pages(buff->page, 0);
-	}
-
 	return err;
 }
 
 void aq_ring_rx_deinit(struct aq_ring_s *self)
 {
 	if (!self)
-		goto err_exit;
+		return;
 
 	for (; self->sw_head != self->sw_tail;
 		self->sw_head = aq_ring_next_dx(self, self->sw_head)) {
 		struct aq_ring_buff_s *buff = &self->buff_ring[self->sw_head];
 
-		dma_unmap_page(aq_nic_get_dev(self->aq_nic), buff->pa,
-			       AQ_CFG_RX_FRAME_MAX, DMA_FROM_DEVICE);
-
-		__free_pages(buff->page, 0);
+		aq_free_rxpage(&buff->rxdata, aq_nic_get_dev(self->aq_nic));
 	}
-
-err_exit:;
 }
 
 void aq_ring_free(struct aq_ring_s *self)
 {
 	if (!self)
-		goto err_exit;
+		return;
 
 	kfree(self->buff_ring);
+	self->buff_ring = NULL;
 
-	if (self->dx_ring)
+	if (self->dx_ring) {
 		dma_free_coherent(aq_nic_get_dev(self->aq_nic),
 				  self->size * self->dx_size, self->dx_ring,
 				  self->dx_ring_pa);
+		self->dx_ring = NULL;
+	}
+}
+
+void aq_ring_hwts_rx_free(struct aq_ring_s *self)
+{
+	if (!self)
+		return;
+
+	if (self->dx_ring) {
+		dma_free_coherent(aq_nic_get_dev(self->aq_nic),
+				  self->size * self->dx_size + AQ_CFG_RXDS_DEF,
+				  self->dx_ring, self->dx_ring_pa);
+		self->dx_ring = NULL;
+	}
+}
+
+unsigned int aq_ring_fill_stats_data(struct aq_ring_s *self, u64 *data)
+{
+	unsigned int count;
+	unsigned int start;
+
+	if (self->ring_type == ATL_RING_RX) {
+		/* This data should mimic aq_ethtool_queue_rx_stat_names structure */
+		do {
+			count = 0;
+			start = u64_stats_fetch_begin(&self->stats.rx.syncp);
+			data[count] = self->stats.rx.packets;
+			data[++count] = self->stats.rx.jumbo_packets;
+			data[++count] = self->stats.rx.lro_packets;
+			data[++count] = self->stats.rx.errors;
+			data[++count] = self->stats.rx.alloc_fails;
+			data[++count] = self->stats.rx.skb_alloc_fails;
+			data[++count] = self->stats.rx.polls;
+			data[++count] = self->stats.rx.pg_flips;
+			data[++count] = self->stats.rx.pg_reuses;
+			data[++count] = self->stats.rx.pg_losts;
+			data[++count] = self->stats.rx.xdp_aborted;
+			data[++count] = self->stats.rx.xdp_drop;
+			data[++count] = self->stats.rx.xdp_pass;
+			data[++count] = self->stats.rx.xdp_tx;
+			data[++count] = self->stats.rx.xdp_invalid;
+			data[++count] = self->stats.rx.xdp_redirect;
+		} while (u64_stats_fetch_retry(&self->stats.rx.syncp, start));
+	} else {
+		/* This data should mimic aq_ethtool_queue_tx_stat_names structure */
+		do {
+			count = 0;
+			start = u64_stats_fetch_begin(&self->stats.tx.syncp);
+			data[count] = self->stats.tx.packets;
+			data[++count] = self->stats.tx.queue_restarts;
+		} while (u64_stats_fetch_retry(&self->stats.tx.syncp, start));
+	}
 
-err_exit:;
+	return ++count;
 }
diff --git a/drivers/net/ethernet/aquantia/atlantic/aq_ring.h b/drivers/net/ethernet/aquantia/atlantic/aq_ring.h
index eecd6d1c4d73..d627ace850ff 100644
--- a/drivers/net/ethernet/aquantia/atlantic/aq_ring.h
+++ b/drivers/net/ethernet/aquantia/atlantic/aq_ring.h
@@ -1,10 +1,8 @@
-/*
- * aQuantia Corporation Network Driver
- * Copyright (C) 2014-2017 aQuantia Corporation. All rights reserved
+/* SPDX-License-Identifier: GPL-2.0-only */
+/* Atlantic Network Driver
  *
- * This program is free software; you can redistribute it and/or modify it
- * under the terms and conditions of the GNU General Public License,
- * version 2, as published by the Free Software Foundation.
+ * Copyright (C) 2014-2019 aQuantia Corporation
+ * Copyright (C) 2019-2020 Marvell International Ltd.
  */
 
 /* File aq_ring.h: Declaration of functions for Rx/Tx rings. */
@@ -13,8 +11,20 @@
 #define AQ_RING_H
 
 #include "aq_common.h"
+#include "aq_vec.h"
+
+#define AQ_XDP_HEADROOM		ALIGN(max(NET_SKB_PAD, XDP_PACKET_HEADROOM), 8)
+#define AQ_XDP_TAILROOM		SKB_DATA_ALIGN(sizeof(struct skb_shared_info))
 
 struct page;
+struct aq_nic_cfg_s;
+
+struct aq_rxpage {
+	struct page *page;
+	dma_addr_t daddr;
+	unsigned int order;
+	unsigned int pg_off;
+};
 
 /*           TxC       SOP        DX         EOP
  *         +----------+----------+----------+-----------
@@ -22,7 +32,7 @@ struct page;
  *         +----------+----------+----------+-----------
  * 4/8bytes|len pkt   |len pkt   |          | skb
  *         +----------+----------+----------+-----------
- * 4/8bytes|is_txc    |len,flags |len       |len,is_eop
+ * 4/8bytes|is_gso    |len,flags |len       |len,is_eop
  *         +----------+----------+----------+-----------
  *
  *  This aq_ring_buff_s doesn't have endianness dependency.
@@ -30,27 +40,22 @@ struct page;
  */
 struct __packed aq_ring_buff_s {
 	union {
+		/* RX/TX */
+		dma_addr_t pa;
 		/* RX */
 		struct {
 			u32 rss_hash;
 			u16 next;
 			u8 is_hash_l4;
 			u8 rsvd1;
-			struct page *page;
+			struct aq_rxpage rxdata;
+			u16 vlan_rx_tag;
 		};
 		/* EOP */
 		struct {
 			dma_addr_t pa_eop;
 			struct sk_buff *skb;
-		};
-		/* DX */
-		struct {
-			dma_addr_t pa;
-		};
-		/* SOP */
-		struct {
-			dma_addr_t pa_sop;
-			u32 len_pkt_sop;
+			struct xdp_frame *xdpf;
 		};
 		/* TxC */
 		struct {
@@ -61,6 +66,7 @@ struct __packed aq_ring_buff_s {
 			u8 is_ipv6:1;
 			u8 rsvd2:7;
 			u32 len_pkt;
+			u16 vlan_tx_tag;
 		};
 	};
 	union {
@@ -72,28 +78,48 @@ struct __packed aq_ring_buff_s {
 			u32 is_cso_err:1;
 			u32 is_sop:1;
 			u32 is_eop:1;
-			u32 is_txc:1;
+			u32 is_gso_tcp:1;
+			u32 is_gso_udp:1;
 			u32 is_mapped:1;
 			u32 is_cleaned:1;
 			u32 is_error:1;
-			u32 rsvd3:6;
+			u32 is_vlan:1;
+			u32 is_lro:1;
+			u32 rsvd3:3;
+			u16 eop_index;
+			u16 rsvd4;
 		};
-		u32 flags;
+		u64 flags;
 	};
 };
 
 struct aq_ring_stats_rx_s {
+	struct u64_stats_sync syncp;	/* must be first */
 	u64 errors;
 	u64 packets;
 	u64 bytes;
 	u64 lro_packets;
 	u64 jumbo_packets;
+	u64 alloc_fails;
+	u64 skb_alloc_fails;
+	u64 polls;
+	u64 pg_losts;
+	u64 pg_flips;
+	u64 pg_reuses;
+	u64 xdp_aborted;
+	u64 xdp_drop;
+	u64 xdp_pass;
+	u64 xdp_tx;
+	u64 xdp_invalid;
+	u64 xdp_redirect;
 };
 
 struct aq_ring_stats_tx_s {
+	struct u64_stats_sync syncp;	/* must be first */
 	u64 errors;
 	u64 packets;
 	u64 bytes;
+	u64 queue_restarts;
 };
 
 union aq_ring_stats_s {
@@ -101,8 +127,12 @@ union aq_ring_stats_s {
 	struct aq_ring_stats_tx_s tx;
 };
 
+enum atl_ring_type {
+	ATL_RING_TX,
+	ATL_RING_RX,
+};
+
 struct aq_ring_s {
-	struct aq_obj_s header;
 	struct aq_ring_buff_s *buff_ring;
 	u8 *dx_ring;		/* descriptors ring, dma shared mem */
 	struct aq_nic_s *aq_nic;
@@ -113,8 +143,15 @@ struct aq_ring_s {
 	unsigned int size;	/* descriptors number */
 	unsigned int dx_size;	/* TX or RX descriptor size,  */
 				/* stored here for fater math */
+	u16 page_order;
+	u16 page_offset;
+	u16 frame_max;
+	u16 tail_size;
 	union aq_ring_stats_s stats;
 	dma_addr_t dx_ring_pa;
+	struct bpf_prog *xdp_prog;
+	enum atl_ring_type ring_type;
+	struct xdp_rxq_info xdp_rxq;
 };
 
 struct aq_ring_param_s {
@@ -123,6 +160,16 @@ struct aq_ring_param_s {
 	cpumask_t affinity_mask;
 };
 
+static inline void *aq_buf_vaddr(struct aq_rxpage *rxpage)
+{
+	return page_to_virt(rxpage->page) + rxpage->pg_off;
+}
+
+static inline dma_addr_t aq_buf_daddr(struct aq_rxpage *rxpage)
+{
+	return rxpage->daddr + rxpage->pg_off;
+}
+
 static inline unsigned int aq_ring_next_dx(struct aq_ring_s *self,
 					   unsigned int dx)
 {
@@ -136,19 +183,36 @@ static inline unsigned int aq_ring_avail_dx(struct aq_ring_s *self)
 		self->sw_head - self->sw_tail - 1);
 }
 
-struct aq_ring_s *aq_ring_tx_alloc(struct aq_ring_s *self,
-				   struct aq_nic_s *aq_nic,
-				   unsigned int idx,
-				   struct aq_nic_cfg_s *aq_nic_cfg);
-struct aq_ring_s *aq_ring_rx_alloc(struct aq_ring_s *self,
-				   struct aq_nic_s *aq_nic,
-				   unsigned int idx,
-				   struct aq_nic_cfg_s *aq_nic_cfg);
-int aq_ring_init(struct aq_ring_s *self);
+int aq_ring_tx_alloc(struct aq_ring_s *self,
+		     struct aq_nic_s *aq_nic,
+		     unsigned int idx,
+		     struct aq_nic_cfg_s *aq_nic_cfg);
+int aq_ring_rx_alloc(struct aq_ring_s *self,
+		     struct aq_nic_s *aq_nic,
+		     unsigned int idx,
+		     struct aq_nic_cfg_s *aq_nic_cfg);
+
+int aq_ring_init(struct aq_ring_s *self, const enum atl_ring_type ring_type);
 void aq_ring_rx_deinit(struct aq_ring_s *self);
 void aq_ring_free(struct aq_ring_s *self);
-void aq_ring_tx_clean(struct aq_ring_s *self);
-int aq_ring_rx_clean(struct aq_ring_s *self, int *work_done, int budget);
+void aq_ring_update_queue_state(struct aq_ring_s *ring);
+void aq_ring_queue_wake(struct aq_ring_s *ring);
+void aq_ring_queue_stop(struct aq_ring_s *ring);
+bool aq_ring_tx_clean(struct aq_ring_s *self);
+int aq_xdp_xmit(struct net_device *dev, int num_frames,
+		struct xdp_frame **frames, u32 flags);
+int aq_ring_rx_clean(struct aq_ring_s *self,
+		     struct napi_struct *napi,
+		     int *work_done,
+		     int budget);
 int aq_ring_rx_fill(struct aq_ring_s *self);
 
+int aq_ring_hwts_rx_alloc(struct aq_ring_s *self,
+			  struct aq_nic_s *aq_nic, unsigned int idx,
+			  unsigned int size, unsigned int dx_size);
+void aq_ring_hwts_rx_free(struct aq_ring_s *self);
+void aq_ring_hwts_rx_clean(struct aq_ring_s *self, struct aq_nic_s *aq_nic);
+
+unsigned int aq_ring_fill_stats_data(struct aq_ring_s *self, u64 *data);
+
 #endif /* AQ_RING_H */
diff --git a/drivers/net/ethernet/aquantia/atlantic/aq_rss.h b/drivers/net/ethernet/aquantia/atlantic/aq_rss.h
index 1db6eb20a8f2..39b1f43c8ad7 100644
--- a/drivers/net/ethernet/aquantia/atlantic/aq_rss.h
+++ b/drivers/net/ethernet/aquantia/atlantic/aq_rss.h
@@ -1,10 +1,7 @@
+/* SPDX-License-Identifier: GPL-2.0-only */
 /*
  * aQuantia Corporation Network Driver
  * Copyright (C) 2014-2017 aQuantia Corporation. All rights reserved
- *
- * This program is free software; you can redistribute it and/or modify it
- * under the terms and conditions of the GNU General Public License,
- * version 2, as published by the Free Software Foundation.
  */
 
 /* File aq_rss.h: Receive Side Scaling definitions. */
diff --git a/drivers/net/ethernet/aquantia/atlantic/aq_utils.h b/drivers/net/ethernet/aquantia/atlantic/aq_utils.h
index f6012b34abe6..d3a0b2ec0279 100644
--- a/drivers/net/ethernet/aquantia/atlantic/aq_utils.h
+++ b/drivers/net/ethernet/aquantia/atlantic/aq_utils.h
@@ -1,10 +1,7 @@
+/* SPDX-License-Identifier: GPL-2.0-only */
 /*
  * aQuantia Corporation Network Driver
  * Copyright (C) 2014-2017 aQuantia Corporation. All rights reserved
- *
- * This program is free software; you can redistribute it and/or modify it
- * under the terms and conditions of the GNU General Public License,
- * version 2, as published by the Free Software Foundation.
  */
 
 /* File aq_utils.h: Useful macro and structures used in all layers of driver. */
@@ -14,13 +11,6 @@
 
 #include "aq_common.h"
 
-#define AQ_DIMOF(_ARY_)  ARRAY_SIZE(_ARY_)
-
-struct aq_obj_s {
-	spinlock_t lock; /* spinlock for nic/rings processing */
-	atomic_t flags;
-};
-
 static inline void aq_utils_obj_set(atomic_t *flags, u32 mask)
 {
 	unsigned long flags_old, flags_new;
diff --git a/drivers/net/ethernet/aquantia/atlantic/aq_vec.c b/drivers/net/ethernet/aquantia/atlantic/aq_vec.c
index ad5b4d4dac7f..9769ab4f9bef 100644
--- a/drivers/net/ethernet/aquantia/atlantic/aq_vec.c
+++ b/drivers/net/ethernet/aquantia/atlantic/aq_vec.c
@@ -1,10 +1,8 @@
-/*
- * aQuantia Corporation Network Driver
- * Copyright (C) 2014-2017 aQuantia Corporation. All rights reserved
+// SPDX-License-Identifier: GPL-2.0-only
+/* Atlantic Network Driver
  *
- * This program is free software; you can redistribute it and/or modify it
- * under the terms and conditions of the GNU General Public License,
- * version 2, as published by the Free Software Foundation.
+ * Copyright (C) 2014-2019 aQuantia Corporation
+ * Copyright (C) 2019-2020 Marvell International Ltd.
  */
 
 /* File aq_vec.c: Definition of common structure for vector of Rx and Tx rings.
@@ -12,15 +10,9 @@
  */
 
 #include "aq_vec.h"
-#include "aq_nic.h"
-#include "aq_ring.h"
-#include "aq_hw.h"
-
-#include <linux/netdevice.h>
 
 struct aq_vec_s {
-	struct aq_obj_s header;
-	struct aq_hw_ops *aq_hw_ops;
+	const struct aq_hw_ops *aq_hw_ops;
 	struct aq_hw_s *aq_hw;
 	struct aq_nic_s *aq_nic;
 	unsigned int tx_rings;
@@ -34,22 +26,23 @@ struct aq_vec_s {
 #define AQ_VEC_RX_ID 1
 
 static int aq_vec_poll(struct napi_struct *napi, int budget)
-__releases(&self->lock)
-__acquires(&self->lock)
 {
 	struct aq_vec_s *self = container_of(napi, struct aq_vec_s, napi);
+	unsigned int sw_tail_old = 0U;
 	struct aq_ring_s *ring = NULL;
+	bool was_tx_cleaned = true;
+	unsigned int i = 0U;
 	int work_done = 0;
 	int err = 0;
-	unsigned int i = 0U;
-	unsigned int sw_tail_old = 0U;
-	bool was_tx_cleaned = false;
 
 	if (!self) {
 		err = -EINVAL;
-	} else if (spin_trylock(&self->header.lock)) {
-		for (i = 0U, ring = self->ring[0];
-			self->tx_rings > i; ++i, ring = self->ring[i]) {
+	} else {
+		for (i = 0U; self->tx_rings > i; ++i) {
+			ring = self->ring[i];
+			u64_stats_update_begin(&ring[AQ_VEC_RX_ID].stats.rx.syncp);
+			ring[AQ_VEC_RX_ID].stats.rx.polls++;
+			u64_stats_update_end(&ring[AQ_VEC_RX_ID].stats.rx.syncp);
 			if (self->aq_hw_ops->hw_ring_tx_head_update) {
 				err = self->aq_hw_ops->hw_ring_tx_head_update(
 							self->aq_hw,
@@ -60,14 +53,8 @@ __acquires(&self->lock)
 
 			if (ring[AQ_VEC_TX_ID].sw_head !=
 			    ring[AQ_VEC_TX_ID].hw_head) {
-				aq_ring_tx_clean(&ring[AQ_VEC_TX_ID]);
-
-				if (aq_ring_avail_dx(&ring[AQ_VEC_TX_ID]) >
-				    AQ_CFG_SKB_FRAGS_MAX) {
-					aq_nic_ndev_queue_start(self->aq_nic,
-						ring[AQ_VEC_TX_ID].idx);
-				}
-				was_tx_cleaned = true;
+				was_tx_cleaned = aq_ring_tx_clean(&ring[AQ_VEC_TX_ID]);
+				aq_ring_update_queue_state(&ring[AQ_VEC_TX_ID]);
 			}
 
 			err = self->aq_hw_ops->hw_ring_rx_receive(self->aq_hw,
@@ -78,6 +65,7 @@ __acquires(&self->lock)
 			if (ring[AQ_VEC_RX_ID].sw_head !=
 				ring[AQ_VEC_RX_ID].hw_head) {
 				err = aq_ring_rx_clean(&ring[AQ_VEC_RX_ID],
+						       napi,
 						       &work_done,
 						       budget - work_done);
 				if (err < 0)
@@ -97,7 +85,8 @@ __acquires(&self->lock)
 			}
 		}
 
-		if (was_tx_cleaned)
+err_exit:
+		if (!was_tx_cleaned)
 			work_done = budget;
 
 		if (work_done < budget) {
@@ -105,9 +94,6 @@ __acquires(&self->lock)
 			self->aq_hw_ops->hw_irq_enable(self->aq_hw,
 					1U << self->aq_ring_param.vec_idx);
 		}
-
-err_exit:
-		spin_unlock(&self->header.lock);
 	}
 
 	return work_done;
@@ -117,15 +103,10 @@ struct aq_vec_s *aq_vec_alloc(struct aq_nic_s *aq_nic, unsigned int idx,
 			      struct aq_nic_cfg_s *aq_nic_cfg)
 {
 	struct aq_vec_s *self = NULL;
-	struct aq_ring_s *ring = NULL;
-	unsigned int i = 0U;
-	int err = 0;
 
 	self = kzalloc(sizeof(*self), GFP_KERNEL);
-	if (!self) {
-		err = -ENOMEM;
+	if (!self)
 		goto err_exit;
-	}
 
 	self->aq_nic = aq_nic;
 	self->aq_ring_param.vec_idx = idx;
@@ -138,44 +119,65 @@ struct aq_vec_s *aq_vec_alloc(struct aq_nic_s *aq_nic, unsigned int idx,
 	self->tx_rings = 0;
 	self->rx_rings = 0;
 
-	netif_napi_add(aq_nic_get_ndev(aq_nic), &self->napi,
-		       aq_vec_poll, AQ_CFG_NAPI_WEIGHT);
+	netif_napi_add(aq_nic_get_ndev(aq_nic), &self->napi, aq_vec_poll);
+
+err_exit:
+	return self;
+}
+
+int aq_vec_ring_alloc(struct aq_vec_s *self, struct aq_nic_s *aq_nic,
+		      unsigned int idx, struct aq_nic_cfg_s *aq_nic_cfg)
+{
+	struct aq_ring_s *ring = NULL;
+	unsigned int i = 0U;
+	int err = 0;
 
 	for (i = 0; i < aq_nic_cfg->tcs; ++i) {
-		unsigned int idx_ring = AQ_NIC_TCVEC2RING(self->nic,
-						self->tx_rings,
-						self->aq_ring_param.vec_idx);
+		const unsigned int idx_ring = AQ_NIC_CFG_TCVEC2RING(aq_nic_cfg,
+								    i, idx);
 
-		ring = aq_ring_tx_alloc(&self->ring[i][AQ_VEC_TX_ID], aq_nic,
-					idx_ring, aq_nic_cfg);
-		if (!ring) {
-			err = -ENOMEM;
+		ring = &self->ring[i][AQ_VEC_TX_ID];
+		err = aq_ring_tx_alloc(ring, aq_nic, idx_ring, aq_nic_cfg);
+		if (err)
 			goto err_exit;
-		}
 
 		++self->tx_rings;
 
 		aq_nic_set_tx_ring(aq_nic, idx_ring, ring);
 
-		ring = aq_ring_rx_alloc(&self->ring[i][AQ_VEC_RX_ID], aq_nic,
-					idx_ring, aq_nic_cfg);
-		if (!ring) {
+		ring = &self->ring[i][AQ_VEC_RX_ID];
+		if (xdp_rxq_info_reg(&ring->xdp_rxq,
+				     aq_nic->ndev, idx,
+				     self->napi.napi_id) < 0) {
+			err = -ENOMEM;
+			goto err_exit;
+		}
+		if (xdp_rxq_info_reg_mem_model(&ring->xdp_rxq,
+					       MEM_TYPE_PAGE_SHARED, NULL) < 0) {
+			xdp_rxq_info_unreg(&ring->xdp_rxq);
 			err = -ENOMEM;
 			goto err_exit;
 		}
 
+		err = aq_ring_rx_alloc(ring, aq_nic, idx_ring, aq_nic_cfg);
+		if (err) {
+			xdp_rxq_info_unreg(&ring->xdp_rxq);
+			goto err_exit;
+		}
+
 		++self->rx_rings;
 	}
 
 err_exit:
 	if (err < 0) {
-		aq_vec_free(self);
+		aq_vec_ring_free(self);
 		self = NULL;
 	}
-	return self;
+
+	return err;
 }
 
-int aq_vec_init(struct aq_vec_s *self, struct aq_hw_ops *aq_hw_ops,
+int aq_vec_init(struct aq_vec_s *self, const struct aq_hw_ops *aq_hw_ops,
 		struct aq_hw_s *aq_hw)
 {
 	struct aq_ring_s *ring = NULL;
@@ -185,11 +187,9 @@ int aq_vec_init(struct aq_vec_s *self, struct aq_hw_ops *aq_hw_ops,
 	self->aq_hw_ops = aq_hw_ops;
 	self->aq_hw = aq_hw;
 
-	spin_lock_init(&self->header.lock);
-
-	for (i = 0U, ring = self->ring[0];
-		self->tx_rings > i; ++i, ring = self->ring[i]) {
-		err = aq_ring_init(&ring[AQ_VEC_TX_ID]);
+	for (i = 0U; self->tx_rings > i; ++i) {
+		ring = self->ring[i];
+		err = aq_ring_init(&ring[AQ_VEC_TX_ID], ATL_RING_TX);
 		if (err < 0)
 			goto err_exit;
 
@@ -199,7 +199,7 @@ int aq_vec_init(struct aq_vec_s *self, struct aq_hw_ops *aq_hw_ops,
 		if (err < 0)
 			goto err_exit;
 
-		err = aq_ring_init(&ring[AQ_VEC_RX_ID]);
+		err = aq_ring_init(&ring[AQ_VEC_RX_ID], ATL_RING_RX);
 		if (err < 0)
 			goto err_exit;
 
@@ -229,8 +229,8 @@ int aq_vec_start(struct aq_vec_s *self)
 	unsigned int i = 0U;
 	int err = 0;
 
-	for (i = 0U, ring = self->ring[0];
-		self->tx_rings > i; ++i, ring = self->ring[i]) {
+	for (i = 0U; self->tx_rings > i; ++i) {
+		ring = self->ring[i];
 		err = self->aq_hw_ops->hw_ring_tx_start(self->aq_hw,
 							&ring[AQ_VEC_TX_ID]);
 		if (err < 0)
@@ -253,8 +253,8 @@ void aq_vec_stop(struct aq_vec_s *self)
 	struct aq_ring_s *ring = NULL;
 	unsigned int i = 0U;
 
-	for (i = 0U, ring = self->ring[0];
-		self->tx_rings > i; ++i, ring = self->ring[i]) {
+	for (i = 0U; self->tx_rings > i; ++i) {
+		ring = self->ring[i];
 		self->aq_hw_ops->hw_ring_tx_stop(self->aq_hw,
 						 &ring[AQ_VEC_TX_ID]);
 
@@ -273,32 +273,46 @@ void aq_vec_deinit(struct aq_vec_s *self)
 	if (!self)
 		goto err_exit;
 
-	for (i = 0U, ring = self->ring[0];
-		self->tx_rings > i; ++i, ring = self->ring[i]) {
+	for (i = 0U; self->tx_rings > i; ++i) {
+		ring = self->ring[i];
 		aq_ring_tx_clean(&ring[AQ_VEC_TX_ID]);
 		aq_ring_rx_deinit(&ring[AQ_VEC_RX_ID]);
 	}
+
 err_exit:;
 }
 
 void aq_vec_free(struct aq_vec_s *self)
 {
+	if (!self)
+		goto err_exit;
+
+	netif_napi_del(&self->napi);
+
+	kfree(self);
+
+err_exit:;
+}
+
+void aq_vec_ring_free(struct aq_vec_s *self)
+{
 	struct aq_ring_s *ring = NULL;
 	unsigned int i = 0U;
 
 	if (!self)
 		goto err_exit;
 
-	for (i = 0U, ring = self->ring[0];
-		self->tx_rings > i; ++i, ring = self->ring[i]) {
+	for (i = 0U; self->tx_rings > i; ++i) {
+		ring = self->ring[i];
 		aq_ring_free(&ring[AQ_VEC_TX_ID]);
-		aq_ring_free(&ring[AQ_VEC_RX_ID]);
+		if (i < self->rx_rings) {
+			xdp_rxq_info_unreg(&ring[AQ_VEC_RX_ID].xdp_rxq);
+			aq_ring_free(&ring[AQ_VEC_RX_ID]);
+		}
 	}
 
-	netif_napi_del(&self->napi);
-
-	kfree(self);
-
+	self->tx_rings = 0;
+	self->rx_rings = 0;
 err_exit:;
 }
 
@@ -321,15 +335,13 @@ irqreturn_t aq_vec_isr_legacy(int irq, void *private)
 {
 	struct aq_vec_s *self = private;
 	u64 irq_mask = 0U;
-	irqreturn_t err = 0;
+	int err;
 
-	if (!self) {
-		err = -EINVAL;
-		goto err_exit;
-	}
+	if (!self)
+		return IRQ_NONE;
 	err = self->aq_hw_ops->hw_irq_read(self->aq_hw, &irq_mask);
 	if (err < 0)
-		goto err_exit;
+		return IRQ_NONE;
 
 	if (irq_mask) {
 		self->aq_hw_ops->hw_irq_disable(self->aq_hw,
@@ -337,11 +349,10 @@ irqreturn_t aq_vec_isr_legacy(int irq, void *private)
 		napi_schedule(&self->napi);
 	} else {
 		self->aq_hw_ops->hw_irq_enable(self->aq_hw, 1U);
-		err = IRQ_NONE;
+		return IRQ_NONE;
 	}
 
-err_exit:
-	return err >= 0 ? IRQ_HANDLED : IRQ_NONE;
+	return IRQ_HANDLED;
 }
 
 cpumask_t *aq_vec_get_affinity_mask(struct aq_vec_s *self)
@@ -349,48 +360,20 @@ cpumask_t *aq_vec_get_affinity_mask(struct aq_vec_s *self)
 	return &self->aq_ring_param.affinity_mask;
 }
 
-void aq_vec_add_stats(struct aq_vec_s *self,
-		      struct aq_ring_stats_rx_s *stats_rx,
-		      struct aq_ring_stats_tx_s *stats_tx)
+bool aq_vec_is_valid_tc(struct aq_vec_s *self, const unsigned int tc)
 {
-	struct aq_ring_s *ring = NULL;
-	unsigned int r = 0U;
-
-	for (r = 0U, ring = self->ring[0];
-		self->tx_rings > r; ++r, ring = self->ring[r]) {
-		struct aq_ring_stats_tx_s *tx = &ring[AQ_VEC_TX_ID].stats.tx;
-		struct aq_ring_stats_rx_s *rx = &ring[AQ_VEC_RX_ID].stats.rx;
-
-		stats_rx->packets += rx->packets;
-		stats_rx->bytes += rx->bytes;
-		stats_rx->errors += rx->errors;
-		stats_rx->jumbo_packets += rx->jumbo_packets;
-		stats_rx->lro_packets += rx->lro_packets;
-
-		stats_tx->packets += tx->packets;
-		stats_tx->bytes += tx->bytes;
-		stats_tx->errors += tx->errors;
-	}
+	return tc < self->rx_rings && tc < self->tx_rings;
 }
 
-int aq_vec_get_sw_stats(struct aq_vec_s *self, u64 *data, unsigned int *p_count)
+unsigned int aq_vec_get_sw_stats(struct aq_vec_s *self, const unsigned int tc, u64 *data)
 {
-	unsigned int count = 0U;
-	struct aq_ring_stats_rx_s stats_rx;
-	struct aq_ring_stats_tx_s stats_tx;
-
-	memset(&stats_rx, 0U, sizeof(struct aq_ring_stats_rx_s));
-	memset(&stats_tx, 0U, sizeof(struct aq_ring_stats_tx_s));
-	aq_vec_add_stats(self, &stats_rx, &stats_tx);
+	unsigned int count;
 
-	data[count] += stats_rx.packets;
-	data[++count] += stats_tx.packets;
-	data[++count] += stats_rx.jumbo_packets;
-	data[++count] += stats_rx.lro_packets;
-	data[++count] += stats_rx.errors;
+	if (!aq_vec_is_valid_tc(self, tc))
+		return 0;
 
-	if (p_count)
-		*p_count = ++count;
+	count = aq_ring_fill_stats_data(&self->ring[tc][AQ_VEC_RX_ID], data);
+	count += aq_ring_fill_stats_data(&self->ring[tc][AQ_VEC_TX_ID], data + count);
 
-	return 0;
+	return count;
 }
diff --git a/drivers/net/ethernet/aquantia/atlantic/aq_vec.h b/drivers/net/ethernet/aquantia/atlantic/aq_vec.h
index 6c68b184236c..78fac609b71d 100644
--- a/drivers/net/ethernet/aquantia/atlantic/aq_vec.h
+++ b/drivers/net/ethernet/aquantia/atlantic/aq_vec.h
@@ -1,10 +1,8 @@
-/*
- * aQuantia Corporation Network Driver
- * Copyright (C) 2014-2017 aQuantia Corporation. All rights reserved
+/* SPDX-License-Identifier: GPL-2.0-only */
+/* Atlantic Network Driver
  *
- * This program is free software; you can redistribute it and/or modify it
- * under the terms and conditions of the GNU General Public License,
- * version 2, as published by the Free Software Foundation.
+ * Copyright (C) 2014-2019 aQuantia Corporation
+ * Copyright (C) 2019-2020 Marvell International Ltd.
  */
 
 /* File aq_vec.h: Definition of common structures for vector of Rx and Tx rings.
@@ -15,10 +13,18 @@
 #define AQ_VEC_H
 
 #include "aq_common.h"
+#include "aq_nic.h"
+#include "aq_ring.h"
+#include "aq_hw.h"
+
 #include <linux/irqreturn.h>
+#include <linux/filter.h>
+#include <linux/netdevice.h>
 
 struct aq_hw_s;
 struct aq_hw_ops;
+struct aq_nic_s;
+struct aq_nic_cfg_s;
 struct aq_ring_stats_rx_s;
 struct aq_ring_stats_tx_s;
 
@@ -26,17 +32,17 @@ irqreturn_t aq_vec_isr(int irq, void *private);
 irqreturn_t aq_vec_isr_legacy(int irq, void *private);
 struct aq_vec_s *aq_vec_alloc(struct aq_nic_s *aq_nic, unsigned int idx,
 			      struct aq_nic_cfg_s *aq_nic_cfg);
-int aq_vec_init(struct aq_vec_s *self, struct aq_hw_ops *aq_hw_ops,
+int aq_vec_ring_alloc(struct aq_vec_s *self, struct aq_nic_s *aq_nic,
+		      unsigned int idx, struct aq_nic_cfg_s *aq_nic_cfg);
+int aq_vec_init(struct aq_vec_s *self, const struct aq_hw_ops *aq_hw_ops,
 		struct aq_hw_s *aq_hw);
 void aq_vec_deinit(struct aq_vec_s *self);
 void aq_vec_free(struct aq_vec_s *self);
+void aq_vec_ring_free(struct aq_vec_s *self);
 int aq_vec_start(struct aq_vec_s *self);
 void aq_vec_stop(struct aq_vec_s *self);
 cpumask_t *aq_vec_get_affinity_mask(struct aq_vec_s *self);
-int aq_vec_get_sw_stats(struct aq_vec_s *self, u64 *data,
-			unsigned int *p_count);
-void aq_vec_add_stats(struct aq_vec_s *self,
-		      struct aq_ring_stats_rx_s *stats_rx,
-		      struct aq_ring_stats_tx_s *stats_tx);
+bool aq_vec_is_valid_tc(struct aq_vec_s *self, const unsigned int tc);
+unsigned int aq_vec_get_sw_stats(struct aq_vec_s *self, const unsigned int tc, u64 *data);
 
 #endif /* AQ_VEC_H */
diff --git a/drivers/net/ethernet/aquantia/atlantic/hw_atl/hw_atl_a0.c b/drivers/net/ethernet/aquantia/atlantic/hw_atl/hw_atl_a0.c
index faeb4935ef3e..8de2cdd09213 100644
--- a/drivers/net/ethernet/aquantia/atlantic/hw_atl/hw_atl_a0.c
+++ b/drivers/net/ethernet/aquantia/atlantic/hw_atl/hw_atl_a0.c
@@ -1,10 +1,8 @@
-/*
- * aQuantia Corporation Network Driver
- * Copyright (C) 2014-2017 aQuantia Corporation. All rights reserved
+// SPDX-License-Identifier: GPL-2.0-only
+/* Atlantic Network Driver
  *
- * This program is free software; you can redistribute it and/or modify it
- * under the terms and conditions of the GNU General Public License,
- * version 2, as published by the Free Software Foundation.
+ * Copyright (C) 2014-2019 aQuantia Corporation
+ * Copyright (C) 2019-2020 Marvell International Ltd.
  */
 
 /* File hw_atl_a0.c: Definition of Atlantic hardware specific functions. */
@@ -12,67 +10,110 @@
 #include "../aq_hw.h"
 #include "../aq_hw_utils.h"
 #include "../aq_ring.h"
+#include "../aq_nic.h"
 #include "hw_atl_a0.h"
 #include "hw_atl_utils.h"
 #include "hw_atl_llh.h"
 #include "hw_atl_a0_internal.h"
 
-static int hw_atl_a0_get_hw_caps(struct aq_hw_s *self,
-				 struct aq_hw_caps_s *aq_hw_caps)
-{
-	memcpy(aq_hw_caps, &hw_atl_a0_hw_caps_, sizeof(*aq_hw_caps));
-	return 0;
-}
-
-static struct aq_hw_s *hw_atl_a0_create(struct aq_pci_func_s *aq_pci_func,
-					unsigned int port,
-					struct aq_hw_ops *ops)
-{
-	struct hw_atl_s *self = NULL;
-
-	self = kzalloc(sizeof(*self), GFP_KERNEL);
-	if (!self)
-		goto err_exit;
-
-	self->base.aq_pci_func = aq_pci_func;
+#define DEFAULT_A0_BOARD_BASIC_CAPABILITIES	     \
+	.is_64_dma = true,			     \
+	.op64bit = false,			     \
+	.msix_irqs = 4U,			     \
+	.irq_mask = ~0U,			     \
+	.vecs = HW_ATL_A0_RSS_MAX,		     \
+	.tcs_max = HW_ATL_A0_TC_MAX,		     \
+	.rxd_alignment = 1U,			     \
+	.rxd_size = HW_ATL_A0_RXD_SIZE,		     \
+	.rxds_max = HW_ATL_A0_MAX_RXD,		     \
+	.rxds_min = HW_ATL_A0_MIN_RXD,		     \
+	.txd_alignment = 1U,			     \
+	.txd_size = HW_ATL_A0_TXD_SIZE,		     \
+	.txds_max = HW_ATL_A0_MAX_TXD,		     \
+	.txds_min = HW_ATL_A0_MIN_RXD,		     \
+	.txhwb_alignment = 4096U,		     \
+	.tx_rings = HW_ATL_A0_TX_RINGS,		     \
+	.rx_rings = HW_ATL_A0_RX_RINGS,		     \
+	.hw_features = NETIF_F_HW_CSUM |	     \
+			NETIF_F_RXHASH |	     \
+			NETIF_F_RXCSUM |	     \
+			NETIF_F_SG |		     \
+			NETIF_F_TSO |		     \
+			NETIF_F_NTUPLE |	     \
+			NETIF_F_HW_VLAN_CTAG_FILTER, \
+	.hw_priv_flags = IFF_UNICAST_FLT,	     \
+	.flow_control = true,			     \
+	.mtu = HW_ATL_A0_MTU_JUMBO,		     \
+	.mac_regs_count = 88,			     \
+	.hw_alive_check_addr = 0x10U
+
+const struct aq_hw_caps_s hw_atl_a0_caps_aqc100 = {
+	DEFAULT_A0_BOARD_BASIC_CAPABILITIES,
+	.media_type = AQ_HW_MEDIA_TYPE_FIBRE,
+	.link_speed_msk = AQ_NIC_RATE_5G |
+			  AQ_NIC_RATE_2G5 |
+			  AQ_NIC_RATE_1G |
+			  AQ_NIC_RATE_100M,
+};
 
-	self->base.not_ff_addr = 0x10U;
+const struct aq_hw_caps_s hw_atl_a0_caps_aqc107 = {
+	DEFAULT_A0_BOARD_BASIC_CAPABILITIES,
+	.media_type = AQ_HW_MEDIA_TYPE_TP,
+	.link_speed_msk = AQ_NIC_RATE_10G |
+			  AQ_NIC_RATE_5G |
+			  AQ_NIC_RATE_2G5 |
+			  AQ_NIC_RATE_1G |
+			  AQ_NIC_RATE_100M,
+};
 
-err_exit:
-	return (struct aq_hw_s *)self;
-}
+const struct aq_hw_caps_s hw_atl_a0_caps_aqc108 = {
+	DEFAULT_A0_BOARD_BASIC_CAPABILITIES,
+	.media_type = AQ_HW_MEDIA_TYPE_TP,
+	.link_speed_msk = AQ_NIC_RATE_5G |
+			  AQ_NIC_RATE_2G5 |
+			  AQ_NIC_RATE_1G |
+			  AQ_NIC_RATE_100M,
+};
 
-static void hw_atl_a0_destroy(struct aq_hw_s *self)
-{
-	kfree(self);
-}
+const struct aq_hw_caps_s hw_atl_a0_caps_aqc109 = {
+	DEFAULT_A0_BOARD_BASIC_CAPABILITIES,
+	.media_type = AQ_HW_MEDIA_TYPE_TP,
+	.link_speed_msk = AQ_NIC_RATE_2G5 |
+			  AQ_NIC_RATE_1G |
+			  AQ_NIC_RATE_100M,
+};
 
 static int hw_atl_a0_hw_reset(struct aq_hw_s *self)
 {
 	int err = 0;
+	u32 val;
 
-	glb_glb_reg_res_dis_set(self, 1U);
-	pci_pci_reg_res_dis_set(self, 0U);
-	rx_rx_reg_res_dis_set(self, 0U);
-	tx_tx_reg_res_dis_set(self, 0U);
+	hw_atl_glb_glb_reg_res_dis_set(self, 1U);
+	hw_atl_pci_pci_reg_res_dis_set(self, 0U);
+	hw_atl_rx_rx_reg_res_dis_set(self, 0U);
+	hw_atl_tx_tx_reg_res_dis_set(self, 0U);
 
 	HW_ATL_FLUSH();
-	glb_soft_res_set(self, 1);
+	hw_atl_glb_soft_res_set(self, 1);
 
 	/* check 10 times by 1ms */
-	AQ_HW_WAIT_FOR(glb_soft_res_get(self) == 0, 1000U, 10U);
+	err = readx_poll_timeout_atomic(hw_atl_glb_soft_res_get,
+					self, val, val == 0,
+					1000U, 10000U);
 	if (err < 0)
 		goto err_exit;
 
-	itr_irq_reg_res_dis_set(self, 0U);
-	itr_res_irq_set(self, 1U);
+	hw_atl_itr_irq_reg_res_dis_set(self, 0U);
+	hw_atl_itr_res_irq_set(self, 1U);
 
 	/* check 10 times by 1ms */
-	AQ_HW_WAIT_FOR(itr_res_irq_get(self) == 0, 1000U, 10U);
+	err = readx_poll_timeout_atomic(hw_atl_itr_res_irq_get,
+					self, val, val == 0,
+					1000U, 10000U);
 	if (err < 0)
 		goto err_exit;
 
-	hw_atl_utils_mpi_set(self, MPI_RESET, 0x0U);
+	self->aq_fw_ops->set_state(self, MPI_RESET);
 
 	err = aq_hw_err_from_flags(self);
 
@@ -82,57 +123,59 @@ err_exit:
 
 static int hw_atl_a0_hw_qos_set(struct aq_hw_s *self)
 {
-	u32 tc = 0U;
-	u32 buff_size = 0U;
-	unsigned int i_priority = 0U;
 	bool is_rx_flow_control = false;
+	unsigned int i_priority = 0U;
+	u32 buff_size = 0U;
+	u32 tc = 0U;
 
 	/* TPS Descriptor rate init */
-	tps_tx_pkt_shed_desc_rate_curr_time_res_set(self, 0x0U);
-	tps_tx_pkt_shed_desc_rate_lim_set(self, 0xA);
+	hw_atl_tps_tx_pkt_shed_desc_rate_curr_time_res_set(self, 0x0U);
+	hw_atl_tps_tx_pkt_shed_desc_rate_lim_set(self, 0xA);
 
 	/* TPS VM init */
-	tps_tx_pkt_shed_desc_vm_arb_mode_set(self, 0U);
+	hw_atl_tps_tx_pkt_shed_desc_vm_arb_mode_set(self, 0U);
 
 	/* TPS TC credits init */
-	tps_tx_pkt_shed_desc_tc_arb_mode_set(self, 0U);
-	tps_tx_pkt_shed_data_arb_mode_set(self, 0U);
+	hw_atl_tps_tx_pkt_shed_desc_tc_arb_mode_set(self, 0U);
+	hw_atl_tps_tx_pkt_shed_data_arb_mode_set(self, 0U);
 
-	tps_tx_pkt_shed_tc_data_max_credit_set(self, 0xFFF, 0U);
-	tps_tx_pkt_shed_tc_data_weight_set(self, 0x64, 0U);
-	tps_tx_pkt_shed_desc_tc_max_credit_set(self, 0x50, 0U);
-	tps_tx_pkt_shed_desc_tc_weight_set(self, 0x1E, 0U);
+	hw_atl_tps_tx_pkt_shed_tc_data_max_credit_set(self, 0U, 0xFFF);
+	hw_atl_tps_tx_pkt_shed_tc_data_weight_set(self, 0U, 0x64);
+	hw_atl_tps_tx_pkt_shed_desc_tc_max_credit_set(self, 0U, 0x50);
+	hw_atl_tps_tx_pkt_shed_desc_tc_weight_set(self, 0U, 0x1E);
 
 	/* Tx buf size */
 	buff_size = HW_ATL_A0_TXBUF_MAX;
 
-	tpb_tx_pkt_buff_size_per_tc_set(self, buff_size, tc);
-	tpb_tx_buff_hi_threshold_per_tc_set(self,
-					    (buff_size * (1024 / 32U) * 66U) /
-					    100U, tc);
-	tpb_tx_buff_lo_threshold_per_tc_set(self,
-					    (buff_size * (1024 / 32U) * 50U) /
-					    100U, tc);
+	hw_atl_tpb_tx_pkt_buff_size_per_tc_set(self, buff_size, tc);
+	hw_atl_tpb_tx_buff_hi_threshold_per_tc_set(self,
+						   (buff_size *
+						   (1024 / 32U) * 66U) /
+						   100U, tc);
+	hw_atl_tpb_tx_buff_lo_threshold_per_tc_set(self,
+						   (buff_size *
+						   (1024 / 32U) * 50U) /
+						   100U, tc);
 
 	/* QoS Rx buf size per TC */
 	tc = 0;
-	is_rx_flow_control = (AQ_NIC_FC_RX & self->aq_nic_cfg->flow_control);
+	is_rx_flow_control = (AQ_NIC_FC_RX & self->aq_nic_cfg->fc.req);
 	buff_size = HW_ATL_A0_RXBUF_MAX;
 
-	rpb_rx_pkt_buff_size_per_tc_set(self, buff_size, tc);
-	rpb_rx_buff_hi_threshold_per_tc_set(self,
-					    (buff_size *
-					    (1024U / 32U) * 66U) /
-					    100U, tc);
-	rpb_rx_buff_lo_threshold_per_tc_set(self,
-					    (buff_size *
-					    (1024U / 32U) * 50U) /
-					    100U, tc);
-	rpb_rx_xoff_en_per_tc_set(self, is_rx_flow_control ? 1U : 0U, tc);
+	hw_atl_rpb_rx_pkt_buff_size_per_tc_set(self, buff_size, tc);
+	hw_atl_rpb_rx_buff_hi_threshold_per_tc_set(self,
+						   (buff_size *
+						   (1024U / 32U) * 66U) /
+						   100U, tc);
+	hw_atl_rpb_rx_buff_lo_threshold_per_tc_set(self,
+						   (buff_size *
+						   (1024U / 32U) * 50U) /
+						   100U, tc);
+	hw_atl_rpb_rx_xoff_en_per_tc_set(self, is_rx_flow_control ? 1U : 0U, tc);
 
 	/* QoS 802.1p priority -> TC mapping */
 	for (i_priority = 8U; i_priority--;)
-		rpf_rpb_user_priority_tc_map_set(self, i_priority, 0U);
+		hw_atl_rpf_rpb_user_priority_tc_map_set(self, i_priority, 0U);
 
 	return aq_hw_err_from_flags(self);
 }
@@ -140,20 +183,21 @@ static int hw_atl_a0_hw_qos_set(struct aq_hw_s *self)
 static int hw_atl_a0_hw_rss_hash_set(struct aq_hw_s *self,
 				     struct aq_rss_parameters *rss_params)
 {
-	struct aq_nic_cfg_s *cfg = NULL;
-	int err = 0;
-	unsigned int i = 0U;
+	struct aq_nic_cfg_s *cfg = self->aq_nic_cfg;
 	unsigned int addr = 0U;
-
-	cfg = self->aq_nic_cfg;
+	unsigned int i = 0U;
+	int err = 0;
+	u32 val;
 
 	for (i = 10, addr = 0U; i--; ++addr) {
 		u32 key_data = cfg->is_rss ?
 			__swab32(rss_params->hash_secret_key[i]) : 0U;
-		rpf_rss_key_wr_data_set(self, key_data);
-		rpf_rss_key_addr_set(self, addr);
-		rpf_rss_key_wr_en_set(self, 1U);
-		AQ_HW_WAIT_FOR(rpf_rss_key_wr_en_get(self) == 0, 1000U, 10U);
+		hw_atl_rpf_rss_key_wr_data_set(self, key_data);
+		hw_atl_rpf_rss_key_addr_set(self, addr);
+		hw_atl_rpf_rss_key_wr_en_set(self, 1U);
+		err = readx_poll_timeout_atomic(hw_atl_rpf_rss_key_wr_en_get,
+						self, val, val == 0,
+						1000U, 10000U);
 		if (err < 0)
 			goto err_exit;
 	}
@@ -167,12 +211,13 @@ err_exit:
 static int hw_atl_a0_hw_rss_set(struct aq_hw_s *self,
 				struct aq_rss_parameters *rss_params)
 {
-	u8 *indirection_table =	rss_params->indirection_table;
-	u32 i = 0U;
 	u32 num_rss_queues = max(1U, self->aq_nic_cfg->num_rss_queues);
+	u8 *indirection_table =	rss_params->indirection_table;
+	u16 bitary[1 + (HW_ATL_A0_RSS_REDIRECTION_MAX *
+		   HW_ATL_A0_RSS_REDIRECTION_BITS / 16U)];
 	int err = 0;
-	u16 bitary[(HW_ATL_A0_RSS_REDIRECTION_MAX *
-					HW_ATL_A0_RSS_REDIRECTION_BITS / 16U)];
+	u32 i = 0U;
+	u32 val;
 
 	memset(bitary, 0, sizeof(bitary));
 
@@ -182,11 +227,13 @@ static int hw_atl_a0_hw_rss_set(struct aq_hw_s *self,
 			((i * 3U) & 0xFU));
 	}
 
-	for (i = AQ_DIMOF(bitary); i--;) {
-		rpf_rss_redir_tbl_wr_data_set(self, bitary[i]);
-		rpf_rss_redir_tbl_addr_set(self, i);
-		rpf_rss_redir_wr_en_set(self, 1U);
-		AQ_HW_WAIT_FOR(rpf_rss_redir_wr_en_get(self) == 0, 1000U, 10U);
+	for (i = ARRAY_SIZE(bitary); i--;) {
+		hw_atl_rpf_rss_redir_tbl_wr_data_set(self, bitary[i]);
+		hw_atl_rpf_rss_redir_tbl_addr_set(self, i);
+		hw_atl_rpf_rss_redir_wr_en_set(self, 1U);
+		err = readx_poll_timeout_atomic(hw_atl_rpf_rss_redir_wr_en_get,
+						self, val, val == 0,
+						1000U, 10000U);
 		if (err < 0)
 			goto err_exit;
 	}
@@ -201,35 +248,34 @@ static int hw_atl_a0_hw_offload_set(struct aq_hw_s *self,
 				    struct aq_nic_cfg_s *aq_nic_cfg)
 {
 	/* TX checksums offloads*/
-	tpo_ipv4header_crc_offload_en_set(self, 1);
-	tpo_tcp_udp_crc_offload_en_set(self, 1);
+	hw_atl_tpo_ipv4header_crc_offload_en_set(self, 1);
+	hw_atl_tpo_tcp_udp_crc_offload_en_set(self, 1);
 
 	/* RX checksums offloads*/
-	rpo_ipv4header_crc_offload_en_set(self, 1);
-	rpo_tcp_udp_crc_offload_en_set(self, 1);
+	hw_atl_rpo_ipv4header_crc_offload_en_set(self, 1);
+	hw_atl_rpo_tcp_udp_crc_offload_en_set(self, 1);
 
 	/* LSO offloads*/
-	tdm_large_send_offload_en_set(self, 0xFFFFFFFFU);
+	hw_atl_tdm_large_send_offload_en_set(self, 0xFFFFFFFFU);
 
 	return aq_hw_err_from_flags(self);
 }
 
 static int hw_atl_a0_hw_init_tx_path(struct aq_hw_s *self)
 {
-	thm_lso_tcp_flag_of_first_pkt_set(self, 0x0FF6U);
-	thm_lso_tcp_flag_of_middle_pkt_set(self, 0x0FF6U);
-	thm_lso_tcp_flag_of_last_pkt_set(self, 0x0F7FU);
+	hw_atl_thm_lso_tcp_flag_of_first_pkt_set(self, 0x0FF6U);
+	hw_atl_thm_lso_tcp_flag_of_middle_pkt_set(self, 0x0FF6U);
+	hw_atl_thm_lso_tcp_flag_of_last_pkt_set(self, 0x0F7FU);
 
 	/* Tx interrupts */
-	tdm_tx_desc_wr_wb_irq_en_set(self, 1U);
+	hw_atl_tdm_tx_desc_wr_wb_irq_en_set(self, 1U);
 
 	/* misc */
-	aq_hw_write_reg(self, 0x00007040U, IS_CHIP_FEATURE(TPO2) ?
-			0x00010000U : 0x00000000U);
-	tdm_tx_dca_en_set(self, 0U);
-	tdm_tx_dca_mode_set(self, 0U);
+	aq_hw_write_reg(self, 0x00007040U, 0x00000000U);
+	hw_atl_tdm_tx_dca_en_set(self, 0U);
+	hw_atl_tdm_tx_dca_mode_set(self, 0U);
 
-	tpb_tx_path_scp_ins_en_set(self, 1U);
+	hw_atl_tpb_tx_path_scp_ins_en_set(self, 1U);
 
 	return aq_hw_err_from_flags(self);
 }
@@ -240,60 +286,61 @@ static int hw_atl_a0_hw_init_rx_path(struct aq_hw_s *self)
 	int i;
 
 	/* Rx TC/RSS number config */
-	rpb_rpf_rx_traf_class_mode_set(self, 1U);
+	hw_atl_rpb_rpf_rx_traf_class_mode_set(self, 1U);
 
 	/* Rx flow control */
-	rpb_rx_flow_ctl_mode_set(self, 1U);
+	hw_atl_rpb_rx_flow_ctl_mode_set(self, 1U);
 
 	/* RSS Ring selection */
-	reg_rx_flr_rss_control1set(self, cfg->is_rss ?
-					0xB3333333U : 0x00000000U);
+	hw_atl_reg_rx_flr_rss_control1set(self, cfg->is_rss ?
+					  0xB3333333U : 0x00000000U);
 
 	/* Multicast filters */
 	for (i = HW_ATL_A0_MAC_MAX; i--;) {
-		rpfl2_uc_flr_en_set(self, (i == 0U) ? 1U : 0U, i);
-		rpfl2unicast_flr_act_set(self, 1U, i);
+		hw_atl_rpfl2_uc_flr_en_set(self, (i == 0U) ? 1U : 0U, i);
+		hw_atl_rpfl2unicast_flr_act_set(self, 1U, i);
 	}
 
-	reg_rx_flr_mcst_flr_msk_set(self, 0x00000000U);
-	reg_rx_flr_mcst_flr_set(self, 0x00010FFFU, 0U);
+	hw_atl_reg_rx_flr_mcst_flr_msk_set(self, 0x00000000U);
+	hw_atl_reg_rx_flr_mcst_flr_set(self, 0x00010FFFU, 0U);
 
 	/* Vlan filters */
-	rpf_vlan_outer_etht_set(self, 0x88A8U);
-	rpf_vlan_inner_etht_set(self, 0x8100U);
-	rpf_vlan_prom_mode_en_set(self, 1);
+	hw_atl_rpf_vlan_outer_etht_set(self, 0x88A8U);
+	hw_atl_rpf_vlan_inner_etht_set(self, 0x8100U);
+	hw_atl_rpf_vlan_prom_mode_en_set(self, 1);
 
 	/* Rx Interrupts */
-	rdm_rx_desc_wr_wb_irq_en_set(self, 1U);
+	hw_atl_rdm_rx_desc_wr_wb_irq_en_set(self, 1U);
 
 	/* misc */
-	rpfl2broadcast_flr_act_set(self, 1U);
-	rpfl2broadcast_count_threshold_set(self, 0xFFFFU & (~0U / 256U));
+	hw_atl_rpfl2broadcast_flr_act_set(self, 1U);
+	hw_atl_rpfl2broadcast_count_threshold_set(self, 0xFFFFU & (~0U / 256U));
 
-	rdm_rx_dca_en_set(self, 0U);
-	rdm_rx_dca_mode_set(self, 0U);
+	hw_atl_rdm_rx_dca_en_set(self, 0U);
+	hw_atl_rdm_rx_dca_mode_set(self, 0U);
 
 	return aq_hw_err_from_flags(self);
 }
 
-static int hw_atl_a0_hw_mac_addr_set(struct aq_hw_s *self, u8 *mac_addr)
+static int hw_atl_a0_hw_mac_addr_set(struct aq_hw_s *self, const u8 *mac_addr)
 {
-	int err = 0;
 	unsigned int h = 0U;
 	unsigned int l = 0U;
+	int err = 0;
 
 	if (!mac_addr) {
 		err = -EINVAL;
 		goto err_exit;
 	}
+
 	h = (mac_addr[0] << 8) | (mac_addr[1]);
 	l = (mac_addr[2] << 24) | (mac_addr[3] << 16) |
-		(mac_addr[4] << 8) | mac_addr[5];
+	    (mac_addr[4] << 8) | mac_addr[5];
 
-	rpfl2_uc_flr_en_set(self, 0U, HW_ATL_A0_MAC);
-	rpfl2unicast_dest_addresslsw_set(self, l, HW_ATL_A0_MAC);
-	rpfl2unicast_dest_addressmsw_set(self, h, HW_ATL_A0_MAC);
-	rpfl2_uc_flr_en_set(self, 1U, HW_ATL_A0_MAC);
+	hw_atl_rpfl2_uc_flr_en_set(self, 0U, HW_ATL_A0_MAC);
+	hw_atl_rpfl2unicast_dest_addresslsw_set(self, l, HW_ATL_A0_MAC);
+	hw_atl_rpfl2unicast_dest_addressmsw_set(self, h, HW_ATL_A0_MAC);
+	hw_atl_rpfl2_uc_flr_en_set(self, 1U, HW_ATL_A0_MAC);
 
 	err = aq_hw_err_from_flags(self);
 
@@ -301,56 +348,53 @@ err_exit:
 	return err;
 }
 
-static int hw_atl_a0_hw_init(struct aq_hw_s *self,
-			     struct aq_nic_cfg_s *aq_nic_cfg,
-			     u8 *mac_addr)
+static int hw_atl_a0_hw_init(struct aq_hw_s *self, const u8 *mac_addr)
 {
 	static u32 aq_hw_atl_igcr_table_[4][2] = {
-		{ 0x20000000U, 0x20000000U }, /* AQ_IRQ_INVALID */
-		{ 0x20000080U, 0x20000080U }, /* AQ_IRQ_LEGACY */
-		{ 0x20000021U, 0x20000025U }, /* AQ_IRQ_MSI */
-		{ 0x20000022U, 0x20000026U }  /* AQ_IRQ_MSIX */
+		[AQ_HW_IRQ_INVALID] = { 0x20000000U, 0x20000000U },
+		[AQ_HW_IRQ_INTX]    = { 0x20000080U, 0x20000080U },
+		[AQ_HW_IRQ_MSI]     = { 0x20000021U, 0x20000025U },
+		[AQ_HW_IRQ_MSIX]    = { 0x20000022U, 0x20000026U },
 	};
-
+	struct aq_nic_cfg_s *aq_nic_cfg = self->aq_nic_cfg;
 	int err = 0;
 
-	self->aq_nic_cfg = aq_nic_cfg;
-
-	hw_atl_utils_hw_chip_features_init(self,
-					   &PHAL_ATLANTIC_A0->chip_features);
-
 	hw_atl_a0_hw_init_tx_path(self);
 	hw_atl_a0_hw_init_rx_path(self);
 
 	hw_atl_a0_hw_mac_addr_set(self, mac_addr);
 
-	hw_atl_utils_mpi_set(self, MPI_INIT, aq_nic_cfg->link_speed_msk);
+	self->aq_fw_ops->set_link_speed(self, aq_nic_cfg->link_speed_msk);
+	self->aq_fw_ops->set_state(self, MPI_INIT);
 
-	reg_tx_dma_debug_ctl_set(self, 0x800000b8U);
-	reg_tx_dma_debug_ctl_set(self, 0x000000b8U);
+	hw_atl_reg_tx_dma_debug_ctl_set(self, 0x800000b8U);
+	hw_atl_reg_tx_dma_debug_ctl_set(self, 0x000000b8U);
 
 	hw_atl_a0_hw_qos_set(self);
 	hw_atl_a0_hw_rss_set(self, &aq_nic_cfg->aq_rss);
 	hw_atl_a0_hw_rss_hash_set(self, &aq_nic_cfg->aq_rss);
 
+	/* Reset link status and read out initial hardware counters */
+	self->aq_link_status.mbps = 0;
+	self->aq_fw_ops->update_stats(self);
+
 	err = aq_hw_err_from_flags(self);
 	if (err < 0)
 		goto err_exit;
 
 	/* Interrupts */
-	reg_irq_glb_ctl_set(self,
-			    aq_hw_atl_igcr_table_[aq_nic_cfg->irq_type]
-						 [(aq_nic_cfg->vecs > 1U) ?
-						 1 : 0]);
+	hw_atl_reg_irq_glb_ctl_set(self,
+				   aq_hw_atl_igcr_table_[aq_nic_cfg->irq_type]
+					[(aq_nic_cfg->vecs > 1U) ? 1 : 0]);
 
-	itr_irq_auto_masklsw_set(self, aq_nic_cfg->aq_hw_caps->irq_mask);
+	hw_atl_itr_irq_auto_masklsw_set(self, aq_nic_cfg->aq_hw_caps->irq_mask);
 
 	/* Interrupts */
-	reg_gen_irq_map_set(self,
-			    ((HW_ATL_A0_ERR_INT << 0x18) | (1U << 0x1F)) |
-			    ((HW_ATL_A0_ERR_INT << 0x10) | (1U << 0x17)) |
-			    ((HW_ATL_A0_ERR_INT << 8) | (1U << 0xF)) |
-			    ((HW_ATL_A0_ERR_INT) | (1U << 0x7)), 0U);
+	hw_atl_reg_gen_irq_map_set(self,
+				   ((HW_ATL_A0_ERR_INT << 0x18) | (1U << 0x1F)) |
+				   ((HW_ATL_A0_ERR_INT << 0x10) | (1U << 0x17)) |
+				   ((HW_ATL_A0_ERR_INT << 8) | (1U << 0xF)) |
+				   ((HW_ATL_A0_ERR_INT) | (1U << 0x7)), 0U);
 
 	hw_atl_a0_hw_offload_set(self, aq_nic_cfg);
 
@@ -361,28 +405,32 @@ err_exit:
 static int hw_atl_a0_hw_ring_tx_start(struct aq_hw_s *self,
 				      struct aq_ring_s *ring)
 {
-	tdm_tx_desc_en_set(self, 1, ring->idx);
+	hw_atl_tdm_tx_desc_en_set(self, 1, ring->idx);
+
 	return aq_hw_err_from_flags(self);
 }
 
 static int hw_atl_a0_hw_ring_rx_start(struct aq_hw_s *self,
 				      struct aq_ring_s *ring)
 {
-	rdm_rx_desc_en_set(self, 1, ring->idx);
+	hw_atl_rdm_rx_desc_en_set(self, 1, ring->idx);
+
 	return aq_hw_err_from_flags(self);
 }
 
 static int hw_atl_a0_hw_start(struct aq_hw_s *self)
 {
-	tpb_tx_buff_en_set(self, 1);
-	rpb_rx_buff_en_set(self, 1);
+	hw_atl_tpb_tx_buff_en_set(self, 1);
+	hw_atl_rpb_rx_buff_en_set(self, 1);
+
 	return aq_hw_err_from_flags(self);
 }
 
 static int hw_atl_a0_hw_tx_ring_tail_update(struct aq_hw_s *self,
 					    struct aq_ring_s *ring)
 {
-	reg_tx_dma_desc_tail_ptr_set(self, ring->sw_tail, ring->idx);
+	hw_atl_reg_tx_dma_desc_tail_ptr_set(self, ring->sw_tail, ring->idx);
+
 	return 0;
 }
 
@@ -393,8 +441,8 @@ static int hw_atl_a0_hw_ring_tx_xmit(struct aq_hw_s *self,
 	struct aq_ring_buff_s *buff = NULL;
 	struct hw_atl_txd_s *txd = NULL;
 	unsigned int buff_pa_len = 0U;
-	unsigned int pkt_len = 0U;
 	unsigned int frag_count = 0U;
+	unsigned int pkt_len = 0U;
 	bool is_gso = false;
 
 	buff = &ring->buff_ring[ring->sw_tail];
@@ -409,7 +457,7 @@ static int hw_atl_a0_hw_ring_tx_xmit(struct aq_hw_s *self,
 
 		buff = &ring->buff_ring[ring->sw_tail];
 
-		if (buff->is_txc) {
+		if (buff->is_gso_tcp) {
 			txd->ctl |= (buff->len_l3 << 31) |
 				(buff->len_l2 << 24) |
 				HW_ATL_A0_TXD_CTL_CMD_TCP |
@@ -458,6 +506,7 @@ static int hw_atl_a0_hw_ring_tx_xmit(struct aq_hw_s *self,
 	}
 
 	hw_atl_a0_hw_tx_ring_tail_update(self, ring);
+
 	return aq_hw_err_from_flags(self);
 }
 
@@ -465,39 +514,40 @@ static int hw_atl_a0_hw_ring_rx_init(struct aq_hw_s *self,
 				     struct aq_ring_s *aq_ring,
 				     struct aq_ring_param_s *aq_ring_param)
 {
-	u32 dma_desc_addr_lsw = (u32)aq_ring->dx_ring_pa;
 	u32 dma_desc_addr_msw = (u32)(((u64)aq_ring->dx_ring_pa) >> 32);
+	u32 dma_desc_addr_lsw = (u32)aq_ring->dx_ring_pa;
 
-	rdm_rx_desc_en_set(self, false, aq_ring->idx);
+	hw_atl_rdm_rx_desc_en_set(self, false, aq_ring->idx);
 
-	rdm_rx_desc_head_splitting_set(self, 0U, aq_ring->idx);
+	hw_atl_rdm_rx_desc_head_splitting_set(self, 0U, aq_ring->idx);
 
-	reg_rx_dma_desc_base_addresslswset(self, dma_desc_addr_lsw,
-					   aq_ring->idx);
+	hw_atl_reg_rx_dma_desc_base_addresslswset(self, dma_desc_addr_lsw,
+						  aq_ring->idx);
 
-	reg_rx_dma_desc_base_addressmswset(self,
-					   dma_desc_addr_msw, aq_ring->idx);
+	hw_atl_reg_rx_dma_desc_base_addressmswset(self,
+						  dma_desc_addr_msw,
+						  aq_ring->idx);
 
-	rdm_rx_desc_len_set(self, aq_ring->size / 8U, aq_ring->idx);
+	hw_atl_rdm_rx_desc_len_set(self, aq_ring->size / 8U, aq_ring->idx);
 
-	rdm_rx_desc_data_buff_size_set(self,
-				       AQ_CFG_RX_FRAME_MAX / 1024U,
-				       aq_ring->idx);
+	hw_atl_rdm_rx_desc_data_buff_size_set(self,
+					      aq_ring->frame_max / 1024U,
+					      aq_ring->idx);
 
-	rdm_rx_desc_head_buff_size_set(self, 0U, aq_ring->idx);
-	rdm_rx_desc_head_splitting_set(self, 0U, aq_ring->idx);
-	rpo_rx_desc_vlan_stripping_set(self, 0U, aq_ring->idx);
+	hw_atl_rdm_rx_desc_head_buff_size_set(self, 0U, aq_ring->idx);
+	hw_atl_rdm_rx_desc_head_splitting_set(self, 0U, aq_ring->idx);
+	hw_atl_rpo_rx_desc_vlan_stripping_set(self, 0U, aq_ring->idx);
 
 	/* Rx ring set mode */
 
 	/* Mapping interrupt vector */
-	itr_irq_map_rx_set(self, aq_ring_param->vec_idx, aq_ring->idx);
-	itr_irq_map_en_rx_set(self, true, aq_ring->idx);
+	hw_atl_itr_irq_map_rx_set(self, aq_ring_param->vec_idx, aq_ring->idx);
+	hw_atl_itr_irq_map_en_rx_set(self, true, aq_ring->idx);
 
-	rdm_cpu_id_set(self, aq_ring_param->cpu, aq_ring->idx);
-	rdm_rx_desc_dca_en_set(self, 0U, aq_ring->idx);
-	rdm_rx_head_dca_en_set(self, 0U, aq_ring->idx);
-	rdm_rx_pld_dca_en_set(self, 0U, aq_ring->idx);
+	hw_atl_rdm_cpu_id_set(self, aq_ring_param->cpu, aq_ring->idx);
+	hw_atl_rdm_rx_desc_dca_en_set(self, 0U, aq_ring->idx);
+	hw_atl_rdm_rx_head_dca_en_set(self, 0U, aq_ring->idx);
+	hw_atl_rdm_rx_pld_dca_en_set(self, 0U, aq_ring->idx);
 
 	return aq_hw_err_from_flags(self);
 }
@@ -506,28 +556,28 @@ static int hw_atl_a0_hw_ring_tx_init(struct aq_hw_s *self,
 				     struct aq_ring_s *aq_ring,
 				     struct aq_ring_param_s *aq_ring_param)
 {
-	u32 dma_desc_lsw_addr = (u32)aq_ring->dx_ring_pa;
 	u32 dma_desc_msw_addr = (u32)(((u64)aq_ring->dx_ring_pa) >> 32);
+	u32 dma_desc_lsw_addr = (u32)aq_ring->dx_ring_pa;
 
-	reg_tx_dma_desc_base_addresslswset(self, dma_desc_lsw_addr,
-					   aq_ring->idx);
+	hw_atl_reg_tx_dma_desc_base_addresslswset(self, dma_desc_lsw_addr,
+						  aq_ring->idx);
 
-	reg_tx_dma_desc_base_addressmswset(self, dma_desc_msw_addr,
-					   aq_ring->idx);
+	hw_atl_reg_tx_dma_desc_base_addressmswset(self, dma_desc_msw_addr,
+						  aq_ring->idx);
 
-	tdm_tx_desc_len_set(self, aq_ring->size / 8U, aq_ring->idx);
+	hw_atl_tdm_tx_desc_len_set(self, aq_ring->size / 8U, aq_ring->idx);
 
 	hw_atl_a0_hw_tx_ring_tail_update(self, aq_ring);
 
 	/* Set Tx threshold */
-	tdm_tx_desc_wr_wb_threshold_set(self, 0U, aq_ring->idx);
+	hw_atl_tdm_tx_desc_wr_wb_threshold_set(self, 0U, aq_ring->idx);
 
 	/* Mapping interrupt vector */
-	itr_irq_map_tx_set(self, aq_ring_param->vec_idx, aq_ring->idx);
-	itr_irq_map_en_tx_set(self, true, aq_ring->idx);
+	hw_atl_itr_irq_map_tx_set(self, aq_ring_param->vec_idx, aq_ring->idx);
+	hw_atl_itr_irq_map_en_tx_set(self, true, aq_ring->idx);
 
-	tdm_cpu_id_set(self, aq_ring_param->cpu, aq_ring->idx);
-	tdm_tx_desc_dca_en_set(self, 0U, aq_ring->idx);
+	hw_atl_tdm_cpu_id_set(self, aq_ring_param->cpu, aq_ring->idx);
+	hw_atl_tdm_tx_desc_dca_en_set(self, 0U, aq_ring->idx);
 
 	return aq_hw_err_from_flags(self);
 }
@@ -548,7 +598,7 @@ static int hw_atl_a0_hw_ring_rx_fill(struct aq_hw_s *self,
 		rxd->hdr_addr = 0U;
 	}
 
-	reg_rx_dma_desc_tail_ptr_set(self, sw_tail_old, ring->idx);
+	hw_atl_reg_rx_dma_desc_tail_ptr_set(self, sw_tail_old, ring->idx);
 
 	return aq_hw_err_from_flags(self);
 }
@@ -556,14 +606,14 @@ static int hw_atl_a0_hw_ring_rx_fill(struct aq_hw_s *self,
 static int hw_atl_a0_hw_ring_tx_head_update(struct aq_hw_s *self,
 					    struct aq_ring_s *ring)
 {
+	unsigned int hw_head = hw_atl_tdm_tx_desc_head_ptr_get(self, ring->idx);
 	int err = 0;
-	unsigned int hw_head_ = tdm_tx_desc_head_ptr_get(self, ring->idx);
 
-	if (aq_utils_obj_test(&self->header.flags, AQ_HW_FLAG_ERR_UNPLUG)) {
+	if (aq_utils_obj_test(&self->flags, AQ_HW_FLAG_ERR_UNPLUG)) {
 		err = -ENXIO;
 		goto err_exit;
 	}
-	ring->hw_head = hw_head_;
+	ring->hw_head = hw_head;
 	err = aq_hw_err_from_flags(self);
 
 err_exit:
@@ -573,8 +623,6 @@ err_exit:
 static int hw_atl_a0_hw_ring_rx_receive(struct aq_hw_s *self,
 					struct aq_ring_s *ring)
 {
-	struct device *ndev = aq_nic_get_dev(ring->aq_nic);
-
 	for (; ring->hw_head != ring->sw_tail;
 		ring->hw_head = aq_ring_next_dx(ring, ring->hw_head)) {
 		struct aq_ring_buff_s *buff = NULL;
@@ -587,15 +635,16 @@ static int hw_atl_a0_hw_ring_rx_receive(struct aq_hw_s *self,
 
 		if (!(rxd_wb->status & 0x5U)) { /* RxD is not done */
 			if ((1U << 4) &
-				reg_rx_dma_desc_status_get(self, ring->idx)) {
-			rdm_rx_desc_en_set(self, false, ring->idx);
-			rdm_rx_desc_res_set(self, true, ring->idx);
-			rdm_rx_desc_res_set(self, false, ring->idx);
-			rdm_rx_desc_en_set(self, true, ring->idx);
+			hw_atl_reg_rx_dma_desc_status_get(self, ring->idx)) {
+				hw_atl_rdm_rx_desc_en_set(self, false, ring->idx);
+				hw_atl_rdm_rx_desc_res_set(self, true, ring->idx);
+				hw_atl_rdm_rx_desc_res_set(self, false, ring->idx);
+				hw_atl_rdm_rx_desc_en_set(self, true, ring->idx);
 			}
 
 			if (ring->hw_head ||
-			    (rdm_rx_desc_head_ptr_get(self, ring->idx) < 2U)) {
+			    (hw_atl_rdm_rx_desc_head_ptr_get(self,
+							     ring->idx) < 2U)) {
 				break;
 			} else if (!(rxd_wb->status & 0x1U)) {
 				struct hw_atl_rxd_wb_s *rxd_wb1 =
@@ -629,13 +678,17 @@ static int hw_atl_a0_hw_ring_rx_receive(struct aq_hw_s *self,
 				buff->is_udp_cso = (is_err & 0x10U) ? 0 : 1;
 			else if (0x0U == (pkt_type & 0x1CU))
 				buff->is_tcp_cso = (is_err & 0x10U) ? 0 : 1;
+
+			/* Checksum offload workaround for small packets */
+			if (rxd_wb->pkt_len <= 60) {
+				buff->is_ip_cso = 0U;
+				buff->is_cso_err = 0U;
+			}
 		}
 
 		is_err &= ~0x18U;
 		is_err &= ~0x04U;
 
-		dma_unmap_page(ndev, buff->pa, buff->len, DMA_FROM_DEVICE);
-
 		if (is_err || rxd_wb->type & 0x1000U) {
 			/* status error or DMA error */
 			buff->is_error = 1U;
@@ -653,9 +706,9 @@ static int hw_atl_a0_hw_ring_rx_receive(struct aq_hw_s *self,
 
 			if (HW_ATL_A0_RXD_WB_STAT2_EOP & rxd_wb->status) {
 				buff->len = rxd_wb->pkt_len %
-					AQ_CFG_RX_FRAME_MAX;
+					    ring->frame_max;
 				buff->len = buff->len ?
-					buff->len : AQ_CFG_RX_FRAME_MAX;
+					buff->len : ring->frame_max;
 				buff->next = 0U;
 				buff->is_eop = 1U;
 			} else {
@@ -672,26 +725,27 @@ static int hw_atl_a0_hw_ring_rx_receive(struct aq_hw_s *self,
 
 static int hw_atl_a0_hw_irq_enable(struct aq_hw_s *self, u64 mask)
 {
-	itr_irq_msk_setlsw_set(self, LODWORD(mask) |
+	hw_atl_itr_irq_msk_setlsw_set(self, LODWORD(mask) |
 			       (1U << HW_ATL_A0_ERR_INT));
+
 	return aq_hw_err_from_flags(self);
 }
 
 static int hw_atl_a0_hw_irq_disable(struct aq_hw_s *self, u64 mask)
 {
-	itr_irq_msk_clearlsw_set(self, LODWORD(mask));
-	itr_irq_status_clearlsw_set(self, LODWORD(mask));
-
-	if ((1U << 16) & reg_gen_irq_status_get(self))
+	hw_atl_itr_irq_msk_clearlsw_set(self, LODWORD(mask));
+	hw_atl_itr_irq_status_clearlsw_set(self, LODWORD(mask));
 
-		atomic_inc(&PHAL_ATLANTIC_A0->dpc);
+	if ((1U << 16) & hw_atl_reg_gen_irq_status_get(self))
+		atomic_inc(&self->dpc);
 
 	return aq_hw_err_from_flags(self);
 }
 
 static int hw_atl_a0_hw_irq_read(struct aq_hw_s *self, u64 *mask)
 {
-	*mask = itr_irq_statuslsw_get(self);
+	*mask = hw_atl_itr_irq_statuslsw_get(self);
+
 	return aq_hw_err_from_flags(self);
 }
 
@@ -700,20 +754,22 @@ static int hw_atl_a0_hw_irq_read(struct aq_hw_s *self, u64 *mask)
 static int hw_atl_a0_hw_packet_filter_set(struct aq_hw_s *self,
 					  unsigned int packet_filter)
 {
+	struct aq_nic_cfg_s *cfg = self->aq_nic_cfg;
 	unsigned int i = 0U;
 
-	rpfl2promiscuous_mode_en_set(self, IS_FILTER_ENABLED(IFF_PROMISC));
-	rpfl2multicast_flr_en_set(self, IS_FILTER_ENABLED(IFF_MULTICAST), 0);
-	rpfl2broadcast_en_set(self, IS_FILTER_ENABLED(IFF_BROADCAST));
+	hw_atl_rpfl2promiscuous_mode_en_set(self,
+					    IS_FILTER_ENABLED(IFF_PROMISC));
+	hw_atl_rpfl2multicast_flr_en_set(self,
+					 IS_FILTER_ENABLED(IFF_MULTICAST), 0);
+	hw_atl_rpfl2broadcast_en_set(self, IS_FILTER_ENABLED(IFF_BROADCAST));
 
-	self->aq_nic_cfg->is_mc_list_enabled =
-			IS_FILTER_ENABLED(IFF_MULTICAST);
+	cfg->is_mc_list_enabled = IS_FILTER_ENABLED(IFF_MULTICAST);
 
 	for (i = HW_ATL_A0_MAC_MIN; i < HW_ATL_A0_MAC_MAX; ++i)
-		rpfl2_uc_flr_en_set(self,
-				    (self->aq_nic_cfg->is_mc_list_enabled &&
-				    (i <= self->aq_nic_cfg->mc_list_count)) ?
-				    1U : 0U, i);
+		hw_atl_rpfl2_uc_flr_en_set(self,
+					   (cfg->is_mc_list_enabled &&
+					    (i <= cfg->mc_list_count)) ? 1U : 0U,
+					   i);
 
 	return aq_hw_err_from_flags(self);
 }
@@ -722,35 +778,36 @@ static int hw_atl_a0_hw_packet_filter_set(struct aq_hw_s *self,
 
 static int hw_atl_a0_hw_multicast_list_set(struct aq_hw_s *self,
 					   u8 ar_mac
-					   [AQ_CFG_MULTICAST_ADDRESS_MAX]
+					   [AQ_HW_MULTICAST_ADDRESS_MAX]
 					   [ETH_ALEN],
 					   u32 count)
 {
+	struct aq_nic_cfg_s *cfg = self->aq_nic_cfg;
 	int err = 0;
 
 	if (count > (HW_ATL_A0_MAC_MAX - HW_ATL_A0_MAC_MIN)) {
-		err = EBADRQC;
+		err = -EBADRQC;
 		goto err_exit;
 	}
-	for (self->aq_nic_cfg->mc_list_count = 0U;
-			self->aq_nic_cfg->mc_list_count < count;
-			++self->aq_nic_cfg->mc_list_count) {
-		u32 i = self->aq_nic_cfg->mc_list_count;
+	for (cfg->mc_list_count = 0U; cfg->mc_list_count < count; ++cfg->mc_list_count) {
+		u32 i = cfg->mc_list_count;
 		u32 h = (ar_mac[i][0] << 8) | (ar_mac[i][1]);
 		u32 l = (ar_mac[i][2] << 24) | (ar_mac[i][3] << 16) |
-					(ar_mac[i][4] << 8) | ar_mac[i][5];
+			(ar_mac[i][4] << 8) | ar_mac[i][5];
 
-		rpfl2_uc_flr_en_set(self, 0U, HW_ATL_A0_MAC_MIN + i);
+		hw_atl_rpfl2_uc_flr_en_set(self, 0U, HW_ATL_A0_MAC_MIN + i);
 
-		rpfl2unicast_dest_addresslsw_set(self,
-						 l, HW_ATL_A0_MAC_MIN + i);
+		hw_atl_rpfl2unicast_dest_addresslsw_set(self,
+							l,
+							HW_ATL_A0_MAC_MIN + i);
 
-		rpfl2unicast_dest_addressmsw_set(self,
-						 h, HW_ATL_A0_MAC_MIN + i);
+		hw_atl_rpfl2unicast_dest_addressmsw_set(self,
+							h,
+							HW_ATL_A0_MAC_MIN + i);
 
-		rpfl2_uc_flr_en_set(self,
-				    (self->aq_nic_cfg->is_mc_list_enabled),
-				    HW_ATL_A0_MAC_MIN + i);
+		hw_atl_rpfl2_uc_flr_en_set(self,
+					   (cfg->is_mc_list_enabled),
+					   HW_ATL_A0_MAC_MIN + i);
 	}
 
 	err = aq_hw_err_from_flags(self);
@@ -759,24 +816,23 @@ err_exit:
 	return err;
 }
 
-static int hw_atl_a0_hw_interrupt_moderation_set(struct aq_hw_s *self,
-						 bool itr_enabled)
+static int hw_atl_a0_hw_interrupt_moderation_set(struct aq_hw_s *self)
 {
 	unsigned int i = 0U;
+	u32 itr_rx;
 
-	if (itr_enabled && self->aq_nic_cfg->itr) {
-		if (self->aq_nic_cfg->itr != 0xFFFFU) {
+	if (self->aq_nic_cfg->itr) {
+		if (self->aq_nic_cfg->itr != AQ_CFG_INTERRUPT_MODERATION_AUTO) {
 			u32 itr_ = (self->aq_nic_cfg->itr >> 1);
 
 			itr_ = min(AQ_CFG_IRQ_MASK, itr_);
 
-			PHAL_ATLANTIC_A0->itr_rx = 0x80000000U |
-					(itr_ << 0x10);
+			itr_rx = 0x80000000U | (itr_ << 0x10);
 		} else  {
 			u32 n = 0xFFFFU & aq_hw_read_reg(self, 0x00002A00U);
 
 			if (n < self->aq_link_status.mbps) {
-				PHAL_ATLANTIC_A0->itr_rx = 0U;
+				itr_rx = 0U;
 			} else {
 				static unsigned int hw_timers_tbl_[] = {
 					0x01CU, /* 10Gbit */
@@ -791,8 +847,7 @@ static int hw_atl_a0_hw_interrupt_moderation_set(struct aq_hw_s *self,
 					hw_atl_utils_mbps_2_speed_index(
 						self->aq_link_status.mbps);
 
-				PHAL_ATLANTIC_A0->itr_rx =
-					0x80000000U |
+				itr_rx = 0x80000000U |
 					(hw_timers_tbl_[speed_index] << 0x10U);
 			}
 
@@ -800,11 +855,11 @@ static int hw_atl_a0_hw_interrupt_moderation_set(struct aq_hw_s *self,
 			aq_hw_write_reg(self, 0x00002A00U, 0x8D000000U);
 		}
 	} else {
-		PHAL_ATLANTIC_A0->itr_rx = 0U;
+		itr_rx = 0U;
 	}
 
 	for (i = HW_ATL_A0_RINGS_MAX; i--;)
-		reg_irq_thr_set(self, PHAL_ATLANTIC_A0->itr_rx, i);
+		hw_atl_reg_irq_thr_set(self, itr_rx, i);
 
 	return aq_hw_err_from_flags(self);
 }
@@ -812,47 +867,88 @@ static int hw_atl_a0_hw_interrupt_moderation_set(struct aq_hw_s *self,
 static int hw_atl_a0_hw_stop(struct aq_hw_s *self)
 {
 	hw_atl_a0_hw_irq_disable(self, HW_ATL_A0_INT_MASK);
+
 	return aq_hw_err_from_flags(self);
 }
 
 static int hw_atl_a0_hw_ring_tx_stop(struct aq_hw_s *self,
 				     struct aq_ring_s *ring)
 {
-	tdm_tx_desc_en_set(self, 0U, ring->idx);
+	hw_atl_tdm_tx_desc_en_set(self, 0U, ring->idx);
+
 	return aq_hw_err_from_flags(self);
 }
 
 static int hw_atl_a0_hw_ring_rx_stop(struct aq_hw_s *self,
 				     struct aq_ring_s *ring)
 {
-	rdm_rx_desc_en_set(self, 0U, ring->idx);
+	hw_atl_rdm_rx_desc_en_set(self, 0U, ring->idx);
+
 	return aq_hw_err_from_flags(self);
 }
 
-static int hw_atl_a0_hw_set_speed(struct aq_hw_s *self, u32 speed)
+static int hw_atl_a0_hw_fl3l4_clear(struct aq_hw_s *self,
+				    struct aq_rx_filter_l3l4 *data)
 {
-	int err = 0;
+	u8 location = data->location;
 
-	err = hw_atl_utils_mpi_set_speed(self, speed, MPI_INIT);
-	if (err < 0)
-		goto err_exit;
+	if (!data->is_ipv6) {
+		hw_atl_rpfl3l4_cmd_clear(self, location);
+		hw_atl_rpf_l4_spd_set(self, 0U, location);
+		hw_atl_rpf_l4_dpd_set(self, 0U, location);
+		hw_atl_rpfl3l4_ipv4_src_addr_clear(self, location);
+		hw_atl_rpfl3l4_ipv4_dest_addr_clear(self, location);
+	} else {
+		int i;
 
-err_exit:
-	return err;
+		for (i = 0; i < HW_ATL_RX_CNT_REG_ADDR_IPV6; ++i) {
+			hw_atl_rpfl3l4_cmd_clear(self, location + i);
+			hw_atl_rpf_l4_spd_set(self, 0U, location + i);
+			hw_atl_rpf_l4_dpd_set(self, 0U, location + i);
+		}
+		hw_atl_rpfl3l4_ipv6_src_addr_clear(self, location);
+		hw_atl_rpfl3l4_ipv6_dest_addr_clear(self, location);
+	}
+
+	return aq_hw_err_from_flags(self);
 }
 
-static struct aq_hw_ops hw_atl_ops_ = {
-	.create               = hw_atl_a0_create,
-	.destroy              = hw_atl_a0_destroy,
-	.get_hw_caps          = hw_atl_a0_get_hw_caps,
+static int hw_atl_a0_hw_fl3l4_set(struct aq_hw_s *self,
+				  struct aq_rx_filter_l3l4 *data)
+{
+	u8 location = data->location;
+
+	hw_atl_a0_hw_fl3l4_clear(self, data);
 
-	.hw_get_mac_permanent = hw_atl_utils_get_mac_permanent,
+	if (data->cmd) {
+		if (!data->is_ipv6) {
+			hw_atl_rpfl3l4_ipv4_dest_addr_set(self,
+							  location,
+							  data->ip_dst[0]);
+			hw_atl_rpfl3l4_ipv4_src_addr_set(self,
+							 location,
+							 data->ip_src[0]);
+		} else {
+			hw_atl_rpfl3l4_ipv6_dest_addr_set(self,
+							  location,
+							  data->ip_dst);
+			hw_atl_rpfl3l4_ipv6_src_addr_set(self,
+							 location,
+							 data->ip_src);
+		}
+	}
+	hw_atl_rpf_l4_dpd_set(self, data->p_dst, location);
+	hw_atl_rpf_l4_spd_set(self, data->p_src, location);
+	hw_atl_rpfl3l4_cmd_set(self, location, data->cmd);
+
+	return aq_hw_err_from_flags(self);
+}
+
+const struct aq_hw_ops hw_atl_ops_a0 = {
+	.hw_soft_reset        = hw_atl_utils_soft_reset,
+	.hw_prepare           = hw_atl_utils_initfw,
 	.hw_set_mac_address   = hw_atl_a0_hw_mac_addr_set,
-	.hw_get_link_status   = hw_atl_utils_mpi_get_link_status,
-	.hw_set_link_speed    = hw_atl_a0_hw_set_speed,
 	.hw_init              = hw_atl_a0_hw_init,
-	.hw_deinit            = hw_atl_utils_hw_deinit,
-	.hw_set_power         = hw_atl_utils_hw_set_power,
 	.hw_reset             = hw_atl_a0_hw_reset,
 	.hw_start             = hw_atl_a0_hw_start,
 	.hw_ring_tx_start     = hw_atl_a0_hw_ring_tx_start,
@@ -874,6 +970,7 @@ static struct aq_hw_ops hw_atl_ops_ = {
 	.hw_ring_rx_init             = hw_atl_a0_hw_ring_rx_init,
 	.hw_ring_tx_init             = hw_atl_a0_hw_ring_tx_init,
 	.hw_packet_filter_set        = hw_atl_a0_hw_packet_filter_set,
+	.hw_filter_l3l4_set          = hw_atl_a0_hw_fl3l4_set,
 	.hw_multicast_list_set       = hw_atl_a0_hw_multicast_list_set,
 	.hw_interrupt_moderation_set = hw_atl_a0_hw_interrupt_moderation_set,
 	.hw_rss_set                  = hw_atl_a0_hw_rss_set,
@@ -882,17 +979,3 @@ static struct aq_hw_ops hw_atl_ops_ = {
 	.hw_get_hw_stats             = hw_atl_utils_get_hw_stats,
 	.hw_get_fw_version           = hw_atl_utils_get_fw_version,
 };
-
-struct aq_hw_ops *hw_atl_a0_get_ops_by_id(struct pci_dev *pdev)
-{
-	bool is_vid_ok = (pdev->vendor == PCI_VENDOR_ID_AQUANTIA);
-	bool is_did_ok = ((pdev->device == HW_ATL_DEVICE_ID_0001) ||
-			(pdev->device == HW_ATL_DEVICE_ID_D100) ||
-			(pdev->device == HW_ATL_DEVICE_ID_D107) ||
-			(pdev->device == HW_ATL_DEVICE_ID_D108) ||
-			(pdev->device == HW_ATL_DEVICE_ID_D109));
-
-	bool is_rev_ok = (pdev->revision == 1U);
-
-	return (is_vid_ok && is_did_ok && is_rev_ok) ? &hw_atl_ops_ : NULL;
-}
diff --git a/drivers/net/ethernet/aquantia/atlantic/hw_atl/hw_atl_a0.h b/drivers/net/ethernet/aquantia/atlantic/hw_atl/hw_atl_a0.h
index 6e1d527954c9..1a294ad4d17d 100644
--- a/drivers/net/ethernet/aquantia/atlantic/hw_atl/hw_atl_a0.h
+++ b/drivers/net/ethernet/aquantia/atlantic/hw_atl/hw_atl_a0.h
@@ -1,10 +1,7 @@
+/* SPDX-License-Identifier: GPL-2.0-only */
 /*
  * aQuantia Corporation Network Driver
  * Copyright (C) 2014-2017 aQuantia Corporation. All rights reserved
- *
- * This program is free software; you can redistribute it and/or modify it
- * under the terms and conditions of the GNU General Public License,
- * version 2, as published by the Free Software Foundation.
  */
 
 /* File hw_atl_a0.h: Declaration of abstract interface for Atlantic hardware
@@ -16,19 +13,11 @@
 
 #include "../aq_common.h"
 
-#ifndef PCI_VENDOR_ID_AQUANTIA
+extern const struct aq_hw_caps_s hw_atl_a0_caps_aqc100;
+extern const struct aq_hw_caps_s hw_atl_a0_caps_aqc107;
+extern const struct aq_hw_caps_s hw_atl_a0_caps_aqc108;
+extern const struct aq_hw_caps_s hw_atl_a0_caps_aqc109;
 
-#define PCI_VENDOR_ID_AQUANTIA  0x1D6A
-#define HW_ATL_DEVICE_ID_0001   0x0001
-#define HW_ATL_DEVICE_ID_D100   0xD100
-#define HW_ATL_DEVICE_ID_D107   0xD107
-#define HW_ATL_DEVICE_ID_D108   0xD108
-#define HW_ATL_DEVICE_ID_D109   0xD109
-
-#define HW_ATL_NIC_NAME "aQuantia AQtion 5Gbit Network Adapter"
-
-#endif
-
-struct aq_hw_ops *hw_atl_a0_get_ops_by_id(struct pci_dev *pdev);
+extern const struct aq_hw_ops hw_atl_ops_a0;
 
 #endif /* HW_ATL_A0_H */
diff --git a/drivers/net/ethernet/aquantia/atlantic/hw_atl/hw_atl_a0_internal.h b/drivers/net/ethernet/aquantia/atlantic/hw_atl/hw_atl_a0_internal.h
index 0592a0330cf0..1b9b2e5601ed 100644
--- a/drivers/net/ethernet/aquantia/atlantic/hw_atl/hw_atl_a0_internal.h
+++ b/drivers/net/ethernet/aquantia/atlantic/hw_atl/hw_atl_a0_internal.h
@@ -1,10 +1,7 @@
+/* SPDX-License-Identifier: GPL-2.0-only */
 /*
  * aQuantia Corporation Network Driver
  * Copyright (C) 2014-2017 aQuantia Corporation. All rights reserved
- *
- * This program is free software; you can redistribute it and/or modify it
- * under the terms and conditions of the GNU General Public License,
- * version 2, as published by the Free Software Foundation.
  */
 
 /* File hw_atl_a0_internal.h: Definition of Atlantic A0 chip specific
@@ -62,12 +59,6 @@
 #define HW_ATL_A0_MPI_SPEED_MSK       0xFFFFU
 #define HW_ATL_A0_MPI_SPEED_SHIFT     16U
 
-#define HW_ATL_A0_RATE_10G            BIT(0)
-#define HW_ATL_A0_RATE_5G             BIT(1)
-#define HW_ATL_A0_RATE_2G5            BIT(3)
-#define HW_ATL_A0_RATE_1G             BIT(4)
-#define HW_ATL_A0_RATE_100M           BIT(5)
-
 #define HW_ATL_A0_TXBUF_MAX 160U
 #define HW_ATL_A0_RXBUF_MAX 320U
 
@@ -88,69 +79,12 @@
 
 #define HW_ATL_A0_FW_VER_EXPECTED 0x01050006U
 
-/* Hardware tx descriptor */
-struct __packed hw_atl_txd_s {
-	u64 buf_addr;
-	u32 ctl;
-	u32 ctl2; /* 63..46 - payload length, 45 - ctx enable, 44 - ctx index */
-};
-
-/* Hardware tx context descriptor */
-struct __packed hw_atl_txc_s {
-	u32 rsvd;
-	u32 len;
-	u32 ctl;
-	u32 len2;
-};
-
-/* Hardware rx descriptor */
-struct __packed hw_atl_rxd_s {
-	u64 buf_addr;
-	u64 hdr_addr;
-};
-
-/* Hardware rx descriptor writeback */
-struct __packed hw_atl_rxd_wb_s {
-	u32 type;
-	u32 rss_hash;
-	u16 status;
-	u16 pkt_len;
-	u16 next_desc_ptr;
-	u16 vlan;
-};
-
-/* HW layer capabilities */
-static struct aq_hw_caps_s hw_atl_a0_hw_caps_ = {
-	.ports = 1U,
-	.is_64_dma = true,
-	.msix_irqs = 4U,
-	.irq_mask = ~0U,
-	.vecs = HW_ATL_A0_RSS_MAX,
-	.tcs = HW_ATL_A0_TC_MAX,
-	.rxd_alignment = 1U,
-	.rxd_size = HW_ATL_A0_RXD_SIZE,
-	.rxds = 248U,
-	.txd_alignment = 1U,
-	.txd_size = HW_ATL_A0_TXD_SIZE,
-	.txds = 8U * 1024U,
-	.txhwb_alignment = 4096U,
-	.tx_rings = HW_ATL_A0_TX_RINGS,
-	.rx_rings = HW_ATL_A0_RX_RINGS,
-	.hw_features = NETIF_F_HW_CSUM |
-			NETIF_F_RXCSUM |
-			NETIF_F_RXHASH |
-			NETIF_F_SG |
-			NETIF_F_TSO,
-	.hw_priv_flags = IFF_UNICAST_FLT,
-	.link_speed_msk = (HW_ATL_A0_RATE_10G |
-			HW_ATL_A0_RATE_5G |
-			HW_ATL_A0_RATE_2G5 |
-			HW_ATL_A0_RATE_1G |
-			HW_ATL_A0_RATE_100M),
-	.flow_control = true,
-	.mtu = HW_ATL_A0_MTU_JUMBO,
-	.mac_regs_count = 88,
-	.fw_ver_expected = HW_ATL_A0_FW_VER_EXPECTED,
-};
+#define HW_ATL_A0_MIN_RXD \
+	(ALIGN(AQ_CFG_SKB_FRAGS_MAX + 1U, AQ_HW_RXD_MULTIPLE))
+#define HW_ATL_A0_MIN_TXD \
+	(ALIGN(AQ_CFG_SKB_FRAGS_MAX + 1U, AQ_HW_TXD_MULTIPLE))
+
+#define HW_ATL_A0_MAX_RXD 8184U
+#define HW_ATL_A0_MAX_TXD 8184U
 
 #endif /* HW_ATL_A0_INTERNAL_H */
diff --git a/drivers/net/ethernet/aquantia/atlantic/hw_atl/hw_atl_b0.c b/drivers/net/ethernet/aquantia/atlantic/hw_atl/hw_atl_b0.c
index 1bceb7358e5c..493432d036b9 100644
--- a/drivers/net/ethernet/aquantia/atlantic/hw_atl/hw_atl_b0.c
+++ b/drivers/net/ethernet/aquantia/atlantic/hw_atl/hw_atl_b0.c
@@ -1,10 +1,8 @@
-/*
- * aQuantia Corporation Network Driver
- * Copyright (C) 2014-2017 aQuantia Corporation. All rights reserved
+// SPDX-License-Identifier: GPL-2.0-only
+/* Atlantic Network Driver
  *
- * This program is free software; you can redistribute it and/or modify it
- * under the terms and conditions of the GNU General Public License,
- * version 2, as published by the Free Software Foundation.
+ * Copyright (C) 2014-2019 aQuantia Corporation
+ * Copyright (C) 2019-2020 Marvell International Ltd.
  */
 
 /* File hw_atl_b0.c: Definition of Atlantic hardware specific functions. */
@@ -12,148 +10,220 @@
 #include "../aq_hw.h"
 #include "../aq_hw_utils.h"
 #include "../aq_ring.h"
+#include "../aq_nic.h"
+#include "../aq_phy.h"
 #include "hw_atl_b0.h"
 #include "hw_atl_utils.h"
 #include "hw_atl_llh.h"
 #include "hw_atl_b0_internal.h"
+#include "hw_atl_llh_internal.h"
+
+#define DEFAULT_B0_BOARD_BASIC_CAPABILITIES \
+	.is_64_dma = true,		  \
+	.op64bit = false,		  \
+	.msix_irqs = 8U,		  \
+	.irq_mask = ~0U,		  \
+	.vecs = HW_ATL_B0_RSS_MAX,	  \
+	.tcs_max = HW_ATL_B0_TC_MAX,	  \
+	.rxd_alignment = 1U,		  \
+	.rxd_size = HW_ATL_B0_RXD_SIZE,   \
+	.rxds_max = HW_ATL_B0_MAX_RXD,    \
+	.rxds_min = HW_ATL_B0_MIN_RXD,    \
+	.txd_alignment = 1U,		  \
+	.txd_size = HW_ATL_B0_TXD_SIZE,   \
+	.txds_max = HW_ATL_B0_MAX_TXD,    \
+	.txds_min = HW_ATL_B0_MIN_TXD,    \
+	.txhwb_alignment = 4096U,	  \
+	.tx_rings = HW_ATL_B0_TX_RINGS,   \
+	.rx_rings = HW_ATL_B0_RX_RINGS,   \
+	.hw_features = NETIF_F_HW_CSUM |  \
+			NETIF_F_RXCSUM |  \
+			NETIF_F_RXHASH |  \
+			NETIF_F_SG |      \
+			NETIF_F_TSO |     \
+			NETIF_F_TSO6 |    \
+			NETIF_F_LRO |     \
+			NETIF_F_NTUPLE |  \
+			NETIF_F_HW_VLAN_CTAG_FILTER | \
+			NETIF_F_HW_VLAN_CTAG_RX |     \
+			NETIF_F_HW_VLAN_CTAG_TX |     \
+			NETIF_F_GSO_UDP_L4      |     \
+			NETIF_F_GSO_PARTIAL |         \
+			NETIF_F_HW_TC,                \
+	.hw_priv_flags = IFF_UNICAST_FLT, \
+	.flow_control = true,		  \
+	.mtu = HW_ATL_B0_MTU_JUMBO,	  \
+	.mac_regs_count = 88,		  \
+	.hw_alive_check_addr = 0x10U
+
+const struct aq_hw_caps_s hw_atl_b0_caps_aqc100 = {
+	DEFAULT_B0_BOARD_BASIC_CAPABILITIES,
+	.media_type = AQ_HW_MEDIA_TYPE_FIBRE,
+	.link_speed_msk = AQ_NIC_RATE_10G |
+			  AQ_NIC_RATE_5G |
+			  AQ_NIC_RATE_2G5 |
+			  AQ_NIC_RATE_1G |
+			  AQ_NIC_RATE_100M,
+};
 
-static int hw_atl_b0_get_hw_caps(struct aq_hw_s *self,
-				 struct aq_hw_caps_s *aq_hw_caps)
-{
-	memcpy(aq_hw_caps, &hw_atl_b0_hw_caps_, sizeof(*aq_hw_caps));
-	return 0;
-}
-
-static struct aq_hw_s *hw_atl_b0_create(struct aq_pci_func_s *aq_pci_func,
-					unsigned int port,
-					struct aq_hw_ops *ops)
-{
-	struct hw_atl_s *self = NULL;
-
-	self = kzalloc(sizeof(*self), GFP_KERNEL);
-	if (!self)
-		goto err_exit;
+const struct aq_hw_caps_s hw_atl_b0_caps_aqc107 = {
+	DEFAULT_B0_BOARD_BASIC_CAPABILITIES,
+	.media_type = AQ_HW_MEDIA_TYPE_TP,
+	.link_speed_msk = AQ_NIC_RATE_10G |
+			  AQ_NIC_RATE_5G |
+			  AQ_NIC_RATE_2G5 |
+			  AQ_NIC_RATE_1G |
+			  AQ_NIC_RATE_100M,
+};
 
-	self->base.aq_pci_func = aq_pci_func;
+const struct aq_hw_caps_s hw_atl_b0_caps_aqc108 = {
+	DEFAULT_B0_BOARD_BASIC_CAPABILITIES,
+	.media_type = AQ_HW_MEDIA_TYPE_TP,
+	.link_speed_msk = AQ_NIC_RATE_5G |
+			  AQ_NIC_RATE_2G5 |
+			  AQ_NIC_RATE_1G |
+			  AQ_NIC_RATE_100M,
+};
 
-	self->base.not_ff_addr = 0x10U;
+const struct aq_hw_caps_s hw_atl_b0_caps_aqc109 = {
+	DEFAULT_B0_BOARD_BASIC_CAPABILITIES,
+	.media_type = AQ_HW_MEDIA_TYPE_TP,
+	.link_speed_msk = AQ_NIC_RATE_2G5 |
+			  AQ_NIC_RATE_1G |
+			  AQ_NIC_RATE_100M,
+};
 
-err_exit:
-	return (struct aq_hw_s *)self;
-}
+const struct aq_hw_caps_s hw_atl_b0_caps_aqc111 = {
+	DEFAULT_B0_BOARD_BASIC_CAPABILITIES,
+	.media_type = AQ_HW_MEDIA_TYPE_TP,
+	.link_speed_msk = AQ_NIC_RATE_5G |
+			  AQ_NIC_RATE_2G5 |
+			  AQ_NIC_RATE_1G |
+			  AQ_NIC_RATE_100M,
+	.quirks = AQ_NIC_QUIRK_BAD_PTP,
+};
 
-static void hw_atl_b0_destroy(struct aq_hw_s *self)
-{
-	kfree(self);
-}
+const struct aq_hw_caps_s hw_atl_b0_caps_aqc112 = {
+	DEFAULT_B0_BOARD_BASIC_CAPABILITIES,
+	.media_type = AQ_HW_MEDIA_TYPE_TP,
+	.link_speed_msk = AQ_NIC_RATE_2G5 |
+			  AQ_NIC_RATE_1G  |
+			  AQ_NIC_RATE_100M,
+	.quirks = AQ_NIC_QUIRK_BAD_PTP,
+};
 
 static int hw_atl_b0_hw_reset(struct aq_hw_s *self)
 {
 	int err = 0;
 
-	glb_glb_reg_res_dis_set(self, 1U);
-	pci_pci_reg_res_dis_set(self, 0U);
-	rx_rx_reg_res_dis_set(self, 0U);
-	tx_tx_reg_res_dis_set(self, 0U);
+	err = hw_atl_utils_soft_reset(self);
+	if (err)
+		return err;
 
-	HW_ATL_FLUSH();
-	glb_soft_res_set(self, 1);
+	self->aq_fw_ops->set_state(self, MPI_RESET);
 
-	/* check 10 times by 1ms */
-	AQ_HW_WAIT_FOR(glb_soft_res_get(self) == 0, 1000U, 10U);
-	if (err < 0)
-		goto err_exit;
+	err = aq_hw_err_from_flags(self);
+
+	return err;
+}
 
-	itr_irq_reg_res_dis_set(self, 0U);
-	itr_res_irq_set(self, 1U);
+int hw_atl_b0_set_fc(struct aq_hw_s *self, u32 fc, u32 tc)
+{
+	hw_atl_rpb_rx_xoff_en_per_tc_set(self, !!(fc & AQ_NIC_FC_RX), tc);
 
-	/* check 10 times by 1ms */
-	AQ_HW_WAIT_FOR(itr_res_irq_get(self) == 0, 1000U, 10U);
-	if (err < 0)
-		goto err_exit;
+	return 0;
+}
 
-	hw_atl_utils_mpi_set(self, MPI_RESET, 0x0U);
+static int hw_atl_b0_tc_ptp_set(struct aq_hw_s *self)
+{
+	/* Init TC2 for PTP_TX */
+	hw_atl_tpb_tx_pkt_buff_size_per_tc_set(self, HW_ATL_B0_PTP_TXBUF_SIZE,
+					       AQ_HW_PTP_TC);
 
-	err = aq_hw_err_from_flags(self);
+	/* Init TC2 for PTP_RX */
+	hw_atl_rpb_rx_pkt_buff_size_per_tc_set(self, HW_ATL_B0_PTP_RXBUF_SIZE,
+					       AQ_HW_PTP_TC);
+	/* No flow control for PTP */
+	hw_atl_rpb_rx_xoff_en_per_tc_set(self, 0U, AQ_HW_PTP_TC);
 
-err_exit:
-	return err;
+	return aq_hw_err_from_flags(self);
 }
 
 static int hw_atl_b0_hw_qos_set(struct aq_hw_s *self)
 {
+	struct aq_nic_cfg_s *cfg = self->aq_nic_cfg;
+	u32 tx_buff_size = HW_ATL_B0_TXBUF_MAX;
+	u32 rx_buff_size = HW_ATL_B0_RXBUF_MAX;
+	unsigned int prio = 0U;
 	u32 tc = 0U;
-	u32 buff_size = 0U;
-	unsigned int i_priority = 0U;
-	bool is_rx_flow_control = false;
+
+	if (cfg->is_ptp) {
+		tx_buff_size -= HW_ATL_B0_PTP_TXBUF_SIZE;
+		rx_buff_size -= HW_ATL_B0_PTP_RXBUF_SIZE;
+	}
 
 	/* TPS Descriptor rate init */
-	tps_tx_pkt_shed_desc_rate_curr_time_res_set(self, 0x0U);
-	tps_tx_pkt_shed_desc_rate_lim_set(self, 0xA);
+	hw_atl_tps_tx_pkt_shed_desc_rate_curr_time_res_set(self, 0x0U);
+	hw_atl_tps_tx_pkt_shed_desc_rate_lim_set(self, 0xA);
 
 	/* TPS VM init */
-	tps_tx_pkt_shed_desc_vm_arb_mode_set(self, 0U);
-
-	/* TPS TC credits init */
-	tps_tx_pkt_shed_desc_tc_arb_mode_set(self, 0U);
-	tps_tx_pkt_shed_data_arb_mode_set(self, 0U);
-
-	tps_tx_pkt_shed_tc_data_max_credit_set(self, 0xFFF, 0U);
-	tps_tx_pkt_shed_tc_data_weight_set(self, 0x64, 0U);
-	tps_tx_pkt_shed_desc_tc_max_credit_set(self, 0x50, 0U);
-	tps_tx_pkt_shed_desc_tc_weight_set(self, 0x1E, 0U);
-
-	/* Tx buf size */
-	buff_size = HW_ATL_B0_TXBUF_MAX;
-
-	tpb_tx_pkt_buff_size_per_tc_set(self, buff_size, tc);
-	tpb_tx_buff_hi_threshold_per_tc_set(self,
-					    (buff_size * (1024 / 32U) * 66U) /
-					    100U, tc);
-	tpb_tx_buff_lo_threshold_per_tc_set(self,
-					    (buff_size * (1024 / 32U) * 50U) /
-					    100U, tc);
-
-	/* QoS Rx buf size per TC */
-	tc = 0;
-	is_rx_flow_control = (AQ_NIC_FC_RX & self->aq_nic_cfg->flow_control);
-	buff_size = HW_ATL_B0_RXBUF_MAX;
-
-	rpb_rx_pkt_buff_size_per_tc_set(self, buff_size, tc);
-	rpb_rx_buff_hi_threshold_per_tc_set(self,
-					    (buff_size *
-					    (1024U / 32U) * 66U) /
-					    100U, tc);
-	rpb_rx_buff_lo_threshold_per_tc_set(self,
-					    (buff_size *
-					    (1024U / 32U) * 50U) /
-					    100U, tc);
-	rpb_rx_xoff_en_per_tc_set(self, is_rx_flow_control ? 1U : 0U, tc);
+	hw_atl_tps_tx_pkt_shed_desc_vm_arb_mode_set(self, 0U);
+
+	tx_buff_size /= cfg->tcs;
+	rx_buff_size /= cfg->tcs;
+	for (tc = 0; tc < cfg->tcs; tc++) {
+		u32 threshold = 0U;
+
+		/* Tx buf size TC0 */
+		hw_atl_tpb_tx_pkt_buff_size_per_tc_set(self, tx_buff_size, tc);
+
+		threshold = (tx_buff_size * (1024 / 32U) * 66U) / 100U;
+		hw_atl_tpb_tx_buff_hi_threshold_per_tc_set(self, threshold, tc);
+
+		threshold = (tx_buff_size * (1024 / 32U) * 50U) / 100U;
+		hw_atl_tpb_tx_buff_lo_threshold_per_tc_set(self, threshold, tc);
+
+		/* QoS Rx buf size per TC */
+		hw_atl_rpb_rx_pkt_buff_size_per_tc_set(self, rx_buff_size, tc);
+
+		threshold = (rx_buff_size * (1024U / 32U) * 66U) / 100U;
+		hw_atl_rpb_rx_buff_hi_threshold_per_tc_set(self, threshold, tc);
+
+		threshold = (rx_buff_size * (1024U / 32U) * 50U) / 100U;
+		hw_atl_rpb_rx_buff_lo_threshold_per_tc_set(self, threshold, tc);
+
+		hw_atl_b0_set_fc(self, self->aq_nic_cfg->fc.req, tc);
+	}
+
+	if (cfg->is_ptp)
+		hw_atl_b0_tc_ptp_set(self);
 
 	/* QoS 802.1p priority -> TC mapping */
-	for (i_priority = 8U; i_priority--;)
-		rpf_rpb_user_priority_tc_map_set(self, i_priority, 0U);
+	for (prio = 0; prio < 8; ++prio)
+		hw_atl_rpf_rpb_user_priority_tc_map_set(self, prio,
+							cfg->prio_tc_map[prio]);
 
 	return aq_hw_err_from_flags(self);
 }
 
-static int hw_atl_b0_hw_rss_hash_set(struct aq_hw_s *self,
-				     struct aq_rss_parameters *rss_params)
+int hw_atl_b0_hw_rss_hash_set(struct aq_hw_s *self,
+			      struct aq_rss_parameters *rss_params)
 {
-	struct aq_nic_cfg_s *cfg = NULL;
-	int err = 0;
-	unsigned int i = 0U;
+	struct aq_nic_cfg_s *cfg = self->aq_nic_cfg;
 	unsigned int addr = 0U;
-
-	cfg = self->aq_nic_cfg;
+	unsigned int i = 0U;
+	int err = 0;
+	u32 val;
 
 	for (i = 10, addr = 0U; i--; ++addr) {
 		u32 key_data = cfg->is_rss ?
 			__swab32(rss_params->hash_secret_key[i]) : 0U;
-		rpf_rss_key_wr_data_set(self, key_data);
-		rpf_rss_key_addr_set(self, addr);
-		rpf_rss_key_wr_en_set(self, 1U);
-		AQ_HW_WAIT_FOR(rpf_rss_key_wr_en_get(self) == 0, 1000U, 10U);
+		hw_atl_rpf_rss_key_wr_data_set(self, key_data);
+		hw_atl_rpf_rss_key_addr_set(self, addr);
+		hw_atl_rpf_rss_key_wr_en_set(self, 1U);
+		err = readx_poll_timeout_atomic(hw_atl_rpf_rss_key_wr_en_get,
+						self, val, val == 0,
+						1000U, 10000U);
 		if (err < 0)
 			goto err_exit;
 	}
@@ -167,12 +237,13 @@ err_exit:
 static int hw_atl_b0_hw_rss_set(struct aq_hw_s *self,
 				struct aq_rss_parameters *rss_params)
 {
-	u8 *indirection_table =	rss_params->indirection_table;
-	u32 i = 0U;
 	u32 num_rss_queues = max(1U, self->aq_nic_cfg->num_rss_queues);
+	u8 *indirection_table =	rss_params->indirection_table;
+	u16 bitary[1 + (HW_ATL_B0_RSS_REDIRECTION_MAX *
+		   HW_ATL_B0_RSS_REDIRECTION_BITS / 16U)];
 	int err = 0;
-	u16 bitary[(HW_ATL_B0_RSS_REDIRECTION_MAX *
-					HW_ATL_B0_RSS_REDIRECTION_BITS / 16U)];
+	u32 i = 0U;
+	u32 val;
 
 	memset(bitary, 0, sizeof(bitary));
 
@@ -182,11 +253,13 @@ static int hw_atl_b0_hw_rss_set(struct aq_hw_s *self,
 			((i * 3U) & 0xFU));
 	}
 
-	for (i = AQ_DIMOF(bitary); i--;) {
-		rpf_rss_redir_tbl_wr_data_set(self, bitary[i]);
-		rpf_rss_redir_tbl_addr_set(self, i);
-		rpf_rss_redir_wr_en_set(self, 1U);
-		AQ_HW_WAIT_FOR(rpf_rss_redir_wr_en_get(self) == 0, 1000U, 10U);
+	for (i = ARRAY_SIZE(bitary); i--;) {
+		hw_atl_rpf_rss_redir_tbl_wr_data_set(self, bitary[i]);
+		hw_atl_rpf_rss_redir_tbl_addr_set(self, i);
+		hw_atl_rpf_rss_redir_wr_en_set(self, 1U);
+		err = readx_poll_timeout_atomic(hw_atl_rpf_rss_redir_wr_en_get,
+						self, val, val == 0,
+						1000U, 10000U);
 		if (err < 0)
 			goto err_exit;
 	}
@@ -197,134 +270,274 @@ err_exit:
 	return err;
 }
 
-static int hw_atl_b0_hw_offload_set(struct aq_hw_s *self,
-				    struct aq_nic_cfg_s *aq_nic_cfg)
+int hw_atl_b0_hw_offload_set(struct aq_hw_s *self,
+			     struct aq_nic_cfg_s *aq_nic_cfg)
 {
+	u64 rxcsum = !!(aq_nic_cfg->features & NETIF_F_RXCSUM);
 	unsigned int i;
 
 	/* TX checksums offloads*/
-	tpo_ipv4header_crc_offload_en_set(self, 1);
-	tpo_tcp_udp_crc_offload_en_set(self, 1);
+	hw_atl_tpo_ipv4header_crc_offload_en_set(self, 1);
+	hw_atl_tpo_tcp_udp_crc_offload_en_set(self, 1);
 
 	/* RX checksums offloads*/
-	rpo_ipv4header_crc_offload_en_set(self, 1);
-	rpo_tcp_udp_crc_offload_en_set(self, 1);
+	hw_atl_rpo_ipv4header_crc_offload_en_set(self, rxcsum);
+	hw_atl_rpo_tcp_udp_crc_offload_en_set(self, rxcsum);
 
 	/* LSO offloads*/
-	tdm_large_send_offload_en_set(self, 0xFFFFFFFFU);
+	hw_atl_tdm_large_send_offload_en_set(self, 0xFFFFFFFFU);
+
+	/* Outer VLAN tag offload */
+	hw_atl_rpo_outer_vlan_tag_mode_set(self, 1U);
 
-/* LRO offloads */
+	/* LRO offloads */
 	{
 		unsigned int val = (8U < HW_ATL_B0_LRO_RXD_MAX) ? 0x3U :
 			((4U < HW_ATL_B0_LRO_RXD_MAX) ? 0x2U :
 			((2U < HW_ATL_B0_LRO_RXD_MAX) ? 0x1U : 0x0));
 
 		for (i = 0; i < HW_ATL_B0_RINGS_MAX; i++)
-			rpo_lro_max_num_of_descriptors_set(self, val, i);
+			hw_atl_rpo_lro_max_num_of_descriptors_set(self, val, i);
 
-		rpo_lro_time_base_divider_set(self, 0x61AU);
-		rpo_lro_inactive_interval_set(self, 0);
-		rpo_lro_max_coalescing_interval_set(self, 2);
+		hw_atl_rpo_lro_time_base_divider_set(self, 0x61AU);
+		hw_atl_rpo_lro_inactive_interval_set(self, 0);
+		/* the LRO timebase divider is 5 uS (0x61a),
+		 * which is multiplied by 50(0x32)
+		 * to get a maximum coalescing interval of 250 uS,
+		 * which is the default value
+		 */
+		hw_atl_rpo_lro_max_coalescing_interval_set(self, 50);
 
-		rpo_lro_qsessions_lim_set(self, 1U);
+		hw_atl_rpo_lro_qsessions_lim_set(self, 1U);
 
-		rpo_lro_total_desc_lim_set(self, 2U);
+		hw_atl_rpo_lro_total_desc_lim_set(self, 2U);
 
-		rpo_lro_patch_optimization_en_set(self, 0U);
+		hw_atl_rpo_lro_patch_optimization_en_set(self, 1U);
 
-		rpo_lro_min_pay_of_first_pkt_set(self, 10U);
+		hw_atl_rpo_lro_min_pay_of_first_pkt_set(self, 10U);
 
-		rpo_lro_pkt_lim_set(self, 1U);
+		hw_atl_rpo_lro_pkt_lim_set(self, 1U);
 
-		rpo_lro_en_set(self, aq_nic_cfg->is_lro ? 0xFFFFFFFFU : 0U);
+		hw_atl_rpo_lro_en_set(self,
+				      aq_nic_cfg->is_lro ? 0xFFFFFFFFU : 0U);
+		hw_atl_itr_rsc_en_set(self,
+				      aq_nic_cfg->is_lro ? 0xFFFFFFFFU : 0U);
+
+		hw_atl_itr_rsc_delay_set(self, 1U);
 	}
+
+	return aq_hw_err_from_flags(self);
+}
+
+static int hw_atl_b0_hw_init_tx_tc_rate_limit(struct aq_hw_s *self)
+{
+	static const u32 max_weight = BIT(HW_ATL_TPS_DATA_TCTWEIGHT_WIDTH) - 1;
+	/* Scale factor is based on the number of bits in fractional portion */
+	static const u32 scale = BIT(HW_ATL_TPS_DESC_RATE_Y_WIDTH);
+	static const u32 frac_msk = HW_ATL_TPS_DESC_RATE_Y_MSK >>
+				    HW_ATL_TPS_DESC_RATE_Y_SHIFT;
+	const u32 link_speed = self->aq_link_status.mbps;
+	struct aq_nic_cfg_s *nic_cfg = self->aq_nic_cfg;
+	unsigned long num_min_rated_tcs = 0;
+	u32 tc_weight[AQ_CFG_TCS_MAX];
+	u32 fixed_max_credit;
+	u8 min_rate_msk = 0;
+	u32 sum_weight = 0;
+	int tc;
+
+	/* By default max_credit is based upon MTU (in unit of 64b) */
+	fixed_max_credit = nic_cfg->aq_hw_caps->mtu / 64;
+
+	if (link_speed) {
+		min_rate_msk = nic_cfg->tc_min_rate_msk &
+			       (BIT(nic_cfg->tcs) - 1);
+		num_min_rated_tcs = hweight8(min_rate_msk);
+	}
+
+	/* First, calculate weights where min_rate is specified */
+	if (num_min_rated_tcs) {
+		for (tc = 0; tc != nic_cfg->tcs; tc++) {
+			if (!nic_cfg->tc_min_rate[tc]) {
+				tc_weight[tc] = 0;
+				continue;
+			}
+
+			tc_weight[tc] = (-1L + link_speed +
+					 nic_cfg->tc_min_rate[tc] *
+					 max_weight) /
+					link_speed;
+			tc_weight[tc] = min(tc_weight[tc], max_weight);
+			sum_weight += tc_weight[tc];
+		}
+	}
+
+	/* WSP, if min_rate is set for at least one TC.
+	 * RR otherwise.
+	 *
+	 * NB! MAC FW sets arb mode itself if PTP is enabled. We shouldn't
+	 * overwrite it here in that case.
+	 */
+	if (!nic_cfg->is_ptp)
+		hw_atl_tps_tx_pkt_shed_data_arb_mode_set(self, min_rate_msk ? 1U : 0U);
+
+	/* Data TC Arbiter takes precedence over Descriptor TC Arbiter,
+	 * leave Descriptor TC Arbiter as RR.
+	 */
+	hw_atl_tps_tx_pkt_shed_desc_tc_arb_mode_set(self, 0U);
+
+	hw_atl_tps_tx_desc_rate_mode_set(self, nic_cfg->is_qos ? 1U : 0U);
+
+	for (tc = 0; tc != nic_cfg->tcs; tc++) {
+		const u32 en = (nic_cfg->tc_max_rate[tc] != 0) ? 1U : 0U;
+		const u32 desc = AQ_NIC_CFG_TCVEC2RING(nic_cfg, tc, 0);
+		u32 weight, max_credit;
+
+		hw_atl_tps_tx_pkt_shed_desc_tc_max_credit_set(self, tc,
+							      fixed_max_credit);
+		hw_atl_tps_tx_pkt_shed_desc_tc_weight_set(self, tc, 0x1E);
+
+		if (num_min_rated_tcs) {
+			weight = tc_weight[tc];
+
+			if (!weight && sum_weight < max_weight)
+				weight = (max_weight - sum_weight) /
+					 (nic_cfg->tcs - num_min_rated_tcs);
+			else if (!weight)
+				weight = 0x64;
+
+			max_credit = max(8 * weight, fixed_max_credit);
+		} else {
+			weight = 0x64;
+			max_credit = 0xFFF;
+		}
+
+		hw_atl_tps_tx_pkt_shed_tc_data_weight_set(self, tc, weight);
+		hw_atl_tps_tx_pkt_shed_tc_data_max_credit_set(self, tc,
+							      max_credit);
+
+		hw_atl_tps_tx_desc_rate_en_set(self, desc, en);
+
+		if (en) {
+			/* Nominal rate is always 10G */
+			const u32 rate = 10000U * scale /
+					 nic_cfg->tc_max_rate[tc];
+			const u32 rate_int = rate >>
+					     HW_ATL_TPS_DESC_RATE_Y_WIDTH;
+			const u32 rate_frac = rate & frac_msk;
+
+			hw_atl_tps_tx_desc_rate_x_set(self, desc, rate_int);
+			hw_atl_tps_tx_desc_rate_y_set(self, desc, rate_frac);
+		} else {
+			/* A value of 1 indicates the queue is not
+			 * rate controlled.
+			 */
+			hw_atl_tps_tx_desc_rate_x_set(self, desc, 1U);
+			hw_atl_tps_tx_desc_rate_y_set(self, desc, 0U);
+		}
+	}
+	for (tc = nic_cfg->tcs; tc != AQ_CFG_TCS_MAX; tc++) {
+		const u32 desc = AQ_NIC_CFG_TCVEC2RING(nic_cfg, tc, 0);
+
+		hw_atl_tps_tx_desc_rate_en_set(self, desc, 0U);
+		hw_atl_tps_tx_desc_rate_x_set(self, desc, 1U);
+		hw_atl_tps_tx_desc_rate_y_set(self, desc, 0U);
+	}
+
 	return aq_hw_err_from_flags(self);
 }
 
 static int hw_atl_b0_hw_init_tx_path(struct aq_hw_s *self)
 {
-	thm_lso_tcp_flag_of_first_pkt_set(self, 0x0FF6U);
-	thm_lso_tcp_flag_of_middle_pkt_set(self, 0x0FF6U);
-	thm_lso_tcp_flag_of_last_pkt_set(self, 0x0F7FU);
+	struct aq_nic_cfg_s *nic_cfg = self->aq_nic_cfg;
+
+	/* Tx TC/Queue number config */
+	hw_atl_tpb_tps_tx_tc_mode_set(self, nic_cfg->tc_mode);
+
+	hw_atl_thm_lso_tcp_flag_of_first_pkt_set(self, 0x0FF6U);
+	hw_atl_thm_lso_tcp_flag_of_middle_pkt_set(self, 0x0FF6U);
+	hw_atl_thm_lso_tcp_flag_of_last_pkt_set(self, 0x0F7FU);
 
 	/* Tx interrupts */
-	tdm_tx_desc_wr_wb_irq_en_set(self, 1U);
+	hw_atl_tdm_tx_desc_wr_wb_irq_en_set(self, 1U);
 
 	/* misc */
-	aq_hw_write_reg(self, 0x00007040U, IS_CHIP_FEATURE(TPO2) ?
+	aq_hw_write_reg(self, 0x00007040U, ATL_HW_IS_CHIP_FEATURE(self, TPO2) ?
 			0x00010000U : 0x00000000U);
-	tdm_tx_dca_en_set(self, 0U);
-	tdm_tx_dca_mode_set(self, 0U);
+	hw_atl_tdm_tx_dca_en_set(self, 0U);
+	hw_atl_tdm_tx_dca_mode_set(self, 0U);
 
-	tpb_tx_path_scp_ins_en_set(self, 1U);
+	hw_atl_tpb_tx_path_scp_ins_en_set(self, 1U);
 
 	return aq_hw_err_from_flags(self);
 }
 
+void hw_atl_b0_hw_init_rx_rss_ctrl1(struct aq_hw_s *self)
+{
+	struct aq_nic_cfg_s *cfg = self->aq_nic_cfg;
+	u32 rss_ctrl1 = HW_ATL_RSS_DISABLED;
+
+	if (cfg->is_rss)
+		rss_ctrl1 = (cfg->tc_mode == AQ_TC_MODE_8TCS) ?
+			    HW_ATL_RSS_ENABLED_8TCS_2INDEX_BITS :
+			    HW_ATL_RSS_ENABLED_4TCS_3INDEX_BITS;
+
+	hw_atl_reg_rx_flr_rss_control1set(self, rss_ctrl1);
+}
+
 static int hw_atl_b0_hw_init_rx_path(struct aq_hw_s *self)
 {
 	struct aq_nic_cfg_s *cfg = self->aq_nic_cfg;
 	int i;
 
 	/* Rx TC/RSS number config */
-	rpb_rpf_rx_traf_class_mode_set(self, 1U);
+	hw_atl_rpb_rpf_rx_traf_class_mode_set(self, cfg->tc_mode);
 
 	/* Rx flow control */
-	rpb_rx_flow_ctl_mode_set(self, 1U);
+	hw_atl_rpb_rx_flow_ctl_mode_set(self, 1U);
 
 	/* RSS Ring selection */
-	reg_rx_flr_rss_control1set(self, cfg->is_rss ?
-					0xB3333333U : 0x00000000U);
+	hw_atl_b0_hw_init_rx_rss_ctrl1(self);
 
 	/* Multicast filters */
 	for (i = HW_ATL_B0_MAC_MAX; i--;) {
-		rpfl2_uc_flr_en_set(self, (i == 0U) ? 1U : 0U, i);
-		rpfl2unicast_flr_act_set(self, 1U, i);
+		hw_atl_rpfl2_uc_flr_en_set(self, (i == 0U) ? 1U : 0U, i);
+		hw_atl_rpfl2unicast_flr_act_set(self, 1U, i);
 	}
 
-	reg_rx_flr_mcst_flr_msk_set(self, 0x00000000U);
-	reg_rx_flr_mcst_flr_set(self, 0x00010FFFU, 0U);
+	hw_atl_reg_rx_flr_mcst_flr_msk_set(self, 0x00000000U);
+	hw_atl_reg_rx_flr_mcst_flr_set(self, 0x00010FFFU, 0U);
 
 	/* Vlan filters */
-	rpf_vlan_outer_etht_set(self, 0x88A8U);
-	rpf_vlan_inner_etht_set(self, 0x8100U);
+	hw_atl_rpf_vlan_outer_etht_set(self, 0x88A8U);
+	hw_atl_rpf_vlan_inner_etht_set(self, 0x8100U);
 
-	if (cfg->vlan_id) {
-		rpf_vlan_flr_act_set(self, 1U, 0U);
-		rpf_vlan_id_flr_set(self, 0U, 0U);
-		rpf_vlan_flr_en_set(self, 0U, 0U);
+	hw_atl_rpf_vlan_prom_mode_en_set(self, 1);
 
-		rpf_vlan_accept_untagged_packets_set(self, 1U);
-		rpf_vlan_untagged_act_set(self, 1U);
-
-		rpf_vlan_flr_act_set(self, 1U, 1U);
-		rpf_vlan_id_flr_set(self, cfg->vlan_id, 0U);
-		rpf_vlan_flr_en_set(self, 1U, 1U);
-	} else {
-		rpf_vlan_prom_mode_en_set(self, 1);
-	}
+	// Always accept untagged packets
+	hw_atl_rpf_vlan_accept_untagged_packets_set(self, 1U);
+	hw_atl_rpf_vlan_untagged_act_set(self, 1U);
 
 	/* Rx Interrupts */
-	rdm_rx_desc_wr_wb_irq_en_set(self, 1U);
+	hw_atl_rdm_rx_desc_wr_wb_irq_en_set(self, 1U);
 
 	/* misc */
-	aq_hw_write_reg(self, 0x00005040U,
-			IS_CHIP_FEATURE(RPF2) ? 0x000F0000U : 0x00000000U);
+	aq_hw_write_reg(self, 0x00005040U, ATL_HW_IS_CHIP_FEATURE(self, RPF2) ?
+			0x000F0000U : 0x00000000U);
 
-	rpfl2broadcast_flr_act_set(self, 1U);
-	rpfl2broadcast_count_threshold_set(self, 0xFFFFU & (~0U / 256U));
+	hw_atl_rpfl2broadcast_flr_act_set(self, 1U);
+	hw_atl_rpfl2broadcast_count_threshold_set(self, 0xFFFFU & (~0U / 256U));
 
-	rdm_rx_dca_en_set(self, 0U);
-	rdm_rx_dca_mode_set(self, 0U);
+	hw_atl_rdm_rx_dca_en_set(self, 0U);
+	hw_atl_rdm_rx_dca_mode_set(self, 0U);
 
 	return aq_hw_err_from_flags(self);
 }
 
-static int hw_atl_b0_hw_mac_addr_set(struct aq_hw_s *self, u8 *mac_addr)
+int hw_atl_b0_hw_mac_addr_set(struct aq_hw_s *self, const u8 *mac_addr)
 {
-	int err = 0;
 	unsigned int h = 0U;
 	unsigned int l = 0U;
+	int err = 0;
 
 	if (!mac_addr) {
 		err = -EINVAL;
@@ -334,10 +547,10 @@ static int hw_atl_b0_hw_mac_addr_set(struct aq_hw_s *self, u8 *mac_addr)
 	l = (mac_addr[2] << 24) | (mac_addr[3] << 16) |
 		(mac_addr[4] << 8) | mac_addr[5];
 
-	rpfl2_uc_flr_en_set(self, 0U, HW_ATL_B0_MAC);
-	rpfl2unicast_dest_addresslsw_set(self, l, HW_ATL_B0_MAC);
-	rpfl2unicast_dest_addressmsw_set(self, h, HW_ATL_B0_MAC);
-	rpfl2_uc_flr_en_set(self, 1U, HW_ATL_B0_MAC);
+	hw_atl_rpfl2_uc_flr_en_set(self, 0U, HW_ATL_B0_MAC);
+	hw_atl_rpfl2unicast_dest_addresslsw_set(self, l, HW_ATL_B0_MAC);
+	hw_atl_rpfl2unicast_dest_addressmsw_set(self, h, HW_ATL_B0_MAC);
+	hw_atl_rpfl2_uc_flr_en_set(self, 1U, HW_ATL_B0_MAC);
 
 	err = aq_hw_err_from_flags(self);
 
@@ -345,51 +558,69 @@ err_exit:
 	return err;
 }
 
-static int hw_atl_b0_hw_init(struct aq_hw_s *self,
-			     struct aq_nic_cfg_s *aq_nic_cfg,
-			     u8 *mac_addr)
+static int hw_atl_b0_hw_init(struct aq_hw_s *self, const u8 *mac_addr)
 {
 	static u32 aq_hw_atl_igcr_table_[4][2] = {
-		{ 0x20000000U, 0x20000000U }, /* AQ_IRQ_INVALID */
-		{ 0x20000080U, 0x20000080U }, /* AQ_IRQ_LEGACY */
-		{ 0x20000021U, 0x20000025U }, /* AQ_IRQ_MSI */
-		{ 0x20000022U, 0x20000026U }  /* AQ_IRQ_MSIX */
+		[AQ_HW_IRQ_INVALID] = { 0x20000000U, 0x20000000U },
+		[AQ_HW_IRQ_INTX]    = { 0x20000080U, 0x20000080U },
+		[AQ_HW_IRQ_MSI]     = { 0x20000021U, 0x20000025U },
+		[AQ_HW_IRQ_MSIX]    = { 0x20000022U, 0x20000026U },
 	};
-
+	struct aq_nic_cfg_s *aq_nic_cfg = self->aq_nic_cfg;
 	int err = 0;
+	u32 val;
 
-	self->aq_nic_cfg = aq_nic_cfg;
-
-	hw_atl_utils_hw_chip_features_init(self,
-					   &PHAL_ATLANTIC_B0->chip_features);
 
 	hw_atl_b0_hw_init_tx_path(self);
 	hw_atl_b0_hw_init_rx_path(self);
 
 	hw_atl_b0_hw_mac_addr_set(self, mac_addr);
 
-	hw_atl_utils_mpi_set(self, MPI_INIT, aq_nic_cfg->link_speed_msk);
+	self->aq_fw_ops->set_link_speed(self, aq_nic_cfg->link_speed_msk);
+	self->aq_fw_ops->set_state(self, MPI_INIT);
 
 	hw_atl_b0_hw_qos_set(self);
 	hw_atl_b0_hw_rss_set(self, &aq_nic_cfg->aq_rss);
 	hw_atl_b0_hw_rss_hash_set(self, &aq_nic_cfg->aq_rss);
 
+	/* Force limit MRRS on RDM/TDM to 2K */
+	val = aq_hw_read_reg(self, HW_ATL_PCI_REG_CONTROL6_ADR);
+	aq_hw_write_reg(self, HW_ATL_PCI_REG_CONTROL6_ADR,
+			(val & ~0x707) | 0x404);
+
+	/* TX DMA total request limit. B0 hardware is not capable to
+	 * handle more than (8K-MRRS) incoming DMA data.
+	 * Value 24 in 256byte units
+	 */
+	aq_hw_write_reg(self, HW_ATL_TX_DMA_TOTAL_REQ_LIMIT_ADR, 24);
+
+	/* Reset link status and read out initial hardware counters */
+	self->aq_link_status.mbps = 0;
+	self->aq_fw_ops->update_stats(self);
+
 	err = aq_hw_err_from_flags(self);
 	if (err < 0)
 		goto err_exit;
 
 	/* Interrupts */
-	reg_irq_glb_ctl_set(self,
-			    aq_hw_atl_igcr_table_[aq_nic_cfg->irq_type]
+	hw_atl_reg_irq_glb_ctl_set(self,
+				   aq_hw_atl_igcr_table_[aq_nic_cfg->irq_type]
 						 [(aq_nic_cfg->vecs > 1U) ?
 						 1 : 0]);
 
-	itr_irq_auto_masklsw_set(self, aq_nic_cfg->aq_hw_caps->irq_mask);
+	hw_atl_itr_irq_auto_masklsw_set(self, aq_nic_cfg->aq_hw_caps->irq_mask);
 
 	/* Interrupts */
-	reg_gen_irq_map_set(self,
-			    ((HW_ATL_B0_ERR_INT << 0x18) | (1U << 0x1F)) |
-			    ((HW_ATL_B0_ERR_INT << 0x10) | (1U << 0x17)), 0U);
+	hw_atl_reg_gen_irq_map_set(self,
+				   ((HW_ATL_B0_ERR_INT << 0x18) |
+				    (1U << 0x1F)) |
+				   ((HW_ATL_B0_ERR_INT << 0x10) |
+				    (1U << 0x17)), 0U);
+
+	/* Enable link interrupt */
+	if (aq_nic_cfg->link_irq_vec)
+		hw_atl_reg_gen_irq_map_set(self, BIT(7) |
+					   aq_nic_cfg->link_irq_vec, 3U);
 
 	hw_atl_b0_hw_offload_set(self, aq_nic_cfg);
 
@@ -397,43 +628,45 @@ err_exit:
 	return err;
 }
 
-static int hw_atl_b0_hw_ring_tx_start(struct aq_hw_s *self,
-				      struct aq_ring_s *ring)
+int hw_atl_b0_hw_ring_tx_start(struct aq_hw_s *self, struct aq_ring_s *ring)
 {
-	tdm_tx_desc_en_set(self, 1, ring->idx);
+	hw_atl_tdm_tx_desc_en_set(self, 1, ring->idx);
+
 	return aq_hw_err_from_flags(self);
 }
 
-static int hw_atl_b0_hw_ring_rx_start(struct aq_hw_s *self,
-				      struct aq_ring_s *ring)
+int hw_atl_b0_hw_ring_rx_start(struct aq_hw_s *self, struct aq_ring_s *ring)
 {
-	rdm_rx_desc_en_set(self, 1, ring->idx);
+	hw_atl_rdm_rx_desc_en_set(self, 1, ring->idx);
+
 	return aq_hw_err_from_flags(self);
 }
 
-static int hw_atl_b0_hw_start(struct aq_hw_s *self)
+int hw_atl_b0_hw_start(struct aq_hw_s *self)
 {
-	tpb_tx_buff_en_set(self, 1);
-	rpb_rx_buff_en_set(self, 1);
+	hw_atl_tpb_tx_buff_en_set(self, 1);
+	hw_atl_rpb_rx_buff_en_set(self, 1);
+
 	return aq_hw_err_from_flags(self);
 }
 
 static int hw_atl_b0_hw_tx_ring_tail_update(struct aq_hw_s *self,
 					    struct aq_ring_s *ring)
 {
-	reg_tx_dma_desc_tail_ptr_set(self, ring->sw_tail, ring->idx);
+	hw_atl_reg_tx_dma_desc_tail_ptr_set(self, ring->sw_tail, ring->idx);
+
 	return 0;
 }
 
-static int hw_atl_b0_hw_ring_tx_xmit(struct aq_hw_s *self,
-				     struct aq_ring_s *ring,
-				     unsigned int frags)
+int hw_atl_b0_hw_ring_tx_xmit(struct aq_hw_s *self, struct aq_ring_s *ring,
+			      unsigned int frags)
 {
 	struct aq_ring_buff_s *buff = NULL;
 	struct hw_atl_txd_s *txd = NULL;
 	unsigned int buff_pa_len = 0U;
-	unsigned int pkt_len = 0U;
 	unsigned int frag_count = 0U;
+	unsigned int pkt_len = 0U;
+	bool is_vlan = false;
 	bool is_gso = false;
 
 	buff = &ring->buff_ring[ring->sw_tail];
@@ -448,36 +681,45 @@ static int hw_atl_b0_hw_ring_tx_xmit(struct aq_hw_s *self,
 
 		buff = &ring->buff_ring[ring->sw_tail];
 
-		if (buff->is_txc) {
+		if (buff->is_gso_tcp || buff->is_gso_udp) {
+			if (buff->is_gso_tcp)
+				txd->ctl |= HW_ATL_B0_TXD_CTL_CMD_TCP;
+			txd->ctl |= HW_ATL_B0_TXD_CTL_DESC_TYPE_TXC;
 			txd->ctl |= (buff->len_l3 << 31) |
-				(buff->len_l2 << 24) |
-				HW_ATL_B0_TXD_CTL_CMD_TCP |
-				HW_ATL_B0_TXD_CTL_DESC_TYPE_TXC;
-			txd->ctl2 |= (buff->mss << 16) |
-				(buff->len_l4 << 8) |
-				(buff->len_l3 >> 1);
+				    (buff->len_l2 << 24);
+			txd->ctl2 |= (buff->mss << 16);
+			is_gso = true;
 
 			pkt_len -= (buff->len_l4 +
 				    buff->len_l3 +
 				    buff->len_l2);
-			is_gso = true;
-
 			if (buff->is_ipv6)
 				txd->ctl |= HW_ATL_B0_TXD_CTL_CMD_IPV6;
-		} else {
+			txd->ctl2 |= (buff->len_l4 << 8) |
+				     (buff->len_l3 >> 1);
+		}
+		if (buff->is_vlan) {
+			txd->ctl |= HW_ATL_B0_TXD_CTL_DESC_TYPE_TXC;
+			txd->ctl |= buff->vlan_tx_tag << 4;
+			is_vlan = true;
+		}
+		if (!buff->is_gso_tcp && !buff->is_gso_udp && !buff->is_vlan) {
 			buff_pa_len = buff->len;
 
 			txd->buf_addr = buff->pa;
 			txd->ctl |= (HW_ATL_B0_TXD_CTL_BLEN &
 						((u32)buff_pa_len << 4));
 			txd->ctl |= HW_ATL_B0_TXD_CTL_DESC_TYPE_TXD;
+
 			/* PAY_LEN */
 			txd->ctl2 |= HW_ATL_B0_TXD_CTL2_LEN & (pkt_len << 14);
 
-			if (is_gso) {
-				txd->ctl |= HW_ATL_B0_TXD_CTL_CMD_LSO;
+			if (is_gso || is_vlan) {
+				/* enable tx context */
 				txd->ctl2 |= HW_ATL_B0_TXD_CTL2_CTX_EN;
 			}
+			if (is_gso)
+				txd->ctl |= HW_ATL_B0_TXD_CTL_CMD_LSO;
 
 			/* Tx checksum offloads */
 			if (buff->is_ip_cso)
@@ -486,94 +728,97 @@ static int hw_atl_b0_hw_ring_tx_xmit(struct aq_hw_s *self,
 			if (buff->is_udp_cso || buff->is_tcp_cso)
 				txd->ctl |= HW_ATL_B0_TXD_CTL_CMD_TUCSO;
 
+			if (is_vlan)
+				txd->ctl |= HW_ATL_B0_TXD_CTL_CMD_VLAN;
+
 			if (unlikely(buff->is_eop)) {
 				txd->ctl |= HW_ATL_B0_TXD_CTL_EOP;
 				txd->ctl |= HW_ATL_B0_TXD_CTL_CMD_WB;
 				is_gso = false;
+				is_vlan = false;
 			}
 		}
-
 		ring->sw_tail = aq_ring_next_dx(ring, ring->sw_tail);
 	}
 
 	hw_atl_b0_hw_tx_ring_tail_update(self, ring);
+
 	return aq_hw_err_from_flags(self);
 }
 
-static int hw_atl_b0_hw_ring_rx_init(struct aq_hw_s *self,
-				     struct aq_ring_s *aq_ring,
-				     struct aq_ring_param_s *aq_ring_param)
+int hw_atl_b0_hw_ring_rx_init(struct aq_hw_s *self, struct aq_ring_s *aq_ring,
+			      struct aq_ring_param_s *aq_ring_param)
 {
-	u32 dma_desc_addr_lsw = (u32)aq_ring->dx_ring_pa;
 	u32 dma_desc_addr_msw = (u32)(((u64)aq_ring->dx_ring_pa) >> 32);
+	u32 vlan_rx_stripping = self->aq_nic_cfg->is_vlan_rx_strip;
+	u32 dma_desc_addr_lsw = (u32)aq_ring->dx_ring_pa;
 
-	rdm_rx_desc_en_set(self, false, aq_ring->idx);
+	hw_atl_rdm_rx_desc_en_set(self, false, aq_ring->idx);
 
-	rdm_rx_desc_head_splitting_set(self, 0U, aq_ring->idx);
+	hw_atl_rdm_rx_desc_head_splitting_set(self, 0U, aq_ring->idx);
 
-	reg_rx_dma_desc_base_addresslswset(self, dma_desc_addr_lsw,
-					   aq_ring->idx);
+	hw_atl_reg_rx_dma_desc_base_addresslswset(self, dma_desc_addr_lsw,
+						  aq_ring->idx);
 
-	reg_rx_dma_desc_base_addressmswset(self,
-					   dma_desc_addr_msw, aq_ring->idx);
+	hw_atl_reg_rx_dma_desc_base_addressmswset(self,
+						  dma_desc_addr_msw, aq_ring->idx);
 
-	rdm_rx_desc_len_set(self, aq_ring->size / 8U, aq_ring->idx);
+	hw_atl_rdm_rx_desc_len_set(self, aq_ring->size / 8U, aq_ring->idx);
 
-	rdm_rx_desc_data_buff_size_set(self,
-				       AQ_CFG_RX_FRAME_MAX / 1024U,
+	hw_atl_rdm_rx_desc_data_buff_size_set(self,
+					      aq_ring->frame_max / 1024U,
 				       aq_ring->idx);
 
-	rdm_rx_desc_head_buff_size_set(self, 0U, aq_ring->idx);
-	rdm_rx_desc_head_splitting_set(self, 0U, aq_ring->idx);
-	rpo_rx_desc_vlan_stripping_set(self, 0U, aq_ring->idx);
+	hw_atl_rdm_rx_desc_head_buff_size_set(self, 0U, aq_ring->idx);
+	hw_atl_rdm_rx_desc_head_splitting_set(self, 0U, aq_ring->idx);
+	hw_atl_rpo_rx_desc_vlan_stripping_set(self, !!vlan_rx_stripping,
+					      aq_ring->idx);
 
 	/* Rx ring set mode */
 
 	/* Mapping interrupt vector */
-	itr_irq_map_rx_set(self, aq_ring_param->vec_idx, aq_ring->idx);
-	itr_irq_map_en_rx_set(self, true, aq_ring->idx);
+	hw_atl_itr_irq_map_rx_set(self, aq_ring_param->vec_idx, aq_ring->idx);
+	hw_atl_itr_irq_map_en_rx_set(self, true, aq_ring->idx);
 
-	rdm_cpu_id_set(self, aq_ring_param->cpu, aq_ring->idx);
-	rdm_rx_desc_dca_en_set(self, 0U, aq_ring->idx);
-	rdm_rx_head_dca_en_set(self, 0U, aq_ring->idx);
-	rdm_rx_pld_dca_en_set(self, 0U, aq_ring->idx);
+	hw_atl_rdm_cpu_id_set(self, aq_ring_param->cpu, aq_ring->idx);
+	hw_atl_rdm_rx_desc_dca_en_set(self, 0U, aq_ring->idx);
+	hw_atl_rdm_rx_head_dca_en_set(self, 0U, aq_ring->idx);
+	hw_atl_rdm_rx_pld_dca_en_set(self, 0U, aq_ring->idx);
 
 	return aq_hw_err_from_flags(self);
 }
 
-static int hw_atl_b0_hw_ring_tx_init(struct aq_hw_s *self,
-				     struct aq_ring_s *aq_ring,
-				     struct aq_ring_param_s *aq_ring_param)
+int hw_atl_b0_hw_ring_tx_init(struct aq_hw_s *self, struct aq_ring_s *aq_ring,
+			      struct aq_ring_param_s *aq_ring_param)
 {
-	u32 dma_desc_lsw_addr = (u32)aq_ring->dx_ring_pa;
 	u32 dma_desc_msw_addr = (u32)(((u64)aq_ring->dx_ring_pa) >> 32);
+	u32 dma_desc_lsw_addr = (u32)aq_ring->dx_ring_pa;
 
-	reg_tx_dma_desc_base_addresslswset(self, dma_desc_lsw_addr,
-					   aq_ring->idx);
+	hw_atl_reg_tx_dma_desc_base_addresslswset(self, dma_desc_lsw_addr,
+						  aq_ring->idx);
 
-	reg_tx_dma_desc_base_addressmswset(self, dma_desc_msw_addr,
-					   aq_ring->idx);
+	hw_atl_reg_tx_dma_desc_base_addressmswset(self, dma_desc_msw_addr,
+						  aq_ring->idx);
 
-	tdm_tx_desc_len_set(self, aq_ring->size / 8U, aq_ring->idx);
+	hw_atl_tdm_tx_desc_len_set(self, aq_ring->size / 8U, aq_ring->idx);
 
 	hw_atl_b0_hw_tx_ring_tail_update(self, aq_ring);
 
 	/* Set Tx threshold */
-	tdm_tx_desc_wr_wb_threshold_set(self, 0U, aq_ring->idx);
+	hw_atl_tdm_tx_desc_wr_wb_threshold_set(self, 0U, aq_ring->idx);
 
 	/* Mapping interrupt vector */
-	itr_irq_map_tx_set(self, aq_ring_param->vec_idx, aq_ring->idx);
-	itr_irq_map_en_tx_set(self, true, aq_ring->idx);
+	hw_atl_itr_irq_map_tx_set(self, aq_ring_param->vec_idx, aq_ring->idx);
+	hw_atl_itr_irq_map_en_tx_set(self, true, aq_ring->idx);
 
-	tdm_cpu_id_set(self, aq_ring_param->cpu, aq_ring->idx);
-	tdm_tx_desc_dca_en_set(self, 0U, aq_ring->idx);
+	hw_atl_tdm_cpu_id_set(self, aq_ring_param->cpu, aq_ring->idx);
+	hw_atl_tdm_tx_desc_dca_en_set(self, 0U, aq_ring->idx);
 
 	return aq_hw_err_from_flags(self);
 }
 
-static int hw_atl_b0_hw_ring_rx_fill(struct aq_hw_s *self,
-				     struct aq_ring_s *ring,
-				     unsigned int sw_tail_old)
+int hw_atl_b0_hw_ring_rx_fill(struct aq_hw_s *self, struct aq_ring_s *ring,
+			      unsigned int sw_tail_old)
 {
 	for (; sw_tail_old != ring->sw_tail;
 		sw_tail_old = aq_ring_next_dx(ring, sw_tail_old)) {
@@ -587,21 +832,70 @@ static int hw_atl_b0_hw_ring_rx_fill(struct aq_hw_s *self,
 		rxd->hdr_addr = 0U;
 	}
 
-	reg_rx_dma_desc_tail_ptr_set(self, sw_tail_old, ring->idx);
+	hw_atl_reg_rx_dma_desc_tail_ptr_set(self, sw_tail_old, ring->idx);
 
 	return aq_hw_err_from_flags(self);
 }
 
-static int hw_atl_b0_hw_ring_tx_head_update(struct aq_hw_s *self,
-					    struct aq_ring_s *ring)
+static int hw_atl_b0_hw_ring_hwts_rx_fill(struct aq_hw_s *self,
+					  struct aq_ring_s *ring)
+{
+	unsigned int i;
+
+	for (i = aq_ring_avail_dx(ring); i--;
+			ring->sw_tail = aq_ring_next_dx(ring, ring->sw_tail)) {
+		struct hw_atl_rxd_s *rxd =
+			(struct hw_atl_rxd_s *)
+			&ring->dx_ring[ring->sw_tail * HW_ATL_B0_RXD_SIZE];
+
+		rxd->buf_addr = ring->dx_ring_pa + ring->size * ring->dx_size;
+		rxd->hdr_addr = 0U;
+	}
+	/* Make sure descriptors are updated before bump tail*/
+	wmb();
+
+	hw_atl_reg_rx_dma_desc_tail_ptr_set(self, ring->sw_tail, ring->idx);
+
+	return aq_hw_err_from_flags(self);
+}
+
+static int hw_atl_b0_hw_ring_hwts_rx_receive(struct aq_hw_s *self,
+					     struct aq_ring_s *ring)
+{
+	while (ring->hw_head != ring->sw_tail) {
+		struct hw_atl_rxd_hwts_wb_s *hwts_wb =
+			(struct hw_atl_rxd_hwts_wb_s *)
+			(ring->dx_ring + (ring->hw_head * HW_ATL_B0_RXD_SIZE));
+
+		/* RxD is not done */
+		if (!(hwts_wb->sec_lw0 & 0x1U))
+			break;
+
+		ring->hw_head = aq_ring_next_dx(ring, ring->hw_head);
+	}
+
+	return aq_hw_err_from_flags(self);
+}
+
+int hw_atl_b0_hw_ring_tx_head_update(struct aq_hw_s *self,
+				     struct aq_ring_s *ring)
 {
+	unsigned int hw_head_;
 	int err = 0;
-	unsigned int hw_head_ = tdm_tx_desc_head_ptr_get(self, ring->idx);
 
-	if (aq_utils_obj_test(&self->header.flags, AQ_HW_FLAG_ERR_UNPLUG)) {
+	hw_head_ = hw_atl_tdm_tx_desc_head_ptr_get(self, ring->idx);
+
+	if (aq_utils_obj_test(&self->flags, AQ_HW_FLAG_ERR_UNPLUG)) {
 		err = -ENXIO;
 		goto err_exit;
 	}
+
+	/* Validate that the new hw_head_ is reasonable. */
+	if (hw_head_ >= ring->size) {
+		err = -ENXIO;
+		goto err_exit;
+	}
+
 	ring->hw_head = hw_head_;
 	err = aq_hw_err_from_flags(self);
 
@@ -609,20 +903,17 @@ err_exit:
 	return err;
 }
 
-static int hw_atl_b0_hw_ring_rx_receive(struct aq_hw_s *self,
-					struct aq_ring_s *ring)
+int hw_atl_b0_hw_ring_rx_receive(struct aq_hw_s *self, struct aq_ring_s *ring)
 {
-	struct device *ndev = aq_nic_get_dev(ring->aq_nic);
-
 	for (; ring->hw_head != ring->sw_tail;
 		ring->hw_head = aq_ring_next_dx(ring, ring->hw_head)) {
 		struct aq_ring_buff_s *buff = NULL;
 		struct hw_atl_rxd_wb_s *rxd_wb = (struct hw_atl_rxd_wb_s *)
 			&ring->dx_ring[ring->hw_head * HW_ATL_B0_RXD_SIZE];
 
-		unsigned int is_err = 1U;
 		unsigned int is_rx_check_sum_enabled = 0U;
 		unsigned int pkt_type = 0U;
+		u8 rx_stat = 0U;
 
 		if (!(rxd_wb->status & 0x1U)) { /* RxD is not done */
 			break;
@@ -630,62 +921,81 @@ static int hw_atl_b0_hw_ring_rx_receive(struct aq_hw_s *self,
 
 		buff = &ring->buff_ring[ring->hw_head];
 
-		is_err = (0x0000003CU & rxd_wb->status);
+		buff->flags = 0U;
+		buff->is_hash_l4 = 0U;
 
-		is_rx_check_sum_enabled = (rxd_wb->type) & (0x3U << 19);
-		is_err &= ~0x20U; /* exclude validity bit */
+		rx_stat = (0x0000003CU & rxd_wb->status) >> 2;
 
-		pkt_type = 0xFFU & (rxd_wb->type >> 4);
+		is_rx_check_sum_enabled = (rxd_wb->type >> 19) & 0x3U;
 
-		if (is_rx_check_sum_enabled) {
-			if (0x0U == (pkt_type & 0x3U))
-				buff->is_ip_cso = (is_err & 0x08U) ? 0U : 1U;
+		pkt_type = (rxd_wb->type & HW_ATL_B0_RXD_WB_STAT_PKTTYPE) >>
+			   HW_ATL_B0_RXD_WB_STAT_PKTTYPE_SHIFT;
 
+		if (is_rx_check_sum_enabled & BIT(0) &&
+		    (0x0U == (pkt_type & 0x3U)))
+			buff->is_ip_cso = (rx_stat & BIT(1)) ? 0U : 1U;
+
+		if (is_rx_check_sum_enabled & BIT(1)) {
 			if (0x4U == (pkt_type & 0x1CU))
-				buff->is_udp_cso = buff->is_cso_err ? 0U : 1U;
+				buff->is_udp_cso = (rx_stat & BIT(2)) ? 0U :
+						   !!(rx_stat & BIT(3));
 			else if (0x0U == (pkt_type & 0x1CU))
-				buff->is_tcp_cso = buff->is_cso_err ? 0U : 1U;
+				buff->is_tcp_cso = (rx_stat & BIT(2)) ? 0U :
+						   !!(rx_stat & BIT(3));
+		}
+		buff->is_cso_err = !!(rx_stat & 0x6);
+		/* Checksum offload workaround for small packets */
+		if (unlikely(rxd_wb->pkt_len <= 60)) {
+			buff->is_ip_cso = 0U;
+			buff->is_cso_err = 0U;
 		}
 
-		is_err &= ~0x18U;
-
-		dma_unmap_page(ndev, buff->pa, buff->len, DMA_FROM_DEVICE);
+		if (self->aq_nic_cfg->is_vlan_rx_strip &&
+		    ((pkt_type & HW_ATL_B0_RXD_WB_PKTTYPE_VLAN) ||
+		     (pkt_type & HW_ATL_B0_RXD_WB_PKTTYPE_VLAN_DOUBLE))) {
+			buff->is_vlan = 1;
+			buff->vlan_rx_tag = le16_to_cpu(rxd_wb->vlan);
+		}
 
-		if (is_err || rxd_wb->type & 0x1000U) {
-			/* status error or DMA error */
+		if ((rx_stat & BIT(0)) || rxd_wb->type & 0x1000U) {
+			/* MAC error or DMA error */
 			buff->is_error = 1U;
-		} else {
-			if (self->aq_nic_cfg->is_rss) {
-				/* last 4 byte */
-				u16 rss_type = rxd_wb->type & 0xFU;
-
-				if (rss_type && rss_type < 0x8U) {
-					buff->is_hash_l4 = (rss_type == 0x4 ||
-					rss_type == 0x5);
-					buff->rss_hash = rxd_wb->rss_hash;
-				}
+		}
+		if (self->aq_nic_cfg->is_rss) {
+			/* last 4 byte */
+			u16 rss_type = rxd_wb->type & 0xFU;
+
+			if (rss_type && rss_type < 0x8U) {
+				buff->is_hash_l4 = (rss_type == 0x4 ||
+				rss_type == 0x5);
+				buff->rss_hash = rxd_wb->rss_hash;
 			}
+		}
 
-			if (HW_ATL_B0_RXD_WB_STAT2_EOP & rxd_wb->status) {
-				buff->len = rxd_wb->pkt_len %
-					AQ_CFG_RX_FRAME_MAX;
-				buff->len = buff->len ?
-					buff->len : AQ_CFG_RX_FRAME_MAX;
-				buff->next = 0U;
-				buff->is_eop = 1U;
+		buff->is_lro = !!(HW_ATL_B0_RXD_WB_STAT2_RSCCNT &
+				  rxd_wb->status);
+		if (HW_ATL_B0_RXD_WB_STAT2_EOP & rxd_wb->status) {
+			buff->len = rxd_wb->pkt_len %
+				ring->frame_max;
+			buff->len = buff->len ?
+				buff->len : ring->frame_max;
+			buff->next = 0U;
+			buff->is_eop = 1U;
+		} else {
+			buff->len =
+				rxd_wb->pkt_len > ring->frame_max ?
+				ring->frame_max : rxd_wb->pkt_len;
+
+			if (buff->is_lro) {
+				/* LRO */
+				buff->next = rxd_wb->next_desc_ptr;
+				++ring->stats.rx.lro_packets;
 			} else {
-				if (HW_ATL_B0_RXD_WB_STAT2_RSCCNT &
-					rxd_wb->status) {
-					/* LRO */
-					buff->next = rxd_wb->next_desc_ptr;
-					++ring->stats.rx.lro_packets;
-				} else {
-					/* jumbo */
-					buff->next =
-						aq_ring_next_dx(ring,
-								ring->hw_head);
-					++ring->stats.rx.jumbo_packets;
-				}
+				/* jumbo */
+				buff->next =
+					aq_ring_next_dx(ring,
+							ring->hw_head);
+				++ring->stats.rx.jumbo_packets;
 			}
 		}
 	}
@@ -693,50 +1003,64 @@ static int hw_atl_b0_hw_ring_rx_receive(struct aq_hw_s *self,
 	return aq_hw_err_from_flags(self);
 }
 
-static int hw_atl_b0_hw_irq_enable(struct aq_hw_s *self, u64 mask)
+int hw_atl_b0_hw_irq_enable(struct aq_hw_s *self, u64 mask)
 {
-	itr_irq_msk_setlsw_set(self, LODWORD(mask));
+	hw_atl_itr_irq_msk_setlsw_set(self, LODWORD(mask));
+
 	return aq_hw_err_from_flags(self);
 }
 
-static int hw_atl_b0_hw_irq_disable(struct aq_hw_s *self, u64 mask)
+int hw_atl_b0_hw_irq_disable(struct aq_hw_s *self, u64 mask)
 {
-	itr_irq_msk_clearlsw_set(self, LODWORD(mask));
-	itr_irq_status_clearlsw_set(self, LODWORD(mask));
+	hw_atl_itr_irq_msk_clearlsw_set(self, LODWORD(mask));
+	hw_atl_itr_irq_status_clearlsw_set(self, LODWORD(mask));
+
+	atomic_inc(&self->dpc);
 
-	atomic_inc(&PHAL_ATLANTIC_B0->dpc);
 	return aq_hw_err_from_flags(self);
 }
 
-static int hw_atl_b0_hw_irq_read(struct aq_hw_s *self, u64 *mask)
+int hw_atl_b0_hw_irq_read(struct aq_hw_s *self, u64 *mask)
 {
-	*mask = itr_irq_statuslsw_get(self);
+	*mask = hw_atl_itr_irq_statuslsw_get(self);
+
 	return aq_hw_err_from_flags(self);
 }
 
 #define IS_FILTER_ENABLED(_F_) ((packet_filter & (_F_)) ? 1U : 0U)
 
-static int hw_atl_b0_hw_packet_filter_set(struct aq_hw_s *self,
-					  unsigned int packet_filter)
+int hw_atl_b0_hw_packet_filter_set(struct aq_hw_s *self,
+				   unsigned int packet_filter)
 {
+	struct aq_nic_cfg_s *cfg = self->aq_nic_cfg;
 	unsigned int i = 0U;
+	u32 vlan_promisc;
+	u32 l2_promisc;
+
+	l2_promisc = IS_FILTER_ENABLED(IFF_PROMISC) ||
+		     !!(cfg->priv_flags & BIT(AQ_HW_LOOPBACK_DMA_NET));
+	vlan_promisc = l2_promisc || cfg->is_vlan_force_promisc;
+
+	hw_atl_rpfl2promiscuous_mode_en_set(self, l2_promisc);
 
-	rpfl2promiscuous_mode_en_set(self, IS_FILTER_ENABLED(IFF_PROMISC));
-	rpfl2multicast_flr_en_set(self,
-				  IS_FILTER_ENABLED(IFF_MULTICAST), 0);
+	hw_atl_rpf_vlan_prom_mode_en_set(self, vlan_promisc);
 
-	rpfl2_accept_all_mc_packets_set(self,
-					IS_FILTER_ENABLED(IFF_ALLMULTI));
+	hw_atl_rpfl2multicast_flr_en_set(self,
+					 IS_FILTER_ENABLED(IFF_ALLMULTI) &&
+					 IS_FILTER_ENABLED(IFF_MULTICAST), 0);
 
-	rpfl2broadcast_en_set(self, IS_FILTER_ENABLED(IFF_BROADCAST));
+	hw_atl_rpfl2_accept_all_mc_packets_set(self,
+					      IS_FILTER_ENABLED(IFF_ALLMULTI) &&
+					      IS_FILTER_ENABLED(IFF_MULTICAST));
+
+	hw_atl_rpfl2broadcast_en_set(self, IS_FILTER_ENABLED(IFF_BROADCAST));
 
-	self->aq_nic_cfg->is_mc_list_enabled = IS_FILTER_ENABLED(IFF_MULTICAST);
 
 	for (i = HW_ATL_B0_MAC_MIN; i < HW_ATL_B0_MAC_MAX; ++i)
-		rpfl2_uc_flr_en_set(self,
-				    (self->aq_nic_cfg->is_mc_list_enabled &&
-				    (i <= self->aq_nic_cfg->mc_list_count)) ?
-				    1U : 0U, i);
+		hw_atl_rpfl2_uc_flr_en_set(self,
+					   (cfg->is_mc_list_enabled &&
+					    (i <= cfg->mc_list_count)) ?
+					   1U : 0U, i);
 
 	return aq_hw_err_from_flags(self);
 }
@@ -745,35 +1069,36 @@ static int hw_atl_b0_hw_packet_filter_set(struct aq_hw_s *self,
 
 static int hw_atl_b0_hw_multicast_list_set(struct aq_hw_s *self,
 					   u8 ar_mac
-					   [AQ_CFG_MULTICAST_ADDRESS_MAX]
+					   [AQ_HW_MULTICAST_ADDRESS_MAX]
 					   [ETH_ALEN],
 					   u32 count)
 {
 	int err = 0;
+	struct aq_nic_cfg_s *cfg = self->aq_nic_cfg;
 
 	if (count > (HW_ATL_B0_MAC_MAX - HW_ATL_B0_MAC_MIN)) {
 		err = -EBADRQC;
 		goto err_exit;
 	}
-	for (self->aq_nic_cfg->mc_list_count = 0U;
-			self->aq_nic_cfg->mc_list_count < count;
-			++self->aq_nic_cfg->mc_list_count) {
-		u32 i = self->aq_nic_cfg->mc_list_count;
+	for (cfg->mc_list_count = 0U;
+			cfg->mc_list_count < count;
+			++cfg->mc_list_count) {
+		u32 i = cfg->mc_list_count;
 		u32 h = (ar_mac[i][0] << 8) | (ar_mac[i][1]);
 		u32 l = (ar_mac[i][2] << 24) | (ar_mac[i][3] << 16) |
 					(ar_mac[i][4] << 8) | ar_mac[i][5];
 
-		rpfl2_uc_flr_en_set(self, 0U, HW_ATL_B0_MAC_MIN + i);
+		hw_atl_rpfl2_uc_flr_en_set(self, 0U, HW_ATL_B0_MAC_MIN + i);
 
-		rpfl2unicast_dest_addresslsw_set(self,
-						 l, HW_ATL_B0_MAC_MIN + i);
+		hw_atl_rpfl2unicast_dest_addresslsw_set(self, l,
+							HW_ATL_B0_MAC_MIN + i);
 
-		rpfl2unicast_dest_addressmsw_set(self,
-						 h, HW_ATL_B0_MAC_MIN + i);
+		hw_atl_rpfl2unicast_dest_addressmsw_set(self, h,
+							HW_ATL_B0_MAC_MIN + i);
 
-		rpfl2_uc_flr_en_set(self,
-				    (self->aq_nic_cfg->is_mc_list_enabled),
-				    HW_ATL_B0_MAC_MIN + i);
+		hw_atl_rpfl2_uc_flr_en_set(self,
+					   (cfg->is_mc_list_enabled),
+					   HW_ATL_B0_MAC_MIN + i);
 	}
 
 	err = aq_hw_err_from_flags(self);
@@ -782,39 +1107,45 @@ err_exit:
 	return err;
 }
 
-static int hw_atl_b0_hw_interrupt_moderation_set(struct aq_hw_s *self,
-						 bool itr_enabled)
+static int hw_atl_b0_hw_interrupt_moderation_set(struct aq_hw_s *self)
 {
 	unsigned int i = 0U;
-
-	if (itr_enabled && self->aq_nic_cfg->itr) {
-		tdm_tx_desc_wr_wb_irq_en_set(self, 0U);
-		tdm_tdm_intr_moder_en_set(self, 1U);
-		rdm_rx_desc_wr_wb_irq_en_set(self, 0U);
-		rdm_rdm_intr_moder_en_set(self, 1U);
-
-		PHAL_ATLANTIC_B0->itr_tx = 2U;
-		PHAL_ATLANTIC_B0->itr_rx = 2U;
-
-		if (self->aq_nic_cfg->itr != 0xFFFFU) {
-			unsigned int max_timer = self->aq_nic_cfg->itr / 2U;
-			unsigned int min_timer = self->aq_nic_cfg->itr / 32U;
-
-			max_timer = min(0x1FFU, max_timer);
-			min_timer = min(0xFFU, min_timer);
-
-			PHAL_ATLANTIC_B0->itr_tx |= min_timer << 0x8U;
-			PHAL_ATLANTIC_B0->itr_tx |= max_timer << 0x10U;
-			PHAL_ATLANTIC_B0->itr_rx |= min_timer << 0x8U;
-			PHAL_ATLANTIC_B0->itr_rx |= max_timer << 0x10U;
+	u32 itr_tx = 2U;
+	u32 itr_rx = 2U;
+
+	switch (self->aq_nic_cfg->itr) {
+	case  AQ_CFG_INTERRUPT_MODERATION_ON:
+	case  AQ_CFG_INTERRUPT_MODERATION_AUTO:
+		hw_atl_tdm_tx_desc_wr_wb_irq_en_set(self, 0U);
+		hw_atl_tdm_tdm_intr_moder_en_set(self, 1U);
+		hw_atl_rdm_rx_desc_wr_wb_irq_en_set(self, 0U);
+		hw_atl_rdm_rdm_intr_moder_en_set(self, 1U);
+
+		if (self->aq_nic_cfg->itr == AQ_CFG_INTERRUPT_MODERATION_ON) {
+			/* HW timers are in 2us units */
+			int tx_max_timer = self->aq_nic_cfg->tx_itr / 2;
+			int tx_min_timer = tx_max_timer / 2;
+
+			int rx_max_timer = self->aq_nic_cfg->rx_itr / 2;
+			int rx_min_timer = rx_max_timer / 2;
+
+			tx_max_timer = min(HW_ATL_INTR_MODER_MAX, tx_max_timer);
+			tx_min_timer = min(HW_ATL_INTR_MODER_MIN, tx_min_timer);
+			rx_max_timer = min(HW_ATL_INTR_MODER_MAX, rx_max_timer);
+			rx_min_timer = min(HW_ATL_INTR_MODER_MIN, rx_min_timer);
+
+			itr_tx |= tx_min_timer << 0x8U;
+			itr_tx |= tx_max_timer << 0x10U;
+			itr_rx |= rx_min_timer << 0x8U;
+			itr_rx |= rx_max_timer << 0x10U;
 		} else {
 			static unsigned int hw_atl_b0_timers_table_tx_[][2] = {
-				{0xffU, 0xffU}, /* 10Gbit */
-				{0xffU, 0x1ffU}, /* 5Gbit */
-				{0xffU, 0x1ffU}, /* 5Gbit 5GS */
-				{0xffU, 0x1ffU}, /* 2.5Gbit */
-				{0xffU, 0x1ffU}, /* 1Gbit */
-				{0xffU, 0x1ffU}, /* 100Mbit */
+				{0xfU, 0xffU}, /* 10Gbit */
+				{0xfU, 0x1ffU}, /* 5Gbit */
+				{0xfU, 0x1ffU}, /* 5Gbit 5GS */
+				{0xfU, 0x1ffU}, /* 2.5Gbit */
+				{0xfU, 0x1ffU}, /* 1Gbit */
+				{0xfU, 0x1ffU}, /* 100Mbit */
 			};
 
 			static unsigned int hw_atl_b0_timers_table_rx_[][2] = {
@@ -830,34 +1161,36 @@ static int hw_atl_b0_hw_interrupt_moderation_set(struct aq_hw_s *self,
 					hw_atl_utils_mbps_2_speed_index(
 						self->aq_link_status.mbps);
 
-			PHAL_ATLANTIC_B0->itr_tx |=
-				hw_atl_b0_timers_table_tx_[speed_index]
-				[0] << 0x8U; /* set min timer value */
-			PHAL_ATLANTIC_B0->itr_tx |=
-				hw_atl_b0_timers_table_tx_[speed_index]
-				[1] << 0x10U; /* set max timer value */
-
-			PHAL_ATLANTIC_B0->itr_rx |=
-				hw_atl_b0_timers_table_rx_[speed_index]
-				[0] << 0x8U; /* set min timer value */
-			PHAL_ATLANTIC_B0->itr_rx |=
-				hw_atl_b0_timers_table_rx_[speed_index]
-				[1] << 0x10U; /* set max timer value */
+			/* Update user visible ITR settings */
+			self->aq_nic_cfg->tx_itr = hw_atl_b0_timers_table_tx_
+							[speed_index][1] * 2;
+			self->aq_nic_cfg->rx_itr = hw_atl_b0_timers_table_rx_
+							[speed_index][1] * 2;
+
+			itr_tx |= hw_atl_b0_timers_table_tx_
+						[speed_index][0] << 0x8U;
+			itr_tx |= hw_atl_b0_timers_table_tx_
+						[speed_index][1] << 0x10U;
+
+			itr_rx |= hw_atl_b0_timers_table_rx_
+						[speed_index][0] << 0x8U;
+			itr_rx |= hw_atl_b0_timers_table_rx_
+						[speed_index][1] << 0x10U;
 		}
-	} else {
-		tdm_tx_desc_wr_wb_irq_en_set(self, 1U);
-		tdm_tdm_intr_moder_en_set(self, 0U);
-		rdm_rx_desc_wr_wb_irq_en_set(self, 1U);
-		rdm_rdm_intr_moder_en_set(self, 0U);
-		PHAL_ATLANTIC_B0->itr_tx = 0U;
-		PHAL_ATLANTIC_B0->itr_rx = 0U;
+		break;
+	case AQ_CFG_INTERRUPT_MODERATION_OFF:
+		hw_atl_tdm_tx_desc_wr_wb_irq_en_set(self, 1U);
+		hw_atl_tdm_tdm_intr_moder_en_set(self, 0U);
+		hw_atl_rdm_rx_desc_wr_wb_irq_en_set(self, 1U);
+		hw_atl_rdm_rdm_intr_moder_en_set(self, 0U);
+		itr_tx = 0U;
+		itr_rx = 0U;
+		break;
 	}
 
 	for (i = HW_ATL_B0_RINGS_MAX; i--;) {
-		reg_tx_intr_moder_ctrl_set(self,
-					   PHAL_ATLANTIC_B0->itr_tx, i);
-		reg_rx_intr_moder_ctrl_set(self,
-					   PHAL_ATLANTIC_B0->itr_rx, i);
+		hw_atl_reg_tx_intr_moder_ctrl_set(self, itr_tx, i);
+		hw_atl_reg_rx_intr_moder_ctrl_set(self, itr_rx, i);
 	}
 
 	return aq_hw_err_from_flags(self);
@@ -865,48 +1198,598 @@ static int hw_atl_b0_hw_interrupt_moderation_set(struct aq_hw_s *self,
 
 static int hw_atl_b0_hw_stop(struct aq_hw_s *self)
 {
+	int err;
+	u32 val;
+
 	hw_atl_b0_hw_irq_disable(self, HW_ATL_B0_INT_MASK);
+
+	/* Invalidate Descriptor Cache to prevent writing to the cached
+	 * descriptors and to the data pointer of those descriptors
+	 */
+	hw_atl_rdm_rx_dma_desc_cache_init_tgl(self);
+
+	err = aq_hw_err_from_flags(self);
+
+	if (err)
+		goto err_exit;
+
+	readx_poll_timeout_atomic(hw_atl_rdm_rx_dma_desc_cache_init_done_get,
+				  self, val, val == 1, 1000U, 10000U);
+
+err_exit:
+	return err;
+}
+
+int hw_atl_b0_hw_ring_tx_stop(struct aq_hw_s *self, struct aq_ring_s *ring)
+{
+	hw_atl_tdm_tx_desc_en_set(self, 0U, ring->idx);
+
 	return aq_hw_err_from_flags(self);
 }
 
-static int hw_atl_b0_hw_ring_tx_stop(struct aq_hw_s *self,
-				     struct aq_ring_s *ring)
+int hw_atl_b0_hw_ring_rx_stop(struct aq_hw_s *self, struct aq_ring_s *ring)
 {
-	tdm_tx_desc_en_set(self, 0U, ring->idx);
+	hw_atl_rdm_rx_desc_en_set(self, 0U, ring->idx);
+
 	return aq_hw_err_from_flags(self);
 }
 
-static int hw_atl_b0_hw_ring_rx_stop(struct aq_hw_s *self,
-				     struct aq_ring_s *ring)
+#define get_ptp_ts_val_u64(self, indx) \
+	((u64)(hw_atl_pcs_ptp_clock_get(self, indx) & 0xffff))
+
+static void hw_atl_b0_get_ptp_ts(struct aq_hw_s *self, u64 *stamp)
+{
+	u64 ns;
+
+	hw_atl_pcs_ptp_clock_read_enable(self, 1);
+	hw_atl_pcs_ptp_clock_read_enable(self, 0);
+	ns = (get_ptp_ts_val_u64(self, 0) +
+	      (get_ptp_ts_val_u64(self, 1) << 16)) * NSEC_PER_SEC +
+	     (get_ptp_ts_val_u64(self, 3) +
+	      (get_ptp_ts_val_u64(self, 4) << 16));
+
+	*stamp = ns + self->ptp_clk_offset;
+}
+
+static void hw_atl_b0_adj_params_get(u64 freq, s64 adj, u32 *ns, u32 *fns)
+{
+	/* For accuracy, the digit is extended */
+	s64 base_ns = ((adj + NSEC_PER_SEC) * NSEC_PER_SEC);
+	u64 nsi_frac = 0;
+	u64 nsi;
+
+	base_ns = div64_s64(base_ns, freq);
+	nsi = div64_u64(base_ns, NSEC_PER_SEC);
+
+	if (base_ns != nsi * NSEC_PER_SEC) {
+		s64 divisor = div64_s64((s64)NSEC_PER_SEC * NSEC_PER_SEC,
+					base_ns - nsi * NSEC_PER_SEC);
+		nsi_frac = div64_s64(AQ_FRAC_PER_NS * NSEC_PER_SEC, divisor);
+	}
+
+	*ns = (u32)nsi;
+	*fns = (u32)nsi_frac;
+}
+
+static void
+hw_atl_b0_mac_adj_param_calc(struct hw_fw_request_ptp_adj_freq *ptp_adj_freq,
+			     u64 phyfreq, u64 macfreq)
 {
-	rdm_rx_desc_en_set(self, 0U, ring->idx);
+	s64 adj_fns_val;
+	s64 fns_in_sec_phy = phyfreq * (ptp_adj_freq->fns_phy +
+					AQ_FRAC_PER_NS * ptp_adj_freq->ns_phy);
+	s64 fns_in_sec_mac = macfreq * (ptp_adj_freq->fns_mac +
+					AQ_FRAC_PER_NS * ptp_adj_freq->ns_mac);
+	s64 fault_in_sec_phy = AQ_FRAC_PER_NS * NSEC_PER_SEC - fns_in_sec_phy;
+	s64 fault_in_sec_mac = AQ_FRAC_PER_NS * NSEC_PER_SEC - fns_in_sec_mac;
+	/* MAC MCP counter freq is macfreq / 4 */
+	s64 diff_in_mcp_overflow = (fault_in_sec_mac - fault_in_sec_phy) *
+				   4 * AQ_FRAC_PER_NS;
+
+	diff_in_mcp_overflow = div64_s64(diff_in_mcp_overflow,
+					 AQ_HW_MAC_COUNTER_HZ);
+	adj_fns_val = (ptp_adj_freq->fns_mac + AQ_FRAC_PER_NS *
+		       ptp_adj_freq->ns_mac) + diff_in_mcp_overflow;
+
+	ptp_adj_freq->mac_ns_adj = div64_s64(adj_fns_val, AQ_FRAC_PER_NS);
+	ptp_adj_freq->mac_fns_adj = adj_fns_val - ptp_adj_freq->mac_ns_adj *
+				    AQ_FRAC_PER_NS;
+}
+
+static int hw_atl_b0_adj_sys_clock(struct aq_hw_s *self, s64 delta)
+{
+	self->ptp_clk_offset += delta;
+
+	self->aq_fw_ops->adjust_ptp(self, self->ptp_clk_offset);
+
+	return 0;
+}
+
+static int hw_atl_b0_set_sys_clock(struct aq_hw_s *self, u64 time, u64 ts)
+{
+	s64 delta = time - (self->ptp_clk_offset + ts);
+
+	return hw_atl_b0_adj_sys_clock(self, delta);
+}
+
+static int hw_atl_b0_ts_to_sys_clock(struct aq_hw_s *self, u64 ts, u64 *time)
+{
+	*time = self->ptp_clk_offset + ts;
+	return 0;
+}
+
+static int hw_atl_b0_adj_clock_freq(struct aq_hw_s *self, s32 ppb)
+{
+	struct hw_fw_request_iface fwreq;
+	size_t size;
+
+	memset(&fwreq, 0, sizeof(fwreq));
+
+	fwreq.msg_id = HW_AQ_FW_REQUEST_PTP_ADJ_FREQ;
+	hw_atl_b0_adj_params_get(AQ_HW_MAC_COUNTER_HZ, ppb,
+				 &fwreq.ptp_adj_freq.ns_mac,
+				 &fwreq.ptp_adj_freq.fns_mac);
+	hw_atl_b0_adj_params_get(AQ_HW_PHY_COUNTER_HZ, ppb,
+				 &fwreq.ptp_adj_freq.ns_phy,
+				 &fwreq.ptp_adj_freq.fns_phy);
+	hw_atl_b0_mac_adj_param_calc(&fwreq.ptp_adj_freq,
+				     AQ_HW_PHY_COUNTER_HZ,
+				     AQ_HW_MAC_COUNTER_HZ);
+
+	size = sizeof(fwreq.msg_id) + sizeof(fwreq.ptp_adj_freq);
+	return self->aq_fw_ops->send_fw_request(self, &fwreq, size);
+}
+
+static int hw_atl_b0_gpio_pulse(struct aq_hw_s *self, u32 index,
+				u64 start, u32 period)
+{
+	struct hw_fw_request_iface fwreq;
+	size_t size;
+
+	memset(&fwreq, 0, sizeof(fwreq));
+
+	fwreq.msg_id = HW_AQ_FW_REQUEST_PTP_GPIO_CTRL;
+	fwreq.ptp_gpio_ctrl.index = index;
+	fwreq.ptp_gpio_ctrl.period = period;
+	/* Apply time offset */
+	fwreq.ptp_gpio_ctrl.start = start;
+
+	size = sizeof(fwreq.msg_id) + sizeof(fwreq.ptp_gpio_ctrl);
+	return self->aq_fw_ops->send_fw_request(self, &fwreq, size);
+}
+
+static int hw_atl_b0_extts_gpio_enable(struct aq_hw_s *self, u32 index,
+				       u32 enable)
+{
+	/* Enable/disable Sync1588 GPIO Timestamping */
+	aq_phy_write_reg(self, MDIO_MMD_PCS, 0xc611, enable ? 0x71 : 0);
+
+	return 0;
+}
+
+static int hw_atl_b0_get_sync_ts(struct aq_hw_s *self, u64 *ts)
+{
+	u64 sec_l;
+	u64 sec_h;
+	u64 nsec_l;
+	u64 nsec_h;
+
+	if (!ts)
+		return -1;
+
+	/* PTP external GPIO clock seconds count 15:0 */
+	sec_l = aq_phy_read_reg(self, MDIO_MMD_PCS, 0xc914);
+	/* PTP external GPIO clock seconds count 31:16 */
+	sec_h = aq_phy_read_reg(self, MDIO_MMD_PCS, 0xc915);
+	/* PTP external GPIO clock nanoseconds count 15:0 */
+	nsec_l = aq_phy_read_reg(self, MDIO_MMD_PCS, 0xc916);
+	/* PTP external GPIO clock nanoseconds count 31:16 */
+	nsec_h = aq_phy_read_reg(self, MDIO_MMD_PCS, 0xc917);
+
+	*ts = (nsec_h << 16) + nsec_l + ((sec_h << 16) + sec_l) * NSEC_PER_SEC;
+
+	return 0;
+}
+
+static u16 hw_atl_b0_rx_extract_ts(struct aq_hw_s *self, u8 *p,
+				   unsigned int len, u64 *timestamp)
+{
+	unsigned int offset = 14;
+	struct ethhdr *eth;
+	__be64 sec;
+	__be32 ns;
+	u8 *ptr;
+
+	if (len <= offset || !timestamp)
+		return 0;
+
+	/* The TIMESTAMP in the end of package has following format:
+	 * (big-endian)
+	 *   struct {
+	 *     uint64_t sec;
+	 *     uint32_t ns;
+	 *     uint16_t stream_id;
+	 *   };
+	 */
+	ptr = p + (len - offset);
+	memcpy(&sec, ptr, sizeof(sec));
+	ptr += sizeof(sec);
+	memcpy(&ns, ptr, sizeof(ns));
+
+	*timestamp = (be64_to_cpu(sec) & 0xffffffffffffllu) * NSEC_PER_SEC +
+		     be32_to_cpu(ns) + self->ptp_clk_offset;
+
+	eth = (struct ethhdr *)p;
+
+	return (eth->h_proto == htons(ETH_P_1588)) ? 12 : 14;
+}
+
+static int hw_atl_b0_extract_hwts(struct aq_hw_s *self, u8 *p, unsigned int len,
+				  u64 *timestamp)
+{
+	struct hw_atl_rxd_hwts_wb_s *hwts_wb = (struct hw_atl_rxd_hwts_wb_s *)p;
+	u64 tmp, sec, ns;
+
+	sec = 0;
+	tmp = (hwts_wb->sec_lw0 >> 2) & 0x3ff;
+	sec += tmp;
+	tmp = (u64)((hwts_wb->sec_lw1 >> 16) & 0xffff) << 10;
+	sec += tmp;
+	tmp = (u64)(hwts_wb->sec_hw & 0xfff) << 26;
+	sec += tmp;
+	tmp = (u64)((hwts_wb->sec_hw >> 22) & 0x3ff) << 38;
+	sec += tmp;
+	ns = sec * NSEC_PER_SEC + hwts_wb->ns;
+	if (timestamp)
+		*timestamp = ns + self->ptp_clk_offset;
+	return 0;
+}
+
+static int hw_atl_b0_hw_fl3l4_clear(struct aq_hw_s *self,
+				    struct aq_rx_filter_l3l4 *data)
+{
+	u8 location = data->location;
+
+	if (!data->is_ipv6) {
+		hw_atl_rpfl3l4_cmd_clear(self, location);
+		hw_atl_rpf_l4_spd_set(self, 0U, location);
+		hw_atl_rpf_l4_dpd_set(self, 0U, location);
+		hw_atl_rpfl3l4_ipv4_src_addr_clear(self, location);
+		hw_atl_rpfl3l4_ipv4_dest_addr_clear(self, location);
+	} else {
+		int i;
+
+		for (i = 0; i < HW_ATL_RX_CNT_REG_ADDR_IPV6; ++i) {
+			hw_atl_rpfl3l4_cmd_clear(self, location + i);
+			hw_atl_rpf_l4_spd_set(self, 0U, location + i);
+			hw_atl_rpf_l4_dpd_set(self, 0U, location + i);
+		}
+		hw_atl_rpfl3l4_ipv6_src_addr_clear(self, location);
+		hw_atl_rpfl3l4_ipv6_dest_addr_clear(self, location);
+	}
+
 	return aq_hw_err_from_flags(self);
 }
 
-static int hw_atl_b0_hw_set_speed(struct aq_hw_s *self, u32 speed)
+static int hw_atl_b0_hw_fl3l4_set(struct aq_hw_s *self,
+				  struct aq_rx_filter_l3l4 *data)
 {
-	int err = 0;
+	u8 location = data->location;
+
+	hw_atl_b0_hw_fl3l4_clear(self, data);
+
+	if (data->cmd & (HW_ATL_RX_ENABLE_CMP_DEST_ADDR_L3 |
+			 HW_ATL_RX_ENABLE_CMP_SRC_ADDR_L3)) {
+		if (!data->is_ipv6) {
+			hw_atl_rpfl3l4_ipv4_dest_addr_set(self,
+							  location,
+							  data->ip_dst[0]);
+			hw_atl_rpfl3l4_ipv4_src_addr_set(self,
+							 location,
+							 data->ip_src[0]);
+		} else {
+			hw_atl_rpfl3l4_ipv6_dest_addr_set(self,
+							  location,
+							  data->ip_dst);
+			hw_atl_rpfl3l4_ipv6_src_addr_set(self,
+							 location,
+							 data->ip_src);
+		}
+	}
 
-	err = hw_atl_utils_mpi_set_speed(self, speed, MPI_INIT);
-	if (err < 0)
-		goto err_exit;
+	if (data->cmd & (HW_ATL_RX_ENABLE_CMP_DEST_PORT_L4 |
+			 HW_ATL_RX_ENABLE_CMP_SRC_PORT_L4)) {
+		hw_atl_rpf_l4_dpd_set(self, data->p_dst, location);
+		hw_atl_rpf_l4_spd_set(self, data->p_src, location);
+	}
 
-err_exit:
-	return err;
+	hw_atl_rpfl3l4_cmd_set(self, location, data->cmd);
+
+	return aq_hw_err_from_flags(self);
 }
 
-static struct aq_hw_ops hw_atl_ops_ = {
-	.create               = hw_atl_b0_create,
-	.destroy              = hw_atl_b0_destroy,
-	.get_hw_caps          = hw_atl_b0_get_hw_caps,
+static int hw_atl_b0_hw_fl2_set(struct aq_hw_s *self,
+				struct aq_rx_filter_l2 *data)
+{
+	hw_atl_rpf_etht_flr_en_set(self, 1U, data->location);
+	hw_atl_rpf_etht_flr_set(self, data->ethertype, data->location);
+	hw_atl_rpf_etht_user_priority_en_set(self,
+					     !!data->user_priority_en,
+					     data->location);
+	if (data->user_priority_en)
+		hw_atl_rpf_etht_user_priority_set(self,
+						  data->user_priority,
+						  data->location);
+
+	if (data->queue < 0) {
+		hw_atl_rpf_etht_flr_act_set(self, 0U, data->location);
+		hw_atl_rpf_etht_rx_queue_en_set(self, 0U, data->location);
+	} else {
+		hw_atl_rpf_etht_flr_act_set(self, 1U, data->location);
+		hw_atl_rpf_etht_rx_queue_en_set(self, 1U, data->location);
+		hw_atl_rpf_etht_rx_queue_set(self, data->queue, data->location);
+	}
+
+	return aq_hw_err_from_flags(self);
+}
 
-	.hw_get_mac_permanent = hw_atl_utils_get_mac_permanent,
+static int hw_atl_b0_hw_fl2_clear(struct aq_hw_s *self,
+				  struct aq_rx_filter_l2 *data)
+{
+	hw_atl_rpf_etht_flr_en_set(self, 0U, data->location);
+	hw_atl_rpf_etht_flr_set(self, 0U, data->location);
+	hw_atl_rpf_etht_user_priority_en_set(self, 0U, data->location);
+
+	return aq_hw_err_from_flags(self);
+}
+
+/*
+ * @brief Set VLAN filter table
+ * @details Configure VLAN filter table to accept (and assign the queue) traffic
+ *  for the particular vlan ids.
+ * Note: use this function under vlan promisc mode not to lost the traffic
+ *
+ * @param aq_hw_s
+ * @param aq_rx_filter_vlan VLAN filter configuration
+ * @return 0 - OK, <0 - error
+ */
+static int hw_atl_b0_hw_vlan_set(struct aq_hw_s *self,
+				 struct aq_rx_filter_vlan *aq_vlans)
+{
+	int i;
+
+	for (i = 0; i < AQ_VLAN_MAX_FILTERS; i++) {
+		hw_atl_rpf_vlan_flr_en_set(self, 0U, i);
+		hw_atl_rpf_vlan_rxq_en_flr_set(self, 0U, i);
+		if (aq_vlans[i].enable) {
+			hw_atl_rpf_vlan_id_flr_set(self,
+						   aq_vlans[i].vlan_id,
+						   i);
+			hw_atl_rpf_vlan_flr_act_set(self, 1U, i);
+			hw_atl_rpf_vlan_flr_en_set(self, 1U, i);
+			if (aq_vlans[i].queue != 0xFF) {
+				hw_atl_rpf_vlan_rxq_flr_set(self,
+							    aq_vlans[i].queue,
+							    i);
+				hw_atl_rpf_vlan_rxq_en_flr_set(self, 1U, i);
+			}
+		}
+	}
+
+	return aq_hw_err_from_flags(self);
+}
+
+static int hw_atl_b0_hw_vlan_ctrl(struct aq_hw_s *self, bool enable)
+{
+	/* set promisc in case of disabing the vland filter */
+	hw_atl_rpf_vlan_prom_mode_en_set(self, !enable);
+
+	return aq_hw_err_from_flags(self);
+}
+
+int hw_atl_b0_set_loopback(struct aq_hw_s *self, u32 mode, bool enable)
+{
+	switch (mode) {
+	case AQ_HW_LOOPBACK_DMA_SYS:
+		hw_atl_tpb_tx_dma_sys_lbk_en_set(self, enable);
+		hw_atl_rpb_dma_sys_lbk_set(self, enable);
+		break;
+	case AQ_HW_LOOPBACK_PKT_SYS:
+		hw_atl_tpo_tx_pkt_sys_lbk_en_set(self, enable);
+		hw_atl_rpf_tpo_to_rpf_sys_lbk_set(self, enable);
+		break;
+	case AQ_HW_LOOPBACK_DMA_NET:
+		hw_atl_rpf_vlan_prom_mode_en_set(self, enable);
+		hw_atl_rpfl2promiscuous_mode_en_set(self, enable);
+		hw_atl_tpb_tx_tx_clk_gate_en_set(self, !enable);
+		hw_atl_tpb_tx_dma_net_lbk_en_set(self, enable);
+		hw_atl_rpb_dma_net_lbk_set(self, enable);
+		break;
+	default:
+		return -EINVAL;
+	}
+
+	return 0;
+}
+
+static u32 hw_atl_b0_ts_ready_and_latch_high_get(struct aq_hw_s *self)
+{
+	if (hw_atl_ts_ready_get(self) && hw_atl_ts_ready_latch_high_get(self))
+		return 1;
+
+	return 0;
+}
+
+static int hw_atl_b0_get_mac_temp(struct aq_hw_s *self, u32 *temp)
+{
+	bool ts_disabled;
+	int err;
+	u32 val;
+	u32 ts;
+
+	ts_disabled = (hw_atl_ts_power_down_get(self) == 1U);
+
+	if (ts_disabled) {
+		// Set AFE Temperature Sensor to on (off by default)
+		hw_atl_ts_power_down_set(self, 0U);
+
+		// Reset internal capacitors, biasing, and counters
+		hw_atl_ts_reset_set(self, 1);
+		hw_atl_ts_reset_set(self, 0);
+	}
+
+	err = readx_poll_timeout(hw_atl_b0_ts_ready_and_latch_high_get, self,
+				 val, val == 1, 10000U, 500000U);
+	if (err)
+		return err;
+
+	ts = hw_atl_ts_data_get(self);
+	*temp = ts * ts * 16 / 100000 + 60 * ts - 83410;
+
+	if (ts_disabled) {
+		// Set AFE Temperature Sensor back to off
+		hw_atl_ts_power_down_set(self, 1U);
+	}
+
+	return 0;
+}
+
+#define START_TRANSMIT 0x5001
+#define START_READ_TRANSMIT 0x5101
+#define STOP_TRANSMIT 0x3001
+#define REPEAT_TRANSMIT 0x1001
+#define REPEAT_NACK_TRANSMIT 0x1011
+
+static int hw_atl_b0_smb0_wait_result(struct aq_hw_s *self, bool expect_ack)
+{
+	int err;
+	u32 val;
+
+	err = readx_poll_timeout(hw_atl_smb0_byte_transfer_complete_get,
+				 self, val, val == 1, 100U, 10000U);
+	if (err)
+		return err;
+	if (hw_atl_smb0_receive_acknowledged_get(self) != expect_ack)
+		return -EIO;
+	return 0;
+}
+
+/* Starts an I2C/SMBUS write to a given address. addr is in 7-bit format,
+ * the read/write bit is not part of it.
+ */
+static int hw_atl_b0_smb0_start_write(struct aq_hw_s *self, u32 addr)
+{
+	hw_atl_smb0_tx_data_set(self, (addr << 1) | 0);
+	hw_atl_smb0_provisioning2_set(self, START_TRANSMIT);
+	return hw_atl_b0_smb0_wait_result(self, 0);
+}
+
+/* Writes a single byte as part of an ongoing write started by start_write. */
+static int hw_atl_b0_smb0_write_byte(struct aq_hw_s *self, u32 data)
+{
+	hw_atl_smb0_tx_data_set(self, data);
+	hw_atl_smb0_provisioning2_set(self, REPEAT_TRANSMIT);
+	return hw_atl_b0_smb0_wait_result(self, 0);
+}
+
+/* Starts an I2C/SMBUS read to a given address. addr is in 7-bit format,
+ * the read/write bit is not part of it.
+ */
+static int hw_atl_b0_smb0_start_read(struct aq_hw_s *self, u32 addr)
+{
+	int err;
+
+	hw_atl_smb0_tx_data_set(self, (addr << 1) | 1);
+	hw_atl_smb0_provisioning2_set(self, START_READ_TRANSMIT);
+	err = hw_atl_b0_smb0_wait_result(self, 0);
+	if (err)
+		return err;
+	if (hw_atl_smb0_repeated_start_detect_get(self) == 0)
+		return -EIO;
+	return 0;
+}
+
+/* Reads a single byte as part of an ongoing read started by start_read. */
+static int hw_atl_b0_smb0_read_byte(struct aq_hw_s *self)
+{
+	int err;
+
+	hw_atl_smb0_provisioning2_set(self, REPEAT_TRANSMIT);
+	err = hw_atl_b0_smb0_wait_result(self, 0);
+	if (err)
+		return err;
+	return hw_atl_smb0_rx_data_get(self);
+}
+
+/* Reads the last byte of an ongoing read. */
+static int hw_atl_b0_smb0_read_byte_nack(struct aq_hw_s *self)
+{
+	int err;
+
+	hw_atl_smb0_provisioning2_set(self, REPEAT_NACK_TRANSMIT);
+	err = hw_atl_b0_smb0_wait_result(self, 1);
+	if (err)
+		return err;
+	return hw_atl_smb0_rx_data_get(self);
+}
+
+/* Sends a stop condition and ends a transfer. */
+static void hw_atl_b0_smb0_stop(struct aq_hw_s *self)
+{
+	hw_atl_smb0_provisioning2_set(self, STOP_TRANSMIT);
+}
+
+static int hw_atl_b0_read_module_eeprom(struct aq_hw_s *self, u8 dev_addr,
+					u8 reg_start_addr, int len, u8 *data)
+{
+	int i, b;
+	int err;
+	u32 val;
+
+	/* Wait for SMBUS0 to be idle */
+	err = readx_poll_timeout(hw_atl_smb0_bus_busy_get, self,
+				 val, val == 0, 100U, 10000U);
+	if (err)
+		return err;
+
+	err = hw_atl_b0_smb0_start_write(self, dev_addr);
+	if (err)
+		goto out;
+
+	err = hw_atl_b0_smb0_write_byte(self, reg_start_addr);
+	if (err)
+		goto out;
+
+	err = hw_atl_b0_smb0_start_read(self, dev_addr);
+	if (err)
+		goto out;
+
+	for (i = 0; i < len - 1; i++) {
+		b = hw_atl_b0_smb0_read_byte(self);
+		if (b < 0) {
+			err = b;
+			goto out;
+		}
+		data[i] = (u8)b;
+	}
+
+	b = hw_atl_b0_smb0_read_byte_nack(self);
+	if (b < 0) {
+		err = b;
+		goto out;
+	}
+	data[i] = (u8)b;
+
+out:
+	hw_atl_b0_smb0_stop(self);
+	return err;
+}
+
+const struct aq_hw_ops hw_atl_ops_b0 = {
+	.hw_soft_reset        = hw_atl_utils_soft_reset,
+	.hw_prepare           = hw_atl_utils_initfw,
 	.hw_set_mac_address   = hw_atl_b0_hw_mac_addr_set,
-	.hw_get_link_status   = hw_atl_utils_mpi_get_link_status,
-	.hw_set_link_speed    = hw_atl_b0_hw_set_speed,
 	.hw_init              = hw_atl_b0_hw_init,
-	.hw_deinit            = hw_atl_utils_hw_deinit,
-	.hw_set_power         = hw_atl_utils_hw_set_power,
 	.hw_reset             = hw_atl_b0_hw_reset,
 	.hw_start             = hw_atl_b0_hw_start,
 	.hw_ring_tx_start     = hw_atl_b0_hw_ring_tx_start,
@@ -928,25 +1811,37 @@ static struct aq_hw_ops hw_atl_ops_ = {
 	.hw_ring_rx_init             = hw_atl_b0_hw_ring_rx_init,
 	.hw_ring_tx_init             = hw_atl_b0_hw_ring_tx_init,
 	.hw_packet_filter_set        = hw_atl_b0_hw_packet_filter_set,
+	.hw_filter_l2_set            = hw_atl_b0_hw_fl2_set,
+	.hw_filter_l2_clear          = hw_atl_b0_hw_fl2_clear,
+	.hw_filter_l3l4_set          = hw_atl_b0_hw_fl3l4_set,
+	.hw_filter_vlan_set          = hw_atl_b0_hw_vlan_set,
+	.hw_filter_vlan_ctrl         = hw_atl_b0_hw_vlan_ctrl,
 	.hw_multicast_list_set       = hw_atl_b0_hw_multicast_list_set,
 	.hw_interrupt_moderation_set = hw_atl_b0_hw_interrupt_moderation_set,
 	.hw_rss_set                  = hw_atl_b0_hw_rss_set,
 	.hw_rss_hash_set             = hw_atl_b0_hw_rss_hash_set,
+	.hw_tc_rate_limit_set        = hw_atl_b0_hw_init_tx_tc_rate_limit,
 	.hw_get_regs                 = hw_atl_utils_hw_get_regs,
 	.hw_get_hw_stats             = hw_atl_utils_get_hw_stats,
 	.hw_get_fw_version           = hw_atl_utils_get_fw_version,
-};
 
-struct aq_hw_ops *hw_atl_b0_get_ops_by_id(struct pci_dev *pdev)
-{
-	bool is_vid_ok = (pdev->vendor == PCI_VENDOR_ID_AQUANTIA);
-	bool is_did_ok = ((pdev->device == HW_ATL_DEVICE_ID_0001) ||
-			(pdev->device == HW_ATL_DEVICE_ID_D100) ||
-			(pdev->device == HW_ATL_DEVICE_ID_D107) ||
-			(pdev->device == HW_ATL_DEVICE_ID_D108) ||
-			(pdev->device == HW_ATL_DEVICE_ID_D109));
-
-	bool is_rev_ok = (pdev->revision == 2U);
-
-	return (is_vid_ok && is_did_ok && is_rev_ok) ? &hw_atl_ops_ : NULL;
-}
+	.hw_ring_hwts_rx_fill        = hw_atl_b0_hw_ring_hwts_rx_fill,
+	.hw_ring_hwts_rx_receive     = hw_atl_b0_hw_ring_hwts_rx_receive,
+
+	.hw_get_ptp_ts           = hw_atl_b0_get_ptp_ts,
+	.hw_adj_sys_clock        = hw_atl_b0_adj_sys_clock,
+	.hw_set_sys_clock        = hw_atl_b0_set_sys_clock,
+	.hw_ts_to_sys_clock      = hw_atl_b0_ts_to_sys_clock,
+	.hw_adj_clock_freq       = hw_atl_b0_adj_clock_freq,
+	.hw_gpio_pulse           = hw_atl_b0_gpio_pulse,
+	.hw_extts_gpio_enable    = hw_atl_b0_extts_gpio_enable,
+	.hw_get_sync_ts          = hw_atl_b0_get_sync_ts,
+	.rx_extract_ts           = hw_atl_b0_rx_extract_ts,
+	.extract_hwts            = hw_atl_b0_extract_hwts,
+	.hw_set_offload          = hw_atl_b0_hw_offload_set,
+	.hw_set_loopback         = hw_atl_b0_set_loopback,
+	.hw_set_fc               = hw_atl_b0_set_fc,
+
+	.hw_get_mac_temp         = hw_atl_b0_get_mac_temp,
+	.hw_read_module_eeprom   = hw_atl_b0_read_module_eeprom,
+};
diff --git a/drivers/net/ethernet/aquantia/atlantic/hw_atl/hw_atl_b0.h b/drivers/net/ethernet/aquantia/atlantic/hw_atl/hw_atl_b0.h
index a1e1bce6c1f3..5298846dd9f7 100644
--- a/drivers/net/ethernet/aquantia/atlantic/hw_atl/hw_atl_b0.h
+++ b/drivers/net/ethernet/aquantia/atlantic/hw_atl/hw_atl_b0.h
@@ -1,10 +1,8 @@
-/*
- * aQuantia Corporation Network Driver
- * Copyright (C) 2014-2017 aQuantia Corporation. All rights reserved
+/* SPDX-License-Identifier: GPL-2.0-only */
+/* Atlantic Network Driver
  *
- * This program is free software; you can redistribute it and/or modify it
- * under the terms and conditions of the GNU General Public License,
- * version 2, as published by the Free Software Foundation.
+ * Copyright (C) 2014-2019 aQuantia Corporation
+ * Copyright (C) 2019-2020 Marvell International Ltd.
  */
 
 /* File hw_atl_b0.h: Declaration of abstract interface for Atlantic hardware
@@ -16,19 +14,62 @@
 
 #include "../aq_common.h"
 
-#ifndef PCI_VENDOR_ID_AQUANTIA
+extern const struct aq_hw_caps_s hw_atl_b0_caps_aqc100;
+extern const struct aq_hw_caps_s hw_atl_b0_caps_aqc107;
+extern const struct aq_hw_caps_s hw_atl_b0_caps_aqc108;
+extern const struct aq_hw_caps_s hw_atl_b0_caps_aqc109;
+extern const struct aq_hw_caps_s hw_atl_b0_caps_aqc111;
+extern const struct aq_hw_caps_s hw_atl_b0_caps_aqc112;
 
-#define PCI_VENDOR_ID_AQUANTIA  0x1D6A
-#define HW_ATL_DEVICE_ID_0001   0x0001
-#define HW_ATL_DEVICE_ID_D100   0xD100
-#define HW_ATL_DEVICE_ID_D107   0xD107
-#define HW_ATL_DEVICE_ID_D108   0xD108
-#define HW_ATL_DEVICE_ID_D109   0xD109
+#define hw_atl_b0_caps_aqc100s hw_atl_b0_caps_aqc100
+#define hw_atl_b0_caps_aqc107s hw_atl_b0_caps_aqc107
+#define hw_atl_b0_caps_aqc108s hw_atl_b0_caps_aqc108
+#define hw_atl_b0_caps_aqc109s hw_atl_b0_caps_aqc109
+#define hw_atl_b0_caps_aqc111s hw_atl_b0_caps_aqc111
+#define hw_atl_b0_caps_aqc112s hw_atl_b0_caps_aqc112
 
-#define HW_ATL_NIC_NAME "aQuantia AQtion 5Gbit Network Adapter"
+extern const struct aq_hw_ops hw_atl_ops_b0;
 
-#endif
+#define hw_atl_ops_b1 hw_atl_ops_b0
 
-struct aq_hw_ops *hw_atl_b0_get_ops_by_id(struct pci_dev *pdev);
+int hw_atl_b0_hw_rss_hash_set(struct aq_hw_s *self,
+			      struct aq_rss_parameters *rss_params);
+int hw_atl_b0_hw_offload_set(struct aq_hw_s *self,
+			     struct aq_nic_cfg_s *aq_nic_cfg);
+
+int hw_atl_b0_hw_ring_tx_start(struct aq_hw_s *self, struct aq_ring_s *ring);
+int hw_atl_b0_hw_ring_rx_start(struct aq_hw_s *self, struct aq_ring_s *ring);
+
+int hw_atl_b0_hw_ring_rx_init(struct aq_hw_s *self, struct aq_ring_s *aq_ring,
+			      struct aq_ring_param_s *aq_ring_param);
+int hw_atl_b0_hw_ring_rx_fill(struct aq_hw_s *self, struct aq_ring_s *ring,
+			      unsigned int sw_tail_old);
+int hw_atl_b0_hw_ring_rx_receive(struct aq_hw_s *self, struct aq_ring_s *ring);
+
+int hw_atl_b0_hw_ring_tx_init(struct aq_hw_s *self, struct aq_ring_s *aq_ring,
+			      struct aq_ring_param_s *aq_ring_param);
+int hw_atl_b0_hw_ring_tx_xmit(struct aq_hw_s *self, struct aq_ring_s *ring,
+			      unsigned int frags);
+int hw_atl_b0_hw_ring_tx_head_update(struct aq_hw_s *self,
+				     struct aq_ring_s *ring);
+
+int hw_atl_b0_hw_ring_tx_stop(struct aq_hw_s *self, struct aq_ring_s *ring);
+int hw_atl_b0_hw_ring_rx_stop(struct aq_hw_s *self, struct aq_ring_s *ring);
+
+void hw_atl_b0_hw_init_rx_rss_ctrl1(struct aq_hw_s *self);
+
+int hw_atl_b0_hw_mac_addr_set(struct aq_hw_s *self, const u8 *mac_addr);
+
+int hw_atl_b0_set_fc(struct aq_hw_s *self, u32 fc, u32 tc);
+int hw_atl_b0_set_loopback(struct aq_hw_s *self, u32 mode, bool enable);
+
+int hw_atl_b0_hw_start(struct aq_hw_s *self);
+
+int hw_atl_b0_hw_irq_enable(struct aq_hw_s *self, u64 mask);
+int hw_atl_b0_hw_irq_disable(struct aq_hw_s *self, u64 mask);
+int hw_atl_b0_hw_irq_read(struct aq_hw_s *self, u64 *mask);
+
+int hw_atl_b0_hw_packet_filter_set(struct aq_hw_s *self,
+				   unsigned int packet_filter);
 
 #endif /* HW_ATL_B0_H */
diff --git a/drivers/net/ethernet/aquantia/atlantic/hw_atl/hw_atl_b0_internal.h b/drivers/net/ethernet/aquantia/atlantic/hw_atl/hw_atl_b0_internal.h
index f3957e930340..cf460d61a45e 100644
--- a/drivers/net/ethernet/aquantia/atlantic/hw_atl/hw_atl_b0_internal.h
+++ b/drivers/net/ethernet/aquantia/atlantic/hw_atl/hw_atl_b0_internal.h
@@ -1,10 +1,7 @@
+/* SPDX-License-Identifier: GPL-2.0-only */
 /*
  * aQuantia Corporation Network Driver
- * Copyright (C) 2014-2017 aQuantia Corporation. All rights reserved
- *
- * This program is free software; you can redistribute it and/or modify it
- * under the terms and conditions of the GNU General Public License,
- * version 2, as published by the Free Software Foundation.
+ * Copyright (C) 2014-2019 aQuantia Corporation. All rights reserved
  */
 
 /* File hw_atl_b0_internal.h: Definition of Atlantic B0 chip specific
@@ -16,7 +13,7 @@
 
 #include "../aq_common.h"
 
-#define HW_ATL_B0_MTU_JUMBO (16000U)
+#define HW_ATL_B0_MTU_JUMBO  16352U
 #define HW_ATL_B0_MTU        1514U
 
 #define HW_ATL_B0_TX_RINGS 4U
@@ -67,24 +64,21 @@
 #define HW_ATL_B0_MPI_SPEED_MSK         0xFFFFU
 #define HW_ATL_B0_MPI_SPEED_SHIFT       16U
 
-#define HW_ATL_B0_RATE_10G              BIT(0)
-#define HW_ATL_B0_RATE_5G               BIT(1)
-#define HW_ATL_B0_RATE_2G5              BIT(3)
-#define HW_ATL_B0_RATE_1G               BIT(4)
-#define HW_ATL_B0_RATE_100M             BIT(5)
+#define HW_ATL_B0_TXBUF_MAX              160U
+#define HW_ATL_B0_PTP_TXBUF_SIZE           8U
 
-#define HW_ATL_B0_TXBUF_MAX  160U
-#define HW_ATL_B0_RXBUF_MAX  320U
+#define HW_ATL_B0_RXBUF_MAX              320U
+#define HW_ATL_B0_PTP_RXBUF_SIZE          16U
 
 #define HW_ATL_B0_RSS_REDIRECTION_MAX 64U
 #define HW_ATL_B0_RSS_REDIRECTION_BITS 3U
 #define HW_ATL_B0_RSS_HASHKEY_BITS 320U
 
 #define HW_ATL_B0_TCRSS_4_8  1
-#define HW_ATL_B0_TC_MAX 1U
+#define HW_ATL_B0_TC_MAX 8U
 #define HW_ATL_B0_RSS_MAX 8U
 
-#define HW_ATL_B0_LRO_RXD_MAX 2U
+#define HW_ATL_B0_LRO_RXD_MAX 16U
 #define HW_ATL_B0_RS_SLIP_ENABLED  0U
 
 /* (256k -1(max pay_len) - 54(header)) */
@@ -116,10 +110,17 @@
 #define HW_ATL_B0_RXD_NCEA0 (0x1)
 
 #define HW_ATL_B0_RXD_WB_STAT_RSSTYPE (0x0000000F)
+#define HW_ATL_B0_RXD_WB_STAT_RSSTYPE_SHIFT (0x0)
 #define HW_ATL_B0_RXD_WB_STAT_PKTTYPE (0x00000FF0)
+#define HW_ATL_B0_RXD_WB_STAT_PKTTYPE_SHIFT (0x4)
 #define HW_ATL_B0_RXD_WB_STAT_RXCTRL  (0x00180000)
+#define HW_ATL_B0_RXD_WB_STAT_RXCTRL_SHIFT (0x13)
 #define HW_ATL_B0_RXD_WB_STAT_SPLHDR  (0x00200000)
 #define HW_ATL_B0_RXD_WB_STAT_HDRLEN  (0xFFC00000)
+#define HW_ATL_B0_RXD_WB_STAT_HDRLEN_SHIFT (0x16)
+
+#define HW_ATL_B0_RXD_WB_PKTTYPE_VLAN          BIT(5)
+#define HW_ATL_B0_RXD_WB_PKTTYPE_VLAN_DOUBLE   BIT(6)
 
 #define HW_ATL_B0_RXD_WB_STAT2_DD      (0x0001)
 #define HW_ATL_B0_RXD_WB_STAT2_EOP     (0x0002)
@@ -139,70 +140,21 @@
 
 #define HW_ATL_B0_FW_VER_EXPECTED 0x01050006U
 
-/* Hardware tx descriptor */
-struct __packed hw_atl_txd_s {
-	u64 buf_addr;
-	u32 ctl;
-	u32 ctl2; /* 63..46 - payload length, 45 - ctx enable, 44 - ctx index */
-};
-
-/* Hardware tx context descriptor */
-struct __packed hw_atl_txc_s {
-	u32 rsvd;
-	u32 len;
-	u32 ctl;
-	u32 len2;
-};
-
-/* Hardware rx descriptor */
-struct __packed hw_atl_rxd_s {
-	u64 buf_addr;
-	u64 hdr_addr;
-};
-
-/* Hardware rx descriptor writeback */
-struct __packed hw_atl_rxd_wb_s {
-	u32 type;
-	u32 rss_hash;
-	u16 status;
-	u16 pkt_len;
-	u16 next_desc_ptr;
-	u16 vlan;
-};
+#define HW_ATL_INTR_MODER_MAX  0x1FF
+#define HW_ATL_INTR_MODER_MIN  0xFF
+
+#define HW_ATL_B0_MIN_RXD \
+	(ALIGN(AQ_CFG_SKB_FRAGS_MAX + 1U, AQ_HW_RXD_MULTIPLE))
+#define HW_ATL_B0_MIN_TXD \
+	(ALIGN(AQ_CFG_SKB_FRAGS_MAX + 1U, AQ_HW_TXD_MULTIPLE))
+
+#define HW_ATL_B0_MAX_RXD 8184U
+#define HW_ATL_B0_MAX_TXD 8184U
+
+#define HW_ATL_RSS_DISABLED 0x00000000U
+#define HW_ATL_RSS_ENABLED_8TCS_2INDEX_BITS 0xA2222222U
+#define HW_ATL_RSS_ENABLED_4TCS_3INDEX_BITS 0x80003333U
 
 /* HW layer capabilities */
-static struct aq_hw_caps_s hw_atl_b0_hw_caps_ = {
-	.ports = 1U,
-	.is_64_dma = true,
-	.msix_irqs = 4U,
-	.irq_mask = ~0U,
-	.vecs = HW_ATL_B0_RSS_MAX,
-	.tcs = HW_ATL_B0_TC_MAX,
-	.rxd_alignment = 1U,
-	.rxd_size = HW_ATL_B0_RXD_SIZE,
-	.rxds = 8U * 1024U,
-	.txd_alignment = 1U,
-	.txd_size = HW_ATL_B0_TXD_SIZE,
-	.txds = 8U * 1024U,
-	.txhwb_alignment = 4096U,
-	.tx_rings = HW_ATL_B0_TX_RINGS,
-	.rx_rings = HW_ATL_B0_RX_RINGS,
-	.hw_features = NETIF_F_HW_CSUM |
-			NETIF_F_RXCSUM |
-			NETIF_F_RXHASH |
-			NETIF_F_SG |
-			NETIF_F_TSO |
-			NETIF_F_LRO,
-	.hw_priv_flags = IFF_UNICAST_FLT,
-	.link_speed_msk = (HW_ATL_B0_RATE_10G |
-			HW_ATL_B0_RATE_5G |
-			HW_ATL_B0_RATE_2G5 |
-			HW_ATL_B0_RATE_1G |
-			HW_ATL_B0_RATE_100M),
-	.flow_control = true,
-	.mtu = HW_ATL_B0_MTU_JUMBO,
-	.mac_regs_count = 88,
-	.fw_ver_expected = HW_ATL_B0_FW_VER_EXPECTED,
-};
 
 #endif /* HW_ATL_B0_INTERNAL_H */
diff --git a/drivers/net/ethernet/aquantia/atlantic/hw_atl/hw_atl_llh.c b/drivers/net/ethernet/aquantia/atlantic/hw_atl/hw_atl_llh.c
index 3de651afa8c7..d07af1271d59 100644
--- a/drivers/net/ethernet/aquantia/atlantic/hw_atl/hw_atl_llh.c
+++ b/drivers/net/ethernet/aquantia/atlantic/hw_atl/hw_atl_llh.c
@@ -1,10 +1,8 @@
-/*
- * aQuantia Corporation Network Driver
- * Copyright (C) 2014-2017 aQuantia Corporation. All rights reserved
+// SPDX-License-Identifier: GPL-2.0-only
+/* Atlantic Network Driver
  *
- * This program is free software; you can redistribute it and/or modify it
- * under the terms and conditions of the GNU General Public License,
- * version 2, as published by the Free Software Foundation.
+ * Copyright (C) 2014-2019 aQuantia Corporation
+ * Copyright (C) 2019-2020 Marvell International Ltd.
  */
 
 /* File hw_atl_llh.c: Definitions of bitfield and register access functions for
@@ -15,112 +13,178 @@
 #include "hw_atl_llh_internal.h"
 #include "../aq_hw_utils.h"
 
-/* global */
-void reg_glb_cpu_sem_set(struct aq_hw_s *aq_hw, u32 glb_cpu_sem, u32 semaphore)
+void hw_atl_ts_reset_set(struct aq_hw_s *aq_hw, u32 val)
 {
-	aq_hw_write_reg(aq_hw, glb_cpu_sem_adr(semaphore), glb_cpu_sem);
+	aq_hw_write_reg_bit(aq_hw, HW_ATL_TS_RESET_ADR,
+			    HW_ATL_TS_RESET_MSK,
+			    HW_ATL_TS_RESET_SHIFT,
+			    val);
 }
 
-u32 reg_glb_cpu_sem_get(struct aq_hw_s *aq_hw, u32 semaphore)
+void hw_atl_ts_power_down_set(struct aq_hw_s *aq_hw, u32 val)
 {
-	return aq_hw_read_reg(aq_hw, glb_cpu_sem_adr(semaphore));
+	aq_hw_write_reg_bit(aq_hw, HW_ATL_TS_POWER_DOWN_ADR,
+			    HW_ATL_TS_POWER_DOWN_MSK,
+			    HW_ATL_TS_POWER_DOWN_SHIFT,
+			    val);
 }
 
-void glb_glb_reg_res_dis_set(struct aq_hw_s *aq_hw, u32 glb_reg_res_dis)
+u32 hw_atl_ts_power_down_get(struct aq_hw_s *aq_hw)
 {
-	aq_hw_write_reg_bit(aq_hw, glb_reg_res_dis_adr,
-			    glb_reg_res_dis_msk,
-			    glb_reg_res_dis_shift,
-			    glb_reg_res_dis);
+	return aq_hw_read_reg_bit(aq_hw, HW_ATL_TS_POWER_DOWN_ADR,
+				  HW_ATL_TS_POWER_DOWN_MSK,
+				  HW_ATL_TS_POWER_DOWN_SHIFT);
 }
 
-void glb_soft_res_set(struct aq_hw_s *aq_hw, u32 soft_res)
+u32 hw_atl_ts_ready_get(struct aq_hw_s *aq_hw)
 {
-	aq_hw_write_reg_bit(aq_hw, glb_soft_res_adr, glb_soft_res_msk,
-			    glb_soft_res_shift, soft_res);
+	return aq_hw_read_reg_bit(aq_hw, HW_ATL_TS_READY_ADR,
+				  HW_ATL_TS_READY_MSK,
+				  HW_ATL_TS_READY_SHIFT);
 }
 
-u32 glb_soft_res_get(struct aq_hw_s *aq_hw)
+u32 hw_atl_ts_ready_latch_high_get(struct aq_hw_s *aq_hw)
 {
-	return aq_hw_read_reg_bit(aq_hw, glb_soft_res_adr,
-				  glb_soft_res_msk,
-				  glb_soft_res_shift);
+	return aq_hw_read_reg_bit(aq_hw, HW_ATL_TS_READY_LATCH_HIGH_ADR,
+				  HW_ATL_TS_READY_LATCH_HIGH_MSK,
+				  HW_ATL_TS_READY_LATCH_HIGH_SHIFT);
 }
 
-u32 reg_rx_dma_stat_counter7get(struct aq_hw_s *aq_hw)
+u32 hw_atl_ts_data_get(struct aq_hw_s *aq_hw)
 {
-	return aq_hw_read_reg(aq_hw, rx_dma_stat_counter7_adr);
+	return aq_hw_read_reg_bit(aq_hw, HW_ATL_TS_DATA_OUT_ADR,
+				  HW_ATL_TS_DATA_OUT_MSK,
+				  HW_ATL_TS_DATA_OUT_SHIFT);
 }
 
-u32 reg_glb_mif_id_get(struct aq_hw_s *aq_hw)
+u32 hw_atl_smb0_bus_busy_get(struct aq_hw_s *aq_hw)
 {
-	return aq_hw_read_reg(aq_hw, glb_mif_id_adr);
+	return aq_hw_read_reg_bit(aq_hw, HW_ATL_SMB0_BUS_BUSY_ADR,
+				HW_ATL_SMB0_BUS_BUSY_MSK,
+				HW_ATL_SMB0_BUS_BUSY_SHIFT);
 }
 
-/* stats */
-u32 rpb_rx_dma_drop_pkt_cnt_get(struct aq_hw_s *aq_hw)
+u32 hw_atl_smb0_byte_transfer_complete_get(struct aq_hw_s *aq_hw)
+{
+	return aq_hw_read_reg_bit(aq_hw, HW_ATL_SMB0_BYTE_TRANSFER_COMPLETE_ADR,
+				HW_ATL_SMB0_BYTE_TRANSFER_COMPLETE_MSK,
+				HW_ATL_SMB0_BYTE_TRANSFER_COMPLETE_SHIFT);
+}
+
+u32 hw_atl_smb0_receive_acknowledged_get(struct aq_hw_s *aq_hw)
+{
+	return aq_hw_read_reg_bit(aq_hw, HW_ATL_SMB0_RX_ACKNOWLEDGED_ADR,
+				HW_ATL_SMB0_RX_ACKNOWLEDGED_MSK,
+				HW_ATL_SMB0_RX_ACKNOWLEDGED_SHIFT);
+}
+
+u32 hw_atl_smb0_repeated_start_detect_get(struct aq_hw_s *aq_hw)
+{
+	return aq_hw_read_reg_bit(aq_hw, HW_ATL_SMB0_REPEATED_START_DETECT_ADR,
+				HW_ATL_SMB0_REPEATED_START_DETECT_MSK,
+				HW_ATL_SMB0_REPEATED_START_DETECT_SHIFT);
+}
+
+u32 hw_atl_smb0_rx_data_get(struct aq_hw_s *aq_hw)
+{
+	return aq_hw_read_reg(aq_hw, HW_ATL_SMB0_RECEIVED_DATA_ADR);
+}
+
+void hw_atl_smb0_tx_data_set(struct aq_hw_s *aq_hw, u32 data)
+{
+	return aq_hw_write_reg(aq_hw, HW_ATL_SMB0_TRANSMITTED_DATA_ADR, data);
+}
+
+void hw_atl_smb0_provisioning2_set(struct aq_hw_s *aq_hw, u32 data)
+{
+	return aq_hw_write_reg(aq_hw, HW_ATL_SMB0_PROVISIONING2_ADR, data);
+}
+
+/* global */
+void hw_atl_reg_glb_cpu_sem_set(struct aq_hw_s *aq_hw, u32 glb_cpu_sem,
+				u32 semaphore)
+{
+	aq_hw_write_reg(aq_hw, HW_ATL_GLB_CPU_SEM_ADR(semaphore), glb_cpu_sem);
+}
+
+u32 hw_atl_reg_glb_cpu_sem_get(struct aq_hw_s *aq_hw, u32 semaphore)
+{
+	return aq_hw_read_reg(aq_hw, HW_ATL_GLB_CPU_SEM_ADR(semaphore));
+}
+
+void hw_atl_glb_glb_reg_res_dis_set(struct aq_hw_s *aq_hw, u32 glb_reg_res_dis)
 {
-	return aq_hw_read_reg(aq_hw, rpb_rx_dma_drop_pkt_cnt_adr);
+	aq_hw_write_reg_bit(aq_hw, HW_ATL_GLB_REG_RES_DIS_ADR,
+			    HW_ATL_GLB_REG_RES_DIS_MSK,
+			    HW_ATL_GLB_REG_RES_DIS_SHIFT,
+			    glb_reg_res_dis);
 }
 
-u32 stats_rx_dma_good_octet_counterlsw_get(struct aq_hw_s *aq_hw)
+void hw_atl_glb_soft_res_set(struct aq_hw_s *aq_hw, u32 soft_res)
 {
-	return aq_hw_read_reg(aq_hw, stats_rx_dma_good_octet_counterlsw__adr);
+	aq_hw_write_reg_bit(aq_hw, HW_ATL_GLB_SOFT_RES_ADR,
+			    HW_ATL_GLB_SOFT_RES_MSK,
+			    HW_ATL_GLB_SOFT_RES_SHIFT, soft_res);
 }
 
-u32 stats_rx_dma_good_pkt_counterlsw_get(struct aq_hw_s *aq_hw)
+u32 hw_atl_glb_soft_res_get(struct aq_hw_s *aq_hw)
 {
-	return aq_hw_read_reg(aq_hw, stats_rx_dma_good_pkt_counterlsw__adr);
+	return aq_hw_read_reg_bit(aq_hw, HW_ATL_GLB_SOFT_RES_ADR,
+				  HW_ATL_GLB_SOFT_RES_MSK,
+				  HW_ATL_GLB_SOFT_RES_SHIFT);
 }
 
-u32 stats_tx_dma_good_octet_counterlsw_get(struct aq_hw_s *aq_hw)
+u32 hw_atl_reg_glb_mif_id_get(struct aq_hw_s *aq_hw)
 {
-	return aq_hw_read_reg(aq_hw, stats_tx_dma_good_octet_counterlsw__adr);
+	return aq_hw_read_reg(aq_hw, HW_ATL_GLB_MIF_ID_ADR);
 }
 
-u32 stats_tx_dma_good_pkt_counterlsw_get(struct aq_hw_s *aq_hw)
+/* stats */
+u32 hw_atl_rpb_rx_dma_drop_pkt_cnt_get(struct aq_hw_s *aq_hw)
 {
-	return aq_hw_read_reg(aq_hw, stats_tx_dma_good_pkt_counterlsw__adr);
+	return aq_hw_read_reg(aq_hw, HW_ATL_RPB_RX_DMA_DROP_PKT_CNT_ADR);
 }
 
-u32 stats_rx_dma_good_octet_countermsw_get(struct aq_hw_s *aq_hw)
+u64 hw_atl_stats_rx_dma_good_octet_counter_get(struct aq_hw_s *aq_hw)
 {
-	return aq_hw_read_reg(aq_hw, stats_rx_dma_good_octet_countermsw__adr);
+	return aq_hw_read_reg64(aq_hw, HW_ATL_STATS_RX_DMA_GOOD_OCTET_COUNTERLSW);
 }
 
-u32 stats_rx_dma_good_pkt_countermsw_get(struct aq_hw_s *aq_hw)
+u64 hw_atl_stats_rx_dma_good_pkt_counter_get(struct aq_hw_s *aq_hw)
 {
-	return aq_hw_read_reg(aq_hw, stats_rx_dma_good_pkt_countermsw__adr);
+	return aq_hw_read_reg64(aq_hw, HW_ATL_STATS_RX_DMA_GOOD_PKT_COUNTERLSW);
 }
 
-u32 stats_tx_dma_good_octet_countermsw_get(struct aq_hw_s *aq_hw)
+u64 hw_atl_stats_tx_dma_good_octet_counter_get(struct aq_hw_s *aq_hw)
 {
-	return aq_hw_read_reg(aq_hw, stats_tx_dma_good_octet_countermsw__adr);
+	return aq_hw_read_reg64(aq_hw, HW_ATL_STATS_TX_DMA_GOOD_OCTET_COUNTERLSW);
 }
 
-u32 stats_tx_dma_good_pkt_countermsw_get(struct aq_hw_s *aq_hw)
+u64 hw_atl_stats_tx_dma_good_pkt_counter_get(struct aq_hw_s *aq_hw)
 {
-	return aq_hw_read_reg(aq_hw, stats_tx_dma_good_pkt_countermsw__adr);
+	return aq_hw_read_reg64(aq_hw, HW_ATL_STATS_TX_DMA_GOOD_PKT_COUNTERLSW);
 }
 
 /* interrupt */
-void itr_irq_auto_masklsw_set(struct aq_hw_s *aq_hw, u32 irq_auto_masklsw)
+void hw_atl_itr_irq_auto_masklsw_set(struct aq_hw_s *aq_hw,
+				     u32 irq_auto_masklsw)
 {
-	aq_hw_write_reg(aq_hw, itr_iamrlsw_adr, irq_auto_masklsw);
+	aq_hw_write_reg(aq_hw, HW_ATL_ITR_IAMRLSW_ADR, irq_auto_masklsw);
 }
 
-void itr_irq_map_en_rx_set(struct aq_hw_s *aq_hw, u32 irq_map_en_rx, u32 rx)
+void hw_atl_itr_irq_map_en_rx_set(struct aq_hw_s *aq_hw, u32 irq_map_en_rx,
+				  u32 rx)
 {
 /* register address for bitfield imr_rx{r}_en */
 	static u32 itr_imr_rxren_adr[32] = {
 			0x00002100U, 0x00002100U, 0x00002104U, 0x00002104U,
-			0x00002108U, 0x00002108U, 0x0000210cU, 0x0000210cU,
+			0x00002108U, 0x00002108U, 0x0000210CU, 0x0000210CU,
 			0x00002110U, 0x00002110U, 0x00002114U, 0x00002114U,
-			0x00002118U, 0x00002118U, 0x0000211cU, 0x0000211cU,
+			0x00002118U, 0x00002118U, 0x0000211CU, 0x0000211CU,
 			0x00002120U, 0x00002120U, 0x00002124U, 0x00002124U,
-			0x00002128U, 0x00002128U, 0x0000212cU, 0x0000212cU,
+			0x00002128U, 0x00002128U, 0x0000212CU, 0x0000212CU,
 			0x00002130U, 0x00002130U, 0x00002134U, 0x00002134U,
-			0x00002138U, 0x00002138U, 0x0000213cU, 0x0000213cU
+			0x00002138U, 0x00002138U, 0x0000213CU, 0x0000213CU
 		};
 
 /* bitmask for bitfield imr_rx{r}_en */
@@ -149,18 +213,19 @@ void itr_irq_map_en_rx_set(struct aq_hw_s *aq_hw, u32 irq_map_en_rx, u32 rx)
 			    irq_map_en_rx);
 }
 
-void itr_irq_map_en_tx_set(struct aq_hw_s *aq_hw, u32 irq_map_en_tx, u32 tx)
+void hw_atl_itr_irq_map_en_tx_set(struct aq_hw_s *aq_hw, u32 irq_map_en_tx,
+				  u32 tx)
 {
 /* register address for bitfield imr_tx{t}_en */
 	static u32 itr_imr_txten_adr[32] = {
 			0x00002100U, 0x00002100U, 0x00002104U, 0x00002104U,
-			0x00002108U, 0x00002108U, 0x0000210cU, 0x0000210cU,
+			0x00002108U, 0x00002108U, 0x0000210CU, 0x0000210CU,
 			0x00002110U, 0x00002110U, 0x00002114U, 0x00002114U,
-			0x00002118U, 0x00002118U, 0x0000211cU, 0x0000211cU,
+			0x00002118U, 0x00002118U, 0x0000211CU, 0x0000211CU,
 			0x00002120U, 0x00002120U, 0x00002124U, 0x00002124U,
-			0x00002128U, 0x00002128U, 0x0000212cU, 0x0000212cU,
+			0x00002128U, 0x00002128U, 0x0000212CU, 0x0000212CU,
 			0x00002130U, 0x00002130U, 0x00002134U, 0x00002134U,
-			0x00002138U, 0x00002138U, 0x0000213cU, 0x0000213cU
+			0x00002138U, 0x00002138U, 0x0000213CU, 0x0000213CU
 		};
 
 /* bitmask for bitfield imr_tx{t}_en */
@@ -189,30 +254,30 @@ void itr_irq_map_en_tx_set(struct aq_hw_s *aq_hw, u32 irq_map_en_tx, u32 tx)
 			    irq_map_en_tx);
 }
 
-void itr_irq_map_rx_set(struct aq_hw_s *aq_hw, u32 irq_map_rx, u32 rx)
+void hw_atl_itr_irq_map_rx_set(struct aq_hw_s *aq_hw, u32 irq_map_rx, u32 rx)
 {
 /* register address for bitfield imr_rx{r}[4:0] */
 	static u32 itr_imr_rxr_adr[32] = {
 			0x00002100U, 0x00002100U, 0x00002104U, 0x00002104U,
-			0x00002108U, 0x00002108U, 0x0000210cU, 0x0000210cU,
+			0x00002108U, 0x00002108U, 0x0000210CU, 0x0000210CU,
 			0x00002110U, 0x00002110U, 0x00002114U, 0x00002114U,
-			0x00002118U, 0x00002118U, 0x0000211cU, 0x0000211cU,
+			0x00002118U, 0x00002118U, 0x0000211CU, 0x0000211CU,
 			0x00002120U, 0x00002120U, 0x00002124U, 0x00002124U,
-			0x00002128U, 0x00002128U, 0x0000212cU, 0x0000212cU,
+			0x00002128U, 0x00002128U, 0x0000212CU, 0x0000212CU,
 			0x00002130U, 0x00002130U, 0x00002134U, 0x00002134U,
-			0x00002138U, 0x00002138U, 0x0000213cU, 0x0000213cU
+			0x00002138U, 0x00002138U, 0x0000213CU, 0x0000213CU
 		};
 
 /* bitmask for bitfield imr_rx{r}[4:0] */
 	static u32 itr_imr_rxr_msk[32] = {
-			0x00001f00U, 0x0000001fU, 0x00001f00U, 0x0000001fU,
-			0x00001f00U, 0x0000001fU, 0x00001f00U, 0x0000001fU,
-			0x00001f00U, 0x0000001fU, 0x00001f00U, 0x0000001fU,
-			0x00001f00U, 0x0000001fU, 0x00001f00U, 0x0000001fU,
-			0x00001f00U, 0x0000001fU, 0x00001f00U, 0x0000001fU,
-			0x00001f00U, 0x0000001fU, 0x00001f00U, 0x0000001fU,
-			0x00001f00U, 0x0000001fU, 0x00001f00U, 0x0000001fU,
-			0x00001f00U, 0x0000001fU, 0x00001f00U, 0x0000001fU
+			0x00001f00U, 0x0000001FU, 0x00001F00U, 0x0000001FU,
+			0x00001f00U, 0x0000001FU, 0x00001F00U, 0x0000001FU,
+			0x00001f00U, 0x0000001FU, 0x00001F00U, 0x0000001FU,
+			0x00001f00U, 0x0000001FU, 0x00001F00U, 0x0000001FU,
+			0x00001f00U, 0x0000001FU, 0x00001F00U, 0x0000001FU,
+			0x00001f00U, 0x0000001FU, 0x00001F00U, 0x0000001FU,
+			0x00001f00U, 0x0000001FU, 0x00001F00U, 0x0000001FU,
+			0x00001f00U, 0x0000001FU, 0x00001F00U, 0x0000001FU
 		};
 
 /* lower bit position of bitfield imr_rx{r}[4:0] */
@@ -229,30 +294,30 @@ void itr_irq_map_rx_set(struct aq_hw_s *aq_hw, u32 irq_map_rx, u32 rx)
 			    irq_map_rx);
 }
 
-void itr_irq_map_tx_set(struct aq_hw_s *aq_hw, u32 irq_map_tx, u32 tx)
+void hw_atl_itr_irq_map_tx_set(struct aq_hw_s *aq_hw, u32 irq_map_tx, u32 tx)
 {
 /* register address for bitfield imr_tx{t}[4:0] */
 	static u32 itr_imr_txt_adr[32] = {
 			0x00002100U, 0x00002100U, 0x00002104U, 0x00002104U,
-			0x00002108U, 0x00002108U, 0x0000210cU, 0x0000210cU,
+			0x00002108U, 0x00002108U, 0x0000210CU, 0x0000210CU,
 			0x00002110U, 0x00002110U, 0x00002114U, 0x00002114U,
-			0x00002118U, 0x00002118U, 0x0000211cU, 0x0000211cU,
+			0x00002118U, 0x00002118U, 0x0000211CU, 0x0000211CU,
 			0x00002120U, 0x00002120U, 0x00002124U, 0x00002124U,
-			0x00002128U, 0x00002128U, 0x0000212cU, 0x0000212cU,
+			0x00002128U, 0x00002128U, 0x0000212CU, 0x0000212CU,
 			0x00002130U, 0x00002130U, 0x00002134U, 0x00002134U,
-			0x00002138U, 0x00002138U, 0x0000213cU, 0x0000213cU
+			0x00002138U, 0x00002138U, 0x0000213CU, 0x0000213CU
 		};
 
 /* bitmask for bitfield imr_tx{t}[4:0] */
 	static u32 itr_imr_txt_msk[32] = {
-			0x1f000000U, 0x001f0000U, 0x1f000000U, 0x001f0000U,
-			0x1f000000U, 0x001f0000U, 0x1f000000U, 0x001f0000U,
-			0x1f000000U, 0x001f0000U, 0x1f000000U, 0x001f0000U,
-			0x1f000000U, 0x001f0000U, 0x1f000000U, 0x001f0000U,
-			0x1f000000U, 0x001f0000U, 0x1f000000U, 0x001f0000U,
-			0x1f000000U, 0x001f0000U, 0x1f000000U, 0x001f0000U,
-			0x1f000000U, 0x001f0000U, 0x1f000000U, 0x001f0000U,
-			0x1f000000U, 0x001f0000U, 0x1f000000U, 0x001f0000U
+			0x1f000000U, 0x001F0000U, 0x1F000000U, 0x001F0000U,
+			0x1f000000U, 0x001F0000U, 0x1F000000U, 0x001F0000U,
+			0x1f000000U, 0x001F0000U, 0x1F000000U, 0x001F0000U,
+			0x1f000000U, 0x001F0000U, 0x1F000000U, 0x001F0000U,
+			0x1f000000U, 0x001F0000U, 0x1F000000U, 0x001F0000U,
+			0x1f000000U, 0x001F0000U, 0x1F000000U, 0x001F0000U,
+			0x1f000000U, 0x001F0000U, 0x1F000000U, 0x001F0000U,
+			0x1f000000U, 0x001F0000U, 0x1F000000U, 0x001F0000U
 		};
 
 /* lower bit position of bitfield imr_tx{t}[4:0] */
@@ -269,429 +334,520 @@ void itr_irq_map_tx_set(struct aq_hw_s *aq_hw, u32 irq_map_tx, u32 tx)
 			    irq_map_tx);
 }
 
-void itr_irq_msk_clearlsw_set(struct aq_hw_s *aq_hw, u32 irq_msk_clearlsw)
+void hw_atl_itr_irq_msk_clearlsw_set(struct aq_hw_s *aq_hw,
+				     u32 irq_msk_clearlsw)
+{
+	aq_hw_write_reg(aq_hw, HW_ATL_ITR_IMCRLSW_ADR, irq_msk_clearlsw);
+}
+
+void hw_atl_itr_irq_msk_setlsw_set(struct aq_hw_s *aq_hw, u32 irq_msk_setlsw)
 {
-	aq_hw_write_reg(aq_hw, itr_imcrlsw_adr, irq_msk_clearlsw);
+	aq_hw_write_reg(aq_hw, HW_ATL_ITR_IMSRLSW_ADR, irq_msk_setlsw);
 }
 
-void itr_irq_msk_setlsw_set(struct aq_hw_s *aq_hw, u32 irq_msk_setlsw)
+void hw_atl_itr_irq_reg_res_dis_set(struct aq_hw_s *aq_hw, u32 irq_reg_res_dis)
 {
-	aq_hw_write_reg(aq_hw, itr_imsrlsw_adr, irq_msk_setlsw);
+	aq_hw_write_reg_bit(aq_hw, HW_ATL_ITR_REG_RES_DSBL_ADR,
+			    HW_ATL_ITR_REG_RES_DSBL_MSK,
+			    HW_ATL_ITR_REG_RES_DSBL_SHIFT, irq_reg_res_dis);
 }
 
-void itr_irq_reg_res_dis_set(struct aq_hw_s *aq_hw, u32 irq_reg_res_dis)
+void hw_atl_itr_irq_status_clearlsw_set(struct aq_hw_s *aq_hw,
+					u32 irq_status_clearlsw)
 {
-	aq_hw_write_reg_bit(aq_hw, itr_reg_res_dsbl_adr,
-			    itr_reg_res_dsbl_msk,
-			    itr_reg_res_dsbl_shift, irq_reg_res_dis);
+	aq_hw_write_reg(aq_hw, HW_ATL_ITR_ISCRLSW_ADR, irq_status_clearlsw);
 }
 
-void itr_irq_status_clearlsw_set(struct aq_hw_s *aq_hw,
-				 u32 irq_status_clearlsw)
+u32 hw_atl_itr_irq_statuslsw_get(struct aq_hw_s *aq_hw)
 {
-	aq_hw_write_reg(aq_hw, itr_iscrlsw_adr, irq_status_clearlsw);
+	return aq_hw_read_reg(aq_hw, HW_ATL_ITR_ISRLSW_ADR);
 }
 
-u32 itr_irq_statuslsw_get(struct aq_hw_s *aq_hw)
+u32 hw_atl_itr_res_irq_get(struct aq_hw_s *aq_hw)
 {
-	return aq_hw_read_reg(aq_hw, itr_isrlsw_adr);
+	return aq_hw_read_reg_bit(aq_hw, HW_ATL_ITR_RES_ADR, HW_ATL_ITR_RES_MSK,
+				  HW_ATL_ITR_RES_SHIFT);
 }
 
-u32 itr_res_irq_get(struct aq_hw_s *aq_hw)
+void hw_atl_itr_res_irq_set(struct aq_hw_s *aq_hw, u32 res_irq)
 {
-	return aq_hw_read_reg_bit(aq_hw, itr_res_adr, itr_res_msk,
-				  itr_res_shift);
+	aq_hw_write_reg_bit(aq_hw, HW_ATL_ITR_RES_ADR, HW_ATL_ITR_RES_MSK,
+			    HW_ATL_ITR_RES_SHIFT, res_irq);
 }
 
-void itr_res_irq_set(struct aq_hw_s *aq_hw, u32 res_irq)
+/* set RSC interrupt */
+void hw_atl_itr_rsc_en_set(struct aq_hw_s *aq_hw, u32 enable)
 {
-	aq_hw_write_reg_bit(aq_hw, itr_res_adr, itr_res_msk,
-			    itr_res_shift, res_irq);
+	aq_hw_write_reg(aq_hw, HW_ATL_ITR_RSC_EN_ADR, enable);
+}
+
+/* set RSC delay */
+void hw_atl_itr_rsc_delay_set(struct aq_hw_s *aq_hw, u32 delay)
+{
+	aq_hw_write_reg_bit(aq_hw, HW_ATL_ITR_RSC_DELAY_ADR,
+			    HW_ATL_ITR_RSC_DELAY_MSK,
+			    HW_ATL_ITR_RSC_DELAY_SHIFT,
+			    delay);
 }
 
 /* rdm */
-void rdm_cpu_id_set(struct aq_hw_s *aq_hw, u32 cpuid, u32 dca)
+void hw_atl_rdm_cpu_id_set(struct aq_hw_s *aq_hw, u32 cpuid, u32 dca)
 {
-	aq_hw_write_reg_bit(aq_hw, rdm_dcadcpuid_adr(dca),
-			    rdm_dcadcpuid_msk,
-			    rdm_dcadcpuid_shift, cpuid);
+	aq_hw_write_reg_bit(aq_hw, HW_ATL_RDM_DCADCPUID_ADR(dca),
+			    HW_ATL_RDM_DCADCPUID_MSK,
+			    HW_ATL_RDM_DCADCPUID_SHIFT, cpuid);
 }
 
-void rdm_rx_dca_en_set(struct aq_hw_s *aq_hw, u32 rx_dca_en)
+void hw_atl_rdm_rx_dca_en_set(struct aq_hw_s *aq_hw, u32 rx_dca_en)
 {
-	aq_hw_write_reg_bit(aq_hw, rdm_dca_en_adr, rdm_dca_en_msk,
-			    rdm_dca_en_shift, rx_dca_en);
+	aq_hw_write_reg_bit(aq_hw, HW_ATL_RDM_DCA_EN_ADR, HW_ATL_RDM_DCA_EN_MSK,
+			    HW_ATL_RDM_DCA_EN_SHIFT, rx_dca_en);
 }
 
-void rdm_rx_dca_mode_set(struct aq_hw_s *aq_hw, u32 rx_dca_mode)
+void hw_atl_rdm_rx_dca_mode_set(struct aq_hw_s *aq_hw, u32 rx_dca_mode)
 {
-	aq_hw_write_reg_bit(aq_hw, rdm_dca_mode_adr, rdm_dca_mode_msk,
-			    rdm_dca_mode_shift, rx_dca_mode);
+	aq_hw_write_reg_bit(aq_hw, HW_ATL_RDM_DCA_MODE_ADR,
+			    HW_ATL_RDM_DCA_MODE_MSK,
+			    HW_ATL_RDM_DCA_MODE_SHIFT, rx_dca_mode);
 }
 
-void rdm_rx_desc_data_buff_size_set(struct aq_hw_s *aq_hw,
-				    u32 rx_desc_data_buff_size, u32 descriptor)
+void hw_atl_rdm_rx_desc_data_buff_size_set(struct aq_hw_s *aq_hw,
+					   u32 rx_desc_data_buff_size,
+					   u32 descriptor)
 {
-	aq_hw_write_reg_bit(aq_hw, rdm_descddata_size_adr(descriptor),
-			    rdm_descddata_size_msk,
-			    rdm_descddata_size_shift,
+	aq_hw_write_reg_bit(aq_hw, HW_ATL_RDM_DESCDDATA_SIZE_ADR(descriptor),
+			    HW_ATL_RDM_DESCDDATA_SIZE_MSK,
+			    HW_ATL_RDM_DESCDDATA_SIZE_SHIFT,
 			    rx_desc_data_buff_size);
 }
 
-void rdm_rx_desc_dca_en_set(struct aq_hw_s *aq_hw, u32 rx_desc_dca_en, u32 dca)
+void hw_atl_rdm_rx_desc_dca_en_set(struct aq_hw_s *aq_hw, u32 rx_desc_dca_en,
+				   u32 dca)
 {
-	aq_hw_write_reg_bit(aq_hw, rdm_dcaddesc_en_adr(dca),
-			    rdm_dcaddesc_en_msk,
-			    rdm_dcaddesc_en_shift,
+	aq_hw_write_reg_bit(aq_hw, HW_ATL_RDM_DCADDESC_EN_ADR(dca),
+			    HW_ATL_RDM_DCADDESC_EN_MSK,
+			    HW_ATL_RDM_DCADDESC_EN_SHIFT,
 			    rx_desc_dca_en);
 }
 
-void rdm_rx_desc_en_set(struct aq_hw_s *aq_hw, u32 rx_desc_en, u32 descriptor)
+void hw_atl_rdm_rx_desc_en_set(struct aq_hw_s *aq_hw, u32 rx_desc_en,
+			       u32 descriptor)
 {
-	aq_hw_write_reg_bit(aq_hw, rdm_descden_adr(descriptor),
-			    rdm_descden_msk,
-			    rdm_descden_shift,
+	aq_hw_write_reg_bit(aq_hw, HW_ATL_RDM_DESCDEN_ADR(descriptor),
+			    HW_ATL_RDM_DESCDEN_MSK,
+			    HW_ATL_RDM_DESCDEN_SHIFT,
 			    rx_desc_en);
 }
 
-void rdm_rx_desc_head_buff_size_set(struct aq_hw_s *aq_hw,
-				    u32 rx_desc_head_buff_size, u32 descriptor)
+void hw_atl_rdm_rx_desc_head_buff_size_set(struct aq_hw_s *aq_hw,
+					   u32 rx_desc_head_buff_size,
+					   u32 descriptor)
 {
-	aq_hw_write_reg_bit(aq_hw, rdm_descdhdr_size_adr(descriptor),
-			    rdm_descdhdr_size_msk,
-			    rdm_descdhdr_size_shift,
+	aq_hw_write_reg_bit(aq_hw, HW_ATL_RDM_DESCDHDR_SIZE_ADR(descriptor),
+			    HW_ATL_RDM_DESCDHDR_SIZE_MSK,
+			    HW_ATL_RDM_DESCDHDR_SIZE_SHIFT,
 			    rx_desc_head_buff_size);
 }
 
-void rdm_rx_desc_head_splitting_set(struct aq_hw_s *aq_hw,
-				    u32 rx_desc_head_splitting, u32 descriptor)
+void hw_atl_rdm_rx_desc_head_splitting_set(struct aq_hw_s *aq_hw,
+					   u32 rx_desc_head_splitting,
+					   u32 descriptor)
 {
-	aq_hw_write_reg_bit(aq_hw, rdm_descdhdr_split_adr(descriptor),
-			    rdm_descdhdr_split_msk,
-			    rdm_descdhdr_split_shift,
+	aq_hw_write_reg_bit(aq_hw, HW_ATL_RDM_DESCDHDR_SPLIT_ADR(descriptor),
+			    HW_ATL_RDM_DESCDHDR_SPLIT_MSK,
+			    HW_ATL_RDM_DESCDHDR_SPLIT_SHIFT,
 			    rx_desc_head_splitting);
 }
 
-u32 rdm_rx_desc_head_ptr_get(struct aq_hw_s *aq_hw, u32 descriptor)
+u32 hw_atl_rdm_rx_desc_head_ptr_get(struct aq_hw_s *aq_hw, u32 descriptor)
 {
-	return aq_hw_read_reg_bit(aq_hw, rdm_descdhd_adr(descriptor),
-				  rdm_descdhd_msk, rdm_descdhd_shift);
+	return aq_hw_read_reg_bit(aq_hw, HW_ATL_RDM_DESCDHD_ADR(descriptor),
+				  HW_ATL_RDM_DESCDHD_MSK,
+				  HW_ATL_RDM_DESCDHD_SHIFT);
 }
 
-void rdm_rx_desc_len_set(struct aq_hw_s *aq_hw, u32 rx_desc_len, u32 descriptor)
+void hw_atl_rdm_rx_desc_len_set(struct aq_hw_s *aq_hw, u32 rx_desc_len,
+				u32 descriptor)
 {
-	aq_hw_write_reg_bit(aq_hw, rdm_descdlen_adr(descriptor),
-			    rdm_descdlen_msk, rdm_descdlen_shift,
+	aq_hw_write_reg_bit(aq_hw, HW_ATL_RDM_DESCDLEN_ADR(descriptor),
+			    HW_ATL_RDM_DESCDLEN_MSK, HW_ATL_RDM_DESCDLEN_SHIFT,
 			    rx_desc_len);
 }
 
-void rdm_rx_desc_res_set(struct aq_hw_s *aq_hw, u32 rx_desc_res, u32 descriptor)
+void hw_atl_rdm_rx_desc_res_set(struct aq_hw_s *aq_hw, u32 rx_desc_res,
+				u32 descriptor)
 {
-	aq_hw_write_reg_bit(aq_hw, rdm_descdreset_adr(descriptor),
-			    rdm_descdreset_msk, rdm_descdreset_shift,
+	aq_hw_write_reg_bit(aq_hw, HW_ATL_RDM_DESCDRESET_ADR(descriptor),
+			    HW_ATL_RDM_DESCDRESET_MSK,
+			    HW_ATL_RDM_DESCDRESET_SHIFT,
 			    rx_desc_res);
 }
 
-void rdm_rx_desc_wr_wb_irq_en_set(struct aq_hw_s *aq_hw,
-				  u32 rx_desc_wr_wb_irq_en)
+void hw_atl_rdm_rx_desc_wr_wb_irq_en_set(struct aq_hw_s *aq_hw,
+					 u32 rx_desc_wr_wb_irq_en)
 {
-	aq_hw_write_reg_bit(aq_hw, rdm_int_desc_wrb_en_adr,
-			    rdm_int_desc_wrb_en_msk,
-			    rdm_int_desc_wrb_en_shift,
+	aq_hw_write_reg_bit(aq_hw, HW_ATL_RDM_INT_DESC_WRB_EN_ADR,
+			    HW_ATL_RDM_INT_DESC_WRB_EN_MSK,
+			    HW_ATL_RDM_INT_DESC_WRB_EN_SHIFT,
 			    rx_desc_wr_wb_irq_en);
 }
 
-void rdm_rx_head_dca_en_set(struct aq_hw_s *aq_hw, u32 rx_head_dca_en, u32 dca)
+void hw_atl_rdm_rx_head_dca_en_set(struct aq_hw_s *aq_hw, u32 rx_head_dca_en,
+				   u32 dca)
 {
-	aq_hw_write_reg_bit(aq_hw, rdm_dcadhdr_en_adr(dca),
-			    rdm_dcadhdr_en_msk,
-			    rdm_dcadhdr_en_shift,
+	aq_hw_write_reg_bit(aq_hw, HW_ATL_RDM_DCADHDR_EN_ADR(dca),
+			    HW_ATL_RDM_DCADHDR_EN_MSK,
+			    HW_ATL_RDM_DCADHDR_EN_SHIFT,
 			    rx_head_dca_en);
 }
 
-void rdm_rx_pld_dca_en_set(struct aq_hw_s *aq_hw, u32 rx_pld_dca_en, u32 dca)
+void hw_atl_rdm_rx_pld_dca_en_set(struct aq_hw_s *aq_hw, u32 rx_pld_dca_en,
+				  u32 dca)
 {
-	aq_hw_write_reg_bit(aq_hw, rdm_dcadpay_en_adr(dca),
-			    rdm_dcadpay_en_msk, rdm_dcadpay_en_shift,
+	aq_hw_write_reg_bit(aq_hw, HW_ATL_RDM_DCADPAY_EN_ADR(dca),
+			    HW_ATL_RDM_DCADPAY_EN_MSK,
+			    HW_ATL_RDM_DCADPAY_EN_SHIFT,
 			    rx_pld_dca_en);
 }
 
-void rdm_rdm_intr_moder_en_set(struct aq_hw_s *aq_hw, u32 rdm_intr_moder_en)
+void hw_atl_rdm_rdm_intr_moder_en_set(struct aq_hw_s *aq_hw,
+				      u32 rdm_intr_moder_en)
 {
-	aq_hw_write_reg_bit(aq_hw, rdm_int_rim_en_adr,
-			    rdm_int_rim_en_msk,
-			    rdm_int_rim_en_shift,
+	aq_hw_write_reg_bit(aq_hw, HW_ATL_RDM_INT_RIM_EN_ADR,
+			    HW_ATL_RDM_INT_RIM_EN_MSK,
+			    HW_ATL_RDM_INT_RIM_EN_SHIFT,
 			    rdm_intr_moder_en);
 }
 
 /* reg */
-void reg_gen_irq_map_set(struct aq_hw_s *aq_hw, u32 gen_intr_map, u32 regidx)
+void hw_atl_reg_gen_irq_map_set(struct aq_hw_s *aq_hw, u32 gen_intr_map,
+				u32 regidx)
 {
-	aq_hw_write_reg(aq_hw, gen_intr_map_adr(regidx), gen_intr_map);
+	aq_hw_write_reg(aq_hw, HW_ATL_GEN_INTR_MAP_ADR(regidx), gen_intr_map);
 }
 
-u32 reg_gen_irq_status_get(struct aq_hw_s *aq_hw)
+u32 hw_atl_reg_gen_irq_status_get(struct aq_hw_s *aq_hw)
 {
-	return aq_hw_read_reg(aq_hw, gen_intr_stat_adr);
+	return aq_hw_read_reg(aq_hw, HW_ATL_GEN_INTR_STAT_ADR);
 }
 
-void reg_irq_glb_ctl_set(struct aq_hw_s *aq_hw, u32 intr_glb_ctl)
+void hw_atl_reg_irq_glb_ctl_set(struct aq_hw_s *aq_hw, u32 intr_glb_ctl)
 {
-	aq_hw_write_reg(aq_hw, intr_glb_ctl_adr, intr_glb_ctl);
+	aq_hw_write_reg(aq_hw, HW_ATL_INTR_GLB_CTL_ADR, intr_glb_ctl);
 }
 
-void reg_irq_thr_set(struct aq_hw_s *aq_hw, u32 intr_thr, u32 throttle)
+void hw_atl_reg_irq_thr_set(struct aq_hw_s *aq_hw, u32 intr_thr, u32 throttle)
 {
-	aq_hw_write_reg(aq_hw, intr_thr_adr(throttle), intr_thr);
+	aq_hw_write_reg(aq_hw, HW_ATL_INTR_THR_ADR(throttle), intr_thr);
 }
 
-void reg_rx_dma_desc_base_addresslswset(struct aq_hw_s *aq_hw,
-					u32 rx_dma_desc_base_addrlsw,
-					u32 descriptor)
+void hw_atl_reg_rx_dma_desc_base_addresslswset(struct aq_hw_s *aq_hw,
+					       u32 rx_dma_desc_base_addrlsw,
+					       u32 descriptor)
 {
-	aq_hw_write_reg(aq_hw, rx_dma_desc_base_addrlsw_adr(descriptor),
+	aq_hw_write_reg(aq_hw, HW_ATL_RX_DMA_DESC_BASE_ADDRLSW_ADR(descriptor),
 			rx_dma_desc_base_addrlsw);
 }
 
-void reg_rx_dma_desc_base_addressmswset(struct aq_hw_s *aq_hw,
-					u32 rx_dma_desc_base_addrmsw,
-					u32 descriptor)
+void hw_atl_reg_rx_dma_desc_base_addressmswset(struct aq_hw_s *aq_hw,
+					       u32 rx_dma_desc_base_addrmsw,
+					       u32 descriptor)
 {
-	aq_hw_write_reg(aq_hw, rx_dma_desc_base_addrmsw_adr(descriptor),
+	aq_hw_write_reg(aq_hw, HW_ATL_RX_DMA_DESC_BASE_ADDRMSW_ADR(descriptor),
 			rx_dma_desc_base_addrmsw);
 }
 
-u32 reg_rx_dma_desc_status_get(struct aq_hw_s *aq_hw, u32 descriptor)
+u32 hw_atl_reg_rx_dma_desc_status_get(struct aq_hw_s *aq_hw, u32 descriptor)
 {
-	return aq_hw_read_reg(aq_hw, rx_dma_desc_stat_adr(descriptor));
+	return aq_hw_read_reg(aq_hw, HW_ATL_RX_DMA_DESC_STAT_ADR(descriptor));
 }
 
-void reg_rx_dma_desc_tail_ptr_set(struct aq_hw_s *aq_hw,
-				  u32 rx_dma_desc_tail_ptr, u32 descriptor)
+void hw_atl_reg_rx_dma_desc_tail_ptr_set(struct aq_hw_s *aq_hw,
+					 u32 rx_dma_desc_tail_ptr,
+					 u32 descriptor)
 {
-	aq_hw_write_reg(aq_hw, rx_dma_desc_tail_ptr_adr(descriptor),
+	aq_hw_write_reg(aq_hw, HW_ATL_RX_DMA_DESC_TAIL_PTR_ADR(descriptor),
 			rx_dma_desc_tail_ptr);
 }
 
-void reg_rx_flr_mcst_flr_msk_set(struct aq_hw_s *aq_hw, u32 rx_flr_mcst_flr_msk)
+void hw_atl_reg_rx_flr_mcst_flr_msk_set(struct aq_hw_s *aq_hw,
+					u32 rx_flr_mcst_flr_msk)
 {
-	aq_hw_write_reg(aq_hw, rx_flr_mcst_flr_msk_adr, rx_flr_mcst_flr_msk);
+	aq_hw_write_reg(aq_hw, HW_ATL_RX_FLR_MCST_FLR_MSK_ADR,
+			rx_flr_mcst_flr_msk);
 }
 
-void reg_rx_flr_mcst_flr_set(struct aq_hw_s *aq_hw, u32 rx_flr_mcst_flr,
-			     u32 filter)
+void hw_atl_reg_rx_flr_mcst_flr_set(struct aq_hw_s *aq_hw, u32 rx_flr_mcst_flr,
+				    u32 filter)
 {
-	aq_hw_write_reg(aq_hw, rx_flr_mcst_flr_adr(filter), rx_flr_mcst_flr);
+	aq_hw_write_reg(aq_hw, HW_ATL_RX_FLR_MCST_FLR_ADR(filter),
+			rx_flr_mcst_flr);
 }
 
-void reg_rx_flr_rss_control1set(struct aq_hw_s *aq_hw, u32 rx_flr_rss_control1)
+void hw_atl_reg_rx_flr_rss_control1set(struct aq_hw_s *aq_hw,
+				       u32 rx_flr_rss_control1)
 {
-	aq_hw_write_reg(aq_hw, rx_flr_rss_control1_adr, rx_flr_rss_control1);
+	aq_hw_write_reg(aq_hw, HW_ATL_RX_FLR_RSS_CONTROL1_ADR,
+			rx_flr_rss_control1);
 }
 
-void reg_rx_flr_control2_set(struct aq_hw_s *aq_hw, u32 rx_filter_control2)
+void hw_atl_reg_rx_flr_control2_set(struct aq_hw_s *aq_hw,
+				    u32 rx_filter_control2)
 {
-	aq_hw_write_reg(aq_hw, rx_flr_control2_adr, rx_filter_control2);
+	aq_hw_write_reg(aq_hw, HW_ATL_RX_FLR_CONTROL2_ADR, rx_filter_control2);
 }
 
-void reg_rx_intr_moder_ctrl_set(struct aq_hw_s *aq_hw,
-				u32 rx_intr_moderation_ctl,
-				u32 queue)
+void hw_atl_reg_rx_intr_moder_ctrl_set(struct aq_hw_s *aq_hw,
+				       u32 rx_intr_moderation_ctl,
+				       u32 queue)
 {
-	aq_hw_write_reg(aq_hw, rx_intr_moderation_ctl_adr(queue),
+	aq_hw_write_reg(aq_hw, HW_ATL_RX_INTR_MODERATION_CTL_ADR(queue),
 			rx_intr_moderation_ctl);
 }
 
-void reg_tx_dma_debug_ctl_set(struct aq_hw_s *aq_hw, u32 tx_dma_debug_ctl)
+void hw_atl_reg_tx_dma_debug_ctl_set(struct aq_hw_s *aq_hw,
+				     u32 tx_dma_debug_ctl)
 {
-	aq_hw_write_reg(aq_hw, tx_dma_debug_ctl_adr, tx_dma_debug_ctl);
+	aq_hw_write_reg(aq_hw, HW_ATL_TX_DMA_DEBUG_CTL_ADR, tx_dma_debug_ctl);
 }
 
-void reg_tx_dma_desc_base_addresslswset(struct aq_hw_s *aq_hw,
-					u32 tx_dma_desc_base_addrlsw,
-					u32 descriptor)
+void hw_atl_reg_tx_dma_desc_base_addresslswset(struct aq_hw_s *aq_hw,
+					       u32 tx_dma_desc_base_addrlsw,
+					       u32 descriptor)
 {
-	aq_hw_write_reg(aq_hw, tx_dma_desc_base_addrlsw_adr(descriptor),
+	aq_hw_write_reg(aq_hw, HW_ATL_TX_DMA_DESC_BASE_ADDRLSW_ADR(descriptor),
 			tx_dma_desc_base_addrlsw);
 }
 
-void reg_tx_dma_desc_base_addressmswset(struct aq_hw_s *aq_hw,
-					u32 tx_dma_desc_base_addrmsw,
-					u32 descriptor)
+void hw_atl_reg_tx_dma_desc_base_addressmswset(struct aq_hw_s *aq_hw,
+					       u32 tx_dma_desc_base_addrmsw,
+					       u32 descriptor)
 {
-	aq_hw_write_reg(aq_hw, tx_dma_desc_base_addrmsw_adr(descriptor),
+	aq_hw_write_reg(aq_hw, HW_ATL_TX_DMA_DESC_BASE_ADDRMSW_ADR(descriptor),
 			tx_dma_desc_base_addrmsw);
 }
 
-void reg_tx_dma_desc_tail_ptr_set(struct aq_hw_s *aq_hw,
-				  u32 tx_dma_desc_tail_ptr, u32 descriptor)
+void hw_atl_reg_tx_dma_desc_tail_ptr_set(struct aq_hw_s *aq_hw,
+					 u32 tx_dma_desc_tail_ptr,
+					 u32 descriptor)
 {
-	aq_hw_write_reg(aq_hw, tx_dma_desc_tail_ptr_adr(descriptor),
+	aq_hw_write_reg(aq_hw, HW_ATL_TX_DMA_DESC_TAIL_PTR_ADR(descriptor),
 			tx_dma_desc_tail_ptr);
 }
 
-void reg_tx_intr_moder_ctrl_set(struct aq_hw_s *aq_hw,
-				u32 tx_intr_moderation_ctl,
-				u32 queue)
+void hw_atl_reg_tx_intr_moder_ctrl_set(struct aq_hw_s *aq_hw,
+				       u32 tx_intr_moderation_ctl,
+				       u32 queue)
 {
-	aq_hw_write_reg(aq_hw, tx_intr_moderation_ctl_adr(queue),
+	aq_hw_write_reg(aq_hw, HW_ATL_TX_INTR_MODERATION_CTL_ADR(queue),
 			tx_intr_moderation_ctl);
 }
 
 /* RPB: rx packet buffer */
-void rpb_dma_sys_lbk_set(struct aq_hw_s *aq_hw, u32 dma_sys_lbk)
+void hw_atl_rpb_dma_sys_lbk_set(struct aq_hw_s *aq_hw, u32 dma_sys_lbk)
+{
+	aq_hw_write_reg_bit(aq_hw, HW_ATL_RPB_DMA_SYS_LBK_ADR,
+			    HW_ATL_RPB_DMA_SYS_LBK_MSK,
+			    HW_ATL_RPB_DMA_SYS_LBK_SHIFT, dma_sys_lbk);
+}
+
+void hw_atl_rpb_dma_net_lbk_set(struct aq_hw_s *aq_hw, u32 dma_net_lbk)
 {
-	aq_hw_write_reg_bit(aq_hw, rpb_dma_sys_lbk_adr,
-			    rpb_dma_sys_lbk_msk,
-			    rpb_dma_sys_lbk_shift, dma_sys_lbk);
+	aq_hw_write_reg_bit(aq_hw, HW_ATL_RPB_DMA_NET_LBK_ADR,
+			    HW_ATL_RPB_DMA_NET_LBK_MSK,
+			    HW_ATL_RPB_DMA_NET_LBK_SHIFT, dma_net_lbk);
 }
 
-void rpb_rpf_rx_traf_class_mode_set(struct aq_hw_s *aq_hw,
-				    u32 rx_traf_class_mode)
+void hw_atl_rpb_rpf_rx_traf_class_mode_set(struct aq_hw_s *aq_hw,
+					   u32 rx_traf_class_mode)
 {
-	aq_hw_write_reg_bit(aq_hw, rpb_rpf_rx_tc_mode_adr,
-			    rpb_rpf_rx_tc_mode_msk,
-			    rpb_rpf_rx_tc_mode_shift,
+	aq_hw_write_reg_bit(aq_hw, HW_ATL_RPB_RPF_RX_TC_MODE_ADR,
+			    HW_ATL_RPB_RPF_RX_TC_MODE_MSK,
+			    HW_ATL_RPB_RPF_RX_TC_MODE_SHIFT,
 			    rx_traf_class_mode);
 }
 
-void rpb_rx_buff_en_set(struct aq_hw_s *aq_hw, u32 rx_buff_en)
+u32 hw_atl_rpb_rpf_rx_traf_class_mode_get(struct aq_hw_s *aq_hw)
 {
-	aq_hw_write_reg_bit(aq_hw, rpb_rx_buf_en_adr, rpb_rx_buf_en_msk,
-			    rpb_rx_buf_en_shift, rx_buff_en);
+	return aq_hw_read_reg_bit(aq_hw, HW_ATL_RPB_RPF_RX_TC_MODE_ADR,
+			HW_ATL_RPB_RPF_RX_TC_MODE_MSK,
+			HW_ATL_RPB_RPF_RX_TC_MODE_SHIFT);
 }
 
-void rpb_rx_buff_hi_threshold_per_tc_set(struct aq_hw_s *aq_hw,
-					 u32 rx_buff_hi_threshold_per_tc,
-					 u32 buffer)
+void hw_atl_rpb_rx_buff_en_set(struct aq_hw_s *aq_hw, u32 rx_buff_en)
+{
+	aq_hw_write_reg_bit(aq_hw, HW_ATL_RPB_RX_BUF_EN_ADR,
+			    HW_ATL_RPB_RX_BUF_EN_MSK,
+			    HW_ATL_RPB_RX_BUF_EN_SHIFT, rx_buff_en);
+}
+
+void hw_atl_rpb_rx_buff_hi_threshold_per_tc_set(struct aq_hw_s *aq_hw,
+						u32 rx_buff_hi_threshold_per_tc,
+						u32 buffer)
 {
-	aq_hw_write_reg_bit(aq_hw, rpb_rxbhi_thresh_adr(buffer),
-			    rpb_rxbhi_thresh_msk, rpb_rxbhi_thresh_shift,
+	aq_hw_write_reg_bit(aq_hw, HW_ATL_RPB_RXBHI_THRESH_ADR(buffer),
+			    HW_ATL_RPB_RXBHI_THRESH_MSK,
+			    HW_ATL_RPB_RXBHI_THRESH_SHIFT,
 			    rx_buff_hi_threshold_per_tc);
 }
 
-void rpb_rx_buff_lo_threshold_per_tc_set(struct aq_hw_s *aq_hw,
-					 u32 rx_buff_lo_threshold_per_tc,
-					 u32 buffer)
+void hw_atl_rpb_rx_buff_lo_threshold_per_tc_set(struct aq_hw_s *aq_hw,
+						u32 rx_buff_lo_threshold_per_tc,
+						u32 buffer)
 {
-	aq_hw_write_reg_bit(aq_hw, rpb_rxblo_thresh_adr(buffer),
-			    rpb_rxblo_thresh_msk,
-			    rpb_rxblo_thresh_shift,
+	aq_hw_write_reg_bit(aq_hw, HW_ATL_RPB_RXBLO_THRESH_ADR(buffer),
+			    HW_ATL_RPB_RXBLO_THRESH_MSK,
+			    HW_ATL_RPB_RXBLO_THRESH_SHIFT,
 			    rx_buff_lo_threshold_per_tc);
 }
 
-void rpb_rx_flow_ctl_mode_set(struct aq_hw_s *aq_hw, u32 rx_flow_ctl_mode)
+void hw_atl_rpb_rx_flow_ctl_mode_set(struct aq_hw_s *aq_hw, u32 rx_flow_ctl_mode)
+{
+	aq_hw_write_reg_bit(aq_hw, HW_ATL_RPB_RX_FC_MODE_ADR,
+			    HW_ATL_RPB_RX_FC_MODE_MSK,
+			    HW_ATL_RPB_RX_FC_MODE_SHIFT, rx_flow_ctl_mode);
+}
+
+void hw_atl_rdm_rx_dma_desc_cache_init_tgl(struct aq_hw_s *aq_hw)
+{
+	u32 val;
+
+	val = aq_hw_read_reg_bit(aq_hw, HW_ATL_RDM_RX_DMA_DESC_CACHE_INIT_ADR,
+				 HW_ATL_RDM_RX_DMA_DESC_CACHE_INIT_MSK,
+				 HW_ATL_RDM_RX_DMA_DESC_CACHE_INIT_SHIFT);
+
+	aq_hw_write_reg_bit(aq_hw, HW_ATL_RDM_RX_DMA_DESC_CACHE_INIT_ADR,
+			    HW_ATL_RDM_RX_DMA_DESC_CACHE_INIT_MSK,
+			    HW_ATL_RDM_RX_DMA_DESC_CACHE_INIT_SHIFT,
+			    val ^ 1);
+}
+
+u32 hw_atl_rdm_rx_dma_desc_cache_init_done_get(struct aq_hw_s *aq_hw)
 {
-	aq_hw_write_reg_bit(aq_hw, rpb_rx_fc_mode_adr,
-			    rpb_rx_fc_mode_msk,
-			    rpb_rx_fc_mode_shift, rx_flow_ctl_mode);
+	return aq_hw_read_reg_bit(aq_hw, RDM_RX_DMA_DESC_CACHE_INIT_DONE_ADR,
+				  RDM_RX_DMA_DESC_CACHE_INIT_DONE_MSK,
+				  RDM_RX_DMA_DESC_CACHE_INIT_DONE_SHIFT);
 }
 
-void rpb_rx_pkt_buff_size_per_tc_set(struct aq_hw_s *aq_hw,
-				     u32 rx_pkt_buff_size_per_tc, u32 buffer)
+void hw_atl_rpb_rx_pkt_buff_size_per_tc_set(struct aq_hw_s *aq_hw,
+					    u32 rx_pkt_buff_size_per_tc, u32 buffer)
 {
-	aq_hw_write_reg_bit(aq_hw, rpb_rxbbuf_size_adr(buffer),
-			    rpb_rxbbuf_size_msk, rpb_rxbbuf_size_shift,
+	aq_hw_write_reg_bit(aq_hw, HW_ATL_RPB_RXBBUF_SIZE_ADR(buffer),
+			    HW_ATL_RPB_RXBBUF_SIZE_MSK,
+			    HW_ATL_RPB_RXBBUF_SIZE_SHIFT,
 			    rx_pkt_buff_size_per_tc);
 }
 
-void rpb_rx_xoff_en_per_tc_set(struct aq_hw_s *aq_hw, u32 rx_xoff_en_per_tc,
-			       u32 buffer)
+void hw_atl_rpb_rx_xoff_en_per_tc_set(struct aq_hw_s *aq_hw,
+				      u32 rx_xoff_en_per_tc, u32 buffer)
 {
-	aq_hw_write_reg_bit(aq_hw, rpb_rxbxoff_en_adr(buffer),
-			    rpb_rxbxoff_en_msk, rpb_rxbxoff_en_shift,
+	aq_hw_write_reg_bit(aq_hw, HW_ATL_RPB_RXBXOFF_EN_ADR(buffer),
+			    HW_ATL_RPB_RXBXOFF_EN_MSK,
+			    HW_ATL_RPB_RXBXOFF_EN_SHIFT,
 			    rx_xoff_en_per_tc);
 }
 
 /* rpf */
 
-void rpfl2broadcast_count_threshold_set(struct aq_hw_s *aq_hw,
-					u32 l2broadcast_count_threshold)
+void hw_atl_rpfl2broadcast_count_threshold_set(struct aq_hw_s *aq_hw,
+					       u32 l2broadcast_count_threshold)
 {
-	aq_hw_write_reg_bit(aq_hw, rpfl2bc_thresh_adr,
-			    rpfl2bc_thresh_msk,
-			    rpfl2bc_thresh_shift,
+	aq_hw_write_reg_bit(aq_hw, HW_ATL_RPFL2BC_THRESH_ADR,
+			    HW_ATL_RPFL2BC_THRESH_MSK,
+			    HW_ATL_RPFL2BC_THRESH_SHIFT,
 			    l2broadcast_count_threshold);
 }
 
-void rpfl2broadcast_en_set(struct aq_hw_s *aq_hw, u32 l2broadcast_en)
+void hw_atl_rpfl2broadcast_en_set(struct aq_hw_s *aq_hw, u32 l2broadcast_en)
 {
-	aq_hw_write_reg_bit(aq_hw, rpfl2bc_en_adr, rpfl2bc_en_msk,
-			    rpfl2bc_en_shift, l2broadcast_en);
+	aq_hw_write_reg_bit(aq_hw, HW_ATL_RPFL2BC_EN_ADR, HW_ATL_RPFL2BC_EN_MSK,
+			    HW_ATL_RPFL2BC_EN_SHIFT, l2broadcast_en);
 }
 
-void rpfl2broadcast_flr_act_set(struct aq_hw_s *aq_hw, u32 l2broadcast_flr_act)
+void hw_atl_rpfl2broadcast_flr_act_set(struct aq_hw_s *aq_hw,
+				       u32 l2broadcast_flr_act)
 {
-	aq_hw_write_reg_bit(aq_hw, rpfl2bc_act_adr, rpfl2bc_act_msk,
-			    rpfl2bc_act_shift, l2broadcast_flr_act);
+	aq_hw_write_reg_bit(aq_hw, HW_ATL_RPFL2BC_ACT_ADR,
+			    HW_ATL_RPFL2BC_ACT_MSK,
+			    HW_ATL_RPFL2BC_ACT_SHIFT, l2broadcast_flr_act);
 }
 
-void rpfl2multicast_flr_en_set(struct aq_hw_s *aq_hw, u32 l2multicast_flr_en,
-			       u32 filter)
+void hw_atl_rpfl2multicast_flr_en_set(struct aq_hw_s *aq_hw,
+				      u32 l2multicast_flr_en,
+				      u32 filter)
+{
+	aq_hw_write_reg_bit(aq_hw, HW_ATL_RPFL2MC_ENF_ADR(filter),
+			    HW_ATL_RPFL2MC_ENF_MSK,
+			    HW_ATL_RPFL2MC_ENF_SHIFT, l2multicast_flr_en);
+}
+
+u32 hw_atl_rpfl2promiscuous_mode_en_get(struct aq_hw_s *aq_hw)
 {
-	aq_hw_write_reg_bit(aq_hw, rpfl2mc_enf_adr(filter),
-			    rpfl2mc_enf_msk,
-			    rpfl2mc_enf_shift, l2multicast_flr_en);
+	return aq_hw_read_reg_bit(aq_hw, HW_ATL_RPFL2PROMIS_MODE_ADR,
+				  HW_ATL_RPFL2PROMIS_MODE_MSK,
+				  HW_ATL_RPFL2PROMIS_MODE_SHIFT);
 }
 
-void rpfl2promiscuous_mode_en_set(struct aq_hw_s *aq_hw,
-				  u32 l2promiscuous_mode_en)
+void hw_atl_rpfl2promiscuous_mode_en_set(struct aq_hw_s *aq_hw,
+					 u32 l2promiscuous_mode_en)
 {
-	aq_hw_write_reg_bit(aq_hw, rpfl2promis_mode_adr,
-			    rpfl2promis_mode_msk,
-			    rpfl2promis_mode_shift,
+	aq_hw_write_reg_bit(aq_hw, HW_ATL_RPFL2PROMIS_MODE_ADR,
+			    HW_ATL_RPFL2PROMIS_MODE_MSK,
+			    HW_ATL_RPFL2PROMIS_MODE_SHIFT,
 			    l2promiscuous_mode_en);
 }
 
-void rpfl2unicast_flr_act_set(struct aq_hw_s *aq_hw, u32 l2unicast_flr_act,
-			      u32 filter)
+void hw_atl_rpfl2unicast_flr_act_set(struct aq_hw_s *aq_hw,
+				     u32 l2unicast_flr_act,
+				     u32 filter)
 {
-	aq_hw_write_reg_bit(aq_hw, rpfl2uc_actf_adr(filter),
-			    rpfl2uc_actf_msk, rpfl2uc_actf_shift,
+	aq_hw_write_reg_bit(aq_hw, HW_ATL_RPFL2UC_ACTF_ADR(filter),
+			    HW_ATL_RPFL2UC_ACTF_MSK, HW_ATL_RPFL2UC_ACTF_SHIFT,
 			    l2unicast_flr_act);
 }
 
-void rpfl2_uc_flr_en_set(struct aq_hw_s *aq_hw, u32 l2unicast_flr_en,
-			 u32 filter)
+void hw_atl_rpfl2_uc_flr_en_set(struct aq_hw_s *aq_hw, u32 l2unicast_flr_en,
+				u32 filter)
 {
-	aq_hw_write_reg_bit(aq_hw, rpfl2uc_enf_adr(filter),
-			    rpfl2uc_enf_msk,
-			    rpfl2uc_enf_shift, l2unicast_flr_en);
+	aq_hw_write_reg_bit(aq_hw, HW_ATL_RPFL2UC_ENF_ADR(filter),
+			    HW_ATL_RPFL2UC_ENF_MSK,
+			    HW_ATL_RPFL2UC_ENF_SHIFT, l2unicast_flr_en);
 }
 
-void rpfl2unicast_dest_addresslsw_set(struct aq_hw_s *aq_hw,
-				      u32 l2unicast_dest_addresslsw,
-				      u32 filter)
+void hw_atl_rpfl2unicast_dest_addresslsw_set(struct aq_hw_s *aq_hw,
+					     u32 l2unicast_dest_addresslsw,
+					     u32 filter)
 {
-	aq_hw_write_reg(aq_hw, rpfl2uc_daflsw_adr(filter),
+	aq_hw_write_reg(aq_hw, HW_ATL_RPFL2UC_DAFLSW_ADR(filter),
 			l2unicast_dest_addresslsw);
 }
 
-void rpfl2unicast_dest_addressmsw_set(struct aq_hw_s *aq_hw,
-				      u32 l2unicast_dest_addressmsw,
-				      u32 filter)
+void hw_atl_rpfl2unicast_dest_addressmsw_set(struct aq_hw_s *aq_hw,
+					     u32 l2unicast_dest_addressmsw,
+					     u32 filter)
 {
-	aq_hw_write_reg_bit(aq_hw, rpfl2uc_dafmsw_adr(filter),
-			    rpfl2uc_dafmsw_msk, rpfl2uc_dafmsw_shift,
+	aq_hw_write_reg_bit(aq_hw, HW_ATL_RPFL2UC_DAFMSW_ADR(filter),
+			    HW_ATL_RPFL2UC_DAFMSW_MSK,
+			    HW_ATL_RPFL2UC_DAFMSW_SHIFT,
 			    l2unicast_dest_addressmsw);
 }
 
-void rpfl2_accept_all_mc_packets_set(struct aq_hw_s *aq_hw,
-				     u32 l2_accept_all_mc_packets)
+void hw_atl_rpfl2_accept_all_mc_packets_set(struct aq_hw_s *aq_hw,
+					    u32 l2_accept_all_mc_packets)
 {
-	aq_hw_write_reg_bit(aq_hw, rpfl2mc_accept_all_adr,
-			    rpfl2mc_accept_all_msk,
-			    rpfl2mc_accept_all_shift,
+	aq_hw_write_reg_bit(aq_hw, HW_ATL_RPFL2MC_ACCEPT_ALL_ADR,
+			    HW_ATL_RPFL2MC_ACCEPT_ALL_MSK,
+			    HW_ATL_RPFL2MC_ACCEPT_ALL_SHIFT,
 			    l2_accept_all_mc_packets);
 }
 
-void rpf_rpb_user_priority_tc_map_set(struct aq_hw_s *aq_hw,
-				      u32 user_priority_tc_map, u32 tc)
+void hw_atl_rpf_rpb_user_priority_tc_map_set(struct aq_hw_s *aq_hw,
+					     u32 user_priority, u32 tc)
 {
 /* register address for bitfield rx_tc_up{t}[2:0] */
 	static u32 rpf_rpb_rx_tc_upt_adr[8] = {
-			0x000054c4U, 0x000054c4U, 0x000054c4U, 0x000054c4U,
-			0x000054c4U, 0x000054c4U, 0x000054c4U, 0x000054c4U
+			0x000054c4U, 0x000054C4U, 0x000054C4U, 0x000054C4U,
+			0x000054c4U, 0x000054C4U, 0x000054C4U, 0x000054C4U
 		};
 
 /* bitmask for bitfield rx_tc_up{t}[2:0] */
@@ -705,279 +861,350 @@ void rpf_rpb_user_priority_tc_map_set(struct aq_hw_s *aq_hw,
 			0U, 4U, 8U, 12U, 16U, 20U, 24U, 28U
 		};
 
-	aq_hw_write_reg_bit(aq_hw, rpf_rpb_rx_tc_upt_adr[tc],
-			    rpf_rpb_rx_tc_upt_msk[tc],
-			    rpf_rpb_rx_tc_upt_shft[tc],
-			    user_priority_tc_map);
+	aq_hw_write_reg_bit(aq_hw, rpf_rpb_rx_tc_upt_adr[user_priority],
+			    rpf_rpb_rx_tc_upt_msk[user_priority],
+			    rpf_rpb_rx_tc_upt_shft[user_priority], tc);
 }
 
-void rpf_rss_key_addr_set(struct aq_hw_s *aq_hw, u32 rss_key_addr)
+void hw_atl_rpf_rss_key_addr_set(struct aq_hw_s *aq_hw, u32 rss_key_addr)
 {
-	aq_hw_write_reg_bit(aq_hw, rpf_rss_key_addr_adr,
-			    rpf_rss_key_addr_msk,
-			    rpf_rss_key_addr_shift,
+	aq_hw_write_reg_bit(aq_hw, HW_ATL_RPF_RSS_KEY_ADDR_ADR,
+			    HW_ATL_RPF_RSS_KEY_ADDR_MSK,
+			    HW_ATL_RPF_RSS_KEY_ADDR_SHIFT,
 			    rss_key_addr);
 }
 
-void rpf_rss_key_wr_data_set(struct aq_hw_s *aq_hw, u32 rss_key_wr_data)
+void hw_atl_rpf_rss_key_wr_data_set(struct aq_hw_s *aq_hw, u32 rss_key_wr_data)
 {
-	aq_hw_write_reg(aq_hw, rpf_rss_key_wr_data_adr,
+	aq_hw_write_reg(aq_hw, HW_ATL_RPF_RSS_KEY_WR_DATA_ADR,
 			rss_key_wr_data);
 }
 
-u32 rpf_rss_key_wr_en_get(struct aq_hw_s *aq_hw)
+u32 hw_atl_rpf_rss_key_wr_en_get(struct aq_hw_s *aq_hw)
 {
-	return aq_hw_read_reg_bit(aq_hw, rpf_rss_key_wr_eni_adr,
-				  rpf_rss_key_wr_eni_msk,
-				  rpf_rss_key_wr_eni_shift);
+	return aq_hw_read_reg_bit(aq_hw, HW_ATL_RPF_RSS_KEY_WR_ENI_ADR,
+				  HW_ATL_RPF_RSS_KEY_WR_ENI_MSK,
+				  HW_ATL_RPF_RSS_KEY_WR_ENI_SHIFT);
 }
 
-void rpf_rss_key_wr_en_set(struct aq_hw_s *aq_hw, u32 rss_key_wr_en)
+void hw_atl_rpf_rss_key_wr_en_set(struct aq_hw_s *aq_hw, u32 rss_key_wr_en)
 {
-	aq_hw_write_reg_bit(aq_hw, rpf_rss_key_wr_eni_adr,
-			    rpf_rss_key_wr_eni_msk,
-			    rpf_rss_key_wr_eni_shift,
+	aq_hw_write_reg_bit(aq_hw, HW_ATL_RPF_RSS_KEY_WR_ENI_ADR,
+			    HW_ATL_RPF_RSS_KEY_WR_ENI_MSK,
+			    HW_ATL_RPF_RSS_KEY_WR_ENI_SHIFT,
 			    rss_key_wr_en);
 }
 
-void rpf_rss_redir_tbl_addr_set(struct aq_hw_s *aq_hw, u32 rss_redir_tbl_addr)
+void hw_atl_rpf_rss_redir_tbl_addr_set(struct aq_hw_s *aq_hw,
+				       u32 rss_redir_tbl_addr)
 {
-	aq_hw_write_reg_bit(aq_hw, rpf_rss_redir_addr_adr,
-			    rpf_rss_redir_addr_msk,
-			    rpf_rss_redir_addr_shift, rss_redir_tbl_addr);
+	aq_hw_write_reg_bit(aq_hw, HW_ATL_RPF_RSS_REDIR_ADDR_ADR,
+			    HW_ATL_RPF_RSS_REDIR_ADDR_MSK,
+			    HW_ATL_RPF_RSS_REDIR_ADDR_SHIFT,
+			    rss_redir_tbl_addr);
 }
 
-void rpf_rss_redir_tbl_wr_data_set(struct aq_hw_s *aq_hw,
-				   u32 rss_redir_tbl_wr_data)
+void hw_atl_rpf_rss_redir_tbl_wr_data_set(struct aq_hw_s *aq_hw,
+					  u32 rss_redir_tbl_wr_data)
 {
-	aq_hw_write_reg_bit(aq_hw, rpf_rss_redir_wr_data_adr,
-			    rpf_rss_redir_wr_data_msk,
-			    rpf_rss_redir_wr_data_shift,
+	aq_hw_write_reg_bit(aq_hw, HW_ATL_RPF_RSS_REDIR_WR_DATA_ADR,
+			    HW_ATL_RPF_RSS_REDIR_WR_DATA_MSK,
+			    HW_ATL_RPF_RSS_REDIR_WR_DATA_SHIFT,
 			    rss_redir_tbl_wr_data);
 }
 
-u32 rpf_rss_redir_wr_en_get(struct aq_hw_s *aq_hw)
+u32 hw_atl_rpf_rss_redir_wr_en_get(struct aq_hw_s *aq_hw)
 {
-	return aq_hw_read_reg_bit(aq_hw, rpf_rss_redir_wr_eni_adr,
-				  rpf_rss_redir_wr_eni_msk,
-				  rpf_rss_redir_wr_eni_shift);
+	return aq_hw_read_reg_bit(aq_hw, HW_ATL_RPF_RSS_REDIR_WR_ENI_ADR,
+				  HW_ATL_RPF_RSS_REDIR_WR_ENI_MSK,
+				  HW_ATL_RPF_RSS_REDIR_WR_ENI_SHIFT);
 }
 
-void rpf_rss_redir_wr_en_set(struct aq_hw_s *aq_hw, u32 rss_redir_wr_en)
+void hw_atl_rpf_rss_redir_wr_en_set(struct aq_hw_s *aq_hw, u32 rss_redir_wr_en)
 {
-	aq_hw_write_reg_bit(aq_hw, rpf_rss_redir_wr_eni_adr,
-			    rpf_rss_redir_wr_eni_msk,
-			    rpf_rss_redir_wr_eni_shift, rss_redir_wr_en);
+	aq_hw_write_reg_bit(aq_hw, HW_ATL_RPF_RSS_REDIR_WR_ENI_ADR,
+			    HW_ATL_RPF_RSS_REDIR_WR_ENI_MSK,
+			    HW_ATL_RPF_RSS_REDIR_WR_ENI_SHIFT, rss_redir_wr_en);
 }
 
-void rpf_tpo_to_rpf_sys_lbk_set(struct aq_hw_s *aq_hw, u32 tpo_to_rpf_sys_lbk)
+void hw_atl_rpf_tpo_to_rpf_sys_lbk_set(struct aq_hw_s *aq_hw,
+				       u32 tpo_to_rpf_sys_lbk)
 {
-	aq_hw_write_reg_bit(aq_hw, rpf_tpo_rpf_sys_lbk_adr,
-			    rpf_tpo_rpf_sys_lbk_msk,
-			    rpf_tpo_rpf_sys_lbk_shift,
+	aq_hw_write_reg_bit(aq_hw, HW_ATL_RPF_TPO_RPF_SYS_LBK_ADR,
+			    HW_ATL_RPF_TPO_RPF_SYS_LBK_MSK,
+			    HW_ATL_RPF_TPO_RPF_SYS_LBK_SHIFT,
 			    tpo_to_rpf_sys_lbk);
 }
 
-void rpf_vlan_inner_etht_set(struct aq_hw_s *aq_hw, u32 vlan_inner_etht)
+void hw_atl_rpf_vlan_inner_etht_set(struct aq_hw_s *aq_hw, u32 vlan_inner_etht)
 {
-	aq_hw_write_reg_bit(aq_hw, rpf_vl_inner_tpid_adr,
-			    rpf_vl_inner_tpid_msk,
-			    rpf_vl_inner_tpid_shift,
+	aq_hw_write_reg_bit(aq_hw, HW_ATL_RPF_VL_INNER_TPID_ADR,
+			    HW_ATL_RPF_VL_INNER_TPID_MSK,
+			    HW_ATL_RPF_VL_INNER_TPID_SHIFT,
 			    vlan_inner_etht);
 }
 
-void rpf_vlan_outer_etht_set(struct aq_hw_s *aq_hw, u32 vlan_outer_etht)
+void hw_atl_rpf_vlan_outer_etht_set(struct aq_hw_s *aq_hw, u32 vlan_outer_etht)
 {
-	aq_hw_write_reg_bit(aq_hw, rpf_vl_outer_tpid_adr,
-			    rpf_vl_outer_tpid_msk,
-			    rpf_vl_outer_tpid_shift,
+	aq_hw_write_reg_bit(aq_hw, HW_ATL_RPF_VL_OUTER_TPID_ADR,
+			    HW_ATL_RPF_VL_OUTER_TPID_MSK,
+			    HW_ATL_RPF_VL_OUTER_TPID_SHIFT,
 			    vlan_outer_etht);
 }
 
-void rpf_vlan_prom_mode_en_set(struct aq_hw_s *aq_hw, u32 vlan_prom_mode_en)
+void hw_atl_rpf_vlan_prom_mode_en_set(struct aq_hw_s *aq_hw,
+				      u32 vlan_prom_mode_en)
 {
-	aq_hw_write_reg_bit(aq_hw, rpf_vl_promis_mode_adr,
-			    rpf_vl_promis_mode_msk,
-			    rpf_vl_promis_mode_shift,
+	aq_hw_write_reg_bit(aq_hw, HW_ATL_RPF_VL_PROMIS_MODE_ADR,
+			    HW_ATL_RPF_VL_PROMIS_MODE_MSK,
+			    HW_ATL_RPF_VL_PROMIS_MODE_SHIFT,
 			    vlan_prom_mode_en);
 }
 
-void rpf_vlan_accept_untagged_packets_set(struct aq_hw_s *aq_hw,
-					  u32 vlan_accept_untagged_packets)
+u32 hw_atl_rpf_vlan_prom_mode_en_get(struct aq_hw_s *aq_hw)
 {
-	aq_hw_write_reg_bit(aq_hw, rpf_vl_accept_untagged_mode_adr,
-			    rpf_vl_accept_untagged_mode_msk,
-			    rpf_vl_accept_untagged_mode_shift,
-			    vlan_accept_untagged_packets);
+	return aq_hw_read_reg_bit(aq_hw, HW_ATL_RPF_VL_PROMIS_MODE_ADR,
+				  HW_ATL_RPF_VL_PROMIS_MODE_MSK,
+				  HW_ATL_RPF_VL_PROMIS_MODE_SHIFT);
 }
 
-void rpf_vlan_untagged_act_set(struct aq_hw_s *aq_hw, u32 vlan_untagged_act)
+void hw_atl_rpf_vlan_accept_untagged_packets_set(struct aq_hw_s *aq_hw,
+						 u32 vlan_acc_untagged_packets)
 {
-	aq_hw_write_reg_bit(aq_hw, rpf_vl_untagged_act_adr,
-			    rpf_vl_untagged_act_msk,
-			    rpf_vl_untagged_act_shift,
+	aq_hw_write_reg_bit(aq_hw, HW_ATL_RPF_VL_ACCEPT_UNTAGGED_MODE_ADR,
+			    HW_ATL_RPF_VL_ACCEPT_UNTAGGED_MODE_MSK,
+			    HW_ATL_RPF_VL_ACCEPT_UNTAGGED_MODE_SHIFT,
+			    vlan_acc_untagged_packets);
+}
+
+void hw_atl_rpf_vlan_untagged_act_set(struct aq_hw_s *aq_hw,
+				      u32 vlan_untagged_act)
+{
+	aq_hw_write_reg_bit(aq_hw, HW_ATL_RPF_VL_UNTAGGED_ACT_ADR,
+			    HW_ATL_RPF_VL_UNTAGGED_ACT_MSK,
+			    HW_ATL_RPF_VL_UNTAGGED_ACT_SHIFT,
 			    vlan_untagged_act);
 }
 
-void rpf_vlan_flr_en_set(struct aq_hw_s *aq_hw, u32 vlan_flr_en, u32 filter)
+void hw_atl_rpf_vlan_flr_en_set(struct aq_hw_s *aq_hw, u32 vlan_flr_en,
+				u32 filter)
 {
-	aq_hw_write_reg_bit(aq_hw, rpf_vl_en_f_adr(filter),
-			    rpf_vl_en_f_msk,
-			    rpf_vl_en_f_shift,
+	aq_hw_write_reg_bit(aq_hw, HW_ATL_RPF_VL_EN_F_ADR(filter),
+			    HW_ATL_RPF_VL_EN_F_MSK,
+			    HW_ATL_RPF_VL_EN_F_SHIFT,
 			    vlan_flr_en);
 }
 
-void rpf_vlan_flr_act_set(struct aq_hw_s *aq_hw, u32 vlan_flr_act, u32 filter)
+void hw_atl_rpf_vlan_flr_act_set(struct aq_hw_s *aq_hw, u32 vlan_flr_act,
+				 u32 filter)
 {
-	aq_hw_write_reg_bit(aq_hw, rpf_vl_act_f_adr(filter),
-			    rpf_vl_act_f_msk,
-			    rpf_vl_act_f_shift,
+	aq_hw_write_reg_bit(aq_hw, HW_ATL_RPF_VL_ACT_F_ADR(filter),
+			    HW_ATL_RPF_VL_ACT_F_MSK,
+			    HW_ATL_RPF_VL_ACT_F_SHIFT,
 			    vlan_flr_act);
 }
 
-void rpf_vlan_id_flr_set(struct aq_hw_s *aq_hw, u32 vlan_id_flr, u32 filter)
+void hw_atl_rpf_vlan_id_flr_set(struct aq_hw_s *aq_hw, u32 vlan_id_flr,
+				u32 filter)
 {
-	aq_hw_write_reg_bit(aq_hw, rpf_vl_id_f_adr(filter),
-			    rpf_vl_id_f_msk,
-			    rpf_vl_id_f_shift,
+	aq_hw_write_reg_bit(aq_hw, HW_ATL_RPF_VL_ID_F_ADR(filter),
+			    HW_ATL_RPF_VL_ID_F_MSK,
+			    HW_ATL_RPF_VL_ID_F_SHIFT,
 			    vlan_id_flr);
 }
 
-void rpf_etht_flr_en_set(struct aq_hw_s *aq_hw, u32 etht_flr_en, u32 filter)
+void hw_atl_rpf_vlan_rxq_en_flr_set(struct aq_hw_s *aq_hw, u32 vlan_rxq_en,
+				    u32 filter)
+{
+	aq_hw_write_reg_bit(aq_hw, HW_ATL_RPF_VL_RXQ_EN_F_ADR(filter),
+			    HW_ATL_RPF_VL_RXQ_EN_F_MSK,
+			    HW_ATL_RPF_VL_RXQ_EN_F_SHIFT,
+			    vlan_rxq_en);
+}
+
+void hw_atl_rpf_vlan_rxq_flr_set(struct aq_hw_s *aq_hw, u32 vlan_rxq,
+				 u32 filter)
 {
-	aq_hw_write_reg_bit(aq_hw, rpf_et_enf_adr(filter),
-			    rpf_et_enf_msk,
-			    rpf_et_enf_shift, etht_flr_en);
+	aq_hw_write_reg_bit(aq_hw, HW_ATL_RPF_VL_RXQ_F_ADR(filter),
+			    HW_ATL_RPF_VL_RXQ_F_MSK,
+			    HW_ATL_RPF_VL_RXQ_F_SHIFT,
+			    vlan_rxq);
+};
+
+void hw_atl_rpf_etht_flr_en_set(struct aq_hw_s *aq_hw, u32 etht_flr_en,
+				u32 filter)
+{
+	aq_hw_write_reg_bit(aq_hw, HW_ATL_RPF_ET_ENF_ADR(filter),
+			    HW_ATL_RPF_ET_ENF_MSK,
+			    HW_ATL_RPF_ET_ENF_SHIFT, etht_flr_en);
 }
 
-void rpf_etht_user_priority_en_set(struct aq_hw_s *aq_hw,
-				   u32 etht_user_priority_en, u32 filter)
+void hw_atl_rpf_etht_user_priority_en_set(struct aq_hw_s *aq_hw,
+					  u32 etht_user_priority_en, u32 filter)
 {
-	aq_hw_write_reg_bit(aq_hw, rpf_et_upfen_adr(filter),
-			    rpf_et_upfen_msk, rpf_et_upfen_shift,
+	aq_hw_write_reg_bit(aq_hw, HW_ATL_RPF_ET_UPFEN_ADR(filter),
+			    HW_ATL_RPF_ET_UPFEN_MSK, HW_ATL_RPF_ET_UPFEN_SHIFT,
 			    etht_user_priority_en);
 }
 
-void rpf_etht_rx_queue_en_set(struct aq_hw_s *aq_hw, u32 etht_rx_queue_en,
-			      u32 filter)
+void hw_atl_rpf_etht_rx_queue_en_set(struct aq_hw_s *aq_hw,
+				     u32 etht_rx_queue_en,
+				     u32 filter)
 {
-	aq_hw_write_reg_bit(aq_hw, rpf_et_rxqfen_adr(filter),
-			    rpf_et_rxqfen_msk, rpf_et_rxqfen_shift,
+	aq_hw_write_reg_bit(aq_hw, HW_ATL_RPF_ET_RXQFEN_ADR(filter),
+			    HW_ATL_RPF_ET_RXQFEN_MSK,
+			    HW_ATL_RPF_ET_RXQFEN_SHIFT,
 			    etht_rx_queue_en);
 }
 
-void rpf_etht_user_priority_set(struct aq_hw_s *aq_hw, u32 etht_user_priority,
-				u32 filter)
+void hw_atl_rpf_etht_user_priority_set(struct aq_hw_s *aq_hw,
+				       u32 etht_user_priority,
+				       u32 filter)
 {
-	aq_hw_write_reg_bit(aq_hw, rpf_et_upf_adr(filter),
-			    rpf_et_upf_msk,
-			    rpf_et_upf_shift, etht_user_priority);
+	aq_hw_write_reg_bit(aq_hw, HW_ATL_RPF_ET_UPF_ADR(filter),
+			    HW_ATL_RPF_ET_UPF_MSK,
+			    HW_ATL_RPF_ET_UPF_SHIFT, etht_user_priority);
 }
 
-void rpf_etht_rx_queue_set(struct aq_hw_s *aq_hw, u32 etht_rx_queue,
-			   u32 filter)
+void hw_atl_rpf_etht_rx_queue_set(struct aq_hw_s *aq_hw, u32 etht_rx_queue,
+				  u32 filter)
 {
-	aq_hw_write_reg_bit(aq_hw, rpf_et_rxqf_adr(filter),
-			    rpf_et_rxqf_msk,
-			    rpf_et_rxqf_shift, etht_rx_queue);
+	aq_hw_write_reg_bit(aq_hw, HW_ATL_RPF_ET_RXQF_ADR(filter),
+			    HW_ATL_RPF_ET_RXQF_MSK,
+			    HW_ATL_RPF_ET_RXQF_SHIFT, etht_rx_queue);
 }
 
-void rpf_etht_mgt_queue_set(struct aq_hw_s *aq_hw, u32 etht_mgt_queue,
-			    u32 filter)
+void hw_atl_rpf_etht_mgt_queue_set(struct aq_hw_s *aq_hw, u32 etht_mgt_queue,
+				   u32 filter)
 {
-	aq_hw_write_reg_bit(aq_hw, rpf_et_mng_rxqf_adr(filter),
-			    rpf_et_mng_rxqf_msk, rpf_et_mng_rxqf_shift,
+	aq_hw_write_reg_bit(aq_hw, HW_ATL_RPF_ET_MNG_RXQF_ADR(filter),
+			    HW_ATL_RPF_ET_MNG_RXQF_MSK,
+			    HW_ATL_RPF_ET_MNG_RXQF_SHIFT,
 			    etht_mgt_queue);
 }
 
-void rpf_etht_flr_act_set(struct aq_hw_s *aq_hw, u32 etht_flr_act, u32 filter)
+void hw_atl_rpf_etht_flr_act_set(struct aq_hw_s *aq_hw, u32 etht_flr_act,
+				 u32 filter)
 {
-	aq_hw_write_reg_bit(aq_hw, rpf_et_actf_adr(filter),
-			    rpf_et_actf_msk,
-			    rpf_et_actf_shift, etht_flr_act);
+	aq_hw_write_reg_bit(aq_hw, HW_ATL_RPF_ET_ACTF_ADR(filter),
+			    HW_ATL_RPF_ET_ACTF_MSK,
+			    HW_ATL_RPF_ET_ACTF_SHIFT, etht_flr_act);
 }
 
-void rpf_etht_flr_set(struct aq_hw_s *aq_hw, u32 etht_flr, u32 filter)
+void hw_atl_rpf_etht_flr_set(struct aq_hw_s *aq_hw, u32 etht_flr, u32 filter)
 {
-	aq_hw_write_reg_bit(aq_hw, rpf_et_valf_adr(filter),
-			    rpf_et_valf_msk,
-			    rpf_et_valf_shift, etht_flr);
+	aq_hw_write_reg_bit(aq_hw, HW_ATL_RPF_ET_VALF_ADR(filter),
+			    HW_ATL_RPF_ET_VALF_MSK,
+			    HW_ATL_RPF_ET_VALF_SHIFT, etht_flr);
+}
+
+void hw_atl_rpf_l4_spd_set(struct aq_hw_s *aq_hw, u32 val, u32 filter)
+{
+	aq_hw_write_reg_bit(aq_hw, HW_ATL_RPF_L4_SPD_ADR(filter),
+			    HW_ATL_RPF_L4_SPD_MSK,
+			    HW_ATL_RPF_L4_SPD_SHIFT, val);
+}
+
+void hw_atl_rpf_l4_dpd_set(struct aq_hw_s *aq_hw, u32 val, u32 filter)
+{
+	aq_hw_write_reg_bit(aq_hw, HW_ATL_RPF_L4_DPD_ADR(filter),
+			    HW_ATL_RPF_L4_DPD_MSK,
+			    HW_ATL_RPF_L4_DPD_SHIFT, val);
 }
 
 /* RPO: rx packet offload */
-void rpo_ipv4header_crc_offload_en_set(struct aq_hw_s *aq_hw,
-				       u32 ipv4header_crc_offload_en)
+void hw_atl_rpo_ipv4header_crc_offload_en_set(struct aq_hw_s *aq_hw,
+					      u32 ipv4header_crc_offload_en)
 {
-	aq_hw_write_reg_bit(aq_hw, rpo_ipv4chk_en_adr,
-			    rpo_ipv4chk_en_msk,
-			    rpo_ipv4chk_en_shift,
+	aq_hw_write_reg_bit(aq_hw, HW_ATL_RPO_IPV4CHK_EN_ADR,
+			    HW_ATL_RPO_IPV4CHK_EN_MSK,
+			    HW_ATL_RPO_IPV4CHK_EN_SHIFT,
 			    ipv4header_crc_offload_en);
 }
 
-void rpo_rx_desc_vlan_stripping_set(struct aq_hw_s *aq_hw,
-				    u32 rx_desc_vlan_stripping, u32 descriptor)
+void hw_atl_rpo_rx_desc_vlan_stripping_set(struct aq_hw_s *aq_hw,
+					   u32 rx_desc_vlan_stripping,
+					   u32 descriptor)
 {
-	aq_hw_write_reg_bit(aq_hw, rpo_descdvl_strip_adr(descriptor),
-			    rpo_descdvl_strip_msk,
-			    rpo_descdvl_strip_shift,
+	aq_hw_write_reg_bit(aq_hw, HW_ATL_RPO_DESCDVL_STRIP_ADR(descriptor),
+			    HW_ATL_RPO_DESCDVL_STRIP_MSK,
+			    HW_ATL_RPO_DESCDVL_STRIP_SHIFT,
 			    rx_desc_vlan_stripping);
 }
 
-void rpo_tcp_udp_crc_offload_en_set(struct aq_hw_s *aq_hw,
-				    u32 tcp_udp_crc_offload_en)
+void hw_atl_rpo_outer_vlan_tag_mode_set(void *context,
+					u32 outervlantagmode)
 {
-	aq_hw_write_reg_bit(aq_hw, rpol4chk_en_adr, rpol4chk_en_msk,
-			    rpol4chk_en_shift, tcp_udp_crc_offload_en);
+	aq_hw_write_reg_bit(context, HW_ATL_RPO_OUTER_VL_INS_MODE_ADR,
+			    HW_ATL_RPO_OUTER_VL_INS_MODE_MSK,
+			    HW_ATL_RPO_OUTER_VL_INS_MODE_SHIFT,
+			    outervlantagmode);
 }
 
-void rpo_lro_en_set(struct aq_hw_s *aq_hw, u32 lro_en)
+u32 hw_atl_rpo_outer_vlan_tag_mode_get(void *context)
 {
-	aq_hw_write_reg(aq_hw, rpo_lro_en_adr, lro_en);
+	return aq_hw_read_reg_bit(context, HW_ATL_RPO_OUTER_VL_INS_MODE_ADR,
+				  HW_ATL_RPO_OUTER_VL_INS_MODE_MSK,
+				  HW_ATL_RPO_OUTER_VL_INS_MODE_SHIFT);
 }
 
-void rpo_lro_patch_optimization_en_set(struct aq_hw_s *aq_hw,
-				       u32 lro_patch_optimization_en)
+void hw_atl_rpo_tcp_udp_crc_offload_en_set(struct aq_hw_s *aq_hw,
+					   u32 tcp_udp_crc_offload_en)
 {
-	aq_hw_write_reg_bit(aq_hw, rpo_lro_ptopt_en_adr,
-			    rpo_lro_ptopt_en_msk,
-			    rpo_lro_ptopt_en_shift,
+	aq_hw_write_reg_bit(aq_hw, HW_ATL_RPOL4CHK_EN_ADR,
+			    HW_ATL_RPOL4CHK_EN_MSK,
+			    HW_ATL_RPOL4CHK_EN_SHIFT, tcp_udp_crc_offload_en);
+}
+
+void hw_atl_rpo_lro_en_set(struct aq_hw_s *aq_hw, u32 lro_en)
+{
+	aq_hw_write_reg(aq_hw, HW_ATL_RPO_LRO_EN_ADR, lro_en);
+}
+
+void hw_atl_rpo_lro_patch_optimization_en_set(struct aq_hw_s *aq_hw,
+					      u32 lro_patch_optimization_en)
+{
+	aq_hw_write_reg_bit(aq_hw, HW_ATL_RPO_LRO_PTOPT_EN_ADR,
+			    HW_ATL_RPO_LRO_PTOPT_EN_MSK,
+			    HW_ATL_RPO_LRO_PTOPT_EN_SHIFT,
 			    lro_patch_optimization_en);
 }
 
-void rpo_lro_qsessions_lim_set(struct aq_hw_s *aq_hw,
-			       u32 lro_qsessions_lim)
+void hw_atl_rpo_lro_qsessions_lim_set(struct aq_hw_s *aq_hw,
+				      u32 lro_qsessions_lim)
 {
-	aq_hw_write_reg_bit(aq_hw, rpo_lro_qses_lmt_adr,
-			    rpo_lro_qses_lmt_msk,
-			    rpo_lro_qses_lmt_shift,
+	aq_hw_write_reg_bit(aq_hw, HW_ATL_RPO_LRO_QSES_LMT_ADR,
+			    HW_ATL_RPO_LRO_QSES_LMT_MSK,
+			    HW_ATL_RPO_LRO_QSES_LMT_SHIFT,
 			    lro_qsessions_lim);
 }
 
-void rpo_lro_total_desc_lim_set(struct aq_hw_s *aq_hw, u32 lro_total_desc_lim)
+void hw_atl_rpo_lro_total_desc_lim_set(struct aq_hw_s *aq_hw,
+				       u32 lro_total_desc_lim)
 {
-	aq_hw_write_reg_bit(aq_hw, rpo_lro_tot_dsc_lmt_adr,
-			    rpo_lro_tot_dsc_lmt_msk,
-			    rpo_lro_tot_dsc_lmt_shift,
+	aq_hw_write_reg_bit(aq_hw, HW_ATL_RPO_LRO_TOT_DSC_LMT_ADR,
+			    HW_ATL_RPO_LRO_TOT_DSC_LMT_MSK,
+			    HW_ATL_RPO_LRO_TOT_DSC_LMT_SHIFT,
 			    lro_total_desc_lim);
 }
 
-void rpo_lro_min_pay_of_first_pkt_set(struct aq_hw_s *aq_hw,
-				      u32 lro_min_pld_of_first_pkt)
+void hw_atl_rpo_lro_min_pay_of_first_pkt_set(struct aq_hw_s *aq_hw,
+					     u32 lro_min_pld_of_first_pkt)
 {
-	aq_hw_write_reg_bit(aq_hw, rpo_lro_pkt_min_adr,
-			    rpo_lro_pkt_min_msk,
-			    rpo_lro_pkt_min_shift,
+	aq_hw_write_reg_bit(aq_hw, HW_ATL_RPO_LRO_PKT_MIN_ADR,
+			    HW_ATL_RPO_LRO_PKT_MIN_MSK,
+			    HW_ATL_RPO_LRO_PKT_MIN_SHIFT,
 			    lro_min_pld_of_first_pkt);
 }
 
-void rpo_lro_pkt_lim_set(struct aq_hw_s *aq_hw, u32 lro_pkt_lim)
+void hw_atl_rpo_lro_pkt_lim_set(struct aq_hw_s *aq_hw, u32 lro_pkt_lim)
 {
-	aq_hw_write_reg(aq_hw, rpo_lro_rsc_max_adr, lro_pkt_lim);
+	aq_hw_write_reg(aq_hw, HW_ATL_RPO_LRO_RSC_MAX_ADR, lro_pkt_lim);
 }
 
-void rpo_lro_max_num_of_descriptors_set(struct aq_hw_s *aq_hw,
-					u32 lro_max_number_of_descriptors,
-					u32 lro)
+void hw_atl_rpo_lro_max_num_of_descriptors_set(struct aq_hw_s *aq_hw,
+					       u32 lro_max_number_of_descriptors,
+					       u32 lro)
 {
 /* Register address for bitfield lro{L}_des_max[1:0] */
 	static u32 rpo_lro_ldes_max_adr[32] = {
@@ -1017,378 +1244,653 @@ void rpo_lro_max_num_of_descriptors_set(struct aq_hw_s *aq_hw,
 			    lro_max_number_of_descriptors);
 }
 
-void rpo_lro_time_base_divider_set(struct aq_hw_s *aq_hw,
-				   u32 lro_time_base_divider)
+void hw_atl_rpo_lro_time_base_divider_set(struct aq_hw_s *aq_hw,
+					  u32 lro_time_base_divider)
 {
-	aq_hw_write_reg_bit(aq_hw, rpo_lro_tb_div_adr,
-			    rpo_lro_tb_div_msk,
-			    rpo_lro_tb_div_shift,
+	aq_hw_write_reg_bit(aq_hw, HW_ATL_RPO_LRO_TB_DIV_ADR,
+			    HW_ATL_RPO_LRO_TB_DIV_MSK,
+			    HW_ATL_RPO_LRO_TB_DIV_SHIFT,
 			    lro_time_base_divider);
 }
 
-void rpo_lro_inactive_interval_set(struct aq_hw_s *aq_hw,
-				   u32 lro_inactive_interval)
+void hw_atl_rpo_lro_inactive_interval_set(struct aq_hw_s *aq_hw,
+					  u32 lro_inactive_interval)
 {
-	aq_hw_write_reg_bit(aq_hw, rpo_lro_ina_ival_adr,
-			    rpo_lro_ina_ival_msk,
-			    rpo_lro_ina_ival_shift,
+	aq_hw_write_reg_bit(aq_hw, HW_ATL_RPO_LRO_INA_IVAL_ADR,
+			    HW_ATL_RPO_LRO_INA_IVAL_MSK,
+			    HW_ATL_RPO_LRO_INA_IVAL_SHIFT,
 			    lro_inactive_interval);
 }
 
-void rpo_lro_max_coalescing_interval_set(struct aq_hw_s *aq_hw,
-					 u32 lro_max_coalescing_interval)
+void hw_atl_rpo_lro_max_coalescing_interval_set(struct aq_hw_s *aq_hw,
+						u32 lro_max_coal_interval)
 {
-	aq_hw_write_reg_bit(aq_hw, rpo_lro_max_ival_adr,
-			    rpo_lro_max_ival_msk,
-			    rpo_lro_max_ival_shift,
-			    lro_max_coalescing_interval);
+	aq_hw_write_reg_bit(aq_hw, HW_ATL_RPO_LRO_MAX_IVAL_ADR,
+			    HW_ATL_RPO_LRO_MAX_IVAL_MSK,
+			    HW_ATL_RPO_LRO_MAX_IVAL_SHIFT,
+			    lro_max_coal_interval);
 }
 
 /* rx */
-void rx_rx_reg_res_dis_set(struct aq_hw_s *aq_hw, u32 rx_reg_res_dis)
+void hw_atl_rx_rx_reg_res_dis_set(struct aq_hw_s *aq_hw, u32 rx_reg_res_dis)
 {
-	aq_hw_write_reg_bit(aq_hw, rx_reg_res_dsbl_adr,
-			    rx_reg_res_dsbl_msk,
-			    rx_reg_res_dsbl_shift,
+	aq_hw_write_reg_bit(aq_hw, HW_ATL_RX_REG_RES_DSBL_ADR,
+			    HW_ATL_RX_REG_RES_DSBL_MSK,
+			    HW_ATL_RX_REG_RES_DSBL_SHIFT,
 			    rx_reg_res_dis);
 }
 
 /* tdm */
-void tdm_cpu_id_set(struct aq_hw_s *aq_hw, u32 cpuid, u32 dca)
+void hw_atl_tdm_cpu_id_set(struct aq_hw_s *aq_hw, u32 cpuid, u32 dca)
 {
-	aq_hw_write_reg_bit(aq_hw, tdm_dcadcpuid_adr(dca),
-			    tdm_dcadcpuid_msk,
-			    tdm_dcadcpuid_shift, cpuid);
+	aq_hw_write_reg_bit(aq_hw, HW_ATL_TDM_DCADCPUID_ADR(dca),
+			    HW_ATL_TDM_DCADCPUID_MSK,
+			    HW_ATL_TDM_DCADCPUID_SHIFT, cpuid);
 }
 
-void tdm_large_send_offload_en_set(struct aq_hw_s *aq_hw,
-				   u32 large_send_offload_en)
+void hw_atl_tdm_large_send_offload_en_set(struct aq_hw_s *aq_hw,
+					  u32 large_send_offload_en)
 {
-	aq_hw_write_reg(aq_hw, tdm_lso_en_adr, large_send_offload_en);
+	aq_hw_write_reg(aq_hw, HW_ATL_TDM_LSO_EN_ADR, large_send_offload_en);
 }
 
-void tdm_tx_dca_en_set(struct aq_hw_s *aq_hw, u32 tx_dca_en)
+void hw_atl_tdm_tx_dca_en_set(struct aq_hw_s *aq_hw, u32 tx_dca_en)
 {
-	aq_hw_write_reg_bit(aq_hw, tdm_dca_en_adr, tdm_dca_en_msk,
-			    tdm_dca_en_shift, tx_dca_en);
+	aq_hw_write_reg_bit(aq_hw, HW_ATL_TDM_DCA_EN_ADR, HW_ATL_TDM_DCA_EN_MSK,
+			    HW_ATL_TDM_DCA_EN_SHIFT, tx_dca_en);
 }
 
-void tdm_tx_dca_mode_set(struct aq_hw_s *aq_hw, u32 tx_dca_mode)
+void hw_atl_tdm_tx_dca_mode_set(struct aq_hw_s *aq_hw, u32 tx_dca_mode)
 {
-	aq_hw_write_reg_bit(aq_hw, tdm_dca_mode_adr, tdm_dca_mode_msk,
-			    tdm_dca_mode_shift, tx_dca_mode);
+	aq_hw_write_reg_bit(aq_hw, HW_ATL_TDM_DCA_MODE_ADR,
+			    HW_ATL_TDM_DCA_MODE_MSK,
+			    HW_ATL_TDM_DCA_MODE_SHIFT, tx_dca_mode);
 }
 
-void tdm_tx_desc_dca_en_set(struct aq_hw_s *aq_hw, u32 tx_desc_dca_en, u32 dca)
+void hw_atl_tdm_tx_desc_dca_en_set(struct aq_hw_s *aq_hw, u32 tx_desc_dca_en,
+				   u32 dca)
 {
-	aq_hw_write_reg_bit(aq_hw, tdm_dcaddesc_en_adr(dca),
-			    tdm_dcaddesc_en_msk, tdm_dcaddesc_en_shift,
+	aq_hw_write_reg_bit(aq_hw, HW_ATL_TDM_DCADDESC_EN_ADR(dca),
+			    HW_ATL_TDM_DCADDESC_EN_MSK,
+			    HW_ATL_TDM_DCADDESC_EN_SHIFT,
 			    tx_desc_dca_en);
 }
 
-void tdm_tx_desc_en_set(struct aq_hw_s *aq_hw, u32 tx_desc_en, u32 descriptor)
+void hw_atl_tdm_tx_desc_en_set(struct aq_hw_s *aq_hw, u32 tx_desc_en,
+			       u32 descriptor)
 {
-	aq_hw_write_reg_bit(aq_hw, tdm_descden_adr(descriptor),
-			    tdm_descden_msk,
-			    tdm_descden_shift,
+	aq_hw_write_reg_bit(aq_hw, HW_ATL_TDM_DESCDEN_ADR(descriptor),
+			    HW_ATL_TDM_DESCDEN_MSK,
+			    HW_ATL_TDM_DESCDEN_SHIFT,
 			    tx_desc_en);
 }
 
-u32 tdm_tx_desc_head_ptr_get(struct aq_hw_s *aq_hw, u32 descriptor)
+u32 hw_atl_tdm_tx_desc_head_ptr_get(struct aq_hw_s *aq_hw, u32 descriptor)
 {
-	return aq_hw_read_reg_bit(aq_hw, tdm_descdhd_adr(descriptor),
-				  tdm_descdhd_msk, tdm_descdhd_shift);
+	return aq_hw_read_reg_bit(aq_hw, HW_ATL_TDM_DESCDHD_ADR(descriptor),
+				  HW_ATL_TDM_DESCDHD_MSK,
+				  HW_ATL_TDM_DESCDHD_SHIFT);
 }
 
-void tdm_tx_desc_len_set(struct aq_hw_s *aq_hw, u32 tx_desc_len,
-			 u32 descriptor)
+void hw_atl_tdm_tx_desc_len_set(struct aq_hw_s *aq_hw, u32 tx_desc_len,
+				u32 descriptor)
 {
-	aq_hw_write_reg_bit(aq_hw, tdm_descdlen_adr(descriptor),
-			    tdm_descdlen_msk,
-			    tdm_descdlen_shift,
+	aq_hw_write_reg_bit(aq_hw, HW_ATL_TDM_DESCDLEN_ADR(descriptor),
+			    HW_ATL_TDM_DESCDLEN_MSK,
+			    HW_ATL_TDM_DESCDLEN_SHIFT,
 			    tx_desc_len);
 }
 
-void tdm_tx_desc_wr_wb_irq_en_set(struct aq_hw_s *aq_hw,
-				  u32 tx_desc_wr_wb_irq_en)
+void hw_atl_tdm_tx_desc_wr_wb_irq_en_set(struct aq_hw_s *aq_hw,
+					 u32 tx_desc_wr_wb_irq_en)
 {
-	aq_hw_write_reg_bit(aq_hw, tdm_int_desc_wrb_en_adr,
-			    tdm_int_desc_wrb_en_msk,
-			    tdm_int_desc_wrb_en_shift,
+	aq_hw_write_reg_bit(aq_hw, HW_ATL_TDM_INT_DESC_WRB_EN_ADR,
+			    HW_ATL_TDM_INT_DESC_WRB_EN_MSK,
+			    HW_ATL_TDM_INT_DESC_WRB_EN_SHIFT,
 			    tx_desc_wr_wb_irq_en);
 }
 
-void tdm_tx_desc_wr_wb_threshold_set(struct aq_hw_s *aq_hw,
-				     u32 tx_desc_wr_wb_threshold,
-				     u32 descriptor)
+void hw_atl_tdm_tx_desc_wr_wb_threshold_set(struct aq_hw_s *aq_hw,
+					    u32 tx_desc_wr_wb_threshold,
+					    u32 descriptor)
 {
-	aq_hw_write_reg_bit(aq_hw, tdm_descdwrb_thresh_adr(descriptor),
-			    tdm_descdwrb_thresh_msk,
-			    tdm_descdwrb_thresh_shift,
+	aq_hw_write_reg_bit(aq_hw, HW_ATL_TDM_DESCDWRB_THRESH_ADR(descriptor),
+			    HW_ATL_TDM_DESCDWRB_THRESH_MSK,
+			    HW_ATL_TDM_DESCDWRB_THRESH_SHIFT,
 			    tx_desc_wr_wb_threshold);
 }
 
-void tdm_tdm_intr_moder_en_set(struct aq_hw_s *aq_hw,
-			       u32 tdm_irq_moderation_en)
+void hw_atl_tdm_tdm_intr_moder_en_set(struct aq_hw_s *aq_hw,
+				      u32 tdm_irq_moderation_en)
 {
-	aq_hw_write_reg_bit(aq_hw, tdm_int_mod_en_adr,
-			    tdm_int_mod_en_msk,
-			    tdm_int_mod_en_shift,
+	aq_hw_write_reg_bit(aq_hw, HW_ATL_TDM_INT_MOD_EN_ADR,
+			    HW_ATL_TDM_INT_MOD_EN_MSK,
+			    HW_ATL_TDM_INT_MOD_EN_SHIFT,
 			    tdm_irq_moderation_en);
 }
 
 /* thm */
-void thm_lso_tcp_flag_of_first_pkt_set(struct aq_hw_s *aq_hw,
-				       u32 lso_tcp_flag_of_first_pkt)
+void hw_atl_thm_lso_tcp_flag_of_first_pkt_set(struct aq_hw_s *aq_hw,
+					      u32 lso_tcp_flag_of_first_pkt)
 {
-	aq_hw_write_reg_bit(aq_hw, thm_lso_tcp_flag_first_adr,
-			    thm_lso_tcp_flag_first_msk,
-			    thm_lso_tcp_flag_first_shift,
+	aq_hw_write_reg_bit(aq_hw, HW_ATL_THM_LSO_TCP_FLAG_FIRST_ADR,
+			    HW_ATL_THM_LSO_TCP_FLAG_FIRST_MSK,
+			    HW_ATL_THM_LSO_TCP_FLAG_FIRST_SHIFT,
 			    lso_tcp_flag_of_first_pkt);
 }
 
-void thm_lso_tcp_flag_of_last_pkt_set(struct aq_hw_s *aq_hw,
-				      u32 lso_tcp_flag_of_last_pkt)
+void hw_atl_thm_lso_tcp_flag_of_last_pkt_set(struct aq_hw_s *aq_hw,
+					     u32 lso_tcp_flag_of_last_pkt)
 {
-	aq_hw_write_reg_bit(aq_hw, thm_lso_tcp_flag_last_adr,
-			    thm_lso_tcp_flag_last_msk,
-			    thm_lso_tcp_flag_last_shift,
+	aq_hw_write_reg_bit(aq_hw, HW_ATL_THM_LSO_TCP_FLAG_LAST_ADR,
+			    HW_ATL_THM_LSO_TCP_FLAG_LAST_MSK,
+			    HW_ATL_THM_LSO_TCP_FLAG_LAST_SHIFT,
 			    lso_tcp_flag_of_last_pkt);
 }
 
-void thm_lso_tcp_flag_of_middle_pkt_set(struct aq_hw_s *aq_hw,
-					u32 lso_tcp_flag_of_middle_pkt)
+void hw_atl_thm_lso_tcp_flag_of_middle_pkt_set(struct aq_hw_s *aq_hw,
+					       u32 lso_tcp_flag_of_middle_pkt)
 {
-	aq_hw_write_reg_bit(aq_hw, thm_lso_tcp_flag_mid_adr,
-			    thm_lso_tcp_flag_mid_msk,
-			    thm_lso_tcp_flag_mid_shift,
+	aq_hw_write_reg_bit(aq_hw, HW_ATL_THM_LSO_TCP_FLAG_MID_ADR,
+			    HW_ATL_THM_LSO_TCP_FLAG_MID_MSK,
+			    HW_ATL_THM_LSO_TCP_FLAG_MID_SHIFT,
 			    lso_tcp_flag_of_middle_pkt);
 }
 
 /* TPB: tx packet buffer */
-void tpb_tx_buff_en_set(struct aq_hw_s *aq_hw, u32 tx_buff_en)
+void hw_atl_tpb_tx_buff_en_set(struct aq_hw_s *aq_hw, u32 tx_buff_en)
 {
-	aq_hw_write_reg_bit(aq_hw, tpb_tx_buf_en_adr, tpb_tx_buf_en_msk,
-			    tpb_tx_buf_en_shift, tx_buff_en);
+	aq_hw_write_reg_bit(aq_hw, HW_ATL_TPB_TX_BUF_EN_ADR,
+			    HW_ATL_TPB_TX_BUF_EN_MSK,
+			    HW_ATL_TPB_TX_BUF_EN_SHIFT, tx_buff_en);
 }
 
-void tpb_tx_buff_hi_threshold_per_tc_set(struct aq_hw_s *aq_hw,
-					 u32 tx_buff_hi_threshold_per_tc,
+u32 hw_atl_tpb_tps_tx_tc_mode_get(struct aq_hw_s *aq_hw)
+{
+	return aq_hw_read_reg_bit(aq_hw, HW_ATL_TPB_TX_TC_MODE_ADDR,
+			HW_ATL_TPB_TX_TC_MODE_MSK,
+			HW_ATL_TPB_TX_TC_MODE_SHIFT);
+}
+
+void hw_atl_tpb_tps_tx_tc_mode_set(struct aq_hw_s *aq_hw,
+				   u32 tx_traf_class_mode)
+{
+	aq_hw_write_reg_bit(aq_hw, HW_ATL_TPB_TX_TC_MODE_ADDR,
+			HW_ATL_TPB_TX_TC_MODE_MSK,
+			HW_ATL_TPB_TX_TC_MODE_SHIFT,
+			tx_traf_class_mode);
+}
+
+void hw_atl_tpb_tx_buff_hi_threshold_per_tc_set(struct aq_hw_s *aq_hw,
+						u32 tx_buff_hi_threshold_per_tc,
 					 u32 buffer)
 {
-	aq_hw_write_reg_bit(aq_hw, tpb_txbhi_thresh_adr(buffer),
-			    tpb_txbhi_thresh_msk, tpb_txbhi_thresh_shift,
+	aq_hw_write_reg_bit(aq_hw, HW_ATL_TPB_TXBHI_THRESH_ADR(buffer),
+			    HW_ATL_TPB_TXBHI_THRESH_MSK,
+			    HW_ATL_TPB_TXBHI_THRESH_SHIFT,
 			    tx_buff_hi_threshold_per_tc);
 }
 
-void tpb_tx_buff_lo_threshold_per_tc_set(struct aq_hw_s *aq_hw,
-					 u32 tx_buff_lo_threshold_per_tc,
+void hw_atl_tpb_tx_buff_lo_threshold_per_tc_set(struct aq_hw_s *aq_hw,
+						u32 tx_buff_lo_threshold_per_tc,
 					 u32 buffer)
 {
-	aq_hw_write_reg_bit(aq_hw, tpb_txblo_thresh_adr(buffer),
-			    tpb_txblo_thresh_msk, tpb_txblo_thresh_shift,
+	aq_hw_write_reg_bit(aq_hw, HW_ATL_TPB_TXBLO_THRESH_ADR(buffer),
+			    HW_ATL_TPB_TXBLO_THRESH_MSK,
+			    HW_ATL_TPB_TXBLO_THRESH_SHIFT,
 			    tx_buff_lo_threshold_per_tc);
 }
 
-void tpb_tx_dma_sys_lbk_en_set(struct aq_hw_s *aq_hw, u32 tx_dma_sys_lbk_en)
+void hw_atl_tpb_tx_dma_sys_lbk_en_set(struct aq_hw_s *aq_hw, u32 tx_dma_sys_lbk_en)
 {
-	aq_hw_write_reg_bit(aq_hw, tpb_dma_sys_lbk_adr,
-			    tpb_dma_sys_lbk_msk,
-			    tpb_dma_sys_lbk_shift,
+	aq_hw_write_reg_bit(aq_hw, HW_ATL_TPB_DMA_SYS_LBK_ADR,
+			    HW_ATL_TPB_DMA_SYS_LBK_MSK,
+			    HW_ATL_TPB_DMA_SYS_LBK_SHIFT,
 			    tx_dma_sys_lbk_en);
 }
 
-void tpb_tx_pkt_buff_size_per_tc_set(struct aq_hw_s *aq_hw,
-				     u32 tx_pkt_buff_size_per_tc, u32 buffer)
+void hw_atl_tpb_tx_dma_net_lbk_en_set(struct aq_hw_s *aq_hw,
+				      u32 tx_dma_net_lbk_en)
 {
-	aq_hw_write_reg_bit(aq_hw, tpb_txbbuf_size_adr(buffer),
-			    tpb_txbbuf_size_msk,
-			    tpb_txbbuf_size_shift,
+	aq_hw_write_reg_bit(aq_hw, HW_ATL_TPB_DMA_NET_LBK_ADR,
+			    HW_ATL_TPB_DMA_NET_LBK_MSK,
+			    HW_ATL_TPB_DMA_NET_LBK_SHIFT,
+			    tx_dma_net_lbk_en);
+}
+
+void hw_atl_tpb_tx_tx_clk_gate_en_set(struct aq_hw_s *aq_hw,
+				      u32 tx_clk_gate_en)
+{
+	aq_hw_write_reg_bit(aq_hw, HW_ATL_TPB_TX_CLK_GATE_EN_ADR,
+			    HW_ATL_TPB_TX_CLK_GATE_EN_MSK,
+			    HW_ATL_TPB_TX_CLK_GATE_EN_SHIFT,
+			    tx_clk_gate_en);
+}
+
+void hw_atl_tpb_tx_pkt_buff_size_per_tc_set(struct aq_hw_s *aq_hw,
+
+					    u32 tx_pkt_buff_size_per_tc, u32 buffer)
+{
+	aq_hw_write_reg_bit(aq_hw, HW_ATL_TPB_TXBBUF_SIZE_ADR(buffer),
+			    HW_ATL_TPB_TXBBUF_SIZE_MSK,
+			    HW_ATL_TPB_TXBBUF_SIZE_SHIFT,
 			    tx_pkt_buff_size_per_tc);
 }
 
-void tpb_tx_path_scp_ins_en_set(struct aq_hw_s *aq_hw, u32 tx_path_scp_ins_en)
+void hw_atl_tpb_tx_path_scp_ins_en_set(struct aq_hw_s *aq_hw, u32 tx_path_scp_ins_en)
 {
-	aq_hw_write_reg_bit(aq_hw, tpb_tx_scp_ins_en_adr,
-			    tpb_tx_scp_ins_en_msk,
-			    tpb_tx_scp_ins_en_shift,
+	aq_hw_write_reg_bit(aq_hw, HW_ATL_TPB_TX_SCP_INS_EN_ADR,
+			    HW_ATL_TPB_TX_SCP_INS_EN_MSK,
+			    HW_ATL_TPB_TX_SCP_INS_EN_SHIFT,
 			    tx_path_scp_ins_en);
 }
 
 /* TPO: tx packet offload */
-void tpo_ipv4header_crc_offload_en_set(struct aq_hw_s *aq_hw,
-				       u32 ipv4header_crc_offload_en)
+void hw_atl_tpo_ipv4header_crc_offload_en_set(struct aq_hw_s *aq_hw,
+					      u32 ipv4header_crc_offload_en)
 {
-	aq_hw_write_reg_bit(aq_hw, tpo_ipv4chk_en_adr,
-			    tpo_ipv4chk_en_msk,
-			    tpo_ipv4chk_en_shift,
+	aq_hw_write_reg_bit(aq_hw, HW_ATL_TPO_IPV4CHK_EN_ADR,
+			    HW_ATL_TPO_IPV4CHK_EN_MSK,
+			    HW_ATL_TPO_IPV4CHK_EN_SHIFT,
 			    ipv4header_crc_offload_en);
 }
 
-void tpo_tcp_udp_crc_offload_en_set(struct aq_hw_s *aq_hw,
-				    u32 tcp_udp_crc_offload_en)
+void hw_atl_tpo_tcp_udp_crc_offload_en_set(struct aq_hw_s *aq_hw,
+					   u32 tcp_udp_crc_offload_en)
 {
-	aq_hw_write_reg_bit(aq_hw, tpol4chk_en_adr,
-			    tpol4chk_en_msk,
-			    tpol4chk_en_shift,
+	aq_hw_write_reg_bit(aq_hw, HW_ATL_TPOL4CHK_EN_ADR,
+			    HW_ATL_TPOL4CHK_EN_MSK,
+			    HW_ATL_TPOL4CHK_EN_SHIFT,
 			    tcp_udp_crc_offload_en);
 }
 
-void tpo_tx_pkt_sys_lbk_en_set(struct aq_hw_s *aq_hw, u32 tx_pkt_sys_lbk_en)
+void hw_atl_tpo_tx_pkt_sys_lbk_en_set(struct aq_hw_s *aq_hw,
+				      u32 tx_pkt_sys_lbk_en)
 {
-	aq_hw_write_reg_bit(aq_hw, tpo_pkt_sys_lbk_adr,
-			    tpo_pkt_sys_lbk_msk,
-			    tpo_pkt_sys_lbk_shift,
+	aq_hw_write_reg_bit(aq_hw, HW_ATL_TPO_PKT_SYS_LBK_ADR,
+			    HW_ATL_TPO_PKT_SYS_LBK_MSK,
+			    HW_ATL_TPO_PKT_SYS_LBK_SHIFT,
 			    tx_pkt_sys_lbk_en);
 }
 
 /* TPS: tx packet scheduler */
-void tps_tx_pkt_shed_data_arb_mode_set(struct aq_hw_s *aq_hw,
-				       u32 tx_pkt_shed_data_arb_mode)
+void hw_atl_tps_tx_pkt_shed_data_arb_mode_set(struct aq_hw_s *aq_hw,
+					      u32 tx_pkt_shed_data_arb_mode)
 {
-	aq_hw_write_reg_bit(aq_hw, tps_data_tc_arb_mode_adr,
-			    tps_data_tc_arb_mode_msk,
-			    tps_data_tc_arb_mode_shift,
+	aq_hw_write_reg_bit(aq_hw, HW_ATL_TPS_DATA_TC_ARB_MODE_ADR,
+			    HW_ATL_TPS_DATA_TC_ARB_MODE_MSK,
+			    HW_ATL_TPS_DATA_TC_ARB_MODE_SHIFT,
 			    tx_pkt_shed_data_arb_mode);
 }
 
-void tps_tx_pkt_shed_desc_rate_curr_time_res_set(struct aq_hw_s *aq_hw,
-						 u32 curr_time_res)
+void hw_atl_tps_tx_pkt_shed_desc_rate_curr_time_res_set(struct aq_hw_s *aq_hw,
+							u32 curr_time_res)
 {
-	aq_hw_write_reg_bit(aq_hw, tps_desc_rate_ta_rst_adr,
-			    tps_desc_rate_ta_rst_msk,
-			    tps_desc_rate_ta_rst_shift,
+	aq_hw_write_reg_bit(aq_hw, HW_ATL_TPS_DESC_RATE_TA_RST_ADR,
+			    HW_ATL_TPS_DESC_RATE_TA_RST_MSK,
+			    HW_ATL_TPS_DESC_RATE_TA_RST_SHIFT,
 			    curr_time_res);
 }
 
-void tps_tx_pkt_shed_desc_rate_lim_set(struct aq_hw_s *aq_hw,
-				       u32 tx_pkt_shed_desc_rate_lim)
+void hw_atl_tps_tx_pkt_shed_desc_rate_lim_set(struct aq_hw_s *aq_hw,
+					      u32 tx_pkt_shed_desc_rate_lim)
 {
-	aq_hw_write_reg_bit(aq_hw, tps_desc_rate_lim_adr,
-			    tps_desc_rate_lim_msk,
-			    tps_desc_rate_lim_shift,
+	aq_hw_write_reg_bit(aq_hw, HW_ATL_TPS_DESC_RATE_LIM_ADR,
+			    HW_ATL_TPS_DESC_RATE_LIM_MSK,
+			    HW_ATL_TPS_DESC_RATE_LIM_SHIFT,
 			    tx_pkt_shed_desc_rate_lim);
 }
 
-void tps_tx_pkt_shed_desc_tc_arb_mode_set(struct aq_hw_s *aq_hw,
-					  u32 tx_pkt_shed_desc_tc_arb_mode)
+void hw_atl_tps_tx_pkt_shed_desc_tc_arb_mode_set(struct aq_hw_s *aq_hw,
+						 u32 arb_mode)
+{
+	aq_hw_write_reg_bit(aq_hw, HW_ATL_TPS_DESC_TC_ARB_MODE_ADR,
+			    HW_ATL_TPS_DESC_TC_ARB_MODE_MSK,
+			    HW_ATL_TPS_DESC_TC_ARB_MODE_SHIFT,
+			    arb_mode);
+}
+
+void hw_atl_tps_tx_pkt_shed_desc_tc_max_credit_set(struct aq_hw_s *aq_hw,
+						   const u32 tc,
+						   const u32 max_credit)
+{
+	aq_hw_write_reg_bit(aq_hw, HW_ATL_TPS_DESC_TCTCREDIT_MAX_ADR(tc),
+			    HW_ATL_TPS_DESC_TCTCREDIT_MAX_MSK,
+			    HW_ATL_TPS_DESC_TCTCREDIT_MAX_SHIFT,
+			    max_credit);
+}
+
+void hw_atl_tps_tx_pkt_shed_desc_tc_weight_set(struct aq_hw_s *aq_hw,
+					       const u32 tc,
+					       const u32 weight)
 {
-	aq_hw_write_reg_bit(aq_hw, tps_desc_tc_arb_mode_adr,
-			    tps_desc_tc_arb_mode_msk,
-			    tps_desc_tc_arb_mode_shift,
-			    tx_pkt_shed_desc_tc_arb_mode);
+	aq_hw_write_reg_bit(aq_hw, HW_ATL_TPS_DESC_TCTWEIGHT_ADR(tc),
+			    HW_ATL_TPS_DESC_TCTWEIGHT_MSK,
+			    HW_ATL_TPS_DESC_TCTWEIGHT_SHIFT,
+			    weight);
 }
 
-void tps_tx_pkt_shed_desc_tc_max_credit_set(struct aq_hw_s *aq_hw,
-					    u32 tx_pkt_shed_desc_tc_max_credit,
-					    u32 tc)
+void hw_atl_tps_tx_pkt_shed_desc_vm_arb_mode_set(struct aq_hw_s *aq_hw,
+						 u32 arb_mode)
 {
-	aq_hw_write_reg_bit(aq_hw, tps_desc_tctcredit_max_adr(tc),
-			    tps_desc_tctcredit_max_msk,
-			    tps_desc_tctcredit_max_shift,
-			    tx_pkt_shed_desc_tc_max_credit);
+	aq_hw_write_reg_bit(aq_hw, HW_ATL_TPS_DESC_VM_ARB_MODE_ADR,
+			    HW_ATL_TPS_DESC_VM_ARB_MODE_MSK,
+			    HW_ATL_TPS_DESC_VM_ARB_MODE_SHIFT,
+			    arb_mode);
 }
 
-void tps_tx_pkt_shed_desc_tc_weight_set(struct aq_hw_s *aq_hw,
-					u32 tx_pkt_shed_desc_tc_weight, u32 tc)
+void hw_atl_tps_tx_pkt_shed_tc_data_max_credit_set(struct aq_hw_s *aq_hw,
+						   const u32 tc,
+						   const u32 max_credit)
 {
-	aq_hw_write_reg_bit(aq_hw, tps_desc_tctweight_adr(tc),
-			    tps_desc_tctweight_msk,
-			    tps_desc_tctweight_shift,
-			    tx_pkt_shed_desc_tc_weight);
+	aq_hw_write_reg_bit(aq_hw, HW_ATL_TPS_DATA_TCTCREDIT_MAX_ADR(tc),
+			    HW_ATL_TPS_DATA_TCTCREDIT_MAX_MSK,
+			    HW_ATL_TPS_DATA_TCTCREDIT_MAX_SHIFT,
+			    max_credit);
 }
 
-void tps_tx_pkt_shed_desc_vm_arb_mode_set(struct aq_hw_s *aq_hw,
-					  u32 tx_pkt_shed_desc_vm_arb_mode)
+void hw_atl_tps_tx_pkt_shed_tc_data_weight_set(struct aq_hw_s *aq_hw,
+					       const u32 tc,
+					       const u32 weight)
 {
-	aq_hw_write_reg_bit(aq_hw, tps_desc_vm_arb_mode_adr,
-			    tps_desc_vm_arb_mode_msk,
-			    tps_desc_vm_arb_mode_shift,
-			    tx_pkt_shed_desc_vm_arb_mode);
+	aq_hw_write_reg_bit(aq_hw, HW_ATL_TPS_DATA_TCTWEIGHT_ADR(tc),
+			    HW_ATL_TPS_DATA_TCTWEIGHT_MSK,
+			    HW_ATL_TPS_DATA_TCTWEIGHT_SHIFT,
+			    weight);
 }
 
-void tps_tx_pkt_shed_tc_data_max_credit_set(struct aq_hw_s *aq_hw,
-					    u32 tx_pkt_shed_tc_data_max_credit,
-					    u32 tc)
+void hw_atl_tps_tx_desc_rate_mode_set(struct aq_hw_s *aq_hw,
+				      const u32 rate_mode)
 {
-	aq_hw_write_reg_bit(aq_hw, tps_data_tctcredit_max_adr(tc),
-			    tps_data_tctcredit_max_msk,
-			    tps_data_tctcredit_max_shift,
-			    tx_pkt_shed_tc_data_max_credit);
+	aq_hw_write_reg_bit(aq_hw, HW_ATL_TPS_TX_DESC_RATE_MODE_ADR,
+			    HW_ATL_TPS_TX_DESC_RATE_MODE_MSK,
+			    HW_ATL_TPS_TX_DESC_RATE_MODE_SHIFT,
+			    rate_mode);
 }
 
-void tps_tx_pkt_shed_tc_data_weight_set(struct aq_hw_s *aq_hw,
-					u32 tx_pkt_shed_tc_data_weight, u32 tc)
+void hw_atl_tps_tx_desc_rate_en_set(struct aq_hw_s *aq_hw, const u32 desc,
+				    const u32 enable)
 {
-	aq_hw_write_reg_bit(aq_hw, tps_data_tctweight_adr(tc),
-			    tps_data_tctweight_msk,
-			    tps_data_tctweight_shift,
-			    tx_pkt_shed_tc_data_weight);
+	aq_hw_write_reg_bit(aq_hw, HW_ATL_TPS_DESC_RATE_EN_ADR(desc),
+			    HW_ATL_TPS_DESC_RATE_EN_MSK,
+			    HW_ATL_TPS_DESC_RATE_EN_SHIFT,
+			    enable);
+}
+
+void hw_atl_tps_tx_desc_rate_x_set(struct aq_hw_s *aq_hw, const u32 desc,
+				   const u32 rate_int)
+{
+	aq_hw_write_reg_bit(aq_hw, HW_ATL_TPS_DESC_RATE_X_ADR(desc),
+			    HW_ATL_TPS_DESC_RATE_X_MSK,
+			    HW_ATL_TPS_DESC_RATE_X_SHIFT,
+			    rate_int);
+}
+
+void hw_atl_tps_tx_desc_rate_y_set(struct aq_hw_s *aq_hw, const u32 desc,
+				   const u32 rate_frac)
+{
+	aq_hw_write_reg_bit(aq_hw, HW_ATL_TPS_DESC_RATE_Y_ADR(desc),
+			    HW_ATL_TPS_DESC_RATE_Y_MSK,
+			    HW_ATL_TPS_DESC_RATE_Y_SHIFT,
+			    rate_frac);
 }
 
 /* tx */
-void tx_tx_reg_res_dis_set(struct aq_hw_s *aq_hw, u32 tx_reg_res_dis)
+void hw_atl_tx_tx_reg_res_dis_set(struct aq_hw_s *aq_hw, u32 tx_reg_res_dis)
 {
-	aq_hw_write_reg_bit(aq_hw, tx_reg_res_dsbl_adr,
-			    tx_reg_res_dsbl_msk,
-			    tx_reg_res_dsbl_shift, tx_reg_res_dis);
+	aq_hw_write_reg_bit(aq_hw, HW_ATL_TX_REG_RES_DSBL_ADR,
+			    HW_ATL_TX_REG_RES_DSBL_MSK,
+			    HW_ATL_TX_REG_RES_DSBL_SHIFT, tx_reg_res_dis);
 }
 
 /* msm */
-u32 msm_reg_access_status_get(struct aq_hw_s *aq_hw)
+u32 hw_atl_msm_reg_access_status_get(struct aq_hw_s *aq_hw)
 {
-	return aq_hw_read_reg_bit(aq_hw, msm_reg_access_busy_adr,
-				  msm_reg_access_busy_msk,
-				  msm_reg_access_busy_shift);
+	return aq_hw_read_reg_bit(aq_hw, HW_ATL_MSM_REG_ACCESS_BUSY_ADR,
+				  HW_ATL_MSM_REG_ACCESS_BUSY_MSK,
+				  HW_ATL_MSM_REG_ACCESS_BUSY_SHIFT);
 }
 
-void msm_reg_addr_for_indirect_addr_set(struct aq_hw_s *aq_hw,
-					u32 reg_addr_for_indirect_addr)
+void hw_atl_msm_reg_addr_for_indirect_addr_set(struct aq_hw_s *aq_hw,
+					       u32 reg_addr_for_indirect_addr)
 {
-	aq_hw_write_reg_bit(aq_hw, msm_reg_addr_adr,
-			    msm_reg_addr_msk,
-			    msm_reg_addr_shift,
+	aq_hw_write_reg_bit(aq_hw, HW_ATL_MSM_REG_ADDR_ADR,
+			    HW_ATL_MSM_REG_ADDR_MSK,
+			    HW_ATL_MSM_REG_ADDR_SHIFT,
 			    reg_addr_for_indirect_addr);
 }
 
-void msm_reg_rd_strobe_set(struct aq_hw_s *aq_hw, u32 reg_rd_strobe)
+void hw_atl_msm_reg_rd_strobe_set(struct aq_hw_s *aq_hw, u32 reg_rd_strobe)
 {
-	aq_hw_write_reg_bit(aq_hw, msm_reg_rd_strobe_adr,
-			    msm_reg_rd_strobe_msk,
-			    msm_reg_rd_strobe_shift,
+	aq_hw_write_reg_bit(aq_hw, HW_ATL_MSM_REG_RD_STROBE_ADR,
+			    HW_ATL_MSM_REG_RD_STROBE_MSK,
+			    HW_ATL_MSM_REG_RD_STROBE_SHIFT,
 			    reg_rd_strobe);
 }
 
-u32 msm_reg_rd_data_get(struct aq_hw_s *aq_hw)
+u32 hw_atl_msm_reg_rd_data_get(struct aq_hw_s *aq_hw)
 {
-	return aq_hw_read_reg(aq_hw, msm_reg_rd_data_adr);
+	return aq_hw_read_reg(aq_hw, HW_ATL_MSM_REG_RD_DATA_ADR);
 }
 
-void msm_reg_wr_data_set(struct aq_hw_s *aq_hw, u32 reg_wr_data)
+void hw_atl_msm_reg_wr_data_set(struct aq_hw_s *aq_hw, u32 reg_wr_data)
 {
-	aq_hw_write_reg(aq_hw, msm_reg_wr_data_adr, reg_wr_data);
+	aq_hw_write_reg(aq_hw, HW_ATL_MSM_REG_WR_DATA_ADR, reg_wr_data);
 }
 
-void msm_reg_wr_strobe_set(struct aq_hw_s *aq_hw, u32 reg_wr_strobe)
+void hw_atl_msm_reg_wr_strobe_set(struct aq_hw_s *aq_hw, u32 reg_wr_strobe)
 {
-	aq_hw_write_reg_bit(aq_hw, msm_reg_wr_strobe_adr,
-			    msm_reg_wr_strobe_msk,
-			    msm_reg_wr_strobe_shift,
+	aq_hw_write_reg_bit(aq_hw, HW_ATL_MSM_REG_WR_STROBE_ADR,
+			    HW_ATL_MSM_REG_WR_STROBE_MSK,
+			    HW_ATL_MSM_REG_WR_STROBE_SHIFT,
 			    reg_wr_strobe);
 }
 
 /* pci */
-void pci_pci_reg_res_dis_set(struct aq_hw_s *aq_hw, u32 pci_reg_res_dis)
+void hw_atl_pci_pci_reg_res_dis_set(struct aq_hw_s *aq_hw, u32 pci_reg_res_dis)
 {
-	aq_hw_write_reg_bit(aq_hw, pci_reg_res_dsbl_adr,
-			    pci_reg_res_dsbl_msk,
-			    pci_reg_res_dsbl_shift,
+	aq_hw_write_reg_bit(aq_hw, HW_ATL_PCI_REG_RES_DSBL_ADR,
+			    HW_ATL_PCI_REG_RES_DSBL_MSK,
+			    HW_ATL_PCI_REG_RES_DSBL_SHIFT,
 			    pci_reg_res_dis);
 }
 
-void reg_glb_cpu_scratch_scp_set(struct aq_hw_s *aq_hw, u32 glb_cpu_scratch_scp,
-				 u32 scratch_scp)
+void hw_atl_reg_glb_cpu_scratch_scp_set(struct aq_hw_s *aq_hw,
+					u32 glb_cpu_scratch_scp,
+					u32 scratch_scp)
 {
-	aq_hw_write_reg(aq_hw, glb_cpu_scratch_scp_adr(scratch_scp),
+	aq_hw_write_reg(aq_hw, HW_ATL_GLB_CPU_SCRATCH_SCP_ADR(scratch_scp),
 			glb_cpu_scratch_scp);
 }
+
+void hw_atl_pcs_ptp_clock_read_enable(struct aq_hw_s *aq_hw,
+				      u32 ptp_clock_read_enable)
+{
+	aq_hw_write_reg_bit(aq_hw, HW_ATL_PCS_PTP_CLOCK_READ_ENABLE_ADR,
+			    HW_ATL_PCS_PTP_CLOCK_READ_ENABLE_MSK,
+			    HW_ATL_PCS_PTP_CLOCK_READ_ENABLE_SHIFT,
+			    ptp_clock_read_enable);
+}
+
+u32 hw_atl_pcs_ptp_clock_get(struct aq_hw_s *aq_hw, u32 index)
+{
+	return aq_hw_read_reg(aq_hw, HW_ATL_PCS_PTP_TS_VAL_ADDR(index));
+}
+
+void hw_atl_mcp_up_force_intr_set(struct aq_hw_s *aq_hw, u32 up_force_intr)
+{
+	aq_hw_write_reg_bit(aq_hw, HW_ATL_MCP_UP_FORCE_INTERRUPT_ADR,
+			    HW_ATL_MCP_UP_FORCE_INTERRUPT_MSK,
+			    HW_ATL_MCP_UP_FORCE_INTERRUPT_SHIFT,
+			    up_force_intr);
+}
+
+void hw_atl_rpfl3l4_ipv4_dest_addr_clear(struct aq_hw_s *aq_hw, u8 location)
+{
+	aq_hw_write_reg(aq_hw, HW_ATL_RPF_L3_DSTA_ADR(location), 0U);
+}
+
+void hw_atl_rpfl3l4_ipv4_src_addr_clear(struct aq_hw_s *aq_hw, u8 location)
+{
+	aq_hw_write_reg(aq_hw, HW_ATL_RPF_L3_SRCA_ADR(location), 0U);
+}
+
+void hw_atl_rpfl3l4_cmd_clear(struct aq_hw_s *aq_hw, u8 location)
+{
+	aq_hw_write_reg(aq_hw, HW_ATL_RPF_L3_REG_CTRL_ADR(location), 0U);
+}
+
+void hw_atl_rpfl3l4_ipv6_dest_addr_clear(struct aq_hw_s *aq_hw, u8 location)
+{
+	int i;
+
+	for (i = 0; i < 4; ++i)
+		aq_hw_write_reg(aq_hw,
+				HW_ATL_RPF_L3_DSTA_ADR(location + i),
+				0U);
+}
+
+void hw_atl_rpfl3l4_ipv6_src_addr_clear(struct aq_hw_s *aq_hw, u8 location)
+{
+	int i;
+
+	for (i = 0; i < 4; ++i)
+		aq_hw_write_reg(aq_hw,
+				HW_ATL_RPF_L3_SRCA_ADR(location + i),
+				0U);
+}
+
+void hw_atl_rpfl3l4_ipv4_dest_addr_set(struct aq_hw_s *aq_hw, u8 location,
+				       u32 ipv4_dest)
+{
+	aq_hw_write_reg(aq_hw, HW_ATL_RPF_L3_DSTA_ADR(location),
+			ipv4_dest);
+}
+
+void hw_atl_rpfl3l4_ipv4_src_addr_set(struct aq_hw_s *aq_hw, u8 location,
+				      u32 ipv4_src)
+{
+	aq_hw_write_reg(aq_hw,
+			HW_ATL_RPF_L3_SRCA_ADR(location),
+			ipv4_src);
+}
+
+void hw_atl_rpfl3l4_cmd_set(struct aq_hw_s *aq_hw, u8 location, u32 cmd)
+{
+	aq_hw_write_reg(aq_hw, HW_ATL_RPF_L3_REG_CTRL_ADR(location), cmd);
+}
+
+void hw_atl_rpfl3l4_ipv6_src_addr_set(struct aq_hw_s *aq_hw, u8 location,
+				      u32 *ipv6_src)
+{
+	int i;
+
+	for (i = 0; i < 4; ++i)
+		aq_hw_write_reg(aq_hw,
+				HW_ATL_RPF_L3_SRCA_ADR(location + i),
+				ipv6_src[3 - i]);
+}
+
+void hw_atl_rpfl3l4_ipv6_dest_addr_set(struct aq_hw_s *aq_hw, u8 location,
+				       u32 *ipv6_dest)
+{
+	int i;
+
+	for (i = 0; i < 4; ++i)
+		aq_hw_write_reg(aq_hw,
+				HW_ATL_RPF_L3_DSTA_ADR(location + i),
+				ipv6_dest[3 - i]);
+}
+
+u32 hw_atl_sem_ram_get(struct aq_hw_s *self)
+{
+	return hw_atl_reg_glb_cpu_sem_get(self, HW_ATL_FW_SM_RAM);
+}
+
+u32 hw_atl_sem_mdio_get(struct aq_hw_s *self)
+{
+	return hw_atl_reg_glb_cpu_sem_get(self, HW_ATL_FW_SM_MDIO);
+}
+
+u32 hw_atl_sem_reset1_get(struct aq_hw_s *self)
+{
+	return hw_atl_reg_glb_cpu_sem_get(self, HW_ATL_FW_SM_RESET1);
+}
+
+u32 hw_atl_sem_reset2_get(struct aq_hw_s *self)
+{
+	return hw_atl_reg_glb_cpu_sem_get(self, HW_ATL_FW_SM_RESET2);
+}
+
+u32 hw_atl_scrpad_get(struct aq_hw_s *aq_hw, u32 scratch_scp)
+{
+	return aq_hw_read_reg(aq_hw,
+			      HW_ATL_GLB_CPU_SCRATCH_SCP_ADR(scratch_scp));
+}
+
+u32 hw_atl_scrpad12_get(struct aq_hw_s *self)
+{
+	return  hw_atl_scrpad_get(self, 0xB);
+}
+
+u32 hw_atl_scrpad25_get(struct aq_hw_s *self)
+{
+	return hw_atl_scrpad_get(self, 0x18);
+}
+
+void hw_atl_glb_mdio_iface1_set(struct aq_hw_s *aq_hw, u32 value)
+{
+	aq_hw_write_reg(aq_hw, HW_ATL_GLB_MDIO_IFACE_N_ADR(1), value);
+}
+
+u32 hw_atl_glb_mdio_iface1_get(struct aq_hw_s *aq_hw)
+{
+	return aq_hw_read_reg(aq_hw, HW_ATL_GLB_MDIO_IFACE_N_ADR(1));
+}
+
+void hw_atl_glb_mdio_iface2_set(struct aq_hw_s *aq_hw, u32 value)
+{
+	aq_hw_write_reg(aq_hw, HW_ATL_GLB_MDIO_IFACE_N_ADR(2), value);
+}
+
+u32 hw_atl_glb_mdio_iface2_get(struct aq_hw_s *aq_hw)
+{
+	return aq_hw_read_reg(aq_hw, HW_ATL_GLB_MDIO_IFACE_N_ADR(2));
+}
+
+void hw_atl_glb_mdio_iface3_set(struct aq_hw_s *aq_hw, u32 value)
+{
+	aq_hw_write_reg(aq_hw, HW_ATL_GLB_MDIO_IFACE_N_ADR(3), value);
+}
+
+u32 hw_atl_glb_mdio_iface3_get(struct aq_hw_s *aq_hw)
+{
+	return aq_hw_read_reg(aq_hw, HW_ATL_GLB_MDIO_IFACE_N_ADR(3));
+}
+
+void hw_atl_glb_mdio_iface4_set(struct aq_hw_s *aq_hw, u32 value)
+{
+	aq_hw_write_reg(aq_hw, HW_ATL_GLB_MDIO_IFACE_N_ADR(4), value);
+}
+
+u32 hw_atl_glb_mdio_iface4_get(struct aq_hw_s *aq_hw)
+{
+	return aq_hw_read_reg(aq_hw, HW_ATL_GLB_MDIO_IFACE_N_ADR(4));
+}
+
+void hw_atl_glb_mdio_iface5_set(struct aq_hw_s *aq_hw, u32 value)
+{
+	aq_hw_write_reg(aq_hw, HW_ATL_GLB_MDIO_IFACE_N_ADR(5), value);
+}
+
+u32 hw_atl_glb_mdio_iface5_get(struct aq_hw_s *aq_hw)
+{
+	return aq_hw_read_reg(aq_hw, HW_ATL_GLB_MDIO_IFACE_N_ADR(5));
+}
+
+u32 hw_atl_mdio_busy_get(struct aq_hw_s *aq_hw)
+{
+	return aq_hw_read_reg_bit(aq_hw, HW_ATL_MDIO_BUSY_ADR,
+				  HW_ATL_MDIO_BUSY_MSK,
+				  HW_ATL_MDIO_BUSY_SHIFT);
+}
diff --git a/drivers/net/ethernet/aquantia/atlantic/hw_atl/hw_atl_llh.h b/drivers/net/ethernet/aquantia/atlantic/hw_atl/hw_atl_llh.h
index ed1085b95adb..5fd506acacb5 100644
--- a/drivers/net/ethernet/aquantia/atlantic/hw_atl/hw_atl_llh.h
+++ b/drivers/net/ethernet/aquantia/atlantic/hw_atl/hw_atl_llh.h
@@ -1,10 +1,8 @@
-/*
- * aQuantia Corporation Network Driver
- * Copyright (C) 2014-2017 aQuantia Corporation. All rights reserved
+/* SPDX-License-Identifier: GPL-2.0-only */
+/* Atlantic Network Driver
  *
- * This program is free software; you can redistribute it and/or modify it
- * under the terms and conditions of the GNU General Public License,
- * version 2, as published by the Free Software Foundation.
+ * Copyright (C) 2014-2019 aQuantia Corporation
+ * Copyright (C) 2019-2020 Marvell International Ltd.
  */
 
 /* File hw_atl_llh.h: Declarations of bitfield and register access functions for
@@ -18,660 +16,878 @@
 
 struct aq_hw_s;
 
+/* set temperature sense reset */
+void hw_atl_ts_reset_set(struct aq_hw_s *aq_hw, u32 val);
+
+/* set temperature sense power down */
+void hw_atl_ts_power_down_set(struct aq_hw_s *aq_hw, u32 val);
+
+/* get temperature sense power down */
+u32 hw_atl_ts_power_down_get(struct aq_hw_s *aq_hw);
+
+/* get temperature sense ready */
+u32 hw_atl_ts_ready_get(struct aq_hw_s *aq_hw);
+
+/* get temperature sense ready latch high */
+u32 hw_atl_ts_ready_latch_high_get(struct aq_hw_s *aq_hw);
+
+/* get temperature sense data */
+u32 hw_atl_ts_data_get(struct aq_hw_s *aq_hw);
+
+/* SMBUS0 bus busy */
+u32 hw_atl_smb0_bus_busy_get(struct aq_hw_s *aq_hw);
+
+/* SMBUS0 byte transfer complete */
+u32 hw_atl_smb0_byte_transfer_complete_get(struct aq_hw_s *aq_hw);
+
+/* SMBUS0 receive acknowledged */
+u32 hw_atl_smb0_receive_acknowledged_get(struct aq_hw_s *aq_hw);
+
+/* SMBUS0 set transmitted data (only leftmost byte of data valid) */
+void hw_atl_smb0_tx_data_set(struct aq_hw_s *aq_hw, u32 data);
+
+/* SMBUS0 provisioning2 command register */
+void hw_atl_smb0_provisioning2_set(struct aq_hw_s *aq_hw, u32 data);
+
+/* SMBUS0 repeated start detect */
+u32 hw_atl_smb0_repeated_start_detect_get(struct aq_hw_s *aq_hw);
+
+/* SMBUS0 received data register */
+u32 hw_atl_smb0_rx_data_get(struct aq_hw_s *aq_hw);
+
 /* global */
 
 /* set global microprocessor semaphore */
-void reg_glb_cpu_sem_set(struct aq_hw_s *aq_hw,	u32 glb_cpu_sem,
-			 u32 semaphore);
+void hw_atl_reg_glb_cpu_sem_set(struct aq_hw_s *aq_hw,	u32 glb_cpu_sem,
+				u32 semaphore);
 
 /* get global microprocessor semaphore */
-u32 reg_glb_cpu_sem_get(struct aq_hw_s *aq_hw, u32 semaphore);
+u32 hw_atl_reg_glb_cpu_sem_get(struct aq_hw_s *aq_hw, u32 semaphore);
 
 /* set global register reset disable */
-void glb_glb_reg_res_dis_set(struct aq_hw_s *aq_hw, u32 glb_reg_res_dis);
+void hw_atl_glb_glb_reg_res_dis_set(struct aq_hw_s *aq_hw, u32 glb_reg_res_dis);
 
 /* set soft reset */
-void glb_soft_res_set(struct aq_hw_s *aq_hw, u32 soft_res);
+void hw_atl_glb_soft_res_set(struct aq_hw_s *aq_hw, u32 soft_res);
 
 /* get soft reset */
-u32 glb_soft_res_get(struct aq_hw_s *aq_hw);
+u32 hw_atl_glb_soft_res_get(struct aq_hw_s *aq_hw);
 
 /* stats */
 
-u32 rpb_rx_dma_drop_pkt_cnt_get(struct aq_hw_s *aq_hw);
-
-/* get rx dma good octet counter lsw */
-u32 stats_rx_dma_good_octet_counterlsw_get(struct aq_hw_s *aq_hw);
-
-/* get rx dma good packet counter lsw */
-u32 stats_rx_dma_good_pkt_counterlsw_get(struct aq_hw_s *aq_hw);
-
-/* get tx dma good octet counter lsw */
-u32 stats_tx_dma_good_octet_counterlsw_get(struct aq_hw_s *aq_hw);
+u32 hw_atl_rpb_rx_dma_drop_pkt_cnt_get(struct aq_hw_s *aq_hw);
 
-/* get tx dma good packet counter lsw */
-u32 stats_tx_dma_good_pkt_counterlsw_get(struct aq_hw_s *aq_hw);
+/* get rx dma good octet counter */
+u64 hw_atl_stats_rx_dma_good_octet_counter_get(struct aq_hw_s *aq_hw);
 
-/* get rx dma good octet counter msw */
-u32 stats_rx_dma_good_octet_countermsw_get(struct aq_hw_s *aq_hw);
+/* get rx dma good packet counter */
+u64 hw_atl_stats_rx_dma_good_pkt_counter_get(struct aq_hw_s *aq_hw);
 
-/* get rx dma good packet counter msw */
-u32 stats_rx_dma_good_pkt_countermsw_get(struct aq_hw_s *aq_hw);
+/* get tx dma good octet counter */
+u64 hw_atl_stats_tx_dma_good_octet_counter_get(struct aq_hw_s *aq_hw);
 
-/* get tx dma good octet counter msw */
-u32 stats_tx_dma_good_octet_countermsw_get(struct aq_hw_s *aq_hw);
-
-/* get tx dma good packet counter msw */
-u32 stats_tx_dma_good_pkt_countermsw_get(struct aq_hw_s *aq_hw);
+/* get tx dma good packet counter */
+u64 hw_atl_stats_tx_dma_good_pkt_counter_get(struct aq_hw_s *aq_hw);
 
 /* get msm rx errors counter register */
-u32 reg_mac_msm_rx_errs_cnt_get(struct aq_hw_s *aq_hw);
+u32 hw_atl_reg_mac_msm_rx_errs_cnt_get(struct aq_hw_s *aq_hw);
 
 /* get msm rx unicast frames counter register */
-u32 reg_mac_msm_rx_ucst_frm_cnt_get(struct aq_hw_s *aq_hw);
+u32 hw_atl_reg_mac_msm_rx_ucst_frm_cnt_get(struct aq_hw_s *aq_hw);
 
 /* get msm rx multicast frames counter register */
-u32 reg_mac_msm_rx_mcst_frm_cnt_get(struct aq_hw_s *aq_hw);
+u32 hw_atl_reg_mac_msm_rx_mcst_frm_cnt_get(struct aq_hw_s *aq_hw);
 
 /* get msm rx broadcast frames counter register */
-u32 reg_mac_msm_rx_bcst_frm_cnt_get(struct aq_hw_s *aq_hw);
+u32 hw_atl_reg_mac_msm_rx_bcst_frm_cnt_get(struct aq_hw_s *aq_hw);
 
 /* get msm rx broadcast octets counter register 1 */
-u32 reg_mac_msm_rx_bcst_octets_counter1get(struct aq_hw_s *aq_hw);
+u32 hw_atl_reg_mac_msm_rx_bcst_octets_counter1get(struct aq_hw_s *aq_hw);
 
 /* get msm rx unicast octets counter register 0 */
-u32 reg_mac_msm_rx_ucst_octets_counter0get(struct aq_hw_s *aq_hw);
-
-/* get rx dma statistics counter 7 */
-u32 reg_rx_dma_stat_counter7get(struct aq_hw_s *aq_hw);
+u32 hw_atl_reg_mac_msm_rx_ucst_octets_counter0get(struct aq_hw_s *aq_hw);
 
 /* get msm tx errors counter register */
-u32 reg_mac_msm_tx_errs_cnt_get(struct aq_hw_s *aq_hw);
+u32 hw_atl_reg_mac_msm_tx_errs_cnt_get(struct aq_hw_s *aq_hw);
 
 /* get msm tx unicast frames counter register */
-u32 reg_mac_msm_tx_ucst_frm_cnt_get(struct aq_hw_s *aq_hw);
+u32 hw_atl_reg_mac_msm_tx_ucst_frm_cnt_get(struct aq_hw_s *aq_hw);
 
 /* get msm tx multicast frames counter register */
-u32 reg_mac_msm_tx_mcst_frm_cnt_get(struct aq_hw_s *aq_hw);
+u32 hw_atl_reg_mac_msm_tx_mcst_frm_cnt_get(struct aq_hw_s *aq_hw);
 
 /* get msm tx broadcast frames counter register */
-u32 reg_mac_msm_tx_bcst_frm_cnt_get(struct aq_hw_s *aq_hw);
+u32 hw_atl_reg_mac_msm_tx_bcst_frm_cnt_get(struct aq_hw_s *aq_hw);
 
 /* get msm tx multicast octets counter register 1 */
-u32 reg_mac_msm_tx_mcst_octets_counter1get(struct aq_hw_s *aq_hw);
+u32 hw_atl_reg_mac_msm_tx_mcst_octets_counter1get(struct aq_hw_s *aq_hw);
 
 /* get msm tx broadcast octets counter register 1 */
-u32 reg_mac_msm_tx_bcst_octets_counter1get(struct aq_hw_s *aq_hw);
+u32 hw_atl_reg_mac_msm_tx_bcst_octets_counter1get(struct aq_hw_s *aq_hw);
 
 /* get msm tx unicast octets counter register 0 */
-u32 reg_mac_msm_tx_ucst_octets_counter0get(struct aq_hw_s *aq_hw);
+u32 hw_atl_reg_mac_msm_tx_ucst_octets_counter0get(struct aq_hw_s *aq_hw);
 
 /* get global mif identification */
-u32 reg_glb_mif_id_get(struct aq_hw_s *aq_hw);
+u32 hw_atl_reg_glb_mif_id_get(struct aq_hw_s *aq_hw);
 
 /* interrupt */
 
 /* set interrupt auto mask lsw */
-void itr_irq_auto_masklsw_set(struct aq_hw_s *aq_hw, u32 irq_auto_masklsw);
+void hw_atl_itr_irq_auto_masklsw_set(struct aq_hw_s *aq_hw,
+				     u32 irq_auto_masklsw);
 
 /* set interrupt mapping enable rx */
-void itr_irq_map_en_rx_set(struct aq_hw_s *aq_hw, u32 irq_map_en_rx, u32 rx);
+void hw_atl_itr_irq_map_en_rx_set(struct aq_hw_s *aq_hw, u32 irq_map_en_rx,
+				  u32 rx);
 
 /* set interrupt mapping enable tx */
-void itr_irq_map_en_tx_set(struct aq_hw_s *aq_hw, u32 irq_map_en_tx, u32 tx);
+void hw_atl_itr_irq_map_en_tx_set(struct aq_hw_s *aq_hw, u32 irq_map_en_tx,
+				  u32 tx);
 
 /* set interrupt mapping rx */
-void itr_irq_map_rx_set(struct aq_hw_s *aq_hw, u32 irq_map_rx, u32 rx);
+void hw_atl_itr_irq_map_rx_set(struct aq_hw_s *aq_hw, u32 irq_map_rx, u32 rx);
 
 /* set interrupt mapping tx */
-void itr_irq_map_tx_set(struct aq_hw_s *aq_hw, u32 irq_map_tx, u32 tx);
+void hw_atl_itr_irq_map_tx_set(struct aq_hw_s *aq_hw, u32 irq_map_tx, u32 tx);
 
 /* set interrupt mask clear lsw */
-void itr_irq_msk_clearlsw_set(struct aq_hw_s *aq_hw, u32 irq_msk_clearlsw);
+void hw_atl_itr_irq_msk_clearlsw_set(struct aq_hw_s *aq_hw,
+				     u32 irq_msk_clearlsw);
 
 /* set interrupt mask set lsw */
-void itr_irq_msk_setlsw_set(struct aq_hw_s *aq_hw, u32 irq_msk_setlsw);
+void hw_atl_itr_irq_msk_setlsw_set(struct aq_hw_s *aq_hw, u32 irq_msk_setlsw);
 
 /* set interrupt register reset disable */
-void itr_irq_reg_res_dis_set(struct aq_hw_s *aq_hw, u32 irq_reg_res_dis);
+void hw_atl_itr_irq_reg_res_dis_set(struct aq_hw_s *aq_hw, u32 irq_reg_res_dis);
 
 /* set interrupt status clear lsw */
-void itr_irq_status_clearlsw_set(struct aq_hw_s *aq_hw,
-				 u32 irq_status_clearlsw);
+void hw_atl_itr_irq_status_clearlsw_set(struct aq_hw_s *aq_hw,
+					u32 irq_status_clearlsw);
 
 /* get interrupt status lsw */
-u32 itr_irq_statuslsw_get(struct aq_hw_s *aq_hw);
+u32 hw_atl_itr_irq_statuslsw_get(struct aq_hw_s *aq_hw);
 
 /* get reset interrupt */
-u32 itr_res_irq_get(struct aq_hw_s *aq_hw);
+u32 hw_atl_itr_res_irq_get(struct aq_hw_s *aq_hw);
 
 /* set reset interrupt */
-void itr_res_irq_set(struct aq_hw_s *aq_hw, u32 res_irq);
+void hw_atl_itr_res_irq_set(struct aq_hw_s *aq_hw, u32 res_irq);
+
+/* set RSC interrupt */
+void hw_atl_itr_rsc_en_set(struct aq_hw_s *aq_hw, u32 enable);
+
+/* set RSC delay */
+void hw_atl_itr_rsc_delay_set(struct aq_hw_s *aq_hw, u32 delay);
 
 /* rdm */
 
 /* set cpu id */
-void rdm_cpu_id_set(struct aq_hw_s *aq_hw, u32 cpuid, u32 dca);
+void hw_atl_rdm_cpu_id_set(struct aq_hw_s *aq_hw, u32 cpuid, u32 dca);
 
 /* set rx dca enable */
-void rdm_rx_dca_en_set(struct aq_hw_s *aq_hw, u32 rx_dca_en);
+void hw_atl_rdm_rx_dca_en_set(struct aq_hw_s *aq_hw, u32 rx_dca_en);
 
 /* set rx dca mode */
-void rdm_rx_dca_mode_set(struct aq_hw_s *aq_hw, u32 rx_dca_mode);
+void hw_atl_rdm_rx_dca_mode_set(struct aq_hw_s *aq_hw, u32 rx_dca_mode);
 
 /* set rx descriptor data buffer size */
-void rdm_rx_desc_data_buff_size_set(struct aq_hw_s *aq_hw,
-				    u32 rx_desc_data_buff_size,
+void hw_atl_rdm_rx_desc_data_buff_size_set(struct aq_hw_s *aq_hw,
+					   u32 rx_desc_data_buff_size,
 				    u32 descriptor);
 
 /* set rx descriptor dca enable */
-void rdm_rx_desc_dca_en_set(struct aq_hw_s *aq_hw, u32 rx_desc_dca_en,
-			    u32 dca);
+void hw_atl_rdm_rx_desc_dca_en_set(struct aq_hw_s *aq_hw, u32 rx_desc_dca_en,
+				   u32 dca);
 
 /* set rx descriptor enable */
-void rdm_rx_desc_en_set(struct aq_hw_s *aq_hw, u32 rx_desc_en,
-			u32 descriptor);
+void hw_atl_rdm_rx_desc_en_set(struct aq_hw_s *aq_hw, u32 rx_desc_en,
+			       u32 descriptor);
 
 /* set rx descriptor header splitting */
-void rdm_rx_desc_head_splitting_set(struct aq_hw_s *aq_hw,
-				    u32 rx_desc_head_splitting,
+void hw_atl_rdm_rx_desc_head_splitting_set(struct aq_hw_s *aq_hw,
+					   u32 rx_desc_head_splitting,
 				    u32 descriptor);
 
 /* get rx descriptor head pointer */
-u32 rdm_rx_desc_head_ptr_get(struct aq_hw_s *aq_hw, u32 descriptor);
+u32 hw_atl_rdm_rx_desc_head_ptr_get(struct aq_hw_s *aq_hw, u32 descriptor);
 
 /* set rx descriptor length */
-void rdm_rx_desc_len_set(struct aq_hw_s *aq_hw, u32 rx_desc_len,
-			 u32 descriptor);
+void hw_atl_rdm_rx_desc_len_set(struct aq_hw_s *aq_hw, u32 rx_desc_len,
+				u32 descriptor);
 
 /* set rx descriptor write-back interrupt enable */
-void rdm_rx_desc_wr_wb_irq_en_set(struct aq_hw_s *aq_hw,
-				  u32 rx_desc_wr_wb_irq_en);
+void hw_atl_rdm_rx_desc_wr_wb_irq_en_set(struct aq_hw_s *aq_hw,
+					 u32 rx_desc_wr_wb_irq_en);
 
 /* set rx header dca enable */
-void rdm_rx_head_dca_en_set(struct aq_hw_s *aq_hw, u32 rx_head_dca_en,
-			    u32 dca);
+void hw_atl_rdm_rx_head_dca_en_set(struct aq_hw_s *aq_hw, u32 rx_head_dca_en,
+				   u32 dca);
 
 /* set rx payload dca enable */
-void rdm_rx_pld_dca_en_set(struct aq_hw_s *aq_hw, u32 rx_pld_dca_en, u32 dca);
+void hw_atl_rdm_rx_pld_dca_en_set(struct aq_hw_s *aq_hw, u32 rx_pld_dca_en,
+				  u32 dca);
 
 /* set rx descriptor header buffer size */
-void rdm_rx_desc_head_buff_size_set(struct aq_hw_s *aq_hw,
-				    u32 rx_desc_head_buff_size,
-				    u32 descriptor);
+void hw_atl_rdm_rx_desc_head_buff_size_set(struct aq_hw_s *aq_hw,
+					   u32 rx_desc_head_buff_size,
+					   u32 descriptor);
 
 /* set rx descriptor reset */
-void rdm_rx_desc_res_set(struct aq_hw_s *aq_hw, u32 rx_desc_res,
-			 u32 descriptor);
+void hw_atl_rdm_rx_desc_res_set(struct aq_hw_s *aq_hw, u32 rx_desc_res,
+				u32 descriptor);
 
 /* Set RDM Interrupt Moderation Enable */
-void rdm_rdm_intr_moder_en_set(struct aq_hw_s *aq_hw, u32 rdm_intr_moder_en);
+void hw_atl_rdm_rdm_intr_moder_en_set(struct aq_hw_s *aq_hw,
+				      u32 rdm_intr_moder_en);
 
 /* reg */
 
 /* set general interrupt mapping register */
-void reg_gen_irq_map_set(struct aq_hw_s *aq_hw, u32 gen_intr_map, u32 regidx);
+void hw_atl_reg_gen_irq_map_set(struct aq_hw_s *aq_hw, u32 gen_intr_map,
+				u32 regidx);
 
 /* get general interrupt status register */
-u32 reg_gen_irq_status_get(struct aq_hw_s *aq_hw);
+u32 hw_atl_reg_gen_irq_status_get(struct aq_hw_s *aq_hw);
 
 /* set interrupt global control register */
-void reg_irq_glb_ctl_set(struct aq_hw_s *aq_hw, u32 intr_glb_ctl);
+void hw_atl_reg_irq_glb_ctl_set(struct aq_hw_s *aq_hw, u32 intr_glb_ctl);
 
 /* set interrupt throttle register */
-void reg_irq_thr_set(struct aq_hw_s *aq_hw, u32 intr_thr, u32 throttle);
+void hw_atl_reg_irq_thr_set(struct aq_hw_s *aq_hw, u32 intr_thr, u32 throttle);
 
 /* set rx dma descriptor base address lsw */
-void reg_rx_dma_desc_base_addresslswset(struct aq_hw_s *aq_hw,
-					u32 rx_dma_desc_base_addrlsw,
+void hw_atl_reg_rx_dma_desc_base_addresslswset(struct aq_hw_s *aq_hw,
+					       u32 rx_dma_desc_base_addrlsw,
 					u32 descriptor);
 
 /* set rx dma descriptor base address msw */
-void reg_rx_dma_desc_base_addressmswset(struct aq_hw_s *aq_hw,
-					u32 rx_dma_desc_base_addrmsw,
+void hw_atl_reg_rx_dma_desc_base_addressmswset(struct aq_hw_s *aq_hw,
+					       u32 rx_dma_desc_base_addrmsw,
 					u32 descriptor);
 
 /* get rx dma descriptor status register */
-u32 reg_rx_dma_desc_status_get(struct aq_hw_s *aq_hw, u32 descriptor);
+u32 hw_atl_reg_rx_dma_desc_status_get(struct aq_hw_s *aq_hw, u32 descriptor);
 
 /* set rx dma descriptor tail pointer register */
-void reg_rx_dma_desc_tail_ptr_set(struct aq_hw_s *aq_hw,
-				  u32 rx_dma_desc_tail_ptr,
+void hw_atl_reg_rx_dma_desc_tail_ptr_set(struct aq_hw_s *aq_hw,
+					 u32 rx_dma_desc_tail_ptr,
 				  u32 descriptor);
 
 /* set rx filter multicast filter mask register */
-void reg_rx_flr_mcst_flr_msk_set(struct aq_hw_s *aq_hw,
-				 u32 rx_flr_mcst_flr_msk);
+void hw_atl_reg_rx_flr_mcst_flr_msk_set(struct aq_hw_s *aq_hw,
+					u32 rx_flr_mcst_flr_msk);
 
 /* set rx filter multicast filter register */
-void reg_rx_flr_mcst_flr_set(struct aq_hw_s *aq_hw, u32 rx_flr_mcst_flr,
-			     u32 filter);
+void hw_atl_reg_rx_flr_mcst_flr_set(struct aq_hw_s *aq_hw, u32 rx_flr_mcst_flr,
+				    u32 filter);
 
 /* set rx filter rss control register 1 */
-void reg_rx_flr_rss_control1set(struct aq_hw_s *aq_hw,
-				u32 rx_flr_rss_control1);
+void hw_atl_reg_rx_flr_rss_control1set(struct aq_hw_s *aq_hw,
+				       u32 rx_flr_rss_control1);
 
 /* Set RX Filter Control Register 2 */
-void reg_rx_flr_control2_set(struct aq_hw_s *aq_hw, u32 rx_flr_control2);
+void hw_atl_reg_rx_flr_control2_set(struct aq_hw_s *aq_hw, u32 rx_flr_control2);
 
 /* Set RX Interrupt Moderation Control Register */
-void reg_rx_intr_moder_ctrl_set(struct aq_hw_s *aq_hw,
-				u32 rx_intr_moderation_ctl,
+void hw_atl_reg_rx_intr_moder_ctrl_set(struct aq_hw_s *aq_hw,
+				       u32 rx_intr_moderation_ctl,
 				u32 queue);
 
 /* set tx dma debug control */
-void reg_tx_dma_debug_ctl_set(struct aq_hw_s *aq_hw, u32 tx_dma_debug_ctl);
+void hw_atl_reg_tx_dma_debug_ctl_set(struct aq_hw_s *aq_hw,
+				     u32 tx_dma_debug_ctl);
 
 /* set tx dma descriptor base address lsw */
-void reg_tx_dma_desc_base_addresslswset(struct aq_hw_s *aq_hw,
-					u32 tx_dma_desc_base_addrlsw,
+void hw_atl_reg_tx_dma_desc_base_addresslswset(struct aq_hw_s *aq_hw,
+					       u32 tx_dma_desc_base_addrlsw,
 					u32 descriptor);
 
 /* set tx dma descriptor base address msw */
-void reg_tx_dma_desc_base_addressmswset(struct aq_hw_s *aq_hw,
-					u32 tx_dma_desc_base_addrmsw,
+void hw_atl_reg_tx_dma_desc_base_addressmswset(struct aq_hw_s *aq_hw,
+					       u32 tx_dma_desc_base_addrmsw,
 					u32 descriptor);
 
 /* set tx dma descriptor tail pointer register */
-void reg_tx_dma_desc_tail_ptr_set(struct aq_hw_s *aq_hw,
-				  u32 tx_dma_desc_tail_ptr,
-				  u32 descriptor);
+void hw_atl_reg_tx_dma_desc_tail_ptr_set(struct aq_hw_s *aq_hw,
+					 u32 tx_dma_desc_tail_ptr,
+					 u32 descriptor);
 
 /* Set TX Interrupt Moderation Control Register */
-void reg_tx_intr_moder_ctrl_set(struct aq_hw_s *aq_hw,
-				u32 tx_intr_moderation_ctl,
-				u32 queue);
+void hw_atl_reg_tx_intr_moder_ctrl_set(struct aq_hw_s *aq_hw,
+				       u32 tx_intr_moderation_ctl,
+				       u32 queue);
 
 /* set global microprocessor scratch pad */
-void reg_glb_cpu_scratch_scp_set(struct aq_hw_s *aq_hw,
-				 u32 glb_cpu_scratch_scp, u32 scratch_scp);
+void hw_atl_reg_glb_cpu_scratch_scp_set(struct aq_hw_s *aq_hw,
+					u32 glb_cpu_scratch_scp,
+					u32 scratch_scp);
 
 /* rpb */
 
 /* set dma system loopback */
-void rpb_dma_sys_lbk_set(struct aq_hw_s *aq_hw, u32 dma_sys_lbk);
+void hw_atl_rpb_dma_sys_lbk_set(struct aq_hw_s *aq_hw, u32 dma_sys_lbk);
+
+/* set dma network loopback */
+void hw_atl_rpb_dma_net_lbk_set(struct aq_hw_s *aq_hw, u32 dma_net_lbk);
 
 /* set rx traffic class mode */
-void rpb_rpf_rx_traf_class_mode_set(struct aq_hw_s *aq_hw,
-				    u32 rx_traf_class_mode);
+void hw_atl_rpb_rpf_rx_traf_class_mode_set(struct aq_hw_s *aq_hw,
+					   u32 rx_traf_class_mode);
+
+/* get rx traffic class mode */
+u32 hw_atl_rpb_rpf_rx_traf_class_mode_get(struct aq_hw_s *aq_hw);
 
 /* set rx buffer enable */
-void rpb_rx_buff_en_set(struct aq_hw_s *aq_hw, u32 rx_buff_en);
+void hw_atl_rpb_rx_buff_en_set(struct aq_hw_s *aq_hw, u32 rx_buff_en);
 
 /* set rx buffer high threshold (per tc) */
-void rpb_rx_buff_hi_threshold_per_tc_set(struct aq_hw_s *aq_hw,
-					 u32 rx_buff_hi_threshold_per_tc,
-					 u32 buffer);
+void hw_atl_rpb_rx_buff_hi_threshold_per_tc_set(struct aq_hw_s *aq_hw,
+						u32 rx_buff_hi_threshold_per_tc,
+						u32 buffer);
 
 /* set rx buffer low threshold (per tc) */
-void rpb_rx_buff_lo_threshold_per_tc_set(struct aq_hw_s *aq_hw,
-					 u32 rx_buff_lo_threshold_per_tc,
+void hw_atl_rpb_rx_buff_lo_threshold_per_tc_set(struct aq_hw_s *aq_hw,
+						u32 rx_buff_lo_threshold_per_tc,
 					 u32 buffer);
 
 /* set rx flow control mode */
-void rpb_rx_flow_ctl_mode_set(struct aq_hw_s *aq_hw, u32 rx_flow_ctl_mode);
+void hw_atl_rpb_rx_flow_ctl_mode_set(struct aq_hw_s *aq_hw,
+				     u32 rx_flow_ctl_mode);
 
 /* set rx packet buffer size (per tc) */
-void rpb_rx_pkt_buff_size_per_tc_set(struct aq_hw_s *aq_hw,
-				     u32 rx_pkt_buff_size_per_tc,
-				     u32 buffer);
+void hw_atl_rpb_rx_pkt_buff_size_per_tc_set(struct aq_hw_s *aq_hw,
+					    u32 rx_pkt_buff_size_per_tc,
+					    u32 buffer);
+
+/* toggle rdm rx dma descriptor cache init */
+void hw_atl_rdm_rx_dma_desc_cache_init_tgl(struct aq_hw_s *aq_hw);
+
+/* get rdm rx dma descriptor cache init done */
+u32 hw_atl_rdm_rx_dma_desc_cache_init_done_get(struct aq_hw_s *aq_hw);
 
 /* set rx xoff enable (per tc) */
-void rpb_rx_xoff_en_per_tc_set(struct aq_hw_s *aq_hw, u32 rx_xoff_en_per_tc,
-			       u32 buffer);
+void hw_atl_rpb_rx_xoff_en_per_tc_set(struct aq_hw_s *aq_hw,
+				      u32 rx_xoff_en_per_tc,
+				      u32 buffer);
 
 /* rpf */
 
 /* set l2 broadcast count threshold */
-void rpfl2broadcast_count_threshold_set(struct aq_hw_s *aq_hw,
-					u32 l2broadcast_count_threshold);
+void hw_atl_rpfl2broadcast_count_threshold_set(struct aq_hw_s *aq_hw,
+					       u32 l2broadcast_count_threshold);
 
 /* set l2 broadcast enable */
-void rpfl2broadcast_en_set(struct aq_hw_s *aq_hw, u32 l2broadcast_en);
+void hw_atl_rpfl2broadcast_en_set(struct aq_hw_s *aq_hw, u32 l2broadcast_en);
 
 /* set l2 broadcast filter action */
-void rpfl2broadcast_flr_act_set(struct aq_hw_s *aq_hw,
-				u32 l2broadcast_flr_act);
+void hw_atl_rpfl2broadcast_flr_act_set(struct aq_hw_s *aq_hw,
+				       u32 l2broadcast_flr_act);
 
 /* set l2 multicast filter enable */
-void rpfl2multicast_flr_en_set(struct aq_hw_s *aq_hw, u32 l2multicast_flr_en,
-			       u32 filter);
+void hw_atl_rpfl2multicast_flr_en_set(struct aq_hw_s *aq_hw,
+				      u32 l2multicast_flr_en,
+				      u32 filter);
+
+/* get l2 promiscuous mode enable */
+u32 hw_atl_rpfl2promiscuous_mode_en_get(struct aq_hw_s *aq_hw);
 
 /* set l2 promiscuous mode enable */
-void rpfl2promiscuous_mode_en_set(struct aq_hw_s *aq_hw,
-				  u32 l2promiscuous_mode_en);
+void hw_atl_rpfl2promiscuous_mode_en_set(struct aq_hw_s *aq_hw,
+					 u32 l2promiscuous_mode_en);
 
 /* set l2 unicast filter action */
-void rpfl2unicast_flr_act_set(struct aq_hw_s *aq_hw, u32 l2unicast_flr_act,
-			      u32 filter);
+void hw_atl_rpfl2unicast_flr_act_set(struct aq_hw_s *aq_hw,
+				     u32 l2unicast_flr_act,
+				     u32 filter);
 
 /* set l2 unicast filter enable */
-void rpfl2_uc_flr_en_set(struct aq_hw_s *aq_hw, u32 l2unicast_flr_en,
-			 u32 filter);
+void hw_atl_rpfl2_uc_flr_en_set(struct aq_hw_s *aq_hw, u32 l2unicast_flr_en,
+				u32 filter);
 
 /* set l2 unicast destination address lsw */
-void rpfl2unicast_dest_addresslsw_set(struct aq_hw_s *aq_hw,
-				      u32 l2unicast_dest_addresslsw,
+void hw_atl_rpfl2unicast_dest_addresslsw_set(struct aq_hw_s *aq_hw,
+					     u32 l2unicast_dest_addresslsw,
 				      u32 filter);
 
 /* set l2 unicast destination address msw */
-void rpfl2unicast_dest_addressmsw_set(struct aq_hw_s *aq_hw,
-				      u32 l2unicast_dest_addressmsw,
+void hw_atl_rpfl2unicast_dest_addressmsw_set(struct aq_hw_s *aq_hw,
+					     u32 l2unicast_dest_addressmsw,
 				      u32 filter);
 
 /* Set L2 Accept all Multicast packets */
-void rpfl2_accept_all_mc_packets_set(struct aq_hw_s *aq_hw,
-				     u32 l2_accept_all_mc_packets);
+void hw_atl_rpfl2_accept_all_mc_packets_set(struct aq_hw_s *aq_hw,
+					    u32 l2_accept_all_mc_packets);
 
 /* set user-priority tc mapping */
-void rpf_rpb_user_priority_tc_map_set(struct aq_hw_s *aq_hw,
-				      u32 user_priority_tc_map, u32 tc);
+void hw_atl_rpf_rpb_user_priority_tc_map_set(struct aq_hw_s *aq_hw,
+					     u32 user_priority_tc_map, u32 tc);
 
 /* set rss key address */
-void rpf_rss_key_addr_set(struct aq_hw_s *aq_hw, u32 rss_key_addr);
+void hw_atl_rpf_rss_key_addr_set(struct aq_hw_s *aq_hw, u32 rss_key_addr);
 
 /* set rss key write data */
-void rpf_rss_key_wr_data_set(struct aq_hw_s *aq_hw, u32 rss_key_wr_data);
+void hw_atl_rpf_rss_key_wr_data_set(struct aq_hw_s *aq_hw, u32 rss_key_wr_data);
 
 /* get rss key write enable */
-u32 rpf_rss_key_wr_en_get(struct aq_hw_s *aq_hw);
+u32 hw_atl_rpf_rss_key_wr_en_get(struct aq_hw_s *aq_hw);
 
 /* set rss key write enable */
-void rpf_rss_key_wr_en_set(struct aq_hw_s *aq_hw, u32 rss_key_wr_en);
+void hw_atl_rpf_rss_key_wr_en_set(struct aq_hw_s *aq_hw, u32 rss_key_wr_en);
 
 /* set rss redirection table address */
-void rpf_rss_redir_tbl_addr_set(struct aq_hw_s *aq_hw,
-				u32 rss_redir_tbl_addr);
+void hw_atl_rpf_rss_redir_tbl_addr_set(struct aq_hw_s *aq_hw,
+				       u32 rss_redir_tbl_addr);
 
 /* set rss redirection table write data */
-void rpf_rss_redir_tbl_wr_data_set(struct aq_hw_s *aq_hw,
-				   u32 rss_redir_tbl_wr_data);
+void hw_atl_rpf_rss_redir_tbl_wr_data_set(struct aq_hw_s *aq_hw,
+					  u32 rss_redir_tbl_wr_data);
 
 /* get rss redirection write enable */
-u32 rpf_rss_redir_wr_en_get(struct aq_hw_s *aq_hw);
+u32 hw_atl_rpf_rss_redir_wr_en_get(struct aq_hw_s *aq_hw);
 
 /* set rss redirection write enable */
-void rpf_rss_redir_wr_en_set(struct aq_hw_s *aq_hw, u32 rss_redir_wr_en);
+void hw_atl_rpf_rss_redir_wr_en_set(struct aq_hw_s *aq_hw, u32 rss_redir_wr_en);
 
 /* set tpo to rpf system loopback */
-void rpf_tpo_to_rpf_sys_lbk_set(struct aq_hw_s *aq_hw,
-				u32 tpo_to_rpf_sys_lbk);
+void hw_atl_rpf_tpo_to_rpf_sys_lbk_set(struct aq_hw_s *aq_hw,
+				       u32 tpo_to_rpf_sys_lbk);
 
 /* set vlan inner ethertype */
-void rpf_vlan_inner_etht_set(struct aq_hw_s *aq_hw, u32 vlan_inner_etht);
+void hw_atl_rpf_vlan_inner_etht_set(struct aq_hw_s *aq_hw, u32 vlan_inner_etht);
 
 /* set vlan outer ethertype */
-void rpf_vlan_outer_etht_set(struct aq_hw_s *aq_hw, u32 vlan_outer_etht);
+void hw_atl_rpf_vlan_outer_etht_set(struct aq_hw_s *aq_hw, u32 vlan_outer_etht);
 
 /* set vlan promiscuous mode enable */
-void rpf_vlan_prom_mode_en_set(struct aq_hw_s *aq_hw, u32 vlan_prom_mode_en);
+void hw_atl_rpf_vlan_prom_mode_en_set(struct aq_hw_s *aq_hw,
+				      u32 vlan_prom_mode_en);
+
+/* Get VLAN promiscuous mode enable */
+u32 hw_atl_rpf_vlan_prom_mode_en_get(struct aq_hw_s *aq_hw);
 
 /* Set VLAN untagged action */
-void rpf_vlan_untagged_act_set(struct aq_hw_s *aq_hw, u32 vlan_untagged_act);
+void hw_atl_rpf_vlan_untagged_act_set(struct aq_hw_s *aq_hw,
+				      u32 vlan_untagged_act);
 
 /* Set VLAN accept untagged packets */
-void rpf_vlan_accept_untagged_packets_set(struct aq_hw_s *aq_hw,
-					  u32 vlan_accept_untagged_packets);
+void hw_atl_rpf_vlan_accept_untagged_packets_set(struct aq_hw_s *aq_hw,
+						 u32 vlan_acc_untagged_packets);
 
 /* Set VLAN filter enable */
-void rpf_vlan_flr_en_set(struct aq_hw_s *aq_hw, u32 vlan_flr_en, u32 filter);
+void hw_atl_rpf_vlan_flr_en_set(struct aq_hw_s *aq_hw, u32 vlan_flr_en,
+				u32 filter);
 
 /* Set VLAN Filter Action */
-void rpf_vlan_flr_act_set(struct aq_hw_s *aq_hw, u32 vlan_filter_act,
-			  u32 filter);
+void hw_atl_rpf_vlan_flr_act_set(struct aq_hw_s *aq_hw, u32 vlan_filter_act,
+				 u32 filter);
 
 /* Set VLAN ID Filter */
-void rpf_vlan_id_flr_set(struct aq_hw_s *aq_hw, u32 vlan_id_flr, u32 filter);
+void hw_atl_rpf_vlan_id_flr_set(struct aq_hw_s *aq_hw, u32 vlan_id_flr,
+				u32 filter);
+
+/* Set VLAN RX queue assignment enable */
+void hw_atl_rpf_vlan_rxq_en_flr_set(struct aq_hw_s *aq_hw, u32 vlan_rxq_en,
+				    u32 filter);
+
+/* Set VLAN RX queue */
+void hw_atl_rpf_vlan_rxq_flr_set(struct aq_hw_s *aq_hw, u32 vlan_rxq,
+				 u32 filter);
 
 /* set ethertype filter enable */
-void rpf_etht_flr_en_set(struct aq_hw_s *aq_hw, u32 etht_flr_en, u32 filter);
+void hw_atl_rpf_etht_flr_en_set(struct aq_hw_s *aq_hw, u32 etht_flr_en,
+				u32 filter);
 
 /* set  ethertype user-priority enable */
-void rpf_etht_user_priority_en_set(struct aq_hw_s *aq_hw,
-				   u32 etht_user_priority_en, u32 filter);
+void hw_atl_rpf_etht_user_priority_en_set(struct aq_hw_s *aq_hw,
+					  u32 etht_user_priority_en,
+					  u32 filter);
 
 /* set  ethertype rx queue enable */
-void rpf_etht_rx_queue_en_set(struct aq_hw_s *aq_hw, u32 etht_rx_queue_en,
-			      u32 filter);
+void hw_atl_rpf_etht_rx_queue_en_set(struct aq_hw_s *aq_hw,
+				     u32 etht_rx_queue_en,
+				     u32 filter);
 
 /* set ethertype rx queue */
-void rpf_etht_rx_queue_set(struct aq_hw_s *aq_hw, u32 etht_rx_queue,
-			   u32 filter);
+void hw_atl_rpf_etht_rx_queue_set(struct aq_hw_s *aq_hw, u32 etht_rx_queue,
+				  u32 filter);
 
 /* set ethertype user-priority */
-void rpf_etht_user_priority_set(struct aq_hw_s *aq_hw, u32 etht_user_priority,
-				u32 filter);
+void hw_atl_rpf_etht_user_priority_set(struct aq_hw_s *aq_hw,
+				       u32 etht_user_priority,
+				       u32 filter);
 
 /* set ethertype management queue */
-void rpf_etht_mgt_queue_set(struct aq_hw_s *aq_hw, u32 etht_mgt_queue,
-			    u32 filter);
+void hw_atl_rpf_etht_mgt_queue_set(struct aq_hw_s *aq_hw, u32 etht_mgt_queue,
+				   u32 filter);
 
 /* set ethertype filter action */
-void rpf_etht_flr_act_set(struct aq_hw_s *aq_hw, u32 etht_flr_act,
-			  u32 filter);
+void hw_atl_rpf_etht_flr_act_set(struct aq_hw_s *aq_hw, u32 etht_flr_act,
+				 u32 filter);
 
 /* set ethertype filter */
-void rpf_etht_flr_set(struct aq_hw_s *aq_hw, u32 etht_flr, u32 filter);
+void hw_atl_rpf_etht_flr_set(struct aq_hw_s *aq_hw, u32 etht_flr, u32 filter);
+
+/* set L4 source port */
+void hw_atl_rpf_l4_spd_set(struct aq_hw_s *aq_hw, u32 val, u32 filter);
+
+/* set L4 destination port */
+void hw_atl_rpf_l4_dpd_set(struct aq_hw_s *aq_hw, u32 val, u32 filter);
 
 /* rpo */
 
 /* set ipv4 header checksum offload enable */
-void rpo_ipv4header_crc_offload_en_set(struct aq_hw_s *aq_hw,
-				       u32 ipv4header_crc_offload_en);
+void hw_atl_rpo_ipv4header_crc_offload_en_set(struct aq_hw_s *aq_hw,
+					      u32 ipv4header_crc_offload_en);
 
 /* set rx descriptor vlan stripping */
-void rpo_rx_desc_vlan_stripping_set(struct aq_hw_s *aq_hw,
-				    u32 rx_desc_vlan_stripping,
-				    u32 descriptor);
+void hw_atl_rpo_rx_desc_vlan_stripping_set(struct aq_hw_s *aq_hw,
+					   u32 rx_desc_vlan_stripping,
+					   u32 descriptor);
+
+void hw_atl_rpo_outer_vlan_tag_mode_set(void *context,
+					u32 outervlantagmode);
+
+u32 hw_atl_rpo_outer_vlan_tag_mode_get(void *context);
 
 /* set tcp/udp checksum offload enable */
-void rpo_tcp_udp_crc_offload_en_set(struct aq_hw_s *aq_hw,
-				    u32 tcp_udp_crc_offload_en);
+void hw_atl_rpo_tcp_udp_crc_offload_en_set(struct aq_hw_s *aq_hw,
+					   u32 tcp_udp_crc_offload_en);
 
 /* Set LRO Patch Optimization Enable. */
-void rpo_lro_patch_optimization_en_set(struct aq_hw_s *aq_hw,
-				       u32 lro_patch_optimization_en);
+void hw_atl_rpo_lro_patch_optimization_en_set(struct aq_hw_s *aq_hw,
+					      u32 lro_patch_optimization_en);
 
 /* Set Large Receive Offload Enable */
-void rpo_lro_en_set(struct aq_hw_s *aq_hw, u32 lro_en);
+void hw_atl_rpo_lro_en_set(struct aq_hw_s *aq_hw, u32 lro_en);
 
 /* Set LRO Q Sessions Limit */
-void rpo_lro_qsessions_lim_set(struct aq_hw_s *aq_hw, u32 lro_qsessions_lim);
+void hw_atl_rpo_lro_qsessions_lim_set(struct aq_hw_s *aq_hw,
+				      u32 lro_qsessions_lim);
 
 /* Set LRO Total Descriptor Limit */
-void rpo_lro_total_desc_lim_set(struct aq_hw_s *aq_hw, u32 lro_total_desc_lim);
+void hw_atl_rpo_lro_total_desc_lim_set(struct aq_hw_s *aq_hw,
+				       u32 lro_total_desc_lim);
 
 /* Set LRO Min Payload of First Packet */
-void rpo_lro_min_pay_of_first_pkt_set(struct aq_hw_s *aq_hw,
-				      u32 lro_min_pld_of_first_pkt);
+void hw_atl_rpo_lro_min_pay_of_first_pkt_set(struct aq_hw_s *aq_hw,
+					     u32 lro_min_pld_of_first_pkt);
 
 /* Set LRO Packet Limit */
-void rpo_lro_pkt_lim_set(struct aq_hw_s *aq_hw, u32 lro_packet_lim);
+void hw_atl_rpo_lro_pkt_lim_set(struct aq_hw_s *aq_hw, u32 lro_packet_lim);
 
 /* Set LRO Max Number of Descriptors */
-void rpo_lro_max_num_of_descriptors_set(struct aq_hw_s *aq_hw,
-					u32 lro_max_desc_num, u32 lro);
+void hw_atl_rpo_lro_max_num_of_descriptors_set(struct aq_hw_s *aq_hw,
+					       u32 lro_max_desc_num, u32 lro);
 
 /* Set LRO Time Base Divider */
-void rpo_lro_time_base_divider_set(struct aq_hw_s *aq_hw,
-				   u32 lro_time_base_divider);
+void hw_atl_rpo_lro_time_base_divider_set(struct aq_hw_s *aq_hw,
+					  u32 lro_time_base_divider);
 
 /*Set LRO Inactive Interval */
-void rpo_lro_inactive_interval_set(struct aq_hw_s *aq_hw,
-				   u32 lro_inactive_interval);
+void hw_atl_rpo_lro_inactive_interval_set(struct aq_hw_s *aq_hw,
+					  u32 lro_inactive_interval);
 
 /*Set LRO Max Coalescing Interval */
-void rpo_lro_max_coalescing_interval_set(struct aq_hw_s *aq_hw,
-					 u32 lro_max_coalescing_interval);
+void hw_atl_rpo_lro_max_coalescing_interval_set(struct aq_hw_s *aq_hw,
+						u32 lro_max_coal_interval);
 
 /* rx */
 
 /* set rx register reset disable */
-void rx_rx_reg_res_dis_set(struct aq_hw_s *aq_hw, u32 rx_reg_res_dis);
+void hw_atl_rx_rx_reg_res_dis_set(struct aq_hw_s *aq_hw, u32 rx_reg_res_dis);
 
 /* tdm */
 
 /* set cpu id */
-void tdm_cpu_id_set(struct aq_hw_s *aq_hw, u32 cpuid, u32 dca);
+void hw_atl_tdm_cpu_id_set(struct aq_hw_s *aq_hw, u32 cpuid, u32 dca);
 
 /* set large send offload enable */
-void tdm_large_send_offload_en_set(struct aq_hw_s *aq_hw,
-				   u32 large_send_offload_en);
+void hw_atl_tdm_large_send_offload_en_set(struct aq_hw_s *aq_hw,
+					  u32 large_send_offload_en);
 
 /* set tx descriptor enable */
-void tdm_tx_desc_en_set(struct aq_hw_s *aq_hw, u32 tx_desc_en, u32 descriptor);
+void hw_atl_tdm_tx_desc_en_set(struct aq_hw_s *aq_hw, u32 tx_desc_en,
+			       u32 descriptor);
 
 /* set tx dca enable */
-void tdm_tx_dca_en_set(struct aq_hw_s *aq_hw, u32 tx_dca_en);
+void hw_atl_tdm_tx_dca_en_set(struct aq_hw_s *aq_hw, u32 tx_dca_en);
 
 /* set tx dca mode */
-void tdm_tx_dca_mode_set(struct aq_hw_s *aq_hw, u32 tx_dca_mode);
+void hw_atl_tdm_tx_dca_mode_set(struct aq_hw_s *aq_hw, u32 tx_dca_mode);
 
 /* set tx descriptor dca enable */
-void tdm_tx_desc_dca_en_set(struct aq_hw_s *aq_hw, u32 tx_desc_dca_en, u32 dca);
+void hw_atl_tdm_tx_desc_dca_en_set(struct aq_hw_s *aq_hw, u32 tx_desc_dca_en,
+				   u32 dca);
 
 /* get tx descriptor head pointer */
-u32 tdm_tx_desc_head_ptr_get(struct aq_hw_s *aq_hw, u32 descriptor);
+u32 hw_atl_tdm_tx_desc_head_ptr_get(struct aq_hw_s *aq_hw, u32 descriptor);
 
 /* set tx descriptor length */
-void tdm_tx_desc_len_set(struct aq_hw_s *aq_hw, u32 tx_desc_len,
-			 u32 descriptor);
+void hw_atl_tdm_tx_desc_len_set(struct aq_hw_s *aq_hw, u32 tx_desc_len,
+				u32 descriptor);
 
 /* set tx descriptor write-back interrupt enable */
-void tdm_tx_desc_wr_wb_irq_en_set(struct aq_hw_s *aq_hw,
-				  u32 tx_desc_wr_wb_irq_en);
+void hw_atl_tdm_tx_desc_wr_wb_irq_en_set(struct aq_hw_s *aq_hw,
+					 u32 tx_desc_wr_wb_irq_en);
 
 /* set tx descriptor write-back threshold */
-void tdm_tx_desc_wr_wb_threshold_set(struct aq_hw_s *aq_hw,
-				     u32 tx_desc_wr_wb_threshold,
+void hw_atl_tdm_tx_desc_wr_wb_threshold_set(struct aq_hw_s *aq_hw,
+					    u32 tx_desc_wr_wb_threshold,
 				     u32 descriptor);
 
 /* Set TDM Interrupt Moderation Enable */
-void tdm_tdm_intr_moder_en_set(struct aq_hw_s *aq_hw,
-			       u32 tdm_irq_moderation_en);
+void hw_atl_tdm_tdm_intr_moder_en_set(struct aq_hw_s *aq_hw,
+				      u32 tdm_irq_moderation_en);
 /* thm */
 
 /* set lso tcp flag of first packet */
-void thm_lso_tcp_flag_of_first_pkt_set(struct aq_hw_s *aq_hw,
-				       u32 lso_tcp_flag_of_first_pkt);
+void hw_atl_thm_lso_tcp_flag_of_first_pkt_set(struct aq_hw_s *aq_hw,
+					      u32 lso_tcp_flag_of_first_pkt);
 
 /* set lso tcp flag of last packet */
-void thm_lso_tcp_flag_of_last_pkt_set(struct aq_hw_s *aq_hw,
-				      u32 lso_tcp_flag_of_last_pkt);
+void hw_atl_thm_lso_tcp_flag_of_last_pkt_set(struct aq_hw_s *aq_hw,
+					     u32 lso_tcp_flag_of_last_pkt);
 
 /* set lso tcp flag of middle packet */
-void thm_lso_tcp_flag_of_middle_pkt_set(struct aq_hw_s *aq_hw,
-					u32 lso_tcp_flag_of_middle_pkt);
+void hw_atl_thm_lso_tcp_flag_of_middle_pkt_set(struct aq_hw_s *aq_hw,
+					       u32 lso_tcp_flag_of_middle_pkt);
 
 /* tpb */
 
+/* set TX Traffic Class Mode */
+void hw_atl_tpb_tps_tx_tc_mode_set(struct aq_hw_s *aq_hw,
+				   u32 tx_traf_class_mode);
+
+/* get TX Traffic Class Mode */
+u32 hw_atl_tpb_tps_tx_tc_mode_get(struct aq_hw_s *aq_hw);
+
 /* set tx buffer enable */
-void tpb_tx_buff_en_set(struct aq_hw_s *aq_hw, u32 tx_buff_en);
+void hw_atl_tpb_tx_buff_en_set(struct aq_hw_s *aq_hw, u32 tx_buff_en);
 
 /* set tx buffer high threshold (per tc) */
-void tpb_tx_buff_hi_threshold_per_tc_set(struct aq_hw_s *aq_hw,
-					 u32 tx_buff_hi_threshold_per_tc,
+void hw_atl_tpb_tx_buff_hi_threshold_per_tc_set(struct aq_hw_s *aq_hw,
+						u32 tx_buff_hi_threshold_per_tc,
 					 u32 buffer);
 
 /* set tx buffer low threshold (per tc) */
-void tpb_tx_buff_lo_threshold_per_tc_set(struct aq_hw_s *aq_hw,
-					 u32 tx_buff_lo_threshold_per_tc,
+void hw_atl_tpb_tx_buff_lo_threshold_per_tc_set(struct aq_hw_s *aq_hw,
+						u32 tx_buff_lo_threshold_per_tc,
 					 u32 buffer);
 
 /* set tx dma system loopback enable */
-void tpb_tx_dma_sys_lbk_en_set(struct aq_hw_s *aq_hw, u32 tx_dma_sys_lbk_en);
+void hw_atl_tpb_tx_dma_sys_lbk_en_set(struct aq_hw_s *aq_hw, u32 tx_dma_sys_lbk_en);
+
+/* set tx dma network loopback enable */
+void hw_atl_tpb_tx_dma_net_lbk_en_set(struct aq_hw_s *aq_hw,
+				      u32 tx_dma_net_lbk_en);
+
+/* set tx clock gating enable */
+void hw_atl_tpb_tx_tx_clk_gate_en_set(struct aq_hw_s *aq_hw,
+				      u32 tx_clk_gate_en);
 
 /* set tx packet buffer size (per tc) */
-void tpb_tx_pkt_buff_size_per_tc_set(struct aq_hw_s *aq_hw,
-				     u32 tx_pkt_buff_size_per_tc, u32 buffer);
+void hw_atl_tpb_tx_pkt_buff_size_per_tc_set(struct aq_hw_s *aq_hw,
+					    u32 tx_pkt_buff_size_per_tc,
+					    u32 buffer);
 
 /* set tx path pad insert enable */
-void tpb_tx_path_scp_ins_en_set(struct aq_hw_s *aq_hw, u32 tx_path_scp_ins_en);
+void hw_atl_tpb_tx_path_scp_ins_en_set(struct aq_hw_s *aq_hw, u32 tx_path_scp_ins_en);
 
 /* tpo */
 
 /* set ipv4 header checksum offload enable */
-void tpo_ipv4header_crc_offload_en_set(struct aq_hw_s *aq_hw,
-				       u32 ipv4header_crc_offload_en);
+void hw_atl_tpo_ipv4header_crc_offload_en_set(struct aq_hw_s *aq_hw,
+					      u32 ipv4header_crc_offload_en);
 
 /* set tcp/udp checksum offload enable */
-void tpo_tcp_udp_crc_offload_en_set(struct aq_hw_s *aq_hw,
-				    u32 tcp_udp_crc_offload_en);
+void hw_atl_tpo_tcp_udp_crc_offload_en_set(struct aq_hw_s *aq_hw,
+					   u32 tcp_udp_crc_offload_en);
 
 /* set tx pkt system loopback enable */
-void tpo_tx_pkt_sys_lbk_en_set(struct aq_hw_s *aq_hw, u32 tx_pkt_sys_lbk_en);
+void hw_atl_tpo_tx_pkt_sys_lbk_en_set(struct aq_hw_s *aq_hw,
+				      u32 tx_pkt_sys_lbk_en);
 
 /* tps */
 
 /* set tx packet scheduler data arbitration mode */
-void tps_tx_pkt_shed_data_arb_mode_set(struct aq_hw_s *aq_hw,
-				       u32 tx_pkt_shed_data_arb_mode);
+void hw_atl_tps_tx_pkt_shed_data_arb_mode_set(struct aq_hw_s *aq_hw,
+					      u32 tx_pkt_shed_data_arb_mode);
 
 /* set tx packet scheduler descriptor rate current time reset */
-void tps_tx_pkt_shed_desc_rate_curr_time_res_set(struct aq_hw_s *aq_hw,
-						 u32 curr_time_res);
+void hw_atl_tps_tx_pkt_shed_desc_rate_curr_time_res_set(struct aq_hw_s *aq_hw,
+							u32 curr_time_res);
 
 /* set tx packet scheduler descriptor rate limit */
-void tps_tx_pkt_shed_desc_rate_lim_set(struct aq_hw_s *aq_hw,
-				       u32 tx_pkt_shed_desc_rate_lim);
+void hw_atl_tps_tx_pkt_shed_desc_rate_lim_set(struct aq_hw_s *aq_hw,
+					      u32 tx_pkt_shed_desc_rate_lim);
 
 /* set tx packet scheduler descriptor tc arbitration mode */
-void tps_tx_pkt_shed_desc_tc_arb_mode_set(struct aq_hw_s *aq_hw,
-					  u32 tx_pkt_shed_desc_tc_arb_mode);
+void hw_atl_tps_tx_pkt_shed_desc_tc_arb_mode_set(struct aq_hw_s *aq_hw,
+						 u32 arb_mode);
 
 /* set tx packet scheduler descriptor tc max credit */
-void tps_tx_pkt_shed_desc_tc_max_credit_set(struct aq_hw_s *aq_hw,
-					    u32 tx_pkt_shed_desc_tc_max_credit,
-					    u32 tc);
+void hw_atl_tps_tx_pkt_shed_desc_tc_max_credit_set(struct aq_hw_s *aq_hw,
+						   const u32 tc,
+						   const u32 max_credit);
 
 /* set tx packet scheduler descriptor tc weight */
-void tps_tx_pkt_shed_desc_tc_weight_set(struct aq_hw_s *aq_hw,
-					u32 tx_pkt_shed_desc_tc_weight,
-					u32 tc);
+void hw_atl_tps_tx_pkt_shed_desc_tc_weight_set(struct aq_hw_s *aq_hw,
+					       const u32 tc,
+					       const u32 weight);
 
 /* set tx packet scheduler descriptor vm arbitration mode */
-void tps_tx_pkt_shed_desc_vm_arb_mode_set(struct aq_hw_s *aq_hw,
-					  u32 tx_pkt_shed_desc_vm_arb_mode);
+void hw_atl_tps_tx_pkt_shed_desc_vm_arb_mode_set(struct aq_hw_s *aq_hw,
+						 u32 arb_mode);
 
 /* set tx packet scheduler tc data max credit */
-void tps_tx_pkt_shed_tc_data_max_credit_set(struct aq_hw_s *aq_hw,
-					    u32 tx_pkt_shed_tc_data_max_credit,
-					    u32 tc);
+void hw_atl_tps_tx_pkt_shed_tc_data_max_credit_set(struct aq_hw_s *aq_hw,
+						   const u32 tc,
+						   const u32 max_credit);
 
 /* set tx packet scheduler tc data weight */
-void tps_tx_pkt_shed_tc_data_weight_set(struct aq_hw_s *aq_hw,
-					u32 tx_pkt_shed_tc_data_weight,
-					u32 tc);
+void hw_atl_tps_tx_pkt_shed_tc_data_weight_set(struct aq_hw_s *aq_hw,
+					       const u32 tc,
+					       const u32 weight);
+
+/* set tx descriptor rate mode */
+void hw_atl_tps_tx_desc_rate_mode_set(struct aq_hw_s *aq_hw,
+				      const u32 rate_mode);
+
+/* set tx packet scheduler descriptor rate enable */
+void hw_atl_tps_tx_desc_rate_en_set(struct aq_hw_s *aq_hw, const u32 desc,
+				    const u32 enable);
+
+/* set tx packet scheduler descriptor rate integral value */
+void hw_atl_tps_tx_desc_rate_x_set(struct aq_hw_s *aq_hw, const u32 desc,
+				   const u32 rate_int);
+
+/* set tx packet scheduler descriptor rate fractional value */
+void hw_atl_tps_tx_desc_rate_y_set(struct aq_hw_s *aq_hw, const u32 desc,
+				   const u32 rate_frac);
 
 /* tx */
 
 /* set tx register reset disable */
-void tx_tx_reg_res_dis_set(struct aq_hw_s *aq_hw, u32 tx_reg_res_dis);
+void hw_atl_tx_tx_reg_res_dis_set(struct aq_hw_s *aq_hw, u32 tx_reg_res_dis);
 
 /* msm */
 
 /* get register access status */
-u32 msm_reg_access_status_get(struct aq_hw_s *aq_hw);
+u32 hw_atl_msm_reg_access_status_get(struct aq_hw_s *aq_hw);
 
 /* set  register address for indirect address */
-void msm_reg_addr_for_indirect_addr_set(struct aq_hw_s *aq_hw,
-					u32 reg_addr_for_indirect_addr);
+void hw_atl_msm_reg_addr_for_indirect_addr_set(struct aq_hw_s *aq_hw,
+					       u32 reg_addr_for_indirect_addr);
 
 /* set register read strobe */
-void msm_reg_rd_strobe_set(struct aq_hw_s *aq_hw, u32 reg_rd_strobe);
+void hw_atl_msm_reg_rd_strobe_set(struct aq_hw_s *aq_hw, u32 reg_rd_strobe);
 
 /* get  register read data */
-u32 msm_reg_rd_data_get(struct aq_hw_s *aq_hw);
+u32 hw_atl_msm_reg_rd_data_get(struct aq_hw_s *aq_hw);
 
 /* set  register write data */
-void msm_reg_wr_data_set(struct aq_hw_s *aq_hw, u32 reg_wr_data);
+void hw_atl_msm_reg_wr_data_set(struct aq_hw_s *aq_hw, u32 reg_wr_data);
 
 /* set register write strobe */
-void msm_reg_wr_strobe_set(struct aq_hw_s *aq_hw, u32 reg_wr_strobe);
+void hw_atl_msm_reg_wr_strobe_set(struct aq_hw_s *aq_hw, u32 reg_wr_strobe);
 
 /* pci */
 
 /* set pci register reset disable */
-void pci_pci_reg_res_dis_set(struct aq_hw_s *aq_hw, u32 pci_reg_res_dis);
+void hw_atl_pci_pci_reg_res_dis_set(struct aq_hw_s *aq_hw, u32 pci_reg_res_dis);
+
+/* pcs */
+void hw_atl_pcs_ptp_clock_read_enable(struct aq_hw_s *aq_hw,
+				      u32 ptp_clock_read_enable);
+
+u32 hw_atl_pcs_ptp_clock_get(struct aq_hw_s *aq_hw, u32 index);
+
+/* set uP Force Interrupt */
+void hw_atl_mcp_up_force_intr_set(struct aq_hw_s *aq_hw, u32 up_force_intr);
+
+/* clear ipv4 filter destination address */
+void hw_atl_rpfl3l4_ipv4_dest_addr_clear(struct aq_hw_s *aq_hw, u8 location);
+
+/* clear ipv4 filter source address */
+void hw_atl_rpfl3l4_ipv4_src_addr_clear(struct aq_hw_s *aq_hw, u8 location);
+
+/* clear command for filter l3-l4 */
+void hw_atl_rpfl3l4_cmd_clear(struct aq_hw_s *aq_hw, u8 location);
+
+/* clear ipv6 filter destination address */
+void hw_atl_rpfl3l4_ipv6_dest_addr_clear(struct aq_hw_s *aq_hw, u8 location);
+
+/* clear ipv6 filter source address */
+void hw_atl_rpfl3l4_ipv6_src_addr_clear(struct aq_hw_s *aq_hw, u8 location);
+
+/* set ipv4 filter destination address */
+void hw_atl_rpfl3l4_ipv4_dest_addr_set(struct aq_hw_s *aq_hw, u8 location,
+				       u32 ipv4_dest);
+
+/* set ipv4 filter source address */
+void hw_atl_rpfl3l4_ipv4_src_addr_set(struct aq_hw_s *aq_hw, u8 location,
+				      u32 ipv4_src);
+
+/* set command for filter l3-l4 */
+void hw_atl_rpfl3l4_cmd_set(struct aq_hw_s *aq_hw, u8 location, u32 cmd);
+
+/* set ipv6 filter source address */
+void hw_atl_rpfl3l4_ipv6_src_addr_set(struct aq_hw_s *aq_hw, u8 location,
+				      u32 *ipv6_src);
+
+/* set ipv6 filter destination address */
+void hw_atl_rpfl3l4_ipv6_dest_addr_set(struct aq_hw_s *aq_hw, u8 location,
+				       u32 *ipv6_dest);
+
+/* set Global MDIO Interface 1 */
+void hw_atl_glb_mdio_iface1_set(struct aq_hw_s *hw, u32 value);
+
+/* get Global MDIO Interface 1 */
+u32 hw_atl_glb_mdio_iface1_get(struct aq_hw_s *hw);
+
+/* set Global MDIO Interface 2 */
+void hw_atl_glb_mdio_iface2_set(struct aq_hw_s *hw, u32 value);
+
+/* get Global MDIO Interface 2 */
+u32 hw_atl_glb_mdio_iface2_get(struct aq_hw_s *hw);
+
+/* set Global MDIO Interface 3 */
+void hw_atl_glb_mdio_iface3_set(struct aq_hw_s *hw, u32 value);
+
+/* get Global MDIO Interface 3 */
+u32 hw_atl_glb_mdio_iface3_get(struct aq_hw_s *hw);
+
+/* set Global MDIO Interface 4 */
+void hw_atl_glb_mdio_iface4_set(struct aq_hw_s *hw, u32 value);
+
+/* get Global MDIO Interface 4 */
+u32 hw_atl_glb_mdio_iface4_get(struct aq_hw_s *hw);
+
+/* set Global MDIO Interface 5 */
+void hw_atl_glb_mdio_iface5_set(struct aq_hw_s *hw, u32 value);
+
+/* get Global MDIO Interface 5 */
+u32 hw_atl_glb_mdio_iface5_get(struct aq_hw_s *hw);
+
+u32 hw_atl_mdio_busy_get(struct aq_hw_s *aq_hw);
+
+/* get global microprocessor ram semaphore */
+u32 hw_atl_sem_ram_get(struct aq_hw_s *self);
+
+/* get global microprocessor mdio semaphore */
+u32 hw_atl_sem_mdio_get(struct aq_hw_s *self);
+
+u32 hw_atl_sem_reset1_get(struct aq_hw_s *self);
+u32 hw_atl_sem_reset2_get(struct aq_hw_s *self);
+
+/* get global microprocessor scratch pad register */
+u32 hw_atl_scrpad_get(struct aq_hw_s *aq_hw, u32 scratch_scp);
+
+/* get global microprocessor scratch pad 12 register */
+u32 hw_atl_scrpad12_get(struct aq_hw_s *self);
+
+/* get global microprocessor scratch pad 25 register */
+u32 hw_atl_scrpad25_get(struct aq_hw_s *self);
 
 #endif /* HW_ATL_LLH_H */
diff --git a/drivers/net/ethernet/aquantia/atlantic/hw_atl/hw_atl_llh_internal.h b/drivers/net/ethernet/aquantia/atlantic/hw_atl/hw_atl_llh_internal.h
index 5527fc0e5942..fce30d90b6cb 100644
--- a/drivers/net/ethernet/aquantia/atlantic/hw_atl/hw_atl_llh_internal.h
+++ b/drivers/net/ethernet/aquantia/atlantic/hw_atl/hw_atl_llh_internal.h
@@ -1,10 +1,8 @@
-/*
- * aQuantia Corporation Network Driver
- * Copyright (C) 2014-2017 aQuantia Corporation. All rights reserved
+/* SPDX-License-Identifier: GPL-2.0-only */
+/* Atlantic Network Driver
  *
- * This program is free software; you can redistribute it and/or modify it
- * under the terms and conditions of the GNU General Public License,
- * version 2, as published by the Free Software Foundation.
+ * Copyright (C) 2014-2019 aQuantia Corporation
+ * Copyright (C) 2019-2020 Marvell International Ltd.
  */
 
 /* File hw_atl_llh_internal.h: Preprocessor definitions
@@ -14,95 +12,167 @@
 #ifndef HW_ATL_LLH_INTERNAL_H
 #define HW_ATL_LLH_INTERNAL_H
 
+/* COM Temperature Sense Reset Bitfield Definitions */
+#define HW_ATL_TS_RESET_ADR 0x00003100
+#define HW_ATL_TS_RESET_MSK 0x00000004
+#define HW_ATL_TS_RESET_SHIFT 2
+#define HW_ATL_TS_RESET_WIDTH 1
+
+/* COM Temperature Sense Power Down Bitfield Definitions */
+#define HW_ATL_TS_POWER_DOWN_ADR 0x00003100
+#define HW_ATL_TS_POWER_DOWN_MSK 0x00000001
+#define HW_ATL_TS_POWER_DOWN_SHIFT 0
+#define HW_ATL_TS_POWER_DOWN_WIDTH 1
+
+/* COM Temperature Sense Ready Bitfield Definitions */
+#define HW_ATL_TS_READY_ADR 0x00003120
+#define HW_ATL_TS_READY_MSK 0x80000000
+#define HW_ATL_TS_READY_SHIFT 31
+#define HW_ATL_TS_READY_WIDTH 1
+
+/*  COM Temperature Sense Ready Latch High Bitfield Definitions */
+#define HW_ATL_TS_READY_LATCH_HIGH_ADR 0x00003120
+#define HW_ATL_TS_READY_LATCH_HIGH_MSK 0x40000000
+#define HW_ATL_TS_READY_LATCH_HIGH_SHIFT 30
+#define HW_ATL_TS_READY_LATCH_HIGH_WIDTH 1
+
+/* COM Temperature Sense Data Out [B:0] Bitfield Definitions */
+#define HW_ATL_TS_DATA_OUT_ADR 0x00003120
+#define HW_ATL_TS_DATA_OUT_MSK 0x00000FFF
+#define HW_ATL_TS_DATA_OUT_SHIFT 0
+#define HW_ATL_TS_DATA_OUT_WIDTH 12
+
+/* SMBUS0 Received Data register */
+#define HW_ATL_SMB0_RECEIVED_DATA_ADR 0x00000748
+/* SMBUS0 Transmitted Data register */
+#define HW_ATL_SMB0_TRANSMITTED_DATA_ADR 0x00000608
+
+/* SMBUS0 Global Provisioning 2 register */
+#define HW_ATL_SMB0_PROVISIONING2_ADR 0x00000604
+
+/* SMBUS0 Bus Busy Bitfield Definitions */
+#define HW_ATL_SMB0_BUS_BUSY_ADR 0x00000744
+#define HW_ATL_SMB0_BUS_BUSY_MSK 0x00000080
+#define HW_ATL_SMB0_BUS_BUSY_SHIFT 7
+#define HW_ATL_SMB0_BUS_BUSY_WIDTH 1
+
+/* SMBUS0 Byte Transfer Complete Bitfield Definitions */
+#define HW_ATL_SMB0_BYTE_TRANSFER_COMPLETE_ADR 0x00000744
+#define HW_ATL_SMB0_BYTE_TRANSFER_COMPLETE_MSK 0x00000002
+#define HW_ATL_SMB0_BYTE_TRANSFER_COMPLETE_SHIFT 1
+#define HW_ATL_SMB0_BYTE_TRANSFER_COMPLETE_WIDTH 1
+
+/* SMBUS0 Receive Acknowledge Bitfield Definitions */
+#define HW_ATL_SMB0_RX_ACKNOWLEDGED_ADR 0x00000744
+#define HW_ATL_SMB0_RX_ACKNOWLEDGED_MSK 0x00000100
+#define HW_ATL_SMB0_RX_ACKNOWLEDGED_SHIFT 8
+#define HW_ATL_SMB0_RX_ACKNOWLEDGED_WIDTH 1
+
+/* SMBUS0 Repeated Start Detect Bitfield Definitions */
+#define HW_ATL_SMB0_REPEATED_START_DETECT_ADR 0x00000744
+#define HW_ATL_SMB0_REPEATED_START_DETECT_MSK 0x00000004
+#define HW_ATL_SMB0_REPEATED_START_DETECT_SHIFT 2
+#define HW_ATL_SMB0_REPEATED_START_DETECT_WIDTH 1
+
 /* global microprocessor semaphore  definitions
  * base address: 0x000003a0
  * parameter: semaphore {s} | stride size 0x4 | range [0, 15]
  */
-#define glb_cpu_sem_adr(semaphore)  (0x000003a0u + (semaphore) * 0x4)
+#define HW_ATL_GLB_CPU_SEM_ADR(semaphore)  (0x000003a0u + (semaphore) * 0x4)
 /* register address for bitfield rx dma good octet counter lsw [1f:0] */
-#define stats_rx_dma_good_octet_counterlsw__adr 0x00006808
+#define HW_ATL_STATS_RX_DMA_GOOD_OCTET_COUNTERLSW 0x00006808
 /* register address for bitfield rx dma good packet counter lsw [1f:0] */
-#define stats_rx_dma_good_pkt_counterlsw__adr 0x00006800
+#define HW_ATL_STATS_RX_DMA_GOOD_PKT_COUNTERLSW 0x00006800
 /* register address for bitfield tx dma good octet counter lsw [1f:0] */
-#define stats_tx_dma_good_octet_counterlsw__adr 0x00008808
+#define HW_ATL_STATS_TX_DMA_GOOD_OCTET_COUNTERLSW 0x00008808
 /* register address for bitfield tx dma good packet counter lsw [1f:0] */
-#define stats_tx_dma_good_pkt_counterlsw__adr 0x00008800
+#define HW_ATL_STATS_TX_DMA_GOOD_PKT_COUNTERLSW 0x00008800
 
 /* register address for bitfield rx dma good octet counter msw [3f:20] */
-#define stats_rx_dma_good_octet_countermsw__adr 0x0000680c
+#define HW_ATL_STATS_RX_DMA_GOOD_OCTET_COUNTERMSW 0x0000680c
 /* register address for bitfield rx dma good packet counter msw [3f:20] */
-#define stats_rx_dma_good_pkt_countermsw__adr 0x00006804
+#define HW_ATL_STATS_RX_DMA_GOOD_PKT_COUNTERMSW 0x00006804
 /* register address for bitfield tx dma good octet counter msw [3f:20] */
-#define stats_tx_dma_good_octet_countermsw__adr 0x0000880c
+#define HW_ATL_STATS_TX_DMA_GOOD_OCTET_COUNTERMSW 0x0000880c
 /* register address for bitfield tx dma good packet counter msw [3f:20] */
-#define stats_tx_dma_good_pkt_countermsw__adr 0x00008804
+#define HW_ATL_STATS_TX_DMA_GOOD_PKT_COUNTERMSW 0x00008804
 
 /* preprocessor definitions for msm rx errors counter register */
-#define mac_msm_rx_errs_cnt_adr 0x00000120u
+#define HW_ATL_MAC_MSM_RX_ERRS_CNT_ADR 0x00000120u
 
 /* preprocessor definitions for msm rx unicast frames counter register */
-#define mac_msm_rx_ucst_frm_cnt_adr 0x000000e0u
+#define HW_ATL_MAC_MSM_RX_UCST_FRM_CNT_ADR 0x000000e0u
 
 /* preprocessor definitions for msm rx multicast frames counter register */
-#define mac_msm_rx_mcst_frm_cnt_adr 0x000000e8u
+#define HW_ATL_MAC_MSM_RX_MCST_FRM_CNT_ADR 0x000000e8u
 
 /* preprocessor definitions for msm rx broadcast frames counter register */
-#define mac_msm_rx_bcst_frm_cnt_adr 0x000000f0u
+#define HW_ATL_MAC_MSM_RX_BCST_FRM_CNT_ADR 0x000000f0u
 
 /* preprocessor definitions for msm rx broadcast octets counter register 1 */
-#define mac_msm_rx_bcst_octets_counter1_adr 0x000001b0u
+#define HW_ATL_MAC_MSM_RX_BCST_OCTETS_COUNTER1_ADR 0x000001b0u
 
 /* preprocessor definitions for msm rx broadcast octets counter register 2 */
-#define mac_msm_rx_bcst_octets_counter2_adr 0x000001b4u
+#define HW_ATL_MAC_MSM_RX_BCST_OCTETS_COUNTER2_ADR 0x000001b4u
 
 /* preprocessor definitions for msm rx unicast octets counter register 0 */
-#define mac_msm_rx_ucst_octets_counter0_adr 0x000001b8u
-
-/* preprocessor definitions for rx dma statistics counter 7 */
-#define rx_dma_stat_counter7_adr 0x00006818u
+#define HW_ATL_MAC_MSM_RX_UCST_OCTETS_COUNTER0_ADR 0x000001b8u
 
 /* preprocessor definitions for msm tx unicast frames counter register */
-#define mac_msm_tx_ucst_frm_cnt_adr 0x00000108u
+#define HW_ATL_MAC_MSM_TX_UCST_FRM_CNT_ADR 0x00000108u
 
 /* preprocessor definitions for msm tx multicast frames counter register */
-#define mac_msm_tx_mcst_frm_cnt_adr 0x00000110u
+#define HW_ATL_MAC_MSM_TX_MCST_FRM_CNT_ADR 0x00000110u
 
 /* preprocessor definitions for global mif identification */
-#define glb_mif_id_adr 0x0000001cu
+#define HW_ATL_GLB_MIF_ID_ADR 0x0000001cu
 
 /* register address for bitfield iamr_lsw[1f:0] */
-#define itr_iamrlsw_adr 0x00002090
+#define HW_ATL_ITR_IAMRLSW_ADR 0x00002090
 /* register address for bitfield rx dma drop packet counter [1f:0] */
-#define rpb_rx_dma_drop_pkt_cnt_adr 0x00006818
+#define HW_ATL_RPB_RX_DMA_DROP_PKT_CNT_ADR 0x00006818
 
 /* register address for bitfield imcr_lsw[1f:0] */
-#define itr_imcrlsw_adr 0x00002070
+#define HW_ATL_ITR_IMCRLSW_ADR 0x00002070
 /* register address for bitfield imsr_lsw[1f:0] */
-#define itr_imsrlsw_adr 0x00002060
+#define HW_ATL_ITR_IMSRLSW_ADR 0x00002060
 /* register address for bitfield itr_reg_res_dsbl */
-#define itr_reg_res_dsbl_adr 0x00002300
+#define HW_ATL_ITR_REG_RES_DSBL_ADR 0x00002300
 /* bitmask for bitfield itr_reg_res_dsbl */
-#define itr_reg_res_dsbl_msk 0x20000000
+#define HW_ATL_ITR_REG_RES_DSBL_MSK 0x20000000
 /* lower bit position of bitfield itr_reg_res_dsbl */
-#define itr_reg_res_dsbl_shift 29
+#define HW_ATL_ITR_REG_RES_DSBL_SHIFT 29
 /* register address for bitfield iscr_lsw[1f:0] */
-#define itr_iscrlsw_adr 0x00002050
+#define HW_ATL_ITR_ISCRLSW_ADR 0x00002050
 /* register address for bitfield isr_lsw[1f:0] */
-#define itr_isrlsw_adr 0x00002000
+#define HW_ATL_ITR_ISRLSW_ADR 0x00002000
 /* register address for bitfield itr_reset */
-#define itr_res_adr 0x00002300
+#define HW_ATL_ITR_RES_ADR 0x00002300
 /* bitmask for bitfield itr_reset */
-#define itr_res_msk 0x80000000
+#define HW_ATL_ITR_RES_MSK 0x80000000
 /* lower bit position of bitfield itr_reset */
-#define itr_res_shift 31
+#define HW_ATL_ITR_RES_SHIFT 31
+
+/* register address for bitfield  rsc_en */
+#define HW_ATL_ITR_RSC_EN_ADR 0x00002200
+
+/* register address for bitfield  rsc_delay */
+#define HW_ATL_ITR_RSC_DELAY_ADR 0x00002204
+/* bitmask for bitfield  rsc_delay */
+#define HW_ATL_ITR_RSC_DELAY_MSK 0x0000000f
+/* width of bitfield  rsc_delay */
+#define HW_ATL_ITR_RSC_DELAY_WIDTH 4
+/* lower bit position of bitfield  rsc_delay */
+#define HW_ATL_ITR_RSC_DELAY_SHIFT 0
+
 /* register address for bitfield dca{d}_cpuid[7:0] */
-#define rdm_dcadcpuid_adr(dca) (0x00006100 + (dca) * 0x4)
+#define HW_ATL_RDM_DCADCPUID_ADR(dca) (0x00006100 + (dca) * 0x4)
 /* bitmask for bitfield dca{d}_cpuid[7:0] */
-#define rdm_dcadcpuid_msk 0x000000ff
+#define HW_ATL_RDM_DCADCPUID_MSK 0x000000ff
 /* lower bit position of bitfield dca{d}_cpuid[7:0] */
-#define rdm_dcadcpuid_shift 0
+#define HW_ATL_RDM_DCADCPUID_SHIFT 0
 /* register address for bitfield dca_en */
-#define rdm_dca_en_adr 0x00006180
+#define HW_ATL_RDM_DCA_EN_ADR 0x00006180
 
 /* rx dca_en bitfield definitions
  * preprocessor definitions for the bitfield "dca_en".
@@ -110,17 +180,17 @@
  */
 
 /* register address for bitfield dca_en */
-#define rdm_dca_en_adr 0x00006180
+#define HW_ATL_RDM_DCA_EN_ADR 0x00006180
 /* bitmask for bitfield dca_en */
-#define rdm_dca_en_msk 0x80000000
+#define HW_ATL_RDM_DCA_EN_MSK 0x80000000
 /* inverted bitmask for bitfield dca_en */
-#define rdm_dca_en_mskn 0x7fffffff
+#define HW_ATL_RDM_DCA_EN_MSKN 0x7fffffff
 /* lower bit position of bitfield dca_en */
-#define rdm_dca_en_shift 31
+#define HW_ATL_RDM_DCA_EN_SHIFT 31
 /* width of bitfield dca_en */
-#define rdm_dca_en_width 1
+#define HW_ATL_RDM_DCA_EN_WIDTH 1
 /* default value of bitfield dca_en */
-#define rdm_dca_en_default 0x1
+#define HW_ATL_RDM_DCA_EN_DEFAULT 0x1
 
 /* rx dca_mode[3:0] bitfield definitions
  * preprocessor definitions for the bitfield "dca_mode[3:0]".
@@ -128,17 +198,17 @@
  */
 
 /* register address for bitfield dca_mode[3:0] */
-#define rdm_dca_mode_adr 0x00006180
+#define HW_ATL_RDM_DCA_MODE_ADR 0x00006180
 /* bitmask for bitfield dca_mode[3:0] */
-#define rdm_dca_mode_msk 0x0000000f
+#define HW_ATL_RDM_DCA_MODE_MSK 0x0000000f
 /* inverted bitmask for bitfield dca_mode[3:0] */
-#define rdm_dca_mode_mskn 0xfffffff0
+#define HW_ATL_RDM_DCA_MODE_MSKN 0xfffffff0
 /* lower bit position of bitfield dca_mode[3:0] */
-#define rdm_dca_mode_shift 0
+#define HW_ATL_RDM_DCA_MODE_SHIFT 0
 /* width of bitfield dca_mode[3:0] */
-#define rdm_dca_mode_width 4
+#define HW_ATL_RDM_DCA_MODE_WIDTH 4
 /* default value of bitfield dca_mode[3:0] */
-#define rdm_dca_mode_default 0x0
+#define HW_ATL_RDM_DCA_MODE_DEFAULT 0x0
 
 /* rx desc{d}_data_size[4:0] bitfield definitions
  * preprocessor definitions for the bitfield "desc{d}_data_size[4:0]".
@@ -147,17 +217,18 @@
  */
 
 /* register address for bitfield desc{d}_data_size[4:0] */
-#define rdm_descddata_size_adr(descriptor) (0x00005b18 + (descriptor) * 0x20)
+#define HW_ATL_RDM_DESCDDATA_SIZE_ADR(descriptor) \
+	(0x00005b18 + (descriptor) * 0x20)
 /* bitmask for bitfield desc{d}_data_size[4:0] */
-#define rdm_descddata_size_msk 0x0000001f
+#define HW_ATL_RDM_DESCDDATA_SIZE_MSK 0x0000001f
 /* inverted bitmask for bitfield desc{d}_data_size[4:0] */
-#define rdm_descddata_size_mskn 0xffffffe0
+#define HW_ATL_RDM_DESCDDATA_SIZE_MSKN 0xffffffe0
 /* lower bit position of bitfield desc{d}_data_size[4:0] */
-#define rdm_descddata_size_shift 0
+#define HW_ATL_RDM_DESCDDATA_SIZE_SHIFT 0
 /* width of bitfield desc{d}_data_size[4:0] */
-#define rdm_descddata_size_width 5
+#define HW_ATL_RDM_DESCDDATA_SIZE_WIDTH 5
 /* default value of bitfield desc{d}_data_size[4:0] */
-#define rdm_descddata_size_default 0x0
+#define HW_ATL_RDM_DESCDDATA_SIZE_DEFAULT 0x0
 
 /* rx dca{d}_desc_en bitfield definitions
  * preprocessor definitions for the bitfield "dca{d}_desc_en".
@@ -166,17 +237,17 @@
  */
 
 /* register address for bitfield dca{d}_desc_en */
-#define rdm_dcaddesc_en_adr(dca) (0x00006100 + (dca) * 0x4)
+#define HW_ATL_RDM_DCADDESC_EN_ADR(dca) (0x00006100 + (dca) * 0x4)
 /* bitmask for bitfield dca{d}_desc_en */
-#define rdm_dcaddesc_en_msk 0x80000000
+#define HW_ATL_RDM_DCADDESC_EN_MSK 0x80000000
 /* inverted bitmask for bitfield dca{d}_desc_en */
-#define rdm_dcaddesc_en_mskn 0x7fffffff
+#define HW_ATL_RDM_DCADDESC_EN_MSKN 0x7fffffff
 /* lower bit position of bitfield dca{d}_desc_en */
-#define rdm_dcaddesc_en_shift 31
+#define HW_ATL_RDM_DCADDESC_EN_SHIFT 31
 /* width of bitfield dca{d}_desc_en */
-#define rdm_dcaddesc_en_width 1
+#define HW_ATL_RDM_DCADDESC_EN_WIDTH 1
 /* default value of bitfield dca{d}_desc_en */
-#define rdm_dcaddesc_en_default 0x0
+#define HW_ATL_RDM_DCADDESC_EN_DEFAULT 0x0
 
 /* rx desc{d}_en bitfield definitions
  * preprocessor definitions for the bitfield "desc{d}_en".
@@ -185,17 +256,17 @@
  */
 
 /* register address for bitfield desc{d}_en */
-#define rdm_descden_adr(descriptor) (0x00005b08 + (descriptor) * 0x20)
+#define HW_ATL_RDM_DESCDEN_ADR(descriptor) (0x00005b08 + (descriptor) * 0x20)
 /* bitmask for bitfield desc{d}_en */
-#define rdm_descden_msk 0x80000000
+#define HW_ATL_RDM_DESCDEN_MSK 0x80000000
 /* inverted bitmask for bitfield desc{d}_en */
-#define rdm_descden_mskn 0x7fffffff
+#define HW_ATL_RDM_DESCDEN_MSKN 0x7fffffff
 /* lower bit position of bitfield desc{d}_en */
-#define rdm_descden_shift 31
+#define HW_ATL_RDM_DESCDEN_SHIFT 31
 /* width of bitfield desc{d}_en */
-#define rdm_descden_width 1
+#define HW_ATL_RDM_DESCDEN_WIDTH 1
 /* default value of bitfield desc{d}_en */
-#define rdm_descden_default 0x0
+#define HW_ATL_RDM_DESCDEN_DEFAULT 0x0
 
 /* rx desc{d}_hdr_size[4:0] bitfield definitions
  * preprocessor definitions for the bitfield "desc{d}_hdr_size[4:0]".
@@ -204,17 +275,18 @@
  */
 
 /* register address for bitfield desc{d}_hdr_size[4:0] */
-#define rdm_descdhdr_size_adr(descriptor) (0x00005b18 + (descriptor) * 0x20)
+#define HW_ATL_RDM_DESCDHDR_SIZE_ADR(descriptor) \
+	(0x00005b18 + (descriptor) * 0x20)
 /* bitmask for bitfield desc{d}_hdr_size[4:0] */
-#define rdm_descdhdr_size_msk 0x00001f00
+#define HW_ATL_RDM_DESCDHDR_SIZE_MSK 0x00001f00
 /* inverted bitmask for bitfield desc{d}_hdr_size[4:0] */
-#define rdm_descdhdr_size_mskn 0xffffe0ff
+#define HW_ATL_RDM_DESCDHDR_SIZE_MSKN 0xffffe0ff
 /* lower bit position of bitfield desc{d}_hdr_size[4:0] */
-#define rdm_descdhdr_size_shift 8
+#define HW_ATL_RDM_DESCDHDR_SIZE_SHIFT 8
 /* width of bitfield desc{d}_hdr_size[4:0] */
-#define rdm_descdhdr_size_width 5
+#define HW_ATL_RDM_DESCDHDR_SIZE_WIDTH 5
 /* default value of bitfield desc{d}_hdr_size[4:0] */
-#define rdm_descdhdr_size_default 0x0
+#define HW_ATL_RDM_DESCDHDR_SIZE_DEFAULT 0x0
 
 /* rx desc{d}_hdr_split bitfield definitions
  * preprocessor definitions for the bitfield "desc{d}_hdr_split".
@@ -223,17 +295,18 @@
  */
 
 /* register address for bitfield desc{d}_hdr_split */
-#define rdm_descdhdr_split_adr(descriptor) (0x00005b08 + (descriptor) * 0x20)
+#define HW_ATL_RDM_DESCDHDR_SPLIT_ADR(descriptor) \
+	(0x00005b08 + (descriptor) * 0x20)
 /* bitmask for bitfield desc{d}_hdr_split */
-#define rdm_descdhdr_split_msk 0x10000000
+#define HW_ATL_RDM_DESCDHDR_SPLIT_MSK 0x10000000
 /* inverted bitmask for bitfield desc{d}_hdr_split */
-#define rdm_descdhdr_split_mskn 0xefffffff
+#define HW_ATL_RDM_DESCDHDR_SPLIT_MSKN 0xefffffff
 /* lower bit position of bitfield desc{d}_hdr_split */
-#define rdm_descdhdr_split_shift 28
+#define HW_ATL_RDM_DESCDHDR_SPLIT_SHIFT 28
 /* width of bitfield desc{d}_hdr_split */
-#define rdm_descdhdr_split_width 1
+#define HW_ATL_RDM_DESCDHDR_SPLIT_WIDTH 1
 /* default value of bitfield desc{d}_hdr_split */
-#define rdm_descdhdr_split_default 0x0
+#define HW_ATL_RDM_DESCDHDR_SPLIT_DEFAULT 0x0
 
 /* rx desc{d}_hd[c:0] bitfield definitions
  * preprocessor definitions for the bitfield "desc{d}_hd[c:0]".
@@ -242,15 +315,15 @@
  */
 
 /* register address for bitfield desc{d}_hd[c:0] */
-#define rdm_descdhd_adr(descriptor) (0x00005b0c + (descriptor) * 0x20)
+#define HW_ATL_RDM_DESCDHD_ADR(descriptor) (0x00005b0c + (descriptor) * 0x20)
 /* bitmask for bitfield desc{d}_hd[c:0] */
-#define rdm_descdhd_msk 0x00001fff
+#define HW_ATL_RDM_DESCDHD_MSK 0x00001fff
 /* inverted bitmask for bitfield desc{d}_hd[c:0] */
-#define rdm_descdhd_mskn 0xffffe000
+#define HW_ATL_RDM_DESCDHD_MSKN 0xffffe000
 /* lower bit position of bitfield desc{d}_hd[c:0] */
-#define rdm_descdhd_shift 0
+#define HW_ATL_RDM_DESCDHD_SHIFT 0
 /* width of bitfield desc{d}_hd[c:0] */
-#define rdm_descdhd_width 13
+#define HW_ATL_RDM_DESCDHD_WIDTH 13
 
 /* rx desc{d}_len[9:0] bitfield definitions
  * preprocessor definitions for the bitfield "desc{d}_len[9:0]".
@@ -259,17 +332,17 @@
  */
 
 /* register address for bitfield desc{d}_len[9:0] */
-#define rdm_descdlen_adr(descriptor) (0x00005b08 + (descriptor) * 0x20)
+#define HW_ATL_RDM_DESCDLEN_ADR(descriptor) (0x00005b08 + (descriptor) * 0x20)
 /* bitmask for bitfield desc{d}_len[9:0] */
-#define rdm_descdlen_msk 0x00001ff8
+#define HW_ATL_RDM_DESCDLEN_MSK 0x00001ff8
 /* inverted bitmask for bitfield desc{d}_len[9:0] */
-#define rdm_descdlen_mskn 0xffffe007
+#define HW_ATL_RDM_DESCDLEN_MSKN 0xffffe007
 /* lower bit position of bitfield desc{d}_len[9:0] */
-#define rdm_descdlen_shift 3
+#define HW_ATL_RDM_DESCDLEN_SHIFT 3
 /* width of bitfield desc{d}_len[9:0] */
-#define rdm_descdlen_width 10
+#define HW_ATL_RDM_DESCDLEN_WIDTH 10
 /* default value of bitfield desc{d}_len[9:0] */
-#define rdm_descdlen_default 0x0
+#define HW_ATL_RDM_DESCDLEN_DEFAULT 0x0
 
 /* rx desc{d}_reset bitfield definitions
  * preprocessor definitions for the bitfield "desc{d}_reset".
@@ -278,17 +351,54 @@
  */
 
 /* register address for bitfield desc{d}_reset */
-#define rdm_descdreset_adr(descriptor) (0x00005b08 + (descriptor) * 0x20)
+#define HW_ATL_RDM_DESCDRESET_ADR(descriptor) (0x00005b08 + (descriptor) * 0x20)
 /* bitmask for bitfield desc{d}_reset */
-#define rdm_descdreset_msk 0x02000000
+#define HW_ATL_RDM_DESCDRESET_MSK 0x02000000
 /* inverted bitmask for bitfield desc{d}_reset */
-#define rdm_descdreset_mskn 0xfdffffff
+#define HW_ATL_RDM_DESCDRESET_MSKN 0xfdffffff
 /* lower bit position of bitfield desc{d}_reset */
-#define rdm_descdreset_shift 25
+#define HW_ATL_RDM_DESCDRESET_SHIFT 25
 /* width of bitfield desc{d}_reset */
-#define rdm_descdreset_width 1
+#define HW_ATL_RDM_DESCDRESET_WIDTH 1
 /* default value of bitfield desc{d}_reset */
-#define rdm_descdreset_default 0x0
+#define HW_ATL_RDM_DESCDRESET_DEFAULT 0x0
+
+/* rdm_desc_init_i bitfield definitions
+ * preprocessor definitions for the bitfield rdm_desc_init_i.
+ * port="pif_rdm_desc_init_i"
+ */
+
+/* register address for bitfield rdm_desc_init_i */
+#define HW_ATL_RDM_RX_DMA_DESC_CACHE_INIT_ADR 0x00005a00
+/* bitmask for bitfield rdm_desc_init_i */
+#define HW_ATL_RDM_RX_DMA_DESC_CACHE_INIT_MSK 0xffffffff
+/* inverted bitmask for bitfield rdm_desc_init_i */
+#define HW_ATL_RDM_RX_DMA_DESC_CACHE_INIT_MSKN 0x00000000
+/* lower bit position of bitfield  rdm_desc_init_i */
+#define HW_ATL_RDM_RX_DMA_DESC_CACHE_INIT_SHIFT 0
+/* width of bitfield rdm_desc_init_i */
+#define HW_ATL_RDM_RX_DMA_DESC_CACHE_INIT_WIDTH 32
+/* default value of bitfield rdm_desc_init_i */
+#define HW_ATL_RDM_RX_DMA_DESC_CACHE_INIT_DEFAULT 0x0
+
+/* rdm_desc_init_done_i bitfield definitions
+ * preprocessor definitions for the bitfield rdm_desc_init_done_i.
+ * port="pif_rdm_desc_init_done_i"
+ */
+
+/* register address for bitfield rdm_desc_init_done_i */
+#define RDM_RX_DMA_DESC_CACHE_INIT_DONE_ADR 0x00005a10
+/* bitmask for bitfield rdm_desc_init_done_i */
+#define RDM_RX_DMA_DESC_CACHE_INIT_DONE_MSK 0x00000001U
+/* inverted bitmask for bitfield rdm_desc_init_done_i */
+#define RDM_RX_DMA_DESC_CACHE_INIT_DONE_MSKN 0xfffffffe
+/* lower bit position of bitfield  rdm_desc_init_done_i */
+#define RDM_RX_DMA_DESC_CACHE_INIT_DONE_SHIFT 0U
+/* width of bitfield rdm_desc_init_done_i */
+#define RDM_RX_DMA_DESC_CACHE_INIT_DONE_WIDTH 1
+/* default value of bitfield rdm_desc_init_done_i */
+#define RDM_RX_DMA_DESC_CACHE_INIT_DONE_DEFAULT 0x0
+
 
 /* rx int_desc_wrb_en bitfield definitions
  * preprocessor definitions for the bitfield "int_desc_wrb_en".
@@ -296,17 +406,17 @@
  */
 
 /* register address for bitfield int_desc_wrb_en */
-#define rdm_int_desc_wrb_en_adr 0x00005a30
+#define HW_ATL_RDM_INT_DESC_WRB_EN_ADR 0x00005a30
 /* bitmask for bitfield int_desc_wrb_en */
-#define rdm_int_desc_wrb_en_msk 0x00000004
+#define HW_ATL_RDM_INT_DESC_WRB_EN_MSK 0x00000004
 /* inverted bitmask for bitfield int_desc_wrb_en */
-#define rdm_int_desc_wrb_en_mskn 0xfffffffb
+#define HW_ATL_RDM_INT_DESC_WRB_EN_MSKN 0xfffffffb
 /* lower bit position of bitfield int_desc_wrb_en */
-#define rdm_int_desc_wrb_en_shift 2
+#define HW_ATL_RDM_INT_DESC_WRB_EN_SHIFT 2
 /* width of bitfield int_desc_wrb_en */
-#define rdm_int_desc_wrb_en_width 1
+#define HW_ATL_RDM_INT_DESC_WRB_EN_WIDTH 1
 /* default value of bitfield int_desc_wrb_en */
-#define rdm_int_desc_wrb_en_default 0x0
+#define HW_ATL_RDM_INT_DESC_WRB_EN_DEFAULT 0x0
 
 /* rx dca{d}_hdr_en bitfield definitions
  * preprocessor definitions for the bitfield "dca{d}_hdr_en".
@@ -315,17 +425,17 @@
  */
 
 /* register address for bitfield dca{d}_hdr_en */
-#define rdm_dcadhdr_en_adr(dca) (0x00006100 + (dca) * 0x4)
+#define HW_ATL_RDM_DCADHDR_EN_ADR(dca) (0x00006100 + (dca) * 0x4)
 /* bitmask for bitfield dca{d}_hdr_en */
-#define rdm_dcadhdr_en_msk 0x40000000
+#define HW_ATL_RDM_DCADHDR_EN_MSK 0x40000000
 /* inverted bitmask for bitfield dca{d}_hdr_en */
-#define rdm_dcadhdr_en_mskn 0xbfffffff
+#define HW_ATL_RDM_DCADHDR_EN_MSKN 0xbfffffff
 /* lower bit position of bitfield dca{d}_hdr_en */
-#define rdm_dcadhdr_en_shift 30
+#define HW_ATL_RDM_DCADHDR_EN_SHIFT 30
 /* width of bitfield dca{d}_hdr_en */
-#define rdm_dcadhdr_en_width 1
+#define HW_ATL_RDM_DCADHDR_EN_WIDTH 1
 /* default value of bitfield dca{d}_hdr_en */
-#define rdm_dcadhdr_en_default 0x0
+#define HW_ATL_RDM_DCADHDR_EN_DEFAULT 0x0
 
 /* rx dca{d}_pay_en bitfield definitions
  * preprocessor definitions for the bitfield "dca{d}_pay_en".
@@ -334,17 +444,17 @@
  */
 
 /* register address for bitfield dca{d}_pay_en */
-#define rdm_dcadpay_en_adr(dca) (0x00006100 + (dca) * 0x4)
+#define HW_ATL_RDM_DCADPAY_EN_ADR(dca) (0x00006100 + (dca) * 0x4)
 /* bitmask for bitfield dca{d}_pay_en */
-#define rdm_dcadpay_en_msk 0x20000000
+#define HW_ATL_RDM_DCADPAY_EN_MSK 0x20000000
 /* inverted bitmask for bitfield dca{d}_pay_en */
-#define rdm_dcadpay_en_mskn 0xdfffffff
+#define HW_ATL_RDM_DCADPAY_EN_MSKN 0xdfffffff
 /* lower bit position of bitfield dca{d}_pay_en */
-#define rdm_dcadpay_en_shift 29
+#define HW_ATL_RDM_DCADPAY_EN_SHIFT 29
 /* width of bitfield dca{d}_pay_en */
-#define rdm_dcadpay_en_width 1
+#define HW_ATL_RDM_DCADPAY_EN_WIDTH 1
 /* default value of bitfield dca{d}_pay_en */
-#define rdm_dcadpay_en_default 0x0
+#define HW_ATL_RDM_DCADPAY_EN_DEFAULT 0x0
 
 /* RX rdm_int_rim_en Bitfield Definitions
  * Preprocessor definitions for the bitfield "rdm_int_rim_en".
@@ -352,51 +462,51 @@
  */
 
 /* Register address for bitfield rdm_int_rim_en */
-#define rdm_int_rim_en_adr 0x00005A30
+#define HW_ATL_RDM_INT_RIM_EN_ADR 0x00005A30
 /* Bitmask for bitfield rdm_int_rim_en */
-#define rdm_int_rim_en_msk 0x00000008
+#define HW_ATL_RDM_INT_RIM_EN_MSK 0x00000008
 /* Inverted bitmask for bitfield rdm_int_rim_en */
-#define rdm_int_rim_en_mskn 0xFFFFFFF7
+#define HW_ATL_RDM_INT_RIM_EN_MSKN 0xFFFFFFF7
 /* Lower bit position of bitfield rdm_int_rim_en */
-#define rdm_int_rim_en_shift 3
+#define HW_ATL_RDM_INT_RIM_EN_SHIFT 3
 /* Width of bitfield rdm_int_rim_en */
-#define rdm_int_rim_en_width 1
+#define HW_ATL_RDM_INT_RIM_EN_WIDTH 1
 /* Default value of bitfield rdm_int_rim_en */
-#define rdm_int_rim_en_default 0x0
+#define HW_ATL_RDM_INT_RIM_EN_DEFAULT 0x0
 
 /* general interrupt mapping register definitions
  * preprocessor definitions for general interrupt mapping register
  * base address: 0x00002180
  * parameter: regidx {f} | stride size 0x4 | range [0, 3]
  */
-#define gen_intr_map_adr(regidx) (0x00002180u + (regidx) * 0x4)
+#define HW_ATL_GEN_INTR_MAP_ADR(regidx) (0x00002180u + (regidx) * 0x4)
 
 /* general interrupt status register definitions
  * preprocessor definitions for general interrupt status register
  * address: 0x000021A0
  */
 
-#define gen_intr_stat_adr 0x000021A4U
+#define HW_ATL_GEN_INTR_STAT_ADR 0x000021A4U
 
 /* interrupt global control register  definitions
  * preprocessor definitions for interrupt global control register
  * address: 0x00002300
  */
-#define intr_glb_ctl_adr 0x00002300u
+#define HW_ATL_INTR_GLB_CTL_ADR 0x00002300u
 
 /* interrupt throttle register definitions
  * preprocessor definitions for interrupt throttle register
  * base address: 0x00002800
  * parameter: throttle {t} | stride size 0x4 | range [0, 31]
  */
-#define intr_thr_adr(throttle) (0x00002800u + (throttle) * 0x4)
+#define HW_ATL_INTR_THR_ADR(throttle) (0x00002800u + (throttle) * 0x4)
 
 /* rx dma descriptor base address lsw definitions
  * preprocessor definitions for rx dma descriptor base address lsw
  * base address: 0x00005b00
  * parameter: descriptor {d} | stride size 0x20 | range [0, 31]
  */
-#define rx_dma_desc_base_addrlsw_adr(descriptor) \
+#define HW_ATL_RX_DMA_DESC_BASE_ADDRLSW_ADR(descriptor) \
 (0x00005b00u + (descriptor) * 0x20)
 
 /* rx dma descriptor base address msw definitions
@@ -404,7 +514,7 @@
  * base address: 0x00005b04
  * parameter: descriptor {d} | stride size 0x20 | range [0, 31]
  */
-#define rx_dma_desc_base_addrmsw_adr(descriptor) \
+#define HW_ATL_RX_DMA_DESC_BASE_ADDRMSW_ADR(descriptor) \
 (0x00005b04u + (descriptor) * 0x20)
 
 /* rx dma descriptor status register definitions
@@ -412,46 +522,48 @@
  * base address: 0x00005b14
  * parameter: descriptor {d} | stride size 0x20 | range [0, 31]
  */
-#define rx_dma_desc_stat_adr(descriptor) (0x00005b14u + (descriptor) * 0x20)
+#define HW_ATL_RX_DMA_DESC_STAT_ADR(descriptor) \
+	(0x00005b14u + (descriptor) * 0x20)
 
 /* rx dma descriptor tail pointer register definitions
  * preprocessor definitions for rx dma descriptor tail pointer register
  * base address: 0x00005b10
  * parameter: descriptor {d} | stride size 0x20 | range [0, 31]
  */
-#define rx_dma_desc_tail_ptr_adr(descriptor) (0x00005b10u + (descriptor) * 0x20)
+#define HW_ATL_RX_DMA_DESC_TAIL_PTR_ADR(descriptor) \
+	(0x00005b10u + (descriptor) * 0x20)
 
 /* rx interrupt moderation control register definitions
  * Preprocessor definitions for RX Interrupt Moderation Control Register
  * Base Address: 0x00005A40
  * Parameter: RIM {R} | stride size 0x4 | range [0, 31]
  */
-#define rx_intr_moderation_ctl_adr(rim) (0x00005A40u + (rim) * 0x4)
+#define HW_ATL_RX_INTR_MODERATION_CTL_ADR(rim) (0x00005A40u + (rim) * 0x4)
 
 /* rx filter multicast filter mask register definitions
  * preprocessor definitions for rx filter multicast filter mask register
  * address: 0x00005270
  */
-#define rx_flr_mcst_flr_msk_adr 0x00005270u
+#define HW_ATL_RX_FLR_MCST_FLR_MSK_ADR 0x00005270u
 
 /* rx filter multicast filter register definitions
  * preprocessor definitions for rx filter multicast filter register
  * base address: 0x00005250
  * parameter: filter {f} | stride size 0x4 | range [0, 7]
  */
-#define rx_flr_mcst_flr_adr(filter) (0x00005250u + (filter) * 0x4)
+#define HW_ATL_RX_FLR_MCST_FLR_ADR(filter) (0x00005250u + (filter) * 0x4)
 
 /* RX Filter RSS Control Register 1 Definitions
  * Preprocessor definitions for RX Filter RSS Control Register 1
  * Address: 0x000054C0
  */
-#define rx_flr_rss_control1_adr 0x000054C0u
+#define HW_ATL_RX_FLR_RSS_CONTROL1_ADR 0x000054C0u
 
 /* RX Filter Control Register 2 Definitions
  * Preprocessor definitions for RX Filter Control Register 2
  * Address: 0x00005104
  */
-#define rx_flr_control2_adr 0x00005104u
+#define HW_ATL_RX_FLR_CONTROL2_ADR 0x00005104u
 
 /* tx tx dma debug control [1f:0] bitfield definitions
  * preprocessor definitions for the bitfield "tx dma debug control [1f:0]".
@@ -459,24 +571,24 @@
  */
 
 /* register address for bitfield tx dma debug control [1f:0] */
-#define tdm_tx_dma_debug_ctl_adr 0x00008920
+#define HW_ATL_TDM_TX_DMA_DEBUG_CTL_ADR 0x00008920
 /* bitmask for bitfield tx dma debug control [1f:0] */
-#define tdm_tx_dma_debug_ctl_msk 0xffffffff
+#define HW_ATL_TDM_TX_DMA_DEBUG_CTL_MSK 0xffffffff
 /* inverted bitmask for bitfield tx dma debug control [1f:0] */
-#define tdm_tx_dma_debug_ctl_mskn 0x00000000
+#define HW_ATL_TDM_TX_DMA_DEBUG_CTL_MSKN 0x00000000
 /* lower bit position of bitfield tx dma debug control [1f:0] */
-#define tdm_tx_dma_debug_ctl_shift 0
+#define HW_ATL_TDM_TX_DMA_DEBUG_CTL_SHIFT 0
 /* width of bitfield tx dma debug control [1f:0] */
-#define tdm_tx_dma_debug_ctl_width 32
+#define HW_ATL_TDM_TX_DMA_DEBUG_CTL_WIDTH 32
 /* default value of bitfield tx dma debug control [1f:0] */
-#define tdm_tx_dma_debug_ctl_default 0x0
+#define HW_ATL_TDM_TX_DMA_DEBUG_CTL_DEFAULT 0x0
 
 /* tx dma descriptor base address lsw definitions
  * preprocessor definitions for tx dma descriptor base address lsw
  * base address: 0x00007c00
  * parameter: descriptor {d} | stride size 0x40 | range [0, 31]
  */
-#define tx_dma_desc_base_addrlsw_adr(descriptor) \
+#define HW_ATL_TX_DMA_DESC_BASE_ADDRLSW_ADR(descriptor) \
 	(0x00007c00u + (descriptor) * 0x40)
 
 /* tx dma descriptor tail pointer register definitions
@@ -484,7 +596,8 @@
  * base address: 0x00007c10
  *  parameter: descriptor {d} | stride size 0x40 | range [0, 31]
  */
-#define tx_dma_desc_tail_ptr_adr(descriptor) (0x00007c10u + (descriptor) * 0x40)
+#define HW_ATL_TX_DMA_DESC_TAIL_PTR_ADR(descriptor) \
+	(0x00007c10u + (descriptor) * 0x40)
 
 /* rx dma_sys_loopback bitfield definitions
  * preprocessor definitions for the bitfield "dma_sys_loopback".
@@ -492,17 +605,35 @@
  */
 
 /* register address for bitfield dma_sys_loopback */
-#define rpb_dma_sys_lbk_adr 0x00005000
+#define HW_ATL_RPB_DMA_SYS_LBK_ADR 0x00005000
 /* bitmask for bitfield dma_sys_loopback */
-#define rpb_dma_sys_lbk_msk 0x00000040
+#define HW_ATL_RPB_DMA_SYS_LBK_MSK 0x00000040
 /* inverted bitmask for bitfield dma_sys_loopback */
-#define rpb_dma_sys_lbk_mskn 0xffffffbf
+#define HW_ATL_RPB_DMA_SYS_LBK_MSKN 0xffffffbf
 /* lower bit position of bitfield dma_sys_loopback */
-#define rpb_dma_sys_lbk_shift 6
+#define HW_ATL_RPB_DMA_SYS_LBK_SHIFT 6
 /* width of bitfield dma_sys_loopback */
-#define rpb_dma_sys_lbk_width 1
+#define HW_ATL_RPB_DMA_SYS_LBK_WIDTH 1
 /* default value of bitfield dma_sys_loopback */
-#define rpb_dma_sys_lbk_default 0x0
+#define HW_ATL_RPB_DMA_SYS_LBK_DEFAULT 0x0
+
+/* rx dma_net_loopback bitfield definitions
+ * preprocessor definitions for the bitfield "dma_net_loopback".
+ * port="pif_rpb_dma_net_lbk_i"
+ */
+
+/* register address for bitfield dma_net_loopback */
+#define HW_ATL_RPB_DMA_NET_LBK_ADR 0x00005000
+/* bitmask for bitfield dma_net_loopback */
+#define HW_ATL_RPB_DMA_NET_LBK_MSK 0x00000010
+/* inverted bitmask for bitfield dma_net_loopback */
+#define HW_ATL_RPB_DMA_NET_LBK_MSKN 0xffffffef
+/* lower bit position of bitfield dma_net_loopback */
+#define HW_ATL_RPB_DMA_NET_LBK_SHIFT 4
+/* width of bitfield dma_net_loopback */
+#define HW_ATL_RPB_DMA_NET_LBK_WIDTH 1
+/* default value of bitfield dma_net_loopback */
+#define HW_ATL_RPB_DMA_NET_LBK_DEFAULT 0x0
 
 /* rx rx_tc_mode bitfield definitions
  * preprocessor definitions for the bitfield "rx_tc_mode".
@@ -510,17 +641,17 @@
  */
 
 /* register address for bitfield rx_tc_mode */
-#define rpb_rpf_rx_tc_mode_adr 0x00005700
+#define HW_ATL_RPB_RPF_RX_TC_MODE_ADR 0x00005700
 /* bitmask for bitfield rx_tc_mode */
-#define rpb_rpf_rx_tc_mode_msk 0x00000100
+#define HW_ATL_RPB_RPF_RX_TC_MODE_MSK 0x00000100
 /* inverted bitmask for bitfield rx_tc_mode */
-#define rpb_rpf_rx_tc_mode_mskn 0xfffffeff
+#define HW_ATL_RPB_RPF_RX_TC_MODE_MSKN 0xfffffeff
 /* lower bit position of bitfield rx_tc_mode */
-#define rpb_rpf_rx_tc_mode_shift 8
+#define HW_ATL_RPB_RPF_RX_TC_MODE_SHIFT 8
 /* width of bitfield rx_tc_mode */
-#define rpb_rpf_rx_tc_mode_width 1
+#define HW_ATL_RPB_RPF_RX_TC_MODE_WIDTH 1
 /* default value of bitfield rx_tc_mode */
-#define rpb_rpf_rx_tc_mode_default 0x0
+#define HW_ATL_RPB_RPF_RX_TC_MODE_DEFAULT 0x0
 
 /* rx rx_buf_en bitfield definitions
  * preprocessor definitions for the bitfield "rx_buf_en".
@@ -528,17 +659,17 @@
  */
 
 /* register address for bitfield rx_buf_en */
-#define rpb_rx_buf_en_adr 0x00005700
+#define HW_ATL_RPB_RX_BUF_EN_ADR 0x00005700
 /* bitmask for bitfield rx_buf_en */
-#define rpb_rx_buf_en_msk 0x00000001
+#define HW_ATL_RPB_RX_BUF_EN_MSK 0x00000001
 /* inverted bitmask for bitfield rx_buf_en */
-#define rpb_rx_buf_en_mskn 0xfffffffe
+#define HW_ATL_RPB_RX_BUF_EN_MSKN 0xfffffffe
 /* lower bit position of bitfield rx_buf_en */
-#define rpb_rx_buf_en_shift 0
+#define HW_ATL_RPB_RX_BUF_EN_SHIFT 0
 /* width of bitfield rx_buf_en */
-#define rpb_rx_buf_en_width 1
+#define HW_ATL_RPB_RX_BUF_EN_WIDTH 1
 /* default value of bitfield rx_buf_en */
-#define rpb_rx_buf_en_default 0x0
+#define HW_ATL_RPB_RX_BUF_EN_DEFAULT 0x0
 
 /* rx rx{b}_hi_thresh[d:0] bitfield definitions
  * preprocessor definitions for the bitfield "rx{b}_hi_thresh[d:0]".
@@ -547,17 +678,17 @@
  */
 
 /* register address for bitfield rx{b}_hi_thresh[d:0] */
-#define rpb_rxbhi_thresh_adr(buffer) (0x00005714 + (buffer) * 0x10)
+#define HW_ATL_RPB_RXBHI_THRESH_ADR(buffer) (0x00005714 + (buffer) * 0x10)
 /* bitmask for bitfield rx{b}_hi_thresh[d:0] */
-#define rpb_rxbhi_thresh_msk 0x3fff0000
+#define HW_ATL_RPB_RXBHI_THRESH_MSK 0x3fff0000
 /* inverted bitmask for bitfield rx{b}_hi_thresh[d:0] */
-#define rpb_rxbhi_thresh_mskn 0xc000ffff
+#define HW_ATL_RPB_RXBHI_THRESH_MSKN 0xc000ffff
 /* lower bit position of bitfield rx{b}_hi_thresh[d:0] */
-#define rpb_rxbhi_thresh_shift 16
+#define HW_ATL_RPB_RXBHI_THRESH_SHIFT 16
 /* width of bitfield rx{b}_hi_thresh[d:0] */
-#define rpb_rxbhi_thresh_width 14
+#define HW_ATL_RPB_RXBHI_THRESH_WIDTH 14
 /* default value of bitfield rx{b}_hi_thresh[d:0] */
-#define rpb_rxbhi_thresh_default 0x0
+#define HW_ATL_RPB_RXBHI_THRESH_DEFAULT 0x0
 
 /* rx rx{b}_lo_thresh[d:0] bitfield definitions
  * preprocessor definitions for the bitfield "rx{b}_lo_thresh[d:0]".
@@ -566,17 +697,17 @@
  */
 
 /* register address for bitfield rx{b}_lo_thresh[d:0] */
-#define rpb_rxblo_thresh_adr(buffer) (0x00005714 + (buffer) * 0x10)
+#define HW_ATL_RPB_RXBLO_THRESH_ADR(buffer) (0x00005714 + (buffer) * 0x10)
 /* bitmask for bitfield rx{b}_lo_thresh[d:0] */
-#define rpb_rxblo_thresh_msk 0x00003fff
+#define HW_ATL_RPB_RXBLO_THRESH_MSK 0x00003fff
 /* inverted bitmask for bitfield rx{b}_lo_thresh[d:0] */
-#define rpb_rxblo_thresh_mskn 0xffffc000
+#define HW_ATL_RPB_RXBLO_THRESH_MSKN 0xffffc000
 /* lower bit position of bitfield rx{b}_lo_thresh[d:0] */
-#define rpb_rxblo_thresh_shift 0
+#define HW_ATL_RPB_RXBLO_THRESH_SHIFT 0
 /* width of bitfield rx{b}_lo_thresh[d:0] */
-#define rpb_rxblo_thresh_width 14
+#define HW_ATL_RPB_RXBLO_THRESH_WIDTH 14
 /* default value of bitfield rx{b}_lo_thresh[d:0] */
-#define rpb_rxblo_thresh_default 0x0
+#define HW_ATL_RPB_RXBLO_THRESH_DEFAULT 0x0
 
 /* rx rx_fc_mode[1:0] bitfield definitions
  * preprocessor definitions for the bitfield "rx_fc_mode[1:0]".
@@ -584,17 +715,17 @@
  */
 
 /* register address for bitfield rx_fc_mode[1:0] */
-#define rpb_rx_fc_mode_adr 0x00005700
+#define HW_ATL_RPB_RX_FC_MODE_ADR 0x00005700
 /* bitmask for bitfield rx_fc_mode[1:0] */
-#define rpb_rx_fc_mode_msk 0x00000030
+#define HW_ATL_RPB_RX_FC_MODE_MSK 0x00000030
 /* inverted bitmask for bitfield rx_fc_mode[1:0] */
-#define rpb_rx_fc_mode_mskn 0xffffffcf
+#define HW_ATL_RPB_RX_FC_MODE_MSKN 0xffffffcf
 /* lower bit position of bitfield rx_fc_mode[1:0] */
-#define rpb_rx_fc_mode_shift 4
+#define HW_ATL_RPB_RX_FC_MODE_SHIFT 4
 /* width of bitfield rx_fc_mode[1:0] */
-#define rpb_rx_fc_mode_width 2
+#define HW_ATL_RPB_RX_FC_MODE_WIDTH 2
 /* default value of bitfield rx_fc_mode[1:0] */
-#define rpb_rx_fc_mode_default 0x0
+#define HW_ATL_RPB_RX_FC_MODE_DEFAULT 0x0
 
 /* rx rx{b}_buf_size[8:0] bitfield definitions
  * preprocessor definitions for the bitfield "rx{b}_buf_size[8:0]".
@@ -603,17 +734,17 @@
  */
 
 /* register address for bitfield rx{b}_buf_size[8:0] */
-#define rpb_rxbbuf_size_adr(buffer) (0x00005710 + (buffer) * 0x10)
+#define HW_ATL_RPB_RXBBUF_SIZE_ADR(buffer) (0x00005710 + (buffer) * 0x10)
 /* bitmask for bitfield rx{b}_buf_size[8:0] */
-#define rpb_rxbbuf_size_msk 0x000001ff
+#define HW_ATL_RPB_RXBBUF_SIZE_MSK 0x000001ff
 /* inverted bitmask for bitfield rx{b}_buf_size[8:0] */
-#define rpb_rxbbuf_size_mskn 0xfffffe00
+#define HW_ATL_RPB_RXBBUF_SIZE_MSKN 0xfffffe00
 /* lower bit position of bitfield rx{b}_buf_size[8:0] */
-#define rpb_rxbbuf_size_shift 0
+#define HW_ATL_RPB_RXBBUF_SIZE_SHIFT 0
 /* width of bitfield rx{b}_buf_size[8:0] */
-#define rpb_rxbbuf_size_width 9
+#define HW_ATL_RPB_RXBBUF_SIZE_WIDTH 9
 /* default value of bitfield rx{b}_buf_size[8:0] */
-#define rpb_rxbbuf_size_default 0x0
+#define HW_ATL_RPB_RXBBUF_SIZE_DEFAULT 0x0
 
 /* rx rx{b}_xoff_en bitfield definitions
  * preprocessor definitions for the bitfield "rx{b}_xoff_en".
@@ -622,17 +753,17 @@
  */
 
 /* register address for bitfield rx{b}_xoff_en */
-#define rpb_rxbxoff_en_adr(buffer) (0x00005714 + (buffer) * 0x10)
+#define HW_ATL_RPB_RXBXOFF_EN_ADR(buffer) (0x00005714 + (buffer) * 0x10)
 /* bitmask for bitfield rx{b}_xoff_en */
-#define rpb_rxbxoff_en_msk 0x80000000
+#define HW_ATL_RPB_RXBXOFF_EN_MSK 0x80000000
 /* inverted bitmask for bitfield rx{b}_xoff_en */
-#define rpb_rxbxoff_en_mskn 0x7fffffff
+#define HW_ATL_RPB_RXBXOFF_EN_MSKN 0x7fffffff
 /* lower bit position of bitfield rx{b}_xoff_en */
-#define rpb_rxbxoff_en_shift 31
+#define HW_ATL_RPB_RXBXOFF_EN_SHIFT 31
 /* width of bitfield rx{b}_xoff_en */
-#define rpb_rxbxoff_en_width 1
+#define HW_ATL_RPB_RXBXOFF_EN_WIDTH 1
 /* default value of bitfield rx{b}_xoff_en */
-#define rpb_rxbxoff_en_default 0x0
+#define HW_ATL_RPB_RXBXOFF_EN_DEFAULT 0x0
 
 /* rx l2_bc_thresh[f:0] bitfield definitions
  * preprocessor definitions for the bitfield "l2_bc_thresh[f:0]".
@@ -640,17 +771,17 @@
  */
 
 /* register address for bitfield l2_bc_thresh[f:0] */
-#define rpfl2bc_thresh_adr 0x00005100
+#define HW_ATL_RPFL2BC_THRESH_ADR 0x00005100
 /* bitmask for bitfield l2_bc_thresh[f:0] */
-#define rpfl2bc_thresh_msk 0xffff0000
+#define HW_ATL_RPFL2BC_THRESH_MSK 0xffff0000
 /* inverted bitmask for bitfield l2_bc_thresh[f:0] */
-#define rpfl2bc_thresh_mskn 0x0000ffff
+#define HW_ATL_RPFL2BC_THRESH_MSKN 0x0000ffff
 /* lower bit position of bitfield l2_bc_thresh[f:0] */
-#define rpfl2bc_thresh_shift 16
+#define HW_ATL_RPFL2BC_THRESH_SHIFT 16
 /* width of bitfield l2_bc_thresh[f:0] */
-#define rpfl2bc_thresh_width 16
+#define HW_ATL_RPFL2BC_THRESH_WIDTH 16
 /* default value of bitfield l2_bc_thresh[f:0] */
-#define rpfl2bc_thresh_default 0x0
+#define HW_ATL_RPFL2BC_THRESH_DEFAULT 0x0
 
 /* rx l2_bc_en bitfield definitions
  * preprocessor definitions for the bitfield "l2_bc_en".
@@ -658,17 +789,17 @@
  */
 
 /* register address for bitfield l2_bc_en */
-#define rpfl2bc_en_adr 0x00005100
+#define HW_ATL_RPFL2BC_EN_ADR 0x00005100
 /* bitmask for bitfield l2_bc_en */
-#define rpfl2bc_en_msk 0x00000001
+#define HW_ATL_RPFL2BC_EN_MSK 0x00000001
 /* inverted bitmask for bitfield l2_bc_en */
-#define rpfl2bc_en_mskn 0xfffffffe
+#define HW_ATL_RPFL2BC_EN_MSKN 0xfffffffe
 /* lower bit position of bitfield l2_bc_en */
-#define rpfl2bc_en_shift 0
+#define HW_ATL_RPFL2BC_EN_SHIFT 0
 /* width of bitfield l2_bc_en */
-#define rpfl2bc_en_width 1
+#define HW_ATL_RPFL2BC_EN_WIDTH 1
 /* default value of bitfield l2_bc_en */
-#define rpfl2bc_en_default 0x0
+#define HW_ATL_RPFL2BC_EN_DEFAULT 0x0
 
 /* rx l2_bc_act[2:0] bitfield definitions
  * preprocessor definitions for the bitfield "l2_bc_act[2:0]".
@@ -676,17 +807,17 @@
  */
 
 /* register address for bitfield l2_bc_act[2:0] */
-#define rpfl2bc_act_adr 0x00005100
+#define HW_ATL_RPFL2BC_ACT_ADR 0x00005100
 /* bitmask for bitfield l2_bc_act[2:0] */
-#define rpfl2bc_act_msk 0x00007000
+#define HW_ATL_RPFL2BC_ACT_MSK 0x00007000
 /* inverted bitmask for bitfield l2_bc_act[2:0] */
-#define rpfl2bc_act_mskn 0xffff8fff
+#define HW_ATL_RPFL2BC_ACT_MSKN 0xffff8fff
 /* lower bit position of bitfield l2_bc_act[2:0] */
-#define rpfl2bc_act_shift 12
+#define HW_ATL_RPFL2BC_ACT_SHIFT 12
 /* width of bitfield l2_bc_act[2:0] */
-#define rpfl2bc_act_width 3
+#define HW_ATL_RPFL2BC_ACT_WIDTH 3
 /* default value of bitfield l2_bc_act[2:0] */
-#define rpfl2bc_act_default 0x0
+#define HW_ATL_RPFL2BC_ACT_DEFAULT 0x0
 
 /* rx l2_mc_en{f} bitfield definitions
  * preprocessor definitions for the bitfield "l2_mc_en{f}".
@@ -695,17 +826,17 @@
  */
 
 /* register address for bitfield l2_mc_en{f} */
-#define rpfl2mc_enf_adr(filter) (0x00005250 + (filter) * 0x4)
+#define HW_ATL_RPFL2MC_ENF_ADR(filter) (0x00005250 + (filter) * 0x4)
 /* bitmask for bitfield l2_mc_en{f} */
-#define rpfl2mc_enf_msk 0x80000000
+#define HW_ATL_RPFL2MC_ENF_MSK 0x80000000
 /* inverted bitmask for bitfield l2_mc_en{f} */
-#define rpfl2mc_enf_mskn 0x7fffffff
+#define HW_ATL_RPFL2MC_ENF_MSKN 0x7fffffff
 /* lower bit position of bitfield l2_mc_en{f} */
-#define rpfl2mc_enf_shift 31
+#define HW_ATL_RPFL2MC_ENF_SHIFT 31
 /* width of bitfield l2_mc_en{f} */
-#define rpfl2mc_enf_width 1
+#define HW_ATL_RPFL2MC_ENF_WIDTH 1
 /* default value of bitfield l2_mc_en{f} */
-#define rpfl2mc_enf_default 0x0
+#define HW_ATL_RPFL2MC_ENF_DEFAULT 0x0
 
 /* rx l2_promis_mode bitfield definitions
  * preprocessor definitions for the bitfield "l2_promis_mode".
@@ -713,17 +844,17 @@
  */
 
 /* register address for bitfield l2_promis_mode */
-#define rpfl2promis_mode_adr 0x00005100
+#define HW_ATL_RPFL2PROMIS_MODE_ADR 0x00005100
 /* bitmask for bitfield l2_promis_mode */
-#define rpfl2promis_mode_msk 0x00000008
+#define HW_ATL_RPFL2PROMIS_MODE_MSK 0x00000008
 /* inverted bitmask for bitfield l2_promis_mode */
-#define rpfl2promis_mode_mskn 0xfffffff7
+#define HW_ATL_RPFL2PROMIS_MODE_MSKN 0xfffffff7
 /* lower bit position of bitfield l2_promis_mode */
-#define rpfl2promis_mode_shift 3
+#define HW_ATL_RPFL2PROMIS_MODE_SHIFT 3
 /* width of bitfield l2_promis_mode */
-#define rpfl2promis_mode_width 1
+#define HW_ATL_RPFL2PROMIS_MODE_WIDTH 1
 /* default value of bitfield l2_promis_mode */
-#define rpfl2promis_mode_default 0x0
+#define HW_ATL_RPFL2PROMIS_MODE_DEFAULT 0x0
 
 /* rx l2_uc_act{f}[2:0] bitfield definitions
  * preprocessor definitions for the bitfield "l2_uc_act{f}[2:0]".
@@ -732,17 +863,17 @@
  */
 
 /* register address for bitfield l2_uc_act{f}[2:0] */
-#define rpfl2uc_actf_adr(filter) (0x00005114 + (filter) * 0x8)
+#define HW_ATL_RPFL2UC_ACTF_ADR(filter) (0x00005114 + (filter) * 0x8)
 /* bitmask for bitfield l2_uc_act{f}[2:0] */
-#define rpfl2uc_actf_msk 0x00070000
+#define HW_ATL_RPFL2UC_ACTF_MSK 0x00070000
 /* inverted bitmask for bitfield l2_uc_act{f}[2:0] */
-#define rpfl2uc_actf_mskn 0xfff8ffff
+#define HW_ATL_RPFL2UC_ACTF_MSKN 0xfff8ffff
 /* lower bit position of bitfield l2_uc_act{f}[2:0] */
-#define rpfl2uc_actf_shift 16
+#define HW_ATL_RPFL2UC_ACTF_SHIFT 16
 /* width of bitfield l2_uc_act{f}[2:0] */
-#define rpfl2uc_actf_width 3
+#define HW_ATL_RPFL2UC_ACTF_WIDTH 3
 /* default value of bitfield l2_uc_act{f}[2:0] */
-#define rpfl2uc_actf_default 0x0
+#define HW_ATL_RPFL2UC_ACTF_DEFAULT 0x0
 
 /* rx l2_uc_en{f} bitfield definitions
  * preprocessor definitions for the bitfield "l2_uc_en{f}".
@@ -751,26 +882,26 @@
  */
 
 /* register address for bitfield l2_uc_en{f} */
-#define rpfl2uc_enf_adr(filter) (0x00005114 + (filter) * 0x8)
+#define HW_ATL_RPFL2UC_ENF_ADR(filter) (0x00005114 + (filter) * 0x8)
 /* bitmask for bitfield l2_uc_en{f} */
-#define rpfl2uc_enf_msk 0x80000000
+#define HW_ATL_RPFL2UC_ENF_MSK 0x80000000
 /* inverted bitmask for bitfield l2_uc_en{f} */
-#define rpfl2uc_enf_mskn 0x7fffffff
+#define HW_ATL_RPFL2UC_ENF_MSKN 0x7fffffff
 /* lower bit position of bitfield l2_uc_en{f} */
-#define rpfl2uc_enf_shift 31
+#define HW_ATL_RPFL2UC_ENF_SHIFT 31
 /* width of bitfield l2_uc_en{f} */
-#define rpfl2uc_enf_width 1
+#define HW_ATL_RPFL2UC_ENF_WIDTH 1
 /* default value of bitfield l2_uc_en{f} */
-#define rpfl2uc_enf_default 0x0
+#define HW_ATL_RPFL2UC_ENF_DEFAULT 0x0
 
 /* register address for bitfield l2_uc_da{f}_lsw[1f:0] */
-#define rpfl2uc_daflsw_adr(filter) (0x00005110 + (filter) * 0x8)
+#define HW_ATL_RPFL2UC_DAFLSW_ADR(filter) (0x00005110 + (filter) * 0x8)
 /* register address for bitfield l2_uc_da{f}_msw[f:0] */
-#define rpfl2uc_dafmsw_adr(filter) (0x00005114 + (filter) * 0x8)
+#define HW_ATL_RPFL2UC_DAFMSW_ADR(filter) (0x00005114 + (filter) * 0x8)
 /* bitmask for bitfield l2_uc_da{f}_msw[f:0] */
-#define rpfl2uc_dafmsw_msk 0x0000ffff
+#define HW_ATL_RPFL2UC_DAFMSW_MSK 0x0000ffff
 /* lower bit position of bitfield l2_uc_da{f}_msw[f:0] */
-#define rpfl2uc_dafmsw_shift 0
+#define HW_ATL_RPFL2UC_DAFMSW_SHIFT 0
 
 /* rx l2_mc_accept_all bitfield definitions
  * Preprocessor definitions for the bitfield "l2_mc_accept_all".
@@ -778,22 +909,22 @@
  */
 
 /* Register address for bitfield l2_mc_accept_all */
-#define rpfl2mc_accept_all_adr 0x00005270
+#define HW_ATL_RPFL2MC_ACCEPT_ALL_ADR 0x00005270
 /* Bitmask for bitfield l2_mc_accept_all */
-#define rpfl2mc_accept_all_msk 0x00004000
+#define HW_ATL_RPFL2MC_ACCEPT_ALL_MSK 0x00004000
 /* Inverted bitmask for bitfield l2_mc_accept_all */
-#define rpfl2mc_accept_all_mskn 0xFFFFBFFF
+#define HW_ATL_RPFL2MC_ACCEPT_ALL_MSKN 0xFFFFBFFF
 /* Lower bit position of bitfield l2_mc_accept_all */
-#define rpfl2mc_accept_all_shift 14
+#define HW_ATL_RPFL2MC_ACCEPT_ALL_SHIFT 14
 /* Width of bitfield l2_mc_accept_all */
-#define rpfl2mc_accept_all_width 1
+#define HW_ATL_RPFL2MC_ACCEPT_ALL_WIDTH 1
 /* Default value of bitfield l2_mc_accept_all */
-#define rpfl2mc_accept_all_default 0x0
+#define HW_ATL_RPFL2MC_ACCEPT_ALL_DEFAULT 0x0
 
 /* width of bitfield rx_tc_up{t}[2:0] */
-#define rpf_rpb_rx_tc_upt_width 3
+#define HW_ATL_RPF_RPB_RX_TC_UPT_WIDTH 3
 /* default value of bitfield rx_tc_up{t}[2:0] */
-#define rpf_rpb_rx_tc_upt_default 0x0
+#define HW_ATL_RPF_RPB_RX_TC_UPT_DEFAULT 0x0
 
 /* rx rss_key_addr[4:0] bitfield definitions
  * preprocessor definitions for the bitfield "rss_key_addr[4:0]".
@@ -801,17 +932,17 @@
  */
 
 /* register address for bitfield rss_key_addr[4:0] */
-#define rpf_rss_key_addr_adr 0x000054d0
+#define HW_ATL_RPF_RSS_KEY_ADDR_ADR 0x000054d0
 /* bitmask for bitfield rss_key_addr[4:0] */
-#define rpf_rss_key_addr_msk 0x0000001f
+#define HW_ATL_RPF_RSS_KEY_ADDR_MSK 0x0000001f
 /* inverted bitmask for bitfield rss_key_addr[4:0] */
-#define rpf_rss_key_addr_mskn 0xffffffe0
+#define HW_ATL_RPF_RSS_KEY_ADDR_MSKN 0xffffffe0
 /* lower bit position of bitfield rss_key_addr[4:0] */
-#define rpf_rss_key_addr_shift 0
+#define HW_ATL_RPF_RSS_KEY_ADDR_SHIFT 0
 /* width of bitfield rss_key_addr[4:0] */
-#define rpf_rss_key_addr_width 5
+#define HW_ATL_RPF_RSS_KEY_ADDR_WIDTH 5
 /* default value of bitfield rss_key_addr[4:0] */
-#define rpf_rss_key_addr_default 0x0
+#define HW_ATL_RPF_RSS_KEY_ADDR_DEFAULT 0x0
 
 /* rx rss_key_wr_data[1f:0] bitfield definitions
  * preprocessor definitions for the bitfield "rss_key_wr_data[1f:0]".
@@ -819,17 +950,17 @@
  */
 
 /* register address for bitfield rss_key_wr_data[1f:0] */
-#define rpf_rss_key_wr_data_adr 0x000054d4
+#define HW_ATL_RPF_RSS_KEY_WR_DATA_ADR 0x000054d4
 /* bitmask for bitfield rss_key_wr_data[1f:0] */
-#define rpf_rss_key_wr_data_msk 0xffffffff
+#define HW_ATL_RPF_RSS_KEY_WR_DATA_MSK 0xffffffff
 /* inverted bitmask for bitfield rss_key_wr_data[1f:0] */
-#define rpf_rss_key_wr_data_mskn 0x00000000
+#define HW_ATL_RPF_RSS_KEY_WR_DATA_MSKN 0x00000000
 /* lower bit position of bitfield rss_key_wr_data[1f:0] */
-#define rpf_rss_key_wr_data_shift 0
+#define HW_ATL_RPF_RSS_KEY_WR_DATA_SHIFT 0
 /* width of bitfield rss_key_wr_data[1f:0] */
-#define rpf_rss_key_wr_data_width 32
+#define HW_ATL_RPF_RSS_KEY_WR_DATA_WIDTH 32
 /* default value of bitfield rss_key_wr_data[1f:0] */
-#define rpf_rss_key_wr_data_default 0x0
+#define HW_ATL_RPF_RSS_KEY_WR_DATA_DEFAULT 0x0
 
 /* rx rss_key_wr_en_i bitfield definitions
  * preprocessor definitions for the bitfield "rss_key_wr_en_i".
@@ -837,17 +968,17 @@
  */
 
 /* register address for bitfield rss_key_wr_en_i */
-#define rpf_rss_key_wr_eni_adr 0x000054d0
+#define HW_ATL_RPF_RSS_KEY_WR_ENI_ADR 0x000054d0
 /* bitmask for bitfield rss_key_wr_en_i */
-#define rpf_rss_key_wr_eni_msk 0x00000020
+#define HW_ATL_RPF_RSS_KEY_WR_ENI_MSK 0x00000020
 /* inverted bitmask for bitfield rss_key_wr_en_i */
-#define rpf_rss_key_wr_eni_mskn 0xffffffdf
+#define HW_ATL_RPF_RSS_KEY_WR_ENI_MSKN 0xffffffdf
 /* lower bit position of bitfield rss_key_wr_en_i */
-#define rpf_rss_key_wr_eni_shift 5
+#define HW_ATL_RPF_RSS_KEY_WR_ENI_SHIFT 5
 /* width of bitfield rss_key_wr_en_i */
-#define rpf_rss_key_wr_eni_width 1
+#define HW_ATL_RPF_RSS_KEY_WR_ENI_WIDTH 1
 /* default value of bitfield rss_key_wr_en_i */
-#define rpf_rss_key_wr_eni_default 0x0
+#define HW_ATL_RPF_RSS_KEY_WR_ENI_DEFAULT 0x0
 
 /* rx rss_redir_addr[3:0] bitfield definitions
  * preprocessor definitions for the bitfield "rss_redir_addr[3:0]".
@@ -855,17 +986,17 @@
  */
 
 /* register address for bitfield rss_redir_addr[3:0] */
-#define rpf_rss_redir_addr_adr 0x000054e0
+#define HW_ATL_RPF_RSS_REDIR_ADDR_ADR 0x000054e0
 /* bitmask for bitfield rss_redir_addr[3:0] */
-#define rpf_rss_redir_addr_msk 0x0000000f
+#define HW_ATL_RPF_RSS_REDIR_ADDR_MSK 0x0000000f
 /* inverted bitmask for bitfield rss_redir_addr[3:0] */
-#define rpf_rss_redir_addr_mskn 0xfffffff0
+#define HW_ATL_RPF_RSS_REDIR_ADDR_MSKN 0xfffffff0
 /* lower bit position of bitfield rss_redir_addr[3:0] */
-#define rpf_rss_redir_addr_shift 0
+#define HW_ATL_RPF_RSS_REDIR_ADDR_SHIFT 0
 /* width of bitfield rss_redir_addr[3:0] */
-#define rpf_rss_redir_addr_width 4
+#define HW_ATL_RPF_RSS_REDIR_ADDR_WIDTH 4
 /* default value of bitfield rss_redir_addr[3:0] */
-#define rpf_rss_redir_addr_default 0x0
+#define HW_ATL_RPF_RSS_REDIR_ADDR_DEFAULT 0x0
 
 /* rx rss_redir_wr_data[f:0] bitfield definitions
  * preprocessor definitions for the bitfield "rss_redir_wr_data[f:0]".
@@ -873,17 +1004,17 @@
  */
 
 /* register address for bitfield rss_redir_wr_data[f:0] */
-#define rpf_rss_redir_wr_data_adr 0x000054e4
+#define HW_ATL_RPF_RSS_REDIR_WR_DATA_ADR 0x000054e4
 /* bitmask for bitfield rss_redir_wr_data[f:0] */
-#define rpf_rss_redir_wr_data_msk 0x0000ffff
+#define HW_ATL_RPF_RSS_REDIR_WR_DATA_MSK 0x0000ffff
 /* inverted bitmask for bitfield rss_redir_wr_data[f:0] */
-#define rpf_rss_redir_wr_data_mskn 0xffff0000
+#define HW_ATL_RPF_RSS_REDIR_WR_DATA_MSKN 0xffff0000
 /* lower bit position of bitfield rss_redir_wr_data[f:0] */
-#define rpf_rss_redir_wr_data_shift 0
+#define HW_ATL_RPF_RSS_REDIR_WR_DATA_SHIFT 0
 /* width of bitfield rss_redir_wr_data[f:0] */
-#define rpf_rss_redir_wr_data_width 16
+#define HW_ATL_RPF_RSS_REDIR_WR_DATA_WIDTH 16
 /* default value of bitfield rss_redir_wr_data[f:0] */
-#define rpf_rss_redir_wr_data_default 0x0
+#define HW_ATL_RPF_RSS_REDIR_WR_DATA_DEFAULT 0x0
 
 /* rx rss_redir_wr_en_i bitfield definitions
  * preprocessor definitions for the bitfield "rss_redir_wr_en_i".
@@ -891,17 +1022,17 @@
  */
 
 /* register address for bitfield rss_redir_wr_en_i */
-#define rpf_rss_redir_wr_eni_adr 0x000054e0
+#define HW_ATL_RPF_RSS_REDIR_WR_ENI_ADR 0x000054e0
 /* bitmask for bitfield rss_redir_wr_en_i */
-#define rpf_rss_redir_wr_eni_msk 0x00000010
+#define HW_ATL_RPF_RSS_REDIR_WR_ENI_MSK 0x00000010
 /* inverted bitmask for bitfield rss_redir_wr_en_i */
-#define rpf_rss_redir_wr_eni_mskn 0xffffffef
+#define HW_ATL_RPF_RSS_REDIR_WR_ENI_MSKN 0xffffffef
 /* lower bit position of bitfield rss_redir_wr_en_i */
-#define rpf_rss_redir_wr_eni_shift 4
+#define HW_ATL_RPF_RSS_REDIR_WR_ENI_SHIFT 4
 /* width of bitfield rss_redir_wr_en_i */
-#define rpf_rss_redir_wr_eni_width 1
+#define HW_ATL_RPF_RSS_REDIR_WR_ENI_WIDTH 1
 /* default value of bitfield rss_redir_wr_en_i */
-#define rpf_rss_redir_wr_eni_default 0x0
+#define HW_ATL_RPF_RSS_REDIR_WR_ENI_DEFAULT 0x0
 
 /* rx tpo_rpf_sys_loopback bitfield definitions
  * preprocessor definitions for the bitfield "tpo_rpf_sys_loopback".
@@ -909,17 +1040,17 @@
  */
 
 /* register address for bitfield tpo_rpf_sys_loopback */
-#define rpf_tpo_rpf_sys_lbk_adr 0x00005000
+#define HW_ATL_RPF_TPO_RPF_SYS_LBK_ADR 0x00005000
 /* bitmask for bitfield tpo_rpf_sys_loopback */
-#define rpf_tpo_rpf_sys_lbk_msk 0x00000100
+#define HW_ATL_RPF_TPO_RPF_SYS_LBK_MSK 0x00000100
 /* inverted bitmask for bitfield tpo_rpf_sys_loopback */
-#define rpf_tpo_rpf_sys_lbk_mskn 0xfffffeff
+#define HW_ATL_RPF_TPO_RPF_SYS_LBK_MSKN 0xfffffeff
 /* lower bit position of bitfield tpo_rpf_sys_loopback */
-#define rpf_tpo_rpf_sys_lbk_shift 8
+#define HW_ATL_RPF_TPO_RPF_SYS_LBK_SHIFT 8
 /* width of bitfield tpo_rpf_sys_loopback */
-#define rpf_tpo_rpf_sys_lbk_width 1
+#define HW_ATL_RPF_TPO_RPF_SYS_LBK_WIDTH 1
 /* default value of bitfield tpo_rpf_sys_loopback */
-#define rpf_tpo_rpf_sys_lbk_default 0x0
+#define HW_ATL_RPF_TPO_RPF_SYS_LBK_DEFAULT 0x0
 
 /* rx vl_inner_tpid[f:0] bitfield definitions
  * preprocessor definitions for the bitfield "vl_inner_tpid[f:0]".
@@ -927,17 +1058,17 @@
  */
 
 /* register address for bitfield vl_inner_tpid[f:0] */
-#define rpf_vl_inner_tpid_adr 0x00005284
+#define HW_ATL_RPF_VL_INNER_TPID_ADR 0x00005284
 /* bitmask for bitfield vl_inner_tpid[f:0] */
-#define rpf_vl_inner_tpid_msk 0x0000ffff
+#define HW_ATL_RPF_VL_INNER_TPID_MSK 0x0000ffff
 /* inverted bitmask for bitfield vl_inner_tpid[f:0] */
-#define rpf_vl_inner_tpid_mskn 0xffff0000
+#define HW_ATL_RPF_VL_INNER_TPID_MSKN 0xffff0000
 /* lower bit position of bitfield vl_inner_tpid[f:0] */
-#define rpf_vl_inner_tpid_shift 0
+#define HW_ATL_RPF_VL_INNER_TPID_SHIFT 0
 /* width of bitfield vl_inner_tpid[f:0] */
-#define rpf_vl_inner_tpid_width 16
+#define HW_ATL_RPF_VL_INNER_TPID_WIDTH 16
 /* default value of bitfield vl_inner_tpid[f:0] */
-#define rpf_vl_inner_tpid_default 0x8100
+#define HW_ATL_RPF_VL_INNER_TPID_DEFAULT 0x8100
 
 /* rx vl_outer_tpid[f:0] bitfield definitions
  * preprocessor definitions for the bitfield "vl_outer_tpid[f:0]".
@@ -945,17 +1076,17 @@
  */
 
 /* register address for bitfield vl_outer_tpid[f:0] */
-#define rpf_vl_outer_tpid_adr 0x00005284
+#define HW_ATL_RPF_VL_OUTER_TPID_ADR 0x00005284
 /* bitmask for bitfield vl_outer_tpid[f:0] */
-#define rpf_vl_outer_tpid_msk 0xffff0000
+#define HW_ATL_RPF_VL_OUTER_TPID_MSK 0xffff0000
 /* inverted bitmask for bitfield vl_outer_tpid[f:0] */
-#define rpf_vl_outer_tpid_mskn 0x0000ffff
+#define HW_ATL_RPF_VL_OUTER_TPID_MSKN 0x0000ffff
 /* lower bit position of bitfield vl_outer_tpid[f:0] */
-#define rpf_vl_outer_tpid_shift 16
+#define HW_ATL_RPF_VL_OUTER_TPID_SHIFT 16
 /* width of bitfield vl_outer_tpid[f:0] */
-#define rpf_vl_outer_tpid_width 16
+#define HW_ATL_RPF_VL_OUTER_TPID_WIDTH 16
 /* default value of bitfield vl_outer_tpid[f:0] */
-#define rpf_vl_outer_tpid_default 0x88a8
+#define HW_ATL_RPF_VL_OUTER_TPID_DEFAULT 0x88a8
 
 /* rx vl_promis_mode bitfield definitions
  * preprocessor definitions for the bitfield "vl_promis_mode".
@@ -963,17 +1094,17 @@
  */
 
 /* register address for bitfield vl_promis_mode */
-#define rpf_vl_promis_mode_adr 0x00005280
+#define HW_ATL_RPF_VL_PROMIS_MODE_ADR 0x00005280
 /* bitmask for bitfield vl_promis_mode */
-#define rpf_vl_promis_mode_msk 0x00000002
+#define HW_ATL_RPF_VL_PROMIS_MODE_MSK 0x00000002
 /* inverted bitmask for bitfield vl_promis_mode */
-#define rpf_vl_promis_mode_mskn 0xfffffffd
+#define HW_ATL_RPF_VL_PROMIS_MODE_MSKN 0xfffffffd
 /* lower bit position of bitfield vl_promis_mode */
-#define rpf_vl_promis_mode_shift 1
+#define HW_ATL_RPF_VL_PROMIS_MODE_SHIFT 1
 /* width of bitfield vl_promis_mode */
-#define rpf_vl_promis_mode_width 1
+#define HW_ATL_RPF_VL_PROMIS_MODE_WIDTH 1
 /* default value of bitfield vl_promis_mode */
-#define rpf_vl_promis_mode_default 0x0
+#define HW_ATL_RPF_VL_PROMIS_MODE_DEFAULT 0x0
 
 /* RX vl_accept_untagged_mode Bitfield Definitions
  * Preprocessor definitions for the bitfield "vl_accept_untagged_mode".
@@ -981,17 +1112,17 @@
  */
 
 /* Register address for bitfield vl_accept_untagged_mode */
-#define rpf_vl_accept_untagged_mode_adr 0x00005280
+#define HW_ATL_RPF_VL_ACCEPT_UNTAGGED_MODE_ADR 0x00005280
 /* Bitmask for bitfield vl_accept_untagged_mode */
-#define rpf_vl_accept_untagged_mode_msk 0x00000004
+#define HW_ATL_RPF_VL_ACCEPT_UNTAGGED_MODE_MSK 0x00000004
 /* Inverted bitmask for bitfield vl_accept_untagged_mode */
-#define rpf_vl_accept_untagged_mode_mskn 0xFFFFFFFB
+#define HW_ATL_RPF_VL_ACCEPT_UNTAGGED_MODE_MSKN 0xFFFFFFFB
 /* Lower bit position of bitfield vl_accept_untagged_mode */
-#define rpf_vl_accept_untagged_mode_shift 2
+#define HW_ATL_RPF_VL_ACCEPT_UNTAGGED_MODE_SHIFT 2
 /* Width of bitfield vl_accept_untagged_mode */
-#define rpf_vl_accept_untagged_mode_width 1
+#define HW_ATL_RPF_VL_ACCEPT_UNTAGGED_MODE_WIDTH 1
 /* Default value of bitfield vl_accept_untagged_mode */
-#define rpf_vl_accept_untagged_mode_default 0x0
+#define HW_ATL_RPF_VL_ACCEPT_UNTAGGED_MODE_DEFAULT 0x0
 
 /* rX vl_untagged_act[2:0] Bitfield Definitions
  * Preprocessor definitions for the bitfield "vl_untagged_act[2:0]".
@@ -999,17 +1130,17 @@
  */
 
 /* Register address for bitfield vl_untagged_act[2:0] */
-#define rpf_vl_untagged_act_adr 0x00005280
+#define HW_ATL_RPF_VL_UNTAGGED_ACT_ADR 0x00005280
 /* Bitmask for bitfield vl_untagged_act[2:0] */
-#define rpf_vl_untagged_act_msk 0x00000038
+#define HW_ATL_RPF_VL_UNTAGGED_ACT_MSK 0x00000038
 /* Inverted bitmask for bitfield vl_untagged_act[2:0] */
-#define rpf_vl_untagged_act_mskn 0xFFFFFFC7
+#define HW_ATL_RPF_VL_UNTAGGED_ACT_MSKN 0xFFFFFFC7
 /* Lower bit position of bitfield vl_untagged_act[2:0] */
-#define rpf_vl_untagged_act_shift 3
+#define HW_ATL_RPF_VL_UNTAGGED_ACT_SHIFT 3
 /* Width of bitfield vl_untagged_act[2:0] */
-#define rpf_vl_untagged_act_width 3
+#define HW_ATL_RPF_VL_UNTAGGED_ACT_WIDTH 3
 /* Default value of bitfield vl_untagged_act[2:0] */
-#define rpf_vl_untagged_act_default 0x0
+#define HW_ATL_RPF_VL_UNTAGGED_ACT_DEFAULT 0x0
 
 /* RX vl_en{F} Bitfield Definitions
  * Preprocessor definitions for the bitfield "vl_en{F}".
@@ -1018,17 +1149,17 @@
  */
 
 /* Register address for bitfield vl_en{F} */
-#define rpf_vl_en_f_adr(filter) (0x00005290 + (filter) * 0x4)
+#define HW_ATL_RPF_VL_EN_F_ADR(filter) (0x00005290 + (filter) * 0x4)
 /* Bitmask for bitfield vl_en{F} */
-#define rpf_vl_en_f_msk 0x80000000
+#define HW_ATL_RPF_VL_EN_F_MSK 0x80000000
 /* Inverted bitmask for bitfield vl_en{F} */
-#define rpf_vl_en_f_mskn 0x7FFFFFFF
+#define HW_ATL_RPF_VL_EN_F_MSKN 0x7FFFFFFF
 /* Lower bit position of bitfield vl_en{F} */
-#define rpf_vl_en_f_shift 31
+#define HW_ATL_RPF_VL_EN_F_SHIFT 31
 /* Width of bitfield vl_en{F} */
-#define rpf_vl_en_f_width 1
+#define HW_ATL_RPF_VL_EN_F_WIDTH 1
 /* Default value of bitfield vl_en{F} */
-#define rpf_vl_en_f_default 0x0
+#define HW_ATL_RPF_VL_EN_F_DEFAULT 0x0
 
 /* RX vl_act{F}[2:0] Bitfield Definitions
  * Preprocessor definitions for the bitfield "vl_act{F}[2:0]".
@@ -1037,17 +1168,17 @@
  */
 
 /* Register address for bitfield vl_act{F}[2:0] */
-#define rpf_vl_act_f_adr(filter) (0x00005290 + (filter) * 0x4)
+#define HW_ATL_RPF_VL_ACT_F_ADR(filter) (0x00005290 + (filter) * 0x4)
 /* Bitmask for bitfield vl_act{F}[2:0] */
-#define rpf_vl_act_f_msk 0x00070000
+#define HW_ATL_RPF_VL_ACT_F_MSK 0x00070000
 /* Inverted bitmask for bitfield vl_act{F}[2:0] */
-#define rpf_vl_act_f_mskn 0xFFF8FFFF
+#define HW_ATL_RPF_VL_ACT_F_MSKN 0xFFF8FFFF
 /* Lower bit position of bitfield vl_act{F}[2:0] */
-#define rpf_vl_act_f_shift 16
+#define HW_ATL_RPF_VL_ACT_F_SHIFT 16
 /* Width of bitfield vl_act{F}[2:0] */
-#define rpf_vl_act_f_width 3
+#define HW_ATL_RPF_VL_ACT_F_WIDTH 3
 /* Default value of bitfield vl_act{F}[2:0] */
-#define rpf_vl_act_f_default 0x0
+#define HW_ATL_RPF_VL_ACT_F_DEFAULT 0x0
 
 /* RX vl_id{F}[B:0] Bitfield Definitions
  * Preprocessor definitions for the bitfield "vl_id{F}[B:0]".
@@ -1056,36 +1187,55 @@
  */
 
 /* Register address for bitfield vl_id{F}[B:0] */
-#define rpf_vl_id_f_adr(filter) (0x00005290 + (filter) * 0x4)
+#define HW_ATL_RPF_VL_ID_F_ADR(filter) (0x00005290 + (filter) * 0x4)
 /* Bitmask for bitfield vl_id{F}[B:0] */
-#define rpf_vl_id_f_msk 0x00000FFF
+#define HW_ATL_RPF_VL_ID_F_MSK 0x00000FFF
 /* Inverted bitmask for bitfield vl_id{F}[B:0] */
-#define rpf_vl_id_f_mskn 0xFFFFF000
+#define HW_ATL_RPF_VL_ID_F_MSKN 0xFFFFF000
 /* Lower bit position of bitfield vl_id{F}[B:0] */
-#define rpf_vl_id_f_shift 0
+#define HW_ATL_RPF_VL_ID_F_SHIFT 0
 /* Width of bitfield vl_id{F}[B:0] */
-#define rpf_vl_id_f_width 12
+#define HW_ATL_RPF_VL_ID_F_WIDTH 12
 /* Default value of bitfield vl_id{F}[B:0] */
-#define rpf_vl_id_f_default 0x0
+#define HW_ATL_RPF_VL_ID_F_DEFAULT 0x0
 
-/* RX et_en{F} Bitfield Definitions
- * Preprocessor definitions for the bitfield "et_en{F}".
+/* RX vl_rxq_en{F} Bitfield Definitions
+ * Preprocessor definitions for the bitfield "vl_rxq{F}".
+ * Parameter: filter {F} | stride size 0x4 | range [0, 15]
+ * PORT="pif_rpf_vl_rxq_en_i"
+ */
+
+/* Register address for bitfield vl_rxq_en{F} */
+#define HW_ATL_RPF_VL_RXQ_EN_F_ADR(filter) (0x00005290 + (filter) * 0x4)
+/* Bitmask for bitfield vl_rxq_en{F} */
+#define HW_ATL_RPF_VL_RXQ_EN_F_MSK 0x10000000
+/* Inverted bitmask for bitfield vl_rxq_en{F}[ */
+#define HW_ATL_RPF_VL_RXQ_EN_F_MSKN 0xEFFFFFFF
+/* Lower bit position of bitfield vl_rxq_en{F} */
+#define HW_ATL_RPF_VL_RXQ_EN_F_SHIFT 28
+/* Width of bitfield vl_rxq_en{F} */
+#define HW_ATL_RPF_VL_RXQ_EN_F_WIDTH 1
+/* Default value of bitfield vl_rxq_en{F} */
+#define HW_ATL_RPF_VL_RXQ_EN_F_DEFAULT 0x0
+
+/* RX vl_rxq{F}[4:0] Bitfield Definitions
+ * Preprocessor definitions for the bitfield "vl_rxq{F}[4:0]".
  * Parameter: filter {F} | stride size 0x4 | range [0, 15]
- * PORT="pif_rpf_et_en_i[0]"
- */
-
-/* Register address for bitfield et_en{F} */
-#define rpf_et_en_f_adr(filter) (0x00005300 + (filter) * 0x4)
-/* Bitmask for bitfield et_en{F} */
-#define rpf_et_en_f_msk 0x80000000
-/* Inverted bitmask for bitfield et_en{F} */
-#define rpf_et_en_f_mskn 0x7FFFFFFF
-/* Lower bit position of bitfield et_en{F} */
-#define rpf_et_en_f_shift 31
-/* Width of bitfield et_en{F} */
-#define rpf_et_en_f_width 1
-/* Default value of bitfield et_en{F} */
-#define rpf_et_en_f_default 0x0
+ * PORT="pif_rpf_vl_rxq0_i[4:0]"
+ */
+
+/* Register address for bitfield vl_rxq{F}[4:0] */
+#define HW_ATL_RPF_VL_RXQ_F_ADR(filter) (0x00005290 + (filter) * 0x4)
+/* Bitmask for bitfield vl_rxq{F}[4:0] */
+#define HW_ATL_RPF_VL_RXQ_F_MSK 0x01F00000
+/* Inverted bitmask for bitfield vl_rxq{F}[4:0] */
+#define HW_ATL_RPF_VL_RXQ_F_MSKN 0xFE0FFFFF
+/* Lower bit position of bitfield vl_rxq{F}[4:0] */
+#define HW_ATL_RPF_VL_RXQ_F_SHIFT 20
+/* Width of bitfield vl_rxw{F}[4:0] */
+#define HW_ATL_RPF_VL_RXQ_F_WIDTH 5
+/* Default value of bitfield vl_rxq{F}[4:0] */
+#define HW_ATL_RPF_VL_RXQ_F_DEFAULT 0x0
 
 /* rx et_en{f} bitfield definitions
  * preprocessor definitions for the bitfield "et_en{f}".
@@ -1094,17 +1244,17 @@
  */
 
 /* register address for bitfield et_en{f} */
-#define rpf_et_enf_adr(filter) (0x00005300 + (filter) * 0x4)
+#define HW_ATL_RPF_ET_ENF_ADR(filter) (0x00005300 + (filter) * 0x4)
 /* bitmask for bitfield et_en{f} */
-#define rpf_et_enf_msk 0x80000000
+#define HW_ATL_RPF_ET_ENF_MSK 0x80000000
 /* inverted bitmask for bitfield et_en{f} */
-#define rpf_et_enf_mskn 0x7fffffff
+#define HW_ATL_RPF_ET_ENF_MSKN 0x7fffffff
 /* lower bit position of bitfield et_en{f} */
-#define rpf_et_enf_shift 31
+#define HW_ATL_RPF_ET_ENF_SHIFT 31
 /* width of bitfield et_en{f} */
-#define rpf_et_enf_width 1
+#define HW_ATL_RPF_ET_ENF_WIDTH 1
 /* default value of bitfield et_en{f} */
-#define rpf_et_enf_default 0x0
+#define HW_ATL_RPF_ET_ENF_DEFAULT 0x0
 
 /* rx et_up{f}_en bitfield definitions
  * preprocessor definitions for the bitfield "et_up{f}_en".
@@ -1113,17 +1263,17 @@
  */
 
 /* register address for bitfield et_up{f}_en */
-#define rpf_et_upfen_adr(filter) (0x00005300 + (filter) * 0x4)
+#define HW_ATL_RPF_ET_UPFEN_ADR(filter) (0x00005300 + (filter) * 0x4)
 /* bitmask for bitfield et_up{f}_en */
-#define rpf_et_upfen_msk 0x40000000
+#define HW_ATL_RPF_ET_UPFEN_MSK 0x40000000
 /* inverted bitmask for bitfield et_up{f}_en */
-#define rpf_et_upfen_mskn 0xbfffffff
+#define HW_ATL_RPF_ET_UPFEN_MSKN 0xbfffffff
 /* lower bit position of bitfield et_up{f}_en */
-#define rpf_et_upfen_shift 30
+#define HW_ATL_RPF_ET_UPFEN_SHIFT 30
 /* width of bitfield et_up{f}_en */
-#define rpf_et_upfen_width 1
+#define HW_ATL_RPF_ET_UPFEN_WIDTH 1
 /* default value of bitfield et_up{f}_en */
-#define rpf_et_upfen_default 0x0
+#define HW_ATL_RPF_ET_UPFEN_DEFAULT 0x0
 
 /* rx et_rxq{f}_en bitfield definitions
  * preprocessor definitions for the bitfield "et_rxq{f}_en".
@@ -1132,17 +1282,17 @@
  */
 
 /* register address for bitfield et_rxq{f}_en */
-#define rpf_et_rxqfen_adr(filter) (0x00005300 + (filter) * 0x4)
+#define HW_ATL_RPF_ET_RXQFEN_ADR(filter) (0x00005300 + (filter) * 0x4)
 /* bitmask for bitfield et_rxq{f}_en */
-#define rpf_et_rxqfen_msk 0x20000000
+#define HW_ATL_RPF_ET_RXQFEN_MSK 0x20000000
 /* inverted bitmask for bitfield et_rxq{f}_en */
-#define rpf_et_rxqfen_mskn 0xdfffffff
+#define HW_ATL_RPF_ET_RXQFEN_MSKN 0xdfffffff
 /* lower bit position of bitfield et_rxq{f}_en */
-#define rpf_et_rxqfen_shift 29
+#define HW_ATL_RPF_ET_RXQFEN_SHIFT 29
 /* width of bitfield et_rxq{f}_en */
-#define rpf_et_rxqfen_width 1
+#define HW_ATL_RPF_ET_RXQFEN_WIDTH 1
 /* default value of bitfield et_rxq{f}_en */
-#define rpf_et_rxqfen_default 0x0
+#define HW_ATL_RPF_ET_RXQFEN_DEFAULT 0x0
 
 /* rx et_up{f}[2:0] bitfield definitions
  * preprocessor definitions for the bitfield "et_up{f}[2:0]".
@@ -1151,17 +1301,17 @@
  */
 
 /* register address for bitfield et_up{f}[2:0] */
-#define rpf_et_upf_adr(filter) (0x00005300 + (filter) * 0x4)
+#define HW_ATL_RPF_ET_UPF_ADR(filter) (0x00005300 + (filter) * 0x4)
 /* bitmask for bitfield et_up{f}[2:0] */
-#define rpf_et_upf_msk 0x1c000000
+#define HW_ATL_RPF_ET_UPF_MSK 0x1c000000
 /* inverted bitmask for bitfield et_up{f}[2:0] */
-#define rpf_et_upf_mskn 0xe3ffffff
+#define HW_ATL_RPF_ET_UPF_MSKN 0xe3ffffff
 /* lower bit position of bitfield et_up{f}[2:0] */
-#define rpf_et_upf_shift 26
+#define HW_ATL_RPF_ET_UPF_SHIFT 26
 /* width of bitfield et_up{f}[2:0] */
-#define rpf_et_upf_width 3
+#define HW_ATL_RPF_ET_UPF_WIDTH 3
 /* default value of bitfield et_up{f}[2:0] */
-#define rpf_et_upf_default 0x0
+#define HW_ATL_RPF_ET_UPF_DEFAULT 0x0
 
 /* rx et_rxq{f}[4:0] bitfield definitions
  * preprocessor definitions for the bitfield "et_rxq{f}[4:0]".
@@ -1170,17 +1320,17 @@
  */
 
 /* register address for bitfield et_rxq{f}[4:0] */
-#define rpf_et_rxqf_adr(filter) (0x00005300 + (filter) * 0x4)
+#define HW_ATL_RPF_ET_RXQF_ADR(filter) (0x00005300 + (filter) * 0x4)
 /* bitmask for bitfield et_rxq{f}[4:0] */
-#define rpf_et_rxqf_msk 0x01f00000
+#define HW_ATL_RPF_ET_RXQF_MSK 0x01f00000
 /* inverted bitmask for bitfield et_rxq{f}[4:0] */
-#define rpf_et_rxqf_mskn 0xfe0fffff
+#define HW_ATL_RPF_ET_RXQF_MSKN 0xfe0fffff
 /* lower bit position of bitfield et_rxq{f}[4:0] */
-#define rpf_et_rxqf_shift 20
+#define HW_ATL_RPF_ET_RXQF_SHIFT 20
 /* width of bitfield et_rxq{f}[4:0] */
-#define rpf_et_rxqf_width 5
+#define HW_ATL_RPF_ET_RXQF_WIDTH 5
 /* default value of bitfield et_rxq{f}[4:0] */
-#define rpf_et_rxqf_default 0x0
+#define HW_ATL_RPF_ET_RXQF_DEFAULT 0x0
 
 /* rx et_mng_rxq{f} bitfield definitions
  * preprocessor definitions for the bitfield "et_mng_rxq{f}".
@@ -1189,17 +1339,17 @@
  */
 
 /* register address for bitfield et_mng_rxq{f} */
-#define rpf_et_mng_rxqf_adr(filter) (0x00005300 + (filter) * 0x4)
+#define HW_ATL_RPF_ET_MNG_RXQF_ADR(filter) (0x00005300 + (filter) * 0x4)
 /* bitmask for bitfield et_mng_rxq{f} */
-#define rpf_et_mng_rxqf_msk 0x00080000
+#define HW_ATL_RPF_ET_MNG_RXQF_MSK 0x00080000
 /* inverted bitmask for bitfield et_mng_rxq{f} */
-#define rpf_et_mng_rxqf_mskn 0xfff7ffff
+#define HW_ATL_RPF_ET_MNG_RXQF_MSKN 0xfff7ffff
 /* lower bit position of bitfield et_mng_rxq{f} */
-#define rpf_et_mng_rxqf_shift 19
+#define HW_ATL_RPF_ET_MNG_RXQF_SHIFT 19
 /* width of bitfield et_mng_rxq{f} */
-#define rpf_et_mng_rxqf_width 1
+#define HW_ATL_RPF_ET_MNG_RXQF_WIDTH 1
 /* default value of bitfield et_mng_rxq{f} */
-#define rpf_et_mng_rxqf_default 0x0
+#define HW_ATL_RPF_ET_MNG_RXQF_DEFAULT 0x0
 
 /* rx et_act{f}[2:0] bitfield definitions
  * preprocessor definitions for the bitfield "et_act{f}[2:0]".
@@ -1208,17 +1358,17 @@
  */
 
 /* register address for bitfield et_act{f}[2:0] */
-#define rpf_et_actf_adr(filter) (0x00005300 + (filter) * 0x4)
+#define HW_ATL_RPF_ET_ACTF_ADR(filter) (0x00005300 + (filter) * 0x4)
 /* bitmask for bitfield et_act{f}[2:0] */
-#define rpf_et_actf_msk 0x00070000
+#define HW_ATL_RPF_ET_ACTF_MSK 0x00070000
 /* inverted bitmask for bitfield et_act{f}[2:0] */
-#define rpf_et_actf_mskn 0xfff8ffff
+#define HW_ATL_RPF_ET_ACTF_MSKN 0xfff8ffff
 /* lower bit position of bitfield et_act{f}[2:0] */
-#define rpf_et_actf_shift 16
+#define HW_ATL_RPF_ET_ACTF_SHIFT 16
 /* width of bitfield et_act{f}[2:0] */
-#define rpf_et_actf_width 3
+#define HW_ATL_RPF_ET_ACTF_WIDTH 3
 /* default value of bitfield et_act{f}[2:0] */
-#define rpf_et_actf_default 0x0
+#define HW_ATL_RPF_ET_ACTF_DEFAULT 0x0
 
 /* rx et_val{f}[f:0] bitfield definitions
  * preprocessor definitions for the bitfield "et_val{f}[f:0]".
@@ -1227,17 +1377,101 @@
  */
 
 /* register address for bitfield et_val{f}[f:0] */
-#define rpf_et_valf_adr(filter) (0x00005300 + (filter) * 0x4)
+#define HW_ATL_RPF_ET_VALF_ADR(filter) (0x00005300 + (filter) * 0x4)
 /* bitmask for bitfield et_val{f}[f:0] */
-#define rpf_et_valf_msk 0x0000ffff
+#define HW_ATL_RPF_ET_VALF_MSK 0x0000ffff
 /* inverted bitmask for bitfield et_val{f}[f:0] */
-#define rpf_et_valf_mskn 0xffff0000
+#define HW_ATL_RPF_ET_VALF_MSKN 0xffff0000
 /* lower bit position of bitfield et_val{f}[f:0] */
-#define rpf_et_valf_shift 0
+#define HW_ATL_RPF_ET_VALF_SHIFT 0
 /* width of bitfield et_val{f}[f:0] */
-#define rpf_et_valf_width 16
+#define HW_ATL_RPF_ET_VALF_WIDTH 16
 /* default value of bitfield et_val{f}[f:0] */
-#define rpf_et_valf_default 0x0
+#define HW_ATL_RPF_ET_VALF_DEFAULT 0x0
+
+/* RX l3_l4_en{F} Bitfield Definitions
+ * Preprocessor definitions for the bitfield "l3_l4_en{F}".
+ * Parameter: filter {F} | stride size 0x4 | range [0, 7]
+ * PORT="pif_rpf_l3_l4_en_i[0]"
+ */
+
+#define HW_ATL_RPF_L3_REG_CTRL_ADR(filter) (0x00005380 + (filter) * 0x4)
+
+/* RX rpf_l3_sa{D}[1F:0] Bitfield Definitions
+ * Preprocessor definitions for the bitfield "l3_sa{D}[1F:0]".
+ * Parameter: location {D} | stride size 0x4 | range [0, 7]
+ * PORT="pif_rpf_l3_sa0_i[31:0]"
+ */
+
+/* Register address for bitfield pif_rpf_l3_sa0_i[31:0] */
+#define HW_ATL_RPF_L3_SRCA_ADR(filter) (0x000053B0 + (filter) * 0x4)
+/* Bitmask for bitfield l3_sa0[1F:0] */
+#define HW_ATL_RPF_L3_SRCA_MSK 0xFFFFFFFFu
+/* Inverted bitmask for bitfield l3_sa0[1F:0] */
+#define HW_ATL_RPF_L3_SRCA_MSKN 0xFFFFFFFFu
+/* Lower bit position of bitfield l3_sa0[1F:0] */
+#define HW_ATL_RPF_L3_SRCA_SHIFT 0
+/* Width of bitfield l3_sa0[1F:0] */
+#define HW_ATL_RPF_L3_SRCA_WIDTH 32
+/* Default value of bitfield l3_sa0[1F:0] */
+#define HW_ATL_RPF_L3_SRCA_DEFAULT 0x0
+
+/* RX rpf_l3_da{D}[1F:0] Bitfield Definitions
+ * Preprocessor definitions for the bitfield "l3_da{D}[1F:0]".
+ * Parameter: location {D} | stride size 0x4 | range [0, 7]
+ * PORT="pif_rpf_l3_da0_i[31:0]"
+ */
+
+ /* Register address for bitfield pif_rpf_l3_da0_i[31:0] */
+#define HW_ATL_RPF_L3_DSTA_ADR(filter) (0x000053D0 + (filter) * 0x4)
+/* Bitmask for bitfield l3_da0[1F:0] */
+#define HW_ATL_RPF_L3_DSTA_MSK 0xFFFFFFFFu
+/* Inverted bitmask for bitfield l3_da0[1F:0] */
+#define HW_ATL_RPF_L3_DSTA_MSKN 0xFFFFFFFFu
+/* Lower bit position of bitfield l3_da0[1F:0] */
+#define HW_ATL_RPF_L3_DSTA_SHIFT 0
+/* Width of bitfield l3_da0[1F:0] */
+#define HW_ATL_RPF_L3_DSTA_WIDTH 32
+/* Default value of bitfield l3_da0[1F:0] */
+#define HW_ATL_RPF_L3_DSTA_DEFAULT 0x0
+
+/* RX l4_sp{D}[F:0] Bitfield Definitions
+ * Preprocessor definitions for the bitfield "l4_sp{D}[F:0]".
+ * Parameter: srcport {D} | stride size 0x4 | range [0, 7]
+ * PORT="pif_rpf_l4_sp0_i[15:0]"
+ */
+
+/* Register address for bitfield l4_sp{D}[F:0] */
+#define HW_ATL_RPF_L4_SPD_ADR(srcport) (0x00005400u + (srcport) * 0x4)
+/* Bitmask for bitfield l4_sp{D}[F:0] */
+#define HW_ATL_RPF_L4_SPD_MSK 0x0000FFFFu
+/* Inverted bitmask for bitfield l4_sp{D}[F:0] */
+#define HW_ATL_RPF_L4_SPD_MSKN 0xFFFF0000u
+/* Lower bit position of bitfield l4_sp{D}[F:0] */
+#define HW_ATL_RPF_L4_SPD_SHIFT 0
+/* Width of bitfield l4_sp{D}[F:0] */
+#define HW_ATL_RPF_L4_SPD_WIDTH 16
+/* Default value of bitfield l4_sp{D}[F:0] */
+#define HW_ATL_RPF_L4_SPD_DEFAULT 0x0
+
+/* RX l4_dp{D}[F:0] Bitfield Definitions
+ * Preprocessor definitions for the bitfield "l4_dp{D}[F:0]".
+ * Parameter: destport {D} | stride size 0x4 | range [0, 7]
+ * PORT="pif_rpf_l4_dp0_i[15:0]"
+ */
+
+/* Register address for bitfield l4_dp{D}[F:0] */
+#define HW_ATL_RPF_L4_DPD_ADR(destport) (0x00005420u + (destport) * 0x4)
+/* Bitmask for bitfield l4_dp{D}[F:0] */
+#define HW_ATL_RPF_L4_DPD_MSK 0x0000FFFFu
+/* Inverted bitmask for bitfield l4_dp{D}[F:0] */
+#define HW_ATL_RPF_L4_DPD_MSKN 0xFFFF0000u
+/* Lower bit position of bitfield l4_dp{D}[F:0] */
+#define HW_ATL_RPF_L4_DPD_SHIFT 0
+/* Width of bitfield l4_dp{D}[F:0] */
+#define HW_ATL_RPF_L4_DPD_WIDTH 16
+/* Default value of bitfield l4_dp{D}[F:0] */
+#define HW_ATL_RPF_L4_DPD_DEFAULT 0x0
 
 /* rx ipv4_chk_en bitfield definitions
  * preprocessor definitions for the bitfield "ipv4_chk_en".
@@ -1245,17 +1479,17 @@
  */
 
 /* register address for bitfield ipv4_chk_en */
-#define rpo_ipv4chk_en_adr 0x00005580
+#define HW_ATL_RPO_IPV4CHK_EN_ADR 0x00005580
 /* bitmask for bitfield ipv4_chk_en */
-#define rpo_ipv4chk_en_msk 0x00000002
+#define HW_ATL_RPO_IPV4CHK_EN_MSK 0x00000002
 /* inverted bitmask for bitfield ipv4_chk_en */
-#define rpo_ipv4chk_en_mskn 0xfffffffd
+#define HW_ATL_RPO_IPV4CHK_EN_MSKN 0xfffffffd
 /* lower bit position of bitfield ipv4_chk_en */
-#define rpo_ipv4chk_en_shift 1
+#define HW_ATL_RPO_IPV4CHK_EN_SHIFT 1
 /* width of bitfield ipv4_chk_en */
-#define rpo_ipv4chk_en_width 1
+#define HW_ATL_RPO_IPV4CHK_EN_WIDTH 1
 /* default value of bitfield ipv4_chk_en */
-#define rpo_ipv4chk_en_default 0x0
+#define HW_ATL_RPO_IPV4CHK_EN_DEFAULT 0x0
 
 /* rx desc{d}_vl_strip bitfield definitions
  * preprocessor definitions for the bitfield "desc{d}_vl_strip".
@@ -1264,17 +1498,18 @@
  */
 
 /* register address for bitfield desc{d}_vl_strip */
-#define rpo_descdvl_strip_adr(descriptor) (0x00005b08 + (descriptor) * 0x20)
+#define HW_ATL_RPO_DESCDVL_STRIP_ADR(descriptor) \
+	(0x00005b08 + (descriptor) * 0x20)
 /* bitmask for bitfield desc{d}_vl_strip */
-#define rpo_descdvl_strip_msk 0x20000000
+#define HW_ATL_RPO_DESCDVL_STRIP_MSK 0x20000000
 /* inverted bitmask for bitfield desc{d}_vl_strip */
-#define rpo_descdvl_strip_mskn 0xdfffffff
+#define HW_ATL_RPO_DESCDVL_STRIP_MSKN 0xdfffffff
 /* lower bit position of bitfield desc{d}_vl_strip */
-#define rpo_descdvl_strip_shift 29
+#define HW_ATL_RPO_DESCDVL_STRIP_SHIFT 29
 /* width of bitfield desc{d}_vl_strip */
-#define rpo_descdvl_strip_width 1
+#define HW_ATL_RPO_DESCDVL_STRIP_WIDTH 1
 /* default value of bitfield desc{d}_vl_strip */
-#define rpo_descdvl_strip_default 0x0
+#define HW_ATL_RPO_DESCDVL_STRIP_DEFAULT 0x0
 
 /* rx l4_chk_en bitfield definitions
  * preprocessor definitions for the bitfield "l4_chk_en".
@@ -1282,17 +1517,35 @@
  */
 
 /* register address for bitfield l4_chk_en */
-#define rpol4chk_en_adr 0x00005580
+#define HW_ATL_RPOL4CHK_EN_ADR 0x00005580
 /* bitmask for bitfield l4_chk_en */
-#define rpol4chk_en_msk 0x00000001
+#define HW_ATL_RPOL4CHK_EN_MSK 0x00000001
 /* inverted bitmask for bitfield l4_chk_en */
-#define rpol4chk_en_mskn 0xfffffffe
+#define HW_ATL_RPOL4CHK_EN_MSKN 0xfffffffe
 /* lower bit position of bitfield l4_chk_en */
-#define rpol4chk_en_shift 0
+#define HW_ATL_RPOL4CHK_EN_SHIFT 0
 /* width of bitfield l4_chk_en */
-#define rpol4chk_en_width 1
+#define HW_ATL_RPOL4CHK_EN_WIDTH 1
 /* default value of bitfield l4_chk_en */
-#define rpol4chk_en_default 0x0
+#define HW_ATL_RPOL4CHK_EN_DEFAULT 0x0
+
+/* RX outer_vl_ins_mode Bitfield Definitions
+ * Preprocessor definitions for the bitfield "outer_vl_ins_mode".
+ * PORT="pif_rpo_outer_vl_mode_i"
+ */
+
+/* Register address for bitfield outer_vl_ins_mode */
+#define HW_ATL_RPO_OUTER_VL_INS_MODE_ADR 0x00005580
+/* Bitmask for bitfield outer_vl_ins_mode */
+#define HW_ATL_RPO_OUTER_VL_INS_MODE_MSK 0x00000004
+/* Inverted bitmask for bitfield outer_vl_ins_mode */
+#define HW_ATL_RPO_OUTER_VL_INS_MODE_MSKN 0xFFFFFFFB
+/* Lower bit position of bitfield outer_vl_ins_mode */
+#define HW_ATL_RPO_OUTER_VL_INS_MODE_SHIFT 2
+/* Width of bitfield outer_vl_ins_mode */
+#define HW_ATL_RPO_OUTER_VL_INS_MODE_WIDTH 1
+/* Default value of bitfield outer_vl_ins_mode */
+#define HW_ATL_RPO_OUTER_VL_INS_MODE_DEFAULT 0x0
 
 /* rx reg_res_dsbl bitfield definitions
  * preprocessor definitions for the bitfield "reg_res_dsbl".
@@ -1300,17 +1553,17 @@
  */
 
 /* register address for bitfield reg_res_dsbl */
-#define rx_reg_res_dsbl_adr 0x00005000
+#define HW_ATL_RX_REG_RES_DSBL_ADR 0x00005000
 /* bitmask for bitfield reg_res_dsbl */
-#define rx_reg_res_dsbl_msk 0x20000000
+#define HW_ATL_RX_REG_RES_DSBL_MSK 0x20000000
 /* inverted bitmask for bitfield reg_res_dsbl */
-#define rx_reg_res_dsbl_mskn 0xdfffffff
+#define HW_ATL_RX_REG_RES_DSBL_MSKN 0xdfffffff
 /* lower bit position of bitfield reg_res_dsbl */
-#define rx_reg_res_dsbl_shift 29
+#define HW_ATL_RX_REG_RES_DSBL_SHIFT 29
 /* width of bitfield reg_res_dsbl */
-#define rx_reg_res_dsbl_width 1
+#define HW_ATL_RX_REG_RES_DSBL_WIDTH 1
 /* default value of bitfield reg_res_dsbl */
-#define rx_reg_res_dsbl_default 0x1
+#define HW_ATL_RX_REG_RES_DSBL_DEFAULT 0x1
 
 /* tx dca{d}_cpuid[7:0] bitfield definitions
  * preprocessor definitions for the bitfield "dca{d}_cpuid[7:0]".
@@ -1319,17 +1572,17 @@
  */
 
 /* register address for bitfield dca{d}_cpuid[7:0] */
-#define tdm_dcadcpuid_adr(dca) (0x00008400 + (dca) * 0x4)
+#define HW_ATL_TDM_DCADCPUID_ADR(dca) (0x00008400 + (dca) * 0x4)
 /* bitmask for bitfield dca{d}_cpuid[7:0] */
-#define tdm_dcadcpuid_msk 0x000000ff
+#define HW_ATL_TDM_DCADCPUID_MSK 0x000000ff
 /* inverted bitmask for bitfield dca{d}_cpuid[7:0] */
-#define tdm_dcadcpuid_mskn 0xffffff00
+#define HW_ATL_TDM_DCADCPUID_MSKN 0xffffff00
 /* lower bit position of bitfield dca{d}_cpuid[7:0] */
-#define tdm_dcadcpuid_shift 0
+#define HW_ATL_TDM_DCADCPUID_SHIFT 0
 /* width of bitfield dca{d}_cpuid[7:0] */
-#define tdm_dcadcpuid_width 8
+#define HW_ATL_TDM_DCADCPUID_WIDTH 8
 /* default value of bitfield dca{d}_cpuid[7:0] */
-#define tdm_dcadcpuid_default 0x0
+#define HW_ATL_TDM_DCADCPUID_DEFAULT 0x0
 
 /* tx lso_en[1f:0] bitfield definitions
  * preprocessor definitions for the bitfield "lso_en[1f:0]".
@@ -1337,17 +1590,17 @@
  */
 
 /* register address for bitfield lso_en[1f:0] */
-#define tdm_lso_en_adr 0x00007810
+#define HW_ATL_TDM_LSO_EN_ADR 0x00007810
 /* bitmask for bitfield lso_en[1f:0] */
-#define tdm_lso_en_msk 0xffffffff
+#define HW_ATL_TDM_LSO_EN_MSK 0xffffffff
 /* inverted bitmask for bitfield lso_en[1f:0] */
-#define tdm_lso_en_mskn 0x00000000
+#define HW_ATL_TDM_LSO_EN_MSKN 0x00000000
 /* lower bit position of bitfield lso_en[1f:0] */
-#define tdm_lso_en_shift 0
+#define HW_ATL_TDM_LSO_EN_SHIFT 0
 /* width of bitfield lso_en[1f:0] */
-#define tdm_lso_en_width 32
+#define HW_ATL_TDM_LSO_EN_WIDTH 32
 /* default value of bitfield lso_en[1f:0] */
-#define tdm_lso_en_default 0x0
+#define HW_ATL_TDM_LSO_EN_DEFAULT 0x0
 
 /* tx dca_en bitfield definitions
  * preprocessor definitions for the bitfield "dca_en".
@@ -1355,17 +1608,17 @@
  */
 
 /* register address for bitfield dca_en */
-#define tdm_dca_en_adr 0x00008480
+#define HW_ATL_TDM_DCA_EN_ADR 0x00008480
 /* bitmask for bitfield dca_en */
-#define tdm_dca_en_msk 0x80000000
+#define HW_ATL_TDM_DCA_EN_MSK 0x80000000
 /* inverted bitmask for bitfield dca_en */
-#define tdm_dca_en_mskn 0x7fffffff
+#define HW_ATL_TDM_DCA_EN_MSKN 0x7fffffff
 /* lower bit position of bitfield dca_en */
-#define tdm_dca_en_shift 31
+#define HW_ATL_TDM_DCA_EN_SHIFT 31
 /* width of bitfield dca_en */
-#define tdm_dca_en_width 1
+#define HW_ATL_TDM_DCA_EN_WIDTH 1
 /* default value of bitfield dca_en */
-#define tdm_dca_en_default 0x1
+#define HW_ATL_TDM_DCA_EN_DEFAULT 0x1
 
 /* tx dca_mode[3:0] bitfield definitions
  * preprocessor definitions for the bitfield "dca_mode[3:0]".
@@ -1373,17 +1626,17 @@
  */
 
 /* register address for bitfield dca_mode[3:0] */
-#define tdm_dca_mode_adr 0x00008480
+#define HW_ATL_TDM_DCA_MODE_ADR 0x00008480
 /* bitmask for bitfield dca_mode[3:0] */
-#define tdm_dca_mode_msk 0x0000000f
+#define HW_ATL_TDM_DCA_MODE_MSK 0x0000000f
 /* inverted bitmask for bitfield dca_mode[3:0] */
-#define tdm_dca_mode_mskn 0xfffffff0
+#define HW_ATL_TDM_DCA_MODE_MSKN 0xfffffff0
 /* lower bit position of bitfield dca_mode[3:0] */
-#define tdm_dca_mode_shift 0
+#define HW_ATL_TDM_DCA_MODE_SHIFT 0
 /* width of bitfield dca_mode[3:0] */
-#define tdm_dca_mode_width 4
+#define HW_ATL_TDM_DCA_MODE_WIDTH 4
 /* default value of bitfield dca_mode[3:0] */
-#define tdm_dca_mode_default 0x0
+#define HW_ATL_TDM_DCA_MODE_DEFAULT 0x0
 
 /* tx dca{d}_desc_en bitfield definitions
  * preprocessor definitions for the bitfield "dca{d}_desc_en".
@@ -1392,17 +1645,17 @@
  */
 
 /* register address for bitfield dca{d}_desc_en */
-#define tdm_dcaddesc_en_adr(dca) (0x00008400 + (dca) * 0x4)
+#define HW_ATL_TDM_DCADDESC_EN_ADR(dca) (0x00008400 + (dca) * 0x4)
 /* bitmask for bitfield dca{d}_desc_en */
-#define tdm_dcaddesc_en_msk 0x80000000
+#define HW_ATL_TDM_DCADDESC_EN_MSK 0x80000000
 /* inverted bitmask for bitfield dca{d}_desc_en */
-#define tdm_dcaddesc_en_mskn 0x7fffffff
+#define HW_ATL_TDM_DCADDESC_EN_MSKN 0x7fffffff
 /* lower bit position of bitfield dca{d}_desc_en */
-#define tdm_dcaddesc_en_shift 31
+#define HW_ATL_TDM_DCADDESC_EN_SHIFT 31
 /* width of bitfield dca{d}_desc_en */
-#define tdm_dcaddesc_en_width 1
+#define HW_ATL_TDM_DCADDESC_EN_WIDTH 1
 /* default value of bitfield dca{d}_desc_en */
-#define tdm_dcaddesc_en_default 0x0
+#define HW_ATL_TDM_DCADDESC_EN_DEFAULT 0x0
 
 /* tx desc{d}_en bitfield definitions
  * preprocessor definitions for the bitfield "desc{d}_en".
@@ -1411,17 +1664,17 @@
  */
 
 /* register address for bitfield desc{d}_en */
-#define tdm_descden_adr(descriptor) (0x00007c08 + (descriptor) * 0x40)
+#define HW_ATL_TDM_DESCDEN_ADR(descriptor) (0x00007c08 + (descriptor) * 0x40)
 /* bitmask for bitfield desc{d}_en */
-#define tdm_descden_msk 0x80000000
+#define HW_ATL_TDM_DESCDEN_MSK 0x80000000
 /* inverted bitmask for bitfield desc{d}_en */
-#define tdm_descden_mskn 0x7fffffff
+#define HW_ATL_TDM_DESCDEN_MSKN 0x7fffffff
 /* lower bit position of bitfield desc{d}_en */
-#define tdm_descden_shift 31
+#define HW_ATL_TDM_DESCDEN_SHIFT 31
 /* width of bitfield desc{d}_en */
-#define tdm_descden_width 1
+#define HW_ATL_TDM_DESCDEN_WIDTH 1
 /* default value of bitfield desc{d}_en */
-#define tdm_descden_default 0x0
+#define HW_ATL_TDM_DESCDEN_DEFAULT 0x0
 
 /* tx desc{d}_hd[c:0] bitfield definitions
  * preprocessor definitions for the bitfield "desc{d}_hd[c:0]".
@@ -1430,15 +1683,15 @@
  */
 
 /* register address for bitfield desc{d}_hd[c:0] */
-#define tdm_descdhd_adr(descriptor) (0x00007c0c + (descriptor) * 0x40)
+#define HW_ATL_TDM_DESCDHD_ADR(descriptor) (0x00007c0c + (descriptor) * 0x40)
 /* bitmask for bitfield desc{d}_hd[c:0] */
-#define tdm_descdhd_msk 0x00001fff
+#define HW_ATL_TDM_DESCDHD_MSK 0x00001fff
 /* inverted bitmask for bitfield desc{d}_hd[c:0] */
-#define tdm_descdhd_mskn 0xffffe000
+#define HW_ATL_TDM_DESCDHD_MSKN 0xffffe000
 /* lower bit position of bitfield desc{d}_hd[c:0] */
-#define tdm_descdhd_shift 0
+#define HW_ATL_TDM_DESCDHD_SHIFT 0
 /* width of bitfield desc{d}_hd[c:0] */
-#define tdm_descdhd_width 13
+#define HW_ATL_TDM_DESCDHD_WIDTH 13
 
 /* tx desc{d}_len[9:0] bitfield definitions
  * preprocessor definitions for the bitfield "desc{d}_len[9:0]".
@@ -1447,17 +1700,17 @@
  */
 
 /* register address for bitfield desc{d}_len[9:0] */
-#define tdm_descdlen_adr(descriptor) (0x00007c08 + (descriptor) * 0x40)
+#define HW_ATL_TDM_DESCDLEN_ADR(descriptor) (0x00007c08 + (descriptor) * 0x40)
 /* bitmask for bitfield desc{d}_len[9:0] */
-#define tdm_descdlen_msk 0x00001ff8
+#define HW_ATL_TDM_DESCDLEN_MSK 0x00001ff8
 /* inverted bitmask for bitfield desc{d}_len[9:0] */
-#define tdm_descdlen_mskn 0xffffe007
+#define HW_ATL_TDM_DESCDLEN_MSKN 0xffffe007
 /* lower bit position of bitfield desc{d}_len[9:0] */
-#define tdm_descdlen_shift 3
+#define HW_ATL_TDM_DESCDLEN_SHIFT 3
 /* width of bitfield desc{d}_len[9:0] */
-#define tdm_descdlen_width 10
+#define HW_ATL_TDM_DESCDLEN_WIDTH 10
 /* default value of bitfield desc{d}_len[9:0] */
-#define tdm_descdlen_default 0x0
+#define HW_ATL_TDM_DESCDLEN_DEFAULT 0x0
 
 /* tx int_desc_wrb_en bitfield definitions
  * preprocessor definitions for the bitfield "int_desc_wrb_en".
@@ -1465,17 +1718,17 @@
  */
 
 /* register address for bitfield int_desc_wrb_en */
-#define tdm_int_desc_wrb_en_adr 0x00007b40
+#define HW_ATL_TDM_INT_DESC_WRB_EN_ADR 0x00007b40
 /* bitmask for bitfield int_desc_wrb_en */
-#define tdm_int_desc_wrb_en_msk 0x00000002
+#define HW_ATL_TDM_INT_DESC_WRB_EN_MSK 0x00000002
 /* inverted bitmask for bitfield int_desc_wrb_en */
-#define tdm_int_desc_wrb_en_mskn 0xfffffffd
+#define HW_ATL_TDM_INT_DESC_WRB_EN_MSKN 0xfffffffd
 /* lower bit position of bitfield int_desc_wrb_en */
-#define tdm_int_desc_wrb_en_shift 1
+#define HW_ATL_TDM_INT_DESC_WRB_EN_SHIFT 1
 /* width of bitfield int_desc_wrb_en */
-#define tdm_int_desc_wrb_en_width 1
+#define HW_ATL_TDM_INT_DESC_WRB_EN_WIDTH 1
 /* default value of bitfield int_desc_wrb_en */
-#define tdm_int_desc_wrb_en_default 0x0
+#define HW_ATL_TDM_INT_DESC_WRB_EN_DEFAULT 0x0
 
 /* tx desc{d}_wrb_thresh[6:0] bitfield definitions
  * preprocessor definitions for the bitfield "desc{d}_wrb_thresh[6:0]".
@@ -1484,17 +1737,18 @@
  */
 
 /* register address for bitfield desc{d}_wrb_thresh[6:0] */
-#define tdm_descdwrb_thresh_adr(descriptor) (0x00007c18 + (descriptor) * 0x40)
+#define HW_ATL_TDM_DESCDWRB_THRESH_ADR(descriptor) \
+	(0x00007c18 + (descriptor) * 0x40)
 /* bitmask for bitfield desc{d}_wrb_thresh[6:0] */
-#define tdm_descdwrb_thresh_msk 0x00007f00
+#define HW_ATL_TDM_DESCDWRB_THRESH_MSK 0x00007f00
 /* inverted bitmask for bitfield desc{d}_wrb_thresh[6:0] */
-#define tdm_descdwrb_thresh_mskn 0xffff80ff
+#define HW_ATL_TDM_DESCDWRB_THRESH_MSKN 0xffff80ff
 /* lower bit position of bitfield desc{d}_wrb_thresh[6:0] */
-#define tdm_descdwrb_thresh_shift 8
+#define HW_ATL_TDM_DESCDWRB_THRESH_SHIFT 8
 /* width of bitfield desc{d}_wrb_thresh[6:0] */
-#define tdm_descdwrb_thresh_width 7
+#define HW_ATL_TDM_DESCDWRB_THRESH_WIDTH 7
 /* default value of bitfield desc{d}_wrb_thresh[6:0] */
-#define tdm_descdwrb_thresh_default 0x0
+#define HW_ATL_TDM_DESCDWRB_THRESH_DEFAULT 0x0
 
 /* tx lso_tcp_flag_first[b:0] bitfield definitions
  * preprocessor definitions for the bitfield "lso_tcp_flag_first[b:0]".
@@ -1502,17 +1756,17 @@
  */
 
 /* register address for bitfield lso_tcp_flag_first[b:0] */
-#define thm_lso_tcp_flag_first_adr 0x00007820
+#define HW_ATL_THM_LSO_TCP_FLAG_FIRST_ADR 0x00007820
 /* bitmask for bitfield lso_tcp_flag_first[b:0] */
-#define thm_lso_tcp_flag_first_msk 0x00000fff
+#define HW_ATL_THM_LSO_TCP_FLAG_FIRST_MSK 0x00000fff
 /* inverted bitmask for bitfield lso_tcp_flag_first[b:0] */
-#define thm_lso_tcp_flag_first_mskn 0xfffff000
+#define HW_ATL_THM_LSO_TCP_FLAG_FIRST_MSKN 0xfffff000
 /* lower bit position of bitfield lso_tcp_flag_first[b:0] */
-#define thm_lso_tcp_flag_first_shift 0
+#define HW_ATL_THM_LSO_TCP_FLAG_FIRST_SHIFT 0
 /* width of bitfield lso_tcp_flag_first[b:0] */
-#define thm_lso_tcp_flag_first_width 12
+#define HW_ATL_THM_LSO_TCP_FLAG_FIRST_WIDTH 12
 /* default value of bitfield lso_tcp_flag_first[b:0] */
-#define thm_lso_tcp_flag_first_default 0x0
+#define HW_ATL_THM_LSO_TCP_FLAG_FIRST_DEFAULT 0x0
 
 /* tx lso_tcp_flag_last[b:0] bitfield definitions
  * preprocessor definitions for the bitfield "lso_tcp_flag_last[b:0]".
@@ -1520,17 +1774,17 @@
  */
 
 /* register address for bitfield lso_tcp_flag_last[b:0] */
-#define thm_lso_tcp_flag_last_adr 0x00007824
+#define HW_ATL_THM_LSO_TCP_FLAG_LAST_ADR 0x00007824
 /* bitmask for bitfield lso_tcp_flag_last[b:0] */
-#define thm_lso_tcp_flag_last_msk 0x00000fff
+#define HW_ATL_THM_LSO_TCP_FLAG_LAST_MSK 0x00000fff
 /* inverted bitmask for bitfield lso_tcp_flag_last[b:0] */
-#define thm_lso_tcp_flag_last_mskn 0xfffff000
+#define HW_ATL_THM_LSO_TCP_FLAG_LAST_MSKN 0xfffff000
 /* lower bit position of bitfield lso_tcp_flag_last[b:0] */
-#define thm_lso_tcp_flag_last_shift 0
+#define HW_ATL_THM_LSO_TCP_FLAG_LAST_SHIFT 0
 /* width of bitfield lso_tcp_flag_last[b:0] */
-#define thm_lso_tcp_flag_last_width 12
+#define HW_ATL_THM_LSO_TCP_FLAG_LAST_WIDTH 12
 /* default value of bitfield lso_tcp_flag_last[b:0] */
-#define thm_lso_tcp_flag_last_default 0x0
+#define HW_ATL_THM_LSO_TCP_FLAG_LAST_DEFAULT 0x0
 
 /* tx lso_tcp_flag_mid[b:0] bitfield definitions
  * preprocessor definitions for the bitfield "lso_tcp_flag_mid[b:0]".
@@ -1538,17 +1792,17 @@
  */
 
 /* Register address for bitfield lro_rsc_max[1F:0] */
-#define rpo_lro_rsc_max_adr 0x00005598
+#define HW_ATL_RPO_LRO_RSC_MAX_ADR 0x00005598
 /* Bitmask for bitfield lro_rsc_max[1F:0] */
-#define rpo_lro_rsc_max_msk 0xFFFFFFFF
+#define HW_ATL_RPO_LRO_RSC_MAX_MSK 0xFFFFFFFF
 /* Inverted bitmask for bitfield lro_rsc_max[1F:0] */
-#define rpo_lro_rsc_max_mskn 0x00000000
+#define HW_ATL_RPO_LRO_RSC_MAX_MSKN 0x00000000
 /* Lower bit position of bitfield lro_rsc_max[1F:0] */
-#define rpo_lro_rsc_max_shift 0
+#define HW_ATL_RPO_LRO_RSC_MAX_SHIFT 0
 /* Width of bitfield lro_rsc_max[1F:0] */
-#define rpo_lro_rsc_max_width 32
+#define HW_ATL_RPO_LRO_RSC_MAX_WIDTH 32
 /* Default value of bitfield lro_rsc_max[1F:0] */
-#define rpo_lro_rsc_max_default 0x0
+#define HW_ATL_RPO_LRO_RSC_MAX_DEFAULT 0x0
 
 /* RX lro_en[1F:0] Bitfield Definitions
  * Preprocessor definitions for the bitfield "lro_en[1F:0]".
@@ -1556,17 +1810,17 @@
  */
 
 /* Register address for bitfield lro_en[1F:0] */
-#define rpo_lro_en_adr 0x00005590
+#define HW_ATL_RPO_LRO_EN_ADR 0x00005590
 /* Bitmask for bitfield lro_en[1F:0] */
-#define rpo_lro_en_msk 0xFFFFFFFF
+#define HW_ATL_RPO_LRO_EN_MSK 0xFFFFFFFF
 /* Inverted bitmask for bitfield lro_en[1F:0] */
-#define rpo_lro_en_mskn 0x00000000
+#define HW_ATL_RPO_LRO_EN_MSKN 0x00000000
 /* Lower bit position of bitfield lro_en[1F:0] */
-#define rpo_lro_en_shift 0
+#define HW_ATL_RPO_LRO_EN_SHIFT 0
 /* Width of bitfield lro_en[1F:0] */
-#define rpo_lro_en_width 32
+#define HW_ATL_RPO_LRO_EN_WIDTH 32
 /* Default value of bitfield lro_en[1F:0] */
-#define rpo_lro_en_default 0x0
+#define HW_ATL_RPO_LRO_EN_DEFAULT 0x0
 
 /* RX lro_ptopt_en Bitfield Definitions
  * Preprocessor definitions for the bitfield "lro_ptopt_en".
@@ -1574,17 +1828,17 @@
  */
 
 /* Register address for bitfield lro_ptopt_en */
-#define rpo_lro_ptopt_en_adr 0x00005594
+#define HW_ATL_RPO_LRO_PTOPT_EN_ADR 0x00005594
 /* Bitmask for bitfield lro_ptopt_en */
-#define rpo_lro_ptopt_en_msk 0x00008000
+#define HW_ATL_RPO_LRO_PTOPT_EN_MSK 0x00008000
 /* Inverted bitmask for bitfield lro_ptopt_en */
-#define rpo_lro_ptopt_en_mskn 0xFFFF7FFF
+#define HW_ATL_RPO_LRO_PTOPT_EN_MSKN 0xFFFF7FFF
 /* Lower bit position of bitfield lro_ptopt_en */
-#define rpo_lro_ptopt_en_shift 15
+#define HW_ATL_RPO_LRO_PTOPT_EN_SHIFT 15
 /* Width of bitfield lro_ptopt_en */
-#define rpo_lro_ptopt_en_width 1
+#define HW_ATL_RPO_LRO_PTOPT_EN_WIDTH 1
 /* Default value of bitfield lro_ptopt_en */
-#define rpo_lro_ptopt_en_defalt 0x1
+#define HW_ATL_RPO_LRO_PTOPT_EN_DEFALT 0x1
 
 /* RX lro_q_ses_lmt Bitfield Definitions
  * Preprocessor definitions for the bitfield "lro_q_ses_lmt".
@@ -1592,17 +1846,17 @@
  */
 
 /* Register address for bitfield lro_q_ses_lmt */
-#define rpo_lro_qses_lmt_adr 0x00005594
+#define HW_ATL_RPO_LRO_QSES_LMT_ADR 0x00005594
 /* Bitmask for bitfield lro_q_ses_lmt */
-#define rpo_lro_qses_lmt_msk 0x00003000
+#define HW_ATL_RPO_LRO_QSES_LMT_MSK 0x00003000
 /* Inverted bitmask for bitfield lro_q_ses_lmt */
-#define rpo_lro_qses_lmt_mskn 0xFFFFCFFF
+#define HW_ATL_RPO_LRO_QSES_LMT_MSKN 0xFFFFCFFF
 /* Lower bit position of bitfield lro_q_ses_lmt */
-#define rpo_lro_qses_lmt_shift 12
+#define HW_ATL_RPO_LRO_QSES_LMT_SHIFT 12
 /* Width of bitfield lro_q_ses_lmt */
-#define rpo_lro_qses_lmt_width 2
+#define HW_ATL_RPO_LRO_QSES_LMT_WIDTH 2
 /* Default value of bitfield lro_q_ses_lmt */
-#define rpo_lro_qses_lmt_default 0x1
+#define HW_ATL_RPO_LRO_QSES_LMT_DEFAULT 0x1
 
 /* RX lro_tot_dsc_lmt[1:0] Bitfield Definitions
  * Preprocessor definitions for the bitfield "lro_tot_dsc_lmt[1:0]".
@@ -1610,17 +1864,17 @@
  */
 
 /* Register address for bitfield lro_tot_dsc_lmt[1:0] */
-#define rpo_lro_tot_dsc_lmt_adr 0x00005594
+#define HW_ATL_RPO_LRO_TOT_DSC_LMT_ADR 0x00005594
 /* Bitmask for bitfield lro_tot_dsc_lmt[1:0] */
-#define rpo_lro_tot_dsc_lmt_msk 0x00000060
+#define HW_ATL_RPO_LRO_TOT_DSC_LMT_MSK 0x00000060
 /* Inverted bitmask for bitfield lro_tot_dsc_lmt[1:0] */
-#define rpo_lro_tot_dsc_lmt_mskn 0xFFFFFF9F
+#define HW_ATL_RPO_LRO_TOT_DSC_LMT_MSKN 0xFFFFFF9F
 /* Lower bit position of bitfield lro_tot_dsc_lmt[1:0] */
-#define rpo_lro_tot_dsc_lmt_shift 5
+#define HW_ATL_RPO_LRO_TOT_DSC_LMT_SHIFT 5
 /* Width of bitfield lro_tot_dsc_lmt[1:0] */
-#define rpo_lro_tot_dsc_lmt_width 2
+#define HW_ATL_RPO_LRO_TOT_DSC_LMT_WIDTH 2
 /* Default value of bitfield lro_tot_dsc_lmt[1:0] */
-#define rpo_lro_tot_dsc_lmt_defalt 0x1
+#define HW_ATL_RPO_LRO_TOT_DSC_LMT_DEFALT 0x1
 
 /* RX lro_pkt_min[4:0] Bitfield Definitions
  * Preprocessor definitions for the bitfield "lro_pkt_min[4:0]".
@@ -1628,22 +1882,22 @@
  */
 
 /* Register address for bitfield lro_pkt_min[4:0] */
-#define rpo_lro_pkt_min_adr 0x00005594
+#define HW_ATL_RPO_LRO_PKT_MIN_ADR 0x00005594
 /* Bitmask for bitfield lro_pkt_min[4:0] */
-#define rpo_lro_pkt_min_msk 0x0000001F
+#define HW_ATL_RPO_LRO_PKT_MIN_MSK 0x0000001F
 /* Inverted bitmask for bitfield lro_pkt_min[4:0] */
-#define rpo_lro_pkt_min_mskn 0xFFFFFFE0
+#define HW_ATL_RPO_LRO_PKT_MIN_MSKN 0xFFFFFFE0
 /* Lower bit position of bitfield lro_pkt_min[4:0] */
-#define rpo_lro_pkt_min_shift 0
+#define HW_ATL_RPO_LRO_PKT_MIN_SHIFT 0
 /* Width of bitfield lro_pkt_min[4:0] */
-#define rpo_lro_pkt_min_width 5
+#define HW_ATL_RPO_LRO_PKT_MIN_WIDTH 5
 /* Default value of bitfield lro_pkt_min[4:0] */
-#define rpo_lro_pkt_min_default 0x8
+#define HW_ATL_RPO_LRO_PKT_MIN_DEFAULT 0x8
 
 /* Width of bitfield lro{L}_des_max[1:0] */
-#define rpo_lro_ldes_max_width 2
+#define HW_ATL_RPO_LRO_LDES_MAX_WIDTH 2
 /* Default value of bitfield lro{L}_des_max[1:0] */
-#define rpo_lro_ldes_max_default 0x0
+#define HW_ATL_RPO_LRO_LDES_MAX_DEFAULT 0x0
 
 /* RX lro_tb_div[11:0] Bitfield Definitions
  * Preprocessor definitions for the bitfield "lro_tb_div[11:0]".
@@ -1651,17 +1905,17 @@
  */
 
 /* Register address for bitfield lro_tb_div[11:0] */
-#define rpo_lro_tb_div_adr 0x00005620
+#define HW_ATL_RPO_LRO_TB_DIV_ADR 0x00005620
 /* Bitmask for bitfield lro_tb_div[11:0] */
-#define rpo_lro_tb_div_msk 0xFFF00000
+#define HW_ATL_RPO_LRO_TB_DIV_MSK 0xFFF00000
 /* Inverted bitmask for bitfield lro_tb_div[11:0] */
-#define rpo_lro_tb_div_mskn 0x000FFFFF
+#define HW_ATL_RPO_LRO_TB_DIV_MSKN 0x000FFFFF
 /* Lower bit position of bitfield lro_tb_div[11:0] */
-#define rpo_lro_tb_div_shift 20
+#define HW_ATL_RPO_LRO_TB_DIV_SHIFT 20
 /* Width of bitfield lro_tb_div[11:0] */
-#define rpo_lro_tb_div_width 12
+#define HW_ATL_RPO_LRO_TB_DIV_WIDTH 12
 /* Default value of bitfield lro_tb_div[11:0] */
-#define rpo_lro_tb_div_default 0xC35
+#define HW_ATL_RPO_LRO_TB_DIV_DEFAULT 0xC35
 
 /* RX lro_ina_ival[9:0] Bitfield Definitions
  *   Preprocessor definitions for the bitfield "lro_ina_ival[9:0]".
@@ -1669,17 +1923,17 @@
  */
 
 /* Register address for bitfield lro_ina_ival[9:0] */
-#define rpo_lro_ina_ival_adr 0x00005620
+#define HW_ATL_RPO_LRO_INA_IVAL_ADR 0x00005620
 /* Bitmask for bitfield lro_ina_ival[9:0] */
-#define rpo_lro_ina_ival_msk 0x000FFC00
+#define HW_ATL_RPO_LRO_INA_IVAL_MSK 0x000FFC00
 /* Inverted bitmask for bitfield lro_ina_ival[9:0] */
-#define rpo_lro_ina_ival_mskn 0xFFF003FF
+#define HW_ATL_RPO_LRO_INA_IVAL_MSKN 0xFFF003FF
 /* Lower bit position of bitfield lro_ina_ival[9:0] */
-#define rpo_lro_ina_ival_shift 10
+#define HW_ATL_RPO_LRO_INA_IVAL_SHIFT 10
 /* Width of bitfield lro_ina_ival[9:0] */
-#define rpo_lro_ina_ival_width 10
+#define HW_ATL_RPO_LRO_INA_IVAL_WIDTH 10
 /* Default value of bitfield lro_ina_ival[9:0] */
-#define rpo_lro_ina_ival_default 0xA
+#define HW_ATL_RPO_LRO_INA_IVAL_DEFAULT 0xA
 
 /* RX lro_max_ival[9:0] Bitfield Definitions
  * Preprocessor definitions for the bitfield "lro_max_ival[9:0]".
@@ -1687,17 +1941,17 @@
  */
 
 /* Register address for bitfield lro_max_ival[9:0] */
-#define rpo_lro_max_ival_adr 0x00005620
+#define HW_ATL_RPO_LRO_MAX_IVAL_ADR 0x00005620
 /* Bitmask for bitfield lro_max_ival[9:0] */
-#define rpo_lro_max_ival_msk 0x000003FF
+#define HW_ATL_RPO_LRO_MAX_IVAL_MSK 0x000003FF
 /* Inverted bitmask for bitfield lro_max_ival[9:0] */
-#define rpo_lro_max_ival_mskn 0xFFFFFC00
+#define HW_ATL_RPO_LRO_MAX_IVAL_MSKN 0xFFFFFC00
 /* Lower bit position of bitfield lro_max_ival[9:0] */
-#define rpo_lro_max_ival_shift 0
+#define HW_ATL_RPO_LRO_MAX_IVAL_SHIFT 0
 /* Width of bitfield lro_max_ival[9:0] */
-#define rpo_lro_max_ival_width 10
+#define HW_ATL_RPO_LRO_MAX_IVAL_WIDTH 10
 /* Default value of bitfield lro_max_ival[9:0] */
-#define rpo_lro_max_ival_default 0x19
+#define HW_ATL_RPO_LRO_MAX_IVAL_DEFAULT 0x19
 
 /* TX dca{D}_cpuid[7:0] Bitfield Definitions
  * Preprocessor definitions for the bitfield "dca{D}_cpuid[7:0]".
@@ -1706,17 +1960,17 @@
  */
 
 /* Register address for bitfield dca{D}_cpuid[7:0] */
-#define tdm_dca_dcpuid_adr(dca) (0x00008400 + (dca) * 0x4)
+#define HW_ATL_TDM_DCA_DCPUID_ADR(dca) (0x00008400 + (dca) * 0x4)
 /* Bitmask for bitfield dca{D}_cpuid[7:0] */
-#define tdm_dca_dcpuid_msk 0x000000FF
+#define HW_ATL_TDM_DCA_DCPUID_MSK 0x000000FF
 /* Inverted bitmask for bitfield dca{D}_cpuid[7:0] */
-#define tdm_dca_dcpuid_mskn 0xFFFFFF00
+#define HW_ATL_TDM_DCA_DCPUID_MSKN 0xFFFFFF00
 /* Lower bit position of bitfield dca{D}_cpuid[7:0] */
-#define tdm_dca_dcpuid_shift 0
+#define HW_ATL_TDM_DCA_DCPUID_SHIFT 0
 /* Width of bitfield dca{D}_cpuid[7:0] */
-#define tdm_dca_dcpuid_width 8
+#define HW_ATL_TDM_DCA_DCPUID_WIDTH 8
 /* Default value of bitfield dca{D}_cpuid[7:0] */
-#define tdm_dca_dcpuid_default 0x0
+#define HW_ATL_TDM_DCA_DCPUID_DEFAULT 0x0
 
 /* TX dca{D}_desc_en Bitfield Definitions
  * Preprocessor definitions for the bitfield "dca{D}_desc_en".
@@ -1725,17 +1979,17 @@
  */
 
 /* Register address for bitfield dca{D}_desc_en */
-#define tdm_dca_ddesc_en_adr(dca) (0x00008400 + (dca) * 0x4)
+#define HW_ATL_TDM_DCA_DDESC_EN_ADR(dca) (0x00008400 + (dca) * 0x4)
 /* Bitmask for bitfield dca{D}_desc_en */
-#define tdm_dca_ddesc_en_msk 0x80000000
+#define HW_ATL_TDM_DCA_DDESC_EN_MSK 0x80000000
 /* Inverted bitmask for bitfield dca{D}_desc_en */
-#define tdm_dca_ddesc_en_mskn 0x7FFFFFFF
+#define HW_ATL_TDM_DCA_DDESC_EN_MSKN 0x7FFFFFFF
 /* Lower bit position of bitfield dca{D}_desc_en */
-#define tdm_dca_ddesc_en_shift 31
+#define HW_ATL_TDM_DCA_DDESC_EN_SHIFT 31
 /* Width of bitfield dca{D}_desc_en */
-#define tdm_dca_ddesc_en_width 1
+#define HW_ATL_TDM_DCA_DDESC_EN_WIDTH 1
 /* Default value of bitfield dca{D}_desc_en */
-#define tdm_dca_ddesc_en_default 0x0
+#define HW_ATL_TDM_DCA_DDESC_EN_DEFAULT 0x0
 
 /* TX desc{D}_en Bitfield Definitions
  * Preprocessor definitions for the bitfield "desc{D}_en".
@@ -1744,17 +1998,17 @@
  */
 
 /* Register address for bitfield desc{D}_en */
-#define tdm_desc_den_adr(descriptor) (0x00007C08 + (descriptor) * 0x40)
+#define HW_ATL_TDM_DESC_DEN_ADR(descriptor) (0x00007C08 + (descriptor) * 0x40)
 /* Bitmask for bitfield desc{D}_en */
-#define tdm_desc_den_msk 0x80000000
+#define HW_ATL_TDM_DESC_DEN_MSK 0x80000000
 /* Inverted bitmask for bitfield desc{D}_en */
-#define tdm_desc_den_mskn 0x7FFFFFFF
+#define HW_ATL_TDM_DESC_DEN_MSKN 0x7FFFFFFF
 /* Lower bit position of bitfield desc{D}_en */
-#define tdm_desc_den_shift 31
+#define HW_ATL_TDM_DESC_DEN_SHIFT 31
 /* Width of bitfield desc{D}_en */
-#define tdm_desc_den_width 1
+#define HW_ATL_TDM_DESC_DEN_WIDTH 1
 /* Default value of bitfield desc{D}_en */
-#define tdm_desc_den_default 0x0
+#define HW_ATL_TDM_DESC_DEN_DEFAULT 0x0
 
 /* TX desc{D}_hd[C:0] Bitfield Definitions
  * Preprocessor definitions for the bitfield "desc{D}_hd[C:0]".
@@ -1763,15 +2017,15 @@
  */
 
 /* Register address for bitfield desc{D}_hd[C:0] */
-#define tdm_desc_dhd_adr(descriptor) (0x00007C0C + (descriptor) * 0x40)
+#define HW_ATL_TDM_DESC_DHD_ADR(descriptor) (0x00007C0C + (descriptor) * 0x40)
 /* Bitmask for bitfield desc{D}_hd[C:0] */
-#define tdm_desc_dhd_msk 0x00001FFF
+#define HW_ATL_TDM_DESC_DHD_MSK 0x00001FFF
 /* Inverted bitmask for bitfield desc{D}_hd[C:0] */
-#define tdm_desc_dhd_mskn 0xFFFFE000
+#define HW_ATL_TDM_DESC_DHD_MSKN 0xFFFFE000
 /* Lower bit position of bitfield desc{D}_hd[C:0] */
-#define tdm_desc_dhd_shift 0
+#define HW_ATL_TDM_DESC_DHD_SHIFT 0
 /* Width of bitfield desc{D}_hd[C:0] */
-#define tdm_desc_dhd_width 13
+#define HW_ATL_TDM_DESC_DHD_WIDTH 13
 
 /* TX desc{D}_len[9:0] Bitfield Definitions
  * Preprocessor definitions for the bitfield "desc{D}_len[9:0]".
@@ -1780,17 +2034,17 @@
  */
 
 /* Register address for bitfield desc{D}_len[9:0] */
-#define tdm_desc_dlen_adr(descriptor) (0x00007C08 + (descriptor) * 0x40)
+#define HW_ATL_TDM_DESC_DLEN_ADR(descriptor) (0x00007C08 + (descriptor) * 0x40)
 /* Bitmask for bitfield desc{D}_len[9:0] */
-#define tdm_desc_dlen_msk 0x00001FF8
+#define HW_ATL_TDM_DESC_DLEN_MSK 0x00001FF8
 /* Inverted bitmask for bitfield desc{D}_len[9:0] */
-#define tdm_desc_dlen_mskn 0xFFFFE007
+#define HW_ATL_TDM_DESC_DLEN_MSKN 0xFFFFE007
 /* Lower bit position of bitfield desc{D}_len[9:0] */
-#define tdm_desc_dlen_shift 3
+#define HW_ATL_TDM_DESC_DLEN_SHIFT 3
 /* Width of bitfield desc{D}_len[9:0] */
-#define tdm_desc_dlen_width 10
+#define HW_ATL_TDM_DESC_DLEN_WIDTH 10
 /* Default value of bitfield desc{D}_len[9:0] */
-#define tdm_desc_dlen_default 0x0
+#define HW_ATL_TDM_DESC_DLEN_DEFAULT 0x0
 
 /* TX desc{D}_wrb_thresh[6:0] Bitfield Definitions
  * Preprocessor definitions for the bitfield "desc{D}_wrb_thresh[6:0]".
@@ -1799,18 +2053,18 @@
  */
 
 /* Register address for bitfield desc{D}_wrb_thresh[6:0] */
-#define tdm_desc_dwrb_thresh_adr(descriptor) \
+#define HW_ATL_TDM_DESC_DWRB_THRESH_ADR(descriptor) \
 	(0x00007C18 + (descriptor) * 0x40)
 /* Bitmask for bitfield desc{D}_wrb_thresh[6:0] */
-#define tdm_desc_dwrb_thresh_msk 0x00007F00
+#define HW_ATL_TDM_DESC_DWRB_THRESH_MSK 0x00007F00
 /* Inverted bitmask for bitfield desc{D}_wrb_thresh[6:0] */
-#define tdm_desc_dwrb_thresh_mskn 0xFFFF80FF
+#define HW_ATL_TDM_DESC_DWRB_THRESH_MSKN 0xFFFF80FF
 /* Lower bit position of bitfield desc{D}_wrb_thresh[6:0] */
-#define tdm_desc_dwrb_thresh_shift 8
+#define HW_ATL_TDM_DESC_DWRB_THRESH_SHIFT 8
 /* Width of bitfield desc{D}_wrb_thresh[6:0] */
-#define tdm_desc_dwrb_thresh_width 7
+#define HW_ATL_TDM_DESC_DWRB_THRESH_WIDTH 7
 /* Default value of bitfield desc{D}_wrb_thresh[6:0] */
-#define tdm_desc_dwrb_thresh_default 0x0
+#define HW_ATL_TDM_DESC_DWRB_THRESH_DEFAULT 0x0
 
 /* TX tdm_int_mod_en Bitfield Definitions
  * Preprocessor definitions for the bitfield "tdm_int_mod_en".
@@ -1818,34 +2072,70 @@
  */
 
 /* Register address for bitfield tdm_int_mod_en */
-#define tdm_int_mod_en_adr 0x00007B40
+#define HW_ATL_TDM_INT_MOD_EN_ADR 0x00007B40
 /* Bitmask for bitfield tdm_int_mod_en */
-#define tdm_int_mod_en_msk 0x00000010
+#define HW_ATL_TDM_INT_MOD_EN_MSK 0x00000010
 /* Inverted bitmask for bitfield tdm_int_mod_en */
-#define tdm_int_mod_en_mskn 0xFFFFFFEF
+#define HW_ATL_TDM_INT_MOD_EN_MSKN 0xFFFFFFEF
 /* Lower bit position of bitfield tdm_int_mod_en */
-#define tdm_int_mod_en_shift 4
+#define HW_ATL_TDM_INT_MOD_EN_SHIFT 4
 /* Width of bitfield tdm_int_mod_en */
-#define tdm_int_mod_en_width 1
+#define HW_ATL_TDM_INT_MOD_EN_WIDTH 1
 /* Default value of bitfield tdm_int_mod_en */
-#define tdm_int_mod_en_default 0x0
+#define HW_ATL_TDM_INT_MOD_EN_DEFAULT 0x0
 
 /* TX lso_tcp_flag_mid[B:0] Bitfield Definitions
  * Preprocessor definitions for the bitfield "lso_tcp_flag_mid[B:0]".
  * PORT="pif_thm_lso_tcp_flag_mid_i[11:0]"
  */
 /* register address for bitfield lso_tcp_flag_mid[b:0] */
-#define thm_lso_tcp_flag_mid_adr 0x00007820
+#define HW_ATL_THM_LSO_TCP_FLAG_MID_ADR 0x00007820
 /* bitmask for bitfield lso_tcp_flag_mid[b:0] */
-#define thm_lso_tcp_flag_mid_msk 0x0fff0000
+#define HW_ATL_THM_LSO_TCP_FLAG_MID_MSK 0x0fff0000
 /* inverted bitmask for bitfield lso_tcp_flag_mid[b:0] */
-#define thm_lso_tcp_flag_mid_mskn 0xf000ffff
+#define HW_ATL_THM_LSO_TCP_FLAG_MID_MSKN 0xf000ffff
 /* lower bit position of bitfield lso_tcp_flag_mid[b:0] */
-#define thm_lso_tcp_flag_mid_shift 16
+#define HW_ATL_THM_LSO_TCP_FLAG_MID_SHIFT 16
 /* width of bitfield lso_tcp_flag_mid[b:0] */
-#define thm_lso_tcp_flag_mid_width 12
+#define HW_ATL_THM_LSO_TCP_FLAG_MID_WIDTH 12
 /* default value of bitfield lso_tcp_flag_mid[b:0] */
-#define thm_lso_tcp_flag_mid_default 0x0
+#define HW_ATL_THM_LSO_TCP_FLAG_MID_DEFAULT 0x0
+
+/* tx tx_tc_mode bitfield definitions
+ * preprocessor definitions for the bitfield "tx_tc_mode".
+ * port="pif_tpb_tx_tc_mode_i,pif_tps_tx_tc_mode_i"
+ */
+
+/* register address for bitfield tx_tc_mode */
+#define HW_ATL_TPB_TX_TC_MODE_ADDR 0x00007900
+/* bitmask for bitfield tx_tc_mode */
+#define HW_ATL_TPB_TX_TC_MODE_MSK 0x00000100
+/* inverted bitmask for bitfield tx_tc_mode */
+#define HW_ATL_TPB_TX_TC_MODE_MSKN 0xFFFFFEFF
+/* lower bit position of bitfield tx_tc_mode */
+#define HW_ATL_TPB_TX_TC_MODE_SHIFT 8
+/* width of bitfield tx_tc_mode */
+#define HW_ATL_TPB_TX_TC_MODE_WIDTH 1
+/* default value of bitfield tx_tc_mode */
+#define HW_ATL_TPB_TX_TC_MODE_DEFAULT 0x0
+
+/* tx tx_desc_rate_mode bitfield definitions
+ * preprocessor definitions for the bitfield "tx_desc_rate_mode".
+ * port="pif_tps_desc_rate_mode_i"
+ */
+
+/* register address for bitfield tx_desc_rate_mode */
+#define HW_ATL_TPS_TX_DESC_RATE_MODE_ADR 0x00007900
+/* bitmask for bitfield tx_desc_rate_mode */
+#define HW_ATL_TPS_TX_DESC_RATE_MODE_MSK 0x00000080
+/* inverted bitmask for bitfield tx_desc_rate_mode */
+#define HW_ATL_TPS_TX_DESC_RATE_MODE_MSKN 0xFFFFFF7F
+/* lower bit position of bitfield tx_desc_rate_mode */
+#define HW_ATL_TPS_TX_DESC_RATE_MODE_SHIFT 7
+/* width of bitfield tx_desc_rate_mode */
+#define HW_ATL_TPS_TX_DESC_RATE_MODE_WIDTH 1
+/* default value of bitfield tx_desc_rate_mode */
+#define HW_ATL_TPS_TX_DESC_RATE_MODE_DEFAULT 0x0
 
 /* tx tx_buf_en bitfield definitions
  * preprocessor definitions for the bitfield "tx_buf_en".
@@ -1853,17 +2143,17 @@
  */
 
 /* register address for bitfield tx_buf_en */
-#define tpb_tx_buf_en_adr 0x00007900
+#define HW_ATL_TPB_TX_BUF_EN_ADR 0x00007900
 /* bitmask for bitfield tx_buf_en */
-#define tpb_tx_buf_en_msk 0x00000001
+#define HW_ATL_TPB_TX_BUF_EN_MSK 0x00000001
 /* inverted bitmask for bitfield tx_buf_en */
-#define tpb_tx_buf_en_mskn 0xfffffffe
+#define HW_ATL_TPB_TX_BUF_EN_MSKN 0xfffffffe
 /* lower bit position of bitfield tx_buf_en */
-#define tpb_tx_buf_en_shift 0
+#define HW_ATL_TPB_TX_BUF_EN_SHIFT 0
 /* width of bitfield tx_buf_en */
-#define tpb_tx_buf_en_width 1
+#define HW_ATL_TPB_TX_BUF_EN_WIDTH 1
 /* default value of bitfield tx_buf_en */
-#define tpb_tx_buf_en_default 0x0
+#define HW_ATL_TPB_TX_BUF_EN_DEFAULT 0x0
 
 /* tx tx{b}_hi_thresh[c:0] bitfield definitions
  * preprocessor definitions for the bitfield "tx{b}_hi_thresh[c:0]".
@@ -1872,17 +2162,17 @@
  */
 
 /* register address for bitfield tx{b}_hi_thresh[c:0] */
-#define tpb_txbhi_thresh_adr(buffer) (0x00007914 + (buffer) * 0x10)
+#define HW_ATL_TPB_TXBHI_THRESH_ADR(buffer) (0x00007914 + (buffer) * 0x10)
 /* bitmask for bitfield tx{b}_hi_thresh[c:0] */
-#define tpb_txbhi_thresh_msk 0x1fff0000
+#define HW_ATL_TPB_TXBHI_THRESH_MSK 0x1fff0000
 /* inverted bitmask for bitfield tx{b}_hi_thresh[c:0] */
-#define tpb_txbhi_thresh_mskn 0xe000ffff
+#define HW_ATL_TPB_TXBHI_THRESH_MSKN 0xe000ffff
 /* lower bit position of bitfield tx{b}_hi_thresh[c:0] */
-#define tpb_txbhi_thresh_shift 16
+#define HW_ATL_TPB_TXBHI_THRESH_SHIFT 16
 /* width of bitfield tx{b}_hi_thresh[c:0] */
-#define tpb_txbhi_thresh_width 13
+#define HW_ATL_TPB_TXBHI_THRESH_WIDTH 13
 /* default value of bitfield tx{b}_hi_thresh[c:0] */
-#define tpb_txbhi_thresh_default 0x0
+#define HW_ATL_TPB_TXBHI_THRESH_DEFAULT 0x0
 
 /* tx tx{b}_lo_thresh[c:0] bitfield definitions
  * preprocessor definitions for the bitfield "tx{b}_lo_thresh[c:0]".
@@ -1891,17 +2181,17 @@
  */
 
 /* register address for bitfield tx{b}_lo_thresh[c:0] */
-#define tpb_txblo_thresh_adr(buffer) (0x00007914 + (buffer) * 0x10)
+#define HW_ATL_TPB_TXBLO_THRESH_ADR(buffer) (0x00007914 + (buffer) * 0x10)
 /* bitmask for bitfield tx{b}_lo_thresh[c:0] */
-#define tpb_txblo_thresh_msk 0x00001fff
+#define HW_ATL_TPB_TXBLO_THRESH_MSK 0x00001fff
 /* inverted bitmask for bitfield tx{b}_lo_thresh[c:0] */
-#define tpb_txblo_thresh_mskn 0xffffe000
+#define HW_ATL_TPB_TXBLO_THRESH_MSKN 0xffffe000
 /* lower bit position of bitfield tx{b}_lo_thresh[c:0] */
-#define tpb_txblo_thresh_shift 0
+#define HW_ATL_TPB_TXBLO_THRESH_SHIFT 0
 /* width of bitfield tx{b}_lo_thresh[c:0] */
-#define tpb_txblo_thresh_width 13
+#define HW_ATL_TPB_TXBLO_THRESH_WIDTH 13
 /* default value of bitfield tx{b}_lo_thresh[c:0] */
-#define tpb_txblo_thresh_default 0x0
+#define HW_ATL_TPB_TXBLO_THRESH_DEFAULT 0x0
 
 /* tx dma_sys_loopback bitfield definitions
  * preprocessor definitions for the bitfield "dma_sys_loopback".
@@ -1909,17 +2199,35 @@
  */
 
 /* register address for bitfield dma_sys_loopback */
-#define tpb_dma_sys_lbk_adr 0x00007000
+#define HW_ATL_TPB_DMA_SYS_LBK_ADR 0x00007000
 /* bitmask for bitfield dma_sys_loopback */
-#define tpb_dma_sys_lbk_msk 0x00000040
+#define HW_ATL_TPB_DMA_SYS_LBK_MSK 0x00000040
 /* inverted bitmask for bitfield dma_sys_loopback */
-#define tpb_dma_sys_lbk_mskn 0xffffffbf
+#define HW_ATL_TPB_DMA_SYS_LBK_MSKN 0xffffffbf
 /* lower bit position of bitfield dma_sys_loopback */
-#define tpb_dma_sys_lbk_shift 6
+#define HW_ATL_TPB_DMA_SYS_LBK_SHIFT 6
 /* width of bitfield dma_sys_loopback */
-#define tpb_dma_sys_lbk_width 1
+#define HW_ATL_TPB_DMA_SYS_LBK_WIDTH 1
 /* default value of bitfield dma_sys_loopback */
-#define tpb_dma_sys_lbk_default 0x0
+#define HW_ATL_TPB_DMA_SYS_LBK_DEFAULT 0x0
+
+/* tx dma_net_loopback bitfield definitions
+ * preprocessor definitions for the bitfield "dma_net_loopback".
+ * port="pif_tpb_dma_net_lbk_i"
+ */
+
+/* register address for bitfield dma_net_loopback */
+#define HW_ATL_TPB_DMA_NET_LBK_ADR 0x00007000
+/* bitmask for bitfield dma_net_loopback */
+#define HW_ATL_TPB_DMA_NET_LBK_MSK 0x00000010
+/* inverted bitmask for bitfield dma_net_loopback */
+#define HW_ATL_TPB_DMA_NET_LBK_MSKN 0xffffffef
+/* lower bit position of bitfield dma_net_loopback */
+#define HW_ATL_TPB_DMA_NET_LBK_SHIFT 4
+/* width of bitfield dma_net_loopback */
+#define HW_ATL_TPB_DMA_NET_LBK_WIDTH 1
+/* default value of bitfield dma_net_loopback */
+#define HW_ATL_TPB_DMA_NET_LBK_DEFAULT 0x0
 
 /* tx tx{b}_buf_size[7:0] bitfield definitions
  * preprocessor definitions for the bitfield "tx{b}_buf_size[7:0]".
@@ -1928,17 +2236,17 @@
  */
 
 /* register address for bitfield tx{b}_buf_size[7:0] */
-#define tpb_txbbuf_size_adr(buffer) (0x00007910 + (buffer) * 0x10)
+#define HW_ATL_TPB_TXBBUF_SIZE_ADR(buffer) (0x00007910 + (buffer) * 0x10)
 /* bitmask for bitfield tx{b}_buf_size[7:0] */
-#define tpb_txbbuf_size_msk 0x000000ff
+#define HW_ATL_TPB_TXBBUF_SIZE_MSK 0x000000ff
 /* inverted bitmask for bitfield tx{b}_buf_size[7:0] */
-#define tpb_txbbuf_size_mskn 0xffffff00
+#define HW_ATL_TPB_TXBBUF_SIZE_MSKN 0xffffff00
 /* lower bit position of bitfield tx{b}_buf_size[7:0] */
-#define tpb_txbbuf_size_shift 0
+#define HW_ATL_TPB_TXBBUF_SIZE_SHIFT 0
 /* width of bitfield tx{b}_buf_size[7:0] */
-#define tpb_txbbuf_size_width 8
+#define HW_ATL_TPB_TXBBUF_SIZE_WIDTH 8
 /* default value of bitfield tx{b}_buf_size[7:0] */
-#define tpb_txbbuf_size_default 0x0
+#define HW_ATL_TPB_TXBBUF_SIZE_DEFAULT 0x0
 
 /* tx tx_scp_ins_en bitfield definitions
  * preprocessor definitions for the bitfield "tx_scp_ins_en".
@@ -1946,17 +2254,35 @@
  */
 
 /* register address for bitfield tx_scp_ins_en */
-#define tpb_tx_scp_ins_en_adr 0x00007900
+#define HW_ATL_TPB_TX_SCP_INS_EN_ADR 0x00007900
 /* bitmask for bitfield tx_scp_ins_en */
-#define tpb_tx_scp_ins_en_msk 0x00000004
+#define HW_ATL_TPB_TX_SCP_INS_EN_MSK 0x00000004
 /* inverted bitmask for bitfield tx_scp_ins_en */
-#define tpb_tx_scp_ins_en_mskn 0xfffffffb
+#define HW_ATL_TPB_TX_SCP_INS_EN_MSKN 0xfffffffb
 /* lower bit position of bitfield tx_scp_ins_en */
-#define tpb_tx_scp_ins_en_shift 2
+#define HW_ATL_TPB_TX_SCP_INS_EN_SHIFT 2
 /* width of bitfield tx_scp_ins_en */
-#define tpb_tx_scp_ins_en_width 1
+#define HW_ATL_TPB_TX_SCP_INS_EN_WIDTH 1
 /* default value of bitfield tx_scp_ins_en */
-#define tpb_tx_scp_ins_en_default 0x0
+#define HW_ATL_TPB_TX_SCP_INS_EN_DEFAULT 0x0
+
+/* tx tx_clk_gate_en bitfield definitions
+ * preprocessor definitions for the bitfield "tx_clk_gate_en".
+ * port="pif_tpb_clk_gate_en_i"
+ */
+
+/* register address for bitfield tx_clk_gate_en */
+#define HW_ATL_TPB_TX_CLK_GATE_EN_ADR 0x00007900
+/* bitmask for bitfield tx_clk_gate_en */
+#define HW_ATL_TPB_TX_CLK_GATE_EN_MSK 0x00000010
+/* inverted bitmask for bitfield tx_clk_gate_en */
+#define HW_ATL_TPB_TX_CLK_GATE_EN_MSKN 0xffffffef
+/* lower bit position of bitfield tx_clk_gate_en */
+#define HW_ATL_TPB_TX_CLK_GATE_EN_SHIFT 4
+/* width of bitfield tx_clk_gate_en */
+#define HW_ATL_TPB_TX_CLK_GATE_EN_WIDTH 1
+/* default value of bitfield tx_clk_gate_en */
+#define HW_ATL_TPB_TX_CLK_GATE_EN_DEFAULT 0x1
 
 /* tx ipv4_chk_en bitfield definitions
  * preprocessor definitions for the bitfield "ipv4_chk_en".
@@ -1964,17 +2290,17 @@
  */
 
 /* register address for bitfield ipv4_chk_en */
-#define tpo_ipv4chk_en_adr 0x00007800
+#define HW_ATL_TPO_IPV4CHK_EN_ADR 0x00007800
 /* bitmask for bitfield ipv4_chk_en */
-#define tpo_ipv4chk_en_msk 0x00000002
+#define HW_ATL_TPO_IPV4CHK_EN_MSK 0x00000002
 /* inverted bitmask for bitfield ipv4_chk_en */
-#define tpo_ipv4chk_en_mskn 0xfffffffd
+#define HW_ATL_TPO_IPV4CHK_EN_MSKN 0xfffffffd
 /* lower bit position of bitfield ipv4_chk_en */
-#define tpo_ipv4chk_en_shift 1
+#define HW_ATL_TPO_IPV4CHK_EN_SHIFT 1
 /* width of bitfield ipv4_chk_en */
-#define tpo_ipv4chk_en_width 1
+#define HW_ATL_TPO_IPV4CHK_EN_WIDTH 1
 /* default value of bitfield ipv4_chk_en */
-#define tpo_ipv4chk_en_default 0x0
+#define HW_ATL_TPO_IPV4CHK_EN_DEFAULT 0x0
 
 /* tx l4_chk_en bitfield definitions
  * preprocessor definitions for the bitfield "l4_chk_en".
@@ -1982,17 +2308,17 @@
  */
 
 /* register address for bitfield l4_chk_en */
-#define tpol4chk_en_adr 0x00007800
+#define HW_ATL_TPOL4CHK_EN_ADR 0x00007800
 /* bitmask for bitfield l4_chk_en */
-#define tpol4chk_en_msk 0x00000001
+#define HW_ATL_TPOL4CHK_EN_MSK 0x00000001
 /* inverted bitmask for bitfield l4_chk_en */
-#define tpol4chk_en_mskn 0xfffffffe
+#define HW_ATL_TPOL4CHK_EN_MSKN 0xfffffffe
 /* lower bit position of bitfield l4_chk_en */
-#define tpol4chk_en_shift 0
+#define HW_ATL_TPOL4CHK_EN_SHIFT 0
 /* width of bitfield l4_chk_en */
-#define tpol4chk_en_width 1
+#define HW_ATL_TPOL4CHK_EN_WIDTH 1
 /* default value of bitfield l4_chk_en */
-#define tpol4chk_en_default 0x0
+#define HW_ATL_TPOL4CHK_EN_DEFAULT 0x0
 
 /* tx pkt_sys_loopback bitfield definitions
  * preprocessor definitions for the bitfield "pkt_sys_loopback".
@@ -2000,17 +2326,17 @@
  */
 
 /* register address for bitfield pkt_sys_loopback */
-#define tpo_pkt_sys_lbk_adr 0x00007000
+#define HW_ATL_TPO_PKT_SYS_LBK_ADR 0x00007000
 /* bitmask for bitfield pkt_sys_loopback */
-#define tpo_pkt_sys_lbk_msk 0x00000080
+#define HW_ATL_TPO_PKT_SYS_LBK_MSK 0x00000080
 /* inverted bitmask for bitfield pkt_sys_loopback */
-#define tpo_pkt_sys_lbk_mskn 0xffffff7f
+#define HW_ATL_TPO_PKT_SYS_LBK_MSKN 0xffffff7f
 /* lower bit position of bitfield pkt_sys_loopback */
-#define tpo_pkt_sys_lbk_shift 7
+#define HW_ATL_TPO_PKT_SYS_LBK_SHIFT 7
 /* width of bitfield pkt_sys_loopback */
-#define tpo_pkt_sys_lbk_width 1
+#define HW_ATL_TPO_PKT_SYS_LBK_WIDTH 1
 /* default value of bitfield pkt_sys_loopback */
-#define tpo_pkt_sys_lbk_default 0x0
+#define HW_ATL_TPO_PKT_SYS_LBK_DEFAULT 0x0
 
 /* tx data_tc_arb_mode bitfield definitions
  * preprocessor definitions for the bitfield "data_tc_arb_mode".
@@ -2018,17 +2344,69 @@
  */
 
 /* register address for bitfield data_tc_arb_mode */
-#define tps_data_tc_arb_mode_adr 0x00007100
+#define HW_ATL_TPS_DATA_TC_ARB_MODE_ADR 0x00007100
 /* bitmask for bitfield data_tc_arb_mode */
-#define tps_data_tc_arb_mode_msk 0x00000001
+#define HW_ATL_TPS_DATA_TC_ARB_MODE_MSK 0x00000001
 /* inverted bitmask for bitfield data_tc_arb_mode */
-#define tps_data_tc_arb_mode_mskn 0xfffffffe
+#define HW_ATL_TPS_DATA_TC_ARB_MODE_MSKN 0xfffffffe
 /* lower bit position of bitfield data_tc_arb_mode */
-#define tps_data_tc_arb_mode_shift 0
+#define HW_ATL_TPS_DATA_TC_ARB_MODE_SHIFT 0
 /* width of bitfield data_tc_arb_mode */
-#define tps_data_tc_arb_mode_width 1
+#define HW_ATL_TPS_DATA_TC_ARB_MODE_WIDTH 1
 /* default value of bitfield data_tc_arb_mode */
-#define tps_data_tc_arb_mode_default 0x0
+#define HW_ATL_TPS_DATA_TC_ARB_MODE_DEFAULT 0x0
+
+/* tx desc{r}_rate_en bitfield definitions
+ * preprocessor definitions for the bitfield "desc{r}_rate_en".
+ * port="pif_tps_desc_rate_en_i[0]"
+ */
+
+/* register address for bitfield desc{r}_rate_en */
+#define HW_ATL_TPS_DESC_RATE_EN_ADR(desc) (0x00007408 + (desc) * 0x10)
+/* bitmask for bitfield desc{r}_rate_en */
+#define HW_ATL_TPS_DESC_RATE_EN_MSK 0x80000000
+/* inverted bitmask for bitfield desc{r}_rate_en */
+#define HW_ATL_TPS_DESC_RATE_EN_MSKN 0x7FFFFFFF
+/* lower bit position of bitfield desc{r}_rate_en */
+#define HW_ATL_TPS_DESC_RATE_EN_SHIFT 31
+/* width of bitfield desc{r}_rate_en */
+#define HW_ATL_TPS_DESC_RATE_EN_WIDTH 1
+/* default value of bitfield desc{r}_rate_en */
+#define HW_ATL_TPS_DESC_RATE_EN_DEFAULT 0x0
+
+/* tx desc{r}_rate_x bitfield definitions
+ * preprocessor definitions for the bitfield "desc{r}_rate_x".
+ * port="pif_tps_desc0_rate_x"
+ */
+/* register address for bitfield desc{r}_rate_x */
+#define HW_ATL_TPS_DESC_RATE_X_ADR(desc) (0x00007408 + (desc) * 0x10)
+/* bitmask for bitfield desc{r}_rate_x */
+#define HW_ATL_TPS_DESC_RATE_X_MSK 0x03FF0000
+/* inverted bitmask for bitfield desc{r}_rate_x */
+#define HW_ATL_TPS_DESC_RATE_X_MSKN 0xFC00FFFF
+/* lower bit position of bitfield desc{r}_rate_x */
+#define HW_ATL_TPS_DESC_RATE_X_SHIFT 16
+/* width of bitfield desc{r}_rate_x */
+#define HW_ATL_TPS_DESC_RATE_X_WIDTH 10
+/* default value of bitfield desc{r}_rate_x */
+#define HW_ATL_TPS_DESC_RATE_X_DEFAULT 0x0
+
+/* tx desc{r}_rate_y bitfield definitions
+ * preprocessor definitions for the bitfield "desc{r}_rate_y".
+ * port="pif_tps_desc0_rate_y"
+ */
+/* register address for bitfield desc{r}_rate_y */
+#define HW_ATL_TPS_DESC_RATE_Y_ADR(desc) (0x00007408 + (desc) * 0x10)
+/* bitmask for bitfield desc{r}_rate_y */
+#define HW_ATL_TPS_DESC_RATE_Y_MSK 0x00003FFF
+/* inverted bitmask for bitfield desc{r}_rate_y */
+#define HW_ATL_TPS_DESC_RATE_Y_MSKN 0xFFFFC000
+/* lower bit position of bitfield desc{r}_rate_y */
+#define HW_ATL_TPS_DESC_RATE_Y_SHIFT 0
+/* width of bitfield desc{r}_rate_y */
+#define HW_ATL_TPS_DESC_RATE_Y_WIDTH 14
+/* default value of bitfield desc{r}_rate_y */
+#define HW_ATL_TPS_DESC_RATE_Y_DEFAULT 0x0
 
 /* tx desc_rate_ta_rst bitfield definitions
  * preprocessor definitions for the bitfield "desc_rate_ta_rst".
@@ -2036,17 +2414,17 @@
  */
 
 /* register address for bitfield desc_rate_ta_rst */
-#define tps_desc_rate_ta_rst_adr 0x00007310
+#define HW_ATL_TPS_DESC_RATE_TA_RST_ADR 0x00007310
 /* bitmask for bitfield desc_rate_ta_rst */
-#define tps_desc_rate_ta_rst_msk 0x80000000
+#define HW_ATL_TPS_DESC_RATE_TA_RST_MSK 0x80000000
 /* inverted bitmask for bitfield desc_rate_ta_rst */
-#define tps_desc_rate_ta_rst_mskn 0x7fffffff
+#define HW_ATL_TPS_DESC_RATE_TA_RST_MSKN 0x7fffffff
 /* lower bit position of bitfield desc_rate_ta_rst */
-#define tps_desc_rate_ta_rst_shift 31
+#define HW_ATL_TPS_DESC_RATE_TA_RST_SHIFT 31
 /* width of bitfield desc_rate_ta_rst */
-#define tps_desc_rate_ta_rst_width 1
+#define HW_ATL_TPS_DESC_RATE_TA_RST_WIDTH 1
 /* default value of bitfield desc_rate_ta_rst */
-#define tps_desc_rate_ta_rst_default 0x0
+#define HW_ATL_TPS_DESC_RATE_TA_RST_DEFAULT 0x0
 
 /* tx desc_rate_limit[a:0] bitfield definitions
  * preprocessor definitions for the bitfield "desc_rate_limit[a:0]".
@@ -2054,17 +2432,17 @@
  */
 
 /* register address for bitfield desc_rate_limit[a:0] */
-#define tps_desc_rate_lim_adr 0x00007310
+#define HW_ATL_TPS_DESC_RATE_LIM_ADR 0x00007310
 /* bitmask for bitfield desc_rate_limit[a:0] */
-#define tps_desc_rate_lim_msk 0x000007ff
+#define HW_ATL_TPS_DESC_RATE_LIM_MSK 0x000007ff
 /* inverted bitmask for bitfield desc_rate_limit[a:0] */
-#define tps_desc_rate_lim_mskn 0xfffff800
+#define HW_ATL_TPS_DESC_RATE_LIM_MSKN 0xfffff800
 /* lower bit position of bitfield desc_rate_limit[a:0] */
-#define tps_desc_rate_lim_shift 0
+#define HW_ATL_TPS_DESC_RATE_LIM_SHIFT 0
 /* width of bitfield desc_rate_limit[a:0] */
-#define tps_desc_rate_lim_width 11
+#define HW_ATL_TPS_DESC_RATE_LIM_WIDTH 11
 /* default value of bitfield desc_rate_limit[a:0] */
-#define tps_desc_rate_lim_default 0x0
+#define HW_ATL_TPS_DESC_RATE_LIM_DEFAULT 0x0
 
 /* tx desc_tc_arb_mode[1:0] bitfield definitions
  * preprocessor definitions for the bitfield "desc_tc_arb_mode[1:0]".
@@ -2072,17 +2450,17 @@
  */
 
 /* register address for bitfield desc_tc_arb_mode[1:0] */
-#define tps_desc_tc_arb_mode_adr 0x00007200
+#define HW_ATL_TPS_DESC_TC_ARB_MODE_ADR 0x00007200
 /* bitmask for bitfield desc_tc_arb_mode[1:0] */
-#define tps_desc_tc_arb_mode_msk 0x00000003
+#define HW_ATL_TPS_DESC_TC_ARB_MODE_MSK 0x00000003
 /* inverted bitmask for bitfield desc_tc_arb_mode[1:0] */
-#define tps_desc_tc_arb_mode_mskn 0xfffffffc
+#define HW_ATL_TPS_DESC_TC_ARB_MODE_MSKN 0xfffffffc
 /* lower bit position of bitfield desc_tc_arb_mode[1:0] */
-#define tps_desc_tc_arb_mode_shift 0
+#define HW_ATL_TPS_DESC_TC_ARB_MODE_SHIFT 0
 /* width of bitfield desc_tc_arb_mode[1:0] */
-#define tps_desc_tc_arb_mode_width 2
+#define HW_ATL_TPS_DESC_TC_ARB_MODE_WIDTH 2
 /* default value of bitfield desc_tc_arb_mode[1:0] */
-#define tps_desc_tc_arb_mode_default 0x0
+#define HW_ATL_TPS_DESC_TC_ARB_MODE_DEFAULT 0x0
 
 /* tx desc_tc{t}_credit_max[b:0] bitfield definitions
  * preprocessor definitions for the bitfield "desc_tc{t}_credit_max[b:0]".
@@ -2091,17 +2469,17 @@
  */
 
 /* register address for bitfield desc_tc{t}_credit_max[b:0] */
-#define tps_desc_tctcredit_max_adr(tc) (0x00007210 + (tc) * 0x4)
+#define HW_ATL_TPS_DESC_TCTCREDIT_MAX_ADR(tc) (0x00007210 + (tc) * 0x4)
 /* bitmask for bitfield desc_tc{t}_credit_max[b:0] */
-#define tps_desc_tctcredit_max_msk 0x0fff0000
+#define HW_ATL_TPS_DESC_TCTCREDIT_MAX_MSK 0x0fff0000
 /* inverted bitmask for bitfield desc_tc{t}_credit_max[b:0] */
-#define tps_desc_tctcredit_max_mskn 0xf000ffff
+#define HW_ATL_TPS_DESC_TCTCREDIT_MAX_MSKN 0xf000ffff
 /* lower bit position of bitfield desc_tc{t}_credit_max[b:0] */
-#define tps_desc_tctcredit_max_shift 16
+#define HW_ATL_TPS_DESC_TCTCREDIT_MAX_SHIFT 16
 /* width of bitfield desc_tc{t}_credit_max[b:0] */
-#define tps_desc_tctcredit_max_width 12
+#define HW_ATL_TPS_DESC_TCTCREDIT_MAX_WIDTH 12
 /* default value of bitfield desc_tc{t}_credit_max[b:0] */
-#define tps_desc_tctcredit_max_default 0x0
+#define HW_ATL_TPS_DESC_TCTCREDIT_MAX_DEFAULT 0x0
 
 /* tx desc_tc{t}_weight[8:0] bitfield definitions
  * preprocessor definitions for the bitfield "desc_tc{t}_weight[8:0]".
@@ -2110,17 +2488,17 @@
  */
 
 /* register address for bitfield desc_tc{t}_weight[8:0] */
-#define tps_desc_tctweight_adr(tc) (0x00007210 + (tc) * 0x4)
+#define HW_ATL_TPS_DESC_TCTWEIGHT_ADR(tc) (0x00007210 + (tc) * 0x4)
 /* bitmask for bitfield desc_tc{t}_weight[8:0] */
-#define tps_desc_tctweight_msk 0x000001ff
+#define HW_ATL_TPS_DESC_TCTWEIGHT_MSK 0x000001ff
 /* inverted bitmask for bitfield desc_tc{t}_weight[8:0] */
-#define tps_desc_tctweight_mskn 0xfffffe00
+#define HW_ATL_TPS_DESC_TCTWEIGHT_MSKN 0xfffffe00
 /* lower bit position of bitfield desc_tc{t}_weight[8:0] */
-#define tps_desc_tctweight_shift 0
+#define HW_ATL_TPS_DESC_TCTWEIGHT_SHIFT 0
 /* width of bitfield desc_tc{t}_weight[8:0] */
-#define tps_desc_tctweight_width 9
+#define HW_ATL_TPS_DESC_TCTWEIGHT_WIDTH 9
 /* default value of bitfield desc_tc{t}_weight[8:0] */
-#define tps_desc_tctweight_default 0x0
+#define HW_ATL_TPS_DESC_TCTWEIGHT_DEFAULT 0x0
 
 /* tx desc_vm_arb_mode bitfield definitions
  * preprocessor definitions for the bitfield "desc_vm_arb_mode".
@@ -2128,17 +2506,17 @@
  */
 
 /* register address for bitfield desc_vm_arb_mode */
-#define tps_desc_vm_arb_mode_adr 0x00007300
+#define HW_ATL_TPS_DESC_VM_ARB_MODE_ADR 0x00007300
 /* bitmask for bitfield desc_vm_arb_mode */
-#define tps_desc_vm_arb_mode_msk 0x00000001
+#define HW_ATL_TPS_DESC_VM_ARB_MODE_MSK 0x00000001
 /* inverted bitmask for bitfield desc_vm_arb_mode */
-#define tps_desc_vm_arb_mode_mskn 0xfffffffe
+#define HW_ATL_TPS_DESC_VM_ARB_MODE_MSKN 0xfffffffe
 /* lower bit position of bitfield desc_vm_arb_mode */
-#define tps_desc_vm_arb_mode_shift 0
+#define HW_ATL_TPS_DESC_VM_ARB_MODE_SHIFT 0
 /* width of bitfield desc_vm_arb_mode */
-#define tps_desc_vm_arb_mode_width 1
+#define HW_ATL_TPS_DESC_VM_ARB_MODE_WIDTH 1
 /* default value of bitfield desc_vm_arb_mode */
-#define tps_desc_vm_arb_mode_default 0x0
+#define HW_ATL_TPS_DESC_VM_ARB_MODE_DEFAULT 0x0
 
 /* tx data_tc{t}_credit_max[b:0] bitfield definitions
  * preprocessor definitions for the bitfield "data_tc{t}_credit_max[b:0]".
@@ -2147,17 +2525,17 @@
  */
 
 /* register address for bitfield data_tc{t}_credit_max[b:0] */
-#define tps_data_tctcredit_max_adr(tc) (0x00007110 + (tc) * 0x4)
+#define HW_ATL_TPS_DATA_TCTCREDIT_MAX_ADR(tc) (0x00007110 + (tc) * 0x4)
 /* bitmask for bitfield data_tc{t}_credit_max[b:0] */
-#define tps_data_tctcredit_max_msk 0x0fff0000
+#define HW_ATL_TPS_DATA_TCTCREDIT_MAX_MSK 0x0fff0000
 /* inverted bitmask for bitfield data_tc{t}_credit_max[b:0] */
-#define tps_data_tctcredit_max_mskn 0xf000ffff
+#define HW_ATL_TPS_DATA_TCTCREDIT_MAX_MSKN 0xf000ffff
 /* lower bit position of bitfield data_tc{t}_credit_max[b:0] */
-#define tps_data_tctcredit_max_shift 16
+#define HW_ATL_TPS_DATA_TCTCREDIT_MAX_SHIFT 16
 /* width of bitfield data_tc{t}_credit_max[b:0] */
-#define tps_data_tctcredit_max_width 12
+#define HW_ATL_TPS_DATA_TCTCREDIT_MAX_WIDTH 12
 /* default value of bitfield data_tc{t}_credit_max[b:0] */
-#define tps_data_tctcredit_max_default 0x0
+#define HW_ATL_TPS_DATA_TCTCREDIT_MAX_DEFAULT 0x0
 
 /* tx data_tc{t}_weight[8:0] bitfield definitions
  * preprocessor definitions for the bitfield "data_tc{t}_weight[8:0]".
@@ -2166,17 +2544,17 @@
  */
 
 /* register address for bitfield data_tc{t}_weight[8:0] */
-#define tps_data_tctweight_adr(tc) (0x00007110 + (tc) * 0x4)
+#define HW_ATL_TPS_DATA_TCTWEIGHT_ADR(tc) (0x00007110 + (tc) * 0x4)
 /* bitmask for bitfield data_tc{t}_weight[8:0] */
-#define tps_data_tctweight_msk 0x000001ff
+#define HW_ATL_TPS_DATA_TCTWEIGHT_MSK 0x000001ff
 /* inverted bitmask for bitfield data_tc{t}_weight[8:0] */
-#define tps_data_tctweight_mskn 0xfffffe00
+#define HW_ATL_TPS_DATA_TCTWEIGHT_MSKN 0xfffffe00
 /* lower bit position of bitfield data_tc{t}_weight[8:0] */
-#define tps_data_tctweight_shift 0
+#define HW_ATL_TPS_DATA_TCTWEIGHT_SHIFT 0
 /* width of bitfield data_tc{t}_weight[8:0] */
-#define tps_data_tctweight_width 9
+#define HW_ATL_TPS_DATA_TCTWEIGHT_WIDTH 9
 /* default value of bitfield data_tc{t}_weight[8:0] */
-#define tps_data_tctweight_default 0x0
+#define HW_ATL_TPS_DATA_TCTWEIGHT_DEFAULT 0x0
 
 /* tx reg_res_dsbl bitfield definitions
  * preprocessor definitions for the bitfield "reg_res_dsbl".
@@ -2184,17 +2562,17 @@
  */
 
 /* register address for bitfield reg_res_dsbl */
-#define tx_reg_res_dsbl_adr 0x00007000
+#define HW_ATL_TX_REG_RES_DSBL_ADR 0x00007000
 /* bitmask for bitfield reg_res_dsbl */
-#define tx_reg_res_dsbl_msk 0x20000000
+#define HW_ATL_TX_REG_RES_DSBL_MSK 0x20000000
 /* inverted bitmask for bitfield reg_res_dsbl */
-#define tx_reg_res_dsbl_mskn 0xdfffffff
+#define HW_ATL_TX_REG_RES_DSBL_MSKN 0xdfffffff
 /* lower bit position of bitfield reg_res_dsbl */
-#define tx_reg_res_dsbl_shift 29
+#define HW_ATL_TX_REG_RES_DSBL_SHIFT 29
 /* width of bitfield reg_res_dsbl */
-#define tx_reg_res_dsbl_width 1
+#define HW_ATL_TX_REG_RES_DSBL_WIDTH 1
 /* default value of bitfield reg_res_dsbl */
-#define tx_reg_res_dsbl_default 0x1
+#define HW_ATL_TX_REG_RES_DSBL_DEFAULT 0x1
 
 /* mac_phy register access busy bitfield definitions
  * preprocessor definitions for the bitfield "register access busy".
@@ -2202,15 +2580,15 @@
  */
 
 /* register address for bitfield register access busy */
-#define msm_reg_access_busy_adr 0x00004400
+#define HW_ATL_MSM_REG_ACCESS_BUSY_ADR 0x00004400
 /* bitmask for bitfield register access busy */
-#define msm_reg_access_busy_msk 0x00001000
+#define HW_ATL_MSM_REG_ACCESS_BUSY_MSK 0x00001000
 /* inverted bitmask for bitfield register access busy */
-#define msm_reg_access_busy_mskn 0xffffefff
+#define HW_ATL_MSM_REG_ACCESS_BUSY_MSKN 0xffffefff
 /* lower bit position of bitfield register access busy */
-#define msm_reg_access_busy_shift 12
+#define HW_ATL_MSM_REG_ACCESS_BUSY_SHIFT 12
 /* width of bitfield register access busy */
-#define msm_reg_access_busy_width 1
+#define HW_ATL_MSM_REG_ACCESS_BUSY_WIDTH 1
 
 /* mac_phy msm register address[7:0] bitfield definitions
  * preprocessor definitions for the bitfield "msm register address[7:0]".
@@ -2218,17 +2596,17 @@
  */
 
 /* register address for bitfield msm register address[7:0] */
-#define msm_reg_addr_adr 0x00004400
+#define HW_ATL_MSM_REG_ADDR_ADR 0x00004400
 /* bitmask for bitfield msm register address[7:0] */
-#define msm_reg_addr_msk 0x000000ff
+#define HW_ATL_MSM_REG_ADDR_MSK 0x000000ff
 /* inverted bitmask for bitfield msm register address[7:0] */
-#define msm_reg_addr_mskn 0xffffff00
+#define HW_ATL_MSM_REG_ADDR_MSKN 0xffffff00
 /* lower bit position of bitfield msm register address[7:0] */
-#define msm_reg_addr_shift 0
+#define HW_ATL_MSM_REG_ADDR_SHIFT 0
 /* width of bitfield msm register address[7:0] */
-#define msm_reg_addr_width 8
+#define HW_ATL_MSM_REG_ADDR_WIDTH 8
 /* default value of bitfield msm register address[7:0] */
-#define msm_reg_addr_default 0x0
+#define HW_ATL_MSM_REG_ADDR_DEFAULT 0x0
 
 /* mac_phy register read strobe bitfield definitions
  * preprocessor definitions for the bitfield "register read strobe".
@@ -2236,17 +2614,17 @@
  */
 
 /* register address for bitfield register read strobe */
-#define msm_reg_rd_strobe_adr 0x00004400
+#define HW_ATL_MSM_REG_RD_STROBE_ADR 0x00004400
 /* bitmask for bitfield register read strobe */
-#define msm_reg_rd_strobe_msk 0x00000200
+#define HW_ATL_MSM_REG_RD_STROBE_MSK 0x00000200
 /* inverted bitmask for bitfield register read strobe */
-#define msm_reg_rd_strobe_mskn 0xfffffdff
+#define HW_ATL_MSM_REG_RD_STROBE_MSKN 0xfffffdff
 /* lower bit position of bitfield register read strobe */
-#define msm_reg_rd_strobe_shift 9
+#define HW_ATL_MSM_REG_RD_STROBE_SHIFT 9
 /* width of bitfield register read strobe */
-#define msm_reg_rd_strobe_width 1
+#define HW_ATL_MSM_REG_RD_STROBE_WIDTH 1
 /* default value of bitfield register read strobe */
-#define msm_reg_rd_strobe_default 0x0
+#define HW_ATL_MSM_REG_RD_STROBE_DEFAULT 0x0
 
 /* mac_phy msm register read data[31:0] bitfield definitions
  * preprocessor definitions for the bitfield "msm register read data[31:0]".
@@ -2254,15 +2632,15 @@
  */
 
 /* register address for bitfield msm register read data[31:0] */
-#define msm_reg_rd_data_adr 0x00004408
+#define HW_ATL_MSM_REG_RD_DATA_ADR 0x00004408
 /* bitmask for bitfield msm register read data[31:0] */
-#define msm_reg_rd_data_msk 0xffffffff
+#define HW_ATL_MSM_REG_RD_DATA_MSK 0xffffffff
 /* inverted bitmask for bitfield msm register read data[31:0] */
-#define msm_reg_rd_data_mskn 0x00000000
+#define HW_ATL_MSM_REG_RD_DATA_MSKN 0x00000000
 /* lower bit position of bitfield msm register read data[31:0] */
-#define msm_reg_rd_data_shift 0
+#define HW_ATL_MSM_REG_RD_DATA_SHIFT 0
 /* width of bitfield msm register read data[31:0] */
-#define msm_reg_rd_data_width 32
+#define HW_ATL_MSM_REG_RD_DATA_WIDTH 32
 
 /* mac_phy msm register write data[31:0] bitfield definitions
  * preprocessor definitions for the bitfield "msm register write data[31:0]".
@@ -2270,17 +2648,17 @@
  */
 
 /* register address for bitfield msm register write data[31:0] */
-#define msm_reg_wr_data_adr 0x00004404
+#define HW_ATL_MSM_REG_WR_DATA_ADR 0x00004404
 /* bitmask for bitfield msm register write data[31:0] */
-#define msm_reg_wr_data_msk 0xffffffff
+#define HW_ATL_MSM_REG_WR_DATA_MSK 0xffffffff
 /* inverted bitmask for bitfield msm register write data[31:0] */
-#define msm_reg_wr_data_mskn 0x00000000
+#define HW_ATL_MSM_REG_WR_DATA_MSKN 0x00000000
 /* lower bit position of bitfield msm register write data[31:0] */
-#define msm_reg_wr_data_shift 0
+#define HW_ATL_MSM_REG_WR_DATA_SHIFT 0
 /* width of bitfield msm register write data[31:0] */
-#define msm_reg_wr_data_width 32
+#define HW_ATL_MSM_REG_WR_DATA_WIDTH 32
 /* default value of bitfield msm register write data[31:0] */
-#define msm_reg_wr_data_default 0x0
+#define HW_ATL_MSM_REG_WR_DATA_DEFAULT 0x0
 
 /* mac_phy register write strobe bitfield definitions
  * preprocessor definitions for the bitfield "register write strobe".
@@ -2288,17 +2666,33 @@
  */
 
 /* register address for bitfield register write strobe */
-#define msm_reg_wr_strobe_adr 0x00004400
+#define HW_ATL_MSM_REG_WR_STROBE_ADR 0x00004400
 /* bitmask for bitfield register write strobe */
-#define msm_reg_wr_strobe_msk 0x00000100
+#define HW_ATL_MSM_REG_WR_STROBE_MSK 0x00000100
 /* inverted bitmask for bitfield register write strobe */
-#define msm_reg_wr_strobe_mskn 0xfffffeff
+#define HW_ATL_MSM_REG_WR_STROBE_MSKN 0xfffffeff
 /* lower bit position of bitfield register write strobe */
-#define msm_reg_wr_strobe_shift 8
+#define HW_ATL_MSM_REG_WR_STROBE_SHIFT 8
 /* width of bitfield register write strobe */
-#define msm_reg_wr_strobe_width 1
+#define HW_ATL_MSM_REG_WR_STROBE_WIDTH 1
 /* default value of bitfield register write strobe */
-#define msm_reg_wr_strobe_default 0x0
+#define HW_ATL_MSM_REG_WR_STROBE_DEFAULT 0x0
+
+/* register address for bitfield PTP Digital Clock Read Enable */
+#define HW_ATL_PCS_PTP_CLOCK_READ_ENABLE_ADR 0x00004628
+/* bitmask for bitfield PTP Digital Clock Read Enable */
+#define HW_ATL_PCS_PTP_CLOCK_READ_ENABLE_MSK 0x00000010
+/* inverted bitmask for bitfield PTP Digital Clock Read Enable */
+#define HW_ATL_PCS_PTP_CLOCK_READ_ENABLE_MSKN 0xFFFFFFEF
+/* lower bit position of bitfield PTP Digital Clock Read Enable */
+#define HW_ATL_PCS_PTP_CLOCK_READ_ENABLE_SHIFT 4
+/* width of bitfield PTP Digital Clock Read Enable */
+#define HW_ATL_PCS_PTP_CLOCK_READ_ENABLE_WIDTH 1
+/* default value of bitfield PTP Digital Clock Read Enable */
+#define HW_ATL_PCS_PTP_CLOCK_READ_ENABLE_DEFAULT 0x0
+
+/* register address for ptp counter reading */
+#define HW_ATL_PCS_PTP_TS_VAL_ADDR(index) (0x00004900 + (index) * 0x4)
 
 /* mif soft reset bitfield definitions
  * preprocessor definitions for the bitfield "soft reset".
@@ -2306,17 +2700,17 @@
  */
 
 /* register address for bitfield soft reset */
-#define glb_soft_res_adr 0x00000000
+#define HW_ATL_GLB_SOFT_RES_ADR 0x00000000
 /* bitmask for bitfield soft reset */
-#define glb_soft_res_msk 0x00008000
+#define HW_ATL_GLB_SOFT_RES_MSK 0x00008000
 /* inverted bitmask for bitfield soft reset */
-#define glb_soft_res_mskn 0xffff7fff
+#define HW_ATL_GLB_SOFT_RES_MSKN 0xffff7fff
 /* lower bit position of bitfield soft reset */
-#define glb_soft_res_shift 15
+#define HW_ATL_GLB_SOFT_RES_SHIFT 15
 /* width of bitfield soft reset */
-#define glb_soft_res_width 1
+#define HW_ATL_GLB_SOFT_RES_WIDTH 1
 /* default value of bitfield soft reset */
-#define glb_soft_res_default 0x0
+#define HW_ATL_GLB_SOFT_RES_DEFAULT 0x0
 
 /* mif register reset disable bitfield definitions
  * preprocessor definitions for the bitfield "register reset disable".
@@ -2324,32 +2718,35 @@
  */
 
 /* register address for bitfield register reset disable */
-#define glb_reg_res_dis_adr 0x00000000
+#define HW_ATL_GLB_REG_RES_DIS_ADR 0x00000000
 /* bitmask for bitfield register reset disable */
-#define glb_reg_res_dis_msk 0x00004000
+#define HW_ATL_GLB_REG_RES_DIS_MSK 0x00004000
 /* inverted bitmask for bitfield register reset disable */
-#define glb_reg_res_dis_mskn 0xffffbfff
+#define HW_ATL_GLB_REG_RES_DIS_MSKN 0xffffbfff
 /* lower bit position of bitfield register reset disable */
-#define glb_reg_res_dis_shift 14
+#define HW_ATL_GLB_REG_RES_DIS_SHIFT 14
 /* width of bitfield register reset disable */
-#define glb_reg_res_dis_width 1
+#define HW_ATL_GLB_REG_RES_DIS_WIDTH 1
 /* default value of bitfield register reset disable */
-#define glb_reg_res_dis_default 0x1
+#define HW_ATL_GLB_REG_RES_DIS_DEFAULT 0x1
 
 /* tx dma debug control definitions */
-#define tx_dma_debug_ctl_adr 0x00008920u
+#define HW_ATL_TX_DMA_DEBUG_CTL_ADR 0x00008920u
 
 /* tx dma descriptor base address msw definitions */
-#define tx_dma_desc_base_addrmsw_adr(descriptor) \
+#define HW_ATL_TX_DMA_DESC_BASE_ADDRMSW_ADR(descriptor) \
 			(0x00007c04u + (descriptor) * 0x40)
 
+/* tx dma total request limit */
+#define HW_ATL_TX_DMA_TOTAL_REQ_LIMIT_ADR 0x00007b20u
+
 /* tx interrupt moderation control register definitions
  * Preprocessor definitions for TX Interrupt Moderation Control Register
  * Base Address: 0x00008980
  * Parameter: queue {Q} | stride size 0x4 | range [0, 31]
  */
 
-#define tx_intr_moderation_ctl_adr(queue) (0x00008980u + (queue) * 0x4)
+#define HW_ATL_TX_INTR_MODERATION_CTL_ADR(queue) (0x00008980u + (queue) * 0x4)
 
 /* pcie reg_res_dsbl bitfield definitions
  * preprocessor definitions for the bitfield "reg_res_dsbl".
@@ -2357,19 +2754,157 @@
  */
 
 /* register address for bitfield reg_res_dsbl */
-#define pci_reg_res_dsbl_adr 0x00001000
+#define HW_ATL_PCI_REG_RES_DSBL_ADR 0x00001000
 /* bitmask for bitfield reg_res_dsbl */
-#define pci_reg_res_dsbl_msk 0x20000000
+#define HW_ATL_PCI_REG_RES_DSBL_MSK 0x20000000
 /* inverted bitmask for bitfield reg_res_dsbl */
-#define pci_reg_res_dsbl_mskn 0xdfffffff
+#define HW_ATL_PCI_REG_RES_DSBL_MSKN 0xdfffffff
 /* lower bit position of bitfield reg_res_dsbl */
-#define pci_reg_res_dsbl_shift 29
+#define HW_ATL_PCI_REG_RES_DSBL_SHIFT 29
 /* width of bitfield reg_res_dsbl */
-#define pci_reg_res_dsbl_width 1
+#define HW_ATL_PCI_REG_RES_DSBL_WIDTH 1
 /* default value of bitfield reg_res_dsbl */
-#define pci_reg_res_dsbl_default 0x1
+#define HW_ATL_PCI_REG_RES_DSBL_DEFAULT 0x1
+
+/* PCI core control register */
+#define HW_ATL_PCI_REG_CONTROL6_ADR 0x1014u
 
 /* global microprocessor scratch pad definitions */
-#define glb_cpu_scratch_scp_adr(scratch_scp) (0x00000300u + (scratch_scp) * 0x4)
+#define HW_ATL_GLB_CPU_SCRATCH_SCP_ADR(scratch_scp) \
+	(0x00000300u + (scratch_scp) * 0x4)
+
+/* register address for bitfield uP Force Interrupt */
+#define HW_ATL_MCP_UP_FORCE_INTERRUPT_ADR 0x00000404
+/* bitmask for bitfield uP Force Interrupt */
+#define HW_ATL_MCP_UP_FORCE_INTERRUPT_MSK 0x00000002
+/* inverted bitmask for bitfield uP Force Interrupt */
+#define HW_ATL_MCP_UP_FORCE_INTERRUPT_MSKN 0xFFFFFFFD
+/* lower bit position of bitfield uP Force Interrupt */
+#define HW_ATL_MCP_UP_FORCE_INTERRUPT_SHIFT 1
+/* width of bitfield uP Force Interrupt */
+#define HW_ATL_MCP_UP_FORCE_INTERRUPT_WIDTH 1
+/* default value of bitfield uP Force Interrupt */
+#define HW_ATL_MCP_UP_FORCE_INTERRUPT_DEFAULT 0x0
+
+/* Preprocessor definitions for Global MDIO Interfaces
+ * Address: 0x00000280 + 0x4 * Number of interface
+ */
+#define HW_ATL_GLB_MDIO_IFACE_ADDR_BEGIN   0x00000280u
+
+#define HW_ATL_GLB_MDIO_IFACE_N_ADR(number) \
+	(HW_ATL_GLB_MDIO_IFACE_ADDR_BEGIN + (((number) - 1) * 0x4))
+
+/* MIF MDIO Busy Bitfield Definitions
+ * Preprocessor definitions for the bitfield "MDIO Busy".
+ * PORT="mdio_pif_busy_o"
+ */
+
+/* Register address for bitfield MDIO Busy */
+#define HW_ATL_MDIO_BUSY_ADR 0x00000284
+/* Bitmask for bitfield MDIO Busy */
+#define HW_ATL_MDIO_BUSY_MSK 0x80000000
+/* Inverted bitmask for bitfield MDIO Busy */
+#define HW_ATL_MDIO_BUSY_MSKN 0x7FFFFFFF
+/* Lower bit position of bitfield MDIO Busy */
+#define HW_ATL_MDIO_BUSY_SHIFT 31
+/* Width of bitfield MDIO Busy */
+#define HW_ATL_MDIO_BUSY_WIDTH 1
+
+/* MIF MDIO Execute Operation Bitfield Definitions
+ * Preprocessor definitions for the bitfield "MDIO Execute Operation".
+ * PORT="pif_mdio_op_start_i"
+ */
+
+/* Register address for bitfield MDIO Execute Operation */
+#define HW_ATL_MDIO_EXECUTE_OPERATION_ADR 0x00000284
+/* Bitmask for bitfield MDIO Execute Operation */
+#define HW_ATL_MDIO_EXECUTE_OPERATION_MSK 0x00008000
+/* Inverted bitmask for bitfield MDIO Execute Operation */
+#define HW_ATL_MDIO_EXECUTE_OPERATION_MSKN 0xFFFF7FFF
+/* Lower bit position of bitfield MDIO Execute Operation */
+#define HW_ATL_MDIO_EXECUTE_OPERATION_SHIFT 15
+/* Width of bitfield MDIO Execute Operation */
+#define HW_ATL_MDIO_EXECUTE_OPERATION_WIDTH 1
+/* Default value of bitfield MDIO Execute Operation */
+#define HW_ATL_MDIO_EXECUTE_OPERATION_DEFAULT 0x0
+
+/* MIF Op Mode [1:0] Bitfield Definitions
+ * Preprocessor definitions for the bitfield "Op Mode [1:0]".
+ * PORT="pif_mdio_mode_i[1:0]"
+ */
+
+/* Register address for bitfield Op Mode [1:0] */
+#define HW_ATL_MDIO_OP_MODE_ADR 0x00000284
+/* Bitmask for bitfield Op Mode [1:0] */
+#define HW_ATL_MDIO_OP_MODE_MSK 0x00003000
+/* Inverted bitmask for bitfield Op Mode [1:0] */
+#define HW_ATL_MDIO_OP_MODE_MSKN 0xFFFFCFFF
+/* Lower bit position of bitfield Op Mode [1:0] */
+#define HW_ATL_MDIO_OP_MODE_SHIFT 12
+/* Width of bitfield Op Mode [1:0] */
+#define HW_ATL_MDIO_OP_MODE_WIDTH 2
+/* Default value of bitfield Op Mode [1:0] */
+#define HW_ATL_MDIO_OP_MODE_DEFAULT 0x0
+
+/* MIF PHY address Bitfield Definitions
+ * Preprocessor definitions for the bitfield "PHY address".
+ * PORT="pif_mdio_phy_addr_i[9:0]"
+ */
+
+/* Register address for bitfield PHY address */
+#define HW_ATL_MDIO_PHY_ADDRESS_ADR 0x00000284
+/* Bitmask for bitfield PHY address */
+#define HW_ATL_MDIO_PHY_ADDRESS_MSK 0x000003FF
+/* Inverted bitmask for bitfield PHY address */
+#define HW_ATL_MDIO_PHY_ADDRESS_MSKN 0xFFFFFC00
+/* Lower bit position of bitfield PHY address */
+#define HW_ATL_MDIO_PHY_ADDRESS_SHIFT 0
+/* Width of bitfield PHY address */
+#define HW_ATL_MDIO_PHY_ADDRESS_WIDTH 10
+/* Default value of bitfield PHY address */
+#define HW_ATL_MDIO_PHY_ADDRESS_DEFAULT 0x0
+
+/* MIF MDIO WriteData [F:0] Bitfield Definitions
+ * Preprocessor definitions for the bitfield "MDIO WriteData [F:0]".
+ * PORT="pif_mdio_wdata_i[15:0]"
+ */
+
+/* Register address for bitfield MDIO WriteData [F:0] */
+#define HW_ATL_MDIO_WRITE_DATA_ADR 0x00000288
+/* Bitmask for bitfield MDIO WriteData [F:0] */
+#define HW_ATL_MDIO_WRITE_DATA_MSK 0x0000FFFF
+/* Inverted bitmask for bitfield MDIO WriteData [F:0] */
+#define HW_ATL_MDIO_WRITE_DATA_MSKN 0xFFFF0000
+/* Lower bit position of bitfield MDIO WriteData [F:0] */
+#define HW_ATL_MDIO_WRITE_DATA_SHIFT 0
+/* Width of bitfield MDIO WriteData [F:0] */
+#define HW_ATL_MDIO_WRITE_DATA_WIDTH 16
+/* Default value of bitfield MDIO WriteData [F:0] */
+#define HW_ATL_MDIO_WRITE_DATA_DEFAULT 0x0
+
+/* MIF MDIO Address [F:0] Bitfield Definitions
+ * Preprocessor definitions for the bitfield "MDIO Address [F:0]".
+ * PORT="pif_mdio_addr_i[15:0]"
+ */
+
+/* Register address for bitfield MDIO Address [F:0] */
+#define HW_ATL_MDIO_ADDRESS_ADR 0x0000028C
+/* Bitmask for bitfield MDIO Address [F:0] */
+#define HW_ATL_MDIO_ADDRESS_MSK 0x0000FFFF
+/* Inverted bitmask for bitfield MDIO Address [F:0] */
+#define HW_ATL_MDIO_ADDRESS_MSKN 0xFFFF0000
+/* Lower bit position of bitfield MDIO Address [F:0] */
+#define HW_ATL_MDIO_ADDRESS_SHIFT 0
+/* Width of bitfield MDIO Address [F:0] */
+#define HW_ATL_MDIO_ADDRESS_WIDTH 16
+/* Default value of bitfield MDIO Address [F:0] */
+#define HW_ATL_MDIO_ADDRESS_DEFAULT 0x0
+
+#define HW_ATL_MIF_RESET_TIMEOUT_ADR 0x00000348
+
+#define HW_ATL_FW_SM_MDIO       0x0U
+#define HW_ATL_FW_SM_RAM        0x2U
+#define HW_ATL_FW_SM_RESET1     0x3U
+#define HW_ATL_FW_SM_RESET2     0x4U
 
 #endif /* HW_ATL_LLH_INTERNAL_H */
diff --git a/drivers/net/ethernet/aquantia/atlantic/hw_atl/hw_atl_utils.c b/drivers/net/ethernet/aquantia/atlantic/hw_atl/hw_atl_utils.c
index 8d6d8f5804da..7e88d7234b14 100644
--- a/drivers/net/ethernet/aquantia/atlantic/hw_atl/hw_atl_utils.c
+++ b/drivers/net/ethernet/aquantia/atlantic/hw_atl/hw_atl_utils.c
@@ -1,125 +1,477 @@
-/*
- * aQuantia Corporation Network Driver
- * Copyright (C) 2014-2017 aQuantia Corporation. All rights reserved
+// SPDX-License-Identifier: GPL-2.0-only
+/* Atlantic Network Driver
  *
- * This program is free software; you can redistribute it and/or modify it
- * under the terms and conditions of the GNU General Public License,
- * version 2, as published by the Free Software Foundation.
+ * Copyright (C) 2014-2019 aQuantia Corporation
+ * Copyright (C) 2019-2020 Marvell International Ltd.
  */
 
 /* File hw_atl_utils.c: Definition of common functions for Atlantic hardware
  * abstraction layer.
  */
 
-#include "../aq_hw.h"
+#include "../aq_nic.h"
 #include "../aq_hw_utils.h"
-#include "../aq_pci_func.h"
-#include "../aq_ring.h"
-#include "../aq_vec.h"
 #include "hw_atl_utils.h"
 #include "hw_atl_llh.h"
+#include "hw_atl_llh_internal.h"
 
 #include <linux/random.h>
 
 #define HW_ATL_UCP_0X370_REG    0x0370U
 
-#define HW_ATL_FW_SM_RAM        0x2U
+#define HW_ATL_MIF_CMD          0x0200U
+#define HW_ATL_MIF_ADDR         0x0208U
+#define HW_ATL_MIF_VAL          0x020CU
+
+#define HW_ATL_MPI_RPC_ADDR     0x0334U
+#define HW_ATL_RPC_CONTROL_ADR  0x0338U
+#define HW_ATL_RPC_STATE_ADR    0x033CU
+
+#define HW_ATL_MPI_FW_VERSION	0x18
 #define HW_ATL_MPI_CONTROL_ADR  0x0368U
 #define HW_ATL_MPI_STATE_ADR    0x036CU
 
-#define HW_ATL_MPI_STATE_MSK    0x00FFU
-#define HW_ATL_MPI_STATE_SHIFT  0U
-#define HW_ATL_MPI_SPEED_MSK    0xFFFFU
-#define HW_ATL_MPI_SPEED_SHIFT  16U
+#define HW_ATL_MPI_STATE_MSK      0x00FFU
+#define HW_ATL_MPI_STATE_SHIFT    0U
+#define HW_ATL_MPI_SPEED_MSK      0x00FF0000U
+#define HW_ATL_MPI_SPEED_SHIFT    16U
+#define HW_ATL_MPI_DIRTY_WAKE_MSK 0x02000000U
+
+#define HW_ATL_MPI_DAISY_CHAIN_STATUS	0x704
+#define HW_ATL_MPI_BOOT_EXIT_CODE	0x388
+
+#define HW_ATL_MAC_PHY_CONTROL	0x4000
+#define HW_ATL_MAC_PHY_MPI_RESET_BIT 0x1D
+
+#define HW_ATL_FW_VER_1X 0x01050006U
+#define HW_ATL_FW_VER_2X 0x02000000U
+#define HW_ATL_FW_VER_3X 0x03000000U
+#define HW_ATL_FW_VER_4X 0x04000000U
+
+#define FORCE_FLASHLESS 0
+
+enum mcp_area {
+	MCP_AREA_CONFIG = 0x80000000,
+	MCP_AREA_SETTINGS = 0x20000000,
+};
+
+static int hw_atl_utils_mpi_set_state(struct aq_hw_s *self,
+				      enum hal_atl_utils_fw_state_e state);
+static u32 hw_atl_utils_get_mpi_mbox_tid(struct aq_hw_s *self);
+static u32 hw_atl_utils_mpi_get_state(struct aq_hw_s *self);
+static u32 hw_atl_utils_mif_cmd_get(struct aq_hw_s *self);
+static u32 hw_atl_utils_mif_addr_get(struct aq_hw_s *self);
+static u32 hw_atl_utils_rpc_state_get(struct aq_hw_s *self);
+static u32 aq_fw1x_rpc_get(struct aq_hw_s *self);
+
+int hw_atl_utils_initfw(struct aq_hw_s *self, const struct aq_fw_ops **fw_ops)
+{
+	int err = 0;
+
+	hw_atl_utils_hw_chip_features_init(self,
+					   &self->chip_features);
+
+	self->fw_ver_actual = hw_atl_utils_get_fw_version(self);
+
+	if (hw_atl_utils_ver_match(HW_ATL_FW_VER_1X, self->fw_ver_actual)) {
+		*fw_ops = &aq_fw_1x_ops;
+	} else if (hw_atl_utils_ver_match(HW_ATL_FW_VER_2X, self->fw_ver_actual)) {
+		*fw_ops = &aq_fw_2x_ops;
+	} else if (hw_atl_utils_ver_match(HW_ATL_FW_VER_3X, self->fw_ver_actual)) {
+		*fw_ops = &aq_fw_2x_ops;
+	} else if (hw_atl_utils_ver_match(HW_ATL_FW_VER_4X, self->fw_ver_actual)) {
+		*fw_ops = &aq_fw_2x_ops;
+	} else {
+		aq_pr_err("Bad FW version detected: %x\n",
+			  self->fw_ver_actual);
+		return -EOPNOTSUPP;
+	}
+	self->aq_fw_ops = *fw_ops;
+	err = self->aq_fw_ops->init(self);
+
+	return err;
+}
+
+static int hw_atl_utils_soft_reset_flb(struct aq_hw_s *self)
+{
+	u32 gsr, val;
+	int k = 0;
+
+	aq_hw_write_reg(self, 0x404, 0x40e1);
+	AQ_HW_SLEEP(50);
+
+	/* Cleanup SPI */
+	val = aq_hw_read_reg(self, 0x53C);
+	aq_hw_write_reg(self, 0x53C, val | 0x10);
+
+	gsr = aq_hw_read_reg(self, HW_ATL_GLB_SOFT_RES_ADR);
+	aq_hw_write_reg(self, HW_ATL_GLB_SOFT_RES_ADR, (gsr & 0xBFFF) | 0x8000);
+
+	/* Kickstart MAC */
+	aq_hw_write_reg(self, 0x404, 0x80e0);
+	aq_hw_write_reg(self, 0x32a8, 0x0);
+	aq_hw_write_reg(self, 0x520, 0x1);
+
+	/* Reset SPI again because of possible interrupted SPI burst */
+	val = aq_hw_read_reg(self, 0x53C);
+	aq_hw_write_reg(self, 0x53C, val | 0x10);
+	AQ_HW_SLEEP(10);
+	/* Clear SPI reset state */
+	aq_hw_write_reg(self, 0x53C, val & ~0x10);
+
+	aq_hw_write_reg(self, 0x404, 0x180e0);
+
+	for (k = 0; k < 1000; k++) {
+		u32 flb_status = aq_hw_read_reg(self,
+						HW_ATL_MPI_DAISY_CHAIN_STATUS);
+
+		flb_status = flb_status & 0x10;
+		if (flb_status)
+			break;
+		AQ_HW_SLEEP(10);
+	}
+	if (k == 1000) {
+		aq_pr_err("MAC kickstart failed\n");
+		return -EIO;
+	}
+
+	/* FW reset */
+	aq_hw_write_reg(self, 0x404, 0x80e0);
+	AQ_HW_SLEEP(50);
+	aq_hw_write_reg(self, 0x3a0, 0x1);
+
+	/* Kickstart PHY - skipped */
+
+	/* Global software reset*/
+	hw_atl_rx_rx_reg_res_dis_set(self, 0U);
+	hw_atl_tx_tx_reg_res_dis_set(self, 0U);
+	aq_hw_write_reg_bit(self, HW_ATL_MAC_PHY_CONTROL,
+			    BIT(HW_ATL_MAC_PHY_MPI_RESET_BIT),
+			    HW_ATL_MAC_PHY_MPI_RESET_BIT, 0x0);
+	gsr = aq_hw_read_reg(self, HW_ATL_GLB_SOFT_RES_ADR);
+	aq_hw_write_reg(self, HW_ATL_GLB_SOFT_RES_ADR, (gsr & 0xBFFF) | 0x8000);
+
+	for (k = 0; k < 1000; k++) {
+		u32 fw_state = aq_hw_read_reg(self, HW_ATL_MPI_FW_VERSION);
+
+		if (fw_state)
+			break;
+		AQ_HW_SLEEP(10);
+	}
+	if (k == 1000) {
+		aq_pr_err("FW kickstart failed\n");
+		return -EIO;
+	}
+	/* Old FW requires fixed delay after init */
+	AQ_HW_SLEEP(15);
+
+	return 0;
+}
+
+static int hw_atl_utils_soft_reset_rbl(struct aq_hw_s *self)
+{
+	u32 gsr, val, rbl_status;
+	int k;
+
+	aq_hw_write_reg(self, 0x404, 0x40e1);
+	aq_hw_write_reg(self, 0x3a0, 0x1);
+	aq_hw_write_reg(self, 0x32a8, 0x0);
+
+	/* Alter RBL status */
+	aq_hw_write_reg(self, 0x388, 0xDEAD);
+
+	/* Cleanup SPI */
+	val = aq_hw_read_reg(self, 0x53C);
+	aq_hw_write_reg(self, 0x53C, val | 0x10);
+
+	/* Global software reset*/
+	hw_atl_rx_rx_reg_res_dis_set(self, 0U);
+	hw_atl_tx_tx_reg_res_dis_set(self, 0U);
+	aq_hw_write_reg_bit(self, HW_ATL_MAC_PHY_CONTROL,
+			    BIT(HW_ATL_MAC_PHY_MPI_RESET_BIT),
+			    HW_ATL_MAC_PHY_MPI_RESET_BIT, 0x0);
+	gsr = aq_hw_read_reg(self, HW_ATL_GLB_SOFT_RES_ADR);
+	aq_hw_write_reg(self, HW_ATL_GLB_SOFT_RES_ADR,
+			(gsr & 0xFFFFBFFF) | 0x8000);
+
+	if (FORCE_FLASHLESS)
+		aq_hw_write_reg(self, 0x534, 0x0);
+
+	aq_hw_write_reg(self, 0x404, 0x40e0);
+
+	/* Wait for RBL boot */
+	for (k = 0; k < 1000; k++) {
+		rbl_status = aq_hw_read_reg(self, 0x388) & 0xFFFF;
+		if (rbl_status && rbl_status != 0xDEAD)
+			break;
+		AQ_HW_SLEEP(10);
+	}
+	if (!rbl_status || rbl_status == 0xDEAD) {
+		aq_pr_err("RBL Restart failed");
+		return -EIO;
+	}
+
+	/* Restore NVR */
+	if (FORCE_FLASHLESS)
+		aq_hw_write_reg(self, 0x534, 0xA0);
+
+	if (rbl_status == 0xF1A7) {
+		aq_pr_err("No FW detected. Dynamic FW load not implemented\n");
+		return -EOPNOTSUPP;
+	}
+
+	for (k = 0; k < 1000; k++) {
+		u32 fw_state = aq_hw_read_reg(self, HW_ATL_MPI_FW_VERSION);
+
+		if (fw_state)
+			break;
+		AQ_HW_SLEEP(10);
+	}
+	if (k == 1000) {
+		aq_pr_err("FW kickstart failed\n");
+		return -EIO;
+	}
+	/* Old FW requires fixed delay after init */
+	AQ_HW_SLEEP(15);
+
+	return 0;
+}
 
-static int hw_atl_utils_fw_downld_dwords(struct aq_hw_s *self, u32 a,
-					 u32 *p, u32 cnt)
+int hw_atl_utils_soft_reset(struct aq_hw_s *self)
+{
+	int ver = hw_atl_utils_get_fw_version(self);
+	u32 boot_exit_code = 0;
+	u32 val;
+	int k;
+
+	for (k = 0; k < 1000; ++k) {
+		u32 flb_status = aq_hw_read_reg(self,
+						HW_ATL_MPI_DAISY_CHAIN_STATUS);
+		boot_exit_code = aq_hw_read_reg(self,
+						HW_ATL_MPI_BOOT_EXIT_CODE);
+		if (flb_status != 0x06000000 || boot_exit_code != 0)
+			break;
+	}
+
+	if (k == 1000) {
+		aq_pr_err("Neither RBL nor FLB firmware started\n");
+		return -EOPNOTSUPP;
+	}
+
+	self->rbl_enabled = (boot_exit_code != 0);
+
+	if (hw_atl_utils_ver_match(HW_ATL_FW_VER_1X, ver)) {
+		int err = 0;
+
+		/* FW 1.x may bootup in an invalid POWER state (WOL feature).
+		 * We should work around this by forcing its state back to DEINIT
+		 */
+		hw_atl_utils_mpi_set_state(self, MPI_DEINIT);
+		err = readx_poll_timeout_atomic(hw_atl_utils_mpi_get_state,
+						self, val,
+						(val & HW_ATL_MPI_STATE_MSK) ==
+						 MPI_DEINIT,
+						10, 10000U);
+		if (err)
+			return err;
+	} else if (hw_atl_utils_ver_match(HW_ATL_FW_VER_4X, ver)) {
+		u64 sem_timeout = aq_hw_read_reg(self, HW_ATL_MIF_RESET_TIMEOUT_ADR);
+
+		/* Acquire 2 semaphores before issuing reset for FW 4.x */
+		if (sem_timeout > 3000)
+			sem_timeout = 3000;
+		sem_timeout = sem_timeout * 1000;
+
+		if (sem_timeout != 0) {
+			int err;
+
+			err = readx_poll_timeout_atomic(hw_atl_sem_reset1_get, self, val,
+							val == 1U, 1U, sem_timeout);
+			if (err)
+				aq_pr_err("reset sema1 timeout");
+
+			err = readx_poll_timeout_atomic(hw_atl_sem_reset2_get, self, val,
+							val == 1U, 1U, sem_timeout);
+			if (err)
+				aq_pr_err("reset sema2 timeout");
+		}
+	}
+
+	if (self->rbl_enabled)
+		return hw_atl_utils_soft_reset_rbl(self);
+	else
+		return hw_atl_utils_soft_reset_flb(self);
+}
+
+int hw_atl_utils_fw_downld_dwords(struct aq_hw_s *self, u32 a,
+				  u32 *p, u32 cnt)
 {
 	int err = 0;
+	u32 val;
 
-	AQ_HW_WAIT_FOR(reg_glb_cpu_sem_get(self,
-					   HW_ATL_FW_SM_RAM) == 1U,
-					   1U, 10000U);
+	err = readx_poll_timeout_atomic(hw_atl_sem_ram_get,
+					self, val, val == 1U,
+					1U, 10000U);
 
 	if (err < 0) {
 		bool is_locked;
 
-		reg_glb_cpu_sem_set(self, 1U, HW_ATL_FW_SM_RAM);
-		is_locked = reg_glb_cpu_sem_get(self, HW_ATL_FW_SM_RAM);
+		hw_atl_reg_glb_cpu_sem_set(self, 1U, HW_ATL_FW_SM_RAM);
+		is_locked = hw_atl_sem_ram_get(self);
 		if (!is_locked) {
 			err = -ETIME;
 			goto err_exit;
 		}
 	}
 
-	aq_hw_write_reg(self, 0x00000208U, a);
-
-	for (++cnt; --cnt;) {
-		u32 i = 0U;
+	aq_hw_write_reg(self, HW_ATL_MIF_ADDR, a);
 
-		aq_hw_write_reg(self, 0x00000200U, 0x00008000U);
+	for (++cnt; --cnt && !err;) {
+		aq_hw_write_reg(self, HW_ATL_MIF_CMD, 0x00008000U);
 
-		for (i = 1024U;
-			(0x100U & aq_hw_read_reg(self, 0x00000200U)) && --i;) {
-		}
+		if (ATL_HW_IS_CHIP_FEATURE(self, REVISION_B1))
+			err = readx_poll_timeout_atomic(hw_atl_utils_mif_addr_get,
+							self, val, val != a,
+							1U, 1000U);
+		else
+			err = readx_poll_timeout_atomic(hw_atl_utils_mif_cmd_get,
+							self, val,
+							!(val & 0x100),
+							1U, 1000U);
 
-		*(p++) = aq_hw_read_reg(self, 0x0000020CU);
+		*(p++) = aq_hw_read_reg(self, HW_ATL_MIF_VAL);
+		a += 4;
 	}
 
-	reg_glb_cpu_sem_set(self, 1U, HW_ATL_FW_SM_RAM);
+	hw_atl_reg_glb_cpu_sem_set(self, 1U, HW_ATL_FW_SM_RAM);
 
 err_exit:
 	return err;
 }
 
-static int hw_atl_utils_fw_upload_dwords(struct aq_hw_s *self, u32 a, u32 *p,
-					 u32 cnt)
+static int hw_atl_utils_write_b1_mbox(struct aq_hw_s *self, u32 addr,
+				      u32 *p, u32 cnt, enum mcp_area area)
 {
+	u32 data_offset = 0;
+	u32 offset = addr;
 	int err = 0;
-	bool is_locked;
+	u32 val;
 
-	is_locked = reg_glb_cpu_sem_get(self, HW_ATL_FW_SM_RAM);
-	if (!is_locked) {
-		err = -ETIME;
-		goto err_exit;
+	switch (area) {
+	case MCP_AREA_CONFIG:
+		offset -= self->rpc_addr;
+		break;
+
+	case MCP_AREA_SETTINGS:
+		offset -= self->settings_addr;
+		break;
 	}
 
-	aq_hw_write_reg(self, 0x00000208U, a);
+	offset = offset / sizeof(u32);
+
+	for (; data_offset < cnt; ++data_offset, ++offset) {
+		aq_hw_write_reg(self, 0x328, p[data_offset]);
+		aq_hw_write_reg(self, 0x32C,
+				(area | (0xFFFF & (offset * 4))));
+		hw_atl_mcp_up_force_intr_set(self, 1);
+		/* 1000 times by 10us = 10ms */
+		err = readx_poll_timeout_atomic(hw_atl_scrpad12_get,
+						self, val,
+						(val & 0xF0000000) !=
+						area,
+						10U, 10000U);
+
+		if (err < 0)
+			break;
+	}
 
-	for (++cnt; --cnt;) {
-		u32 i = 0U;
+	return err;
+}
 
-		aq_hw_write_reg(self, 0x0000020CU, *(p++));
-		aq_hw_write_reg(self, 0x00000200U, 0xC000U);
+static int hw_atl_utils_write_b0_mbox(struct aq_hw_s *self, u32 addr,
+				      u32 *p, u32 cnt)
+{
+	u32 offset = 0;
+	int err = 0;
+	u32 val;
 
-		for (i = 1024U;
-			(0x100U & aq_hw_read_reg(self, 0x00000200U)) && --i;) {
-		}
+	aq_hw_write_reg(self, 0x208, addr);
+
+	for (; offset < cnt; ++offset) {
+		aq_hw_write_reg(self, 0x20C, p[offset]);
+		aq_hw_write_reg(self, 0x200, 0xC000);
+
+		err = readx_poll_timeout_atomic(hw_atl_utils_mif_cmd_get,
+						self, val,
+						(val & 0x100) == 0U,
+						10U, 10000U);
+
+		if (err < 0)
+			break;
 	}
 
-	reg_glb_cpu_sem_set(self, 1U, HW_ATL_FW_SM_RAM);
+	return err;
+}
+
+static int hw_atl_utils_fw_upload_dwords(struct aq_hw_s *self, u32 addr, u32 *p,
+					 u32 cnt, enum mcp_area area)
+{
+	int err = 0;
+	u32 val;
+
+	err = readx_poll_timeout_atomic(hw_atl_sem_ram_get, self,
+					val, val == 1U,
+					10U, 100000U);
+	if (err < 0)
+		goto err_exit;
+
+	if (ATL_HW_IS_CHIP_FEATURE(self, REVISION_B1))
+		err = hw_atl_utils_write_b1_mbox(self, addr, p, cnt, area);
+	else
+		err = hw_atl_utils_write_b0_mbox(self, addr, p, cnt);
+
+	hw_atl_reg_glb_cpu_sem_set(self, 1U, HW_ATL_FW_SM_RAM);
+
+	if (err < 0)
+		goto err_exit;
+
+	err = aq_hw_err_from_flags(self);
 
 err_exit:
 	return err;
 }
 
-static int hw_atl_utils_ver_match(u32 ver_expected, u32 ver_actual)
+int hw_atl_write_fwcfg_dwords(struct aq_hw_s *self, u32 *p, u32 cnt)
+{
+	return hw_atl_utils_fw_upload_dwords(self, self->rpc_addr, p,
+					     cnt, MCP_AREA_CONFIG);
+}
+
+int hw_atl_write_fwsettings_dwords(struct aq_hw_s *self, u32 offset, u32 *p,
+				   u32 cnt)
+{
+	return hw_atl_utils_fw_upload_dwords(self, self->settings_addr + offset,
+					     p, cnt, MCP_AREA_SETTINGS);
+}
+
+bool hw_atl_utils_ver_match(u32 ver_expected, u32 ver_actual)
 {
-	int err = 0;
 	const u32 dw_major_mask = 0xff000000U;
 	const u32 dw_minor_mask = 0x00ffffffU;
+	bool ver_match;
 
-	err = (dw_major_mask & (ver_expected ^ ver_actual)) ? -EOPNOTSUPP : 0;
-	if (err < 0)
+	ver_match = (dw_major_mask & (ver_expected ^ ver_actual)) ? false : true;
+	if (!ver_match)
 		goto err_exit;
-	err = ((dw_minor_mask & ver_expected) > (dw_minor_mask & ver_actual)) ?
-		-EOPNOTSUPP : 0;
+	ver_match = ((dw_minor_mask & ver_expected) > (dw_minor_mask & ver_actual)) ?
+		false : true;
+
 err_exit:
-	return err;
+	return ver_match;
 }
 
 static int hw_atl_utils_init_ucp(struct aq_hw_s *self,
-				 struct aq_hw_caps_s *aq_hw_caps)
+				 const struct aq_hw_caps_s *aq_hw_caps)
 {
 	int err = 0;
 
@@ -133,20 +485,21 @@ static int hw_atl_utils_init_ucp(struct aq_hw_s *self,
 		aq_hw_write_reg(self, HW_ATL_UCP_0X370_REG, ucp_0x370);
 	}
 
-	reg_glb_cpu_scratch_scp_set(self, 0x00000000U, 25U);
+	hw_atl_reg_glb_cpu_scratch_scp_set(self, 0x00000000U, 25U);
 
 	/* check 10 times by 1ms */
-	AQ_HW_WAIT_FOR(0U != (PHAL_ATLANTIC_A0->mbox_addr =
-			aq_hw_read_reg(self, 0x360U)), 1000U, 10U);
+	err = readx_poll_timeout_atomic(hw_atl_scrpad25_get,
+					self, self->mbox_addr,
+					self->mbox_addr != 0U,
+					1000U, 10000U);
+	err = readx_poll_timeout_atomic(aq_fw1x_rpc_get, self,
+					self->rpc_addr,
+					self->rpc_addr != 0U,
+					1000U, 100000U);
 
-	err = hw_atl_utils_ver_match(aq_hw_caps->fw_ver_expected,
-				     aq_hw_read_reg(self, 0x18U));
 	return err;
 }
 
-#define HW_ATL_RPC_CONTROL_ADR 0x0338U
-#define HW_ATL_RPC_STATE_ADR   0x033CU
-
 struct aq_hw_atl_utils_fw_rpc_tid_s {
 	union {
 		u32 val;
@@ -159,23 +512,22 @@ struct aq_hw_atl_utils_fw_rpc_tid_s {
 
 #define hw_atl_utils_fw_rpc_init(_H_) hw_atl_utils_fw_rpc_wait(_H_, NULL)
 
-static int hw_atl_utils_fw_rpc_call(struct aq_hw_s *self, unsigned int rpc_size)
+int hw_atl_utils_fw_rpc_call(struct aq_hw_s *self, unsigned int rpc_size)
 {
-	int err = 0;
 	struct aq_hw_atl_utils_fw_rpc_tid_s sw;
+	int err = 0;
 
-	if (!IS_CHIP_FEATURE(MIPS)) {
+	if (!ATL_HW_IS_CHIP_FEATURE(self, MIPS)) {
 		err = -1;
 		goto err_exit;
 	}
-	err = hw_atl_utils_fw_upload_dwords(self, PHAL_ATLANTIC->rpc_addr,
-					    (u32 *)(void *)&PHAL_ATLANTIC->rpc,
-					    (rpc_size + sizeof(u32) -
-					    sizeof(u8)) / sizeof(u32));
+	err = hw_atl_write_fwcfg_dwords(self, (u32 *)(void *)&self->rpc,
+					(rpc_size + sizeof(u32) -
+					 sizeof(u8)) / sizeof(u32));
 	if (err < 0)
 		goto err_exit;
 
-	sw.tid = 0xFFFFU & (++PHAL_ATLANTIC->rpc_tid);
+	sw.tid = 0xFFFFU & (++self->rpc_tid);
 	sw.len = (u16)rpc_size;
 	aq_hw_write_reg(self, HW_ATL_RPC_CONTROL_ADR, sw.val);
 
@@ -183,61 +535,72 @@ err_exit:
 	return err;
 }
 
-static int hw_atl_utils_fw_rpc_wait(struct aq_hw_s *self,
-				    struct hw_aq_atl_utils_fw_rpc **rpc)
+int hw_atl_utils_fw_rpc_wait(struct aq_hw_s *self,
+			     struct hw_atl_utils_fw_rpc **rpc)
 {
-	int err = 0;
 	struct aq_hw_atl_utils_fw_rpc_tid_s sw;
 	struct aq_hw_atl_utils_fw_rpc_tid_s fw;
+	int err = 0;
 
 	do {
 		sw.val = aq_hw_read_reg(self, HW_ATL_RPC_CONTROL_ADR);
 
-		PHAL_ATLANTIC->rpc_tid = sw.tid;
+		self->rpc_tid = sw.tid;
 
-		AQ_HW_WAIT_FOR(sw.tid ==
-				(fw.val =
-				aq_hw_read_reg(self, HW_ATL_RPC_STATE_ADR),
-				fw.tid), 1000U, 100U);
+		err = readx_poll_timeout_atomic(hw_atl_utils_rpc_state_get,
+						self, fw.val,
+						sw.tid == fw.tid,
+						1000U, 100000U);
+		if (err < 0)
+			goto err_exit;
+
+		err = aq_hw_err_from_flags(self);
 		if (err < 0)
 			goto err_exit;
 
 		if (fw.len == 0xFFFFU) {
+			if (sw.len > sizeof(self->rpc)) {
+				printk(KERN_INFO "Invalid sw len: %x\n", sw.len);
+				err = -EINVAL;
+				goto err_exit;
+			}
 			err = hw_atl_utils_fw_rpc_call(self, sw.len);
 			if (err < 0)
 				goto err_exit;
 		}
 	} while (sw.tid != fw.tid || 0xFFFFU == fw.len);
-	if (err < 0)
-		goto err_exit;
 
 	if (rpc) {
 		if (fw.len) {
+			if (fw.len > sizeof(self->rpc)) {
+				printk(KERN_INFO "Invalid fw len: %x\n", fw.len);
+				err = -EINVAL;
+				goto err_exit;
+			}
 			err =
 			hw_atl_utils_fw_downld_dwords(self,
-						      PHAL_ATLANTIC->rpc_addr,
+						      self->rpc_addr,
 						      (u32 *)(void *)
-						      &PHAL_ATLANTIC->rpc,
+						      &self->rpc,
 						      (fw.len + sizeof(u32) -
-						      sizeof(u8)) /
+						       sizeof(u8)) /
 						      sizeof(u32));
 			if (err < 0)
 				goto err_exit;
 		}
 
-		*rpc = &PHAL_ATLANTIC->rpc;
+		*rpc = &self->rpc;
 	}
 
 err_exit:
 	return err;
 }
 
-static int hw_atl_utils_mpi_create(struct aq_hw_s *self,
-				   struct aq_hw_caps_s *aq_hw_caps)
+static int hw_atl_utils_mpi_create(struct aq_hw_s *self)
 {
 	int err = 0;
 
-	err = hw_atl_utils_init_ucp(self, aq_hw_caps);
+	err = hw_atl_utils_init_ucp(self, self->aq_nic_cfg->aq_hw_caps);
 	if (err < 0)
 		goto err_exit;
 
@@ -249,80 +612,103 @@ err_exit:
 	return err;
 }
 
+int hw_atl_utils_mpi_read_mbox(struct aq_hw_s *self,
+			       struct hw_atl_utils_mbox_header *pmbox)
+{
+	return hw_atl_utils_fw_downld_dwords(self,
+					     self->mbox_addr,
+					     (u32 *)(void *)pmbox,
+					     sizeof(*pmbox) / sizeof(u32));
+}
+
 void hw_atl_utils_mpi_read_stats(struct aq_hw_s *self,
-				 struct hw_aq_atl_utils_mbox *pmbox)
+				 struct hw_atl_utils_mbox *pmbox)
 {
 	int err = 0;
 
 	err = hw_atl_utils_fw_downld_dwords(self,
-					    PHAL_ATLANTIC->mbox_addr,
+					    self->mbox_addr,
 					    (u32 *)(void *)pmbox,
 					    sizeof(*pmbox) / sizeof(u32));
 	if (err < 0)
 		goto err_exit;
 
-	if (pmbox != &PHAL_ATLANTIC->mbox)
-		memcpy(pmbox, &PHAL_ATLANTIC->mbox, sizeof(*pmbox));
-
-	if (IS_CHIP_FEATURE(REVISION_A0)) {
+	if (ATL_HW_IS_CHIP_FEATURE(self, REVISION_A0)) {
 		unsigned int mtu = self->aq_nic_cfg ?
 					self->aq_nic_cfg->mtu : 1514U;
 		pmbox->stats.ubrc = pmbox->stats.uprc * mtu;
 		pmbox->stats.ubtc = pmbox->stats.uptc * mtu;
-		pmbox->stats.dpc = atomic_read(&PHAL_ATLANTIC_A0->dpc);
+		pmbox->stats.dpc = atomic_read(&self->dpc);
 	} else {
-		pmbox->stats.dpc = reg_rx_dma_stat_counter7get(self);
+		pmbox->stats.dpc = hw_atl_rpb_rx_dma_drop_pkt_cnt_get(self);
 	}
 
 err_exit:;
 }
 
-int hw_atl_utils_mpi_set_speed(struct aq_hw_s *self, u32 speed,
-			       enum hal_atl_utils_fw_state_e state)
+static int hw_atl_utils_mpi_set_speed(struct aq_hw_s *self, u32 speed)
 {
-	u32 ucp_0x368 = 0;
+	u32 val = aq_hw_read_reg(self, HW_ATL_MPI_CONTROL_ADR);
 
-	ucp_0x368 = (speed << HW_ATL_MPI_SPEED_SHIFT) | state;
-	aq_hw_write_reg(self, HW_ATL_MPI_CONTROL_ADR, ucp_0x368);
+	val = val & ~HW_ATL_MPI_SPEED_MSK;
+	val |= speed << HW_ATL_MPI_SPEED_SHIFT;
+	aq_hw_write_reg(self, HW_ATL_MPI_CONTROL_ADR, val);
 
 	return 0;
 }
 
-void hw_atl_utils_mpi_set(struct aq_hw_s *self,
-			  enum hal_atl_utils_fw_state_e state, u32 speed)
+static int hw_atl_utils_mpi_set_state(struct aq_hw_s *self,
+				      enum hal_atl_utils_fw_state_e state)
 {
-	int err = 0;
+	u32 val = aq_hw_read_reg(self, HW_ATL_MPI_CONTROL_ADR);
+	struct hw_atl_utils_mbox_header mbox;
 	u32 transaction_id = 0;
+	int err = 0;
 
 	if (state == MPI_RESET) {
-		hw_atl_utils_mpi_read_stats(self, &PHAL_ATLANTIC->mbox);
+		hw_atl_utils_mpi_read_mbox(self, &mbox);
 
-		transaction_id = PHAL_ATLANTIC->mbox.transaction_id;
+		transaction_id = mbox.transaction_id;
 
-		AQ_HW_WAIT_FOR(transaction_id !=
-				(hw_atl_utils_mpi_read_stats
-					(self, &PHAL_ATLANTIC->mbox),
-					PHAL_ATLANTIC->mbox.transaction_id),
-					1000U, 100U);
+		err = readx_poll_timeout_atomic(hw_atl_utils_get_mpi_mbox_tid,
+						self, mbox.transaction_id,
+						transaction_id !=
+						mbox.transaction_id,
+						1000U, 100000U);
 		if (err < 0)
 			goto err_exit;
 	}
+	/* On interface DEINIT we disable DW (raise bit)
+	 * Otherwise enable DW (clear bit)
+	 */
+	if (state == MPI_DEINIT || state == MPI_POWER)
+		val |= HW_ATL_MPI_DIRTY_WAKE_MSK;
+	else
+		val &= ~HW_ATL_MPI_DIRTY_WAKE_MSK;
 
-	err = hw_atl_utils_mpi_set_speed(self, speed, state);
+	/* Set new state bits */
+	val = val & ~HW_ATL_MPI_STATE_MSK;
+	val |= state & HW_ATL_MPI_STATE_MSK;
 
-err_exit:;
+	aq_hw_write_reg(self, HW_ATL_MPI_CONTROL_ADR, val);
+
+err_exit:
+	return err;
 }
 
-int hw_atl_utils_mpi_get_link_status(struct aq_hw_s *self,
-				     struct aq_hw_link_status_s *link_status)
+int hw_atl_utils_mpi_get_link_status(struct aq_hw_s *self)
 {
-	u32 cp0x036C = aq_hw_read_reg(self, HW_ATL_MPI_STATE_ADR);
-	u32 link_speed_mask = cp0x036C >> HW_ATL_MPI_SPEED_SHIFT;
+	struct aq_hw_link_status_s *link_status = &self->aq_link_status;
+	u32 mpi_state;
+	u32 speed;
+
+	mpi_state = hw_atl_utils_mpi_get_state(self);
+	speed = mpi_state >> HW_ATL_MPI_SPEED_SHIFT;
 
-	if (!link_speed_mask) {
+	if (!speed) {
 		link_status->mbps = 0U;
 	} else {
-		switch (link_speed_mask) {
+		switch (speed) {
 		case HAL_ATLANTIC_RATE_10G:
 			link_status->mbps = 10000U;
 			break;
@@ -332,7 +718,7 @@ int hw_atl_utils_mpi_get_link_status(struct aq_hw_s *self,
 			link_status->mbps = 5000U;
 			break;
 
-		case HAL_ATLANTIC_RATE_2GS:
+		case HAL_ATLANTIC_RATE_2G5:
 			link_status->mbps = 2500U;
 			break;
 
@@ -345,35 +731,26 @@ int hw_atl_utils_mpi_get_link_status(struct aq_hw_s *self,
 			break;
 
 		default:
-			link_status->mbps = 0U;
-			break;
+			return -EBUSY;
 		}
 	}
+	link_status->full_duplex = true;
 
 	return 0;
 }
 
 int hw_atl_utils_get_mac_permanent(struct aq_hw_s *self,
-				   struct aq_hw_caps_s *aq_hw_caps,
 				   u8 *mac)
 {
+	u32 mac_addr[2];
+	u32 efuse_addr;
 	int err = 0;
 	u32 h = 0U;
 	u32 l = 0U;
-	u32 mac_addr[2];
-
-	self->mmio = aq_pci_func_get_mmio(self->aq_pci_func);
-
-	hw_atl_utils_hw_chip_features_init(self,
-					   &PHAL_ATLANTIC_A0->chip_features);
-
-	err = hw_atl_utils_mpi_create(self, aq_hw_caps);
-	if (err < 0)
-		goto err_exit;
 
 	if (!aq_hw_read_reg(self, HW_ATL_UCP_0X370_REG)) {
-		unsigned int rnd = 0;
 		unsigned int ucp_0x370 = 0;
+		unsigned int rnd = 0;
 
 		get_random_bytes(&rnd, sizeof(unsigned int));
 
@@ -381,11 +758,10 @@ int hw_atl_utils_get_mac_permanent(struct aq_hw_s *self,
 		aq_hw_write_reg(self, HW_ATL_UCP_0X370_REG, ucp_0x370);
 	}
 
-	err = hw_atl_utils_fw_downld_dwords(self,
-					    aq_hw_read_reg(self, 0x00000374U) +
-					    (40U * 4U),
-					    mac_addr,
-					    AQ_DIMOF(mac_addr));
+	efuse_addr = aq_hw_read_reg(self, 0x00000374U);
+
+	err = hw_atl_utils_fw_downld_dwords(self, efuse_addr + (40U * 4U),
+					    mac_addr, ARRAY_SIZE(mac_addr));
 	if (err < 0) {
 		mac_addr[0] = 0U;
 		mac_addr[1] = 0U;
@@ -399,9 +775,9 @@ int hw_atl_utils_get_mac_permanent(struct aq_hw_s *self,
 
 	if ((mac[0] & 0x01U) || ((mac[0] | mac[1] | mac[2]) == 0x00U)) {
 		/* chip revision */
-		l = 0xE3000000U
-			| (0xFFFFU & aq_hw_read_reg(self, HW_ATL_UCP_0X370_REG))
-			| (0x00 << 16);
+		l = 0xE3000000U |
+		    (0xFFFFU & aq_hw_read_reg(self, HW_ATL_UCP_0X370_REG)) |
+		    (0x00 << 16);
 		h = 0x8001300EU;
 
 		mac[5] = (u8)(0xFFU & l);
@@ -416,7 +792,6 @@ int hw_atl_utils_get_mac_permanent(struct aq_hw_s *self,
 		mac[0] = (u8)(0xFFU & h);
 	}
 
-err_exit:
 	return err;
 }
 
@@ -448,84 +823,104 @@ unsigned int hw_atl_utils_mbps_2_speed_index(unsigned int mbps)
 	default:
 		break;
 	}
+
 	return ret;
 }
 
 void hw_atl_utils_hw_chip_features_init(struct aq_hw_s *self, u32 *p)
 {
-	u32 chip_features = 0U;
-	u32 val = reg_glb_mif_id_get(self);
+	u32 val = hw_atl_reg_glb_mif_id_get(self);
 	u32 mif_rev = val & 0xFFU;
+	u32 chip_features = 0U;
 
-	if ((3U & mif_rev) == 1U) {
-		chip_features |=
-			HAL_ATLANTIC_UTILS_CHIP_REVISION_A0 |
-			HAL_ATLANTIC_UTILS_CHIP_MPI_AQ |
-			HAL_ATLANTIC_UTILS_CHIP_MIPS;
-	} else if ((3U & mif_rev) == 2U) {
-		chip_features |=
-			HAL_ATLANTIC_UTILS_CHIP_REVISION_B0 |
-			HAL_ATLANTIC_UTILS_CHIP_MPI_AQ |
-			HAL_ATLANTIC_UTILS_CHIP_MIPS |
-			HAL_ATLANTIC_UTILS_CHIP_TPO2 |
-			HAL_ATLANTIC_UTILS_CHIP_RPF2;
+	chip_features |= ATL_HW_CHIP_ATLANTIC;
+
+	if ((0xFU & mif_rev) == 1U) {
+		chip_features |= ATL_HW_CHIP_REVISION_A0 |
+			ATL_HW_CHIP_MPI_AQ |
+			ATL_HW_CHIP_MIPS;
+	} else if ((0xFU & mif_rev) == 2U) {
+		chip_features |= ATL_HW_CHIP_REVISION_B0 |
+			ATL_HW_CHIP_MPI_AQ |
+			ATL_HW_CHIP_MIPS |
+			ATL_HW_CHIP_TPO2 |
+			ATL_HW_CHIP_RPF2;
+	} else if ((0xFU & mif_rev) == 0xAU) {
+		chip_features |= ATL_HW_CHIP_REVISION_B1 |
+			ATL_HW_CHIP_MPI_AQ |
+			ATL_HW_CHIP_MIPS |
+			ATL_HW_CHIP_TPO2 |
+			ATL_HW_CHIP_RPF2;
 	}
 
 	*p = chip_features;
 }
 
-int hw_atl_utils_hw_deinit(struct aq_hw_s *self)
+static int hw_atl_fw1x_deinit(struct aq_hw_s *self)
 {
-	hw_atl_utils_mpi_set(self, MPI_DEINIT, 0x0U);
+	hw_atl_utils_mpi_set_speed(self, 0);
+	hw_atl_utils_mpi_set_state(self, MPI_DEINIT);
+
 	return 0;
 }
 
-int hw_atl_utils_hw_set_power(struct aq_hw_s *self,
-			      unsigned int power_state)
+int hw_atl_utils_update_stats(struct aq_hw_s *self)
 {
-	hw_atl_utils_mpi_set(self, MPI_POWER, 0x0U);
-	return 0;
-}
+	struct aq_stats_s *cs = &self->curr_stats;
+	struct aq_stats_s curr_stats = *cs;
+	struct hw_atl_utils_mbox mbox;
+	bool corrupted_stats = false;
+
+	hw_atl_utils_mpi_read_stats(self, &mbox);
+
+#define AQ_SDELTA(_N_)  \
+do { \
+	if (!corrupted_stats && \
+	    ((s64)(mbox.stats._N_ - self->last_stats._N_)) >= 0) \
+		curr_stats._N_ += mbox.stats._N_ - self->last_stats._N_; \
+	else \
+		corrupted_stats = true; \
+} while (0)
+
+	if (self->aq_link_status.mbps) {
+		AQ_SDELTA(uprc);
+		AQ_SDELTA(mprc);
+		AQ_SDELTA(bprc);
+		AQ_SDELTA(erpt);
+
+		AQ_SDELTA(uptc);
+		AQ_SDELTA(mptc);
+		AQ_SDELTA(bptc);
+		AQ_SDELTA(erpr);
+
+		AQ_SDELTA(ubrc);
+		AQ_SDELTA(ubtc);
+		AQ_SDELTA(mbrc);
+		AQ_SDELTA(mbtc);
+		AQ_SDELTA(bbrc);
+		AQ_SDELTA(bbtc);
+		AQ_SDELTA(dpc);
+
+		if (!corrupted_stats)
+			*cs = curr_stats;
+	}
+#undef AQ_SDELTA
 
-int hw_atl_utils_get_hw_stats(struct aq_hw_s *self,
-			      u64 *data, unsigned int *p_count)
-{
-	struct hw_atl_stats_s *stats = NULL;
-	int i = 0;
-
-	hw_atl_utils_mpi_read_stats(self, &PHAL_ATLANTIC->mbox);
-
-	stats = &PHAL_ATLANTIC->mbox.stats;
-
-	data[i] = stats->uprc + stats->mprc + stats->bprc;
-	data[++i] = stats->uprc;
-	data[++i] = stats->mprc;
-	data[++i] = stats->bprc;
-	data[++i] = stats->erpt;
-	data[++i] = stats->uptc + stats->mptc + stats->bptc;
-	data[++i] = stats->uptc;
-	data[++i] = stats->mptc;
-	data[++i] = stats->bptc;
-	data[++i] = stats->ubrc;
-	data[++i] = stats->ubtc;
-	data[++i] = stats->mbrc;
-	data[++i] = stats->mbtc;
-	data[++i] = stats->bbrc;
-	data[++i] = stats->bbtc;
-	data[++i] = stats->ubrc + stats->mbrc + stats->bbrc;
-	data[++i] = stats->ubtc + stats->mbtc + stats->bbtc;
-	data[++i] = stats_rx_dma_good_pkt_counterlsw_get(self);
-	data[++i] = stats_tx_dma_good_pkt_counterlsw_get(self);
-	data[++i] = stats_rx_dma_good_octet_counterlsw_get(self);
-	data[++i] = stats_tx_dma_good_octet_counterlsw_get(self);
-	data[++i] = stats->dpc;
-
-	if (p_count)
-		*p_count = ++i;
+	cs->dma_pkt_rc = hw_atl_stats_rx_dma_good_pkt_counter_get(self);
+	cs->dma_pkt_tc = hw_atl_stats_tx_dma_good_pkt_counter_get(self);
+	cs->dma_oct_rc = hw_atl_stats_rx_dma_good_octet_counter_get(self);
+	cs->dma_oct_tc = hw_atl_stats_tx_dma_good_octet_counter_get(self);
+
+	memcpy(&self->last_stats, &mbox.stats, sizeof(mbox.stats));
 
 	return 0;
 }
 
+struct aq_stats_s *hw_atl_utils_get_hw_stats(struct aq_hw_s *self)
+{
+	return &self->curr_stats;
+}
+
 static const u32 hw_atl_utils_hw_mac_regs[] = {
 	0x00005580U, 0x00005590U, 0x000055B0U, 0x000055B4U,
 	0x000055C0U, 0x00005B00U, 0x00005B04U, 0x00005B08U,
@@ -552,19 +947,153 @@ static const u32 hw_atl_utils_hw_mac_regs[] = {
 };
 
 int hw_atl_utils_hw_get_regs(struct aq_hw_s *self,
-			     struct aq_hw_caps_s *aq_hw_caps,
+			     const struct aq_hw_caps_s *aq_hw_caps,
 			     u32 *regs_buff)
 {
 	unsigned int i = 0U;
 
 	for (i = 0; i < aq_hw_caps->mac_regs_count; i++)
 		regs_buff[i] = aq_hw_read_reg(self,
-			hw_atl_utils_hw_mac_regs[i]);
+					      hw_atl_utils_hw_mac_regs[i]);
+
 	return 0;
 }
 
-int hw_atl_utils_get_fw_version(struct aq_hw_s *self, u32 *fw_version)
+u32 hw_atl_utils_get_fw_version(struct aq_hw_s *self)
 {
-	*fw_version = aq_hw_read_reg(self, 0x18U);
-	return 0;
+	return aq_hw_read_reg(self, HW_ATL_MPI_FW_VERSION);
+}
+
+static int aq_fw1x_set_wake_magic(struct aq_hw_s *self, bool wol_enabled,
+				  const u8 *mac)
+{
+	struct hw_atl_utils_fw_rpc *prpc = NULL;
+	unsigned int rpc_size = 0U;
+	int err = 0;
+
+	err = hw_atl_utils_fw_rpc_wait(self, &prpc);
+	if (err < 0)
+		goto err_exit;
+
+	memset(prpc, 0, sizeof(*prpc));
+
+	if (wol_enabled) {
+		rpc_size = offsetof(struct hw_atl_utils_fw_rpc, msg_wol_add) +
+			   sizeof(prpc->msg_wol_add);
+
+
+		prpc->msg_id = HAL_ATLANTIC_UTILS_FW_MSG_WOL_ADD;
+		prpc->msg_wol_add.priority =
+				HAL_ATLANTIC_UTILS_FW_MSG_WOL_PRIOR;
+		prpc->msg_wol_add.pattern_id =
+				HAL_ATLANTIC_UTILS_FW_MSG_WOL_PATTERN;
+		prpc->msg_wol_add.packet_type =
+				HAL_ATLANTIC_UTILS_FW_MSG_WOL_MAG_PKT;
+
+		ether_addr_copy((u8 *)&prpc->msg_wol_add.magic_packet_pattern,
+				mac);
+	} else {
+		rpc_size = sizeof(prpc->msg_wol_remove) +
+			   offsetof(struct hw_atl_utils_fw_rpc, msg_wol_remove);
+
+		prpc->msg_id = HAL_ATLANTIC_UTILS_FW_MSG_WOL_DEL;
+		prpc->msg_wol_add.pattern_id =
+				HAL_ATLANTIC_UTILS_FW_MSG_WOL_PATTERN;
+	}
+
+	err = hw_atl_utils_fw_rpc_call(self, rpc_size);
+
+err_exit:
+	return err;
 }
+
+static int aq_fw1x_set_power(struct aq_hw_s *self, unsigned int power_state,
+			     const u8 *mac)
+{
+	struct hw_atl_utils_fw_rpc *prpc = NULL;
+	unsigned int rpc_size = 0U;
+	int err = 0;
+
+	if (self->aq_nic_cfg->wol & WAKE_MAGIC) {
+		err = aq_fw1x_set_wake_magic(self, 1, mac);
+
+		if (err < 0)
+			goto err_exit;
+
+		rpc_size = sizeof(prpc->msg_id) +
+			   sizeof(prpc->msg_enable_wakeup);
+
+		err = hw_atl_utils_fw_rpc_wait(self, &prpc);
+
+		if (err < 0)
+			goto err_exit;
+
+		memset(prpc, 0, rpc_size);
+
+		prpc->msg_id = HAL_ATLANTIC_UTILS_FW_MSG_ENABLE_WAKEUP;
+		prpc->msg_enable_wakeup.pattern_mask = 0x00000002;
+
+		err = hw_atl_utils_fw_rpc_call(self, rpc_size);
+		if (err < 0)
+			goto err_exit;
+	}
+	hw_atl_utils_mpi_set_speed(self, 0);
+	hw_atl_utils_mpi_set_state(self, MPI_POWER);
+
+err_exit:
+	return err;
+}
+
+static u32 hw_atl_utils_get_mpi_mbox_tid(struct aq_hw_s *self)
+{
+	struct hw_atl_utils_mbox_header mbox;
+
+	hw_atl_utils_mpi_read_mbox(self, &mbox);
+
+	return mbox.transaction_id;
+}
+
+static u32 hw_atl_utils_mpi_get_state(struct aq_hw_s *self)
+{
+	return aq_hw_read_reg(self, HW_ATL_MPI_STATE_ADR);
+}
+
+static u32 hw_atl_utils_mif_cmd_get(struct aq_hw_s *self)
+{
+	return aq_hw_read_reg(self, HW_ATL_MIF_CMD);
+}
+
+static u32 hw_atl_utils_mif_addr_get(struct aq_hw_s *self)
+{
+	return aq_hw_read_reg(self, HW_ATL_MIF_ADDR);
+}
+
+static u32 hw_atl_utils_rpc_state_get(struct aq_hw_s *self)
+{
+	return aq_hw_read_reg(self, HW_ATL_RPC_STATE_ADR);
+}
+
+static u32 aq_fw1x_rpc_get(struct aq_hw_s *self)
+{
+	return aq_hw_read_reg(self, HW_ATL_MPI_RPC_ADDR);
+}
+
+const struct aq_fw_ops aq_fw_1x_ops = {
+	.init = hw_atl_utils_mpi_create,
+	.deinit = hw_atl_fw1x_deinit,
+	.reset = NULL,
+	.get_mac_permanent = hw_atl_utils_get_mac_permanent,
+	.set_link_speed = hw_atl_utils_mpi_set_speed,
+	.set_state = hw_atl_utils_mpi_set_state,
+	.update_link_status = hw_atl_utils_mpi_get_link_status,
+	.update_stats = hw_atl_utils_update_stats,
+	.get_mac_temp = NULL,
+	.get_phy_temp = NULL,
+	.set_power = aq_fw1x_set_power,
+	.set_eee_rate = NULL,
+	.get_eee_rate = NULL,
+	.set_flow_control = NULL,
+	.send_fw_request = NULL,
+	.enable_ptp = NULL,
+	.led_control = NULL,
+};
diff --git a/drivers/net/ethernet/aquantia/atlantic/hw_atl/hw_atl_utils.h b/drivers/net/ethernet/aquantia/atlantic/hw_atl/hw_atl_utils.h
index b8e3d88f0879..f6b990b7f5b4 100644
--- a/drivers/net/ethernet/aquantia/atlantic/hw_atl/hw_atl_utils.h
+++ b/drivers/net/ethernet/aquantia/atlantic/hw_atl/hw_atl_utils.h
@@ -1,10 +1,8 @@
-/*
- * aQuantia Corporation Network Driver
- * Copyright (C) 2014-2017 aQuantia Corporation. All rights reserved
+/* SPDX-License-Identifier: GPL-2.0-only */
+/* Atlantic Network Driver
  *
- * This program is free software; you can redistribute it and/or modify it
- * under the terms and conditions of the GNU General Public License,
- * version 2, as published by the Free Software Foundation.
+ * Copyright (C) 2014-2019 aQuantia Corporation
+ * Copyright (C) 2019-2020 Marvell International Ltd.
  */
 
 /* File hw_atl_utils.h: Declaration of common functions for Atlantic hardware
@@ -14,10 +12,47 @@
 #ifndef HW_ATL_UTILS_H
 #define HW_ATL_UTILS_H
 
-#include "../aq_common.h"
-
 #define HW_ATL_FLUSH() { (void)aq_hw_read_reg(self, 0x10); }
 
+/* Hardware tx descriptor */
+struct __packed hw_atl_txd_s {
+	u64 buf_addr;
+	u32 ctl;
+	u32 ctl2; /* 63..46 - payload length, 45 - ctx enable, 44 - ctx index */
+};
+
+/* Hardware tx context descriptor */
+struct __packed hw_atl_txc_s {
+	u32 rsvd;
+	u32 len;
+	u32 ctl;
+	u32 len2;
+};
+
+/* Hardware rx descriptor */
+struct __packed hw_atl_rxd_s {
+	u64 buf_addr;
+	u64 hdr_addr;
+};
+
+/* Hardware rx descriptor writeback */
+struct __packed hw_atl_rxd_wb_s {
+	u32 type;
+	u32 rss_hash;
+	u16 status;
+	u16 pkt_len;
+	u16 next_desc_ptr;
+	__le16 vlan;
+};
+
+/* Hardware rx HW TIMESTAMP writeback */
+struct __packed hw_atl_rxd_hwts_wb_s {
+	u32 sec_hw;
+	u32 ns;
+	u32 sec_lw0;
+	u32 sec_lw1;
+};
+
 struct __packed hw_atl_stats_s {
 	u32 uprc;
 	u32 mprc;
@@ -36,122 +71,342 @@ struct __packed hw_atl_stats_s {
 	u32 dpc;
 };
 
-union __packed ip_addr {
-	struct {
-		u8 addr[16];
-	} v6;
-	struct {
-		u8 padding[12];
-		u8 addr[4];
-	} v4;
+struct __packed drv_msg_enable_wakeup {
+	union {
+		u32 pattern_mask;
+
+		struct {
+			u32 reason_arp_v4_pkt : 1;
+			u32 reason_ipv4_ping_pkt : 1;
+			u32 reason_ipv6_ns_pkt : 1;
+			u32 reason_ipv6_ping_pkt : 1;
+			u32 reason_link_up : 1;
+			u32 reason_link_down : 1;
+			u32 reason_maximum : 1;
+		};
+	};
+
+	union {
+		u32 offload_mask;
+	};
+};
+
+struct __packed magic_packet_pattern_s {
+	u8 mac_addr[ETH_ALEN];
+};
+
+struct __packed drv_msg_wol_add {
+	u32 priority;
+	u32 packet_type;
+	u32 pattern_id;
+	u32 next_pattern_offset;
+
+	struct magic_packet_pattern_s magic_packet_pattern;
+};
+
+struct __packed drv_msg_wol_remove {
+	u32 id;
+};
+
+struct __packed hw_atl_utils_mbox_header {
+	u32 version;
+	u32 transaction_id;
+	u32 error;
 };
 
-struct __packed hw_aq_atl_utils_fw_rpc {
+struct __packed hw_atl_ptp_offset {
+	u16 ingress_100;
+	u16 egress_100;
+	u16 ingress_1000;
+	u16 egress_1000;
+	u16 ingress_2500;
+	u16 egress_2500;
+	u16 ingress_5000;
+	u16 egress_5000;
+	u16 ingress_10000;
+	u16 egress_10000;
+};
+
+struct __packed hw_atl_cable_diag {
+	u8 fault;
+	u8 distance;
+	u8 far_distance;
+	u8 reserved;
+};
+
+enum gpio_pin_function {
+	GPIO_PIN_FUNCTION_NC,
+	GPIO_PIN_FUNCTION_VAUX_ENABLE,
+	GPIO_PIN_FUNCTION_EFUSE_BURN_ENABLE,
+	GPIO_PIN_FUNCTION_SFP_PLUS_DETECT,
+	GPIO_PIN_FUNCTION_TX_DISABLE,
+	GPIO_PIN_FUNCTION_RATE_SEL_0,
+	GPIO_PIN_FUNCTION_RATE_SEL_1,
+	GPIO_PIN_FUNCTION_TX_FAULT,
+	GPIO_PIN_FUNCTION_PTP0,
+	GPIO_PIN_FUNCTION_PTP1,
+	GPIO_PIN_FUNCTION_PTP2,
+	GPIO_PIN_FUNCTION_SIZE
+};
+
+struct __packed hw_atl_info {
+	u8 reserved[6];
+	u16 phy_fault_code;
+	u16 phy_temperature;
+	u8 cable_len;
+	u8 reserved1;
+	struct hw_atl_cable_diag cable_diag_data[4];
+	struct hw_atl_ptp_offset ptp_offset;
+	u8 reserved2[12];
+	u32 caps_lo;
+	u32 caps_hi;
+	u32 reserved_datapath;
+	u32 reserved3[7];
+	u32 reserved_simpleresp[3];
+	u32 reserved_linkstat[7];
+	u32 reserved_wakes_count;
+	u32 reserved_eee_stat[12];
+	u32 tx_stuck_cnt;
+	u32 setting_address;
+	u32 setting_length;
+	u32 caps_ex;
+	enum gpio_pin_function gpio_pin[3];
+	u32 pcie_aer_dump[18];
+	u16 snr_margin[4];
+};
+
+struct __packed hw_atl_utils_mbox {
+	struct hw_atl_utils_mbox_header header;
+	struct hw_atl_stats_s stats;
+	struct hw_atl_info info;
+};
+
+struct __packed offload_ip_info {
+	u8 v4_local_addr_count;
+	u8 v4_addr_count;
+	u8 v6_local_addr_count;
+	u8 v6_addr_count;
+	u32 v4_addr;
+	u32 v4_prefix;
+	u32 v6_addr;
+	u32 v6_prefix;
+};
+
+struct __packed offload_port_info {
+	u16 udp_port_count;
+	u16 tcp_port_count;
+	u32 udp_port;
+	u32 tcp_port;
+};
+
+struct __packed offload_ka_info {
+	u16 v4_ka_count;
+	u16 v6_ka_count;
+	u32 retry_count;
+	u32 retry_interval;
+	u32 v4_ka;
+	u32 v6_ka;
+};
+
+struct __packed offload_rr_info {
+	u32 rr_count;
+	u32 rr_buf_len;
+	u32 rr_id_x;
+	u32 rr_buf;
+};
+
+struct __packed offload_info {
+	u32 version;
+	u32 len;
+	u8 mac_addr[ETH_ALEN];
+
+	u8 reserved[2];
+
+	struct offload_ip_info ips;
+	struct offload_port_info ports;
+	struct offload_ka_info kas;
+	struct offload_rr_info rrs;
+};
+
+struct __packed hw_atl_utils_fw_rpc {
 	u32 msg_id;
 
 	union {
-		struct {
-			u32 pong;
-		} msg_ping;
+		/* fw1x structures */
+		struct drv_msg_wol_add msg_wol_add;
+		struct drv_msg_wol_remove msg_wol_remove;
+		struct drv_msg_enable_wakeup msg_enable_wakeup;
+		/* fw2x structures */
+		struct offload_info fw2x_offloads;
+	};
+};
 
-		struct {
-			u8 mac_addr[6];
-			u32 ip_addr_cnt;
+/* Mailbox FW Request interface */
+struct __packed hw_fw_request_ptp_gpio_ctrl {
+	u32 index;
+	u32 period;
+	u64 start;
+};
 
-			struct {
-				union ip_addr addr;
-				union ip_addr mask;
-			} ip[1];
-		} msg_arp;
+struct __packed hw_fw_request_ptp_adj_freq {
+	u32 ns_mac;
+	u32 fns_mac;
+	u32 ns_phy;
+	u32 fns_phy;
+	u32 mac_ns_adj;
+	u32 mac_fns_adj;
+};
 
-		struct {
-			u32 len;
-			u8 packet[1514U];
-		} msg_inject;
+struct __packed hw_fw_request_ptp_adj_clock {
+	u32 ns;
+	u32 sec;
+	int sign;
+};
 
-		struct {
-			u32 priority;
-			u32 wol_packet_type;
-			u16 friendly_name_len;
-			u16 friendly_name[65];
-			u32 pattern_id;
-			u32 next_wol_pattern_offset;
-
-			union {
-				struct {
-					u32 flags;
-					u8 ipv4_source_address[4];
-					u8 ipv4_dest_address[4];
-					u16 tcp_source_port_number;
-					u16 tcp_dest_port_number;
-				} ipv4_tcp_syn_parameters;
-
-				struct {
-					u32 flags;
-					u8 ipv6_source_address[16];
-					u8 ipv6_dest_address[16];
-					u16 tcp_source_port_number;
-					u16 tcp_dest_port_number;
-				} ipv6_tcp_syn_parameters;
-
-				struct {
-					u32 flags;
-				} eapol_request_id_message_parameters;
-
-				struct {
-					u32 flags;
-					u32 mask_offset;
-					u32 mask_size;
-					u32 pattern_offset;
-					u32 pattern_size;
-				} wol_bit_map_pattern;
-			} wol_pattern;
-		} msg_wol;
+#define HW_AQ_FW_REQUEST_PTP_GPIO_CTRL	         0x11
+#define HW_AQ_FW_REQUEST_PTP_ADJ_FREQ	         0x12
+#define HW_AQ_FW_REQUEST_PTP_ADJ_CLOCK	         0x13
 
-		struct {
-			u32 is_wake_on_link_down;
-			u32 is_wake_on_link_up;
-		} msg_wolink;
+struct __packed hw_fw_request_iface {
+	u32 msg_id;
+	union {
+		/* PTP FW Request */
+		struct hw_fw_request_ptp_gpio_ctrl ptp_gpio_ctrl;
+		struct hw_fw_request_ptp_adj_freq ptp_adj_freq;
+		struct hw_fw_request_ptp_adj_clock ptp_adj_clock;
 	};
 };
 
-struct __packed hw_aq_atl_utils_mbox {
-	u32 version;
-	u32 transaction_id;
-	int error;
-	struct hw_atl_stats_s stats;
+struct __packed hw_atl_utils_settings {
+	u32 mtu;
+	u32 downshift_retry_count;
+	u32 link_pause_frame_quanta_100m;
+	u32 link_pause_frame_threshold_100m;
+	u32 link_pause_frame_quanta_1g;
+	u32 link_pause_frame_threshold_1g;
+	u32 link_pause_frame_quanta_2p5g;
+	u32 link_pause_frame_threshold_2p5g;
+	u32 link_pause_frame_quanta_5g;
+	u32 link_pause_frame_threshold_5g;
+	u32 link_pause_frame_quanta_10g;
+	u32 link_pause_frame_threshold_10g;
+	u32 pfc_quanta_class_0;
+	u32 pfc_threshold_class_0;
+	u32 pfc_quanta_class_1;
+	u32 pfc_threshold_class_1;
+	u32 pfc_quanta_class_2;
+	u32 pfc_threshold_class_2;
+	u32 pfc_quanta_class_3;
+	u32 pfc_threshold_class_3;
+	u32 pfc_quanta_class_4;
+	u32 pfc_threshold_class_4;
+	u32 pfc_quanta_class_5;
+	u32 pfc_threshold_class_5;
+	u32 pfc_quanta_class_6;
+	u32 pfc_threshold_class_6;
+	u32 pfc_quanta_class_7;
+	u32 pfc_threshold_class_7;
+	u32 eee_link_down_timeout;
+	u32 eee_link_up_timeout;
+	u32 eee_max_link_drops;
+	u32 eee_rates_mask;
+	u32 wake_timer;
+	u32 thermal_shutdown_off_temp;
+	u32 thermal_shutdown_warning_temp;
+	u32 thermal_shutdown_cold_temp;
+	u32 msm_options;
+	u32 dac_cable_serdes_modes;
+	u32 media_detect;
+};
+
+enum macsec_msg_type {
+	macsec_cfg_msg = 0,
+	macsec_add_rx_sc_msg,
+	macsec_add_tx_sc_msg,
+	macsec_add_rx_sa_msg,
+	macsec_add_tx_sa_msg,
+	macsec_get_stats_msg,
+};
+
+struct __packed macsec_cfg_request {
+	u32 enabled;
+	u32 egress_threshold;
+	u32 ingress_threshold;
+	u32 interrupts_enabled;
 };
 
-struct __packed hw_atl_s {
-	struct aq_hw_s base;
-	struct hw_aq_atl_utils_mbox mbox;
-	u64 speed;
-	u32 itr_tx;
-	u32 itr_rx;
-	unsigned int chip_features;
-	u32 fw_ver_actual;
-	atomic_t dpc;
-	u32 mbox_addr;
-	u32 rpc_addr;
-	u32 rpc_tid;
-	struct hw_aq_atl_utils_fw_rpc rpc;
+struct __packed macsec_msg_fw_request {
+	u32 msg_id; /* not used */
+	u32 msg_type;
+	struct macsec_cfg_request cfg;
 };
 
-#define SELF ((struct hw_atl_s *)self)
+struct __packed macsec_msg_fw_response {
+	u32 result;
+};
+
+enum hw_atl_rx_action_with_traffic {
+	HW_ATL_RX_DISCARD,
+	HW_ATL_RX_HOST,
+	HW_ATL_RX_MNGMNT,
+	HW_ATL_RX_HOST_AND_MNGMNT,
+	HW_ATL_RX_WOL
+};
+
+struct aq_rx_filter_vlan {
+	u8 enable;
+	u8 location;
+	u16 vlan_id;
+	u8 queue;
+};
+
+#define HW_ATL_VLAN_MAX_FILTERS         16U
+
+struct aq_rx_filter_l2 {
+	s8 queue;
+	u8 location;
+	u8 user_priority_en;
+	u8 user_priority;
+	u16 ethertype;
+};
+
+struct aq_rx_filter_l3l4 {
+	u32 cmd;
+	u8 location;
+	u32 ip_dst[4];
+	u32 ip_src[4];
+	u16 p_dst;
+	u16 p_src;
+	u8 is_ipv6;
+};
+
+enum hw_atl_rx_protocol_value_l3l4 {
+	HW_ATL_RX_TCP,
+	HW_ATL_RX_UDP,
+	HW_ATL_RX_SCTP,
+	HW_ATL_RX_ICMP
+};
+
+enum hw_atl_rx_ctrl_registers_l3l4 {
+	HW_ATL_RX_ENABLE_MNGMNT_QUEUE_L3L4 = BIT(22),
+	HW_ATL_RX_ENABLE_QUEUE_L3L4        = BIT(23),
+	HW_ATL_RX_ENABLE_ARP_FLTR_L3       = BIT(24),
+	HW_ATL_RX_ENABLE_CMP_PROT_L4       = BIT(25),
+	HW_ATL_RX_ENABLE_CMP_DEST_PORT_L4  = BIT(26),
+	HW_ATL_RX_ENABLE_CMP_SRC_PORT_L4   = BIT(27),
+	HW_ATL_RX_ENABLE_CMP_DEST_ADDR_L3  = BIT(28),
+	HW_ATL_RX_ENABLE_CMP_SRC_ADDR_L3   = BIT(29),
+	HW_ATL_RX_ENABLE_L3_IPV6           = BIT(30),
+	HW_ATL_RX_ENABLE_FLTR_L3L4         = BIT(31)
+};
 
-#define PHAL_ATLANTIC ((struct hw_atl_s *)((void *)(self)))
-#define PHAL_ATLANTIC_A0 ((struct hw_atl_s *)((void *)(self)))
-#define PHAL_ATLANTIC_B0 ((struct hw_atl_s *)((void *)(self)))
+#define HW_ATL_RX_QUEUE_FL3L4_SHIFT       8U
+#define HW_ATL_RX_ACTION_FL3F4_SHIFT      16U
 
-#define HAL_ATLANTIC_UTILS_CHIP_MIPS         0x00000001U
-#define HAL_ATLANTIC_UTILS_CHIP_TPO2         0x00000002U
-#define HAL_ATLANTIC_UTILS_CHIP_RPF2         0x00000004U
-#define HAL_ATLANTIC_UTILS_CHIP_MPI_AQ       0x00000010U
-#define HAL_ATLANTIC_UTILS_CHIP_REVISION_A0  0x01000000U
-#define HAL_ATLANTIC_UTILS_CHIP_REVISION_B0  0x02000000U
+#define HW_ATL_RX_CNT_REG_ADDR_IPV6       4U
 
-#define IS_CHIP_FEATURE(_F_) (HAL_ATLANTIC_UTILS_CHIP_##_F_ & \
-				PHAL_ATLANTIC->chip_features)
+#define HW_ATL_GET_REG_LOCATION_FL3L4(location) \
+	((location) - AQ_RX_FIRST_LOC_FL3L4)
 
 enum hal_atl_utils_fw_state_e {
 	MPI_DEINIT = 0,
@@ -163,48 +418,225 @@ enum hal_atl_utils_fw_state_e {
 #define HAL_ATLANTIC_RATE_10G        BIT(0)
 #define HAL_ATLANTIC_RATE_5G         BIT(1)
 #define HAL_ATLANTIC_RATE_5GSR       BIT(2)
-#define HAL_ATLANTIC_RATE_2GS        BIT(3)
+#define HAL_ATLANTIC_RATE_2G5        BIT(3)
 #define HAL_ATLANTIC_RATE_1G         BIT(4)
 #define HAL_ATLANTIC_RATE_100M       BIT(5)
 #define HAL_ATLANTIC_RATE_INVALID    BIT(6)
 
+#define HAL_ATLANTIC_UTILS_FW_MSG_WOL_ADD       0x4U
+#define HAL_ATLANTIC_UTILS_FW_MSG_WOL_PRIOR     0x10000000U
+#define HAL_ATLANTIC_UTILS_FW_MSG_WOL_PATTERN   0x1U
+#define HAL_ATLANTIC_UTILS_FW_MSG_WOL_MAG_PKT   0x2U
+#define HAL_ATLANTIC_UTILS_FW_MSG_WOL_DEL       0x5U
+#define HAL_ATLANTIC_UTILS_FW_MSG_ENABLE_WAKEUP 0x6U
+
+enum hw_atl_fw2x_rate {
+	FW2X_RATE_100M    = 0x20,
+	FW2X_RATE_1G      = 0x100,
+	FW2X_RATE_2G5     = 0x200,
+	FW2X_RATE_5G      = 0x400,
+	FW2X_RATE_10G     = 0x800,
+};
+
+/* 0x370
+ * Link capabilities resolution register
+ */
+enum hw_atl_fw2x_caps_lo {
+	CAPS_LO_10BASET_HD        = 0,
+	CAPS_LO_10BASET_FD,
+	CAPS_LO_100BASETX_HD,
+	CAPS_LO_100BASET4_HD,
+	CAPS_LO_100BASET2_HD,
+	CAPS_LO_100BASETX_FD      = 5,
+	CAPS_LO_100BASET2_FD,
+	CAPS_LO_1000BASET_HD,
+	CAPS_LO_1000BASET_FD,
+	CAPS_LO_2P5GBASET_FD,
+	CAPS_LO_5GBASET_FD        = 10,
+	CAPS_LO_10GBASET_FD,
+	CAPS_LO_AUTONEG,
+	CAPS_LO_SMBUS_READ,
+	CAPS_LO_SMBUS_WRITE,
+	CAPS_LO_MACSEC            = 15,
+	CAPS_LO_RESERVED1,
+	CAPS_LO_WAKE_ON_LINK_FORCED,
+	CAPS_LO_HIGH_TEMP_WARNING = 29,
+	CAPS_LO_DRIVER_SCRATCHPAD = 30,
+	CAPS_LO_GLOBAL_FAULT      = 31
+};
+
+/* 0x374
+ * Status register
+ */
+enum hw_atl_fw2x_caps_hi {
+	CAPS_HI_TPO2EN            = 0,
+	CAPS_HI_10BASET_EEE,
+	CAPS_HI_RESERVED2,
+	CAPS_HI_PAUSE,
+	CAPS_HI_ASYMMETRIC_PAUSE,
+	CAPS_HI_100BASETX_EEE     = 5,
+	CAPS_HI_PHY_BUF_SEND,
+	CAPS_HI_PHY_BUF_RECV,
+	CAPS_HI_1000BASET_FD_EEE,
+	CAPS_HI_2P5GBASET_FD_EEE,
+	CAPS_HI_5GBASET_FD_EEE    = 10,
+	CAPS_HI_10GBASET_FD_EEE,
+	CAPS_HI_FW_REQUEST,
+	CAPS_HI_PHY_LOG,
+	CAPS_HI_EEE_AUTO_DISABLE_SETTINGS,
+	CAPS_HI_PFC               = 15,
+	CAPS_HI_WAKE_ON_LINK,
+	CAPS_HI_CABLE_DIAG,
+	CAPS_HI_TEMPERATURE,
+	CAPS_HI_DOWNSHIFT,
+	CAPS_HI_PTP_AVB_EN_FW2X   = 20,
+	CAPS_HI_THERMAL_SHUTDOWN,
+	CAPS_HI_LINK_DROP,
+	CAPS_HI_SLEEP_PROXY,
+	CAPS_HI_WOL,
+	CAPS_HI_MAC_STOP          = 25,
+	CAPS_HI_EXT_LOOPBACK,
+	CAPS_HI_INT_LOOPBACK,
+	CAPS_HI_EFUSE_AGENT,
+	CAPS_HI_WOL_TIMER,
+	CAPS_HI_STATISTICS        = 30,
+	CAPS_HI_TRANSACTION_ID,
+};
+
+/* 0x36C
+ * Control register
+ */
+enum hw_atl_fw2x_ctrl {
+	CTRL_RESERVED1            = 0,
+	CTRL_RESERVED2,
+	CTRL_RESERVED3,
+	CTRL_PAUSE,
+	CTRL_ASYMMETRIC_PAUSE,
+	CTRL_RESERVED4            = 5,
+	CTRL_RESERVED5,
+	CTRL_RESERVED6,
+	CTRL_1GBASET_FD_EEE,
+	CTRL_2P5GBASET_FD_EEE,
+	CTRL_5GBASET_FD_EEE       = 10,
+	CTRL_10GBASET_FD_EEE,
+	CTRL_THERMAL_SHUTDOWN,
+	CTRL_PHY_LOGS,
+	CTRL_EEE_AUTO_DISABLE,
+	CTRL_PFC                  = 15,
+	CTRL_WAKE_ON_LINK,
+	CTRL_CABLE_DIAG,
+	CTRL_TEMPERATURE,
+	CTRL_DOWNSHIFT,
+	CTRL_PTP_AVB              = 20,
+	CTRL_RESERVED7,
+	CTRL_LINK_DROP,
+	CTRL_SLEEP_PROXY,
+	CTRL_WOL,
+	CTRL_MAC_STOP             = 25,
+	CTRL_EXT_LOOPBACK,
+	CTRL_INT_LOOPBACK,
+	CTRL_RESERVED8,
+	CTRL_WOL_TIMER,
+	CTRL_STATISTICS           = 30,
+	CTRL_FORCE_RECONNECT,
+};
+
+enum hw_atl_caps_ex {
+	CAPS_EX_LED_CONTROL       =  0,
+	CAPS_EX_LED0_MODE_LO,
+	CAPS_EX_LED0_MODE_HI,
+	CAPS_EX_LED1_MODE_LO,
+	CAPS_EX_LED1_MODE_HI,
+	CAPS_EX_LED2_MODE_LO      =  5,
+	CAPS_EX_LED2_MODE_HI,
+	CAPS_EX_RESERVED07,
+	CAPS_EX_RESERVED08,
+	CAPS_EX_RESERVED09,
+	CAPS_EX_RESERVED10        = 10,
+	CAPS_EX_RESERVED11,
+	CAPS_EX_RESERVED12,
+	CAPS_EX_RESERVED13,
+	CAPS_EX_RESERVED14,
+	CAPS_EX_RESERVED15        = 15,
+	CAPS_EX_PHY_PTP_EN,
+	CAPS_EX_MAC_PTP_EN,
+	CAPS_EX_EXT_CLK_EN,
+	CAPS_EX_SCHED_DMA_EN,
+	CAPS_EX_PTP_GPIO_EN       = 20,
+	CAPS_EX_UPDATE_SETTINGS,
+	CAPS_EX_PHY_CTRL_TS_PIN,
+	CAPS_EX_SNR_OPERATING_MARGIN,
+	CAPS_EX_RESERVED24,
+	CAPS_EX_RESERVED25        = 25,
+	CAPS_EX_RESERVED26,
+	CAPS_EX_RESERVED27,
+	CAPS_EX_RESERVED28,
+	CAPS_EX_RESERVED29,
+	CAPS_EX_RESERVED30        = 30,
+	CAPS_EX_RESERVED31
+};
+
+struct aq_hw_s;
+struct aq_fw_ops;
+struct aq_hw_caps_s;
+struct aq_hw_link_status_s;
+
+int hw_atl_utils_initfw(struct aq_hw_s *self, const struct aq_fw_ops **fw_ops);
+
+int hw_atl_utils_soft_reset(struct aq_hw_s *self);
+
 void hw_atl_utils_hw_chip_features_init(struct aq_hw_s *self, u32 *p);
 
+int hw_atl_utils_mpi_read_mbox(struct aq_hw_s *self,
+			       struct hw_atl_utils_mbox_header *pmbox);
+
 void hw_atl_utils_mpi_read_stats(struct aq_hw_s *self,
-				 struct hw_aq_atl_utils_mbox *pmbox);
+				 struct hw_atl_utils_mbox *pmbox);
 
 void hw_atl_utils_mpi_set(struct aq_hw_s *self,
 			  enum hal_atl_utils_fw_state_e state,
 			  u32 speed);
 
-int hw_atl_utils_mpi_set_speed(struct aq_hw_s *self, u32 speed,
-			       enum hal_atl_utils_fw_state_e state);
-
-int hw_atl_utils_mpi_get_link_status(struct aq_hw_s *self,
-				     struct aq_hw_link_status_s *link_status);
+int hw_atl_utils_mpi_get_link_status(struct aq_hw_s *self);
 
 int hw_atl_utils_get_mac_permanent(struct aq_hw_s *self,
-				   struct aq_hw_caps_s *aq_hw_caps,
 				   u8 *mac);
 
 unsigned int hw_atl_utils_mbps_2_speed_index(unsigned int mbps);
 
 int hw_atl_utils_hw_get_regs(struct aq_hw_s *self,
-			     struct aq_hw_caps_s *aq_hw_caps,
+			     const struct aq_hw_caps_s *aq_hw_caps,
 			     u32 *regs_buff);
 
-int hw_atl_utils_hw_get_settings(struct aq_hw_s *self,
-				 struct ethtool_cmd *cmd);
-
 int hw_atl_utils_hw_set_power(struct aq_hw_s *self,
 			      unsigned int power_state);
 
 int hw_atl_utils_hw_deinit(struct aq_hw_s *self);
 
-int hw_atl_utils_get_fw_version(struct aq_hw_s *self, u32 *fw_version);
+u32 hw_atl_utils_get_fw_version(struct aq_hw_s *self);
+
+int hw_atl_utils_update_stats(struct aq_hw_s *self);
+
+struct aq_stats_s *hw_atl_utils_get_hw_stats(struct aq_hw_s *self);
+
+int hw_atl_utils_fw_downld_dwords(struct aq_hw_s *self, u32 a,
+				  u32 *p, u32 cnt);
+
+int hw_atl_write_fwcfg_dwords(struct aq_hw_s *self, u32 *p, u32 cnt);
+
+int hw_atl_write_fwsettings_dwords(struct aq_hw_s *self, u32 offset, u32 *p,
+				   u32 cnt);
+
+int hw_atl_utils_fw_set_wol(struct aq_hw_s *self, bool wol_enabled, u8 *mac);
+
+int hw_atl_utils_fw_rpc_call(struct aq_hw_s *self, unsigned int rpc_size);
+
+int hw_atl_utils_fw_rpc_wait(struct aq_hw_s *self,
+			     struct hw_atl_utils_fw_rpc **rpc);
+
+bool hw_atl_utils_ver_match(u32 ver_expected, u32 ver_actual);
 
-int hw_atl_utils_get_hw_stats(struct aq_hw_s *self,
-			      u64 *data,
-			      unsigned int *p_count);
+extern const struct aq_fw_ops aq_fw_1x_ops;
+extern const struct aq_fw_ops aq_fw_2x_ops;
 
 #endif /* HW_ATL_UTILS_H */
diff --git a/drivers/net/ethernet/aquantia/atlantic/hw_atl/hw_atl_utils_fw2x.c b/drivers/net/ethernet/aquantia/atlantic/hw_atl/hw_atl_utils_fw2x.c
new file mode 100644
index 000000000000..4d4cfbc91e19
--- /dev/null
+++ b/drivers/net/ethernet/aquantia/atlantic/hw_atl/hw_atl_utils_fw2x.c
@@ -0,0 +1,732 @@
+// SPDX-License-Identifier: GPL-2.0-only
+/* Atlantic Network Driver
+ *
+ * Copyright (C) 2014-2019 aQuantia Corporation
+ * Copyright (C) 2019-2020 Marvell International Ltd.
+ */
+
+/* File hw_atl_utils_fw2x.c: Definition of firmware 2.x functions for
+ * Atlantic hardware abstraction layer.
+ */
+
+#include "../aq_hw.h"
+#include "../aq_hw_utils.h"
+#include "../aq_pci_func.h"
+#include "../aq_ring.h"
+#include "../aq_vec.h"
+#include "../aq_nic.h"
+#include "hw_atl_utils.h"
+#include "hw_atl_llh.h"
+
+#define HW_ATL_FW2X_MPI_LED_ADDR         0x31c
+#define HW_ATL_FW2X_MPI_RPC_ADDR         0x334
+
+#define HW_ATL_FW2X_MPI_MBOX_ADDR        0x360
+#define HW_ATL_FW2X_MPI_EFUSE_ADDR       0x364
+#define HW_ATL_FW2X_MPI_CONTROL_ADDR     0x368
+#define HW_ATL_FW2X_MPI_CONTROL2_ADDR    0x36C
+#define HW_ATL_FW2X_MPI_STATE_ADDR       0x370
+#define HW_ATL_FW2X_MPI_STATE2_ADDR      0x374
+
+#define HW_ATL_FW3X_EXT_CONTROL_ADDR     0x378
+#define HW_ATL_FW3X_EXT_STATE_ADDR       0x37c
+
+#define HW_ATL_FW3X_PTP_ADJ_LSW_ADDR	 0x50a0
+#define HW_ATL_FW3X_PTP_ADJ_MSW_ADDR	 0x50a4
+
+#define HW_ATL_FW2X_CAP_PAUSE            BIT(CAPS_HI_PAUSE)
+#define HW_ATL_FW2X_CAP_ASYM_PAUSE       BIT(CAPS_HI_ASYMMETRIC_PAUSE)
+#define HW_ATL_FW2X_CAP_SLEEP_PROXY      BIT(CAPS_HI_SLEEP_PROXY)
+#define HW_ATL_FW2X_CAP_WOL              BIT(CAPS_HI_WOL)
+
+#define HW_ATL_FW2X_CTRL_WAKE_ON_LINK     BIT(CTRL_WAKE_ON_LINK)
+#define HW_ATL_FW2X_CTRL_SLEEP_PROXY      BIT(CTRL_SLEEP_PROXY)
+#define HW_ATL_FW2X_CTRL_WOL              BIT(CTRL_WOL)
+#define HW_ATL_FW2X_CTRL_LINK_DROP        BIT(CTRL_LINK_DROP)
+#define HW_ATL_FW2X_CTRL_PAUSE            BIT(CTRL_PAUSE)
+#define HW_ATL_FW2X_CTRL_TEMPERATURE      BIT(CTRL_TEMPERATURE)
+#define HW_ATL_FW2X_CTRL_ASYMMETRIC_PAUSE BIT(CTRL_ASYMMETRIC_PAUSE)
+#define HW_ATL_FW2X_CTRL_INT_LOOPBACK     BIT(CTRL_INT_LOOPBACK)
+#define HW_ATL_FW2X_CTRL_EXT_LOOPBACK     BIT(CTRL_EXT_LOOPBACK)
+#define HW_ATL_FW2X_CTRL_DOWNSHIFT        BIT(CTRL_DOWNSHIFT)
+#define HW_ATL_FW2X_CTRL_FORCE_RECONNECT  BIT(CTRL_FORCE_RECONNECT)
+
+#define HW_ATL_FW2X_CAP_EEE_1G_MASK      BIT(CAPS_HI_1000BASET_FD_EEE)
+#define HW_ATL_FW2X_CAP_EEE_2G5_MASK     BIT(CAPS_HI_2P5GBASET_FD_EEE)
+#define HW_ATL_FW2X_CAP_EEE_5G_MASK      BIT(CAPS_HI_5GBASET_FD_EEE)
+#define HW_ATL_FW2X_CAP_EEE_10G_MASK     BIT(CAPS_HI_10GBASET_FD_EEE)
+
+#define HW_ATL_FW2X_CAP_MACSEC           BIT(CAPS_LO_MACSEC)
+
+#define HAL_ATLANTIC_WOL_FILTERS_COUNT   8
+#define HAL_ATLANTIC_UTILS_FW2X_MSG_WOL  0x0E
+
+#define HW_ATL_FW_VER_LED                0x03010026U
+#define HW_ATL_FW_VER_MEDIA_CONTROL      0x0301005aU
+
+struct __packed fw2x_msg_wol_pattern {
+	u8 mask[16];
+	u32 crc;
+};
+
+struct __packed fw2x_msg_wol {
+	u32 msg_id;
+	u8 hw_addr[ETH_ALEN];
+	u8 magic_packet_enabled;
+	u8 filter_count;
+	struct fw2x_msg_wol_pattern filter[HAL_ATLANTIC_WOL_FILTERS_COUNT];
+	u8 link_up_enabled;
+	u8 link_down_enabled;
+	u16 reserved;
+	u32 link_up_timeout;
+	u32 link_down_timeout;
+};
+
+static int aq_fw2x_set_link_speed(struct aq_hw_s *self, u32 speed);
+static int aq_fw2x_set_state(struct aq_hw_s *self,
+			     enum hal_atl_utils_fw_state_e state);
+
+static u32 aq_fw2x_mbox_get(struct aq_hw_s *self);
+static u32 aq_fw2x_rpc_get(struct aq_hw_s *self);
+static int aq_fw2x_settings_get(struct aq_hw_s *self, u32 *addr);
+static u32 aq_fw2x_state_get(struct aq_hw_s *self);
+static u32 aq_fw2x_state2_get(struct aq_hw_s *self);
+
+static int aq_fw2x_init(struct aq_hw_s *self)
+{
+	int err = 0;
+
+	/* check 10 times by 1ms */
+	err = readx_poll_timeout_atomic(aq_fw2x_mbox_get,
+					self, self->mbox_addr,
+					self->mbox_addr != 0U,
+					1000U, 10000U);
+
+	err = readx_poll_timeout_atomic(aq_fw2x_rpc_get,
+					self, self->rpc_addr,
+					self->rpc_addr != 0U,
+					1000U, 100000U);
+
+	err = aq_fw2x_settings_get(self, &self->settings_addr);
+
+	return err;
+}
+
+static int aq_fw2x_deinit(struct aq_hw_s *self)
+{
+	int err = aq_fw2x_set_link_speed(self, 0);
+
+	if (!err)
+		err = aq_fw2x_set_state(self, MPI_DEINIT);
+
+	return err;
+}
+
+static enum hw_atl_fw2x_rate link_speed_mask_2fw2x_ratemask(u32 speed)
+{
+	enum hw_atl_fw2x_rate rate = 0;
+
+	if (speed & AQ_NIC_RATE_10G)
+		rate |= FW2X_RATE_10G;
+
+	if (speed & AQ_NIC_RATE_5G)
+		rate |= FW2X_RATE_5G;
+
+	if (speed & AQ_NIC_RATE_2G5)
+		rate |= FW2X_RATE_2G5;
+
+	if (speed & AQ_NIC_RATE_1G)
+		rate |= FW2X_RATE_1G;
+
+	if (speed & AQ_NIC_RATE_100M)
+		rate |= FW2X_RATE_100M;
+
+	return rate;
+}
+
+static u32 fw2x_to_eee_mask(u32 speed)
+{
+	u32 rate = 0;
+
+	if (speed & HW_ATL_FW2X_CAP_EEE_10G_MASK)
+		rate |= AQ_NIC_RATE_EEE_10G;
+	if (speed & HW_ATL_FW2X_CAP_EEE_5G_MASK)
+		rate |= AQ_NIC_RATE_EEE_5G;
+	if (speed & HW_ATL_FW2X_CAP_EEE_2G5_MASK)
+		rate |= AQ_NIC_RATE_EEE_2G5;
+	if (speed & HW_ATL_FW2X_CAP_EEE_1G_MASK)
+		rate |= AQ_NIC_RATE_EEE_1G;
+
+	return rate;
+}
+
+static u32 eee_mask_to_fw2x(u32 speed)
+{
+	u32 rate = 0;
+
+	if (speed & AQ_NIC_RATE_EEE_10G)
+		rate |= HW_ATL_FW2X_CAP_EEE_10G_MASK;
+	if (speed & AQ_NIC_RATE_EEE_5G)
+		rate |= HW_ATL_FW2X_CAP_EEE_5G_MASK;
+	if (speed & AQ_NIC_RATE_EEE_2G5)
+		rate |= HW_ATL_FW2X_CAP_EEE_2G5_MASK;
+	if (speed & AQ_NIC_RATE_EEE_1G)
+		rate |= HW_ATL_FW2X_CAP_EEE_1G_MASK;
+
+	return rate;
+}
+
+static int aq_fw2x_set_link_speed(struct aq_hw_s *self, u32 speed)
+{
+	u32 val = link_speed_mask_2fw2x_ratemask(speed);
+
+	aq_hw_write_reg(self, HW_ATL_FW2X_MPI_CONTROL_ADDR, val);
+
+	return 0;
+}
+
+static void aq_fw2x_upd_flow_control_bits(struct aq_hw_s *self,
+					  u32 *mpi_state, u32 fc)
+{
+	*mpi_state &= ~(HW_ATL_FW2X_CTRL_PAUSE |
+			HW_ATL_FW2X_CTRL_ASYMMETRIC_PAUSE);
+
+	switch (fc) {
+	/* There is not explicit mode of RX only pause frames,
+	 * thus, we join this mode with FC full.
+	 * FC full is either Rx, either Tx, or both.
+	 */
+	case AQ_NIC_FC_FULL:
+	case AQ_NIC_FC_RX:
+		*mpi_state |= HW_ATL_FW2X_CTRL_PAUSE |
+			      HW_ATL_FW2X_CTRL_ASYMMETRIC_PAUSE;
+		break;
+	case AQ_NIC_FC_TX:
+		*mpi_state |= HW_ATL_FW2X_CTRL_ASYMMETRIC_PAUSE;
+		break;
+	}
+}
+
+static void aq_fw2x_upd_eee_rate_bits(struct aq_hw_s *self, u32 *mpi_opts,
+				      u32 eee_speeds)
+{
+	*mpi_opts &= ~(HW_ATL_FW2X_CAP_EEE_1G_MASK |
+		       HW_ATL_FW2X_CAP_EEE_2G5_MASK |
+		       HW_ATL_FW2X_CAP_EEE_5G_MASK |
+		       HW_ATL_FW2X_CAP_EEE_10G_MASK);
+
+	*mpi_opts |= eee_mask_to_fw2x(eee_speeds);
+}
+
+static int aq_fw2x_set_state(struct aq_hw_s *self,
+			     enum hal_atl_utils_fw_state_e state)
+{
+	u32 mpi_state = aq_hw_read_reg(self, HW_ATL_FW2X_MPI_CONTROL2_ADDR);
+	struct aq_nic_cfg_s *cfg = self->aq_nic_cfg;
+
+	switch (state) {
+	case MPI_INIT:
+		mpi_state &= ~BIT(CAPS_HI_LINK_DROP);
+		aq_fw2x_upd_eee_rate_bits(self, &mpi_state, cfg->eee_speeds);
+		aq_fw2x_upd_flow_control_bits(self, &mpi_state,
+					      self->aq_nic_cfg->fc.req);
+		break;
+	case MPI_DEINIT:
+		mpi_state |= BIT(CAPS_HI_LINK_DROP);
+		break;
+	case MPI_RESET:
+	case MPI_POWER:
+		/* No actions */
+		break;
+	}
+	aq_hw_write_reg(self, HW_ATL_FW2X_MPI_CONTROL2_ADDR, mpi_state);
+
+	return 0;
+}
+
+static int aq_fw2x_update_link_status(struct aq_hw_s *self)
+{
+	struct aq_hw_link_status_s *link_status = &self->aq_link_status;
+	u32 mpi_state;
+	u32 speed;
+
+	mpi_state = aq_hw_read_reg(self, HW_ATL_FW2X_MPI_STATE_ADDR);
+	speed = mpi_state & (FW2X_RATE_100M | FW2X_RATE_1G |
+			     FW2X_RATE_2G5 | FW2X_RATE_5G |
+			     FW2X_RATE_10G);
+
+	if (speed) {
+		if (speed & FW2X_RATE_10G)
+			link_status->mbps = 10000;
+		else if (speed & FW2X_RATE_5G)
+			link_status->mbps = 5000;
+		else if (speed & FW2X_RATE_2G5)
+			link_status->mbps = 2500;
+		else if (speed & FW2X_RATE_1G)
+			link_status->mbps = 1000;
+		else if (speed & FW2X_RATE_100M)
+			link_status->mbps = 100;
+		else
+			link_status->mbps = 10000;
+	} else {
+		link_status->mbps = 0;
+	}
+	link_status->full_duplex = true;
+
+	return 0;
+}
+
+static int aq_fw2x_get_mac_permanent(struct aq_hw_s *self, u8 *mac)
+{
+	u32 efuse_addr = aq_hw_read_reg(self, HW_ATL_FW2X_MPI_EFUSE_ADDR);
+	u32 mac_addr[2] = { 0 };
+	int err = 0;
+
+	if (efuse_addr != 0) {
+		err = hw_atl_utils_fw_downld_dwords(self,
+						    efuse_addr + (40U * 4U),
+						    mac_addr,
+						    ARRAY_SIZE(mac_addr));
+		if (err)
+			return err;
+		mac_addr[0] = __swab32(mac_addr[0]);
+		mac_addr[1] = __swab32(mac_addr[1]);
+	}
+
+	ether_addr_copy(mac, (u8 *)mac_addr);
+
+	return err;
+}
+
+static int aq_fw2x_update_stats(struct aq_hw_s *self)
+{
+	u32 mpi_opts = aq_hw_read_reg(self, HW_ATL_FW2X_MPI_CONTROL2_ADDR);
+	u32 orig_stats_val = mpi_opts & BIT(CAPS_HI_STATISTICS);
+	u32 stats_val;
+	int err = 0;
+
+	/* Toggle statistics bit for FW to update */
+	mpi_opts = mpi_opts ^ BIT(CAPS_HI_STATISTICS);
+	aq_hw_write_reg(self, HW_ATL_FW2X_MPI_CONTROL2_ADDR, mpi_opts);
+
+	/* Wait FW to report back */
+	err = readx_poll_timeout_atomic(aq_fw2x_state2_get,
+					self, stats_val,
+					orig_stats_val != (stats_val &
+					BIT(CAPS_HI_STATISTICS)),
+					1U, 10000U);
+	if (err)
+		return err;
+
+	return hw_atl_utils_update_stats(self);
+}
+
+static int aq_fw2x_get_phy_temp(struct aq_hw_s *self, int *temp)
+{
+	u32 mpi_opts = aq_hw_read_reg(self, HW_ATL_FW2X_MPI_CONTROL2_ADDR);
+	u32 temp_val = mpi_opts & HW_ATL_FW2X_CTRL_TEMPERATURE;
+	u32 phy_temp_offset;
+	u32 temp_res;
+	int err = 0;
+	u32 val;
+
+	phy_temp_offset = self->mbox_addr + offsetof(struct hw_atl_utils_mbox,
+						     info.phy_temperature);
+
+	/* Toggle statistics bit for FW to 0x36C.18 (CTRL_TEMPERATURE) */
+	mpi_opts = mpi_opts ^ HW_ATL_FW2X_CTRL_TEMPERATURE;
+	aq_hw_write_reg(self, HW_ATL_FW2X_MPI_CONTROL2_ADDR, mpi_opts);
+	/* Wait FW to report back */
+	err = readx_poll_timeout_atomic(aq_fw2x_state2_get, self, val,
+					temp_val !=
+					(val & HW_ATL_FW2X_CTRL_TEMPERATURE),
+					1U, 10000U);
+	err = hw_atl_utils_fw_downld_dwords(self, phy_temp_offset,
+					    &temp_res, 1);
+
+	if (err)
+		return err;
+
+	/* Convert PHY temperature from 1/256 degree Celsius
+	 * to 1/1000 degree Celsius.
+	 */
+	*temp = (int16_t)(temp_res & 0xFFFF) * 1000 / 256;
+
+	return 0;
+}
+
+static int aq_fw2x_set_wol(struct aq_hw_s *self, const u8 *mac)
+{
+	struct hw_atl_utils_fw_rpc *rpc = NULL;
+	struct offload_info *info = NULL;
+	u32 wol_bits = 0;
+	u32 rpc_size;
+	int err = 0;
+	u32 val;
+
+	if (self->aq_nic_cfg->wol & WAKE_PHY) {
+		aq_hw_write_reg(self, HW_ATL_FW2X_MPI_CONTROL2_ADDR,
+				HW_ATL_FW2X_CTRL_LINK_DROP);
+		readx_poll_timeout_atomic(aq_fw2x_state2_get, self, val,
+					  (val &
+					   HW_ATL_FW2X_CTRL_LINK_DROP) != 0,
+					  1000, 100000);
+		wol_bits |= HW_ATL_FW2X_CTRL_WAKE_ON_LINK;
+	}
+
+	if (self->aq_nic_cfg->wol & WAKE_MAGIC) {
+		wol_bits |= HW_ATL_FW2X_CTRL_SLEEP_PROXY |
+			    HW_ATL_FW2X_CTRL_WOL;
+
+		err = hw_atl_utils_fw_rpc_wait(self, &rpc);
+		if (err < 0)
+			goto err_exit;
+
+		rpc_size = sizeof(*info) +
+			   offsetof(struct hw_atl_utils_fw_rpc, fw2x_offloads);
+		memset(rpc, 0, rpc_size);
+		info = &rpc->fw2x_offloads;
+		memcpy(info->mac_addr, mac, ETH_ALEN);
+		info->len = sizeof(*info);
+
+		err = hw_atl_utils_fw_rpc_call(self, rpc_size);
+		if (err < 0)
+			goto err_exit;
+	}
+
+	aq_hw_write_reg(self, HW_ATL_FW2X_MPI_CONTROL2_ADDR, wol_bits);
+
+err_exit:
+	return err;
+}
+
+static int aq_fw2x_set_power(struct aq_hw_s *self, unsigned int power_state,
+			     const u8 *mac)
+{
+	int err = 0;
+
+	if (self->aq_nic_cfg->wol)
+		err = aq_fw2x_set_wol(self, mac);
+
+	return err;
+}
+
+static int aq_fw2x_send_fw_request(struct aq_hw_s *self,
+				   const struct hw_fw_request_iface *fw_req,
+				   size_t size)
+{
+	u32 ctrl2, orig_ctrl2;
+	u32 dword_cnt;
+	int err = 0;
+	u32 val;
+
+	/* Write data to drvIface Mailbox */
+	dword_cnt = size / sizeof(u32);
+	if (size % sizeof(u32))
+		dword_cnt++;
+	err = hw_atl_write_fwcfg_dwords(self, (void *)fw_req, dword_cnt);
+	if (err < 0)
+		goto err_exit;
+
+	/* Toggle statistics bit for FW to update */
+	ctrl2 = aq_hw_read_reg(self, HW_ATL_FW2X_MPI_CONTROL2_ADDR);
+	orig_ctrl2 = ctrl2 & BIT(CAPS_HI_FW_REQUEST);
+	ctrl2 = ctrl2 ^ BIT(CAPS_HI_FW_REQUEST);
+	aq_hw_write_reg(self, HW_ATL_FW2X_MPI_CONTROL2_ADDR, ctrl2);
+
+	/* Wait FW to report back */
+	err = readx_poll_timeout_atomic(aq_fw2x_state2_get, self, val,
+					orig_ctrl2 != (val &
+						       BIT(CAPS_HI_FW_REQUEST)),
+					1U, 10000U);
+
+err_exit:
+	return err;
+}
+
+static void aq_fw3x_enable_ptp(struct aq_hw_s *self, int enable)
+{
+	u32 ptp_opts = aq_hw_read_reg(self, HW_ATL_FW3X_EXT_STATE_ADDR);
+	u32 all_ptp_features = BIT(CAPS_EX_PHY_PTP_EN) |
+						   BIT(CAPS_EX_PTP_GPIO_EN);
+
+	if (enable)
+		ptp_opts |= all_ptp_features;
+	else
+		ptp_opts &= ~all_ptp_features;
+
+	aq_hw_write_reg(self, HW_ATL_FW3X_EXT_CONTROL_ADDR, ptp_opts);
+}
+
+static void aq_fw3x_adjust_ptp(struct aq_hw_s *self, uint64_t adj)
+{
+	aq_hw_write_reg(self, HW_ATL_FW3X_PTP_ADJ_LSW_ADDR,
+			(adj >>  0) & 0xffffffff);
+	aq_hw_write_reg(self, HW_ATL_FW3X_PTP_ADJ_MSW_ADDR,
+			(adj >> 32) & 0xffffffff);
+}
+
+static int aq_fw2x_led_control(struct aq_hw_s *self, u32 mode)
+{
+	if (self->fw_ver_actual < HW_ATL_FW_VER_LED)
+		return -EOPNOTSUPP;
+
+	aq_hw_write_reg(self, HW_ATL_FW2X_MPI_LED_ADDR, mode);
+
+	return 0;
+}
+
+static int aq_fw2x_set_eee_rate(struct aq_hw_s *self, u32 speed)
+{
+	u32 mpi_opts = aq_hw_read_reg(self, HW_ATL_FW2X_MPI_CONTROL2_ADDR);
+
+	aq_fw2x_upd_eee_rate_bits(self, &mpi_opts, speed);
+
+	aq_hw_write_reg(self, HW_ATL_FW2X_MPI_CONTROL2_ADDR, mpi_opts);
+
+	return 0;
+}
+
+static int aq_fw2x_get_eee_rate(struct aq_hw_s *self, u32 *rate,
+				u32 *supported_rates)
+{
+	u32 mpi_state;
+	u32 caps_hi;
+	int err = 0;
+	u32 offset;
+
+	offset = self->mbox_addr + offsetof(struct hw_atl_utils_mbox,
+					    info.caps_hi);
+
+	err = hw_atl_utils_fw_downld_dwords(self, offset, &caps_hi, 1);
+
+	if (err)
+		return err;
+
+	*supported_rates = fw2x_to_eee_mask(caps_hi);
+
+	mpi_state = aq_fw2x_state2_get(self);
+	*rate = fw2x_to_eee_mask(mpi_state);
+
+	return err;
+}
+
+static int aq_fw2x_renegotiate(struct aq_hw_s *self)
+{
+	u32 mpi_opts = aq_hw_read_reg(self, HW_ATL_FW2X_MPI_CONTROL2_ADDR);
+
+	mpi_opts |= BIT(CTRL_FORCE_RECONNECT);
+
+	aq_hw_write_reg(self, HW_ATL_FW2X_MPI_CONTROL2_ADDR, mpi_opts);
+
+	return 0;
+}
+
+static int aq_fw2x_set_flow_control(struct aq_hw_s *self)
+{
+	u32 mpi_state = aq_hw_read_reg(self, HW_ATL_FW2X_MPI_CONTROL2_ADDR);
+
+	aq_fw2x_upd_flow_control_bits(self, &mpi_state,
+				      self->aq_nic_cfg->fc.req);
+
+	aq_hw_write_reg(self, HW_ATL_FW2X_MPI_CONTROL2_ADDR, mpi_state);
+
+	return 0;
+}
+
+static u32 aq_fw2x_get_flow_control(struct aq_hw_s *self, u32 *fcmode)
+{
+	u32 mpi_state = aq_fw2x_state2_get(self);
+	*fcmode = 0;
+
+	if (mpi_state & HW_ATL_FW2X_CAP_PAUSE)
+		*fcmode |= AQ_NIC_FC_RX;
+
+	if (mpi_state & HW_ATL_FW2X_CAP_ASYM_PAUSE)
+		*fcmode |= AQ_NIC_FC_TX;
+
+	return 0;
+}
+
+static int aq_fw2x_set_phyloopback(struct aq_hw_s *self, u32 mode, bool enable)
+{
+	u32 mpi_opts;
+
+	switch (mode) {
+	case AQ_HW_LOOPBACK_PHYINT_SYS:
+		mpi_opts = aq_hw_read_reg(self, HW_ATL_FW2X_MPI_CONTROL2_ADDR);
+		if (enable)
+			mpi_opts |= HW_ATL_FW2X_CTRL_INT_LOOPBACK;
+		else
+			mpi_opts &= ~HW_ATL_FW2X_CTRL_INT_LOOPBACK;
+		aq_hw_write_reg(self, HW_ATL_FW2X_MPI_CONTROL2_ADDR, mpi_opts);
+		break;
+	case AQ_HW_LOOPBACK_PHYEXT_SYS:
+		mpi_opts = aq_hw_read_reg(self, HW_ATL_FW2X_MPI_CONTROL2_ADDR);
+		if (enable)
+			mpi_opts |= HW_ATL_FW2X_CTRL_EXT_LOOPBACK;
+		else
+			mpi_opts &= ~HW_ATL_FW2X_CTRL_EXT_LOOPBACK;
+		aq_hw_write_reg(self, HW_ATL_FW2X_MPI_CONTROL2_ADDR, mpi_opts);
+		break;
+	default:
+		return -EINVAL;
+	}
+
+	return 0;
+}
+
+static u32 aq_fw2x_mbox_get(struct aq_hw_s *self)
+{
+	return aq_hw_read_reg(self, HW_ATL_FW2X_MPI_MBOX_ADDR);
+}
+
+static u32 aq_fw2x_rpc_get(struct aq_hw_s *self)
+{
+	return aq_hw_read_reg(self, HW_ATL_FW2X_MPI_RPC_ADDR);
+}
+
+static int aq_fw2x_settings_get(struct aq_hw_s *self, u32 *addr)
+{
+	int err = 0;
+	u32 offset;
+
+	offset = self->mbox_addr + offsetof(struct hw_atl_utils_mbox,
+					    info.setting_address);
+
+	err = hw_atl_utils_fw_downld_dwords(self, offset, addr, 1);
+
+	return err;
+}
+
+static u32 aq_fw2x_state_get(struct aq_hw_s *self)
+{
+	return aq_hw_read_reg(self, HW_ATL_FW2X_MPI_STATE_ADDR);
+}
+
+static u32 aq_fw2x_state2_get(struct aq_hw_s *self)
+{
+	return aq_hw_read_reg(self, HW_ATL_FW2X_MPI_STATE2_ADDR);
+}
+
+static int aq_fw2x_set_downshift(struct aq_hw_s *self, u32 counter)
+{
+	int err = 0;
+	u32 mpi_opts;
+	u32 offset;
+
+	offset = offsetof(struct hw_atl_utils_settings, downshift_retry_count);
+	err = hw_atl_write_fwsettings_dwords(self, offset, &counter, 1);
+	if (err)
+		return err;
+
+	mpi_opts = aq_hw_read_reg(self, HW_ATL_FW2X_MPI_CONTROL2_ADDR);
+	if (counter)
+		mpi_opts |= HW_ATL_FW2X_CTRL_DOWNSHIFT;
+	else
+		mpi_opts &= ~HW_ATL_FW2X_CTRL_DOWNSHIFT;
+	aq_hw_write_reg(self, HW_ATL_FW2X_MPI_CONTROL2_ADDR, mpi_opts);
+
+	return err;
+}
+
+static int aq_fw2x_set_media_detect(struct aq_hw_s *self, bool on)
+{
+	u32 enable;
+	u32 offset;
+
+	if (self->fw_ver_actual < HW_ATL_FW_VER_MEDIA_CONTROL)
+		return -EOPNOTSUPP;
+
+	offset = offsetof(struct hw_atl_utils_settings, media_detect);
+	enable = on;
+
+	return hw_atl_write_fwsettings_dwords(self, offset, &enable, 1);
+}
+
+static u32 aq_fw2x_get_link_capabilities(struct aq_hw_s *self)
+{
+	int err = 0;
+	u32 offset;
+	u32 val;
+
+	offset = self->mbox_addr +
+		 offsetof(struct hw_atl_utils_mbox, info.caps_lo);
+
+	err = hw_atl_utils_fw_downld_dwords(self, offset, &val, 1);
+
+	if (err)
+		return 0;
+
+	return val;
+}
+
+static int aq_fw2x_send_macsec_req(struct aq_hw_s *hw,
+				   struct macsec_msg_fw_request *req,
+				   struct macsec_msg_fw_response *response)
+{
+	u32 low_status, low_req = 0;
+	u32 dword_cnt;
+	u32 caps_lo;
+	u32 offset;
+	int err;
+
+	if (!req || !response)
+		return -EINVAL;
+
+	caps_lo = aq_fw2x_get_link_capabilities(hw);
+	if (!(caps_lo & BIT(CAPS_LO_MACSEC)))
+		return -EOPNOTSUPP;
+
+	/* Write macsec request to cfg memory */
+	dword_cnt = (sizeof(*req) + sizeof(u32) - 1) / sizeof(u32);
+	err = hw_atl_write_fwcfg_dwords(hw, (void *)req, dword_cnt);
+	if (err < 0)
+		return err;
+
+	/* Toggle 0x368.CAPS_LO_MACSEC bit */
+	low_req = aq_hw_read_reg(hw, HW_ATL_FW2X_MPI_CONTROL_ADDR);
+	low_req ^= HW_ATL_FW2X_CAP_MACSEC;
+	aq_hw_write_reg(hw, HW_ATL_FW2X_MPI_CONTROL_ADDR, low_req);
+
+	/* Wait FW to report back */
+	err = readx_poll_timeout_atomic(aq_fw2x_state_get, hw, low_status,
+		low_req != (low_status & BIT(CAPS_LO_MACSEC)), 1U, 10000U);
+	if (err)
+		return -EIO;
+
+	/* Read status of write operation */
+	offset = hw->rpc_addr + sizeof(u32);
+	err = hw_atl_utils_fw_downld_dwords(hw, offset, (u32 *)(void *)response,
+					    sizeof(*response) / sizeof(u32));
+
+	return err;
+}
+
+const struct aq_fw_ops aq_fw_2x_ops = {
+	.init               = aq_fw2x_init,
+	.deinit             = aq_fw2x_deinit,
+	.reset              = NULL,
+	.renegotiate        = aq_fw2x_renegotiate,
+	.get_mac_permanent  = aq_fw2x_get_mac_permanent,
+	.set_link_speed     = aq_fw2x_set_link_speed,
+	.set_state          = aq_fw2x_set_state,
+	.update_link_status = aq_fw2x_update_link_status,
+	.update_stats       = aq_fw2x_update_stats,
+	.get_mac_temp       = NULL,
+	.get_phy_temp       = aq_fw2x_get_phy_temp,
+	.set_power          = aq_fw2x_set_power,
+	.set_eee_rate       = aq_fw2x_set_eee_rate,
+	.get_eee_rate       = aq_fw2x_get_eee_rate,
+	.set_flow_control   = aq_fw2x_set_flow_control,
+	.get_flow_control   = aq_fw2x_get_flow_control,
+	.send_fw_request    = aq_fw2x_send_fw_request,
+	.enable_ptp         = aq_fw3x_enable_ptp,
+	.led_control        = aq_fw2x_led_control,
+	.set_phyloopback    = aq_fw2x_set_phyloopback,
+	.set_downshift      = aq_fw2x_set_downshift,
+	.set_media_detect   = aq_fw2x_set_media_detect,
+	.adjust_ptp         = aq_fw3x_adjust_ptp,
+	.get_link_capabilities = aq_fw2x_get_link_capabilities,
+	.send_macsec_req    = aq_fw2x_send_macsec_req,
+};
diff --git a/drivers/net/ethernet/aquantia/atlantic/hw_atl2/hw_atl2.c b/drivers/net/ethernet/aquantia/atlantic/hw_atl2/hw_atl2.c
new file mode 100644
index 000000000000..b0ed572e88c6
--- /dev/null
+++ b/drivers/net/ethernet/aquantia/atlantic/hw_atl2/hw_atl2.c
@@ -0,0 +1,863 @@
+// SPDX-License-Identifier: GPL-2.0-only
+/* Atlantic Network Driver
+ * Copyright (C) 2020 Marvell International Ltd.
+ */
+
+#include "aq_hw.h"
+#include "aq_hw_utils.h"
+#include "aq_ring.h"
+#include "aq_nic.h"
+#include "hw_atl/hw_atl_b0.h"
+#include "hw_atl/hw_atl_utils.h"
+#include "hw_atl/hw_atl_llh.h"
+#include "hw_atl/hw_atl_llh_internal.h"
+#include "hw_atl2_utils.h"
+#include "hw_atl2_llh.h"
+#include "hw_atl2_internal.h"
+#include "hw_atl2_llh_internal.h"
+
+static int hw_atl2_act_rslvr_table_set(struct aq_hw_s *self, u8 location,
+				       u32 tag, u32 mask, u32 action);
+
+#define DEFAULT_BOARD_BASIC_CAPABILITIES \
+	.is_64_dma = true,		  \
+	.op64bit = true,		  \
+	.msix_irqs = 8U,		  \
+	.irq_mask = ~0U,		  \
+	.vecs = HW_ATL2_RSS_MAX,	  \
+	.tcs_max = HW_ATL2_TC_MAX,	  \
+	.rxd_alignment = 1U,		  \
+	.rxd_size = HW_ATL2_RXD_SIZE,   \
+	.rxds_max = HW_ATL2_MAX_RXD,    \
+	.rxds_min = HW_ATL2_MIN_RXD,    \
+	.txd_alignment = 1U,		  \
+	.txd_size = HW_ATL2_TXD_SIZE,   \
+	.txds_max = HW_ATL2_MAX_TXD,    \
+	.txds_min = HW_ATL2_MIN_TXD,    \
+	.txhwb_alignment = 4096U,	  \
+	.tx_rings = HW_ATL2_TX_RINGS,   \
+	.rx_rings = HW_ATL2_RX_RINGS,   \
+	.hw_features = NETIF_F_HW_CSUM |  \
+			NETIF_F_RXCSUM |  \
+			NETIF_F_RXHASH |  \
+			NETIF_F_SG |      \
+			NETIF_F_TSO |     \
+			NETIF_F_TSO6 |    \
+			NETIF_F_LRO |     \
+			NETIF_F_NTUPLE |  \
+			NETIF_F_HW_VLAN_CTAG_FILTER | \
+			NETIF_F_HW_VLAN_CTAG_RX |     \
+			NETIF_F_HW_VLAN_CTAG_TX |     \
+			NETIF_F_GSO_UDP_L4      |     \
+			NETIF_F_GSO_PARTIAL     |     \
+			NETIF_F_HW_TC,                \
+	.hw_priv_flags = IFF_UNICAST_FLT, \
+	.flow_control = true,		  \
+	.mtu = HW_ATL2_MTU_JUMBO,	  \
+	.mac_regs_count = 72,		  \
+	.hw_alive_check_addr = 0x10U,     \
+	.priv_data_len = sizeof(struct hw_atl2_priv)
+
+const struct aq_hw_caps_s hw_atl2_caps_aqc113 = {
+	DEFAULT_BOARD_BASIC_CAPABILITIES,
+	.media_type = AQ_HW_MEDIA_TYPE_TP,
+	.link_speed_msk = AQ_NIC_RATE_10G |
+			  AQ_NIC_RATE_5G  |
+			  AQ_NIC_RATE_2G5 |
+			  AQ_NIC_RATE_1G  |
+			  AQ_NIC_RATE_100M      |
+			  AQ_NIC_RATE_10M,
+};
+
+const struct aq_hw_caps_s hw_atl2_caps_aqc115c = {
+	DEFAULT_BOARD_BASIC_CAPABILITIES,
+	.media_type = AQ_HW_MEDIA_TYPE_TP,
+	.link_speed_msk = AQ_NIC_RATE_2G5 |
+			  AQ_NIC_RATE_1G  |
+			  AQ_NIC_RATE_100M      |
+			  AQ_NIC_RATE_10M,
+};
+
+const struct aq_hw_caps_s hw_atl2_caps_aqc116c = {
+	DEFAULT_BOARD_BASIC_CAPABILITIES,
+	.media_type = AQ_HW_MEDIA_TYPE_TP,
+	.link_speed_msk = AQ_NIC_RATE_1G  |
+			  AQ_NIC_RATE_100M      |
+			  AQ_NIC_RATE_10M,
+};
+
+static u32 hw_atl2_sem_act_rslvr_get(struct aq_hw_s *self)
+{
+	return hw_atl_reg_glb_cpu_sem_get(self, HW_ATL2_FW_SM_ACT_RSLVR);
+}
+
+static int hw_atl2_hw_reset(struct aq_hw_s *self)
+{
+	struct hw_atl2_priv *priv = self->priv;
+	int err;
+
+	err = hw_atl2_utils_soft_reset(self);
+	if (err)
+		return err;
+
+	memset(priv, 0, sizeof(*priv));
+
+	self->aq_fw_ops->set_state(self, MPI_RESET);
+
+	err = aq_hw_err_from_flags(self);
+
+	return err;
+}
+
+static int hw_atl2_hw_queue_to_tc_map_set(struct aq_hw_s *self)
+{
+	struct aq_nic_cfg_s *cfg = self->aq_nic_cfg;
+	unsigned int tcs, q_per_tc;
+	unsigned int tc, q;
+	u32 rx_map = 0;
+	u32 tx_map = 0;
+
+	hw_atl2_tpb_tx_tc_q_rand_map_en_set(self, 1U);
+
+	switch (cfg->tc_mode) {
+	case AQ_TC_MODE_8TCS:
+		tcs = 8;
+		q_per_tc = 4;
+		break;
+	case AQ_TC_MODE_4TCS:
+		tcs = 4;
+		q_per_tc = 8;
+		break;
+	default:
+		return -EINVAL;
+	}
+
+	for (tc = 0; tc != tcs; tc++) {
+		unsigned int tc_q_offset = tc * q_per_tc;
+
+		for (q = tc_q_offset; q != tc_q_offset + q_per_tc; q++) {
+			rx_map |= tc << HW_ATL2_RX_Q_TC_MAP_SHIFT(q);
+			if (HW_ATL2_RX_Q_TC_MAP_ADR(q) !=
+			    HW_ATL2_RX_Q_TC_MAP_ADR(q + 1)) {
+				aq_hw_write_reg(self,
+						HW_ATL2_RX_Q_TC_MAP_ADR(q),
+						rx_map);
+				rx_map = 0;
+			}
+
+			tx_map |= tc << HW_ATL2_TX_Q_TC_MAP_SHIFT(q);
+			if (HW_ATL2_TX_Q_TC_MAP_ADR(q) !=
+			    HW_ATL2_TX_Q_TC_MAP_ADR(q + 1)) {
+				aq_hw_write_reg(self,
+						HW_ATL2_TX_Q_TC_MAP_ADR(q),
+						tx_map);
+				tx_map = 0;
+			}
+		}
+	}
+
+	return aq_hw_err_from_flags(self);
+}
+
+static int hw_atl2_hw_qos_set(struct aq_hw_s *self)
+{
+	struct aq_nic_cfg_s *cfg = self->aq_nic_cfg;
+	u32 tx_buff_size = HW_ATL2_TXBUF_MAX;
+	u32 rx_buff_size = HW_ATL2_RXBUF_MAX;
+	unsigned int prio = 0U;
+	u32 tc = 0U;
+
+	/* TPS Descriptor rate init */
+	hw_atl_tps_tx_pkt_shed_desc_rate_curr_time_res_set(self, 0x0U);
+	hw_atl_tps_tx_pkt_shed_desc_rate_lim_set(self, 0xA);
+
+	/* TPS VM init */
+	hw_atl_tps_tx_pkt_shed_desc_vm_arb_mode_set(self, 0U);
+
+	tx_buff_size /= cfg->tcs;
+	rx_buff_size /= cfg->tcs;
+	for (tc = 0; tc < cfg->tcs; tc++) {
+		u32 threshold = 0U;
+
+		/* Tx buf size TC0 */
+		hw_atl_tpb_tx_pkt_buff_size_per_tc_set(self, tx_buff_size, tc);
+
+		threshold = (tx_buff_size * (1024 / 32U) * 66U) / 100U;
+		hw_atl_tpb_tx_buff_hi_threshold_per_tc_set(self, threshold, tc);
+
+		threshold = (tx_buff_size * (1024 / 32U) * 50U) / 100U;
+		hw_atl_tpb_tx_buff_lo_threshold_per_tc_set(self, threshold, tc);
+
+		/* QoS Rx buf size per TC */
+		hw_atl_rpb_rx_pkt_buff_size_per_tc_set(self, rx_buff_size, tc);
+
+		threshold = (rx_buff_size * (1024U / 32U) * 66U) / 100U;
+		hw_atl_rpb_rx_buff_hi_threshold_per_tc_set(self, threshold, tc);
+
+		threshold = (rx_buff_size * (1024U / 32U) * 50U) / 100U;
+		hw_atl_rpb_rx_buff_lo_threshold_per_tc_set(self, threshold, tc);
+
+		hw_atl_b0_set_fc(self, self->aq_nic_cfg->fc.req, tc);
+	}
+
+	/* QoS 802.1p priority -> TC mapping */
+	for (prio = 0; prio < 8; ++prio)
+		hw_atl_rpf_rpb_user_priority_tc_map_set(self, prio,
+							cfg->prio_tc_map[prio]);
+
+	/* ATL2 Apply ring to TC mapping */
+	hw_atl2_hw_queue_to_tc_map_set(self);
+
+	return aq_hw_err_from_flags(self);
+}
+
+static int hw_atl2_hw_rss_set(struct aq_hw_s *self,
+			      struct aq_rss_parameters *rss_params)
+{
+	u8 *indirection_table = rss_params->indirection_table;
+	const u32 num_tcs = aq_hw_num_tcs(self);
+	u32 rpf_redir2_enable;
+	int tc;
+	int i;
+
+	rpf_redir2_enable = num_tcs > 4 ? 1 : 0;
+
+	hw_atl2_rpf_redirection_table2_select_set(self, rpf_redir2_enable);
+
+	for (i = HW_ATL2_RSS_REDIRECTION_MAX; i--;) {
+		for (tc = 0; tc != num_tcs; tc++) {
+			hw_atl2_new_rpf_rss_redir_set(self, tc, i,
+						      tc *
+						      aq_hw_q_per_tc(self) +
+						      indirection_table[i]);
+		}
+	}
+
+	return aq_hw_err_from_flags(self);
+}
+
+static int hw_atl2_hw_init_tx_tc_rate_limit(struct aq_hw_s *self)
+{
+	static const u32 max_weight = BIT(HW_ATL2_TPS_DATA_TCTWEIGHT_WIDTH) - 1;
+	/* Scale factor is based on the number of bits in fractional portion */
+	static const u32 scale = BIT(HW_ATL_TPS_DESC_RATE_Y_WIDTH);
+	static const u32 frac_msk = HW_ATL_TPS_DESC_RATE_Y_MSK >>
+				    HW_ATL_TPS_DESC_RATE_Y_SHIFT;
+	const u32 link_speed = self->aq_link_status.mbps;
+	struct aq_nic_cfg_s *nic_cfg = self->aq_nic_cfg;
+	unsigned long num_min_rated_tcs = 0;
+	u32 tc_weight[AQ_CFG_TCS_MAX];
+	u32 fixed_max_credit_4b;
+	u32 fixed_max_credit;
+	u8 min_rate_msk = 0;
+	u32 sum_weight = 0;
+	int tc;
+
+	/* By default max_credit is based upon MTU (in unit of 64b) */
+	fixed_max_credit = nic_cfg->aq_hw_caps->mtu / 64;
+	/* in unit of 4b */
+	fixed_max_credit_4b = nic_cfg->aq_hw_caps->mtu / 4;
+
+	if (link_speed) {
+		min_rate_msk = nic_cfg->tc_min_rate_msk &
+			       (BIT(nic_cfg->tcs) - 1);
+		num_min_rated_tcs = hweight8(min_rate_msk);
+	}
+
+	/* First, calculate weights where min_rate is specified */
+	if (num_min_rated_tcs) {
+		for (tc = 0; tc != nic_cfg->tcs; tc++) {
+			if (!nic_cfg->tc_min_rate[tc]) {
+				tc_weight[tc] = 0;
+				continue;
+			}
+
+			tc_weight[tc] = (-1L + link_speed +
+					 nic_cfg->tc_min_rate[tc] *
+					 max_weight) /
+					link_speed;
+			tc_weight[tc] = min(tc_weight[tc], max_weight);
+			sum_weight += tc_weight[tc];
+		}
+	}
+
+	/* WSP, if min_rate is set for at least one TC.
+	 * RR otherwise.
+	 */
+	hw_atl2_tps_tx_pkt_shed_data_arb_mode_set(self, min_rate_msk ? 1U : 0U);
+	/* Data TC Arbiter takes precedence over Descriptor TC Arbiter,
+	 * leave Descriptor TC Arbiter as RR.
+	 */
+	hw_atl_tps_tx_pkt_shed_desc_tc_arb_mode_set(self, 0U);
+
+	hw_atl_tps_tx_desc_rate_mode_set(self, nic_cfg->is_qos ? 1U : 0U);
+
+	for (tc = 0; tc != nic_cfg->tcs; tc++) {
+		const u32 en = (nic_cfg->tc_max_rate[tc] != 0) ? 1U : 0U;
+		const u32 desc = AQ_NIC_CFG_TCVEC2RING(nic_cfg, tc, 0);
+		u32 weight, max_credit;
+
+		hw_atl_tps_tx_pkt_shed_desc_tc_max_credit_set(self, tc,
+							      fixed_max_credit);
+		hw_atl_tps_tx_pkt_shed_desc_tc_weight_set(self, tc, 0x1E);
+
+		if (num_min_rated_tcs) {
+			weight = tc_weight[tc];
+
+			if (!weight && sum_weight < max_weight)
+				weight = (max_weight - sum_weight) /
+					 (nic_cfg->tcs - num_min_rated_tcs);
+			else if (!weight)
+				weight = 0x640;
+
+			max_credit = max(2 * weight, fixed_max_credit_4b);
+		} else {
+			weight = 0x640;
+			max_credit = 0xFFF0;
+		}
+
+		hw_atl2_tps_tx_pkt_shed_tc_data_weight_set(self, tc, weight);
+		hw_atl2_tps_tx_pkt_shed_tc_data_max_credit_set(self, tc,
+							       max_credit);
+
+		hw_atl_tps_tx_desc_rate_en_set(self, desc, en);
+
+		if (en) {
+			/* Nominal rate is always 10G */
+			const u32 rate = 10000U * scale /
+					 nic_cfg->tc_max_rate[tc];
+			const u32 rate_int = rate >>
+					     HW_ATL_TPS_DESC_RATE_Y_WIDTH;
+			const u32 rate_frac = rate & frac_msk;
+
+			hw_atl_tps_tx_desc_rate_x_set(self, desc, rate_int);
+			hw_atl_tps_tx_desc_rate_y_set(self, desc, rate_frac);
+		} else {
+			/* A value of 1 indicates the queue is not
+			 * rate controlled.
+			 */
+			hw_atl_tps_tx_desc_rate_x_set(self, desc, 1U);
+			hw_atl_tps_tx_desc_rate_y_set(self, desc, 0U);
+		}
+	}
+	for (tc = nic_cfg->tcs; tc != AQ_CFG_TCS_MAX; tc++) {
+		const u32 desc = AQ_NIC_CFG_TCVEC2RING(nic_cfg, tc, 0);
+
+		hw_atl_tps_tx_desc_rate_en_set(self, desc, 0U);
+		hw_atl_tps_tx_desc_rate_x_set(self, desc, 1U);
+		hw_atl_tps_tx_desc_rate_y_set(self, desc, 0U);
+	}
+
+	return aq_hw_err_from_flags(self);
+}
+
+static int hw_atl2_hw_init_tx_path(struct aq_hw_s *self)
+{
+	struct aq_nic_cfg_s *nic_cfg = self->aq_nic_cfg;
+
+	/* Tx TC/RSS number config */
+	hw_atl_tpb_tps_tx_tc_mode_set(self, nic_cfg->tc_mode);
+
+	hw_atl_thm_lso_tcp_flag_of_first_pkt_set(self, 0x0FF6U);
+	hw_atl_thm_lso_tcp_flag_of_middle_pkt_set(self, 0x0FF6U);
+	hw_atl_thm_lso_tcp_flag_of_last_pkt_set(self, 0x0F7FU);
+
+	/* Tx interrupts */
+	hw_atl_tdm_tx_desc_wr_wb_irq_en_set(self, 1U);
+
+	/* misc */
+	hw_atl_tdm_tx_dca_en_set(self, 0U);
+	hw_atl_tdm_tx_dca_mode_set(self, 0U);
+
+	hw_atl_tpb_tx_path_scp_ins_en_set(self, 1U);
+
+	hw_atl2_tpb_tx_buf_clk_gate_en_set(self, 0U);
+
+	return aq_hw_err_from_flags(self);
+}
+
+static void hw_atl2_hw_init_new_rx_filters(struct aq_hw_s *self)
+{
+	u8 *prio_tc_map = self->aq_nic_cfg->prio_tc_map;
+	struct hw_atl2_priv *priv = self->priv;
+	u16 action;
+	u8 index;
+	int i;
+
+	/* Action Resolver Table (ART) is used by RPF to decide which action
+	 * to take with a packet based upon input tag and tag mask, where:
+	 *  - input tag is a combination of 3-bit VLan Prio (PTP) and
+	 *    29-bit concatenation of all tags from filter block;
+	 *  - tag mask is a mask used for matching against input tag.
+	 * The input_tag is compared with the all the Requested_tags in the
+	 * Record table to find a match. Action field of the selected matched
+	 * REC entry is used for further processing. If multiple entries match,
+	 * the lowest REC entry, Action field will be selected.
+	 */
+	hw_atl2_rpf_act_rslvr_section_en_set(self, 0xFFFF);
+	hw_atl2_rpfl2_uc_flr_tag_set(self, HW_ATL2_RPF_TAG_BASE_UC,
+				     HW_ATL2_MAC_UC);
+	hw_atl2_rpfl2_bc_flr_tag_set(self, HW_ATL2_RPF_TAG_BASE_UC);
+
+	/* FW reserves the beginning of ART, thus all driver entries must
+	 * start from the offset specified in FW caps.
+	 */
+	index = priv->art_base_index + HW_ATL2_RPF_L2_PROMISC_OFF_INDEX;
+	hw_atl2_act_rslvr_table_set(self, index, 0,
+				    HW_ATL2_RPF_TAG_UC_MASK |
+					HW_ATL2_RPF_TAG_ALLMC_MASK,
+				    HW_ATL2_ACTION_DROP);
+
+	index = priv->art_base_index + HW_ATL2_RPF_VLAN_PROMISC_OFF_INDEX;
+	hw_atl2_act_rslvr_table_set(self, index, 0,
+				    HW_ATL2_RPF_TAG_VLAN_MASK |
+					HW_ATL2_RPF_TAG_UNTAG_MASK,
+				    HW_ATL2_ACTION_DROP);
+
+	/* Configure ART to map given VLan Prio (PCP) to the TC index for
+	 * RSS redirection table.
+	 */
+	for (i = 0; i < 8; i++) {
+		action = HW_ATL2_ACTION_ASSIGN_TC(prio_tc_map[i]);
+
+		index = priv->art_base_index + HW_ATL2_RPF_PCP_TO_TC_INDEX + i;
+		hw_atl2_act_rslvr_table_set(self, index,
+					    i << HW_ATL2_RPF_TAG_PCP_OFFSET,
+					    HW_ATL2_RPF_TAG_PCP_MASK, action);
+	}
+}
+
+static void hw_atl2_hw_new_rx_filter_vlan_promisc(struct aq_hw_s *self,
+						  bool promisc)
+{
+	u16 off_action = (!promisc &&
+			  !hw_atl_rpfl2promiscuous_mode_en_get(self)) ?
+				HW_ATL2_ACTION_DROP : HW_ATL2_ACTION_DISABLE;
+	struct hw_atl2_priv *priv = self->priv;
+	u8 index;
+
+	index = priv->art_base_index + HW_ATL2_RPF_VLAN_PROMISC_OFF_INDEX;
+	hw_atl2_act_rslvr_table_set(self, index, 0,
+				    HW_ATL2_RPF_TAG_VLAN_MASK |
+				    HW_ATL2_RPF_TAG_UNTAG_MASK, off_action);
+}
+
+static void hw_atl2_hw_new_rx_filter_promisc(struct aq_hw_s *self, bool promisc)
+{
+	u16 off_action = promisc ? HW_ATL2_ACTION_DISABLE : HW_ATL2_ACTION_DROP;
+	struct hw_atl2_priv *priv = self->priv;
+	bool vlan_promisc_enable;
+	u8 index;
+
+	index = priv->art_base_index + HW_ATL2_RPF_L2_PROMISC_OFF_INDEX;
+	hw_atl2_act_rslvr_table_set(self, index, 0,
+				    HW_ATL2_RPF_TAG_UC_MASK |
+				    HW_ATL2_RPF_TAG_ALLMC_MASK,
+				    off_action);
+
+	/* turn VLAN promisc mode too */
+	vlan_promisc_enable = hw_atl_rpf_vlan_prom_mode_en_get(self);
+	hw_atl2_hw_new_rx_filter_vlan_promisc(self, promisc |
+					      vlan_promisc_enable);
+}
+
+static int hw_atl2_act_rslvr_table_set(struct aq_hw_s *self, u8 location,
+				       u32 tag, u32 mask, u32 action)
+{
+	u32 val;
+	int err;
+
+	err = readx_poll_timeout_atomic(hw_atl2_sem_act_rslvr_get,
+					self, val, val == 1,
+					1, 10000U);
+	if (err)
+		return err;
+
+	hw_atl2_rpf_act_rslvr_record_set(self, location, tag, mask,
+					 action);
+
+	hw_atl_reg_glb_cpu_sem_set(self, 1, HW_ATL2_FW_SM_ACT_RSLVR);
+
+	return err;
+}
+
+static int hw_atl2_hw_init_rx_path(struct aq_hw_s *self)
+{
+	struct aq_nic_cfg_s *cfg = self->aq_nic_cfg;
+	int i;
+
+	/* Rx TC/RSS number config */
+	hw_atl_rpb_rpf_rx_traf_class_mode_set(self, cfg->tc_mode);
+
+	/* Rx flow control */
+	hw_atl_rpb_rx_flow_ctl_mode_set(self, 1U);
+
+	hw_atl2_rpf_rss_hash_type_set(self, HW_ATL2_RPF_RSS_HASH_TYPE_ALL);
+
+	/* RSS Ring selection */
+	hw_atl_b0_hw_init_rx_rss_ctrl1(self);
+
+	/* Multicast filters */
+	for (i = HW_ATL2_MAC_MAX; i--;) {
+		hw_atl_rpfl2_uc_flr_en_set(self, (i == 0U) ? 1U : 0U, i);
+		hw_atl_rpfl2unicast_flr_act_set(self, 1U, i);
+	}
+
+	hw_atl_reg_rx_flr_mcst_flr_msk_set(self, 0x00000000U);
+	hw_atl_reg_rx_flr_mcst_flr_set(self, HW_ATL_MCAST_FLT_ANY_TO_HOST, 0U);
+
+	/* Vlan filters */
+	hw_atl_rpf_vlan_outer_etht_set(self, ETH_P_8021AD);
+	hw_atl_rpf_vlan_inner_etht_set(self, ETH_P_8021Q);
+
+	hw_atl_rpf_vlan_prom_mode_en_set(self, 1);
+
+	/* Always accept untagged packets */
+	hw_atl_rpf_vlan_accept_untagged_packets_set(self, 1U);
+	hw_atl_rpf_vlan_untagged_act_set(self, 1U);
+
+	hw_atl2_hw_init_new_rx_filters(self);
+
+	/* Rx Interrupts */
+	hw_atl_rdm_rx_desc_wr_wb_irq_en_set(self, 1U);
+
+	hw_atl_rpfl2broadcast_flr_act_set(self, 1U);
+	hw_atl_rpfl2broadcast_count_threshold_set(self, 0xFFFFU & (~0U / 256U));
+
+	hw_atl_rdm_rx_dca_en_set(self, 0U);
+	hw_atl_rdm_rx_dca_mode_set(self, 0U);
+
+	return aq_hw_err_from_flags(self);
+}
+
+static int hw_atl2_hw_init(struct aq_hw_s *self, const u8 *mac_addr)
+{
+	static u32 aq_hw_atl2_igcr_table_[4][2] = {
+		[AQ_HW_IRQ_INVALID] = { 0x20000000U, 0x20000000U },
+		[AQ_HW_IRQ_INTX]    = { 0x20000080U, 0x20000080U },
+		[AQ_HW_IRQ_MSI]     = { 0x20000021U, 0x20000025U },
+		[AQ_HW_IRQ_MSIX]    = { 0x20000022U, 0x20000026U },
+	};
+
+	struct aq_nic_cfg_s *aq_nic_cfg = self->aq_nic_cfg;
+	struct hw_atl2_priv *priv = self->priv;
+	u8 base_index, count;
+	int err;
+
+	err = hw_atl2_utils_get_action_resolve_table_caps(self, &base_index,
+							  &count);
+	if (err)
+		return err;
+
+	priv->art_base_index = 8 * base_index;
+
+	hw_atl2_init_launchtime(self);
+
+	hw_atl2_hw_init_tx_path(self);
+	hw_atl2_hw_init_rx_path(self);
+
+	hw_atl_b0_hw_mac_addr_set(self, mac_addr);
+
+	self->aq_fw_ops->set_link_speed(self, aq_nic_cfg->link_speed_msk);
+	self->aq_fw_ops->set_state(self, MPI_INIT);
+
+	hw_atl2_hw_qos_set(self);
+	hw_atl2_hw_rss_set(self, &aq_nic_cfg->aq_rss);
+	hw_atl_b0_hw_rss_hash_set(self, &aq_nic_cfg->aq_rss);
+
+	hw_atl2_rpf_new_enable_set(self, 1);
+
+	/* Reset link status and read out initial hardware counters */
+	self->aq_link_status.mbps = 0;
+	self->aq_fw_ops->update_stats(self);
+
+	err = aq_hw_err_from_flags(self);
+	if (err < 0)
+		goto err_exit;
+
+	/* Interrupts */
+	hw_atl_reg_irq_glb_ctl_set(self,
+				   aq_hw_atl2_igcr_table_[aq_nic_cfg->irq_type]
+						 [(aq_nic_cfg->vecs > 1U) ?
+						  1 : 0]);
+
+	hw_atl_itr_irq_auto_masklsw_set(self, aq_nic_cfg->aq_hw_caps->irq_mask);
+
+	/* Interrupts */
+	hw_atl_reg_gen_irq_map_set(self,
+				   ((HW_ATL2_ERR_INT << 0x18) |
+				    (1U << 0x1F)) |
+				   ((HW_ATL2_ERR_INT << 0x10) |
+				    (1U << 0x17)), 0U);
+
+	hw_atl_b0_hw_offload_set(self, aq_nic_cfg);
+
+err_exit:
+	return err;
+}
+
+static int hw_atl2_hw_ring_rx_init(struct aq_hw_s *self,
+				   struct aq_ring_s *aq_ring,
+				   struct aq_ring_param_s *aq_ring_param)
+{
+	return hw_atl_b0_hw_ring_rx_init(self, aq_ring, aq_ring_param);
+}
+
+static int hw_atl2_hw_ring_tx_init(struct aq_hw_s *self,
+				   struct aq_ring_s *aq_ring,
+				   struct aq_ring_param_s *aq_ring_param)
+{
+	return hw_atl_b0_hw_ring_tx_init(self, aq_ring, aq_ring_param);
+}
+
+#define IS_FILTER_ENABLED(_F_) ((packet_filter & (_F_)) ? 1U : 0U)
+
+static int hw_atl2_hw_packet_filter_set(struct aq_hw_s *self,
+					unsigned int packet_filter)
+{
+	hw_atl2_hw_new_rx_filter_promisc(self, IS_FILTER_ENABLED(IFF_PROMISC));
+
+	return hw_atl_b0_hw_packet_filter_set(self, packet_filter);
+}
+
+#undef IS_FILTER_ENABLED
+
+static int hw_atl2_hw_multicast_list_set(struct aq_hw_s *self,
+					 u8 ar_mac
+					 [AQ_HW_MULTICAST_ADDRESS_MAX]
+					 [ETH_ALEN],
+					 u32 count)
+{
+	struct aq_nic_cfg_s *cfg = self->aq_nic_cfg;
+	int err = 0;
+
+	if (count > (HW_ATL2_MAC_MAX - HW_ATL2_MAC_MIN)) {
+		err = -EBADRQC;
+		goto err_exit;
+	}
+	for (cfg->mc_list_count = 0U;
+			cfg->mc_list_count < count;
+			++cfg->mc_list_count) {
+		u32 i = cfg->mc_list_count;
+		u32 h = (ar_mac[i][0] << 8) | (ar_mac[i][1]);
+		u32 l = (ar_mac[i][2] << 24) | (ar_mac[i][3] << 16) |
+					(ar_mac[i][4] << 8) | ar_mac[i][5];
+
+		hw_atl_rpfl2_uc_flr_en_set(self, 0U, HW_ATL2_MAC_MIN + i);
+
+		hw_atl_rpfl2unicast_dest_addresslsw_set(self, l,
+							HW_ATL2_MAC_MIN + i);
+
+		hw_atl_rpfl2unicast_dest_addressmsw_set(self, h,
+							HW_ATL2_MAC_MIN + i);
+
+		hw_atl2_rpfl2_uc_flr_tag_set(self, 1, HW_ATL2_MAC_MIN + i);
+
+		hw_atl_rpfl2_uc_flr_en_set(self, (cfg->is_mc_list_enabled),
+					   HW_ATL2_MAC_MIN + i);
+	}
+
+	err = aq_hw_err_from_flags(self);
+
+err_exit:
+	return err;
+}
+
+static int hw_atl2_hw_interrupt_moderation_set(struct aq_hw_s *self)
+{
+	unsigned int i = 0U;
+	u32 itr_tx = 2U;
+	u32 itr_rx = 2U;
+
+	switch (self->aq_nic_cfg->itr) {
+	case  AQ_CFG_INTERRUPT_MODERATION_ON:
+	case  AQ_CFG_INTERRUPT_MODERATION_AUTO:
+		hw_atl_tdm_tx_desc_wr_wb_irq_en_set(self, 0U);
+		hw_atl_tdm_tdm_intr_moder_en_set(self, 1U);
+		hw_atl_rdm_rx_desc_wr_wb_irq_en_set(self, 0U);
+		hw_atl_rdm_rdm_intr_moder_en_set(self, 1U);
+
+		if (self->aq_nic_cfg->itr == AQ_CFG_INTERRUPT_MODERATION_ON) {
+			/* HW timers are in 2us units */
+			int tx_max_timer = self->aq_nic_cfg->tx_itr / 2;
+			int tx_min_timer = tx_max_timer / 2;
+
+			int rx_max_timer = self->aq_nic_cfg->rx_itr / 2;
+			int rx_min_timer = rx_max_timer / 2;
+
+			tx_max_timer = min(HW_ATL2_INTR_MODER_MAX,
+					   tx_max_timer);
+			tx_min_timer = min(HW_ATL2_INTR_MODER_MIN,
+					   tx_min_timer);
+			rx_max_timer = min(HW_ATL2_INTR_MODER_MAX,
+					   rx_max_timer);
+			rx_min_timer = min(HW_ATL2_INTR_MODER_MIN,
+					   rx_min_timer);
+
+			itr_tx |= tx_min_timer << 0x8U;
+			itr_tx |= tx_max_timer << 0x10U;
+			itr_rx |= rx_min_timer << 0x8U;
+			itr_rx |= rx_max_timer << 0x10U;
+		} else {
+			static unsigned int hw_atl2_timers_table_tx_[][2] = {
+				{0xfU, 0xffU}, /* 10Gbit */
+				{0xfU, 0x1ffU}, /* 5Gbit */
+				{0xfU, 0x1ffU}, /* 5Gbit 5GS */
+				{0xfU, 0x1ffU}, /* 2.5Gbit */
+				{0xfU, 0x1ffU}, /* 1Gbit */
+				{0xfU, 0x1ffU}, /* 100Mbit */
+			};
+			static unsigned int hw_atl2_timers_table_rx_[][2] = {
+				{0x6U, 0x38U},/* 10Gbit */
+				{0xCU, 0x70U},/* 5Gbit */
+				{0xCU, 0x70U},/* 5Gbit 5GS */
+				{0x18U, 0xE0U},/* 2.5Gbit */
+				{0x30U, 0x80U},/* 1Gbit */
+				{0x4U, 0x50U},/* 100Mbit */
+			};
+			unsigned int mbps = self->aq_link_status.mbps;
+			unsigned int speed_index;
+
+			speed_index = hw_atl_utils_mbps_2_speed_index(mbps);
+
+			/* Update user visible ITR settings */
+			self->aq_nic_cfg->tx_itr = hw_atl2_timers_table_tx_
+							[speed_index][1] * 2;
+			self->aq_nic_cfg->rx_itr = hw_atl2_timers_table_rx_
+							[speed_index][1] * 2;
+
+			itr_tx |= hw_atl2_timers_table_tx_
+						[speed_index][0] << 0x8U;
+			itr_tx |= hw_atl2_timers_table_tx_
+						[speed_index][1] << 0x10U;
+
+			itr_rx |= hw_atl2_timers_table_rx_
+						[speed_index][0] << 0x8U;
+			itr_rx |= hw_atl2_timers_table_rx_
+						[speed_index][1] << 0x10U;
+		}
+		break;
+	case AQ_CFG_INTERRUPT_MODERATION_OFF:
+		hw_atl_tdm_tx_desc_wr_wb_irq_en_set(self, 1U);
+		hw_atl_tdm_tdm_intr_moder_en_set(self, 0U);
+		hw_atl_rdm_rx_desc_wr_wb_irq_en_set(self, 1U);
+		hw_atl_rdm_rdm_intr_moder_en_set(self, 0U);
+		itr_tx = 0U;
+		itr_rx = 0U;
+		break;
+	}
+
+	for (i = HW_ATL2_RINGS_MAX; i--;) {
+		hw_atl2_reg_tx_intr_moder_ctrl_set(self, itr_tx, i);
+		hw_atl_reg_rx_intr_moder_ctrl_set(self, itr_rx, i);
+	}
+
+	return aq_hw_err_from_flags(self);
+}
+
+static int hw_atl2_hw_stop(struct aq_hw_s *self)
+{
+	hw_atl_b0_hw_irq_disable(self, HW_ATL2_INT_MASK);
+
+	return 0;
+}
+
+static struct aq_stats_s *hw_atl2_utils_get_hw_stats(struct aq_hw_s *self)
+{
+	return &self->curr_stats;
+}
+
+static int hw_atl2_hw_vlan_set(struct aq_hw_s *self,
+			       struct aq_rx_filter_vlan *aq_vlans)
+{
+	struct hw_atl2_priv *priv = self->priv;
+	u32 queue;
+	u8 index;
+	int i;
+
+	hw_atl_rpf_vlan_prom_mode_en_set(self, 1U);
+
+	for (i = 0; i < HW_ATL_VLAN_MAX_FILTERS; i++) {
+		queue = HW_ATL2_ACTION_ASSIGN_QUEUE(aq_vlans[i].queue);
+
+		hw_atl_rpf_vlan_flr_en_set(self, 0U, i);
+		hw_atl_rpf_vlan_rxq_en_flr_set(self, 0U, i);
+		index = priv->art_base_index + HW_ATL2_RPF_VLAN_USER_INDEX + i;
+		hw_atl2_act_rslvr_table_set(self, index, 0, 0,
+					    HW_ATL2_ACTION_DISABLE);
+		if (aq_vlans[i].enable) {
+			hw_atl_rpf_vlan_id_flr_set(self,
+						   aq_vlans[i].vlan_id, i);
+			hw_atl_rpf_vlan_flr_act_set(self, 1U, i);
+			hw_atl_rpf_vlan_flr_en_set(self, 1U, i);
+
+			if (aq_vlans[i].queue != 0xFF) {
+				hw_atl_rpf_vlan_rxq_flr_set(self,
+							    aq_vlans[i].queue,
+							    i);
+				hw_atl_rpf_vlan_rxq_en_flr_set(self, 1U, i);
+
+				hw_atl2_rpf_vlan_flr_tag_set(self, i + 2, i);
+
+				index = priv->art_base_index +
+					HW_ATL2_RPF_VLAN_USER_INDEX + i;
+				hw_atl2_act_rslvr_table_set(self, index,
+					(i + 2) << HW_ATL2_RPF_TAG_VLAN_OFFSET,
+					HW_ATL2_RPF_TAG_VLAN_MASK, queue);
+			} else {
+				hw_atl2_rpf_vlan_flr_tag_set(self, 1, i);
+			}
+		}
+	}
+
+	return aq_hw_err_from_flags(self);
+}
+
+static int hw_atl2_hw_vlan_ctrl(struct aq_hw_s *self, bool enable)
+{
+	/* set promisc in case of disabing the vlan filter */
+	hw_atl_rpf_vlan_prom_mode_en_set(self, !enable);
+	hw_atl2_hw_new_rx_filter_vlan_promisc(self, !enable);
+
+	return aq_hw_err_from_flags(self);
+}
+
+const struct aq_hw_ops hw_atl2_ops = {
+	.hw_soft_reset        = hw_atl2_utils_soft_reset,
+	.hw_prepare           = hw_atl2_utils_initfw,
+	.hw_set_mac_address   = hw_atl_b0_hw_mac_addr_set,
+	.hw_init              = hw_atl2_hw_init,
+	.hw_reset             = hw_atl2_hw_reset,
+	.hw_start             = hw_atl_b0_hw_start,
+	.hw_ring_tx_start     = hw_atl_b0_hw_ring_tx_start,
+	.hw_ring_tx_stop      = hw_atl_b0_hw_ring_tx_stop,
+	.hw_ring_rx_start     = hw_atl_b0_hw_ring_rx_start,
+	.hw_ring_rx_stop      = hw_atl_b0_hw_ring_rx_stop,
+	.hw_stop              = hw_atl2_hw_stop,
+
+	.hw_ring_tx_xmit         = hw_atl_b0_hw_ring_tx_xmit,
+	.hw_ring_tx_head_update  = hw_atl_b0_hw_ring_tx_head_update,
+
+	.hw_ring_rx_receive      = hw_atl_b0_hw_ring_rx_receive,
+	.hw_ring_rx_fill         = hw_atl_b0_hw_ring_rx_fill,
+
+	.hw_irq_enable           = hw_atl_b0_hw_irq_enable,
+	.hw_irq_disable          = hw_atl_b0_hw_irq_disable,
+	.hw_irq_read             = hw_atl_b0_hw_irq_read,
+
+	.hw_ring_rx_init             = hw_atl2_hw_ring_rx_init,
+	.hw_ring_tx_init             = hw_atl2_hw_ring_tx_init,
+	.hw_packet_filter_set        = hw_atl2_hw_packet_filter_set,
+	.hw_filter_vlan_set          = hw_atl2_hw_vlan_set,
+	.hw_filter_vlan_ctrl         = hw_atl2_hw_vlan_ctrl,
+	.hw_multicast_list_set       = hw_atl2_hw_multicast_list_set,
+	.hw_interrupt_moderation_set = hw_atl2_hw_interrupt_moderation_set,
+	.hw_rss_set                  = hw_atl2_hw_rss_set,
+	.hw_rss_hash_set             = hw_atl_b0_hw_rss_hash_set,
+	.hw_tc_rate_limit_set        = hw_atl2_hw_init_tx_tc_rate_limit,
+	.hw_get_hw_stats             = hw_atl2_utils_get_hw_stats,
+	.hw_get_fw_version           = hw_atl2_utils_get_fw_version,
+	.hw_set_offload              = hw_atl_b0_hw_offload_set,
+	.hw_set_loopback             = hw_atl_b0_set_loopback,
+	.hw_set_fc                   = hw_atl_b0_set_fc,
+};
diff --git a/drivers/net/ethernet/aquantia/atlantic/hw_atl2/hw_atl2.h b/drivers/net/ethernet/aquantia/atlantic/hw_atl2/hw_atl2.h
new file mode 100644
index 000000000000..346f0dc9912e
--- /dev/null
+++ b/drivers/net/ethernet/aquantia/atlantic/hw_atl2/hw_atl2.h
@@ -0,0 +1,16 @@
+/* SPDX-License-Identifier: GPL-2.0-only */
+/* Atlantic Network Driver
+ * Copyright (C) 2020 Marvell International Ltd.
+ */
+
+#ifndef HW_ATL2_H
+#define HW_ATL2_H
+
+#include "aq_common.h"
+
+extern const struct aq_hw_caps_s hw_atl2_caps_aqc113;
+extern const struct aq_hw_caps_s hw_atl2_caps_aqc115c;
+extern const struct aq_hw_caps_s hw_atl2_caps_aqc116c;
+extern const struct aq_hw_ops hw_atl2_ops;
+
+#endif /* HW_ATL2_H */
diff --git a/drivers/net/ethernet/aquantia/atlantic/hw_atl2/hw_atl2_internal.h b/drivers/net/ethernet/aquantia/atlantic/hw_atl2/hw_atl2_internal.h
new file mode 100644
index 000000000000..5a89bb8722f9
--- /dev/null
+++ b/drivers/net/ethernet/aquantia/atlantic/hw_atl2/hw_atl2_internal.h
@@ -0,0 +1,127 @@
+/* SPDX-License-Identifier: GPL-2.0-only */
+/* Atlantic Network Driver
+ * Copyright (C) 2020 Marvell International Ltd.
+ */
+
+#ifndef HW_ATL2_INTERNAL_H
+#define HW_ATL2_INTERNAL_H
+
+#include "hw_atl2_utils.h"
+
+#define HW_ATL2_MTU_JUMBO  16352U
+#define HW_ATL2_MTU        1514U
+
+#define HW_ATL2_TX_RINGS 4U
+#define HW_ATL2_RX_RINGS 4U
+
+#define HW_ATL2_RINGS_MAX 32U
+#define HW_ATL2_TXD_SIZE       (16U)
+#define HW_ATL2_RXD_SIZE       (16U)
+
+#define HW_ATL2_MAC_UC   0U
+#define HW_ATL2_MAC_MIN  1U
+#define HW_ATL2_MAC_MAX  38U
+
+/* interrupts */
+#define HW_ATL2_ERR_INT 8U
+#define HW_ATL2_INT_MASK  (0xFFFFFFFFU)
+
+#define HW_ATL2_TXBUF_MAX              128U
+#define HW_ATL2_RXBUF_MAX              192U
+
+#define HW_ATL2_RSS_REDIRECTION_MAX 64U
+
+#define HW_ATL2_TC_MAX 8U
+#define HW_ATL2_RSS_MAX 8U
+
+#define HW_ATL2_INTR_MODER_MAX  0x1FF
+#define HW_ATL2_INTR_MODER_MIN  0xFF
+
+#define HW_ATL2_MIN_RXD \
+	(ALIGN(AQ_CFG_SKB_FRAGS_MAX + 1U, AQ_HW_RXD_MULTIPLE))
+#define HW_ATL2_MIN_TXD \
+	(ALIGN(AQ_CFG_SKB_FRAGS_MAX + 1U, AQ_HW_TXD_MULTIPLE))
+
+#define HW_ATL2_MAX_RXD 8184U
+#define HW_ATL2_MAX_TXD 8184U
+
+#define HW_ATL2_FW_SM_ACT_RSLVR  0x3U
+
+#define HW_ATL2_RPF_TAG_UC_OFFSET      0x0
+#define HW_ATL2_RPF_TAG_ALLMC_OFFSET   0x6
+#define HW_ATL2_RPF_TAG_ET_OFFSET      0x7
+#define HW_ATL2_RPF_TAG_VLAN_OFFSET    0xA
+#define HW_ATL2_RPF_TAG_UNTAG_OFFSET   0xE
+#define HW_ATL2_RPF_TAG_L3_V4_OFFSET   0xF
+#define HW_ATL2_RPF_TAG_L3_V6_OFFSET   0x12
+#define HW_ATL2_RPF_TAG_L4_OFFSET      0x15
+#define HW_ATL2_RPF_TAG_L4_FLEX_OFFSET 0x18
+#define HW_ATL2_RPF_TAG_FLEX_OFFSET    0x1B
+#define HW_ATL2_RPF_TAG_PCP_OFFSET     0x1D
+
+#define HW_ATL2_RPF_TAG_UC_MASK    (0x0000003F << HW_ATL2_RPF_TAG_UC_OFFSET)
+#define HW_ATL2_RPF_TAG_ALLMC_MASK (0x00000001 << HW_ATL2_RPF_TAG_ALLMC_OFFSET)
+#define HW_ATL2_RPF_TAG_UNTAG_MASK (0x00000001 << HW_ATL2_RPF_TAG_UNTAG_OFFSET)
+#define HW_ATL2_RPF_TAG_VLAN_MASK  (0x0000000F << HW_ATL2_RPF_TAG_VLAN_OFFSET)
+#define HW_ATL2_RPF_TAG_ET_MASK    (0x00000007 << HW_ATL2_RPF_TAG_ET_OFFSET)
+#define HW_ATL2_RPF_TAG_L3_V4_MASK (0x00000007 << HW_ATL2_RPF_TAG_L3_V4_OFFSET)
+#define HW_ATL2_RPF_TAG_L3_V6_MASK (0x00000007 << HW_ATL2_RPF_TAG_L3_V6_OFFSET)
+#define HW_ATL2_RPF_TAG_L4_MASK    (0x00000007 << HW_ATL2_RPF_TAG_L4_OFFSET)
+#define HW_ATL2_RPF_TAG_PCP_MASK   (0x00000007 << HW_ATL2_RPF_TAG_PCP_OFFSET)
+
+#define HW_ATL2_RPF_TAG_BASE_UC    BIT(HW_ATL2_RPF_TAG_UC_OFFSET)
+#define HW_ATL2_RPF_TAG_BASE_ALLMC BIT(HW_ATL2_RPF_TAG_ALLMC_OFFSET)
+#define HW_ATL2_RPF_TAG_BASE_UNTAG BIT(HW_ATL2_RPF_TAG_UNTAG_OFFSET)
+#define HW_ATL2_RPF_TAG_BASE_VLAN  BIT(HW_ATL2_RPF_TAG_VLAN_OFFSET)
+
+enum HW_ATL2_RPF_ART_INDEX {
+	HW_ATL2_RPF_L2_PROMISC_OFF_INDEX,
+	HW_ATL2_RPF_VLAN_PROMISC_OFF_INDEX,
+	HW_ATL2_RPF_L3L4_USER_INDEX	= 8,
+	HW_ATL2_RPF_ET_PCP_USER_INDEX	= HW_ATL2_RPF_L3L4_USER_INDEX + 16,
+	HW_ATL2_RPF_VLAN_USER_INDEX	= HW_ATL2_RPF_ET_PCP_USER_INDEX + 16,
+	HW_ATL2_RPF_PCP_TO_TC_INDEX	= HW_ATL2_RPF_VLAN_USER_INDEX +
+					  HW_ATL_VLAN_MAX_FILTERS,
+};
+
+#define HW_ATL2_ACTION(ACTION, RSS, INDEX, VALID) \
+	((((ACTION) & 0x3U) << 8) | \
+	(((RSS) & 0x1U) << 7) | \
+	(((INDEX) & 0x3FU) << 2) | \
+	(((VALID) & 0x1U) << 0))
+
+#define HW_ATL2_ACTION_DROP HW_ATL2_ACTION(0, 0, 0, 1)
+#define HW_ATL2_ACTION_DISABLE HW_ATL2_ACTION(0, 0, 0, 0)
+#define HW_ATL2_ACTION_ASSIGN_QUEUE(QUEUE) HW_ATL2_ACTION(1, 0, (QUEUE), 1)
+#define HW_ATL2_ACTION_ASSIGN_TC(TC) HW_ATL2_ACTION(1, 1, (TC), 1)
+
+enum HW_ATL2_RPF_RSS_HASH_TYPE {
+	HW_ATL2_RPF_RSS_HASH_TYPE_NONE = 0,
+	HW_ATL2_RPF_RSS_HASH_TYPE_IPV4 = BIT(0),
+	HW_ATL2_RPF_RSS_HASH_TYPE_IPV4_TCP = BIT(1),
+	HW_ATL2_RPF_RSS_HASH_TYPE_IPV4_UDP = BIT(2),
+	HW_ATL2_RPF_RSS_HASH_TYPE_IPV6 = BIT(3),
+	HW_ATL2_RPF_RSS_HASH_TYPE_IPV6_TCP = BIT(4),
+	HW_ATL2_RPF_RSS_HASH_TYPE_IPV6_UDP = BIT(5),
+	HW_ATL2_RPF_RSS_HASH_TYPE_IPV6_EX = BIT(6),
+	HW_ATL2_RPF_RSS_HASH_TYPE_IPV6_EX_TCP = BIT(7),
+	HW_ATL2_RPF_RSS_HASH_TYPE_IPV6_EX_UDP = BIT(8),
+	HW_ATL2_RPF_RSS_HASH_TYPE_ALL = HW_ATL2_RPF_RSS_HASH_TYPE_IPV4 |
+					HW_ATL2_RPF_RSS_HASH_TYPE_IPV4_TCP |
+					HW_ATL2_RPF_RSS_HASH_TYPE_IPV4_UDP |
+					HW_ATL2_RPF_RSS_HASH_TYPE_IPV6 |
+					HW_ATL2_RPF_RSS_HASH_TYPE_IPV6_TCP |
+					HW_ATL2_RPF_RSS_HASH_TYPE_IPV6_UDP |
+					HW_ATL2_RPF_RSS_HASH_TYPE_IPV6_EX |
+					HW_ATL2_RPF_RSS_HASH_TYPE_IPV6_EX_TCP |
+					HW_ATL2_RPF_RSS_HASH_TYPE_IPV6_EX_UDP,
+};
+
+#define HW_ATL_MCAST_FLT_ANY_TO_HOST 0x00010FFFU
+
+struct hw_atl2_priv {
+	struct statistics_s last_stats;
+	unsigned int art_base_index;
+};
+
+#endif /* HW_ATL2_INTERNAL_H */
diff --git a/drivers/net/ethernet/aquantia/atlantic/hw_atl2/hw_atl2_llh.c b/drivers/net/ethernet/aquantia/atlantic/hw_atl2/hw_atl2_llh.c
new file mode 100644
index 000000000000..cd954b11d24a
--- /dev/null
+++ b/drivers/net/ethernet/aquantia/atlantic/hw_atl2/hw_atl2_llh.c
@@ -0,0 +1,234 @@
+// SPDX-License-Identifier: GPL-2.0-only
+/* Atlantic Network Driver
+ * Copyright (C) 2020 Marvell International Ltd.
+ */
+
+#include "hw_atl2_llh.h"
+#include "hw_atl2_llh_internal.h"
+#include "aq_hw_utils.h"
+
+void hw_atl2_rpf_redirection_table2_select_set(struct aq_hw_s *aq_hw,
+					       u32 select)
+{
+	aq_hw_write_reg_bit(aq_hw, HW_ATL2_RPF_PIF_RPF_REDIR2_ENI_ADR,
+			    HW_ATL2_RPF_PIF_RPF_REDIR2_ENI_MSK,
+			    HW_ATL2_RPF_PIF_RPF_REDIR2_ENI_SHIFT, select);
+}
+
+void hw_atl2_rpf_rss_hash_type_set(struct aq_hw_s *aq_hw, u32 rss_hash_type)
+{
+	aq_hw_write_reg_bit(aq_hw, HW_ATL2_RPF_PIF_RPF_RSS_HASH_TYPEI_ADR,
+			    HW_ATL2_RPF_PIF_RPF_RSS_HASH_TYPEI_MSK,
+			    HW_ATL2_RPF_PIF_RPF_RSS_HASH_TYPEI_SHIFT,
+			    rss_hash_type);
+}
+
+/* rpf */
+
+void hw_atl2_rpf_new_enable_set(struct aq_hw_s *aq_hw, u32 enable)
+{
+	aq_hw_write_reg_bit(aq_hw, HW_ATL2_RPF_NEW_EN_ADR,
+			    HW_ATL2_RPF_NEW_EN_MSK,
+			    HW_ATL2_RPF_NEW_EN_SHIFT,
+			    enable);
+}
+
+void hw_atl2_rpfl2_uc_flr_tag_set(struct aq_hw_s *aq_hw, u32 tag, u32 filter)
+{
+	aq_hw_write_reg_bit(aq_hw, HW_ATL2_RPFL2UC_TAG_ADR(filter),
+			    HW_ATL2_RPFL2UC_TAG_MSK,
+			    HW_ATL2_RPFL2UC_TAG_SHIFT,
+			    tag);
+}
+
+void hw_atl2_rpfl2_bc_flr_tag_set(struct aq_hw_s *aq_hw, u32 tag)
+{
+	aq_hw_write_reg_bit(aq_hw, HW_ATL2_RPF_L2_BC_TAG_ADR,
+			    HW_ATL2_RPF_L2_BC_TAG_MSK,
+			    HW_ATL2_RPF_L2_BC_TAG_SHIFT,
+			    tag);
+}
+
+void hw_atl2_new_rpf_rss_redir_set(struct aq_hw_s *aq_hw, u32 tc, u32 index,
+				   u32 queue)
+{
+	aq_hw_write_reg_bit(aq_hw, HW_ATL2_RPF_RSS_REDIR_ADR(tc, index),
+			    HW_ATL2_RPF_RSS_REDIR_MSK(tc),
+			    HW_ATL2_RPF_RSS_REDIR_SHIFT(tc),
+			    queue);
+}
+
+void hw_atl2_rpf_vlan_flr_tag_set(struct aq_hw_s *aq_hw, u32 tag, u32 filter)
+{
+	aq_hw_write_reg_bit(aq_hw, HW_ATL2_RPF_VL_TAG_ADR(filter),
+			    HW_ATL2_RPF_VL_TAG_MSK,
+			    HW_ATL2_RPF_VL_TAG_SHIFT,
+			    tag);
+}
+
+/* TX */
+
+void hw_atl2_tpb_tx_tc_q_rand_map_en_set(struct aq_hw_s *aq_hw,
+					 const u32 tc_q_rand_map_en)
+{
+	aq_hw_write_reg_bit(aq_hw, HW_ATL2_TPB_TX_TC_Q_RAND_MAP_EN_ADR,
+			    HW_ATL2_TPB_TX_TC_Q_RAND_MAP_EN_MSK,
+			    HW_ATL2_TPB_TX_TC_Q_RAND_MAP_EN_SHIFT,
+			    tc_q_rand_map_en);
+}
+
+void hw_atl2_tpb_tx_buf_clk_gate_en_set(struct aq_hw_s *aq_hw, u32 clk_gate_en)
+{
+	aq_hw_write_reg_bit(aq_hw, HW_ATL2_TPB_TX_BUF_CLK_GATE_EN_ADR,
+			    HW_ATL2_TPB_TX_BUF_CLK_GATE_EN_MSK,
+			    HW_ATL2_TPB_TX_BUF_CLK_GATE_EN_SHIFT,
+			    clk_gate_en);
+}
+
+void hw_atl2_reg_tx_intr_moder_ctrl_set(struct aq_hw_s *aq_hw,
+					u32 tx_intr_moderation_ctl,
+					u32 queue)
+{
+	aq_hw_write_reg(aq_hw, HW_ATL2_TX_INTR_MODERATION_CTL_ADR(queue),
+			tx_intr_moderation_ctl);
+}
+
+void hw_atl2_tps_tx_pkt_shed_data_arb_mode_set(struct aq_hw_s *aq_hw,
+					       const u32 data_arb_mode)
+{
+	aq_hw_write_reg_bit(aq_hw, HW_ATL2_TPS_DATA_TC_ARB_MODE_ADR,
+			    HW_ATL2_TPS_DATA_TC_ARB_MODE_MSK,
+			    HW_ATL2_TPS_DATA_TC_ARB_MODE_SHIFT,
+			    data_arb_mode);
+}
+
+void hw_atl2_tps_tx_pkt_shed_tc_data_max_credit_set(struct aq_hw_s *aq_hw,
+						    const u32 tc,
+						    const u32 max_credit)
+{
+	aq_hw_write_reg_bit(aq_hw, HW_ATL2_TPS_DATA_TCTCREDIT_MAX_ADR(tc),
+			    HW_ATL2_TPS_DATA_TCTCREDIT_MAX_MSK,
+			    HW_ATL2_TPS_DATA_TCTCREDIT_MAX_SHIFT,
+			    max_credit);
+}
+
+void hw_atl2_tps_tx_pkt_shed_tc_data_weight_set(struct aq_hw_s *aq_hw,
+						const u32 tc,
+						const u32 weight)
+{
+	aq_hw_write_reg_bit(aq_hw, HW_ATL2_TPS_DATA_TCTWEIGHT_ADR(tc),
+			    HW_ATL2_TPS_DATA_TCTWEIGHT_MSK,
+			    HW_ATL2_TPS_DATA_TCTWEIGHT_SHIFT,
+			    weight);
+}
+
+u32 hw_atl2_get_hw_version(struct aq_hw_s *aq_hw)
+{
+	return aq_hw_read_reg(aq_hw, HW_ATL2_FPGA_VER_ADR);
+}
+
+void hw_atl2_init_launchtime(struct aq_hw_s *aq_hw)
+{
+	u32 hw_ver = hw_atl2_get_hw_version(aq_hw);
+
+	aq_hw_write_reg_bit(aq_hw, HW_ATL2_LT_CTRL_ADR,
+			    HW_ATL2_LT_CTRL_CLK_RATIO_MSK,
+			    HW_ATL2_LT_CTRL_CLK_RATIO_SHIFT,
+			    hw_ver  < HW_ATL2_FPGA_VER_U32(1, 0, 0, 0) ?
+			    HW_ATL2_LT_CTRL_CLK_RATIO_FULL_SPEED :
+			    hw_ver >= HW_ATL2_FPGA_VER_U32(1, 0, 85, 2) ?
+			    HW_ATL2_LT_CTRL_CLK_RATIO_HALF_SPEED :
+			    HW_ATL2_LT_CTRL_CLK_RATIO_QUATER_SPEED);
+}
+
+/* set action resolver record */
+void hw_atl2_rpf_act_rslvr_record_set(struct aq_hw_s *aq_hw, u8 location,
+				      u32 tag, u32 mask, u32 action)
+{
+	aq_hw_write_reg(aq_hw,
+			HW_ATL2_RPF_ACT_RSLVR_REQ_TAG_ADR(location),
+			tag);
+	aq_hw_write_reg(aq_hw,
+			HW_ATL2_RPF_ACT_RSLVR_TAG_MASK_ADR(location),
+			mask);
+	aq_hw_write_reg(aq_hw,
+			HW_ATL2_RPF_ACT_RSLVR_ACTN_ADR(location),
+			action);
+}
+
+void hw_atl2_rpf_act_rslvr_section_en_set(struct aq_hw_s *aq_hw, u32 sections)
+{
+	aq_hw_write_reg_bit(aq_hw, HW_ATL2_RPF_REC_TAB_EN_ADR,
+			    HW_ATL2_RPF_REC_TAB_EN_MSK,
+			    HW_ATL2_RPF_REC_TAB_EN_SHIFT,
+			    sections);
+}
+
+void hw_atl2_mif_shared_buf_get(struct aq_hw_s *aq_hw, int offset, u32 *data,
+				int len)
+{
+	int j = 0;
+	int i;
+
+	for (i = offset; i < offset + len; i++, j++)
+		data[j] = aq_hw_read_reg(aq_hw,
+					 HW_ATL2_MIF_SHARED_BUFFER_IN_ADR(i));
+}
+
+void hw_atl2_mif_shared_buf_write(struct aq_hw_s *aq_hw, int offset, u32 *data,
+				  int len)
+{
+	int j = 0;
+	int i;
+
+	for (i = offset; i < offset + len; i++, j++)
+		aq_hw_write_reg(aq_hw, HW_ATL2_MIF_SHARED_BUFFER_IN_ADR(i),
+				data[j]);
+}
+
+void hw_atl2_mif_shared_buf_read(struct aq_hw_s *aq_hw, int offset, u32 *data,
+				 int len)
+{
+	int j = 0;
+	int i;
+
+	for (i = offset; i < offset + len; i++, j++)
+		data[j] = aq_hw_read_reg(aq_hw,
+					 HW_ATL2_MIF_SHARED_BUFFER_OUT_ADR(i));
+}
+
+void hw_atl2_mif_host_finished_write_set(struct aq_hw_s *aq_hw, u32 finish)
+{
+	aq_hw_write_reg_bit(aq_hw, HW_ATL2_MIF_HOST_FINISHED_WRITE_ADR,
+			    HW_ATL2_MIF_HOST_FINISHED_WRITE_MSK,
+			    HW_ATL2_MIF_HOST_FINISHED_WRITE_SHIFT,
+			    finish);
+}
+
+u32 hw_atl2_mif_mcp_finished_read_get(struct aq_hw_s *aq_hw)
+{
+	return aq_hw_read_reg_bit(aq_hw, HW_ATL2_MIF_MCP_FINISHED_READ_ADR,
+				  HW_ATL2_MIF_MCP_FINISHED_READ_MSK,
+				  HW_ATL2_MIF_MCP_FINISHED_READ_SHIFT);
+}
+
+u32 hw_atl2_mif_mcp_boot_reg_get(struct aq_hw_s *aq_hw)
+{
+	return aq_hw_read_reg(aq_hw, HW_ATL2_MIF_BOOT_REG_ADR);
+}
+
+void hw_atl2_mif_mcp_boot_reg_set(struct aq_hw_s *aq_hw, u32 val)
+{
+	return aq_hw_write_reg(aq_hw, HW_ATL2_MIF_BOOT_REG_ADR, val);
+}
+
+u32 hw_atl2_mif_host_req_int_get(struct aq_hw_s *aq_hw)
+{
+	return aq_hw_read_reg(aq_hw, HW_ATL2_MCP_HOST_REQ_INT_ADR);
+}
+
+void hw_atl2_mif_host_req_int_clr(struct aq_hw_s *aq_hw, u32 val)
+{
+	return aq_hw_write_reg(aq_hw, HW_ATL2_MCP_HOST_REQ_INT_CLR_ADR,
+			       val);
+}
diff --git a/drivers/net/ethernet/aquantia/atlantic/hw_atl2/hw_atl2_llh.h b/drivers/net/ethernet/aquantia/atlantic/hw_atl2/hw_atl2_llh.h
new file mode 100644
index 000000000000..98c7a4621297
--- /dev/null
+++ b/drivers/net/ethernet/aquantia/atlantic/hw_atl2/hw_atl2_llh.h
@@ -0,0 +1,102 @@
+/* SPDX-License-Identifier: GPL-2.0-only */
+/* Atlantic Network Driver
+ * Copyright (C) 2020 Marvell International Ltd.
+ */
+
+#ifndef HW_ATL2_LLH_H
+#define HW_ATL2_LLH_H
+
+#include <linux/types.h>
+
+struct aq_hw_s;
+
+/* Set TX Interrupt Moderation Control Register */
+void hw_atl2_reg_tx_intr_moder_ctrl_set(struct aq_hw_s *aq_hw,
+					u32 tx_intr_moderation_ctl,
+					u32 queue);
+
+/* Set Redirection Table 2 Select */
+void hw_atl2_rpf_redirection_table2_select_set(struct aq_hw_s *aq_hw,
+					       u32 select);
+
+/** Set RSS HASH type */
+void hw_atl2_rpf_rss_hash_type_set(struct aq_hw_s *aq_hw, u32 rss_hash_type);
+
+/* set new RPF enable */
+void hw_atl2_rpf_new_enable_set(struct aq_hw_s *aq_hw, u32 enable);
+
+/* set l2 unicast filter tag */
+void hw_atl2_rpfl2_uc_flr_tag_set(struct aq_hw_s *aq_hw, u32 tag, u32 filter);
+
+/* set l2 broadcast filter tag */
+void hw_atl2_rpfl2_bc_flr_tag_set(struct aq_hw_s *aq_hw, u32 tag);
+
+/* set new rss redirection table */
+void hw_atl2_new_rpf_rss_redir_set(struct aq_hw_s *aq_hw, u32 tc, u32 index,
+				   u32 queue);
+
+/* Set VLAN filter tag */
+void hw_atl2_rpf_vlan_flr_tag_set(struct aq_hw_s *aq_hw, u32 tag, u32 filter);
+
+/* set tx random TC-queue mapping enable bit */
+void hw_atl2_tpb_tx_tc_q_rand_map_en_set(struct aq_hw_s *aq_hw,
+					 const u32 tc_q_rand_map_en);
+
+/* set tx buffer clock gate enable */
+void hw_atl2_tpb_tx_buf_clk_gate_en_set(struct aq_hw_s *aq_hw, u32 clk_gate_en);
+
+void hw_atl2_tps_tx_pkt_shed_data_arb_mode_set(struct aq_hw_s *aq_hw,
+					       const u32 data_arb_mode);
+
+/* set tx packet scheduler tc data max credit */
+void hw_atl2_tps_tx_pkt_shed_tc_data_max_credit_set(struct aq_hw_s *aq_hw,
+						    const u32 tc,
+						    const u32 max_credit);
+
+/* set tx packet scheduler tc data weight */
+void hw_atl2_tps_tx_pkt_shed_tc_data_weight_set(struct aq_hw_s *aq_hw,
+						const u32 tc,
+						const u32 weight);
+
+u32 hw_atl2_get_hw_version(struct aq_hw_s *aq_hw);
+
+void hw_atl2_init_launchtime(struct aq_hw_s *aq_hw);
+
+/* set action resolver record */
+void hw_atl2_rpf_act_rslvr_record_set(struct aq_hw_s *aq_hw, u8 location,
+				      u32 tag, u32 mask, u32 action);
+
+/* set enable action resolver section */
+void hw_atl2_rpf_act_rslvr_section_en_set(struct aq_hw_s *aq_hw, u32 sections);
+
+/* get data from firmware shared input buffer */
+void hw_atl2_mif_shared_buf_get(struct aq_hw_s *aq_hw, int offset, u32 *data,
+				int len);
+
+/* set data into firmware shared input buffer */
+void hw_atl2_mif_shared_buf_write(struct aq_hw_s *aq_hw, int offset, u32 *data,
+				  int len);
+
+/* get data from firmware shared output buffer */
+void hw_atl2_mif_shared_buf_read(struct aq_hw_s *aq_hw, int offset, u32 *data,
+				 int len);
+
+/* set host finished write shared buffer indication */
+void hw_atl2_mif_host_finished_write_set(struct aq_hw_s *aq_hw, u32 finish);
+
+/* get mcp finished read shared buffer indication */
+u32 hw_atl2_mif_mcp_finished_read_get(struct aq_hw_s *aq_hw);
+
+/* get mcp boot register */
+u32 hw_atl2_mif_mcp_boot_reg_get(struct aq_hw_s *aq_hw);
+
+/* set mcp boot register */
+void hw_atl2_mif_mcp_boot_reg_set(struct aq_hw_s *aq_hw, u32 val);
+
+/* get host interrupt request */
+u32 hw_atl2_mif_host_req_int_get(struct aq_hw_s *aq_hw);
+
+/* clear host interrupt request */
+void hw_atl2_mif_host_req_int_clr(struct aq_hw_s *aq_hw, u32 val);
+
+#endif /* HW_ATL2_LLH_H */
diff --git a/drivers/net/ethernet/aquantia/atlantic/hw_atl2/hw_atl2_llh_internal.h b/drivers/net/ethernet/aquantia/atlantic/hw_atl2/hw_atl2_llh_internal.h
new file mode 100644
index 000000000000..e34c5cda061e
--- /dev/null
+++ b/drivers/net/ethernet/aquantia/atlantic/hw_atl2/hw_atl2_llh_internal.h
@@ -0,0 +1,391 @@
+/* SPDX-License-Identifier: GPL-2.0-only */
+/* Atlantic Network Driver
+ * Copyright (C) 2020 Marvell International Ltd.
+ */
+
+#ifndef HW_ATL2_LLH_INTERNAL_H
+#define HW_ATL2_LLH_INTERNAL_H
+
+/* RX pif_rpf_redir_2_en_i Bitfield Definitions
+ * PORT="pif_rpf_redir_2_en_i"
+ */
+#define HW_ATL2_RPF_PIF_RPF_REDIR2_ENI_ADR 0x000054C8
+#define HW_ATL2_RPF_PIF_RPF_REDIR2_ENI_MSK 0x00001000
+#define HW_ATL2_RPF_PIF_RPF_REDIR2_ENI_MSKN 0xFFFFEFFF
+#define HW_ATL2_RPF_PIF_RPF_REDIR2_ENI_SHIFT 12
+#define HW_ATL2_RPF_PIF_RPF_REDIR2_ENI_WIDTH 1
+#define HW_ATL2_RPF_PIF_RPF_REDIR2_ENI_DEFAULT 0x0
+
+/* RX pif_rpf_rss_hash_type_i Bitfield Definitions
+ */
+#define HW_ATL2_RPF_PIF_RPF_RSS_HASH_TYPEI_ADR 0x000054C8
+#define HW_ATL2_RPF_PIF_RPF_RSS_HASH_TYPEI_MSK 0x000001FF
+#define HW_ATL2_RPF_PIF_RPF_RSS_HASH_TYPEI_MSKN 0xFFFFFE00
+#define HW_ATL2_RPF_PIF_RPF_RSS_HASH_TYPEI_SHIFT 0
+#define HW_ATL2_RPF_PIF_RPF_RSS_HASH_TYPEI_WIDTH 9
+
+/* rx rpf_new_rpf_en bitfield definitions
+ * preprocessor definitions for the bitfield "rpf_new_rpf_en_i".
+ * port="pif_rpf_new_rpf_en_i
+ */
+
+/* register address for bitfield rpf_new_rpf_en */
+#define HW_ATL2_RPF_NEW_EN_ADR 0x00005104
+/* bitmask for bitfield rpf_new_rpf_en */
+#define HW_ATL2_RPF_NEW_EN_MSK 0x00000800
+/* inverted bitmask for bitfield rpf_new_rpf_en */
+#define HW_ATL2_RPF_NEW_EN_MSKN 0xfffff7ff
+/* lower bit position of bitfield rpf_new_rpf_en */
+#define HW_ATL2_RPF_NEW_EN_SHIFT 11
+/* width of bitfield rpf_new_rpf_en */
+#define HW_ATL2_RPF_NEW_EN_WIDTH 1
+/* default value of bitfield rpf_new_rpf_en */
+#define HW_ATL2_RPF_NEW_EN_DEFAULT 0x0
+
+/* rx l2_uc_req_tag0{f}[5:0] bitfield definitions
+ * preprocessor definitions for the bitfield "l2_uc_req_tag0{f}[7:0]".
+ * parameter: filter {f} | stride size 0x8 | range [0, 37]
+ * port="pif_rpf_l2_uc_req_tag0[5:0]"
+ */
+
+/* register address for bitfield l2_uc_req_tag0{f}[2:0] */
+#define HW_ATL2_RPFL2UC_TAG_ADR(filter) (0x00005114 + (filter) * 0x8)
+/* bitmask for bitfield l2_uc_req_tag0{f}[2:0] */
+#define HW_ATL2_RPFL2UC_TAG_MSK 0x0FC00000
+/* inverted bitmask for bitfield l2_uc_req_tag0{f}[2:0] */
+#define HW_ATL2_RPFL2UC_TAG_MSKN 0xF03FFFFF
+/* lower bit position of bitfield l2_uc_req_tag0{f}[2:0] */
+#define HW_ATL2_RPFL2UC_TAG_SHIFT 22
+/* width of bitfield l2_uc_req_tag0{f}[2:0] */
+#define HW_ATL2_RPFL2UC_TAG_WIDTH 6
+/* default value of bitfield l2_uc_req_tag0{f}[2:0] */
+#define HW_ATL2_RPFL2UC_TAG_DEFAULT 0x0
+
+/* rpf_l2_bc_req_tag[5:0] bitfield definitions
+ * preprocessor definitions for the bitfield "rpf_l2_bc_req_tag[5:0]".
+ * port="pifrpf_l2_bc_req_tag_i[5:0]"
+ */
+
+/* register address for bitfield rpf_l2_bc_req_tag */
+#define HW_ATL2_RPF_L2_BC_TAG_ADR 0x000050F0
+/* bitmask for bitfield rpf_l2_bc_req_tag */
+#define HW_ATL2_RPF_L2_BC_TAG_MSK 0x0000003F
+/* inverted bitmask for bitfield rpf_l2_bc_req_tag */
+#define HW_ATL2_RPF_L2_BC_TAG_MSKN 0xffffffc0
+/* lower bit position of bitfield rpf_l2_bc_req_tag */
+#define HW_ATL2_RPF_L2_BC_TAG_SHIFT 0
+/* width of bitfield rpf_l2_bc_req_tag */
+#define HW_ATL2_RPF_L2_BC_TAG_WIDTH 6
+/* default value of bitfield rpf_l2_bc_req_tag */
+#define HW_ATL2_RPF_L2_BC_TAG_DEFAULT 0x0
+
+/* rx rpf_rss_red1_data_[4:0] bitfield definitions
+ * preprocessor definitions for the bitfield "rpf_rss_red1_data[4:0]".
+ * port="pif_rpf_rss_red1_data_i[4:0]"
+ */
+
+/* register address for bitfield rpf_rss_red1_data[4:0] */
+#define HW_ATL2_RPF_RSS_REDIR_ADR(TC, INDEX) (0x00006200 + \
+					(0x100 * !!((TC) > 3)) + (INDEX) * 4)
+/* bitmask for bitfield rpf_rss_red1_data[4:0] */
+#define HW_ATL2_RPF_RSS_REDIR_MSK(TC)  (0x00000001F << (5 * ((TC) % 4)))
+/* lower bit position of bitfield rpf_rss_red1_data[4:0] */
+#define HW_ATL2_RPF_RSS_REDIR_SHIFT(TC) (5 * ((TC) % 4))
+/* width of bitfield rpf_rss_red1_data[4:0] */
+#define HW_ATL2_RPF_RSS_REDIR_WIDTH 5
+/* default value of bitfield rpf_rss_red1_data[4:0] */
+#define HW_ATL2_RPF_RSS_REDIR_DEFAULT 0x0
+
+/* rx vlan_req_tag0{f}[3:0] bitfield definitions
+ * preprocessor definitions for the bitfield "vlan_req_tag0{f}[3:0]".
+ * parameter: filter {f} | stride size 0x4 | range [0, 15]
+ * port="pif_rpf_vlan_req_tag0[3:0]"
+ */
+
+/* register address for bitfield vlan_req_tag0{f}[3:0] */
+#define HW_ATL2_RPF_VL_TAG_ADR(filter) (0x00005290 + (filter) * 0x4)
+/* bitmask for bitfield vlan_req_tag0{f}[3:0] */
+#define HW_ATL2_RPF_VL_TAG_MSK 0x0000F000
+/* inverted bitmask for bitfield vlan_req_tag0{f}[3:0] */
+#define HW_ATL2_RPF_VL_TAG_MSKN 0xFFFF0FFF
+/* lower bit position of bitfield vlan_req_tag0{f}[3:0] */
+#define HW_ATL2_RPF_VL_TAG_SHIFT 12
+/* width of bitfield vlan_req_tag0{f}[3:0] */
+#define HW_ATL2_RPF_VL_TAG_WIDTH 4
+/* default value of bitfield vlan_req_tag0{f}[3:0] */
+#define HW_ATL2_RPF_VL_TAG_DEFAULT 0x0
+
+/* RX rx_q{Q}_tc_map[2:0] Bitfield Definitions
+ * Preprocessor definitions for the bitfield "rx_q{Q}_tc_map[2:0]".
+ * Parameter: Queue {Q} | bit-level stride | range [0, 31]
+ * PORT="pif_rx_q0_tc_map_i[2:0]"
+ */
+
+/* Register address for bitfield rx_q{Q}_tc_map[2:0] */
+#define HW_ATL2_RX_Q_TC_MAP_ADR(queue) \
+	(((queue) < 32) ? 0x00005900 + ((queue) / 8) * 4 : 0)
+/* Lower bit position of bitfield rx_q{Q}_tc_map[2:0] */
+#define HW_ATL2_RX_Q_TC_MAP_SHIFT(queue) \
+	(((queue) < 32) ? ((queue) * 4) % 32 : 0)
+/* Width of bitfield rx_q{Q}_tc_map[2:0] */
+#define HW_ATL2_RX_Q_TC_MAP_WIDTH 3
+/* Default value of bitfield rx_q{Q}_tc_map[2:0] */
+#define HW_ATL2_RX_Q_TC_MAP_DEFAULT 0x0
+
+/* tx tx_tc_q_rand_map_en bitfield definitions
+ * preprocessor definitions for the bitfield "tx_tc_q_rand_map_en".
+ * port="pif_tpb_tx_tc_q_rand_map_en_i"
+ */
+
+/* register address for bitfield tx_tc_q_rand_map_en */
+#define HW_ATL2_TPB_TX_TC_Q_RAND_MAP_EN_ADR 0x00007900
+/* bitmask for bitfield tx_tc_q_rand_map_en */
+#define HW_ATL2_TPB_TX_TC_Q_RAND_MAP_EN_MSK 0x00000200
+/* inverted bitmask for bitfield tx_tc_q_rand_map_en */
+#define HW_ATL2_TPB_TX_TC_Q_RAND_MAP_EN_MSKN 0xFFFFFDFF
+/* lower bit position of bitfield tx_tc_q_rand_map_en */
+#define HW_ATL2_TPB_TX_TC_Q_RAND_MAP_EN_SHIFT 9
+/* width of bitfield tx_tc_q_rand_map_en */
+#define HW_ATL2_TPB_TX_TC_Q_RAND_MAP_EN_WIDTH 1
+/* default value of bitfield tx_tc_q_rand_map_en */
+#define HW_ATL2_TPB_TX_TC_Q_RAND_MAP_EN_DEFAULT 0x0
+
+/* tx tx_buffer_clk_gate_en bitfield definitions
+ * preprocessor definitions for the bitfield "tx_buffer_clk_gate_en".
+ * port="pif_tpb_tx_buffer_clk_gate_en_i"
+ */
+
+/* register address for bitfield tx_buffer_clk_gate_en */
+#define HW_ATL2_TPB_TX_BUF_CLK_GATE_EN_ADR 0x00007900
+/* bitmask for bitfield tx_buffer_clk_gate_en */
+#define HW_ATL2_TPB_TX_BUF_CLK_GATE_EN_MSK 0x00000020
+/* inverted bitmask for bitfield tx_buffer_clk_gate_en */
+#define HW_ATL2_TPB_TX_BUF_CLK_GATE_EN_MSKN 0xffffffdf
+/* lower bit position of bitfield tx_buffer_clk_gate_en */
+#define HW_ATL2_TPB_TX_BUF_CLK_GATE_EN_SHIFT 5
+/* width of bitfield tx_buffer_clk_gate_en */
+#define HW_ATL2_TPB_TX_BUF_CLK_GATE_EN_WIDTH 1
+/* default value of bitfield tx_buffer_clk_gate_en */
+#define HW_ATL2_TPB_TX_BUF_CLK_GATE_EN_DEFAULT 0x0
+
+/* tx tx_q_tc_map{q} bitfield definitions
+ * preprocessor definitions for the bitfield "tx_q_tc_map{q}".
+ * parameter: queue {q} | bit-level stride | range [0, 31]
+ * port="pif_tpb_tx_q_tc_map0_i[2:0]"
+ */
+
+/* register address for bitfield tx_q_tc_map{q} */
+#define HW_ATL2_TX_Q_TC_MAP_ADR(queue) \
+	(((queue) < 32) ? 0x0000799C + ((queue) / 4) * 4 : 0)
+/* lower bit position of bitfield tx_q_tc_map{q} */
+#define HW_ATL2_TX_Q_TC_MAP_SHIFT(queue) \
+	(((queue) < 32) ? ((queue) * 8) % 32 : 0)
+/* width of bitfield tx_q_tc_map{q} */
+#define HW_ATL2_TX_Q_TC_MAP_WIDTH 3
+/* default value of bitfield tx_q_tc_map{q} */
+#define HW_ATL2_TX_Q_TC_MAP_DEFAULT 0x0
+
+/* tx data_tc_arb_mode bitfield definitions
+ * preprocessor definitions for the bitfield "data_tc_arb_mode".
+ * port="pif_tps_data_tc_arb_mode_i"
+ */
+
+/* register address for bitfield data_tc_arb_mode */
+#define HW_ATL2_TPS_DATA_TC_ARB_MODE_ADR 0x00007100
+/* bitmask for bitfield data_tc_arb_mode */
+#define HW_ATL2_TPS_DATA_TC_ARB_MODE_MSK 0x00000003
+/* inverted bitmask for bitfield data_tc_arb_mode */
+#define HW_ATL2_TPS_DATA_TC_ARB_MODE_MSKN 0xfffffffc
+/* lower bit position of bitfield data_tc_arb_mode */
+#define HW_ATL2_TPS_DATA_TC_ARB_MODE_SHIFT 0
+/* width of bitfield data_tc_arb_mode */
+#define HW_ATL2_TPS_DATA_TC_ARB_MODE_WIDTH 2
+/* default value of bitfield data_tc_arb_mode */
+#define HW_ATL2_TPS_DATA_TC_ARB_MODE_DEFAULT 0x0
+
+/* tx data_tc{t}_credit_max[f:0] bitfield definitions
+ * preprocessor definitions for the bitfield "data_tc{t}_credit_max[f:0]".
+ * parameter: tc {t} | stride size 0x4 | range [0, 7]
+ * port="pif_tps_data_tc0_credit_max_i[15:0]"
+ */
+
+/* register address for bitfield data_tc{t}_credit_max[f:0] */
+#define HW_ATL2_TPS_DATA_TCTCREDIT_MAX_ADR(tc) (0x00007110 + (tc) * 0x4)
+/* bitmask for bitfield data_tc{t}_credit_max[f:0] */
+#define HW_ATL2_TPS_DATA_TCTCREDIT_MAX_MSK 0xffff0000
+/* inverted bitmask for bitfield data_tc{t}_credit_max[f:0] */
+#define HW_ATL2_TPS_DATA_TCTCREDIT_MAX_MSKN 0x0000ffff
+/* lower bit position of bitfield data_tc{t}_credit_max[f:0] */
+#define HW_ATL2_TPS_DATA_TCTCREDIT_MAX_SHIFT 16
+/* width of bitfield data_tc{t}_credit_max[f:0] */
+#define HW_ATL2_TPS_DATA_TCTCREDIT_MAX_WIDTH 16
+/* default value of bitfield data_tc{t}_credit_max[f:0] */
+#define HW_ATL2_TPS_DATA_TCTCREDIT_MAX_DEFAULT 0x0
+
+/* tx data_tc{t}_weight[e:0] bitfield definitions
+ * preprocessor definitions for the bitfield "data_tc{t}_weight[e:0]".
+ * parameter: tc {t} | stride size 0x4 | range [0, 7]
+ * port="pif_tps_data_tc0_weight_i[14:0]"
+ */
+
+/* register address for bitfield data_tc{t}_weight[e:0] */
+#define HW_ATL2_TPS_DATA_TCTWEIGHT_ADR(tc) (0x00007110 + (tc) * 0x4)
+/* bitmask for bitfield data_tc{t}_weight[e:0] */
+#define HW_ATL2_TPS_DATA_TCTWEIGHT_MSK 0x00007fff
+/* inverted bitmask for bitfield data_tc{t}_weight[e:0] */
+#define HW_ATL2_TPS_DATA_TCTWEIGHT_MSKN 0xffff8000
+/* lower bit position of bitfield data_tc{t}_weight[e:0] */
+#define HW_ATL2_TPS_DATA_TCTWEIGHT_SHIFT 0
+/* width of bitfield data_tc{t}_weight[e:0] */
+#define HW_ATL2_TPS_DATA_TCTWEIGHT_WIDTH 15
+/* default value of bitfield data_tc{t}_weight[e:0] */
+#define HW_ATL2_TPS_DATA_TCTWEIGHT_DEFAULT 0x0
+
+/* tx interrupt moderation control register definitions
+ * Preprocessor definitions for TX Interrupt Moderation Control Register
+ * Base Address: 0x00007c28
+ * Parameter: queue {Q} | stride size 0x4 | range [0, 31]
+ */
+
+#define HW_ATL2_TX_INTR_MODERATION_CTL_ADR(queue) (0x00007c28u + (queue) * 0x40)
+
+/* Launch time control register */
+#define HW_ATL2_LT_CTRL_ADR 0x00007a1c
+
+#define HW_ATL2_LT_CTRL_AVB_LEN_CMP_TRSHLD_MSK 0xFFFF0000
+#define HW_ATL2_LT_CTRL_AVB_LEN_CMP_TRSHLD_SHIFT 16
+
+#define HW_ATL2_LT_CTRL_CLK_RATIO_MSK 0x0000FF00
+#define HW_ATL2_LT_CTRL_CLK_RATIO_SHIFT 8
+#define HW_ATL2_LT_CTRL_CLK_RATIO_QUATER_SPEED 4
+#define HW_ATL2_LT_CTRL_CLK_RATIO_HALF_SPEED 2
+#define HW_ATL2_LT_CTRL_CLK_RATIO_FULL_SPEED 1
+
+#define HW_ATL2_LT_CTRL_25G_MODE_SUPPORT_MSK 0x00000008
+#define HW_ATL2_LT_CTRL_25G_MODE_SUPPORT_SHIFT 3
+
+#define HW_ATL2_LT_CTRL_LINK_SPEED_MSK 0x00000007
+#define HW_ATL2_LT_CTRL_LINK_SPEED_SHIFT 0
+
+/* FPGA VER register */
+#define HW_ATL2_FPGA_VER_ADR 0x000000f4
+#define HW_ATL2_FPGA_VER_U32(mj, mi, bl, rv) \
+	((((mj) & 0xff) << 24) | \
+	 (((mi) & 0xff) << 16) | \
+	 (((bl) & 0xff) << 8) | \
+	 (((rv) & 0xff) << 0))
+
+/* ahb_mem_addr{f}[31:0] Bitfield Definitions
+ * Preprocessor definitions for the bitfield "ahb_mem_addr{f}[31:0]".
+ * Parameter: filter {f} | stride size 0x10 | range [0, 127]
+ * PORT="ahb_mem_addr{f}[31:0]"
+ */
+
+/* Register address for bitfield ahb_mem_addr{f}[31:0] */
+#define HW_ATL2_RPF_ACT_RSLVR_REQ_TAG_ADR(filter) \
+	(0x00014000u + (filter) * 0x10)
+/* Bitmask for bitfield ahb_mem_addr{f}[31:0] */
+#define HW_ATL2_RPF_ACT_RSLVR_REQ_TAG_MSK 0xFFFFFFFFu
+/* Inverted bitmask for bitfield ahb_mem_addr{f}[31:0] */
+#define HW_ATL2_RPF_ACT_RSLVR_REQ_TAG_MSKN 0x00000000u
+/* Lower bit position of bitfield ahb_mem_addr{f}[31:0] */
+#define HW_ATL2_RPF_ACT_RSLVR_REQ_TAG_SHIFT 0
+/* Width of bitfield ahb_mem_addr{f}[31:0] */
+#define HW_ATL2_RPF_ACT_RSLVR_REQ_TAG_WIDTH 31
+/* Default value of bitfield ahb_mem_addr{f}[31:0] */
+#define HW_ATL2_RPF_ACT_RSLVR_REQ_TAG_DEFAULT 0x0
+
+/* Register address for bitfield ahb_mem_addr{f}[31:0] */
+#define HW_ATL2_RPF_ACT_RSLVR_TAG_MASK_ADR(filter) \
+	(0x00014004u + (filter) * 0x10)
+/* Bitmask for bitfield ahb_mem_addr{f}[31:0] */
+#define HW_ATL2_RPF_ACT_RSLVR_TAG_MASK_MSK 0xFFFFFFFFu
+/* Inverted bitmask for bitfield ahb_mem_addr{f}[31:0] */
+#define HW_ATL2_RPF_ACT_RSLVR_TAG_MASK_MSKN 0x00000000u
+/* Lower bit position of bitfield ahb_mem_addr{f}[31:0] */
+#define HW_ATL2_RPF_ACT_RSLVR_TAG_MASK_SHIFT 0
+/* Width of bitfield ahb_mem_addr{f}[31:0] */
+#define HW_ATL2_RPF_ACT_RSLVR_TAG_MASK_WIDTH 31
+/* Default value of bitfield ahb_mem_addr{f}[31:0] */
+#define HW_ATL2_RPF_ACT_RSLVR_TAG_MASK_DEFAULT 0x0
+
+/* Register address for bitfield ahb_mem_addr{f}[31:0] */
+#define HW_ATL2_RPF_ACT_RSLVR_ACTN_ADR(filter) \
+	(0x00014008u + (filter) * 0x10)
+/* Bitmask for bitfield ahb_mem_addr{f}[31:0] */
+#define HW_ATL2_RPF_ACT_RSLVR_ACTN_MSK 0x000007FFu
+/* Inverted bitmask for bitfield ahb_mem_addr{f}[31:0] */
+#define HW_ATL2_RPF_ACT_RSLVR_ACTN_MSKN 0xFFFFF800u
+/* Lower bit position of bitfield ahb_mem_addr{f}[31:0] */
+#define HW_ATL2_RPF_ACT_RSLVR_ACTN_SHIFT 0
+/* Width of bitfield ahb_mem_addr{f}[31:0] */
+#define HW_ATL2_RPF_ACT_RSLVR_ACTN_WIDTH 10
+/* Default value of bitfield ahb_mem_addr{f}[31:0] */
+#define HW_ATL2_RPF_ACT_RSLVR_ACTN_DEFAULT 0x0
+
+/* rpf_rec_tab_en[15:0] Bitfield Definitions
+ * Preprocessor definitions for the bitfield "rpf_rec_tab_en[15:0]".
+ * PORT="pif_rpf_rec_tab_en[15:0]"
+ */
+/* Register address for bitfield rpf_rec_tab_en[15:0] */
+#define HW_ATL2_RPF_REC_TAB_EN_ADR 0x00006ff0u
+/* Bitmask for bitfield rpf_rec_tab_en[15:0] */
+#define HW_ATL2_RPF_REC_TAB_EN_MSK 0x0000FFFFu
+/* Inverted bitmask for bitfield rpf_rec_tab_en[15:0] */
+#define HW_ATL2_RPF_REC_TAB_EN_MSKN 0xFFFF0000u
+/* Lower bit position of bitfield rpf_rec_tab_en[15:0] */
+#define HW_ATL2_RPF_REC_TAB_EN_SHIFT 0
+/* Width of bitfield rpf_rec_tab_en[15:0] */
+#define HW_ATL2_RPF_REC_TAB_EN_WIDTH 16
+/* Default value of bitfield rpf_rec_tab_en[15:0] */
+#define HW_ATL2_RPF_REC_TAB_EN_DEFAULT 0x0
+
+/* Register address for firmware shared input buffer */
+#define HW_ATL2_MIF_SHARED_BUFFER_IN_ADR(dword) (0x00012000U + (dword) * 0x4U)
+/* Register address for firmware shared output buffer */
+#define HW_ATL2_MIF_SHARED_BUFFER_OUT_ADR(dword) (0x00013000U + (dword) * 0x4U)
+
+/* pif_host_finished_buf_wr_i Bitfield Definitions
+ * Preprocessor definitions for the bitfield "pif_host_finished_buf_wr_i".
+ * PORT="pif_host_finished_buf_wr_i"
+ */
+/* Register address for bitfield rpif_host_finished_buf_wr_i */
+#define HW_ATL2_MIF_HOST_FINISHED_WRITE_ADR 0x00000e00u
+/* Bitmask for bitfield pif_host_finished_buf_wr_i */
+#define HW_ATL2_MIF_HOST_FINISHED_WRITE_MSK 0x00000001u
+/* Inverted bitmask for bitfield pif_host_finished_buf_wr_i */
+#define HW_ATL2_MIF_HOST_FINISHED_WRITE_MSKN 0xFFFFFFFEu
+/* Lower bit position of bitfield pif_host_finished_buf_wr_i */
+#define HW_ATL2_MIF_HOST_FINISHED_WRITE_SHIFT 0
+/* Width of bitfield pif_host_finished_buf_wr_i */
+#define HW_ATL2_MIF_HOST_FINISHED_WRITE_WIDTH 1
+/* Default value of bitfield pif_host_finished_buf_wr_i */
+#define HW_ATL2_MIF_HOST_FINISHED_WRITE_DEFAULT 0x0
+
+/* pif_mcp_finished_buf_rd_i Bitfield Definitions
+ * Preprocessor definitions for the bitfield "pif_mcp_finished_buf_rd_i".
+ * PORT="pif_mcp_finished_buf_rd_i"
+ */
+/* Register address for bitfield pif_mcp_finished_buf_rd_i */
+#define HW_ATL2_MIF_MCP_FINISHED_READ_ADR 0x00000e04u
+/* Bitmask for bitfield pif_mcp_finished_buf_rd_i */
+#define HW_ATL2_MIF_MCP_FINISHED_READ_MSK 0x00000001u
+/* Inverted bitmask for bitfield pif_mcp_finished_buf_rd_i */
+#define HW_ATL2_MIF_MCP_FINISHED_READ_MSKN 0xFFFFFFFEu
+/* Lower bit position of bitfield pif_mcp_finished_buf_rd_i */
+#define HW_ATL2_MIF_MCP_FINISHED_READ_SHIFT 0
+/* Width of bitfield pif_mcp_finished_buf_rd_i */
+#define HW_ATL2_MIF_MCP_FINISHED_READ_WIDTH 1
+/* Default value of bitfield pif_mcp_finished_buf_rd_i */
+#define HW_ATL2_MIF_MCP_FINISHED_READ_DEFAULT 0x0
+
+/* Register address for bitfield pif_mcp_boot_reg */
+#define HW_ATL2_MIF_BOOT_REG_ADR 0x00003040u
+
+#define HW_ATL2_MCP_HOST_REQ_INT_READY BIT(0)
+
+#define HW_ATL2_MCP_HOST_REQ_INT_ADR 0x00000F00u
+#define HW_ATL2_MCP_HOST_REQ_INT_SET_ADR 0x00000F04u
+#define HW_ATL2_MCP_HOST_REQ_INT_CLR_ADR 0x00000F08u
+
+#endif /* HW_ATL2_LLH_INTERNAL_H */
diff --git a/drivers/net/ethernet/aquantia/atlantic/hw_atl2/hw_atl2_utils.c b/drivers/net/ethernet/aquantia/atlantic/hw_atl2/hw_atl2_utils.c
new file mode 100644
index 000000000000..0fe6257d9c08
--- /dev/null
+++ b/drivers/net/ethernet/aquantia/atlantic/hw_atl2/hw_atl2_utils.c
@@ -0,0 +1,130 @@
+// SPDX-License-Identifier: GPL-2.0-only
+/* Atlantic Network Driver
+ * Copyright (C) 2020 Marvell International Ltd.
+ */
+
+#include <linux/iopoll.h>
+
+#include "aq_hw_utils.h"
+#include "hw_atl/hw_atl_utils.h"
+#include "hw_atl2_utils.h"
+#include "hw_atl2_llh.h"
+#include "hw_atl2_llh_internal.h"
+
+#define HW_ATL2_FW_VER_1X          0x01000000U
+
+#define AQ_A2_BOOT_STARTED         BIT(0x18)
+#define AQ_A2_CRASH_INIT           BIT(0x1B)
+#define AQ_A2_BOOT_CODE_FAILED     BIT(0x1C)
+#define AQ_A2_FW_INIT_FAILED       BIT(0x1D)
+#define AQ_A2_FW_INIT_COMP_SUCCESS BIT(0x1F)
+
+#define AQ_A2_FW_BOOT_FAILED_MASK (AQ_A2_CRASH_INIT | \
+				   AQ_A2_BOOT_CODE_FAILED | \
+				   AQ_A2_FW_INIT_FAILED)
+#define AQ_A2_FW_BOOT_COMPLETE_MASK (AQ_A2_FW_BOOT_FAILED_MASK | \
+				     AQ_A2_FW_INIT_COMP_SUCCESS)
+
+#define AQ_A2_FW_BOOT_REQ_REBOOT        BIT(0x0)
+#define AQ_A2_FW_BOOT_REQ_HOST_BOOT     BIT(0x8)
+#define AQ_A2_FW_BOOT_REQ_MAC_FAST_BOOT BIT(0xA)
+#define AQ_A2_FW_BOOT_REQ_PHY_FAST_BOOT BIT(0xB)
+
+int hw_atl2_utils_initfw(struct aq_hw_s *self, const struct aq_fw_ops **fw_ops)
+{
+	int err;
+
+	self->fw_ver_actual = hw_atl2_utils_get_fw_version(self);
+
+	if (hw_atl_utils_ver_match(HW_ATL2_FW_VER_1X, self->fw_ver_actual)) {
+		*fw_ops = &aq_a2_fw_ops;
+	} else {
+		aq_pr_err("Bad FW version detected: %x, but continue\n",
+			  self->fw_ver_actual);
+		*fw_ops = &aq_a2_fw_ops;
+	}
+	aq_pr_trace("Detect ATL2FW %x\n", self->fw_ver_actual);
+	self->aq_fw_ops = *fw_ops;
+	err = self->aq_fw_ops->init(self);
+
+	self->chip_features |= ATL_HW_CHIP_ANTIGUA;
+
+	return err;
+}
+
+static bool hw_atl2_mcp_boot_complete(struct aq_hw_s *self)
+{
+	u32 rbl_status;
+
+	rbl_status = hw_atl2_mif_mcp_boot_reg_get(self);
+	if (rbl_status & AQ_A2_FW_BOOT_COMPLETE_MASK)
+		return true;
+
+	/* Host boot requested */
+	if (hw_atl2_mif_host_req_int_get(self) & HW_ATL2_MCP_HOST_REQ_INT_READY)
+		return true;
+
+	return false;
+}
+
+int hw_atl2_utils_soft_reset(struct aq_hw_s *self)
+{
+	bool rbl_complete = false;
+	u32 rbl_status = 0;
+	u32 rbl_request;
+	int err;
+
+	hw_atl2_mif_host_req_int_clr(self, 0x01);
+	rbl_request = AQ_A2_FW_BOOT_REQ_REBOOT;
+#ifdef AQ_CFG_FAST_START
+	rbl_request |= AQ_A2_FW_BOOT_REQ_MAC_FAST_BOOT;
+#endif
+	hw_atl2_mif_mcp_boot_reg_set(self, rbl_request);
+
+	/* Wait for RBL boot */
+	err = readx_poll_timeout_atomic(hw_atl2_mif_mcp_boot_reg_get, self,
+				rbl_status,
+				((rbl_status & AQ_A2_BOOT_STARTED) &&
+				 (rbl_status != 0xFFFFFFFFu)),
+				10, 200000);
+	if (err) {
+		aq_pr_err("Boot code hanged");
+		goto err_exit;
+	}
+
+	err = readx_poll_timeout_atomic(hw_atl2_mcp_boot_complete, self,
+					rbl_complete,
+					rbl_complete,
+					10, 2000000);
+
+	if (err) {
+		aq_pr_err("FW Restart timed out");
+		goto err_exit;
+	}
+
+	rbl_status = hw_atl2_mif_mcp_boot_reg_get(self);
+
+	if (rbl_status & AQ_A2_FW_BOOT_FAILED_MASK) {
+		err = -EIO;
+		aq_pr_err("FW Restart failed");
+		goto err_exit;
+	}
+
+	if (hw_atl2_mif_host_req_int_get(self) &
+	    HW_ATL2_MCP_HOST_REQ_INT_READY) {
+		err = -EIO;
+		aq_pr_err("No FW detected. Dynamic FW load not implemented");
+		goto err_exit;
+	}
+
+	if (self->aq_fw_ops) {
+		err = self->aq_fw_ops->init(self);
+		if (err) {
+			aq_pr_err("FW Init failed");
+			goto err_exit;
+		}
+	}
+
+err_exit:
+	return err;
+}
diff --git a/drivers/net/ethernet/aquantia/atlantic/hw_atl2/hw_atl2_utils.h b/drivers/net/ethernet/aquantia/atlantic/hw_atl2/hw_atl2_utils.h
new file mode 100644
index 000000000000..6bad64c77b87
--- /dev/null
+++ b/drivers/net/ethernet/aquantia/atlantic/hw_atl2/hw_atl2_utils.h
@@ -0,0 +1,636 @@
+/* SPDX-License-Identifier: GPL-2.0-only */
+/* Atlantic Network Driver
+ * Copyright (C) 2020 Marvell International Ltd.
+ */
+
+#ifndef HW_ATL2_UTILS_H
+#define HW_ATL2_UTILS_H
+
+#include "aq_hw.h"
+
+/* F W    A P I */
+
+struct link_options_s {
+	u8 link_up:1;
+	u8 link_renegotiate:1;
+	u8 minimal_link_speed:1;
+	u8 internal_loopback:1;
+	u8 external_loopback:1;
+	u8 rate_10M_hd:1;
+	u8 rate_100M_hd:1;
+	u8 rate_1G_hd:1;
+
+	u8 rate_10M:1;
+	u8 rate_100M:1;
+	u8 rate_1G:1;
+	u8 rate_2P5G:1;
+	u8 rate_N2P5G:1;
+	u8 rate_5G:1;
+	u8 rate_N5G:1;
+	u8 rate_10G:1;
+
+	u8 eee_100M:1;
+	u8 eee_1G:1;
+	u8 eee_2P5G:1;
+	u8 eee_5G:1;
+	u8 eee_10G:1;
+	u8 rsvd3:3;
+
+	u8 pause_rx:1;
+	u8 pause_tx:1;
+	u8 rsvd4:1;
+	u8 downshift:1;
+	u8 downshift_retry:4;
+};
+
+struct link_control_s {
+	u8 mode:4;
+	u8 disable_crc_corruption:1;
+	u8 discard_short_frames:1;
+	u8 flow_control_mode:1;
+	u8 disable_length_check:1;
+
+	u8 discard_errored_frames:1;
+	u8 control_frame_enable:1;
+	u8 enable_tx_padding:1;
+	u8 enable_crc_forwarding:1;
+	u8 enable_frame_padding_removal_rx: 1;
+	u8 promiscuous_mode: 1;
+	u8 rsvd:2;
+
+	u16 rsvd2;
+};
+
+struct thermal_shutdown_s {
+	u8 enable:1;
+	u8 warning_enable:1;
+	u8 rsvd:6;
+
+	u8 shutdown_temperature;
+	u8 cold_temperature;
+	u8 warning_temperature;
+};
+
+struct mac_address_s {
+	u8 mac_address[6];
+};
+
+struct mac_address_aligned_s {
+	struct mac_address_s aligned;
+	u16 rsvd;
+};
+
+struct sleep_proxy_s {
+	struct wake_on_lan_s {
+		u8 wake_on_magic_packet:1;
+		u8 wake_on_pattern:1;
+		u8 wake_on_link_up:1;
+		u8 wake_on_link_down:1;
+		u8 wake_on_ping:1;
+		u8 wake_on_timer:1;
+		u8 rsvd:2;
+
+		u8 rsvd2;
+		u16 rsvd3;
+
+		u32 link_up_timeout;
+		u32 link_down_timeout;
+		u32 timer;
+	} wake_on_lan;
+
+	struct {
+		u32 mask[4];
+		u32 crc32;
+	} wake_up_pattern[8];
+
+	struct __packed {
+		u8 arp_responder:1;
+		u8 echo_responder:1;
+		u8 igmp_client:1;
+		u8 echo_truncate:1;
+		u8 address_guard:1;
+		u8 ignore_fragmented:1;
+		u8 rsvd:2;
+
+		u16 echo_max_len;
+		u8 rsvd2;
+	} ipv4_offload;
+
+	u32 ipv4_offload_addr[8];
+	u32 reserved[8];
+
+	struct __packed {
+		u8 ns_responder:1;
+		u8 echo_responder:1;
+		u8 mld_client:1;
+		u8 echo_truncate:1;
+		u8 address_guard:1;
+		u8 rsvd:3;
+
+		u16 echo_max_len;
+		u8 rsvd2;
+	} ipv6_offload;
+
+	u32 ipv6_offload_addr[16][4];
+
+	struct {
+		u16 port[16];
+	} tcp_port_offload;
+
+	struct {
+		u16 port[16];
+	} udp_port_offload;
+
+	struct {
+		u32 retry_count;
+		u32 retry_interval;
+	} ka4_offload;
+
+	struct {
+		u32 timeout;
+		u16 local_port;
+		u16 remote_port;
+		u8 remote_mac_addr[6];
+		u16 rsvd;
+		u32 rsvd2;
+		u32 rsvd3;
+		u16 rsvd4;
+		u16 win_size;
+		u32 seq_num;
+		u32 ack_num;
+		u32 local_ip;
+		u32 remote_ip;
+	} ka4_connection[16];
+
+	struct {
+		u32 retry_count;
+		u32 retry_interval;
+	} ka6_offload;
+
+	struct {
+		u32 timeout;
+		u16 local_port;
+		u16 remote_port;
+		u8 remote_mac_addr[6];
+		u16 rsvd;
+		u32 rsvd2;
+		u32 rsvd3;
+		u16 rsvd4;
+		u16 win_size;
+		u32 seq_num;
+		u32 ack_num;
+		u32 local_ip[4];
+		u32 remote_ip[4];
+	} ka6_connection[16];
+
+	struct {
+		u32 rr_count;
+		u32 rr_buf_len;
+		u32 idx_offset;
+		u32 rr__offset;
+	} mdns_offload;
+};
+
+struct pause_quanta_s {
+	u16 quanta_10M;
+	u16 threshold_10M;
+	u16 quanta_100M;
+	u16 threshold_100M;
+	u16 quanta_1G;
+	u16 threshold_1G;
+	u16 quanta_2P5G;
+	u16 threshold_2P5G;
+	u16 quanta_5G;
+	u16 threshold_5G;
+	u16 quanta_10G;
+	u16 threshold_10G;
+};
+
+struct data_buffer_status_s {
+	u32 data_offset;
+	u32 data_length;
+};
+
+struct device_caps_s {
+	u8 finite_flashless:1;
+	u8 cable_diag:1;
+	u8 ncsi:1;
+	u8 avb:1;
+	u8 rsvd:4;
+
+	u8 rsvd2;
+	u16 rsvd3;
+	u32 rsvd4;
+};
+
+struct version_s {
+	struct bundle_version_t {
+		u8 major;
+		u8 minor;
+		u16 build;
+	} bundle;
+	struct mac_version_t {
+		u8 major;
+		u8 minor;
+		u16 build;
+	} mac;
+	struct phy_version_t {
+		u8 major;
+		u8 minor;
+		u16 build;
+	} phy;
+	u32 drv_iface_ver:4;
+	u32 rsvd:28;
+};
+
+struct link_status_s {
+	u8 link_state:4;
+	u8 link_rate:4;
+
+	u8 pause_tx:1;
+	u8 pause_rx:1;
+	u8 eee:1;
+	u8 duplex:1;
+	u8 rsvd:4;
+
+	u16 rsvd2;
+};
+
+struct wol_status_s {
+	u8 wake_count;
+	u8 wake_reason;
+
+	u16 wake_up_packet_length :12;
+	u16 wake_up_pattern_number :3;
+	u16 rsvd:1;
+
+	u32 wake_up_packet[379];
+};
+
+struct mac_health_monitor_s {
+	u8 mac_ready:1;
+	u8 mac_fault:1;
+	u8 mac_flashless_finished:1;
+	u8 rsvd:5;
+
+	u8 mac_temperature;
+	u16 mac_heart_beat;
+	u16 mac_fault_code;
+	u16 rsvd2;
+};
+
+struct phy_health_monitor_s {
+	u8 phy_ready:1;
+	u8 phy_fault:1;
+	u8 phy_hot_warning:1;
+	u8 rsvd:5;
+
+	u8 phy_temperature;
+	u16 phy_heart_beat;
+	u16 phy_fault_code;
+	u16 rsvd2;
+};
+
+struct device_link_caps_s {
+	u8 rsvd:3;
+	u8 internal_loopback:1;
+	u8 external_loopback:1;
+	u8 rate_10M_hd:1;
+	u8 rate_100M_hd:1;
+	u8 rate_1G_hd:1;
+
+	u8 rate_10M:1;
+	u8 rate_100M:1;
+	u8 rate_1G:1;
+	u8 rate_2P5G:1;
+	u8 rate_N2P5G:1;
+	u8 rate_5G:1;
+	u8 rate_N5G:1;
+	u8 rate_10G:1;
+
+	u8 rsvd3:1;
+	u8 eee_100M:1;
+	u8 eee_1G:1;
+	u8 eee_2P5G:1;
+	u8 rsvd4:1;
+	u8 eee_5G:1;
+	u8 rsvd5:1;
+	u8 eee_10G:1;
+
+	u8 pause_rx:1;
+	u8 pause_tx:1;
+	u8 pfc:1;
+	u8 downshift:1;
+	u8 downshift_retry:4;
+};
+
+struct sleep_proxy_caps_s {
+	u8 ipv4_offload:1;
+	u8 ipv6_offload:1;
+	u8 tcp_port_offload:1;
+	u8 udp_port_offload:1;
+	u8 ka4_offload:1;
+	u8 ka6_offload:1;
+	u8 mdns_offload:1;
+	u8 wake_on_ping:1;
+
+	u8 wake_on_magic_packet:1;
+	u8 wake_on_pattern:1;
+	u8 wake_on_timer:1;
+	u8 wake_on_link:1;
+	u8 wake_patterns_count:4;
+
+	u8 ipv4_count;
+	u8 ipv6_count;
+
+	u8 tcp_port_offload_count;
+	u8 udp_port_offload_count;
+
+	u8 tcp4_ka_count;
+	u8 tcp6_ka_count;
+
+	u8 igmp_offload:1;
+	u8 mld_offload:1;
+	u8 rsvd:6;
+
+	u8 rsvd2;
+	u16 rsvd3;
+};
+
+struct lkp_link_caps_s {
+	u8 rsvd:5;
+	u8 rate_10M_hd:1;
+	u8 rate_100M_hd:1;
+	u8 rate_1G_hd:1;
+
+	u8 rate_10M:1;
+	u8 rate_100M:1;
+	u8 rate_1G:1;
+	u8 rate_2P5G:1;
+	u8 rate_N2P5G:1;
+	u8 rate_5G:1;
+	u8 rate_N5G:1;
+	u8 rate_10G:1;
+
+	u8 rsvd2:1;
+	u8 eee_100M:1;
+	u8 eee_1G:1;
+	u8 eee_2P5G:1;
+	u8 rsvd3:1;
+	u8 eee_5G:1;
+	u8 rsvd4:1;
+	u8 eee_10G:1;
+
+	u8 pause_rx:1;
+	u8 pause_tx:1;
+	u8 rsvd5:6;
+};
+
+struct core_dump_s {
+	u32 reg0;
+	u32 reg1;
+	u32 reg2;
+
+	u32 hi;
+	u32 lo;
+
+	u32 regs[32];
+};
+
+struct trace_s {
+	u32 sync_counter;
+	u32 mem_buffer[0x1ff];
+};
+
+struct cable_diag_control_s {
+	u8 toggle :1;
+	u8 rsvd:7;
+
+	u8 wait_timeout_sec;
+	u16 rsvd2;
+};
+
+struct cable_diag_lane_data_s {
+	u8 result_code;
+	u8 dist;
+	u8 far_dist;
+	u8 rsvd;
+};
+
+struct cable_diag_status_s {
+	struct cable_diag_lane_data_s lane_data[4];
+	u8 transact_id;
+	u8 status:4;
+	u8 rsvd:4;
+	u16 rsvd2;
+};
+
+struct statistics_a0_s {
+	struct {
+		u32 link_up;
+		u32 link_down;
+	} link;
+
+	struct {
+		u64 tx_unicast_octets;
+		u64 tx_multicast_octets;
+		u64 tx_broadcast_octets;
+		u64 rx_unicast_octets;
+		u64 rx_multicast_octets;
+		u64 rx_broadcast_octets;
+
+		u32 tx_unicast_frames;
+		u32 tx_multicast_frames;
+		u32 tx_broadcast_frames;
+		u32 tx_errors;
+
+		u32 rx_unicast_frames;
+		u32 rx_multicast_frames;
+		u32 rx_broadcast_frames;
+		u32 rx_dropped_frames;
+		u32 rx_error_frames;
+
+		u32 tx_good_frames;
+		u32 rx_good_frames;
+		u32 reserve_fw_gap;
+	} msm;
+	u32 main_loop_cycles;
+	u32 reserve_fw_gap;
+};
+
+struct __packed statistics_b0_s {
+	u64 rx_good_octets;
+	u64 rx_pause_frames;
+	u64 rx_good_frames;
+	u64 rx_errors;
+	u64 rx_unicast_frames;
+	u64 rx_multicast_frames;
+	u64 rx_broadcast_frames;
+
+	u64 tx_good_octets;
+	u64 tx_pause_frames;
+	u64 tx_good_frames;
+	u64 tx_errors;
+	u64 tx_unicast_frames;
+	u64 tx_multicast_frames;
+	u64 tx_broadcast_frames;
+
+	u32 main_loop_cycles;
+};
+
+struct __packed statistics_s {
+	union __packed {
+		struct statistics_a0_s a0;
+		struct statistics_b0_s b0;
+	};
+};
+
+struct filter_caps_s {
+	u8 l2_filters_base_index:6;
+	u8 flexible_filter_mask:2;
+	u8 l2_filter_count;
+	u8 ethertype_filter_base_index;
+	u8 ethertype_filter_count;
+
+	u8 vlan_filter_base_index;
+	u8 vlan_filter_count;
+	u8 l3_ip4_filter_base_index:4;
+	u8 l3_ip4_filter_count:4;
+	u8 l3_ip6_filter_base_index:4;
+	u8 l3_ip6_filter_count:4;
+
+	u8 l4_filter_base_index:4;
+	u8 l4_filter_count:4;
+	u8 l4_flex_filter_base_index:4;
+	u8 l4_flex_filter_count:4;
+	u8 rslv_tbl_base_index;
+	u8 rslv_tbl_count;
+};
+
+struct request_policy_s {
+	struct {
+		u8 all:1;
+		u8 mcast:1;
+		u8 rx_queue_tc_index:5;
+		u8 queue_or_tc:1;
+	} promisc;
+
+	struct {
+		u8 accept:1;
+		u8 rsvd:1;
+		u8 rx_queue_tc_index:5;
+		u8 queue_or_tc:1;
+	} bcast;
+
+	struct {
+		u8 accept:1;
+		u8 rsvd:1;
+		u8 rx_queue_tc_index:5;
+		u8 queue_or_tc:1;
+	} mcast;
+
+	u8 rsvd:8;
+};
+
+struct fw_interface_in {
+	u32 mtu;
+	u32 rsvd1;
+	struct mac_address_aligned_s mac_address;
+	struct link_control_s link_control;
+	u32 rsvd2;
+	struct link_options_s link_options;
+	u32 rsvd3;
+	struct thermal_shutdown_s thermal_shutdown;
+	u32 rsvd4;
+	struct sleep_proxy_s sleep_proxy;
+	u32 rsvd5;
+	struct pause_quanta_s pause_quanta[8];
+	struct cable_diag_control_s cable_diag_control;
+	u32 rsvd6;
+	struct data_buffer_status_s data_buffer_status;
+	u32 rsvd7;
+	struct request_policy_s request_policy;
+};
+
+struct transaction_counter_s {
+	u16 transaction_cnt_a;
+	u16 transaction_cnt_b;
+};
+
+struct management_status_s {
+	struct mac_address_s mac_address;
+	u16 vlan;
+
+	struct{
+		u32 enable : 1;
+		u32 rsvd:31;
+	} flags;
+
+	u32 rsvd1;
+	u32 rsvd2;
+	u32 rsvd3;
+	u32 rsvd4;
+	u32 rsvd5;
+};
+
+struct __packed fw_interface_out {
+	struct transaction_counter_s transaction_id;
+	struct version_s version;
+	struct link_status_s link_status;
+	struct wol_status_s wol_status;
+	u32 rsvd;
+	u32 rsvd2;
+	struct mac_health_monitor_s mac_health_monitor;
+	u32 rsvd3;
+	u32 rsvd4;
+	struct phy_health_monitor_s phy_health_monitor;
+	u32 rsvd5;
+	u32 rsvd6;
+	struct cable_diag_status_s cable_diag_status;
+	u32 rsvd7;
+	struct device_link_caps_s device_link_caps;
+	u32 rsvd8;
+	struct sleep_proxy_caps_s sleep_proxy_caps;
+	u32 rsvd9;
+	struct lkp_link_caps_s lkp_link_caps;
+	u32 rsvd10;
+	struct core_dump_s core_dump;
+	u32 rsvd11;
+	struct statistics_s stats;
+	struct filter_caps_s filter_caps;
+	struct device_caps_s device_caps;
+	u32 rsvd13;
+	struct management_status_s management_status;
+	u32 reserve[21];
+	struct trace_s trace;
+};
+
+#define  AQ_A2_FW_LINK_RATE_INVALID 0
+#define  AQ_A2_FW_LINK_RATE_10M     1
+#define  AQ_A2_FW_LINK_RATE_100M    2
+#define  AQ_A2_FW_LINK_RATE_1G      3
+#define  AQ_A2_FW_LINK_RATE_2G5     4
+#define  AQ_A2_FW_LINK_RATE_5G      5
+#define  AQ_A2_FW_LINK_RATE_10G     6
+
+#define  AQ_HOST_MODE_INVALID      0U
+#define  AQ_HOST_MODE_ACTIVE       1U
+#define  AQ_HOST_MODE_SLEEP_PROXY  2U
+#define  AQ_HOST_MODE_LOW_POWER    3U
+#define  AQ_HOST_MODE_SHUTDOWN     4U
+
+#define  AQ_A2_FW_INTERFACE_A0     0
+#define  AQ_A2_FW_INTERFACE_B0     1
+
+int hw_atl2_utils_initfw(struct aq_hw_s *self, const struct aq_fw_ops **fw_ops);
+
+int hw_atl2_utils_soft_reset(struct aq_hw_s *self);
+
+u32 hw_atl2_utils_get_fw_version(struct aq_hw_s *self);
+
+int hw_atl2_utils_get_action_resolve_table_caps(struct aq_hw_s *self,
+						u8 *base_index, u8 *count);
+
+extern const struct aq_fw_ops aq_a2_fw_ops;
+
+#endif /* HW_ATL2_UTILS_H */
diff --git a/drivers/net/ethernet/aquantia/atlantic/hw_atl2/hw_atl2_utils_fw.c b/drivers/net/ethernet/aquantia/atlantic/hw_atl2/hw_atl2_utils_fw.c
new file mode 100644
index 000000000000..7370e3f76b62
--- /dev/null
+++ b/drivers/net/ethernet/aquantia/atlantic/hw_atl2/hw_atl2_utils_fw.c
@@ -0,0 +1,655 @@
+// SPDX-License-Identifier: GPL-2.0-only
+/* Atlantic Network Driver
+ * Copyright (C) 2020 Marvell International Ltd.
+ */
+
+#include <linux/iopoll.h>
+
+#include "aq_hw.h"
+#include "aq_hw_utils.h"
+#include "aq_nic.h"
+#include "hw_atl/hw_atl_llh.h"
+#include "hw_atl2_utils.h"
+#include "hw_atl2_llh.h"
+#include "hw_atl2_internal.h"
+
+#define AQ_A2_FW_READ_TRY_MAX 1000
+
+#define hw_atl2_shared_buffer_write(HW, ITEM, VARIABLE) \
+{\
+	BUILD_BUG_ON_MSG((offsetof(struct fw_interface_in, ITEM) % \
+			 sizeof(u32)) != 0,\
+			 "Unaligned write " # ITEM);\
+	BUILD_BUG_ON_MSG((sizeof(VARIABLE) %  sizeof(u32)) != 0,\
+			 "Unaligned write length " # ITEM);\
+	hw_atl2_mif_shared_buf_write(HW,\
+		(offsetof(struct fw_interface_in, ITEM) / sizeof(u32)),\
+		(u32 *)&(VARIABLE), sizeof(VARIABLE) / sizeof(u32));\
+}
+
+#define hw_atl2_shared_buffer_get(HW, ITEM, VARIABLE) \
+{\
+	BUILD_BUG_ON_MSG((offsetof(struct fw_interface_in, ITEM) % \
+			 sizeof(u32)) != 0,\
+			 "Unaligned get " # ITEM);\
+	BUILD_BUG_ON_MSG((sizeof(VARIABLE) %  sizeof(u32)) != 0,\
+			 "Unaligned get length " # ITEM);\
+	hw_atl2_mif_shared_buf_get(HW, \
+		(offsetof(struct fw_interface_in, ITEM) / sizeof(u32)),\
+		(u32 *)&(VARIABLE), \
+		sizeof(VARIABLE) / sizeof(u32));\
+}
+
+/* This should never be used on non atomic fields,
+ * treat any > u32 read as non atomic.
+ */
+#define hw_atl2_shared_buffer_read(HW, ITEM, VARIABLE) \
+{\
+	BUILD_BUG_ON_MSG((offsetof(struct fw_interface_out, ITEM) % \
+			 sizeof(u32)) != 0,\
+			 "Unaligned read " # ITEM);\
+	BUILD_BUG_ON_MSG((sizeof(VARIABLE) %  sizeof(u32)) != 0,\
+			 "Unaligned read length " # ITEM);\
+	BUILD_BUG_ON_MSG(sizeof(VARIABLE) > sizeof(u32),\
+			 "Non atomic read " # ITEM);\
+	hw_atl2_mif_shared_buf_read(HW, \
+		(offsetof(struct fw_interface_out, ITEM) / sizeof(u32)),\
+		(u32 *)&(VARIABLE), sizeof(VARIABLE) / sizeof(u32));\
+}
+
+#define hw_atl2_shared_buffer_read_safe(HW, ITEM, DATA) \
+({\
+	BUILD_BUG_ON_MSG((offsetof(struct fw_interface_out, ITEM) % \
+			 sizeof(u32)) != 0,\
+			 "Unaligned read_safe " # ITEM);\
+	BUILD_BUG_ON_MSG((sizeof(((struct fw_interface_out *)0)->ITEM) % \
+			 sizeof(u32)) != 0,\
+			 "Unaligned read_safe length " # ITEM);\
+	hw_atl2_shared_buffer_read_block((HW), \
+		(offsetof(struct fw_interface_out, ITEM) / sizeof(u32)),\
+		sizeof(((struct fw_interface_out *)0)->ITEM) / sizeof(u32),\
+		(DATA));\
+})
+
+static int hw_atl2_shared_buffer_read_block(struct aq_hw_s *self,
+					    u32 offset, u32 dwords, void *data)
+{
+	struct transaction_counter_s tid1, tid2;
+	int cnt = 0;
+
+	do {
+		do {
+			hw_atl2_shared_buffer_read(self, transaction_id, tid1);
+			cnt++;
+			if (cnt > AQ_A2_FW_READ_TRY_MAX)
+				return -ETIME;
+			if (tid1.transaction_cnt_a != tid1.transaction_cnt_b)
+				mdelay(1);
+		} while (tid1.transaction_cnt_a != tid1.transaction_cnt_b);
+
+		hw_atl2_mif_shared_buf_read(self, offset, (u32 *)data, dwords);
+
+		hw_atl2_shared_buffer_read(self, transaction_id, tid2);
+
+		cnt++;
+		if (cnt > AQ_A2_FW_READ_TRY_MAX)
+			return -ETIME;
+	} while (tid2.transaction_cnt_a != tid2.transaction_cnt_b ||
+		 tid1.transaction_cnt_a != tid2.transaction_cnt_a);
+
+	return 0;
+}
+
+static inline int hw_atl2_shared_buffer_finish_ack(struct aq_hw_s *self)
+{
+	u32 val;
+	int err;
+
+	hw_atl2_mif_host_finished_write_set(self, 1U);
+	err = readx_poll_timeout_atomic(hw_atl2_mif_mcp_finished_read_get,
+					self, val, val == 0U,
+					100, 100000U);
+	WARN(err, "hw_atl2_shared_buffer_finish_ack");
+
+	return err;
+}
+
+static int aq_a2_fw_init(struct aq_hw_s *self)
+{
+	struct link_control_s link_control;
+	u32 mtu;
+	u32 val;
+	int err;
+
+	hw_atl2_shared_buffer_get(self, link_control, link_control);
+	link_control.mode = AQ_HOST_MODE_ACTIVE;
+	hw_atl2_shared_buffer_write(self, link_control, link_control);
+
+	hw_atl2_shared_buffer_get(self, mtu, mtu);
+	mtu = HW_ATL2_MTU_JUMBO;
+	hw_atl2_shared_buffer_write(self, mtu, mtu);
+
+	hw_atl2_mif_host_finished_write_set(self, 1U);
+	err = readx_poll_timeout_atomic(hw_atl2_mif_mcp_finished_read_get,
+					self, val, val == 0U,
+					100, 5000000U);
+	WARN(err, "hw_atl2_shared_buffer_finish_ack");
+
+	return err;
+}
+
+static int aq_a2_fw_deinit(struct aq_hw_s *self)
+{
+	struct link_control_s link_control;
+
+	hw_atl2_shared_buffer_get(self, link_control, link_control);
+	link_control.mode = AQ_HOST_MODE_SHUTDOWN;
+	hw_atl2_shared_buffer_write(self, link_control, link_control);
+
+	return hw_atl2_shared_buffer_finish_ack(self);
+}
+
+static void a2_link_speed_mask2fw(u32 speed,
+				  struct link_options_s *link_options)
+{
+	link_options->rate_10G = !!(speed & AQ_NIC_RATE_10G);
+	link_options->rate_5G = !!(speed & AQ_NIC_RATE_5G);
+	link_options->rate_N5G = link_options->rate_5G;
+	link_options->rate_2P5G = !!(speed & AQ_NIC_RATE_2G5);
+	link_options->rate_N2P5G = link_options->rate_2P5G;
+	link_options->rate_1G = !!(speed & AQ_NIC_RATE_1G);
+	link_options->rate_100M = !!(speed & AQ_NIC_RATE_100M);
+	link_options->rate_10M = !!(speed & AQ_NIC_RATE_10M);
+
+	link_options->rate_1G_hd = !!(speed & AQ_NIC_RATE_1G_HALF);
+	link_options->rate_100M_hd = !!(speed & AQ_NIC_RATE_100M_HALF);
+	link_options->rate_10M_hd = !!(speed & AQ_NIC_RATE_10M_HALF);
+}
+
+static u32 a2_fw_dev_to_eee_mask(struct device_link_caps_s *device_link_caps)
+{
+	u32 rate = 0;
+
+	if (device_link_caps->eee_10G)
+		rate |= AQ_NIC_RATE_EEE_10G;
+	if (device_link_caps->eee_5G)
+		rate |= AQ_NIC_RATE_EEE_5G;
+	if (device_link_caps->eee_2P5G)
+		rate |= AQ_NIC_RATE_EEE_2G5;
+	if (device_link_caps->eee_1G)
+		rate |= AQ_NIC_RATE_EEE_1G;
+	if (device_link_caps->eee_100M)
+		rate |= AQ_NIC_RATE_EEE_100M;
+
+	return rate;
+}
+
+static u32 a2_fw_lkp_to_mask(struct lkp_link_caps_s *lkp_link_caps)
+{
+	u32 rate = 0;
+
+	if (lkp_link_caps->rate_10G)
+		rate |= AQ_NIC_RATE_10G;
+	if (lkp_link_caps->rate_5G)
+		rate |= AQ_NIC_RATE_5G;
+	if (lkp_link_caps->rate_2P5G)
+		rate |= AQ_NIC_RATE_2G5;
+	if (lkp_link_caps->rate_1G)
+		rate |= AQ_NIC_RATE_1G;
+	if (lkp_link_caps->rate_1G_hd)
+		rate |= AQ_NIC_RATE_1G_HALF;
+	if (lkp_link_caps->rate_100M)
+		rate |= AQ_NIC_RATE_100M;
+	if (lkp_link_caps->rate_100M_hd)
+		rate |= AQ_NIC_RATE_100M_HALF;
+	if (lkp_link_caps->rate_10M)
+		rate |= AQ_NIC_RATE_10M;
+	if (lkp_link_caps->rate_10M_hd)
+		rate |= AQ_NIC_RATE_10M_HALF;
+
+	if (lkp_link_caps->eee_10G)
+		rate |= AQ_NIC_RATE_EEE_10G;
+	if (lkp_link_caps->eee_5G)
+		rate |= AQ_NIC_RATE_EEE_5G;
+	if (lkp_link_caps->eee_2P5G)
+		rate |= AQ_NIC_RATE_EEE_2G5;
+	if (lkp_link_caps->eee_1G)
+		rate |= AQ_NIC_RATE_EEE_1G;
+	if (lkp_link_caps->eee_100M)
+		rate |= AQ_NIC_RATE_EEE_100M;
+
+	return rate;
+}
+
+static int aq_a2_fw_set_link_speed(struct aq_hw_s *self, u32 speed)
+{
+	struct link_options_s link_options;
+
+	hw_atl2_shared_buffer_get(self, link_options, link_options);
+	link_options.link_up = 1U;
+	a2_link_speed_mask2fw(speed, &link_options);
+	hw_atl2_shared_buffer_write(self, link_options, link_options);
+
+	return hw_atl2_shared_buffer_finish_ack(self);
+}
+
+static void aq_a2_fw_set_mpi_flow_control(struct aq_hw_s *self,
+					  struct link_options_s *link_options)
+{
+	u32 flow_control = self->aq_nic_cfg->fc.req;
+
+	link_options->pause_rx = !!(flow_control & AQ_NIC_FC_RX);
+	link_options->pause_tx = !!(flow_control & AQ_NIC_FC_TX);
+}
+
+static void aq_a2_fw_upd_eee_rate_bits(struct aq_hw_s *self,
+				       struct link_options_s *link_options,
+				       u32 eee_speeds)
+{
+	link_options->eee_10G =  !!(eee_speeds & AQ_NIC_RATE_EEE_10G);
+	link_options->eee_5G = !!(eee_speeds & AQ_NIC_RATE_EEE_5G);
+	link_options->eee_2P5G = !!(eee_speeds & AQ_NIC_RATE_EEE_2G5);
+	link_options->eee_1G = !!(eee_speeds & AQ_NIC_RATE_EEE_1G);
+	link_options->eee_100M = !!(eee_speeds & AQ_NIC_RATE_EEE_100M);
+}
+
+static int aq_a2_fw_set_state(struct aq_hw_s *self,
+			      enum hal_atl_utils_fw_state_e state)
+{
+	struct link_options_s link_options;
+
+	hw_atl2_shared_buffer_get(self, link_options, link_options);
+
+	switch (state) {
+	case MPI_INIT:
+		link_options.link_up = 1U;
+		aq_a2_fw_upd_eee_rate_bits(self, &link_options,
+					   self->aq_nic_cfg->eee_speeds);
+		aq_a2_fw_set_mpi_flow_control(self, &link_options);
+		break;
+	case MPI_DEINIT:
+		link_options.link_up = 0U;
+		break;
+	case MPI_RESET:
+	case MPI_POWER:
+		/* No actions */
+		break;
+	}
+
+	hw_atl2_shared_buffer_write(self, link_options, link_options);
+
+	return hw_atl2_shared_buffer_finish_ack(self);
+}
+
+static int aq_a2_fw_update_link_status(struct aq_hw_s *self)
+{
+	struct lkp_link_caps_s lkp_link_caps;
+	struct link_status_s link_status;
+
+	hw_atl2_shared_buffer_read(self, link_status, link_status);
+
+	switch (link_status.link_rate) {
+	case AQ_A2_FW_LINK_RATE_10G:
+		self->aq_link_status.mbps = 10000;
+		break;
+	case AQ_A2_FW_LINK_RATE_5G:
+		self->aq_link_status.mbps = 5000;
+		break;
+	case AQ_A2_FW_LINK_RATE_2G5:
+		self->aq_link_status.mbps = 2500;
+		break;
+	case AQ_A2_FW_LINK_RATE_1G:
+		self->aq_link_status.mbps = 1000;
+		break;
+	case AQ_A2_FW_LINK_RATE_100M:
+		self->aq_link_status.mbps = 100;
+		break;
+	case AQ_A2_FW_LINK_RATE_10M:
+		self->aq_link_status.mbps = 10;
+		break;
+	default:
+		self->aq_link_status.mbps = 0;
+	}
+	self->aq_link_status.full_duplex = link_status.duplex;
+
+	hw_atl2_shared_buffer_read(self, lkp_link_caps, lkp_link_caps);
+
+	self->aq_link_status.lp_link_speed_msk =
+				 a2_fw_lkp_to_mask(&lkp_link_caps);
+	self->aq_link_status.lp_flow_control =
+				((lkp_link_caps.pause_rx) ? AQ_NIC_FC_RX : 0) |
+				((lkp_link_caps.pause_tx) ? AQ_NIC_FC_TX : 0);
+
+	return 0;
+}
+
+static int aq_a2_fw_get_mac_permanent(struct aq_hw_s *self, u8 *mac)
+{
+	struct mac_address_aligned_s mac_address;
+
+	hw_atl2_shared_buffer_get(self, mac_address, mac_address);
+	ether_addr_copy(mac, (u8 *)mac_address.aligned.mac_address);
+
+	return 0;
+}
+
+static void aq_a2_fill_a0_stats(struct aq_hw_s *self,
+				struct statistics_s *stats)
+{
+	struct hw_atl2_priv *priv = self->priv;
+	struct aq_stats_s *cs = &self->curr_stats;
+	struct aq_stats_s curr_stats = *cs;
+	bool corrupted_stats = false;
+
+#define AQ_SDELTA(_N, _F)  \
+do { \
+	if (!corrupted_stats && \
+	    ((s64)(stats->a0.msm._F - priv->last_stats.a0.msm._F)) >= 0) \
+		curr_stats._N += stats->a0.msm._F - priv->last_stats.a0.msm._F;\
+	else \
+		corrupted_stats = true; \
+} while (0)
+
+	if (self->aq_link_status.mbps) {
+		AQ_SDELTA(uprc, rx_unicast_frames);
+		AQ_SDELTA(mprc, rx_multicast_frames);
+		AQ_SDELTA(bprc, rx_broadcast_frames);
+		AQ_SDELTA(erpr, rx_error_frames);
+
+		AQ_SDELTA(uptc, tx_unicast_frames);
+		AQ_SDELTA(mptc, tx_multicast_frames);
+		AQ_SDELTA(bptc, tx_broadcast_frames);
+		AQ_SDELTA(erpt, tx_errors);
+
+		AQ_SDELTA(ubrc, rx_unicast_octets);
+		AQ_SDELTA(ubtc, tx_unicast_octets);
+		AQ_SDELTA(mbrc, rx_multicast_octets);
+		AQ_SDELTA(mbtc, tx_multicast_octets);
+		AQ_SDELTA(bbrc, rx_broadcast_octets);
+		AQ_SDELTA(bbtc, tx_broadcast_octets);
+
+		if (!corrupted_stats)
+			*cs = curr_stats;
+	}
+#undef AQ_SDELTA
+
+}
+
+static void aq_a2_fill_b0_stats(struct aq_hw_s *self,
+				struct statistics_s *stats)
+{
+	struct hw_atl2_priv *priv = self->priv;
+	struct aq_stats_s *cs = &self->curr_stats;
+	struct aq_stats_s curr_stats = *cs;
+	bool corrupted_stats = false;
+
+#define AQ_SDELTA(_N, _F)  \
+do { \
+	if (!corrupted_stats && \
+	    ((s64)(stats->b0._F - priv->last_stats.b0._F)) >= 0) \
+		curr_stats._N += stats->b0._F - priv->last_stats.b0._F; \
+	else \
+		corrupted_stats = true; \
+} while (0)
+
+	if (self->aq_link_status.mbps) {
+		AQ_SDELTA(uprc, rx_unicast_frames);
+		AQ_SDELTA(mprc, rx_multicast_frames);
+		AQ_SDELTA(bprc, rx_broadcast_frames);
+		AQ_SDELTA(erpr, rx_errors);
+		AQ_SDELTA(brc, rx_good_octets);
+
+		AQ_SDELTA(uptc, tx_unicast_frames);
+		AQ_SDELTA(mptc, tx_multicast_frames);
+		AQ_SDELTA(bptc, tx_broadcast_frames);
+		AQ_SDELTA(erpt, tx_errors);
+		AQ_SDELTA(btc, tx_good_octets);
+
+		if (!corrupted_stats)
+			*cs = curr_stats;
+	}
+#undef AQ_SDELTA
+}
+
+static int aq_a2_fw_update_stats(struct aq_hw_s *self)
+{
+	struct aq_stats_s *cs = &self->curr_stats;
+	struct hw_atl2_priv *priv = self->priv;
+	struct statistics_s stats;
+	struct version_s version;
+	int err;
+
+	err = hw_atl2_shared_buffer_read_safe(self, version, &version);
+	if (err)
+		return err;
+
+	err = hw_atl2_shared_buffer_read_safe(self, stats, &stats);
+	if (err)
+		return err;
+
+	if (version.drv_iface_ver == AQ_A2_FW_INTERFACE_A0)
+		aq_a2_fill_a0_stats(self, &stats);
+	else
+		aq_a2_fill_b0_stats(self, &stats);
+
+	cs->dma_pkt_rc = hw_atl_stats_rx_dma_good_pkt_counter_get(self);
+	cs->dma_pkt_tc = hw_atl_stats_tx_dma_good_pkt_counter_get(self);
+	cs->dma_oct_rc = hw_atl_stats_rx_dma_good_octet_counter_get(self);
+	cs->dma_oct_tc = hw_atl_stats_tx_dma_good_octet_counter_get(self);
+	cs->dpc = hw_atl_rpb_rx_dma_drop_pkt_cnt_get(self);
+
+	memcpy(&priv->last_stats, &stats, sizeof(stats));
+
+	return 0;
+}
+
+static int aq_a2_fw_get_phy_temp(struct aq_hw_s *self, int *temp)
+{
+	struct phy_health_monitor_s phy_health_monitor;
+
+	hw_atl2_shared_buffer_read_safe(self, phy_health_monitor,
+					&phy_health_monitor);
+
+	*temp = (int8_t)phy_health_monitor.phy_temperature * 1000;
+	return 0;
+}
+
+static int aq_a2_fw_get_mac_temp(struct aq_hw_s *self, int *temp)
+{
+	/* There's only one temperature sensor on A2, use it for
+	 * both MAC and PHY.
+	 */
+	return aq_a2_fw_get_phy_temp(self, temp);
+}
+
+static int aq_a2_fw_set_wol_params(struct aq_hw_s *self, const u8 *mac, u32 wol)
+{
+	struct mac_address_aligned_s mac_address;
+	struct link_control_s link_control;
+	struct wake_on_lan_s wake_on_lan;
+
+	memcpy(mac_address.aligned.mac_address, mac, ETH_ALEN);
+	hw_atl2_shared_buffer_write(self, mac_address, mac_address);
+
+	memset(&wake_on_lan, 0, sizeof(wake_on_lan));
+
+	if (wol & WAKE_MAGIC)
+		wake_on_lan.wake_on_magic_packet = 1U;
+
+	if (wol & (WAKE_PHY | AQ_FW_WAKE_ON_LINK_RTPM))
+		wake_on_lan.wake_on_link_up = 1U;
+
+	hw_atl2_shared_buffer_write(self, sleep_proxy, wake_on_lan);
+
+	hw_atl2_shared_buffer_get(self, link_control, link_control);
+	link_control.mode = AQ_HOST_MODE_SLEEP_PROXY;
+	hw_atl2_shared_buffer_write(self, link_control, link_control);
+
+	return hw_atl2_shared_buffer_finish_ack(self);
+}
+
+static int aq_a2_fw_set_power(struct aq_hw_s *self, unsigned int power_state,
+			      const u8 *mac)
+{
+	u32 wol = self->aq_nic_cfg->wol;
+	int err = 0;
+
+	if (wol)
+		err = aq_a2_fw_set_wol_params(self, mac, wol);
+
+	return err;
+}
+
+static int aq_a2_fw_set_eee_rate(struct aq_hw_s *self, u32 speed)
+{
+	struct link_options_s link_options;
+
+	hw_atl2_shared_buffer_get(self, link_options, link_options);
+
+	aq_a2_fw_upd_eee_rate_bits(self, &link_options, speed);
+
+	hw_atl2_shared_buffer_write(self, link_options, link_options);
+
+	return hw_atl2_shared_buffer_finish_ack(self);
+}
+
+static int aq_a2_fw_get_eee_rate(struct aq_hw_s *self, u32 *rate,
+				 u32 *supported_rates)
+{
+	struct device_link_caps_s device_link_caps;
+	struct lkp_link_caps_s lkp_link_caps;
+
+	hw_atl2_shared_buffer_read(self, device_link_caps, device_link_caps);
+	hw_atl2_shared_buffer_read(self, lkp_link_caps, lkp_link_caps);
+
+	*supported_rates = a2_fw_dev_to_eee_mask(&device_link_caps);
+	*rate = a2_fw_lkp_to_mask(&lkp_link_caps);
+
+	return 0;
+}
+
+static int aq_a2_fw_renegotiate(struct aq_hw_s *self)
+{
+	struct link_options_s link_options;
+	int err;
+
+	hw_atl2_shared_buffer_get(self, link_options, link_options);
+	link_options.link_renegotiate = 1U;
+	hw_atl2_shared_buffer_write(self, link_options, link_options);
+
+	err = hw_atl2_shared_buffer_finish_ack(self);
+
+	/* We should put renegotiate status back to zero
+	 * after command completes
+	 */
+	link_options.link_renegotiate = 0U;
+	hw_atl2_shared_buffer_write(self, link_options, link_options);
+
+	return err;
+}
+
+static int aq_a2_fw_set_flow_control(struct aq_hw_s *self)
+{
+	struct link_options_s link_options;
+
+	hw_atl2_shared_buffer_get(self, link_options, link_options);
+
+	aq_a2_fw_set_mpi_flow_control(self, &link_options);
+
+	hw_atl2_shared_buffer_write(self, link_options, link_options);
+
+	return hw_atl2_shared_buffer_finish_ack(self);
+}
+
+static u32 aq_a2_fw_get_flow_control(struct aq_hw_s *self, u32 *fcmode)
+{
+	struct link_status_s link_status;
+
+	hw_atl2_shared_buffer_read(self, link_status, link_status);
+
+	*fcmode = ((link_status.pause_rx) ? AQ_NIC_FC_RX : 0) |
+		  ((link_status.pause_tx) ? AQ_NIC_FC_TX : 0);
+	return 0;
+}
+
+static int aq_a2_fw_set_phyloopback(struct aq_hw_s *self, u32 mode, bool enable)
+{
+	struct link_options_s link_options;
+
+	hw_atl2_shared_buffer_get(self, link_options, link_options);
+
+	switch (mode) {
+	case AQ_HW_LOOPBACK_PHYINT_SYS:
+		link_options.internal_loopback = enable;
+		break;
+	case AQ_HW_LOOPBACK_PHYEXT_SYS:
+		link_options.external_loopback = enable;
+		break;
+	default:
+		return -EINVAL;
+	}
+
+	hw_atl2_shared_buffer_write(self, link_options, link_options);
+
+	return hw_atl2_shared_buffer_finish_ack(self);
+}
+
+u32 hw_atl2_utils_get_fw_version(struct aq_hw_s *self)
+{
+	struct version_s version;
+
+	hw_atl2_shared_buffer_read_safe(self, version, &version);
+
+	/* A2 FW version is stored in reverse order */
+	return version.bundle.major << 24 |
+	       version.bundle.minor << 16 |
+	       version.bundle.build;
+}
+
+int hw_atl2_utils_get_action_resolve_table_caps(struct aq_hw_s *self,
+						u8 *base_index, u8 *count)
+{
+	struct filter_caps_s filter_caps;
+	int err;
+
+	err = hw_atl2_shared_buffer_read_safe(self, filter_caps, &filter_caps);
+	if (err)
+		return err;
+
+	*base_index = filter_caps.rslv_tbl_base_index;
+	*count = filter_caps.rslv_tbl_count;
+	return 0;
+}
+
+static int aq_a2_fw_set_downshift(struct aq_hw_s *self, u32 counter)
+{
+	struct link_options_s link_options;
+
+	hw_atl2_shared_buffer_get(self, link_options, link_options);
+	link_options.downshift = !!counter;
+	link_options.downshift_retry = counter;
+	hw_atl2_shared_buffer_write(self, link_options, link_options);
+
+	return hw_atl2_shared_buffer_finish_ack(self);
+}
+
+const struct aq_fw_ops aq_a2_fw_ops = {
+	.init               = aq_a2_fw_init,
+	.deinit             = aq_a2_fw_deinit,
+	.reset              = NULL,
+	.renegotiate        = aq_a2_fw_renegotiate,
+	.get_mac_permanent  = aq_a2_fw_get_mac_permanent,
+	.set_link_speed     = aq_a2_fw_set_link_speed,
+	.set_state          = aq_a2_fw_set_state,
+	.update_link_status = aq_a2_fw_update_link_status,
+	.update_stats       = aq_a2_fw_update_stats,
+	.set_power          = aq_a2_fw_set_power,
+	.get_mac_temp       = aq_a2_fw_get_mac_temp,
+	.get_phy_temp       = aq_a2_fw_get_phy_temp,
+	.set_eee_rate       = aq_a2_fw_set_eee_rate,
+	.get_eee_rate       = aq_a2_fw_get_eee_rate,
+	.set_flow_control   = aq_a2_fw_set_flow_control,
+	.get_flow_control   = aq_a2_fw_get_flow_control,
+	.set_phyloopback    = aq_a2_fw_set_phyloopback,
+	.set_downshift      = aq_a2_fw_set_downshift,
+};
diff --git a/drivers/net/ethernet/aquantia/atlantic/macsec/MSS_Egress_registers.h b/drivers/net/ethernet/aquantia/atlantic/macsec/MSS_Egress_registers.h
new file mode 100644
index 000000000000..71d08ea80b54
--- /dev/null
+++ b/drivers/net/ethernet/aquantia/atlantic/macsec/MSS_Egress_registers.h
@@ -0,0 +1,73 @@
+/* SPDX-License-Identifier: GPL-2.0-only */
+/* Atlantic Network Driver
+ * Copyright (C) 2020 Marvell International Ltd.
+ */
+
+#ifndef MSS_EGRESS_REGS_HEADER
+#define MSS_EGRESS_REGS_HEADER
+
+#define MSS_EGRESS_CTL_REGISTER_ADDR 0x00005002
+#define MSS_EGRESS_SA_EXPIRED_STATUS_REGISTER_ADDR 0x00005060
+#define MSS_EGRESS_SA_THRESHOLD_EXPIRED_STATUS_REGISTER_ADDR 0x00005062
+#define MSS_EGRESS_LUT_ADDR_CTL_REGISTER_ADDR 0x00005080
+#define MSS_EGRESS_LUT_CTL_REGISTER_ADDR 0x00005081
+#define MSS_EGRESS_LUT_DATA_CTL_REGISTER_ADDR 0x000050A0
+
+struct mss_egress_ctl_register {
+	union {
+		struct {
+			unsigned int soft_reset : 1;
+			unsigned int drop_kay_packet : 1;
+			unsigned int drop_egprc_lut_miss : 1;
+			unsigned int gcm_start : 1;
+			unsigned int gcm_test_mode : 1;
+			unsigned int unmatched_use_sc_0 : 1;
+			unsigned int drop_invalid_sa_sc_packets : 1;
+			unsigned int reserved0 : 1;
+			/* Should always be set to 0. */
+			unsigned int external_classification_enable : 1;
+			unsigned int icv_lsb_8bytes_enable : 1;
+			unsigned int high_prio : 1;
+			unsigned int clear_counter : 1;
+			unsigned int clear_global_time : 1;
+			unsigned int ethertype_explicit_sectag_lsb : 3;
+		} bits_0;
+		unsigned short word_0;
+	};
+	union {
+		struct {
+			unsigned int ethertype_explicit_sectag_msb : 13;
+			unsigned int reserved0 : 3;
+		} bits_1;
+		unsigned short word_1;
+	};
+};
+
+struct mss_egress_lut_addr_ctl_register {
+	union {
+		struct {
+			unsigned int lut_addr : 9;
+			unsigned int reserved0 : 3;
+			/* 0x0 : Egress MAC Control FIlter (CTLF) LUT
+			 * 0x1 : Egress Classification LUT
+			 * 0x2 : Egress SC/SA LUT
+			 * 0x3 : Egress SMIB
+			 */
+			unsigned int lut_select : 4;
+		} bits_0;
+		unsigned short word_0;
+	};
+};
+
+struct mss_egress_lut_ctl_register {
+	union {
+		struct {
+			unsigned int reserved0 : 14;
+			unsigned int lut_read : 1;
+			unsigned int lut_write : 1;
+		} bits_0;
+		unsigned short word_0;
+	};
+};
+
+#endif /* MSS_EGRESS_REGS_HEADER */
diff --git a/drivers/net/ethernet/aquantia/atlantic/macsec/MSS_Ingress_registers.h b/drivers/net/ethernet/aquantia/atlantic/macsec/MSS_Ingress_registers.h
new file mode 100644
index 000000000000..d4c00d9a0fc6
--- /dev/null
+++ b/drivers/net/ethernet/aquantia/atlantic/macsec/MSS_Ingress_registers.h
@@ -0,0 +1,77 @@
+/* SPDX-License-Identifier: GPL-2.0-only */
+/* Atlantic Network Driver
+ * Copyright (C) 2020 Marvell International Ltd.
+ */
+
+#ifndef MSS_INGRESS_REGS_HEADER
+#define MSS_INGRESS_REGS_HEADER
+
+#define MSS_INGRESS_CTL_REGISTER_ADDR 0x0000800E
+#define MSS_INGRESS_LUT_ADDR_CTL_REGISTER_ADDR 0x00008080
+#define MSS_INGRESS_LUT_CTL_REGISTER_ADDR 0x00008081
+#define MSS_INGRESS_LUT_DATA_CTL_REGISTER_ADDR 0x000080A0
+
+struct mss_ingress_ctl_register {
+	union {
+		struct {
+			unsigned int soft_reset : 1;
+			unsigned int operation_point_to_point : 1;
+			unsigned int create_sci : 1;
+			/* Unused  */
+			unsigned int mask_short_length_error : 1;
+			unsigned int drop_kay_packet : 1;
+			unsigned int drop_igprc_miss : 1;
+			/* Unused  */
+			unsigned int check_icv : 1;
+			unsigned int clear_global_time : 1;
+			unsigned int clear_count : 1;
+			unsigned int high_prio : 1;
+			unsigned int remove_sectag : 1;
+			unsigned int global_validate_frames : 2;
+			unsigned int icv_lsb_8bytes_enabled : 1;
+			unsigned int reserved0 : 2;
+		} bits_0;
+		unsigned short word_0;
+	};
+	union {
+		struct {
+			unsigned int reserved0 : 16;
+		} bits_1;
+		unsigned short word_1;
+	};
+};
+
+struct mss_ingress_lut_addr_ctl_register {
+	union {
+		struct {
+			unsigned int lut_addr : 9;
+			unsigned int reserved0 : 3;
+			/* 0x0 : Ingress Pre-Security MAC Control FIlter
+			 *       (IGPRCTLF) LUT
+			 * 0x1 : Ingress Pre-Security Classification LUT (IGPRC)
+			 * 0x2 : Ingress Packet Format (IGPFMT) SAKey LUT
+			 * 0x3 : Ingress Packet Format (IGPFMT) SC/SA LUT
+			 * 0x4 : Ingress Post-Security Classification LUT
+			 *       (IGPOC)
+			 * 0x5 : Ingress Post-Security MAC Control Filter
+			 *       (IGPOCTLF) LUT
+			 * 0x6 : Ingress MIB (IGMIB)
+			 */
+			unsigned int lut_select : 4;
+		} bits_0;
+		unsigned short word_0;
+	};
+};
+
+struct mss_ingress_lut_ctl_register {
+	union {
+		struct {
+			unsigned int reserved0 : 14;
+			unsigned int lut_read : 1;
+			unsigned int lut_write : 1;
+		} bits_0;
+		unsigned short word_0;
+	};
+};
+
+#endif /* MSS_INGRESS_REGS_HEADER */
diff --git a/drivers/net/ethernet/aquantia/atlantic/macsec/macsec_api.c b/drivers/net/ethernet/aquantia/atlantic/macsec/macsec_api.c
new file mode 100644
index 000000000000..431924959520
--- /dev/null
+++ b/drivers/net/ethernet/aquantia/atlantic/macsec/macsec_api.c
@@ -0,0 +1,2477 @@
+// SPDX-License-Identifier: GPL-2.0-only
+/* Atlantic Network Driver
+ * Copyright (C) 2020 Marvell International Ltd.
+ */
+
+#include "macsec_api.h"
+#include <linux/mdio.h>
+#include "MSS_Ingress_registers.h"
+#include "MSS_Egress_registers.h"
+#include "aq_phy.h"
+
+#define AQ_API_CALL_SAFE(func, ...)                                            \
+({                                                                             \
+	int ret;                                                               \
+	do {                                                                   \
+		ret = aq_mss_mdio_sem_get(hw);                                 \
+		if (unlikely(ret))                                             \
+			break;                                                 \
+									       \
+		ret = func(__VA_ARGS__);                                       \
+									       \
+		aq_mss_mdio_sem_put(hw);                                       \
+	} while (0);                                                           \
+	ret;                                                                   \
+})
+
+/*******************************************************************************
+ *                               MDIO wrappers
+ ******************************************************************************/
+static int aq_mss_mdio_sem_get(struct aq_hw_s *hw)
+{
+	u32 val;
+
+	return readx_poll_timeout_atomic(hw_atl_sem_mdio_get, hw, val,
+					 val == 1U, 10U, 100000U);
+}
+
+static void aq_mss_mdio_sem_put(struct aq_hw_s *hw)
+{
+	hw_atl_reg_glb_cpu_sem_set(hw, 1U, HW_ATL_FW_SM_MDIO);
+}
+
+static int aq_mss_mdio_read(struct aq_hw_s *hw, u16 mmd, u16 addr, u16 *data)
+{
+	*data = aq_mdio_read_word(hw, mmd, addr);
+	return (*data != 0xffff) ? 0 : -ETIME;
+}
+
+static int aq_mss_mdio_write(struct aq_hw_s *hw, u16 mmd, u16 addr, u16 data)
+{
+	aq_mdio_write_word(hw, mmd, addr, data);
+	return 0;
+}
+
+/*******************************************************************************
+ *                          MACSEC config and status
+ ******************************************************************************/
+
+static int set_raw_ingress_record(struct aq_hw_s *hw, u16 *packed_record,
+				  u8 num_words, u8 table_id,
+				  u16 table_index)
+{
+	struct mss_ingress_lut_addr_ctl_register lut_sel_reg;
+	struct mss_ingress_lut_ctl_register lut_op_reg;
+
+	unsigned int i;
+
+	/* NOTE: MSS registers must always be read/written as adjacent pairs.
+	 * For instance, to write either or both 1E.80A0 and 80A1, we have to:
+	 * 1. Write 1E.80A0 first
+	 * 2. Then write 1E.80A1
+	 *
+	 * For HHD devices: These writes need to be performed consecutively, and
+	 * to ensure this we use the PIF mailbox to delegate the reads/writes to
+	 * the FW.
+	 *
+	 * For EUR devices: Not need to use the PIF mailbox; it is safe to
+	 * write to the registers directly.
+	 */
+
+	/* Write the packed record words to the data buffer registers. */
+	for (i = 0; i < num_words; i += 2) {
+		aq_mss_mdio_write(hw, MDIO_MMD_VEND1,
+				  MSS_INGRESS_LUT_DATA_CTL_REGISTER_ADDR + i,
+				  packed_record[i]);
+		aq_mss_mdio_write(hw, MDIO_MMD_VEND1,
+				  MSS_INGRESS_LUT_DATA_CTL_REGISTER_ADDR + i +
+					  1,
+				  packed_record[i + 1]);
+	}
+
+	/* Clear out the unused data buffer registers. */
+	for (i = num_words; i < 24; i += 2) {
+		aq_mss_mdio_write(hw, MDIO_MMD_VEND1,
+				  MSS_INGRESS_LUT_DATA_CTL_REGISTER_ADDR + i,
+				  0);
+		aq_mss_mdio_write(hw, MDIO_MMD_VEND1,
+			MSS_INGRESS_LUT_DATA_CTL_REGISTER_ADDR + i + 1, 0);
+	}
+
+	/* Select the table and row index to write to */
+	lut_sel_reg.bits_0.lut_select = table_id;
+	lut_sel_reg.bits_0.lut_addr = table_index;
+
+	lut_op_reg.bits_0.lut_read = 0;
+	lut_op_reg.bits_0.lut_write = 1;
+
+	aq_mss_mdio_write(hw, MDIO_MMD_VEND1,
+			  MSS_INGRESS_LUT_ADDR_CTL_REGISTER_ADDR,
+			  lut_sel_reg.word_0);
+	aq_mss_mdio_write(hw, MDIO_MMD_VEND1, MSS_INGRESS_LUT_CTL_REGISTER_ADDR,
+			  lut_op_reg.word_0);
+
+	return 0;
+}
+
+/*! Read the specified Ingress LUT table row.
+ *  packed_record - [OUT] The table row data (raw).
+ */
+static int get_raw_ingress_record(struct aq_hw_s *hw, u16 *packed_record,
+				  u8 num_words, u8 table_id,
+				  u16 table_index)
+{
+	struct mss_ingress_lut_addr_ctl_register lut_sel_reg;
+	struct mss_ingress_lut_ctl_register lut_op_reg;
+	int ret;
+
+	unsigned int i;
+
+	/* Select the table and row index to read */
+	lut_sel_reg.bits_0.lut_select = table_id;
+	lut_sel_reg.bits_0.lut_addr = table_index;
+
+	lut_op_reg.bits_0.lut_read = 1;
+	lut_op_reg.bits_0.lut_write = 0;
+
+	ret = aq_mss_mdio_write(hw, MDIO_MMD_VEND1,
+				MSS_INGRESS_LUT_ADDR_CTL_REGISTER_ADDR,
+				lut_sel_reg.word_0);
+	if (unlikely(ret))
+		return ret;
+	ret = aq_mss_mdio_write(hw, MDIO_MMD_VEND1,
+				MSS_INGRESS_LUT_CTL_REGISTER_ADDR,
+				lut_op_reg.word_0);
+	if (unlikely(ret))
+		return ret;
+
+	memset(packed_record, 0, sizeof(u16) * num_words);
+
+	for (i = 0; i < num_words; i += 2) {
+		ret = aq_mss_mdio_read(hw, MDIO_MMD_VEND1,
+				       MSS_INGRESS_LUT_DATA_CTL_REGISTER_ADDR +
+					       i,
+				       &packed_record[i]);
+		if (unlikely(ret))
+			return ret;
+		ret = aq_mss_mdio_read(hw, MDIO_MMD_VEND1,
+				       MSS_INGRESS_LUT_DATA_CTL_REGISTER_ADDR +
+					       i + 1,
+				       &packed_record[i + 1]);
+		if (unlikely(ret))
+			return ret;
+	}
+
+	return 0;
+}
+
+/*! Write packed_record to the specified Egress LUT table row. */
+static int set_raw_egress_record(struct aq_hw_s *hw, u16 *packed_record,
+				 u8 num_words, u8 table_id,
+				 u16 table_index)
+{
+	struct mss_egress_lut_addr_ctl_register lut_sel_reg;
+	struct mss_egress_lut_ctl_register lut_op_reg;
+
+	unsigned int i;
+
+	/* Write the packed record words to the data buffer registers. */
+	for (i = 0; i < num_words; i += 2) {
+		aq_mss_mdio_write(hw, MDIO_MMD_VEND1,
+				  MSS_EGRESS_LUT_DATA_CTL_REGISTER_ADDR + i,
+				  packed_record[i]);
+		aq_mss_mdio_write(hw, MDIO_MMD_VEND1,
+				  MSS_EGRESS_LUT_DATA_CTL_REGISTER_ADDR + i + 1,
+				  packed_record[i + 1]);
+	}
+
+	/* Clear out the unused data buffer registers. */
+	for (i = num_words; i < 28; i += 2) {
+		aq_mss_mdio_write(hw, MDIO_MMD_VEND1,
+				  MSS_EGRESS_LUT_DATA_CTL_REGISTER_ADDR + i, 0);
+		aq_mss_mdio_write(hw, MDIO_MMD_VEND1,
+				  MSS_EGRESS_LUT_DATA_CTL_REGISTER_ADDR + i + 1,
+				  0);
+	}
+
+	/* Select the table and row index to write to */
+	lut_sel_reg.bits_0.lut_select = table_id;
+	lut_sel_reg.bits_0.lut_addr = table_index;
+
+	lut_op_reg.bits_0.lut_read = 0;
+	lut_op_reg.bits_0.lut_write = 1;
+
+	aq_mss_mdio_write(hw, MDIO_MMD_VEND1,
+			  MSS_EGRESS_LUT_ADDR_CTL_REGISTER_ADDR,
+			  lut_sel_reg.word_0);
+	aq_mss_mdio_write(hw, MDIO_MMD_VEND1, MSS_EGRESS_LUT_CTL_REGISTER_ADDR,
+			  lut_op_reg.word_0);
+
+	return 0;
+}
+
+static int get_raw_egress_record(struct aq_hw_s *hw, u16 *packed_record,
+				 u8 num_words, u8 table_id,
+				 u16 table_index)
+{
+	struct mss_egress_lut_addr_ctl_register lut_sel_reg;
+	struct mss_egress_lut_ctl_register lut_op_reg;
+	int ret;
+
+	unsigned int i;
+
+	/* Select the table and row index to read */
+	lut_sel_reg.bits_0.lut_select = table_id;
+	lut_sel_reg.bits_0.lut_addr = table_index;
+
+	lut_op_reg.bits_0.lut_read = 1;
+	lut_op_reg.bits_0.lut_write = 0;
+
+	ret = aq_mss_mdio_write(hw, MDIO_MMD_VEND1,
+				MSS_EGRESS_LUT_ADDR_CTL_REGISTER_ADDR,
+				lut_sel_reg.word_0);
+	if (unlikely(ret))
+		return ret;
+	ret = aq_mss_mdio_write(hw, MDIO_MMD_VEND1,
+				MSS_EGRESS_LUT_CTL_REGISTER_ADDR,
+				lut_op_reg.word_0);
+	if (unlikely(ret))
+		return ret;
+
+	memset(packed_record, 0, sizeof(u16) * num_words);
+
+	for (i = 0; i < num_words; i += 2) {
+		ret = aq_mss_mdio_read(hw, MDIO_MMD_VEND1,
+				       MSS_EGRESS_LUT_DATA_CTL_REGISTER_ADDR +
+					       i,
+				       &packed_record[i]);
+		if (unlikely(ret))
+			return ret;
+		ret = aq_mss_mdio_read(hw, MDIO_MMD_VEND1,
+				       MSS_EGRESS_LUT_DATA_CTL_REGISTER_ADDR +
+					       i + 1,
+				       &packed_record[i + 1]);
+		if (unlikely(ret))
+			return ret;
+	}
+
+	return 0;
+}
+
+static int
+set_ingress_prectlf_record(struct aq_hw_s *hw,
+			   const struct aq_mss_ingress_prectlf_record *rec,
+			   u16 table_index)
+{
+	u16 packed_record[6];
+
+	if (table_index >= NUMROWS_INGRESSPRECTLFRECORD)
+		return -EINVAL;
+
+	memset(packed_record, 0, sizeof(u16) * 6);
+
+	packed_record[0] = rec->sa_da[0] & 0xFFFF;
+	packed_record[1] = (rec->sa_da[0] >> 16) & 0xFFFF;
+	packed_record[2] = rec->sa_da[1] & 0xFFFF;
+	packed_record[3] = rec->eth_type & 0xFFFF;
+	packed_record[4] = rec->match_mask & 0xFFFF;
+	packed_record[5] = rec->match_type & 0xF;
+	packed_record[5] |= (rec->action & 0x1) << 4;
+
+	return set_raw_ingress_record(hw, packed_record, 6, 0,
+				      ROWOFFSET_INGRESSPRECTLFRECORD +
+					      table_index);
+}
+
+int aq_mss_set_ingress_prectlf_record(struct aq_hw_s *hw,
+	const struct aq_mss_ingress_prectlf_record *rec,
+	u16 table_index)
+{
+	return AQ_API_CALL_SAFE(set_ingress_prectlf_record, hw, rec,
+				table_index);
+}
+
+static int get_ingress_prectlf_record(struct aq_hw_s *hw,
+				      struct aq_mss_ingress_prectlf_record *rec,
+				      u16 table_index)
+{
+	u16 packed_record[6];
+	int ret;
+
+	if (table_index >= NUMROWS_INGRESSPRECTLFRECORD)
+		return -EINVAL;
+
+	/* If the row that we want to read is odd, first read the previous even
+	 * row, throw that value away, and finally read the desired row.
+	 * This is a workaround for EUR devices that allows us to read
+	 * odd-numbered rows.  For HHD devices: this workaround will not work,
+	 * so don't bother; odd-numbered rows are not readable.
+	 */
+	if ((table_index % 2) > 0) {
+		ret = get_raw_ingress_record(hw, packed_record, 6, 0,
+					     ROWOFFSET_INGRESSPRECTLFRECORD +
+						     table_index - 1);
+		if (unlikely(ret))
+			return ret;
+	}
+
+	ret = get_raw_ingress_record(hw, packed_record, 6, 0,
+				     ROWOFFSET_INGRESSPRECTLFRECORD +
+					     table_index);
+	if (unlikely(ret))
+		return ret;
+
+	rec->sa_da[0] = packed_record[0];
+	rec->sa_da[0] |= packed_record[1] << 16;
+
+	rec->sa_da[1] = packed_record[2];
+
+	rec->eth_type = packed_record[3];
+
+	rec->match_mask = packed_record[4];
+
+	rec->match_type = packed_record[5] & 0xF;
+
+	rec->action = (packed_record[5] >> 4) & 0x1;
+
+	return 0;
+}
+
+int aq_mss_get_ingress_prectlf_record(struct aq_hw_s *hw,
+				      struct aq_mss_ingress_prectlf_record *rec,
+				      u16 table_index)
+{
+	memset(rec, 0, sizeof(*rec));
+
+	return AQ_API_CALL_SAFE(get_ingress_prectlf_record, hw, rec,
+				table_index);
+}
+
+static int
+set_ingress_preclass_record(struct aq_hw_s *hw,
+			    const struct aq_mss_ingress_preclass_record *rec,
+			    u16 table_index)
+{
+	u16 packed_record[20];
+
+	if (table_index >= NUMROWS_INGRESSPRECLASSRECORD)
+		return -EINVAL;
+
+	memset(packed_record, 0, sizeof(u16) * 20);
+
+	packed_record[0] = rec->sci[0] & 0xFFFF;
+	packed_record[1] = (rec->sci[0] >> 16) & 0xFFFF;
+
+	packed_record[2] = rec->sci[1] & 0xFFFF;
+	packed_record[3] = (rec->sci[1] >> 16) & 0xFFFF;
+
+	packed_record[4] = rec->tci & 0xFF;
+
+	packed_record[4] |= (rec->encr_offset & 0xFF) << 8;
+
+	packed_record[5] = rec->eth_type & 0xFFFF;
+
+	packed_record[6] = rec->snap[0] & 0xFFFF;
+	packed_record[7] = (rec->snap[0] >> 16) & 0xFFFF;
+
+	packed_record[8] = rec->snap[1] & 0xFF;
+
+	packed_record[8] |= (rec->llc & 0xFF) << 8;
+	packed_record[9] = (rec->llc >> 8) & 0xFFFF;
+
+	packed_record[10] = rec->mac_sa[0] & 0xFFFF;
+	packed_record[11] = (rec->mac_sa[0] >> 16) & 0xFFFF;
+
+	packed_record[12] = rec->mac_sa[1] & 0xFFFF;
+
+	packed_record[13] = rec->mac_da[0] & 0xFFFF;
+	packed_record[14] = (rec->mac_da[0] >> 16) & 0xFFFF;
+
+	packed_record[15] = rec->mac_da[1] & 0xFFFF;
+
+	packed_record[16] = rec->lpbk_packet & 0x1;
+
+	packed_record[16] |= (rec->an_mask & 0x3) << 1;
+
+	packed_record[16] |= (rec->tci_mask & 0x3F) << 3;
+
+	packed_record[16] |= (rec->sci_mask & 0x7F) << 9;
+	packed_record[17] = (rec->sci_mask >> 7) & 0x1;
+
+	packed_record[17] |= (rec->eth_type_mask & 0x3) << 1;
+
+	packed_record[17] |= (rec->snap_mask & 0x1F) << 3;
+
+	packed_record[17] |= (rec->llc_mask & 0x7) << 8;
+
+	packed_record[17] |= (rec->_802_2_encapsulate & 0x1) << 11;
+
+	packed_record[17] |= (rec->sa_mask & 0xF) << 12;
+	packed_record[18] = (rec->sa_mask >> 4) & 0x3;
+
+	packed_record[18] |= (rec->da_mask & 0x3F) << 2;
+
+	packed_record[18] |= (rec->lpbk_mask & 0x1) << 8;
+
+	packed_record[18] |= (rec->sc_idx & 0x1F) << 9;
+
+	packed_record[18] |= (rec->proc_dest & 0x1) << 14;
+
+	packed_record[18] |= (rec->action & 0x1) << 15;
+	packed_record[19] = (rec->action >> 1) & 0x1;
+
+	packed_record[19] |= (rec->ctrl_unctrl & 0x1) << 1;
+
+	packed_record[19] |= (rec->sci_from_table & 0x1) << 2;
+
+	packed_record[19] |= (rec->reserved & 0xF) << 3;
+
+	packed_record[19] |= (rec->valid & 0x1) << 7;
+
+	return set_raw_ingress_record(hw, packed_record, 20, 1,
+				      ROWOFFSET_INGRESSPRECLASSRECORD +
+					      table_index);
+}
+
+int aq_mss_set_ingress_preclass_record(struct aq_hw_s *hw,
+	const struct aq_mss_ingress_preclass_record *rec,
+	u16 table_index)
+{
+	int err = AQ_API_CALL_SAFE(set_ingress_preclass_record, hw, rec,
+				   table_index);
+
+	WARN_ONCE(err, "%s failed with %d\n", __func__, err);
+
+	return err;
+}
+
+static int
+get_ingress_preclass_record(struct aq_hw_s *hw,
+			    struct aq_mss_ingress_preclass_record *rec,
+			    u16 table_index)
+{
+	u16 packed_record[20];
+	int ret;
+
+	if (table_index >= NUMROWS_INGRESSPRECLASSRECORD)
+		return -EINVAL;
+
+	/* If the row that we want to read is odd, first read the previous even
+	 * row, throw that value away, and finally read the desired row.
+	 */
+	if ((table_index % 2) > 0) {
+		ret = get_raw_ingress_record(hw, packed_record, 20, 1,
+					     ROWOFFSET_INGRESSPRECLASSRECORD +
+						     table_index - 1);
+		if (unlikely(ret))
+			return ret;
+	}
+
+	ret = get_raw_ingress_record(hw, packed_record, 20, 1,
+				     ROWOFFSET_INGRESSPRECLASSRECORD +
+					     table_index);
+	if (unlikely(ret))
+		return ret;
+
+	rec->sci[0] = packed_record[0];
+	rec->sci[0] |= packed_record[1] << 16;
+
+	rec->sci[1] = packed_record[2];
+	rec->sci[1] |= packed_record[3] << 16;
+
+	rec->tci = packed_record[4] & 0xFF;
+
+	rec->encr_offset = (packed_record[4] >> 8) & 0xFF;
+
+	rec->eth_type = packed_record[5];
+
+	rec->snap[0] = packed_record[6];
+	rec->snap[0] |= packed_record[7] << 16;
+
+	rec->snap[1] = packed_record[8] & 0xFF;
+
+	rec->llc = (packed_record[8] >> 8) & 0xFF;
+	rec->llc |= packed_record[9] << 8;
+
+	rec->mac_sa[0] = packed_record[10];
+	rec->mac_sa[0] |= packed_record[11] << 16;
+
+	rec->mac_sa[1] = packed_record[12];
+
+	rec->mac_da[0] = packed_record[13];
+	rec->mac_da[0] |= packed_record[14] << 16;
+
+	rec->mac_da[1] = packed_record[15];
+
+	rec->lpbk_packet = packed_record[16] & 0x1;
+
+	rec->an_mask = (packed_record[16] >> 1) & 0x3;
+
+	rec->tci_mask = (packed_record[16] >> 3) & 0x3F;
+
+	rec->sci_mask = (packed_record[16] >> 9) & 0x7F;
+	rec->sci_mask |= (packed_record[17] & 0x1) << 7;
+
+	rec->eth_type_mask = (packed_record[17] >> 1) & 0x3;
+
+	rec->snap_mask = (packed_record[17] >> 3) & 0x1F;
+
+	rec->llc_mask = (packed_record[17] >> 8) & 0x7;
+
+	rec->_802_2_encapsulate = (packed_record[17] >> 11) & 0x1;
+
+	rec->sa_mask = (packed_record[17] >> 12) & 0xF;
+	rec->sa_mask |= (packed_record[18] & 0x3) << 4;
+
+	rec->da_mask = (packed_record[18] >> 2) & 0x3F;
+
+	rec->lpbk_mask = (packed_record[18] >> 8) & 0x1;
+
+	rec->sc_idx = (packed_record[18] >> 9) & 0x1F;
+
+	rec->proc_dest = (packed_record[18] >> 14) & 0x1;
+
+	rec->action = (packed_record[18] >> 15) & 0x1;
+	rec->action |= (packed_record[19] & 0x1) << 1;
+
+	rec->ctrl_unctrl = (packed_record[19] >> 1) & 0x1;
+
+	rec->sci_from_table = (packed_record[19] >> 2) & 0x1;
+
+	rec->reserved = (packed_record[19] >> 3) & 0xF;
+
+	rec->valid = (packed_record[19] >> 7) & 0x1;
+
+	return 0;
+}
+
+int aq_mss_get_ingress_preclass_record(struct aq_hw_s *hw,
+	struct aq_mss_ingress_preclass_record *rec,
+	u16 table_index)
+{
+	memset(rec, 0, sizeof(*rec));
+
+	return AQ_API_CALL_SAFE(get_ingress_preclass_record, hw, rec,
+				table_index);
+}
+
+static int set_ingress_sc_record(struct aq_hw_s *hw,
+				 const struct aq_mss_ingress_sc_record *rec,
+				 u16 table_index)
+{
+	u16 packed_record[8];
+
+	if (table_index >= NUMROWS_INGRESSSCRECORD)
+		return -EINVAL;
+
+	memset(packed_record, 0, sizeof(u16) * 8);
+
+	packed_record[0] = rec->stop_time & 0xFFFF;
+	packed_record[1] = (rec->stop_time >> 16) & 0xFFFF;
+
+	packed_record[2] = rec->start_time & 0xFFFF;
+	packed_record[3] = (rec->start_time >> 16) & 0xFFFF;
+
+	packed_record[4] = rec->validate_frames & 0x3;
+
+	packed_record[4] |= (rec->replay_protect & 0x1) << 2;
+
+	packed_record[4] |= (rec->anti_replay_window & 0x1FFF) << 3;
+	packed_record[5] = (rec->anti_replay_window >> 13) & 0xFFFF;
+	packed_record[6] = (rec->anti_replay_window >> 29) & 0x7;
+
+	packed_record[6] |= (rec->receiving & 0x1) << 3;
+
+	packed_record[6] |= (rec->fresh & 0x1) << 4;
+
+	packed_record[6] |= (rec->an_rol & 0x1) << 5;
+
+	packed_record[6] |= (rec->reserved & 0x3FF) << 6;
+	packed_record[7] = (rec->reserved >> 10) & 0x7FFF;
+
+	packed_record[7] |= (rec->valid & 0x1) << 15;
+
+	return set_raw_ingress_record(hw, packed_record, 8, 3,
+				      ROWOFFSET_INGRESSSCRECORD + table_index);
+}
+
+int aq_mss_set_ingress_sc_record(struct aq_hw_s *hw,
+				 const struct aq_mss_ingress_sc_record *rec,
+				 u16 table_index)
+{
+	int err = AQ_API_CALL_SAFE(set_ingress_sc_record, hw, rec, table_index);
+
+	WARN_ONCE(err, "%s failed with %d\n", __func__, err);
+
+	return err;
+}
+
+static int get_ingress_sc_record(struct aq_hw_s *hw,
+				 struct aq_mss_ingress_sc_record *rec,
+				 u16 table_index)
+{
+	u16 packed_record[8];
+	int ret;
+
+	if (table_index >= NUMROWS_INGRESSSCRECORD)
+		return -EINVAL;
+
+	ret = get_raw_ingress_record(hw, packed_record, 8, 3,
+				     ROWOFFSET_INGRESSSCRECORD + table_index);
+	if (unlikely(ret))
+		return ret;
+
+	rec->stop_time = packed_record[0];
+	rec->stop_time |= packed_record[1] << 16;
+
+	rec->start_time = packed_record[2];
+	rec->start_time |= packed_record[3] << 16;
+
+	rec->validate_frames = packed_record[4] & 0x3;
+
+	rec->replay_protect = (packed_record[4] >> 2) & 0x1;
+
+	rec->anti_replay_window = (packed_record[4] >> 3) & 0x1FFF;
+	rec->anti_replay_window |= packed_record[5] << 13;
+	rec->anti_replay_window |= (packed_record[6] & 0x7) << 29;
+
+	rec->receiving = (packed_record[6] >> 3) & 0x1;
+
+	rec->fresh = (packed_record[6] >> 4) & 0x1;
+
+	rec->an_rol = (packed_record[6] >> 5) & 0x1;
+
+	rec->reserved = (packed_record[6] >> 6) & 0x3FF;
+	rec->reserved |= (packed_record[7] & 0x7FFF) << 10;
+
+	rec->valid = (packed_record[7] >> 15) & 0x1;
+
+	return 0;
+}
+
+int aq_mss_get_ingress_sc_record(struct aq_hw_s *hw,
+				 struct aq_mss_ingress_sc_record *rec,
+				 u16 table_index)
+{
+	memset(rec, 0, sizeof(*rec));
+
+	return AQ_API_CALL_SAFE(get_ingress_sc_record, hw, rec, table_index);
+}
+
+static int set_ingress_sa_record(struct aq_hw_s *hw,
+				 const struct aq_mss_ingress_sa_record *rec,
+				 u16 table_index)
+{
+	u16 packed_record[8];
+
+	if (table_index >= NUMROWS_INGRESSSARECORD)
+		return -EINVAL;
+
+	memset(packed_record, 0, sizeof(u16) * 8);
+
+	packed_record[0] = rec->stop_time & 0xFFFF;
+	packed_record[1] = (rec->stop_time >> 16) & 0xFFFF;
+
+	packed_record[2] = rec->start_time & 0xFFFF;
+	packed_record[3] = (rec->start_time >> 16) & 0xFFFF;
+
+	packed_record[4] = rec->next_pn & 0xFFFF;
+	packed_record[5] = (rec->next_pn >> 16) & 0xFFFF;
+
+	packed_record[6] = rec->sat_nextpn & 0x1;
+
+	packed_record[6] |= (rec->in_use & 0x1) << 1;
+
+	packed_record[6] |= (rec->fresh & 0x1) << 2;
+
+	packed_record[6] |= (rec->reserved & 0x1FFF) << 3;
+	packed_record[7] = (rec->reserved >> 13) & 0x7FFF;
+
+	packed_record[7] |= (rec->valid & 0x1) << 15;
+
+	return set_raw_ingress_record(hw, packed_record, 8, 3,
+				      ROWOFFSET_INGRESSSARECORD + table_index);
+}
+
+int aq_mss_set_ingress_sa_record(struct aq_hw_s *hw,
+				 const struct aq_mss_ingress_sa_record *rec,
+				 u16 table_index)
+{
+	int err = AQ_API_CALL_SAFE(set_ingress_sa_record, hw, rec, table_index);
+
+	WARN_ONCE(err, "%s failed with %d\n", __func__, err);
+
+	return err;
+}
+
+static int get_ingress_sa_record(struct aq_hw_s *hw,
+				 struct aq_mss_ingress_sa_record *rec,
+				 u16 table_index)
+{
+	u16 packed_record[8];
+	int ret;
+
+	if (table_index >= NUMROWS_INGRESSSARECORD)
+		return -EINVAL;
+
+	ret = get_raw_ingress_record(hw, packed_record, 8, 3,
+				     ROWOFFSET_INGRESSSARECORD + table_index);
+	if (unlikely(ret))
+		return ret;
+
+	rec->stop_time = packed_record[0];
+	rec->stop_time |= packed_record[1] << 16;
+
+	rec->start_time = packed_record[2];
+	rec->start_time |= packed_record[3] << 16;
+
+	rec->next_pn = packed_record[4];
+	rec->next_pn |= packed_record[5] << 16;
+
+	rec->sat_nextpn = packed_record[6] & 0x1;
+
+	rec->in_use = (packed_record[6] >> 1) & 0x1;
+
+	rec->fresh = (packed_record[6] >> 2) & 0x1;
+
+	rec->reserved = (packed_record[6] >> 3) & 0x1FFF;
+	rec->reserved |= (packed_record[7] & 0x7FFF) << 13;
+
+	rec->valid = (packed_record[7] >> 15) & 0x1;
+
+	return 0;
+}
+
+int aq_mss_get_ingress_sa_record(struct aq_hw_s *hw,
+				 struct aq_mss_ingress_sa_record *rec,
+				 u16 table_index)
+{
+	memset(rec, 0, sizeof(*rec));
+
+	return AQ_API_CALL_SAFE(get_ingress_sa_record, hw, rec, table_index);
+}
+
+static int
+set_ingress_sakey_record(struct aq_hw_s *hw,
+			 const struct aq_mss_ingress_sakey_record *rec,
+			 u16 table_index)
+{
+	u16 packed_record[18];
+	int ret;
+
+	if (table_index >= NUMROWS_INGRESSSAKEYRECORD)
+		return -EINVAL;
+
+	memset(packed_record, 0, sizeof(u16) * 18);
+
+	packed_record[0] = rec->key[0] & 0xFFFF;
+	packed_record[1] = (rec->key[0] >> 16) & 0xFFFF;
+
+	packed_record[2] = rec->key[1] & 0xFFFF;
+	packed_record[3] = (rec->key[1] >> 16) & 0xFFFF;
+
+	packed_record[4] = rec->key[2] & 0xFFFF;
+	packed_record[5] = (rec->key[2] >> 16) & 0xFFFF;
+
+	packed_record[6] = rec->key[3] & 0xFFFF;
+	packed_record[7] = (rec->key[3] >> 16) & 0xFFFF;
+
+	packed_record[8] = rec->key[4] & 0xFFFF;
+	packed_record[9] = (rec->key[4] >> 16) & 0xFFFF;
+
+	packed_record[10] = rec->key[5] & 0xFFFF;
+	packed_record[11] = (rec->key[5] >> 16) & 0xFFFF;
+
+	packed_record[12] = rec->key[6] & 0xFFFF;
+	packed_record[13] = (rec->key[6] >> 16) & 0xFFFF;
+
+	packed_record[14] = rec->key[7] & 0xFFFF;
+	packed_record[15] = (rec->key[7] >> 16) & 0xFFFF;
+
+	packed_record[16] = rec->key_len & 0x3;
+
+	ret = set_raw_ingress_record(hw, packed_record, 18, 2,
+				     ROWOFFSET_INGRESSSAKEYRECORD +
+				     table_index);
+
+	memzero_explicit(packed_record, sizeof(packed_record));
+	return ret;
+}
+
+int aq_mss_set_ingress_sakey_record(struct aq_hw_s *hw,
+	const struct aq_mss_ingress_sakey_record *rec,
+	u16 table_index)
+{
+	int err = AQ_API_CALL_SAFE(set_ingress_sakey_record, hw, rec,
+				   table_index);
+
+	WARN_ONCE(err, "%s failed with %d\n", __func__, err);
+
+	return err;
+}
+
+static int get_ingress_sakey_record(struct aq_hw_s *hw,
+				    struct aq_mss_ingress_sakey_record *rec,
+				    u16 table_index)
+{
+	u16 packed_record[18];
+	int ret;
+
+	if (table_index >= NUMROWS_INGRESSSAKEYRECORD)
+		return -EINVAL;
+
+	ret = get_raw_ingress_record(hw, packed_record, 18, 2,
+				     ROWOFFSET_INGRESSSAKEYRECORD +
+					     table_index);
+	if (unlikely(ret))
+		return ret;
+
+	rec->key[0] = packed_record[0];
+	rec->key[0] |= packed_record[1] << 16;
+
+	rec->key[1] = packed_record[2];
+	rec->key[1] |= packed_record[3] << 16;
+
+	rec->key[2] = packed_record[4];
+	rec->key[2] |= packed_record[5] << 16;
+
+	rec->key[3] = packed_record[6];
+	rec->key[3] |= packed_record[7] << 16;
+
+	rec->key[4] = packed_record[8];
+	rec->key[4] |= packed_record[9] << 16;
+
+	rec->key[5] = packed_record[10];
+	rec->key[5] |= packed_record[11] << 16;
+
+	rec->key[6] = packed_record[12];
+	rec->key[6] |= packed_record[13] << 16;
+
+	rec->key[7] = packed_record[14];
+	rec->key[7] |= packed_record[15] << 16;
+
+	rec->key_len = packed_record[16] & 0x3;
+
+	return 0;
+}
+
+int aq_mss_get_ingress_sakey_record(struct aq_hw_s *hw,
+				    struct aq_mss_ingress_sakey_record *rec,
+				    u16 table_index)
+{
+	memset(rec, 0, sizeof(*rec));
+
+	return AQ_API_CALL_SAFE(get_ingress_sakey_record, hw, rec, table_index);
+}
+
+static int
+set_ingress_postclass_record(struct aq_hw_s *hw,
+			     const struct aq_mss_ingress_postclass_record *rec,
+			     u16 table_index)
+{
+	u16 packed_record[8];
+
+	if (table_index >= NUMROWS_INGRESSPOSTCLASSRECORD)
+		return -EINVAL;
+
+	memset(packed_record, 0, sizeof(u16) * 8);
+
+	packed_record[0] = rec->byte0 & 0xFF;
+
+	packed_record[0] |= (rec->byte1 & 0xFF) << 8;
+
+	packed_record[1] = rec->byte2 & 0xFF;
+
+	packed_record[1] |= (rec->byte3 & 0xFF) << 8;
+
+	packed_record[2] = rec->eth_type & 0xFFFF;
+
+	packed_record[3] = rec->eth_type_valid & 0x1;
+
+	packed_record[3] |= (rec->vlan_id & 0xFFF) << 1;
+
+	packed_record[3] |= (rec->vlan_up & 0x7) << 13;
+
+	packed_record[4] = rec->vlan_valid & 0x1;
+
+	packed_record[4] |= (rec->sai & 0x1F) << 1;
+
+	packed_record[4] |= (rec->sai_hit & 0x1) << 6;
+
+	packed_record[4] |= (rec->eth_type_mask & 0xF) << 7;
+
+	packed_record[4] |= (rec->byte3_location & 0x1F) << 11;
+	packed_record[5] = (rec->byte3_location >> 5) & 0x1;
+
+	packed_record[5] |= (rec->byte3_mask & 0x3) << 1;
+
+	packed_record[5] |= (rec->byte2_location & 0x3F) << 3;
+
+	packed_record[5] |= (rec->byte2_mask & 0x3) << 9;
+
+	packed_record[5] |= (rec->byte1_location & 0x1F) << 11;
+	packed_record[6] = (rec->byte1_location >> 5) & 0x1;
+
+	packed_record[6] |= (rec->byte1_mask & 0x3) << 1;
+
+	packed_record[6] |= (rec->byte0_location & 0x3F) << 3;
+
+	packed_record[6] |= (rec->byte0_mask & 0x3) << 9;
+
+	packed_record[6] |= (rec->eth_type_valid_mask & 0x3) << 11;
+
+	packed_record[6] |= (rec->vlan_id_mask & 0x7) << 13;
+	packed_record[7] = (rec->vlan_id_mask >> 3) & 0x1;
+
+	packed_record[7] |= (rec->vlan_up_mask & 0x3) << 1;
+
+	packed_record[7] |= (rec->vlan_valid_mask & 0x3) << 3;
+
+	packed_record[7] |= (rec->sai_mask & 0x3) << 5;
+
+	packed_record[7] |= (rec->sai_hit_mask & 0x3) << 7;
+
+	packed_record[7] |= (rec->firstlevel_actions & 0x1) << 9;
+
+	packed_record[7] |= (rec->secondlevel_actions & 0x1) << 10;
+
+	packed_record[7] |= (rec->reserved & 0xF) << 11;
+
+	packed_record[7] |= (rec->valid & 0x1) << 15;
+
+	return set_raw_ingress_record(hw, packed_record, 8, 4,
+				      ROWOFFSET_INGRESSPOSTCLASSRECORD +
+					      table_index);
+}
+
+int aq_mss_set_ingress_postclass_record(struct aq_hw_s *hw,
+	const struct aq_mss_ingress_postclass_record *rec,
+	u16 table_index)
+{
+	return AQ_API_CALL_SAFE(set_ingress_postclass_record, hw, rec,
+				table_index);
+}
+
+static int
+get_ingress_postclass_record(struct aq_hw_s *hw,
+			     struct aq_mss_ingress_postclass_record *rec,
+			     u16 table_index)
+{
+	u16 packed_record[8];
+	int ret;
+
+	if (table_index >= NUMROWS_INGRESSPOSTCLASSRECORD)
+		return -EINVAL;
+
+	/* If the row that we want to read is odd, first read the previous even
+	 * row, throw that value away, and finally read the desired row.
+	 */
+	if ((table_index % 2) > 0) {
+		ret = get_raw_ingress_record(hw, packed_record, 8, 4,
+					     ROWOFFSET_INGRESSPOSTCLASSRECORD +
+						     table_index - 1);
+		if (unlikely(ret))
+			return ret;
+	}
+
+	ret = get_raw_ingress_record(hw, packed_record, 8, 4,
+				     ROWOFFSET_INGRESSPOSTCLASSRECORD +
+					     table_index);
+	if (unlikely(ret))
+		return ret;
+
+	rec->byte0 = packed_record[0] & 0xFF;
+
+	rec->byte1 = (packed_record[0] >> 8) & 0xFF;
+
+	rec->byte2 = packed_record[1] & 0xFF;
+
+	rec->byte3 = (packed_record[1] >> 8) & 0xFF;
+
+	rec->eth_type = packed_record[2];
+
+	rec->eth_type_valid = packed_record[3] & 0x1;
+
+	rec->vlan_id = (packed_record[3] >> 1) & 0xFFF;
+
+	rec->vlan_up = (packed_record[3] >> 13) & 0x7;
+
+	rec->vlan_valid = packed_record[4] & 0x1;
+
+	rec->sai = (packed_record[4] >> 1) & 0x1F;
+
+	rec->sai_hit = (packed_record[4] >> 6) & 0x1;
+
+	rec->eth_type_mask = (packed_record[4] >> 7) & 0xF;
+
+	rec->byte3_location = (packed_record[4] >> 11) & 0x1F;
+	rec->byte3_location |= (packed_record[5] & 0x1) << 5;
+
+	rec->byte3_mask = (packed_record[5] >> 1) & 0x3;
+
+	rec->byte2_location = (packed_record[5] >> 3) & 0x3F;
+
+	rec->byte2_mask = (packed_record[5] >> 9) & 0x3;
+
+	rec->byte1_location = (packed_record[5] >> 11) & 0x1F;
+	rec->byte1_location |= (packed_record[6] & 0x1) << 5;
+
+	rec->byte1_mask = (packed_record[6] >> 1) & 0x3;
+
+	rec->byte0_location = (packed_record[6] >> 3) & 0x3F;
+
+	rec->byte0_mask = (packed_record[6] >> 9) & 0x3;
+
+	rec->eth_type_valid_mask = (packed_record[6] >> 11) & 0x3;
+
+	rec->vlan_id_mask = (packed_record[6] >> 13) & 0x7;
+	rec->vlan_id_mask |= (packed_record[7] & 0x1) << 3;
+
+	rec->vlan_up_mask = (packed_record[7] >> 1) & 0x3;
+
+	rec->vlan_valid_mask = (packed_record[7] >> 3) & 0x3;
+
+	rec->sai_mask = (packed_record[7] >> 5) & 0x3;
+
+	rec->sai_hit_mask = (packed_record[7] >> 7) & 0x3;
+
+	rec->firstlevel_actions = (packed_record[7] >> 9) & 0x1;
+
+	rec->secondlevel_actions = (packed_record[7] >> 10) & 0x1;
+
+	rec->reserved = (packed_record[7] >> 11) & 0xF;
+
+	rec->valid = (packed_record[7] >> 15) & 0x1;
+
+	return 0;
+}
+
+int aq_mss_get_ingress_postclass_record(struct aq_hw_s *hw,
+	struct aq_mss_ingress_postclass_record *rec,
+	u16 table_index)
+{
+	memset(rec, 0, sizeof(*rec));
+
+	return AQ_API_CALL_SAFE(get_ingress_postclass_record, hw, rec,
+				table_index);
+}
+
+static int
+set_ingress_postctlf_record(struct aq_hw_s *hw,
+			    const struct aq_mss_ingress_postctlf_record *rec,
+			    u16 table_index)
+{
+	u16 packed_record[6];
+
+	if (table_index >= NUMROWS_INGRESSPOSTCTLFRECORD)
+		return -EINVAL;
+
+	memset(packed_record, 0, sizeof(u16) * 6);
+
+	packed_record[0] = rec->sa_da[0] & 0xFFFF;
+	packed_record[1] = (rec->sa_da[0] >> 16) & 0xFFFF;
+
+	packed_record[2] = rec->sa_da[1] & 0xFFFF;
+
+	packed_record[3] = rec->eth_type & 0xFFFF;
+
+	packed_record[4] = rec->match_mask & 0xFFFF;
+
+	packed_record[5] = rec->match_type & 0xF;
+
+	packed_record[5] |= (rec->action & 0x1) << 4;
+
+	return set_raw_ingress_record(hw, packed_record, 6, 5,
+				      ROWOFFSET_INGRESSPOSTCTLFRECORD +
+					      table_index);
+}
+
+int aq_mss_set_ingress_postctlf_record(struct aq_hw_s *hw,
+	const struct aq_mss_ingress_postctlf_record *rec,
+	u16 table_index)
+{
+	return AQ_API_CALL_SAFE(set_ingress_postctlf_record, hw, rec,
+				table_index);
+}
+
+static int
+get_ingress_postctlf_record(struct aq_hw_s *hw,
+			    struct aq_mss_ingress_postctlf_record *rec,
+			    u16 table_index)
+{
+	u16 packed_record[6];
+	int ret;
+
+	if (table_index >= NUMROWS_INGRESSPOSTCTLFRECORD)
+		return -EINVAL;
+
+	/* If the row that we want to read is odd, first read the previous even
+	 * row, throw that value away, and finally read the desired row.
+	 */
+	if ((table_index % 2) > 0) {
+		ret = get_raw_ingress_record(hw, packed_record, 6, 5,
+					     ROWOFFSET_INGRESSPOSTCTLFRECORD +
+						     table_index - 1);
+		if (unlikely(ret))
+			return ret;
+	}
+
+	ret = get_raw_ingress_record(hw, packed_record, 6, 5,
+				     ROWOFFSET_INGRESSPOSTCTLFRECORD +
+					     table_index);
+	if (unlikely(ret))
+		return ret;
+
+	rec->sa_da[0] = packed_record[0];
+	rec->sa_da[0] |= packed_record[1] << 16;
+
+	rec->sa_da[1] = packed_record[2];
+
+	rec->eth_type = packed_record[3];
+
+	rec->match_mask = packed_record[4];
+
+	rec->match_type = packed_record[5] & 0xF;
+
+	rec->action = (packed_record[5] >> 4) & 0x1;
+
+	return 0;
+}
+
+int aq_mss_get_ingress_postctlf_record(struct aq_hw_s *hw,
+	struct aq_mss_ingress_postctlf_record *rec,
+	u16 table_index)
+{
+	memset(rec, 0, sizeof(*rec));
+
+	return AQ_API_CALL_SAFE(get_ingress_postctlf_record, hw, rec,
+				table_index);
+}
+
+static int set_egress_ctlf_record(struct aq_hw_s *hw,
+				  const struct aq_mss_egress_ctlf_record *rec,
+				  u16 table_index)
+{
+	u16 packed_record[6];
+
+	if (table_index >= NUMROWS_EGRESSCTLFRECORD)
+		return -EINVAL;
+
+	memset(packed_record, 0, sizeof(u16) * 6);
+
+	packed_record[0] = rec->sa_da[0] & 0xFFFF;
+	packed_record[1] = (rec->sa_da[0] >> 16) & 0xFFFF;
+
+	packed_record[2] = rec->sa_da[1] & 0xFFFF;
+
+	packed_record[3] = rec->eth_type & 0xFFFF;
+
+	packed_record[4] = rec->match_mask & 0xFFFF;
+
+	packed_record[5] = rec->match_type & 0xF;
+
+	packed_record[5] |= (rec->action & 0x1) << 4;
+
+	return set_raw_egress_record(hw, packed_record, 6, 0,
+				     ROWOFFSET_EGRESSCTLFRECORD + table_index);
+}
+
+int aq_mss_set_egress_ctlf_record(struct aq_hw_s *hw,
+				  const struct aq_mss_egress_ctlf_record *rec,
+				  u16 table_index)
+{
+	return AQ_API_CALL_SAFE(set_egress_ctlf_record, hw, rec, table_index);
+}
+
+static int get_egress_ctlf_record(struct aq_hw_s *hw,
+				  struct aq_mss_egress_ctlf_record *rec,
+				  u16 table_index)
+{
+	u16 packed_record[6];
+	int ret;
+
+	if (table_index >= NUMROWS_EGRESSCTLFRECORD)
+		return -EINVAL;
+
+	/* If the row that we want to read is odd, first read the previous even
+	 * row, throw that value away, and finally read the desired row.
+	 */
+	if ((table_index % 2) > 0) {
+		ret = get_raw_egress_record(hw, packed_record, 6, 0,
+					    ROWOFFSET_EGRESSCTLFRECORD +
+						    table_index - 1);
+		if (unlikely(ret))
+			return ret;
+	}
+
+	ret = get_raw_egress_record(hw, packed_record, 6, 0,
+				    ROWOFFSET_EGRESSCTLFRECORD + table_index);
+	if (unlikely(ret))
+		return ret;
+
+	rec->sa_da[0] = packed_record[0];
+	rec->sa_da[0] |= packed_record[1] << 16;
+
+	rec->sa_da[1] = packed_record[2];
+
+	rec->eth_type = packed_record[3];
+
+	rec->match_mask = packed_record[4];
+
+	rec->match_type = packed_record[5] & 0xF;
+
+	rec->action = (packed_record[5] >> 4) & 0x1;
+
+	return 0;
+}
+
+int aq_mss_get_egress_ctlf_record(struct aq_hw_s *hw,
+				  struct aq_mss_egress_ctlf_record *rec,
+				  u16 table_index)
+{
+	memset(rec, 0, sizeof(*rec));
+
+	return AQ_API_CALL_SAFE(get_egress_ctlf_record, hw, rec, table_index);
+}
+
+static int set_egress_class_record(struct aq_hw_s *hw,
+				   const struct aq_mss_egress_class_record *rec,
+				   u16 table_index)
+{
+	u16 packed_record[28];
+
+	if (table_index >= NUMROWS_EGRESSCLASSRECORD)
+		return -EINVAL;
+
+	memset(packed_record, 0, sizeof(u16) * 28);
+
+	packed_record[0] = rec->vlan_id & 0xFFF;
+
+	packed_record[0] |= (rec->vlan_up & 0x7) << 12;
+
+	packed_record[0] |= (rec->vlan_valid & 0x1) << 15;
+
+	packed_record[1] = rec->byte3 & 0xFF;
+
+	packed_record[1] |= (rec->byte2 & 0xFF) << 8;
+
+	packed_record[2] = rec->byte1 & 0xFF;
+
+	packed_record[2] |= (rec->byte0 & 0xFF) << 8;
+
+	packed_record[3] = rec->tci & 0xFF;
+
+	packed_record[3] |= (rec->sci[0] & 0xFF) << 8;
+	packed_record[4] = (rec->sci[0] >> 8) & 0xFFFF;
+	packed_record[5] = (rec->sci[0] >> 24) & 0xFF;
+
+	packed_record[5] |= (rec->sci[1] & 0xFF) << 8;
+	packed_record[6] = (rec->sci[1] >> 8) & 0xFFFF;
+	packed_record[7] = (rec->sci[1] >> 24) & 0xFF;
+
+	packed_record[7] |= (rec->eth_type & 0xFF) << 8;
+	packed_record[8] = (rec->eth_type >> 8) & 0xFF;
+
+	packed_record[8] |= (rec->snap[0] & 0xFF) << 8;
+	packed_record[9] = (rec->snap[0] >> 8) & 0xFFFF;
+	packed_record[10] = (rec->snap[0] >> 24) & 0xFF;
+
+	packed_record[10] |= (rec->snap[1] & 0xFF) << 8;
+
+	packed_record[11] = rec->llc & 0xFFFF;
+	packed_record[12] = (rec->llc >> 16) & 0xFF;
+
+	packed_record[12] |= (rec->mac_sa[0] & 0xFF) << 8;
+	packed_record[13] = (rec->mac_sa[0] >> 8) & 0xFFFF;
+	packed_record[14] = (rec->mac_sa[0] >> 24) & 0xFF;
+
+	packed_record[14] |= (rec->mac_sa[1] & 0xFF) << 8;
+	packed_record[15] = (rec->mac_sa[1] >> 8) & 0xFF;
+
+	packed_record[15] |= (rec->mac_da[0] & 0xFF) << 8;
+	packed_record[16] = (rec->mac_da[0] >> 8) & 0xFFFF;
+	packed_record[17] = (rec->mac_da[0] >> 24) & 0xFF;
+
+	packed_record[17] |= (rec->mac_da[1] & 0xFF) << 8;
+	packed_record[18] = (rec->mac_da[1] >> 8) & 0xFF;
+
+	packed_record[18] |= (rec->pn & 0xFF) << 8;
+	packed_record[19] = (rec->pn >> 8) & 0xFFFF;
+	packed_record[20] = (rec->pn >> 24) & 0xFF;
+
+	packed_record[20] |= (rec->byte3_location & 0x3F) << 8;
+
+	packed_record[20] |= (rec->byte3_mask & 0x1) << 14;
+
+	packed_record[20] |= (rec->byte2_location & 0x1) << 15;
+	packed_record[21] = (rec->byte2_location >> 1) & 0x1F;
+
+	packed_record[21] |= (rec->byte2_mask & 0x1) << 5;
+
+	packed_record[21] |= (rec->byte1_location & 0x3F) << 6;
+
+	packed_record[21] |= (rec->byte1_mask & 0x1) << 12;
+
+	packed_record[21] |= (rec->byte0_location & 0x7) << 13;
+	packed_record[22] = (rec->byte0_location >> 3) & 0x7;
+
+	packed_record[22] |= (rec->byte0_mask & 0x1) << 3;
+
+	packed_record[22] |= (rec->vlan_id_mask & 0x3) << 4;
+
+	packed_record[22] |= (rec->vlan_up_mask & 0x1) << 6;
+
+	packed_record[22] |= (rec->vlan_valid_mask & 0x1) << 7;
+
+	packed_record[22] |= (rec->tci_mask & 0xFF) << 8;
+
+	packed_record[23] = rec->sci_mask & 0xFF;
+
+	packed_record[23] |= (rec->eth_type_mask & 0x3) << 8;
+
+	packed_record[23] |= (rec->snap_mask & 0x1F) << 10;
+
+	packed_record[23] |= (rec->llc_mask & 0x1) << 15;
+	packed_record[24] = (rec->llc_mask >> 1) & 0x3;
+
+	packed_record[24] |= (rec->sa_mask & 0x3F) << 2;
+
+	packed_record[24] |= (rec->da_mask & 0x3F) << 8;
+
+	packed_record[24] |= (rec->pn_mask & 0x3) << 14;
+	packed_record[25] = (rec->pn_mask >> 2) & 0x3;
+
+	packed_record[25] |= (rec->eight02dot2 & 0x1) << 2;
+
+	packed_record[25] |= (rec->tci_sc & 0x1) << 3;
+
+	packed_record[25] |= (rec->tci_87543 & 0x1) << 4;
+
+	packed_record[25] |= (rec->exp_sectag_en & 0x1) << 5;
+
+	packed_record[25] |= (rec->sc_idx & 0x1F) << 6;
+
+	packed_record[25] |= (rec->sc_sa & 0x3) << 11;
+
+	packed_record[25] |= (rec->debug & 0x1) << 13;
+
+	packed_record[25] |= (rec->action & 0x3) << 14;
+
+	packed_record[26] = (rec->valid & 0x1) << 3;
+
+	return set_raw_egress_record(hw, packed_record, 28, 1,
+				     ROWOFFSET_EGRESSCLASSRECORD + table_index);
+}
+
+int aq_mss_set_egress_class_record(struct aq_hw_s *hw,
+				   const struct aq_mss_egress_class_record *rec,
+				   u16 table_index)
+{
+	return AQ_API_CALL_SAFE(set_egress_class_record, hw, rec, table_index);
+}
+
+static int get_egress_class_record(struct aq_hw_s *hw,
+				   struct aq_mss_egress_class_record *rec,
+				   u16 table_index)
+{
+	u16 packed_record[28];
+	int ret;
+
+	if (table_index >= NUMROWS_EGRESSCLASSRECORD)
+		return -EINVAL;
+
+	/* If the row that we want to read is odd, first read the previous even
+	 * row, throw that value away, and finally read the desired row.
+	 */
+	if ((table_index % 2) > 0) {
+		ret = get_raw_egress_record(hw, packed_record, 28, 1,
+					    ROWOFFSET_EGRESSCLASSRECORD +
+						    table_index - 1);
+		if (unlikely(ret))
+			return ret;
+	}
+
+	ret = get_raw_egress_record(hw, packed_record, 28, 1,
+				    ROWOFFSET_EGRESSCLASSRECORD + table_index);
+	if (unlikely(ret))
+		return ret;
+
+	rec->vlan_id = packed_record[0] & 0xFFF;
+
+	rec->vlan_up = (packed_record[0] >> 12) & 0x7;
+
+	rec->vlan_valid = (packed_record[0] >> 15) & 0x1;
+
+	rec->byte3 = packed_record[1] & 0xFF;
+
+	rec->byte2 = (packed_record[1] >> 8) & 0xFF;
+
+	rec->byte1 = packed_record[2] & 0xFF;
+
+	rec->byte0 = (packed_record[2] >> 8) & 0xFF;
+
+	rec->tci = packed_record[3] & 0xFF;
+
+	rec->sci[0] = (packed_record[3] >> 8) & 0xFF;
+	rec->sci[0] |= packed_record[4] << 8;
+	rec->sci[0] |= (packed_record[5] & 0xFF) << 24;
+
+	rec->sci[1] = (packed_record[5] >> 8) & 0xFF;
+	rec->sci[1] |= packed_record[6] << 8;
+	rec->sci[1] |= (packed_record[7] & 0xFF) << 24;
+
+	rec->eth_type = (packed_record[7] >> 8) & 0xFF;
+	rec->eth_type |= (packed_record[8] & 0xFF) << 8;
+
+	rec->snap[0] = (packed_record[8] >> 8) & 0xFF;
+	rec->snap[0] |= packed_record[9] << 8;
+	rec->snap[0] |= (packed_record[10] & 0xFF) << 24;
+
+	rec->snap[1] = (packed_record[10] >> 8) & 0xFF;
+
+	rec->llc = packed_record[11];
+	rec->llc |= (packed_record[12] & 0xFF) << 16;
+
+	rec->mac_sa[0] = (packed_record[12] >> 8) & 0xFF;
+	rec->mac_sa[0] |= packed_record[13] << 8;
+	rec->mac_sa[0] |= (packed_record[14] & 0xFF) << 24;
+
+	rec->mac_sa[1] = (packed_record[14] >> 8) & 0xFF;
+	rec->mac_sa[1] |= (packed_record[15] & 0xFF) << 8;
+
+	rec->mac_da[0] = (packed_record[15] >> 8) & 0xFF;
+	rec->mac_da[0] |= packed_record[16] << 8;
+	rec->mac_da[0] |= (packed_record[17] & 0xFF) << 24;
+
+	rec->mac_da[1] = (packed_record[17] >> 8) & 0xFF;
+	rec->mac_da[1] |= (packed_record[18] & 0xFF) << 8;
+
+	rec->pn = (packed_record[18] >> 8) & 0xFF;
+	rec->pn |= packed_record[19] << 8;
+	rec->pn |= (packed_record[20] & 0xFF) << 24;
+
+	rec->byte3_location = (packed_record[20] >> 8) & 0x3F;
+
+	rec->byte3_mask = (packed_record[20] >> 14) & 0x1;
+
+	rec->byte2_location = (packed_record[20] >> 15) & 0x1;
+	rec->byte2_location |= (packed_record[21] & 0x1F) << 1;
+
+	rec->byte2_mask = (packed_record[21] >> 5) & 0x1;
+
+	rec->byte1_location = (packed_record[21] >> 6) & 0x3F;
+
+	rec->byte1_mask = (packed_record[21] >> 12) & 0x1;
+
+	rec->byte0_location = (packed_record[21] >> 13) & 0x7;
+	rec->byte0_location |= (packed_record[22] & 0x7) << 3;
+
+	rec->byte0_mask = (packed_record[22] >> 3) & 0x1;
+
+	rec->vlan_id_mask = (packed_record[22] >> 4) & 0x3;
+
+	rec->vlan_up_mask = (packed_record[22] >> 6) & 0x1;
+
+	rec->vlan_valid_mask = (packed_record[22] >> 7) & 0x1;
+
+	rec->tci_mask = (packed_record[22] >> 8) & 0xFF;
+
+	rec->sci_mask = packed_record[23] & 0xFF;
+
+	rec->eth_type_mask = (packed_record[23] >> 8) & 0x3;
+
+	rec->snap_mask = (packed_record[23] >> 10) & 0x1F;
+
+	rec->llc_mask = (packed_record[23] >> 15) & 0x1;
+	rec->llc_mask |= (packed_record[24] & 0x3) << 1;
+
+	rec->sa_mask = (packed_record[24] >> 2) & 0x3F;
+
+	rec->da_mask = (packed_record[24] >> 8) & 0x3F;
+
+	rec->pn_mask = (packed_record[24] >> 14) & 0x3;
+	rec->pn_mask |= (packed_record[25] & 0x3) << 2;
+
+	rec->eight02dot2 = (packed_record[25] >> 2) & 0x1;
+
+	rec->tci_sc = (packed_record[25] >> 3) & 0x1;
+
+	rec->tci_87543 = (packed_record[25] >> 4) & 0x1;
+
+	rec->exp_sectag_en = (packed_record[25] >> 5) & 0x1;
+
+	rec->sc_idx = (packed_record[25] >> 6) & 0x1F;
+
+	rec->sc_sa = (packed_record[25] >> 11) & 0x3;
+
+	rec->debug = (packed_record[25] >> 13) & 0x1;
+
+	rec->action = (packed_record[25] >> 14) & 0x3;
+
+	rec->valid = (packed_record[26] >> 3) & 0x1;
+
+	return 0;
+}
+
+int aq_mss_get_egress_class_record(struct aq_hw_s *hw,
+				   struct aq_mss_egress_class_record *rec,
+				   u16 table_index)
+{
+	memset(rec, 0, sizeof(*rec));
+
+	return AQ_API_CALL_SAFE(get_egress_class_record, hw, rec, table_index);
+}
+
+static int set_egress_sc_record(struct aq_hw_s *hw,
+				const struct aq_mss_egress_sc_record *rec,
+				u16 table_index)
+{
+	u16 packed_record[8];
+
+	if (table_index >= NUMROWS_EGRESSSCRECORD)
+		return -EINVAL;
+
+	memset(packed_record, 0, sizeof(u16) * 8);
+
+	packed_record[0] = rec->start_time & 0xFFFF;
+	packed_record[1] = (rec->start_time >> 16) & 0xFFFF;
+
+	packed_record[2] = rec->stop_time & 0xFFFF;
+	packed_record[3] = (rec->stop_time >> 16) & 0xFFFF;
+
+	packed_record[4] = rec->curr_an & 0x3;
+
+	packed_record[4] |= (rec->an_roll & 0x1) << 2;
+
+	packed_record[4] |= (rec->tci & 0x3F) << 3;
+
+	packed_record[4] |= (rec->enc_off & 0x7F) << 9;
+	packed_record[5] = (rec->enc_off >> 7) & 0x1;
+
+	packed_record[5] |= (rec->protect & 0x1) << 1;
+
+	packed_record[5] |= (rec->recv & 0x1) << 2;
+
+	packed_record[5] |= (rec->fresh & 0x1) << 3;
+
+	packed_record[5] |= (rec->sak_len & 0x3) << 4;
+
+	packed_record[7] = (rec->valid & 0x1) << 15;
+
+	return set_raw_egress_record(hw, packed_record, 8, 2,
+				     ROWOFFSET_EGRESSSCRECORD + table_index);
+}
+
+int aq_mss_set_egress_sc_record(struct aq_hw_s *hw,
+				const struct aq_mss_egress_sc_record *rec,
+				u16 table_index)
+{
+	return AQ_API_CALL_SAFE(set_egress_sc_record, hw, rec, table_index);
+}
+
+static int get_egress_sc_record(struct aq_hw_s *hw,
+				struct aq_mss_egress_sc_record *rec,
+				u16 table_index)
+{
+	u16 packed_record[8];
+	int ret;
+
+	if (table_index >= NUMROWS_EGRESSSCRECORD)
+		return -EINVAL;
+
+	ret = get_raw_egress_record(hw, packed_record, 8, 2,
+				    ROWOFFSET_EGRESSSCRECORD + table_index);
+	if (unlikely(ret))
+		return ret;
+
+	rec->start_time = packed_record[0];
+	rec->start_time |= packed_record[1] << 16;
+
+	rec->stop_time = packed_record[2];
+	rec->stop_time |= packed_record[3] << 16;
+
+	rec->curr_an = packed_record[4] & 0x3;
+
+	rec->an_roll = (packed_record[4] >> 2) & 0x1;
+
+	rec->tci = (packed_record[4] >> 3) & 0x3F;
+
+	rec->enc_off = (packed_record[4] >> 9) & 0x7F;
+	rec->enc_off |= (packed_record[5] & 0x1) << 7;
+
+	rec->protect = (packed_record[5] >> 1) & 0x1;
+
+	rec->recv = (packed_record[5] >> 2) & 0x1;
+
+	rec->fresh = (packed_record[5] >> 3) & 0x1;
+
+	rec->sak_len = (packed_record[5] >> 4) & 0x3;
+
+	rec->valid = (packed_record[7] >> 15) & 0x1;
+
+	return 0;
+}
+
+int aq_mss_get_egress_sc_record(struct aq_hw_s *hw,
+				struct aq_mss_egress_sc_record *rec,
+				u16 table_index)
+{
+	memset(rec, 0, sizeof(*rec));
+
+	return AQ_API_CALL_SAFE(get_egress_sc_record, hw, rec, table_index);
+}
+
+static int set_egress_sa_record(struct aq_hw_s *hw,
+				const struct aq_mss_egress_sa_record *rec,
+				u16 table_index)
+{
+	u16 packed_record[8];
+
+	if (table_index >= NUMROWS_EGRESSSARECORD)
+		return -EINVAL;
+
+	memset(packed_record, 0, sizeof(u16) * 8);
+
+	packed_record[0] = rec->start_time & 0xFFFF;
+	packed_record[1] = (rec->start_time >> 16) & 0xFFFF;
+
+	packed_record[2] = rec->stop_time & 0xFFFF;
+	packed_record[3] = (rec->stop_time >> 16) & 0xFFFF;
+
+	packed_record[4] = rec->next_pn & 0xFFFF;
+	packed_record[5] = (rec->next_pn >> 16) & 0xFFFF;
+
+	packed_record[6] = rec->sat_pn & 0x1;
+
+	packed_record[6] |= (rec->fresh & 0x1) << 1;
+
+	packed_record[7] = (rec->valid & 0x1) << 15;
+
+	return set_raw_egress_record(hw, packed_record, 8, 2,
+				     ROWOFFSET_EGRESSSARECORD + table_index);
+}
+
+int aq_mss_set_egress_sa_record(struct aq_hw_s *hw,
+				const struct aq_mss_egress_sa_record *rec,
+				u16 table_index)
+{
+	int err = AQ_API_CALL_SAFE(set_egress_sa_record, hw, rec, table_index);
+
+	WARN_ONCE(err, "%s failed with %d\n", __func__, err);
+
+	return err;
+}
+
+static int get_egress_sa_record(struct aq_hw_s *hw,
+				struct aq_mss_egress_sa_record *rec,
+				u16 table_index)
+{
+	u16 packed_record[8];
+	int ret;
+
+	if (table_index >= NUMROWS_EGRESSSARECORD)
+		return -EINVAL;
+
+	ret = get_raw_egress_record(hw, packed_record, 8, 2,
+				    ROWOFFSET_EGRESSSARECORD + table_index);
+	if (unlikely(ret))
+		return ret;
+
+	rec->start_time = packed_record[0];
+	rec->start_time |= packed_record[1] << 16;
+
+	rec->stop_time = packed_record[2];
+	rec->stop_time |= packed_record[3] << 16;
+
+	rec->next_pn = packed_record[4];
+	rec->next_pn |= packed_record[5] << 16;
+
+	rec->sat_pn = packed_record[6] & 0x1;
+
+	rec->fresh = (packed_record[6] >> 1) & 0x1;
+
+	rec->valid = (packed_record[7] >> 15) & 0x1;
+
+	return 0;
+}
+
+int aq_mss_get_egress_sa_record(struct aq_hw_s *hw,
+				struct aq_mss_egress_sa_record *rec,
+				u16 table_index)
+{
+	memset(rec, 0, sizeof(*rec));
+
+	return AQ_API_CALL_SAFE(get_egress_sa_record, hw, rec, table_index);
+}
+
+static int set_egress_sakey_record(struct aq_hw_s *hw,
+				   const struct aq_mss_egress_sakey_record *rec,
+				   u16 table_index)
+{
+	u16 packed_record[16];
+	int ret;
+
+	if (table_index >= NUMROWS_EGRESSSAKEYRECORD)
+		return -EINVAL;
+
+	memset(packed_record, 0, sizeof(u16) * 16);
+
+	packed_record[0] = rec->key[0] & 0xFFFF;
+	packed_record[1] = (rec->key[0] >> 16) & 0xFFFF;
+
+	packed_record[2] = rec->key[1] & 0xFFFF;
+	packed_record[3] = (rec->key[1] >> 16) & 0xFFFF;
+
+	packed_record[4] = rec->key[2] & 0xFFFF;
+	packed_record[5] = (rec->key[2] >> 16) & 0xFFFF;
+
+	packed_record[6] = rec->key[3] & 0xFFFF;
+	packed_record[7] = (rec->key[3] >> 16) & 0xFFFF;
+
+	packed_record[8] = rec->key[4] & 0xFFFF;
+	packed_record[9] = (rec->key[4] >> 16) & 0xFFFF;
+
+	packed_record[10] = rec->key[5] & 0xFFFF;
+	packed_record[11] = (rec->key[5] >> 16) & 0xFFFF;
+
+	packed_record[12] = rec->key[6] & 0xFFFF;
+	packed_record[13] = (rec->key[6] >> 16) & 0xFFFF;
+
+	packed_record[14] = rec->key[7] & 0xFFFF;
+	packed_record[15] = (rec->key[7] >> 16) & 0xFFFF;
+
+	ret = set_raw_egress_record(hw, packed_record, 8, 2,
+				    ROWOFFSET_EGRESSSAKEYRECORD + table_index);
+	if (unlikely(ret))
+		goto clear_key;
+	ret = set_raw_egress_record(hw, packed_record + 8, 8, 2,
+				    ROWOFFSET_EGRESSSAKEYRECORD + table_index -
+					    32);
+
+clear_key:
+	memzero_explicit(packed_record, sizeof(packed_record));
+	return ret;
+}
+
+int aq_mss_set_egress_sakey_record(struct aq_hw_s *hw,
+				   const struct aq_mss_egress_sakey_record *rec,
+				   u16 table_index)
+{
+	int err = AQ_API_CALL_SAFE(set_egress_sakey_record, hw, rec,
+				   table_index);
+
+	WARN_ONCE(err, "%s failed with %d\n", __func__, err);
+
+	return err;
+}
+
+static int get_egress_sakey_record(struct aq_hw_s *hw,
+				   struct aq_mss_egress_sakey_record *rec,
+				   u16 table_index)
+{
+	u16 packed_record[16];
+	int ret;
+
+	if (table_index >= NUMROWS_EGRESSSAKEYRECORD)
+		return -EINVAL;
+
+	ret = get_raw_egress_record(hw, packed_record, 8, 2,
+				    ROWOFFSET_EGRESSSAKEYRECORD + table_index);
+	if (unlikely(ret))
+		return ret;
+	ret = get_raw_egress_record(hw, packed_record + 8, 8, 2,
+				    ROWOFFSET_EGRESSSAKEYRECORD + table_index -
+					    32);
+	if (unlikely(ret))
+		return ret;
+
+	rec->key[0] = packed_record[0];
+	rec->key[0] |= packed_record[1] << 16;
+
+	rec->key[1] = packed_record[2];
+	rec->key[1] |= packed_record[3] << 16;
+
+	rec->key[2] = packed_record[4];
+	rec->key[2] |= packed_record[5] << 16;
+
+	rec->key[3] = packed_record[6];
+	rec->key[3] |= packed_record[7] << 16;
+
+	rec->key[4] = packed_record[8];
+	rec->key[4] |= packed_record[9] << 16;
+
+	rec->key[5] = packed_record[10];
+	rec->key[5] |= packed_record[11] << 16;
+
+	rec->key[6] = packed_record[12];
+	rec->key[6] |= packed_record[13] << 16;
+
+	rec->key[7] = packed_record[14];
+	rec->key[7] |= packed_record[15] << 16;
+
+	return 0;
+}
+
+int aq_mss_get_egress_sakey_record(struct aq_hw_s *hw,
+				   struct aq_mss_egress_sakey_record *rec,
+				   u16 table_index)
+{
+	memset(rec, 0, sizeof(*rec));
+
+	return AQ_API_CALL_SAFE(get_egress_sakey_record, hw, rec, table_index);
+}
+
+static int get_egress_sc_counters(struct aq_hw_s *hw,
+				  struct aq_mss_egress_sc_counters *counters,
+				  u16 sc_index)
+{
+	u16 packed_record[4];
+	int ret;
+
+	if (sc_index >= NUMROWS_EGRESSSCRECORD)
+		return -EINVAL;
+
+	ret = get_raw_egress_record(hw, packed_record, 4, 3, sc_index * 8 + 4);
+	if (unlikely(ret))
+		return ret;
+	counters->sc_protected_pkts[0] =
+		packed_record[0] | (packed_record[1] << 16);
+	counters->sc_protected_pkts[1] =
+		packed_record[2] | (packed_record[3] << 16);
+
+	ret = get_raw_egress_record(hw, packed_record, 4, 3, sc_index * 8 + 5);
+	if (unlikely(ret))
+		return ret;
+	counters->sc_encrypted_pkts[0] =
+		packed_record[0] | (packed_record[1] << 16);
+	counters->sc_encrypted_pkts[1] =
+		packed_record[2] | (packed_record[3] << 16);
+
+	ret = get_raw_egress_record(hw, packed_record, 4, 3, sc_index * 8 + 6);
+	if (unlikely(ret))
+		return ret;
+	counters->sc_protected_octets[0] =
+		packed_record[0] | (packed_record[1] << 16);
+	counters->sc_protected_octets[1] =
+		packed_record[2] | (packed_record[3] << 16);
+
+	ret = get_raw_egress_record(hw, packed_record, 4, 3, sc_index * 8 + 7);
+	if (unlikely(ret))
+		return ret;
+	counters->sc_encrypted_octets[0] =
+		packed_record[0] | (packed_record[1] << 16);
+	counters->sc_encrypted_octets[1] =
+		packed_record[2] | (packed_record[3] << 16);
+
+	return 0;
+}
+
+int aq_mss_get_egress_sc_counters(struct aq_hw_s *hw,
+				  struct aq_mss_egress_sc_counters *counters,
+				  u16 sc_index)
+{
+	memset(counters, 0, sizeof(*counters));
+
+	return AQ_API_CALL_SAFE(get_egress_sc_counters, hw, counters, sc_index);
+}
+
+static int get_egress_sa_counters(struct aq_hw_s *hw,
+				  struct aq_mss_egress_sa_counters *counters,
+				  u16 sa_index)
+{
+	u16 packed_record[4];
+	int ret;
+
+	if (sa_index >= NUMROWS_EGRESSSARECORD)
+		return -EINVAL;
+
+	ret = get_raw_egress_record(hw, packed_record, 4, 3, sa_index * 8 + 0);
+	if (unlikely(ret))
+		return ret;
+	counters->sa_hit_drop_redirect[0] =
+		packed_record[0] | (packed_record[1] << 16);
+	counters->sa_hit_drop_redirect[1] =
+		packed_record[2] | (packed_record[3] << 16);
+
+	ret = get_raw_egress_record(hw, packed_record, 4, 3, sa_index * 8 + 1);
+	if (unlikely(ret))
+		return ret;
+	counters->sa_protected2_pkts[0] =
+		packed_record[0] | (packed_record[1] << 16);
+	counters->sa_protected2_pkts[1] =
+		packed_record[2] | (packed_record[3] << 16);
+
+	ret = get_raw_egress_record(hw, packed_record, 4, 3, sa_index * 8 + 2);
+	if (unlikely(ret))
+		return ret;
+	counters->sa_protected_pkts[0] =
+		packed_record[0] | (packed_record[1] << 16);
+	counters->sa_protected_pkts[1] =
+		packed_record[2] | (packed_record[3] << 16);
+
+	ret = get_raw_egress_record(hw, packed_record, 4, 3, sa_index * 8 + 3);
+	if (unlikely(ret))
+		return ret;
+	counters->sa_encrypted_pkts[0] =
+		packed_record[0] | (packed_record[1] << 16);
+	counters->sa_encrypted_pkts[1] =
+		packed_record[2] | (packed_record[3] << 16);
+
+	return 0;
+}
+
+int aq_mss_get_egress_sa_counters(struct aq_hw_s *hw,
+				  struct aq_mss_egress_sa_counters *counters,
+				  u16 sa_index)
+{
+	memset(counters, 0, sizeof(*counters));
+
+	return AQ_API_CALL_SAFE(get_egress_sa_counters, hw, counters, sa_index);
+}
+
+static int
+get_egress_common_counters(struct aq_hw_s *hw,
+			   struct aq_mss_egress_common_counters *counters)
+{
+	u16 packed_record[4];
+	int ret;
+
+	ret = get_raw_egress_record(hw, packed_record, 4, 3, 256 + 0);
+	if (unlikely(ret))
+		return ret;
+	counters->ctl_pkt[0] = packed_record[0] | (packed_record[1] << 16);
+	counters->ctl_pkt[1] = packed_record[2] | (packed_record[3] << 16);
+
+	ret = get_raw_egress_record(hw, packed_record, 4, 3, 256 + 1);
+	if (unlikely(ret))
+		return ret;
+	counters->unknown_sa_pkts[0] =
+		packed_record[0] | (packed_record[1] << 16);
+	counters->unknown_sa_pkts[1] =
+		packed_record[2] | (packed_record[3] << 16);
+
+	ret = get_raw_egress_record(hw, packed_record, 4, 3, 256 + 2);
+	if (unlikely(ret))
+		return ret;
+	counters->untagged_pkts[0] =
+		packed_record[0] | (packed_record[1] << 16);
+	counters->untagged_pkts[1] =
+		packed_record[2] | (packed_record[3] << 16);
+
+	ret = get_raw_egress_record(hw, packed_record, 4, 3, 256 + 3);
+	if (unlikely(ret))
+		return ret;
+	counters->too_long[0] = packed_record[0] | (packed_record[1] << 16);
+	counters->too_long[1] = packed_record[2] | (packed_record[3] << 16);
+
+	ret = get_raw_egress_record(hw, packed_record, 4, 3, 256 + 4);
+	if (unlikely(ret))
+		return ret;
+	counters->ecc_error_pkts[0] =
+		packed_record[0] | (packed_record[1] << 16);
+	counters->ecc_error_pkts[1] =
+		packed_record[2] | (packed_record[3] << 16);
+
+	ret = get_raw_egress_record(hw, packed_record, 4, 3, 256 + 5);
+	if (unlikely(ret))
+		return ret;
+	counters->unctrl_hit_drop_redir[0] =
+		packed_record[0] | (packed_record[1] << 16);
+	counters->unctrl_hit_drop_redir[1] =
+		packed_record[2] | (packed_record[3] << 16);
+
+	return 0;
+}
+
+int aq_mss_get_egress_common_counters(struct aq_hw_s *hw,
+	struct aq_mss_egress_common_counters *counters)
+{
+	memset(counters, 0, sizeof(*counters));
+
+	return AQ_API_CALL_SAFE(get_egress_common_counters, hw, counters);
+}
+
+static int clear_egress_counters(struct aq_hw_s *hw)
+{
+	struct mss_egress_ctl_register ctl_reg;
+	int ret;
+
+	memset(&ctl_reg, 0, sizeof(ctl_reg));
+
+	ret = aq_mss_mdio_read(hw, MDIO_MMD_VEND1, MSS_EGRESS_CTL_REGISTER_ADDR,
+			       &ctl_reg.word_0);
+	if (unlikely(ret))
+		return ret;
+	ret = aq_mss_mdio_read(hw, MDIO_MMD_VEND1,
+			       MSS_EGRESS_CTL_REGISTER_ADDR + 4,
+			       &ctl_reg.word_1);
+	if (unlikely(ret))
+		return ret;
+
+	/* Toggle the Egress MIB clear bit 0->1->0 */
+	ctl_reg.bits_0.clear_counter = 0;
+	ret = aq_mss_mdio_write(hw, MDIO_MMD_VEND1,
+				MSS_EGRESS_CTL_REGISTER_ADDR, ctl_reg.word_0);
+	if (unlikely(ret))
+		return ret;
+	ret = aq_mss_mdio_write(hw, MDIO_MMD_VEND1,
+				MSS_EGRESS_CTL_REGISTER_ADDR + 4,
+				ctl_reg.word_1);
+	if (unlikely(ret))
+		return ret;
+
+	ctl_reg.bits_0.clear_counter = 1;
+	ret = aq_mss_mdio_write(hw, MDIO_MMD_VEND1,
+				MSS_EGRESS_CTL_REGISTER_ADDR, ctl_reg.word_0);
+	if (unlikely(ret))
+		return ret;
+	ret = aq_mss_mdio_write(hw, MDIO_MMD_VEND1,
+				MSS_EGRESS_CTL_REGISTER_ADDR + 4,
+				ctl_reg.word_1);
+	if (unlikely(ret))
+		return ret;
+
+	ctl_reg.bits_0.clear_counter = 0;
+	ret = aq_mss_mdio_write(hw, MDIO_MMD_VEND1,
+				MSS_EGRESS_CTL_REGISTER_ADDR, ctl_reg.word_0);
+	if (unlikely(ret))
+		return ret;
+	ret = aq_mss_mdio_write(hw, MDIO_MMD_VEND1,
+				MSS_EGRESS_CTL_REGISTER_ADDR + 4,
+				ctl_reg.word_1);
+	if (unlikely(ret))
+		return ret;
+
+	return 0;
+}
+
+int aq_mss_clear_egress_counters(struct aq_hw_s *hw)
+{
+	return AQ_API_CALL_SAFE(clear_egress_counters, hw);
+}
+
+static int get_ingress_sa_counters(struct aq_hw_s *hw,
+				   struct aq_mss_ingress_sa_counters *counters,
+				   u16 sa_index)
+{
+	u16 packed_record[4];
+	int ret;
+
+	if (sa_index >= NUMROWS_INGRESSSARECORD)
+		return -EINVAL;
+
+	ret = get_raw_ingress_record(hw, packed_record, 4, 6,
+				     sa_index * 12 + 0);
+	if (unlikely(ret))
+		return ret;
+	counters->untagged_hit_pkts[0] =
+		packed_record[0] | (packed_record[1] << 16);
+	counters->untagged_hit_pkts[1] =
+		packed_record[2] | (packed_record[3] << 16);
+
+	ret = get_raw_ingress_record(hw, packed_record, 4, 6,
+				     sa_index * 12 + 1);
+	if (unlikely(ret))
+		return ret;
+	counters->ctrl_hit_drop_redir_pkts[0] =
+		packed_record[0] | (packed_record[1] << 16);
+	counters->ctrl_hit_drop_redir_pkts[1] =
+		packed_record[2] | (packed_record[3] << 16);
+
+	ret = get_raw_ingress_record(hw, packed_record, 4, 6,
+				     sa_index * 12 + 2);
+	if (unlikely(ret))
+		return ret;
+	counters->not_using_sa[0] = packed_record[0] | (packed_record[1] << 16);
+	counters->not_using_sa[1] = packed_record[2] | (packed_record[3] << 16);
+
+	ret = get_raw_ingress_record(hw, packed_record, 4, 6,
+				     sa_index * 12 + 3);
+	if (unlikely(ret))
+		return ret;
+	counters->unused_sa[0] = packed_record[0] | (packed_record[1] << 16);
+	counters->unused_sa[1] = packed_record[2] | (packed_record[3] << 16);
+
+	ret = get_raw_ingress_record(hw, packed_record, 4, 6,
+				     sa_index * 12 + 4);
+	if (unlikely(ret))
+		return ret;
+	counters->not_valid_pkts[0] =
+		packed_record[0] | (packed_record[1] << 16);
+	counters->not_valid_pkts[1] =
+		packed_record[2] | (packed_record[3] << 16);
+
+	ret = get_raw_ingress_record(hw, packed_record, 4, 6,
+				     sa_index * 12 + 5);
+	if (unlikely(ret))
+		return ret;
+	counters->invalid_pkts[0] = packed_record[0] | (packed_record[1] << 16);
+	counters->invalid_pkts[1] = packed_record[2] | (packed_record[3] << 16);
+
+	ret = get_raw_ingress_record(hw, packed_record, 4, 6,
+				     sa_index * 12 + 6);
+	if (unlikely(ret))
+		return ret;
+	counters->ok_pkts[0] = packed_record[0] | (packed_record[1] << 16);
+	counters->ok_pkts[1] = packed_record[2] | (packed_record[3] << 16);
+
+	ret = get_raw_ingress_record(hw, packed_record, 4, 6,
+				     sa_index * 12 + 7);
+	if (unlikely(ret))
+		return ret;
+	counters->late_pkts[0] = packed_record[0] | (packed_record[1] << 16);
+	counters->late_pkts[1] = packed_record[2] | (packed_record[3] << 16);
+
+	ret = get_raw_ingress_record(hw, packed_record, 4, 6,
+				     sa_index * 12 + 8);
+	if (unlikely(ret))
+		return ret;
+	counters->delayed_pkts[0] = packed_record[0] | (packed_record[1] << 16);
+	counters->delayed_pkts[1] = packed_record[2] | (packed_record[3] << 16);
+
+	ret = get_raw_ingress_record(hw, packed_record, 4, 6,
+				     sa_index * 12 + 9);
+	if (unlikely(ret))
+		return ret;
+	counters->unchecked_pkts[0] =
+		packed_record[0] | (packed_record[1] << 16);
+	counters->unchecked_pkts[1] =
+		packed_record[2] | (packed_record[3] << 16);
+
+	ret = get_raw_ingress_record(hw, packed_record, 4, 6,
+				     sa_index * 12 + 10);
+	if (unlikely(ret))
+		return ret;
+	counters->validated_octets[0] =
+		packed_record[0] | (packed_record[1] << 16);
+	counters->validated_octets[1] =
+		packed_record[2] | (packed_record[3] << 16);
+
+	ret = get_raw_ingress_record(hw, packed_record, 4, 6,
+				     sa_index * 12 + 11);
+	if (unlikely(ret))
+		return ret;
+	counters->decrypted_octets[0] =
+		packed_record[0] | (packed_record[1] << 16);
+	counters->decrypted_octets[1] =
+		packed_record[2] | (packed_record[3] << 16);
+
+	return 0;
+}
+
+int aq_mss_get_ingress_sa_counters(struct aq_hw_s *hw,
+				   struct aq_mss_ingress_sa_counters *counters,
+				   u16 sa_index)
+{
+	memset(counters, 0, sizeof(*counters));
+
+	return AQ_API_CALL_SAFE(get_ingress_sa_counters, hw, counters,
+				sa_index);
+}
+
+static int
+get_ingress_common_counters(struct aq_hw_s *hw,
+			    struct aq_mss_ingress_common_counters *counters)
+{
+	u16 packed_record[4];
+	int ret;
+
+	ret = get_raw_ingress_record(hw, packed_record, 4, 6, 385 + 0);
+	if (unlikely(ret))
+		return ret;
+	counters->ctl_pkts[0] = packed_record[0] | (packed_record[1] << 16);
+	counters->ctl_pkts[1] = packed_record[2] | (packed_record[3] << 16);
+
+	ret = get_raw_ingress_record(hw, packed_record, 4, 6, 385 + 1);
+	if (unlikely(ret))
+		return ret;
+	counters->tagged_miss_pkts[0] =
+		packed_record[0] | (packed_record[1] << 16);
+	counters->tagged_miss_pkts[1] =
+		packed_record[2] | (packed_record[3] << 16);
+
+	ret = get_raw_ingress_record(hw, packed_record, 4, 6, 385 + 2);
+	if (unlikely(ret))
+		return ret;
+	counters->untagged_miss_pkts[0] =
+		packed_record[0] | (packed_record[1] << 16);
+	counters->untagged_miss_pkts[1] =
+		packed_record[2] | (packed_record[3] << 16);
+
+	ret = get_raw_ingress_record(hw, packed_record, 4, 6, 385 + 3);
+	if (unlikely(ret))
+		return ret;
+	counters->notag_pkts[0] = packed_record[0] | (packed_record[1] << 16);
+	counters->notag_pkts[1] = packed_record[2] | (packed_record[3] << 16);
+
+	ret = get_raw_ingress_record(hw, packed_record, 4, 6, 385 + 4);
+	if (unlikely(ret))
+		return ret;
+	counters->untagged_pkts[0] =
+		packed_record[0] | (packed_record[1] << 16);
+	counters->untagged_pkts[1] =
+		packed_record[2] | (packed_record[3] << 16);
+
+	ret = get_raw_ingress_record(hw, packed_record, 4, 6, 385 + 5);
+	if (unlikely(ret))
+		return ret;
+	counters->bad_tag_pkts[0] = packed_record[0] | (packed_record[1] << 16);
+	counters->bad_tag_pkts[1] = packed_record[2] | (packed_record[3] << 16);
+
+	ret = get_raw_ingress_record(hw, packed_record, 4, 6, 385 + 6);
+	if (unlikely(ret))
+		return ret;
+	counters->no_sci_pkts[0] = packed_record[0] | (packed_record[1] << 16);
+	counters->no_sci_pkts[1] = packed_record[2] | (packed_record[3] << 16);
+
+	ret = get_raw_ingress_record(hw, packed_record, 4, 6, 385 + 7);
+	if (unlikely(ret))
+		return ret;
+	counters->unknown_sci_pkts[0] =
+		packed_record[0] | (packed_record[1] << 16);
+	counters->unknown_sci_pkts[1] =
+		packed_record[2] | (packed_record[3] << 16);
+
+	ret = get_raw_ingress_record(hw, packed_record, 4, 6, 385 + 8);
+	if (unlikely(ret))
+		return ret;
+	counters->ctrl_prt_pass_pkts[0] =
+		packed_record[0] | (packed_record[1] << 16);
+	counters->ctrl_prt_pass_pkts[1] =
+		packed_record[2] | (packed_record[3] << 16);
+
+	ret = get_raw_ingress_record(hw, packed_record, 4, 6, 385 + 9);
+	if (unlikely(ret))
+		return ret;
+	counters->unctrl_prt_pass_pkts[0] =
+		packed_record[0] | (packed_record[1] << 16);
+	counters->unctrl_prt_pass_pkts[1] =
+		packed_record[2] | (packed_record[3] << 16);
+
+	ret = get_raw_ingress_record(hw, packed_record, 4, 6, 385 + 10);
+	if (unlikely(ret))
+		return ret;
+	counters->ctrl_prt_fail_pkts[0] =
+		packed_record[0] | (packed_record[1] << 16);
+	counters->ctrl_prt_fail_pkts[1] =
+		packed_record[2] | (packed_record[3] << 16);
+
+	ret = get_raw_ingress_record(hw, packed_record, 4, 6, 385 + 11);
+	if (unlikely(ret))
+		return ret;
+	counters->unctrl_prt_fail_pkts[0] =
+		packed_record[0] | (packed_record[1] << 16);
+	counters->unctrl_prt_fail_pkts[1] =
+		packed_record[2] | (packed_record[3] << 16);
+
+	ret = get_raw_ingress_record(hw, packed_record, 4, 6, 385 + 12);
+	if (unlikely(ret))
+		return ret;
+	counters->too_long_pkts[0] =
+		packed_record[0] | (packed_record[1] << 16);
+	counters->too_long_pkts[1] =
+		packed_record[2] | (packed_record[3] << 16);
+
+	ret = get_raw_ingress_record(hw, packed_record, 4, 6, 385 + 13);
+	if (unlikely(ret))
+		return ret;
+	counters->igpoc_ctl_pkts[0] =
+		packed_record[0] | (packed_record[1] << 16);
+	counters->igpoc_ctl_pkts[1] =
+		packed_record[2] | (packed_record[3] << 16);
+
+	ret = get_raw_ingress_record(hw, packed_record, 4, 6, 385 + 14);
+	if (unlikely(ret))
+		return ret;
+	counters->ecc_error_pkts[0] =
+		packed_record[0] | (packed_record[1] << 16);
+	counters->ecc_error_pkts[1] =
+		packed_record[2] | (packed_record[3] << 16);
+
+	ret = get_raw_ingress_record(hw, packed_record, 4, 6, 385 + 15);
+	if (unlikely(ret))
+		return ret;
+	counters->unctrl_hit_drop_redir[0] =
+		packed_record[0] | (packed_record[1] << 16);
+	counters->unctrl_hit_drop_redir[1] =
+		packed_record[2] | (packed_record[3] << 16);
+
+	return 0;
+}
+
+int aq_mss_get_ingress_common_counters(struct aq_hw_s *hw,
+	struct aq_mss_ingress_common_counters *counters)
+{
+	memset(counters, 0, sizeof(*counters));
+
+	return AQ_API_CALL_SAFE(get_ingress_common_counters, hw, counters);
+}
+
+static int clear_ingress_counters(struct aq_hw_s *hw)
+{
+	struct mss_ingress_ctl_register ctl_reg;
+	int ret;
+
+	memset(&ctl_reg, 0, sizeof(ctl_reg));
+
+	ret = aq_mss_mdio_read(hw, MDIO_MMD_VEND1,
+			       MSS_INGRESS_CTL_REGISTER_ADDR, &ctl_reg.word_0);
+	if (unlikely(ret))
+		return ret;
+	ret = aq_mss_mdio_read(hw, MDIO_MMD_VEND1,
+			       MSS_INGRESS_CTL_REGISTER_ADDR + 4,
+			       &ctl_reg.word_1);
+	if (unlikely(ret))
+		return ret;
+
+	/* Toggle the Ingress MIB clear bit 0->1->0 */
+	ctl_reg.bits_0.clear_count = 0;
+	ret = aq_mss_mdio_write(hw, MDIO_MMD_VEND1,
+				MSS_INGRESS_CTL_REGISTER_ADDR, ctl_reg.word_0);
+	if (unlikely(ret))
+		return ret;
+	ret = aq_mss_mdio_write(hw, MDIO_MMD_VEND1,
+				MSS_INGRESS_CTL_REGISTER_ADDR + 4,
+				ctl_reg.word_1);
+	if (unlikely(ret))
+		return ret;
+
+	ctl_reg.bits_0.clear_count = 1;
+	ret = aq_mss_mdio_write(hw, MDIO_MMD_VEND1,
+				MSS_INGRESS_CTL_REGISTER_ADDR, ctl_reg.word_0);
+	if (unlikely(ret))
+		return ret;
+	ret = aq_mss_mdio_write(hw, MDIO_MMD_VEND1,
+				MSS_INGRESS_CTL_REGISTER_ADDR + 4,
+				ctl_reg.word_1);
+	if (unlikely(ret))
+		return ret;
+
+	ctl_reg.bits_0.clear_count = 0;
+	ret = aq_mss_mdio_write(hw, MDIO_MMD_VEND1,
+				MSS_INGRESS_CTL_REGISTER_ADDR, ctl_reg.word_0);
+	if (unlikely(ret))
+		return ret;
+	ret = aq_mss_mdio_write(hw, MDIO_MMD_VEND1,
+				MSS_INGRESS_CTL_REGISTER_ADDR + 4,
+				ctl_reg.word_1);
+	if (unlikely(ret))
+		return ret;
+
+	return 0;
+}
+
+int aq_mss_clear_ingress_counters(struct aq_hw_s *hw)
+{
+	return AQ_API_CALL_SAFE(clear_ingress_counters, hw);
+}
+
+static int get_egress_sa_expired(struct aq_hw_s *hw, u32 *expired)
+{
+	u16 val;
+	int ret;
+
+	ret = aq_mss_mdio_read(hw, MDIO_MMD_VEND1,
+			       MSS_EGRESS_SA_EXPIRED_STATUS_REGISTER_ADDR,
+			       &val);
+	if (unlikely(ret))
+		return ret;
+
+	*expired = val;
+
+	ret = aq_mss_mdio_read(hw, MDIO_MMD_VEND1,
+			       MSS_EGRESS_SA_EXPIRED_STATUS_REGISTER_ADDR + 1,
+			       &val);
+	if (unlikely(ret))
+		return ret;
+
+	*expired |= val << 16;
+
+	return 0;
+}
+
+int aq_mss_get_egress_sa_expired(struct aq_hw_s *hw, u32 *expired)
+{
+	*expired = 0;
+
+	return AQ_API_CALL_SAFE(get_egress_sa_expired, hw, expired);
+}
+
+static int get_egress_sa_threshold_expired(struct aq_hw_s *hw,
+					   u32 *expired)
+{
+	u16 val;
+	int ret;
+
+	ret = aq_mss_mdio_read(hw, MDIO_MMD_VEND1,
+		MSS_EGRESS_SA_THRESHOLD_EXPIRED_STATUS_REGISTER_ADDR, &val);
+	if (unlikely(ret))
+		return ret;
+
+	*expired = val;
+
+	ret = aq_mss_mdio_read(hw, MDIO_MMD_VEND1,
+		MSS_EGRESS_SA_THRESHOLD_EXPIRED_STATUS_REGISTER_ADDR + 1, &val);
+	if (unlikely(ret))
+		return ret;
+
+	*expired |= val << 16;
+
+	return 0;
+}
+
+int aq_mss_get_egress_sa_threshold_expired(struct aq_hw_s *hw,
+					   u32 *expired)
+{
+	*expired = 0;
+
+	return AQ_API_CALL_SAFE(get_egress_sa_threshold_expired, hw, expired);
+}
+
+static int set_egress_sa_expired(struct aq_hw_s *hw, u32 expired)
+{
+	int ret;
+
+	ret = aq_mss_mdio_write(hw, MDIO_MMD_VEND1,
+				MSS_EGRESS_SA_EXPIRED_STATUS_REGISTER_ADDR,
+				expired & 0xFFFF);
+	if (unlikely(ret))
+		return ret;
+
+	ret = aq_mss_mdio_write(hw, MDIO_MMD_VEND1,
+				MSS_EGRESS_SA_EXPIRED_STATUS_REGISTER_ADDR + 1,
+				expired >> 16);
+	if (unlikely(ret))
+		return ret;
+
+	return 0;
+}
+
+int aq_mss_set_egress_sa_expired(struct aq_hw_s *hw, u32 expired)
+{
+	return AQ_API_CALL_SAFE(set_egress_sa_expired, hw, expired);
+}
+
+static int set_egress_sa_threshold_expired(struct aq_hw_s *hw, u32 expired)
+{
+	int ret;
+
+	ret = aq_mss_mdio_write(hw, MDIO_MMD_VEND1,
+		MSS_EGRESS_SA_THRESHOLD_EXPIRED_STATUS_REGISTER_ADDR,
+		expired & 0xFFFF);
+	if (unlikely(ret))
+		return ret;
+
+	ret = aq_mss_mdio_write(hw, MDIO_MMD_VEND1,
+		MSS_EGRESS_SA_THRESHOLD_EXPIRED_STATUS_REGISTER_ADDR + 1,
+		expired >> 16);
+	if (unlikely(ret))
+		return ret;
+
+	return 0;
+}
+
+int aq_mss_set_egress_sa_threshold_expired(struct aq_hw_s *hw, u32 expired)
+{
+	return AQ_API_CALL_SAFE(set_egress_sa_threshold_expired, hw, expired);
+}
diff --git a/drivers/net/ethernet/aquantia/atlantic/macsec/macsec_api.h b/drivers/net/ethernet/aquantia/atlantic/macsec/macsec_api.h
new file mode 100644
index 000000000000..ff03cc462a37
--- /dev/null
+++ b/drivers/net/ethernet/aquantia/atlantic/macsec/macsec_api.h
@@ -0,0 +1,323 @@
+/* SPDX-License-Identifier: GPL-2.0-only */
+/* Atlantic Network Driver
+ * Copyright (C) 2020 Marvell International Ltd.
+ */
+
+#ifndef __MACSEC_API_H__
+#define __MACSEC_API_H__
+
+#include "aq_hw.h"
+#include "macsec_struct.h"
+
+#define NUMROWS_INGRESSPRECTLFRECORD 24
+#define ROWOFFSET_INGRESSPRECTLFRECORD 0
+
+#define NUMROWS_INGRESSPRECLASSRECORD 48
+#define ROWOFFSET_INGRESSPRECLASSRECORD 0
+
+#define NUMROWS_INGRESSPOSTCLASSRECORD 48
+#define ROWOFFSET_INGRESSPOSTCLASSRECORD 0
+
+#define NUMROWS_INGRESSSCRECORD 32
+#define ROWOFFSET_INGRESSSCRECORD 0
+
+#define NUMROWS_INGRESSSARECORD 32
+#define ROWOFFSET_INGRESSSARECORD 32
+
+#define NUMROWS_INGRESSSAKEYRECORD 32
+#define ROWOFFSET_INGRESSSAKEYRECORD 0
+
+#define NUMROWS_INGRESSPOSTCTLFRECORD 24
+#define ROWOFFSET_INGRESSPOSTCTLFRECORD 0
+
+#define NUMROWS_EGRESSCTLFRECORD 24
+#define ROWOFFSET_EGRESSCTLFRECORD 0
+
+#define NUMROWS_EGRESSCLASSRECORD 48
+#define ROWOFFSET_EGRESSCLASSRECORD 0
+
+#define NUMROWS_EGRESSSCRECORD 32
+#define ROWOFFSET_EGRESSSCRECORD 0
+
+#define NUMROWS_EGRESSSARECORD 32
+#define ROWOFFSET_EGRESSSARECORD 32
+
+#define NUMROWS_EGRESSSAKEYRECORD 32
+#define ROWOFFSET_EGRESSSAKEYRECORD 96
+
+/*!  Read the raw table data from the specified row of the Egress CTL
+ *   Filter table, and unpack it into the fields of rec.
+ *  rec - [OUT] The raw table row data will be unpacked into the fields of rec.
+ *  table_index - The table row to read (max 23).
+ */
+int aq_mss_get_egress_ctlf_record(struct aq_hw_s *hw,
+				  struct aq_mss_egress_ctlf_record *rec,
+				  u16 table_index);
+
+/*!  Pack the fields of rec, and write the packed data into the
+ *   specified row of the Egress CTL Filter table.
+ *  rec - [IN] The bitfield values to write to the table row.
+ *  table_index - The table row to write(max 23).
+ */
+int aq_mss_set_egress_ctlf_record(struct aq_hw_s *hw,
+				  const struct aq_mss_egress_ctlf_record *rec,
+				  u16 table_index);
+
+/*!  Read the raw table data from the specified row of the Egress
+ *   Packet Classifier table, and unpack it into the fields of rec.
+ *  rec - [OUT] The raw table row data will be unpacked into the fields of rec.
+ *  table_index - The table row to read (max 47).
+ */
+int aq_mss_get_egress_class_record(struct aq_hw_s *hw,
+				   struct aq_mss_egress_class_record *rec,
+				   u16 table_index);
+
+/*!  Pack the fields of rec, and write the packed data into the
+ *   specified row of the Egress Packet Classifier table.
+ *  rec - [IN] The bitfield values to write to the table row.
+ *  table_index - The table row to write (max 47).
+ */
+int aq_mss_set_egress_class_record(struct aq_hw_s *hw,
+				   const struct aq_mss_egress_class_record *rec,
+				   u16 table_index);
+
+/*!  Read the raw table data from the specified row of the Egress SC
+ *   Lookup table, and unpack it into the fields of rec.
+ *  rec - [OUT] The raw table row data will be unpacked into the fields of rec.
+ *  table_index - The table row to read (max 31).
+ */
+int aq_mss_get_egress_sc_record(struct aq_hw_s *hw,
+				struct aq_mss_egress_sc_record *rec,
+				u16 table_index);
+
+/*!  Pack the fields of rec, and write the packed data into the
+ *   specified row of the Egress SC Lookup table.
+ *  rec - [IN] The bitfield values to write to the table row.
+ *  table_index - The table row to write (max 31).
+ */
+int aq_mss_set_egress_sc_record(struct aq_hw_s *hw,
+				const struct aq_mss_egress_sc_record *rec,
+				u16 table_index);
+
+/*!  Read the raw table data from the specified row of the Egress SA
+ *   Lookup table, and unpack it into the fields of rec.
+ *  rec - [OUT] The raw table row data will be unpacked into the fields of rec.
+ *  table_index - The table row to read (max 31).
+ */
+int aq_mss_get_egress_sa_record(struct aq_hw_s *hw,
+				struct aq_mss_egress_sa_record *rec,
+				u16 table_index);
+
+/*!  Pack the fields of rec, and write the packed data into the
+ *   specified row of the Egress SA Lookup table.
+ *  rec  - [IN] The bitfield values to write to the table row.
+ *  table_index - The table row to write (max 31).
+ */
+int aq_mss_set_egress_sa_record(struct aq_hw_s *hw,
+				const struct aq_mss_egress_sa_record *rec,
+				u16 table_index);
+
+/*!  Read the raw table data from the specified row of the Egress SA
+ *   Key Lookup table, and unpack it into the fields of rec.
+ *  rec - [OUT] The raw table row data will be unpacked into the fields of rec.
+ *  table_index - The table row to read (max 31).
+ */
+int aq_mss_get_egress_sakey_record(struct aq_hw_s *hw,
+				   struct aq_mss_egress_sakey_record *rec,
+				   u16 table_index);
+
+/*!  Pack the fields of rec, and write the packed data into the
+ *   specified row of the Egress SA Key Lookup table.
+ *  rec - [IN] The bitfield values to write to the table row.
+ *  table_index - The table row to write (max 31).
+ */
+int aq_mss_set_egress_sakey_record(struct aq_hw_s *hw,
+				   const struct aq_mss_egress_sakey_record *rec,
+				   u16 table_index);
+
+/*!  Read the raw table data from the specified row of the Ingress
+ *   Pre-MACSec CTL Filter table, and unpack it into the fields of rec.
+ *  rec - [OUT] The raw table row data will be unpacked into the fields of rec.
+ *  table_index - The table row to read (max 23).
+ */
+int aq_mss_get_ingress_prectlf_record(struct aq_hw_s *hw,
+				      struct aq_mss_ingress_prectlf_record *rec,
+				      u16 table_index);
+
+/*!  Pack the fields of rec, and write the packed data into the
+ *   specified row of the Ingress Pre-MACSec CTL Filter table.
+ *  rec - [IN] The bitfield values to write to the table row.
+ *  table_index - The table row to write(max 23).
+ */
+int aq_mss_set_ingress_prectlf_record(struct aq_hw_s *hw,
+	const struct aq_mss_ingress_prectlf_record *rec,
+	u16 table_index);
+
+/*!  Read the raw table data from the specified row of the Ingress
+ *   Pre-MACSec Packet Classifier table, and unpack it into the fields of rec.
+ *  rec - [OUT] The raw table row data will be unpacked into the fields of rec.
+ *  table_index - The table row to read (max 47).
+ */
+int aq_mss_get_ingress_preclass_record(struct aq_hw_s *hw,
+	struct aq_mss_ingress_preclass_record *rec,
+	u16 table_index);
+
+/*!  Pack the fields of rec, and write the packed data into the
+ *   specified row of the Ingress Pre-MACSec Packet Classifier table.
+ *  rec - [IN] The bitfield values to write to the table row.
+ *  table_index - The table row to write(max 47).
+ */
+int aq_mss_set_ingress_preclass_record(struct aq_hw_s *hw,
+	const struct aq_mss_ingress_preclass_record *rec,
+	u16 table_index);
+
+/*!  Read the raw table data from the specified row of the Ingress SC
+ *   Lookup table, and unpack it into the fields of rec.
+ *  rec - [OUT] The raw table row data will be unpacked into the fields of rec.
+ *  table_index - The table row to read (max 31).
+ */
+int aq_mss_get_ingress_sc_record(struct aq_hw_s *hw,
+				 struct aq_mss_ingress_sc_record *rec,
+				 u16 table_index);
+
+/*!  Pack the fields of rec, and write the packed data into the
+ *   specified row of the Ingress SC Lookup table.
+ *  rec - [IN] The bitfield values to write to the table row.
+ *  table_index - The table row to write(max 31).
+ */
+int aq_mss_set_ingress_sc_record(struct aq_hw_s *hw,
+				 const struct aq_mss_ingress_sc_record *rec,
+				 u16 table_index);
+
+/*!  Read the raw table data from the specified row of the Ingress SA
+ *   Lookup table, and unpack it into the fields of rec.
+ *  rec - [OUT] The raw table row data will be unpacked into the fields of rec.
+ *  table_index - The table row to read (max 31).
+ */
+int aq_mss_get_ingress_sa_record(struct aq_hw_s *hw,
+				 struct aq_mss_ingress_sa_record *rec,
+				 u16 table_index);
+
+/*!  Pack the fields of rec, and write the packed data into the
+ *   specified row of the Ingress SA Lookup table.
+ *  rec - [IN] The bitfield values to write to the table row.
+ *  table_index - The table row to write(max 31).
+ */
+int aq_mss_set_ingress_sa_record(struct aq_hw_s *hw,
+				 const struct aq_mss_ingress_sa_record *rec,
+				 u16 table_index);
+
+/*!  Read the raw table data from the specified row of the Ingress SA
+ *   Key Lookup table, and unpack it into the fields of rec.
+ *  rec - [OUT] The raw table row data will be unpacked into the fields of rec.
+ *  table_index - The table row to read (max 31).
+ */
+int aq_mss_get_ingress_sakey_record(struct aq_hw_s *hw,
+				    struct aq_mss_ingress_sakey_record *rec,
+				    u16 table_index);
+
+/*!  Pack the fields of rec, and write the packed data into the
+ *   specified row of the Ingress SA Key Lookup table.
+ *  rec - [IN] The bitfield values to write to the table row.
+ *  table_index - The table row to write(max 31).
+ */
+int aq_mss_set_ingress_sakey_record(struct aq_hw_s *hw,
+	const struct aq_mss_ingress_sakey_record *rec,
+	u16 table_index);
+
+/*!  Read the raw table data from the specified row of the Ingress
+ *   Post-MACSec Packet Classifier table, and unpack it into the
+ *   fields of rec.
+ *  rec - [OUT] The raw table row data will be unpacked into the fields of rec.
+ *  table_index - The table row to read (max 48).
+ */
+int aq_mss_get_ingress_postclass_record(struct aq_hw_s *hw,
+	struct aq_mss_ingress_postclass_record *rec,
+	u16 table_index);
+
+/*!  Pack the fields of rec, and write the packed data into the
+ *   specified row of the Ingress Post-MACSec Packet Classifier table.
+ *  rec - [IN] The bitfield values to write to the table row.
+ *  table_index - The table row to write(max 48).
+ */
+int aq_mss_set_ingress_postclass_record(struct aq_hw_s *hw,
+	const struct aq_mss_ingress_postclass_record *rec,
+	u16 table_index);
+
+/*!  Read the raw table data from the specified row of the Ingress
+ *   Post-MACSec CTL Filter table, and unpack it into the fields of rec.
+ *  rec - [OUT] The raw table row data will be unpacked into the fields of rec.
+ *  table_index - The table row to read (max 23).
+ */
+int aq_mss_get_ingress_postctlf_record(struct aq_hw_s *hw,
+	struct aq_mss_ingress_postctlf_record *rec,
+	u16 table_index);
+
+/*!  Pack the fields of rec, and write the packed data into the
+ *   specified row of the Ingress Post-MACSec CTL Filter table.
+ *  rec - [IN] The bitfield values to write to the table row.
+ *  table_index - The table row to write(max 23).
+ */
+int aq_mss_set_ingress_postctlf_record(struct aq_hw_s *hw,
+	const struct aq_mss_ingress_postctlf_record *rec,
+	u16 table_index);
+
+/*!  Read the counters for the specified SC, and unpack them into the
+ *   fields of counters.
+ *  counters - [OUT] The raw table row data will be unpacked here.
+ *  sc_index - The table row to read (max 31).
+ */
+int aq_mss_get_egress_sc_counters(struct aq_hw_s *hw,
+				  struct aq_mss_egress_sc_counters *counters,
+				  u16 sc_index);
+
+/*!  Read the counters for the specified SA, and unpack them into the
+ *   fields of counters.
+ *  counters - [OUT] The raw table row data will be unpacked here.
+ *  sa_index - The table row to read (max 31).
+ */
+int aq_mss_get_egress_sa_counters(struct aq_hw_s *hw,
+				  struct aq_mss_egress_sa_counters *counters,
+				  u16 sa_index);
+
+/*!  Read the counters for the common egress counters, and unpack them
+ *   into the fields of counters.
+ *  counters - [OUT] The raw table row data will be unpacked here.
+ */
+int aq_mss_get_egress_common_counters(struct aq_hw_s *hw,
+	struct aq_mss_egress_common_counters *counters);
+
+/*!  Clear all Egress counters to 0.*/
+int aq_mss_clear_egress_counters(struct aq_hw_s *hw);
+
+/*!  Read the counters for the specified SA, and unpack them into the
+ *   fields of counters.
+ *  counters - [OUT] The raw table row data will be unpacked here.
+ *  sa_index - The table row to read (max 31).
+ */
+int aq_mss_get_ingress_sa_counters(struct aq_hw_s *hw,
+				   struct aq_mss_ingress_sa_counters *counters,
+				   u16 sa_index);
+
+/*!  Read the counters for the common ingress counters, and unpack them
+ *   into the fields of counters.
+ *  counters - [OUT] The raw table row data will be unpacked here.
+ */
+int aq_mss_get_ingress_common_counters(struct aq_hw_s *hw,
+	struct aq_mss_ingress_common_counters *counters);
+
+/*!  Clear all Ingress counters to 0. */
+int aq_mss_clear_ingress_counters(struct aq_hw_s *hw);
+
+/*!  Get Egress SA expired. */
+int aq_mss_get_egress_sa_expired(struct aq_hw_s *hw, u32 *expired);
+/*!  Get Egress SA threshold expired. */
+int aq_mss_get_egress_sa_threshold_expired(struct aq_hw_s *hw,
+					   u32 *expired);
+/*!  Set Egress SA expired. */
+int aq_mss_set_egress_sa_expired(struct aq_hw_s *hw, u32 expired);
+/*!  Set Egress SA threshold expired. */
+int aq_mss_set_egress_sa_threshold_expired(struct aq_hw_s *hw,
+					   u32 expired);
+
+#endif /* __MACSEC_API_H__ */
diff --git a/drivers/net/ethernet/aquantia/atlantic/macsec/macsec_struct.h b/drivers/net/ethernet/aquantia/atlantic/macsec/macsec_struct.h
new file mode 100644
index 000000000000..c2fda80fe1cc
--- /dev/null
+++ b/drivers/net/ethernet/aquantia/atlantic/macsec/macsec_struct.h
@@ -0,0 +1,914 @@
+/* SPDX-License-Identifier: GPL-2.0-only */
+/* Atlantic Network Driver
+ * Copyright (C) 2020 Marvell International Ltd.
+ */
+
+#ifndef _MACSEC_STRUCT_H_
+#define _MACSEC_STRUCT_H_
+
+/*! Represents the bitfields of a single row in the Egress CTL Filter
+ *  table.
+ */
+struct aq_mss_egress_ctlf_record {
+	/*! This is used to store the 48 bit value used to compare SA, DA or
+	 *  halfDA+half SA value.
+	 */
+	u32 sa_da[2];
+	/*! This is used to store the 16 bit ethertype value used for
+	 *  comparison.
+	 */
+	u32 eth_type;
+	/*! The match mask is per-nibble. 0 means don't care, i.e. every value
+	 *  will match successfully. The total data is 64 bit, i.e. 16 nibbles
+	 *  masks.
+	 */
+	u32 match_mask;
+	/*! 0: No compare, i.e. This entry is not used
+	 *  1: compare DA only
+	 *  2: compare SA only
+	 *  3: compare half DA + half SA
+	 *  4: compare ether type only
+	 *  5: compare DA + ethertype
+	 *  6: compare SA + ethertype
+	 *  7: compare DA+ range.
+	 */
+	u32 match_type;
+	/*! 0: Bypass the remaining modules if matched.
+	 *  1: Forward to next module for more classifications.
+	 */
+	u32 action;
+};
+
+/*! Represents the bitfields of a single row in the Egress Packet
+ *  Classifier table.
+ */
+struct aq_mss_egress_class_record {
+	/*! VLAN ID field. */
+	u32 vlan_id;
+	/*! VLAN UP field. */
+	u32 vlan_up;
+	/*! VLAN Present in the Packet. */
+	u32 vlan_valid;
+	/*! The 8 bit value used to compare with extracted value for byte 3. */
+	u32 byte3;
+	/*! The 8 bit value used to compare with extracted value for byte 2. */
+	u32 byte2;
+	/*! The 8 bit value used to compare with extracted value for byte 1. */
+	u32 byte1;
+	/*! The 8 bit value used to compare with extracted value for byte 0. */
+	u32 byte0;
+	/*! The 8 bit TCI field used to compare with extracted value. */
+	u32 tci;
+	/*! The 64 bit SCI field in the SecTAG. */
+	u32 sci[2];
+	/*! The 16 bit Ethertype (in the clear) field used to compare with
+	 *  extracted value.
+	 */
+	u32 eth_type;
+	/*! This is to specify the 40bit SNAP header if the SNAP header's mask
+	 *  is enabled.
+	 */
+	u32 snap[2];
+	/*! This is to specify the 24bit LLC header if the LLC header's mask is
+	 *  enabled.
+	 */
+	u32 llc;
+	/*! The 48 bit MAC_SA field used to compare with extracted value. */
+	u32 mac_sa[2];
+	/*! The 48 bit MAC_DA field used to compare with extracted value. */
+	u32 mac_da[2];
+	/*! The 32 bit Packet number used to compare with extracted value. */
+	u32 pn;
+	/*! 0~63: byte location used extracted by packets comparator, which
+	 *  can be anything from the first 64 bytes of the MAC packets.
+	 *  This byte location counted from MAC' DA address. i.e. set to 0
+	 *  will point to byte 0 of DA address.
+	 */
+	u32 byte3_location;
+	/*! 0: don't care
+	 *  1: enable comparison of extracted byte pointed by byte 3 location.
+	 */
+	u32 byte3_mask;
+	/*! 0~63: byte location used extracted by packets comparator, which
+	 *  can be anything from the first 64 bytes of the MAC packets.
+	 *  This byte location counted from MAC' DA address. i.e. set to 0
+	 *  will point to byte 0 of DA address.
+	 */
+	u32 byte2_location;
+	/*! 0: don't care
+	 *  1: enable comparison of extracted byte pointed by byte 2 location.
+	 */
+	u32 byte2_mask;
+	/*! 0~63: byte location used extracted by packets comparator, which
+	 *  can be anything from the first 64 bytes of the MAC packets.
+	 *  This byte location counted from MAC' DA address. i.e. set to 0
+	 *  will point to byte 0 of DA address.
+	 */
+	u32 byte1_location;
+	/*! 0: don't care
+	 *  1: enable comparison of extracted byte pointed by byte 1 location.
+	 */
+	u32 byte1_mask;
+	/*! 0~63: byte location used extracted by packets comparator, which
+	 *  can be anything from the first 64 bytes of the MAC packets.
+	 *  This byte location counted from MAC' DA address. i.e. set to 0
+	 *  will point to byte 0 of DA address.
+	 */
+	u32 byte0_location;
+	/*! 0: don't care
+	 *  1: enable comparison of extracted byte pointed by byte 0 location.
+	 */
+	u32 byte0_mask;
+	/*! Mask is per-byte.
+	 *  0: don't care
+	 *  1: enable comparison of extracted VLAN ID field.
+	 */
+	u32 vlan_id_mask;
+	/*! 0: don't care
+	 *  1: enable comparison of extracted VLAN UP field.
+	 */
+	u32 vlan_up_mask;
+	/*! 0: don't care
+	 *  1: enable comparison of extracted VLAN Valid field.
+	 */
+	u32 vlan_valid_mask;
+	/*! This is bit mask to enable comparison the 8 bit TCI field,
+	 *  including the AN field.
+	 *  For explicit SECTAG, AN is hardware controlled. For sending
+	 *  packet w/ explicit SECTAG, rest of the TCI fields are directly
+	 *  from the SECTAG.
+	 */
+	u32 tci_mask;
+	/*! Mask is per-byte.
+	 *  0: don't care
+	 *  1: enable comparison of SCI
+	 *  Note: If this field is not 0, this means the input packet's
+	 *  SECTAG is explicitly tagged and MACSEC module will only update
+	 *  the MSDU.
+	 *  PN number is hardware controlled.
+	 */
+	u32 sci_mask;
+	/*! Mask is per-byte.
+	 *  0: don't care
+	 *  1: enable comparison of Ethertype.
+	 */
+	u32 eth_type_mask;
+	/*! Mask is per-byte.
+	 *  0: don't care and no SNAP header exist.
+	 *  1: compare the SNAP header.
+	 *  If this bit is set to 1, the extracted filed will assume the
+	 *  SNAP header exist as encapsulated in 802.3 (RFC 1042). I.E. the
+	 *  next 5 bytes after the LLC header is SNAP header.
+	 */
+	u32 snap_mask;
+	/*! 0: don't care and no LLC header exist.
+	 *  1: compare the LLC header.
+	 *  If this bit is set to 1, the extracted filed will assume the
+	 *  LLC header exist as encapsulated in 802.3 (RFC 1042). I.E. the
+	 *  next three bytes after the 802.3MAC header is LLC header.
+	 */
+	u32 llc_mask;
+	/*! Mask is per-byte.
+	 *  0: don't care
+	 *  1: enable comparison of MAC_SA.
+	 */
+	u32 sa_mask;
+	/*! Mask is per-byte.
+	 *  0: don't care
+	 *  1: enable comparison of MAC_DA.
+	 */
+	u32 da_mask;
+	/*! Mask is per-byte. */
+	u32 pn_mask;
+	/*! Reserved. This bit should be always 0. */
+	u32 eight02dot2;
+	/*! 1: For explicit sectag case use TCI_SC from table
+	 *  0: use TCI_SC from explicit sectag.
+	 */
+	u32 tci_sc;
+	/*! 1: For explicit sectag case,use TCI_V,ES,SCB,E,C from table
+	 *  0: use TCI_V,ES,SCB,E,C from explicit sectag.
+	 */
+	u32 tci_87543;
+	/*! 1: indicates that incoming packet has explicit sectag. */
+	u32 exp_sectag_en;
+	/*! If packet matches and tagged as controlled-packet, this SC/SA
+	 *  index is used for later SC and SA table lookup.
+	 */
+	u32 sc_idx;
+	/*! This field is used to specify how many SA entries are
+	 *  associated with 1 SC entry.
+	 *  2'b00: 1 SC has 4 SA.
+	 *  SC index is equivalent to {SC_Index[4:2], 1'b0}.
+	 *  SA index is equivalent to {SC_Index[4:2], SC entry's current AN[1:0]
+	 *  2'b10: 1 SC has 2 SA.
+	 *  SC index is equivalent to SC_Index[4:1]
+	 *  SA index is equivalent to {SC_Index[4:1], SC entry's current AN[0]}
+	 *  2'b11: 1 SC has 1 SA. No SC entry exists for the specific SA.
+	 *  SA index is equivalent to SC_Index[4:0]
+	 *  Note: if specified as 2'b11, hardware AN roll over is not
+	 *  supported.
+	 */
+	u32 sc_sa;
+	/*! 0: the packets will be sent to MAC FIFO
+	 *  1: The packets will be sent to Debug/Loopback FIFO.
+	 *  If the above's action is drop, this bit has no meaning.
+	 */
+	u32 debug;
+	/*! 0: forward to remaining modules
+	 *  1: bypass the next encryption modules. This packet is considered
+	 *     un-control packet.
+	 *  2: drop
+	 *  3: Reserved.
+	 */
+	u32 action;
+	/*! 0: Not valid entry. This entry is not used
+	 *  1: valid entry.
+	 */
+	u32 valid;
+};
+
+/*! Represents the bitfields of a single row in the Egress SC Lookup table. */
+struct aq_mss_egress_sc_record {
+	/*! This is to specify when the SC was first used. Set by HW. */
+	u32 start_time;
+	/*! This is to specify when the SC was last used. Set by HW. */
+	u32 stop_time;
+	/*! This is to specify which of the SA entries are used by current HW.
+	 *  Note: This value need to be set by SW after reset.  It will be
+	 *  automatically updated by HW, if AN roll over is enabled.
+	 */
+	u32 curr_an;
+	/*! 0: Clear the SA Valid Bit after PN expiry.
+	 *  1: Do not Clear the SA Valid bit after PN expiry of the current SA.
+	 *  When the Enable AN roll over is set, S/W does not need to
+	 *  program the new SA's and the H/W will automatically roll over
+	 *  between the SA's without session expiry.
+	 *  For normal operation, Enable AN Roll over will be set to '0'
+	 *  and in which case, the SW needs to program the new SA values
+	 *  after the current PN expires.
+	 */
+	u32 an_roll;
+	/*! This is the TCI field used if packet is not explicitly tagged. */
+	u32 tci;
+	/*! This value indicates the offset where the decryption will start.
+	 *  [[Values of 0, 4, 8-50].
+	 */
+	u32 enc_off;
+	/*! 0: Do not protect frames, all the packets will be forwarded
+	 *     unchanged. MIB counter (OutPktsUntagged) will be updated.
+	 *  1: Protect.
+	 */
+	u32 protect;
+	/*! 0: when none of the SA related to SC has inUse set.
+	 *  1: when either of the SA related to the SC has inUse set.
+	 *  This bit is set by HW.
+	 */
+	u32 recv;
+	/*! 0: H/W Clears this bit on the first use.
+	 *  1: SW updates this entry, when programming the SC Table.
+	 */
+	u32 fresh;
+	/*! AES Key size
+	 *  00 - 128bits
+	 *  01 - 192bits
+	 *  10 - 256bits
+	 *  11 - Reserved.
+	 */
+	u32 sak_len;
+	/*! 0: Invalid SC
+	 *  1: Valid SC.
+	 */
+	u32 valid;
+};
+
+/*! Represents the bitfields of a single row in the Egress SA Lookup table. */
+struct aq_mss_egress_sa_record {
+	/*! This is to specify when the SC was first used. Set by HW. */
+	u32 start_time;
+	/*! This is to specify when the SC was last used. Set by HW. */
+	u32 stop_time;
+	/*! This is set by SW and updated by HW to store the Next PN number
+	 *  used for encryption.
+	 */
+	u32 next_pn;
+	/*! The Next_PN number is going to wrapped around from 0xFFFF_FFFF
+	 *  to 0. set by HW.
+	 */
+	u32 sat_pn;
+	/*! 0: This SA is in use.
+	 *  1: This SA is Fresh and set by SW.
+	 */
+	u32 fresh;
+	/*! 0: Invalid SA
+	 *  1: Valid SA.
+	 */
+	u32 valid;
+};
+
+/*! Represents the bitfields of a single row in the Egress SA Key
+ *  Lookup table.
+ */
+struct aq_mss_egress_sakey_record {
+	/*! Key for AES-GCM processing. */
+	u32 key[8];
+};
+
+/*! Represents the bitfields of a single row in the Ingress Pre-MACSec
+ *  CTL Filter table.
+ */
+struct aq_mss_ingress_prectlf_record {
+	/*! This is used to store the 48 bit value used to compare SA, DA
+	 *  or halfDA+half SA value.
+	 */
+	u32 sa_da[2];
+	/*! This is used to store the 16 bit ethertype value used for
+	 *  comparison.
+	 */
+	u32 eth_type;
+	/*! The match mask is per-nibble. 0 means don't care, i.e. every
+	 *  value will match successfully. The total data is 64 bit, i.e.
+	 *  16 nibbles masks.
+	 */
+	u32 match_mask;
+	/*! 0: No compare, i.e. This entry is not used
+	 *  1: compare DA only
+	 *  2: compare SA only
+	 *  3: compare half DA + half SA
+	 *  4: compare ether type only
+	 *  5: compare DA + ethertype
+	 *  6: compare SA + ethertype
+	 *  7: compare DA+ range.
+	 */
+	u32 match_type;
+	/*! 0: Bypass the remaining modules if matched.
+	 *  1: Forward to next module for more classifications.
+	 */
+	u32 action;
+};
+
+/*! Represents the bitfields of a single row in the Ingress Pre-MACSec
+ *  Packet Classifier table.
+ */
+struct aq_mss_ingress_preclass_record {
+	/*! The 64 bit SCI field used to compare with extracted value.
+	 *  Should have SCI value in case TCI[SCI_SEND] == 0. This will be
+	 *  used for ICV calculation.
+	 */
+	u32 sci[2];
+	/*! The 8 bit TCI field used to compare with extracted value. */
+	u32 tci;
+	/*! 8 bit encryption offset. */
+	u32 encr_offset;
+	/*! The 16 bit Ethertype (in the clear) field used to compare with
+	 *  extracted value.
+	 */
+	u32 eth_type;
+	/*! This is to specify the 40bit SNAP header if the SNAP header's
+	 *  mask is enabled.
+	 */
+	u32 snap[2];
+	/*! This is to specify the 24bit LLC header if the LLC header's
+	 *  mask is enabled.
+	 */
+	u32 llc;
+	/*! The 48 bit MAC_SA field used to compare with extracted value. */
+	u32 mac_sa[2];
+	/*! The 48 bit MAC_DA field used to compare with extracted value. */
+	u32 mac_da[2];
+	/*! 0: this is to compare with non-LPBK packet
+	 *  1: this is to compare with LPBK packet.
+	 *  This value is used to compare with a controlled-tag which goes
+	 *  with the packet when looped back from Egress port.
+	 */
+	u32 lpbk_packet;
+	/*! The value of this bit mask will affects how the SC index and SA
+	 *  index created.
+	 *  2'b00: 1 SC has 4 SA.
+	 *    SC index is equivalent to {SC_Index[4:2], 1'b0}.
+	 *    SA index is equivalent to {SC_Index[4:2], SECTAG's AN[1:0]}
+	 *    Here AN bits are not compared.
+	 *  2'b10: 1 SC has 2 SA.
+	 *    SC index is equivalent to SC_Index[4:1]
+	 *    SA index is equivalent to {SC_Index[4:1], SECTAG's AN[0]}
+	 *    Compare AN[1] field only
+	 *  2'b11: 1 SC has 1 SA. No SC entry exists for the specific SA.
+	 *    SA index is equivalent to SC_Index[4:0]
+	 *    AN[1:0] bits are compared.
+	 *    NOTE: This design is to supports different usage of AN. User
+	 *    can either ping-pong buffer 2 SA by using only the AN[0] bit.
+	 *    Or use 4 SA per SC by use AN[1:0] bits. Or even treat each SA
+	 *    as independent. i.e. AN[1:0] is just another matching pointer
+	 *    to select SA.
+	 */
+	u32 an_mask;
+	/*! This is bit mask to enable comparison the upper 6 bits TCI
+	 *  field, which does not include the AN field.
+	 *  0: don't compare
+	 *  1: enable comparison of the bits.
+	 */
+	u32 tci_mask;
+	/*! 0: don't care
+	 *  1: enable comparison of SCI.
+	 */
+	u32 sci_mask;
+	/*! Mask is per-byte.
+	 *  0: don't care
+	 *  1: enable comparison of Ethertype.
+	 */
+	u32 eth_type_mask;
+	/*! Mask is per-byte.
+	 *  0: don't care and no SNAP header exist.
+	 *  1: compare the SNAP header.
+	 *  If this bit is set to 1, the extracted filed will assume the
+	 *  SNAP header exist as encapsulated in 802.3 (RFC 1042). I.E. the
+	 *  next 5 bytes after the LLC header is SNAP header.
+	 */
+	u32 snap_mask;
+	/*! Mask is per-byte.
+	 *  0: don't care and no LLC header exist.
+	 *  1: compare the LLC header.
+	 *  If this bit is set to 1, the extracted filed will assume the
+	 *  LLC header exist as encapsulated in 802.3 (RFC 1042). I.E. the
+	 *  next three bytes after the 802.3MAC header is LLC header.
+	 */
+	u32 llc_mask;
+	/*! Reserved. This bit should be always 0. */
+	u32 _802_2_encapsulate;
+	/*! Mask is per-byte.
+	 *  0: don't care
+	 *  1: enable comparison of MAC_SA.
+	 */
+	u32 sa_mask;
+	/*! Mask is per-byte.
+	 *  0: don't care
+	 *  1: enable comparison of MAC_DA.
+	 */
+	u32 da_mask;
+	/*! 0: don't care
+	 *  1: enable checking if this is loopback packet or not.
+	 */
+	u32 lpbk_mask;
+	/*! If packet matches and tagged as controlled-packet. This SC/SA
+	 *  index is used for later SC and SA table lookup.
+	 */
+	u32 sc_idx;
+	/*! 0: the packets will be sent to MAC FIFO
+	 *  1: The packets will be sent to Debug/Loopback FIFO.
+	 *  If the above's action is drop. This bit has no meaning.
+	 */
+	u32 proc_dest;
+	/*! 0: Process: Forward to next two modules for 802.1AE decryption.
+	 *  1: Process but keep SECTAG: Forward to next two modules for
+	 *     802.1AE decryption but keep the MACSEC header with added error
+	 *     code information. ICV will be stripped for all control packets.
+	 *  2: Bypass: Bypass the next two decryption modules but processed
+	 *     by post-classification.
+	 *  3: Drop: drop this packet and update counts accordingly.
+	 */
+	u32 action;
+	/*! 0: This is a controlled-port packet if matched.
+	 *  1: This is an uncontrolled-port packet if matched.
+	 */
+	u32 ctrl_unctrl;
+	/*! Use the SCI value from the Table if 'SC' bit of the input
+	 *  packet is not present.
+	 */
+	u32 sci_from_table;
+	/*! Reserved. */
+	u32 reserved;
+	/*! 0: Not valid entry. This entry is not used
+	 *  1: valid entry.
+	 */
+	u32 valid;
+};
+
+/*! Represents the bitfields of a single row in the Ingress SC Lookup table. */
+struct aq_mss_ingress_sc_record {
+	/*! This is to specify when the SC was first used. Set by HW. */
+	u32 stop_time;
+	/*! This is to specify when the SC was first used. Set by HW. */
+	u32 start_time;
+	/*! 0: Strict
+	 *  1: Check
+	 *  2: Disabled.
+	 */
+	u32 validate_frames;
+	/*! 1: Replay control enabled.
+	 *  0: replay control disabled.
+	 */
+	u32 replay_protect;
+	/*! This is to specify the window range for anti-replay. Default is 0.
+	 *  0: is strict order enforcement.
+	 */
+	u32 anti_replay_window;
+	/*! 0: when none of the SA related to SC has inUse set.
+	 *  1: when either of the SA related to the SC has inUse set.
+	 *  This bit is set by HW.
+	 */
+	u32 receiving;
+	/*! 0: when hardware processed the SC for the first time, it clears
+	 *     this bit
+	 *  1: This bit is set by SW, when it sets up the SC.
+	 */
+	u32 fresh;
+	/*! 0: The AN number will not automatically roll over if Next_PN is
+	 *     saturated.
+	 *  1: The AN number will automatically roll over if Next_PN is
+	 *     saturated.
+	 *  Rollover is valid only after expiry. Normal roll over between
+	 *  SA's should be normal process.
+	 */
+	u32 an_rol;
+	/*! Reserved. */
+	u32 reserved;
+	/*! 0: Invalid SC
+	 *  1: Valid SC.
+	 */
+	u32 valid;
+};
+
+/*! Represents the bitfields of a single row in the Ingress SA Lookup table. */
+struct aq_mss_ingress_sa_record {
+	/*! This is to specify when the SC was first used. Set by HW. */
+	u32 stop_time;
+	/*! This is to specify when the SC was first used. Set by HW. */
+	u32 start_time;
+	/*! This is updated by HW to store the expected NextPN number for
+	 *  anti-replay.
+	 */
+	u32 next_pn;
+	/*! The Next_PN number is going to wrapped around from 0XFFFF_FFFF
+	 *  to 0. set by HW.
+	 */
+	u32 sat_nextpn;
+	/*! 0: This SA is not yet used.
+	 *  1: This SA is inUse.
+	 */
+	u32 in_use;
+	/*! 0: when hardware processed the SC for the first time, it clears
+	 *     this timer
+	 *  1: This bit is set by SW, when it sets up the SC.
+	 */
+	u32 fresh;
+	/*! Reserved. */
+	u32 reserved;
+	/*! 0: Invalid SA.
+	 *  1: Valid SA.
+	 */
+	u32 valid;
+};
+
+/*! Represents the bitfields of a single row in the Ingress SA Key
+ *  Lookup table.
+ */
+struct aq_mss_ingress_sakey_record {
+	/*! Key for AES-GCM processing. */
+	u32 key[8];
+	/*! AES key size
+	 *  00 - 128bits
+	 *  01 - 192bits
+	 *  10 - 256bits
+	 *  11 - reserved.
+	 */
+	u32 key_len;
+};
+
+/*! Represents the bitfields of a single row in the Ingress Post-
+ *  MACSec Packet Classifier table.
+ */
+struct aq_mss_ingress_postclass_record {
+	/*! The 8 bit value used to compare with extracted value for byte 0. */
+	u32 byte0;
+	/*! The 8 bit value used to compare with extracted value for byte 1. */
+	u32 byte1;
+	/*! The 8 bit value used to compare with extracted value for byte 2. */
+	u32 byte2;
+	/*! The 8 bit value used to compare with extracted value for byte 3. */
+	u32 byte3;
+	/*! Ethertype in the packet. */
+	u32 eth_type;
+	/*! Ether Type value > 1500 (0x5dc). */
+	u32 eth_type_valid;
+	/*! VLAN ID after parsing. */
+	u32 vlan_id;
+	/*! VLAN priority after parsing. */
+	u32 vlan_up;
+	/*! Valid VLAN coding. */
+	u32 vlan_valid;
+	/*! SA index. */
+	u32 sai;
+	/*! SAI hit, i.e. controlled packet. */
+	u32 sai_hit;
+	/*! Mask for payload ethertype field. */
+	u32 eth_type_mask;
+	/*! 0~63: byte location used extracted by packets comparator, which
+	 *  can be anything from the first 64 bytes of the MAC packets.
+	 *  This byte location counted from MAC' DA address. i.e. set to 0
+	 *  will point to byte 0 of DA address.
+	 */
+	u32 byte3_location;
+	/*! Mask for Byte Offset 3. */
+	u32 byte3_mask;
+	/*! 0~63: byte location used extracted by packets comparator, which
+	 *  can be anything from the first 64 bytes of the MAC packets.
+	 *  This byte location counted from MAC' DA address. i.e. set to 0
+	 *  will point to byte 0 of DA address.
+	 */
+	u32 byte2_location;
+	/*! Mask for Byte Offset 2. */
+	u32 byte2_mask;
+	/*! 0~63: byte location used extracted by packets comparator, which
+	 *  can be anything from the first 64 bytes of the MAC packets.
+	 *  This byte location counted from MAC' DA address. i.e. set to 0
+	 *  will point to byte 0 of DA address.
+	 */
+	u32 byte1_location;
+	/*! Mask for Byte Offset 1. */
+	u32 byte1_mask;
+	/*! 0~63: byte location used extracted by packets comparator, which
+	 *  can be anything from the first 64 bytes of the MAC packets.
+	 *  This byte location counted from MAC' DA address. i.e. set to 0
+	 *  will point to byte 0 of DA address.
+	 */
+	u32 byte0_location;
+	/*! Mask for Byte Offset 0. */
+	u32 byte0_mask;
+	/*! Mask for Ethertype valid field. Indicates 802.3 vs. Other. */
+	u32 eth_type_valid_mask;
+	/*! Mask for VLAN ID field. */
+	u32 vlan_id_mask;
+	/*! Mask for VLAN UP field. */
+	u32 vlan_up_mask;
+	/*! Mask for VLAN valid field. */
+	u32 vlan_valid_mask;
+	/*! Mask for SAI. */
+	u32 sai_mask;
+	/*! Mask for SAI_HIT. */
+	u32 sai_hit_mask;
+	/*! Action if only first level matches and second level does not.
+	 *  0: pass
+	 *  1: drop (fail).
+	 */
+	u32 firstlevel_actions;
+	/*! Action if both first and second level matched.
+	 *  0: pass
+	 *  1: drop (fail).
+	 */
+	u32 secondlevel_actions;
+	/*! Reserved. */
+	u32 reserved;
+	/*! 0: Not valid entry. This entry is not used
+	 *  1: valid entry.
+	 */
+	u32 valid;
+};
+
+/*! Represents the bitfields of a single row in the Ingress Post-
+ *  MACSec CTL Filter table.
+ */
+struct aq_mss_ingress_postctlf_record {
+	/*! This is used to store the 48 bit value used to compare SA, DA
+	 *  or halfDA+half SA value.
+	 */
+	u32 sa_da[2];
+	/*! This is used to store the 16 bit ethertype value used for
+	 *  comparison.
+	 */
+	u32 eth_type;
+	/*! The match mask is per-nibble. 0 means don't care, i.e. every
+	 *  value will match successfully. The total data is 64 bit, i.e.
+	 *  16 nibbles masks.
+	 */
+	u32 match_mask;
+	/*! 0: No compare, i.e. This entry is not used
+	 *  1: compare DA only
+	 *  2: compare SA only
+	 *  3: compare half DA + half SA
+	 *  4: compare ether type only
+	 *  5: compare DA + ethertype
+	 *  6: compare SA + ethertype
+	 *  7: compare DA+ range.
+	 */
+	u32 match_type;
+	/*! 0: Bypass the remaining modules if matched.
+	 *  1: Forward to next module for more classifications.
+	 */
+	u32 action;
+};
+
+/*! Represents the Egress MIB counters for a single SC. Counters are
+ *  64 bits, lower 32 bits in field[0].
+ */
+struct aq_mss_egress_sc_counters {
+	/*! The number of integrity protected but not encrypted packets
+	 *  for this transmitting SC.
+	 */
+	u32 sc_protected_pkts[2];
+	/*! The number of integrity protected and encrypted packets for
+	 *  this transmitting SC.
+	 */
+	u32 sc_encrypted_pkts[2];
+	/*! The number of plain text octets that are integrity protected
+	 *  but not encrypted on the transmitting SC.
+	 */
+	u32 sc_protected_octets[2];
+	/*! The number of plain text octets that are integrity protected
+	 *  and encrypted on the transmitting SC.
+	 */
+	u32 sc_encrypted_octets[2];
+};
+
+/*! Represents the Egress MIB counters for a single SA. Counters are
+ *  64 bits, lower 32 bits in field[0].
+ */
+struct aq_mss_egress_sa_counters {
+	/*! The number of dropped packets for this transmitting SA. */
+	u32 sa_hit_drop_redirect[2];
+	/*! TODO */
+	u32 sa_protected2_pkts[2];
+	/*! The number of integrity protected but not encrypted packets
+	 *  for this transmitting SA.
+	 */
+	u32 sa_protected_pkts[2];
+	/*! The number of integrity protected and encrypted packets for
+	 *  this transmitting SA.
+	 */
+	u32 sa_encrypted_pkts[2];
+};
+
+/*! Represents the common Egress MIB counters; the counter not
+ *  associated with a particular SC/SA. Counters are 64 bits, lower 32
+ *  bits in field[0].
+ */
+struct aq_mss_egress_common_counters {
+	/*! The number of transmitted packets classified as MAC_CTL packets. */
+	u32 ctl_pkt[2];
+	/*! The number of transmitted packets that did not match any rows
+	 *  in the Egress Packet Classifier table.
+	 */
+	u32 unknown_sa_pkts[2];
+	/*! The number of transmitted packets where the SC table entry has
+	 *  protect=0 (so packets are forwarded unchanged).
+	 */
+	u32 untagged_pkts[2];
+	/*! The number of transmitted packets discarded because the packet
+	 *  length is greater than the ifMtu of the Common Port interface.
+	 */
+	u32 too_long[2];
+	/*! The number of transmitted packets for which table memory was
+	 *  affected by an ECC error during processing.
+	 */
+	u32 ecc_error_pkts[2];
+	/*! The number of transmitted packets for where the matched row in
+	 *  the Egress Packet Classifier table has action=drop.
+	 */
+	u32 unctrl_hit_drop_redir[2];
+};
+
+/*! Represents the Ingress MIB counters for a single SA. Counters are
+ *  64 bits, lower 32 bits in field[0].
+ */
+struct aq_mss_ingress_sa_counters {
+	/*! For this SA, the number of received packets without a SecTAG. */
+	u32 untagged_hit_pkts[2];
+	/*! For this SA, the number of received packets that were dropped. */
+	u32 ctrl_hit_drop_redir_pkts[2];
+	/*! For this SA which is not currently in use, the number of
+	 *  received packets that have been discarded, and have either the
+	 *  packets encrypted or the matched row in the Ingress SC Lookup
+	 *  table has validate_frames=Strict.
+	 */
+	u32 not_using_sa[2];
+	/*! For this SA which is not currently in use, the number of
+	 *  received, unencrypted, packets with the matched row in the
+	 *  Ingress SC Lookup table has validate_frames!=Strict.
+	 */
+	u32 unused_sa[2];
+	/*! For this SA, the number discarded packets with the condition
+	 *  that the packets are not valid and one of the following
+	 *  conditions are true: either the matched row in the Ingress SC
+	 *  Lookup table has validate_frames=Strict or the packets
+	 *  encrypted.
+	 */
+	u32 not_valid_pkts[2];
+	/*! For this SA, the number of packets with the condition that the
+	 *  packets are not valid and the matched row in the Ingress SC
+	 *  Lookup table has validate_frames=Check.
+	 */
+	u32 invalid_pkts[2];
+	/*! For this SA, the number of validated packets. */
+	u32 ok_pkts[2];
+	/*! For this SC, the number of received packets that have been
+	 *  discarded with the condition: the matched row in the Ingress
+	 *  SC Lookup table has replay_protect=1 and the PN of the packet
+	 *  is lower than the lower bound replay check PN.
+	 */
+	u32 late_pkts[2];
+	/*! For this SA, the number of packets with the condition that the
+	 *  PN of the packets is lower than the lower bound replay
+	 *  protection PN.
+	 */
+	u32 delayed_pkts[2];
+	/*! For this SC, the number of packets with the following condition:
+	 *  - the matched row in the Ingress SC Lookup table has
+	 *    replay_protect=0 or
+	 *  - the matched row in the Ingress SC Lookup table has
+	 *    replay_protect=1 and the packet is not encrypted and the
+	 *    integrity check has failed or
+	 *  - the matched row in the Ingress SC Lookup table has
+	 *    replay_protect=1 and the packet is encrypted and integrity
+	 *    check has failed.
+	 */
+	u32 unchecked_pkts[2];
+	/*! The number of octets of plaintext recovered from received
+	 *  packets that were integrity protected but not encrypted.
+	 */
+	u32 validated_octets[2];
+	/*! The number of octets of plaintext recovered from received
+	 *  packets that were integrity protected and encrypted.
+	 */
+	u32 decrypted_octets[2];
+};
+
+/*! Represents the common Ingress MIB counters; the counter not
+ *  associated with a particular SA. Counters are 64 bits, lower 32
+ *  bits in field[0].
+ */
+struct aq_mss_ingress_common_counters {
+	/*! The number of received packets classified as MAC_CTL packets. */
+	u32 ctl_pkts[2];
+	/*! The number of received packets with the MAC security tag
+	 *  (SecTAG), not matching any rows in the Ingress Pre-MACSec
+	 *  Packet Classifier table.
+	 */
+	u32 tagged_miss_pkts[2];
+	/*! The number of received packets without the MAC security tag
+	 *  (SecTAG), not matching any rows in the Ingress Pre-MACSec
+	 *  Packet Classifier table.
+	 */
+	u32 untagged_miss_pkts[2];
+	/*! The number of received packets discarded without the MAC
+	 *  security tag (SecTAG) and with the matched row in the Ingress
+	 *  SC Lookup table having validate_frames=Strict.
+	 */
+	u32 notag_pkts[2];
+	/*! The number of received packets without the MAC security tag
+	 *  (SecTAG) and with the matched row in the Ingress SC Lookup
+	 *  table having validate_frames!=Strict.
+	 */
+	u32 untagged_pkts[2];
+	/*! The number of received packets discarded with an invalid
+	 *  SecTAG or a zero value PN or an invalid ICV.
+	 */
+	u32 bad_tag_pkts[2];
+	/*! The number of received packets discarded with unknown SCI
+	 *  information with the condition:
+	 *  the matched row in the Ingress SC Lookup table has
+	 *  validate_frames=Strict or the C bit in the SecTAG is set.
+	 */
+	u32 no_sci_pkts[2];
+	/*! The number of received packets with unknown SCI with the condition:
+	 *  The matched row in the Ingress SC Lookup table has
+	 *  validate_frames!=Strict and the C bit in the SecTAG is not set.
+	 */
+	u32 unknown_sci_pkts[2];
+	/*! The number of received packets by the controlled port service
+	 *  that passed the Ingress Post-MACSec Packet Classifier table
+	 *  check.
+	 */
+	u32 ctrl_prt_pass_pkts[2];
+	/*! The number of received packets by the uncontrolled port
+	 *  service that passed the Ingress Post-MACSec Packet Classifier
+	 *  table check.
+	 */
+	u32 unctrl_prt_pass_pkts[2];
+	/*! The number of received packets by the controlled port service
+	 *  that failed the Ingress Post-MACSec Packet Classifier table
+	 *  check.
+	 */
+	u32 ctrl_prt_fail_pkts[2];
+	/*! The number of received packets by the uncontrolled port
+	 *  service that failed the Ingress Post-MACSec Packet Classifier
+	 *  table check.
+	 */
+	u32 unctrl_prt_fail_pkts[2];
+	/*! The number of received packets discarded because the packet
+	 *  length is greater than the ifMtu of the Common Port interface.
+	 */
+	u32 too_long_pkts[2];
+	/*! The number of received packets classified as MAC_CTL by the
+	 *  Ingress Post-MACSec CTL Filter table.
+	 */
+	u32 igpoc_ctl_pkts[2];
+	/*! The number of received packets for which table memory was
+	 *  affected by an ECC error during processing.
+	 */
+	u32 ecc_error_pkts[2];
+	/*! The number of received packets by the uncontrolled port
+	 *  service that were dropped.
+	 */
+	u32 unctrl_hit_drop_redir[2];
+};
+
+#endif
diff --git a/drivers/net/ethernet/aquantia/atlantic/ver.h b/drivers/net/ethernet/aquantia/atlantic/ver.h
deleted file mode 100644
index 0de858d215c2..000000000000
--- a/drivers/net/ethernet/aquantia/atlantic/ver.h
+++ /dev/null
@@ -1,18 +0,0 @@
-/*
- * aQuantia Corporation Network Driver
- * Copyright (C) 2014-2017 aQuantia Corporation. All rights reserved
- *
- * This program is free software; you can redistribute it and/or modify it
- * under the terms and conditions of the GNU General Public License,
- * version 2, as published by the Free Software Foundation.
- */
-
-#ifndef VER_H
-#define VER_H
-
-#define NIC_MAJOR_DRIVER_VERSION           1
-#define NIC_MINOR_DRIVER_VERSION           5
-#define NIC_BUILD_DRIVER_VERSION           345
-#define NIC_REVISION_DRIVER_VERSION        0
-
-#endif /* VER_H */
diff --git a/drivers/net/ethernet/arc/Kconfig b/drivers/net/ethernet/arc/Kconfig
index e743ddf46343..0d400a7d8d91 100644
--- a/drivers/net/ethernet/arc/Kconfig
+++ b/drivers/net/ethernet/arc/Kconfig
@@ -1,3 +1,4 @@
+# SPDX-License-Identifier: GPL-2.0-only
 #
 # ARC EMAC network device configuration
 #
@@ -5,7 +6,7 @@
 config NET_VENDOR_ARC
 	bool "ARC devices"
 	default y
-	---help---
+	help
 	  If you have a network (Ethernet) card belonging to this class, say Y.
 
 	  Note that the answer to this question doesn't directly affect the
@@ -20,21 +21,14 @@ config ARC_EMAC_CORE
 	depends on ARC || ARCH_ROCKCHIP || COMPILE_TEST
 	select MII
 	select PHYLIB
-
-config ARC_EMAC
-	tristate "ARC EMAC support"
-	select ARC_EMAC_CORE
-	depends on OF_IRQ && OF_NET && HAS_DMA && (ARC || COMPILE_TEST)
-	---help---
-	  On some legacy ARC (Synopsys) FPGA boards such as ARCAngel4/ML50x
-	  non-standard on-chip ethernet device ARC EMAC 10/100 is used.
-	  Say Y here if you have such a board.  If unsure, say N.
+	select CRC32
 
 config EMAC_ROCKCHIP
 	tristate "Rockchip EMAC support"
 	select ARC_EMAC_CORE
-	depends on OF_IRQ && OF_NET && REGULATOR && HAS_DMA && (ARCH_ROCKCHIP || COMPILE_TEST)
-	---help---
+	depends on OF_IRQ && REGULATOR
+	depends on ARCH_ROCKCHIP || COMPILE_TEST
+	help
 	  Support for Rockchip RK3036/RK3066/RK3188 EMAC ethernet controllers.
 	  This selects Rockchip SoC glue layer support for the
 	  emac device driver. This driver is used for RK3036/RK3066/RK3188
diff --git a/drivers/net/ethernet/arc/Makefile b/drivers/net/ethernet/arc/Makefile
index 79108af553fb..23586eefec44 100644
--- a/drivers/net/ethernet/arc/Makefile
+++ b/drivers/net/ethernet/arc/Makefile
@@ -1,8 +1,8 @@
+# SPDX-License-Identifier: GPL-2.0-only
 #
 # Makefile for the ARC network device drivers.
 #
 
 arc_emac-objs := emac_main.o emac_mdio.o
 obj-$(CONFIG_ARC_EMAC_CORE) += arc_emac.o
-obj-$(CONFIG_ARC_EMAC) += emac_arc.o
 obj-$(CONFIG_EMAC_ROCKCHIP) += emac_rockchip.o
diff --git a/drivers/net/ethernet/arc/emac.h b/drivers/net/ethernet/arc/emac.h
index e4feb712d4f2..0e244f0e25fd 100644
--- a/drivers/net/ethernet/arc/emac.h
+++ b/drivers/net/ethernet/arc/emac.h
@@ -1,3 +1,4 @@
+/* SPDX-License-Identifier: GPL-2.0 */
 /*
  * Copyright (C) 2004-2013 Synopsys, Inc. (www.synopsys.com)
  *
@@ -129,7 +130,6 @@ struct arc_emac_mdio_bus_data {
  */
 struct arc_emac_priv {
 	const char *drv_name;
-	const char *drv_version;
 	void (*set_mac_speed)(void *priv, unsigned int speed);
 
 	/* Devices */
@@ -158,6 +158,8 @@ struct arc_emac_priv {
 	unsigned int link;
 	unsigned int duplex;
 	unsigned int speed;
+
+	unsigned int rx_missed_errors;
 };
 
 /**
@@ -218,6 +220,6 @@ static inline void arc_reg_clr(struct arc_emac_priv *priv, int reg, int mask)
 int arc_mdio_probe(struct arc_emac_priv *priv);
 int arc_mdio_remove(struct arc_emac_priv *priv);
 int arc_emac_probe(struct net_device *ndev, int interface);
-int arc_emac_remove(struct net_device *ndev);
+void arc_emac_remove(struct net_device *ndev);
 
 #endif /* ARC_EMAC_H */
diff --git a/drivers/net/ethernet/arc/emac_arc.c b/drivers/net/ethernet/arc/emac_arc.c
deleted file mode 100644
index ffd180570920..000000000000
--- a/drivers/net/ethernet/arc/emac_arc.c
+++ /dev/null
@@ -1,96 +0,0 @@
-/**
- * emac_arc.c - ARC EMAC specific glue layer
- *
- * Copyright (C) 2014 Romain Perier
- *
- * Romain Perier  <romain.perier@gmail.com>
- *
- * This program is free software; you can redistribute it and/or modify
- * it under the terms of the GNU General Public License as published by
- * the Free Software Foundation; either version 2 of the License, or
- * (at your option) any later version.
- *
- * This program is distributed in the hope that it will be useful,
- * but WITHOUT ANY WARRANTY; without even the implied warranty of
- * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
- * GNU General Public License for more details.
- */
-
-#include <linux/etherdevice.h>
-#include <linux/module.h>
-#include <linux/of_net.h>
-#include <linux/platform_device.h>
-
-#include "emac.h"
-
-#define DRV_NAME    "emac_arc"
-#define DRV_VERSION "1.0"
-
-static int emac_arc_probe(struct platform_device *pdev)
-{
-	struct device *dev = &pdev->dev;
-	struct net_device *ndev;
-	struct arc_emac_priv *priv;
-	int interface, err;
-
-	if (!dev->of_node)
-		return -ENODEV;
-
-	ndev = alloc_etherdev(sizeof(struct arc_emac_priv));
-	if (!ndev)
-		return -ENOMEM;
-	platform_set_drvdata(pdev, ndev);
-	SET_NETDEV_DEV(ndev, dev);
-
-	priv = netdev_priv(ndev);
-	priv->drv_name = DRV_NAME;
-	priv->drv_version = DRV_VERSION;
-
-	interface = of_get_phy_mode(dev->of_node);
-	if (interface < 0)
-		interface = PHY_INTERFACE_MODE_MII;
-
-	priv->clk = devm_clk_get(dev, "hclk");
-	if (IS_ERR(priv->clk)) {
-		dev_err(dev, "failed to retrieve host clock from device tree\n");
-		err = -EINVAL;
-		goto out_netdev;
-	}
-
-	err = arc_emac_probe(ndev, interface);
-out_netdev:
-	if (err)
-		free_netdev(ndev);
-	return err;
-}
-
-static int emac_arc_remove(struct platform_device *pdev)
-{
-	struct net_device *ndev = platform_get_drvdata(pdev);
-	int err;
-
-	err = arc_emac_remove(ndev);
-	free_netdev(ndev);
-	return err;
-}
-
-static const struct of_device_id emac_arc_dt_ids[] = {
-	{ .compatible = "snps,arc-emac" },
-	{ /* Sentinel */ }
-};
-MODULE_DEVICE_TABLE(of, emac_arc_dt_ids);
-
-static struct platform_driver emac_arc_driver = {
-	.probe = emac_arc_probe,
-	.remove = emac_arc_remove,
-	.driver = {
-		.name = DRV_NAME,
-		.of_match_table  = emac_arc_dt_ids,
-	},
-};
-
-module_platform_driver(emac_arc_driver);
-
-MODULE_AUTHOR("Romain Perier <romain.perier@gmail.com>");
-MODULE_DESCRIPTION("ARC EMAC platform driver");
-MODULE_LICENSE("GPL");
diff --git a/drivers/net/ethernet/arc/emac_main.c b/drivers/net/ethernet/arc/emac_main.c
index 68de2f2652f2..8283aeee35fb 100644
--- a/drivers/net/ethernet/arc/emac_main.c
+++ b/drivers/net/ethernet/arc/emac_main.c
@@ -1,10 +1,7 @@
+// SPDX-License-Identifier: GPL-2.0-only
 /*
  * Copyright (C) 2004-2013 Synopsys, Inc. (www.synopsys.com)
  *
- * This program is free software; you can redistribute it and/or modify
- * it under the terms of the GNU General Public License version 2 as
- * published by the Free Software Foundation.
- *
  * Driver for the ARC EMAC 10100 (hardware revision 5)
  *
  * Contributors:
@@ -18,14 +15,16 @@
 #include <linux/interrupt.h>
 #include <linux/io.h>
 #include <linux/module.h>
+#include <linux/of.h>
 #include <linux/of_address.h>
 #include <linux/of_irq.h>
 #include <linux/of_mdio.h>
 #include <linux/of_net.h>
-#include <linux/of_platform.h>
 
 #include "emac.h"
 
+static void arc_emac_restart(struct net_device *ndev);
+
 /**
  * arc_emac_tx_avail - Return the number of available slots in the tx ring.
  * @priv: Pointer to ARC EMAC private data structure.
@@ -92,8 +91,7 @@ static void arc_emac_get_drvinfo(struct net_device *ndev,
 {
 	struct arc_emac_priv *priv = netdev_priv(ndev);
 
-	strlcpy(info->driver, priv->drv_name, sizeof(info->driver));
-	strlcpy(info->version, priv->drv_version, sizeof(info->version));
+	strscpy(info->driver, priv->drv_name, sizeof(info->driver));
 }
 
 static const struct ethtool_ops arc_emac_ethtool_ops = {
@@ -113,6 +111,7 @@ static void arc_emac_tx_clean(struct net_device *ndev)
 {
 	struct arc_emac_priv *priv = netdev_priv(ndev);
 	struct net_device_stats *stats = &ndev->stats;
+	struct device *dev = ndev->dev.parent;
 	unsigned int i;
 
 	for (i = 0; i < TX_BD_NUM; i++) {
@@ -142,11 +141,11 @@ static void arc_emac_tx_clean(struct net_device *ndev)
 			stats->tx_bytes += skb->len;
 		}
 
-		dma_unmap_single(&ndev->dev, dma_unmap_addr(tx_buff, addr),
+		dma_unmap_single(dev, dma_unmap_addr(tx_buff, addr),
 				 dma_unmap_len(tx_buff, len), DMA_TO_DEVICE);
 
 		/* return the sk_buff to system */
-		dev_kfree_skb_irq(skb);
+		dev_consume_skb_irq(skb);
 
 		txbd->data = 0;
 		txbd->info = 0;
@@ -176,6 +175,7 @@ static void arc_emac_tx_clean(struct net_device *ndev)
 static int arc_emac_rx(struct net_device *ndev, int budget)
 {
 	struct arc_emac_priv *priv = netdev_priv(ndev);
+	struct device *dev = ndev->dev.parent;
 	unsigned int work_done;
 
 	for (work_done = 0; work_done < budget; work_done++) {
@@ -210,39 +210,48 @@ static int arc_emac_rx(struct net_device *ndev, int budget)
 			continue;
 		}
 
-		pktlen = info & LEN_MASK;
-		stats->rx_packets++;
-		stats->rx_bytes += pktlen;
-		skb = rx_buff->skb;
-		skb_put(skb, pktlen);
-		skb->dev = ndev;
-		skb->protocol = eth_type_trans(skb, ndev);
-
-		dma_unmap_single(&ndev->dev, dma_unmap_addr(rx_buff, addr),
-				 dma_unmap_len(rx_buff, len), DMA_FROM_DEVICE);
-
-		/* Prepare the BD for next cycle */
-		rx_buff->skb = netdev_alloc_skb_ip_align(ndev,
-							 EMAC_BUFFER_SIZE);
-		if (unlikely(!rx_buff->skb)) {
+		/* Prepare the BD for next cycle. netif_receive_skb()
+		 * only if new skb was allocated and mapped to avoid holes
+		 * in the RX fifo.
+		 */
+		skb = netdev_alloc_skb_ip_align(ndev, EMAC_BUFFER_SIZE);
+		if (unlikely(!skb)) {
+			if (net_ratelimit())
+				netdev_err(ndev, "cannot allocate skb\n");
+			/* Return ownership to EMAC */
+			rxbd->info = cpu_to_le32(FOR_EMAC | EMAC_BUFFER_SIZE);
 			stats->rx_errors++;
-			/* Because receive_skb is below, increment rx_dropped */
 			stats->rx_dropped++;
 			continue;
 		}
 
-		/* receive_skb only if new skb was allocated to avoid holes */
-		netif_receive_skb(skb);
-
-		addr = dma_map_single(&ndev->dev, (void *)rx_buff->skb->data,
+		addr = dma_map_single(dev, (void *)skb->data,
 				      EMAC_BUFFER_SIZE, DMA_FROM_DEVICE);
-		if (dma_mapping_error(&ndev->dev, addr)) {
+		if (dma_mapping_error(dev, addr)) {
 			if (net_ratelimit())
-				netdev_err(ndev, "cannot dma map\n");
-			dev_kfree_skb(rx_buff->skb);
+				netdev_err(ndev, "cannot map dma buffer\n");
+			dev_kfree_skb(skb);
+			/* Return ownership to EMAC */
+			rxbd->info = cpu_to_le32(FOR_EMAC | EMAC_BUFFER_SIZE);
 			stats->rx_errors++;
+			stats->rx_dropped++;
 			continue;
 		}
+
+		/* unmap previosly mapped skb */
+		dma_unmap_single(dev, dma_unmap_addr(rx_buff, addr),
+				 dma_unmap_len(rx_buff, len), DMA_FROM_DEVICE);
+
+		pktlen = info & LEN_MASK;
+		stats->rx_packets++;
+		stats->rx_bytes += pktlen;
+		skb_put(rx_buff->skb, pktlen);
+		rx_buff->skb->dev = ndev;
+		rx_buff->skb->protocol = eth_type_trans(rx_buff->skb, ndev);
+
+		netif_receive_skb(rx_buff->skb);
+
+		rx_buff->skb = skb;
 		dma_unmap_addr_set(rx_buff, addr, addr);
 		dma_unmap_len_set(rx_buff, len, EMAC_BUFFER_SIZE);
 
@@ -259,6 +268,53 @@ static int arc_emac_rx(struct net_device *ndev, int budget)
 }
 
 /**
+ * arc_emac_rx_miss_handle - handle R_MISS register
+ * @ndev:	Pointer to the net_device structure.
+ */
+static void arc_emac_rx_miss_handle(struct net_device *ndev)
+{
+	struct arc_emac_priv *priv = netdev_priv(ndev);
+	struct net_device_stats *stats = &ndev->stats;
+	unsigned int miss;
+
+	miss = arc_reg_get(priv, R_MISS);
+	if (miss) {
+		stats->rx_errors += miss;
+		stats->rx_missed_errors += miss;
+		priv->rx_missed_errors += miss;
+	}
+}
+
+/**
+ * arc_emac_rx_stall_check - check RX stall
+ * @ndev:	Pointer to the net_device structure.
+ * @budget:	How many BDs requested to process on 1 call.
+ * @work_done:	How many BDs processed
+ *
+ * Under certain conditions EMAC stop reception of incoming packets and
+ * continuously increment R_MISS register instead of saving data into
+ * provided buffer. This function detect that condition and restart
+ * EMAC.
+ */
+static void arc_emac_rx_stall_check(struct net_device *ndev,
+				    int budget, unsigned int work_done)
+{
+	struct arc_emac_priv *priv = netdev_priv(ndev);
+	struct arc_emac_bd *rxbd;
+
+	if (work_done)
+		priv->rx_missed_errors = 0;
+
+	if (priv->rx_missed_errors && budget) {
+		rxbd = &priv->rxbd[priv->last_rx_bd];
+		if (le32_to_cpu(rxbd->info) & FOR_EMAC) {
+			arc_emac_restart(ndev);
+			priv->rx_missed_errors = 0;
+		}
+	}
+}
+
+/**
  * arc_emac_poll - NAPI poll handler.
  * @napi:	Pointer to napi_struct structure.
  * @budget:	How many BDs to process on 1 call.
@@ -272,6 +328,7 @@ static int arc_emac_poll(struct napi_struct *napi, int budget)
 	unsigned int work_done;
 
 	arc_emac_tx_clean(ndev);
+	arc_emac_rx_miss_handle(ndev);
 
 	work_done = arc_emac_rx(ndev, budget);
 	if (work_done < budget) {
@@ -279,6 +336,8 @@ static int arc_emac_poll(struct napi_struct *napi, int budget)
 		arc_reg_or(priv, R_ENABLE, RXINT_MASK | TXINT_MASK);
 	}
 
+	arc_emac_rx_stall_check(ndev, budget, work_done);
+
 	return work_done;
 }
 
@@ -320,6 +379,8 @@ static irqreturn_t arc_emac_intr(int irq, void *dev_instance)
 		if (status & MSER_MASK) {
 			stats->rx_missed_errors += 0x100;
 			stats->rx_errors += 0x100;
+			priv->rx_missed_errors += 0x100;
+			napi_schedule(&priv->napi);
 		}
 
 		if (status & RXCR_MASK) {
@@ -364,12 +425,14 @@ static int arc_emac_open(struct net_device *ndev)
 {
 	struct arc_emac_priv *priv = netdev_priv(ndev);
 	struct phy_device *phy_dev = ndev->phydev;
+	struct device *dev = ndev->dev.parent;
 	int i;
 
 	phy_dev->autoneg = AUTONEG_ENABLE;
 	phy_dev->speed = 0;
 	phy_dev->duplex = 0;
-	phy_dev->advertising &= phy_dev->supported;
+	linkmode_and(phy_dev->advertising, phy_dev->advertising,
+		     phy_dev->supported);
 
 	priv->last_rx_bd = 0;
 
@@ -385,9 +448,9 @@ static int arc_emac_open(struct net_device *ndev)
 		if (unlikely(!rx_buff->skb))
 			return -ENOMEM;
 
-		addr = dma_map_single(&ndev->dev, (void *)rx_buff->skb->data,
+		addr = dma_map_single(dev, (void *)rx_buff->skb->data,
 				      EMAC_BUFFER_SIZE, DMA_FROM_DEVICE);
-		if (dma_mapping_error(&ndev->dev, addr)) {
+		if (dma_mapping_error(dev, addr)) {
 			netdev_err(ndev, "cannot dma map\n");
 			dev_kfree_skb(rx_buff->skb);
 			return -ENOMEM;
@@ -488,6 +551,7 @@ static void arc_emac_set_rx_mode(struct net_device *ndev)
 static void arc_free_tx_queue(struct net_device *ndev)
 {
 	struct arc_emac_priv *priv = netdev_priv(ndev);
+	struct device *dev = ndev->dev.parent;
 	unsigned int i;
 
 	for (i = 0; i < TX_BD_NUM; i++) {
@@ -495,7 +559,7 @@ static void arc_free_tx_queue(struct net_device *ndev)
 		struct buffer_state *tx_buff = &priv->tx_buff[i];
 
 		if (tx_buff->skb) {
-			dma_unmap_single(&ndev->dev,
+			dma_unmap_single(dev,
 					 dma_unmap_addr(tx_buff, addr),
 					 dma_unmap_len(tx_buff, len),
 					 DMA_TO_DEVICE);
@@ -519,6 +583,7 @@ static void arc_free_tx_queue(struct net_device *ndev)
 static void arc_free_rx_queue(struct net_device *ndev)
 {
 	struct arc_emac_priv *priv = netdev_priv(ndev);
+	struct device *dev = ndev->dev.parent;
 	unsigned int i;
 
 	for (i = 0; i < RX_BD_NUM; i++) {
@@ -526,7 +591,7 @@ static void arc_free_rx_queue(struct net_device *ndev)
 		struct buffer_state *rx_buff = &priv->rx_buff[i];
 
 		if (rx_buff->skb) {
-			dma_unmap_single(&ndev->dev,
+			dma_unmap_single(dev,
 					 dma_unmap_addr(rx_buff, addr),
 					 dma_unmap_len(rx_buff, len),
 					 DMA_FROM_DEVICE);
@@ -613,12 +678,13 @@ static struct net_device_stats *arc_emac_stats(struct net_device *ndev)
  *
  * This function is invoked from upper layers to initiate transmission.
  */
-static int arc_emac_tx(struct sk_buff *skb, struct net_device *ndev)
+static netdev_tx_t arc_emac_tx(struct sk_buff *skb, struct net_device *ndev)
 {
 	struct arc_emac_priv *priv = netdev_priv(ndev);
 	unsigned int len, *txbd_curr = &priv->txbd_curr;
 	struct net_device_stats *stats = &ndev->stats;
 	__le32 *info = &priv->txbd[*txbd_curr].info;
+	struct device *dev = ndev->dev.parent;
 	dma_addr_t addr;
 
 	if (skb_padto(skb, ETH_ZLEN))
@@ -632,10 +698,9 @@ static int arc_emac_tx(struct sk_buff *skb, struct net_device *ndev)
 		return NETDEV_TX_BUSY;
 	}
 
-	addr = dma_map_single(&ndev->dev, (void *)skb->data, len,
-			      DMA_TO_DEVICE);
+	addr = dma_map_single(dev, (void *)skb->data, len, DMA_TO_DEVICE);
 
-	if (unlikely(dma_mapping_error(&ndev->dev, addr))) {
+	if (unlikely(dma_mapping_error(dev, addr))) {
 		stats->tx_dropped++;
 		stats->tx_errors++;
 		dev_kfree_skb_any(skb);
@@ -713,13 +778,70 @@ static int arc_emac_set_address(struct net_device *ndev, void *p)
 	if (!is_valid_ether_addr(addr->sa_data))
 		return -EADDRNOTAVAIL;
 
-	memcpy(ndev->dev_addr, addr->sa_data, ndev->addr_len);
+	eth_hw_addr_set(ndev, addr->sa_data);
 
 	arc_emac_set_address_internal(ndev);
 
 	return 0;
 }
 
+/**
+ * arc_emac_restart - Restart EMAC
+ * @ndev:	Pointer to net_device structure.
+ *
+ * This function do hardware reset of EMAC in order to restore
+ * network packets reception.
+ */
+static void arc_emac_restart(struct net_device *ndev)
+{
+	struct arc_emac_priv *priv = netdev_priv(ndev);
+	struct net_device_stats *stats = &ndev->stats;
+	int i;
+
+	if (net_ratelimit())
+		netdev_warn(ndev, "restarting stalled EMAC\n");
+
+	netif_stop_queue(ndev);
+
+	/* Disable interrupts */
+	arc_reg_clr(priv, R_ENABLE, RXINT_MASK | TXINT_MASK | ERR_MASK);
+
+	/* Disable EMAC */
+	arc_reg_clr(priv, R_CTRL, EN_MASK);
+
+	/* Return the sk_buff to system */
+	arc_free_tx_queue(ndev);
+
+	/* Clean Tx BD's */
+	priv->txbd_curr = 0;
+	priv->txbd_dirty = 0;
+	memset(priv->txbd, 0, TX_RING_SZ);
+
+	for (i = 0; i < RX_BD_NUM; i++) {
+		struct arc_emac_bd *rxbd = &priv->rxbd[i];
+		unsigned int info = le32_to_cpu(rxbd->info);
+
+		if (!(info & FOR_EMAC)) {
+			stats->rx_errors++;
+			stats->rx_dropped++;
+		}
+		/* Return ownership to EMAC */
+		rxbd->info = cpu_to_le32(FOR_EMAC | EMAC_BUFFER_SIZE);
+	}
+	priv->last_rx_bd = 0;
+
+	/* Make sure info is visible to EMAC before enable */
+	wmb();
+
+	/* Enable interrupts */
+	arc_reg_set(priv, R_ENABLE, RXINT_MASK | TXINT_MASK | ERR_MASK);
+
+	/* Enable EMAC */
+	arc_reg_or(priv, R_CTRL, EN_MASK);
+
+	netif_start_queue(ndev);
+}
+
 static const struct net_device_ops arc_emac_netdev_ops = {
 	.ndo_open		= arc_emac_open,
 	.ndo_stop		= arc_emac_stop,
@@ -727,6 +849,7 @@ static const struct net_device_ops arc_emac_netdev_ops = {
 	.ndo_set_mac_address	= arc_emac_set_address,
 	.ndo_get_stats		= arc_emac_stats,
 	.ndo_set_rx_mode	= arc_emac_set_rx_mode,
+	.ndo_eth_ioctl		= phy_do_ioctl_running,
 #ifdef CONFIG_NET_POLL_CONTROLLER
 	.ndo_poll_controller	= arc_emac_poll_controller,
 #endif
@@ -739,7 +862,6 @@ int arc_emac_probe(struct net_device *ndev, int interface)
 	struct device_node *phy_node;
 	struct phy_device *phydev = NULL;
 	struct arc_emac_priv *priv;
-	const char *mac_addr;
 	unsigned int id, clock_frequency, irq;
 	int err;
 
@@ -824,11 +946,8 @@ int arc_emac_probe(struct net_device *ndev, int interface)
 	}
 
 	/* Get MAC address from device tree */
-	mac_addr = of_get_mac_address(dev->of_node);
-
-	if (mac_addr)
-		memcpy(ndev->dev_addr, mac_addr, ETH_ALEN);
-	else
+	err = of_get_ethdev_address(dev->of_node, ndev);
+	if (err)
 		eth_hw_addr_random(ndev);
 
 	arc_emac_set_address_internal(ndev);
@@ -867,7 +986,8 @@ int arc_emac_probe(struct net_device *ndev, int interface)
 	dev_info(dev, "connected to %s phy with id 0x%x\n",
 		 phydev->drv->name, phydev->phy_id);
 
-	netif_napi_add(ndev, &priv->napi, arc_emac_poll, ARC_EMAC_NAPI_WEIGHT);
+	netif_napi_add_weight(ndev, &priv->napi, arc_emac_poll,
+			      ARC_EMAC_NAPI_WEIGHT);
 
 	err = register_netdev(ndev);
 	if (err) {
@@ -893,7 +1013,7 @@ out_put_node:
 }
 EXPORT_SYMBOL_GPL(arc_emac_probe);
 
-int arc_emac_remove(struct net_device *ndev)
+void arc_emac_remove(struct net_device *ndev)
 {
 	struct arc_emac_priv *priv = netdev_priv(ndev);
 
@@ -904,8 +1024,6 @@ int arc_emac_remove(struct net_device *ndev)
 
 	if (!IS_ERR(priv->clk))
 		clk_disable_unprepare(priv->clk);
-
-	return 0;
 }
 EXPORT_SYMBOL_GPL(arc_emac_remove);
 
diff --git a/drivers/net/ethernet/arc/emac_mdio.c b/drivers/net/ethernet/arc/emac_mdio.c
index a22403c688c9..078b1a72c161 100644
--- a/drivers/net/ethernet/arc/emac_mdio.c
+++ b/drivers/net/ethernet/arc/emac_mdio.c
@@ -1,3 +1,4 @@
+// SPDX-License-Identifier: GPL-2.0
 /*
  * Copyright (C) 2004-2013 Synopsys, Inc. (www.synopsys.com)
  *
@@ -131,6 +132,8 @@ int arc_mdio_probe(struct arc_emac_priv *priv)
 {
 	struct arc_emac_mdio_bus_data *data = &priv->bus_data;
 	struct device_node *np = priv->dev->of_node;
+	const char *name = "Synopsys MII Bus";
+	struct device_node *mdio_node;
 	struct mii_bus *bus;
 	int error;
 
@@ -141,7 +144,7 @@ int arc_mdio_probe(struct arc_emac_priv *priv)
 	priv->bus = bus;
 	bus->priv = priv;
 	bus->parent = priv->dev;
-	bus->name = "Synopsys MII Bus";
+	bus->name = name;
 	bus->read = &arc_mdio_read;
 	bus->write = &arc_mdio_write;
 	bus->reset = &arc_mdio_reset;
@@ -150,9 +153,9 @@ int arc_mdio_probe(struct arc_emac_priv *priv)
 	data->reset_gpio = devm_gpiod_get_optional(priv->dev, "phy-reset",
 						   GPIOD_OUT_LOW);
 	if (IS_ERR(data->reset_gpio)) {
-		error = PTR_ERR(data->reset_gpio);
-		dev_err(priv->dev, "Failed to request gpio: %d\n", error);
-		return error;
+		mdiobus_free(bus);
+		return dev_err_probe(priv->dev, PTR_ERR(data->reset_gpio),
+				     "Failed to request gpio\n");
 	}
 
 	of_property_read_u32(np, "phy-reset-duration", &data->msec);
@@ -162,11 +165,17 @@ int arc_mdio_probe(struct arc_emac_priv *priv)
 
 	snprintf(bus->id, MII_BUS_ID_SIZE, "%s", bus->name);
 
-	error = of_mdiobus_register(bus, priv->dev->of_node);
+	/* Backwards compatibility for EMAC nodes without MDIO subnode. */
+	mdio_node = of_get_child_by_name(np, "mdio");
+	if (!mdio_node)
+		mdio_node = of_node_get(np);
+
+	error = of_mdiobus_register(bus, mdio_node);
+	of_node_put(mdio_node);
 	if (error) {
-		dev_err(priv->dev, "cannot register MDIO bus %s\n", bus->name);
 		mdiobus_free(bus);
-		return error;
+		return dev_err_probe(priv->dev, error,
+				     "cannot register MDIO bus %s\n", name);
 	}
 
 	return 0;
diff --git a/drivers/net/ethernet/arc/emac_rockchip.c b/drivers/net/ethernet/arc/emac_rockchip.c
index e278e3d96ee0..780e70ea1c22 100644
--- a/drivers/net/ethernet/arc/emac_rockchip.c
+++ b/drivers/net/ethernet/arc/emac_rockchip.c
@@ -1,17 +1,8 @@
-/**
+// SPDX-License-Identifier: GPL-2.0-or-later
+/*
  * emac-rockchip.c - Rockchip EMAC specific glue layer
  *
  * Copyright (C) 2014 Romain Perier <romain.perier@gmail.com>
- *
- * This program is free software; you can redistribute it and/or modify
- * it under the terms of the GNU General Public License as published by
- * the Free Software Foundation; either version 2 of the License, or
- * (at your option) any later version.
- *
- * This program is distributed in the hope that it will be useful,
- * but WITHOUT ANY WARRANTY; without even the implied warranty of
- * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
- * GNU General Public License for more details.
  */
 
 #include <linux/etherdevice.h>
@@ -25,7 +16,6 @@
 #include "emac.h"
 
 #define DRV_NAME        "rockchip_emac"
-#define DRV_VERSION     "1.1"
 
 struct emac_rockchip_soc_data {
 	unsigned int grf_offset;
@@ -106,8 +96,9 @@ static int emac_rockchip_probe(struct platform_device *pdev)
 	struct net_device *ndev;
 	struct rockchip_priv_data *priv;
 	const struct of_device_id *match;
+	phy_interface_t interface;
 	u32 data;
-	int err, interface;
+	int err;
 
 	if (!pdev->dev.of_node)
 		return -ENODEV;
@@ -120,10 +111,11 @@ static int emac_rockchip_probe(struct platform_device *pdev)
 
 	priv = netdev_priv(ndev);
 	priv->emac.drv_name = DRV_NAME;
-	priv->emac.drv_version = DRV_VERSION;
 	priv->emac.set_mac_speed = emac_rockchip_set_mac_speed;
 
-	interface = of_get_phy_mode(dev->of_node);
+	err = of_get_phy_mode(dev->of_node, &interface);
+	if (err)
+		goto out_netdev;
 
 	/* RK3036/RK3066/RK3188 SoCs only support RMII */
 	if (interface != PHY_INTERFACE_MODE_RMII) {
@@ -169,8 +161,10 @@ static int emac_rockchip_probe(struct platform_device *pdev)
 	/* Optional regulator for PHY */
 	priv->regulator = devm_regulator_get_optional(dev, "phy");
 	if (IS_ERR(priv->regulator)) {
-		if (PTR_ERR(priv->regulator) == -EPROBE_DEFER)
-			return -EPROBE_DEFER;
+		if (PTR_ERR(priv->regulator) == -EPROBE_DEFER) {
+			err = -EPROBE_DEFER;
+			goto out_clk_disable;
+		}
 		dev_err(dev, "no regulator found\n");
 		priv->regulator = NULL;
 	}
@@ -199,9 +193,11 @@ static int emac_rockchip_probe(struct platform_device *pdev)
 
 	/* RMII interface needs always a rate of 50MHz */
 	err = clk_set_rate(priv->refclk, 50000000);
-	if (err)
+	if (err) {
 		dev_err(dev,
 			"failed to change reference clock rate (%d)\n", err);
+		goto out_regulator_disable;
+	}
 
 	if (priv->soc_data->need_div_macclk) {
 		priv->macclk = devm_clk_get(dev, "macclk");
@@ -220,19 +216,24 @@ static int emac_rockchip_probe(struct platform_device *pdev)
 
 		/* RMII TX/RX needs always a rate of 25MHz */
 		err = clk_set_rate(priv->macclk, 25000000);
-		if (err)
+		if (err) {
 			dev_err(dev,
 				"failed to change mac clock rate (%d)\n", err);
+			goto out_clk_disable_macclk;
+		}
 	}
 
 	err = arc_emac_probe(ndev, interface);
 	if (err) {
 		dev_err(dev, "failed to probe arc emac (%d)\n", err);
-		goto out_regulator_disable;
+		goto out_clk_disable_macclk;
 	}
 
 	return 0;
 
+out_clk_disable_macclk:
+	if (priv->soc_data->need_div_macclk)
+		clk_disable_unprepare(priv->macclk);
 out_regulator_disable:
 	if (priv->regulator)
 		regulator_disable(priv->regulator);
@@ -243,21 +244,22 @@ out_netdev:
 	return err;
 }
 
-static int emac_rockchip_remove(struct platform_device *pdev)
+static void emac_rockchip_remove(struct platform_device *pdev)
 {
 	struct net_device *ndev = platform_get_drvdata(pdev);
 	struct rockchip_priv_data *priv = netdev_priv(ndev);
-	int err;
 
-	err = arc_emac_remove(ndev);
+	arc_emac_remove(ndev);
 
 	clk_disable_unprepare(priv->refclk);
 
 	if (priv->regulator)
 		regulator_disable(priv->regulator);
 
+	if (priv->soc_data->need_div_macclk)
+		clk_disable_unprepare(priv->macclk);
+
 	free_netdev(ndev);
-	return err;
 }
 
 static struct platform_driver emac_rockchip_driver = {
diff --git a/drivers/net/ethernet/asix/Kconfig b/drivers/net/ethernet/asix/Kconfig
new file mode 100644
index 000000000000..eed02453314c
--- /dev/null
+++ b/drivers/net/ethernet/asix/Kconfig
@@ -0,0 +1,35 @@
+#
+# Asix network device configuration
+#
+
+config NET_VENDOR_ASIX
+	bool "Asix devices"
+	default y
+	help
+	  If you have a network (Ethernet, non-USB, not NE2000 compatible)
+	  interface based on a chip from ASIX, say Y.
+
+if NET_VENDOR_ASIX
+
+config SPI_AX88796C
+	tristate "Asix AX88796C-SPI support"
+	select PHYLIB
+	depends on SPI
+	depends on GPIOLIB
+	help
+	  Say Y here if you intend to use ASIX AX88796C attached in SPI mode.
+
+config SPI_AX88796C_COMPRESSION
+	bool "SPI transfer compression"
+	default n
+	depends on SPI_AX88796C
+	help
+	  Say Y here to enable SPI transfer compression. It saves up
+	  to 24 dummy cycles during each transfer which may noticeably
+	  speed up short transfers. This sets the default value that is
+	  inherited by network interfaces during probe. It can be
+	  changed at run time via spi-compression ethtool tunable.
+
+	  If unsure say N.
+
+endif # NET_VENDOR_ASIX
diff --git a/drivers/net/ethernet/asix/Makefile b/drivers/net/ethernet/asix/Makefile
new file mode 100644
index 000000000000..0bfbbb042634
--- /dev/null
+++ b/drivers/net/ethernet/asix/Makefile
@@ -0,0 +1,6 @@
+#
+# Makefile for the Asix network device drivers.
+#
+
+obj-$(CONFIG_SPI_AX88796C) += ax88796c.o
+ax88796c-y := ax88796c_main.o ax88796c_ioctl.o ax88796c_spi.o
diff --git a/drivers/net/ethernet/asix/ax88796c_ioctl.c b/drivers/net/ethernet/asix/ax88796c_ioctl.c
new file mode 100644
index 000000000000..7d2fe2e5af92
--- /dev/null
+++ b/drivers/net/ethernet/asix/ax88796c_ioctl.c
@@ -0,0 +1,239 @@
+// SPDX-License-Identifier: GPL-2.0-only
+/*
+ * Copyright (c) 2010 ASIX Electronics Corporation
+ * Copyright (c) 2020 Samsung Electronics Co., Ltd.
+ *
+ * ASIX AX88796C SPI Fast Ethernet Linux driver
+ */
+
+#define pr_fmt(fmt)	"ax88796c: " fmt
+
+#include <linux/bitmap.h>
+#include <linux/iopoll.h>
+#include <linux/phy.h>
+#include <linux/netdevice.h>
+
+#include "ax88796c_main.h"
+#include "ax88796c_ioctl.h"
+
+static const char ax88796c_priv_flag_names[][ETH_GSTRING_LEN] = {
+	"SPICompression",
+};
+
+static void
+ax88796c_get_drvinfo(struct net_device *ndev, struct ethtool_drvinfo *info)
+{
+	/* Inherit standard device info */
+	strscpy(info->driver, DRV_NAME, sizeof(info->driver));
+}
+
+static u32 ax88796c_get_msglevel(struct net_device *ndev)
+{
+	struct ax88796c_device *ax_local = to_ax88796c_device(ndev);
+
+	return ax_local->msg_enable;
+}
+
+static void ax88796c_set_msglevel(struct net_device *ndev, u32 level)
+{
+	struct ax88796c_device *ax_local = to_ax88796c_device(ndev);
+
+	ax_local->msg_enable = level;
+}
+
+static void
+ax88796c_get_pauseparam(struct net_device *ndev, struct ethtool_pauseparam *pause)
+{
+	struct ax88796c_device *ax_local = to_ax88796c_device(ndev);
+
+	pause->tx_pause = !!(ax_local->flowctrl & AX_FC_TX);
+	pause->rx_pause = !!(ax_local->flowctrl & AX_FC_RX);
+	pause->autoneg = (ax_local->flowctrl & AX_FC_ANEG) ?
+		AUTONEG_ENABLE :
+		AUTONEG_DISABLE;
+}
+
+static int
+ax88796c_set_pauseparam(struct net_device *ndev, struct ethtool_pauseparam *pause)
+{
+	struct ax88796c_device *ax_local = to_ax88796c_device(ndev);
+	int fc;
+
+	/* The following logic comes from phylink_ethtool_set_pauseparam() */
+	fc = pause->tx_pause ? AX_FC_TX : 0;
+	fc |= pause->rx_pause ? AX_FC_RX : 0;
+	fc |= pause->autoneg ? AX_FC_ANEG : 0;
+
+	ax_local->flowctrl = fc;
+
+	if (pause->autoneg) {
+		phy_set_asym_pause(ax_local->phydev, pause->tx_pause,
+				   pause->rx_pause);
+	} else {
+		int maccr = 0;
+
+		phy_set_asym_pause(ax_local->phydev, 0, 0);
+		maccr |= (ax_local->flowctrl & AX_FC_RX) ? MACCR_RXFC_ENABLE : 0;
+		maccr |= (ax_local->flowctrl & AX_FC_TX) ? MACCR_TXFC_ENABLE : 0;
+
+		mutex_lock(&ax_local->spi_lock);
+
+		maccr |= AX_READ(&ax_local->ax_spi, P0_MACCR) &
+			~(MACCR_TXFC_ENABLE | MACCR_RXFC_ENABLE);
+		AX_WRITE(&ax_local->ax_spi, maccr, P0_MACCR);
+
+		mutex_unlock(&ax_local->spi_lock);
+	}
+
+	return 0;
+}
+
+static int ax88796c_get_regs_len(struct net_device *ndev)
+{
+	return AX88796C_REGDUMP_LEN + AX88796C_PHY_REGDUMP_LEN;
+}
+
+static void
+ax88796c_get_regs(struct net_device *ndev, struct ethtool_regs *regs, void *_p)
+{
+	struct ax88796c_device *ax_local = to_ax88796c_device(ndev);
+	int offset, i;
+	u16 *p = _p;
+
+	memset(p, 0, ax88796c_get_regs_len(ndev));
+
+	mutex_lock(&ax_local->spi_lock);
+
+	for (offset = 0; offset < AX88796C_REGDUMP_LEN; offset += 2) {
+		if (!test_bit(offset / 2, ax88796c_no_regs_mask))
+			*p = AX_READ(&ax_local->ax_spi, offset);
+		p++;
+	}
+
+	mutex_unlock(&ax_local->spi_lock);
+
+	for (i = 0; i < AX88796C_PHY_REGDUMP_LEN / 2; i++) {
+		*p = phy_read(ax_local->phydev, i);
+		p++;
+	}
+}
+
+static void
+ax88796c_get_strings(struct net_device *ndev, u32 stringset, u8 *data)
+{
+	switch (stringset) {
+	case ETH_SS_PRIV_FLAGS:
+		memcpy(data, ax88796c_priv_flag_names,
+		       sizeof(ax88796c_priv_flag_names));
+		break;
+	}
+}
+
+static int
+ax88796c_get_sset_count(struct net_device *ndev, int stringset)
+{
+	int ret = 0;
+
+	switch (stringset) {
+	case ETH_SS_PRIV_FLAGS:
+		ret = ARRAY_SIZE(ax88796c_priv_flag_names);
+		break;
+	default:
+		ret = -EOPNOTSUPP;
+	}
+
+	return ret;
+}
+
+static int ax88796c_set_priv_flags(struct net_device *ndev, u32 flags)
+{
+	struct ax88796c_device *ax_local = to_ax88796c_device(ndev);
+
+	if (flags & ~AX_PRIV_FLAGS_MASK)
+		return -EOPNOTSUPP;
+
+	if ((ax_local->priv_flags ^ flags) & AX_CAP_COMP)
+		if (netif_running(ndev))
+			return -EBUSY;
+
+	ax_local->priv_flags = flags;
+
+	return 0;
+}
+
+static u32 ax88796c_get_priv_flags(struct net_device *ndev)
+{
+	struct ax88796c_device *ax_local = to_ax88796c_device(ndev);
+
+	return ax_local->priv_flags;
+}
+
+int ax88796c_mdio_read(struct mii_bus *mdiobus, int phy_id, int loc)
+{
+	struct ax88796c_device *ax_local = mdiobus->priv;
+	int ret;
+
+	mutex_lock(&ax_local->spi_lock);
+	AX_WRITE(&ax_local->ax_spi, MDIOCR_RADDR(loc)
+			| MDIOCR_FADDR(phy_id) | MDIOCR_READ, P2_MDIOCR);
+
+	ret = read_poll_timeout(AX_READ, ret,
+				(ret != 0),
+				0, jiffies_to_usecs(HZ / 100), false,
+				&ax_local->ax_spi, P2_MDIOCR);
+	if (!ret)
+		ret = AX_READ(&ax_local->ax_spi, P2_MDIODR);
+
+	mutex_unlock(&ax_local->spi_lock);
+
+	return ret;
+}
+
+int
+ax88796c_mdio_write(struct mii_bus *mdiobus, int phy_id, int loc, u16 val)
+{
+	struct ax88796c_device *ax_local = mdiobus->priv;
+	int ret;
+
+	mutex_lock(&ax_local->spi_lock);
+	AX_WRITE(&ax_local->ax_spi, val, P2_MDIODR);
+
+	AX_WRITE(&ax_local->ax_spi,
+		 MDIOCR_RADDR(loc) | MDIOCR_FADDR(phy_id)
+		 | MDIOCR_WRITE, P2_MDIOCR);
+
+	ret = read_poll_timeout(AX_READ, ret,
+				((ret & MDIOCR_VALID) != 0), 0,
+				jiffies_to_usecs(HZ / 100), false,
+				&ax_local->ax_spi, P2_MDIOCR);
+	mutex_unlock(&ax_local->spi_lock);
+
+	return ret;
+}
+
+const struct ethtool_ops ax88796c_ethtool_ops = {
+	.get_drvinfo		= ax88796c_get_drvinfo,
+	.get_link		= ethtool_op_get_link,
+	.get_msglevel		= ax88796c_get_msglevel,
+	.set_msglevel		= ax88796c_set_msglevel,
+	.get_link_ksettings	= phy_ethtool_get_link_ksettings,
+	.set_link_ksettings	= phy_ethtool_set_link_ksettings,
+	.nway_reset		= phy_ethtool_nway_reset,
+	.get_pauseparam		= ax88796c_get_pauseparam,
+	.set_pauseparam		= ax88796c_set_pauseparam,
+	.get_regs_len		= ax88796c_get_regs_len,
+	.get_regs		= ax88796c_get_regs,
+	.get_strings		= ax88796c_get_strings,
+	.get_sset_count		= ax88796c_get_sset_count,
+	.get_priv_flags		= ax88796c_get_priv_flags,
+	.set_priv_flags		= ax88796c_set_priv_flags,
+};
+
+int ax88796c_ioctl(struct net_device *ndev, struct ifreq *ifr, int cmd)
+{
+	int ret;
+
+	ret = phy_mii_ioctl(ndev->phydev, ifr, cmd);
+
+	return ret;
+}
diff --git a/drivers/net/ethernet/asix/ax88796c_ioctl.h b/drivers/net/ethernet/asix/ax88796c_ioctl.h
new file mode 100644
index 000000000000..34d2a7dcc5ef
--- /dev/null
+++ b/drivers/net/ethernet/asix/ax88796c_ioctl.h
@@ -0,0 +1,26 @@
+/* SPDX-License-Identifier: GPL-2.0-only */
+/*
+ * Copyright (c) 2010 ASIX Electronics Corporation
+ * Copyright (c) 2020 Samsung Electronics Co., Ltd.
+ *
+ * ASIX AX88796C SPI Fast Ethernet Linux driver
+ */
+
+#ifndef _AX88796C_IOCTL_H
+#define _AX88796C_IOCTL_H
+
+#include <linux/ethtool.h>
+#include <linux/netdevice.h>
+
+#include "ax88796c_main.h"
+
+extern const struct ethtool_ops ax88796c_ethtool_ops;
+
+bool ax88796c_check_power(const struct ax88796c_device *ax_local);
+bool ax88796c_check_power_and_wake(struct ax88796c_device *ax_local);
+void ax88796c_set_power_saving(struct ax88796c_device *ax_local, u8 ps_level);
+int ax88796c_mdio_read(struct mii_bus *mdiobus, int phy_id, int loc);
+int ax88796c_mdio_write(struct mii_bus *mdiobus, int phy_id, int loc, u16 val);
+int ax88796c_ioctl(struct net_device *dev, struct ifreq *ifr, int cmd);
+
+#endif
diff --git a/drivers/net/ethernet/asix/ax88796c_main.c b/drivers/net/ethernet/asix/ax88796c_main.c
new file mode 100644
index 000000000000..11e8996b33d7
--- /dev/null
+++ b/drivers/net/ethernet/asix/ax88796c_main.c
@@ -0,0 +1,1166 @@
+// SPDX-License-Identifier: GPL-2.0-only
+/*
+ * Copyright (c) 2010 ASIX Electronics Corporation
+ * Copyright (c) 2020 Samsung Electronics Co., Ltd.
+ *
+ * ASIX AX88796C SPI Fast Ethernet Linux driver
+ */
+
+#define pr_fmt(fmt)	"ax88796c: " fmt
+
+#include "ax88796c_main.h"
+#include "ax88796c_ioctl.h"
+
+#include <linux/bitmap.h>
+#include <linux/etherdevice.h>
+#include <linux/iopoll.h>
+#include <linux/lockdep.h>
+#include <linux/mdio.h>
+#include <linux/minmax.h>
+#include <linux/module.h>
+#include <linux/netdevice.h>
+#include <linux/of.h>
+#include <linux/phy.h>
+#include <linux/skbuff.h>
+#include <linux/spi/spi.h>
+
+static int comp = IS_ENABLED(CONFIG_SPI_AX88796C_COMPRESSION);
+static int msg_enable = NETIF_MSG_PROBE |
+			NETIF_MSG_LINK |
+			NETIF_MSG_RX_ERR |
+			NETIF_MSG_TX_ERR;
+
+static const char *no_regs_list = "80018001,e1918001,8001a001,fc0d0000";
+unsigned long ax88796c_no_regs_mask[AX88796C_REGDUMP_LEN / (sizeof(unsigned long) * 8)];
+
+module_param(msg_enable, int, 0444);
+MODULE_PARM_DESC(msg_enable, "Message mask (see linux/netdevice.h for bitmap)");
+
+static int ax88796c_soft_reset(struct ax88796c_device *ax_local)
+{
+	u16 temp;
+	int ret;
+
+	lockdep_assert_held(&ax_local->spi_lock);
+
+	AX_WRITE(&ax_local->ax_spi, PSR_RESET, P0_PSR);
+	AX_WRITE(&ax_local->ax_spi, PSR_RESET_CLR, P0_PSR);
+
+	ret = read_poll_timeout(AX_READ, ret,
+				(ret & PSR_DEV_READY),
+				0, jiffies_to_usecs(160 * HZ / 1000), false,
+				&ax_local->ax_spi, P0_PSR);
+	if (ret)
+		return ret;
+
+	temp = AX_READ(&ax_local->ax_spi, P4_SPICR);
+	if (ax_local->priv_flags & AX_CAP_COMP) {
+		AX_WRITE(&ax_local->ax_spi,
+			 (temp | SPICR_RCEN | SPICR_QCEN), P4_SPICR);
+		ax_local->ax_spi.comp = 1;
+	} else {
+		AX_WRITE(&ax_local->ax_spi,
+			 (temp & ~(SPICR_RCEN | SPICR_QCEN)), P4_SPICR);
+		ax_local->ax_spi.comp = 0;
+	}
+
+	return 0;
+}
+
+static int ax88796c_reload_eeprom(struct ax88796c_device *ax_local)
+{
+	int ret;
+
+	lockdep_assert_held(&ax_local->spi_lock);
+
+	AX_WRITE(&ax_local->ax_spi, EECR_RELOAD, P3_EECR);
+
+	ret = read_poll_timeout(AX_READ, ret,
+				(ret & PSR_DEV_READY),
+				0, jiffies_to_usecs(2 * HZ / 1000), false,
+				&ax_local->ax_spi, P0_PSR);
+	if (ret) {
+		dev_err(&ax_local->spi->dev,
+			"timeout waiting for reload eeprom\n");
+		return ret;
+	}
+
+	return 0;
+}
+
+static void ax88796c_set_hw_multicast(struct net_device *ndev)
+{
+	struct ax88796c_device *ax_local = to_ax88796c_device(ndev);
+	int mc_count = netdev_mc_count(ndev);
+	u16 rx_ctl = RXCR_AB;
+
+	lockdep_assert_held(&ax_local->spi_lock);
+
+	memset(ax_local->multi_filter, 0, AX_MCAST_FILTER_SIZE);
+
+	if (ndev->flags & IFF_PROMISC) {
+		rx_ctl |= RXCR_PRO;
+
+	} else if (ndev->flags & IFF_ALLMULTI || mc_count > AX_MAX_MCAST) {
+		rx_ctl |= RXCR_AMALL;
+
+	} else if (mc_count == 0) {
+		/* just broadcast and directed */
+	} else {
+		u32 crc_bits;
+		int i;
+		struct netdev_hw_addr *ha;
+
+		netdev_for_each_mc_addr(ha, ndev) {
+			crc_bits = ether_crc(ETH_ALEN, ha->addr);
+			ax_local->multi_filter[crc_bits >> 29] |=
+						(1 << ((crc_bits >> 26) & 7));
+		}
+
+		for (i = 0; i < 4; i++) {
+			AX_WRITE(&ax_local->ax_spi,
+				 ((ax_local->multi_filter[i * 2 + 1] << 8) |
+				  ax_local->multi_filter[i * 2]), P3_MFAR(i));
+		}
+	}
+
+	AX_WRITE(&ax_local->ax_spi, rx_ctl, P2_RXCR);
+}
+
+static void ax88796c_set_mac_addr(struct net_device *ndev)
+{
+	struct ax88796c_device *ax_local = to_ax88796c_device(ndev);
+
+	lockdep_assert_held(&ax_local->spi_lock);
+
+	AX_WRITE(&ax_local->ax_spi, ((u16)(ndev->dev_addr[4] << 8) |
+			(u16)ndev->dev_addr[5]), P3_MACASR0);
+	AX_WRITE(&ax_local->ax_spi, ((u16)(ndev->dev_addr[2] << 8) |
+			(u16)ndev->dev_addr[3]), P3_MACASR1);
+	AX_WRITE(&ax_local->ax_spi, ((u16)(ndev->dev_addr[0] << 8) |
+			(u16)ndev->dev_addr[1]), P3_MACASR2);
+}
+
+static void ax88796c_load_mac_addr(struct net_device *ndev)
+{
+	struct ax88796c_device *ax_local = to_ax88796c_device(ndev);
+	u8 addr[ETH_ALEN];
+	u16 temp;
+
+	lockdep_assert_held(&ax_local->spi_lock);
+
+	/* Try the device tree first */
+	if (!platform_get_ethdev_address(&ax_local->spi->dev, ndev) &&
+	    is_valid_ether_addr(ndev->dev_addr)) {
+		if (netif_msg_probe(ax_local))
+			dev_info(&ax_local->spi->dev,
+				 "MAC address read from device tree\n");
+		return;
+	}
+
+	/* Read the MAC address from AX88796C */
+	temp = AX_READ(&ax_local->ax_spi, P3_MACASR0);
+	addr[5] = (u8)temp;
+	addr[4] = (u8)(temp >> 8);
+
+	temp = AX_READ(&ax_local->ax_spi, P3_MACASR1);
+	addr[3] = (u8)temp;
+	addr[2] = (u8)(temp >> 8);
+
+	temp = AX_READ(&ax_local->ax_spi, P3_MACASR2);
+	addr[1] = (u8)temp;
+	addr[0] = (u8)(temp >> 8);
+
+	if (is_valid_ether_addr(addr)) {
+		eth_hw_addr_set(ndev, addr);
+		if (netif_msg_probe(ax_local))
+			dev_info(&ax_local->spi->dev,
+				 "MAC address read from ASIX chip\n");
+		return;
+	}
+
+	/* Use random address if none found */
+	if (netif_msg_probe(ax_local))
+		dev_info(&ax_local->spi->dev, "Use random MAC address\n");
+	eth_hw_addr_random(ndev);
+}
+
+static void ax88796c_proc_tx_hdr(struct tx_pkt_info *info, u8 ip_summed)
+{
+	u16 pkt_len_bar = (~info->pkt_len & TX_HDR_SOP_PKTLENBAR);
+
+	/* Prepare SOP header */
+	info->sop.flags_len = info->pkt_len |
+		((ip_summed == CHECKSUM_NONE) ||
+		 (ip_summed == CHECKSUM_UNNECESSARY) ? TX_HDR_SOP_DICF : 0);
+
+	info->sop.seq_lenbar = ((info->seq_num << 11) & TX_HDR_SOP_SEQNUM)
+				| pkt_len_bar;
+	cpu_to_be16s(&info->sop.flags_len);
+	cpu_to_be16s(&info->sop.seq_lenbar);
+
+	/* Prepare Segment header */
+	info->seg.flags_seqnum_seglen = TX_HDR_SEG_FS | TX_HDR_SEG_LS
+						| info->pkt_len;
+
+	info->seg.eo_so_seglenbar = pkt_len_bar;
+
+	cpu_to_be16s(&info->seg.flags_seqnum_seglen);
+	cpu_to_be16s(&info->seg.eo_so_seglenbar);
+
+	/* Prepare EOP header */
+	info->eop.seq_len = ((info->seq_num << 11) &
+			     TX_HDR_EOP_SEQNUM) | info->pkt_len;
+	info->eop.seqbar_lenbar = ((~info->seq_num << 11) &
+				   TX_HDR_EOP_SEQNUMBAR) | pkt_len_bar;
+
+	cpu_to_be16s(&info->eop.seq_len);
+	cpu_to_be16s(&info->eop.seqbar_lenbar);
+}
+
+static int
+ax88796c_check_free_pages(struct ax88796c_device *ax_local, u8 need_pages)
+{
+	u8 free_pages;
+	u16 tmp;
+
+	lockdep_assert_held(&ax_local->spi_lock);
+
+	free_pages = AX_READ(&ax_local->ax_spi, P0_TFBFCR) & TX_FREEBUF_MASK;
+	if (free_pages < need_pages) {
+		/* schedule free page interrupt */
+		tmp = AX_READ(&ax_local->ax_spi, P0_TFBFCR)
+				& TFBFCR_SCHE_FREE_PAGE;
+		AX_WRITE(&ax_local->ax_spi, tmp | TFBFCR_TX_PAGE_SET |
+				TFBFCR_SET_FREE_PAGE(need_pages),
+				P0_TFBFCR);
+		return -ENOMEM;
+	}
+
+	return 0;
+}
+
+static struct sk_buff *
+ax88796c_tx_fixup(struct net_device *ndev, struct sk_buff_head *q)
+{
+	struct ax88796c_device *ax_local = to_ax88796c_device(ndev);
+	u8 spi_len = ax_local->ax_spi.comp ? 1 : 4;
+	struct sk_buff *skb;
+	struct tx_pkt_info info;
+	struct skb_data *entry;
+	u16 pkt_len;
+	u8 padlen, seq_num;
+	u8 need_pages;
+	int headroom;
+	int tailroom;
+
+	if (skb_queue_empty(q))
+		return NULL;
+
+	skb = skb_peek(q);
+	pkt_len = skb->len;
+	need_pages = (pkt_len + TX_OVERHEAD + 127) >> 7;
+	if (ax88796c_check_free_pages(ax_local, need_pages) != 0)
+		return NULL;
+
+	headroom = skb_headroom(skb);
+	tailroom = skb_tailroom(skb);
+	padlen = round_up(pkt_len, 4) - pkt_len;
+	seq_num = ++ax_local->seq_num & 0x1F;
+
+	info.pkt_len = pkt_len;
+
+	if (skb_cloned(skb) ||
+	    (headroom < (TX_OVERHEAD + spi_len)) ||
+	    (tailroom < (padlen + TX_EOP_SIZE))) {
+		size_t h = max((TX_OVERHEAD + spi_len) - headroom, 0);
+		size_t t = max((padlen + TX_EOP_SIZE) - tailroom, 0);
+
+		if (pskb_expand_head(skb, h, t, GFP_KERNEL))
+			return NULL;
+	}
+
+	info.seq_num = seq_num;
+	ax88796c_proc_tx_hdr(&info, skb->ip_summed);
+
+	/* SOP and SEG header */
+	memcpy(skb_push(skb, TX_OVERHEAD), &info.tx_overhead, TX_OVERHEAD);
+
+	/* Write SPI TXQ header */
+	memcpy(skb_push(skb, spi_len), ax88796c_tx_cmd_buf, spi_len);
+
+	/* Make 32-bit alignment */
+	skb_put(skb, padlen);
+
+	/* EOP header */
+	skb_put_data(skb, &info.eop, TX_EOP_SIZE);
+
+	skb_unlink(skb, q);
+
+	entry = (struct skb_data *)skb->cb;
+	memset(entry, 0, sizeof(*entry));
+	entry->len = pkt_len;
+
+	if (netif_msg_pktdata(ax_local)) {
+		char pfx[IFNAMSIZ + 7];
+
+		snprintf(pfx, sizeof(pfx), "%s:     ", ndev->name);
+
+		netdev_info(ndev, "TX packet len %d, total len %d, seq %d\n",
+			    pkt_len, skb->len, seq_num);
+
+		netdev_info(ndev, "  SPI Header:\n");
+		print_hex_dump(KERN_INFO, pfx, DUMP_PREFIX_OFFSET, 16, 1,
+			       skb->data, 4, 0);
+
+		netdev_info(ndev, "  TX SOP:\n");
+		print_hex_dump(KERN_INFO, pfx, DUMP_PREFIX_OFFSET, 16, 1,
+			       skb->data + 4, TX_OVERHEAD, 0);
+
+		netdev_info(ndev, "  TX packet:\n");
+		print_hex_dump(KERN_INFO, pfx, DUMP_PREFIX_OFFSET, 16, 1,
+			       skb->data + 4 + TX_OVERHEAD,
+			       skb->len - TX_EOP_SIZE - 4 - TX_OVERHEAD, 0);
+
+		netdev_info(ndev, "  TX EOP:\n");
+		print_hex_dump(KERN_INFO, pfx, DUMP_PREFIX_OFFSET, 16, 1,
+			       skb->data + skb->len - 4, 4, 0);
+	}
+
+	return skb;
+}
+
+static int ax88796c_hard_xmit(struct ax88796c_device *ax_local)
+{
+	struct ax88796c_pcpu_stats *stats;
+	struct sk_buff *tx_skb;
+	struct skb_data *entry;
+	unsigned long flags;
+
+	lockdep_assert_held(&ax_local->spi_lock);
+
+	stats = this_cpu_ptr(ax_local->stats);
+	tx_skb = ax88796c_tx_fixup(ax_local->ndev, &ax_local->tx_wait_q);
+
+	if (!tx_skb) {
+		this_cpu_inc(ax_local->stats->tx_dropped);
+		return 0;
+	}
+	entry = (struct skb_data *)tx_skb->cb;
+
+	AX_WRITE(&ax_local->ax_spi,
+		 (TSNR_TXB_START | TSNR_PKT_CNT(1)), P0_TSNR);
+
+	axspi_write_txq(&ax_local->ax_spi, tx_skb->data, tx_skb->len);
+
+	if (((AX_READ(&ax_local->ax_spi, P0_TSNR) & TXNR_TXB_IDLE) == 0) ||
+	    ((ISR_TXERR & AX_READ(&ax_local->ax_spi, P0_ISR)) != 0)) {
+		/* Ack tx error int */
+		AX_WRITE(&ax_local->ax_spi, ISR_TXERR, P0_ISR);
+
+		this_cpu_inc(ax_local->stats->tx_dropped);
+
+		if (net_ratelimit())
+			netif_err(ax_local, tx_err, ax_local->ndev,
+				  "TX FIFO error, re-initialize the TX bridge\n");
+
+		/* Reinitial tx bridge */
+		AX_WRITE(&ax_local->ax_spi, TXNR_TXB_REINIT |
+			AX_READ(&ax_local->ax_spi, P0_TSNR), P0_TSNR);
+		ax_local->seq_num = 0;
+	} else {
+		flags = u64_stats_update_begin_irqsave(&stats->syncp);
+		u64_stats_inc(&stats->tx_packets);
+		u64_stats_add(&stats->tx_bytes, entry->len);
+		u64_stats_update_end_irqrestore(&stats->syncp, flags);
+	}
+
+	entry->state = tx_done;
+	dev_kfree_skb(tx_skb);
+
+	return 1;
+}
+
+static netdev_tx_t
+ax88796c_start_xmit(struct sk_buff *skb, struct net_device *ndev)
+{
+	struct ax88796c_device *ax_local = to_ax88796c_device(ndev);
+
+	skb_queue_tail(&ax_local->tx_wait_q, skb);
+	if (skb_queue_len(&ax_local->tx_wait_q) > TX_QUEUE_HIGH_WATER)
+		netif_stop_queue(ndev);
+
+	set_bit(EVENT_TX, &ax_local->flags);
+	schedule_work(&ax_local->ax_work);
+
+	return NETDEV_TX_OK;
+}
+
+static void
+ax88796c_skb_return(struct ax88796c_device *ax_local,
+		    struct sk_buff *skb, struct rx_header *rxhdr)
+{
+	struct net_device *ndev = ax_local->ndev;
+	struct ax88796c_pcpu_stats *stats;
+	unsigned long flags;
+	int status;
+
+	stats = this_cpu_ptr(ax_local->stats);
+
+	do {
+		if (!(ndev->features & NETIF_F_RXCSUM))
+			break;
+
+		/* checksum error bit is set */
+		if ((rxhdr->flags & RX_HDR3_L3_ERR) ||
+		    (rxhdr->flags & RX_HDR3_L4_ERR))
+			break;
+
+		/* Other types may be indicated by more than one bit. */
+		if ((rxhdr->flags & RX_HDR3_L4_TYPE_TCP) ||
+		    (rxhdr->flags & RX_HDR3_L4_TYPE_UDP))
+			skb->ip_summed = CHECKSUM_UNNECESSARY;
+	} while (0);
+
+	flags = u64_stats_update_begin_irqsave(&stats->syncp);
+	u64_stats_inc(&stats->rx_packets);
+	u64_stats_add(&stats->rx_bytes, skb->len);
+	u64_stats_update_end_irqrestore(&stats->syncp, flags);
+
+	skb->dev = ndev;
+	skb->protocol = eth_type_trans(skb, ax_local->ndev);
+
+	netif_info(ax_local, rx_status, ndev, "< rx, len %zu, type 0x%x\n",
+		   skb->len + sizeof(struct ethhdr), skb->protocol);
+
+	status = netif_rx(skb);
+	if (status != NET_RX_SUCCESS && net_ratelimit())
+		netif_info(ax_local, rx_err, ndev,
+			   "netif_rx status %d\n", status);
+}
+
+static void
+ax88796c_rx_fixup(struct ax88796c_device *ax_local, struct sk_buff *rx_skb)
+{
+	struct rx_header *rxhdr = (struct rx_header *)rx_skb->data;
+	struct net_device *ndev = ax_local->ndev;
+	u16 len;
+
+	be16_to_cpus(&rxhdr->flags_len);
+	be16_to_cpus(&rxhdr->seq_lenbar);
+	be16_to_cpus(&rxhdr->flags);
+
+	if ((rxhdr->flags_len & RX_HDR1_PKT_LEN) !=
+			 (~rxhdr->seq_lenbar & 0x7FF)) {
+		netif_err(ax_local, rx_err, ndev, "Header error\n");
+
+		this_cpu_inc(ax_local->stats->rx_frame_errors);
+		kfree_skb(rx_skb);
+		return;
+	}
+
+	if ((rxhdr->flags_len & RX_HDR1_MII_ERR) ||
+	    (rxhdr->flags_len & RX_HDR1_CRC_ERR)) {
+		netif_err(ax_local, rx_err, ndev, "CRC or MII error\n");
+
+		this_cpu_inc(ax_local->stats->rx_crc_errors);
+		kfree_skb(rx_skb);
+		return;
+	}
+
+	len = rxhdr->flags_len & RX_HDR1_PKT_LEN;
+	if (netif_msg_pktdata(ax_local)) {
+		char pfx[IFNAMSIZ + 7];
+
+		snprintf(pfx, sizeof(pfx), "%s:     ", ndev->name);
+		netdev_info(ndev, "RX data, total len %d, packet len %d\n",
+			    rx_skb->len, len);
+
+		netdev_info(ndev, "  Dump RX packet header:");
+		print_hex_dump(KERN_INFO, pfx, DUMP_PREFIX_OFFSET, 16, 1,
+			       rx_skb->data, sizeof(*rxhdr), 0);
+
+		netdev_info(ndev, "  Dump RX packet:");
+		print_hex_dump(KERN_INFO, pfx, DUMP_PREFIX_OFFSET, 16, 1,
+			       rx_skb->data + sizeof(*rxhdr), len, 0);
+	}
+
+	skb_pull(rx_skb, sizeof(*rxhdr));
+	pskb_trim(rx_skb, len);
+
+	ax88796c_skb_return(ax_local, rx_skb, rxhdr);
+}
+
+static int ax88796c_receive(struct net_device *ndev)
+{
+	struct ax88796c_device *ax_local = to_ax88796c_device(ndev);
+	struct skb_data *entry;
+	u16 w_count, pkt_len;
+	struct sk_buff *skb;
+	u8 pkt_cnt;
+
+	lockdep_assert_held(&ax_local->spi_lock);
+
+	/* check rx packet and total word count */
+	AX_WRITE(&ax_local->ax_spi, AX_READ(&ax_local->ax_spi, P0_RTWCR)
+		  | RTWCR_RX_LATCH, P0_RTWCR);
+
+	pkt_cnt = AX_READ(&ax_local->ax_spi, P0_RXBCR2) & RXBCR2_PKT_MASK;
+	if (!pkt_cnt)
+		return 0;
+
+	pkt_len = AX_READ(&ax_local->ax_spi, P0_RCPHR) & 0x7FF;
+
+	w_count = round_up(pkt_len + 6, 4) >> 1;
+
+	skb = netdev_alloc_skb(ndev, w_count * 2);
+	if (!skb) {
+		AX_WRITE(&ax_local->ax_spi, RXBCR1_RXB_DISCARD, P0_RXBCR1);
+		this_cpu_inc(ax_local->stats->rx_dropped);
+		return 0;
+	}
+	entry = (struct skb_data *)skb->cb;
+
+	AX_WRITE(&ax_local->ax_spi, RXBCR1_RXB_START | w_count, P0_RXBCR1);
+
+	axspi_read_rxq(&ax_local->ax_spi,
+		       skb_put(skb, w_count * 2), skb->len);
+
+	/* Check if rx bridge is idle */
+	if ((AX_READ(&ax_local->ax_spi, P0_RXBCR2) & RXBCR2_RXB_IDLE) == 0) {
+		if (net_ratelimit())
+			netif_err(ax_local, rx_err, ndev,
+				  "Rx Bridge is not idle\n");
+		AX_WRITE(&ax_local->ax_spi, RXBCR2_RXB_REINIT, P0_RXBCR2);
+
+		entry->state = rx_err;
+	} else {
+		entry->state = rx_done;
+	}
+
+	AX_WRITE(&ax_local->ax_spi, ISR_RXPKT, P0_ISR);
+
+	ax88796c_rx_fixup(ax_local, skb);
+
+	return 1;
+}
+
+static int ax88796c_process_isr(struct ax88796c_device *ax_local)
+{
+	struct net_device *ndev = ax_local->ndev;
+	int todo = 0;
+	u16 isr;
+
+	lockdep_assert_held(&ax_local->spi_lock);
+
+	isr = AX_READ(&ax_local->ax_spi, P0_ISR);
+	AX_WRITE(&ax_local->ax_spi, isr, P0_ISR);
+
+	netif_dbg(ax_local, intr, ndev, "  ISR 0x%04x\n", isr);
+
+	if (isr & ISR_TXERR) {
+		netif_dbg(ax_local, intr, ndev, "  TXERR interrupt\n");
+		AX_WRITE(&ax_local->ax_spi, TXNR_TXB_REINIT, P0_TSNR);
+		ax_local->seq_num = 0x1f;
+	}
+
+	if (isr & ISR_TXPAGES) {
+		netif_dbg(ax_local, intr, ndev, "  TXPAGES interrupt\n");
+		set_bit(EVENT_TX, &ax_local->flags);
+	}
+
+	if (isr & ISR_LINK) {
+		netif_dbg(ax_local, intr, ndev, "  Link change interrupt\n");
+		phy_mac_interrupt(ax_local->ndev->phydev);
+	}
+
+	if (isr & ISR_RXPKT) {
+		netif_dbg(ax_local, intr, ndev, "  RX interrupt\n");
+		todo = ax88796c_receive(ax_local->ndev);
+	}
+
+	return todo;
+}
+
+static irqreturn_t ax88796c_interrupt(int irq, void *dev_instance)
+{
+	struct ax88796c_device *ax_local;
+	struct net_device *ndev;
+
+	ndev = dev_instance;
+	if (!ndev) {
+		pr_err("irq %d for unknown device.\n", irq);
+		return IRQ_RETVAL(0);
+	}
+	ax_local = to_ax88796c_device(ndev);
+
+	disable_irq_nosync(irq);
+
+	netif_dbg(ax_local, intr, ndev, "Interrupt occurred\n");
+
+	set_bit(EVENT_INTR, &ax_local->flags);
+	schedule_work(&ax_local->ax_work);
+
+	return IRQ_HANDLED;
+}
+
+static void ax88796c_work(struct work_struct *work)
+{
+	struct ax88796c_device *ax_local =
+			container_of(work, struct ax88796c_device, ax_work);
+
+	mutex_lock(&ax_local->spi_lock);
+
+	if (test_bit(EVENT_SET_MULTI, &ax_local->flags)) {
+		ax88796c_set_hw_multicast(ax_local->ndev);
+		clear_bit(EVENT_SET_MULTI, &ax_local->flags);
+	}
+
+	if (test_bit(EVENT_INTR, &ax_local->flags)) {
+		AX_WRITE(&ax_local->ax_spi, IMR_MASKALL, P0_IMR);
+
+		while (ax88796c_process_isr(ax_local))
+			/* nothing */;
+
+		clear_bit(EVENT_INTR, &ax_local->flags);
+
+		AX_WRITE(&ax_local->ax_spi, IMR_DEFAULT, P0_IMR);
+
+		enable_irq(ax_local->ndev->irq);
+	}
+
+	if (test_bit(EVENT_TX, &ax_local->flags)) {
+		while (skb_queue_len(&ax_local->tx_wait_q)) {
+			if (!ax88796c_hard_xmit(ax_local))
+				break;
+		}
+
+		clear_bit(EVENT_TX, &ax_local->flags);
+
+		if (netif_queue_stopped(ax_local->ndev) &&
+		    (skb_queue_len(&ax_local->tx_wait_q) < TX_QUEUE_LOW_WATER))
+			netif_wake_queue(ax_local->ndev);
+	}
+
+	mutex_unlock(&ax_local->spi_lock);
+}
+
+static void ax88796c_get_stats64(struct net_device *ndev,
+				 struct rtnl_link_stats64 *stats)
+{
+	struct ax88796c_device *ax_local = to_ax88796c_device(ndev);
+	u32 rx_frame_errors = 0, rx_crc_errors = 0;
+	u32 rx_dropped = 0, tx_dropped = 0;
+	unsigned int start;
+	int cpu;
+
+	for_each_possible_cpu(cpu) {
+		struct ax88796c_pcpu_stats *s;
+		u64 rx_packets, rx_bytes;
+		u64 tx_packets, tx_bytes;
+
+		s = per_cpu_ptr(ax_local->stats, cpu);
+
+		do {
+			start = u64_stats_fetch_begin(&s->syncp);
+			rx_packets = u64_stats_read(&s->rx_packets);
+			rx_bytes   = u64_stats_read(&s->rx_bytes);
+			tx_packets = u64_stats_read(&s->tx_packets);
+			tx_bytes   = u64_stats_read(&s->tx_bytes);
+		} while (u64_stats_fetch_retry(&s->syncp, start));
+
+		stats->rx_packets += rx_packets;
+		stats->rx_bytes   += rx_bytes;
+		stats->tx_packets += tx_packets;
+		stats->tx_bytes   += tx_bytes;
+
+		rx_dropped      += s->rx_dropped;
+		tx_dropped      += s->tx_dropped;
+		rx_frame_errors += s->rx_frame_errors;
+		rx_crc_errors   += s->rx_crc_errors;
+	}
+
+	stats->rx_dropped = rx_dropped;
+	stats->tx_dropped = tx_dropped;
+	stats->rx_frame_errors = rx_frame_errors;
+	stats->rx_crc_errors = rx_crc_errors;
+}
+
+static void ax88796c_set_mac(struct  ax88796c_device *ax_local)
+{
+	u16 maccr;
+
+	maccr = (ax_local->link) ? MACCR_RXEN : 0;
+
+	switch (ax_local->speed) {
+	case SPEED_100:
+		maccr |= MACCR_SPEED_100;
+		break;
+	case SPEED_10:
+	case SPEED_UNKNOWN:
+		break;
+	default:
+		return;
+	}
+
+	switch (ax_local->duplex) {
+	case DUPLEX_FULL:
+		maccr |= MACCR_SPEED_100;
+		break;
+	case DUPLEX_HALF:
+	case DUPLEX_UNKNOWN:
+		break;
+	default:
+		return;
+	}
+
+	if (ax_local->flowctrl & AX_FC_ANEG &&
+	    ax_local->phydev->autoneg) {
+		maccr |= ax_local->pause ? MACCR_RXFC_ENABLE : 0;
+		maccr |= !ax_local->pause != !ax_local->asym_pause ?
+			MACCR_TXFC_ENABLE : 0;
+	} else {
+		maccr |= (ax_local->flowctrl & AX_FC_RX) ? MACCR_RXFC_ENABLE : 0;
+		maccr |= (ax_local->flowctrl & AX_FC_TX) ? MACCR_TXFC_ENABLE : 0;
+	}
+
+	mutex_lock(&ax_local->spi_lock);
+
+	maccr |= AX_READ(&ax_local->ax_spi, P0_MACCR) &
+		~(MACCR_DUPLEX_FULL | MACCR_SPEED_100 |
+		  MACCR_TXFC_ENABLE | MACCR_RXFC_ENABLE);
+	AX_WRITE(&ax_local->ax_spi, maccr, P0_MACCR);
+
+	mutex_unlock(&ax_local->spi_lock);
+}
+
+static void ax88796c_handle_link_change(struct net_device *ndev)
+{
+	struct ax88796c_device *ax_local = to_ax88796c_device(ndev);
+	struct phy_device *phydev = ndev->phydev;
+	bool update = false;
+
+	if (phydev->link && (ax_local->speed != phydev->speed ||
+			     ax_local->duplex != phydev->duplex ||
+			     ax_local->pause != phydev->pause ||
+			     ax_local->asym_pause != phydev->asym_pause)) {
+		ax_local->speed = phydev->speed;
+		ax_local->duplex = phydev->duplex;
+		ax_local->pause = phydev->pause;
+		ax_local->asym_pause = phydev->asym_pause;
+		update = true;
+	}
+
+	if (phydev->link != ax_local->link) {
+		if (!phydev->link) {
+			ax_local->speed = SPEED_UNKNOWN;
+			ax_local->duplex = DUPLEX_UNKNOWN;
+		}
+
+		ax_local->link = phydev->link;
+		update = true;
+	}
+
+	if (update)
+		ax88796c_set_mac(ax_local);
+
+	if (net_ratelimit())
+		phy_print_status(ndev->phydev);
+}
+
+static void ax88796c_set_csums(struct ax88796c_device *ax_local)
+{
+	struct net_device *ndev = ax_local->ndev;
+
+	lockdep_assert_held(&ax_local->spi_lock);
+
+	if (ndev->features & NETIF_F_RXCSUM) {
+		AX_WRITE(&ax_local->ax_spi, COERCR0_DEFAULT, P4_COERCR0);
+		AX_WRITE(&ax_local->ax_spi, COERCR1_DEFAULT, P4_COERCR1);
+	} else {
+		AX_WRITE(&ax_local->ax_spi, 0, P4_COERCR0);
+		AX_WRITE(&ax_local->ax_spi, 0, P4_COERCR1);
+	}
+
+	if (ndev->features & NETIF_F_HW_CSUM) {
+		AX_WRITE(&ax_local->ax_spi, COETCR0_DEFAULT, P4_COETCR0);
+		AX_WRITE(&ax_local->ax_spi, COETCR1_TXPPPE, P4_COETCR1);
+	} else {
+		AX_WRITE(&ax_local->ax_spi, 0, P4_COETCR0);
+		AX_WRITE(&ax_local->ax_spi, 0, P4_COETCR1);
+	}
+}
+
+static int
+ax88796c_open(struct net_device *ndev)
+{
+	struct ax88796c_device *ax_local = to_ax88796c_device(ndev);
+	unsigned long irq_flag = 0;
+	int fc = AX_FC_NONE;
+	int ret;
+	u16 t;
+
+	ret = request_irq(ndev->irq, ax88796c_interrupt,
+			  irq_flag, ndev->name, ndev);
+	if (ret) {
+		netdev_err(ndev, "unable to get IRQ %d (errno=%d).\n",
+			   ndev->irq, ret);
+		return ret;
+	}
+
+	mutex_lock(&ax_local->spi_lock);
+
+	ret = ax88796c_soft_reset(ax_local);
+	if (ret < 0) {
+		free_irq(ndev->irq, ndev);
+		mutex_unlock(&ax_local->spi_lock);
+		return ret;
+	}
+	ax_local->seq_num = 0x1f;
+
+	ax88796c_set_mac_addr(ndev);
+	ax88796c_set_csums(ax_local);
+
+	/* Disable stuffing packet */
+	t = AX_READ(&ax_local->ax_spi, P1_RXBSPCR);
+	t &= ~RXBSPCR_STUF_ENABLE;
+	AX_WRITE(&ax_local->ax_spi, t, P1_RXBSPCR);
+
+	/* Enable RX packet process */
+	AX_WRITE(&ax_local->ax_spi, RPPER_RXEN, P1_RPPER);
+
+	t = AX_READ(&ax_local->ax_spi, P0_FER);
+	t |= FER_RXEN | FER_TXEN | FER_BSWAP | FER_IRQ_PULL;
+	AX_WRITE(&ax_local->ax_spi, t, P0_FER);
+
+	/* Setup LED mode */
+	AX_WRITE(&ax_local->ax_spi,
+		 (LCR_LED0_EN | LCR_LED0_DUPLEX | LCR_LED1_EN |
+		 LCR_LED1_100MODE), P2_LCR0);
+	AX_WRITE(&ax_local->ax_spi,
+		 (AX_READ(&ax_local->ax_spi, P2_LCR1) & LCR_LED2_MASK) |
+		 LCR_LED2_EN | LCR_LED2_LINK, P2_LCR1);
+
+	/* Disable PHY auto-polling */
+	AX_WRITE(&ax_local->ax_spi, PCR_PHYID(AX88796C_PHY_ID), P2_PCR);
+
+	/* Enable MAC interrupts */
+	AX_WRITE(&ax_local->ax_spi, IMR_DEFAULT, P0_IMR);
+
+	mutex_unlock(&ax_local->spi_lock);
+
+	/* Setup flow-control configuration */
+	phy_support_asym_pause(ax_local->phydev);
+
+	if (linkmode_test_bit(ETHTOOL_LINK_MODE_Pause_BIT,
+			      ax_local->phydev->advertising) ||
+	    linkmode_test_bit(ETHTOOL_LINK_MODE_Asym_Pause_BIT,
+			      ax_local->phydev->advertising))
+		fc |= AX_FC_ANEG;
+
+	fc |= linkmode_test_bit(ETHTOOL_LINK_MODE_Pause_BIT,
+				ax_local->phydev->advertising) ? AX_FC_RX : 0;
+	fc |= (linkmode_test_bit(ETHTOOL_LINK_MODE_Pause_BIT,
+				 ax_local->phydev->advertising) !=
+	       linkmode_test_bit(ETHTOOL_LINK_MODE_Asym_Pause_BIT,
+				 ax_local->phydev->advertising)) ? AX_FC_TX : 0;
+	ax_local->flowctrl = fc;
+
+	phy_start(ax_local->ndev->phydev);
+
+	netif_start_queue(ndev);
+
+	spi_message_init(&ax_local->ax_spi.rx_msg);
+
+	return 0;
+}
+
+static int
+ax88796c_close(struct net_device *ndev)
+{
+	struct ax88796c_device *ax_local = to_ax88796c_device(ndev);
+
+	phy_stop(ndev->phydev);
+
+	/* We lock the mutex early not only to protect the device
+	 * against concurrent access, but also avoid waking up the
+	 * queue in ax88796c_work(). phy_stop() needs to be called
+	 * before because it locks the mutex to access SPI.
+	 */
+	mutex_lock(&ax_local->spi_lock);
+
+	netif_stop_queue(ndev);
+
+	/* No more work can be scheduled now. Make any pending work,
+	 * including one already waiting for the mutex to be unlocked,
+	 * NOP.
+	 */
+	netif_dbg(ax_local, ifdown, ndev, "clearing bits\n");
+	clear_bit(EVENT_SET_MULTI, &ax_local->flags);
+	clear_bit(EVENT_INTR, &ax_local->flags);
+	clear_bit(EVENT_TX, &ax_local->flags);
+
+	/* Disable MAC interrupts */
+	AX_WRITE(&ax_local->ax_spi, IMR_MASKALL, P0_IMR);
+	__skb_queue_purge(&ax_local->tx_wait_q);
+	ax88796c_soft_reset(ax_local);
+
+	mutex_unlock(&ax_local->spi_lock);
+
+	cancel_work_sync(&ax_local->ax_work);
+
+	free_irq(ndev->irq, ndev);
+
+	return 0;
+}
+
+static int
+ax88796c_set_features(struct net_device *ndev, netdev_features_t features)
+{
+	struct ax88796c_device *ax_local = to_ax88796c_device(ndev);
+	netdev_features_t changed = features ^ ndev->features;
+
+	if (!(changed & (NETIF_F_RXCSUM | NETIF_F_HW_CSUM)))
+		return 0;
+
+	ndev->features = features;
+
+	if (changed & (NETIF_F_RXCSUM | NETIF_F_HW_CSUM))
+		ax88796c_set_csums(ax_local);
+
+	return 0;
+}
+
+static const struct net_device_ops ax88796c_netdev_ops = {
+	.ndo_open		= ax88796c_open,
+	.ndo_stop		= ax88796c_close,
+	.ndo_start_xmit		= ax88796c_start_xmit,
+	.ndo_get_stats64	= ax88796c_get_stats64,
+	.ndo_eth_ioctl		= ax88796c_ioctl,
+	.ndo_set_mac_address	= eth_mac_addr,
+	.ndo_set_features	= ax88796c_set_features,
+};
+
+static int ax88796c_hard_reset(struct ax88796c_device *ax_local)
+{
+	struct device *dev = (struct device *)&ax_local->spi->dev;
+	struct gpio_desc *reset_gpio;
+
+	/* reset info */
+	reset_gpio = gpiod_get(dev, "reset", 0);
+	if (IS_ERR(reset_gpio)) {
+		dev_err(dev, "Could not get 'reset' GPIO: %ld", PTR_ERR(reset_gpio));
+		return PTR_ERR(reset_gpio);
+	}
+
+	/* set reset */
+	gpiod_direction_output(reset_gpio, 1);
+	msleep(100);
+	gpiod_direction_output(reset_gpio, 0);
+	gpiod_put(reset_gpio);
+	msleep(20);
+
+	return 0;
+}
+
+static int ax88796c_probe(struct spi_device *spi)
+{
+	char phy_id[MII_BUS_ID_SIZE + 3];
+	struct ax88796c_device *ax_local;
+	struct net_device *ndev;
+	u16 temp;
+	int ret;
+
+	ndev = devm_alloc_etherdev(&spi->dev, sizeof(*ax_local));
+	if (!ndev)
+		return -ENOMEM;
+
+	SET_NETDEV_DEV(ndev, &spi->dev);
+
+	ax_local = to_ax88796c_device(ndev);
+
+	dev_set_drvdata(&spi->dev, ax_local);
+	ax_local->spi = spi;
+	ax_local->ax_spi.spi = spi;
+
+	ax_local->stats =
+		devm_netdev_alloc_pcpu_stats(&spi->dev,
+					     struct ax88796c_pcpu_stats);
+	if (!ax_local->stats)
+		return -ENOMEM;
+
+	ax_local->ndev = ndev;
+	ax_local->priv_flags |= comp ? AX_CAP_COMP : 0;
+	ax_local->msg_enable = msg_enable;
+	mutex_init(&ax_local->spi_lock);
+
+	ax_local->mdiobus = devm_mdiobus_alloc(&spi->dev);
+	if (!ax_local->mdiobus)
+		return -ENOMEM;
+
+	ax_local->mdiobus->priv = ax_local;
+	ax_local->mdiobus->read = ax88796c_mdio_read;
+	ax_local->mdiobus->write = ax88796c_mdio_write;
+	ax_local->mdiobus->name = "ax88976c-mdiobus";
+	ax_local->mdiobus->phy_mask = (u32)~BIT(AX88796C_PHY_ID);
+	ax_local->mdiobus->parent = &spi->dev;
+
+	snprintf(ax_local->mdiobus->id, MII_BUS_ID_SIZE,
+		 "ax88796c-%s.%u", dev_name(&spi->dev), spi_get_chipselect(spi, 0));
+
+	ret = devm_mdiobus_register(&spi->dev, ax_local->mdiobus);
+	if (ret < 0) {
+		dev_err(&spi->dev, "Could not register MDIO bus\n");
+		return ret;
+	}
+
+	if (netif_msg_probe(ax_local)) {
+		dev_info(&spi->dev, "AX88796C-SPI Configuration:\n");
+		dev_info(&spi->dev, "    Compression : %s\n",
+			 ax_local->priv_flags & AX_CAP_COMP ? "ON" : "OFF");
+	}
+
+	ndev->irq = spi->irq;
+	ndev->netdev_ops = &ax88796c_netdev_ops;
+	ndev->ethtool_ops = &ax88796c_ethtool_ops;
+	ndev->hw_features |= NETIF_F_HW_CSUM | NETIF_F_RXCSUM;
+	ndev->features |= NETIF_F_HW_CSUM | NETIF_F_RXCSUM;
+	ndev->needed_headroom = TX_OVERHEAD;
+	ndev->needed_tailroom = TX_EOP_SIZE;
+
+	mutex_lock(&ax_local->spi_lock);
+
+	/* ax88796c gpio reset */
+	ax88796c_hard_reset(ax_local);
+
+	/* Reset AX88796C */
+	ret = ax88796c_soft_reset(ax_local);
+	if (ret < 0) {
+		ret = -ENODEV;
+		mutex_unlock(&ax_local->spi_lock);
+		goto err;
+	}
+	/* Check board revision */
+	temp = AX_READ(&ax_local->ax_spi, P2_CRIR);
+	if ((temp & 0xF) != 0x0) {
+		dev_err(&spi->dev, "spi read failed: %d\n", temp);
+		ret = -ENODEV;
+		mutex_unlock(&ax_local->spi_lock);
+		goto err;
+	}
+
+	/*Reload EEPROM*/
+	ax88796c_reload_eeprom(ax_local);
+
+	ax88796c_load_mac_addr(ndev);
+
+	if (netif_msg_probe(ax_local))
+		dev_info(&spi->dev,
+			 "irq %d, MAC addr %02X:%02X:%02X:%02X:%02X:%02X\n",
+			 ndev->irq,
+			 ndev->dev_addr[0], ndev->dev_addr[1],
+			 ndev->dev_addr[2], ndev->dev_addr[3],
+			 ndev->dev_addr[4], ndev->dev_addr[5]);
+
+	/* Disable power saving */
+	AX_WRITE(&ax_local->ax_spi, (AX_READ(&ax_local->ax_spi, P0_PSCR)
+				     & PSCR_PS_MASK) | PSCR_PS_D0, P0_PSCR);
+
+	mutex_unlock(&ax_local->spi_lock);
+
+	INIT_WORK(&ax_local->ax_work, ax88796c_work);
+
+	skb_queue_head_init(&ax_local->tx_wait_q);
+
+	snprintf(phy_id, MII_BUS_ID_SIZE + 3, PHY_ID_FMT,
+		 ax_local->mdiobus->id, AX88796C_PHY_ID);
+	ax_local->phydev = phy_connect(ax_local->ndev, phy_id,
+				       ax88796c_handle_link_change,
+				       PHY_INTERFACE_MODE_MII);
+	if (IS_ERR(ax_local->phydev)) {
+		ret = PTR_ERR(ax_local->phydev);
+		goto err;
+	}
+	ax_local->phydev->irq = PHY_POLL;
+
+	ret = devm_register_netdev(&spi->dev, ndev);
+	if (ret) {
+		dev_err(&spi->dev, "failed to register a network device\n");
+		goto err_phy_dis;
+	}
+
+	netif_info(ax_local, probe, ndev, "%s %s registered\n",
+		   dev_driver_string(&spi->dev),
+		   dev_name(&spi->dev));
+	phy_attached_info(ax_local->phydev);
+
+	return 0;
+
+err_phy_dis:
+	phy_disconnect(ax_local->phydev);
+err:
+	return ret;
+}
+
+static void ax88796c_remove(struct spi_device *spi)
+{
+	struct ax88796c_device *ax_local = dev_get_drvdata(&spi->dev);
+	struct net_device *ndev = ax_local->ndev;
+
+	phy_disconnect(ndev->phydev);
+
+	netif_info(ax_local, probe, ndev, "removing network device %s %s\n",
+		   dev_driver_string(&spi->dev),
+		   dev_name(&spi->dev));
+}
+
+#ifdef CONFIG_OF
+static const struct of_device_id ax88796c_dt_ids[] = {
+	{ .compatible = "asix,ax88796c" },
+	{},
+};
+MODULE_DEVICE_TABLE(of, ax88796c_dt_ids);
+#endif
+
+static const struct spi_device_id asix_id[] = {
+	{ "ax88796c", 0 },
+	{ }
+};
+MODULE_DEVICE_TABLE(spi, asix_id);
+
+static struct spi_driver ax88796c_spi_driver = {
+	.driver = {
+		.name = DRV_NAME,
+		.of_match_table = of_match_ptr(ax88796c_dt_ids),
+	},
+	.probe = ax88796c_probe,
+	.remove = ax88796c_remove,
+	.id_table = asix_id,
+};
+
+static __init int ax88796c_spi_init(void)
+{
+	int ret;
+
+	bitmap_zero(ax88796c_no_regs_mask, AX88796C_REGDUMP_LEN);
+	ret = bitmap_parse(no_regs_list, 35,
+			   ax88796c_no_regs_mask, AX88796C_REGDUMP_LEN);
+	if (ret) {
+		bitmap_fill(ax88796c_no_regs_mask, AX88796C_REGDUMP_LEN);
+		pr_err("Invalid bitmap description, masking all registers\n");
+	}
+
+	return spi_register_driver(&ax88796c_spi_driver);
+}
+
+static __exit void ax88796c_spi_exit(void)
+{
+	spi_unregister_driver(&ax88796c_spi_driver);
+}
+
+module_init(ax88796c_spi_init);
+module_exit(ax88796c_spi_exit);
+
+MODULE_AUTHOR("Łukasz Stelmach <l.stelmach@samsung.com>");
+MODULE_DESCRIPTION("ASIX AX88796C SPI Ethernet driver");
+MODULE_LICENSE("GPL");
diff --git a/drivers/net/ethernet/asix/ax88796c_main.h b/drivers/net/ethernet/asix/ax88796c_main.h
new file mode 100644
index 000000000000..68a09edecab8
--- /dev/null
+++ b/drivers/net/ethernet/asix/ax88796c_main.h
@@ -0,0 +1,570 @@
+/* SPDX-License-Identifier: GPL-2.0-only */
+/*
+ * Copyright (c) 2010 ASIX Electronics Corporation
+ * Copyright (c) 2020 Samsung Electronics
+ *
+ * ASIX AX88796C SPI Fast Ethernet Linux driver
+ */
+
+#ifndef _AX88796C_MAIN_H
+#define _AX88796C_MAIN_H
+
+#include <linux/netdevice.h>
+#include <linux/mii.h>
+
+#include "ax88796c_spi.h"
+
+/* These identify the driver base version and may not be removed. */
+#define DRV_NAME	"ax88796c"
+#define ADP_NAME	"ASIX AX88796C SPI Ethernet Adapter"
+
+#define TX_QUEUE_HIGH_WATER		45	/* Tx queue high water mark */
+#define TX_QUEUE_LOW_WATER		20	/* Tx queue low water mark */
+
+#define AX88796C_REGDUMP_LEN		256
+#define AX88796C_PHY_REGDUMP_LEN	14
+#define AX88796C_PHY_ID			0x10
+
+#define TX_OVERHEAD     sizeof_field(struct tx_pkt_info, tx_overhead)
+#define TX_EOP_SIZE			4
+
+#define AX_MCAST_FILTER_SIZE		8
+#define AX_MAX_MCAST			64
+#define AX_MAX_CLK                      80000000
+#define TX_HDR_SOP_DICF			0x8000
+#define TX_HDR_SOP_CPHI			0x4000
+#define TX_HDR_SOP_INT			0x2000
+#define TX_HDR_SOP_MDEQ			0x1000
+#define TX_HDR_SOP_PKTLEN		0x07FF
+#define TX_HDR_SOP_SEQNUM		0xF800
+#define TX_HDR_SOP_PKTLENBAR		0x07FF
+
+#define TX_HDR_SEG_FS			0x8000
+#define TX_HDR_SEG_LS			0x4000
+#define TX_HDR_SEG_SEGNUM		0x3800
+#define TX_HDR_SEG_SEGLEN		0x0700
+#define TX_HDR_SEG_EOFST		0xC000
+#define TX_HDR_SEG_SOFST		0x3800
+#define TX_HDR_SEG_SEGLENBAR		0x07FF
+
+#define TX_HDR_EOP_SEQNUM		0xF800
+#define TX_HDR_EOP_PKTLEN		0x07FF
+#define TX_HDR_EOP_SEQNUMBAR		0xF800
+#define TX_HDR_EOP_PKTLENBAR		0x07FF
+
+/* Rx header fields mask */
+#define RX_HDR1_MCBC			0x8000
+#define RX_HDR1_STUFF_PKT		0x4000
+#define RX_HDR1_MII_ERR			0x2000
+#define RX_HDR1_CRC_ERR			0x1000
+#define RX_HDR1_PKT_LEN			0x07FF
+
+#define RX_HDR2_SEQ_NUM			0xF800
+#define RX_HDR2_PKT_LEN_BAR		0x7FFF
+
+#define RX_HDR3_PE			0x8000
+#define RX_HDR3_L3_TYPE_IPV4V6		0x6000
+#define RX_HDR3_L3_TYPE_IP		0x4000
+#define RX_HDR3_L3_TYPE_IPV6		0x2000
+#define RX_HDR3_L4_TYPE_ICMPV6		0x1400
+#define RX_HDR3_L4_TYPE_TCP		0x1000
+#define RX_HDR3_L4_TYPE_IGMP		0x0c00
+#define RX_HDR3_L4_TYPE_ICMP		0x0800
+#define RX_HDR3_L4_TYPE_UDP		0x0400
+#define RX_HDR3_L3_ERR			0x0200
+#define RX_HDR3_L4_ERR			0x0100
+#define RX_HDR3_PRIORITY(x)		((x) << 4)
+#define RX_HDR3_STRIP			0x0008
+#define RX_HDR3_VLAN_ID			0x0007
+
+struct ax88796c_pcpu_stats {
+	u64_stats_t rx_packets;
+	u64_stats_t rx_bytes;
+	u64_stats_t tx_packets;
+	u64_stats_t tx_bytes;
+	struct u64_stats_sync syncp;
+	u32 rx_dropped;
+	u32 tx_dropped;
+	u32 rx_frame_errors;
+	u32 rx_crc_errors;
+};
+
+struct ax88796c_device {
+	struct spi_device	*spi;
+	struct net_device	*ndev;
+	struct ax88796c_pcpu_stats __percpu *stats;
+
+	struct work_struct	ax_work;
+
+	struct mutex		spi_lock; /* device access */
+
+	struct sk_buff_head	tx_wait_q;
+
+	struct axspi_data	ax_spi;
+
+	struct mii_bus		*mdiobus;
+	struct phy_device	*phydev;
+
+	int			msg_enable;
+
+	u16			seq_num;
+
+	u8			multi_filter[AX_MCAST_FILTER_SIZE];
+
+	int			link;
+	int			speed;
+	int			duplex;
+	int			pause;
+	int			asym_pause;
+	int			flowctrl;
+		#define AX_FC_NONE		0
+		#define AX_FC_RX		BIT(0)
+		#define AX_FC_TX		BIT(1)
+		#define AX_FC_ANEG		BIT(2)
+
+	u32			priv_flags;
+		#define AX_CAP_COMP		BIT(0)
+		#define AX_PRIV_FLAGS_MASK	(AX_CAP_COMP)
+
+	unsigned long		flags;
+		#define EVENT_INTR		0
+		#define EVENT_TX		1
+		#define EVENT_SET_MULTI		2
+
+};
+
+#define to_ax88796c_device(ndev) ((struct ax88796c_device *)netdev_priv(ndev))
+
+enum skb_state {
+	illegal = 0,
+	tx_done,
+	rx_done,
+	rx_err,
+};
+
+struct skb_data {
+	enum skb_state state;
+	size_t len;
+};
+
+/* A88796C register definition */
+	/* Definition of PAGE0 */
+#define P0_PSR		(0x00)
+	#define PSR_DEV_READY		BIT(7)
+	#define PSR_RESET		(0 << 15)
+	#define PSR_RESET_CLR		BIT(15)
+#define P0_BOR		(0x02)
+#define P0_FER		(0x04)
+	#define FER_IPALM		BIT(0)
+	#define FER_DCRC		BIT(1)
+	#define FER_RH3M		BIT(2)
+	#define FER_HEADERSWAP		BIT(7)
+	#define FER_WSWAP		BIT(8)
+	#define FER_BSWAP		BIT(9)
+	#define FER_INTHI		BIT(10)
+	#define FER_INTLO		(0 << 10)
+	#define FER_IRQ_PULL		BIT(11)
+	#define FER_RXEN		BIT(14)
+	#define FER_TXEN		BIT(15)
+#define P0_ISR		(0x06)
+	#define ISR_RXPKT		BIT(0)
+	#define ISR_MDQ			BIT(4)
+	#define ISR_TXT			BIT(5)
+	#define ISR_TXPAGES		BIT(6)
+	#define ISR_TXERR		BIT(8)
+	#define ISR_LINK		BIT(9)
+#define P0_IMR		(0x08)
+	#define IMR_RXPKT		BIT(0)
+	#define IMR_MDQ			BIT(4)
+	#define IMR_TXT			BIT(5)
+	#define IMR_TXPAGES		BIT(6)
+	#define IMR_TXERR		BIT(8)
+	#define IMR_LINK		BIT(9)
+	#define IMR_MASKALL		(0xFFFF)
+	#define IMR_DEFAULT		(IMR_TXERR)
+#define P0_WFCR		(0x0A)
+	#define WFCR_PMEIND		BIT(0) /* PME indication */
+	#define WFCR_PMETYPE		BIT(1) /* PME I/O type */
+	#define WFCR_PMEPOL		BIT(2) /* PME polarity */
+	#define WFCR_PMERST		BIT(3) /* Reset PME */
+	#define WFCR_SLEEP		BIT(4) /* Enable sleep mode */
+	#define WFCR_WAKEUP		BIT(5) /* Enable wakeup mode */
+	#define WFCR_WAITEVENT		BIT(6) /* Reserved */
+	#define WFCR_CLRWAKE		BIT(7) /* Clear wakeup */
+	#define WFCR_LINKCH		BIT(8) /* Enable link change */
+	#define WFCR_MAGICP		BIT(9) /* Enable magic packet */
+	#define WFCR_WAKEF		BIT(10) /* Enable wakeup frame */
+	#define WFCR_PMEEN		BIT(11) /* Enable PME pin */
+	#define WFCR_LINKCHS		BIT(12) /* Link change status */
+	#define WFCR_MAGICPS		BIT(13) /* Magic packet status */
+	#define WFCR_WAKEFS		BIT(14) /* Wakeup frame status */
+	#define WFCR_PMES		BIT(15) /* PME pin status */
+#define P0_PSCR		(0x0C)
+	#define PSCR_PS_MASK		(0xFFF0)
+	#define PSCR_PS_D0		(0)
+	#define PSCR_PS_D1		BIT(0)
+	#define PSCR_PS_D2		BIT(1)
+	#define PSCR_FPS		BIT(3) /* Enable fiber mode PS */
+	#define PSCR_SWPS		BIT(4) /* Enable software */
+						 /* PS control */
+	#define PSCR_WOLPS		BIT(5) /* Enable WOL PS */
+	#define PSCR_SWWOL		BIT(6) /* Enable software select */
+						 /* WOL PS */
+	#define PSCR_PHYOSC		BIT(7) /* Internal PHY OSC control */
+	#define PSCR_FOFEF		BIT(8) /* Force PHY generate FEF */
+	#define PSCR_FOF		BIT(9) /* Force PHY in fiber mode */
+	#define PSCR_PHYPD		BIT(10) /* PHY power down. */
+						  /* Active high */
+	#define PSCR_PHYRST		BIT(11) /* PHY reset signal. */
+						  /* Active low */
+	#define PSCR_PHYCSIL		BIT(12) /* PHY cable energy detect */
+	#define PSCR_PHYCOFF		BIT(13) /* PHY cable off */
+	#define PSCR_PHYLINK		BIT(14) /* PHY link status */
+	#define PSCR_EEPOK		BIT(15) /* EEPROM load complete */
+#define P0_MACCR	(0x0E)
+	#define MACCR_RXEN		BIT(0) /* Enable RX */
+	#define MACCR_DUPLEX_FULL	BIT(1) /* 1: Full, 0: Half */
+	#define MACCR_SPEED_100		BIT(2) /* 1: 100Mbps, 0: 10Mbps */
+	#define MACCR_RXFC_ENABLE	BIT(3)
+	#define MACCR_RXFC_MASK		0xFFF7
+	#define MACCR_TXFC_ENABLE	BIT(4)
+	#define MACCR_TXFC_MASK		0xFFEF
+	#define MACCR_PSI		BIT(6) /* Software Cable-Off */
+					       /* Power Saving Interrupt */
+	#define MACCR_PF		BIT(7)
+	#define MACCR_PMM_BITS		8
+	#define MACCR_PMM_MASK		(0x1F00)
+	#define MACCR_PMM_RESET		BIT(8)
+	#define MACCR_PMM_WAIT		(2 << 8)
+	#define MACCR_PMM_READY		(3 << 8)
+	#define MACCR_PMM_D1		(4 << 8)
+	#define MACCR_PMM_D2		(5 << 8)
+	#define MACCR_PMM_WAKE		(7 << 8)
+	#define MACCR_PMM_D1_WAKE	(8 << 8)
+	#define MACCR_PMM_D2_WAKE	(9 << 8)
+	#define MACCR_PMM_SLEEP		(10 << 8)
+	#define MACCR_PMM_PHY_RESET	(11 << 8)
+	#define MACCR_PMM_SOFT_D1	(16 << 8)
+	#define MACCR_PMM_SOFT_D2	(17 << 8)
+#define P0_TFBFCR	(0x10)
+	#define TFBFCR_SCHE_FREE_PAGE	0xE07F
+	#define TFBFCR_FREE_PAGE_BITS	0x07
+	#define TFBFCR_FREE_PAGE_LATCH	BIT(6)
+	#define TFBFCR_SET_FREE_PAGE(x)	(((x) & 0x3F) << TFBFCR_FREE_PAGE_BITS)
+	#define TFBFCR_TX_PAGE_SET	BIT(13)
+	#define TFBFCR_MANU_ENTX	BIT(15)
+	#define TX_FREEBUF_MASK		0x003F
+	#define TX_DPTSTART		0x4000
+
+#define P0_TSNR		(0x12)
+	#define TXNR_TXB_ERR		BIT(5)
+	#define TXNR_TXB_IDLE		BIT(6)
+	#define TSNR_PKT_CNT(x)		(((x) & 0x3F) << 8)
+	#define TXNR_TXB_REINIT		BIT(14)
+	#define TSNR_TXB_START		BIT(15)
+#define P0_RTDPR	(0x14)
+#define P0_RXBCR1	(0x16)
+	#define RXBCR1_RXB_DISCARD	BIT(14)
+	#define RXBCR1_RXB_START	BIT(15)
+#define P0_RXBCR2	(0x18)
+	#define RXBCR2_PKT_MASK		(0xFF)
+	#define RXBCR2_RXPC_MASK	(0x7F)
+	#define RXBCR2_RXB_READY	BIT(13)
+	#define RXBCR2_RXB_IDLE		BIT(14)
+	#define RXBCR2_RXB_REINIT	BIT(15)
+#define P0_RTWCR	(0x1A)
+	#define RTWCR_RXWC_MASK		(0x3FFF)
+	#define RTWCR_RX_LATCH		BIT(15)
+#define P0_RCPHR	(0x1C)
+
+	/* Definition of PAGE1 */
+#define P1_RPPER	(0x22)
+	#define RPPER_RXEN		BIT(0)
+#define P1_MRCR		(0x28)
+#define P1_MDR		(0x2A)
+#define P1_RMPR		(0x2C)
+#define P1_TMPR		(0x2E)
+#define P1_RXBSPCR	(0x30)
+	#define RXBSPCR_STUF_WORD_CNT(x)	(((x) & 0x7000) >> 12)
+	#define RXBSPCR_STUF_ENABLE		BIT(15)
+#define P1_MCR		(0x32)
+	#define MCR_SBP			BIT(8)
+	#define MCR_SM			BIT(9)
+	#define MCR_CRCENLAN		BIT(11)
+	#define MCR_STP			BIT(12)
+	/* Definition of PAGE2 */
+#define P2_CIR		(0x42)
+#define P2_PCR		(0x44)
+	#define PCR_POLL_EN		BIT(0)
+	#define PCR_POLL_FLOWCTRL	BIT(1)
+	#define PCR_POLL_BMCR		BIT(2)
+	#define PCR_PHYID(x)		((x) << 8)
+#define P2_PHYSR	(0x46)
+#define P2_MDIODR	(0x48)
+#define P2_MDIOCR	(0x4A)
+	#define MDIOCR_RADDR(x)		((x) & 0x1F)
+	#define MDIOCR_FADDR(x)		(((x) & 0x1F) << 8)
+	#define MDIOCR_VALID		BIT(13)
+	#define MDIOCR_READ		BIT(14)
+	#define MDIOCR_WRITE		BIT(15)
+#define P2_LCR0		(0x4C)
+	#define LCR_LED0_EN		BIT(0)
+	#define LCR_LED0_100MODE	BIT(1)
+	#define LCR_LED0_DUPLEX		BIT(2)
+	#define LCR_LED0_LINK		BIT(3)
+	#define LCR_LED0_ACT		BIT(4)
+	#define LCR_LED0_COL		BIT(5)
+	#define LCR_LED0_10MODE		BIT(6)
+	#define LCR_LED0_DUPCOL		BIT(7)
+	#define LCR_LED1_EN		BIT(8)
+	#define LCR_LED1_100MODE	BIT(9)
+	#define LCR_LED1_DUPLEX		BIT(10)
+	#define LCR_LED1_LINK		BIT(11)
+	#define LCR_LED1_ACT		BIT(12)
+	#define LCR_LED1_COL		BIT(13)
+	#define LCR_LED1_10MODE		BIT(14)
+	#define LCR_LED1_DUPCOL		BIT(15)
+#define P2_LCR1		(0x4E)
+	#define LCR_LED2_MASK		(0xFF00)
+	#define LCR_LED2_EN		BIT(0)
+	#define LCR_LED2_100MODE	BIT(1)
+	#define LCR_LED2_DUPLEX		BIT(2)
+	#define LCR_LED2_LINK		BIT(3)
+	#define LCR_LED2_ACT		BIT(4)
+	#define LCR_LED2_COL		BIT(5)
+	#define LCR_LED2_10MODE		BIT(6)
+	#define LCR_LED2_DUPCOL		BIT(7)
+#define P2_IPGCR	(0x50)
+#define P2_CRIR		(0x52)
+#define P2_FLHWCR	(0x54)
+#define P2_RXCR		(0x56)
+	#define RXCR_PRO		BIT(0)
+	#define RXCR_AMALL		BIT(1)
+	#define RXCR_SEP		BIT(2)
+	#define RXCR_AB			BIT(3)
+	#define RXCR_AM			BIT(4)
+	#define RXCR_AP			BIT(5)
+	#define RXCR_ARP		BIT(6)
+#define P2_JLCR		(0x58)
+#define P2_MPLR		(0x5C)
+
+	/* Definition of PAGE3 */
+#define P3_MACASR0	(0x62)
+	#define P3_MACASR(x)		(P3_MACASR0 + 2 * (x))
+	#define MACASR_LOWBYTE_MASK	0x00FF
+	#define MACASR_HIGH_BITS	0x08
+#define P3_MACASR1	(0x64)
+#define P3_MACASR2	(0x66)
+#define P3_MFAR01	(0x68)
+#define P3_MFAR_BASE	(0x68)
+	#define P3_MFAR(x)		(P3_MFAR_BASE + 2 * (x))
+
+#define P3_MFAR23	(0x6A)
+#define P3_MFAR45	(0x6C)
+#define P3_MFAR67	(0x6E)
+#define P3_VID0FR	(0x70)
+#define P3_VID1FR	(0x72)
+#define P3_EECSR	(0x74)
+#define P3_EEDR		(0x76)
+#define P3_EECR		(0x78)
+	#define EECR_ADDR_MASK		(0x00FF)
+	#define EECR_READ_ACT		BIT(8)
+	#define EECR_WRITE_ACT		BIT(9)
+	#define EECR_WRITE_DISABLE	BIT(10)
+	#define EECR_WRITE_ENABLE	BIT(11)
+	#define EECR_EE_READY		BIT(13)
+	#define EECR_RELOAD		BIT(14)
+	#define EECR_RESET		BIT(15)
+#define P3_TPCR		(0x7A)
+	#define TPCR_PATT_MASK		(0xFF)
+	#define TPCR_RAND_PKT_EN	BIT(14)
+	#define TPCR_FIXED_PKT_EN	BIT(15)
+#define P3_TPLR		(0x7C)
+	/* Definition of PAGE4 */
+#define P4_SPICR	(0x8A)
+	#define SPICR_RCEN		BIT(0)
+	#define SPICR_QCEN		BIT(1)
+	#define SPICR_RBRE		BIT(3)
+	#define SPICR_PMM		BIT(4)
+	#define SPICR_LOOPBACK		BIT(8)
+	#define SPICR_CORE_RES_CLR	BIT(10)
+	#define SPICR_SPI_RES_CLR	BIT(11)
+#define P4_SPIISMR	(0x8C)
+
+#define P4_COERCR0	(0x92)
+	#define COERCR0_RXIPCE		BIT(0)
+	#define COERCR0_RXIPVE		BIT(1)
+	#define COERCR0_RXV6PE		BIT(2)
+	#define COERCR0_RXTCPE		BIT(3)
+	#define COERCR0_RXUDPE		BIT(4)
+	#define COERCR0_RXICMP		BIT(5)
+	#define COERCR0_RXIGMP		BIT(6)
+	#define COERCR0_RXICV6		BIT(7)
+
+	#define COERCR0_RXTCPV6		BIT(8)
+	#define COERCR0_RXUDPV6		BIT(9)
+	#define COERCR0_RXICMV6		BIT(10)
+	#define COERCR0_RXIGMV6		BIT(11)
+	#define COERCR0_RXICV6V6	BIT(12)
+
+	#define COERCR0_DEFAULT		(COERCR0_RXIPCE | COERCR0_RXV6PE | \
+					 COERCR0_RXTCPE | COERCR0_RXUDPE | \
+					 COERCR0_RXTCPV6 | COERCR0_RXUDPV6)
+#define P4_COERCR1	(0x94)
+	#define COERCR1_IPCEDP		BIT(0)
+	#define COERCR1_IPVEDP		BIT(1)
+	#define COERCR1_V6VEDP		BIT(2)
+	#define COERCR1_TCPEDP		BIT(3)
+	#define COERCR1_UDPEDP		BIT(4)
+	#define COERCR1_ICMPDP		BIT(5)
+	#define COERCR1_IGMPDP		BIT(6)
+	#define COERCR1_ICV6DP		BIT(7)
+	#define COERCR1_RX64TE		BIT(8)
+	#define COERCR1_RXPPPE		BIT(9)
+	#define COERCR1_TCP6DP		BIT(10)
+	#define COERCR1_UDP6DP		BIT(11)
+	#define COERCR1_IC6DP		BIT(12)
+	#define COERCR1_IG6DP		BIT(13)
+	#define COERCR1_ICV66DP		BIT(14)
+	#define COERCR1_RPCE		BIT(15)
+
+	#define COERCR1_DEFAULT		(COERCR1_RXPPPE)
+
+#define P4_COETCR0	(0x96)
+	#define COETCR0_TXIP		BIT(0)
+	#define COETCR0_TXTCP		BIT(1)
+	#define COETCR0_TXUDP		BIT(2)
+	#define COETCR0_TXICMP		BIT(3)
+	#define COETCR0_TXIGMP		BIT(4)
+	#define COETCR0_TXICV6		BIT(5)
+	#define COETCR0_TXTCPV6		BIT(8)
+	#define COETCR0_TXUDPV6		BIT(9)
+	#define COETCR0_TXICMV6		BIT(10)
+	#define COETCR0_TXIGMV6		BIT(11)
+	#define COETCR0_TXICV6V6	BIT(12)
+
+	#define COETCR0_DEFAULT		(COETCR0_TXIP | COETCR0_TXTCP | \
+					 COETCR0_TXUDP | COETCR0_TXTCPV6 | \
+					 COETCR0_TXUDPV6)
+#define P4_COETCR1	(0x98)
+	#define COETCR1_TX64TE		BIT(0)
+	#define COETCR1_TXPPPE		BIT(1)
+
+#define P4_COECEDR	(0x9A)
+#define P4_L2CECR	(0x9C)
+
+	/* Definition of PAGE5 */
+#define P5_WFTR		(0xA2)
+	#define WFTR_2MS		(0x01)
+	#define WFTR_4MS		(0x02)
+	#define WFTR_8MS		(0x03)
+	#define WFTR_16MS		(0x04)
+	#define WFTR_32MS		(0x05)
+	#define WFTR_64MS		(0x06)
+	#define WFTR_128MS		(0x07)
+	#define WFTR_256MS		(0x08)
+	#define WFTR_512MS		(0x09)
+	#define WFTR_1024MS		(0x0A)
+	#define WFTR_2048MS		(0x0B)
+	#define WFTR_4096MS		(0x0C)
+	#define WFTR_8192MS		(0x0D)
+	#define WFTR_16384MS		(0x0E)
+	#define WFTR_32768MS		(0x0F)
+#define P5_WFCCR	(0xA4)
+#define P5_WFCR03	(0xA6)
+	#define WFCR03_F0_EN		BIT(0)
+	#define WFCR03_F1_EN		BIT(4)
+	#define WFCR03_F2_EN		BIT(8)
+	#define WFCR03_F3_EN		BIT(12)
+#define P5_WFCR47	(0xA8)
+	#define WFCR47_F4_EN		BIT(0)
+	#define WFCR47_F5_EN		BIT(4)
+	#define WFCR47_F6_EN		BIT(8)
+	#define WFCR47_F7_EN		BIT(12)
+#define P5_WF0BMR0	(0xAA)
+#define P5_WF0BMR1	(0xAC)
+#define P5_WF0CR	(0xAE)
+#define P5_WF0OBR	(0xB0)
+#define P5_WF1BMR0	(0xB2)
+#define P5_WF1BMR1	(0xB4)
+#define P5_WF1CR	(0xB6)
+#define P5_WF1OBR	(0xB8)
+#define P5_WF2BMR0	(0xBA)
+#define P5_WF2BMR1	(0xBC)
+
+	/* Definition of PAGE6 */
+#define P6_WF2CR	(0xC2)
+#define P6_WF2OBR	(0xC4)
+#define P6_WF3BMR0	(0xC6)
+#define P6_WF3BMR1	(0xC8)
+#define P6_WF3CR	(0xCA)
+#define P6_WF3OBR	(0xCC)
+#define P6_WF4BMR0	(0xCE)
+#define P6_WF4BMR1	(0xD0)
+#define P6_WF4CR	(0xD2)
+#define P6_WF4OBR	(0xD4)
+#define P6_WF5BMR0	(0xD6)
+#define P6_WF5BMR1	(0xD8)
+#define P6_WF5CR	(0xDA)
+#define P6_WF5OBR	(0xDC)
+
+/* Definition of PAGE7 */
+#define P7_WF6BMR0	(0xE2)
+#define P7_WF6BMR1	(0xE4)
+#define P7_WF6CR	(0xE6)
+#define P7_WF6OBR	(0xE8)
+#define P7_WF7BMR0	(0xEA)
+#define P7_WF7BMR1	(0xEC)
+#define P7_WF7CR	(0xEE)
+#define P7_WF7OBR	(0xF0)
+#define P7_WFR01	(0xF2)
+#define P7_WFR23	(0xF4)
+#define P7_WFR45	(0xF6)
+#define P7_WFR67	(0xF8)
+#define P7_WFPC0	(0xFA)
+#define P7_WFPC1	(0xFC)
+
+/* Tx headers structure */
+struct tx_sop_header {
+	/* bit 15-11: flags, bit 10-0: packet length */
+	u16 flags_len;
+	/* bit 15-11: sequence number, bit 11-0: packet length bar */
+	u16 seq_lenbar;
+};
+
+struct tx_segment_header {
+	/* bit 15-14: flags, bit 13-11: segment number */
+	/* bit 10-0: segment length */
+	u16 flags_seqnum_seglen;
+	/* bit 15-14: end offset, bit 13-11: start offset */
+	/* bit 10-0: segment length bar */
+	u16 eo_so_seglenbar;
+};
+
+struct tx_eop_header {
+	/* bit 15-11: sequence number, bit 10-0: packet length */
+	u16 seq_len;
+	/* bit 15-11: sequence number bar, bit 10-0: packet length bar */
+	u16 seqbar_lenbar;
+};
+
+struct tx_pkt_info {
+	struct_group(tx_overhead,
+		struct tx_sop_header sop;
+		struct tx_segment_header seg;
+	);
+	struct tx_eop_header eop;
+	u16 pkt_len;
+	u16 seq_num;
+};
+
+/* Rx headers structure */
+struct rx_header {
+	u16 flags_len;
+	u16 seq_lenbar;
+	u16 flags;
+};
+
+extern unsigned long ax88796c_no_regs_mask[];
+
+#endif /* #ifndef _AX88796C_MAIN_H */
diff --git a/drivers/net/ethernet/asix/ax88796c_spi.c b/drivers/net/ethernet/asix/ax88796c_spi.c
new file mode 100644
index 000000000000..0710e716d682
--- /dev/null
+++ b/drivers/net/ethernet/asix/ax88796c_spi.c
@@ -0,0 +1,115 @@
+// SPDX-License-Identifier: GPL-2.0-only
+/*
+ * Copyright (c) 2010 ASIX Electronics Corporation
+ * Copyright (c) 2020 Samsung Electronics Co., Ltd.
+ *
+ * ASIX AX88796C SPI Fast Ethernet Linux driver
+ */
+
+#define pr_fmt(fmt)	"ax88796c: " fmt
+
+#include <linux/string.h>
+#include <linux/spi/spi.h>
+
+#include "ax88796c_spi.h"
+
+const u8 ax88796c_rx_cmd_buf[5] = {AX_SPICMD_READ_RXQ, 0xFF, 0xFF, 0xFF, 0xFF};
+const u8 ax88796c_tx_cmd_buf[4] = {AX_SPICMD_WRITE_TXQ, 0xFF, 0xFF, 0xFF};
+
+/* driver bus management functions */
+int axspi_wakeup(struct axspi_data *ax_spi)
+{
+	int ret;
+
+	ax_spi->cmd_buf[0] = AX_SPICMD_EXIT_PWD;	/* OP */
+	ret = spi_write(ax_spi->spi, ax_spi->cmd_buf, 1);
+	if (ret)
+		dev_err(&ax_spi->spi->dev, "%s() failed: ret = %d\n", __func__, ret);
+	return ret;
+}
+
+int axspi_read_status(struct axspi_data *ax_spi, struct spi_status *status)
+{
+	int ret;
+
+	/* OP */
+	ax_spi->cmd_buf[0] = AX_SPICMD_READ_STATUS;
+	ret = spi_write_then_read(ax_spi->spi, ax_spi->cmd_buf, 1, (u8 *)status, 3);
+	if (ret)
+		dev_err(&ax_spi->spi->dev, "%s() failed: ret = %d\n", __func__, ret);
+	else
+		le16_to_cpus(&status->isr);
+
+	return ret;
+}
+
+int axspi_read_rxq(struct axspi_data *ax_spi, void *data, int len)
+{
+	struct spi_transfer *xfer = ax_spi->spi_rx_xfer;
+	int ret;
+
+	memcpy(ax_spi->cmd_buf, ax88796c_rx_cmd_buf, 5);
+
+	xfer->tx_buf = ax_spi->cmd_buf;
+	xfer->rx_buf = NULL;
+	xfer->len = ax_spi->comp ? 2 : 5;
+	xfer->bits_per_word = 8;
+	spi_message_add_tail(xfer, &ax_spi->rx_msg);
+
+	xfer++;
+	xfer->rx_buf = data;
+	xfer->tx_buf = NULL;
+	xfer->len = len;
+	xfer->bits_per_word = 8;
+	spi_message_add_tail(xfer, &ax_spi->rx_msg);
+	ret = spi_sync(ax_spi->spi, &ax_spi->rx_msg);
+	if (ret)
+		dev_err(&ax_spi->spi->dev, "%s() failed: ret = %d\n", __func__, ret);
+
+	return ret;
+}
+
+int axspi_write_txq(const struct axspi_data *ax_spi, void *data, int len)
+{
+	return spi_write(ax_spi->spi, data, len);
+}
+
+u16 axspi_read_reg(struct axspi_data *ax_spi, u8 reg)
+{
+	int ret;
+	int len = ax_spi->comp ? 3 : 4;
+
+	ax_spi->cmd_buf[0] = 0x03;	/* OP code read register */
+	ax_spi->cmd_buf[1] = reg;	/* register address */
+	ax_spi->cmd_buf[2] = 0xFF;	/* dumy cycle */
+	ax_spi->cmd_buf[3] = 0xFF;	/* dumy cycle */
+	ret = spi_write_then_read(ax_spi->spi,
+				  ax_spi->cmd_buf, len,
+				  ax_spi->rx_buf, 2);
+	if (ret) {
+		dev_err(&ax_spi->spi->dev,
+			"%s() failed: ret = %d\n", __func__, ret);
+		return 0xFFFF;
+	}
+
+	le16_to_cpus((u16 *)ax_spi->rx_buf);
+
+	return *(u16 *)ax_spi->rx_buf;
+}
+
+int axspi_write_reg(struct axspi_data *ax_spi, u8 reg, u16 value)
+{
+	int ret;
+
+	memset(ax_spi->cmd_buf, 0, sizeof(ax_spi->cmd_buf));
+	ax_spi->cmd_buf[0] = AX_SPICMD_WRITE_REG;	/* OP code read register */
+	ax_spi->cmd_buf[1] = reg;			/* register address */
+	ax_spi->cmd_buf[2] = value;
+	ax_spi->cmd_buf[3] = value >> 8;
+
+	ret = spi_write(ax_spi->spi, ax_spi->cmd_buf, 4);
+	if (ret)
+		dev_err(&ax_spi->spi->dev, "%s() failed: ret = %d\n", __func__, ret);
+	return ret;
+}
+
diff --git a/drivers/net/ethernet/asix/ax88796c_spi.h b/drivers/net/ethernet/asix/ax88796c_spi.h
new file mode 100644
index 000000000000..5bcf91f603fb
--- /dev/null
+++ b/drivers/net/ethernet/asix/ax88796c_spi.h
@@ -0,0 +1,69 @@
+/* SPDX-License-Identifier: GPL-2.0-only */
+/*
+ * Copyright (c) 2010 ASIX Electronics Corporation
+ * Copyright (c) 2020 Samsung Electronics Co., Ltd.
+ *
+ * ASIX AX88796C SPI Fast Ethernet Linux driver
+ */
+
+#ifndef _AX88796C_SPI_H
+#define _AX88796C_SPI_H
+
+#include <linux/spi/spi.h>
+#include <linux/types.h>
+
+/* Definition of SPI command */
+#define AX_SPICMD_WRITE_TXQ		0x02
+#define AX_SPICMD_READ_REG		0x03
+#define AX_SPICMD_READ_STATUS		0x05
+#define AX_SPICMD_READ_RXQ		0x0B
+#define AX_SPICMD_BIDIR_WRQ		0xB2
+#define AX_SPICMD_WRITE_REG		0xD8
+#define AX_SPICMD_EXIT_PWD		0xAB
+
+extern const u8 ax88796c_rx_cmd_buf[];
+extern const u8 ax88796c_tx_cmd_buf[];
+
+struct axspi_data {
+	struct spi_device	*spi;
+	struct spi_message	rx_msg;
+	struct spi_transfer	spi_rx_xfer[2];
+	u8			cmd_buf[6];
+	u8			rx_buf[6];
+	u8			comp;
+};
+
+struct spi_status {
+	u16 isr;
+	u8 status;
+#	define AX_STATUS_READY		0x80
+};
+
+int axspi_read_rxq(struct axspi_data *ax_spi, void *data, int len);
+int axspi_write_txq(const struct axspi_data *ax_spi, void *data, int len);
+u16 axspi_read_reg(struct axspi_data *ax_spi, u8 reg);
+int axspi_write_reg(struct axspi_data *ax_spi, u8 reg, u16 value);
+int axspi_read_status(struct axspi_data *ax_spi, struct spi_status *status);
+int axspi_wakeup(struct axspi_data *ax_spi);
+
+static inline u16 AX_READ(struct axspi_data *ax_spi, u8 offset)
+{
+	return axspi_read_reg(ax_spi, offset);
+}
+
+static inline int AX_WRITE(struct axspi_data *ax_spi, u16 value, u8 offset)
+{
+	return axspi_write_reg(ax_spi, offset, value);
+}
+
+static inline int AX_READ_STATUS(struct axspi_data *ax_spi,
+				 struct spi_status *status)
+{
+	return axspi_read_status(ax_spi, status);
+}
+
+static inline int AX_WAKEUP(struct axspi_data *ax_spi)
+{
+	return axspi_wakeup(ax_spi);
+}
+#endif
diff --git a/drivers/net/ethernet/atheros/Kconfig b/drivers/net/ethernet/atheros/Kconfig
index e05b25675333..bec5cdf8d1da 100644
--- a/drivers/net/ethernet/atheros/Kconfig
+++ b/drivers/net/ethernet/atheros/Kconfig
@@ -1,3 +1,4 @@
+# SPDX-License-Identifier: GPL-2.0-only
 #
 # Atheros device configuration
 #
@@ -5,8 +6,8 @@
 config NET_VENDOR_ATHEROS
 	bool "Atheros devices"
 	default y
-	depends on PCI
-	---help---
+	depends on PCI || ATH79 || COMPILE_TEST
+	help
 	  If you have a network (Ethernet) card belonging to this class, say Y.
 
 	  Note that the answer to this question doesn't directly affect the
@@ -16,12 +17,21 @@ config NET_VENDOR_ATHEROS
 
 if NET_VENDOR_ATHEROS
 
+config AG71XX
+	tristate "Atheros AR7XXX/AR9XXX built-in ethernet mac support"
+	depends on ATH79 || COMPILE_TEST
+	select PHYLINK
+	imply NET_SELFTESTS
+	help
+	  If you wish to compile a kernel for AR7XXX/91XXX and enable
+	  ethernet support, then you should always answer Y to this.
+
 config ATL2
 	tristate "Atheros L2 Fast Ethernet support"
 	depends on PCI
 	select CRC32
 	select MII
-	---help---
+	help
 	  This driver supports the Atheros L2 fast ethernet adapter.
 
 	  To compile this driver as a module, choose M here.  The module
@@ -32,7 +42,7 @@ config ATL1
 	depends on PCI
 	select CRC32
 	select MII
-	---help---
+	help
 	  This driver supports the Atheros/Attansic L1 gigabit ethernet
 	  adapter.
 
@@ -44,7 +54,7 @@ config ATL1E
 	depends on PCI
 	select CRC32
 	select MII
-	---help---
+	help
 	  This driver supports the Atheros L1E gigabit ethernet adapter.
 
 	  To compile this driver as a module, choose M here.  The module
@@ -55,7 +65,7 @@ config ATL1C
 	depends on PCI
 	select CRC32
 	select MII
-	---help---
+	help
 	  This driver supports the Atheros L1C gigabit ethernet adapter.
 
 	  To compile this driver as a module, choose M here.  The module
diff --git a/drivers/net/ethernet/atheros/Makefile b/drivers/net/ethernet/atheros/Makefile
index 5cf1c65bbce9..aca696cb6425 100644
--- a/drivers/net/ethernet/atheros/Makefile
+++ b/drivers/net/ethernet/atheros/Makefile
@@ -1,7 +1,9 @@
+# SPDX-License-Identifier: GPL-2.0
 #
 # Makefile for the Atheros network device drivers.
 #
 
+obj-$(CONFIG_AG71XX) += ag71xx.o
 obj-$(CONFIG_ATL1) += atlx/
 obj-$(CONFIG_ATL2) += atlx/
 obj-$(CONFIG_ATL1E) += atl1e/
diff --git a/drivers/net/ethernet/atheros/ag71xx.c b/drivers/net/ethernet/atheros/ag71xx.c
new file mode 100644
index 000000000000..cbc730c7cff2
--- /dev/null
+++ b/drivers/net/ethernet/atheros/ag71xx.c
@@ -0,0 +1,2035 @@
+// SPDX-License-Identifier: GPL-2.0
+/*  Atheros AR71xx built-in ethernet mac driver
+ *
+ *  Copyright (C) 2019 Oleksij Rempel <o.rempel@pengutronix.de>
+ *
+ *  List of authors contributed to this driver before mainlining:
+ *  Alexander Couzens <lynxis@fe80.eu>
+ *  Christian Lamparter <chunkeey@gmail.com>
+ *  Chuanhong Guo <gch981213@gmail.com>
+ *  Daniel F. Dickinson <cshored@thecshore.com>
+ *  David Bauer <mail@david-bauer.net>
+ *  Felix Fietkau <nbd@nbd.name>
+ *  Gabor Juhos <juhosg@freemail.hu>
+ *  Hauke Mehrtens <hauke@hauke-m.de>
+ *  Johann Neuhauser <johann@it-neuhauser.de>
+ *  John Crispin <john@phrozen.org>
+ *  Jo-Philipp Wich <jo@mein.io>
+ *  Koen Vandeputte <koen.vandeputte@ncentric.com>
+ *  Lucian Cristian <lucian.cristian@gmail.com>
+ *  Matt Merhar <mattmerhar@protonmail.com>
+ *  Milan Krstic <milan.krstic@gmail.com>
+ *  Petr Štetiar <ynezz@true.cz>
+ *  Rosen Penev <rosenp@gmail.com>
+ *  Stephen Walker <stephendwalker+github@gmail.com>
+ *  Vittorio Gambaletta <openwrt@vittgam.net>
+ *  Weijie Gao <hackpascal@gmail.com>
+ *  Imre Kaloz <kaloz@openwrt.org>
+ */
+
+#include <linux/if_vlan.h>
+#include <linux/mfd/syscon.h>
+#include <linux/of.h>
+#include <linux/of_mdio.h>
+#include <linux/of_net.h>
+#include <linux/platform_device.h>
+#include <linux/phylink.h>
+#include <linux/regmap.h>
+#include <linux/reset.h>
+#include <linux/clk.h>
+#include <linux/io.h>
+#include <net/selftests.h>
+
+/* For our NAPI weight bigger does *NOT* mean better - it means more
+ * D-cache misses and lots more wasted cycles than we'll ever
+ * possibly gain from saving instructions.
+ */
+#define AG71XX_NAPI_WEIGHT	32
+#define AG71XX_OOM_REFILL	(1 + HZ / 10)
+
+#define AG71XX_INT_ERR	(AG71XX_INT_RX_BE | AG71XX_INT_TX_BE)
+#define AG71XX_INT_TX	(AG71XX_INT_TX_PS)
+#define AG71XX_INT_RX	(AG71XX_INT_RX_PR | AG71XX_INT_RX_OF)
+
+#define AG71XX_INT_POLL	(AG71XX_INT_RX | AG71XX_INT_TX)
+#define AG71XX_INT_INIT	(AG71XX_INT_ERR | AG71XX_INT_POLL)
+
+#define AG71XX_TX_MTU_LEN	1540
+
+#define AG71XX_TX_RING_SPLIT		512
+#define AG71XX_TX_RING_DS_PER_PKT	DIV_ROUND_UP(AG71XX_TX_MTU_LEN, \
+						     AG71XX_TX_RING_SPLIT)
+#define AG71XX_TX_RING_SIZE_DEFAULT	128
+#define AG71XX_RX_RING_SIZE_DEFAULT	256
+
+#define AG71XX_MDIO_RETRY	1000
+#define AG71XX_MDIO_DELAY	5
+#define AG71XX_MDIO_MAX_CLK	5000000
+
+/* Register offsets */
+#define AG71XX_REG_MAC_CFG1	0x0000
+#define MAC_CFG1_TXE		BIT(0)	/* Tx Enable */
+#define MAC_CFG1_STX		BIT(1)	/* Synchronize Tx Enable */
+#define MAC_CFG1_RXE		BIT(2)	/* Rx Enable */
+#define MAC_CFG1_SRX		BIT(3)	/* Synchronize Rx Enable */
+#define MAC_CFG1_TFC		BIT(4)	/* Tx Flow Control Enable */
+#define MAC_CFG1_RFC		BIT(5)	/* Rx Flow Control Enable */
+#define MAC_CFG1_SR		BIT(31)	/* Soft Reset */
+#define MAC_CFG1_INIT	(MAC_CFG1_RXE | MAC_CFG1_TXE | \
+			 MAC_CFG1_SRX | MAC_CFG1_STX)
+
+#define AG71XX_REG_MAC_CFG2	0x0004
+#define MAC_CFG2_FDX		BIT(0)
+#define MAC_CFG2_PAD_CRC_EN	BIT(2)
+#define MAC_CFG2_LEN_CHECK	BIT(4)
+#define MAC_CFG2_IF_1000	BIT(9)
+#define MAC_CFG2_IF_10_100	BIT(8)
+
+#define AG71XX_REG_MAC_MFL	0x0010
+
+#define AG71XX_REG_MII_CFG	0x0020
+#define MII_CFG_CLK_DIV_4	0
+#define MII_CFG_CLK_DIV_6	2
+#define MII_CFG_CLK_DIV_8	3
+#define MII_CFG_CLK_DIV_10	4
+#define MII_CFG_CLK_DIV_14	5
+#define MII_CFG_CLK_DIV_20	6
+#define MII_CFG_CLK_DIV_28	7
+#define MII_CFG_CLK_DIV_34	8
+#define MII_CFG_CLK_DIV_42	9
+#define MII_CFG_CLK_DIV_50	10
+#define MII_CFG_CLK_DIV_58	11
+#define MII_CFG_CLK_DIV_66	12
+#define MII_CFG_CLK_DIV_74	13
+#define MII_CFG_CLK_DIV_82	14
+#define MII_CFG_CLK_DIV_98	15
+#define MII_CFG_RESET		BIT(31)
+
+#define AG71XX_REG_MII_CMD	0x0024
+#define MII_CMD_READ		BIT(0)
+
+#define AG71XX_REG_MII_ADDR	0x0028
+#define MII_ADDR_SHIFT		8
+
+#define AG71XX_REG_MII_CTRL	0x002c
+#define AG71XX_REG_MII_STATUS	0x0030
+#define AG71XX_REG_MII_IND	0x0034
+#define MII_IND_BUSY		BIT(0)
+#define MII_IND_INVALID		BIT(2)
+
+#define AG71XX_REG_MAC_IFCTL	0x0038
+#define MAC_IFCTL_SPEED		BIT(16)
+
+#define AG71XX_REG_MAC_ADDR1	0x0040
+#define AG71XX_REG_MAC_ADDR2	0x0044
+#define AG71XX_REG_FIFO_CFG0	0x0048
+#define FIFO_CFG0_WTM		BIT(0)	/* Watermark Module */
+#define FIFO_CFG0_RXS		BIT(1)	/* Rx System Module */
+#define FIFO_CFG0_RXF		BIT(2)	/* Rx Fabric Module */
+#define FIFO_CFG0_TXS		BIT(3)	/* Tx System Module */
+#define FIFO_CFG0_TXF		BIT(4)	/* Tx Fabric Module */
+#define FIFO_CFG0_ALL	(FIFO_CFG0_WTM | FIFO_CFG0_RXS | FIFO_CFG0_RXF \
+			| FIFO_CFG0_TXS | FIFO_CFG0_TXF)
+#define FIFO_CFG0_INIT	(FIFO_CFG0_ALL << FIFO_CFG0_ENABLE_SHIFT)
+
+#define FIFO_CFG0_ENABLE_SHIFT	8
+
+#define AG71XX_REG_FIFO_CFG1	0x004c
+#define AG71XX_REG_FIFO_CFG2	0x0050
+#define AG71XX_REG_FIFO_CFG3	0x0054
+#define AG71XX_REG_FIFO_CFG4	0x0058
+#define FIFO_CFG4_DE		BIT(0)	/* Drop Event */
+#define FIFO_CFG4_DV		BIT(1)	/* RX_DV Event */
+#define FIFO_CFG4_FC		BIT(2)	/* False Carrier */
+#define FIFO_CFG4_CE		BIT(3)	/* Code Error */
+#define FIFO_CFG4_CR		BIT(4)	/* CRC error */
+#define FIFO_CFG4_LM		BIT(5)	/* Length Mismatch */
+#define FIFO_CFG4_LO		BIT(6)	/* Length out of range */
+#define FIFO_CFG4_OK		BIT(7)	/* Packet is OK */
+#define FIFO_CFG4_MC		BIT(8)	/* Multicast Packet */
+#define FIFO_CFG4_BC		BIT(9)	/* Broadcast Packet */
+#define FIFO_CFG4_DR		BIT(10)	/* Dribble */
+#define FIFO_CFG4_CF		BIT(11)	/* Control Frame */
+#define FIFO_CFG4_PF		BIT(12)	/* Pause Frame */
+#define FIFO_CFG4_UO		BIT(13)	/* Unsupported Opcode */
+#define FIFO_CFG4_VT		BIT(14)	/* VLAN tag detected */
+#define FIFO_CFG4_LE		BIT(15)	/* Long Event */
+#define FIFO_CFG4_FT		BIT(16)	/* Frame Truncated */
+#define FIFO_CFG4_UC		BIT(17)	/* Unicast Packet */
+#define FIFO_CFG4_INIT	(FIFO_CFG4_DE | FIFO_CFG4_DV | FIFO_CFG4_FC | \
+			 FIFO_CFG4_CE | FIFO_CFG4_CR | FIFO_CFG4_LM | \
+			 FIFO_CFG4_LO | FIFO_CFG4_OK | FIFO_CFG4_MC | \
+			 FIFO_CFG4_BC | FIFO_CFG4_DR | FIFO_CFG4_LE | \
+			 FIFO_CFG4_CF | FIFO_CFG4_PF | FIFO_CFG4_UO | \
+			 FIFO_CFG4_VT)
+
+#define AG71XX_REG_FIFO_CFG5	0x005c
+#define FIFO_CFG5_DE		BIT(0)	/* Drop Event */
+#define FIFO_CFG5_DV		BIT(1)	/* RX_DV Event */
+#define FIFO_CFG5_FC		BIT(2)	/* False Carrier */
+#define FIFO_CFG5_CE		BIT(3)	/* Code Error */
+#define FIFO_CFG5_CR		BIT(4)  /* CRC error */
+#define FIFO_CFG5_LM		BIT(5)	/* Length Mismatch */
+#define FIFO_CFG5_LO		BIT(6)	/* Length Out of Range */
+#define FIFO_CFG5_OK		BIT(7)	/* Packet is OK */
+#define FIFO_CFG5_MC		BIT(8)	/* Multicast Packet */
+#define FIFO_CFG5_BC		BIT(9)	/* Broadcast Packet */
+#define FIFO_CFG5_DR		BIT(10)	/* Dribble */
+#define FIFO_CFG5_CF		BIT(11)	/* Control Frame */
+#define FIFO_CFG5_PF		BIT(12)	/* Pause Frame */
+#define FIFO_CFG5_UO		BIT(13)	/* Unsupported Opcode */
+#define FIFO_CFG5_VT		BIT(14)	/* VLAN tag detected */
+#define FIFO_CFG5_LE		BIT(15)	/* Long Event */
+#define FIFO_CFG5_FT		BIT(16)	/* Frame Truncated */
+#define FIFO_CFG5_UC		BIT(17)	/* Unicast Packet */
+#define FIFO_CFG5_SF		BIT(18)	/* Short Frame */
+#define FIFO_CFG5_BM		BIT(19)	/* Byte Mode */
+#define FIFO_CFG5_INIT	(FIFO_CFG5_DE | FIFO_CFG5_DV | FIFO_CFG5_FC | \
+			 FIFO_CFG5_CE | FIFO_CFG5_LM | FIFO_CFG5_LO | \
+			 FIFO_CFG5_OK | FIFO_CFG5_MC | FIFO_CFG5_BC | \
+			 FIFO_CFG5_DR | FIFO_CFG5_CF | FIFO_CFG5_UO | \
+			 FIFO_CFG5_VT | FIFO_CFG5_LE | FIFO_CFG5_FT | \
+			 FIFO_CFG5_UC | FIFO_CFG5_SF)
+
+#define AG71XX_REG_TX_CTRL	0x0180
+#define TX_CTRL_TXE		BIT(0)	/* Tx Enable */
+
+#define AG71XX_REG_TX_DESC	0x0184
+#define AG71XX_REG_TX_STATUS	0x0188
+#define TX_STATUS_PS		BIT(0)	/* Packet Sent */
+#define TX_STATUS_UR		BIT(1)	/* Tx Underrun */
+#define TX_STATUS_BE		BIT(3)	/* Bus Error */
+
+#define AG71XX_REG_RX_CTRL	0x018c
+#define RX_CTRL_RXE		BIT(0)	/* Rx Enable */
+
+#define AG71XX_DMA_RETRY	10
+#define AG71XX_DMA_DELAY	1
+
+#define AG71XX_REG_RX_DESC	0x0190
+#define AG71XX_REG_RX_STATUS	0x0194
+#define RX_STATUS_PR		BIT(0)	/* Packet Received */
+#define RX_STATUS_OF		BIT(2)	/* Rx Overflow */
+#define RX_STATUS_BE		BIT(3)	/* Bus Error */
+
+#define AG71XX_REG_INT_ENABLE	0x0198
+#define AG71XX_REG_INT_STATUS	0x019c
+#define AG71XX_INT_TX_PS	BIT(0)
+#define AG71XX_INT_TX_UR	BIT(1)
+#define AG71XX_INT_TX_BE	BIT(3)
+#define AG71XX_INT_RX_PR	BIT(4)
+#define AG71XX_INT_RX_OF	BIT(6)
+#define AG71XX_INT_RX_BE	BIT(7)
+
+#define AG71XX_REG_FIFO_DEPTH	0x01a8
+#define AG71XX_REG_RX_SM	0x01b0
+#define AG71XX_REG_TX_SM	0x01b4
+
+#define AG71XX_DEFAULT_MSG_ENABLE	\
+	(NETIF_MSG_DRV			\
+	| NETIF_MSG_PROBE		\
+	| NETIF_MSG_LINK		\
+	| NETIF_MSG_TIMER		\
+	| NETIF_MSG_IFDOWN		\
+	| NETIF_MSG_IFUP		\
+	| NETIF_MSG_RX_ERR		\
+	| NETIF_MSG_TX_ERR)
+
+struct ag71xx_statistic {
+	unsigned short offset;
+	u32 mask;
+	const char name[ETH_GSTRING_LEN];
+};
+
+static const struct ag71xx_statistic ag71xx_statistics[] = {
+	{ 0x0080, GENMASK(17, 0), "Tx/Rx 64 Byte", },
+	{ 0x0084, GENMASK(17, 0), "Tx/Rx 65-127 Byte", },
+	{ 0x0088, GENMASK(17, 0), "Tx/Rx 128-255 Byte", },
+	{ 0x008C, GENMASK(17, 0), "Tx/Rx 256-511 Byte", },
+	{ 0x0090, GENMASK(17, 0), "Tx/Rx 512-1023 Byte", },
+	{ 0x0094, GENMASK(17, 0), "Tx/Rx 1024-1518 Byte", },
+	{ 0x0098, GENMASK(17, 0), "Tx/Rx 1519-1522 Byte VLAN", },
+	{ 0x009C, GENMASK(23, 0), "Rx Byte", },
+	{ 0x00A0, GENMASK(17, 0), "Rx Packet", },
+	{ 0x00A4, GENMASK(11, 0), "Rx FCS Error", },
+	{ 0x00A8, GENMASK(17, 0), "Rx Multicast Packet", },
+	{ 0x00AC, GENMASK(21, 0), "Rx Broadcast Packet", },
+	{ 0x00B0, GENMASK(17, 0), "Rx Control Frame Packet", },
+	{ 0x00B4, GENMASK(11, 0), "Rx Pause Frame Packet", },
+	{ 0x00B8, GENMASK(11, 0), "Rx Unknown OPCode Packet", },
+	{ 0x00BC, GENMASK(11, 0), "Rx Alignment Error", },
+	{ 0x00C0, GENMASK(15, 0), "Rx Frame Length Error", },
+	{ 0x00C4, GENMASK(11, 0), "Rx Code Error", },
+	{ 0x00C8, GENMASK(11, 0), "Rx Carrier Sense Error", },
+	{ 0x00CC, GENMASK(11, 0), "Rx Undersize Packet", },
+	{ 0x00D0, GENMASK(11, 0), "Rx Oversize Packet", },
+	{ 0x00D4, GENMASK(11, 0), "Rx Fragments", },
+	{ 0x00D8, GENMASK(11, 0), "Rx Jabber", },
+	{ 0x00DC, GENMASK(11, 0), "Rx Dropped Packet", },
+	{ 0x00E0, GENMASK(23, 0), "Tx Byte", },
+	{ 0x00E4, GENMASK(17, 0), "Tx Packet", },
+	{ 0x00E8, GENMASK(17, 0), "Tx Multicast Packet", },
+	{ 0x00EC, GENMASK(17, 0), "Tx Broadcast Packet", },
+	{ 0x00F0, GENMASK(11, 0), "Tx Pause Control Frame", },
+	{ 0x00F4, GENMASK(11, 0), "Tx Deferral Packet", },
+	{ 0x00F8, GENMASK(11, 0), "Tx Excessive Deferral Packet", },
+	{ 0x00FC, GENMASK(11, 0), "Tx Single Collision Packet", },
+	{ 0x0100, GENMASK(11, 0), "Tx Multiple Collision", },
+	{ 0x0104, GENMASK(11, 0), "Tx Late Collision Packet", },
+	{ 0x0108, GENMASK(11, 0), "Tx Excessive Collision Packet", },
+	{ 0x010C, GENMASK(12, 0), "Tx Total Collision", },
+	{ 0x0110, GENMASK(11, 0), "Tx Pause Frames Honored", },
+	{ 0x0114, GENMASK(11, 0), "Tx Drop Frame", },
+	{ 0x0118, GENMASK(11, 0), "Tx Jabber Frame", },
+	{ 0x011C, GENMASK(11, 0), "Tx FCS Error", },
+	{ 0x0120, GENMASK(11, 0), "Tx Control Frame", },
+	{ 0x0124, GENMASK(11, 0), "Tx Oversize Frame", },
+	{ 0x0128, GENMASK(11, 0), "Tx Undersize Frame", },
+	{ 0x012C, GENMASK(11, 0), "Tx Fragment", },
+};
+
+#define DESC_EMPTY		BIT(31)
+#define DESC_MORE		BIT(24)
+#define DESC_PKTLEN_M		0xfff
+struct ag71xx_desc {
+	u32 data;
+	u32 ctrl;
+	u32 next;
+	u32 pad;
+} __aligned(4);
+
+#define AG71XX_DESC_SIZE	roundup(sizeof(struct ag71xx_desc), \
+					L1_CACHE_BYTES)
+
+struct ag71xx_buf {
+	union {
+		struct {
+			struct sk_buff *skb;
+			unsigned int len;
+		} tx;
+		struct {
+			dma_addr_t dma_addr;
+			void *rx_buf;
+		} rx;
+	};
+};
+
+struct ag71xx_ring {
+	/* "Hot" fields in the data path. */
+	unsigned int curr;
+	unsigned int dirty;
+
+	/* "Cold" fields - not used in the data path. */
+	struct ag71xx_buf *buf;
+	u16 order;
+	u16 desc_split;
+	dma_addr_t descs_dma;
+	u8 *descs_cpu;
+};
+
+enum ag71xx_type {
+	AR7100,
+	AR7240,
+	AR9130,
+	AR9330,
+	AR9340,
+	QCA9530,
+	QCA9550,
+};
+
+struct ag71xx_dcfg {
+	u32 max_frame_len;
+	const u32 *fifodata;
+	u16 desc_pktlen_mask;
+	bool tx_hang_workaround;
+	enum ag71xx_type type;
+};
+
+struct ag71xx {
+	/* Critical data related to the per-packet data path are clustered
+	 * early in this structure to help improve the D-cache footprint.
+	 */
+	struct ag71xx_ring rx_ring ____cacheline_aligned;
+	struct ag71xx_ring tx_ring ____cacheline_aligned;
+
+	u16 rx_buf_size;
+	u8 rx_buf_offset;
+
+	struct net_device *ndev;
+	struct platform_device *pdev;
+	struct napi_struct napi;
+	u32 msg_enable;
+	const struct ag71xx_dcfg *dcfg;
+
+	/* From this point onwards we're not looking at per-packet fields. */
+	void __iomem *mac_base;
+
+	struct ag71xx_desc *stop_desc;
+	dma_addr_t stop_desc_dma;
+
+	phy_interface_t phy_if_mode;
+	struct phylink *phylink;
+	struct phylink_config phylink_config;
+
+	struct delayed_work restart_work;
+	struct timer_list oom_timer;
+
+	struct reset_control *mac_reset;
+
+	u32 fifodata[3];
+	int mac_idx;
+
+	struct clk *clk_mdio;
+};
+
+static int ag71xx_desc_empty(struct ag71xx_desc *desc)
+{
+	return (desc->ctrl & DESC_EMPTY) != 0;
+}
+
+static struct ag71xx_desc *ag71xx_ring_desc(struct ag71xx_ring *ring, int idx)
+{
+	return (struct ag71xx_desc *)&ring->descs_cpu[idx * AG71XX_DESC_SIZE];
+}
+
+static int ag71xx_ring_size_order(int size)
+{
+	return fls(size - 1);
+}
+
+static bool ag71xx_is(struct ag71xx *ag, enum ag71xx_type type)
+{
+	return ag->dcfg->type == type;
+}
+
+static void ag71xx_wr(struct ag71xx *ag, unsigned int reg, u32 value)
+{
+	iowrite32(value, ag->mac_base + reg);
+	/* flush write */
+	(void)ioread32(ag->mac_base + reg);
+}
+
+static u32 ag71xx_rr(struct ag71xx *ag, unsigned int reg)
+{
+	return ioread32(ag->mac_base + reg);
+}
+
+static void ag71xx_sb(struct ag71xx *ag, unsigned int reg, u32 mask)
+{
+	void __iomem *r;
+
+	r = ag->mac_base + reg;
+	iowrite32(ioread32(r) | mask, r);
+	/* flush write */
+	(void)ioread32(r);
+}
+
+static void ag71xx_cb(struct ag71xx *ag, unsigned int reg, u32 mask)
+{
+	void __iomem *r;
+
+	r = ag->mac_base + reg;
+	iowrite32(ioread32(r) & ~mask, r);
+	/* flush write */
+	(void)ioread32(r);
+}
+
+static void ag71xx_int_enable(struct ag71xx *ag, u32 ints)
+{
+	ag71xx_sb(ag, AG71XX_REG_INT_ENABLE, ints);
+}
+
+static void ag71xx_int_disable(struct ag71xx *ag, u32 ints)
+{
+	ag71xx_cb(ag, AG71XX_REG_INT_ENABLE, ints);
+}
+
+static int ag71xx_do_ioctl(struct net_device *ndev, struct ifreq *ifr, int cmd)
+{
+	struct ag71xx *ag = netdev_priv(ndev);
+
+	return phylink_mii_ioctl(ag->phylink, ifr, cmd);
+}
+
+static void ag71xx_get_drvinfo(struct net_device *ndev,
+			       struct ethtool_drvinfo *info)
+{
+	struct ag71xx *ag = netdev_priv(ndev);
+
+	strscpy(info->driver, "ag71xx", sizeof(info->driver));
+	strscpy(info->bus_info, of_node_full_name(ag->pdev->dev.of_node),
+		sizeof(info->bus_info));
+}
+
+static int ag71xx_get_link_ksettings(struct net_device *ndev,
+				   struct ethtool_link_ksettings *kset)
+{
+	struct ag71xx *ag = netdev_priv(ndev);
+
+	return phylink_ethtool_ksettings_get(ag->phylink, kset);
+}
+
+static int ag71xx_set_link_ksettings(struct net_device *ndev,
+				   const struct ethtool_link_ksettings *kset)
+{
+	struct ag71xx *ag = netdev_priv(ndev);
+
+	return phylink_ethtool_ksettings_set(ag->phylink, kset);
+}
+
+static int ag71xx_ethtool_nway_reset(struct net_device *ndev)
+{
+	struct ag71xx *ag = netdev_priv(ndev);
+
+	return phylink_ethtool_nway_reset(ag->phylink);
+}
+
+static void ag71xx_ethtool_get_pauseparam(struct net_device *ndev,
+					  struct ethtool_pauseparam *pause)
+{
+	struct ag71xx *ag = netdev_priv(ndev);
+
+	phylink_ethtool_get_pauseparam(ag->phylink, pause);
+}
+
+static int ag71xx_ethtool_set_pauseparam(struct net_device *ndev,
+					 struct ethtool_pauseparam *pause)
+{
+	struct ag71xx *ag = netdev_priv(ndev);
+
+	return phylink_ethtool_set_pauseparam(ag->phylink, pause);
+}
+
+static void ag71xx_ethtool_get_strings(struct net_device *netdev, u32 sset,
+				       u8 *data)
+{
+	int i;
+
+	switch (sset) {
+	case ETH_SS_STATS:
+		for (i = 0; i < ARRAY_SIZE(ag71xx_statistics); i++)
+			ethtool_puts(&data, ag71xx_statistics[i].name);
+		break;
+	case ETH_SS_TEST:
+		net_selftest_get_strings(data);
+		break;
+	}
+}
+
+static void ag71xx_ethtool_get_stats(struct net_device *ndev,
+				     struct ethtool_stats *stats, u64 *data)
+{
+	struct ag71xx *ag = netdev_priv(ndev);
+	int i;
+
+	for (i = 0; i < ARRAY_SIZE(ag71xx_statistics); i++)
+		*data++ = ag71xx_rr(ag, ag71xx_statistics[i].offset)
+				& ag71xx_statistics[i].mask;
+}
+
+static int ag71xx_ethtool_get_sset_count(struct net_device *ndev, int sset)
+{
+	switch (sset) {
+	case ETH_SS_STATS:
+		return ARRAY_SIZE(ag71xx_statistics);
+	case ETH_SS_TEST:
+		return net_selftest_get_count();
+	default:
+		return -EOPNOTSUPP;
+	}
+}
+
+static const struct ethtool_ops ag71xx_ethtool_ops = {
+	.get_drvinfo			= ag71xx_get_drvinfo,
+	.get_link			= ethtool_op_get_link,
+	.get_ts_info			= ethtool_op_get_ts_info,
+	.get_link_ksettings		= ag71xx_get_link_ksettings,
+	.set_link_ksettings		= ag71xx_set_link_ksettings,
+	.nway_reset			= ag71xx_ethtool_nway_reset,
+	.get_pauseparam			= ag71xx_ethtool_get_pauseparam,
+	.set_pauseparam			= ag71xx_ethtool_set_pauseparam,
+	.get_strings			= ag71xx_ethtool_get_strings,
+	.get_ethtool_stats		= ag71xx_ethtool_get_stats,
+	.get_sset_count			= ag71xx_ethtool_get_sset_count,
+	.self_test			= net_selftest,
+};
+
+static int ag71xx_mdio_wait_busy(struct ag71xx *ag)
+{
+	struct net_device *ndev = ag->ndev;
+	int i;
+
+	for (i = 0; i < AG71XX_MDIO_RETRY; i++) {
+		u32 busy;
+
+		udelay(AG71XX_MDIO_DELAY);
+
+		busy = ag71xx_rr(ag, AG71XX_REG_MII_IND);
+		if (!busy)
+			return 0;
+
+		udelay(AG71XX_MDIO_DELAY);
+	}
+
+	netif_err(ag, link, ndev, "MDIO operation timed out\n");
+
+	return -ETIMEDOUT;
+}
+
+static int ag71xx_mdio_mii_read(struct mii_bus *bus, int addr, int reg)
+{
+	struct ag71xx *ag = bus->priv;
+	int err, val;
+
+	err = ag71xx_mdio_wait_busy(ag);
+	if (err)
+		return err;
+
+	ag71xx_wr(ag, AG71XX_REG_MII_ADDR,
+		  ((addr & 0x1f) << MII_ADDR_SHIFT) | (reg & 0xff));
+	/* enable read mode */
+	ag71xx_wr(ag, AG71XX_REG_MII_CMD, MII_CMD_READ);
+
+	err = ag71xx_mdio_wait_busy(ag);
+	if (err)
+		return err;
+
+	val = ag71xx_rr(ag, AG71XX_REG_MII_STATUS);
+	/* disable read mode */
+	ag71xx_wr(ag, AG71XX_REG_MII_CMD, 0);
+
+	netif_dbg(ag, link, ag->ndev, "mii_read: addr=%04x, reg=%04x, value=%04x\n",
+		  addr, reg, val);
+
+	return val;
+}
+
+static int ag71xx_mdio_mii_write(struct mii_bus *bus, int addr, int reg,
+				 u16 val)
+{
+	struct ag71xx *ag = bus->priv;
+
+	netif_dbg(ag, link, ag->ndev, "mii_write: addr=%04x, reg=%04x, value=%04x\n",
+		  addr, reg, val);
+
+	ag71xx_wr(ag, AG71XX_REG_MII_ADDR,
+		  ((addr & 0x1f) << MII_ADDR_SHIFT) | (reg & 0xff));
+	ag71xx_wr(ag, AG71XX_REG_MII_CTRL, val);
+
+	return ag71xx_mdio_wait_busy(ag);
+}
+
+static const u32 ar71xx_mdio_div_table[] = {
+	4, 4, 6, 8, 10, 14, 20, 28,
+};
+
+static const u32 ar7240_mdio_div_table[] = {
+	2, 2, 4, 6, 8, 12, 18, 26, 32, 40, 48, 56, 62, 70, 78, 96,
+};
+
+static const u32 ar933x_mdio_div_table[] = {
+	4, 4, 6, 8, 10, 14, 20, 28, 34, 42, 50, 58, 66, 74, 82, 98,
+};
+
+static int ag71xx_mdio_get_divider(struct ag71xx *ag, u32 *div)
+{
+	unsigned long ref_clock;
+	const u32 *table;
+	int ndivs, i;
+
+	ref_clock = clk_get_rate(ag->clk_mdio);
+	if (!ref_clock)
+		return -EINVAL;
+
+	if (ag71xx_is(ag, AR9330) || ag71xx_is(ag, AR9340)) {
+		table = ar933x_mdio_div_table;
+		ndivs = ARRAY_SIZE(ar933x_mdio_div_table);
+	} else if (ag71xx_is(ag, AR7240)) {
+		table = ar7240_mdio_div_table;
+		ndivs = ARRAY_SIZE(ar7240_mdio_div_table);
+	} else {
+		table = ar71xx_mdio_div_table;
+		ndivs = ARRAY_SIZE(ar71xx_mdio_div_table);
+	}
+
+	for (i = 0; i < ndivs; i++) {
+		unsigned long t;
+
+		t = ref_clock / table[i];
+		if (t <= AG71XX_MDIO_MAX_CLK) {
+			*div = i;
+			return 0;
+		}
+	}
+
+	return -ENOENT;
+}
+
+static int ag71xx_mdio_reset(struct mii_bus *bus)
+{
+	struct ag71xx *ag = bus->priv;
+	int err;
+	u32 t;
+
+	err = ag71xx_mdio_get_divider(ag, &t);
+	if (err)
+		return err;
+
+	ag71xx_wr(ag, AG71XX_REG_MII_CFG, t | MII_CFG_RESET);
+	usleep_range(100, 200);
+
+	ag71xx_wr(ag, AG71XX_REG_MII_CFG, t);
+	usleep_range(100, 200);
+
+	return 0;
+}
+
+static int ag71xx_mdio_probe(struct ag71xx *ag)
+{
+	struct device *dev = &ag->pdev->dev;
+	struct net_device *ndev = ag->ndev;
+	struct reset_control *mdio_reset;
+	static struct mii_bus *mii_bus;
+	struct device_node *np, *mnp;
+	int err;
+
+	np = dev->of_node;
+
+	ag->clk_mdio = devm_clk_get_enabled(dev, "mdio");
+	if (IS_ERR(ag->clk_mdio)) {
+		netif_err(ag, probe, ndev, "Failed to get mdio clk.\n");
+		return PTR_ERR(ag->clk_mdio);
+	}
+
+	mii_bus = devm_mdiobus_alloc(dev);
+	if (!mii_bus)
+		return -ENOMEM;
+
+	mdio_reset = devm_reset_control_get_exclusive(dev, "mdio");
+	if (IS_ERR(mdio_reset)) {
+		netif_err(ag, probe, ndev, "Failed to get reset mdio.\n");
+		return PTR_ERR(mdio_reset);
+	}
+
+	mii_bus->name = "ag71xx_mdio";
+	mii_bus->read = ag71xx_mdio_mii_read;
+	mii_bus->write = ag71xx_mdio_mii_write;
+	mii_bus->reset = ag71xx_mdio_reset;
+	mii_bus->priv = ag;
+	mii_bus->parent = dev;
+	snprintf(mii_bus->id, MII_BUS_ID_SIZE, "%s.%d", np->name, ag->mac_idx);
+
+	reset_control_assert(mdio_reset);
+	msleep(100);
+	reset_control_deassert(mdio_reset);
+	msleep(200);
+
+	mnp = of_get_child_by_name(np, "mdio");
+	err = devm_of_mdiobus_register(dev, mii_bus, mnp);
+	of_node_put(mnp);
+	if (err)
+		return err;
+
+	return 0;
+}
+
+static void ag71xx_hw_stop(struct ag71xx *ag)
+{
+	/* disable all interrupts and stop the rx/tx engine */
+	ag71xx_wr(ag, AG71XX_REG_INT_ENABLE, 0);
+	ag71xx_wr(ag, AG71XX_REG_RX_CTRL, 0);
+	ag71xx_wr(ag, AG71XX_REG_TX_CTRL, 0);
+}
+
+static bool ag71xx_check_dma_stuck(struct ag71xx *ag)
+{
+	unsigned long timestamp;
+	u32 rx_sm, tx_sm, rx_fd;
+
+	timestamp = READ_ONCE(netdev_get_tx_queue(ag->ndev, 0)->trans_start);
+	if (likely(time_before(jiffies, timestamp + HZ / 10)))
+		return false;
+
+	if (!netif_carrier_ok(ag->ndev))
+		return false;
+
+	rx_sm = ag71xx_rr(ag, AG71XX_REG_RX_SM);
+	if ((rx_sm & 0x7) == 0x3 && ((rx_sm >> 4) & 0x7) == 0x6)
+		return true;
+
+	tx_sm = ag71xx_rr(ag, AG71XX_REG_TX_SM);
+	rx_fd = ag71xx_rr(ag, AG71XX_REG_FIFO_DEPTH);
+	if (((tx_sm >> 4) & 0x7) == 0 && ((rx_sm & 0x7) == 0) &&
+	    ((rx_sm >> 4) & 0x7) == 0 && rx_fd == 0)
+		return true;
+
+	return false;
+}
+
+static int ag71xx_tx_packets(struct ag71xx *ag, bool flush, int budget)
+{
+	struct ag71xx_ring *ring = &ag->tx_ring;
+	int sent = 0, bytes_compl = 0, n = 0;
+	struct net_device *ndev = ag->ndev;
+	int ring_mask, ring_size;
+	bool dma_stuck = false;
+
+	ring_mask = BIT(ring->order) - 1;
+	ring_size = BIT(ring->order);
+
+	netif_dbg(ag, tx_queued, ndev, "processing TX ring\n");
+
+	while (ring->dirty + n != ring->curr) {
+		struct ag71xx_desc *desc;
+		struct sk_buff *skb;
+		unsigned int i;
+
+		i = (ring->dirty + n) & ring_mask;
+		desc = ag71xx_ring_desc(ring, i);
+		skb = ring->buf[i].tx.skb;
+
+		if (!flush && !ag71xx_desc_empty(desc)) {
+			if (ag->dcfg->tx_hang_workaround &&
+			    ag71xx_check_dma_stuck(ag)) {
+				schedule_delayed_work(&ag->restart_work,
+						      HZ / 2);
+				dma_stuck = true;
+			}
+			break;
+		}
+
+		if (flush)
+			desc->ctrl |= DESC_EMPTY;
+
+		n++;
+		if (!skb)
+			continue;
+
+		napi_consume_skb(skb, budget);
+		ring->buf[i].tx.skb = NULL;
+
+		bytes_compl += ring->buf[i].tx.len;
+
+		sent++;
+		ring->dirty += n;
+
+		while (n > 0) {
+			ag71xx_wr(ag, AG71XX_REG_TX_STATUS, TX_STATUS_PS);
+			n--;
+		}
+	}
+
+	netif_dbg(ag, tx_done, ndev, "%d packets sent out\n", sent);
+
+	if (!sent)
+		return 0;
+
+	ag->ndev->stats.tx_bytes += bytes_compl;
+	ag->ndev->stats.tx_packets += sent;
+
+	netdev_completed_queue(ag->ndev, sent, bytes_compl);
+	if ((ring->curr - ring->dirty) < (ring_size * 3) / 4)
+		netif_wake_queue(ag->ndev);
+
+	if (!dma_stuck)
+		cancel_delayed_work(&ag->restart_work);
+
+	return sent;
+}
+
+static void ag71xx_dma_wait_stop(struct ag71xx *ag)
+{
+	struct net_device *ndev = ag->ndev;
+	int i;
+
+	for (i = 0; i < AG71XX_DMA_RETRY; i++) {
+		u32 rx, tx;
+
+		mdelay(AG71XX_DMA_DELAY);
+
+		rx = ag71xx_rr(ag, AG71XX_REG_RX_CTRL) & RX_CTRL_RXE;
+		tx = ag71xx_rr(ag, AG71XX_REG_TX_CTRL) & TX_CTRL_TXE;
+		if (!rx && !tx)
+			return;
+	}
+
+	netif_err(ag, hw, ndev, "DMA stop operation timed out\n");
+}
+
+static void ag71xx_dma_reset(struct ag71xx *ag)
+{
+	struct net_device *ndev = ag->ndev;
+	u32 val;
+	int i;
+
+	/* stop RX and TX */
+	ag71xx_wr(ag, AG71XX_REG_RX_CTRL, 0);
+	ag71xx_wr(ag, AG71XX_REG_TX_CTRL, 0);
+
+	/* give the hardware some time to really stop all rx/tx activity
+	 * clearing the descriptors too early causes random memory corruption
+	 */
+	ag71xx_dma_wait_stop(ag);
+
+	/* clear descriptor addresses */
+	ag71xx_wr(ag, AG71XX_REG_TX_DESC, ag->stop_desc_dma);
+	ag71xx_wr(ag, AG71XX_REG_RX_DESC, ag->stop_desc_dma);
+
+	/* clear pending RX/TX interrupts */
+	for (i = 0; i < 256; i++) {
+		ag71xx_wr(ag, AG71XX_REG_RX_STATUS, RX_STATUS_PR);
+		ag71xx_wr(ag, AG71XX_REG_TX_STATUS, TX_STATUS_PS);
+	}
+
+	/* clear pending errors */
+	ag71xx_wr(ag, AG71XX_REG_RX_STATUS, RX_STATUS_BE | RX_STATUS_OF);
+	ag71xx_wr(ag, AG71XX_REG_TX_STATUS, TX_STATUS_BE | TX_STATUS_UR);
+
+	val = ag71xx_rr(ag, AG71XX_REG_RX_STATUS);
+	if (val)
+		netif_err(ag, hw, ndev, "unable to clear DMA Rx status: %08x\n",
+			  val);
+
+	val = ag71xx_rr(ag, AG71XX_REG_TX_STATUS);
+
+	/* mask out reserved bits */
+	val &= ~0xff000000;
+
+	if (val)
+		netif_err(ag, hw, ndev, "unable to clear DMA Tx status: %08x\n",
+			  val);
+}
+
+static void ag71xx_hw_setup(struct ag71xx *ag)
+{
+	u32 init = MAC_CFG1_INIT;
+
+	/* setup MAC configuration registers */
+	ag71xx_wr(ag, AG71XX_REG_MAC_CFG1, init);
+
+	ag71xx_sb(ag, AG71XX_REG_MAC_CFG2,
+		  MAC_CFG2_PAD_CRC_EN | MAC_CFG2_LEN_CHECK);
+
+	/* setup max frame length to zero */
+	ag71xx_wr(ag, AG71XX_REG_MAC_MFL, 0);
+
+	/* setup FIFO configuration registers */
+	ag71xx_wr(ag, AG71XX_REG_FIFO_CFG0, FIFO_CFG0_INIT);
+	ag71xx_wr(ag, AG71XX_REG_FIFO_CFG1, ag->fifodata[0]);
+	ag71xx_wr(ag, AG71XX_REG_FIFO_CFG2, ag->fifodata[1]);
+	ag71xx_wr(ag, AG71XX_REG_FIFO_CFG4, FIFO_CFG4_INIT);
+	ag71xx_wr(ag, AG71XX_REG_FIFO_CFG5, FIFO_CFG5_INIT);
+}
+
+static unsigned int ag71xx_max_frame_len(unsigned int mtu)
+{
+	return ETH_HLEN + VLAN_HLEN + mtu + ETH_FCS_LEN;
+}
+
+static void ag71xx_hw_set_macaddr(struct ag71xx *ag, const unsigned char *mac)
+{
+	u32 t;
+
+	t = (((u32)mac[5]) << 24) | (((u32)mac[4]) << 16)
+	  | (((u32)mac[3]) << 8) | ((u32)mac[2]);
+
+	ag71xx_wr(ag, AG71XX_REG_MAC_ADDR1, t);
+
+	t = (((u32)mac[1]) << 24) | (((u32)mac[0]) << 16);
+	ag71xx_wr(ag, AG71XX_REG_MAC_ADDR2, t);
+}
+
+static void ag71xx_fast_reset(struct ag71xx *ag)
+{
+	struct net_device *dev = ag->ndev;
+	u32 rx_ds;
+	u32 mii_reg;
+
+	ag71xx_hw_stop(ag);
+
+	mii_reg = ag71xx_rr(ag, AG71XX_REG_MII_CFG);
+	rx_ds = ag71xx_rr(ag, AG71XX_REG_RX_DESC);
+
+	ag71xx_tx_packets(ag, true, 0);
+
+	reset_control_assert(ag->mac_reset);
+	usleep_range(10, 20);
+	reset_control_deassert(ag->mac_reset);
+	usleep_range(10, 20);
+
+	ag71xx_dma_reset(ag);
+	ag71xx_hw_setup(ag);
+	ag->tx_ring.curr = 0;
+	ag->tx_ring.dirty = 0;
+	netdev_reset_queue(ag->ndev);
+
+	/* setup max frame length */
+	ag71xx_wr(ag, AG71XX_REG_MAC_MFL,
+		  ag71xx_max_frame_len(ag->ndev->mtu));
+
+	ag71xx_wr(ag, AG71XX_REG_RX_DESC, rx_ds);
+	ag71xx_wr(ag, AG71XX_REG_TX_DESC, ag->tx_ring.descs_dma);
+	ag71xx_wr(ag, AG71XX_REG_MII_CFG, mii_reg);
+
+	ag71xx_hw_set_macaddr(ag, dev->dev_addr);
+}
+
+static void ag71xx_hw_start(struct ag71xx *ag)
+{
+	/* start RX engine */
+	ag71xx_wr(ag, AG71XX_REG_RX_CTRL, RX_CTRL_RXE);
+
+	/* enable interrupts */
+	ag71xx_wr(ag, AG71XX_REG_INT_ENABLE, AG71XX_INT_INIT);
+
+	netif_wake_queue(ag->ndev);
+}
+
+static void ag71xx_mac_config(struct phylink_config *config, unsigned int mode,
+			      const struct phylink_link_state *state)
+{
+	struct ag71xx *ag = netdev_priv(to_net_dev(config->dev));
+
+	if (phylink_autoneg_inband(mode))
+		return;
+
+	if (!ag71xx_is(ag, AR7100) && !ag71xx_is(ag, AR9130))
+		ag71xx_fast_reset(ag);
+
+	if (ag->tx_ring.desc_split) {
+		ag->fifodata[2] &= 0xffff;
+		ag->fifodata[2] |= ((2048 - ag->tx_ring.desc_split) / 4) << 16;
+	}
+
+	ag71xx_wr(ag, AG71XX_REG_FIFO_CFG3, ag->fifodata[2]);
+}
+
+static void ag71xx_mac_link_down(struct phylink_config *config,
+				 unsigned int mode, phy_interface_t interface)
+{
+	struct ag71xx *ag = netdev_priv(to_net_dev(config->dev));
+
+	ag71xx_hw_stop(ag);
+}
+
+static void ag71xx_mac_link_up(struct phylink_config *config,
+			       struct phy_device *phy,
+			       unsigned int mode, phy_interface_t interface,
+			       int speed, int duplex,
+			       bool tx_pause, bool rx_pause)
+{
+	struct ag71xx *ag = netdev_priv(to_net_dev(config->dev));
+	u32 cfg1, cfg2;
+	u32 ifctl;
+	u32 fifo5;
+
+	cfg2 = ag71xx_rr(ag, AG71XX_REG_MAC_CFG2);
+	cfg2 &= ~(MAC_CFG2_IF_1000 | MAC_CFG2_IF_10_100 | MAC_CFG2_FDX);
+	cfg2 |= duplex ? MAC_CFG2_FDX : 0;
+
+	ifctl = ag71xx_rr(ag, AG71XX_REG_MAC_IFCTL);
+	ifctl &= ~(MAC_IFCTL_SPEED);
+
+	fifo5 = ag71xx_rr(ag, AG71XX_REG_FIFO_CFG5);
+	fifo5 &= ~FIFO_CFG5_BM;
+
+	switch (speed) {
+	case SPEED_1000:
+		cfg2 |= MAC_CFG2_IF_1000;
+		fifo5 |= FIFO_CFG5_BM;
+		break;
+	case SPEED_100:
+		cfg2 |= MAC_CFG2_IF_10_100;
+		ifctl |= MAC_IFCTL_SPEED;
+		break;
+	case SPEED_10:
+		cfg2 |= MAC_CFG2_IF_10_100;
+		break;
+	default:
+		return;
+	}
+
+	ag71xx_wr(ag, AG71XX_REG_MAC_CFG2, cfg2);
+	ag71xx_wr(ag, AG71XX_REG_FIFO_CFG5, fifo5);
+	ag71xx_wr(ag, AG71XX_REG_MAC_IFCTL, ifctl);
+
+	cfg1 = ag71xx_rr(ag, AG71XX_REG_MAC_CFG1);
+	cfg1 &= ~(MAC_CFG1_TFC | MAC_CFG1_RFC);
+	if (tx_pause)
+		cfg1 |= MAC_CFG1_TFC;
+
+	if (rx_pause)
+		cfg1 |= MAC_CFG1_RFC;
+	ag71xx_wr(ag, AG71XX_REG_MAC_CFG1, cfg1);
+
+	ag71xx_hw_start(ag);
+}
+
+static const struct phylink_mac_ops ag71xx_phylink_mac_ops = {
+	.mac_config = ag71xx_mac_config,
+	.mac_link_down = ag71xx_mac_link_down,
+	.mac_link_up = ag71xx_mac_link_up,
+};
+
+static int ag71xx_phylink_setup(struct ag71xx *ag)
+{
+	struct phylink *phylink;
+
+	ag->phylink_config.dev = &ag->ndev->dev;
+	ag->phylink_config.type = PHYLINK_NETDEV;
+	ag->phylink_config.mac_capabilities = MAC_SYM_PAUSE | MAC_ASYM_PAUSE |
+		MAC_10 | MAC_100 | MAC_1000FD;
+
+	if ((ag71xx_is(ag, AR9330) && ag->mac_idx == 0) ||
+	    ag71xx_is(ag, AR9340) ||
+	    ag71xx_is(ag, QCA9530) ||
+	    (ag71xx_is(ag, QCA9550) && ag->mac_idx == 1))
+		__set_bit(PHY_INTERFACE_MODE_MII,
+			  ag->phylink_config.supported_interfaces);
+
+	if ((ag71xx_is(ag, AR9330) && ag->mac_idx == 1) ||
+	    (ag71xx_is(ag, AR9340) && ag->mac_idx == 1) ||
+	    (ag71xx_is(ag, QCA9530) && ag->mac_idx == 1))
+		__set_bit(PHY_INTERFACE_MODE_GMII,
+			  ag->phylink_config.supported_interfaces);
+
+	if (ag71xx_is(ag, QCA9550) && ag->mac_idx == 0)
+		__set_bit(PHY_INTERFACE_MODE_SGMII,
+			  ag->phylink_config.supported_interfaces);
+
+	if (ag71xx_is(ag, AR9340) && ag->mac_idx == 0)
+		__set_bit(PHY_INTERFACE_MODE_RMII,
+			  ag->phylink_config.supported_interfaces);
+
+	if ((ag71xx_is(ag, AR9340) && ag->mac_idx == 0) ||
+	    (ag71xx_is(ag, QCA9550) && ag->mac_idx == 1))
+		__set_bit(PHY_INTERFACE_MODE_RGMII,
+			  ag->phylink_config.supported_interfaces);
+
+	phylink = phylink_create(&ag->phylink_config, ag->pdev->dev.fwnode,
+				 ag->phy_if_mode, &ag71xx_phylink_mac_ops);
+	if (IS_ERR(phylink))
+		return PTR_ERR(phylink);
+
+	ag->phylink = phylink;
+	return 0;
+}
+
+static void ag71xx_ring_tx_clean(struct ag71xx *ag)
+{
+	struct ag71xx_ring *ring = &ag->tx_ring;
+	int ring_mask = BIT(ring->order) - 1;
+	u32 bytes_compl = 0, pkts_compl = 0;
+	struct net_device *ndev = ag->ndev;
+
+	while (ring->curr != ring->dirty) {
+		struct ag71xx_desc *desc;
+		u32 i = ring->dirty & ring_mask;
+
+		desc = ag71xx_ring_desc(ring, i);
+		if (!ag71xx_desc_empty(desc)) {
+			desc->ctrl = 0;
+			ndev->stats.tx_errors++;
+		}
+
+		if (ring->buf[i].tx.skb) {
+			bytes_compl += ring->buf[i].tx.len;
+			pkts_compl++;
+			dev_kfree_skb_any(ring->buf[i].tx.skb);
+		}
+		ring->buf[i].tx.skb = NULL;
+		ring->dirty++;
+	}
+
+	/* flush descriptors */
+	wmb();
+
+	netdev_completed_queue(ndev, pkts_compl, bytes_compl);
+}
+
+static void ag71xx_ring_tx_init(struct ag71xx *ag)
+{
+	struct ag71xx_ring *ring = &ag->tx_ring;
+	int ring_size = BIT(ring->order);
+	int ring_mask = ring_size - 1;
+	int i;
+
+	for (i = 0; i < ring_size; i++) {
+		struct ag71xx_desc *desc = ag71xx_ring_desc(ring, i);
+
+		desc->next = (u32)(ring->descs_dma +
+			AG71XX_DESC_SIZE * ((i + 1) & ring_mask));
+
+		desc->ctrl = DESC_EMPTY;
+		ring->buf[i].tx.skb = NULL;
+	}
+
+	/* flush descriptors */
+	wmb();
+
+	ring->curr = 0;
+	ring->dirty = 0;
+	netdev_reset_queue(ag->ndev);
+}
+
+static void ag71xx_ring_rx_clean(struct ag71xx *ag)
+{
+	struct ag71xx_ring *ring = &ag->rx_ring;
+	int ring_size = BIT(ring->order);
+	int i;
+
+	if (!ring->buf)
+		return;
+
+	for (i = 0; i < ring_size; i++)
+		if (ring->buf[i].rx.rx_buf) {
+			dma_unmap_single(&ag->pdev->dev,
+					 ring->buf[i].rx.dma_addr,
+					 ag->rx_buf_size, DMA_FROM_DEVICE);
+			skb_free_frag(ring->buf[i].rx.rx_buf);
+		}
+}
+
+static int ag71xx_buffer_size(struct ag71xx *ag)
+{
+	return ag->rx_buf_size +
+	       SKB_DATA_ALIGN(sizeof(struct skb_shared_info));
+}
+
+static bool ag71xx_fill_rx_buf(struct ag71xx *ag, struct ag71xx_buf *buf,
+			       int offset,
+			       void *(*alloc)(unsigned int size))
+{
+	struct ag71xx_ring *ring = &ag->rx_ring;
+	struct ag71xx_desc *desc;
+	void *data;
+
+	desc = ag71xx_ring_desc(ring, buf - &ring->buf[0]);
+
+	data = alloc(ag71xx_buffer_size(ag));
+	if (!data)
+		return false;
+
+	buf->rx.rx_buf = data;
+	buf->rx.dma_addr = dma_map_single(&ag->pdev->dev, data, ag->rx_buf_size,
+					  DMA_FROM_DEVICE);
+	if (dma_mapping_error(&ag->pdev->dev, buf->rx.dma_addr)) {
+		skb_free_frag(data);
+		buf->rx.rx_buf = NULL;
+		return false;
+	}
+	desc->data = (u32)buf->rx.dma_addr + offset;
+	return true;
+}
+
+static int ag71xx_ring_rx_init(struct ag71xx *ag)
+{
+	struct ag71xx_ring *ring = &ag->rx_ring;
+	struct net_device *ndev = ag->ndev;
+	int ring_mask = BIT(ring->order) - 1;
+	int ring_size = BIT(ring->order);
+	unsigned int i;
+	int ret;
+
+	ret = 0;
+	for (i = 0; i < ring_size; i++) {
+		struct ag71xx_desc *desc = ag71xx_ring_desc(ring, i);
+
+		desc->next = (u32)(ring->descs_dma +
+			AG71XX_DESC_SIZE * ((i + 1) & ring_mask));
+
+		netif_dbg(ag, rx_status, ndev, "RX desc at %p, next is %08x\n",
+			  desc, desc->next);
+	}
+
+	for (i = 0; i < ring_size; i++) {
+		struct ag71xx_desc *desc = ag71xx_ring_desc(ring, i);
+
+		if (!ag71xx_fill_rx_buf(ag, &ring->buf[i], ag->rx_buf_offset,
+					netdev_alloc_frag)) {
+			ret = -ENOMEM;
+			break;
+		}
+
+		desc->ctrl = DESC_EMPTY;
+	}
+
+	/* flush descriptors */
+	wmb();
+
+	ring->curr = 0;
+	ring->dirty = 0;
+
+	return ret;
+}
+
+static int ag71xx_ring_rx_refill(struct ag71xx *ag)
+{
+	struct ag71xx_ring *ring = &ag->rx_ring;
+	int ring_mask = BIT(ring->order) - 1;
+	int offset = ag->rx_buf_offset;
+	unsigned int count;
+
+	count = 0;
+	for (; ring->curr - ring->dirty > 0; ring->dirty++) {
+		struct ag71xx_desc *desc;
+		unsigned int i;
+
+		i = ring->dirty & ring_mask;
+		desc = ag71xx_ring_desc(ring, i);
+
+		if (!ring->buf[i].rx.rx_buf &&
+		    !ag71xx_fill_rx_buf(ag, &ring->buf[i], offset,
+					napi_alloc_frag))
+			break;
+
+		desc->ctrl = DESC_EMPTY;
+		count++;
+	}
+
+	/* flush descriptors */
+	wmb();
+
+	netif_dbg(ag, rx_status, ag->ndev, "%u rx descriptors refilled\n",
+		  count);
+
+	return count;
+}
+
+static int ag71xx_rings_init(struct ag71xx *ag)
+{
+	struct ag71xx_ring *tx = &ag->tx_ring;
+	struct ag71xx_ring *rx = &ag->rx_ring;
+	int ring_size, tx_size;
+
+	ring_size = BIT(tx->order) + BIT(rx->order);
+	tx_size = BIT(tx->order);
+
+	tx->buf = kcalloc(ring_size, sizeof(*tx->buf), GFP_KERNEL);
+	if (!tx->buf)
+		return -ENOMEM;
+
+	tx->descs_cpu = dma_alloc_coherent(&ag->pdev->dev,
+					   ring_size * AG71XX_DESC_SIZE,
+					   &tx->descs_dma, GFP_KERNEL);
+	if (!tx->descs_cpu) {
+		kfree(tx->buf);
+		tx->buf = NULL;
+		return -ENOMEM;
+	}
+
+	rx->buf = &tx->buf[tx_size];
+	rx->descs_cpu = ((void *)tx->descs_cpu) + tx_size * AG71XX_DESC_SIZE;
+	rx->descs_dma = tx->descs_dma + tx_size * AG71XX_DESC_SIZE;
+
+	ag71xx_ring_tx_init(ag);
+	return ag71xx_ring_rx_init(ag);
+}
+
+static void ag71xx_rings_free(struct ag71xx *ag)
+{
+	struct ag71xx_ring *tx = &ag->tx_ring;
+	struct ag71xx_ring *rx = &ag->rx_ring;
+	int ring_size;
+
+	ring_size = BIT(tx->order) + BIT(rx->order);
+
+	if (tx->descs_cpu)
+		dma_free_coherent(&ag->pdev->dev, ring_size * AG71XX_DESC_SIZE,
+				  tx->descs_cpu, tx->descs_dma);
+
+	kfree(tx->buf);
+
+	tx->descs_cpu = NULL;
+	rx->descs_cpu = NULL;
+	tx->buf = NULL;
+	rx->buf = NULL;
+}
+
+static void ag71xx_rings_cleanup(struct ag71xx *ag)
+{
+	ag71xx_ring_rx_clean(ag);
+	ag71xx_ring_tx_clean(ag);
+	ag71xx_rings_free(ag);
+
+	netdev_reset_queue(ag->ndev);
+}
+
+static void ag71xx_hw_init(struct ag71xx *ag)
+{
+	ag71xx_hw_stop(ag);
+
+	ag71xx_sb(ag, AG71XX_REG_MAC_CFG1, MAC_CFG1_SR);
+	usleep_range(20, 30);
+
+	reset_control_assert(ag->mac_reset);
+	msleep(100);
+	reset_control_deassert(ag->mac_reset);
+	msleep(200);
+
+	ag71xx_hw_setup(ag);
+
+	ag71xx_dma_reset(ag);
+}
+
+static int ag71xx_hw_enable(struct ag71xx *ag)
+{
+	int ret;
+
+	ret = ag71xx_rings_init(ag);
+	if (ret)
+		return ret;
+
+	napi_enable(&ag->napi);
+	ag71xx_wr(ag, AG71XX_REG_TX_DESC, ag->tx_ring.descs_dma);
+	ag71xx_wr(ag, AG71XX_REG_RX_DESC, ag->rx_ring.descs_dma);
+	netif_start_queue(ag->ndev);
+
+	return 0;
+}
+
+static void ag71xx_hw_disable(struct ag71xx *ag)
+{
+	netif_stop_queue(ag->ndev);
+
+	ag71xx_hw_stop(ag);
+	ag71xx_dma_reset(ag);
+
+	napi_disable(&ag->napi);
+	timer_delete_sync(&ag->oom_timer);
+
+	ag71xx_rings_cleanup(ag);
+}
+
+static int ag71xx_open(struct net_device *ndev)
+{
+	struct ag71xx *ag = netdev_priv(ndev);
+	unsigned int max_frame_len;
+	int ret;
+
+	ret = phylink_of_phy_connect(ag->phylink, ag->pdev->dev.of_node, 0);
+	if (ret) {
+		netif_err(ag, link, ndev, "phylink_of_phy_connect filed with err: %i\n",
+			  ret);
+		return ret;
+	}
+
+	max_frame_len = ag71xx_max_frame_len(ndev->mtu);
+	ag->rx_buf_size =
+		SKB_DATA_ALIGN(max_frame_len + NET_SKB_PAD + NET_IP_ALIGN);
+
+	/* setup max frame length */
+	ag71xx_wr(ag, AG71XX_REG_MAC_MFL, max_frame_len);
+	ag71xx_hw_set_macaddr(ag, ndev->dev_addr);
+
+	ret = ag71xx_hw_enable(ag);
+	if (ret)
+		goto err;
+
+	phylink_start(ag->phylink);
+
+	return 0;
+
+err:
+	ag71xx_rings_cleanup(ag);
+	phylink_disconnect_phy(ag->phylink);
+	return ret;
+}
+
+static int ag71xx_stop(struct net_device *ndev)
+{
+	struct ag71xx *ag = netdev_priv(ndev);
+
+	phylink_stop(ag->phylink);
+	phylink_disconnect_phy(ag->phylink);
+	ag71xx_hw_disable(ag);
+
+	return 0;
+}
+
+static int ag71xx_fill_dma_desc(struct ag71xx_ring *ring, u32 addr, int len)
+{
+	int i, ring_mask, ndesc, split;
+	struct ag71xx_desc *desc;
+
+	ring_mask = BIT(ring->order) - 1;
+	ndesc = 0;
+	split = ring->desc_split;
+
+	if (!split)
+		split = len;
+
+	while (len > 0) {
+		unsigned int cur_len = len;
+
+		i = (ring->curr + ndesc) & ring_mask;
+		desc = ag71xx_ring_desc(ring, i);
+
+		if (!ag71xx_desc_empty(desc))
+			return -1;
+
+		if (cur_len > split) {
+			cur_len = split;
+
+			/*  TX will hang if DMA transfers <= 4 bytes,
+			 * make sure next segment is more than 4 bytes long.
+			 */
+			if (len <= split + 4)
+				cur_len -= 4;
+		}
+
+		desc->data = addr;
+		addr += cur_len;
+		len -= cur_len;
+
+		if (len > 0)
+			cur_len |= DESC_MORE;
+
+		/* prevent early tx attempt of this descriptor */
+		if (!ndesc)
+			cur_len |= DESC_EMPTY;
+
+		desc->ctrl = cur_len;
+		ndesc++;
+	}
+
+	return ndesc;
+}
+
+static netdev_tx_t ag71xx_hard_start_xmit(struct sk_buff *skb,
+					  struct net_device *ndev)
+{
+	int i, n, ring_min, ring_mask, ring_size;
+	struct ag71xx *ag = netdev_priv(ndev);
+	struct ag71xx_ring *ring;
+	struct ag71xx_desc *desc;
+	dma_addr_t dma_addr;
+
+	ring = &ag->tx_ring;
+	ring_mask = BIT(ring->order) - 1;
+	ring_size = BIT(ring->order);
+
+	if (skb->len <= 4) {
+		netif_dbg(ag, tx_err, ndev, "packet len is too small\n");
+		goto err_drop;
+	}
+
+	dma_addr = dma_map_single(&ag->pdev->dev, skb->data, skb->len,
+				  DMA_TO_DEVICE);
+	if (dma_mapping_error(&ag->pdev->dev, dma_addr)) {
+		netif_dbg(ag, tx_err, ndev, "DMA mapping error\n");
+		goto err_drop;
+	}
+
+	i = ring->curr & ring_mask;
+	desc = ag71xx_ring_desc(ring, i);
+
+	/* setup descriptor fields */
+	n = ag71xx_fill_dma_desc(ring, (u32)dma_addr,
+				 skb->len & ag->dcfg->desc_pktlen_mask);
+	if (n < 0)
+		goto err_drop_unmap;
+
+	i = (ring->curr + n - 1) & ring_mask;
+	ring->buf[i].tx.len = skb->len;
+	ring->buf[i].tx.skb = skb;
+
+	netdev_sent_queue(ndev, skb->len);
+
+	skb_tx_timestamp(skb);
+
+	desc->ctrl &= ~DESC_EMPTY;
+	ring->curr += n;
+
+	/* flush descriptor */
+	wmb();
+
+	ring_min = 2;
+	if (ring->desc_split)
+		ring_min *= AG71XX_TX_RING_DS_PER_PKT;
+
+	if (ring->curr - ring->dirty >= ring_size - ring_min) {
+		netif_dbg(ag, tx_err, ndev, "tx queue full\n");
+		netif_stop_queue(ndev);
+	}
+
+	netif_dbg(ag, tx_queued, ndev, "packet injected into TX queue\n");
+
+	/* enable TX engine */
+	ag71xx_wr(ag, AG71XX_REG_TX_CTRL, TX_CTRL_TXE);
+
+	return NETDEV_TX_OK;
+
+err_drop_unmap:
+	dma_unmap_single(&ag->pdev->dev, dma_addr, skb->len, DMA_TO_DEVICE);
+
+err_drop:
+	ndev->stats.tx_dropped++;
+
+	dev_kfree_skb(skb);
+	return NETDEV_TX_OK;
+}
+
+static void ag71xx_oom_timer_handler(struct timer_list *t)
+{
+	struct ag71xx *ag = timer_container_of(ag, t, oom_timer);
+
+	napi_schedule(&ag->napi);
+}
+
+static void ag71xx_tx_timeout(struct net_device *ndev, unsigned int txqueue)
+{
+	struct ag71xx *ag = netdev_priv(ndev);
+
+	netif_err(ag, tx_err, ndev, "tx timeout\n");
+
+	schedule_delayed_work(&ag->restart_work, 1);
+}
+
+static void ag71xx_restart_work_func(struct work_struct *work)
+{
+	struct ag71xx *ag = container_of(work, struct ag71xx,
+					 restart_work.work);
+
+	rtnl_lock();
+	ag71xx_hw_disable(ag);
+	ag71xx_hw_enable(ag);
+
+	phylink_stop(ag->phylink);
+	phylink_start(ag->phylink);
+
+	rtnl_unlock();
+}
+
+static int ag71xx_rx_packets(struct ag71xx *ag, int limit)
+{
+	struct net_device *ndev = ag->ndev;
+	int ring_mask, ring_size, done = 0;
+	unsigned int pktlen_mask, offset;
+	struct ag71xx_ring *ring;
+	struct sk_buff *skb;
+	LIST_HEAD(rx_list);
+
+	ring = &ag->rx_ring;
+	pktlen_mask = ag->dcfg->desc_pktlen_mask;
+	offset = ag->rx_buf_offset;
+	ring_mask = BIT(ring->order) - 1;
+	ring_size = BIT(ring->order);
+
+	netif_dbg(ag, rx_status, ndev, "rx packets, limit=%d, curr=%u, dirty=%u\n",
+		  limit, ring->curr, ring->dirty);
+
+	while (done < limit) {
+		unsigned int i = ring->curr & ring_mask;
+		struct ag71xx_desc *desc = ag71xx_ring_desc(ring, i);
+		int pktlen;
+
+		if (ag71xx_desc_empty(desc))
+			break;
+
+		if ((ring->dirty + ring_size) == ring->curr) {
+			WARN_ONCE(1, "RX out of ring");
+			break;
+		}
+
+		ag71xx_wr(ag, AG71XX_REG_RX_STATUS, RX_STATUS_PR);
+
+		pktlen = desc->ctrl & pktlen_mask;
+		pktlen -= ETH_FCS_LEN;
+
+		dma_unmap_single(&ag->pdev->dev, ring->buf[i].rx.dma_addr,
+				 ag->rx_buf_size, DMA_FROM_DEVICE);
+
+		ndev->stats.rx_packets++;
+		ndev->stats.rx_bytes += pktlen;
+
+		skb = napi_build_skb(ring->buf[i].rx.rx_buf, ag71xx_buffer_size(ag));
+		if (!skb) {
+			ndev->stats.rx_errors++;
+			skb_free_frag(ring->buf[i].rx.rx_buf);
+			goto next;
+		}
+
+		skb_reserve(skb, offset);
+		skb_put(skb, pktlen);
+
+		skb->dev = ndev;
+		skb->ip_summed = CHECKSUM_NONE;
+		skb->protocol = eth_type_trans(skb, ndev);
+		list_add_tail(&skb->list, &rx_list);
+
+next:
+		ring->buf[i].rx.rx_buf = NULL;
+		done++;
+
+		ring->curr++;
+	}
+
+	ag71xx_ring_rx_refill(ag);
+
+	netif_receive_skb_list(&rx_list);
+
+	netif_dbg(ag, rx_status, ndev, "rx finish, curr=%u, dirty=%u, done=%d\n",
+		  ring->curr, ring->dirty, done);
+
+	return done;
+}
+
+static int ag71xx_poll(struct napi_struct *napi, int limit)
+{
+	struct ag71xx *ag = container_of(napi, struct ag71xx, napi);
+	struct ag71xx_ring *rx_ring = &ag->rx_ring;
+	int rx_ring_size = BIT(rx_ring->order);
+	struct net_device *ndev = ag->ndev;
+	int tx_done, rx_done;
+	u32 status;
+
+	tx_done = ag71xx_tx_packets(ag, false, limit);
+
+	netif_dbg(ag, rx_status, ndev, "processing RX ring\n");
+	rx_done = ag71xx_rx_packets(ag, limit);
+
+	if (!rx_ring->buf[rx_ring->dirty % rx_ring_size].rx.rx_buf)
+		goto oom;
+
+	status = ag71xx_rr(ag, AG71XX_REG_RX_STATUS);
+	if (unlikely(status & RX_STATUS_OF)) {
+		ag71xx_wr(ag, AG71XX_REG_RX_STATUS, RX_STATUS_OF);
+		ndev->stats.rx_fifo_errors++;
+
+		/* restart RX */
+		ag71xx_wr(ag, AG71XX_REG_RX_CTRL, RX_CTRL_RXE);
+	}
+
+	if (rx_done < limit) {
+		if (status & RX_STATUS_PR)
+			goto more;
+
+		status = ag71xx_rr(ag, AG71XX_REG_TX_STATUS);
+		if (status & TX_STATUS_PS)
+			goto more;
+
+		netif_dbg(ag, rx_status, ndev, "disable polling mode, rx=%d, tx=%d,limit=%d\n",
+			  rx_done, tx_done, limit);
+
+		napi_complete(napi);
+
+		/* enable interrupts */
+		ag71xx_int_enable(ag, AG71XX_INT_POLL);
+		return rx_done;
+	}
+
+more:
+	netif_dbg(ag, rx_status, ndev, "stay in polling mode, rx=%d, tx=%d, limit=%d\n",
+		  rx_done, tx_done, limit);
+	return limit;
+
+oom:
+	netif_err(ag, rx_err, ndev, "out of memory\n");
+
+	mod_timer(&ag->oom_timer, jiffies + AG71XX_OOM_REFILL);
+	napi_complete(napi);
+	return 0;
+}
+
+static irqreturn_t ag71xx_interrupt(int irq, void *dev_id)
+{
+	struct net_device *ndev = dev_id;
+	struct ag71xx *ag;
+	u32 status;
+
+	ag = netdev_priv(ndev);
+	status = ag71xx_rr(ag, AG71XX_REG_INT_STATUS);
+
+	if (unlikely(!status))
+		return IRQ_NONE;
+
+	if (unlikely(status & AG71XX_INT_ERR)) {
+		if (status & AG71XX_INT_TX_BE) {
+			ag71xx_wr(ag, AG71XX_REG_TX_STATUS, TX_STATUS_BE);
+			netif_err(ag, intr, ndev, "TX BUS error\n");
+		}
+		if (status & AG71XX_INT_RX_BE) {
+			ag71xx_wr(ag, AG71XX_REG_RX_STATUS, RX_STATUS_BE);
+			netif_err(ag, intr, ndev, "RX BUS error\n");
+		}
+	}
+
+	if (likely(status & AG71XX_INT_POLL)) {
+		ag71xx_int_disable(ag, AG71XX_INT_POLL);
+		netif_dbg(ag, intr, ndev, "enable polling mode\n");
+		napi_schedule(&ag->napi);
+	}
+
+	return IRQ_HANDLED;
+}
+
+static int ag71xx_change_mtu(struct net_device *ndev, int new_mtu)
+{
+	struct ag71xx *ag = netdev_priv(ndev);
+
+	WRITE_ONCE(ndev->mtu, new_mtu);
+	ag71xx_wr(ag, AG71XX_REG_MAC_MFL,
+		  ag71xx_max_frame_len(ndev->mtu));
+
+	return 0;
+}
+
+static const struct net_device_ops ag71xx_netdev_ops = {
+	.ndo_open		= ag71xx_open,
+	.ndo_stop		= ag71xx_stop,
+	.ndo_start_xmit		= ag71xx_hard_start_xmit,
+	.ndo_eth_ioctl		= ag71xx_do_ioctl,
+	.ndo_tx_timeout		= ag71xx_tx_timeout,
+	.ndo_change_mtu		= ag71xx_change_mtu,
+	.ndo_set_mac_address	= eth_mac_addr,
+	.ndo_validate_addr	= eth_validate_addr,
+};
+
+static const u32 ar71xx_addr_ar7100[] = {
+	0x19000000, 0x1a000000,
+};
+
+static int ag71xx_probe(struct platform_device *pdev)
+{
+	struct device_node *np = pdev->dev.of_node;
+	const struct ag71xx_dcfg *dcfg;
+	struct net_device *ndev;
+	struct resource *res;
+	struct clk *clk_eth;
+	int tx_size, err, i;
+	struct ag71xx *ag;
+
+	if (!np)
+		return -ENODEV;
+
+	ndev = devm_alloc_etherdev(&pdev->dev, sizeof(*ag));
+	if (!ndev)
+		return -ENOMEM;
+
+	res = platform_get_resource(pdev, IORESOURCE_MEM, 0);
+	if (!res)
+		return -EINVAL;
+
+	dcfg = of_device_get_match_data(&pdev->dev);
+	if (!dcfg)
+		return -EINVAL;
+
+	ag = netdev_priv(ndev);
+	ag->mac_idx = -1;
+	for (i = 0; i < ARRAY_SIZE(ar71xx_addr_ar7100); i++) {
+		if (ar71xx_addr_ar7100[i] == res->start)
+			ag->mac_idx = i;
+	}
+
+	if (ag->mac_idx < 0) {
+		netif_err(ag, probe, ndev, "unknown mac idx\n");
+		return -EINVAL;
+	}
+
+	clk_eth = devm_clk_get_enabled(&pdev->dev, "eth");
+	if (IS_ERR(clk_eth))
+		return dev_err_probe(&pdev->dev, PTR_ERR(clk_eth),
+				     "Failed to get eth clk.");
+
+	SET_NETDEV_DEV(ndev, &pdev->dev);
+
+	ag->pdev = pdev;
+	ag->ndev = ndev;
+	ag->dcfg = dcfg;
+	ag->msg_enable = netif_msg_init(-1, AG71XX_DEFAULT_MSG_ENABLE);
+	memcpy(ag->fifodata, dcfg->fifodata, sizeof(ag->fifodata));
+
+	ag->mac_reset = devm_reset_control_get(&pdev->dev, "mac");
+	if (IS_ERR(ag->mac_reset))
+		return dev_err_probe(&pdev->dev, PTR_ERR(ag->mac_reset),
+				     "missing mac reset");
+
+	ag->mac_base = devm_ioremap_resource(&pdev->dev, res);
+	if (IS_ERR(ag->mac_base))
+		return PTR_ERR(ag->mac_base);
+
+	/* ensure that HW is in manual polling mode before interrupts are
+	 * activated. Otherwise ag71xx_interrupt might call napi_schedule
+	 * before it is initialized by netif_napi_add.
+	 */
+	ag71xx_int_disable(ag, AG71XX_INT_POLL);
+
+	ndev->irq = platform_get_irq(pdev, 0);
+	err = devm_request_irq(&pdev->dev, ndev->irq, ag71xx_interrupt,
+			       0x0, dev_name(&pdev->dev), ndev);
+	if (err) {
+		netif_err(ag, probe, ndev, "unable to request IRQ %d\n",
+			  ndev->irq);
+		return err;
+	}
+
+	ndev->netdev_ops = &ag71xx_netdev_ops;
+	ndev->ethtool_ops = &ag71xx_ethtool_ops;
+
+	INIT_DELAYED_WORK(&ag->restart_work, ag71xx_restart_work_func);
+	timer_setup(&ag->oom_timer, ag71xx_oom_timer_handler, 0);
+
+	tx_size = AG71XX_TX_RING_SIZE_DEFAULT;
+	ag->rx_ring.order = ag71xx_ring_size_order(AG71XX_RX_RING_SIZE_DEFAULT);
+
+	ndev->min_mtu = 68;
+	ndev->max_mtu = dcfg->max_frame_len - ag71xx_max_frame_len(0);
+
+	ag->rx_buf_offset = NET_SKB_PAD;
+	if (!ag71xx_is(ag, AR7100) && !ag71xx_is(ag, AR9130))
+		ag->rx_buf_offset += NET_IP_ALIGN;
+
+	if (ag71xx_is(ag, AR7100)) {
+		ag->tx_ring.desc_split = AG71XX_TX_RING_SPLIT;
+		tx_size *= AG71XX_TX_RING_DS_PER_PKT;
+	}
+	ag->tx_ring.order = ag71xx_ring_size_order(tx_size);
+
+	ag->stop_desc = dmam_alloc_coherent(&pdev->dev,
+					    sizeof(struct ag71xx_desc),
+					    &ag->stop_desc_dma, GFP_KERNEL);
+	if (!ag->stop_desc)
+		return -ENOMEM;
+
+	ag->stop_desc->data = 0;
+	ag->stop_desc->ctrl = 0;
+	ag->stop_desc->next = (u32)ag->stop_desc_dma;
+
+	err = of_get_ethdev_address(np, ndev);
+	if (err == -EPROBE_DEFER)
+		return err;
+	if (err) {
+		netif_err(ag, probe, ndev, "invalid MAC address, using random address\n");
+		eth_hw_addr_random(ndev);
+	}
+
+	err = of_get_phy_mode(np, &ag->phy_if_mode);
+	if (err) {
+		netif_err(ag, probe, ndev, "missing phy-mode property in DT\n");
+		return err;
+	}
+
+	netif_napi_add_weight(ndev, &ag->napi, ag71xx_poll,
+			      AG71XX_NAPI_WEIGHT);
+
+	ag71xx_wr(ag, AG71XX_REG_MAC_CFG1, 0);
+
+	ag71xx_hw_init(ag);
+
+	err = ag71xx_mdio_probe(ag);
+	if (err)
+		return err;
+
+	err = ag71xx_phylink_setup(ag);
+	if (err)
+		return dev_err_probe(&pdev->dev, err,
+				     "failed to setup phylink");
+
+	err = devm_register_netdev(&pdev->dev, ndev);
+	if (err) {
+		netif_err(ag, probe, ndev, "unable to register net device\n");
+		return err;
+	}
+
+	netif_info(ag, probe, ndev, "Atheros AG71xx at 0x%08lx, irq %d, mode:%s\n",
+		   (unsigned long)ag->mac_base, ndev->irq,
+		   phy_modes(ag->phy_if_mode));
+
+	return 0;
+}
+
+static const u32 ar71xx_fifo_ar7100[] = {
+	0x0fff0000, 0x00001fff, 0x00780fff,
+};
+
+static const u32 ar71xx_fifo_ar9130[] = {
+	0x0fff0000, 0x00001fff, 0x008001ff,
+};
+
+static const u32 ar71xx_fifo_ar9330[] = {
+	0x0010ffff, 0x015500aa, 0x01f00140,
+};
+
+static const struct ag71xx_dcfg ag71xx_dcfg_ar7100 = {
+	.type = AR7100,
+	.fifodata = ar71xx_fifo_ar7100,
+	.max_frame_len = 1540,
+	.desc_pktlen_mask = SZ_4K - 1,
+	.tx_hang_workaround = false,
+};
+
+static const struct ag71xx_dcfg ag71xx_dcfg_ar7240 = {
+	.type = AR7240,
+	.fifodata = ar71xx_fifo_ar7100,
+	.max_frame_len = 1540,
+	.desc_pktlen_mask = SZ_4K - 1,
+	.tx_hang_workaround = true,
+};
+
+static const struct ag71xx_dcfg ag71xx_dcfg_ar9130 = {
+	.type = AR9130,
+	.fifodata = ar71xx_fifo_ar9130,
+	.max_frame_len = 1540,
+	.desc_pktlen_mask = SZ_4K - 1,
+	.tx_hang_workaround = false,
+};
+
+static const struct ag71xx_dcfg ag71xx_dcfg_ar9330 = {
+	.type = AR9330,
+	.fifodata = ar71xx_fifo_ar9330,
+	.max_frame_len = 1540,
+	.desc_pktlen_mask = SZ_4K - 1,
+	.tx_hang_workaround = true,
+};
+
+static const struct ag71xx_dcfg ag71xx_dcfg_ar9340 = {
+	.type = AR9340,
+	.fifodata = ar71xx_fifo_ar9330,
+	.max_frame_len = SZ_16K - 1,
+	.desc_pktlen_mask = SZ_16K - 1,
+	.tx_hang_workaround = true,
+};
+
+static const struct ag71xx_dcfg ag71xx_dcfg_qca9530 = {
+	.type = QCA9530,
+	.fifodata = ar71xx_fifo_ar9330,
+	.max_frame_len = SZ_16K - 1,
+	.desc_pktlen_mask = SZ_16K - 1,
+	.tx_hang_workaround = true,
+};
+
+static const struct ag71xx_dcfg ag71xx_dcfg_qca9550 = {
+	.type = QCA9550,
+	.fifodata = ar71xx_fifo_ar9330,
+	.max_frame_len = 1540,
+	.desc_pktlen_mask = SZ_16K - 1,
+	.tx_hang_workaround = true,
+};
+
+static const struct of_device_id ag71xx_match[] = {
+	{ .compatible = "qca,ar7100-eth", .data = &ag71xx_dcfg_ar7100 },
+	{ .compatible = "qca,ar7240-eth", .data = &ag71xx_dcfg_ar7240 },
+	{ .compatible = "qca,ar7241-eth", .data = &ag71xx_dcfg_ar7240 },
+	{ .compatible = "qca,ar7242-eth", .data = &ag71xx_dcfg_ar7240 },
+	{ .compatible = "qca,ar9130-eth", .data = &ag71xx_dcfg_ar9130 },
+	{ .compatible = "qca,ar9330-eth", .data = &ag71xx_dcfg_ar9330 },
+	{ .compatible = "qca,ar9340-eth", .data = &ag71xx_dcfg_ar9340 },
+	{ .compatible = "qca,qca9530-eth", .data = &ag71xx_dcfg_qca9530 },
+	{ .compatible = "qca,qca9550-eth", .data = &ag71xx_dcfg_qca9550 },
+	{ .compatible = "qca,qca9560-eth", .data = &ag71xx_dcfg_qca9550 },
+	{}
+};
+MODULE_DEVICE_TABLE(of, ag71xx_match);
+
+static struct platform_driver ag71xx_driver = {
+	.probe		= ag71xx_probe,
+	.driver = {
+		.name	= "ag71xx",
+		.of_match_table = ag71xx_match,
+	}
+};
+
+module_platform_driver(ag71xx_driver);
+MODULE_DESCRIPTION("Atheros AR71xx built-in ethernet mac driver");
+MODULE_LICENSE("GPL v2");
diff --git a/drivers/net/ethernet/atheros/alx/Makefile b/drivers/net/ethernet/atheros/alx/Makefile
index ed4a605874a3..fec7885a599b 100644
--- a/drivers/net/ethernet/atheros/alx/Makefile
+++ b/drivers/net/ethernet/atheros/alx/Makefile
@@ -1,2 +1,3 @@
+# SPDX-License-Identifier: GPL-2.0-only
 obj-$(CONFIG_ALX) += alx.o
 alx-objs := main.o ethtool.o hw.o
diff --git a/drivers/net/ethernet/atheros/alx/alx.h b/drivers/net/ethernet/atheros/alx/alx.h
index 78c5de467426..693006c5a498 100644
--- a/drivers/net/ethernet/atheros/alx/alx.h
+++ b/drivers/net/ethernet/atheros/alx/alx.h
@@ -137,9 +137,10 @@ struct alx_priv {
 
 	/* protects hw.stats */
 	spinlock_t stats_lock;
+
+	struct mutex mtx;
 };
 
 extern const struct ethtool_ops alx_ethtool_ops;
-extern const char alx_drv_name[];
 
 #endif
diff --git a/drivers/net/ethernet/atheros/alx/ethtool.c b/drivers/net/ethernet/atheros/alx/ethtool.c
index 2f4eabf652e8..7f6b69a52367 100644
--- a/drivers/net/ethernet/atheros/alx/ethtool.c
+++ b/drivers/net/ethernet/atheros/alx/ethtool.c
@@ -163,8 +163,10 @@ static int alx_get_link_ksettings(struct net_device *netdev,
 		}
 	}
 
+	mutex_lock(&alx->mtx);
 	cmd->base.speed = hw->link_speed;
 	cmd->base.duplex = hw->duplex;
+	mutex_unlock(&alx->mtx);
 
 	ethtool_convert_legacy_u32_to_link_mode(cmd->link_modes.supported,
 						supported);
@@ -181,8 +183,7 @@ static int alx_set_link_ksettings(struct net_device *netdev,
 	struct alx_hw *hw = &alx->hw;
 	u32 adv_cfg;
 	u32 advertising;
-
-	ASSERT_RTNL();
+	int ret;
 
 	ethtool_convert_link_mode_to_legacy_u32(&advertising,
 						cmd->link_modes.advertising);
@@ -200,7 +201,12 @@ static int alx_set_link_ksettings(struct net_device *netdev,
 	}
 
 	hw->adv_cfg = adv_cfg;
-	return alx_setup_speed_duplex(hw, adv_cfg, hw->flowctrl);
+
+	mutex_lock(&alx->mtx);
+	ret = alx_setup_speed_duplex(hw, adv_cfg, hw->flowctrl);
+	mutex_unlock(&alx->mtx);
+
+	return ret;
 }
 
 static void alx_get_pauseparam(struct net_device *netdev,
@@ -209,10 +215,12 @@ static void alx_get_pauseparam(struct net_device *netdev,
 	struct alx_priv *alx = netdev_priv(netdev);
 	struct alx_hw *hw = &alx->hw;
 
+	mutex_lock(&alx->mtx);
 	pause->autoneg = !!(hw->flowctrl & ALX_FC_ANEG &&
 			    hw->adv_cfg & ADVERTISED_Autoneg);
 	pause->tx_pause = !!(hw->flowctrl & ALX_FC_TX);
 	pause->rx_pause = !!(hw->flowctrl & ALX_FC_RX);
+	mutex_unlock(&alx->mtx);
 }
 
 
@@ -232,7 +240,7 @@ static int alx_set_pauseparam(struct net_device *netdev,
 	if (pause->autoneg)
 		fc |= ALX_FC_ANEG;
 
-	ASSERT_RTNL();
+	mutex_lock(&alx->mtx);
 
 	/* restart auto-neg for auto-mode */
 	if (hw->adv_cfg & ADVERTISED_Autoneg) {
@@ -245,8 +253,10 @@ static int alx_set_pauseparam(struct net_device *netdev,
 
 	if (reconfig_phy) {
 		err = alx_setup_speed_duplex(hw, hw->adv_cfg, fc);
-		if (err)
+		if (err) {
+			mutex_unlock(&alx->mtx);
 			return err;
+		}
 	}
 
 	/* flow control on mac */
@@ -254,6 +264,7 @@ static int alx_set_pauseparam(struct net_device *netdev,
 		alx_cfg_mac_flowcontrol(hw, fc);
 
 	hw->flowctrl = fc;
+	mutex_unlock(&alx->mtx);
 
 	return 0;
 }
@@ -281,9 +292,8 @@ static void alx_get_ethtool_stats(struct net_device *netdev,
 	spin_lock(&alx->stats_lock);
 
 	alx_update_hw_stats(hw);
-	BUILD_BUG_ON(sizeof(hw->stats) - offsetof(struct alx_hw_stats, rx_ok) <
-		     ALX_NUM_STATS * sizeof(u64));
-	memcpy(data, &hw->stats.rx_ok, ALX_NUM_STATS * sizeof(u64));
+	BUILD_BUG_ON(sizeof(hw->stats) != ALX_NUM_STATS * sizeof(u64));
+	memcpy(data, &hw->stats, sizeof(hw->stats));
 
 	spin_unlock(&alx->stats_lock);
 }
diff --git a/drivers/net/ethernet/atheros/alx/main.c b/drivers/net/ethernet/atheros/alx/main.c
index 567ee54504bc..ad6d6abd885f 100644
--- a/drivers/net/ethernet/atheros/alx/main.c
+++ b/drivers/net/ethernet/atheros/alx/main.c
@@ -1,5 +1,5 @@
 /*
- * Copyright (c) 2013 Johannes Berg <johannes@sipsolutions.net>
+ * Copyright (c) 2013, 2021 Johannes Berg <johannes@sipsolutions.net>
  *
  *  This file is free software: you may copy, redistribute and/or modify it
  *  under the terms of the GNU General Public License as published by the
@@ -39,7 +39,6 @@
 #include <linux/ipv6.h>
 #include <linux/if_vlan.h>
 #include <linux/mdio.h>
-#include <linux/aer.h>
 #include <linux/bitops.h>
 #include <linux/netdevice.h>
 #include <linux/etherdevice.h>
@@ -49,7 +48,7 @@
 #include "hw.h"
 #include "reg.h"
 
-const char alx_drv_name[] = "alx";
+static const char alx_drv_name[] = "alx";
 
 static void alx_free_txbuf(struct alx_tx_queue *txq, int entry)
 {
@@ -607,7 +606,7 @@ static int alx_set_mac_address(struct net_device *netdev, void *data)
 	if (netdev->addr_assign_type & NET_ADDR_RANDOM)
 		netdev->addr_assign_type ^= NET_ADDR_RANDOM;
 
-	memcpy(netdev->dev_addr, addr->sa_data, netdev->addr_len);
+	eth_hw_addr_set(netdev, addr->sa_data);
 	memcpy(hw->mac_addr, addr->sa_data, netdev->addr_len);
 	alx_set_macaddr(hw, hw->mac_addr);
 
@@ -660,10 +659,9 @@ static int alx_alloc_rings(struct alx_priv *alx)
 			    alx->num_txq +
 			    sizeof(struct alx_rrd) * alx->rx_ringsz +
 			    sizeof(struct alx_rfd) * alx->rx_ringsz;
-	alx->descmem.virt = dma_zalloc_coherent(&alx->hw.pdev->dev,
-						alx->descmem.size,
-						&alx->descmem.dma,
-						GFP_KERNEL);
+	alx->descmem.virt = dma_alloc_coherent(&alx->hw.pdev->dev,
+					       alx->descmem.size,
+					       &alx->descmem.dma, GFP_KERNEL);
 	if (!alx->descmem.virt)
 		return -ENOMEM;
 
@@ -753,7 +751,7 @@ static int alx_alloc_napis(struct alx_priv *alx)
 			goto err_out;
 
 		np->alx = alx;
-		netif_napi_add(alx->dev, &np->napi, alx_poll, 64);
+		netif_napi_add(alx->dev, &np->napi, alx_poll);
 		alx->qnapi[i] = np;
 	}
 
@@ -903,7 +901,7 @@ static int alx_init_intr(struct alx_priv *alx)
 	int ret;
 
 	ret = pci_alloc_irq_vectors(alx->hw.pdev, 1, 1,
-			PCI_IRQ_MSI | PCI_IRQ_LEGACY);
+			PCI_IRQ_MSI | PCI_IRQ_INTX);
 	if (ret < 0)
 		return ret;
 
@@ -1092,8 +1090,9 @@ static int alx_init_sw(struct alx_priv *alx)
 		      ALX_MAC_CTRL_RXFC_EN |
 		      ALX_MAC_CTRL_TXFC_EN |
 		      7 << ALX_MAC_CTRL_PRMBLEN_SHIFT;
+	mutex_init(&alx->mtx);
 
-	return err;
+	return 0;
 }
 
 
@@ -1123,6 +1122,8 @@ static void alx_halt(struct alx_priv *alx)
 {
 	struct alx_hw *hw = &alx->hw;
 
+	lockdep_assert_held(&alx->mtx);
+
 	alx_netif_stop(alx);
 	hw->link_speed = SPEED_UNKNOWN;
 	hw->duplex = DUPLEX_UNKNOWN;
@@ -1148,6 +1149,8 @@ static void alx_configure(struct alx_priv *alx)
 
 static void alx_activate(struct alx_priv *alx)
 {
+	lockdep_assert_held(&alx->mtx);
+
 	/* hardware setting lost, restore it */
 	alx_reinit_rings(alx);
 	alx_configure(alx);
@@ -1162,7 +1165,7 @@ static void alx_activate(struct alx_priv *alx)
 
 static void alx_reinit(struct alx_priv *alx)
 {
-	ASSERT_RTNL();
+	lockdep_assert_held(&alx->mtx);
 
 	alx_halt(alx);
 	alx_activate(alx);
@@ -1173,12 +1176,15 @@ static int alx_change_mtu(struct net_device *netdev, int mtu)
 	struct alx_priv *alx = netdev_priv(netdev);
 	int max_frame = ALX_MAX_FRAME_LEN(mtu);
 
-	netdev->mtu = mtu;
+	WRITE_ONCE(netdev->mtu, mtu);
 	alx->hw.mtu = mtu;
 	alx->rxbuf_size = max(max_frame, ALX_DEF_RXBUF_SIZE);
 	netdev_update_features(netdev);
-	if (netif_running(netdev))
+	if (netif_running(netdev)) {
+		mutex_lock(&alx->mtx);
 		alx_reinit(alx);
+		mutex_unlock(&alx->mtx);
+	}
 	return 0;
 }
 
@@ -1250,8 +1256,14 @@ out_disable_adv_intr:
 
 static void __alx_stop(struct alx_priv *alx)
 {
-	alx_halt(alx);
+	lockdep_assert_held(&alx->mtx);
+
 	alx_free_irq(alx);
+
+	cancel_work_sync(&alx->link_check_wk);
+	cancel_work_sync(&alx->reset_wk);
+
+	alx_halt(alx);
 	alx_free_rings(alx);
 	alx_free_napis(alx);
 }
@@ -1279,16 +1291,16 @@ static void alx_check_link(struct alx_priv *alx)
 	struct alx_hw *hw = &alx->hw;
 	unsigned long flags;
 	int old_speed;
-	u8 old_duplex;
 	int err;
 
+	lockdep_assert_held(&alx->mtx);
+
 	/* clear PHY internal interrupt status, otherwise the main
 	 * interrupt status will be asserted forever
 	 */
 	alx_clear_phy_intr(hw);
 
 	old_speed = hw->link_speed;
-	old_duplex = hw->duplex;
 	err = alx_read_phy_link(hw);
 	if (err < 0)
 		goto reset;
@@ -1337,12 +1349,24 @@ reset:
 
 static int alx_open(struct net_device *netdev)
 {
-	return __alx_open(netdev_priv(netdev), false);
+	struct alx_priv *alx = netdev_priv(netdev);
+	int ret;
+
+	mutex_lock(&alx->mtx);
+	ret = __alx_open(alx, false);
+	mutex_unlock(&alx->mtx);
+
+	return ret;
 }
 
 static int alx_stop(struct net_device *netdev)
 {
-	__alx_stop(netdev_priv(netdev));
+	struct alx_priv *alx = netdev_priv(netdev);
+
+	mutex_lock(&alx->mtx);
+	__alx_stop(alx);
+	mutex_unlock(&alx->mtx);
+
 	return 0;
 }
 
@@ -1352,18 +1376,18 @@ static void alx_link_check(struct work_struct *work)
 
 	alx = container_of(work, struct alx_priv, link_check_wk);
 
-	rtnl_lock();
+	mutex_lock(&alx->mtx);
 	alx_check_link(alx);
-	rtnl_unlock();
+	mutex_unlock(&alx->mtx);
 }
 
 static void alx_reset(struct work_struct *work)
 {
 	struct alx_priv *alx = container_of(work, struct alx_priv, reset_wk);
 
-	rtnl_lock();
+	mutex_lock(&alx->mtx);
 	alx_reinit(alx);
-	rtnl_unlock();
+	mutex_unlock(&alx->mtx);
 }
 
 static int alx_tpd_req(struct sk_buff *skb)
@@ -1419,10 +1443,7 @@ static int alx_tso(struct sk_buff *skb, struct alx_txd *first)
 							 0, IPPROTO_TCP, 0);
 		first->word1 |= 1 << TPD_IPV4_SHIFT;
 	} else if (skb_is_gso_v6(skb)) {
-		ipv6_hdr(skb)->payload_len = 0;
-		tcp_hdr(skb)->check = ~csum_ipv6_magic(&ipv6_hdr(skb)->saddr,
-						       &ipv6_hdr(skb)->daddr,
-						       0, IPPROTO_TCP, 0);
+		tcp_v6_gso_csum_prep(skb);
 		/* LSOv2: the first TPD only provides the packet length */
 		first->adrl.l.pkt_len = skb->len;
 		first->word1 |= 1 << TPD_LSO_V2_SHIFT;
@@ -1468,9 +1489,7 @@ static int alx_map_tx_skb(struct alx_tx_queue *txq, struct sk_buff *skb)
 	tpd->len = cpu_to_le16(maplen);
 
 	for (f = 0; f < skb_shinfo(skb)->nr_frags; f++) {
-		struct skb_frag_struct *frag;
-
-		frag = &skb_shinfo(skb)->frags[f];
+		skb_frag_t *frag = &skb_shinfo(skb)->frags[f];
 
 		if (++txq->write_idx == txq->count)
 			txq->write_idx = 0;
@@ -1558,7 +1577,7 @@ static netdev_tx_t alx_start_xmit(struct sk_buff *skb,
 	return alx_start_xmit_ring(skb, alx_tx_queue_mapping(alx, skb));
 }
 
-static void alx_tx_timeout(struct net_device *dev)
+static void alx_tx_timeout(struct net_device *dev, unsigned int txqueue)
 {
 	struct alx_priv *alx = netdev_priv(dev);
 
@@ -1684,7 +1703,7 @@ static const struct net_device_ops alx_netdev_ops = {
 	.ndo_validate_addr      = eth_validate_addr,
 	.ndo_set_mac_address    = alx_set_mac_address,
 	.ndo_change_mtu         = alx_change_mtu,
-	.ndo_do_ioctl           = alx_ioctl,
+	.ndo_eth_ioctl           = alx_ioctl,
 	.ndo_tx_timeout         = alx_tx_timeout,
 	.ndo_fix_features	= alx_fix_features,
 #ifdef CONFIG_NET_POLL_CONTROLLER
@@ -1725,7 +1744,6 @@ static int alx_probe(struct pci_dev *pdev, const struct pci_device_id *ent)
 		goto out_pci_disable;
 	}
 
-	pci_enable_pcie_error_reporting(pdev);
 	pci_set_master(pdev);
 
 	if (!pdev->pm_cap) {
@@ -1775,6 +1793,8 @@ static int alx_probe(struct pci_dev *pdev, const struct pci_device_id *ent)
 		goto out_unmap;
 	}
 
+	mutex_lock(&alx->mtx);
+
 	alx_reset_pcie(hw);
 
 	phy_configured = alx_phy_configured(hw);
@@ -1785,7 +1805,7 @@ static int alx_probe(struct pci_dev *pdev, const struct pci_device_id *ent)
 	err = alx_reset_mac(hw);
 	if (err) {
 		dev_err(&pdev->dev, "MAC Reset failed, error = %d\n", err);
-		goto out_unmap;
+		goto out_unlock;
 	}
 
 	/* setup link to put it in a known good starting state */
@@ -1795,7 +1815,7 @@ static int alx_probe(struct pci_dev *pdev, const struct pci_device_id *ent)
 			dev_err(&pdev->dev,
 				"failed to configure PHY speed/duplex (err=%d)\n",
 				err);
-			goto out_unmap;
+			goto out_unlock;
 		}
 	}
 
@@ -1813,7 +1833,7 @@ static int alx_probe(struct pci_dev *pdev, const struct pci_device_id *ent)
 	}
 
 	memcpy(hw->mac_addr, hw->perm_addr, ETH_ALEN);
-	memcpy(netdev->dev_addr, hw->mac_addr, ETH_ALEN);
+	eth_hw_addr_set(netdev, hw->mac_addr);
 	memcpy(netdev->perm_addr, hw->perm_addr, ETH_ALEN);
 
 	hw->mdio.prtad = 0;
@@ -1828,9 +1848,11 @@ static int alx_probe(struct pci_dev *pdev, const struct pci_device_id *ent)
 	if (!alx_get_phy_info(hw)) {
 		dev_err(&pdev->dev, "failed to identify PHY\n");
 		err = -EIO;
-		goto out_unmap;
+		goto out_unlock;
 	}
 
+	mutex_unlock(&alx->mtx);
+
 	INIT_WORK(&alx->link_check_wk, alx_link_check);
 	INIT_WORK(&alx->reset_wk, alx_reset);
 	netif_carrier_off(netdev);
@@ -1847,6 +1869,8 @@ static int alx_probe(struct pci_dev *pdev, const struct pci_device_id *ent)
 
 	return 0;
 
+out_unlock:
+	mutex_unlock(&alx->mtx);
 out_unmap:
 	iounmap(hw->hw_addr);
 out_free_netdev:
@@ -1863,9 +1887,6 @@ static void alx_remove(struct pci_dev *pdev)
 	struct alx_priv *alx = pci_get_drvdata(pdev);
 	struct alx_hw *hw = &alx->hw;
 
-	cancel_work_sync(&alx->link_check_wk);
-	cancel_work_sync(&alx->reset_wk);
-
 	/* restore permanent mac address */
 	alx_set_macaddr(hw, hw->perm_addr);
 
@@ -1873,45 +1894,59 @@ static void alx_remove(struct pci_dev *pdev)
 	iounmap(hw->hw_addr);
 	pci_release_mem_regions(pdev);
 
-	pci_disable_pcie_error_reporting(pdev);
 	pci_disable_device(pdev);
 
+	mutex_destroy(&alx->mtx);
+
 	free_netdev(alx->dev);
 }
 
-#ifdef CONFIG_PM_SLEEP
 static int alx_suspend(struct device *dev)
 {
-	struct pci_dev *pdev = to_pci_dev(dev);
-	struct alx_priv *alx = pci_get_drvdata(pdev);
+	struct alx_priv *alx = dev_get_drvdata(dev);
 
 	if (!netif_running(alx->dev))
 		return 0;
+
+	rtnl_lock();
 	netif_device_detach(alx->dev);
+
+	mutex_lock(&alx->mtx);
 	__alx_stop(alx);
+	mutex_unlock(&alx->mtx);
+	rtnl_unlock();
+
 	return 0;
 }
 
 static int alx_resume(struct device *dev)
 {
-	struct pci_dev *pdev = to_pci_dev(dev);
-	struct alx_priv *alx = pci_get_drvdata(pdev);
+	struct alx_priv *alx = dev_get_drvdata(dev);
 	struct alx_hw *hw = &alx->hw;
+	int err;
 
+	rtnl_lock();
+	mutex_lock(&alx->mtx);
 	alx_reset_phy(hw);
 
-	if (!netif_running(alx->dev))
-		return 0;
+	if (!netif_running(alx->dev)) {
+		err = 0;
+		goto unlock;
+	}
+
+	err = __alx_open(alx, true);
+	if (err)
+		goto unlock;
+
 	netif_device_attach(alx->dev);
-	return __alx_open(alx, true);
-}
 
-static SIMPLE_DEV_PM_OPS(alx_pm_ops, alx_suspend, alx_resume);
-#define ALX_PM_OPS      (&alx_pm_ops)
-#else
-#define ALX_PM_OPS      NULL
-#endif
+unlock:
+	mutex_unlock(&alx->mtx);
+	rtnl_unlock();
+	return err;
+}
 
+static DEFINE_SIMPLE_DEV_PM_OPS(alx_pm_ops, alx_suspend, alx_resume);
 
 static pci_ers_result_t alx_pci_error_detected(struct pci_dev *pdev,
 					       pci_channel_state_t state)
@@ -1922,7 +1957,7 @@ static pci_ers_result_t alx_pci_error_detected(struct pci_dev *pdev,
 
 	dev_info(&pdev->dev, "pci error detected\n");
 
-	rtnl_lock();
+	mutex_lock(&alx->mtx);
 
 	if (netif_running(netdev)) {
 		netif_device_detach(netdev);
@@ -1934,7 +1969,7 @@ static pci_ers_result_t alx_pci_error_detected(struct pci_dev *pdev,
 	else
 		pci_disable_device(pdev);
 
-	rtnl_unlock();
+	mutex_unlock(&alx->mtx);
 
 	return rc;
 }
@@ -1947,7 +1982,7 @@ static pci_ers_result_t alx_pci_error_slot_reset(struct pci_dev *pdev)
 
 	dev_info(&pdev->dev, "pci error slot reset\n");
 
-	rtnl_lock();
+	mutex_lock(&alx->mtx);
 
 	if (pci_enable_device(pdev)) {
 		dev_err(&pdev->dev, "Failed to re-enable PCI device after reset\n");
@@ -1960,9 +1995,7 @@ static pci_ers_result_t alx_pci_error_slot_reset(struct pci_dev *pdev)
 	if (!alx_reset_mac(hw))
 		rc = PCI_ERS_RESULT_RECOVERED;
 out:
-	pci_cleanup_aer_uncorrect_error_status(pdev);
-
-	rtnl_unlock();
+	mutex_unlock(&alx->mtx);
 
 	return rc;
 }
@@ -1974,14 +2007,14 @@ static void alx_pci_error_resume(struct pci_dev *pdev)
 
 	dev_info(&pdev->dev, "pci error resume\n");
 
-	rtnl_lock();
+	mutex_lock(&alx->mtx);
 
 	if (netif_running(netdev)) {
 		alx_activate(alx);
 		netif_device_attach(netdev);
 	}
 
-	rtnl_unlock();
+	mutex_unlock(&alx->mtx);
 }
 
 static const struct pci_error_handlers alx_err_handlers = {
@@ -2012,13 +2045,13 @@ static struct pci_driver alx_driver = {
 	.probe       = alx_probe,
 	.remove      = alx_remove,
 	.err_handler = &alx_err_handlers,
-	.driver.pm   = ALX_PM_OPS,
+	.driver.pm   = pm_sleep_ptr(&alx_pm_ops),
 };
 
 module_pci_driver(alx_driver);
 MODULE_DEVICE_TABLE(pci, alx_pci_tbl);
 MODULE_AUTHOR("Johannes Berg <johannes@sipsolutions.net>");
-MODULE_AUTHOR("Qualcomm Corporation, <nic-devel@qualcomm.com>");
+MODULE_AUTHOR("Qualcomm Corporation");
 MODULE_DESCRIPTION(
 	"Qualcomm Atheros(R) AR816x/AR817x PCI-E Ethernet Network Driver");
 MODULE_LICENSE("GPL");
diff --git a/drivers/net/ethernet/atheros/atl1c/Makefile b/drivers/net/ethernet/atheros/atl1c/Makefile
index c37d966952ee..02d025029554 100644
--- a/drivers/net/ethernet/atheros/atl1c/Makefile
+++ b/drivers/net/ethernet/atheros/atl1c/Makefile
@@ -1,2 +1,3 @@
+# SPDX-License-Identifier: GPL-2.0-only
 obj-$(CONFIG_ATL1C) += atl1c.o
 atl1c-objs := atl1c_main.o atl1c_hw.o atl1c_ethtool.o
diff --git a/drivers/net/ethernet/atheros/atl1c/atl1c.h b/drivers/net/ethernet/atheros/atl1c/atl1c.h
index c46b489ce9b4..63ba64dbb731 100644
--- a/drivers/net/ethernet/atheros/atl1c/atl1c.h
+++ b/drivers/net/ethernet/atheros/atl1c/atl1c.h
@@ -1,22 +1,9 @@
+/* SPDX-License-Identifier: GPL-2.0-or-later */
 /*
  * Copyright(c) 2008 - 2009 Atheros Corporation. All rights reserved.
  *
  * Derived from Intel e1000 driver
  * Copyright(c) 1999 - 2005 Intel Corporation. All rights reserved.
- *
- * This program is free software; you can redistribute it and/or modify it
- * under the terms of the GNU General Public License as published by the Free
- * Software Foundation; either version 2 of the License, or (at your option)
- * any later version.
- *
- * This program is distributed in the hope that it will be useful, but WITHOUT
- * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
- * FITNESS FOR A PARTICULAR PURPOSE.  See the GNU General Public License for
- * more details.
- *
- * You should have received a copy of the GNU General Public License along with
- * this program; if not, write to the Free Software Foundation, Inc., 59
- * Temple Place - Suite 330, Boston, MA  02111-1307, USA.
  */
 
 #ifndef _ATL1C_H_
@@ -76,7 +63,7 @@
 
 #define AT_MAX_RECEIVE_QUEUE    4
 #define AT_DEF_RECEIVE_QUEUE	1
-#define AT_MAX_TRANSMIT_QUEUE	2
+#define AT_MAX_TRANSMIT_QUEUE  4
 
 #define AT_DMA_HI_ADDR_MASK     0xffffffff00000000ULL
 #define AT_DMA_LO_ADDR_MASK     0x00000000ffffffffULL
@@ -254,6 +241,8 @@ struct atl1c_tpd_ext_desc {
 #define RRS_PACKET_PROT_IS_IPV6_ONLY(word) \
 	((((word) >> RRS_PROT_ID_SHIFT) & RRS_PROT_ID_MASK) == 6)
 
+#define RRS_MT_PROT_ID_TCPUDP	BIT(19)
+
 struct atl1c_recv_ret_status {
 	__le32  word0;
 	__le32	rss_hash;
@@ -302,11 +291,7 @@ enum atl1c_nic_type {
 	athr_l2c_b2,
 	athr_l1d,
 	athr_l1d_2,
-};
-
-enum atl1c_trans_queue {
-	atl1c_trans_normal = 0,
-	atl1c_trans_high = 1
+	athr_mt,
 };
 
 struct atl1c_hw_stats {
@@ -380,6 +365,7 @@ struct atl1c_hw {
 	u16 phy_id1;
 	u16 phy_id2;
 
+	spinlock_t intr_mask_lock;	/* protect the intr_mask */
 	u32 intr_mask;
 
 	u8 preamble_len;
@@ -484,13 +470,16 @@ struct atl1c_buffer {
 
 /* transimit packet descriptor (tpd) ring */
 struct atl1c_tpd_ring {
+	struct atl1c_adapter *adapter;
 	void *desc;		/* descriptor ring virtual address */
 	dma_addr_t dma;		/* descriptor ring physical address */
+	u16 num;
 	u16 size;		/* descriptor ring length in bytes */
 	u16 count;		/* number of descriptors in the ring */
 	u16 next_to_use;
 	atomic_t next_to_clean;
 	struct atl1c_buffer *buffer_info;
+	struct napi_struct napi;
 };
 
 /* receive free descriptor (rfd) ring */
@@ -506,26 +495,27 @@ struct atl1c_rfd_ring {
 
 /* receive return descriptor (rrd) ring */
 struct atl1c_rrd_ring {
+	struct atl1c_adapter *adapter;
 	void *desc;		/* descriptor ring virtual address */
 	dma_addr_t dma;		/* descriptor ring physical address */
+	u16 num;
 	u16 size;		/* descriptor ring length in bytes */
 	u16 count;		/* number of descriptors in the ring */
 	u16 next_to_use;
 	u16 next_to_clean;
+	struct napi_struct napi;
 };
 
 /* board specific private data structure */
 struct atl1c_adapter {
 	struct net_device   *netdev;
 	struct pci_dev      *pdev;
-	struct napi_struct  napi;
-	struct page         *rx_page;
-	unsigned int	    rx_page_offset;
-	unsigned int	    rx_frag_size;
 	struct atl1c_hw        hw;
 	struct atl1c_hw_stats  hw_stats;
 	struct mii_if_info  mii;    /* MII interface info */
 	u16 rx_buffer_len;
+	unsigned int tx_queue_count;
+	unsigned int rx_queue_count;
 
 	unsigned long flags;
 #define __AT_TESTING        0x0001
@@ -551,8 +541,8 @@ struct atl1c_adapter {
 	/* All Descriptor memory */
 	struct atl1c_ring_header ring_header;
 	struct atl1c_tpd_ring tpd_ring[AT_MAX_TRANSMIT_QUEUE];
-	struct atl1c_rfd_ring rfd_ring;
-	struct atl1c_rrd_ring rrd_ring;
+	struct atl1c_rfd_ring rfd_ring[AT_MAX_RECEIVE_QUEUE];
+	struct atl1c_rrd_ring rrd_ring[AT_MAX_RECEIVE_QUEUE];
 	u32 bd_number;     /* board number;*/
 };
 
@@ -596,7 +586,6 @@ struct atl1c_adapter {
 		readl(((a)->hw_addr + reg) + ((offset) << 2)))
 
 extern char atl1c_driver_name[];
-extern char atl1c_driver_version[];
 
 void atl1c_reinit_locked(struct atl1c_adapter *adapter);
 s32 atl1c_reset_hw(struct atl1c_hw *hw);
diff --git a/drivers/net/ethernet/atheros/atl1c/atl1c_ethtool.c b/drivers/net/ethernet/atheros/atl1c/atl1c_ethtool.c
index cfe86a20c899..0bce122c68f1 100644
--- a/drivers/net/ethernet/atheros/atl1c/atl1c_ethtool.c
+++ b/drivers/net/ethernet/atheros/atl1c/atl1c_ethtool.c
@@ -1,23 +1,9 @@
+// SPDX-License-Identifier: GPL-2.0-or-later
 /*
  * Copyright(c) 2009 - 2009 Atheros Corporation. All rights reserved.
  *
  * Derived from Intel e1000 driver
  * Copyright(c) 1999 - 2005 Intel Corporation. All rights reserved.
- *
- * This program is free software; you can redistribute it and/or modify it
- * under the terms of the GNU General Public License as published by the Free
- * Software Foundation; either version 2 of the License, or (at your option)
- * any later version.
- *
- * This program is distributed in the hope that it will be useful, but WITHOUT
- * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
- * FITNESS FOR A PARTICULAR PURPOSE.  See the GNU General Public License for
- * more details.
- *
- * You should have received a copy of the GNU General Public License along with
- * this program; if not, write to the Free Software Foundation, Inc., 59
- * Temple Place - Suite 330, Boston, MA  02111-1307, USA.
- *
  */
 
 #include <linux/netdevice.h>
@@ -209,8 +195,8 @@ static int atl1c_get_eeprom(struct net_device *netdev,
 	first_dword = eeprom->offset >> 2;
 	last_dword = (eeprom->offset + eeprom->len - 1) >> 2;
 
-	eeprom_buff = kmalloc(sizeof(u32) *
-			(last_dword - first_dword + 1), GFP_KERNEL);
+	eeprom_buff = kmalloc_array(last_dword - first_dword + 1, sizeof(u32),
+				    GFP_KERNEL);
 	if (eeprom_buff == NULL)
 		return -ENOMEM;
 
@@ -234,10 +220,8 @@ static void atl1c_get_drvinfo(struct net_device *netdev,
 {
 	struct atl1c_adapter *adapter = netdev_priv(netdev);
 
-	strlcpy(drvinfo->driver,  atl1c_driver_name, sizeof(drvinfo->driver));
-	strlcpy(drvinfo->version, atl1c_driver_version,
-		sizeof(drvinfo->version));
-	strlcpy(drvinfo->bus_info, pci_name(adapter->pdev),
+	strscpy(drvinfo->driver,  atl1c_driver_name, sizeof(drvinfo->driver));
+	strscpy(drvinfo->bus_info, pci_name(adapter->pdev),
 		sizeof(drvinfo->bus_info));
 }
 
diff --git a/drivers/net/ethernet/atheros/atl1c/atl1c_hw.c b/drivers/net/ethernet/atheros/atl1c/atl1c_hw.c
index 73efdb05a490..f19370c33444 100644
--- a/drivers/net/ethernet/atheros/atl1c/atl1c_hw.c
+++ b/drivers/net/ethernet/atheros/atl1c/atl1c_hw.c
@@ -1,22 +1,9 @@
+// SPDX-License-Identifier: GPL-2.0-or-later
 /*
  * Copyright(c) 2007 Atheros Corporation. All rights reserved.
  *
  * Derived from Intel e1000 driver
  * Copyright(c) 1999 - 2005 Intel Corporation. All rights reserved.
- *
- * This program is free software; you can redistribute it and/or modify it
- * under the terms of the GNU General Public License as published by the Free
- * Software Foundation; either version 2 of the License, or (at your option)
- * any later version.
- *
- * This program is distributed in the hope that it will be useful, but WITHOUT
- * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
- * FITNESS FOR A PARTICULAR PURPOSE.  See the GNU General Public License for
- * more details.
- *
- * You should have received a copy of the GNU General Public License along with
- * this program; if not, write to the Free Software Foundation, Inc., 59
- * Temple Place - Suite 330, Boston, MA  02111-1307, USA.
  */
 #include <linux/pci.h>
 #include <linux/delay.h>
@@ -607,6 +594,11 @@ int atl1c_phy_init(struct atl1c_hw *hw)
 	int ret_val;
 	u16 mii_bmcr_data = BMCR_RESET;
 
+	if (hw->nic_type == athr_mt) {
+		hw->phy_configured = true;
+		return 0;
+	}
+
 	if ((atl1c_read_phy_reg(hw, MII_PHYSID1, &hw->phy_id1) != 0) ||
 		(atl1c_read_phy_reg(hw, MII_PHYSID2, &hw->phy_id2) != 0)) {
 		dev_err(&pdev->dev, "Error get phy ID\n");
@@ -649,6 +641,23 @@ int atl1c_phy_init(struct atl1c_hw *hw)
 	return 0;
 }
 
+bool atl1c_get_link_status(struct atl1c_hw *hw)
+{
+	u16 phy_data;
+
+	if (hw->nic_type == athr_mt) {
+		u32 spd;
+
+		AT_READ_REG(hw, REG_MT_SPEED, &spd);
+		return !!spd;
+	}
+
+	/* MII_BMSR must be read twice */
+	atl1c_read_phy_reg(hw, MII_BMSR, &phy_data);
+	atl1c_read_phy_reg(hw, MII_BMSR, &phy_data);
+	return !!(phy_data & BMSR_LSTATUS);
+}
+
 /*
  * Detects the current speed and duplex settings of the hardware.
  *
@@ -661,6 +670,15 @@ int atl1c_get_speed_and_duplex(struct atl1c_hw *hw, u16 *speed, u16 *duplex)
 	int err;
 	u16 phy_data;
 
+	if (hw->nic_type == athr_mt) {
+		u32 spd;
+
+		AT_READ_REG(hw, REG_MT_SPEED, &spd);
+		*speed = spd;
+		*duplex = FULL_DUPLEX;
+		return 0;
+	}
+
 	/* Read   PHY Specific Status Register (17) */
 	err = atl1c_read_phy_reg(hw, MII_GIGA_PSSR, &phy_data);
 	if (err)
@@ -699,15 +717,12 @@ int atl1c_phy_to_ps_link(struct atl1c_hw *hw)
 	int ret = 0;
 	u16 autoneg_advertised = ADVERTISED_10baseT_Half;
 	u16 save_autoneg_advertised;
-	u16 phy_data;
 	u16 mii_lpa_data;
 	u16 speed = SPEED_0;
 	u16 duplex = FULL_DUPLEX;
 	int i;
 
-	atl1c_read_phy_reg(hw, MII_BMSR, &phy_data);
-	atl1c_read_phy_reg(hw, MII_BMSR, &phy_data);
-	if (phy_data & BMSR_LSTATUS) {
+	if (atl1c_get_link_status(hw)) {
 		atl1c_read_phy_reg(hw, MII_LPA, &mii_lpa_data);
 		if (mii_lpa_data & LPA_10FULL)
 			autoneg_advertised = ADVERTISED_10baseT_Full;
@@ -730,9 +745,7 @@ int atl1c_phy_to_ps_link(struct atl1c_hw *hw)
 		if (mii_lpa_data) {
 			for (i = 0; i < AT_SUSPEND_LINK_TIMEOUT; i++) {
 				mdelay(100);
-				atl1c_read_phy_reg(hw, MII_BMSR, &phy_data);
-				atl1c_read_phy_reg(hw, MII_BMSR, &phy_data);
-				if (phy_data & BMSR_LSTATUS) {
+				if (atl1c_get_link_status(hw)) {
 					if (atl1c_get_speed_and_duplex(hw, &speed,
 									&duplex) != 0)
 						dev_dbg(&pdev->dev,
diff --git a/drivers/net/ethernet/atheros/atl1c/atl1c_hw.h b/drivers/net/ethernet/atheros/atl1c/atl1c_hw.h
index 21d8c4dbdbe1..c567c920628f 100644
--- a/drivers/net/ethernet/atheros/atl1c/atl1c_hw.h
+++ b/drivers/net/ethernet/atheros/atl1c/atl1c_hw.h
@@ -1,22 +1,9 @@
+/* SPDX-License-Identifier: GPL-2.0-or-later */
 /*
  * Copyright(c) 2008 - 2009 Atheros Corporation. All rights reserved.
  *
  * Derived from Intel e1000 driver
  * Copyright(c) 1999 - 2005 Intel Corporation. All rights reserved.
- *
- * This program is free software; you can redistribute it and/or modify it
- * under the terms of the GNU General Public License as published by the Free
- * Software Foundation; either version 2 of the License, or (at your option)
- * any later version.
- *
- * This program is distributed in the hope that it will be useful, but WITHOUT
- * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
- * FITNESS FOR A PARTICULAR PURPOSE.  See the GNU General Public License for
- * more details.
- *
- * You should have received a copy of the GNU General Public License along with
- * this program; if not, write to the Free Software Foundation, Inc., 59
- * Temple Place - Suite 330, Boston, MA  02111-1307, USA.
  */
 
 #ifndef _ATL1C_HW_H_
@@ -39,6 +26,7 @@ void atl1c_phy_disable(struct atl1c_hw *hw);
 void atl1c_hw_set_mac_addr(struct atl1c_hw *hw, u8 *mac_addr);
 int atl1c_phy_reset(struct atl1c_hw *hw);
 int atl1c_read_mac_addr(struct atl1c_hw *hw);
+bool atl1c_get_link_status(struct atl1c_hw *hw);
 int atl1c_get_speed_and_duplex(struct atl1c_hw *hw, u16 *speed, u16 *duplex);
 u32 atl1c_hash_mc_addr(struct atl1c_hw *hw, u8 *mc_addr);
 void atl1c_hash_set(struct atl1c_hw *hw, u32 hash_value);
@@ -540,15 +528,24 @@ void atl1c_post_phy_linkchg(struct atl1c_hw *hw, u16 link_speed);
 #define REG_RX_BASE_ADDR_HI		0x1540
 #define REG_TX_BASE_ADDR_HI		0x1544
 #define REG_RFD0_HEAD_ADDR_LO		0x1550
+#define REG_RFD1_HEAD_ADDR_LO          0x1554
+#define REG_RFD2_HEAD_ADDR_LO          0x1558
+#define REG_RFD3_HEAD_ADDR_LO          0x155C
 #define REG_RFD_RING_SIZE		0x1560
 #define RFD_RING_SIZE_MASK		0x0FFF
 #define REG_RX_BUF_SIZE			0x1564
 #define RX_BUF_SIZE_MASK		0xFFFF
 #define REG_RRD0_HEAD_ADDR_LO		0x1568
+#define REG_RRD1_HEAD_ADDR_LO          0x156C
+#define REG_RRD2_HEAD_ADDR_LO          0x1570
+#define REG_RRD3_HEAD_ADDR_LO          0x1574
 #define REG_RRD_RING_SIZE		0x1578
 #define RRD_RING_SIZE_MASK		0x0FFF
 #define REG_TPD_PRI1_ADDR_LO		0x157C
 #define REG_TPD_PRI0_ADDR_LO		0x1580
+#define REG_TPD_PRI2_ADDR_LO           0x1F10
+#define REG_TPD_PRI3_ADDR_LO           0x1F14
+
 #define REG_TPD_RING_SIZE		0x1584
 #define TPD_RING_SIZE_MASK		0xFFFF
 
@@ -667,15 +664,26 @@ void atl1c_post_phy_linkchg(struct atl1c_hw *hw, u16 link_speed);
 /* Mail box */
 #define MB_RFDX_PROD_IDX_MASK		0xFFFF
 #define REG_MB_RFD0_PROD_IDX		0x15E0
+#define REG_MB_RFD1_PROD_IDX           0x15E4
+#define REG_MB_RFD2_PROD_IDX           0x15E8
+#define REG_MB_RFD3_PROD_IDX           0x15EC
 
 #define REG_TPD_PRI1_PIDX               0x15F0	/* 16bit,hi-tpd producer idx */
 #define REG_TPD_PRI0_PIDX		0x15F2	/* 16bit,lo-tpd producer idx */
 #define REG_TPD_PRI1_CIDX		0x15F4	/* 16bit,hi-tpd consumer idx */
 #define REG_TPD_PRI0_CIDX		0x15F6	/* 16bit,lo-tpd consumer idx */
+#define REG_TPD_PRI3_PIDX              0x1F18
+#define REG_TPD_PRI2_PIDX              0x1F1A
+#define REG_TPD_PRI3_CIDX              0x1F1C
+#define REG_TPD_PRI2_CIDX              0x1F1E
+
 
 #define REG_MB_RFD01_CONS_IDX		0x15F8
 #define MB_RFD0_CONS_IDX_MASK		0x0000FFFF
 #define MB_RFD1_CONS_IDX_MASK		0xFFFF0000
+#define REG_MB_RFD23_CONS_IDX          0x15FC
+#define MB_RFD2_CONS_IDX_MASK          0x0000FFFF
+#define MB_RFD3_CONS_IDX_MASK          0xFFFF0000
 
 /* Interrupt Status Register */
 #define REG_ISR    			0x1600
@@ -699,7 +707,7 @@ void atl1c_post_phy_linkchg(struct atl1c_hw *hw, u16 link_speed);
 /* GPHY low power state interrupt */
 #define ISR_GPHY_LPW           		0x00002000
 #define ISR_TXQ_TO_RST			0x00004000
-#define ISR_TX_PKT			0x00008000
+#define ISR_TX_PKT_0                   0x00008000
 #define ISR_RX_PKT_0			0x00010000
 #define ISR_RX_PKT_1			0x00020000
 #define ISR_RX_PKT_2			0x00040000
@@ -711,6 +719,9 @@ void atl1c_post_phy_linkchg(struct atl1c_hw *hw, u16 link_speed);
 #define ISR_NFERR_DETECTED		0x01000000
 #define ISR_CERR_DETECTED		0x02000000
 #define ISR_PHY_LINKDOWN		0x04000000
+#define ISR_TX_PKT_1                   0x10000000
+#define ISR_TX_PKT_2                   0x20000000
+#define ISR_TX_PKT_3                   0x40000000
 #define ISR_DIS_INT			0x80000000
 
 /* Interrupt Mask Register */
@@ -725,11 +736,15 @@ void atl1c_post_phy_linkchg(struct atl1c_hw *hw, u16 link_speed);
 		ISR_TXQ_TO_RST  |\
 		ISR_DMAW_TO_RST	|\
 		ISR_GPHY	|\
-		ISR_TX_PKT	|\
-		ISR_RX_PKT_0	|\
 		ISR_GPHY_LPW    |\
 		ISR_PHY_LINKDOWN)
 
+#define ISR_TX_PKT     (			\
+	ISR_TX_PKT_0    |			\
+	ISR_TX_PKT_1    |			\
+	ISR_TX_PKT_2    |			\
+	ISR_TX_PKT_3)
+
 #define ISR_RX_PKT 	(\
 	ISR_RX_PKT_0    |\
 	ISR_RX_PKT_1    |\
@@ -777,6 +792,14 @@ void atl1c_post_phy_linkchg(struct atl1c_hw *hw, u16 link_speed);
 #define REG_DEBUG_DATA0 		0x1900
 #define REG_DEBUG_DATA1 		0x1904
 
+#define REG_MT_MAGIC			0x1F00
+#define REG_MT_MODE			0x1F04
+#define REG_MT_SPEED			0x1F08
+#define REG_MT_VERSION			0x1F0C
+
+#define MT_MAGIC			0xaabb1234
+#define MT_MODE_4Q			BIT(0)
+
 #define L1D_MPW_PHYID1			0xD01C  /* V7 */
 #define L1D_MPW_PHYID2			0xD01D  /* V1-V6 */
 #define L1D_MPW_PHYID3			0xD01E  /* V8 */
diff --git a/drivers/net/ethernet/atheros/atl1c/atl1c_main.c b/drivers/net/ethernet/atheros/atl1c/atl1c_main.c
index 7e913d8331c3..7efa3fc257b3 100644
--- a/drivers/net/ethernet/atheros/atl1c/atl1c_main.c
+++ b/drivers/net/ethernet/atheros/atl1c/atl1c_main.c
@@ -1,29 +1,14 @@
+// SPDX-License-Identifier: GPL-2.0-or-later
 /*
  * Copyright(c) 2008 - 2009 Atheros Corporation. All rights reserved.
  *
  * Derived from Intel e1000 driver
  * Copyright(c) 1999 - 2005 Intel Corporation. All rights reserved.
- *
- * This program is free software; you can redistribute it and/or modify it
- * under the terms of the GNU General Public License as published by the Free
- * Software Foundation; either version 2 of the License, or (at your option)
- * any later version.
- *
- * This program is distributed in the hope that it will be useful, but WITHOUT
- * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
- * FITNESS FOR A PARTICULAR PURPOSE.  See the GNU General Public License for
- * more details.
- *
- * You should have received a copy of the GNU General Public License along with
- * this program; if not, write to the Free Software Foundation, Inc., 59
- * Temple Place - Suite 330, Boston, MA  02111-1307, USA.
  */
 
 #include "atl1c.h"
 
-#define ATL1C_DRV_VERSION "1.0.1.1-NAPI"
 char atl1c_driver_name[] = "atl1c";
-char atl1c_driver_version[] = ATL1C_DRV_VERSION;
 
 /*
  * atl1c_pci_tbl - PCI Device ID Table
@@ -47,23 +32,55 @@ static const struct pci_device_id atl1c_pci_tbl[] = {
 MODULE_DEVICE_TABLE(pci, atl1c_pci_tbl);
 
 MODULE_AUTHOR("Jie Yang");
-MODULE_AUTHOR("Qualcomm Atheros Inc., <nic-devel@qualcomm.com>");
+MODULE_AUTHOR("Qualcomm Atheros Inc.");
 MODULE_DESCRIPTION("Qualcomm Atheros 100/1000M Ethernet Network Driver");
 MODULE_LICENSE("GPL");
-MODULE_VERSION(ATL1C_DRV_VERSION);
+
+struct atl1c_qregs {
+	u16 tpd_addr_lo;
+	u16 tpd_prod;
+	u16 tpd_cons;
+	u16 rfd_addr_lo;
+	u16 rrd_addr_lo;
+	u16 rfd_prod;
+	u32 tx_isr;
+	u32 rx_isr;
+};
+
+static struct atl1c_qregs atl1c_qregs[AT_MAX_TRANSMIT_QUEUE] = {
+	{
+		REG_TPD_PRI0_ADDR_LO, REG_TPD_PRI0_PIDX, REG_TPD_PRI0_CIDX,
+		REG_RFD0_HEAD_ADDR_LO, REG_RRD0_HEAD_ADDR_LO,
+		REG_MB_RFD0_PROD_IDX, ISR_TX_PKT_0, ISR_RX_PKT_0
+	},
+	{
+		REG_TPD_PRI1_ADDR_LO, REG_TPD_PRI1_PIDX, REG_TPD_PRI1_CIDX,
+		REG_RFD1_HEAD_ADDR_LO, REG_RRD1_HEAD_ADDR_LO,
+		REG_MB_RFD1_PROD_IDX, ISR_TX_PKT_1, ISR_RX_PKT_1
+	},
+	{
+		REG_TPD_PRI2_ADDR_LO, REG_TPD_PRI2_PIDX, REG_TPD_PRI2_CIDX,
+		REG_RFD2_HEAD_ADDR_LO, REG_RRD2_HEAD_ADDR_LO,
+		REG_MB_RFD2_PROD_IDX, ISR_TX_PKT_2, ISR_RX_PKT_2
+	},
+	{
+		REG_TPD_PRI3_ADDR_LO, REG_TPD_PRI3_PIDX, REG_TPD_PRI3_CIDX,
+		REG_RFD3_HEAD_ADDR_LO, REG_RRD3_HEAD_ADDR_LO,
+		REG_MB_RFD3_PROD_IDX, ISR_TX_PKT_3, ISR_RX_PKT_3
+	},
+};
 
 static int atl1c_stop_mac(struct atl1c_hw *hw);
 static void atl1c_disable_l0s_l1(struct atl1c_hw *hw);
 static void atl1c_set_aspm(struct atl1c_hw *hw, u16 link_speed);
 static void atl1c_start_mac(struct atl1c_adapter *adapter);
-static void atl1c_clean_rx_irq(struct atl1c_adapter *adapter,
-		   int *work_done, int work_to_do);
 static int atl1c_up(struct atl1c_adapter *adapter);
 static void atl1c_down(struct atl1c_adapter *adapter);
 static int atl1c_reset_mac(struct atl1c_hw *hw);
 static void atl1c_reset_dma_ring(struct atl1c_adapter *adapter);
 static int atl1c_configure(struct atl1c_adapter *adapter);
-static int atl1c_alloc_rx_buffer(struct atl1c_adapter *adapter);
+static int atl1c_alloc_rx_buffer(struct atl1c_adapter *adapter, u32 queue,
+				 bool napi_mode);
 
 
 static const u32 atl1c_default_msg = NETIF_MSG_DRV | NETIF_MSG_PROBE |
@@ -190,16 +207,6 @@ static inline void atl1c_irq_disable(struct atl1c_adapter *adapter)
 	synchronize_irq(adapter->pdev->irq);
 }
 
-/**
- * atl1c_irq_reset - reset interrupt confiure on the NIC
- * @adapter: board private structure
- */
-static inline void atl1c_irq_reset(struct atl1c_adapter *adapter)
-{
-	atomic_set(&adapter->irq_sem, 1);
-	atl1c_irq_enable(adapter);
-}
-
 /*
  * atl1c_wait_until_idle - wait up to AT_HW_MAX_IDLE_DELAY reads
  * of the idle status register until the device is actually idle
@@ -220,11 +227,12 @@ static u32 atl1c_wait_until_idle(struct atl1c_hw *hw, u32 modu_ctrl)
 
 /**
  * atl1c_phy_config - Timer Call-back
- * @data: pointer to netdev cast into an unsigned long
+ * @t: timer list containing pointer to netdev cast into an unsigned long
  */
-static void atl1c_phy_config(unsigned long data)
+static void atl1c_phy_config(struct timer_list *t)
 {
-	struct atl1c_adapter *adapter = (struct atl1c_adapter *) data;
+	struct atl1c_adapter *adapter = timer_container_of(adapter, t,
+							   phy_config_timer);
 	struct atl1c_hw *hw = &adapter->hw;
 	unsigned long flags;
 
@@ -235,7 +243,6 @@ static void atl1c_phy_config(unsigned long data)
 
 void atl1c_reinit_locked(struct atl1c_adapter *adapter)
 {
-	WARN_ON(in_interrupt());
 	atl1c_down(adapter);
 	atl1c_up(adapter);
 	clear_bit(__AT_RESETTING, &adapter->flags);
@@ -248,15 +255,14 @@ static void atl1c_check_link_status(struct atl1c_adapter *adapter)
 	struct pci_dev    *pdev   = adapter->pdev;
 	int err;
 	unsigned long flags;
-	u16 speed, duplex, phy_data;
+	u16 speed, duplex;
+	bool link;
 
 	spin_lock_irqsave(&adapter->mdio_lock, flags);
-	/* MII_BMSR must read twise */
-	atl1c_read_phy_reg(hw, MII_BMSR, &phy_data);
-	atl1c_read_phy_reg(hw, MII_BMSR, &phy_data);
+	link = atl1c_get_link_status(hw);
 	spin_unlock_irqrestore(&adapter->mdio_lock, flags);
 
-	if ((phy_data & BMSR_LSTATUS) == 0) {
+	if (!link) {
 		/* link down */
 		netif_carrier_off(netdev);
 		hw->hibernate = true;
@@ -300,16 +306,13 @@ static void atl1c_link_chg_event(struct atl1c_adapter *adapter)
 {
 	struct net_device *netdev = adapter->netdev;
 	struct pci_dev    *pdev   = adapter->pdev;
-	u16 phy_data;
-	u16 link_up;
+	bool link;
 
 	spin_lock(&adapter->mdio_lock);
-	atl1c_read_phy_reg(&adapter->hw, MII_BMSR, &phy_data);
-	atl1c_read_phy_reg(&adapter->hw, MII_BMSR, &phy_data);
+	link = atl1c_get_link_status(&adapter->hw);
 	spin_unlock(&adapter->mdio_lock);
-	link_up = phy_data & BMSR_LSTATUS;
 	/* notify upper layer link down ASAP */
-	if (!link_up) {
+	if (!link) {
 		if (netif_carrier_ok(netdev)) {
 			/* old link state: Up */
 			netif_carrier_off(netdev);
@@ -354,15 +357,16 @@ static void atl1c_common_task(struct work_struct *work)
 
 static void atl1c_del_timer(struct atl1c_adapter *adapter)
 {
-	del_timer_sync(&adapter->phy_config_timer);
+	timer_delete_sync(&adapter->phy_config_timer);
 }
 
 
 /**
  * atl1c_tx_timeout - Respond to a Tx Hang
  * @netdev: network interface device structure
+ * @txqueue: index of hanging tx queue
  */
-static void atl1c_tx_timeout(struct net_device *netdev)
+static void atl1c_tx_timeout(struct net_device *netdev, unsigned int txqueue)
 {
 	struct atl1c_adapter *adapter = netdev_priv(netdev);
 
@@ -451,7 +455,7 @@ static void atl1c_restore_vlan(struct atl1c_adapter *adapter)
 }
 
 /**
- * atl1c_set_mac - Change the Ethernet Address of the NIC
+ * atl1c_set_mac_addr - Change the Ethernet Address of the NIC
  * @netdev: network interface device structure
  * @p: pointer to an address structure
  *
@@ -468,7 +472,7 @@ static int atl1c_set_mac_addr(struct net_device *netdev, void *p)
 	if (netif_running(netdev))
 		return -EBUSY;
 
-	memcpy(netdev->dev_addr, addr->sa_data, netdev->addr_len);
+	eth_hw_addr_set(netdev, addr->sa_data);
 	memcpy(adapter->hw.mac_addr, addr->sa_data, netdev->addr_len);
 
 	atl1c_hw_set_mac_addr(&adapter->hw, adapter->hw.mac_addr);
@@ -479,20 +483,18 @@ static int atl1c_set_mac_addr(struct net_device *netdev, void *p)
 static void atl1c_set_rxbufsize(struct atl1c_adapter *adapter,
 				struct net_device *dev)
 {
-	unsigned int head_size;
 	int mtu = dev->mtu;
 
 	adapter->rx_buffer_len = mtu > AT_RX_BUF_SIZE ?
 		roundup(mtu + ETH_HLEN + ETH_FCS_LEN + VLAN_HLEN, 8) : AT_RX_BUF_SIZE;
-
-	head_size = SKB_DATA_ALIGN(adapter->rx_buffer_len + NET_SKB_PAD) +
-		    SKB_DATA_ALIGN(sizeof(struct skb_shared_info));
-	adapter->rx_frag_size = roundup_pow_of_two(head_size);
 }
 
 static netdev_features_t atl1c_fix_features(struct net_device *netdev,
 	netdev_features_t features)
 {
+	struct atl1c_adapter *adapter = netdev_priv(netdev);
+	struct atl1c_hw *hw = &adapter->hw;
+
 	/*
 	 * Since there is no support for separate rx/tx vlan accel
 	 * enable/disable make sure tx flag is always in same state as rx.
@@ -502,8 +504,10 @@ static netdev_features_t atl1c_fix_features(struct net_device *netdev,
 	else
 		features &= ~NETIF_F_HW_VLAN_CTAG_TX;
 
-	if (netdev->mtu > MAX_TSO_FRAME_SIZE)
-		features &= ~(NETIF_F_TSO | NETIF_F_TSO6);
+	if (hw->nic_type != athr_mt) {
+		if (netdev->mtu > MAX_TSO_FRAME_SIZE)
+			features &= ~(NETIF_F_TSO | NETIF_F_TSO6);
+	}
 
 	return features;
 }
@@ -530,9 +534,12 @@ static void atl1c_set_max_mtu(struct net_device *netdev)
 	case athr_l1d:
 	case athr_l1d_2:
 		netdev->max_mtu = MAX_JUMBO_FRAME_SIZE -
-				  (ETH_HLEN + ETH_FCS_LEN + VLAN_HLEN);
+			(ETH_HLEN + ETH_FCS_LEN + VLAN_HLEN);
 		break;
-	/* The 10/100 devices don't support jumbo packets, max_mtu 1500 */
+	case athr_mt:
+		netdev->max_mtu = 9500;
+		break;
+		/* The 10/100 devices don't support jumbo packets, max_mtu 1500 */
 	default:
 		netdev->max_mtu = ETH_DATA_LEN;
 		break;
@@ -554,7 +561,7 @@ static int atl1c_change_mtu(struct net_device *netdev, int new_mtu)
 	if (netif_running(netdev)) {
 		while (test_and_set_bit(__AT_RESETTING, &adapter->flags))
 			msleep(1);
-		netdev->mtu = new_mtu;
+		WRITE_ONCE(netdev->mtu, new_mtu);
 		adapter->hw.max_frame_size = new_mtu;
 		atl1c_set_rxbufsize(adapter, netdev);
 		atl1c_down(adapter);
@@ -657,29 +664,26 @@ static int atl1c_alloc_queues(struct atl1c_adapter *adapter)
 	return 0;
 }
 
-static void atl1c_set_mac_type(struct atl1c_hw *hw)
+static enum atl1c_nic_type atl1c_get_mac_type(struct pci_dev *pdev,
+					      u8 __iomem *hw_addr)
 {
-	switch (hw->device_id) {
+	switch (pdev->device) {
 	case PCI_DEVICE_ID_ATTANSIC_L2C:
-		hw->nic_type = athr_l2c;
-		break;
+		return athr_l2c;
 	case PCI_DEVICE_ID_ATTANSIC_L1C:
-		hw->nic_type = athr_l1c;
-		break;
+		return athr_l1c;
 	case PCI_DEVICE_ID_ATHEROS_L2C_B:
-		hw->nic_type = athr_l2c_b;
-		break;
+		return athr_l2c_b;
 	case PCI_DEVICE_ID_ATHEROS_L2C_B2:
-		hw->nic_type = athr_l2c_b2;
-		break;
+		return athr_l2c_b2;
 	case PCI_DEVICE_ID_ATHEROS_L1D:
-		hw->nic_type = athr_l1d;
-		break;
+		return athr_l1d;
 	case PCI_DEVICE_ID_ATHEROS_L1D_2_0:
-		hw->nic_type = athr_l1d_2;
-		break;
+		if (readl(hw_addr + REG_MT_MAGIC) == MT_MAGIC)
+			return athr_mt;
+		return athr_l1d_2;
 	default:
-		break;
+		return athr_l1c;
 	}
 }
 
@@ -687,7 +691,6 @@ static int atl1c_setup_mac_funcs(struct atl1c_hw *hw)
 {
 	u32 link_ctrl_data;
 
-	atl1c_set_mac_type(hw);
 	AT_READ_REG(hw, REG_LINK_CTRL, &link_ctrl_data);
 
 	hw->ctrl_flags = ATL1C_INTR_MODRT_ENABLE  |
@@ -778,14 +781,14 @@ static int atl1c_sw_init(struct atl1c_adapter *adapter)
 	struct atl1c_hw *hw   = &adapter->hw;
 	struct pci_dev	*pdev = adapter->pdev;
 	u32 revision;
-
+	int i;
 
 	adapter->wol = 0;
 	device_set_wakeup_enable(&pdev->dev, false);
 	adapter->link_speed = SPEED_0;
 	adapter->link_duplex = FULL_DUPLEX;
 	adapter->tpd_ring[0].count = 1024;
-	adapter->rfd_ring.count = 512;
+	adapter->rfd_ring[0].count = 512;
 
 	hw->vendor_id = pdev->vendor;
 	hw->device_id = pdev->device;
@@ -803,6 +806,10 @@ static int atl1c_sw_init(struct atl1c_adapter *adapter)
 	atl1c_patch_assign(hw);
 
 	hw->intr_mask = IMR_NORMAL_MASK;
+	for (i = 0; i < adapter->tx_queue_count; ++i)
+		hw->intr_mask |= atl1c_qregs[i].tx_isr;
+	for (i = 0; i < adapter->rx_queue_count; ++i)
+		hw->intr_mask |= atl1c_qregs[i].rx_isr;
 	hw->phy_configured = false;
 	hw->preamble_len = 7;
 	hw->max_frame_size = adapter->netdev->mtu;
@@ -828,32 +835,34 @@ static int atl1c_sw_init(struct atl1c_adapter *adapter)
 	atl1c_set_rxbufsize(adapter, adapter->netdev);
 	atomic_set(&adapter->irq_sem, 1);
 	spin_lock_init(&adapter->mdio_lock);
+	spin_lock_init(&adapter->hw.intr_mask_lock);
 	set_bit(__AT_DOWN, &adapter->flags);
 
 	return 0;
 }
 
 static inline void atl1c_clean_buffer(struct pci_dev *pdev,
-				struct atl1c_buffer *buffer_info)
+				      struct atl1c_buffer *buffer_info,
+				      int budget)
 {
 	u16 pci_driection;
 	if (buffer_info->flags & ATL1C_BUFFER_FREE)
 		return;
 	if (buffer_info->dma) {
 		if (buffer_info->flags & ATL1C_PCIMAP_FROMDEVICE)
-			pci_driection = PCI_DMA_FROMDEVICE;
+			pci_driection = DMA_FROM_DEVICE;
 		else
-			pci_driection = PCI_DMA_TODEVICE;
+			pci_driection = DMA_TO_DEVICE;
 
 		if (buffer_info->flags & ATL1C_PCIMAP_SINGLE)
-			pci_unmap_single(pdev, buffer_info->dma,
-					buffer_info->length, pci_driection);
+			dma_unmap_single(&pdev->dev, buffer_info->dma,
+					 buffer_info->length, pci_driection);
 		else if (buffer_info->flags & ATL1C_PCIMAP_PAGE)
-			pci_unmap_page(pdev, buffer_info->dma,
-					buffer_info->length, pci_driection);
+			dma_unmap_page(&pdev->dev, buffer_info->dma,
+				       buffer_info->length, pci_driection);
 	}
 	if (buffer_info->skb)
-		dev_consume_skb_any(buffer_info->skb);
+		napi_consume_skb(buffer_info->skb, budget);
 	buffer_info->dma = 0;
 	buffer_info->skb = NULL;
 	ATL1C_SET_BUFFER_STATE(buffer_info, ATL1C_BUFFER_FREE);
@@ -861,11 +870,12 @@ static inline void atl1c_clean_buffer(struct pci_dev *pdev,
 /**
  * atl1c_clean_tx_ring - Free Tx-skb
  * @adapter: board private structure
+ * @queue: idx of transmit queue
  */
 static void atl1c_clean_tx_ring(struct atl1c_adapter *adapter,
-				enum atl1c_trans_queue type)
+				u32 queue)
 {
-	struct atl1c_tpd_ring *tpd_ring = &adapter->tpd_ring[type];
+	struct atl1c_tpd_ring *tpd_ring = &adapter->tpd_ring[queue];
 	struct atl1c_buffer *buffer_info;
 	struct pci_dev *pdev = adapter->pdev;
 	u16 index, ring_count;
@@ -873,10 +883,10 @@ static void atl1c_clean_tx_ring(struct atl1c_adapter *adapter,
 	ring_count = tpd_ring->count;
 	for (index = 0; index < ring_count; index++) {
 		buffer_info = &tpd_ring->buffer_info[index];
-		atl1c_clean_buffer(pdev, buffer_info);
+		atl1c_clean_buffer(pdev, buffer_info, 0);
 	}
 
-	netdev_reset_queue(adapter->netdev);
+	netdev_tx_reset_queue(netdev_get_tx_queue(adapter->netdev, queue));
 
 	/* Zero out Tx-buffers */
 	memset(tpd_ring->desc, 0, sizeof(struct atl1c_tpd_desc) *
@@ -888,18 +898,19 @@ static void atl1c_clean_tx_ring(struct atl1c_adapter *adapter,
 /**
  * atl1c_clean_rx_ring - Free rx-reservation skbs
  * @adapter: board private structure
+ * @queue: idx of transmit queue
  */
-static void atl1c_clean_rx_ring(struct atl1c_adapter *adapter)
+static void atl1c_clean_rx_ring(struct atl1c_adapter *adapter, u32 queue)
 {
-	struct atl1c_rfd_ring *rfd_ring = &adapter->rfd_ring;
-	struct atl1c_rrd_ring *rrd_ring = &adapter->rrd_ring;
+	struct atl1c_rfd_ring *rfd_ring = &adapter->rfd_ring[queue];
+	struct atl1c_rrd_ring *rrd_ring = &adapter->rrd_ring[queue];
 	struct atl1c_buffer *buffer_info;
 	struct pci_dev *pdev = adapter->pdev;
 	int j;
 
 	for (j = 0; j < rfd_ring->count; j++) {
 		buffer_info = &rfd_ring->buffer_info[j];
-		atl1c_clean_buffer(pdev, buffer_info);
+		atl1c_clean_buffer(pdev, buffer_info, 0);
 	}
 	/* zero out the descriptor ring */
 	memset(rfd_ring->desc, 0, rfd_ring->size);
@@ -915,26 +926,28 @@ static void atl1c_clean_rx_ring(struct atl1c_adapter *adapter)
 static void atl1c_init_ring_ptrs(struct atl1c_adapter *adapter)
 {
 	struct atl1c_tpd_ring *tpd_ring = adapter->tpd_ring;
-	struct atl1c_rfd_ring *rfd_ring = &adapter->rfd_ring;
-	struct atl1c_rrd_ring *rrd_ring = &adapter->rrd_ring;
+	struct atl1c_rfd_ring *rfd_ring = adapter->rfd_ring;
+	struct atl1c_rrd_ring *rrd_ring = adapter->rrd_ring;
 	struct atl1c_buffer *buffer_info;
 	int i, j;
 
-	for (i = 0; i < AT_MAX_TRANSMIT_QUEUE; i++) {
+	for (i = 0; i < adapter->tx_queue_count; i++) {
 		tpd_ring[i].next_to_use = 0;
 		atomic_set(&tpd_ring[i].next_to_clean, 0);
 		buffer_info = tpd_ring[i].buffer_info;
 		for (j = 0; j < tpd_ring->count; j++)
 			ATL1C_SET_BUFFER_STATE(&buffer_info[i],
-					ATL1C_BUFFER_FREE);
-	}
-	rfd_ring->next_to_use = 0;
-	rfd_ring->next_to_clean = 0;
-	rrd_ring->next_to_use = 0;
-	rrd_ring->next_to_clean = 0;
-	for (j = 0; j < rfd_ring->count; j++) {
-		buffer_info = &rfd_ring->buffer_info[j];
-		ATL1C_SET_BUFFER_STATE(buffer_info, ATL1C_BUFFER_FREE);
+					       ATL1C_BUFFER_FREE);
+	}
+	for (i = 0; i < adapter->rx_queue_count; i++) {
+		rfd_ring[i].next_to_use = 0;
+		rfd_ring[i].next_to_clean = 0;
+		rrd_ring[i].next_to_use = 0;
+		rrd_ring[i].next_to_clean = 0;
+		for (j = 0; j < rfd_ring[i].count; j++) {
+			buffer_info = &rfd_ring[i].buffer_info[j];
+			ATL1C_SET_BUFFER_STATE(buffer_info, ATL1C_BUFFER_FREE);
+		}
 	}
 }
 
@@ -948,25 +961,21 @@ static void atl1c_free_ring_resources(struct atl1c_adapter *adapter)
 {
 	struct pci_dev *pdev = adapter->pdev;
 
-	pci_free_consistent(pdev, adapter->ring_header.size,
-					adapter->ring_header.desc,
-					adapter->ring_header.dma);
+	dma_free_coherent(&pdev->dev, adapter->ring_header.size,
+			  adapter->ring_header.desc, adapter->ring_header.dma);
 	adapter->ring_header.desc = NULL;
 
 	/* Note: just free tdp_ring.buffer_info,
-	*  it contain rfd_ring.buffer_info, do not double free */
+	 * it contain rfd_ring.buffer_info, do not double free
+	 */
 	if (adapter->tpd_ring[0].buffer_info) {
 		kfree(adapter->tpd_ring[0].buffer_info);
 		adapter->tpd_ring[0].buffer_info = NULL;
 	}
-	if (adapter->rx_page) {
-		put_page(adapter->rx_page);
-		adapter->rx_page = NULL;
-	}
 }
 
 /**
- * atl1c_setup_mem_resources - allocate Tx / RX descriptor resources
+ * atl1c_setup_ring_resources - allocate Tx / RX descriptor resources
  * @adapter: board private structure
  *
  * Return 0 on success, negative on failure
@@ -975,51 +984,60 @@ static int atl1c_setup_ring_resources(struct atl1c_adapter *adapter)
 {
 	struct pci_dev *pdev = adapter->pdev;
 	struct atl1c_tpd_ring *tpd_ring = adapter->tpd_ring;
-	struct atl1c_rfd_ring *rfd_ring = &adapter->rfd_ring;
-	struct atl1c_rrd_ring *rrd_ring = &adapter->rrd_ring;
+	struct atl1c_rfd_ring *rfd_ring = adapter->rfd_ring;
+	struct atl1c_rrd_ring *rrd_ring = adapter->rrd_ring;
 	struct atl1c_ring_header *ring_header = &adapter->ring_header;
+	int tqc = adapter->tx_queue_count;
+	int rqc = adapter->rx_queue_count;
 	int size;
 	int i;
 	int count = 0;
-	int rx_desc_count = 0;
 	u32 offset = 0;
 
-	rrd_ring->count = rfd_ring->count;
-	for (i = 1; i < AT_MAX_TRANSMIT_QUEUE; i++)
+	/* Even though only one tpd queue is actually used, the "high"
+	 * priority tpd queue also gets initialized
+	 */
+	if (tqc == 1)
+		tqc = 2;
+
+	for (i = 1; i < tqc; i++)
 		tpd_ring[i].count = tpd_ring[0].count;
 
-	/* 2 tpd queue, one high priority queue,
-	 * another normal priority queue */
-	size = sizeof(struct atl1c_buffer) * (tpd_ring->count * 2 +
-		rfd_ring->count);
+	size = sizeof(struct atl1c_buffer) * (tpd_ring->count * tqc +
+					      rfd_ring->count * rqc);
 	tpd_ring->buffer_info = kzalloc(size, GFP_KERNEL);
 	if (unlikely(!tpd_ring->buffer_info))
 		goto err_nomem;
 
-	for (i = 0; i < AT_MAX_TRANSMIT_QUEUE; i++) {
-		tpd_ring[i].buffer_info =
-			(tpd_ring->buffer_info + count);
+	for (i = 0; i < tqc; i++) {
+		tpd_ring[i].adapter = adapter;
+		tpd_ring[i].num = i;
+		tpd_ring[i].buffer_info = (tpd_ring->buffer_info + count);
 		count += tpd_ring[i].count;
 	}
 
-	rfd_ring->buffer_info =
-		(tpd_ring->buffer_info + count);
-	count += rfd_ring->count;
-	rx_desc_count += rfd_ring->count;
+	for (i = 0; i < rqc; i++) {
+		rrd_ring[i].adapter = adapter;
+		rrd_ring[i].num = i;
+		rrd_ring[i].count = rfd_ring[0].count;
+		rfd_ring[i].count = rfd_ring[0].count;
+		rfd_ring[i].buffer_info = (tpd_ring->buffer_info + count);
+		count += rfd_ring->count;
+	}
 
 	/*
 	 * real ring DMA buffer
 	 * each ring/block may need up to 8 bytes for alignment, hence the
 	 * additional bytes tacked onto the end.
 	 */
-	ring_header->size = size =
-		sizeof(struct atl1c_tpd_desc) * tpd_ring->count * 2 +
-		sizeof(struct atl1c_rx_free_desc) * rx_desc_count +
-		sizeof(struct atl1c_recv_ret_status) * rx_desc_count +
+	ring_header->size =
+		sizeof(struct atl1c_tpd_desc) * tpd_ring->count * tqc +
+		sizeof(struct atl1c_rx_free_desc) * rfd_ring->count * rqc +
+		sizeof(struct atl1c_recv_ret_status) * rfd_ring->count * rqc +
 		8 * 4;
 
-	ring_header->desc = dma_zalloc_coherent(&pdev->dev, ring_header->size,
-						&ring_header->dma, GFP_KERNEL);
+	ring_header->desc = dma_alloc_coherent(&pdev->dev, ring_header->size,
+					       &ring_header->dma, GFP_KERNEL);
 	if (unlikely(!ring_header->desc)) {
 		dev_err(&pdev->dev, "could not get memory for DMA buffer\n");
 		goto err_nomem;
@@ -1028,25 +1046,28 @@ static int atl1c_setup_ring_resources(struct atl1c_adapter *adapter)
 
 	tpd_ring[0].dma = roundup(ring_header->dma, 8);
 	offset = tpd_ring[0].dma - ring_header->dma;
-	for (i = 0; i < AT_MAX_TRANSMIT_QUEUE; i++) {
+	for (i = 0; i < tqc; i++) {
 		tpd_ring[i].dma = ring_header->dma + offset;
-		tpd_ring[i].desc = (u8 *) ring_header->desc + offset;
+		tpd_ring[i].desc = (u8 *)ring_header->desc + offset;
 		tpd_ring[i].size =
 			sizeof(struct atl1c_tpd_desc) * tpd_ring[i].count;
 		offset += roundup(tpd_ring[i].size, 8);
 	}
-	/* init RFD ring */
-	rfd_ring->dma = ring_header->dma + offset;
-	rfd_ring->desc = (u8 *) ring_header->desc + offset;
-	rfd_ring->size = sizeof(struct atl1c_rx_free_desc) * rfd_ring->count;
-	offset += roundup(rfd_ring->size, 8);
+	for (i = 0; i < rqc; i++) {
+		/* init RFD ring */
+		rfd_ring[i].dma = ring_header->dma + offset;
+		rfd_ring[i].desc = (u8 *)ring_header->desc + offset;
+		rfd_ring[i].size = sizeof(struct atl1c_rx_free_desc) *
+			rfd_ring[i].count;
+		offset += roundup(rfd_ring[i].size, 8);
 
-	/* init RRD ring */
-	rrd_ring->dma = ring_header->dma + offset;
-	rrd_ring->desc = (u8 *) ring_header->desc + offset;
-	rrd_ring->size = sizeof(struct atl1c_recv_ret_status) *
-		rrd_ring->count;
-	offset += roundup(rrd_ring->size, 8);
+		/* init RRD ring */
+		rrd_ring[i].dma = ring_header->dma + offset;
+		rrd_ring[i].desc = (u8 *)ring_header->desc + offset;
+		rrd_ring[i].size = sizeof(struct atl1c_recv_ret_status) *
+			rrd_ring[i].count;
+		offset += roundup(rrd_ring[i].size, 8);
+	}
 
 	return 0;
 
@@ -1058,31 +1079,34 @@ err_nomem:
 static void atl1c_configure_des_ring(struct atl1c_adapter *adapter)
 {
 	struct atl1c_hw *hw = &adapter->hw;
-	struct atl1c_rfd_ring *rfd_ring = &adapter->rfd_ring;
-	struct atl1c_rrd_ring *rrd_ring = &adapter->rrd_ring;
-	struct atl1c_tpd_ring *tpd_ring = (struct atl1c_tpd_ring *)
-				adapter->tpd_ring;
+	struct atl1c_rfd_ring *rfd_ring = adapter->rfd_ring;
+	struct atl1c_rrd_ring *rrd_ring = adapter->rrd_ring;
+	struct atl1c_tpd_ring *tpd_ring = adapter->tpd_ring;
+	int i;
+	int tx_queue_count = adapter->tx_queue_count;
+
+	if (tx_queue_count == 1)
+		tx_queue_count = 2;
 
 	/* TPD */
 	AT_WRITE_REG(hw, REG_TX_BASE_ADDR_HI,
-			(u32)((tpd_ring[atl1c_trans_normal].dma &
-				AT_DMA_HI_ADDR_MASK) >> 32));
+		     (u32)((tpd_ring[0].dma & AT_DMA_HI_ADDR_MASK) >> 32));
 	/* just enable normal priority TX queue */
-	AT_WRITE_REG(hw, REG_TPD_PRI0_ADDR_LO,
-			(u32)(tpd_ring[atl1c_trans_normal].dma &
-				AT_DMA_LO_ADDR_MASK));
-	AT_WRITE_REG(hw, REG_TPD_PRI1_ADDR_LO,
-			(u32)(tpd_ring[atl1c_trans_high].dma &
-				AT_DMA_LO_ADDR_MASK));
+	for (i = 0; i < tx_queue_count; i++) {
+		AT_WRITE_REG(hw, atl1c_qregs[i].tpd_addr_lo,
+			     (u32)(tpd_ring[i].dma & AT_DMA_LO_ADDR_MASK));
+	}
 	AT_WRITE_REG(hw, REG_TPD_RING_SIZE,
 			(u32)(tpd_ring[0].count & TPD_RING_SIZE_MASK));
 
 
 	/* RFD */
 	AT_WRITE_REG(hw, REG_RX_BASE_ADDR_HI,
-			(u32)((rfd_ring->dma & AT_DMA_HI_ADDR_MASK) >> 32));
-	AT_WRITE_REG(hw, REG_RFD0_HEAD_ADDR_LO,
-			(u32)(rfd_ring->dma & AT_DMA_LO_ADDR_MASK));
+		     (u32)((rfd_ring->dma & AT_DMA_HI_ADDR_MASK) >> 32));
+	for (i = 0; i < adapter->rx_queue_count; i++) {
+		AT_WRITE_REG(hw, atl1c_qregs[i].rfd_addr_lo,
+			     (u32)(rfd_ring[i].dma & AT_DMA_LO_ADDR_MASK));
+	}
 
 	AT_WRITE_REG(hw, REG_RFD_RING_SIZE,
 			rfd_ring->count & RFD_RING_SIZE_MASK);
@@ -1090,8 +1114,10 @@ static void atl1c_configure_des_ring(struct atl1c_adapter *adapter)
 			adapter->rx_buffer_len & RX_BUF_SIZE_MASK);
 
 	/* RRD */
-	AT_WRITE_REG(hw, REG_RRD0_HEAD_ADDR_LO,
-			(u32)(rrd_ring->dma & AT_DMA_LO_ADDR_MASK));
+	for (i = 0; i < adapter->rx_queue_count; i++) {
+		AT_WRITE_REG(hw, atl1c_qregs[i].rrd_addr_lo,
+			     (u32)(rrd_ring[i].dma & AT_DMA_LO_ADDR_MASK));
+	}
 	AT_WRITE_REG(hw, REG_RRD_RING_SIZE,
 			(rrd_ring->count & RRD_RING_SIZE_MASK));
 
@@ -1201,7 +1227,7 @@ static void atl1c_start_mac(struct atl1c_adapter *adapter)
 	struct atl1c_hw *hw = &adapter->hw;
 	u32 mac, txq, rxq;
 
-	hw->mac_duplex = adapter->link_duplex == FULL_DUPLEX ? true : false;
+	hw->mac_duplex = adapter->link_duplex == FULL_DUPLEX;
 	hw->mac_speed = adapter->link_speed == SPEED_1000 ?
 		atl1c_mac_speed_1000 : atl1c_mac_speed_10_100;
 
@@ -1372,7 +1398,7 @@ static void atl1c_set_aspm(struct atl1c_hw *hw, u16 link_speed)
 }
 
 /**
- * atl1c_configure - Configure Transmit&Receive Unit after Reset
+ * atl1c_configure_mac - Configure Transmit&Receive Unit after Reset
  * @adapter: board private structure
  *
  * Configure the Tx /Rx unit of the MAC after a reset.
@@ -1444,14 +1470,28 @@ static int atl1c_configure(struct atl1c_adapter *adapter)
 {
 	struct net_device *netdev = adapter->netdev;
 	int num;
+	int i;
+
+	if (adapter->hw.nic_type == athr_mt) {
+		u32 mode;
+
+		AT_READ_REG(&adapter->hw, REG_MT_MODE, &mode);
+		if (adapter->rx_queue_count == 4)
+			mode |= MT_MODE_4Q;
+		else
+			mode &= ~MT_MODE_4Q;
+		AT_WRITE_REG(&adapter->hw, REG_MT_MODE, mode);
+	}
 
 	atl1c_init_ring_ptrs(adapter);
 	atl1c_set_multi(netdev);
 	atl1c_restore_vlan(adapter);
 
-	num = atl1c_alloc_rx_buffer(adapter);
-	if (unlikely(num == 0))
-		return -ENOMEM;
+	for (i = 0; i < adapter->rx_queue_count; ++i) {
+		num = atl1c_alloc_rx_buffer(adapter, i, false);
+		if (unlikely(num == 0))
+			return -ENOMEM;
+	}
 
 	if (atl1c_configure_mac(adapter))
 		return -EIO;
@@ -1545,20 +1585,22 @@ static inline void atl1c_clear_phy_int(struct atl1c_adapter *adapter)
 	spin_unlock(&adapter->mdio_lock);
 }
 
-static bool atl1c_clean_tx_irq(struct atl1c_adapter *adapter,
-				enum atl1c_trans_queue type)
+static int atl1c_clean_tx(struct napi_struct *napi, int budget)
 {
-	struct atl1c_tpd_ring *tpd_ring = &adapter->tpd_ring[type];
+	struct atl1c_tpd_ring *tpd_ring =
+		container_of(napi, struct atl1c_tpd_ring, napi);
+	struct atl1c_adapter *adapter = tpd_ring->adapter;
+	struct netdev_queue *txq =
+		netdev_get_tx_queue(napi->dev, tpd_ring->num);
 	struct atl1c_buffer *buffer_info;
 	struct pci_dev *pdev = adapter->pdev;
 	u16 next_to_clean = atomic_read(&tpd_ring->next_to_clean);
 	u16 hw_next_to_clean;
-	u16 reg;
 	unsigned int total_bytes = 0, total_packets = 0;
+	unsigned long flags;
 
-	reg = type == atl1c_trans_high ? REG_TPD_PRI1_CIDX : REG_TPD_PRI0_CIDX;
-
-	AT_READ_REGW(&adapter->hw, reg, &hw_next_to_clean);
+	AT_READ_REGW(&adapter->hw, atl1c_qregs[tpd_ring->num].tpd_cons,
+		     &hw_next_to_clean);
 
 	while (next_to_clean != hw_next_to_clean) {
 		buffer_info = &tpd_ring->buffer_info[next_to_clean];
@@ -1566,20 +1608,58 @@ static bool atl1c_clean_tx_irq(struct atl1c_adapter *adapter,
 			total_bytes += buffer_info->skb->len;
 			total_packets++;
 		}
-		atl1c_clean_buffer(pdev, buffer_info);
+		atl1c_clean_buffer(pdev, buffer_info, budget);
 		if (++next_to_clean == tpd_ring->count)
 			next_to_clean = 0;
 		atomic_set(&tpd_ring->next_to_clean, next_to_clean);
 	}
 
-	netdev_completed_queue(adapter->netdev, total_packets, total_bytes);
+	netdev_tx_completed_queue(txq, total_packets, total_bytes);
+
+	if (netif_tx_queue_stopped(txq) && netif_carrier_ok(adapter->netdev))
+		netif_tx_wake_queue(txq);
+
+	if (total_packets < budget) {
+		napi_complete_done(napi, total_packets);
+		spin_lock_irqsave(&adapter->hw.intr_mask_lock, flags);
+		adapter->hw.intr_mask |= atl1c_qregs[tpd_ring->num].tx_isr;
+		AT_WRITE_REG(&adapter->hw, REG_IMR, adapter->hw.intr_mask);
+		spin_unlock_irqrestore(&adapter->hw.intr_mask_lock, flags);
+		return total_packets;
+	}
+	return budget;
+}
+
+static void atl1c_intr_rx_tx(struct atl1c_adapter *adapter, u32 status)
+{
+	struct atl1c_hw *hw = &adapter->hw;
+	u32 intr_mask;
+	int i;
 
-	if (netif_queue_stopped(adapter->netdev) &&
-			netif_carrier_ok(adapter->netdev)) {
-		netif_wake_queue(adapter->netdev);
+	spin_lock(&hw->intr_mask_lock);
+	intr_mask = hw->intr_mask;
+	for (i = 0; i < adapter->rx_queue_count; ++i) {
+		if (!(status & atl1c_qregs[i].rx_isr))
+			continue;
+		if (napi_schedule_prep(&adapter->rrd_ring[i].napi)) {
+			intr_mask &= ~atl1c_qregs[i].rx_isr;
+			__napi_schedule(&adapter->rrd_ring[i].napi);
+		}
+	}
+	for (i = 0; i < adapter->tx_queue_count; ++i) {
+		if (!(status & atl1c_qregs[i].tx_isr))
+			continue;
+		if (napi_schedule_prep(&adapter->tpd_ring[i].napi)) {
+			intr_mask &= ~atl1c_qregs[i].tx_isr;
+			__napi_schedule(&adapter->tpd_ring[i].napi);
+		}
 	}
 
-	return true;
+	if (hw->intr_mask != intr_mask) {
+		hw->intr_mask = intr_mask;
+		AT_WRITE_REG(hw, REG_IMR, hw->intr_mask);
+	}
+	spin_unlock(&hw->intr_mask_lock);
 }
 
 /**
@@ -1612,15 +1692,8 @@ static irqreturn_t atl1c_intr(int irq, void *data)
 			atl1c_clear_phy_int(adapter);
 		/* Ack ISR */
 		AT_WRITE_REG(hw, REG_ISR, status | ISR_DIS_INT);
-		if (status & ISR_RX_PKT) {
-			if (likely(napi_schedule_prep(&adapter->napi))) {
-				hw->intr_mask &= ~ISR_RX_PKT;
-				AT_WRITE_REG(hw, REG_IMR, hw->intr_mask);
-				__napi_schedule(&adapter->napi);
-			}
-		}
-		if (status & ISR_TX_PKT)
-			atl1c_clean_tx_irq(adapter, atl1c_trans_normal);
+		if (status & (ISR_RX_PKT | ISR_TX_PKT))
+			atl1c_intr_rx_tx(adapter, status);
 
 		handled = IRQ_HANDLED;
 		/* check if PCIE PHY Link down */
@@ -1657,6 +1730,11 @@ static irqreturn_t atl1c_intr(int irq, void *data)
 static inline void atl1c_rx_checksum(struct atl1c_adapter *adapter,
 		  struct sk_buff *skb, struct atl1c_recv_ret_status *prrs)
 {
+	if (adapter->hw.nic_type == athr_mt) {
+		if (prrs->word3 & RRS_MT_PROT_ID_TCPUDP)
+			skb->ip_summed = CHECKSUM_UNNECESSARY;
+		return;
+	}
 	/*
 	 * The pid field in RRS in not correct sometimes, so we
 	 * cannot figure out if the packet is fragmented or not,
@@ -1665,38 +1743,11 @@ static inline void atl1c_rx_checksum(struct atl1c_adapter *adapter,
 	skb_checksum_none_assert(skb);
 }
 
-static struct sk_buff *atl1c_alloc_skb(struct atl1c_adapter *adapter)
+static int atl1c_alloc_rx_buffer(struct atl1c_adapter *adapter, u32 queue,
+				 bool napi_mode)
 {
-	struct sk_buff *skb;
-	struct page *page;
-
-	if (adapter->rx_frag_size > PAGE_SIZE)
-		return netdev_alloc_skb(adapter->netdev,
-					adapter->rx_buffer_len);
-
-	page = adapter->rx_page;
-	if (!page) {
-		adapter->rx_page = page = alloc_page(GFP_ATOMIC);
-		if (unlikely(!page))
-			return NULL;
-		adapter->rx_page_offset = 0;
-	}
-
-	skb = build_skb(page_address(page) + adapter->rx_page_offset,
-			adapter->rx_frag_size);
-	if (likely(skb)) {
-		adapter->rx_page_offset += adapter->rx_frag_size;
-		if (adapter->rx_page_offset >= PAGE_SIZE)
-			adapter->rx_page = NULL;
-		else
-			get_page(page);
-	}
-	return skb;
-}
-
-static int atl1c_alloc_rx_buffer(struct atl1c_adapter *adapter)
-{
-	struct atl1c_rfd_ring *rfd_ring = &adapter->rfd_ring;
+	struct atl1c_rfd_ring *rfd_ring = &adapter->rfd_ring[queue];
+	struct atl1c_rrd_ring *rrd_ring = &adapter->rrd_ring[queue];
 	struct pci_dev *pdev = adapter->pdev;
 	struct atl1c_buffer *buffer_info, *next_info;
 	struct sk_buff *skb;
@@ -1715,13 +1766,27 @@ static int atl1c_alloc_rx_buffer(struct atl1c_adapter *adapter)
 	while (next_info->flags & ATL1C_BUFFER_FREE) {
 		rfd_desc = ATL1C_RFD_DESC(rfd_ring, rfd_next_to_use);
 
-		skb = atl1c_alloc_skb(adapter);
+		/* When DMA RX address is set to something like
+		 * 0x....fc0, it will be very likely to cause DMA
+		 * RFD overflow issue.
+		 *
+		 * To work around it, we apply rx skb with 64 bytes
+		 * longer space, and offset the address whenever
+		 * 0x....fc0 is detected.
+		 */
+		if (likely(napi_mode))
+			skb = napi_alloc_skb(&rrd_ring->napi, adapter->rx_buffer_len + 64);
+		else
+			skb = netdev_alloc_skb(adapter->netdev, adapter->rx_buffer_len + 64);
 		if (unlikely(!skb)) {
 			if (netif_msg_rx_err(adapter))
 				dev_warn(&pdev->dev, "alloc rx buffer failed\n");
 			break;
 		}
 
+		if (((unsigned long)skb->data & 0xfff) == 0xfc0)
+			skb_reserve(skb, 64);
+
 		/*
 		 * Make buffer alignment 2 beyond a 16 byte boundary
 		 * this will result in a 16 byte aligned IP header after
@@ -1731,15 +1796,14 @@ static int atl1c_alloc_rx_buffer(struct atl1c_adapter *adapter)
 		ATL1C_SET_BUFFER_STATE(buffer_info, ATL1C_BUFFER_BUSY);
 		buffer_info->skb = skb;
 		buffer_info->length = adapter->rx_buffer_len;
-		mapping = pci_map_single(pdev, vir_addr,
-						buffer_info->length,
-						PCI_DMA_FROMDEVICE);
-		if (unlikely(pci_dma_mapping_error(pdev, mapping))) {
+		mapping = dma_map_single(&pdev->dev, vir_addr,
+					 buffer_info->length, DMA_FROM_DEVICE);
+		if (unlikely(dma_mapping_error(&pdev->dev, mapping))) {
 			dev_kfree_skb(skb);
 			buffer_info->skb = NULL;
 			buffer_info->length = 0;
 			ATL1C_SET_BUFFER_STATE(buffer_info, ATL1C_BUFFER_FREE);
-			netif_warn(adapter, rx_err, adapter->netdev, "RX pci_map_single failed");
+			netif_warn(adapter, rx_err, adapter->netdev, "RX dma_map_single failed");
 			break;
 		}
 		buffer_info->dma = mapping;
@@ -1758,8 +1822,8 @@ static int atl1c_alloc_rx_buffer(struct atl1c_adapter *adapter)
 		/* TODO: update mailbox here */
 		wmb();
 		rfd_ring->next_to_use = rfd_next_to_use;
-		AT_WRITE_REG(&adapter->hw, REG_MB_RFD0_PROD_IDX,
-			rfd_ring->next_to_use & MB_RFDX_PROD_IDX_MASK);
+		AT_WRITE_REG(&adapter->hw, atl1c_qregs[queue].rfd_prod,
+			     rfd_ring->next_to_use & MB_RFDX_PROD_IDX_MASK);
 	}
 
 	return num_alloc;
@@ -1797,22 +1861,33 @@ static void atl1c_clean_rfd(struct atl1c_rfd_ring *rfd_ring,
 	rfd_ring->next_to_clean = rfd_index;
 }
 
-static void atl1c_clean_rx_irq(struct atl1c_adapter *adapter,
-		   int *work_done, int work_to_do)
+/**
+ * atl1c_clean_rx - NAPI Rx polling callback
+ * @napi: napi info
+ * @budget: limit of packets to clean
+ */
+static int atl1c_clean_rx(struct napi_struct *napi, int budget)
 {
+	struct atl1c_rrd_ring *rrd_ring =
+		container_of(napi, struct atl1c_rrd_ring, napi);
+	struct atl1c_adapter *adapter = rrd_ring->adapter;
 	u16 rfd_num, rfd_index;
-	u16 count = 0;
 	u16 length;
 	struct pci_dev *pdev = adapter->pdev;
 	struct net_device *netdev  = adapter->netdev;
-	struct atl1c_rfd_ring *rfd_ring = &adapter->rfd_ring;
-	struct atl1c_rrd_ring *rrd_ring = &adapter->rrd_ring;
+	struct atl1c_rfd_ring *rfd_ring = &adapter->rfd_ring[rrd_ring->num];
 	struct sk_buff *skb;
 	struct atl1c_recv_ret_status *rrs;
 	struct atl1c_buffer *buffer_info;
+	int work_done = 0;
+	unsigned long flags;
+
+	/* Keep link state information with original netdev */
+	if (!netif_carrier_ok(adapter->netdev))
+		goto quit_polling;
 
 	while (1) {
-		if (*work_done >= work_to_do)
+		if (work_done >= budget)
 			break;
 		rrs = ATL1C_RRD_DESC(rrd_ring, rrd_ring->next_to_clean);
 		if (likely(RRS_RXD_IS_VALID(rrs->word3))) {
@@ -1831,10 +1906,10 @@ rrs_checked:
 		atl1c_clean_rrd(rrd_ring, rrs, rfd_num);
 		if (rrs->word3 & (RRS_RX_ERR_SUM | RRS_802_3_LEN_ERR)) {
 			atl1c_clean_rfd(rfd_ring, rrs, rfd_num);
-				if (netif_msg_rx_err(adapter))
-					dev_warn(&pdev->dev,
-						"wrong packet! rrs word3 is %x\n",
-						rrs->word3);
+			if (netif_msg_rx_err(adapter))
+				dev_warn(&pdev->dev,
+					 "wrong packet! rrs word3 is %x\n",
+					 rrs->word3);
 			continue;
 		}
 
@@ -1845,8 +1920,8 @@ rrs_checked:
 			rfd_index = (rrs->word0 >> RRS_RX_RFD_INDEX_SHIFT) &
 					RRS_RX_RFD_INDEX_MASK;
 			buffer_info = &rfd_ring->buffer_info[rfd_index];
-			pci_unmap_single(pdev, buffer_info->dma,
-				buffer_info->length, PCI_DMA_FROMDEVICE);
+			dma_unmap_single(&pdev->dev, buffer_info->dma,
+					 buffer_info->length, DMA_FROM_DEVICE);
 			skb = buffer_info->skb;
 		} else {
 			/* TODO */
@@ -1866,35 +1941,20 @@ rrs_checked:
 			vlan = le16_to_cpu(vlan);
 			__vlan_hwaccel_put_tag(skb, htons(ETH_P_8021Q), vlan);
 		}
-		netif_receive_skb(skb);
+		napi_gro_receive(napi, skb);
 
-		(*work_done)++;
-		count++;
+		work_done++;
 	}
-	if (count)
-		atl1c_alloc_rx_buffer(adapter);
-}
-
-/**
- * atl1c_clean - NAPI Rx polling callback
- */
-static int atl1c_clean(struct napi_struct *napi, int budget)
-{
-	struct atl1c_adapter *adapter =
-			container_of(napi, struct atl1c_adapter, napi);
-	int work_done = 0;
-
-	/* Keep link state information with original netdev */
-	if (!netif_carrier_ok(adapter->netdev))
-		goto quit_polling;
-	/* just enable one RXQ */
-	atl1c_clean_rx_irq(adapter, &work_done, budget);
+	if (work_done)
+		atl1c_alloc_rx_buffer(adapter, rrd_ring->num, true);
 
 	if (work_done < budget) {
 quit_polling:
 		napi_complete_done(napi, work_done);
-		adapter->hw.intr_mask |= ISR_RX_PKT;
+		spin_lock_irqsave(&adapter->hw.intr_mask_lock, flags);
+		adapter->hw.intr_mask |= atl1c_qregs[rrd_ring->num].rx_isr;
 		AT_WRITE_REG(&adapter->hw, REG_IMR, adapter->hw.intr_mask);
+		spin_unlock_irqrestore(&adapter->hw.intr_mask_lock, flags);
 	}
 	return work_done;
 }
@@ -1916,9 +1976,9 @@ static void atl1c_netpoll(struct net_device *netdev)
 }
 #endif
 
-static inline u16 atl1c_tpd_avail(struct atl1c_adapter *adapter, enum atl1c_trans_queue type)
+static inline u16 atl1c_tpd_avail(struct atl1c_adapter *adapter, u32 queue)
 {
-	struct atl1c_tpd_ring *tpd_ring = &adapter->tpd_ring[type];
+	struct atl1c_tpd_ring *tpd_ring = &adapter->tpd_ring[queue];
 	u16 next_to_use = 0;
 	u16 next_to_clean = 0;
 
@@ -1936,9 +1996,9 @@ static inline u16 atl1c_tpd_avail(struct atl1c_adapter *adapter, enum atl1c_tran
  * there is enough tpd to use
  */
 static struct atl1c_tpd_desc *atl1c_get_tpd(struct atl1c_adapter *adapter,
-	enum atl1c_trans_queue type)
+					    u32 queue)
 {
-	struct atl1c_tpd_ring *tpd_ring = &adapter->tpd_ring[type];
+	struct atl1c_tpd_ring *tpd_ring = &adapter->tpd_ring[queue];
 	struct atl1c_tpd_desc *tpd_desc;
 	u16 next_to_use = 0;
 
@@ -1968,7 +2028,7 @@ static u16 atl1c_cal_tpd_req(const struct sk_buff *skb)
 	tpd_req = skb_shinfo(skb)->nr_frags + 1;
 
 	if (skb_is_gso(skb)) {
-		proto_hdr_len = skb_transport_offset(skb) + tcp_hdrlen(skb);
+		proto_hdr_len = skb_tcp_all_headers(skb);
 		if (proto_hdr_len < skb_headlen(skb))
 			tpd_req++;
 		if (skb_shinfo(skb)->gso_type & SKB_GSO_TCPV6)
@@ -1980,7 +2040,7 @@ static u16 atl1c_cal_tpd_req(const struct sk_buff *skb)
 static int atl1c_tso_csum(struct atl1c_adapter *adapter,
 			  struct sk_buff *skb,
 			  struct atl1c_tpd_desc **tpd,
-			  enum atl1c_trans_queue type)
+			  u32 queue)
 {
 	struct pci_dev *pdev = adapter->pdev;
 	unsigned short offload_type;
@@ -2000,10 +2060,13 @@ static int atl1c_tso_csum(struct atl1c_adapter *adapter,
 			real_len = (((unsigned char *)ip_hdr(skb) - skb->data)
 					+ ntohs(ip_hdr(skb)->tot_len));
 
-			if (real_len < skb->len)
-				pskb_trim(skb, real_len);
+			if (real_len < skb->len) {
+				err = pskb_trim(skb, real_len);
+				if (err)
+					return err;
+			}
 
-			hdr_len = (skb_transport_offset(skb) + tcp_hdrlen(skb));
+			hdr_len = skb_tcp_all_headers(skb);
 			if (unlikely(skb->len == hdr_len)) {
 				/* only xsum need */
 				if (netif_msg_tx_queued(adapter))
@@ -2025,10 +2088,10 @@ static int atl1c_tso_csum(struct atl1c_adapter *adapter,
 				*(struct atl1c_tpd_ext_desc **)(tpd);
 
 			memset(etpd, 0, sizeof(struct atl1c_tpd_ext_desc));
-			*tpd = atl1c_get_tpd(adapter, type);
+			*tpd = atl1c_get_tpd(adapter, queue);
 			ipv6_hdr(skb)->payload_len = 0;
 			/* check payload == 0 byte ? */
-			hdr_len = (skb_transport_offset(skb) + tcp_hdrlen(skb));
+			hdr_len = skb_tcp_all_headers(skb);
 			if (unlikely(skb->len == hdr_len)) {
 				/* only xsum need */
 				if (netif_msg_tx_queued(adapter))
@@ -2036,10 +2099,8 @@ static int atl1c_tso_csum(struct atl1c_adapter *adapter,
 						"IPV6 tso with zero data??\n");
 				goto check_sum;
 			} else
-				tcp_hdr(skb)->check = ~csum_ipv6_magic(
-						&ipv6_hdr(skb)->saddr,
-						&ipv6_hdr(skb)->daddr,
-						0, IPPROTO_TCP, 0);
+				tcp_v6_gso_csum_prep(skb);
+
 			etpd->word1 |= 1 << TPD_LSO_EN_SHIFT;
 			etpd->word1 |= 1 << TPD_LSO_VER_SHIFT;
 			etpd->pkt_len = cpu_to_le32(skb->len);
@@ -2079,9 +2140,9 @@ check_sum:
 
 static void atl1c_tx_rollback(struct atl1c_adapter *adpt,
 			      struct atl1c_tpd_desc *first_tpd,
-			      enum atl1c_trans_queue type)
+			      u32 queue)
 {
-	struct atl1c_tpd_ring *tpd_ring = &adpt->tpd_ring[type];
+	struct atl1c_tpd_ring *tpd_ring = &adpt->tpd_ring[queue];
 	struct atl1c_buffer *buffer_info;
 	struct atl1c_tpd_desc *tpd;
 	u16 first_index, index;
@@ -2091,7 +2152,7 @@ static void atl1c_tx_rollback(struct atl1c_adapter *adpt,
 	while (index != tpd_ring->next_to_use) {
 		tpd = ATL1C_TPD_DESC(tpd_ring, index);
 		buffer_info = &tpd_ring->buffer_info[index];
-		atl1c_clean_buffer(adpt->pdev, buffer_info);
+		atl1c_clean_buffer(adpt->pdev, buffer_info, 0);
 		memset(tpd, 0, sizeof(struct atl1c_tpd_desc));
 		if (++index == tpd_ring->count)
 			index = 0;
@@ -2100,8 +2161,8 @@ static void atl1c_tx_rollback(struct atl1c_adapter *adpt,
 }
 
 static int atl1c_tx_map(struct atl1c_adapter *adapter,
-		      struct sk_buff *skb, struct atl1c_tpd_desc *tpd,
-			enum atl1c_trans_queue type)
+			struct sk_buff *skb, struct atl1c_tpd_desc *tpd,
+			u32 queue)
 {
 	struct atl1c_tpd_desc *use_tpd = NULL;
 	struct atl1c_buffer *buffer_info = NULL;
@@ -2117,15 +2178,16 @@ static int atl1c_tx_map(struct atl1c_adapter *adapter,
 	tso = (tpd->word1 >> TPD_LSO_EN_SHIFT) & TPD_LSO_EN_MASK;
 	if (tso) {
 		/* TSO */
-		map_len = hdr_len = skb_transport_offset(skb) + tcp_hdrlen(skb);
+		hdr_len = skb_tcp_all_headers(skb);
+		map_len = hdr_len;
 		use_tpd = tpd;
 
 		buffer_info = atl1c_get_tx_buffer(adapter, use_tpd);
 		buffer_info->length = map_len;
-		buffer_info->dma = pci_map_single(adapter->pdev,
-					skb->data, hdr_len, PCI_DMA_TODEVICE);
-		if (unlikely(pci_dma_mapping_error(adapter->pdev,
-						   buffer_info->dma)))
+		buffer_info->dma = dma_map_single(&adapter->pdev->dev,
+						  skb->data, hdr_len,
+						  DMA_TO_DEVICE);
+		if (unlikely(dma_mapping_error(&adapter->pdev->dev, buffer_info->dma)))
 			goto err_dma;
 		ATL1C_SET_BUFFER_STATE(buffer_info, ATL1C_BUFFER_BUSY);
 		ATL1C_SET_PCIMAP_TYPE(buffer_info, ATL1C_PCIMAP_SINGLE,
@@ -2141,16 +2203,16 @@ static int atl1c_tx_map(struct atl1c_adapter *adapter,
 		if (mapped_len == 0)
 			use_tpd = tpd;
 		else {
-			use_tpd = atl1c_get_tpd(adapter, type);
+			use_tpd = atl1c_get_tpd(adapter, queue);
 			memcpy(use_tpd, tpd, sizeof(struct atl1c_tpd_desc));
 		}
 		buffer_info = atl1c_get_tx_buffer(adapter, use_tpd);
 		buffer_info->length = buf_len - mapped_len;
 		buffer_info->dma =
-			pci_map_single(adapter->pdev, skb->data + mapped_len,
-					buffer_info->length, PCI_DMA_TODEVICE);
-		if (unlikely(pci_dma_mapping_error(adapter->pdev,
-						   buffer_info->dma)))
+			dma_map_single(&adapter->pdev->dev,
+				       skb->data + mapped_len,
+				       buffer_info->length, DMA_TO_DEVICE);
+		if (unlikely(dma_mapping_error(&adapter->pdev->dev, buffer_info->dma)))
 			goto err_dma;
 
 		ATL1C_SET_BUFFER_STATE(buffer_info, ATL1C_BUFFER_BUSY);
@@ -2161,11 +2223,9 @@ static int atl1c_tx_map(struct atl1c_adapter *adapter,
 	}
 
 	for (f = 0; f < nr_frags; f++) {
-		struct skb_frag_struct *frag;
-
-		frag = &skb_shinfo(skb)->frags[f];
+		skb_frag_t *frag = &skb_shinfo(skb)->frags[f];
 
-		use_tpd = atl1c_get_tpd(adapter, type);
+		use_tpd = atl1c_get_tpd(adapter, queue);
 		memcpy(use_tpd, tpd, sizeof(struct atl1c_tpd_desc));
 
 		buffer_info = atl1c_get_tx_buffer(adapter, use_tpd);
@@ -2198,23 +2258,22 @@ err_dma:
 	return -1;
 }
 
-static void atl1c_tx_queue(struct atl1c_adapter *adapter, struct sk_buff *skb,
-			   struct atl1c_tpd_desc *tpd, enum atl1c_trans_queue type)
+static void atl1c_tx_queue(struct atl1c_adapter *adapter, u32 queue)
 {
-	struct atl1c_tpd_ring *tpd_ring = &adapter->tpd_ring[type];
-	u16 reg;
+	struct atl1c_tpd_ring *tpd_ring = &adapter->tpd_ring[queue];
 
-	reg = type == atl1c_trans_high ? REG_TPD_PRI1_PIDX : REG_TPD_PRI0_PIDX;
-	AT_WRITE_REGW(&adapter->hw, reg, tpd_ring->next_to_use);
+	AT_WRITE_REGW(&adapter->hw, atl1c_qregs[queue].tpd_prod,
+		      tpd_ring->next_to_use);
 }
 
 static netdev_tx_t atl1c_xmit_frame(struct sk_buff *skb,
 					  struct net_device *netdev)
 {
 	struct atl1c_adapter *adapter = netdev_priv(netdev);
-	u16 tpd_req = 1;
+	u32 queue = skb_get_queue_mapping(skb);
+	struct netdev_queue *txq = netdev_get_tx_queue(netdev, queue);
 	struct atl1c_tpd_desc *tpd;
-	enum atl1c_trans_queue type = atl1c_trans_normal;
+	u16 tpd_req;
 
 	if (test_bit(__AT_DOWN, &adapter->flags)) {
 		dev_kfree_skb_any(skb);
@@ -2223,16 +2282,18 @@ static netdev_tx_t atl1c_xmit_frame(struct sk_buff *skb,
 
 	tpd_req = atl1c_cal_tpd_req(skb);
 
-	if (atl1c_tpd_avail(adapter, type) < tpd_req) {
+	if (atl1c_tpd_avail(adapter, queue) < tpd_req) {
 		/* no enough descriptor, just stop queue */
-		netif_stop_queue(netdev);
+		atl1c_tx_queue(adapter, queue);
+		netif_tx_stop_queue(txq);
 		return NETDEV_TX_BUSY;
 	}
 
-	tpd = atl1c_get_tpd(adapter, type);
+	tpd = atl1c_get_tpd(adapter, queue);
 
 	/* do TSO and check sum */
-	if (atl1c_tso_csum(adapter, skb, &tpd, type) != 0) {
+	if (atl1c_tso_csum(adapter, skb, &tpd, queue) != 0) {
+		atl1c_tx_queue(adapter, queue);
 		dev_kfree_skb_any(skb);
 		return NETDEV_TX_OK;
 	}
@@ -2250,15 +2311,17 @@ static netdev_tx_t atl1c_xmit_frame(struct sk_buff *skb,
 	if (skb_network_offset(skb) != ETH_HLEN)
 		tpd->word1 |= 1 << TPD_ETH_TYPE_SHIFT; /* Ethernet frame */
 
-	if (atl1c_tx_map(adapter, skb, tpd, type) < 0) {
+	if (atl1c_tx_map(adapter, skb, tpd, queue) < 0) {
 		netif_info(adapter, tx_done, adapter->netdev,
-			   "tx-skb droppted due to dma error\n");
+			   "tx-skb dropped due to dma error\n");
 		/* roll back tpd/buffer */
-		atl1c_tx_rollback(adapter, tpd, type);
+		atl1c_tx_rollback(adapter, tpd, queue);
 		dev_kfree_skb_any(skb);
 	} else {
-		netdev_sent_queue(adapter->netdev, skb->len);
-		atl1c_tx_queue(adapter, skb, tpd, type);
+		bool more = netdev_xmit_more();
+
+		if (__netdev_tx_sent_queue(txq, skb->len, more))
+			atl1c_tx_queue(adapter, queue);
 	}
 
 	return NETDEV_TX_OK;
@@ -2312,16 +2375,19 @@ static int atl1c_request_irq(struct atl1c_adapter *adapter)
 
 static void atl1c_reset_dma_ring(struct atl1c_adapter *adapter)
 {
+	int i;
 	/* release tx-pending skbs and reset tx/rx ring index */
-	atl1c_clean_tx_ring(adapter, atl1c_trans_normal);
-	atl1c_clean_tx_ring(adapter, atl1c_trans_high);
-	atl1c_clean_rx_ring(adapter);
+	for (i = 0; i < adapter->tx_queue_count; ++i)
+		atl1c_clean_tx_ring(adapter, i);
+	for (i = 0; i < adapter->rx_queue_count; ++i)
+		atl1c_clean_rx_ring(adapter, i);
 }
 
 static int atl1c_up(struct atl1c_adapter *adapter)
 {
 	struct net_device *netdev = adapter->netdev;
 	int err;
+	int i;
 
 	netif_carrier_off(netdev);
 
@@ -2335,19 +2401,24 @@ static int atl1c_up(struct atl1c_adapter *adapter)
 
 	atl1c_check_link_status(adapter);
 	clear_bit(__AT_DOWN, &adapter->flags);
-	napi_enable(&adapter->napi);
+	for (i = 0; i < adapter->tx_queue_count; ++i)
+		napi_enable(&adapter->tpd_ring[i].napi);
+	for (i = 0; i < adapter->rx_queue_count; ++i)
+		napi_enable(&adapter->rrd_ring[i].napi);
 	atl1c_irq_enable(adapter);
 	netif_start_queue(netdev);
 	return err;
 
 err_up:
-	atl1c_clean_rx_ring(adapter);
+	for (i = 0; i < adapter->rx_queue_count; ++i)
+		atl1c_clean_rx_ring(adapter, i);
 	return err;
 }
 
 static void atl1c_down(struct atl1c_adapter *adapter)
 {
 	struct net_device *netdev = adapter->netdev;
+	int i;
 
 	atl1c_del_timer(adapter);
 	adapter->work_event = 0; /* clear all event */
@@ -2355,7 +2426,10 @@ static void atl1c_down(struct atl1c_adapter *adapter)
 	 * reschedule our watchdog timer */
 	set_bit(__AT_DOWN, &adapter->flags);
 	netif_carrier_off(netdev);
-	napi_disable(&adapter->napi);
+	for (i = 0; i < adapter->tx_queue_count; ++i)
+		napi_disable(&adapter->tpd_ring[i].napi);
+	for (i = 0; i < adapter->rx_queue_count; ++i)
+		napi_disable(&adapter->rrd_ring[i].napi);
 	atl1c_irq_disable(adapter);
 	atl1c_free_irq(adapter);
 	/* disable ASPM if device inactive */
@@ -2433,8 +2507,7 @@ static int atl1c_close(struct net_device *netdev)
 
 static int atl1c_suspend(struct device *dev)
 {
-	struct pci_dev *pdev = to_pci_dev(dev);
-	struct net_device *netdev = pci_get_drvdata(pdev);
+	struct net_device *netdev = dev_get_drvdata(dev);
 	struct atl1c_adapter *adapter = netdev_priv(netdev);
 	struct atl1c_hw *hw = &adapter->hw;
 	u32 wufc = adapter->wol;
@@ -2448,7 +2521,7 @@ static int atl1c_suspend(struct device *dev)
 
 	if (wufc)
 		if (atl1c_phy_to_ps_link(hw) != 0)
-			dev_dbg(&pdev->dev, "phy power saving failed");
+			dev_dbg(dev, "phy power saving failed");
 
 	atl1c_power_saving(hw, wufc);
 
@@ -2458,8 +2531,7 @@ static int atl1c_suspend(struct device *dev)
 #ifdef CONFIG_PM_SLEEP
 static int atl1c_resume(struct device *dev)
 {
-	struct pci_dev *pdev = to_pci_dev(dev);
-	struct net_device *netdev = pci_get_drvdata(pdev);
+	struct net_device *netdev = dev_get_drvdata(dev);
 	struct atl1c_adapter *adapter = netdev_priv(netdev);
 
 	AT_WRITE_REG(&adapter->hw, REG_WOL_CTRL, 0);
@@ -2469,12 +2541,6 @@ static int atl1c_resume(struct device *dev)
 	atl1c_reset_mac(&adapter->hw);
 	atl1c_phy_init(&adapter->hw);
 
-#if 0
-	AT_READ_REG(&adapter->hw, REG_PM_CTRLSTAT, &pm_data);
-	pm_data &= ~PM_CTRLSTAT_PME_EN;
-	AT_WRITE_REG(&adapter->hw, REG_PM_CTRLSTAT, pm_data);
-#endif
-
 	netif_device_attach(netdev);
 	if (netif_running(netdev))
 		atl1c_up(adapter);
@@ -2503,7 +2569,7 @@ static const struct net_device_ops atl1c_netdev_ops = {
 	.ndo_change_mtu		= atl1c_change_mtu,
 	.ndo_fix_features	= atl1c_fix_features,
 	.ndo_set_features	= atl1c_set_features,
-	.ndo_do_ioctl		= atl1c_ioctl,
+	.ndo_eth_ioctl		= atl1c_ioctl,
 	.ndo_tx_timeout		= atl1c_tx_timeout,
 	.ndo_get_stats		= atl1c_get_stats,
 #ifdef CONFIG_NET_POLL_CONTROLLER
@@ -2548,15 +2614,16 @@ static int atl1c_probe(struct pci_dev *pdev, const struct pci_device_id *ent)
 	struct net_device *netdev;
 	struct atl1c_adapter *adapter;
 	static int cards_found;
-
+	u8 __iomem *hw_addr;
+	enum atl1c_nic_type nic_type;
+	u32 queue_count = 1;
 	int err = 0;
+	int i;
 
 	/* enable device (incl. PCI PM wakeup and hotplug setup) */
 	err = pci_enable_device_mem(pdev);
-	if (err) {
-		dev_err(&pdev->dev, "cannot enable PCI device\n");
-		return err;
-	}
+	if (err)
+		return dev_err_probe(&pdev->dev, err, "cannot enable PCI device\n");
 
 	/*
 	 * The atl1c chip can DMA to 64-bit addresses, but it uses a single
@@ -2568,8 +2635,8 @@ static int atl1c_probe(struct pci_dev *pdev, const struct pci_device_id *ent)
 	 * various kernel subsystems to support the mechanics required by a
 	 * fixed-high-32-bit system.
 	 */
-	if ((pci_set_dma_mask(pdev, DMA_BIT_MASK(32)) != 0) ||
-	    (pci_set_consistent_dma_mask(pdev, DMA_BIT_MASK(32)) != 0)) {
+	err = dma_set_mask_and_coherent(&pdev->dev, DMA_BIT_MASK(32));
+	if (err) {
 		dev_err(&pdev->dev, "No usable DMA configuration,aborting\n");
 		goto err_dma;
 	}
@@ -2582,7 +2649,18 @@ static int atl1c_probe(struct pci_dev *pdev, const struct pci_device_id *ent)
 
 	pci_set_master(pdev);
 
-	netdev = alloc_etherdev(sizeof(struct atl1c_adapter));
+	hw_addr = pci_ioremap_bar(pdev, 0);
+	if (!hw_addr) {
+		err = -EIO;
+		dev_err(&pdev->dev, "cannot map device registers\n");
+		goto err_ioremap;
+	}
+
+	nic_type = atl1c_get_mac_type(pdev, hw_addr);
+	if (nic_type == athr_mt)
+		queue_count = 4;
+
+	netdev = alloc_etherdev_mq(sizeof(struct atl1c_adapter), queue_count);
 	if (netdev == NULL) {
 		err = -ENOMEM;
 		goto err_alloc_etherdev;
@@ -2598,13 +2676,11 @@ static int atl1c_probe(struct pci_dev *pdev, const struct pci_device_id *ent)
 	adapter->netdev = netdev;
 	adapter->pdev = pdev;
 	adapter->hw.adapter = adapter;
+	adapter->hw.nic_type = nic_type;
 	adapter->msg_enable = netif_msg_init(-1, atl1c_default_msg);
-	adapter->hw.hw_addr = ioremap(pci_resource_start(pdev, 0), pci_resource_len(pdev, 0));
-	if (!adapter->hw.hw_addr) {
-		err = -EIO;
-		dev_err(&pdev->dev, "cannot map device registers\n");
-		goto err_ioremap;
-	}
+	adapter->hw.hw_addr = hw_addr;
+	adapter->tx_queue_count = queue_count;
+	adapter->rx_queue_count = queue_count;
 
 	/* init mii data */
 	adapter->mii.dev = netdev;
@@ -2612,9 +2688,14 @@ static int atl1c_probe(struct pci_dev *pdev, const struct pci_device_id *ent)
 	adapter->mii.mdio_write = atl1c_mdio_write;
 	adapter->mii.phy_id_mask = 0x1f;
 	adapter->mii.reg_num_mask = MDIO_CTRL_REG_MASK;
-	netif_napi_add(netdev, &adapter->napi, atl1c_clean, 64);
-	setup_timer(&adapter->phy_config_timer, atl1c_phy_config,
-			(unsigned long)adapter);
+	netif_threaded_enable(netdev);
+	for (i = 0; i < adapter->rx_queue_count; ++i)
+		netif_napi_add(netdev, &adapter->rrd_ring[i].napi,
+			       atl1c_clean_rx);
+	for (i = 0; i < adapter->tx_queue_count; ++i)
+		netif_napi_add_tx(netdev, &adapter->tpd_ring[i].napi,
+				  atl1c_clean_tx);
+	timer_setup(&adapter->phy_config_timer, atl1c_phy_config, 0);
 	/* setup the private structure */
 	err = atl1c_sw_init(adapter);
 	if (err) {
@@ -2646,7 +2727,7 @@ static int atl1c_probe(struct pci_dev *pdev, const struct pci_device_id *ent)
 		/* got a random MAC address, set NET_ADDR_RANDOM to netdev */
 		netdev->addr_assign_type = NET_ADDR_RANDOM;
 	}
-	memcpy(netdev->dev_addr, adapter->hw.mac_addr, netdev->addr_len);
+	eth_hw_addr_set(netdev, adapter->hw.mac_addr);
 	if (netif_msg_probe(adapter))
 		dev_dbg(&pdev->dev, "mac address : %pM\n",
 			adapter->hw.mac_addr);
@@ -2660,19 +2741,17 @@ static int atl1c_probe(struct pci_dev *pdev, const struct pci_device_id *ent)
 		goto err_register;
 	}
 
-	if (netif_msg_probe(adapter))
-		dev_info(&pdev->dev, "version %s\n", ATL1C_DRV_VERSION);
 	cards_found++;
 	return 0;
 
 err_reset:
 err_register:
 err_sw_init:
-	iounmap(adapter->hw.hw_addr);
 err_init_netdev:
-err_ioremap:
 	free_netdev(netdev);
 err_alloc_etherdev:
+	iounmap(hw_addr);
+err_ioremap:
 	pci_release_regions(pdev);
 err_pci_reg:
 err_dma:
@@ -2730,7 +2809,7 @@ static pci_ers_result_t atl1c_io_error_detected(struct pci_dev *pdev,
 
 	pci_disable_device(pdev);
 
-	/* Request a slot slot reset. */
+	/* Request a slot reset. */
 	return PCI_ERS_RESULT_NEED_RESET;
 }
 
diff --git a/drivers/net/ethernet/atheros/atl1e/Makefile b/drivers/net/ethernet/atheros/atl1e/Makefile
index bc11be824e76..8506694054a7 100644
--- a/drivers/net/ethernet/atheros/atl1e/Makefile
+++ b/drivers/net/ethernet/atheros/atl1e/Makefile
@@ -1,2 +1,3 @@
+# SPDX-License-Identifier: GPL-2.0-only
 obj-$(CONFIG_ATL1E)	+= atl1e.o
 atl1e-objs		+= atl1e_main.o atl1e_hw.o atl1e_ethtool.o atl1e_param.o
diff --git a/drivers/net/ethernet/atheros/atl1e/atl1e.h b/drivers/net/ethernet/atheros/atl1e/atl1e.h
index 632bb843aed6..9fcad783c939 100644
--- a/drivers/net/ethernet/atheros/atl1e/atl1e.h
+++ b/drivers/net/ethernet/atheros/atl1e/atl1e.h
@@ -1,23 +1,10 @@
+/* SPDX-License-Identifier: GPL-2.0-or-later */
 /*
  * Copyright(c) 2007 Atheros Corporation. All rights reserved.
  * Copyright(c) 2007 xiong huang <xiong.huang@atheros.com>
  *
  * Derived from Intel e1000 driver
  * Copyright(c) 1999 - 2005 Intel Corporation. All rights reserved.
- *
- * This program is free software; you can redistribute it and/or modify it
- * under the terms of the GNU General Public License as published by the Free
- * Software Foundation; either version 2 of the License, or (at your option)
- * any later version.
- *
- * This program is distributed in the hope that it will be useful, but WITHOUT
- * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
- * FITNESS FOR A PARTICULAR PURPOSE.  See the GNU General Public License for
- * more details.
- *
- * You should have received a copy of the GNU General Public License along with
- * this program; if not, write to the Free Software Foundation, Inc., 59
- * Temple Place - Suite 330, Boston, MA  02111-1307, USA.
  */
 
 #ifndef _ATL1E_H_
@@ -495,7 +482,6 @@ struct atl1e_adapter {
 		readl(((a)->hw_addr + reg) + ((offset) << 2)))
 
 extern char atl1e_driver_name[];
-extern char atl1e_driver_version[];
 
 void atl1e_check_options(struct atl1e_adapter *adapter);
 int atl1e_up(struct atl1e_adapter *adapter);
diff --git a/drivers/net/ethernet/atheros/atl1e/atl1e_ethtool.c b/drivers/net/ethernet/atheros/atl1e/atl1e_ethtool.c
index cb489e7e8374..68f1832a198d 100644
--- a/drivers/net/ethernet/atheros/atl1e/atl1e_ethtool.c
+++ b/drivers/net/ethernet/atheros/atl1e/atl1e_ethtool.c
@@ -1,23 +1,9 @@
+// SPDX-License-Identifier: GPL-2.0-or-later
 /*
  * Copyright(c) 2007 Atheros Corporation. All rights reserved.
  *
  * Derived from Intel e1000 driver
  * Copyright(c) 1999 - 2005 Intel Corporation. All rights reserved.
- *
- * This program is free software; you can redistribute it and/or modify it
- * under the terms of the GNU General Public License as published by the Free
- * Software Foundation; either version 2 of the License, or (at your option)
- * any later version.
- *
- * This program is distributed in the hope that it will be useful, but WITHOUT
- * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
- * FITNESS FOR A PARTICULAR PURPOSE.  See the GNU General Public License for
- * more details.
- *
- * You should have received a copy of the GNU General Public License along with
- * this program; if not, write to the Free Software Foundation, Inc., 59
- * Temple Place - Suite 330, Boston, MA  02111-1307, USA.
- *
  */
 
 #include <linux/netdevice.h>
@@ -236,8 +222,8 @@ static int atl1e_get_eeprom(struct net_device *netdev,
 	first_dword = eeprom->offset >> 2;
 	last_dword = (eeprom->offset + eeprom->len - 1) >> 2;
 
-	eeprom_buff = kmalloc(sizeof(u32) *
-			(last_dword - first_dword + 1), GFP_KERNEL);
+	eeprom_buff = kmalloc_array(last_dword - first_dword + 1, sizeof(u32),
+				    GFP_KERNEL);
 	if (eeprom_buff == NULL)
 		return -ENOMEM;
 
@@ -320,11 +306,9 @@ static void atl1e_get_drvinfo(struct net_device *netdev,
 {
 	struct atl1e_adapter *adapter = netdev_priv(netdev);
 
-	strlcpy(drvinfo->driver,  atl1e_driver_name, sizeof(drvinfo->driver));
-	strlcpy(drvinfo->version, atl1e_driver_version,
-		sizeof(drvinfo->version));
-	strlcpy(drvinfo->fw_version, "L1e", sizeof(drvinfo->fw_version));
-	strlcpy(drvinfo->bus_info, pci_name(adapter->pdev),
+	strscpy(drvinfo->driver,  atl1e_driver_name, sizeof(drvinfo->driver));
+	strscpy(drvinfo->fw_version, "L1e", sizeof(drvinfo->fw_version));
+	strscpy(drvinfo->bus_info, pci_name(adapter->pdev),
 		sizeof(drvinfo->bus_info));
 }
 
diff --git a/drivers/net/ethernet/atheros/atl1e/atl1e_hw.c b/drivers/net/ethernet/atheros/atl1e/atl1e_hw.c
index 113565da155f..fa89bc27f62d 100644
--- a/drivers/net/ethernet/atheros/atl1e/atl1e_hw.c
+++ b/drivers/net/ethernet/atheros/atl1e/atl1e_hw.c
@@ -1,22 +1,9 @@
+// SPDX-License-Identifier: GPL-2.0-or-later
 /*
  * Copyright(c) 2007 Atheros Corporation. All rights reserved.
  *
  * Derived from Intel e1000 driver
  * Copyright(c) 1999 - 2005 Intel Corporation. All rights reserved.
- *
- * This program is free software; you can redistribute it and/or modify it
- * under the terms of the GNU General Public License as published by the Free
- * Software Foundation; either version 2 of the License, or (at your option)
- * any later version.
- *
- * This program is distributed in the hope that it will be useful, but WITHOUT
- * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
- * FITNESS FOR A PARTICULAR PURPOSE.  See the GNU General Public License for
- * more details.
- *
- * You should have received a copy of the GNU General Public License along with
- * this program; if not, write to the Free Software Foundation, Inc., 59
- * Temple Place - Suite 330, Boston, MA  02111-1307, USA.
  */
 #include <linux/pci.h>
 #include <linux/delay.h>
diff --git a/drivers/net/ethernet/atheros/atl1e/atl1e_hw.h b/drivers/net/ethernet/atheros/atl1e/atl1e_hw.h
index 88a6271de5bc..3193f7d26945 100644
--- a/drivers/net/ethernet/atheros/atl1e/atl1e_hw.h
+++ b/drivers/net/ethernet/atheros/atl1e/atl1e_hw.h
@@ -1,22 +1,9 @@
+/* SPDX-License-Identifier: GPL-2.0-or-later */
 /*
  * Copyright(c) 2007 Atheros Corporation. All rights reserved.
  *
  * Derived from Intel e1000 driver
  * Copyright(c) 1999 - 2005 Intel Corporation. All rights reserved.
- *
- * This program is free software; you can redistribute it and/or modify it
- * under the terms of the GNU General Public License as published by the Free
- * Software Foundation; either version 2 of the License, or (at your option)
- * any later version.
- *
- * This program is distributed in the hope that it will be useful, but WITHOUT
- * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
- * FITNESS FOR A PARTICULAR PURPOSE.  See the GNU General Public License for
- * more details.
- *
- * You should have received a copy of the GNU General Public License along with
- * this program; if not, write to the Free Software Foundation, Inc., 59
- * Temple Place - Suite 330, Boston, MA  02111-1307, USA.
  */
 
 #ifndef _ATHL1E_HW_H_
diff --git a/drivers/net/ethernet/atheros/atl1e/atl1e_main.c b/drivers/net/ethernet/atheros/atl1e/atl1e_main.c
index 4f7e195af0bc..40290028580b 100644
--- a/drivers/net/ethernet/atheros/atl1e/atl1e_main.c
+++ b/drivers/net/ethernet/atheros/atl1e/atl1e_main.c
@@ -1,30 +1,14 @@
+// SPDX-License-Identifier: GPL-2.0-or-later
 /*
  * Copyright(c) 2007 Atheros Corporation. All rights reserved.
  *
  * Derived from Intel e1000 driver
  * Copyright(c) 1999 - 2005 Intel Corporation. All rights reserved.
- *
- * This program is free software; you can redistribute it and/or modify it
- * under the terms of the GNU General Public License as published by the Free
- * Software Foundation; either version 2 of the License, or (at your option)
- * any later version.
- *
- * This program is distributed in the hope that it will be useful, but WITHOUT
- * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
- * FITNESS FOR A PARTICULAR PURPOSE.  See the GNU General Public License for
- * more details.
- *
- * You should have received a copy of the GNU General Public License along with
- * this program; if not, write to the Free Software Foundation, Inc., 59
- * Temple Place - Suite 330, Boston, MA  02111-1307, USA.
  */
 
 #include "atl1e.h"
 
-#define DRV_VERSION "1.0.0.7-NAPI"
-
 char atl1e_driver_name[] = "ATL1E";
-char atl1e_driver_version[] = DRV_VERSION;
 #define PCI_DEVICE_ID_ATTANSIC_L1E      0x1026
 /*
  * atl1e_pci_tbl - PCI Device ID Table
@@ -46,7 +30,6 @@ MODULE_DEVICE_TABLE(pci, atl1e_pci_tbl);
 MODULE_AUTHOR("Atheros Corporation, <xiong.huang@atheros.com>, Jie Yang <jie.yang@atheros.com>");
 MODULE_DESCRIPTION("Atheros 1000M Ethernet Network Driver");
 MODULE_LICENSE("GPL");
-MODULE_VERSION(DRV_VERSION);
 
 static void atl1e_setup_mac_ctrl(struct atl1e_adapter *adapter);
 
@@ -128,11 +111,12 @@ static inline void atl1e_irq_reset(struct atl1e_adapter *adapter)
 
 /**
  * atl1e_phy_config - Timer Call-back
- * @data: pointer to netdev cast into an unsigned long
+ * @t: timer list containing pointer to netdev cast into an unsigned long
  */
-static void atl1e_phy_config(unsigned long data)
+static void atl1e_phy_config(struct timer_list *t)
 {
-	struct atl1e_adapter *adapter = (struct atl1e_adapter *) data;
+	struct atl1e_adapter *adapter = timer_container_of(adapter, t,
+							   phy_config_timer);
 	struct atl1e_hw *hw = &adapter->hw;
 	unsigned long flags;
 
@@ -143,8 +127,6 @@ static void atl1e_phy_config(unsigned long data)
 
 void atl1e_reinit_locked(struct atl1e_adapter *adapter)
 {
-
-	WARN_ON(in_interrupt());
 	while (test_and_set_bit(__AT_RESETTING, &adapter->flags))
 		msleep(1);
 	atl1e_down(adapter);
@@ -212,7 +194,7 @@ static int atl1e_check_link(struct atl1e_adapter *adapter)
 
 /**
  * atl1e_link_chg_task - deal with link change event Out of interrupt context
- * @netdev: network interface device structure
+ * @work: work struct with driver info
  */
 static void atl1e_link_chg_task(struct work_struct *work)
 {
@@ -250,7 +232,7 @@ static void atl1e_link_chg_event(struct atl1e_adapter *adapter)
 
 static void atl1e_del_timer(struct atl1e_adapter *adapter)
 {
-	del_timer_sync(&adapter->phy_config_timer);
+	timer_delete_sync(&adapter->phy_config_timer);
 }
 
 static void atl1e_cancel_work(struct atl1e_adapter *adapter)
@@ -262,8 +244,9 @@ static void atl1e_cancel_work(struct atl1e_adapter *adapter)
 /**
  * atl1e_tx_timeout - Respond to a Tx Hang
  * @netdev: network interface device structure
+ * @txqueue: the index of the hanging queue
  */
-static void atl1e_tx_timeout(struct net_device *netdev)
+static void atl1e_tx_timeout(struct net_device *netdev, unsigned int txqueue)
 {
 	struct atl1e_adapter *adapter = netdev_priv(netdev);
 
@@ -374,7 +357,7 @@ static void atl1e_restore_vlan(struct atl1e_adapter *adapter)
 }
 
 /**
- * atl1e_set_mac - Change the Ethernet Address of the NIC
+ * atl1e_set_mac_addr - Change the Ethernet Address of the NIC
  * @netdev: network interface device structure
  * @p: pointer to an address structure
  *
@@ -391,7 +374,7 @@ static int atl1e_set_mac_addr(struct net_device *netdev, void *p)
 	if (netif_running(netdev))
 		return -EBUSY;
 
-	memcpy(netdev->dev_addr, addr->sa_data, netdev->addr_len);
+	eth_hw_addr_set(netdev, addr->sa_data);
 	memcpy(adapter->hw.mac_addr, addr->sa_data, netdev->addr_len);
 
 	atl1e_hw_set_mac_addr(&adapter->hw);
@@ -445,7 +428,7 @@ static int atl1e_change_mtu(struct net_device *netdev, int new_mtu)
 	if (netif_running(netdev)) {
 		while (test_and_set_bit(__AT_RESETTING, &adapter->flags))
 			msleep(1);
-		netdev->mtu = new_mtu;
+		WRITE_ONCE(netdev->mtu, new_mtu);
 		adapter->hw.max_frame_size = new_mtu;
 		adapter->hw.rx_jumbo_th = (max_frame + 7) >> 3;
 		atl1e_down(adapter);
@@ -472,7 +455,9 @@ static void atl1e_mdio_write(struct net_device *netdev, int phy_id,
 {
 	struct atl1e_adapter *adapter = netdev_priv(netdev);
 
-	atl1e_write_phy_reg(&adapter->hw, reg_num & MDIO_REG_ADDR_MASK, val);
+	if (atl1e_write_phy_reg(&adapter->hw,
+				reg_num & MDIO_REG_ADDR_MASK, val))
+		netdev_err(netdev, "write phy register failed\n");
 }
 
 static int atl1e_mii_ioctl(struct net_device *netdev,
@@ -668,11 +653,13 @@ static void atl1e_clean_tx_ring(struct atl1e_adapter *adapter)
 		tx_buffer = &tx_ring->tx_buffer[index];
 		if (tx_buffer->dma) {
 			if (tx_buffer->flags & ATL1E_TX_PCIMAP_SINGLE)
-				pci_unmap_single(pdev, tx_buffer->dma,
-					tx_buffer->length, PCI_DMA_TODEVICE);
+				dma_unmap_single(&pdev->dev, tx_buffer->dma,
+						 tx_buffer->length,
+						 DMA_TO_DEVICE);
 			else if (tx_buffer->flags & ATL1E_TX_PCIMAP_PAGE)
-				pci_unmap_page(pdev, tx_buffer->dma,
-					tx_buffer->length, PCI_DMA_TODEVICE);
+				dma_unmap_page(&pdev->dev, tx_buffer->dma,
+					       tx_buffer->length,
+					       DMA_TO_DEVICE);
 			tx_buffer->dma = 0;
 		}
 	}
@@ -788,8 +775,8 @@ static void atl1e_free_ring_resources(struct atl1e_adapter *adapter)
 	atl1e_clean_rx_ring(adapter);
 
 	if (adapter->ring_vir_addr) {
-		pci_free_consistent(pdev, adapter->ring_size,
-				adapter->ring_vir_addr, adapter->ring_dma);
+		dma_free_coherent(&pdev->dev, adapter->ring_size,
+				  adapter->ring_vir_addr, adapter->ring_dma);
 		adapter->ring_vir_addr = NULL;
 	}
 
@@ -800,7 +787,7 @@ static void atl1e_free_ring_resources(struct atl1e_adapter *adapter)
 }
 
 /**
- * atl1e_setup_mem_resources - allocate Tx / RX descriptor resources
+ * atl1e_setup_ring_resources - allocate Tx / RX descriptor resources
  * @adapter: board private structure
  *
  * Return 0 on success, negative on failure
@@ -824,11 +811,12 @@ static int atl1e_setup_ring_resources(struct atl1e_adapter *adapter)
 	/* real ring DMA buffer */
 
 	size = adapter->ring_size;
-	adapter->ring_vir_addr = pci_zalloc_consistent(pdev, adapter->ring_size,
-						       &adapter->ring_dma);
+	adapter->ring_vir_addr = dma_alloc_coherent(&pdev->dev,
+						    adapter->ring_size,
+						    &adapter->ring_dma, GFP_KERNEL);
 	if (adapter->ring_vir_addr == NULL) {
 		netdev_err(adapter->netdev,
-			   "pci_alloc_consistent failed, size = D%d\n", size);
+			   "dma_alloc_coherent failed, size = D%d\n", size);
 		return -ENOMEM;
 	}
 
@@ -878,14 +866,17 @@ static int atl1e_setup_ring_resources(struct atl1e_adapter *adapter)
 		netdev_err(adapter->netdev, "offset(%d) > ring size(%d) !!\n",
 			   offset, adapter->ring_size);
 		err = -1;
-		goto failed;
+		goto free_buffer;
 	}
 
 	return 0;
+free_buffer:
+	kfree(tx_ring->tx_buffer);
+	tx_ring->tx_buffer = NULL;
 failed:
 	if (adapter->ring_vir_addr != NULL) {
-		pci_free_consistent(pdev, adapter->ring_size,
-				adapter->ring_vir_addr, adapter->ring_dma);
+		dma_free_coherent(&pdev->dev, adapter->ring_size,
+				  adapter->ring_vir_addr, adapter->ring_dma);
 		adapter->ring_vir_addr = NULL;
 	}
 	return err;
@@ -1247,16 +1238,20 @@ static bool atl1e_clean_tx_irq(struct atl1e_adapter *adapter)
 		tx_buffer = &tx_ring->tx_buffer[next_to_clean];
 		if (tx_buffer->dma) {
 			if (tx_buffer->flags & ATL1E_TX_PCIMAP_SINGLE)
-				pci_unmap_single(adapter->pdev, tx_buffer->dma,
-					tx_buffer->length, PCI_DMA_TODEVICE);
+				dma_unmap_single(&adapter->pdev->dev,
+						 tx_buffer->dma,
+						 tx_buffer->length,
+						 DMA_TO_DEVICE);
 			else if (tx_buffer->flags & ATL1E_TX_PCIMAP_PAGE)
-				pci_unmap_page(adapter->pdev, tx_buffer->dma,
-					tx_buffer->length, PCI_DMA_TODEVICE);
+				dma_unmap_page(&adapter->pdev->dev,
+					       tx_buffer->dma,
+					       tx_buffer->length,
+					       DMA_TO_DEVICE);
 			tx_buffer->dma = 0;
 		}
 
 		if (tx_buffer->skb) {
-			dev_kfree_skb_irq(tx_buffer->skb);
+			dev_consume_skb_irq(tx_buffer->skb);
 			tx_buffer->skb = NULL;
 		}
 
@@ -1509,6 +1504,8 @@ fatal_err:
 
 /**
  * atl1e_clean - NAPI Rx polling callback
+ * @napi: napi info
+ * @budget: number of packets to clean
  */
 static int atl1e_clean(struct napi_struct *napi, int budget)
 {
@@ -1615,8 +1612,7 @@ static u16 atl1e_cal_tdp_req(const struct sk_buff *skb)
 	if (skb_is_gso(skb)) {
 		if (skb->protocol == htons(ETH_P_IP) ||
 		   (skb_shinfo(skb)->gso_type == SKB_GSO_TCPV6)) {
-			proto_hdr_len = skb_transport_offset(skb) +
-					tcp_hdrlen(skb);
+			proto_hdr_len = skb_tcp_all_headers(skb);
 			if (proto_hdr_len < skb_headlen(skb)) {
 				tpd_req += ((skb_headlen(skb) - proto_hdr_len +
 					   MAX_TX_BUF_LEN - 1) >>
@@ -1648,10 +1644,13 @@ static int atl1e_tso_csum(struct atl1e_adapter *adapter,
 			real_len = (((unsigned char *)ip_hdr(skb) - skb->data)
 					+ ntohs(ip_hdr(skb)->tot_len));
 
-			if (real_len < skb->len)
-				pskb_trim(skb, real_len);
+			if (real_len < skb->len) {
+				err = pskb_trim(skb, real_len);
+				if (err)
+					return err;
+			}
 
-			hdr_len = (skb_transport_offset(skb) + tcp_hdrlen(skb));
+			hdr_len = skb_tcp_all_headers(skb);
 			if (unlikely(skb->len == hdr_len)) {
 				/* only xsum need */
 				netdev_warn(adapter->netdev,
@@ -1719,13 +1718,15 @@ static int atl1e_tx_map(struct atl1e_adapter *adapter,
 	segment = (tpd->word3 >> TPD_SEGMENT_EN_SHIFT) & TPD_SEGMENT_EN_MASK;
 	if (segment) {
 		/* TSO */
-		map_len = hdr_len = skb_transport_offset(skb) + tcp_hdrlen(skb);
+		hdr_len = skb_tcp_all_headers(skb);
+		map_len = hdr_len;
 		use_tpd = tpd;
 
 		tx_buffer = atl1e_get_tx_buffer(adapter, use_tpd);
 		tx_buffer->length = map_len;
-		tx_buffer->dma = pci_map_single(adapter->pdev,
-					skb->data, hdr_len, PCI_DMA_TODEVICE);
+		tx_buffer->dma = dma_map_single(&adapter->pdev->dev,
+						skb->data, hdr_len,
+						DMA_TO_DEVICE);
 		if (dma_mapping_error(&adapter->pdev->dev, tx_buffer->dma))
 			return -ENOSPC;
 
@@ -1753,8 +1754,9 @@ static int atl1e_tx_map(struct atl1e_adapter *adapter,
 			((buf_len - mapped_len) >= MAX_TX_BUF_LEN) ?
 			MAX_TX_BUF_LEN : (buf_len - mapped_len);
 		tx_buffer->dma =
-			pci_map_single(adapter->pdev, skb->data + mapped_len,
-					map_len, PCI_DMA_TODEVICE);
+			dma_map_single(&adapter->pdev->dev,
+				       skb->data + mapped_len, map_len,
+				       DMA_TO_DEVICE);
 
 		if (dma_mapping_error(&adapter->pdev->dev, tx_buffer->dma)) {
 			/* We need to unwind the mappings we've done */
@@ -1763,8 +1765,10 @@ static int atl1e_tx_map(struct atl1e_adapter *adapter,
 			while (adapter->tx_ring.next_to_use != ring_end) {
 				tpd = atl1e_get_tpd(adapter);
 				tx_buffer = atl1e_get_tx_buffer(adapter, tpd);
-				pci_unmap_single(adapter->pdev, tx_buffer->dma,
-						 tx_buffer->length, PCI_DMA_TODEVICE);
+				dma_unmap_single(&adapter->pdev->dev,
+						 tx_buffer->dma,
+						 tx_buffer->length,
+						 DMA_TO_DEVICE);
 			}
 			/* Reset the tx rings next pointer */
 			adapter->tx_ring.next_to_use = ring_start;
@@ -1780,11 +1784,10 @@ static int atl1e_tx_map(struct atl1e_adapter *adapter,
 	}
 
 	for (f = 0; f < nr_frags; f++) {
-		const struct skb_frag_struct *frag;
+		const skb_frag_t *frag = &skb_shinfo(skb)->frags[f];
 		u16 i;
 		u16 seg_num;
 
-		frag = &skb_shinfo(skb)->frags[f];
 		buf_len = skb_frag_size(frag);
 
 		seg_num = (buf_len + MAX_TX_BUF_LEN - 1) / MAX_TX_BUF_LEN;
@@ -2250,7 +2253,7 @@ static const struct net_device_ops atl1e_netdev_ops = {
 	.ndo_fix_features	= atl1e_fix_features,
 	.ndo_set_features	= atl1e_set_features,
 	.ndo_change_mtu		= atl1e_change_mtu,
-	.ndo_do_ioctl		= atl1e_ioctl,
+	.ndo_eth_ioctl		= atl1e_ioctl,
 	.ndo_tx_timeout		= atl1e_tx_timeout,
 #ifdef CONFIG_NET_POLL_CONTROLLER
 	.ndo_poll_controller	= atl1e_netpoll,
@@ -2300,10 +2303,8 @@ static int atl1e_probe(struct pci_dev *pdev, const struct pci_device_id *ent)
 	int err = 0;
 
 	err = pci_enable_device(pdev);
-	if (err) {
-		dev_err(&pdev->dev, "cannot enable PCI device\n");
-		return err;
-	}
+	if (err)
+		return dev_err_probe(&pdev->dev, err, "cannot enable PCI device\n");
 
 	/*
 	 * The atl1e chip can DMA to 64-bit addresses, but it uses a single
@@ -2315,8 +2316,8 @@ static int atl1e_probe(struct pci_dev *pdev, const struct pci_device_id *ent)
 	 * various kernel subsystems to support the mechanics required by a
 	 * fixed-high-32-bit system.
 	 */
-	if ((pci_set_dma_mask(pdev, DMA_BIT_MASK(32)) != 0) ||
-	    (pci_set_consistent_dma_mask(pdev, DMA_BIT_MASK(32)) != 0)) {
+	err = dma_set_mask_and_coherent(&pdev->dev, DMA_BIT_MASK(32));
+	if (err) {
 		dev_err(&pdev->dev, "No usable DMA configuration,aborting\n");
 		goto err_dma;
 	}
@@ -2359,10 +2360,9 @@ static int atl1e_probe(struct pci_dev *pdev, const struct pci_device_id *ent)
 	adapter->mii.phy_id_mask = 0x1f;
 	adapter->mii.reg_num_mask = MDIO_REG_ADDR_MASK;
 
-	netif_napi_add(netdev, &adapter->napi, atl1e_clean, 64);
+	netif_napi_add(netdev, &adapter->napi, atl1e_clean);
 
-	setup_timer(&adapter->phy_config_timer, atl1e_phy_config,
-		    (unsigned long)adapter);
+	timer_setup(&adapter->phy_config_timer, atl1e_phy_config, 0);
 
 	/* get user settings */
 	atl1e_check_options(adapter);
@@ -2396,12 +2396,12 @@ static int atl1e_probe(struct pci_dev *pdev, const struct pci_device_id *ent)
 		goto err_eeprom;
 	}
 
-	memcpy(netdev->dev_addr, adapter->hw.mac_addr, netdev->addr_len);
+	eth_hw_addr_set(netdev, adapter->hw.mac_addr);
 	netdev_dbg(netdev, "mac address : %pM\n", adapter->hw.mac_addr);
 
 	INIT_WORK(&adapter->reset_task, atl1e_reset_task);
 	INIT_WORK(&adapter->link_chg_task, atl1e_link_chg_task);
-	netif_set_gso_max_size(netdev, MAX_TSO_SEG_SIZE);
+	netif_set_tso_max_size(netdev, MAX_TSO_SEG_SIZE);
 	err = register_netdev(netdev);
 	if (err) {
 		netdev_err(netdev, "register netdevice failed\n");
@@ -2488,7 +2488,7 @@ atl1e_io_error_detected(struct pci_dev *pdev, pci_channel_state_t state)
 
 	pci_disable_device(pdev);
 
-	/* Request a slot slot reset. */
+	/* Request a slot reset. */
 	return PCI_ERS_RESULT_NEED_RESET;
 }
 
diff --git a/drivers/net/ethernet/atheros/atl1e/atl1e_param.c b/drivers/net/ethernet/atheros/atl1e/atl1e_param.c
index fa314282c9ad..6b1d6df8da97 100644
--- a/drivers/net/ethernet/atheros/atl1e/atl1e_param.c
+++ b/drivers/net/ethernet/atheros/atl1e/atl1e_param.c
@@ -1,22 +1,9 @@
+// SPDX-License-Identifier: GPL-2.0-or-later
 /*
  * Copyright(c) 2007 Atheros Corporation. All rights reserved.
  *
  * Derived from Intel e1000 driver
  * Copyright(c) 1999 - 2005 Intel Corporation. All rights reserved.
- *
- * This program is free software; you can redistribute it and/or modify it
- * under the terms of the GNU General Public License as published by the Free
- * Software Foundation; either version 2 of the License, or (at your option)
- * any later version.
- *
- * This program is distributed in the hope that it will be useful, but WITHOUT
- * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
- * FITNESS FOR A PARTICULAR PURPOSE.  See the GNU General Public License for
- * more details.
- *
- * You should have received a copy of the GNU General Public License along with
- * this program; if not, write to the Free Software Foundation, Inc., 59
- * Temple Place - Suite 330, Boston, MA  02111-1307, USA.
  */
 
 #include <linux/netdevice.h>
diff --git a/drivers/net/ethernet/atheros/atlx/Makefile b/drivers/net/ethernet/atheros/atlx/Makefile
index e4f6022ca552..df030e421ff3 100644
--- a/drivers/net/ethernet/atheros/atlx/Makefile
+++ b/drivers/net/ethernet/atheros/atlx/Makefile
@@ -1,3 +1,4 @@
+# SPDX-License-Identifier: GPL-2.0-only
 obj-$(CONFIG_ATL1)	+= atl1.o
 obj-$(CONFIG_ATL2)	+= atl2.o
 
diff --git a/drivers/net/ethernet/atheros/atlx/atl1.c b/drivers/net/ethernet/atheros/atlx/atl1.c
index 83d2db2abb45..98a4d089270e 100644
--- a/drivers/net/ethernet/atheros/atlx/atl1.c
+++ b/drivers/net/ethernet/atheros/atlx/atl1.c
@@ -1,3 +1,4 @@
+// SPDX-License-Identifier: GPL-2.0-or-later
 /*
  * Copyright(c) 2005 - 2006 Attansic Corporation. All rights reserved.
  * Copyright(c) 2006 - 2007 Chris Snook <csnook@redhat.com>
@@ -6,23 +7,6 @@
  * Derived from Intel e1000 driver
  * Copyright(c) 1999 - 2005 Intel Corporation. All rights reserved.
  *
- * This program is free software; you can redistribute it and/or modify it
- * under the terms of the GNU General Public License as published by the Free
- * Software Foundation; either version 2 of the License, or (at your option)
- * any later version.
- *
- * This program is distributed in the hope that it will be useful, but WITHOUT
- * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
- * FITNESS FOR A PARTICULAR PURPOSE.  See the GNU General Public License for
- * more details.
- *
- * You should have received a copy of the GNU General Public License along with
- * this program; if not, write to the Free Software Foundation, Inc., 59
- * Temple Place - Suite 330, Boston, MA  02111-1307, USA.
- *
- * The full GNU General Public License is included in this distribution in the
- * file called COPYING.
- *
  * Contact Information:
  * Xiong Huang <xiong.huang@atheros.com>
  * Jie Yang <jie.yang@atheros.com>
@@ -63,7 +47,6 @@
 #include <linux/jiffies.h>
 #include <linux/mii.h>
 #include <linux/module.h>
-#include <linux/moduleparam.h>
 #include <linux/net.h>
 #include <linux/netdevice.h>
 #include <linux/pci.h>
@@ -82,12 +65,10 @@
 
 #include "atl1.h"
 
-#define ATLX_DRIVER_VERSION "2.1.3"
 MODULE_AUTHOR("Xiong Huang <xiong.huang@atheros.com>, "
 	      "Chris Snook <csnook@redhat.com>, "
 	      "Jay Cliburn <jcliburn@gmail.com>");
 MODULE_LICENSE("GPL");
-MODULE_VERSION(ATLX_DRIVER_VERSION);
 
 /* Temporary hack for merging atl1 and atl2 */
 #include "atlx.c"
@@ -1030,7 +1011,7 @@ static int atl1_mii_ioctl(struct net_device *netdev, struct ifreq *ifr, int cmd)
 }
 
 /**
- * atl1_setup_mem_resources - allocate Tx / RX descriptor resources
+ * atl1_setup_ring_resources - allocate Tx / RX descriptor resources
  * @adapter: board private structure
  *
  * Return 0 on success, negative on failure
@@ -1061,7 +1042,7 @@ static s32 atl1_setup_ring_resources(struct atl1_adapter *adapter)
 	 * each ring/block may need up to 8 bytes for alignment, hence the
 	 * additional 40 bytes tacked onto the end.
 	 */
-	ring_header->size = size =
+	ring_header->size =
 		sizeof(struct tx_packet_desc) * tpd_ring->count
 		+ sizeof(struct rx_free_desc) * rfd_ring->count
 		+ sizeof(struct rx_return_desc) * rrd_ring->count
@@ -1069,16 +1050,14 @@ static s32 atl1_setup_ring_resources(struct atl1_adapter *adapter)
 		+ sizeof(struct stats_msg_block)
 		+ 40;
 
-	ring_header->desc = pci_alloc_consistent(pdev, ring_header->size,
-		&ring_header->dma);
+	ring_header->desc = dma_alloc_coherent(&pdev->dev, ring_header->size,
+					       &ring_header->dma, GFP_KERNEL);
 	if (unlikely(!ring_header->desc)) {
 		if (netif_msg_drv(adapter))
-			dev_err(&pdev->dev, "pci_alloc_consistent failed\n");
+			dev_err(&pdev->dev, "dma_alloc_coherent failed\n");
 		goto err_nomem;
 	}
 
-	memset(ring_header->desc, 0, ring_header->size);
-
 	/* init TPD ring */
 	tpd_ring->dma = ring_header->dma;
 	offset = (tpd_ring->dma & 0x7) ? (8 - (ring_header->dma & 0x7)) : 0;
@@ -1157,8 +1136,8 @@ static void atl1_clean_rx_ring(struct atl1_adapter *adapter)
 	for (i = 0; i < rfd_ring->count; i++) {
 		buffer_info = &rfd_ring->buffer_info[i];
 		if (buffer_info->dma) {
-			pci_unmap_page(pdev, buffer_info->dma,
-				buffer_info->length, PCI_DMA_FROMDEVICE);
+			dma_unmap_page(&pdev->dev, buffer_info->dma,
+				       buffer_info->length, DMA_FROM_DEVICE);
 			buffer_info->dma = 0;
 		}
 		if (buffer_info->skb) {
@@ -1196,8 +1175,8 @@ static void atl1_clean_tx_ring(struct atl1_adapter *adapter)
 	for (i = 0; i < tpd_ring->count; i++) {
 		buffer_info = &tpd_ring->buffer_info[i];
 		if (buffer_info->dma) {
-			pci_unmap_page(pdev, buffer_info->dma,
-				buffer_info->length, PCI_DMA_TODEVICE);
+			dma_unmap_page(&pdev->dev, buffer_info->dma,
+				       buffer_info->length, DMA_TO_DEVICE);
 			buffer_info->dma = 0;
 		}
 	}
@@ -1238,8 +1217,8 @@ static void atl1_free_ring_resources(struct atl1_adapter *adapter)
 	atl1_clean_rx_ring(adapter);
 
 	kfree(tpd_ring->buffer_info);
-	pci_free_consistent(pdev, ring_header->size, ring_header->desc,
-		ring_header->dma);
+	dma_free_coherent(&pdev->dev, ring_header->size, ring_header->desc,
+			  ring_header->dma);
 
 	tpd_ring->buffer_info = NULL;
 	tpd_ring->desc = NULL;
@@ -1722,7 +1701,7 @@ static void atl1_inc_smb(struct atl1_adapter *adapter)
 	adapter->soft_stats.scc += smb->tx_1_col;
 	adapter->soft_stats.mcc += smb->tx_2_col;
 	adapter->soft_stats.latecol += smb->tx_late_col;
-	adapter->soft_stats.tx_underun += smb->tx_underrun;
+	adapter->soft_stats.tx_underrun += smb->tx_underrun;
 	adapter->soft_stats.tx_trunc += smb->tx_trunc;
 	adapter->soft_stats.tx_pause += smb->tx_pause;
 
@@ -1882,14 +1861,21 @@ static u16 atl1_alloc_rx_buffers(struct atl1_adapter *adapter)
 			break;
 		}
 
-		buffer_info->alloced = 1;
-		buffer_info->skb = skb;
-		buffer_info->length = (u16) adapter->rx_buffer_len;
 		page = virt_to_page(skb->data);
 		offset = offset_in_page(skb->data);
-		buffer_info->dma = pci_map_page(pdev, page, offset,
+		buffer_info->dma = dma_map_page(&pdev->dev, page, offset,
 						adapter->rx_buffer_len,
-						PCI_DMA_FROMDEVICE);
+						DMA_FROM_DEVICE);
+		if (dma_mapping_error(&pdev->dev, buffer_info->dma)) {
+			kfree_skb(skb);
+			adapter->soft_stats.rx_dropped++;
+			break;
+		}
+
+		buffer_info->alloced = 1;
+		buffer_info->skb = skb;
+		buffer_info->length = (u16)adapter->rx_buffer_len;
+
 		rfd_desc->buffer_addr = cpu_to_le64(buffer_info->dma);
 		rfd_desc->buf_len = cpu_to_le16(adapter->rx_buffer_len);
 		rfd_desc->coalese = 0;
@@ -2013,8 +1999,8 @@ rrd_ok:
 		}
 
 		/* Good Receive */
-		pci_unmap_page(adapter->pdev, buffer_info->dma,
-			       buffer_info->length, PCI_DMA_FROMDEVICE);
+		dma_unmap_page(&adapter->pdev->dev, buffer_info->dma,
+			       buffer_info->length, DMA_FROM_DEVICE);
 		buffer_info->dma = 0;
 		skb = buffer_info->skb;
 		length = le16_to_cpu(rrd->xsz.xsum_sz.pkt_size);
@@ -2083,13 +2069,13 @@ static int atl1_intr_tx(struct atl1_adapter *adapter)
 	while (cmb_tpd_next_to_clean != sw_tpd_next_to_clean) {
 		buffer_info = &tpd_ring->buffer_info[sw_tpd_next_to_clean];
 		if (buffer_info->dma) {
-			pci_unmap_page(adapter->pdev, buffer_info->dma,
-				       buffer_info->length, PCI_DMA_TODEVICE);
+			dma_unmap_page(&adapter->pdev->dev, buffer_info->dma,
+				       buffer_info->length, DMA_TO_DEVICE);
 			buffer_info->dma = 0;
 		}
 
 		if (buffer_info->skb) {
-			dev_kfree_skb_irq(buffer_info->skb);
+			dev_consume_skb_irq(buffer_info->skb);
 			buffer_info->skb = NULL;
 		}
 
@@ -2134,9 +2120,12 @@ static int atl1_tso(struct atl1_adapter *adapter, struct sk_buff *skb,
 
 			real_len = (((unsigned char *)iph - skb->data) +
 				ntohs(iph->tot_len));
-			if (real_len < skb->len)
-				pskb_trim(skb, real_len);
-			hdr_len = (skb_transport_offset(skb) + tcp_hdrlen(skb));
+			if (real_len < skb->len) {
+				err = pskb_trim(skb, real_len);
+				if (err)
+					return err;
+			}
+			hdr_len = skb_tcp_all_headers(skb);
 			if (skb->len == hdr_len) {
 				iph->check = 0;
 				tcp_hdr(skb)->check =
@@ -2201,8 +2190,8 @@ static int atl1_tx_csum(struct atl1_adapter *adapter, struct sk_buff *skb,
 	return 0;
 }
 
-static void atl1_tx_map(struct atl1_adapter *adapter, struct sk_buff *skb,
-	struct tx_packet_desc *ptpd)
+static bool atl1_tx_map(struct atl1_adapter *adapter, struct sk_buff *skb,
+			struct tx_packet_desc *ptpd)
 {
 	struct atl1_tpd_ring *tpd_ring = &adapter->tpd_ring;
 	struct atl1_buffer *buffer_info;
@@ -2212,6 +2201,7 @@ static void atl1_tx_map(struct atl1_adapter *adapter, struct sk_buff *skb,
 	unsigned int nr_frags;
 	unsigned int f;
 	int retval;
+	u16 first_mapped;
 	u16 next_to_use;
 	u16 data_len;
 	u8 hdr_len;
@@ -2219,6 +2209,7 @@ static void atl1_tx_map(struct atl1_adapter *adapter, struct sk_buff *skb,
 	buf_len -= skb->data_len;
 	nr_frags = skb_shinfo(skb)->nr_frags;
 	next_to_use = atomic_read(&tpd_ring->next_to_use);
+	first_mapped = next_to_use;
 	buffer_info = &tpd_ring->buffer_info[next_to_use];
 	BUG_ON(buffer_info->skb);
 	/* put skb in last TPD */
@@ -2227,13 +2218,15 @@ static void atl1_tx_map(struct atl1_adapter *adapter, struct sk_buff *skb,
 	retval = (ptpd->word3 >> TPD_SEGMENT_EN_SHIFT) & TPD_SEGMENT_EN_MASK;
 	if (retval) {
 		/* TSO */
-		hdr_len = skb_transport_offset(skb) + tcp_hdrlen(skb);
+		hdr_len = skb_tcp_all_headers(skb);
 		buffer_info->length = hdr_len;
 		page = virt_to_page(skb->data);
 		offset = offset_in_page(skb->data);
-		buffer_info->dma = pci_map_page(adapter->pdev, page,
+		buffer_info->dma = dma_map_page(&adapter->pdev->dev, page,
 						offset, hdr_len,
-						PCI_DMA_TODEVICE);
+						DMA_TO_DEVICE);
+		if (dma_mapping_error(&adapter->pdev->dev, buffer_info->dma))
+			goto dma_err;
 
 		if (++next_to_use == tpd_ring->count)
 			next_to_use = 0;
@@ -2256,9 +2249,13 @@ static void atl1_tx_map(struct atl1_adapter *adapter, struct sk_buff *skb,
 					(hdr_len + i * ATL1_MAX_TX_BUF_LEN));
 				offset = offset_in_page(skb->data +
 					(hdr_len + i * ATL1_MAX_TX_BUF_LEN));
-				buffer_info->dma = pci_map_page(adapter->pdev,
-					page, offset, buffer_info->length,
-					PCI_DMA_TODEVICE);
+				buffer_info->dma = dma_map_page(&adapter->pdev->dev,
+								page, offset,
+								buffer_info->length,
+								DMA_TO_DEVICE);
+				if (dma_mapping_error(&adapter->pdev->dev,
+						      buffer_info->dma))
+					goto dma_err;
 				if (++next_to_use == tpd_ring->count)
 					next_to_use = 0;
 			}
@@ -2268,17 +2265,19 @@ static void atl1_tx_map(struct atl1_adapter *adapter, struct sk_buff *skb,
 		buffer_info->length = buf_len;
 		page = virt_to_page(skb->data);
 		offset = offset_in_page(skb->data);
-		buffer_info->dma = pci_map_page(adapter->pdev, page,
-			offset, buf_len, PCI_DMA_TODEVICE);
+		buffer_info->dma = dma_map_page(&adapter->pdev->dev, page,
+						offset, buf_len,
+						DMA_TO_DEVICE);
+		if (dma_mapping_error(&adapter->pdev->dev, buffer_info->dma))
+			goto dma_err;
 		if (++next_to_use == tpd_ring->count)
 			next_to_use = 0;
 	}
 
 	for (f = 0; f < nr_frags; f++) {
-		const struct skb_frag_struct *frag;
+		const skb_frag_t *frag = &skb_shinfo(skb)->frags[f];
 		u16 i, nseg;
 
-		frag = &skb_shinfo(skb)->frags[f];
 		buf_len = skb_frag_size(frag);
 
 		nseg = (buf_len + ATL1_MAX_TX_BUF_LEN - 1) /
@@ -2294,6 +2293,9 @@ static void atl1_tx_map(struct atl1_adapter *adapter, struct sk_buff *skb,
 			buffer_info->dma = skb_frag_dma_map(&adapter->pdev->dev,
 				frag, i * ATL1_MAX_TX_BUF_LEN,
 				buffer_info->length, DMA_TO_DEVICE);
+			if (dma_mapping_error(&adapter->pdev->dev,
+					      buffer_info->dma))
+				goto dma_err;
 
 			if (++next_to_use == tpd_ring->count)
 				next_to_use = 0;
@@ -2302,6 +2304,22 @@ static void atl1_tx_map(struct atl1_adapter *adapter, struct sk_buff *skb,
 
 	/* last tpd's buffer-info */
 	buffer_info->skb = skb;
+
+	return true;
+
+ dma_err:
+	while (first_mapped != next_to_use) {
+		buffer_info = &tpd_ring->buffer_info[first_mapped];
+		dma_unmap_page(&adapter->pdev->dev,
+			       buffer_info->dma,
+			       buffer_info->length,
+			       DMA_TO_DEVICE);
+		buffer_info->dma = 0;
+
+		if (++first_mapped == tpd_ring->count)
+			first_mapped = 0;
+	}
+	return false;
 }
 
 static void atl1_tx_queue(struct atl1_adapter *adapter, u16 count,
@@ -2372,10 +2390,8 @@ static netdev_tx_t atl1_xmit_frame(struct sk_buff *skb,
 
 	len = skb_headlen(skb);
 
-	if (unlikely(skb->len <= 0)) {
-		dev_kfree_skb_any(skb);
-		return NETDEV_TX_OK;
-	}
+	if (unlikely(skb->len <= 0))
+		goto drop_packet;
 
 	nr_frags = skb_shinfo(skb)->nr_frags;
 	for (f = 0; f < nr_frags; f++) {
@@ -2387,12 +2403,10 @@ static netdev_tx_t atl1_xmit_frame(struct sk_buff *skb,
 	mss = skb_shinfo(skb)->gso_size;
 	if (mss) {
 		if (skb->protocol == htons(ETH_P_IP)) {
-			proto_hdr_len = (skb_transport_offset(skb) +
-					 tcp_hdrlen(skb));
-			if (unlikely(proto_hdr_len > len)) {
-				dev_kfree_skb_any(skb);
-				return NETDEV_TX_OK;
-			}
+			proto_hdr_len = skb_tcp_all_headers(skb);
+			if (unlikely(proto_hdr_len > len))
+				goto drop_packet;
+
 			/* need additional TPD ? */
 			if (proto_hdr_len != len)
 				count += (len - proto_hdr_len +
@@ -2424,23 +2438,25 @@ static netdev_tx_t atl1_xmit_frame(struct sk_buff *skb,
 	}
 
 	tso = atl1_tso(adapter, skb, ptpd);
-	if (tso < 0) {
-		dev_kfree_skb_any(skb);
-		return NETDEV_TX_OK;
-	}
+	if (tso < 0)
+		goto drop_packet;
 
 	if (!tso) {
 		ret_val = atl1_tx_csum(adapter, skb, ptpd);
-		if (ret_val < 0) {
-			dev_kfree_skb_any(skb);
-			return NETDEV_TX_OK;
-		}
+		if (ret_val < 0)
+			goto drop_packet;
 	}
 
-	atl1_tx_map(adapter, skb, ptpd);
+	if (!atl1_tx_map(adapter, skb, ptpd))
+		goto drop_packet;
+
 	atl1_tx_queue(adapter, count, ptpd);
 	atl1_update_mailbox(adapter);
-	mmiowb();
+	return NETDEV_TX_OK;
+
+drop_packet:
+	adapter->soft_stats.tx_errors++;
+	dev_kfree_skb_any(skb);
 	return NETDEV_TX_OK;
 }
 
@@ -2465,7 +2481,7 @@ static int atl1_rings_clean(struct napi_struct *napi, int budget)
 
 static inline int atl1_sched_rings_clean(struct atl1_adapter* adapter)
 {
-	if (!napi_schedule_prep(&adapter->napi))
+	if (!napi_schedule(&adapter->napi))
 		/* It is possible in case even the RX/TX ints are disabled via IMR
 		 * register the ISR bits are set anyway (but do not produce IRQ).
 		 * To handle such situation the napi functions used to check is
@@ -2473,8 +2489,6 @@ static inline int atl1_sched_rings_clean(struct atl1_adapter* adapter)
 		 */
 		return 0;
 
-	__napi_schedule(&adapter->napi);
-
 	/*
 	 * Disable RX/TX ints via IMR register if it is
 	 * allowed. NAPI handler must reenable them in same
@@ -2573,11 +2587,12 @@ static irqreturn_t atl1_intr(int irq, void *data)
 
 /**
  * atl1_phy_config - Timer Call-back
- * @data: pointer to netdev cast into an unsigned long
+ * @t: timer_list containing pointer to netdev cast into an unsigned long
  */
-static void atl1_phy_config(unsigned long data)
+static void atl1_phy_config(struct timer_list *t)
 {
-	struct atl1_adapter *adapter = (struct atl1_adapter *)data;
+	struct atl1_adapter *adapter = timer_container_of(adapter, t,
+							  phy_config_timer);
 	struct atl1_hw *hw = &adapter->hw;
 	unsigned long flags;
 
@@ -2661,7 +2676,7 @@ static void atl1_down(struct atl1_adapter *adapter)
 
 	napi_disable(&adapter->napi);
 	netif_stop_queue(netdev);
-	del_timer_sync(&adapter->phy_config_timer);
+	timer_delete_sync(&adapter->phy_config_timer);
 	adapter->phy_timer_pending = false;
 
 	atlx_irq_disable(adapter);
@@ -2707,7 +2722,7 @@ static int atl1_change_mtu(struct net_device *netdev, int new_mtu)
 	adapter->rx_buffer_len = (max_frame + 7) & ~7;
 	adapter->hw.rx_jumbo_th = adapter->rx_buffer_len / 8;
 
-	netdev->mtu = new_mtu;
+	WRITE_ONCE(netdev->mtu, new_mtu);
 	if (netif_running(netdev)) {
 		atl1_down(adapter);
 		atl1_up(adapter);
@@ -2773,8 +2788,7 @@ static int atl1_close(struct net_device *netdev)
 #ifdef CONFIG_PM_SLEEP
 static int atl1_suspend(struct device *dev)
 {
-	struct pci_dev *pdev = to_pci_dev(dev);
-	struct net_device *netdev = pci_get_drvdata(pdev);
+	struct net_device *netdev = dev_get_drvdata(dev);
 	struct atl1_adapter *adapter = netdev_priv(netdev);
 	struct atl1_hw *hw = &adapter->hw;
 	u32 ctrl = 0;
@@ -2799,7 +2813,7 @@ static int atl1_suspend(struct device *dev)
 		val = atl1_get_speed_and_duplex(hw, &speed, &duplex);
 		if (val) {
 			if (netif_msg_ifdown(adapter))
-				dev_printk(KERN_DEBUG, &pdev->dev,
+				dev_printk(KERN_DEBUG, dev,
 					"error getting speed/duplex\n");
 			goto disable_wol;
 		}
@@ -2856,8 +2870,7 @@ static int atl1_suspend(struct device *dev)
 
 static int atl1_resume(struct device *dev)
 {
-	struct pci_dev *pdev = to_pci_dev(dev);
-	struct net_device *netdev = pci_get_drvdata(pdev);
+	struct net_device *netdev = dev_get_drvdata(dev);
 	struct atl1_adapter *adapter = netdev_priv(netdev);
 
 	iowrite32(0, adapter->hw.hw_addr + REG_WOL_CTRL);
@@ -2907,7 +2920,7 @@ static const struct net_device_ops atl1_netdev_ops = {
 	.ndo_change_mtu		= atl1_change_mtu,
 	.ndo_fix_features	= atlx_fix_features,
 	.ndo_set_features	= atlx_set_features,
-	.ndo_do_ioctl		= atlx_ioctl,
+	.ndo_eth_ioctl		= atlx_ioctl,
 	.ndo_tx_timeout		= atlx_tx_timeout,
 #ifdef CONFIG_NET_POLL_CONTROLLER
 	.ndo_poll_controller	= atl1_poll_controller,
@@ -2946,7 +2959,7 @@ static int atl1_probe(struct pci_dev *pdev, const struct pci_device_id *ent)
 	 * various kernel subsystems to support the mechanics required by a
 	 * fixed-high-32-bit system.
 	 */
-	err = pci_set_dma_mask(pdev, DMA_BIT_MASK(32));
+	err = dma_set_mask(&pdev->dev, DMA_BIT_MASK(32));
 	if (err) {
 		dev_err(&pdev->dev, "no usable DMA configuration\n");
 		goto err_dma;
@@ -2987,8 +3000,6 @@ static int atl1_probe(struct pci_dev *pdev, const struct pci_device_id *ent)
 	/* get device revision number */
 	adapter->hw.dev_rev = ioread16(adapter->hw.hw_addr +
 		(REG_MASTER_CTRL + 2));
-	if (netif_msg_probe(adapter))
-		dev_info(&pdev->dev, "version %s\n", ATLX_DRIVER_VERSION);
 
 	/* set default ring resource counts */
 	adapter->rfd_ring.count = adapter->rrd_ring.count = ATL1_DEFAULT_RFD;
@@ -3002,7 +3013,7 @@ static int atl1_probe(struct pci_dev *pdev, const struct pci_device_id *ent)
 
 	netdev->netdev_ops = &atl1_netdev_ops;
 	netdev->watchdog_timeo = 5 * HZ;
-	netif_napi_add(netdev, &adapter->napi, atl1_rings_clean, 64);
+	netif_napi_add(netdev, &adapter->napi, atl1_rings_clean);
 
 	netdev->ethtool_ops = &atl1_ethtool_ops;
 	adapter->bd_number = cards_found;
@@ -3051,7 +3062,7 @@ static int atl1_probe(struct pci_dev *pdev, const struct pci_device_id *ent)
 		/* mark random mac */
 		netdev->addr_assign_type = NET_ADDR_RANDOM;
 	}
-	memcpy(netdev->dev_addr, adapter->hw.mac_addr, netdev->addr_len);
+	eth_hw_addr_set(netdev, adapter->hw.mac_addr);
 
 	if (!is_valid_ether_addr(netdev->dev_addr)) {
 		err = -EIO;
@@ -3071,8 +3082,7 @@ static int atl1_probe(struct pci_dev *pdev, const struct pci_device_id *ent)
 	/* assume we have no link for now */
 	netif_carrier_off(netdev);
 
-	setup_timer(&adapter->phy_config_timer, atl1_phy_config,
-		    (unsigned long)adapter);
+	timer_setup(&adapter->phy_config_timer, atl1_phy_config, 0);
 	adapter->phy_timer_pending = false;
 
 	INIT_WORK(&adapter->reset_dev_task, atl1_reset_dev_task);
@@ -3180,7 +3190,7 @@ static struct atl1_stats atl1_gstrings_stats[] = {
 	{"tx_deferred_ok", ATL1_STAT(soft_stats.deffer)},
 	{"tx_single_coll_ok", ATL1_STAT(soft_stats.scc)},
 	{"tx_multi_coll_ok", ATL1_STAT(soft_stats.mcc)},
-	{"tx_underun", ATL1_STAT(soft_stats.tx_underun)},
+	{"tx_underrun", ATL1_STAT(soft_stats.tx_underrun)},
 	{"tx_trunc", ATL1_STAT(soft_stats.tx_trunc)},
 	{"tx_pause", ATL1_STAT(soft_stats.tx_pause)},
 	{"rx_pause", ATL1_STAT(soft_stats.rx_pause)},
@@ -3278,7 +3288,6 @@ static int atl1_set_link_ksettings(struct net_device *netdev,
 	u16 phy_data;
 	int ret_val = 0;
 	u16 old_media_type = hw->media_type;
-	u32 advertising;
 
 	if (netif_running(adapter->netdev)) {
 		if (netif_msg_link(adapter))
@@ -3312,25 +3321,7 @@ static int atl1_set_link_ksettings(struct net_device *netdev,
 				hw->media_type = MEDIA_TYPE_10M_HALF;
 		}
 	}
-	switch (hw->media_type) {
-	case MEDIA_TYPE_AUTO_SENSOR:
-		advertising =
-		    ADVERTISED_10baseT_Half |
-		    ADVERTISED_10baseT_Full |
-		    ADVERTISED_100baseT_Half |
-		    ADVERTISED_100baseT_Full |
-		    ADVERTISED_1000baseT_Full |
-		    ADVERTISED_Autoneg | ADVERTISED_TP;
-		break;
-	case MEDIA_TYPE_1000M_FULL:
-		advertising =
-		    ADVERTISED_1000baseT_Full |
-		    ADVERTISED_Autoneg | ADVERTISED_TP;
-		break;
-	default:
-		advertising = 0;
-		break;
-	}
+
 	if (atl1_phy_setup_autoneg_adv(hw)) {
 		ret_val = -EINVAL;
 		if (netif_msg_link(adapter))
@@ -3385,10 +3376,8 @@ static void atl1_get_drvinfo(struct net_device *netdev,
 {
 	struct atl1_adapter *adapter = netdev_priv(netdev);
 
-	strlcpy(drvinfo->driver, ATLX_DRIVER_NAME, sizeof(drvinfo->driver));
-	strlcpy(drvinfo->version, ATLX_DRIVER_VERSION,
-		sizeof(drvinfo->version));
-	strlcpy(drvinfo->bus_info, pci_name(adapter->pdev),
+	strscpy(drvinfo->driver, ATLX_DRIVER_NAME, sizeof(drvinfo->driver));
+	strscpy(drvinfo->bus_info, pci_name(adapter->pdev),
 		sizeof(drvinfo->bus_info));
 }
 
@@ -3484,7 +3473,9 @@ static void atl1_get_regs(struct net_device *netdev, struct ethtool_regs *regs,
 }
 
 static void atl1_get_ringparam(struct net_device *netdev,
-	struct ethtool_ringparam *ring)
+			       struct ethtool_ringparam *ring,
+			       struct kernel_ethtool_ringparam *kernel_ring,
+			       struct netlink_ext_ack *extack)
 {
 	struct atl1_adapter *adapter = netdev_priv(netdev);
 	struct atl1_tpd_ring *txdr = &adapter->tpd_ring;
@@ -3497,7 +3488,9 @@ static void atl1_get_ringparam(struct net_device *netdev,
 }
 
 static int atl1_set_ringparam(struct net_device *netdev,
-	struct ethtool_ringparam *ring)
+			      struct ethtool_ringparam *ring,
+			      struct kernel_ethtool_ringparam *kernel_ring,
+			      struct netlink_ext_ack *extack)
 {
 	struct atl1_adapter *adapter = netdev_priv(netdev);
 	struct atl1_tpd_ring *tpdr = &adapter->tpd_ring;
diff --git a/drivers/net/ethernet/atheros/atlx/atl1.h b/drivers/net/ethernet/atheros/atlx/atl1.h
index 34a58cd846a0..6b264c5127a5 100644
--- a/drivers/net/ethernet/atheros/atlx/atl1.h
+++ b/drivers/net/ethernet/atheros/atlx/atl1.h
@@ -1,3 +1,4 @@
+/* SPDX-License-Identifier: GPL-2.0-or-later */
 /*
  * Copyright(c) 2005 - 2006 Attansic Corporation. All rights reserved.
  * Copyright(c) 2006 - 2007 Chris Snook <csnook@redhat.com>
@@ -5,20 +6,6 @@
  *
  * Derived from Intel e1000 driver
  * Copyright(c) 1999 - 2005 Intel Corporation. All rights reserved.
- *
- * This program is free software; you can redistribute it and/or modify it
- * under the terms of the GNU General Public License as published by the Free
- * Software Foundation; either version 2 of the License, or (at your option)
- * any later version.
- *
- * This program is distributed in the hope that it will be useful, but WITHOUT
- * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
- * FITNESS FOR A PARTICULAR PURPOSE.  See the GNU General Public License for
- * more details.
- *
- * You should have received a copy of the GNU General Public License along with
- * this program; if not, write to the Free Software Foundation, Inc., 59
- * Temple Place - Suite 330, Boston, MA  02111-1307, USA.
  */
 
 #ifndef ATL1_H
@@ -681,7 +668,7 @@ struct atl1_sft_stats {
 	u64 scc;		/* packets TX after a single collision */
 	u64 mcc;		/* packets TX after multiple collisions */
 	u64 latecol;		/* TX packets w/ late collisions */
-	u64 tx_underun;		/* TX packets aborted due to TX FIFO underrun
+	u64 tx_underrun;	/* TX packets aborted due to TX FIFO underrun
 				 * or TRD FIFO underrun */
 	u64 tx_trunc;		/* TX packets truncated due to size > MTU */
 	u64 rx_pause;		/* num Pause packets received. */
diff --git a/drivers/net/ethernet/atheros/atlx/atl2.c b/drivers/net/ethernet/atheros/atlx/atl2.c
index 77a1c03255de..280e2f5f4aa5 100644
--- a/drivers/net/ethernet/atheros/atlx/atl2.c
+++ b/drivers/net/ethernet/atheros/atlx/atl2.c
@@ -1,23 +1,10 @@
+// SPDX-License-Identifier: GPL-2.0-or-later
 /*
  * Copyright(c) 2006 - 2007 Atheros Corporation. All rights reserved.
  * Copyright(c) 2007 - 2008 Chris Snook <csnook@redhat.com>
  *
  * Derived from Intel e1000 driver
  * Copyright(c) 1999 - 2005 Intel Corporation. All rights reserved.
- *
- * This program is free software; you can redistribute it and/or modify it
- * under the terms of the GNU General Public License as published by the Free
- * Software Foundation; either version 2 of the License, or (at your option)
- * any later version.
- *
- * This program is distributed in the hope that it will be useful, but WITHOUT
- * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
- * FITNESS FOR A PARTICULAR PURPOSE.  See the GNU General Public License for
- * more details.
- *
- * You should have received a copy of the GNU General Public License along with
- * this program; if not, write to the Free Software Foundation, Inc., 59
- * Temple Place - Suite 330, Boston, MA  02111-1307, USA.
  */
 
 #include <linux/atomic.h>
@@ -49,18 +36,12 @@
 
 #include "atl2.h"
 
-#define ATL2_DRV_VERSION "2.2.3"
-
 static const char atl2_driver_name[] = "atl2";
-static const char atl2_driver_string[] = "Atheros(R) L2 Ethernet Driver";
-static const char atl2_copyright[] = "Copyright (c) 2007 Atheros Corporation.";
-static const char atl2_driver_version[] = ATL2_DRV_VERSION;
 static const struct ethtool_ops atl2_ethtool_ops;
 
 MODULE_AUTHOR("Atheros Corporation <xiong.huang@atheros.com>, Chris Snook <csnook@redhat.com>");
 MODULE_DESCRIPTION("Atheros Fast Ethernet Network Driver");
 MODULE_LICENSE("GPL");
-MODULE_VERSION(ATL2_DRV_VERSION);
 
 /*
  * atl2_pci_tbl - PCI Device ID Table
@@ -300,11 +281,10 @@ static s32 atl2_setup_ring_resources(struct atl2_adapter *adapter)
 		adapter->txs_ring_size * 4 + 7 +	/* dword align */
 		adapter->rxd_ring_size * 1536 + 127;	/* 128bytes align */
 
-	adapter->ring_vir_addr = pci_alloc_consistent(pdev, size,
-		&adapter->ring_dma);
+	adapter->ring_vir_addr = dma_alloc_coherent(&pdev->dev, size,
+						    &adapter->ring_dma, GFP_KERNEL);
 	if (!adapter->ring_vir_addr)
 		return -ENOMEM;
-	memset(adapter->ring_vir_addr, 0, adapter->ring_size);
 
 	/* Init TXD Ring */
 	adapter->txd_dma = adapter->ring_dma ;
@@ -553,7 +533,7 @@ static void atl2_intr_tx(struct atl2_adapter *adapter)
 			netdev->stats.tx_aborted_errors++;
 		if (txs->late_col)
 			netdev->stats.tx_window_errors++;
-		if (txs->underun)
+		if (txs->underrun)
 			netdev->stats.tx_fifo_errors++;
 	} while (1);
 
@@ -683,8 +663,8 @@ static int atl2_request_irq(struct atl2_adapter *adapter)
 static void atl2_free_ring_resources(struct atl2_adapter *adapter)
 {
 	struct pci_dev *pdev = adapter->pdev;
-	pci_free_consistent(pdev, adapter->ring_size, adapter->ring_vir_addr,
-		adapter->ring_dma);
+	dma_free_coherent(&pdev->dev, adapter->ring_size,
+			  adapter->ring_vir_addr, adapter->ring_dma);
 }
 
 /**
@@ -772,8 +752,8 @@ static void atl2_down(struct atl2_adapter *adapter)
 
 	atl2_irq_disable(adapter);
 
-	del_timer_sync(&adapter->watchdog_timer);
-	del_timer_sync(&adapter->phy_config_timer);
+	timer_delete_sync(&adapter->watchdog_timer);
+	timer_delete_sync(&adapter->phy_config_timer);
 	clear_bit(0, &adapter->cfg_phy);
 
 	netif_carrier_off(netdev);
@@ -908,8 +888,7 @@ static netdev_tx_t atl2_xmit_frame(struct sk_buff *skb,
 	ATL2_WRITE_REGW(&adapter->hw, REG_MB_TXD_WR_IDX,
 		(adapter->txd_write_ptr >> 2));
 
-	mmiowb();
-	dev_kfree_skb_any(skb);
+	dev_consume_skb_any(skb);
 	return NETDEV_TX_OK;
 }
 
@@ -926,7 +905,7 @@ static int atl2_change_mtu(struct net_device *netdev, int new_mtu)
 	struct atl2_hw *hw = &adapter->hw;
 
 	/* set MTU */
-	netdev->mtu = new_mtu;
+	WRITE_ONCE(netdev->mtu, new_mtu);
 	hw->max_frame_size = new_mtu;
 	ATL2_WRITE_REG(hw, REG_MTU, new_mtu + ETH_HLEN +
 		       VLAN_HLEN + ETH_FCS_LEN);
@@ -952,7 +931,7 @@ static int atl2_set_mac(struct net_device *netdev, void *p)
 	if (netif_running(netdev))
 		return -EBUSY;
 
-	memcpy(netdev->dev_addr, addr->sa_data, netdev->addr_len);
+	eth_hw_addr_set(netdev, addr->sa_data);
 	memcpy(adapter->hw.mac_addr, addr->sa_data, netdev->addr_len);
 
 	atl2_set_mac_addr(&adapter->hw);
@@ -1015,8 +994,9 @@ static int atl2_ioctl(struct net_device *netdev, struct ifreq *ifr, int cmd)
 /**
  * atl2_tx_timeout - Respond to a Tx Hang
  * @netdev: network interface device structure
+ * @txqueue: index of the hanging transmit queue
  */
-static void atl2_tx_timeout(struct net_device *netdev)
+static void atl2_tx_timeout(struct net_device *netdev, unsigned int txqueue)
 {
 	struct atl2_adapter *adapter = netdev_priv(netdev);
 
@@ -1026,11 +1006,12 @@ static void atl2_tx_timeout(struct net_device *netdev)
 
 /**
  * atl2_watchdog - Timer Call-back
- * @data: pointer to netdev cast into an unsigned long
+ * @t: timer list containing a pointer to netdev cast into an unsigned long
  */
-static void atl2_watchdog(unsigned long data)
+static void atl2_watchdog(struct timer_list *t)
 {
-	struct atl2_adapter *adapter = (struct atl2_adapter *) data;
+	struct atl2_adapter *adapter = timer_container_of(adapter, t,
+							  watchdog_timer);
 
 	if (!test_bit(__ATL2_DOWN, &adapter->flags)) {
 		u32 drop_rxd, drop_rxs;
@@ -1051,11 +1032,12 @@ static void atl2_watchdog(unsigned long data)
 
 /**
  * atl2_phy_config - Timer Call-back
- * @data: pointer to netdev cast into an unsigned long
+ * @t: timer list containing a pointer to netdev cast into an unsigned long
  */
-static void atl2_phy_config(unsigned long data)
+static void atl2_phy_config(struct timer_list *t)
 {
-	struct atl2_adapter *adapter = (struct atl2_adapter *) data;
+	struct atl2_adapter *adapter = timer_container_of(adapter, t,
+							  phy_config_timer);
 	struct atl2_hw *hw = &adapter->hw;
 	unsigned long flags;
 
@@ -1105,7 +1087,6 @@ err_up:
 
 static void atl2_reinit_locked(struct atl2_adapter *adapter)
 {
-	WARN_ON(in_interrupt());
 	while (test_and_set_bit(__ATL2_RESETTING, &adapter->flags))
 		msleep(1);
 	atl2_down(adapter);
@@ -1255,6 +1236,7 @@ static int atl2_check_link(struct atl2_adapter *adapter)
 
 /**
  * atl2_link_chg_task - deal with link change event Out of interrupt context
+ * @work: pointer to work struct with private info
  */
 static void atl2_link_chg_task(struct work_struct *work)
 {
@@ -1312,7 +1294,7 @@ static const struct net_device_ops atl2_netdev_ops = {
 	.ndo_change_mtu		= atl2_change_mtu,
 	.ndo_fix_features	= atl2_fix_features,
 	.ndo_set_features	= atl2_set_features,
-	.ndo_do_ioctl		= atl2_ioctl,
+	.ndo_eth_ioctl		= atl2_ioctl,
 	.ndo_tx_timeout		= atl2_tx_timeout,
 #ifdef CONFIG_NET_POLL_CONTROLLER
 	.ndo_poll_controller	= atl2_poll_controller,
@@ -1334,13 +1316,11 @@ static int atl2_probe(struct pci_dev *pdev, const struct pci_device_id *ent)
 {
 	struct net_device *netdev;
 	struct atl2_adapter *adapter;
-	static int cards_found;
+	static int cards_found = 0;
 	unsigned long mmio_start;
 	int mmio_len;
 	int err;
 
-	cards_found = 0;
-
 	err = pci_enable_device(pdev);
 	if (err)
 		return err;
@@ -1350,8 +1330,8 @@ static int atl2_probe(struct pci_dev *pdev, const struct pci_device_id *ent)
 	 * until the kernel has the proper infrastructure to support 64-bit DMA
 	 * on these devices.
 	 */
-	if (pci_set_dma_mask(pdev, DMA_BIT_MASK(32)) &&
-		pci_set_consistent_dma_mask(pdev, DMA_BIT_MASK(32))) {
+	if (dma_set_mask(&pdev->dev, DMA_BIT_MASK(32)) &&
+	    dma_set_coherent_mask(&pdev->dev, DMA_BIT_MASK(32))) {
 		printk(KERN_ERR "atl2: No usable DMA configuration, aborting\n");
 		err = -EIO;
 		goto err_dma;
@@ -1398,7 +1378,7 @@ static int atl2_probe(struct pci_dev *pdev, const struct pci_device_id *ent)
 	netdev->watchdog_timeo = 5 * HZ;
 	netdev->min_mtu = 40;
 	netdev->max_mtu = ETH_DATA_LEN + VLAN_HLEN;
-	strncpy(netdev->name, pci_name(pdev), sizeof(netdev->name) - 1);
+	strscpy(netdev->name, pci_name(pdev), sizeof(netdev->name));
 
 	netdev->mem_start = mmio_start;
 	netdev->mem_end = mmio_start + mmio_len;
@@ -1426,7 +1406,7 @@ static int atl2_probe(struct pci_dev *pdev, const struct pci_device_id *ent)
 
 	/* copy the MAC address out of the EEPROM */
 	atl2_read_mac_addr(&adapter->hw);
-	memcpy(netdev->dev_addr, adapter->hw.mac_addr, netdev->addr_len);
+	eth_hw_addr_set(netdev, adapter->hw.mac_addr);
 	if (!is_valid_ether_addr(netdev->dev_addr)) {
 		err = -EIO;
 		goto err_eeprom;
@@ -1434,11 +1414,9 @@ static int atl2_probe(struct pci_dev *pdev, const struct pci_device_id *ent)
 
 	atl2_check_options(adapter);
 
-	setup_timer(&adapter->watchdog_timer, atl2_watchdog,
-		    (unsigned long)adapter);
+	timer_setup(&adapter->watchdog_timer, atl2_watchdog, 0);
 
-	setup_timer(&adapter->phy_config_timer, atl2_phy_config,
-		    (unsigned long)adapter);
+	timer_setup(&adapter->phy_config_timer, atl2_phy_config, 0);
 
 	INIT_WORK(&adapter->reset_task, atl2_reset_task);
 	INIT_WORK(&adapter->link_chg_task, atl2_link_chg_task);
@@ -1491,8 +1469,8 @@ static void atl2_remove(struct pci_dev *pdev)
 	 * explicitly disable watchdog tasks from being rescheduled  */
 	set_bit(__ATL2_DOWN, &adapter->flags);
 
-	del_timer_sync(&adapter->watchdog_timer);
-	del_timer_sync(&adapter->phy_config_timer);
+	timer_delete_sync(&adapter->watchdog_timer);
+	timer_delete_sync(&adapter->phy_config_timer);
 	cancel_work_sync(&adapter->reset_task);
 	cancel_work_sync(&adapter->link_chg_task);
 
@@ -1698,32 +1676,7 @@ static struct pci_driver atl2_driver = {
 	.shutdown = atl2_shutdown,
 };
 
-/**
- * atl2_init_module - Driver Registration Routine
- *
- * atl2_init_module is the first routine called when the driver is
- * loaded. All it does is register with the PCI subsystem.
- */
-static int __init atl2_init_module(void)
-{
-	printk(KERN_INFO "%s - version %s\n", atl2_driver_string,
-		atl2_driver_version);
-	printk(KERN_INFO "%s\n", atl2_copyright);
-	return pci_register_driver(&atl2_driver);
-}
-module_init(atl2_init_module);
-
-/**
- * atl2_exit_module - Driver Exit Cleanup Routine
- *
- * atl2_exit_module is called just before the driver is removed
- * from memory.
- */
-static void __exit atl2_exit_module(void)
-{
-	pci_unregister_driver(&atl2_driver);
-}
-module_exit(atl2_exit_module);
+module_pci_driver(atl2_driver);
 
 static void atl2_read_pci_cfg(struct atl2_hw *hw, u32 reg, u16 *value)
 {
@@ -1942,8 +1895,8 @@ static int atl2_get_eeprom(struct net_device *netdev,
 	first_dword = eeprom->offset >> 2;
 	last_dword = (eeprom->offset + eeprom->len - 1) >> 2;
 
-	eeprom_buff = kmalloc(sizeof(u32) * (last_dword - first_dword + 1),
-		GFP_KERNEL);
+	eeprom_buff = kmalloc_array(last_dword - first_dword + 1, sizeof(u32),
+				    GFP_KERNEL);
 	if (!eeprom_buff)
 		return -ENOMEM;
 
@@ -2028,11 +1981,9 @@ static void atl2_get_drvinfo(struct net_device *netdev,
 {
 	struct atl2_adapter *adapter = netdev_priv(netdev);
 
-	strlcpy(drvinfo->driver,  atl2_driver_name, sizeof(drvinfo->driver));
-	strlcpy(drvinfo->version, atl2_driver_version,
-		sizeof(drvinfo->version));
-	strlcpy(drvinfo->fw_version, "L2", sizeof(drvinfo->fw_version));
-	strlcpy(drvinfo->bus_info, pci_name(adapter->pdev),
+	strscpy(drvinfo->driver,  atl2_driver_name, sizeof(drvinfo->driver));
+	strscpy(drvinfo->fw_version, "L2", sizeof(drvinfo->fw_version));
+	strscpy(drvinfo->bus_info, pci_name(adapter->pdev),
 		sizeof(drvinfo->bus_info));
 }
 
@@ -2577,7 +2528,6 @@ static s32 atl2_write_phy_reg(struct atl2_hw *hw, u32 reg_addr, u16 phy_data)
  */
 static s32 atl2_phy_setup_autoneg_adv(struct atl2_hw *hw)
 {
-	s32 ret_val;
 	s16 mii_autoneg_adv_reg;
 
 	/* Read the MII Auto-Neg Advertisement Register (Address 4). */
@@ -2633,12 +2583,7 @@ static s32 atl2_phy_setup_autoneg_adv(struct atl2_hw *hw)
 
 	hw->mii_autoneg_adv_reg = mii_autoneg_adv_reg;
 
-	ret_val = atl2_write_phy_reg(hw, MII_ADVERTISE, mii_autoneg_adv_reg);
-
-	if (ret_val)
-		return ret_val;
-
-	return 0;
+	return atl2_write_phy_reg(hw, MII_ADVERTISE, mii_autoneg_adv_reg);
 }
 
 /*
@@ -2947,7 +2892,7 @@ static int atl2_validate_option(int *value, struct atl2_option *opt)
 			if (*value == ent->i) {
 				if (ent->str[0] != '\0')
 					printk(KERN_INFO "%s\n", ent->str);
-			return 0;
+				return 0;
 			}
 		}
 		break;
diff --git a/drivers/net/ethernet/atheros/atlx/atl2.h b/drivers/net/ethernet/atheros/atlx/atl2.h
index c64a6bdfa7ae..d97613bd15d5 100644
--- a/drivers/net/ethernet/atheros/atlx/atl2.h
+++ b/drivers/net/ethernet/atheros/atlx/atl2.h
@@ -1,3 +1,4 @@
+/* SPDX-License-Identifier: GPL-2.0-or-later */
 /* atl2.h -- atl2 driver definitions
  *
  * Copyright(c) 2007 Atheros Corporation. All rights reserved.
@@ -6,20 +7,6 @@
  *
  * Derived from Intel e1000 driver
  * Copyright(c) 1999 - 2005 Intel Corporation. All rights reserved.
- *
- * This program is free software; you can redistribute it and/or modify it
- * under the terms of the GNU General Public License as published by the Free
- * Software Foundation; either version 2 of the License, or (at your option)
- * any later version.
- *
- * This program is distributed in the hope that it will be useful, but WITHOUT
- * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
- * FITNESS FOR A PARTICULAR PURPOSE.  See the GNU General Public License for
- * more details.
- *
- * You should have received a copy of the GNU General Public License along with
- * this program; if not, write to the Free Software Foundation, Inc., 59
- * Temple Place - Suite 330, Boston, MA  02111-1307, USA.
  */
 
 #ifndef _ATL2_H_
@@ -260,7 +247,7 @@ struct tx_pkt_status {
 	unsigned multi_col:1;
 	unsigned late_col:1;
 	unsigned abort_col:1;
-	unsigned underun:1;	/* current packet is aborted
+	unsigned underrun:1;	/* current packet is aborted
 				 * due to txram underrun */
 	unsigned:3;		/* reserved */
 	unsigned update:1;	/* always 1'b1 in tx_status_buf */
diff --git a/drivers/net/ethernet/atheros/atlx/atlx.c b/drivers/net/ethernet/atheros/atlx/atlx.c
index 46a622cceee4..e8cfbf4ff1b5 100644
--- a/drivers/net/ethernet/atheros/atlx/atlx.c
+++ b/drivers/net/ethernet/atheros/atlx/atlx.c
@@ -1,3 +1,4 @@
+// SPDX-License-Identifier: GPL-2.0-or-later
 /* atlx.c -- common functions for Attansic network drivers
  *
  * Copyright(c) 2005 - 2006 Attansic Corporation. All rights reserved.
@@ -7,20 +8,6 @@
  *
  * Derived from Intel e1000 driver
  * Copyright(c) 1999 - 2005 Intel Corporation. All rights reserved.
- *
- * This program is free software; you can redistribute it and/or modify it
- * under the terms of the GNU General Public License as published by the Free
- * Software Foundation; either version 2 of the License, or (at your option)
- * any later version.
- *
- * This program is distributed in the hope that it will be useful, but WITHOUT
- * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
- * FITNESS FOR A PARTICULAR PURPOSE.  See the GNU General Public License for
- * more details.
- *
- * You should have received a copy of the GNU General Public License along with
- * this program; if not, write to the Free Software Foundation, Inc., 59
- * Temple Place - Suite 330, Boston, MA  02111-1307, USA.
  */
 
 /* Including this file like a header is a temporary hack, I promise. -- CHS */
@@ -82,7 +69,7 @@ static int atlx_set_mac(struct net_device *netdev, void *p)
 	if (!is_valid_ether_addr(addr->sa_data))
 		return -EADDRNOTAVAIL;
 
-	memcpy(netdev->dev_addr, addr->sa_data, netdev->addr_len);
+	eth_hw_addr_set(netdev, addr->sa_data);
 	memcpy(adapter->hw.mac_addr, addr->sa_data, netdev->addr_len);
 
 	atlx_set_mac_addr(&adapter->hw);
@@ -196,7 +183,7 @@ static void atlx_clear_phy_int(struct atlx_adapter *adapter)
  * atlx_tx_timeout - Respond to a Tx Hang
  * @netdev: network interface device structure
  */
-static void atlx_tx_timeout(struct net_device *netdev)
+static void atlx_tx_timeout(struct net_device *netdev, unsigned int txqueue)
 {
 	struct atlx_adapter *adapter = netdev_priv(netdev);
 	/* Do the reset outside of interrupt context */
diff --git a/drivers/net/ethernet/atheros/atlx/atlx.h b/drivers/net/ethernet/atheros/atlx/atlx.h
index 448f5dcc02e6..7f5d4e24eb9f 100644
--- a/drivers/net/ethernet/atheros/atlx/atlx.h
+++ b/drivers/net/ethernet/atheros/atlx/atlx.h
@@ -1,3 +1,4 @@
+/* SPDX-License-Identifier: GPL-2.0-or-later */
 /* atlx_hw.h -- common hardware definitions for Attansic network drivers
  *
  * Copyright(c) 2005 - 2006 Attansic Corporation. All rights reserved.
@@ -7,20 +8,6 @@
  *
  * Derived from Intel e1000 driver
  * Copyright(c) 1999 - 2005 Intel Corporation. All rights reserved.
- *
- * This program is free software; you can redistribute it and/or modify it
- * under the terms of the GNU General Public License as published by the Free
- * Software Foundation; either version 2 of the License, or (at your option)
- * any later version.
- *
- * This program is distributed in the hope that it will be useful, but WITHOUT
- * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
- * FITNESS FOR A PARTICULAR PURPOSE.  See the GNU General Public License for
- * more details.
- *
- * You should have received a copy of the GNU General Public License along with
- * this program; if not, write to the Free Software Foundation, Inc., 59
- * Temple Place - Suite 330, Boston, MA  02111-1307, USA.
  */
 
 #ifndef ATLX_H
diff --git a/drivers/net/ethernet/aurora/Kconfig b/drivers/net/ethernet/aurora/Kconfig
deleted file mode 100644
index 8ba7f8ff3434..000000000000
--- a/drivers/net/ethernet/aurora/Kconfig
+++ /dev/null
@@ -1,21 +0,0 @@
-config NET_VENDOR_AURORA
-	bool "Aurora VLSI devices"
-	help
-	  If you have a network (Ethernet) device belonging to this class,
-	  say Y.
-
-	  Note that the answer to this question doesn't directly affect the
-	  kernel: saying N will just cause the configurator to skip all
-	  questions about Aurora devices. If you say Y, you will be asked
-	  for your specific device in the following questions.
-
-if NET_VENDOR_AURORA
-
-config AURORA_NB8800
-	tristate "Aurora AU-NB8800 support"
-	depends on HAS_DMA
-	select PHYLIB
-	help
-	 Support for the AU-NB8800 gigabit Ethernet controller.
-
-endif
diff --git a/drivers/net/ethernet/aurora/Makefile b/drivers/net/ethernet/aurora/Makefile
deleted file mode 100644
index 6cb528a2fc26..000000000000
--- a/drivers/net/ethernet/aurora/Makefile
+++ /dev/null
@@ -1 +0,0 @@
-obj-$(CONFIG_AURORA_NB8800) += nb8800.o
diff --git a/drivers/net/ethernet/aurora/nb8800.c b/drivers/net/ethernet/aurora/nb8800.c
deleted file mode 100644
index 5711fbbd6ae3..000000000000
--- a/drivers/net/ethernet/aurora/nb8800.c
+++ /dev/null
@@ -1,1545 +0,0 @@
-/*
- * Copyright (C) 2015 Mans Rullgard <mans@mansr.com>
- *
- * Mostly rewritten, based on driver from Sigma Designs.  Original
- * copyright notice below.
- *
- *
- * Driver for tangox SMP864x/SMP865x/SMP867x/SMP868x builtin Ethernet Mac.
- *
- * Copyright (C) 2005 Maxime Bizon <mbizon@freebox.fr>
- *
- * This program is free software; you can redistribute it and/or modify
- * it under the terms of the GNU General Public License as published by
- * the Free Software Foundation; either version 2 of the License, or
- * (at your option) any later version.
- *
- * This program is distributed in the hope that it will be useful,
- * but WITHOUT ANY WARRANTY; without even the implied warranty of
- * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
- * GNU General Public License for more details.
- */
-
-#include <linux/module.h>
-#include <linux/etherdevice.h>
-#include <linux/delay.h>
-#include <linux/ethtool.h>
-#include <linux/interrupt.h>
-#include <linux/platform_device.h>
-#include <linux/of_device.h>
-#include <linux/of_mdio.h>
-#include <linux/of_net.h>
-#include <linux/dma-mapping.h>
-#include <linux/phy.h>
-#include <linux/cache.h>
-#include <linux/jiffies.h>
-#include <linux/io.h>
-#include <linux/iopoll.h>
-#include <asm/barrier.h>
-
-#include "nb8800.h"
-
-static void nb8800_tx_done(struct net_device *dev);
-static int nb8800_dma_stop(struct net_device *dev);
-
-static inline u8 nb8800_readb(struct nb8800_priv *priv, int reg)
-{
-	return readb_relaxed(priv->base + reg);
-}
-
-static inline u32 nb8800_readl(struct nb8800_priv *priv, int reg)
-{
-	return readl_relaxed(priv->base + reg);
-}
-
-static inline void nb8800_writeb(struct nb8800_priv *priv, int reg, u8 val)
-{
-	writeb_relaxed(val, priv->base + reg);
-}
-
-static inline void nb8800_writew(struct nb8800_priv *priv, int reg, u16 val)
-{
-	writew_relaxed(val, priv->base + reg);
-}
-
-static inline void nb8800_writel(struct nb8800_priv *priv, int reg, u32 val)
-{
-	writel_relaxed(val, priv->base + reg);
-}
-
-static inline void nb8800_maskb(struct nb8800_priv *priv, int reg,
-				u32 mask, u32 val)
-{
-	u32 old = nb8800_readb(priv, reg);
-	u32 new = (old & ~mask) | (val & mask);
-
-	if (new != old)
-		nb8800_writeb(priv, reg, new);
-}
-
-static inline void nb8800_maskl(struct nb8800_priv *priv, int reg,
-				u32 mask, u32 val)
-{
-	u32 old = nb8800_readl(priv, reg);
-	u32 new = (old & ~mask) | (val & mask);
-
-	if (new != old)
-		nb8800_writel(priv, reg, new);
-}
-
-static inline void nb8800_modb(struct nb8800_priv *priv, int reg, u8 bits,
-			       bool set)
-{
-	nb8800_maskb(priv, reg, bits, set ? bits : 0);
-}
-
-static inline void nb8800_setb(struct nb8800_priv *priv, int reg, u8 bits)
-{
-	nb8800_maskb(priv, reg, bits, bits);
-}
-
-static inline void nb8800_clearb(struct nb8800_priv *priv, int reg, u8 bits)
-{
-	nb8800_maskb(priv, reg, bits, 0);
-}
-
-static inline void nb8800_modl(struct nb8800_priv *priv, int reg, u32 bits,
-			       bool set)
-{
-	nb8800_maskl(priv, reg, bits, set ? bits : 0);
-}
-
-static inline void nb8800_setl(struct nb8800_priv *priv, int reg, u32 bits)
-{
-	nb8800_maskl(priv, reg, bits, bits);
-}
-
-static inline void nb8800_clearl(struct nb8800_priv *priv, int reg, u32 bits)
-{
-	nb8800_maskl(priv, reg, bits, 0);
-}
-
-static int nb8800_mdio_wait(struct mii_bus *bus)
-{
-	struct nb8800_priv *priv = bus->priv;
-	u32 val;
-
-	return readl_poll_timeout_atomic(priv->base + NB8800_MDIO_CMD,
-					 val, !(val & MDIO_CMD_GO), 1, 1000);
-}
-
-static int nb8800_mdio_cmd(struct mii_bus *bus, u32 cmd)
-{
-	struct nb8800_priv *priv = bus->priv;
-	int err;
-
-	err = nb8800_mdio_wait(bus);
-	if (err)
-		return err;
-
-	nb8800_writel(priv, NB8800_MDIO_CMD, cmd);
-	udelay(10);
-	nb8800_writel(priv, NB8800_MDIO_CMD, cmd | MDIO_CMD_GO);
-
-	return nb8800_mdio_wait(bus);
-}
-
-static int nb8800_mdio_read(struct mii_bus *bus, int phy_id, int reg)
-{
-	struct nb8800_priv *priv = bus->priv;
-	u32 val;
-	int err;
-
-	err = nb8800_mdio_cmd(bus, MDIO_CMD_ADDR(phy_id) | MDIO_CMD_REG(reg));
-	if (err)
-		return err;
-
-	val = nb8800_readl(priv, NB8800_MDIO_STS);
-	if (val & MDIO_STS_ERR)
-		return 0xffff;
-
-	return val & 0xffff;
-}
-
-static int nb8800_mdio_write(struct mii_bus *bus, int phy_id, int reg, u16 val)
-{
-	u32 cmd = MDIO_CMD_ADDR(phy_id) | MDIO_CMD_REG(reg) |
-		MDIO_CMD_DATA(val) | MDIO_CMD_WR;
-
-	return nb8800_mdio_cmd(bus, cmd);
-}
-
-static void nb8800_mac_tx(struct net_device *dev, bool enable)
-{
-	struct nb8800_priv *priv = netdev_priv(dev);
-
-	while (nb8800_readl(priv, NB8800_TXC_CR) & TCR_EN)
-		cpu_relax();
-
-	nb8800_modb(priv, NB8800_TX_CTL1, TX_EN, enable);
-}
-
-static void nb8800_mac_rx(struct net_device *dev, bool enable)
-{
-	nb8800_modb(netdev_priv(dev), NB8800_RX_CTL, RX_EN, enable);
-}
-
-static void nb8800_mac_af(struct net_device *dev, bool enable)
-{
-	nb8800_modb(netdev_priv(dev), NB8800_RX_CTL, RX_AF_EN, enable);
-}
-
-static void nb8800_start_rx(struct net_device *dev)
-{
-	nb8800_setl(netdev_priv(dev), NB8800_RXC_CR, RCR_EN);
-}
-
-static int nb8800_alloc_rx(struct net_device *dev, unsigned int i, bool napi)
-{
-	struct nb8800_priv *priv = netdev_priv(dev);
-	struct nb8800_rx_desc *rxd = &priv->rx_descs[i];
-	struct nb8800_rx_buf *rxb = &priv->rx_bufs[i];
-	int size = L1_CACHE_ALIGN(RX_BUF_SIZE);
-	dma_addr_t dma_addr;
-	struct page *page;
-	unsigned long offset;
-	void *data;
-
-	data = napi ? napi_alloc_frag(size) : netdev_alloc_frag(size);
-	if (!data)
-		return -ENOMEM;
-
-	page = virt_to_head_page(data);
-	offset = data - page_address(page);
-
-	dma_addr = dma_map_page(&dev->dev, page, offset, RX_BUF_SIZE,
-				DMA_FROM_DEVICE);
-
-	if (dma_mapping_error(&dev->dev, dma_addr)) {
-		skb_free_frag(data);
-		return -ENOMEM;
-	}
-
-	rxb->page = page;
-	rxb->offset = offset;
-	rxd->desc.s_addr = dma_addr;
-
-	return 0;
-}
-
-static void nb8800_receive(struct net_device *dev, unsigned int i,
-			   unsigned int len)
-{
-	struct nb8800_priv *priv = netdev_priv(dev);
-	struct nb8800_rx_desc *rxd = &priv->rx_descs[i];
-	struct page *page = priv->rx_bufs[i].page;
-	int offset = priv->rx_bufs[i].offset;
-	void *data = page_address(page) + offset;
-	dma_addr_t dma = rxd->desc.s_addr;
-	struct sk_buff *skb;
-	unsigned int size;
-	int err;
-
-	size = len <= RX_COPYBREAK ? len : RX_COPYHDR;
-
-	skb = napi_alloc_skb(&priv->napi, size);
-	if (!skb) {
-		netdev_err(dev, "rx skb allocation failed\n");
-		dev->stats.rx_dropped++;
-		return;
-	}
-
-	if (len <= RX_COPYBREAK) {
-		dma_sync_single_for_cpu(&dev->dev, dma, len, DMA_FROM_DEVICE);
-		memcpy(skb_put(skb, len), data, len);
-		dma_sync_single_for_device(&dev->dev, dma, len,
-					   DMA_FROM_DEVICE);
-	} else {
-		err = nb8800_alloc_rx(dev, i, true);
-		if (err) {
-			netdev_err(dev, "rx buffer allocation failed\n");
-			dev->stats.rx_dropped++;
-			dev_kfree_skb(skb);
-			return;
-		}
-
-		dma_unmap_page(&dev->dev, dma, RX_BUF_SIZE, DMA_FROM_DEVICE);
-		memcpy(skb_put(skb, RX_COPYHDR), data, RX_COPYHDR);
-		skb_add_rx_frag(skb, skb_shinfo(skb)->nr_frags, page,
-				offset + RX_COPYHDR, len - RX_COPYHDR,
-				RX_BUF_SIZE);
-	}
-
-	skb->protocol = eth_type_trans(skb, dev);
-	napi_gro_receive(&priv->napi, skb);
-}
-
-static void nb8800_rx_error(struct net_device *dev, u32 report)
-{
-	if (report & RX_LENGTH_ERR)
-		dev->stats.rx_length_errors++;
-
-	if (report & RX_FCS_ERR)
-		dev->stats.rx_crc_errors++;
-
-	if (report & RX_FIFO_OVERRUN)
-		dev->stats.rx_fifo_errors++;
-
-	if (report & RX_ALIGNMENT_ERROR)
-		dev->stats.rx_frame_errors++;
-
-	dev->stats.rx_errors++;
-}
-
-static int nb8800_poll(struct napi_struct *napi, int budget)
-{
-	struct net_device *dev = napi->dev;
-	struct nb8800_priv *priv = netdev_priv(dev);
-	struct nb8800_rx_desc *rxd;
-	unsigned int last = priv->rx_eoc;
-	unsigned int next;
-	int work = 0;
-
-	nb8800_tx_done(dev);
-
-again:
-	do {
-		struct nb8800_rx_buf *rxb;
-		unsigned int len;
-
-		next = (last + 1) % RX_DESC_COUNT;
-
-		rxb = &priv->rx_bufs[next];
-		rxd = &priv->rx_descs[next];
-
-		if (!rxd->report)
-			break;
-
-		len = RX_BYTES_TRANSFERRED(rxd->report);
-
-		if (IS_RX_ERROR(rxd->report))
-			nb8800_rx_error(dev, rxd->report);
-		else
-			nb8800_receive(dev, next, len);
-
-		dev->stats.rx_packets++;
-		dev->stats.rx_bytes += len;
-
-		if (rxd->report & RX_MULTICAST_PKT)
-			dev->stats.multicast++;
-
-		rxd->report = 0;
-		last = next;
-		work++;
-	} while (work < budget);
-
-	if (work) {
-		priv->rx_descs[last].desc.config |= DESC_EOC;
-		wmb();	/* ensure new EOC is written before clearing old */
-		priv->rx_descs[priv->rx_eoc].desc.config &= ~DESC_EOC;
-		priv->rx_eoc = last;
-		nb8800_start_rx(dev);
-	}
-
-	if (work < budget) {
-		nb8800_writel(priv, NB8800_RX_ITR, priv->rx_itr_irq);
-
-		/* If a packet arrived after we last checked but
-		 * before writing RX_ITR, the interrupt will be
-		 * delayed, so we retrieve it now.
-		 */
-		if (priv->rx_descs[next].report)
-			goto again;
-
-		napi_complete_done(napi, work);
-	}
-
-	return work;
-}
-
-static void __nb8800_tx_dma_start(struct net_device *dev)
-{
-	struct nb8800_priv *priv = netdev_priv(dev);
-	struct nb8800_tx_buf *txb;
-	u32 txc_cr;
-
-	txb = &priv->tx_bufs[priv->tx_queue];
-	if (!txb->ready)
-		return;
-
-	txc_cr = nb8800_readl(priv, NB8800_TXC_CR);
-	if (txc_cr & TCR_EN)
-		return;
-
-	nb8800_writel(priv, NB8800_TX_DESC_ADDR, txb->dma_desc);
-	wmb();		/* ensure desc addr is written before starting DMA */
-	nb8800_writel(priv, NB8800_TXC_CR, txc_cr | TCR_EN);
-
-	priv->tx_queue = (priv->tx_queue + txb->chain_len) % TX_DESC_COUNT;
-}
-
-static void nb8800_tx_dma_start(struct net_device *dev)
-{
-	struct nb8800_priv *priv = netdev_priv(dev);
-
-	spin_lock_irq(&priv->tx_lock);
-	__nb8800_tx_dma_start(dev);
-	spin_unlock_irq(&priv->tx_lock);
-}
-
-static void nb8800_tx_dma_start_irq(struct net_device *dev)
-{
-	struct nb8800_priv *priv = netdev_priv(dev);
-
-	spin_lock(&priv->tx_lock);
-	__nb8800_tx_dma_start(dev);
-	spin_unlock(&priv->tx_lock);
-}
-
-static int nb8800_xmit(struct sk_buff *skb, struct net_device *dev)
-{
-	struct nb8800_priv *priv = netdev_priv(dev);
-	struct nb8800_tx_desc *txd;
-	struct nb8800_tx_buf *txb;
-	struct nb8800_dma_desc *desc;
-	dma_addr_t dma_addr;
-	unsigned int dma_len;
-	unsigned int align;
-	unsigned int next;
-
-	if (atomic_read(&priv->tx_free) <= NB8800_DESC_LOW) {
-		netif_stop_queue(dev);
-		return NETDEV_TX_BUSY;
-	}
-
-	align = (8 - (uintptr_t)skb->data) & 7;
-
-	dma_len = skb->len - align;
-	dma_addr = dma_map_single(&dev->dev, skb->data + align,
-				  dma_len, DMA_TO_DEVICE);
-
-	if (dma_mapping_error(&dev->dev, dma_addr)) {
-		netdev_err(dev, "tx dma mapping error\n");
-		kfree_skb(skb);
-		dev->stats.tx_dropped++;
-		return NETDEV_TX_OK;
-	}
-
-	if (atomic_dec_return(&priv->tx_free) <= NB8800_DESC_LOW) {
-		netif_stop_queue(dev);
-		skb->xmit_more = 0;
-	}
-
-	next = priv->tx_next;
-	txb = &priv->tx_bufs[next];
-	txd = &priv->tx_descs[next];
-	desc = &txd->desc[0];
-
-	next = (next + 1) % TX_DESC_COUNT;
-
-	if (align) {
-		memcpy(txd->buf, skb->data, align);
-
-		desc->s_addr =
-			txb->dma_desc + offsetof(struct nb8800_tx_desc, buf);
-		desc->n_addr = txb->dma_desc + sizeof(txd->desc[0]);
-		desc->config = DESC_BTS(2) | DESC_DS | align;
-
-		desc++;
-	}
-
-	desc->s_addr = dma_addr;
-	desc->n_addr = priv->tx_bufs[next].dma_desc;
-	desc->config = DESC_BTS(2) | DESC_DS | DESC_EOF | dma_len;
-
-	if (!skb->xmit_more)
-		desc->config |= DESC_EOC;
-
-	txb->skb = skb;
-	txb->dma_addr = dma_addr;
-	txb->dma_len = dma_len;
-
-	if (!priv->tx_chain) {
-		txb->chain_len = 1;
-		priv->tx_chain = txb;
-	} else {
-		priv->tx_chain->chain_len++;
-	}
-
-	netdev_sent_queue(dev, skb->len);
-
-	priv->tx_next = next;
-
-	if (!skb->xmit_more) {
-		smp_wmb();
-		priv->tx_chain->ready = true;
-		priv->tx_chain = NULL;
-		nb8800_tx_dma_start(dev);
-	}
-
-	return NETDEV_TX_OK;
-}
-
-static void nb8800_tx_error(struct net_device *dev, u32 report)
-{
-	if (report & TX_LATE_COLLISION)
-		dev->stats.collisions++;
-
-	if (report & TX_PACKET_DROPPED)
-		dev->stats.tx_dropped++;
-
-	if (report & TX_FIFO_UNDERRUN)
-		dev->stats.tx_fifo_errors++;
-
-	dev->stats.tx_errors++;
-}
-
-static void nb8800_tx_done(struct net_device *dev)
-{
-	struct nb8800_priv *priv = netdev_priv(dev);
-	unsigned int limit = priv->tx_next;
-	unsigned int done = priv->tx_done;
-	unsigned int packets = 0;
-	unsigned int len = 0;
-
-	while (done != limit) {
-		struct nb8800_tx_desc *txd = &priv->tx_descs[done];
-		struct nb8800_tx_buf *txb = &priv->tx_bufs[done];
-		struct sk_buff *skb;
-
-		if (!txd->report)
-			break;
-
-		skb = txb->skb;
-		len += skb->len;
-
-		dma_unmap_single(&dev->dev, txb->dma_addr, txb->dma_len,
-				 DMA_TO_DEVICE);
-
-		if (IS_TX_ERROR(txd->report)) {
-			nb8800_tx_error(dev, txd->report);
-			kfree_skb(skb);
-		} else {
-			consume_skb(skb);
-		}
-
-		dev->stats.tx_packets++;
-		dev->stats.tx_bytes += TX_BYTES_TRANSFERRED(txd->report);
-		dev->stats.collisions += TX_EARLY_COLLISIONS(txd->report);
-
-		txb->skb = NULL;
-		txb->ready = false;
-		txd->report = 0;
-
-		done = (done + 1) % TX_DESC_COUNT;
-		packets++;
-	}
-
-	if (packets) {
-		smp_mb__before_atomic();
-		atomic_add(packets, &priv->tx_free);
-		netdev_completed_queue(dev, packets, len);
-		netif_wake_queue(dev);
-		priv->tx_done = done;
-	}
-}
-
-static irqreturn_t nb8800_irq(int irq, void *dev_id)
-{
-	struct net_device *dev = dev_id;
-	struct nb8800_priv *priv = netdev_priv(dev);
-	irqreturn_t ret = IRQ_NONE;
-	u32 val;
-
-	/* tx interrupt */
-	val = nb8800_readl(priv, NB8800_TXC_SR);
-	if (val) {
-		nb8800_writel(priv, NB8800_TXC_SR, val);
-
-		if (val & TSR_DI)
-			nb8800_tx_dma_start_irq(dev);
-
-		if (val & TSR_TI)
-			napi_schedule_irqoff(&priv->napi);
-
-		if (unlikely(val & TSR_DE))
-			netdev_err(dev, "TX DMA error\n");
-
-		/* should never happen with automatic status retrieval */
-		if (unlikely(val & TSR_TO))
-			netdev_err(dev, "TX Status FIFO overflow\n");
-
-		ret = IRQ_HANDLED;
-	}
-
-	/* rx interrupt */
-	val = nb8800_readl(priv, NB8800_RXC_SR);
-	if (val) {
-		nb8800_writel(priv, NB8800_RXC_SR, val);
-
-		if (likely(val & (RSR_RI | RSR_DI))) {
-			nb8800_writel(priv, NB8800_RX_ITR, priv->rx_itr_poll);
-			napi_schedule_irqoff(&priv->napi);
-		}
-
-		if (unlikely(val & RSR_DE))
-			netdev_err(dev, "RX DMA error\n");
-
-		/* should never happen with automatic status retrieval */
-		if (unlikely(val & RSR_RO))
-			netdev_err(dev, "RX Status FIFO overflow\n");
-
-		ret = IRQ_HANDLED;
-	}
-
-	return ret;
-}
-
-static void nb8800_mac_config(struct net_device *dev)
-{
-	struct nb8800_priv *priv = netdev_priv(dev);
-	bool gigabit = priv->speed == SPEED_1000;
-	u32 mac_mode_mask = RGMII_MODE | HALF_DUPLEX | GMAC_MODE;
-	u32 mac_mode = 0;
-	u32 slot_time;
-	u32 phy_clk;
-	u32 ict;
-
-	if (!priv->duplex)
-		mac_mode |= HALF_DUPLEX;
-
-	if (gigabit) {
-		if (priv->phy_mode == PHY_INTERFACE_MODE_RGMII)
-			mac_mode |= RGMII_MODE;
-
-		mac_mode |= GMAC_MODE;
-		phy_clk = 125000000;
-
-		/* Should be 512 but register is only 8 bits */
-		slot_time = 255;
-	} else {
-		phy_clk = 25000000;
-		slot_time = 128;
-	}
-
-	ict = DIV_ROUND_UP(phy_clk, clk_get_rate(priv->clk));
-
-	nb8800_writeb(priv, NB8800_IC_THRESHOLD, ict);
-	nb8800_writeb(priv, NB8800_SLOT_TIME, slot_time);
-	nb8800_maskb(priv, NB8800_MAC_MODE, mac_mode_mask, mac_mode);
-}
-
-static void nb8800_pause_config(struct net_device *dev)
-{
-	struct nb8800_priv *priv = netdev_priv(dev);
-	struct phy_device *phydev = dev->phydev;
-	u32 rxcr;
-
-	if (priv->pause_aneg) {
-		if (!phydev || !phydev->link)
-			return;
-
-		priv->pause_rx = phydev->pause;
-		priv->pause_tx = phydev->pause ^ phydev->asym_pause;
-	}
-
-	nb8800_modb(priv, NB8800_RX_CTL, RX_PAUSE_EN, priv->pause_rx);
-
-	rxcr = nb8800_readl(priv, NB8800_RXC_CR);
-	if (!!(rxcr & RCR_FL) == priv->pause_tx)
-		return;
-
-	if (netif_running(dev)) {
-		napi_disable(&priv->napi);
-		netif_tx_lock_bh(dev);
-		nb8800_dma_stop(dev);
-		nb8800_modl(priv, NB8800_RXC_CR, RCR_FL, priv->pause_tx);
-		nb8800_start_rx(dev);
-		netif_tx_unlock_bh(dev);
-		napi_enable(&priv->napi);
-	} else {
-		nb8800_modl(priv, NB8800_RXC_CR, RCR_FL, priv->pause_tx);
-	}
-}
-
-static void nb8800_link_reconfigure(struct net_device *dev)
-{
-	struct nb8800_priv *priv = netdev_priv(dev);
-	struct phy_device *phydev = dev->phydev;
-	int change = 0;
-
-	if (phydev->link) {
-		if (phydev->speed != priv->speed) {
-			priv->speed = phydev->speed;
-			change = 1;
-		}
-
-		if (phydev->duplex != priv->duplex) {
-			priv->duplex = phydev->duplex;
-			change = 1;
-		}
-
-		if (change)
-			nb8800_mac_config(dev);
-
-		nb8800_pause_config(dev);
-	}
-
-	if (phydev->link != priv->link) {
-		priv->link = phydev->link;
-		change = 1;
-	}
-
-	if (change)
-		phy_print_status(phydev);
-}
-
-static void nb8800_update_mac_addr(struct net_device *dev)
-{
-	struct nb8800_priv *priv = netdev_priv(dev);
-	int i;
-
-	for (i = 0; i < ETH_ALEN; i++)
-		nb8800_writeb(priv, NB8800_SRC_ADDR(i), dev->dev_addr[i]);
-
-	for (i = 0; i < ETH_ALEN; i++)
-		nb8800_writeb(priv, NB8800_UC_ADDR(i), dev->dev_addr[i]);
-}
-
-static int nb8800_set_mac_address(struct net_device *dev, void *addr)
-{
-	struct sockaddr *sock = addr;
-
-	if (netif_running(dev))
-		return -EBUSY;
-
-	ether_addr_copy(dev->dev_addr, sock->sa_data);
-	nb8800_update_mac_addr(dev);
-
-	return 0;
-}
-
-static void nb8800_mc_init(struct net_device *dev, int val)
-{
-	struct nb8800_priv *priv = netdev_priv(dev);
-
-	nb8800_writeb(priv, NB8800_MC_INIT, val);
-	readb_poll_timeout_atomic(priv->base + NB8800_MC_INIT, val, !val,
-				  1, 1000);
-}
-
-static void nb8800_set_rx_mode(struct net_device *dev)
-{
-	struct nb8800_priv *priv = netdev_priv(dev);
-	struct netdev_hw_addr *ha;
-	int i;
-
-	if (dev->flags & (IFF_PROMISC | IFF_ALLMULTI)) {
-		nb8800_mac_af(dev, false);
-		return;
-	}
-
-	nb8800_mac_af(dev, true);
-	nb8800_mc_init(dev, 0);
-
-	netdev_for_each_mc_addr(ha, dev) {
-		for (i = 0; i < ETH_ALEN; i++)
-			nb8800_writeb(priv, NB8800_MC_ADDR(i), ha->addr[i]);
-
-		nb8800_mc_init(dev, 0xff);
-	}
-}
-
-#define RX_DESC_SIZE (RX_DESC_COUNT * sizeof(struct nb8800_rx_desc))
-#define TX_DESC_SIZE (TX_DESC_COUNT * sizeof(struct nb8800_tx_desc))
-
-static void nb8800_dma_free(struct net_device *dev)
-{
-	struct nb8800_priv *priv = netdev_priv(dev);
-	unsigned int i;
-
-	if (priv->rx_bufs) {
-		for (i = 0; i < RX_DESC_COUNT; i++)
-			if (priv->rx_bufs[i].page)
-				put_page(priv->rx_bufs[i].page);
-
-		kfree(priv->rx_bufs);
-		priv->rx_bufs = NULL;
-	}
-
-	if (priv->tx_bufs) {
-		for (i = 0; i < TX_DESC_COUNT; i++)
-			kfree_skb(priv->tx_bufs[i].skb);
-
-		kfree(priv->tx_bufs);
-		priv->tx_bufs = NULL;
-	}
-
-	if (priv->rx_descs) {
-		dma_free_coherent(dev->dev.parent, RX_DESC_SIZE, priv->rx_descs,
-				  priv->rx_desc_dma);
-		priv->rx_descs = NULL;
-	}
-
-	if (priv->tx_descs) {
-		dma_free_coherent(dev->dev.parent, TX_DESC_SIZE, priv->tx_descs,
-				  priv->tx_desc_dma);
-		priv->tx_descs = NULL;
-	}
-}
-
-static void nb8800_dma_reset(struct net_device *dev)
-{
-	struct nb8800_priv *priv = netdev_priv(dev);
-	struct nb8800_rx_desc *rxd;
-	struct nb8800_tx_desc *txd;
-	unsigned int i;
-
-	for (i = 0; i < RX_DESC_COUNT; i++) {
-		dma_addr_t rx_dma = priv->rx_desc_dma + i * sizeof(*rxd);
-
-		rxd = &priv->rx_descs[i];
-		rxd->desc.n_addr = rx_dma + sizeof(*rxd);
-		rxd->desc.r_addr =
-			rx_dma + offsetof(struct nb8800_rx_desc, report);
-		rxd->desc.config = priv->rx_dma_config;
-		rxd->report = 0;
-	}
-
-	rxd->desc.n_addr = priv->rx_desc_dma;
-	rxd->desc.config |= DESC_EOC;
-
-	priv->rx_eoc = RX_DESC_COUNT - 1;
-
-	for (i = 0; i < TX_DESC_COUNT; i++) {
-		struct nb8800_tx_buf *txb = &priv->tx_bufs[i];
-		dma_addr_t r_dma = txb->dma_desc +
-			offsetof(struct nb8800_tx_desc, report);
-
-		txd = &priv->tx_descs[i];
-		txd->desc[0].r_addr = r_dma;
-		txd->desc[1].r_addr = r_dma;
-		txd->report = 0;
-	}
-
-	priv->tx_next = 0;
-	priv->tx_queue = 0;
-	priv->tx_done = 0;
-	atomic_set(&priv->tx_free, TX_DESC_COUNT);
-
-	nb8800_writel(priv, NB8800_RX_DESC_ADDR, priv->rx_desc_dma);
-
-	wmb();		/* ensure all setup is written before starting */
-}
-
-static int nb8800_dma_init(struct net_device *dev)
-{
-	struct nb8800_priv *priv = netdev_priv(dev);
-	unsigned int n_rx = RX_DESC_COUNT;
-	unsigned int n_tx = TX_DESC_COUNT;
-	unsigned int i;
-	int err;
-
-	priv->rx_descs = dma_alloc_coherent(dev->dev.parent, RX_DESC_SIZE,
-					    &priv->rx_desc_dma, GFP_KERNEL);
-	if (!priv->rx_descs)
-		goto err_out;
-
-	priv->rx_bufs = kcalloc(n_rx, sizeof(*priv->rx_bufs), GFP_KERNEL);
-	if (!priv->rx_bufs)
-		goto err_out;
-
-	for (i = 0; i < n_rx; i++) {
-		err = nb8800_alloc_rx(dev, i, false);
-		if (err)
-			goto err_out;
-	}
-
-	priv->tx_descs = dma_alloc_coherent(dev->dev.parent, TX_DESC_SIZE,
-					    &priv->tx_desc_dma, GFP_KERNEL);
-	if (!priv->tx_descs)
-		goto err_out;
-
-	priv->tx_bufs = kcalloc(n_tx, sizeof(*priv->tx_bufs), GFP_KERNEL);
-	if (!priv->tx_bufs)
-		goto err_out;
-
-	for (i = 0; i < n_tx; i++)
-		priv->tx_bufs[i].dma_desc =
-			priv->tx_desc_dma + i * sizeof(struct nb8800_tx_desc);
-
-	nb8800_dma_reset(dev);
-
-	return 0;
-
-err_out:
-	nb8800_dma_free(dev);
-
-	return -ENOMEM;
-}
-
-static int nb8800_dma_stop(struct net_device *dev)
-{
-	struct nb8800_priv *priv = netdev_priv(dev);
-	struct nb8800_tx_buf *txb = &priv->tx_bufs[0];
-	struct nb8800_tx_desc *txd = &priv->tx_descs[0];
-	int retry = 5;
-	u32 txcr;
-	u32 rxcr;
-	int err;
-	unsigned int i;
-
-	/* wait for tx to finish */
-	err = readl_poll_timeout_atomic(priv->base + NB8800_TXC_CR, txcr,
-					!(txcr & TCR_EN) &&
-					priv->tx_done == priv->tx_next,
-					1000, 1000000);
-	if (err)
-		return err;
-
-	/* The rx DMA only stops if it reaches the end of chain.
-	 * To make this happen, we set the EOC flag on all rx
-	 * descriptors, put the device in loopback mode, and send
-	 * a few dummy frames.  The interrupt handler will ignore
-	 * these since NAPI is disabled and no real frames are in
-	 * the tx queue.
-	 */
-
-	for (i = 0; i < RX_DESC_COUNT; i++)
-		priv->rx_descs[i].desc.config |= DESC_EOC;
-
-	txd->desc[0].s_addr =
-		txb->dma_desc + offsetof(struct nb8800_tx_desc, buf);
-	txd->desc[0].config = DESC_BTS(2) | DESC_DS | DESC_EOF | DESC_EOC | 8;
-	memset(txd->buf, 0, sizeof(txd->buf));
-
-	nb8800_mac_af(dev, false);
-	nb8800_setb(priv, NB8800_MAC_MODE, LOOPBACK_EN);
-
-	do {
-		nb8800_writel(priv, NB8800_TX_DESC_ADDR, txb->dma_desc);
-		wmb();
-		nb8800_writel(priv, NB8800_TXC_CR, txcr | TCR_EN);
-
-		err = readl_poll_timeout_atomic(priv->base + NB8800_RXC_CR,
-						rxcr, !(rxcr & RCR_EN),
-						1000, 100000);
-	} while (err && --retry);
-
-	nb8800_mac_af(dev, true);
-	nb8800_clearb(priv, NB8800_MAC_MODE, LOOPBACK_EN);
-	nb8800_dma_reset(dev);
-
-	return retry ? 0 : -ETIMEDOUT;
-}
-
-static void nb8800_pause_adv(struct net_device *dev)
-{
-	struct nb8800_priv *priv = netdev_priv(dev);
-	struct phy_device *phydev = dev->phydev;
-	u32 adv = 0;
-
-	if (!phydev)
-		return;
-
-	if (priv->pause_rx)
-		adv |= ADVERTISED_Pause | ADVERTISED_Asym_Pause;
-	if (priv->pause_tx)
-		adv ^= ADVERTISED_Asym_Pause;
-
-	phydev->supported |= adv;
-	phydev->advertising |= adv;
-}
-
-static int nb8800_open(struct net_device *dev)
-{
-	struct nb8800_priv *priv = netdev_priv(dev);
-	struct phy_device *phydev;
-	int err;
-
-	/* clear any pending interrupts */
-	nb8800_writel(priv, NB8800_RXC_SR, 0xf);
-	nb8800_writel(priv, NB8800_TXC_SR, 0xf);
-
-	err = nb8800_dma_init(dev);
-	if (err)
-		return err;
-
-	err = request_irq(dev->irq, nb8800_irq, 0, dev_name(&dev->dev), dev);
-	if (err)
-		goto err_free_dma;
-
-	nb8800_mac_rx(dev, true);
-	nb8800_mac_tx(dev, true);
-
-	phydev = of_phy_connect(dev, priv->phy_node,
-				nb8800_link_reconfigure, 0,
-				priv->phy_mode);
-	if (!phydev) {
-		err = -ENODEV;
-		goto err_free_irq;
-	}
-
-	nb8800_pause_adv(dev);
-
-	netdev_reset_queue(dev);
-	napi_enable(&priv->napi);
-	netif_start_queue(dev);
-
-	nb8800_start_rx(dev);
-	phy_start(phydev);
-
-	return 0;
-
-err_free_irq:
-	free_irq(dev->irq, dev);
-err_free_dma:
-	nb8800_dma_free(dev);
-
-	return err;
-}
-
-static int nb8800_stop(struct net_device *dev)
-{
-	struct nb8800_priv *priv = netdev_priv(dev);
-	struct phy_device *phydev = dev->phydev;
-
-	phy_stop(phydev);
-
-	netif_stop_queue(dev);
-	napi_disable(&priv->napi);
-
-	nb8800_dma_stop(dev);
-	nb8800_mac_rx(dev, false);
-	nb8800_mac_tx(dev, false);
-
-	phy_disconnect(phydev);
-
-	free_irq(dev->irq, dev);
-
-	nb8800_dma_free(dev);
-
-	return 0;
-}
-
-static int nb8800_ioctl(struct net_device *dev, struct ifreq *rq, int cmd)
-{
-	return phy_mii_ioctl(dev->phydev, rq, cmd);
-}
-
-static const struct net_device_ops nb8800_netdev_ops = {
-	.ndo_open		= nb8800_open,
-	.ndo_stop		= nb8800_stop,
-	.ndo_start_xmit		= nb8800_xmit,
-	.ndo_set_mac_address	= nb8800_set_mac_address,
-	.ndo_set_rx_mode	= nb8800_set_rx_mode,
-	.ndo_do_ioctl		= nb8800_ioctl,
-	.ndo_validate_addr	= eth_validate_addr,
-};
-
-static void nb8800_get_pauseparam(struct net_device *dev,
-				  struct ethtool_pauseparam *pp)
-{
-	struct nb8800_priv *priv = netdev_priv(dev);
-
-	pp->autoneg = priv->pause_aneg;
-	pp->rx_pause = priv->pause_rx;
-	pp->tx_pause = priv->pause_tx;
-}
-
-static int nb8800_set_pauseparam(struct net_device *dev,
-				 struct ethtool_pauseparam *pp)
-{
-	struct nb8800_priv *priv = netdev_priv(dev);
-	struct phy_device *phydev = dev->phydev;
-
-	priv->pause_aneg = pp->autoneg;
-	priv->pause_rx = pp->rx_pause;
-	priv->pause_tx = pp->tx_pause;
-
-	nb8800_pause_adv(dev);
-
-	if (!priv->pause_aneg)
-		nb8800_pause_config(dev);
-	else if (phydev)
-		phy_start_aneg(phydev);
-
-	return 0;
-}
-
-static const char nb8800_stats_names[][ETH_GSTRING_LEN] = {
-	"rx_bytes_ok",
-	"rx_frames_ok",
-	"rx_undersize_frames",
-	"rx_fragment_frames",
-	"rx_64_byte_frames",
-	"rx_127_byte_frames",
-	"rx_255_byte_frames",
-	"rx_511_byte_frames",
-	"rx_1023_byte_frames",
-	"rx_max_size_frames",
-	"rx_oversize_frames",
-	"rx_bad_fcs_frames",
-	"rx_broadcast_frames",
-	"rx_multicast_frames",
-	"rx_control_frames",
-	"rx_pause_frames",
-	"rx_unsup_control_frames",
-	"rx_align_error_frames",
-	"rx_overrun_frames",
-	"rx_jabber_frames",
-	"rx_bytes",
-	"rx_frames",
-
-	"tx_bytes_ok",
-	"tx_frames_ok",
-	"tx_64_byte_frames",
-	"tx_127_byte_frames",
-	"tx_255_byte_frames",
-	"tx_511_byte_frames",
-	"tx_1023_byte_frames",
-	"tx_max_size_frames",
-	"tx_oversize_frames",
-	"tx_broadcast_frames",
-	"tx_multicast_frames",
-	"tx_control_frames",
-	"tx_pause_frames",
-	"tx_underrun_frames",
-	"tx_single_collision_frames",
-	"tx_multi_collision_frames",
-	"tx_deferred_collision_frames",
-	"tx_late_collision_frames",
-	"tx_excessive_collision_frames",
-	"tx_bytes",
-	"tx_frames",
-	"tx_collisions",
-};
-
-#define NB8800_NUM_STATS ARRAY_SIZE(nb8800_stats_names)
-
-static int nb8800_get_sset_count(struct net_device *dev, int sset)
-{
-	if (sset == ETH_SS_STATS)
-		return NB8800_NUM_STATS;
-
-	return -EOPNOTSUPP;
-}
-
-static void nb8800_get_strings(struct net_device *dev, u32 sset, u8 *buf)
-{
-	if (sset == ETH_SS_STATS)
-		memcpy(buf, &nb8800_stats_names, sizeof(nb8800_stats_names));
-}
-
-static u32 nb8800_read_stat(struct net_device *dev, int index)
-{
-	struct nb8800_priv *priv = netdev_priv(dev);
-
-	nb8800_writeb(priv, NB8800_STAT_INDEX, index);
-
-	return nb8800_readl(priv, NB8800_STAT_DATA);
-}
-
-static void nb8800_get_ethtool_stats(struct net_device *dev,
-				     struct ethtool_stats *estats, u64 *st)
-{
-	unsigned int i;
-	u32 rx, tx;
-
-	for (i = 0; i < NB8800_NUM_STATS / 2; i++) {
-		rx = nb8800_read_stat(dev, i);
-		tx = nb8800_read_stat(dev, i | 0x80);
-		st[i] = rx;
-		st[i + NB8800_NUM_STATS / 2] = tx;
-	}
-}
-
-static const struct ethtool_ops nb8800_ethtool_ops = {
-	.nway_reset		= phy_ethtool_nway_reset,
-	.get_link		= ethtool_op_get_link,
-	.get_pauseparam		= nb8800_get_pauseparam,
-	.set_pauseparam		= nb8800_set_pauseparam,
-	.get_sset_count		= nb8800_get_sset_count,
-	.get_strings		= nb8800_get_strings,
-	.get_ethtool_stats	= nb8800_get_ethtool_stats,
-	.get_link_ksettings	= phy_ethtool_get_link_ksettings,
-	.set_link_ksettings	= phy_ethtool_set_link_ksettings,
-};
-
-static int nb8800_hw_init(struct net_device *dev)
-{
-	struct nb8800_priv *priv = netdev_priv(dev);
-	u32 val;
-
-	val = TX_RETRY_EN | TX_PAD_EN | TX_APPEND_FCS;
-	nb8800_writeb(priv, NB8800_TX_CTL1, val);
-
-	/* Collision retry count */
-	nb8800_writeb(priv, NB8800_TX_CTL2, 5);
-
-	val = RX_PAD_STRIP | RX_AF_EN;
-	nb8800_writeb(priv, NB8800_RX_CTL, val);
-
-	/* Chosen by fair dice roll */
-	nb8800_writeb(priv, NB8800_RANDOM_SEED, 4);
-
-	/* TX cycles per deferral period */
-	nb8800_writeb(priv, NB8800_TX_SDP, 12);
-
-	/* The following three threshold values have been
-	 * experimentally determined for good results.
-	 */
-
-	/* RX/TX FIFO threshold for partial empty (64-bit entries) */
-	nb8800_writeb(priv, NB8800_PE_THRESHOLD, 0);
-
-	/* RX/TX FIFO threshold for partial full (64-bit entries) */
-	nb8800_writeb(priv, NB8800_PF_THRESHOLD, 255);
-
-	/* Buffer size for transmit (64-bit entries) */
-	nb8800_writeb(priv, NB8800_TX_BUFSIZE, 64);
-
-	/* Configure tx DMA */
-
-	val = nb8800_readl(priv, NB8800_TXC_CR);
-	val &= TCR_LE;		/* keep endian setting */
-	val |= TCR_DM;		/* DMA descriptor mode */
-	val |= TCR_RS;		/* automatically store tx status  */
-	val |= TCR_DIE;		/* interrupt on DMA chain completion */
-	val |= TCR_TFI(7);	/* interrupt after 7 frames transmitted */
-	val |= TCR_BTS(2);	/* 32-byte bus transaction size */
-	nb8800_writel(priv, NB8800_TXC_CR, val);
-
-	/* TX complete interrupt after 10 ms or 7 frames (see above) */
-	val = clk_get_rate(priv->clk) / 100;
-	nb8800_writel(priv, NB8800_TX_ITR, val);
-
-	/* Configure rx DMA */
-
-	val = nb8800_readl(priv, NB8800_RXC_CR);
-	val &= RCR_LE;		/* keep endian setting */
-	val |= RCR_DM;		/* DMA descriptor mode */
-	val |= RCR_RS;		/* automatically store rx status */
-	val |= RCR_DIE;		/* interrupt at end of DMA chain */
-	val |= RCR_RFI(7);	/* interrupt after 7 frames received */
-	val |= RCR_BTS(2);	/* 32-byte bus transaction size */
-	nb8800_writel(priv, NB8800_RXC_CR, val);
-
-	/* The rx interrupt can fire before the DMA has completed
-	 * unless a small delay is added.  50 us is hopefully enough.
-	 */
-	priv->rx_itr_irq = clk_get_rate(priv->clk) / 20000;
-
-	/* In NAPI poll mode we want to disable interrupts, but the
-	 * hardware does not permit this.  Delay 10 ms instead.
-	 */
-	priv->rx_itr_poll = clk_get_rate(priv->clk) / 100;
-
-	nb8800_writel(priv, NB8800_RX_ITR, priv->rx_itr_irq);
-
-	priv->rx_dma_config = RX_BUF_SIZE | DESC_BTS(2) | DESC_DS | DESC_EOF;
-
-	/* Flow control settings */
-
-	/* Pause time of 0.1 ms */
-	val = 100000 / 512;
-	nb8800_writeb(priv, NB8800_PQ1, val >> 8);
-	nb8800_writeb(priv, NB8800_PQ2, val & 0xff);
-
-	/* Auto-negotiate by default */
-	priv->pause_aneg = true;
-	priv->pause_rx = true;
-	priv->pause_tx = true;
-
-	nb8800_mc_init(dev, 0);
-
-	return 0;
-}
-
-static int nb8800_tangox_init(struct net_device *dev)
-{
-	struct nb8800_priv *priv = netdev_priv(dev);
-	u32 pad_mode = PAD_MODE_MII;
-
-	switch (priv->phy_mode) {
-	case PHY_INTERFACE_MODE_MII:
-	case PHY_INTERFACE_MODE_GMII:
-		pad_mode = PAD_MODE_MII;
-		break;
-
-	case PHY_INTERFACE_MODE_RGMII:
-		pad_mode = PAD_MODE_RGMII;
-		break;
-
-	case PHY_INTERFACE_MODE_RGMII_TXID:
-		pad_mode = PAD_MODE_RGMII | PAD_MODE_GTX_CLK_DELAY;
-		break;
-
-	default:
-		dev_err(dev->dev.parent, "unsupported phy mode %s\n",
-			phy_modes(priv->phy_mode));
-		return -EINVAL;
-	}
-
-	nb8800_writeb(priv, NB8800_TANGOX_PAD_MODE, pad_mode);
-
-	return 0;
-}
-
-static int nb8800_tangox_reset(struct net_device *dev)
-{
-	struct nb8800_priv *priv = netdev_priv(dev);
-	int clk_div;
-
-	nb8800_writeb(priv, NB8800_TANGOX_RESET, 0);
-	usleep_range(1000, 10000);
-	nb8800_writeb(priv, NB8800_TANGOX_RESET, 1);
-
-	wmb();		/* ensure reset is cleared before proceeding */
-
-	clk_div = DIV_ROUND_UP(clk_get_rate(priv->clk), 2 * MAX_MDC_CLOCK);
-	nb8800_writew(priv, NB8800_TANGOX_MDIO_CLKDIV, clk_div);
-
-	return 0;
-}
-
-static const struct nb8800_ops nb8800_tangox_ops = {
-	.init	= nb8800_tangox_init,
-	.reset	= nb8800_tangox_reset,
-};
-
-static int nb8800_tango4_init(struct net_device *dev)
-{
-	struct nb8800_priv *priv = netdev_priv(dev);
-	int err;
-
-	err = nb8800_tangox_init(dev);
-	if (err)
-		return err;
-
-	/* On tango4 interrupt on DMA completion per frame works and gives
-	 * better performance despite generating more rx interrupts.
-	 */
-
-	/* Disable unnecessary interrupt on rx completion */
-	nb8800_clearl(priv, NB8800_RXC_CR, RCR_RFI(7));
-
-	/* Request interrupt on descriptor DMA completion */
-	priv->rx_dma_config |= DESC_ID;
-
-	return 0;
-}
-
-static const struct nb8800_ops nb8800_tango4_ops = {
-	.init	= nb8800_tango4_init,
-	.reset	= nb8800_tangox_reset,
-};
-
-static const struct of_device_id nb8800_dt_ids[] = {
-	{
-		.compatible = "aurora,nb8800",
-	},
-	{
-		.compatible = "sigma,smp8642-ethernet",
-		.data = &nb8800_tangox_ops,
-	},
-	{
-		.compatible = "sigma,smp8734-ethernet",
-		.data = &nb8800_tango4_ops,
-	},
-	{ }
-};
-MODULE_DEVICE_TABLE(of, nb8800_dt_ids);
-
-static int nb8800_probe(struct platform_device *pdev)
-{
-	const struct of_device_id *match;
-	const struct nb8800_ops *ops = NULL;
-	struct nb8800_priv *priv;
-	struct resource *res;
-	struct net_device *dev;
-	struct mii_bus *bus;
-	const unsigned char *mac;
-	void __iomem *base;
-	int irq;
-	int ret;
-
-	match = of_match_device(nb8800_dt_ids, &pdev->dev);
-	if (match)
-		ops = match->data;
-
-	irq = platform_get_irq(pdev, 0);
-	if (irq <= 0) {
-		dev_err(&pdev->dev, "No IRQ\n");
-		return -EINVAL;
-	}
-
-	res = platform_get_resource(pdev, IORESOURCE_MEM, 0);
-	base = devm_ioremap_resource(&pdev->dev, res);
-	if (IS_ERR(base))
-		return PTR_ERR(base);
-
-	dev_dbg(&pdev->dev, "AU-NB8800 Ethernet at %pa\n", &res->start);
-
-	dev = alloc_etherdev(sizeof(*priv));
-	if (!dev)
-		return -ENOMEM;
-
-	platform_set_drvdata(pdev, dev);
-	SET_NETDEV_DEV(dev, &pdev->dev);
-
-	priv = netdev_priv(dev);
-	priv->base = base;
-
-	priv->phy_mode = of_get_phy_mode(pdev->dev.of_node);
-	if (priv->phy_mode < 0)
-		priv->phy_mode = PHY_INTERFACE_MODE_RGMII;
-
-	priv->clk = devm_clk_get(&pdev->dev, NULL);
-	if (IS_ERR(priv->clk)) {
-		dev_err(&pdev->dev, "failed to get clock\n");
-		ret = PTR_ERR(priv->clk);
-		goto err_free_dev;
-	}
-
-	ret = clk_prepare_enable(priv->clk);
-	if (ret)
-		goto err_free_dev;
-
-	spin_lock_init(&priv->tx_lock);
-
-	if (ops && ops->reset) {
-		ret = ops->reset(dev);
-		if (ret)
-			goto err_disable_clk;
-	}
-
-	bus = devm_mdiobus_alloc(&pdev->dev);
-	if (!bus) {
-		ret = -ENOMEM;
-		goto err_disable_clk;
-	}
-
-	bus->name = "nb8800-mii";
-	bus->read = nb8800_mdio_read;
-	bus->write = nb8800_mdio_write;
-	bus->parent = &pdev->dev;
-	snprintf(bus->id, MII_BUS_ID_SIZE, "%lx.nb8800-mii",
-		 (unsigned long)res->start);
-	bus->priv = priv;
-
-	ret = of_mdiobus_register(bus, pdev->dev.of_node);
-	if (ret) {
-		dev_err(&pdev->dev, "failed to register MII bus\n");
-		goto err_disable_clk;
-	}
-
-	if (of_phy_is_fixed_link(pdev->dev.of_node)) {
-		ret = of_phy_register_fixed_link(pdev->dev.of_node);
-		if (ret < 0) {
-			dev_err(&pdev->dev, "bad fixed-link spec\n");
-			goto err_free_bus;
-		}
-		priv->phy_node = of_node_get(pdev->dev.of_node);
-	}
-
-	if (!priv->phy_node)
-		priv->phy_node = of_parse_phandle(pdev->dev.of_node,
-						  "phy-handle", 0);
-
-	if (!priv->phy_node) {
-		dev_err(&pdev->dev, "no PHY specified\n");
-		ret = -ENODEV;
-		goto err_free_bus;
-	}
-
-	priv->mii_bus = bus;
-
-	ret = nb8800_hw_init(dev);
-	if (ret)
-		goto err_deregister_fixed_link;
-
-	if (ops && ops->init) {
-		ret = ops->init(dev);
-		if (ret)
-			goto err_deregister_fixed_link;
-	}
-
-	dev->netdev_ops = &nb8800_netdev_ops;
-	dev->ethtool_ops = &nb8800_ethtool_ops;
-	dev->flags |= IFF_MULTICAST;
-	dev->irq = irq;
-
-	mac = of_get_mac_address(pdev->dev.of_node);
-	if (mac)
-		ether_addr_copy(dev->dev_addr, mac);
-
-	if (!is_valid_ether_addr(dev->dev_addr))
-		eth_hw_addr_random(dev);
-
-	nb8800_update_mac_addr(dev);
-
-	netif_carrier_off(dev);
-
-	ret = register_netdev(dev);
-	if (ret) {
-		netdev_err(dev, "failed to register netdev\n");
-		goto err_free_dma;
-	}
-
-	netif_napi_add(dev, &priv->napi, nb8800_poll, NAPI_POLL_WEIGHT);
-
-	netdev_info(dev, "MAC address %pM\n", dev->dev_addr);
-
-	return 0;
-
-err_free_dma:
-	nb8800_dma_free(dev);
-err_deregister_fixed_link:
-	if (of_phy_is_fixed_link(pdev->dev.of_node))
-		of_phy_deregister_fixed_link(pdev->dev.of_node);
-err_free_bus:
-	of_node_put(priv->phy_node);
-	mdiobus_unregister(bus);
-err_disable_clk:
-	clk_disable_unprepare(priv->clk);
-err_free_dev:
-	free_netdev(dev);
-
-	return ret;
-}
-
-static int nb8800_remove(struct platform_device *pdev)
-{
-	struct net_device *ndev = platform_get_drvdata(pdev);
-	struct nb8800_priv *priv = netdev_priv(ndev);
-
-	unregister_netdev(ndev);
-	if (of_phy_is_fixed_link(pdev->dev.of_node))
-		of_phy_deregister_fixed_link(pdev->dev.of_node);
-	of_node_put(priv->phy_node);
-
-	mdiobus_unregister(priv->mii_bus);
-
-	clk_disable_unprepare(priv->clk);
-
-	nb8800_dma_free(ndev);
-	free_netdev(ndev);
-
-	return 0;
-}
-
-static struct platform_driver nb8800_driver = {
-	.driver = {
-		.name		= "nb8800",
-		.of_match_table	= nb8800_dt_ids,
-	},
-	.probe	= nb8800_probe,
-	.remove	= nb8800_remove,
-};
-
-module_platform_driver(nb8800_driver);
-
-MODULE_DESCRIPTION("Aurora AU-NB8800 Ethernet driver");
-MODULE_AUTHOR("Mans Rullgard <mans@mansr.com>");
-MODULE_LICENSE("GPL");
diff --git a/drivers/net/ethernet/aurora/nb8800.h b/drivers/net/ethernet/aurora/nb8800.h
deleted file mode 100644
index 6ec4a956e1e5..000000000000
--- a/drivers/net/ethernet/aurora/nb8800.h
+++ /dev/null
@@ -1,315 +0,0 @@
-#ifndef _NB8800_H_
-#define _NB8800_H_
-
-#include <linux/types.h>
-#include <linux/skbuff.h>
-#include <linux/phy.h>
-#include <linux/clk.h>
-#include <linux/bitops.h>
-
-#define RX_DESC_COUNT			256
-#define TX_DESC_COUNT			256
-
-#define NB8800_DESC_LOW			4
-
-#define RX_BUF_SIZE			1552
-
-#define RX_COPYBREAK			256
-#define RX_COPYHDR			128
-
-#define MAX_MDC_CLOCK			2500000
-
-/* Stargate Solutions SSN8800 core registers */
-#define NB8800_TX_CTL1			0x000
-#define TX_TPD				BIT(5)
-#define TX_APPEND_FCS			BIT(4)
-#define TX_PAD_EN			BIT(3)
-#define TX_RETRY_EN			BIT(2)
-#define TX_EN				BIT(0)
-
-#define NB8800_TX_CTL2			0x001
-
-#define NB8800_RX_CTL			0x004
-#define RX_BC_DISABLE			BIT(7)
-#define RX_RUNT				BIT(6)
-#define RX_AF_EN			BIT(5)
-#define RX_PAUSE_EN			BIT(3)
-#define RX_SEND_CRC			BIT(2)
-#define RX_PAD_STRIP			BIT(1)
-#define RX_EN				BIT(0)
-
-#define NB8800_RANDOM_SEED		0x008
-#define NB8800_TX_SDP			0x14
-#define NB8800_TX_TPDP1			0x18
-#define NB8800_TX_TPDP2			0x19
-#define NB8800_SLOT_TIME		0x1c
-
-#define NB8800_MDIO_CMD			0x020
-#define MDIO_CMD_GO			BIT(31)
-#define MDIO_CMD_WR			BIT(26)
-#define MDIO_CMD_ADDR(x)		((x) << 21)
-#define MDIO_CMD_REG(x)			((x) << 16)
-#define MDIO_CMD_DATA(x)		((x) <<	 0)
-
-#define NB8800_MDIO_STS			0x024
-#define MDIO_STS_ERR			BIT(31)
-
-#define NB8800_MC_ADDR(i)		(0x028 + (i))
-#define NB8800_MC_INIT			0x02e
-#define NB8800_UC_ADDR(i)		(0x03c + (i))
-
-#define NB8800_MAC_MODE			0x044
-#define RGMII_MODE			BIT(7)
-#define HALF_DUPLEX			BIT(4)
-#define BURST_EN			BIT(3)
-#define LOOPBACK_EN			BIT(2)
-#define GMAC_MODE			BIT(0)
-
-#define NB8800_IC_THRESHOLD		0x050
-#define NB8800_PE_THRESHOLD		0x051
-#define NB8800_PF_THRESHOLD		0x052
-#define NB8800_TX_BUFSIZE		0x054
-#define NB8800_FIFO_CTL			0x056
-#define NB8800_PQ1			0x060
-#define NB8800_PQ2			0x061
-#define NB8800_SRC_ADDR(i)		(0x06a + (i))
-#define NB8800_STAT_DATA		0x078
-#define NB8800_STAT_INDEX		0x07c
-#define NB8800_STAT_CLEAR		0x07d
-
-#define NB8800_SLEEP_MODE		0x07e
-#define SLEEP_MODE			BIT(0)
-
-#define NB8800_WAKEUP			0x07f
-#define WAKEUP				BIT(0)
-
-/* Aurora NB8800 host interface registers */
-#define NB8800_TXC_CR			0x100
-#define TCR_LK				BIT(12)
-#define TCR_DS				BIT(11)
-#define TCR_BTS(x)			(((x) & 0x7) << 8)
-#define TCR_DIE				BIT(7)
-#define TCR_TFI(x)			(((x) & 0x7) << 4)
-#define TCR_LE				BIT(3)
-#define TCR_RS				BIT(2)
-#define TCR_DM				BIT(1)
-#define TCR_EN				BIT(0)
-
-#define NB8800_TXC_SR			0x104
-#define TSR_DE				BIT(3)
-#define TSR_DI				BIT(2)
-#define TSR_TO				BIT(1)
-#define TSR_TI				BIT(0)
-
-#define NB8800_TX_SAR			0x108
-#define NB8800_TX_DESC_ADDR		0x10c
-
-#define NB8800_TX_REPORT_ADDR		0x110
-#define TX_BYTES_TRANSFERRED(x)		(((x) >> 16) & 0xffff)
-#define TX_FIRST_DEFERRAL		BIT(7)
-#define TX_EARLY_COLLISIONS(x)		(((x) >> 3) & 0xf)
-#define TX_LATE_COLLISION		BIT(2)
-#define TX_PACKET_DROPPED		BIT(1)
-#define TX_FIFO_UNDERRUN		BIT(0)
-#define IS_TX_ERROR(r)			((r) & 0x07)
-
-#define NB8800_TX_FIFO_SR		0x114
-#define NB8800_TX_ITR			0x118
-
-#define NB8800_RXC_CR			0x200
-#define RCR_FL				BIT(13)
-#define RCR_LK				BIT(12)
-#define RCR_DS				BIT(11)
-#define RCR_BTS(x)			(((x) & 7) << 8)
-#define RCR_DIE				BIT(7)
-#define RCR_RFI(x)			(((x) & 7) << 4)
-#define RCR_LE				BIT(3)
-#define RCR_RS				BIT(2)
-#define RCR_DM				BIT(1)
-#define RCR_EN				BIT(0)
-
-#define NB8800_RXC_SR			0x204
-#define RSR_DE				BIT(3)
-#define RSR_DI				BIT(2)
-#define RSR_RO				BIT(1)
-#define RSR_RI				BIT(0)
-
-#define NB8800_RX_SAR			0x208
-#define NB8800_RX_DESC_ADDR		0x20c
-
-#define NB8800_RX_REPORT_ADDR		0x210
-#define RX_BYTES_TRANSFERRED(x)		(((x) >> 16) & 0xFFFF)
-#define RX_MULTICAST_PKT		BIT(9)
-#define RX_BROADCAST_PKT		BIT(8)
-#define RX_LENGTH_ERR			BIT(7)
-#define RX_FCS_ERR			BIT(6)
-#define RX_RUNT_PKT			BIT(5)
-#define RX_FIFO_OVERRUN			BIT(4)
-#define RX_LATE_COLLISION		BIT(3)
-#define RX_ALIGNMENT_ERROR		BIT(2)
-#define RX_ERROR_MASK			0xfc
-#define IS_RX_ERROR(r)			((r) & RX_ERROR_MASK)
-
-#define NB8800_RX_FIFO_SR		0x214
-#define NB8800_RX_ITR			0x218
-
-/* Sigma Designs SMP86xx additional registers */
-#define NB8800_TANGOX_PAD_MODE		0x400
-#define PAD_MODE_MASK			0x7
-#define PAD_MODE_MII			0x0
-#define PAD_MODE_RGMII			0x1
-#define PAD_MODE_GTX_CLK_INV		BIT(3)
-#define PAD_MODE_GTX_CLK_DELAY		BIT(4)
-
-#define NB8800_TANGOX_MDIO_CLKDIV	0x420
-#define NB8800_TANGOX_RESET		0x424
-
-/* Hardware DMA descriptor */
-struct nb8800_dma_desc {
-	u32				s_addr;	/* start address */
-	u32				n_addr;	/* next descriptor address */
-	u32				r_addr;	/* report address */
-	u32				config;
-} __aligned(8);
-
-#define DESC_ID				BIT(23)
-#define DESC_EOC			BIT(22)
-#define DESC_EOF			BIT(21)
-#define DESC_LK				BIT(20)
-#define DESC_DS				BIT(19)
-#define DESC_BTS(x)			(((x) & 0x7) << 16)
-
-/* DMA descriptor and associated data for rx.
- * Allocated from coherent memory.
- */
-struct nb8800_rx_desc {
-	/* DMA descriptor */
-	struct nb8800_dma_desc		desc;
-
-	/* Status report filled in by hardware */
-	u32				report;
-};
-
-/* Address of buffer on rx ring */
-struct nb8800_rx_buf {
-	struct page			*page;
-	unsigned long			offset;
-};
-
-/* DMA descriptors and associated data for tx.
- * Allocated from coherent memory.
- */
-struct nb8800_tx_desc {
-	/* DMA descriptor.  The second descriptor is used if packet
-	 * data is unaligned.
-	 */
-	struct nb8800_dma_desc		desc[2];
-
-	/* Status report filled in by hardware */
-	u32				report;
-
-	/* Bounce buffer for initial unaligned part of packet */
-	u8				buf[8] __aligned(8);
-};
-
-/* Packet in tx queue */
-struct nb8800_tx_buf {
-	/* Currently queued skb */
-	struct sk_buff			*skb;
-
-	/* DMA address of the first descriptor */
-	dma_addr_t			dma_desc;
-
-	/* DMA address of packet data */
-	dma_addr_t			dma_addr;
-
-	/* Length of DMA mapping, less than skb->len if alignment
-	 * buffer is used.
-	 */
-	unsigned int			dma_len;
-
-	/* Number of packets in chain starting here */
-	unsigned int			chain_len;
-
-	/* Packet chain ready to be submitted to hardware */
-	bool				ready;
-};
-
-struct nb8800_priv {
-	struct napi_struct		napi;
-
-	void __iomem			*base;
-
-	/* RX DMA descriptors */
-	struct nb8800_rx_desc		*rx_descs;
-
-	/* RX buffers referenced by DMA descriptors */
-	struct nb8800_rx_buf		*rx_bufs;
-
-	/* Current end of chain */
-	u32				rx_eoc;
-
-	/* Value for rx interrupt time register in NAPI interrupt mode */
-	u32				rx_itr_irq;
-
-	/* Value for rx interrupt time register in NAPI poll mode */
-	u32				rx_itr_poll;
-
-	/* Value for config field of rx DMA descriptors */
-	u32				rx_dma_config;
-
-	/* TX DMA descriptors */
-	struct nb8800_tx_desc		*tx_descs;
-
-	/* TX packet queue */
-	struct nb8800_tx_buf		*tx_bufs;
-
-	/* Number of free tx queue entries */
-	atomic_t			tx_free;
-
-	/* First free tx queue entry */
-	u32				tx_next;
-
-	/* Next buffer to transmit */
-	u32				tx_queue;
-
-	/* Start of current packet chain */
-	struct nb8800_tx_buf		*tx_chain;
-
-	/* Next buffer to reclaim */
-	u32				tx_done;
-
-	/* Lock for DMA activation */
-	spinlock_t			tx_lock;
-
-	struct mii_bus			*mii_bus;
-	struct device_node		*phy_node;
-
-	/* PHY connection type from DT */
-	int				phy_mode;
-
-	/* Current link status */
-	int				speed;
-	int				duplex;
-	int				link;
-
-	/* Pause settings */
-	bool				pause_aneg;
-	bool				pause_rx;
-	bool				pause_tx;
-
-	/* DMA base address of rx descriptors, see rx_descs above */
-	dma_addr_t			rx_desc_dma;
-
-	/* DMA base address of tx descriptors, see tx_descs above */
-	dma_addr_t			tx_desc_dma;
-
-	struct clk			*clk;
-};
-
-struct nb8800_ops {
-	int				(*init)(struct net_device *dev);
-	int				(*reset)(struct net_device *dev);
-};
-
-#endif /* _NB8800_H_ */
diff --git a/drivers/net/ethernet/broadcom/Kconfig b/drivers/net/ethernet/broadcom/Kconfig
index 96413808c726..9fdef874f5ca 100644
--- a/drivers/net/ethernet/broadcom/Kconfig
+++ b/drivers/net/ethernet/broadcom/Kconfig
@@ -1,3 +1,4 @@
+# SPDX-License-Identifier: GPL-2.0-only
 #
 # Broadcom device configuration
 #
@@ -7,14 +8,14 @@ config NET_VENDOR_BROADCOM
 	default y
 	depends on (SSB_POSSIBLE && HAS_DMA) || PCI || BCM63XX || \
 		   SIBYTE_SB1xxx_SOC
-	---help---
+	help
 	  If you have a network (Ethernet) chipset belonging to this class,
 	  say Y.
 
 	  Note that the answer to this question does not directly affect
-	  the kernel: saying N will just case the configurator to skip all
-	  the questions regarding AMD chipsets. If you say Y, you will be asked
-	  for your specific chipset/driver in the following questions.
+	  the kernel: saying N will just cause the configurator to skip all
+	  the questions regarding Broadcom chipsets. If you say Y, you will
+	  be asked for your specific chipset/driver in the following questions.
 
 if NET_VENDOR_BROADCOM
 
@@ -24,7 +25,7 @@ config B44
 	select SSB
 	select MII
 	select PHYLIB
-	---help---
+	help
 	  If you have a network (Ethernet) controller of this type, say Y
 	  or M here.
 
@@ -50,6 +51,14 @@ config B44_PCI
 	depends on B44_PCI_AUTOSELECT && B44_PCICORE_AUTOSELECT
 	default y
 
+config BCM4908_ENET
+	tristate "Broadcom BCM4908 internal mac support"
+	depends on ARCH_BCMBCA || COMPILE_TEST
+	default y if ARCH_BCMBCA
+	help
+	  This driver supports Ethernet controller integrated into Broadcom
+	  BCM4908 family SoCs.
+
 config BCM63XX_ENET
 	tristate "Broadcom 63xx internal mac support"
 	depends on BCM63XX
@@ -61,10 +70,14 @@ config BCM63XX_ENET
 
 config BCMGENET
 	tristate "Broadcom GENET internal MAC support"
-	select MII
+	depends on HAS_IOMEM
+	depends on PTP_1588_CLOCK_OPTIONAL || !ARCH_BCM2835
 	select PHYLIB
 	select FIXED_PHY
 	select BCM7XXX_PHY
+	select MDIO_BCM_UNIMAC
+	select DIMLIB
+	select BROADCOM_PHY if ARCH_BCM2835
 	help
 	  This driver supports the built-in Ethernet MACs found in the
 	  Broadcom BCM7xxx Set Top Box family chipset.
@@ -74,7 +87,7 @@ config BNX2
 	depends on PCI
 	select CRC32
 	select FW_LOADER
-	---help---
+	help
 	  This driver supports QLogic bnx2 gigabit Ethernet cards.
 
 	  To compile this driver as a module, choose M here: the module
@@ -85,7 +98,7 @@ config CNIC
 	depends on PCI && (IPV6 || IPV6=n)
 	select BNX2
 	select UIO
-	---help---
+	help
 	  This driver supports offload features of QLogic bnx2 gigabit
 	  Ethernet cards.
 
@@ -96,7 +109,7 @@ config SB1250_MAC
 	tristate "SB1250 Gigabit Ethernet support"
 	depends on SIBYTE_SB1xxx_SOC
 	select PHYLIB
-	---help---
+	help
 	  This driver supports Gigabit Ethernet interfaces based on the
 	  Broadcom SiByte family of System-On-a-Chip parts.  They include
 	  the BCM1120, BCM1125, BCM1125H, BCM1250, BCM1255, BCM1280, BCM1455
@@ -108,9 +121,10 @@ config SB1250_MAC
 config TIGON3
 	tristate "Broadcom Tigon3 support"
 	depends on PCI
+	depends on PTP_1588_CLOCK_OPTIONAL
+	select CRC32
 	select PHYLIB
-	imply PTP_1588_CLOCK
-	---help---
+	help
 	  This driver supports Broadcom Tigon3 based gigabit Ethernet cards.
 
 	  To compile this driver as a module, choose M here: the module
@@ -120,18 +134,18 @@ config TIGON3_HWMON
 	bool "Broadcom Tigon3 HWMON support"
 	default y
 	depends on TIGON3 && HWMON && !(TIGON3=y && HWMON=m)
-	---help---
+	help
 	  Say Y if you want to expose the thermal sensor on Tigon3 devices.
 
 config BNX2X
 	tristate "Broadcom NetXtremeII 10Gb support"
 	depends on PCI
-	imply PTP_1588_CLOCK
+	depends on PTP_1588_CLOCK_OPTIONAL
 	select FW_LOADER
 	select ZLIB_INFLATE
-	select LIBCRC32C
+	select CRC32
 	select MDIO
-	---help---
+	help
 	  This driver supports Broadcom NetXtremeII 10 gigabit Ethernet cards.
 	  To compile this driver as a module, choose M here: the module
 	  will be called bnx2x.  This is recommended.
@@ -140,7 +154,7 @@ config BNX2X_SRIOV
 	bool "Broadcom 578xx and 57712 SR-IOV support"
 	depends on BNX2X && PCI_IOV
 	default y
-	---help---
+	help
 	  This configuration parameter enables Single Root Input Output
 	  Virtualization support in the 578xx and 57712 products. This
 	  allows for virtual function acceleration in virtual environments.
@@ -155,12 +169,11 @@ config BGMAC
 config BGMAC_BCMA
 	tristate "Broadcom iProc GBit BCMA support"
 	depends on BCMA && BCMA_HOST_SOC
-	depends on HAS_DMA
 	depends on BCM47XX || ARCH_BCM_5301X || COMPILE_TEST
 	select BGMAC
 	select PHYLIB
 	select FIXED_PHY
-	---help---
+	help
 	  This driver supports GBit MAC and BCM4706 GBit MAC cores on BCMA bus.
 	  They can be found on BCM47xx SoCs and provide gigabit ethernet.
 	  In case of using this driver on BCM4706 it's also requires to enable
@@ -168,23 +181,22 @@ config BGMAC_BCMA
 
 config BGMAC_PLATFORM
 	tristate "Broadcom iProc GBit platform support"
-	depends on HAS_DMA
 	depends on ARCH_BCM_IPROC || COMPILE_TEST
-	depends on OF
 	select BGMAC
 	select PHYLIB
 	select FIXED_PHY
 	default ARCH_BCM_IPROC
-	---help---
+	help
 	  Say Y here if you want to use the Broadcom iProc Gigabit Ethernet
 	  controller through the generic platform interface
 
 config SYSTEMPORT
 	tristate "Broadcom SYSTEMPORT internal MAC support"
-	depends on OF
-	select MII
+	depends on HAS_IOMEM
+	depends on NET_DSA || !NET_DSA
 	select PHYLIB
 	select FIXED_PHY
+	select DIMLIB
 	help
 	  This driver supports the built-in Ethernet MACs found in the
 	  Broadcom BCM7xxx Set Top Box family chipset using an internal
@@ -193,9 +205,14 @@ config SYSTEMPORT
 config BNXT
 	tristate "Broadcom NetXtreme-C/E support"
 	depends on PCI
+	depends on PTP_1588_CLOCK_OPTIONAL
 	select FW_LOADER
-	select LIBCRC32C
-	---help---
+	select CRC32
+	select NET_DEVLINK
+	select PAGE_POOL
+	select DIMLIB
+	select AUXILIARY_BUS
+	help
 	  This driver supports Broadcom NetXtreme-C/E 10/25/40/50 gigabit
 	  Ethernet cards.  To compile this driver as a module, choose M here:
 	  the module will be called bnxt_en.  This is recommended.
@@ -204,19 +221,57 @@ config BNXT_SRIOV
 	bool "Broadcom NetXtreme-C/E SR-IOV support"
 	depends on BNXT && PCI_IOV
 	default y
-	---help---
+	help
 	  This configuration parameter enables Single Root Input Output
 	  Virtualization support in the NetXtreme-C/E products. This
 	  allows for virtual function acceleration in virtual environments.
 
+config BNXT_FLOWER_OFFLOAD
+	bool "TC Flower offload support for NetXtreme-C/E"
+	depends on BNXT
+	default y
+	help
+	  This configuration parameter enables TC Flower packet classifier
+	  offload for eswitch.  This option enables SR-IOV switchdev eswitch
+	  offload.
+
 config BNXT_DCB
 	bool "Data Center Bridging (DCB) Support"
 	default n
 	depends on BNXT && DCB
-	---help---
+	help
 	  Say Y here if you want to use Data Center Bridging (DCB) in the
 	  driver.
 
 	  If unsure, say N.
 
+config BNXT_HWMON
+	bool "Broadcom NetXtreme-C/E HWMON support"
+	default y
+	depends on BNXT && HWMON && !(BNXT=y && HWMON=m)
+	help
+	  Say Y if you want to expose the thermal sensor data on NetXtreme-C/E
+	  devices, via the hwmon sysfs interface.
+
+config BNGE
+	tristate "Broadcom Ethernet device support"
+	depends on PCI
+	select NET_DEVLINK
+	select PAGE_POOL
+	help
+	  This driver supports Broadcom 50/100/200/400/800 gigabit Ethernet cards.
+	  The module will be called bng_en. To compile this driver as a module,
+	  choose M here.
+
+config BCMASP
+	tristate "Broadcom ASP 2.0 Ethernet support"
+	depends on ARCH_BRCMSTB || COMPILE_TEST
+	default ARCH_BRCMSTB
+	depends on OF
+	select PHYLIB
+	select MDIO_BCM_UNIMAC
+	help
+	  This configuration enables the Broadcom ASP 2.0 Ethernet controller
+	  driver which is present in Broadcom STB SoCs such as 72165.
+
 endif # NET_VENDOR_BROADCOM
diff --git a/drivers/net/ethernet/broadcom/Makefile b/drivers/net/ethernet/broadcom/Makefile
index 79f2372c66ec..10cc1c92ecfc 100644
--- a/drivers/net/ethernet/broadcom/Makefile
+++ b/drivers/net/ethernet/broadcom/Makefile
@@ -1,8 +1,10 @@
+# SPDX-License-Identifier: GPL-2.0
 #
 # Makefile for the Broadcom network device drivers.
 #
 
 obj-$(CONFIG_B44) += b44.o
+obj-$(CONFIG_BCM4908_ENET) += bcm4908_enet.o
 obj-$(CONFIG_BCM63XX_ENET) += bcm63xx_enet.o
 obj-$(CONFIG_BCMGENET) += genet/
 obj-$(CONFIG_BNX2) += bnx2.o
@@ -15,3 +17,5 @@ obj-$(CONFIG_BGMAC_BCMA) += bgmac-bcma.o bgmac-bcma-mdio.o
 obj-$(CONFIG_BGMAC_PLATFORM) += bgmac-platform.o
 obj-$(CONFIG_SYSTEMPORT) += bcmsysport.o
 obj-$(CONFIG_BNXT) += bnxt/
+obj-$(CONFIG_BCMASP) += asp2/
+obj-$(CONFIG_BNGE) += bnge/
diff --git a/drivers/net/ethernet/broadcom/asp2/Makefile b/drivers/net/ethernet/broadcom/asp2/Makefile
new file mode 100644
index 000000000000..e07550315f83
--- /dev/null
+++ b/drivers/net/ethernet/broadcom/asp2/Makefile
@@ -0,0 +1,2 @@
+obj-$(CONFIG_BCMASP) += bcm-asp.o
+bcm-asp-objs := bcmasp.o bcmasp_intf.o bcmasp_ethtool.o
diff --git a/drivers/net/ethernet/broadcom/asp2/bcmasp.c b/drivers/net/ethernet/broadcom/asp2/bcmasp.c
new file mode 100644
index 000000000000..fd35f4b4dc50
--- /dev/null
+++ b/drivers/net/ethernet/broadcom/asp2/bcmasp.c
@@ -0,0 +1,1469 @@
+// SPDX-License-Identifier: GPL-2.0
+/*
+ * Broadcom STB ASP 2.0 Driver
+ *
+ * Copyright (c) 2023 Broadcom
+ */
+#include <linux/etherdevice.h>
+#include <linux/if_vlan.h>
+#include <linux/init.h>
+#include <linux/interrupt.h>
+#include <linux/module.h>
+#include <linux/kernel.h>
+#include <linux/platform_device.h>
+#include <linux/of.h>
+#include <linux/of_address.h>
+#include <linux/of_platform.h>
+#include <linux/clk.h>
+
+#include "bcmasp.h"
+#include "bcmasp_intf_defs.h"
+
+static void _intr2_mask_clear(struct bcmasp_priv *priv, u32 mask)
+{
+	intr2_core_wl(priv, mask, ASP_INTR2_MASK_CLEAR);
+	priv->irq_mask &= ~mask;
+}
+
+static void _intr2_mask_set(struct bcmasp_priv *priv, u32 mask)
+{
+	intr2_core_wl(priv, mask, ASP_INTR2_MASK_SET);
+	priv->irq_mask |= mask;
+}
+
+void bcmasp_enable_phy_irq(struct bcmasp_intf *intf, int en)
+{
+	struct bcmasp_priv *priv = intf->parent;
+
+	/* Only supported with internal phys */
+	if (!intf->internal_phy)
+		return;
+
+	if (en)
+		_intr2_mask_clear(priv, ASP_INTR2_PHY_EVENT(intf->channel));
+	else
+		_intr2_mask_set(priv, ASP_INTR2_PHY_EVENT(intf->channel));
+}
+
+void bcmasp_enable_tx_irq(struct bcmasp_intf *intf, int en)
+{
+	struct bcmasp_priv *priv = intf->parent;
+
+	if (en)
+		_intr2_mask_clear(priv, ASP_INTR2_TX_DESC(intf->channel));
+	else
+		_intr2_mask_set(priv, ASP_INTR2_TX_DESC(intf->channel));
+}
+EXPORT_SYMBOL_GPL(bcmasp_enable_tx_irq);
+
+void bcmasp_enable_rx_irq(struct bcmasp_intf *intf, int en)
+{
+	struct bcmasp_priv *priv = intf->parent;
+
+	if (en)
+		_intr2_mask_clear(priv, ASP_INTR2_RX_ECH(intf->channel));
+	else
+		_intr2_mask_set(priv, ASP_INTR2_RX_ECH(intf->channel));
+}
+EXPORT_SYMBOL_GPL(bcmasp_enable_rx_irq);
+
+static void bcmasp_intr2_mask_set_all(struct bcmasp_priv *priv)
+{
+	_intr2_mask_set(priv, 0xffffffff);
+	priv->irq_mask = 0xffffffff;
+}
+
+static void bcmasp_intr2_clear_all(struct bcmasp_priv *priv)
+{
+	intr2_core_wl(priv, 0xffffffff, ASP_INTR2_CLEAR);
+}
+
+static void bcmasp_intr2_handling(struct bcmasp_intf *intf, u32 status)
+{
+	if (status & ASP_INTR2_RX_ECH(intf->channel)) {
+		if (likely(napi_schedule_prep(&intf->rx_napi))) {
+			bcmasp_enable_rx_irq(intf, 0);
+			__napi_schedule_irqoff(&intf->rx_napi);
+		}
+	}
+
+	if (status & ASP_INTR2_TX_DESC(intf->channel)) {
+		if (likely(napi_schedule_prep(&intf->tx_napi))) {
+			bcmasp_enable_tx_irq(intf, 0);
+			__napi_schedule_irqoff(&intf->tx_napi);
+		}
+	}
+
+	if (status & ASP_INTR2_PHY_EVENT(intf->channel))
+		phy_mac_interrupt(intf->ndev->phydev);
+}
+
+static irqreturn_t bcmasp_isr(int irq, void *data)
+{
+	struct bcmasp_priv *priv = data;
+	struct bcmasp_intf *intf;
+	u32 status;
+
+	status = intr2_core_rl(priv, ASP_INTR2_STATUS) &
+		~intr2_core_rl(priv, ASP_INTR2_MASK_STATUS);
+
+	intr2_core_wl(priv, status, ASP_INTR2_CLEAR);
+
+	if (unlikely(status == 0)) {
+		dev_warn(&priv->pdev->dev, "l2 spurious interrupt\n");
+		return IRQ_NONE;
+	}
+
+	/* Handle intferfaces */
+	list_for_each_entry(intf, &priv->intfs, list)
+		bcmasp_intr2_handling(intf, status);
+
+	return IRQ_HANDLED;
+}
+
+void bcmasp_flush_rx_port(struct bcmasp_intf *intf)
+{
+	struct bcmasp_priv *priv = intf->parent;
+	u32 mask;
+
+	switch (intf->port) {
+	case 0:
+		mask = ASP_CTRL_UMAC0_FLUSH_MASK;
+		break;
+	case 1:
+		mask = ASP_CTRL_UMAC1_FLUSH_MASK;
+		break;
+	case 2:
+		mask = ASP_CTRL_SPB_FLUSH_MASK;
+		break;
+	default:
+		/* Not valid port */
+		return;
+	}
+
+	rx_ctrl_core_wl(priv, mask, ASP_RX_CTRL_FLUSH);
+}
+
+static void bcmasp_netfilt_hw_en_wake(struct bcmasp_priv *priv,
+				      struct bcmasp_net_filter *nfilt)
+{
+	rx_filter_core_wl(priv, ASP_RX_FILTER_NET_OFFSET_L3_1(64),
+			  ASP_RX_FILTER_NET_OFFSET(nfilt->hw_index));
+
+	rx_filter_core_wl(priv, ASP_RX_FILTER_NET_OFFSET_L2(32) |
+			  ASP_RX_FILTER_NET_OFFSET_L3_0(32) |
+			  ASP_RX_FILTER_NET_OFFSET_L3_1(96) |
+			  ASP_RX_FILTER_NET_OFFSET_L4(32),
+			  ASP_RX_FILTER_NET_OFFSET(nfilt->hw_index + 1));
+
+	rx_filter_core_wl(priv, ASP_RX_FILTER_NET_CFG_CH(nfilt->port + 8) |
+			  ASP_RX_FILTER_NET_CFG_EN |
+			  ASP_RX_FILTER_NET_CFG_L2_EN |
+			  ASP_RX_FILTER_NET_CFG_L3_EN |
+			  ASP_RX_FILTER_NET_CFG_L4_EN |
+			  ASP_RX_FILTER_NET_CFG_L3_FRM(2) |
+			  ASP_RX_FILTER_NET_CFG_L4_FRM(2) |
+			  ASP_RX_FILTER_NET_CFG_UMC(nfilt->port),
+			  ASP_RX_FILTER_NET_CFG(nfilt->hw_index));
+
+	rx_filter_core_wl(priv, ASP_RX_FILTER_NET_CFG_CH(nfilt->port + 8) |
+			  ASP_RX_FILTER_NET_CFG_EN |
+			  ASP_RX_FILTER_NET_CFG_L2_EN |
+			  ASP_RX_FILTER_NET_CFG_L3_EN |
+			  ASP_RX_FILTER_NET_CFG_L4_EN |
+			  ASP_RX_FILTER_NET_CFG_L3_FRM(2) |
+			  ASP_RX_FILTER_NET_CFG_L4_FRM(2) |
+			  ASP_RX_FILTER_NET_CFG_UMC(nfilt->port),
+			  ASP_RX_FILTER_NET_CFG(nfilt->hw_index + 1));
+}
+
+#define MAX_WAKE_FILTER_SIZE		256
+enum asp_netfilt_reg_type {
+	ASP_NETFILT_MATCH = 0,
+	ASP_NETFILT_MASK,
+	ASP_NETFILT_MAX
+};
+
+static int bcmasp_netfilt_get_reg_offset(struct bcmasp_priv *priv,
+					 struct bcmasp_net_filter *nfilt,
+					 enum asp_netfilt_reg_type reg_type,
+					 u32 offset)
+{
+	u32 block_index, filter_sel;
+
+	if (offset < 32) {
+		block_index = ASP_RX_FILTER_NET_L2;
+		filter_sel = nfilt->hw_index;
+	} else if (offset < 64) {
+		block_index = ASP_RX_FILTER_NET_L2;
+		filter_sel = nfilt->hw_index + 1;
+	} else if (offset < 96) {
+		block_index = ASP_RX_FILTER_NET_L3_0;
+		filter_sel = nfilt->hw_index;
+	} else if (offset < 128) {
+		block_index = ASP_RX_FILTER_NET_L3_0;
+		filter_sel = nfilt->hw_index + 1;
+	} else if (offset < 160) {
+		block_index = ASP_RX_FILTER_NET_L3_1;
+		filter_sel = nfilt->hw_index;
+	} else if (offset < 192) {
+		block_index = ASP_RX_FILTER_NET_L3_1;
+		filter_sel = nfilt->hw_index + 1;
+	} else if (offset < 224) {
+		block_index = ASP_RX_FILTER_NET_L4;
+		filter_sel = nfilt->hw_index;
+	} else if (offset < 256) {
+		block_index = ASP_RX_FILTER_NET_L4;
+		filter_sel = nfilt->hw_index + 1;
+	} else {
+		return -EINVAL;
+	}
+
+	switch (reg_type) {
+	case ASP_NETFILT_MATCH:
+		return ASP_RX_FILTER_NET_PAT(filter_sel, block_index,
+					     (offset % 32));
+	case ASP_NETFILT_MASK:
+		return ASP_RX_FILTER_NET_MASK(filter_sel, block_index,
+					      (offset % 32));
+	default:
+		return -EINVAL;
+	}
+}
+
+static void bcmasp_netfilt_wr(struct bcmasp_priv *priv,
+			      struct bcmasp_net_filter *nfilt,
+			      enum asp_netfilt_reg_type reg_type,
+			      u32 val, u32 offset)
+{
+	int reg_offset;
+
+	/* HW only accepts 4 byte aligned writes */
+	if (!IS_ALIGNED(offset, 4) || offset > MAX_WAKE_FILTER_SIZE)
+		return;
+
+	reg_offset = bcmasp_netfilt_get_reg_offset(priv, nfilt, reg_type,
+						   offset);
+
+	rx_filter_core_wl(priv, val, reg_offset);
+}
+
+static u32 bcmasp_netfilt_rd(struct bcmasp_priv *priv,
+			     struct bcmasp_net_filter *nfilt,
+			     enum asp_netfilt_reg_type reg_type,
+			     u32 offset)
+{
+	int reg_offset;
+
+	/* HW only accepts 4 byte aligned writes */
+	if (!IS_ALIGNED(offset, 4) || offset > MAX_WAKE_FILTER_SIZE)
+		return 0;
+
+	reg_offset = bcmasp_netfilt_get_reg_offset(priv, nfilt, reg_type,
+						   offset);
+
+	return rx_filter_core_rl(priv, reg_offset);
+}
+
+static int bcmasp_netfilt_wr_m_wake(struct bcmasp_priv *priv,
+				    struct bcmasp_net_filter *nfilt,
+				    u32 offset, void *match, void *mask,
+				    size_t size)
+{
+	u32 shift, mask_val = 0, match_val = 0;
+	bool first_byte = true;
+
+	if ((offset + size) > MAX_WAKE_FILTER_SIZE)
+		return -EINVAL;
+
+	while (size--) {
+		/* The HW only accepts 4 byte aligned writes, so if we
+		 * begin unaligned or if remaining bytes less than 4,
+		 * we need to read then write to avoid losing current
+		 * register state
+		 */
+		if (first_byte && (!IS_ALIGNED(offset, 4) || size < 3)) {
+			match_val = bcmasp_netfilt_rd(priv, nfilt,
+						      ASP_NETFILT_MATCH,
+						      ALIGN_DOWN(offset, 4));
+			mask_val = bcmasp_netfilt_rd(priv, nfilt,
+						     ASP_NETFILT_MASK,
+						     ALIGN_DOWN(offset, 4));
+		}
+
+		shift = (3 - (offset % 4)) * 8;
+		match_val &= ~GENMASK(shift + 7, shift);
+		mask_val &= ~GENMASK(shift + 7, shift);
+		match_val |= (u32)(*((u8 *)match) << shift);
+		mask_val |= (u32)(*((u8 *)mask) << shift);
+
+		/* If last byte or last byte of word, write to reg */
+		if (!size || ((offset % 4) == 3)) {
+			bcmasp_netfilt_wr(priv, nfilt, ASP_NETFILT_MATCH,
+					  match_val, ALIGN_DOWN(offset, 4));
+			bcmasp_netfilt_wr(priv, nfilt, ASP_NETFILT_MASK,
+					  mask_val, ALIGN_DOWN(offset, 4));
+			first_byte = true;
+		} else {
+			first_byte = false;
+		}
+
+		offset++;
+		match++;
+		mask++;
+	}
+
+	return 0;
+}
+
+static void bcmasp_netfilt_reset_hw(struct bcmasp_priv *priv,
+				    struct bcmasp_net_filter *nfilt)
+{
+	int i;
+
+	for (i = 0; i < MAX_WAKE_FILTER_SIZE; i += 4) {
+		bcmasp_netfilt_wr(priv, nfilt, ASP_NETFILT_MATCH, 0, i);
+		bcmasp_netfilt_wr(priv, nfilt, ASP_NETFILT_MASK, 0, i);
+	}
+}
+
+static void bcmasp_netfilt_tcpip4_wr(struct bcmasp_priv *priv,
+				     struct bcmasp_net_filter *nfilt,
+				     struct ethtool_tcpip4_spec *match,
+				     struct ethtool_tcpip4_spec *mask,
+				     u32 offset)
+{
+	__be16 val_16, mask_16;
+
+	val_16 = htons(ETH_P_IP);
+	mask_16 = htons(0xFFFF);
+	bcmasp_netfilt_wr_m_wake(priv, nfilt, (ETH_ALEN * 2) + offset,
+				 &val_16, &mask_16, sizeof(val_16));
+	bcmasp_netfilt_wr_m_wake(priv, nfilt, ETH_HLEN + offset + 1,
+				 &match->tos, &mask->tos,
+				 sizeof(match->tos));
+	bcmasp_netfilt_wr_m_wake(priv, nfilt, ETH_HLEN + offset + 12,
+				 &match->ip4src, &mask->ip4src,
+				 sizeof(match->ip4src));
+	bcmasp_netfilt_wr_m_wake(priv, nfilt, ETH_HLEN + offset + 16,
+				 &match->ip4dst, &mask->ip4dst,
+				 sizeof(match->ip4dst));
+	bcmasp_netfilt_wr_m_wake(priv, nfilt, ETH_HLEN + offset + 20,
+				 &match->psrc, &mask->psrc,
+				 sizeof(match->psrc));
+	bcmasp_netfilt_wr_m_wake(priv, nfilt, ETH_HLEN + offset + 22,
+				 &match->pdst, &mask->pdst,
+				 sizeof(match->pdst));
+}
+
+static void bcmasp_netfilt_tcpip6_wr(struct bcmasp_priv *priv,
+				     struct bcmasp_net_filter *nfilt,
+				     struct ethtool_tcpip6_spec *match,
+				     struct ethtool_tcpip6_spec *mask,
+				     u32 offset)
+{
+	__be16 val_16, mask_16;
+
+	val_16 = htons(ETH_P_IPV6);
+	mask_16 = htons(0xFFFF);
+	bcmasp_netfilt_wr_m_wake(priv, nfilt, (ETH_ALEN * 2) + offset,
+				 &val_16, &mask_16, sizeof(val_16));
+	val_16 = htons(match->tclass << 4);
+	mask_16 = htons(mask->tclass << 4);
+	bcmasp_netfilt_wr_m_wake(priv, nfilt, ETH_HLEN + offset,
+				 &val_16, &mask_16, sizeof(val_16));
+	bcmasp_netfilt_wr_m_wake(priv, nfilt, ETH_HLEN + offset + 8,
+				 &match->ip6src, &mask->ip6src,
+				 sizeof(match->ip6src));
+	bcmasp_netfilt_wr_m_wake(priv, nfilt, ETH_HLEN + offset + 24,
+				 &match->ip6dst, &mask->ip6dst,
+				 sizeof(match->ip6dst));
+	bcmasp_netfilt_wr_m_wake(priv, nfilt, ETH_HLEN + offset + 40,
+				 &match->psrc, &mask->psrc,
+				 sizeof(match->psrc));
+	bcmasp_netfilt_wr_m_wake(priv, nfilt, ETH_HLEN + offset + 42,
+				 &match->pdst, &mask->pdst,
+				 sizeof(match->pdst));
+}
+
+static int bcmasp_netfilt_wr_to_hw(struct bcmasp_priv *priv,
+				   struct bcmasp_net_filter *nfilt)
+{
+	struct ethtool_rx_flow_spec *fs = &nfilt->fs;
+	unsigned int offset = 0;
+	__be16 val_16, mask_16;
+	u8 val_8, mask_8;
+
+	/* Currently only supports wake filters */
+	if (!nfilt->wake_filter)
+		return -EINVAL;
+
+	bcmasp_netfilt_reset_hw(priv, nfilt);
+
+	if (fs->flow_type & FLOW_MAC_EXT) {
+		bcmasp_netfilt_wr_m_wake(priv, nfilt, 0, &fs->h_ext.h_dest,
+					 &fs->m_ext.h_dest,
+					 sizeof(fs->h_ext.h_dest));
+	}
+
+	if ((fs->flow_type & FLOW_EXT) &&
+	    (fs->m_ext.vlan_etype || fs->m_ext.vlan_tci)) {
+		bcmasp_netfilt_wr_m_wake(priv, nfilt, (ETH_ALEN * 2),
+					 &fs->h_ext.vlan_etype,
+					 &fs->m_ext.vlan_etype,
+					 sizeof(fs->h_ext.vlan_etype));
+		bcmasp_netfilt_wr_m_wake(priv, nfilt, ((ETH_ALEN * 2) + 2),
+					 &fs->h_ext.vlan_tci,
+					 &fs->m_ext.vlan_tci,
+					 sizeof(fs->h_ext.vlan_tci));
+		offset += VLAN_HLEN;
+	}
+
+	switch (fs->flow_type & ~(FLOW_EXT | FLOW_MAC_EXT)) {
+	case ETHER_FLOW:
+		bcmasp_netfilt_wr_m_wake(priv, nfilt, 0,
+					 &fs->h_u.ether_spec.h_dest,
+					 &fs->m_u.ether_spec.h_dest,
+					 sizeof(fs->h_u.ether_spec.h_dest));
+		bcmasp_netfilt_wr_m_wake(priv, nfilt, ETH_ALEN,
+					 &fs->h_u.ether_spec.h_source,
+					 &fs->m_u.ether_spec.h_source,
+					 sizeof(fs->h_u.ether_spec.h_source));
+		bcmasp_netfilt_wr_m_wake(priv, nfilt, (ETH_ALEN * 2) + offset,
+					 &fs->h_u.ether_spec.h_proto,
+					 &fs->m_u.ether_spec.h_proto,
+					 sizeof(fs->h_u.ether_spec.h_proto));
+
+		break;
+	case IP_USER_FLOW:
+		val_16 = htons(ETH_P_IP);
+		mask_16 = htons(0xFFFF);
+		bcmasp_netfilt_wr_m_wake(priv, nfilt, (ETH_ALEN * 2) + offset,
+					 &val_16, &mask_16, sizeof(val_16));
+		bcmasp_netfilt_wr_m_wake(priv, nfilt, ETH_HLEN + offset + 1,
+					 &fs->h_u.usr_ip4_spec.tos,
+					 &fs->m_u.usr_ip4_spec.tos,
+					 sizeof(fs->h_u.usr_ip4_spec.tos));
+		bcmasp_netfilt_wr_m_wake(priv, nfilt, ETH_HLEN + offset + 9,
+					 &fs->h_u.usr_ip4_spec.proto,
+					 &fs->m_u.usr_ip4_spec.proto,
+					 sizeof(fs->h_u.usr_ip4_spec.proto));
+		bcmasp_netfilt_wr_m_wake(priv, nfilt, ETH_HLEN + offset + 12,
+					 &fs->h_u.usr_ip4_spec.ip4src,
+					 &fs->m_u.usr_ip4_spec.ip4src,
+					 sizeof(fs->h_u.usr_ip4_spec.ip4src));
+		bcmasp_netfilt_wr_m_wake(priv, nfilt, ETH_HLEN + offset + 16,
+					 &fs->h_u.usr_ip4_spec.ip4dst,
+					 &fs->m_u.usr_ip4_spec.ip4dst,
+					 sizeof(fs->h_u.usr_ip4_spec.ip4dst));
+		if (!fs->m_u.usr_ip4_spec.l4_4_bytes)
+			break;
+
+		/* Only supports 20 byte IPv4 header */
+		val_8 = 0x45;
+		mask_8 = 0xFF;
+		bcmasp_netfilt_wr_m_wake(priv, nfilt, ETH_HLEN + offset,
+					 &val_8, &mask_8, sizeof(val_8));
+		bcmasp_netfilt_wr_m_wake(priv, nfilt,
+					 ETH_HLEN + 20 + offset,
+					 &fs->h_u.usr_ip4_spec.l4_4_bytes,
+					 &fs->m_u.usr_ip4_spec.l4_4_bytes,
+					 sizeof(fs->h_u.usr_ip4_spec.l4_4_bytes)
+					 );
+		break;
+	case TCP_V4_FLOW:
+		val_8 = IPPROTO_TCP;
+		mask_8 = 0xFF;
+		bcmasp_netfilt_tcpip4_wr(priv, nfilt, &fs->h_u.tcp_ip4_spec,
+					 &fs->m_u.tcp_ip4_spec, offset);
+		bcmasp_netfilt_wr_m_wake(priv, nfilt, ETH_HLEN + offset + 9,
+					 &val_8, &mask_8, sizeof(val_8));
+		break;
+	case UDP_V4_FLOW:
+		val_8 = IPPROTO_UDP;
+		mask_8 = 0xFF;
+		bcmasp_netfilt_tcpip4_wr(priv, nfilt, &fs->h_u.udp_ip4_spec,
+					 &fs->m_u.udp_ip4_spec, offset);
+
+		bcmasp_netfilt_wr_m_wake(priv, nfilt, ETH_HLEN + offset + 9,
+					 &val_8, &mask_8, sizeof(val_8));
+		break;
+	case TCP_V6_FLOW:
+		val_8 = IPPROTO_TCP;
+		mask_8 = 0xFF;
+		bcmasp_netfilt_tcpip6_wr(priv, nfilt, &fs->h_u.tcp_ip6_spec,
+					 &fs->m_u.tcp_ip6_spec, offset);
+		bcmasp_netfilt_wr_m_wake(priv, nfilt, ETH_HLEN + offset + 6,
+					 &val_8, &mask_8, sizeof(val_8));
+		break;
+	case UDP_V6_FLOW:
+		val_8 = IPPROTO_UDP;
+		mask_8 = 0xFF;
+		bcmasp_netfilt_tcpip6_wr(priv, nfilt, &fs->h_u.udp_ip6_spec,
+					 &fs->m_u.udp_ip6_spec, offset);
+		bcmasp_netfilt_wr_m_wake(priv, nfilt, ETH_HLEN + offset + 6,
+					 &val_8, &mask_8, sizeof(val_8));
+		break;
+	}
+
+	bcmasp_netfilt_hw_en_wake(priv, nfilt);
+
+	return 0;
+}
+
+void bcmasp_netfilt_suspend(struct bcmasp_intf *intf)
+{
+	struct bcmasp_priv *priv = intf->parent;
+	bool write = false;
+	int ret, i;
+
+	/* Write all filters to HW */
+	for (i = 0; i < priv->num_net_filters; i++) {
+		/* If the filter does not match the port, skip programming. */
+		if (!priv->net_filters[i].claimed ||
+		    priv->net_filters[i].port != intf->port)
+			continue;
+
+		if (i > 0 && (i % 2) &&
+		    priv->net_filters[i].wake_filter &&
+		    priv->net_filters[i - 1].wake_filter)
+			continue;
+
+		ret = bcmasp_netfilt_wr_to_hw(priv, &priv->net_filters[i]);
+		if (!ret)
+			write = true;
+	}
+
+	/* Successfully programmed at least one wake filter
+	 * so enable top level wake config
+	 */
+	if (write)
+		rx_filter_core_wl(priv, (ASP_RX_FILTER_OPUT_EN |
+				  ASP_RX_FILTER_LNR_MD |
+				  ASP_RX_FILTER_GEN_WK_EN |
+				  ASP_RX_FILTER_NT_FLT_EN),
+				  ASP_RX_FILTER_BLK_CTRL);
+}
+
+int bcmasp_netfilt_get_all_active(struct bcmasp_intf *intf, u32 *rule_locs,
+				  u32 *rule_cnt)
+{
+	struct bcmasp_priv *priv = intf->parent;
+	int j = 0, i;
+
+	for (i = 0; i < priv->num_net_filters; i++) {
+		if (!priv->net_filters[i].claimed ||
+		    priv->net_filters[i].port != intf->port)
+			continue;
+
+		if (i > 0 && (i % 2) &&
+		    priv->net_filters[i].wake_filter &&
+		    priv->net_filters[i - 1].wake_filter)
+			continue;
+
+		if (j == *rule_cnt)
+			return -EMSGSIZE;
+
+		rule_locs[j++] = priv->net_filters[i].fs.location;
+	}
+
+	*rule_cnt = j;
+
+	return 0;
+}
+
+int bcmasp_netfilt_get_active(struct bcmasp_intf *intf)
+{
+	struct bcmasp_priv *priv = intf->parent;
+	int cnt = 0, i;
+
+	for (i = 0; i < priv->num_net_filters; i++) {
+		if (!priv->net_filters[i].claimed ||
+		    priv->net_filters[i].port != intf->port)
+			continue;
+
+		/* Skip over a wake filter pair */
+		if (i > 0 && (i % 2) &&
+		    priv->net_filters[i].wake_filter &&
+		    priv->net_filters[i - 1].wake_filter)
+			continue;
+
+		cnt++;
+	}
+
+	return cnt;
+}
+
+bool bcmasp_netfilt_check_dup(struct bcmasp_intf *intf,
+			      struct ethtool_rx_flow_spec *fs)
+{
+	struct bcmasp_priv *priv = intf->parent;
+	struct ethtool_rx_flow_spec *cur;
+	size_t fs_size = 0;
+	int i;
+
+	for (i = 0; i < priv->num_net_filters; i++) {
+		if (!priv->net_filters[i].claimed ||
+		    priv->net_filters[i].port != intf->port)
+			continue;
+
+		cur = &priv->net_filters[i].fs;
+
+		if (cur->flow_type != fs->flow_type ||
+		    cur->ring_cookie != fs->ring_cookie)
+			continue;
+
+		switch (fs->flow_type & ~(FLOW_EXT | FLOW_MAC_EXT)) {
+		case ETHER_FLOW:
+			fs_size = sizeof(struct ethhdr);
+			break;
+		case IP_USER_FLOW:
+			fs_size = sizeof(struct ethtool_usrip4_spec);
+			break;
+		case TCP_V6_FLOW:
+		case UDP_V6_FLOW:
+			fs_size = sizeof(struct ethtool_tcpip6_spec);
+			break;
+		case TCP_V4_FLOW:
+		case UDP_V4_FLOW:
+			fs_size = sizeof(struct ethtool_tcpip4_spec);
+			break;
+		default:
+			continue;
+		}
+
+		if (memcmp(&cur->h_u, &fs->h_u, fs_size) ||
+		    memcmp(&cur->m_u, &fs->m_u, fs_size))
+			continue;
+
+		if (cur->flow_type & FLOW_EXT) {
+			if (cur->h_ext.vlan_etype != fs->h_ext.vlan_etype ||
+			    cur->m_ext.vlan_etype != fs->m_ext.vlan_etype ||
+			    cur->h_ext.vlan_tci != fs->h_ext.vlan_tci ||
+			    cur->m_ext.vlan_tci != fs->m_ext.vlan_tci ||
+			    cur->h_ext.data[0] != fs->h_ext.data[0])
+				continue;
+		}
+		if (cur->flow_type & FLOW_MAC_EXT) {
+			if (memcmp(&cur->h_ext.h_dest,
+				   &fs->h_ext.h_dest, ETH_ALEN) ||
+			    memcmp(&cur->m_ext.h_dest,
+				   &fs->m_ext.h_dest, ETH_ALEN))
+				continue;
+		}
+
+		return true;
+	}
+
+	return false;
+}
+
+/* If no network filter found, return open filter.
+ * If no more open filters return NULL
+ */
+struct bcmasp_net_filter *bcmasp_netfilt_get_init(struct bcmasp_intf *intf,
+						  u32 loc, bool wake_filter,
+						  bool init)
+{
+	struct bcmasp_net_filter *nfilter = NULL;
+	struct bcmasp_priv *priv = intf->parent;
+	int i, open_index = -1;
+
+	/* Check whether we exceed the filter table capacity */
+	if (loc != RX_CLS_LOC_ANY && loc >= priv->num_net_filters)
+		return ERR_PTR(-EINVAL);
+
+	/* If the filter location is busy (already claimed) and we are initializing
+	 * the filter (insertion), return a busy error code.
+	 */
+	if (loc != RX_CLS_LOC_ANY && init && priv->net_filters[loc].claimed)
+		return ERR_PTR(-EBUSY);
+
+	/* We need two filters for wake-up, so we cannot use an odd filter */
+	if (wake_filter && loc != RX_CLS_LOC_ANY && (loc % 2))
+		return ERR_PTR(-EINVAL);
+
+	/* Initialize the loop index based on the desired location or from 0 */
+	i = loc == RX_CLS_LOC_ANY ? 0 : loc;
+
+	for ( ; i < priv->num_net_filters; i++) {
+		/* Found matching network filter */
+		if (!init &&
+		    priv->net_filters[i].claimed &&
+		    priv->net_filters[i].hw_index == i &&
+		    priv->net_filters[i].port == intf->port)
+			return &priv->net_filters[i];
+
+		/* If we don't need a new filter or new filter already found */
+		if (!init || open_index >= 0)
+			continue;
+
+		/* Wake filter conslidates two filters to cover more bytes
+		 * Wake filter is open if...
+		 * 1. It is an even filter
+		 * 2. The current and next filter is not claimed
+		 */
+		if (wake_filter && !(i % 2) && !priv->net_filters[i].claimed &&
+		    !priv->net_filters[i + 1].claimed)
+			open_index = i;
+		else if (!priv->net_filters[i].claimed)
+			open_index = i;
+	}
+
+	if (open_index >= 0) {
+		nfilter = &priv->net_filters[open_index];
+		nfilter->claimed = true;
+		nfilter->port = intf->port;
+		nfilter->hw_index = open_index;
+	}
+
+	if (wake_filter && open_index >= 0) {
+		/* Claim next filter */
+		priv->net_filters[open_index + 1].claimed = true;
+		priv->net_filters[open_index + 1].wake_filter = true;
+		nfilter->wake_filter = true;
+	}
+
+	return nfilter ? nfilter : ERR_PTR(-EINVAL);
+}
+
+void bcmasp_netfilt_release(struct bcmasp_intf *intf,
+			    struct bcmasp_net_filter *nfilt)
+{
+	struct bcmasp_priv *priv = intf->parent;
+
+	if (nfilt->wake_filter) {
+		memset(&priv->net_filters[nfilt->hw_index + 1], 0,
+		       sizeof(struct bcmasp_net_filter));
+	}
+
+	memset(nfilt, 0, sizeof(struct bcmasp_net_filter));
+}
+
+static void bcmasp_addr_to_uint(unsigned char *addr, u32 *high, u32 *low)
+{
+	*high = (u32)(addr[0] << 8 | addr[1]);
+	*low = (u32)(addr[2] << 24 | addr[3] << 16 | addr[4] << 8 |
+		     addr[5]);
+}
+
+static void bcmasp_set_mda_filter(struct bcmasp_intf *intf,
+				  const unsigned char *addr,
+				  unsigned char *mask,
+				  unsigned int i)
+{
+	struct bcmasp_priv *priv = intf->parent;
+	u32 addr_h, addr_l, mask_h, mask_l;
+
+	/* Set local copy */
+	ether_addr_copy(priv->mda_filters[i].mask, mask);
+	ether_addr_copy(priv->mda_filters[i].addr, addr);
+
+	/* Write to HW */
+	bcmasp_addr_to_uint(priv->mda_filters[i].mask, &mask_h, &mask_l);
+	bcmasp_addr_to_uint(priv->mda_filters[i].addr, &addr_h, &addr_l);
+	rx_filter_core_wl(priv, addr_h, ASP_RX_FILTER_MDA_PAT_H(i));
+	rx_filter_core_wl(priv, addr_l, ASP_RX_FILTER_MDA_PAT_L(i));
+	rx_filter_core_wl(priv, mask_h, ASP_RX_FILTER_MDA_MSK_H(i));
+	rx_filter_core_wl(priv, mask_l, ASP_RX_FILTER_MDA_MSK_L(i));
+}
+
+static void bcmasp_en_mda_filter(struct bcmasp_intf *intf, bool en,
+				 unsigned int i)
+{
+	struct bcmasp_priv *priv = intf->parent;
+
+	if (priv->mda_filters[i].en == en)
+		return;
+
+	priv->mda_filters[i].en = en;
+	priv->mda_filters[i].port = intf->port;
+
+	rx_filter_core_wl(priv, ((intf->channel + priv->tx_chan_offset) |
+			  (en << ASP_RX_FILTER_MDA_CFG_EN_SHIFT) |
+			  ASP_RX_FILTER_MDA_CFG_UMC_SEL(intf->port)),
+			  ASP_RX_FILTER_MDA_CFG(i));
+}
+
+/* There are 32 MDA filters shared between all ports, we reserve 4 filters per
+ * port for the following.
+ * - Promisc: Filter to allow all packets when promisc is enabled
+ * - All Multicast
+ * - Broadcast
+ * - Own address
+ *
+ * The reserved filters are identified as so.
+ * - Promisc: (index * 4) + 0
+ * - All Multicast: (index * 4) + 1
+ * - Broadcast: (index * 4) + 2
+ * - Own address: (index * 4) + 3
+ */
+enum asp_rx_filter_id {
+	ASP_RX_FILTER_MDA_PROMISC = 0,
+	ASP_RX_FILTER_MDA_ALLMULTI,
+	ASP_RX_FILTER_MDA_BROADCAST,
+	ASP_RX_FILTER_MDA_OWN_ADDR,
+	ASP_RX_FILTER_MDA_RES_MAX,
+};
+
+#define ASP_RX_FILT_MDA(intf, name)	(((intf)->index * \
+					  ASP_RX_FILTER_MDA_RES_MAX) \
+					 + ASP_RX_FILTER_MDA_##name)
+
+static int bcmasp_total_res_mda_cnt(struct bcmasp_priv *priv)
+{
+	return list_count_nodes(&priv->intfs) * ASP_RX_FILTER_MDA_RES_MAX;
+}
+
+void bcmasp_set_promisc(struct bcmasp_intf *intf, bool en)
+{
+	unsigned int i = ASP_RX_FILT_MDA(intf, PROMISC);
+	unsigned char promisc[ETH_ALEN];
+
+	eth_zero_addr(promisc);
+	/* Set mask to 00:00:00:00:00:00 to match all packets */
+	bcmasp_set_mda_filter(intf, promisc, promisc, i);
+	bcmasp_en_mda_filter(intf, en, i);
+}
+
+void bcmasp_set_allmulti(struct bcmasp_intf *intf, bool en)
+{
+	unsigned char allmulti[] = {0x01, 0x00, 0x00, 0x00, 0x00, 0x00};
+	unsigned int i = ASP_RX_FILT_MDA(intf, ALLMULTI);
+
+	/* Set mask to 01:00:00:00:00:00 to match all multicast */
+	bcmasp_set_mda_filter(intf, allmulti, allmulti, i);
+	bcmasp_en_mda_filter(intf, en, i);
+}
+
+void bcmasp_set_broad(struct bcmasp_intf *intf, bool en)
+{
+	unsigned int i = ASP_RX_FILT_MDA(intf, BROADCAST);
+	unsigned char addr[ETH_ALEN];
+
+	eth_broadcast_addr(addr);
+	bcmasp_set_mda_filter(intf, addr, addr, i);
+	bcmasp_en_mda_filter(intf, en, i);
+}
+
+void bcmasp_set_oaddr(struct bcmasp_intf *intf, const unsigned char *addr,
+		      bool en)
+{
+	unsigned char mask[] = {0xff, 0xff, 0xff, 0xff, 0xff, 0xff};
+	unsigned int i = ASP_RX_FILT_MDA(intf, OWN_ADDR);
+
+	bcmasp_set_mda_filter(intf, addr, mask, i);
+	bcmasp_en_mda_filter(intf, en, i);
+}
+
+void bcmasp_disable_all_filters(struct bcmasp_intf *intf)
+{
+	struct bcmasp_priv *priv = intf->parent;
+	unsigned int i;
+	int res_count;
+
+	res_count = bcmasp_total_res_mda_cnt(intf->parent);
+
+	/* Disable all filters held by this port */
+	for (i = res_count; i < priv->num_mda_filters; i++) {
+		if (priv->mda_filters[i].en &&
+		    priv->mda_filters[i].port == intf->port)
+			bcmasp_en_mda_filter(intf, 0, i);
+	}
+}
+
+static int bcmasp_combine_set_filter(struct bcmasp_intf *intf,
+				     unsigned char *addr, unsigned char *mask,
+				     int i)
+{
+	struct bcmasp_priv *priv = intf->parent;
+	u64 addr1, addr2, mask1, mask2, mask3;
+
+	/* Switch to u64 to help with the calculations */
+	addr1 = ether_addr_to_u64(priv->mda_filters[i].addr);
+	mask1 = ether_addr_to_u64(priv->mda_filters[i].mask);
+	addr2 = ether_addr_to_u64(addr);
+	mask2 = ether_addr_to_u64(mask);
+
+	/* Check if one filter resides within the other */
+	mask3 = mask1 & mask2;
+	if (mask3 == mask1 && ((addr1 & mask1) == (addr2 & mask1))) {
+		/* Filter 2 resides within filter 1, so everything is good */
+		return 0;
+	} else if (mask3 == mask2 && ((addr1 & mask2) == (addr2 & mask2))) {
+		/* Filter 1 resides within filter 2, so swap filters */
+		bcmasp_set_mda_filter(intf, addr, mask, i);
+		return 0;
+	}
+
+	/* Unable to combine */
+	return -EINVAL;
+}
+
+int bcmasp_set_en_mda_filter(struct bcmasp_intf *intf, unsigned char *addr,
+			     unsigned char *mask)
+{
+	struct bcmasp_priv *priv = intf->parent;
+	int ret, res_count;
+	unsigned int i;
+
+	res_count = bcmasp_total_res_mda_cnt(intf->parent);
+
+	for (i = res_count; i < priv->num_mda_filters; i++) {
+		/* If filter not enabled or belongs to another port skip */
+		if (!priv->mda_filters[i].en ||
+		    priv->mda_filters[i].port != intf->port)
+			continue;
+
+		/* Attempt to combine filters */
+		ret = bcmasp_combine_set_filter(intf, addr, mask, i);
+		if (!ret) {
+			intf->mib.filters_combine_cnt++;
+			return 0;
+		}
+	}
+
+	/* Create new filter if possible */
+	for (i = res_count; i < priv->num_mda_filters; i++) {
+		if (priv->mda_filters[i].en)
+			continue;
+
+		bcmasp_set_mda_filter(intf, addr, mask, i);
+		bcmasp_en_mda_filter(intf, 1, i);
+		return 0;
+	}
+
+	/* No room for new filter */
+	return -EINVAL;
+}
+
+static void bcmasp_core_init_filters(struct bcmasp_priv *priv)
+{
+	unsigned int i;
+
+	/* Disable all filters and reset software view since the HW
+	 * can lose context while in deep sleep suspend states
+	 */
+	for (i = 0; i < priv->num_mda_filters; i++) {
+		rx_filter_core_wl(priv, 0x0, ASP_RX_FILTER_MDA_CFG(i));
+		priv->mda_filters[i].en = 0;
+	}
+
+	for (i = 0; i < priv->num_net_filters; i++)
+		rx_filter_core_wl(priv, 0x0, ASP_RX_FILTER_NET_CFG(i));
+
+	/* Top level filter enable bit should be enabled at all times, set
+	 * GEN_WAKE_CLEAR to clear the network filter wake-up which would
+	 * otherwise be sticky
+	 */
+	rx_filter_core_wl(priv, (ASP_RX_FILTER_OPUT_EN |
+			  ASP_RX_FILTER_MDA_EN |
+			  ASP_RX_FILTER_GEN_WK_CLR |
+			  ASP_RX_FILTER_NT_FLT_EN),
+			  ASP_RX_FILTER_BLK_CTRL);
+}
+
+/* ASP core initialization */
+static void bcmasp_core_init(struct bcmasp_priv *priv)
+{
+	rx_edpkt_core_wl(priv, 0x1b, ASP_EDPKT_BURST_BUF_PSCAL_TOUT);
+	rx_edpkt_core_wl(priv, 0x3e8, ASP_EDPKT_BURST_BUF_WRITE_TOUT);
+
+	rx_edpkt_core_wl(priv, ASP_EDPKT_ENABLE_EN, ASP_EDPKT_ENABLE);
+
+	/* Disable and clear both UniMAC's wake-up interrupts to avoid
+	 * sticky interrupts.
+	 */
+	_intr2_mask_set(priv, ASP_INTR2_UMC0_WAKE | ASP_INTR2_UMC1_WAKE);
+	intr2_core_wl(priv, ASP_INTR2_UMC0_WAKE | ASP_INTR2_UMC1_WAKE,
+		      ASP_INTR2_CLEAR);
+}
+
+static void bcmasp_core_clock_select_many(struct bcmasp_priv *priv, bool slow)
+{
+	u32 reg;
+
+	reg = ctrl2_core_rl(priv, ASP_CTRL2_CORE_CLOCK_SELECT);
+	if (slow)
+		reg &= ~ASP_CTRL2_CORE_CLOCK_SELECT_MAIN;
+	else
+		reg |= ASP_CTRL2_CORE_CLOCK_SELECT_MAIN;
+	ctrl2_core_wl(priv, reg, ASP_CTRL2_CORE_CLOCK_SELECT);
+
+	reg = ctrl2_core_rl(priv, ASP_CTRL2_CPU_CLOCK_SELECT);
+	if (slow)
+		reg &= ~ASP_CTRL2_CPU_CLOCK_SELECT_MAIN;
+	else
+		reg |= ASP_CTRL2_CPU_CLOCK_SELECT_MAIN;
+	ctrl2_core_wl(priv, reg, ASP_CTRL2_CPU_CLOCK_SELECT);
+}
+
+static void bcmasp_core_clock_select_one(struct bcmasp_priv *priv, bool slow)
+{
+	u32 reg;
+
+	reg = ctrl_core_rl(priv, ASP_CTRL_CORE_CLOCK_SELECT);
+	if (slow)
+		reg &= ~ASP_CTRL_CORE_CLOCK_SELECT_MAIN;
+	else
+		reg |= ASP_CTRL_CORE_CLOCK_SELECT_MAIN;
+	ctrl_core_wl(priv, reg, ASP_CTRL_CORE_CLOCK_SELECT);
+}
+
+static void bcmasp_core_clock_select_one_ctrl2(struct bcmasp_priv *priv, bool slow)
+{
+	u32 reg;
+
+	reg = ctrl2_core_rl(priv, ASP_CTRL2_CORE_CLOCK_SELECT);
+	if (slow)
+		reg &= ~ASP_CTRL2_CORE_CLOCK_SELECT_MAIN;
+	else
+		reg |= ASP_CTRL2_CORE_CLOCK_SELECT_MAIN;
+	ctrl2_core_wl(priv, reg, ASP_CTRL2_CORE_CLOCK_SELECT);
+}
+
+static void bcmasp_core_clock_set_ll(struct bcmasp_priv *priv, u32 clr, u32 set)
+{
+	u32 reg;
+
+	reg = ctrl_core_rl(priv, ASP_CTRL_CLOCK_CTRL);
+	reg &= ~clr;
+	reg |= set;
+	ctrl_core_wl(priv, reg, ASP_CTRL_CLOCK_CTRL);
+
+	reg = ctrl_core_rl(priv, ASP_CTRL_SCRATCH_0);
+	reg &= ~clr;
+	reg |= set;
+	ctrl_core_wl(priv, reg, ASP_CTRL_SCRATCH_0);
+}
+
+static void bcmasp_core_clock_set(struct bcmasp_priv *priv, u32 clr, u32 set)
+{
+	unsigned long flags;
+
+	spin_lock_irqsave(&priv->clk_lock, flags);
+	bcmasp_core_clock_set_ll(priv, clr, set);
+	spin_unlock_irqrestore(&priv->clk_lock, flags);
+}
+
+void bcmasp_core_clock_set_intf(struct bcmasp_intf *intf, bool en)
+{
+	u32 intf_mask = ASP_CTRL_CLOCK_CTRL_ASP_RGMII_DIS(intf->port);
+	struct bcmasp_priv *priv = intf->parent;
+	unsigned long flags;
+	u32 reg;
+
+	/* When enabling an interface, if the RX or TX clocks were not enabled,
+	 * enable them. Conversely, while disabling an interface, if this is
+	 * the last one enabled, we can turn off the shared RX and TX clocks as
+	 * well. We control enable bits which is why we test for equality on
+	 * the RGMII clock bit mask.
+	 */
+	spin_lock_irqsave(&priv->clk_lock, flags);
+	if (en) {
+		intf_mask |= ASP_CTRL_CLOCK_CTRL_ASP_TX_DISABLE |
+			     ASP_CTRL_CLOCK_CTRL_ASP_RX_DISABLE;
+		bcmasp_core_clock_set_ll(priv, intf_mask, 0);
+	} else {
+		reg = ctrl_core_rl(priv, ASP_CTRL_SCRATCH_0) | intf_mask;
+		if ((reg & ASP_CTRL_CLOCK_CTRL_ASP_RGMII_MASK) ==
+		    ASP_CTRL_CLOCK_CTRL_ASP_RGMII_MASK)
+			intf_mask |= ASP_CTRL_CLOCK_CTRL_ASP_TX_DISABLE |
+				     ASP_CTRL_CLOCK_CTRL_ASP_RX_DISABLE;
+		bcmasp_core_clock_set_ll(priv, 0, intf_mask);
+	}
+	spin_unlock_irqrestore(&priv->clk_lock, flags);
+}
+
+static irqreturn_t bcmasp_isr_wol(int irq, void *data)
+{
+	struct bcmasp_priv *priv = data;
+	u32 status;
+
+	/* No L3 IRQ, so we good */
+	if (priv->wol_irq <= 0)
+		goto irq_handled;
+
+	status = wakeup_intr2_core_rl(priv, ASP_WAKEUP_INTR2_STATUS) &
+		~wakeup_intr2_core_rl(priv, ASP_WAKEUP_INTR2_MASK_STATUS);
+	wakeup_intr2_core_wl(priv, status, ASP_WAKEUP_INTR2_CLEAR);
+
+irq_handled:
+	pm_wakeup_event(&priv->pdev->dev, 0);
+	return IRQ_HANDLED;
+}
+
+static int bcmasp_get_and_request_irq(struct bcmasp_priv *priv, int i)
+{
+	struct platform_device *pdev = priv->pdev;
+	int irq, ret;
+
+	irq = platform_get_irq_optional(pdev, i);
+	if (irq < 0)
+		return irq;
+
+	ret = devm_request_irq(&pdev->dev, irq, bcmasp_isr_wol, 0,
+			       pdev->name, priv);
+	if (ret)
+		return ret;
+
+	return irq;
+}
+
+static void bcmasp_init_wol(struct bcmasp_priv *priv)
+{
+	struct platform_device *pdev = priv->pdev;
+	struct device *dev = &pdev->dev;
+	int irq;
+
+	irq = bcmasp_get_and_request_irq(priv, 1);
+	if (irq < 0) {
+		dev_warn(dev, "Failed to init WoL irq: %d\n", irq);
+		return;
+	}
+
+	priv->wol_irq = irq;
+	priv->wol_irq_enabled_mask = 0;
+	device_set_wakeup_capable(&pdev->dev, 1);
+}
+
+void bcmasp_enable_wol(struct bcmasp_intf *intf, bool en)
+{
+	struct bcmasp_priv *priv = intf->parent;
+	struct device *dev = &priv->pdev->dev;
+
+	if (en) {
+		if (priv->wol_irq_enabled_mask) {
+			set_bit(intf->port, &priv->wol_irq_enabled_mask);
+			return;
+		}
+
+		/* First enable */
+		set_bit(intf->port, &priv->wol_irq_enabled_mask);
+		enable_irq_wake(priv->wol_irq);
+		device_set_wakeup_enable(dev, 1);
+	} else {
+		if (!priv->wol_irq_enabled_mask)
+			return;
+
+		clear_bit(intf->port, &priv->wol_irq_enabled_mask);
+		if (priv->wol_irq_enabled_mask)
+			return;
+
+		/* Last disable */
+		disable_irq_wake(priv->wol_irq);
+		device_set_wakeup_enable(dev, 0);
+	}
+}
+
+static void bcmasp_wol_irq_destroy(struct bcmasp_priv *priv)
+{
+	if (priv->wol_irq > 0)
+		free_irq(priv->wol_irq, priv);
+}
+
+static void bcmasp_eee_fixup(struct bcmasp_intf *intf, bool en)
+{
+	u32 reg, phy_lpi_overwrite;
+
+	reg = rx_edpkt_core_rl(intf->parent, ASP_EDPKT_SPARE_REG);
+	phy_lpi_overwrite = intf->internal_phy ? ASP_EDPKT_SPARE_REG_EPHY_LPI :
+			    ASP_EDPKT_SPARE_REG_GPHY_LPI;
+
+	if (en)
+		reg |= phy_lpi_overwrite;
+	else
+		reg &= ~phy_lpi_overwrite;
+
+	rx_edpkt_core_wl(intf->parent, reg, ASP_EDPKT_SPARE_REG);
+
+	usleep_range(50, 100);
+}
+
+static const struct bcmasp_plat_data v21_plat_data = {
+	.core_clock_select = bcmasp_core_clock_select_one,
+	.num_mda_filters = 32,
+	.num_net_filters = 32,
+	.tx_chan_offset = 8,
+	.rx_ctrl_offset = 0x0,
+};
+
+static const struct bcmasp_plat_data v22_plat_data = {
+	.core_clock_select = bcmasp_core_clock_select_many,
+	.eee_fixup = bcmasp_eee_fixup,
+	.num_mda_filters = 32,
+	.num_net_filters = 32,
+	.tx_chan_offset = 8,
+	.rx_ctrl_offset = 0x0,
+};
+
+static const struct bcmasp_plat_data v30_plat_data = {
+	.core_clock_select = bcmasp_core_clock_select_one_ctrl2,
+	.num_mda_filters = 20,
+	.num_net_filters = 16,
+	.tx_chan_offset = 0,
+	.rx_ctrl_offset = 0x10000,
+};
+
+static void bcmasp_set_pdata(struct bcmasp_priv *priv, const struct bcmasp_plat_data *pdata)
+{
+	priv->core_clock_select = pdata->core_clock_select;
+	priv->eee_fixup = pdata->eee_fixup;
+	priv->num_mda_filters = pdata->num_mda_filters;
+	priv->num_net_filters = pdata->num_net_filters;
+	priv->tx_chan_offset = pdata->tx_chan_offset;
+	priv->rx_ctrl_offset = pdata->rx_ctrl_offset;
+}
+
+static const struct of_device_id bcmasp_of_match[] = {
+	{ .compatible = "brcm,asp-v2.1", .data = &v21_plat_data },
+	{ .compatible = "brcm,asp-v2.2", .data = &v22_plat_data },
+	{ .compatible = "brcm,asp-v3.0", .data = &v30_plat_data },
+	{ /* sentinel */ },
+};
+MODULE_DEVICE_TABLE(of, bcmasp_of_match);
+
+static const struct of_device_id bcmasp_mdio_of_match[] = {
+	{ .compatible = "brcm,asp-v2.1-mdio", },
+	{ .compatible = "brcm,asp-v2.2-mdio", },
+	{ .compatible = "brcm,asp-v3.0-mdio", },
+	{ /* sentinel */ },
+};
+MODULE_DEVICE_TABLE(of, bcmasp_mdio_of_match);
+
+static void bcmasp_remove_intfs(struct bcmasp_priv *priv)
+{
+	struct bcmasp_intf *intf, *n;
+
+	list_for_each_entry_safe(intf, n, &priv->intfs, list) {
+		list_del(&intf->list);
+		bcmasp_interface_destroy(intf);
+	}
+}
+
+static int bcmasp_probe(struct platform_device *pdev)
+{
+	const struct bcmasp_plat_data *pdata;
+	struct device *dev = &pdev->dev;
+	struct device_node *ports_node;
+	struct bcmasp_priv *priv;
+	struct bcmasp_intf *intf;
+	int ret = 0, count = 0;
+	unsigned int i;
+
+	priv = devm_kzalloc(dev, sizeof(*priv), GFP_KERNEL);
+	if (!priv)
+		return -ENOMEM;
+
+	priv->irq = platform_get_irq(pdev, 0);
+	if (priv->irq <= 0)
+		return -EINVAL;
+
+	priv->clk = devm_clk_get_optional_enabled(dev, "sw_asp");
+	if (IS_ERR(priv->clk))
+		return dev_err_probe(dev, PTR_ERR(priv->clk),
+				     "failed to request clock\n");
+
+	/* Base from parent node */
+	priv->base = devm_platform_ioremap_resource(pdev, 0);
+	if (IS_ERR(priv->base))
+		return dev_err_probe(dev, PTR_ERR(priv->base), "failed to iomap\n");
+
+	ret = dma_set_mask_and_coherent(&pdev->dev, DMA_BIT_MASK(40));
+	if (ret)
+		return dev_err_probe(dev, ret, "unable to set DMA mask: %d\n", ret);
+
+	dev_set_drvdata(&pdev->dev, priv);
+	priv->pdev = pdev;
+	spin_lock_init(&priv->mda_lock);
+	spin_lock_init(&priv->clk_lock);
+	mutex_init(&priv->wol_lock);
+	mutex_init(&priv->net_lock);
+	INIT_LIST_HEAD(&priv->intfs);
+
+	pdata = device_get_match_data(&pdev->dev);
+	if (!pdata)
+		return dev_err_probe(dev, -EINVAL, "unable to find platform data\n");
+
+	bcmasp_set_pdata(priv, pdata);
+
+	/* Enable all clocks to ensure successful probing */
+	bcmasp_core_clock_set(priv, ASP_CTRL_CLOCK_CTRL_ASP_ALL_DISABLE, 0);
+
+	/* Switch to the main clock */
+	priv->core_clock_select(priv, false);
+
+	bcmasp_intr2_mask_set_all(priv);
+	bcmasp_intr2_clear_all(priv);
+
+	ret = devm_request_irq(&pdev->dev, priv->irq, bcmasp_isr, 0,
+			       pdev->name, priv);
+	if (ret)
+		return dev_err_probe(dev, ret, "failed to request ASP interrupt: %d", ret);
+
+	/* Register mdio child nodes */
+	of_platform_populate(dev->of_node, bcmasp_mdio_of_match, NULL, dev);
+
+	/* ASP specific initialization, Needs to be done regardless of
+	 * how many interfaces come up.
+	 */
+	bcmasp_core_init(priv);
+
+	priv->mda_filters = devm_kcalloc(dev, priv->num_mda_filters,
+					 sizeof(*priv->mda_filters), GFP_KERNEL);
+	if (!priv->mda_filters)
+		return -ENOMEM;
+
+	priv->net_filters = devm_kcalloc(dev, priv->num_net_filters,
+					 sizeof(*priv->net_filters), GFP_KERNEL);
+	if (!priv->net_filters)
+		return -ENOMEM;
+
+	bcmasp_core_init_filters(priv);
+
+	ports_node = of_find_node_by_name(dev->of_node, "ethernet-ports");
+	if (!ports_node) {
+		dev_warn(dev, "No ports found\n");
+		return -EINVAL;
+	}
+
+	i = 0;
+	for_each_available_child_of_node_scoped(ports_node, intf_node) {
+		intf = bcmasp_interface_create(priv, intf_node, i);
+		if (!intf) {
+			dev_err(dev, "Cannot create eth interface %d\n", i);
+			bcmasp_remove_intfs(priv);
+			ret = -ENOMEM;
+			goto of_put_exit;
+		}
+		list_add_tail(&intf->list, &priv->intfs);
+		i++;
+	}
+
+	/* Check and enable WoL */
+	bcmasp_init_wol(priv);
+
+	/* Drop the clock reference count now and let ndo_open()/ndo_close()
+	 * manage it for us from now on.
+	 */
+	bcmasp_core_clock_set(priv, 0, ASP_CTRL_CLOCK_CTRL_ASP_ALL_DISABLE);
+
+	clk_disable_unprepare(priv->clk);
+
+	/* Now do the registration of the network ports which will take care
+	 * of managing the clock properly.
+	 */
+	list_for_each_entry(intf, &priv->intfs, list) {
+		ret = register_netdev(intf->ndev);
+		if (ret) {
+			netdev_err(intf->ndev,
+				   "failed to register net_device: %d\n", ret);
+			bcmasp_wol_irq_destroy(priv);
+			bcmasp_remove_intfs(priv);
+			goto of_put_exit;
+		}
+		count++;
+	}
+
+	dev_info(dev, "Initialized %d port(s)\n", count);
+
+of_put_exit:
+	of_node_put(ports_node);
+	return ret;
+}
+
+static void bcmasp_remove(struct platform_device *pdev)
+{
+	struct bcmasp_priv *priv = dev_get_drvdata(&pdev->dev);
+
+	if (!priv)
+		return;
+
+	bcmasp_wol_irq_destroy(priv);
+	bcmasp_remove_intfs(priv);
+}
+
+static void bcmasp_shutdown(struct platform_device *pdev)
+{
+	bcmasp_remove(pdev);
+}
+
+static int __maybe_unused bcmasp_suspend(struct device *d)
+{
+	struct bcmasp_priv *priv = dev_get_drvdata(d);
+	struct bcmasp_intf *intf;
+	int ret;
+
+	list_for_each_entry(intf, &priv->intfs, list) {
+		ret = bcmasp_interface_suspend(intf);
+		if (ret)
+			break;
+	}
+
+	ret = clk_prepare_enable(priv->clk);
+	if (ret)
+		return ret;
+
+	/* Whether Wake-on-LAN is enabled or not, we can always disable
+	 * the shared TX clock
+	 */
+	bcmasp_core_clock_set(priv, 0, ASP_CTRL_CLOCK_CTRL_ASP_TX_DISABLE);
+
+	priv->core_clock_select(priv, true);
+
+	clk_disable_unprepare(priv->clk);
+
+	return ret;
+}
+
+static int __maybe_unused bcmasp_resume(struct device *d)
+{
+	struct bcmasp_priv *priv = dev_get_drvdata(d);
+	struct bcmasp_intf *intf;
+	int ret;
+
+	ret = clk_prepare_enable(priv->clk);
+	if (ret)
+		return ret;
+
+	/* Switch to the main clock domain */
+	priv->core_clock_select(priv, false);
+
+	/* Re-enable all clocks for re-initialization */
+	bcmasp_core_clock_set(priv, ASP_CTRL_CLOCK_CTRL_ASP_ALL_DISABLE, 0);
+
+	bcmasp_core_init(priv);
+	bcmasp_core_init_filters(priv);
+
+	/* And disable them to let the network devices take care of them */
+	bcmasp_core_clock_set(priv, 0, ASP_CTRL_CLOCK_CTRL_ASP_ALL_DISABLE);
+
+	clk_disable_unprepare(priv->clk);
+
+	list_for_each_entry(intf, &priv->intfs, list) {
+		ret = bcmasp_interface_resume(intf);
+		if (ret)
+			break;
+	}
+
+	return ret;
+}
+
+static SIMPLE_DEV_PM_OPS(bcmasp_pm_ops,
+			 bcmasp_suspend, bcmasp_resume);
+
+static struct platform_driver bcmasp_driver = {
+	.probe = bcmasp_probe,
+	.remove = bcmasp_remove,
+	.shutdown = bcmasp_shutdown,
+	.driver = {
+		.name = "brcm,asp-v2",
+		.of_match_table = bcmasp_of_match,
+		.pm = &bcmasp_pm_ops,
+	},
+};
+module_platform_driver(bcmasp_driver);
+
+MODULE_DESCRIPTION("Broadcom ASP 2.0 Ethernet controller driver");
+MODULE_ALIAS("platform:brcm,asp-v2");
+MODULE_LICENSE("GPL");
diff --git a/drivers/net/ethernet/broadcom/asp2/bcmasp.h b/drivers/net/ethernet/broadcom/asp2/bcmasp.h
new file mode 100644
index 000000000000..74adfdb50e11
--- /dev/null
+++ b/drivers/net/ethernet/broadcom/asp2/bcmasp.h
@@ -0,0 +1,600 @@
+/* SPDX-License-Identifier: GPL-2.0 */
+#ifndef __BCMASP_H
+#define __BCMASP_H
+
+#include <linux/netdevice.h>
+#include <linux/phy.h>
+#include <linux/io-64-nonatomic-hi-lo.h>
+#include <uapi/linux/ethtool.h>
+
+#define ASP_INTR2_OFFSET			0x1000
+#define  ASP_INTR2_STATUS			0x0
+#define  ASP_INTR2_SET				0x4
+#define  ASP_INTR2_CLEAR			0x8
+#define  ASP_INTR2_MASK_STATUS			0xc
+#define  ASP_INTR2_MASK_SET			0x10
+#define  ASP_INTR2_MASK_CLEAR			0x14
+
+#define ASP_INTR2_RX_ECH(intr)			BIT(intr)
+#define ASP_INTR2_TX_DESC(intr)			BIT((intr) + 14)
+#define ASP_INTR2_UMC0_WAKE			BIT(22)
+#define ASP_INTR2_UMC1_WAKE			BIT(28)
+#define ASP_INTR2_PHY_EVENT(intr)		((intr) ? BIT(30) | BIT(31) : \
+						BIT(24) | BIT(25))
+
+#define ASP_WAKEUP_INTR2_OFFSET			0x1200
+#define  ASP_WAKEUP_INTR2_STATUS		0x0
+#define  ASP_WAKEUP_INTR2_SET			0x4
+#define  ASP_WAKEUP_INTR2_CLEAR			0x8
+#define  ASP_WAKEUP_INTR2_MASK_STATUS		0xc
+#define  ASP_WAKEUP_INTR2_MASK_SET		0x10
+#define  ASP_WAKEUP_INTR2_MASK_CLEAR		0x14
+#define ASP_WAKEUP_INTR2_MPD_0			BIT(0)
+#define ASP_WAKEUP_INTR2_MPD_1			BIT(1)
+#define ASP_WAKEUP_INTR2_FILT_0			BIT(2)
+#define ASP_WAKEUP_INTR2_FILT_1			BIT(3)
+#define ASP_WAKEUP_INTR2_FW			BIT(4)
+
+#define ASP_CTRL2_OFFSET			0x2000
+#define  ASP_CTRL2_CORE_CLOCK_SELECT		0x0
+#define   ASP_CTRL2_CORE_CLOCK_SELECT_MAIN	BIT(0)
+#define  ASP_CTRL2_CPU_CLOCK_SELECT		0x4
+#define   ASP_CTRL2_CPU_CLOCK_SELECT_MAIN	BIT(0)
+
+#define ASP_TX_ANALYTICS_OFFSET			0x4c000
+#define  ASP_TX_ANALYTICS_CTRL			0x0
+
+#define ASP_RX_ANALYTICS_OFFSET			0x98000
+#define  ASP_RX_ANALYTICS_CTRL			0x0
+
+#define ASP_RX_CTRL_OFFSET			0x9f000
+#define ASP_RX_CTRL_UMAC_0_FRAME_COUNT		0x8
+#define ASP_RX_CTRL_UMAC_1_FRAME_COUNT		0xc
+#define ASP_RX_CTRL_FB_0_FRAME_COUNT		0x14
+#define ASP_RX_CTRL_FB_1_FRAME_COUNT		0x18
+#define ASP_RX_CTRL_FB_8_FRAME_COUNT		0x1c
+#define ASP_RX_CTRL_FB_9_FRAME_COUNT		0x20
+#define ASP_RX_CTRL_FB_10_FRAME_COUNT		0x24
+#define ASP_RX_CTRL_FB_OUT_FRAME_COUNT		0x28
+#define ASP_RX_CTRL_FB_FILT_OUT_FRAME_COUNT	0x2c
+#define ASP_RX_CTRL_FLUSH			0x30
+#define  ASP_CTRL_UMAC0_FLUSH_MASK             (BIT(0) | BIT(12))
+#define  ASP_CTRL_UMAC1_FLUSH_MASK             (BIT(1) | BIT(13))
+#define  ASP_CTRL_SPB_FLUSH_MASK               (BIT(8) | BIT(20))
+#define ASP_RX_CTRL_FB_RX_FIFO_DEPTH		0x38
+
+#define ASP_RX_FILTER_OFFSET			0x80000
+#define  ASP_RX_FILTER_BLK_CTRL			0x0
+#define   ASP_RX_FILTER_OPUT_EN			BIT(0)
+#define   ASP_RX_FILTER_MDA_EN			BIT(1)
+#define   ASP_RX_FILTER_LNR_MD			BIT(2)
+#define   ASP_RX_FILTER_GEN_WK_EN		BIT(3)
+#define   ASP_RX_FILTER_GEN_WK_CLR		BIT(4)
+#define   ASP_RX_FILTER_NT_FLT_EN		BIT(5)
+#define  ASP_RX_FILTER_MDA_CFG(sel)		(((sel) * 0x14) + 0x100)
+#define   ASP_RX_FILTER_MDA_CFG_EN_SHIFT	8
+#define   ASP_RX_FILTER_MDA_CFG_UMC_SEL(sel)	((sel) > 1 ? BIT(17) : \
+						 BIT((sel) + 9))
+#define  ASP_RX_FILTER_MDA_PAT_H(sel)		(((sel) * 0x14) + 0x104)
+#define  ASP_RX_FILTER_MDA_PAT_L(sel)		(((sel) * 0x14) + 0x108)
+#define  ASP_RX_FILTER_MDA_MSK_H(sel)		(((sel) * 0x14) + 0x10c)
+#define  ASP_RX_FILTER_MDA_MSK_L(sel)		(((sel) * 0x14) + 0x110)
+#define  ASP_RX_FILTER_MDA_CFG(sel)		(((sel) * 0x14) + 0x100)
+#define  ASP_RX_FILTER_MDA_PAT_H(sel)		(((sel) * 0x14) + 0x104)
+#define  ASP_RX_FILTER_MDA_PAT_L(sel)		(((sel) * 0x14) + 0x108)
+#define  ASP_RX_FILTER_MDA_MSK_H(sel)		(((sel) * 0x14) + 0x10c)
+#define  ASP_RX_FILTER_MDA_MSK_L(sel)		(((sel) * 0x14) + 0x110)
+#define  ASP_RX_FILTER_NET_CFG(sel)		(((sel) * 0xa04) + 0x400)
+#define   ASP_RX_FILTER_NET_CFG_CH(sel)		((sel) << 0)
+#define   ASP_RX_FILTER_NET_CFG_EN		BIT(9)
+#define   ASP_RX_FILTER_NET_CFG_L2_EN		BIT(10)
+#define   ASP_RX_FILTER_NET_CFG_L3_EN		BIT(11)
+#define   ASP_RX_FILTER_NET_CFG_L4_EN		BIT(12)
+#define   ASP_RX_FILTER_NET_CFG_L3_FRM(sel)	((sel) << 13)
+#define   ASP_RX_FILTER_NET_CFG_L4_FRM(sel)	((sel) << 15)
+#define   ASP_RX_FILTER_NET_CFG_UMC(sel)	BIT((sel) + 19)
+#define   ASP_RX_FILTER_NET_CFG_DMA_EN		BIT(27)
+
+#define  ASP_RX_FILTER_NET_OFFSET_MAX		32
+#define  ASP_RX_FILTER_NET_PAT(sel, block, off) \
+		(((sel) * 0xa04) + ((block) * 0x200) + (off) + 0x600)
+#define  ASP_RX_FILTER_NET_MASK(sel, block, off) \
+		(((sel) * 0xa04) + ((block) * 0x200) + (off) + 0x700)
+
+#define  ASP_RX_FILTER_NET_OFFSET(sel)		(((sel) * 0xa04) + 0xe00)
+#define   ASP_RX_FILTER_NET_OFFSET_L2(val)	((val) << 0)
+#define   ASP_RX_FILTER_NET_OFFSET_L3_0(val)	((val) << 8)
+#define   ASP_RX_FILTER_NET_OFFSET_L3_1(val)	((val) << 16)
+#define   ASP_RX_FILTER_NET_OFFSET_L4(val)	((val) << 24)
+
+enum asp_rx_net_filter_block {
+	ASP_RX_FILTER_NET_L2 = 0,
+	ASP_RX_FILTER_NET_L3_0,
+	ASP_RX_FILTER_NET_L3_1,
+	ASP_RX_FILTER_NET_L4,
+	ASP_RX_FILTER_NET_BLOCK_MAX
+};
+
+#define ASP_EDPKT_OFFSET			0x9c000
+#define  ASP_EDPKT_ENABLE			0x4
+#define   ASP_EDPKT_ENABLE_EN			BIT(0)
+#define  ASP_EDPKT_HDR_CFG			0xc
+#define   ASP_EDPKT_HDR_SZ_SHIFT		2
+#define   ASP_EDPKT_HDR_SZ_32			0
+#define   ASP_EDPKT_HDR_SZ_64			1
+#define   ASP_EDPKT_HDR_SZ_96			2
+#define   ASP_EDPKT_HDR_SZ_128			3
+#define ASP_EDPKT_BURST_BUF_PSCAL_TOUT		0x10
+#define ASP_EDPKT_BURST_BUF_WRITE_TOUT		0x14
+#define ASP_EDPKT_BURST_BUF_READ_TOUT		0x18
+#define ASP_EDPKT_RX_TS_COUNTER			0x38
+#define  ASP_EDPKT_ENDI				0x48
+#define   ASP_EDPKT_ENDI_DESC_SHIFT		8
+#define   ASP_EDPKT_ENDI_NO_BT_SWP		0
+#define   ASP_EDPKT_ENDI_BT_SWP_WD		1
+#define ASP_EDPKT_RX_PKT_CNT			0x138
+#define ASP_EDPKT_HDR_EXTR_CNT			0x13c
+#define ASP_EDPKT_HDR_OUT_CNT			0x140
+#define ASP_EDPKT_SPARE_REG			0x174
+#define  ASP_EDPKT_SPARE_REG_EPHY_LPI		BIT(4)
+#define  ASP_EDPKT_SPARE_REG_GPHY_LPI		BIT(3)
+
+#define ASP_CTRL_OFFSET				0x101000
+#define  ASP_CTRL_ASP_SW_INIT			0x04
+#define   ASP_CTRL_ASP_SW_INIT_ACPUSS_CORE	BIT(0)
+#define   ASP_CTRL_ASP_SW_INIT_ASP_TX		BIT(1)
+#define   ASP_CTRL_ASP_SW_INIT_AS_RX		BIT(2)
+#define   ASP_CTRL_ASP_SW_INIT_ASP_RGMII_UMAC0	BIT(3)
+#define   ASP_CTRL_ASP_SW_INIT_ASP_RGMII_UMAC1	BIT(4)
+#define   ASP_CTRL_ASP_SW_INIT_ASP_XMEMIF	BIT(5)
+#define  ASP_CTRL_CLOCK_CTRL			0x04
+#define   ASP_CTRL_CLOCK_CTRL_ASP_TX_DISABLE	BIT(0)
+#define   ASP_CTRL_CLOCK_CTRL_ASP_RX_DISABLE	BIT(1)
+#define   ASP_CTRL_CLOCK_CTRL_ASP_RGMII_SHIFT	2
+#define   ASP_CTRL_CLOCK_CTRL_ASP_RGMII_MASK	(0x7 << ASP_CTRL_CLOCK_CTRL_ASP_RGMII_SHIFT)
+#define   ASP_CTRL_CLOCK_CTRL_ASP_RGMII_DIS(x)	BIT(ASP_CTRL_CLOCK_CTRL_ASP_RGMII_SHIFT + (x))
+#define   ASP_CTRL_CLOCK_CTRL_ASP_ALL_DISABLE	GENMASK(4, 0)
+#define  ASP_CTRL_CORE_CLOCK_SELECT		0x08
+#define   ASP_CTRL_CORE_CLOCK_SELECT_MAIN	BIT(0)
+#define  ASP_CTRL_SCRATCH_0			0x0c
+
+struct bcmasp_tx_cb {
+	struct sk_buff		*skb;
+	unsigned int		bytes_sent;
+	bool			last;
+
+	DEFINE_DMA_UNMAP_ADDR(dma_addr);
+	DEFINE_DMA_UNMAP_LEN(dma_len);
+};
+
+struct bcmasp_res {
+	/* Per interface resources */
+	/* Port */
+	void __iomem		*umac;
+	void __iomem		*umac2fb;
+	void __iomem		*rgmii;
+
+	/* TX slowpath/configuration */
+	void __iomem		*tx_spb_ctrl;
+	void __iomem		*tx_spb_top;
+	void __iomem		*tx_epkt_core;
+	void __iomem		*tx_pause_ctrl;
+};
+
+#define DESC_ADDR(x)		((x) & GENMASK_ULL(39, 0))
+#define DESC_FLAGS(x)		((x) & GENMASK_ULL(63, 40))
+
+struct bcmasp_desc {
+	u64		buf;
+	#define DESC_CHKSUM	BIT_ULL(40)
+	#define DESC_CRC_ERR	BIT_ULL(41)
+	#define DESC_RX_SYM_ERR	BIT_ULL(42)
+	#define DESC_NO_OCT_ALN BIT_ULL(43)
+	#define DESC_PKT_TRUC	BIT_ULL(44)
+	/*  39:0 (TX/RX) bits 0-39 of buf addr
+	 *    40 (RX) checksum
+	 *    41 (RX) crc_error
+	 *    42 (RX) rx_symbol_error
+	 *    43 (RX) non_octet_aligned
+	 *    44 (RX) pkt_truncated
+	 *    45 Reserved
+	 * 56:46 (RX) mac_filter_id
+	 * 60:57 (RX) rx_port_num (0-unicmac0, 1-unimac1)
+	 *    61 Reserved
+	 * 63:62 (TX) forward CRC, overwrite CRC
+	 */
+	u32		size;
+	u32		flags;
+	#define DESC_INT_EN     BIT(0)
+	#define DESC_SOF	BIT(1)
+	#define DESC_EOF	BIT(2)
+	#define DESC_EPKT_CMD   BIT(3)
+	#define DESC_SCRAM_ST   BIT(8)
+	#define DESC_SCRAM_END  BIT(9)
+	#define DESC_PCPP       BIT(10)
+	#define DESC_PPPP       BIT(11)
+	/*     0 (TX) tx_int_en
+	 *     1 (TX/RX) SOF
+	 *     2 (TX/RX) EOF
+	 *     3 (TX) epkt_command
+	 *   6:4 (TX) PA
+	 *     7 (TX) pause at desc end
+	 *     8 (TX) scram_start
+	 *     9 (TX) scram_end
+	 *    10 (TX) PCPP
+	 *    11 (TX) PPPP
+	 * 14:12 Reserved
+	 *    15 (TX) pid ch Valid
+	 * 19:16 (TX) data_pkt_type
+	 * 32:20 (TX) pid_channel (RX) nw_filter_id
+	 */
+};
+
+struct bcmasp_intf;
+
+struct bcmasp_intf_stats64 {
+	/* Rx Stats */
+	u64_stats_t	rx_packets;
+	u64_stats_t	rx_bytes;
+	u64_stats_t	rx_errors;
+	u64_stats_t	rx_dropped;
+	u64_stats_t	rx_crc_errs;
+	u64_stats_t	rx_sym_errs;
+
+	/* Tx Stats*/
+	u64_stats_t	tx_packets;
+	u64_stats_t	tx_bytes;
+
+	struct u64_stats_sync		syncp;
+};
+
+struct bcmasp_mib_counters {
+	u32	edpkt_ts;
+	u32	edpkt_rx_pkt_cnt;
+	u32	edpkt_hdr_ext_cnt;
+	u32	edpkt_hdr_out_cnt;
+	u32	umac_frm_cnt;
+	u32	fb_frm_cnt;
+	u32	fb_rx_fifo_depth;
+	u32	fb_out_frm_cnt;
+	u32	fb_filt_out_frm_cnt;
+	u32	alloc_rx_skb_failed;
+	u32	tx_dma_failed;
+	u32	mc_filters_full_cnt;
+	u32	uc_filters_full_cnt;
+	u32	filters_combine_cnt;
+	u32	promisc_filters_cnt;
+	u32	tx_realloc_offload_failed;
+	u32	tx_timeout_cnt;
+};
+
+struct bcmasp_intf_ops {
+	unsigned long (*rx_desc_read)(struct bcmasp_intf *intf);
+	void (*rx_buffer_write)(struct bcmasp_intf *intf, dma_addr_t addr);
+	void (*rx_desc_write)(struct bcmasp_intf *intf, dma_addr_t addr);
+	unsigned long (*tx_read)(struct bcmasp_intf *intf);
+	void (*tx_write)(struct bcmasp_intf *intf, dma_addr_t addr);
+};
+
+struct bcmasp_priv;
+
+struct bcmasp_intf {
+	struct list_head		list;
+	struct net_device		*ndev;
+	struct bcmasp_priv		*parent;
+
+	/* ASP Ch */
+	int				channel;
+	int				port;
+	const struct bcmasp_intf_ops	*ops;
+
+	/* Used for splitting shared resources */
+	int				index;
+
+	struct napi_struct		tx_napi;
+	/* TX ring, starts on a new cacheline boundary */
+	void __iomem			*tx_spb_dma;
+	int				tx_spb_index;
+	int				tx_spb_clean_index;
+	struct bcmasp_desc		*tx_spb_cpu;
+	dma_addr_t			tx_spb_dma_addr;
+	dma_addr_t			tx_spb_dma_valid;
+	dma_addr_t			tx_spb_dma_read;
+	struct bcmasp_tx_cb		*tx_cbs;
+
+	/* RX ring, starts on a new cacheline boundary */
+	void __iomem			*rx_edpkt_cfg;
+	void __iomem			*rx_edpkt_dma;
+	int				rx_edpkt_index;
+	int				rx_buf_order;
+	struct bcmasp_desc		*rx_edpkt_cpu;
+	dma_addr_t			rx_edpkt_dma_addr;
+	dma_addr_t			rx_edpkt_dma_read;
+	dma_addr_t			rx_edpkt_dma_valid;
+
+	/* RX buffer prefetcher ring*/
+	void				*rx_ring_cpu;
+	dma_addr_t			rx_ring_dma;
+	dma_addr_t			rx_ring_dma_valid;
+	struct napi_struct		rx_napi;
+
+	struct bcmasp_res		res;
+	unsigned int			crc_fwd;
+
+	/* PHY device */
+	struct device_node		*phy_dn;
+	struct device_node		*ndev_dn;
+	phy_interface_t			phy_interface;
+	bool				internal_phy;
+	int				old_pause;
+	int				old_link;
+	int				old_duplex;
+
+	u32				msg_enable;
+
+	/* Statistics */
+	struct bcmasp_intf_stats64	stats64;
+	struct bcmasp_mib_counters	mib;
+
+	u32				wolopts;
+	u8				sopass[SOPASS_MAX];
+};
+
+#define NUM_NET_FILTERS				32
+struct bcmasp_net_filter {
+	struct ethtool_rx_flow_spec	fs;
+
+	bool				claimed;
+	bool				wake_filter;
+
+	int				port;
+	unsigned int			hw_index;
+};
+
+#define NUM_MDA_FILTERS				32
+struct bcmasp_mda_filter {
+	/* Current owner of this filter */
+	int		port;
+	bool		en;
+	u8		addr[ETH_ALEN];
+	u8		mask[ETH_ALEN];
+};
+
+struct bcmasp_plat_data {
+	void (*core_clock_select)(struct bcmasp_priv *priv, bool slow);
+	void (*eee_fixup)(struct bcmasp_intf *priv, bool en);
+	unsigned int num_mda_filters;
+	unsigned int num_net_filters;
+	unsigned int tx_chan_offset;
+	unsigned int rx_ctrl_offset;
+};
+
+struct bcmasp_priv {
+	struct platform_device		*pdev;
+	struct clk			*clk;
+
+	int				irq;
+	u32				irq_mask;
+
+	/* Used if shared wol irq */
+	struct mutex			wol_lock;
+	int				wol_irq;
+	unsigned long			wol_irq_enabled_mask;
+
+	void (*core_clock_select)(struct bcmasp_priv *priv, bool slow);
+	void (*eee_fixup)(struct bcmasp_intf *intf, bool en);
+	unsigned int			num_mda_filters;
+	unsigned int			num_net_filters;
+	unsigned int			tx_chan_offset;
+	unsigned int			rx_ctrl_offset;
+
+	void __iomem			*base;
+
+	struct list_head		intfs;
+
+	struct bcmasp_mda_filter	*mda_filters;
+
+	/* MAC destination address filters lock */
+	spinlock_t			mda_lock;
+
+	/* Protects accesses to ASP_CTRL_CLOCK_CTRL */
+	spinlock_t			clk_lock;
+
+	struct bcmasp_net_filter	*net_filters;
+
+	/* Network filter lock */
+	struct mutex			net_lock;
+};
+
+static inline unsigned long bcmasp_intf_rx_desc_read(struct bcmasp_intf *intf)
+{
+	return intf->ops->rx_desc_read(intf);
+}
+
+static inline void bcmasp_intf_rx_buffer_write(struct bcmasp_intf *intf,
+					       dma_addr_t addr)
+{
+	intf->ops->rx_buffer_write(intf, addr);
+}
+
+static inline void bcmasp_intf_rx_desc_write(struct bcmasp_intf *intf,
+					     dma_addr_t addr)
+{
+	intf->ops->rx_desc_write(intf, addr);
+}
+
+static inline unsigned long bcmasp_intf_tx_read(struct bcmasp_intf *intf)
+{
+	return intf->ops->tx_read(intf);
+}
+
+static inline void bcmasp_intf_tx_write(struct bcmasp_intf *intf,
+					dma_addr_t addr)
+{
+	intf->ops->tx_write(intf, addr);
+}
+
+#define __BCMASP_IO_MACRO(name, m)					\
+static inline u32 name##_rl(struct bcmasp_intf *intf, u32 off)		\
+{									\
+	u32 reg = readl_relaxed(intf->m + off);				\
+	return reg;							\
+}									\
+static inline void name##_wl(struct bcmasp_intf *intf, u32 val, u32 off)\
+{									\
+	writel_relaxed(val, intf->m + off);				\
+}
+
+#define BCMASP_IO_MACRO(name)		__BCMASP_IO_MACRO(name, res.name)
+#define BCMASP_FP_IO_MACRO(name)	__BCMASP_IO_MACRO(name, name)
+
+BCMASP_IO_MACRO(umac);
+BCMASP_IO_MACRO(umac2fb);
+BCMASP_IO_MACRO(rgmii);
+BCMASP_FP_IO_MACRO(tx_spb_dma);
+BCMASP_IO_MACRO(tx_spb_ctrl);
+BCMASP_IO_MACRO(tx_spb_top);
+BCMASP_IO_MACRO(tx_epkt_core);
+BCMASP_IO_MACRO(tx_pause_ctrl);
+BCMASP_FP_IO_MACRO(rx_edpkt_dma);
+BCMASP_FP_IO_MACRO(rx_edpkt_cfg);
+
+#define __BCMASP_FP_IO_MACRO_Q(name, m)					\
+static inline u64 name##_rq(struct bcmasp_intf *intf, u32 off)		\
+{									\
+	u64 reg = readq_relaxed(intf->m + off);				\
+	return reg;							\
+}									\
+static inline void name##_wq(struct bcmasp_intf *intf, u64 val, u32 off)\
+{									\
+	writeq_relaxed(val, intf->m + off);				\
+}
+
+#define BCMASP_FP_IO_MACRO_Q(name)	__BCMASP_FP_IO_MACRO_Q(name, name)
+
+BCMASP_FP_IO_MACRO_Q(tx_spb_dma);
+BCMASP_FP_IO_MACRO_Q(rx_edpkt_dma);
+BCMASP_FP_IO_MACRO_Q(rx_edpkt_cfg);
+
+#define PKT_OFFLOAD_NOP			(0 << 28)
+#define PKT_OFFLOAD_HDR_OP		(1 << 28)
+#define  PKT_OFFLOAD_HDR_WRBACK		BIT(19)
+#define  PKT_OFFLOAD_HDR_COUNT(x)	((x) << 16)
+#define  PKT_OFFLOAD_HDR_SIZE_1(x)	((x) << 4)
+#define  PKT_OFFLOAD_HDR_SIZE_2(x)	(x)
+#define  PKT_OFFLOAD_HDR2_SIZE_2(x)	((x) << 24)
+#define  PKT_OFFLOAD_HDR2_SIZE_3(x)	((x) << 12)
+#define  PKT_OFFLOAD_HDR2_SIZE_4(x)	(x)
+#define PKT_OFFLOAD_EPKT_OP		(2 << 28)
+#define  PKT_OFFLOAD_EPKT_WRBACK	BIT(23)
+#define  PKT_OFFLOAD_EPKT_IP(x)		((x) << 21)
+#define  PKT_OFFLOAD_EPKT_TP(x)		((x) << 19)
+#define  PKT_OFFLOAD_EPKT_LEN(x)	((x) << 16)
+#define  PKT_OFFLOAD_EPKT_CSUM_L4	BIT(15)
+#define  PKT_OFFLOAD_EPKT_CSUM_L3	BIT(14)
+#define  PKT_OFFLOAD_EPKT_ID(x)		((x) << 12)
+#define  PKT_OFFLOAD_EPKT_SEQ(x)	((x) << 10)
+#define  PKT_OFFLOAD_EPKT_TS(x)		((x) << 8)
+#define  PKT_OFFLOAD_EPKT_BLOC(x)	(x)
+#define PKT_OFFLOAD_END_OP		(7 << 28)
+
+struct bcmasp_pkt_offload {
+	__be32		nop;
+	__be32		header;
+	__be32		header2;
+	__be32		epkt;
+	__be32		end;
+};
+
+#define BCMASP_CORE_IO_MACRO(name, offset)				\
+static inline u32 name##_core_rl(struct bcmasp_priv *priv,		\
+				 u32 off)				\
+{									\
+	u32 reg = readl_relaxed(priv->base + (offset) + off);		\
+	return reg;							\
+}									\
+static inline void name##_core_wl(struct bcmasp_priv *priv,		\
+				  u32 val, u32 off)			\
+{									\
+	writel_relaxed(val, priv->base + (offset) + off);		\
+}
+
+BCMASP_CORE_IO_MACRO(intr2, ASP_INTR2_OFFSET);
+BCMASP_CORE_IO_MACRO(wakeup_intr2, ASP_WAKEUP_INTR2_OFFSET);
+BCMASP_CORE_IO_MACRO(tx_analytics, ASP_TX_ANALYTICS_OFFSET);
+BCMASP_CORE_IO_MACRO(rx_analytics, ASP_RX_ANALYTICS_OFFSET);
+BCMASP_CORE_IO_MACRO(rx_filter, ASP_RX_FILTER_OFFSET);
+BCMASP_CORE_IO_MACRO(rx_edpkt, ASP_EDPKT_OFFSET);
+BCMASP_CORE_IO_MACRO(ctrl, ASP_CTRL_OFFSET);
+BCMASP_CORE_IO_MACRO(ctrl2, ASP_CTRL2_OFFSET);
+
+#define BCMASP_CORE_IO_MACRO_OFFSET(name, offset)			\
+static inline u32 name##_core_rl(struct bcmasp_priv *priv,		\
+				 u32 off)				\
+{									\
+	u32 reg = readl_relaxed(priv->base + priv->name##_offset +	\
+				(offset) + off);			\
+	return reg;							\
+}									\
+static inline void name##_core_wl(struct bcmasp_priv *priv,		\
+				  u32 val, u32 off)			\
+{									\
+	writel_relaxed(val, priv->base + priv->name##_offset +		\
+		       (offset) + off);					\
+}
+BCMASP_CORE_IO_MACRO_OFFSET(rx_ctrl, ASP_RX_CTRL_OFFSET);
+
+struct bcmasp_intf *bcmasp_interface_create(struct bcmasp_priv *priv,
+					    struct device_node *ndev_dn, int i);
+
+void bcmasp_interface_destroy(struct bcmasp_intf *intf);
+
+void bcmasp_enable_tx_irq(struct bcmasp_intf *intf, int en);
+
+void bcmasp_enable_rx_irq(struct bcmasp_intf *intf, int en);
+
+void bcmasp_enable_phy_irq(struct bcmasp_intf *intf, int en);
+
+void bcmasp_flush_rx_port(struct bcmasp_intf *intf);
+
+extern const struct ethtool_ops bcmasp_ethtool_ops;
+
+int bcmasp_interface_suspend(struct bcmasp_intf *intf);
+
+int bcmasp_interface_resume(struct bcmasp_intf *intf);
+
+void bcmasp_set_promisc(struct bcmasp_intf *intf, bool en);
+
+void bcmasp_set_allmulti(struct bcmasp_intf *intf, bool en);
+
+void bcmasp_set_broad(struct bcmasp_intf *intf, bool en);
+
+void bcmasp_set_oaddr(struct bcmasp_intf *intf, const unsigned char *addr,
+		      bool en);
+
+int bcmasp_set_en_mda_filter(struct bcmasp_intf *intf, unsigned char *addr,
+			     unsigned char *mask);
+
+void bcmasp_disable_all_filters(struct bcmasp_intf *intf);
+
+void bcmasp_core_clock_set_intf(struct bcmasp_intf *intf, bool en);
+
+struct bcmasp_net_filter *bcmasp_netfilt_get_init(struct bcmasp_intf *intf,
+						  u32 loc, bool wake_filter,
+						  bool init);
+
+bool bcmasp_netfilt_check_dup(struct bcmasp_intf *intf,
+			      struct ethtool_rx_flow_spec *fs);
+
+void bcmasp_netfilt_release(struct bcmasp_intf *intf,
+			    struct bcmasp_net_filter *nfilt);
+
+int bcmasp_netfilt_get_active(struct bcmasp_intf *intf);
+
+int bcmasp_netfilt_get_all_active(struct bcmasp_intf *intf, u32 *rule_locs,
+				  u32 *rule_cnt);
+
+void bcmasp_netfilt_suspend(struct bcmasp_intf *intf);
+
+void bcmasp_enable_wol(struct bcmasp_intf *intf, bool en);
+#endif
diff --git a/drivers/net/ethernet/broadcom/asp2/bcmasp_ethtool.c b/drivers/net/ethernet/broadcom/asp2/bcmasp_ethtool.c
new file mode 100644
index 000000000000..63f1a8c3a7fb
--- /dev/null
+++ b/drivers/net/ethernet/broadcom/asp2/bcmasp_ethtool.c
@@ -0,0 +1,434 @@
+// SPDX-License-Identifier: GPL-2.0
+#define pr_fmt(fmt)				"bcmasp_ethtool: " fmt
+
+#include <linux/unaligned.h>
+#include <linux/ethtool.h>
+#include <linux/netdevice.h>
+#include <linux/platform_device.h>
+
+#include "bcmasp.h"
+#include "bcmasp_intf_defs.h"
+
+enum bcmasp_stat_type {
+	BCMASP_STAT_RX_CTRL,
+	BCMASP_STAT_RX_CTRL_PER_INTF,
+	BCMASP_STAT_SOFT,
+};
+
+struct bcmasp_stats {
+	char stat_string[ETH_GSTRING_LEN];
+	enum bcmasp_stat_type type;
+	u32 reg_offset;
+};
+
+#define STAT_BCMASP_SOFT_MIB(str) { \
+	.stat_string = str, \
+	.type = BCMASP_STAT_SOFT, \
+}
+
+#define STAT_BCMASP_OFFSET(str, _type, offset) { \
+	.stat_string = str, \
+	.type = _type, \
+	.reg_offset = offset, \
+}
+
+#define STAT_BCMASP_RX_CTRL(str, offset) \
+	STAT_BCMASP_OFFSET(str, BCMASP_STAT_RX_CTRL, offset)
+#define STAT_BCMASP_RX_CTRL_PER_INTF(str, offset) \
+	STAT_BCMASP_OFFSET(str, BCMASP_STAT_RX_CTRL_PER_INTF, offset)
+
+/* Must match the order of struct bcmasp_mib_counters */
+static const struct bcmasp_stats bcmasp_gstrings_stats[] = {
+	/* ASP RX control */
+	STAT_BCMASP_RX_CTRL_PER_INTF("Frames From Unimac",
+				     ASP_RX_CTRL_UMAC_0_FRAME_COUNT),
+	STAT_BCMASP_RX_CTRL_PER_INTF("Frames From Port",
+				     ASP_RX_CTRL_FB_0_FRAME_COUNT),
+	STAT_BCMASP_RX_CTRL_PER_INTF("RX Buffer FIFO Depth",
+				     ASP_RX_CTRL_FB_RX_FIFO_DEPTH),
+	STAT_BCMASP_RX_CTRL("Frames Out(Buffer)",
+			    ASP_RX_CTRL_FB_OUT_FRAME_COUNT),
+	STAT_BCMASP_RX_CTRL("Frames Out(Filters)",
+			    ASP_RX_CTRL_FB_FILT_OUT_FRAME_COUNT),
+	/* Software maintained statistics */
+	STAT_BCMASP_SOFT_MIB("RX SKB Alloc Failed"),
+	STAT_BCMASP_SOFT_MIB("TX DMA Failed"),
+	STAT_BCMASP_SOFT_MIB("Multicast Filters Full"),
+	STAT_BCMASP_SOFT_MIB("Unicast Filters Full"),
+	STAT_BCMASP_SOFT_MIB("MDA Filters Combined"),
+	STAT_BCMASP_SOFT_MIB("Promisc Filter Set"),
+	STAT_BCMASP_SOFT_MIB("TX Realloc For Offload Failed"),
+	STAT_BCMASP_SOFT_MIB("Tx Timeout Count"),
+};
+
+#define BCMASP_STATS_LEN	ARRAY_SIZE(bcmasp_gstrings_stats)
+
+static int bcmasp_get_sset_count(struct net_device *dev, int string_set)
+{
+	switch (string_set) {
+	case ETH_SS_STATS:
+		return BCMASP_STATS_LEN;
+	default:
+		return -EOPNOTSUPP;
+	}
+}
+
+static void bcmasp_get_strings(struct net_device *dev, u32 stringset,
+			       u8 *data)
+{
+	const char *str;
+	unsigned int i;
+
+	switch (stringset) {
+	case ETH_SS_STATS:
+		for (i = 0; i < BCMASP_STATS_LEN; i++) {
+			str = bcmasp_gstrings_stats[i].stat_string;
+			ethtool_puts(&data, str);
+		}
+		break;
+	default:
+		return;
+	}
+}
+
+static void bcmasp_update_mib_counters(struct bcmasp_intf *intf)
+{
+	unsigned int i;
+
+	for (i = 0; i < BCMASP_STATS_LEN; i++) {
+		const struct bcmasp_stats *s;
+		u32 offset, val;
+		char *p;
+
+		s = &bcmasp_gstrings_stats[i];
+		offset = s->reg_offset;
+		switch (s->type) {
+		case BCMASP_STAT_SOFT:
+			continue;
+		case BCMASP_STAT_RX_CTRL:
+			val = rx_ctrl_core_rl(intf->parent, offset);
+			break;
+		case BCMASP_STAT_RX_CTRL_PER_INTF:
+			offset += sizeof(u32) * intf->port;
+			val = rx_ctrl_core_rl(intf->parent, offset);
+			break;
+		default:
+			continue;
+		}
+		p = (char *)(&intf->mib) + (i * sizeof(u32));
+		put_unaligned(val, (u32 *)p);
+	}
+}
+
+static void bcmasp_get_ethtool_stats(struct net_device *dev,
+				     struct ethtool_stats *stats,
+				     u64 *data)
+{
+	struct bcmasp_intf *intf = netdev_priv(dev);
+	unsigned int i;
+	char *p;
+
+	if (netif_running(dev))
+		bcmasp_update_mib_counters(intf);
+
+	for (i = 0; i < BCMASP_STATS_LEN; i++) {
+		p = (char *)(&intf->mib) + (i * sizeof(u32));
+		data[i] = *(u32 *)p;
+	}
+}
+
+static void bcmasp_get_drvinfo(struct net_device *dev,
+			       struct ethtool_drvinfo *info)
+{
+	strscpy(info->driver, "bcmasp", sizeof(info->driver));
+	strscpy(info->bus_info, dev_name(dev->dev.parent),
+		sizeof(info->bus_info));
+}
+
+static u32 bcmasp_get_msglevel(struct net_device *dev)
+{
+	struct bcmasp_intf *intf = netdev_priv(dev);
+
+	return intf->msg_enable;
+}
+
+static void bcmasp_set_msglevel(struct net_device *dev, u32 level)
+{
+	struct bcmasp_intf *intf = netdev_priv(dev);
+
+	intf->msg_enable = level;
+}
+
+#define BCMASP_SUPPORTED_WAKE   (WAKE_MAGIC | WAKE_MAGICSECURE | WAKE_FILTER)
+static void bcmasp_get_wol(struct net_device *dev, struct ethtool_wolinfo *wol)
+{
+	struct bcmasp_intf *intf = netdev_priv(dev);
+
+	wol->supported = BCMASP_SUPPORTED_WAKE;
+	wol->wolopts = intf->wolopts;
+	memset(wol->sopass, 0, sizeof(wol->sopass));
+
+	if (wol->wolopts & WAKE_MAGICSECURE)
+		memcpy(wol->sopass, intf->sopass, sizeof(intf->sopass));
+}
+
+static int bcmasp_set_wol(struct net_device *dev, struct ethtool_wolinfo *wol)
+{
+	struct bcmasp_intf *intf = netdev_priv(dev);
+	struct bcmasp_priv *priv = intf->parent;
+	struct device *kdev = &priv->pdev->dev;
+
+	if (!device_can_wakeup(kdev))
+		return -EOPNOTSUPP;
+
+	/* Interface Specific */
+	intf->wolopts = wol->wolopts;
+	if (intf->wolopts & WAKE_MAGICSECURE)
+		memcpy(intf->sopass, wol->sopass, sizeof(wol->sopass));
+
+	mutex_lock(&priv->wol_lock);
+	bcmasp_enable_wol(intf, !!intf->wolopts);
+	mutex_unlock(&priv->wol_lock);
+
+	return 0;
+}
+
+static int bcmasp_flow_insert(struct net_device *dev, struct ethtool_rxnfc *cmd)
+{
+	struct bcmasp_intf *intf = netdev_priv(dev);
+	struct bcmasp_net_filter *nfilter;
+	u32 loc = cmd->fs.location;
+	bool wake = false;
+
+	if (cmd->fs.ring_cookie == RX_CLS_FLOW_WAKE)
+		wake = true;
+
+	/* Currently only supports WAKE filters */
+	if (!wake)
+		return -EOPNOTSUPP;
+
+	switch (cmd->fs.flow_type & ~(FLOW_EXT | FLOW_MAC_EXT)) {
+	case ETHER_FLOW:
+	case IP_USER_FLOW:
+	case TCP_V4_FLOW:
+	case UDP_V4_FLOW:
+	case TCP_V6_FLOW:
+	case UDP_V6_FLOW:
+		break;
+	default:
+		return -EOPNOTSUPP;
+	}
+
+	/* Check if filter already exists */
+	if (bcmasp_netfilt_check_dup(intf, &cmd->fs))
+		return -EINVAL;
+
+	nfilter = bcmasp_netfilt_get_init(intf, loc, wake, true);
+	if (IS_ERR(nfilter))
+		return PTR_ERR(nfilter);
+
+	/* Return the location where we did insert the filter */
+	cmd->fs.location = nfilter->hw_index;
+	memcpy(&nfilter->fs, &cmd->fs, sizeof(struct ethtool_rx_flow_spec));
+
+	/* Since we only support wake filters, defer register programming till
+	 * suspend time.
+	 */
+	return 0;
+}
+
+static int bcmasp_flow_delete(struct net_device *dev, struct ethtool_rxnfc *cmd)
+{
+	struct bcmasp_intf *intf = netdev_priv(dev);
+	struct bcmasp_net_filter *nfilter;
+
+	nfilter = bcmasp_netfilt_get_init(intf, cmd->fs.location, false, false);
+	if (IS_ERR(nfilter))
+		return PTR_ERR(nfilter);
+
+	bcmasp_netfilt_release(intf, nfilter);
+
+	return 0;
+}
+
+static int bcmasp_flow_get(struct bcmasp_intf *intf, struct ethtool_rxnfc *cmd)
+{
+	struct bcmasp_net_filter *nfilter;
+
+	nfilter = bcmasp_netfilt_get_init(intf, cmd->fs.location, false, false);
+	if (IS_ERR(nfilter))
+		return PTR_ERR(nfilter);
+
+	memcpy(&cmd->fs, &nfilter->fs, sizeof(nfilter->fs));
+
+	cmd->data = intf->parent->num_net_filters;
+
+	return 0;
+}
+
+static int bcmasp_set_rxnfc(struct net_device *dev, struct ethtool_rxnfc *cmd)
+{
+	struct bcmasp_intf *intf = netdev_priv(dev);
+	int ret = -EOPNOTSUPP;
+
+	mutex_lock(&intf->parent->net_lock);
+
+	switch (cmd->cmd) {
+	case ETHTOOL_SRXCLSRLINS:
+		ret = bcmasp_flow_insert(dev, cmd);
+		break;
+	case ETHTOOL_SRXCLSRLDEL:
+		ret = bcmasp_flow_delete(dev, cmd);
+		break;
+	default:
+		break;
+	}
+
+	mutex_unlock(&intf->parent->net_lock);
+
+	return ret;
+}
+
+static int bcmasp_get_rxnfc(struct net_device *dev, struct ethtool_rxnfc *cmd,
+			    u32 *rule_locs)
+{
+	struct bcmasp_intf *intf = netdev_priv(dev);
+	int err = 0;
+
+	mutex_lock(&intf->parent->net_lock);
+
+	switch (cmd->cmd) {
+	case ETHTOOL_GRXCLSRLCNT:
+		cmd->rule_cnt = bcmasp_netfilt_get_active(intf);
+		/* We support specifying rule locations */
+		cmd->data |= RX_CLS_LOC_SPECIAL;
+		break;
+	case ETHTOOL_GRXCLSRULE:
+		err = bcmasp_flow_get(intf, cmd);
+		break;
+	case ETHTOOL_GRXCLSRLALL:
+		err = bcmasp_netfilt_get_all_active(intf, rule_locs, &cmd->rule_cnt);
+		cmd->data = intf->parent->num_net_filters;
+		break;
+	default:
+		err = -EOPNOTSUPP;
+		break;
+	}
+
+	mutex_unlock(&intf->parent->net_lock);
+
+	return err;
+}
+
+static int bcmasp_get_eee(struct net_device *dev, struct ethtool_keee *e)
+{
+	if (!dev->phydev)
+		return -ENODEV;
+
+	return phy_ethtool_get_eee(dev->phydev, e);
+}
+
+static int bcmasp_set_eee(struct net_device *dev, struct ethtool_keee *e)
+{
+	if (!dev->phydev)
+		return -ENODEV;
+
+	return phy_ethtool_set_eee(dev->phydev, e);
+}
+
+static void bcmasp_get_eth_mac_stats(struct net_device *dev,
+				     struct ethtool_eth_mac_stats *mac_stats)
+{
+	struct bcmasp_intf *intf = netdev_priv(dev);
+
+	mac_stats->FramesTransmittedOK = umac_rl(intf, UMC_GTPOK);
+	mac_stats->SingleCollisionFrames = umac_rl(intf, UMC_GTSCL);
+	mac_stats->MultipleCollisionFrames = umac_rl(intf, UMC_GTMCL);
+	mac_stats->FramesReceivedOK = umac_rl(intf, UMC_GRPOK);
+	mac_stats->FrameCheckSequenceErrors = umac_rl(intf, UMC_GRFCS);
+	mac_stats->AlignmentErrors = umac_rl(intf, UMC_GRALN);
+	mac_stats->OctetsTransmittedOK = umac_rl(intf, UMC_GTBYT);
+	mac_stats->FramesWithDeferredXmissions = umac_rl(intf, UMC_GTDRF);
+	mac_stats->LateCollisions = umac_rl(intf, UMC_GTLCL);
+	mac_stats->FramesAbortedDueToXSColls = umac_rl(intf, UMC_GTXCL);
+	mac_stats->OctetsReceivedOK = umac_rl(intf, UMC_GRBYT);
+	mac_stats->MulticastFramesXmittedOK = umac_rl(intf, UMC_GTMCA);
+	mac_stats->BroadcastFramesXmittedOK = umac_rl(intf, UMC_GTBCA);
+	mac_stats->FramesWithExcessiveDeferral = umac_rl(intf, UMC_GTEDF);
+	mac_stats->MulticastFramesReceivedOK = umac_rl(intf, UMC_GRMCA);
+	mac_stats->BroadcastFramesReceivedOK = umac_rl(intf, UMC_GRBCA);
+}
+
+static const struct ethtool_rmon_hist_range bcmasp_rmon_ranges[] = {
+	{    0,   64},
+	{   65,  127},
+	{  128,  255},
+	{  256,  511},
+	{  512, 1023},
+	{ 1024, 1518},
+	{ 1519, 1522},
+	{}
+};
+
+static void bcmasp_get_rmon_stats(struct net_device *dev,
+				  struct ethtool_rmon_stats *rmon_stats,
+				  const struct ethtool_rmon_hist_range **ranges)
+{
+	struct bcmasp_intf *intf = netdev_priv(dev);
+
+	*ranges = bcmasp_rmon_ranges;
+
+	rmon_stats->undersize_pkts = umac_rl(intf, UMC_RRUND);
+	rmon_stats->oversize_pkts = umac_rl(intf, UMC_GROVR);
+	rmon_stats->fragments = umac_rl(intf, UMC_RRFRG);
+	rmon_stats->jabbers = umac_rl(intf, UMC_GRJBR);
+
+	rmon_stats->hist[0] = umac_rl(intf, UMC_GR64);
+	rmon_stats->hist[1] = umac_rl(intf, UMC_GR127);
+	rmon_stats->hist[2] = umac_rl(intf, UMC_GR255);
+	rmon_stats->hist[3] = umac_rl(intf, UMC_GR511);
+	rmon_stats->hist[4] = umac_rl(intf, UMC_GR1023);
+	rmon_stats->hist[5] = umac_rl(intf, UMC_GR1518);
+	rmon_stats->hist[6] = umac_rl(intf, UMC_GRMGV);
+
+	rmon_stats->hist_tx[0] = umac_rl(intf, UMC_TR64);
+	rmon_stats->hist_tx[1] = umac_rl(intf, UMC_TR127);
+	rmon_stats->hist_tx[2] = umac_rl(intf, UMC_TR255);
+	rmon_stats->hist_tx[3] = umac_rl(intf, UMC_TR511);
+	rmon_stats->hist_tx[4] = umac_rl(intf, UMC_TR1023);
+	rmon_stats->hist_tx[5] = umac_rl(intf, UMC_TR1518);
+	rmon_stats->hist_tx[6] = umac_rl(intf, UMC_TRMGV);
+}
+
+static void bcmasp_get_eth_ctrl_stats(struct net_device *dev,
+				      struct ethtool_eth_ctrl_stats *ctrl_stats)
+{
+	struct bcmasp_intf *intf = netdev_priv(dev);
+
+	ctrl_stats->MACControlFramesTransmitted = umac_rl(intf, UMC_GTXCF);
+	ctrl_stats->MACControlFramesReceived = umac_rl(intf, UMC_GRXCF);
+	ctrl_stats->UnsupportedOpcodesReceived = umac_rl(intf, UMC_GRXUO);
+}
+
+const struct ethtool_ops bcmasp_ethtool_ops = {
+	.get_drvinfo		= bcmasp_get_drvinfo,
+	.get_link		= ethtool_op_get_link,
+	.get_link_ksettings	= phy_ethtool_get_link_ksettings,
+	.set_link_ksettings	= phy_ethtool_set_link_ksettings,
+	.get_msglevel		= bcmasp_get_msglevel,
+	.set_msglevel		= bcmasp_set_msglevel,
+	.get_wol		= bcmasp_get_wol,
+	.set_wol		= bcmasp_set_wol,
+	.get_rxnfc		= bcmasp_get_rxnfc,
+	.set_rxnfc		= bcmasp_set_rxnfc,
+	.set_eee		= bcmasp_set_eee,
+	.get_eee		= bcmasp_get_eee,
+	.get_eth_mac_stats	= bcmasp_get_eth_mac_stats,
+	.get_rmon_stats		= bcmasp_get_rmon_stats,
+	.get_eth_ctrl_stats	= bcmasp_get_eth_ctrl_stats,
+	.get_strings		= bcmasp_get_strings,
+	.get_ethtool_stats	= bcmasp_get_ethtool_stats,
+	.get_sset_count		= bcmasp_get_sset_count,
+	.get_ts_info		= ethtool_op_get_ts_info,
+	.nway_reset		= phy_ethtool_nway_reset,
+};
diff --git a/drivers/net/ethernet/broadcom/asp2/bcmasp_intf.c b/drivers/net/ethernet/broadcom/asp2/bcmasp_intf.c
new file mode 100644
index 000000000000..b9973956c480
--- /dev/null
+++ b/drivers/net/ethernet/broadcom/asp2/bcmasp_intf.c
@@ -0,0 +1,1423 @@
+// SPDX-License-Identifier: GPL-2.0
+#define pr_fmt(fmt)			"bcmasp_intf: " fmt
+
+#include <asm/byteorder.h>
+#include <linux/brcmphy.h>
+#include <linux/clk.h>
+#include <linux/delay.h>
+#include <linux/etherdevice.h>
+#include <linux/netdevice.h>
+#include <linux/of_net.h>
+#include <linux/of_mdio.h>
+#include <linux/phy.h>
+#include <linux/phy_fixed.h>
+#include <linux/ptp_classify.h>
+#include <linux/platform_device.h>
+#include <net/ip.h>
+#include <net/ipv6.h>
+
+#include "bcmasp.h"
+#include "bcmasp_intf_defs.h"
+
+static int incr_ring(int index, int ring_count)
+{
+	index++;
+	if (index == ring_count)
+		return 0;
+
+	return index;
+}
+
+/* Points to last byte of descriptor */
+static dma_addr_t incr_last_byte(dma_addr_t addr, dma_addr_t beg,
+				 int ring_count)
+{
+	dma_addr_t end = beg + (ring_count * DESC_SIZE);
+
+	addr += DESC_SIZE;
+	if (addr > end)
+		return beg + DESC_SIZE - 1;
+
+	return addr;
+}
+
+/* Points to first byte of descriptor */
+static dma_addr_t incr_first_byte(dma_addr_t addr, dma_addr_t beg,
+				  int ring_count)
+{
+	dma_addr_t end = beg + (ring_count * DESC_SIZE);
+
+	addr += DESC_SIZE;
+	if (addr >= end)
+		return beg;
+
+	return addr;
+}
+
+static void bcmasp_enable_tx(struct bcmasp_intf *intf, int en)
+{
+	if (en) {
+		tx_spb_ctrl_wl(intf, TX_SPB_CTRL_ENABLE_EN, TX_SPB_CTRL_ENABLE);
+		tx_epkt_core_wl(intf, (TX_EPKT_C_CFG_MISC_EN |
+				TX_EPKT_C_CFG_MISC_PT |
+				(intf->port << TX_EPKT_C_CFG_MISC_PS_SHIFT)),
+				TX_EPKT_C_CFG_MISC);
+	} else {
+		tx_spb_ctrl_wl(intf, 0x0, TX_SPB_CTRL_ENABLE);
+		tx_epkt_core_wl(intf, 0x0, TX_EPKT_C_CFG_MISC);
+	}
+}
+
+static void bcmasp_enable_rx(struct bcmasp_intf *intf, int en)
+{
+	if (en)
+		rx_edpkt_cfg_wl(intf, RX_EDPKT_CFG_ENABLE_EN,
+				RX_EDPKT_CFG_ENABLE);
+	else
+		rx_edpkt_cfg_wl(intf, 0x0, RX_EDPKT_CFG_ENABLE);
+}
+
+static void bcmasp_set_rx_mode(struct net_device *dev)
+{
+	unsigned char mask[] = {0xff, 0xff, 0xff, 0xff, 0xff, 0xff};
+	struct bcmasp_intf *intf = netdev_priv(dev);
+	struct netdev_hw_addr *ha;
+	int ret;
+
+	spin_lock_bh(&intf->parent->mda_lock);
+
+	bcmasp_disable_all_filters(intf);
+
+	if (dev->flags & IFF_PROMISC)
+		goto set_promisc;
+
+	bcmasp_set_promisc(intf, 0);
+
+	bcmasp_set_broad(intf, 1);
+
+	bcmasp_set_oaddr(intf, dev->dev_addr, 1);
+
+	if (dev->flags & IFF_ALLMULTI) {
+		bcmasp_set_allmulti(intf, 1);
+	} else {
+		bcmasp_set_allmulti(intf, 0);
+
+		netdev_for_each_mc_addr(ha, dev) {
+			ret = bcmasp_set_en_mda_filter(intf, ha->addr, mask);
+			if (ret) {
+				intf->mib.mc_filters_full_cnt++;
+				goto set_promisc;
+			}
+		}
+	}
+
+	netdev_for_each_uc_addr(ha, dev) {
+		ret = bcmasp_set_en_mda_filter(intf, ha->addr, mask);
+		if (ret) {
+			intf->mib.uc_filters_full_cnt++;
+			goto set_promisc;
+		}
+	}
+
+	spin_unlock_bh(&intf->parent->mda_lock);
+	return;
+
+set_promisc:
+	bcmasp_set_promisc(intf, 1);
+	intf->mib.promisc_filters_cnt++;
+
+	/* disable all filters used by this port */
+	bcmasp_disable_all_filters(intf);
+
+	spin_unlock_bh(&intf->parent->mda_lock);
+}
+
+static void bcmasp_clean_txcb(struct bcmasp_intf *intf, int index)
+{
+	struct bcmasp_tx_cb *txcb = &intf->tx_cbs[index];
+
+	txcb->skb = NULL;
+	dma_unmap_addr_set(txcb, dma_addr, 0);
+	dma_unmap_len_set(txcb, dma_len, 0);
+	txcb->last = false;
+}
+
+static int tx_spb_ring_full(struct bcmasp_intf *intf, int cnt)
+{
+	int next_index, i;
+
+	/* Check if we have enough room for cnt descriptors */
+	for (i = 0; i < cnt; i++) {
+		next_index = incr_ring(intf->tx_spb_index, DESC_RING_COUNT);
+		if (next_index == intf->tx_spb_clean_index)
+			return 1;
+	}
+
+	return 0;
+}
+
+static struct sk_buff *bcmasp_csum_offload(struct net_device *dev,
+					   struct sk_buff *skb,
+					   bool *csum_hw)
+{
+	struct bcmasp_intf *intf = netdev_priv(dev);
+	u32 header = 0, header2 = 0, epkt = 0;
+	struct bcmasp_pkt_offload *offload;
+	unsigned int header_cnt = 0;
+	u8 ip_proto;
+	int ret;
+
+	if (skb->ip_summed != CHECKSUM_PARTIAL)
+		return skb;
+
+	ret = skb_cow_head(skb, sizeof(*offload));
+	if (ret < 0) {
+		intf->mib.tx_realloc_offload_failed++;
+		goto help;
+	}
+
+	switch (skb->protocol) {
+	case htons(ETH_P_IP):
+		header |= PKT_OFFLOAD_HDR_SIZE_2((ip_hdrlen(skb) >> 8) & 0xf);
+		header2 |= PKT_OFFLOAD_HDR2_SIZE_2(ip_hdrlen(skb) & 0xff);
+		epkt |= PKT_OFFLOAD_EPKT_IP(0);
+		ip_proto = ip_hdr(skb)->protocol;
+		header_cnt += 2;
+		break;
+	case htons(ETH_P_IPV6):
+		header |= PKT_OFFLOAD_HDR_SIZE_2((IP6_HLEN >> 8) & 0xf);
+		header2 |= PKT_OFFLOAD_HDR2_SIZE_2(IP6_HLEN & 0xff);
+		epkt |= PKT_OFFLOAD_EPKT_IP(1);
+		ip_proto = ipv6_hdr(skb)->nexthdr;
+		header_cnt += 2;
+		break;
+	default:
+		goto help;
+	}
+
+	switch (ip_proto) {
+	case IPPROTO_TCP:
+		header2 |= PKT_OFFLOAD_HDR2_SIZE_3(tcp_hdrlen(skb));
+		epkt |= PKT_OFFLOAD_EPKT_TP(0) | PKT_OFFLOAD_EPKT_CSUM_L4;
+		header_cnt++;
+		break;
+	case IPPROTO_UDP:
+		header2 |= PKT_OFFLOAD_HDR2_SIZE_3(UDP_HLEN);
+		epkt |= PKT_OFFLOAD_EPKT_TP(1) | PKT_OFFLOAD_EPKT_CSUM_L4;
+		header_cnt++;
+		break;
+	default:
+		goto help;
+	}
+
+	offload = (struct bcmasp_pkt_offload *)skb_push(skb, sizeof(*offload));
+
+	header |= PKT_OFFLOAD_HDR_OP | PKT_OFFLOAD_HDR_COUNT(header_cnt) |
+		  PKT_OFFLOAD_HDR_SIZE_1(ETH_HLEN);
+	epkt |= PKT_OFFLOAD_EPKT_OP;
+
+	offload->nop = htonl(PKT_OFFLOAD_NOP);
+	offload->header = htonl(header);
+	offload->header2 = htonl(header2);
+	offload->epkt = htonl(epkt);
+	offload->end = htonl(PKT_OFFLOAD_END_OP);
+	*csum_hw = true;
+
+	return skb;
+
+help:
+	skb_checksum_help(skb);
+
+	return skb;
+}
+
+static unsigned long bcmasp_rx_edpkt_dma_rq(struct bcmasp_intf *intf)
+{
+	return rx_edpkt_dma_rq(intf, RX_EDPKT_DMA_VALID);
+}
+
+static void bcmasp_rx_edpkt_cfg_wq(struct bcmasp_intf *intf, dma_addr_t addr)
+{
+	rx_edpkt_cfg_wq(intf, addr, RX_EDPKT_RING_BUFFER_READ);
+}
+
+static void bcmasp_rx_edpkt_dma_wq(struct bcmasp_intf *intf, dma_addr_t addr)
+{
+	rx_edpkt_dma_wq(intf, addr, RX_EDPKT_DMA_READ);
+}
+
+static unsigned long bcmasp_tx_spb_dma_rq(struct bcmasp_intf *intf)
+{
+	return tx_spb_dma_rq(intf, TX_SPB_DMA_READ);
+}
+
+static void bcmasp_tx_spb_dma_wq(struct bcmasp_intf *intf, dma_addr_t addr)
+{
+	tx_spb_dma_wq(intf, addr, TX_SPB_DMA_VALID);
+}
+
+static const struct bcmasp_intf_ops bcmasp_intf_ops = {
+	.rx_desc_read = bcmasp_rx_edpkt_dma_rq,
+	.rx_buffer_write = bcmasp_rx_edpkt_cfg_wq,
+	.rx_desc_write = bcmasp_rx_edpkt_dma_wq,
+	.tx_read = bcmasp_tx_spb_dma_rq,
+	.tx_write = bcmasp_tx_spb_dma_wq,
+};
+
+static netdev_tx_t bcmasp_xmit(struct sk_buff *skb, struct net_device *dev)
+{
+	struct bcmasp_intf *intf = netdev_priv(dev);
+	unsigned int total_bytes, size;
+	int spb_index, nr_frags, i, j;
+	struct bcmasp_tx_cb *txcb;
+	dma_addr_t mapping, valid;
+	struct bcmasp_desc *desc;
+	bool csum_hw = false;
+	struct device *kdev;
+	skb_frag_t *frag;
+
+	kdev = &intf->parent->pdev->dev;
+
+	nr_frags = skb_shinfo(skb)->nr_frags;
+
+	if (tx_spb_ring_full(intf, nr_frags + 1)) {
+		netif_stop_queue(dev);
+		if (net_ratelimit())
+			netdev_err(dev, "Tx Ring Full!\n");
+		return NETDEV_TX_BUSY;
+	}
+
+	/* Save skb len before adding csum offload header */
+	total_bytes = skb->len;
+	skb = bcmasp_csum_offload(dev, skb, &csum_hw);
+	if (!skb)
+		return NETDEV_TX_OK;
+
+	spb_index = intf->tx_spb_index;
+	valid = intf->tx_spb_dma_valid;
+	for (i = 0; i <= nr_frags; i++) {
+		if (!i) {
+			size = skb_headlen(skb);
+			if (!nr_frags && size < (ETH_ZLEN + ETH_FCS_LEN)) {
+				if (skb_put_padto(skb, ETH_ZLEN + ETH_FCS_LEN))
+					return NETDEV_TX_OK;
+				size = skb->len;
+			}
+			mapping = dma_map_single(kdev, skb->data, size,
+						 DMA_TO_DEVICE);
+		} else {
+			frag = &skb_shinfo(skb)->frags[i - 1];
+			size = skb_frag_size(frag);
+			mapping = skb_frag_dma_map(kdev, frag, 0, size,
+						   DMA_TO_DEVICE);
+		}
+
+		if (dma_mapping_error(kdev, mapping)) {
+			intf->mib.tx_dma_failed++;
+			spb_index = intf->tx_spb_index;
+			for (j = 0; j < i; j++) {
+				bcmasp_clean_txcb(intf, spb_index);
+				spb_index = incr_ring(spb_index,
+						      DESC_RING_COUNT);
+			}
+			/* Rewind so we do not have a hole */
+			spb_index = intf->tx_spb_index;
+			dev_kfree_skb(skb);
+			return NETDEV_TX_OK;
+		}
+
+		txcb = &intf->tx_cbs[spb_index];
+		desc = &intf->tx_spb_cpu[spb_index];
+		memset(desc, 0, sizeof(*desc));
+		txcb->skb = skb;
+		txcb->bytes_sent = total_bytes;
+		dma_unmap_addr_set(txcb, dma_addr, mapping);
+		dma_unmap_len_set(txcb, dma_len, size);
+		if (!i) {
+			desc->flags |= DESC_SOF;
+			if (csum_hw)
+				desc->flags |= DESC_EPKT_CMD;
+		}
+
+		if (i == nr_frags) {
+			desc->flags |= DESC_EOF;
+			txcb->last = true;
+		}
+
+		desc->buf = mapping;
+		desc->size = size;
+		desc->flags |= DESC_INT_EN;
+
+		netif_dbg(intf, tx_queued, dev,
+			  "%s dma_buf=%pad dma_len=0x%x flags=0x%x index=0x%x\n",
+			  __func__, &mapping, desc->size, desc->flags,
+			  spb_index);
+
+		spb_index = incr_ring(spb_index, DESC_RING_COUNT);
+		valid = incr_last_byte(valid, intf->tx_spb_dma_addr,
+				       DESC_RING_COUNT);
+	}
+
+	/* Ensure all descriptors have been written to DRAM for the
+	 * hardware to see up-to-date contents.
+	 */
+	wmb();
+
+	intf->tx_spb_index = spb_index;
+	intf->tx_spb_dma_valid = valid;
+
+	skb_tx_timestamp(skb);
+
+	bcmasp_intf_tx_write(intf, intf->tx_spb_dma_valid);
+
+	if (tx_spb_ring_full(intf, MAX_SKB_FRAGS + 1))
+		netif_stop_queue(dev);
+
+	return NETDEV_TX_OK;
+}
+
+static void bcmasp_netif_start(struct net_device *dev)
+{
+	struct bcmasp_intf *intf = netdev_priv(dev);
+
+	bcmasp_set_rx_mode(dev);
+	napi_enable(&intf->tx_napi);
+	napi_enable(&intf->rx_napi);
+
+	bcmasp_enable_rx_irq(intf, 1);
+	bcmasp_enable_tx_irq(intf, 1);
+	bcmasp_enable_phy_irq(intf, 1);
+
+	phy_start(dev->phydev);
+}
+
+static void umac_reset(struct bcmasp_intf *intf)
+{
+	umac_wl(intf, 0x0, UMC_CMD);
+	umac_wl(intf, UMC_CMD_SW_RESET, UMC_CMD);
+	usleep_range(10, 100);
+	/* We hold the umac in reset and bring it out of
+	 * reset when phy link is up.
+	 */
+}
+
+static void umac_set_hw_addr(struct bcmasp_intf *intf,
+			     const unsigned char *addr)
+{
+	u32 mac0 = (addr[0] << 24) | (addr[1] << 16) | (addr[2] << 8) |
+		    addr[3];
+	u32 mac1 = (addr[4] << 8) | addr[5];
+
+	umac_wl(intf, mac0, UMC_MAC0);
+	umac_wl(intf, mac1, UMC_MAC1);
+}
+
+static void umac_enable_set(struct bcmasp_intf *intf, u32 mask,
+			    unsigned int enable)
+{
+	u32 reg;
+
+	reg = umac_rl(intf, UMC_CMD);
+	if (reg & UMC_CMD_SW_RESET)
+		return;
+	if (enable)
+		reg |= mask;
+	else
+		reg &= ~mask;
+	umac_wl(intf, reg, UMC_CMD);
+
+	/* UniMAC stops on a packet boundary, wait for a full-sized packet
+	 * to be processed (1 msec).
+	 */
+	if (enable == 0)
+		usleep_range(1000, 2000);
+}
+
+static void umac_init(struct bcmasp_intf *intf)
+{
+	umac_wl(intf, 0x800, UMC_FRM_LEN);
+	umac_wl(intf, 0xffff, UMC_PAUSE_CNTRL);
+	umac_wl(intf, 0x800, UMC_RX_MAX_PKT_SZ);
+}
+
+static int bcmasp_tx_reclaim(struct bcmasp_intf *intf)
+{
+	struct bcmasp_intf_stats64 *stats = &intf->stats64;
+	struct device *kdev = &intf->parent->pdev->dev;
+	unsigned long read, released = 0;
+	struct bcmasp_tx_cb *txcb;
+	struct bcmasp_desc *desc;
+	dma_addr_t mapping;
+
+	read = bcmasp_intf_tx_read(intf);
+	while (intf->tx_spb_dma_read != read) {
+		txcb = &intf->tx_cbs[intf->tx_spb_clean_index];
+		mapping = dma_unmap_addr(txcb, dma_addr);
+
+		dma_unmap_single(kdev, mapping,
+				 dma_unmap_len(txcb, dma_len),
+				 DMA_TO_DEVICE);
+
+		if (txcb->last) {
+			dev_consume_skb_any(txcb->skb);
+
+			u64_stats_update_begin(&stats->syncp);
+			u64_stats_inc(&stats->tx_packets);
+			u64_stats_add(&stats->tx_bytes, txcb->bytes_sent);
+			u64_stats_update_end(&stats->syncp);
+		}
+
+		desc = &intf->tx_spb_cpu[intf->tx_spb_clean_index];
+
+		netif_dbg(intf, tx_done, intf->ndev,
+			  "%s dma_buf=%pad dma_len=0x%x flags=0x%x c_index=0x%x\n",
+			  __func__, &mapping, desc->size, desc->flags,
+			  intf->tx_spb_clean_index);
+
+		bcmasp_clean_txcb(intf, intf->tx_spb_clean_index);
+		released++;
+
+		intf->tx_spb_clean_index = incr_ring(intf->tx_spb_clean_index,
+						     DESC_RING_COUNT);
+		intf->tx_spb_dma_read = incr_first_byte(intf->tx_spb_dma_read,
+							intf->tx_spb_dma_addr,
+							DESC_RING_COUNT);
+	}
+
+	return released;
+}
+
+static int bcmasp_tx_poll(struct napi_struct *napi, int budget)
+{
+	struct bcmasp_intf *intf =
+		container_of(napi, struct bcmasp_intf, tx_napi);
+	int released = 0;
+
+	released = bcmasp_tx_reclaim(intf);
+
+	napi_complete(&intf->tx_napi);
+
+	bcmasp_enable_tx_irq(intf, 1);
+
+	if (released)
+		netif_wake_queue(intf->ndev);
+
+	return 0;
+}
+
+static int bcmasp_rx_poll(struct napi_struct *napi, int budget)
+{
+	struct bcmasp_intf *intf =
+		container_of(napi, struct bcmasp_intf, rx_napi);
+	struct bcmasp_intf_stats64 *stats = &intf->stats64;
+	struct device *kdev = &intf->parent->pdev->dev;
+	unsigned long processed = 0;
+	struct bcmasp_desc *desc;
+	struct sk_buff *skb;
+	dma_addr_t valid;
+	void *data;
+	u64 flags;
+	u32 len;
+
+	valid = bcmasp_intf_rx_desc_read(intf) + 1;
+	if (valid == intf->rx_edpkt_dma_addr + DESC_RING_SIZE)
+		valid = intf->rx_edpkt_dma_addr;
+
+	while ((processed < budget) && (valid != intf->rx_edpkt_dma_read)) {
+		desc = &intf->rx_edpkt_cpu[intf->rx_edpkt_index];
+
+		/* Ensure that descriptor has been fully written to DRAM by
+		 * hardware before reading by the CPU
+		 */
+		rmb();
+
+		/* Calculate virt addr by offsetting from physical addr */
+		data = intf->rx_ring_cpu +
+			(DESC_ADDR(desc->buf) - intf->rx_ring_dma);
+
+		flags = DESC_FLAGS(desc->buf);
+		if (unlikely(flags & (DESC_CRC_ERR | DESC_RX_SYM_ERR))) {
+			if (net_ratelimit()) {
+				netif_err(intf, rx_status, intf->ndev,
+					  "flags=0x%llx\n", flags);
+			}
+
+			u64_stats_update_begin(&stats->syncp);
+			if (flags & DESC_CRC_ERR)
+				u64_stats_inc(&stats->rx_crc_errs);
+			if (flags & DESC_RX_SYM_ERR)
+				u64_stats_inc(&stats->rx_sym_errs);
+			u64_stats_update_end(&stats->syncp);
+
+			goto next;
+		}
+
+		dma_sync_single_for_cpu(kdev, DESC_ADDR(desc->buf), desc->size,
+					DMA_FROM_DEVICE);
+
+		len = desc->size;
+
+		skb = napi_alloc_skb(napi, len);
+		if (!skb) {
+			u64_stats_update_begin(&stats->syncp);
+			u64_stats_inc(&stats->rx_dropped);
+			u64_stats_update_end(&stats->syncp);
+			intf->mib.alloc_rx_skb_failed++;
+
+			goto next;
+		}
+
+		skb_put(skb, len);
+		memcpy(skb->data, data, len);
+
+		skb_pull(skb, 2);
+		len -= 2;
+		if (likely(intf->crc_fwd)) {
+			skb_trim(skb, len - ETH_FCS_LEN);
+			len -= ETH_FCS_LEN;
+		}
+
+		if ((intf->ndev->features & NETIF_F_RXCSUM) &&
+		    (desc->buf & DESC_CHKSUM))
+			skb->ip_summed = CHECKSUM_UNNECESSARY;
+
+		skb->protocol = eth_type_trans(skb, intf->ndev);
+
+		napi_gro_receive(napi, skb);
+
+		u64_stats_update_begin(&stats->syncp);
+		u64_stats_inc(&stats->rx_packets);
+		u64_stats_add(&stats->rx_bytes, len);
+		u64_stats_update_end(&stats->syncp);
+
+next:
+		bcmasp_intf_rx_buffer_write(intf, (DESC_ADDR(desc->buf) +
+					    desc->size));
+
+		processed++;
+		intf->rx_edpkt_dma_read =
+			incr_first_byte(intf->rx_edpkt_dma_read,
+					intf->rx_edpkt_dma_addr,
+					DESC_RING_COUNT);
+		intf->rx_edpkt_index = incr_ring(intf->rx_edpkt_index,
+						 DESC_RING_COUNT);
+	}
+
+	bcmasp_intf_rx_desc_write(intf, intf->rx_edpkt_dma_read);
+
+	if (processed < budget && napi_complete_done(&intf->rx_napi, processed))
+		bcmasp_enable_rx_irq(intf, 1);
+
+	return processed;
+}
+
+static void bcmasp_adj_link(struct net_device *dev)
+{
+	struct bcmasp_intf *intf = netdev_priv(dev);
+	struct phy_device *phydev = dev->phydev;
+	u32 cmd_bits = 0, reg;
+	int changed = 0;
+
+	if (intf->old_link != phydev->link) {
+		changed = 1;
+		intf->old_link = phydev->link;
+	}
+
+	if (intf->old_duplex != phydev->duplex) {
+		changed = 1;
+		intf->old_duplex = phydev->duplex;
+	}
+
+	switch (phydev->speed) {
+	case SPEED_2500:
+		cmd_bits = UMC_CMD_SPEED_2500;
+		break;
+	case SPEED_1000:
+		cmd_bits = UMC_CMD_SPEED_1000;
+		break;
+	case SPEED_100:
+		cmd_bits = UMC_CMD_SPEED_100;
+		break;
+	case SPEED_10:
+		cmd_bits = UMC_CMD_SPEED_10;
+		break;
+	default:
+		break;
+	}
+	cmd_bits <<= UMC_CMD_SPEED_SHIFT;
+
+	if (phydev->duplex == DUPLEX_HALF)
+		cmd_bits |= UMC_CMD_HD_EN;
+
+	if (intf->old_pause != phydev->pause) {
+		changed = 1;
+		intf->old_pause = phydev->pause;
+	}
+
+	if (!phydev->pause)
+		cmd_bits |= UMC_CMD_RX_PAUSE_IGNORE | UMC_CMD_TX_PAUSE_IGNORE;
+
+	if (!changed)
+		return;
+
+	if (phydev->link) {
+		reg = umac_rl(intf, UMC_CMD);
+		reg &= ~((UMC_CMD_SPEED_MASK << UMC_CMD_SPEED_SHIFT) |
+			UMC_CMD_HD_EN | UMC_CMD_RX_PAUSE_IGNORE |
+			UMC_CMD_TX_PAUSE_IGNORE);
+		reg |= cmd_bits;
+		if (reg & UMC_CMD_SW_RESET) {
+			reg &= ~UMC_CMD_SW_RESET;
+			umac_wl(intf, reg, UMC_CMD);
+			udelay(2);
+			reg |= UMC_CMD_TX_EN | UMC_CMD_RX_EN | UMC_CMD_PROMISC;
+		}
+		umac_wl(intf, reg, UMC_CMD);
+
+		umac_wl(intf, phydev->eee_cfg.tx_lpi_timer, UMC_EEE_LPI_TIMER);
+		reg = umac_rl(intf, UMC_EEE_CTRL);
+		if (phydev->enable_tx_lpi)
+			reg |= EEE_EN;
+		else
+			reg &= ~EEE_EN;
+		umac_wl(intf, reg, UMC_EEE_CTRL);
+	}
+
+	reg = rgmii_rl(intf, RGMII_OOB_CNTRL);
+	if (phydev->link)
+		reg |= RGMII_LINK;
+	else
+		reg &= ~RGMII_LINK;
+	rgmii_wl(intf, reg, RGMII_OOB_CNTRL);
+
+	if (changed)
+		phy_print_status(phydev);
+}
+
+static int bcmasp_alloc_buffers(struct bcmasp_intf *intf)
+{
+	struct device *kdev = &intf->parent->pdev->dev;
+	struct page *buffer_pg;
+
+	/* Alloc RX */
+	intf->rx_buf_order = get_order(RING_BUFFER_SIZE);
+	buffer_pg = alloc_pages(GFP_KERNEL, intf->rx_buf_order);
+	if (!buffer_pg)
+		return -ENOMEM;
+
+	intf->rx_ring_cpu = page_to_virt(buffer_pg);
+	intf->rx_ring_dma = dma_map_page(kdev, buffer_pg, 0, RING_BUFFER_SIZE,
+					 DMA_FROM_DEVICE);
+	if (dma_mapping_error(kdev, intf->rx_ring_dma))
+		goto free_rx_buffer;
+
+	intf->rx_edpkt_cpu = dma_alloc_coherent(kdev, DESC_RING_SIZE,
+						&intf->rx_edpkt_dma_addr, GFP_KERNEL);
+	if (!intf->rx_edpkt_cpu)
+		goto free_rx_buffer_dma;
+
+	/* Alloc TX */
+	intf->tx_spb_cpu = dma_alloc_coherent(kdev, DESC_RING_SIZE,
+					      &intf->tx_spb_dma_addr, GFP_KERNEL);
+	if (!intf->tx_spb_cpu)
+		goto free_rx_edpkt_dma;
+
+	intf->tx_cbs = kcalloc(DESC_RING_COUNT, sizeof(struct bcmasp_tx_cb),
+			       GFP_KERNEL);
+	if (!intf->tx_cbs)
+		goto free_tx_spb_dma;
+
+	return 0;
+
+free_tx_spb_dma:
+	dma_free_coherent(kdev, DESC_RING_SIZE, intf->tx_spb_cpu,
+			  intf->tx_spb_dma_addr);
+free_rx_edpkt_dma:
+	dma_free_coherent(kdev, DESC_RING_SIZE, intf->rx_edpkt_cpu,
+			  intf->rx_edpkt_dma_addr);
+free_rx_buffer_dma:
+	dma_unmap_page(kdev, intf->rx_ring_dma, RING_BUFFER_SIZE,
+		       DMA_FROM_DEVICE);
+free_rx_buffer:
+	__free_pages(buffer_pg, intf->rx_buf_order);
+
+	return -ENOMEM;
+}
+
+static void bcmasp_reclaim_free_buffers(struct bcmasp_intf *intf)
+{
+	struct device *kdev = &intf->parent->pdev->dev;
+
+	/* RX buffers */
+	dma_free_coherent(kdev, DESC_RING_SIZE, intf->rx_edpkt_cpu,
+			  intf->rx_edpkt_dma_addr);
+	dma_unmap_page(kdev, intf->rx_ring_dma, RING_BUFFER_SIZE,
+		       DMA_FROM_DEVICE);
+	__free_pages(virt_to_page(intf->rx_ring_cpu), intf->rx_buf_order);
+
+	/* TX buffers */
+	dma_free_coherent(kdev, DESC_RING_SIZE, intf->tx_spb_cpu,
+			  intf->tx_spb_dma_addr);
+	kfree(intf->tx_cbs);
+}
+
+static void bcmasp_init_rx(struct bcmasp_intf *intf)
+{
+	/* Restart from index 0 */
+	intf->rx_ring_dma_valid = intf->rx_ring_dma + RING_BUFFER_SIZE - 1;
+	intf->rx_edpkt_dma_valid = intf->rx_edpkt_dma_addr + (DESC_RING_SIZE - 1);
+	intf->rx_edpkt_dma_read = intf->rx_edpkt_dma_addr;
+	intf->rx_edpkt_index = 0;
+
+	/* Make sure channels are disabled */
+	rx_edpkt_cfg_wl(intf, 0x0, RX_EDPKT_CFG_ENABLE);
+
+	/* Rx SPB */
+	rx_edpkt_cfg_wq(intf, intf->rx_ring_dma, RX_EDPKT_RING_BUFFER_READ);
+	rx_edpkt_cfg_wq(intf, intf->rx_ring_dma, RX_EDPKT_RING_BUFFER_WRITE);
+	rx_edpkt_cfg_wq(intf, intf->rx_ring_dma, RX_EDPKT_RING_BUFFER_BASE);
+	rx_edpkt_cfg_wq(intf, intf->rx_ring_dma_valid,
+			RX_EDPKT_RING_BUFFER_END);
+	rx_edpkt_cfg_wq(intf, intf->rx_ring_dma_valid,
+			RX_EDPKT_RING_BUFFER_VALID);
+
+	/* EDPKT */
+	rx_edpkt_cfg_wl(intf, (RX_EDPKT_CFG_CFG0_RBUF_4K <<
+			RX_EDPKT_CFG_CFG0_DBUF_SHIFT) |
+		       (RX_EDPKT_CFG_CFG0_64_ALN <<
+			RX_EDPKT_CFG_CFG0_BALN_SHIFT) |
+		       (RX_EDPKT_CFG_CFG0_EFRM_STUF),
+			RX_EDPKT_CFG_CFG0);
+	rx_edpkt_dma_wq(intf, intf->rx_edpkt_dma_addr, RX_EDPKT_DMA_WRITE);
+	rx_edpkt_dma_wq(intf, intf->rx_edpkt_dma_addr, RX_EDPKT_DMA_READ);
+	rx_edpkt_dma_wq(intf, intf->rx_edpkt_dma_addr, RX_EDPKT_DMA_BASE);
+	rx_edpkt_dma_wq(intf, intf->rx_edpkt_dma_valid, RX_EDPKT_DMA_END);
+	rx_edpkt_dma_wq(intf, intf->rx_edpkt_dma_valid, RX_EDPKT_DMA_VALID);
+
+	umac2fb_wl(intf, UMAC2FB_CFG_DEFAULT_EN | ((intf->channel + 11) <<
+		   UMAC2FB_CFG_CHID_SHIFT) | (0xd << UMAC2FB_CFG_OK_SEND_SHIFT),
+		   UMAC2FB_CFG);
+}
+
+
+static void bcmasp_init_tx(struct bcmasp_intf *intf)
+{
+	/* Restart from index 0 */
+	intf->tx_spb_dma_valid = intf->tx_spb_dma_addr + DESC_RING_SIZE - 1;
+	intf->tx_spb_dma_read = intf->tx_spb_dma_addr;
+	intf->tx_spb_index = 0;
+	intf->tx_spb_clean_index = 0;
+	memset(intf->tx_cbs, 0, sizeof(struct bcmasp_tx_cb) * DESC_RING_COUNT);
+
+	/* Make sure channels are disabled */
+	tx_spb_ctrl_wl(intf, 0x0, TX_SPB_CTRL_ENABLE);
+	tx_epkt_core_wl(intf, 0x0, TX_EPKT_C_CFG_MISC);
+
+	/* Tx SPB */
+	tx_spb_ctrl_wl(intf, ((intf->channel + 8) << TX_SPB_CTRL_XF_BID_SHIFT),
+		       TX_SPB_CTRL_XF_CTRL2);
+
+	if (intf->parent->tx_chan_offset)
+		tx_pause_ctrl_wl(intf, (1 << (intf->channel + 8)), TX_PAUSE_MAP_VECTOR);
+	tx_spb_top_wl(intf, 0x1e, TX_SPB_TOP_BLKOUT);
+
+	tx_spb_dma_wq(intf, intf->tx_spb_dma_addr, TX_SPB_DMA_READ);
+	tx_spb_dma_wq(intf, intf->tx_spb_dma_addr, TX_SPB_DMA_BASE);
+	tx_spb_dma_wq(intf, intf->tx_spb_dma_valid, TX_SPB_DMA_END);
+	tx_spb_dma_wq(intf, intf->tx_spb_dma_valid, TX_SPB_DMA_VALID);
+}
+
+static void bcmasp_ephy_enable_set(struct bcmasp_intf *intf, bool enable)
+{
+	u32 mask = RGMII_EPHY_CFG_IDDQ_BIAS | RGMII_EPHY_CFG_EXT_PWRDOWN |
+		   RGMII_EPHY_CFG_IDDQ_GLOBAL;
+	u32 reg;
+
+	reg = rgmii_rl(intf, RGMII_EPHY_CNTRL);
+	if (enable) {
+		reg &= ~RGMII_EPHY_CK25_DIS;
+		rgmii_wl(intf, reg, RGMII_EPHY_CNTRL);
+		mdelay(1);
+
+		reg &= ~mask;
+		reg |= RGMII_EPHY_RESET;
+		rgmii_wl(intf, reg, RGMII_EPHY_CNTRL);
+		mdelay(1);
+
+		reg &= ~RGMII_EPHY_RESET;
+	} else {
+		reg |= mask | RGMII_EPHY_RESET;
+		rgmii_wl(intf, reg, RGMII_EPHY_CNTRL);
+		mdelay(1);
+		reg |= RGMII_EPHY_CK25_DIS;
+	}
+	rgmii_wl(intf, reg, RGMII_EPHY_CNTRL);
+	mdelay(1);
+
+	/* Set or clear the LED control override to avoid lighting up LEDs
+	 * while the EPHY is powered off and drawing unnecessary current.
+	 */
+	reg = rgmii_rl(intf, RGMII_SYS_LED_CNTRL);
+	if (enable)
+		reg &= ~RGMII_SYS_LED_CNTRL_LINK_OVRD;
+	else
+		reg |= RGMII_SYS_LED_CNTRL_LINK_OVRD;
+	rgmii_wl(intf, reg, RGMII_SYS_LED_CNTRL);
+}
+
+static void bcmasp_rgmii_mode_en_set(struct bcmasp_intf *intf, bool enable)
+{
+	u32 reg;
+
+	reg = rgmii_rl(intf, RGMII_OOB_CNTRL);
+	reg &= ~RGMII_OOB_DIS;
+	if (enable)
+		reg |= RGMII_MODE_EN;
+	else
+		reg &= ~RGMII_MODE_EN;
+	rgmii_wl(intf, reg, RGMII_OOB_CNTRL);
+}
+
+static void bcmasp_netif_deinit(struct net_device *dev)
+{
+	struct bcmasp_intf *intf = netdev_priv(dev);
+	u32 reg, timeout = 1000;
+
+	napi_disable(&intf->tx_napi);
+
+	bcmasp_enable_tx(intf, 0);
+
+	/* Flush any TX packets in the pipe */
+	tx_spb_dma_wl(intf, TX_SPB_DMA_FIFO_FLUSH, TX_SPB_DMA_FIFO_CTRL);
+	do {
+		reg = tx_spb_dma_rl(intf, TX_SPB_DMA_FIFO_STATUS);
+		if (!(reg & TX_SPB_DMA_FIFO_FLUSH))
+			break;
+		usleep_range(1000, 2000);
+	} while (timeout-- > 0);
+	tx_spb_dma_wl(intf, 0x0, TX_SPB_DMA_FIFO_CTRL);
+
+	bcmasp_tx_reclaim(intf);
+
+	umac_enable_set(intf, UMC_CMD_TX_EN, 0);
+
+	phy_stop(dev->phydev);
+
+	umac_enable_set(intf, UMC_CMD_RX_EN, 0);
+
+	bcmasp_flush_rx_port(intf);
+	usleep_range(1000, 2000);
+	bcmasp_enable_rx(intf, 0);
+
+	napi_disable(&intf->rx_napi);
+
+	/* Disable interrupts */
+	bcmasp_enable_tx_irq(intf, 0);
+	bcmasp_enable_rx_irq(intf, 0);
+	bcmasp_enable_phy_irq(intf, 0);
+
+	netif_napi_del(&intf->tx_napi);
+	netif_napi_del(&intf->rx_napi);
+}
+
+static int bcmasp_stop(struct net_device *dev)
+{
+	struct bcmasp_intf *intf = netdev_priv(dev);
+
+	netif_dbg(intf, ifdown, dev, "bcmasp stop\n");
+
+	/* Stop tx from updating HW */
+	netif_tx_disable(dev);
+
+	bcmasp_netif_deinit(dev);
+
+	bcmasp_reclaim_free_buffers(intf);
+
+	phy_disconnect(dev->phydev);
+
+	/* Disable internal EPHY or external PHY */
+	if (intf->internal_phy)
+		bcmasp_ephy_enable_set(intf, false);
+	else
+		bcmasp_rgmii_mode_en_set(intf, false);
+
+	/* Disable the interface clocks */
+	bcmasp_core_clock_set_intf(intf, false);
+
+	clk_disable_unprepare(intf->parent->clk);
+
+	return 0;
+}
+
+static void bcmasp_configure_port(struct bcmasp_intf *intf)
+{
+	u32 reg, id_mode_dis = 0;
+
+	reg = rgmii_rl(intf, RGMII_PORT_CNTRL);
+	reg &= ~RGMII_PORT_MODE_MASK;
+
+	switch (intf->phy_interface) {
+	case PHY_INTERFACE_MODE_RGMII:
+		/* RGMII_NO_ID: TXC transitions at the same time as TXD
+		 *		(requires PCB or receiver-side delay)
+		 * RGMII:	Add 2ns delay on TXC (90 degree shift)
+		 *
+		 * ID is implicitly disabled for 100Mbps (RG)MII operation.
+		 */
+		id_mode_dis = RGMII_ID_MODE_DIS;
+		fallthrough;
+	case PHY_INTERFACE_MODE_RGMII_TXID:
+		reg |= RGMII_PORT_MODE_EXT_GPHY;
+		break;
+	case PHY_INTERFACE_MODE_MII:
+		reg |= RGMII_PORT_MODE_EXT_EPHY;
+		break;
+	default:
+		break;
+	}
+
+	if (intf->internal_phy)
+		reg |= RGMII_PORT_MODE_EPHY;
+
+	rgmii_wl(intf, reg, RGMII_PORT_CNTRL);
+
+	reg = rgmii_rl(intf, RGMII_OOB_CNTRL);
+	reg &= ~RGMII_ID_MODE_DIS;
+	reg |= id_mode_dis;
+	rgmii_wl(intf, reg, RGMII_OOB_CNTRL);
+}
+
+static int bcmasp_netif_init(struct net_device *dev, bool phy_connect)
+{
+	struct bcmasp_intf *intf = netdev_priv(dev);
+	phy_interface_t phy_iface = intf->phy_interface;
+	u32 phy_flags = PHY_BRCM_AUTO_PWRDWN_ENABLE |
+			PHY_BRCM_DIS_TXCRXC_NOENRGY |
+			PHY_BRCM_IDDQ_SUSPEND;
+	struct phy_device *phydev = NULL;
+	int ret;
+
+	/* Always enable interface clocks */
+	bcmasp_core_clock_set_intf(intf, true);
+
+	/* Enable internal PHY or external PHY before any MAC activity */
+	if (intf->internal_phy)
+		bcmasp_ephy_enable_set(intf, true);
+	else
+		bcmasp_rgmii_mode_en_set(intf, true);
+	bcmasp_configure_port(intf);
+
+	/* This is an ugly quirk but we have not been correctly
+	 * interpreting the phy_interface values and we have done that
+	 * across different drivers, so at least we are consistent in
+	 * our mistakes.
+	 *
+	 * When the Generic PHY driver is in use either the PHY has
+	 * been strapped or programmed correctly by the boot loader so
+	 * we should stick to our incorrect interpretation since we
+	 * have validated it.
+	 *
+	 * Now when a dedicated PHY driver is in use, we need to
+	 * reverse the meaning of the phy_interface_mode values to
+	 * something that the PHY driver will interpret and act on such
+	 * that we have two mistakes canceling themselves so to speak.
+	 * We only do this for the two modes that GENET driver
+	 * officially supports on Broadcom STB chips:
+	 * PHY_INTERFACE_MODE_RGMII and PHY_INTERFACE_MODE_RGMII_TXID.
+	 * Other modes are not *officially* supported with the boot
+	 * loader and the scripted environment generating Device Tree
+	 * blobs for those platforms.
+	 *
+	 * Note that internal PHY and fixed-link configurations are not
+	 * affected because they use different phy_interface_t values
+	 * or the Generic PHY driver.
+	 */
+	switch (phy_iface) {
+	case PHY_INTERFACE_MODE_RGMII:
+		phy_iface = PHY_INTERFACE_MODE_RGMII_ID;
+		break;
+	case PHY_INTERFACE_MODE_RGMII_TXID:
+		phy_iface = PHY_INTERFACE_MODE_RGMII_RXID;
+		break;
+	default:
+		break;
+	}
+
+	if (phy_connect) {
+		phydev = of_phy_connect(dev, intf->phy_dn,
+					bcmasp_adj_link, phy_flags,
+					phy_iface);
+		if (!phydev) {
+			ret = -ENODEV;
+			netdev_err(dev, "could not attach to PHY\n");
+			goto err_phy_disable;
+		}
+
+		if (intf->internal_phy)
+			dev->phydev->irq = PHY_MAC_INTERRUPT;
+
+		/* Indicate that the MAC is responsible for PHY PM */
+		phydev->mac_managed_pm = true;
+
+		/* Set phylib's copy of the LPI timer */
+		phydev->eee_cfg.tx_lpi_timer = umac_rl(intf, UMC_EEE_LPI_TIMER);
+	}
+
+	umac_reset(intf);
+
+	umac_init(intf);
+
+	umac_set_hw_addr(intf, dev->dev_addr);
+
+	intf->old_duplex = -1;
+	intf->old_link = -1;
+	intf->old_pause = -1;
+
+	bcmasp_init_tx(intf);
+	netif_napi_add_tx(intf->ndev, &intf->tx_napi, bcmasp_tx_poll);
+	bcmasp_enable_tx(intf, 1);
+
+	bcmasp_init_rx(intf);
+	netif_napi_add(intf->ndev, &intf->rx_napi, bcmasp_rx_poll);
+	bcmasp_enable_rx(intf, 1);
+
+	intf->crc_fwd = !!(umac_rl(intf, UMC_CMD) & UMC_CMD_CRC_FWD);
+
+	bcmasp_netif_start(dev);
+
+	netif_start_queue(dev);
+
+	return 0;
+
+err_phy_disable:
+	if (intf->internal_phy)
+		bcmasp_ephy_enable_set(intf, false);
+	else
+		bcmasp_rgmii_mode_en_set(intf, false);
+	return ret;
+}
+
+static int bcmasp_open(struct net_device *dev)
+{
+	struct bcmasp_intf *intf = netdev_priv(dev);
+	int ret;
+
+	netif_dbg(intf, ifup, dev, "bcmasp open\n");
+
+	ret = bcmasp_alloc_buffers(intf);
+	if (ret)
+		return ret;
+
+	ret = clk_prepare_enable(intf->parent->clk);
+	if (ret)
+		goto err_free_mem;
+
+	ret = bcmasp_netif_init(dev, true);
+	if (ret) {
+		clk_disable_unprepare(intf->parent->clk);
+		goto err_free_mem;
+	}
+
+	return ret;
+
+err_free_mem:
+	bcmasp_reclaim_free_buffers(intf);
+
+	return ret;
+}
+
+static void bcmasp_tx_timeout(struct net_device *dev, unsigned int txqueue)
+{
+	struct bcmasp_intf *intf = netdev_priv(dev);
+
+	netif_dbg(intf, tx_err, dev, "transmit timeout!\n");
+	intf->mib.tx_timeout_cnt++;
+}
+
+static int bcmasp_get_phys_port_name(struct net_device *dev,
+				     char *name, size_t len)
+{
+	struct bcmasp_intf *intf = netdev_priv(dev);
+
+	if (snprintf(name, len, "p%d", intf->port) >= len)
+		return -EINVAL;
+
+	return 0;
+}
+
+static void bcmasp_get_stats64(struct net_device *dev,
+			       struct rtnl_link_stats64 *stats)
+{
+	struct bcmasp_intf *intf = netdev_priv(dev);
+	struct bcmasp_intf_stats64 *lstats;
+	unsigned int start;
+
+	lstats = &intf->stats64;
+
+	do {
+		start = u64_stats_fetch_begin(&lstats->syncp);
+		stats->rx_packets = u64_stats_read(&lstats->rx_packets);
+		stats->rx_bytes = u64_stats_read(&lstats->rx_bytes);
+		stats->rx_dropped = u64_stats_read(&lstats->rx_dropped);
+		stats->rx_crc_errors = u64_stats_read(&lstats->rx_crc_errs);
+		stats->rx_frame_errors = u64_stats_read(&lstats->rx_sym_errs);
+		stats->rx_errors = stats->rx_crc_errors + stats->rx_frame_errors;
+
+		stats->tx_packets = u64_stats_read(&lstats->tx_packets);
+		stats->tx_bytes = u64_stats_read(&lstats->tx_bytes);
+	} while (u64_stats_fetch_retry(&lstats->syncp, start));
+}
+
+static const struct net_device_ops bcmasp_netdev_ops = {
+	.ndo_open		= bcmasp_open,
+	.ndo_stop		= bcmasp_stop,
+	.ndo_start_xmit		= bcmasp_xmit,
+	.ndo_tx_timeout		= bcmasp_tx_timeout,
+	.ndo_set_rx_mode	= bcmasp_set_rx_mode,
+	.ndo_get_phys_port_name	= bcmasp_get_phys_port_name,
+	.ndo_eth_ioctl		= phy_do_ioctl_running,
+	.ndo_set_mac_address	= eth_mac_addr,
+	.ndo_get_stats64	= bcmasp_get_stats64,
+};
+
+static void bcmasp_map_res(struct bcmasp_priv *priv, struct bcmasp_intf *intf)
+{
+	/* Per port */
+	intf->res.umac = priv->base + UMC_OFFSET(intf);
+	intf->res.umac2fb = priv->base + (UMAC2FB_OFFSET + priv->rx_ctrl_offset +
+					  (intf->port * 0x4));
+	intf->res.rgmii = priv->base + RGMII_OFFSET(intf);
+
+	/* Per ch */
+	intf->tx_spb_dma = priv->base + TX_SPB_DMA_OFFSET(intf);
+	intf->res.tx_spb_ctrl = priv->base + TX_SPB_CTRL_OFFSET(intf);
+	intf->res.tx_spb_top = priv->base + TX_SPB_TOP_OFFSET(intf);
+	intf->res.tx_epkt_core = priv->base + TX_EPKT_C_OFFSET(intf);
+	intf->res.tx_pause_ctrl = priv->base + TX_PAUSE_CTRL_OFFSET(intf);
+
+	intf->rx_edpkt_dma = priv->base + RX_EDPKT_DMA_OFFSET(intf);
+	intf->rx_edpkt_cfg = priv->base + RX_EDPKT_CFG_OFFSET(intf);
+}
+
+struct bcmasp_intf *bcmasp_interface_create(struct bcmasp_priv *priv,
+					    struct device_node *ndev_dn, int i)
+{
+	struct device *dev = &priv->pdev->dev;
+	struct bcmasp_intf *intf;
+	struct net_device *ndev;
+	int ch, port, ret;
+
+	if (of_property_read_u32(ndev_dn, "reg", &port)) {
+		dev_warn(dev, "%s: invalid port number\n", ndev_dn->name);
+		goto err;
+	}
+
+	if (of_property_read_u32(ndev_dn, "brcm,channel", &ch)) {
+		dev_warn(dev, "%s: invalid ch number\n", ndev_dn->name);
+		goto err;
+	}
+
+	ndev = alloc_etherdev(sizeof(struct bcmasp_intf));
+	if (!ndev) {
+		dev_warn(dev, "%s: unable to alloc ndev\n", ndev_dn->name);
+		goto err;
+	}
+	intf = netdev_priv(ndev);
+
+	intf->parent = priv;
+	intf->ndev = ndev;
+	intf->channel = ch;
+	intf->port = port;
+	intf->ndev_dn = ndev_dn;
+	intf->index = i;
+
+	ret = of_get_phy_mode(ndev_dn, &intf->phy_interface);
+	if (ret < 0) {
+		dev_err(dev, "invalid PHY mode property\n");
+		goto err_free_netdev;
+	}
+
+	if (intf->phy_interface == PHY_INTERFACE_MODE_INTERNAL)
+		intf->internal_phy = true;
+
+	intf->phy_dn = of_parse_phandle(ndev_dn, "phy-handle", 0);
+	if (!intf->phy_dn && of_phy_is_fixed_link(ndev_dn)) {
+		ret = of_phy_register_fixed_link(ndev_dn);
+		if (ret) {
+			dev_warn(dev, "%s: failed to register fixed PHY\n",
+				 ndev_dn->name);
+			goto err_free_netdev;
+		}
+		intf->phy_dn = ndev_dn;
+	}
+
+	/* Map resource */
+	bcmasp_map_res(priv, intf);
+
+	if ((!phy_interface_mode_is_rgmii(intf->phy_interface) &&
+	     intf->phy_interface != PHY_INTERFACE_MODE_MII &&
+	     intf->phy_interface != PHY_INTERFACE_MODE_INTERNAL) ||
+	    (intf->port != 1 && intf->internal_phy)) {
+		netdev_err(intf->ndev, "invalid PHY mode: %s for port %d\n",
+			   phy_modes(intf->phy_interface), intf->port);
+		ret = -EINVAL;
+		goto err_free_netdev;
+	}
+
+	ret = of_get_ethdev_address(ndev_dn, ndev);
+	if (ret) {
+		netdev_warn(ndev, "using random Ethernet MAC\n");
+		eth_hw_addr_random(ndev);
+	}
+
+	SET_NETDEV_DEV(ndev, dev);
+	intf->ops = &bcmasp_intf_ops;
+	ndev->netdev_ops = &bcmasp_netdev_ops;
+	ndev->ethtool_ops = &bcmasp_ethtool_ops;
+	intf->msg_enable = netif_msg_init(-1, NETIF_MSG_DRV |
+					  NETIF_MSG_PROBE |
+					  NETIF_MSG_LINK);
+	ndev->features |= NETIF_F_IP_CSUM | NETIF_F_IPV6_CSUM | NETIF_F_SG |
+			  NETIF_F_RXCSUM;
+	ndev->hw_features |= ndev->features;
+	ndev->needed_headroom += sizeof(struct bcmasp_pkt_offload);
+
+	netdev_sw_irq_coalesce_default_on(ndev);
+
+	return intf;
+
+err_free_netdev:
+	free_netdev(ndev);
+err:
+	return NULL;
+}
+
+void bcmasp_interface_destroy(struct bcmasp_intf *intf)
+{
+	if (intf->ndev->reg_state == NETREG_REGISTERED)
+		unregister_netdev(intf->ndev);
+	if (of_phy_is_fixed_link(intf->ndev_dn))
+		of_phy_deregister_fixed_link(intf->ndev_dn);
+	free_netdev(intf->ndev);
+}
+
+static void bcmasp_suspend_to_wol(struct bcmasp_intf *intf)
+{
+	struct net_device *ndev = intf->ndev;
+	u32 reg;
+
+	reg = umac_rl(intf, UMC_MPD_CTRL);
+	if (intf->wolopts & (WAKE_MAGIC | WAKE_MAGICSECURE))
+		reg |= UMC_MPD_CTRL_MPD_EN;
+	reg &= ~UMC_MPD_CTRL_PSW_EN;
+	if (intf->wolopts & WAKE_MAGICSECURE) {
+		/* Program the SecureOn password */
+		umac_wl(intf, get_unaligned_be16(&intf->sopass[0]),
+			UMC_PSW_MS);
+		umac_wl(intf, get_unaligned_be32(&intf->sopass[2]),
+			UMC_PSW_LS);
+		reg |= UMC_MPD_CTRL_PSW_EN;
+	}
+	umac_wl(intf, reg, UMC_MPD_CTRL);
+
+	if (intf->wolopts & WAKE_FILTER)
+		bcmasp_netfilt_suspend(intf);
+
+	/* Bring UniMAC out of reset if needed and enable RX */
+	reg = umac_rl(intf, UMC_CMD);
+	if (reg & UMC_CMD_SW_RESET)
+		reg &= ~UMC_CMD_SW_RESET;
+
+	reg |= UMC_CMD_RX_EN | UMC_CMD_PROMISC;
+	umac_wl(intf, reg, UMC_CMD);
+
+	umac_enable_set(intf, UMC_CMD_RX_EN, 1);
+
+	if (intf->parent->wol_irq > 0) {
+		wakeup_intr2_core_wl(intf->parent, 0xffffffff,
+				     ASP_WAKEUP_INTR2_MASK_CLEAR);
+	}
+
+	if (ndev->phydev && ndev->phydev->eee_cfg.eee_enabled &&
+	    intf->parent->eee_fixup)
+		intf->parent->eee_fixup(intf, true);
+
+	netif_dbg(intf, wol, ndev, "entered WOL mode\n");
+}
+
+int bcmasp_interface_suspend(struct bcmasp_intf *intf)
+{
+	struct device *kdev = &intf->parent->pdev->dev;
+	struct net_device *dev = intf->ndev;
+
+	if (!netif_running(dev))
+		return 0;
+
+	netif_device_detach(dev);
+
+	bcmasp_netif_deinit(dev);
+
+	if (!intf->wolopts) {
+		if (intf->internal_phy)
+			bcmasp_ephy_enable_set(intf, false);
+		else
+			bcmasp_rgmii_mode_en_set(intf, false);
+
+		/* If Wake-on-LAN is disabled, we can safely
+		 * disable the network interface clocks.
+		 */
+		bcmasp_core_clock_set_intf(intf, false);
+	}
+
+	if (device_may_wakeup(kdev) && intf->wolopts)
+		bcmasp_suspend_to_wol(intf);
+
+	clk_disable_unprepare(intf->parent->clk);
+
+	return 0;
+}
+
+static void bcmasp_resume_from_wol(struct bcmasp_intf *intf)
+{
+	u32 reg;
+
+	if (intf->ndev->phydev && intf->ndev->phydev->eee_cfg.eee_enabled &&
+	    intf->parent->eee_fixup)
+		intf->parent->eee_fixup(intf, false);
+
+	reg = umac_rl(intf, UMC_MPD_CTRL);
+	reg &= ~UMC_MPD_CTRL_MPD_EN;
+	umac_wl(intf, reg, UMC_MPD_CTRL);
+
+	if (intf->parent->wol_irq > 0) {
+		wakeup_intr2_core_wl(intf->parent, 0xffffffff,
+				     ASP_WAKEUP_INTR2_MASK_SET);
+	}
+}
+
+int bcmasp_interface_resume(struct bcmasp_intf *intf)
+{
+	struct net_device *dev = intf->ndev;
+	int ret;
+
+	if (!netif_running(dev))
+		return 0;
+
+	ret = clk_prepare_enable(intf->parent->clk);
+	if (ret)
+		return ret;
+
+	ret = bcmasp_netif_init(dev, false);
+	if (ret)
+		goto out;
+
+	bcmasp_resume_from_wol(intf);
+
+	netif_device_attach(dev);
+
+	return 0;
+
+out:
+	clk_disable_unprepare(intf->parent->clk);
+	return ret;
+}
diff --git a/drivers/net/ethernet/broadcom/asp2/bcmasp_intf_defs.h b/drivers/net/ethernet/broadcom/asp2/bcmasp_intf_defs.h
new file mode 100644
index 000000000000..af7418348e81
--- /dev/null
+++ b/drivers/net/ethernet/broadcom/asp2/bcmasp_intf_defs.h
@@ -0,0 +1,256 @@
+/* SPDX-License-Identifier: GPL-2.0 */
+#ifndef __BCMASP_INTF_DEFS_H
+#define __BCMASP_INTF_DEFS_H
+
+#define UMC_OFFSET(intf)		\
+	((((intf)->port) * 0x800) + 0xc000)
+#define  UMC_CMD			0x008
+#define   UMC_CMD_TX_EN			BIT(0)
+#define   UMC_CMD_RX_EN			BIT(1)
+#define   UMC_CMD_SPEED_SHIFT		0x2
+#define    UMC_CMD_SPEED_MASK		0x3
+#define    UMC_CMD_SPEED_10		0x0
+#define    UMC_CMD_SPEED_100		0x1
+#define    UMC_CMD_SPEED_1000		0x2
+#define    UMC_CMD_SPEED_2500		0x3
+#define   UMC_CMD_PROMISC		BIT(4)
+#define   UMC_CMD_PAD_EN		BIT(5)
+#define   UMC_CMD_CRC_FWD		BIT(6)
+#define   UMC_CMD_PAUSE_FWD		BIT(7)
+#define   UMC_CMD_RX_PAUSE_IGNORE	BIT(8)
+#define   UMC_CMD_TX_ADDR_INS		BIT(9)
+#define   UMC_CMD_HD_EN			BIT(10)
+#define   UMC_CMD_SW_RESET		BIT(13)
+#define   UMC_CMD_LCL_LOOP_EN		BIT(15)
+#define   UMC_CMD_AUTO_CONFIG		BIT(22)
+#define   UMC_CMD_CNTL_FRM_EN		BIT(23)
+#define   UMC_CMD_NO_LEN_CHK		BIT(24)
+#define   UMC_CMD_RMT_LOOP_EN		BIT(25)
+#define   UMC_CMD_PRBL_EN		BIT(27)
+#define   UMC_CMD_TX_PAUSE_IGNORE	BIT(28)
+#define   UMC_CMD_TX_RX_EN		BIT(29)
+#define   UMC_CMD_RUNT_FILTER_DIS	BIT(30)
+#define  UMC_MAC0			0x0c
+#define  UMC_MAC1			0x10
+#define  UMC_FRM_LEN			0x14
+#define  UMC_EEE_CTRL			0x64
+#define   EN_LPI_RX_PAUSE		BIT(0)
+#define   EN_LPI_TX_PFC			BIT(1)
+#define   EN_LPI_TX_PAUSE		BIT(2)
+#define   EEE_EN			BIT(3)
+#define   RX_FIFO_CHECK			BIT(4)
+#define   EEE_TX_CLK_DIS		BIT(5)
+#define   DIS_EEE_10M			BIT(6)
+#define   LP_IDLE_PREDICTION_MODE	BIT(7)
+#define  UMC_EEE_LPI_TIMER		0x68
+#define  UMC_PAUSE_CNTRL		0x330
+#define  UMC_TX_FLUSH			0x334
+#define  UMC_GR64			0x400
+#define  UMC_GR127			0x404
+#define  UMC_GR255			0x408
+#define  UMC_GR511			0x40c
+#define  UMC_GR1023			0x410
+#define  UMC_GR1518			0x414
+#define  UMC_GRMGV			0x418
+#define  UMC_GR2047			0x41c
+#define  UMC_GR4095			0x420
+#define  UMC_GR9216			0x424
+#define  UMC_GRPKT			0x428
+#define  UMC_GRBYT			0x42c
+#define  UMC_GRMCA			0x430
+#define  UMC_GRBCA			0x434
+#define  UMC_GRFCS			0x438
+#define  UMC_GRXCF			0x43c
+#define  UMC_GRXPF			0x440
+#define  UMC_GRXUO			0x444
+#define  UMC_GRALN			0x448
+#define  UMC_GRFLR			0x44c
+#define  UMC_GRCDE			0x450
+#define  UMC_GRFCR			0x454
+#define  UMC_GROVR			0x458
+#define  UMC_GRJBR			0x45c
+#define  UMC_GRMTUE			0x460
+#define  UMC_GRPOK			0x464
+#define  UMC_GRUC			0x468
+#define  UMC_GRPPP			0x46c
+#define  UMC_GRMCRC			0x470
+#define  UMC_TR64			0x480
+#define  UMC_TR127			0x484
+#define  UMC_TR255			0x488
+#define  UMC_TR511			0x48c
+#define  UMC_TR1023			0x490
+#define  UMC_TR1518			0x494
+#define  UMC_TRMGV			0x498
+#define  UMC_TR2047			0x49c
+#define  UMC_TR4095			0x4a0
+#define  UMC_TR9216			0x4a4
+#define  UMC_GTPKT			0x4a8
+#define  UMC_GTMCA			0x4ac
+#define  UMC_GTBCA			0x4b0
+#define  UMC_GTXPF			0x4b4
+#define  UMC_GTXCF			0x4b8
+#define  UMC_GTFCS			0x4bc
+#define  UMC_GTOVR			0x4c0
+#define  UMC_GTDRF			0x4c4
+#define  UMC_GTEDF			0x4c8
+#define  UMC_GTSCL			0x4cc
+#define  UMC_GTMCL			0x4d0
+#define  UMC_GTLCL			0x4d4
+#define  UMC_GTXCL			0x4d8
+#define  UMC_GTFRG			0x4dc
+#define  UMC_GTNCL			0x4e0
+#define  UMC_GTJBR			0x4e4
+#define  UMC_GTBYT			0x4e8
+#define  UMC_GTPOK			0x4ec
+#define  UMC_GTUC			0x4f0
+#define  UMC_RRPKT			0x500
+#define  UMC_RRUND			0x504
+#define  UMC_RRFRG			0x508
+#define  UMC_RRBYT			0x50c
+#define  UMC_MIB_CNTRL			0x580
+#define   UMC_MIB_CNTRL_RX_CNT_RST	BIT(0)
+#define   UMC_MIB_CNTRL_RUNT_CNT_RST	BIT(1)
+#define   UMC_MIB_CNTRL_TX_CNT_RST	BIT(2)
+#define  UMC_RX_MAX_PKT_SZ		0x608
+#define  UMC_MPD_CTRL			0x620
+#define   UMC_MPD_CTRL_MPD_EN		BIT(0)
+#define   UMC_MPD_CTRL_PSW_EN		BIT(27)
+#define  UMC_PSW_MS			0x624
+#define  UMC_PSW_LS			0x628
+
+#define UMAC2FB_OFFSET			0x9f044
+#define  UMAC2FB_CFG			0x0
+#define   UMAC2FB_CFG_OPUT_EN		BIT(0)
+#define   UMAC2FB_CFG_VLAN_EN		BIT(1)
+#define   UMAC2FB_CFG_SNAP_EN		BIT(2)
+#define   UMAC2FB_CFG_BCM_TG_EN		BIT(3)
+#define   UMAC2FB_CFG_IPUT_EN		BIT(4)
+#define   UMAC2FB_CFG_CHID_SHIFT	8
+#define   UMAC2FB_CFG_OK_SEND_SHIFT	24
+#define   UMAC2FB_CFG_DEFAULT_EN	\
+		(UMAC2FB_CFG_OPUT_EN | UMAC2FB_CFG_VLAN_EN \
+		| UMAC2FB_CFG_SNAP_EN | UMAC2FB_CFG_IPUT_EN)
+
+#define RGMII_OFFSET(intf)	\
+	((((intf)->port) * 0x100) + 0xd000)
+#define  RGMII_EPHY_CNTRL		0x00
+#define    RGMII_EPHY_CFG_IDDQ_BIAS	BIT(0)
+#define    RGMII_EPHY_CFG_EXT_PWRDOWN	BIT(1)
+#define    RGMII_EPHY_CFG_FORCE_DLL_EN	BIT(2)
+#define    RGMII_EPHY_CFG_IDDQ_GLOBAL	BIT(3)
+#define    RGMII_EPHY_CK25_DIS		BIT(4)
+#define    RGMII_EPHY_RESET		BIT(7)
+#define  RGMII_OOB_CNTRL		0x0c
+#define   RGMII_LINK			BIT(4)
+#define   RGMII_OOB_DIS			BIT(5)
+#define   RGMII_MODE_EN			BIT(6)
+#define   RGMII_ID_MODE_DIS		BIT(16)
+
+#define RGMII_PORT_CNTRL		0x60
+#define   RGMII_PORT_MODE_EPHY		0
+#define   RGMII_PORT_MODE_GPHY		1
+#define   RGMII_PORT_MODE_EXT_EPHY	2
+#define   RGMII_PORT_MODE_EXT_GPHY	3
+#define   RGMII_PORT_MODE_EXT_RVMII	4
+#define   RGMII_PORT_MODE_MASK		GENMASK(2, 0)
+
+#define RGMII_SYS_LED_CNTRL		0x74
+#define  RGMII_SYS_LED_CNTRL_LINK_OVRD	BIT(15)
+
+#define TX_SPB_DMA_OFFSET(intf) \
+	((((intf)->channel) * 0x30) + 0x48180)
+#define  TX_SPB_DMA_READ		0x00
+#define  TX_SPB_DMA_BASE		0x08
+#define  TX_SPB_DMA_END			0x10
+#define  TX_SPB_DMA_VALID		0x18
+#define  TX_SPB_DMA_FIFO_CTRL		0x20
+#define   TX_SPB_DMA_FIFO_FLUSH		BIT(0)
+#define  TX_SPB_DMA_FIFO_STATUS		0x24
+
+#define TX_SPB_CTRL_OFFSET(intf) \
+	((((intf)->channel) * 0x68) + 0x49340)
+#define  TX_SPB_CTRL_ENABLE		0x0
+#define   TX_SPB_CTRL_ENABLE_EN		BIT(0)
+#define  TX_SPB_CTRL_XF_CTRL2		0x20
+#define   TX_SPB_CTRL_XF_BID_SHIFT	16
+
+#define TX_SPB_TOP_OFFSET(intf) \
+	((((intf)->channel) * 0x1c) + 0x4a0e0)
+#define TX_SPB_TOP_BLKOUT		0x0
+#define TX_SPB_TOP_SPRE_BW_CTRL		0x4
+
+#define TX_EPKT_C_OFFSET(intf) \
+	((((intf)->channel) * 0x120) + 0x40900)
+#define  TX_EPKT_C_CFG_MISC		0x0
+#define   TX_EPKT_C_CFG_MISC_EN		BIT(0)
+#define   TX_EPKT_C_CFG_MISC_PT		BIT(1)
+#define   TX_EPKT_C_CFG_MISC_PS_SHIFT	14
+#define   TX_EPKT_C_CFG_MISC_FD_SHIFT	20
+
+#define TX_PAUSE_CTRL_OFFSET(intf) \
+	((((intf)->channel * 0xc) + 0x49a20))
+#define  TX_PAUSE_MAP_VECTOR		0x8
+
+#define RX_EDPKT_DMA_OFFSET(intf) \
+	((((intf)->channel) * 0x38) + 0x9ca00)
+#define  RX_EDPKT_DMA_WRITE		0x00
+#define  RX_EDPKT_DMA_READ		0x08
+#define  RX_EDPKT_DMA_BASE		0x10
+#define  RX_EDPKT_DMA_END		0x18
+#define  RX_EDPKT_DMA_VALID		0x20
+#define  RX_EDPKT_DMA_FULLNESS		0x28
+#define  RX_EDPKT_DMA_MIN_THRES		0x2c
+#define  RX_EDPKT_DMA_CH_XONOFF		0x30
+
+#define RX_EDPKT_CFG_OFFSET(intf) \
+	((((intf)->channel) * 0x70) + 0x9c600)
+#define  RX_EDPKT_CFG_CFG0		0x0
+#define   RX_EDPKT_CFG_CFG0_DBUF_SHIFT	9
+#define    RX_EDPKT_CFG_CFG0_RBUF	0x0
+#define    RX_EDPKT_CFG_CFG0_RBUF_4K	0x1
+#define    RX_EDPKT_CFG_CFG0_BUF_4K	0x2
+/* EFRM STUFF, 0 = no byte stuff, 1 = two byte stuff */
+#define   RX_EDPKT_CFG_CFG0_EFRM_STUF	BIT(11)
+#define   RX_EDPKT_CFG_CFG0_BALN_SHIFT	12
+#define    RX_EDPKT_CFG_CFG0_NO_ALN	0
+#define    RX_EDPKT_CFG_CFG0_4_ALN	2
+#define    RX_EDPKT_CFG_CFG0_64_ALN	6
+#define  RX_EDPKT_RING_BUFFER_WRITE	0x38
+#define  RX_EDPKT_RING_BUFFER_READ	0x40
+#define  RX_EDPKT_RING_BUFFER_BASE	0x48
+#define  RX_EDPKT_RING_BUFFER_END	0x50
+#define  RX_EDPKT_RING_BUFFER_VALID	0x58
+#define  RX_EDPKT_CFG_ENABLE		0x6c
+#define   RX_EDPKT_CFG_ENABLE_EN	BIT(0)
+
+#define RX_SPB_DMA_OFFSET(intf) \
+	((((intf)->channel) * 0x30) + 0xa0000)
+#define  RX_SPB_DMA_READ		0x00
+#define  RX_SPB_DMA_BASE		0x08
+#define  RX_SPB_DMA_END			0x10
+#define  RX_SPB_DMA_VALID		0x18
+#define  RX_SPB_DMA_FIFO_CTRL		0x20
+#define   RX_SPB_DMA_FIFO_FLUSH		BIT(0)
+#define  RX_SPB_DMA_FIFO_STATUS		0x24
+
+#define RX_SPB_CTRL_OFFSET(intf) \
+	((((intf)->channel - 6) * 0x68) + 0xa1000)
+#define  RX_SPB_CTRL_ENABLE		0x00
+#define   RX_SPB_CTRL_ENABLE_EN		BIT(0)
+
+#define RX_PAUSE_CTRL_OFFSET(intf) \
+	((((intf)->channel - 6) * 0x4) + 0xa1138)
+#define  RX_PAUSE_MAP_VECTOR		0x00
+
+#define RX_SPB_TOP_CTRL_OFFSET(intf) \
+	((((intf)->channel - 6) * 0x14) + 0xa2000)
+#define  RX_SPB_TOP_BLKOUT		0x00
+
+#define NUM_4K_BUFFERS			32
+#define RING_BUFFER_SIZE		(PAGE_SIZE * NUM_4K_BUFFERS)
+
+#define DESC_RING_COUNT			(64 * NUM_4K_BUFFERS)
+#define DESC_SIZE			16
+#define DESC_RING_SIZE			(DESC_RING_COUNT * DESC_SIZE)
+
+#endif
diff --git a/drivers/net/ethernet/broadcom/b44.c b/drivers/net/ethernet/broadcom/b44.c
index 5b95bb48ce97..0353359c3fe9 100644
--- a/drivers/net/ethernet/broadcom/b44.c
+++ b/drivers/net/ethernet/broadcom/b44.c
@@ -40,7 +40,6 @@
 #include "b44.h"
 
 #define DRV_MODULE_NAME		"b44"
-#define DRV_MODULE_VERSION	"2.0"
 #define DRV_DESCRIPTION		"Broadcom 44xx/47xx 10/100 PCI ethernet driver"
 
 #define B44_DEF_MSG_ENABLE	  \
@@ -97,7 +96,6 @@
 MODULE_AUTHOR("Felix Fietkau, Florian Schirmer, Pekka Pietikainen, David S. Miller");
 MODULE_DESCRIPTION(DRV_DESCRIPTION);
 MODULE_LICENSE("GPL");
-MODULE_VERSION(DRV_MODULE_VERSION);
 
 static int b44_debug = -1;	/* -1 == use B44_DEF_MSG_ENABLE as value */
 module_param(b44_debug, int, 0);
@@ -198,29 +196,8 @@ static int b44_wait_bit(struct b44 *bp, unsigned long reg,
 	return 0;
 }
 
-static inline void __b44_cam_read(struct b44 *bp, unsigned char *data, int index)
-{
-	u32 val;
-
-	bw32(bp, B44_CAM_CTRL, (CAM_CTRL_READ |
-			    (index << CAM_CTRL_INDEX_SHIFT)));
-
-	b44_wait_bit(bp, B44_CAM_CTRL, CAM_CTRL_BUSY, 100, 1);
-
-	val = br32(bp, B44_CAM_DATA_LO);
-
-	data[2] = (val >> 24) & 0xFF;
-	data[3] = (val >> 16) & 0xFF;
-	data[4] = (val >> 8) & 0xFF;
-	data[5] = (val >> 0) & 0xFF;
-
-	val = br32(bp, B44_CAM_DATA_HI);
-
-	data[0] = (val >> 8) & 0xFF;
-	data[1] = (val >> 0) & 0xFF;
-}
-
-static inline void __b44_cam_write(struct b44 *bp, unsigned char *data, int index)
+static inline void __b44_cam_write(struct b44 *bp,
+				   const unsigned char *data, int index)
 {
 	u32 val;
 
@@ -511,9 +488,6 @@ static void b44_stats_update(struct b44 *bp)
 		*val++ += br32(bp, reg);
 	}
 
-	/* Pad */
-	reg += 8*4UL;
-
 	for (reg = B44_RX_GOOD_O; reg <= B44_RX_NPAUSE; reg += 4UL) {
 		*val++ += br32(bp, reg);
 	}
@@ -599,9 +573,9 @@ static void b44_check_phy(struct b44 *bp)
 	}
 }
 
-static void b44_timer(unsigned long __opaque)
+static void b44_timer(struct timer_list *t)
 {
-	struct b44 *bp = (struct b44 *) __opaque;
+	struct b44 *bp = timer_container_of(bp, t, timer);
 
 	spin_lock_irq(&bp->lock);
 
@@ -638,7 +612,7 @@ static void b44_tx(struct b44 *bp)
 		bytes_compl += skb->len;
 		pkts_compl++;
 
-		dev_kfree_skb_irq(skb);
+		dev_consume_skb_irq(skb);
 	}
 
 	netdev_completed_queue(bp->dev, pkts_compl, bytes_compl);
@@ -951,7 +925,7 @@ irq_ack:
 	return IRQ_RETVAL(handled);
 }
 
-static void b44_tx_timeout(struct net_device *dev)
+static void b44_tx_timeout(struct net_device *dev, unsigned int txqueue)
 {
 	struct b44 *bp = netdev_priv(dev);
 
@@ -1012,7 +986,7 @@ static netdev_tx_t b44_start_xmit(struct sk_buff *skb, struct net_device *dev)
 		}
 
 		skb_copy_from_linear_data(skb, skb_put(bounce_skb, len), len);
-		dev_kfree_skb_any(skb);
+		dev_consume_skb_any(skb);
 		skb = bounce_skb;
 	}
 
@@ -1068,13 +1042,13 @@ static int b44_change_mtu(struct net_device *dev, int new_mtu)
 		/* We'll just catch it later when the
 		 * device is up'd.
 		 */
-		dev->mtu = new_mtu;
+		WRITE_ONCE(dev->mtu, new_mtu);
 		return 0;
 	}
 
 	spin_lock_irq(&bp->lock);
 	b44_halt(bp);
-	dev->mtu = new_mtu;
+	WRITE_ONCE(dev->mtu, new_mtu);
 	b44_init_rings(bp);
 	b44_init_hw(bp, B44_FULL_RESET);
 	spin_unlock_irq(&bp->lock);
@@ -1205,7 +1179,7 @@ static int b44_alloc_consistent(struct b44 *bp, gfp_t gfp)
 	bp->rx_ring = dma_alloc_coherent(bp->sdev->dma_dev, size,
 					 &bp->rx_ring_dma, gfp);
 	if (!bp->rx_ring) {
-		/* Allocation may have failed due to pci_alloc_consistent
+		/* Allocation may have failed due to dma_alloc_coherent
 		   insisting on use of GFP_DMA, which is more restrictive
 		   than necessary...  */
 		struct dma_desc *rx_ring;
@@ -1388,7 +1362,7 @@ static int b44_set_mac_addr(struct net_device *dev, void *p)
 	if (!is_valid_ether_addr(addr->sa_data))
 		return -EINVAL;
 
-	memcpy(dev->dev_addr, addr->sa_data, dev->addr_len);
+	eth_hw_addr_set(dev, addr->sa_data);
 
 	spin_lock_irq(&bp->lock);
 
@@ -1474,10 +1448,8 @@ static int b44_open(struct net_device *dev)
 		goto out;
 	}
 
-	init_timer(&bp->timer);
+	timer_setup(&bp->timer, b44_timer, 0);
 	bp->timer.expires = jiffies + HZ;
-	bp->timer.data = (unsigned long) bp;
-	bp->timer.function = b44_timer;
 	add_timer(&bp->timer);
 
 	b44_enable_ints(bp);
@@ -1514,15 +1486,18 @@ static void bwfilter_table(struct b44 *bp, u8 *pp, u32 bytes, u32 table_offset)
 	}
 }
 
-static int b44_magic_pattern(u8 *macaddr, u8 *ppattern, u8 *pmask, int offset)
+static int b44_magic_pattern(const u8 *macaddr, u8 *ppattern, u8 *pmask,
+			     int offset)
 {
 	int magicsync = 6;
 	int k, j, len = offset;
 	int ethaddr_bytes = ETH_ALEN;
 
 	memset(ppattern + offset, 0xff, magicsync);
-	for (j = 0; j < magicsync; j++)
-		set_bit(len++, (unsigned long *) pmask);
+	for (j = 0; j < magicsync; j++) {
+		pmask[len >> 3] |= BIT(len & 7);
+		len++;
+	}
 
 	for (j = 0; j < B44_MAX_PATTERNS; j++) {
 		if ((B44_PATTERN_SIZE - len) >= ETH_ALEN)
@@ -1534,7 +1509,8 @@ static int b44_magic_pattern(u8 *macaddr, u8 *ppattern, u8 *pmask, int offset)
 		for (k = 0; k< ethaddr_bytes; k++) {
 			ppattern[offset + magicsync +
 				(j * ETH_ALEN) + k] = macaddr[k];
-			set_bit(len++, (unsigned long *) pmask);
+			pmask[len >> 3] |= BIT(len & 7);
+			len++;
 		}
 	}
 	return len - 1;
@@ -1560,8 +1536,8 @@ static void b44_setup_pseudo_magicp(struct b44 *bp)
 	plen0 = b44_magic_pattern(bp->dev->dev_addr, pwol_pattern, pwol_mask,
 				  B44_ETHIPV4UDP_HLEN);
 
-   	bwfilter_table(bp, pwol_pattern, B44_PATTERN_SIZE, B44_PATTERN_BASE);
-   	bwfilter_table(bp, pwol_mask, B44_PMASK_SIZE, B44_PMASK_BASE);
+	bwfilter_table(bp, pwol_pattern, B44_PATTERN_SIZE, B44_PATTERN_BASE);
+	bwfilter_table(bp, pwol_mask, B44_PMASK_SIZE, B44_PMASK_BASE);
 
 	/* Raw ethernet II magic packet pattern - pattern 1 */
 	memset(pwol_pattern, 0, B44_PATTERN_SIZE);
@@ -1569,9 +1545,9 @@ static void b44_setup_pseudo_magicp(struct b44 *bp)
 	plen1 = b44_magic_pattern(bp->dev->dev_addr, pwol_pattern, pwol_mask,
 				  ETH_HLEN);
 
-   	bwfilter_table(bp, pwol_pattern, B44_PATTERN_SIZE,
+	bwfilter_table(bp, pwol_pattern, B44_PATTERN_SIZE,
 		       B44_PATTERN_BASE + B44_PATTERN_SIZE);
-  	bwfilter_table(bp, pwol_mask, B44_PMASK_SIZE,
+	bwfilter_table(bp, pwol_mask, B44_PMASK_SIZE,
 		       B44_PMASK_BASE + B44_PMASK_SIZE);
 
 	/* Ipv6 magic packet pattern - pattern 2 */
@@ -1580,9 +1556,9 @@ static void b44_setup_pseudo_magicp(struct b44 *bp)
 	plen2 = b44_magic_pattern(bp->dev->dev_addr, pwol_pattern, pwol_mask,
 				  B44_ETHIPV6UDP_HLEN);
 
-   	bwfilter_table(bp, pwol_pattern, B44_PATTERN_SIZE,
+	bwfilter_table(bp, pwol_pattern, B44_PATTERN_SIZE,
 		       B44_PATTERN_BASE + B44_PATTERN_SIZE + B44_PATTERN_SIZE);
-  	bwfilter_table(bp, pwol_mask, B44_PMASK_SIZE,
+	bwfilter_table(bp, pwol_mask, B44_PMASK_SIZE,
 		       B44_PMASK_BASE + B44_PMASK_SIZE + B44_PMASK_SIZE);
 
 	kfree(pwol_pattern);
@@ -1635,9 +1611,9 @@ static void b44_setup_wol(struct b44 *bp)
 		val = br32(bp, B44_DEVCTRL);
 		bw32(bp, B44_DEVCTRL, val | DEVCTRL_MPM | DEVCTRL_PFE);
 
- 	} else {
- 		b44_setup_pseudo_magicp(bp);
- 	}
+	} else {
+		b44_setup_pseudo_magicp(bp);
+	}
 	b44_setup_wol_pci(bp);
 }
 
@@ -1652,7 +1628,7 @@ static int b44_close(struct net_device *dev)
 
 	napi_disable(&bp->napi);
 
-	del_timer_sync(&bp->timer);
+	timer_delete_sync(&bp->timer);
 
 	spin_lock_irq(&bp->lock);
 
@@ -1682,7 +1658,7 @@ static void b44_get_stats64(struct net_device *dev,
 	unsigned int start;
 
 	do {
-		start = u64_stats_fetch_begin_irq(&hwstat->syncp);
+		start = u64_stats_fetch_begin(&hwstat->syncp);
 
 		/* Convert HW stats into rtnl_link_stats64 stats. */
 		nstat->rx_packets = hwstat->rx_pkts;
@@ -1716,7 +1692,7 @@ static void b44_get_stats64(struct net_device *dev,
 		/* Carrier lost counter seems to be broken for some devices */
 		nstat->tx_carrier_errors = hwstat->tx_carrier_lost;
 #endif
-	} while (u64_stats_fetch_retry_irq(&hwstat->syncp, start));
+	} while (u64_stats_fetch_retry(&hwstat->syncp, start));
 
 }
 
@@ -1761,7 +1737,7 @@ static void __b44_set_rx_mode(struct net_device *dev)
 			__b44_cam_write(bp, zero, i);
 
 		bw32(bp, B44_RXCONFIG, val);
-        	val = br32(bp, B44_CAM_CTRL);
+		val = br32(bp, B44_CAM_CTRL);
 	        bw32(bp, B44_CAM_CTRL, val | CAM_CTRL_ENABLE);
 	}
 }
@@ -1792,14 +1768,13 @@ static void b44_get_drvinfo (struct net_device *dev, struct ethtool_drvinfo *inf
 	struct b44 *bp = netdev_priv(dev);
 	struct ssb_bus *bus = bp->sdev->bus;
 
-	strlcpy(info->driver, DRV_MODULE_NAME, sizeof(info->driver));
-	strlcpy(info->version, DRV_MODULE_VERSION, sizeof(info->version));
+	strscpy(info->driver, DRV_MODULE_NAME, sizeof(info->driver));
 	switch (bus->bustype) {
 	case SSB_BUSTYPE_PCI:
-		strlcpy(info->bus_info, pci_name(bus->host_pci), sizeof(info->bus_info));
+		strscpy(info->bus_info, pci_name(bus->host_pci), sizeof(info->bus_info));
 		break;
 	case SSB_BUSTYPE_SSB:
-		strlcpy(info->bus_info, "SSB", sizeof(info->bus_info));
+		strscpy(info->bus_info, "SSB", sizeof(info->bus_info));
 		break;
 	case SSB_BUSTYPE_PCMCIA:
 	case SSB_BUSTYPE_SDIO:
@@ -1818,11 +1793,9 @@ static int b44_nway_reset(struct net_device *dev)
 	b44_readphy(bp, MII_BMCR, &bmcr);
 	b44_readphy(bp, MII_BMCR, &bmcr);
 	r = -EINVAL;
-	if (bmcr & BMCR_ANENABLE) {
-		b44_writephy(bp, MII_BMCR,
-			     bmcr | BMCR_ANRESTART);
-		r = 0;
-	}
+	if (bmcr & BMCR_ANENABLE)
+		r = b44_writephy(bp, MII_BMCR,
+				 bmcr | BMCR_ANRESTART);
 	spin_unlock_irq(&bp->lock);
 
 	return r;
@@ -1836,7 +1809,9 @@ static int b44_get_link_ksettings(struct net_device *dev,
 
 	if (bp->flags & B44_FLAG_EXTERNAL_PHY) {
 		BUG_ON(!dev->phydev);
-		return phy_ethtool_ksettings_get(dev->phydev, cmd);
+		phy_ethtool_ksettings_get(dev->phydev, cmd);
+
+		return 0;
 	}
 
 	supported = (SUPPORTED_Autoneg);
@@ -1962,7 +1937,9 @@ static int b44_set_link_ksettings(struct net_device *dev,
 }
 
 static void b44_get_ringparam(struct net_device *dev,
-			      struct ethtool_ringparam *ering)
+			      struct ethtool_ringparam *ering,
+			      struct kernel_ethtool_ringparam *kernel_ering,
+			      struct netlink_ext_ack *extack)
 {
 	struct b44 *bp = netdev_priv(dev);
 
@@ -1973,7 +1950,9 @@ static void b44_get_ringparam(struct net_device *dev,
 }
 
 static int b44_set_ringparam(struct net_device *dev,
-			     struct ethtool_ringparam *ering)
+			     struct ethtool_ringparam *ering,
+			     struct kernel_ethtool_ringparam *kernel_ering,
+			     struct netlink_ext_ack *extack)
 {
 	struct b44 *bp = netdev_priv(dev);
 
@@ -2030,12 +2009,14 @@ static int b44_set_pauseparam(struct net_device *dev,
 		bp->flags |= B44_FLAG_TX_PAUSE;
 	else
 		bp->flags &= ~B44_FLAG_TX_PAUSE;
-	if (bp->flags & B44_FLAG_PAUSE_AUTO) {
-		b44_halt(bp);
-		b44_init_rings(bp);
-		b44_init_hw(bp, B44_FULL_RESET);
-	} else {
-		__b44_set_flow_ctrl(bp, bp->flags);
+	if (netif_running(dev)) {
+		if (bp->flags & B44_FLAG_PAUSE_AUTO) {
+			b44_halt(bp);
+			b44_init_rings(bp);
+			b44_init_hw(bp, B44_FULL_RESET);
+		} else {
+			__b44_set_flow_ctrl(bp, bp->flags);
+		}
 	}
 	spin_unlock_irq(&bp->lock);
 
@@ -2079,12 +2060,12 @@ static void b44_get_ethtool_stats(struct net_device *dev,
 	do {
 		data_src = &hwstat->tx_good_octets;
 		data_dst = data;
-		start = u64_stats_fetch_begin_irq(&hwstat->syncp);
+		start = u64_stats_fetch_begin(&hwstat->syncp);
 
 		for (i = 0; i < ARRAY_SIZE(b44_gstrings); i++)
 			*data_dst++ = *data_src++;
 
-	} while (u64_stats_fetch_retry_irq(&hwstat->syncp, start));
+	} while (u64_stats_fetch_retry(&hwstat->syncp, start));
 }
 
 static void b44_get_wol(struct net_device *dev, struct ethtool_wolinfo *wol)
@@ -2174,7 +2155,7 @@ static int b44_get_invariants(struct b44 *bp)
 	 * valid PHY address. */
 	bp->phy_addr &= 0x1F;
 
-	memcpy(bp->dev->dev_addr, addr, ETH_ALEN);
+	eth_hw_addr_set(bp->dev, addr);
 
 	if (!is_valid_ether_addr(&bp->dev->dev_addr[0])){
 		pr_err("Invalid MAC address found in EEPROM\n");
@@ -2201,7 +2182,7 @@ static const struct net_device_ops b44_netdev_ops = {
 	.ndo_set_rx_mode	= b44_set_rx_mode,
 	.ndo_set_mac_address	= b44_set_mac_addr,
 	.ndo_validate_addr	= eth_validate_addr,
-	.ndo_do_ioctl		= b44_ioctl,
+	.ndo_eth_ioctl		= b44_ioctl,
 	.ndo_tx_timeout		= b44_tx_timeout,
 	.ndo_change_mtu		= b44_change_mtu,
 #ifdef CONFIG_NET_POLL_CONTROLLER
@@ -2213,12 +2194,12 @@ static void b44_adjust_link(struct net_device *dev)
 {
 	struct b44 *bp = netdev_priv(dev);
 	struct phy_device *phydev = dev->phydev;
-	bool status_changed = 0;
+	bool status_changed = false;
 
 	BUG_ON(!phydev);
 
 	if (bp->old_link != phydev->link) {
-		status_changed = 1;
+		status_changed = true;
 		bp->old_link = phydev->link;
 	}
 
@@ -2226,11 +2207,11 @@ static void b44_adjust_link(struct net_device *dev)
 	if (phydev->link) {
 		if ((phydev->duplex == DUPLEX_HALF) &&
 		    (bp->flags & B44_FLAG_FULL_DUPLEX)) {
-			status_changed = 1;
+			status_changed = true;
 			bp->flags &= ~B44_FLAG_FULL_DUPLEX;
 		} else if ((phydev->duplex == DUPLEX_FULL) &&
 			   !(bp->flags & B44_FLAG_FULL_DUPLEX)) {
-			status_changed = 1;
+			status_changed = true;
 			bp->flags |= B44_FLAG_FULL_DUPLEX;
 		}
 	}
@@ -2248,6 +2229,7 @@ static void b44_adjust_link(struct net_device *dev)
 
 static int b44_register_phy_one(struct b44 *bp)
 {
+	__ETHTOOL_DECLARE_LINK_MODE_MASK(mask) = { 0, };
 	struct mii_bus *mii_bus;
 	struct ssb_device *sdev = bp->sdev;
 	struct phy_device *phydev;
@@ -2303,11 +2285,12 @@ static int b44_register_phy_one(struct b44 *bp)
 	}
 
 	/* mask with MAC supported features */
-	phydev->supported &= (SUPPORTED_100baseT_Half |
-			      SUPPORTED_100baseT_Full |
-			      SUPPORTED_Autoneg |
-			      SUPPORTED_MII);
-	phydev->advertising = phydev->supported;
+	linkmode_set_bit(ETHTOOL_LINK_MODE_100baseT_Half_BIT, mask);
+	linkmode_set_bit(ETHTOOL_LINK_MODE_100baseT_Full_BIT, mask);
+	linkmode_set_bit(ETHTOOL_LINK_MODE_Autoneg_BIT, mask);
+	linkmode_set_bit(ETHTOOL_LINK_MODE_MII_BIT, mask);
+	linkmode_and(phydev->supported, phydev->supported, mask);
+	linkmode_copy(phydev->advertising, phydev->supported);
 
 	bp->old_link = 0;
 	bp->phy_addr = phydev->mdio.addr;
@@ -2345,8 +2328,6 @@ static int b44_init_one(struct ssb_device *sdev,
 
 	instance++;
 
-	pr_info_once("%s version %s\n", DRV_DESCRIPTION, DRV_MODULE_VERSION);
-
 	dev = alloc_etherdev(sizeof(*bp));
 	if (!dev) {
 		err = -ENOMEM;
@@ -2366,12 +2347,13 @@ static int b44_init_one(struct ssb_device *sdev,
 	bp->msg_enable = netif_msg_init(b44_debug, B44_DEF_MSG_ENABLE);
 
 	spin_lock_init(&bp->lock);
+	u64_stats_init(&bp->hw_stats.syncp);
 
 	bp->rx_pending = B44_DEF_RX_RING_PENDING;
 	bp->tx_pending = B44_DEF_TX_RING_PENDING;
 
 	dev->netdev_ops = &b44_netdev_ops;
-	netif_napi_add(dev, &bp->napi, b44_poll, 64);
+	netif_napi_add(dev, &bp->napi, b44_poll);
 	dev->watchdog_timeo = B44_TX_TIMEOUT;
 	dev->min_mtu = B44_MIN_MTU;
 	dev->max_mtu = B44_MAX_MTU;
@@ -2385,7 +2367,8 @@ static int b44_init_one(struct ssb_device *sdev,
 		goto err_out_free_dev;
 	}
 
-	if (dma_set_mask_and_coherent(sdev->dma_dev, DMA_BIT_MASK(30))) {
+	err = dma_set_mask_and_coherent(sdev->dma_dev, DMA_BIT_MASK(30));
+	if (err) {
 		dev_err(sdev->dev,
 			"Required 30BIT DMA mask unsupported by the system\n");
 		goto err_out_powerdown;
@@ -2490,7 +2473,7 @@ static int b44_suspend(struct ssb_device *sdev, pm_message_t state)
 	if (!netif_running(dev))
 		return 0;
 
-	del_timer_sync(&bp->timer);
+	timer_delete_sync(&bp->timer);
 
 	spin_lock_irq(&bp->lock);
 
@@ -2587,7 +2570,7 @@ static int __init b44_init(void)
 	unsigned int dma_desc_align_size = dma_get_cache_alignment();
 	int err;
 
-	/* Setup paramaters for syncing RX/TX DMA descriptors */
+	/* Setup parameters for syncing RX/TX DMA descriptors */
 	dma_desc_sync_size = max_t(unsigned int, dma_desc_align_size, sizeof(struct dma_desc));
 
 	err = b44_pci_init();
diff --git a/drivers/net/ethernet/broadcom/b44.h b/drivers/net/ethernet/broadcom/b44.h
index 89d2cf341163..b3e36ca0fd19 100644
--- a/drivers/net/ethernet/broadcom/b44.h
+++ b/drivers/net/ethernet/broadcom/b44.h
@@ -1,3 +1,4 @@
+/* SPDX-License-Identifier: GPL-2.0 */
 #ifndef _B44_H
 #define _B44_H
 
diff --git a/drivers/net/ethernet/broadcom/bcm4908_enet.c b/drivers/net/ethernet/broadcom/bcm4908_enet.c
new file mode 100644
index 000000000000..203e8d0dd04b
--- /dev/null
+++ b/drivers/net/ethernet/broadcom/bcm4908_enet.c
@@ -0,0 +1,798 @@
+// SPDX-License-Identifier: GPL-2.0-only
+/*
+ * Copyright (C) 2021 Rafał Miłecki <rafal@milecki.pl>
+ */
+
+#include <linux/delay.h>
+#include <linux/etherdevice.h>
+#include <linux/if_vlan.h>
+#include <linux/interrupt.h>
+#include <linux/module.h>
+#include <linux/of.h>
+#include <linux/of_net.h>
+#include <linux/platform_device.h>
+#include <linux/slab.h>
+#include <linux/string.h>
+
+#include "bcm4908_enet.h"
+#include "unimac.h"
+
+#define ENET_DMA_CH_RX_CFG			ENET_DMA_CH0_CFG
+#define ENET_DMA_CH_TX_CFG			ENET_DMA_CH1_CFG
+#define ENET_DMA_CH_RX_STATE_RAM		ENET_DMA_CH0_STATE_RAM
+#define ENET_DMA_CH_TX_STATE_RAM		ENET_DMA_CH1_STATE_RAM
+
+#define ENET_TX_BDS_NUM				200
+#define ENET_RX_BDS_NUM				200
+#define ENET_RX_BDS_NUM_MAX			8192
+
+#define ENET_DMA_INT_DEFAULTS			(ENET_DMA_CH_CFG_INT_DONE | \
+						 ENET_DMA_CH_CFG_INT_NO_DESC | \
+						 ENET_DMA_CH_CFG_INT_BUFF_DONE)
+#define ENET_DMA_MAX_BURST_LEN			8 /* in 64 bit words */
+
+#define ENET_MTU_MAX				ETH_DATA_LEN /* Is it possible to support 2044? */
+#define BRCM_MAX_TAG_LEN			6
+#define ENET_MAX_ETH_OVERHEAD			(ETH_HLEN + BRCM_MAX_TAG_LEN + VLAN_HLEN + \
+						 ETH_FCS_LEN + 4) /* 32 */
+
+#define ENET_RX_SKB_BUF_SIZE			(NET_SKB_PAD + NET_IP_ALIGN + \
+						 ETH_HLEN + BRCM_MAX_TAG_LEN + VLAN_HLEN + \
+						 ENET_MTU_MAX + ETH_FCS_LEN + 4)
+#define ENET_RX_SKB_BUF_ALLOC_SIZE		(SKB_DATA_ALIGN(ENET_RX_SKB_BUF_SIZE) + \
+						 SKB_DATA_ALIGN(sizeof(struct skb_shared_info)))
+#define ENET_RX_BUF_DMA_OFFSET			(NET_SKB_PAD + NET_IP_ALIGN)
+#define ENET_RX_BUF_DMA_SIZE			(ENET_RX_SKB_BUF_SIZE - ENET_RX_BUF_DMA_OFFSET)
+
+struct bcm4908_enet_dma_ring_bd {
+	__le32 ctl;
+	__le32 addr;
+} __packed;
+
+struct bcm4908_enet_dma_ring_slot {
+	union {
+		void *buf;			/* RX */
+		struct sk_buff *skb;		/* TX */
+	};
+	unsigned int len;
+	dma_addr_t dma_addr;
+};
+
+struct bcm4908_enet_dma_ring {
+	int is_tx;
+	int read_idx;
+	int write_idx;
+	int length;
+	u16 cfg_block;
+	u16 st_ram_block;
+	struct napi_struct napi;
+
+	union {
+		void *cpu_addr;
+		struct bcm4908_enet_dma_ring_bd *buf_desc;
+	};
+	dma_addr_t dma_addr;
+
+	struct bcm4908_enet_dma_ring_slot *slots;
+};
+
+struct bcm4908_enet {
+	struct device *dev;
+	struct net_device *netdev;
+	void __iomem *base;
+	int irq_tx;
+
+	struct bcm4908_enet_dma_ring tx_ring;
+	struct bcm4908_enet_dma_ring rx_ring;
+};
+
+/***
+ * R/W ops
+ */
+
+static u32 enet_read(struct bcm4908_enet *enet, u16 offset)
+{
+	return readl(enet->base + offset);
+}
+
+static void enet_write(struct bcm4908_enet *enet, u16 offset, u32 value)
+{
+	writel(value, enet->base + offset);
+}
+
+static void enet_maskset(struct bcm4908_enet *enet, u16 offset, u32 mask, u32 set)
+{
+	u32 val;
+
+	WARN_ON(set & ~mask);
+
+	val = enet_read(enet, offset);
+	val = (val & ~mask) | (set & mask);
+	enet_write(enet, offset, val);
+}
+
+static void enet_set(struct bcm4908_enet *enet, u16 offset, u32 set)
+{
+	enet_maskset(enet, offset, set, set);
+}
+
+static u32 enet_umac_read(struct bcm4908_enet *enet, u16 offset)
+{
+	return enet_read(enet, ENET_UNIMAC + offset);
+}
+
+static void enet_umac_write(struct bcm4908_enet *enet, u16 offset, u32 value)
+{
+	enet_write(enet, ENET_UNIMAC + offset, value);
+}
+
+static void enet_umac_set(struct bcm4908_enet *enet, u16 offset, u32 set)
+{
+	enet_set(enet, ENET_UNIMAC + offset, set);
+}
+
+/***
+ * Helpers
+ */
+
+static void bcm4908_enet_set_mtu(struct bcm4908_enet *enet, int mtu)
+{
+	enet_umac_write(enet, UMAC_MAX_FRAME_LEN, mtu + ENET_MAX_ETH_OVERHEAD);
+}
+
+/***
+ * DMA ring ops
+ */
+
+static void bcm4908_enet_dma_ring_intrs_on(struct bcm4908_enet *enet,
+					   struct bcm4908_enet_dma_ring *ring)
+{
+	enet_write(enet, ring->cfg_block + ENET_DMA_CH_CFG_INT_MASK, ENET_DMA_INT_DEFAULTS);
+}
+
+static void bcm4908_enet_dma_ring_intrs_off(struct bcm4908_enet *enet,
+					    struct bcm4908_enet_dma_ring *ring)
+{
+	enet_write(enet, ring->cfg_block + ENET_DMA_CH_CFG_INT_MASK, 0);
+}
+
+static void bcm4908_enet_dma_ring_intrs_ack(struct bcm4908_enet *enet,
+					    struct bcm4908_enet_dma_ring *ring)
+{
+	enet_write(enet, ring->cfg_block + ENET_DMA_CH_CFG_INT_STAT, ENET_DMA_INT_DEFAULTS);
+}
+
+/***
+ * DMA
+ */
+
+static int bcm4908_dma_alloc_buf_descs(struct bcm4908_enet *enet,
+				       struct bcm4908_enet_dma_ring *ring)
+{
+	int size = ring->length * sizeof(struct bcm4908_enet_dma_ring_bd);
+	struct device *dev = enet->dev;
+
+	ring->cpu_addr = dma_alloc_coherent(dev, size, &ring->dma_addr, GFP_KERNEL);
+	if (!ring->cpu_addr)
+		return -ENOMEM;
+
+	if (((uintptr_t)ring->cpu_addr) & (0x40 - 1)) {
+		dev_err(dev, "Invalid DMA ring alignment\n");
+		goto err_free_buf_descs;
+	}
+
+	ring->slots = kcalloc(ring->length, sizeof(*ring->slots), GFP_KERNEL);
+	if (!ring->slots)
+		goto err_free_buf_descs;
+
+	return 0;
+
+err_free_buf_descs:
+	dma_free_coherent(dev, size, ring->cpu_addr, ring->dma_addr);
+	ring->cpu_addr = NULL;
+	return -ENOMEM;
+}
+
+static void bcm4908_enet_dma_free(struct bcm4908_enet *enet)
+{
+	struct bcm4908_enet_dma_ring *tx_ring = &enet->tx_ring;
+	struct bcm4908_enet_dma_ring *rx_ring = &enet->rx_ring;
+	struct device *dev = enet->dev;
+	int size;
+
+	size = rx_ring->length * sizeof(struct bcm4908_enet_dma_ring_bd);
+	if (rx_ring->cpu_addr)
+		dma_free_coherent(dev, size, rx_ring->cpu_addr, rx_ring->dma_addr);
+	kfree(rx_ring->slots);
+
+	size = tx_ring->length * sizeof(struct bcm4908_enet_dma_ring_bd);
+	if (tx_ring->cpu_addr)
+		dma_free_coherent(dev, size, tx_ring->cpu_addr, tx_ring->dma_addr);
+	kfree(tx_ring->slots);
+}
+
+static int bcm4908_enet_dma_alloc(struct bcm4908_enet *enet)
+{
+	struct bcm4908_enet_dma_ring *tx_ring = &enet->tx_ring;
+	struct bcm4908_enet_dma_ring *rx_ring = &enet->rx_ring;
+	struct device *dev = enet->dev;
+	int err;
+
+	tx_ring->length = ENET_TX_BDS_NUM;
+	tx_ring->is_tx = 1;
+	tx_ring->cfg_block = ENET_DMA_CH_TX_CFG;
+	tx_ring->st_ram_block = ENET_DMA_CH_TX_STATE_RAM;
+	err = bcm4908_dma_alloc_buf_descs(enet, tx_ring);
+	if (err) {
+		dev_err(dev, "Failed to alloc TX buf descriptors: %d\n", err);
+		return err;
+	}
+
+	rx_ring->length = ENET_RX_BDS_NUM;
+	rx_ring->is_tx = 0;
+	rx_ring->cfg_block = ENET_DMA_CH_RX_CFG;
+	rx_ring->st_ram_block = ENET_DMA_CH_RX_STATE_RAM;
+	err = bcm4908_dma_alloc_buf_descs(enet, rx_ring);
+	if (err) {
+		dev_err(dev, "Failed to alloc RX buf descriptors: %d\n", err);
+		bcm4908_enet_dma_free(enet);
+		return err;
+	}
+
+	return 0;
+}
+
+static void bcm4908_enet_dma_reset(struct bcm4908_enet *enet)
+{
+	struct bcm4908_enet_dma_ring *rings[] = { &enet->rx_ring, &enet->tx_ring };
+	int i;
+
+	/* Disable the DMA controller and channel */
+	for (i = 0; i < ARRAY_SIZE(rings); i++)
+		enet_write(enet, rings[i]->cfg_block + ENET_DMA_CH_CFG, 0);
+	enet_maskset(enet, ENET_DMA_CONTROLLER_CFG, ENET_DMA_CTRL_CFG_MASTER_EN, 0);
+
+	/* Reset channels state */
+	for (i = 0; i < ARRAY_SIZE(rings); i++) {
+		struct bcm4908_enet_dma_ring *ring = rings[i];
+
+		enet_write(enet, ring->st_ram_block + ENET_DMA_CH_STATE_RAM_BASE_DESC_PTR, 0);
+		enet_write(enet, ring->st_ram_block + ENET_DMA_CH_STATE_RAM_STATE_DATA, 0);
+		enet_write(enet, ring->st_ram_block + ENET_DMA_CH_STATE_RAM_DESC_LEN_STATUS, 0);
+		enet_write(enet, ring->st_ram_block + ENET_DMA_CH_STATE_RAM_DESC_BASE_BUFPTR, 0);
+	}
+}
+
+static int bcm4908_enet_dma_alloc_rx_buf(struct bcm4908_enet *enet, unsigned int idx)
+{
+	struct bcm4908_enet_dma_ring_bd *buf_desc = &enet->rx_ring.buf_desc[idx];
+	struct bcm4908_enet_dma_ring_slot *slot = &enet->rx_ring.slots[idx];
+	struct device *dev = enet->dev;
+	u32 tmp;
+	int err;
+
+	slot->buf = napi_alloc_frag(ENET_RX_SKB_BUF_ALLOC_SIZE);
+	if (!slot->buf)
+		return -ENOMEM;
+
+	slot->dma_addr = dma_map_single(dev, slot->buf + ENET_RX_BUF_DMA_OFFSET,
+					ENET_RX_BUF_DMA_SIZE, DMA_FROM_DEVICE);
+	err = dma_mapping_error(dev, slot->dma_addr);
+	if (err) {
+		dev_err(dev, "Failed to map DMA buffer: %d\n", err);
+		skb_free_frag(slot->buf);
+		slot->buf = NULL;
+		return err;
+	}
+
+	tmp = ENET_RX_BUF_DMA_SIZE << DMA_CTL_LEN_DESC_BUFLENGTH_SHIFT;
+	tmp |= DMA_CTL_STATUS_OWN;
+	if (idx == enet->rx_ring.length - 1)
+		tmp |= DMA_CTL_STATUS_WRAP;
+	buf_desc->ctl = cpu_to_le32(tmp);
+	buf_desc->addr = cpu_to_le32(slot->dma_addr);
+
+	return 0;
+}
+
+static void bcm4908_enet_dma_ring_init(struct bcm4908_enet *enet,
+				       struct bcm4908_enet_dma_ring *ring)
+{
+	int reset_channel = 0; /* We support only 1 main channel (with TX and RX) */
+	int reset_subch = ring->is_tx ? 1 : 0;
+
+	/* Reset the DMA channel */
+	enet_write(enet, ENET_DMA_CTRL_CHANNEL_RESET, BIT(reset_channel * 2 + reset_subch));
+	enet_write(enet, ENET_DMA_CTRL_CHANNEL_RESET, 0);
+
+	enet_write(enet, ring->cfg_block + ENET_DMA_CH_CFG, 0);
+	enet_write(enet, ring->cfg_block + ENET_DMA_CH_CFG_MAX_BURST, ENET_DMA_MAX_BURST_LEN);
+	enet_write(enet, ring->cfg_block + ENET_DMA_CH_CFG_INT_MASK, 0);
+
+	enet_write(enet, ring->st_ram_block + ENET_DMA_CH_STATE_RAM_BASE_DESC_PTR,
+		   (uint32_t)ring->dma_addr);
+
+	ring->read_idx = 0;
+	ring->write_idx = 0;
+}
+
+static void bcm4908_enet_dma_uninit(struct bcm4908_enet *enet)
+{
+	struct bcm4908_enet_dma_ring *rx_ring = &enet->rx_ring;
+	struct bcm4908_enet_dma_ring_slot *slot;
+	struct device *dev = enet->dev;
+	int i;
+
+	for (i = rx_ring->length - 1; i >= 0; i--) {
+		slot = &rx_ring->slots[i];
+		if (!slot->buf)
+			continue;
+		dma_unmap_single(dev, slot->dma_addr, slot->len, DMA_FROM_DEVICE);
+		skb_free_frag(slot->buf);
+		slot->buf = NULL;
+	}
+}
+
+static int bcm4908_enet_dma_init(struct bcm4908_enet *enet)
+{
+	struct bcm4908_enet_dma_ring *rx_ring = &enet->rx_ring;
+	struct device *dev = enet->dev;
+	int err;
+	int i;
+
+	for (i = 0; i < rx_ring->length; i++) {
+		err = bcm4908_enet_dma_alloc_rx_buf(enet, i);
+		if (err) {
+			dev_err(dev, "Failed to alloc RX buffer: %d\n", err);
+			bcm4908_enet_dma_uninit(enet);
+			return err;
+		}
+	}
+
+	bcm4908_enet_dma_ring_init(enet, &enet->tx_ring);
+	bcm4908_enet_dma_ring_init(enet, &enet->rx_ring);
+
+	return 0;
+}
+
+static void bcm4908_enet_dma_tx_ring_enable(struct bcm4908_enet *enet,
+					    struct bcm4908_enet_dma_ring *ring)
+{
+	enet_write(enet, ring->cfg_block + ENET_DMA_CH_CFG, ENET_DMA_CH_CFG_ENABLE);
+}
+
+static void bcm4908_enet_dma_tx_ring_disable(struct bcm4908_enet *enet,
+					     struct bcm4908_enet_dma_ring *ring)
+{
+	enet_write(enet, ring->cfg_block + ENET_DMA_CH_CFG, 0);
+}
+
+static void bcm4908_enet_dma_rx_ring_enable(struct bcm4908_enet *enet,
+					    struct bcm4908_enet_dma_ring *ring)
+{
+	enet_set(enet, ring->cfg_block + ENET_DMA_CH_CFG, ENET_DMA_CH_CFG_ENABLE);
+}
+
+static void bcm4908_enet_dma_rx_ring_disable(struct bcm4908_enet *enet,
+					     struct bcm4908_enet_dma_ring *ring)
+{
+	unsigned long deadline;
+	u32 tmp;
+
+	enet_maskset(enet, ring->cfg_block + ENET_DMA_CH_CFG, ENET_DMA_CH_CFG_ENABLE, 0);
+
+	deadline = jiffies + usecs_to_jiffies(2000);
+	do {
+		tmp = enet_read(enet, ring->cfg_block + ENET_DMA_CH_CFG);
+		if (!(tmp & ENET_DMA_CH_CFG_ENABLE))
+			return;
+		enet_maskset(enet, ring->cfg_block + ENET_DMA_CH_CFG, ENET_DMA_CH_CFG_ENABLE, 0);
+		usleep_range(10, 30);
+	} while (!time_after_eq(jiffies, deadline));
+
+	dev_warn(enet->dev, "Timeout waiting for DMA TX stop\n");
+}
+
+/***
+ * Ethernet driver
+ */
+
+static void bcm4908_enet_gmac_init(struct bcm4908_enet *enet)
+{
+	u32 cmd;
+
+	bcm4908_enet_set_mtu(enet, enet->netdev->mtu);
+
+	cmd = enet_umac_read(enet, UMAC_CMD);
+	enet_umac_write(enet, UMAC_CMD, cmd | CMD_SW_RESET);
+	enet_umac_write(enet, UMAC_CMD, cmd & ~CMD_SW_RESET);
+
+	enet_set(enet, ENET_FLUSH, ENET_FLUSH_RXFIFO_FLUSH | ENET_FLUSH_TXFIFO_FLUSH);
+	enet_maskset(enet, ENET_FLUSH, ENET_FLUSH_RXFIFO_FLUSH | ENET_FLUSH_TXFIFO_FLUSH, 0);
+
+	enet_set(enet, ENET_MIB_CTRL, ENET_MIB_CTRL_CLR_MIB);
+	enet_maskset(enet, ENET_MIB_CTRL, ENET_MIB_CTRL_CLR_MIB, 0);
+
+	cmd = enet_umac_read(enet, UMAC_CMD);
+	cmd &= ~(CMD_SPEED_MASK << CMD_SPEED_SHIFT);
+	cmd &= ~CMD_TX_EN;
+	cmd &= ~CMD_RX_EN;
+	cmd |= CMD_SPEED_1000 << CMD_SPEED_SHIFT;
+	enet_umac_write(enet, UMAC_CMD, cmd);
+
+	enet_maskset(enet, ENET_GMAC_STATUS,
+		     ENET_GMAC_STATUS_ETH_SPEED_MASK |
+		     ENET_GMAC_STATUS_HD |
+		     ENET_GMAC_STATUS_AUTO_CFG_EN |
+		     ENET_GMAC_STATUS_LINK_UP,
+		     ENET_GMAC_STATUS_ETH_SPEED_1000 |
+		     ENET_GMAC_STATUS_AUTO_CFG_EN |
+		     ENET_GMAC_STATUS_LINK_UP);
+}
+
+static irqreturn_t bcm4908_enet_irq_handler(int irq, void *dev_id)
+{
+	struct bcm4908_enet *enet = dev_id;
+	struct bcm4908_enet_dma_ring *ring;
+
+	ring = (irq == enet->irq_tx) ? &enet->tx_ring : &enet->rx_ring;
+
+	bcm4908_enet_dma_ring_intrs_off(enet, ring);
+	bcm4908_enet_dma_ring_intrs_ack(enet, ring);
+
+	napi_schedule(&ring->napi);
+
+	return IRQ_HANDLED;
+}
+
+static int bcm4908_enet_open(struct net_device *netdev)
+{
+	struct bcm4908_enet *enet = netdev_priv(netdev);
+	struct bcm4908_enet_dma_ring *tx_ring = &enet->tx_ring;
+	struct bcm4908_enet_dma_ring *rx_ring = &enet->rx_ring;
+	struct device *dev = enet->dev;
+	int err;
+
+	err = request_irq(netdev->irq, bcm4908_enet_irq_handler, 0, "enet", enet);
+	if (err) {
+		dev_err(dev, "Failed to request IRQ %d: %d\n", netdev->irq, err);
+		return err;
+	}
+
+	if (enet->irq_tx > 0) {
+		err = request_irq(enet->irq_tx, bcm4908_enet_irq_handler, 0,
+				  "tx", enet);
+		if (err) {
+			dev_err(dev, "Failed to request IRQ %d: %d\n",
+				enet->irq_tx, err);
+			free_irq(netdev->irq, enet);
+			return err;
+		}
+	}
+
+	bcm4908_enet_gmac_init(enet);
+	bcm4908_enet_dma_reset(enet);
+	bcm4908_enet_dma_init(enet);
+
+	enet_umac_set(enet, UMAC_CMD, CMD_TX_EN | CMD_RX_EN);
+
+	enet_set(enet, ENET_DMA_CONTROLLER_CFG, ENET_DMA_CTRL_CFG_MASTER_EN);
+	enet_maskset(enet, ENET_DMA_CONTROLLER_CFG, ENET_DMA_CTRL_CFG_FLOWC_CH1_EN, 0);
+
+	if (enet->irq_tx > 0) {
+		napi_enable(&tx_ring->napi);
+		bcm4908_enet_dma_ring_intrs_ack(enet, tx_ring);
+		bcm4908_enet_dma_ring_intrs_on(enet, tx_ring);
+	}
+
+	bcm4908_enet_dma_rx_ring_enable(enet, rx_ring);
+	napi_enable(&rx_ring->napi);
+	netif_carrier_on(netdev);
+	netif_start_queue(netdev);
+	bcm4908_enet_dma_ring_intrs_ack(enet, rx_ring);
+	bcm4908_enet_dma_ring_intrs_on(enet, rx_ring);
+
+	return 0;
+}
+
+static int bcm4908_enet_stop(struct net_device *netdev)
+{
+	struct bcm4908_enet *enet = netdev_priv(netdev);
+	struct bcm4908_enet_dma_ring *tx_ring = &enet->tx_ring;
+	struct bcm4908_enet_dma_ring *rx_ring = &enet->rx_ring;
+
+	netif_stop_queue(netdev);
+	netif_carrier_off(netdev);
+	napi_disable(&rx_ring->napi);
+	napi_disable(&tx_ring->napi);
+	netdev_reset_queue(netdev);
+
+	bcm4908_enet_dma_rx_ring_disable(enet, &enet->rx_ring);
+	bcm4908_enet_dma_tx_ring_disable(enet, &enet->tx_ring);
+
+	bcm4908_enet_dma_uninit(enet);
+
+	free_irq(enet->irq_tx, enet);
+	free_irq(enet->netdev->irq, enet);
+
+	return 0;
+}
+
+static netdev_tx_t bcm4908_enet_start_xmit(struct sk_buff *skb, struct net_device *netdev)
+{
+	struct bcm4908_enet *enet = netdev_priv(netdev);
+	struct bcm4908_enet_dma_ring *ring = &enet->tx_ring;
+	struct bcm4908_enet_dma_ring_slot *slot;
+	struct device *dev = enet->dev;
+	struct bcm4908_enet_dma_ring_bd *buf_desc;
+	int free_buf_descs;
+	u32 tmp;
+
+	/* Free transmitted skbs */
+	if (enet->irq_tx < 0 &&
+	    !(le32_to_cpu(ring->buf_desc[ring->read_idx].ctl) & DMA_CTL_STATUS_OWN))
+		napi_schedule(&enet->tx_ring.napi);
+
+	/* Don't use the last empty buf descriptor */
+	if (ring->read_idx <= ring->write_idx)
+		free_buf_descs = ring->read_idx - ring->write_idx + ring->length;
+	else
+		free_buf_descs = ring->read_idx - ring->write_idx;
+	if (free_buf_descs < 2) {
+		netif_stop_queue(netdev);
+		return NETDEV_TX_BUSY;
+	}
+
+	/* Hardware removes OWN bit after sending data */
+	buf_desc = &ring->buf_desc[ring->write_idx];
+	if (unlikely(le32_to_cpu(buf_desc->ctl) & DMA_CTL_STATUS_OWN)) {
+		netif_stop_queue(netdev);
+		return NETDEV_TX_BUSY;
+	}
+
+	slot = &ring->slots[ring->write_idx];
+	slot->skb = skb;
+	slot->len = skb->len;
+	slot->dma_addr = dma_map_single(dev, skb->data, skb->len, DMA_TO_DEVICE);
+	if (unlikely(dma_mapping_error(dev, slot->dma_addr)))
+		return NETDEV_TX_BUSY;
+
+	tmp = skb->len << DMA_CTL_LEN_DESC_BUFLENGTH_SHIFT;
+	tmp |= DMA_CTL_STATUS_OWN;
+	tmp |= DMA_CTL_STATUS_SOP;
+	tmp |= DMA_CTL_STATUS_EOP;
+	tmp |= DMA_CTL_STATUS_APPEND_CRC;
+	if (ring->write_idx + 1 == ring->length - 1)
+		tmp |= DMA_CTL_STATUS_WRAP;
+
+	netdev_sent_queue(enet->netdev, skb->len);
+
+	buf_desc->addr = cpu_to_le32((uint32_t)slot->dma_addr);
+	buf_desc->ctl = cpu_to_le32(tmp);
+
+	bcm4908_enet_dma_tx_ring_enable(enet, &enet->tx_ring);
+
+	if (++ring->write_idx == ring->length - 1)
+		ring->write_idx = 0;
+
+	return NETDEV_TX_OK;
+}
+
+static int bcm4908_enet_poll_rx(struct napi_struct *napi, int weight)
+{
+	struct bcm4908_enet_dma_ring *rx_ring = container_of(napi, struct bcm4908_enet_dma_ring, napi);
+	struct bcm4908_enet *enet = container_of(rx_ring, struct bcm4908_enet, rx_ring);
+	struct device *dev = enet->dev;
+	int handled = 0;
+
+	while (handled < weight) {
+		struct bcm4908_enet_dma_ring_bd *buf_desc;
+		struct bcm4908_enet_dma_ring_slot slot;
+		struct sk_buff *skb;
+		u32 ctl;
+		int len;
+		int err;
+
+		buf_desc = &enet->rx_ring.buf_desc[enet->rx_ring.read_idx];
+		ctl = le32_to_cpu(buf_desc->ctl);
+		if (ctl & DMA_CTL_STATUS_OWN)
+			break;
+
+		slot = enet->rx_ring.slots[enet->rx_ring.read_idx];
+
+		/* Provide new buffer before unpinning the old one */
+		err = bcm4908_enet_dma_alloc_rx_buf(enet, enet->rx_ring.read_idx);
+		if (err)
+			break;
+
+		if (++enet->rx_ring.read_idx == enet->rx_ring.length)
+			enet->rx_ring.read_idx = 0;
+
+		len = (ctl & DMA_CTL_LEN_DESC_BUFLENGTH) >> DMA_CTL_LEN_DESC_BUFLENGTH_SHIFT;
+
+		if (len < ETH_ZLEN ||
+		    (ctl & (DMA_CTL_STATUS_SOP | DMA_CTL_STATUS_EOP)) != (DMA_CTL_STATUS_SOP | DMA_CTL_STATUS_EOP)) {
+			skb_free_frag(slot.buf);
+			enet->netdev->stats.rx_dropped++;
+			break;
+		}
+
+		dma_unmap_single(dev, slot.dma_addr, ENET_RX_BUF_DMA_SIZE, DMA_FROM_DEVICE);
+
+		skb = build_skb(slot.buf, ENET_RX_SKB_BUF_ALLOC_SIZE);
+		if (unlikely(!skb)) {
+			skb_free_frag(slot.buf);
+			enet->netdev->stats.rx_dropped++;
+			break;
+		}
+		skb_reserve(skb, ENET_RX_BUF_DMA_OFFSET);
+		skb_put(skb, len - ETH_FCS_LEN);
+		skb->protocol = eth_type_trans(skb, enet->netdev);
+
+		netif_receive_skb(skb);
+
+		enet->netdev->stats.rx_packets++;
+		enet->netdev->stats.rx_bytes += len;
+
+		handled++;
+	}
+
+	if (handled < weight) {
+		napi_complete_done(napi, handled);
+		bcm4908_enet_dma_ring_intrs_on(enet, rx_ring);
+	}
+
+	/* Hardware could disable ring if it run out of descriptors */
+	bcm4908_enet_dma_rx_ring_enable(enet, &enet->rx_ring);
+
+	return handled;
+}
+
+static int bcm4908_enet_poll_tx(struct napi_struct *napi, int weight)
+{
+	struct bcm4908_enet_dma_ring *tx_ring = container_of(napi, struct bcm4908_enet_dma_ring, napi);
+	struct bcm4908_enet *enet = container_of(tx_ring, struct bcm4908_enet, tx_ring);
+	struct bcm4908_enet_dma_ring_bd *buf_desc;
+	struct bcm4908_enet_dma_ring_slot *slot;
+	struct device *dev = enet->dev;
+	unsigned int bytes = 0;
+	int handled = 0;
+
+	while (handled < weight && tx_ring->read_idx != tx_ring->write_idx) {
+		buf_desc = &tx_ring->buf_desc[tx_ring->read_idx];
+		if (le32_to_cpu(buf_desc->ctl) & DMA_CTL_STATUS_OWN)
+			break;
+		slot = &tx_ring->slots[tx_ring->read_idx];
+
+		dma_unmap_single(dev, slot->dma_addr, slot->len, DMA_TO_DEVICE);
+		dev_kfree_skb(slot->skb);
+
+		handled++;
+		bytes += slot->len;
+
+		if (++tx_ring->read_idx == tx_ring->length)
+			tx_ring->read_idx = 0;
+	}
+
+	netdev_completed_queue(enet->netdev, handled, bytes);
+	enet->netdev->stats.tx_packets += handled;
+	enet->netdev->stats.tx_bytes += bytes;
+
+	if (handled < weight) {
+		napi_complete_done(napi, handled);
+		bcm4908_enet_dma_ring_intrs_on(enet, tx_ring);
+	}
+
+	if (netif_queue_stopped(enet->netdev))
+		netif_wake_queue(enet->netdev);
+
+	return handled;
+}
+
+static int bcm4908_enet_change_mtu(struct net_device *netdev, int new_mtu)
+{
+	struct bcm4908_enet *enet = netdev_priv(netdev);
+
+	bcm4908_enet_set_mtu(enet, new_mtu);
+
+	return 0;
+}
+
+static const struct net_device_ops bcm4908_enet_netdev_ops = {
+	.ndo_open = bcm4908_enet_open,
+	.ndo_stop = bcm4908_enet_stop,
+	.ndo_start_xmit = bcm4908_enet_start_xmit,
+	.ndo_set_mac_address = eth_mac_addr,
+	.ndo_change_mtu = bcm4908_enet_change_mtu,
+};
+
+static int bcm4908_enet_probe(struct platform_device *pdev)
+{
+	struct device *dev = &pdev->dev;
+	struct net_device *netdev;
+	struct bcm4908_enet *enet;
+	int err;
+
+	netdev = devm_alloc_etherdev(dev, sizeof(*enet));
+	if (!netdev)
+		return -ENOMEM;
+
+	enet = netdev_priv(netdev);
+	enet->dev = dev;
+	enet->netdev = netdev;
+
+	enet->base = devm_platform_ioremap_resource(pdev, 0);
+	if (IS_ERR(enet->base)) {
+		dev_err(dev, "Failed to map registers: %ld\n", PTR_ERR(enet->base));
+		return PTR_ERR(enet->base);
+	}
+
+	netdev->irq = platform_get_irq_byname(pdev, "rx");
+	if (netdev->irq < 0)
+		return netdev->irq;
+
+	enet->irq_tx = platform_get_irq_byname(pdev, "tx");
+
+	err = dma_set_coherent_mask(dev, DMA_BIT_MASK(32));
+	if (err)
+		return err;
+
+	err = bcm4908_enet_dma_alloc(enet);
+	if (err)
+		return err;
+
+	SET_NETDEV_DEV(netdev, &pdev->dev);
+	err = of_get_ethdev_address(dev->of_node, netdev);
+	if (err == -EPROBE_DEFER)
+		goto err_dma_free;
+	if (err)
+		eth_hw_addr_random(netdev);
+	netdev->netdev_ops = &bcm4908_enet_netdev_ops;
+	netdev->min_mtu = ETH_ZLEN;
+	netdev->mtu = ETH_DATA_LEN;
+	netdev->max_mtu = ENET_MTU_MAX;
+	netif_napi_add_tx(netdev, &enet->tx_ring.napi, bcm4908_enet_poll_tx);
+	netif_napi_add(netdev, &enet->rx_ring.napi, bcm4908_enet_poll_rx);
+
+	err = register_netdev(netdev);
+	if (err)
+		goto err_dma_free;
+
+	platform_set_drvdata(pdev, enet);
+
+	return 0;
+
+err_dma_free:
+	bcm4908_enet_dma_free(enet);
+
+	return err;
+}
+
+static void bcm4908_enet_remove(struct platform_device *pdev)
+{
+	struct bcm4908_enet *enet = platform_get_drvdata(pdev);
+
+	unregister_netdev(enet->netdev);
+	netif_napi_del(&enet->rx_ring.napi);
+	netif_napi_del(&enet->tx_ring.napi);
+	bcm4908_enet_dma_free(enet);
+}
+
+static const struct of_device_id bcm4908_enet_of_match[] = {
+	{ .compatible = "brcm,bcm4908-enet"},
+	{},
+};
+
+static struct platform_driver bcm4908_enet_driver = {
+	.driver = {
+		.name = "bcm4908_enet",
+		.of_match_table = bcm4908_enet_of_match,
+	},
+	.probe	= bcm4908_enet_probe,
+	.remove = bcm4908_enet_remove,
+};
+module_platform_driver(bcm4908_enet_driver);
+
+MODULE_DESCRIPTION("Broadcom BCM4908 Gigabit Ethernet driver");
+MODULE_LICENSE("GPL v2");
+MODULE_DEVICE_TABLE(of, bcm4908_enet_of_match);
diff --git a/drivers/net/ethernet/broadcom/bcm4908_enet.h b/drivers/net/ethernet/broadcom/bcm4908_enet.h
new file mode 100644
index 000000000000..8a3ede2da537
--- /dev/null
+++ b/drivers/net/ethernet/broadcom/bcm4908_enet.h
@@ -0,0 +1,96 @@
+/* SPDX-License-Identifier: GPL-2.0-only */
+#ifndef __BCM4908_ENET_H
+#define __BCM4908_ENET_H
+
+#define ENET_CONTROL					0x000
+#define ENET_MIB_CTRL					0x004
+#define  ENET_MIB_CTRL_CLR_MIB				0x00000001
+#define ENET_RX_ERR_MASK				0x008
+#define ENET_MIB_MAX_PKT_SIZE				0x00C
+#define  ENET_MIB_MAX_PKT_SIZE_VAL			0x00003fff
+#define ENET_DIAG_OUT					0x01c
+#define ENET_ENABLE_DROP_PKT				0x020
+#define ENET_IRQ_ENABLE					0x024
+#define  ENET_IRQ_ENABLE_OVFL				0x00000001
+#define ENET_GMAC_STATUS				0x028
+#define  ENET_GMAC_STATUS_ETH_SPEED_MASK		0x00000003
+#define  ENET_GMAC_STATUS_ETH_SPEED_10			0x00000000
+#define  ENET_GMAC_STATUS_ETH_SPEED_100			0x00000001
+#define  ENET_GMAC_STATUS_ETH_SPEED_1000		0x00000002
+#define  ENET_GMAC_STATUS_HD				0x00000004
+#define  ENET_GMAC_STATUS_AUTO_CFG_EN			0x00000008
+#define  ENET_GMAC_STATUS_LINK_UP			0x00000010
+#define ENET_IRQ_STATUS					0x02c
+#define  ENET_IRQ_STATUS_OVFL				0x00000001
+#define ENET_OVERFLOW_COUNTER				0x030
+#define ENET_FLUSH					0x034
+#define  ENET_FLUSH_RXFIFO_FLUSH			0x00000001
+#define  ENET_FLUSH_TXFIFO_FLUSH			0x00000002
+#define ENET_RSV_SELECT					0x038
+#define ENET_BP_FORCE					0x03c
+#define  ENET_BP_FORCE_FORCE				0x00000001
+#define ENET_DMA_RX_OK_TO_SEND_COUNT			0x040
+#define  ENET_DMA_RX_OK_TO_SEND_COUNT_VAL		0x0000000f
+#define ENET_TX_CRC_CTRL				0x044
+#define ENET_MIB					0x200
+#define ENET_UNIMAC					0x400
+#define ENET_DMA					0x800
+#define ENET_DMA_CONTROLLER_CFG				0x800
+#define  ENET_DMA_CTRL_CFG_MASTER_EN			0x00000001
+#define  ENET_DMA_CTRL_CFG_FLOWC_CH1_EN			0x00000002
+#define  ENET_DMA_CTRL_CFG_FLOWC_CH3_EN			0x00000004
+#define ENET_DMA_FLOWCTL_CH1_THRESH_LO			0x804
+#define ENET_DMA_FLOWCTL_CH1_THRESH_HI			0x808
+#define ENET_DMA_FLOWCTL_CH1_ALLOC			0x80c
+#define  ENET_DMA_FLOWCTL_CH1_ALLOC_FORCE		0x80000000
+#define ENET_DMA_FLOWCTL_CH3_THRESH_LO			0x810
+#define ENET_DMA_FLOWCTL_CH3_THRESH_HI			0x814
+#define ENET_DMA_FLOWCTL_CH3_ALLOC			0x818
+#define ENET_DMA_FLOWCTL_CH5_THRESH_LO			0x81C
+#define ENET_DMA_FLOWCTL_CH5_THRESH_HI			0x820
+#define ENET_DMA_FLOWCTL_CH5_ALLOC			0x824
+#define ENET_DMA_FLOWCTL_CH7_THRESH_LO			0x828
+#define ENET_DMA_FLOWCTL_CH7_THRESH_HI			0x82C
+#define ENET_DMA_FLOWCTL_CH7_ALLOC			0x830
+#define ENET_DMA_CTRL_CHANNEL_RESET			0x834
+#define ENET_DMA_CTRL_CHANNEL_DEBUG			0x838
+#define ENET_DMA_CTRL_GLOBAL_INTERRUPT_STATUS		0x840
+#define ENET_DMA_CTRL_GLOBAL_INTERRUPT_MASK		0x844
+#define ENET_DMA_CH0_CFG				0xa00		/* RX */
+#define ENET_DMA_CH1_CFG				0xa10		/* TX */
+#define ENET_DMA_CH0_STATE_RAM				0xc00		/* RX */
+#define ENET_DMA_CH1_STATE_RAM				0xc10		/* TX */
+
+#define ENET_DMA_CH_CFG					0x00		/* assorted configuration */
+#define  ENET_DMA_CH_CFG_ENABLE				0x00000001	/* set to enable channel */
+#define  ENET_DMA_CH_CFG_PKT_HALT			0x00000002	/* idle after an EOP flag is detected */
+#define  ENET_DMA_CH_CFG_BURST_HALT			0x00000004	/* idle after finish current memory burst */
+#define ENET_DMA_CH_CFG_INT_STAT			0x04		/* interrupts control and status */
+#define ENET_DMA_CH_CFG_INT_MASK			0x08		/* interrupts mask */
+#define  ENET_DMA_CH_CFG_INT_BUFF_DONE			0x00000001	/* buffer done */
+#define  ENET_DMA_CH_CFG_INT_DONE			0x00000002	/* packet xfer complete */
+#define  ENET_DMA_CH_CFG_INT_NO_DESC			0x00000004	/* no valid descriptors */
+#define  ENET_DMA_CH_CFG_INT_RX_ERROR			0x00000008	/* rxdma detect client protocol error */
+#define ENET_DMA_CH_CFG_MAX_BURST			0x0c		/* max burst length permitted */
+#define  ENET_DMA_CH_CFG_MAX_BURST_DESCSIZE_SEL		0x00040000	/* DMA Descriptor Size Selection */
+#define ENET_DMA_CH_CFG_SIZE				0x10
+
+#define ENET_DMA_CH_STATE_RAM_BASE_DESC_PTR		0x00		/* descriptor ring start address */
+#define ENET_DMA_CH_STATE_RAM_STATE_DATA		0x04		/* state/bytes done/ring offset */
+#define ENET_DMA_CH_STATE_RAM_DESC_LEN_STATUS		0x08		/* buffer descriptor status and len */
+#define ENET_DMA_CH_STATE_RAM_DESC_BASE_BUFPTR		0x0c		/* buffer descrpitor current processing */
+#define ENET_DMA_CH_STATE_RAM_SIZE			0x10
+
+#define DMA_CTL_STATUS_APPEND_CRC			0x00000100
+#define DMA_CTL_STATUS_APPEND_BRCM_TAG			0x00000200
+#define DMA_CTL_STATUS_PRIO				0x00000C00  /* Prio for Tx */
+#define DMA_CTL_STATUS_WRAP				0x00001000  /* */
+#define DMA_CTL_STATUS_SOP				0x00002000  /* first buffer in packet */
+#define DMA_CTL_STATUS_EOP				0x00004000  /* last buffer in packet */
+#define DMA_CTL_STATUS_OWN				0x00008000  /* cleared by DMA, set by SW */
+#define DMA_CTL_LEN_DESC_BUFLENGTH			0x0fff0000
+#define DMA_CTL_LEN_DESC_BUFLENGTH_SHIFT		16
+#define DMA_CTL_LEN_DESC_MULTICAST			0x40000000
+#define DMA_CTL_LEN_DESC_USEFPM				0x80000000
+
+#endif
diff --git a/drivers/net/ethernet/broadcom/bcm63xx_enet.c b/drivers/net/ethernet/broadcom/bcm63xx_enet.c
index 50d88d3e03b6..92204fea1f08 100644
--- a/drivers/net/ethernet/broadcom/bcm63xx_enet.c
+++ b/drivers/net/ethernet/broadcom/bcm63xx_enet.c
@@ -1,21 +1,8 @@
+// SPDX-License-Identifier: GPL-2.0-or-later
 /*
  * Driver for BCM963xx builtin Ethernet mac
  *
  * Copyright (C) 2008 Maxime Bizon <mbizon@freebox.fr>
- *
- * This program is free software; you can redistribute it and/or modify
- * it under the terms of the GNU General Public License as published by
- * the Free Software Foundation; either version 2 of the License, or
- * (at your option) any later version.
- *
- * This program is distributed in the hope that it will be useful,
- * but WITHOUT ANY WARRANTY; without even the implied warranty of
- * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
- * GNU General Public License for more details.
- *
- * You should have received a copy of the GNU General Public License
- * along with this program; if not, write to the Free Software
- * Foundation, Inc., 675 Mass Ave, Cambridge, MA 02139, USA.
  */
 #include <linux/init.h>
 #include <linux/interrupt.h>
@@ -35,7 +22,6 @@
 #include "bcm63xx_enet.h"
 
 static char bcm_enet_driver_name[] = "bcm63xx_enet";
-static char bcm_enet_driver_version[] = "1.0";
 
 static int copybreak __read_mostly = 128;
 module_param(copybreak, int, 0);
@@ -234,7 +220,7 @@ static void bcm_enet_mdio_write_mii(struct net_device *dev, int mii_id,
 /*
  * refill rx queue
  */
-static int bcm_enet_refill_rx(struct net_device *dev)
+static int bcm_enet_refill_rx(struct net_device *dev, bool napi_mode)
 {
 	struct bcm_enet_priv *priv;
 
@@ -242,26 +228,29 @@ static int bcm_enet_refill_rx(struct net_device *dev)
 
 	while (priv->rx_desc_count < priv->rx_ring_size) {
 		struct bcm_enet_desc *desc;
-		struct sk_buff *skb;
-		dma_addr_t p;
 		int desc_idx;
 		u32 len_stat;
 
 		desc_idx = priv->rx_dirty_desc;
 		desc = &priv->rx_desc_cpu[desc_idx];
 
-		if (!priv->rx_skb[desc_idx]) {
-			skb = netdev_alloc_skb(dev, priv->rx_skb_size);
-			if (!skb)
+		if (!priv->rx_buf[desc_idx]) {
+			void *buf;
+
+			if (likely(napi_mode))
+				buf = napi_alloc_frag(priv->rx_frag_size);
+			else
+				buf = netdev_alloc_frag(priv->rx_frag_size);
+			if (unlikely(!buf))
 				break;
-			priv->rx_skb[desc_idx] = skb;
-			p = dma_map_single(&priv->pdev->dev, skb->data,
-					   priv->rx_skb_size,
-					   DMA_FROM_DEVICE);
-			desc->address = p;
+			priv->rx_buf[desc_idx] = buf;
+			desc->address = dma_map_single(&priv->pdev->dev,
+						       buf + priv->rx_buf_offset,
+						       priv->rx_buf_size,
+						       DMA_FROM_DEVICE);
 		}
 
-		len_stat = priv->rx_skb_size << DMADESC_LENGTH_SHIFT;
+		len_stat = priv->rx_buf_size << DMADESC_LENGTH_SHIFT;
 		len_stat |= DMADESC_OWNER_MASK;
 		if (priv->rx_dirty_desc == priv->rx_ring_size - 1) {
 			len_stat |= (DMADESC_WRAP_MASK >> priv->dma_desc_shift);
@@ -295,16 +284,13 @@ static int bcm_enet_refill_rx(struct net_device *dev)
 /*
  * timer callback to defer refill rx queue in case we're OOM
  */
-static void bcm_enet_refill_rx_timer(unsigned long data)
+static void bcm_enet_refill_rx_timer(struct timer_list *t)
 {
-	struct net_device *dev;
-	struct bcm_enet_priv *priv;
-
-	dev = (struct net_device *)data;
-	priv = netdev_priv(dev);
+	struct bcm_enet_priv *priv = timer_container_of(priv, t, rx_timeout);
+	struct net_device *dev = priv->net_dev;
 
 	spin_lock(&priv->rx_lock);
-	bcm_enet_refill_rx((struct net_device *)data);
+	bcm_enet_refill_rx(dev, false);
 	spin_unlock(&priv->rx_lock);
 }
 
@@ -314,10 +300,12 @@ static void bcm_enet_refill_rx_timer(unsigned long data)
 static int bcm_enet_receive_queue(struct net_device *dev, int budget)
 {
 	struct bcm_enet_priv *priv;
+	struct list_head rx_list;
 	struct device *kdev;
 	int processed;
 
 	priv = netdev_priv(dev);
+	INIT_LIST_HEAD(&rx_list);
 	kdev = &priv->pdev->dev;
 	processed = 0;
 
@@ -332,6 +320,7 @@ static int bcm_enet_receive_queue(struct net_device *dev, int budget)
 		int desc_idx;
 		u32 len_stat;
 		unsigned int len;
+		void *buf;
 
 		desc_idx = priv->rx_curr_desc;
 		desc = &priv->rx_desc_cpu[desc_idx];
@@ -350,7 +339,6 @@ static int bcm_enet_receive_queue(struct net_device *dev, int budget)
 		priv->rx_curr_desc++;
 		if (priv->rx_curr_desc == priv->rx_ring_size)
 			priv->rx_curr_desc = 0;
-		priv->rx_desc_count--;
 
 		/* if the packet does not have start of packet _and_
 		 * end of packet flag set, then just recycle it */
@@ -377,16 +365,14 @@ static int bcm_enet_receive_queue(struct net_device *dev, int budget)
 		}
 
 		/* valid packet */
-		skb = priv->rx_skb[desc_idx];
+		buf = priv->rx_buf[desc_idx];
 		len = (len_stat & DMADESC_LENGTH_MASK) >> DMADESC_LENGTH_SHIFT;
 		/* don't include FCS */
 		len -= 4;
 
 		if (len < copybreak) {
-			struct sk_buff *nskb;
-
-			nskb = napi_alloc_skb(&priv->napi, len);
-			if (!nskb) {
+			skb = napi_alloc_skb(&priv->napi, len);
+			if (unlikely(!skb)) {
 				/* forget packet, just rearm desc */
 				dev->stats.rx_dropped++;
 				continue;
@@ -394,26 +380,36 @@ static int bcm_enet_receive_queue(struct net_device *dev, int budget)
 
 			dma_sync_single_for_cpu(kdev, desc->address,
 						len, DMA_FROM_DEVICE);
-			memcpy(nskb->data, skb->data, len);
+			memcpy(skb->data, buf + priv->rx_buf_offset, len);
 			dma_sync_single_for_device(kdev, desc->address,
 						   len, DMA_FROM_DEVICE);
-			skb = nskb;
 		} else {
-			dma_unmap_single(&priv->pdev->dev, desc->address,
-					 priv->rx_skb_size, DMA_FROM_DEVICE);
-			priv->rx_skb[desc_idx] = NULL;
+			dma_unmap_single(kdev, desc->address,
+					 priv->rx_buf_size, DMA_FROM_DEVICE);
+			priv->rx_buf[desc_idx] = NULL;
+
+			skb = napi_build_skb(buf, priv->rx_frag_size);
+			if (unlikely(!skb)) {
+				skb_free_frag(buf);
+				dev->stats.rx_dropped++;
+				continue;
+			}
+			skb_reserve(skb, priv->rx_buf_offset);
 		}
 
 		skb_put(skb, len);
 		skb->protocol = eth_type_trans(skb, dev);
 		dev->stats.rx_packets++;
 		dev->stats.rx_bytes += len;
-		netif_receive_skb(skb);
+		list_add_tail(&skb->list, &rx_list);
 
-	} while (--budget > 0);
+	} while (processed < budget);
+
+	netif_receive_skb_list(&rx_list);
+	priv->rx_desc_count -= processed;
 
 	if (processed || !priv->rx_desc_count) {
-		bcm_enet_refill_rx(dev);
+		bcm_enet_refill_rx(dev, true);
 
 		/* kick rx dma */
 		enet_dmac_writel(priv, priv->dma_chan_en_mask,
@@ -427,12 +423,14 @@ static int bcm_enet_receive_queue(struct net_device *dev, int budget)
 /*
  * try to or force reclaim of transmitted buffers
  */
-static int bcm_enet_tx_reclaim(struct net_device *dev, int force)
+static int bcm_enet_tx_reclaim(struct net_device *dev, int force, int budget)
 {
 	struct bcm_enet_priv *priv;
+	unsigned int bytes;
 	int released;
 
 	priv = netdev_priv(dev);
+	bytes = 0;
 	released = 0;
 
 	while (priv->tx_desc_count < priv->tx_ring_size) {
@@ -469,10 +467,13 @@ static int bcm_enet_tx_reclaim(struct net_device *dev, int force)
 		if (desc->len_stat & DMADESC_UNDER_MASK)
 			dev->stats.tx_errors++;
 
-		dev_kfree_skb(skb);
+		bytes += skb->len;
+		napi_consume_skb(skb, budget);
 		released++;
 	}
 
+	netdev_completed_queue(dev, released, bytes);
+
 	if (netif_queue_stopped(dev) && released)
 		netif_wake_queue(dev);
 
@@ -498,7 +499,7 @@ static int bcm_enet_poll(struct napi_struct *napi, int budget)
 			 ENETDMAC_IR, priv->tx_chan);
 
 	/* reclaim sent skb */
-	bcm_enet_tx_reclaim(dev, 0);
+	bcm_enet_tx_reclaim(dev, 0, budget);
 
 	spin_lock(&priv->rx_lock);
 	rx_work_done = bcm_enet_receive_queue(dev, budget);
@@ -571,12 +572,13 @@ static irqreturn_t bcm_enet_isr_dma(int irq, void *dev_id)
 /*
  * tx request callback
  */
-static int bcm_enet_start_xmit(struct sk_buff *skb, struct net_device *dev)
+static netdev_tx_t
+bcm_enet_start_xmit(struct sk_buff *skb, struct net_device *dev)
 {
 	struct bcm_enet_priv *priv;
 	struct bcm_enet_desc *desc;
 	u32 len_stat;
-	int ret;
+	netdev_tx_t ret;
 
 	priv = netdev_priv(dev);
 
@@ -609,8 +611,7 @@ static int bcm_enet_start_xmit(struct sk_buff *skb, struct net_device *dev)
 			dev_kfree_skb(skb);
 			skb = nskb;
 		}
-		data = skb_put(skb, needed);
-		memset(data, 0, needed);
+		data = skb_put_zero(skb, needed);
 	}
 
 	/* point to the next available desc */
@@ -639,8 +640,11 @@ static int bcm_enet_start_xmit(struct sk_buff *skb, struct net_device *dev)
 	desc->len_stat = len_stat;
 	wmb();
 
+	netdev_sent_queue(dev, skb->len);
+
 	/* kick tx dma */
-	enet_dmac_writel(priv, priv->dma_chan_en_mask,
+	if (!netdev_xmit_more() || !priv->tx_desc_count)
+		enet_dmac_writel(priv, priv->dma_chan_en_mask,
 				 ENETDMAC_CHANCFG, priv->tx_chan);
 
 	/* stop queue if no more desc available */
@@ -666,7 +670,7 @@ static int bcm_enet_set_mac_address(struct net_device *dev, void *p)
 	u32 val;
 
 	priv = netdev_priv(dev);
-	memcpy(dev->dev_addr, addr->sa_data, ETH_ALEN);
+	eth_hw_addr_set(dev, addr->sa_data);
 
 	/* use perfect match register 0 to store my mac address */
 	val = (dev->dev_addr[2] << 24) | (dev->dev_addr[3] << 16) |
@@ -862,6 +866,24 @@ static void bcm_enet_adjust_link(struct net_device *dev)
 		priv->pause_tx ? "tx" : "off");
 }
 
+static void bcm_enet_free_rx_buf_ring(struct device *kdev, struct bcm_enet_priv *priv)
+{
+	int i;
+
+	for (i = 0; i < priv->rx_ring_size; i++) {
+		struct bcm_enet_desc *desc;
+
+		if (!priv->rx_buf[i])
+			continue;
+
+		desc = &priv->rx_desc_cpu[i];
+		dma_unmap_single(kdev, desc->address, priv->rx_buf_size,
+				 DMA_FROM_DEVICE);
+		skb_free_frag(priv->rx_buf[i]);
+	}
+	kfree(priv->rx_buf);
+}
+
 /*
  * open callback, allocate dma rings & buffers and start rx operation
  */
@@ -894,19 +916,10 @@ static int bcm_enet_open(struct net_device *dev)
 		}
 
 		/* mask with MAC supported features */
-		phydev->supported &= (SUPPORTED_10baseT_Half |
-				      SUPPORTED_10baseT_Full |
-				      SUPPORTED_100baseT_Half |
-				      SUPPORTED_100baseT_Full |
-				      SUPPORTED_Autoneg |
-				      SUPPORTED_Pause |
-				      SUPPORTED_MII);
-		phydev->advertising = phydev->supported;
-
-		if (priv->pause_auto && priv->pause_rx && priv->pause_tx)
-			phydev->advertising |= SUPPORTED_Pause;
-		else
-			phydev->advertising &= ~SUPPORTED_Pause;
+		phy_support_sym_pause(phydev);
+		phy_set_max_speed(phydev, SPEED_100);
+		phy_set_sym_pause(phydev, priv->pause_rx, priv->pause_rx,
+				  priv->pause_auto);
 
 		phy_attached_info(phydev);
 
@@ -948,7 +961,7 @@ static int bcm_enet_open(struct net_device *dev)
 
 	/* allocate rx dma ring */
 	size = priv->rx_ring_size * sizeof(struct bcm_enet_desc);
-	p = dma_zalloc_coherent(kdev, size, &priv->rx_desc_dma, GFP_KERNEL);
+	p = dma_alloc_coherent(kdev, size, &priv->rx_desc_dma, GFP_KERNEL);
 	if (!p) {
 		ret = -ENOMEM;
 		goto out_freeirq_tx;
@@ -959,7 +972,7 @@ static int bcm_enet_open(struct net_device *dev)
 
 	/* allocate tx dma ring */
 	size = priv->tx_ring_size * sizeof(struct bcm_enet_desc);
-	p = dma_zalloc_coherent(kdev, size, &priv->tx_desc_dma, GFP_KERNEL);
+	p = dma_alloc_coherent(kdev, size, &priv->tx_desc_dma, GFP_KERNEL);
 	if (!p) {
 		ret = -ENOMEM;
 		goto out_free_rx_ring;
@@ -980,10 +993,10 @@ static int bcm_enet_open(struct net_device *dev)
 	priv->tx_curr_desc = 0;
 	spin_lock_init(&priv->tx_lock);
 
-	/* init & fill rx ring with skbs */
-	priv->rx_skb = kcalloc(priv->rx_ring_size, sizeof(struct sk_buff *),
+	/* init & fill rx ring with buffers */
+	priv->rx_buf = kcalloc(priv->rx_ring_size, sizeof(void *),
 			       GFP_KERNEL);
-	if (!priv->rx_skb) {
+	if (!priv->rx_buf) {
 		ret = -ENOMEM;
 		goto out_free_tx_skb;
 	}
@@ -1000,8 +1013,8 @@ static int bcm_enet_open(struct net_device *dev)
 		enet_dmac_writel(priv, ENETDMA_BUFALLOC_FORCE_MASK | 0,
 				ENETDMAC_BUFALLOC, priv->rx_chan);
 
-	if (bcm_enet_refill_rx(dev)) {
-		dev_err(kdev, "cannot allocate rx skb queue\n");
+	if (bcm_enet_refill_rx(dev, false)) {
+		dev_err(kdev, "cannot allocate rx buffer queue\n");
 		ret = -ENOMEM;
 		goto out;
 	}
@@ -1063,7 +1076,8 @@ static int bcm_enet_open(struct net_device *dev)
 	val = enet_readl(priv, ENET_CTL_REG);
 	val |= ENET_CTL_ENABLE_MASK;
 	enet_writel(priv, val, ENET_CTL_REG);
-	enet_dma_writel(priv, ENETDMA_CFG_EN_MASK, ENETDMA_CFG_REG);
+	if (priv->dma_has_sram)
+		enet_dma_writel(priv, ENETDMA_CFG_EN_MASK, ENETDMA_CFG_REG);
 	enet_dmac_writel(priv, priv->dma_chan_en_mask,
 			 ENETDMAC_CHANCFG, priv->rx_chan);
 
@@ -1094,18 +1108,7 @@ static int bcm_enet_open(struct net_device *dev)
 	return 0;
 
 out:
-	for (i = 0; i < priv->rx_ring_size; i++) {
-		struct bcm_enet_desc *desc;
-
-		if (!priv->rx_skb[i])
-			continue;
-
-		desc = &priv->rx_desc_cpu[i];
-		dma_unmap_single(kdev, desc->address, priv->rx_skb_size,
-				 DMA_FROM_DEVICE);
-		kfree_skb(priv->rx_skb[i]);
-	}
-	kfree(priv->rx_skb);
+	bcm_enet_free_rx_buf_ring(kdev, priv);
 
 out_free_tx_skb:
 	kfree(priv->tx_skb);
@@ -1184,7 +1187,6 @@ static int bcm_enet_stop(struct net_device *dev)
 {
 	struct bcm_enet_priv *priv;
 	struct device *kdev;
-	int i;
 
 	priv = netdev_priv(dev);
 	kdev = &priv->pdev->dev;
@@ -1193,7 +1195,7 @@ static int bcm_enet_stop(struct net_device *dev)
 	napi_disable(&priv->napi);
 	if (priv->has_phy)
 		phy_stop(dev->phydev);
-	del_timer_sync(&priv->rx_timeout);
+	timer_delete_sync(&priv->rx_timeout);
 
 	/* mask all interrupts */
 	enet_writel(priv, 0, ENET_IRMASK_REG);
@@ -1209,23 +1211,12 @@ static int bcm_enet_stop(struct net_device *dev)
 	bcm_enet_disable_mac(priv);
 
 	/* force reclaim of all tx buffers */
-	bcm_enet_tx_reclaim(dev, 1);
-
-	/* free the rx skb ring */
-	for (i = 0; i < priv->rx_ring_size; i++) {
-		struct bcm_enet_desc *desc;
+	bcm_enet_tx_reclaim(dev, 1, 0);
 
-		if (!priv->rx_skb[i])
-			continue;
-
-		desc = &priv->rx_desc_cpu[i];
-		dma_unmap_single(kdev, desc->address, priv->rx_skb_size,
-				 DMA_FROM_DEVICE);
-		kfree_skb(priv->rx_skb[i]);
-	}
+	/* free the rx buffer ring */
+	bcm_enet_free_rx_buf_ring(kdev, priv);
 
 	/* free remaining allocated memory */
-	kfree(priv->rx_skb);
 	kfree(priv->tx_skb);
 	dma_free_coherent(kdev, priv->rx_desc_alloc_size,
 			  priv->rx_desc_cpu, priv->rx_desc_dma);
@@ -1239,6 +1230,9 @@ static int bcm_enet_stop(struct net_device *dev)
 	if (priv->has_phy)
 		phy_disconnect(dev->phydev);
 
+	/* reset BQL after forced tx reclaim to prevent kernel panic */
+	netdev_reset_queue(dev);
+
 	return 0;
 }
 
@@ -1327,11 +1321,8 @@ static const u32 unused_mib_regs[] = {
 static void bcm_enet_get_drvinfo(struct net_device *netdev,
 				 struct ethtool_drvinfo *drvinfo)
 {
-	strlcpy(drvinfo->driver, bcm_enet_driver_name, sizeof(drvinfo->driver));
-	strlcpy(drvinfo->version, bcm_enet_driver_version,
-		sizeof(drvinfo->version));
-	strlcpy(drvinfo->fw_version, "N/A", sizeof(drvinfo->fw_version));
-	strlcpy(drvinfo->bus_info, "bcm63xx", sizeof(drvinfo->bus_info));
+	strscpy(drvinfo->driver, bcm_enet_driver_name, sizeof(drvinfo->driver));
+	strscpy(drvinfo->bus_info, "bcm63xx", sizeof(drvinfo->bus_info));
 }
 
 static int bcm_enet_get_sset_count(struct net_device *netdev,
@@ -1348,14 +1339,14 @@ static int bcm_enet_get_sset_count(struct net_device *netdev,
 static void bcm_enet_get_strings(struct net_device *netdev,
 				 u32 stringset, u8 *data)
 {
+	const char *str;
 	int i;
 
 	switch (stringset) {
 	case ETH_SS_STATS:
 		for (i = 0; i < BCM_ENET_STATS_LEN; i++) {
-			memcpy(data + i * ETH_GSTRING_LEN,
-			       bcm_enet_gstrings_stats[i].stat_string,
-			       ETH_GSTRING_LEN);
+			str = bcm_enet_gstrings_stats[i].stat_string;
+			ethtool_puts(&data, str);
 		}
 		break;
 	}
@@ -1453,7 +1444,10 @@ static int bcm_enet_get_link_ksettings(struct net_device *dev,
 	if (priv->has_phy) {
 		if (!dev->phydev)
 			return -ENODEV;
-		return phy_ethtool_ksettings_get(dev->phydev, cmd);
+
+		phy_ethtool_ksettings_get(dev->phydev, cmd);
+
+		return 0;
 	} else {
 		cmd->base.autoneg = 0;
 		cmd->base.speed = (priv->force_speed_100) ?
@@ -1503,8 +1497,11 @@ static int bcm_enet_set_link_ksettings(struct net_device *dev,
 	}
 }
 
-static void bcm_enet_get_ringparam(struct net_device *dev,
-				   struct ethtool_ringparam *ering)
+static void
+bcm_enet_get_ringparam(struct net_device *dev,
+		       struct ethtool_ringparam *ering,
+		       struct kernel_ethtool_ringparam *kernel_ering,
+		       struct netlink_ext_ack *extack)
 {
 	struct bcm_enet_priv *priv;
 
@@ -1518,7 +1515,9 @@ static void bcm_enet_get_ringparam(struct net_device *dev,
 }
 
 static int bcm_enet_set_ringparam(struct net_device *dev,
-				  struct ethtool_ringparam *ering)
+				  struct ethtool_ringparam *ering,
+				  struct kernel_ethtool_ringparam *kernel_ering,
+				  struct netlink_ext_ack *extack)
 {
 	struct bcm_enet_priv *priv;
 	int was_running;
@@ -1647,10 +1646,13 @@ static int bcm_enet_change_mtu(struct net_device *dev, int new_mtu)
 	 * align rx buffer size to dma burst len, account FCS since
 	 * it's appended
 	 */
-	priv->rx_skb_size = ALIGN(actual_mtu + ETH_FCS_LEN,
+	priv->rx_buf_size = ALIGN(actual_mtu + ETH_FCS_LEN,
 				  priv->dma_maxburst * 4);
 
-	dev->mtu = new_mtu;
+	priv->rx_frag_size = SKB_DATA_ALIGN(priv->rx_buf_offset + priv->rx_buf_size) +
+					    SKB_DATA_ALIGN(sizeof(struct skb_shared_info));
+
+	WRITE_ONCE(dev->mtu, new_mtu);
 	return 0;
 }
 
@@ -1702,7 +1704,7 @@ static const struct net_device_ops bcm_enet_ops = {
 	.ndo_start_xmit		= bcm_enet_start_xmit,
 	.ndo_set_mac_address	= bcm_enet_set_mac_address,
 	.ndo_set_rx_mode	= bcm_enet_set_multicast_list,
-	.ndo_do_ioctl		= bcm_enet_ioctl,
+	.ndo_eth_ioctl		= bcm_enet_ioctl,
 	.ndo_change_mtu		= bcm_enet_change_mtu,
 };
 
@@ -1714,23 +1716,19 @@ static int bcm_enet_probe(struct platform_device *pdev)
 	struct bcm_enet_priv *priv;
 	struct net_device *dev;
 	struct bcm63xx_enet_platform_data *pd;
-	struct resource *res_mem, *res_irq, *res_irq_rx, *res_irq_tx;
+	int irq, irq_rx, irq_tx;
 	struct mii_bus *bus;
-	const char *clk_name;
 	int i, ret;
 
-	/* stop if shared driver failed, assume driver->probe will be
-	 * called in the same order we register devices (correct ?) */
 	if (!bcm_enet_shared_base[0])
-		return -ENODEV;
+		return -EPROBE_DEFER;
 
-	res_irq = platform_get_resource(pdev, IORESOURCE_IRQ, 0);
-	res_irq_rx = platform_get_resource(pdev, IORESOURCE_IRQ, 1);
-	res_irq_tx = platform_get_resource(pdev, IORESOURCE_IRQ, 2);
-	if (!res_irq || !res_irq_rx || !res_irq_tx)
+	irq = platform_get_irq(pdev, 0);
+	irq_rx = platform_get_irq(pdev, 1);
+	irq_tx = platform_get_irq(pdev, 2);
+	if (irq < 0 || irq_rx < 0 || irq_tx < 0)
 		return -ENODEV;
 
-	ret = 0;
 	dev = alloc_etherdev(sizeof(*priv));
 	if (!dev)
 		return -ENOMEM;
@@ -1738,40 +1736,30 @@ static int bcm_enet_probe(struct platform_device *pdev)
 
 	priv->enet_is_sw = false;
 	priv->dma_maxburst = BCMENET_DMA_MAXBURST;
+	priv->rx_buf_offset = NET_SKB_PAD;
 
 	ret = bcm_enet_change_mtu(dev, dev->mtu);
 	if (ret)
 		goto out;
 
-	res_mem = platform_get_resource(pdev, IORESOURCE_MEM, 0);
-	priv->base = devm_ioremap_resource(&pdev->dev, res_mem);
+	priv->base = devm_platform_ioremap_resource(pdev, 0);
 	if (IS_ERR(priv->base)) {
 		ret = PTR_ERR(priv->base);
 		goto out;
 	}
 
-	dev->irq = priv->irq = res_irq->start;
-	priv->irq_rx = res_irq_rx->start;
-	priv->irq_tx = res_irq_tx->start;
-	priv->mac_id = pdev->id;
-
-	/* get rx & tx dma channel id for this mac */
-	if (priv->mac_id == 0) {
-		priv->rx_chan = 0;
-		priv->tx_chan = 1;
-		clk_name = "enet0";
-	} else {
-		priv->rx_chan = 2;
-		priv->tx_chan = 3;
-		clk_name = "enet1";
-	}
+	dev->irq = priv->irq = irq;
+	priv->irq_rx = irq_rx;
+	priv->irq_tx = irq_tx;
 
-	priv->mac_clk = clk_get(&pdev->dev, clk_name);
+	priv->mac_clk = devm_clk_get(&pdev->dev, "enet");
 	if (IS_ERR(priv->mac_clk)) {
 		ret = PTR_ERR(priv->mac_clk);
 		goto out;
 	}
-	clk_prepare_enable(priv->mac_clk);
+	ret = clk_prepare_enable(priv->mac_clk);
+	if (ret)
+		goto out;
 
 	/* initialize default and fetch platform data */
 	priv->rx_ring_size = BCMENET_DEF_RX_DESC;
@@ -1779,7 +1767,7 @@ static int bcm_enet_probe(struct platform_device *pdev)
 
 	pd = dev_get_platdata(&pdev->dev);
 	if (pd) {
-		memcpy(dev->dev_addr, pd->mac_addr, ETH_ALEN);
+		eth_hw_addr_set(dev, pd->mac_addr);
 		priv->has_phy = pd->has_phy;
 		priv->phy_id = pd->phy_id;
 		priv->has_phy_interrupt = pd->has_phy_interrupt;
@@ -1795,17 +1783,21 @@ static int bcm_enet_probe(struct platform_device *pdev)
 		priv->dma_chan_width = pd->dma_chan_width;
 		priv->dma_has_sram = pd->dma_has_sram;
 		priv->dma_desc_shift = pd->dma_desc_shift;
+		priv->rx_chan = pd->rx_chan;
+		priv->tx_chan = pd->tx_chan;
 	}
 
-	if (priv->mac_id == 0 && priv->has_phy && !priv->use_external_mii) {
+	if (priv->has_phy && !priv->use_external_mii) {
 		/* using internal PHY, enable clock */
-		priv->phy_clk = clk_get(&pdev->dev, "ephy");
+		priv->phy_clk = devm_clk_get(&pdev->dev, "ephy");
 		if (IS_ERR(priv->phy_clk)) {
 			ret = PTR_ERR(priv->phy_clk);
 			priv->phy_clk = NULL;
-			goto out_put_clk_mac;
+			goto out_disable_clk_mac;
 		}
-		clk_prepare_enable(priv->phy_clk);
+		ret = clk_prepare_enable(priv->phy_clk);
+		if (ret)
+			goto out_disable_clk_mac;
 	}
 
 	/* do minimal hardware init to be able to probe mii bus */
@@ -1826,7 +1818,7 @@ static int bcm_enet_probe(struct platform_device *pdev)
 		bus->priv = priv;
 		bus->read = bcm_enet_mdio_read_phylib;
 		bus->write = bcm_enet_mdio_write_phylib;
-		sprintf(bus->id, "%s-%d", pdev->name, priv->mac_id);
+		sprintf(bus->id, "%s-%d", pdev->name, pdev->id);
 
 		/* only probe bus where we think the PHY is, because
 		 * the mdio read operation return 0 instead of 0xffff
@@ -1855,9 +1847,7 @@ static int bcm_enet_probe(struct platform_device *pdev)
 	spin_lock_init(&priv->rx_lock);
 
 	/* init rx timeout (used for oom) */
-	init_timer(&priv->rx_timeout);
-	priv->rx_timeout.function = bcm_enet_refill_rx_timer;
-	priv->rx_timeout.data = (unsigned long)dev;
+	timer_setup(&priv->rx_timeout, bcm_enet_refill_rx_timer, 0);
 
 	/* init the mib update lock&work */
 	mutex_init(&priv->mib_update_lock);
@@ -1869,7 +1859,7 @@ static int bcm_enet_probe(struct platform_device *pdev)
 
 	/* register netdevice */
 	dev->netdev_ops = &bcm_enet_ops;
-	netif_napi_add(dev, &priv->napi, bcm_enet_poll, 16);
+	netif_napi_add_weight(dev, &priv->napi, bcm_enet_poll, 16);
 
 	dev->ethtool_ops = &bcm_enet_ethtool_ops;
 	/* MTU range: 46 - 2028 */
@@ -1899,14 +1889,10 @@ out_free_mdio:
 out_uninit_hw:
 	/* turn off mdc clock */
 	enet_writel(priv, 0, ENET_MIISC_REG);
-	if (priv->phy_clk) {
-		clk_disable_unprepare(priv->phy_clk);
-		clk_put(priv->phy_clk);
-	}
+	clk_disable_unprepare(priv->phy_clk);
 
-out_put_clk_mac:
+out_disable_clk_mac:
 	clk_disable_unprepare(priv->mac_clk);
-	clk_put(priv->mac_clk);
 out:
 	free_netdev(dev);
 	return ret;
@@ -1916,7 +1902,7 @@ out:
 /*
  * exit func, stops hardware and unregisters netdevice
  */
-static int bcm_enet_remove(struct platform_device *pdev)
+static void bcm_enet_remove(struct platform_device *pdev)
 {
 	struct bcm_enet_priv *priv;
 	struct net_device *dev;
@@ -1942,23 +1928,17 @@ static int bcm_enet_remove(struct platform_device *pdev)
 	}
 
 	/* disable hw block clocks */
-	if (priv->phy_clk) {
-		clk_disable_unprepare(priv->phy_clk);
-		clk_put(priv->phy_clk);
-	}
+	clk_disable_unprepare(priv->phy_clk);
 	clk_disable_unprepare(priv->mac_clk);
-	clk_put(priv->mac_clk);
 
 	free_netdev(dev);
-	return 0;
 }
 
-struct platform_driver bcm63xx_enet_driver = {
+static struct platform_driver bcm63xx_enet_driver = {
 	.probe	= bcm_enet_probe,
-	.remove	= bcm_enet_remove,
+	.remove = bcm_enet_remove,
 	.driver	= {
 		.name	= "bcm63xx_enet",
-		.owner  = THIS_MODULE,
 	},
 };
 
@@ -2019,9 +1999,9 @@ static inline int bcm_enet_port_is_rgmii(int portid)
 /*
  * enet sw PHY polling
  */
-static void swphy_poll_timer(unsigned long data)
+static void swphy_poll_timer(struct timer_list *t)
 {
-	struct bcm_enet_priv *priv = (struct bcm_enet_priv *)data;
+	struct bcm_enet_priv *priv = timer_container_of(priv, t, swphy_poll);
 	unsigned int i;
 
 	for (i = 0; i < priv->num_ports; i++) {
@@ -2154,7 +2134,6 @@ static int bcm_enetsw_open(struct net_device *dev)
 		goto out_freeirq_tx;
 	}
 
-	memset(p, 0, size);
 	priv->rx_desc_alloc_size = size;
 	priv->rx_desc_cpu = p;
 
@@ -2167,14 +2146,13 @@ static int bcm_enetsw_open(struct net_device *dev)
 		goto out_free_rx_ring;
 	}
 
-	memset(p, 0, size);
 	priv->tx_desc_alloc_size = size;
 	priv->tx_desc_cpu = p;
 
-	priv->tx_skb = kzalloc(sizeof(struct sk_buff *) * priv->tx_ring_size,
+	priv->tx_skb = kcalloc(priv->tx_ring_size, sizeof(struct sk_buff *),
 			       GFP_KERNEL);
 	if (!priv->tx_skb) {
-		dev_err(kdev, "cannot allocate rx skb queue\n");
+		dev_err(kdev, "cannot allocate tx skb queue\n");
 		ret = -ENOMEM;
 		goto out_free_tx_ring;
 	}
@@ -2184,11 +2162,11 @@ static int bcm_enetsw_open(struct net_device *dev)
 	priv->tx_curr_desc = 0;
 	spin_lock_init(&priv->tx_lock);
 
-	/* init & fill rx ring with skbs */
-	priv->rx_skb = kzalloc(sizeof(struct sk_buff *) * priv->rx_ring_size,
+	/* init & fill rx ring with buffers */
+	priv->rx_buf = kcalloc(priv->rx_ring_size, sizeof(void *),
 			       GFP_KERNEL);
-	if (!priv->rx_skb) {
-		dev_err(kdev, "cannot allocate rx skb queue\n");
+	if (!priv->rx_buf) {
+		dev_err(kdev, "cannot allocate rx buffer queue\n");
 		ret = -ENOMEM;
 		goto out_free_tx_skb;
 	}
@@ -2235,8 +2213,8 @@ static int bcm_enetsw_open(struct net_device *dev)
 	enet_dma_writel(priv, ENETDMA_BUFALLOC_FORCE_MASK | 0,
 			ENETDMA_BUFALLOC_REG(priv->rx_chan));
 
-	if (bcm_enet_refill_rx(dev)) {
-		dev_err(kdev, "cannot allocate rx skb queue\n");
+	if (bcm_enet_refill_rx(dev, false)) {
+		dev_err(kdev, "cannot allocate rx buffer queue\n");
 		ret = -ENOMEM;
 		goto out;
 	}
@@ -2330,26 +2308,12 @@ static int bcm_enetsw_open(struct net_device *dev)
 	}
 
 	/* start phy polling timer */
-	init_timer(&priv->swphy_poll);
-	priv->swphy_poll.function = swphy_poll_timer;
-	priv->swphy_poll.data = (unsigned long)priv;
-	priv->swphy_poll.expires = jiffies;
-	add_timer(&priv->swphy_poll);
+	timer_setup(&priv->swphy_poll, swphy_poll_timer, 0);
+	mod_timer(&priv->swphy_poll, jiffies);
 	return 0;
 
 out:
-	for (i = 0; i < priv->rx_ring_size; i++) {
-		struct bcm_enet_desc *desc;
-
-		if (!priv->rx_skb[i])
-			continue;
-
-		desc = &priv->rx_desc_cpu[i];
-		dma_unmap_single(kdev, desc->address, priv->rx_skb_size,
-				 DMA_FROM_DEVICE);
-		kfree_skb(priv->rx_skb[i]);
-	}
-	kfree(priv->rx_skb);
+	bcm_enet_free_rx_buf_ring(kdev, priv);
 
 out_free_tx_skb:
 	kfree(priv->tx_skb);
@@ -2378,15 +2342,14 @@ static int bcm_enetsw_stop(struct net_device *dev)
 {
 	struct bcm_enet_priv *priv;
 	struct device *kdev;
-	int i;
 
 	priv = netdev_priv(dev);
 	kdev = &priv->pdev->dev;
 
-	del_timer_sync(&priv->swphy_poll);
+	timer_delete_sync(&priv->swphy_poll);
 	netif_stop_queue(dev);
 	napi_disable(&priv->napi);
-	del_timer_sync(&priv->rx_timeout);
+	timer_delete_sync(&priv->rx_timeout);
 
 	/* mask all interrupts */
 	enet_dmac_writel(priv, 0, ENETDMAC_IRMASK, priv->rx_chan);
@@ -2397,23 +2360,12 @@ static int bcm_enetsw_stop(struct net_device *dev)
 	bcm_enet_disable_dma(priv, priv->rx_chan);
 
 	/* force reclaim of all tx buffers */
-	bcm_enet_tx_reclaim(dev, 1);
+	bcm_enet_tx_reclaim(dev, 1, 0);
 
-	/* free the rx skb ring */
-	for (i = 0; i < priv->rx_ring_size; i++) {
-		struct bcm_enet_desc *desc;
-
-		if (!priv->rx_skb[i])
-			continue;
-
-		desc = &priv->rx_desc_cpu[i];
-		dma_unmap_single(kdev, desc->address, priv->rx_skb_size,
-				 DMA_FROM_DEVICE);
-		kfree_skb(priv->rx_skb[i]);
-	}
+	/* free the rx buffer ring */
+	bcm_enet_free_rx_buf_ring(kdev, priv);
 
 	/* free remaining allocated memory */
-	kfree(priv->rx_skb);
 	kfree(priv->tx_skb);
 	dma_free_coherent(kdev, priv->rx_desc_alloc_size,
 			  priv->rx_desc_cpu, priv->rx_desc_dma);
@@ -2423,6 +2375,9 @@ static int bcm_enetsw_stop(struct net_device *dev)
 		free_irq(priv->irq_tx, dev);
 	free_irq(priv->irq_rx, dev);
 
+	/* reset BQL after forced tx reclaim to prevent kernel panic */
+	netdev_reset_queue(dev);
+
 	return 0;
 }
 
@@ -2494,7 +2449,7 @@ static const struct net_device_ops bcm_enetsw_ops = {
 	.ndo_stop		= bcm_enetsw_stop,
 	.ndo_start_xmit		= bcm_enet_start_xmit,
 	.ndo_change_mtu		= bcm_enet_change_mtu,
-	.ndo_do_ioctl		= bcm_enetsw_ioctl,
+	.ndo_eth_ioctl		= bcm_enetsw_ioctl,
 };
 
 
@@ -2548,14 +2503,14 @@ static const struct bcm_enet_stats bcm_enetsw_gstrings_stats[] = {
 static void bcm_enetsw_get_strings(struct net_device *netdev,
 				   u32 stringset, u8 *data)
 {
+	const char *str;
 	int i;
 
 	switch (stringset) {
 	case ETH_SS_STATS:
 		for (i = 0; i < BCM_ENETSW_STATS_LEN; i++) {
-			memcpy(data + i * ETH_GSTRING_LEN,
-			       bcm_enetsw_gstrings_stats[i].stat_string,
-			       ETH_GSTRING_LEN);
+			str = bcm_enetsw_gstrings_stats[i].stat_string;
+			ethtool_puts(&data, str);
 		}
 		break;
 	}
@@ -2575,10 +2530,8 @@ static int bcm_enetsw_get_sset_count(struct net_device *netdev,
 static void bcm_enetsw_get_drvinfo(struct net_device *netdev,
 				   struct ethtool_drvinfo *drvinfo)
 {
-	strncpy(drvinfo->driver, bcm_enet_driver_name, 32);
-	strncpy(drvinfo->version, bcm_enet_driver_version, 32);
-	strncpy(drvinfo->fw_version, "N/A", 32);
-	strncpy(drvinfo->bus_info, "bcm63xx", 32);
+	strscpy(drvinfo->driver, bcm_enet_driver_name, sizeof(drvinfo->driver));
+	strscpy(drvinfo->bus_info, "bcm63xx", sizeof(drvinfo->bus_info));
 }
 
 static void bcm_enetsw_get_ethtool_stats(struct net_device *netdev,
@@ -2629,8 +2582,11 @@ static void bcm_enetsw_get_ethtool_stats(struct net_device *netdev,
 	}
 }
 
-static void bcm_enetsw_get_ringparam(struct net_device *dev,
-				     struct ethtool_ringparam *ering)
+static void
+bcm_enetsw_get_ringparam(struct net_device *dev,
+			 struct ethtool_ringparam *ering,
+			 struct kernel_ethtool_ringparam *kernel_ering,
+			 struct netlink_ext_ack *extack)
 {
 	struct bcm_enet_priv *priv;
 
@@ -2645,8 +2601,11 @@ static void bcm_enetsw_get_ringparam(struct net_device *dev,
 	ering->tx_pending = priv->tx_ring_size;
 }
 
-static int bcm_enetsw_set_ringparam(struct net_device *dev,
-				    struct ethtool_ringparam *ering)
+static int
+bcm_enetsw_set_ringparam(struct net_device *dev,
+			 struct ethtool_ringparam *ering,
+			 struct kernel_ethtool_ringparam *kernel_ering,
+			 struct netlink_ext_ack *extack)
 {
 	struct bcm_enet_priv *priv;
 	int was_running;
@@ -2672,7 +2631,7 @@ static int bcm_enetsw_set_ringparam(struct net_device *dev,
 	return 0;
 }
 
-static struct ethtool_ops bcm_enetsw_ethtool_ops = {
+static const struct ethtool_ops bcm_enetsw_ethtool_ops = {
 	.get_strings		= bcm_enetsw_get_strings,
 	.get_sset_count		= bcm_enetsw_get_sset_count,
 	.get_ethtool_stats      = bcm_enetsw_get_ethtool_stats,
@@ -2690,11 +2649,8 @@ static int bcm_enetsw_probe(struct platform_device *pdev)
 	struct resource *res_mem;
 	int ret, irq_rx, irq_tx;
 
-	/* stop if shared driver failed, assume driver->probe will be
-	 * called in the same order we register devices (correct ?)
-	 */
 	if (!bcm_enet_shared_base[0])
-		return -ENODEV;
+		return -EPROBE_DEFER;
 
 	res_mem = platform_get_resource(pdev, IORESOURCE_MEM, 0);
 	irq_rx = platform_get_irq(pdev, 0);
@@ -2702,12 +2658,10 @@ static int bcm_enetsw_probe(struct platform_device *pdev)
 	if (!res_mem || irq_rx < 0)
 		return -ENODEV;
 
-	ret = 0;
 	dev = alloc_etherdev(sizeof(*priv));
 	if (!dev)
 		return -ENOMEM;
 	priv = netdev_priv(dev);
-	memset(priv, 0, sizeof(*priv));
 
 	/* initialize default and fetch platform data */
 	priv->enet_is_sw = true;
@@ -2716,10 +2670,11 @@ static int bcm_enetsw_probe(struct platform_device *pdev)
 	priv->rx_ring_size = BCMENET_DEF_RX_DESC;
 	priv->tx_ring_size = BCMENET_DEF_TX_DESC;
 	priv->dma_maxburst = BCMENETSW_DMA_MAXBURST;
+	priv->rx_buf_offset = NET_SKB_PAD + NET_IP_ALIGN;
 
 	pd = dev_get_platdata(&pdev->dev);
 	if (pd) {
-		memcpy(dev->dev_addr, pd->mac_addr, ETH_ALEN);
+		eth_hw_addr_set(dev, pd->mac_addr);
 		memcpy(priv->used_ports, pd->used_ports,
 		       sizeof(pd->used_ports));
 		priv->num_ports = pd->num_ports;
@@ -2733,37 +2688,31 @@ static int bcm_enetsw_probe(struct platform_device *pdev)
 	if (ret)
 		goto out;
 
-	if (!request_mem_region(res_mem->start, resource_size(res_mem),
-				"bcm63xx_enetsw")) {
-		ret = -EBUSY;
+	priv->base = devm_ioremap_resource(&pdev->dev, res_mem);
+	if (IS_ERR(priv->base)) {
+		ret = PTR_ERR(priv->base);
 		goto out;
 	}
 
-	priv->base = ioremap(res_mem->start, resource_size(res_mem));
-	if (priv->base == NULL) {
-		ret = -ENOMEM;
-		goto out_release_mem;
-	}
-
-	priv->mac_clk = clk_get(&pdev->dev, "enetsw");
+	priv->mac_clk = devm_clk_get(&pdev->dev, "enetsw");
 	if (IS_ERR(priv->mac_clk)) {
 		ret = PTR_ERR(priv->mac_clk);
-		goto out_unmap;
+		goto out;
 	}
-	clk_enable(priv->mac_clk);
+	ret = clk_prepare_enable(priv->mac_clk);
+	if (ret)
+		goto out;
 
 	priv->rx_chan = 0;
 	priv->tx_chan = 1;
 	spin_lock_init(&priv->rx_lock);
 
 	/* init rx timeout (used for oom) */
-	init_timer(&priv->rx_timeout);
-	priv->rx_timeout.function = bcm_enet_refill_rx_timer;
-	priv->rx_timeout.data = (unsigned long)dev;
+	timer_setup(&priv->rx_timeout, bcm_enet_refill_rx_timer, 0);
 
 	/* register netdevice */
 	dev->netdev_ops = &bcm_enetsw_ops;
-	netif_napi_add(dev, &priv->napi, bcm_enet_poll, 16);
+	netif_napi_add_weight(dev, &priv->napi, bcm_enet_poll, 16);
 	dev->ethtool_ops = &bcm_enetsw_ethtool_ops;
 	SET_NETDEV_DEV(dev, &pdev->dev);
 
@@ -2771,7 +2720,7 @@ static int bcm_enetsw_probe(struct platform_device *pdev)
 
 	ret = register_netdev(dev);
 	if (ret)
-		goto out_put_clk;
+		goto out_disable_clk;
 
 	netif_carrier_off(dev);
 	platform_set_drvdata(pdev, dev);
@@ -2780,14 +2729,8 @@ static int bcm_enetsw_probe(struct platform_device *pdev)
 
 	return 0;
 
-out_put_clk:
-	clk_put(priv->mac_clk);
-
-out_unmap:
-	iounmap(priv->base);
-
-out_release_mem:
-	release_mem_region(res_mem->start, resource_size(res_mem));
+out_disable_clk:
+	clk_disable_unprepare(priv->mac_clk);
 out:
 	free_netdev(dev);
 	return ret;
@@ -2795,47 +2738,39 @@ out:
 
 
 /* exit func, stops hardware and unregisters netdevice */
-static int bcm_enetsw_remove(struct platform_device *pdev)
+static void bcm_enetsw_remove(struct platform_device *pdev)
 {
 	struct bcm_enet_priv *priv;
 	struct net_device *dev;
-	struct resource *res;
 
 	/* stop netdevice */
 	dev = platform_get_drvdata(pdev);
 	priv = netdev_priv(dev);
 	unregister_netdev(dev);
 
-	/* release device resources */
-	iounmap(priv->base);
-	res = platform_get_resource(pdev, IORESOURCE_MEM, 0);
-	release_mem_region(res->start, resource_size(res));
+	clk_disable_unprepare(priv->mac_clk);
 
 	free_netdev(dev);
-	return 0;
 }
 
-struct platform_driver bcm63xx_enetsw_driver = {
+static struct platform_driver bcm63xx_enetsw_driver = {
 	.probe	= bcm_enetsw_probe,
-	.remove	= bcm_enetsw_remove,
+	.remove = bcm_enetsw_remove,
 	.driver	= {
 		.name	= "bcm63xx_enetsw",
-		.owner  = THIS_MODULE,
 	},
 };
 
 /* reserve & remap memory space shared between all macs */
 static int bcm_enet_shared_probe(struct platform_device *pdev)
 {
-	struct resource *res;
 	void __iomem *p[3];
 	unsigned int i;
 
 	memset(bcm_enet_shared_base, 0, sizeof(bcm_enet_shared_base));
 
 	for (i = 0; i < 3; i++) {
-		res = platform_get_resource(pdev, IORESOURCE_MEM, i);
-		p[i] = devm_ioremap_resource(&pdev->dev, res);
+		p[i] = devm_platform_ioremap_resource(pdev, i);
 		if (IS_ERR(p[i]))
 			return PTR_ERR(p[i]);
 	}
@@ -2845,20 +2780,13 @@ static int bcm_enet_shared_probe(struct platform_device *pdev)
 	return 0;
 }
 
-static int bcm_enet_shared_remove(struct platform_device *pdev)
-{
-	return 0;
-}
-
 /* this "shared" driver is needed because both macs share a single
  * address space
  */
 struct platform_driver bcm63xx_enet_shared_driver = {
 	.probe	= bcm_enet_shared_probe,
-	.remove	= bcm_enet_shared_remove,
 	.driver	= {
 		.name	= "bcm63xx_enet_shared",
-		.owner  = THIS_MODULE,
 	},
 };
 
diff --git a/drivers/net/ethernet/broadcom/bcm63xx_enet.h b/drivers/net/ethernet/broadcom/bcm63xx_enet.h
index 0a1b7b2e55bd..78f1830fb3cb 100644
--- a/drivers/net/ethernet/broadcom/bcm63xx_enet.h
+++ b/drivers/net/ethernet/broadcom/bcm63xx_enet.h
@@ -1,3 +1,4 @@
+/* SPDX-License-Identifier: GPL-2.0 */
 #ifndef BCM63XX_ENET_H_
 #define BCM63XX_ENET_H_
 
@@ -8,7 +9,6 @@
 #include <linux/platform_device.h>
 
 #include <bcm63xx_regs.h>
-#include <bcm63xx_irq.h>
 #include <bcm63xx_io.h>
 #include <bcm63xx_iudma.h>
 
@@ -193,9 +193,6 @@ struct bcm_enet_mib_counters {
 
 struct bcm_enet_priv {
 
-	/* mac id (from platform device id) */
-	int mac_id;
-
 	/* base remapped address of device */
 	void __iomem *base;
 
@@ -233,11 +230,17 @@ struct bcm_enet_priv {
 	/* next dirty rx descriptor to refill */
 	int rx_dirty_desc;
 
-	/* size of allocated rx skbs */
-	unsigned int rx_skb_size;
+	/* size of allocated rx buffers */
+	unsigned int rx_buf_size;
+
+	/* allocated rx buffer offset */
+	unsigned int rx_buf_offset;
+
+	/* size of allocated rx frag */
+	unsigned int rx_frag_size;
 
-	/* list of skb given to hw for rx */
-	struct sk_buff **rx_skb;
+	/* list of buffer given to hw for rx */
+	void **rx_buf;
 
 	/* used when rx skb allocation failed, so we defer rx queue
 	 * refill */
diff --git a/drivers/net/ethernet/broadcom/bcmsysport.c b/drivers/net/ethernet/broadcom/bcmsysport.c
index 099b374c1b17..bc4e1f3b3752 100644
--- a/drivers/net/ethernet/broadcom/bcmsysport.c
+++ b/drivers/net/ethernet/broadcom/bcmsysport.c
@@ -1,11 +1,8 @@
+// SPDX-License-Identifier: GPL-2.0-only
 /*
  * Broadcom BCM7xxx System Port Ethernet MAC driver
  *
  * Copyright (C) 2014 Broadcom Corporation
- *
- * This program is free software; you can redistribute it and/or modify
- * it under the terms of the GNU General Public License version 2 as
- * published by the Free Software Foundation.
  */
 
 #define pr_fmt(fmt)	KBUILD_MODNAME ": " fmt
@@ -15,6 +12,7 @@
 #include <linux/module.h>
 #include <linux/kernel.h>
 #include <linux/netdevice.h>
+#include <linux/dsa/brcm.h>
 #include <linux/etherdevice.h>
 #include <linux/platform_device.h>
 #include <linux/of.h>
@@ -23,35 +21,12 @@
 #include <linux/phy.h>
 #include <linux/phy_fixed.h>
 #include <net/dsa.h>
+#include <linux/clk.h>
 #include <net/ip.h>
 #include <net/ipv6.h>
 
 #include "bcmsysport.h"
 
-/* I/O accessors register helpers */
-#define BCM_SYSPORT_IO_MACRO(name, offset) \
-static inline u32 name##_readl(struct bcm_sysport_priv *priv, u32 off)	\
-{									\
-	u32 reg = __raw_readl(priv->base + offset + off);		\
-	return reg;							\
-}									\
-static inline void name##_writel(struct bcm_sysport_priv *priv,		\
-				  u32 val, u32 off)			\
-{									\
-	__raw_writel(val, priv->base + offset + off);			\
-}									\
-
-BCM_SYSPORT_IO_MACRO(intrl2_0, SYS_PORT_INTRL2_0_OFFSET);
-BCM_SYSPORT_IO_MACRO(intrl2_1, SYS_PORT_INTRL2_1_OFFSET);
-BCM_SYSPORT_IO_MACRO(umac, SYS_PORT_UMAC_OFFSET);
-BCM_SYSPORT_IO_MACRO(gib, SYS_PORT_GIB_OFFSET);
-BCM_SYSPORT_IO_MACRO(tdma, SYS_PORT_TDMA_OFFSET);
-BCM_SYSPORT_IO_MACRO(rxchk, SYS_PORT_RXCHK_OFFSET);
-BCM_SYSPORT_IO_MACRO(txchk, SYS_PORT_TXCHK_OFFSET);
-BCM_SYSPORT_IO_MACRO(rbuf, SYS_PORT_RBUF_OFFSET);
-BCM_SYSPORT_IO_MACRO(tbuf, SYS_PORT_TBUF_OFFSET);
-BCM_SYSPORT_IO_MACRO(topctrl, SYS_PORT_TOPCTRL_OFFSET);
-
 /* On SYSTEMPORT Lite, any register after RDMA_STATUS has the exact
  * same layout, except it has been moved by 4 bytes up, *sigh*
  */
@@ -59,14 +34,14 @@ static inline u32 rdma_readl(struct bcm_sysport_priv *priv, u32 off)
 {
 	if (priv->is_lite && off >= RDMA_STATUS)
 		off += 4;
-	return __raw_readl(priv->base + SYS_PORT_RDMA_OFFSET + off);
+	return readl_relaxed(priv->base + SYS_PORT_RDMA_OFFSET + off);
 }
 
 static inline void rdma_writel(struct bcm_sysport_priv *priv, u32 val, u32 off)
 {
 	if (priv->is_lite && off >= RDMA_STATUS)
 		off += 4;
-	__raw_writel(val, priv->base + SYS_PORT_RDMA_OFFSET + off);
+	writel_relaxed(val, priv->base + SYS_PORT_RDMA_OFFSET + off);
 }
 
 static inline u32 tdma_control_bit(struct bcm_sysport_priv *priv, u32 bit)
@@ -110,30 +85,25 @@ static inline void dma_desc_set_addr(struct bcm_sysport_priv *priv,
 				     dma_addr_t addr)
 {
 #ifdef CONFIG_PHYS_ADDR_T_64BIT
-	__raw_writel(upper_32_bits(addr) & DESC_ADDR_HI_MASK,
+	writel_relaxed(upper_32_bits(addr) & DESC_ADDR_HI_MASK,
 		     d + DESC_ADDR_HI_STATUS_LEN);
 #endif
-	__raw_writel(lower_32_bits(addr), d + DESC_ADDR_LO);
-}
-
-static inline void tdma_port_write_desc_addr(struct bcm_sysport_priv *priv,
-					     struct dma_desc *desc,
-					     unsigned int port)
-{
-	/* Ports are latched, so write upper address first */
-	tdma_writel(priv, desc->addr_status_len, TDMA_WRITE_PORT_HI(port));
-	tdma_writel(priv, desc->addr_lo, TDMA_WRITE_PORT_LO(port));
+	writel_relaxed(lower_32_bits(addr), d + DESC_ADDR_LO);
 }
 
 /* Ethtool operations */
-static int bcm_sysport_set_rx_csum(struct net_device *dev,
-				   netdev_features_t wanted)
+static void bcm_sysport_set_rx_csum(struct net_device *dev,
+				    netdev_features_t wanted)
 {
 	struct bcm_sysport_priv *priv = netdev_priv(dev);
 	u32 reg;
 
 	priv->rx_chk_en = !!(wanted & NETIF_F_RXCSUM);
 	reg = rxchk_readl(priv, RXCHK_CONTROL);
+	/* Clear L2 header checks, which would prevent BPDUs
+	 * from being received.
+	 */
+	reg &= ~RXCHK_L2_HDR_DIS;
 	if (priv->rx_chk_en)
 		reg |= RXCHK_EN;
 	else
@@ -157,12 +127,10 @@ static int bcm_sysport_set_rx_csum(struct net_device *dev,
 		reg &= ~RXCHK_BRCM_TAG_EN;
 
 	rxchk_writel(priv, reg, RXCHK_CONTROL);
-
-	return 0;
 }
 
-static int bcm_sysport_set_tx_csum(struct net_device *dev,
-				   netdev_features_t wanted)
+static void bcm_sysport_set_tx_csum(struct net_device *dev,
+				    netdev_features_t wanted)
 {
 	struct bcm_sysport_priv *priv = netdev_priv(dev);
 	u32 reg;
@@ -170,30 +138,51 @@ static int bcm_sysport_set_tx_csum(struct net_device *dev,
 	/* Hardware transmit checksum requires us to enable the Transmit status
 	 * block prepended to the packet contents
 	 */
-	priv->tsb_en = !!(wanted & (NETIF_F_IP_CSUM | NETIF_F_IPV6_CSUM));
+	priv->tsb_en = !!(wanted & (NETIF_F_IP_CSUM | NETIF_F_IPV6_CSUM |
+				    NETIF_F_HW_VLAN_CTAG_TX));
 	reg = tdma_readl(priv, TDMA_CONTROL);
 	if (priv->tsb_en)
 		reg |= tdma_control_bit(priv, TSB_EN);
 	else
 		reg &= ~tdma_control_bit(priv, TSB_EN);
+	/* Indicating that software inserts Broadcom tags is needed for the TX
+	 * checksum to be computed correctly when using VLAN HW acceleration,
+	 * else it has no effect, so it can always be turned on.
+	 */
+	if (netdev_uses_dsa(dev))
+		reg |= tdma_control_bit(priv, SW_BRCM_TAG);
+	else
+		reg &= ~tdma_control_bit(priv, SW_BRCM_TAG);
 	tdma_writel(priv, reg, TDMA_CONTROL);
 
-	return 0;
+	/* Default TPID is ETH_P_8021AD, change to ETH_P_8021Q */
+	if (wanted & NETIF_F_HW_VLAN_CTAG_TX)
+		tdma_writel(priv, ETH_P_8021Q, TDMA_TPID);
 }
 
 static int bcm_sysport_set_features(struct net_device *dev,
 				    netdev_features_t features)
 {
-	netdev_features_t changed = features ^ dev->features;
-	netdev_features_t wanted = dev->wanted_features;
-	int ret = 0;
+	struct bcm_sysport_priv *priv = netdev_priv(dev);
+	int ret;
 
-	if (changed & NETIF_F_RXCSUM)
-		ret = bcm_sysport_set_rx_csum(dev, wanted);
-	if (changed & (NETIF_F_IP_CSUM | NETIF_F_IPV6_CSUM))
-		ret = bcm_sysport_set_tx_csum(dev, wanted);
+	ret = clk_prepare_enable(priv->clk);
+	if (ret)
+		return ret;
 
-	return ret;
+	/* Read CRC forward */
+	if (!priv->is_lite)
+		priv->crc_fwd = !!(umac_readl(priv, UMAC_CMD) & CMD_CRC_FWD);
+	else
+		priv->crc_fwd = !((gib_readl(priv, GIB_CONTROL) &
+				  GIB_FCS_STRIP) >> GIB_FCS_STRIP_SHIFT);
+
+	bcm_sysport_set_rx_csum(dev, features);
+	bcm_sysport_set_tx_csum(dev, features);
+
+	clk_disable_unprepare(priv->clk);
+
+	return 0;
 }
 
 /* Hardware counters must be kept in sync because the order/offset
@@ -201,10 +190,10 @@ static int bcm_sysport_set_features(struct net_device *dev,
  */
 static const struct bcm_sysport_stats bcm_sysport_gstrings_stats[] = {
 	/* general stats */
-	STAT_NETDEV(rx_packets),
-	STAT_NETDEV(tx_packets),
-	STAT_NETDEV(rx_bytes),
-	STAT_NETDEV(tx_bytes),
+	STAT_NETDEV64(rx_packets),
+	STAT_NETDEV64(tx_packets),
+	STAT_NETDEV64(rx_bytes),
+	STAT_NETDEV64(tx_bytes),
 	STAT_NETDEV(rx_errors),
 	STAT_NETDEV(tx_errors),
 	STAT_NETDEV(rx_dropped),
@@ -282,9 +271,13 @@ static const struct bcm_sysport_stats bcm_sysport_gstrings_stats[] = {
 	/* RBUF misc statistics */
 	STAT_RBUF("rbuf_ovflow_cnt", mib.rbuf_ovflow_cnt, RBUF_OVFL_DISC_CNTR),
 	STAT_RBUF("rbuf_err_cnt", mib.rbuf_err_cnt, RBUF_ERR_PKT_CNTR),
+	/* RDMA misc statistics */
+	STAT_RDMA("rdma_ovflow_cnt", mib.rdma_ovflow_cnt, RDMA_OVFL_DISC_CNTR),
 	STAT_MIB_SOFT("alloc_rx_buff_failed", mib.alloc_rx_buff_failed),
 	STAT_MIB_SOFT("rx_dma_failed", mib.rx_dma_failed),
 	STAT_MIB_SOFT("tx_dma_failed", mib.tx_dma_failed),
+	STAT_MIB_SOFT("tx_realloc_tsb", mib.tx_realloc_tsb),
+	STAT_MIB_SOFT("tx_realloc_tsb_failed", mib.tx_realloc_tsb_failed),
 	/* Per TX-queue statistics are dynamically appended */
 };
 
@@ -293,9 +286,8 @@ static const struct bcm_sysport_stats bcm_sysport_gstrings_stats[] = {
 static void bcm_sysport_get_drvinfo(struct net_device *dev,
 				    struct ethtool_drvinfo *info)
 {
-	strlcpy(info->driver, KBUILD_MODNAME, sizeof(info->driver));
-	strlcpy(info->version, "0.1", sizeof(info->version));
-	strlcpy(info->bus_info, "platform", sizeof(info->bus_info));
+	strscpy(info->driver, KBUILD_MODNAME, sizeof(info->driver));
+	strscpy(info->bus_info, "platform", sizeof(info->bus_info));
 }
 
 static u32 bcm_sysport_get_msglvl(struct net_device *dev)
@@ -316,8 +308,10 @@ static inline bool bcm_sysport_lite_stat_valid(enum bcm_sysport_stat_type type)
 {
 	switch (type) {
 	case BCM_SYSPORT_STAT_NETDEV:
+	case BCM_SYSPORT_STAT_NETDEV64:
 	case BCM_SYSPORT_STAT_RXCHK:
 	case BCM_SYSPORT_STAT_RBUF:
+	case BCM_SYSPORT_STAT_RDMA:
 	case BCM_SYSPORT_STAT_SOFT:
 		return true;
 	default:
@@ -352,32 +346,22 @@ static void bcm_sysport_get_strings(struct net_device *dev,
 {
 	struct bcm_sysport_priv *priv = netdev_priv(dev);
 	const struct bcm_sysport_stats *s;
-	char buf[128];
-	int i, j;
+	int i;
 
 	switch (stringset) {
 	case ETH_SS_STATS:
-		for (i = 0, j = 0; i < BCM_SYSPORT_STATS_LEN; i++) {
+		for (i = 0; i < BCM_SYSPORT_STATS_LEN; i++) {
 			s = &bcm_sysport_gstrings_stats[i];
 			if (priv->is_lite &&
 			    !bcm_sysport_lite_stat_valid(s->type))
 				continue;
 
-			memcpy(data + j * ETH_GSTRING_LEN, s->stat_string,
-			       ETH_GSTRING_LEN);
-			j++;
+			ethtool_puts(&data, s->stat_string);
 		}
 
 		for (i = 0; i < dev->num_tx_queues; i++) {
-			snprintf(buf, sizeof(buf), "txq%d_packets", i);
-			memcpy(data + j * ETH_GSTRING_LEN, buf,
-			       ETH_GSTRING_LEN);
-			j++;
-
-			snprintf(buf, sizeof(buf), "txq%d_bytes", i);
-			memcpy(data + j * ETH_GSTRING_LEN, buf,
-			       ETH_GSTRING_LEN);
-			j++;
+			ethtool_sprintf(&data, "txq%d_packets", i);
+			ethtool_sprintf(&data, "txq%d_bytes", i);
 		}
 		break;
 	default:
@@ -398,6 +382,7 @@ static void bcm_sysport_update_mib_counters(struct bcm_sysport_priv *priv)
 		s = &bcm_sysport_gstrings_stats[i];
 		switch (s->type) {
 		case BCM_SYSPORT_STAT_NETDEV:
+		case BCM_SYSPORT_STAT_NETDEV64:
 		case BCM_SYSPORT_STAT_SOFT:
 			continue;
 		case BCM_SYSPORT_STAT_MIB_RX:
@@ -420,6 +405,14 @@ static void bcm_sysport_update_mib_counters(struct bcm_sysport_priv *priv)
 			if (val == ~0)
 				rbuf_writel(priv, 0, s->reg_offset);
 			break;
+		case BCM_SYSPORT_STAT_RDMA:
+			if (!priv->is_lite)
+				continue;
+
+			val = rdma_readl(priv, s->reg_offset);
+			if (val == ~0)
+				rdma_writel(priv, 0, s->reg_offset);
+			break;
 		}
 
 		j += s->stat_sizeof;
@@ -430,15 +423,44 @@ static void bcm_sysport_update_mib_counters(struct bcm_sysport_priv *priv)
 	netif_dbg(priv, hw, priv->netdev, "updated MIB counters\n");
 }
 
+static void bcm_sysport_update_tx_stats(struct bcm_sysport_priv *priv,
+					u64 *tx_bytes, u64 *tx_packets)
+{
+	struct bcm_sysport_tx_ring *ring;
+	u64 bytes = 0, packets = 0;
+	unsigned int start;
+	unsigned int q;
+
+	for (q = 0; q < priv->netdev->num_tx_queues; q++) {
+		ring = &priv->tx_rings[q];
+		do {
+			start = u64_stats_fetch_begin(&priv->syncp);
+			bytes = ring->bytes;
+			packets = ring->packets;
+		} while (u64_stats_fetch_retry(&priv->syncp, start));
+
+		*tx_bytes += bytes;
+		*tx_packets += packets;
+	}
+}
+
 static void bcm_sysport_get_stats(struct net_device *dev,
 				  struct ethtool_stats *stats, u64 *data)
 {
 	struct bcm_sysport_priv *priv = netdev_priv(dev);
+	struct bcm_sysport_stats64 *stats64 = &priv->stats64;
+	struct u64_stats_sync *syncp = &priv->syncp;
 	struct bcm_sysport_tx_ring *ring;
+	u64 tx_bytes = 0, tx_packets = 0;
+	unsigned int start;
 	int i, j;
 
-	if (netif_running(dev))
+	if (netif_running(dev)) {
 		bcm_sysport_update_mib_counters(priv);
+		bcm_sysport_update_tx_stats(priv, &tx_bytes, &tx_packets);
+		stats64->tx_bytes = tx_bytes;
+		stats64->tx_packets = tx_packets;
+	}
 
 	for (i =  0, j = 0; i < BCM_SYSPORT_STATS_LEN; i++) {
 		const struct bcm_sysport_stats *s;
@@ -447,10 +469,23 @@ static void bcm_sysport_get_stats(struct net_device *dev,
 		s = &bcm_sysport_gstrings_stats[i];
 		if (s->type == BCM_SYSPORT_STAT_NETDEV)
 			p = (char *)&dev->stats;
+		else if (s->type == BCM_SYSPORT_STAT_NETDEV64)
+			p = (char *)stats64;
 		else
 			p = (char *)priv;
+
+		if (priv->is_lite && !bcm_sysport_lite_stat_valid(s->type))
+			continue;
 		p += s->stat_offset;
-		data[j] = *(unsigned long *)p;
+
+		if (s->stat_sizeof == sizeof(u64) &&
+		    s->type == BCM_SYSPORT_STAT_NETDEV64) {
+			do {
+				start = u64_stats_fetch_begin(syncp);
+				data[i] = *(u64 *)p;
+			} while (u64_stats_fetch_retry(syncp, start));
+		} else
+			data[i] = *(u32 *)p;
 		j++;
 	}
 
@@ -475,19 +510,14 @@ static void bcm_sysport_get_wol(struct net_device *dev,
 				struct ethtool_wolinfo *wol)
 {
 	struct bcm_sysport_priv *priv = netdev_priv(dev);
-	u32 reg;
 
-	wol->supported = WAKE_MAGIC | WAKE_MAGICSECURE;
+	wol->supported = WAKE_MAGIC | WAKE_MAGICSECURE | WAKE_FILTER;
 	wol->wolopts = priv->wolopts;
 
 	if (!(priv->wolopts & WAKE_MAGICSECURE))
 		return;
 
-	/* Return the programmed SecureOn password */
-	reg = umac_readl(priv, UMAC_PSW_MS);
-	put_unaligned_be16(reg, &wol->sopass[0]);
-	reg = umac_readl(priv, UMAC_PSW_LS);
-	put_unaligned_be32(reg, &wol->sopass[2]);
+	memcpy(wol->sopass, priv->sopass, sizeof(priv->sopass));
 }
 
 static int bcm_sysport_set_wol(struct net_device *dev,
@@ -495,7 +525,7 @@ static int bcm_sysport_set_wol(struct net_device *dev,
 {
 	struct bcm_sysport_priv *priv = netdev_priv(dev);
 	struct device *kdev = &priv->pdev->dev;
-	u32 supported = WAKE_MAGIC | WAKE_MAGICSECURE;
+	u32 supported = WAKE_MAGIC | WAKE_MAGICSECURE | WAKE_FILTER;
 
 	if (!device_can_wakeup(kdev))
 		return -ENOTSUPP;
@@ -503,13 +533,8 @@ static int bcm_sysport_set_wol(struct net_device *dev,
 	if (wol->wolopts & ~supported)
 		return -EINVAL;
 
-	/* Program the SecureOn password */
-	if (wol->wolopts & WAKE_MAGICSECURE) {
-		umac_writel(priv, get_unaligned_be16(&wol->sopass[0]),
-			    UMAC_PSW_MS);
-		umac_writel(priv, get_unaligned_be32(&wol->sopass[2]),
-			    UMAC_PSW_LS);
-	}
+	if (wol->wolopts & WAKE_MAGICSECURE)
+		memcpy(priv->sopass, wol->sopass, sizeof(priv->sopass));
 
 	/* Flag the device and relevant IRQ as wakeup capable */
 	if (wol->wolopts) {
@@ -530,8 +555,38 @@ static int bcm_sysport_set_wol(struct net_device *dev,
 	return 0;
 }
 
+static void bcm_sysport_set_rx_coalesce(struct bcm_sysport_priv *priv,
+					u32 usecs, u32 pkts)
+{
+	u32 reg;
+
+	reg = rdma_readl(priv, RDMA_MBDONE_INTR);
+	reg &= ~(RDMA_INTR_THRESH_MASK |
+		 RDMA_TIMEOUT_MASK << RDMA_TIMEOUT_SHIFT);
+	reg |= pkts;
+	reg |= DIV_ROUND_UP(usecs * 1000, 8192) << RDMA_TIMEOUT_SHIFT;
+	rdma_writel(priv, reg, RDMA_MBDONE_INTR);
+}
+
+static void bcm_sysport_set_tx_coalesce(struct bcm_sysport_tx_ring *ring,
+					struct ethtool_coalesce *ec)
+{
+	struct bcm_sysport_priv *priv = ring->priv;
+	u32 reg;
+
+	reg = tdma_readl(priv, TDMA_DESC_RING_INTR_CONTROL(ring->index));
+	reg &= ~(RING_INTR_THRESH_MASK |
+		 RING_TIMEOUT_MASK << RING_TIMEOUT_SHIFT);
+	reg |= ec->tx_max_coalesced_frames;
+	reg |= DIV_ROUND_UP(ec->tx_coalesce_usecs * 1000, 8192) <<
+			    RING_TIMEOUT_SHIFT;
+	tdma_writel(priv, reg, TDMA_DESC_RING_INTR_CONTROL(ring->index));
+}
+
 static int bcm_sysport_get_coalesce(struct net_device *dev,
-				    struct ethtool_coalesce *ec)
+				    struct ethtool_coalesce *ec,
+				    struct kernel_ethtool_coalesce *kernel_coal,
+				    struct netlink_ext_ack *extack)
 {
 	struct bcm_sysport_priv *priv = netdev_priv(dev);
 	u32 reg;
@@ -545,16 +600,20 @@ static int bcm_sysport_get_coalesce(struct net_device *dev,
 
 	ec->rx_coalesce_usecs = (reg >> RDMA_TIMEOUT_SHIFT) * 8192 / 1000;
 	ec->rx_max_coalesced_frames = reg & RDMA_INTR_THRESH_MASK;
+	ec->use_adaptive_rx_coalesce = priv->dim.use_dim;
 
 	return 0;
 }
 
 static int bcm_sysport_set_coalesce(struct net_device *dev,
-				    struct ethtool_coalesce *ec)
+				    struct ethtool_coalesce *ec,
+				    struct kernel_ethtool_coalesce *kernel_coal,
+				    struct netlink_ext_ack *extack)
 {
 	struct bcm_sysport_priv *priv = netdev_priv(dev);
+	struct dim_cq_moder moder;
+	u32 usecs, pkts;
 	unsigned int i;
-	u32 reg;
 
 	/* Base system clock is 125Mhz, DMA timeout is this reference clock
 	 * divided by 1024, which yield roughly 8.192 us, our maximum value has
@@ -570,30 +629,31 @@ static int bcm_sysport_set_coalesce(struct net_device *dev,
 	    (ec->rx_coalesce_usecs == 0 && ec->rx_max_coalesced_frames == 0))
 		return -EINVAL;
 
-	for (i = 0; i < dev->num_tx_queues; i++) {
-		reg = tdma_readl(priv, TDMA_DESC_RING_INTR_CONTROL(i));
-		reg &= ~(RING_INTR_THRESH_MASK |
-			 RING_TIMEOUT_MASK << RING_TIMEOUT_SHIFT);
-		reg |= ec->tx_max_coalesced_frames;
-		reg |= DIV_ROUND_UP(ec->tx_coalesce_usecs * 1000, 8192) <<
-			 RING_TIMEOUT_SHIFT;
-		tdma_writel(priv, reg, TDMA_DESC_RING_INTR_CONTROL(i));
+	for (i = 0; i < dev->num_tx_queues; i++)
+		bcm_sysport_set_tx_coalesce(&priv->tx_rings[i], ec);
+
+	priv->rx_coalesce_usecs = ec->rx_coalesce_usecs;
+	priv->rx_max_coalesced_frames = ec->rx_max_coalesced_frames;
+	usecs = priv->rx_coalesce_usecs;
+	pkts = priv->rx_max_coalesced_frames;
+
+	if (ec->use_adaptive_rx_coalesce && !priv->dim.use_dim) {
+		moder = net_dim_get_def_rx_moderation(priv->dim.dim.mode);
+		usecs = moder.usec;
+		pkts = moder.pkts;
 	}
 
-	reg = rdma_readl(priv, RDMA_MBDONE_INTR);
-	reg &= ~(RDMA_INTR_THRESH_MASK |
-		 RDMA_TIMEOUT_MASK << RDMA_TIMEOUT_SHIFT);
-	reg |= ec->rx_max_coalesced_frames;
-	reg |= DIV_ROUND_UP(ec->rx_coalesce_usecs * 1000, 8192) <<
-			    RDMA_TIMEOUT_SHIFT;
-	rdma_writel(priv, reg, RDMA_MBDONE_INTR);
+	priv->dim.use_dim = ec->use_adaptive_rx_coalesce;
+
+	/* Apply desired coalescing parameters */
+	bcm_sysport_set_rx_coalesce(priv, usecs, pkts);
 
 	return 0;
 }
 
 static void bcm_sysport_free_cb(struct bcm_sysport_cb *cb)
 {
-	dev_kfree_skb_any(cb->skb);
+	dev_consume_skb_any(cb->skb);
 	cb->skb = NULL;
 	dma_unmap_addr_set(cb, dma_addr, 0);
 }
@@ -607,7 +667,8 @@ static struct sk_buff *bcm_sysport_rx_refill(struct bcm_sysport_priv *priv,
 	dma_addr_t mapping;
 
 	/* Allocate a new SKB for a new packet */
-	skb = netdev_alloc_skb(priv->netdev, RX_BUF_LENGTH);
+	skb = __netdev_alloc_skb(priv->netdev, RX_BUF_LENGTH,
+				 GFP_ATOMIC | __GFP_NOWARN);
 	if (!skb) {
 		priv->mib.alloc_rx_buff_failed++;
 		netif_err(priv, rx_err, ndev, "SKB alloc failed\n");
@@ -649,8 +710,7 @@ static int bcm_sysport_alloc_rx_bufs(struct bcm_sysport_priv *priv)
 	for (i = 0; i < priv->num_rx_bds; i++) {
 		cb = &priv->rx_cbs[i];
 		skb = bcm_sysport_rx_refill(priv, cb);
-		if (skb)
-			dev_kfree_skb(skb);
+		dev_kfree_skb(skb);
 		if (!cb->skb)
 			return -ENOMEM;
 	}
@@ -662,8 +722,10 @@ static int bcm_sysport_alloc_rx_bufs(struct bcm_sysport_priv *priv)
 static unsigned int bcm_sysport_desc_rx(struct bcm_sysport_priv *priv,
 					unsigned int budget)
 {
+	struct bcm_sysport_stats64 *stats64 = &priv->stats64;
 	struct net_device *ndev = priv->netdev;
 	unsigned int processed = 0, to_process;
+	unsigned int processed_bytes = 0;
 	struct bcm_sysport_cb *cb;
 	struct sk_buff *skb;
 	unsigned int p_index;
@@ -755,6 +817,7 @@ static unsigned int bcm_sysport_desc_rx(struct bcm_sysport_priv *priv,
 		 */
 		skb_pull(skb, sizeof(*rsb) + 2);
 		len -= (sizeof(*rsb) + 2);
+		processed_bytes += len;
 
 		/* UniMAC may forward CRC */
 		if (priv->crc_fwd) {
@@ -765,6 +828,10 @@ static unsigned int bcm_sysport_desc_rx(struct bcm_sysport_priv *priv,
 		skb->protocol = eth_type_trans(skb, ndev);
 		ndev->stats.rx_packets++;
 		ndev->stats.rx_bytes += len;
+		u64_stats_update_begin(&priv->syncp);
+		stats64->rx_packets++;
+		stats64->rx_bytes += len;
+		u64_stats_update_end(&priv->syncp);
 
 		napi_gro_receive(&priv->napi, skb);
 next:
@@ -775,6 +842,9 @@ next:
 			priv->rx_read_ptr = 0;
 	}
 
+	priv->dim.packets = processed;
+	priv->dim.bytes = processed_bytes;
+
 	return processed;
 }
 
@@ -787,17 +857,15 @@ static void bcm_sysport_tx_reclaim_one(struct bcm_sysport_tx_ring *ring,
 	struct device *kdev = &priv->pdev->dev;
 
 	if (cb->skb) {
-		ring->bytes += cb->skb->len;
 		*bytes_compl += cb->skb->len;
 		dma_unmap_single(kdev, dma_unmap_addr(cb, dma_addr),
 				 dma_unmap_len(cb, dma_len),
 				 DMA_TO_DEVICE);
-		ring->packets++;
 		(*pkts_compl)++;
 		bcm_sysport_free_cb(cb);
 	/* SKB fragment */
 	} else if (dma_unmap_addr(cb, dma_addr)) {
-		ring->bytes += dma_unmap_len(cb, dma_len);
+		*bytes_compl += dma_unmap_len(cb, dma_len);
 		dma_unmap_page(kdev, dma_unmap_addr(cb, dma_addr),
 			       dma_unmap_len(cb, dma_len), DMA_TO_DEVICE);
 		dma_unmap_addr_set(cb, dma_addr, 0);
@@ -808,10 +876,12 @@ static void bcm_sysport_tx_reclaim_one(struct bcm_sysport_tx_ring *ring,
 static unsigned int __bcm_sysport_tx_reclaim(struct bcm_sysport_priv *priv,
 					     struct bcm_sysport_tx_ring *ring)
 {
-	struct net_device *ndev = priv->netdev;
-	unsigned int c_index, last_c_index, last_tx_cn, num_tx_cbs;
 	unsigned int pkts_compl = 0, bytes_compl = 0;
+	struct net_device *ndev = priv->netdev;
+	unsigned int txbds_processed = 0;
 	struct bcm_sysport_cb *cb;
+	unsigned int txbds_ready;
+	unsigned int c_index;
 	u32 hw_ind;
 
 	/* Clear status before servicing to reduce spurious interrupts */
@@ -824,31 +894,30 @@ static unsigned int __bcm_sysport_tx_reclaim(struct bcm_sysport_priv *priv,
 	/* Compute how many descriptors have been processed since last call */
 	hw_ind = tdma_readl(priv, TDMA_DESC_RING_PROD_CONS_INDEX(ring->index));
 	c_index = (hw_ind >> RING_CONS_INDEX_SHIFT) & RING_CONS_INDEX_MASK;
-	ring->p_index = (hw_ind & RING_PROD_INDEX_MASK);
-
-	last_c_index = ring->c_index;
-	num_tx_cbs = ring->size;
-
-	c_index &= (num_tx_cbs - 1);
-
-	if (c_index >= last_c_index)
-		last_tx_cn = c_index - last_c_index;
-	else
-		last_tx_cn = num_tx_cbs - last_c_index + c_index;
+	txbds_ready = (c_index - ring->c_index) & RING_CONS_INDEX_MASK;
 
 	netif_dbg(priv, tx_done, ndev,
-		  "ring=%d c_index=%d last_tx_cn=%d last_c_index=%d\n",
-		  ring->index, c_index, last_tx_cn, last_c_index);
+		  "ring=%d old_c_index=%u c_index=%u txbds_ready=%u\n",
+		  ring->index, ring->c_index, c_index, txbds_ready);
 
-	while (last_tx_cn-- > 0) {
-		cb = ring->cbs + last_c_index;
+	while (txbds_processed < txbds_ready) {
+		cb = &ring->cbs[ring->clean_index];
 		bcm_sysport_tx_reclaim_one(ring, cb, &bytes_compl, &pkts_compl);
 
 		ring->desc_count++;
-		last_c_index++;
-		last_c_index &= (num_tx_cbs - 1);
+		txbds_processed++;
+
+		if (likely(ring->clean_index < ring->size - 1))
+			ring->clean_index++;
+		else
+			ring->clean_index = 0;
 	}
 
+	u64_stats_update_begin(&priv->syncp);
+	ring->packets += pkts_compl;
+	ring->bytes += bytes_compl;
+	u64_stats_update_end(&priv->syncp);
+
 	ring->c_index = c_index;
 
 	netif_dbg(priv, tx_done, ndev,
@@ -924,6 +993,7 @@ static int bcm_sysport_poll(struct napi_struct *napi, int budget)
 {
 	struct bcm_sysport_priv *priv =
 		container_of(napi, struct bcm_sysport_priv, napi);
+	struct dim_sample dim_sample = {};
 	unsigned int work_done = 0;
 
 	work_done = bcm_sysport_desc_rx(priv, budget);
@@ -946,24 +1016,90 @@ static int bcm_sysport_poll(struct napi_struct *napi, int budget)
 		intrl2_0_mask_clear(priv, INTRL2_0_RDMA_MBDONE);
 	}
 
+	if (priv->dim.use_dim) {
+		dim_update_sample(priv->dim.event_ctr, priv->dim.packets,
+				  priv->dim.bytes, &dim_sample);
+		net_dim(&priv->dim.dim, &dim_sample);
+	}
+
 	return work_done;
 }
 
+static void mpd_enable_set(struct bcm_sysport_priv *priv, bool enable)
+{
+	u32 reg, bit;
+
+	reg = umac_readl(priv, UMAC_MPD_CTRL);
+	if (enable)
+		reg |= MPD_EN;
+	else
+		reg &= ~MPD_EN;
+	umac_writel(priv, reg, UMAC_MPD_CTRL);
+
+	if (priv->is_lite)
+		bit = RBUF_ACPI_EN_LITE;
+	else
+		bit = RBUF_ACPI_EN;
+
+	reg = rbuf_readl(priv, RBUF_CONTROL);
+	if (enable)
+		reg |= bit;
+	else
+		reg &= ~bit;
+	rbuf_writel(priv, reg, RBUF_CONTROL);
+}
+
 static void bcm_sysport_resume_from_wol(struct bcm_sysport_priv *priv)
 {
+	unsigned int index;
 	u32 reg;
 
-	/* Stop monitoring MPD interrupt */
-	intrl2_0_mask_set(priv, INTRL2_0_MPD);
+	/* Disable RXCHK, active filters and Broadcom tag matching */
+	reg = rxchk_readl(priv, RXCHK_CONTROL);
+	reg &= ~(RXCHK_BRCM_TAG_MATCH_MASK <<
+		 RXCHK_BRCM_TAG_MATCH_SHIFT | RXCHK_EN | RXCHK_BRCM_TAG_EN);
+	rxchk_writel(priv, reg, RXCHK_CONTROL);
+
+	/* Make sure we restore correct CID index in case HW lost
+	 * its context during deep idle state
+	 */
+	for_each_set_bit(index, priv->filters, RXCHK_BRCM_TAG_MAX) {
+		rxchk_writel(priv, priv->filters_loc[index] <<
+			     RXCHK_BRCM_TAG_CID_SHIFT, RXCHK_BRCM_TAG(index));
+		rxchk_writel(priv, 0xff00ffff, RXCHK_BRCM_TAG_MASK(index));
+	}
 
 	/* Clear the MagicPacket detection logic */
-	reg = umac_readl(priv, UMAC_MPD_CTRL);
-	reg &= ~MPD_EN;
-	umac_writel(priv, reg, UMAC_MPD_CTRL);
+	mpd_enable_set(priv, false);
+
+	reg = intrl2_0_readl(priv, INTRL2_CPU_STATUS);
+	if (reg & INTRL2_0_MPD)
+		netdev_info(priv->netdev, "Wake-on-LAN (MPD) interrupt!\n");
+
+	if (reg & INTRL2_0_BRCM_MATCH_TAG) {
+		reg = rxchk_readl(priv, RXCHK_BRCM_TAG_MATCH_STATUS) &
+				  RXCHK_BRCM_TAG_MATCH_MASK;
+		netdev_info(priv->netdev,
+			    "Wake-on-LAN (filters 0x%02x) interrupt!\n", reg);
+	}
 
 	netif_dbg(priv, wol, priv->netdev, "resumed from WOL\n");
 }
 
+static void bcm_sysport_dim_work(struct work_struct *work)
+{
+	struct dim *dim = container_of(work, struct dim, work);
+	struct bcm_sysport_net_dim *ndim =
+			container_of(dim, struct bcm_sysport_net_dim, dim);
+	struct bcm_sysport_priv *priv =
+			container_of(ndim, struct bcm_sysport_priv, dim);
+	struct dim_cq_moder cur_profile = net_dim_get_rx_moderation(dim->mode,
+								    dim->profile_ix);
+
+	bcm_sysport_set_rx_coalesce(priv, cur_profile.usec, cur_profile.pkts);
+	dim->state = DIM_START_MEASURE;
+}
+
 /* RX and misc interrupt routine */
 static irqreturn_t bcm_sysport_rx_isr(int irq, void *dev_id)
 {
@@ -982,6 +1118,7 @@ static irqreturn_t bcm_sysport_rx_isr(int irq, void *dev_id)
 	}
 
 	if (priv->irq0_stat & INTRL2_0_RDMA_MBDONE) {
+		priv->dim.event_ctr++;
 		if (likely(napi_schedule_prep(&priv->napi))) {
 			/* disable RX interrupts */
 			intrl2_0_mask_set(priv, INTRL2_0_RDMA_MBDONE);
@@ -995,11 +1132,6 @@ static irqreturn_t bcm_sysport_rx_isr(int irq, void *dev_id)
 	if (priv->irq0_stat & INTRL2_0_TX_RING_FULL)
 		bcm_sysport_tx_reclaim_all(priv);
 
-	if (priv->irq0_stat & INTRL2_0_MPD) {
-		netdev_info(priv->netdev, "Wake-on-LAN interrupt!\n");
-		bcm_sysport_resume_from_wol(priv);
-	}
-
 	if (!priv->is_lite)
 		goto out;
 
@@ -1080,36 +1212,45 @@ static void bcm_sysport_poll_controller(struct net_device *dev)
 static struct sk_buff *bcm_sysport_insert_tsb(struct sk_buff *skb,
 					      struct net_device *dev)
 {
+	struct bcm_sysport_priv *priv = netdev_priv(dev);
 	struct sk_buff *nskb;
 	struct bcm_tsb *tsb;
 	u32 csum_info;
 	u8 ip_proto;
 	u16 csum_start;
-	u16 ip_ver;
+	__be16 ip_ver;
 
 	/* Re-allocate SKB if needed */
 	if (unlikely(skb_headroom(skb) < sizeof(*tsb))) {
 		nskb = skb_realloc_headroom(skb, sizeof(*tsb));
-		dev_kfree_skb(skb);
 		if (!nskb) {
+			dev_kfree_skb_any(skb);
+			priv->mib.tx_realloc_tsb_failed++;
 			dev->stats.tx_errors++;
 			dev->stats.tx_dropped++;
 			return NULL;
 		}
+		dev_consume_skb_any(skb);
 		skb = nskb;
+		priv->mib.tx_realloc_tsb++;
 	}
 
-	tsb = (struct bcm_tsb *)skb_push(skb, sizeof(*tsb));
+	tsb = skb_push(skb, sizeof(*tsb));
 	/* Zero-out TSB by default */
 	memset(tsb, 0, sizeof(*tsb));
 
+	if (skb_vlan_tag_present(skb)) {
+		tsb->pcp_dei_vid = skb_vlan_tag_get_prio(skb) & PCP_DEI_MASK;
+		tsb->pcp_dei_vid |= (u32)skb_vlan_tag_get_id(skb) << VID_SHIFT;
+	}
+
 	if (skb->ip_summed == CHECKSUM_PARTIAL) {
-		ip_ver = htons(skb->protocol);
+		ip_ver = skb->protocol;
 		switch (ip_ver) {
-		case ETH_P_IP:
+		case htons(ETH_P_IP):
 			ip_proto = ip_hdr(skb)->protocol;
 			break;
-		case ETH_P_IPV6:
+		case htons(ETH_P_IPV6):
 			ip_proto = ipv6_hdr(skb)->nexthdr;
 			break;
 		default:
@@ -1118,12 +1259,16 @@ static struct sk_buff *bcm_sysport_insert_tsb(struct sk_buff *skb,
 
 		/* Get the checksum offset and the L4 (transport) offset */
 		csum_start = skb_checksum_start_offset(skb) - sizeof(*tsb);
+		/* Account for the HW inserted VLAN tag */
+		if (skb_vlan_tag_present(skb))
+			csum_start += VLAN_HLEN;
 		csum_info = (csum_start + skb->csum_offset) & L4_CSUM_PTR_MASK;
 		csum_info |= (csum_start << L4_PTR_SHIFT);
 
 		if (ip_proto == IPPROTO_TCP || ip_proto == IPPROTO_UDP) {
 			csum_info |= L4_LENGTH_VALID;
-			if (ip_proto == IPPROTO_UDP && ip_ver == ETH_P_IP)
+			if (ip_proto == IPPROTO_UDP &&
+			    ip_ver == htons(ETH_P_IP))
 				csum_info |= L4_UDP;
 		} else {
 			csum_info = 0;
@@ -1141,13 +1286,12 @@ static netdev_tx_t bcm_sysport_xmit(struct sk_buff *skb,
 	struct bcm_sysport_priv *priv = netdev_priv(dev);
 	struct device *kdev = &priv->pdev->dev;
 	struct bcm_sysport_tx_ring *ring;
+	unsigned long flags, desc_flags;
 	struct bcm_sysport_cb *cb;
 	struct netdev_queue *txq;
-	struct dma_desc *desc;
+	u32 len_status, addr_lo;
 	unsigned int skb_len;
-	unsigned long flags;
 	dma_addr_t mapping;
-	u32 len_status;
 	u16 queue;
 	int ret;
 
@@ -1164,18 +1308,6 @@ static netdev_tx_t bcm_sysport_xmit(struct sk_buff *skb,
 		goto out;
 	}
 
-	/* The Ethernet switch we are interfaced with needs packets to be at
-	 * least 64 bytes (including FCS) otherwise they will be discarded when
-	 * they enter the switch port logic. When Broadcom tags are enabled, we
-	 * need to make sure that packets are at least 68 bytes
-	 * (including FCS and tag) because the length verification is done after
-	 * the Broadcom tag is stripped off the ingress packet.
-	 */
-	if (skb_put_padto(skb, ETH_ZLEN + ENET_BRCM_TAG_LEN)) {
-		ret = NETDEV_TX_OK;
-		goto out;
-	}
-
 	/* Insert TSB and checksum infos */
 	if (priv->tsb_en) {
 		skb = bcm_sysport_insert_tsb(skb, dev);
@@ -1193,6 +1325,7 @@ static netdev_tx_t bcm_sysport_xmit(struct sk_buff *skb,
 		netif_err(priv, tx_err, dev, "DMA map failed at %p (len=%d)\n",
 			  skb->data, skb_len);
 		ret = NETDEV_TX_OK;
+		dev_kfree_skb_any(skb);
 		goto out;
 	}
 
@@ -1202,32 +1335,26 @@ static netdev_tx_t bcm_sysport_xmit(struct sk_buff *skb,
 	dma_unmap_addr_set(cb, dma_addr, mapping);
 	dma_unmap_len_set(cb, dma_len, skb_len);
 
-	/* Fetch a descriptor entry from our pool */
-	desc = ring->desc_cpu;
-
-	desc->addr_lo = lower_32_bits(mapping);
+	addr_lo = lower_32_bits(mapping);
 	len_status = upper_32_bits(mapping) & DESC_ADDR_HI_MASK;
 	len_status |= (skb_len << DESC_LEN_SHIFT);
 	len_status |= (DESC_SOP | DESC_EOP | TX_STATUS_APP_CRC) <<
 		       DESC_STATUS_SHIFT;
 	if (skb->ip_summed == CHECKSUM_PARTIAL)
 		len_status |= (DESC_L4_CSUM << DESC_STATUS_SHIFT);
+	if (skb_vlan_tag_present(skb))
+		len_status |= (TX_STATUS_VLAN_VID_TSB << DESC_STATUS_SHIFT);
 
 	ring->curr_desc++;
 	if (ring->curr_desc == ring->size)
 		ring->curr_desc = 0;
 	ring->desc_count--;
 
-	/* Ensure write completion of the descriptor status/length
-	 * in DRAM before the System Port WRITE_PORT register latches
-	 * the value
-	 */
-	wmb();
-	desc->addr_status_len = len_status;
-	wmb();
-
-	/* Write this descriptor address to the RING write port */
-	tdma_port_write_desc_addr(priv, desc, ring->index);
+	/* Ports are latched, so write upper address first */
+	spin_lock_irqsave(&priv->desc_lock, desc_flags);
+	tdma_writel(priv, len_status, TDMA_WRITE_PORT_HI(ring->index));
+	tdma_writel(priv, addr_lo, TDMA_WRITE_PORT_LO(ring->index));
+	spin_unlock_irqrestore(&priv->desc_lock, desc_flags);
 
 	/* Check ring space and update SW control flow */
 	if (ring->desc_count == 0)
@@ -1242,7 +1369,7 @@ out:
 	return ret;
 }
 
-static void bcm_sysport_tx_timeout(struct net_device *dev)
+static void bcm_sysport_tx_timeout(struct net_device *dev, unsigned int txqueue)
 {
 	netdev_warn(dev, "transmit timeout!\n");
 
@@ -1318,28 +1445,47 @@ out:
 		phy_print_status(phydev);
 }
 
+static void bcm_sysport_init_dim(struct bcm_sysport_priv *priv,
+				 void (*cb)(struct work_struct *work))
+{
+	struct bcm_sysport_net_dim *dim = &priv->dim;
+
+	INIT_WORK(&dim->dim.work, cb);
+	dim->dim.mode = DIM_CQ_PERIOD_MODE_START_FROM_EQE;
+	dim->event_ctr = 0;
+	dim->packets = 0;
+	dim->bytes = 0;
+}
+
+static void bcm_sysport_init_rx_coalesce(struct bcm_sysport_priv *priv)
+{
+	struct bcm_sysport_net_dim *dim = &priv->dim;
+	struct dim_cq_moder moder;
+	u32 usecs, pkts;
+
+	usecs = priv->rx_coalesce_usecs;
+	pkts = priv->rx_max_coalesced_frames;
+
+	/* If DIM was enabled, re-apply default parameters */
+	if (dim->use_dim) {
+		moder = net_dim_get_def_rx_moderation(dim->dim.mode);
+		usecs = moder.usec;
+		pkts = moder.pkts;
+	}
+
+	bcm_sysport_set_rx_coalesce(priv, usecs, pkts);
+}
+
 static int bcm_sysport_init_tx_ring(struct bcm_sysport_priv *priv,
 				    unsigned int index)
 {
 	struct bcm_sysport_tx_ring *ring = &priv->tx_rings[index];
-	struct device *kdev = &priv->pdev->dev;
 	size_t size;
-	void *p;
 	u32 reg;
 
 	/* Simple descriptors partitioning for now */
 	size = 256;
 
-	/* We just need one DMA descriptor which is DMA-able, since writing to
-	 * the port will allocate a new descriptor in its internal linked-list
-	 */
-	p = dma_zalloc_coherent(kdev, sizeof(struct dma_desc), &ring->desc_dma,
-				GFP_KERNEL);
-	if (!p) {
-		netif_err(priv, hw, priv->netdev, "DMA alloc failed\n");
-		return -ENOMEM;
-	}
-
 	ring->cbs = kcalloc(size, sizeof(struct bcm_sysport_cb), GFP_KERNEL);
 	if (!ring->cbs) {
 		netif_err(priv, hw, priv->netdev, "CB allocation failed\n");
@@ -1349,11 +1495,11 @@ static int bcm_sysport_init_tx_ring(struct bcm_sysport_priv *priv,
 	/* Initialize SW view of the ring */
 	spin_lock_init(&ring->lock);
 	ring->priv = priv;
-	netif_tx_napi_add(priv->netdev, &ring->napi, bcm_sysport_tx_poll, 64);
+	netif_napi_add_tx(priv->netdev, &ring->napi, bcm_sysport_tx_poll);
 	ring->index = index;
 	ring->size = size;
+	ring->clean_index = 0;
 	ring->alloc_size = ring->size;
-	ring->desc_cpu = p;
 	ring->desc_count = ring->size;
 	ring->curr_desc = 0;
 
@@ -1362,8 +1508,42 @@ static int bcm_sysport_init_tx_ring(struct bcm_sysport_priv *priv,
 	tdma_writel(priv, 0, TDMA_DESC_RING_COUNT(index));
 	tdma_writel(priv, 1, TDMA_DESC_RING_INTR_CONTROL(index));
 	tdma_writel(priv, 0, TDMA_DESC_RING_PROD_CONS_INDEX(index));
-	tdma_writel(priv, RING_IGNORE_STATUS, TDMA_DESC_RING_MAPPING(index));
-	tdma_writel(priv, 0, TDMA_DESC_RING_PCP_DEI_VID(index));
+
+	/* Configure QID and port mapping */
+	reg = tdma_readl(priv, TDMA_DESC_RING_MAPPING(index));
+	reg &= ~(RING_QID_MASK | RING_PORT_ID_MASK << RING_PORT_ID_SHIFT);
+	if (ring->inspect) {
+		reg |= ring->switch_queue & RING_QID_MASK;
+		reg |= ring->switch_port << RING_PORT_ID_SHIFT;
+	} else {
+		reg |= RING_IGNORE_STATUS;
+	}
+	tdma_writel(priv, reg, TDMA_DESC_RING_MAPPING(index));
+	reg = 0;
+	/* Adjust the packet size calculations if SYSTEMPORT is responsible
+	 * for HW insertion of VLAN tags
+	 */
+	if (priv->netdev->features & NETIF_F_HW_VLAN_CTAG_TX)
+		reg = VLAN_HLEN << RING_PKT_SIZE_ADJ_SHIFT;
+	tdma_writel(priv, reg, TDMA_DESC_RING_PCP_DEI_VID(index));
+
+	/* Enable ACB algorithm 2 */
+	reg = tdma_readl(priv, TDMA_CONTROL);
+	reg |= tdma_control_bit(priv, ACB_ALGO);
+	tdma_writel(priv, reg, TDMA_CONTROL);
+
+	/* Do not use tdma_control_bit() here because TSB_SWAP1 collides
+	 * with the original definition of ACB_ALGO
+	 */
+	reg = tdma_readl(priv, TDMA_CONTROL);
+	if (priv->is_lite)
+		reg &= ~BIT(TSB_SWAP1);
+	/* Set a correct TSB format based on host endian */
+	if (!IS_ENABLED(CONFIG_CPU_BIG_ENDIAN))
+		reg |= tdma_control_bit(priv, TSB_SWAP0);
+	else
+		reg &= ~tdma_control_bit(priv, TSB_SWAP0);
+	tdma_writel(priv, reg, TDMA_CONTROL);
 
 	/* Program the number of descriptors as MAX_THRESHOLD and half of
 	 * its size for the hysteresis trigger
@@ -1380,8 +1560,9 @@ static int bcm_sysport_init_tx_ring(struct bcm_sysport_priv *priv,
 	napi_enable(&ring->napi);
 
 	netif_dbg(priv, hw, priv->netdev,
-		  "TDMA cfg, size=%d, desc_cpu=%p\n",
-		  ring->size, ring->desc_cpu);
+		  "TDMA cfg, size=%d, switch q=%d,port=%d\n",
+		  ring->size, ring->switch_queue,
+		  ring->switch_port);
 
 	return 0;
 }
@@ -1390,7 +1571,6 @@ static void bcm_sysport_fini_tx_ring(struct bcm_sysport_priv *priv,
 				     unsigned int index)
 {
 	struct bcm_sysport_tx_ring *ring = &priv->tx_rings[index];
-	struct device *kdev = &priv->pdev->dev;
 	u32 reg;
 
 	/* Caller should stop the TDMA engine */
@@ -1412,12 +1592,6 @@ static void bcm_sysport_fini_tx_ring(struct bcm_sysport_priv *priv,
 
 	kfree(ring->cbs);
 	ring->cbs = NULL;
-
-	if (ring->desc_dma) {
-		dma_free_coherent(kdev, sizeof(struct dma_desc),
-				  ring->desc_cpu, ring->desc_dma);
-		ring->desc_dma = 0;
-	}
 	ring->size = 0;
 	ring->alloc_size = 0;
 
@@ -1526,8 +1700,6 @@ static int bcm_sysport_init_rx_ring(struct bcm_sysport_priv *priv)
 	rdma_writel(priv, 0, RDMA_END_ADDR_HI);
 	rdma_writel(priv, priv->num_rx_desc_words - 1, RDMA_END_ADDR_LO);
 
-	rdma_writel(priv, 1, RDMA_MBDONE_INTR);
-
 	netif_dbg(priv, hw, priv->netdev,
 		  "RDMA cfg, num_rx_bds=%d, rx_bds=%p\n",
 		  priv->num_rx_bds, priv->rx_bds);
@@ -1626,7 +1798,7 @@ static inline void umac_reset(struct bcm_sysport_priv *priv)
 }
 
 static void umac_set_hw_addr(struct bcm_sysport_priv *priv,
-			     unsigned char *addr)
+			     const unsigned char *addr)
 {
 	u32 mac0 = (addr[0] << 24) | (addr[1] << 16) | (addr[2] << 8) |
 		    addr[3];
@@ -1658,7 +1830,7 @@ static int bcm_sysport_change_mac(struct net_device *dev, void *p)
 	if (!is_valid_ether_addr(addr->sa_data))
 		return -EINVAL;
 
-	memcpy(dev->dev_addr, addr->sa_data, dev->addr_len);
+	eth_hw_addr_set(dev, addr->sa_data);
 
 	/* interface is disabled, changes to MAC will be reflected on next
 	 * open call
@@ -1671,22 +1843,23 @@ static int bcm_sysport_change_mac(struct net_device *dev, void *p)
 	return 0;
 }
 
-static struct net_device_stats *bcm_sysport_get_nstats(struct net_device *dev)
+static void bcm_sysport_get_stats64(struct net_device *dev,
+				    struct rtnl_link_stats64 *stats)
 {
 	struct bcm_sysport_priv *priv = netdev_priv(dev);
-	unsigned long tx_bytes = 0, tx_packets = 0;
-	struct bcm_sysport_tx_ring *ring;
-	unsigned int q;
+	struct bcm_sysport_stats64 *stats64 = &priv->stats64;
+	unsigned int start;
 
-	for (q = 0; q < dev->num_tx_queues; q++) {
-		ring = &priv->tx_rings[q];
-		tx_bytes += ring->bytes;
-		tx_packets += ring->packets;
-	}
+	netdev_stats_to_stats64(stats, &dev->stats);
 
-	dev->stats.tx_bytes = tx_bytes;
-	dev->stats.tx_packets = tx_packets;
-	return &dev->stats;
+	bcm_sysport_update_tx_stats(priv, &stats->tx_bytes,
+				    &stats->tx_packets);
+
+	do {
+		start = u64_stats_fetch_begin(&priv->syncp);
+		stats->rx_packets = stats64->rx_packets;
+		stats->rx_bytes = stats64->rx_bytes;
+	} while (u64_stats_fetch_retry(&priv->syncp, start));
 }
 
 static void bcm_sysport_netif_start(struct net_device *dev)
@@ -1694,6 +1867,8 @@ static void bcm_sysport_netif_start(struct net_device *dev)
 	struct bcm_sysport_priv *priv = netdev_priv(dev);
 
 	/* Enable NAPI */
+	bcm_sysport_init_dim(priv, bcm_sysport_dim_work);
+	bcm_sysport_init_rx_coalesce(priv);
 	napi_enable(&priv->napi);
 
 	/* Enable RX interrupt and TX ring full interrupt */
@@ -1706,9 +1881,6 @@ static void bcm_sysport_netif_start(struct net_device *dev)
 		intrl2_1_mask_clear(priv, 0xffffffff);
 	else
 		intrl2_0_mask_clear(priv, INTRL2_0_TDMA_MBDONE_MASK);
-
-	/* Last call before we start the real business */
-	netif_tx_start_all_queues(dev);
 }
 
 static void rbuf_init(struct bcm_sysport_priv *priv)
@@ -1718,10 +1890,14 @@ static void rbuf_init(struct bcm_sysport_priv *priv)
 	reg = rbuf_readl(priv, RBUF_CONTROL);
 	reg |= RBUF_4B_ALGN | RBUF_RSB_EN;
 	/* Set a correct RSB format on SYSTEMPORT Lite */
-	if (priv->is_lite) {
+	if (priv->is_lite)
 		reg &= ~RBUF_RSB_SWAP1;
+
+	/* Set a correct RSB format based on host endian */
+	if (!IS_ENABLED(CONFIG_CPU_BIG_ENDIAN))
 		reg |= RBUF_RSB_SWAP0;
-	}
+	else
+		reg &= ~RBUF_RSB_SWAP0;
 	rbuf_writel(priv, reg, RBUF_CONTROL);
 }
 
@@ -1737,15 +1913,17 @@ static inline void bcm_sysport_mask_all_intrs(struct bcm_sysport_priv *priv)
 
 static inline void gib_set_pad_extension(struct bcm_sysport_priv *priv)
 {
-	u32 __maybe_unused reg;
+	u32 reg;
 
-	/* Include Broadcom tag in pad extension */
+	reg = gib_readl(priv, GIB_CONTROL);
+	/* Include Broadcom tag in pad extension and fix up IPG_LENGTH */
 	if (netdev_uses_dsa(priv->netdev)) {
-		reg = gib_readl(priv, GIB_CONTROL);
 		reg &= ~(GIB_PAD_EXTENSION_MASK << GIB_PAD_EXTENSION_SHIFT);
 		reg |= ENET_BRCM_TAG_LEN << GIB_PAD_EXTENSION_SHIFT;
-		gib_writel(priv, reg, GIB_CONTROL);
 	}
+	reg &= ~(GIB_IPG_LEN_MASK << GIB_IPG_LEN_SHIFT);
+	reg |= 12 << GIB_IPG_LEN_SHIFT;
+	gib_writel(priv, reg, GIB_CONTROL);
 }
 
 static int bcm_sysport_open(struct net_device *dev)
@@ -1755,6 +1933,12 @@ static int bcm_sysport_open(struct net_device *dev)
 	unsigned int i;
 	int ret;
 
+	ret = clk_prepare_enable(priv->clk);
+	if (ret) {
+		netdev_err(dev, "could not enable priv clock\n");
+		return ret;
+	}
+
 	/* Reset UniMAC */
 	umac_reset(priv);
 
@@ -1773,23 +1957,25 @@ static int bcm_sysport_open(struct net_device *dev)
 	else
 		gib_set_pad_extension(priv);
 
+	/* Apply features again in case we changed them while interface was
+	 * down
+	 */
+	bcm_sysport_set_features(dev, dev->features);
+
 	/* Set MAC address */
 	umac_set_hw_addr(priv, dev->dev_addr);
 
-	/* Read CRC forward */
-	if (!priv->is_lite)
-		priv->crc_fwd = !!(umac_readl(priv, UMAC_CMD) & CMD_CRC_FWD);
-	else
-		priv->crc_fwd = !!(gib_readl(priv, GIB_CONTROL) &
-				   GIB_FCS_STRIP);
-
 	phydev = of_phy_connect(dev, priv->phy_dn, bcm_sysport_adj_link,
 				0, priv->phy_interface);
 	if (!phydev) {
 		netdev_err(dev, "could not attach to PHY\n");
-		return -ENODEV;
+		ret = -ENODEV;
+		goto out_clk_disable;
 	}
 
+	/* Indicate that the MAC is responsible for PHY PM */
+	phydev->mac_managed_pm = true;
+
 	/* Reset house keeping link status */
 	priv->old_duplex = -1;
 	priv->old_link = -1;
@@ -1814,6 +2000,7 @@ static int bcm_sysport_open(struct net_device *dev)
 	}
 
 	/* Initialize both hardware and software ring */
+	spin_lock_init(&priv->desc_lock);
 	for (i = 0; i < dev->num_tx_queues; i++) {
 		ret = bcm_sysport_init_tx_ring(priv, i);
 		if (ret) {
@@ -1848,6 +2035,8 @@ static int bcm_sysport_open(struct net_device *dev)
 
 	bcm_sysport_netif_start(dev);
 
+	netif_tx_start_all_queues(dev);
+
 	return 0;
 
 out_clear_rx_int:
@@ -1863,6 +2052,8 @@ out_free_irq0:
 	free_irq(priv->irq0, dev);
 out_phy_disconnect:
 	phy_disconnect(phydev);
+out_clk_disable:
+	clk_disable_unprepare(priv->clk);
 	return ret;
 }
 
@@ -1871,8 +2062,9 @@ static void bcm_sysport_netif_stop(struct net_device *dev)
 	struct bcm_sysport_priv *priv = netdev_priv(dev);
 
 	/* stop all software from updating hardware */
-	netif_tx_stop_all_queues(dev);
+	netif_tx_disable(dev);
 	napi_disable(&priv->napi);
+	cancel_work_sync(&priv->dim.dim.work);
 	phy_stop(dev->phydev);
 
 	/* mask all interrupts */
@@ -1920,10 +2112,139 @@ static int bcm_sysport_stop(struct net_device *dev)
 	/* Disconnect from PHY */
 	phy_disconnect(dev->phydev);
 
+	clk_disable_unprepare(priv->clk);
+
+	return 0;
+}
+
+static int bcm_sysport_rule_find(struct bcm_sysport_priv *priv,
+				 u64 location)
+{
+	unsigned int index;
+	u32 reg;
+
+	for_each_set_bit(index, priv->filters, RXCHK_BRCM_TAG_MAX) {
+		reg = rxchk_readl(priv, RXCHK_BRCM_TAG(index));
+		reg >>= RXCHK_BRCM_TAG_CID_SHIFT;
+		reg &= RXCHK_BRCM_TAG_CID_MASK;
+		if (reg == location)
+			return index;
+	}
+
+	return -EINVAL;
+}
+
+static int bcm_sysport_rule_get(struct bcm_sysport_priv *priv,
+				struct ethtool_rxnfc *nfc)
+{
+	int index;
+
+	/* This is not a rule that we know about */
+	index = bcm_sysport_rule_find(priv, nfc->fs.location);
+	if (index < 0)
+		return -EOPNOTSUPP;
+
+	nfc->fs.ring_cookie = RX_CLS_FLOW_WAKE;
+
+	return 0;
+}
+
+static int bcm_sysport_rule_set(struct bcm_sysport_priv *priv,
+				struct ethtool_rxnfc *nfc)
+{
+	unsigned int index;
+	u32 reg;
+
+	/* We cannot match locations greater than what the classification ID
+	 * permits (256 entries)
+	 */
+	if (nfc->fs.location > RXCHK_BRCM_TAG_CID_MASK)
+		return -E2BIG;
+
+	/* We cannot support flows that are not destined for a wake-up */
+	if (nfc->fs.ring_cookie != RX_CLS_FLOW_WAKE)
+		return -EOPNOTSUPP;
+
+	index = find_first_zero_bit(priv->filters, RXCHK_BRCM_TAG_MAX);
+	if (index >= RXCHK_BRCM_TAG_MAX)
+		/* All filters are already in use, we cannot match more rules */
+		return -ENOSPC;
+
+	/* Location is the classification ID, and index is the position
+	 * within one of our 8 possible filters to be programmed
+	 */
+	reg = rxchk_readl(priv, RXCHK_BRCM_TAG(index));
+	reg &= ~(RXCHK_BRCM_TAG_CID_MASK << RXCHK_BRCM_TAG_CID_SHIFT);
+	reg |= nfc->fs.location << RXCHK_BRCM_TAG_CID_SHIFT;
+	rxchk_writel(priv, reg, RXCHK_BRCM_TAG(index));
+	rxchk_writel(priv, 0xff00ffff, RXCHK_BRCM_TAG_MASK(index));
+
+	priv->filters_loc[index] = nfc->fs.location;
+	set_bit(index, priv->filters);
+
+	return 0;
+}
+
+static int bcm_sysport_rule_del(struct bcm_sysport_priv *priv,
+				u64 location)
+{
+	int index;
+
+	/* This is not a rule that we know about */
+	index = bcm_sysport_rule_find(priv, location);
+	if (index < 0)
+		return -EOPNOTSUPP;
+
+	/* No need to disable this filter if it was enabled, this will
+	 * be taken care of during suspend time by bcm_sysport_suspend_to_wol
+	 */
+	clear_bit(index, priv->filters);
+	priv->filters_loc[index] = 0;
+
 	return 0;
 }
 
+static int bcm_sysport_get_rxnfc(struct net_device *dev,
+				 struct ethtool_rxnfc *nfc, u32 *rule_locs)
+{
+	struct bcm_sysport_priv *priv = netdev_priv(dev);
+	int ret = -EOPNOTSUPP;
+
+	switch (nfc->cmd) {
+	case ETHTOOL_GRXCLSRULE:
+		ret = bcm_sysport_rule_get(priv, nfc);
+		break;
+	default:
+		break;
+	}
+
+	return ret;
+}
+
+static int bcm_sysport_set_rxnfc(struct net_device *dev,
+				 struct ethtool_rxnfc *nfc)
+{
+	struct bcm_sysport_priv *priv = netdev_priv(dev);
+	int ret = -EOPNOTSUPP;
+
+	switch (nfc->cmd) {
+	case ETHTOOL_SRXCLSRLINS:
+		ret = bcm_sysport_rule_set(priv, nfc);
+		break;
+	case ETHTOOL_SRXCLSRLDEL:
+		ret = bcm_sysport_rule_del(priv, nfc->fs.location);
+		break;
+	default:
+		break;
+	}
+
+	return ret;
+}
+
 static const struct ethtool_ops bcm_sysport_ethtool_ops = {
+	.supported_coalesce_params = ETHTOOL_COALESCE_USECS |
+				     ETHTOOL_COALESCE_MAX_FRAMES |
+				     ETHTOOL_COALESCE_USE_ADAPTIVE_RX,
 	.get_drvinfo		= bcm_sysport_get_drvinfo,
 	.get_msglevel		= bcm_sysport_get_msglvl,
 	.set_msglevel		= bcm_sysport_set_msglvl,
@@ -1937,8 +2258,32 @@ static const struct ethtool_ops bcm_sysport_ethtool_ops = {
 	.set_coalesce		= bcm_sysport_set_coalesce,
 	.get_link_ksettings     = phy_ethtool_get_link_ksettings,
 	.set_link_ksettings     = phy_ethtool_set_link_ksettings,
+	.get_rxnfc		= bcm_sysport_get_rxnfc,
+	.set_rxnfc		= bcm_sysport_set_rxnfc,
 };
 
+static u16 bcm_sysport_select_queue(struct net_device *dev, struct sk_buff *skb,
+				    struct net_device *sb_dev)
+{
+	struct bcm_sysport_priv *priv = netdev_priv(dev);
+	u16 queue = skb_get_queue_mapping(skb);
+	struct bcm_sysport_tx_ring *tx_ring;
+	unsigned int q, port;
+
+	if (!netdev_uses_dsa(dev))
+		return netdev_pick_tx(dev, skb, NULL);
+
+	/* DSA tagging layer will have configured the correct queue */
+	q = BRCM_TAG_GET_QUEUE(queue);
+	port = BRCM_TAG_GET_PORT(queue);
+	tx_ring = priv->ring_map[q + port * priv->per_port_num_tx_queues];
+
+	if (unlikely(!tx_ring))
+		return netdev_pick_tx(dev, skb, NULL);
+
+	return tx_ring->index;
+}
+
 static const struct net_device_ops bcm_sysport_netdev_ops = {
 	.ndo_start_xmit		= bcm_sysport_xmit,
 	.ndo_tx_timeout		= bcm_sysport_tx_timeout,
@@ -1950,9 +2295,125 @@ static const struct net_device_ops bcm_sysport_netdev_ops = {
 #ifdef CONFIG_NET_POLL_CONTROLLER
 	.ndo_poll_controller	= bcm_sysport_poll_controller,
 #endif
-	.ndo_get_stats		= bcm_sysport_get_nstats,
+	.ndo_get_stats64	= bcm_sysport_get_stats64,
+	.ndo_select_queue	= bcm_sysport_select_queue,
 };
 
+static int bcm_sysport_map_queues(struct net_device *dev,
+				  struct net_device *slave_dev)
+{
+	struct dsa_port *dp = dsa_port_from_netdev(slave_dev);
+	struct bcm_sysport_priv *priv = netdev_priv(dev);
+	struct bcm_sysport_tx_ring *ring;
+	unsigned int num_tx_queues;
+	unsigned int q, qp, port;
+
+	/* We can't be setting up queue inspection for non directly attached
+	 * switches
+	 */
+	if (dp->ds->index)
+		return 0;
+
+	port = dp->index;
+
+	/* On SYSTEMPORT Lite we have twice as less queues, so we cannot do a
+	 * 1:1 mapping, we can only do a 2:1 mapping. By reducing the number of
+	 * per-port (slave_dev) network devices queue, we achieve just that.
+	 * This need to happen now before any slave network device is used such
+	 * it accurately reflects the number of real TX queues.
+	 */
+	if (priv->is_lite)
+		netif_set_real_num_tx_queues(slave_dev,
+					     slave_dev->num_tx_queues / 2);
+
+	num_tx_queues = slave_dev->real_num_tx_queues;
+
+	if (priv->per_port_num_tx_queues &&
+	    priv->per_port_num_tx_queues != num_tx_queues)
+		netdev_warn(slave_dev, "asymmetric number of per-port queues\n");
+
+	priv->per_port_num_tx_queues = num_tx_queues;
+
+	for (q = 0, qp = 0; q < dev->num_tx_queues && qp < num_tx_queues;
+	     q++) {
+		ring = &priv->tx_rings[q];
+
+		if (ring->inspect)
+			continue;
+
+		/* Just remember the mapping actual programming done
+		 * during bcm_sysport_init_tx_ring
+		 */
+		ring->switch_queue = qp;
+		ring->switch_port = port;
+		ring->inspect = true;
+		priv->ring_map[qp + port * num_tx_queues] = ring;
+		qp++;
+	}
+
+	return 0;
+}
+
+static int bcm_sysport_unmap_queues(struct net_device *dev,
+				    struct net_device *slave_dev)
+{
+	struct dsa_port *dp = dsa_port_from_netdev(slave_dev);
+	struct bcm_sysport_priv *priv = netdev_priv(dev);
+	struct bcm_sysport_tx_ring *ring;
+	unsigned int num_tx_queues;
+	unsigned int q, qp, port;
+
+	port = dp->index;
+
+	num_tx_queues = slave_dev->real_num_tx_queues;
+
+	for (q = 0; q < dev->num_tx_queues; q++) {
+		ring = &priv->tx_rings[q];
+
+		if (ring->switch_port != port)
+			continue;
+
+		if (!ring->inspect)
+			continue;
+
+		ring->inspect = false;
+		qp = ring->switch_queue;
+		priv->ring_map[qp + port * num_tx_queues] = NULL;
+	}
+
+	return 0;
+}
+
+static int bcm_sysport_netdevice_event(struct notifier_block *nb,
+				       unsigned long event, void *ptr)
+{
+	struct net_device *dev = netdev_notifier_info_to_dev(ptr);
+	struct netdev_notifier_changeupper_info *info = ptr;
+	struct bcm_sysport_priv *priv;
+	int ret = 0;
+
+	priv = container_of(nb, struct bcm_sysport_priv, netdev_notifier);
+	if (priv->netdev != dev)
+		return NOTIFY_DONE;
+
+	switch (event) {
+	case NETDEV_CHANGEUPPER:
+		if (dev->netdev_ops != &bcm_sysport_netdev_ops)
+			return NOTIFY_DONE;
+
+		if (!dsa_user_dev_check(info->upper_dev))
+			return NOTIFY_DONE;
+
+		if (info->linking)
+			ret = bcm_sysport_map_queues(dev, info->upper_dev);
+		else
+			ret = bcm_sysport_unmap_queues(dev, info->upper_dev);
+		break;
+	}
+
+	return notifier_from_errno(ret);
+}
+
 #define REV_FMT	"v%2x.%02x"
 
 static const struct bcm_sysport_hw_params bcm_sysport_params[] = {
@@ -1984,17 +2445,22 @@ static int bcm_sysport_probe(struct platform_device *pdev)
 	struct bcm_sysport_priv *priv;
 	struct device_node *dn;
 	struct net_device *dev;
-	const void *macaddr;
-	struct resource *r;
 	u32 txq, rxq;
 	int ret;
 
 	dn = pdev->dev.of_node;
-	r = platform_get_resource(pdev, IORESOURCE_MEM, 0);
 	of_id = of_match_node(bcm_sysport_of_match, dn);
 	if (!of_id || !of_id->data)
 		return -EINVAL;
 
+	ret = dma_set_mask_and_coherent(&pdev->dev, DMA_BIT_MASK(40));
+	if (ret)
+		ret = dma_set_mask_and_coherent(&pdev->dev, DMA_BIT_MASK(32));
+	if (ret) {
+		dev_err(&pdev->dev, "unable to set DMA mask: %d\n", ret);
+		return ret;
+	}
+
 	/* Fairly quickly we need to know the type of adapter we have */
 	params = of_id->data;
 
@@ -2015,27 +2481,37 @@ static int bcm_sysport_probe(struct platform_device *pdev)
 	/* Initialize private members */
 	priv = netdev_priv(dev);
 
+	priv->clk = devm_clk_get_optional(&pdev->dev, "sw_sysport");
+	if (IS_ERR(priv->clk)) {
+		ret = PTR_ERR(priv->clk);
+		goto err_free_netdev;
+	}
+
 	/* Allocate number of TX rings */
 	priv->tx_rings = devm_kcalloc(&pdev->dev, txq,
 				      sizeof(struct bcm_sysport_tx_ring),
 				      GFP_KERNEL);
-	if (!priv->tx_rings)
-		return -ENOMEM;
+	if (!priv->tx_rings) {
+		ret = -ENOMEM;
+		goto err_free_netdev;
+	}
 
 	priv->is_lite = params->is_lite;
 	priv->num_rx_desc_words = params->num_rx_desc_words;
 
 	priv->irq0 = platform_get_irq(pdev, 0);
-	if (!priv->is_lite)
+	if (!priv->is_lite) {
 		priv->irq1 = platform_get_irq(pdev, 1);
-	priv->wol_irq = platform_get_irq(pdev, 2);
+		priv->wol_irq = platform_get_irq_optional(pdev, 2);
+	} else {
+		priv->wol_irq = platform_get_irq_optional(pdev, 1);
+	}
 	if (priv->irq0 <= 0 || (priv->irq1 <= 0 && !priv->is_lite)) {
-		dev_err(&pdev->dev, "invalid interrupts\n");
 		ret = -EINVAL;
 		goto err_free_netdev;
 	}
 
-	priv->base = devm_ioremap_resource(&pdev->dev, r);
+	priv->base = devm_platform_ioremap_resource(pdev, 0);
 	if (IS_ERR(priv->base)) {
 		ret = PTR_ERR(priv->base);
 		goto err_free_netdev;
@@ -2044,9 +2520,9 @@ static int bcm_sysport_probe(struct platform_device *pdev)
 	priv->netdev = dev;
 	priv->pdev = pdev;
 
-	priv->phy_interface = of_get_phy_mode(dn);
+	ret = of_get_phy_mode(dn, &priv->phy_interface);
 	/* Default to GMII interface mode */
-	if (priv->phy_interface < 0)
+	if (ret)
 		priv->phy_interface = PHY_INTERFACE_MODE_GMII;
 
 	/* In the case of a fixed PHY, the DT node associated
@@ -2063,23 +2539,24 @@ static int bcm_sysport_probe(struct platform_device *pdev)
 	}
 
 	/* Initialize netdevice members */
-	macaddr = of_get_mac_address(dn);
-	if (!macaddr || !is_valid_ether_addr(macaddr)) {
+	ret = of_get_ethdev_address(dn, dev);
+	if (ret) {
 		dev_warn(&pdev->dev, "using random Ethernet MAC\n");
 		eth_hw_addr_random(dev);
-	} else {
-		ether_addr_copy(dev->dev_addr, macaddr);
 	}
 
 	SET_NETDEV_DEV(dev, &pdev->dev);
 	dev_set_drvdata(&pdev->dev, dev);
 	dev->ethtool_ops = &bcm_sysport_ethtool_ops;
 	dev->netdev_ops = &bcm_sysport_netdev_ops;
-	netif_napi_add(dev, &priv->napi, bcm_sysport_poll, 64);
+	netif_napi_add(dev, &priv->napi, bcm_sysport_poll);
 
-	/* HW supported features, none enabled by default */
-	dev->hw_features |= NETIF_F_RXCSUM | NETIF_F_HIGHDMA |
-				NETIF_F_IP_CSUM | NETIF_F_IPV6_CSUM;
+	dev->features |= NETIF_F_RXCSUM | NETIF_F_HIGHDMA |
+			 NETIF_F_IP_CSUM | NETIF_F_IPV6_CSUM |
+			 NETIF_F_HW_VLAN_CTAG_TX;
+	dev->hw_features |= dev->features;
+	dev->vlan_features |= dev->features;
+	dev->max_mtu = UMAC_MAX_MTU_SIZE;
 
 	/* Request the WOL interrupt and advertise suspend if available */
 	priv->wol_irq_disabled = 1;
@@ -2088,6 +2565,12 @@ static int bcm_sysport_probe(struct platform_device *pdev)
 	if (!ret)
 		device_set_wakeup_capable(&pdev->dev, 1);
 
+	priv->wol_clk = devm_clk_get_optional(&pdev->dev, "sw_sysportwol");
+	if (IS_ERR(priv->wol_clk)) {
+		ret = PTR_ERR(priv->wol_clk);
+		goto err_deregister_fixed_link;
+	}
+
 	/* Set the needed headroom once and for all */
 	BUILD_BUG_ON(sizeof(struct bcm_tsb) != 8);
 	dev->needed_headroom += sizeof(struct bcm_tsb);
@@ -2095,22 +2578,45 @@ static int bcm_sysport_probe(struct platform_device *pdev)
 	/* libphy will adjust the link state accordingly */
 	netif_carrier_off(dev);
 
+	priv->rx_max_coalesced_frames = 1;
+	u64_stats_init(&priv->syncp);
+
+	priv->netdev_notifier.notifier_call = bcm_sysport_netdevice_event;
+
+	ret = register_netdevice_notifier(&priv->netdev_notifier);
+	if (ret) {
+		dev_err(&pdev->dev, "failed to register DSA notifier\n");
+		goto err_deregister_fixed_link;
+	}
+
 	ret = register_netdev(dev);
 	if (ret) {
 		dev_err(&pdev->dev, "failed to register net_device\n");
-		goto err_deregister_fixed_link;
+		goto err_deregister_notifier;
+	}
+
+	ret = clk_prepare_enable(priv->clk);
+	if (ret) {
+		dev_err(&pdev->dev, "could not enable priv clock\n");
+		goto err_deregister_netdev;
 	}
 
 	priv->rev = topctrl_readl(priv, REV_CNTL) & REV_MASK;
 	dev_info(&pdev->dev,
-		 "Broadcom SYSTEMPORT%s" REV_FMT
-		 " at 0x%p (irqs: %d, %d, TXQs: %d, RXQs: %d)\n",
+		 "Broadcom SYSTEMPORT%s " REV_FMT
+		 " (irqs: %d, %d, TXQs: %d, RXQs: %d)\n",
 		 priv->is_lite ? " Lite" : "",
 		 (priv->rev >> 8) & 0xff, priv->rev & 0xff,
-		 priv->base, priv->irq0, priv->irq1, txq, rxq);
+		 priv->irq0, priv->irq1, txq, rxq);
+
+	clk_disable_unprepare(priv->clk);
 
 	return 0;
 
+err_deregister_netdev:
+	unregister_netdev(dev);
+err_deregister_notifier:
+	unregister_netdevice_notifier(&priv->netdev_notifier);
 err_deregister_fixed_link:
 	if (of_phy_is_fixed_link(dn))
 		of_phy_deregister_fixed_link(dn);
@@ -2119,38 +2625,65 @@ err_free_netdev:
 	return ret;
 }
 
-static int bcm_sysport_remove(struct platform_device *pdev)
+static void bcm_sysport_remove(struct platform_device *pdev)
 {
 	struct net_device *dev = dev_get_drvdata(&pdev->dev);
+	struct bcm_sysport_priv *priv = netdev_priv(dev);
 	struct device_node *dn = pdev->dev.of_node;
 
 	/* Not much to do, ndo_close has been called
 	 * and we use managed allocations
 	 */
+	unregister_netdevice_notifier(&priv->netdev_notifier);
 	unregister_netdev(dev);
 	if (of_phy_is_fixed_link(dn))
 		of_phy_deregister_fixed_link(dn);
 	free_netdev(dev);
 	dev_set_drvdata(&pdev->dev, NULL);
-
-	return 0;
 }
 
-#ifdef CONFIG_PM_SLEEP
 static int bcm_sysport_suspend_to_wol(struct bcm_sysport_priv *priv)
 {
 	struct net_device *ndev = priv->netdev;
 	unsigned int timeout = 1000;
+	unsigned int index, i = 0;
 	u32 reg;
 
-	/* Password has already been programmed */
 	reg = umac_readl(priv, UMAC_MPD_CTRL);
-	reg |= MPD_EN;
+	if (priv->wolopts & (WAKE_MAGIC | WAKE_MAGICSECURE))
+		reg |= MPD_EN;
 	reg &= ~PSW_EN;
-	if (priv->wolopts & WAKE_MAGICSECURE)
+	if (priv->wolopts & WAKE_MAGICSECURE) {
+		/* Program the SecureOn password */
+		umac_writel(priv, get_unaligned_be16(&priv->sopass[0]),
+			    UMAC_PSW_MS);
+		umac_writel(priv, get_unaligned_be32(&priv->sopass[2]),
+			    UMAC_PSW_LS);
 		reg |= PSW_EN;
+	}
 	umac_writel(priv, reg, UMAC_MPD_CTRL);
 
+	if (priv->wolopts & WAKE_FILTER) {
+		/* Turn on ACPI matching to steal packets from RBUF */
+		reg = rbuf_readl(priv, RBUF_CONTROL);
+		if (priv->is_lite)
+			reg |= RBUF_ACPI_EN_LITE;
+		else
+			reg |= RBUF_ACPI_EN;
+		rbuf_writel(priv, reg, RBUF_CONTROL);
+
+		/* Enable RXCHK, active filters and Broadcom tag matching */
+		reg = rxchk_readl(priv, RXCHK_CONTROL);
+		reg &= ~(RXCHK_BRCM_TAG_MATCH_MASK <<
+			 RXCHK_BRCM_TAG_MATCH_SHIFT);
+		for_each_set_bit(index, priv->filters, RXCHK_BRCM_TAG_MAX) {
+			reg |= BIT(RXCHK_BRCM_TAG_MATCH_SHIFT + i);
+			i++;
+		}
+		reg |= RXCHK_EN | RXCHK_BRCM_TAG_EN;
+		rxchk_writel(priv, reg, RXCHK_CONTROL);
+	}
+
 	/* Make sure RBUF entered WoL mode as result */
 	do {
 		reg = rbuf_readl(priv, RBUF_STATUS);
@@ -2162,9 +2695,7 @@ static int bcm_sysport_suspend_to_wol(struct bcm_sysport_priv *priv)
 
 	/* Do not leave the UniMAC RBUF matching only MPD packets */
 	if (!timeout) {
-		reg = umac_readl(priv, UMAC_MPD_CTRL);
-		reg &= ~MPD_EN;
-		umac_writel(priv, reg, UMAC_MPD_CTRL);
+		mpd_enable_set(priv, false);
 		netif_err(priv, wol, ndev, "failed to enter WOL mode\n");
 		return -ETIMEDOUT;
 	}
@@ -2172,15 +2703,12 @@ static int bcm_sysport_suspend_to_wol(struct bcm_sysport_priv *priv)
 	/* UniMAC receive needs to be turned on */
 	umac_enable_set(priv, CMD_RX_EN, 1);
 
-	/* Enable the interrupt wake-up source */
-	intrl2_0_mask_clear(priv, INTRL2_0_MPD);
-
 	netif_dbg(priv, wol, ndev, "entered WOL mode\n");
 
 	return 0;
 }
 
-static int bcm_sysport_suspend(struct device *d)
+static int __maybe_unused bcm_sysport_suspend(struct device *d)
 {
 	struct net_device *dev = dev_get_drvdata(d);
 	struct bcm_sysport_priv *priv = netdev_priv(dev);
@@ -2191,12 +2719,12 @@ static int bcm_sysport_suspend(struct device *d)
 	if (!netif_running(dev))
 		return 0;
 
+	netif_device_detach(dev);
+
 	bcm_sysport_netif_stop(dev);
 
 	phy_suspend(dev->phydev);
 
-	netif_device_detach(dev);
-
 	/* Disable UniMAC RX */
 	umac_enable_set(priv, CMD_RX_EN, 0);
 
@@ -2236,25 +2764,40 @@ static int bcm_sysport_suspend(struct device *d)
 	bcm_sysport_fini_rx_ring(priv);
 
 	/* Get prepared for Wake-on-LAN */
-	if (device_may_wakeup(d) && priv->wolopts)
+	if (device_may_wakeup(d) && priv->wolopts) {
+		clk_prepare_enable(priv->wol_clk);
 		ret = bcm_sysport_suspend_to_wol(priv);
+	}
+
+	clk_disable_unprepare(priv->clk);
 
 	return ret;
 }
 
-static int bcm_sysport_resume(struct device *d)
+static int __maybe_unused bcm_sysport_resume(struct device *d)
 {
 	struct net_device *dev = dev_get_drvdata(d);
 	struct bcm_sysport_priv *priv = netdev_priv(dev);
 	unsigned int i;
-	u32 reg;
 	int ret;
 
 	if (!netif_running(dev))
 		return 0;
 
+	ret = clk_prepare_enable(priv->clk);
+	if (ret) {
+		netdev_err(dev, "could not enable priv clock\n");
+		return ret;
+	}
+
+	if (priv->wolopts)
+		clk_disable_unprepare(priv->wol_clk);
+
 	umac_reset(priv);
 
+	/* Disable the UniMAC RX/TX */
+	umac_enable_set(priv, CMD_RX_EN | CMD_TX_EN, 0);
+
 	/* We may have been suspended and never received a WOL event that
 	 * would turn off MPD detection, take care of that now
 	 */
@@ -2280,8 +2823,6 @@ static int bcm_sysport_resume(struct device *d)
 		goto out_free_rx_ring;
 	}
 
-	netif_device_attach(dev);
-
 	/* RX pipe enable */
 	topctrl_writel(priv, 0, RX_FLUSH_CNTL);
 
@@ -2291,12 +2832,8 @@ static int bcm_sysport_resume(struct device *d)
 		goto out_free_rx_ring;
 	}
 
-	/* Enable rxhck */
-	if (priv->rx_chk_en) {
-		reg = rxchk_readl(priv, RXCHK_CONTROL);
-		reg |= RXCHK_EN;
-		rxchk_writel(priv, reg, RXCHK_CONTROL);
-	}
+	/* Restore enabled features */
+	bcm_sysport_set_features(dev, dev->features);
 
 	rbuf_init(priv);
 
@@ -2326,6 +2863,8 @@ static int bcm_sysport_resume(struct device *d)
 
 	bcm_sysport_netif_start(dev);
 
+	netif_device_attach(dev);
+
 	return 0;
 
 out_free_rx_ring:
@@ -2333,16 +2872,16 @@ out_free_rx_ring:
 out_free_tx_rings:
 	for (i = 0; i < dev->num_tx_queues; i++)
 		bcm_sysport_fini_tx_ring(priv, i);
+	clk_disable_unprepare(priv->clk);
 	return ret;
 }
-#endif
 
 static SIMPLE_DEV_PM_OPS(bcm_sysport_pm_ops,
 		bcm_sysport_suspend, bcm_sysport_resume);
 
 static struct platform_driver bcm_sysport_driver = {
 	.probe	= bcm_sysport_probe,
-	.remove	= bcm_sysport_remove,
+	.remove = bcm_sysport_remove,
 	.driver =  {
 		.name = "brcm-systemport",
 		.of_match_table = bcm_sysport_of_match,
diff --git a/drivers/net/ethernet/broadcom/bcmsysport.h b/drivers/net/ethernet/broadcom/bcmsysport.h
index 77a51c167a69..a34296f989f1 100644
--- a/drivers/net/ethernet/broadcom/bcmsysport.h
+++ b/drivers/net/ethernet/broadcom/bcmsysport.h
@@ -1,17 +1,19 @@
+/* SPDX-License-Identifier: GPL-2.0-only */
 /*
  * Broadcom BCM7xxx System Port Ethernet MAC driver
  *
  * Copyright (C) 2014 Broadcom Corporation
- *
- * This program is free software; you can redistribute it and/or modify
- * it under the terms of the GNU General Public License version 2 as
- * published by the Free Software Foundation.
  */
 
 #ifndef __BCM_SYSPORT_H
 #define __BCM_SYSPORT_H
 
+#include <linux/bitmap.h>
+#include <linux/ethtool.h>
 #include <linux/if_vlan.h>
+#include <linux/dim.h>
+
+#include "unimac.h"
 
 /* Receive/transmit descriptor format */
 #define DESC_ADDR_HI_STATUS_LEN	0x00
@@ -154,14 +156,18 @@ struct bcm_rsb {
 #define  RXCHK_PARSE_AUTH		(1 << 22)
 
 #define RXCHK_BRCM_TAG0			0x04
-#define RXCHK_BRCM_TAG(i)		((i) * RXCHK_BRCM_TAG0)
+#define RXCHK_BRCM_TAG(i)		((i) * 0x4 + RXCHK_BRCM_TAG0)
 #define RXCHK_BRCM_TAG0_MASK		0x24
-#define RXCHK_BRCM_TAG_MASK(i)		((i) * RXCHK_BRCM_TAG0_MASK)
+#define RXCHK_BRCM_TAG_MASK(i)		((i) * 0x4 + RXCHK_BRCM_TAG0_MASK)
 #define RXCHK_BRCM_TAG_MATCH_STATUS	0x44
 #define RXCHK_ETHERTYPE			0x48
 #define RXCHK_BAD_CSUM_CNTR		0x4C
 #define RXCHK_OTHER_DISC_CNTR		0x50
 
+#define RXCHK_BRCM_TAG_MAX		8
+#define RXCHK_BRCM_TAG_CID_SHIFT	16
+#define RXCHK_BRCM_TAG_CID_MASK		0xff
+
 /* TXCHCK offsets and defines */
 #define SYS_PORT_TXCHK_OFFSET		0x380
 #define TXCHK_PKT_RDY_THRESH		0x00
@@ -184,6 +190,7 @@ struct bcm_rsb {
 #define  RBUF_RSB_SWAP0			(1 << 22)
 #define  RBUF_RSB_SWAP1			(1 << 23)
 #define  RBUF_ACPI_EN			(1 << 23)
+#define  RBUF_ACPI_EN_LITE		(1 << 24)
 
 #define RBUF_PKT_RDY_THRESH		0x04
 
@@ -208,39 +215,6 @@ struct bcm_rsb {
 /* UniMAC offset and defines */
 #define SYS_PORT_UMAC_OFFSET		0x800
 
-#define UMAC_CMD			0x008
-#define  CMD_TX_EN			(1 << 0)
-#define  CMD_RX_EN			(1 << 1)
-#define  CMD_SPEED_SHIFT		2
-#define  CMD_SPEED_10			0
-#define  CMD_SPEED_100			1
-#define  CMD_SPEED_1000			2
-#define  CMD_SPEED_2500			3
-#define  CMD_SPEED_MASK			3
-#define  CMD_PROMISC			(1 << 4)
-#define  CMD_PAD_EN			(1 << 5)
-#define  CMD_CRC_FWD			(1 << 6)
-#define  CMD_PAUSE_FWD			(1 << 7)
-#define  CMD_RX_PAUSE_IGNORE		(1 << 8)
-#define  CMD_TX_ADDR_INS		(1 << 9)
-#define  CMD_HD_EN			(1 << 10)
-#define  CMD_SW_RESET			(1 << 13)
-#define  CMD_LCL_LOOP_EN		(1 << 15)
-#define  CMD_AUTO_CONFIG		(1 << 22)
-#define  CMD_CNTL_FRM_EN		(1 << 23)
-#define  CMD_NO_LEN_CHK			(1 << 24)
-#define  CMD_RMT_LOOP_EN		(1 << 25)
-#define  CMD_PRBL_EN			(1 << 27)
-#define  CMD_TX_PAUSE_IGNORE		(1 << 28)
-#define  CMD_TX_RX_EN			(1 << 29)
-#define  CMD_RUNT_FILTER_DIS		(1 << 30)
-
-#define UMAC_MAC0			0x00c
-#define UMAC_MAC1			0x010
-#define UMAC_MAX_FRAME_LEN		0x014
-
-#define UMAC_TX_FLUSH			0x334
-
 #define UMAC_MIB_START			0x400
 
 /* There is a 0xC gap between the end of RX and beginning of TX stats and then
@@ -277,7 +251,8 @@ struct bcm_rsb {
 #define  GIB_GTX_CLK_EXT_CLK		(0 << GIB_GTX_CLK_SEL_SHIFT)
 #define  GIB_GTX_CLK_125MHZ		(1 << GIB_GTX_CLK_SEL_SHIFT)
 #define  GIB_GTX_CLK_250MHZ		(2 << GIB_GTX_CLK_SEL_SHIFT)
-#define  GIB_FCS_STRIP			(1 << 6)
+#define  GIB_FCS_STRIP_SHIFT		6
+#define  GIB_FCS_STRIP			(1 << GIB_FCS_STRIP_SHIFT)
 #define  GIB_LCL_LOOP_EN		(1 << 7)
 #define  GIB_LCL_LOOP_TXEN		(1 << 8)
 #define  GIB_RMT_LOOP_EN		(1 << 9)
@@ -315,6 +290,7 @@ struct bcm_rsb {
 
 #define RDMA_WRITE_PTR_HI		0x1010
 #define RDMA_WRITE_PTR_LO		0x1014
+#define RDMA_OVFL_DISC_CNTR		0x1018
 #define RDMA_PROD_INDEX			0x1018
 #define  RDMA_PROD_INDEX_MASK		0xffff
 
@@ -404,7 +380,7 @@ struct bcm_rsb {
 #define  RING_CONS_INDEX_MASK		0xffff
 
 #define RING_MAPPING			0x14
-#define  RING_QID_MASK			0x3
+#define  RING_QID_MASK			0x7
 #define  RING_PORT_ID_SHIFT		3
 #define  RING_PORT_ID_MASK		0x7
 #define  RING_IGNORE_STATUS		(1 << 6)
@@ -449,7 +425,8 @@ struct bcm_rsb {
 /* Uses 2 bits on SYSTEMPORT Lite and shifts everything by 1 bit, we
  * keep the SYSTEMPORT layout here and adjust with tdma_control_bit()
  */
-#define  TSB_SWAP			2
+#define  TSB_SWAP0			2
+#define  TSB_SWAP1			3
 #define  ACB_ALGO			3
 #define  BUF_DATA_OFFSET_SHIFT		4
 #define  BUF_DATA_OFFSET_MASK		0x3ff
@@ -506,21 +483,15 @@ struct bcm_rsb {
 
 #define TDMA_DEBUG			0x64c
 
-/* Transmit/Receive descriptor */
-struct dma_desc {
-	u32	addr_status_len;
-	u32	addr_lo;
-};
-
 /* Number of Receive hardware descriptor words */
 #define SP_NUM_HW_RX_DESC_WORDS		1024
-#define SP_LT_NUM_HW_RX_DESC_WORDS	256
+#define SP_LT_NUM_HW_RX_DESC_WORDS	512
 
 /* Internal linked-list RAM size */
 #define SP_NUM_TX_DESC			1536
 #define SP_LT_NUM_TX_DESC		256
 
-#define WORDS_PER_DESC			(sizeof(struct dma_desc) / sizeof(u32))
+#define WORDS_PER_DESC			2
 
 /* Rx/Tx common counter group.*/
 struct bcm_sysport_pkt_counters {
@@ -595,19 +566,24 @@ struct bcm_sysport_mib {
 	u32 rxchk_other_pkt_disc;
 	u32 rbuf_ovflow_cnt;
 	u32 rbuf_err_cnt;
+	u32 rdma_ovflow_cnt;
 	u32 alloc_rx_buff_failed;
 	u32 rx_dma_failed;
 	u32 tx_dma_failed;
+	u32 tx_realloc_tsb;
+	u32 tx_realloc_tsb_failed;
 };
 
 /* HW maintains a large list of counters */
 enum bcm_sysport_stat_type {
 	BCM_SYSPORT_STAT_NETDEV = -1,
+	BCM_SYSPORT_STAT_NETDEV64,
 	BCM_SYSPORT_STAT_MIB_RX,
 	BCM_SYSPORT_STAT_MIB_TX,
 	BCM_SYSPORT_STAT_RUNT,
 	BCM_SYSPORT_STAT_RXCHK,
 	BCM_SYSPORT_STAT_RBUF,
+	BCM_SYSPORT_STAT_RDMA,
 	BCM_SYSPORT_STAT_SOFT,
 };
 
@@ -619,6 +595,13 @@ enum bcm_sysport_stat_type {
 	.type = BCM_SYSPORT_STAT_NETDEV, \
 }
 
+#define STAT_NETDEV64(m) { \
+	.stat_string = __stringify(m), \
+	.stat_sizeof = sizeof(((struct bcm_sysport_stats64 *)0)->m), \
+	.stat_offset = offsetof(struct bcm_sysport_stats64, m), \
+	.type = BCM_SYSPORT_STAT_NETDEV64, \
+}
+
 #define STAT_MIB(str, m, _type) { \
 	.stat_string = str, \
 	.stat_sizeof = sizeof(((struct bcm_sysport_priv *)0)->m), \
@@ -647,6 +630,14 @@ enum bcm_sysport_stat_type {
 	.reg_offset = ofs, \
 }
 
+#define STAT_RDMA(str, m, ofs) { \
+	.stat_string = str, \
+	.stat_sizeof = sizeof(((struct bcm_sysport_priv *)0)->m), \
+	.stat_offset = offsetof(struct bcm_sysport_priv, m), \
+	.type = BCM_SYSPORT_STAT_RDMA, \
+	.reg_offset = ofs, \
+}
+
 /* TX bytes and packets */
 #define NUM_SYSPORT_TXQ_STAT	2
 
@@ -659,6 +650,14 @@ struct bcm_sysport_stats {
 	u16 reg_offset;
 };
 
+struct bcm_sysport_stats64 {
+	/* 64bit stats on 32bit/64bit Machine */
+	u64	rx_packets;
+	u64	rx_bytes;
+	u64	tx_packets;
+	u64	tx_bytes;
+};
+
 /* Software house keeping helper structure */
 struct bcm_sysport_cb {
 	struct sk_buff	*skb;		/* SKB for RX packets */
@@ -678,23 +677,32 @@ struct bcm_sysport_hw_params {
 	unsigned int	num_rx_desc_words;
 };
 
+struct bcm_sysport_net_dim {
+	u16			use_dim;
+	u16			event_ctr;
+	unsigned long		packets;
+	unsigned long		bytes;
+	struct dim		dim;
+};
+
 /* Software view of the TX ring */
 struct bcm_sysport_tx_ring {
 	spinlock_t	lock;		/* Ring lock for tx reclaim/xmit */
 	struct napi_struct napi;	/* NAPI per tx queue */
-	dma_addr_t	desc_dma;	/* DMA cookie */
 	unsigned int	index;		/* Ring index */
 	unsigned int	size;		/* Ring current size */
 	unsigned int	alloc_size;	/* Ring one-time allocated size */
 	unsigned int	desc_count;	/* Number of descriptors */
 	unsigned int	curr_desc;	/* Current descriptor */
 	unsigned int	c_index;	/* Last consumer index */
-	unsigned int	p_index;	/* Current producer index */
+	unsigned int	clean_index;	/* Current clean index */
 	struct bcm_sysport_cb *cbs;	/* Transmit control blocks */
-	struct dma_desc	*desc_cpu;	/* CPU view of the descriptor */
 	struct bcm_sysport_priv *priv;	/* private context backpointer */
 	unsigned long	packets;	/* packets statistics */
 	unsigned long	bytes;		/* bytes statistics */
+	unsigned int	switch_queue;	/* switch port queue number */
+	unsigned int	switch_port;	/* switch port queue number */
+	bool		inspect;	/* inspect switch port and queue */
 };
 
 /* Driver private structure */
@@ -714,6 +722,7 @@ struct bcm_sysport_priv {
 	int			wol_irq;
 
 	/* Transmit rings */
+	spinlock_t		desc_lock;
 	struct bcm_sysport_tx_ring *tx_rings;
 
 	/* Receive queue */
@@ -723,6 +732,10 @@ struct bcm_sysport_priv {
 	unsigned int		rx_read_ptr;
 	unsigned int		rx_c_index;
 
+	struct bcm_sysport_net_dim	dim;
+	u32			rx_max_coalesced_frames;
+	u32			rx_coalesce_usecs;
+
 	/* PHY device */
 	struct device_node	*phy_dn;
 	phy_interface_t		phy_interface;
@@ -736,12 +749,51 @@ struct bcm_sysport_priv {
 	unsigned int		crc_fwd:1;
 	u16			rev;
 	u32			wolopts;
+	u8			sopass[SOPASS_MAX];
 	unsigned int		wol_irq_disabled:1;
+	struct clk		*clk;
+	struct clk		*wol_clk;
 
 	/* MIB related fields */
 	struct bcm_sysport_mib	mib;
 
 	/* Ethtool */
 	u32			msg_enable;
+	DECLARE_BITMAP(filters, RXCHK_BRCM_TAG_MAX);
+	u32			filters_loc[RXCHK_BRCM_TAG_MAX];
+
+	struct bcm_sysport_stats64	stats64;
+
+	/* For atomic update generic 64bit value on 32bit Machine */
+	struct u64_stats_sync	syncp;
+
+	/* map information between switch port queues and local queues */
+	struct notifier_block	netdev_notifier;
+	unsigned int		per_port_num_tx_queues;
+	struct bcm_sysport_tx_ring *ring_map[DSA_MAX_PORTS * 8];
+
 };
+
+/* I/O accessors register helpers */
+#define BCM_SYSPORT_IO_MACRO(name, offset) \
+static inline u32 name##_readl(struct bcm_sysport_priv *priv, u32 off)	\
+{									\
+	u32 reg = readl_relaxed(priv->base + (offset) + off);		\
+	return reg;							\
+}									\
+static inline void name##_writel(struct bcm_sysport_priv *priv,		\
+				  u32 val, u32 off)			\
+{									\
+	writel_relaxed(val, priv->base + (offset) + off);		\
+}									\
+
+BCM_SYSPORT_IO_MACRO(intrl2_0, SYS_PORT_INTRL2_0_OFFSET);
+BCM_SYSPORT_IO_MACRO(intrl2_1, SYS_PORT_INTRL2_1_OFFSET);
+BCM_SYSPORT_IO_MACRO(umac, SYS_PORT_UMAC_OFFSET);
+BCM_SYSPORT_IO_MACRO(gib, SYS_PORT_GIB_OFFSET);
+BCM_SYSPORT_IO_MACRO(tdma, SYS_PORT_TDMA_OFFSET);
+BCM_SYSPORT_IO_MACRO(rxchk, SYS_PORT_RXCHK_OFFSET);
+BCM_SYSPORT_IO_MACRO(rbuf, SYS_PORT_RBUF_OFFSET);
+BCM_SYSPORT_IO_MACRO(topctrl, SYS_PORT_TOPCTRL_OFFSET);
+
 #endif /* __BCM_SYSPORT_H */
diff --git a/drivers/net/ethernet/broadcom/bgmac-bcma-mdio.c b/drivers/net/ethernet/broadcom/bgmac-bcma-mdio.c
index 6ce80cbcb48e..50b8e97a811d 100644
--- a/drivers/net/ethernet/broadcom/bgmac-bcma-mdio.c
+++ b/drivers/net/ethernet/broadcom/bgmac-bcma-mdio.c
@@ -10,6 +10,7 @@
 
 #include <linux/bcma/bcma.h>
 #include <linux/brcmphy.h>
+#include <linux/of_mdio.h>
 #include "bgmac.h"
 
 static bool bcma_mdio_wait_value(struct bcma_device *core, u16 reg, u32 mask,
@@ -211,6 +212,7 @@ struct mii_bus *bcma_mdio_mii_register(struct bgmac *bgmac)
 {
 	struct bcma_device *core = bgmac->bcma.core;
 	struct mii_bus *mii_bus;
+	struct device_node *np;
 	int err;
 
 	mii_bus = mdiobus_alloc();
@@ -229,7 +231,10 @@ struct mii_bus *bcma_mdio_mii_register(struct bgmac *bgmac)
 	mii_bus->parent = &core->dev;
 	mii_bus->phy_mask = ~(1 << bgmac->phyaddr);
 
-	err = mdiobus_register(mii_bus);
+	np = of_get_child_by_name(core->dev.of_node, "mdio");
+
+	err = of_mdiobus_register(mii_bus, np);
+	of_node_put(np);
 	if (err) {
 		dev_err(&core->dev, "Registration of mii bus failed\n");
 		goto err_free_bus;
@@ -255,4 +260,5 @@ void bcma_mdio_mii_unregister(struct mii_bus *mii_bus)
 EXPORT_SYMBOL_GPL(bcma_mdio_mii_unregister);
 
 MODULE_AUTHOR("Rafał Miłecki");
+MODULE_DESCRIPTION("Broadcom iProc GBit BCMA MDIO helpers");
 MODULE_LICENSE("GPL");
diff --git a/drivers/net/ethernet/broadcom/bgmac-bcma.c b/drivers/net/ethernet/broadcom/bgmac-bcma.c
index 6322594ab260..36f9bad28e6a 100644
--- a/drivers/net/ethernet/broadcom/bgmac-bcma.c
+++ b/drivers/net/ethernet/broadcom/bgmac-bcma.c
@@ -11,6 +11,7 @@
 #include <linux/bcma/bcma.h>
 #include <linux/brcmphy.h>
 #include <linux/etherdevice.h>
+#include <linux/of_mdio.h>
 #include <linux/of_net.h>
 #include "bgmac.h"
 
@@ -86,17 +87,28 @@ static int bcma_phy_connect(struct bgmac *bgmac)
 	struct phy_device *phy_dev;
 	char bus_id[MII_BUS_ID_SIZE + 3];
 
+	/* DT info should be the most accurate */
+	phy_dev = of_phy_get_and_connect(bgmac->net_dev, bgmac->dev->of_node,
+					 bgmac_adjust_link);
+	if (phy_dev)
+		return 0;
+
 	/* Connect to the PHY */
-	snprintf(bus_id, sizeof(bus_id), PHY_ID_FMT, bgmac->mii_bus->id,
-		 bgmac->phyaddr);
-	phy_dev = phy_connect(bgmac->net_dev, bus_id, bgmac_adjust_link,
-			      PHY_INTERFACE_MODE_MII);
-	if (IS_ERR(phy_dev)) {
-		dev_err(bgmac->dev, "PHY connection failed\n");
-		return PTR_ERR(phy_dev);
+	if (bgmac->mii_bus && bgmac->phyaddr != BGMAC_PHY_NOREGS) {
+		snprintf(bus_id, sizeof(bus_id), PHY_ID_FMT, bgmac->mii_bus->id,
+			 bgmac->phyaddr);
+		phy_dev = phy_connect(bgmac->net_dev, bus_id, bgmac_adjust_link,
+				      PHY_INTERFACE_MODE_MII);
+		if (IS_ERR(phy_dev)) {
+			dev_err(bgmac->dev, "PHY connection failed\n");
+			return PTR_ERR(phy_dev);
+		}
+
+		return 0;
 	}
 
-	return 0;
+	/* Assume a fixed link to the switch port */
+	return bgmac_phy_connect_direct(bgmac);
 }
 
 static const struct bcma_device_id bgmac_bcma_tbl[] = {
@@ -115,7 +127,7 @@ static int bgmac_probe(struct bcma_device *core)
 	struct ssb_sprom *sprom = &core->bus->sprom;
 	struct mii_bus *mii_bus;
 	struct bgmac *bgmac;
-	const u8 *mac = NULL;
+	const u8 *mac;
 	int err;
 
 	bgmac = bgmac_alloc(&core->dev);
@@ -128,11 +140,12 @@ static int bgmac_probe(struct bcma_device *core)
 
 	bcma_set_drvdata(core, bgmac);
 
-	if (bgmac->dev->of_node)
-		mac = of_get_mac_address(bgmac->dev->of_node);
+	err = of_get_ethdev_address(bgmac->dev->of_node, bgmac->net_dev);
+	if (err == -EPROBE_DEFER)
+		return err;
 
 	/* If no MAC address assigned via device tree, check SPROM */
-	if (!mac) {
+	if (err) {
 		switch (core->core_unit) {
 		case 0:
 			mac = sprom->et0mac;
@@ -149,10 +162,9 @@ static int bgmac_probe(struct bcma_device *core)
 			err = -ENOTSUPP;
 			goto err;
 		}
+		eth_hw_addr_set(bgmac->net_dev, mac);
 	}
 
-	ether_addr_copy(bgmac->net_dev->dev_addr, mac);
-
 	/* On BCM4706 we need common core to access PHY */
 	if (core->id.id == BCMA_CORE_4706_MAC_GBIT &&
 	    !core->bus->drv_gmac_cmn.core) {
@@ -184,13 +196,19 @@ static int bgmac_probe(struct bcma_device *core)
 
 	if (!bgmac_is_bcm4707_family(core) &&
 	    !(ci->id == BCMA_CHIP_ID_BCM53573 && core->core_unit == 1)) {
+		struct phy_device *phydev;
+
 		mii_bus = bcma_mdio_mii_register(bgmac);
 		if (IS_ERR(mii_bus)) {
 			err = PTR_ERR(mii_bus);
 			goto err;
 		}
-
 		bgmac->mii_bus = mii_bus;
+
+		phydev = mdiobus_get_phy(bgmac->mii_bus, bgmac->phyaddr);
+		if (ci->id == BCMA_CHIP_ID_BCM53573 && phydev &&
+		    (phydev->drv->phy_id & phydev->drv->phy_id_mask) == PHY_ID_BCM54210E)
+			phydev->dev_flags |= PHY_BRCM_EN_MASTER_MODE;
 	}
 
 	if (core->bus->hosttype == BCMA_HOSTTYPE_PCI) {
@@ -211,7 +229,7 @@ static int bgmac_probe(struct bcma_device *core)
 	/* BCM 471X/535X family */
 	case BCMA_CHIP_ID_BCM4716:
 		bgmac->feature_flags |= BGMAC_FEAT_CLKCTLST;
-		/* fallthrough */
+		fallthrough;
 	case BCMA_CHIP_ID_BCM47162:
 		bgmac->feature_flags |= BGMAC_FEAT_FLW_CTRL2;
 		bgmac->feature_flags |= BGMAC_FEAT_SET_RXQ_CLK;
@@ -222,12 +240,12 @@ static int bgmac_probe(struct bcma_device *core)
 		bgmac->feature_flags |= BGMAC_FEAT_CLKCTLST;
 		bgmac->feature_flags |= BGMAC_FEAT_FLW_CTRL1;
 		bgmac->feature_flags |= BGMAC_FEAT_SW_TYPE_PHY;
-		if (ci->pkg == BCMA_PKG_ID_BCM47188 ||
-		    ci->pkg == BCMA_PKG_ID_BCM47186) {
+		if ((ci->id == BCMA_CHIP_ID_BCM5357 && ci->pkg == BCMA_PKG_ID_BCM47186) ||
+		    (ci->id == BCMA_CHIP_ID_BCM53572 && ci->pkg == BCMA_PKG_ID_BCM47188)) {
 			bgmac->feature_flags |= BGMAC_FEAT_SW_TYPE_RGMII;
 			bgmac->feature_flags |= BGMAC_FEAT_IOST_ATTACHED;
 		}
-		if (ci->pkg == BCMA_PKG_ID_BCM5358)
+		if (ci->id == BCMA_CHIP_ID_BCM5357 && ci->pkg == BCMA_PKG_ID_BCM5358)
 			bgmac->feature_flags |= BGMAC_FEAT_SW_TYPE_EPHYRMII;
 		break;
 	case BCMA_CHIP_ID_BCM53573:
@@ -291,10 +309,7 @@ static int bgmac_probe(struct bcma_device *core)
 	bgmac->cco_ctl_maskset = bcma_bgmac_cco_ctl_maskset;
 	bgmac->get_bus_clock = bcma_bgmac_get_bus_clock;
 	bgmac->cmn_maskset32 = bcma_bgmac_cmn_maskset32;
-	if (bgmac->mii_bus)
-		bgmac->phy_connect = bcma_phy_connect;
-	else
-		bgmac->phy_connect = bgmac_phy_connect_direct;
+	bgmac->phy_connect = bcma_phy_connect;
 
 	err = bgmac_enet_probe(bgmac);
 	if (err)
@@ -317,7 +332,6 @@ static void bgmac_remove(struct bcma_device *core)
 	bcma_mdio_mii_unregister(bgmac->mii_bus);
 	bgmac_enet_remove(bgmac);
 	bcma_set_drvdata(core, NULL);
-	kfree(bgmac);
 }
 
 static struct bcma_driver bgmac_bcma_driver = {
@@ -348,4 +362,5 @@ module_init(bgmac_init)
 module_exit(bgmac_exit)
 
 MODULE_AUTHOR("Rafał Miłecki");
+MODULE_DESCRIPTION("Broadcom iProc GBit BCMA interface driver");
 MODULE_LICENSE("GPL");
diff --git a/drivers/net/ethernet/broadcom/bgmac-platform.c b/drivers/net/ethernet/broadcom/bgmac-platform.c
index 73aca97a96bc..4e266ce41180 100644
--- a/drivers/net/ethernet/broadcom/bgmac-platform.c
+++ b/drivers/net/ethernet/broadcom/bgmac-platform.c
@@ -50,11 +50,14 @@ static u32 platform_bgmac_idm_read(struct bgmac *bgmac, u16 offset)
 
 static void platform_bgmac_idm_write(struct bgmac *bgmac, u16 offset, u32 value)
 {
-	return writel(value, bgmac->plat.idm_base + offset);
+	writel(value, bgmac->plat.idm_base + offset);
 }
 
 static bool platform_bgmac_clk_enabled(struct bgmac *bgmac)
 {
+	if (!bgmac->plat.idm_base)
+		return true;
+
 	if ((bgmac_idm_read(bgmac, BCMA_IOCTL) & BGMAC_CLK_EN) != BGMAC_CLK_EN)
 		return false;
 	if (bgmac_idm_read(bgmac, BCMA_RESET_CTL) & BCMA_RESET_CTL_RESET)
@@ -66,6 +69,9 @@ static void platform_bgmac_clk_enable(struct bgmac *bgmac, u32 flags)
 {
 	u32 val;
 
+	if (!bgmac->plat.idm_base)
+		return;
+
 	/* The Reset Control register only contains a single bit to show if the
 	 * controller is currently in reset.  Do a sanity check here, just in
 	 * case the bootloader happened to leave the device in reset.
@@ -125,6 +131,7 @@ static void bgmac_nicpm_speed_set(struct net_device *net_dev)
 	switch (bgmac->net_dev->phydev->speed) {
 	default:
 		netdev_err(net_dev, "Unsupported speed. Defaulting to 1000Mb\n");
+		fallthrough;
 	case SPEED_1000:
 		val |= NICPM_IOMUX_CTRL_SPD_1000M << NICPM_IOMUX_CTRL_SPD_SHIFT;
 		break;
@@ -164,9 +171,10 @@ static int platform_phy_connect(struct bgmac *bgmac)
 static int bgmac_probe(struct platform_device *pdev)
 {
 	struct device_node *np = pdev->dev.of_node;
+	struct device_node *phy_node;
 	struct bgmac *bgmac;
 	struct resource *regs;
-	const u8 *mac_addr;
+	int ret;
 
 	bgmac = bgmac_alloc(&pdev->dev);
 	if (!bgmac)
@@ -180,42 +188,38 @@ static int bgmac_probe(struct platform_device *pdev)
 	bgmac->feature_flags |= BGMAC_FEAT_CMDCFG_SR_REV4;
 	bgmac->feature_flags |= BGMAC_FEAT_TX_MASK_SETUP;
 	bgmac->feature_flags |= BGMAC_FEAT_RX_MASK_SETUP;
+	bgmac->feature_flags |= BGMAC_FEAT_IDM_MASK;
 
 	bgmac->dev = &pdev->dev;
 	bgmac->dma_dev = &pdev->dev;
 
-	mac_addr = of_get_mac_address(np);
-	if (mac_addr)
-		ether_addr_copy(bgmac->net_dev->dev_addr, mac_addr);
-	else
-		dev_warn(&pdev->dev, "MAC address not present in device tree\n");
+	ret = of_get_ethdev_address(np, bgmac->net_dev);
+	if (ret == -EPROBE_DEFER)
+		return ret;
+
+	if (ret)
+		dev_warn(&pdev->dev,
+			 "MAC address not present in device tree\n");
 
 	bgmac->irq = platform_get_irq(pdev, 0);
-	if (bgmac->irq < 0) {
-		dev_err(&pdev->dev, "Unable to obtain IRQ\n");
+	if (bgmac->irq < 0)
 		return bgmac->irq;
-	}
 
-	regs = platform_get_resource_byname(pdev, IORESOURCE_MEM, "amac_base");
-	if (!regs) {
-		dev_err(&pdev->dev, "Unable to obtain base resource\n");
-		return -EINVAL;
-	}
-
-	bgmac->plat.base = devm_ioremap_resource(&pdev->dev, regs);
+	bgmac->plat.base =
+		devm_platform_ioremap_resource_byname(pdev, "amac_base");
 	if (IS_ERR(bgmac->plat.base))
 		return PTR_ERR(bgmac->plat.base);
 
+	/* The idm_base resource is optional for some platforms */
 	regs = platform_get_resource_byname(pdev, IORESOURCE_MEM, "idm_base");
-	if (!regs) {
-		dev_err(&pdev->dev, "Unable to obtain idm resource\n");
-		return -EINVAL;
+	if (regs) {
+		bgmac->plat.idm_base = devm_ioremap_resource(&pdev->dev, regs);
+		if (IS_ERR(bgmac->plat.idm_base))
+			return PTR_ERR(bgmac->plat.idm_base);
+		bgmac->feature_flags &= ~BGMAC_FEAT_IDM_MASK;
 	}
 
-	bgmac->plat.idm_base = devm_ioremap_resource(&pdev->dev, regs);
-	if (IS_ERR(bgmac->plat.idm_base))
-		return PTR_ERR(bgmac->plat.idm_base);
-
+	/* The nicpm_base resource is optional for some platforms */
 	regs = platform_get_resource_byname(pdev, IORESOURCE_MEM, "nicpm_base");
 	if (regs) {
 		bgmac->plat.nicpm_base = devm_ioremap_resource(&pdev->dev,
@@ -233,7 +237,9 @@ static int bgmac_probe(struct platform_device *pdev)
 	bgmac->cco_ctl_maskset = platform_bgmac_cco_ctl_maskset;
 	bgmac->get_bus_clock = platform_bgmac_get_bus_clock;
 	bgmac->cmn_maskset32 = platform_bgmac_cmn_maskset32;
-	if (of_parse_phandle(np, "phy-handle", 0)) {
+	phy_node = of_parse_phandle(np, "phy-handle", 0);
+	if (phy_node) {
+		of_node_put(phy_node);
 		bgmac->phy_connect = platform_phy_connect;
 	} else {
 		bgmac->phy_connect = bgmac_phy_connect_direct;
@@ -243,13 +249,11 @@ static int bgmac_probe(struct platform_device *pdev)
 	return bgmac_enet_probe(bgmac);
 }
 
-static int bgmac_remove(struct platform_device *pdev)
+static void bgmac_remove(struct platform_device *pdev)
 {
 	struct bgmac *bgmac = platform_get_drvdata(pdev);
 
 	bgmac_enet_remove(bgmac);
-
-	return 0;
 }
 
 #ifdef CONFIG_PM
@@ -297,4 +301,5 @@ static struct platform_driver bgmac_enet_driver = {
 };
 
 module_platform_driver(bgmac_enet_driver);
+MODULE_DESCRIPTION("Broadcom iProc GBit platform interface driver");
 MODULE_LICENSE("GPL");
diff --git a/drivers/net/ethernet/broadcom/bgmac.c b/drivers/net/ethernet/broadcom/bgmac.c
index ba4d2e145bb9..3e9c57196a39 100644
--- a/drivers/net/ethernet/broadcom/bgmac.c
+++ b/drivers/net/ethernet/broadcom/bgmac.c
@@ -15,6 +15,7 @@
 #include <linux/bcm47xx_nvram.h>
 #include <linux/phy.h>
 #include <linux/phy_fixed.h>
+#include <net/dsa.h>
 #include "bgmac.h"
 
 static bool bgmac_wait_value(struct bgmac *bgmac, u16 reg, u32 mask,
@@ -171,7 +172,7 @@ static netdev_tx_t bgmac_dma_tx_add(struct bgmac *bgmac,
 	flags = 0;
 
 	for (i = 0; i < nr_frags; i++) {
-		struct skb_frag_struct *frag = &skb_shinfo(skb)->frags[i];
+		skb_frag_t *frag = &skb_shinfo(skb)->frags[i];
 		int len = skb_frag_size(frag);
 
 		index = (index + 1) % BGMAC_TX_RING_SLOTS;
@@ -188,8 +189,8 @@ static netdev_tx_t bgmac_dma_tx_add(struct bgmac *bgmac,
 	}
 
 	slot->skb = skb;
-	ring->end += nr_frags + 1;
 	netdev_sent_queue(net_dev, skb->len);
+	ring->end += nr_frags + 1;
 
 	wmb();
 
@@ -235,7 +236,6 @@ static void bgmac_dma_tx_free(struct bgmac *bgmac, struct bgmac_dma_ring *ring)
 {
 	struct device *dma_dev = bgmac->dma_dev;
 	int empty_slot;
-	bool freed = false;
 	unsigned bytes_compl = 0, pkts_compl = 0;
 
 	/* The last slot that hardware didn't consume yet */
@@ -278,7 +278,6 @@ static void bgmac_dma_tx_free(struct bgmac *bgmac, struct bgmac_dma_ring *ring)
 
 		slot->dma_addr = 0;
 		ring->start++;
-		freed = true;
 	}
 
 	if (!pkts_compl)
@@ -532,7 +531,8 @@ static void bgmac_dma_tx_ring_free(struct bgmac *bgmac,
 	int i;
 
 	for (i = 0; i < BGMAC_TX_RING_SLOTS; i++) {
-		int len = dma_desc[i].ctl1 & BGMAC_DESC_CTL1_LEN;
+		u32 ctl1 = le32_to_cpu(dma_desc[i].ctl1);
+		unsigned int len = ctl1 & BGMAC_DESC_CTL1_LEN;
 
 		slot = &ring->slots[i];
 		dev_kfree_skb(slot->skb);
@@ -616,15 +616,16 @@ static int bgmac_dma_alloc(struct bgmac *bgmac)
 	static const u16 ring_base[] = { BGMAC_DMA_BASE0, BGMAC_DMA_BASE1,
 					 BGMAC_DMA_BASE2, BGMAC_DMA_BASE3, };
 	int size; /* ring size: different for Tx and Rx */
-	int err;
 	int i;
 
 	BUILD_BUG_ON(BGMAC_MAX_TX_RINGS > ARRAY_SIZE(ring_base));
 	BUILD_BUG_ON(BGMAC_MAX_RX_RINGS > ARRAY_SIZE(ring_base));
 
-	if (!(bgmac_idm_read(bgmac, BCMA_IOST) & BCMA_IOST_DMA64)) {
-		dev_err(bgmac->dev, "Core does not report 64-bit DMA\n");
-		return -ENOTSUPP;
+	if (!(bgmac->feature_flags & BGMAC_FEAT_IDM_MASK)) {
+		if (!(bgmac_idm_read(bgmac, BCMA_IOST) & BCMA_IOST_DMA64)) {
+			dev_err(bgmac->dev, "Core does not report 64-bit DMA\n");
+			return -ENOTSUPP;
+		}
 	}
 
 	for (i = 0; i < BGMAC_MAX_TX_RINGS; i++) {
@@ -633,9 +634,9 @@ static int bgmac_dma_alloc(struct bgmac *bgmac)
 
 		/* Alloc ring of descriptors */
 		size = BGMAC_TX_RING_SLOTS * sizeof(struct bgmac_dma_desc);
-		ring->cpu_base = dma_zalloc_coherent(dma_dev, size,
-						     &ring->dma_base,
-						     GFP_KERNEL);
+		ring->cpu_base = dma_alloc_coherent(dma_dev, size,
+						    &ring->dma_base,
+						    GFP_KERNEL);
 		if (!ring->cpu_base) {
 			dev_err(bgmac->dev, "Allocation of TX ring 0x%X failed\n",
 				ring->mmio_base);
@@ -658,13 +659,12 @@ static int bgmac_dma_alloc(struct bgmac *bgmac)
 
 		/* Alloc ring of descriptors */
 		size = BGMAC_RX_RING_SLOTS * sizeof(struct bgmac_dma_desc);
-		ring->cpu_base = dma_zalloc_coherent(dma_dev, size,
-						     &ring->dma_base,
-						     GFP_KERNEL);
+		ring->cpu_base = dma_alloc_coherent(dma_dev, size,
+						    &ring->dma_base,
+						    GFP_KERNEL);
 		if (!ring->cpu_base) {
 			dev_err(bgmac->dev, "Allocation of RX ring 0x%X failed\n",
 				ring->mmio_base);
-			err = -ENOMEM;
 			goto err_dma_free;
 		}
 
@@ -746,36 +746,36 @@ error:
 /* TODO: can we just drop @force? Can we don't reset MAC at all if there is
  * nothing to change? Try if after stabilizng driver.
  */
-static void bgmac_cmdcfg_maskset(struct bgmac *bgmac, u32 mask, u32 set,
-				 bool force)
+static void bgmac_umac_cmd_maskset(struct bgmac *bgmac, u32 mask, u32 set,
+				   bool force)
 {
-	u32 cmdcfg = bgmac_read(bgmac, BGMAC_CMDCFG);
+	u32 cmdcfg = bgmac_umac_read(bgmac, UMAC_CMD);
 	u32 new_val = (cmdcfg & mask) | set;
 	u32 cmdcfg_sr;
 
 	if (bgmac->feature_flags & BGMAC_FEAT_CMDCFG_SR_REV4)
-		cmdcfg_sr = BGMAC_CMDCFG_SR_REV4;
+		cmdcfg_sr = CMD_SW_RESET;
 	else
-		cmdcfg_sr = BGMAC_CMDCFG_SR_REV0;
+		cmdcfg_sr = CMD_SW_RESET_OLD;
 
-	bgmac_set(bgmac, BGMAC_CMDCFG, cmdcfg_sr);
+	bgmac_umac_maskset(bgmac, UMAC_CMD, ~0, cmdcfg_sr);
 	udelay(2);
 
 	if (new_val != cmdcfg || force)
-		bgmac_write(bgmac, BGMAC_CMDCFG, new_val);
+		bgmac_umac_write(bgmac, UMAC_CMD, new_val);
 
-	bgmac_mask(bgmac, BGMAC_CMDCFG, ~cmdcfg_sr);
+	bgmac_umac_maskset(bgmac, UMAC_CMD, ~cmdcfg_sr, 0);
 	udelay(2);
 }
 
-static void bgmac_write_mac_address(struct bgmac *bgmac, u8 *addr)
+static void bgmac_write_mac_address(struct bgmac *bgmac, const u8 *addr)
 {
 	u32 tmp;
 
 	tmp = (addr[0] << 24) | (addr[1] << 16) | (addr[2] << 8) | addr[3];
-	bgmac_write(bgmac, BGMAC_MACADDR_HIGH, tmp);
+	bgmac_umac_write(bgmac, UMAC_MAC0, tmp);
 	tmp = (addr[4] << 8) | addr[5];
-	bgmac_write(bgmac, BGMAC_MACADDR_LOW, tmp);
+	bgmac_umac_write(bgmac, UMAC_MAC1, tmp);
 }
 
 static void bgmac_set_rx_mode(struct net_device *net_dev)
@@ -783,9 +783,9 @@ static void bgmac_set_rx_mode(struct net_device *net_dev)
 	struct bgmac *bgmac = netdev_priv(net_dev);
 
 	if (net_dev->flags & IFF_PROMISC)
-		bgmac_cmdcfg_maskset(bgmac, ~0, BGMAC_CMDCFG_PROM, true);
+		bgmac_umac_cmd_maskset(bgmac, ~0, CMD_PROMISC, true);
 	else
-		bgmac_cmdcfg_maskset(bgmac, ~BGMAC_CMDCFG_PROM, 0, true);
+		bgmac_umac_cmd_maskset(bgmac, ~CMD_PROMISC, 0, true);
 }
 
 #if 0 /* We don't use that regs yet */
@@ -825,21 +825,21 @@ static void bgmac_clear_mib(struct bgmac *bgmac)
 /* http://bcm-v4.sipsolutions.net/mac-gbit/gmac/gmac_speed */
 static void bgmac_mac_speed(struct bgmac *bgmac)
 {
-	u32 mask = ~(BGMAC_CMDCFG_ES_MASK | BGMAC_CMDCFG_HD);
+	u32 mask = ~(CMD_SPEED_MASK << CMD_SPEED_SHIFT | CMD_HD_EN);
 	u32 set = 0;
 
 	switch (bgmac->mac_speed) {
 	case SPEED_10:
-		set |= BGMAC_CMDCFG_ES_10;
+		set |= CMD_SPEED_10 << CMD_SPEED_SHIFT;
 		break;
 	case SPEED_100:
-		set |= BGMAC_CMDCFG_ES_100;
+		set |= CMD_SPEED_100 << CMD_SPEED_SHIFT;
 		break;
 	case SPEED_1000:
-		set |= BGMAC_CMDCFG_ES_1000;
+		set |= CMD_SPEED_1000 << CMD_SPEED_SHIFT;
 		break;
 	case SPEED_2500:
-		set |= BGMAC_CMDCFG_ES_2500;
+		set |= CMD_SPEED_2500 << CMD_SPEED_SHIFT;
 		break;
 	default:
 		dev_err(bgmac->dev, "Unsupported speed: %d\n",
@@ -847,17 +847,19 @@ static void bgmac_mac_speed(struct bgmac *bgmac)
 	}
 
 	if (bgmac->mac_duplex == DUPLEX_HALF)
-		set |= BGMAC_CMDCFG_HD;
+		set |= CMD_HD_EN;
 
-	bgmac_cmdcfg_maskset(bgmac, mask, set, true);
+	bgmac_umac_cmd_maskset(bgmac, mask, set, true);
 }
 
 static void bgmac_miiconfig(struct bgmac *bgmac)
 {
 	if (bgmac->feature_flags & BGMAC_FEAT_FORCE_SPEED_2500) {
-		bgmac_idm_write(bgmac, BCMA_IOCTL,
-				bgmac_idm_read(bgmac, BCMA_IOCTL) | 0x40 |
-				BGMAC_BCMA_IOCTL_SW_CLKEN);
+		if (!(bgmac->feature_flags & BGMAC_FEAT_IDM_MASK)) {
+			bgmac_idm_write(bgmac, BCMA_IOCTL,
+					bgmac_idm_read(bgmac, BCMA_IOCTL) |
+					0x40 | BGMAC_BCMA_IOCTL_SW_CLKEN);
+		}
 		bgmac->mac_speed = SPEED_2500;
 		bgmac->mac_duplex = DUPLEX_FULL;
 		bgmac_mac_speed(bgmac);
@@ -874,11 +876,36 @@ static void bgmac_miiconfig(struct bgmac *bgmac)
 	}
 }
 
+static void bgmac_chip_reset_idm_config(struct bgmac *bgmac)
+{
+	u32 iost;
+
+	iost = bgmac_idm_read(bgmac, BCMA_IOST);
+	if (bgmac->feature_flags & BGMAC_FEAT_IOST_ATTACHED)
+		iost &= ~BGMAC_BCMA_IOST_ATTACHED;
+
+	/* 3GMAC: for BCM4707 & BCM47094, only do core reset at bgmac_probe() */
+	if (!(bgmac->feature_flags & BGMAC_FEAT_NO_RESET)) {
+		u32 flags = 0;
+
+		if (iost & BGMAC_BCMA_IOST_ATTACHED) {
+			flags = BGMAC_BCMA_IOCTL_SW_CLKEN;
+			if (bgmac->in_init || !bgmac->has_robosw)
+				flags |= BGMAC_BCMA_IOCTL_SW_RESET;
+		}
+		bgmac_clk_enable(bgmac, flags);
+	}
+
+	if (iost & BGMAC_BCMA_IOST_ATTACHED && (bgmac->in_init || !bgmac->has_robosw))
+		bgmac_idm_write(bgmac, BCMA_IOCTL,
+				bgmac_idm_read(bgmac, BCMA_IOCTL) &
+				~BGMAC_BCMA_IOCTL_SW_RESET);
+}
+
 /* http://bcm-v4.sipsolutions.net/mac-gbit/gmac/chipreset */
 static void bgmac_chip_reset(struct bgmac *bgmac)
 {
 	u32 cmdcfg_sr;
-	u32 iost;
 	int i;
 
 	if (bgmac_clk_enabled(bgmac)) {
@@ -890,7 +917,7 @@ static void bgmac_chip_reset(struct bgmac *bgmac)
 		for (i = 0; i < BGMAC_MAX_TX_RINGS; i++)
 			bgmac_dma_tx_reset(bgmac, &bgmac->tx_ring[i]);
 
-		bgmac_cmdcfg_maskset(bgmac, ~0, BGMAC_CMDCFG_ML, false);
+		bgmac_umac_cmd_maskset(bgmac, ~0, CMD_LCL_LOOP_EN, false);
 		udelay(1);
 
 		for (i = 0; i < BGMAC_MAX_RX_RINGS; i++)
@@ -899,20 +926,8 @@ static void bgmac_chip_reset(struct bgmac *bgmac)
 		/* TODO: Clear software multicast filter list */
 	}
 
-	iost = bgmac_idm_read(bgmac, BCMA_IOST);
-	if (bgmac->feature_flags & BGMAC_FEAT_IOST_ATTACHED)
-		iost &= ~BGMAC_BCMA_IOST_ATTACHED;
-
-	/* 3GMAC: for BCM4707 & BCM47094, only do core reset at bgmac_probe() */
-	if (!(bgmac->feature_flags & BGMAC_FEAT_NO_RESET)) {
-		u32 flags = 0;
-		if (iost & BGMAC_BCMA_IOST_ATTACHED) {
-			flags = BGMAC_BCMA_IOCTL_SW_CLKEN;
-			if (!bgmac->has_robosw)
-				flags |= BGMAC_BCMA_IOCTL_SW_RESET;
-		}
-		bgmac_clk_enable(bgmac, flags);
-	}
+	if (!(bgmac->feature_flags & BGMAC_FEAT_IDM_MASK))
+		bgmac_chip_reset_idm_config(bgmac);
 
 	/* Request Misc PLL for corerev > 2 */
 	if (bgmac->feature_flags & BGMAC_FEAT_MISC_PLL_REQ) {
@@ -970,40 +985,35 @@ static void bgmac_chip_reset(struct bgmac *bgmac)
 				      BGMAC_CHIPCTL_7_IF_TYPE_RGMII);
 	}
 
-	if (iost & BGMAC_BCMA_IOST_ATTACHED && !bgmac->has_robosw)
-		bgmac_idm_write(bgmac, BCMA_IOCTL,
-				bgmac_idm_read(bgmac, BCMA_IOCTL) &
-				~BGMAC_BCMA_IOCTL_SW_RESET);
-
 	/* http://bcm-v4.sipsolutions.net/mac-gbit/gmac/gmac_reset
-	 * Specs don't say about using BGMAC_CMDCFG_SR, but in this routine
-	 * BGMAC_CMDCFG is read _after_ putting chip in a reset. So it has to
+	 * Specs don't say about using UMAC_CMD_SR, but in this routine
+	 * UMAC_CMD is read _after_ putting chip in a reset. So it has to
 	 * be keps until taking MAC out of the reset.
 	 */
 	if (bgmac->feature_flags & BGMAC_FEAT_CMDCFG_SR_REV4)
-		cmdcfg_sr = BGMAC_CMDCFG_SR_REV4;
+		cmdcfg_sr = CMD_SW_RESET;
 	else
-		cmdcfg_sr = BGMAC_CMDCFG_SR_REV0;
-
-	bgmac_cmdcfg_maskset(bgmac,
-			     ~(BGMAC_CMDCFG_TE |
-			       BGMAC_CMDCFG_RE |
-			       BGMAC_CMDCFG_RPI |
-			       BGMAC_CMDCFG_TAI |
-			       BGMAC_CMDCFG_HD |
-			       BGMAC_CMDCFG_ML |
-			       BGMAC_CMDCFG_CFE |
-			       BGMAC_CMDCFG_RL |
-			       BGMAC_CMDCFG_RED |
-			       BGMAC_CMDCFG_PE |
-			       BGMAC_CMDCFG_TPI |
-			       BGMAC_CMDCFG_PAD_EN |
-			       BGMAC_CMDCFG_PF),
-			     BGMAC_CMDCFG_PROM |
-			     BGMAC_CMDCFG_NLC |
-			     BGMAC_CMDCFG_CFE |
-			     cmdcfg_sr,
-			     false);
+		cmdcfg_sr = CMD_SW_RESET_OLD;
+
+	bgmac_umac_cmd_maskset(bgmac,
+			       ~(CMD_TX_EN |
+				 CMD_RX_EN |
+				 CMD_RX_PAUSE_IGNORE |
+				 CMD_TX_ADDR_INS |
+				 CMD_HD_EN |
+				 CMD_LCL_LOOP_EN |
+				 CMD_CNTL_FRM_EN |
+				 CMD_RMT_LOOP_EN |
+				 CMD_RX_ERR_DISC |
+				 CMD_PRBL_EN |
+				 CMD_TX_PAUSE_IGNORE |
+				 CMD_PAD_EN |
+				 CMD_PAUSE_FWD),
+			       CMD_PROMISC |
+			       CMD_NO_LEN_CHK |
+			       CMD_CNTL_FRM_EN |
+			       cmdcfg_sr,
+			       false);
 	bgmac->mac_speed = SPEED_UNKNOWN;
 	bgmac->mac_duplex = DUPLEX_UNKNOWN;
 
@@ -1039,16 +1049,16 @@ static void bgmac_enable(struct bgmac *bgmac)
 	u32 mode;
 
 	if (bgmac->feature_flags & BGMAC_FEAT_CMDCFG_SR_REV4)
-		cmdcfg_sr = BGMAC_CMDCFG_SR_REV4;
+		cmdcfg_sr = CMD_SW_RESET;
 	else
-		cmdcfg_sr = BGMAC_CMDCFG_SR_REV0;
+		cmdcfg_sr = CMD_SW_RESET_OLD;
 
-	cmdcfg = bgmac_read(bgmac, BGMAC_CMDCFG);
-	bgmac_cmdcfg_maskset(bgmac, ~(BGMAC_CMDCFG_TE | BGMAC_CMDCFG_RE),
-			     cmdcfg_sr, true);
+	cmdcfg = bgmac_umac_read(bgmac, UMAC_CMD);
+	bgmac_umac_cmd_maskset(bgmac, ~(CMD_TX_EN | CMD_RX_EN),
+			       cmdcfg_sr, true);
 	udelay(2);
-	cmdcfg |= BGMAC_CMDCFG_TE | BGMAC_CMDCFG_RE;
-	bgmac_write(bgmac, BGMAC_CMDCFG, cmdcfg);
+	cmdcfg |= CMD_TX_EN | CMD_RX_EN;
+	bgmac_umac_write(bgmac, UMAC_CMD, cmdcfg);
 
 	mode = (bgmac_read(bgmac, BGMAC_DEV_STATUS) & BGMAC_DS_MM_MASK) >>
 		BGMAC_DS_MM_SHIFT;
@@ -1068,7 +1078,7 @@ static void bgmac_enable(struct bgmac *bgmac)
 			fl_ctl = 0x03cb04cb;
 
 		bgmac_write(bgmac, BGMAC_FLOW_CTL_THRESH, fl_ctl);
-		bgmac_write(bgmac, BGMAC_PAUSE_CTL, 0x27fff);
+		bgmac_umac_write(bgmac, UMAC_PAUSE_CTRL, 0x27fff);
 	}
 
 	if (bgmac->feature_flags & BGMAC_FEAT_SET_RXQ_CLK) {
@@ -1095,18 +1105,18 @@ static void bgmac_chip_init(struct bgmac *bgmac)
 	bgmac_write(bgmac, BGMAC_INT_RECV_LAZY, 1 << BGMAC_IRL_FC_SHIFT);
 
 	/* Enable 802.3x tx flow control (honor received PAUSE frames) */
-	bgmac_cmdcfg_maskset(bgmac, ~BGMAC_CMDCFG_RPI, 0, true);
+	bgmac_umac_cmd_maskset(bgmac, ~CMD_RX_PAUSE_IGNORE, 0, true);
 
 	bgmac_set_rx_mode(bgmac->net_dev);
 
 	bgmac_write_mac_address(bgmac, bgmac->net_dev->dev_addr);
 
 	if (bgmac->loopback)
-		bgmac_cmdcfg_maskset(bgmac, ~0, BGMAC_CMDCFG_ML, false);
+		bgmac_umac_cmd_maskset(bgmac, ~0, CMD_LCL_LOOP_EN, false);
 	else
-		bgmac_cmdcfg_maskset(bgmac, ~BGMAC_CMDCFG_ML, 0, false);
+		bgmac_umac_cmd_maskset(bgmac, ~CMD_LCL_LOOP_EN, 0, false);
 
-	bgmac_write(bgmac, BGMAC_RXMAX_LENGTH, 32 + ETHER_MAX_LEN);
+	bgmac_umac_write(bgmac, UMAC_MAX_FRAME_LEN, 32 + ETHER_MAX_LEN);
 
 	bgmac_chip_intrs_on(bgmac);
 
@@ -1177,7 +1187,7 @@ static int bgmac_open(struct net_device *net_dev)
 	bgmac_chip_init(bgmac);
 
 	err = request_irq(bgmac->irq, bgmac_interrupt, IRQF_SHARED,
-			  KBUILD_MODNAME, net_dev);
+			  net_dev->name, net_dev);
 	if (err < 0) {
 		dev_err(bgmac->dev, "IRQ request error: %d!\n", err);
 		bgmac_dma_cleanup(bgmac);
@@ -1231,19 +1241,19 @@ static int bgmac_set_mac_address(struct net_device *net_dev, void *addr)
 	if (ret < 0)
 		return ret;
 
-	ether_addr_copy(net_dev->dev_addr, sa->sa_data);
+	eth_hw_addr_set(net_dev, sa->sa_data);
 	bgmac_write_mac_address(bgmac, net_dev->dev_addr);
 
 	eth_commit_mac_addr_change(net_dev, addr);
 	return 0;
 }
 
-static int bgmac_ioctl(struct net_device *net_dev, struct ifreq *ifr, int cmd)
+static int bgmac_change_mtu(struct net_device *net_dev, int mtu)
 {
-	if (!netif_running(net_dev))
-		return -EINVAL;
+	struct bgmac *bgmac = netdev_priv(net_dev);
 
-	return phy_mii_ioctl(net_dev->phydev, ifr, cmd);
+	bgmac_umac_write(bgmac, UMAC_MAX_FRAME_LEN, 32 + mtu);
+	return 0;
 }
 
 static const struct net_device_ops bgmac_netdev_ops = {
@@ -1253,7 +1263,8 @@ static const struct net_device_ops bgmac_netdev_ops = {
 	.ndo_set_rx_mode	= bgmac_set_rx_mode,
 	.ndo_set_mac_address	= bgmac_set_mac_address,
 	.ndo_validate_addr	= eth_validate_addr,
-	.ndo_do_ioctl           = bgmac_ioctl,
+	.ndo_eth_ioctl           = phy_do_ioctl_running,
+	.ndo_change_mtu		= bgmac_change_mtu,
 };
 
 /**************************************************
@@ -1356,8 +1367,7 @@ static void bgmac_get_strings(struct net_device *dev, u32 stringset,
 		return;
 
 	for (i = 0; i < BGMAC_STATS_LEN; i++)
-		strlcpy(data + i * ETH_GSTRING_LEN,
-			bgmac_get_strings_stats[i].name, ETH_GSTRING_LEN);
+		ethtool_puts(&data, bgmac_get_strings_stats[i].name);
 }
 
 static void bgmac_get_ethtool_stats(struct net_device *dev,
@@ -1384,8 +1394,8 @@ static void bgmac_get_ethtool_stats(struct net_device *dev,
 static void bgmac_get_drvinfo(struct net_device *net_dev,
 			      struct ethtool_drvinfo *info)
 {
-	strlcpy(info->driver, KBUILD_MODNAME, sizeof(info->driver));
-	strlcpy(info->bus_info, "AXI", sizeof(info->bus_info));
+	strscpy(info->driver, KBUILD_MODNAME, sizeof(info->driver));
+	strscpy(info->bus_info, "AXI", sizeof(info->bus_info));
 }
 
 static const struct ethtool_ops bgmac_ethtool_ops = {
@@ -1436,10 +1446,10 @@ int bgmac_phy_connect_direct(struct bgmac *bgmac)
 	struct phy_device *phy_dev;
 	int err;
 
-	phy_dev = fixed_phy_register(PHY_POLL, &fphy_status, -1, NULL);
-	if (!phy_dev || IS_ERR(phy_dev)) {
+	phy_dev = fixed_phy_register(&fphy_status, NULL);
+	if (IS_ERR(phy_dev)) {
 		dev_err(bgmac->dev, "Failed to register fixed PHY device\n");
-		return -ENODEV;
+		return PTR_ERR(phy_dev);
 	}
 
 	err = phy_connect_direct(bgmac->net_dev, phy_dev, bgmac_adjust_link,
@@ -1479,6 +1489,8 @@ int bgmac_enet_probe(struct bgmac *bgmac)
 	struct net_device *net_dev = bgmac->net_dev;
 	int err;
 
+	bgmac->in_init = true;
+
 	net_dev->irq = bgmac->irq;
 	SET_NETDEV_DEV(net_dev, bgmac->dev);
 	dev_set_drvdata(bgmac->dev, bgmac);
@@ -1496,9 +1508,13 @@ int bgmac_enet_probe(struct bgmac *bgmac)
 	 */
 	bgmac_clk_enable(bgmac, 0);
 
+	bgmac_chip_intrs_off(bgmac);
+
 	/* This seems to be fixing IRQ by assigning OOB #6 to the core */
-	if (bgmac->feature_flags & BGMAC_FEAT_IRQ_ID_OOB_6)
-		bgmac_idm_write(bgmac, BCMA_OOB_SEL_OUT_A30, 0x86);
+	if (!(bgmac->feature_flags & BGMAC_FEAT_IDM_MASK)) {
+		if (bgmac->feature_flags & BGMAC_FEAT_IRQ_ID_OOB_6)
+			bgmac_idm_write(bgmac, BCMA_OOB_SEL_OUT_A30, 0x86);
+	}
 
 	bgmac_chip_reset(bgmac);
 
@@ -1512,7 +1528,7 @@ int bgmac_enet_probe(struct bgmac *bgmac)
 	if (bcm47xx_nvram_getenv("et0_no_txint", NULL, 0) == 0)
 		bgmac->int_mask &= ~BGMAC_IS_TX_MASK;
 
-	netif_napi_add(net_dev, &bgmac->napi, bgmac_poll, BGMAC_WEIGHT);
+	netif_napi_add(net_dev, &bgmac->napi, bgmac_poll);
 
 	err = bgmac_phy_connect(bgmac);
 	if (err) {
@@ -1524,6 +1540,11 @@ int bgmac_enet_probe(struct bgmac *bgmac)
 	net_dev->hw_features = net_dev->features;
 	net_dev->vlan_features = net_dev->features;
 
+	/* Omit FCS from max MTU size */
+	net_dev->max_mtu = BGMAC_RX_MAX_FRAME_SIZE - ETH_FCS_LEN;
+
+	bgmac->in_init = false;
+
 	err = register_netdev(bgmac->net_dev);
 	if (err) {
 		dev_err(bgmac->dev, "Cannot register net device\n");
@@ -1550,7 +1571,6 @@ void bgmac_enet_remove(struct bgmac *bgmac)
 	phy_disconnect(bgmac->net_dev->phydev);
 	netif_napi_del(&bgmac->napi);
 	bgmac_dma_free(bgmac);
-	free_netdev(bgmac->net_dev);
 }
 EXPORT_SYMBOL_GPL(bgmac_enet_remove);
 
@@ -1605,4 +1625,5 @@ int bgmac_enet_resume(struct bgmac *bgmac)
 EXPORT_SYMBOL_GPL(bgmac_enet_resume);
 
 MODULE_AUTHOR("Rafał Miłecki");
+MODULE_DESCRIPTION("Broadcom iProc GBit driver");
 MODULE_LICENSE("GPL");
diff --git a/drivers/net/ethernet/broadcom/bgmac.h b/drivers/net/ethernet/broadcom/bgmac.h
index c1818766c501..6fee9a41839c 100644
--- a/drivers/net/ethernet/broadcom/bgmac.h
+++ b/drivers/net/ethernet/broadcom/bgmac.h
@@ -1,8 +1,11 @@
+/* SPDX-License-Identifier: GPL-2.0 */
 #ifndef _BGMAC_H
 #define _BGMAC_H
 
 #include <linux/netdevice.h>
 
+#include "unimac.h"
+
 #define BGMAC_DEV_CTL				0x000
 #define  BGMAC_DC_TSM				0x00000002
 #define  BGMAC_DC_CFCO				0x00000004
@@ -168,47 +171,7 @@
 #define BGMAC_RX_NONPAUSE_PKTS			0x420
 #define BGMAC_RX_SACHANGES			0x424
 #define BGMAC_RX_UNI_PKTS			0x428
-#define BGMAC_UNIMAC_VERSION			0x800
-#define BGMAC_HDBKP_CTL				0x804
-#define BGMAC_CMDCFG				0x808		/* Configuration */
-#define  BGMAC_CMDCFG_TE			0x00000001	/* Set to activate TX */
-#define  BGMAC_CMDCFG_RE			0x00000002	/* Set to activate RX */
-#define  BGMAC_CMDCFG_ES_MASK			0x0000000c	/* Ethernet speed see gmac_speed */
-#define   BGMAC_CMDCFG_ES_10			0x00000000
-#define   BGMAC_CMDCFG_ES_100			0x00000004
-#define   BGMAC_CMDCFG_ES_1000			0x00000008
-#define   BGMAC_CMDCFG_ES_2500			0x0000000C
-#define  BGMAC_CMDCFG_PROM			0x00000010	/* Set to activate promiscuous mode */
-#define  BGMAC_CMDCFG_PAD_EN			0x00000020
-#define  BGMAC_CMDCFG_CF			0x00000040
-#define  BGMAC_CMDCFG_PF			0x00000080
-#define  BGMAC_CMDCFG_RPI			0x00000100	/* Unset to enable 802.3x tx flow control */
-#define  BGMAC_CMDCFG_TAI			0x00000200
-#define  BGMAC_CMDCFG_HD			0x00000400	/* Set if in half duplex mode */
-#define  BGMAC_CMDCFG_HD_SHIFT			10
-#define  BGMAC_CMDCFG_SR_REV0			0x00000800	/* Set to reset mode, for core rev 0-3 */
-#define  BGMAC_CMDCFG_SR_REV4			0x00002000	/* Set to reset mode, for core rev >= 4 */
-#define  BGMAC_CMDCFG_ML			0x00008000	/* Set to activate mac loopback mode */
-#define  BGMAC_CMDCFG_AE			0x00400000
-#define  BGMAC_CMDCFG_CFE			0x00800000
-#define  BGMAC_CMDCFG_NLC			0x01000000
-#define  BGMAC_CMDCFG_RL			0x02000000
-#define  BGMAC_CMDCFG_RED			0x04000000
-#define  BGMAC_CMDCFG_PE			0x08000000
-#define  BGMAC_CMDCFG_TPI			0x10000000
-#define  BGMAC_CMDCFG_AT			0x20000000
-#define BGMAC_MACADDR_HIGH			0x80c		/* High 4 octets of own mac address */
-#define BGMAC_MACADDR_LOW			0x810		/* Low 2 octets of own mac address */
-#define BGMAC_RXMAX_LENGTH			0x814		/* Max receive frame length with vlan tag */
-#define BGMAC_PAUSEQUANTA			0x818
-#define BGMAC_MAC_MODE				0x844
-#define BGMAC_OUTERTAG				0x848
-#define BGMAC_INNERTAG				0x84c
-#define BGMAC_TXIPG				0x85c
-#define BGMAC_PAUSE_CTL				0xb30
-#define BGMAC_TX_FLUSH				0xb34
-#define BGMAC_RX_STATUS				0xb38
-#define BGMAC_TX_STATUS				0xb3c
+#define BGMAC_UNIMAC				0x800
 
 /* BCMA GMAC core specific IO Control (BCMA_IOCTL) flags */
 #define BGMAC_BCMA_IOCTL_SW_CLKEN		0x00000004	/* PHY Clock Enable */
@@ -350,7 +313,7 @@
 #define BGMAC_DESC_CTL0_IOC			0x20000000	/* IRQ on complete */
 #define BGMAC_DESC_CTL0_EOF			0x40000000	/* End of frame */
 #define BGMAC_DESC_CTL0_SOF			0x80000000	/* Start of frame */
-#define BGMAC_DESC_CTL1_LEN			0x00001FFF
+#define BGMAC_DESC_CTL1_LEN			0x00003FFF
 
 #define BGMAC_PHY_NOREGS			BRCM_PSEUDO_PHY_ADDR
 #define BGMAC_PHY_MASK				0x1F
@@ -365,7 +328,7 @@
 #define BGMAC_RX_FRAME_OFFSET			30		/* There are 2 unused bytes between header and real data */
 #define BGMAC_RX_BUF_OFFSET			(NET_SKB_PAD + NET_IP_ALIGN - \
 						 BGMAC_RX_FRAME_OFFSET)
-#define BGMAC_RX_MAX_FRAME_SIZE			1536		/* Copied from b44/tg3 */
+#define BGMAC_RX_MAX_FRAME_SIZE			1536
 #define BGMAC_RX_BUF_SIZE			(BGMAC_RX_FRAME_OFFSET + BGMAC_RX_MAX_FRAME_SIZE)
 #define BGMAC_RX_ALLOC_SIZE			(SKB_DATA_ALIGN(BGMAC_RX_BUF_SIZE + BGMAC_RX_BUF_OFFSET) + \
 						 SKB_DATA_ALIGN(sizeof(struct skb_shared_info)))
@@ -400,8 +363,6 @@
 #define BGMAC_CHIPCTL_7_IF_TYPE_MII		0x00000040
 #define BGMAC_CHIPCTL_7_IF_TYPE_RGMII		0x00000080
 
-#define BGMAC_WEIGHT	64
-
 #define ETHER_MAX_LEN	(ETH_FRAME_LEN + ETH_FCS_LEN)
 
 /* Feature Flags */
@@ -425,6 +386,7 @@
 #define BGMAC_FEAT_CC4_IF_SW_TYPE	BIT(17)
 #define BGMAC_FEAT_CC4_IF_SW_TYPE_RGMII	BIT(18)
 #define BGMAC_FEAT_CC7_IF_TYPE_RGMII	BIT(19)
+#define BGMAC_FEAT_IDM_MASK		BIT(20)
 
 struct bgmac_slot_info {
 	union {
@@ -477,9 +439,9 @@ struct bgmac_rx_header {
 struct bgmac {
 	union {
 		struct {
-			void *base;
-			void *idm_base;
-			void *nicpm_base;
+			void __iomem *base;
+			void __iomem *idm_base;
+			void __iomem *nicpm_base;
 		} plat;
 		struct {
 			struct bcma_device *core;
@@ -509,6 +471,8 @@ struct bgmac {
 	int irq;
 	u32 int_mask;
 
+	bool in_init;
+
 	/* Current MAC state */
 	int mac_speed;
 	int mac_duplex;
@@ -553,6 +517,16 @@ static inline void bgmac_write(struct bgmac *bgmac, u16 offset, u32 value)
 	bgmac->write(bgmac, offset, value);
 }
 
+static inline u32 bgmac_umac_read(struct bgmac *bgmac, u16 offset)
+{
+	return bgmac_read(bgmac, BGMAC_UNIMAC + offset);
+}
+
+static inline void bgmac_umac_write(struct bgmac *bgmac, u16 offset, u32 value)
+{
+	bgmac_write(bgmac, BGMAC_UNIMAC + offset, value);
+}
+
 static inline u32 bgmac_idm_read(struct bgmac *bgmac, u16 offset)
 {
 	return bgmac->idm_read(bgmac, offset);
@@ -606,6 +580,11 @@ static inline void bgmac_set(struct bgmac *bgmac, u16 offset, u32 set)
 	bgmac_maskset(bgmac, offset, ~0, set);
 }
 
+static inline void bgmac_umac_maskset(struct bgmac *bgmac, u16 offset, u32 mask, u32 set)
+{
+	bgmac_maskset(bgmac, BGMAC_UNIMAC + offset, mask, set);
+}
+
 static inline int bgmac_phy_connect(struct bgmac *bgmac)
 {
 	return bgmac->phy_connect(bgmac);
diff --git a/drivers/net/ethernet/broadcom/bnge/Makefile b/drivers/net/ethernet/broadcom/bnge/Makefile
new file mode 100644
index 000000000000..6142d9c57f49
--- /dev/null
+++ b/drivers/net/ethernet/broadcom/bnge/Makefile
@@ -0,0 +1,12 @@
+# SPDX-License-Identifier: GPL-2.0-only
+
+obj-$(CONFIG_BNGE) += bng_en.o
+
+bng_en-y := bnge_core.o \
+	    bnge_devlink.o \
+	    bnge_hwrm.o \
+	    bnge_hwrm_lib.o \
+	    bnge_rmem.o \
+	    bnge_resc.o \
+	    bnge_netdev.o \
+	    bnge_ethtool.o
diff --git a/drivers/net/ethernet/broadcom/bnge/bnge.h b/drivers/net/ethernet/broadcom/bnge/bnge.h
new file mode 100644
index 000000000000..7aed5f81cd51
--- /dev/null
+++ b/drivers/net/ethernet/broadcom/bnge/bnge.h
@@ -0,0 +1,245 @@
+/* SPDX-License-Identifier: GPL-2.0 */
+/* Copyright (c) 2025 Broadcom */
+
+#ifndef _BNGE_H_
+#define _BNGE_H_
+
+#define DRV_NAME	"bng_en"
+#define DRV_SUMMARY	"Broadcom 800G Ethernet Linux Driver"
+
+#include <linux/etherdevice.h>
+#include <linux/bnxt/hsi.h>
+#include "bnge_rmem.h"
+#include "bnge_resc.h"
+
+#define DRV_VER_MAJ	1
+#define DRV_VER_MIN	15
+#define DRV_VER_UPD	1
+
+extern char bnge_driver_name[];
+
+enum board_idx {
+	BCM57708,
+};
+
+struct bnge_pf_info {
+	u16	fw_fid;
+	u16	port_id;
+	u8	mac_addr[ETH_ALEN];
+};
+
+#define INVALID_HW_RING_ID      ((u16)-1)
+
+enum {
+	BNGE_FW_CAP_SHORT_CMD				= BIT_ULL(0),
+	BNGE_FW_CAP_LLDP_AGENT				= BIT_ULL(1),
+	BNGE_FW_CAP_DCBX_AGENT				= BIT_ULL(2),
+	BNGE_FW_CAP_IF_CHANGE				= BIT_ULL(3),
+	BNGE_FW_CAP_KONG_MB_CHNL			= BIT_ULL(4),
+	BNGE_FW_CAP_ERROR_RECOVERY			= BIT_ULL(5),
+	BNGE_FW_CAP_PKG_VER				= BIT_ULL(6),
+	BNGE_FW_CAP_CFA_ADV_FLOW			= BIT_ULL(7),
+	BNGE_FW_CAP_CFA_RFS_RING_TBL_IDX_V2		= BIT_ULL(8),
+	BNGE_FW_CAP_PCIE_STATS_SUPPORTED		= BIT_ULL(9),
+	BNGE_FW_CAP_EXT_STATS_SUPPORTED			= BIT_ULL(10),
+	BNGE_FW_CAP_ERR_RECOVER_RELOAD			= BIT_ULL(11),
+	BNGE_FW_CAP_HOT_RESET				= BIT_ULL(12),
+	BNGE_FW_CAP_RX_ALL_PKT_TS			= BIT_ULL(13),
+	BNGE_FW_CAP_VLAN_RX_STRIP			= BIT_ULL(14),
+	BNGE_FW_CAP_VLAN_TX_INSERT			= BIT_ULL(15),
+	BNGE_FW_CAP_EXT_HW_STATS_SUPPORTED		= BIT_ULL(16),
+	BNGE_FW_CAP_LIVEPATCH				= BIT_ULL(17),
+	BNGE_FW_CAP_HOT_RESET_IF			= BIT_ULL(18),
+	BNGE_FW_CAP_RING_MONITOR			= BIT_ULL(19),
+	BNGE_FW_CAP_DBG_QCAPS				= BIT_ULL(20),
+	BNGE_FW_CAP_THRESHOLD_TEMP_SUPPORTED		= BIT_ULL(21),
+	BNGE_FW_CAP_DFLT_VLAN_TPID_PCP			= BIT_ULL(22),
+	BNGE_FW_CAP_VNIC_TUNNEL_TPA			= BIT_ULL(23),
+	BNGE_FW_CAP_CFA_NTUPLE_RX_EXT_IP_PROTO		= BIT_ULL(24),
+	BNGE_FW_CAP_CFA_RFS_RING_TBL_IDX_V3		= BIT_ULL(25),
+	BNGE_FW_CAP_VNIC_RE_FLUSH			= BIT_ULL(26),
+};
+
+enum {
+	BNGE_EN_ROCE_V1					= BIT_ULL(0),
+	BNGE_EN_ROCE_V2					= BIT_ULL(1),
+	BNGE_EN_STRIP_VLAN				= BIT_ULL(2),
+	BNGE_EN_SHARED_CHNL				= BIT_ULL(3),
+	BNGE_EN_UDP_GSO_SUPP				= BIT_ULL(4),
+};
+
+#define BNGE_EN_ROCE		(BNGE_EN_ROCE_V1 | BNGE_EN_ROCE_V2)
+
+enum {
+	BNGE_RSS_CAP_RSS_HASH_TYPE_DELTA		= BIT(0),
+	BNGE_RSS_CAP_UDP_RSS_CAP			= BIT(1),
+	BNGE_RSS_CAP_NEW_RSS_CAP			= BIT(2),
+	BNGE_RSS_CAP_RSS_TCAM				= BIT(3),
+	BNGE_RSS_CAP_AH_V4_RSS_CAP			= BIT(4),
+	BNGE_RSS_CAP_AH_V6_RSS_CAP			= BIT(5),
+	BNGE_RSS_CAP_ESP_V4_RSS_CAP			= BIT(6),
+	BNGE_RSS_CAP_ESP_V6_RSS_CAP			= BIT(7),
+};
+
+#define BNGE_MAX_QUEUE		8
+struct bnge_queue_info {
+	u8      queue_id;
+	u8      queue_profile;
+};
+
+struct bnge_dev {
+	struct device	*dev;
+	struct pci_dev	*pdev;
+	struct net_device	*netdev;
+	u64	dsn;
+#define BNGE_VPD_FLD_LEN	32
+	char		board_partno[BNGE_VPD_FLD_LEN];
+	char		board_serialno[BNGE_VPD_FLD_LEN];
+
+	void __iomem	*bar0;
+	void __iomem	*bar1;
+
+	u16		chip_num;
+	u8		chip_rev;
+
+#if BITS_PER_LONG == 32
+	/* ensure atomic 64-bit doorbell writes on 32-bit systems. */
+	spinlock_t	db_lock;
+#endif
+	int		db_offset; /* db_offset within db_size */
+	int		db_size;
+
+	/* HWRM members */
+	u16			hwrm_cmd_seq;
+	u16			hwrm_cmd_kong_seq;
+	struct dma_pool		*hwrm_dma_pool;
+	struct hlist_head	hwrm_pending_list;
+	u16			hwrm_max_req_len;
+	u16			hwrm_max_ext_req_len;
+	unsigned int		hwrm_cmd_timeout;
+	unsigned int		hwrm_cmd_max_timeout;
+	struct mutex		hwrm_cmd_lock;	/* serialize hwrm messages */
+
+	struct hwrm_ver_get_output	ver_resp;
+#define FW_VER_STR_LEN		32
+	char			fw_ver_str[FW_VER_STR_LEN];
+	char			hwrm_ver_supp[FW_VER_STR_LEN];
+	char			nvm_cfg_ver[FW_VER_STR_LEN];
+	u64			fw_ver_code;
+#define BNGE_FW_VER_CODE(maj, min, bld, rsv)			\
+	((u64)(maj) << 48 | (u64)(min) << 32 | (u64)(bld) << 16 | (rsv))
+
+	struct bnge_pf_info	pf;
+
+	unsigned long           state;
+#define BNGE_STATE_DRV_REGISTERED      0
+#define BNGE_STATE_OPEN			1
+
+	u64			fw_cap;
+
+	/* Backing stores */
+	struct bnge_ctx_mem_info	*ctx;
+
+	u64			flags;
+
+	struct bnge_hw_resc	hw_resc;
+
+	u16			tso_max_segs;
+
+	int			max_fltr;
+#define BNGE_L2_FLTR_MAX_FLTR	1024
+
+	u32			*rss_indir_tbl;
+#define BNGE_RSS_TABLE_ENTRIES	64
+#define BNGE_RSS_TABLE_SIZE		(BNGE_RSS_TABLE_ENTRIES * 4)
+#define BNGE_RSS_TABLE_MAX_TBL	8
+#define BNGE_MAX_RSS_TABLE_SIZE				\
+	(BNGE_RSS_TABLE_SIZE * BNGE_RSS_TABLE_MAX_TBL)
+#define BNGE_MAX_RSS_TABLE_ENTRIES				\
+	(BNGE_RSS_TABLE_ENTRIES * BNGE_RSS_TABLE_MAX_TBL)
+	u16			rss_indir_tbl_entries;
+
+	u32			rss_cap;
+	u32			rss_hash_cfg;
+
+	u16			rx_nr_rings;
+	u16			tx_nr_rings;
+	u16			tx_nr_rings_per_tc;
+	/* Number of NQs */
+	u16			nq_nr_rings;
+
+	/* Aux device resources */
+	u16			aux_num_msix;
+	u16			aux_num_stat_ctxs;
+
+	u16			max_mtu;
+#define BNGE_MAX_MTU		9500
+
+	u16			hw_ring_stats_size;
+#define BNGE_NUM_RX_RING_STATS			8
+#define BNGE_NUM_TX_RING_STATS			8
+#define BNGE_NUM_TPA_RING_STATS			6
+#define BNGE_RING_STATS_SIZE					\
+	((BNGE_NUM_RX_RING_STATS + BNGE_NUM_TX_RING_STATS +	\
+	  BNGE_NUM_TPA_RING_STATS) * 8)
+
+	u16			max_tpa_v2;
+#define BNGE_SUPPORTS_TPA(bd)	((bd)->max_tpa_v2)
+
+	u8                      num_tc;
+	u8			max_tc;
+	u8			max_lltc;	/* lossless TCs */
+	struct bnge_queue_info	q_info[BNGE_MAX_QUEUE];
+	u8			tc_to_qidx[BNGE_MAX_QUEUE];
+	u8			q_ids[BNGE_MAX_QUEUE];
+	u8			max_q;
+	u8			port_count;
+
+	struct bnge_irq		*irq_tbl;
+	u16			irqs_acquired;
+};
+
+static inline bool bnge_is_roce_en(struct bnge_dev *bd)
+{
+	return bd->flags & BNGE_EN_ROCE;
+}
+
+static inline bool bnge_is_agg_reqd(struct bnge_dev *bd)
+{
+	if (bd->netdev) {
+		struct bnge_net *bn = netdev_priv(bd->netdev);
+
+		if (bn->priv_flags & BNGE_NET_EN_TPA ||
+		    bn->priv_flags & BNGE_NET_EN_JUMBO)
+			return true;
+		else
+			return false;
+	}
+
+	return true;
+}
+
+static inline void bnge_writeq(struct bnge_dev *bd, u64 val,
+			       void __iomem *addr)
+{
+#if BITS_PER_LONG == 32
+	spin_lock(&bd->db_lock);
+	lo_hi_writeq(val, addr);
+	spin_unlock(&bd->db_lock);
+#else
+	writeq(val, addr);
+#endif
+}
+
+/* For TX and RX ring doorbells */
+static inline void bnge_db_write(struct bnge_dev *bd, struct bnge_db_info *db,
+				 u32 idx)
+{
+	bnge_writeq(bd, db->db_key64 | DB_RING_IDX(db, idx),
+		    db->doorbell);
+}
+
+bool bnge_aux_registered(struct bnge_dev *bd);
+u16 bnge_aux_get_msix(struct bnge_dev *bd);
+
+#endif /* _BNGE_H_ */
diff --git a/drivers/net/ethernet/broadcom/bnge/bnge_core.c b/drivers/net/ethernet/broadcom/bnge/bnge_core.c
new file mode 100644
index 000000000000..2c72dd34d50d
--- /dev/null
+++ b/drivers/net/ethernet/broadcom/bnge/bnge_core.c
@@ -0,0 +1,404 @@
+// SPDX-License-Identifier: GPL-2.0
+// Copyright (c) 2025 Broadcom.
+
+#include <linux/init.h>
+#include <linux/crash_dump.h>
+#include <linux/module.h>
+#include <linux/pci.h>
+
+#include "bnge.h"
+#include "bnge_devlink.h"
+#include "bnge_hwrm.h"
+#include "bnge_hwrm_lib.h"
+
+MODULE_LICENSE("GPL");
+MODULE_DESCRIPTION(DRV_SUMMARY);
+
+char bnge_driver_name[] = DRV_NAME;
+
+static const struct {
+	char *name;
+} board_info[] = {
+	[BCM57708] = { "Broadcom BCM57708 50Gb/100Gb/200Gb/400Gb/800Gb Ethernet" },
+};
+
+static const struct pci_device_id bnge_pci_tbl[] = {
+	{ PCI_VDEVICE(BROADCOM, 0x1780), .driver_data = BCM57708 },
+	/* Required last entry */
+	{0, }
+};
+MODULE_DEVICE_TABLE(pci, bnge_pci_tbl);
+
+static void bnge_print_device_info(struct pci_dev *pdev, enum board_idx idx)
+{
+	struct device *dev = &pdev->dev;
+
+	dev_info(dev, "%s found at mem %lx\n", board_info[idx].name,
+		 (long)pci_resource_start(pdev, 0));
+
+	pcie_print_link_status(pdev);
+}
+
+bool bnge_aux_registered(struct bnge_dev *bd)
+{
+	return false;
+}
+
+static void bnge_nvm_cfg_ver_get(struct bnge_dev *bd)
+{
+	struct hwrm_nvm_get_dev_info_output nvm_info;
+
+	if (!bnge_hwrm_nvm_dev_info(bd, &nvm_info))
+		snprintf(bd->nvm_cfg_ver, FW_VER_STR_LEN, "%d.%d.%d",
+			 nvm_info.nvm_cfg_ver_maj, nvm_info.nvm_cfg_ver_min,
+			 nvm_info.nvm_cfg_ver_upd);
+}
+
+static int bnge_func_qcaps(struct bnge_dev *bd)
+{
+	int rc;
+
+	rc = bnge_hwrm_func_qcaps(bd);
+	if (rc)
+		return rc;
+
+	rc = bnge_hwrm_queue_qportcfg(bd);
+	if (rc) {
+		dev_err(bd->dev, "query qportcfg failure rc: %d\n", rc);
+		return rc;
+	}
+
+	rc = bnge_hwrm_func_resc_qcaps(bd);
+	if (rc) {
+		dev_err(bd->dev, "query resc caps failure rc: %d\n", rc);
+		return rc;
+	}
+
+	rc = bnge_hwrm_func_qcfg(bd);
+	if (rc) {
+		dev_err(bd->dev, "query config failure rc: %d\n", rc);
+		return rc;
+	}
+
+	rc = bnge_hwrm_vnic_qcaps(bd);
+	if (rc) {
+		dev_err(bd->dev, "vnic caps failure rc: %d\n", rc);
+		return rc;
+	}
+
+	return 0;
+}
+
+static void bnge_fw_unregister_dev(struct bnge_dev *bd)
+{
+	/* ctx mem free after unrgtr only */
+	bnge_hwrm_func_drv_unrgtr(bd);
+	bnge_free_ctx_mem(bd);
+}
+
+static void bnge_set_dflt_rss_hash_type(struct bnge_dev *bd)
+{
+	bd->rss_hash_cfg = VNIC_RSS_CFG_REQ_HASH_TYPE_IPV4 |
+			   VNIC_RSS_CFG_REQ_HASH_TYPE_TCP_IPV4 |
+			   VNIC_RSS_CFG_REQ_HASH_TYPE_IPV6 |
+			   VNIC_RSS_CFG_REQ_HASH_TYPE_TCP_IPV6 |
+			   VNIC_RSS_CFG_REQ_HASH_TYPE_UDP_IPV4 |
+			   VNIC_RSS_CFG_REQ_HASH_TYPE_UDP_IPV6;
+}
+
+static int bnge_fw_register_dev(struct bnge_dev *bd)
+{
+	int rc;
+
+	bd->fw_cap = 0;
+	rc = bnge_hwrm_ver_get(bd);
+	if (rc) {
+		dev_err(bd->dev, "Get Version command failed rc: %d\n", rc);
+		return rc;
+	}
+
+	bnge_nvm_cfg_ver_get(bd);
+
+	rc = bnge_hwrm_func_reset(bd);
+	if (rc) {
+		dev_err(bd->dev, "Failed to reset function rc: %d\n", rc);
+		return rc;
+	}
+
+	bnge_hwrm_fw_set_time(bd);
+
+	rc =  bnge_hwrm_func_drv_rgtr(bd);
+	if (rc) {
+		dev_err(bd->dev, "Failed to rgtr with firmware rc: %d\n", rc);
+		return rc;
+	}
+
+	rc = bnge_alloc_ctx_mem(bd);
+	if (rc) {
+		dev_err(bd->dev, "Failed to allocate ctx mem rc: %d\n", rc);
+		goto err_func_unrgtr;
+	}
+
+	/* Get the resources and configuration from firmware */
+	rc = bnge_func_qcaps(bd);
+	if (rc) {
+		dev_err(bd->dev, "Failed initial configuration rc: %d\n", rc);
+		rc = -ENODEV;
+		goto err_func_unrgtr;
+	}
+
+	bnge_set_dflt_rss_hash_type(bd);
+
+	return 0;
+
+err_func_unrgtr:
+	bnge_fw_unregister_dev(bd);
+	return rc;
+}
+
+static void bnge_pci_disable(struct pci_dev *pdev)
+{
+	pci_release_regions(pdev);
+	if (pci_is_enabled(pdev))
+		pci_disable_device(pdev);
+}
+
+static int bnge_pci_enable(struct pci_dev *pdev)
+{
+	int rc;
+
+	rc = pci_enable_device(pdev);
+	if (rc) {
+		dev_err(&pdev->dev, "Cannot enable PCI device, aborting\n");
+		return rc;
+	}
+
+	if (!(pci_resource_flags(pdev, 0) & IORESOURCE_MEM)) {
+		dev_err(&pdev->dev,
+			"Cannot find PCI device base address, aborting\n");
+		rc = -ENODEV;
+		goto err_pci_disable;
+	}
+
+	rc = pci_request_regions(pdev, bnge_driver_name);
+	if (rc) {
+		dev_err(&pdev->dev, "Cannot obtain PCI resources, aborting\n");
+		goto err_pci_disable;
+	}
+
+	dma_set_mask_and_coherent(&pdev->dev, DMA_BIT_MASK(64));
+
+	pci_set_master(pdev);
+
+	return 0;
+
+err_pci_disable:
+	pci_disable_device(pdev);
+	return rc;
+}
+
+static void bnge_unmap_bars(struct pci_dev *pdev)
+{
+	struct bnge_dev *bd = pci_get_drvdata(pdev);
+
+	if (bd->bar1) {
+		pci_iounmap(pdev, bd->bar1);
+		bd->bar1 = NULL;
+	}
+
+	if (bd->bar0) {
+		pci_iounmap(pdev, bd->bar0);
+		bd->bar0 = NULL;
+	}
+}
+
+static void bnge_set_max_func_irqs(struct bnge_dev *bd,
+				   unsigned int max_irqs)
+{
+	bd->hw_resc.max_irqs = max_irqs;
+}
+
+static int bnge_get_max_irq(struct pci_dev *pdev)
+{
+	u16 ctrl;
+
+	pci_read_config_word(pdev, pdev->msix_cap + PCI_MSIX_FLAGS, &ctrl);
+	return (ctrl & PCI_MSIX_FLAGS_QSIZE) + 1;
+}
+
+static int bnge_map_db_bar(struct bnge_dev *bd)
+{
+	if (!bd->db_size)
+		return -ENODEV;
+
+	bd->bar1 = pci_iomap(bd->pdev, 2, bd->db_size);
+	if (!bd->bar1)
+		return -ENOMEM;
+
+	return 0;
+}
+
+static int bnge_probe_one(struct pci_dev *pdev, const struct pci_device_id *ent)
+{
+	unsigned int max_irqs;
+	struct bnge_dev *bd;
+	int rc;
+
+	if (pci_is_bridge(pdev))
+		return -ENODEV;
+
+	if (!pdev->msix_cap) {
+		dev_err(&pdev->dev, "MSIX capability missing, aborting\n");
+		return -ENODEV;
+	}
+
+	if (is_kdump_kernel()) {
+		pci_clear_master(pdev);
+		pcie_flr(pdev);
+	}
+
+	rc = bnge_pci_enable(pdev);
+	if (rc)
+		return rc;
+
+	bnge_print_device_info(pdev, ent->driver_data);
+
+	bd = bnge_devlink_alloc(pdev);
+	if (!bd) {
+		dev_err(&pdev->dev, "Devlink allocation failed\n");
+		rc = -ENOMEM;
+		goto err_pci_disable;
+	}
+
+	bd->bar0 = pci_ioremap_bar(pdev, 0);
+	if (!bd->bar0) {
+		dev_err(&pdev->dev, "Failed mapping BAR-0, aborting\n");
+		rc = -ENOMEM;
+		goto err_devl_free;
+	}
+
+	rc = bnge_init_hwrm_resources(bd);
+	if (rc)
+		goto err_bar_unmap;
+
+	rc = bnge_fw_register_dev(bd);
+	if (rc) {
+		dev_err(&pdev->dev, "Failed to register with firmware rc = %d\n", rc);
+		goto err_hwrm_cleanup;
+	}
+
+	bnge_devlink_register(bd);
+
+	max_irqs = bnge_get_max_irq(pdev);
+	bnge_set_max_func_irqs(bd, max_irqs);
+
+	bnge_aux_init_dflt_config(bd);
+
+	rc = bnge_net_init_dflt_config(bd);
+	if (rc) {
+		dev_err(&pdev->dev, "Error setting up default cfg to netdev rc = %d\n",
+			rc);
+		goto err_fw_reg;
+	}
+
+	rc = bnge_map_db_bar(bd);
+	if (rc) {
+		dev_err(&pdev->dev, "Failed mapping doorbell BAR rc = %d, aborting\n",
+			rc);
+		goto err_config_uninit;
+	}
+
+#if BITS_PER_LONG == 32
+	spin_lock_init(&bd->db_lock);
+#endif
+
+	rc = bnge_alloc_irqs(bd);
+	if (rc) {
+		dev_err(&pdev->dev, "Error IRQ allocation rc = %d\n", rc);
+		goto err_config_uninit;
+	}
+
+	rc = bnge_netdev_alloc(bd, max_irqs);
+	if (rc)
+		goto err_free_irq;
+
+	pci_save_state(pdev);
+
+	return 0;
+
+err_free_irq:
+	bnge_free_irqs(bd);
+
+err_config_uninit:
+	bnge_net_uninit_dflt_config(bd);
+
+err_fw_reg:
+	bnge_devlink_unregister(bd);
+	bnge_fw_unregister_dev(bd);
+
+err_hwrm_cleanup:
+	bnge_cleanup_hwrm_resources(bd);
+
+err_bar_unmap:
+	bnge_unmap_bars(pdev);
+
+err_devl_free:
+	bnge_devlink_free(bd);
+
+err_pci_disable:
+	bnge_pci_disable(pdev);
+	return rc;
+}
+
+static void bnge_remove_one(struct pci_dev *pdev)
+{
+	struct bnge_dev *bd = pci_get_drvdata(pdev);
+
+	bnge_netdev_free(bd);
+
+	bnge_free_irqs(bd);
+
+	bnge_net_uninit_dflt_config(bd);
+
+	bnge_devlink_unregister(bd);
+
+	bnge_fw_unregister_dev(bd);
+
+	bnge_cleanup_hwrm_resources(bd);
+
+	bnge_unmap_bars(pdev);
+
+	bnge_devlink_free(bd);
+
+	bnge_pci_disable(pdev);
+}
+
+static void bnge_shutdown(struct pci_dev *pdev)
+{
+	pci_disable_device(pdev);
+
+	if (system_state == SYSTEM_POWER_OFF) {
+		pci_wake_from_d3(pdev, 0);
+		pci_set_power_state(pdev, PCI_D3hot);
+	}
+}
+
+static struct pci_driver bnge_driver = {
+	.name		= bnge_driver_name,
+	.id_table	= bnge_pci_tbl,
+	.probe		= bnge_probe_one,
+	.remove		= bnge_remove_one,
+	.shutdown	= bnge_shutdown,
+};
+
+static int __init bnge_init_module(void)
+{
+	return pci_register_driver(&bnge_driver);
+}
+module_init(bnge_init_module);
+
+static void __exit bnge_exit_module(void)
+{
+	pci_unregister_driver(&bnge_driver);
+}
+module_exit(bnge_exit_module);
diff --git a/drivers/net/ethernet/broadcom/bnge/bnge_db.h b/drivers/net/ethernet/broadcom/bnge/bnge_db.h
new file mode 100644
index 000000000000..950ed582f1d8
--- /dev/null
+++ b/drivers/net/ethernet/broadcom/bnge/bnge_db.h
@@ -0,0 +1,34 @@
+/* SPDX-License-Identifier: GPL-2.0 */
+/* Copyright (c) 2025 Broadcom */
+
+#ifndef _BNGE_DB_H_
+#define _BNGE_DB_H_
+
+/* 64-bit doorbell */
+#define DBR_EPOCH_SFT					24
+#define DBR_TOGGLE_SFT					25
+#define DBR_XID_SFT					32
+#define DBR_PATH_L2					(0x1ULL << 56)
+#define DBR_VALID					(0x1ULL << 58)
+#define DBR_TYPE_SQ					(0x0ULL << 60)
+#define DBR_TYPE_SRQ					(0x2ULL << 60)
+#define DBR_TYPE_CQ					(0x4ULL << 60)
+#define DBR_TYPE_CQ_ARMALL				(0x6ULL << 60)
+#define DBR_TYPE_NQ					(0xaULL << 60)
+#define DBR_TYPE_NQ_ARM					(0xbULL << 60)
+#define DBR_TYPE_NQ_MASK				(0xeULL << 60)
+
+struct bnge_db_info {
+	void __iomem		*doorbell;
+	u64			db_key64;
+	u32			db_ring_mask;
+	u32			db_epoch_mask;
+	u8			db_epoch_shift;
+};
+
+#define DB_EPOCH(db, idx)	(((idx) & (db)->db_epoch_mask) <<	\
+				 ((db)->db_epoch_shift))
+#define DB_RING_IDX(db, idx)	(((idx) & (db)->db_ring_mask) |		\
+				 DB_EPOCH(db, idx))
+
+#endif /* _BNGE_DB_H_ */
diff --git a/drivers/net/ethernet/broadcom/bnge/bnge_devlink.c b/drivers/net/ethernet/broadcom/bnge/bnge_devlink.c
new file mode 100644
index 000000000000..a987afebd64d
--- /dev/null
+++ b/drivers/net/ethernet/broadcom/bnge/bnge_devlink.c
@@ -0,0 +1,306 @@
+// SPDX-License-Identifier: GPL-2.0
+// Copyright (c) 2025 Broadcom.
+
+#include <linux/unaligned.h>
+#include <linux/pci.h>
+#include <linux/types.h>
+#include <net/devlink.h>
+
+#include "bnge.h"
+#include "bnge_devlink.h"
+#include "bnge_hwrm_lib.h"
+
+static int bnge_dl_info_put(struct bnge_dev *bd, struct devlink_info_req *req,
+			    enum bnge_dl_version_type type, const char *key,
+			    char *buf)
+{
+	if (!strlen(buf))
+		return 0;
+
+	if (!strcmp(key, DEVLINK_INFO_VERSION_GENERIC_FW_NCSI) ||
+	    !strcmp(key, DEVLINK_INFO_VERSION_GENERIC_FW_ROCE))
+		return 0;
+
+	switch (type) {
+	case BNGE_VERSION_FIXED:
+		return devlink_info_version_fixed_put(req, key, buf);
+	case BNGE_VERSION_RUNNING:
+		return devlink_info_version_running_put(req, key, buf);
+	case BNGE_VERSION_STORED:
+		return devlink_info_version_stored_put(req, key, buf);
+	}
+
+	return 0;
+}
+
+static void bnge_vpd_read_info(struct bnge_dev *bd)
+{
+	struct pci_dev *pdev = bd->pdev;
+	unsigned int vpd_size, kw_len;
+	int pos, size;
+	u8 *vpd_data;
+
+	vpd_data = pci_vpd_alloc(pdev, &vpd_size);
+	if (IS_ERR(vpd_data)) {
+		pci_warn(pdev, "Unable to read VPD\n");
+		return;
+	}
+
+	pos = pci_vpd_find_ro_info_keyword(vpd_data, vpd_size,
+					   PCI_VPD_RO_KEYWORD_PARTNO, &kw_len);
+	if (pos < 0)
+		goto read_sn;
+
+	size = min_t(int, kw_len, BNGE_VPD_FLD_LEN - 1);
+	memcpy(bd->board_partno, &vpd_data[pos], size);
+
+read_sn:
+	pos = pci_vpd_find_ro_info_keyword(vpd_data, vpd_size,
+					   PCI_VPD_RO_KEYWORD_SERIALNO,
+					   &kw_len);
+	if (pos < 0)
+		goto exit;
+
+	size = min_t(int, kw_len, BNGE_VPD_FLD_LEN - 1);
+	memcpy(bd->board_serialno, &vpd_data[pos], size);
+
+exit:
+	kfree(vpd_data);
+}
+
+#define HWRM_FW_VER_STR_LEN	16
+
+static int bnge_devlink_info_get(struct devlink *devlink,
+				 struct devlink_info_req *req,
+				 struct netlink_ext_ack *extack)
+{
+	struct hwrm_nvm_get_dev_info_output nvm_dev_info;
+	struct bnge_dev *bd = devlink_priv(devlink);
+	struct hwrm_ver_get_output *ver_resp;
+	char mgmt_ver[FW_VER_STR_LEN];
+	char roce_ver[FW_VER_STR_LEN];
+	char ncsi_ver[FW_VER_STR_LEN];
+	char buf[32];
+
+	int rc;
+
+	if (bd->dsn) {
+		char buf[32];
+		u8 dsn[8];
+		int rc;
+
+		put_unaligned_le64(bd->dsn, dsn);
+		sprintf(buf, "%02X-%02X-%02X-%02X-%02X-%02X-%02X-%02X",
+			dsn[7], dsn[6], dsn[5], dsn[4],
+			dsn[3], dsn[2], dsn[1], dsn[0]);
+		rc = devlink_info_serial_number_put(req, buf);
+		if (rc) {
+			NL_SET_ERR_MSG_MOD(extack, "Failed to set dsn");
+			return rc;
+		}
+	}
+
+	if (strlen(bd->board_serialno)) {
+		rc = devlink_info_board_serial_number_put(req,
+							  bd->board_serialno);
+		if (rc) {
+			NL_SET_ERR_MSG_MOD(extack,
+					   "Failed to set board serial number");
+			return rc;
+		}
+	}
+
+	rc = bnge_dl_info_put(bd, req, BNGE_VERSION_FIXED,
+			      DEVLINK_INFO_VERSION_GENERIC_BOARD_ID,
+			      bd->board_partno);
+	if (rc) {
+		NL_SET_ERR_MSG_MOD(extack, "Failed to set board part number");
+		return rc;
+	}
+
+	/* More information from HWRM ver get command */
+	sprintf(buf, "%X", bd->chip_num);
+	rc = bnge_dl_info_put(bd, req, BNGE_VERSION_FIXED,
+			      DEVLINK_INFO_VERSION_GENERIC_ASIC_ID, buf);
+	if (rc) {
+		NL_SET_ERR_MSG_MOD(extack, "Failed to set asic id");
+		return rc;
+	}
+
+	ver_resp = &bd->ver_resp;
+	sprintf(buf, "%c%d", 'A' + ver_resp->chip_rev, ver_resp->chip_metal);
+	rc = bnge_dl_info_put(bd, req, BNGE_VERSION_FIXED,
+			      DEVLINK_INFO_VERSION_GENERIC_ASIC_REV, buf);
+	if (rc) {
+		NL_SET_ERR_MSG_MOD(extack, "Failed to set asic info");
+		return rc;
+	}
+
+	rc = bnge_dl_info_put(bd, req, BNGE_VERSION_RUNNING,
+			      DEVLINK_INFO_VERSION_GENERIC_FW_PSID,
+			      bd->nvm_cfg_ver);
+	if (rc) {
+		NL_SET_ERR_MSG_MOD(extack, "Failed to set firmware version");
+		return rc;
+	}
+
+	buf[0] = 0;
+	strncat(buf, ver_resp->active_pkg_name, HWRM_FW_VER_STR_LEN);
+	rc = bnge_dl_info_put(bd, req, BNGE_VERSION_RUNNING,
+			      DEVLINK_INFO_VERSION_GENERIC_FW, buf);
+	if (rc) {
+		NL_SET_ERR_MSG_MOD(extack,
+				   "Failed to set firmware generic version");
+		return rc;
+	}
+
+	if (ver_resp->flags & VER_GET_RESP_FLAGS_EXT_VER_AVAIL) {
+		snprintf(mgmt_ver, FW_VER_STR_LEN, "%d.%d.%d.%d",
+			 ver_resp->hwrm_fw_major, ver_resp->hwrm_fw_minor,
+			 ver_resp->hwrm_fw_build, ver_resp->hwrm_fw_patch);
+
+		snprintf(ncsi_ver, FW_VER_STR_LEN, "%d.%d.%d.%d",
+			 ver_resp->mgmt_fw_major, ver_resp->mgmt_fw_minor,
+			 ver_resp->mgmt_fw_build, ver_resp->mgmt_fw_patch);
+
+		snprintf(roce_ver, FW_VER_STR_LEN, "%d.%d.%d.%d",
+			 ver_resp->roce_fw_major, ver_resp->roce_fw_minor,
+			 ver_resp->roce_fw_build, ver_resp->roce_fw_patch);
+	} else {
+		snprintf(mgmt_ver, FW_VER_STR_LEN, "%d.%d.%d.%d",
+			 ver_resp->hwrm_fw_maj_8b, ver_resp->hwrm_fw_min_8b,
+			 ver_resp->hwrm_fw_bld_8b, ver_resp->hwrm_fw_rsvd_8b);
+
+		snprintf(ncsi_ver, FW_VER_STR_LEN, "%d.%d.%d.%d",
+			 ver_resp->mgmt_fw_maj_8b, ver_resp->mgmt_fw_min_8b,
+			 ver_resp->mgmt_fw_bld_8b, ver_resp->mgmt_fw_rsvd_8b);
+
+		snprintf(roce_ver, FW_VER_STR_LEN, "%d.%d.%d.%d",
+			 ver_resp->roce_fw_maj_8b, ver_resp->roce_fw_min_8b,
+			 ver_resp->roce_fw_bld_8b, ver_resp->roce_fw_rsvd_8b);
+	}
+	rc = bnge_dl_info_put(bd, req, BNGE_VERSION_RUNNING,
+			      DEVLINK_INFO_VERSION_GENERIC_FW_MGMT, mgmt_ver);
+	if (rc) {
+		NL_SET_ERR_MSG_MOD(extack,
+				   "Failed to set firmware mgmt version");
+		return rc;
+	}
+
+	rc = bnge_dl_info_put(bd, req, BNGE_VERSION_RUNNING,
+			      DEVLINK_INFO_VERSION_GENERIC_FW_MGMT_API,
+			      bd->hwrm_ver_supp);
+	if (rc) {
+		NL_SET_ERR_MSG_MOD(extack,
+				   "Failed to set firmware mgmt api version");
+		return rc;
+	}
+
+	rc = bnge_dl_info_put(bd, req, BNGE_VERSION_RUNNING,
+			      DEVLINK_INFO_VERSION_GENERIC_FW_NCSI, ncsi_ver);
+	if (rc) {
+		NL_SET_ERR_MSG_MOD(extack,
+				   "Failed to set ncsi firmware version");
+		return rc;
+	}
+
+	rc = bnge_dl_info_put(bd, req, BNGE_VERSION_RUNNING,
+			      DEVLINK_INFO_VERSION_GENERIC_FW_ROCE, roce_ver);
+	if (rc) {
+		NL_SET_ERR_MSG_MOD(extack, "Failed to set roce firmware version");
+		return rc;
+	}
+
+	rc = bnge_hwrm_nvm_dev_info(bd, &nvm_dev_info);
+	if (!(nvm_dev_info.flags & NVM_GET_DEV_INFO_RESP_FLAGS_FW_VER_VALID))
+		return 0;
+
+	buf[0] = 0;
+	strncat(buf, nvm_dev_info.pkg_name, HWRM_FW_VER_STR_LEN);
+	rc = bnge_dl_info_put(bd, req, BNGE_VERSION_STORED,
+			      DEVLINK_INFO_VERSION_GENERIC_FW, buf);
+	if (rc) {
+		NL_SET_ERR_MSG_MOD(extack,
+				   "Failed to set roce firmware version");
+		return rc;
+	}
+
+	snprintf(mgmt_ver, FW_VER_STR_LEN, "%d.%d.%d.%d",
+		 nvm_dev_info.hwrm_fw_major, nvm_dev_info.hwrm_fw_minor,
+		 nvm_dev_info.hwrm_fw_build, nvm_dev_info.hwrm_fw_patch);
+	rc = bnge_dl_info_put(bd, req, BNGE_VERSION_STORED,
+			      DEVLINK_INFO_VERSION_GENERIC_FW_MGMT, mgmt_ver);
+	if (rc) {
+		NL_SET_ERR_MSG_MOD(extack,
+				   "Failed to set stored firmware version");
+		return rc;
+	}
+
+	snprintf(ncsi_ver, FW_VER_STR_LEN, "%d.%d.%d.%d",
+		 nvm_dev_info.mgmt_fw_major, nvm_dev_info.mgmt_fw_minor,
+		 nvm_dev_info.mgmt_fw_build, nvm_dev_info.mgmt_fw_patch);
+	rc = bnge_dl_info_put(bd, req, BNGE_VERSION_STORED,
+			      DEVLINK_INFO_VERSION_GENERIC_FW_NCSI, ncsi_ver);
+	if (rc) {
+		NL_SET_ERR_MSG_MOD(extack,
+				   "Failed to set stored ncsi firmware version");
+		return rc;
+	}
+
+	snprintf(roce_ver, FW_VER_STR_LEN, "%d.%d.%d.%d",
+		 nvm_dev_info.roce_fw_major, nvm_dev_info.roce_fw_minor,
+		 nvm_dev_info.roce_fw_build, nvm_dev_info.roce_fw_patch);
+	rc = bnge_dl_info_put(bd, req, BNGE_VERSION_STORED,
+			      DEVLINK_INFO_VERSION_GENERIC_FW_ROCE, roce_ver);
+	if (rc)
+		NL_SET_ERR_MSG_MOD(extack,
+				   "Failed to set stored roce firmware version");
+
+	return rc;
+}
+
+static const struct devlink_ops bnge_devlink_ops = {
+	.info_get = bnge_devlink_info_get,
+};
+
+void bnge_devlink_free(struct bnge_dev *bd)
+{
+	struct devlink *devlink = priv_to_devlink(bd);
+
+	devlink_free(devlink);
+}
+
+struct bnge_dev *bnge_devlink_alloc(struct pci_dev *pdev)
+{
+	struct devlink *devlink;
+	struct bnge_dev *bd;
+
+	devlink = devlink_alloc(&bnge_devlink_ops, sizeof(*bd), &pdev->dev);
+	if (!devlink)
+		return NULL;
+
+	bd = devlink_priv(devlink);
+	pci_set_drvdata(pdev, bd);
+	bd->dev = &pdev->dev;
+	bd->pdev = pdev;
+
+	bd->dsn = pci_get_dsn(pdev);
+	if (!bd->dsn)
+		pci_warn(pdev, "Failed to get DSN\n");
+
+	bnge_vpd_read_info(bd);
+
+	return bd;
+}
+
+void bnge_devlink_register(struct bnge_dev *bd)
+{
+	struct devlink *devlink = priv_to_devlink(bd);
+	devlink_register(devlink);
+}
+
+void bnge_devlink_unregister(struct bnge_dev *bd)
+{
+	struct devlink *devlink = priv_to_devlink(bd);
+	devlink_unregister(devlink);
+}
diff --git a/drivers/net/ethernet/broadcom/bnge/bnge_devlink.h b/drivers/net/ethernet/broadcom/bnge/bnge_devlink.h
new file mode 100644
index 000000000000..c6575255e650
--- /dev/null
+++ b/drivers/net/ethernet/broadcom/bnge/bnge_devlink.h
@@ -0,0 +1,18 @@
+/* SPDX-License-Identifier: GPL-2.0 */
+/* Copyright (c) 2025 Broadcom */
+
+#ifndef _BNGE_DEVLINK_H_
+#define _BNGE_DEVLINK_H_
+
+enum bnge_dl_version_type {
+	BNGE_VERSION_FIXED,
+	BNGE_VERSION_RUNNING,
+	BNGE_VERSION_STORED,
+};
+
+void bnge_devlink_free(struct bnge_dev *bd);
+struct bnge_dev *bnge_devlink_alloc(struct pci_dev *pdev);
+void bnge_devlink_register(struct bnge_dev *bd);
+void bnge_devlink_unregister(struct bnge_dev *bd);
+
+#endif /* _BNGE_DEVLINK_H_ */
diff --git a/drivers/net/ethernet/broadcom/bnge/bnge_ethtool.c b/drivers/net/ethernet/broadcom/bnge/bnge_ethtool.c
new file mode 100644
index 000000000000..569371c1b4f2
--- /dev/null
+++ b/drivers/net/ethernet/broadcom/bnge/bnge_ethtool.c
@@ -0,0 +1,33 @@
+// SPDX-License-Identifier: GPL-2.0
+// Copyright (c) 2025 Broadcom.
+
+#include <linux/unaligned.h>
+#include <linux/pci.h>
+#include <linux/types.h>
+#include <net/devlink.h>
+#include <linux/ethtool.h>
+#include <linux/etherdevice.h>
+#include <linux/ethtool_netlink.h>
+
+#include "bnge.h"
+#include "bnge_ethtool.h"
+
+static void bnge_get_drvinfo(struct net_device *dev,
+			     struct ethtool_drvinfo *info)
+{
+	struct bnge_net *bn = netdev_priv(dev);
+	struct bnge_dev *bd = bn->bd;
+
+	strscpy(info->driver, DRV_NAME, sizeof(info->driver));
+	strscpy(info->fw_version, bd->fw_ver_str, sizeof(info->fw_version));
+	strscpy(info->bus_info, pci_name(bd->pdev), sizeof(info->bus_info));
+}
+
+static const struct ethtool_ops bnge_ethtool_ops = {
+	.get_drvinfo		= bnge_get_drvinfo,
+};
+
+void bnge_set_ethtool_ops(struct net_device *dev)
+{
+	dev->ethtool_ops = &bnge_ethtool_ops;
+}
diff --git a/drivers/net/ethernet/broadcom/bnge/bnge_ethtool.h b/drivers/net/ethernet/broadcom/bnge/bnge_ethtool.h
new file mode 100644
index 000000000000..21e96a0976d5
--- /dev/null
+++ b/drivers/net/ethernet/broadcom/bnge/bnge_ethtool.h
@@ -0,0 +1,9 @@
+/* SPDX-License-Identifier: GPL-2.0 */
+/* Copyright (c) 2025 Broadcom */
+
+#ifndef _BNGE_ETHTOOL_H_
+#define _BNGE_ETHTOOL_H_
+
+void bnge_set_ethtool_ops(struct net_device *dev);
+
+#endif /* _BNGE_ETHTOOL_H_ */
diff --git a/drivers/net/ethernet/broadcom/bnge/bnge_hwrm.c b/drivers/net/ethernet/broadcom/bnge/bnge_hwrm.c
new file mode 100644
index 000000000000..0f971af24142
--- /dev/null
+++ b/drivers/net/ethernet/broadcom/bnge/bnge_hwrm.c
@@ -0,0 +1,508 @@
+// SPDX-License-Identifier: GPL-2.0
+// Copyright (c) 2025 Broadcom.
+
+#include <asm/byteorder.h>
+#include <linux/dma-mapping.h>
+#include <linux/dmapool.h>
+#include <linux/delay.h>
+#include <linux/errno.h>
+#include <linux/kernel.h>
+#include <linux/list.h>
+#include <linux/pci.h>
+
+#include "bnge.h"
+#include "bnge_hwrm.h"
+
+static u64 bnge_cal_sentinel(struct bnge_hwrm_ctx *ctx, u16 req_type)
+{
+	return (((uintptr_t)ctx) + req_type) ^ BNGE_HWRM_SENTINEL;
+}
+
+int bnge_hwrm_req_create(struct bnge_dev *bd, void **req, u16 req_type,
+			 u32 req_len)
+{
+	struct bnge_hwrm_ctx *ctx;
+	dma_addr_t dma_handle;
+	u8 *req_addr;
+
+	if (req_len > BNGE_HWRM_CTX_OFFSET)
+		return -E2BIG;
+
+	req_addr = dma_pool_alloc(bd->hwrm_dma_pool, GFP_KERNEL | __GFP_ZERO,
+				  &dma_handle);
+	if (!req_addr)
+		return -ENOMEM;
+
+	ctx = (struct bnge_hwrm_ctx *)(req_addr + BNGE_HWRM_CTX_OFFSET);
+	/* safety first, sentinel used to check for invalid requests */
+	ctx->sentinel = bnge_cal_sentinel(ctx, req_type);
+	ctx->req_len = req_len;
+	ctx->req = (struct input *)req_addr;
+	ctx->resp = (struct output *)(req_addr + BNGE_HWRM_RESP_OFFSET);
+	ctx->dma_handle = dma_handle;
+	ctx->flags = 0; /* __GFP_ZERO, but be explicit regarding ownership */
+	ctx->timeout = bd->hwrm_cmd_timeout ?: BNGE_DFLT_HWRM_CMD_TIMEOUT;
+	ctx->allocated = BNGE_HWRM_DMA_SIZE - BNGE_HWRM_CTX_OFFSET;
+	ctx->gfp = GFP_KERNEL;
+	ctx->slice_addr = NULL;
+
+	/* initialize common request fields */
+	ctx->req->req_type = cpu_to_le16(req_type);
+	ctx->req->resp_addr = cpu_to_le64(dma_handle + BNGE_HWRM_RESP_OFFSET);
+	ctx->req->cmpl_ring = cpu_to_le16(BNGE_HWRM_NO_CMPL_RING);
+	ctx->req->target_id = cpu_to_le16(BNGE_HWRM_TARGET);
+	*req = ctx->req;
+
+	return 0;
+}
+
+static struct bnge_hwrm_ctx *__hwrm_ctx_get(struct bnge_dev *bd, u8 *req_addr)
+{
+	void *ctx_addr = req_addr + BNGE_HWRM_CTX_OFFSET;
+	struct input *req = (struct input *)req_addr;
+	struct bnge_hwrm_ctx *ctx = ctx_addr;
+	u64 sentinel;
+
+	if (!req) {
+		dev_err(bd->dev, "null HWRM request");
+		dump_stack();
+		return NULL;
+	}
+
+	/* HWRM API has no type safety, verify sentinel to validate address */
+	sentinel = bnge_cal_sentinel(ctx, le16_to_cpu(req->req_type));
+	if (ctx->sentinel != sentinel) {
+		dev_err(bd->dev, "HWRM sentinel mismatch, req_type = %u\n",
+			(u32)le16_to_cpu(req->req_type));
+		dump_stack();
+		return NULL;
+	}
+
+	return ctx;
+}
+
+void bnge_hwrm_req_timeout(struct bnge_dev *bd,
+			   void *req, unsigned int timeout)
+{
+	struct bnge_hwrm_ctx *ctx = __hwrm_ctx_get(bd, req);
+
+	if (ctx)
+		ctx->timeout = timeout;
+}
+
+void bnge_hwrm_req_alloc_flags(struct bnge_dev *bd, void *req, gfp_t gfp)
+{
+	struct bnge_hwrm_ctx *ctx = __hwrm_ctx_get(bd, req);
+
+	if (ctx)
+		ctx->gfp = gfp;
+}
+
+void bnge_hwrm_req_flags(struct bnge_dev *bd, void *req,
+			 enum bnge_hwrm_ctx_flags flags)
+{
+	struct bnge_hwrm_ctx *ctx = __hwrm_ctx_get(bd, req);
+
+	if (ctx)
+		ctx->flags |= (flags & BNGE_HWRM_API_FLAGS);
+}
+
+void *bnge_hwrm_req_hold(struct bnge_dev *bd, void *req)
+{
+	struct bnge_hwrm_ctx *ctx = __hwrm_ctx_get(bd, req);
+	struct input *input = (struct input *)req;
+
+	if (!ctx)
+		return NULL;
+
+	if (ctx->flags & BNGE_HWRM_INTERNAL_CTX_OWNED) {
+		dev_err(bd->dev, "HWRM context already owned, req_type = %u\n",
+			(u32)le16_to_cpu(input->req_type));
+		dump_stack();
+		return NULL;
+	}
+
+	ctx->flags |= BNGE_HWRM_INTERNAL_CTX_OWNED;
+	return ((u8 *)req) + BNGE_HWRM_RESP_OFFSET;
+}
+
+static void __hwrm_ctx_invalidate(struct bnge_dev *bd,
+				  struct bnge_hwrm_ctx *ctx)
+{
+	void *addr = ((u8 *)ctx) - BNGE_HWRM_CTX_OFFSET;
+	dma_addr_t dma_handle = ctx->dma_handle; /* save before invalidate */
+
+	/* unmap any auxiliary DMA slice */
+	if (ctx->slice_addr)
+		dma_free_coherent(bd->dev, ctx->slice_size,
+				  ctx->slice_addr, ctx->slice_handle);
+
+	/* invalidate, ensure ownership, sentinel and dma_handle are cleared */
+	memset(ctx, 0, sizeof(struct bnge_hwrm_ctx));
+
+	/* return the buffer to the DMA pool */
+	if (dma_handle)
+		dma_pool_free(bd->hwrm_dma_pool, addr, dma_handle);
+}
+
+void bnge_hwrm_req_drop(struct bnge_dev *bd, void *req)
+{
+	struct bnge_hwrm_ctx *ctx = __hwrm_ctx_get(bd, req);
+
+	if (ctx)
+		__hwrm_ctx_invalidate(bd, ctx);
+}
+
+static int bnge_map_hwrm_error(u32 hwrm_err)
+{
+	switch (hwrm_err) {
+	case HWRM_ERR_CODE_SUCCESS:
+		return 0;
+	case HWRM_ERR_CODE_RESOURCE_LOCKED:
+		return -EROFS;
+	case HWRM_ERR_CODE_RESOURCE_ACCESS_DENIED:
+		return -EACCES;
+	case HWRM_ERR_CODE_RESOURCE_ALLOC_ERROR:
+		return -ENOSPC;
+	case HWRM_ERR_CODE_INVALID_PARAMS:
+	case HWRM_ERR_CODE_INVALID_FLAGS:
+	case HWRM_ERR_CODE_INVALID_ENABLES:
+	case HWRM_ERR_CODE_UNSUPPORTED_TLV:
+	case HWRM_ERR_CODE_UNSUPPORTED_OPTION_ERR:
+		return -EINVAL;
+	case HWRM_ERR_CODE_NO_BUFFER:
+		return -ENOMEM;
+	case HWRM_ERR_CODE_HOT_RESET_PROGRESS:
+	case HWRM_ERR_CODE_BUSY:
+		return -EAGAIN;
+	case HWRM_ERR_CODE_CMD_NOT_SUPPORTED:
+		return -EOPNOTSUPP;
+	case HWRM_ERR_CODE_PF_UNAVAILABLE:
+		return -ENODEV;
+	default:
+		return -EIO;
+	}
+}
+
+static struct bnge_hwrm_wait_token *
+bnge_hwrm_create_token(struct bnge_dev *bd, enum bnge_hwrm_chnl dst)
+{
+	struct bnge_hwrm_wait_token *token;
+
+	token = kzalloc(sizeof(*token), GFP_KERNEL);
+	if (!token)
+		return NULL;
+
+	mutex_lock(&bd->hwrm_cmd_lock);
+
+	token->dst = dst;
+	token->state = BNGE_HWRM_PENDING;
+	if (dst == BNGE_HWRM_CHNL_CHIMP) {
+		token->seq_id = bd->hwrm_cmd_seq++;
+		hlist_add_head_rcu(&token->node, &bd->hwrm_pending_list);
+	} else {
+		token->seq_id = bd->hwrm_cmd_kong_seq++;
+	}
+
+	return token;
+}
+
+static void
+bnge_hwrm_destroy_token(struct bnge_dev *bd, struct bnge_hwrm_wait_token *token)
+{
+	if (token->dst == BNGE_HWRM_CHNL_CHIMP) {
+		hlist_del_rcu(&token->node);
+		kfree_rcu(token, rcu);
+	} else {
+		kfree(token);
+	}
+	mutex_unlock(&bd->hwrm_cmd_lock);
+}
+
+static void bnge_hwrm_req_dbg(struct bnge_dev *bd, struct input *req)
+{
+	u32 ring = le16_to_cpu(req->cmpl_ring);
+	u32 type = le16_to_cpu(req->req_type);
+	u32 tgt = le16_to_cpu(req->target_id);
+	u32 seq = le16_to_cpu(req->seq_id);
+	char opt[32] = "\n";
+
+	if (unlikely(ring != (u16)BNGE_HWRM_NO_CMPL_RING))
+		snprintf(opt, 16, " ring %d\n", ring);
+
+	if (unlikely(tgt != BNGE_HWRM_TARGET))
+		snprintf(opt + strlen(opt) - 1, 16, " tgt 0x%x\n", tgt);
+
+	dev_dbg(bd->dev, "sent hwrm req_type 0x%x seq id 0x%x%s",
+		type, seq, opt);
+}
+
+#define bnge_hwrm_err(bd, ctx, fmt, ...)		\
+	do {							       \
+		if ((ctx)->flags & BNGE_HWRM_CTX_SILENT)	       \
+			dev_dbg((bd)->dev, fmt, __VA_ARGS__);       \
+		else						       \
+			dev_err((bd)->dev, fmt, __VA_ARGS__);       \
+	} while (0)
+
+static int __hwrm_send_ctx(struct bnge_dev *bd, struct bnge_hwrm_ctx *ctx)
+{
+	u32 doorbell_offset = BNGE_GRCPF_REG_CHIMP_COMM_TRIGGER;
+	enum bnge_hwrm_chnl dst = BNGE_HWRM_CHNL_CHIMP;
+	u32 bar_offset = BNGE_GRCPF_REG_CHIMP_COMM;
+	struct bnge_hwrm_wait_token *token = NULL;
+	u16 max_req_len = BNGE_HWRM_MAX_REQ_LEN;
+	unsigned int i, timeout, tmo_count;
+	u32 *data = (u32 *)ctx->req;
+	u32 msg_len = ctx->req_len;
+	int rc = -EBUSY;
+	u32 req_type;
+	u16 len = 0;
+	u8 *valid;
+
+	if (ctx->flags & BNGE_HWRM_INTERNAL_RESP_DIRTY)
+		memset(ctx->resp, 0, PAGE_SIZE);
+
+	req_type = le16_to_cpu(ctx->req->req_type);
+
+	if (msg_len > BNGE_HWRM_MAX_REQ_LEN &&
+	    msg_len > bd->hwrm_max_ext_req_len) {
+		dev_warn(bd->dev, "oversized hwrm request, req_type 0x%x",
+			 req_type);
+		rc = -E2BIG;
+		goto exit;
+	}
+
+	token = bnge_hwrm_create_token(bd, dst);
+	if (!token) {
+		rc = -ENOMEM;
+		goto exit;
+	}
+	ctx->req->seq_id = cpu_to_le16(token->seq_id);
+
+	/* Ensure any associated DMA buffers are written before doorbell */
+	wmb();
+
+	/* Write request msg to hwrm channel */
+	__iowrite32_copy(bd->bar0 + bar_offset, data, msg_len / 4);
+
+	for (i = msg_len; i < max_req_len; i += 4)
+		writel(0, bd->bar0 + bar_offset + i);
+
+	/* Ring channel doorbell */
+	writel(1, bd->bar0 + doorbell_offset);
+
+	bnge_hwrm_req_dbg(bd, ctx->req);
+
+	/* Limit timeout to an upper limit */
+	timeout = min(ctx->timeout,
+		      bd->hwrm_cmd_max_timeout ?: BNGE_HWRM_CMD_MAX_TIMEOUT);
+	/* convert timeout to usec */
+	timeout *= 1000;
+
+	i = 0;
+	/* Short timeout for the first few iterations:
+	 * number of loops = number of loops for short timeout +
+	 * number of loops for standard timeout.
+	 */
+	tmo_count = BNGE_HWRM_SHORT_TIMEOUT_COUNTER;
+	timeout = timeout - BNGE_HWRM_SHORT_MIN_TIMEOUT *
+			BNGE_HWRM_SHORT_TIMEOUT_COUNTER;
+	tmo_count += DIV_ROUND_UP(timeout, BNGE_HWRM_MIN_TIMEOUT);
+
+	if (le16_to_cpu(ctx->req->cmpl_ring) != INVALID_HW_RING_ID) {
+		/* Wait until hwrm response cmpl interrupt is processed */
+		while (READ_ONCE(token->state) < BNGE_HWRM_COMPLETE &&
+		       i++ < tmo_count) {
+			/* on first few passes, just barely sleep */
+			if (i < BNGE_HWRM_SHORT_TIMEOUT_COUNTER) {
+				usleep_range(BNGE_HWRM_SHORT_MIN_TIMEOUT,
+					     BNGE_HWRM_SHORT_MAX_TIMEOUT);
+			} else {
+				usleep_range(BNGE_HWRM_MIN_TIMEOUT,
+					     BNGE_HWRM_MAX_TIMEOUT);
+			}
+		}
+
+		if (READ_ONCE(token->state) != BNGE_HWRM_COMPLETE) {
+			bnge_hwrm_err(bd, ctx, "No hwrm cmpl received: 0x%x\n",
+				      req_type);
+			goto exit;
+		}
+		len = le16_to_cpu(READ_ONCE(ctx->resp->resp_len));
+		valid = ((u8 *)ctx->resp) + len - 1;
+	} else {
+		__le16 seen_out_of_seq = ctx->req->seq_id; /* will never see */
+		int j;
+
+		/* Check if response len is updated */
+		for (i = 0; i < tmo_count; i++) {
+			if (token &&
+			    READ_ONCE(token->state) == BNGE_HWRM_DEFERRED) {
+				bnge_hwrm_destroy_token(bd, token);
+				token = NULL;
+			}
+
+			len = le16_to_cpu(READ_ONCE(ctx->resp->resp_len));
+			if (len) {
+				__le16 resp_seq = READ_ONCE(ctx->resp->seq_id);
+
+				if (resp_seq == ctx->req->seq_id)
+					break;
+				if (resp_seq != seen_out_of_seq) {
+					dev_warn(bd->dev, "Discarding out of seq response: 0x%x for msg {0x%x 0x%x}\n",
+						 le16_to_cpu(resp_seq), req_type, le16_to_cpu(ctx->req->seq_id));
+					seen_out_of_seq = resp_seq;
+				}
+			}
+
+			/* on first few passes, just barely sleep */
+			if (i < BNGE_HWRM_SHORT_TIMEOUT_COUNTER) {
+				usleep_range(BNGE_HWRM_SHORT_MIN_TIMEOUT,
+					     BNGE_HWRM_SHORT_MAX_TIMEOUT);
+			} else {
+				usleep_range(BNGE_HWRM_MIN_TIMEOUT,
+					     BNGE_HWRM_MAX_TIMEOUT);
+			}
+		}
+
+		if (i >= tmo_count) {
+			bnge_hwrm_err(bd, ctx,
+				      "Error (timeout: %u) msg {0x%x 0x%x} len:%d\n",
+				      bnge_hwrm_timeout(i), req_type,
+				      le16_to_cpu(ctx->req->seq_id), len);
+			goto exit;
+		}
+
+		/* Last byte of resp contains valid bit */
+		valid = ((u8 *)ctx->resp) + len - 1;
+		for (j = 0; j < BNGE_HWRM_FIN_WAIT_USEC; ) {
+			/* make sure we read from updated DMA memory */
+			dma_rmb();
+			if (*valid)
+				break;
+			if (j < 10) {
+				udelay(1);
+				j++;
+			} else {
+				usleep_range(20, 30);
+				j += 20;
+			}
+		}
+
+		if (j >= BNGE_HWRM_FIN_WAIT_USEC) {
+			bnge_hwrm_err(bd, ctx, "Error (timeout: %u) msg {0x%x 0x%x} len:%d v:%d\n",
+				      bnge_hwrm_timeout(i) + j, req_type,
+				      le16_to_cpu(ctx->req->seq_id), len, *valid);
+			goto exit;
+		}
+	}
+
+	/* Zero valid bit for compatibility.  Valid bit in an older spec
+	 * may become a new field in a newer spec.  We must make sure that
+	 * a new field not implemented by old spec will read zero.
+	 */
+	*valid = 0;
+	rc = le16_to_cpu(ctx->resp->error_code);
+	if (rc == HWRM_ERR_CODE_BUSY && !(ctx->flags & BNGE_HWRM_CTX_SILENT))
+		dev_warn(bd->dev, "FW returned busy, hwrm req_type 0x%x\n",
+			 req_type);
+	else if (rc && rc != HWRM_ERR_CODE_PF_UNAVAILABLE)
+		bnge_hwrm_err(bd, ctx, "hwrm req_type 0x%x seq id 0x%x error %d\n",
+			      req_type, le16_to_cpu(ctx->req->seq_id), rc);
+	rc = bnge_map_hwrm_error(rc);
+
+exit:
+	if (token)
+		bnge_hwrm_destroy_token(bd, token);
+	if (ctx->flags & BNGE_HWRM_INTERNAL_CTX_OWNED)
+		ctx->flags |= BNGE_HWRM_INTERNAL_RESP_DIRTY;
+	else
+		__hwrm_ctx_invalidate(bd, ctx);
+	return rc;
+}
+
+int bnge_hwrm_req_send(struct bnge_dev *bd, void *req)
+{
+	struct bnge_hwrm_ctx *ctx = __hwrm_ctx_get(bd, req);
+
+	if (!ctx)
+		return -EINVAL;
+
+	return __hwrm_send_ctx(bd, ctx);
+}
+
+int bnge_hwrm_req_send_silent(struct bnge_dev *bd, void *req)
+{
+	bnge_hwrm_req_flags(bd, req, BNGE_HWRM_CTX_SILENT);
+	return bnge_hwrm_req_send(bd, req);
+}
+
+void *
+bnge_hwrm_req_dma_slice(struct bnge_dev *bd, void *req, u32 size,
+			dma_addr_t *dma_handle)
+{
+	struct bnge_hwrm_ctx *ctx = __hwrm_ctx_get(bd, req);
+	u8 *end = ((u8 *)req) + BNGE_HWRM_DMA_SIZE;
+	struct input *input = req;
+	u8 *addr, *req_addr = req;
+	u32 max_offset, offset;
+
+	if (!ctx)
+		return NULL;
+
+	max_offset = BNGE_HWRM_DMA_SIZE - ctx->allocated;
+	offset = max_offset - size;
+	offset = ALIGN_DOWN(offset, BNGE_HWRM_DMA_ALIGN);
+	addr = req_addr + offset;
+
+	if (addr < req_addr + max_offset && req_addr + ctx->req_len <= addr) {
+		ctx->allocated = end - addr;
+		*dma_handle = ctx->dma_handle + offset;
+		return addr;
+	}
+
+	if (ctx->slice_addr) {
+		dev_err(bd->dev, "HWRM refusing to reallocate DMA slice, req_type = %u\n",
+			(u32)le16_to_cpu(input->req_type));
+		dump_stack();
+		return NULL;
+	}
+
+	addr = dma_alloc_coherent(bd->dev, size, dma_handle, ctx->gfp);
+	if (!addr)
+		return NULL;
+
+	ctx->slice_addr = addr;
+	ctx->slice_size = size;
+	ctx->slice_handle = *dma_handle;
+
+	return addr;
+}
+
+void bnge_cleanup_hwrm_resources(struct bnge_dev *bd)
+{
+	struct bnge_hwrm_wait_token *token;
+
+	dma_pool_destroy(bd->hwrm_dma_pool);
+	bd->hwrm_dma_pool = NULL;
+
+	rcu_read_lock();
+	hlist_for_each_entry_rcu(token, &bd->hwrm_pending_list, node)
+		WRITE_ONCE(token->state, BNGE_HWRM_CANCELLED);
+	rcu_read_unlock();
+}
+
+int bnge_init_hwrm_resources(struct bnge_dev *bd)
+{
+	bd->hwrm_dma_pool = dma_pool_create("bnge_hwrm", bd->dev,
+					    BNGE_HWRM_DMA_SIZE,
+					    BNGE_HWRM_DMA_ALIGN, 0);
+	if (!bd->hwrm_dma_pool)
+		return -ENOMEM;
+
+	INIT_HLIST_HEAD(&bd->hwrm_pending_list);
+	mutex_init(&bd->hwrm_cmd_lock);
+
+	return 0;
+}
diff --git a/drivers/net/ethernet/broadcom/bnge/bnge_hwrm.h b/drivers/net/ethernet/broadcom/bnge/bnge_hwrm.h
new file mode 100644
index 000000000000..83794a12cc81
--- /dev/null
+++ b/drivers/net/ethernet/broadcom/bnge/bnge_hwrm.h
@@ -0,0 +1,110 @@
+/* SPDX-License-Identifier: GPL-2.0 */
+/* Copyright (c) 2025 Broadcom */
+
+#ifndef _BNGE_HWRM_H_
+#define _BNGE_HWRM_H_
+
+#include <linux/bnxt/hsi.h>
+
+enum bnge_hwrm_ctx_flags {
+	BNGE_HWRM_INTERNAL_CTX_OWNED	= BIT(0),
+	BNGE_HWRM_INTERNAL_RESP_DIRTY	= BIT(1),
+	BNGE_HWRM_CTX_SILENT		= BIT(2),
+	BNGE_HWRM_FULL_WAIT		= BIT(3),
+};
+
+#define BNGE_HWRM_API_FLAGS (BNGE_HWRM_CTX_SILENT | BNGE_HWRM_FULL_WAIT)
+
+struct bnge_hwrm_ctx {
+	u64 sentinel;
+	dma_addr_t dma_handle;
+	struct output *resp;
+	struct input *req;
+	dma_addr_t slice_handle;
+	void *slice_addr;
+	u32 slice_size;
+	u32 req_len;
+	enum bnge_hwrm_ctx_flags flags;
+	unsigned int timeout;
+	u32 allocated;
+	gfp_t gfp;
+};
+
+enum bnge_hwrm_wait_state {
+	BNGE_HWRM_PENDING,
+	BNGE_HWRM_DEFERRED,
+	BNGE_HWRM_COMPLETE,
+	BNGE_HWRM_CANCELLED,
+};
+
+enum bnge_hwrm_chnl { BNGE_HWRM_CHNL_CHIMP, BNGE_HWRM_CHNL_KONG };
+
+struct bnge_hwrm_wait_token {
+	struct rcu_head rcu;
+	struct hlist_node node;
+	enum bnge_hwrm_wait_state state;
+	enum bnge_hwrm_chnl dst;
+	u16 seq_id;
+};
+
+#define BNGE_DFLT_HWRM_CMD_TIMEOUT		500
+
+#define BNGE_GRCPF_REG_CHIMP_COMM		0x0
+#define BNGE_GRCPF_REG_CHIMP_COMM_TRIGGER	0x100
+
+#define BNGE_HWRM_MAX_REQ_LEN		(bd->hwrm_max_req_len)
+#define BNGE_HWRM_SHORT_REQ_LEN		sizeof(struct hwrm_short_input)
+#define BNGE_HWRM_CMD_MAX_TIMEOUT	40000U
+#define BNGE_SHORT_HWRM_CMD_TIMEOUT	20
+#define BNGE_HWRM_CMD_TIMEOUT		(bd->hwrm_cmd_timeout)
+#define BNGE_HWRM_RESET_TIMEOUT		((BNGE_HWRM_CMD_TIMEOUT) * 4)
+#define BNGE_HWRM_TARGET		0xffff
+#define BNGE_HWRM_NO_CMPL_RING		-1
+#define BNGE_HWRM_REQ_MAX_SIZE		128
+#define BNGE_HWRM_DMA_SIZE		(2 * PAGE_SIZE) /* space for req+resp */
+#define BNGE_HWRM_RESP_RESERVED		PAGE_SIZE
+#define BNGE_HWRM_RESP_OFFSET		(BNGE_HWRM_DMA_SIZE -		\
+					 BNGE_HWRM_RESP_RESERVED)
+#define BNGE_HWRM_CTX_OFFSET		(BNGE_HWRM_RESP_OFFSET -	\
+					 sizeof(struct bnge_hwrm_ctx))
+#define BNGE_HWRM_DMA_ALIGN		16
+#define BNGE_HWRM_SENTINEL		0xb6e1f68a12e9a7eb /* arbitrary value */
+#define BNGE_HWRM_SHORT_MIN_TIMEOUT		3
+#define BNGE_HWRM_SHORT_MAX_TIMEOUT		10
+#define BNGE_HWRM_SHORT_TIMEOUT_COUNTER		5
+
+#define BNGE_HWRM_MIN_TIMEOUT		25
+#define BNGE_HWRM_MAX_TIMEOUT		40
+
+static inline unsigned int bnge_hwrm_timeout(unsigned int n)
+{
+	return n <= BNGE_HWRM_SHORT_TIMEOUT_COUNTER ?
+		n * BNGE_HWRM_SHORT_MIN_TIMEOUT :
+		BNGE_HWRM_SHORT_TIMEOUT_COUNTER *
+			BNGE_HWRM_SHORT_MIN_TIMEOUT +
+			(n - BNGE_HWRM_SHORT_TIMEOUT_COUNTER) *
+				BNGE_HWRM_MIN_TIMEOUT;
+}
+
+#define BNGE_HWRM_FIN_WAIT_USEC	50000
+
+void bnge_cleanup_hwrm_resources(struct bnge_dev *bd);
+int bnge_init_hwrm_resources(struct bnge_dev *bd);
+
+int bnge_hwrm_req_create(struct bnge_dev *bd, void **req, u16 req_type,
+			 u32 req_len);
+#define bnge_hwrm_req_init(bd, req, req_type) \
+	bnge_hwrm_req_create((bd), (void **)&(req), (req_type),	\
+			     sizeof(*(req)))
+void *bnge_hwrm_req_hold(struct bnge_dev *bd, void *req);
+void bnge_hwrm_req_drop(struct bnge_dev *bd, void *req);
+void bnge_hwrm_req_flags(struct bnge_dev *bd, void *req,
+			 enum bnge_hwrm_ctx_flags flags);
+void bnge_hwrm_req_timeout(struct bnge_dev *bd, void *req,
+			   unsigned int timeout);
+int bnge_hwrm_req_send(struct bnge_dev *bd, void *req);
+int bnge_hwrm_req_send_silent(struct bnge_dev *bd, void *req);
+void bnge_hwrm_req_alloc_flags(struct bnge_dev *bd, void *req, gfp_t flags);
+void *bnge_hwrm_req_dma_slice(struct bnge_dev *bd, void *req, u32 size,
+			      dma_addr_t *dma);
+#endif /* _BNGE_HWRM_H_ */
diff --git a/drivers/net/ethernet/broadcom/bnge/bnge_hwrm_lib.c b/drivers/net/ethernet/broadcom/bnge/bnge_hwrm_lib.c
new file mode 100644
index 000000000000..198f49b40dbf
--- /dev/null
+++ b/drivers/net/ethernet/broadcom/bnge/bnge_hwrm_lib.c
@@ -0,0 +1,1185 @@
+// SPDX-License-Identifier: GPL-2.0
+// Copyright (c) 2025 Broadcom.
+
+#include <linux/errno.h>
+#include <linux/kernel.h>
+#include <linux/mm.h>
+#include <linux/pci.h>
+#include <linux/bnxt/hsi.h>
+#include <linux/if_vlan.h>
+#include <net/netdev_queues.h>
+
+#include "bnge.h"
+#include "bnge_hwrm.h"
+#include "bnge_hwrm_lib.h"
+#include "bnge_rmem.h"
+#include "bnge_resc.h"
+
+int bnge_hwrm_ver_get(struct bnge_dev *bd)
+{
+	u32 dev_caps_cfg, hwrm_ver, hwrm_spec_code;
+	u16 fw_maj, fw_min, fw_bld, fw_rsv;
+	struct hwrm_ver_get_output *resp;
+	struct hwrm_ver_get_input *req;
+	int rc;
+
+	rc = bnge_hwrm_req_init(bd, req, HWRM_VER_GET);
+	if (rc)
+		return rc;
+
+	bnge_hwrm_req_flags(bd, req, BNGE_HWRM_FULL_WAIT);
+	bd->hwrm_max_req_len = HWRM_MAX_REQ_LEN;
+	req->hwrm_intf_maj = HWRM_VERSION_MAJOR;
+	req->hwrm_intf_min = HWRM_VERSION_MINOR;
+	req->hwrm_intf_upd = HWRM_VERSION_UPDATE;
+
+	resp = bnge_hwrm_req_hold(bd, req);
+	rc = bnge_hwrm_req_send(bd, req);
+	if (rc)
+		goto hwrm_ver_get_exit;
+
+	memcpy(&bd->ver_resp, resp, sizeof(struct hwrm_ver_get_output));
+
+	hwrm_spec_code = resp->hwrm_intf_maj_8b << 16 |
+			 resp->hwrm_intf_min_8b << 8 |
+			 resp->hwrm_intf_upd_8b;
+	hwrm_ver = HWRM_VERSION_MAJOR << 16 | HWRM_VERSION_MINOR << 8 |
+			HWRM_VERSION_UPDATE;
+
+	if (hwrm_spec_code > hwrm_ver)
+		snprintf(bd->hwrm_ver_supp, FW_VER_STR_LEN, "%d.%d.%d",
+			 HWRM_VERSION_MAJOR, HWRM_VERSION_MINOR,
+			 HWRM_VERSION_UPDATE);
+	else
+		snprintf(bd->hwrm_ver_supp, FW_VER_STR_LEN, "%d.%d.%d",
+			 resp->hwrm_intf_maj_8b, resp->hwrm_intf_min_8b,
+			 resp->hwrm_intf_upd_8b);
+
+	fw_maj = le16_to_cpu(resp->hwrm_fw_major);
+	fw_min = le16_to_cpu(resp->hwrm_fw_minor);
+	fw_bld = le16_to_cpu(resp->hwrm_fw_build);
+	fw_rsv = le16_to_cpu(resp->hwrm_fw_patch);
+
+	bd->fw_ver_code = BNGE_FW_VER_CODE(fw_maj, fw_min, fw_bld, fw_rsv);
+	snprintf(bd->fw_ver_str, FW_VER_STR_LEN, "%d.%d.%d.%d",
+		 fw_maj, fw_min, fw_bld, fw_rsv);
+
+	if (strlen(resp->active_pkg_name)) {
+		int fw_ver_len = strlen(bd->fw_ver_str);
+
+		snprintf(bd->fw_ver_str + fw_ver_len,
+			 FW_VER_STR_LEN - fw_ver_len - 1, "/pkg %s",
+			 resp->active_pkg_name);
+		bd->fw_cap |= BNGE_FW_CAP_PKG_VER;
+	}
+
+	bd->hwrm_cmd_timeout = le16_to_cpu(resp->def_req_timeout);
+	if (!bd->hwrm_cmd_timeout)
+		bd->hwrm_cmd_timeout = BNGE_DFLT_HWRM_CMD_TIMEOUT;
+	bd->hwrm_cmd_max_timeout = le16_to_cpu(resp->max_req_timeout) * 1000;
+	if (!bd->hwrm_cmd_max_timeout)
+		bd->hwrm_cmd_max_timeout = BNGE_HWRM_CMD_MAX_TIMEOUT;
+	else if (bd->hwrm_cmd_max_timeout > BNGE_HWRM_CMD_MAX_TIMEOUT)
+		dev_warn(bd->dev, "Default HWRM commands max timeout increased to %d seconds\n",
+			 bd->hwrm_cmd_max_timeout / 1000);
+
+	bd->hwrm_max_req_len = le16_to_cpu(resp->max_req_win_len);
+	bd->hwrm_max_ext_req_len = le16_to_cpu(resp->max_ext_req_len);
+
+	if (bd->hwrm_max_ext_req_len < HWRM_MAX_REQ_LEN)
+		bd->hwrm_max_ext_req_len = HWRM_MAX_REQ_LEN;
+
+	bd->chip_num = le16_to_cpu(resp->chip_num);
+	bd->chip_rev = resp->chip_rev;
+
+	dev_caps_cfg = le32_to_cpu(resp->dev_caps_cfg);
+	if ((dev_caps_cfg & VER_GET_RESP_DEV_CAPS_CFG_SHORT_CMD_SUPPORTED) &&
+	    (dev_caps_cfg & VER_GET_RESP_DEV_CAPS_CFG_SHORT_CMD_REQUIRED))
+		bd->fw_cap |= BNGE_FW_CAP_SHORT_CMD;
+
+	if (dev_caps_cfg & VER_GET_RESP_DEV_CAPS_CFG_KONG_MB_CHNL_SUPPORTED)
+		bd->fw_cap |= BNGE_FW_CAP_KONG_MB_CHNL;
+
+	if (dev_caps_cfg &
+	    VER_GET_RESP_DEV_CAPS_CFG_CFA_ADV_FLOW_MGNT_SUPPORTED)
+		bd->fw_cap |= BNGE_FW_CAP_CFA_ADV_FLOW;
+
+hwrm_ver_get_exit:
+	bnge_hwrm_req_drop(bd, req);
+	return rc;
+}
+
+int
+bnge_hwrm_nvm_dev_info(struct bnge_dev *bd,
+		       struct hwrm_nvm_get_dev_info_output *nvm_info)
+{
+	struct hwrm_nvm_get_dev_info_output *resp;
+	struct hwrm_nvm_get_dev_info_input *req;
+	int rc;
+
+	rc = bnge_hwrm_req_init(bd, req, HWRM_NVM_GET_DEV_INFO);
+	if (rc)
+		return rc;
+
+	resp = bnge_hwrm_req_hold(bd, req);
+	rc = bnge_hwrm_req_send(bd, req);
+	if (!rc)
+		memcpy(nvm_info, resp, sizeof(*resp));
+	bnge_hwrm_req_drop(bd, req);
+	return rc;
+}
+
+int bnge_hwrm_func_reset(struct bnge_dev *bd)
+{
+	struct hwrm_func_reset_input *req;
+	int rc;
+
+	rc = bnge_hwrm_req_init(bd, req, HWRM_FUNC_RESET);
+	if (rc)
+		return rc;
+
+	req->enables = 0;
+	bnge_hwrm_req_timeout(bd, req, BNGE_HWRM_RESET_TIMEOUT);
+	return bnge_hwrm_req_send(bd, req);
+}
+
+int bnge_hwrm_fw_set_time(struct bnge_dev *bd)
+{
+	struct hwrm_fw_set_time_input *req;
+	struct tm tm;
+	int rc;
+
+	time64_to_tm(ktime_get_real_seconds(), 0, &tm);
+
+	rc = bnge_hwrm_req_init(bd, req, HWRM_FW_SET_TIME);
+	if (rc)
+		return rc;
+
+	req->year = cpu_to_le16(1900 + tm.tm_year);
+	req->month = 1 + tm.tm_mon;
+	req->day = tm.tm_mday;
+	req->hour = tm.tm_hour;
+	req->minute = tm.tm_min;
+	req->second = tm.tm_sec;
+	return bnge_hwrm_req_send(bd, req);
+}
+
+int bnge_hwrm_func_drv_rgtr(struct bnge_dev *bd)
+{
+	struct hwrm_func_drv_rgtr_output *resp;
+	struct hwrm_func_drv_rgtr_input *req;
+	u32 flags;
+	int rc;
+
+	rc = bnge_hwrm_req_init(bd, req, HWRM_FUNC_DRV_RGTR);
+	if (rc)
+		return rc;
+
+	req->enables = cpu_to_le32(FUNC_DRV_RGTR_REQ_ENABLES_OS_TYPE |
+				   FUNC_DRV_RGTR_REQ_ENABLES_VER |
+				   FUNC_DRV_RGTR_REQ_ENABLES_ASYNC_EVENT_FWD);
+
+	req->os_type = cpu_to_le16(FUNC_DRV_RGTR_REQ_OS_TYPE_LINUX);
+	flags = FUNC_DRV_RGTR_REQ_FLAGS_16BIT_VER_MODE;
+
+	req->flags = cpu_to_le32(flags);
+	req->ver_maj_8b = DRV_VER_MAJ;
+	req->ver_min_8b = DRV_VER_MIN;
+	req->ver_upd_8b = DRV_VER_UPD;
+	req->ver_maj = cpu_to_le16(DRV_VER_MAJ);
+	req->ver_min = cpu_to_le16(DRV_VER_MIN);
+	req->ver_upd = cpu_to_le16(DRV_VER_UPD);
+
+	resp = bnge_hwrm_req_hold(bd, req);
+	rc = bnge_hwrm_req_send(bd, req);
+	if (!rc) {
+		set_bit(BNGE_STATE_DRV_REGISTERED, &bd->state);
+		if (resp->flags &
+		    cpu_to_le32(FUNC_DRV_RGTR_RESP_FLAGS_IF_CHANGE_SUPPORTED))
+			bd->fw_cap |= BNGE_FW_CAP_IF_CHANGE;
+	}
+	bnge_hwrm_req_drop(bd, req);
+	return rc;
+}
+
+int bnge_hwrm_func_drv_unrgtr(struct bnge_dev *bd)
+{
+	struct hwrm_func_drv_unrgtr_input *req;
+	int rc;
+
+	if (!test_and_clear_bit(BNGE_STATE_DRV_REGISTERED, &bd->state))
+		return 0;
+
+	rc = bnge_hwrm_req_init(bd, req, HWRM_FUNC_DRV_UNRGTR);
+	if (rc)
+		return rc;
+	return bnge_hwrm_req_send(bd, req);
+}
+
+static void bnge_init_ctx_initializer(struct bnge_ctx_mem_type *ctxm,
+				      u8 init_val, u8 init_offset,
+				      bool init_mask_set)
+{
+	ctxm->init_value = init_val;
+	ctxm->init_offset = BNGE_CTX_INIT_INVALID_OFFSET;
+	if (init_mask_set)
+		ctxm->init_offset = init_offset * 4;
+	else
+		ctxm->init_value = 0;
+}
+
+static int bnge_alloc_all_ctx_pg_info(struct bnge_dev *bd, int ctx_max)
+{
+	struct bnge_ctx_mem_info *ctx = bd->ctx;
+	u16 type;
+
+	for (type = 0; type < ctx_max; type++) {
+		struct bnge_ctx_mem_type *ctxm = &ctx->ctx_arr[type];
+		int n = 1;
+
+		if (!ctxm->max_entries)
+			continue;
+
+		if (ctxm->instance_bmap)
+			n = hweight32(ctxm->instance_bmap);
+		ctxm->pg_info = kcalloc(n, sizeof(*ctxm->pg_info), GFP_KERNEL);
+		if (!ctxm->pg_info)
+			return -ENOMEM;
+	}
+
+	return 0;
+}
+
+#define BNGE_CTX_INIT_VALID(flags)	\
+	(!!((flags) &			\
+	    FUNC_BACKING_STORE_QCAPS_V2_RESP_FLAGS_ENABLE_CTX_KIND_INIT))
+
+int bnge_hwrm_func_backing_store_qcaps(struct bnge_dev *bd)
+{
+	struct hwrm_func_backing_store_qcaps_v2_output *resp;
+	struct hwrm_func_backing_store_qcaps_v2_input *req;
+	struct bnge_ctx_mem_info *ctx;
+	u16 type;
+	int rc;
+
+	if (bd->ctx)
+		return 0;
+
+	rc = bnge_hwrm_req_init(bd, req, HWRM_FUNC_BACKING_STORE_QCAPS_V2);
+	if (rc)
+		return rc;
+
+	ctx = kzalloc(sizeof(*ctx), GFP_KERNEL);
+	if (!ctx)
+		return -ENOMEM;
+	bd->ctx = ctx;
+
+	resp = bnge_hwrm_req_hold(bd, req);
+
+	for (type = 0; type < BNGE_CTX_V2_MAX; ) {
+		struct bnge_ctx_mem_type *ctxm = &ctx->ctx_arr[type];
+		u8 init_val, init_off, i;
+		__le32 *p;
+		u32 flags;
+
+		req->type = cpu_to_le16(type);
+		rc = bnge_hwrm_req_send(bd, req);
+		if (rc)
+			goto ctx_done;
+		flags = le32_to_cpu(resp->flags);
+		type = le16_to_cpu(resp->next_valid_type);
+		if (!(flags &
+		      FUNC_BACKING_STORE_QCAPS_V2_RESP_FLAGS_TYPE_VALID))
+			continue;
+
+		ctxm->type = le16_to_cpu(resp->type);
+		ctxm->entry_size = le16_to_cpu(resp->entry_size);
+		ctxm->flags = flags;
+		ctxm->instance_bmap = le32_to_cpu(resp->instance_bit_map);
+		ctxm->entry_multiple = resp->entry_multiple;
+		ctxm->max_entries = le32_to_cpu(resp->max_num_entries);
+		ctxm->min_entries = le32_to_cpu(resp->min_num_entries);
+		init_val = resp->ctx_init_value;
+		init_off = resp->ctx_init_offset;
+		bnge_init_ctx_initializer(ctxm, init_val, init_off,
+					  BNGE_CTX_INIT_VALID(flags));
+		ctxm->split_entry_cnt = min_t(u8, resp->subtype_valid_cnt,
+					      BNGE_MAX_SPLIT_ENTRY);
+		for (i = 0, p = &resp->split_entry_0; i < ctxm->split_entry_cnt;
+		     i++, p++)
+			ctxm->split[i] = le32_to_cpu(*p);
+	}
+	rc = bnge_alloc_all_ctx_pg_info(bd, BNGE_CTX_V2_MAX);
+
+ctx_done:
+	bnge_hwrm_req_drop(bd, req);
+	return rc;
+}
+
+static void bnge_hwrm_set_pg_attr(struct bnge_ring_mem_info *rmem, u8 *pg_attr,
+				  __le64 *pg_dir)
+{
+	if (!rmem->nr_pages)
+		return;
+
+	BNGE_SET_CTX_PAGE_ATTR(*pg_attr);
+	if (rmem->depth >= 1) {
+		if (rmem->depth == 2)
+			*pg_attr |= 2;
+		else
+			*pg_attr |= 1;
+		*pg_dir = cpu_to_le64(rmem->dma_pg_tbl);
+	} else {
+		*pg_dir = cpu_to_le64(rmem->dma_arr[0]);
+	}
+}
+
+int bnge_hwrm_func_backing_store(struct bnge_dev *bd,
+				 struct bnge_ctx_mem_type *ctxm,
+				 bool last)
+{
+	struct hwrm_func_backing_store_cfg_v2_input *req;
+	u32 instance_bmap = ctxm->instance_bmap;
+	int i, j, rc = 0, n = 1;
+	__le32 *p;
+
+	if (!(ctxm->flags & BNGE_CTX_MEM_TYPE_VALID) || !ctxm->pg_info)
+		return 0;
+
+	if (instance_bmap)
+		n = hweight32(ctxm->instance_bmap);
+	else
+		instance_bmap = 1;
+
+	rc = bnge_hwrm_req_init(bd, req, HWRM_FUNC_BACKING_STORE_CFG_V2);
+	if (rc)
+		return rc;
+	bnge_hwrm_req_hold(bd, req);
+	req->type = cpu_to_le16(ctxm->type);
+	req->entry_size = cpu_to_le16(ctxm->entry_size);
+	req->subtype_valid_cnt = ctxm->split_entry_cnt;
+	for (i = 0, p = &req->split_entry_0; i < ctxm->split_entry_cnt; i++)
+		p[i] = cpu_to_le32(ctxm->split[i]);
+	for (i = 0, j = 0; j < n && !rc; i++) {
+		struct bnge_ctx_pg_info *ctx_pg;
+
+		if (!(instance_bmap & (1 << i)))
+			continue;
+		req->instance = cpu_to_le16(i);
+		ctx_pg = &ctxm->pg_info[j++];
+		if (!ctx_pg->entries)
+			continue;
+		req->num_entries = cpu_to_le32(ctx_pg->entries);
+		bnge_hwrm_set_pg_attr(&ctx_pg->ring_mem,
+				      &req->page_size_pbl_level,
+				      &req->page_dir);
+		if (last && j == n)
+			req->flags =
+				cpu_to_le32(BNGE_BS_CFG_ALL_DONE);
+		rc = bnge_hwrm_req_send(bd, req);
+	}
+	bnge_hwrm_req_drop(bd, req);
+
+	return rc;
+}
+
+static int bnge_hwrm_get_rings(struct bnge_dev *bd)
+{
+	struct bnge_hw_resc *hw_resc = &bd->hw_resc;
+	struct hwrm_func_qcfg_output *resp;
+	struct hwrm_func_qcfg_input *req;
+	u16 cp, stats;
+	u16 rx, tx;
+	int rc;
+
+	rc = bnge_hwrm_req_init(bd, req, HWRM_FUNC_QCFG);
+	if (rc)
+		return rc;
+
+	req->fid = cpu_to_le16(0xffff);
+	resp = bnge_hwrm_req_hold(bd, req);
+	rc = bnge_hwrm_req_send(bd, req);
+	if (rc) {
+		bnge_hwrm_req_drop(bd, req);
+		return rc;
+	}
+
+	hw_resc->resv_tx_rings = le16_to_cpu(resp->alloc_tx_rings);
+	hw_resc->resv_rx_rings = le16_to_cpu(resp->alloc_rx_rings);
+	hw_resc->resv_hw_ring_grps =
+		le32_to_cpu(resp->alloc_hw_ring_grps);
+	hw_resc->resv_vnics = le16_to_cpu(resp->alloc_vnics);
+	hw_resc->resv_rsscos_ctxs = le16_to_cpu(resp->alloc_rsscos_ctx);
+	cp = le16_to_cpu(resp->alloc_cmpl_rings);
+	stats = le16_to_cpu(resp->alloc_stat_ctx);
+	hw_resc->resv_irqs = cp;
+	rx = hw_resc->resv_rx_rings;
+	tx = hw_resc->resv_tx_rings;
+	if (bnge_is_agg_reqd(bd))
+		rx >>= 1;
+	if (cp < (rx + tx)) {
+		rc = bnge_fix_rings_count(&rx, &tx, cp, false);
+		if (rc)
+			goto get_rings_exit;
+		if (bnge_is_agg_reqd(bd))
+			rx <<= 1;
+		hw_resc->resv_rx_rings = rx;
+		hw_resc->resv_tx_rings = tx;
+	}
+	hw_resc->resv_irqs = le16_to_cpu(resp->alloc_msix);
+	hw_resc->resv_hw_ring_grps = rx;
+	hw_resc->resv_cp_rings = cp;
+	hw_resc->resv_stat_ctxs = stats;
+
+get_rings_exit:
+	bnge_hwrm_req_drop(bd, req);
+	return rc;
+}
+
+static struct hwrm_func_cfg_input *
+__bnge_hwrm_reserve_pf_rings(struct bnge_dev *bd, struct bnge_hw_rings *hwr)
+{
+	struct hwrm_func_cfg_input *req;
+	u32 enables = 0;
+
+	if (bnge_hwrm_req_init(bd, req, HWRM_FUNC_QCFG))
+		return NULL;
+
+	req->fid = cpu_to_le16(0xffff);
+	enables |= hwr->tx ? FUNC_CFG_REQ_ENABLES_NUM_TX_RINGS : 0;
+	req->num_tx_rings = cpu_to_le16(hwr->tx);
+
+	enables |= hwr->rx ? FUNC_CFG_REQ_ENABLES_NUM_RX_RINGS : 0;
+	enables |= hwr->stat ? FUNC_CFG_REQ_ENABLES_NUM_STAT_CTXS : 0;
+	enables |= hwr->nq ? FUNC_CFG_REQ_ENABLES_NUM_MSIX : 0;
+	enables |= hwr->cmpl ? FUNC_CFG_REQ_ENABLES_NUM_CMPL_RINGS : 0;
+	enables |= hwr->vnic ? FUNC_CFG_REQ_ENABLES_NUM_VNICS : 0;
+	enables |= hwr->rss_ctx ? FUNC_CFG_REQ_ENABLES_NUM_RSSCOS_CTXS : 0;
+
+	req->num_rx_rings = cpu_to_le16(hwr->rx);
+	req->num_rsscos_ctxs = cpu_to_le16(hwr->rss_ctx);
+	req->num_cmpl_rings = cpu_to_le16(hwr->cmpl);
+	req->num_msix = cpu_to_le16(hwr->nq);
+	req->num_stat_ctxs = cpu_to_le16(hwr->stat);
+	req->num_vnics = cpu_to_le16(hwr->vnic);
+	req->enables = cpu_to_le32(enables);
+
+	return req;
+}
+
+static int
+bnge_hwrm_reserve_pf_rings(struct bnge_dev *bd, struct bnge_hw_rings *hwr)
+{
+	struct hwrm_func_cfg_input *req;
+	int rc;
+
+	req = __bnge_hwrm_reserve_pf_rings(bd, hwr);
+	if (!req)
+		return -ENOMEM;
+
+	if (!req->enables) {
+		bnge_hwrm_req_drop(bd, req);
+		return 0;
+	}
+
+	rc = bnge_hwrm_req_send(bd, req);
+	if (rc)
+		return rc;
+
+	return bnge_hwrm_get_rings(bd);
+}
+
+int bnge_hwrm_reserve_rings(struct bnge_dev *bd, struct bnge_hw_rings *hwr)
+{
+	return bnge_hwrm_reserve_pf_rings(bd, hwr);
+}
+
+int bnge_hwrm_func_qcfg(struct bnge_dev *bd)
+{
+	struct hwrm_func_qcfg_output *resp;
+	struct hwrm_func_qcfg_input *req;
+	int rc;
+
+	rc = bnge_hwrm_req_init(bd, req, HWRM_FUNC_QCFG);
+	if (rc)
+		return rc;
+
+	req->fid = cpu_to_le16(0xffff);
+	resp = bnge_hwrm_req_hold(bd, req);
+	rc = bnge_hwrm_req_send(bd, req);
+	if (rc)
+		goto func_qcfg_exit;
+
+	bd->max_mtu = le16_to_cpu(resp->max_mtu_configured);
+	if (!bd->max_mtu)
+		bd->max_mtu = BNGE_MAX_MTU;
+
+	if (bd->db_size)
+		goto func_qcfg_exit;
+
+	bd->db_offset = le16_to_cpu(resp->legacy_l2_db_size_kb) * 1024;
+	bd->db_size = PAGE_ALIGN(le16_to_cpu(resp->l2_doorbell_bar_size_kb) *
+			1024);
+	if (!bd->db_size || bd->db_size > pci_resource_len(bd->pdev, 2) ||
+	    bd->db_size <= bd->db_offset)
+		bd->db_size = pci_resource_len(bd->pdev, 2);
+
+func_qcfg_exit:
+	bnge_hwrm_req_drop(bd, req);
+	return rc;
+}
+
+int bnge_hwrm_func_resc_qcaps(struct bnge_dev *bd)
+{
+	struct hwrm_func_resource_qcaps_output *resp;
+	struct bnge_hw_resc *hw_resc = &bd->hw_resc;
+	struct hwrm_func_resource_qcaps_input *req;
+	int rc;
+
+	rc = bnge_hwrm_req_init(bd, req, HWRM_FUNC_RESOURCE_QCAPS);
+	if (rc)
+		return rc;
+
+	req->fid = cpu_to_le16(0xffff);
+	resp = bnge_hwrm_req_hold(bd, req);
+	rc = bnge_hwrm_req_send_silent(bd, req);
+	if (rc)
+		goto hwrm_func_resc_qcaps_exit;
+
+	hw_resc->max_tx_sch_inputs = le16_to_cpu(resp->max_tx_scheduler_inputs);
+	hw_resc->min_rsscos_ctxs = le16_to_cpu(resp->min_rsscos_ctx);
+	hw_resc->max_rsscos_ctxs = le16_to_cpu(resp->max_rsscos_ctx);
+	hw_resc->min_cp_rings = le16_to_cpu(resp->min_cmpl_rings);
+	hw_resc->max_cp_rings = le16_to_cpu(resp->max_cmpl_rings);
+	hw_resc->min_tx_rings = le16_to_cpu(resp->min_tx_rings);
+	hw_resc->max_tx_rings = le16_to_cpu(resp->max_tx_rings);
+	hw_resc->min_rx_rings = le16_to_cpu(resp->min_rx_rings);
+	hw_resc->max_rx_rings = le16_to_cpu(resp->max_rx_rings);
+	hw_resc->min_hw_ring_grps = le16_to_cpu(resp->min_hw_ring_grps);
+	hw_resc->max_hw_ring_grps = le16_to_cpu(resp->max_hw_ring_grps);
+	hw_resc->min_l2_ctxs = le16_to_cpu(resp->min_l2_ctxs);
+	hw_resc->max_l2_ctxs = le16_to_cpu(resp->max_l2_ctxs);
+	hw_resc->min_vnics = le16_to_cpu(resp->min_vnics);
+	hw_resc->max_vnics = le16_to_cpu(resp->max_vnics);
+	hw_resc->min_stat_ctxs = le16_to_cpu(resp->min_stat_ctx);
+	hw_resc->max_stat_ctxs = le16_to_cpu(resp->max_stat_ctx);
+
+	hw_resc->max_nqs = le16_to_cpu(resp->max_msix);
+	hw_resc->max_hw_ring_grps = hw_resc->max_rx_rings;
+
+hwrm_func_resc_qcaps_exit:
+	bnge_hwrm_req_drop(bd, req);
+	return rc;
+}
+
+int bnge_hwrm_func_qcaps(struct bnge_dev *bd)
+{
+	struct hwrm_func_qcaps_output *resp;
+	struct hwrm_func_qcaps_input *req;
+	struct bnge_pf_info *pf = &bd->pf;
+	u32 flags;
+	int rc;
+
+	rc = bnge_hwrm_req_init(bd, req, HWRM_FUNC_QCAPS);
+	if (rc)
+		return rc;
+
+	req->fid = cpu_to_le16(0xffff);
+	resp = bnge_hwrm_req_hold(bd, req);
+	rc = bnge_hwrm_req_send(bd, req);
+	if (rc)
+		goto hwrm_func_qcaps_exit;
+
+	flags = le32_to_cpu(resp->flags);
+	if (flags & FUNC_QCAPS_RESP_FLAGS_ROCE_V1_SUPPORTED)
+		bd->flags |= BNGE_EN_ROCE_V1;
+	if (flags & FUNC_QCAPS_RESP_FLAGS_ROCE_V2_SUPPORTED)
+		bd->flags |= BNGE_EN_ROCE_V2;
+
+	pf->fw_fid = le16_to_cpu(resp->fid);
+	pf->port_id = le16_to_cpu(resp->port_id);
+	memcpy(pf->mac_addr, resp->mac_address, ETH_ALEN);
+
+	bd->tso_max_segs = le16_to_cpu(resp->max_tso_segs);
+
+hwrm_func_qcaps_exit:
+	bnge_hwrm_req_drop(bd, req);
+	return rc;
+}
+
+int bnge_hwrm_vnic_qcaps(struct bnge_dev *bd)
+{
+	struct hwrm_vnic_qcaps_output *resp;
+	struct hwrm_vnic_qcaps_input *req;
+	int rc;
+
+	bd->hw_ring_stats_size = sizeof(struct ctx_hw_stats);
+	bd->rss_cap &= ~BNGE_RSS_CAP_NEW_RSS_CAP;
+
+	rc = bnge_hwrm_req_init(bd, req, HWRM_VNIC_QCAPS);
+	if (rc)
+		return rc;
+
+	resp = bnge_hwrm_req_hold(bd, req);
+	rc = bnge_hwrm_req_send(bd, req);
+	if (!rc) {
+		u32 flags = le32_to_cpu(resp->flags);
+
+		if (flags & VNIC_QCAPS_RESP_FLAGS_VLAN_STRIP_CAP)
+			bd->fw_cap |= BNGE_FW_CAP_VLAN_RX_STRIP;
+		if (flags & VNIC_QCAPS_RESP_FLAGS_RSS_HASH_TYPE_DELTA_CAP)
+			bd->rss_cap |= BNGE_RSS_CAP_RSS_HASH_TYPE_DELTA;
+		if (flags & VNIC_QCAPS_RESP_FLAGS_RSS_PROF_TCAM_MODE_ENABLED)
+			bd->rss_cap |= BNGE_RSS_CAP_RSS_TCAM;
+		bd->max_tpa_v2 = le16_to_cpu(resp->max_aggs_supported);
+		if (bd->max_tpa_v2)
+			bd->hw_ring_stats_size = BNGE_RING_STATS_SIZE;
+		if (flags & VNIC_QCAPS_RESP_FLAGS_HW_TUNNEL_TPA_CAP)
+			bd->fw_cap |= BNGE_FW_CAP_VNIC_TUNNEL_TPA;
+		if (flags & VNIC_QCAPS_RESP_FLAGS_RSS_IPSEC_AH_SPI_IPV4_CAP)
+			bd->rss_cap |= BNGE_RSS_CAP_AH_V4_RSS_CAP;
+		if (flags & VNIC_QCAPS_RESP_FLAGS_RSS_IPSEC_AH_SPI_IPV6_CAP)
+			bd->rss_cap |= BNGE_RSS_CAP_AH_V6_RSS_CAP;
+		if (flags & VNIC_QCAPS_RESP_FLAGS_RSS_IPSEC_ESP_SPI_IPV4_CAP)
+			bd->rss_cap |= BNGE_RSS_CAP_ESP_V4_RSS_CAP;
+		if (flags & VNIC_QCAPS_RESP_FLAGS_RSS_IPSEC_ESP_SPI_IPV6_CAP)
+			bd->rss_cap |= BNGE_RSS_CAP_ESP_V6_RSS_CAP;
+	}
+	bnge_hwrm_req_drop(bd, req);
+
+	return rc;
+}
+
+#define BNGE_CNPQ(q_profile)	\
+		((q_profile) ==	\
+		 QUEUE_QPORTCFG_RESP_QUEUE_ID0_SERVICE_PROFILE_LOSSY_ROCE_CNP)
+
+int bnge_hwrm_queue_qportcfg(struct bnge_dev *bd)
+{
+	struct hwrm_queue_qportcfg_output *resp;
+	struct hwrm_queue_qportcfg_input *req;
+	u8 i, j, *qptr;
+	bool no_rdma;
+	int rc;
+
+	rc = bnge_hwrm_req_init(bd, req, HWRM_QUEUE_QPORTCFG);
+	if (rc)
+		return rc;
+
+	resp = bnge_hwrm_req_hold(bd, req);
+	rc = bnge_hwrm_req_send(bd, req);
+	if (rc)
+		goto qportcfg_exit;
+
+	if (!resp->max_configurable_queues) {
+		rc = -EINVAL;
+		goto qportcfg_exit;
+	}
+	bd->max_tc = resp->max_configurable_queues;
+	bd->max_lltc = resp->max_configurable_lossless_queues;
+	if (bd->max_tc > BNGE_MAX_QUEUE)
+		bd->max_tc = BNGE_MAX_QUEUE;
+
+	no_rdma = !bnge_is_roce_en(bd);
+	qptr = &resp->queue_id0;
+	for (i = 0, j = 0; i < bd->max_tc; i++) {
+		bd->q_info[j].queue_id = *qptr;
+		bd->q_ids[i] = *qptr++;
+		bd->q_info[j].queue_profile = *qptr++;
+		bd->tc_to_qidx[j] = j;
+		if (!BNGE_CNPQ(bd->q_info[j].queue_profile) || no_rdma)
+			j++;
+	}
+	bd->max_q = bd->max_tc;
+	bd->max_tc = max_t(u8, j, 1);
+
+	if (resp->queue_cfg_info & QUEUE_QPORTCFG_RESP_QUEUE_CFG_INFO_ASYM_CFG)
+		bd->max_tc = 1;
+
+	if (bd->max_lltc > bd->max_tc)
+		bd->max_lltc = bd->max_tc;
+
+qportcfg_exit:
+	bnge_hwrm_req_drop(bd, req);
+	return rc;
+}
+
+int bnge_hwrm_vnic_set_hds(struct bnge_net *bn, struct bnge_vnic_info *vnic)
+{
+	u16 hds_thresh = (u16)bn->netdev->cfg_pending->hds_thresh;
+	struct hwrm_vnic_plcmodes_cfg_input *req;
+	struct bnge_dev *bd = bn->bd;
+	int rc;
+
+	rc = bnge_hwrm_req_init(bd, req, HWRM_VNIC_PLCMODES_CFG);
+	if (rc)
+		return rc;
+
+	req->flags = cpu_to_le32(VNIC_PLCMODES_CFG_REQ_FLAGS_JUMBO_PLACEMENT);
+	req->enables = cpu_to_le32(BNGE_PLC_EN_JUMBO_THRES_VALID);
+	req->jumbo_thresh = cpu_to_le16(bn->rx_buf_use_size);
+
+	if (bnge_is_agg_reqd(bd)) {
+		req->flags |= cpu_to_le32(VNIC_PLCMODES_CFG_REQ_FLAGS_HDS_IPV4 |
+					  VNIC_PLCMODES_CFG_REQ_FLAGS_HDS_IPV6);
+		req->enables |=
+			cpu_to_le32(BNGE_PLC_EN_HDS_THRES_VALID);
+		req->hds_threshold = cpu_to_le16(hds_thresh);
+	}
+	req->vnic_id = cpu_to_le32(vnic->fw_vnic_id);
+	return bnge_hwrm_req_send(bd, req);
+}
+
+int bnge_hwrm_vnic_ctx_alloc(struct bnge_dev *bd,
+			     struct bnge_vnic_info *vnic, u16 ctx_idx)
+{
+	struct hwrm_vnic_rss_cos_lb_ctx_alloc_output *resp;
+	struct hwrm_vnic_rss_cos_lb_ctx_alloc_input *req;
+	int rc;
+
+	rc = bnge_hwrm_req_init(bd, req, HWRM_VNIC_RSS_COS_LB_CTX_ALLOC);
+	if (rc)
+		return rc;
+
+	resp = bnge_hwrm_req_hold(bd, req);
+	rc = bnge_hwrm_req_send(bd, req);
+	if (!rc)
+		vnic->fw_rss_cos_lb_ctx[ctx_idx] =
+			le16_to_cpu(resp->rss_cos_lb_ctx_id);
+	bnge_hwrm_req_drop(bd, req);
+
+	return rc;
+}
+
+static void
+__bnge_hwrm_vnic_set_rss(struct bnge_net *bn,
+			 struct hwrm_vnic_rss_cfg_input *req,
+			 struct bnge_vnic_info *vnic)
+{
+	struct bnge_dev *bd = bn->bd;
+
+	bnge_fill_hw_rss_tbl(bn, vnic);
+	req->flags |= VNIC_RSS_CFG_REQ_FLAGS_IPSEC_HASH_TYPE_CFG_SUPPORT;
+
+	req->hash_type = cpu_to_le32(bd->rss_hash_cfg);
+	req->hash_mode_flags = VNIC_RSS_CFG_REQ_HASH_MODE_FLAGS_DEFAULT;
+	req->ring_grp_tbl_addr = cpu_to_le64(vnic->rss_table_dma_addr);
+	req->hash_key_tbl_addr = cpu_to_le64(vnic->rss_hash_key_dma_addr);
+}
+
+int bnge_hwrm_vnic_set_rss(struct bnge_net *bn,
+			   struct bnge_vnic_info *vnic, bool set_rss)
+{
+	struct hwrm_vnic_rss_cfg_input *req;
+	struct bnge_dev *bd = bn->bd;
+	dma_addr_t ring_tbl_map;
+	u32 i, nr_ctxs;
+	int rc;
+
+	rc = bnge_hwrm_req_init(bd, req, HWRM_VNIC_RSS_CFG);
+	if (rc)
+		return rc;
+
+	req->vnic_id = cpu_to_le16(vnic->fw_vnic_id);
+	if (!set_rss)
+		return bnge_hwrm_req_send(bd, req);
+
+	__bnge_hwrm_vnic_set_rss(bn, req, vnic);
+	ring_tbl_map = vnic->rss_table_dma_addr;
+	nr_ctxs = bnge_cal_nr_rss_ctxs(bd->rx_nr_rings);
+
+	bnge_hwrm_req_hold(bd, req);
+	for (i = 0; i < nr_ctxs; ring_tbl_map += BNGE_RSS_TABLE_SIZE, i++) {
+		req->ring_grp_tbl_addr = cpu_to_le64(ring_tbl_map);
+		req->ring_table_pair_index = i;
+		req->rss_ctx_idx = cpu_to_le16(vnic->fw_rss_cos_lb_ctx[i]);
+		rc = bnge_hwrm_req_send(bd, req);
+		if (rc)
+			goto exit;
+	}
+
+exit:
+	bnge_hwrm_req_drop(bd, req);
+	return rc;
+}
+
+int bnge_hwrm_vnic_cfg(struct bnge_net *bn, struct bnge_vnic_info *vnic)
+{
+	struct bnge_rx_ring_info *rxr = &bn->rx_ring[0];
+	struct hwrm_vnic_cfg_input *req;
+	struct bnge_dev *bd = bn->bd;
+	int rc;
+
+	rc = bnge_hwrm_req_init(bd, req, HWRM_VNIC_CFG);
+	if (rc)
+		return rc;
+
+	req->default_rx_ring_id =
+		cpu_to_le16(rxr->rx_ring_struct.fw_ring_id);
+	req->default_cmpl_ring_id =
+		cpu_to_le16(bnge_cp_ring_for_rx(rxr));
+	req->enables =
+		cpu_to_le32(VNIC_CFG_REQ_ENABLES_DEFAULT_RX_RING_ID |
+			    VNIC_CFG_REQ_ENABLES_DEFAULT_CMPL_RING_ID);
+	vnic->mru = bd->netdev->mtu + ETH_HLEN + VLAN_HLEN;
+	req->mru = cpu_to_le16(vnic->mru);
+
+	req->vnic_id = cpu_to_le16(vnic->fw_vnic_id);
+
+	if (bd->flags & BNGE_EN_STRIP_VLAN)
+		req->flags |= cpu_to_le32(VNIC_CFG_REQ_FLAGS_VLAN_STRIP_MODE);
+	if (vnic->vnic_id == BNGE_VNIC_DEFAULT && bnge_aux_registered(bd))
+		req->flags |= cpu_to_le32(BNGE_VNIC_CFG_ROCE_DUAL_MODE);
+
+	return bnge_hwrm_req_send(bd, req);
+}
+
+void bnge_hwrm_update_rss_hash_cfg(struct bnge_net *bn)
+{
+	struct bnge_vnic_info *vnic = &bn->vnic_info[BNGE_VNIC_DEFAULT];
+	struct hwrm_vnic_rss_qcfg_output *resp;
+	struct hwrm_vnic_rss_qcfg_input *req;
+	struct bnge_dev *bd = bn->bd;
+
+	if (bnge_hwrm_req_init(bd, req, HWRM_VNIC_RSS_QCFG))
+		return;
+
+	req->vnic_id = cpu_to_le16(vnic->fw_vnic_id);
+	/* all contexts configured to same hash_type, zero always exists */
+	req->rss_ctx_idx = cpu_to_le16(vnic->fw_rss_cos_lb_ctx[0]);
+	resp = bnge_hwrm_req_hold(bd, req);
+	if (!bnge_hwrm_req_send(bd, req))
+		bd->rss_hash_cfg =
+			le32_to_cpu(resp->hash_type) ?: bd->rss_hash_cfg;
+	bnge_hwrm_req_drop(bd, req);
+}
+
+int bnge_hwrm_l2_filter_free(struct bnge_dev *bd, struct bnge_l2_filter *fltr)
+{
+	struct hwrm_cfa_l2_filter_free_input *req;
+	int rc;
+
+	rc = bnge_hwrm_req_init(bd, req, HWRM_CFA_L2_FILTER_FREE);
+	if (rc)
+		return rc;
+
+	req->l2_filter_id = fltr->base.filter_id;
+	return bnge_hwrm_req_send(bd, req);
+}
+
+int bnge_hwrm_l2_filter_alloc(struct bnge_dev *bd, struct bnge_l2_filter *fltr)
+{
+	struct hwrm_cfa_l2_filter_alloc_output *resp;
+	struct hwrm_cfa_l2_filter_alloc_input *req;
+	int rc;
+
+	rc = bnge_hwrm_req_init(bd, req, HWRM_CFA_L2_FILTER_ALLOC);
+	if (rc)
+		return rc;
+
+	req->flags = cpu_to_le32(CFA_L2_FILTER_ALLOC_REQ_FLAGS_PATH_RX);
+
+	req->flags |= cpu_to_le32(CFA_L2_FILTER_ALLOC_REQ_FLAGS_OUTERMOST);
+	req->dst_id = cpu_to_le16(fltr->base.fw_vnic_id);
+	req->enables =
+		cpu_to_le32(CFA_L2_FILTER_ALLOC_REQ_ENABLES_L2_ADDR |
+			    CFA_L2_FILTER_ALLOC_REQ_ENABLES_DST_ID |
+			    CFA_L2_FILTER_ALLOC_REQ_ENABLES_L2_ADDR_MASK);
+	ether_addr_copy(req->l2_addr, fltr->l2_key.dst_mac_addr);
+	eth_broadcast_addr(req->l2_addr_mask);
+
+	if (fltr->l2_key.vlan) {
+		req->enables |=
+			cpu_to_le32(CFA_L2_FILTER_ALLOC_REQ_ENABLES_L2_IVLAN |
+				CFA_L2_FILTER_ALLOC_REQ_ENABLES_L2_IVLAN_MASK |
+				CFA_L2_FILTER_ALLOC_REQ_ENABLES_NUM_VLANS);
+		req->num_vlans = 1;
+		req->l2_ivlan = cpu_to_le16(fltr->l2_key.vlan);
+		req->l2_ivlan_mask = cpu_to_le16(0xfff);
+	}
+
+	resp = bnge_hwrm_req_hold(bd, req);
+	rc = bnge_hwrm_req_send(bd, req);
+	if (!rc)
+		fltr->base.filter_id = resp->l2_filter_id;
+
+	bnge_hwrm_req_drop(bd, req);
+	return rc;
+}
+
+int bnge_hwrm_cfa_l2_set_rx_mask(struct bnge_dev *bd,
+				 struct bnge_vnic_info *vnic)
+{
+	struct hwrm_cfa_l2_set_rx_mask_input *req;
+	int rc;
+
+	rc = bnge_hwrm_req_init(bd, req, HWRM_CFA_L2_SET_RX_MASK);
+	if (rc)
+		return rc;
+
+	req->vnic_id = cpu_to_le32(vnic->fw_vnic_id);
+	if (vnic->rx_mask & CFA_L2_SET_RX_MASK_REQ_MASK_MCAST) {
+		req->num_mc_entries = cpu_to_le32(vnic->mc_list_count);
+		req->mc_tbl_addr = cpu_to_le64(vnic->mc_list_mapping);
+	}
+	req->mask = cpu_to_le32(vnic->rx_mask);
+	return bnge_hwrm_req_send_silent(bd, req);
+}
+
+int bnge_hwrm_vnic_alloc(struct bnge_dev *bd, struct bnge_vnic_info *vnic,
+			 unsigned int nr_rings)
+{
+	struct hwrm_vnic_alloc_output *resp;
+	struct hwrm_vnic_alloc_input *req;
+	unsigned int i;
+	int rc;
+
+	rc = bnge_hwrm_req_init(bd, req, HWRM_VNIC_ALLOC);
+	if (rc)
+		return rc;
+
+	for (i = 0; i < BNGE_MAX_CTX_PER_VNIC; i++)
+		vnic->fw_rss_cos_lb_ctx[i] = INVALID_HW_RING_ID;
+	if (vnic->vnic_id == BNGE_VNIC_DEFAULT)
+		req->flags = cpu_to_le32(VNIC_ALLOC_REQ_FLAGS_DEFAULT);
+
+	resp = bnge_hwrm_req_hold(bd, req);
+	rc = bnge_hwrm_req_send(bd, req);
+	if (!rc)
+		vnic->fw_vnic_id = le32_to_cpu(resp->vnic_id);
+	bnge_hwrm_req_drop(bd, req);
+	return rc;
+}
+
+void bnge_hwrm_vnic_free_one(struct bnge_dev *bd, struct bnge_vnic_info *vnic)
+{
+	if (vnic->fw_vnic_id != INVALID_HW_RING_ID) {
+		struct hwrm_vnic_free_input *req;
+
+		if (bnge_hwrm_req_init(bd, req, HWRM_VNIC_FREE))
+			return;
+
+		req->vnic_id = cpu_to_le32(vnic->fw_vnic_id);
+
+		bnge_hwrm_req_send(bd, req);
+		vnic->fw_vnic_id = INVALID_HW_RING_ID;
+	}
+}
+
+void bnge_hwrm_vnic_ctx_free_one(struct bnge_dev *bd,
+				 struct bnge_vnic_info *vnic, u16 ctx_idx)
+{
+	struct hwrm_vnic_rss_cos_lb_ctx_free_input *req;
+
+	if (bnge_hwrm_req_init(bd, req, HWRM_VNIC_RSS_COS_LB_CTX_FREE))
+		return;
+
+	req->rss_cos_lb_ctx_id =
+		cpu_to_le16(vnic->fw_rss_cos_lb_ctx[ctx_idx]);
+
+	bnge_hwrm_req_send(bd, req);
+	vnic->fw_rss_cos_lb_ctx[ctx_idx] = INVALID_HW_RING_ID;
+}
+
+void bnge_hwrm_stat_ctx_free(struct bnge_net *bn)
+{
+	struct hwrm_stat_ctx_free_input *req;
+	struct bnge_dev *bd = bn->bd;
+	int i;
+
+	if (bnge_hwrm_req_init(bd, req, HWRM_STAT_CTX_FREE))
+		return;
+
+	bnge_hwrm_req_hold(bd, req);
+	for (i = 0; i < bd->nq_nr_rings; i++) {
+		struct bnge_napi *bnapi = bn->bnapi[i];
+		struct bnge_nq_ring_info *nqr = &bnapi->nq_ring;
+
+		if (nqr->hw_stats_ctx_id != INVALID_STATS_CTX_ID) {
+			req->stat_ctx_id = cpu_to_le32(nqr->hw_stats_ctx_id);
+			bnge_hwrm_req_send(bd, req);
+
+			nqr->hw_stats_ctx_id = INVALID_STATS_CTX_ID;
+		}
+	}
+	bnge_hwrm_req_drop(bd, req);
+}
+
+int bnge_hwrm_stat_ctx_alloc(struct bnge_net *bn)
+{
+	struct hwrm_stat_ctx_alloc_output *resp;
+	struct hwrm_stat_ctx_alloc_input *req;
+	struct bnge_dev *bd = bn->bd;
+	int rc, i;
+
+	rc = bnge_hwrm_req_init(bd, req, HWRM_STAT_CTX_ALLOC);
+	if (rc)
+		return rc;
+
+	req->stats_dma_length = cpu_to_le16(bd->hw_ring_stats_size);
+	req->update_period_ms = cpu_to_le32(bn->stats_coal_ticks / 1000);
+
+	resp = bnge_hwrm_req_hold(bd, req);
+	for (i = 0; i < bd->nq_nr_rings; i++) {
+		struct bnge_napi *bnapi = bn->bnapi[i];
+		struct bnge_nq_ring_info *nqr = &bnapi->nq_ring;
+
+		req->stats_dma_addr = cpu_to_le64(nqr->stats.hw_stats_map);
+
+		rc = bnge_hwrm_req_send(bd, req);
+		if (rc)
+			break;
+
+		nqr->hw_stats_ctx_id = le32_to_cpu(resp->stat_ctx_id);
+		bn->grp_info[i].fw_stats_ctx = nqr->hw_stats_ctx_id;
+	}
+	bnge_hwrm_req_drop(bd, req);
+	return rc;
+}
+
+int hwrm_ring_free_send_msg(struct bnge_net *bn,
+			    struct bnge_ring_struct *ring,
+			    u32 ring_type, int cmpl_ring_id)
+{
+	struct hwrm_ring_free_input *req;
+	struct bnge_dev *bd = bn->bd;
+	int rc;
+
+	rc = bnge_hwrm_req_init(bd, req, HWRM_RING_FREE);
+	if (rc)
+		goto exit;
+
+	req->cmpl_ring = cpu_to_le16(cmpl_ring_id);
+	req->ring_type = ring_type;
+	req->ring_id = cpu_to_le16(ring->fw_ring_id);
+
+	bnge_hwrm_req_hold(bd, req);
+	rc = bnge_hwrm_req_send(bd, req);
+	bnge_hwrm_req_drop(bd, req);
+exit:
+	if (rc) {
+		netdev_err(bd->netdev, "hwrm_ring_free type %d failed. rc:%d\n", ring_type, rc);
+		return -EIO;
+	}
+	return 0;
+}
+
+int hwrm_ring_alloc_send_msg(struct bnge_net *bn,
+			     struct bnge_ring_struct *ring,
+			     u32 ring_type, u32 map_index)
+{
+	struct bnge_ring_mem_info *rmem = &ring->ring_mem;
+	struct bnge_ring_grp_info *grp_info;
+	struct hwrm_ring_alloc_output *resp;
+	struct hwrm_ring_alloc_input *req;
+	struct bnge_dev *bd = bn->bd;
+	u16 ring_id, flags = 0;
+	int rc;
+
+	rc = bnge_hwrm_req_init(bd, req, HWRM_RING_ALLOC);
+	if (rc)
+		goto exit;
+
+	req->enables = 0;
+	if (rmem->nr_pages > 1) {
+		req->page_tbl_addr = cpu_to_le64(rmem->dma_pg_tbl);
+		/* Page size is in log2 units */
+		req->page_size = BNGE_PAGE_SHIFT;
+		req->page_tbl_depth = 1;
+	} else {
+		req->page_tbl_addr =  cpu_to_le64(rmem->dma_arr[0]);
+	}
+	req->fbo = 0;
+	/* Association of ring index with doorbell index and MSIX number */
+	req->logical_id = cpu_to_le16(map_index);
+
+	switch (ring_type) {
+	case HWRM_RING_ALLOC_TX: {
+		struct bnge_tx_ring_info *txr;
+
+		txr = container_of(ring, struct bnge_tx_ring_info,
+				   tx_ring_struct);
+		req->ring_type = RING_ALLOC_REQ_RING_TYPE_TX;
+		/* Association of transmit ring with completion ring */
+		grp_info = &bn->grp_info[ring->grp_idx];
+		req->cmpl_ring_id = cpu_to_le16(bnge_cp_ring_for_tx(txr));
+		req->length = cpu_to_le32(bn->tx_ring_mask + 1);
+		req->stat_ctx_id = cpu_to_le32(grp_info->fw_stats_ctx);
+		req->queue_id = cpu_to_le16(ring->queue_id);
+		req->flags = cpu_to_le16(flags);
+		break;
+	}
+	case HWRM_RING_ALLOC_RX:
+		req->ring_type = RING_ALLOC_REQ_RING_TYPE_RX;
+		req->length = cpu_to_le32(bn->rx_ring_mask + 1);
+
+		/* Association of rx ring with stats context */
+		grp_info = &bn->grp_info[ring->grp_idx];
+		req->rx_buf_size = cpu_to_le16(bn->rx_buf_use_size);
+		req->stat_ctx_id = cpu_to_le32(grp_info->fw_stats_ctx);
+		req->enables |=
+			cpu_to_le32(RING_ALLOC_REQ_ENABLES_RX_BUF_SIZE_VALID);
+		if (NET_IP_ALIGN == 2)
+			flags = RING_ALLOC_REQ_FLAGS_RX_SOP_PAD;
+		req->flags = cpu_to_le16(flags);
+		break;
+	case HWRM_RING_ALLOC_AGG:
+		req->ring_type = RING_ALLOC_REQ_RING_TYPE_RX_AGG;
+		/* Association of agg ring with rx ring */
+		grp_info = &bn->grp_info[ring->grp_idx];
+		req->rx_ring_id = cpu_to_le16(grp_info->rx_fw_ring_id);
+		req->rx_buf_size = cpu_to_le16(BNGE_RX_PAGE_SIZE);
+		req->stat_ctx_id = cpu_to_le32(grp_info->fw_stats_ctx);
+		req->enables |=
+			cpu_to_le32(RING_ALLOC_REQ_ENABLES_RX_RING_ID_VALID |
+				    RING_ALLOC_REQ_ENABLES_RX_BUF_SIZE_VALID);
+		req->length = cpu_to_le32(bn->rx_agg_ring_mask + 1);
+		break;
+	case HWRM_RING_ALLOC_CMPL:
+		req->ring_type = RING_ALLOC_REQ_RING_TYPE_L2_CMPL;
+		req->length = cpu_to_le32(bn->cp_ring_mask + 1);
+		/* Association of cp ring with nq */
+		grp_info = &bn->grp_info[map_index];
+		req->nq_ring_id = cpu_to_le16(grp_info->nq_fw_ring_id);
+		req->cq_handle = cpu_to_le64(ring->handle);
+		req->enables |=
+			cpu_to_le32(RING_ALLOC_REQ_ENABLES_NQ_RING_ID_VALID);
+		break;
+	case HWRM_RING_ALLOC_NQ:
+		req->ring_type = RING_ALLOC_REQ_RING_TYPE_NQ;
+		req->length = cpu_to_le32(bn->cp_ring_mask + 1);
+		req->int_mode = RING_ALLOC_REQ_INT_MODE_MSIX;
+		break;
+	default:
+		netdev_err(bn->netdev, "hwrm alloc invalid ring type %d\n", ring_type);
+		return -EINVAL;
+	}
+
+	resp = bnge_hwrm_req_hold(bd, req);
+	rc = bnge_hwrm_req_send(bd, req);
+	ring_id = le16_to_cpu(resp->ring_id);
+	bnge_hwrm_req_drop(bd, req);
+
+exit:
+	if (rc) {
+		netdev_err(bd->netdev, "hwrm_ring_alloc type %d failed. rc:%d\n", ring_type, rc);
+		return -EIO;
+	}
+	ring->fw_ring_id = ring_id;
+	return rc;
+}
+
+int bnge_hwrm_set_async_event_cr(struct bnge_dev *bd, int idx)
+{
+	struct hwrm_func_cfg_input *req;
+	int rc;
+
+	rc = bnge_hwrm_req_init(bd, req, HWRM_FUNC_CFG);
+	if (rc)
+		return rc;
+
+	req->fid = cpu_to_le16(0xffff);
+	req->enables = cpu_to_le32(FUNC_CFG_REQ_ENABLES_ASYNC_EVENT_CR);
+	req->async_event_cr = cpu_to_le16(idx);
+	return bnge_hwrm_req_send(bd, req);
+}
diff --git a/drivers/net/ethernet/broadcom/bnge/bnge_hwrm_lib.h b/drivers/net/ethernet/broadcom/bnge/bnge_hwrm_lib.h
new file mode 100644
index 000000000000..042f28e84a05
--- /dev/null
+++ b/drivers/net/ethernet/broadcom/bnge/bnge_hwrm_lib.h
@@ -0,0 +1,58 @@
+/* SPDX-License-Identifier: GPL-2.0 */
+/* Copyright (c) 2025 Broadcom */
+
+#ifndef _BNGE_HWRM_LIB_H_
+#define _BNGE_HWRM_LIB_H_
+
+#define BNGE_PLC_EN_JUMBO_THRES_VALID		\
+	VNIC_PLCMODES_CFG_REQ_ENABLES_JUMBO_THRESH_VALID
+#define BNGE_PLC_EN_HDS_THRES_VALID		\
+	VNIC_PLCMODES_CFG_REQ_ENABLES_HDS_THRESHOLD_VALID
+#define BNGE_VNIC_CFG_ROCE_DUAL_MODE		\
+	VNIC_CFG_REQ_FLAGS_ROCE_DUAL_VNIC_MODE
+
+int bnge_hwrm_ver_get(struct bnge_dev *bd);
+int bnge_hwrm_func_reset(struct bnge_dev *bd);
+int bnge_hwrm_fw_set_time(struct bnge_dev *bd);
+int bnge_hwrm_func_drv_rgtr(struct bnge_dev *bd);
+int bnge_hwrm_func_drv_unrgtr(struct bnge_dev *bd);
+int bnge_hwrm_vnic_qcaps(struct bnge_dev *bd);
+int bnge_hwrm_nvm_dev_info(struct bnge_dev *bd,
+			   struct hwrm_nvm_get_dev_info_output *nvm_dev_info);
+int bnge_hwrm_func_backing_store(struct bnge_dev *bd,
+				 struct bnge_ctx_mem_type *ctxm,
+				 bool last);
+int bnge_hwrm_func_backing_store_qcaps(struct bnge_dev *bd);
+int bnge_hwrm_reserve_rings(struct bnge_dev *bd,
+			    struct bnge_hw_rings *hwr);
+int bnge_hwrm_func_qcaps(struct bnge_dev *bd);
+int bnge_hwrm_vnic_qcaps(struct bnge_dev *bd);
+int bnge_hwrm_func_qcfg(struct bnge_dev *bd);
+int bnge_hwrm_func_resc_qcaps(struct bnge_dev *bd);
+int bnge_hwrm_queue_qportcfg(struct bnge_dev *bd);
+
+int bnge_hwrm_vnic_set_hds(struct bnge_net *bn, struct bnge_vnic_info *vnic);
+int bnge_hwrm_vnic_ctx_alloc(struct bnge_dev *bd,
+			     struct bnge_vnic_info *vnic, u16 ctx_idx);
+int bnge_hwrm_vnic_set_rss(struct bnge_net *bn,
+			   struct bnge_vnic_info *vnic, bool set_rss);
+int bnge_hwrm_vnic_cfg(struct bnge_net *bn, struct bnge_vnic_info *vnic);
+void bnge_hwrm_update_rss_hash_cfg(struct bnge_net *bn);
+int bnge_hwrm_vnic_alloc(struct bnge_dev *bd, struct bnge_vnic_info *vnic,
+			 unsigned int nr_rings);
+void bnge_hwrm_vnic_free_one(struct bnge_dev *bd, struct bnge_vnic_info *vnic);
+void bnge_hwrm_vnic_ctx_free_one(struct bnge_dev *bd,
+				 struct bnge_vnic_info *vnic, u16 ctx_idx);
+int bnge_hwrm_l2_filter_free(struct bnge_dev *bd, struct bnge_l2_filter *fltr);
+int bnge_hwrm_l2_filter_alloc(struct bnge_dev *bd, struct bnge_l2_filter *fltr);
+int bnge_hwrm_cfa_l2_set_rx_mask(struct bnge_dev *bd,
+				 struct bnge_vnic_info *vnic);
+void bnge_hwrm_stat_ctx_free(struct bnge_net *bn);
+int bnge_hwrm_stat_ctx_alloc(struct bnge_net *bn);
+int hwrm_ring_free_send_msg(struct bnge_net *bn, struct bnge_ring_struct *ring,
+			    u32 ring_type, int cmpl_ring_id);
+int hwrm_ring_alloc_send_msg(struct bnge_net *bn,
+			     struct bnge_ring_struct *ring,
+			     u32 ring_type, u32 map_index);
+int bnge_hwrm_set_async_event_cr(struct bnge_dev *bd, int idx);
+#endif /* _BNGE_HWRM_LIB_H_ */
diff --git a/drivers/net/ethernet/broadcom/bnge/bnge_netdev.c b/drivers/net/ethernet/broadcom/bnge/bnge_netdev.c
new file mode 100644
index 000000000000..832eeb960bd2
--- /dev/null
+++ b/drivers/net/ethernet/broadcom/bnge/bnge_netdev.c
@@ -0,0 +1,2485 @@
+// SPDX-License-Identifier: GPL-2.0
+// Copyright (c) 2025 Broadcom.
+
+#include <asm/byteorder.h>
+#include <linux/dma-mapping.h>
+#include <linux/dmapool.h>
+#include <linux/delay.h>
+#include <linux/errno.h>
+#include <linux/kernel.h>
+#include <linux/list.h>
+#include <linux/pci.h>
+#include <linux/netdevice.h>
+#include <linux/etherdevice.h>
+#include <linux/if.h>
+#include <net/ip.h>
+#include <linux/skbuff.h>
+#include <net/page_pool/helpers.h>
+
+#include "bnge.h"
+#include "bnge_hwrm_lib.h"
+#include "bnge_ethtool.h"
+#include "bnge_rmem.h"
+
+#define BNGE_RING_TO_TC_OFF(bd, tx)	\
+	((tx) % (bd)->tx_nr_rings_per_tc)
+
+#define BNGE_RING_TO_TC(bd, tx)		\
+	((tx) / (bd)->tx_nr_rings_per_tc)
+
+#define BNGE_TC_TO_RING_BASE(bd, tc)	\
+	((tc) * (bd)->tx_nr_rings_per_tc)
+
+static void bnge_free_stats_mem(struct bnge_net *bn,
+				struct bnge_stats_mem *stats)
+{
+	struct bnge_dev *bd = bn->bd;
+
+	if (stats->hw_stats) {
+		dma_free_coherent(bd->dev, stats->len, stats->hw_stats,
+				  stats->hw_stats_map);
+		stats->hw_stats = NULL;
+	}
+}
+
+static int bnge_alloc_stats_mem(struct bnge_net *bn,
+				struct bnge_stats_mem *stats)
+{
+	struct bnge_dev *bd = bn->bd;
+
+	stats->hw_stats = dma_alloc_coherent(bd->dev, stats->len,
+					     &stats->hw_stats_map, GFP_KERNEL);
+	if (!stats->hw_stats)
+		return -ENOMEM;
+
+	return 0;
+}
+
+static void bnge_free_ring_stats(struct bnge_net *bn)
+{
+	struct bnge_dev *bd = bn->bd;
+	int i;
+
+	if (!bn->bnapi)
+		return;
+
+	for (i = 0; i < bd->nq_nr_rings; i++) {
+		struct bnge_napi *bnapi = bn->bnapi[i];
+		struct bnge_nq_ring_info *nqr = &bnapi->nq_ring;
+
+		bnge_free_stats_mem(bn, &nqr->stats);
+	}
+}
+
+static int bnge_alloc_ring_stats(struct bnge_net *bn)
+{
+	struct bnge_dev *bd = bn->bd;
+	u32 size, i;
+	int rc;
+
+	size = bd->hw_ring_stats_size;
+
+	for (i = 0; i < bd->nq_nr_rings; i++) {
+		struct bnge_napi *bnapi = bn->bnapi[i];
+		struct bnge_nq_ring_info *nqr = &bnapi->nq_ring;
+
+		nqr->stats.len = size;
+		rc = bnge_alloc_stats_mem(bn, &nqr->stats);
+		if (rc)
+			goto err_free_ring_stats;
+
+		nqr->hw_stats_ctx_id = INVALID_STATS_CTX_ID;
+	}
+	return 0;
+
+err_free_ring_stats:
+	bnge_free_ring_stats(bn);
+	return rc;
+}
+
+static void bnge_free_nq_desc_arr(struct bnge_nq_ring_info *nqr)
+{
+	struct bnge_ring_struct *ring = &nqr->ring_struct;
+
+	kfree(nqr->desc_ring);
+	nqr->desc_ring = NULL;
+	ring->ring_mem.pg_arr = NULL;
+	kfree(nqr->desc_mapping);
+	nqr->desc_mapping = NULL;
+	ring->ring_mem.dma_arr = NULL;
+}
+
+static void bnge_free_cp_desc_arr(struct bnge_cp_ring_info *cpr)
+{
+	struct bnge_ring_struct *ring = &cpr->ring_struct;
+
+	kfree(cpr->desc_ring);
+	cpr->desc_ring = NULL;
+	ring->ring_mem.pg_arr = NULL;
+	kfree(cpr->desc_mapping);
+	cpr->desc_mapping = NULL;
+	ring->ring_mem.dma_arr = NULL;
+}
+
+static int bnge_alloc_nq_desc_arr(struct bnge_nq_ring_info *nqr, int n)
+{
+	nqr->desc_ring = kcalloc(n, sizeof(*nqr->desc_ring), GFP_KERNEL);
+	if (!nqr->desc_ring)
+		return -ENOMEM;
+
+	nqr->desc_mapping = kcalloc(n, sizeof(*nqr->desc_mapping), GFP_KERNEL);
+	if (!nqr->desc_mapping)
+		goto err_free_desc_ring;
+	return 0;
+
+err_free_desc_ring:
+	kfree(nqr->desc_ring);
+	nqr->desc_ring = NULL;
+	return -ENOMEM;
+}
+
+static int bnge_alloc_cp_desc_arr(struct bnge_cp_ring_info *cpr, int n)
+{
+	cpr->desc_ring = kcalloc(n, sizeof(*cpr->desc_ring), GFP_KERNEL);
+	if (!cpr->desc_ring)
+		return -ENOMEM;
+
+	cpr->desc_mapping = kcalloc(n, sizeof(*cpr->desc_mapping), GFP_KERNEL);
+	if (!cpr->desc_mapping)
+		goto err_free_desc_ring;
+	return 0;
+
+err_free_desc_ring:
+	kfree(cpr->desc_ring);
+	cpr->desc_ring = NULL;
+	return -ENOMEM;
+}
+
+static void bnge_free_nq_arrays(struct bnge_net *bn)
+{
+	struct bnge_dev *bd = bn->bd;
+	int i;
+
+	for (i = 0; i < bd->nq_nr_rings; i++) {
+		struct bnge_napi *bnapi = bn->bnapi[i];
+
+		bnge_free_nq_desc_arr(&bnapi->nq_ring);
+	}
+}
+
+static int bnge_alloc_nq_arrays(struct bnge_net *bn)
+{
+	struct bnge_dev *bd = bn->bd;
+	int i, rc;
+
+	for (i = 0; i < bd->nq_nr_rings; i++) {
+		struct bnge_napi *bnapi = bn->bnapi[i];
+
+		rc = bnge_alloc_nq_desc_arr(&bnapi->nq_ring, bn->cp_nr_pages);
+		if (rc)
+			goto err_free_nq_arrays;
+	}
+	return 0;
+
+err_free_nq_arrays:
+	bnge_free_nq_arrays(bn);
+	return rc;
+}
+
+static void bnge_free_nq_tree(struct bnge_net *bn)
+{
+	struct bnge_dev *bd = bn->bd;
+	int i;
+
+	for (i = 0; i < bd->nq_nr_rings; i++) {
+		struct bnge_napi *bnapi = bn->bnapi[i];
+		struct bnge_nq_ring_info *nqr;
+		struct bnge_ring_struct *ring;
+		int j;
+
+		nqr = &bnapi->nq_ring;
+		ring = &nqr->ring_struct;
+
+		bnge_free_ring(bd, &ring->ring_mem);
+
+		if (!nqr->cp_ring_arr)
+			continue;
+
+		for (j = 0; j < nqr->cp_ring_count; j++) {
+			struct bnge_cp_ring_info *cpr = &nqr->cp_ring_arr[j];
+
+			ring = &cpr->ring_struct;
+			bnge_free_ring(bd, &ring->ring_mem);
+			bnge_free_cp_desc_arr(cpr);
+		}
+		kfree(nqr->cp_ring_arr);
+		nqr->cp_ring_arr = NULL;
+		nqr->cp_ring_count = 0;
+	}
+}
+
+static int alloc_one_cp_ring(struct bnge_net *bn,
+			     struct bnge_cp_ring_info *cpr)
+{
+	struct bnge_ring_mem_info *rmem;
+	struct bnge_ring_struct *ring;
+	struct bnge_dev *bd = bn->bd;
+	int rc;
+
+	rc = bnge_alloc_cp_desc_arr(cpr, bn->cp_nr_pages);
+	if (rc)
+		return -ENOMEM;
+	ring = &cpr->ring_struct;
+	rmem = &ring->ring_mem;
+	rmem->nr_pages = bn->cp_nr_pages;
+	rmem->page_size = HW_CMPD_RING_SIZE;
+	rmem->pg_arr = (void **)cpr->desc_ring;
+	rmem->dma_arr = cpr->desc_mapping;
+	rmem->flags = BNGE_RMEM_RING_PTE_FLAG;
+	rc = bnge_alloc_ring(bd, rmem);
+	if (rc)
+		goto err_free_cp_desc_arr;
+	return rc;
+
+err_free_cp_desc_arr:
+	bnge_free_cp_desc_arr(cpr);
+	return rc;
+}
+
+static int bnge_alloc_nq_tree(struct bnge_net *bn)
+{
+	struct bnge_dev *bd = bn->bd;
+	int i, j, ulp_msix, rc;
+	int tcs = 1;
+
+	ulp_msix = bnge_aux_get_msix(bd);
+	for (i = 0, j = 0; i < bd->nq_nr_rings; i++) {
+		bool sh = !!(bd->flags & BNGE_EN_SHARED_CHNL);
+		struct bnge_napi *bnapi = bn->bnapi[i];
+		struct bnge_nq_ring_info *nqr;
+		struct bnge_cp_ring_info *cpr;
+		struct bnge_ring_struct *ring;
+		int cp_count = 0, k;
+		int rx = 0, tx = 0;
+
+		nqr = &bnapi->nq_ring;
+		nqr->bnapi = bnapi;
+		ring = &nqr->ring_struct;
+
+		rc = bnge_alloc_ring(bd, &ring->ring_mem);
+		if (rc)
+			goto err_free_nq_tree;
+
+		ring->map_idx = ulp_msix + i;
+
+		if (i < bd->rx_nr_rings) {
+			cp_count++;
+			rx = 1;
+		}
+
+		if ((sh && i < bd->tx_nr_rings) ||
+		    (!sh && i >= bd->rx_nr_rings)) {
+			cp_count += tcs;
+			tx = 1;
+		}
+
+		nqr->cp_ring_arr = kcalloc(cp_count, sizeof(*cpr),
+					   GFP_KERNEL);
+		if (!nqr->cp_ring_arr) {
+			rc = -ENOMEM;
+			goto err_free_nq_tree;
+		}
+
+		nqr->cp_ring_count = cp_count;
+
+		for (k = 0; k < cp_count; k++) {
+			cpr = &nqr->cp_ring_arr[k];
+			rc = alloc_one_cp_ring(bn, cpr);
+			if (rc)
+				goto err_free_nq_tree;
+
+			cpr->bnapi = bnapi;
+			cpr->cp_idx = k;
+			if (!k && rx) {
+				bn->rx_ring[i].rx_cpr = cpr;
+				cpr->cp_ring_type = BNGE_NQ_HDL_TYPE_RX;
+			} else {
+				int n, tc = k - rx;
+
+				n = BNGE_TC_TO_RING_BASE(bd, tc) + j;
+				bn->tx_ring[n].tx_cpr = cpr;
+				cpr->cp_ring_type = BNGE_NQ_HDL_TYPE_TX;
+			}
+		}
+		if (tx)
+			j++;
+	}
+	return 0;
+
+err_free_nq_tree:
+	bnge_free_nq_tree(bn);
+	return rc;
+}
+
+static bool bnge_separate_head_pool(struct bnge_rx_ring_info *rxr)
+{
+	return rxr->need_head_pool || PAGE_SIZE > BNGE_RX_PAGE_SIZE;
+}
+
+static void bnge_free_one_rx_ring_bufs(struct bnge_net *bn,
+				       struct bnge_rx_ring_info *rxr)
+{
+	int i, max_idx;
+
+	if (!rxr->rx_buf_ring)
+		return;
+
+	max_idx = bn->rx_nr_pages * RX_DESC_CNT;
+
+	for (i = 0; i < max_idx; i++) {
+		struct bnge_sw_rx_bd *rx_buf = &rxr->rx_buf_ring[i];
+		void *data = rx_buf->data;
+
+		if (!data)
+			continue;
+
+		rx_buf->data = NULL;
+		page_pool_free_va(rxr->head_pool, data, true);
+	}
+}
+
+static void bnge_free_one_agg_ring_bufs(struct bnge_net *bn,
+					struct bnge_rx_ring_info *rxr)
+{
+	int i, max_idx;
+
+	if (!rxr->rx_agg_buf_ring)
+		return;
+
+	max_idx = bn->rx_agg_nr_pages * RX_DESC_CNT;
+
+	for (i = 0; i < max_idx; i++) {
+		struct bnge_sw_rx_agg_bd *rx_agg_buf = &rxr->rx_agg_buf_ring[i];
+		netmem_ref netmem = rx_agg_buf->netmem;
+
+		if (!netmem)
+			continue;
+
+		rx_agg_buf->netmem = 0;
+		__clear_bit(i, rxr->rx_agg_bmap);
+
+		page_pool_recycle_direct_netmem(rxr->page_pool, netmem);
+	}
+}
+
+static void bnge_free_one_rx_ring_pair_bufs(struct bnge_net *bn,
+					    struct bnge_rx_ring_info *rxr)
+{
+	bnge_free_one_rx_ring_bufs(bn, rxr);
+	bnge_free_one_agg_ring_bufs(bn, rxr);
+}
+
+static void bnge_free_rx_ring_pair_bufs(struct bnge_net *bn)
+{
+	struct bnge_dev *bd = bn->bd;
+	int i;
+
+	if (!bn->rx_ring)
+		return;
+
+	for (i = 0; i < bd->rx_nr_rings; i++)
+		bnge_free_one_rx_ring_pair_bufs(bn, &bn->rx_ring[i]);
+}
+
+static void bnge_free_all_rings_bufs(struct bnge_net *bn)
+{
+	bnge_free_rx_ring_pair_bufs(bn);
+}
+
+static void bnge_free_rx_rings(struct bnge_net *bn)
+{
+	struct bnge_dev *bd = bn->bd;
+	int i;
+
+	for (i = 0; i < bd->rx_nr_rings; i++) {
+		struct bnge_rx_ring_info *rxr = &bn->rx_ring[i];
+		struct bnge_ring_struct *ring;
+
+		page_pool_destroy(rxr->page_pool);
+		page_pool_destroy(rxr->head_pool);
+		rxr->page_pool = rxr->head_pool = NULL;
+
+		kfree(rxr->rx_agg_bmap);
+		rxr->rx_agg_bmap = NULL;
+
+		ring = &rxr->rx_ring_struct;
+		bnge_free_ring(bd, &ring->ring_mem);
+
+		ring = &rxr->rx_agg_ring_struct;
+		bnge_free_ring(bd, &ring->ring_mem);
+	}
+}
+
+static int bnge_alloc_rx_page_pool(struct bnge_net *bn,
+				   struct bnge_rx_ring_info *rxr,
+				   int numa_node)
+{
+	const unsigned int agg_size_fac = PAGE_SIZE / BNGE_RX_PAGE_SIZE;
+	const unsigned int rx_size_fac = PAGE_SIZE / SZ_4K;
+	struct page_pool_params pp = { 0 };
+	struct bnge_dev *bd = bn->bd;
+	struct page_pool *pool;
+
+	pp.pool_size = bn->rx_agg_ring_size / agg_size_fac;
+	pp.nid = numa_node;
+	pp.netdev = bn->netdev;
+	pp.dev = bd->dev;
+	pp.dma_dir = bn->rx_dir;
+	pp.max_len = PAGE_SIZE;
+	pp.flags = PP_FLAG_DMA_MAP | PP_FLAG_DMA_SYNC_DEV |
+		   PP_FLAG_ALLOW_UNREADABLE_NETMEM;
+	pp.queue_idx = rxr->bnapi->index;
+
+	pool = page_pool_create(&pp);
+	if (IS_ERR(pool))
+		return PTR_ERR(pool);
+	rxr->page_pool = pool;
+
+	rxr->need_head_pool = page_pool_is_unreadable(pool);
+	if (bnge_separate_head_pool(rxr)) {
+		pp.pool_size = min(bn->rx_ring_size / rx_size_fac, 1024);
+		pp.flags = PP_FLAG_DMA_MAP | PP_FLAG_DMA_SYNC_DEV;
+		pool = page_pool_create(&pp);
+		if (IS_ERR(pool))
+			goto err_destroy_pp;
+	} else {
+		page_pool_get(pool);
+	}
+	rxr->head_pool = pool;
+	return 0;
+
+err_destroy_pp:
+	page_pool_destroy(rxr->page_pool);
+	rxr->page_pool = NULL;
+	return PTR_ERR(pool);
+}
+
+static void bnge_enable_rx_page_pool(struct bnge_rx_ring_info *rxr)
+{
+	page_pool_enable_direct_recycling(rxr->head_pool, &rxr->bnapi->napi);
+	page_pool_enable_direct_recycling(rxr->page_pool, &rxr->bnapi->napi);
+}
+
+static int bnge_alloc_rx_agg_bmap(struct bnge_net *bn,
+				  struct bnge_rx_ring_info *rxr)
+{
+	u16 mem_size;
+
+	rxr->rx_agg_bmap_size = bn->rx_agg_ring_mask + 1;
+	mem_size = rxr->rx_agg_bmap_size / 8;
+	rxr->rx_agg_bmap = kzalloc(mem_size, GFP_KERNEL);
+	if (!rxr->rx_agg_bmap)
+		return -ENOMEM;
+
+	return 0;
+}
+
+static int bnge_alloc_rx_rings(struct bnge_net *bn)
+{
+	int i, rc = 0, agg_rings = 0, cpu;
+	struct bnge_dev *bd = bn->bd;
+
+	if (bnge_is_agg_reqd(bd))
+		agg_rings = 1;
+
+	for (i = 0; i < bd->rx_nr_rings; i++) {
+		struct bnge_rx_ring_info *rxr = &bn->rx_ring[i];
+		struct bnge_ring_struct *ring;
+		int cpu_node;
+
+		ring = &rxr->rx_ring_struct;
+
+		cpu = cpumask_local_spread(i, dev_to_node(bd->dev));
+		cpu_node = cpu_to_node(cpu);
+		netdev_dbg(bn->netdev, "Allocating page pool for rx_ring[%d] on numa_node: %d\n",
+			   i, cpu_node);
+		rc = bnge_alloc_rx_page_pool(bn, rxr, cpu_node);
+		if (rc)
+			goto err_free_rx_rings;
+		bnge_enable_rx_page_pool(rxr);
+
+		rc = bnge_alloc_ring(bd, &ring->ring_mem);
+		if (rc)
+			goto err_free_rx_rings;
+
+		ring->grp_idx = i;
+		if (agg_rings) {
+			ring = &rxr->rx_agg_ring_struct;
+			rc = bnge_alloc_ring(bd, &ring->ring_mem);
+			if (rc)
+				goto err_free_rx_rings;
+
+			ring->grp_idx = i;
+			rc = bnge_alloc_rx_agg_bmap(bn, rxr);
+			if (rc)
+				goto err_free_rx_rings;
+		}
+	}
+	return rc;
+
+err_free_rx_rings:
+	bnge_free_rx_rings(bn);
+	return rc;
+}
+
+static void bnge_free_tx_rings(struct bnge_net *bn)
+{
+	struct bnge_dev *bd = bn->bd;
+	int i;
+
+	for (i = 0; i < bd->tx_nr_rings; i++) {
+		struct bnge_tx_ring_info *txr = &bn->tx_ring[i];
+		struct bnge_ring_struct *ring;
+
+		ring = &txr->tx_ring_struct;
+
+		bnge_free_ring(bd, &ring->ring_mem);
+	}
+}
+
+static int bnge_alloc_tx_rings(struct bnge_net *bn)
+{
+	struct bnge_dev *bd = bn->bd;
+	int i, j, rc;
+
+	for (i = 0, j = 0; i < bd->tx_nr_rings; i++) {
+		struct bnge_tx_ring_info *txr = &bn->tx_ring[i];
+		struct bnge_ring_struct *ring;
+		u8 qidx;
+
+		ring = &txr->tx_ring_struct;
+
+		rc = bnge_alloc_ring(bd, &ring->ring_mem);
+		if (rc)
+			goto err_free_tx_rings;
+
+		ring->grp_idx = txr->bnapi->index;
+		qidx = bd->tc_to_qidx[j];
+		ring->queue_id = bd->q_info[qidx].queue_id;
+		if (BNGE_RING_TO_TC_OFF(bd, i) == (bd->tx_nr_rings_per_tc - 1))
+			j++;
+	}
+	return 0;
+
+err_free_tx_rings:
+	bnge_free_tx_rings(bn);
+	return rc;
+}
+
+static void bnge_free_vnic_attributes(struct bnge_net *bn)
+{
+	struct pci_dev *pdev = bn->bd->pdev;
+	struct bnge_vnic_info *vnic;
+	int i;
+
+	if (!bn->vnic_info)
+		return;
+
+	for (i = 0; i < bn->nr_vnics; i++) {
+		vnic = &bn->vnic_info[i];
+
+		kfree(vnic->uc_list);
+		vnic->uc_list = NULL;
+
+		if (vnic->mc_list) {
+			dma_free_coherent(&pdev->dev, vnic->mc_list_size,
+					  vnic->mc_list, vnic->mc_list_mapping);
+			vnic->mc_list = NULL;
+		}
+
+		if (vnic->rss_table) {
+			dma_free_coherent(&pdev->dev, vnic->rss_table_size,
+					  vnic->rss_table,
+					  vnic->rss_table_dma_addr);
+			vnic->rss_table = NULL;
+		}
+
+		vnic->rss_hash_key = NULL;
+		vnic->flags = 0;
+	}
+}
+
+static int bnge_alloc_vnic_attributes(struct bnge_net *bn)
+{
+	struct bnge_dev *bd = bn->bd;
+	struct bnge_vnic_info *vnic;
+	int i, size;
+
+	for (i = 0; i < bn->nr_vnics; i++) {
+		vnic = &bn->vnic_info[i];
+
+		if (vnic->flags & BNGE_VNIC_UCAST_FLAG) {
+			int mem_size = (BNGE_MAX_UC_ADDRS - 1) * ETH_ALEN;
+
+			vnic->uc_list = kmalloc(mem_size, GFP_KERNEL);
+			if (!vnic->uc_list)
+				goto err_free_vnic_attributes;
+		}
+
+		if (vnic->flags & BNGE_VNIC_MCAST_FLAG) {
+			vnic->mc_list_size = BNGE_MAX_MC_ADDRS * ETH_ALEN;
+			vnic->mc_list =
+				dma_alloc_coherent(bd->dev,
+						   vnic->mc_list_size,
+						   &vnic->mc_list_mapping,
+						   GFP_KERNEL);
+			if (!vnic->mc_list)
+				goto err_free_vnic_attributes;
+		}
+
+		/* Allocate rss table and hash key */
+		size = L1_CACHE_ALIGN(BNGE_MAX_RSS_TABLE_SIZE);
+
+		vnic->rss_table_size = size + HW_HASH_KEY_SIZE;
+		vnic->rss_table = dma_alloc_coherent(bd->dev,
+						     vnic->rss_table_size,
+						     &vnic->rss_table_dma_addr,
+						     GFP_KERNEL);
+		if (!vnic->rss_table)
+			goto err_free_vnic_attributes;
+
+		vnic->rss_hash_key = ((void *)vnic->rss_table) + size;
+		vnic->rss_hash_key_dma_addr = vnic->rss_table_dma_addr + size;
+	}
+	return 0;
+
+err_free_vnic_attributes:
+	bnge_free_vnic_attributes(bn);
+	return -ENOMEM;
+}
+
+static int bnge_alloc_vnics(struct bnge_net *bn)
+{
+	int num_vnics;
+
+	/* Allocate only 1 VNIC for now
+	 * Additional VNICs will be added based on RFS/NTUPLE in future patches
+	 */
+	num_vnics = 1;
+
+	bn->vnic_info = kcalloc(num_vnics, sizeof(struct bnge_vnic_info),
+				GFP_KERNEL);
+	if (!bn->vnic_info)
+		return -ENOMEM;
+
+	bn->nr_vnics = num_vnics;
+
+	return 0;
+}
+
+static void bnge_free_vnics(struct bnge_net *bn)
+{
+	kfree(bn->vnic_info);
+	bn->vnic_info = NULL;
+	bn->nr_vnics = 0;
+}
+
+static void bnge_free_ring_grps(struct bnge_net *bn)
+{
+	kfree(bn->grp_info);
+	bn->grp_info = NULL;
+}
+
+static int bnge_init_ring_grps(struct bnge_net *bn)
+{
+	struct bnge_dev *bd = bn->bd;
+	int i;
+
+	bn->grp_info = kcalloc(bd->nq_nr_rings,
+			       sizeof(struct bnge_ring_grp_info),
+			       GFP_KERNEL);
+	if (!bn->grp_info)
+		return -ENOMEM;
+	for (i = 0; i < bd->nq_nr_rings; i++) {
+		bn->grp_info[i].fw_stats_ctx = INVALID_HW_RING_ID;
+		bn->grp_info[i].fw_grp_id = INVALID_HW_RING_ID;
+		bn->grp_info[i].rx_fw_ring_id = INVALID_HW_RING_ID;
+		bn->grp_info[i].agg_fw_ring_id = INVALID_HW_RING_ID;
+		bn->grp_info[i].nq_fw_ring_id = INVALID_HW_RING_ID;
+	}
+
+	return 0;
+}
+
+static void bnge_free_core(struct bnge_net *bn)
+{
+	bnge_free_vnic_attributes(bn);
+	bnge_free_tx_rings(bn);
+	bnge_free_rx_rings(bn);
+	bnge_free_nq_tree(bn);
+	bnge_free_nq_arrays(bn);
+	bnge_free_ring_stats(bn);
+	bnge_free_ring_grps(bn);
+	bnge_free_vnics(bn);
+	kfree(bn->tx_ring_map);
+	bn->tx_ring_map = NULL;
+	kfree(bn->tx_ring);
+	bn->tx_ring = NULL;
+	kfree(bn->rx_ring);
+	bn->rx_ring = NULL;
+	kfree(bn->bnapi);
+	bn->bnapi = NULL;
+}
+
+static int bnge_alloc_core(struct bnge_net *bn)
+{
+	struct bnge_dev *bd = bn->bd;
+	int i, j, size, arr_size;
+	int rc = -ENOMEM;
+	void *bnapi;
+
+	arr_size = L1_CACHE_ALIGN(sizeof(struct bnge_napi *) *
+			bd->nq_nr_rings);
+	size = L1_CACHE_ALIGN(sizeof(struct bnge_napi));
+	bnapi = kzalloc(arr_size + size * bd->nq_nr_rings, GFP_KERNEL);
+	if (!bnapi)
+		return rc;
+
+	bn->bnapi = bnapi;
+	bnapi += arr_size;
+	for (i = 0; i < bd->nq_nr_rings; i++, bnapi += size) {
+		struct bnge_nq_ring_info *nqr;
+
+		bn->bnapi[i] = bnapi;
+		bn->bnapi[i]->index = i;
+		bn->bnapi[i]->bn = bn;
+		nqr = &bn->bnapi[i]->nq_ring;
+		nqr->ring_struct.ring_mem.flags = BNGE_RMEM_RING_PTE_FLAG;
+	}
+
+	bn->rx_ring = kcalloc(bd->rx_nr_rings,
+			      sizeof(struct bnge_rx_ring_info),
+			      GFP_KERNEL);
+	if (!bn->rx_ring)
+		goto err_free_core;
+
+	for (i = 0; i < bd->rx_nr_rings; i++) {
+		struct bnge_rx_ring_info *rxr = &bn->rx_ring[i];
+
+		rxr->rx_ring_struct.ring_mem.flags =
+			BNGE_RMEM_RING_PTE_FLAG;
+		rxr->rx_agg_ring_struct.ring_mem.flags =
+			BNGE_RMEM_RING_PTE_FLAG;
+		rxr->bnapi = bn->bnapi[i];
+		bn->bnapi[i]->rx_ring = &bn->rx_ring[i];
+	}
+
+	bn->tx_ring = kcalloc(bd->tx_nr_rings,
+			      sizeof(struct bnge_tx_ring_info),
+			      GFP_KERNEL);
+	if (!bn->tx_ring)
+		goto err_free_core;
+
+	bn->tx_ring_map = kcalloc(bd->tx_nr_rings, sizeof(u16),
+				  GFP_KERNEL);
+	if (!bn->tx_ring_map)
+		goto err_free_core;
+
+	if (bd->flags & BNGE_EN_SHARED_CHNL)
+		j = 0;
+	else
+		j = bd->rx_nr_rings;
+
+	for (i = 0; i < bd->tx_nr_rings; i++) {
+		struct bnge_tx_ring_info *txr = &bn->tx_ring[i];
+		struct bnge_napi *bnapi2;
+		int k;
+
+		txr->tx_ring_struct.ring_mem.flags = BNGE_RMEM_RING_PTE_FLAG;
+		bn->tx_ring_map[i] = i;
+		k = j + BNGE_RING_TO_TC_OFF(bd, i);
+
+		bnapi2 = bn->bnapi[k];
+		txr->txq_index = i;
+		txr->tx_napi_idx =
+			BNGE_RING_TO_TC(bd, txr->txq_index);
+		bnapi2->tx_ring[txr->tx_napi_idx] = txr;
+		txr->bnapi = bnapi2;
+	}
+
+	rc = bnge_alloc_ring_stats(bn);
+	if (rc)
+		goto err_free_core;
+
+	rc = bnge_alloc_vnics(bn);
+	if (rc)
+		goto err_free_core;
+
+	rc = bnge_alloc_nq_arrays(bn);
+	if (rc)
+		goto err_free_core;
+
+	bnge_init_ring_struct(bn);
+
+	rc = bnge_alloc_rx_rings(bn);
+	if (rc)
+		goto err_free_core;
+
+	rc = bnge_alloc_tx_rings(bn);
+	if (rc)
+		goto err_free_core;
+
+	rc = bnge_alloc_nq_tree(bn);
+	if (rc)
+		goto err_free_core;
+
+	bn->vnic_info[BNGE_VNIC_DEFAULT].flags |= BNGE_VNIC_RSS_FLAG |
+						  BNGE_VNIC_MCAST_FLAG |
+						  BNGE_VNIC_UCAST_FLAG;
+	rc = bnge_alloc_vnic_attributes(bn);
+	if (rc)
+		goto err_free_core;
+	return 0;
+
+err_free_core:
+	bnge_free_core(bn);
+	return rc;
+}
+
+u16 bnge_cp_ring_for_rx(struct bnge_rx_ring_info *rxr)
+{
+	return rxr->rx_cpr->ring_struct.fw_ring_id;
+}
+
+u16 bnge_cp_ring_for_tx(struct bnge_tx_ring_info *txr)
+{
+	return txr->tx_cpr->ring_struct.fw_ring_id;
+}
+
+static void bnge_db_nq(struct bnge_net *bn, struct bnge_db_info *db, u32 idx)
+{
+	bnge_writeq(bn->bd, db->db_key64 | DBR_TYPE_NQ_MASK |
+		    DB_RING_IDX(db, idx), db->doorbell);
+}
+
+static void bnge_db_cq(struct bnge_net *bn, struct bnge_db_info *db, u32 idx)
+{
+	bnge_writeq(bn->bd, db->db_key64 | DBR_TYPE_CQ_ARMALL |
+		    DB_RING_IDX(db, idx), db->doorbell);
+}
+
+static int bnge_cp_num_to_irq_num(struct bnge_net *bn, int n)
+{
+	struct bnge_napi *bnapi = bn->bnapi[n];
+	struct bnge_nq_ring_info *nqr;
+
+	nqr = &bnapi->nq_ring;
+
+	return nqr->ring_struct.map_idx;
+}
+
+static irqreturn_t bnge_msix(int irq, void *dev_instance)
+{
+	/* NAPI scheduling to be added in a future patch */
+	return IRQ_HANDLED;
+}
+
+static void bnge_init_nq_tree(struct bnge_net *bn)
+{
+	struct bnge_dev *bd = bn->bd;
+	int i, j;
+
+	for (i = 0; i < bd->nq_nr_rings; i++) {
+		struct bnge_nq_ring_info *nqr = &bn->bnapi[i]->nq_ring;
+		struct bnge_ring_struct *ring = &nqr->ring_struct;
+
+		ring->fw_ring_id = INVALID_HW_RING_ID;
+		for (j = 0; j < nqr->cp_ring_count; j++) {
+			struct bnge_cp_ring_info *cpr = &nqr->cp_ring_arr[j];
+
+			ring = &cpr->ring_struct;
+			ring->fw_ring_id = INVALID_HW_RING_ID;
+		}
+	}
+}
+
+static netmem_ref __bnge_alloc_rx_netmem(struct bnge_net *bn,
+					 dma_addr_t *mapping,
+					 struct bnge_rx_ring_info *rxr,
+					 unsigned int *offset,
+					 gfp_t gfp)
+{
+	netmem_ref netmem;
+
+	if (PAGE_SIZE > BNGE_RX_PAGE_SIZE) {
+		netmem = page_pool_alloc_frag_netmem(rxr->page_pool, offset,
+						     BNGE_RX_PAGE_SIZE, gfp);
+	} else {
+		netmem = page_pool_alloc_netmems(rxr->page_pool, gfp);
+		*offset = 0;
+	}
+	if (!netmem)
+		return 0;
+
+	*mapping = page_pool_get_dma_addr_netmem(netmem) + *offset;
+	return netmem;
+}
+
+static u8 *__bnge_alloc_rx_frag(struct bnge_net *bn, dma_addr_t *mapping,
+				struct bnge_rx_ring_info *rxr,
+				gfp_t gfp)
+{
+	unsigned int offset;
+	struct page *page;
+
+	page = page_pool_alloc_frag(rxr->head_pool, &offset,
+				    bn->rx_buf_size, gfp);
+	if (!page)
+		return NULL;
+
+	*mapping = page_pool_get_dma_addr(page) + bn->rx_dma_offset + offset;
+	return page_address(page) + offset;
+}
+
+static int bnge_alloc_rx_data(struct bnge_net *bn,
+			      struct bnge_rx_ring_info *rxr,
+			      u16 prod, gfp_t gfp)
+{
+	struct bnge_sw_rx_bd *rx_buf = &rxr->rx_buf_ring[RING_RX(bn, prod)];
+	struct rx_bd *rxbd;
+	dma_addr_t mapping;
+	u8 *data;
+
+	rxbd = &rxr->rx_desc_ring[RX_RING(bn, prod)][RX_IDX(prod)];
+	data = __bnge_alloc_rx_frag(bn, &mapping, rxr, gfp);
+	if (!data)
+		return -ENOMEM;
+
+	rx_buf->data = data;
+	rx_buf->data_ptr = data + bn->rx_offset;
+	rx_buf->mapping = mapping;
+
+	rxbd->rx_bd_haddr = cpu_to_le64(mapping);
+
+	return 0;
+}
+
+static int bnge_alloc_one_rx_ring_bufs(struct bnge_net *bn,
+				       struct bnge_rx_ring_info *rxr,
+				       int ring_nr)
+{
+	u32 prod = rxr->rx_prod;
+	int i, rc = 0;
+
+	for (i = 0; i < bn->rx_ring_size; i++) {
+		rc = bnge_alloc_rx_data(bn, rxr, prod, GFP_KERNEL);
+		if (rc)
+			break;
+		prod = NEXT_RX(prod);
+	}
+
+	/* Abort if not a single buffer can be allocated */
+	if (rc && !i) {
+		netdev_err(bn->netdev,
+			   "RX ring %d: allocated %d/%d buffers, abort\n",
+			   ring_nr, i, bn->rx_ring_size);
+		return rc;
+	}
+
+	rxr->rx_prod = prod;
+
+	if (i < bn->rx_ring_size)
+		netdev_warn(bn->netdev,
+			    "RX ring %d: allocated %d/%d buffers, continuing\n",
+			    ring_nr, i, bn->rx_ring_size);
+	return 0;
+}
+
+static u16 bnge_find_next_agg_idx(struct bnge_rx_ring_info *rxr, u16 idx)
+{
+	u16 next, max = rxr->rx_agg_bmap_size;
+
+	next = find_next_zero_bit(rxr->rx_agg_bmap, max, idx);
+	if (next >= max)
+		next = find_first_zero_bit(rxr->rx_agg_bmap, max);
+	return next;
+}
+
+static int bnge_alloc_rx_netmem(struct bnge_net *bn,
+				struct bnge_rx_ring_info *rxr,
+				u16 prod, gfp_t gfp)
+{
+	struct bnge_sw_rx_agg_bd *rx_agg_buf;
+	u16 sw_prod = rxr->rx_sw_agg_prod;
+	unsigned int offset = 0;
+	struct rx_bd *rxbd;
+	dma_addr_t mapping;
+	netmem_ref netmem;
+
+	rxbd = &rxr->rx_agg_desc_ring[RX_AGG_RING(bn, prod)][RX_IDX(prod)];
+	netmem = __bnge_alloc_rx_netmem(bn, &mapping, rxr, &offset, gfp);
+	if (!netmem)
+		return -ENOMEM;
+
+	if (unlikely(test_bit(sw_prod, rxr->rx_agg_bmap)))
+		sw_prod = bnge_find_next_agg_idx(rxr, sw_prod);
+
+	__set_bit(sw_prod, rxr->rx_agg_bmap);
+	rx_agg_buf = &rxr->rx_agg_buf_ring[sw_prod];
+	rxr->rx_sw_agg_prod = RING_RX_AGG(bn, NEXT_RX_AGG(sw_prod));
+
+	rx_agg_buf->netmem = netmem;
+	rx_agg_buf->offset = offset;
+	rx_agg_buf->mapping = mapping;
+	rxbd->rx_bd_haddr = cpu_to_le64(mapping);
+	rxbd->rx_bd_opaque = sw_prod;
+	return 0;
+}
+
+static int bnge_alloc_one_agg_ring_bufs(struct bnge_net *bn,
+					struct bnge_rx_ring_info *rxr,
+					int ring_nr)
+{
+	u32 prod = rxr->rx_agg_prod;
+	int i, rc = 0;
+
+	for (i = 0; i < bn->rx_agg_ring_size; i++) {
+		rc = bnge_alloc_rx_netmem(bn, rxr, prod, GFP_KERNEL);
+		if (rc)
+			break;
+		prod = NEXT_RX_AGG(prod);
+	}
+
+	if (rc && i < MAX_SKB_FRAGS) {
+		netdev_err(bn->netdev,
+			   "Agg ring %d: allocated %d/%d buffers (min %d), abort\n",
+			   ring_nr, i, bn->rx_agg_ring_size, MAX_SKB_FRAGS);
+		goto err_free_one_agg_ring_bufs;
+	}
+
+	rxr->rx_agg_prod = prod;
+
+	if (i < bn->rx_agg_ring_size)
+		netdev_warn(bn->netdev,
+			    "Agg ring %d: allocated %d/%d buffers, continuing\n",
+			    ring_nr, i, bn->rx_agg_ring_size);
+	return 0;
+
+err_free_one_agg_ring_bufs:
+	bnge_free_one_agg_ring_bufs(bn, rxr);
+	return -ENOMEM;
+}
+
+static int bnge_alloc_one_rx_ring_pair_bufs(struct bnge_net *bn, int ring_nr)
+{
+	struct bnge_rx_ring_info *rxr = &bn->rx_ring[ring_nr];
+	int rc;
+
+	rc = bnge_alloc_one_rx_ring_bufs(bn, rxr, ring_nr);
+	if (rc)
+		return rc;
+
+	if (bnge_is_agg_reqd(bn->bd)) {
+		rc = bnge_alloc_one_agg_ring_bufs(bn, rxr, ring_nr);
+		if (rc)
+			goto err_free_one_rx_ring_bufs;
+	}
+	return 0;
+
+err_free_one_rx_ring_bufs:
+	bnge_free_one_rx_ring_bufs(bn, rxr);
+	return rc;
+}
+
+static void bnge_init_rxbd_pages(struct bnge_ring_struct *ring, u32 type)
+{
+	struct rx_bd **rx_desc_ring;
+	u32 prod;
+	int i;
+
+	rx_desc_ring = (struct rx_bd **)ring->ring_mem.pg_arr;
+	for (i = 0, prod = 0; i < ring->ring_mem.nr_pages; i++) {
+		struct rx_bd *rxbd = rx_desc_ring[i];
+		int j;
+
+		for (j = 0; j < RX_DESC_CNT; j++, rxbd++, prod++) {
+			rxbd->rx_bd_len_flags_type = cpu_to_le32(type);
+			rxbd->rx_bd_opaque = prod;
+		}
+	}
+}
+
+static void bnge_init_one_rx_ring_rxbd(struct bnge_net *bn,
+				       struct bnge_rx_ring_info *rxr)
+{
+	struct bnge_ring_struct *ring;
+	u32 type;
+
+	type = (bn->rx_buf_use_size << RX_BD_LEN_SHIFT) |
+		RX_BD_TYPE_RX_PACKET_BD | RX_BD_FLAGS_EOP;
+
+	if (NET_IP_ALIGN == 2)
+		type |= RX_BD_FLAGS_SOP;
+
+	ring = &rxr->rx_ring_struct;
+	bnge_init_rxbd_pages(ring, type);
+	ring->fw_ring_id = INVALID_HW_RING_ID;
+}
+
+static void bnge_init_one_agg_ring_rxbd(struct bnge_net *bn,
+					struct bnge_rx_ring_info *rxr)
+{
+	struct bnge_ring_struct *ring;
+	u32 type;
+
+	ring = &rxr->rx_agg_ring_struct;
+	ring->fw_ring_id = INVALID_HW_RING_ID;
+	if (bnge_is_agg_reqd(bn->bd)) {
+		type = ((u32)BNGE_RX_PAGE_SIZE << RX_BD_LEN_SHIFT) |
+			RX_BD_TYPE_RX_AGG_BD | RX_BD_FLAGS_SOP;
+
+		bnge_init_rxbd_pages(ring, type);
+	}
+}
+
+static void bnge_init_one_rx_ring_pair(struct bnge_net *bn, int ring_nr)
+{
+	struct bnge_rx_ring_info *rxr;
+
+	rxr = &bn->rx_ring[ring_nr];
+	bnge_init_one_rx_ring_rxbd(bn, rxr);
+
+	netif_queue_set_napi(bn->netdev, ring_nr, NETDEV_QUEUE_TYPE_RX,
+			     &rxr->bnapi->napi);
+
+	bnge_init_one_agg_ring_rxbd(bn, rxr);
+}
+
+static int bnge_alloc_rx_ring_pair_bufs(struct bnge_net *bn)
+{
+	int i, rc;
+
+	for (i = 0; i < bn->bd->rx_nr_rings; i++) {
+		rc = bnge_alloc_one_rx_ring_pair_bufs(bn, i);
+		if (rc)
+			goto err_free_rx_ring_pair_bufs;
+	}
+	return 0;
+
+err_free_rx_ring_pair_bufs:
+	bnge_free_rx_ring_pair_bufs(bn);
+	return rc;
+}
+
+static void bnge_init_rx_rings(struct bnge_net *bn)
+{
+	int i;
+
+#define BNGE_RX_OFFSET (NET_SKB_PAD + NET_IP_ALIGN)
+#define BNGE_RX_DMA_OFFSET NET_SKB_PAD
+	bn->rx_offset = BNGE_RX_OFFSET;
+	bn->rx_dma_offset = BNGE_RX_DMA_OFFSET;
+
+	for (i = 0; i < bn->bd->rx_nr_rings; i++)
+		bnge_init_one_rx_ring_pair(bn, i);
+}
+
+static void bnge_init_tx_rings(struct bnge_net *bn)
+{
+	int i;
+
+	bn->tx_wake_thresh = max(bn->tx_ring_size / 2, BNGE_MIN_TX_DESC_CNT);
+
+	for (i = 0; i < bn->bd->tx_nr_rings; i++) {
+		struct bnge_tx_ring_info *txr = &bn->tx_ring[i];
+		struct bnge_ring_struct *ring = &txr->tx_ring_struct;
+
+		ring->fw_ring_id = INVALID_HW_RING_ID;
+
+		netif_queue_set_napi(bn->netdev, i, NETDEV_QUEUE_TYPE_TX,
+				     &txr->bnapi->napi);
+	}
+}
+
+static void bnge_init_vnics(struct bnge_net *bn)
+{
+	struct bnge_vnic_info *vnic0 = &bn->vnic_info[BNGE_VNIC_DEFAULT];
+	int i;
+
+	for (i = 0; i < bn->nr_vnics; i++) {
+		struct bnge_vnic_info *vnic = &bn->vnic_info[i];
+		int j;
+
+		vnic->fw_vnic_id = INVALID_HW_RING_ID;
+		vnic->vnic_id = i;
+		for (j = 0; j < BNGE_MAX_CTX_PER_VNIC; j++)
+			vnic->fw_rss_cos_lb_ctx[j] = INVALID_HW_RING_ID;
+
+		if (bn->vnic_info[i].rss_hash_key) {
+			if (i == BNGE_VNIC_DEFAULT) {
+				u8 *key = (void *)vnic->rss_hash_key;
+				int k;
+
+				if (!bn->rss_hash_key_valid &&
+				    !bn->rss_hash_key_updated) {
+					get_random_bytes(bn->rss_hash_key,
+							 HW_HASH_KEY_SIZE);
+					bn->rss_hash_key_updated = true;
+				}
+
+				memcpy(vnic->rss_hash_key, bn->rss_hash_key,
+				       HW_HASH_KEY_SIZE);
+
+				if (!bn->rss_hash_key_updated)
+					continue;
+
+				bn->rss_hash_key_updated = false;
+				bn->rss_hash_key_valid = true;
+
+				bn->toeplitz_prefix = 0;
+				for (k = 0; k < 8; k++) {
+					bn->toeplitz_prefix <<= 8;
+					bn->toeplitz_prefix |= key[k];
+				}
+			} else {
+				memcpy(vnic->rss_hash_key, vnic0->rss_hash_key,
+				       HW_HASH_KEY_SIZE);
+			}
+		}
+	}
+}
+
+static void bnge_set_db_mask(struct bnge_net *bn, struct bnge_db_info *db,
+			     u32 ring_type)
+{
+	switch (ring_type) {
+	case HWRM_RING_ALLOC_TX:
+		db->db_ring_mask = bn->tx_ring_mask;
+		break;
+	case HWRM_RING_ALLOC_RX:
+		db->db_ring_mask = bn->rx_ring_mask;
+		break;
+	case HWRM_RING_ALLOC_AGG:
+		db->db_ring_mask = bn->rx_agg_ring_mask;
+		break;
+	case HWRM_RING_ALLOC_CMPL:
+	case HWRM_RING_ALLOC_NQ:
+		db->db_ring_mask = bn->cp_ring_mask;
+		break;
+	}
+	db->db_epoch_mask = db->db_ring_mask + 1;
+	db->db_epoch_shift = DBR_EPOCH_SFT - ilog2(db->db_epoch_mask);
+}
+
+static void bnge_set_db(struct bnge_net *bn, struct bnge_db_info *db,
+			u32 ring_type, u32 map_idx, u32 xid)
+{
+	struct bnge_dev *bd = bn->bd;
+
+	switch (ring_type) {
+	case HWRM_RING_ALLOC_TX:
+		db->db_key64 = DBR_PATH_L2 | DBR_TYPE_SQ;
+		break;
+	case HWRM_RING_ALLOC_RX:
+	case HWRM_RING_ALLOC_AGG:
+		db->db_key64 = DBR_PATH_L2 | DBR_TYPE_SRQ;
+		break;
+	case HWRM_RING_ALLOC_CMPL:
+		db->db_key64 = DBR_PATH_L2;
+		break;
+	case HWRM_RING_ALLOC_NQ:
+		db->db_key64 = DBR_PATH_L2;
+		break;
+	}
+	db->db_key64 |= ((u64)xid << DBR_XID_SFT) | DBR_VALID;
+
+	db->doorbell = bd->bar1 + bd->db_offset;
+	bnge_set_db_mask(bn, db, ring_type);
+}
+
+static int bnge_hwrm_cp_ring_alloc(struct bnge_net *bn,
+				   struct bnge_cp_ring_info *cpr)
+{
+	const u32 type = HWRM_RING_ALLOC_CMPL;
+	struct bnge_napi *bnapi = cpr->bnapi;
+	struct bnge_ring_struct *ring;
+	u32 map_idx = bnapi->index;
+	int rc;
+
+	ring = &cpr->ring_struct;
+	ring->handle = BNGE_SET_NQ_HDL(cpr);
+	rc = hwrm_ring_alloc_send_msg(bn, ring, type, map_idx);
+	if (rc)
+		return rc;
+
+	bnge_set_db(bn, &cpr->cp_db, type, map_idx, ring->fw_ring_id);
+	bnge_db_cq(bn, &cpr->cp_db, cpr->cp_raw_cons);
+
+	return 0;
+}
+
+static int bnge_hwrm_tx_ring_alloc(struct bnge_net *bn,
+				   struct bnge_tx_ring_info *txr, u32 tx_idx)
+{
+	struct bnge_ring_struct *ring = &txr->tx_ring_struct;
+	const u32 type = HWRM_RING_ALLOC_TX;
+	int rc;
+
+	rc = hwrm_ring_alloc_send_msg(bn, ring, type, tx_idx);
+	if (rc)
+		return rc;
+
+	bnge_set_db(bn, &txr->tx_db, type, tx_idx, ring->fw_ring_id);
+
+	return 0;
+}
+
+static int bnge_hwrm_rx_agg_ring_alloc(struct bnge_net *bn,
+				       struct bnge_rx_ring_info *rxr)
+{
+	struct bnge_ring_struct *ring = &rxr->rx_agg_ring_struct;
+	u32 type = HWRM_RING_ALLOC_AGG;
+	struct bnge_dev *bd = bn->bd;
+	u32 grp_idx = ring->grp_idx;
+	u32 map_idx;
+	int rc;
+
+	map_idx = grp_idx + bd->rx_nr_rings;
+	rc = hwrm_ring_alloc_send_msg(bn, ring, type, map_idx);
+	if (rc)
+		return rc;
+
+	bnge_set_db(bn, &rxr->rx_agg_db, type, map_idx,
+		    ring->fw_ring_id);
+	bnge_db_write(bn->bd, &rxr->rx_agg_db, rxr->rx_agg_prod);
+	bnge_db_write(bn->bd, &rxr->rx_db, rxr->rx_prod);
+	bn->grp_info[grp_idx].agg_fw_ring_id = ring->fw_ring_id;
+
+	return 0;
+}
+
+static int bnge_hwrm_rx_ring_alloc(struct bnge_net *bn,
+				   struct bnge_rx_ring_info *rxr)
+{
+	struct bnge_ring_struct *ring = &rxr->rx_ring_struct;
+	struct bnge_napi *bnapi = rxr->bnapi;
+	u32 type = HWRM_RING_ALLOC_RX;
+	u32 map_idx = bnapi->index;
+	int rc;
+
+	rc = hwrm_ring_alloc_send_msg(bn, ring, type, map_idx);
+	if (rc)
+		return rc;
+
+	bnge_set_db(bn, &rxr->rx_db, type, map_idx, ring->fw_ring_id);
+	bn->grp_info[map_idx].rx_fw_ring_id = ring->fw_ring_id;
+
+	return 0;
+}
+
+static int bnge_hwrm_ring_alloc(struct bnge_net *bn)
+{
+	struct bnge_dev *bd = bn->bd;
+	bool agg_rings;
+	int i, rc = 0;
+
+	agg_rings = !!(bnge_is_agg_reqd(bd));
+	for (i = 0; i < bd->nq_nr_rings; i++) {
+		struct bnge_napi *bnapi = bn->bnapi[i];
+		struct bnge_nq_ring_info *nqr = &bnapi->nq_ring;
+		struct bnge_ring_struct *ring = &nqr->ring_struct;
+		u32 type = HWRM_RING_ALLOC_NQ;
+		u32 map_idx = ring->map_idx;
+		unsigned int vector;
+
+		vector = bd->irq_tbl[map_idx].vector;
+		disable_irq_nosync(vector);
+		rc = hwrm_ring_alloc_send_msg(bn, ring, type, map_idx);
+		if (rc) {
+			enable_irq(vector);
+			goto err_out;
+		}
+		bnge_set_db(bn, &nqr->nq_db, type, map_idx, ring->fw_ring_id);
+		bnge_db_nq(bn, &nqr->nq_db, nqr->nq_raw_cons);
+		enable_irq(vector);
+		bn->grp_info[i].nq_fw_ring_id = ring->fw_ring_id;
+
+		if (!i) {
+			rc = bnge_hwrm_set_async_event_cr(bd, ring->fw_ring_id);
+			if (rc)
+				netdev_warn(bn->netdev, "Failed to set async event completion ring.\n");
+		}
+	}
+
+	for (i = 0; i < bd->tx_nr_rings; i++) {
+		struct bnge_tx_ring_info *txr = &bn->tx_ring[i];
+
+		rc = bnge_hwrm_cp_ring_alloc(bn, txr->tx_cpr);
+		if (rc)
+			goto err_out;
+		rc = bnge_hwrm_tx_ring_alloc(bn, txr, i);
+		if (rc)
+			goto err_out;
+	}
+
+	for (i = 0; i < bd->rx_nr_rings; i++) {
+		struct bnge_rx_ring_info *rxr = &bn->rx_ring[i];
+		struct bnge_cp_ring_info *cpr;
+		struct bnge_ring_struct *ring;
+		struct bnge_napi *bnapi;
+		u32 map_idx, type;
+
+		rc = bnge_hwrm_rx_ring_alloc(bn, rxr);
+		if (rc)
+			goto err_out;
+		/* If we have agg rings, post agg buffers first. */
+		if (!agg_rings)
+			bnge_db_write(bn->bd, &rxr->rx_db, rxr->rx_prod);
+
+		cpr = rxr->rx_cpr;
+		bnapi = rxr->bnapi;
+		type = HWRM_RING_ALLOC_CMPL;
+		map_idx = bnapi->index;
+
+		ring = &cpr->ring_struct;
+		ring->handle = BNGE_SET_NQ_HDL(cpr);
+		rc = hwrm_ring_alloc_send_msg(bn, ring, type, map_idx);
+		if (rc)
+			goto err_out;
+		bnge_set_db(bn, &cpr->cp_db, type, map_idx,
+			    ring->fw_ring_id);
+		bnge_db_cq(bn, &cpr->cp_db, cpr->cp_raw_cons);
+	}
+
+	if (agg_rings) {
+		for (i = 0; i < bd->rx_nr_rings; i++) {
+			rc = bnge_hwrm_rx_agg_ring_alloc(bn, &bn->rx_ring[i]);
+			if (rc)
+				goto err_out;
+		}
+	}
+err_out:
+	return rc;
+}
+
+void bnge_fill_hw_rss_tbl(struct bnge_net *bn, struct bnge_vnic_info *vnic)
+{
+	__le16 *ring_tbl = vnic->rss_table;
+	struct bnge_rx_ring_info *rxr;
+	struct bnge_dev *bd = bn->bd;
+	u16 tbl_size, i;
+
+	tbl_size = bnge_get_rxfh_indir_size(bd);
+
+	for (i = 0; i < tbl_size; i++) {
+		u16 ring_id, j;
+
+		j = bd->rss_indir_tbl[i];
+		rxr = &bn->rx_ring[j];
+
+		ring_id = rxr->rx_ring_struct.fw_ring_id;
+		*ring_tbl++ = cpu_to_le16(ring_id);
+		ring_id = bnge_cp_ring_for_rx(rxr);
+		*ring_tbl++ = cpu_to_le16(ring_id);
+	}
+}
+
+static int bnge_hwrm_vnic_rss_cfg(struct bnge_net *bn,
+				  struct bnge_vnic_info *vnic)
+{
+	int rc;
+
+	rc = bnge_hwrm_vnic_set_rss(bn, vnic, true);
+	if (rc) {
+		netdev_err(bn->netdev, "hwrm vnic %d set rss failure rc: %d\n",
+			   vnic->vnic_id, rc);
+		return rc;
+	}
+	rc = bnge_hwrm_vnic_cfg(bn, vnic);
+	if (rc)
+		netdev_err(bn->netdev, "hwrm vnic %d cfg failure rc: %d\n",
+			   vnic->vnic_id, rc);
+	return rc;
+}
+
+static int bnge_setup_vnic(struct bnge_net *bn, struct bnge_vnic_info *vnic)
+{
+	struct bnge_dev *bd = bn->bd;
+	int rc, i, nr_ctxs;
+
+	nr_ctxs = bnge_cal_nr_rss_ctxs(bd->rx_nr_rings);
+	for (i = 0; i < nr_ctxs; i++) {
+		rc = bnge_hwrm_vnic_ctx_alloc(bd, vnic, i);
+		if (rc) {
+			netdev_err(bn->netdev, "hwrm vnic %d ctx %d alloc failure rc: %d\n",
+				   vnic->vnic_id, i, rc);
+			return -ENOMEM;
+		}
+		bn->rsscos_nr_ctxs++;
+	}
+
+	rc = bnge_hwrm_vnic_rss_cfg(bn, vnic);
+	if (rc)
+		return rc;
+
+	if (bnge_is_agg_reqd(bd)) {
+		rc = bnge_hwrm_vnic_set_hds(bn, vnic);
+		if (rc)
+			netdev_err(bn->netdev, "hwrm vnic %d set hds failure rc: %d\n",
+				   vnic->vnic_id, rc);
+	}
+	return rc;
+}
+
+static void bnge_del_l2_filter(struct bnge_net *bn, struct bnge_l2_filter *fltr)
+{
+	if (!refcount_dec_and_test(&fltr->refcnt))
+		return;
+	hlist_del_rcu(&fltr->base.hash);
+	kfree_rcu(fltr, base.rcu);
+}
+
+static void bnge_init_l2_filter(struct bnge_net *bn,
+				struct bnge_l2_filter *fltr,
+				struct bnge_l2_key *key, u32 idx)
+{
+	struct hlist_head *head;
+
+	ether_addr_copy(fltr->l2_key.dst_mac_addr, key->dst_mac_addr);
+	fltr->l2_key.vlan = key->vlan;
+	fltr->base.type = BNGE_FLTR_TYPE_L2;
+
+	head = &bn->l2_fltr_hash_tbl[idx];
+	hlist_add_head_rcu(&fltr->base.hash, head);
+	refcount_set(&fltr->refcnt, 1);
+}
+
+static struct bnge_l2_filter *__bnge_lookup_l2_filter(struct bnge_net *bn,
+						      struct bnge_l2_key *key,
+						      u32 idx)
+{
+	struct bnge_l2_filter *fltr;
+	struct hlist_head *head;
+
+	head = &bn->l2_fltr_hash_tbl[idx];
+	hlist_for_each_entry_rcu(fltr, head, base.hash) {
+		struct bnge_l2_key *l2_key = &fltr->l2_key;
+
+		if (ether_addr_equal(l2_key->dst_mac_addr, key->dst_mac_addr) &&
+		    l2_key->vlan == key->vlan)
+			return fltr;
+	}
+	return NULL;
+}
+
+static struct bnge_l2_filter *bnge_lookup_l2_filter(struct bnge_net *bn,
+						    struct bnge_l2_key *key,
+						    u32 idx)
+{
+	struct bnge_l2_filter *fltr;
+
+	rcu_read_lock();
+	fltr = __bnge_lookup_l2_filter(bn, key, idx);
+	if (fltr)
+		refcount_inc(&fltr->refcnt);
+	rcu_read_unlock();
+	return fltr;
+}
+
+static struct bnge_l2_filter *bnge_alloc_l2_filter(struct bnge_net *bn,
+						   struct bnge_l2_key *key,
+						   gfp_t gfp)
+{
+	struct bnge_l2_filter *fltr;
+	u32 idx;
+
+	idx = jhash2(&key->filter_key, BNGE_L2_KEY_SIZE, bn->hash_seed) &
+	      BNGE_L2_FLTR_HASH_MASK;
+	fltr = bnge_lookup_l2_filter(bn, key, idx);
+	if (fltr)
+		return fltr;
+
+	fltr = kzalloc(sizeof(*fltr), gfp);
+	if (!fltr)
+		return ERR_PTR(-ENOMEM);
+
+	bnge_init_l2_filter(bn, fltr, key, idx);
+	return fltr;
+}
+
+static int bnge_hwrm_set_vnic_filter(struct bnge_net *bn, u16 vnic_id, u16 idx,
+				     const u8 *mac_addr)
+{
+	struct bnge_l2_filter *fltr;
+	struct bnge_l2_key key;
+	int rc;
+
+	ether_addr_copy(key.dst_mac_addr, mac_addr);
+	key.vlan = 0;
+	fltr = bnge_alloc_l2_filter(bn, &key, GFP_KERNEL);
+	if (IS_ERR(fltr))
+		return PTR_ERR(fltr);
+
+	fltr->base.fw_vnic_id = bn->vnic_info[vnic_id].fw_vnic_id;
+	rc = bnge_hwrm_l2_filter_alloc(bn->bd, fltr);
+	if (rc)
+		goto err_del_l2_filter;
+	bn->vnic_info[vnic_id].l2_filters[idx] = fltr;
+	return rc;
+
+err_del_l2_filter:
+	bnge_del_l2_filter(bn, fltr);
+	return rc;
+}
+
+static bool bnge_mc_list_updated(struct bnge_net *bn, u32 *rx_mask)
+{
+	struct bnge_vnic_info *vnic = &bn->vnic_info[BNGE_VNIC_DEFAULT];
+	struct net_device *dev = bn->netdev;
+	struct netdev_hw_addr *ha;
+	int mc_count = 0, off = 0;
+	bool update = false;
+	u8 *haddr;
+
+	netdev_for_each_mc_addr(ha, dev) {
+		if (mc_count >= BNGE_MAX_MC_ADDRS) {
+			*rx_mask |= CFA_L2_SET_RX_MASK_REQ_MASK_ALL_MCAST;
+			vnic->mc_list_count = 0;
+			return false;
+		}
+		haddr = ha->addr;
+		if (!ether_addr_equal(haddr, vnic->mc_list + off)) {
+			memcpy(vnic->mc_list + off, haddr, ETH_ALEN);
+			update = true;
+		}
+		off += ETH_ALEN;
+		mc_count++;
+	}
+	if (mc_count)
+		*rx_mask |= CFA_L2_SET_RX_MASK_REQ_MASK_MCAST;
+
+	if (mc_count != vnic->mc_list_count) {
+		vnic->mc_list_count = mc_count;
+		update = true;
+	}
+	return update;
+}
+
+static bool bnge_uc_list_updated(struct bnge_net *bn)
+{
+	struct bnge_vnic_info *vnic = &bn->vnic_info[BNGE_VNIC_DEFAULT];
+	struct net_device *dev = bn->netdev;
+	struct netdev_hw_addr *ha;
+	int off = 0;
+
+	if (netdev_uc_count(dev) != (vnic->uc_filter_count - 1))
+		return true;
+
+	netdev_for_each_uc_addr(ha, dev) {
+		if (!ether_addr_equal(ha->addr, vnic->uc_list + off))
+			return true;
+
+		off += ETH_ALEN;
+	}
+	return false;
+}
+
+static bool bnge_promisc_ok(struct bnge_net *bn)
+{
+	return true;
+}
+
+static int bnge_cfg_def_vnic(struct bnge_net *bn)
+{
+	struct bnge_vnic_info *vnic = &bn->vnic_info[BNGE_VNIC_DEFAULT];
+	struct net_device *dev = bn->netdev;
+	struct bnge_dev *bd = bn->bd;
+	struct netdev_hw_addr *ha;
+	int i, off = 0, rc;
+	bool uc_update;
+
+	netif_addr_lock_bh(dev);
+	uc_update = bnge_uc_list_updated(bn);
+	netif_addr_unlock_bh(dev);
+
+	if (!uc_update)
+		goto skip_uc;
+
+	for (i = 1; i < vnic->uc_filter_count; i++) {
+		struct bnge_l2_filter *fltr = vnic->l2_filters[i];
+
+		bnge_hwrm_l2_filter_free(bd, fltr);
+		bnge_del_l2_filter(bn, fltr);
+	}
+
+	vnic->uc_filter_count = 1;
+
+	netif_addr_lock_bh(dev);
+	if (netdev_uc_count(dev) > (BNGE_MAX_UC_ADDRS - 1)) {
+		vnic->rx_mask |= CFA_L2_SET_RX_MASK_REQ_MASK_PROMISCUOUS;
+	} else {
+		netdev_for_each_uc_addr(ha, dev) {
+			memcpy(vnic->uc_list + off, ha->addr, ETH_ALEN);
+			off += ETH_ALEN;
+			vnic->uc_filter_count++;
+		}
+	}
+	netif_addr_unlock_bh(dev);
+
+	for (i = 1, off = 0; i < vnic->uc_filter_count; i++, off += ETH_ALEN) {
+		rc = bnge_hwrm_set_vnic_filter(bn, 0, i, vnic->uc_list + off);
+		if (rc) {
+			netdev_err(dev, "HWRM vnic filter failure rc: %d\n", rc);
+			vnic->uc_filter_count = i;
+			return rc;
+		}
+	}
+
+skip_uc:
+	if ((vnic->rx_mask & CFA_L2_SET_RX_MASK_REQ_MASK_PROMISCUOUS) &&
+	    !bnge_promisc_ok(bn))
+		vnic->rx_mask &= ~CFA_L2_SET_RX_MASK_REQ_MASK_PROMISCUOUS;
+	rc = bnge_hwrm_cfa_l2_set_rx_mask(bd, vnic);
+	if (rc && (vnic->rx_mask & CFA_L2_SET_RX_MASK_REQ_MASK_MCAST)) {
+		netdev_info(dev, "Failed setting MC filters rc: %d, turning on ALL_MCAST mode\n",
+			    rc);
+		vnic->rx_mask &= ~CFA_L2_SET_RX_MASK_REQ_MASK_MCAST;
+		vnic->rx_mask |= CFA_L2_SET_RX_MASK_REQ_MASK_ALL_MCAST;
+		vnic->mc_list_count = 0;
+		rc = bnge_hwrm_cfa_l2_set_rx_mask(bd, vnic);
+	}
+	if (rc)
+		netdev_err(dev, "HWRM cfa l2 rx mask failure rc: %d\n",
+			   rc);
+
+	return rc;
+}
+
+static void bnge_hwrm_vnic_free(struct bnge_net *bn)
+{
+	int i;
+
+	for (i = 0; i < bn->nr_vnics; i++)
+		bnge_hwrm_vnic_free_one(bn->bd, &bn->vnic_info[i]);
+}
+
+static void bnge_hwrm_vnic_ctx_free(struct bnge_net *bn)
+{
+	int i, j;
+
+	for (i = 0; i < bn->nr_vnics; i++) {
+		struct bnge_vnic_info *vnic = &bn->vnic_info[i];
+
+		for (j = 0; j < BNGE_MAX_CTX_PER_VNIC; j++) {
+			if (vnic->fw_rss_cos_lb_ctx[j] != INVALID_HW_RING_ID)
+				bnge_hwrm_vnic_ctx_free_one(bn->bd, vnic, j);
+		}
+	}
+	bn->rsscos_nr_ctxs = 0;
+}
+
+static void bnge_hwrm_clear_vnic_filter(struct bnge_net *bn)
+{
+	struct bnge_vnic_info *vnic = &bn->vnic_info[BNGE_VNIC_DEFAULT];
+	int i;
+
+	for (i = 0; i < vnic->uc_filter_count; i++) {
+		struct bnge_l2_filter *fltr = vnic->l2_filters[i];
+
+		bnge_hwrm_l2_filter_free(bn->bd, fltr);
+		bnge_del_l2_filter(bn, fltr);
+	}
+
+	vnic->uc_filter_count = 0;
+}
+
+static void bnge_clear_vnic(struct bnge_net *bn)
+{
+	bnge_hwrm_clear_vnic_filter(bn);
+	bnge_hwrm_vnic_free(bn);
+	bnge_hwrm_vnic_ctx_free(bn);
+}
+
+static void bnge_hwrm_rx_ring_free(struct bnge_net *bn,
+				   struct bnge_rx_ring_info *rxr,
+				   bool close_path)
+{
+	struct bnge_ring_struct *ring = &rxr->rx_ring_struct;
+	u32 grp_idx = rxr->bnapi->index;
+	u32 cmpl_ring_id;
+
+	if (ring->fw_ring_id == INVALID_HW_RING_ID)
+		return;
+
+	cmpl_ring_id = bnge_cp_ring_for_rx(rxr);
+	hwrm_ring_free_send_msg(bn, ring,
+				RING_FREE_REQ_RING_TYPE_RX,
+				close_path ? cmpl_ring_id :
+				INVALID_HW_RING_ID);
+	ring->fw_ring_id = INVALID_HW_RING_ID;
+	bn->grp_info[grp_idx].rx_fw_ring_id = INVALID_HW_RING_ID;
+}
+
+static void bnge_hwrm_rx_agg_ring_free(struct bnge_net *bn,
+				       struct bnge_rx_ring_info *rxr,
+				       bool close_path)
+{
+	struct bnge_ring_struct *ring = &rxr->rx_agg_ring_struct;
+	u32 grp_idx = rxr->bnapi->index;
+	u32 cmpl_ring_id;
+
+	if (ring->fw_ring_id == INVALID_HW_RING_ID)
+		return;
+
+	cmpl_ring_id = bnge_cp_ring_for_rx(rxr);
+	hwrm_ring_free_send_msg(bn, ring, RING_FREE_REQ_RING_TYPE_RX_AGG,
+				close_path ? cmpl_ring_id :
+				INVALID_HW_RING_ID);
+	ring->fw_ring_id = INVALID_HW_RING_ID;
+	bn->grp_info[grp_idx].agg_fw_ring_id = INVALID_HW_RING_ID;
+}
+
+static void bnge_hwrm_tx_ring_free(struct bnge_net *bn,
+				   struct bnge_tx_ring_info *txr,
+				   bool close_path)
+{
+	struct bnge_ring_struct *ring = &txr->tx_ring_struct;
+	u32 cmpl_ring_id;
+
+	if (ring->fw_ring_id == INVALID_HW_RING_ID)
+		return;
+
+	cmpl_ring_id = close_path ? bnge_cp_ring_for_tx(txr) :
+		       INVALID_HW_RING_ID;
+	hwrm_ring_free_send_msg(bn, ring, RING_FREE_REQ_RING_TYPE_TX,
+				cmpl_ring_id);
+	ring->fw_ring_id = INVALID_HW_RING_ID;
+}
+
+static void bnge_hwrm_cp_ring_free(struct bnge_net *bn,
+				   struct bnge_cp_ring_info *cpr)
+{
+	struct bnge_ring_struct *ring;
+
+	ring = &cpr->ring_struct;
+	if (ring->fw_ring_id == INVALID_HW_RING_ID)
+		return;
+
+	hwrm_ring_free_send_msg(bn, ring, RING_FREE_REQ_RING_TYPE_L2_CMPL,
+				INVALID_HW_RING_ID);
+	ring->fw_ring_id = INVALID_HW_RING_ID;
+}
+
+static void bnge_hwrm_ring_free(struct bnge_net *bn, bool close_path)
+{
+	struct bnge_dev *bd = bn->bd;
+	int i;
+
+	if (!bn->bnapi)
+		return;
+
+	for (i = 0; i < bd->tx_nr_rings; i++)
+		bnge_hwrm_tx_ring_free(bn, &bn->tx_ring[i], close_path);
+
+	for (i = 0; i < bd->rx_nr_rings; i++) {
+		bnge_hwrm_rx_ring_free(bn, &bn->rx_ring[i], close_path);
+		bnge_hwrm_rx_agg_ring_free(bn, &bn->rx_ring[i], close_path);
+	}
+
+	for (i = 0; i < bd->nq_nr_rings; i++) {
+		struct bnge_napi *bnapi = bn->bnapi[i];
+		struct bnge_nq_ring_info *nqr;
+		struct bnge_ring_struct *ring;
+		int j;
+
+		nqr = &bnapi->nq_ring;
+		for (j = 0; j < nqr->cp_ring_count && nqr->cp_ring_arr; j++)
+			bnge_hwrm_cp_ring_free(bn, &nqr->cp_ring_arr[j]);
+
+		ring = &nqr->ring_struct;
+		if (ring->fw_ring_id != INVALID_HW_RING_ID) {
+			hwrm_ring_free_send_msg(bn, ring,
+						RING_FREE_REQ_RING_TYPE_NQ,
+						INVALID_HW_RING_ID);
+			ring->fw_ring_id = INVALID_HW_RING_ID;
+			bn->grp_info[i].nq_fw_ring_id = INVALID_HW_RING_ID;
+		}
+	}
+}
+
+static void bnge_setup_msix(struct bnge_net *bn)
+{
+	struct net_device *dev = bn->netdev;
+	struct bnge_dev *bd = bn->bd;
+	int len, i;
+
+	len = sizeof(bd->irq_tbl[0].name);
+	for (i = 0; i < bd->nq_nr_rings; i++) {
+		int map_idx = bnge_cp_num_to_irq_num(bn, i);
+		char *attr;
+
+		if (bd->flags & BNGE_EN_SHARED_CHNL)
+			attr = "TxRx";
+		else if (i < bd->rx_nr_rings)
+			attr = "rx";
+		else
+			attr = "tx";
+
+		snprintf(bd->irq_tbl[map_idx].name, len, "%s-%s-%d", dev->name,
+			 attr, i);
+		bd->irq_tbl[map_idx].handler = bnge_msix;
+	}
+}
+
+static int bnge_setup_interrupts(struct bnge_net *bn)
+{
+	struct net_device *dev = bn->netdev;
+	struct bnge_dev *bd = bn->bd;
+
+	bnge_setup_msix(bn);
+
+	return netif_set_real_num_queues(dev, bd->tx_nr_rings, bd->rx_nr_rings);
+}
+
+static void bnge_hwrm_resource_free(struct bnge_net *bn, bool close_path)
+{
+	bnge_clear_vnic(bn);
+	bnge_hwrm_ring_free(bn, close_path);
+	bnge_hwrm_stat_ctx_free(bn);
+}
+
+static void bnge_free_irq(struct bnge_net *bn)
+{
+	struct bnge_dev *bd = bn->bd;
+	struct bnge_irq *irq;
+	int i;
+
+	for (i = 0; i < bd->nq_nr_rings; i++) {
+		int map_idx = bnge_cp_num_to_irq_num(bn, i);
+
+		irq = &bd->irq_tbl[map_idx];
+		if (irq->requested) {
+			if (irq->have_cpumask) {
+				irq_set_affinity_hint(irq->vector, NULL);
+				free_cpumask_var(irq->cpu_mask);
+				irq->have_cpumask = 0;
+			}
+			free_irq(irq->vector, bn->bnapi[i]);
+		}
+
+		irq->requested = 0;
+	}
+}
+
+static int bnge_request_irq(struct bnge_net *bn)
+{
+	struct bnge_dev *bd = bn->bd;
+	int i, rc;
+
+	rc = bnge_setup_interrupts(bn);
+	if (rc) {
+		netdev_err(bn->netdev, "bnge_setup_interrupts err: %d\n", rc);
+		return rc;
+	}
+	for (i = 0; i < bd->nq_nr_rings; i++) {
+		int map_idx = bnge_cp_num_to_irq_num(bn, i);
+		struct bnge_irq *irq = &bd->irq_tbl[map_idx];
+
+		rc = request_irq(irq->vector, irq->handler, 0, irq->name,
+				 bn->bnapi[i]);
+		if (rc)
+			goto err_free_irq;
+
+		netif_napi_set_irq_locked(&bn->bnapi[i]->napi, irq->vector);
+		irq->requested = 1;
+
+		if (zalloc_cpumask_var(&irq->cpu_mask, GFP_KERNEL)) {
+			int numa_node = dev_to_node(&bd->pdev->dev);
+
+			irq->have_cpumask = 1;
+			cpumask_set_cpu(cpumask_local_spread(i, numa_node),
+					irq->cpu_mask);
+			rc = irq_set_affinity_hint(irq->vector, irq->cpu_mask);
+			if (rc) {
+				netdev_warn(bn->netdev,
+					    "Set affinity failed, IRQ = %d\n",
+					    irq->vector);
+				goto err_free_irq;
+			}
+		}
+	}
+	return 0;
+
+err_free_irq:
+	bnge_free_irq(bn);
+	return rc;
+}
+
+static int bnge_init_chip(struct bnge_net *bn)
+{
+	struct bnge_vnic_info *vnic = &bn->vnic_info[BNGE_VNIC_DEFAULT];
+	struct bnge_dev *bd = bn->bd;
+	int rc;
+
+#define BNGE_DEF_STATS_COAL_TICKS	 1000000
+	bn->stats_coal_ticks = BNGE_DEF_STATS_COAL_TICKS;
+
+	rc = bnge_hwrm_stat_ctx_alloc(bn);
+	if (rc) {
+		netdev_err(bn->netdev, "hwrm stat ctx alloc failure rc: %d\n", rc);
+		goto err_out;
+	}
+
+	rc = bnge_hwrm_ring_alloc(bn);
+	if (rc) {
+		netdev_err(bn->netdev, "hwrm ring alloc failure rc: %d\n", rc);
+		goto err_out;
+	}
+
+	rc = bnge_hwrm_vnic_alloc(bd, vnic, bd->rx_nr_rings);
+	if (rc) {
+		netdev_err(bn->netdev, "hwrm vnic alloc failure rc: %d\n", rc);
+		goto err_out;
+	}
+
+	rc = bnge_setup_vnic(bn, vnic);
+	if (rc)
+		goto err_out;
+
+	if (bd->rss_cap & BNGE_RSS_CAP_RSS_HASH_TYPE_DELTA)
+		bnge_hwrm_update_rss_hash_cfg(bn);
+
+	/* Filter for default vnic 0 */
+	rc = bnge_hwrm_set_vnic_filter(bn, 0, 0, bn->netdev->dev_addr);
+	if (rc) {
+		netdev_err(bn->netdev, "HWRM vnic filter failure rc: %d\n", rc);
+		goto err_out;
+	}
+	vnic->uc_filter_count = 1;
+
+	vnic->rx_mask = 0;
+
+	if (bn->netdev->flags & IFF_BROADCAST)
+		vnic->rx_mask |= CFA_L2_SET_RX_MASK_REQ_MASK_BCAST;
+
+	if (bn->netdev->flags & IFF_PROMISC)
+		vnic->rx_mask |= CFA_L2_SET_RX_MASK_REQ_MASK_PROMISCUOUS;
+
+	if (bn->netdev->flags & IFF_ALLMULTI) {
+		vnic->rx_mask |= CFA_L2_SET_RX_MASK_REQ_MASK_ALL_MCAST;
+		vnic->mc_list_count = 0;
+	} else if (bn->netdev->flags & IFF_MULTICAST) {
+		u32 mask = 0;
+
+		bnge_mc_list_updated(bn, &mask);
+		vnic->rx_mask |= mask;
+	}
+
+	rc = bnge_cfg_def_vnic(bn);
+	if (rc)
+		goto err_out;
+	return 0;
+
+err_out:
+	bnge_hwrm_resource_free(bn, 0);
+	return rc;
+}
+
+static int bnge_napi_poll(struct napi_struct *napi, int budget)
+{
+	int work_done = 0;
+
+	/* defer NAPI implementation to next patch series */
+	napi_complete_done(napi, work_done);
+
+	return work_done;
+}
+
+static void bnge_init_napi(struct bnge_net *bn)
+{
+	struct bnge_dev *bd = bn->bd;
+	struct bnge_napi *bnapi;
+	int i;
+
+	for (i = 0; i < bd->nq_nr_rings; i++) {
+		bnapi = bn->bnapi[i];
+		netif_napi_add_config_locked(bn->netdev, &bnapi->napi,
+					     bnge_napi_poll, bnapi->index);
+	}
+}
+
+static void bnge_del_napi(struct bnge_net *bn)
+{
+	struct bnge_dev *bd = bn->bd;
+	int i;
+
+	for (i = 0; i < bd->rx_nr_rings; i++)
+		netif_queue_set_napi(bn->netdev, i, NETDEV_QUEUE_TYPE_RX, NULL);
+	for (i = 0; i < bd->tx_nr_rings; i++)
+		netif_queue_set_napi(bn->netdev, i, NETDEV_QUEUE_TYPE_TX, NULL);
+
+	for (i = 0; i < bd->nq_nr_rings; i++) {
+		struct bnge_napi *bnapi = bn->bnapi[i];
+
+		__netif_napi_del_locked(&bnapi->napi);
+	}
+
+	/* Wait for RCU grace period after removing NAPI instances */
+	synchronize_net();
+}
+
+static int bnge_init_nic(struct bnge_net *bn)
+{
+	int rc;
+
+	bnge_init_nq_tree(bn);
+
+	bnge_init_rx_rings(bn);
+	rc = bnge_alloc_rx_ring_pair_bufs(bn);
+	if (rc)
+		return rc;
+
+	bnge_init_tx_rings(bn);
+
+	rc = bnge_init_ring_grps(bn);
+	if (rc)
+		goto err_free_rx_ring_pair_bufs;
+
+	bnge_init_vnics(bn);
+
+	rc = bnge_init_chip(bn);
+	if (rc)
+		goto err_free_ring_grps;
+	return rc;
+
+err_free_ring_grps:
+	bnge_free_ring_grps(bn);
+	return rc;
+
+err_free_rx_ring_pair_bufs:
+	bnge_free_rx_ring_pair_bufs(bn);
+	return rc;
+}
+
+static int bnge_open_core(struct bnge_net *bn)
+{
+	struct bnge_dev *bd = bn->bd;
+	int rc;
+
+	netif_carrier_off(bn->netdev);
+
+	rc = bnge_reserve_rings(bd);
+	if (rc) {
+		netdev_err(bn->netdev, "bnge_reserve_rings err: %d\n", rc);
+		return rc;
+	}
+
+	rc = bnge_alloc_core(bn);
+	if (rc) {
+		netdev_err(bn->netdev, "bnge_alloc_core err: %d\n", rc);
+		return rc;
+	}
+
+	bnge_init_napi(bn);
+	rc = bnge_request_irq(bn);
+	if (rc) {
+		netdev_err(bn->netdev, "bnge_request_irq err: %d\n", rc);
+		goto err_del_napi;
+	}
+
+	rc = bnge_init_nic(bn);
+	if (rc) {
+		netdev_err(bn->netdev, "bnge_init_nic err: %d\n", rc);
+		goto err_free_irq;
+	}
+	set_bit(BNGE_STATE_OPEN, &bd->state);
+	return 0;
+
+err_free_irq:
+	bnge_free_irq(bn);
+err_del_napi:
+	bnge_del_napi(bn);
+	bnge_free_core(bn);
+	return rc;
+}
+
+static netdev_tx_t bnge_start_xmit(struct sk_buff *skb, struct net_device *dev)
+{
+	dev_kfree_skb_any(skb);
+
+	return NETDEV_TX_OK;
+}
+
+static int bnge_open(struct net_device *dev)
+{
+	struct bnge_net *bn = netdev_priv(dev);
+	int rc;
+
+	rc = bnge_open_core(bn);
+	if (rc)
+		netdev_err(dev, "bnge_open_core err: %d\n", rc);
+
+	return rc;
+}
+
+static int bnge_shutdown_nic(struct bnge_net *bn)
+{
+	/* TODO: close_path = 0 until we make NAPI functional */
+	bnge_hwrm_resource_free(bn, 0);
+	return 0;
+}
+
+static void bnge_close_core(struct bnge_net *bn)
+{
+	struct bnge_dev *bd = bn->bd;
+
+	clear_bit(BNGE_STATE_OPEN, &bd->state);
+	bnge_shutdown_nic(bn);
+	bnge_free_all_rings_bufs(bn);
+	bnge_free_irq(bn);
+	bnge_del_napi(bn);
+
+	bnge_free_core(bn);
+}
+
+static int bnge_close(struct net_device *dev)
+{
+	struct bnge_net *bn = netdev_priv(dev);
+
+	bnge_close_core(bn);
+
+	return 0;
+}
+
+static const struct net_device_ops bnge_netdev_ops = {
+	.ndo_open		= bnge_open,
+	.ndo_stop		= bnge_close,
+	.ndo_start_xmit		= bnge_start_xmit,
+};
+
+static void bnge_init_mac_addr(struct bnge_dev *bd)
+{
+	eth_hw_addr_set(bd->netdev, bd->pf.mac_addr);
+}
+
+static void bnge_set_tpa_flags(struct bnge_dev *bd)
+{
+	struct bnge_net *bn = netdev_priv(bd->netdev);
+
+	bn->priv_flags &= ~BNGE_NET_EN_TPA;
+
+	if (bd->netdev->features & NETIF_F_LRO)
+		bn->priv_flags |= BNGE_NET_EN_LRO;
+	else if (bd->netdev->features & NETIF_F_GRO_HW)
+		bn->priv_flags |= BNGE_NET_EN_GRO;
+}
+
+static void bnge_init_l2_fltr_tbl(struct bnge_net *bn)
+{
+	int i;
+
+	for (i = 0; i < BNGE_L2_FLTR_HASH_SIZE; i++)
+		INIT_HLIST_HEAD(&bn->l2_fltr_hash_tbl[i]);
+	get_random_bytes(&bn->hash_seed, sizeof(bn->hash_seed));
+}
+
+void bnge_set_ring_params(struct bnge_dev *bd)
+{
+	struct bnge_net *bn = netdev_priv(bd->netdev);
+	u32 ring_size, rx_size, rx_space, max_rx_cmpl;
+	u32 agg_factor = 0, agg_ring_size = 0;
+
+	/* 8 for CRC and VLAN */
+	rx_size = SKB_DATA_ALIGN(bn->netdev->mtu + ETH_HLEN + NET_IP_ALIGN + 8);
+
+	rx_space = rx_size + ALIGN(NET_SKB_PAD, 8) +
+		SKB_DATA_ALIGN(sizeof(struct skb_shared_info));
+
+	bn->rx_copy_thresh = BNGE_RX_COPY_THRESH;
+	ring_size = bn->rx_ring_size;
+	bn->rx_agg_ring_size = 0;
+	bn->rx_agg_nr_pages = 0;
+
+	if (bn->priv_flags & BNGE_NET_EN_TPA)
+		agg_factor = min_t(u32, 4, 65536 / BNGE_RX_PAGE_SIZE);
+
+	bn->priv_flags &= ~BNGE_NET_EN_JUMBO;
+	if (rx_space > PAGE_SIZE) {
+		u32 jumbo_factor;
+
+		bn->priv_flags |= BNGE_NET_EN_JUMBO;
+		jumbo_factor = PAGE_ALIGN(bn->netdev->mtu - 40) >> PAGE_SHIFT;
+		if (jumbo_factor > agg_factor)
+			agg_factor = jumbo_factor;
+	}
+	if (agg_factor) {
+		if (ring_size > BNGE_MAX_RX_DESC_CNT_JUM_ENA) {
+			ring_size = BNGE_MAX_RX_DESC_CNT_JUM_ENA;
+			netdev_warn(bn->netdev, "RX ring size reduced from %d to %d due to jumbo ring\n",
+				    bn->rx_ring_size, ring_size);
+			bn->rx_ring_size = ring_size;
+		}
+		agg_ring_size = ring_size * agg_factor;
+
+		bn->rx_agg_nr_pages = bnge_adjust_pow_two(agg_ring_size,
+							  RX_DESC_CNT);
+		if (bn->rx_agg_nr_pages > MAX_RX_AGG_PAGES) {
+			u32 tmp = agg_ring_size;
+
+			bn->rx_agg_nr_pages = MAX_RX_AGG_PAGES;
+			agg_ring_size = MAX_RX_AGG_PAGES * RX_DESC_CNT - 1;
+			netdev_warn(bn->netdev, "RX agg ring size %d reduced to %d.\n",
+				    tmp, agg_ring_size);
+		}
+		bn->rx_agg_ring_size = agg_ring_size;
+		bn->rx_agg_ring_mask = (bn->rx_agg_nr_pages * RX_DESC_CNT) - 1;
+
+		rx_size = SKB_DATA_ALIGN(BNGE_RX_COPY_THRESH + NET_IP_ALIGN);
+		rx_space = rx_size + NET_SKB_PAD +
+			SKB_DATA_ALIGN(sizeof(struct skb_shared_info));
+	}
+
+	bn->rx_buf_use_size = rx_size;
+	bn->rx_buf_size = rx_space;
+
+	bn->rx_nr_pages = bnge_adjust_pow_two(ring_size, RX_DESC_CNT);
+	bn->rx_ring_mask = (bn->rx_nr_pages * RX_DESC_CNT) - 1;
+
+	ring_size = bn->tx_ring_size;
+	bn->tx_nr_pages = bnge_adjust_pow_two(ring_size, TX_DESC_CNT);
+	bn->tx_ring_mask = (bn->tx_nr_pages * TX_DESC_CNT) - 1;
+
+	max_rx_cmpl = bn->rx_ring_size;
+
+	if (bn->priv_flags & BNGE_NET_EN_TPA)
+		max_rx_cmpl += bd->max_tpa_v2;
+	ring_size = max_rx_cmpl * 2 + agg_ring_size + bn->tx_ring_size;
+	bn->cp_ring_size = ring_size;
+
+	bn->cp_nr_pages = bnge_adjust_pow_two(ring_size, CP_DESC_CNT);
+	if (bn->cp_nr_pages > MAX_CP_PAGES) {
+		bn->cp_nr_pages = MAX_CP_PAGES;
+		bn->cp_ring_size = MAX_CP_PAGES * CP_DESC_CNT - 1;
+		netdev_warn(bn->netdev, "completion ring size %d reduced to %d.\n",
+			    ring_size, bn->cp_ring_size);
+	}
+	bn->cp_bit = bn->cp_nr_pages * CP_DESC_CNT;
+	bn->cp_ring_mask = bn->cp_bit - 1;
+}
+
+int bnge_netdev_alloc(struct bnge_dev *bd, int max_irqs)
+{
+	struct net_device *netdev;
+	struct bnge_net *bn;
+	int rc;
+
+	netdev = alloc_etherdev_mqs(sizeof(*bn), max_irqs * BNGE_MAX_QUEUE,
+				    max_irqs);
+	if (!netdev)
+		return -ENOMEM;
+
+	SET_NETDEV_DEV(netdev, bd->dev);
+	bd->netdev = netdev;
+
+	netdev->netdev_ops = &bnge_netdev_ops;
+
+	bnge_set_ethtool_ops(netdev);
+
+	bn = netdev_priv(netdev);
+	bn->netdev = netdev;
+	bn->bd = bd;
+
+	netdev->min_mtu = ETH_ZLEN;
+	netdev->max_mtu = bd->max_mtu;
+
+	netdev->hw_features = NETIF_F_IP_CSUM |
+			      NETIF_F_IPV6_CSUM |
+			      NETIF_F_SG |
+			      NETIF_F_TSO |
+			      NETIF_F_TSO6 |
+			      NETIF_F_GSO_UDP_TUNNEL |
+			      NETIF_F_GSO_GRE |
+			      NETIF_F_GSO_IPXIP4 |
+			      NETIF_F_GSO_UDP_TUNNEL_CSUM |
+			      NETIF_F_GSO_GRE_CSUM |
+			      NETIF_F_GSO_PARTIAL |
+			      NETIF_F_RXHASH |
+			      NETIF_F_RXCSUM |
+			      NETIF_F_GRO;
+
+	if (bd->flags & BNGE_EN_UDP_GSO_SUPP)
+		netdev->hw_features |= NETIF_F_GSO_UDP_L4;
+
+	if (BNGE_SUPPORTS_TPA(bd))
+		netdev->hw_features |= NETIF_F_LRO;
+
+	netdev->hw_enc_features = NETIF_F_IP_CSUM |
+				  NETIF_F_IPV6_CSUM |
+				  NETIF_F_SG |
+				  NETIF_F_TSO |
+				  NETIF_F_TSO6 |
+				  NETIF_F_GSO_UDP_TUNNEL |
+				  NETIF_F_GSO_GRE |
+				  NETIF_F_GSO_UDP_TUNNEL_CSUM |
+				  NETIF_F_GSO_GRE_CSUM |
+				  NETIF_F_GSO_IPXIP4 |
+				  NETIF_F_GSO_PARTIAL;
+
+	if (bd->flags & BNGE_EN_UDP_GSO_SUPP)
+		netdev->hw_enc_features |= NETIF_F_GSO_UDP_L4;
+
+	netdev->gso_partial_features = NETIF_F_GSO_UDP_TUNNEL_CSUM |
+				       NETIF_F_GSO_GRE_CSUM;
+
+	netdev->vlan_features = netdev->hw_features | NETIF_F_HIGHDMA;
+	if (bd->fw_cap & BNGE_FW_CAP_VLAN_RX_STRIP)
+		netdev->hw_features |= BNGE_HW_FEATURE_VLAN_ALL_RX;
+	if (bd->fw_cap & BNGE_FW_CAP_VLAN_TX_INSERT)
+		netdev->hw_features |= BNGE_HW_FEATURE_VLAN_ALL_TX;
+
+	if (BNGE_SUPPORTS_TPA(bd))
+		netdev->hw_features |= NETIF_F_GRO_HW;
+
+	netdev->features |= netdev->hw_features | NETIF_F_HIGHDMA;
+
+	if (netdev->features & NETIF_F_GRO_HW)
+		netdev->features &= ~NETIF_F_LRO;
+
+	netdev->priv_flags |= IFF_UNICAST_FLT;
+
+	netif_set_tso_max_size(netdev, GSO_MAX_SIZE);
+	if (bd->tso_max_segs)
+		netif_set_tso_max_segs(netdev, bd->tso_max_segs);
+
+	bn->rx_ring_size = BNGE_DEFAULT_RX_RING_SIZE;
+	bn->tx_ring_size = BNGE_DEFAULT_TX_RING_SIZE;
+	bn->rx_dir = DMA_FROM_DEVICE;
+
+	bnge_set_tpa_flags(bd);
+	bnge_set_ring_params(bd);
+
+	bnge_init_l2_fltr_tbl(bn);
+	bnge_init_mac_addr(bd);
+
+	netdev->request_ops_lock = true;
+	rc = register_netdev(netdev);
+	if (rc) {
+		dev_err(bd->dev, "Register netdev failed rc: %d\n", rc);
+		goto err_netdev;
+	}
+
+	return 0;
+
+err_netdev:
+	free_netdev(netdev);
+	return rc;
+}
+
+void bnge_netdev_free(struct bnge_dev *bd)
+{
+	struct net_device *netdev = bd->netdev;
+
+	unregister_netdev(netdev);
+	free_netdev(netdev);
+	bd->netdev = NULL;
+}
diff --git a/drivers/net/ethernet/broadcom/bnge/bnge_netdev.h b/drivers/net/ethernet/broadcom/bnge/bnge_netdev.h
new file mode 100644
index 000000000000..fb3b961536ba
--- /dev/null
+++ b/drivers/net/ethernet/broadcom/bnge/bnge_netdev.h
@@ -0,0 +1,454 @@
+/* SPDX-License-Identifier: GPL-2.0 */
+/* Copyright (c) 2025 Broadcom */
+
+#ifndef _BNGE_NETDEV_H_
+#define _BNGE_NETDEV_H_
+
+#include <linux/bnxt/hsi.h>
+#include <linux/io-64-nonatomic-lo-hi.h>
+#include <linux/refcount.h>
+#include "bnge_db.h"
+
+struct tx_bd {
+	__le32 tx_bd_len_flags_type;
+	#define TX_BD_TYPE					(0x3f << 0)
+	#define TX_BD_TYPE_SHORT_TX_BD				(0x00 << 0)
+	#define TX_BD_TYPE_LONG_TX_BD				(0x10 << 0)
+	#define TX_BD_FLAGS_PACKET_END				(1 << 6)
+	#define TX_BD_FLAGS_NO_CMPL				(1 << 7)
+	#define TX_BD_FLAGS_BD_CNT				(0x1f << 8)
+	#define TX_BD_FLAGS_BD_CNT_SHIFT			8
+	#define TX_BD_FLAGS_LHINT				(3 << 13)
+	#define TX_BD_FLAGS_LHINT_SHIFT				13
+	#define TX_BD_FLAGS_LHINT_512_AND_SMALLER		(0 << 13)
+	#define TX_BD_FLAGS_LHINT_512_TO_1023			(1 << 13)
+	#define TX_BD_FLAGS_LHINT_1024_TO_2047			(2 << 13)
+	#define TX_BD_FLAGS_LHINT_2048_AND_LARGER		(3 << 13)
+	#define TX_BD_FLAGS_COAL_NOW				(1 << 15)
+	#define TX_BD_LEN					(0xffff << 16)
+	#define TX_BD_LEN_SHIFT					16
+	u32 tx_bd_opaque;
+	__le64 tx_bd_haddr;
+} __packed;
+
+struct rx_bd {
+	__le32 rx_bd_len_flags_type;
+	#define RX_BD_TYPE					(0x3f << 0)
+	#define RX_BD_TYPE_RX_PACKET_BD				0x4
+	#define RX_BD_TYPE_RX_BUFFER_BD				0x5
+	#define RX_BD_TYPE_RX_AGG_BD				0x6
+	#define RX_BD_TYPE_16B_BD_SIZE				(0 << 4)
+	#define RX_BD_TYPE_32B_BD_SIZE				(1 << 4)
+	#define RX_BD_TYPE_48B_BD_SIZE				(2 << 4)
+	#define RX_BD_TYPE_64B_BD_SIZE				(3 << 4)
+	#define RX_BD_FLAGS_SOP					(1 << 6)
+	#define RX_BD_FLAGS_EOP					(1 << 7)
+	#define RX_BD_FLAGS_BUFFERS				(3 << 8)
+	#define RX_BD_FLAGS_1_BUFFER_PACKET			(0 << 8)
+	#define RX_BD_FLAGS_2_BUFFER_PACKET			(1 << 8)
+	#define RX_BD_FLAGS_3_BUFFER_PACKET			(2 << 8)
+	#define RX_BD_FLAGS_4_BUFFER_PACKET			(3 << 8)
+	#define RX_BD_LEN					(0xffff << 16)
+	#define RX_BD_LEN_SHIFT					16
+	u32 rx_bd_opaque;
+	__le64 rx_bd_haddr;
+};
+
+struct tx_cmp {
+	__le32 tx_cmp_flags_type;
+	#define CMP_TYPE					(0x3f << 0)
+	#define CMP_TYPE_TX_L2_CMP				0
+	#define CMP_TYPE_TX_L2_COAL_CMP				2
+	#define CMP_TYPE_TX_L2_PKT_TS_CMP			4
+	#define CMP_TYPE_RX_L2_CMP				17
+	#define CMP_TYPE_RX_AGG_CMP				18
+	#define CMP_TYPE_RX_L2_TPA_START_CMP			19
+	#define CMP_TYPE_RX_L2_TPA_END_CMP			21
+	#define CMP_TYPE_RX_TPA_AGG_CMP				22
+	#define CMP_TYPE_RX_L2_V3_CMP				23
+	#define CMP_TYPE_RX_L2_TPA_START_V3_CMP			25
+	#define CMP_TYPE_STATUS_CMP				32
+	#define CMP_TYPE_REMOTE_DRIVER_REQ			34
+	#define CMP_TYPE_REMOTE_DRIVER_RESP			36
+	#define CMP_TYPE_ERROR_STATUS				48
+	#define CMPL_BASE_TYPE_STAT_EJECT			0x1aUL
+	#define CMPL_BASE_TYPE_HWRM_DONE			0x20UL
+	#define CMPL_BASE_TYPE_HWRM_FWD_REQ			0x22UL
+	#define CMPL_BASE_TYPE_HWRM_FWD_RESP			0x24UL
+	#define CMPL_BASE_TYPE_HWRM_ASYNC_EVENT			0x2eUL
+	#define TX_CMP_FLAGS_ERROR				(1 << 6)
+	#define TX_CMP_FLAGS_PUSH				(1 << 7)
+	u32 tx_cmp_opaque;
+	__le32 tx_cmp_errors_v;
+	#define TX_CMP_V					(1 << 0)
+	#define TX_CMP_ERRORS_BUFFER_ERROR			(7 << 1)
+	#define TX_CMP_ERRORS_BUFFER_ERROR_NO_ERROR		0
+	#define TX_CMP_ERRORS_BUFFER_ERROR_BAD_FORMAT		2
+	#define TX_CMP_ERRORS_BUFFER_ERROR_INVALID_STAG		4
+	#define TX_CMP_ERRORS_BUFFER_ERROR_STAG_BOUNDS		5
+	#define TX_CMP_ERRORS_ZERO_LENGTH_PKT			(1 << 4)
+	#define TX_CMP_ERRORS_EXCESSIVE_BD_LEN			(1 << 5)
+	#define TX_CMP_ERRORS_DMA_ERROR				(1 << 6)
+	#define TX_CMP_ERRORS_HINT_TOO_SHORT			(1 << 7)
+	__le32 sq_cons_idx;
+	#define TX_CMP_SQ_CONS_IDX_MASK				0x00ffffff
+};
+
+struct bnge_sw_tx_bd {
+	struct sk_buff		*skb;
+	DEFINE_DMA_UNMAP_ADDR(mapping);
+	DEFINE_DMA_UNMAP_LEN(len);
+	struct page		*page;
+	u8			is_ts_pkt;
+	u8			is_push;
+	u8			action;
+	unsigned short		nr_frags;
+	union {
+		u16		rx_prod;
+		u16		txts_prod;
+	};
+};
+
+struct bnge_sw_rx_bd {
+	void			*data;
+	u8			*data_ptr;
+	dma_addr_t		mapping;
+};
+
+struct bnge_sw_rx_agg_bd {
+	netmem_ref		netmem;
+	unsigned int		offset;
+	dma_addr_t		mapping;
+};
+
+#define HWRM_RING_ALLOC_TX	0x1
+#define HWRM_RING_ALLOC_RX	0x2
+#define HWRM_RING_ALLOC_AGG	0x4
+#define HWRM_RING_ALLOC_CMPL	0x8
+#define HWRM_RING_ALLOC_NQ	0x10
+
+struct bnge_ring_grp_info {
+	u16	fw_stats_ctx;
+	u16	fw_grp_id;
+	u16	rx_fw_ring_id;
+	u16	agg_fw_ring_id;
+	u16	nq_fw_ring_id;
+};
+
+#define BNGE_RX_COPY_THRESH     256
+
+#define BNGE_HW_FEATURE_VLAN_ALL_RX	\
+		(NETIF_F_HW_VLAN_CTAG_RX | NETIF_F_HW_VLAN_STAG_RX)
+#define BNGE_HW_FEATURE_VLAN_ALL_TX	\
+		(NETIF_F_HW_VLAN_CTAG_TX | NETIF_F_HW_VLAN_STAG_TX)
+
+enum {
+	BNGE_NET_EN_GRO		= BIT(0),
+	BNGE_NET_EN_LRO		= BIT(1),
+	BNGE_NET_EN_JUMBO	= BIT(2),
+};
+
+#define BNGE_NET_EN_TPA		(BNGE_NET_EN_GRO | BNGE_NET_EN_LRO)
+
+/* Minimum TX BDs for a TX packet with MAX_SKB_FRAGS + 1. We need one extra
+ * BD because the first TX BD is always a long BD.
+ */
+#define BNGE_MIN_TX_DESC_CNT	(MAX_SKB_FRAGS + 2)
+
+#define RX_RING(bn, x)	(((x) & (bn)->rx_ring_mask) >> (BNGE_PAGE_SHIFT - 4))
+#define RX_AGG_RING(bn, x)	(((x) & (bn)->rx_agg_ring_mask) >>	\
+				 (BNGE_PAGE_SHIFT - 4))
+#define RX_IDX(x)	((x) & (RX_DESC_CNT - 1))
+
+#define TX_RING(bn, x)	(((x) & (bn)->tx_ring_mask) >> (BNGE_PAGE_SHIFT - 4))
+#define TX_IDX(x)	((x) & (TX_DESC_CNT - 1))
+
+#define CP_RING(x)	(((x) & ~(CP_DESC_CNT - 1)) >> (BNGE_PAGE_SHIFT - 4))
+#define CP_IDX(x)	((x) & (CP_DESC_CNT - 1))
+
+#define RING_RX(bn, idx)	((idx) & (bn)->rx_ring_mask)
+#define NEXT_RX(idx)		((idx) + 1)
+
+#define RING_RX_AGG(bn, idx)	((idx) & (bn)->rx_agg_ring_mask)
+#define NEXT_RX_AGG(idx)	((idx) + 1)
+
+#define BNGE_NQ_HDL_TYPE_SHIFT	24
+#define BNGE_NQ_HDL_TYPE_RX	0x00
+#define BNGE_NQ_HDL_TYPE_TX	0x01
+
+struct bnge_net {
+	struct bnge_dev		*bd;
+	struct net_device	*netdev;
+
+	u32			priv_flags;
+
+	u32			rx_ring_size;
+	u32			rx_buf_size;
+	u32			rx_buf_use_size; /* usable size */
+	u32			rx_agg_ring_size;
+	u32			rx_copy_thresh;
+	u32			rx_ring_mask;
+	u32			rx_agg_ring_mask;
+	u16			rx_nr_pages;
+	u16			rx_agg_nr_pages;
+
+	u32			tx_ring_size;
+	u32			tx_ring_mask;
+	u16			tx_nr_pages;
+
+	/* NQs and Completion rings */
+	u32			cp_ring_size;
+	u32			cp_ring_mask;
+	u32			cp_bit;
+	u16			cp_nr_pages;
+
+#define BNGE_L2_FLTR_HASH_SIZE	32
+#define BNGE_L2_FLTR_HASH_MASK	(BNGE_L2_FLTR_HASH_SIZE - 1)
+	struct hlist_head	l2_fltr_hash_tbl[BNGE_L2_FLTR_HASH_SIZE];
+	u32			hash_seed;
+	u64			toeplitz_prefix;
+
+	struct bnge_napi		**bnapi;
+
+	struct bnge_rx_ring_info	*rx_ring;
+	struct bnge_tx_ring_info	*tx_ring;
+
+	u16				*tx_ring_map;
+	enum dma_data_direction		rx_dir;
+
+	/* grp_info indexed by napi/nq index */
+	struct bnge_ring_grp_info	*grp_info;
+	struct bnge_vnic_info		*vnic_info;
+	int				nr_vnics;
+	int				total_irqs;
+
+	u32			tx_wake_thresh;
+	u16			rx_offset;
+	u16			rx_dma_offset;
+
+	u8			rss_hash_key[HW_HASH_KEY_SIZE];
+	u8			rss_hash_key_valid:1;
+	u8			rss_hash_key_updated:1;
+	int			rsscos_nr_ctxs;
+	u32			stats_coal_ticks;
+};
+
+#define BNGE_DEFAULT_RX_RING_SIZE	511
+#define BNGE_DEFAULT_TX_RING_SIZE	511
+
+int bnge_netdev_alloc(struct bnge_dev *bd, int max_irqs);
+void bnge_netdev_free(struct bnge_dev *bd);
+void bnge_set_ring_params(struct bnge_dev *bd);
+
+#if (BNGE_PAGE_SHIFT == 16)
+#define MAX_RX_PAGES_AGG_ENA	1
+#define MAX_RX_PAGES		4
+#define MAX_RX_AGG_PAGES	4
+#define MAX_TX_PAGES		1
+#define MAX_CP_PAGES		16
+#else
+#define MAX_RX_PAGES_AGG_ENA	8
+#define MAX_RX_PAGES		32
+#define MAX_RX_AGG_PAGES	32
+#define MAX_TX_PAGES		8
+#define MAX_CP_PAGES		128
+#endif
+
+#define BNGE_RX_PAGE_SIZE		(1 << BNGE_RX_PAGE_SHIFT)
+
+#define RX_DESC_CNT			(BNGE_PAGE_SIZE / sizeof(struct rx_bd))
+#define TX_DESC_CNT			(BNGE_PAGE_SIZE / sizeof(struct tx_bd))
+#define CP_DESC_CNT			(BNGE_PAGE_SIZE / sizeof(struct tx_cmp))
+#define SW_RXBD_RING_SIZE		(sizeof(struct bnge_sw_rx_bd) * RX_DESC_CNT)
+#define HW_RXBD_RING_SIZE		(sizeof(struct rx_bd) * RX_DESC_CNT)
+#define SW_RXBD_AGG_RING_SIZE		(sizeof(struct bnge_sw_rx_agg_bd) * RX_DESC_CNT)
+#define SW_TXBD_RING_SIZE		(sizeof(struct bnge_sw_tx_bd) * TX_DESC_CNT)
+#define HW_TXBD_RING_SIZE		(sizeof(struct tx_bd) * TX_DESC_CNT)
+#define HW_CMPD_RING_SIZE		(sizeof(struct tx_cmp) * CP_DESC_CNT)
+#define BNGE_MAX_RX_DESC_CNT		(RX_DESC_CNT * MAX_RX_PAGES - 1)
+#define BNGE_MAX_RX_DESC_CNT_JUM_ENA	(RX_DESC_CNT * MAX_RX_PAGES_AGG_ENA - 1)
+#define BNGE_MAX_RX_JUM_DESC_CNT	(RX_DESC_CNT * MAX_RX_AGG_PAGES - 1)
+#define BNGE_MAX_TX_DESC_CNT		(TX_DESC_CNT * MAX_TX_PAGES - 1)
+
+#define BNGE_MAX_TXR_PER_NAPI	8
+
+#define bnge_for_each_napi_tx(iter, bnapi, txr)		\
+	for (iter = 0, txr = (bnapi)->tx_ring[0]; txr;	\
+	     txr = (iter < BNGE_MAX_TXR_PER_NAPI - 1) ?	\
+	     (bnapi)->tx_ring[++iter] : NULL)
+
+#define BNGE_SET_NQ_HDL(cpr)						\
+	(((cpr)->cp_ring_type << BNGE_NQ_HDL_TYPE_SHIFT) | (cpr)->cp_idx)
+
+struct bnge_stats_mem {
+	u64		*sw_stats;
+	u64		*hw_masks;
+	void		*hw_stats;
+	dma_addr_t	hw_stats_map;
+	int		len;
+};
+
+struct bnge_cp_ring_info {
+	struct bnge_napi	*bnapi;
+	dma_addr_t		*desc_mapping;
+	struct tx_cmp		**desc_ring;
+	struct bnge_ring_struct	ring_struct;
+	u8			cp_ring_type;
+	u8			cp_idx;
+	u32			cp_raw_cons;
+	struct bnge_db_info	cp_db;
+};
+
+struct bnge_nq_ring_info {
+	struct bnge_napi	*bnapi;
+	dma_addr_t		*desc_mapping;
+	struct nqe_cn		**desc_ring;
+	struct bnge_ring_struct	ring_struct;
+	u32			nq_raw_cons;
+	struct bnge_db_info	nq_db;
+
+	struct bnge_stats_mem	stats;
+	u32			hw_stats_ctx_id;
+
+	int				cp_ring_count;
+	struct bnge_cp_ring_info	*cp_ring_arr;
+};
+
+struct bnge_rx_ring_info {
+	struct bnge_napi	*bnapi;
+	struct bnge_cp_ring_info	*rx_cpr;
+	u16			rx_prod;
+	u16			rx_agg_prod;
+	u16			rx_sw_agg_prod;
+	u16			rx_next_cons;
+	struct bnge_db_info	rx_db;
+	struct bnge_db_info	rx_agg_db;
+
+	struct rx_bd		*rx_desc_ring[MAX_RX_PAGES];
+	struct bnge_sw_rx_bd	*rx_buf_ring;
+
+	struct rx_bd			*rx_agg_desc_ring[MAX_RX_AGG_PAGES];
+	struct bnge_sw_rx_agg_bd	*rx_agg_buf_ring;
+
+	unsigned long		*rx_agg_bmap;
+	u16			rx_agg_bmap_size;
+
+	dma_addr_t		rx_desc_mapping[MAX_RX_PAGES];
+	dma_addr_t		rx_agg_desc_mapping[MAX_RX_AGG_PAGES];
+
+	struct bnge_ring_struct	rx_ring_struct;
+	struct bnge_ring_struct	rx_agg_ring_struct;
+	struct page_pool	*page_pool;
+	struct page_pool	*head_pool;
+	bool			need_head_pool;
+};
+
+struct bnge_tx_ring_info {
+	struct bnge_napi	*bnapi;
+	struct bnge_cp_ring_info	*tx_cpr;
+	u16			tx_prod;
+	u16			tx_cons;
+	u16			tx_hw_cons;
+	u16			txq_index;
+	u8			tx_napi_idx;
+	u8			kick_pending;
+	struct bnge_db_info	tx_db;
+
+	struct tx_bd		*tx_desc_ring[MAX_TX_PAGES];
+	struct bnge_sw_tx_bd	*tx_buf_ring;
+
+	dma_addr_t		tx_desc_mapping[MAX_TX_PAGES];
+
+	u32			dev_state;
+#define BNGE_DEV_STATE_CLOSING	0x1
+
+	struct bnge_ring_struct	tx_ring_struct;
+};
+
+struct bnge_napi {
+	struct napi_struct		napi;
+	struct bnge_net			*bn;
+	int				index;
+
+	struct bnge_nq_ring_info	nq_ring;
+	struct bnge_rx_ring_info	*rx_ring;
+	struct bnge_tx_ring_info	*tx_ring[BNGE_MAX_TXR_PER_NAPI];
+};
+
+#define INVALID_STATS_CTX_ID	-1
+#define BNGE_VNIC_DEFAULT	0
+#define BNGE_MAX_UC_ADDRS	4
+
+struct bnge_vnic_info {
+	u16		fw_vnic_id;
+#define BNGE_MAX_CTX_PER_VNIC	8
+	u16		fw_rss_cos_lb_ctx[BNGE_MAX_CTX_PER_VNIC];
+	u16		mru;
+	/* index 0 always dev_addr */
+	struct bnge_l2_filter *l2_filters[BNGE_MAX_UC_ADDRS];
+	u16		uc_filter_count;
+	u8		*uc_list;
+	dma_addr_t	rss_table_dma_addr;
+	__le16		*rss_table;
+	dma_addr_t	rss_hash_key_dma_addr;
+	u64		*rss_hash_key;
+	int		rss_table_size;
+#define BNGE_RSS_TABLE_ENTRIES		64
+#define BNGE_RSS_TABLE_SIZE		(BNGE_RSS_TABLE_ENTRIES * 4)
+#define BNGE_RSS_TABLE_MAX_TBL		8
+#define BNGE_MAX_RSS_TABLE_SIZE			\
+	(BNGE_RSS_TABLE_SIZE * BNGE_RSS_TABLE_MAX_TBL)
+	u32		rx_mask;
+
+	u8		*mc_list;
+	int		mc_list_size;
+	int		mc_list_count;
+	dma_addr_t	mc_list_mapping;
+#define BNGE_MAX_MC_ADDRS	16
+
+	u32		flags;
+#define BNGE_VNIC_RSS_FLAG	1
+#define BNGE_VNIC_MCAST_FLAG	4
+#define BNGE_VNIC_UCAST_FLAG	8
+	u32		vnic_id;
+};
+
+struct bnge_filter_base {
+	struct hlist_node	hash;
+	struct list_head	list;
+	__le64			filter_id;
+	u8			type;
+#define BNGE_FLTR_TYPE_L2	2
+	u8			flags;
+	u16			rxq;
+	u16			fw_vnic_id;
+	u16			vf_idx;
+	unsigned long		state;
+#define BNGE_FLTR_VALID		0
+#define BNGE_FLTR_FW_DELETED	2
+
+	struct rcu_head         rcu;
+};
+
+struct bnge_l2_key {
+	union {
+		struct {
+			u8	dst_mac_addr[ETH_ALEN];
+			u16	vlan;
+		};
+		u32	filter_key;
+	};
+};
+
+#define BNGE_L2_KEY_SIZE	(sizeof(struct bnge_l2_key) / 4)
+struct bnge_l2_filter {
+	/* base filter must be the first member */
+	struct bnge_filter_base	base;
+	struct bnge_l2_key	l2_key;
+	refcount_t		refcnt;
+};
+
+u16 bnge_cp_ring_for_rx(struct bnge_rx_ring_info *rxr);
+u16 bnge_cp_ring_for_tx(struct bnge_tx_ring_info *txr);
+void bnge_fill_hw_rss_tbl(struct bnge_net *bn, struct bnge_vnic_info *vnic);
+#endif /* _BNGE_NETDEV_H_ */
diff --git a/drivers/net/ethernet/broadcom/bnge/bnge_resc.c b/drivers/net/ethernet/broadcom/bnge/bnge_resc.c
new file mode 100644
index 000000000000..62ebe03a0dcf
--- /dev/null
+++ b/drivers/net/ethernet/broadcom/bnge/bnge_resc.c
@@ -0,0 +1,605 @@
+// SPDX-License-Identifier: GPL-2.0
+// Copyright (c) 2025 Broadcom.
+
+#include <linux/init.h>
+#include <linux/module.h>
+#include <linux/pci.h>
+#include <linux/ethtool.h>
+#include <linux/netdevice.h>
+
+#include "bnge.h"
+#include "bnge_hwrm.h"
+#include "bnge_hwrm_lib.h"
+#include "bnge_resc.h"
+
+static u16 bnge_num_tx_to_cp(struct bnge_dev *bd, u16 tx)
+{
+	u16 tcs = bd->num_tc;
+
+	if (!tcs)
+		tcs = 1;
+
+	return tx / tcs;
+}
+
+static u16 bnge_get_max_func_irqs(struct bnge_dev *bd)
+{
+	struct bnge_hw_resc *hw_resc = &bd->hw_resc;
+
+	return min_t(u16, hw_resc->max_irqs, hw_resc->max_nqs);
+}
+
+static unsigned int bnge_get_max_func_stat_ctxs(struct bnge_dev *bd)
+{
+	return bd->hw_resc.max_stat_ctxs;
+}
+
+static unsigned int bnge_get_max_func_cp_rings(struct bnge_dev *bd)
+{
+	return bd->hw_resc.max_cp_rings;
+}
+
+static int bnge_aux_get_dflt_msix(struct bnge_dev *bd)
+{
+	int roce_msix = BNGE_MAX_ROCE_MSIX;
+
+	return min_t(int, roce_msix, num_online_cpus() + 1);
+}
+
+u16 bnge_aux_get_msix(struct bnge_dev *bd)
+{
+	if (bnge_is_roce_en(bd))
+		return bd->aux_num_msix;
+
+	return 0;
+}
+
+static void bnge_aux_set_msix_num(struct bnge_dev *bd, u16 num)
+{
+	if (bnge_is_roce_en(bd))
+		bd->aux_num_msix = num;
+}
+
+static u16 bnge_aux_get_stat_ctxs(struct bnge_dev *bd)
+{
+	if (bnge_is_roce_en(bd))
+		return bd->aux_num_stat_ctxs;
+
+	return 0;
+}
+
+static void bnge_aux_set_stat_ctxs(struct bnge_dev *bd, u16 num_aux_ctx)
+{
+	if (bnge_is_roce_en(bd))
+		bd->aux_num_stat_ctxs = num_aux_ctx;
+}
+
+static u16 bnge_func_stat_ctxs_demand(struct bnge_dev *bd)
+{
+	return bd->nq_nr_rings + bnge_aux_get_stat_ctxs(bd);
+}
+
+static int bnge_get_dflt_aux_stat_ctxs(struct bnge_dev *bd)
+{
+	int stat_ctx = 0;
+
+	if (bnge_is_roce_en(bd)) {
+		stat_ctx = BNGE_MIN_ROCE_STAT_CTXS;
+
+		if (!bd->pf.port_id && bd->port_count > 1)
+			stat_ctx++;
+	}
+
+	return stat_ctx;
+}
+
+static u16 bnge_nqs_demand(struct bnge_dev *bd)
+{
+	return bd->nq_nr_rings + bnge_aux_get_msix(bd);
+}
+
+static u16 bnge_cprs_demand(struct bnge_dev *bd)
+{
+	return bd->tx_nr_rings + bd->rx_nr_rings;
+}
+
+static u16 bnge_get_avail_msix(struct bnge_dev *bd, int num)
+{
+	u16 max_irq = bnge_get_max_func_irqs(bd);
+	u16 total_demand = bd->nq_nr_rings + num;
+
+	if (max_irq < total_demand) {
+		num = max_irq - bd->nq_nr_rings;
+		if (num <= 0)
+			return 0;
+	}
+
+	return num;
+}
+
+static u16 bnge_num_cp_to_tx(struct bnge_dev *bd, u16 tx_chunks)
+{
+	return tx_chunks * bd->num_tc;
+}
+
+int bnge_fix_rings_count(u16 *rx, u16 *tx, u16 max, bool shared)
+{
+	u16 _rx = *rx, _tx = *tx;
+
+	if (shared) {
+		*rx = min_t(u16, _rx, max);
+		*tx = min_t(u16, _tx, max);
+	} else {
+		if (max < 2)
+			return -ENOMEM;
+		while (_rx + _tx > max) {
+			if (_rx > _tx && _rx > 1)
+				_rx--;
+			else if (_tx > 1)
+				_tx--;
+		}
+		*rx = _rx;
+		*tx = _tx;
+	}
+
+	return 0;
+}
+
+static int bnge_adjust_rings(struct bnge_dev *bd, u16 *rx,
+			     u16 *tx, u16 max_nq, bool sh)
+{
+	u16 tx_chunks = bnge_num_tx_to_cp(bd, *tx);
+
+	if (tx_chunks != *tx) {
+		u16 tx_saved = tx_chunks, rc;
+
+		rc = bnge_fix_rings_count(rx, &tx_chunks, max_nq, sh);
+		if (rc)
+			return rc;
+		if (tx_chunks != tx_saved)
+			*tx = bnge_num_cp_to_tx(bd, tx_chunks);
+		return 0;
+	}
+
+	return bnge_fix_rings_count(rx, tx, max_nq, sh);
+}
+
+int bnge_cal_nr_rss_ctxs(u16 rx_rings)
+{
+	if (!rx_rings)
+		return 0;
+
+	return bnge_adjust_pow_two(rx_rings - 1,
+				   BNGE_RSS_TABLE_ENTRIES);
+}
+
+static u16 bnge_rss_ctxs_in_use(struct bnge_dev *bd,
+				struct bnge_hw_rings *hwr)
+{
+	return bnge_cal_nr_rss_ctxs(hwr->grp);
+}
+
+static u16 bnge_get_total_vnics(struct bnge_dev *bd, u16 rx_rings)
+{
+	return 1;
+}
+
+u32 bnge_get_rxfh_indir_size(struct bnge_dev *bd)
+{
+	return bnge_cal_nr_rss_ctxs(bd->rx_nr_rings) *
+	       BNGE_RSS_TABLE_ENTRIES;
+}
+
+static void bnge_set_dflt_rss_indir_tbl(struct bnge_dev *bd)
+{
+	u16 max_entries, pad;
+	u32 *rss_indir_tbl;
+	int i;
+
+	max_entries = bnge_get_rxfh_indir_size(bd);
+	rss_indir_tbl = &bd->rss_indir_tbl[0];
+
+	for (i = 0; i < max_entries; i++)
+		rss_indir_tbl[i] = ethtool_rxfh_indir_default(i,
+							      bd->rx_nr_rings);
+
+	pad = bd->rss_indir_tbl_entries - max_entries;
+	if (pad)
+		memset(&rss_indir_tbl[i], 0, pad * sizeof(*rss_indir_tbl));
+}
+
+static void bnge_copy_reserved_rings(struct bnge_dev *bd,
+				     struct bnge_hw_rings *hwr)
+{
+	struct bnge_hw_resc *hw_resc = &bd->hw_resc;
+
+	hwr->tx = hw_resc->resv_tx_rings;
+	hwr->rx = hw_resc->resv_rx_rings;
+	hwr->nq = hw_resc->resv_irqs;
+	hwr->cmpl = hw_resc->resv_cp_rings;
+	hwr->grp = hw_resc->resv_hw_ring_grps;
+	hwr->vnic = hw_resc->resv_vnics;
+	hwr->stat = hw_resc->resv_stat_ctxs;
+	hwr->rss_ctx = hw_resc->resv_rsscos_ctxs;
+}
+
+static bool bnge_rings_ok(struct bnge_hw_rings *hwr)
+{
+	return hwr->tx && hwr->rx && hwr->nq && hwr->grp && hwr->vnic &&
+	       hwr->stat && hwr->cmpl;
+}
+
+static bool bnge_need_reserve_rings(struct bnge_dev *bd)
+{
+	struct bnge_hw_resc *hw_resc = &bd->hw_resc;
+	u16 cprs = bnge_cprs_demand(bd);
+	u16 rx = bd->rx_nr_rings, stat;
+	u16 nqs = bnge_nqs_demand(bd);
+	u16 vnic;
+
+	if (hw_resc->resv_tx_rings != bd->tx_nr_rings)
+		return true;
+
+	vnic = bnge_get_total_vnics(bd, rx);
+
+	if (bnge_is_agg_reqd(bd))
+		rx <<= 1;
+	stat = bnge_func_stat_ctxs_demand(bd);
+	if (hw_resc->resv_rx_rings != rx || hw_resc->resv_cp_rings != cprs ||
+	    hw_resc->resv_vnics != vnic || hw_resc->resv_stat_ctxs != stat)
+		return true;
+	if (hw_resc->resv_irqs != nqs)
+		return true;
+
+	return false;
+}
+
+int bnge_reserve_rings(struct bnge_dev *bd)
+{
+	u16 aux_dflt_msix = bnge_aux_get_dflt_msix(bd);
+	struct bnge_hw_rings hwr = {0};
+	u16 rx_rings, old_rx_rings;
+	u16 nq = bd->nq_nr_rings;
+	u16 aux_msix = 0;
+	bool sh = false;
+	u16 tx_cp;
+	int rc;
+
+	if (!bnge_need_reserve_rings(bd))
+		return 0;
+
+	if (!bnge_aux_registered(bd)) {
+		aux_msix = bnge_get_avail_msix(bd, aux_dflt_msix);
+		if (!aux_msix)
+			bnge_aux_set_stat_ctxs(bd, 0);
+
+		if (aux_msix > aux_dflt_msix)
+			aux_msix = aux_dflt_msix;
+		hwr.nq = nq + aux_msix;
+	} else {
+		hwr.nq = bnge_nqs_demand(bd);
+	}
+
+	hwr.tx = bd->tx_nr_rings;
+	hwr.rx = bd->rx_nr_rings;
+	if (bd->flags & BNGE_EN_SHARED_CHNL)
+		sh = true;
+	hwr.cmpl = hwr.rx + hwr.tx;
+
+	hwr.vnic = bnge_get_total_vnics(bd, hwr.rx);
+
+	if (bnge_is_agg_reqd(bd))
+		hwr.rx <<= 1;
+	hwr.grp = bd->rx_nr_rings;
+	hwr.rss_ctx = bnge_rss_ctxs_in_use(bd, &hwr);
+	hwr.stat = bnge_func_stat_ctxs_demand(bd);
+	old_rx_rings = bd->hw_resc.resv_rx_rings;
+
+	rc = bnge_hwrm_reserve_rings(bd, &hwr);
+	if (rc)
+		return rc;
+
+	bnge_copy_reserved_rings(bd, &hwr);
+
+	rx_rings = hwr.rx;
+	if (bnge_is_agg_reqd(bd)) {
+		if (hwr.rx >= 2)
+			rx_rings = hwr.rx >> 1;
+		else
+			return -ENOMEM;
+	}
+
+	rx_rings = min_t(u16, rx_rings, hwr.grp);
+	hwr.nq = min_t(u16, hwr.nq, bd->nq_nr_rings);
+	if (hwr.stat > bnge_aux_get_stat_ctxs(bd))
+		hwr.stat -= bnge_aux_get_stat_ctxs(bd);
+	hwr.nq = min_t(u16, hwr.nq, hwr.stat);
+
+	/* Adjust the rings */
+	rc = bnge_adjust_rings(bd, &rx_rings, &hwr.tx, hwr.nq, sh);
+	if (bnge_is_agg_reqd(bd))
+		hwr.rx = rx_rings << 1;
+	tx_cp = hwr.tx;
+	hwr.nq = sh ? max_t(u16, tx_cp, rx_rings) : tx_cp + rx_rings;
+	bd->tx_nr_rings = hwr.tx;
+
+	if (rx_rings != bd->rx_nr_rings)
+		dev_warn(bd->dev, "RX rings resv reduced to %d than earlier %d requested\n",
+			 rx_rings, bd->rx_nr_rings);
+
+	bd->rx_nr_rings = rx_rings;
+	bd->nq_nr_rings = hwr.nq;
+
+	if (!bnge_rings_ok(&hwr))
+		return -ENOMEM;
+
+	if (old_rx_rings != bd->hw_resc.resv_rx_rings)
+		bnge_set_dflt_rss_indir_tbl(bd);
+
+	if (!bnge_aux_registered(bd)) {
+		u16 resv_msix, resv_ctx, aux_ctxs;
+		struct bnge_hw_resc *hw_resc;
+
+		hw_resc = &bd->hw_resc;
+		resv_msix = hw_resc->resv_irqs - bd->nq_nr_rings;
+		aux_msix = min_t(u16, resv_msix, aux_msix);
+		bnge_aux_set_msix_num(bd, aux_msix);
+		resv_ctx = hw_resc->resv_stat_ctxs  - bd->nq_nr_rings;
+		aux_ctxs = min(resv_ctx, bnge_aux_get_stat_ctxs(bd));
+		bnge_aux_set_stat_ctxs(bd, aux_ctxs);
+	}
+
+	return rc;
+}
+
+int bnge_alloc_irqs(struct bnge_dev *bd)
+{
+	u16 aux_msix, tx_cp, num_entries;
+	int i, irqs_demand, rc;
+	u16 max, min = 1;
+
+	irqs_demand = bnge_nqs_demand(bd);
+	max = bnge_get_max_func_irqs(bd);
+	if (irqs_demand > max)
+		irqs_demand = max;
+
+	if (!(bd->flags & BNGE_EN_SHARED_CHNL))
+		min = 2;
+
+	irqs_demand = pci_alloc_irq_vectors(bd->pdev, min, irqs_demand,
+					    PCI_IRQ_MSIX);
+	aux_msix = bnge_aux_get_msix(bd);
+	if (irqs_demand < 0 || irqs_demand < aux_msix) {
+		rc = -ENODEV;
+		goto err_free_irqs;
+	}
+
+	num_entries = irqs_demand;
+	if (pci_msix_can_alloc_dyn(bd->pdev))
+		num_entries = max;
+	bd->irq_tbl = kcalloc(num_entries, sizeof(*bd->irq_tbl), GFP_KERNEL);
+	if (!bd->irq_tbl) {
+		rc = -ENOMEM;
+		goto err_free_irqs;
+	}
+
+	for (i = 0; i < irqs_demand; i++)
+		bd->irq_tbl[i].vector = pci_irq_vector(bd->pdev, i);
+
+	bd->irqs_acquired = irqs_demand;
+	/* Reduce rings based upon num of vectors allocated.
+	 * We dont need to consider NQs as they have been calculated
+	 * and must be more than irqs_demand.
+	 */
+	rc = bnge_adjust_rings(bd, &bd->rx_nr_rings,
+			       &bd->tx_nr_rings,
+			       irqs_demand - aux_msix, min == 1);
+	if (rc)
+		goto err_free_irqs;
+
+	tx_cp = bnge_num_tx_to_cp(bd, bd->tx_nr_rings);
+	bd->nq_nr_rings = (min == 1) ?
+		max_t(u16, tx_cp, bd->rx_nr_rings) :
+		tx_cp + bd->rx_nr_rings;
+
+	/* Readjust tx_nr_rings_per_tc */
+	if (!bd->num_tc)
+		bd->tx_nr_rings_per_tc = bd->tx_nr_rings;
+
+	return 0;
+
+err_free_irqs:
+	dev_err(bd->dev, "Failed to allocate IRQs err = %d\n", rc);
+	bnge_free_irqs(bd);
+	return rc;
+}
+
+void bnge_free_irqs(struct bnge_dev *bd)
+{
+	pci_free_irq_vectors(bd->pdev);
+	kfree(bd->irq_tbl);
+	bd->irq_tbl = NULL;
+}
+
+static void _bnge_get_max_rings(struct bnge_dev *bd, u16 *max_rx,
+				u16 *max_tx, u16 *max_nq)
+{
+	struct bnge_hw_resc *hw_resc = &bd->hw_resc;
+	u16 max_ring_grps = 0, max_cp;
+	int rc;
+
+	*max_tx = hw_resc->max_tx_rings;
+	*max_rx = hw_resc->max_rx_rings;
+	*max_nq = min_t(int, bnge_get_max_func_irqs(bd),
+			hw_resc->max_stat_ctxs);
+	max_ring_grps = hw_resc->max_hw_ring_grps;
+	if (bnge_is_agg_reqd(bd))
+		*max_rx >>= 1;
+
+	max_cp = bnge_get_max_func_cp_rings(bd);
+
+	/* Fix RX and TX rings according to number of CPs available */
+	rc = bnge_fix_rings_count(max_rx, max_tx, max_cp, false);
+	if (rc) {
+		*max_rx = 0;
+		*max_tx = 0;
+	}
+
+	*max_rx = min_t(int, *max_rx, max_ring_grps);
+}
+
+static int bnge_get_max_rings(struct bnge_dev *bd, u16 *max_rx,
+			      u16 *max_tx, bool shared)
+{
+	u16 rx, tx, nq;
+
+	_bnge_get_max_rings(bd, &rx, &tx, &nq);
+	*max_rx = rx;
+	*max_tx = tx;
+	if (!rx || !tx || !nq)
+		return -ENOMEM;
+
+	return bnge_fix_rings_count(max_rx, max_tx, nq, shared);
+}
+
+static int bnge_get_dflt_rings(struct bnge_dev *bd, u16 *max_rx, u16 *max_tx,
+			       bool shared)
+{
+	int rc;
+
+	rc = bnge_get_max_rings(bd, max_rx, max_tx, shared);
+	if (rc) {
+		dev_info(bd->dev, "Not enough rings available\n");
+		return rc;
+	}
+
+	if (bnge_is_roce_en(bd)) {
+		int max_cp, max_stat, max_irq;
+
+		/* Reserve minimum resources for RoCE */
+		max_cp = bnge_get_max_func_cp_rings(bd);
+		max_stat = bnge_get_max_func_stat_ctxs(bd);
+		max_irq = bnge_get_max_func_irqs(bd);
+		if (max_cp <= BNGE_MIN_ROCE_CP_RINGS ||
+		    max_irq <= BNGE_MIN_ROCE_CP_RINGS ||
+		    max_stat <= BNGE_MIN_ROCE_STAT_CTXS)
+			return 0;
+
+		max_cp -= BNGE_MIN_ROCE_CP_RINGS;
+		max_irq -= BNGE_MIN_ROCE_CP_RINGS;
+		max_stat -= BNGE_MIN_ROCE_STAT_CTXS;
+		max_cp = min_t(u16, max_cp, max_irq);
+		max_cp = min_t(u16, max_cp, max_stat);
+		rc = bnge_adjust_rings(bd, max_rx, max_tx, max_cp, shared);
+		if (rc)
+			rc = 0;
+	}
+
+	return rc;
+}
+
+/* In initial default shared ring setting, each shared ring must have a
+ * RX/TX ring pair.
+ */
+static void bnge_trim_dflt_sh_rings(struct bnge_dev *bd)
+{
+	bd->nq_nr_rings = min_t(u16, bd->tx_nr_rings_per_tc, bd->rx_nr_rings);
+	bd->rx_nr_rings = bd->nq_nr_rings;
+	bd->tx_nr_rings_per_tc = bd->nq_nr_rings;
+	bd->tx_nr_rings = bd->tx_nr_rings_per_tc;
+}
+
+static int bnge_net_init_dflt_rings(struct bnge_dev *bd, bool sh)
+{
+	u16 dflt_rings, max_rx_rings, max_tx_rings;
+	int rc;
+
+	if (sh)
+		bd->flags |= BNGE_EN_SHARED_CHNL;
+
+	dflt_rings = netif_get_num_default_rss_queues();
+
+	rc = bnge_get_dflt_rings(bd, &max_rx_rings, &max_tx_rings, sh);
+	if (rc)
+		return rc;
+	bd->rx_nr_rings = min_t(u16, dflt_rings, max_rx_rings);
+	bd->tx_nr_rings_per_tc = min_t(u16, dflt_rings, max_tx_rings);
+	if (sh)
+		bnge_trim_dflt_sh_rings(bd);
+	else
+		bd->nq_nr_rings = bd->tx_nr_rings_per_tc + bd->rx_nr_rings;
+	bd->tx_nr_rings = bd->tx_nr_rings_per_tc;
+
+	rc = bnge_reserve_rings(bd);
+	if (rc && rc != -ENODEV)
+		dev_warn(bd->dev, "Unable to reserve tx rings\n");
+	bd->tx_nr_rings_per_tc = bd->tx_nr_rings;
+	if (sh)
+		bnge_trim_dflt_sh_rings(bd);
+
+	/* Rings may have been reduced, re-reserve them again */
+	if (bnge_need_reserve_rings(bd)) {
+		rc = bnge_reserve_rings(bd);
+		if (rc && rc != -ENODEV)
+			dev_warn(bd->dev, "Fewer rings reservation failed\n");
+		bd->tx_nr_rings_per_tc = bd->tx_nr_rings;
+	}
+	if (rc) {
+		bd->tx_nr_rings = 0;
+		bd->rx_nr_rings = 0;
+	}
+
+	return rc;
+}
+
+static int bnge_alloc_rss_indir_tbl(struct bnge_dev *bd)
+{
+	u16 entries;
+
+	entries = BNGE_MAX_RSS_TABLE_ENTRIES;
+
+	bd->rss_indir_tbl_entries = entries;
+	bd->rss_indir_tbl =
+		kmalloc_array(entries, sizeof(*bd->rss_indir_tbl), GFP_KERNEL);
+	if (!bd->rss_indir_tbl)
+		return -ENOMEM;
+
+	return 0;
+}
+
+int bnge_net_init_dflt_config(struct bnge_dev *bd)
+{
+	struct bnge_hw_resc *hw_resc;
+	int rc;
+
+	rc = bnge_alloc_rss_indir_tbl(bd);
+	if (rc)
+		return rc;
+
+	rc = bnge_net_init_dflt_rings(bd, true);
+	if (rc)
+		goto err_free_tbl;
+
+	hw_resc = &bd->hw_resc;
+	bd->max_fltr = hw_resc->max_rx_em_flows + hw_resc->max_rx_wm_flows +
+		       BNGE_L2_FLTR_MAX_FLTR;
+
+	return 0;
+
+err_free_tbl:
+	kfree(bd->rss_indir_tbl);
+	bd->rss_indir_tbl = NULL;
+	return rc;
+}
+
+void bnge_net_uninit_dflt_config(struct bnge_dev *bd)
+{
+	kfree(bd->rss_indir_tbl);
+	bd->rss_indir_tbl = NULL;
+}
+
+void bnge_aux_init_dflt_config(struct bnge_dev *bd)
+{
+	bd->aux_num_msix = bnge_aux_get_dflt_msix(bd);
+	bd->aux_num_stat_ctxs = bnge_get_dflt_aux_stat_ctxs(bd);
+}
diff --git a/drivers/net/ethernet/broadcom/bnge/bnge_resc.h b/drivers/net/ethernet/broadcom/bnge/bnge_resc.h
new file mode 100644
index 000000000000..0d6213b27580
--- /dev/null
+++ b/drivers/net/ethernet/broadcom/bnge/bnge_resc.h
@@ -0,0 +1,96 @@
+/* SPDX-License-Identifier: GPL-2.0 */
+/* Copyright (c) 2025 Broadcom */
+
+#ifndef _BNGE_RESC_H_
+#define _BNGE_RESC_H_
+
+#include "bnge_netdev.h"
+#include "bnge_rmem.h"
+
+struct bnge_hw_resc {
+	u16	min_rsscos_ctxs;
+	u16	max_rsscos_ctxs;
+	u16	resv_rsscos_ctxs;
+	u16	min_cp_rings;
+	u16	max_cp_rings;
+	u16	resv_cp_rings;
+	u16	min_tx_rings;
+	u16	max_tx_rings;
+	u16	resv_tx_rings;
+	u16	max_tx_sch_inputs;
+	u16	min_rx_rings;
+	u16	max_rx_rings;
+	u16	resv_rx_rings;
+	u16	min_hw_ring_grps;
+	u16	max_hw_ring_grps;
+	u16	resv_hw_ring_grps;
+	u16	min_l2_ctxs;
+	u16	max_l2_ctxs;
+	u16	min_vnics;
+	u16	max_vnics;
+	u16	resv_vnics;
+	u16	min_stat_ctxs;
+	u16	max_stat_ctxs;
+	u16	resv_stat_ctxs;
+	u16	max_nqs;
+	u16	max_irqs;
+	u16	resv_irqs;
+	u32	max_encap_records;
+	u32	max_decap_records;
+	u32	max_tx_em_flows;
+	u32	max_tx_wm_flows;
+	u32	max_rx_em_flows;
+	u32	max_rx_wm_flows;
+};
+
+struct bnge_hw_rings {
+	u16 tx;
+	u16 rx;
+	u16 grp;
+	u16 nq;
+	u16 cmpl;
+	u16 stat;
+	u16 vnic;
+	u16 rss_ctx;
+};
+
+/* "TXRX", 2 hypens, plus maximum integer */
+#define BNGE_IRQ_NAME_EXTRA	17
+struct bnge_irq {
+	irq_handler_t	handler;
+	unsigned int	vector;
+	u8		requested:1;
+	u8		have_cpumask:1;
+	char		name[IFNAMSIZ + BNGE_IRQ_NAME_EXTRA];
+	cpumask_var_t	cpu_mask;
+};
+
+int bnge_reserve_rings(struct bnge_dev *bd);
+int bnge_fix_rings_count(u16 *rx, u16 *tx, u16 max, bool shared);
+int bnge_alloc_irqs(struct bnge_dev *bd);
+void bnge_free_irqs(struct bnge_dev *bd);
+int bnge_net_init_dflt_config(struct bnge_dev *bd);
+void bnge_net_uninit_dflt_config(struct bnge_dev *bd);
+void bnge_aux_init_dflt_config(struct bnge_dev *bd);
+u32 bnge_get_rxfh_indir_size(struct bnge_dev *bd);
+int bnge_cal_nr_rss_ctxs(u16 rx_rings);
+
+static inline u32
+bnge_adjust_pow_two(u32 total_ent, u16 ent_per_blk)
+{
+	u32 blks = total_ent / ent_per_blk;
+
+	if (blks == 0 || blks == 1)
+		return ++blks;
+
+	if (!is_power_of_2(blks))
+		blks = roundup_pow_of_two(blks);
+
+	return blks;
+}
+
+#define BNGE_MAX_ROCE_MSIX		64
+#define BNGE_MIN_ROCE_CP_RINGS		2
+#define BNGE_MIN_ROCE_STAT_CTXS		1
+
+#endif /* _BNGE_RESC_H_ */
diff --git a/drivers/net/ethernet/broadcom/bnge/bnge_rmem.c b/drivers/net/ethernet/broadcom/bnge/bnge_rmem.c
new file mode 100644
index 000000000000..79f5ce2e5d08
--- /dev/null
+++ b/drivers/net/ethernet/broadcom/bnge/bnge_rmem.c
@@ -0,0 +1,499 @@
+// SPDX-License-Identifier: GPL-2.0
+// Copyright (c) 2025 Broadcom.
+
+#include <linux/etherdevice.h>
+#include <linux/init.h>
+#include <linux/module.h>
+#include <linux/pci.h>
+#include <linux/mm.h>
+#include <linux/dma-mapping.h>
+#include <linux/vmalloc.h>
+#include <linux/crash_dump.h>
+#include <linux/bnxt/hsi.h>
+
+#include "bnge.h"
+#include "bnge_hwrm_lib.h"
+#include "bnge_rmem.h"
+
+static void bnge_init_ctx_mem(struct bnge_ctx_mem_type *ctxm,
+			      void *p, int len)
+{
+	u8 init_val = ctxm->init_value;
+	u16 offset = ctxm->init_offset;
+	u8 *p2 = p;
+	int i;
+
+	if (!init_val)
+		return;
+	if (offset == BNGE_CTX_INIT_INVALID_OFFSET) {
+		memset(p, init_val, len);
+		return;
+	}
+	for (i = 0; i < len; i += ctxm->entry_size)
+		*(p2 + i + offset) = init_val;
+}
+
+void bnge_free_ring(struct bnge_dev *bd, struct bnge_ring_mem_info *rmem)
+{
+	struct pci_dev *pdev = bd->pdev;
+	int i;
+
+	if (!rmem->pg_arr)
+		goto skip_pages;
+
+	for (i = 0; i < rmem->nr_pages; i++) {
+		if (!rmem->pg_arr[i])
+			continue;
+
+		dma_free_coherent(&pdev->dev, rmem->page_size,
+				  rmem->pg_arr[i], rmem->dma_arr[i]);
+
+		rmem->pg_arr[i] = NULL;
+	}
+skip_pages:
+	if (rmem->pg_tbl) {
+		size_t pg_tbl_size = rmem->nr_pages * 8;
+
+		if (rmem->flags & BNGE_RMEM_USE_FULL_PAGE_FLAG)
+			pg_tbl_size = rmem->page_size;
+		dma_free_coherent(&pdev->dev, pg_tbl_size,
+				  rmem->pg_tbl, rmem->dma_pg_tbl);
+		rmem->pg_tbl = NULL;
+	}
+	if (rmem->vmem_size && *rmem->vmem) {
+		vfree(*rmem->vmem);
+		*rmem->vmem = NULL;
+	}
+}
+
+int bnge_alloc_ring(struct bnge_dev *bd, struct bnge_ring_mem_info *rmem)
+{
+	struct pci_dev *pdev = bd->pdev;
+	u64 valid_bit = 0;
+	int i;
+
+	if (rmem->flags & (BNGE_RMEM_VALID_PTE_FLAG | BNGE_RMEM_RING_PTE_FLAG))
+		valid_bit = PTU_PTE_VALID;
+
+	if ((rmem->nr_pages > 1 || rmem->depth > 0) && !rmem->pg_tbl) {
+		size_t pg_tbl_size = rmem->nr_pages * 8;
+
+		if (rmem->flags & BNGE_RMEM_USE_FULL_PAGE_FLAG)
+			pg_tbl_size = rmem->page_size;
+		rmem->pg_tbl = dma_alloc_coherent(&pdev->dev, pg_tbl_size,
+						  &rmem->dma_pg_tbl,
+						  GFP_KERNEL);
+		if (!rmem->pg_tbl)
+			return -ENOMEM;
+	}
+
+	for (i = 0; i < rmem->nr_pages; i++) {
+		u64 extra_bits = valid_bit;
+
+		rmem->pg_arr[i] = dma_alloc_coherent(&pdev->dev,
+						     rmem->page_size,
+						     &rmem->dma_arr[i],
+						     GFP_KERNEL);
+		if (!rmem->pg_arr[i])
+			goto err_free_ring;
+
+		if (rmem->ctx_mem)
+			bnge_init_ctx_mem(rmem->ctx_mem, rmem->pg_arr[i],
+					  rmem->page_size);
+
+		if (rmem->nr_pages > 1 || rmem->depth > 0) {
+			if (i == rmem->nr_pages - 2 &&
+			    (rmem->flags & BNGE_RMEM_RING_PTE_FLAG))
+				extra_bits |= PTU_PTE_NEXT_TO_LAST;
+			else if (i == rmem->nr_pages - 1 &&
+				 (rmem->flags & BNGE_RMEM_RING_PTE_FLAG))
+				extra_bits |= PTU_PTE_LAST;
+			rmem->pg_tbl[i] =
+				cpu_to_le64(rmem->dma_arr[i] | extra_bits);
+		}
+	}
+
+	if (rmem->vmem_size) {
+		*rmem->vmem = vzalloc(rmem->vmem_size);
+		if (!(*rmem->vmem))
+			goto err_free_ring;
+	}
+	return 0;
+
+err_free_ring:
+	bnge_free_ring(bd, rmem);
+	return -ENOMEM;
+}
+
+static int bnge_alloc_ctx_one_lvl(struct bnge_dev *bd,
+				  struct bnge_ctx_pg_info *ctx_pg)
+{
+	struct bnge_ring_mem_info *rmem = &ctx_pg->ring_mem;
+
+	rmem->page_size = BNGE_PAGE_SIZE;
+	rmem->pg_arr = ctx_pg->ctx_pg_arr;
+	rmem->dma_arr = ctx_pg->ctx_dma_arr;
+	rmem->flags = BNGE_RMEM_VALID_PTE_FLAG;
+	if (rmem->depth >= 1)
+		rmem->flags |= BNGE_RMEM_USE_FULL_PAGE_FLAG;
+	return bnge_alloc_ring(bd, rmem);
+}
+
+static int bnge_alloc_ctx_pg_tbls(struct bnge_dev *bd,
+				  struct bnge_ctx_pg_info *ctx_pg, u32 mem_size,
+				  u8 depth, struct bnge_ctx_mem_type *ctxm)
+{
+	struct bnge_ring_mem_info *rmem = &ctx_pg->ring_mem;
+	int rc;
+
+	if (!mem_size)
+		return -EINVAL;
+
+	ctx_pg->nr_pages = DIV_ROUND_UP(mem_size, BNGE_PAGE_SIZE);
+	if (ctx_pg->nr_pages > MAX_CTX_TOTAL_PAGES) {
+		ctx_pg->nr_pages = 0;
+		return -EINVAL;
+	}
+	if (ctx_pg->nr_pages > MAX_CTX_PAGES || depth > 1) {
+		int nr_tbls, i;
+
+		rmem->depth = 2;
+		ctx_pg->ctx_pg_tbl = kcalloc(MAX_CTX_PAGES, sizeof(ctx_pg),
+					     GFP_KERNEL);
+		if (!ctx_pg->ctx_pg_tbl)
+			return -ENOMEM;
+		nr_tbls = DIV_ROUND_UP(ctx_pg->nr_pages, MAX_CTX_PAGES);
+		rmem->nr_pages = nr_tbls;
+		rc = bnge_alloc_ctx_one_lvl(bd, ctx_pg);
+		if (rc)
+			return rc;
+		for (i = 0; i < nr_tbls; i++) {
+			struct bnge_ctx_pg_info *pg_tbl;
+
+			pg_tbl = kzalloc(sizeof(*pg_tbl), GFP_KERNEL);
+			if (!pg_tbl)
+				return -ENOMEM;
+			ctx_pg->ctx_pg_tbl[i] = pg_tbl;
+			rmem = &pg_tbl->ring_mem;
+			rmem->pg_tbl = ctx_pg->ctx_pg_arr[i];
+			rmem->dma_pg_tbl = ctx_pg->ctx_dma_arr[i];
+			rmem->depth = 1;
+			rmem->nr_pages = MAX_CTX_PAGES;
+			rmem->ctx_mem = ctxm;
+			if (i == (nr_tbls - 1)) {
+				int rem = ctx_pg->nr_pages % MAX_CTX_PAGES;
+
+				if (rem)
+					rmem->nr_pages = rem;
+			}
+			rc = bnge_alloc_ctx_one_lvl(bd, pg_tbl);
+			if (rc)
+				break;
+		}
+	} else {
+		rmem->nr_pages = DIV_ROUND_UP(mem_size, BNGE_PAGE_SIZE);
+		if (rmem->nr_pages > 1 || depth)
+			rmem->depth = 1;
+		rmem->ctx_mem = ctxm;
+		rc = bnge_alloc_ctx_one_lvl(bd, ctx_pg);
+	}
+
+	return rc;
+}
+
+static void bnge_free_ctx_pg_tbls(struct bnge_dev *bd,
+				  struct bnge_ctx_pg_info *ctx_pg)
+{
+	struct bnge_ring_mem_info *rmem = &ctx_pg->ring_mem;
+
+	if (rmem->depth > 1 || ctx_pg->nr_pages > MAX_CTX_PAGES ||
+	    ctx_pg->ctx_pg_tbl) {
+		int i, nr_tbls = rmem->nr_pages;
+
+		for (i = 0; i < nr_tbls; i++) {
+			struct bnge_ctx_pg_info *pg_tbl;
+			struct bnge_ring_mem_info *rmem2;
+
+			pg_tbl = ctx_pg->ctx_pg_tbl[i];
+			if (!pg_tbl)
+				continue;
+			rmem2 = &pg_tbl->ring_mem;
+			bnge_free_ring(bd, rmem2);
+			ctx_pg->ctx_pg_arr[i] = NULL;
+			kfree(pg_tbl);
+			ctx_pg->ctx_pg_tbl[i] = NULL;
+		}
+		kfree(ctx_pg->ctx_pg_tbl);
+		ctx_pg->ctx_pg_tbl = NULL;
+	}
+	bnge_free_ring(bd, rmem);
+	ctx_pg->nr_pages = 0;
+}
+
+static int bnge_setup_ctxm_pg_tbls(struct bnge_dev *bd,
+				   struct bnge_ctx_mem_type *ctxm, u32 entries,
+				   u8 pg_lvl)
+{
+	struct bnge_ctx_pg_info *ctx_pg = ctxm->pg_info;
+	int i, rc = 0, n = 1;
+	u32 mem_size;
+
+	if (!ctxm->entry_size || !ctx_pg)
+		return -EINVAL;
+	if (ctxm->instance_bmap)
+		n = hweight32(ctxm->instance_bmap);
+	if (ctxm->entry_multiple)
+		entries = roundup(entries, ctxm->entry_multiple);
+	entries = clamp_t(u32, entries, ctxm->min_entries, ctxm->max_entries);
+	mem_size = entries * ctxm->entry_size;
+	for (i = 0; i < n && !rc; i++) {
+		ctx_pg[i].entries = entries;
+		rc = bnge_alloc_ctx_pg_tbls(bd, &ctx_pg[i], mem_size, pg_lvl,
+					    ctxm->init_value ? ctxm : NULL);
+	}
+
+	return rc;
+}
+
+static int bnge_backing_store_cfg(struct bnge_dev *bd, u32 ena)
+{
+	struct bnge_ctx_mem_info *ctx = bd->ctx;
+	struct bnge_ctx_mem_type *ctxm;
+	u16 last_type;
+	int rc = 0;
+	u16 type;
+
+	if (!ena)
+		return 0;
+	else if (ena & FUNC_BACKING_STORE_CFG_REQ_ENABLES_TIM)
+		last_type = BNGE_CTX_MAX - 1;
+	else
+		last_type = BNGE_CTX_L2_MAX - 1;
+	ctx->ctx_arr[last_type].last = 1;
+
+	for (type = 0 ; type < BNGE_CTX_V2_MAX; type++) {
+		ctxm = &ctx->ctx_arr[type];
+
+		rc = bnge_hwrm_func_backing_store(bd, ctxm, ctxm->last);
+		if (rc)
+			return rc;
+	}
+
+	return 0;
+}
+
+void bnge_free_ctx_mem(struct bnge_dev *bd)
+{
+	struct bnge_ctx_mem_info *ctx = bd->ctx;
+	u16 type;
+
+	if (!ctx)
+		return;
+
+	for (type = 0; type < BNGE_CTX_V2_MAX; type++) {
+		struct bnge_ctx_mem_type *ctxm = &ctx->ctx_arr[type];
+		struct bnge_ctx_pg_info *ctx_pg = ctxm->pg_info;
+		int i, n = 1;
+
+		if (!ctx_pg)
+			continue;
+		if (ctxm->instance_bmap)
+			n = hweight32(ctxm->instance_bmap);
+		for (i = 0; i < n; i++)
+			bnge_free_ctx_pg_tbls(bd, &ctx_pg[i]);
+
+		kfree(ctx_pg);
+		ctxm->pg_info = NULL;
+	}
+
+	ctx->flags &= ~BNGE_CTX_FLAG_INITED;
+	kfree(ctx);
+	bd->ctx = NULL;
+}
+
+#define FUNC_BACKING_STORE_CFG_REQ_DFLT_ENABLES			\
+	(FUNC_BACKING_STORE_CFG_REQ_ENABLES_QP |		\
+	 FUNC_BACKING_STORE_CFG_REQ_ENABLES_SRQ |		\
+	 FUNC_BACKING_STORE_CFG_REQ_ENABLES_CQ |		\
+	 FUNC_BACKING_STORE_CFG_REQ_ENABLES_VNIC |		\
+	 FUNC_BACKING_STORE_CFG_REQ_ENABLES_STAT)
+
+int bnge_alloc_ctx_mem(struct bnge_dev *bd)
+{
+	struct bnge_ctx_mem_type *ctxm;
+	struct bnge_ctx_mem_info *ctx;
+	u32 l2_qps, qp1_qps, max_qps;
+	u32 ena, entries_sp, entries;
+	u32 srqs, max_srqs, min;
+	u32 num_mr, num_ah;
+	u32 extra_srqs = 0;
+	u32 extra_qps = 0;
+	u32 fast_qpmd_qps;
+	u8 pg_lvl = 1;
+	int i, rc;
+
+	rc = bnge_hwrm_func_backing_store_qcaps(bd);
+	if (rc) {
+		dev_err(bd->dev, "Failed querying ctx mem caps, rc: %d\n", rc);
+		return rc;
+	}
+
+	ctx = bd->ctx;
+	if (!ctx || (ctx->flags & BNGE_CTX_FLAG_INITED))
+		return 0;
+
+	ctxm = &ctx->ctx_arr[BNGE_CTX_QP];
+	l2_qps = ctxm->qp_l2_entries;
+	qp1_qps = ctxm->qp_qp1_entries;
+	fast_qpmd_qps = ctxm->qp_fast_qpmd_entries;
+	max_qps = ctxm->max_entries;
+	ctxm = &ctx->ctx_arr[BNGE_CTX_SRQ];
+	srqs = ctxm->srq_l2_entries;
+	max_srqs = ctxm->max_entries;
+	ena = 0;
+	if (bnge_is_roce_en(bd) && !is_kdump_kernel()) {
+		pg_lvl = 2;
+		extra_qps = min_t(u32, 65536, max_qps - l2_qps - qp1_qps);
+		/* allocate extra qps if fast qp destroy feature enabled */
+		extra_qps += fast_qpmd_qps;
+		extra_srqs = min_t(u32, 8192, max_srqs - srqs);
+		if (fast_qpmd_qps)
+			ena |= FUNC_BACKING_STORE_CFG_REQ_ENABLES_QP_FAST_QPMD;
+	}
+
+	ctxm = &ctx->ctx_arr[BNGE_CTX_QP];
+	rc = bnge_setup_ctxm_pg_tbls(bd, ctxm, l2_qps + qp1_qps + extra_qps,
+				     pg_lvl);
+	if (rc)
+		return rc;
+
+	ctxm = &ctx->ctx_arr[BNGE_CTX_SRQ];
+	rc = bnge_setup_ctxm_pg_tbls(bd, ctxm, srqs + extra_srqs, pg_lvl);
+	if (rc)
+		return rc;
+
+	ctxm = &ctx->ctx_arr[BNGE_CTX_CQ];
+	rc = bnge_setup_ctxm_pg_tbls(bd, ctxm, ctxm->cq_l2_entries +
+				     extra_qps * 2, pg_lvl);
+	if (rc)
+		return rc;
+
+	ctxm = &ctx->ctx_arr[BNGE_CTX_VNIC];
+	rc = bnge_setup_ctxm_pg_tbls(bd, ctxm, ctxm->max_entries, 1);
+	if (rc)
+		return rc;
+
+	ctxm = &ctx->ctx_arr[BNGE_CTX_STAT];
+	rc = bnge_setup_ctxm_pg_tbls(bd, ctxm, ctxm->max_entries, 1);
+	if (rc)
+		return rc;
+
+	if (!bnge_is_roce_en(bd))
+		goto skip_rdma;
+
+	ctxm = &ctx->ctx_arr[BNGE_CTX_MRAV];
+	/* 128K extra is needed to accommodate static AH context
+	 * allocation by f/w.
+	 */
+	num_mr = min_t(u32, ctxm->max_entries / 2, 1024 * 256);
+	num_ah = min_t(u32, num_mr, 1024 * 128);
+	ctxm->split_entry_cnt = BNGE_CTX_MRAV_AV_SPLIT_ENTRY + 1;
+	if (!ctxm->mrav_av_entries || ctxm->mrav_av_entries > num_ah)
+		ctxm->mrav_av_entries = num_ah;
+
+	rc = bnge_setup_ctxm_pg_tbls(bd, ctxm, num_mr + num_ah, 2);
+	if (rc)
+		return rc;
+	ena |= FUNC_BACKING_STORE_CFG_REQ_ENABLES_MRAV;
+
+	ctxm = &ctx->ctx_arr[BNGE_CTX_TIM];
+	rc = bnge_setup_ctxm_pg_tbls(bd, ctxm, l2_qps + qp1_qps + extra_qps, 1);
+	if (rc)
+		return rc;
+	ena |= FUNC_BACKING_STORE_CFG_REQ_ENABLES_TIM;
+
+skip_rdma:
+	ctxm = &ctx->ctx_arr[BNGE_CTX_STQM];
+	min = ctxm->min_entries;
+	entries_sp = ctx->ctx_arr[BNGE_CTX_VNIC].vnic_entries + l2_qps +
+		     2 * (extra_qps + qp1_qps) + min;
+	rc = bnge_setup_ctxm_pg_tbls(bd, ctxm, entries_sp, 2);
+	if (rc)
+		return rc;
+
+	ctxm = &ctx->ctx_arr[BNGE_CTX_FTQM];
+	entries = l2_qps + 2 * (extra_qps + qp1_qps);
+	rc = bnge_setup_ctxm_pg_tbls(bd, ctxm, entries, 2);
+	if (rc)
+		return rc;
+	for (i = 0; i < ctx->tqm_fp_rings_count + 1; i++)
+		ena |= FUNC_BACKING_STORE_CFG_REQ_ENABLES_TQM_SP << i;
+	ena |= FUNC_BACKING_STORE_CFG_REQ_DFLT_ENABLES;
+
+	rc = bnge_backing_store_cfg(bd, ena);
+	if (rc) {
+		dev_err(bd->dev, "Failed configuring ctx mem, rc: %d\n", rc);
+		return rc;
+	}
+	ctx->flags |= BNGE_CTX_FLAG_INITED;
+
+	return 0;
+}
+
+void bnge_init_ring_struct(struct bnge_net *bn)
+{
+	struct bnge_dev *bd = bn->bd;
+	int i, j;
+
+	for (i = 0; i < bd->nq_nr_rings; i++) {
+		struct bnge_napi *bnapi = bn->bnapi[i];
+		struct bnge_ring_mem_info *rmem;
+		struct bnge_nq_ring_info *nqr;
+		struct bnge_rx_ring_info *rxr;
+		struct bnge_tx_ring_info *txr;
+		struct bnge_ring_struct *ring;
+
+		nqr = &bnapi->nq_ring;
+		ring = &nqr->ring_struct;
+		rmem = &ring->ring_mem;
+		rmem->nr_pages = bn->cp_nr_pages;
+		rmem->page_size = HW_CMPD_RING_SIZE;
+		rmem->pg_arr = (void **)nqr->desc_ring;
+		rmem->dma_arr = nqr->desc_mapping;
+		rmem->vmem_size = 0;
+
+		rxr = bnapi->rx_ring;
+		if (!rxr)
+			goto skip_rx;
+
+		ring = &rxr->rx_ring_struct;
+		rmem = &ring->ring_mem;
+		rmem->nr_pages = bn->rx_nr_pages;
+		rmem->page_size = HW_RXBD_RING_SIZE;
+		rmem->pg_arr = (void **)rxr->rx_desc_ring;
+		rmem->dma_arr = rxr->rx_desc_mapping;
+		rmem->vmem_size = SW_RXBD_RING_SIZE * bn->rx_nr_pages;
+		rmem->vmem = (void **)&rxr->rx_buf_ring;
+
+		ring = &rxr->rx_agg_ring_struct;
+		rmem = &ring->ring_mem;
+		rmem->nr_pages = bn->rx_agg_nr_pages;
+		rmem->page_size = HW_RXBD_RING_SIZE;
+		rmem->pg_arr = (void **)rxr->rx_agg_desc_ring;
+		rmem->dma_arr = rxr->rx_agg_desc_mapping;
+		rmem->vmem_size = SW_RXBD_AGG_RING_SIZE * bn->rx_agg_nr_pages;
+		rmem->vmem = (void **)&rxr->rx_agg_buf_ring;
+
+skip_rx:
+		bnge_for_each_napi_tx(j, bnapi, txr) {
+			ring = &txr->tx_ring_struct;
+			rmem = &ring->ring_mem;
+			rmem->nr_pages = bn->tx_nr_pages;
+			rmem->page_size = HW_TXBD_RING_SIZE;
+			rmem->pg_arr = (void **)txr->tx_desc_ring;
+			rmem->dma_arr = txr->tx_desc_mapping;
+			rmem->vmem_size = SW_TXBD_RING_SIZE * bn->tx_nr_pages;
+			rmem->vmem = (void **)&txr->tx_buf_ring;
+		}
+	}
+}
diff --git a/drivers/net/ethernet/broadcom/bnge/bnge_rmem.h b/drivers/net/ethernet/broadcom/bnge/bnge_rmem.h
new file mode 100644
index 000000000000..341c7f81ed09
--- /dev/null
+++ b/drivers/net/ethernet/broadcom/bnge/bnge_rmem.h
@@ -0,0 +1,202 @@
+/* SPDX-License-Identifier: GPL-2.0 */
+/* Copyright (c) 2025 Broadcom */
+
+#ifndef _BNGE_RMEM_H_
+#define _BNGE_RMEM_H_
+
+struct bnge_ctx_mem_type;
+struct bnge_dev;
+struct bnge_net;
+
+#define PTU_PTE_VALID             0x1UL
+#define PTU_PTE_LAST              0x2UL
+#define PTU_PTE_NEXT_TO_LAST      0x4UL
+
+struct bnge_ring_mem_info {
+	/* Number of pages to next level */
+	int			nr_pages;
+	int			page_size;
+	u16			flags;
+#define BNGE_RMEM_VALID_PTE_FLAG	1
+#define BNGE_RMEM_RING_PTE_FLAG		2
+#define BNGE_RMEM_USE_FULL_PAGE_FLAG	4
+
+	u16			depth;
+
+	void			**pg_arr;
+	dma_addr_t		*dma_arr;
+
+	__le64			*pg_tbl;
+	dma_addr_t		dma_pg_tbl;
+
+	int			vmem_size;
+	void			**vmem;
+
+	struct bnge_ctx_mem_type	*ctx_mem;
+};
+
+/* The hardware supports certain page sizes.
+ * Use the supported page sizes to allocate the rings.
+ */
+#if (PAGE_SHIFT < 12)
+#define BNGE_PAGE_SHIFT	12
+#elif (PAGE_SHIFT <= 13)
+#define BNGE_PAGE_SHIFT	PAGE_SHIFT
+#elif (PAGE_SHIFT < 16)
+#define BNGE_PAGE_SHIFT	13
+#else
+#define BNGE_PAGE_SHIFT	16
+#endif
+#define BNGE_PAGE_SIZE	(1 << BNGE_PAGE_SHIFT)
+/* The RXBD length is 16-bit so we can only support page sizes < 64K */
+#if (PAGE_SHIFT > 15)
+#define BNGE_RX_PAGE_SHIFT 15
+#else
+#define BNGE_RX_PAGE_SHIFT PAGE_SHIFT
+#endif
+#define MAX_CTX_PAGES	(BNGE_PAGE_SIZE / 8)
+#define MAX_CTX_TOTAL_PAGES	(MAX_CTX_PAGES * MAX_CTX_PAGES)
+
+struct bnge_ctx_pg_info {
+	u32		entries;
+	u32		nr_pages;
+	void		*ctx_pg_arr[MAX_CTX_PAGES];
+	dma_addr_t	ctx_dma_arr[MAX_CTX_PAGES];
+	struct bnge_ring_mem_info ring_mem;
+	struct bnge_ctx_pg_info **ctx_pg_tbl;
+};
+
+#define BNGE_MAX_TQM_SP_RINGS		1
+#define BNGE_MAX_TQM_FP_RINGS		8
+#define BNGE_MAX_TQM_RINGS		\
+	(BNGE_MAX_TQM_SP_RINGS + BNGE_MAX_TQM_FP_RINGS)
+#define BNGE_BACKING_STORE_CFG_LEGACY_LEN	256
+#define BNGE_SET_CTX_PAGE_ATTR(attr)					\
+do {									\
+	if (BNGE_PAGE_SIZE == 0x2000)					\
+		attr = FUNC_BACKING_STORE_CFG_REQ_SRQ_PG_SIZE_PG_8K;	\
+	else if (BNGE_PAGE_SIZE == 0x10000)				\
+		attr = FUNC_BACKING_STORE_CFG_REQ_QPC_PG_SIZE_PG_64K;	\
+	else								\
+		attr = FUNC_BACKING_STORE_CFG_REQ_QPC_PG_SIZE_PG_4K;	\
+} while (0)
+
+#define BNGE_CTX_MRAV_AV_SPLIT_ENTRY	0
+
+#define BNGE_CTX_QP	\
+	FUNC_BACKING_STORE_QCAPS_V2_REQ_TYPE_QP
+#define BNGE_CTX_SRQ	\
+	FUNC_BACKING_STORE_QCAPS_V2_REQ_TYPE_SRQ
+#define BNGE_CTX_CQ	\
+	FUNC_BACKING_STORE_QCAPS_V2_REQ_TYPE_CQ
+#define BNGE_CTX_VNIC	\
+	FUNC_BACKING_STORE_QCAPS_V2_REQ_TYPE_VNIC
+#define BNGE_CTX_STAT	\
+	FUNC_BACKING_STORE_QCAPS_V2_REQ_TYPE_STAT
+#define BNGE_CTX_STQM	\
+	FUNC_BACKING_STORE_QCAPS_V2_REQ_TYPE_SP_TQM_RING
+#define BNGE_CTX_FTQM	\
+	FUNC_BACKING_STORE_QCAPS_V2_REQ_TYPE_FP_TQM_RING
+#define BNGE_CTX_MRAV	\
+	FUNC_BACKING_STORE_QCAPS_V2_REQ_TYPE_MRAV
+#define BNGE_CTX_TIM	\
+	FUNC_BACKING_STORE_QCAPS_V2_REQ_TYPE_TIM
+#define BNGE_CTX_TCK	\
+	FUNC_BACKING_STORE_QCAPS_V2_REQ_TYPE_TX_CK
+#define BNGE_CTX_RCK	\
+	FUNC_BACKING_STORE_QCAPS_V2_REQ_TYPE_RX_CK
+#define BNGE_CTX_MTQM	\
+	FUNC_BACKING_STORE_QCAPS_V2_REQ_TYPE_MP_TQM_RING
+#define BNGE_CTX_SQDBS	\
+	FUNC_BACKING_STORE_QCAPS_V2_REQ_TYPE_SQ_DB_SHADOW
+#define BNGE_CTX_RQDBS	\
+	FUNC_BACKING_STORE_QCAPS_V2_REQ_TYPE_RQ_DB_SHADOW
+#define BNGE_CTX_SRQDBS	\
+	FUNC_BACKING_STORE_QCAPS_V2_REQ_TYPE_SRQ_DB_SHADOW
+#define BNGE_CTX_CQDBS	\
+	FUNC_BACKING_STORE_QCAPS_V2_REQ_TYPE_CQ_DB_SHADOW
+#define BNGE_CTX_SRT_TRACE	\
+	FUNC_BACKING_STORE_QCAPS_V2_REQ_TYPE_SRT_TRACE
+#define BNGE_CTX_SRT2_TRACE	\
+	FUNC_BACKING_STORE_QCAPS_V2_REQ_TYPE_SRT2_TRACE
+#define BNGE_CTX_CRT_TRACE	\
+	FUNC_BACKING_STORE_QCAPS_V2_REQ_TYPE_CRT_TRACE
+#define BNGE_CTX_CRT2_TRACE	\
+	FUNC_BACKING_STORE_QCAPS_V2_REQ_TYPE_CRT2_TRACE
+#define BNGE_CTX_RIGP0_TRACE	\
+	FUNC_BACKING_STORE_QCAPS_V2_REQ_TYPE_RIGP0_TRACE
+#define BNGE_CTX_L2_HWRM_TRACE	\
+	FUNC_BACKING_STORE_QCAPS_V2_REQ_TYPE_L2_HWRM_TRACE
+#define BNGE_CTX_ROCE_HWRM_TRACE	\
+	FUNC_BACKING_STORE_QCAPS_V2_REQ_TYPE_ROCE_HWRM_TRACE
+
+#define BNGE_CTX_MAX		(BNGE_CTX_TIM + 1)
+#define BNGE_CTX_L2_MAX		(BNGE_CTX_FTQM + 1)
+#define BNGE_CTX_INV		((u16)-1)
+
+#define BNGE_CTX_V2_MAX	\
+	(FUNC_BACKING_STORE_QCAPS_V2_REQ_TYPE_ROCE_HWRM_TRACE + 1)
+
+#define BNGE_BS_CFG_ALL_DONE	\
+	FUNC_BACKING_STORE_CFG_V2_REQ_FLAGS_BS_CFG_ALL_DONE
+
+struct bnge_ctx_mem_type {
+	u16	type;
+	u16	entry_size;
+	u32	flags;
+#define BNGE_CTX_MEM_TYPE_VALID	\
+	FUNC_BACKING_STORE_QCAPS_V2_RESP_FLAGS_TYPE_VALID
+	u32	instance_bmap;
+	u8	init_value;
+	u8	entry_multiple;
+	u16	init_offset;
+#define	BNGE_CTX_INIT_INVALID_OFFSET	0xffff
+	u32	max_entries;
+	u32	min_entries;
+	u8	last:1;
+	u8	split_entry_cnt;
+#define BNGE_MAX_SPLIT_ENTRY	4
+	union {
+		struct {
+			u32	qp_l2_entries;
+			u32	qp_qp1_entries;
+			u32	qp_fast_qpmd_entries;
+		};
+		u32	srq_l2_entries;
+		u32	cq_l2_entries;
+		u32	vnic_entries;
+		struct {
+			u32	mrav_av_entries;
+			u32	mrav_num_entries_units;
+		};
+		u32	split[BNGE_MAX_SPLIT_ENTRY];
+	};
+	struct bnge_ctx_pg_info	*pg_info;
+};
+
+struct bnge_ctx_mem_info {
+	u8	tqm_fp_rings_count;
+	u32	flags;
+#define BNGE_CTX_FLAG_INITED	0x01
+	struct bnge_ctx_mem_type	ctx_arr[BNGE_CTX_V2_MAX];
+};
+
+struct bnge_ring_struct {
+	struct bnge_ring_mem_info	ring_mem;
+
+	u16			fw_ring_id;
+	union {
+		u16		grp_idx;
+		u16		map_idx; /* Used by NQs */
+	};
+	u32			handle;
+	u8			queue_id;
+};
+
+int bnge_alloc_ring(struct bnge_dev *bd, struct bnge_ring_mem_info *rmem);
+void bnge_free_ring(struct bnge_dev *bd, struct bnge_ring_mem_info *rmem);
+int bnge_alloc_ctx_mem(struct bnge_dev *bd);
+void bnge_free_ctx_mem(struct bnge_dev *bd);
+void bnge_init_ring_struct(struct bnge_net *bn);
+
+#endif /* _BNGE_RMEM_H_ */
diff --git a/drivers/net/ethernet/broadcom/bnx2.c b/drivers/net/ethernet/broadcom/bnx2.c
index e3af1f3cb61f..cb1011f6fd30 100644
--- a/drivers/net/ethernet/broadcom/bnx2.c
+++ b/drivers/net/ethernet/broadcom/bnx2.c
@@ -48,7 +48,6 @@
 #include <linux/cache.h>
 #include <linux/firmware.h>
 #include <linux/log2.h>
-#include <linux/aer.h>
 #include <linux/crash_dump.h>
 
 #if IS_ENABLED(CONFIG_CNIC)
@@ -59,8 +58,6 @@
 #include "bnx2_fw.h"
 
 #define DRV_MODULE_NAME		"bnx2"
-#define DRV_MODULE_VERSION	"2.2.6"
-#define DRV_MODULE_RELDATE	"January 29, 2014"
 #define FW_MIPS_FILE_06		"bnx2/bnx2-mips-06-6.2.3.fw"
 #define FW_RV2P_FILE_06		"bnx2/bnx2-rv2p-06-6.0.15.fw"
 #define FW_MIPS_FILE_09		"bnx2/bnx2-mips-09-6.2.1b.fw"
@@ -72,13 +69,9 @@
 /* Time in jiffies before concluding the transmitter is hung. */
 #define TX_TIMEOUT  (5*HZ)
 
-static char version[] =
-	"QLogic " DRV_MODULE_NAME " Gigabit Ethernet Driver v" DRV_MODULE_VERSION " (" DRV_MODULE_RELDATE ")\n";
-
 MODULE_AUTHOR("Michael Chan <mchan@broadcom.com>");
 MODULE_DESCRIPTION("QLogic BCM5706/5708/5709/5716 Driver");
 MODULE_LICENSE("GPL");
-MODULE_VERSION(DRV_MODULE_VERSION);
 MODULE_FIRMWARE(FW_MIPS_FILE_06);
 MODULE_FIRMWARE(FW_RV2P_FILE_06);
 MODULE_FIRMWARE(FW_MIPS_FILE_09);
@@ -87,7 +80,7 @@ MODULE_FIRMWARE(FW_RV2P_FILE_09_Ax);
 
 static int disable_msi = 0;
 
-module_param(disable_msi, int, S_IRUGO);
+module_param(disable_msi, int, 0444);
 MODULE_PARM_DESC(disable_msi, "Disable Message Signaled Interrupt (MSI)");
 
 typedef enum {
@@ -182,12 +175,12 @@ static const struct flash_spec flash_table[] =
 	{0x19000002, 0x5b808201, 0x000500db, 0x03840253, 0xaf020406,
 	 NONBUFFERED_FLAGS, ST_MICRO_FLASH_PAGE_BITS, ST_MICRO_FLASH_PAGE_SIZE,
 	 ST_MICRO_FLASH_BYTE_ADDR_MASK, ST_MICRO_FLASH_BASE_TOTAL_SIZE*2,
-	 "Entry 0101: ST M45PE10 (128kB non-bufferred)"},
+	 "Entry 0101: ST M45PE10 (128kB non-buffered)"},
 	/* Entry 0110: ST M45PE20 (non-buffered flash)*/
 	{0x15000001, 0x57808201, 0x000500db, 0x03840253, 0xaf020406,
 	 NONBUFFERED_FLAGS, ST_MICRO_FLASH_PAGE_BITS, ST_MICRO_FLASH_PAGE_SIZE,
 	 ST_MICRO_FLASH_BYTE_ADDR_MASK, ST_MICRO_FLASH_BASE_TOTAL_SIZE*4,
-	 "Entry 0110: ST M45PE20 (256kB non-bufferred)"},
+	 "Entry 0110: ST M45PE20 (256kB non-buffered)"},
 	/* Saifun SA25F005 (non-buffered flash) */
 	/* strap, cfg1, & write1 need updates */
 	{0x1d000003, 0x5f808201, 0x00050081, 0x03840253, 0xaf020406,
@@ -374,6 +367,7 @@ static void bnx2_setup_cnic_irq_info(struct bnx2 *bp)
 	cp->irq_arr[0].status_blk = (void *)
 		((unsigned long) bnapi->status_blk.msi +
 		(BNX2_SBLK_MSIX_ALIGN_SIZE * sb_id));
+	cp->irq_arr[0].status_blk_map = bp->status_blk_mapping;
 	cp->irq_arr[0].status_blk_num = sb_id;
 	cp->num_irq = 1;
 }
@@ -384,7 +378,7 @@ static int bnx2_register_cnic(struct net_device *dev, struct cnic_ops *ops,
 	struct bnx2 *bp = netdev_priv(dev);
 	struct cnic_eth_dev *cp = &bp->cnic_eth_dev;
 
-	if (ops == NULL)
+	if (!ops)
 		return -EINVAL;
 
 	if (cp->drv_state & CNIC_DRV_STATE_REGD)
@@ -578,7 +572,7 @@ bnx2_write_phy(struct bnx2 *bp, u32 reg, u32 val)
 	}
 
 	if (val1 & BNX2_EMAC_MDIO_COMM_START_BUSY)
-        	ret = -EBUSY;
+		ret = -EBUSY;
 	else
 		ret = 0;
 
@@ -755,13 +749,13 @@ bnx2_alloc_tx_mem(struct bnx2 *bp)
 		struct bnx2_tx_ring_info *txr = &bnapi->tx_ring;
 
 		txr->tx_buf_ring = kzalloc(SW_TXBD_RING_SIZE, GFP_KERNEL);
-		if (txr->tx_buf_ring == NULL)
+		if (!txr->tx_buf_ring)
 			return -ENOMEM;
 
 		txr->tx_desc_ring =
 			dma_alloc_coherent(&bp->pdev->dev, TXBD_RING_SIZE,
 					   &txr->tx_desc_mapping, GFP_KERNEL);
-		if (txr->tx_desc_ring == NULL)
+		if (!txr->tx_desc_ring)
 			return -ENOMEM;
 	}
 	return 0;
@@ -778,8 +772,8 @@ bnx2_alloc_rx_mem(struct bnx2 *bp)
 		int j;
 
 		rxr->rx_buf_ring =
-			vzalloc(SW_RXBD_RING_SIZE * bp->rx_max_ring);
-		if (rxr->rx_buf_ring == NULL)
+			vzalloc(array_size(SW_RXBD_RING_SIZE, bp->rx_max_ring));
+		if (!rxr->rx_buf_ring)
 			return -ENOMEM;
 
 		for (j = 0; j < bp->rx_max_ring; j++) {
@@ -788,15 +782,16 @@ bnx2_alloc_rx_mem(struct bnx2 *bp)
 						   RXBD_RING_SIZE,
 						   &rxr->rx_desc_mapping[j],
 						   GFP_KERNEL);
-			if (rxr->rx_desc_ring[j] == NULL)
+			if (!rxr->rx_desc_ring[j])
 				return -ENOMEM;
 
 		}
 
 		if (bp->rx_pg_ring_size) {
-			rxr->rx_pg_ring = vzalloc(SW_RXPG_RING_SIZE *
-						  bp->rx_max_pg_ring);
-			if (rxr->rx_pg_ring == NULL)
+			rxr->rx_pg_ring =
+				vzalloc(array_size(SW_RXPG_RING_SIZE,
+						   bp->rx_max_pg_ring));
+			if (!rxr->rx_pg_ring)
 				return -ENOMEM;
 
 		}
@@ -807,7 +802,7 @@ bnx2_alloc_rx_mem(struct bnx2 *bp)
 						   RXBD_RING_SIZE,
 						   &rxr->rx_pg_desc_mapping[j],
 						   GFP_KERNEL);
-			if (rxr->rx_pg_desc_ring[j] == NULL)
+			if (!rxr->rx_pg_desc_ring[j])
 				return -ENOMEM;
 
 		}
@@ -843,9 +838,9 @@ bnx2_alloc_stats_blk(struct net_device *dev)
 						 BNX2_SBLK_MSIX_ALIGN_SIZE);
 	bp->status_stats_size = status_blk_size +
 				sizeof(struct statistics_block);
-	status_blk = dma_zalloc_coherent(&bp->pdev->dev, bp->status_stats_size,
-					 &bp->status_blk_mapping, GFP_KERNEL);
-	if (status_blk == NULL)
+	status_blk = dma_alloc_coherent(&bp->pdev->dev, bp->status_stats_size,
+					&bp->status_blk_mapping, GFP_KERNEL);
+	if (!status_blk)
 		return -ENOMEM;
 
 	bp->status_blk = status_blk;
@@ -914,7 +909,7 @@ bnx2_alloc_mem(struct bnx2 *bp)
 						BNX2_PAGE_SIZE,
 						&bp->ctx_blk_mapping[i],
 						GFP_KERNEL);
-			if (bp->ctx_blk[i] == NULL)
+			if (!bp->ctx_blk[i])
 				goto alloc_mem_err;
 		}
 	}
@@ -1342,13 +1337,13 @@ bnx2_set_mac_link(struct bnx2 *bp)
 					val |= BNX2_EMAC_MODE_PORT_MII_10M;
 					break;
 				}
-				/* fall through */
+				fallthrough;
 			case SPEED_100:
 				val |= BNX2_EMAC_MODE_PORT_MII;
 				break;
 			case SPEED_2500:
 				val |= BNX2_EMAC_MODE_25G_MODE;
-				/* fall through */
+				fallthrough;
 			case SPEED_1000:
 				val |= BNX2_EMAC_MODE_PORT_GMII;
 				break;
@@ -1460,7 +1455,7 @@ bnx2_test_and_disable_2g5(struct bnx2 *bp)
 static void
 bnx2_enable_forced_2g5(struct bnx2 *bp)
 {
-	u32 uninitialized_var(bmcr);
+	u32 bmcr;
 	int err;
 
 	if (!(bp->phy_flags & BNX2_PHY_FLAG_2_5G_CAPABLE))
@@ -1504,7 +1499,7 @@ bnx2_enable_forced_2g5(struct bnx2 *bp)
 static void
 bnx2_disable_forced_2g5(struct bnx2 *bp)
 {
-	u32 uninitialized_var(bmcr);
+	u32 bmcr;
 	int err;
 
 	if (!(bp->phy_flags & BNX2_PHY_FLAG_2_5G_CAPABLE))
@@ -2000,26 +1995,26 @@ bnx2_remote_phy_event(struct bnx2 *bp)
 		switch (speed) {
 			case BNX2_LINK_STATUS_10HALF:
 				bp->duplex = DUPLEX_HALF;
-				/* fall through */
+				fallthrough;
 			case BNX2_LINK_STATUS_10FULL:
 				bp->line_speed = SPEED_10;
 				break;
 			case BNX2_LINK_STATUS_100HALF:
 				bp->duplex = DUPLEX_HALF;
-				/* fall through */
+				fallthrough;
 			case BNX2_LINK_STATUS_100BASE_T4:
 			case BNX2_LINK_STATUS_100FULL:
 				bp->line_speed = SPEED_100;
 				break;
 			case BNX2_LINK_STATUS_1000HALF:
 				bp->duplex = DUPLEX_HALF;
-				/* fall through */
+				fallthrough;
 			case BNX2_LINK_STATUS_1000FULL:
 				bp->line_speed = SPEED_1000;
 				break;
 			case BNX2_LINK_STATUS_2500HALF:
 				bp->duplex = DUPLEX_HALF;
-				/* fall through */
+				fallthrough;
 			case BNX2_LINK_STATUS_2500FULL:
 				bp->line_speed = SPEED_2500;
 				break;
@@ -2666,8 +2661,8 @@ bnx2_alloc_bad_rbuf(struct bnx2 *bp)
 	u32 good_mbuf_cnt;
 	u32 val;
 
-	good_mbuf = kmalloc(512 * sizeof(u16), GFP_KERNEL);
-	if (good_mbuf == NULL)
+	good_mbuf = kmalloc_array(512, sizeof(u16), GFP_KERNEL);
+	if (!good_mbuf)
 		return -ENOMEM;
 
 	BNX2_WR(bp, BNX2_MISC_ENABLE_SET_BITS,
@@ -2709,7 +2704,7 @@ bnx2_alloc_bad_rbuf(struct bnx2 *bp)
 }
 
 static void
-bnx2_set_mac_addr(struct bnx2 *bp, u8 *mac_addr, u32 pos)
+bnx2_set_mac_addr(struct bnx2 *bp, const u8 *mac_addr, u32 pos)
 {
 	u32 val;
 
@@ -2735,7 +2730,7 @@ bnx2_alloc_rx_page(struct bnx2 *bp, struct bnx2_rx_ring_info *rxr, u16 index, gf
 	if (!page)
 		return -ENOMEM;
 	mapping = dma_map_page(&bp->pdev->dev, page, 0, PAGE_SIZE,
-			       PCI_DMA_FROMDEVICE);
+			       DMA_FROM_DEVICE);
 	if (dma_mapping_error(&bp->pdev->dev, mapping)) {
 		__free_page(page);
 		return -EIO;
@@ -2758,7 +2753,7 @@ bnx2_free_rx_page(struct bnx2 *bp, struct bnx2_rx_ring_info *rxr, u16 index)
 		return;
 
 	dma_unmap_page(&bp->pdev->dev, dma_unmap_addr(rx_pg, mapping),
-		       PAGE_SIZE, PCI_DMA_FROMDEVICE);
+		       PAGE_SIZE, DMA_FROM_DEVICE);
 
 	__free_page(page);
 	rx_pg->page = NULL;
@@ -2780,7 +2775,7 @@ bnx2_alloc_rx_data(struct bnx2 *bp, struct bnx2_rx_ring_info *rxr, u16 index, gf
 	mapping = dma_map_single(&bp->pdev->dev,
 				 get_l2_fhdr(data),
 				 bp->rx_buf_use_size,
-				 PCI_DMA_FROMDEVICE);
+				 DMA_FROM_DEVICE);
 	if (dma_mapping_error(&bp->pdev->dev, mapping)) {
 		kfree(data);
 		return -EIO;
@@ -2886,7 +2881,7 @@ bnx2_tx_int(struct bnx2 *bp, struct bnx2_napi *bnapi, int budget)
 		}
 
 		dma_unmap_single(&bp->pdev->dev, dma_unmap_addr(tx_buf, mapping),
-			skb_headlen(skb), PCI_DMA_TODEVICE);
+			skb_headlen(skb), DMA_TO_DEVICE);
 
 		tx_buf->skb = NULL;
 		last = tx_buf->nr_frags;
@@ -2900,7 +2895,7 @@ bnx2_tx_int(struct bnx2 *bp, struct bnx2_napi *bnapi, int budget)
 			dma_unmap_page(&bp->pdev->dev,
 				dma_unmap_addr(tx_buf, mapping),
 				skb_frag_size(&skb_shinfo(skb)->frags[i]),
-				PCI_DMA_TODEVICE);
+				DMA_TO_DEVICE);
 		}
 
 		sw_cons = BNX2_NEXT_TX_BD(sw_cons);
@@ -2961,7 +2956,6 @@ bnx2_reuse_rx_skb_pages(struct bnx2 *bp, struct bnx2_rx_ring_info *rxr,
 		shinfo = skb_shinfo(skb);
 		shinfo->nr_frags--;
 		page = skb_frag_page(&shinfo->frags[shinfo->nr_frags]);
-		__skb_frag_set_page(&shinfo->frags[shinfo->nr_frags], NULL);
 
 		cons_rx_pg->page = page;
 		dev_kfree_skb(skb);
@@ -3008,7 +3002,7 @@ bnx2_reuse_rx_data(struct bnx2 *bp, struct bnx2_rx_ring_info *rxr,
 
 	dma_sync_single_for_device(&bp->pdev->dev,
 		dma_unmap_addr(cons_rx_buf, mapping),
-		BNX2_RX_OFFSET + BNX2_RX_COPY_THRESH, PCI_DMA_FROMDEVICE);
+		BNX2_RX_OFFSET + BNX2_RX_COPY_THRESH, DMA_FROM_DEVICE);
 
 	rxr->rx_prod_bseq += bp->rx_buf_use_size;
 
@@ -3049,8 +3043,8 @@ error:
 	}
 
 	dma_unmap_single(&bp->pdev->dev, dma_addr, bp->rx_buf_use_size,
-			 PCI_DMA_FROMDEVICE);
-	skb = build_skb(data, 0);
+			 DMA_FROM_DEVICE);
+	skb = slab_build_skb(data);
 	if (!skb) {
 		kfree(data);
 		goto error;
@@ -3115,7 +3109,7 @@ error:
 			}
 
 			dma_unmap_page(&bp->pdev->dev, mapping_old,
-				       PAGE_SIZE, PCI_DMA_FROMDEVICE);
+				       PAGE_SIZE, DMA_FROM_DEVICE);
 
 			frag_size -= frag_len;
 			skb->data_len += frag_len;
@@ -3185,7 +3179,7 @@ bnx2_rx_int(struct bnx2 *bp, struct bnx2_napi *bnapi, int budget)
 
 		dma_sync_single_for_cpu(&bp->pdev->dev, dma_addr,
 			BNX2_RX_OFFSET + BNX2_RX_COPY_THRESH,
-			PCI_DMA_FROMDEVICE);
+			DMA_FROM_DEVICE);
 
 		next_ring_idx = BNX2_RX_RING_IDX(BNX2_NEXT_RX_BD(sw_cons));
 		next_rx_buf = &rxr->rx_buf_ring[next_ring_idx];
@@ -3225,7 +3219,7 @@ bnx2_rx_int(struct bnx2 *bp, struct bnx2_napi *bnapi, int budget)
 
 		if (len <= bp->rx_copy_thresh) {
 			skb = netdev_alloc_skb(bp->dev, len + 6);
-			if (skb == NULL) {
+			if (!skb) {
 				bnx2_reuse_rx_data(bp, rxr, data, sw_ring_cons,
 						  sw_ring_prod);
 				goto next_rx;
@@ -3285,7 +3279,7 @@ next_rx:
 		sw_cons = BNX2_NEXT_RX_BD(sw_cons);
 		sw_prod = BNX2_NEXT_RX_BD(sw_prod);
 
-		if ((rx_pkt == budget))
+		if (rx_pkt == budget)
 			break;
 
 		/* Refresh hw_cons to see if there is new work */
@@ -3304,8 +3298,6 @@ next_rx:
 
 	BNX2_WR(bp, rxr->rx_bseq_addr, rxr->rx_prod_bseq);
 
-	mmiowb();
-
 	return rx_pkt;
 
 }
@@ -3606,7 +3598,7 @@ bnx2_set_rx_mode(struct net_device *dev)
 		for (i = 0; i < NUM_MC_HASH_REGISTERS; i++) {
 			BNX2_WR(bp, BNX2_EMAC_MULTICAST_HASH0 + (i * 4),
 				0xffffffff);
-        	}
+		}
 		sort_mode |= BNX2_RPM_SORT_USER0_MC_EN;
 	}
 	else {
@@ -3836,7 +3828,7 @@ load_rv2p_fw(struct bnx2 *bp, u32 rv2p_proc,
 	return 0;
 }
 
-static int
+static void
 load_cpu_fw(struct bnx2 *bp, const struct cpu_reg *cpu_reg,
 	    const struct bnx2_mips_fw_file_entry *fw_entry)
 {
@@ -3904,48 +3896,34 @@ load_cpu_fw(struct bnx2 *bp, const struct cpu_reg *cpu_reg,
 	val &= ~cpu_reg->mode_value_halt;
 	bnx2_reg_wr_ind(bp, cpu_reg->state, cpu_reg->state_value_clear);
 	bnx2_reg_wr_ind(bp, cpu_reg->mode, val);
-
-	return 0;
 }
 
-static int
+static void
 bnx2_init_cpus(struct bnx2 *bp)
 {
 	const struct bnx2_mips_fw_file *mips_fw =
 		(const struct bnx2_mips_fw_file *) bp->mips_firmware->data;
 	const struct bnx2_rv2p_fw_file *rv2p_fw =
 		(const struct bnx2_rv2p_fw_file *) bp->rv2p_firmware->data;
-	int rc;
 
 	/* Initialize the RV2P processor. */
 	load_rv2p_fw(bp, RV2P_PROC1, &rv2p_fw->proc1);
 	load_rv2p_fw(bp, RV2P_PROC2, &rv2p_fw->proc2);
 
 	/* Initialize the RX Processor. */
-	rc = load_cpu_fw(bp, &cpu_reg_rxp, &mips_fw->rxp);
-	if (rc)
-		goto init_cpu_err;
+	load_cpu_fw(bp, &cpu_reg_rxp, &mips_fw->rxp);
 
 	/* Initialize the TX Processor. */
-	rc = load_cpu_fw(bp, &cpu_reg_txp, &mips_fw->txp);
-	if (rc)
-		goto init_cpu_err;
+	load_cpu_fw(bp, &cpu_reg_txp, &mips_fw->txp);
 
 	/* Initialize the TX Patch-up Processor. */
-	rc = load_cpu_fw(bp, &cpu_reg_tpat, &mips_fw->tpat);
-	if (rc)
-		goto init_cpu_err;
+	load_cpu_fw(bp, &cpu_reg_tpat, &mips_fw->tpat);
 
 	/* Initialize the Completion Processor. */
-	rc = load_cpu_fw(bp, &cpu_reg_com, &mips_fw->com);
-	if (rc)
-		goto init_cpu_err;
+	load_cpu_fw(bp, &cpu_reg_com, &mips_fw->com);
 
 	/* Initialize the Command Processor. */
-	rc = load_cpu_fw(bp, &cpu_reg_cp, &mips_fw->cp);
-
-init_cpu_err:
-	return rc;
+	load_cpu_fw(bp, &cpu_reg_cp, &mips_fw->cp);
 }
 
 static void
@@ -4561,7 +4539,7 @@ bnx2_nvram_write(struct bnx2 *bp, u32 offset, u8 *data_buf,
 
 	if (align_start || align_end) {
 		align_buf = kmalloc(len32, GFP_KERNEL);
-		if (align_buf == NULL)
+		if (!align_buf)
 			return -ENOMEM;
 		if (align_start) {
 			memcpy(align_buf, start, 4);
@@ -4575,7 +4553,7 @@ bnx2_nvram_write(struct bnx2 *bp, u32 offset, u8 *data_buf,
 
 	if (!(bp->flash_info->flags & BNX2_NV_BUFFERED)) {
 		flash_buffer = kmalloc(264, GFP_KERNEL);
-		if (flash_buffer == NULL) {
+		if (!flash_buffer) {
 			rc = -ENOMEM;
 			goto nvram_write_end;
 		}
@@ -4681,7 +4659,7 @@ bnx2_nvram_write(struct bnx2 *bp, u32 offset, u8 *data_buf,
 
 				if (addr == page_end-4) {
 					cmd_flags = BNX2_NVM_COMMAND_LAST;
-                		}
+				}
 				rc = bnx2_nvram_write_dword(bp, addr,
 					&flash_buffer[i], cmd_flags);
 
@@ -4958,8 +4936,7 @@ bnx2_init_chip(struct bnx2 *bp)
 	} else
 		bnx2_init_context(bp);
 
-	if ((rc = bnx2_init_cpus(bp)) != 0)
-		return rc;
+	bnx2_init_cpus(bp);
 
 	bnx2_init_nvram(bp);
 
@@ -5422,8 +5399,9 @@ bnx2_set_rx_ring_size(struct bnx2 *bp, u32 size)
 
 	bp->rx_buf_use_size = rx_size;
 	/* hw alignment + build_skb() overhead*/
-	bp->rx_buf_size = SKB_DATA_ALIGN(bp->rx_buf_use_size + BNX2_RX_ALIGN) +
-		NET_SKB_PAD + SKB_DATA_ALIGN(sizeof(struct skb_shared_info));
+	bp->rx_buf_size = kmalloc_size_roundup(
+		SKB_DATA_ALIGN(bp->rx_buf_use_size + BNX2_RX_ALIGN) +
+		NET_SKB_PAD + SKB_DATA_ALIGN(sizeof(struct skb_shared_info)));
 	bp->rx_jumbo_thresh = rx_size - BNX2_RX_OFFSET;
 	bp->rx_ring_size = size;
 	bp->rx_max_ring = bnx2_find_max_ring(size, BNX2_MAX_RX_RINGS);
@@ -5440,7 +5418,7 @@ bnx2_free_tx_skbs(struct bnx2 *bp)
 		struct bnx2_tx_ring_info *txr = &bnapi->tx_ring;
 		int j;
 
-		if (txr->tx_buf_ring == NULL)
+		if (!txr->tx_buf_ring)
 			continue;
 
 		for (j = 0; j < BNX2_TX_DESC_CNT; ) {
@@ -5448,7 +5426,7 @@ bnx2_free_tx_skbs(struct bnx2 *bp)
 			struct sk_buff *skb = tx_buf->skb;
 			int k, last;
 
-			if (skb == NULL) {
+			if (!skb) {
 				j = BNX2_NEXT_TX_BD(j);
 				continue;
 			}
@@ -5456,7 +5434,7 @@ bnx2_free_tx_skbs(struct bnx2 *bp)
 			dma_unmap_single(&bp->pdev->dev,
 					 dma_unmap_addr(tx_buf, mapping),
 					 skb_headlen(skb),
-					 PCI_DMA_TODEVICE);
+					 DMA_TO_DEVICE);
 
 			tx_buf->skb = NULL;
 
@@ -5467,7 +5445,7 @@ bnx2_free_tx_skbs(struct bnx2 *bp)
 				dma_unmap_page(&bp->pdev->dev,
 					dma_unmap_addr(tx_buf, mapping),
 					skb_frag_size(&skb_shinfo(skb)->frags[k]),
-					PCI_DMA_TODEVICE);
+					DMA_TO_DEVICE);
 			}
 			dev_kfree_skb(skb);
 		}
@@ -5485,20 +5463,20 @@ bnx2_free_rx_skbs(struct bnx2 *bp)
 		struct bnx2_rx_ring_info *rxr = &bnapi->rx_ring;
 		int j;
 
-		if (rxr->rx_buf_ring == NULL)
+		if (!rxr->rx_buf_ring)
 			return;
 
 		for (j = 0; j < bp->rx_max_ring_idx; j++) {
 			struct bnx2_sw_bd *rx_buf = &rxr->rx_buf_ring[j];
 			u8 *data = rx_buf->data;
 
-			if (data == NULL)
+			if (!data)
 				continue;
 
 			dma_unmap_single(&bp->pdev->dev,
 					 dma_unmap_addr(rx_buf, mapping),
 					 bp->rx_buf_use_size,
-					 PCI_DMA_FROMDEVICE);
+					 DMA_FROM_DEVICE);
 
 			rx_buf->data = NULL;
 
@@ -5818,8 +5796,8 @@ bnx2_run_loopback(struct bnx2 *bp, int loopback_mode)
 	struct l2_fhdr *rx_hdr;
 	int ret = -ENODEV;
 	struct bnx2_napi *bnapi = &bp->bnx2_napi[0], *tx_napi;
-	struct bnx2_tx_ring_info *txr = &bnapi->tx_ring;
-	struct bnx2_rx_ring_info *rxr = &bnapi->rx_ring;
+	struct bnx2_tx_ring_info *txr;
+	struct bnx2_rx_ring_info *rxr;
 
 	tx_napi = bnapi;
 
@@ -5850,7 +5828,7 @@ bnx2_run_loopback(struct bnx2 *bp, int loopback_mode)
 		packet[i] = (unsigned char) (i & 0xff);
 
 	map = dma_map_single(&bp->pdev->dev, skb->data, pkt_size,
-			     PCI_DMA_TODEVICE);
+			     DMA_TO_DEVICE);
 	if (dma_mapping_error(&bp->pdev->dev, map)) {
 		dev_kfree_skb(skb);
 		return -EIO;
@@ -5889,7 +5867,7 @@ bnx2_run_loopback(struct bnx2 *bp, int loopback_mode)
 
 	udelay(5);
 
-	dma_unmap_single(&bp->pdev->dev, map, pkt_size, PCI_DMA_TODEVICE);
+	dma_unmap_single(&bp->pdev->dev, map, pkt_size, DMA_TO_DEVICE);
 	dev_kfree_skb(skb);
 
 	if (bnx2_get_hw_tx_cons(tx_napi) != txr->tx_prod)
@@ -5908,7 +5886,7 @@ bnx2_run_loopback(struct bnx2 *bp, int loopback_mode)
 
 	dma_sync_single_for_cpu(&bp->pdev->dev,
 		dma_unmap_addr(rx_buf, mapping),
-		bp->rx_buf_use_size, PCI_DMA_FROMDEVICE);
+		bp->rx_buf_use_size, DMA_FROM_DEVICE);
 
 	if (rx_hdr->l2_fhdr_status &
 		(L2_FHDR_ERRORS_BAD_CRC |
@@ -6183,9 +6161,9 @@ bnx2_5708_serdes_timer(struct bnx2 *bp)
 }
 
 static void
-bnx2_timer(unsigned long data)
+bnx2_timer(struct timer_list *t)
 {
-	struct bnx2 *bp = (struct bnx2 *) data;
+	struct bnx2 *bp = timer_container_of(bp, t, timer);
 
 	if (!netif_running(bp->dev))
 		return;
@@ -6422,7 +6400,7 @@ bnx2_open(struct net_device *dev)
 				rc = bnx2_request_irq(bp);
 
 			if (rc) {
-				del_timer_sync(&bp->timer);
+				timer_delete_sync(&bp->timer);
 				goto open_err;
 			}
 			bnx2_enable_int(bp);
@@ -6576,7 +6554,7 @@ bnx2_dump_state(struct bnx2 *bp)
 }
 
 static void
-bnx2_tx_timeout(struct net_device *dev)
+bnx2_tx_timeout(struct net_device *dev, unsigned int txqueue)
 {
 	struct bnx2 *bp = netdev_priv(dev);
 
@@ -6667,7 +6645,8 @@ bnx2_start_xmit(struct sk_buff *skb, struct net_device *dev)
 	} else
 		mss = 0;
 
-	mapping = dma_map_single(&bp->pdev->dev, skb->data, len, PCI_DMA_TODEVICE);
+	mapping = dma_map_single(&bp->pdev->dev, skb->data, len,
+				 DMA_TO_DEVICE);
 	if (dma_mapping_error(&bp->pdev->dev, mapping)) {
 		dev_kfree_skb_any(skb);
 		return NETDEV_TX_OK;
@@ -6722,8 +6701,6 @@ bnx2_start_xmit(struct sk_buff *skb, struct net_device *dev)
 	BNX2_WR16(bp, txr->tx_bidx_addr, prod);
 	BNX2_WR(bp, txr->tx_bseq_addr, txr->tx_prod_bseq);
 
-	mmiowb();
-
 	txr->tx_prod = prod;
 
 	if (unlikely(bnx2_tx_avail(bp, txr) <= MAX_SKB_FRAGS)) {
@@ -6750,7 +6727,7 @@ dma_error:
 	tx_buf = &txr->tx_buf_ring[ring_prod];
 	tx_buf->skb = NULL;
 	dma_unmap_single(&bp->pdev->dev, dma_unmap_addr(tx_buf, mapping),
-			 skb_headlen(skb), PCI_DMA_TODEVICE);
+			 skb_headlen(skb), DMA_TO_DEVICE);
 
 	/* unmap remaining mapped pages */
 	for (i = 0; i < last_frag; i++) {
@@ -6759,7 +6736,7 @@ dma_error:
 		tx_buf = &txr->tx_buf_ring[ring_prod];
 		dma_unmap_page(&bp->pdev->dev, dma_unmap_addr(tx_buf, mapping),
 			       skb_frag_size(&skb_shinfo(skb)->frags[i]),
-			       PCI_DMA_TODEVICE);
+			       DMA_TO_DEVICE);
 	}
 
 	dev_kfree_skb_any(skb);
@@ -6775,7 +6752,7 @@ bnx2_close(struct net_device *dev)
 	bnx2_disable_int_sync(bp);
 	bnx2_napi_disable(bp);
 	netif_tx_disable(dev);
-	del_timer_sync(&bp->timer);
+	timer_delete_sync(&bp->timer);
 	bnx2_shutdown_chip(bp);
 	bnx2_free_irq(bp);
 	bnx2_free_skbs(bp);
@@ -6826,7 +6803,7 @@ bnx2_get_stats64(struct net_device *dev, struct rtnl_link_stats64 *net_stats)
 {
 	struct bnx2 *bp = netdev_priv(dev);
 
-	if (bp->stats_blk == NULL)
+	if (!bp->stats_blk)
 		return;
 
 	net_stats->rx_packets =
@@ -7050,10 +7027,9 @@ bnx2_get_drvinfo(struct net_device *dev, struct ethtool_drvinfo *info)
 {
 	struct bnx2 *bp = netdev_priv(dev);
 
-	strlcpy(info->driver, DRV_MODULE_NAME, sizeof(info->driver));
-	strlcpy(info->version, DRV_MODULE_VERSION, sizeof(info->version));
-	strlcpy(info->bus_info, pci_name(bp->pdev), sizeof(info->bus_info));
-	strlcpy(info->fw_version, bp->fw_version, sizeof(info->fw_version));
+	strscpy(info->driver, DRV_MODULE_NAME, sizeof(info->driver));
+	strscpy(info->bus_info, pci_name(bp->pdev), sizeof(info->bus_info));
+	strscpy(info->fw_version, bp->fw_version, sizeof(info->fw_version));
 }
 
 #define BNX2_REGDUMP_LEN		(32 * 1024)
@@ -7217,7 +7193,7 @@ bnx2_get_eeprom_len(struct net_device *dev)
 {
 	struct bnx2 *bp = netdev_priv(dev);
 
-	if (bp->flash_info == NULL)
+	if (!bp->flash_info)
 		return 0;
 
 	return (int) bp->flash_size;
@@ -7251,8 +7227,10 @@ bnx2_set_eeprom(struct net_device *dev, struct ethtool_eeprom *eeprom,
 	return rc;
 }
 
-static int
-bnx2_get_coalesce(struct net_device *dev, struct ethtool_coalesce *coal)
+static int bnx2_get_coalesce(struct net_device *dev,
+			     struct ethtool_coalesce *coal,
+			     struct kernel_ethtool_coalesce *kernel_coal,
+			     struct netlink_ext_ack *extack)
 {
 	struct bnx2 *bp = netdev_priv(dev);
 
@@ -7273,8 +7251,10 @@ bnx2_get_coalesce(struct net_device *dev, struct ethtool_coalesce *coal)
 	return 0;
 }
 
-static int
-bnx2_set_coalesce(struct net_device *dev, struct ethtool_coalesce *coal)
+static int bnx2_set_coalesce(struct net_device *dev,
+			     struct ethtool_coalesce *coal,
+			     struct kernel_ethtool_coalesce *kernel_coal,
+			     struct netlink_ext_ack *extack)
 {
 	struct bnx2 *bp = netdev_priv(dev);
 
@@ -7323,7 +7303,9 @@ bnx2_set_coalesce(struct net_device *dev, struct ethtool_coalesce *coal)
 }
 
 static void
-bnx2_get_ringparam(struct net_device *dev, struct ethtool_ringparam *ering)
+bnx2_get_ringparam(struct net_device *dev, struct ethtool_ringparam *ering,
+		   struct kernel_ethtool_ringparam *kernel_ering,
+		   struct netlink_ext_ack *extack)
 {
 	struct bnx2 *bp = netdev_priv(dev);
 
@@ -7394,7 +7376,9 @@ bnx2_change_ring_size(struct bnx2 *bp, u32 rx, u32 tx, bool reset_irq)
 }
 
 static int
-bnx2_set_ringparam(struct net_device *dev, struct ethtool_ringparam *ering)
+bnx2_set_ringparam(struct net_device *dev, struct ethtool_ringparam *ering,
+		   struct kernel_ethtool_ringparam *kernel_ering,
+		   struct netlink_ext_ack *extack)
 {
 	struct bnx2 *bp = netdev_priv(dev);
 	int rc;
@@ -7678,7 +7662,7 @@ bnx2_get_ethtool_stats(struct net_device *dev,
 	u32 *temp_stats = (u32 *) bp->temp_stats_blk;
 	u8 *stats_len_arr = NULL;
 
-	if (hw_stats == NULL) {
+	if (!hw_stats) {
 		memset(buf, 0, sizeof(u64) * BNX2_NUM_STATS);
 		return;
 	}
@@ -7822,6 +7806,11 @@ static int bnx2_set_channels(struct net_device *dev,
 }
 
 static const struct ethtool_ops bnx2_ethtool_ops = {
+	.supported_coalesce_params = ETHTOOL_COALESCE_USECS |
+				     ETHTOOL_COALESCE_MAX_FRAMES |
+				     ETHTOOL_COALESCE_USECS_IRQ |
+				     ETHTOOL_COALESCE_MAX_FRAMES_IRQ |
+				     ETHTOOL_COALESCE_STATS_BLOCK_USECS,
 	.get_drvinfo		= bnx2_get_drvinfo,
 	.get_regs_len		= bnx2_get_regs_len,
 	.get_regs		= bnx2_get_regs,
@@ -7861,7 +7850,7 @@ bnx2_ioctl(struct net_device *dev, struct ifreq *ifr, int cmd)
 	case SIOCGMIIPHY:
 		data->phy_id = bp->phy_addr;
 
-		/* fallthru */
+		fallthrough;
 	case SIOCGMIIREG: {
 		u32 mii_regval;
 
@@ -7910,7 +7899,7 @@ bnx2_change_mac_addr(struct net_device *dev, void *p)
 	if (!is_valid_ether_addr(addr->sa_data))
 		return -EADDRNOTAVAIL;
 
-	memcpy(dev->dev_addr, addr->sa_data, dev->addr_len);
+	eth_hw_addr_set(dev, addr->sa_data);
 	if (netif_running(dev))
 		bnx2_set_mac_addr(bp, bp->dev->dev_addr, 0);
 
@@ -7923,7 +7912,7 @@ bnx2_change_mtu(struct net_device *dev, int new_mtu)
 {
 	struct bnx2 *bp = netdev_priv(dev);
 
-	dev->mtu = new_mtu;
+	WRITE_ONCE(dev->mtu, new_mtu);
 	return bnx2_change_ring_size(bp, bp->rx_ring_size, bp->tx_ring_size,
 				     false);
 }
@@ -8038,62 +8027,40 @@ bnx2_get_pci_speed(struct bnx2 *bp)
 static void
 bnx2_read_vpd_fw_ver(struct bnx2 *bp)
 {
+	unsigned int len;
 	int rc, i, j;
 	u8 *data;
-	unsigned int block_end, rosize, len;
 
 #define BNX2_VPD_NVRAM_OFFSET	0x300
 #define BNX2_VPD_LEN		128
 #define BNX2_MAX_VER_SLEN	30
 
-	data = kmalloc(256, GFP_KERNEL);
+	data = kmalloc(BNX2_VPD_LEN, GFP_KERNEL);
 	if (!data)
 		return;
 
-	rc = bnx2_nvram_read(bp, BNX2_VPD_NVRAM_OFFSET, data + BNX2_VPD_LEN,
-			     BNX2_VPD_LEN);
+	rc = bnx2_nvram_read(bp, BNX2_VPD_NVRAM_OFFSET, data, BNX2_VPD_LEN);
 	if (rc)
 		goto vpd_done;
 
-	for (i = 0; i < BNX2_VPD_LEN; i += 4) {
-		data[i] = data[i + BNX2_VPD_LEN + 3];
-		data[i + 1] = data[i + BNX2_VPD_LEN + 2];
-		data[i + 2] = data[i + BNX2_VPD_LEN + 1];
-		data[i + 3] = data[i + BNX2_VPD_LEN];
-	}
+	for (i = 0; i < BNX2_VPD_LEN; i += 4)
+		swab32s((u32 *)&data[i]);
 
-	i = pci_vpd_find_tag(data, 0, BNX2_VPD_LEN, PCI_VPD_LRDT_RO_DATA);
-	if (i < 0)
-		goto vpd_done;
-
-	rosize = pci_vpd_lrdt_size(&data[i]);
-	i += PCI_VPD_LRDT_TAG_SIZE;
-	block_end = i + rosize;
-
-	if (block_end > BNX2_VPD_LEN)
-		goto vpd_done;
-
-	j = pci_vpd_find_info_keyword(data, i, rosize,
-				      PCI_VPD_RO_KEYWORD_MFR_ID);
+	j = pci_vpd_find_ro_info_keyword(data, BNX2_VPD_LEN,
+					 PCI_VPD_RO_KEYWORD_MFR_ID, &len);
 	if (j < 0)
 		goto vpd_done;
 
-	len = pci_vpd_info_field_size(&data[j]);
-
-	j += PCI_VPD_INFO_FLD_HDR_SIZE;
-	if (j + len > block_end || len != 4 ||
-	    memcmp(&data[j], "1028", 4))
+	if (len != 4 || memcmp(&data[j], "1028", 4))
 		goto vpd_done;
 
-	j = pci_vpd_find_info_keyword(data, i, rosize,
-				      PCI_VPD_RO_KEYWORD_VENDOR0);
+	j = pci_vpd_find_ro_info_keyword(data, BNX2_VPD_LEN,
+					 PCI_VPD_RO_KEYWORD_VENDOR0,
+					 &len);
 	if (j < 0)
 		goto vpd_done;
 
-	len = pci_vpd_info_field_size(&data[j]);
-
-	j += PCI_VPD_INFO_FLD_HDR_SIZE;
-	if (j + len > block_end || len > BNX2_MAX_VER_SLEN)
+	if (len > BNX2_MAX_VER_SLEN)
 		goto vpd_done;
 
 	memcpy(bp->fw_version, &data[j], len);
@@ -8110,7 +8077,6 @@ bnx2_init_board(struct pci_dev *pdev, struct net_device *dev)
 	int rc, i, j;
 	u32 reg;
 	u64 dma_mask, persist_dma_mask;
-	int err;
 
 	SET_NETDEV_DEV(dev, &pdev->dev);
 	bp = netdev_priv(dev);
@@ -8121,7 +8087,7 @@ bnx2_init_board(struct pci_dev *pdev, struct net_device *dev)
 	bp->temp_stats_blk =
 		kzalloc(sizeof(struct statistics_block), GFP_KERNEL);
 
-	if (bp->temp_stats_blk == NULL) {
+	if (!bp->temp_stats_blk) {
 		rc = -ENOMEM;
 		goto err_out;
 	}
@@ -8193,12 +8159,6 @@ bnx2_init_board(struct pci_dev *pdev, struct net_device *dev)
 		bp->flags |= BNX2_FLAG_PCIE;
 		if (BNX2_CHIP_REV(bp) == BNX2_CHIP_REV_Ax)
 			bp->flags |= BNX2_FLAG_JUMBO_BROKEN;
-
-		/* AER (Advanced Error Reporting) hooks */
-		err = pci_enable_pcie_error_reporting(pdev);
-		if (!err)
-			bp->flags |= BNX2_FLAG_AER_ENABLED;
-
 	} else {
 		bp->pcix_cap = pci_find_capability(pdev, PCI_CAP_ID_PCIX);
 		if (bp->pcix_cap == 0) {
@@ -8229,15 +8189,15 @@ bnx2_init_board(struct pci_dev *pdev, struct net_device *dev)
 		persist_dma_mask = dma_mask = DMA_BIT_MASK(64);
 
 	/* Configure DMA attributes. */
-	if (pci_set_dma_mask(pdev, dma_mask) == 0) {
+	if (dma_set_mask(&pdev->dev, dma_mask) == 0) {
 		dev->features |= NETIF_F_HIGHDMA;
-		rc = pci_set_consistent_dma_mask(pdev, persist_dma_mask);
+		rc = dma_set_coherent_mask(&pdev->dev, persist_dma_mask);
 		if (rc) {
 			dev_err(&pdev->dev,
-				"pci_set_consistent_dma_mask failed, aborting\n");
+				"dma_set_coherent_mask failed, aborting\n");
 			goto err_out_unmap;
 		}
-	} else if ((rc = pci_set_dma_mask(pdev, DMA_BIT_MASK(32))) != 0) {
+	} else if ((rc = dma_set_mask(&pdev->dev, DMA_BIT_MASK(32))) != 0) {
 		dev_err(&pdev->dev, "System does not support DMA, aborting\n");
 		goto err_out_unmap;
 	}
@@ -8252,9 +8212,9 @@ bnx2_init_board(struct pci_dev *pdev, struct net_device *dev)
 		BNX2_WR(bp, PCI_COMMAND, reg);
 	} else if ((BNX2_CHIP_ID(bp) == BNX2_CHIP_ID_5706_A1) &&
 		!(bp->flags & BNX2_FLAG_PCIX)) {
-
 		dev_err(&pdev->dev,
 			"5706 A1 can only be used in a PCIX bus, aborting\n");
+		rc = -EPERM;
 		goto err_out_unmap;
 	}
 
@@ -8462,10 +8422,8 @@ bnx2_init_board(struct pci_dev *pdev, struct net_device *dev)
 	bnx2_set_default_link(bp);
 	bp->req_flow_ctrl = FLOW_CTRL_RX | FLOW_CTRL_TX;
 
-	init_timer(&bp->timer);
+	timer_setup(&bp->timer, bnx2_timer, 0);
 	bp->timer.expires = RUN_AT(BNX2_TIMER_INTERVAL);
-	bp->timer.data = (unsigned long) bp;
-	bp->timer.function = bnx2_timer;
 
 #ifdef BCM_CNIC
 	if (bnx2_shmem_rd(bp, BNX2_ISCSI_INITIATOR) & BNX2_ISCSI_INITIATOR_EN)
@@ -8479,11 +8437,6 @@ bnx2_init_board(struct pci_dev *pdev, struct net_device *dev)
 	return 0;
 
 err_out_unmap:
-	if (bp->flags & BNX2_FLAG_AER_ENABLED) {
-		pci_disable_pcie_error_reporting(pdev);
-		bp->flags &= ~BNX2_FLAG_AER_ENABLED;
-	}
-
 	pci_iounmap(pdev, bp->regview);
 	bp->regview = NULL;
 
@@ -8542,7 +8495,7 @@ bnx2_init_napi(struct bnx2 *bp)
 		else
 			poll = bnx2_poll_msix;
 
-		netif_napi_add(bp->dev, &bp->bnx2_napi[i].napi, poll, 64);
+		netif_napi_add(bp->dev, &bp->bnx2_napi[i].napi, poll);
 		bnapi->bp = bp;
 	}
 }
@@ -8553,7 +8506,7 @@ static const struct net_device_ops bnx2_netdev_ops = {
 	.ndo_stop		= bnx2_close,
 	.ndo_get_stats64	= bnx2_get_stats64,
 	.ndo_set_rx_mode	= bnx2_set_rx_mode,
-	.ndo_do_ioctl		= bnx2_ioctl,
+	.ndo_eth_ioctl		= bnx2_ioctl,
 	.ndo_validate_addr	= eth_validate_addr,
 	.ndo_set_mac_address	= bnx2_change_mac_addr,
 	.ndo_change_mtu		= bnx2_change_mtu,
@@ -8567,15 +8520,11 @@ static const struct net_device_ops bnx2_netdev_ops = {
 static int
 bnx2_init_one(struct pci_dev *pdev, const struct pci_device_id *ent)
 {
-	static int version_printed = 0;
 	struct net_device *dev;
 	struct bnx2 *bp;
 	int rc;
 	char str[40];
 
-	if (version_printed++ == 0)
-		pr_info("%s", version);
-
 	/* dev zeroed in init_etherdev */
 	dev = alloc_etherdev_mq(sizeof(*bp), TX_MAX_RINGS);
 	if (!dev)
@@ -8602,7 +8551,7 @@ bnx2_init_one(struct pci_dev *pdev, const struct pci_device_id *ent)
 	if (is_kdump_kernel())
 		bnx2_wait_dma_complete(bp);
 
-	memcpy(dev->dev_addr, bp->mac_addr, ETH_ALEN);
+	eth_hw_addr_set(dev, bp->mac_addr);
 
 	dev->hw_features = NETIF_F_IP_CSUM | NETIF_F_SG |
 		NETIF_F_TSO | NETIF_F_TSO_ECN |
@@ -8653,7 +8602,7 @@ bnx2_remove_one(struct pci_dev *pdev)
 
 	unregister_netdev(dev);
 
-	del_timer_sync(&bp->timer);
+	timer_delete_sync(&bp->timer);
 	cancel_work_sync(&bp->reset_task);
 
 	pci_iounmap(bp->pdev, bp->regview);
@@ -8661,11 +8610,6 @@ bnx2_remove_one(struct pci_dev *pdev)
 	bnx2_free_stats_blk(dev);
 	kfree(bp->temp_stats_blk);
 
-	if (bp->flags & BNX2_FLAG_AER_ENABLED) {
-		pci_disable_pcie_error_reporting(pdev);
-		bp->flags &= ~BNX2_FLAG_AER_ENABLED;
-	}
-
 	bnx2_release_firmware(bp);
 
 	free_netdev(dev);
@@ -8678,15 +8622,14 @@ bnx2_remove_one(struct pci_dev *pdev)
 static int
 bnx2_suspend(struct device *device)
 {
-	struct pci_dev *pdev = to_pci_dev(device);
-	struct net_device *dev = pci_get_drvdata(pdev);
+	struct net_device *dev = dev_get_drvdata(device);
 	struct bnx2 *bp = netdev_priv(dev);
 
 	if (netif_running(dev)) {
 		cancel_work_sync(&bp->reset_task);
 		bnx2_netif_stop(bp, true);
 		netif_device_detach(dev);
-		del_timer_sync(&bp->timer);
+		timer_delete_sync(&bp->timer);
 		bnx2_shutdown_chip(bp);
 		__bnx2_free_irq(bp);
 		bnx2_free_skbs(bp);
@@ -8698,8 +8641,7 @@ bnx2_suspend(struct device *device)
 static int
 bnx2_resume(struct device *device)
 {
-	struct pci_dev *pdev = to_pci_dev(device);
-	struct net_device *dev = pci_get_drvdata(pdev);
+	struct net_device *dev = dev_get_drvdata(device);
 	struct bnx2 *bp = netdev_priv(dev);
 
 	if (!netif_running(dev))
@@ -8745,7 +8687,7 @@ static pci_ers_result_t bnx2_io_error_detected(struct pci_dev *pdev,
 
 	if (netif_running(dev)) {
 		bnx2_netif_stop(bp, true);
-		del_timer_sync(&bp->timer);
+		timer_delete_sync(&bp->timer);
 		bnx2_reset_nic(bp, BNX2_DRV_MSG_CODE_RESET);
 	}
 
@@ -8791,16 +8733,6 @@ static pci_ers_result_t bnx2_io_slot_reset(struct pci_dev *pdev)
 	}
 	rtnl_unlock();
 
-	if (!(bp->flags & BNX2_FLAG_AER_ENABLED))
-		return result;
-
-	err = pci_cleanup_aer_uncorrect_error_status(pdev);
-	if (err) {
-		dev_err(&pdev->dev,
-			"pci_cleanup_aer_uncorrect_error_status failed 0x%0x\n",
-			 err); /* non-fatal, continue */
-	}
-
 	return result;
 }
 
diff --git a/drivers/net/ethernet/broadcom/bnx2.h b/drivers/net/ethernet/broadcom/bnx2.h
index a09ec47461c9..315b08c64edd 100644
--- a/drivers/net/ethernet/broadcom/bnx2.h
+++ b/drivers/net/ethernet/broadcom/bnx2.h
@@ -6808,7 +6808,6 @@ struct bnx2 {
 #define BNX2_FLAG_JUMBO_BROKEN		0x00000800
 #define BNX2_FLAG_CAN_KEEP_VLAN		0x00001000
 #define BNX2_FLAG_BROKEN_STATS		0x00002000
-#define BNX2_FLAG_AER_ENABLED		0x00004000
 
 	struct bnx2_napi	bnx2_napi[BNX2_MAX_MSIX_VEC];
 
diff --git a/drivers/net/ethernet/broadcom/bnx2x/Makefile b/drivers/net/ethernet/broadcom/bnx2x/Makefile
index 116762daae09..2523cfc7527d 100644
--- a/drivers/net/ethernet/broadcom/bnx2x/Makefile
+++ b/drivers/net/ethernet/broadcom/bnx2x/Makefile
@@ -1,8 +1,9 @@
+# SPDX-License-Identifier: GPL-2.0-only
 #
 # Makefile for Broadcom 10-Gigabit ethernet driver
 #
 
 obj-$(CONFIG_BNX2X) += bnx2x.o
 
-bnx2x-y := bnx2x_main.o bnx2x_link.o bnx2x_cmn.o bnx2x_ethtool.o bnx2x_stats.o bnx2x_dcb.o bnx2x_sp.o
+bnx2x-y := bnx2x_main.o bnx2x_link.o bnx2x_cmn.o bnx2x_ethtool.o bnx2x_stats.o bnx2x_dcb.o bnx2x_sp.o bnx2x_self_test.o
 bnx2x-$(CONFIG_BNX2X_SRIOV) += bnx2x_vfpf.o bnx2x_sriov.o
diff --git a/drivers/net/ethernet/broadcom/bnx2x/bnx2x.h b/drivers/net/ethernet/broadcom/bnx2x/bnx2x.h
index 352beff796ae..9580ab83d387 100644
--- a/drivers/net/ethernet/broadcom/bnx2x/bnx2x.h
+++ b/drivers/net/ethernet/broadcom/bnx2x/bnx2x.h
@@ -32,8 +32,14 @@
  * (you will need to reboot afterwards) */
 /* #define BNX2X_STOP_ON_ERROR */
 
-#define DRV_MODULE_VERSION      "1.712.30-0"
-#define DRV_MODULE_RELDATE      "2014/02/10"
+/* FIXME: Delete the DRV_MODULE_VERSION below, but please be warned
+ * that it is not an easy task because such change has all chances
+ * to break this driver due to amount of abuse of in-kernel interfaces
+ * between modules and FW.
+ *
+ * DO NOT UPDATE DRV_MODULE_VERSION below.
+ */
+#define DRV_MODULE_VERSION      "1.713.36-0"
 #define BNX2X_BC_VER            0x040200
 
 #if defined(CONFIG_DCB)
@@ -166,6 +172,12 @@ do {						\
 #define REG_RD8(bp, offset)		readb(REG_ADDR(bp, offset))
 #define REG_RD16(bp, offset)		readw(REG_ADDR(bp, offset))
 
+#define REG_WR_RELAXED(bp, offset, val)	\
+	writel_relaxed((u32)val, REG_ADDR(bp, offset))
+
+#define REG_WR16_RELAXED(bp, offset, val) \
+	writew_relaxed((u16)val, REG_ADDR(bp, offset))
+
 #define REG_WR(bp, offset, val)		writel((u32)val, REG_ADDR(bp, offset))
 #define REG_WR8(bp, offset, val)	writeb((u8)val, REG_ADDR(bp, offset))
 #define REG_WR16(bp, offset, val)	writew((u16)val, REG_ADDR(bp, offset))
@@ -758,10 +770,8 @@ struct bnx2x_fastpath {
 #if (BNX2X_DB_SHIFT < BNX2X_DB_MIN_SHIFT)
 #error "Min DB doorbell stride is 8"
 #endif
-#define DOORBELL(bp, cid, val) \
-	do { \
-		writel((u32)(val), bp->doorbells + (bp->db_size * (cid))); \
-	} while (0)
+#define DOORBELL_RELAXED(bp, cid, val) \
+	writel_relaxed((u32)(val), (bp)->doorbells + ((bp)->db_size * (cid)))
 
 /* TX CSUM helpers */
 #define SKB_CS_OFF(skb)		(offsetof(struct tcphdr, check) - \
@@ -1252,7 +1262,7 @@ enum {
 
 struct bnx2x_fw_stats_req {
 	struct stats_query_header hdr;
-	struct stats_query_entry query[FP_SB_MAX_E1x+
+	struct stats_query_entry query[FP_SB_MAX_E2 +
 		BNX2X_FIRST_QUEUE_QUERY_IDX];
 };
 
@@ -1261,7 +1271,7 @@ struct bnx2x_fw_stats_data {
 	struct per_port_stats		port;
 	struct per_pf_stats		pf;
 	struct fcoe_statistics_params	fcoe;
-	struct per_queue_stats		queue_stats[1];
+	struct per_queue_stats		queue_stats[];
 };
 
 /* Public slow path states */
@@ -1277,7 +1287,7 @@ enum sp_rtnl_flag {
 	BNX2X_SP_RTNL_HYPERVISOR_VLAN,
 	BNX2X_SP_RTNL_TX_STOP,
 	BNX2X_SP_RTNL_GET_DRV_VERSION,
-	BNX2X_SP_RTNL_CHANGE_UDP_PORT,
+	BNX2X_SP_RTNL_UPDATE_SVID,
 };
 
 enum bnx2x_iov_flag {
@@ -1332,11 +1342,6 @@ enum bnx2x_udp_port_type {
 	BNX2X_UDP_PORT_MAX,
 };
 
-struct bnx2x_udp_tunnel {
-	u16 dst_port;
-	u8 count;
-};
-
 struct bnx2x {
 	/* Fields used in the tx and intr/napi performance paths
 	 * are grouped together in the beginning of the structure
@@ -1481,7 +1486,6 @@ struct bnx2x {
 #define IS_VF_FLAG			(1 << 22)
 #define BC_SUPPORTS_RMMOD_CMD		(1 << 23)
 #define HAS_PHYS_PORT_ID		(1 << 24)
-#define AER_ENABLED			(1 << 25)
 #define PTP_SUPPORTED			(1 << 26)
 #define TX_TIMESTAMPING_EN		(1 << 27)
 
@@ -1504,6 +1508,8 @@ struct bnx2x {
 	bool			cnic_loaded;
 	struct cnic_eth_dev	*(*cnic_probe)(struct net_device *);
 
+	bool                    nic_stopped;
+
 	/* Flag that indicates that we can start looking for FCoE L2 queue
 	 * completions in the default status block.
 	 */
@@ -1529,6 +1535,7 @@ struct bnx2x {
 	struct link_vars	link_vars;
 	u32			link_cnt;
 	struct bnx2x_link_report_data last_reported_link;
+	bool			force_link_down;
 
 	struct mdio_if_info	mdio;
 
@@ -1843,7 +1850,15 @@ struct bnx2x {
 	bool accept_any_vlan;
 
 	/* Vxlan/Geneve related information */
-	struct bnx2x_udp_tunnel udp_tunnel_ports[BNX2X_UDP_PORT_MAX];
+	u16 udp_tunnel_ports[BNX2X_UDP_PORT_MAX];
+
+#define FW_CAP_INVALIDATE_VF_FP_HSI	BIT(0)
+	u32 fw_cap;
+
+	u32 fw_major;
+	u32 fw_minor;
+	u32 fw_rev;
+	u32 fw_eng;
 };
 
 /* Tx queues may be less or equal to Rx queues */
@@ -1967,6 +1982,9 @@ struct bnx2x_func_init_params {
 
 #define skip_queue(bp, idx)	(NO_FCOE(bp) && IS_FCOE_IDX(idx))
 
+/*self test*/
+int bnx2x_idle_chk(struct bnx2x *bp);
+
 /**
  * bnx2x_set_mac_one - configure a single MAC address
  *
@@ -1985,7 +2003,7 @@ struct bnx2x_func_init_params {
  * operation has been successfully scheduled and a negative - if a requested
  * operations has failed.
  */
-int bnx2x_set_mac_one(struct bnx2x *bp, u8 *mac,
+int bnx2x_set_mac_one(struct bnx2x *bp, const u8 *mac,
 		      struct bnx2x_vlan_mac_obj *obj, bool set,
 		      int mac_type, unsigned long *ramrod_flags);
 
@@ -2075,7 +2093,7 @@ void bnx2x_igu_clear_sb_gen(struct bnx2x *bp, u8 func, u8 idu_sb_id,
 			    bool is_pf);
 
 #define BNX2X_ILT_ZALLOC(x, y, size)					\
-	x = dma_zalloc_coherent(&bp->pdev->dev, size, y, GFP_KERNEL)
+	x = dma_alloc_coherent(&bp->pdev->dev, size, y, GFP_KERNEL)
 
 #define BNX2X_ILT_FREE(x, y, size) \
 	do { \
@@ -2186,6 +2204,13 @@ void bnx2x_igu_clear_sb_gen(struct bnx2x *bp, u8 func, u8 idu_sb_id,
 #define PMF_DMAE_C(bp)			(BP_PORT(bp) * MAX_DMAE_C_PER_PORT + \
 					 E1HVN_MAX)
 
+/* Following is the DMAE channel number allocation for the clients.
+ *   MFW: OCBB/OCSD implementations use DMAE channels 14/15 respectively.
+ *   Driver: 0-3 and 8-11 (for PF dmae operations)
+ *           4 and 12 (for stats requests)
+ */
+#define BNX2X_FW_DMAE_C                 13 /* Channel for FW DMAE operations */
+
 /* PCIE link and speed */
 #define PCICFG_LINK_WIDTH		0x1f00000
 #define PCICFG_LINK_WIDTH_SHIFT		20
@@ -2391,7 +2416,6 @@ void bnx2x_igu_clear_sb_gen(struct bnx2x *bp, u8 func, u8 idu_sb_id,
 #define ETH_MAX_RX_CLIENTS_E2		ETH_MAX_RX_CLIENTS_E1H
 #endif
 
-#define BNX2X_VPD_LEN			128
 #define VENDOR_ID_LEN			4
 
 #define VF_ACQUIRE_THRESH		3
@@ -2411,13 +2435,6 @@ int bnx2x_compare_fw_ver(struct bnx2x *bp, u32 load_code, bool print_err);
 #define HC_SEG_ACCESS_ATTN		4
 #define HC_SEG_ACCESS_NORM		0   /*Driver decision 0-1*/
 
-static const u32 dmae_reg_go_c[] = {
-	DMAE_REG_GO_C0, DMAE_REG_GO_C1, DMAE_REG_GO_C2, DMAE_REG_GO_C3,
-	DMAE_REG_GO_C4, DMAE_REG_GO_C5, DMAE_REG_GO_C6, DMAE_REG_GO_C7,
-	DMAE_REG_GO_C8, DMAE_REG_GO_C9, DMAE_REG_GO_C10, DMAE_REG_GO_C11,
-	DMAE_REG_GO_C12, DMAE_REG_GO_C13, DMAE_REG_GO_C14, DMAE_REG_GO_C15
-};
-
 void bnx2x_set_ethtool_ops(struct bnx2x *bp, struct net_device *netdev);
 void bnx2x_notify_link_changed(struct bnx2x *bp);
 
@@ -2508,6 +2525,7 @@ void bnx2x_update_mfw_dump(struct bnx2x *bp);
 void bnx2x_init_ptp(struct bnx2x *bp);
 int bnx2x_configure_ptp_filters(struct bnx2x *bp);
 void bnx2x_set_rx_ts(struct bnx2x *bp, struct sk_buff *skb);
+void bnx2x_register_phc(struct bnx2x *bp);
 
 #define BNX2X_MAX_PHC_DRIFT 31000000
 #define BNX2X_PTP_TX_TIMEOUT
@@ -2516,5 +2534,4 @@ void bnx2x_set_rx_ts(struct bnx2x *bp, struct sk_buff *skb);
  * Meant for implicit re-load flows.
  */
 int bnx2x_vlan_reconfigure_vid(struct bnx2x *bp);
-
 #endif /* bnx2x.h */
diff --git a/drivers/net/ethernet/broadcom/bnx2x/bnx2x_cmn.c b/drivers/net/ethernet/broadcom/bnx2x/bnx2x_cmn.c
index eccb3d1b6abb..e59530357e2c 100644
--- a/drivers/net/ethernet/broadcom/bnx2x/bnx2x_cmn.c
+++ b/drivers/net/ethernet/broadcom/bnx2x/bnx2x_cmn.c
@@ -25,9 +25,9 @@
 #include <linux/ip.h>
 #include <linux/crash_dump.h>
 #include <net/tcp.h>
+#include <net/gro.h>
 #include <net/ipv6.h>
 #include <net/ip6_checksum.h>
-#include <net/busy_poll.h>
 #include <linux/prefetch.h>
 #include "bnx2x_cmn.h"
 #include "bnx2x_init.h"
@@ -44,8 +44,7 @@ static void bnx2x_add_all_napi_cnic(struct bnx2x *bp)
 
 	/* Add NAPI objects */
 	for_each_rx_queue_cnic(bp, i) {
-		netif_napi_add(bp->dev, &bnx2x_fp(bp, i, napi),
-			       bnx2x_poll, NAPI_POLL_WEIGHT);
+		netif_napi_add(bp->dev, &bnx2x_fp(bp, i, napi), bnx2x_poll);
 	}
 }
 
@@ -55,8 +54,7 @@ static void bnx2x_add_all_napi(struct bnx2x *bp)
 
 	/* Add NAPI objects */
 	for_each_eth_queue(bp, i) {
-		netif_napi_add(bp->dev, &bnx2x_fp(bp, i, napi),
-			       bnx2x_poll, NAPI_POLL_WEIGHT);
+		netif_napi_add(bp->dev, &bnx2x_fp(bp, i, napi), bnx2x_poll);
 	}
 }
 
@@ -149,10 +147,11 @@ void bnx2x_fill_fw_str(struct bnx2x *bp, char *buf, size_t buf_len)
 
 		phy_fw_ver[0] = '\0';
 		bnx2x_get_ext_phy_fw_version(&bp->link_params,
-					     phy_fw_ver, PHY_FW_VER_LEN);
-		strlcpy(buf, bp->fw_ver, buf_len);
-		snprintf(buf + strlen(bp->fw_ver), 32 - strlen(bp->fw_ver),
-			 "bc %d.%d.%d%s%s",
+					     phy_fw_ver, sizeof(phy_fw_ver));
+		/* This may become truncated. */
+		scnprintf(buf, buf_len,
+			 "%sbc %d.%d.%d%s%s",
+			 bp->fw_ver,
 			 (bp->common.bc_ver & 0xff0000) >> 16,
 			 (bp->common.bc_ver & 0xff00) >> 8,
 			 (bp->common.bc_ver & 0xff),
@@ -286,6 +285,9 @@ int bnx2x_tx_int(struct bnx2x *bp, struct bnx2x_fp_txdata *txdata)
 	hw_cons = le16_to_cpu(*txdata->tx_cons_sb);
 	sw_cons = txdata->tx_pkt_cons;
 
+	/* Ensure subsequent loads occur after hw_cons */
+	smp_rmb();
+
 	while (sw_cons != hw_cons) {
 		u16 pkt_cons;
 
@@ -502,6 +504,7 @@ static void bnx2x_tpa_start(struct bnx2x_fastpath *fp, u16 queue,
  * @len_on_bd:		total length of the first packet for the
  *			aggregation.
  * @pkt_len:		length of all segments
+ * @num_of_coalesced_segs: count of segments
  *
  * Approximate value of the MSS for this aggregation calculated using
  * the first packet of it.
@@ -670,6 +673,18 @@ static int bnx2x_fill_frag_skb(struct bnx2x *bp, struct bnx2x_fastpath *fp,
 	return 0;
 }
 
+static struct sk_buff *
+bnx2x_build_skb(const struct bnx2x_fastpath *fp, void *data)
+{
+	struct sk_buff *skb;
+
+	if (fp->rx_frag_size)
+		skb = build_skb(data, fp->rx_frag_size);
+	else
+		skb = slab_build_skb(data);
+	return skb;
+}
+
 static void bnx2x_frag_free(const struct bnx2x_fastpath *fp, void *data)
 {
 	if (fp->rx_frag_size)
@@ -685,7 +700,7 @@ static void *bnx2x_frag_alloc(const struct bnx2x_fastpath *fp, gfp_t gfp_mask)
 		if (unlikely(gfpflags_allow_blocking(gfp_mask)))
 			return (void *)__get_free_page(gfp_mask);
 
-		return netdev_alloc_frag(fp->rx_frag_size);
+		return napi_alloc_frag(fp->rx_frag_size);
 	}
 
 	return kmalloc(fp->rx_buf_size + NET_SKB_PAD, gfp_mask);
@@ -738,8 +753,9 @@ static void bnx2x_gro_receive(struct bnx2x *bp, struct bnx2x_fastpath *fp,
 			bnx2x_gro_csum(bp, skb, bnx2x_gro_ipv6_csum);
 			break;
 		default:
-			WARN_ONCE(1, "Error: FW GRO supports only IPv4/IPv6, not 0x%04x\n",
-				  be16_to_cpu(skb->protocol));
+			netdev_WARN_ONCE(bp->dev,
+					 "Error: FW GRO supports only IPv4/IPv6, not 0x%04x\n",
+					 be16_to_cpu(skb->protocol));
 		}
 	}
 #endif
@@ -776,7 +792,7 @@ static void bnx2x_tpa_stop(struct bnx2x *bp, struct bnx2x_fastpath *fp,
 	dma_unmap_single(&bp->pdev->dev, dma_unmap_addr(rx_buf, mapping),
 			 fp->rx_buf_size, DMA_FROM_DEVICE);
 	if (likely(new_data))
-		skb = build_skb(data, fp->rx_frag_size);
+		skb = bnx2x_build_skb(fp, data);
 
 	if (likely(skb)) {
 #ifdef BNX2X_STOP_ON_ERROR
@@ -784,6 +800,7 @@ static void bnx2x_tpa_stop(struct bnx2x *bp, struct bnx2x_fastpath *fp,
 			BNX2X_ERR("skb_put is about to fail...  pad %d  len %d  rx_buf_size %d\n",
 				  pad, len, fp->rx_buf_size);
 			bnx2x_panic();
+			bnx2x_frag_free(fp, new_data);
 			return;
 		}
 #endif
@@ -1042,7 +1059,7 @@ static int bnx2x_rx_int(struct bnx2x_fastpath *fp, int budget)
 						 dma_unmap_addr(rx_buf, mapping),
 						 fp->rx_buf_size,
 						 DMA_FROM_DEVICE);
-				skb = build_skb(data, fp->rx_frag_size);
+				skb = bnx2x_build_skb(fp, data);
 				if (unlikely(!skb)) {
 					bnx2x_frag_free(fp, data);
 					bnx2x_fp_qstats(bp, fp)->
@@ -1260,6 +1277,11 @@ void __bnx2x_link_report(struct bnx2x *bp)
 {
 	struct bnx2x_link_report_data cur_data;
 
+	if (bp->force_link_down) {
+		bp->link_vars.link_up = 0;
+		return;
+	}
+
 	/* reread mf_cfg */
 	if (IS_PF(bp) && !CHIP_IS_E1(bp))
 		bnx2x_read_mf_cfg(bp);
@@ -1904,7 +1926,7 @@ void bnx2x_netif_stop(struct bnx2x *bp, int disable_hw)
 }
 
 u16 bnx2x_select_queue(struct net_device *dev, struct sk_buff *skb,
-		       void *accel_priv, select_queue_fallback_t fallback)
+		       struct net_device *sb_dev)
 {
 	struct bnx2x *bp = netdev_priv(dev);
 
@@ -1914,8 +1936,7 @@ u16 bnx2x_select_queue(struct net_device *dev, struct sk_buff *skb,
 
 		/* Skip VLAN tag if present */
 		if (ether_type == ETH_P_8021Q) {
-			struct vlan_ethhdr *vhdr =
-				(struct vlan_ethhdr *)skb->data;
+			struct vlan_ethhdr *vhdr = skb_vlan_eth_hdr(skb);
 
 			ether_type = ntohs(vhdr->h_vlan_encapsulated_proto);
 		}
@@ -1926,7 +1947,8 @@ u16 bnx2x_select_queue(struct net_device *dev, struct sk_buff *skb,
 	}
 
 	/* select a non-FCoE queue */
-	return fallback(dev, skb) % BNX2X_NUM_ETH_QUEUES(bp);
+	return netdev_pick_tx(dev, skb, NULL) %
+			(BNX2X_NUM_ETH_QUEUES(bp) * bp->max_cos);
 }
 
 void bnx2x_set_num_queues(struct bnx2x *bp)
@@ -1949,6 +1971,7 @@ void bnx2x_set_num_queues(struct bnx2x *bp)
  * bnx2x_set_real_num_queues - configure netdev->real_num_[tx,rx]_queues
  *
  * @bp:		Driver handle
+ * @include_cnic: handle cnic case
  *
  * We currently support for at most 16 Tx queues for each CoS thus we will
  * allocate a multiple of 16 for ETH L2 rings according to the value of the
@@ -2352,26 +2375,30 @@ int bnx2x_compare_fw_ver(struct bnx2x *bp, u32 load_code, bool print_err)
 	/* is another pf loaded on this engine? */
 	if (load_code != FW_MSG_CODE_DRV_LOAD_COMMON_CHIP &&
 	    load_code != FW_MSG_CODE_DRV_LOAD_COMMON) {
-		/* build my FW version dword */
-		u32 my_fw = (BCM_5710_FW_MAJOR_VERSION) +
-			(BCM_5710_FW_MINOR_VERSION << 8) +
-			(BCM_5710_FW_REVISION_VERSION << 16) +
-			(BCM_5710_FW_ENGINEERING_VERSION << 24);
+		u8 loaded_fw_major, loaded_fw_minor, loaded_fw_rev, loaded_fw_eng;
+		u32 loaded_fw;
 
 		/* read loaded FW from chip */
-		u32 loaded_fw = REG_RD(bp, XSEM_REG_PRAM);
+		loaded_fw = REG_RD(bp, XSEM_REG_PRAM);
+
+		loaded_fw_major = loaded_fw & 0xff;
+		loaded_fw_minor = (loaded_fw >> 8) & 0xff;
+		loaded_fw_rev = (loaded_fw >> 16) & 0xff;
+		loaded_fw_eng = (loaded_fw >> 24) & 0xff;
 
-		DP(BNX2X_MSG_SP, "loaded fw %x, my fw %x\n",
-		   loaded_fw, my_fw);
+		DP(BNX2X_MSG_SP, "loaded fw 0x%x major 0x%x minor 0x%x rev 0x%x eng 0x%x\n",
+		   loaded_fw, loaded_fw_major, loaded_fw_minor, loaded_fw_rev, loaded_fw_eng);
 
 		/* abort nic load if version mismatch */
-		if (my_fw != loaded_fw) {
+		if (loaded_fw_major != BCM_5710_FW_MAJOR_VERSION ||
+		    loaded_fw_minor != BCM_5710_FW_MINOR_VERSION ||
+		    loaded_fw_eng != BCM_5710_FW_ENGINEERING_VERSION ||
+		    loaded_fw_rev < BCM_5710_FW_REVISION_VERSION_V15) {
 			if (print_err)
-				BNX2X_ERR("bnx2x with FW %x was already loaded which mismatches my %x FW. Aborting\n",
-					  loaded_fw, my_fw);
+				BNX2X_ERR("loaded FW incompatible. Aborting\n");
 			else
-				BNX2X_DEV_INFO("bnx2x with FW %x was already loaded which mismatches my %x FW, possibly due to MF UNDI\n",
-					       loaded_fw, my_fw);
+				BNX2X_DEV_INFO("loaded FW incompatible, possibly due to MF UNDI\n");
+
 			return -EBUSY;
 		}
 	}
@@ -2482,8 +2509,7 @@ static void bnx2x_bz_fp(struct bnx2x *bp, int index)
 	 */
 	if (bp->dev->features & NETIF_F_LRO)
 		fp->mode = TPA_MODE_LRO;
-	else if (bp->dev->features & NETIF_F_GRO &&
-		 bnx2x_mtu_allows_gro(bp->dev->mtu))
+	else if (bp->dev->features & NETIF_F_GRO_HW)
 		fp->mode = TPA_MODE_GRO;
 	else
 		fp->mode = TPA_MODE_DISABLED;
@@ -2659,7 +2685,8 @@ int bnx2x_nic_load(struct bnx2x *bp, int load_mode)
 	}
 
 	/* Allocated memory for FW statistics  */
-	if (bnx2x_alloc_fw_stats_mem(bp))
+	rc = bnx2x_alloc_fw_stats_mem(bp);
+	if (rc)
 		LOAD_ERROR_EXIT(bp, load_error0);
 
 	/* request pf to initialize status blocks */
@@ -2689,6 +2716,7 @@ int bnx2x_nic_load(struct bnx2x *bp, int load_mode)
 	bnx2x_add_all_napi(bp);
 	DP(NETIF_MSG_IFUP, "napi added\n");
 	bnx2x_napi_enable(bp);
+	bp->nic_stopped = false;
 
 	if (IS_PF(bp)) {
 		/* set pf load just before approaching the MCP */
@@ -2817,6 +2845,7 @@ int bnx2x_nic_load(struct bnx2x *bp, int load_mode)
 		bp->pending_max = 0;
 	}
 
+	bp->force_link_down = false;
 	if (bp->port.pmf) {
 		rc = bnx2x_initial_phy_init(bp, load_mode);
 		if (rc)
@@ -2835,6 +2864,7 @@ int bnx2x_nic_load(struct bnx2x *bp, int load_mode)
 	bnx2x_set_rx_mode_inner(bp);
 
 	if (bp->flags & PTP_SUPPORTED) {
+		bnx2x_register_phc(bp);
 		bnx2x_init_ptp(bp);
 		bnx2x_configure_ptp_filters(bp);
 	}
@@ -2932,6 +2962,7 @@ load_error2:
 load_error1:
 	bnx2x_napi_disable(bp);
 	bnx2x_del_all_napi(bp);
+	bp->nic_stopped = true;
 
 	/* clear pf_load status, as it was already set */
 	if (IS_PF(bp))
@@ -3028,9 +3059,9 @@ int bnx2x_nic_unload(struct bnx2x *bp, int unload_mode, bool keep_link)
 
 	bp->rx_mode = BNX2X_RX_MODE_NONE;
 
-	del_timer_sync(&bp->timer);
+	timer_delete_sync(&bp->timer);
 
-	if (IS_PF(bp)) {
+	if (IS_PF(bp) && !BP_NOMCP(bp)) {
 		/* Set ALWAYS_ALIVE bit in shmem */
 		bp->fw_drv_pulse_wr_seq |= DRV_PULSE_ALWAYS_ALIVE;
 		bnx2x_drv_pulse(bp);
@@ -3048,12 +3079,13 @@ int bnx2x_nic_unload(struct bnx2x *bp, int unload_mode, bool keep_link)
 	/* if VF indicate to PF this function is going down (PF will delete sp
 	 * elements and clear initializations
 	 */
-	if (IS_VF(bp))
+	if (IS_VF(bp)) {
+		bnx2x_clear_vlan_info(bp);
 		bnx2x_vfpf_close_vf(bp);
-	else if (unload_mode != UNLOAD_RECOVERY)
+	} else if (unload_mode != UNLOAD_RECOVERY) {
 		/* if this is a normal/close unload need to clean up chip*/
 		bnx2x_chip_cleanup(bp, unload_mode, keep_link);
-	else {
+	} else {
 		/* Send the UNLOAD_REQUEST to the MCP */
 		bnx2x_send_unload_req(bp, unload_mode);
 
@@ -3066,14 +3098,17 @@ int bnx2x_nic_unload(struct bnx2x *bp, int unload_mode, bool keep_link)
 		if (!CHIP_IS_E1x(bp))
 			bnx2x_pf_disable(bp);
 
-		/* Disable HW interrupts, NAPI */
-		bnx2x_netif_stop(bp, 1);
-		/* Delete all NAPI objects */
-		bnx2x_del_all_napi(bp);
-		if (CNIC_LOADED(bp))
-			bnx2x_del_all_napi_cnic(bp);
-		/* Release IRQs */
-		bnx2x_free_irq(bp);
+		if (!bp->nic_stopped) {
+			/* Disable HW interrupts, NAPI */
+			bnx2x_netif_stop(bp, 1);
+			/* Delete all NAPI objects */
+			bnx2x_del_all_napi(bp);
+			if (CNIC_LOADED(bp))
+				bnx2x_del_all_napi_cnic(bp);
+			/* Release IRQs */
+			bnx2x_free_irq(bp);
+			bp->nic_stopped = true;
+		}
 
 		/* Report UNLOAD_DONE to MCP */
 		bnx2x_send_unload_done(bp, false);
@@ -3116,7 +3151,7 @@ int bnx2x_nic_unload(struct bnx2x *bp, int unload_mode, bool keep_link)
 	bp->cnic_loaded = false;
 
 	/* Clear driver version indication in shmem */
-	if (IS_PF(bp))
+	if (IS_PF(bp) && !BP_NOMCP(bp))
 		bnx2x_update_mng_version(bp);
 
 	/* Check if there are pending parity attentions. If there are - set
@@ -3402,12 +3437,9 @@ static int bnx2x_pkt_req_lin(struct bnx2x *bp, struct sk_buff *skb,
 
 			/* Headers length */
 			if (xmit_type & XMIT_GSO_ENC)
-				hlen = (int)(skb_inner_transport_header(skb) -
-					     skb->data) +
-					     inner_tcp_hdrlen(skb);
+				hlen = skb_inner_tcp_all_headers(skb);
 			else
-				hlen = (int)(skb_transport_header(skb) -
-					     skb->data) + tcp_hdrlen(skb);
+				hlen = skb_tcp_all_headers(skb);
 
 			/* Amount of data (w/o headers) on linear part of SKB*/
 			first_bd_sz = skb_headlen(skb) - hlen;
@@ -3506,7 +3538,7 @@ static u8 bnx2x_set_pbd_csum_enc(struct bnx2x *bp, struct sk_buff *skb,
 				 u32 *parsing_data, u32 xmit_type)
 {
 	*parsing_data |=
-		((((u8 *)skb_inner_transport_header(skb) - skb->data) >> 1) <<
+		((skb_inner_transport_offset(skb) >> 1) <<
 		ETH_TX_PARSE_BD_E2_L4_HDR_START_OFFSET_W_SHIFT) &
 		ETH_TX_PARSE_BD_E2_L4_HDR_START_OFFSET_W;
 
@@ -3515,15 +3547,13 @@ static u8 bnx2x_set_pbd_csum_enc(struct bnx2x *bp, struct sk_buff *skb,
 			ETH_TX_PARSE_BD_E2_TCP_HDR_LENGTH_DW_SHIFT) &
 			ETH_TX_PARSE_BD_E2_TCP_HDR_LENGTH_DW;
 
-		return skb_inner_transport_header(skb) +
-			inner_tcp_hdrlen(skb) - skb->data;
+		return skb_inner_tcp_all_headers(skb);
 	}
 
 	/* We support checksum offload for TCP and UDP only.
 	 * No need to pass the UDP header length - it's a constant.
 	 */
-	return skb_inner_transport_header(skb) +
-		sizeof(struct udphdr) - skb->data;
+	return skb_inner_transport_offset(skb) + sizeof(struct udphdr);
 }
 
 /**
@@ -3540,7 +3570,7 @@ static u8 bnx2x_set_pbd_csum_e2(struct bnx2x *bp, struct sk_buff *skb,
 				u32 *parsing_data, u32 xmit_type)
 {
 	*parsing_data |=
-		((((u8 *)skb_transport_header(skb) - skb->data) >> 1) <<
+		((skb_transport_offset(skb) >> 1) <<
 		ETH_TX_PARSE_BD_E2_L4_HDR_START_OFFSET_W_SHIFT) &
 		ETH_TX_PARSE_BD_E2_L4_HDR_START_OFFSET_W;
 
@@ -3549,12 +3579,12 @@ static u8 bnx2x_set_pbd_csum_e2(struct bnx2x *bp, struct sk_buff *skb,
 			ETH_TX_PARSE_BD_E2_TCP_HDR_LENGTH_DW_SHIFT) &
 			ETH_TX_PARSE_BD_E2_TCP_HDR_LENGTH_DW;
 
-		return skb_transport_header(skb) + tcp_hdrlen(skb) - skb->data;
+		return skb_tcp_all_headers(skb);
 	}
 	/* We support checksum offload for TCP and UDP only.
 	 * No need to pass the UDP header length - it's a constant.
 	 */
-	return skb_transport_header(skb) + sizeof(struct udphdr) - skb->data;
+	return skb_transport_offset(skb) + sizeof(struct udphdr);
 }
 
 /* set FW indication according to inner or outer protocols if tunneled */
@@ -3583,7 +3613,7 @@ static u8 bnx2x_set_pbd_csum(struct bnx2x *bp, struct sk_buff *skb,
 			     struct eth_tx_parse_bd_e1x *pbd,
 			     u32 xmit_type)
 {
-	u8 hlen = (skb_network_header(skb) - skb->data) >> 1;
+	u8 hlen = skb_network_offset(skb) >> 1;
 
 	/* for now NS flag is not used in Linux */
 	pbd->global_data =
@@ -3591,8 +3621,7 @@ static u8 bnx2x_set_pbd_csum(struct bnx2x *bp, struct sk_buff *skb,
 			    ((skb->protocol == cpu_to_be16(ETH_P_8021Q)) <<
 			     ETH_TX_PARSE_BD_E1X_LLC_SNAP_EN_SHIFT));
 
-	pbd->ip_hlen_w = (skb_transport_header(skb) -
-			skb_network_header(skb)) >> 1;
+	pbd->ip_hlen_w = skb_network_header_len(skb) >> 1;
 
 	hlen += pbd->ip_hlen_w;
 
@@ -3637,8 +3666,7 @@ static void bnx2x_update_pbds_gso_enc(struct sk_buff *skb,
 	u8 outerip_off, outerip_len = 0;
 
 	/* from outer IP to transport */
-	hlen_w = (skb_inner_transport_header(skb) -
-		  skb_network_header(skb)) >> 1;
+	hlen_w = skb_inner_transport_offset(skb) >> 1;
 
 	/* transport len */
 	hlen_w += inner_tcp_hdrlen(skb) >> 1;
@@ -3684,7 +3712,7 @@ static void bnx2x_update_pbds_gso_enc(struct sk_buff *skb,
 					0, IPPROTO_TCP, 0));
 	}
 
-	outerip_off = (skb_network_header(skb) - skb->data) >> 1;
+	outerip_off = (skb_network_offset(skb)) >> 1;
 
 	*global_data |=
 		outerip_off |
@@ -3850,9 +3878,12 @@ netdev_tx_t bnx2x_start_xmit(struct sk_buff *skb, struct net_device *dev)
 
 	if (unlikely(skb_shinfo(skb)->tx_flags & SKBTX_HW_TSTAMP)) {
 		if (!(bp->flags & TX_TIMESTAMPING_EN)) {
+			bp->eth_stats.ptp_skip_tx_ts++;
 			BNX2X_ERR("Tx timestamping was not enabled, this packet will not be timestamped\n");
 		} else if (bp->ptp_tx_skb) {
-			BNX2X_ERR("The device supports only a single outstanding packet to timestamp, this packet will not be timestamped\n");
+			bp->eth_stats.ptp_skip_tx_ts++;
+			netdev_err_once(bp->dev,
+					"Device supports only a single outstanding packet to timestamp, this packet won't be timestamped\n");
 		} else {
 			skb_shinfo(skb)->tx_flags |= SKBTX_IN_PROGRESS;
 			/* schedule check for Tx timestamp */
@@ -3883,15 +3914,26 @@ netdev_tx_t bnx2x_start_xmit(struct sk_buff *skb, struct net_device *dev)
 		/* when transmitting in a vf, start bd must hold the ethertype
 		 * for fw to enforce it
 		 */
+		u16 vlan_tci = 0;
 #ifndef BNX2X_STOP_ON_ERROR
-		if (IS_VF(bp))
+		if (IS_VF(bp)) {
 #endif
-			tx_start_bd->vlan_or_ethertype =
-				cpu_to_le16(ntohs(eth->h_proto));
+			/* Still need to consider inband vlan for enforced */
+			if (__vlan_get_tag(skb, &vlan_tci)) {
+				tx_start_bd->vlan_or_ethertype =
+					cpu_to_le16(ntohs(eth->h_proto));
+			} else {
+				tx_start_bd->bd_flags.as_bitfield |=
+					(X_ETH_INBAND_VLAN <<
+					 ETH_TX_BD_FLAGS_VLAN_MODE_SHIFT);
+				tx_start_bd->vlan_or_ethertype =
+					cpu_to_le16(vlan_tci);
+			}
 #ifndef BNX2X_STOP_ON_ERROR
-		else
+		} else {
 			/* used by FW for packet accounting */
 			tx_start_bd->vlan_or_ethertype = cpu_to_le16(pkt_prod);
+		}
 #endif
 	}
 
@@ -4142,11 +4184,10 @@ netdev_tx_t bnx2x_start_xmit(struct sk_buff *skb, struct net_device *dev)
 	wmb();
 
 	txdata->tx_db.data.prod += nbd;
-	barrier();
-
-	DOORBELL(bp, txdata->cid, txdata->tx_db.raw);
+	/* make sure descriptor update is observed by HW */
+	wmb();
 
-	mmiowb();
+	DOORBELL_RELAXED(bp, txdata->cid, txdata->tx_db.raw);
 
 	txdata->tx_bd_prod += nbd;
 
@@ -4205,8 +4246,8 @@ void bnx2x_get_c2s_mapping(struct bnx2x *bp, u8 *c2s_map, u8 *c2s_default)
 /**
  * bnx2x_setup_tc - routine to configure net_device for multi tc
  *
- * @netdev: net device to configure
- * @tc: number of traffic classes to enable
+ * @dev: net device to configure
+ * @num_tc: number of traffic classes to enable
  *
  * callback connected to the ndo_setup_tc function pointer
  */
@@ -4273,15 +4314,17 @@ int bnx2x_setup_tc(struct net_device *dev, u8 num_tc)
 	return 0;
 }
 
-int __bnx2x_setup_tc(struct net_device *dev, u32 handle, __be16 proto,
-		     struct tc_to_netdev *tc)
+int __bnx2x_setup_tc(struct net_device *dev, enum tc_setup_type type,
+		     void *type_data)
 {
-	if (tc->type != TC_SETUP_MQPRIO)
-		return -EINVAL;
+	struct tc_mqprio_qopt *mqprio = type_data;
 
-	tc->mqprio->hw = TC_MQPRIO_HW_OFFLOAD_TCS;
+	if (type != TC_SETUP_QDISC_MQPRIO)
+		return -EOPNOTSUPP;
 
-	return bnx2x_setup_tc(dev, tc->mqprio->num_tc);
+	mqprio->hw = TC_MQPRIO_HW_OFFLOAD_TCS;
+
+	return bnx2x_setup_tc(dev, mqprio->num_tc);
 }
 
 /* called with rtnl_lock */
@@ -4307,7 +4350,7 @@ int bnx2x_change_mac_addr(struct net_device *dev, void *p)
 			return rc;
 	}
 
-	memcpy(dev->dev_addr, addr->sa_data, dev->addr_len);
+	eth_hw_addr_set(dev, addr->sa_data);
 
 	if (netif_running(dev))
 		rc = bnx2x_set_eth_mac(bp, true);
@@ -4859,7 +4902,10 @@ int bnx2x_change_mtu(struct net_device *dev, int new_mtu)
 	 * because the actual alloc size is
 	 * only updated as part of load
 	 */
-	dev->mtu = new_mtu;
+	WRITE_ONCE(dev->mtu, new_mtu);
+
+	if (!bnx2x_mtu_allows_gro(new_mtu))
+		dev->features &= ~NETIF_F_GRO_HW;
 
 	if (IS_PF(bp) && SHMEM2_HAS(bp, curr_cfg))
 		SHMEM2_WR(bp, curr_cfg, CURR_CFG_MET_OS);
@@ -4890,10 +4936,13 @@ netdev_features_t bnx2x_fix_features(struct net_device *dev,
 	}
 
 	/* TPA requires Rx CSUM offloading */
-	if (!(features & NETIF_F_RXCSUM)) {
+	if (!(features & NETIF_F_RXCSUM))
+		features &= ~NETIF_F_LRO;
+
+	if (!(features & NETIF_F_GRO) || !bnx2x_mtu_allows_gro(dev->mtu))
+		features &= ~NETIF_F_GRO_HW;
+	if (features & NETIF_F_GRO_HW)
 		features &= ~NETIF_F_LRO;
-		features &= ~NETIF_F_GRO;
-	}
 
 	return features;
 }
@@ -4920,13 +4969,8 @@ int bnx2x_set_features(struct net_device *dev, netdev_features_t features)
 		}
 	}
 
-	/* if GRO is changed while LRO is enabled, don't force a reload */
-	if ((changes & NETIF_F_GRO) && (features & NETIF_F_LRO))
-		changes &= ~NETIF_F_GRO;
-
-	/* if GRO is changed while HW TPA is off, don't force a reload */
-	if ((changes & NETIF_F_GRO) && bp->disable_tpa)
-		changes &= ~NETIF_F_GRO;
+	/* Don't care about GRO changes */
+	changes &= ~NETIF_F_GRO;
 
 	if (changes)
 		bnx2x_reload = true;
@@ -4943,12 +4987,17 @@ int bnx2x_set_features(struct net_device *dev, netdev_features_t features)
 	return 0;
 }
 
-void bnx2x_tx_timeout(struct net_device *dev)
+void bnx2x_tx_timeout(struct net_device *dev, unsigned int txqueue)
 {
 	struct bnx2x *bp = netdev_priv(dev);
 
-#ifdef BNX2X_STOP_ON_ERROR
+	/* We want the information of the dump logged,
+	 * but calling bnx2x_panic() would kill all chances of recovery.
+	 */
 	if (!bp->panic)
+#ifndef BNX2X_STOP_ON_ERROR
+		bnx2x_panic_dump(bp, false);
+#else
 		bnx2x_panic();
 #endif
 
@@ -4956,8 +5005,9 @@ void bnx2x_tx_timeout(struct net_device *dev)
 	bnx2x_schedule_sp_rtnl(bp, BNX2X_SP_RTNL_TX_TIMEOUT, 0);
 }
 
-int bnx2x_suspend(struct pci_dev *pdev, pm_message_t state)
+static int __maybe_unused bnx2x_suspend(struct device *dev_d)
 {
+	struct pci_dev *pdev = to_pci_dev(dev_d);
 	struct net_device *dev = pci_get_drvdata(pdev);
 	struct bnx2x *bp;
 
@@ -4969,8 +5019,6 @@ int bnx2x_suspend(struct pci_dev *pdev, pm_message_t state)
 
 	rtnl_lock();
 
-	pci_save_state(pdev);
-
 	if (!netif_running(dev)) {
 		rtnl_unlock();
 		return 0;
@@ -4980,15 +5028,14 @@ int bnx2x_suspend(struct pci_dev *pdev, pm_message_t state)
 
 	bnx2x_nic_unload(bp, UNLOAD_CLOSE, false);
 
-	bnx2x_set_power_state(bp, pci_choose_state(pdev, state));
-
 	rtnl_unlock();
 
 	return 0;
 }
 
-int bnx2x_resume(struct pci_dev *pdev)
+static int __maybe_unused bnx2x_resume(struct device *dev_d)
 {
+	struct pci_dev *pdev = to_pci_dev(dev_d);
 	struct net_device *dev = pci_get_drvdata(pdev);
 	struct bnx2x *bp;
 	int rc;
@@ -5006,14 +5053,11 @@ int bnx2x_resume(struct pci_dev *pdev)
 
 	rtnl_lock();
 
-	pci_restore_state(pdev);
-
 	if (!netif_running(dev)) {
 		rtnl_unlock();
 		return 0;
 	}
 
-	bnx2x_set_power_state(bp, PCI_D0);
 	netif_device_attach(dev);
 
 	rc = bnx2x_nic_load(bp, LOAD_OPEN);
@@ -5023,6 +5067,8 @@ int bnx2x_resume(struct pci_dev *pdev)
 	return rc;
 }
 
+SIMPLE_DEV_PM_OPS(bnx2x_pm_ops, bnx2x_suspend, bnx2x_resume);
+
 void bnx2x_set_ctx_validation(struct bnx2x *bp, struct eth_context *cxt,
 			      u32 cid)
 {
diff --git a/drivers/net/ethernet/broadcom/bnx2x/bnx2x_cmn.h b/drivers/net/ethernet/broadcom/bnx2x/bnx2x_cmn.h
index 243cb9748d35..0bc1367fd649 100644
--- a/drivers/net/ethernet/broadcom/bnx2x/bnx2x_cmn.h
+++ b/drivers/net/ethernet/broadcom/bnx2x/bnx2x_cmn.h
@@ -52,7 +52,7 @@ extern int bnx2x_num_queues;
 
 #define BNX2X_PCI_ALLOC(y, size)					\
 ({									\
-	void *x = dma_zalloc_coherent(&bp->pdev->dev, size, y, GFP_KERNEL); \
+	void *x = dma_alloc_coherent(&bp->pdev->dev, size, y, GFP_KERNEL); \
 	if (x)								\
 		DP(NETIF_MSG_HW,					\
 		   "BNX2X_PCI_ALLOC: Physical %Lx Virtual %p\n",	\
@@ -425,6 +425,8 @@ void bnx2x_set_reset_global(struct bnx2x *bp);
 void bnx2x_disable_close_the_gate(struct bnx2x *bp);
 int bnx2x_init_hw_func_cnic(struct bnx2x *bp);
 
+void bnx2x_clear_vlan_info(struct bnx2x *bp);
+
 /**
  * bnx2x_sp_event - handle ramrods completion.
  *
@@ -486,18 +488,19 @@ netdev_tx_t bnx2x_start_xmit(struct sk_buff *skb, struct net_device *dev);
 
 /* setup_tc callback */
 int bnx2x_setup_tc(struct net_device *dev, u8 num_tc);
-int __bnx2x_setup_tc(struct net_device *dev, u32 handle, __be16 proto,
-		     struct tc_to_netdev *tc);
+int __bnx2x_setup_tc(struct net_device *dev, enum tc_setup_type type,
+		     void *type_data);
 
 int bnx2x_get_vf_config(struct net_device *dev, int vf,
 			struct ifla_vf_info *ivi);
 int bnx2x_set_vf_mac(struct net_device *dev, int queue, u8 *mac);
 int bnx2x_set_vf_vlan(struct net_device *netdev, int vf, u16 vlan, u8 qos,
 		      __be16 vlan_proto);
+int bnx2x_set_vf_spoofchk(struct net_device *dev, int idx, bool val);
 
 /* select_queue callback */
 u16 bnx2x_select_queue(struct net_device *dev, struct sk_buff *skb,
-		       void *accel_priv, select_queue_fallback_t fallback);
+		       struct net_device *sb_dev);
 
 static inline void bnx2x_update_rx_prod(struct bnx2x *bp,
 					struct bnx2x_fastpath *fp,
@@ -522,10 +525,8 @@ static inline void bnx2x_update_rx_prod(struct bnx2x *bp,
 	wmb();
 
 	for (i = 0; i < sizeof(rx_prods)/4; i++)
-		REG_WR(bp, fp->ustorm_rx_prods_offset + i*4,
-		       ((u32 *)&rx_prods)[i]);
-
-	mmiowb(); /* keep prod updates ordered */
+		REG_WR_RELAXED(bp, fp->ustorm_rx_prods_offset + i * 4,
+			       ((u32 *)&rx_prods)[i]);
 
 	DP(NETIF_MSG_RX_STATUS,
 	   "queue[%d]:  wrote  bd_prod %u  cqe_prod %u  sge_prod %u\n",
@@ -540,9 +541,7 @@ int bnx2x_change_mac_addr(struct net_device *dev, void *p);
 /* NAPI poll Tx part */
 int bnx2x_tx_int(struct bnx2x *bp, struct bnx2x_fp_txdata *txdata);
 
-/* suspend/resume callbacks */
-int bnx2x_suspend(struct pci_dev *pdev, pm_message_t state);
-int bnx2x_resume(struct pci_dev *pdev);
+extern const struct dev_pm_ops bnx2x_pm_ops;
 
 /* Release IRQ vectors */
 void bnx2x_free_irq(struct bnx2x *bp);
@@ -616,7 +615,7 @@ int bnx2x_set_features(struct net_device *dev, netdev_features_t features);
  *
  * @dev:	net device
  */
-void bnx2x_tx_timeout(struct net_device *dev);
+void bnx2x_tx_timeout(struct net_device *dev, unsigned int txqueue);
 
 /** bnx2x_get_c2s_mapping - read inner-to-outer vlan configuration
  * c2s_map should have BNX2X_MAX_PRIORITY entries.
@@ -651,7 +650,6 @@ static inline void bnx2x_igu_ack_sb_gen(struct bnx2x *bp, u8 igu_sb_id,
 	REG_WR(bp, igu_addr, cmd_data.sb_id_and_flags);
 
 	/* Make sure that ACK is written */
-	mmiowb();
 	barrier();
 }
 
@@ -672,7 +670,6 @@ static inline void bnx2x_hc_ack_sb(struct bnx2x *bp, u8 sb_id,
 	REG_WR(bp, hc_addr, (*(u32 *)&igu_ack));
 
 	/* Make sure that ACK is written */
-	mmiowb();
 	barrier();
 }
 
@@ -828,9 +825,9 @@ static inline void bnx2x_del_all_napi_cnic(struct bnx2x *bp)
 	int i;
 
 	for_each_rx_queue_cnic(bp, i) {
-		napi_hash_del(&bnx2x_fp(bp, i, napi));
-		netif_napi_del(&bnx2x_fp(bp, i, napi));
+		__netif_napi_del(&bnx2x_fp(bp, i, napi));
 	}
+	synchronize_net();
 }
 
 static inline void bnx2x_del_all_napi(struct bnx2x *bp)
@@ -838,9 +835,9 @@ static inline void bnx2x_del_all_napi(struct bnx2x *bp)
 	int i;
 
 	for_each_eth_queue(bp, i) {
-		napi_hash_del(&bnx2x_fp(bp, i, napi));
-		netif_napi_del(&bnx2x_fp(bp, i, napi));
+		__netif_napi_del(&bnx2x_fp(bp, i, napi));
 	}
+	synchronize_net();
 }
 
 int bnx2x_set_int_mode(struct bnx2x *bp);
@@ -963,12 +960,12 @@ static inline int bnx2x_func_start(struct bnx2x *bp)
 		start_params->network_cos_mode = STATIC_COS;
 	else /* CHIP_IS_E1X */
 		start_params->network_cos_mode = FW_WRR;
-	if (bp->udp_tunnel_ports[BNX2X_UDP_PORT_VXLAN].count) {
-		port = bp->udp_tunnel_ports[BNX2X_UDP_PORT_VXLAN].dst_port;
+	if (bp->udp_tunnel_ports[BNX2X_UDP_PORT_VXLAN]) {
+		port = bp->udp_tunnel_ports[BNX2X_UDP_PORT_VXLAN];
 		start_params->vxlan_dst_port = port;
 	}
-	if (bp->udp_tunnel_ports[BNX2X_UDP_PORT_GENEVE].count) {
-		port = bp->udp_tunnel_ports[BNX2X_UDP_PORT_GENEVE].dst_port;
+	if (bp->udp_tunnel_ports[BNX2X_UDP_PORT_GENEVE]) {
+		port = bp->udp_tunnel_ports[BNX2X_UDP_PORT_GENEVE];
 		start_params->geneve_dst_port = port;
 	}
 
@@ -1005,9 +1002,6 @@ static inline void bnx2x_set_fw_mac_addr(__le16 *fw_hi, __le16 *fw_mid,
 static inline void bnx2x_free_rx_mem_pool(struct bnx2x *bp,
 					  struct bnx2x_alloc_pool *pool)
 {
-	if (!pool->page)
-		return;
-
 	put_page(pool->page);
 
 	pool->page = NULL;
@@ -1018,6 +1012,9 @@ static inline void bnx2x_free_rx_sge_range(struct bnx2x *bp,
 {
 	int i;
 
+	if (!fp->page_pool.page)
+		return;
+
 	if (fp->mode == TPA_MODE_DISABLED)
 		return;
 
@@ -1110,7 +1107,7 @@ static inline u8 bnx2x_get_path_func_num(struct bnx2x *bp)
 		for (i = 0; i < E1H_FUNC_MAX / 2; i++) {
 			u32 func_config =
 				MF_CFG_RD(bp,
-					  func_mf_config[BP_PORT(bp) + 2 * i].
+					  func_mf_config[BP_PATH(bp) + 2 * i].
 					  config);
 			func_num +=
 				((func_config & FUNC_MF_CFG_FUNC_HIDE) ? 0 : 1);
diff --git a/drivers/net/ethernet/broadcom/bnx2x/bnx2x_dcb.c b/drivers/net/ethernet/broadcom/bnx2x/bnx2x_dcb.c
index 2c6ba046d2a8..9af81630c8a4 100644
--- a/drivers/net/ethernet/broadcom/bnx2x/bnx2x_dcb.c
+++ b/drivers/net/ethernet/broadcom/bnx2x/bnx2x_dcb.c
@@ -344,7 +344,7 @@ static void  bnx2x_dcbx_get_pfc_feature(struct bnx2x *bp,
 	}
 }
 
-/* maps unmapped priorities to to the same COS as L2 */
+/* maps unmapped priorities to the same COS as L2 */
 static void bnx2x_dcbx_map_nw(struct bnx2x *bp)
 {
 	int i;
@@ -1145,7 +1145,7 @@ static void bnx2x_dcbx_get_num_pg_traf_type(struct bnx2x *bp,
 					break;
 				}
 			}
-			if (false == pg_found) {
+			if (!pg_found) {
 				data[help_data->num_of_pg].pg = add_pg;
 				data[help_data->num_of_pg].pg_priority =
 						(1 << ttp[add_traf_type]);
@@ -1155,7 +1155,7 @@ static void bnx2x_dcbx_get_num_pg_traf_type(struct bnx2x *bp,
 		}
 		DP(BNX2X_MSG_DCB,
 		   "add_traf_type %d pg_found %s num_of_pg %d\n",
-		   add_traf_type, (false == pg_found) ? "NO" : "YES",
+		   add_traf_type, !pg_found ? "NO" : "YES",
 		   help_data->num_of_pg);
 	}
 }
@@ -1544,8 +1544,7 @@ static void bnx2x_dcbx_2cos_limit_cee_three_pg_to_cos_params(
 			if (pg_entry < DCBX_MAX_NUM_PG_BW_ENTRIES) {
 				entry = 0;
 
-				if (i == (num_of_pri-1) &&
-				    false == b_found_strict)
+				if (i == (num_of_pri-1) && !b_found_strict)
 					/* last entry will be handled separately
 					 * If no priority is strict than last
 					 * entry goes to last queue.
diff --git a/drivers/net/ethernet/broadcom/bnx2x/bnx2x_ethtool.c b/drivers/net/ethernet/broadcom/bnx2x/bnx2x_ethtool.c
index 43423744fdfa..fc8dec37a9e4 100644
--- a/drivers/net/ethernet/broadcom/bnx2x/bnx2x_ethtool.c
+++ b/drivers/net/ethernet/broadcom/bnx2x/bnx2x_ethtool.c
@@ -39,34 +39,34 @@ static const struct {
 	int size;
 	char string[ETH_GSTRING_LEN];
 } bnx2x_q_stats_arr[] = {
-/* 1 */	{ Q_STATS_OFFSET32(total_bytes_received_hi), 8, "[%s]: rx_bytes" },
+/* 1 */	{ Q_STATS_OFFSET32(total_bytes_received_hi), 8, "[%d]: rx_bytes" },
 	{ Q_STATS_OFFSET32(total_unicast_packets_received_hi),
-						8, "[%s]: rx_ucast_packets" },
+						8, "[%d]: rx_ucast_packets" },
 	{ Q_STATS_OFFSET32(total_multicast_packets_received_hi),
-						8, "[%s]: rx_mcast_packets" },
+						8, "[%d]: rx_mcast_packets" },
 	{ Q_STATS_OFFSET32(total_broadcast_packets_received_hi),
-						8, "[%s]: rx_bcast_packets" },
-	{ Q_STATS_OFFSET32(no_buff_discard_hi),	8, "[%s]: rx_discards" },
+						8, "[%d]: rx_bcast_packets" },
+	{ Q_STATS_OFFSET32(no_buff_discard_hi),	8, "[%d]: rx_discards" },
 	{ Q_STATS_OFFSET32(rx_err_discard_pkt),
-					 4, "[%s]: rx_phy_ip_err_discards"},
+					 4, "[%d]: rx_phy_ip_err_discards"},
 	{ Q_STATS_OFFSET32(rx_skb_alloc_failed),
-					 4, "[%s]: rx_skb_alloc_discard" },
-	{ Q_STATS_OFFSET32(hw_csum_err), 4, "[%s]: rx_csum_offload_errors" },
-	{ Q_STATS_OFFSET32(driver_xoff), 4, "[%s]: tx_exhaustion_events" },
-	{ Q_STATS_OFFSET32(total_bytes_transmitted_hi),	8, "[%s]: tx_bytes" },
+					 4, "[%d]: rx_skb_alloc_discard" },
+	{ Q_STATS_OFFSET32(hw_csum_err), 4, "[%d]: rx_csum_offload_errors" },
+	{ Q_STATS_OFFSET32(driver_xoff), 4, "[%d]: tx_exhaustion_events" },
+	{ Q_STATS_OFFSET32(total_bytes_transmitted_hi),	8, "[%d]: tx_bytes" },
 /* 10 */{ Q_STATS_OFFSET32(total_unicast_packets_transmitted_hi),
-						8, "[%s]: tx_ucast_packets" },
+						8, "[%d]: tx_ucast_packets" },
 	{ Q_STATS_OFFSET32(total_multicast_packets_transmitted_hi),
-						8, "[%s]: tx_mcast_packets" },
+						8, "[%d]: tx_mcast_packets" },
 	{ Q_STATS_OFFSET32(total_broadcast_packets_transmitted_hi),
-						8, "[%s]: tx_bcast_packets" },
+						8, "[%d]: tx_bcast_packets" },
 	{ Q_STATS_OFFSET32(total_tpa_aggregations_hi),
-						8, "[%s]: tpa_aggregations" },
+						8, "[%d]: tpa_aggregations" },
 	{ Q_STATS_OFFSET32(total_tpa_aggregated_frames_hi),
-					8, "[%s]: tpa_aggregated_frames"},
-	{ Q_STATS_OFFSET32(total_tpa_bytes_hi),	8, "[%s]: tpa_bytes"},
+					8, "[%d]: tpa_aggregated_frames"},
+	{ Q_STATS_OFFSET32(total_tpa_bytes_hi),	8, "[%d]: tpa_bytes"},
 	{ Q_STATS_OFFSET32(driver_filtered_tx_pkt),
-					4, "[%s]: driver_filtered_tx_pkt" }
+					4, "[%d]: driver_filtered_tx_pkt" }
 };
 
 #define BNX2X_NUM_Q_STATS ARRAY_SIZE(bnx2x_q_stats_arr)
@@ -182,7 +182,9 @@ static const struct {
 	{ STATS_OFFSET32(driver_filtered_tx_pkt),
 				4, false, "driver_filtered_tx_pkt" },
 	{ STATS_OFFSET32(eee_tx_lpi),
-				4, true, "Tx LPI entry count"}
+				4, true, "Tx LPI entry count"},
+	{ STATS_OFFSET32(ptp_skip_tx_ts),
+				4, false, "ptp_skipped_tx_tstamp" },
 };
 
 #define BNX2X_NUM_STATS		ARRAY_SIZE(bnx2x_stats_arr)
@@ -837,8 +839,9 @@ static bool bnx2x_is_wreg_in_chip(struct bnx2x *bp,
 /**
  * bnx2x_read_pages_regs - read "paged" registers
  *
- * @bp		device handle
- * @p		output buffer
+ * @bp:		device handle
+ * @p:		output buffer
+ * @preset:	the preset value
  *
  * Reads "paged" memories: memories that may only be read by first writing to a
  * specific address ("write address") and then reading from a specific address
@@ -1105,13 +1108,34 @@ static void bnx2x_get_drvinfo(struct net_device *dev,
 			      struct ethtool_drvinfo *info)
 {
 	struct bnx2x *bp = netdev_priv(dev);
+	char version[ETHTOOL_FWVERS_LEN];
+	int ext_dev_info_offset;
+	u32 mbi;
+
+	strscpy(info->driver, DRV_MODULE_NAME, sizeof(info->driver));
+
+	if (SHMEM2_HAS(bp, extended_dev_info_shared_addr)) {
+		ext_dev_info_offset = SHMEM2_RD(bp,
+						extended_dev_info_shared_addr);
+		mbi = REG_RD(bp, ext_dev_info_offset +
+			     offsetof(struct extended_dev_info_shared_cfg,
+				      mbi_version));
+		if (mbi) {
+			memset(version, 0, sizeof(version));
+			snprintf(version, ETHTOOL_FWVERS_LEN, "mbi %d.%d.%d ",
+				 (mbi & 0xff000000) >> 24,
+				 (mbi & 0x00ff0000) >> 16,
+				 (mbi & 0x0000ff00) >> 8);
+			strscpy(info->fw_version, version,
+				sizeof(info->fw_version));
+		}
+	}
 
-	strlcpy(info->driver, DRV_MODULE_NAME, sizeof(info->driver));
-	strlcpy(info->version, DRV_MODULE_VERSION, sizeof(info->version));
-
-	bnx2x_fill_fw_str(bp, info->fw_version, sizeof(info->fw_version));
+	memset(version, 0, sizeof(version));
+	bnx2x_fill_fw_str(bp, version, sizeof(version));
+	strlcat(info->fw_version, version, sizeof(info->fw_version));
 
-	strlcpy(info->bus_info, pci_name(bp->pdev), sizeof(info->bus_info));
+	strscpy(info->bus_info, pci_name(bp->pdev), sizeof(info->bus_info));
 }
 
 static void bnx2x_get_wol(struct net_device *dev, struct ethtool_wolinfo *wol)
@@ -1219,9 +1243,9 @@ static int bnx2x_get_eeprom_len(struct net_device *dev)
  * pf B succeeds in taking the same lock since they are from the same port.
  * pf A takes the per pf misc lock. Performs eeprom access.
  * pf A finishes. Unlocks the per pf misc lock.
- * Pf B takes the lock and proceeds to perform it's own access.
+ * Pf B takes the lock and proceeds to perform its own access.
  * pf A unlocks the per port lock, while pf B is still working (!).
- * mcp takes the per port lock and corrupts pf B's access (and/or has it's own
+ * mcp takes the per port lock and corrupts pf B's access (and/or has its own
  * access corrupted by pf B)
  */
 static int bnx2x_acquire_nvram_lock(struct bnx2x *bp)
@@ -1581,7 +1605,8 @@ static int bnx2x_get_module_info(struct net_device *dev,
 	}
 
 	if (!sff8472_comp ||
-	    (diag_type & SFP_EEPROM_DIAG_ADDR_CHANGE_REQ)) {
+	    (diag_type & SFP_EEPROM_DIAG_ADDR_CHANGE_REQ) ||
+	    !(diag_type & SFP_EEPROM_DDM_IMPLEMENTED)) {
 		modinfo->type = ETH_MODULE_SFF_8079;
 		modinfo->eeprom_len = ETH_MODULE_SFF_8079_LEN;
 	} else {
@@ -1853,7 +1878,9 @@ static int bnx2x_set_eeprom(struct net_device *dev,
 }
 
 static int bnx2x_get_coalesce(struct net_device *dev,
-			      struct ethtool_coalesce *coal)
+			      struct ethtool_coalesce *coal,
+			      struct kernel_ethtool_coalesce *kernel_coal,
+			      struct netlink_ext_ack *extack)
 {
 	struct bnx2x *bp = netdev_priv(dev);
 
@@ -1866,7 +1893,9 @@ static int bnx2x_get_coalesce(struct net_device *dev,
 }
 
 static int bnx2x_set_coalesce(struct net_device *dev,
-			      struct ethtool_coalesce *coal)
+			      struct ethtool_coalesce *coal,
+			      struct kernel_ethtool_coalesce *kernel_coal,
+			      struct netlink_ext_ack *extack)
 {
 	struct bnx2x *bp = netdev_priv(dev);
 
@@ -1885,7 +1914,9 @@ static int bnx2x_set_coalesce(struct net_device *dev,
 }
 
 static void bnx2x_get_ringparam(struct net_device *dev,
-				struct ethtool_ringparam *ering)
+				struct ethtool_ringparam *ering,
+				struct kernel_ethtool_ringparam *kernel_ering,
+				struct netlink_ext_ack *extack)
 {
 	struct bnx2x *bp = netdev_priv(dev);
 
@@ -1909,7 +1940,9 @@ static void bnx2x_get_ringparam(struct net_device *dev,
 }
 
 static int bnx2x_set_ringparam(struct net_device *dev,
-			       struct ethtool_ringparam *ering)
+			       struct ethtool_ringparam *ering,
+			       struct kernel_ethtool_ringparam *kernel_ering,
+			       struct netlink_ext_ack *extack)
 {
 	struct bnx2x *bp = netdev_priv(dev);
 
@@ -2048,34 +2081,31 @@ static const char bnx2x_private_arr[BNX2X_PRI_FLAG_LEN][ETH_GSTRING_LEN] = {
 	"Storage only interface"
 };
 
-static u32 bnx2x_eee_to_adv(u32 eee_adv)
+static void bnx2x_eee_to_linkmode(unsigned long *mode, u32 eee_adv)
 {
-	u32 modes = 0;
-
 	if (eee_adv & SHMEM_EEE_100M_ADV)
-		modes |= ADVERTISED_100baseT_Full;
+		linkmode_set_bit(ETHTOOL_LINK_MODE_100baseT_Full_BIT, mode);
 	if (eee_adv & SHMEM_EEE_1G_ADV)
-		modes |= ADVERTISED_1000baseT_Full;
+		linkmode_set_bit(ETHTOOL_LINK_MODE_1000baseT_Full_BIT, mode);
 	if (eee_adv & SHMEM_EEE_10G_ADV)
-		modes |= ADVERTISED_10000baseT_Full;
-
-	return modes;
+		linkmode_set_bit(ETHTOOL_LINK_MODE_10000baseT_Full_BIT, mode);
 }
 
-static u32 bnx2x_adv_to_eee(u32 modes, u32 shift)
+static u32 bnx2x_linkmode_to_eee(const unsigned long *mode, u32 shift)
 {
 	u32 eee_adv = 0;
-	if (modes & ADVERTISED_100baseT_Full)
+
+	if (linkmode_test_bit(ETHTOOL_LINK_MODE_100baseT_Full_BIT, mode))
 		eee_adv |= SHMEM_EEE_100M_ADV;
-	if (modes & ADVERTISED_1000baseT_Full)
+	if (linkmode_test_bit(ETHTOOL_LINK_MODE_1000baseT_Full_BIT, mode))
 		eee_adv |= SHMEM_EEE_1G_ADV;
-	if (modes & ADVERTISED_10000baseT_Full)
+	if (linkmode_test_bit(ETHTOOL_LINK_MODE_10000baseT_Full_BIT, mode))
 		eee_adv |= SHMEM_EEE_10G_ADV;
 
 	return eee_adv << shift;
 }
 
-static int bnx2x_get_eee(struct net_device *dev, struct ethtool_eee *edata)
+static int bnx2x_get_eee(struct net_device *dev, struct ethtool_keee *edata)
 {
 	struct bnx2x *bp = netdev_priv(dev);
 	u32 eee_cfg;
@@ -2087,16 +2117,17 @@ static int bnx2x_get_eee(struct net_device *dev, struct ethtool_eee *edata)
 
 	eee_cfg = bp->link_vars.eee_status;
 
-	edata->supported =
-		bnx2x_eee_to_adv((eee_cfg & SHMEM_EEE_SUPPORTED_MASK) >>
-				 SHMEM_EEE_SUPPORTED_SHIFT);
+	bnx2x_eee_to_linkmode(edata->supported,
+			      (eee_cfg & SHMEM_EEE_SUPPORTED_MASK) >>
+			      SHMEM_EEE_SUPPORTED_SHIFT);
+
+	bnx2x_eee_to_linkmode(edata->advertised,
+			      (eee_cfg & SHMEM_EEE_ADV_STATUS_MASK) >>
+			      SHMEM_EEE_ADV_STATUS_SHIFT);
 
-	edata->advertised =
-		bnx2x_eee_to_adv((eee_cfg & SHMEM_EEE_ADV_STATUS_MASK) >>
-				 SHMEM_EEE_ADV_STATUS_SHIFT);
-	edata->lp_advertised =
-		bnx2x_eee_to_adv((eee_cfg & SHMEM_EEE_LP_ADV_STATUS_MASK) >>
-				 SHMEM_EEE_LP_ADV_STATUS_SHIFT);
+	bnx2x_eee_to_linkmode(edata->lp_advertised,
+			      (eee_cfg & SHMEM_EEE_LP_ADV_STATUS_MASK) >>
+			      SHMEM_EEE_LP_ADV_STATUS_SHIFT);
 
 	/* SHMEM value is in 16u units --> Convert to 1u units. */
 	edata->tx_lpi_timer = (eee_cfg & SHMEM_EEE_TIMER_MASK) << 4;
@@ -2108,7 +2139,7 @@ static int bnx2x_get_eee(struct net_device *dev, struct ethtool_eee *edata)
 	return 0;
 }
 
-static int bnx2x_set_eee(struct net_device *dev, struct ethtool_eee *edata)
+static int bnx2x_set_eee(struct net_device *dev, struct ethtool_keee *edata)
 {
 	struct bnx2x *bp = netdev_priv(dev);
 	u32 eee_cfg;
@@ -2129,8 +2160,8 @@ static int bnx2x_set_eee(struct net_device *dev, struct ethtool_eee *edata)
 		return -EOPNOTSUPP;
 	}
 
-	advertised = bnx2x_adv_to_eee(edata->advertised,
-				      SHMEM_EEE_ADV_STATUS_SHIFT);
+	advertised = bnx2x_linkmode_to_eee(edata->advertised,
+					   SHMEM_EEE_ADV_STATUS_SHIFT);
 	if ((advertised != (eee_cfg & SHMEM_EEE_ADV_STATUS_MASK))) {
 		DP(BNX2X_MSG_ETHTOOL,
 		   "Direct manipulation of EEE advertisement is not supported\n");
@@ -2591,10 +2622,10 @@ static int bnx2x_run_loopback(struct bnx2x *bp, int loopback_mode)
 	wmb();
 
 	txdata->tx_db.data.prod += 2;
-	barrier();
-	DOORBELL(bp, txdata->cid, txdata->tx_db.raw);
+	/* make sure descriptor update is observed by the HW */
+	wmb();
+	DOORBELL_RELAXED(bp, txdata->cid, txdata->tx_db.raw);
 
-	mmiowb();
 	barrier();
 
 	num_pkts++;
@@ -2886,7 +2917,7 @@ static int bnx2x_test_nvram_tbl(struct bnx2x *bp,
 
 static int bnx2x_test_nvram(struct bnx2x *bp)
 {
-	const struct crc_pair nvram_tbl[] = {
+	static const struct crc_pair nvram_tbl[] = {
 		{     0,  0x14 }, /* bootstrap */
 		{  0x14,  0xec }, /* dir */
 		{ 0x100, 0x350 }, /* manuf_info */
@@ -2895,7 +2926,7 @@ static int bnx2x_test_nvram(struct bnx2x *bp)
 		{ 0x708,  0x70 }, /* manuf_key_info */
 		{     0,     0 }
 	};
-	const struct crc_pair nvram_tbl2[] = {
+	static const struct crc_pair nvram_tbl2[] = {
 		{ 0x7e8, 0x350 }, /* manuf_info2 */
 		{ 0xb38,  0xf0 }, /* feature_info */
 		{     0,     0 }
@@ -3153,48 +3184,43 @@ static u32 bnx2x_get_private_flags(struct net_device *dev)
 static void bnx2x_get_strings(struct net_device *dev, u32 stringset, u8 *buf)
 {
 	struct bnx2x *bp = netdev_priv(dev);
-	int i, j, k, start;
-	char queue_name[MAX_QUEUE_NAME_LEN+1];
+	const char *str;
+	int i, j, start;
 
 	switch (stringset) {
 	case ETH_SS_STATS:
-		k = 0;
 		if (is_multi(bp)) {
 			for_each_eth_queue(bp, i) {
-				memset(queue_name, 0, sizeof(queue_name));
-				sprintf(queue_name, "%d", i);
-				for (j = 0; j < BNX2X_NUM_Q_STATS; j++)
-					snprintf(buf + (k + j)*ETH_GSTRING_LEN,
-						ETH_GSTRING_LEN,
-						bnx2x_q_stats_arr[j].string,
-						queue_name);
-				k += BNX2X_NUM_Q_STATS;
+				for (j = 0; j < BNX2X_NUM_Q_STATS; j++) {
+					str = bnx2x_q_stats_arr[j].string;
+					ethtool_sprintf(&buf, str, i);
+				}
 			}
 		}
 
-		for (i = 0, j = 0; i < BNX2X_NUM_STATS; i++) {
+		for (i = 0; i < BNX2X_NUM_STATS; i++) {
 			if (HIDE_PORT_STAT(bp) && IS_PORT_STAT(i))
 				continue;
-			strcpy(buf + (k + j)*ETH_GSTRING_LEN,
-				   bnx2x_stats_arr[i].string);
-			j++;
+			ethtool_puts(&buf, bnx2x_stats_arr[i].string);
 		}
 
 		break;
 
 	case ETH_SS_TEST:
+		if (IS_VF(bp))
+			break;
 		/* First 4 tests cannot be done in MF mode */
 		if (!IS_MF(bp))
 			start = 0;
 		else
 			start = 4;
-		memcpy(buf, bnx2x_tests_str_arr + start,
-		       ETH_GSTRING_LEN * BNX2X_NUM_TESTS(bp));
+		for (i = start; i < BNX2X_NUM_TESTS_SF; i++)
+			ethtool_puts(&buf, bnx2x_tests_str_arr[i]);
 		break;
 
 	case ETH_SS_PRIV_FLAGS:
-		memcpy(buf, bnx2x_private_arr,
-		       ETH_GSTRING_LEN * BNX2X_PRI_FLAG_LEN);
+		for (i = 0; i < BNX2X_PRI_FLAG_LEN; i++)
+			ethtool_puts(&buf, bnx2x_private_arr[i]);
 		break;
 	}
 }
@@ -3292,8 +3318,11 @@ static int bnx2x_set_phys_id(struct net_device *dev,
 	return 0;
 }
 
-static int bnx2x_get_rss_flags(struct bnx2x *bp, struct ethtool_rxnfc *info)
+static int bnx2x_get_rxfh_fields(struct net_device *dev,
+				 struct ethtool_rxfh_fields *info)
 {
+	struct bnx2x *bp = netdev_priv(dev);
+
 	switch (info->flow_type) {
 	case TCP_V4_FLOW:
 	case TCP_V6_FLOW:
@@ -3335,20 +3364,21 @@ static int bnx2x_get_rxnfc(struct net_device *dev, struct ethtool_rxnfc *info,
 	case ETHTOOL_GRXRINGS:
 		info->data = BNX2X_NUM_ETH_QUEUES(bp);
 		return 0;
-	case ETHTOOL_GRXFH:
-		return bnx2x_get_rss_flags(bp, info);
 	default:
 		DP(BNX2X_MSG_ETHTOOL, "Command parameters not supported\n");
 		return -EOPNOTSUPP;
 	}
 }
 
-static int bnx2x_set_rss_flags(struct bnx2x *bp, struct ethtool_rxnfc *info)
+static int bnx2x_set_rxfh_fields(struct net_device *dev,
+				 const struct ethtool_rxfh_fields *info,
+				 struct netlink_ext_ack *extack)
 {
+	struct bnx2x *bp = netdev_priv(dev);
 	int udp_rss_requested;
 
 	DP(BNX2X_MSG_ETHTOOL,
-	   "Set rss flags command parameters: flow type = %d, data = %llu\n",
+	   "Set rss flags command parameters: flow type = %d, data = %u\n",
 	   info->flow_type, info->data);
 
 	switch (info->flow_type) {
@@ -3386,14 +3416,18 @@ static int bnx2x_set_rss_flags(struct bnx2x *bp, struct ethtool_rxnfc *info)
 			DP(BNX2X_MSG_ETHTOOL,
 			   "rss re-configured, UDP 4-tupple %s\n",
 			   udp_rss_requested ? "enabled" : "disabled");
-			return bnx2x_rss(bp, &bp->rss_conf_obj, false, true);
+			if (bp->state == BNX2X_STATE_OPEN)
+				return bnx2x_rss(bp, &bp->rss_conf_obj, false,
+						 true);
 		} else if ((info->flow_type == UDP_V6_FLOW) &&
 			   (bp->rss_conf_obj.udp_rss_v6 != udp_rss_requested)) {
 			bp->rss_conf_obj.udp_rss_v6 = udp_rss_requested;
 			DP(BNX2X_MSG_ETHTOOL,
 			   "rss re-configured, UDP 4-tupple %s\n",
 			   udp_rss_requested ? "enabled" : "disabled");
-			return bnx2x_rss(bp, &bp->rss_conf_obj, false, true);
+			if (bp->state == BNX2X_STATE_OPEN)
+				return bnx2x_rss(bp, &bp->rss_conf_obj, false,
+						 true);
 		}
 		return 0;
 
@@ -3430,34 +3464,20 @@ static int bnx2x_set_rss_flags(struct bnx2x *bp, struct ethtool_rxnfc *info)
 	}
 }
 
-static int bnx2x_set_rxnfc(struct net_device *dev, struct ethtool_rxnfc *info)
-{
-	struct bnx2x *bp = netdev_priv(dev);
-
-	switch (info->cmd) {
-	case ETHTOOL_SRXFH:
-		return bnx2x_set_rss_flags(bp, info);
-	default:
-		DP(BNX2X_MSG_ETHTOOL, "Command parameters not supported\n");
-		return -EOPNOTSUPP;
-	}
-}
-
 static u32 bnx2x_get_rxfh_indir_size(struct net_device *dev)
 {
 	return T_ETH_INDIRECTION_TABLE_SIZE;
 }
 
-static int bnx2x_get_rxfh(struct net_device *dev, u32 *indir, u8 *key,
-			  u8 *hfunc)
+static int bnx2x_get_rxfh(struct net_device *dev,
+			  struct ethtool_rxfh_param *rxfh)
 {
 	struct bnx2x *bp = netdev_priv(dev);
 	u8 ind_table[T_ETH_INDIRECTION_TABLE_SIZE] = {0};
 	size_t i;
 
-	if (hfunc)
-		*hfunc = ETH_RSS_HASH_TOP;
-	if (!indir)
+	rxfh->hfunc = ETH_RSS_HASH_TOP;
+	if (!rxfh->indir)
 		return 0;
 
 	/* Get the current configuration of the RSS indirection table */
@@ -3473,13 +3493,14 @@ static int bnx2x_get_rxfh(struct net_device *dev, u32 *indir, u8 *key,
 	 * queue.
 	 */
 	for (i = 0; i < T_ETH_INDIRECTION_TABLE_SIZE; i++)
-		indir[i] = ind_table[i] - bp->fp->cl_id;
+		rxfh->indir[i] = ind_table[i] - bp->fp->cl_id;
 
 	return 0;
 }
 
-static int bnx2x_set_rxfh(struct net_device *dev, const u32 *indir,
-			  const u8 *key, const u8 hfunc)
+static int bnx2x_set_rxfh(struct net_device *dev,
+			  struct ethtool_rxfh_param *rxfh,
+			  struct netlink_ext_ack *extack)
 {
 	struct bnx2x *bp = netdev_priv(dev);
 	size_t i;
@@ -3487,11 +3508,12 @@ static int bnx2x_set_rxfh(struct net_device *dev, const u32 *indir,
 	/* We require at least one supported parameter to be changed and no
 	 * change in any of the unsupported parameters
 	 */
-	if (key ||
-	    (hfunc != ETH_RSS_HASH_NO_CHANGE && hfunc != ETH_RSS_HASH_TOP))
+	if (rxfh->key ||
+	    (rxfh->hfunc != ETH_RSS_HASH_NO_CHANGE &&
+	     rxfh->hfunc != ETH_RSS_HASH_TOP))
 		return -EOPNOTSUPP;
 
-	if (!indir)
+	if (!rxfh->indir)
 		return 0;
 
 	for (i = 0; i < T_ETH_INDIRECTION_TABLE_SIZE; i++) {
@@ -3504,10 +3526,13 @@ static int bnx2x_set_rxfh(struct net_device *dev, const u32 *indir,
 		 * align the received table to the Client ID of the leading RSS
 		 * queue
 		 */
-		bp->rss_conf_obj.ind_table[i] = indir[i] + bp->fp->cl_id;
+		bp->rss_conf_obj.ind_table[i] = rxfh->indir[i] + bp->fp->cl_id;
 	}
 
-	return bnx2x_config_rss_eth(bp, false);
+	if (bp->state == BNX2X_STATE_OPEN)
+		return bnx2x_config_rss_eth(bp, false);
+
+	return 0;
 }
 
 /**
@@ -3529,6 +3554,7 @@ static void bnx2x_get_channels(struct net_device *dev,
  * bnx2x_change_num_queues - change the number of RSS queues.
  *
  * @bp:			bnx2x private structure
+ * @num_rss:		rss count
  *
  * Re-configure interrupt mode to get the new number of MSI-X
  * vectors and re-add NAPI objects.
@@ -3593,22 +3619,18 @@ static int bnx2x_set_channels(struct net_device *dev,
 }
 
 static int bnx2x_get_ts_info(struct net_device *dev,
-			     struct ethtool_ts_info *info)
+			     struct kernel_ethtool_ts_info *info)
 {
 	struct bnx2x *bp = netdev_priv(dev);
 
 	if (bp->flags & PTP_SUPPORTED) {
 		info->so_timestamping = SOF_TIMESTAMPING_TX_SOFTWARE |
-					SOF_TIMESTAMPING_RX_SOFTWARE |
-					SOF_TIMESTAMPING_SOFTWARE |
 					SOF_TIMESTAMPING_TX_HARDWARE |
 					SOF_TIMESTAMPING_RX_HARDWARE |
 					SOF_TIMESTAMPING_RAW_HARDWARE;
 
 		if (bp->ptp_clock)
 			info->phc_index = ptp_clock_index(bp->ptp_clock);
-		else
-			info->phc_index = -1;
 
 		info->rx_filters = (1 << HWTSTAMP_FILTER_NONE) |
 				   (1 << HWTSTAMP_FILTER_PTP_V1_L4_EVENT) |
@@ -3624,6 +3646,7 @@ static int bnx2x_get_ts_info(struct net_device *dev,
 }
 
 static const struct ethtool_ops bnx2x_ethtool_ops = {
+	.supported_coalesce_params = ETHTOOL_COALESCE_USECS,
 	.get_drvinfo		= bnx2x_get_drvinfo,
 	.get_regs_len		= bnx2x_get_regs_len,
 	.get_regs		= bnx2x_get_regs,
@@ -3652,10 +3675,11 @@ static const struct ethtool_ops bnx2x_ethtool_ops = {
 	.set_phys_id		= bnx2x_set_phys_id,
 	.get_ethtool_stats	= bnx2x_get_ethtool_stats,
 	.get_rxnfc		= bnx2x_get_rxnfc,
-	.set_rxnfc		= bnx2x_set_rxnfc,
 	.get_rxfh_indir_size	= bnx2x_get_rxfh_indir_size,
 	.get_rxfh		= bnx2x_get_rxfh,
 	.set_rxfh		= bnx2x_set_rxfh,
+	.get_rxfh_fields	= bnx2x_get_rxfh_fields,
+	.set_rxfh_fields	= bnx2x_set_rxfh_fields,
 	.get_channels		= bnx2x_get_channels,
 	.set_channels		= bnx2x_set_channels,
 	.get_module_info	= bnx2x_get_module_info,
@@ -3679,10 +3703,11 @@ static const struct ethtool_ops bnx2x_vf_ethtool_ops = {
 	.get_strings		= bnx2x_get_strings,
 	.get_ethtool_stats	= bnx2x_get_ethtool_stats,
 	.get_rxnfc		= bnx2x_get_rxnfc,
-	.set_rxnfc		= bnx2x_set_rxnfc,
 	.get_rxfh_indir_size	= bnx2x_get_rxfh_indir_size,
 	.get_rxfh		= bnx2x_get_rxfh,
 	.set_rxfh		= bnx2x_set_rxfh,
+	.get_rxfh_fields	= bnx2x_get_rxfh_fields,
+	.set_rxfh_fields	= bnx2x_set_rxfh_fields,
 	.get_channels		= bnx2x_get_channels,
 	.set_channels		= bnx2x_set_channels,
 	.get_link_ksettings	= bnx2x_get_vf_link_ksettings,
diff --git a/drivers/net/ethernet/broadcom/bnx2x/bnx2x_fw_defs.h b/drivers/net/ethernet/broadcom/bnx2x/bnx2x_fw_defs.h
index 226ab29f4cb6..9221942290a8 100644
--- a/drivers/net/ethernet/broadcom/bnx2x/bnx2x_fw_defs.h
+++ b/drivers/net/ethernet/broadcom/bnx2x/bnx2x_fw_defs.h
@@ -32,31 +32,31 @@
 	* IRO[142].m2) + ((sbId) * IRO[142].m3))
 #define CSTORM_IGU_MODE_OFFSET (IRO[161].base)
 #define CSTORM_ISCSI_CQ_SIZE_OFFSET(pfId) \
-	(IRO[323].base + ((pfId) * IRO[323].m1))
-#define CSTORM_ISCSI_CQ_SQN_SIZE_OFFSET(pfId) \
 	(IRO[324].base + ((pfId) * IRO[324].m1))
+#define CSTORM_ISCSI_CQ_SQN_SIZE_OFFSET(pfId) \
+	(IRO[325].base + ((pfId) * IRO[325].m1))
 #define CSTORM_ISCSI_EQ_CONS_OFFSET(pfId, iscsiEqId) \
-	(IRO[316].base + ((pfId) * IRO[316].m1) + ((iscsiEqId) * IRO[316].m2))
+	(IRO[317].base + ((pfId) * IRO[317].m1) + ((iscsiEqId) * IRO[317].m2))
 #define CSTORM_ISCSI_EQ_NEXT_EQE_ADDR_OFFSET(pfId, iscsiEqId) \
-	(IRO[318].base + ((pfId) * IRO[318].m1) + ((iscsiEqId) * IRO[318].m2))
+	(IRO[319].base + ((pfId) * IRO[319].m1) + ((iscsiEqId) * IRO[319].m2))
 #define CSTORM_ISCSI_EQ_NEXT_PAGE_ADDR_OFFSET(pfId, iscsiEqId) \
-	(IRO[317].base + ((pfId) * IRO[317].m1) + ((iscsiEqId) * IRO[317].m2))
+	(IRO[318].base + ((pfId) * IRO[318].m1) + ((iscsiEqId) * IRO[318].m2))
 #define CSTORM_ISCSI_EQ_NEXT_PAGE_ADDR_VALID_OFFSET(pfId, iscsiEqId) \
-	(IRO[319].base + ((pfId) * IRO[319].m1) + ((iscsiEqId) * IRO[319].m2))
+	(IRO[320].base + ((pfId) * IRO[320].m1) + ((iscsiEqId) * IRO[320].m2))
 #define CSTORM_ISCSI_EQ_PROD_OFFSET(pfId, iscsiEqId) \
-	(IRO[315].base + ((pfId) * IRO[315].m1) + ((iscsiEqId) * IRO[315].m2))
+	(IRO[316].base + ((pfId) * IRO[316].m1) + ((iscsiEqId) * IRO[316].m2))
 #define CSTORM_ISCSI_EQ_SB_INDEX_OFFSET(pfId, iscsiEqId) \
-	(IRO[321].base + ((pfId) * IRO[321].m1) + ((iscsiEqId) * IRO[321].m2))
+	(IRO[322].base + ((pfId) * IRO[322].m1) + ((iscsiEqId) * IRO[322].m2))
 #define CSTORM_ISCSI_EQ_SB_NUM_OFFSET(pfId, iscsiEqId) \
-	(IRO[320].base + ((pfId) * IRO[320].m1) + ((iscsiEqId) * IRO[320].m2))
+	(IRO[321].base + ((pfId) * IRO[321].m1) + ((iscsiEqId) * IRO[321].m2))
 #define CSTORM_ISCSI_HQ_SIZE_OFFSET(pfId) \
-	(IRO[322].base + ((pfId) * IRO[322].m1))
+	(IRO[323].base + ((pfId) * IRO[323].m1))
 #define CSTORM_ISCSI_NUM_OF_TASKS_OFFSET(pfId) \
-	(IRO[314].base + ((pfId) * IRO[314].m1))
+	(IRO[315].base + ((pfId) * IRO[315].m1))
 #define CSTORM_ISCSI_PAGE_SIZE_LOG_OFFSET(pfId) \
-	(IRO[313].base + ((pfId) * IRO[313].m1))
+	(IRO[314].base + ((pfId) * IRO[314].m1))
 #define CSTORM_ISCSI_PAGE_SIZE_OFFSET(pfId) \
-	(IRO[312].base + ((pfId) * IRO[312].m1))
+	(IRO[313].base + ((pfId) * IRO[313].m1))
 #define CSTORM_RECORD_SLOW_PATH_OFFSET(funcId) \
 	(IRO[155].base + ((funcId) * IRO[155].m1))
 #define CSTORM_SP_STATUS_BLOCK_DATA_OFFSET(pfId) \
@@ -99,81 +99,81 @@
 #define TSTORM_FUNC_EN_OFFSET(funcId) \
 	(IRO[107].base + ((funcId) * IRO[107].m1))
 #define TSTORM_ISCSI_ERROR_BITMAP_OFFSET(pfId) \
-	(IRO[278].base + ((pfId) * IRO[278].m1))
-#define TSTORM_ISCSI_L2_ISCSI_OOO_CID_TABLE_OFFSET(pfId) \
 	(IRO[279].base + ((pfId) * IRO[279].m1))
-#define TSTORM_ISCSI_L2_ISCSI_OOO_CLIENT_ID_TABLE_OFFSET(pfId) \
+#define TSTORM_ISCSI_L2_ISCSI_OOO_CID_TABLE_OFFSET(pfId) \
 	(IRO[280].base + ((pfId) * IRO[280].m1))
-#define TSTORM_ISCSI_L2_ISCSI_OOO_PROD_OFFSET(pfId) \
+#define TSTORM_ISCSI_L2_ISCSI_OOO_CLIENT_ID_TABLE_OFFSET(pfId) \
 	(IRO[281].base + ((pfId) * IRO[281].m1))
+#define TSTORM_ISCSI_L2_ISCSI_OOO_PROD_OFFSET(pfId) \
+	(IRO[282].base + ((pfId) * IRO[282].m1))
 #define TSTORM_ISCSI_NUM_OF_TASKS_OFFSET(pfId) \
-	(IRO[277].base + ((pfId) * IRO[277].m1))
+	(IRO[278].base + ((pfId) * IRO[278].m1))
 #define TSTORM_ISCSI_PAGE_SIZE_LOG_OFFSET(pfId) \
-	(IRO[276].base + ((pfId) * IRO[276].m1))
+	(IRO[277].base + ((pfId) * IRO[277].m1))
 #define TSTORM_ISCSI_PAGE_SIZE_OFFSET(pfId) \
-	(IRO[275].base + ((pfId) * IRO[275].m1))
+	(IRO[276].base + ((pfId) * IRO[276].m1))
 #define TSTORM_ISCSI_RQ_SIZE_OFFSET(pfId) \
-	(IRO[274].base + ((pfId) * IRO[274].m1))
+	(IRO[275].base + ((pfId) * IRO[275].m1))
 #define TSTORM_ISCSI_TCP_LOCAL_ADV_WND_OFFSET(pfId) \
-	(IRO[284].base + ((pfId) * IRO[284].m1))
+	(IRO[285].base + ((pfId) * IRO[285].m1))
 #define TSTORM_ISCSI_TCP_VARS_FLAGS_OFFSET(pfId) \
-	(IRO[270].base + ((pfId) * IRO[270].m1))
-#define TSTORM_ISCSI_TCP_VARS_LSB_LOCAL_MAC_ADDR_OFFSET(pfId) \
 	(IRO[271].base + ((pfId) * IRO[271].m1))
-#define TSTORM_ISCSI_TCP_VARS_MID_LOCAL_MAC_ADDR_OFFSET(pfId) \
+#define TSTORM_ISCSI_TCP_VARS_LSB_LOCAL_MAC_ADDR_OFFSET(pfId) \
 	(IRO[272].base + ((pfId) * IRO[272].m1))
-#define TSTORM_ISCSI_TCP_VARS_MSB_LOCAL_MAC_ADDR_OFFSET(pfId) \
+#define TSTORM_ISCSI_TCP_VARS_MID_LOCAL_MAC_ADDR_OFFSET(pfId) \
 	(IRO[273].base + ((pfId) * IRO[273].m1))
+#define TSTORM_ISCSI_TCP_VARS_MSB_LOCAL_MAC_ADDR_OFFSET(pfId) \
+	(IRO[274].base + ((pfId) * IRO[274].m1))
 #define TSTORM_MAC_FILTER_CONFIG_OFFSET(pfId) \
 	(IRO[206].base + ((pfId) * IRO[206].m1))
 #define TSTORM_RECORD_SLOW_PATH_OFFSET(funcId) \
 	(IRO[109].base + ((funcId) * IRO[109].m1))
 #define TSTORM_TCP_MAX_CWND_OFFSET(pfId) \
-	(IRO[223].base + ((pfId) * IRO[223].m1))
+	(IRO[224].base + ((pfId) * IRO[224].m1))
 #define TSTORM_VF_TO_PF_OFFSET(funcId) \
 	(IRO[108].base + ((funcId) * IRO[108].m1))
-#define USTORM_AGG_DATA_OFFSET (IRO[212].base)
-#define USTORM_AGG_DATA_SIZE (IRO[212].size)
+#define USTORM_AGG_DATA_OFFSET (IRO[213].base)
+#define USTORM_AGG_DATA_SIZE (IRO[213].size)
 #define USTORM_ASSERT_LIST_INDEX_OFFSET	(IRO[181].base)
 #define USTORM_ASSERT_LIST_OFFSET(assertListEntry) \
 	(IRO[180].base + ((assertListEntry) * IRO[180].m1))
 #define USTORM_ETH_PAUSE_ENABLED_OFFSET(portId) \
 	(IRO[187].base + ((portId) * IRO[187].m1))
 #define USTORM_FCOE_EQ_PROD_OFFSET(pfId) \
-	(IRO[325].base + ((pfId) * IRO[325].m1))
+	(IRO[326].base + ((pfId) * IRO[326].m1))
 #define USTORM_FUNC_EN_OFFSET(funcId) \
 	(IRO[182].base + ((funcId) * IRO[182].m1))
 #define USTORM_ISCSI_CQ_SIZE_OFFSET(pfId) \
-	(IRO[289].base + ((pfId) * IRO[289].m1))
-#define USTORM_ISCSI_CQ_SQN_SIZE_OFFSET(pfId) \
 	(IRO[290].base + ((pfId) * IRO[290].m1))
+#define USTORM_ISCSI_CQ_SQN_SIZE_OFFSET(pfId) \
+	(IRO[291].base + ((pfId) * IRO[291].m1))
 #define USTORM_ISCSI_ERROR_BITMAP_OFFSET(pfId) \
-	(IRO[294].base + ((pfId) * IRO[294].m1))
+	(IRO[295].base + ((pfId) * IRO[295].m1))
 #define USTORM_ISCSI_GLOBAL_BUF_PHYS_ADDR_OFFSET(pfId) \
-	(IRO[291].base + ((pfId) * IRO[291].m1))
+	(IRO[292].base + ((pfId) * IRO[292].m1))
 #define USTORM_ISCSI_NUM_OF_TASKS_OFFSET(pfId) \
-	(IRO[287].base + ((pfId) * IRO[287].m1))
+	(IRO[288].base + ((pfId) * IRO[288].m1))
 #define USTORM_ISCSI_PAGE_SIZE_LOG_OFFSET(pfId) \
-	(IRO[286].base + ((pfId) * IRO[286].m1))
+	(IRO[287].base + ((pfId) * IRO[287].m1))
 #define USTORM_ISCSI_PAGE_SIZE_OFFSET(pfId) \
-	(IRO[285].base + ((pfId) * IRO[285].m1))
+	(IRO[286].base + ((pfId) * IRO[286].m1))
 #define USTORM_ISCSI_R2TQ_SIZE_OFFSET(pfId) \
-	(IRO[288].base + ((pfId) * IRO[288].m1))
+	(IRO[289].base + ((pfId) * IRO[289].m1))
 #define USTORM_ISCSI_RQ_BUFFER_SIZE_OFFSET(pfId) \
-	(IRO[292].base + ((pfId) * IRO[292].m1))
-#define USTORM_ISCSI_RQ_SIZE_OFFSET(pfId) \
 	(IRO[293].base + ((pfId) * IRO[293].m1))
+#define USTORM_ISCSI_RQ_SIZE_OFFSET(pfId) \
+	(IRO[294].base + ((pfId) * IRO[294].m1))
 #define USTORM_MEM_WORKAROUND_ADDRESS_OFFSET(pfId) \
 	(IRO[186].base + ((pfId) * IRO[186].m1))
 #define USTORM_RECORD_SLOW_PATH_OFFSET(funcId) \
 	(IRO[184].base + ((funcId) * IRO[184].m1))
 #define USTORM_RX_PRODS_E1X_OFFSET(portId, clientId) \
-	(IRO[215].base + ((portId) * IRO[215].m1) + ((clientId) * \
-	IRO[215].m2))
+	(IRO[216].base + ((portId) * IRO[216].m1) + ((clientId) * \
+	IRO[216].m2))
 #define USTORM_RX_PRODS_E2_OFFSET(qzoneId) \
-	(IRO[216].base + ((qzoneId) * IRO[216].m1))
-#define USTORM_TPA_BTR_OFFSET (IRO[213].base)
-#define USTORM_TPA_BTR_SIZE (IRO[213].size)
+	(IRO[217].base + ((qzoneId) * IRO[217].m1))
+#define USTORM_TPA_BTR_OFFSET (IRO[214].base)
+#define USTORM_TPA_BTR_SIZE (IRO[214].size)
 #define USTORM_VF_TO_PF_OFFSET(funcId) \
 	(IRO[183].base + ((funcId) * IRO[183].m1))
 #define XSTORM_AGG_INT_FINAL_CLEANUP_COMP_TYPE (IRO[67].base)
@@ -188,39 +188,39 @@
 #define XSTORM_FUNC_EN_OFFSET(funcId) \
 	(IRO[47].base + ((funcId) * IRO[47].m1))
 #define XSTORM_ISCSI_HQ_SIZE_OFFSET(pfId) \
-	(IRO[302].base + ((pfId) * IRO[302].m1))
+	(IRO[303].base + ((pfId) * IRO[303].m1))
 #define XSTORM_ISCSI_LOCAL_MAC_ADDR0_OFFSET(pfId) \
-	(IRO[305].base + ((pfId) * IRO[305].m1))
-#define XSTORM_ISCSI_LOCAL_MAC_ADDR1_OFFSET(pfId) \
 	(IRO[306].base + ((pfId) * IRO[306].m1))
-#define XSTORM_ISCSI_LOCAL_MAC_ADDR2_OFFSET(pfId) \
+#define XSTORM_ISCSI_LOCAL_MAC_ADDR1_OFFSET(pfId) \
 	(IRO[307].base + ((pfId) * IRO[307].m1))
-#define XSTORM_ISCSI_LOCAL_MAC_ADDR3_OFFSET(pfId) \
+#define XSTORM_ISCSI_LOCAL_MAC_ADDR2_OFFSET(pfId) \
 	(IRO[308].base + ((pfId) * IRO[308].m1))
-#define XSTORM_ISCSI_LOCAL_MAC_ADDR4_OFFSET(pfId) \
+#define XSTORM_ISCSI_LOCAL_MAC_ADDR3_OFFSET(pfId) \
 	(IRO[309].base + ((pfId) * IRO[309].m1))
-#define XSTORM_ISCSI_LOCAL_MAC_ADDR5_OFFSET(pfId) \
+#define XSTORM_ISCSI_LOCAL_MAC_ADDR4_OFFSET(pfId) \
 	(IRO[310].base + ((pfId) * IRO[310].m1))
-#define XSTORM_ISCSI_LOCAL_VLAN_OFFSET(pfId) \
+#define XSTORM_ISCSI_LOCAL_MAC_ADDR5_OFFSET(pfId) \
 	(IRO[311].base + ((pfId) * IRO[311].m1))
+#define XSTORM_ISCSI_LOCAL_VLAN_OFFSET(pfId) \
+	(IRO[312].base + ((pfId) * IRO[312].m1))
 #define XSTORM_ISCSI_NUM_OF_TASKS_OFFSET(pfId) \
-	(IRO[301].base + ((pfId) * IRO[301].m1))
+	(IRO[302].base + ((pfId) * IRO[302].m1))
 #define XSTORM_ISCSI_PAGE_SIZE_LOG_OFFSET(pfId) \
-	(IRO[300].base + ((pfId) * IRO[300].m1))
+	(IRO[301].base + ((pfId) * IRO[301].m1))
 #define XSTORM_ISCSI_PAGE_SIZE_OFFSET(pfId) \
-	(IRO[299].base + ((pfId) * IRO[299].m1))
+	(IRO[300].base + ((pfId) * IRO[300].m1))
 #define XSTORM_ISCSI_R2TQ_SIZE_OFFSET(pfId) \
-	(IRO[304].base + ((pfId) * IRO[304].m1))
+	(IRO[305].base + ((pfId) * IRO[305].m1))
 #define XSTORM_ISCSI_SQ_SIZE_OFFSET(pfId) \
-	(IRO[303].base + ((pfId) * IRO[303].m1))
+	(IRO[304].base + ((pfId) * IRO[304].m1))
 #define XSTORM_ISCSI_TCP_VARS_ADV_WND_SCL_OFFSET(pfId) \
-	(IRO[298].base + ((pfId) * IRO[298].m1))
+	(IRO[299].base + ((pfId) * IRO[299].m1))
 #define XSTORM_ISCSI_TCP_VARS_FLAGS_OFFSET(pfId) \
-	(IRO[297].base + ((pfId) * IRO[297].m1))
+	(IRO[298].base + ((pfId) * IRO[298].m1))
 #define XSTORM_ISCSI_TCP_VARS_TOS_OFFSET(pfId) \
-	(IRO[296].base + ((pfId) * IRO[296].m1))
+	(IRO[297].base + ((pfId) * IRO[297].m1))
 #define XSTORM_ISCSI_TCP_VARS_TTL_OFFSET(pfId) \
-	(IRO[295].base + ((pfId) * IRO[295].m1))
+	(IRO[296].base + ((pfId) * IRO[296].m1))
 #define XSTORM_RATE_SHAPING_PER_VN_VARS_OFFSET(pfId) \
 	(IRO[44].base + ((pfId) * IRO[44].m1))
 #define XSTORM_RECORD_SLOW_PATH_OFFSET(funcId) \
@@ -233,14 +233,16 @@
 #define XSTORM_SPQ_PROD_OFFSET(funcId) \
 	(IRO[31].base + ((funcId) * IRO[31].m1))
 #define XSTORM_TCP_GLOBAL_DEL_ACK_COUNTER_ENABLED_OFFSET(portId) \
-	(IRO[217].base + ((portId) * IRO[217].m1))
-#define XSTORM_TCP_GLOBAL_DEL_ACK_COUNTER_MAX_COUNT_OFFSET(portId) \
 	(IRO[218].base + ((portId) * IRO[218].m1))
+#define XSTORM_TCP_GLOBAL_DEL_ACK_COUNTER_MAX_COUNT_OFFSET(portId) \
+	(IRO[219].base + ((portId) * IRO[219].m1))
 #define XSTORM_TCP_TX_SWS_TIMER_VAL_OFFSET(pfId) \
-	(IRO[220].base + (((pfId)>>1) * IRO[220].m1) + (((pfId)&1) * \
-	IRO[220].m2))
+	(IRO[221].base + (((pfId)>>1) * IRO[221].m1) + (((pfId)&1) * \
+	IRO[221].m2))
 #define XSTORM_VF_TO_PF_OFFSET(funcId) \
 	(IRO[48].base + ((funcId) * IRO[48].m1))
+#define XSTORM_ETH_FUNCTION_INFO_FP_HSI_VALID_E2_OFFSET(fid)	\
+	(IRO[386].base + ((fid) * IRO[386].m1))
 #define COMMON_ASM_INVALID_ASSERT_OPCODE 0x0
 
 /* eth hsi version */
@@ -330,7 +332,7 @@
 #define TOE_STATE (TOE_CONNECTION_TYPE << PROTOCOL_STATE_BIT_OFFSET)
 #define RDMA_STATE (RDMA_CONNECTION_TYPE << PROTOCOL_STATE_BIT_OFFSET)
 
-/* microcode fixed page page size 4K (chains and ring segments) */
+/* microcode fixed page size 4K (chains and ring segments) */
 #define MC_PAGE_SIZE 4096
 
 /* Number of indices per slow-path SB */
diff --git a/drivers/net/ethernet/broadcom/bnx2x/bnx2x_hsi.h b/drivers/net/ethernet/broadcom/bnx2x/bnx2x_hsi.h
index f8b810313094..611efee75834 100644
--- a/drivers/net/ethernet/broadcom/bnx2x/bnx2x_hsi.h
+++ b/drivers/net/ethernet/broadcom/bnx2x/bnx2x_hsi.h
@@ -1140,6 +1140,11 @@ struct shm_dev_info {				/* size */
 
 };
 
+struct extended_dev_info_shared_cfg {
+	u32 reserved[18];
+	u32 mbi_version;
+	u32 mbi_date;
+};
 
 #if !defined(__LITTLE_ENDIAN) && !defined(__BIG_ENDIAN)
 	#error "Missing either LITTLE_ENDIAN or BIG_ENDIAN definition."
@@ -3019,7 +3024,8 @@ struct afex_stats {
 
 #define BCM_5710_FW_MAJOR_VERSION			7
 #define BCM_5710_FW_MINOR_VERSION			13
-#define BCM_5710_FW_REVISION_VERSION		1
+#define BCM_5710_FW_REVISION_VERSION		21
+#define BCM_5710_FW_REVISION_VERSION_V15	15
 #define BCM_5710_FW_ENGINEERING_VERSION		0
 #define BCM_5710_FW_COMPILE_FLAGS			1
 
@@ -3634,8 +3640,10 @@ struct client_init_rx_data {
 #define CLIENT_INIT_RX_DATA_TPA_EN_IPV6_SHIFT 1
 #define CLIENT_INIT_RX_DATA_TPA_MODE (0x1<<2)
 #define CLIENT_INIT_RX_DATA_TPA_MODE_SHIFT 2
-#define CLIENT_INIT_RX_DATA_RESERVED5 (0x1F<<3)
-#define CLIENT_INIT_RX_DATA_RESERVED5_SHIFT 3
+#define CLIENT_INIT_RX_DATA_TPA_OVER_VLAN_DISABLE (0x1<<3)
+#define CLIENT_INIT_RX_DATA_TPA_OVER_VLAN_DISABLE_SHIFT 3
+#define CLIENT_INIT_RX_DATA_RESERVED5 (0xF<<4)
+#define CLIENT_INIT_RX_DATA_RESERVED5_SHIFT 4
 	u8 vmqueue_mode_en_flg;
 	u8 extra_data_over_sgl_en_flg;
 	u8 cache_line_alignment_log_size;
@@ -3826,7 +3834,7 @@ struct eth_classify_cmd_header {
  */
 struct eth_classify_header {
 	u8 rule_cnt;
-	u8 reserved0;
+	u8 warning_on_error;
 	__le16 reserved1;
 	__le32 echo;
 };
@@ -4747,6 +4755,8 @@ struct tpa_update_ramrod_data {
 	__le32 sge_page_base_hi;
 	__le16 sge_pause_thr_low;
 	__le16 sge_pause_thr_high;
+	u8 tpa_over_vlan_disable;
+	u8 reserved[7];
 };
 
 
@@ -4941,7 +4951,7 @@ struct fairness_vars_per_port {
 	u32 upper_bound;
 	u32 fair_threshold;
 	u32 fairness_timeout;
-	u32 reserved0;
+	u32 size_thr;
 };
 
 /*
@@ -5410,7 +5420,9 @@ struct function_start_data {
 	u8 sd_vlan_force_pri_val;
 	u8 c2s_pri_tt_valid;
 	u8 c2s_pri_default;
-	u8 reserved2[6];
+	u8 tx_vlan_filtering_enable;
+	u8 tx_vlan_filtering_use_pvid;
+	u8 reserved2[4];
 	struct c2s_pri_trans_table_entry c2s_pri_trans_table;
 };
 
@@ -5443,7 +5455,8 @@ struct function_update_data {
 	u8 reserved1;
 	__le16 sd_vlan_tag;
 	__le16 sd_vlan_eth_type;
-	__le16 reserved0;
+	u8 tx_vlan_filtering_pvid_change_flg;
+	u8 reserved0;
 	__le32 reserved2;
 };
 
diff --git a/drivers/net/ethernet/broadcom/bnx2x/bnx2x_init.h b/drivers/net/ethernet/broadcom/bnx2x/bnx2x_init.h
index 46ee2c01f4c5..0a59a09ef82f 100644
--- a/drivers/net/ethernet/broadcom/bnx2x/bnx2x_init.h
+++ b/drivers/net/ethernet/broadcom/bnx2x/bnx2x_init.h
@@ -296,7 +296,6 @@ static inline void bnx2x_dcb_config_qm(struct bnx2x *bp, enum cos_mode mode,
  *    possible, the driver should only write the valid vnics into the internal
  *    ram according to the appropriate port mode.
  */
-#define BITS_TO_BYTES(x) ((x)/8)
 
 /* CMNG constants, as derived from system spec calculations */
 
@@ -449,7 +448,7 @@ static inline void bnx2x_init_fw_wrr(const struct cmng_init_input *input_data,
 				ccd[cos] =
 				    (u32)input_data->cos_min_rate[cos] * 100 *
 				    (T_FAIR_COEF / (8 * 100 * cosWeightSum));
-				 if (ccd[cos] < pdata->fair_vars.fair_threshold
+				if (ccd[cos] < pdata->fair_vars.fair_threshold
 						+ MIN_ABOVE_THRESH) {
 					ccd[cos] =
 					    pdata->fair_vars.fair_threshold +
diff --git a/drivers/net/ethernet/broadcom/bnx2x/bnx2x_init_ops.h b/drivers/net/ethernet/broadcom/bnx2x/bnx2x_init_ops.h
index 1835d2e451c0..fc7fce642666 100644
--- a/drivers/net/ethernet/broadcom/bnx2x/bnx2x_init_ops.h
+++ b/drivers/net/ethernet/broadcom/bnx2x/bnx2x_init_ops.h
@@ -635,11 +635,13 @@ static int bnx2x_ilt_client_mem_op(struct bnx2x *bp, int cli_num,
 {
 	int i, rc;
 	struct bnx2x_ilt *ilt = BP_ILT(bp);
-	struct ilt_client_info *ilt_cli = &ilt->clients[cli_num];
+	struct ilt_client_info *ilt_cli;
 
 	if (!ilt || !ilt->lines)
 		return -1;
 
+	ilt_cli = &ilt->clients[cli_num];
+
 	if (ilt_cli->flags & (ILT_CLIENT_SKIP_INIT | ILT_CLIENT_SKIP_MEM))
 		return 0;
 
diff --git a/drivers/net/ethernet/broadcom/bnx2x/bnx2x_link.c b/drivers/net/ethernet/broadcom/bnx2x/bnx2x_link.c
index 7dd83d0ef0a0..ea310057fe3a 100644
--- a/drivers/net/ethernet/broadcom/bnx2x/bnx2x_link.c
+++ b/drivers/net/ethernet/broadcom/bnx2x/bnx2x_link.c
@@ -588,7 +588,7 @@ static void bnx2x_ets_e3b0_nig_disabled(const struct link_params *params,
 	 * slots for the highest priority.
 	 */
 	REG_WR(bp, (port) ? NIG_REG_P1_TX_ARB_NUM_STRICT_ARB_SLOTS :
-		   NIG_REG_P1_TX_ARB_NUM_STRICT_ARB_SLOTS, 0x100);
+		   NIG_REG_P0_TX_ARB_NUM_STRICT_ARB_SLOTS, 0x100);
 	/* Mapping between the CREDIT_WEIGHT registers and actual client
 	 * numbers
 	 */
@@ -837,49 +837,45 @@ static int bnx2x_ets_e3b0_set_cos_bw(struct bnx2x *bp,
 
 	switch (cos_entry) {
 	case 0:
-	    nig_reg_adress_crd_weight =
-		 (port) ? NIG_REG_P1_TX_ARB_CREDIT_WEIGHT_0 :
-		     NIG_REG_P0_TX_ARB_CREDIT_WEIGHT_0;
-	     pbf_reg_adress_crd_weight = (port) ?
-		 PBF_REG_COS0_WEIGHT_P1 : PBF_REG_COS0_WEIGHT_P0;
-	     break;
+		nig_reg_adress_crd_weight =
+			(port) ? NIG_REG_P1_TX_ARB_CREDIT_WEIGHT_0 :
+			NIG_REG_P0_TX_ARB_CREDIT_WEIGHT_0;
+		pbf_reg_adress_crd_weight = (port) ?
+		    PBF_REG_COS0_WEIGHT_P1 : PBF_REG_COS0_WEIGHT_P0;
+		break;
 	case 1:
-	     nig_reg_adress_crd_weight = (port) ?
-		 NIG_REG_P1_TX_ARB_CREDIT_WEIGHT_1 :
-		 NIG_REG_P0_TX_ARB_CREDIT_WEIGHT_1;
-	     pbf_reg_adress_crd_weight = (port) ?
-		 PBF_REG_COS1_WEIGHT_P1 : PBF_REG_COS1_WEIGHT_P0;
-	     break;
+		nig_reg_adress_crd_weight = (port) ?
+			NIG_REG_P1_TX_ARB_CREDIT_WEIGHT_1 :
+			NIG_REG_P0_TX_ARB_CREDIT_WEIGHT_1;
+		pbf_reg_adress_crd_weight = (port) ?
+			PBF_REG_COS1_WEIGHT_P1 : PBF_REG_COS1_WEIGHT_P0;
+		break;
 	case 2:
-	     nig_reg_adress_crd_weight = (port) ?
-		 NIG_REG_P1_TX_ARB_CREDIT_WEIGHT_2 :
-		 NIG_REG_P0_TX_ARB_CREDIT_WEIGHT_2;
+		nig_reg_adress_crd_weight = (port) ?
+			NIG_REG_P1_TX_ARB_CREDIT_WEIGHT_2 :
+			NIG_REG_P0_TX_ARB_CREDIT_WEIGHT_2;
 
-		 pbf_reg_adress_crd_weight = (port) ?
-		     PBF_REG_COS2_WEIGHT_P1 : PBF_REG_COS2_WEIGHT_P0;
-	     break;
+		pbf_reg_adress_crd_weight = (port) ?
+			PBF_REG_COS2_WEIGHT_P1 : PBF_REG_COS2_WEIGHT_P0;
+		break;
 	case 3:
-	    if (port)
+		if (port)
 			return -EINVAL;
-	     nig_reg_adress_crd_weight =
-		 NIG_REG_P0_TX_ARB_CREDIT_WEIGHT_3;
-	     pbf_reg_adress_crd_weight =
-		 PBF_REG_COS3_WEIGHT_P0;
-	     break;
+		nig_reg_adress_crd_weight = NIG_REG_P0_TX_ARB_CREDIT_WEIGHT_3;
+		pbf_reg_adress_crd_weight = PBF_REG_COS3_WEIGHT_P0;
+		break;
 	case 4:
-	    if (port)
-		return -EINVAL;
-	     nig_reg_adress_crd_weight =
-		 NIG_REG_P0_TX_ARB_CREDIT_WEIGHT_4;
-	     pbf_reg_adress_crd_weight = PBF_REG_COS4_WEIGHT_P0;
-	     break;
+		if (port)
+			return -EINVAL;
+		nig_reg_adress_crd_weight = NIG_REG_P0_TX_ARB_CREDIT_WEIGHT_4;
+		pbf_reg_adress_crd_weight = PBF_REG_COS4_WEIGHT_P0;
+		break;
 	case 5:
-	    if (port)
-		return -EINVAL;
-	     nig_reg_adress_crd_weight =
-		 NIG_REG_P0_TX_ARB_CREDIT_WEIGHT_5;
-	     pbf_reg_adress_crd_weight = PBF_REG_COS5_WEIGHT_P0;
-	     break;
+		if (port)
+			return -EINVAL;
+		nig_reg_adress_crd_weight = NIG_REG_P0_TX_ARB_CREDIT_WEIGHT_5;
+		pbf_reg_adress_crd_weight = PBF_REG_COS5_WEIGHT_P0;
+		break;
 	}
 
 	REG_WR(bp, nig_reg_adress_crd_weight, cos_bw_nig);
@@ -966,7 +962,7 @@ static int bnx2x_ets_e3b0_sp_pri_to_cos_set(const struct link_params *params,
 	if (pri >= max_num_of_cos) {
 		DP(NETIF_MSG_LINK, "bnx2x_ets_e3b0_sp_pri_to_cos_set invalid "
 		   "parameter Illegal strict priority\n");
-	    return -EINVAL;
+		return -EINVAL;
 	}
 
 	if (sp_pri_to_cos[pri] != DCBX_INVALID_COS) {
@@ -1845,28 +1841,28 @@ static int bnx2x_emac_enable(struct link_params *params,
 	bnx2x_bits_en(bp, emac_base + EMAC_REG_EMAC_TX_MODE,
 		      EMAC_TX_MODE_RESET);
 
-		/* pause enable/disable */
-		bnx2x_bits_dis(bp, emac_base + EMAC_REG_EMAC_RX_MODE,
-			       EMAC_RX_MODE_FLOW_EN);
+	/* pause enable/disable */
+	bnx2x_bits_dis(bp, emac_base + EMAC_REG_EMAC_RX_MODE,
+		       EMAC_RX_MODE_FLOW_EN);
 
-		bnx2x_bits_dis(bp,  emac_base + EMAC_REG_EMAC_TX_MODE,
-			       (EMAC_TX_MODE_EXT_PAUSE_EN |
-				EMAC_TX_MODE_FLOW_EN));
-		if (!(params->feature_config_flags &
-		      FEATURE_CONFIG_PFC_ENABLED)) {
-			if (vars->flow_ctrl & BNX2X_FLOW_CTRL_RX)
-				bnx2x_bits_en(bp, emac_base +
-					      EMAC_REG_EMAC_RX_MODE,
-					      EMAC_RX_MODE_FLOW_EN);
-
-			if (vars->flow_ctrl & BNX2X_FLOW_CTRL_TX)
-				bnx2x_bits_en(bp, emac_base +
-					      EMAC_REG_EMAC_TX_MODE,
-					      (EMAC_TX_MODE_EXT_PAUSE_EN |
-					       EMAC_TX_MODE_FLOW_EN));
-		} else
-			bnx2x_bits_en(bp, emac_base + EMAC_REG_EMAC_TX_MODE,
-				      EMAC_TX_MODE_FLOW_EN);
+	bnx2x_bits_dis(bp,  emac_base + EMAC_REG_EMAC_TX_MODE,
+		       (EMAC_TX_MODE_EXT_PAUSE_EN |
+			EMAC_TX_MODE_FLOW_EN));
+	if (!(params->feature_config_flags &
+	      FEATURE_CONFIG_PFC_ENABLED)) {
+		if (vars->flow_ctrl & BNX2X_FLOW_CTRL_RX)
+			bnx2x_bits_en(bp, emac_base +
+				      EMAC_REG_EMAC_RX_MODE,
+				      EMAC_RX_MODE_FLOW_EN);
+
+		if (vars->flow_ctrl & BNX2X_FLOW_CTRL_TX)
+			bnx2x_bits_en(bp, emac_base +
+				      EMAC_REG_EMAC_TX_MODE,
+				      (EMAC_TX_MODE_EXT_PAUSE_EN |
+				       EMAC_TX_MODE_FLOW_EN));
+	} else
+		bnx2x_bits_en(bp, emac_base + EMAC_REG_EMAC_TX_MODE,
+			      EMAC_TX_MODE_FLOW_EN);
 
 	/* KEEP_VLAN_TAG, promiscuous */
 	val = REG_RD(bp, emac_base + EMAC_REG_EMAC_RX_MODE);
@@ -3089,6 +3085,7 @@ static int bnx2x_bsc_read(struct link_params *params,
 			  u8 xfer_cnt,
 			  u32 *data_array)
 {
+	u64 t0, delta;
 	u32 val, i;
 	int rc = 0;
 
@@ -3118,17 +3115,18 @@ static int bnx2x_bsc_read(struct link_params *params,
 	REG_WR(bp, MCP_REG_MCPR_IMC_COMMAND, val);
 
 	/* Poll for completion */
-	i = 0;
+	t0 = ktime_get_ns();
 	val = REG_RD(bp, MCP_REG_MCPR_IMC_COMMAND);
 	while (((val >> MCPR_IMC_COMMAND_IMC_STATUS_BITSHIFT) & 0x3) != 1) {
-		udelay(10);
-		val = REG_RD(bp, MCP_REG_MCPR_IMC_COMMAND);
-		if (i++ > 1000) {
-			DP(NETIF_MSG_LINK, "wr 0 byte timed out after %d try\n",
-								i);
+		delta = ktime_get_ns() - t0;
+		if (delta > 10 * NSEC_PER_MSEC) {
+			DP(NETIF_MSG_LINK, "wr 0 byte timed out after %Lu ns\n",
+					   delta);
 			rc = -EFAULT;
 			break;
 		}
+		usleep_range(10, 20);
+		val = REG_RD(bp, MCP_REG_MCPR_IMC_COMMAND);
 	}
 	if (rc == -EFAULT)
 		return rc;
@@ -3142,16 +3140,18 @@ static int bnx2x_bsc_read(struct link_params *params,
 	REG_WR(bp, MCP_REG_MCPR_IMC_COMMAND, val);
 
 	/* Poll for completion */
-	i = 0;
+	t0 = ktime_get_ns();
 	val = REG_RD(bp, MCP_REG_MCPR_IMC_COMMAND);
 	while (((val >> MCPR_IMC_COMMAND_IMC_STATUS_BITSHIFT) & 0x3) != 1) {
-		udelay(10);
-		val = REG_RD(bp, MCP_REG_MCPR_IMC_COMMAND);
-		if (i++ > 1000) {
-			DP(NETIF_MSG_LINK, "rd op timed out after %d try\n", i);
+		delta = ktime_get_ns() - t0;
+		if (delta > 10 * NSEC_PER_MSEC) {
+			DP(NETIF_MSG_LINK, "rd op timed out after %Lu ns\n",
+					   delta);
 			rc = -EFAULT;
 			break;
 		}
+		usleep_range(10, 20);
+		val = REG_RD(bp, MCP_REG_MCPR_IMC_COMMAND);
 	}
 	if (rc == -EFAULT)
 		return rc;
@@ -4712,14 +4712,14 @@ static void bnx2x_sync_link(struct link_params *params,
 			LINK_STATUS_SPEED_AND_DUPLEX_MASK) {
 		case LINK_10THD:
 			vars->duplex = DUPLEX_HALF;
-			/* Fall thru */
+			fallthrough;
 		case LINK_10TFD:
 			vars->line_speed = SPEED_10;
 			break;
 
 		case LINK_100TXHD:
 			vars->duplex = DUPLEX_HALF;
-			/* Fall thru */
+			fallthrough;
 		case LINK_100T4:
 		case LINK_100TXFD:
 			vars->line_speed = SPEED_100;
@@ -4727,14 +4727,14 @@ static void bnx2x_sync_link(struct link_params *params,
 
 		case LINK_1000THD:
 			vars->duplex = DUPLEX_HALF;
-			/* Fall thru */
+			fallthrough;
 		case LINK_1000TFD:
 			vars->line_speed = SPEED_1000;
 			break;
 
 		case LINK_2500THD:
 			vars->duplex = DUPLEX_HALF;
-			/* Fall thru */
+			fallthrough;
 		case LINK_2500TFD:
 			vars->line_speed = SPEED_2500;
 			break;
@@ -5615,9 +5615,9 @@ static int bnx2x_get_link_speed_duplex(struct bnx2x_phy *phy,
 	return 0;
 }
 
-static int bnx2x_link_settings_status(struct bnx2x_phy *phy,
-				      struct link_params *params,
-				      struct link_vars *vars)
+static u8 bnx2x_link_settings_status(struct bnx2x_phy *phy,
+				     struct link_params *params,
+				     struct link_vars *vars)
 {
 	struct bnx2x *bp = params->bp;
 
@@ -5689,7 +5689,7 @@ static int bnx2x_link_settings_status(struct bnx2x_phy *phy,
 	return rc;
 }
 
-static int bnx2x_warpcore_read_status(struct bnx2x_phy *phy,
+static u8 bnx2x_warpcore_read_status(struct bnx2x_phy *phy,
 				     struct link_params *params,
 				     struct link_vars *vars)
 {
@@ -6163,8 +6163,8 @@ static void bnx2x_link_int_ack(struct link_params *params,
 
 static int bnx2x_null_format_ver(u32 spirom_ver, u8 *str, u16 *len)
 {
-	str[0] = '\0';
-	(*len)--;
+	if (*len)
+		str[0] = '\0';
 	return 0;
 }
 
@@ -6173,12 +6173,13 @@ static int bnx2x_format_ver(u32 num, u8 *str, u16 *len)
 	u16 ret;
 
 	if (*len < 10) {
-		/* Need more than 10chars for this format */
+		/* Need more than 10 chars for this format */
 		bnx2x_null_format_ver(num, str, len);
 		return -EINVAL;
 	}
 
-	ret = scnprintf(str, *len, "%hx.%hx", num >> 16, num);
+	ret = scnprintf(str, *len, "%x.%x", (num >> 16) & 0xFFFF,
+			num & 0xFFFF);
 	*len -= ret;
 	return 0;
 }
@@ -6187,13 +6188,14 @@ static int bnx2x_3_seq_format_ver(u32 num, u8 *str, u16 *len)
 {
 	u16 ret;
 
-	if (*len < 10) {
-		/* Need more than 10chars for this format */
+	if (*len < 9) {
+		/* Need more than 9 chars for this format */
 		bnx2x_null_format_ver(num, str, len);
 		return -EINVAL;
 	}
 
-	ret = scnprintf(str, *len, "%hhx.%hhx.%hhx", num >> 16, num >> 8, num);
+	ret = scnprintf(str, *len, "%x.%x.%x", (num >> 16) & 0xFF,
+			(num >> 8) & 0xFF, num & 0xFF);
 	*len -= ret;
 	return 0;
 }
@@ -6206,7 +6208,7 @@ int bnx2x_get_ext_phy_fw_version(struct link_params *params, u8 *version,
 	int status = 0;
 	u8 *ver_p = version;
 	u16 remain_len = len;
-	if (version == NULL || params == NULL)
+	if (version == NULL || params == NULL || len == 0)
 		return -EINVAL;
 	bp = params->bp;
 
@@ -6339,6 +6341,7 @@ int bnx2x_set_led(struct link_params *params,
 		 */
 		if (!vars->link_up)
 			break;
+		fallthrough;
 	case LED_MODE_ON:
 		if (((params->phy[EXT_PHY1].type ==
 			  PORT_HW_CFG_XGXS_EXT_PHY_TYPE_BCM8727) ||
@@ -6477,9 +6480,9 @@ int bnx2x_test_link(struct link_params *params, struct link_vars *vars,
 			  MDIO_REG_BANK_GP_STATUS,
 			  MDIO_GP_STATUS_TOP_AN_STATUS1,
 			  &gp_status);
-	/* Link is up only if both local phy and external phy are up */
-	if (!(gp_status & MDIO_GP_STATUS_TOP_AN_STATUS1_LINK_STATUS))
-		return -ESRCH;
+		/* Link is up only if both local phy and external phy are up */
+		if (!(gp_status & MDIO_GP_STATUS_TOP_AN_STATUS1_LINK_STATUS))
+			return -ESRCH;
 	}
 	/* In XGXS loopback mode, do not check external PHY */
 	if (params->loopback_mode == LOOPBACK_XGXS)
@@ -6877,7 +6880,8 @@ int bnx2x_link_update(struct link_params *params, struct link_vars *vars)
 			case PORT_HW_CFG_PHY_SELECTION_FIRST_PHY_PRIORITY:
 			/* In this option, the first PHY makes sure to pass the
 			 * traffic through itself only.
-			 * Its not clear how to reset the link on the second phy
+			 * It's not clear how to reset the link on the second
+			 * phy.
 			 */
 				active_external_phy = EXT_PHY1;
 				break;
@@ -7292,8 +7296,8 @@ static int bnx2x_8073_xaui_wa(struct bnx2x *bp, struct bnx2x_phy *phy)
 					DP(NETIF_MSG_LINK,
 					  "XAUI workaround has completed\n");
 					return 0;
-				 }
-				 usleep_range(3000, 6000);
+				}
+				usleep_range(3000, 6000);
 			}
 			break;
 		}
@@ -7367,9 +7371,9 @@ static void bnx2x_8073_specific_func(struct bnx2x_phy *phy,
 	}
 }
 
-static int bnx2x_8073_config_init(struct bnx2x_phy *phy,
-				  struct link_params *params,
-				  struct link_vars *vars)
+static void bnx2x_8073_config_init(struct bnx2x_phy *phy,
+				   struct link_params *params,
+				   struct link_vars *vars)
 {
 	struct bnx2x *bp = params->bp;
 	u16 val = 0, tmp1;
@@ -7430,7 +7434,7 @@ static int bnx2x_8073_config_init(struct bnx2x_phy *phy,
 	if (params->loopback_mode == LOOPBACK_EXT) {
 		bnx2x_807x_force_10G(bp, phy);
 		DP(NETIF_MSG_LINK, "Forced speed 10G on 807X\n");
-		return 0;
+		return;
 	} else {
 		bnx2x_cl45_write(bp, phy,
 				 MDIO_PMA_DEVAD, MDIO_PMA_REG_BCM_CTRL, 0x0002);
@@ -7512,7 +7516,6 @@ static int bnx2x_8073_config_init(struct bnx2x_phy *phy,
 	bnx2x_cl45_write(bp, phy, MDIO_AN_DEVAD, MDIO_AN_REG_CTRL, 0x1200);
 	DP(NETIF_MSG_LINK, "807x Autoneg Restart: Advertise 1G=%x, 10G=%x\n",
 		   ((val & (1<<5)) > 0), ((val & (1<<7)) > 0));
-	return 0;
 }
 
 static u8 bnx2x_8073_read_status(struct bnx2x_phy *phy,
@@ -7679,9 +7682,9 @@ static void bnx2x_8073_link_reset(struct bnx2x_phy *phy,
 /******************************************************************/
 /*			BCM8705 PHY SECTION			  */
 /******************************************************************/
-static int bnx2x_8705_config_init(struct bnx2x_phy *phy,
-				  struct link_params *params,
-				  struct link_vars *vars)
+static void bnx2x_8705_config_init(struct bnx2x_phy *phy,
+				   struct link_params *params,
+				   struct link_vars *vars)
 {
 	struct bnx2x *bp = params->bp;
 	DP(NETIF_MSG_LINK, "init 8705\n");
@@ -7703,7 +7706,6 @@ static int bnx2x_8705_config_init(struct bnx2x_phy *phy,
 			 MDIO_WIS_DEVAD, MDIO_WIS_REG_LASI_CNTL, 0x1);
 	/* BCM8705 doesn't have microcode, hence the 0 */
 	bnx2x_save_spirom_version(bp, params->port, params->shmem_base, 0);
-	return 0;
 }
 
 static u8 bnx2x_8705_read_status(struct bnx2x_phy *phy,
@@ -8890,9 +8892,9 @@ static u8 bnx2x_8706_8726_read_status(struct bnx2x_phy *phy,
 /******************************************************************/
 /*			BCM8706 PHY SECTION			  */
 /******************************************************************/
-static u8 bnx2x_8706_config_init(struct bnx2x_phy *phy,
-				 struct link_params *params,
-				 struct link_vars *vars)
+static void bnx2x_8706_config_init(struct bnx2x_phy *phy,
+				   struct link_params *params,
+				   struct link_vars *vars)
 {
 	u32 tx_en_mode;
 	u16 cnt, val, tmp1;
@@ -8992,13 +8994,11 @@ static u8 bnx2x_8706_config_init(struct bnx2x_phy *phy,
 		bnx2x_cl45_write(bp, phy,
 			MDIO_PMA_DEVAD, MDIO_PMA_REG_DIGITAL_CTRL, tmp1);
 	}
-
-	return 0;
 }
 
-static int bnx2x_8706_read_status(struct bnx2x_phy *phy,
-				  struct link_params *params,
-				  struct link_vars *vars)
+static u8 bnx2x_8706_read_status(struct bnx2x_phy *phy,
+				 struct link_params *params,
+				 struct link_vars *vars)
 {
 	return bnx2x_8706_8726_read_status(phy, params, vars);
 }
@@ -9073,9 +9073,9 @@ static u8 bnx2x_8726_read_status(struct bnx2x_phy *phy,
 }
 
 
-static int bnx2x_8726_config_init(struct bnx2x_phy *phy,
-				  struct link_params *params,
-				  struct link_vars *vars)
+static void bnx2x_8726_config_init(struct bnx2x_phy *phy,
+				   struct link_params *params,
+				   struct link_vars *vars)
 {
 	struct bnx2x *bp = params->bp;
 	DP(NETIF_MSG_LINK, "Initializing BCM8726\n");
@@ -9153,9 +9153,6 @@ static int bnx2x_8726_config_init(struct bnx2x_phy *phy,
 				 MDIO_PMA_REG_8726_TX_CTRL2,
 				 phy->tx_preemphasis[1]);
 	}
-
-	return 0;
-
 }
 
 static void bnx2x_8726_link_reset(struct bnx2x_phy *phy,
@@ -9291,9 +9288,9 @@ static void bnx2x_8727_config_speed(struct bnx2x_phy *phy,
 	}
 }
 
-static int bnx2x_8727_config_init(struct bnx2x_phy *phy,
-				  struct link_params *params,
-				  struct link_vars *vars)
+static void bnx2x_8727_config_init(struct bnx2x_phy *phy,
+				   struct link_params *params,
+				   struct link_vars *vars)
 {
 	u32 tx_en_mode;
 	u16 tmp1, mod_abs, tmp2;
@@ -9373,8 +9370,6 @@ static int bnx2x_8727_config_init(struct bnx2x_phy *phy,
 				 MDIO_PMA_DEVAD, MDIO_PMA_REG_PHY_IDENTIFIER,
 				 (tmp2 & 0x7fff));
 	}
-
-	return 0;
 }
 
 static void bnx2x_8727_handle_mod_abs(struct bnx2x_phy *phy,
@@ -9949,9 +9944,9 @@ static int bnx2x_848xx_cmn_config_init(struct bnx2x_phy *phy,
 	return 0;
 }
 
-static int bnx2x_8481_config_init(struct bnx2x_phy *phy,
-				  struct link_params *params,
-				  struct link_vars *vars)
+static void bnx2x_8481_config_init(struct bnx2x_phy *phy,
+				   struct link_params *params,
+				   struct link_vars *vars)
 {
 	struct bnx2x *bp = params->bp;
 	/* Restore normal power mode*/
@@ -9963,7 +9958,7 @@ static int bnx2x_8481_config_init(struct bnx2x_phy *phy,
 	bnx2x_wait_reset_complete(bp, phy, params);
 
 	bnx2x_cl45_write(bp, phy, MDIO_PMA_DEVAD, MDIO_PMA_REG_CTRL, 1<<15);
-	return bnx2x_848xx_cmn_config_init(phy, params, vars);
+	bnx2x_848xx_cmn_config_init(phy, params, vars);
 }
 
 #define PHY848xx_CMDHDLR_WAIT 300
@@ -10213,8 +10208,8 @@ static u8 bnx2x_84833_get_reset_gpios(struct bnx2x *bp,
 	return reset_gpios;
 }
 
-static int bnx2x_84833_hw_reset_phy(struct bnx2x_phy *phy,
-				struct link_params *params)
+static void bnx2x_84833_hw_reset_phy(struct bnx2x_phy *phy,
+				     struct link_params *params)
 {
 	struct bnx2x *bp = params->bp;
 	u8 reset_gpios;
@@ -10242,8 +10237,6 @@ static int bnx2x_84833_hw_reset_phy(struct bnx2x_phy *phy,
 	udelay(10);
 	DP(NETIF_MSG_LINK, "84833 hw reset on pin values 0x%x\n",
 		reset_gpios);
-
-	return 0;
 }
 
 static int bnx2x_8483x_disable_eee(struct bnx2x_phy *phy,
@@ -10286,9 +10279,9 @@ static int bnx2x_8483x_enable_eee(struct bnx2x_phy *phy,
 }
 
 #define PHY84833_CONSTANT_LATENCY 1193
-static int bnx2x_848x3_config_init(struct bnx2x_phy *phy,
-				   struct link_params *params,
-				   struct link_vars *vars)
+static void bnx2x_848x3_config_init(struct bnx2x_phy *phy,
+				    struct link_params *params,
+				    struct link_vars *vars)
 {
 	struct bnx2x *bp = params->bp;
 	u8 port, initialize = 1;
@@ -10433,7 +10426,7 @@ static int bnx2x_848x3_config_init(struct bnx2x_phy *phy,
 		if (rc) {
 			DP(NETIF_MSG_LINK, "Failed to configure EEE timers\n");
 			bnx2x_8483x_disable_eee(phy, params, vars);
-			return rc;
+			return;
 		}
 
 		if ((phy->req_duplex == DUPLEX_FULL) &&
@@ -10445,7 +10438,7 @@ static int bnx2x_848x3_config_init(struct bnx2x_phy *phy,
 			rc = bnx2x_8483x_disable_eee(phy, params, vars);
 		if (rc) {
 			DP(NETIF_MSG_LINK, "Failed to set EEE advertisement\n");
-			return rc;
+			return;
 		}
 	} else {
 		vars->eee_status &= ~SHMEM_EEE_SUPPORTED_MASK;
@@ -10484,7 +10477,6 @@ static int bnx2x_848x3_config_init(struct bnx2x_phy *phy,
 					  MDIO_84833_TOP_CFG_XGPHY_STRAP1,
 					  (u16)~MDIO_84833_SUPER_ISOLATE);
 	}
-	return rc;
 }
 
 static u8 bnx2x_848xx_read_status(struct bnx2x_phy *phy,
@@ -11041,9 +11033,9 @@ static void bnx2x_54618se_specific_func(struct bnx2x_phy *phy,
 	}
 }
 
-static int bnx2x_54618se_config_init(struct bnx2x_phy *phy,
-					       struct link_params *params,
-					       struct link_vars *vars)
+static void bnx2x_54618se_config_init(struct bnx2x_phy *phy,
+				      struct link_params *params,
+				      struct link_vars *vars)
 {
 	struct bnx2x *bp = params->bp;
 	u8 port;
@@ -11243,8 +11235,6 @@ static int bnx2x_54618se_config_init(struct bnx2x_phy *phy,
 
 	bnx2x_cl22_write(bp, phy,
 			MDIO_PMA_REG_CTRL, autoneg_val);
-
-	return 0;
 }
 
 
@@ -11468,9 +11458,9 @@ static void bnx2x_7101_config_loopback(struct bnx2x_phy *phy,
 			 MDIO_XS_DEVAD, MDIO_XS_SFX7101_XGXS_TEST1, 0x100);
 }
 
-static int bnx2x_7101_config_init(struct bnx2x_phy *phy,
-				  struct link_params *params,
-				  struct link_vars *vars)
+static void bnx2x_7101_config_init(struct bnx2x_phy *phy,
+				   struct link_params *params,
+				   struct link_vars *vars)
 {
 	u16 fw_ver1, fw_ver2, val;
 	struct bnx2x *bp = params->bp;
@@ -11505,7 +11495,6 @@ static int bnx2x_7101_config_init(struct bnx2x_phy *phy,
 			MDIO_PMA_DEVAD, MDIO_PMA_REG_7101_VER2, &fw_ver2);
 	bnx2x_save_spirom_version(bp, params->port,
 				  (u32)(fw_ver1<<16 | fw_ver2), phy->ver_addr);
-	return 0;
 }
 
 static u8 bnx2x_7101_read_status(struct bnx2x_phy *phy,
@@ -11557,7 +11546,7 @@ static int bnx2x_7101_format_ver(u32 spirom_ver, u8 *str, u16 *len)
 	str[2] = (spirom_ver & 0xFF0000) >> 16;
 	str[3] = (spirom_ver & 0xFF000000) >> 24;
 	str[4] = '\0';
-	*len -= 5;
+	*len -= 4;
 	return 0;
 }
 
@@ -11639,14 +11628,14 @@ static const struct bnx2x_phy phy_null = {
 	.speed_cap_mask	= 0,
 	.req_duplex	= 0,
 	.rsrv		= 0,
-	.config_init	= (config_init_t)NULL,
-	.read_status	= (read_status_t)NULL,
-	.link_reset	= (link_reset_t)NULL,
-	.config_loopback = (config_loopback_t)NULL,
-	.format_fw_ver	= (format_fw_ver_t)NULL,
-	.hw_reset	= (hw_reset_t)NULL,
-	.set_link_led	= (set_link_led_t)NULL,
-	.phy_specific_func = (phy_specific_func_t)NULL
+	.config_init	= NULL,
+	.read_status	= NULL,
+	.link_reset	= NULL,
+	.config_loopback = NULL,
+	.format_fw_ver	= NULL,
+	.hw_reset	= NULL,
+	.set_link_led	= NULL,
+	.phy_specific_func = NULL
 };
 
 static const struct bnx2x_phy phy_serdes = {
@@ -11674,14 +11663,14 @@ static const struct bnx2x_phy phy_serdes = {
 	.speed_cap_mask	= 0,
 	.req_duplex	= 0,
 	.rsrv		= 0,
-	.config_init	= (config_init_t)bnx2x_xgxs_config_init,
-	.read_status	= (read_status_t)bnx2x_link_settings_status,
-	.link_reset	= (link_reset_t)bnx2x_int_link_reset,
-	.config_loopback = (config_loopback_t)NULL,
-	.format_fw_ver	= (format_fw_ver_t)NULL,
-	.hw_reset	= (hw_reset_t)NULL,
-	.set_link_led	= (set_link_led_t)NULL,
-	.phy_specific_func = (phy_specific_func_t)NULL
+	.config_init	= bnx2x_xgxs_config_init,
+	.read_status	= bnx2x_link_settings_status,
+	.link_reset	= bnx2x_int_link_reset,
+	.config_loopback = NULL,
+	.format_fw_ver	= NULL,
+	.hw_reset	= NULL,
+	.set_link_led	= NULL,
+	.phy_specific_func = NULL
 };
 
 static const struct bnx2x_phy phy_xgxs = {
@@ -11710,14 +11699,14 @@ static const struct bnx2x_phy phy_xgxs = {
 	.speed_cap_mask	= 0,
 	.req_duplex	= 0,
 	.rsrv		= 0,
-	.config_init	= (config_init_t)bnx2x_xgxs_config_init,
-	.read_status	= (read_status_t)bnx2x_link_settings_status,
-	.link_reset	= (link_reset_t)bnx2x_int_link_reset,
-	.config_loopback = (config_loopback_t)bnx2x_set_xgxs_loopback,
-	.format_fw_ver	= (format_fw_ver_t)NULL,
-	.hw_reset	= (hw_reset_t)NULL,
-	.set_link_led	= (set_link_led_t)NULL,
-	.phy_specific_func = (phy_specific_func_t)bnx2x_xgxs_specific_func
+	.config_init	= bnx2x_xgxs_config_init,
+	.read_status	= bnx2x_link_settings_status,
+	.link_reset	= bnx2x_int_link_reset,
+	.config_loopback = bnx2x_set_xgxs_loopback,
+	.format_fw_ver	= NULL,
+	.hw_reset	= NULL,
+	.set_link_led	= NULL,
+	.phy_specific_func = bnx2x_xgxs_specific_func
 };
 static const struct bnx2x_phy phy_warpcore = {
 	.type		= PORT_HW_CFG_XGXS_EXT_PHY_TYPE_DIRECT,
@@ -11748,14 +11737,14 @@ static const struct bnx2x_phy phy_warpcore = {
 	.speed_cap_mask	= 0,
 	/* req_duplex = */0,
 	/* rsrv = */0,
-	.config_init	= (config_init_t)bnx2x_warpcore_config_init,
-	.read_status	= (read_status_t)bnx2x_warpcore_read_status,
-	.link_reset	= (link_reset_t)bnx2x_warpcore_link_reset,
-	.config_loopback = (config_loopback_t)bnx2x_set_warpcore_loopback,
-	.format_fw_ver	= (format_fw_ver_t)NULL,
-	.hw_reset	= (hw_reset_t)bnx2x_warpcore_hw_reset,
-	.set_link_led	= (set_link_led_t)NULL,
-	.phy_specific_func = (phy_specific_func_t)NULL
+	.config_init	= bnx2x_warpcore_config_init,
+	.read_status	= bnx2x_warpcore_read_status,
+	.link_reset	= bnx2x_warpcore_link_reset,
+	.config_loopback = bnx2x_set_warpcore_loopback,
+	.format_fw_ver	= NULL,
+	.hw_reset	= bnx2x_warpcore_hw_reset,
+	.set_link_led	= NULL,
+	.phy_specific_func = NULL
 };
 
 
@@ -11779,14 +11768,14 @@ static const struct bnx2x_phy phy_7101 = {
 	.speed_cap_mask	= 0,
 	.req_duplex	= 0,
 	.rsrv		= 0,
-	.config_init	= (config_init_t)bnx2x_7101_config_init,
-	.read_status	= (read_status_t)bnx2x_7101_read_status,
-	.link_reset	= (link_reset_t)bnx2x_common_ext_link_reset,
-	.config_loopback = (config_loopback_t)bnx2x_7101_config_loopback,
-	.format_fw_ver	= (format_fw_ver_t)bnx2x_7101_format_ver,
-	.hw_reset	= (hw_reset_t)bnx2x_7101_hw_reset,
-	.set_link_led	= (set_link_led_t)bnx2x_7101_set_link_led,
-	.phy_specific_func = (phy_specific_func_t)NULL
+	.config_init	= bnx2x_7101_config_init,
+	.read_status	= bnx2x_7101_read_status,
+	.link_reset	= bnx2x_common_ext_link_reset,
+	.config_loopback = bnx2x_7101_config_loopback,
+	.format_fw_ver	= bnx2x_7101_format_ver,
+	.hw_reset	= bnx2x_7101_hw_reset,
+	.set_link_led	= bnx2x_7101_set_link_led,
+	.phy_specific_func = NULL
 };
 static const struct bnx2x_phy phy_8073 = {
 	.type		= PORT_HW_CFG_XGXS_EXT_PHY_TYPE_BCM8073,
@@ -11810,14 +11799,14 @@ static const struct bnx2x_phy phy_8073 = {
 	.speed_cap_mask	= 0,
 	.req_duplex	= 0,
 	.rsrv		= 0,
-	.config_init	= (config_init_t)bnx2x_8073_config_init,
-	.read_status	= (read_status_t)bnx2x_8073_read_status,
-	.link_reset	= (link_reset_t)bnx2x_8073_link_reset,
-	.config_loopback = (config_loopback_t)NULL,
-	.format_fw_ver	= (format_fw_ver_t)bnx2x_format_ver,
-	.hw_reset	= (hw_reset_t)NULL,
-	.set_link_led	= (set_link_led_t)NULL,
-	.phy_specific_func = (phy_specific_func_t)bnx2x_8073_specific_func
+	.config_init	= bnx2x_8073_config_init,
+	.read_status	= bnx2x_8073_read_status,
+	.link_reset	= bnx2x_8073_link_reset,
+	.config_loopback = NULL,
+	.format_fw_ver	= bnx2x_format_ver,
+	.hw_reset	= NULL,
+	.set_link_led	= NULL,
+	.phy_specific_func = bnx2x_8073_specific_func
 };
 static const struct bnx2x_phy phy_8705 = {
 	.type		= PORT_HW_CFG_XGXS_EXT_PHY_TYPE_BCM8705,
@@ -11838,14 +11827,14 @@ static const struct bnx2x_phy phy_8705 = {
 	.speed_cap_mask	= 0,
 	.req_duplex	= 0,
 	.rsrv		= 0,
-	.config_init	= (config_init_t)bnx2x_8705_config_init,
-	.read_status	= (read_status_t)bnx2x_8705_read_status,
-	.link_reset	= (link_reset_t)bnx2x_common_ext_link_reset,
-	.config_loopback = (config_loopback_t)NULL,
-	.format_fw_ver	= (format_fw_ver_t)bnx2x_null_format_ver,
-	.hw_reset	= (hw_reset_t)NULL,
-	.set_link_led	= (set_link_led_t)NULL,
-	.phy_specific_func = (phy_specific_func_t)NULL
+	.config_init	= bnx2x_8705_config_init,
+	.read_status	= bnx2x_8705_read_status,
+	.link_reset	= bnx2x_common_ext_link_reset,
+	.config_loopback = NULL,
+	.format_fw_ver	= bnx2x_null_format_ver,
+	.hw_reset	= NULL,
+	.set_link_led	= NULL,
+	.phy_specific_func = NULL
 };
 static const struct bnx2x_phy phy_8706 = {
 	.type		= PORT_HW_CFG_XGXS_EXT_PHY_TYPE_BCM8706,
@@ -11867,14 +11856,14 @@ static const struct bnx2x_phy phy_8706 = {
 	.speed_cap_mask	= 0,
 	.req_duplex	= 0,
 	.rsrv		= 0,
-	.config_init	= (config_init_t)bnx2x_8706_config_init,
-	.read_status	= (read_status_t)bnx2x_8706_read_status,
-	.link_reset	= (link_reset_t)bnx2x_common_ext_link_reset,
-	.config_loopback = (config_loopback_t)NULL,
-	.format_fw_ver	= (format_fw_ver_t)bnx2x_format_ver,
-	.hw_reset	= (hw_reset_t)NULL,
-	.set_link_led	= (set_link_led_t)NULL,
-	.phy_specific_func = (phy_specific_func_t)NULL
+	.config_init	= bnx2x_8706_config_init,
+	.read_status	= bnx2x_8706_read_status,
+	.link_reset	= bnx2x_common_ext_link_reset,
+	.config_loopback = NULL,
+	.format_fw_ver	= bnx2x_format_ver,
+	.hw_reset	= NULL,
+	.set_link_led	= NULL,
+	.phy_specific_func = NULL
 };
 
 static const struct bnx2x_phy phy_8726 = {
@@ -11899,14 +11888,14 @@ static const struct bnx2x_phy phy_8726 = {
 	.speed_cap_mask	= 0,
 	.req_duplex	= 0,
 	.rsrv		= 0,
-	.config_init	= (config_init_t)bnx2x_8726_config_init,
-	.read_status	= (read_status_t)bnx2x_8726_read_status,
-	.link_reset	= (link_reset_t)bnx2x_8726_link_reset,
-	.config_loopback = (config_loopback_t)bnx2x_8726_config_loopback,
-	.format_fw_ver	= (format_fw_ver_t)bnx2x_format_ver,
-	.hw_reset	= (hw_reset_t)NULL,
-	.set_link_led	= (set_link_led_t)NULL,
-	.phy_specific_func = (phy_specific_func_t)NULL
+	.config_init	= bnx2x_8726_config_init,
+	.read_status	= bnx2x_8726_read_status,
+	.link_reset	= bnx2x_8726_link_reset,
+	.config_loopback = bnx2x_8726_config_loopback,
+	.format_fw_ver	= bnx2x_format_ver,
+	.hw_reset	= NULL,
+	.set_link_led	= NULL,
+	.phy_specific_func = NULL
 };
 
 static const struct bnx2x_phy phy_8727 = {
@@ -11930,14 +11919,14 @@ static const struct bnx2x_phy phy_8727 = {
 	.speed_cap_mask	= 0,
 	.req_duplex	= 0,
 	.rsrv		= 0,
-	.config_init	= (config_init_t)bnx2x_8727_config_init,
-	.read_status	= (read_status_t)bnx2x_8727_read_status,
-	.link_reset	= (link_reset_t)bnx2x_8727_link_reset,
-	.config_loopback = (config_loopback_t)NULL,
-	.format_fw_ver	= (format_fw_ver_t)bnx2x_format_ver,
-	.hw_reset	= (hw_reset_t)bnx2x_8727_hw_reset,
-	.set_link_led	= (set_link_led_t)bnx2x_8727_set_link_led,
-	.phy_specific_func = (phy_specific_func_t)bnx2x_8727_specific_func
+	.config_init	= bnx2x_8727_config_init,
+	.read_status	= bnx2x_8727_read_status,
+	.link_reset	= bnx2x_8727_link_reset,
+	.config_loopback = NULL,
+	.format_fw_ver	= bnx2x_format_ver,
+	.hw_reset	= bnx2x_8727_hw_reset,
+	.set_link_led	= bnx2x_8727_set_link_led,
+	.phy_specific_func = bnx2x_8727_specific_func
 };
 static const struct bnx2x_phy phy_8481 = {
 	.type		= PORT_HW_CFG_XGXS_EXT_PHY_TYPE_BCM8481,
@@ -11965,14 +11954,14 @@ static const struct bnx2x_phy phy_8481 = {
 	.speed_cap_mask	= 0,
 	.req_duplex	= 0,
 	.rsrv		= 0,
-	.config_init	= (config_init_t)bnx2x_8481_config_init,
-	.read_status	= (read_status_t)bnx2x_848xx_read_status,
-	.link_reset	= (link_reset_t)bnx2x_8481_link_reset,
-	.config_loopback = (config_loopback_t)NULL,
-	.format_fw_ver	= (format_fw_ver_t)bnx2x_848xx_format_ver,
-	.hw_reset	= (hw_reset_t)bnx2x_8481_hw_reset,
-	.set_link_led	= (set_link_led_t)bnx2x_848xx_set_link_led,
-	.phy_specific_func = (phy_specific_func_t)NULL
+	.config_init	= bnx2x_8481_config_init,
+	.read_status	= bnx2x_848xx_read_status,
+	.link_reset	= bnx2x_8481_link_reset,
+	.config_loopback = NULL,
+	.format_fw_ver	= bnx2x_848xx_format_ver,
+	.hw_reset	= bnx2x_8481_hw_reset,
+	.set_link_led	= bnx2x_848xx_set_link_led,
+	.phy_specific_func = NULL
 };
 
 static const struct bnx2x_phy phy_84823 = {
@@ -12002,14 +11991,14 @@ static const struct bnx2x_phy phy_84823 = {
 	.speed_cap_mask	= 0,
 	.req_duplex	= 0,
 	.rsrv		= 0,
-	.config_init	= (config_init_t)bnx2x_848x3_config_init,
-	.read_status	= (read_status_t)bnx2x_848xx_read_status,
-	.link_reset	= (link_reset_t)bnx2x_848x3_link_reset,
-	.config_loopback = (config_loopback_t)NULL,
-	.format_fw_ver	= (format_fw_ver_t)bnx2x_848xx_format_ver,
-	.hw_reset	= (hw_reset_t)NULL,
-	.set_link_led	= (set_link_led_t)bnx2x_848xx_set_link_led,
-	.phy_specific_func = (phy_specific_func_t)bnx2x_848xx_specific_func
+	.config_init	= bnx2x_848x3_config_init,
+	.read_status	= bnx2x_848xx_read_status,
+	.link_reset	= bnx2x_848x3_link_reset,
+	.config_loopback = NULL,
+	.format_fw_ver	= bnx2x_848xx_format_ver,
+	.hw_reset	= NULL,
+	.set_link_led	= bnx2x_848xx_set_link_led,
+	.phy_specific_func = bnx2x_848xx_specific_func
 };
 
 static const struct bnx2x_phy phy_84833 = {
@@ -12037,14 +12026,14 @@ static const struct bnx2x_phy phy_84833 = {
 	.speed_cap_mask	= 0,
 	.req_duplex	= 0,
 	.rsrv		= 0,
-	.config_init	= (config_init_t)bnx2x_848x3_config_init,
-	.read_status	= (read_status_t)bnx2x_848xx_read_status,
-	.link_reset	= (link_reset_t)bnx2x_848x3_link_reset,
-	.config_loopback = (config_loopback_t)NULL,
-	.format_fw_ver	= (format_fw_ver_t)bnx2x_848xx_format_ver,
-	.hw_reset	= (hw_reset_t)bnx2x_84833_hw_reset_phy,
-	.set_link_led	= (set_link_led_t)bnx2x_848xx_set_link_led,
-	.phy_specific_func = (phy_specific_func_t)bnx2x_848xx_specific_func
+	.config_init	= bnx2x_848x3_config_init,
+	.read_status	= bnx2x_848xx_read_status,
+	.link_reset	= bnx2x_848x3_link_reset,
+	.config_loopback = NULL,
+	.format_fw_ver	= bnx2x_848xx_format_ver,
+	.hw_reset	= bnx2x_84833_hw_reset_phy,
+	.set_link_led	= bnx2x_848xx_set_link_led,
+	.phy_specific_func = bnx2x_848xx_specific_func
 };
 
 static const struct bnx2x_phy phy_84834 = {
@@ -12071,14 +12060,14 @@ static const struct bnx2x_phy phy_84834 = {
 	.speed_cap_mask	= 0,
 	.req_duplex	= 0,
 	.rsrv		= 0,
-	.config_init	= (config_init_t)bnx2x_848x3_config_init,
-	.read_status	= (read_status_t)bnx2x_848xx_read_status,
-	.link_reset	= (link_reset_t)bnx2x_848x3_link_reset,
-	.config_loopback = (config_loopback_t)NULL,
-	.format_fw_ver	= (format_fw_ver_t)bnx2x_848xx_format_ver,
-	.hw_reset	= (hw_reset_t)bnx2x_84833_hw_reset_phy,
-	.set_link_led	= (set_link_led_t)bnx2x_848xx_set_link_led,
-	.phy_specific_func = (phy_specific_func_t)bnx2x_848xx_specific_func
+	.config_init	= bnx2x_848x3_config_init,
+	.read_status	= bnx2x_848xx_read_status,
+	.link_reset	= bnx2x_848x3_link_reset,
+	.config_loopback = NULL,
+	.format_fw_ver	= bnx2x_848xx_format_ver,
+	.hw_reset	= bnx2x_84833_hw_reset_phy,
+	.set_link_led	= bnx2x_848xx_set_link_led,
+	.phy_specific_func = bnx2x_848xx_specific_func
 };
 
 static const struct bnx2x_phy phy_84858 = {
@@ -12105,14 +12094,14 @@ static const struct bnx2x_phy phy_84858 = {
 	.speed_cap_mask	= 0,
 	.req_duplex	= 0,
 	.rsrv		= 0,
-	.config_init	= (config_init_t)bnx2x_848x3_config_init,
-	.read_status	= (read_status_t)bnx2x_848xx_read_status,
-	.link_reset	= (link_reset_t)bnx2x_848x3_link_reset,
-	.config_loopback = (config_loopback_t)NULL,
-	.format_fw_ver	= (format_fw_ver_t)bnx2x_8485x_format_ver,
-	.hw_reset	= (hw_reset_t)bnx2x_84833_hw_reset_phy,
-	.set_link_led	= (set_link_led_t)bnx2x_848xx_set_link_led,
-	.phy_specific_func = (phy_specific_func_t)bnx2x_848xx_specific_func
+	.config_init	= bnx2x_848x3_config_init,
+	.read_status	= bnx2x_848xx_read_status,
+	.link_reset	= bnx2x_848x3_link_reset,
+	.config_loopback = NULL,
+	.format_fw_ver	= bnx2x_8485x_format_ver,
+	.hw_reset	= bnx2x_84833_hw_reset_phy,
+	.set_link_led	= bnx2x_848xx_set_link_led,
+	.phy_specific_func = bnx2x_848xx_specific_func
 };
 
 static const struct bnx2x_phy phy_54618se = {
@@ -12139,14 +12128,14 @@ static const struct bnx2x_phy phy_54618se = {
 	.speed_cap_mask	= 0,
 	/* req_duplex = */0,
 	/* rsrv = */0,
-	.config_init	= (config_init_t)bnx2x_54618se_config_init,
-	.read_status	= (read_status_t)bnx2x_54618se_read_status,
-	.link_reset	= (link_reset_t)bnx2x_54618se_link_reset,
-	.config_loopback = (config_loopback_t)bnx2x_54618se_config_loopback,
-	.format_fw_ver	= (format_fw_ver_t)NULL,
-	.hw_reset	= (hw_reset_t)NULL,
-	.set_link_led	= (set_link_led_t)bnx2x_5461x_set_link_led,
-	.phy_specific_func = (phy_specific_func_t)bnx2x_54618se_specific_func
+	.config_init	= bnx2x_54618se_config_init,
+	.read_status	= bnx2x_54618se_read_status,
+	.link_reset	= bnx2x_54618se_link_reset,
+	.config_loopback = bnx2x_54618se_config_loopback,
+	.format_fw_ver	= NULL,
+	.hw_reset	= NULL,
+	.set_link_led	= bnx2x_5461x_set_link_led,
+	.phy_specific_func = bnx2x_54618se_specific_func
 };
 /*****************************************************************/
 /*                                                               */
@@ -12521,11 +12510,13 @@ static void bnx2x_phy_def_cfg(struct link_params *params,
 	switch (link_config  & PORT_FEATURE_LINK_SPEED_MASK) {
 	case PORT_FEATURE_LINK_SPEED_10M_HALF:
 		phy->req_duplex = DUPLEX_HALF;
+		fallthrough;
 	case PORT_FEATURE_LINK_SPEED_10M_FULL:
 		phy->req_line_speed = SPEED_10;
 		break;
 	case PORT_FEATURE_LINK_SPEED_100M_HALF:
 		phy->req_duplex = DUPLEX_HALF;
+		fallthrough;
 	case PORT_FEATURE_LINK_SPEED_100M_FULL:
 		phy->req_line_speed = SPEED_100;
 		break;
@@ -12672,39 +12663,39 @@ static void bnx2x_init_bmac_loopback(struct link_params *params,
 				     struct link_vars *vars)
 {
 	struct bnx2x *bp = params->bp;
-		vars->link_up = 1;
-		vars->line_speed = SPEED_10000;
-		vars->duplex = DUPLEX_FULL;
-		vars->flow_ctrl = BNX2X_FLOW_CTRL_NONE;
-		vars->mac_type = MAC_TYPE_BMAC;
+	vars->link_up = 1;
+	vars->line_speed = SPEED_10000;
+	vars->duplex = DUPLEX_FULL;
+	vars->flow_ctrl = BNX2X_FLOW_CTRL_NONE;
+	vars->mac_type = MAC_TYPE_BMAC;
 
-		vars->phy_flags = PHY_XGXS_FLAG;
+	vars->phy_flags = PHY_XGXS_FLAG;
 
-		bnx2x_xgxs_deassert(params);
+	bnx2x_xgxs_deassert(params);
 
-		/* Set bmac loopback */
-		bnx2x_bmac_enable(params, vars, 1, 1);
+	/* Set bmac loopback */
+	bnx2x_bmac_enable(params, vars, 1, 1);
 
-		REG_WR(bp, NIG_REG_EGRESS_DRAIN0_MODE + params->port*4, 0);
+	REG_WR(bp, NIG_REG_EGRESS_DRAIN0_MODE + params->port * 4, 0);
 }
 
 static void bnx2x_init_emac_loopback(struct link_params *params,
 				     struct link_vars *vars)
 {
 	struct bnx2x *bp = params->bp;
-		vars->link_up = 1;
-		vars->line_speed = SPEED_1000;
-		vars->duplex = DUPLEX_FULL;
-		vars->flow_ctrl = BNX2X_FLOW_CTRL_NONE;
-		vars->mac_type = MAC_TYPE_EMAC;
+	vars->link_up = 1;
+	vars->line_speed = SPEED_1000;
+	vars->duplex = DUPLEX_FULL;
+	vars->flow_ctrl = BNX2X_FLOW_CTRL_NONE;
+	vars->mac_type = MAC_TYPE_EMAC;
 
-		vars->phy_flags = PHY_XGXS_FLAG;
+	vars->phy_flags = PHY_XGXS_FLAG;
 
-		bnx2x_xgxs_deassert(params);
-		/* Set bmac loopback */
-		bnx2x_emac_enable(params, vars, 1);
-		bnx2x_emac_program(params, vars);
-		REG_WR(bp, NIG_REG_EGRESS_DRAIN0_MODE + params->port*4, 0);
+	bnx2x_xgxs_deassert(params);
+	/* Set bmac loopback */
+	bnx2x_emac_enable(params, vars, 1);
+	bnx2x_emac_program(params, vars);
+	REG_WR(bp, NIG_REG_EGRESS_DRAIN0_MODE + params->port * 4, 0);
 }
 
 static void bnx2x_init_xmac_loopback(struct link_params *params,
@@ -13070,12 +13061,12 @@ int bnx2x_link_reset(struct link_params *params, struct link_vars *vars,
 		REG_WR(bp, NIG_REG_EGRESS_EMAC0_OUT_EN + port*4, 0);
 	}
 
-		if (!CHIP_IS_E3(bp)) {
-			bnx2x_set_bmac_rx(bp, params->chip_id, port, 0);
-		} else {
-			bnx2x_set_xmac_rxtx(params, 0);
-			bnx2x_set_umac_rxtx(params, 0);
-		}
+	if (!CHIP_IS_E3(bp)) {
+		bnx2x_set_bmac_rx(bp, params->chip_id, port, 0);
+	} else {
+		bnx2x_set_xmac_rxtx(params, 0);
+		bnx2x_set_umac_rxtx(params, 0);
+	}
 	/* Disable emac */
 	if (!CHIP_IS_E3(bp))
 		REG_WR(bp, NIG_REG_NIG_EMAC0_EN + port*4, 0);
@@ -13853,7 +13844,7 @@ static void bnx2x_check_kr2_wa(struct link_params *params,
 
 	/* Once KR2 was disabled, wait 5 seconds before checking KR2 recovery
 	 * Since some switches tend to reinit the AN process and clear the
-	 * the advertised BP/NP after ~2 seconds causing the KR2 to be disabled
+	 * advertised BP/NP after ~2 seconds causing the KR2 to be disabled
 	 * and recovered many times
 	 */
 	if (vars->check_kr2_recovery_cnt > 0) {
diff --git a/drivers/net/ethernet/broadcom/bnx2x/bnx2x_link.h b/drivers/net/ethernet/broadcom/bnx2x/bnx2x_link.h
index b7d251108c19..cae03c89dc73 100644
--- a/drivers/net/ethernet/broadcom/bnx2x/bnx2x_link.h
+++ b/drivers/net/ethernet/broadcom/bnx2x/bnx2x_link.h
@@ -62,6 +62,7 @@
 #define SFP_EEPROM_DIAG_TYPE_ADDR		0x5c
 #define SFP_EEPROM_DIAG_TYPE_SIZE		1
 #define SFP_EEPROM_DIAG_ADDR_CHANGE_REQ		(1<<2)
+#define SFP_EEPROM_DDM_IMPLEMENTED		(1<<6)
 #define SFP_EEPROM_SFF_8472_COMP_ADDR		0x5e
 #define SFP_EEPROM_SFF_8472_COMP_SIZE		1
 
@@ -126,15 +127,15 @@ struct link_vars;
 struct link_params;
 struct bnx2x_phy;
 
-typedef u8 (*config_init_t)(struct bnx2x_phy *phy, struct link_params *params,
-			    struct link_vars *vars);
+typedef void (*config_init_t)(struct bnx2x_phy *phy, struct link_params *params,
+			      struct link_vars *vars);
 typedef u8 (*read_status_t)(struct bnx2x_phy *phy, struct link_params *params,
 			    struct link_vars *vars);
 typedef void (*link_reset_t)(struct bnx2x_phy *phy,
 			     struct link_params *params);
 typedef void (*config_loopback_t)(struct bnx2x_phy *phy,
 				  struct link_params *params);
-typedef u8 (*format_fw_ver_t)(u32 raw, u8 *str, u16 *len);
+typedef int (*format_fw_ver_t)(u32 raw, u8 *str, u16 *len);
 typedef void (*hw_reset_t)(struct bnx2x_phy *phy, struct link_params *params);
 typedef void (*set_link_led_t)(struct bnx2x_phy *phy,
 			       struct link_params *params, u8 mode);
diff --git a/drivers/net/ethernet/broadcom/bnx2x/bnx2x_main.c b/drivers/net/ethernet/broadcom/bnx2x/bnx2x_main.c
index 14c236e5bdb1..f0f05d7315ac 100644
--- a/drivers/net/ethernet/broadcom/bnx2x/bnx2x_main.c
+++ b/drivers/net/ethernet/broadcom/bnx2x/bnx2x_main.c
@@ -29,7 +29,6 @@
 #include <linux/slab.h>
 #include <linux/interrupt.h>
 #include <linux/pci.h>
-#include <linux/aer.h>
 #include <linux/init.h>
 #include <linux/netdevice.h>
 #include <linux/etherdevice.h>
@@ -74,52 +73,60 @@
 	__stringify(BCM_5710_FW_MINOR_VERSION) "."	\
 	__stringify(BCM_5710_FW_REVISION_VERSION) "."	\
 	__stringify(BCM_5710_FW_ENGINEERING_VERSION)
+
+#define FW_FILE_VERSION_V15				\
+	__stringify(BCM_5710_FW_MAJOR_VERSION) "."      \
+	__stringify(BCM_5710_FW_MINOR_VERSION) "."	\
+	__stringify(BCM_5710_FW_REVISION_VERSION_V15) "."	\
+	__stringify(BCM_5710_FW_ENGINEERING_VERSION)
+
 #define FW_FILE_NAME_E1		"bnx2x/bnx2x-e1-" FW_FILE_VERSION ".fw"
 #define FW_FILE_NAME_E1H	"bnx2x/bnx2x-e1h-" FW_FILE_VERSION ".fw"
 #define FW_FILE_NAME_E2		"bnx2x/bnx2x-e2-" FW_FILE_VERSION ".fw"
+#define FW_FILE_NAME_E1_V15	"bnx2x/bnx2x-e1-" FW_FILE_VERSION_V15 ".fw"
+#define FW_FILE_NAME_E1H_V15	"bnx2x/bnx2x-e1h-" FW_FILE_VERSION_V15 ".fw"
+#define FW_FILE_NAME_E2_V15	"bnx2x/bnx2x-e2-" FW_FILE_VERSION_V15 ".fw"
 
 /* Time in jiffies before concluding the transmitter is hung */
 #define TX_TIMEOUT		(5*HZ)
 
-static char version[] =
-	"QLogic 5771x/578xx 10/20-Gigabit Ethernet Driver "
-	DRV_MODULE_NAME " " DRV_MODULE_VERSION " (" DRV_MODULE_RELDATE ")\n";
-
 MODULE_AUTHOR("Eliezer Tamir");
 MODULE_DESCRIPTION("QLogic "
 		   "BCM57710/57711/57711E/"
 		   "57712/57712_MF/57800/57800_MF/57810/57810_MF/"
 		   "57840/57840_MF Driver");
 MODULE_LICENSE("GPL");
-MODULE_VERSION(DRV_MODULE_VERSION);
 MODULE_FIRMWARE(FW_FILE_NAME_E1);
 MODULE_FIRMWARE(FW_FILE_NAME_E1H);
 MODULE_FIRMWARE(FW_FILE_NAME_E2);
+MODULE_FIRMWARE(FW_FILE_NAME_E1_V15);
+MODULE_FIRMWARE(FW_FILE_NAME_E1H_V15);
+MODULE_FIRMWARE(FW_FILE_NAME_E2_V15);
 
 int bnx2x_num_queues;
-module_param_named(num_queues, bnx2x_num_queues, int, S_IRUGO);
+module_param_named(num_queues, bnx2x_num_queues, int, 0444);
 MODULE_PARM_DESC(num_queues,
 		 " Set number of queues (default is as a number of CPUs)");
 
 static int disable_tpa;
-module_param(disable_tpa, int, S_IRUGO);
+module_param(disable_tpa, int, 0444);
 MODULE_PARM_DESC(disable_tpa, " Disable the TPA (LRO) feature");
 
 static int int_mode;
-module_param(int_mode, int, S_IRUGO);
+module_param(int_mode, int, 0444);
 MODULE_PARM_DESC(int_mode, " Force interrupt mode other than MSI-X "
 				"(1 INT#x; 2 MSI)");
 
 static int dropless_fc;
-module_param(dropless_fc, int, S_IRUGO);
+module_param(dropless_fc, int, 0444);
 MODULE_PARM_DESC(dropless_fc, " Pause on exhausted host ring");
 
 static int mrrs = -1;
-module_param(mrrs, int, S_IRUGO);
+module_param(mrrs, int, 0444);
 MODULE_PARM_DESC(mrrs, " Force Max Read Req Size (0..3) (for debug)");
 
 static int debug;
-module_param(debug, int, S_IRUGO);
+module_param(debug, int, 0444);
 MODULE_PARM_DESC(debug, " Default debug msglevel");
 
 static struct workqueue_struct *bnx2x_wq;
@@ -281,9 +288,16 @@ static const struct pci_device_id bnx2x_pci_tbl[] = {
 
 MODULE_DEVICE_TABLE(pci, bnx2x_pci_tbl);
 
+const u32 dmae_reg_go_c[] = {
+	DMAE_REG_GO_C0, DMAE_REG_GO_C1, DMAE_REG_GO_C2, DMAE_REG_GO_C3,
+	DMAE_REG_GO_C4, DMAE_REG_GO_C5, DMAE_REG_GO_C6, DMAE_REG_GO_C7,
+	DMAE_REG_GO_C8, DMAE_REG_GO_C9, DMAE_REG_GO_C10, DMAE_REG_GO_C11,
+	DMAE_REG_GO_C12, DMAE_REG_GO_C13, DMAE_REG_GO_C14, DMAE_REG_GO_C15
+};
+
 /* Global resources for unloading a previously loaded device */
 #define BNX2X_PREV_WAIT_NEEDED 1
-static DEFINE_SEMAPHORE(bnx2x_prev_sem);
+static DEFINE_SEMAPHORE(bnx2x_prev_sem, 1);
 static LIST_HEAD(bnx2x_prev_list);
 
 /* Forward declaration */
@@ -745,9 +759,7 @@ static int bnx2x_mc_assert(struct bnx2x *bp)
 		  CHIP_IS_E1(bp) ? "everest1" :
 		  CHIP_IS_E1H(bp) ? "everest1h" :
 		  CHIP_IS_E2(bp) ? "everest2" : "everest3",
-		  BCM_5710_FW_MAJOR_VERSION,
-		  BCM_5710_FW_MINOR_VERSION,
-		  BCM_5710_FW_REVISION_VERSION);
+		  bp->fw_major, bp->fw_minor, bp->fw_rev);
 
 	return rc;
 }
@@ -869,9 +881,6 @@ static void bnx2x_hc_int_disable(struct bnx2x *bp)
 	   "write %x to HC %d (addr 0x%x)\n",
 	   val, port, addr);
 
-	/* flush all outstanding writes */
-	mmiowb();
-
 	REG_WR(bp, addr, val);
 	if (REG_RD(bp, addr) != val)
 		BNX2X_ERR("BUG! Proper val not read from IGU!\n");
@@ -887,9 +896,6 @@ static void bnx2x_igu_int_disable(struct bnx2x *bp)
 
 	DP(NETIF_MSG_IFDOWN, "write %x to IGU\n", val);
 
-	/* flush all outstanding writes */
-	mmiowb();
-
 	REG_WR(bp, IGU_REG_PF_CONFIGURATION, val);
 	if (REG_RD(bp, IGU_REG_PF_CONFIGURATION) != val)
 		BNX2X_ERR("BUG! Proper val not read from IGU!\n");
@@ -1180,9 +1186,18 @@ void bnx2x_panic_dump(struct bnx2x *bp, bool disable_int)
 	}
 #endif
 	if (IS_PF(bp)) {
+		int tmp_msg_en = bp->msg_enable;
+
 		bnx2x_fw_dump(bp);
+		bp->msg_enable |= NETIF_MSG_HW;
+		BNX2X_ERR("Idle check (1st round) ----------\n");
+		bnx2x_idle_chk(bp);
+		BNX2X_ERR("Idle check (2nd round) ----------\n");
+		bnx2x_idle_chk(bp);
+		bp->msg_enable = tmp_msg_en;
 		bnx2x_mc_assert(bp);
 	}
+
 	BNX2X_ERR("end crash dump -----------------\n");
 }
 
@@ -1390,7 +1405,6 @@ int bnx2x_send_final_clnup(struct bnx2x *bp, u8 clnup_func, u32 poll_cnt)
 	u32 op_gen_command = 0;
 	u32 comp_addr = BAR_CSTRORM_INTMEM +
 			CSTORM_FINAL_CLEANUP_COMPLETE_OFFSET(clnup_func);
-	int ret = 0;
 
 	if (REG_RD(bp, comp_addr)) {
 		BNX2X_ERR("Cleanup complete was not 0 before sending\n");
@@ -1415,7 +1429,7 @@ int bnx2x_send_final_clnup(struct bnx2x *bp, u8 clnup_func, u32 poll_cnt)
 	/* Zero completion for next FLR */
 	REG_WR(bp, comp_addr, 0);
 
-	return ret;
+	return 0;
 }
 
 u8 bnx2x_is_pcie_pending(struct pci_dev *dev)
@@ -1595,7 +1609,6 @@ static void bnx2x_hc_int_enable(struct bnx2x *bp)
 	/*
 	 * Ensure that HC_CONFIG is written before leading/trailing edge config
 	 */
-	mmiowb();
 	barrier();
 
 	if (!CHIP_IS_E1(bp)) {
@@ -1611,9 +1624,6 @@ static void bnx2x_hc_int_enable(struct bnx2x *bp)
 		REG_WR(bp, HC_REG_TRAILING_EDGE_0 + port*8, val);
 		REG_WR(bp, HC_REG_LEADING_EDGE_0 + port*8, val);
 	}
-
-	/* Make sure that interrupts are indeed enabled from here on */
-	mmiowb();
 }
 
 static void bnx2x_igu_int_enable(struct bnx2x *bp)
@@ -1674,9 +1684,6 @@ static void bnx2x_igu_int_enable(struct bnx2x *bp)
 
 	REG_WR(bp, IGU_REG_TRAILING_EDGE_LATCH, val);
 	REG_WR(bp, IGU_REG_LEADING_EDGE_LATCH, val);
-
-	/* Make sure that interrupts are indeed enabled from here on */
-	mmiowb();
 }
 
 void bnx2x_int_enable(struct bnx2x *bp)
@@ -1761,7 +1768,7 @@ static bool bnx2x_trylock_hw_lock(struct bnx2x *bp, u32 resource)
  * @bp:	driver handle
  *
  * Returns the recovery leader resource id according to the engine this function
- * belongs to. Currently only only 2 engines is supported.
+ * belongs to. Currently only 2 engines is supported.
  */
 static int bnx2x_get_leader_lock_resource(struct bnx2x *bp)
 {
@@ -2925,6 +2932,10 @@ static void bnx2x_handle_update_svid_cmd(struct bnx2x *bp)
 	func_params.f_obj = &bp->func_obj;
 	func_params.cmd = BNX2X_F_CMD_SWITCH_UPDATE;
 
+	/* Prepare parameters for function state transitions */
+	__set_bit(RAMROD_COMP_WAIT, &func_params.ramrod_flags);
+	__set_bit(RAMROD_RETRY, &func_params.ramrod_flags);
+
 	if (IS_MF_UFP(bp) || IS_MF_BD(bp)) {
 		int func = BP_ABS_FUNC(bp);
 		u32 val;
@@ -3084,9 +3095,9 @@ void bnx2x_func_init(struct bnx2x *bp, struct bnx2x_func_init_params *p)
 /**
  * bnx2x_get_common_flags - Return common flags
  *
- * @bp		device handle
- * @fp		queue handle
- * @zero_stats	TRUE if statistics zeroing is needed
+ * @bp:		device handle
+ * @fp:		queue handle
+ * @zero_stats:	TRUE if statistics zeroing is needed
  *
  * Return the flags that are common for the Tx-only and not normal connections.
  */
@@ -3373,7 +3384,7 @@ static void bnx2x_drv_info_ether_stat(struct bnx2x *bp)
 		&bp->sp_objs->mac_obj;
 	int i;
 
-	strlcpy(ether_stat->version, DRV_MODULE_VERSION,
+	strscpy(ether_stat->version, DRV_MODULE_VERSION,
 		ETH_STAT_INFO_VERSION_LEN);
 
 	/* get DRV_INFO_ETH_STAT_NUM_MACS_REQUIRED macs, placing them in the
@@ -3536,6 +3547,16 @@ static void bnx2x_drv_info_iscsi_stat(struct bnx2x *bp)
  */
 static void bnx2x_config_mf_bw(struct bnx2x *bp)
 {
+	/* Workaround for MFW bug.
+	 * MFW is not supposed to generate BW attention in
+	 * single function mode.
+	 */
+	if (!IS_MF(bp)) {
+		DP(BNX2X_MSG_MCP,
+		   "Ignoring MF BW config in single function mode\n");
+		return;
+	}
+
 	if (bp->link_vars.link_up) {
 		bnx2x_cmng_fns_init(bp, true, CMNG_FNS_MINMAX);
 		bnx2x_link_sync_notify(bp);
@@ -3817,9 +3838,8 @@ static void bnx2x_sp_prod_update(struct bnx2x *bp)
 	 */
 	mb();
 
-	REG_WR16(bp, BAR_XSTRORM_INTMEM + XSTORM_SPQ_PROD_OFFSET(func),
-		 bp->spq_prod_idx);
-	mmiowb();
+	REG_WR16_RELAXED(bp, BAR_XSTRORM_INTMEM + XSTORM_SPQ_PROD_OFFSET(func),
+			 bp->spq_prod_idx);
 }
 
 /**
@@ -4301,7 +4321,8 @@ static void bnx2x_attn_int_deasserted3(struct bnx2x *bp, u32 attn)
 				bnx2x_handle_eee_event(bp);
 
 			if (val & DRV_STATUS_OEM_UPDATE_SVID)
-				bnx2x_handle_update_svid_cmd(bp);
+				bnx2x_schedule_sp_rtnl(bp,
+					BNX2X_SP_RTNL_UPDATE_SVID, 0);
 
 			if (bp->link_vars.periodic_flags &
 			    PERIODIC_FLAGS_LINK_EVENT) {
@@ -5229,7 +5250,6 @@ static void bnx2x_update_eq_prod(struct bnx2x *bp, u16 prod)
 {
 	/* No memory barriers */
 	storm_memset_eq_prod(bp, prod, BP_FUNC(bp));
-	mmiowb(); /* keep prod updates ordered */
 }
 
 static int  bnx2x_cnic_handle_cfc_del(struct bnx2x *bp, u32 cid,
@@ -5761,9 +5781,9 @@ void bnx2x_drv_pulse(struct bnx2x *bp)
 		 bp->fw_drv_pulse_wr_seq);
 }
 
-static void bnx2x_timer(unsigned long data)
+static void bnx2x_timer(struct timer_list *t)
 {
-	struct bnx2x *bp = (struct bnx2x *) data;
+	struct bnx2x *bp = timer_container_of(bp, t, timer);
 
 	if (!netif_running(bp->dev))
 		return;
@@ -6302,11 +6322,11 @@ static void bnx2x_init_internal(struct bnx2x *bp, u32 load_code)
 	case FW_MSG_CODE_DRV_LOAD_COMMON:
 	case FW_MSG_CODE_DRV_LOAD_COMMON_CHIP:
 		bnx2x_init_internal_common(bp);
-		/* no break */
+		fallthrough;
 
 	case FW_MSG_CODE_DRV_LOAD_PORT:
 		/* nothing to do */
-		/* no break */
+		fallthrough;
 
 	case FW_MSG_CODE_DRV_LOAD_FUNCTION:
 		/* internal memory per function is
@@ -6498,7 +6518,6 @@ void bnx2x_nic_init_cnic(struct bnx2x *bp)
 
 	/* flush all */
 	mb();
-	mmiowb();
 }
 
 void bnx2x_pre_irq_nic_init(struct bnx2x *bp)
@@ -6538,7 +6557,6 @@ void bnx2x_post_irq_nic_init(struct bnx2x *bp, u32 load_code)
 
 	/* flush all before enabling interrupts */
 	mb();
-	mmiowb();
 
 	bnx2x_int_enable(bp);
 
@@ -7713,6 +7731,9 @@ static int bnx2x_init_hw_port(struct bnx2x *bp)
 		REG_WR(bp, reg_addr, val);
 	}
 
+	if (CHIP_IS_E3B0(bp))
+		bp->flags |= PTP_SUPPORTED;
+
 	return 0;
 }
 
@@ -7757,12 +7778,10 @@ void bnx2x_igu_clear_sb_gen(struct bnx2x *bp, u8 func, u8 idu_sb_id, bool is_pf)
 	DP(NETIF_MSG_HW, "write 0x%08x to IGU(via GRC) addr 0x%x\n",
 			 data, igu_addr_data);
 	REG_WR(bp, igu_addr_data, data);
-	mmiowb();
 	barrier();
 	DP(NETIF_MSG_HW, "write 0x%08x to IGU(via GRC) addr 0x%x\n",
 			  ctl, igu_addr_ctl);
 	REG_WR(bp, igu_addr_ctl, ctl);
-	mmiowb();
 	barrier();
 
 	/* wait for clean up to finish */
@@ -8408,7 +8427,7 @@ alloc_mem_err:
  * Init service functions
  */
 
-int bnx2x_set_mac_one(struct bnx2x *bp, u8 *mac,
+int bnx2x_set_mac_one(struct bnx2x *bp, const u8 *mac,
 		      struct bnx2x_vlan_mac_obj *obj, bool set,
 		      int mac_type, unsigned long *ramrod_flags)
 {
@@ -8462,6 +8481,7 @@ int bnx2x_set_vlan_one(struct bnx2x *bp, u16 vlan,
 	/* Fill a user request section if needed */
 	if (!test_bit(RAMROD_CONT, ramrod_flags)) {
 		ramrod_param.user_req.u.vlan.vlan = vlan;
+		__set_bit(BNX2X_VLAN, &ramrod_param.user_req.vlan_mac_flags);
 		/* Set the command: ADD or DEL */
 		if (set)
 			ramrod_param.user_req.cmd = BNX2X_VLAN_MAC_ADD;
@@ -8482,6 +8502,34 @@ int bnx2x_set_vlan_one(struct bnx2x *bp, u16 vlan,
 	return rc;
 }
 
+void bnx2x_clear_vlan_info(struct bnx2x *bp)
+{
+	struct bnx2x_vlan_entry *vlan;
+
+	/* Mark that hw forgot all entries */
+	list_for_each_entry(vlan, &bp->vlan_reg, link)
+		vlan->hw = false;
+
+	bp->vlan_cnt = 0;
+}
+
+static int bnx2x_del_all_vlans(struct bnx2x *bp)
+{
+	struct bnx2x_vlan_mac_obj *vlan_obj = &bp->sp_objs[0].vlan_obj;
+	unsigned long ramrod_flags = 0, vlan_flags = 0;
+	int rc;
+
+	__set_bit(RAMROD_COMP_WAIT, &ramrod_flags);
+	__set_bit(BNX2X_VLAN, &vlan_flags);
+	rc = vlan_obj->delete_all(bp, vlan_obj, &vlan_flags, &ramrod_flags);
+	if (rc)
+		return rc;
+
+	bnx2x_clear_vlan_info(bp);
+
+	return 0;
+}
+
 int bnx2x_del_all_macs(struct bnx2x *bp,
 		       struct bnx2x_vlan_mac_obj *mac_obj,
 		       int mac_type, bool wait_for_comp)
@@ -8561,11 +8609,11 @@ int bnx2x_set_int_mode(struct bnx2x *bp)
 			       bp->num_queues,
 			       1 + bp->num_cnic_queues);
 
-		/* falling through... */
+		fallthrough;
 	case BNX2X_INT_MODE_MSI:
 		bnx2x_enable_msi(bp);
 
-		/* falling through... */
+		fallthrough;
 	case BNX2X_INT_MODE_INTX:
 		bp->num_ethernet_queues = 1;
 		bp->num_queues = bp->num_ethernet_queues + bp->num_cnic_queues;
@@ -9108,7 +9156,7 @@ u32 bnx2x_send_unload_req(struct bnx2x *bp, int unload_mode)
 
 	else if (bp->wol) {
 		u32 emac_base = port ? GRCBASE_EMAC1 : GRCBASE_EMAC0;
-		u8 *mac_addr = bp->dev->dev_addr;
+		const u8 *mac_addr = bp->dev->dev_addr;
 		struct pci_dev *pdev = bp->pdev;
 		u32 val;
 		u16 pmc;
@@ -9320,6 +9368,17 @@ void bnx2x_chip_cleanup(struct bnx2x *bp, int unload_mode, bool keep_link)
 		BNX2X_ERR("Failed to schedule DEL commands for UC MACs list: %d\n",
 			  rc);
 
+	/* The whole *vlan_obj structure may be not initialized if VLAN
+	 * filtering offload is not supported by hardware. Currently this is
+	 * true for all hardware covered by CHIP_IS_E1x().
+	 */
+	if (!CHIP_IS_E1x(bp)) {
+		/* Remove all currently configured VLANs */
+		rc = bnx2x_del_all_vlans(bp);
+		if (rc < 0)
+			BNX2X_ERR("Failed to delete all VLANs\n");
+	}
+
 	/* Disable LLH */
 	if (!CHIP_IS_E1(bp))
 		REG_WR(bp, NIG_REG_LLH0_FUNC_EN + port*8, 0);
@@ -9332,7 +9391,7 @@ void bnx2x_chip_cleanup(struct bnx2x *bp, int unload_mode, bool keep_link)
 	/* Schedule the rx_mode command */
 	if (test_bit(BNX2X_FILTER_RX_MODE_PENDING, &bp->sp_state))
 		set_bit(BNX2X_FILTER_RX_MODE_SCHED, &bp->sp_state);
-	else
+	else if (bp->slowpath)
 		bnx2x_set_storm_rx_mode(bp);
 
 	/* Cleanup multicast configuration */
@@ -9407,18 +9466,26 @@ unload_error:
 	 * function stop ramrod is sent, since as part of this ramrod FW access
 	 * PTP registers.
 	 */
-	if (bp->flags & PTP_SUPPORTED)
+	if (bp->flags & PTP_SUPPORTED) {
 		bnx2x_stop_ptp(bp);
+		if (bp->ptp_clock) {
+			ptp_clock_unregister(bp->ptp_clock);
+			bp->ptp_clock = NULL;
+		}
+	}
 
-	/* Disable HW interrupts, NAPI */
-	bnx2x_netif_stop(bp, 1);
-	/* Delete all NAPI objects */
-	bnx2x_del_all_napi(bp);
-	if (CNIC_LOADED(bp))
-		bnx2x_del_all_napi_cnic(bp);
+	if (!bp->nic_stopped) {
+		/* Disable HW interrupts, NAPI */
+		bnx2x_netif_stop(bp, 1);
+		/* Delete all NAPI objects */
+		bnx2x_del_all_napi(bp);
+		if (CNIC_LOADED(bp))
+			bnx2x_del_all_napi_cnic(bp);
 
-	/* Release IRQs */
-	bnx2x_free_irq(bp);
+		/* Release IRQs */
+		bnx2x_free_irq(bp);
+		bp->nic_stopped = true;
+	}
 
 	/* Reset the chip, unless PCI function is offline. If we reach this
 	 * point following a PCI error handling, it means device is really
@@ -9494,7 +9561,6 @@ static void bnx2x_set_234_gates(struct bnx2x *bp, bool close)
 
 	DP(NETIF_MSG_HW | NETIF_MSG_IFUP, "%s gates #2, #3 and #4\n",
 		close ? "closing" : "opening");
-	mmiowb();
 }
 
 #define SHARED_MF_CLP_MAGIC  0x80000000 /* `magic' bit */
@@ -9578,6 +9644,15 @@ static int bnx2x_init_shmem(struct bnx2x *bp)
 
 	do {
 		bp->common.shmem_base = REG_RD(bp, MISC_REG_SHARED_MEM_ADDR);
+
+		/* If we read all 0xFFs, means we are in PCI error state and
+		 * should bail out to avoid crashes on adapter's FW reads.
+		 */
+		if (bp->common.shmem_base == 0xFFFFFFFF) {
+			bp->flags |= NO_MCP_FLAG;
+			return -ENODEV;
+		}
+
 		if (bp->common.shmem_base) {
 			val = SHMEM_RD(bp, validity_map[BP_PORT(bp)]);
 			if (val & SHR_MEM_VALIDITY_MB)
@@ -9609,7 +9684,6 @@ static void bnx2x_pxp_prep(struct bnx2x *bp)
 	if (!CHIP_IS_E1(bp)) {
 		REG_WR(bp, PXP2_REG_RD_START_INIT, 0);
 		REG_WR(bp, PXP2_REG_RQ_RBC_DONE, 0);
-		mmiowb();
 	}
 }
 
@@ -9709,16 +9783,13 @@ static void bnx2x_process_kill_chip_reset(struct bnx2x *bp, bool global)
 	       reset_mask1 & (~not_reset_mask1));
 
 	barrier();
-	mmiowb();
 
 	REG_WR(bp, GRCBASE_MISC + MISC_REGISTERS_RESET_REG_2_SET,
 	       reset_mask2 & (~stay_reset2));
 
 	barrier();
-	mmiowb();
 
 	REG_WR(bp, GRCBASE_MISC + MISC_REGISTERS_RESET_REG_1_SET, reset_mask1);
-	mmiowb();
 }
 
 /**
@@ -9802,9 +9873,6 @@ static int bnx2x_process_kill(struct bnx2x *bp, bool global)
 	REG_WR(bp, MISC_REG_UNPREPARED, 0);
 	barrier();
 
-	/* Make sure all is written to the chip before the reset */
-	mmiowb();
-
 	/* Wait for 1ms to empty GLUE and PCI-E core queues,
 	 * PSWHST, GRC and PSWRD Tetris buffer.
 	 */
@@ -9931,10 +9999,18 @@ static void bnx2x_recovery_failed(struct bnx2x *bp)
  */
 static void bnx2x_parity_recover(struct bnx2x *bp)
 {
-	bool global = false;
 	u32 error_recovered, error_unrecovered;
-	bool is_parity;
+	bool is_parity, global = false;
+#ifdef CONFIG_BNX2X_SRIOV
+	int vf_idx;
 
+	for (vf_idx = 0; vf_idx < bp->requested_nr_virtfn; vf_idx++) {
+		struct bnx2x_virtf *vf = BP_VF(bp, vf_idx);
+
+		if (vf)
+			vf->state = VF_LOST;
+	}
+#endif
 	DP(NETIF_MSG_HW, "Handling parity\n");
 	while (1) {
 		switch (bp->recovery_state) {
@@ -10088,7 +10164,6 @@ static int bnx2x_udp_port_update(struct bnx2x *bp)
 {
 	struct bnx2x_func_switch_update_params *switch_update_params;
 	struct bnx2x_func_state_params func_params = {NULL};
-	struct bnx2x_udp_tunnel *udp_tunnel;
 	u16 vxlan_port = 0, geneve_port = 0;
 	int rc;
 
@@ -10105,15 +10180,13 @@ static int bnx2x_udp_port_update(struct bnx2x *bp)
 	__set_bit(BNX2X_F_UPDATE_TUNNEL_CFG_CHNG,
 		  &switch_update_params->changes);
 
-	if (bp->udp_tunnel_ports[BNX2X_UDP_PORT_GENEVE].count) {
-		udp_tunnel = &bp->udp_tunnel_ports[BNX2X_UDP_PORT_GENEVE];
-		geneve_port = udp_tunnel->dst_port;
+	if (bp->udp_tunnel_ports[BNX2X_UDP_PORT_GENEVE]) {
+		geneve_port = bp->udp_tunnel_ports[BNX2X_UDP_PORT_GENEVE];
 		switch_update_params->geneve_dst_port = geneve_port;
 	}
 
-	if (bp->udp_tunnel_ports[BNX2X_UDP_PORT_VXLAN].count) {
-		udp_tunnel = &bp->udp_tunnel_ports[BNX2X_UDP_PORT_VXLAN];
-		vxlan_port = udp_tunnel->dst_port;
+	if (bp->udp_tunnel_ports[BNX2X_UDP_PORT_VXLAN]) {
+		vxlan_port = bp->udp_tunnel_ports[BNX2X_UDP_PORT_VXLAN];
 		switch_update_params->vxlan_dst_port = vxlan_port;
 	}
 
@@ -10133,94 +10206,26 @@ static int bnx2x_udp_port_update(struct bnx2x *bp)
 	return rc;
 }
 
-static void __bnx2x_add_udp_port(struct bnx2x *bp, u16 port,
-				 enum bnx2x_udp_port_type type)
-{
-	struct bnx2x_udp_tunnel *udp_port = &bp->udp_tunnel_ports[type];
-
-	if (!netif_running(bp->dev) || !IS_PF(bp) || CHIP_IS_E1x(bp))
-		return;
-
-	if (udp_port->count && udp_port->dst_port == port) {
-		udp_port->count++;
-		return;
-	}
-
-	if (udp_port->count) {
-		DP(BNX2X_MSG_SP,
-		   "UDP tunnel [%d] -  destination port limit reached\n",
-		   type);
-		return;
-	}
-
-	udp_port->dst_port = port;
-	udp_port->count = 1;
-	bnx2x_schedule_sp_rtnl(bp, BNX2X_SP_RTNL_CHANGE_UDP_PORT, 0);
-}
-
-static void __bnx2x_del_udp_port(struct bnx2x *bp, u16 port,
-				 enum bnx2x_udp_port_type type)
-{
-	struct bnx2x_udp_tunnel *udp_port = &bp->udp_tunnel_ports[type];
-
-	if (!IS_PF(bp) || CHIP_IS_E1x(bp))
-		return;
-
-	if (!udp_port->count || udp_port->dst_port != port) {
-		DP(BNX2X_MSG_SP, "Invalid UDP tunnel [%d] port\n",
-		   type);
-		return;
-	}
-
-	/* Remove reference, and make certain it's no longer in use */
-	udp_port->count--;
-	if (udp_port->count)
-		return;
-	udp_port->dst_port = 0;
-
-	if (netif_running(bp->dev))
-		bnx2x_schedule_sp_rtnl(bp, BNX2X_SP_RTNL_CHANGE_UDP_PORT, 0);
-	else
-		DP(BNX2X_MSG_SP, "Deleted UDP tunnel [%d] port %d\n",
-		   type, port);
-}
-
-static void bnx2x_udp_tunnel_add(struct net_device *netdev,
-				 struct udp_tunnel_info *ti)
+static int bnx2x_udp_tunnel_sync(struct net_device *netdev, unsigned int table)
 {
 	struct bnx2x *bp = netdev_priv(netdev);
-	u16 t_port = ntohs(ti->port);
-
-	switch (ti->type) {
-	case UDP_TUNNEL_TYPE_VXLAN:
-		__bnx2x_add_udp_port(bp, t_port, BNX2X_UDP_PORT_VXLAN);
-		break;
-	case UDP_TUNNEL_TYPE_GENEVE:
-		__bnx2x_add_udp_port(bp, t_port, BNX2X_UDP_PORT_GENEVE);
-		break;
-	default:
-		break;
-	}
-}
+	struct udp_tunnel_info ti;
 
-static void bnx2x_udp_tunnel_del(struct net_device *netdev,
-				 struct udp_tunnel_info *ti)
-{
-	struct bnx2x *bp = netdev_priv(netdev);
-	u16 t_port = ntohs(ti->port);
+	udp_tunnel_nic_get_port(netdev, table, 0, &ti);
+	bp->udp_tunnel_ports[table] = be16_to_cpu(ti.port);
 
-	switch (ti->type) {
-	case UDP_TUNNEL_TYPE_VXLAN:
-		__bnx2x_del_udp_port(bp, t_port, BNX2X_UDP_PORT_VXLAN);
-		break;
-	case UDP_TUNNEL_TYPE_GENEVE:
-		__bnx2x_del_udp_port(bp, t_port, BNX2X_UDP_PORT_GENEVE);
-		break;
-	default:
-		break;
-	}
+	return bnx2x_udp_port_update(bp);
 }
 
+static const struct udp_tunnel_nic_info bnx2x_udp_tunnels = {
+	.sync_table	= bnx2x_udp_tunnel_sync,
+	.flags		= UDP_TUNNEL_NIC_INFO_OPEN_ONLY,
+	.tables		= {
+		{ .n_entries = 1, .tunnel_types = UDP_TUNNEL_TYPE_VXLAN,  },
+		{ .n_entries = 1, .tunnel_types = UDP_TUNNEL_TYPE_GENEVE, },
+	},
+};
+
 static int bnx2x_close(struct net_device *dev);
 
 /* bnx2x_nic_unload() flushes the bnx2x_wq, thus reset task is
@@ -10270,9 +10275,22 @@ static void bnx2x_sp_rtnl_task(struct work_struct *work)
 		bp->sp_rtnl_state = 0;
 		smp_mb();
 
-		bnx2x_nic_unload(bp, UNLOAD_NORMAL, true);
-		bnx2x_nic_load(bp, LOAD_NORMAL);
+		/* Immediately indicate link as down */
+		bp->link_vars.link_up = 0;
+		bp->force_link_down = true;
+		netif_carrier_off(bp->dev);
+		BNX2X_ERR("Indicating link is down due to Tx-timeout\n");
 
+		bnx2x_nic_unload(bp, UNLOAD_NORMAL, true);
+		/* When ret value shows failure of allocation failure,
+		 * the nic is rebooted again. If open still fails, a error
+		 * message to notify the user.
+		 */
+		if (bnx2x_nic_load(bp, LOAD_NORMAL) == -ENOMEM) {
+			bnx2x_nic_unload(bp, UNLOAD_NORMAL, true);
+			if (bnx2x_nic_load(bp, LOAD_NORMAL))
+				BNX2X_ERR("Open the NIC fails again!\n");
+		}
 		rtnl_unlock();
 		return;
 	}
@@ -10327,23 +10345,8 @@ sp_rtnl_not_reset:
 			       &bp->sp_rtnl_state))
 		bnx2x_update_mng_version(bp);
 
-	if (test_and_clear_bit(BNX2X_SP_RTNL_CHANGE_UDP_PORT,
-			       &bp->sp_rtnl_state)) {
-		if (bnx2x_udp_port_update(bp)) {
-			/* On error, forget configuration */
-			memset(bp->udp_tunnel_ports, 0,
-			       sizeof(struct bnx2x_udp_tunnel) *
-			       BNX2X_UDP_PORT_MAX);
-		} else {
-			/* Since we don't store additional port information,
-			 * if no ports are configured for any feature ask for
-			 * information about currently configured ports.
-			 */
-			if (!bp->udp_tunnel_ports[BNX2X_UDP_PORT_VXLAN].count &&
-			    !bp->udp_tunnel_ports[BNX2X_UDP_PORT_GENEVE].count)
-				udp_tunnel_get_rx_info(bp->dev);
-		}
-	}
+	if (test_and_clear_bit(BNX2X_SP_RTNL_UPDATE_SVID, &bp->sp_rtnl_state))
+		bnx2x_handle_update_svid_cmd(bp);
 
 	/* work which needs rtnl lock not-taken (as it takes the lock itself and
 	 * can be called from other contexts as well)
@@ -11217,7 +11220,7 @@ static void bnx2x_link_settings_supported(struct bnx2x *bp, u32 switch_cfg)
 			   dev_info.port_hw_config[port].external_phy_config),
 			   SHMEM_RD(bp,
 			   dev_info.port_hw_config[port].external_phy_config2));
-			return;
+		return;
 	}
 
 	if (CHIP_IS_E3(bp))
@@ -11718,8 +11721,10 @@ static void bnx2x_get_fcoe_info(struct bnx2x *bp)
 	 * If maximum allowed number of connections is zero -
 	 * disable the feature.
 	 */
-	if (!bp->cnic_eth_dev.max_fcoe_conn)
+	if (!bp->cnic_eth_dev.max_fcoe_conn) {
 		bp->flags |= NO_FCOE_FLAG;
+		eth_zero_addr(bp->fip_mac);
+	}
 }
 
 static void bnx2x_get_cnic_info(struct bnx2x *bp)
@@ -11797,7 +11802,7 @@ static void bnx2x_get_cnic_mac_hwinfo(struct bnx2x *bp)
 		 * as the SAN mac was copied from the primary MAC.
 		 */
 		if (IS_MF_FCOE_AFEX(bp))
-			memcpy(bp->dev->dev_addr, fip_mac, ETH_ALEN);
+			eth_hw_addr_set(bp->dev, fip_mac);
 	} else {
 		val2 = SHMEM_RD(bp, dev_info.port_hw_config[port].
 				iscsi_mac_upper);
@@ -11830,9 +11835,10 @@ static void bnx2x_get_mac_hwinfo(struct bnx2x *bp)
 	u32 val, val2;
 	int func = BP_ABS_FUNC(bp);
 	int port = BP_PORT(bp);
+	u8 addr[ETH_ALEN] = {};
 
 	/* Zero primary MAC configuration */
-	eth_zero_addr(bp->dev->dev_addr);
+	eth_hw_addr_set(bp->dev, addr);
 
 	if (BP_NOMCP(bp)) {
 		BNX2X_ERROR("warning: random MAC workaround active\n");
@@ -11841,8 +11847,10 @@ static void bnx2x_get_mac_hwinfo(struct bnx2x *bp)
 		val2 = MF_CFG_RD(bp, func_mf_config[func].mac_upper);
 		val = MF_CFG_RD(bp, func_mf_config[func].mac_lower);
 		if ((val2 != FUNC_MF_CFG_UPPERMAC_DEFAULT) &&
-		    (val != FUNC_MF_CFG_LOWERMAC_DEFAULT))
-			bnx2x_set_mac_buf(bp->dev->dev_addr, val, val2);
+		    (val != FUNC_MF_CFG_LOWERMAC_DEFAULT)) {
+			bnx2x_set_mac_buf(addr, val, val2);
+			eth_hw_addr_set(bp->dev, addr);
+		}
 
 		if (CNIC_SUPPORT(bp))
 			bnx2x_get_cnic_mac_hwinfo(bp);
@@ -11850,7 +11858,8 @@ static void bnx2x_get_mac_hwinfo(struct bnx2x *bp)
 		/* in SF read MACs from port configuration */
 		val2 = SHMEM_RD(bp, dev_info.port_hw_config[port].mac_upper);
 		val = SHMEM_RD(bp, dev_info.port_hw_config[port].mac_lower);
-		bnx2x_set_mac_buf(bp->dev->dev_addr, val, val2);
+		bnx2x_set_mac_buf(addr, val, val2);
+		eth_hw_addr_set(bp->dev, addr);
 
 		if (CNIC_SUPPORT(bp))
 			bnx2x_get_cnic_mac_hwinfo(bp);
@@ -11915,7 +11924,7 @@ static void validate_set_si_mode(struct bnx2x *bp)
 static int bnx2x_get_hwinfo(struct bnx2x *bp)
 {
 	int /*abs*/func = BP_ABS_FUNC(bp);
-	int vn, mfw_vn;
+	int vn;
 	u32 val = 0, val2 = 0;
 	int rc = 0;
 
@@ -12000,12 +12009,10 @@ static int bnx2x_get_hwinfo(struct bnx2x *bp)
 	/*
 	 * Initialize MF configuration
 	 */
-
 	bp->mf_ov = 0;
 	bp->mf_mode = 0;
 	bp->mf_sub_mode = 0;
 	vn = BP_VN(bp);
-	mfw_vn = BP_FW_MB_IDX(bp);
 
 	if (!CHIP_IS_E1(bp) && !BP_NOMCP(bp)) {
 		BNX2X_DEV_INFO("shmem2base 0x%x, size %d, mfcfg offset %d\n",
@@ -12198,87 +12205,35 @@ static int bnx2x_get_hwinfo(struct bnx2x *bp)
 
 static void bnx2x_read_fwinfo(struct bnx2x *bp)
 {
-	int cnt, i, block_end, rodi;
-	char vpd_start[BNX2X_VPD_LEN+1];
-	char str_id_reg[VENDOR_ID_LEN+1];
-	char str_id_cap[VENDOR_ID_LEN+1];
-	char *vpd_data;
-	char *vpd_extended_data = NULL;
-	u8 len;
+	char str_id[VENDOR_ID_LEN + 1];
+	unsigned int vpd_len, kw_len;
+	u8 *vpd_data;
+	int rodi;
 
-	cnt = pci_read_vpd(bp->pdev, 0, BNX2X_VPD_LEN, vpd_start);
 	memset(bp->fw_ver, 0, sizeof(bp->fw_ver));
 
-	if (cnt < BNX2X_VPD_LEN)
-		goto out_not_found;
-
-	/* VPD RO tag should be first tag after identifier string, hence
-	 * we should be able to find it in first BNX2X_VPD_LEN chars
-	 */
-	i = pci_vpd_find_tag(vpd_start, 0, BNX2X_VPD_LEN,
-			     PCI_VPD_LRDT_RO_DATA);
-	if (i < 0)
-		goto out_not_found;
-
-	block_end = i + PCI_VPD_LRDT_TAG_SIZE +
-		    pci_vpd_lrdt_size(&vpd_start[i]);
-
-	i += PCI_VPD_LRDT_TAG_SIZE;
-
-	if (block_end > BNX2X_VPD_LEN) {
-		vpd_extended_data = kmalloc(block_end, GFP_KERNEL);
-		if (vpd_extended_data  == NULL)
-			goto out_not_found;
-
-		/* read rest of vpd image into vpd_extended_data */
-		memcpy(vpd_extended_data, vpd_start, BNX2X_VPD_LEN);
-		cnt = pci_read_vpd(bp->pdev, BNX2X_VPD_LEN,
-				   block_end - BNX2X_VPD_LEN,
-				   vpd_extended_data + BNX2X_VPD_LEN);
-		if (cnt < (block_end - BNX2X_VPD_LEN))
-			goto out_not_found;
-		vpd_data = vpd_extended_data;
-	} else
-		vpd_data = vpd_start;
-
-	/* now vpd_data holds full vpd content in both cases */
-
-	rodi = pci_vpd_find_info_keyword(vpd_data, i, block_end,
-				   PCI_VPD_RO_KEYWORD_MFR_ID);
-	if (rodi < 0)
-		goto out_not_found;
-
-	len = pci_vpd_info_field_size(&vpd_data[rodi]);
+	vpd_data = pci_vpd_alloc(bp->pdev, &vpd_len);
+	if (IS_ERR(vpd_data))
+		return;
 
-	if (len != VENDOR_ID_LEN)
+	rodi = pci_vpd_find_ro_info_keyword(vpd_data, vpd_len,
+					    PCI_VPD_RO_KEYWORD_MFR_ID, &kw_len);
+	if (rodi < 0 || kw_len != VENDOR_ID_LEN)
 		goto out_not_found;
 
-	rodi += PCI_VPD_INFO_FLD_HDR_SIZE;
-
 	/* vendor specific info */
-	snprintf(str_id_reg, VENDOR_ID_LEN + 1, "%04x", PCI_VENDOR_ID_DELL);
-	snprintf(str_id_cap, VENDOR_ID_LEN + 1, "%04X", PCI_VENDOR_ID_DELL);
-	if (!strncmp(str_id_reg, &vpd_data[rodi], VENDOR_ID_LEN) ||
-	    !strncmp(str_id_cap, &vpd_data[rodi], VENDOR_ID_LEN)) {
-
-		rodi = pci_vpd_find_info_keyword(vpd_data, i, block_end,
-						PCI_VPD_RO_KEYWORD_VENDOR0);
-		if (rodi >= 0) {
-			len = pci_vpd_info_field_size(&vpd_data[rodi]);
-
-			rodi += PCI_VPD_INFO_FLD_HDR_SIZE;
-
-			if (len < 32 && (len + rodi) <= BNX2X_VPD_LEN) {
-				memcpy(bp->fw_ver, &vpd_data[rodi], len);
-				bp->fw_ver[len] = ' ';
-			}
+	snprintf(str_id, VENDOR_ID_LEN + 1, "%04x", PCI_VENDOR_ID_DELL);
+	if (!strncasecmp(str_id, &vpd_data[rodi], VENDOR_ID_LEN)) {
+		rodi = pci_vpd_find_ro_info_keyword(vpd_data, vpd_len,
+						    PCI_VPD_RO_KEYWORD_VENDOR0,
+						    &kw_len);
+		if (rodi >= 0 && kw_len < sizeof(bp->fw_ver)) {
+			memcpy(bp->fw_ver, &vpd_data[rodi], kw_len);
+			bp->fw_ver[kw_len] = ' ';
 		}
-		kfree(vpd_extended_data);
-		return;
 	}
 out_not_found:
-	kfree(vpd_extended_data);
-	return;
+	kfree(vpd_data);
 }
 
 static void bnx2x_set_modes_bitmap(struct bnx2x *bp)
@@ -12352,7 +12307,9 @@ static int bnx2x_init_bp(struct bnx2x *bp)
 		if (rc)
 			return rc;
 	} else {
-		eth_zero_addr(bp->dev->dev_addr);
+		static const u8 zero_addr[ETH_ALEN] = {};
+
+		eth_hw_addr_set(bp->dev, zero_addr);
 	}
 
 	bnx2x_set_modes_bitmap(bp);
@@ -12393,12 +12350,12 @@ static int bnx2x_init_bp(struct bnx2x *bp)
 
 	/* Set TPA flags */
 	if (bp->disable_tpa) {
-		bp->dev->hw_features &= ~NETIF_F_LRO;
-		bp->dev->features &= ~NETIF_F_LRO;
+		bp->dev->hw_features &= ~(NETIF_F_LRO | NETIF_F_GRO_HW);
+		bp->dev->features &= ~(NETIF_F_LRO | NETIF_F_GRO_HW);
 	}
 
 	if (CHIP_IS_E1(bp))
-		bp->dropless_fc = 0;
+		bp->dropless_fc = false;
 	else
 		bp->dropless_fc = dropless_fc | bnx2x_get_dropless_info(bp);
 
@@ -12414,10 +12371,8 @@ static int bnx2x_init_bp(struct bnx2x *bp)
 
 	bp->current_interval = CHIP_REV_IS_SLOW(bp) ? 5*HZ : HZ;
 
-	init_timer(&bp->timer);
+	timer_setup(&bp->timer, bnx2x_timer, 0);
 	bp->timer.expires = jiffies + bp->current_interval;
-	bp->timer.data = (unsigned long) bp;
-	bp->timer.function = bnx2x_timer;
 
 	if (SHMEM2_HAS(bp, dcbx_lldp_params_offset) &&
 	    SHMEM2_HAS(bp, dcbx_lldp_dcbx_stat_offset) &&
@@ -12464,9 +12419,6 @@ static int bnx2x_init_bp(struct bnx2x *bp)
 
 	bp->dump_preset_idx = 1;
 
-	if (CHIP_IS_E3B0(bp))
-		bp->flags |= PTP_SUPPORTED;
-
 	return rc;
 }
 
@@ -12545,9 +12497,6 @@ static int bnx2x_open(struct net_device *dev)
 	if (rc)
 		return rc;
 
-	if (IS_PF(bp))
-		udp_tunnel_get_rx_info(dev);
-
 	return 0;
 }
 
@@ -12729,7 +12678,7 @@ static int bnx2x_set_mc_list(struct bnx2x *bp)
 	} else {
 		/* If no mc addresses are required, flush the configuration */
 		rc = bnx2x_config_mcast(bp, &rparam, BNX2X_MCAST_CMD_DEL);
-		if (rc)
+		if (rc < 0)
 			BNX2X_ERR("Failed to clear multicast configuration %d\n",
 				  rc);
 	}
@@ -12874,19 +12823,6 @@ static int bnx2x_ioctl(struct net_device *dev, struct ifreq *ifr, int cmd)
 	}
 }
 
-#ifdef CONFIG_NET_POLL_CONTROLLER
-static void poll_bnx2x(struct net_device *dev)
-{
-	struct bnx2x *bp = netdev_priv(dev);
-	int i;
-
-	for_each_eth_queue(bp, i) {
-		struct bnx2x_fastpath *fp = &bp->fp[i];
-		napi_schedule(&bnx2x_fp(bp, fp->index, napi));
-	}
-}
-#endif
-
 static int bnx2x_validate_addr(struct net_device *dev)
 {
 	struct bnx2x *bp = netdev_priv(dev);
@@ -12920,6 +12856,24 @@ static netdev_features_t bnx2x_features_check(struct sk_buff *skb,
 					      struct net_device *dev,
 					      netdev_features_t features)
 {
+	/*
+	 * A skb with gso_size + header length > 9700 will cause a
+	 * firmware panic. Drop GSO support.
+	 *
+	 * Eventually the upper layer should not pass these packets down.
+	 *
+	 * For speed, if the gso_size is <= 9000, assume there will
+	 * not be 700 bytes of headers and pass it through. Only do a
+	 * full (slow) validation if the gso_size is > 9000.
+	 *
+	 * (Due to the way SKB_BY_FRAGS works this will also do a full
+	 * validation in that case.)
+	 */
+	if (unlikely(skb_is_gso(skb) &&
+		     (skb_shinfo(skb)->gso_size > 9000) &&
+		     !skb_gso_validate_mac_len(skb, 9700)))
+		features &= ~NETIF_F_GSO_MASK;
+
 	features = vlan_features_check(skb, features);
 	return vxlan_features_check(skb, features);
 }
@@ -12989,13 +12943,6 @@ static void bnx2x_vlan_configure(struct bnx2x *bp, bool set_rx_mode)
 
 int bnx2x_vlan_reconfigure_vid(struct bnx2x *bp)
 {
-	struct bnx2x_vlan_entry *vlan;
-
-	/* The hw forgot all entries after reload */
-	list_for_each_entry(vlan, &bp->vlan_reg, link)
-		vlan->hw = false;
-	bp->vlan_cnt = 0;
-
 	/* Don't set rx mode here. Our caller will do it. */
 	bnx2x_vlan_configure(bp, false);
 
@@ -13068,21 +13015,19 @@ static const struct net_device_ops bnx2x_netdev_ops = {
 	.ndo_set_rx_mode	= bnx2x_set_rx_mode,
 	.ndo_set_mac_address	= bnx2x_change_mac_addr,
 	.ndo_validate_addr	= bnx2x_validate_addr,
-	.ndo_do_ioctl		= bnx2x_ioctl,
+	.ndo_eth_ioctl		= bnx2x_ioctl,
 	.ndo_change_mtu		= bnx2x_change_mtu,
 	.ndo_fix_features	= bnx2x_fix_features,
 	.ndo_set_features	= bnx2x_set_features,
 	.ndo_tx_timeout		= bnx2x_tx_timeout,
 	.ndo_vlan_rx_add_vid	= bnx2x_vlan_rx_add_vid,
 	.ndo_vlan_rx_kill_vid	= bnx2x_vlan_rx_kill_vid,
-#ifdef CONFIG_NET_POLL_CONTROLLER
-	.ndo_poll_controller	= poll_bnx2x,
-#endif
 	.ndo_setup_tc		= __bnx2x_setup_tc,
 #ifdef CONFIG_BNX2X_SRIOV
 	.ndo_set_vf_mac		= bnx2x_set_vf_mac,
 	.ndo_set_vf_vlan	= bnx2x_set_vf_vlan,
 	.ndo_get_vf_config	= bnx2x_get_vf_config,
+	.ndo_set_vf_spoofchk	= bnx2x_set_vf_spoofchk,
 #endif
 #ifdef NETDEV_FCOE_WWNN
 	.ndo_fcoe_get_wwn	= bnx2x_fcoe_get_wwn,
@@ -13091,31 +13036,8 @@ static const struct net_device_ops bnx2x_netdev_ops = {
 	.ndo_get_phys_port_id	= bnx2x_get_phys_port_id,
 	.ndo_set_vf_link_state	= bnx2x_set_vf_link_state,
 	.ndo_features_check	= bnx2x_features_check,
-	.ndo_udp_tunnel_add	= bnx2x_udp_tunnel_add,
-	.ndo_udp_tunnel_del	= bnx2x_udp_tunnel_del,
 };
 
-static int bnx2x_set_coherency_mask(struct bnx2x *bp)
-{
-	struct device *dev = &bp->pdev->dev;
-
-	if (dma_set_mask_and_coherent(dev, DMA_BIT_MASK(64)) != 0 &&
-	    dma_set_mask_and_coherent(dev, DMA_BIT_MASK(32)) != 0) {
-		dev_err(dev, "System does not support DMA, aborting\n");
-		return -EIO;
-	}
-
-	return 0;
-}
-
-static void bnx2x_disable_pcie_error_reporting(struct bnx2x *bp)
-{
-	if (bp->flags & AER_ENABLED) {
-		pci_disable_pcie_error_reporting(bp->pdev);
-		bp->flags &= ~AER_ENABLED;
-	}
-}
-
 static int bnx2x_init_dev(struct bnx2x *bp, struct pci_dev *pdev,
 			  struct net_device *dev, unsigned long board_type)
 {
@@ -13185,9 +13107,11 @@ static int bnx2x_init_dev(struct bnx2x *bp, struct pci_dev *pdev,
 		goto err_out_release;
 	}
 
-	rc = bnx2x_set_coherency_mask(bp);
-	if (rc)
+	rc = dma_set_mask_and_coherent(&bp->pdev->dev, DMA_BIT_MASK(64));
+	if (rc) {
+		dev_err(&bp->pdev->dev, "System does not support DMA, aborting\n");
 		goto err_out_release;
+	}
 
 	dev->mem_start = pci_resource_start(pdev, 0);
 	dev->base_addr = dev->mem_start;
@@ -13226,13 +13150,6 @@ static int bnx2x_init_dev(struct bnx2x *bp, struct pci_dev *pdev,
 	/* Set PCIe reset type to fundamental for EEH recovery */
 	pdev->needs_freset = 1;
 
-	/* AER (Advanced Error reporting) configuration */
-	rc = pci_enable_pcie_error_reporting(pdev);
-	if (!rc)
-		bp->flags |= AER_ENABLED;
-	else
-		BNX2X_DEV_INFO("Failed To configure PCIe AER [%d]\n", rc);
-
 	/*
 	 * Clean the following indirect addresses for all functions since it
 	 * is not used by the driver.
@@ -13268,7 +13185,7 @@ static int bnx2x_init_dev(struct bnx2x *bp, struct pci_dev *pdev,
 
 	dev->hw_features = NETIF_F_SG | NETIF_F_IP_CSUM | NETIF_F_IPV6_CSUM |
 		NETIF_F_TSO | NETIF_F_TSO_ECN | NETIF_F_TSO6 |
-		NETIF_F_RXCSUM | NETIF_F_LRO | NETIF_F_GRO |
+		NETIF_F_RXCSUM | NETIF_F_LRO | NETIF_F_GRO | NETIF_F_GRO_HW |
 		NETIF_F_RXHASH | NETIF_F_HW_VLAN_CTAG_TX;
 	if (!chip_is_e1x) {
 		dev->hw_features |= NETIF_F_GSO_GRE | NETIF_F_GSO_GRE_CSUM |
@@ -13287,6 +13204,9 @@ static int bnx2x_init_dev(struct bnx2x *bp, struct pci_dev *pdev,
 
 		dev->gso_partial_features = NETIF_F_GSO_GRE_CSUM |
 					    NETIF_F_GSO_UDP_TUNNEL_CSUM;
+
+		if (IS_PF(bp))
+			dev->udp_tunnel_nic_info = &bnx2x_udp_tunnels;
 	}
 
 	dev->vlan_features = NETIF_F_SG | NETIF_F_IP_CSUM | NETIF_F_IPV6_CSUM |
@@ -13304,6 +13224,8 @@ static int bnx2x_init_dev(struct bnx2x *bp, struct pci_dev *pdev,
 
 	dev->features |= dev->hw_features | NETIF_F_HW_VLAN_CTAG_RX;
 	dev->features |= NETIF_F_HIGHDMA;
+	if (dev->features & NETIF_F_LRO)
+		dev->features &= ~NETIF_F_GRO_HW;
 
 	/* Add Loopback capability to the device */
 	dev->hw_features |= NETIF_F_LOOPBACK;
@@ -13381,16 +13303,11 @@ static int bnx2x_check_firmware(struct bnx2x *bp)
 	/* Check FW version */
 	offset = be32_to_cpu(fw_hdr->fw_version.offset);
 	fw_ver = firmware->data + offset;
-	if ((fw_ver[0] != BCM_5710_FW_MAJOR_VERSION) ||
-	    (fw_ver[1] != BCM_5710_FW_MINOR_VERSION) ||
-	    (fw_ver[2] != BCM_5710_FW_REVISION_VERSION) ||
-	    (fw_ver[3] != BCM_5710_FW_ENGINEERING_VERSION)) {
+	if (fw_ver[0] != bp->fw_major || fw_ver[1] != bp->fw_minor ||
+	    fw_ver[2] != bp->fw_rev || fw_ver[3] != bp->fw_eng) {
 		BNX2X_ERR("Bad FW version:%d.%d.%d.%d. Should be %d.%d.%d.%d\n",
-		       fw_ver[0], fw_ver[1], fw_ver[2], fw_ver[3],
-		       BCM_5710_FW_MAJOR_VERSION,
-		       BCM_5710_FW_MINOR_VERSION,
-		       BCM_5710_FW_REVISION_VERSION,
-		       BCM_5710_FW_ENGINEERING_VERSION);
+			  fw_ver[0], fw_ver[1], fw_ver[2], fw_ver[3],
+			  bp->fw_major, bp->fw_minor, bp->fw_rev, bp->fw_eng);
 		return -EINVAL;
 	}
 
@@ -13470,32 +13387,49 @@ do {									\
 
 static int bnx2x_init_firmware(struct bnx2x *bp)
 {
-	const char *fw_file_name;
+	const char *fw_file_name, *fw_file_name_v15;
 	struct bnx2x_fw_file_hdr *fw_hdr;
 	int rc;
 
 	if (bp->firmware)
 		return 0;
 
-	if (CHIP_IS_E1(bp))
+	if (CHIP_IS_E1(bp)) {
 		fw_file_name = FW_FILE_NAME_E1;
-	else if (CHIP_IS_E1H(bp))
+		fw_file_name_v15 = FW_FILE_NAME_E1_V15;
+	} else if (CHIP_IS_E1H(bp)) {
 		fw_file_name = FW_FILE_NAME_E1H;
-	else if (!CHIP_IS_E1x(bp))
+		fw_file_name_v15 = FW_FILE_NAME_E1H_V15;
+	} else if (!CHIP_IS_E1x(bp)) {
 		fw_file_name = FW_FILE_NAME_E2;
-	else {
+		fw_file_name_v15 = FW_FILE_NAME_E2_V15;
+	} else {
 		BNX2X_ERR("Unsupported chip revision\n");
 		return -EINVAL;
 	}
+
 	BNX2X_DEV_INFO("Loading %s\n", fw_file_name);
 
 	rc = request_firmware(&bp->firmware, fw_file_name, &bp->pdev->dev);
 	if (rc) {
-		BNX2X_ERR("Can't load firmware file %s\n",
-			  fw_file_name);
-		goto request_firmware_exit;
+		BNX2X_DEV_INFO("Trying to load older fw %s\n", fw_file_name_v15);
+
+		/* try to load prev version */
+		rc = request_firmware(&bp->firmware, fw_file_name_v15, &bp->pdev->dev);
+
+		if (rc)
+			goto request_firmware_exit;
+
+		bp->fw_rev = BCM_5710_FW_REVISION_VERSION_V15;
+	} else {
+		bp->fw_cap |= FW_CAP_INVALIDATE_VF_FP_HSI;
+		bp->fw_rev = BCM_5710_FW_REVISION_VERSION;
 	}
 
+	bp->fw_major = BCM_5710_FW_MAJOR_VERSION;
+	bp->fw_minor = BCM_5710_FW_MINOR_VERSION;
+	bp->fw_eng = BCM_5710_FW_ENGINEERING_VERSION;
+
 	rc = bnx2x_check_firmware(bp);
 	if (rc) {
 		BNX2X_ERR("Corrupt firmware file %s\n", fw_file_name);
@@ -13605,9 +13539,9 @@ static int bnx2x_set_qm_cid_count(struct bnx2x *bp)
 }
 
 /**
- * bnx2x_get_num_none_def_sbs - return the number of none default SBs
- *
- * @dev:	pci device
+ * bnx2x_get_num_non_def_sbs - return the number of none default SBs
+ * @pdev: pci device
+ * @cnic_cnt: count
  *
  */
 static int bnx2x_get_num_non_def_sbs(struct pci_dev *pdev, int cnic_cnt)
@@ -13723,19 +13657,20 @@ static int bnx2x_send_update_drift_ramrod(struct bnx2x *bp, int drift_dir,
 	return bnx2x_func_state_change(bp, &func_params);
 }
 
-static int bnx2x_ptp_adjfreq(struct ptp_clock_info *ptp, s32 ppb)
+static int bnx2x_ptp_adjfine(struct ptp_clock_info *ptp, long scaled_ppm)
 {
 	struct bnx2x *bp = container_of(ptp, struct bnx2x, ptp_clock_info);
 	int rc;
 	int drift_dir = 1;
 	int val, period, period1, period2, dif, dif1, dif2;
 	int best_dif = BNX2X_MAX_PHC_DRIFT, best_period = 0, best_val = 0;
+	s32 ppb = scaled_ppm_to_ppb(scaled_ppm);
 
-	DP(BNX2X_MSG_PTP, "PTP adjfreq called, ppb = %d\n", ppb);
+	DP(BNX2X_MSG_PTP, "PTP adjfine called, ppb = %d\n", ppb);
 
 	if (!netif_running(bp->dev)) {
 		DP(BNX2X_MSG_PTP,
-		   "PTP adjfreq called while the interface is down\n");
+		   "PTP adjfine called while the interface is down\n");
 		return -ENETDOWN;
 	}
 
@@ -13860,7 +13795,7 @@ static int bnx2x_ptp_enable(struct ptp_clock_info *ptp,
 	return -ENOTSUPP;
 }
 
-static void bnx2x_register_phc(struct bnx2x *bp)
+void bnx2x_register_phc(struct bnx2x *bp)
 {
 	/* Fill the ptp_clock_info struct and register PTP clock*/
 	bp->ptp_clock_info.owner = THIS_MODULE;
@@ -13870,7 +13805,7 @@ static void bnx2x_register_phc(struct bnx2x *bp)
 	bp->ptp_clock_info.n_ext_ts = 0;
 	bp->ptp_clock_info.n_per_out = 0;
 	bp->ptp_clock_info.pps = 0;
-	bp->ptp_clock_info.adjfreq = bnx2x_ptp_adjfreq;
+	bp->ptp_clock_info.adjfine = bnx2x_ptp_adjfine;
 	bp->ptp_clock_info.adjtime = bnx2x_ptp_adjtime;
 	bp->ptp_clock_info.gettime64 = bnx2x_ptp_gettime;
 	bp->ptp_clock_info.settime64 = bnx2x_ptp_settime;
@@ -13879,7 +13814,7 @@ static void bnx2x_register_phc(struct bnx2x *bp)
 	bp->ptp_clock = ptp_clock_register(&bp->ptp_clock_info, &bp->pdev->dev);
 	if (IS_ERR(bp->ptp_clock)) {
 		bp->ptp_clock = NULL;
-		BNX2X_ERR("PTP clock registeration failed\n");
+		BNX2X_ERR("PTP clock registration failed\n");
 	}
 }
 
@@ -13888,8 +13823,6 @@ static int bnx2x_init_one(struct pci_dev *pdev,
 {
 	struct net_device *dev = NULL;
 	struct bnx2x *bp;
-	enum pcie_link_width pcie_width;
-	enum pci_bus_speed pcie_speed;
 	int rc, max_non_def_sbs;
 	int rx_count, tx_count, rss_count, doorbell_size;
 	int max_cos_est;
@@ -13993,7 +13926,7 @@ static int bnx2x_init_one(struct pci_dev *pdev,
 			rc = -ENOMEM;
 			goto init_one_freemem;
 		}
-		bp->doorbells = ioremap_nocache(pci_resource_start(pdev, 2),
+		bp->doorbells = ioremap(pci_resource_start(pdev, 2),
 						doorbell_size);
 	}
 	if (!bp->doorbells) {
@@ -14057,23 +13990,12 @@ static int bnx2x_init_one(struct pci_dev *pdev,
 		dev_addr_add(bp->dev, bp->fip_mac, NETDEV_HW_ADDR_T_SAN);
 		rtnl_unlock();
 	}
-	if (pcie_get_minimum_link(bp->pdev, &pcie_speed, &pcie_width) ||
-	    pcie_speed == PCI_SPEED_UNKNOWN ||
-	    pcie_width == PCIE_LNK_WIDTH_UNKNOWN)
-		BNX2X_DEV_INFO("Failed to determine PCI Express Bandwidth\n");
-	else
-		BNX2X_DEV_INFO(
-		       "%s (%c%d) PCI-E x%d %s found at mem %lx, IRQ %d, node addr %pM\n",
-		       board_info[ent->driver_data].name,
-		       (CHIP_REV(bp) >> 12) + 'A', (CHIP_METAL(bp) >> 4),
-		       pcie_width,
-		       pcie_speed == PCIE_SPEED_2_5GT ? "2.5GHz" :
-		       pcie_speed == PCIE_SPEED_5_0GT ? "5.0GHz" :
-		       pcie_speed == PCIE_SPEED_8_0GT ? "8.0GHz" :
-		       "Unknown",
-		       dev->base_addr, bp->pdev->irq, dev->dev_addr);
-
-	bnx2x_register_phc(bp);
+	BNX2X_DEV_INFO(
+	       "%s (%c%d) PCI-E found at mem %lx, IRQ %d, node addr %pM\n",
+	       board_info[ent->driver_data].name,
+	       (CHIP_REV(bp) >> 12) + 'A', (CHIP_METAL(bp) >> 4),
+	       dev->base_addr, bp->pdev->irq, dev->dev_addr);
+	pcie_print_link_status(bp->pdev);
 
 	if (!IS_MF_SD_STORAGE_PERSONALITY_ONLY(bp))
 		bnx2x_set_os_driver_state(bp, OS_DRIVER_STATE_DISABLED);
@@ -14084,8 +14006,6 @@ init_one_freemem:
 	bnx2x_free_mem_bp(bp);
 
 init_one_exit:
-	bnx2x_disable_pcie_error_reporting(bp);
-
 	if (bp->regview)
 		iounmap(bp->regview);
 
@@ -14107,11 +14027,6 @@ static void __bnx2x_remove(struct pci_dev *pdev,
 			   struct bnx2x *bp,
 			   bool remove_netdev)
 {
-	if (bp->ptp_clock) {
-		ptp_clock_unregister(bp->ptp_clock);
-		bp->ptp_clock = NULL;
-	}
-
 	/* Delete storage MAC address */
 	if (!NO_FCOE(bp)) {
 		rtnl_lock();
@@ -14171,7 +14086,6 @@ static void __bnx2x_remove(struct pci_dev *pdev,
 		pci_set_power_state(pdev, PCI_D3hot);
 	}
 
-	bnx2x_disable_pcie_error_reporting(bp);
 	if (remove_netdev) {
 		if (bp->regview)
 			iounmap(bp->regview);
@@ -14223,13 +14137,9 @@ static int bnx2x_eeh_nic_unload(struct bnx2x *bp)
 
 	/* Stop Tx */
 	bnx2x_tx_disable(bp);
-	/* Delete all NAPI objects */
-	bnx2x_del_all_napi(bp);
-	if (CNIC_LOADED(bp))
-		bnx2x_del_all_napi_cnic(bp);
 	netdev_reset_tc(bp->dev);
 
-	del_timer_sync(&bp->timer);
+	timer_delete_sync(&bp->timer);
 	cancel_delayed_work_sync(&bp->sp_task);
 	cancel_delayed_work_sync(&bp->period_task);
 
@@ -14315,7 +14225,10 @@ static pci_ers_result_t bnx2x_io_slot_reset(struct pci_dev *pdev)
 		BNX2X_ERR("IO slot reset --> driver unload\n");
 
 		/* MCP should have been reset; Need to wait for validity */
-		bnx2x_init_shmem(bp);
+		if (bnx2x_init_shmem(bp)) {
+			rtnl_unlock();
+			return PCI_ERS_RESULT_DISCONNECT;
+		}
 
 		if (IS_PF(bp) && SHMEM2_HAS(bp, drv_capabilities_flag)) {
 			u32 v;
@@ -14327,8 +14240,16 @@ static pci_ers_result_t bnx2x_io_slot_reset(struct pci_dev *pdev)
 		}
 		bnx2x_drain_tx_queues(bp);
 		bnx2x_send_unload_req(bp, UNLOAD_RECOVERY);
-		bnx2x_netif_stop(bp, 1);
-		bnx2x_free_irq(bp);
+		if (!bp->nic_stopped) {
+			bnx2x_netif_stop(bp, 1);
+			bnx2x_del_all_napi(bp);
+
+			if (CNIC_LOADED(bp))
+				bnx2x_del_all_napi_cnic(bp);
+
+			bnx2x_free_irq(bp);
+			bp->nic_stopped = true;
+		}
 
 		/* Report UNLOAD_DONE to MCP */
 		bnx2x_send_unload_done(bp, true);
@@ -14353,14 +14274,6 @@ static pci_ers_result_t bnx2x_io_slot_reset(struct pci_dev *pdev)
 
 	rtnl_unlock();
 
-	/* If AER, perform cleanup of the PCIe registers */
-	if (bp->flags & AER_ENABLED) {
-		if (pci_cleanup_aer_uncorrect_error_status(pdev))
-			BNX2X_ERR("pci_cleanup_aer_uncorrect_error_status failed\n");
-		else
-			DP(NETIF_MSG_HW, "pci_cleanup_aer_uncorrect_error_status succeeded\n");
-	}
-
 	return PCI_ERS_RESULT_RECOVERED;
 }
 
@@ -14386,11 +14299,16 @@ static void bnx2x_io_resume(struct pci_dev *pdev)
 	bp->fw_seq = SHMEM_RD(bp, func_mb[BP_FW_MB_IDX(bp)].drv_mb_header) &
 							DRV_MSG_SEQ_NUMBER_MASK;
 
-	if (netif_running(dev))
-		bnx2x_nic_load(bp, LOAD_NORMAL);
+	if (netif_running(dev)) {
+		if (bnx2x_nic_load(bp, LOAD_NORMAL)) {
+			netdev_err(bp->dev, "Error during driver initialization, try unloading/reloading the driver\n");
+			goto done;
+		}
+	}
 
 	netif_device_attach(dev);
 
+done:
 	rtnl_unlock();
 }
 
@@ -14428,8 +14346,7 @@ static struct pci_driver bnx2x_pci_driver = {
 	.id_table    = bnx2x_pci_tbl,
 	.probe       = bnx2x_init_one,
 	.remove      = bnx2x_remove_one,
-	.suspend     = bnx2x_suspend,
-	.resume      = bnx2x_resume,
+	.driver.pm   = &bnx2x_pm_ops,
 	.err_handler = &bnx2x_err_handler,
 #ifdef CONFIG_BNX2X_SRIOV
 	.sriov_configure = bnx2x_sriov_configure,
@@ -14441,8 +14358,6 @@ static int __init bnx2x_init(void)
 {
 	int ret;
 
-	pr_info("%s", version);
-
 	bnx2x_wq = create_singlethread_workqueue("bnx2x");
 	if (bnx2x_wq == NULL) {
 		pr_err("Cannot create workqueue\n");
@@ -14492,9 +14407,7 @@ module_exit(bnx2x_cleanup);
 
 /**
  * bnx2x_set_iscsi_eth_mac_addr - set iSCSI MAC(s).
- *
  * @bp:		driver handle
- * @set:	set or clear the CAM entry
  *
  * This function will wait until the ramrod completion returns.
  * Return 0 if success, -ENODEV if ramrod doesn't return.
@@ -14777,7 +14690,6 @@ static int bnx2x_drv_ctl(struct net_device *dev, struct drv_ctl_info *ctl)
 		if (rc)
 			break;
 
-		mmiowb();
 		barrier();
 
 		/* Start accepting on iSCSI L2 ring */
@@ -14812,7 +14724,6 @@ static int bnx2x_drv_ctl(struct net_device *dev, struct drv_ctl_info *ctl)
 		if (!bnx2x_wait_sp_comp(bp, sp_bits))
 			BNX2X_ERR("rx_mode completion timed out!\n");
 
-		mmiowb();
 		barrier();
 
 		/* Unset iSCSI L2 MAC */
@@ -15000,9 +14911,11 @@ void bnx2x_setup_cnic_irq_info(struct bnx2x *bp)
 	else
 		cp->irq_arr[0].status_blk = (void *)bp->cnic_sb.e1x_sb;
 
+	cp->irq_arr[0].status_blk_map = bp->cnic_sb_mapping;
 	cp->irq_arr[0].status_blk_num =  bnx2x_cnic_fw_sb_id(bp);
 	cp->irq_arr[0].status_blk_num2 = bnx2x_cnic_igu_sb_id(bp);
 	cp->irq_arr[1].status_blk = bp->def_status_blk;
+	cp->irq_arr[1].status_blk_map = bp->def_status_blk_mapping;
 	cp->irq_arr[1].status_blk_num = DEF_SB_ID;
 	cp->irq_arr[1].status_blk_num2 = DEF_SB_IGU_ID;
 
@@ -15192,11 +15105,24 @@ static void bnx2x_ptp_task(struct work_struct *work)
 	u32 val_seq;
 	u64 timestamp, ns;
 	struct skb_shared_hwtstamps shhwtstamps;
+	bool bail = true;
+	int i;
+
+	/* FW may take a while to complete timestamping; try a bit and if it's
+	 * still not complete, may indicate an error state - bail out then.
+	 */
+	for (i = 0; i < 10; i++) {
+		/* Read Tx timestamp registers */
+		val_seq = REG_RD(bp, port ? NIG_REG_P1_TLLH_PTP_BUF_SEQID :
+				 NIG_REG_P0_TLLH_PTP_BUF_SEQID);
+		if (val_seq & 0x10000) {
+			bail = false;
+			break;
+		}
+		msleep(1 << i);
+	}
 
-	/* Read Tx timestamp registers */
-	val_seq = REG_RD(bp, port ? NIG_REG_P1_TLLH_PTP_BUF_SEQID :
-			 NIG_REG_P0_TLLH_PTP_BUF_SEQID);
-	if (val_seq & 0x10000) {
+	if (!bail) {
 		/* There is a valid timestamp value */
 		timestamp = REG_RD(bp, port ? NIG_REG_P1_TLLH_PTP_BUF_TS_MSB :
 				   NIG_REG_P0_TLLH_PTP_BUF_TS_MSB);
@@ -15211,16 +15137,18 @@ static void bnx2x_ptp_task(struct work_struct *work)
 		memset(&shhwtstamps, 0, sizeof(shhwtstamps));
 		shhwtstamps.hwtstamp = ns_to_ktime(ns);
 		skb_tstamp_tx(bp->ptp_tx_skb, &shhwtstamps);
-		dev_kfree_skb_any(bp->ptp_tx_skb);
-		bp->ptp_tx_skb = NULL;
 
 		DP(BNX2X_MSG_PTP, "Tx timestamp, timestamp cycles = %llu, ns = %llu\n",
 		   timestamp, ns);
 	} else {
-		DP(BNX2X_MSG_PTP, "There is no valid Tx timestamp yet\n");
-		/* Reschedule to keep checking for a valid timestamp value */
-		schedule_work(&bp->ptp_task);
+		DP(BNX2X_MSG_PTP,
+		   "Tx timestamp is not recorded (register read=%u)\n",
+		   val_seq);
+		bp->eth_stats.ptp_skip_tx_ts++;
 	}
+
+	dev_kfree_skb_any(bp->ptp_tx_skb);
+	bp->ptp_tx_skb = NULL;
 }
 
 void bnx2x_set_rx_ts(struct bnx2x *bp, struct sk_buff *skb)
@@ -15247,7 +15175,7 @@ void bnx2x_set_rx_ts(struct bnx2x *bp, struct sk_buff *skb)
 }
 
 /* Read the PHC */
-static u64 bnx2x_cyclecounter_read(const struct cyclecounter *cc)
+static u64 bnx2x_cyclecounter_read(struct cyclecounter *cc)
 {
 	struct bnx2x *bp = container_of(cc, struct bnx2x, cyclecounter);
 	int port = BP_PORT(bp);
@@ -15325,27 +15253,48 @@ static int bnx2x_enable_ptp_packets(struct bnx2x *bp)
 	return 0;
 }
 
+#define BNX2X_P2P_DETECT_PARAM_MASK 0x5F5
+#define BNX2X_P2P_DETECT_RULE_MASK 0x3DBB
+#define BNX2X_PTP_TX_ON_PARAM_MASK (BNX2X_P2P_DETECT_PARAM_MASK & 0x6AA)
+#define BNX2X_PTP_TX_ON_RULE_MASK (BNX2X_P2P_DETECT_RULE_MASK & 0x3EEE)
+#define BNX2X_PTP_V1_L4_PARAM_MASK (BNX2X_P2P_DETECT_PARAM_MASK & 0x7EE)
+#define BNX2X_PTP_V1_L4_RULE_MASK (BNX2X_P2P_DETECT_RULE_MASK & 0x3FFE)
+#define BNX2X_PTP_V2_L4_PARAM_MASK (BNX2X_P2P_DETECT_PARAM_MASK & 0x7EA)
+#define BNX2X_PTP_V2_L4_RULE_MASK (BNX2X_P2P_DETECT_RULE_MASK & 0x3FEE)
+#define BNX2X_PTP_V2_L2_PARAM_MASK (BNX2X_P2P_DETECT_PARAM_MASK & 0x6BF)
+#define BNX2X_PTP_V2_L2_RULE_MASK (BNX2X_P2P_DETECT_RULE_MASK & 0x3EFF)
+#define BNX2X_PTP_V2_PARAM_MASK (BNX2X_P2P_DETECT_PARAM_MASK & 0x6AA)
+#define BNX2X_PTP_V2_RULE_MASK (BNX2X_P2P_DETECT_RULE_MASK & 0x3EEE)
+
 int bnx2x_configure_ptp_filters(struct bnx2x *bp)
 {
 	int port = BP_PORT(bp);
+	u32 param, rule;
 	int rc;
 
 	if (!bp->hwtstamp_ioctl_called)
 		return 0;
 
+	param = port ? NIG_REG_P1_TLLH_PTP_PARAM_MASK :
+		NIG_REG_P0_TLLH_PTP_PARAM_MASK;
+	rule = port ? NIG_REG_P1_TLLH_PTP_RULE_MASK :
+		NIG_REG_P0_TLLH_PTP_RULE_MASK;
 	switch (bp->tx_type) {
 	case HWTSTAMP_TX_ON:
 		bp->flags |= TX_TIMESTAMPING_EN;
-		REG_WR(bp, port ? NIG_REG_P1_TLLH_PTP_PARAM_MASK :
-		       NIG_REG_P0_TLLH_PTP_PARAM_MASK, 0x6AA);
-		REG_WR(bp, port ? NIG_REG_P1_TLLH_PTP_RULE_MASK :
-		       NIG_REG_P0_TLLH_PTP_RULE_MASK, 0x3EEE);
+		REG_WR(bp, param, BNX2X_PTP_TX_ON_PARAM_MASK);
+		REG_WR(bp, rule, BNX2X_PTP_TX_ON_RULE_MASK);
 		break;
 	case HWTSTAMP_TX_ONESTEP_SYNC:
+	case HWTSTAMP_TX_ONESTEP_P2P:
 		BNX2X_ERR("One-step timestamping is not supported\n");
 		return -ERANGE;
 	}
 
+	param = port ? NIG_REG_P1_LLH_PTP_PARAM_MASK :
+		NIG_REG_P0_LLH_PTP_PARAM_MASK;
+	rule = port ? NIG_REG_P1_LLH_PTP_RULE_MASK :
+		NIG_REG_P0_LLH_PTP_RULE_MASK;
 	switch (bp->rx_filter) {
 	case HWTSTAMP_FILTER_NONE:
 		break;
@@ -15359,30 +15308,24 @@ int bnx2x_configure_ptp_filters(struct bnx2x *bp)
 	case HWTSTAMP_FILTER_PTP_V1_L4_DELAY_REQ:
 		bp->rx_filter = HWTSTAMP_FILTER_PTP_V1_L4_EVENT;
 		/* Initialize PTP detection for UDP/IPv4 events */
-		REG_WR(bp, port ? NIG_REG_P1_LLH_PTP_PARAM_MASK :
-		       NIG_REG_P0_LLH_PTP_PARAM_MASK, 0x7EE);
-		REG_WR(bp, port ? NIG_REG_P1_LLH_PTP_RULE_MASK :
-		       NIG_REG_P0_LLH_PTP_RULE_MASK, 0x3FFE);
+		REG_WR(bp, param, BNX2X_PTP_V1_L4_PARAM_MASK);
+		REG_WR(bp, rule, BNX2X_PTP_V1_L4_RULE_MASK);
 		break;
 	case HWTSTAMP_FILTER_PTP_V2_L4_EVENT:
 	case HWTSTAMP_FILTER_PTP_V2_L4_SYNC:
 	case HWTSTAMP_FILTER_PTP_V2_L4_DELAY_REQ:
 		bp->rx_filter = HWTSTAMP_FILTER_PTP_V2_L4_EVENT;
 		/* Initialize PTP detection for UDP/IPv4 or UDP/IPv6 events */
-		REG_WR(bp, port ? NIG_REG_P1_LLH_PTP_PARAM_MASK :
-		       NIG_REG_P0_LLH_PTP_PARAM_MASK, 0x7EA);
-		REG_WR(bp, port ? NIG_REG_P1_LLH_PTP_RULE_MASK :
-		       NIG_REG_P0_LLH_PTP_RULE_MASK, 0x3FEE);
+		REG_WR(bp, param, BNX2X_PTP_V2_L4_PARAM_MASK);
+		REG_WR(bp, rule, BNX2X_PTP_V2_L4_RULE_MASK);
 		break;
 	case HWTSTAMP_FILTER_PTP_V2_L2_EVENT:
 	case HWTSTAMP_FILTER_PTP_V2_L2_SYNC:
 	case HWTSTAMP_FILTER_PTP_V2_L2_DELAY_REQ:
 		bp->rx_filter = HWTSTAMP_FILTER_PTP_V2_L2_EVENT;
 		/* Initialize PTP detection L2 events */
-		REG_WR(bp, port ? NIG_REG_P1_LLH_PTP_PARAM_MASK :
-		       NIG_REG_P0_LLH_PTP_PARAM_MASK, 0x6BF);
-		REG_WR(bp, port ? NIG_REG_P1_LLH_PTP_RULE_MASK :
-		       NIG_REG_P0_LLH_PTP_RULE_MASK, 0x3EFF);
+		REG_WR(bp, param, BNX2X_PTP_V2_L2_PARAM_MASK);
+		REG_WR(bp, rule, BNX2X_PTP_V2_L2_RULE_MASK);
 
 		break;
 	case HWTSTAMP_FILTER_PTP_V2_EVENT:
@@ -15390,10 +15333,8 @@ int bnx2x_configure_ptp_filters(struct bnx2x *bp)
 	case HWTSTAMP_FILTER_PTP_V2_DELAY_REQ:
 		bp->rx_filter = HWTSTAMP_FILTER_PTP_V2_EVENT;
 		/* Initialize PTP detection L2, UDP/IPv4 or UDP/IPv6 events */
-		REG_WR(bp, port ? NIG_REG_P1_LLH_PTP_PARAM_MASK :
-		       NIG_REG_P0_LLH_PTP_PARAM_MASK, 0x6AA);
-		REG_WR(bp, port ? NIG_REG_P1_LLH_PTP_RULE_MASK :
-		       NIG_REG_P0_LLH_PTP_RULE_MASK, 0x3EEE);
+		REG_WR(bp, param, BNX2X_PTP_V2_PARAM_MASK);
+		REG_WR(bp, rule, BNX2X_PTP_V2_RULE_MASK);
 		break;
 	}
 
@@ -15422,12 +15363,7 @@ static int bnx2x_hwtstamp_ioctl(struct bnx2x *bp, struct ifreq *ifr)
 	DP(BNX2X_MSG_PTP, "Requested tx_type: %d, requested rx_filters = %d\n",
 	   config.tx_type, config.rx_filter);
 
-	if (config.flags) {
-		BNX2X_ERR("config.flags is reserved for future use\n");
-		return -EINVAL;
-	}
-
-	bp->hwtstamp_ioctl_called = 1;
+	bp->hwtstamp_ioctl_called = true;
 	bp->tx_type = config.tx_type;
 	bp->rx_filter = config.rx_filter;
 
@@ -15509,7 +15445,7 @@ void bnx2x_init_ptp(struct bnx2x *bp)
 		bnx2x_init_cyclecounter(bp);
 		timecounter_init(&bp->timecounter, &bp->cyclecounter,
 				 ktime_to_ns(ktime_get_real()));
-		bp->timecounter_init_done = 1;
+		bp->timecounter_init_done = true;
 	}
 
 	DP(BNX2X_MSG_PTP, "PTP initialization ended successfully\n");
diff --git a/drivers/net/ethernet/broadcom/bnx2x/bnx2x_reg.h b/drivers/net/ethernet/broadcom/bnx2x/bnx2x_reg.h
index a43dea259b12..a018f251d198 100644
--- a/drivers/net/ethernet/broadcom/bnx2x/bnx2x_reg.h
+++ b/drivers/net/ethernet/broadcom/bnx2x/bnx2x_reg.h
@@ -284,12 +284,12 @@
 #define CCM_REG_GR_ARB_TYPE					 0xd015c
 /* [RW 2] Load (FIC0) channel group priority. The lowest priority is 0; the
    highest priority is 3. It is supposed; that the Store channel priority is
-   the compliment to 4 of the rest priorities - Aggregation channel; Load
+   the complement to 4 of the rest priorities - Aggregation channel; Load
    (FIC0) channel and Load (FIC1). */
 #define CCM_REG_GR_LD0_PR					 0xd0164
 /* [RW 2] Load (FIC1) channel group priority. The lowest priority is 0; the
    highest priority is 3. It is supposed; that the Store channel priority is
-   the compliment to 4 of the rest priorities - Aggregation channel; Load
+   the complement to 4 of the rest priorities - Aggregation channel; Load
    (FIC0) channel and Load (FIC1). */
 #define CCM_REG_GR_LD1_PR					 0xd0168
 /* [RW 2] General flags index. */
@@ -868,6 +868,8 @@
 #define DORQ_REG_VF_TYPE_VALUE_0				 0x170258
 #define DORQ_REG_VF_USAGE_CT_LIMIT				 0x170340
 
+extern const u32 dmae_reg_go_c[];
+
 /* [RW 4] Initial activity counter value on the load request; when the
    shortcut is done. */
 #define DORQ_REG_SHRT_ACT_CNT					 0x170070
@@ -2212,7 +2214,7 @@
  * MAC DA 2. The reset default is set to mask out all parameters.
  */
 #define NIG_REG_P0_LLH_PTP_PARAM_MASK				 0x187a0
-/* [RW 14] Mask regiser for the rules used in detecting PTP packets. Set
+/* [RW 14] Mask register for the rules used in detecting PTP packets. Set
  * each bit to 1 to mask out that particular rule. 0-{IPv4 DA 0; UDP DP 0} .
  * 1-{IPv4 DA 0; UDP DP 1} . 2-{IPv4 DA 1; UDP DP 0} . 3-{IPv4 DA 1; UDP DP
  * 1} . 4-{IPv6 DA 0; UDP DP 0} . 5-{IPv6 DA 0; UDP DP 1} . 6-{IPv6 DA 1;
@@ -2381,7 +2383,7 @@
  * MAC DA 2. The reset default is set to mask out all parameters.
  */
 #define NIG_REG_P1_LLH_PTP_PARAM_MASK				 0x187c8
-/* [RW 14] Mask regiser for the rules used in detecting PTP packets. Set
+/* [RW 14] Mask register for the rules used in detecting PTP packets. Set
  * each bit to 1 to mask out that particular rule. 0-{IPv4 DA 0; UDP DP 0} .
  * 1-{IPv4 DA 0; UDP DP 1} . 2-{IPv4 DA 1; UDP DP 0} . 3-{IPv4 DA 1; UDP DP
  * 1} . 4-{IPv6 DA 0; UDP DP 0} . 5-{IPv6 DA 0; UDP DP 1} . 6-{IPv6 DA 1;
@@ -2493,7 +2495,7 @@
  * MAC DA 2. The reset default is set to mask out all parameters.
  */
 #define NIG_REG_P0_TLLH_PTP_PARAM_MASK				 0x187f0
-/* [RW 14] Mask regiser for the rules used in detecting PTP packets. Set
+/* [RW 14] Mask register for the rules used in detecting PTP packets. Set
  * each bit to 1 to mask out that particular rule. 0-{IPv4 DA 0; UDP DP 0} .
  * 1-{IPv4 DA 0; UDP DP 1} . 2-{IPv4 DA 1; UDP DP 0} . 3-{IPv4 DA 1; UDP DP
  * 1} . 4-{IPv6 DA 0; UDP DP 0} . 5-{IPv6 DA 0; UDP DP 1} . 6-{IPv6 DA 1;
@@ -2529,7 +2531,7 @@
  * MAC DA 2. The reset default is set to mask out all parameters.
  */
 #define NIG_REG_P1_TLLH_PTP_PARAM_MASK				 0x187f8
-/* [RW 14] Mask regiser for the rules used in detecting PTP packets. Set
+/* [RW 14] Mask register for the rules used in detecting PTP packets. Set
  * each bit to 1 to mask out that particular rule. 0-{IPv4 DA 0; UDP DP 0} .
  * 1-{IPv4 DA 0; UDP DP 1} . 2-{IPv4 DA 1; UDP DP 0} . 3-{IPv4 DA 1; UDP DP
  * 1} . 4-{IPv6 DA 0; UDP DP 0} . 5-{IPv6 DA 0; UDP DP 1} . 6-{IPv6 DA 1;
@@ -4489,11 +4491,11 @@
 #define TCM_REG_GR_ARB_TYPE					 0x50114
 /* [RW 2] Load (FIC0) channel group priority. The lowest priority is 0; the
    highest priority is 3. It is supposed that the Store channel is the
-   compliment of the other 3 groups. */
+   complement of the other 3 groups. */
 #define TCM_REG_GR_LD0_PR					 0x5011c
 /* [RW 2] Load (FIC1) channel group priority. The lowest priority is 0; the
    highest priority is 3. It is supposed that the Store channel is the
-   compliment of the other 3 groups. */
+   complement of the other 3 groups. */
 #define TCM_REG_GR_LD1_PR					 0x50120
 /* [RW 4] The number of double REG-pairs; loaded from the STORM context and
    sent to STORM; for a specific connection type. The double REG-pairs are
@@ -5020,11 +5022,11 @@
 #define UCM_REG_GR_ARB_TYPE					 0xe0144
 /* [RW 2] Load (FIC0) channel group priority. The lowest priority is 0; the
    highest priority is 3. It is supposed that the Store channel group is
-   compliment to the others. */
+   complement to the others. */
 #define UCM_REG_GR_LD0_PR					 0xe014c
 /* [RW 2] Load (FIC1) channel group priority. The lowest priority is 0; the
    highest priority is 3. It is supposed that the Store channel group is
-   compliment to the others. */
+   complement to the others. */
 #define UCM_REG_GR_LD1_PR					 0xe0150
 /* [RW 2] The queue index for invalidate counter flag decision. */
 #define UCM_REG_INV_CFLG_Q					 0xe00e4
@@ -5523,11 +5525,11 @@
 #define XCM_REG_GR_ARB_TYPE					 0x2020c
 /* [RW 2] Load (FIC0) channel group priority. The lowest priority is 0; the
    highest priority is 3. It is supposed that the Channel group is the
-   compliment of the other 3 groups. */
+   complement of the other 3 groups. */
 #define XCM_REG_GR_LD0_PR					 0x20214
 /* [RW 2] Load (FIC1) channel group priority. The lowest priority is 0; the
    highest priority is 3. It is supposed that the Channel group is the
-   compliment of the other 3 groups. */
+   complement of the other 3 groups. */
 #define XCM_REG_GR_LD1_PR					 0x20218
 /* [RW 1] Input nig0 Interface enable. If 0 - the valid input is
    disregarded; acknowledge output is deasserted; all other signals are
@@ -6218,7 +6220,7 @@
 #define AEU_INPUTS_ATTN_BITS_GPIO0_FUNCTION_0			 (0x1<<2)
 #define AEU_INPUTS_ATTN_BITS_IGU_PARITY_ERROR			 (0x1<<12)
 #define AEU_INPUTS_ATTN_BITS_MCP_LATCHED_ROM_PARITY		 (0x1<<28)
-#define AEU_INPUTS_ATTN_BITS_MCP_LATCHED_SCPAD_PARITY		 (0x1<<31)
+#define AEU_INPUTS_ATTN_BITS_MCP_LATCHED_SCPAD_PARITY		 (0x1U<<31)
 #define AEU_INPUTS_ATTN_BITS_MCP_LATCHED_UMP_RX_PARITY		 (0x1<<29)
 #define AEU_INPUTS_ATTN_BITS_MCP_LATCHED_UMP_TX_PARITY		 (0x1<<30)
 #define AEU_INPUTS_ATTN_BITS_MISC_HW_INTERRUPT			 (0x1<<15)
@@ -7639,6 +7641,82 @@ Theotherbitsarereservedandshouldbezero*/
 	(0x80 | ((_type)&0xf << 3) | ((CDU_CRC8(_cid, _region, _type)) & 0x7))
 #define CDU_RSRVD_INVALIDATE_CONTEXT_VALUE(_val) ((_val) & ~0x80)
 
+/* IdleChk registers */
+#define PXP_REG_HST_VF_DISABLED_ERROR_VALID			 0x1030bc
+#define PXP_REG_HST_VF_DISABLED_ERROR_DATA			 0x1030b8
+#define PXP_REG_HST_PER_VIOLATION_VALID				 0x1030e0
+#define PXP_REG_HST_INCORRECT_ACCESS_VALID			 0x1030cc
+#define PXP2_REG_RD_CPL_ERR_DETAILS				 0x120778
+#define PXP2_REG_RD_CPL_ERR_DETAILS2				 0x12077c
+#define PXP2_REG_RQ_GARB					 0x120748
+#define PBF_REG_DISABLE_NEW_TASK_PROC_Q0			 0x15c1bc
+#define PBF_REG_DISABLE_NEW_TASK_PROC_Q1			 0x15c1c0
+#define PBF_REG_DISABLE_NEW_TASK_PROC_Q2			 0x15c1c4
+#define PBF_REG_DISABLE_NEW_TASK_PROC_Q3			 0x15c1c8
+#define PBF_REG_DISABLE_NEW_TASK_PROC_Q4			 0x15c1cc
+#define PBF_REG_DISABLE_NEW_TASK_PROC_Q5			 0x15c1d0
+#define PBF_REG_CREDIT_Q2					 0x140344
+#define PBF_REG_CREDIT_Q3					 0x140348
+#define PBF_REG_CREDIT_Q4					 0x14034c
+#define PBF_REG_CREDIT_Q5					 0x140350
+#define PBF_REG_INIT_CRD_Q2					 0x15c238
+#define PBF_REG_INIT_CRD_Q3					 0x15c23c
+#define PBF_REG_INIT_CRD_Q4					 0x15c240
+#define PBF_REG_INIT_CRD_Q5					 0x15c244
+#define PBF_REG_TASK_CNT_Q0					 0x140374
+#define PBF_REG_TASK_CNT_Q1					 0x140378
+#define PBF_REG_TASK_CNT_Q2					 0x14037c
+#define PBF_REG_TASK_CNT_Q3					 0x140380
+#define PBF_REG_TASK_CNT_Q4					 0x140384
+#define PBF_REG_TASK_CNT_Q5					 0x140388
+#define PBF_REG_TASK_CNT_LB_Q					 0x140370
+#define QM_REG_BYTECRD0						 0x16e6fc
+#define QM_REG_BYTECRD1						 0x16e700
+#define QM_REG_BYTECRD2						 0x16e704
+#define QM_REG_BYTECRD3						 0x16e7ac
+#define QM_REG_BYTECRD4						 0x16e7b0
+#define QM_REG_BYTECRD5						 0x16e7b4
+#define QM_REG_BYTECRD6						 0x16e7b8
+#define QM_REG_BYTECRDCMDQ_0					 0x16e6e8
+#define QM_REG_BYTECRDERRREG					 0x16e708
+#define MISC_REG_GRC_TIMEOUT_ATTN_FULL_FID			 0xa714
+#define QM_REG_VOQCREDIT_2					 0x1682d8
+#define QM_REG_VOQCREDIT_3					 0x1682dc
+#define QM_REG_VOQCREDIT_5					 0x1682e4
+#define QM_REG_VOQCREDIT_6					 0x1682e8
+#define QM_REG_VOQINITCREDIT_3					 0x16806c
+#define QM_REG_VOQINITCREDIT_6					 0x168078
+#define QM_REG_FWVOQ0TOHWVOQ					 0x16e7bc
+#define QM_REG_FWVOQ1TOHWVOQ					 0x16e7c0
+#define QM_REG_FWVOQ2TOHWVOQ					 0x16e7c4
+#define QM_REG_FWVOQ3TOHWVOQ					 0x16e7c8
+#define QM_REG_FWVOQ4TOHWVOQ					 0x16e7cc
+#define QM_REG_FWVOQ5TOHWVOQ					 0x16e7d0
+#define QM_REG_FWVOQ6TOHWVOQ					 0x16e7d4
+#define QM_REG_FWVOQ7TOHWVOQ					 0x16e7d8
+#define NIG_REG_INGRESS_EOP_PORT0_EMPTY				 0x104ec
+#define NIG_REG_INGRESS_EOP_PORT1_EMPTY				 0x104f8
+#define NIG_REG_INGRESS_RMP0_DSCR_EMPTY				 0x10530
+#define NIG_REG_INGRESS_RMP1_DSCR_EMPTY				 0x10538
+#define NIG_REG_INGRESS_LB_PBF_DELAY_EMPTY			 0x10508
+#define NIG_REG_EGRESS_MNG0_FIFO_EMPTY				 0x10460
+#define NIG_REG_EGRESS_MNG1_FIFO_EMPTY				 0x10474
+#define NIG_REG_EGRESS_DEBUG_FIFO_EMPTY				 0x10418
+#define NIG_REG_EGRESS_DELAY0_EMPTY				 0x10420
+#define NIG_REG_EGRESS_DELAY1_EMPTY				 0x10428
+#define NIG_REG_LLH0_FIFO_EMPTY					 0x10548
+#define NIG_REG_LLH1_FIFO_EMPTY					 0x10558
+#define NIG_REG_P0_TX_MNG_HOST_FIFO_EMPTY			 0x182a8
+#define NIG_REG_P0_TLLH_FIFO_EMPTY				 0x18308
+#define NIG_REG_P0_HBUF_DSCR_EMPTY				 0x18318
+#define NIG_REG_P1_HBUF_DSCR_EMPTY				 0x18348
+#define NIG_REG_P0_RX_MACFIFO_EMPTY				 0x18570
+#define NIG_REG_P0_TX_MACFIFO_EMPTY				 0x18578
+#define NIG_REG_EGRESS_DELAY2_EMPTY				 0x1862c
+#define NIG_REG_EGRESS_DELAY3_EMPTY				 0x18630
+#define NIG_REG_EGRESS_DELAY4_EMPTY				 0x18634
+#define NIG_REG_EGRESS_DELAY5_EMPTY				 0x18638
+
 /******************************************************************************
  * Description:
  *	   Calculates crc 8 on a word value: polynomial 0-1-2-8
@@ -7697,6 +7775,4 @@ static inline u8 calc_crc8(u32 data, u8 crc)
 
 	return crc_res;
 }
-
-
 #endif /* BNX2X_REG_H */
diff --git a/drivers/net/ethernet/broadcom/bnx2x/bnx2x_self_test.c b/drivers/net/ethernet/broadcom/bnx2x/bnx2x_self_test.c
new file mode 100644
index 000000000000..3f8bdad3351c
--- /dev/null
+++ b/drivers/net/ethernet/broadcom/bnx2x/bnx2x_self_test.c
@@ -0,0 +1,3183 @@
+// SPDX-License-Identifier: GPL-2.0
+#include <linux/kernel.h>
+#include <linux/netdevice.h>
+#include "bnx2x.h"
+
+#define NA 0xCD
+
+#define IDLE_CHK_E1			0x01
+#define IDLE_CHK_E1H			0x02
+#define IDLE_CHK_E2			0x04
+#define IDLE_CHK_E3A0			0x08
+#define IDLE_CHK_E3B0			0x10
+
+#define IDLE_CHK_ERROR			1
+#define IDLE_CHK_ERROR_NO_TRAFFIC	2
+#define IDLE_CHK_WARNING		3
+
+#define MAX_FAIL_MSG 256
+
+/* statistics and error reporting */
+static int idle_chk_errors, idle_chk_warnings;
+
+/* masks for all chip types */
+static int is_e1, is_e1h, is_e2, is_e3a0, is_e3b0;
+
+/* struct for the argument list for a predicate in the self test databasei */
+struct st_pred_args {
+	u32 val1; /* value read from first register */
+	u32 val2; /* value read from second register, if applicable */
+	u32 imm1; /* 1st value in predicate condition, left-to-right */
+	u32 imm2; /* 2nd value in predicate condition, left-to-right */
+	u32 imm3; /* 3rd value in predicate condition, left-to-right */
+	u32 imm4; /* 4th value in predicate condition, left-to-right */
+};
+
+/* struct representing self test record - a single test */
+struct st_record {
+	u8 chip_mask;
+	u8 macro;
+	u32 reg1;
+	u32 reg2;
+	u16 loop;
+	u16 incr;
+	int (*bnx2x_predicate)(struct st_pred_args *pred_args);
+	u32 reg3;
+	u8 severity;
+	char *fail_msg;
+	struct st_pred_args pred_args;
+};
+
+/* predicates for self test */
+static int peq(struct st_pred_args *args)
+{
+	return (args->val1 == args->imm1);
+}
+
+static int pneq(struct st_pred_args *args)
+{
+	return (args->val1 != args->imm1);
+}
+
+static int pand_neq(struct st_pred_args *args)
+{
+	return ((args->val1 & args->imm1) != args->imm2);
+}
+
+static int pand_neq_x2(struct st_pred_args *args)
+{
+	return (((args->val1 & args->imm1) != args->imm2) &&
+		((args->val1 & args->imm3) != args->imm4));
+}
+
+static int pneq_err(struct st_pred_args *args)
+{
+	return ((args->val1 != args->imm1) && (idle_chk_errors > args->imm2));
+}
+
+static int pgt(struct st_pred_args *args)
+{
+	return (args->val1 > args->imm1);
+}
+
+static int pneq_r2(struct st_pred_args *args)
+{
+	return (args->val1 != args->val2);
+}
+
+static int plt_sub_r2(struct st_pred_args *args)
+{
+	return (args->val1 < (args->val2 - args->imm1));
+}
+
+static int pne_sub_r2(struct st_pred_args *args)
+{
+	return (args->val1 != (args->val2 - args->imm1));
+}
+
+static int prsh_and_neq(struct st_pred_args *args)
+{
+	return (((args->val1 >> args->imm1) & args->imm2) != args->imm3);
+}
+
+static int peq_neq_r2(struct st_pred_args *args)
+{
+	return ((args->val1 == args->imm1) && (args->val2 != args->imm2));
+}
+
+static int peq_neq_neq_r2(struct st_pred_args *args)
+{
+	return ((args->val1 == args->imm1) && (args->val2 != args->imm2) &&
+		(args->val2 != args->imm3));
+}
+
+/* struct holding the database of self test checks (registers and predicates) */
+/* lines start from 2 since line 1 is heading in csv */
+#define ST_DB_LINES 468
+static struct st_record st_database[ST_DB_LINES] = {
+/*line 2*/{(0x3), 1, 0x2114,
+	NA, 1, 0, pand_neq,
+	NA, IDLE_CHK_ERROR,
+	"PCIE: ucorr_err_status is not 0",
+	{NA, NA, 0x0FF010, 0, NA, NA} },
+
+/*line 3*/{(0x3), 1, 0x2114,
+	NA, 1, 0, pand_neq,
+	NA, IDLE_CHK_WARNING,
+	"PCIE: ucorr_err_status - Unsupported request error",
+	{NA, NA, 0x100000, 0, NA, NA} },
+
+/*line 4*/{(0x3), 1, 0x2120,
+	NA, 1, 0, pand_neq_x2,
+	NA, IDLE_CHK_WARNING,
+	"PCIE: corr_err_status is not 0x2000",
+	{NA, NA, 0x31C1, 0x2000, 0x31C1, 0} },
+
+/*line 5*/{(0x3), 1, 0x2814,
+	NA, 1, 0, pand_neq,
+	NA, IDLE_CHK_ERROR,
+	"PCIE: attentions register is not 0x40100",
+	{NA, NA, ~0x40100, 0, NA, NA} },
+
+/*line 6*/{(0x2), 1, 0x281c,
+	NA, 1, 0, pand_neq,
+	NA, IDLE_CHK_ERROR,
+	"PCIE: attentions register is not 0x40040100",
+	{NA, NA, ~0x40040100, 0, NA, NA} },
+
+/*line 7*/{(0x2), 1, 0x2820,
+	NA, 1, 0, pand_neq,
+	NA, IDLE_CHK_ERROR,
+	"PCIE: attentions register is not 0x40040100",
+	{NA, NA, ~0x40040100, 0, NA, NA} },
+
+/*line 8*/{(0x3), 1, PXP2_REG_PGL_EXP_ROM2,
+	NA, 1, 0, pneq,
+	NA, IDLE_CHK_WARNING,
+	"PXP2: There are outstanding read requests. Not all completios have arrived for read requests on tags that are marked with 0",
+	{NA, NA, 0xffffffff, NA, NA, NA} },
+
+/*line 9*/{(0x3), 2, 0x212c,
+	NA, 4, 4, pneq_err,
+	NA, IDLE_CHK_WARNING,
+	"PCIE: error packet header is not 0",
+	{NA, NA, 0, NA, NA, NA} },
+
+/*line 10*/{(0x1C), 1, 0x2104,
+	NA, 1, 0, pand_neq,
+	NA, IDLE_CHK_ERROR,
+	"PCIE: ucorr_err_status is not 0",
+	{NA, NA, 0x0FD010, 0, NA, NA} },
+
+/*line 11*/{(0x1C), 1, 0x2104,
+	NA, 1, 0, pand_neq,
+	NA, IDLE_CHK_WARNING,
+	"PCIE: ucorr_err_status - Unsupported request error",
+	{NA, NA, 0x100000, 0, NA, NA} },
+
+/*line 12*/{(0x1C), 1, 0x2104,
+	NA, 1, 0, pand_neq,
+	NA, IDLE_CHK_WARNING,
+	"PCIE: ucorr_err_status - Flow Control Protocol Error",
+	{NA, NA, 0x2000, 0, NA, NA} },
+
+/*line 13*/{(0x1C), 1, 0x2110,
+	NA, 1, 0, pand_neq_x2,
+	NA, IDLE_CHK_WARNING,
+	"PCIE: corr_err_status is not 0x2000",
+	{NA, NA, 0x31C1, 0x2000, 0x31C1, 0} },
+
+/*line 14*/{(0x1C), 1, 0x2814,
+	NA, 1, 0, pand_neq,
+	NA, IDLE_CHK_WARNING,
+	"PCIE: TTX_BRIDGE_FORWARD_ERR - Received master request while BME was 0",
+	{NA, NA, 0x2000000, 0, NA, NA} },
+
+/*line 15*/{(0x1C), 1, 0x2814,
+	NA, 1, 0, pand_neq,
+	NA, IDLE_CHK_ERROR,
+	"PCIE: Func 0 1: attentions register is not 0x2040902",
+	{NA, NA, ~0x2040902, 0, NA, NA} },
+
+/*line 16*/{(0x1C), 1, 0x2854,
+	NA, 1, 0, pand_neq,
+	NA, IDLE_CHK_ERROR,
+	"PCIE: Func 2 3 4: attentions register is not 0x10240902",
+	{NA, NA, ~0x10240902, 0, NA, NA} },
+
+/*line 17*/{(0x1C), 1, 0x285c,
+	NA, 1, 0, pand_neq,
+	NA, IDLE_CHK_ERROR,
+	"PCIE: Func 5 6 7: attentions register is not 0x10240902",
+	{NA, NA, ~0x10240902, 0, NA, NA} },
+
+/*line 18*/{(0x18), 1, 0x3040,
+	NA, 1, 0, pand_neq,
+	NA, IDLE_CHK_ERROR,
+	"PCIE: Overflow in DLP2TLP buffer",
+	{NA, NA, 0x2, 0, NA, NA} },
+
+/*line 19*/{(0x1C), 1, PXP2_REG_PGL_EXP_ROM2,
+	NA, 1, 0, pneq,
+	NA, IDLE_CHK_WARNING,
+	"PXP2: There are outstanding read requests for tags 0-31. Not all completios have arrived for read requests on tags that are marked with 0",
+	{NA, NA, 0xffffffff, NA, NA, NA} },
+
+/*line 20*/{(0x1C), 2, 0x211c,
+	NA, 4, 4, pneq_err,
+	NA, IDLE_CHK_WARNING,
+	"PCIE: error packet header is not 0",
+	{NA, NA, 0, NA, NA, NA} },
+
+/*line 21*/{(0x1C), 1, PGLUE_B_REG_INCORRECT_RCV_DETAILS,
+	NA, 1, 0, pneq,
+	NA, IDLE_CHK_ERROR,
+	"PGLUE_B: Packet received from PCIe not according to the rules",
+	{NA, NA, 0, NA, NA, NA} },
+
+/*line 22*/{(0x1C), 1, PGLUE_B_REG_WAS_ERROR_VF_31_0,
+	NA, 1, 0, pneq,
+	NA, IDLE_CHK_WARNING,
+	"PGLUE_B: was_error for VFs 0-31 is not 0",
+	{NA, NA, 0, NA, NA, NA} },
+
+/*line 23*/{(0x1C), 1, PGLUE_B_REG_WAS_ERROR_VF_63_32,
+	NA, 1, 0, pneq,
+	NA, IDLE_CHK_WARNING,
+	"PGLUE_B: was_error for VFs 32-63 is not 0",
+	{NA, NA, 0, NA, NA, NA} },
+
+/*line 24*/{(0x1C), 1, PGLUE_B_REG_WAS_ERROR_VF_95_64,
+	NA, 1, 0, pneq,
+	NA, IDLE_CHK_WARNING,
+	"PGLUE_B: was_error for VFs 64-95 is not 0",
+	{NA, NA, 0, NA, NA, NA} },
+
+/*line 25*/{(0x1C), 1, PGLUE_B_REG_WAS_ERROR_VF_127_96,
+	NA, 1, 0, pneq,
+	NA, IDLE_CHK_WARNING,
+	"PGLUE_B: was_error for VFs 96-127 is not 0",
+	{NA, NA, 0, NA, NA, NA} },
+
+/*line 26*/{(0x1C), 1, PGLUE_B_REG_WAS_ERROR_PF_7_0,
+	NA, 1, 0, pneq,
+	NA, IDLE_CHK_WARNING,
+	"PGLUE_B: was_error for PFs 0-7 is not 0",
+	{NA, NA, 0, NA, NA, NA} },
+
+/*line 27*/{(0x1C), 1, PGLUE_B_REG_RX_ERR_DETAILS,
+	NA, 1, 0, pneq,
+	NA, IDLE_CHK_WARNING,
+	"PGLUE_B: Completion received with error. (2:0) - PFID. (3) - VF_VALID. (9:4) - VFID. (11:10) - Error code : 0 - Completion Timeout; 1 - Unsupported Request; 2 - Completer Abort. (12) - valid bit",
+	{NA, NA, 0, NA, NA, NA} },
+
+/*line 28*/{(0x1C), 1, PGLUE_B_REG_RX_TCPL_ERR_DETAILS,
+	NA, 1, 0, pneq,
+	NA, IDLE_CHK_WARNING,
+	"PGLUE_B: ATS TCPL received with error. (2:0) - PFID. (3) - VF_VALID. (9:4) - VFID. (11:10) - Error code : 0 - Completion Timeout ; 1 - Unsupported Request; 2 - Completer Abort. (16:12) - OTB Entry ID. (17) - valid bit",
+	{NA, NA, 0, NA, NA, NA} },
+
+/*line 29*/{(0x1C), 1, PGLUE_B_REG_TX_ERR_WR_ADD_31_0,
+	NA, 1, 0, pneq,
+	NA, IDLE_CHK_WARNING,
+	"PGLUE_B: Error in master write. Address(31:0) is not 0",
+	{NA, NA, 0, NA, NA, NA} },
+
+/*line 30*/{(0x1C), 1, PGLUE_B_REG_TX_ERR_WR_ADD_63_32,
+	NA, 1, 0, pneq,
+	NA, IDLE_CHK_WARNING,
+	"PGLUE_B: Error in master write. Address(63:32) is not 0",
+	{NA, NA, 0, NA, NA, NA} },
+
+/*line 31*/{(0x1C), 1, PGLUE_B_REG_TX_ERR_WR_DETAILS,
+	NA, 1, 0, pneq,
+	NA, IDLE_CHK_WARNING,
+	"PGLUE_B: Error in master write. Error details register is not 0. (4:0) VQID. (23:21) - PFID. (24) - VF_VALID. (30:25) - VFID",
+	{NA, NA, 0, NA, NA, NA} },
+
+/*line 32*/{(0x1C), 1, PGLUE_B_REG_TX_ERR_WR_DETAILS2,
+	NA, 1, 0, pneq,
+	NA, IDLE_CHK_WARNING,
+	"PGLUE_B: Error in master write. Error details 2nd register is not 0. (21) - was_error set; (22) - BME cleared; (23) - FID_enable cleared; (24) - VF with parent PF FLR_request or IOV_disable_request",
+	{NA, NA, 0, NA, NA, NA} },
+
+/*line 33*/{(0x1C), 1, PGLUE_B_REG_TX_ERR_RD_ADD_31_0,
+	NA, 1, 0, pneq,
+	NA, IDLE_CHK_WARNING,
+	"PGLUE: Error in master read address(31:0) is not 0",
+	{NA, NA, 0, NA, NA, NA} },
+
+/*line 34*/{(0x1C), 1, PGLUE_B_REG_TX_ERR_RD_ADD_63_32,
+	NA, 1, 0, pneq,
+	NA, IDLE_CHK_WARNING,
+	"PGLUE_B: Error in master read address(63:32) is not 0",
+	{NA, NA, 0, NA, NA, NA} },
+
+/*line 35*/{(0x1C), 1, PGLUE_B_REG_TX_ERR_RD_DETAILS,
+	NA, 1, 0, pneq,
+	NA, IDLE_CHK_WARNING,
+	"PGLUE_B: Error in master read Error details register is not 0. (4:0) VQID. (23:21) - PFID. (24) - VF_VALID. (30:25) - VFID",
+	{NA, NA, 0, NA, NA, NA} },
+
+/*line 36*/{(0x1C), 1, PGLUE_B_REG_TX_ERR_RD_DETAILS2,
+	NA, 1, 0, pneq,
+	NA, IDLE_CHK_WARNING,
+	"PGLUE_B: Error in master read Error details 2nd register is not 0. (21) - was_error set; (22) - BME cleared; (23) - FID_enable cleared; (24) - VF with parent PF FLR_request or IOV_disable_request",
+	{NA, NA, 0, NA, NA, NA} },
+
+/*line 37*/{(0x1C), 1, PGLUE_B_REG_VF_LENGTH_VIOLATION_DETAILS,
+	NA, 1, 0, pneq,
+	NA, IDLE_CHK_WARNING,
+	"PGLUE_B: Target VF length violation access",
+	{NA, NA, 0, NA, NA, NA} },
+
+/*line 38*/{(0x1C), 1, PGLUE_B_REG_VF_GRC_SPACE_VIOLATION_DETAILS,
+	NA, 1, 0, pneq,
+	NA, IDLE_CHK_WARNING,
+	"PGLUE_B: Target VF GRC space access failed permission check",
+	{NA, NA, 0, NA, NA, NA} },
+
+/*line 39*/{(0x1C), 1, PGLUE_B_REG_TAGS_63_32,
+	NA, 1, 0, pneq,
+	NA, IDLE_CHK_WARNING,
+	"PGLUE_B: There are outstanding read requests for tags 32-63. Not all completios have arrived for read requests on tags that are marked with 0",
+	{NA, NA, 0xffffffff, NA, NA, NA} },
+
+/*line 40*/{(0x1C), 3, PXP_REG_HST_VF_DISABLED_ERROR_VALID,
+	PXP_REG_HST_VF_DISABLED_ERROR_DATA, 1, 0, pneq,
+	NA, IDLE_CHK_WARNING,
+	"PXP: Access to disabled VF took place",
+	{NA, NA, 0, NA, NA, NA} },
+
+/*line 41*/{(0x1C), 1, PXP_REG_HST_PER_VIOLATION_VALID,
+	NA, 1, 0, pneq,
+	NA, IDLE_CHK_WARNING,
+	"PXP: Zone A permission violation occurred",
+	{NA, NA, 0, NA, NA, NA} },
+
+/*line 42*/{(0x1C), 1, PXP_REG_HST_INCORRECT_ACCESS_VALID,
+	NA, 1, 0, pneq,
+	NA, IDLE_CHK_WARNING,
+	"PXP: Incorrect transaction took place",
+	{NA, NA, 0, NA, NA, NA} },
+
+/*line 43*/{(0x1C), 1, PXP2_REG_RD_CPL_ERR_DETAILS,
+	NA, 1, 0, pneq,
+	NA, IDLE_CHK_WARNING,
+	"PXP2: Completion received with error. Error details register is not 0. (15:0) - ECHO. (28:16) - Sub Request length plus start_offset_2_0 minus 1",
+	{NA, NA, 0, NA, NA, NA} },
+
+/*line 44*/{(0x1C), 1, PXP2_REG_RD_CPL_ERR_DETAILS2,
+	NA, 1, 0, pneq,
+	NA, IDLE_CHK_WARNING,
+	"PXP2: Completion received with error. Error details 2nd register is not 0. (4:0) - VQ ID. (8:5) - client ID. (9) - valid bit",
+	{NA, NA, 0, NA, NA, NA} },
+
+/*line 45*/{(0x1F), 1, PXP2_REG_RQ_VQ0_ENTRY_CNT,
+	NA, 1, 0, pneq,
+	NA, IDLE_CHK_ERROR_NO_TRAFFIC,
+	"PXP2: VQ0 is not empty",
+	{NA, NA, 0, NA, NA, NA} },
+
+/*line 46*/{(0x1F), 1, PXP2_REG_RQ_VQ1_ENTRY_CNT,
+	NA, 1, 0, pneq,
+	NA, IDLE_CHK_ERROR_NO_TRAFFIC,
+	"PXP2: VQ1 is not empty",
+	{NA, NA, 0, NA, NA, NA} },
+
+/*line 47*/{(0x1F), 1, PXP2_REG_RQ_VQ2_ENTRY_CNT,
+	NA, 1, 0, pneq,
+	NA, IDLE_CHK_ERROR_NO_TRAFFIC,
+	"PXP2: VQ2 is not empty",
+	{NA, NA, 0, NA, NA, NA} },
+
+/*line 48*/{(0x1F), 1, PXP2_REG_RQ_VQ3_ENTRY_CNT,
+	NA, 1, 0, pgt,
+	NA, IDLE_CHK_ERROR_NO_TRAFFIC,
+	"PXP2: VQ3 is not empty",
+	{NA, NA, 2, NA, NA, NA} },
+
+/*line 49*/{(0x1F), 1, PXP2_REG_RQ_VQ4_ENTRY_CNT,
+	NA, 1, 0, pneq,
+	NA, IDLE_CHK_ERROR_NO_TRAFFIC,
+	"PXP2: VQ4 is not empty",
+	{NA, NA, 0, NA, NA, NA} },
+
+/*line 50*/{(0x1F), 1, PXP2_REG_RQ_VQ5_ENTRY_CNT,
+	NA, 1, 0, pneq,
+	NA, IDLE_CHK_ERROR_NO_TRAFFIC,
+	"PXP2: VQ5 is not empty",
+	{NA, NA, 0, NA, NA, NA} },
+
+/*line 51*/{(0x1F), 1, PXP2_REG_RQ_VQ6_ENTRY_CNT,
+	NA, 1, 0, pneq,
+	NA, IDLE_CHK_ERROR_NO_TRAFFIC,
+	"PXP2: VQ6 is not empty",
+	{NA, NA, 0, NA, NA, NA} },
+
+/*line 52*/{(0x1F), 1, PXP2_REG_RQ_VQ7_ENTRY_CNT,
+	NA, 1, 0, pneq,
+	NA, IDLE_CHK_ERROR_NO_TRAFFIC,
+	"PXP2: VQ7 is not empty",
+	{NA, NA, 0, NA, NA, NA} },
+
+/*line 53*/{(0x1F), 1, PXP2_REG_RQ_VQ8_ENTRY_CNT,
+	NA, 1, 0, pneq,
+	NA, IDLE_CHK_ERROR_NO_TRAFFIC,
+	"PXP2: VQ8 is not empty",
+	{NA, NA, 0, NA, NA, NA} },
+
+/*line 54*/{(0x1F), 1, PXP2_REG_RQ_VQ9_ENTRY_CNT,
+	NA, 1, 0, pneq,
+	NA, IDLE_CHK_ERROR_NO_TRAFFIC,
+	"PXP2: VQ9 is not empty",
+	{NA, NA, 0, NA, NA, NA} },
+
+/*line 55*/{(0x1F), 1, PXP2_REG_RQ_VQ10_ENTRY_CNT,
+	NA, 1, 0, pneq,
+	NA, IDLE_CHK_ERROR_NO_TRAFFIC,
+	"PXP2: VQ10 is not empty",
+	{NA, NA, 0, NA, NA, NA} },
+
+/*line 56*/{(0x1F), 1, PXP2_REG_RQ_VQ11_ENTRY_CNT,
+	NA, 1, 0, pneq,
+	NA, IDLE_CHK_ERROR_NO_TRAFFIC,
+	"PXP2: VQ11 is not empty",
+	{NA, NA, 0, NA, NA, NA} },
+
+/*line 57*/{(0x1F), 1, PXP2_REG_RQ_VQ12_ENTRY_CNT,
+	NA, 1, 0, pneq,
+	NA, IDLE_CHK_ERROR_NO_TRAFFIC,
+	"PXP2: VQ12 is not empty",
+	{NA, NA, 0, NA, NA, NA} },
+
+/*line 58*/{(0x1F), 1, PXP2_REG_RQ_VQ13_ENTRY_CNT,
+	NA, 1, 0, pneq,
+	NA, IDLE_CHK_ERROR_NO_TRAFFIC,
+	"PXP2: VQ13 is not empty",
+	{NA, NA, 0, NA, NA, NA} },
+
+/*line 59*/{(0x1F), 1, PXP2_REG_RQ_VQ14_ENTRY_CNT,
+	NA, 1, 0, pneq,
+	NA, IDLE_CHK_ERROR_NO_TRAFFIC,
+	"PXP2: VQ14 is not empty",
+	{NA, NA, 0, NA, NA, NA} },
+
+/*line 60*/{(0x1F), 1, PXP2_REG_RQ_VQ15_ENTRY_CNT,
+	NA, 1, 0, pneq,
+	NA, IDLE_CHK_ERROR_NO_TRAFFIC,
+	"PXP2: VQ15 is not empty",
+	{NA, NA, 0, NA, NA, NA} },
+
+/*line 61*/{(0x1F), 1, PXP2_REG_RQ_VQ16_ENTRY_CNT,
+	NA, 1, 0, pneq,
+	NA, IDLE_CHK_ERROR_NO_TRAFFIC,
+	"PXP2: VQ16 is not empty",
+	{NA, NA, 0, NA, NA, NA} },
+
+/*line 62*/{(0x1F), 1, PXP2_REG_RQ_VQ17_ENTRY_CNT,
+	NA, 1, 0, pneq,
+	NA, IDLE_CHK_ERROR_NO_TRAFFIC,
+	"PXP2: VQ17 is not empty",
+	{NA, NA, 0, NA, NA, NA} },
+
+/*line 63*/{(0x1F), 1, PXP2_REG_RQ_VQ18_ENTRY_CNT,
+	NA, 1, 0, pneq,
+	NA, IDLE_CHK_ERROR_NO_TRAFFIC,
+	"PXP2: VQ18 is not empty",
+	{NA, NA, 0, NA, NA, NA} },
+
+/*line 64*/{(0x1F), 1, PXP2_REG_RQ_VQ19_ENTRY_CNT,
+	NA, 1, 0, pneq,
+	NA, IDLE_CHK_ERROR_NO_TRAFFIC,
+	"PXP2: VQ19 is not empty",
+	{NA, NA, 0, NA, NA, NA} },
+
+/*line 65*/{(0x1F), 1, PXP2_REG_RQ_VQ20_ENTRY_CNT,
+	NA, 1, 0, pneq,
+	NA, IDLE_CHK_ERROR_NO_TRAFFIC,
+	"PXP2: VQ20 is not empty",
+	{NA, NA, 0, NA, NA, NA} },
+
+/*line 66*/{(0x1F), 1, PXP2_REG_RQ_VQ21_ENTRY_CNT,
+	NA, 1, 0, pneq,
+	NA, IDLE_CHK_ERROR_NO_TRAFFIC,
+	"PXP2: VQ21 is not empty",
+	{NA, NA, 0, NA, NA, NA} },
+
+/*line 67*/{(0x1F), 1, PXP2_REG_RQ_VQ22_ENTRY_CNT,
+	NA, 1, 0, pneq,
+	NA, IDLE_CHK_ERROR_NO_TRAFFIC,
+	"PXP2: VQ22 is not empty",
+	{NA, NA, 0, NA, NA, NA} },
+
+/*line 68*/{(0x1F), 1, PXP2_REG_RQ_VQ23_ENTRY_CNT,
+	NA, 1, 0, pneq,
+	NA, IDLE_CHK_ERROR_NO_TRAFFIC,
+	"PXP2: VQ23 is not empty",
+	{NA, NA, 0, NA, NA, NA} },
+
+/*line 69*/{(0x1F), 1, PXP2_REG_RQ_VQ24_ENTRY_CNT,
+	NA, 1, 0, pneq,
+	NA, IDLE_CHK_ERROR_NO_TRAFFIC,
+	"PXP2: VQ24 is not empty",
+	{NA, NA, 0, NA, NA, NA} },
+
+/*line 70*/{(0x1F), 1, PXP2_REG_RQ_VQ25_ENTRY_CNT,
+	NA, 1, 0, pneq,
+	NA, IDLE_CHK_ERROR_NO_TRAFFIC,
+	"PXP2: VQ25 is not empty",
+	{NA, NA, 0, NA, NA, NA} },
+
+/*line 71*/{(0x1F), 1, PXP2_REG_RQ_VQ26_ENTRY_CNT,
+	NA, 1, 0, pneq,
+	NA, IDLE_CHK_ERROR_NO_TRAFFIC,
+	"PXP2: VQ26 is not empty",
+	{NA, NA, 0, NA, NA, NA} },
+
+/*line 72*/{(0x1F), 1, PXP2_REG_RQ_VQ27_ENTRY_CNT,
+	NA, 1, 0, pneq,
+	NA, IDLE_CHK_ERROR_NO_TRAFFIC,
+	"PXP2: VQ27 is not empty",
+	{NA, NA, 0, NA, NA, NA} },
+
+/*line 73*/{(0x1F), 1, PXP2_REG_RQ_VQ28_ENTRY_CNT,
+	NA, 1, 0, pneq,
+	NA, IDLE_CHK_ERROR_NO_TRAFFIC,
+	"PXP2: VQ28 is not empty",
+	{NA, NA, 0, NA, NA, NA} },
+
+/*line 74*/{(0x1F), 1, PXP2_REG_RQ_VQ29_ENTRY_CNT,
+	NA, 1, 0, pneq,
+	NA, IDLE_CHK_ERROR_NO_TRAFFIC,
+	"PXP2: VQ29 is not empty",
+	{NA, NA, 0, NA, NA, NA} },
+
+/*line 75*/{(0x1F), 1, PXP2_REG_RQ_VQ30_ENTRY_CNT,
+	NA, 1, 0, pneq,
+	NA, IDLE_CHK_ERROR_NO_TRAFFIC,
+	"PXP2: VQ30 is not empty",
+	{NA, NA, 0, NA, NA, NA} },
+
+/*line 76*/{(0x1F), 1, PXP2_REG_RQ_VQ31_ENTRY_CNT,
+	NA, 1, 0, pneq,
+	NA, IDLE_CHK_ERROR_NO_TRAFFIC,
+	"PXP2: VQ31 is not empty",
+	{NA, NA, 0, NA, NA, NA} },
+
+/*line 77*/{(0x1F), 1, PXP2_REG_RQ_UFIFO_NUM_OF_ENTRY,
+	NA, 1, 0, pneq,
+	NA, IDLE_CHK_ERROR_NO_TRAFFIC,
+	"PXP2: rq_ufifo_num_of_entry is not 0",
+	{NA, NA, 0, NA, NA, NA} },
+
+/*line 78*/{(0x1F), 1, PXP2_REG_RQ_RBC_DONE,
+	NA, 1, 0, pneq,
+	NA, IDLE_CHK_ERROR,
+	"PXP2: rq_rbc_done is not 1",
+	{NA, NA, 1, NA, NA, NA} },
+
+/*line 79*/{(0x1F), 1, PXP2_REG_RQ_CFG_DONE,
+	NA, 1, 0, pneq,
+	NA, IDLE_CHK_ERROR,
+	"PXP2: rq_cfg_done is not 1",
+	{NA, NA, 1, NA, NA, NA} },
+
+/*line 80*/{(0x3), 1, PXP2_REG_PSWRQ_BW_CREDIT,
+	NA, 1, 0, pneq,
+	NA, IDLE_CHK_ERROR_NO_TRAFFIC,
+	"PXP2: rq_read_credit and rq_write_credit are not 3",
+	{NA, NA, 0x1B, NA, NA, NA} },
+
+/*line 81*/{(0x1F), 1, PXP2_REG_RD_START_INIT,
+	NA, 1, 0, pneq,
+	NA, IDLE_CHK_ERROR,
+	"PXP2: rd_start_init is not 1",
+	{NA, NA, 1, NA, NA, NA} },
+
+/*line 82*/{(0x1F), 1, PXP2_REG_RD_INIT_DONE,
+	NA, 1, 0, pneq,
+	NA, IDLE_CHK_ERROR,
+	"PXP2: rd_init_done is not 1",
+	{NA, NA, 1, NA, NA, NA} },
+
+/*line 83*/{(0x1F), 3, PXP2_REG_RD_SR_CNT,
+	PXP2_REG_RD_SR_NUM_CFG, 1, 0, pne_sub_r2,
+	NA, IDLE_CHK_WARNING,
+	"PXP2: rd_sr_cnt is not equal to rd_sr_num_cfg",
+	{NA, NA, 1, NA, NA, NA} },
+
+/*line 84*/{(0x1F), 3, PXP2_REG_RD_BLK_CNT,
+	PXP2_REG_RD_BLK_NUM_CFG, 1, 0, pneq_r2,
+	NA, IDLE_CHK_WARNING,
+	"PXP2: rd_blk_cnt is not equal to rd_blk_num_cfg",
+	{NA, NA, NA, NA, NA, NA} },
+
+/*line 85*/{(0x1F), 3, PXP2_REG_RD_SR_CNT,
+	PXP2_REG_RD_SR_NUM_CFG, 1, 0, plt_sub_r2,
+	NA, IDLE_CHK_ERROR_NO_TRAFFIC,
+	"PXP2: There are more than two unused SRs",
+	{NA, NA, 3, NA, NA, NA} },
+
+/*line 86*/{(0x1F), 3, PXP2_REG_RD_BLK_CNT,
+	PXP2_REG_RD_BLK_NUM_CFG, 1, 0, plt_sub_r2,
+	NA, IDLE_CHK_ERROR_NO_TRAFFIC,
+	"PXP2: There are more than two unused blocks",
+	{NA, NA, 2, NA, NA, NA} },
+
+/*line 87*/{(0x1F), 1, PXP2_REG_RD_PORT_IS_IDLE_0,
+	NA, 1, 0, pneq,
+	NA, IDLE_CHK_ERROR_NO_TRAFFIC,
+	"PXP2: P0 All delivery ports are not idle",
+	{NA, NA, 1, NA, NA, NA} },
+
+/*line 88*/{(0x1F), 1, PXP2_REG_RD_PORT_IS_IDLE_1,
+	NA, 1, 0, pneq,
+	NA, IDLE_CHK_ERROR_NO_TRAFFIC,
+	"PXP2: P1 All delivery ports are not idle",
+	{NA, NA, 1, NA, NA, NA} },
+
+/*line 89*/{(0x1F), 2, PXP2_REG_RD_ALMOST_FULL_0,
+	NA, 11, 4, pneq,
+	NA, IDLE_CHK_ERROR_NO_TRAFFIC,
+	"PXP2: rd_almost_full is not 0",
+	{NA, NA, 0, NA, NA, NA} },
+
+/*line 90*/{(0x1F), 1, PXP2_REG_RD_DISABLE_INPUTS,
+	NA, 1, 0, pneq,
+	NA, IDLE_CHK_ERROR,
+	"PXP2: PSWRD inputs are disabled",
+	{NA, NA, 0, NA, NA, NA} },
+
+/*line 91*/{(0x1F), 1, PXP2_REG_HST_HEADER_FIFO_STATUS,
+	NA, 1, 0, pneq,
+	NA, IDLE_CHK_ERROR_NO_TRAFFIC,
+	"PXP2: HST header FIFO status is not 0",
+	{NA, NA, 0, NA, NA, NA} },
+
+/*line 92*/{(0x1F), 1, PXP2_REG_HST_DATA_FIFO_STATUS,
+	NA, 1, 0, pneq,
+	NA, IDLE_CHK_ERROR_NO_TRAFFIC,
+	"PXP2: HST data FIFO status is not 0",
+	{NA, NA, 0, NA, NA, NA} },
+
+/*line 93*/{(0x3), 1, PXP2_REG_PGL_WRITE_BLOCKED,
+	NA, 1, 0, pneq,
+	NA, IDLE_CHK_ERROR,
+	"PXP2: pgl_write_blocked is not 0",
+	{NA, NA, 0, NA, NA, NA} },
+
+/*line 94*/{(0x3), 1, PXP2_REG_PGL_READ_BLOCKED,
+	NA, 1, 0, pneq,
+	NA, IDLE_CHK_ERROR,
+	"PXP2: pgl_read_blocked is not 0",
+	{NA, NA, 0, NA, NA, NA} },
+
+/*line 95*/{(0x1C), 1, PXP2_REG_PGL_WRITE_BLOCKED,
+	NA, 1, 0, pneq,
+	NA, IDLE_CHK_WARNING,
+	"PXP2: pgl_write_blocked is not 0",
+	{NA, NA, 0, NA, NA, NA} },
+
+/*line 96*/{(0x1C), 1, PXP2_REG_PGL_READ_BLOCKED,
+	NA, 1, 0, pneq,
+	NA, IDLE_CHK_WARNING,
+	"PXP2: pgl_read_blocked is not 0",
+	{NA, NA, 0, NA, NA, NA} },
+
+/*line 97*/{(0x1F), 1, PXP2_REG_PGL_TXW_CDTS,
+	NA, 1, 0, prsh_and_neq,
+	NA, IDLE_CHK_ERROR_NO_TRAFFIC,
+	"PXP2: There is data which is ready",
+	{NA, NA, 17, 1, 0, NA} },
+
+/*line 98*/{(0x1F), 1, PXP_REG_HST_ARB_IS_IDLE,
+	NA, 1, 0, pneq,
+	NA, IDLE_CHK_WARNING,
+	"PXP: HST arbiter is not idle",
+	{NA, NA, 1, NA, NA, NA} },
+
+/*line 99*/{(0x1F), 1, PXP_REG_HST_CLIENTS_WAITING_TO_ARB,
+	NA, 1, 0, pneq,
+	NA, IDLE_CHK_WARNING,
+	"PXP: HST one of the clients is waiting for delivery",
+	{NA, NA, 0, NA, NA, NA} },
+
+/*line 100*/{(0x1E), 1, PXP_REG_HST_DISCARD_INTERNAL_WRITES_STATUS,
+	NA, 1, 0, pneq,
+	NA, IDLE_CHK_WARNING,
+	"PXP: HST Close the gates: Discarding internal writes",
+	{NA, NA, 0, NA, NA, NA} },
+
+/*line 101*/{(0x1E), 1, PXP_REG_HST_DISCARD_DOORBELLS_STATUS,
+	NA, 1, 0, pneq,
+	NA, IDLE_CHK_WARNING,
+	"PXP: HST Close the gates: Discarding doorbells",
+	{NA, NA, 0, NA, NA, NA} },
+
+/*line 102*/{(0x1C), 1, PXP2_REG_RQ_GARB,
+	NA, 1, 0, pand_neq,
+	NA, IDLE_CHK_WARNING,
+	"PXP2: PSWRQ Close the gates is asserted. Check AEU AFTER_INVERT registers for parity errors",
+	{NA, NA, 0x1000, 0, NA, NA} },
+
+/*line 103*/{(0x1F), 1, DMAE_REG_GO_C0,
+	NA, 1, 0, pneq,
+	NA, IDLE_CHK_ERROR_NO_TRAFFIC,
+	"DMAE: command 0 go is not 0",
+	{NA, NA, 0, NA, NA, NA} },
+
+/*line 104*/{(0x1F), 1, DMAE_REG_GO_C1,
+	NA, 1, 0, pneq,
+	NA, IDLE_CHK_ERROR_NO_TRAFFIC,
+	"DMAE: command 1 go is not 0",
+	{NA, NA, 0, NA, NA, NA} },
+
+/*line 105*/{(0x1F), 1, DMAE_REG_GO_C2,
+	NA, 1, 0, pneq,
+	NA, IDLE_CHK_ERROR_NO_TRAFFIC,
+	"DMAE: command 2 go is not 0",
+	{NA, NA, 0, NA, NA, NA} },
+
+/*line 106*/{(0x1F), 1, DMAE_REG_GO_C3,
+	NA, 1, 0, pneq,
+	NA, IDLE_CHK_ERROR_NO_TRAFFIC,
+	"DMAE: command 3 go is not 0",
+	{NA, NA, 0, NA, NA, NA} },
+
+/*line 107*/{(0x1F), 1, DMAE_REG_GO_C4,
+	NA, 1, 0, pneq,
+	NA, IDLE_CHK_ERROR_NO_TRAFFIC,
+	"DMAE: command 4 go is not 0",
+	{NA, NA, 0, NA, NA, NA} },
+
+/*line 108*/{(0x1F), 1, DMAE_REG_GO_C5,
+	NA, 1, 0, pneq,
+	NA, IDLE_CHK_ERROR_NO_TRAFFIC,
+	"DMAE: command 5 go is not 0",
+	{NA, NA, 0, NA, NA, NA} },
+
+/*line 109*/{(0x1F), 1, DMAE_REG_GO_C6,
+	NA, 1, 0, pneq,
+	NA, IDLE_CHK_ERROR_NO_TRAFFIC,
+	"DMAE: command 6 go is not 0",
+	{NA, NA, 0, NA, NA, NA} },
+
+/*line 110*/{(0x1F), 1, DMAE_REG_GO_C7,
+	NA, 1, 0, pneq,
+	NA, IDLE_CHK_ERROR_NO_TRAFFIC,
+	"DMAE: command 7 go is not 0",
+	{NA, NA, 0, NA, NA, NA} },
+
+/*line 111*/{(0x1F), 1, DMAE_REG_GO_C8,
+	NA, 1, 0, pneq,
+	NA, IDLE_CHK_ERROR_NO_TRAFFIC,
+	"DMAE: command 8 go is not 0",
+	{NA, NA, 0, NA, NA, NA} },
+
+/*line 112*/{(0x1F), 1, DMAE_REG_GO_C9,
+	NA, 1, 0, pneq,
+	NA, IDLE_CHK_ERROR_NO_TRAFFIC,
+	"DMAE: command 9 go is not 0",
+	{NA, NA, 0, NA, NA, NA} },
+
+/*line 113*/{(0x1F), 1, DMAE_REG_GO_C10,
+	NA, 1, 0, pneq,
+	NA, IDLE_CHK_ERROR_NO_TRAFFIC,
+	"DMAE: command 10 go is not 0",
+	{NA, NA, 0, NA, NA, NA} },
+
+/*line 114*/{(0x1F), 1, DMAE_REG_GO_C11,
+	NA, 1, 0, pneq,
+	NA, IDLE_CHK_ERROR_NO_TRAFFIC,
+	"DMAE: command 11 go is not 0",
+	{NA, NA, 0, NA, NA, NA} },
+
+/*line 115*/{(0x1F), 1, DMAE_REG_GO_C12,
+	NA, 1, 0, pneq,
+	NA, IDLE_CHK_ERROR_NO_TRAFFIC,
+	"DMAE: command 12 go is not 0",
+	{NA, NA, 0, NA, NA, NA} },
+
+/*line 116*/{(0x1F), 1, DMAE_REG_GO_C13,
+	NA, 1, 0, pneq,
+	NA, IDLE_CHK_ERROR_NO_TRAFFIC,
+	"DMAE: command 13 go is not 0",
+	{NA, NA, 0, NA, NA, NA} },
+
+/*line 117*/{(0x1F), 1, DMAE_REG_GO_C14,
+	NA, 1, 0, pneq,
+	NA, IDLE_CHK_ERROR_NO_TRAFFIC,
+	"DMAE: command 14 go is not 0",
+	{NA, NA, 0, NA, NA, NA} },
+
+/*line 118*/{(0x1F), 1, DMAE_REG_GO_C15,
+	NA, 1, 0, pneq,
+	NA, IDLE_CHK_ERROR_NO_TRAFFIC,
+	"DMAE: command 15 go is not 0",
+	{NA, NA, 0, NA, NA, NA} },
+
+/*line 119*/{(0x1F), 1, CFC_REG_ERROR_VECTOR,
+	NA, 1, 0, pneq,
+	NA, IDLE_CHK_ERROR,
+	"CFC: error vector is not 0",
+	{NA, NA, 0, NA, NA, NA} },
+
+/*line 120*/{(0x1F), 1, CFC_REG_NUM_LCIDS_ARRIVING,
+	NA, 1, 0, pneq,
+	NA, IDLE_CHK_ERROR,
+	"CFC: number of arriving LCIDs is not 0",
+	{NA, NA, 0, NA, NA, NA} },
+
+/*line 121*/{(0x1F), 1, CFC_REG_NUM_LCIDS_ALLOC,
+	NA, 1, 0, pneq,
+	NA, IDLE_CHK_ERROR,
+	"CFC: number of alloc LCIDs is not 0",
+	{NA, NA, 0, NA, NA, NA} },
+
+/*line 122*/{(0x1F), 1, CFC_REG_NUM_LCIDS_LEAVING,
+	NA, 1, 0, pneq,
+	NA, IDLE_CHK_ERROR,
+	"CFC: number of leaving LCIDs is not 0",
+	{NA, NA, 0, NA, NA, NA} },
+
+/*line 123*/{(0x1F), 7, CFC_REG_INFO_RAM,
+	CFC_REG_CID_CAM, (CFC_REG_INFO_RAM_SIZE >> 4), 16, peq_neq_neq_r2,
+	CFC_REG_ACTIVITY_COUNTER, IDLE_CHK_ERROR_NO_TRAFFIC,
+	"CFC: AC is neither 0 nor 2 on connType 0 (ETH)",
+	{NA, NA, 0, 0, 2, NA} },
+
+/*line 124*/{(0x1F), 7, CFC_REG_INFO_RAM,
+	CFC_REG_CID_CAM, (CFC_REG_INFO_RAM_SIZE >> 4), 16, peq_neq_r2,
+	CFC_REG_ACTIVITY_COUNTER, IDLE_CHK_ERROR_NO_TRAFFIC,
+	"CFC: AC is not 0 on connType 1 (TOE)",
+	{NA, NA, 1, 0, NA, NA} },
+
+/*line 125*/{(0x1F), 7, CFC_REG_INFO_RAM,
+	CFC_REG_CID_CAM, (CFC_REG_INFO_RAM_SIZE >> 4), 16, peq_neq_r2,
+	CFC_REG_ACTIVITY_COUNTER, IDLE_CHK_ERROR_NO_TRAFFIC,
+	"CFC: AC is not 0 on connType 3 (iSCSI)",
+	{NA, NA, 3, 0, NA, NA} },
+
+/*line 126*/{(0x1F), 7, CFC_REG_INFO_RAM,
+	CFC_REG_CID_CAM, (CFC_REG_INFO_RAM_SIZE >> 4), 16, peq_neq_r2,
+	CFC_REG_ACTIVITY_COUNTER, IDLE_CHK_ERROR_NO_TRAFFIC,
+	"CFC: AC is not 0 on connType 4 (FCoE)",
+	{NA, NA, 4, 0, NA, NA} },
+
+/*line 127*/{(0x1F), 2, QM_REG_QTASKCTR_0,
+	NA, 64, 4, pneq,
+	NA, IDLE_CHK_ERROR_NO_TRAFFIC,
+	"QM: Queue is not empty",
+	{NA, NA, 0, NA, NA, NA} },
+
+/*line 128*/{(0xF), 3, QM_REG_VOQCREDIT_0,
+	QM_REG_VOQINITCREDIT_0, 1, 0, pneq_r2,
+	NA, IDLE_CHK_ERROR_NO_TRAFFIC,
+	"QM: VOQ_0, VOQ credit is not equal to initial credit",
+	{NA, NA, NA, NA, NA, NA} },
+
+/*line 129*/{(0xF), 3, QM_REG_VOQCREDIT_1,
+	QM_REG_VOQINITCREDIT_1, 1, 0, pneq_r2,
+	NA, IDLE_CHK_ERROR_NO_TRAFFIC,
+	"QM: VOQ_1, VOQ credit is not equal to initial credit",
+	{NA, NA, NA, NA, NA, NA} },
+
+/*line 130*/{(0xF), 3, QM_REG_VOQCREDIT_4,
+	QM_REG_VOQINITCREDIT_4, 1, 0, pneq_r2,
+	NA, IDLE_CHK_ERROR_NO_TRAFFIC,
+	"QM: VOQ_4, VOQ credit is not equal to initial credit",
+	{NA, NA, NA, NA, NA, NA} },
+
+/*line 131*/{(0x3), 3, QM_REG_PORT0BYTECRD,
+	QM_REG_BYTECRDINITVAL, 1, 0, pneq_r2,
+	NA, IDLE_CHK_ERROR_NO_TRAFFIC,
+	"QM: P0 Byte credit is not equal to initial credit",
+	{NA, NA, NA, NA, NA, NA} },
+
+/*line 132*/{(0x3), 3, QM_REG_PORT1BYTECRD,
+	QM_REG_BYTECRDINITVAL, 1, 0, pneq_r2,
+	NA, IDLE_CHK_ERROR_NO_TRAFFIC,
+	"QM: P1 Byte credit is not equal to initial credit",
+	{NA, NA, NA, NA, NA, NA} },
+
+/*line 133*/{(0x1F), 1, CCM_REG_CAM_OCCUP,
+	NA, 1, 0, pneq,
+	NA, IDLE_CHK_ERROR_NO_TRAFFIC,
+	"CCM: XX protection CAM is not empty",
+	{NA, NA, 0, NA, NA, NA} },
+
+/*line 134*/{(0x1F), 1, TCM_REG_CAM_OCCUP,
+	NA, 1, 0, pneq,
+	NA, IDLE_CHK_ERROR_NO_TRAFFIC,
+	"TCM: XX protection CAM is not empty",
+	{NA, NA, 0, NA, NA, NA} },
+
+/*line 135*/{(0x1F), 1, UCM_REG_CAM_OCCUP,
+	NA, 1, 0, pneq,
+	NA, IDLE_CHK_ERROR_NO_TRAFFIC,
+	"UCM: XX protection CAM is not empty",
+	{NA, NA, 0, NA, NA, NA} },
+
+/*line 136*/{(0x1F), 1, XCM_REG_CAM_OCCUP,
+	NA, 1, 0, pneq,
+	NA, IDLE_CHK_ERROR_NO_TRAFFIC,
+	"XCM: XX protection CAM is not empty",
+	{NA, NA, 0, NA, NA, NA} },
+
+/*line 137*/{(0x1F), 1, BRB1_REG_NUM_OF_FULL_BLOCKS,
+	NA, 1, 0, pneq,
+	NA, IDLE_CHK_ERROR_NO_TRAFFIC,
+	"BRB1: BRB is not empty",
+	{NA, NA, 0, NA, NA, NA} },
+
+/*line 138*/{(0x1F), 1, CSEM_REG_SLEEP_THREADS_VALID,
+	NA, 1, 0, pneq,
+	NA, IDLE_CHK_ERROR_NO_TRAFFIC,
+	"CSEM: There are sleeping threads",
+	{NA, NA, 0, NA, NA, NA} },
+
+/*line 139*/{(0x1F), 1, TSEM_REG_SLEEP_THREADS_VALID,
+	NA, 1, 0, pneq,
+	NA, IDLE_CHK_ERROR_NO_TRAFFIC,
+	"TSEM: There are sleeping threads",
+	{NA, NA, 0, NA, NA, NA} },
+
+/*line 140*/{(0x1F), 1, USEM_REG_SLEEP_THREADS_VALID,
+	NA, 1, 0, pneq,
+	NA, IDLE_CHK_ERROR_NO_TRAFFIC,
+	"USEM: There are sleeping threads",
+	{NA, NA, 0, NA, NA, NA} },
+
+/*line 141*/{(0x1F), 1, XSEM_REG_SLEEP_THREADS_VALID,
+	NA, 1, 0, pneq,
+	NA, IDLE_CHK_ERROR_NO_TRAFFIC,
+	"XSEM: There are sleeping threads",
+	{NA, NA, 0, NA, NA, NA} },
+
+/*line 142*/{(0x1F), 1, CSEM_REG_SLOW_EXT_STORE_EMPTY,
+	NA, 1, 0, pneq,
+	NA, IDLE_CHK_ERROR_NO_TRAFFIC,
+	"CSEM: External store FIFO is not empty",
+	{NA, NA, 1, NA, NA, NA} },
+
+/*line 143*/{(0x1F), 1, TSEM_REG_SLOW_EXT_STORE_EMPTY,
+	NA, 1, 0, pneq,
+	NA, IDLE_CHK_ERROR_NO_TRAFFIC,
+	"TSEM: External store FIFO is not empty",
+	{NA, NA, 1, NA, NA, NA} },
+
+/*line 144*/{(0x1F), 1, USEM_REG_SLOW_EXT_STORE_EMPTY,
+	NA, 1, 0, pneq,
+	NA, IDLE_CHK_ERROR_NO_TRAFFIC,
+	"USEM: External store FIFO is not empty",
+	{NA, NA, 1, NA, NA, NA} },
+
+/*line 145*/{(0x1F), 1, XSEM_REG_SLOW_EXT_STORE_EMPTY,
+	NA, 1, 0, pneq,
+	NA, IDLE_CHK_ERROR_NO_TRAFFIC,
+	"XSEM: External store FIFO is not empty",
+	{NA, NA, 1, NA, NA, NA} },
+
+/*line 146*/{(0x1F), 1, CSDM_REG_SYNC_PARSER_EMPTY,
+	NA, 1, 0, pneq,
+	NA, IDLE_CHK_ERROR_NO_TRAFFIC,
+	"CSDM: Parser serial FIFO is not empty",
+	{NA, NA, 1, NA, NA, NA} },
+
+/*line 147*/{(0x1F), 1, TSDM_REG_SYNC_PARSER_EMPTY,
+	NA, 1, 0, pneq,
+	NA, IDLE_CHK_ERROR_NO_TRAFFIC,
+	"TSDM: Parser serial FIFO is not empty",
+	{NA, NA, 1, NA, NA, NA} },
+
+/*line 148*/{(0x1F), 1, USDM_REG_SYNC_PARSER_EMPTY,
+	NA, 1, 0, pneq,
+	NA, IDLE_CHK_ERROR_NO_TRAFFIC,
+	"USDM: Parser serial FIFO is not empty",
+	{NA, NA, 1, NA, NA, NA} },
+
+/*line 149*/{(0x1F), 1, XSDM_REG_SYNC_PARSER_EMPTY,
+	NA, 1, 0, pneq,
+	NA, IDLE_CHK_ERROR_NO_TRAFFIC,
+	"XSDM: Parser serial FIFO is not empty",
+	{NA, NA, 1, NA, NA, NA} },
+
+/*line 150*/{(0x1F), 1, CSDM_REG_SYNC_SYNC_EMPTY,
+	NA, 1, 0, pneq,
+	NA, IDLE_CHK_ERROR_NO_TRAFFIC,
+	"CSDM: Parser SYNC serial FIFO is not empty",
+	{NA, NA, 1, NA, NA, NA} },
+
+/*line 151*/{(0x1F), 1, TSDM_REG_SYNC_SYNC_EMPTY,
+	NA, 1, 0, pneq,
+	NA, IDLE_CHK_ERROR_NO_TRAFFIC,
+	"TSDM: Parser SYNC serial FIFO is not empty",
+	{NA, NA, 1, NA, NA, NA} },
+
+/*line 152*/{(0x1F), 1, USDM_REG_SYNC_SYNC_EMPTY,
+	NA, 1, 0, pneq,
+	NA, IDLE_CHK_ERROR_NO_TRAFFIC,
+	"USDM: Parser SYNC serial FIFO is not empty",
+	{NA, NA, 1, NA, NA, NA} },
+
+/*line 153*/{(0x1F), 1, XSDM_REG_SYNC_SYNC_EMPTY,
+	NA, 1, 0, pneq,
+	NA, IDLE_CHK_ERROR_NO_TRAFFIC,
+	"XSDM: Parser SYNC serial FIFO is not empty",
+	{NA, NA, 1, NA, NA, NA} },
+
+/*line 154*/{(0x1F), 1, CSDM_REG_RSP_PXP_CTRL_RDATA_EMPTY,
+	NA, 1, 0, pneq,
+	NA, IDLE_CHK_ERROR_NO_TRAFFIC,
+	"CSDM: pxp_ctrl rd_data fifo is not empty in sdm_dma_rsp block",
+	{NA, NA, 1, NA, NA, NA} },
+
+/*line 155*/{(0x1F), 1, TSDM_REG_RSP_PXP_CTRL_RDATA_EMPTY,
+	NA, 1, 0, pneq,
+	NA, IDLE_CHK_ERROR_NO_TRAFFIC,
+	"TSDM: pxp_ctrl rd_data fifo is not empty in sdm_dma_rsp block",
+	{NA, NA, 1, NA, NA, NA} },
+
+/*line 156*/{(0x1F), 1, USDM_REG_RSP_PXP_CTRL_RDATA_EMPTY,
+	NA, 1, 0, pneq,
+	NA, IDLE_CHK_ERROR_NO_TRAFFIC,
+	"USDM: pxp_ctrl rd_data fifo is not empty in sdm_dma_rsp block",
+	{NA, NA, 1, NA, NA, NA} },
+
+/*line 157*/{(0x1F), 1, XSDM_REG_RSP_PXP_CTRL_RDATA_EMPTY,
+	NA, 1, 0, pneq,
+	NA, IDLE_CHK_ERROR_NO_TRAFFIC,
+	"XSDM: pxp_ctrl rd_data fifo is not empty in sdm_dma_rsp block",
+	{NA, NA, 1, NA, NA, NA} },
+
+/*line 158*/{(0x1F), 1, DORQ_REG_DQ_FILL_LVLF,
+	NA, 1, 0, pneq,
+	NA, IDLE_CHK_ERROR_NO_TRAFFIC,
+	"DORQ: DORQ queue is not empty",
+	{NA, NA, 0, NA, NA, NA} },
+
+/*line 159*/{(0x1F), 1, CFC_REG_CFC_INT_STS,
+	NA, 1, 0, pneq,
+	NA, IDLE_CHK_ERROR,
+	"CFC: Interrupt status is not 0",
+	{NA, NA, 0, NA, NA, NA} },
+
+/*line 160*/{(0x1F), 1, CDU_REG_CDU_INT_STS,
+	NA, 1, 0, pneq,
+	NA, IDLE_CHK_ERROR,
+	"CDU: Interrupt status is not 0",
+	{NA, NA, 0, NA, NA, NA} },
+
+/*line 161*/{(0x1F), 1, CCM_REG_CCM_INT_STS,
+	NA, 1, 0, pneq,
+	NA, IDLE_CHK_ERROR,
+	"CCM: Interrupt status is not 0",
+	{NA, NA, 0, NA, NA, NA} },
+
+/*line 162*/{(0x1F), 1, TCM_REG_TCM_INT_STS,
+	NA, 1, 0, pneq,
+	NA, IDLE_CHK_ERROR,
+	"TCM: Interrupt status is not 0",
+	{NA, NA, 0, NA, NA, NA} },
+
+/*line 163*/{(0x1F), 1, UCM_REG_UCM_INT_STS,
+	NA, 1, 0, pneq,
+	NA, IDLE_CHK_ERROR,
+	"UCM: Interrupt status is not 0",
+	{NA, NA, 0, NA, NA, NA} },
+
+/*line 164*/{(0x1F), 1, XCM_REG_XCM_INT_STS,
+	NA, 1, 0, pneq,
+	NA, IDLE_CHK_ERROR,
+	"XCM: Interrupt status is not 0",
+	{NA, NA, 0, NA, NA, NA} },
+
+/*line 165*/{(0xF), 1, PBF_REG_PBF_INT_STS,
+	NA, 1, 0, pneq,
+	NA, IDLE_CHK_ERROR,
+	"PBF: Interrupt status is not 0",
+	{NA, NA, 0, NA, NA, NA} },
+
+/*line 166*/{(0x1F), 1, TM_REG_TM_INT_STS,
+	NA, 1, 0, pneq,
+	NA, IDLE_CHK_ERROR,
+	"TIMERS: Interrupt status is not 0",
+	{NA, NA, 0, NA, NA, NA} },
+
+/*line 167*/{(0x1F), 1, DORQ_REG_DORQ_INT_STS,
+	NA, 1, 0, pneq,
+	NA, IDLE_CHK_ERROR,
+	"DORQ: Interrupt status is not 0",
+	{NA, NA, 0, NA, NA, NA} },
+
+/*line 168*/{(0x1F), 1, SRC_REG_SRC_INT_STS,
+	NA, 1, 0, pneq,
+	NA, IDLE_CHK_ERROR,
+	"SRCH: Interrupt status is not 0",
+	{NA, NA, 0, NA, NA, NA} },
+
+/*line 169*/{(0x1F), 1, PRS_REG_PRS_INT_STS,
+	NA, 1, 0, pneq,
+	NA, IDLE_CHK_ERROR,
+	"PRS: Interrupt status is not 0",
+	{NA, NA, 0, NA, NA, NA} },
+
+/*line 170*/{(0x1F), 1, BRB1_REG_BRB1_INT_STS,
+	NA, 1, 0, pand_neq,
+	NA, IDLE_CHK_ERROR,
+	"BRB1: Interrupt status is not 0",
+	{NA, NA, ~0xFC00, 0, NA, NA} },
+
+/*line 171*/{(0x1F), 1, GRCBASE_XPB + PB_REG_PB_INT_STS,
+	NA, 1, 0, pneq,
+	NA, IDLE_CHK_ERROR,
+	"XPB: Interrupt status is not 0",
+	{NA, NA, 0, NA, NA, NA} },
+
+/*line 172*/{(0x1F), 1, GRCBASE_UPB + PB_REG_PB_INT_STS,
+	NA, 1, 0, pneq,
+	NA, IDLE_CHK_ERROR,
+	"UPB: Interrupt status is not 0",
+	{NA, NA, 0, NA, NA, NA} },
+
+/*line 173*/{(0x1), 1, PXP2_REG_PXP2_INT_STS,
+	NA, 1, 0, pneq,
+	NA, IDLE_CHK_WARNING,
+	"PXP2: Interrupt status 0 is not 0",
+	{NA, NA, 0, NA, NA, NA} },
+
+/*line 174*/{(0x1E), 1, PXP2_REG_PXP2_INT_STS_0,
+	NA, 1, 0, pneq,
+	NA, IDLE_CHK_WARNING,
+	"PXP2: Interrupt status 0 is not 0",
+	{NA, NA, 0, NA, NA, NA} },
+
+/*line 175*/{(0x1E), 1, PXP2_REG_PXP2_INT_STS_1,
+	NA, 1, 0, pneq,
+	NA, IDLE_CHK_WARNING,
+	"PXP2: Interrupt status 1 is not 0",
+	{NA, NA, 0, NA, NA, NA} },
+
+/*line 176*/{(0x1F), 1, QM_REG_QM_INT_STS,
+	NA, 1, 0, pneq,
+	NA, IDLE_CHK_ERROR,
+	"QM: Interrupt status is not 0",
+	{NA, NA, 0, NA, NA, NA} },
+
+/*line 177*/{(0x1F), 1, PXP_REG_PXP_INT_STS_0,
+	NA, 1, 0, pneq,
+	NA, IDLE_CHK_WARNING,
+	"PXP: P0 Interrupt status is not 0",
+	{NA, NA, 0, NA, NA, NA} },
+
+/*line 178*/{(0x1F), 1, PXP_REG_PXP_INT_STS_1,
+	NA, 1, 0, pneq,
+	NA, IDLE_CHK_WARNING,
+	"PXP: P1 Interrupt status is not 0",
+	{NA, NA, 0, NA, NA, NA} },
+
+/*line 179*/{(0x1C), 1, PGLUE_B_REG_PGLUE_B_INT_STS,
+	NA, 1, 0, pneq,
+	NA, IDLE_CHK_WARNING,
+	"PGLUE_B: Interrupt status is not 0",
+	{NA, NA, 0, NA, NA, NA} },
+
+/*line 180*/{(0x1F), 1, DORQ_REG_RSPA_CRD_CNT,
+	NA, 1, 0, pneq,
+	NA, IDLE_CHK_ERROR_NO_TRAFFIC,
+	"DORQ: Credit to XCM is not full",
+	{NA, NA, 2, NA, NA, NA} },
+
+/*line 181*/{(0x1F), 1, DORQ_REG_RSPB_CRD_CNT,
+	NA, 1, 0, pneq,
+	NA, IDLE_CHK_ERROR_NO_TRAFFIC,
+	"DORQ: Credit to UCM is not full",
+	{NA, NA, 2, NA, NA, NA} },
+
+/*line 182*/{(0x3), 1, QM_REG_VOQCRDERRREG,
+	NA, 1, 0, pneq,
+	NA, IDLE_CHK_ERROR,
+	"QM: Credit error register is not 0 (byte or credit overflow/underflow)",
+	{NA, NA, 0, NA, NA, NA} },
+
+/*line 183*/{(0x1F), 1, DORQ_REG_DQ_FULL_ST,
+	NA, 1, 0, pneq,
+	NA, IDLE_CHK_ERROR,
+	"DORQ: DORQ queue is full",
+	{NA, NA, 0, NA, NA, NA} },
+
+/*line 184*/{(0x1F), 1, MISC_REG_AEU_AFTER_INVERT_1_FUNC_0,
+	NA, 1, 0, pand_neq,
+	NA, IDLE_CHK_WARNING,
+	"AEU: P0 AFTER_INVERT_1 is not 0",
+	{NA, NA, ~0xCFFC, 0, NA, NA} },
+
+/*line 185*/{(0x1F), 1, MISC_REG_AEU_AFTER_INVERT_2_FUNC_0,
+	NA, 1, 0, pneq,
+	NA, IDLE_CHK_ERROR,
+	"AEU: P0 AFTER_INVERT_2 is not 0",
+	{NA, NA, 0, NA, NA, NA} },
+
+/*line 186*/{(0x1F), 1, MISC_REG_AEU_AFTER_INVERT_3_FUNC_0,
+	NA, 1, 0, pand_neq,
+	NA, IDLE_CHK_ERROR,
+	"AEU: P0 AFTER_INVERT_3 is not 0",
+	{NA, NA, ~0xFFFF0000, 0, NA, NA} },
+
+/*line 187*/{(0x1F), 1, MISC_REG_AEU_AFTER_INVERT_4_FUNC_0,
+	NA, 1, 0, pand_neq,
+	NA, IDLE_CHK_ERROR,
+	"AEU: P0 AFTER_INVERT_4 is not 0",
+	{NA, NA, ~0x801FFFFF, 0, NA, NA} },
+
+/*line 188*/{(0x3), 1, MISC_REG_AEU_AFTER_INVERT_1_FUNC_1,
+	NA, 1, 0, pand_neq,
+	NA, IDLE_CHK_WARNING,
+	"AEU: P1 AFTER_INVERT_1 is not 0",
+	{NA, NA, ~0xCFFC, 0, NA, NA} },
+
+/*line 189*/{(0x3), 1, MISC_REG_AEU_AFTER_INVERT_2_FUNC_1,
+	NA, 1, 0, pneq,
+	NA, IDLE_CHK_ERROR,
+	"AEU: P1 AFTER_INVERT_2 is not 0",
+	{NA, NA, 0, NA, NA, NA} },
+
+/*line 190*/{(0x3), 1, MISC_REG_AEU_AFTER_INVERT_3_FUNC_1,
+	NA, 1, 0, pand_neq,
+	NA, IDLE_CHK_ERROR,
+	"AEU: P1 AFTER_INVERT_3 is not 0",
+	{NA, NA, ~0xFFFF0000, 0, NA, NA} },
+
+/*line 191*/{(0x3), 1, MISC_REG_AEU_AFTER_INVERT_4_FUNC_1,
+	NA, 1, 0, pand_neq,
+	NA, IDLE_CHK_ERROR,
+	"AEU: P1 AFTER_INVERT_4 is not 0",
+	{NA, NA, ~0x801FFFFF, 0, NA, NA} },
+
+/*line 192*/{(0x1F), 1, MISC_REG_AEU_AFTER_INVERT_1_MCP,
+	NA, 1, 0, pand_neq,
+	NA, IDLE_CHK_WARNING,
+	"AEU: MCP AFTER_INVERT_1 is not 0",
+	{NA, NA, ~0xCFFC, 0, NA, NA} },
+
+/*line 193*/{(0x1F), 1, MISC_REG_AEU_AFTER_INVERT_2_MCP,
+	NA, 1, 0, pneq,
+	NA, IDLE_CHK_ERROR,
+	"AEU: MCP AFTER_INVERT_2 is not 0",
+	{NA, NA, 0, NA, NA, NA} },
+
+/*line 194*/{(0x1F), 1, MISC_REG_AEU_AFTER_INVERT_3_MCP,
+	NA, 1, 0, pand_neq,
+	NA, IDLE_CHK_ERROR,
+	"AEU: MCP AFTER_INVERT_3 is not 0",
+	{NA, NA, ~0xFFFF0000, 0, NA, NA} },
+
+/*line 195*/{(0x1F), 1, MISC_REG_AEU_AFTER_INVERT_4_MCP,
+	NA, 1, 0, pand_neq,
+	NA, IDLE_CHK_ERROR,
+	"AEU: MCP AFTER_INVERT_4 is not 0",
+	{NA, NA, ~0x801FFFFF, 0, NA, NA} },
+
+/*line 196*/{(0xF), 5, PBF_REG_P0_CREDIT,
+	PBF_REG_P0_INIT_CRD, 1, 0, pneq_r2,
+	PBF_REG_DISABLE_NEW_TASK_PROC_P0, IDLE_CHK_ERROR_NO_TRAFFIC,
+	"PBF: P0 credit is not equal to init_crd",
+	{NA, NA, NA, NA, NA, NA} },
+
+/*line 197*/{(0xF), 5, PBF_REG_P1_CREDIT,
+	PBF_REG_P1_INIT_CRD, 1, 0, pneq_r2,
+	PBF_REG_DISABLE_NEW_TASK_PROC_P1, IDLE_CHK_ERROR_NO_TRAFFIC,
+	"PBF: P1 credit is not equal to init_crd",
+	{NA, NA, NA, NA, NA, NA} },
+
+/*line 198*/{(0xF), 3, PBF_REG_P4_CREDIT,
+	PBF_REG_P4_INIT_CRD, 1, 0, pneq_r2,
+	NA, IDLE_CHK_ERROR_NO_TRAFFIC,
+	"PBF: P4 credit is not equal to init_crd",
+	{NA, NA, NA, NA, NA, NA} },
+
+/*line 199*/{(0x10), 5, PBF_REG_CREDIT_Q0,
+	PBF_REG_INIT_CRD_Q0, 1, 0, pneq_r2,
+	PBF_REG_DISABLE_NEW_TASK_PROC_Q0, IDLE_CHK_ERROR_NO_TRAFFIC,
+	"PBF: Q0 credit is not equal to init_crd",
+	{NA, NA, NA, NA, NA, NA} },
+
+/*line 200*/{(0x10), 5, PBF_REG_CREDIT_Q1,
+	PBF_REG_INIT_CRD_Q1, 1, 0, pneq_r2,
+	PBF_REG_DISABLE_NEW_TASK_PROC_Q1, IDLE_CHK_ERROR_NO_TRAFFIC,
+	"PBF: Q1 credit is not equal to init_crd",
+	{NA, NA, NA, NA, NA, NA} },
+
+/*line 201*/{(0x10), 5, PBF_REG_CREDIT_Q2,
+	PBF_REG_INIT_CRD_Q2, 1, 0, pneq_r2,
+	PBF_REG_DISABLE_NEW_TASK_PROC_Q2, IDLE_CHK_ERROR_NO_TRAFFIC,
+	"PBF: Q2 credit is not equal to init_crd",
+	{NA, NA, NA, NA, NA, NA} },
+
+/*line 202*/{(0x10), 5, PBF_REG_CREDIT_Q3,
+	PBF_REG_INIT_CRD_Q3, 1, 0, pneq_r2,
+	PBF_REG_DISABLE_NEW_TASK_PROC_Q3, IDLE_CHK_ERROR_NO_TRAFFIC,
+	"PBF: Q3 credit is not equal to init_crd",
+	{NA, NA, NA, NA, NA, NA} },
+
+/*line 203*/{(0x10), 5, PBF_REG_CREDIT_Q4,
+	PBF_REG_INIT_CRD_Q4, 1, 0, pneq_r2,
+	PBF_REG_DISABLE_NEW_TASK_PROC_Q4, IDLE_CHK_ERROR_NO_TRAFFIC,
+	"PBF: Q4 credit is not equal to init_crd",
+	{NA, NA, NA, NA, NA, NA} },
+
+/*line 204*/{(0x10), 5, PBF_REG_CREDIT_Q5,
+	PBF_REG_INIT_CRD_Q5, 1, 0, pneq_r2,
+	PBF_REG_DISABLE_NEW_TASK_PROC_Q5, IDLE_CHK_ERROR_NO_TRAFFIC,
+	"PBF: Q5 credit is not equal to init_crd",
+	{NA, NA, NA, NA, NA, NA} },
+
+/*line 205*/{(0x10), 3, PBF_REG_CREDIT_LB_Q,
+	PBF_REG_INIT_CRD_LB_Q, 1, 0, pneq_r2,
+	NA, IDLE_CHK_ERROR_NO_TRAFFIC,
+	"PBF: LB Q credit is not equal to init_crd",
+	{NA, NA, NA, NA, NA, NA} },
+
+/*line 206*/{(0xF), 1, PBF_REG_P0_TASK_CNT,
+	NA, 1, 0, pneq,
+	NA, IDLE_CHK_ERROR_NO_TRAFFIC,
+	"PBF: P0 task_cnt is not 0",
+	{NA, NA, 0, NA, NA, NA} },
+
+/*line 207*/{(0xF), 1, PBF_REG_P1_TASK_CNT,
+	NA, 1, 0, pneq,
+	NA, IDLE_CHK_ERROR_NO_TRAFFIC,
+	"PBF: P1 task_cnt is not 0",
+	{NA, NA, 0, NA, NA, NA} },
+
+/*line 208*/{(0xF), 1, PBF_REG_P4_TASK_CNT,
+	NA, 1, 0, pneq,
+	NA, IDLE_CHK_ERROR_NO_TRAFFIC,
+	"PBF: P4 task_cnt is not 0",
+	{NA, NA, 0, NA, NA, NA} },
+
+/*line 209*/{(0x10), 1, PBF_REG_TASK_CNT_Q0,
+	NA, 1, 0, pneq,
+	NA, IDLE_CHK_ERROR_NO_TRAFFIC,
+	"PBF: Q0 task_cnt is not 0",
+	{NA, NA, 0, NA, NA, NA} },
+
+/*line 210*/{(0x10), 1, PBF_REG_TASK_CNT_Q1,
+	NA, 1, 0, pneq,
+	NA, IDLE_CHK_ERROR_NO_TRAFFIC,
+	"PBF: Q1 task_cnt is not 0",
+	{NA, NA, 0, NA, NA, NA} },
+
+/*line 211*/{(0x10), 1, PBF_REG_TASK_CNT_Q2,
+	NA, 1, 0, pneq,
+	NA, IDLE_CHK_ERROR_NO_TRAFFIC,
+	"PBF: Q2 task_cnt is not 0",
+	{NA, NA, 0, NA, NA, NA} },
+
+/*line 212*/{(0x10), 1, PBF_REG_TASK_CNT_Q3,
+	NA, 1, 0, pneq,
+	NA, IDLE_CHK_ERROR_NO_TRAFFIC,
+	"PBF: Q3 task_cnt is not 0",
+	{NA, NA, 0, NA, NA, NA} },
+
+/*line 213*/{(0x10), 1, PBF_REG_TASK_CNT_Q4,
+	NA, 1, 0, pneq,
+	NA, IDLE_CHK_ERROR_NO_TRAFFIC,
+	"PBF: Q4 task_cnt is not 0",
+	{NA, NA, 0, NA, NA, NA} },
+
+/*line 214*/{(0x10), 1, PBF_REG_TASK_CNT_Q5,
+	NA, 1, 0, pneq,
+	NA, IDLE_CHK_ERROR_NO_TRAFFIC,
+	"PBF: Q5 task_cnt is not 0",
+	{NA, NA, 0, NA, NA, NA} },
+
+/*line 215*/{(0x10), 1, PBF_REG_TASK_CNT_LB_Q,
+	NA, 1, 0, pneq,
+	NA, IDLE_CHK_ERROR_NO_TRAFFIC,
+	"PBF: LB Q task_cnt is not 0",
+	{NA, NA, 0, NA, NA, NA} },
+
+/*line 216*/{(0x1F), 1, XCM_REG_CFC_INIT_CRD,
+	NA, 1, 0, pneq,
+	NA, IDLE_CHK_ERROR_NO_TRAFFIC,
+	"XCM: CFC_INIT_CRD is not 1",
+	{NA, NA, 1, NA, NA, NA} },
+
+/*line 217*/{(0x1F), 1, UCM_REG_CFC_INIT_CRD,
+	NA, 1, 0, pneq,
+	NA, IDLE_CHK_ERROR_NO_TRAFFIC,
+	"UCM: CFC_INIT_CRD is not 1",
+	{NA, NA, 1, NA, NA, NA} },
+
+/*line 218*/{(0x1F), 1, TCM_REG_CFC_INIT_CRD,
+	NA, 1, 0, pneq,
+	NA, IDLE_CHK_ERROR_NO_TRAFFIC,
+	"TCM: CFC_INIT_CRD is not 1",
+	{NA, NA, 1, NA, NA, NA} },
+
+/*line 219*/{(0x1F), 1, CCM_REG_CFC_INIT_CRD,
+	NA, 1, 0, pneq,
+	NA, IDLE_CHK_ERROR_NO_TRAFFIC,
+	"CCM: CFC_INIT_CRD is not 1",
+	{NA, NA, 1, NA, NA, NA} },
+
+/*line 220*/{(0x1F), 1, XCM_REG_XQM_INIT_CRD,
+	NA, 1, 0, pneq,
+	NA, IDLE_CHK_ERROR_NO_TRAFFIC,
+	"XCM: XQM_INIT_CRD is not 32",
+	{NA, NA, 32, NA, NA, NA} },
+
+/*line 221*/{(0x1F), 1, UCM_REG_UQM_INIT_CRD,
+	NA, 1, 0, pneq,
+	NA, IDLE_CHK_ERROR_NO_TRAFFIC,
+	"UCM: UQM_INIT_CRD is not 32",
+	{NA, NA, 32, NA, NA, NA} },
+
+/*line 222*/{(0x1F), 1, TCM_REG_TQM_INIT_CRD,
+	NA, 1, 0, pneq,
+	NA, IDLE_CHK_ERROR_NO_TRAFFIC,
+	"TCM: TQM_INIT_CRD is not 32",
+	{NA, NA, 32, NA, NA, NA} },
+
+/*line 223*/{(0x1F), 1, CCM_REG_CQM_INIT_CRD,
+	NA, 1, 0, pneq,
+	NA, IDLE_CHK_ERROR_NO_TRAFFIC,
+	"CCM: CQM_INIT_CRD is not 32",
+	{NA, NA, 32, NA, NA, NA} },
+
+/*line 224*/{(0x1F), 1, XCM_REG_TM_INIT_CRD,
+	NA, 1, 0, pneq,
+	NA, IDLE_CHK_ERROR_NO_TRAFFIC,
+	"XCM: TM_INIT_CRD is not 4",
+	{NA, NA, 4, NA, NA, NA} },
+
+/*line 225*/{(0x1F), 1, UCM_REG_TM_INIT_CRD,
+	NA, 1, 0, pneq,
+	NA, IDLE_CHK_ERROR_NO_TRAFFIC,
+	"UCM: TM_INIT_CRD is not 4",
+	{NA, NA, 4, NA, NA, NA} },
+
+/*line 226*/{(0x1F), 1, XCM_REG_FIC0_INIT_CRD,
+	NA, 1, 0, pneq,
+	NA, IDLE_CHK_WARNING,
+	"XCM: FIC0_INIT_CRD is not 64",
+	{NA, NA, 64, NA, NA, NA} },
+
+/*line 227*/{(0x1F), 1, UCM_REG_FIC0_INIT_CRD,
+	NA, 1, 0, pneq,
+	NA, IDLE_CHK_ERROR_NO_TRAFFIC,
+	"UCM: FIC0_INIT_CRD is not 64",
+	{NA, NA, 64, NA, NA, NA} },
+
+/*line 228*/{(0x1F), 1, TCM_REG_FIC0_INIT_CRD,
+	NA, 1, 0, pneq,
+	NA, IDLE_CHK_ERROR_NO_TRAFFIC,
+	"TCM: FIC0_INIT_CRD is not 64",
+	{NA, NA, 64, NA, NA, NA} },
+
+/*line 229*/{(0x1F), 1, CCM_REG_FIC0_INIT_CRD,
+	NA, 1, 0, pneq,
+	NA, IDLE_CHK_ERROR_NO_TRAFFIC,
+	"CCM: FIC0_INIT_CRD is not 64",
+	{NA, NA, 64, NA, NA, NA} },
+
+/*line 230*/{(0x1F), 1, XCM_REG_FIC1_INIT_CRD,
+	NA, 1, 0, pneq,
+	NA, IDLE_CHK_ERROR_NO_TRAFFIC,
+	"XCM: FIC1_INIT_CRD is not 64",
+	{NA, NA, 64, NA, NA, NA} },
+
+/*line 231*/{(0x1F), 1, UCM_REG_FIC1_INIT_CRD,
+	NA, 1, 0, pneq,
+	NA, IDLE_CHK_ERROR_NO_TRAFFIC,
+	"UCM: FIC1_INIT_CRD is not 64",
+	{NA, NA, 64, NA, NA, NA} },
+
+/*line 232*/{(0x1F), 1, TCM_REG_FIC1_INIT_CRD,
+	NA, 1, 0, pneq,
+	NA, IDLE_CHK_ERROR_NO_TRAFFIC,
+	"TCM: FIC1_INIT_CRD is not 64",
+	{NA, NA, 64, NA, NA, NA} },
+
+/*line 233*/{(0x1F), 1, CCM_REG_FIC1_INIT_CRD,
+	NA, 1, 0, pneq,
+	NA, IDLE_CHK_ERROR_NO_TRAFFIC,
+	"CCM: FIC1_INIT_CRD is not 64",
+	{NA, NA, 64, NA, NA, NA} },
+
+/*line 234*/{(0x1), 1, XCM_REG_XX_FREE,
+	NA, 1, 0, pneq,
+	NA, IDLE_CHK_ERROR_NO_TRAFFIC,
+	"XCM: XX_FREE differs from expected 31",
+	{NA, NA, 31, NA, NA, NA} },
+
+/*line 235*/{(0x1E), 1, XCM_REG_XX_FREE,
+	NA, 1, 0, pneq,
+	NA, IDLE_CHK_ERROR_NO_TRAFFIC,
+	"XCM: XX_FREE differs from expected 32",
+	{NA, NA, 32, NA, NA, NA} },
+
+/*line 236*/{(0x1F), 1, UCM_REG_XX_FREE,
+	NA, 1, 0, pneq,
+	NA, IDLE_CHK_ERROR_NO_TRAFFIC,
+	"UCM: XX_FREE differs from expected 27",
+	{NA, NA, 27, NA, NA, NA} },
+
+/*line 237*/{(0x7), 1, TCM_REG_XX_FREE,
+	NA, 1, 0, pneq,
+	NA, IDLE_CHK_ERROR_NO_TRAFFIC,
+	"TCM: XX_FREE differs from expected 32",
+	{NA, NA, 32, NA, NA, NA} },
+
+/*line 238*/{(0x18), 1, TCM_REG_XX_FREE,
+	NA, 1, 0, pneq,
+	NA, IDLE_CHK_ERROR_NO_TRAFFIC,
+	"TCM: XX_FREE differs from expected 29",
+	{NA, NA, 29, NA, NA, NA} },
+
+/*line 239*/{(0x1F), 1, CCM_REG_XX_FREE,
+	NA, 1, 0, pneq,
+	NA, IDLE_CHK_ERROR_NO_TRAFFIC,
+	"CCM: XX_FREE differs from expected 24",
+	{NA, NA, 24, NA, NA, NA} },
+
+/*line 240*/{(0x1F), 1, XSEM_REG_FAST_MEMORY + 0x18000,
+	NA, 1, 0, pneq,
+	NA, IDLE_CHK_ERROR_NO_TRAFFIC,
+	"XSEM: FOC0 credit less than initial credit",
+	{NA, NA, 0, NA, NA, NA} },
+
+/*line 241*/{(0x1F), 1, XSEM_REG_FAST_MEMORY + 0x18040,
+	NA, 1, 0, pneq,
+	NA, IDLE_CHK_ERROR_NO_TRAFFIC,
+	"XSEM: FOC1 credit less than initial credit",
+	{NA, NA, 24, NA, NA, NA} },
+
+/*line 242*/{(0x1F), 1, XSEM_REG_FAST_MEMORY + 0x18080,
+	NA, 1, 0, pneq,
+	NA, IDLE_CHK_ERROR_NO_TRAFFIC,
+	"XSEM: FOC2 credit less than initial credit",
+	{NA, NA, 12, NA, NA, NA} },
+
+/*line 243*/{(0x1F), 1, USEM_REG_FAST_MEMORY + 0x18000,
+	NA, 1, 0, pneq,
+	NA, IDLE_CHK_ERROR_NO_TRAFFIC,
+	"USEM: FOC0 credit less than initial credit",
+	{NA, NA, 26, NA, NA, NA} },
+
+/*line 244*/{(0x1F), 1, USEM_REG_FAST_MEMORY + 0x18040,
+	NA, 1, 0, pneq,
+	NA, IDLE_CHK_ERROR_NO_TRAFFIC,
+	"USEM: FOC1 credit less than initial credit",
+	{NA, NA, 78, NA, NA, NA} },
+
+/*line 245*/{(0x1F), 1, USEM_REG_FAST_MEMORY + 0x18080,
+	NA, 1, 0, pneq,
+	NA, IDLE_CHK_ERROR_NO_TRAFFIC,
+	"USEM: FOC2 credit less than initial credit",
+	{NA, NA, 16, NA, NA, NA} },
+
+/*line 246*/{(0x1F), 1, USEM_REG_FAST_MEMORY + 0x180C0,
+	NA, 1, 0, pneq,
+	NA, IDLE_CHK_ERROR_NO_TRAFFIC,
+	"USEM: FOC3 credit less than initial credit",
+	{NA, NA, 32, NA, NA, NA} },
+
+/*line 247*/{(0x1F), 1, TSEM_REG_FAST_MEMORY + 0x18000,
+	NA, 1, 0, pneq,
+	NA, IDLE_CHK_ERROR_NO_TRAFFIC,
+	"TSEM: FOC0 credit less than initial credit",
+	{NA, NA, 52, NA, NA, NA} },
+
+/*line 248*/{(0x1F), 1, TSEM_REG_FAST_MEMORY + 0x18040,
+	NA, 1, 0, pneq,
+	NA, IDLE_CHK_ERROR_NO_TRAFFIC,
+	"TSEM: FOC1 credit less than initial credit",
+	{NA, NA, 24, NA, NA, NA} },
+
+/*line 249*/{(0x1F), 1, TSEM_REG_FAST_MEMORY + 0x18080,
+	NA, 1, 0, pneq,
+	NA, IDLE_CHK_ERROR_NO_TRAFFIC,
+	"TSEM: FOC2 credit less than initial credit",
+	{NA, NA, 12, NA, NA, NA} },
+
+/*line 250*/{(0x1F), 1, TSEM_REG_FAST_MEMORY + 0x180C0,
+	NA, 1, 0, pneq,
+	NA, IDLE_CHK_ERROR_NO_TRAFFIC,
+	"TSEM: FOC3 credit less than initial credit",
+	{NA, NA, 32, NA, NA, NA} },
+
+/*line 251*/{(0x1F), 1, CSEM_REG_FAST_MEMORY + 0x18000,
+	NA, 1, 0, pneq,
+	NA, IDLE_CHK_ERROR_NO_TRAFFIC,
+	"CSEM: FOC0 credit less than initial credit",
+	{NA, NA, 16, NA, NA, NA} },
+
+/*line 252*/{(0x1F), 1, CSEM_REG_FAST_MEMORY + 0x18040,
+	NA, 1, 0, pneq,
+	NA, IDLE_CHK_ERROR_NO_TRAFFIC,
+	"CSEM: FOC1 credit less than initial credit",
+	{NA, NA, 18, NA, NA, NA} },
+
+/*line 253*/{(0x1F), 1, CSEM_REG_FAST_MEMORY + 0x18080,
+	NA, 1, 0, pneq,
+	NA, IDLE_CHK_ERROR_NO_TRAFFIC,
+	"CSEM: FOC2 credit less than initial credit",
+	{NA, NA, 48, NA, NA, NA} },
+
+/*line 254*/{(0x1F), 1, CSEM_REG_FAST_MEMORY + 0x180C0,
+	NA, 1, 0, pneq,
+	NA, IDLE_CHK_ERROR_NO_TRAFFIC,
+	"CSEM: FOC3 credit less than initial credit",
+	{NA, NA, 14, NA, NA, NA} },
+
+/*line 255*/{(0x1F), 1, PRS_REG_TSDM_CURRENT_CREDIT,
+	NA, 1, 0, pneq,
+	NA, IDLE_CHK_ERROR_NO_TRAFFIC,
+	"PRS: TSDM current credit is not 0",
+	{NA, NA, 0, NA, NA, NA} },
+
+/*line 256*/{(0x1F), 1, PRS_REG_TCM_CURRENT_CREDIT,
+	NA, 1, 0, pneq,
+	NA, IDLE_CHK_ERROR_NO_TRAFFIC,
+	"PRS: TCM current credit is not 0",
+	{NA, NA, 0, NA, NA, NA} },
+
+/*line 257*/{(0x1F), 1, PRS_REG_CFC_LD_CURRENT_CREDIT,
+	NA, 1, 0, pneq,
+	NA, IDLE_CHK_ERROR_NO_TRAFFIC,
+	"PRS: CFC_LD current credit is not 0",
+	{NA, NA, 0, NA, NA, NA} },
+
+/*line 258*/{(0x1F), 1, PRS_REG_CFC_SEARCH_CURRENT_CREDIT,
+	NA, 1, 0, pneq,
+	NA, IDLE_CHK_ERROR_NO_TRAFFIC,
+	"PRS: CFC_SEARCH current credit is not 0",
+	{NA, NA, 0, NA, NA, NA} },
+
+/*line 259*/{(0x1F), 1, PRS_REG_SRC_CURRENT_CREDIT,
+	NA, 1, 0, pneq,
+	NA, IDLE_CHK_ERROR_NO_TRAFFIC,
+	"PRS: SRCH current credit is not 0",
+	{NA, NA, 0, NA, NA, NA} },
+
+/*line 260*/{(0x1F), 1, PRS_REG_PENDING_BRB_PRS_RQ,
+	NA, 1, 0, pneq,
+	NA, IDLE_CHK_ERROR_NO_TRAFFIC,
+	"PRS: PENDING_BRB_PRS_RQ is not 0",
+	{NA, NA, 0, NA, NA, NA} },
+
+/*line 261*/{(0x1F), 2, PRS_REG_PENDING_BRB_CAC0_RQ,
+	NA, 5, 4, pneq,
+	NA, IDLE_CHK_ERROR_NO_TRAFFIC,
+	"PRS: PENDING_BRB_CAC_RQ is not 0",
+	{NA, NA, 0, NA, NA, NA} },
+
+/*line 262*/{(0x1F), 1, PRS_REG_SERIAL_NUM_STATUS_LSB,
+	NA, 1, 0, pneq,
+	NA, IDLE_CHK_ERROR_NO_TRAFFIC,
+	"PRS: SERIAL_NUM_STATUS_LSB is not 0",
+	{NA, NA, 0, NA, NA, NA} },
+
+/*line 263*/{(0x1F), 1, PRS_REG_SERIAL_NUM_STATUS_MSB,
+	NA, 1, 0, pneq,
+	NA, IDLE_CHK_ERROR_NO_TRAFFIC,
+	"PRS: SERIAL_NUM_STATUS_MSB is not 0",
+	{NA, NA, 0, NA, NA, NA} },
+
+/*line 264*/{(0x1F), 1, CDU_REG_ERROR_DATA,
+	NA, 1, 0, pneq,
+	NA, IDLE_CHK_ERROR,
+	"CDU: ERROR_DATA is not 0",
+	{NA, NA, 0, NA, NA, NA} },
+
+/*line 265*/{(0x1F), 1, CCM_REG_STORM_LENGTH_MIS,
+	NA, 1, 0, pneq,
+	NA, IDLE_CHK_ERROR,
+	"CCM: STORM declared message length unequal to actual",
+	{NA, NA, 0, NA, NA, NA} },
+
+/*line 266*/{(0x1F), 1, CCM_REG_CSDM_LENGTH_MIS,
+	NA, 1, 0, pneq,
+	NA, IDLE_CHK_ERROR,
+	"CCM: CSDM declared message length unequal to actual",
+	{NA, NA, 0, NA, NA, NA} },
+
+/*line 267*/{(0x1F), 1, CCM_REG_TSEM_LENGTH_MIS,
+	NA, 1, 0, pneq,
+	NA, IDLE_CHK_ERROR,
+	"CCM: TSEM declared message length unequal to actual",
+	{NA, NA, 0, NA, NA, NA} },
+
+/*line 268*/{(0x1F), 1, CCM_REG_XSEM_LENGTH_MIS,
+	NA, 1, 0, pneq,
+	NA, IDLE_CHK_ERROR,
+	"CCM: XSEM declared message length unequal to actual",
+	{NA, NA, 0, NA, NA, NA} },
+
+/*line 269*/{(0x1F), 1, CCM_REG_USEM_LENGTH_MIS,
+	NA, 1, 0, pneq,
+	NA, IDLE_CHK_ERROR,
+	"CCM: USEM declared message length unequal to actual",
+	{NA, NA, 0, NA, NA, NA} },
+
+/*line 270*/{(0x1F), 1, CCM_REG_PBF_LENGTH_MIS,
+	NA, 1, 0, pneq,
+	NA, IDLE_CHK_ERROR,
+	"CCM: PBF declared message length unequal to actual",
+	{NA, NA, 0, NA, NA, NA} },
+
+/*line 271*/{(0x1F), 1, TCM_REG_STORM_LENGTH_MIS,
+	NA, 1, 0, pneq,
+	NA, IDLE_CHK_ERROR,
+	"TCM: STORM declared message length unequal to actual",
+	{NA, NA, 0, NA, NA, NA} },
+
+/*line 272*/{(0x1F), 1, TCM_REG_TSDM_LENGTH_MIS,
+	NA, 1, 0, pneq,
+	NA, IDLE_CHK_ERROR,
+	"TCM: TSDM declared message length unequal to actual",
+	{NA, NA, 0, NA, NA, NA} },
+
+/*line 273*/{(0x1F), 1, TCM_REG_PRS_LENGTH_MIS,
+	NA, 1, 0, pneq,
+	NA, IDLE_CHK_ERROR,
+	"TCM: PRS declared message length unequal to actual",
+	{NA, NA, 0, NA, NA, NA} },
+
+/*line 274*/{(0x1F), 1, TCM_REG_PBF_LENGTH_MIS,
+	NA, 1, 0, pneq,
+	NA, IDLE_CHK_ERROR,
+	"TCM: PBF declared message length unequal to actual",
+	{NA, NA, 0, NA, NA, NA} },
+
+/*line 275*/{(0x1F), 1, TCM_REG_USEM_LENGTH_MIS,
+	NA, 1, 0, pneq,
+	NA, IDLE_CHK_ERROR,
+	"TCM: USEM declared message length unequal to actual",
+	{NA, NA, 0, NA, NA, NA} },
+
+/*line 276*/{(0x1F), 1, TCM_REG_CSEM_LENGTH_MIS,
+	NA, 1, 0, pneq,
+	NA, IDLE_CHK_ERROR,
+	"TCM: CSEM declared message length unequal to actual",
+	{NA, NA, 0, NA, NA, NA} },
+
+/*line 277*/{(0x1F), 1, UCM_REG_STORM_LENGTH_MIS,
+	NA, 1, 0, pneq,
+	NA, IDLE_CHK_ERROR,
+	"UCM: STORM declared message length unequal to actual",
+	{NA, NA, 0, NA, NA, NA} },
+
+/*line 278*/{(0x1F), 1, UCM_REG_USDM_LENGTH_MIS,
+	NA, 1, 0, pneq,
+	NA, IDLE_CHK_ERROR,
+	"UCM: USDM declared message length unequal to actual",
+	{NA, NA, 0, NA, NA, NA} },
+
+/*line 279*/{(0x1F), 1, UCM_REG_TSEM_LENGTH_MIS,
+	NA, 1, 0, pneq,
+	NA, IDLE_CHK_ERROR,
+	"UCM: TSEM declared message length unequal to actual",
+	{NA, NA, 0, NA, NA, NA} },
+
+/*line 280*/{(0x1F), 1, UCM_REG_CSEM_LENGTH_MIS,
+	NA, 1, 0, pneq,
+	NA, IDLE_CHK_ERROR,
+	"UCM: CSEM declared message length unequal to actual",
+	{NA, NA, 0, NA, NA, NA} },
+
+/*line 281*/{(0x1F), 1, UCM_REG_XSEM_LENGTH_MIS,
+	NA, 1, 0, pneq,
+	NA, IDLE_CHK_ERROR,
+	"UCM: XSEM declared message length unequal to actual",
+	{NA, NA, 0, NA, NA, NA} },
+
+/*line 282*/{(0x1F), 1, UCM_REG_DORQ_LENGTH_MIS,
+	NA, 1, 0, pneq,
+	NA, IDLE_CHK_ERROR,
+	"UCM: DORQ declared message length unequal to actual",
+	{NA, NA, 0, NA, NA, NA} },
+
+/*line 283*/{(0x1F), 1, XCM_REG_STORM_LENGTH_MIS,
+	NA, 1, 0, pneq,
+	NA, IDLE_CHK_ERROR,
+	"XCM: STORM declared message length unequal to actual",
+	{NA, NA, 0, NA, NA, NA} },
+
+/*line 284*/{(0x1F), 1, XCM_REG_XSDM_LENGTH_MIS,
+	NA, 1, 0, pneq,
+	NA, IDLE_CHK_ERROR,
+	"XCM: XSDM declared message length unequal to actual",
+	{NA, NA, 0, NA, NA, NA} },
+
+/*line 285*/{(0x1F), 1, XCM_REG_TSEM_LENGTH_MIS,
+	NA, 1, 0, pneq,
+	NA, IDLE_CHK_ERROR,
+	"XCM: TSEM declared message length unequal to actual",
+	{NA, NA, 0, NA, NA, NA} },
+
+/*line 286*/{(0x1F), 1, XCM_REG_CSEM_LENGTH_MIS,
+	NA, 1, 0, pneq,
+	NA, IDLE_CHK_ERROR,
+	"XCM: CSEM declared message length unequal to actual",
+	{NA, NA, 0, NA, NA, NA} },
+
+/*line 287*/{(0x1F), 1, XCM_REG_USEM_LENGTH_MIS,
+	NA, 1, 0, pneq,
+	NA, IDLE_CHK_ERROR,
+	"XCM: USEM declared message length unequal to actual",
+	{NA, NA, 0, NA, NA, NA} },
+
+/*line 288*/{(0x1F), 1, XCM_REG_DORQ_LENGTH_MIS,
+	NA, 1, 0, pneq,
+	NA, IDLE_CHK_ERROR,
+	"XCM: DORQ declared message length unequal to actual",
+	{NA, NA, 0, NA, NA, NA} },
+
+/*line 289*/{(0x1F), 1, XCM_REG_PBF_LENGTH_MIS,
+	NA, 1, 0, pneq,
+	NA, IDLE_CHK_ERROR,
+	"XCM: PBF declared message length unequal to actual",
+	{NA, NA, 0, NA, NA, NA} },
+
+/*line 290*/{(0x1F), 1, XCM_REG_NIG0_LENGTH_MIS,
+	NA, 1, 0, pneq,
+	NA, IDLE_CHK_ERROR,
+	"XCM: NIG0 declared message length unequal to actual",
+	{NA, NA, 0, NA, NA, NA} },
+
+/*line 291*/{(0x1F), 1, XCM_REG_NIG1_LENGTH_MIS,
+	NA, 1, 0, pneq,
+	NA, IDLE_CHK_ERROR,
+	"XCM: NIG1 declared message length unequal to actual",
+	{NA, NA, 0, NA, NA, NA} },
+
+/*line 292*/{(0x1F), 1, QM_REG_XQM_WRC_FIFOLVL,
+	NA, 1, 0, pneq,
+	NA, IDLE_CHK_ERROR_NO_TRAFFIC,
+	"QM: XQM wrc_fifolvl is not 0",
+	{NA, NA, 0, NA, NA, NA} },
+
+/*line 293*/{(0x1F), 1, QM_REG_UQM_WRC_FIFOLVL,
+	NA, 1, 0, pneq,
+	NA, IDLE_CHK_ERROR_NO_TRAFFIC,
+	"QM: UQM wrc_fifolvl is not 0",
+	{NA, NA, 0, NA, NA, NA} },
+
+/*line 294*/{(0x1F), 1, QM_REG_TQM_WRC_FIFOLVL,
+	NA, 1, 0, pneq,
+	NA, IDLE_CHK_ERROR_NO_TRAFFIC,
+	"QM: TQM wrc_fifolvl is not 0",
+	{NA, NA, 0, NA, NA, NA} },
+
+/*line 295*/{(0x1F), 1, QM_REG_CQM_WRC_FIFOLVL,
+	NA, 1, 0, pneq,
+	NA, IDLE_CHK_ERROR_NO_TRAFFIC,
+	"QM: CQM wrc_fifolvl is not 0",
+	{NA, NA, 0, NA, NA, NA} },
+
+/*line 296*/{(0x1F), 1, QM_REG_QSTATUS_LOW,
+	NA, 1, 0, pneq,
+	NA, IDLE_CHK_ERROR_NO_TRAFFIC,
+	"QM: QSTATUS_LOW is not 0",
+	{NA, NA, 0, NA, NA, NA} },
+
+/*line 297*/{(0x1F), 1, QM_REG_QSTATUS_HIGH,
+	NA, 1, 0, pneq,
+	NA, IDLE_CHK_ERROR_NO_TRAFFIC,
+	"QM: QSTATUS_HIGH is not 0",
+	{NA, NA, 0, NA, NA, NA} },
+
+/*line 298*/{(0x1F), 1, QM_REG_PAUSESTATE0,
+	NA, 1, 0, pneq,
+	NA, IDLE_CHK_ERROR_NO_TRAFFIC,
+	"QM: PAUSESTATE0 is not 0",
+	{NA, NA, 0, NA, NA, NA} },
+
+/*line 299*/{(0x1F), 1, QM_REG_PAUSESTATE1,
+	NA, 1, 0, pneq,
+	NA, IDLE_CHK_ERROR_NO_TRAFFIC,
+	"QM: PAUSESTATE1 is not 0",
+	{NA, NA, 0, NA, NA, NA} },
+
+/*line 300*/{(0x1F), 1, QM_REG_OVFQNUM,
+	NA, 1, 0, pneq,
+	NA, IDLE_CHK_ERROR,
+	"QM: OVFQNUM is not 0",
+	{NA, NA, 0, NA, NA, NA} },
+
+/*line 301*/{(0x1F), 1, QM_REG_OVFERROR,
+	NA, 1, 0, pneq,
+	NA, IDLE_CHK_ERROR,
+	"QM: OVFERROR is not 0",
+	{NA, NA, 0, NA, NA, NA} },
+
+/*line 302*/{(0x1F), 6, QM_REG_PTRTBL,
+	NA, 64, 8, pneq_r2,
+	NA, IDLE_CHK_ERROR_NO_TRAFFIC,
+	"QM: read and write variables not equal",
+	{NA, NA, NA, NA, NA, NA} },
+
+/*line 303*/{(0x1F), 1, BRB1_REG_BRB1_PRTY_STS,
+	NA, 1, 0, pand_neq,
+	NA, IDLE_CHK_WARNING,
+	"BRB1: parity status is not 0",
+	{NA, NA, ~0x8, 0, NA, NA} },
+
+/*line 304*/{(0x1F), 1, CDU_REG_CDU_PRTY_STS,
+	NA, 1, 0, pneq,
+	NA, IDLE_CHK_WARNING,
+	"CDU: parity status is not 0",
+	{NA, NA, 0, NA, NA, NA} },
+
+/*line 305*/{(0x1F), 1, CFC_REG_CFC_PRTY_STS,
+	NA, 1, 0, pand_neq,
+	NA, IDLE_CHK_WARNING,
+	"CFC: parity status is not 0",
+	{NA, NA, ~0x2, 0, NA, NA} },
+
+/*line 306*/{(0x1F), 1, CSDM_REG_CSDM_PRTY_STS,
+	NA, 1, 0, pneq,
+	NA, IDLE_CHK_WARNING,
+	"CSDM: parity status is not 0",
+	{NA, NA, 0, NA, NA, NA} },
+
+/*line 307*/{(0x3), 1, DBG_REG_DBG_PRTY_STS,
+	NA, 1, 0, pneq,
+	NA, IDLE_CHK_WARNING,
+	"DBG: parity status is not 0",
+	{NA, NA, 0, NA, NA, NA} },
+
+/*line 308*/{(0x1F), 1, DMAE_REG_DMAE_PRTY_STS,
+	NA, 1, 0, pneq,
+	NA, IDLE_CHK_WARNING,
+	"DMAE: parity status is not 0",
+	{NA, NA, 0, NA, NA, NA} },
+
+/*line 309*/{(0x1F), 1, DORQ_REG_DORQ_PRTY_STS,
+	NA, 1, 0, pneq,
+	NA, IDLE_CHK_WARNING,
+	"DORQ: parity status is not 0",
+	{NA, NA, 0, NA, NA, NA} },
+
+/*line 310*/{(0x1), 1, TCM_REG_TCM_PRTY_STS,
+	NA, 1, 0, pand_neq,
+	NA, IDLE_CHK_WARNING,
+	"TCM: parity status is not 0",
+	{NA, NA, ~0x3ffc0, 0, NA, NA} },
+
+/*line 311*/{(0x1E), 1, TCM_REG_TCM_PRTY_STS,
+	NA, 1, 0, pneq,
+	NA, IDLE_CHK_WARNING,
+	"TCM: parity status is not 0",
+	{NA, NA, 0, NA, NA, NA} },
+
+/*line 312*/{(0x1), 1, CCM_REG_CCM_PRTY_STS,
+	NA, 1, 0, pand_neq,
+	NA, IDLE_CHK_WARNING,
+	"CCM: parity status is not 0",
+	{NA, NA, ~0x3ffc0, 0, NA, NA} },
+
+/*line 313*/{(0x1E), 1, CCM_REG_CCM_PRTY_STS,
+	NA, 1, 0, pneq,
+	NA, IDLE_CHK_WARNING,
+	"CCM: parity status is not 0",
+	{NA, NA, 0, NA, NA, NA} },
+
+/*line 314*/{(0x1), 1, UCM_REG_UCM_PRTY_STS,
+	NA, 1, 0, pand_neq,
+	NA, IDLE_CHK_WARNING,
+	"UCM: parity status is not 0",
+	{NA, NA, ~0x3ffc0, 0, NA, NA} },
+
+/*line 315*/{(0x1E), 1, UCM_REG_UCM_PRTY_STS,
+	NA, 1, 0, pneq,
+	NA, IDLE_CHK_WARNING,
+	"UCM: parity status is not 0",
+	{NA, NA, 0, NA, NA, NA} },
+
+/*line 316*/{(0x1), 1, XCM_REG_XCM_PRTY_STS,
+	NA, 1, 0, pand_neq,
+	NA, IDLE_CHK_WARNING,
+	"XCM: parity status is not 0",
+	{NA, NA, ~0x3ffc0, 0, NA, NA} },
+
+/*line 317*/{(0x1E), 1, XCM_REG_XCM_PRTY_STS,
+	NA, 1, 0, pneq,
+	NA, IDLE_CHK_WARNING,
+	"XCM: parity status is not 0",
+	{NA, NA, 0, NA, NA, NA} },
+
+/*line 318*/{(0x1), 1, HC_REG_HC_PRTY_STS,
+	NA, 1, 0, pand_neq,
+	NA, IDLE_CHK_WARNING,
+	"HC: parity status is not 0",
+	{NA, NA, ~0x1, 0, NA, NA} },
+
+/*line 319*/{(0x1), 1, MISC_REG_MISC_PRTY_STS,
+	NA, 1, 0, pand_neq,
+	NA, IDLE_CHK_WARNING,
+	"MISC: parity status is not 0",
+	{NA, NA, ~0x1, 0, NA, NA} },
+
+/*line 320*/{(0x1F), 1, PRS_REG_PRS_PRTY_STS,
+	NA, 1, 0, pneq,
+	NA, IDLE_CHK_WARNING,
+	"PRS: parity status is not 0",
+	{NA, NA, 0, NA, NA, NA} },
+
+/*line 321*/{(0x1F), 1, PXP_REG_PXP_PRTY_STS,
+	NA, 1, 0, pneq,
+	NA, IDLE_CHK_WARNING,
+	"PXP: parity status is not 0",
+	{NA, NA, 0, NA, NA, NA} },
+
+/*line 322*/{(0x1F), 1, QM_REG_QM_PRTY_STS,
+	NA, 1, 0, pneq,
+	NA, IDLE_CHK_WARNING,
+	"QM: parity status is not 0",
+	{NA, NA, 0, NA, NA, NA} },
+
+/*line 323*/{(0x1), 1, SRC_REG_SRC_PRTY_STS,
+	NA, 1, 0, pand_neq,
+	NA, IDLE_CHK_WARNING,
+	"SRCH: parity status is not 0",
+	{NA, NA, ~0x4, 0, NA, NA} },
+
+/*line 324*/{(0x1F), 1, TSDM_REG_TSDM_PRTY_STS,
+	NA, 1, 0, pneq,
+	NA, IDLE_CHK_WARNING,
+	"TSDM: parity status is not 0",
+	{NA, NA, 0, NA, NA, NA} },
+
+/*line 325*/{(0x1F), 1, USDM_REG_USDM_PRTY_STS,
+	NA, 1, 0, pand_neq,
+	NA, IDLE_CHK_WARNING,
+	"USDM: parity status is not 0",
+	{NA, NA, ~0x20, 0, NA, NA} },
+
+/*line 326*/{(0x1F), 1, XSDM_REG_XSDM_PRTY_STS,
+	NA, 1, 0, pneq,
+	NA, IDLE_CHK_WARNING,
+	"XSDM: parity status is not 0",
+	{NA, NA, 0, NA, NA, NA} },
+
+/*line 327*/{(0x1F), 1, GRCBASE_XPB + PB_REG_PB_PRTY_STS,
+	NA, 1, 0, pneq,
+	NA, IDLE_CHK_WARNING,
+	"XPB: parity status is not 0",
+	{NA, NA, 0, NA, NA, NA} },
+
+/*line 328*/{(0x1F), 1, GRCBASE_UPB + PB_REG_PB_PRTY_STS,
+	NA, 1, 0, pneq,
+	NA, IDLE_CHK_WARNING,
+	"UPB: parity status is not 0",
+	{NA, NA, 0, NA, NA, NA} },
+
+/*line 329*/{(0x1F), 1, CSEM_REG_CSEM_PRTY_STS_0,
+	NA, 1, 0, pneq,
+	NA, IDLE_CHK_WARNING,
+	"CSEM: parity status 0 is not 0",
+	{NA, NA, 0, NA, NA, NA} },
+
+/*line 330*/{(0x1), 1, PXP2_REG_PXP2_PRTY_STS_0,
+	NA, 1, 0, pand_neq,
+	NA, IDLE_CHK_WARNING,
+	"PXP2: parity status 0 is not 0",
+	{NA, NA, ~0xfff40020, 0, NA, NA} },
+
+/*line 331*/{(0x1E), 1, PXP2_REG_PXP2_PRTY_STS_0,
+	NA, 1, 0, pand_neq,
+	NA, IDLE_CHK_WARNING,
+	"PXP2: parity status 0 is not 0",
+	{NA, NA, ~0x20, 0, NA, NA} },
+
+/*line 332*/{(0x1F), 1, TSEM_REG_TSEM_PRTY_STS_0,
+	NA, 1, 0, pneq,
+	NA, IDLE_CHK_WARNING,
+	"TSEM: parity status 0 is not 0",
+	{NA, NA, 0, NA, NA, NA} },
+
+/*line 333*/{(0x1F), 1, USEM_REG_USEM_PRTY_STS_0,
+	NA, 1, 0, pneq,
+	NA, IDLE_CHK_WARNING,
+	"USEM: parity status 0 is not 0",
+	{NA, NA, 0, NA, NA, NA} },
+
+/*line 334*/{(0x1F), 1, XSEM_REG_XSEM_PRTY_STS_0,
+	NA, 1, 0, pneq,
+	NA, IDLE_CHK_WARNING,
+	"XSEM: parity status 0 is not 0",
+	{NA, NA, 0, NA, NA, NA} },
+
+/*line 335*/{(0x1F), 1, CSEM_REG_CSEM_PRTY_STS_1,
+	NA, 1, 0, pneq,
+	NA, IDLE_CHK_WARNING,
+	"CSEM: parity status 1 is not 0",
+	{NA, NA, 0, NA, NA, NA} },
+
+/*line 336*/{(0x1), 1, PXP2_REG_PXP2_PRTY_STS_1,
+	NA, 1, 0, pand_neq,
+	NA, IDLE_CHK_WARNING,
+	"PXP2: parity status 1 is not 0",
+	{NA, NA, ~0x20, 0, NA, NA} },
+
+/*line 337*/{(0x1E), 1, PXP2_REG_PXP2_PRTY_STS_1,
+	NA, 1, 0, pneq,
+	NA, IDLE_CHK_WARNING,
+	"PXP2: parity status 1 is not 0",
+	{NA, NA, 0, NA, NA, NA} },
+
+/*line 338*/{(0x1F), 1, TSEM_REG_TSEM_PRTY_STS_1,
+	NA, 1, 0, pneq,
+	NA, IDLE_CHK_WARNING,
+	"TSEM: parity status 1 is not 0",
+	{NA, NA, 0, NA, NA, NA} },
+
+/*line 339*/{(0x1F), 1, USEM_REG_USEM_PRTY_STS_1,
+	NA, 1, 0, pneq,
+	NA, IDLE_CHK_WARNING,
+	"USEM: parity status 1 is not 0",
+	{NA, NA, 0, NA, NA, NA} },
+
+/*line 340*/{(0x1F), 1, XSEM_REG_XSEM_PRTY_STS_1,
+	NA, 1, 0, pneq,
+	NA, IDLE_CHK_WARNING,
+	"XSEM: parity status 1 is not 0",
+	{NA, NA, 0, NA, NA, NA} },
+
+/*line 341*/{(0x1C), 1, PGLUE_B_REG_PGLUE_B_PRTY_STS,
+	NA, 1, 0, pneq,
+	NA, IDLE_CHK_WARNING,
+	"PGLUE_B: parity status is not 0",
+	{NA, NA, 0, NA, NA, NA} },
+
+/*line 342*/{(0x2), 2, QM_REG_QTASKCTR_EXT_A_0,
+	NA, 64, 4, pneq,
+	NA, IDLE_CHK_ERROR_NO_TRAFFIC,
+	"QM: Q_EXT_A (upper 64 queues), Queue is not empty",
+	{NA, NA, 0, NA, NA, NA} },
+
+/*line 343*/{(0x2), 1, QM_REG_QSTATUS_LOW_EXT_A,
+	NA, 1, 0, pneq,
+	NA, IDLE_CHK_ERROR,
+	"QM: QSTATUS_LOW_EXT_A is not 0",
+	{NA, NA, 0, NA, NA, NA} },
+
+/*line 344*/{(0x2), 1, QM_REG_QSTATUS_HIGH_EXT_A,
+	NA, 1, 0, pneq,
+	NA, IDLE_CHK_ERROR,
+	"QM: QSTATUS_HIGH_EXT_A is not 0",
+	{NA, NA, 0, NA, NA, NA} },
+
+/*line 345*/{(0x1E), 1, QM_REG_PAUSESTATE2,
+	NA, 1, 0, pneq,
+	NA, IDLE_CHK_ERROR_NO_TRAFFIC,
+	"QM: PAUSESTATE2 is not 0",
+	{NA, NA, 0, NA, NA, NA} },
+
+/*line 346*/{(0x1E), 1, QM_REG_PAUSESTATE3,
+	NA, 1, 0, pneq,
+	NA, IDLE_CHK_ERROR_NO_TRAFFIC,
+	"QM: PAUSESTATE3 is not 0",
+	{NA, NA, 0, NA, NA, NA} },
+
+/*line 347*/{(0x2), 1, QM_REG_PAUSESTATE4,
+	NA, 1, 0, pneq,
+	NA, IDLE_CHK_ERROR,
+	"QM: PAUSESTATE4 is not 0",
+	{NA, NA, 0, NA, NA, NA} },
+
+/*line 348*/{(0x2), 1, QM_REG_PAUSESTATE5,
+	NA, 1, 0, pneq,
+	NA, IDLE_CHK_ERROR,
+	"QM: PAUSESTATE5 is not 0",
+	{NA, NA, 0, NA, NA, NA} },
+
+/*line 349*/{(0x2), 1, QM_REG_PAUSESTATE6,
+	NA, 1, 0, pneq,
+	NA, IDLE_CHK_ERROR,
+	"QM: PAUSESTATE6 is not 0",
+	{NA, NA, 0, NA, NA, NA} },
+
+/*line 350*/{(0x2), 1, QM_REG_PAUSESTATE7,
+	NA, 1, 0, pneq,
+	NA, IDLE_CHK_ERROR,
+	"QM: PAUSESTATE7 is not 0",
+	{NA, NA, 0, NA, NA, NA} },
+
+/*line 351*/{(0x2), 6, QM_REG_PTRTBL_EXT_A,
+	NA, 64, 8, pneq_r2,
+	NA, IDLE_CHK_ERROR_NO_TRAFFIC,
+	"QM: read and write variables not equal in ext table",
+	{NA, NA, NA, NA, NA, NA} },
+
+/*line 352*/{(0x1E), 1, MISC_REG_AEU_SYS_KILL_OCCURRED,
+	NA, NA, NA, pneq,
+	NA, IDLE_CHK_ERROR,
+	"MISC: system kill occurred;",
+	{NA, NA, 0, NA, NA, NA} },
+
+/*line 353*/{(0x1E), 1, MISC_REG_AEU_SYS_KILL_STATUS_0,
+	NA, NA, NA, pneq,
+	NA, IDLE_CHK_ERROR,
+	"MISC: system kill occurred; status_0 register",
+	{NA, NA, 0, NA, NA, NA} },
+
+/*line 354*/{(0x1E), 1, MISC_REG_AEU_SYS_KILL_STATUS_1,
+	NA, NA, NA, pneq,
+	NA, IDLE_CHK_ERROR,
+	"MISC: system kill occurred; status_1 register",
+	{NA, NA, 0, NA, NA, NA} },
+
+/*line 355*/{(0x1E), 1, MISC_REG_AEU_SYS_KILL_STATUS_2,
+	NA, NA, NA, pneq,
+	NA, IDLE_CHK_ERROR,
+	"MISC: system kill occurred; status_2 register",
+	{NA, NA, 0, NA, NA, NA} },
+
+/*line 356*/{(0x1E), 1, MISC_REG_AEU_SYS_KILL_STATUS_3,
+	NA, NA, NA, pneq,
+	NA, IDLE_CHK_ERROR,
+	"MISC: system kill occurred; status_3 register",
+	{NA, NA, 0, NA, NA, NA} },
+
+/*line 357*/{(0x1E), 1, MISC_REG_PCIE_HOT_RESET,
+	NA, NA, NA, pneq,
+	NA, IDLE_CHK_WARNING,
+	"MISC: pcie_rst_b was asserted without perst assertion",
+	{NA, NA, 0, NA, NA, NA} },
+
+/*line 358*/{(0x1F), 1, NIG_REG_NIG_INT_STS_0,
+	NA, NA, NA, pand_neq,
+	NA, IDLE_CHK_ERROR,
+	"NIG: interrupt 0 is active",
+	{NA, NA, ~0x300, 0, NA, NA} },
+
+/*line 359*/{(0x1F), 1, NIG_REG_NIG_INT_STS_0,
+	NA, NA, NA, peq,
+	NA, IDLE_CHK_WARNING,
+	"NIG: Access to BMAC while not active. If tested on FPGA, ignore this warning",
+	{NA, NA, 0x300, NA, NA, NA} },
+
+/*line 360*/{(0x1F), 1, NIG_REG_NIG_INT_STS_1,
+	NA, NA, NA, pand_neq,
+	NA, IDLE_CHK_ERROR,
+	"NIG: interrupt 1 is active",
+	{NA, NA, 0x783FF03, 0, NA, NA} },
+
+/*line 361*/{(0x1F), 1, NIG_REG_NIG_INT_STS_1,
+	NA, NA, NA, pand_neq,
+	NA, IDLE_CHK_WARNING,
+	"NIG: port cos was paused too long",
+	{NA, NA, ~0x783FF0F, 0, NA, NA} },
+
+/*line 362*/{(0x1F), 1, NIG_REG_NIG_INT_STS_1,
+	NA, NA, NA, pand_neq,
+	NA, IDLE_CHK_WARNING,
+	"NIG: Got packets w/o Outer-VLAN in MF mode",
+	{NA, NA, 0xC, 0, NA, NA} },
+
+/*line 363*/{(0x2), 1, NIG_REG_NIG_PRTY_STS,
+	NA, NA, NA, pand_neq,
+	NA, IDLE_CHK_ERROR,
+	"NIG: parity interrupt is active",
+	{NA, NA, ~0xFFC00000, 0, NA, NA} },
+
+/*line 364*/{(0x1C), 1, NIG_REG_NIG_PRTY_STS_0,
+	NA, NA, NA, pand_neq,
+	NA, IDLE_CHK_ERROR,
+	"NIG: parity 0 interrupt is active",
+	{NA, NA, ~0xFFC00000, 0, NA, NA} },
+
+/*line 365*/{(0x4), 1, NIG_REG_NIG_PRTY_STS_1,
+	NA, NA, NA, pand_neq,
+	NA, IDLE_CHK_ERROR,
+	"NIG: parity 1 interrupt is active",
+	{NA, NA, 0xff, 0, NA, NA} },
+
+/*line 366*/{(0x18), 1, NIG_REG_NIG_PRTY_STS_1,
+	NA, NA, NA, pneq,
+	NA, IDLE_CHK_ERROR,
+	"NIG: parity 1 interrupt is active",
+	{NA, NA, 0, NA, NA, NA} },
+
+/*line 367*/{(0x1F), 1, TSEM_REG_TSEM_INT_STS_0,
+	NA, NA, NA, pand_neq,
+	NA, IDLE_CHK_WARNING,
+	"TSEM: interrupt 0 is active",
+	{NA, NA, ~0x10000000, 0, NA, NA} },
+
+/*line 368*/{(0x1F), 1, TSEM_REG_TSEM_INT_STS_0,
+	NA, NA, NA, peq,
+	NA, IDLE_CHK_WARNING,
+	"TSEM: interrupt 0 is active",
+	{NA, NA, 0x10000000, NA, NA, NA} },
+
+/*line 369*/{(0x1F), 1, TSEM_REG_TSEM_INT_STS_1,
+	NA, NA, NA, pneq,
+	NA, IDLE_CHK_ERROR,
+	"TSEM: interrupt 1 is active",
+	{NA, NA, 0, NA, NA, NA} },
+
+/*line 370*/{(0x1F), 1, CSEM_REG_CSEM_INT_STS_0,
+	NA, NA, NA, pand_neq,
+	NA, IDLE_CHK_WARNING,
+	"CSEM: interrupt 0 is active",
+	{NA, NA, ~0x10000000, 0, NA, NA} },
+
+/*line 371*/{(0x1F), 1, CSEM_REG_CSEM_INT_STS_0,
+	NA, NA, NA, peq,
+	NA, IDLE_CHK_WARNING,
+	"CSEM: interrupt 0 is active",
+	{NA, NA, 0x10000000, NA, NA, NA} },
+
+/*line 372*/{(0x1F), 1, CSEM_REG_CSEM_INT_STS_1,
+	NA, NA, NA, pneq,
+	NA, IDLE_CHK_ERROR,
+	"CSEM: interrupt 1 is active",
+	{NA, NA, 0, NA, NA, NA} },
+
+/*line 373*/{(0x1F), 1, USEM_REG_USEM_INT_STS_0,
+	NA, NA, NA, pand_neq,
+	NA, IDLE_CHK_WARNING,
+	"USEM: interrupt 0 is active",
+	{NA, NA, ~0x10000000, 0, NA, NA} },
+
+/*line 374*/{(0x1F), 1, USEM_REG_USEM_INT_STS_0,
+	NA, NA, NA, peq,
+	NA, IDLE_CHK_WARNING,
+	"USEM: interrupt 0 is active",
+	{NA, NA, 0x10000000, NA, NA, NA} },
+
+/*line 375*/{(0x1F), 1, USEM_REG_USEM_INT_STS_1,
+	NA, NA, NA, pneq,
+	NA, IDLE_CHK_ERROR,
+	"USEM: interrupt 1 is active",
+	{NA, NA, 0, NA, NA, NA} },
+
+/*line 376*/{(0x1F), 1, XSEM_REG_XSEM_INT_STS_0,
+	NA, NA, NA, pand_neq,
+	NA, IDLE_CHK_WARNING,
+	"XSEM: interrupt 0 is active",
+	{NA, NA, ~0x10000000, 0, NA, NA} },
+
+/*line 377*/{(0x1F), 1, XSEM_REG_XSEM_INT_STS_0,
+	NA, NA, NA, peq,
+	NA, IDLE_CHK_WARNING,
+	"XSEM: interrupt 0 is active",
+	{NA, NA, 0x10000000, NA, NA, NA} },
+
+/*line 378*/{(0x1F), 1, XSEM_REG_XSEM_INT_STS_1,
+	NA, NA, NA, pneq,
+	NA, IDLE_CHK_ERROR,
+	"XSEM: interrupt 1 is active",
+	{NA, NA, 0, NA, NA, NA} },
+
+/*line 379*/{(0x1F), 1, TSDM_REG_TSDM_INT_STS_0,
+	NA, NA, NA, pneq,
+	NA, IDLE_CHK_ERROR,
+	"TSDM: interrupt 0 is active",
+	{NA, NA, 0, NA, NA, NA} },
+
+/*line 380*/{(0x1F), 1, TSDM_REG_TSDM_INT_STS_1,
+	NA, NA, NA, pneq,
+	NA, IDLE_CHK_ERROR,
+	"TSDM: interrupt 0 is active",
+	{NA, NA, 0, NA, NA, NA} },
+
+/*line 381*/{(0x1F), 1, CSDM_REG_CSDM_INT_STS_0,
+	NA, NA, NA, pneq,
+	NA, IDLE_CHK_ERROR,
+	"CSDM: interrupt 0 is active",
+	{NA, NA, 0, NA, NA, NA} },
+
+/*line 382*/{(0x1F), 1, CSDM_REG_CSDM_INT_STS_1,
+	NA, NA, NA, pneq,
+	NA, IDLE_CHK_ERROR,
+	"CSDM: interrupt 0 is active",
+	{NA, NA, 0, NA, NA, NA} },
+
+/*line 383*/{(0x1F), 1, USDM_REG_USDM_INT_STS_0,
+	NA, NA, NA, pneq,
+	NA, IDLE_CHK_ERROR,
+	"USDM: interrupt 0 is active",
+	{NA, NA, 0, NA, NA, NA} },
+
+/*line 384*/{(0x1F), 1, USDM_REG_USDM_INT_STS_1,
+	NA, NA, NA, pneq,
+	NA, IDLE_CHK_ERROR,
+	"USDM: interrupt 0 is active",
+	{NA, NA, 0, NA, NA, NA} },
+
+/*line 385*/{(0x1F), 1, XSDM_REG_XSDM_INT_STS_0,
+	NA, NA, NA, pneq,
+	NA, IDLE_CHK_ERROR,
+	"XSDM: interrupt 0 is active",
+	{NA, NA, 0, NA, NA, NA} },
+
+/*line 386*/{(0x1F), 1, XSDM_REG_XSDM_INT_STS_1,
+	NA, NA, NA, pneq,
+	NA, IDLE_CHK_ERROR,
+	"XSDM: interrupt 0 is active",
+	{NA, NA, 0, NA, NA, NA} },
+
+/*line 387*/{(0x2), 1, HC_REG_HC_PRTY_STS,
+	NA, 1, 0, pneq,
+	NA, IDLE_CHK_WARNING,
+	"HC: parity status is not 0",
+	{NA, NA, 0, NA, NA, NA} },
+
+/*line 388*/{(0x1E), 1, MISC_REG_MISC_PRTY_STS,
+	NA, 1, 0, pneq,
+	NA, IDLE_CHK_WARNING,
+	"MISC: parity status is not 0",
+	{NA, NA, 0, NA, NA, NA} },
+
+/*line 389*/{(0x1E), 1, SRC_REG_SRC_PRTY_STS,
+	NA, 1, 0, pneq,
+	NA, IDLE_CHK_WARNING,
+	"SRCH: parity status is not 0",
+	{NA, NA, 0, NA, NA, NA} },
+
+/*line 390*/{(0xC), 3, QM_REG_BYTECRD0,
+	QM_REG_BYTECRDINITVAL, 1, 0, pneq_r2,
+	NA, IDLE_CHK_ERROR_NO_TRAFFIC,
+	"QM: Byte credit 0 is not equal to initial credit",
+	{NA, NA, NA, NA, NA, NA} },
+
+/*line 391*/{(0xC), 3, QM_REG_BYTECRD1,
+	QM_REG_BYTECRDINITVAL, 1, 0, pneq_r2,
+	NA, IDLE_CHK_ERROR_NO_TRAFFIC,
+	"QM: Byte credit 1 is not equal to initial credit",
+	{NA, NA, NA, NA, NA, NA} },
+
+/*line 392*/{(0xC), 3, QM_REG_BYTECRD2,
+	QM_REG_BYTECRDINITVAL, 1, 0, pneq_r2,
+	NA, IDLE_CHK_ERROR_NO_TRAFFIC,
+	"QM: Byte credit 2 is not equal to initial credit",
+	{NA, NA, NA, NA, NA, NA} },
+
+/*line 393*/{(0x1C), 1, QM_REG_VOQCRDERRREG,
+	NA, 1, 0, pand_neq,
+	NA, IDLE_CHK_ERROR,
+	"QM: VOQ credit error register is not 0 (VOQ credit overflow/underflow)",
+	{NA, NA, 0xFFFF, 0, NA, NA} },
+
+/*line 394*/{(0x1C), 1, QM_REG_BYTECRDERRREG,
+	NA, 1, 0, pand_neq,
+	NA, IDLE_CHK_ERROR,
+	"QM: Byte credit error register is not 0 (Byte credit overflow/underflow)",
+	{NA, NA, 0xFFF, 0, NA, NA} },
+
+/*line 395*/{(0x1C), 1, PGLUE_B_REG_FLR_REQUEST_VF_31_0,
+	NA, 1, 0, pneq,
+	NA, IDLE_CHK_WARNING,
+	"PGL: FLR request is set for VF addresses 31-0",
+	{NA, NA, 0, NA, NA, NA} },
+
+/*line 396*/{(0x1C), 1, PGLUE_B_REG_FLR_REQUEST_VF_63_32,
+	NA, 1, 0, pneq,
+	NA, IDLE_CHK_WARNING,
+	"PGL: FLR request is set for VF addresses 63-32",
+	{NA, NA, 0, NA, NA, NA} },
+
+/*line 397*/{(0x1C), 1, PGLUE_B_REG_FLR_REQUEST_VF_95_64,
+	NA, 1, 0, pneq,
+	NA, IDLE_CHK_WARNING,
+	"PGL: FLR request is set for VF addresses 95-64",
+	{NA, NA, 0, NA, NA, NA} },
+
+/*line 398*/{(0x1C), 1, PGLUE_B_REG_FLR_REQUEST_VF_127_96,
+	NA, 1, 0, pneq,
+	NA, IDLE_CHK_WARNING,
+	"PGL: FLR request is set for VF addresses 127-96",
+	{NA, NA, 0, NA, NA, NA} },
+
+/*line 399*/{(0x1C), 1, PGLUE_B_REG_FLR_REQUEST_PF_7_0,
+	NA, 1, 0, pneq,
+	NA, IDLE_CHK_WARNING,
+	"PGL: FLR request is set for PF addresses 7-0",
+	{NA, NA, 0, NA, NA, NA} },
+
+/*line 400*/{(0x1C), 1, PGLUE_B_REG_SR_IOV_DISABLED_REQUEST,
+	NA, 1, 0, pneq,
+	NA, IDLE_CHK_WARNING,
+	"PGL: SR-IOV disable request is set",
+	{NA, NA, 0, NA, NA, NA} },
+
+/*line 401*/{(0x1C), 1, PGLUE_B_REG_CFG_SPACE_A_REQUEST,
+	NA, 1, 0, pneq,
+	NA, IDLE_CHK_WARNING,
+	"PGL: Cfg-Space A request is set",
+	{NA, NA, 0, NA, NA, NA} },
+
+/*line 402*/{(0x1C), 1, PGLUE_B_REG_CFG_SPACE_B_REQUEST,
+	NA, 1, 0, pneq,
+	NA, IDLE_CHK_WARNING,
+	"PGL: Cfg-Space B request is set",
+	{NA, NA, 0, NA, NA, NA} },
+
+/*line 403*/{(0x1C), 1, IGU_REG_ERROR_HANDLING_DATA_VALID,
+	NA, NA, 0, pneq,
+	NA, IDLE_CHK_WARNING,
+	"IGU: some unauthorized commands arrived to the IGU. Use igu_dump_fifo utility for more details",
+	{NA, NA, 0, NA, NA, NA} },
+
+/*line 404*/{(0x1C), 1, IGU_REG_ATTN_WRITE_DONE_PENDING,
+	NA, NA, NA, pneq,
+	NA, IDLE_CHK_WARNING,
+	"IGU attention message write done pending is not empty",
+	{NA, NA, 0, NA, NA, NA} },
+
+/*line 405*/{(0x1C), 1, IGU_REG_WRITE_DONE_PENDING,
+	NA, 5, 4, pneq,
+	NA, IDLE_CHK_WARNING,
+	"IGU MSI/MSIX message write done pending is not empty",
+	{NA, NA, 0, NA, NA, NA} },
+
+/*line 406*/{(0x1C), 1, IGU_REG_IGU_PRTY_STS,
+	NA, 1, 0, pneq,
+	NA, IDLE_CHK_WARNING,
+	"IGU: parity status is not 0",
+	{NA, NA, 0, NA, NA, NA} },
+
+/*line 407*/{(0x1E), 3, MISC_REG_GRC_TIMEOUT_ATTN,
+	MISC_REG_AEU_AFTER_INVERT_4_FUNC_0, 1, 0, pand_neq,
+	NA, IDLE_CHK_ERROR,
+	"MISC_REG_GRC_TIMEOUT_ATTN: GRC timeout attention parameters (FUNC_0)",
+	{NA, NA, 0x4000000, 0, NA, NA} },
+
+/*line 408*/{(0x1C), 3, MISC_REG_GRC_TIMEOUT_ATTN_FULL_FID,
+	MISC_REG_AEU_AFTER_INVERT_4_FUNC_0, 1, 0, pand_neq,
+	NA, IDLE_CHK_ERROR,
+	"MISC_REG_GRC_TIMEOUT_ATTN_FULL_FID: GRC timeout attention FID (FUNC_0)",
+	{NA, NA, 0x4000000, 0, NA, NA} },
+
+/*line 409*/{(0x1E), 3, MISC_REG_GRC_TIMEOUT_ATTN,
+	MISC_REG_AEU_AFTER_INVERT_4_FUNC_1, 1, 0, pand_neq,
+	NA, IDLE_CHK_ERROR,
+	"MISC_REG_GRC_TIMEOUT_ATTN: GRC timeout attention parameters (FUNC_1)",
+	{NA, NA, 0x4000000, 0, NA, NA} },
+
+/*line 410*/{(0x1C), 3, MISC_REG_GRC_TIMEOUT_ATTN_FULL_FID,
+	MISC_REG_AEU_AFTER_INVERT_4_FUNC_1, 1, 0, pand_neq,
+	NA, IDLE_CHK_ERROR,
+	"MISC_REG_GRC_TIMEOUT_ATTN_FULL_FID: GRC timeout attention FID (FUNC_1)",
+	{NA, NA, 0x4000000, 0, NA, NA} },
+
+/*line 411*/{(0x1E), 3, MISC_REG_GRC_TIMEOUT_ATTN,
+	MISC_REG_AEU_AFTER_INVERT_4_MCP, 1, 0, pand_neq,
+	NA, IDLE_CHK_ERROR,
+	"MISC_REG_GRC_TIMEOUT_ATTN: GRC timeout attention parameters (MCP)",
+	{NA, NA, 0x4000000, 0, NA, NA} },
+
+/*line 412*/{(0x1C), 3, MISC_REG_GRC_TIMEOUT_ATTN_FULL_FID,
+	MISC_REG_AEU_AFTER_INVERT_4_MCP, 1, 0, pand_neq,
+	NA, IDLE_CHK_ERROR,
+	"MISC_REG_GRC_TIMEOUT_ATTN_FULL_FID: GRC timeout attention FID (MCP)",
+	{NA, NA, 0x4000000, 0, NA, NA} },
+
+/*line 413*/{(0x1C), 1, IGU_REG_SILENT_DROP,
+	NA, 1, 0, pneq,
+	NA, IDLE_CHK_ERROR,
+	"Some messages were not executed in the IGU",
+	{NA, NA, 0, NA, NA, NA} },
+
+/*line 414*/{(0x1C), 1, PXP2_REG_PSWRQ_BW_CREDIT,
+	NA, 1, 0, pneq,
+	NA, IDLE_CHK_ERROR,
+	"PXP2: rq_read_credit and rq_write_credit are not 5",
+	{NA, NA, 0x2D, NA, NA, NA} },
+
+/*line 415*/{(0x1C), 1, IGU_REG_SB_CTRL_FSM,
+	NA, 1, 0, pneq,
+	NA, IDLE_CHK_WARNING,
+	"IGU: block is not in idle. SB_CTRL_FSM should be zero in idle state",
+	{NA, NA, 0, NA, NA, NA} },
+
+/*line 416*/{(0x1C), 1, IGU_REG_INT_HANDLE_FSM,
+	NA, 1, 0, pneq,
+	NA, IDLE_CHK_WARNING,
+	"IGU: block is not in idle. INT_HANDLE_FSM should be zero in idle state",
+	{NA, NA, 0, NA, NA, NA} },
+
+/*line 417*/{(0x1C), 1, IGU_REG_ATTN_FSM,
+	NA, 1, 0, pand_neq,
+	NA, IDLE_CHK_WARNING,
+	"IGU: block is not in idle. SB_ATTN_FSMshould be zeroor two in idle state",
+	{NA, NA, ~0x2, 0, NA, NA} },
+
+/*line 418*/{(0x1C), 1, IGU_REG_CTRL_FSM,
+	NA, 1, 0, pand_neq,
+	NA, IDLE_CHK_WARNING,
+	"IGU: block is not in idle. SB_CTRL_FSM should be zero in idle state",
+	{NA, NA, ~0x1, 0, NA, NA} },
+
+/*line 419*/{(0x1C), 1, IGU_REG_PXP_ARB_FSM,
+	NA, 1, 0, pand_neq,
+	NA, IDLE_CHK_WARNING,
+	"IGU: block is not in idle. SB_ARB_FSM should be zero in idle state",
+	{NA, NA, ~0x1, 0, NA, NA} },
+
+/*line 420*/{(0x1C), 1, IGU_REG_PENDING_BITS_STATUS,
+	NA, 5, 4, pneq,
+	NA, IDLE_CHK_WARNING,
+	"IGU: block is not in idle. There are pending write done",
+	{NA, NA, 0, NA, NA, NA} },
+
+/*line 421*/{(0x10), 3, QM_REG_VOQCREDIT_0,
+	QM_REG_VOQINITCREDIT_0, 1, 0, pneq_r2,
+	NA, IDLE_CHK_ERROR_NO_TRAFFIC,
+	"QM: VOQ_0, VOQ credit is not equal to initial credit",
+	{NA, NA, NA, NA, NA, NA} },
+
+/*line 422*/{(0x10), 3, QM_REG_VOQCREDIT_1,
+	QM_REG_VOQINITCREDIT_1, 1, 0, pneq_r2,
+	NA, IDLE_CHK_ERROR_NO_TRAFFIC,
+	"QM: VOQ_1, VOQ credit is not equal to initial credit",
+	{NA, NA, NA, NA, NA, NA} },
+
+/*line 423*/{(0x10), 3, QM_REG_VOQCREDIT_2,
+	QM_REG_VOQINITCREDIT_2, 1, 0, pneq_r2,
+	NA, IDLE_CHK_ERROR_NO_TRAFFIC,
+	"QM: VOQ_2, VOQ credit is not equal to initial credit",
+	{NA, NA, NA, NA, NA, NA} },
+
+/*line 424*/{(0x10), 3, QM_REG_VOQCREDIT_3,
+	QM_REG_VOQINITCREDIT_3, 1, 0, pneq_r2,
+	NA, IDLE_CHK_ERROR_NO_TRAFFIC,
+	"QM: VOQ_3, VOQ credit is not equal to initial credit",
+	{NA, NA, NA, NA, NA, NA} },
+
+/*line 425*/{(0x10), 3, QM_REG_VOQCREDIT_4,
+	QM_REG_VOQINITCREDIT_4, 1, 0, pneq_r2,
+	NA, IDLE_CHK_ERROR_NO_TRAFFIC,
+	"QM: VOQ_4, VOQ credit is not equal to initial credit",
+	{NA, NA, NA, NA, NA, NA} },
+
+/*line 426*/{(0x10), 3, QM_REG_VOQCREDIT_5,
+	QM_REG_VOQINITCREDIT_5, 1, 0, pneq_r2,
+	NA, IDLE_CHK_ERROR_NO_TRAFFIC,
+	"QM: VOQ_5, VOQ credit is not equal to initial credit",
+	{NA, NA, NA, NA, NA, NA} },
+
+/*line 427*/{(0x10), 3, QM_REG_VOQCREDIT_6,
+	QM_REG_VOQINITCREDIT_6, 1, 0, pneq_r2,
+	NA, IDLE_CHK_ERROR_NO_TRAFFIC,
+	"QM: VOQ_6 (LB VOQ), VOQ credit is not equal to initial credit",
+	{NA, NA, NA, NA, NA, NA} },
+
+/*line 428*/{(0x10), 3, QM_REG_BYTECRD0,
+	QM_REG_BYTECRDINITVAL, 1, 0, pneq_r2,
+	NA, IDLE_CHK_ERROR_NO_TRAFFIC,
+	"QM: Byte credit 0 is not equal to initial credit",
+	{NA, NA, NA, NA, NA, NA} },
+
+/*line 429*/{(0x10), 3, QM_REG_BYTECRD1,
+	QM_REG_BYTECRDINITVAL, 1, 0, pneq_r2,
+	NA, IDLE_CHK_ERROR_NO_TRAFFIC,
+	"QM: Byte credit 1 is not equal to initial credit",
+	{NA, NA, NA, NA, NA, NA} },
+
+/*line 430*/{(0x10), 3, QM_REG_BYTECRD2,
+	QM_REG_BYTECRDINITVAL, 1, 0, pneq_r2,
+	NA, IDLE_CHK_ERROR_NO_TRAFFIC,
+	"QM: Byte credit 2 is not equal to initial credit",
+	{NA, NA, NA, NA, NA, NA} },
+
+/*line 431*/{(0x10), 3, QM_REG_BYTECRD3,
+	QM_REG_BYTECRDINITVAL, 1, 0, pneq_r2,
+	NA, IDLE_CHK_ERROR_NO_TRAFFIC,
+	"QM: Byte credit 3 is not equal to initial credit",
+	{NA, NA, NA, NA, NA, NA} },
+
+/*line 432*/{(0x10), 3, QM_REG_BYTECRD4,
+	QM_REG_BYTECRDINITVAL, 1, 0, pneq_r2,
+	NA, IDLE_CHK_ERROR_NO_TRAFFIC,
+	"QM: Byte credit 4 is not equal to initial credit",
+	{NA, NA, NA, NA, NA, NA} },
+
+/*line 433*/{(0x10), 3, QM_REG_BYTECRD5,
+	QM_REG_BYTECRDINITVAL, 1, 0, pneq_r2,
+	NA, IDLE_CHK_ERROR_NO_TRAFFIC,
+	"QM: Byte credit 5 is not equal to initial credit",
+	{NA, NA, NA, NA, NA, NA} },
+
+/*line 434*/{(0x10), 3, QM_REG_BYTECRD6,
+	QM_REG_BYTECRDINITVAL, 1, 0, pneq_r2,
+	NA, IDLE_CHK_ERROR_NO_TRAFFIC,
+	"QM: Byte credit 6 is not equal to initial credit",
+	{NA, NA, NA, NA, NA, NA} },
+
+/*line 435*/{(0x10), 1, QM_REG_FWVOQ0TOHWVOQ,
+	NA, 1, 0, peq,
+	NA, IDLE_CHK_ERROR,
+	"QM: FwVoq0 is mapped to HwVoq7 (non-TX HwVoq)",
+	{NA, NA, 0x7, NA, NA, NA} },
+
+/*line 436*/{(0x10), 1, QM_REG_FWVOQ1TOHWVOQ,
+	NA, 1, 0, peq,
+	NA, IDLE_CHK_ERROR,
+	"QM: FwVoq1 is mapped to HwVoq7 (non-TX HwVoq)",
+	{NA, NA, 0x7, NA, NA, NA} },
+
+/*line 437*/{(0x10), 1, QM_REG_FWVOQ2TOHWVOQ,
+	NA, 1, 0, peq,
+	NA, IDLE_CHK_ERROR,
+	"QM: FwVoq2 is mapped to HwVoq7 (non-TX HwVoq)",
+	{NA, NA, 0x7, NA, NA, NA} },
+
+/*line 438*/{(0x10), 1, QM_REG_FWVOQ3TOHWVOQ,
+	NA, 1, 0, peq,
+	NA, IDLE_CHK_ERROR,
+	"QM: FwVoq3 is mapped to HwVoq7 (non-TX HwVoq)",
+	{NA, NA, 0x7, NA, NA, NA} },
+
+/*line 439*/{(0x10), 1, QM_REG_FWVOQ4TOHWVOQ,
+	NA, 1, 0, peq,
+	NA, IDLE_CHK_ERROR,
+	"QM: FwVoq4 is mapped to HwVoq7 (non-TX HwVoq)",
+	{NA, NA, 0x7, NA, NA, NA} },
+
+/*line 440*/{(0x10), 1, QM_REG_FWVOQ5TOHWVOQ,
+	NA, 1, 0, peq,
+	NA, IDLE_CHK_ERROR,
+	"QM: FwVoq5 is mapped to HwVoq7 (non-TX HwVoq)",
+	{NA, NA, 0x7, NA, NA, NA} },
+
+/*line 441*/{(0x10), 1, QM_REG_FWVOQ6TOHWVOQ,
+	NA, 1, 0, peq,
+	NA, IDLE_CHK_ERROR,
+	"QM: FwVoq6 is mapped to HwVoq7 (non-TX HwVoq)",
+	{NA, NA, 0x7, NA, NA, NA} },
+
+/*line 442*/{(0x10), 1, QM_REG_FWVOQ7TOHWVOQ,
+	NA, 1, 0, peq,
+	NA, IDLE_CHK_ERROR,
+	"QM: FwVoq7 is mapped to HwVoq7 (non-TX HwVoq)",
+	{NA, NA, 0x7, NA, NA, NA} },
+
+/*line 443*/{(0x1F), 1, NIG_REG_INGRESS_EOP_PORT0_EMPTY,
+	NA, 1, 0, pneq,
+	NA, IDLE_CHK_ERROR_NO_TRAFFIC,
+	"NIG: Port 0 EOP FIFO is not empty",
+	{NA, NA, 1, NA, NA, NA} },
+
+/*line 444*/{(0x1F), 1, NIG_REG_INGRESS_EOP_PORT1_EMPTY,
+	NA, 1, 0, pneq,
+	NA, IDLE_CHK_ERROR_NO_TRAFFIC,
+	"NIG: Port 1 EOP FIFO is not empty",
+	{NA, NA, 1, NA, NA, NA} },
+
+/*line 445*/{(0x1F), 1, NIG_REG_INGRESS_EOP_LB_EMPTY,
+	NA, 1, 0, pneq,
+	NA, IDLE_CHK_ERROR_NO_TRAFFIC,
+	"NIG: LB EOP FIFO is not empty",
+	{NA, NA, 1, NA, NA, NA} },
+
+/*line 446*/{(0x1F), 1, NIG_REG_INGRESS_RMP0_DSCR_EMPTY,
+	NA, 1, 0, pneq,
+	NA, IDLE_CHK_WARNING,
+	"NIG: Port 0 RX MCP descriptor FIFO is not empty",
+	{NA, NA, 1, NA, NA, NA} },
+
+/*line 447*/{(0x1F), 1, NIG_REG_INGRESS_RMP1_DSCR_EMPTY,
+	NA, 1, 0, pneq,
+	NA, IDLE_CHK_WARNING,
+	"NIG: Port 1 RX MCP descriptor FIFO is not empty",
+	{NA, NA, 1, NA, NA, NA} },
+
+/*line 448*/{(0x1F), 1, NIG_REG_INGRESS_LB_PBF_DELAY_EMPTY,
+	NA, 1, 0, pneq,
+	NA, IDLE_CHK_ERROR_NO_TRAFFIC,
+	"NIG: PBF LB FIFO is not empty",
+	{NA, NA, 1, NA, NA, NA} },
+
+/*line 449*/{(0x1F), 1, NIG_REG_EGRESS_MNG0_FIFO_EMPTY,
+	NA, 1, 0, pneq,
+	NA, IDLE_CHK_WARNING,
+	"NIG: Port 0 TX MCP FIFO is not empty",
+	{NA, NA, 1, NA, NA, NA} },
+
+/*line 450*/{(0x1F), 1, NIG_REG_EGRESS_MNG1_FIFO_EMPTY,
+	NA, 1, 0, pneq,
+	NA, IDLE_CHK_WARNING,
+	"NIG: Port 1 TX MCP FIFO is not empty",
+	{NA, NA, 1, NA, NA, NA} },
+
+/*line 451*/{(0x1F), 1, NIG_REG_EGRESS_DEBUG_FIFO_EMPTY,
+	NA, 1, 0, pneq,
+	NA, IDLE_CHK_ERROR_NO_TRAFFIC,
+	"NIG: Debug FIFO is not empty",
+	{NA, NA, 1, NA, NA, NA} },
+
+/*line 452*/{(0x1F), 1, NIG_REG_EGRESS_DELAY0_EMPTY,
+	NA, 1, 0, pneq,
+	NA, IDLE_CHK_ERROR_NO_TRAFFIC,
+	"NIG: PBF IF0 FIFO is not empty",
+	{NA, NA, 1, NA, NA, NA} },
+
+/*line 453*/{(0x1F), 1, NIG_REG_EGRESS_DELAY1_EMPTY,
+	NA, 1, 0, pneq,
+	NA, IDLE_CHK_ERROR_NO_TRAFFIC,
+	"NIG: PBF IF1 FIFO is not empty",
+	{NA, NA, 1, NA, NA, NA} },
+
+/*line 454*/{(0x1F), 1, NIG_REG_LLH0_FIFO_EMPTY,
+	NA, 1, 0, pneq,
+	NA, IDLE_CHK_ERROR_NO_TRAFFIC,
+	"NIG: Port 0 RX LLH FIFO is not empty",
+	{NA, NA, 1, NA, NA, NA} },
+
+/*line 455*/{(0x1F), 1, NIG_REG_LLH1_FIFO_EMPTY,
+	NA, 1, 0, pneq,
+	NA, IDLE_CHK_ERROR_NO_TRAFFIC,
+	"NIG: Port 1 RX LLH FIFO is not empty",
+	{NA, NA, 1, NA, NA, NA} },
+
+/*line 456*/{(0x1C), 1, NIG_REG_P0_TX_MNG_HOST_FIFO_EMPTY,
+	NA, 1, 0, pneq,
+	NA, IDLE_CHK_WARNING,
+	"NIG: Port 0 TX MCP FIFO for traffic going to the host is not empty",
+	{NA, NA, 1, NA, NA, NA} },
+
+/*line 457*/{(0x1C), 1, NIG_REG_P1_TX_MNG_HOST_FIFO_EMPTY,
+	NA, 1, 0, pneq,
+	NA, IDLE_CHK_WARNING,
+	"NIG: Port 1 TX MCP FIFO for traffic going to the host is not empty",
+	{NA, NA, 1, NA, NA, NA} },
+
+/*line 458*/{(0x1C), 1, NIG_REG_P0_TLLH_FIFO_EMPTY,
+	NA, 1, 0, pneq,
+	NA, IDLE_CHK_ERROR_NO_TRAFFIC,
+	"NIG: Port 0 TX LLH FIFO is not empty",
+	{NA, NA, 1, NA, NA, NA} },
+
+/*line 459*/{(0x1C), 1, NIG_REG_P1_TLLH_FIFO_EMPTY,
+	NA, 1, 0, pneq,
+	NA, IDLE_CHK_ERROR_NO_TRAFFIC,
+	"NIG: Port 1 TX LLH FIFO is not empty",
+	{NA, NA, 1, NA, NA, NA} },
+
+/*line 460*/{(0x1C), 1, NIG_REG_P0_HBUF_DSCR_EMPTY,
+	NA, 1, 0, pneq,
+	NA, IDLE_CHK_WARNING,
+	"NIG: Port 0 RX MCP descriptor FIFO for traffic from the host is not empty",
+	{NA, NA, 1, NA, NA, NA} },
+
+/*line 461*/{(0x1C), 1, NIG_REG_P1_HBUF_DSCR_EMPTY,
+	NA, 1, 0, pneq,
+	NA, IDLE_CHK_WARNING,
+	"NIG: Port 1 RX MCP descriptor FIFO for traffic from the host is not empty",
+	{NA, NA, 1, NA, NA, NA} },
+
+/*line 462*/{(0x18), 1, NIG_REG_P0_RX_MACFIFO_EMPTY,
+	NA, 1, 0, pneq,
+	NA, IDLE_CHK_ERROR_NO_TRAFFIC,
+	"NIG: Port 0 RX MAC interface FIFO is not empty",
+	{NA, NA, 1, NA, NA, NA} },
+
+/*line 463*/{(0x18), 1, NIG_REG_P1_RX_MACFIFO_EMPTY,
+	NA, 1, 0, pneq,
+	NA, IDLE_CHK_ERROR_NO_TRAFFIC,
+	"NIG: Port 1 RX MAC interface FIFO is not empty",
+	{NA, NA, 1, NA, NA, NA} },
+
+/*line 464*/{(0x18), 1, NIG_REG_P0_TX_MACFIFO_EMPTY,
+	NA, 1, 0, pneq,
+	NA, IDLE_CHK_ERROR_NO_TRAFFIC,
+	"NIG: Port 0 TX MAC interface FIFO is not empty",
+	{NA, NA, 1, NA, NA, NA} },
+
+/*line 465*/{(0x18), 1, NIG_REG_P1_TX_MACFIFO_EMPTY,
+	NA, 1, 0, pneq,
+	NA, IDLE_CHK_ERROR_NO_TRAFFIC,
+	"NIG: Port 1 TX MAC interface FIFO is not empty",
+	{NA, NA, 1, NA, NA, NA} },
+
+/*line 466*/{(0x10), 1, NIG_REG_EGRESS_DELAY2_EMPTY,
+	NA, 1, 0, pneq,
+	NA, IDLE_CHK_ERROR_NO_TRAFFIC,
+	"NIG: PBF IF2 FIFO is not empty",
+	{NA, NA, 1, NA, NA, NA} },
+
+/*line 467*/{(0x10), 1, NIG_REG_EGRESS_DELAY3_EMPTY,
+	NA, 1, 0, pneq,
+	NA, IDLE_CHK_ERROR_NO_TRAFFIC,
+	"NIG: PBF IF3 FIFO is not empty",
+	{NA, NA, 1, NA, NA, NA} },
+
+/*line 468*/{(0x10), 1, NIG_REG_EGRESS_DELAY4_EMPTY,
+	NA, 1, 0, pneq,
+	NA, IDLE_CHK_ERROR_NO_TRAFFIC,
+	"NIG: PBF IF4 FIFO is not empty",
+	{NA, NA, 1, NA, NA, NA} },
+
+/*line 469*/{(0x10), 1, NIG_REG_EGRESS_DELAY5_EMPTY,
+	NA, 1, 0, pneq,
+	NA, IDLE_CHK_ERROR_NO_TRAFFIC,
+	"NIG: PBF IF5 FIFO is not empty",
+	{NA, NA, 1, NA, NA, NA} },
+};
+
+/* handle self test fails according to severity and type */
+static void bnx2x_self_test_log(struct bnx2x *bp, u8 severity, char *message)
+{
+	switch (severity) {
+	case IDLE_CHK_ERROR:
+		BNX2X_ERR("ERROR %s", message);
+		idle_chk_errors++;
+		break;
+	case IDLE_CHK_ERROR_NO_TRAFFIC:
+		DP(NETIF_MSG_HW, "INFO %s", message);
+		break;
+	case IDLE_CHK_WARNING:
+		DP(NETIF_MSG_HW, "WARNING %s", message);
+		idle_chk_warnings++;
+		break;
+	}
+}
+
+/* specific test for QM rd/wr pointers and rd/wr banks */
+static void bnx2x_idle_chk6(struct bnx2x *bp,
+			    struct st_record *rec, char *message)
+{
+	u32 rd_ptr, wr_ptr, rd_bank, wr_bank;
+	int i;
+
+	for (i = 0; i < rec->loop; i++) {
+		/* read regs */
+		rec->pred_args.val1 =
+			REG_RD(bp, rec->reg1 + i * rec->incr);
+		rec->pred_args.val2 =
+			REG_RD(bp, rec->reg1 + i * rec->incr + 4);
+
+		/* calc read and write pointers */
+		rd_ptr = ((rec->pred_args.val1 & 0x3FFFFFC0) >> 6);
+		wr_ptr = ((((rec->pred_args.val1 & 0xC0000000) >> 30) & 0x3) |
+			((rec->pred_args.val2 & 0x3FFFFF) << 2));
+
+		/* perfrom pointer test */
+		if (rd_ptr != wr_ptr) {
+			snprintf(message, MAX_FAIL_MSG,
+				 "QM: PTRTBL entry %d- rd_ptr is not equal to wr_ptr. Values are 0x%x and 0x%x\n",
+				 i, rd_ptr, wr_ptr);
+			bnx2x_self_test_log(bp, rec->severity, message);
+		}
+
+		/* calculate read and write banks */
+		rd_bank = ((rec->pred_args.val1 & 0x30) >> 4);
+		wr_bank = (rec->pred_args.val1 & 0x03);
+
+		/* perform bank test */
+		if (rd_bank != wr_bank) {
+			snprintf(message, MAX_FAIL_MSG,
+				 "QM: PTRTBL entry %d - rd_bank is not equal to wr_bank. Values are 0x%x 0x%x\n",
+				 i, rd_bank, wr_bank);
+			bnx2x_self_test_log(bp, rec->severity, message);
+		}
+	}
+}
+
+/* specific test for cfc info ram and cid cam */
+static void bnx2x_idle_chk7(struct bnx2x *bp,
+			    struct st_record *rec, char *message)
+{
+	int i;
+
+	/* iterate through lcids */
+	for (i = 0; i < rec->loop; i++) {
+		/* make sure cam entry is valid (bit 0) */
+		if ((REG_RD(bp, (rec->reg2 + i * 4)) & 0x1) != 0x1)
+			continue;
+
+		/* get connection type (multiple reads due to widebus) */
+		REG_RD(bp, (rec->reg1 + i * rec->incr));
+		REG_RD(bp, (rec->reg1 + i * rec->incr + 4));
+		rec->pred_args.val1 =
+			REG_RD(bp, (rec->reg1 + i * rec->incr + 8));
+		REG_RD(bp, (rec->reg1 + i * rec->incr + 12));
+
+		/* obtain connection type */
+		if (is_e1 || is_e1h) {
+			/* E1 E1H (bits 4..7) */
+			rec->pred_args.val1 &= 0x78;
+			rec->pred_args.val1 >>= 3;
+		} else {
+			/* E2 E3A0 E3B0 (bits 26..29) */
+			rec->pred_args.val1 &= 0x1E000000;
+			rec->pred_args.val1 >>= 25;
+		}
+
+		/* get activity counter value */
+		rec->pred_args.val2 = REG_RD(bp, rec->reg3 + i * 4);
+
+		/* validate ac value is legal for con_type at idle state */
+		if (rec->bnx2x_predicate(&rec->pred_args)) {
+			snprintf(message, MAX_FAIL_MSG,
+				 "%s. Values are 0x%x 0x%x\n", rec->fail_msg,
+				 rec->pred_args.val1, rec->pred_args.val2);
+			bnx2x_self_test_log(bp, rec->severity, message);
+		}
+	}
+}
+
+/* self test procedure
+ * scan auto-generated database
+ * for each line:
+ * 1.	compare chip mask
+ * 2.	determine type (according to maro number)
+ * 3.	read registers
+ * 4.	call predicate
+ * 5.	collate results and statistics
+ */
+int bnx2x_idle_chk(struct bnx2x *bp)
+{
+	u16 i;				/* loop counter */
+	u16 st_ind;			/* self test database access index */
+	struct st_record rec;		/* current record variable */
+	char message[MAX_FAIL_MSG];	/* message to log */
+
+	/*init stats*/
+	idle_chk_errors = 0;
+	idle_chk_warnings = 0;
+
+	/*create masks for all chip types*/
+	is_e1	= CHIP_IS_E1(bp);
+	is_e1h	= CHIP_IS_E1H(bp);
+	is_e2	= CHIP_IS_E2(bp);
+	is_e3a0	= CHIP_IS_E3A0(bp);
+	is_e3b0	= CHIP_IS_E3B0(bp);
+
+	/*database main loop*/
+	for (st_ind = 0; st_ind < ST_DB_LINES; st_ind++) {
+		rec = st_database[st_ind];
+
+		/*check if test applies to chip*/
+		if (!((rec.chip_mask & IDLE_CHK_E1) && is_e1) &&
+		    !((rec.chip_mask & IDLE_CHK_E1H) && is_e1h) &&
+		    !((rec.chip_mask & IDLE_CHK_E2) && is_e2) &&
+		    !((rec.chip_mask & IDLE_CHK_E3A0) && is_e3a0) &&
+		    !((rec.chip_mask & IDLE_CHK_E3B0) && is_e3b0))
+			continue;
+
+		/* identify macro */
+		switch (rec.macro) {
+		case 1:
+			/* read single reg and call predicate */
+			rec.pred_args.val1 = REG_RD(bp, rec.reg1);
+			DP(BNX2X_MSG_IDLE, "mac1 add %x\n", rec.reg1);
+			if (rec.bnx2x_predicate(&rec.pred_args)) {
+				snprintf(message, sizeof(message),
+					 "%s.Value is 0x%x\n", rec.fail_msg,
+					 rec.pred_args.val1);
+				bnx2x_self_test_log(bp, rec.severity, message);
+			}
+			break;
+		case 2:
+			/* read repeatedly starting from reg1 and call
+			 * predicate after each read
+			 */
+			for (i = 0; i < rec.loop; i++) {
+				rec.pred_args.val1 =
+					REG_RD(bp, rec.reg1 + i * rec.incr);
+				DP(BNX2X_MSG_IDLE, "mac2 add %x\n", rec.reg1);
+				if (rec.bnx2x_predicate(&rec.pred_args)) {
+					snprintf(message, sizeof(message),
+						 "%s. Value is 0x%x in loop %d\n",
+						 rec.fail_msg,
+						 rec.pred_args.val1, i);
+					bnx2x_self_test_log(bp, rec.severity,
+							    message);
+				}
+			}
+			break;
+		case 3:
+			/* read two regs and call predicate */
+			rec.pred_args.val1 = REG_RD(bp, rec.reg1);
+			rec.pred_args.val2 = REG_RD(bp, rec.reg2);
+			DP(BNX2X_MSG_IDLE, "mac3 add1 %x add2 %x\n",
+			   rec.reg1, rec.reg2);
+			if (rec.bnx2x_predicate(&rec.pred_args)) {
+				snprintf(message, sizeof(message),
+					 "%s. Values are 0x%x 0x%x\n",
+					 rec.fail_msg, rec.pred_args.val1,
+					 rec.pred_args.val2);
+				bnx2x_self_test_log(bp, rec.severity, message);
+			}
+			break;
+		case 4:
+			/*unused to-date*/
+			for (i = 0; i < rec.loop; i++) {
+				rec.pred_args.val1 =
+					REG_RD(bp, rec.reg1 + i * rec.incr);
+				rec.pred_args.val2 =
+					(REG_RD(bp,
+						rec.reg2 + i * rec.incr)) >> 1;
+				if (rec.bnx2x_predicate(&rec.pred_args)) {
+					snprintf(message, sizeof(message),
+						 "%s. Values are 0x%x 0x%x in loop %d\n",
+						 rec.fail_msg,
+						 rec.pred_args.val1,
+						 rec.pred_args.val2, i);
+					bnx2x_self_test_log(bp, rec.severity,
+							    message);
+				}
+			}
+			break;
+		case 5:
+			/* compare two regs, pending
+			 * the value of a condition reg
+			 */
+			rec.pred_args.val1 = REG_RD(bp, rec.reg1);
+			rec.pred_args.val2 = REG_RD(bp, rec.reg2);
+			DP(BNX2X_MSG_IDLE, "mac3 add1 %x add2 %x add3 %x\n",
+			   rec.reg1, rec.reg2, rec.reg3);
+			if (REG_RD(bp, rec.reg3) != 0) {
+				if (rec.bnx2x_predicate(&rec.pred_args)) {
+					snprintf(message, sizeof(message),
+						 "%s. Values are 0x%x 0x%x\n",
+						 rec.fail_msg,
+						 rec.pred_args.val1,
+						 rec.pred_args.val2);
+					bnx2x_self_test_log(bp, rec.severity,
+							    message);
+				}
+			}
+			break;
+		case 6:
+			/* compare read and write pointers
+			 * and read and write banks in QM
+			 */
+			bnx2x_idle_chk6(bp, &rec, message);
+			break;
+		case 7:
+			/* compare cfc info cam with cid cam */
+			bnx2x_idle_chk7(bp, &rec, message);
+			break;
+		default:
+			DP(BNX2X_MSG_IDLE,
+			   "unknown macro in self test data base. macro %d line %d",
+			   rec.macro, st_ind);
+		}
+	}
+
+	/* abort if interface is not running */
+	if (!netif_running(bp->dev))
+		return idle_chk_errors;
+
+	/* return value accorindg to statistics */
+	if (idle_chk_errors == 0) {
+		DP(BNX2X_MSG_IDLE,
+		   "completed successfully (logged %d warnings)\n",
+		   idle_chk_warnings);
+	} else {
+		BNX2X_ERR("failed (with %d errors, %d warnings)\n",
+			  idle_chk_errors, idle_chk_warnings);
+	}
+	return idle_chk_errors;
+}
diff --git a/drivers/net/ethernet/broadcom/bnx2x/bnx2x_sp.c b/drivers/net/ethernet/broadcom/bnx2x/bnx2x_sp.c
index 8baf9d3eb4b1..02c8213915a5 100644
--- a/drivers/net/ethernet/broadcom/bnx2x/bnx2x_sp.c
+++ b/drivers/net/ethernet/broadcom/bnx2x/bnx2x_sp.c
@@ -37,10 +37,12 @@
 /**
  * bnx2x_exe_queue_init - init the Exe Queue object
  *
+ * @bp:		driver handle
  * @o:		pointer to the object
  * @exe_len:	length
  * @owner:	pointer to the owner
  * @validate:	validate function pointer
+ * @remove:	remove function pointer
  * @optimize:	optimize function pointer
  * @exec:	execute function pointer
  * @get:	get function pointer
@@ -103,7 +105,7 @@ static inline int bnx2x_exe_queue_length(struct bnx2x_exe_queue_obj *o)
  *
  * @bp:		driver handle
  * @o:		queue
- * @cmd:	new command to add
+ * @elem:	new command to add
  * @restore:	true - do not optimize the command
  *
  * If the element is optimized or is illegal, frees it.
@@ -277,7 +279,7 @@ static void bnx2x_raw_set_pending(struct bnx2x_raw_obj *o)
  *
  * @bp:		device handle
  * @state:	state which is to be cleared
- * @state_p:	state buffer
+ * @pstate:	state buffer
  *
  */
 static inline int bnx2x_state_wait(struct bnx2x *bp, int state,
@@ -424,8 +426,8 @@ static bool bnx2x_put_credit_vlan_mac(struct bnx2x_vlan_mac_obj *o)
  * @bp:		device handle
  * @o:		vlan_mac object
  *
- * @details: Non-blocking implementation; should be called under execution
- *           queue lock.
+ * Context: Non-blocking implementation; should be called under execution
+ *          queue lock.
  */
 static int __bnx2x_vlan_mac_h_write_trylock(struct bnx2x *bp,
 					    struct bnx2x_vlan_mac_obj *o)
@@ -445,7 +447,7 @@ static int __bnx2x_vlan_mac_h_write_trylock(struct bnx2x *bp,
  * @bp:		device handle
  * @o:		vlan_mac object
  *
- * @details Should be called under execution queue lock; notice it might release
+ * details Should be called under execution queue lock; notice it might release
  *          and reclaim it during its run.
  */
 static void __bnx2x_vlan_mac_h_exec_pending(struct bnx2x *bp,
@@ -475,7 +477,7 @@ static void __bnx2x_vlan_mac_h_exec_pending(struct bnx2x *bp,
  * @o:			vlan_mac object
  * @ramrod_flags:	ramrod flags of missed execution
  *
- * @details Should be called under execution queue lock.
+ * Context: Should be called under execution queue lock.
  */
 static void __bnx2x_vlan_mac_h_pend(struct bnx2x *bp,
 				    struct bnx2x_vlan_mac_obj *o,
@@ -493,7 +495,7 @@ static void __bnx2x_vlan_mac_h_pend(struct bnx2x *bp,
  * @bp:			device handle
  * @o:			vlan_mac object
  *
- * @details Should be called under execution queue lock. Notice if a pending
+ * Context: Should be called under execution queue lock. Notice if a pending
  *          execution exists, it would perform it - possibly releasing and
  *          reclaiming the execution queue lock.
  */
@@ -516,7 +518,7 @@ static void __bnx2x_vlan_mac_h_write_unlock(struct bnx2x *bp,
  * @bp:			device handle
  * @o:			vlan_mac object
  *
- * @details Should be called under the execution queue lock. May sleep. May
+ * Context: Should be called under the execution queue lock. May sleep. May
  *          release and reclaim execution queue lock during its run.
  */
 static int __bnx2x_vlan_mac_h_read_lock(struct bnx2x *bp,
@@ -536,7 +538,7 @@ static int __bnx2x_vlan_mac_h_read_lock(struct bnx2x *bp,
  * @bp:			device handle
  * @o:			vlan_mac object
  *
- * @details May sleep. Claims and releases execution queue lock during its run.
+ * Context: May sleep. Claims and releases execution queue lock during its run.
  */
 int bnx2x_vlan_mac_h_read_lock(struct bnx2x *bp,
 			       struct bnx2x_vlan_mac_obj *o)
@@ -556,7 +558,7 @@ int bnx2x_vlan_mac_h_read_lock(struct bnx2x *bp,
  * @bp:			device handle
  * @o:			vlan_mac object
  *
- * @details Should be called under execution queue lock. Notice if a pending
+ * Context: Should be called under execution queue lock. Notice if a pending
  *          execution exists, it would be performed if this was the last
  *          reader. possibly releasing and reclaiming the execution queue lock.
  */
@@ -591,7 +593,7 @@ static void __bnx2x_vlan_mac_h_read_unlock(struct bnx2x *bp,
  * @bp:			device handle
  * @o:			vlan_mac object
  *
- * @details Notice if a pending execution exists, it would be performed if this
+ * Context: Notice if a pending execution exists, it would be performed if this
  *          was the last reader. Claims and releases the execution queue lock
  *          during its run.
  */
@@ -888,7 +890,7 @@ static void bnx2x_set_one_mac_e2(struct bnx2x *bp,
 		(struct eth_classify_rules_ramrod_data *)(raw->rdata);
 	int rule_cnt = rule_idx + 1, cmd = elem->cmd_data.vlan_mac.cmd;
 	union eth_classify_rule_cmd *rule_entry = &data->rules[rule_idx];
-	bool add = (cmd == BNX2X_VLAN_MAC_ADD) ? true : false;
+	bool add = cmd == BNX2X_VLAN_MAC_ADD;
 	unsigned long *vlan_mac_flags = &elem->cmd_data.vlan_mac.vlan_mac_flags;
 	u8 *mac = elem->cmd_data.vlan_mac.u.mac.mac;
 
@@ -968,7 +970,7 @@ static void bnx2x_set_one_mac_e2(struct bnx2x *bp,
  *
  * @bp:		device handle
  * @o:		queue
- * @type:
+ * @type:	the type of echo
  * @cam_offset:	offset in cam memory
  * @hdr:	pointer to a header to setup
  *
@@ -1073,7 +1075,7 @@ static void bnx2x_set_one_vlan_e2(struct bnx2x *bp,
 	int rule_cnt = rule_idx + 1;
 	union eth_classify_rule_cmd *rule_entry = &data->rules[rule_idx];
 	enum bnx2x_vlan_mac_cmd cmd = elem->cmd_data.vlan_mac.cmd;
-	bool add = (cmd == BNX2X_VLAN_MAC_ADD) ? true : false;
+	bool add = cmd == BNX2X_VLAN_MAC_ADD;
 	u16 vlan = elem->cmd_data.vlan_mac.u.vlan.vlan;
 
 	/* Reset the ramrod data buffer for the first rule */
@@ -1123,7 +1125,7 @@ static void bnx2x_set_one_vlan_mac_e2(struct bnx2x *bp,
 	int rule_cnt = rule_idx + 1;
 	union eth_classify_rule_cmd *rule_entry = &data->rules[rule_idx];
 	enum bnx2x_vlan_mac_cmd cmd = elem->cmd_data.vlan_mac.cmd;
-	bool add = (cmd == BNX2X_VLAN_MAC_ADD) ? true : false;
+	bool add = cmd == BNX2X_VLAN_MAC_ADD;
 	u16 vlan = elem->cmd_data.vlan_mac.u.vlan_mac.vlan;
 	u8 *mac = elem->cmd_data.vlan_mac.u.vlan_mac.mac;
 	u16 inner_mac;
@@ -1608,8 +1610,8 @@ static int __bnx2x_vlan_mac_execute_step(struct bnx2x *bp,
  *
  * @bp:		device handle
  * @o:		bnx2x_vlan_mac_obj
- * @cqe:
- * @cont:	if true schedule next execution chunk
+ * @cqe:	completion element
+ * @ramrod_flags: if set schedule next execution chunk
  *
  */
 static int bnx2x_complete_vlan_mac(struct bnx2x *bp,
@@ -1656,7 +1658,7 @@ static int bnx2x_complete_vlan_mac(struct bnx2x *bp,
  * bnx2x_optimize_vlan_mac - optimize ADD and DEL commands.
  *
  * @bp:		device handle
- * @o:		bnx2x_qable_obj
+ * @qo:		bnx2x_qable_obj
  * @elem:	bnx2x_exeq_elem
  */
 static int bnx2x_optimize_vlan_mac(struct bnx2x *bp,
@@ -1714,10 +1716,10 @@ static int bnx2x_optimize_vlan_mac(struct bnx2x *bp,
  * bnx2x_vlan_mac_get_registry_elem - prepare a registry element
  *
  * @bp:	  device handle
- * @o:
- * @elem:
- * @restore:
- * @re:
+ * @o:	vlan object
+ * @elem: element
+ * @restore: to restore or not
+ * @re: registry
  *
  * prepare a registry element according to the current command request.
  */
@@ -1768,9 +1770,9 @@ static inline int bnx2x_vlan_mac_get_registry_elem(
  * bnx2x_execute_vlan_mac - execute vlan mac command
  *
  * @bp:			device handle
- * @qo:
- * @exe_chunk:
- * @ramrod_flags:
+ * @qo:			bnx2x_qable_obj pointer
+ * @exe_chunk:		chunk
+ * @ramrod_flags:	flags
  *
  * go and send a ramrod!
  */
@@ -2006,8 +2008,8 @@ int bnx2x_config_vlan_mac(struct bnx2x *bp,
  * bnx2x_vlan_mac_del_all - delete elements with given vlan_mac_flags spec
  *
  * @bp:			device handle
- * @o:
- * @vlan_mac_flags:
+ * @o:			vlan object info
+ * @vlan_mac_flags:	vlan flags
  * @ramrod_flags:	execution flags to be used for this deletion
  *
  * if the last operation has completed successfully and there are no
@@ -2591,7 +2593,7 @@ void bnx2x_init_rx_mode_obj(struct bnx2x *bp,
 /********************* Multicast verbs: SET, CLEAR ****************************/
 static inline u8 bnx2x_mcast_bin_from_mac(u8 *mac)
 {
-	return (crc32c_le(0, mac, ETH_ALEN) >> 24) & 0xff;
+	return (crc32c(0, mac, ETH_ALEN) >> 24) & 0xff;
 }
 
 struct bnx2x_mcast_mac_elem {
@@ -2767,7 +2769,7 @@ static int bnx2x_mcast_enqueue_cmd(struct bnx2x *bp,
 /**
  * bnx2x_mcast_get_next_bin - get the next set bin (index)
  *
- * @o:
+ * @o:		multicast object info
  * @last:	index to start looking from (including)
  *
  * returns the next found (set) bin or a negative value if none is found.
@@ -2892,7 +2894,7 @@ static void bnx2x_mcast_set_one_rule_e2(struct bnx2x *bp,
  * bnx2x_mcast_handle_restore_cmd_e2 - restore configuration from the registry
  *
  * @bp:		device handle
- * @o:
+ * @o:		multicast object info
  * @start_bin:	index in the registry to start from (including)
  * @rdata_idx:	index in the ramrod data to start from
  *
@@ -2977,8 +2979,8 @@ static inline void bnx2x_mcast_hdl_pending_del_e2(struct bnx2x *bp,
 
 		cmd_pos->data.macs_num--;
 
-		  DP(BNX2X_MSG_SP, "Deleting MAC. %d left,cnt is %d\n",
-				   cmd_pos->data.macs_num, cnt);
+		DP(BNX2X_MSG_SP, "Deleting MAC. %d left,cnt is %d\n",
+		   cmd_pos->data.macs_num, cnt);
 
 		/* Break if we reached the maximum
 		 * number of rules.
@@ -3202,11 +3204,11 @@ static inline void bnx2x_mcast_hdl_del(struct bnx2x *bp,
 }
 
 /**
- * bnx2x_mcast_handle_current_cmd -
+ * bnx2x_mcast_handle_current_cmd - send command if room
  *
  * @bp:		device handle
- * @p:
- * @cmd:
+ * @p:		ramrod mcast info
+ * @cmd:	command
  * @start_cnt:	first line in the ramrod data that may be used
  *
  * This function is called iff there is enough place for the current command in
@@ -3258,7 +3260,7 @@ static int bnx2x_mcast_validate_e2(struct bnx2x *bp,
 	/* DEL command deletes all currently configured MACs */
 	case BNX2X_MCAST_CMD_DEL:
 		o->set_registry_size(o, 0);
-		/* Don't break */
+		fallthrough;
 
 	/* RESTORE command will restore the entire multicast configuration */
 	case BNX2X_MCAST_CMD_RESTORE:
@@ -3323,7 +3325,7 @@ static void bnx2x_mcast_revert_e2(struct bnx2x *bp,
  * bnx2x_mcast_set_rdata_hdr_e2 - sets a header values
  *
  * @bp:		device handle
- * @p:
+ * @p:		ramrod parameters
  * @len:	number of rules to handle
  */
 static inline void bnx2x_mcast_set_rdata_hdr_e2(struct bnx2x *bp,
@@ -3592,13 +3594,13 @@ static int bnx2x_mcast_validate_e1(struct bnx2x *bp,
 	/* DEL command deletes all currently configured MACs */
 	case BNX2X_MCAST_CMD_DEL:
 		o->set_registry_size(o, 0);
-		/* Don't break */
+		fallthrough;
 
 	/* RESTORE command will restore the entire multicast configuration */
 	case BNX2X_MCAST_CMD_RESTORE:
 		p->mcast_list_len = reg_sz;
-		  DP(BNX2X_MSG_SP, "Command %d, p->mcast_list_len=%d\n",
-				   cmd, p->mcast_list_len);
+		DP(BNX2X_MSG_SP, "Command %d, p->mcast_list_len=%d\n",
+		   cmd, p->mcast_list_len);
 		break;
 
 	case BNX2X_MCAST_CMD_ADD:
@@ -3684,7 +3686,7 @@ static void bnx2x_mcast_set_one_rule_e1(struct bnx2x *bp,
  * bnx2x_mcast_set_rdata_hdr_e1  - set header values in mac_configuration_cmd
  *
  * @bp:		device handle
- * @p:
+ * @p:		ramrod parameters
  * @len:	number of rules to handle
  */
 static inline void bnx2x_mcast_set_rdata_hdr_e1(struct bnx2x *bp,
@@ -3711,7 +3713,7 @@ static inline void bnx2x_mcast_set_rdata_hdr_e1(struct bnx2x *bp,
  * bnx2x_mcast_handle_restore_cmd_e1 - restore command for 57710
  *
  * @bp:		device handle
- * @o:
+ * @o:		multicast info
  * @start_idx:	index in the registry to start from
  * @rdata_idx:	index in the ramrod data to start from
  *
@@ -3735,8 +3737,8 @@ static inline int bnx2x_mcast_handle_restore_cmd_e1(
 
 		i++;
 
-		  DP(BNX2X_MSG_SP, "About to configure %pM mcast MAC\n",
-		     cfg_data.mac);
+		DP(BNX2X_MSG_SP, "About to configure %pM mcast MAC\n",
+		   cfg_data.mac);
 	}
 
 	*rdata_idx = i;
@@ -3798,10 +3800,10 @@ static inline int bnx2x_mcast_handle_pending_cmds_e1(
 /**
  * bnx2x_get_fw_mac_addr - revert the bnx2x_set_fw_mac_addr().
  *
- * @fw_hi:
- * @fw_mid:
- * @fw_lo:
- * @mac:
+ * @fw_hi: address
+ * @fw_mid: address
+ * @fw_lo: address
+ * @mac: mac address
  */
 static inline void bnx2x_get_fw_mac_addr(__le16 *fw_hi, __le16 *fw_mid,
 					 __le16 *fw_lo, u8 *mac)
@@ -3818,7 +3820,7 @@ static inline void bnx2x_get_fw_mac_addr(__le16 *fw_hi, __le16 *fw_mid,
  * bnx2x_mcast_refresh_registry_e1 -
  *
  * @bp:		device handle
- * @cnt:
+ * @o:		multicast info
  *
  * Check the ramrod data first entry flag to see if it's a DELETE or ADD command
  * and update the registry correspondingly: if ADD - allocate a memory and add
@@ -4150,7 +4152,7 @@ void bnx2x_init_mcast_obj(struct bnx2x *bp,
 /*************************** Credit handling **********************************/
 
 /**
- * atomic_add_ifless - add if the result is less than a given value.
+ * __atomic_add_ifless - add if the result is less than a given value.
  *
  * @v:	pointer of type atomic_t
  * @a:	the amount to add to v...
@@ -4178,7 +4180,7 @@ static inline bool __atomic_add_ifless(atomic_t *v, int a, int u)
 }
 
 /**
- * atomic_dec_ifmoe - dec if the result is more or equal than a given value.
+ * __atomic_dec_ifmoe - dec if the result is more or equal than a given value.
  *
  * @v:	pointer of type atomic_t
  * @a:	the amount to dec from v...
@@ -4311,7 +4313,7 @@ static bool bnx2x_credit_pool_get_entry_always_true(
 /**
  * bnx2x_init_credit_pool - initialize credit pool internals.
  *
- * @p:
+ * @p:		credit pool
  * @base:	Base entry in the CAM to use.
  * @credit:	pool size.
  *
@@ -4725,8 +4727,8 @@ static int bnx2x_queue_wait_comp(struct bnx2x *bp,
  * bnx2x_queue_comp_cmd - complete the state change command.
  *
  * @bp:		device handle
- * @o:
- * @cmd:
+ * @o:		queue info
+ * @cmd:	command to exec
  *
  * Checks that the arrived completion is expected.
  */
@@ -5039,7 +5041,6 @@ static inline int bnx2x_q_init(struct bnx2x *bp,
 	/* As no ramrod is sent, complete the command immediately  */
 	o->complete_cmd(bp, o, BNX2X_Q_CMD_INIT);
 
-	mmiowb();
 	smp_mb();
 
 	return 0;
@@ -5478,8 +5479,8 @@ static int bnx2x_queue_send_cmd_e2(struct bnx2x *bp,
  * bnx2x_queue_chk_transition - check state machine of a regular Queue
  *
  * @bp:		device handle
- * @o:
- * @params:
+ * @o:		queue info
+ * @params:	queue state
  *
  * (not Forwarding)
  * It both checks if the requested command is legal in a current
@@ -5736,8 +5737,8 @@ static int bnx2x_func_wait_comp(struct bnx2x *bp,
  * bnx2x_func_state_change_comp - complete the state machine transition
  *
  * @bp:		device handle
- * @o:
- * @cmd:
+ * @o:		function info
+ * @cmd:	more info
  *
  * Called on state change transition. Completes the state
  * machine transition only - no HW interaction.
@@ -5777,8 +5778,8 @@ static inline int bnx2x_func_state_change_comp(struct bnx2x *bp,
  * bnx2x_func_comp_cmd - complete the state change command
  *
  * @bp:		device handle
- * @o:
- * @cmd:
+ * @o:		function info
+ * @cmd:	more info
  *
  * Checks that the arrived completion is expected.
  */
@@ -5797,8 +5798,8 @@ static int bnx2x_func_comp_cmd(struct bnx2x *bp,
  * bnx2x_func_chk_transition - perform function state machine transition
  *
  * @bp:		device handle
- * @o:
- * @params:
+ * @o:		function info
+ * @params:	state parameters
  *
  * It both checks if the requested command is legal in a current
  * state and, if it's legal, sets a `next_state' in the object
@@ -6149,6 +6150,7 @@ static inline int bnx2x_func_send_start(struct bnx2x *bp,
 	rdata->sd_vlan_tag	= cpu_to_le16(start_params->sd_vlan_tag);
 	rdata->path_id		= BP_PATH(bp);
 	rdata->network_cos_mode	= start_params->network_cos_mode;
+	rdata->dmae_cmd_id	= BNX2X_FW_DMAE_C;
 
 	rdata->vxlan_dst_port	= cpu_to_le16(start_params->vxlan_dst_port);
 	rdata->geneve_dst_port	= cpu_to_le16(start_params->geneve_dst_port);
diff --git a/drivers/net/ethernet/broadcom/bnx2x/bnx2x_sp.h b/drivers/net/ethernet/broadcom/bnx2x/bnx2x_sp.h
index 0bf2fd470819..00ca861c80dd 100644
--- a/drivers/net/ethernet/broadcom/bnx2x/bnx2x_sp.h
+++ b/drivers/net/ethernet/broadcom/bnx2x/bnx2x_sp.h
@@ -265,6 +265,7 @@ enum {
 	BNX2X_ETH_MAC,
 	BNX2X_ISCSI_ETH_MAC,
 	BNX2X_NETQ_ETH_MAC,
+	BNX2X_VLAN,
 	BNX2X_DONT_CONSUME_CAM_CREDIT,
 	BNX2X_DONT_CONSUME_CAM_CREDIT_DEST,
 };
@@ -272,7 +273,8 @@ enum {
 #define BNX2X_VLAN_MAC_CMP_MASK	(1 << BNX2X_UC_LIST_MAC | \
 				 1 << BNX2X_ETH_MAC | \
 				 1 << BNX2X_ISCSI_ETH_MAC | \
-				 1 << BNX2X_NETQ_ETH_MAC)
+				 1 << BNX2X_NETQ_ETH_MAC | \
+				 1 << BNX2X_VLAN)
 #define BNX2X_VLAN_MAC_CMP_FLAGS(flags) \
 	((flags) & BNX2X_VLAN_MAC_CMP_MASK)
 
@@ -377,7 +379,7 @@ struct bnx2x_vlan_mac_obj {
 	/**
 	*  Delete all configured elements having the given
 	*  vlan_mac_flags specification. Assumes no pending for
-	*  execution commands. Will schedule all all currently
+	*  execution commands. Will schedule all currently
 	*  configured MACs/VLANs/VLAN-MACs matching the vlan_mac_flags
 	*  specification for deletion and will use the given
 	*  ramrod_flags for the last DEL operation.
@@ -1534,8 +1536,11 @@ void bnx2x_get_rss_ind_table(struct bnx2x_rss_config_obj *rss_obj,
 	((MAX_MAC_CREDIT_E2 - GET_NUM_VFS_PER_PATH(bp) * VF_MAC_CREDIT_CNT) / \
 	 func_num + GET_NUM_VFS_PER_PF(bp) * VF_MAC_CREDIT_CNT)
 
+#define BNX2X_VFS_VLAN_CREDIT(bp)	\
+	(GET_NUM_VFS_PER_PATH(bp) * VF_VLAN_CREDIT_CNT)
+
 #define PF_VLAN_CREDIT_E2(bp, func_num)					 \
-	((MAX_MAC_CREDIT_E2 - GET_NUM_VFS_PER_PATH(bp) * VF_VLAN_CREDIT_CNT) / \
+	((MAX_VLAN_CREDIT_E2 - 1 - BNX2X_VFS_VLAN_CREDIT(bp)) /	\
 	 func_num + GET_NUM_VFS_PER_PF(bp) * VF_VLAN_CREDIT_CNT)
 
 #endif /* BNX2X_SP_VERBS */
diff --git a/drivers/net/ethernet/broadcom/bnx2x/bnx2x_sriov.c b/drivers/net/ethernet/broadcom/bnx2x/bnx2x_sriov.c
index bdfd53b46bc5..12198fc3ab22 100644
--- a/drivers/net/ethernet/broadcom/bnx2x/bnx2x_sriov.c
+++ b/drivers/net/ethernet/broadcom/bnx2x/bnx2x_sriov.c
@@ -100,13 +100,11 @@ static void bnx2x_vf_igu_ack_sb(struct bnx2x *bp, struct bnx2x_virtf *vf,
 	DP(NETIF_MSG_HW, "write 0x%08x to IGU(via GRC) addr 0x%x\n",
 	   cmd_data.sb_id_and_flags, igu_addr_data);
 	REG_WR(bp, igu_addr_data, cmd_data.sb_id_and_flags);
-	mmiowb();
 	barrier();
 
 	DP(NETIF_MSG_HW, "write 0x%08x to IGU(via GRC) addr 0x%x\n",
 	   ctl, igu_addr_ctl);
 	REG_WR(bp, igu_addr_ctl, ctl);
-	mmiowb();
 	barrier();
 }
 
@@ -209,7 +207,10 @@ void bnx2x_vfop_qctor_prep(struct bnx2x *bp,
 	 */
 	__set_bit(BNX2X_Q_FLG_TX_SWITCH, &setup_p->flags);
 	__set_bit(BNX2X_Q_FLG_TX_SEC, &setup_p->flags);
-	__set_bit(BNX2X_Q_FLG_ANTI_SPOOF, &setup_p->flags);
+	if (vf->spoofchk)
+		__set_bit(BNX2X_Q_FLG_ANTI_SPOOF, &setup_p->flags);
+	else
+		__clear_bit(BNX2X_Q_FLG_ANTI_SPOOF, &setup_p->flags);
 
 	/* Setup-op rx parameters */
 	if (test_bit(BNX2X_Q_TYPE_HAS_RX, &q_type)) {
@@ -330,27 +331,6 @@ bnx2x_vf_set_igu_info(struct bnx2x *bp, u8 igu_sb_id, u8 abs_vfid)
 	BP_VFDB(bp)->vf_sbs_pool++;
 }
 
-static inline void bnx2x_vf_vlan_credit(struct bnx2x *bp,
-					struct bnx2x_vlan_mac_obj *obj,
-					atomic_t *counter)
-{
-	struct list_head *pos;
-	int read_lock;
-	int cnt = 0;
-
-	read_lock = bnx2x_vlan_mac_h_read_lock(bp, obj);
-	if (read_lock)
-		DP(BNX2X_MSG_SP, "Failed to take vlan mac read head; continuing anyway\n");
-
-	list_for_each(pos, &obj->head)
-		cnt++;
-
-	if (!read_lock)
-		bnx2x_vlan_mac_h_read_unlock(bp, obj);
-
-	atomic_set(counter, cnt);
-}
-
 static int bnx2x_vf_vlan_mac_clear(struct bnx2x *bp, struct bnx2x_virtf *vf,
 				   int qid, bool drv_only, int type)
 {
@@ -571,7 +551,7 @@ int bnx2x_vf_mcast(struct bnx2x *bp, struct bnx2x_virtf *vf,
 	else
 		set_bit(RAMROD_COMP_WAIT, &mcast.ramrod_flags);
 	if (mc_num) {
-		mc = kzalloc(mc_num * sizeof(struct bnx2x_mcast_list_elem),
+		mc = kcalloc(mc_num, sizeof(struct bnx2x_mcast_list_elem),
 			     GFP_KERNEL);
 		if (!mc) {
 			BNX2X_ERR("Cannot Configure multicasts due to lack of memory\n");
@@ -778,9 +758,18 @@ static void bnx2x_vf_igu_reset(struct bnx2x *bp, struct bnx2x_virtf *vf)
 
 void bnx2x_vf_enable_access(struct bnx2x *bp, u8 abs_vfid)
 {
+	u16 abs_fid;
+
+	abs_fid = FW_VF_HANDLE(abs_vfid);
+
 	/* set the VF-PF association in the FW */
-	storm_memset_vf_to_pf(bp, FW_VF_HANDLE(abs_vfid), BP_FUNC(bp));
-	storm_memset_func_en(bp, FW_VF_HANDLE(abs_vfid), 1);
+	storm_memset_vf_to_pf(bp, abs_fid, BP_FUNC(bp));
+	storm_memset_func_en(bp, abs_fid, 1);
+
+	/* Invalidate fp_hsi version for vfs */
+	if (bp->fw_cap & FW_CAP_INVALIDATE_VF_FP_HSI)
+		REG_WR8(bp, BAR_XSTRORM_INTMEM +
+			    XSTORM_ETH_FUNCTION_INFO_FP_HSI_VALID_E2_OFFSET(abs_fid), 0);
 
 	/* clear vf errors*/
 	bnx2x_vf_semi_clear_err(bp, abs_vfid);
@@ -806,16 +795,20 @@ static void bnx2x_vf_enable_traffic(struct bnx2x *bp, struct bnx2x_virtf *vf)
 
 static u8 bnx2x_vf_is_pcie_pending(struct bnx2x *bp, u8 abs_vfid)
 {
-	struct pci_dev *dev;
 	struct bnx2x_virtf *vf = bnx2x_vf_by_abs_fid(bp, abs_vfid);
+	struct pci_dev *dev;
+	bool pending;
 
 	if (!vf)
 		return false;
 
-	dev = pci_get_bus_and_slot(vf->bus, vf->devfn);
-	if (dev)
-		return bnx2x_is_pcie_pending(dev);
-	return false;
+	dev = pci_get_domain_bus_and_slot(vf->domain, vf->bus, vf->devfn);
+	if (!dev)
+		return false;
+	pending = bnx2x_is_pcie_pending(dev);
+	pci_dev_put(dev);
+
+	return pending;
 }
 
 int bnx2x_vf_flr_clnup_epilog(struct bnx2x *bp, u8 abs_vfid)
@@ -901,6 +894,8 @@ static void bnx2x_vf_flr(struct bnx2x *bp, struct bnx2x_virtf *vf)
 	/* release VF resources */
 	bnx2x_vf_free_resc(bp, vf);
 
+	vf->malicious = false;
+
 	/* re-open the mailbox */
 	bnx2x_vf_enable_mbx(bp, vf->abs_vfid);
 	return;
@@ -1039,6 +1034,13 @@ void bnx2x_iov_init_dmae(struct bnx2x *bp)
 		REG_WR(bp, DMAE_REG_BACKWARD_COMP_EN, 0);
 }
 
+static int bnx2x_vf_domain(struct bnx2x *bp, int vfid)
+{
+	struct pci_dev *dev = bp->pdev;
+
+	return pci_domain_nr(dev->bus);
+}
+
 static int bnx2x_vf_bus(struct bnx2x *bp, int vfid)
 {
 	struct pci_dev *dev = bp->pdev;
@@ -1072,11 +1074,6 @@ static void bnx2x_vf_set_bars(struct bnx2x *bp, struct bnx2x_virtf *vf)
 	}
 }
 
-static int bnx2x_ari_enabled(struct pci_dev *dev)
-{
-	return dev->bus->self && dev->bus->self->ari_enabled;
-}
-
 static int
 bnx2x_get_vf_igu_cam_info(struct bnx2x *bp)
 {
@@ -1208,9 +1205,8 @@ int bnx2x_iov_init_one(struct bnx2x *bp, int int_mode_param,
 		return 0;
 	}
 
-	err = -EIO;
 	/* verify ari is enabled */
-	if (!bnx2x_ari_enabled(bp->pdev)) {
+	if (!pci_ari_enabled(bp->pdev->bus)) {
 		BNX2X_ERR("ARI not supported (check pci bridge ARI forwarding), SRIOV can not be enabled\n");
 		return 0;
 	}
@@ -1240,8 +1236,10 @@ int bnx2x_iov_init_one(struct bnx2x *bp, int int_mode_param,
 		goto failed;
 
 	/* SR-IOV capability was enabled but there are no VFs*/
-	if (iov->total == 0)
+	if (iov->total == 0) {
+		err = 0;
 		goto failed;
+	}
 
 	iov->nr_virtfn = min_t(u16, iov->total, num_vfs_param);
 
@@ -1249,8 +1247,9 @@ int bnx2x_iov_init_one(struct bnx2x *bp, int int_mode_param,
 	   num_vfs_param, iov->nr_virtfn);
 
 	/* allocate the vf array */
-	bp->vfdb->vfs = kzalloc(sizeof(struct bnx2x_virtf) *
-				BNX2X_NR_VIRTFN(bp), GFP_KERNEL);
+	bp->vfdb->vfs = kcalloc(BNX2X_NR_VIRTFN(bp),
+				sizeof(struct bnx2x_virtf),
+				GFP_KERNEL);
 	if (!bp->vfdb->vfs) {
 		BNX2X_ERR("failed to allocate vf array\n");
 		err = -ENOMEM;
@@ -1264,6 +1263,8 @@ int bnx2x_iov_init_one(struct bnx2x *bp, int int_mode_param,
 		bnx2x_vf(bp, i, state) = VF_FREE;
 		mutex_init(&bnx2x_vf(bp, i, op_mutex));
 		bnx2x_vf(bp, i, op_current) = CHANNEL_TLV_NONE;
+		/* enable spoofchk by default */
+		bnx2x_vf(bp, i, spoofchk) = 1;
 	}
 
 	/* re-read the IGU CAM for VFs - index and abs_vfid must be set */
@@ -1274,9 +1275,9 @@ int bnx2x_iov_init_one(struct bnx2x *bp, int int_mode_param,
 	}
 
 	/* allocate the queue arrays for all VFs */
-	bp->vfdb->vfqs = kzalloc(
-		BNX2X_MAX_NUM_VF_QUEUES * sizeof(struct bnx2x_vf_queue),
-		GFP_KERNEL);
+	bp->vfdb->vfqs = kcalloc(BNX2X_MAX_NUM_VF_QUEUES,
+				 sizeof(struct bnx2x_vf_queue),
+				 GFP_KERNEL);
 
 	if (!bp->vfdb->vfqs) {
 		BNX2X_ERR("failed to allocate vf queue array\n");
@@ -1609,6 +1610,7 @@ int bnx2x_iov_nic_init(struct bnx2x *bp)
 		struct bnx2x_virtf *vf = BP_VF(bp, vfid);
 
 		/* fill in the BDF and bars */
+		vf->domain = bnx2x_vf_domain(bp, vfid);
 		vf->bus = bnx2x_vf_bus(bp, vfid);
 		vf->devfn = bnx2x_vf_devfn(bp, vfid);
 		bnx2x_vf_set_bars(bp, vf);
@@ -1821,10 +1823,13 @@ get_vf:
 		DP(BNX2X_MSG_IOV, "got VF [%d:%d] RSS update ramrod\n",
 		   vf->abs_vfid, qidx);
 		bnx2x_vf_handle_rss_update_eqe(bp, vf);
+		fallthrough;
 	case EVENT_RING_OPCODE_VF_FLR:
-	case EVENT_RING_OPCODE_MALICIOUS_VF:
 		/* Do nothing for now */
 		return 0;
+	case EVENT_RING_OPCODE_MALICIOUS_VF:
+		vf->malicious = true;
+		return 0;
 	}
 
 	return 0;
@@ -1866,7 +1871,6 @@ void bnx2x_iov_adjust_stats_req(struct bnx2x *bp)
 {
 	int i;
 	int first_queue_query_index, num_queues_req;
-	dma_addr_t cur_data_offset;
 	struct stats_query_entry *cur_query_entry;
 	u8 stats_count = 0;
 	bool is_fcoe = false;
@@ -1887,10 +1891,6 @@ void bnx2x_iov_adjust_stats_req(struct bnx2x *bp)
 	       BNX2X_NUM_ETH_QUEUES(bp), is_fcoe, first_queue_query_index,
 	       first_queue_query_index + num_queues_req);
 
-	cur_data_offset = bp->fw_stats_data_mapping +
-		offsetof(struct bnx2x_fw_stats_data, queue_stats) +
-		num_queues_req * sizeof(struct per_queue_stats);
-
 	cur_query_entry = &bp->fw_stats_req->
 		query[first_queue_query_index + num_queues_req];
 
@@ -1905,6 +1905,13 @@ void bnx2x_iov_adjust_stats_req(struct bnx2x *bp)
 			continue;
 		}
 
+		if (vf->malicious) {
+			DP_AND((BNX2X_MSG_IOV | BNX2X_MSG_STATS),
+			       "vf %d malicious so no stats for it\n",
+			       vf->abs_vfid);
+			continue;
+		}
+
 		DP_AND((BNX2X_MSG_IOV | BNX2X_MSG_STATS),
 		       "add addresses for vf %d\n", vf->abs_vfid);
 		for_each_vfq(vf, j) {
@@ -1934,7 +1941,6 @@ void bnx2x_iov_adjust_stats_req(struct bnx2x *bp)
 			       cur_query_entry->funcID,
 			       j, cur_query_entry->index);
 			cur_query_entry++;
-			cur_data_offset += sizeof(struct per_queue_stats);
 			stats_count++;
 
 			/* all stats are coalesced to the leading queue */
@@ -2209,7 +2215,7 @@ int bnx2x_vf_free(struct bnx2x *bp, struct bnx2x_virtf *vf)
 		rc = bnx2x_vf_close(bp, vf);
 		if (rc)
 			goto op_err;
-		/* Fallthrough to release resources */
+		fallthrough;	/* to release resources */
 	case VF_ACQUIRED:
 		DP(BNX2X_MSG_IOV, "about to free resources\n");
 		bnx2x_vf_free_resc(bp, vf);
@@ -2378,15 +2384,21 @@ static int bnx2x_set_pf_tx_switching(struct bnx2x *bp, bool enable)
 	/* send the ramrod on all the queues of the PF */
 	for_each_eth_queue(bp, i) {
 		struct bnx2x_fastpath *fp = &bp->fp[i];
+		int tx_idx;
 
 		/* Set the appropriate Queue object */
 		q_params.q_obj = &bnx2x_sp_obj(bp, fp).q_obj;
 
-		/* Update the Queue state */
-		rc = bnx2x_queue_state_change(bp, &q_params);
-		if (rc) {
-			BNX2X_ERR("Failed to configure Tx switching\n");
-			return rc;
+		for (tx_idx = FIRST_TX_COS_INDEX;
+		     tx_idx < fp->max_cos; tx_idx++) {
+			q_params.params.update.cid_index = tx_idx;
+
+			/* Update the Queue state */
+			rc = bnx2x_queue_state_change(bp, &q_params);
+			if (rc) {
+				BNX2X_ERR("Failed to configure Tx switching\n");
+				return rc;
+			}
 		}
 	}
 
@@ -2616,7 +2628,8 @@ int bnx2x_get_vf_config(struct net_device *dev, int vfidx,
 	ivi->qos = 0;
 	ivi->max_tx_rate = 10000; /* always 10G. TBA take from link struct */
 	ivi->min_tx_rate = 0;
-	ivi->spoofchk = 1; /*always enabled */
+	ivi->spoofchk = vf->spoofchk ? 1 : 0;
+	ivi->linkstate = vf->link_cfg;
 	if (vf->state == VF_ENABLED) {
 		/* mac and vlan are in vlan_mac objects */
 		if (bnx2x_validate_vf_sp_objs(bp, vf, false)) {
@@ -2639,10 +2652,10 @@ int bnx2x_get_vf_config(struct net_device *dev, int vfidx,
 		/* vlan */
 		if (bulletin->valid_bitmap & (1 << VLAN_VALID))
 			/* vlan configured by ndo so its in bulletin board */
-			memcpy(&ivi->vlan, &bulletin->vlan, VLAN_HLEN);
+			ivi->vlan = bulletin->vlan;
 		else
 			/* function has not been loaded yet. Show vlans as 0s */
-			memset(&ivi->vlan, 0, VLAN_HLEN);
+			ivi->vlan = 0;
 
 		mutex_unlock(&bp->vfdb->bulletin_mutex);
 	}
@@ -2934,6 +2947,77 @@ out:
 	return rc;
 }
 
+int bnx2x_set_vf_spoofchk(struct net_device *dev, int idx, bool val)
+{
+	struct bnx2x *bp = netdev_priv(dev);
+	struct bnx2x_virtf *vf;
+	int i, rc = 0;
+
+	vf = BP_VF(bp, idx);
+	if (!vf)
+		return -EINVAL;
+
+	/* nothing to do */
+	if (vf->spoofchk == val)
+		return 0;
+
+	vf->spoofchk = val ? 1 : 0;
+
+	DP(BNX2X_MSG_IOV, "%s spoofchk for VF %d\n",
+	   val ? "enabling" : "disabling", idx);
+
+	/* is vf initialized and queue set up? */
+	if (vf->state != VF_ENABLED ||
+	    bnx2x_get_q_logical_state(bp, &bnx2x_leading_vfq(vf, sp_obj)) !=
+	    BNX2X_Q_LOGICAL_STATE_ACTIVE)
+		return rc;
+
+	/* User should be able to see error in system logs */
+	if (!bnx2x_validate_vf_sp_objs(bp, vf, true))
+		return -EINVAL;
+
+	/* send queue update ramrods to configure spoofchk */
+	for_each_vfq(vf, i) {
+		struct bnx2x_queue_state_params q_params = {NULL};
+		struct bnx2x_queue_update_params *update_params;
+
+		q_params.q_obj = &bnx2x_vfq(vf, i, sp_obj);
+
+		/* validate the Q is UP */
+		if (bnx2x_get_q_logical_state(bp, q_params.q_obj) !=
+		    BNX2X_Q_LOGICAL_STATE_ACTIVE)
+			continue;
+
+		__set_bit(RAMROD_COMP_WAIT, &q_params.ramrod_flags);
+		q_params.cmd = BNX2X_Q_CMD_UPDATE;
+		update_params = &q_params.params.update;
+		__set_bit(BNX2X_Q_UPDATE_ANTI_SPOOF_CHNG,
+			  &update_params->update_flags);
+		if (val) {
+			__set_bit(BNX2X_Q_UPDATE_ANTI_SPOOF,
+				  &update_params->update_flags);
+		} else {
+			__clear_bit(BNX2X_Q_UPDATE_ANTI_SPOOF,
+				    &update_params->update_flags);
+		}
+
+		/* Update the Queue state */
+		rc = bnx2x_queue_state_change(bp, &q_params);
+		if (rc) {
+			BNX2X_ERR("Failed to %s spoofchk on VF %d - vfq %d\n",
+				  val ? "enable" : "disable", idx, i);
+			goto out;
+		}
+	}
+out:
+	if (!rc)
+		DP(BNX2X_MSG_IOV,
+		   "%s spoofchk for VF[%d]\n", val ? "Enabled" : "Disabled",
+		   idx);
+
+	return rc;
+}
+
 /* crc is the first field in the bulletin board. Compute the crc over the
  * entire bulletin board excluding the crc field itself. Use the length field
  * as the Bulletin Board was posted by a PF with possibly a different version
@@ -2987,7 +3071,7 @@ enum sample_bulletin_result bnx2x_sample_bulletin(struct bnx2x *bp)
 	if (bulletin->valid_bitmap & 1 << MAC_ADDR_VALID &&
 	    !ether_addr_equal(bulletin->mac, bp->old_bulletin.mac)) {
 		/* update new mac to net device */
-		memcpy(bp->dev->dev_addr, bulletin->mac, ETH_ALEN);
+		eth_hw_addr_set(bp->dev, bulletin->mac);
 	}
 
 	if (bulletin->valid_bitmap & (1 << LINK_VALID)) {
@@ -3042,7 +3126,7 @@ void bnx2x_vf_pci_dealloc(struct bnx2x *bp)
 {
 	BNX2X_PCI_FREE(bp->vf2pf_mbox, bp->vf2pf_mbox_mapping,
 		       sizeof(struct bnx2x_vf_mbx_msg));
-	BNX2X_PCI_FREE(bp->vf2pf_mbox, bp->pf2vf_bulletin_mapping,
+	BNX2X_PCI_FREE(bp->pf2vf_bulletin, bp->pf2vf_bulletin_mapping,
 		       sizeof(union pf_vf_bulletin));
 }
 
diff --git a/drivers/net/ethernet/broadcom/bnx2x/bnx2x_sriov.h b/drivers/net/ethernet/broadcom/bnx2x/bnx2x_sriov.h
index 888d0b6632e8..02a4e557e176 100644
--- a/drivers/net/ethernet/broadcom/bnx2x/bnx2x_sriov.h
+++ b/drivers/net/ethernet/broadcom/bnx2x/bnx2x_sriov.h
@@ -139,8 +139,12 @@ struct bnx2x_virtf {
 #define VF_ACQUIRED	1	/* VF acquired, but not initialized */
 #define VF_ENABLED	2	/* VF Enabled */
 #define VF_RESET	3	/* VF FLR'd, pending cleanup */
+#define VF_LOST		4	/* Recovery while VFs are loaded */
 
 	bool flr_clnup_stage;	/* true during flr cleanup */
+	bool malicious;		/* true if FW indicated so, until FLR */
+	/* 1(true) if spoof check is enabled */
+	u8 spoofchk;
 
 	/* dma */
 	dma_addr_t fw_stat_map;
@@ -181,6 +185,7 @@ struct bnx2x_virtf {
 	u32 error;	/* 0 means all's-well */
 
 	/* BDF */
+	unsigned int domain;
 	unsigned int bus;
 	unsigned int devfn;
 
@@ -499,12 +504,12 @@ enum sample_bulletin_result bnx2x_sample_bulletin(struct bnx2x *bp);
 /* VF side vfpf channel functions */
 int bnx2x_vfpf_acquire(struct bnx2x *bp, u8 tx_count, u8 rx_count);
 int bnx2x_vfpf_release(struct bnx2x *bp);
-int bnx2x_vfpf_release(struct bnx2x *bp);
 int bnx2x_vfpf_init(struct bnx2x *bp);
 void bnx2x_vfpf_close_vf(struct bnx2x *bp);
 int bnx2x_vfpf_setup_q(struct bnx2x *bp, struct bnx2x_fastpath *fp,
 		       bool is_leading);
-int bnx2x_vfpf_config_mac(struct bnx2x *bp, u8 *addr, u8 vf_qid, bool set);
+int bnx2x_vfpf_config_mac(struct bnx2x *bp, const u8 *addr, u8 vf_qid,
+			  bool set);
 int bnx2x_vfpf_config_rss(struct bnx2x *bp,
 			  struct bnx2x_config_rss_params *params);
 int bnx2x_vfpf_set_mcast(struct net_device *dev);
@@ -513,7 +518,7 @@ int bnx2x_vfpf_storm_rx_mode(struct bnx2x *bp);
 static inline void bnx2x_vf_fill_fw_str(struct bnx2x *bp, char *buf,
 					size_t buf_len)
 {
-	strlcpy(buf, bp->acquire_resp.pfdev_info.fw_ver, buf_len);
+	strscpy(buf, bp->acquire_resp.pfdev_info.fw_ver, buf_len);
 }
 
 static inline int bnx2x_vf_ustorm_prods_offset(struct bnx2x *bp,
@@ -581,7 +586,7 @@ static inline int bnx2x_vfpf_release(struct bnx2x *bp) {return 0; }
 static inline int bnx2x_vfpf_init(struct bnx2x *bp) {return 0; }
 static inline void bnx2x_vfpf_close_vf(struct bnx2x *bp) {}
 static inline int bnx2x_vfpf_setup_q(struct bnx2x *bp, struct bnx2x_fastpath *fp, bool is_leading) {return 0; }
-static inline int bnx2x_vfpf_config_mac(struct bnx2x *bp, u8 *addr,
+static inline int bnx2x_vfpf_config_mac(struct bnx2x *bp, const u8 *addr,
 					u8 vf_qid, bool set) {return 0; }
 static inline int bnx2x_vfpf_config_rss(struct bnx2x *bp,
 					struct bnx2x_config_rss_params *params) {return 0; }
diff --git a/drivers/net/ethernet/broadcom/bnx2x/bnx2x_stats.c b/drivers/net/ethernet/broadcom/bnx2x/bnx2x_stats.c
index 7e0919aa450e..ba6729f2f9c0 100644
--- a/drivers/net/ethernet/broadcom/bnx2x/bnx2x_stats.c
+++ b/drivers/net/ethernet/broadcom/bnx2x/bnx2x_stats.c
@@ -847,7 +847,8 @@ static int bnx2x_hw_stats_update(struct bnx2x *bp)
 
 	memcpy(old, new, sizeof(struct nig_stats));
 
-	memcpy(&(estats->rx_stat_ifhcinbadoctets_hi), &(pstats->mac_stx[1]),
+	BUILD_BUG_ON(sizeof(estats->shared) != sizeof(pstats->mac_stx[1]));
+	memcpy(&(estats->shared), &(pstats->mac_stx[1]),
 	       sizeof(struct mac_stx));
 	estats->brb_drop_hi = pstats->brb_drop_hi;
 	estats->brb_drop_lo = pstats->brb_drop_lo;
@@ -1632,9 +1633,9 @@ void bnx2x_stats_init(struct bnx2x *bp)
 			REG_RD(bp, NIG_REG_STAT0_BRB_TRUNCATE + port*0x38);
 	if (!CHIP_IS_E3(bp)) {
 		REG_RD_DMAE(bp, NIG_REG_STAT0_EGRESS_MAC_PKT0 + port*0x50,
-			    &(bp->port.old_nig_stats.egress_mac_pkt0_lo), 2);
+			    &(bp->port.old_nig_stats.egress_mac_pkt0), 2);
 		REG_RD_DMAE(bp, NIG_REG_STAT0_EGRESS_MAC_PKT1 + port*0x50,
-			    &(bp->port.old_nig_stats.egress_mac_pkt1_lo), 2);
+			    &(bp->port.old_nig_stats.egress_mac_pkt1), 2);
 	}
 
 	/* Prepare statistics ramrod data */
diff --git a/drivers/net/ethernet/broadcom/bnx2x/bnx2x_stats.h b/drivers/net/ethernet/broadcom/bnx2x/bnx2x_stats.h
index b2644ed13d06..ae93c078707b 100644
--- a/drivers/net/ethernet/broadcom/bnx2x/bnx2x_stats.h
+++ b/drivers/net/ethernet/broadcom/bnx2x/bnx2x_stats.h
@@ -36,10 +36,14 @@ struct nig_stats {
 	u32 pbf_octets;
 	u32 pbf_packet;
 	u32 safc_inp;
-	u32 egress_mac_pkt0_lo;
-	u32 egress_mac_pkt0_hi;
-	u32 egress_mac_pkt1_lo;
-	u32 egress_mac_pkt1_hi;
+	struct_group(egress_mac_pkt0,
+		u32 egress_mac_pkt0_lo;
+		u32 egress_mac_pkt0_hi;
+	);
+	struct_group(egress_mac_pkt1,
+		u32 egress_mac_pkt1_lo;
+		u32 egress_mac_pkt1_hi;
+	);
 };
 
 enum bnx2x_stats_event {
@@ -83,6 +87,7 @@ struct bnx2x_eth_stats {
 	u32 no_buff_discard_hi;
 	u32 no_buff_discard_lo;
 
+	struct_group(shared,
 	u32 rx_stat_ifhcinbadoctets_hi;
 	u32 rx_stat_ifhcinbadoctets_lo;
 	u32 tx_stat_ifhcoutbadoctets_hi;
@@ -159,6 +164,7 @@ struct bnx2x_eth_stats {
 	u32 tx_stat_dot3statsinternalmactransmiterrors_lo;
 	u32 tx_stat_bmac_ufl_hi;
 	u32 tx_stat_bmac_ufl_lo;
+	);
 
 	u32 pause_frames_received_hi;
 	u32 pause_frames_received_lo;
@@ -207,6 +213,9 @@ struct bnx2x_eth_stats {
 	u32 driver_filtered_tx_pkt;
 	/* src: Clear-on-Read register; Will not survive PMF Migration */
 	u32 eee_tx_lpi;
+
+	/* PTP */
+	u32 ptp_skip_tx_ts;
 };
 
 struct bnx2x_eth_q_stats {
diff --git a/drivers/net/ethernet/broadcom/bnx2x/bnx2x_vfpf.c b/drivers/net/ethernet/broadcom/bnx2x/bnx2x_vfpf.c
index 76a4668c50fe..8946a931e87e 100644
--- a/drivers/net/ethernet/broadcom/bnx2x/bnx2x_vfpf.c
+++ b/drivers/net/ethernet/broadcom/bnx2x/bnx2x_vfpf.c
@@ -170,7 +170,7 @@ static int bnx2x_send_msg2pf(struct bnx2x *bp, u8 *done, dma_addr_t msg_mapping)
 	wmb();
 
 	/* Trigger the PF FW */
-	writeb(1, &zone_data->trigger.vf_pf_channel.addr_valid);
+	writeb_relaxed(1, &zone_data->trigger.vf_pf_channel.addr_valid);
 
 	/* Wait for PF to complete */
 	while ((tout >= 0) && (!*done)) {
@@ -380,13 +380,12 @@ int bnx2x_vfpf_acquire(struct bnx2x *bp, u8 tx_count, u8 rx_count)
 	bp->igu_base_sb = bp->acquire_resp.resc.hw_sbs[0].hw_sb_id;
 	bp->vlan_credit = bp->acquire_resp.resc.num_vlan_filters;
 
-	strlcpy(bp->fw_ver, bp->acquire_resp.pfdev_info.fw_ver,
+	strscpy(bp->fw_ver, bp->acquire_resp.pfdev_info.fw_ver,
 		sizeof(bp->fw_ver));
 
 	if (is_valid_ether_addr(bp->acquire_resp.resc.current_mac_addr))
-		memcpy(bp->dev->dev_addr,
-		       bp->acquire_resp.resc.current_mac_addr,
-		       ETH_ALEN);
+		eth_hw_addr_set(bp->dev,
+				bp->acquire_resp.resc.current_mac_addr);
 
 out:
 	bnx2x_vfpf_finalize(bp, &req->first_tlv);
@@ -530,13 +529,16 @@ void bnx2x_vfpf_close_vf(struct bnx2x *bp)
 	bnx2x_vfpf_finalize(bp, &req->first_tlv);
 
 free_irq:
-	/* Disable HW interrupts, NAPI */
-	bnx2x_netif_stop(bp, 0);
-	/* Delete all NAPI objects */
-	bnx2x_del_all_napi(bp);
-
-	/* Release IRQs */
-	bnx2x_free_irq(bp);
+	if (!bp->nic_stopped) {
+		/* Disable HW interrupts, NAPI */
+		bnx2x_netif_stop(bp, 0);
+		/* Delete all NAPI objects */
+		bnx2x_del_all_napi(bp);
+
+		/* Release IRQs */
+		bnx2x_free_irq(bp);
+		bp->nic_stopped = true;
+	}
 }
 
 static void bnx2x_leading_vfq_init(struct bnx2x *bp, struct bnx2x_virtf *vf,
@@ -722,7 +724,7 @@ out:
 }
 
 /* request pf to add a mac for the vf */
-int bnx2x_vfpf_config_mac(struct bnx2x *bp, u8 *addr, u8 vf_qid, bool set)
+int bnx2x_vfpf_config_mac(struct bnx2x *bp, const u8 *addr, u8 vf_qid, bool set)
 {
 	struct vfpf_set_q_filters_tlv *req = &bp->vf2pf_mbox->req.set_q_filters;
 	struct pfvf_general_resp_tlv *resp = &bp->vf2pf_mbox->resp.general_resp;
@@ -767,7 +769,7 @@ int bnx2x_vfpf_config_mac(struct bnx2x *bp, u8 *addr, u8 vf_qid, bool set)
 		   "vfpf SET MAC failed. Check bulletin board for new posts\n");
 
 		/* copy mac from bulletin to device */
-		memcpy(bp->dev->dev_addr, bulletin.mac, ETH_ALEN);
+		eth_hw_addr_set(bp->dev, bulletin.mac);
 
 		/* check if bulletin board was updated */
 		if (bnx2x_sample_bulletin(bp) == PFVF_BULLETIN_UPDATED) {
@@ -955,7 +957,7 @@ int bnx2x_vfpf_update_vlan(struct bnx2x *bp, u16 vid, u8 vf_qid, bool add)
 	bnx2x_sample_bulletin(bp);
 
 	if (bp->shadow_bulletin.content.valid_bitmap & 1 << VLAN_VALID) {
-		BNX2X_ERR("Hypervisor will dicline the request, avoiding\n");
+		BNX2X_ERR("Hypervisor will decline the request, avoiding\n");
 		rc = -EINVAL;
 		goto out;
 	}
@@ -1177,7 +1179,6 @@ static void bnx2x_vf_mbx_resp_send_msg(struct bnx2x *bp,
 
 	/* ack the FW */
 	storm_memset_vf_mbx_ack(bp, vf->abs_vfid);
-	mmiowb();
 
 	/* copy the response header including status-done field,
 	 * must be last dmae, must be after FW is acked
@@ -1652,13 +1653,9 @@ static int bnx2x_vf_mbx_macvlan_list(struct bnx2x *bp,
 {
 	int i, j;
 	struct bnx2x_vf_mac_vlan_filters *fl = NULL;
-	size_t fsz;
-
-	fsz = tlv->n_mac_vlan_filters *
-	      sizeof(struct bnx2x_vf_mac_vlan_filter) +
-	      sizeof(struct bnx2x_vf_mac_vlan_filters);
 
-	fl = kzalloc(fsz, GFP_KERNEL);
+	fl = kzalloc(struct_size(fl, filters, tlv->n_mac_vlan_filters),
+		     GFP_KERNEL);
 	if (!fl)
 		return -ENOMEM;
 
@@ -2112,6 +2109,18 @@ static void bnx2x_vf_mbx_request(struct bnx2x *bp, struct bnx2x_virtf *vf,
 {
 	int i;
 
+	if (vf->state == VF_LOST) {
+		/* Just ack the FW and return if VFs are lost
+		 * in case of parity error. VFs are supposed to be timedout
+		 * on waiting for PF response.
+		 */
+		DP(BNX2X_MSG_IOV,
+		   "VF 0x%x lost, not handling the request\n", vf->abs_vfid);
+
+		storm_memset_vf_mbx_ack(bp, vf->abs_vfid);
+		return;
+	}
+
 	/* check if tlv type is known */
 	if (bnx2x_tlv_supported(mbx->first_tlv.tl.type)) {
 		/* Lock the per vf op mutex and note the locker's identity.
@@ -2176,7 +2185,6 @@ static void bnx2x_vf_mbx_request(struct bnx2x *bp, struct bnx2x_virtf *vf,
 		 */
 		storm_memset_vf_mbx_ack(bp, vf->abs_vfid);
 		/* Firmware ack should be written before unlocking channel */
-		mmiowb();
 		bnx2x_unlock_vf_pf_channel(bp, vf, mbx->first_tlv.tl.type);
 	}
 }
diff --git a/drivers/net/ethernet/broadcom/bnxt/Makefile b/drivers/net/ethernet/broadcom/bnxt/Makefile
index a7ca45b251cb..ba6c239d52fa 100644
--- a/drivers/net/ethernet/broadcom/bnxt/Makefile
+++ b/drivers/net/ethernet/broadcom/bnxt/Makefile
@@ -1,3 +1,7 @@
+# SPDX-License-Identifier: GPL-2.0-only
 obj-$(CONFIG_BNXT) += bnxt_en.o
 
-bnxt_en-y := bnxt.o bnxt_sriov.o bnxt_ethtool.o bnxt_dcb.o bnxt_ulp.o bnxt_xdp.o
+bnxt_en-y := bnxt.o bnxt_hwrm.o bnxt_sriov.o bnxt_ethtool.o bnxt_dcb.o bnxt_ulp.o bnxt_xdp.o bnxt_ptp.o bnxt_vfr.o bnxt_devlink.o bnxt_dim.o bnxt_coredump.o
+bnxt_en-$(CONFIG_BNXT_FLOWER_OFFLOAD) += bnxt_tc.o
+bnxt_en-$(CONFIG_DEBUG_FS) += bnxt_debugfs.o
+bnxt_en-$(CONFIG_BNXT_HWMON) += bnxt_hwmon.o
diff --git a/drivers/net/ethernet/broadcom/bnxt/bnxt.c b/drivers/net/ethernet/broadcom/bnxt/bnxt.c
index c1cd72a5eccf..3fc33b1b4dfb 100644
--- a/drivers/net/ethernet/broadcom/bnxt/bnxt.c
+++ b/drivers/net/ethernet/broadcom/bnxt/bnxt.c
@@ -1,7 +1,7 @@
 /* Broadcom NetXtreme-C/E network driver.
  *
  * Copyright (c) 2014-2016 Broadcom Corporation
- * Copyright (c) 2016-2017 Broadcom Limited
+ * Copyright (c) 2016-2019 Broadcom Limited
  *
  * This program is free software; you can redistribute it and/or modify
  * it under the terms of the GNU General Public License as published by
@@ -31,10 +31,13 @@
 #include <asm/page.h>
 #include <linux/time.h>
 #include <linux/mii.h>
+#include <linux/mdio.h>
 #include <linux/if.h>
 #include <linux/if_vlan.h>
+#include <linux/if_bridge.h>
 #include <linux/rtc.h>
 #include <linux/bpf.h>
+#include <net/gro.h>
 #include <net/ip.h>
 #include <net/tcp.h>
 #include <net/udp.h>
@@ -45,103 +48,107 @@
 #include <linux/prefetch.h>
 #include <linux/cache.h>
 #include <linux/log2.h>
-#include <linux/aer.h>
 #include <linux/bitmap.h>
 #include <linux/cpu_rmap.h>
+#include <linux/cpumask.h>
+#include <net/pkt_cls.h>
+#include <net/page_pool/helpers.h>
+#include <linux/align.h>
+#include <net/netdev_lock.h>
+#include <net/netdev_queues.h>
+#include <net/netdev_rx_queue.h>
+#include <linux/pci-tph.h>
+#include <linux/bnxt/hsi.h>
 
-#include "bnxt_hsi.h"
 #include "bnxt.h"
+#include "bnxt_hwrm.h"
 #include "bnxt_ulp.h"
 #include "bnxt_sriov.h"
 #include "bnxt_ethtool.h"
 #include "bnxt_dcb.h"
 #include "bnxt_xdp.h"
+#include "bnxt_ptp.h"
+#include "bnxt_vfr.h"
+#include "bnxt_tc.h"
+#include "bnxt_devlink.h"
+#include "bnxt_debugfs.h"
+#include "bnxt_coredump.h"
+#include "bnxt_hwmon.h"
 
 #define BNXT_TX_TIMEOUT		(5 * HZ)
+#define BNXT_DEF_MSG_ENABLE	(NETIF_MSG_DRV | NETIF_MSG_HW | \
+				 NETIF_MSG_TX_ERR)
 
-static const char version[] =
-	"Broadcom NetXtreme-C/E driver " DRV_MODULE_NAME " v" DRV_MODULE_VERSION "\n";
-
+MODULE_IMPORT_NS("NETDEV_INTERNAL");
 MODULE_LICENSE("GPL");
-MODULE_DESCRIPTION("Broadcom BCM573xx network driver");
-MODULE_VERSION(DRV_MODULE_VERSION);
+MODULE_DESCRIPTION("Broadcom NetXtreme network driver");
 
 #define BNXT_RX_OFFSET (NET_SKB_PAD + NET_IP_ALIGN)
 #define BNXT_RX_DMA_OFFSET NET_SKB_PAD
-#define BNXT_RX_COPY_THRESH 256
 
 #define BNXT_TX_PUSH_THRESH 164
 
-enum board_idx {
-	BCM57301,
-	BCM57302,
-	BCM57304,
-	BCM57417_NPAR,
-	BCM58700,
-	BCM57311,
-	BCM57312,
-	BCM57402,
-	BCM57404,
-	BCM57406,
-	BCM57402_NPAR,
-	BCM57407,
-	BCM57412,
-	BCM57414,
-	BCM57416,
-	BCM57417,
-	BCM57412_NPAR,
-	BCM57314,
-	BCM57417_SFP,
-	BCM57416_SFP,
-	BCM57404_NPAR,
-	BCM57406_NPAR,
-	BCM57407_SFP,
-	BCM57407_NPAR,
-	BCM57414_NPAR,
-	BCM57416_NPAR,
-	BCM57452,
-	BCM57454,
-	NETXTREME_E_VF,
-	NETXTREME_C_VF,
-};
-
-/* indexed by enum above */
+/* indexed by enum board_idx */
 static const struct {
 	char *name;
 } board_info[] = {
-	{ "Broadcom BCM57301 NetXtreme-C 10Gb Ethernet" },
-	{ "Broadcom BCM57302 NetXtreme-C 10Gb/25Gb Ethernet" },
-	{ "Broadcom BCM57304 NetXtreme-C 10Gb/25Gb/40Gb/50Gb Ethernet" },
-	{ "Broadcom BCM57417 NetXtreme-E Ethernet Partition" },
-	{ "Broadcom BCM58700 Nitro 1Gb/2.5Gb/10Gb Ethernet" },
-	{ "Broadcom BCM57311 NetXtreme-C 10Gb Ethernet" },
-	{ "Broadcom BCM57312 NetXtreme-C 10Gb/25Gb Ethernet" },
-	{ "Broadcom BCM57402 NetXtreme-E 10Gb Ethernet" },
-	{ "Broadcom BCM57404 NetXtreme-E 10Gb/25Gb Ethernet" },
-	{ "Broadcom BCM57406 NetXtreme-E 10GBase-T Ethernet" },
-	{ "Broadcom BCM57402 NetXtreme-E Ethernet Partition" },
-	{ "Broadcom BCM57407 NetXtreme-E 10GBase-T Ethernet" },
-	{ "Broadcom BCM57412 NetXtreme-E 10Gb Ethernet" },
-	{ "Broadcom BCM57414 NetXtreme-E 10Gb/25Gb Ethernet" },
-	{ "Broadcom BCM57416 NetXtreme-E 10GBase-T Ethernet" },
-	{ "Broadcom BCM57417 NetXtreme-E 10GBase-T Ethernet" },
-	{ "Broadcom BCM57412 NetXtreme-E Ethernet Partition" },
-	{ "Broadcom BCM57314 NetXtreme-C 10Gb/25Gb/40Gb/50Gb Ethernet" },
-	{ "Broadcom BCM57417 NetXtreme-E 10Gb/25Gb Ethernet" },
-	{ "Broadcom BCM57416 NetXtreme-E 10Gb Ethernet" },
-	{ "Broadcom BCM57404 NetXtreme-E Ethernet Partition" },
-	{ "Broadcom BCM57406 NetXtreme-E Ethernet Partition" },
-	{ "Broadcom BCM57407 NetXtreme-E 25Gb Ethernet" },
-	{ "Broadcom BCM57407 NetXtreme-E Ethernet Partition" },
-	{ "Broadcom BCM57414 NetXtreme-E Ethernet Partition" },
-	{ "Broadcom BCM57416 NetXtreme-E Ethernet Partition" },
-	{ "Broadcom BCM57452 NetXtreme-E 10Gb/25Gb/40Gb/50Gb Ethernet" },
-	{ "Broadcom BCM57454 NetXtreme-E 10Gb/25Gb/40Gb/50Gb/100Gb Ethernet" },
-	{ "Broadcom NetXtreme-E Ethernet Virtual Function" },
-	{ "Broadcom NetXtreme-C Ethernet Virtual Function" },
+	[BCM57301] = { "Broadcom BCM57301 NetXtreme-C 10Gb Ethernet" },
+	[BCM57302] = { "Broadcom BCM57302 NetXtreme-C 10Gb/25Gb Ethernet" },
+	[BCM57304] = { "Broadcom BCM57304 NetXtreme-C 10Gb/25Gb/40Gb/50Gb Ethernet" },
+	[BCM57417_NPAR] = { "Broadcom BCM57417 NetXtreme-E Ethernet Partition" },
+	[BCM58700] = { "Broadcom BCM58700 Nitro 1Gb/2.5Gb/10Gb Ethernet" },
+	[BCM57311] = { "Broadcom BCM57311 NetXtreme-C 10Gb Ethernet" },
+	[BCM57312] = { "Broadcom BCM57312 NetXtreme-C 10Gb/25Gb Ethernet" },
+	[BCM57402] = { "Broadcom BCM57402 NetXtreme-E 10Gb Ethernet" },
+	[BCM57404] = { "Broadcom BCM57404 NetXtreme-E 10Gb/25Gb Ethernet" },
+	[BCM57406] = { "Broadcom BCM57406 NetXtreme-E 10GBase-T Ethernet" },
+	[BCM57402_NPAR] = { "Broadcom BCM57402 NetXtreme-E Ethernet Partition" },
+	[BCM57407] = { "Broadcom BCM57407 NetXtreme-E 10GBase-T Ethernet" },
+	[BCM57412] = { "Broadcom BCM57412 NetXtreme-E 10Gb Ethernet" },
+	[BCM57414] = { "Broadcom BCM57414 NetXtreme-E 10Gb/25Gb Ethernet" },
+	[BCM57416] = { "Broadcom BCM57416 NetXtreme-E 10GBase-T Ethernet" },
+	[BCM57417] = { "Broadcom BCM57417 NetXtreme-E 10GBase-T Ethernet" },
+	[BCM57412_NPAR] = { "Broadcom BCM57412 NetXtreme-E Ethernet Partition" },
+	[BCM57314] = { "Broadcom BCM57314 NetXtreme-C 10Gb/25Gb/40Gb/50Gb Ethernet" },
+	[BCM57417_SFP] = { "Broadcom BCM57417 NetXtreme-E 10Gb/25Gb Ethernet" },
+	[BCM57416_SFP] = { "Broadcom BCM57416 NetXtreme-E 10Gb Ethernet" },
+	[BCM57404_NPAR] = { "Broadcom BCM57404 NetXtreme-E Ethernet Partition" },
+	[BCM57406_NPAR] = { "Broadcom BCM57406 NetXtreme-E Ethernet Partition" },
+	[BCM57407_SFP] = { "Broadcom BCM57407 NetXtreme-E 25Gb Ethernet" },
+	[BCM57407_NPAR] = { "Broadcom BCM57407 NetXtreme-E Ethernet Partition" },
+	[BCM57414_NPAR] = { "Broadcom BCM57414 NetXtreme-E Ethernet Partition" },
+	[BCM57416_NPAR] = { "Broadcom BCM57416 NetXtreme-E Ethernet Partition" },
+	[BCM57452] = { "Broadcom BCM57452 NetXtreme-E 10Gb/25Gb/40Gb/50Gb Ethernet" },
+	[BCM57454] = { "Broadcom BCM57454 NetXtreme-E 10Gb/25Gb/40Gb/50Gb/100Gb Ethernet" },
+	[BCM5745x_NPAR] = { "Broadcom BCM5745x NetXtreme-E Ethernet Partition" },
+	[BCM57508] = { "Broadcom BCM57508 NetXtreme-E 10Gb/25Gb/50Gb/100Gb/200Gb Ethernet" },
+	[BCM57504] = { "Broadcom BCM57504 NetXtreme-E 10Gb/25Gb/50Gb/100Gb/200Gb Ethernet" },
+	[BCM57502] = { "Broadcom BCM57502 NetXtreme-E 10Gb/25Gb/50Gb Ethernet" },
+	[BCM57608] = { "Broadcom BCM57608 NetXtreme-E 10Gb/25Gb/50Gb/100Gb/200Gb/400Gb Ethernet" },
+	[BCM57604] = { "Broadcom BCM57604 NetXtreme-E 10Gb/25Gb/50Gb/100Gb/200Gb Ethernet" },
+	[BCM57602] = { "Broadcom BCM57602 NetXtreme-E 10Gb/25Gb/50Gb/100Gb Ethernet" },
+	[BCM57601] = { "Broadcom BCM57601 NetXtreme-E 10Gb/25Gb/50Gb/100Gb/200Gb/400Gb Ethernet" },
+	[BCM57508_NPAR] = { "Broadcom BCM57508 NetXtreme-E Ethernet Partition" },
+	[BCM57504_NPAR] = { "Broadcom BCM57504 NetXtreme-E Ethernet Partition" },
+	[BCM57502_NPAR] = { "Broadcom BCM57502 NetXtreme-E Ethernet Partition" },
+	[BCM58802] = { "Broadcom BCM58802 NetXtreme-S 10Gb/25Gb/40Gb/50Gb Ethernet" },
+	[BCM58804] = { "Broadcom BCM58804 NetXtreme-S 10Gb/25Gb/40Gb/50Gb/100Gb Ethernet" },
+	[BCM58808] = { "Broadcom BCM58808 NetXtreme-S 10Gb/25Gb/40Gb/50Gb/100Gb Ethernet" },
+	[NETXTREME_E_VF] = { "Broadcom NetXtreme-E Ethernet Virtual Function" },
+	[NETXTREME_C_VF] = { "Broadcom NetXtreme-C Ethernet Virtual Function" },
+	[NETXTREME_S_VF] = { "Broadcom NetXtreme-S Ethernet Virtual Function" },
+	[NETXTREME_C_VF_HV] = { "Broadcom NetXtreme-C Virtual Function for Hyper-V" },
+	[NETXTREME_E_VF_HV] = { "Broadcom NetXtreme-E Virtual Function for Hyper-V" },
+	[NETXTREME_E_P5_VF] = { "Broadcom BCM5750X NetXtreme-E Ethernet Virtual Function" },
+	[NETXTREME_E_P5_VF_HV] = { "Broadcom BCM5750X NetXtreme-E Virtual Function for Hyper-V" },
+	[NETXTREME_E_P7_VF] = { "Broadcom BCM5760X Virtual Function" },
+	[NETXTREME_E_P7_VF_HV] = { "Broadcom BCM5760X Virtual Function for Hyper-V" },
 };
 
 static const struct pci_device_id bnxt_pci_tbl[] = {
+	{ PCI_VDEVICE(BROADCOM, 0x1604), .driver_data = BCM5745x_NPAR },
+	{ PCI_VDEVICE(BROADCOM, 0x1605), .driver_data = BCM5745x_NPAR },
+	{ PCI_VDEVICE(BROADCOM, 0x1614), .driver_data = BCM57454 },
 	{ PCI_VDEVICE(BROADCOM, 0x16c0), .driver_data = BCM57417_NPAR },
 	{ PCI_VDEVICE(BROADCOM, 0x16c8), .driver_data = BCM57301 },
 	{ PCI_VDEVICE(BROADCOM, 0x16c9), .driver_data = BCM57302 },
@@ -172,17 +179,47 @@ static const struct pci_device_id bnxt_pci_tbl[] = {
 	{ PCI_VDEVICE(BROADCOM, 0x16ed), .driver_data = BCM57414_NPAR },
 	{ PCI_VDEVICE(BROADCOM, 0x16ee), .driver_data = BCM57416_NPAR },
 	{ PCI_VDEVICE(BROADCOM, 0x16ef), .driver_data = BCM57416_NPAR },
+	{ PCI_VDEVICE(BROADCOM, 0x16f0), .driver_data = BCM58808 },
 	{ PCI_VDEVICE(BROADCOM, 0x16f1), .driver_data = BCM57452 },
-	{ PCI_VDEVICE(BROADCOM, 0x1614), .driver_data = BCM57454 },
+	{ PCI_VDEVICE(BROADCOM, 0x1750), .driver_data = BCM57508 },
+	{ PCI_VDEVICE(BROADCOM, 0x1751), .driver_data = BCM57504 },
+	{ PCI_VDEVICE(BROADCOM, 0x1752), .driver_data = BCM57502 },
+	{ PCI_VDEVICE(BROADCOM, 0x1760), .driver_data = BCM57608 },
+	{ PCI_VDEVICE(BROADCOM, 0x1761), .driver_data = BCM57604 },
+	{ PCI_VDEVICE(BROADCOM, 0x1762), .driver_data = BCM57602 },
+	{ PCI_VDEVICE(BROADCOM, 0x1763), .driver_data = BCM57601 },
+	{ PCI_VDEVICE(BROADCOM, 0x1800), .driver_data = BCM57502_NPAR },
+	{ PCI_VDEVICE(BROADCOM, 0x1801), .driver_data = BCM57504_NPAR },
+	{ PCI_VDEVICE(BROADCOM, 0x1802), .driver_data = BCM57508_NPAR },
+	{ PCI_VDEVICE(BROADCOM, 0x1803), .driver_data = BCM57502_NPAR },
+	{ PCI_VDEVICE(BROADCOM, 0x1804), .driver_data = BCM57504_NPAR },
+	{ PCI_VDEVICE(BROADCOM, 0x1805), .driver_data = BCM57508_NPAR },
+	{ PCI_VDEVICE(BROADCOM, 0xd802), .driver_data = BCM58802 },
+	{ PCI_VDEVICE(BROADCOM, 0xd804), .driver_data = BCM58804 },
 #ifdef CONFIG_BNXT_SRIOV
 	{ PCI_VDEVICE(BROADCOM, 0x1606), .driver_data = NETXTREME_E_VF },
+	{ PCI_VDEVICE(BROADCOM, 0x1607), .driver_data = NETXTREME_E_VF_HV },
+	{ PCI_VDEVICE(BROADCOM, 0x1608), .driver_data = NETXTREME_E_VF_HV },
 	{ PCI_VDEVICE(BROADCOM, 0x1609), .driver_data = NETXTREME_E_VF },
+	{ PCI_VDEVICE(BROADCOM, 0x16bd), .driver_data = NETXTREME_E_VF_HV },
 	{ PCI_VDEVICE(BROADCOM, 0x16c1), .driver_data = NETXTREME_E_VF },
+	{ PCI_VDEVICE(BROADCOM, 0x16c2), .driver_data = NETXTREME_C_VF_HV },
+	{ PCI_VDEVICE(BROADCOM, 0x16c3), .driver_data = NETXTREME_C_VF_HV },
+	{ PCI_VDEVICE(BROADCOM, 0x16c4), .driver_data = NETXTREME_E_VF_HV },
+	{ PCI_VDEVICE(BROADCOM, 0x16c5), .driver_data = NETXTREME_E_VF_HV },
 	{ PCI_VDEVICE(BROADCOM, 0x16cb), .driver_data = NETXTREME_C_VF },
 	{ PCI_VDEVICE(BROADCOM, 0x16d3), .driver_data = NETXTREME_E_VF },
 	{ PCI_VDEVICE(BROADCOM, 0x16dc), .driver_data = NETXTREME_E_VF },
 	{ PCI_VDEVICE(BROADCOM, 0x16e1), .driver_data = NETXTREME_C_VF },
 	{ PCI_VDEVICE(BROADCOM, 0x16e5), .driver_data = NETXTREME_C_VF },
+	{ PCI_VDEVICE(BROADCOM, 0x16e6), .driver_data = NETXTREME_C_VF_HV },
+	{ PCI_VDEVICE(BROADCOM, 0x1806), .driver_data = NETXTREME_E_P5_VF },
+	{ PCI_VDEVICE(BROADCOM, 0x1807), .driver_data = NETXTREME_E_P5_VF },
+	{ PCI_VDEVICE(BROADCOM, 0x1808), .driver_data = NETXTREME_E_P5_VF_HV },
+	{ PCI_VDEVICE(BROADCOM, 0x1809), .driver_data = NETXTREME_E_P5_VF_HV },
+	{ PCI_VDEVICE(BROADCOM, 0x1819), .driver_data = NETXTREME_E_P7_VF },
+	{ PCI_VDEVICE(BROADCOM, 0x181b), .driver_data = NETXTREME_E_P7_VF_HV },
+	{ PCI_VDEVICE(BROADCOM, 0xd800), .driver_data = NETXTREME_S_VF },
 #endif
 	{ 0 }
 };
@@ -191,35 +228,202 @@ MODULE_DEVICE_TABLE(pci, bnxt_pci_tbl);
 
 static const u16 bnxt_vf_req_snif[] = {
 	HWRM_FUNC_CFG,
+	HWRM_FUNC_VF_CFG,
 	HWRM_PORT_PHY_QCFG,
 	HWRM_CFA_L2_FILTER_ALLOC,
 };
 
 static const u16 bnxt_async_events_arr[] = {
 	ASYNC_EVENT_CMPL_EVENT_ID_LINK_STATUS_CHANGE,
+	ASYNC_EVENT_CMPL_EVENT_ID_LINK_SPEED_CHANGE,
 	ASYNC_EVENT_CMPL_EVENT_ID_PF_DRVR_UNLOAD,
 	ASYNC_EVENT_CMPL_EVENT_ID_PORT_CONN_NOT_ALLOWED,
 	ASYNC_EVENT_CMPL_EVENT_ID_VF_CFG_CHANGE,
 	ASYNC_EVENT_CMPL_EVENT_ID_LINK_SPEED_CFG_CHANGE,
+	ASYNC_EVENT_CMPL_EVENT_ID_PORT_PHY_CFG_CHANGE,
+	ASYNC_EVENT_CMPL_EVENT_ID_RESET_NOTIFY,
+	ASYNC_EVENT_CMPL_EVENT_ID_ERROR_RECOVERY,
+	ASYNC_EVENT_CMPL_EVENT_ID_DEBUG_NOTIFICATION,
+	ASYNC_EVENT_CMPL_EVENT_ID_DEFERRED_RESPONSE,
+	ASYNC_EVENT_CMPL_EVENT_ID_RING_MONITOR_MSG,
+	ASYNC_EVENT_CMPL_EVENT_ID_ECHO_REQUEST,
+	ASYNC_EVENT_CMPL_EVENT_ID_PPS_TIMESTAMP,
+	ASYNC_EVENT_CMPL_EVENT_ID_ERROR_REPORT,
+	ASYNC_EVENT_CMPL_EVENT_ID_PHC_UPDATE,
+	ASYNC_EVENT_CMPL_EVENT_ID_DBG_BUF_PRODUCER,
+};
+
+const u16 bnxt_bstore_to_trace[] = {
+	[BNXT_CTX_SRT]		= DBG_LOG_BUFFER_FLUSH_REQ_TYPE_SRT_TRACE,
+	[BNXT_CTX_SRT2]		= DBG_LOG_BUFFER_FLUSH_REQ_TYPE_SRT2_TRACE,
+	[BNXT_CTX_CRT]		= DBG_LOG_BUFFER_FLUSH_REQ_TYPE_CRT_TRACE,
+	[BNXT_CTX_CRT2]		= DBG_LOG_BUFFER_FLUSH_REQ_TYPE_CRT2_TRACE,
+	[BNXT_CTX_RIGP0]	= DBG_LOG_BUFFER_FLUSH_REQ_TYPE_RIGP0_TRACE,
+	[BNXT_CTX_L2HWRM]	= DBG_LOG_BUFFER_FLUSH_REQ_TYPE_L2_HWRM_TRACE,
+	[BNXT_CTX_REHWRM]	= DBG_LOG_BUFFER_FLUSH_REQ_TYPE_ROCE_HWRM_TRACE,
+	[BNXT_CTX_CA0]		= DBG_LOG_BUFFER_FLUSH_REQ_TYPE_CA0_TRACE,
+	[BNXT_CTX_CA1]		= DBG_LOG_BUFFER_FLUSH_REQ_TYPE_CA1_TRACE,
+	[BNXT_CTX_CA2]		= DBG_LOG_BUFFER_FLUSH_REQ_TYPE_CA2_TRACE,
+	[BNXT_CTX_RIGP1]	= DBG_LOG_BUFFER_FLUSH_REQ_TYPE_RIGP1_TRACE,
+	[BNXT_CTX_KONG]		= DBG_LOG_BUFFER_FLUSH_REQ_TYPE_AFM_KONG_HWRM_TRACE,
+	[BNXT_CTX_QPC]		= DBG_LOG_BUFFER_FLUSH_REQ_TYPE_ERR_QPC_TRACE,
+};
+
+static struct workqueue_struct *bnxt_pf_wq;
+
+#define BNXT_IPV6_MASK_ALL {{{ 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, \
+			       0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff }}}
+#define BNXT_IPV6_MASK_NONE {{{ 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0 }}}
+
+const struct bnxt_flow_masks BNXT_FLOW_MASK_NONE = {
+	.ports = {
+		.src = 0,
+		.dst = 0,
+	},
+	.addrs = {
+		.v6addrs = {
+			.src = BNXT_IPV6_MASK_NONE,
+			.dst = BNXT_IPV6_MASK_NONE,
+		},
+	},
+};
+
+const struct bnxt_flow_masks BNXT_FLOW_IPV6_MASK_ALL = {
+	.ports = {
+		.src = cpu_to_be16(0xffff),
+		.dst = cpu_to_be16(0xffff),
+	},
+	.addrs = {
+		.v6addrs = {
+			.src = BNXT_IPV6_MASK_ALL,
+			.dst = BNXT_IPV6_MASK_ALL,
+		},
+	},
+};
+
+const struct bnxt_flow_masks BNXT_FLOW_IPV4_MASK_ALL = {
+	.ports = {
+		.src = cpu_to_be16(0xffff),
+		.dst = cpu_to_be16(0xffff),
+	},
+	.addrs = {
+		.v4addrs = {
+			.src = cpu_to_be32(0xffffffff),
+			.dst = cpu_to_be32(0xffffffff),
+		},
+	},
 };
 
 static bool bnxt_vf_pciid(enum board_idx idx)
 {
-	return (idx == NETXTREME_C_VF || idx == NETXTREME_E_VF);
+	return (idx == NETXTREME_C_VF || idx == NETXTREME_E_VF ||
+		idx == NETXTREME_S_VF || idx == NETXTREME_C_VF_HV ||
+		idx == NETXTREME_E_VF_HV || idx == NETXTREME_E_P5_VF ||
+		idx == NETXTREME_E_P5_VF_HV || idx == NETXTREME_E_P7_VF ||
+		idx == NETXTREME_E_P7_VF_HV);
 }
 
 #define DB_CP_REARM_FLAGS	(DB_KEY_CP | DB_IDX_VALID)
 #define DB_CP_FLAGS		(DB_KEY_CP | DB_IDX_VALID | DB_IRQ_DIS)
-#define DB_CP_IRQ_DIS_FLAGS	(DB_KEY_CP | DB_IRQ_DIS)
 
-#define BNXT_CP_DB_REARM(db, raw_cons)					\
-		writel(DB_CP_REARM_FLAGS | RING_CMP(raw_cons), db)
+#define BNXT_DB_CQ(db, idx)						\
+	writel(DB_CP_FLAGS | DB_RING_IDX(db, idx), (db)->doorbell)
+
+#define BNXT_DB_NQ_P5(db, idx)						\
+	bnxt_writeq(bp, (db)->db_key64 | DBR_TYPE_NQ | DB_RING_IDX(db, idx),\
+		    (db)->doorbell)
+
+#define BNXT_DB_NQ_P7(db, idx)						\
+	bnxt_writeq(bp, (db)->db_key64 | DBR_TYPE_NQ_MASK |		\
+		    DB_RING_IDX(db, idx), (db)->doorbell)
+
+#define BNXT_DB_CQ_ARM(db, idx)						\
+	writel(DB_CP_REARM_FLAGS | DB_RING_IDX(db, idx), (db)->doorbell)
+
+#define BNXT_DB_NQ_ARM_P5(db, idx)					\
+	bnxt_writeq(bp, (db)->db_key64 | DBR_TYPE_NQ_ARM |		\
+		    DB_RING_IDX(db, idx), (db)->doorbell)
+
+static void bnxt_db_nq(struct bnxt *bp, struct bnxt_db_info *db, u32 idx)
+{
+	if (bp->flags & BNXT_FLAG_CHIP_P7)
+		BNXT_DB_NQ_P7(db, idx);
+	else if (bp->flags & BNXT_FLAG_CHIP_P5_PLUS)
+		BNXT_DB_NQ_P5(db, idx);
+	else
+		BNXT_DB_CQ(db, idx);
+}
+
+static void bnxt_db_nq_arm(struct bnxt *bp, struct bnxt_db_info *db, u32 idx)
+{
+	if (bp->flags & BNXT_FLAG_CHIP_P5_PLUS)
+		BNXT_DB_NQ_ARM_P5(db, idx);
+	else
+		BNXT_DB_CQ_ARM(db, idx);
+}
+
+static void bnxt_db_cq(struct bnxt *bp, struct bnxt_db_info *db, u32 idx)
+{
+	if (bp->flags & BNXT_FLAG_CHIP_P5_PLUS)
+		bnxt_writeq(bp, db->db_key64 | DBR_TYPE_CQ_ARMALL |
+			    DB_RING_IDX(db, idx), db->doorbell);
+	else
+		BNXT_DB_CQ(db, idx);
+}
+
+static void bnxt_queue_fw_reset_work(struct bnxt *bp, unsigned long delay)
+{
+	if (!(test_bit(BNXT_STATE_IN_FW_RESET, &bp->state)))
+		return;
+
+	if (BNXT_PF(bp))
+		queue_delayed_work(bnxt_pf_wq, &bp->fw_reset_task, delay);
+	else
+		schedule_delayed_work(&bp->fw_reset_task, delay);
+}
+
+static void __bnxt_queue_sp_work(struct bnxt *bp)
+{
+	if (BNXT_PF(bp))
+		queue_work(bnxt_pf_wq, &bp->sp_task);
+	else
+		schedule_work(&bp->sp_task);
+}
+
+static void bnxt_queue_sp_work(struct bnxt *bp, unsigned int event)
+{
+	set_bit(event, &bp->sp_event);
+	__bnxt_queue_sp_work(bp);
+}
+
+static void bnxt_sched_reset_rxr(struct bnxt *bp, struct bnxt_rx_ring_info *rxr)
+{
+	if (!rxr->bnapi->in_reset) {
+		rxr->bnapi->in_reset = true;
+		if (bp->flags & BNXT_FLAG_CHIP_P5_PLUS)
+			set_bit(BNXT_RESET_TASK_SP_EVENT, &bp->sp_event);
+		else
+			set_bit(BNXT_RST_RING_SP_EVENT, &bp->sp_event);
+		__bnxt_queue_sp_work(bp);
+	}
+	rxr->rx_next_cons = 0xffff;
+}
+
+void bnxt_sched_reset_txr(struct bnxt *bp, struct bnxt_tx_ring_info *txr,
+			  u16 curr)
+{
+	struct bnxt_napi *bnapi = txr->bnapi;
 
-#define BNXT_CP_DB(db, raw_cons)					\
-		writel(DB_CP_FLAGS | RING_CMP(raw_cons), db)
+	if (bnapi->tx_fault)
+		return;
 
-#define BNXT_CP_DB_IRQ_DIS(db)						\
-		writel(DB_CP_IRQ_DIS_FLAGS, db)
+	netdev_err(bp->dev, "Invalid Tx completion (ring:%d tx_hw_cons:%u cons:%u prod:%u curr:%u)",
+		   txr->txq_index, txr->tx_hw_cons,
+		   txr->tx_cons, txr->tx_prod, curr);
+	WARN_ON_ONCE(1);
+	bnapi->tx_fault = 1;
+	bnxt_queue_sp_work(bp, BNXT_RESET_TASK_SP_EVENT);
+}
 
 const u16 bnxt_lhint_arr[] = {
 	TX_BD_FLAGS_LHINT_512_AND_SMALLER,
@@ -243,24 +447,47 @@ const u16 bnxt_lhint_arr[] = {
 	TX_BD_FLAGS_LHINT_2048_AND_LARGER,
 };
 
+static u16 bnxt_xmit_get_cfa_action(struct sk_buff *skb)
+{
+	struct metadata_dst *md_dst = skb_metadata_dst(skb);
+
+	if (!md_dst || md_dst->type != METADATA_HW_PORT_MUX)
+		return 0;
+
+	return md_dst->u.port_info.port_id;
+}
+
+static void bnxt_txr_db_kick(struct bnxt *bp, struct bnxt_tx_ring_info *txr,
+			     u16 prod)
+{
+	/* Sync BD data before updating doorbell */
+	wmb();
+	bnxt_db_write(bp, &txr->tx_db, prod);
+	txr->kick_pending = 0;
+}
+
 static netdev_tx_t bnxt_start_xmit(struct sk_buff *skb, struct net_device *dev)
 {
 	struct bnxt *bp = netdev_priv(dev);
-	struct tx_bd *txbd;
+	struct tx_bd *txbd, *txbd0;
 	struct tx_bd_ext *txbd1;
 	struct netdev_queue *txq;
 	int i;
 	dma_addr_t mapping;
 	unsigned int length, pad = 0;
 	u32 len, free_size, vlan_tag_flags, cfa_action, flags;
-	u16 prod, last_frag;
+	struct bnxt_ptp_cfg *ptp = bp->ptp_cfg;
 	struct pci_dev *pdev = bp->pdev;
+	u16 prod, last_frag, txts_prod;
 	struct bnxt_tx_ring_info *txr;
 	struct bnxt_sw_tx_bd *tx_buf;
+	__le32 lflags = 0;
+	skb_frag_t *frag;
 
 	i = skb_get_queue_mapping(skb);
 	if (unlikely(i >= bp->tx_nr_rings)) {
 		dev_kfree_skb_any(skb);
+		dev_core_stats_tx_dropped_inc(dev);
 		return NETDEV_TX_OK;
 	}
 
@@ -268,26 +495,43 @@ static netdev_tx_t bnxt_start_xmit(struct sk_buff *skb, struct net_device *dev)
 	txr = &bp->tx_ring[bp->tx_ring_map[i]];
 	prod = txr->tx_prod;
 
+#if (MAX_SKB_FRAGS > TX_MAX_FRAGS)
+	if (skb_shinfo(skb)->nr_frags > TX_MAX_FRAGS) {
+		netdev_warn_once(dev, "SKB has too many (%d) fragments, max supported is %d.  SKB will be linearized.\n",
+				 skb_shinfo(skb)->nr_frags, TX_MAX_FRAGS);
+		if (skb_linearize(skb)) {
+			dev_kfree_skb_any(skb);
+			dev_core_stats_tx_dropped_inc(dev);
+			return NETDEV_TX_OK;
+		}
+	}
+#endif
 	free_size = bnxt_tx_avail(bp, txr);
 	if (unlikely(free_size < skb_shinfo(skb)->nr_frags + 2)) {
-		netif_tx_stop_queue(txq);
-		return NETDEV_TX_BUSY;
+		/* We must have raced with NAPI cleanup */
+		if (net_ratelimit() && txr->kick_pending)
+			netif_warn(bp, tx_err, dev,
+				   "bnxt: ring busy w/ flush pending!\n");
+		if (!netif_txq_try_stop(txq, bnxt_tx_avail(bp, txr),
+					bp->tx_wake_thresh))
+			return NETDEV_TX_BUSY;
 	}
 
+	if (unlikely(ipv6_hopopt_jumbo_remove(skb)))
+		goto tx_free;
+
 	length = skb->len;
 	len = skb_headlen(skb);
 	last_frag = skb_shinfo(skb)->nr_frags;
 
-	txbd = &txr->tx_desc_ring[TX_RING(prod)][TX_IDX(prod)];
-
-	txbd->tx_bd_opaque = prod;
+	txbd = &txr->tx_desc_ring[TX_RING(bp, prod)][TX_IDX(prod)];
 
-	tx_buf = &txr->tx_buf_ring[prod];
+	tx_buf = &txr->tx_buf_ring[RING_TX(bp, prod)];
 	tx_buf->skb = skb;
 	tx_buf->nr_frags = last_frag;
 
 	vlan_tag_flags = 0;
-	cfa_action = 0;
+	cfa_action = bnxt_xmit_get_cfa_action(skb);
 	if (skb_vlan_tag_present(skb)) {
 		vlan_tag_flags = TX_BD_CFA_META_KEY_VLAN |
 				 skb_vlan_tag_get(skb);
@@ -298,10 +542,38 @@ static netdev_tx_t bnxt_start_xmit(struct sk_buff *skb, struct net_device *dev)
 			vlan_tag_flags |= 1 << TX_BD_CFA_META_TPID_SHIFT;
 	}
 
-	if (free_size == bp->tx_ring_size && length <= bp->tx_push_thresh) {
+	if (unlikely(skb_shinfo(skb)->tx_flags & SKBTX_HW_TSTAMP) && ptp &&
+	    ptp->tx_tstamp_en) {
+		if (bp->fw_cap & BNXT_FW_CAP_TX_TS_CMP) {
+			lflags |= cpu_to_le32(TX_BD_FLAGS_STAMP);
+			tx_buf->is_ts_pkt = 1;
+			skb_shinfo(skb)->tx_flags |= SKBTX_IN_PROGRESS;
+		} else if (!skb_is_gso(skb)) {
+			u16 seq_id, hdr_off;
+
+			if (!bnxt_ptp_parse(skb, &seq_id, &hdr_off) &&
+			    !bnxt_ptp_get_txts_prod(ptp, &txts_prod)) {
+				if (vlan_tag_flags)
+					hdr_off += VLAN_HLEN;
+				lflags |= cpu_to_le32(TX_BD_FLAGS_STAMP);
+				tx_buf->is_ts_pkt = 1;
+				skb_shinfo(skb)->tx_flags |= SKBTX_IN_PROGRESS;
+
+				ptp->txts_req[txts_prod].tx_seqid = seq_id;
+				ptp->txts_req[txts_prod].tx_hdr_off = hdr_off;
+				tx_buf->txts_prod = txts_prod;
+			}
+		}
+	}
+	if (unlikely(skb->no_fcs))
+		lflags |= cpu_to_le32(TX_BD_FLAGS_NO_CRC);
+
+	if (free_size == bp->tx_ring_size && length <= bp->tx_push_thresh &&
+	    skb_frags_readable(skb) && !lflags) {
 		struct tx_push_buffer *tx_push_buf = txr->tx_push;
 		struct tx_push_bd *tx_push = &tx_push_buf->push_bd;
 		struct tx_bd_ext *tx_push1 = &tx_push->txbd2;
+		void __iomem *db = txr->tx_db.doorbell;
 		void *pdata = tx_push_buf->data;
 		u64 *end;
 		int j, push_len;
@@ -313,7 +585,7 @@ static netdev_tx_t bnxt_start_xmit(struct sk_buff *skb, struct net_device *dev)
 					TX_BD_FLAGS_LHINT_512_AND_SMALLER |
 					TX_BD_FLAGS_COAL_NOW |
 					TX_BD_FLAGS_PACKET_END |
-					(2 << TX_BD_FLAGS_BD_CNT_SHIFT));
+					TX_BD_CNT(2));
 
 		if (skb->ip_summed == CHECKSUM_PARTIAL)
 			tx_push1->tx_bd_hsize_lflags =
@@ -322,7 +594,8 @@ static netdev_tx_t bnxt_start_xmit(struct sk_buff *skb, struct net_device *dev)
 			tx_push1->tx_bd_hsize_lflags = 0;
 
 		tx_push1->tx_bd_cfa_meta = cpu_to_le32(vlan_tag_flags);
-		tx_push1->tx_bd_cfa_action = cpu_to_le32(cfa_action);
+		tx_push1->tx_bd_cfa_action =
+			cpu_to_le32(cfa_action << TX_BD_CFA_ACTION_SHIFT);
 
 		end = pdata + length;
 		end = PTR_ALIGN(end, 8) - 1;
@@ -331,9 +604,9 @@ static netdev_tx_t bnxt_start_xmit(struct sk_buff *skb, struct net_device *dev)
 		skb_copy_from_linear_data(skb, pdata, len);
 		pdata += len;
 		for (j = 0; j < last_frag; j++) {
-			skb_frag_t *frag = &skb_shinfo(skb)->frags[j];
 			void *fptr;
 
+			frag = &skb_shinfo(skb)->frags[j];
 			fptr = skb_frag_address_safe(frag);
 			if (!fptr)
 				goto normal_tx;
@@ -344,13 +617,16 @@ static netdev_tx_t bnxt_start_xmit(struct sk_buff *skb, struct net_device *dev)
 
 		txbd->tx_bd_len_flags_type = tx_push->tx_bd_len_flags_type;
 		txbd->tx_bd_haddr = txr->data_mapping;
+		txbd->tx_bd_opaque = SET_TX_OPAQUE(bp, txr, prod, 2);
 		prod = NEXT_TX(prod);
-		txbd = &txr->tx_desc_ring[TX_RING(prod)][TX_IDX(prod)];
+		tx_push->tx_bd_opaque = txbd->tx_bd_opaque;
+		txbd = &txr->tx_desc_ring[TX_RING(bp, prod)][TX_IDX(prod)];
 		memcpy(txbd, tx_push1, sizeof(*txbd));
 		prod = NEXT_TX(prod);
 		tx_push->doorbell =
-			cpu_to_le32(DB_KEY_TX_PUSH | DB_LONG_TX_PUSH | prod);
-		txr->tx_prod = prod;
+			cpu_to_le32(DB_KEY_TX_PUSH | DB_LONG_TX_PUSH |
+				    DB_RING_IDX(&txr->tx_db, prod));
+		WRITE_ONCE(txr->tx_prod, prod);
 
 		tx_buf->is_push = 1;
 		netdev_tx_sent_queue(txq, skb->len);
@@ -358,12 +634,11 @@ static netdev_tx_t bnxt_start_xmit(struct sk_buff *skb, struct net_device *dev)
 
 		push_len = (length + sizeof(*tx_push) + 7) / 8;
 		if (push_len > 16) {
-			__iowrite64_copy(txr->tx_doorbell, tx_push_buf, 16);
-			__iowrite32_copy(txr->tx_doorbell + 4, tx_push_buf + 1,
+			__iowrite64_copy(db, tx_push_buf, 16);
+			__iowrite32_copy(db + 4, tx_push_buf + 1,
 					 (push_len - 16) << 1);
 		} else {
-			__iowrite64_copy(txr->tx_doorbell, tx_push_buf,
-					 push_len);
+			__iowrite64_copy(db, tx_push_buf, push_len);
 		}
 
 		goto tx_done;
@@ -372,67 +647,76 @@ static netdev_tx_t bnxt_start_xmit(struct sk_buff *skb, struct net_device *dev)
 normal_tx:
 	if (length < BNXT_MIN_PKT_SIZE) {
 		pad = BNXT_MIN_PKT_SIZE - length;
-		if (skb_pad(skb, pad)) {
+		if (skb_pad(skb, pad))
 			/* SKB already freed. */
-			tx_buf->skb = NULL;
-			return NETDEV_TX_OK;
-		}
+			goto tx_kick_pending;
 		length = BNXT_MIN_PKT_SIZE;
 	}
 
 	mapping = dma_map_single(&pdev->dev, skb->data, len, DMA_TO_DEVICE);
 
-	if (unlikely(dma_mapping_error(&pdev->dev, mapping))) {
-		dev_kfree_skb_any(skb);
-		tx_buf->skb = NULL;
-		return NETDEV_TX_OK;
-	}
+	if (unlikely(dma_mapping_error(&pdev->dev, mapping)))
+		goto tx_free;
 
 	dma_unmap_addr_set(tx_buf, mapping, mapping);
 	flags = (len << TX_BD_LEN_SHIFT) | TX_BD_TYPE_LONG_TX_BD |
-		((last_frag + 2) << TX_BD_FLAGS_BD_CNT_SHIFT);
+		TX_BD_CNT(last_frag + 2);
 
 	txbd->tx_bd_haddr = cpu_to_le64(mapping);
+	txbd->tx_bd_opaque = SET_TX_OPAQUE(bp, txr, prod, 2 + last_frag);
 
 	prod = NEXT_TX(prod);
 	txbd1 = (struct tx_bd_ext *)
-		&txr->tx_desc_ring[TX_RING(prod)][TX_IDX(prod)];
+		&txr->tx_desc_ring[TX_RING(bp, prod)][TX_IDX(prod)];
 
-	txbd1->tx_bd_hsize_lflags = 0;
+	txbd1->tx_bd_hsize_lflags = lflags;
 	if (skb_is_gso(skb)) {
+		bool udp_gso = !!(skb_shinfo(skb)->gso_type & SKB_GSO_UDP_L4);
 		u32 hdr_len;
 
-		if (skb->encapsulation)
-			hdr_len = skb_inner_network_offset(skb) +
-				skb_inner_network_header_len(skb) +
-				inner_tcp_hdrlen(skb);
-		else
+		if (skb->encapsulation) {
+			if (udp_gso)
+				hdr_len = skb_inner_transport_offset(skb) +
+					  sizeof(struct udphdr);
+			else
+				hdr_len = skb_inner_tcp_all_headers(skb);
+		} else if (udp_gso) {
 			hdr_len = skb_transport_offset(skb) +
-				tcp_hdrlen(skb);
+				  sizeof(struct udphdr);
+		} else {
+			hdr_len = skb_tcp_all_headers(skb);
+		}
 
-		txbd1->tx_bd_hsize_lflags = cpu_to_le32(TX_BD_FLAGS_LSO |
+		txbd1->tx_bd_hsize_lflags |= cpu_to_le32(TX_BD_FLAGS_LSO |
 					TX_BD_FLAGS_T_IPID |
 					(hdr_len << (TX_BD_HSIZE_SHIFT - 1)));
 		length = skb_shinfo(skb)->gso_size;
 		txbd1->tx_bd_mss = cpu_to_le32(length);
 		length += hdr_len;
 	} else if (skb->ip_summed == CHECKSUM_PARTIAL) {
-		txbd1->tx_bd_hsize_lflags =
+		txbd1->tx_bd_hsize_lflags |=
 			cpu_to_le32(TX_BD_FLAGS_TCP_UDP_CHKSUM);
 		txbd1->tx_bd_mss = 0;
 	}
 
 	length >>= 9;
+	if (unlikely(length >= ARRAY_SIZE(bnxt_lhint_arr))) {
+		dev_warn_ratelimited(&pdev->dev, "Dropped oversize %d bytes TX packet.\n",
+				     skb->len);
+		i = 0;
+		goto tx_dma_error;
+	}
 	flags |= bnxt_lhint_arr[length];
 	txbd->tx_bd_len_flags_type = cpu_to_le32(flags);
 
 	txbd1->tx_bd_cfa_meta = cpu_to_le32(vlan_tag_flags);
-	txbd1->tx_bd_cfa_action = cpu_to_le32(cfa_action);
+	txbd1->tx_bd_cfa_action =
+			cpu_to_le32(cfa_action << TX_BD_CFA_ACTION_SHIFT);
+	txbd0 = txbd;
 	for (i = 0; i < last_frag; i++) {
-		skb_frag_t *frag = &skb_shinfo(skb)->frags[i];
-
+		frag = &skb_shinfo(skb)->frags[i];
 		prod = NEXT_TX(prod);
-		txbd = &txr->tx_desc_ring[TX_RING(prod)][TX_IDX(prod)];
+		txbd = &txr->tx_desc_ring[TX_RING(bp, prod)][TX_IDX(prod)];
 
 		len = skb_frag_size(frag);
 		mapping = skb_frag_dma_map(&pdev->dev, frag, 0, len,
@@ -441,8 +725,9 @@ normal_tx:
 		if (unlikely(dma_mapping_error(&pdev->dev, mapping)))
 			goto tx_dma_error;
 
-		tx_buf = &txr->tx_buf_ring[prod];
-		dma_unmap_addr_set(tx_buf, mapping, mapping);
+		tx_buf = &txr->tx_buf_ring[RING_TX(bp, prod)];
+		netmem_dma_unmap_addr_set(skb_frag_netmem(frag), tx_buf,
+					  mapping, mapping);
 
 		txbd->tx_bd_haddr = cpu_to_le64(mapping);
 
@@ -457,33 +742,31 @@ normal_tx:
 
 	netdev_tx_sent_queue(txq, skb->len);
 
-	/* Sync BD data before updating doorbell */
-	wmb();
+	skb_tx_timestamp(skb);
 
 	prod = NEXT_TX(prod);
-	txr->tx_prod = prod;
+	WRITE_ONCE(txr->tx_prod, prod);
 
-	if (!skb->xmit_more || netif_xmit_stopped(txq))
-		bnxt_db_write(bp, txr->tx_doorbell, DB_KEY_TX | prod);
+	if (!netdev_xmit_more() || netif_xmit_stopped(txq)) {
+		bnxt_txr_db_kick(bp, txr, prod);
+	} else {
+		if (free_size >= bp->tx_wake_thresh)
+			txbd0->tx_bd_len_flags_type |=
+				cpu_to_le32(TX_BD_FLAGS_NO_CMPL);
+		txr->kick_pending = 1;
+	}
 
 tx_done:
 
-	mmiowb();
-
 	if (unlikely(bnxt_tx_avail(bp, txr) <= MAX_SKB_FRAGS + 1)) {
-		if (skb->xmit_more && !tx_buf->is_push)
-			bnxt_db_write(bp, txr->tx_doorbell, DB_KEY_TX | prod);
-
-		netif_tx_stop_queue(txq);
+		if (netdev_xmit_more() && !tx_buf->is_push) {
+			txbd0->tx_bd_len_flags_type &=
+				cpu_to_le32(~TX_BD_FLAGS_NO_CMPL);
+			bnxt_txr_db_kick(bp, txr, prod);
+		}
 
-		/* netif_tx_stop_queue() must be done before checking
-		 * tx index in bnxt_tx_avail() below, because in
-		 * bnxt_tx_int(), we update tx index before checking for
-		 * netif_tx_queue_stopped().
-		 */
-		smp_mb();
-		if (bnxt_tx_avail(bp, txr) > bp->tx_wake_thresh)
-			netif_tx_wake_queue(txq);
+		netif_txq_try_stop(txq, bnxt_tx_avail(bp, txr),
+				   bp->tx_wake_thresh);
 	}
 	return NETDEV_TX_OK;
 
@@ -492,43 +775,77 @@ tx_dma_error:
 
 	/* start back at beginning and unmap skb */
 	prod = txr->tx_prod;
-	tx_buf = &txr->tx_buf_ring[prod];
-	tx_buf->skb = NULL;
+	tx_buf = &txr->tx_buf_ring[RING_TX(bp, prod)];
 	dma_unmap_single(&pdev->dev, dma_unmap_addr(tx_buf, mapping),
-			 skb_headlen(skb), PCI_DMA_TODEVICE);
+			 skb_headlen(skb), DMA_TO_DEVICE);
 	prod = NEXT_TX(prod);
 
 	/* unmap remaining mapped pages */
 	for (i = 0; i < last_frag; i++) {
 		prod = NEXT_TX(prod);
-		tx_buf = &txr->tx_buf_ring[prod];
-		dma_unmap_page(&pdev->dev, dma_unmap_addr(tx_buf, mapping),
-			       skb_frag_size(&skb_shinfo(skb)->frags[i]),
-			       PCI_DMA_TODEVICE);
+		tx_buf = &txr->tx_buf_ring[RING_TX(bp, prod)];
+		frag = &skb_shinfo(skb)->frags[i];
+		netmem_dma_unmap_page_attrs(&pdev->dev,
+					    dma_unmap_addr(tx_buf, mapping),
+					    skb_frag_size(frag),
+					    DMA_TO_DEVICE, 0);
 	}
 
+tx_free:
 	dev_kfree_skb_any(skb);
+tx_kick_pending:
+	if (BNXT_TX_PTP_IS_SET(lflags)) {
+		txr->tx_buf_ring[RING_TX(bp, txr->tx_prod)].is_ts_pkt = 0;
+		atomic64_inc(&bp->ptp_cfg->stats.ts_err);
+		if (!(bp->fw_cap & BNXT_FW_CAP_TX_TS_CMP))
+			/* set SKB to err so PTP worker will clean up */
+			ptp->txts_req[txts_prod].tx_skb = ERR_PTR(-EIO);
+	}
+	if (txr->kick_pending)
+		bnxt_txr_db_kick(bp, txr, txr->tx_prod);
+	txr->tx_buf_ring[RING_TX(bp, txr->tx_prod)].skb = NULL;
+	dev_core_stats_tx_dropped_inc(dev);
 	return NETDEV_TX_OK;
 }
 
-static void bnxt_tx_int(struct bnxt *bp, struct bnxt_napi *bnapi, int nr_pkts)
+/* Returns true if some remaining TX packets not processed. */
+static bool __bnxt_tx_int(struct bnxt *bp, struct bnxt_tx_ring_info *txr,
+			  int budget)
 {
-	struct bnxt_tx_ring_info *txr = bnapi->tx_ring;
 	struct netdev_queue *txq = netdev_get_tx_queue(bp->dev, txr->txq_index);
-	u16 cons = txr->tx_cons;
 	struct pci_dev *pdev = bp->pdev;
-	int i;
+	u16 hw_cons = txr->tx_hw_cons;
 	unsigned int tx_bytes = 0;
+	u16 cons = txr->tx_cons;
+	skb_frag_t *frag;
+	int tx_pkts = 0;
+	bool rc = false;
 
-	for (i = 0; i < nr_pkts; i++) {
+	while (RING_TX(bp, cons) != hw_cons) {
 		struct bnxt_sw_tx_bd *tx_buf;
 		struct sk_buff *skb;
+		bool is_ts_pkt;
 		int j, last;
 
-		tx_buf = &txr->tx_buf_ring[cons];
-		cons = NEXT_TX(cons);
+		tx_buf = &txr->tx_buf_ring[RING_TX(bp, cons)];
 		skb = tx_buf->skb;
+
+		if (unlikely(!skb)) {
+			bnxt_sched_reset_txr(bp, txr, cons);
+			return rc;
+		}
+
+		is_ts_pkt = tx_buf->is_ts_pkt;
+		if (is_ts_pkt && (bp->fw_cap & BNXT_FW_CAP_TX_TS_CMP)) {
+			rc = true;
+			break;
+		}
+
+		cons = NEXT_TX(cons);
+		tx_pkts++;
+		tx_bytes += skb->len;
 		tx_buf->skb = NULL;
+		tx_buf->is_ts_pkt = 0;
 
 		if (tx_buf->is_push) {
 			tx_buf->is_push = 0;
@@ -536,105 +853,138 @@ static void bnxt_tx_int(struct bnxt *bp, struct bnxt_napi *bnapi, int nr_pkts)
 		}
 
 		dma_unmap_single(&pdev->dev, dma_unmap_addr(tx_buf, mapping),
-				 skb_headlen(skb), PCI_DMA_TODEVICE);
+				 skb_headlen(skb), DMA_TO_DEVICE);
 		last = tx_buf->nr_frags;
 
 		for (j = 0; j < last; j++) {
+			frag = &skb_shinfo(skb)->frags[j];
 			cons = NEXT_TX(cons);
-			tx_buf = &txr->tx_buf_ring[cons];
-			dma_unmap_page(
-				&pdev->dev,
-				dma_unmap_addr(tx_buf, mapping),
-				skb_frag_size(&skb_shinfo(skb)->frags[j]),
-				PCI_DMA_TODEVICE);
+			tx_buf = &txr->tx_buf_ring[RING_TX(bp, cons)];
+			netmem_dma_unmap_page_attrs(&pdev->dev,
+						    dma_unmap_addr(tx_buf,
+								   mapping),
+						    skb_frag_size(frag),
+						    DMA_TO_DEVICE, 0);
+		}
+		if (unlikely(is_ts_pkt)) {
+			if (BNXT_CHIP_P5(bp)) {
+				/* PTP worker takes ownership of the skb */
+				bnxt_get_tx_ts_p5(bp, skb, tx_buf->txts_prod);
+				skb = NULL;
+			}
 		}
 
 next_tx_int:
 		cons = NEXT_TX(cons);
 
-		tx_bytes += skb->len;
-		dev_kfree_skb_any(skb);
+		dev_consume_skb_any(skb);
 	}
 
-	netdev_tx_completed_queue(txq, nr_pkts, tx_bytes);
-	txr->tx_cons = cons;
+	WRITE_ONCE(txr->tx_cons, cons);
 
-	/* Need to make the tx_cons update visible to bnxt_start_xmit()
-	 * before checking for netif_tx_queue_stopped().  Without the
-	 * memory barrier, there is a small possibility that bnxt_start_xmit()
-	 * will miss it and cause the queue to be stopped forever.
-	 */
-	smp_mb();
+	__netif_txq_completed_wake(txq, tx_pkts, tx_bytes,
+				   bnxt_tx_avail(bp, txr), bp->tx_wake_thresh,
+				   READ_ONCE(txr->dev_state) == BNXT_DEV_STATE_CLOSING);
+
+	return rc;
+}
+
+static void bnxt_tx_int(struct bnxt *bp, struct bnxt_napi *bnapi, int budget)
+{
+	struct bnxt_tx_ring_info *txr;
+	bool more = false;
+	int i;
 
-	if (unlikely(netif_tx_queue_stopped(txq)) &&
-	    (bnxt_tx_avail(bp, txr) > bp->tx_wake_thresh)) {
-		__netif_tx_lock(txq, smp_processor_id());
-		if (netif_tx_queue_stopped(txq) &&
-		    bnxt_tx_avail(bp, txr) > bp->tx_wake_thresh &&
-		    txr->dev_state != BNXT_DEV_STATE_CLOSING)
-			netif_tx_wake_queue(txq);
-		__netif_tx_unlock(txq);
+	bnxt_for_each_napi_tx(i, bnapi, txr) {
+		if (txr->tx_hw_cons != RING_TX(bp, txr->tx_cons))
+			more |= __bnxt_tx_int(bp, txr, budget);
 	}
+	if (!more)
+		bnapi->events &= ~BNXT_TX_CMP_EVENT;
+}
+
+static bool bnxt_separate_head_pool(struct bnxt_rx_ring_info *rxr)
+{
+	return rxr->need_head_pool || PAGE_SIZE > BNXT_RX_PAGE_SIZE;
 }
 
 static struct page *__bnxt_alloc_rx_page(struct bnxt *bp, dma_addr_t *mapping,
+					 struct bnxt_rx_ring_info *rxr,
+					 unsigned int *offset,
 					 gfp_t gfp)
 {
-	struct device *dev = &bp->pdev->dev;
 	struct page *page;
 
-	page = alloc_page(gfp);
+	if (PAGE_SIZE > BNXT_RX_PAGE_SIZE) {
+		page = page_pool_dev_alloc_frag(rxr->page_pool, offset,
+						BNXT_RX_PAGE_SIZE);
+	} else {
+		page = page_pool_dev_alloc_pages(rxr->page_pool);
+		*offset = 0;
+	}
 	if (!page)
 		return NULL;
 
-	*mapping = dma_map_page_attrs(dev, page, 0, PAGE_SIZE, bp->rx_dir,
-				      DMA_ATTR_WEAK_ORDERING);
-	if (dma_mapping_error(dev, *mapping)) {
-		__free_page(page);
-		return NULL;
-	}
-	*mapping += bp->rx_dma_offset;
+	*mapping = page_pool_get_dma_addr(page) + *offset;
 	return page;
 }
 
-static inline u8 *__bnxt_alloc_rx_data(struct bnxt *bp, dma_addr_t *mapping,
+static netmem_ref __bnxt_alloc_rx_netmem(struct bnxt *bp, dma_addr_t *mapping,
+					 struct bnxt_rx_ring_info *rxr,
+					 unsigned int *offset,
+					 gfp_t gfp)
+{
+	netmem_ref netmem;
+
+	if (PAGE_SIZE > BNXT_RX_PAGE_SIZE) {
+		netmem = page_pool_alloc_frag_netmem(rxr->page_pool, offset, BNXT_RX_PAGE_SIZE, gfp);
+	} else {
+		netmem = page_pool_alloc_netmems(rxr->page_pool, gfp);
+		*offset = 0;
+	}
+	if (!netmem)
+		return 0;
+
+	*mapping = page_pool_get_dma_addr_netmem(netmem) + *offset;
+	return netmem;
+}
+
+static inline u8 *__bnxt_alloc_rx_frag(struct bnxt *bp, dma_addr_t *mapping,
+				       struct bnxt_rx_ring_info *rxr,
 				       gfp_t gfp)
 {
-	u8 *data;
-	struct pci_dev *pdev = bp->pdev;
+	unsigned int offset;
+	struct page *page;
 
-	data = kmalloc(bp->rx_buf_size, gfp);
-	if (!data)
+	page = page_pool_alloc_frag(rxr->head_pool, &offset,
+				    bp->rx_buf_size, gfp);
+	if (!page)
 		return NULL;
 
-	*mapping = dma_map_single_attrs(&pdev->dev, data + bp->rx_dma_offset,
-					bp->rx_buf_use_size, bp->rx_dir,
-					DMA_ATTR_WEAK_ORDERING);
-
-	if (dma_mapping_error(&pdev->dev, *mapping)) {
-		kfree(data);
-		data = NULL;
-	}
-	return data;
+	*mapping = page_pool_get_dma_addr(page) + bp->rx_dma_offset + offset;
+	return page_address(page) + offset;
 }
 
 int bnxt_alloc_rx_data(struct bnxt *bp, struct bnxt_rx_ring_info *rxr,
 		       u16 prod, gfp_t gfp)
 {
-	struct rx_bd *rxbd = &rxr->rx_desc_ring[RX_RING(prod)][RX_IDX(prod)];
-	struct bnxt_sw_rx_bd *rx_buf = &rxr->rx_buf_ring[prod];
+	struct rx_bd *rxbd = &rxr->rx_desc_ring[RX_RING(bp, prod)][RX_IDX(prod)];
+	struct bnxt_sw_rx_bd *rx_buf = &rxr->rx_buf_ring[RING_RX(bp, prod)];
 	dma_addr_t mapping;
 
 	if (BNXT_RX_PAGE_MODE(bp)) {
-		struct page *page = __bnxt_alloc_rx_page(bp, &mapping, gfp);
+		unsigned int offset;
+		struct page *page =
+			__bnxt_alloc_rx_page(bp, &mapping, rxr, &offset, gfp);
 
 		if (!page)
 			return -ENOMEM;
 
+		mapping += bp->rx_dma_offset;
 		rx_buf->data = page;
-		rx_buf->data_ptr = page_address(page) + bp->rx_offset;
+		rx_buf->data_ptr = page_address(page) + offset + bp->rx_offset;
 	} else {
-		u8 *data = __bnxt_alloc_rx_data(bp, &mapping, gfp);
+		u8 *data = __bnxt_alloc_rx_frag(bp, &mapping, rxr, gfp);
 
 		if (!data)
 			return -ENOMEM;
@@ -652,9 +1002,10 @@ void bnxt_reuse_rx_data(struct bnxt_rx_ring_info *rxr, u16 cons, void *data)
 {
 	u16 prod = rxr->rx_prod;
 	struct bnxt_sw_rx_bd *cons_rx_buf, *prod_rx_buf;
+	struct bnxt *bp = rxr->bnapi->bp;
 	struct rx_bd *cons_bd, *prod_bd;
 
-	prod_rx_buf = &rxr->rx_buf_ring[prod];
+	prod_rx_buf = &rxr->rx_buf_ring[RING_RX(bp, prod)];
 	cons_rx_buf = &rxr->rx_buf_ring[cons];
 
 	prod_rx_buf->data = data;
@@ -662,8 +1013,8 @@ void bnxt_reuse_rx_data(struct bnxt_rx_ring_info *rxr, u16 cons, void *data)
 
 	prod_rx_buf->mapping = cons_rx_buf->mapping;
 
-	prod_bd = &rxr->rx_desc_ring[RX_RING(prod)][RX_IDX(prod)];
-	cons_bd = &rxr->rx_desc_ring[RX_RING(cons)][RX_IDX(cons)];
+	prod_bd = &rxr->rx_desc_ring[RX_RING(bp, prod)][RX_IDX(prod)];
+	cons_bd = &rxr->rx_desc_ring[RX_RING(bp, cons)][RX_IDX(cons)];
 
 	prod_bd->rx_bd_haddr = cons_bd->rx_bd_haddr;
 }
@@ -678,56 +1029,29 @@ static inline u16 bnxt_find_next_agg_idx(struct bnxt_rx_ring_info *rxr, u16 idx)
 	return next;
 }
 
-static inline int bnxt_alloc_rx_page(struct bnxt *bp,
-				     struct bnxt_rx_ring_info *rxr,
-				     u16 prod, gfp_t gfp)
+static int bnxt_alloc_rx_netmem(struct bnxt *bp, struct bnxt_rx_ring_info *rxr,
+				u16 prod, gfp_t gfp)
 {
 	struct rx_bd *rxbd =
-		&rxr->rx_agg_desc_ring[RX_RING(prod)][RX_IDX(prod)];
+		&rxr->rx_agg_desc_ring[RX_AGG_RING(bp, prod)][RX_IDX(prod)];
 	struct bnxt_sw_rx_agg_bd *rx_agg_buf;
-	struct pci_dev *pdev = bp->pdev;
-	struct page *page;
-	dma_addr_t mapping;
 	u16 sw_prod = rxr->rx_sw_agg_prod;
 	unsigned int offset = 0;
+	dma_addr_t mapping;
+	netmem_ref netmem;
 
-	if (PAGE_SIZE > BNXT_RX_PAGE_SIZE) {
-		page = rxr->rx_page;
-		if (!page) {
-			page = alloc_page(gfp);
-			if (!page)
-				return -ENOMEM;
-			rxr->rx_page = page;
-			rxr->rx_page_offset = 0;
-		}
-		offset = rxr->rx_page_offset;
-		rxr->rx_page_offset += BNXT_RX_PAGE_SIZE;
-		if (rxr->rx_page_offset == PAGE_SIZE)
-			rxr->rx_page = NULL;
-		else
-			get_page(page);
-	} else {
-		page = alloc_page(gfp);
-		if (!page)
-			return -ENOMEM;
-	}
-
-	mapping = dma_map_page_attrs(&pdev->dev, page, offset,
-				     BNXT_RX_PAGE_SIZE, PCI_DMA_FROMDEVICE,
-				     DMA_ATTR_WEAK_ORDERING);
-	if (dma_mapping_error(&pdev->dev, mapping)) {
-		__free_page(page);
-		return -EIO;
-	}
+	netmem = __bnxt_alloc_rx_netmem(bp, &mapping, rxr, &offset, gfp);
+	if (!netmem)
+		return -ENOMEM;
 
 	if (unlikely(test_bit(sw_prod, rxr->rx_agg_bmap)))
 		sw_prod = bnxt_find_next_agg_idx(rxr, sw_prod);
 
 	__set_bit(sw_prod, rxr->rx_agg_bmap);
 	rx_agg_buf = &rxr->rx_agg_ring[sw_prod];
-	rxr->rx_sw_agg_prod = NEXT_RX_AGG(sw_prod);
+	rxr->rx_sw_agg_prod = RING_RX_AGG(bp, NEXT_RX_AGG(sw_prod));
 
-	rx_agg_buf->page = page;
+	rx_agg_buf->netmem = netmem;
 	rx_agg_buf->offset = offset;
 	rx_agg_buf->mapping = mapping;
 	rxbd->rx_bd_haddr = cpu_to_le64(mapping);
@@ -735,25 +1059,52 @@ static inline int bnxt_alloc_rx_page(struct bnxt *bp,
 	return 0;
 }
 
-static void bnxt_reuse_rx_agg_bufs(struct bnxt_napi *bnapi, u16 cp_cons,
-				   u32 agg_bufs)
+static struct rx_agg_cmp *bnxt_get_agg(struct bnxt *bp,
+				       struct bnxt_cp_ring_info *cpr,
+				       u16 cp_cons, u16 curr)
+{
+	struct rx_agg_cmp *agg;
+
+	cp_cons = RING_CMP(ADV_RAW_CMP(cp_cons, curr));
+	agg = (struct rx_agg_cmp *)
+		&cpr->cp_desc_ring[CP_RING(cp_cons)][CP_IDX(cp_cons)];
+	return agg;
+}
+
+static struct rx_agg_cmp *bnxt_get_tpa_agg_p5(struct bnxt *bp,
+					      struct bnxt_rx_ring_info *rxr,
+					      u16 agg_id, u16 curr)
 {
+	struct bnxt_tpa_info *tpa_info = &rxr->rx_tpa[agg_id];
+
+	return &tpa_info->agg_arr[curr];
+}
+
+static void bnxt_reuse_rx_agg_bufs(struct bnxt_cp_ring_info *cpr, u16 idx,
+				   u16 start, u32 agg_bufs, bool tpa)
+{
+	struct bnxt_napi *bnapi = cpr->bnapi;
 	struct bnxt *bp = bnapi->bp;
-	struct bnxt_cp_ring_info *cpr = &bnapi->cp_ring;
 	struct bnxt_rx_ring_info *rxr = bnapi->rx_ring;
 	u16 prod = rxr->rx_agg_prod;
 	u16 sw_prod = rxr->rx_sw_agg_prod;
+	bool p5_tpa = false;
 	u32 i;
 
+	if ((bp->flags & BNXT_FLAG_CHIP_P5_PLUS) && tpa)
+		p5_tpa = true;
+
 	for (i = 0; i < agg_bufs; i++) {
-		u16 cons;
-		struct rx_agg_cmp *agg;
 		struct bnxt_sw_rx_agg_bd *cons_rx_buf, *prod_rx_buf;
+		struct rx_agg_cmp *agg;
 		struct rx_bd *prod_bd;
-		struct page *page;
+		netmem_ref netmem;
+		u16 cons;
 
-		agg = (struct rx_agg_cmp *)
-			&cpr->cp_desc_ring[CP_RING(cp_cons)][CP_IDX(cp_cons)];
+		if (p5_tpa)
+			agg = bnxt_get_tpa_agg_p5(bp, rxr, idx, start + i);
+		else
+			agg = bnxt_get_agg(bp, cpr, idx, start + i);
 		cons = agg->rx_agg_cmp_opaque;
 		__clear_bit(cons, rxr->rx_agg_bmap);
 
@@ -765,28 +1116,59 @@ static void bnxt_reuse_rx_agg_bufs(struct bnxt_napi *bnapi, u16 cp_cons,
 		cons_rx_buf = &rxr->rx_agg_ring[cons];
 
 		/* It is possible for sw_prod to be equal to cons, so
-		 * set cons_rx_buf->page to NULL first.
+		 * set cons_rx_buf->netmem to 0 first.
 		 */
-		page = cons_rx_buf->page;
-		cons_rx_buf->page = NULL;
-		prod_rx_buf->page = page;
+		netmem = cons_rx_buf->netmem;
+		cons_rx_buf->netmem = 0;
+		prod_rx_buf->netmem = netmem;
 		prod_rx_buf->offset = cons_rx_buf->offset;
 
 		prod_rx_buf->mapping = cons_rx_buf->mapping;
 
-		prod_bd = &rxr->rx_agg_desc_ring[RX_RING(prod)][RX_IDX(prod)];
+		prod_bd = &rxr->rx_agg_desc_ring[RX_AGG_RING(bp, prod)][RX_IDX(prod)];
 
 		prod_bd->rx_bd_haddr = cpu_to_le64(cons_rx_buf->mapping);
 		prod_bd->rx_bd_opaque = sw_prod;
 
 		prod = NEXT_RX_AGG(prod);
-		sw_prod = NEXT_RX_AGG(sw_prod);
-		cp_cons = NEXT_CMP(cp_cons);
+		sw_prod = RING_RX_AGG(bp, NEXT_RX_AGG(sw_prod));
 	}
 	rxr->rx_agg_prod = prod;
 	rxr->rx_sw_agg_prod = sw_prod;
 }
 
+static struct sk_buff *bnxt_rx_multi_page_skb(struct bnxt *bp,
+					      struct bnxt_rx_ring_info *rxr,
+					      u16 cons, void *data, u8 *data_ptr,
+					      dma_addr_t dma_addr,
+					      unsigned int offset_and_len)
+{
+	unsigned int len = offset_and_len & 0xffff;
+	struct page *page = data;
+	u16 prod = rxr->rx_prod;
+	struct sk_buff *skb;
+	int err;
+
+	err = bnxt_alloc_rx_data(bp, rxr, prod, GFP_ATOMIC);
+	if (unlikely(err)) {
+		bnxt_reuse_rx_data(rxr, cons, data);
+		return NULL;
+	}
+	dma_addr -= bp->rx_dma_offset;
+	dma_sync_single_for_cpu(&bp->pdev->dev, dma_addr, BNXT_RX_PAGE_SIZE,
+				bp->rx_dir);
+	skb = napi_build_skb(data_ptr - bp->rx_offset, BNXT_RX_PAGE_SIZE);
+	if (!skb) {
+		page_pool_recycle_direct(rxr->page_pool, page);
+		return NULL;
+	}
+	skb_mark_for_recycle(skb);
+	skb_reserve(skb, bp->rx_offset);
+	__skb_put(skb, len);
+
+	return skb;
+}
+
 static struct sk_buff *bnxt_rx_page_skb(struct bnxt *bp,
 					struct bnxt_rx_ring_info *rxr,
 					u16 cons, void *data, u8 *data_ptr,
@@ -795,7 +1177,7 @@ static struct sk_buff *bnxt_rx_page_skb(struct bnxt *bp,
 {
 	unsigned int payload = offset_and_len >> 16;
 	unsigned int len = offset_and_len & 0xffff;
-	struct skb_frag_struct *frag;
+	skb_frag_t *frag;
 	struct page *page = data;
 	u16 prod = rxr->rx_prod;
 	struct sk_buff *skb;
@@ -807,26 +1189,27 @@ static struct sk_buff *bnxt_rx_page_skb(struct bnxt *bp,
 		return NULL;
 	}
 	dma_addr -= bp->rx_dma_offset;
-	dma_unmap_page_attrs(&bp->pdev->dev, dma_addr, PAGE_SIZE, bp->rx_dir,
-			     DMA_ATTR_WEAK_ORDERING);
+	dma_sync_single_for_cpu(&bp->pdev->dev, dma_addr, BNXT_RX_PAGE_SIZE,
+				bp->rx_dir);
 
 	if (unlikely(!payload))
-		payload = eth_get_headlen(data_ptr, len);
+		payload = eth_get_headlen(bp->dev, data_ptr, len);
 
 	skb = napi_alloc_skb(&rxr->bnapi->napi, payload);
 	if (!skb) {
-		__free_page(page);
+		page_pool_recycle_direct(rxr->page_pool, page);
 		return NULL;
 	}
 
+	skb_mark_for_recycle(skb);
 	off = (void *)data_ptr - page_address(page);
-	skb_add_rx_frag(skb, 0, page, off, len, PAGE_SIZE);
+	skb_add_rx_frag(skb, 0, page, off, len, BNXT_RX_PAGE_SIZE);
 	memcpy(skb->data - NET_IP_ALIGN, data_ptr - NET_IP_ALIGN,
 	       payload + NET_IP_ALIGN);
 
 	frag = &skb_shinfo(skb)->frags[0];
 	skb_frag_size_sub(frag, payload);
-	frag->page_offset += payload;
+	skb_frag_off_add(frag, payload);
 	skb->data_len -= payload;
 	skb->tail += payload;
 
@@ -849,90 +1232,150 @@ static struct sk_buff *bnxt_rx_skb(struct bnxt *bp,
 		return NULL;
 	}
 
-	skb = build_skb(data, 0);
-	dma_unmap_single_attrs(&bp->pdev->dev, dma_addr, bp->rx_buf_use_size,
-			       bp->rx_dir, DMA_ATTR_WEAK_ORDERING);
+	skb = napi_build_skb(data, bp->rx_buf_size);
+	dma_sync_single_for_cpu(&bp->pdev->dev, dma_addr, bp->rx_buf_use_size,
+				bp->rx_dir);
 	if (!skb) {
-		kfree(data);
+		page_pool_free_va(rxr->head_pool, data, true);
 		return NULL;
 	}
 
+	skb_mark_for_recycle(skb);
 	skb_reserve(skb, bp->rx_offset);
 	skb_put(skb, offset_and_len & 0xffff);
 	return skb;
 }
 
-static struct sk_buff *bnxt_rx_pages(struct bnxt *bp, struct bnxt_napi *bnapi,
-				     struct sk_buff *skb, u16 cp_cons,
-				     u32 agg_bufs)
+static u32 __bnxt_rx_agg_netmems(struct bnxt *bp,
+				 struct bnxt_cp_ring_info *cpr,
+				 u16 idx, u32 agg_bufs, bool tpa,
+				 struct sk_buff *skb,
+				 struct xdp_buff *xdp)
 {
-	struct pci_dev *pdev = bp->pdev;
-	struct bnxt_cp_ring_info *cpr = &bnapi->cp_ring;
-	struct bnxt_rx_ring_info *rxr = bnapi->rx_ring;
-	u16 prod = rxr->rx_agg_prod;
-	u32 i;
+	struct bnxt_napi *bnapi = cpr->bnapi;
+	struct skb_shared_info *shinfo;
+	struct bnxt_rx_ring_info *rxr;
+	u32 i, total_frag_len = 0;
+	bool p5_tpa = false;
+	u16 prod;
+
+	rxr = bnapi->rx_ring;
+	prod = rxr->rx_agg_prod;
+
+	if ((bp->flags & BNXT_FLAG_CHIP_P5_PLUS) && tpa)
+		p5_tpa = true;
+
+	if (skb)
+		shinfo = skb_shinfo(skb);
+	else
+		shinfo = xdp_get_shared_info_from_buff(xdp);
 
 	for (i = 0; i < agg_bufs; i++) {
-		u16 cons, frag_len;
-		struct rx_agg_cmp *agg;
 		struct bnxt_sw_rx_agg_bd *cons_rx_buf;
-		struct page *page;
-		dma_addr_t mapping;
+		struct rx_agg_cmp *agg;
+		u16 cons, frag_len;
+		netmem_ref netmem;
 
-		agg = (struct rx_agg_cmp *)
-			&cpr->cp_desc_ring[CP_RING(cp_cons)][CP_IDX(cp_cons)];
+		if (p5_tpa)
+			agg = bnxt_get_tpa_agg_p5(bp, rxr, idx, i);
+		else
+			agg = bnxt_get_agg(bp, cpr, idx, i);
 		cons = agg->rx_agg_cmp_opaque;
 		frag_len = (le32_to_cpu(agg->rx_agg_cmp_len_flags_type) &
 			    RX_AGG_CMP_LEN) >> RX_AGG_CMP_LEN_SHIFT;
 
 		cons_rx_buf = &rxr->rx_agg_ring[cons];
-		skb_fill_page_desc(skb, i, cons_rx_buf->page,
-				   cons_rx_buf->offset, frag_len);
+		if (skb) {
+			skb_add_rx_frag_netmem(skb, i, cons_rx_buf->netmem,
+					       cons_rx_buf->offset,
+					       frag_len, BNXT_RX_PAGE_SIZE);
+		} else {
+			skb_frag_t *frag = &shinfo->frags[i];
+
+			skb_frag_fill_netmem_desc(frag, cons_rx_buf->netmem,
+						  cons_rx_buf->offset,
+						  frag_len);
+			shinfo->nr_frags = i + 1;
+		}
 		__clear_bit(cons, rxr->rx_agg_bmap);
 
-		/* It is possible for bnxt_alloc_rx_page() to allocate
+		/* It is possible for bnxt_alloc_rx_netmem() to allocate
 		 * a sw_prod index that equals the cons index, so we
 		 * need to clear the cons entry now.
 		 */
-		mapping = cons_rx_buf->mapping;
-		page = cons_rx_buf->page;
-		cons_rx_buf->page = NULL;
+		netmem = cons_rx_buf->netmem;
+		cons_rx_buf->netmem = 0;
 
-		if (bnxt_alloc_rx_page(bp, rxr, prod, GFP_ATOMIC) != 0) {
-			struct skb_shared_info *shinfo;
-			unsigned int nr_frags;
+		if (xdp && netmem_is_pfmemalloc(netmem))
+			xdp_buff_set_frag_pfmemalloc(xdp);
 
-			shinfo = skb_shinfo(skb);
-			nr_frags = --shinfo->nr_frags;
-			__skb_frag_set_page(&shinfo->frags[nr_frags], NULL);
-
-			dev_kfree_skb(skb);
+		if (bnxt_alloc_rx_netmem(bp, rxr, prod, GFP_ATOMIC) != 0) {
+			if (skb) {
+				skb->len -= frag_len;
+				skb->data_len -= frag_len;
+				skb->truesize -= BNXT_RX_PAGE_SIZE;
+			}
 
-			cons_rx_buf->page = page;
+			--shinfo->nr_frags;
+			cons_rx_buf->netmem = netmem;
 
-			/* Update prod since possibly some pages have been
+			/* Update prod since possibly some netmems have been
 			 * allocated already.
 			 */
 			rxr->rx_agg_prod = prod;
-			bnxt_reuse_rx_agg_bufs(bnapi, cp_cons, agg_bufs - i);
-			return NULL;
+			bnxt_reuse_rx_agg_bufs(cpr, idx, i, agg_bufs - i, tpa);
+			return 0;
 		}
 
-		dma_unmap_page_attrs(&pdev->dev, mapping, BNXT_RX_PAGE_SIZE,
-				     PCI_DMA_FROMDEVICE,
-				     DMA_ATTR_WEAK_ORDERING);
-
-		skb->data_len += frag_len;
-		skb->len += frag_len;
-		skb->truesize += PAGE_SIZE;
+		page_pool_dma_sync_netmem_for_cpu(rxr->page_pool, netmem, 0,
+						  BNXT_RX_PAGE_SIZE);
 
+		total_frag_len += frag_len;
 		prod = NEXT_RX_AGG(prod);
-		cp_cons = NEXT_CMP(cp_cons);
 	}
 	rxr->rx_agg_prod = prod;
+	return total_frag_len;
+}
+
+static struct sk_buff *bnxt_rx_agg_netmems_skb(struct bnxt *bp,
+					       struct bnxt_cp_ring_info *cpr,
+					       struct sk_buff *skb, u16 idx,
+					       u32 agg_bufs, bool tpa)
+{
+	u32 total_frag_len = 0;
+
+	total_frag_len = __bnxt_rx_agg_netmems(bp, cpr, idx, agg_bufs, tpa,
+					       skb, NULL);
+	if (!total_frag_len) {
+		skb_mark_for_recycle(skb);
+		dev_kfree_skb(skb);
+		return NULL;
+	}
+
 	return skb;
 }
 
+static u32 bnxt_rx_agg_netmems_xdp(struct bnxt *bp,
+				   struct bnxt_cp_ring_info *cpr,
+				   struct xdp_buff *xdp, u16 idx,
+				   u32 agg_bufs, bool tpa)
+{
+	struct skb_shared_info *shinfo = xdp_get_shared_info_from_buff(xdp);
+	u32 total_frag_len = 0;
+
+	if (!xdp_buff_has_frags(xdp))
+		shinfo->nr_frags = 0;
+
+	total_frag_len = __bnxt_rx_agg_netmems(bp, cpr, idx, agg_bufs, tpa,
+					       NULL, xdp);
+	if (total_frag_len) {
+		xdp_buff_set_frags_flag(xdp);
+		shinfo->nr_frags = agg_bufs;
+		shinfo->xdp_frags_size = total_frag_len;
+	}
+	return total_frag_len;
+}
+
 static int bnxt_agg_bufs_valid(struct bnxt *bp, struct bnxt_cp_ring_info *cpr,
 			       u8 agg_bufs, u32 *raw_cons)
 {
@@ -946,9 +1389,9 @@ static int bnxt_agg_bufs_valid(struct bnxt *bp, struct bnxt_cp_ring_info *cpr,
 	return RX_AGG_CMP_VALID(agg, *raw_cons);
 }
 
-static inline struct sk_buff *bnxt_copy_skb(struct bnxt_napi *bnapi, u8 *data,
-					    unsigned int len,
-					    dma_addr_t mapping)
+static struct sk_buff *bnxt_copy_data(struct bnxt_napi *bnapi, u8 *data,
+				      unsigned int len,
+				      dma_addr_t mapping)
 {
 	struct bnxt *bp = bnapi->bp;
 	struct pci_dev *pdev = bp->pdev;
@@ -958,23 +1401,55 @@ static inline struct sk_buff *bnxt_copy_skb(struct bnxt_napi *bnapi, u8 *data,
 	if (!skb)
 		return NULL;
 
-	dma_sync_single_for_cpu(&pdev->dev, mapping, bp->rx_copy_thresh,
+	dma_sync_single_for_cpu(&pdev->dev, mapping, bp->rx_copybreak,
 				bp->rx_dir);
 
 	memcpy(skb->data - NET_IP_ALIGN, data - NET_IP_ALIGN,
 	       len + NET_IP_ALIGN);
 
-	dma_sync_single_for_device(&pdev->dev, mapping, bp->rx_copy_thresh,
+	dma_sync_single_for_device(&pdev->dev, mapping, bp->rx_copybreak,
 				   bp->rx_dir);
 
 	skb_put(skb, len);
+
 	return skb;
 }
 
-static int bnxt_discard_rx(struct bnxt *bp, struct bnxt_napi *bnapi,
+static struct sk_buff *bnxt_copy_skb(struct bnxt_napi *bnapi, u8 *data,
+				     unsigned int len,
+				     dma_addr_t mapping)
+{
+	return bnxt_copy_data(bnapi, data, len, mapping);
+}
+
+static struct sk_buff *bnxt_copy_xdp(struct bnxt_napi *bnapi,
+				     struct xdp_buff *xdp,
+				     unsigned int len,
+				     dma_addr_t mapping)
+{
+	unsigned int metasize = 0;
+	u8 *data = xdp->data;
+	struct sk_buff *skb;
+
+	len = xdp->data_end - xdp->data_meta;
+	metasize = xdp->data - xdp->data_meta;
+	data = xdp->data_meta;
+
+	skb = bnxt_copy_data(bnapi, data, len, mapping);
+	if (!skb)
+		return skb;
+
+	if (metasize) {
+		skb_metadata_set(skb, metasize);
+		__skb_pull(skb, metasize);
+	}
+
+	return skb;
+}
+
+static int bnxt_discard_rx(struct bnxt *bp, struct bnxt_cp_ring_info *cpr,
 			   u32 *raw_cons, void *cmp)
 {
-	struct bnxt_cp_ring_info *cpr = &bnapi->cp_ring;
 	struct rx_cmp *rxcmp = cmp;
 	u32 tmp_raw_cons = *raw_cons;
 	u8 cmp_type, agg_bufs = 0;
@@ -988,9 +1463,10 @@ static int bnxt_discard_rx(struct bnxt *bp, struct bnxt_napi *bnapi,
 	} else if (cmp_type == CMP_TYPE_RX_L2_TPA_END_CMP) {
 		struct rx_tpa_end_cmp *tpa_end = cmp;
 
-		agg_bufs = (le32_to_cpu(tpa_end->rx_tpa_end_cmp_misc_v1) &
-			    RX_TPA_END_CMP_AGG_BUFS) >>
-			   RX_TPA_END_CMP_AGG_BUFS_SHIFT;
+		if (bp->flags & BNXT_FLAG_CHIP_P5_PLUS)
+			return 0;
+
+		agg_bufs = TPA_END_AGG_BUFS(tpa_end);
 	}
 
 	if (agg_bufs) {
@@ -1001,45 +1477,101 @@ static int bnxt_discard_rx(struct bnxt *bp, struct bnxt_napi *bnapi,
 	return 0;
 }
 
-static void bnxt_sched_reset(struct bnxt *bp, struct bnxt_rx_ring_info *rxr)
+static u16 bnxt_alloc_agg_idx(struct bnxt_rx_ring_info *rxr, u16 agg_id)
 {
-	if (!rxr->bnapi->in_reset) {
-		rxr->bnapi->in_reset = true;
-		set_bit(BNXT_RESET_TASK_SP_EVENT, &bp->sp_event);
-		schedule_work(&bp->sp_task);
+	struct bnxt_tpa_idx_map *map = rxr->rx_tpa_idx_map;
+	u16 idx = agg_id & MAX_TPA_P5_MASK;
+
+	if (test_bit(idx, map->agg_idx_bmap))
+		idx = find_first_zero_bit(map->agg_idx_bmap,
+					  BNXT_AGG_IDX_BMAP_SIZE);
+	__set_bit(idx, map->agg_idx_bmap);
+	map->agg_id_tbl[agg_id] = idx;
+	return idx;
+}
+
+static void bnxt_free_agg_idx(struct bnxt_rx_ring_info *rxr, u16 idx)
+{
+	struct bnxt_tpa_idx_map *map = rxr->rx_tpa_idx_map;
+
+	__clear_bit(idx, map->agg_idx_bmap);
+}
+
+static u16 bnxt_lookup_agg_idx(struct bnxt_rx_ring_info *rxr, u16 agg_id)
+{
+	struct bnxt_tpa_idx_map *map = rxr->rx_tpa_idx_map;
+
+	return map->agg_id_tbl[agg_id];
+}
+
+static void bnxt_tpa_metadata(struct bnxt_tpa_info *tpa_info,
+			      struct rx_tpa_start_cmp *tpa_start,
+			      struct rx_tpa_start_cmp_ext *tpa_start1)
+{
+	tpa_info->cfa_code_valid = 1;
+	tpa_info->cfa_code = TPA_START_CFA_CODE(tpa_start1);
+	tpa_info->vlan_valid = 0;
+	if (tpa_info->flags2 & RX_CMP_FLAGS2_META_FORMAT_VLAN) {
+		tpa_info->vlan_valid = 1;
+		tpa_info->metadata =
+			le32_to_cpu(tpa_start1->rx_tpa_start_cmp_metadata);
+	}
+}
+
+static void bnxt_tpa_metadata_v2(struct bnxt_tpa_info *tpa_info,
+				 struct rx_tpa_start_cmp *tpa_start,
+				 struct rx_tpa_start_cmp_ext *tpa_start1)
+{
+	tpa_info->vlan_valid = 0;
+	if (TPA_START_VLAN_VALID(tpa_start)) {
+		u32 tpid_sel = TPA_START_VLAN_TPID_SEL(tpa_start);
+		u32 vlan_proto = ETH_P_8021Q;
+
+		tpa_info->vlan_valid = 1;
+		if (tpid_sel == RX_TPA_START_METADATA1_TPID_8021AD)
+			vlan_proto = ETH_P_8021AD;
+		tpa_info->metadata = vlan_proto << 16 |
+				     TPA_START_METADATA0_TCI(tpa_start1);
 	}
-	rxr->rx_next_cons = 0xffff;
 }
 
 static void bnxt_tpa_start(struct bnxt *bp, struct bnxt_rx_ring_info *rxr,
-			   struct rx_tpa_start_cmp *tpa_start,
+			   u8 cmp_type, struct rx_tpa_start_cmp *tpa_start,
 			   struct rx_tpa_start_cmp_ext *tpa_start1)
 {
-	u8 agg_id = TPA_START_AGG_ID(tpa_start);
-	u16 cons, prod;
-	struct bnxt_tpa_info *tpa_info;
 	struct bnxt_sw_rx_bd *cons_rx_buf, *prod_rx_buf;
+	struct bnxt_tpa_info *tpa_info;
+	u16 cons, prod, agg_id;
 	struct rx_bd *prod_bd;
 	dma_addr_t mapping;
 
+	if (bp->flags & BNXT_FLAG_CHIP_P5_PLUS) {
+		agg_id = TPA_START_AGG_ID_P5(tpa_start);
+		agg_id = bnxt_alloc_agg_idx(rxr, agg_id);
+	} else {
+		agg_id = TPA_START_AGG_ID(tpa_start);
+	}
 	cons = tpa_start->rx_tpa_start_cmp_opaque;
 	prod = rxr->rx_prod;
 	cons_rx_buf = &rxr->rx_buf_ring[cons];
-	prod_rx_buf = &rxr->rx_buf_ring[prod];
+	prod_rx_buf = &rxr->rx_buf_ring[RING_RX(bp, prod)];
 	tpa_info = &rxr->rx_tpa[agg_id];
 
-	if (unlikely(cons != rxr->rx_next_cons)) {
-		bnxt_sched_reset(bp, rxr);
+	if (unlikely(cons != rxr->rx_next_cons ||
+		     TPA_START_ERROR(tpa_start))) {
+		netdev_warn(bp->dev, "TPA cons %x, expected cons %x, error code %x\n",
+			    cons, rxr->rx_next_cons,
+			    TPA_START_ERROR_CODE(tpa_start1));
+		bnxt_sched_reset_rxr(bp, rxr);
 		return;
 	}
-
 	prod_rx_buf->data = tpa_info->data;
 	prod_rx_buf->data_ptr = tpa_info->data_ptr;
 
 	mapping = tpa_info->mapping;
 	prod_rx_buf->mapping = mapping;
 
-	prod_bd = &rxr->rx_desc_ring[RX_RING(prod)][RX_IDX(prod)];
+	prod_bd = &rxr->rx_desc_ring[RX_RING(bp, prod)][RX_IDX(prod)];
 
 	prod_bd->rx_bd_haddr = cpu_to_le64(mapping);
 
@@ -1052,28 +1584,32 @@ static void bnxt_tpa_start(struct bnxt *bp, struct bnxt_rx_ring_info *rxr,
 		le32_to_cpu(tpa_start->rx_tpa_start_cmp_len_flags_type) >>
 				RX_TPA_START_CMP_LEN_SHIFT;
 	if (likely(TPA_START_HASH_VALID(tpa_start))) {
-		u32 hash_type = TPA_START_HASH_TYPE(tpa_start);
-
 		tpa_info->hash_type = PKT_HASH_TYPE_L4;
 		tpa_info->gso_type = SKB_GSO_TCPV4;
+		if (TPA_START_IS_IPV6(tpa_start1))
+			tpa_info->gso_type = SKB_GSO_TCPV6;
 		/* RSS profiles 1 and 3 with extract code 0 for inner 4-tuple */
-		if (hash_type == 3)
+		else if (!BNXT_CHIP_P4_PLUS(bp) &&
+			 TPA_START_HASH_TYPE(tpa_start) == 3)
 			tpa_info->gso_type = SKB_GSO_TCPV6;
 		tpa_info->rss_hash =
 			le32_to_cpu(tpa_start->rx_tpa_start_cmp_rss_hash);
 	} else {
 		tpa_info->hash_type = PKT_HASH_TYPE_NONE;
 		tpa_info->gso_type = 0;
-		if (netif_msg_rx_err(bp))
-			netdev_warn(bp->dev, "TPA packet without valid hash\n");
+		netif_warn(bp, rx_err, bp->dev, "TPA packet without valid hash\n");
 	}
 	tpa_info->flags2 = le32_to_cpu(tpa_start1->rx_tpa_start_cmp_flags2);
-	tpa_info->metadata = le32_to_cpu(tpa_start1->rx_tpa_start_cmp_metadata);
 	tpa_info->hdr_info = le32_to_cpu(tpa_start1->rx_tpa_start_cmp_hdr_info);
+	if (cmp_type == CMP_TYPE_RX_L2_TPA_START_CMP)
+		bnxt_tpa_metadata(tpa_info, tpa_start, tpa_start1);
+	else
+		bnxt_tpa_metadata_v2(tpa_info, tpa_start, tpa_start1);
+	tpa_info->agg_count = 0;
 
 	rxr->rx_prod = NEXT_RX(prod);
-	cons = NEXT_RX(cons);
-	rxr->rx_next_cons = NEXT_RX(cons);
+	cons = RING_RX(bp, NEXT_RX(cons));
+	rxr->rx_next_cons = RING_RX(bp, NEXT_RX(cons));
 	cons_rx_buf = &rxr->rx_buf_ring[cons];
 
 	bnxt_reuse_rx_data(rxr, cons, cons_rx_buf->data);
@@ -1081,13 +1617,37 @@ static void bnxt_tpa_start(struct bnxt *bp, struct bnxt_rx_ring_info *rxr,
 	cons_rx_buf->data = NULL;
 }
 
-static void bnxt_abort_tpa(struct bnxt *bp, struct bnxt_napi *bnapi,
-			   u16 cp_cons, u32 agg_bufs)
+static void bnxt_abort_tpa(struct bnxt_cp_ring_info *cpr, u16 idx, u32 agg_bufs)
 {
 	if (agg_bufs)
-		bnxt_reuse_rx_agg_bufs(bnapi, cp_cons, agg_bufs);
+		bnxt_reuse_rx_agg_bufs(cpr, idx, 0, agg_bufs, true);
 }
 
+#ifdef CONFIG_INET
+static void bnxt_gro_tunnel(struct sk_buff *skb, __be16 ip_proto)
+{
+	struct udphdr *uh = NULL;
+
+	if (ip_proto == htons(ETH_P_IP)) {
+		struct iphdr *iph = (struct iphdr *)skb->data;
+
+		if (iph->protocol == IPPROTO_UDP)
+			uh = (struct udphdr *)(iph + 1);
+	} else {
+		struct ipv6hdr *iph = (struct ipv6hdr *)skb->data;
+
+		if (iph->nexthdr == IPPROTO_UDP)
+			uh = (struct udphdr *)(iph + 1);
+	}
+	if (uh) {
+		if (uh->check)
+			skb_shinfo(skb)->gso_type |= SKB_GSO_UDP_TUNNEL_CSUM;
+		else
+			skb_shinfo(skb)->gso_type |= SKB_GSO_UDP_TUNNEL;
+	}
+}
+#endif
+
 static struct sk_buff *bnxt_gro_func_5731x(struct bnxt_tpa_info *tpa_info,
 					   int payload_off, int tcp_ts,
 					   struct sk_buff *skb)
@@ -1145,28 +1705,39 @@ static struct sk_buff *bnxt_gro_func_5731x(struct bnxt_tpa_info *tpa_info,
 	}
 
 	if (inner_mac_off) { /* tunnel */
-		struct udphdr *uh = NULL;
 		__be16 proto = *((__be16 *)(skb->data + outer_ip_off -
 					    ETH_HLEN - 2));
 
-		if (proto == htons(ETH_P_IP)) {
-			struct iphdr *iph = (struct iphdr *)skb->data;
+		bnxt_gro_tunnel(skb, proto);
+	}
+#endif
+	return skb;
+}
 
-			if (iph->protocol == IPPROTO_UDP)
-				uh = (struct udphdr *)(iph + 1);
-		} else {
-			struct ipv6hdr *iph = (struct ipv6hdr *)skb->data;
+static struct sk_buff *bnxt_gro_func_5750x(struct bnxt_tpa_info *tpa_info,
+					   int payload_off, int tcp_ts,
+					   struct sk_buff *skb)
+{
+#ifdef CONFIG_INET
+	u16 outer_ip_off, inner_ip_off, inner_mac_off;
+	u32 hdr_info = tpa_info->hdr_info;
+	int iphdr_len, nw_off;
 
-			if (iph->nexthdr == IPPROTO_UDP)
-				uh = (struct udphdr *)(iph + 1);
-		}
-		if (uh) {
-			if (uh->check)
-				skb_shinfo(skb)->gso_type |=
-					SKB_GSO_UDP_TUNNEL_CSUM;
-			else
-				skb_shinfo(skb)->gso_type |= SKB_GSO_UDP_TUNNEL;
-		}
+	inner_ip_off = BNXT_TPA_INNER_L3_OFF(hdr_info);
+	inner_mac_off = BNXT_TPA_INNER_L2_OFF(hdr_info);
+	outer_ip_off = BNXT_TPA_OUTER_L3_OFF(hdr_info);
+
+	nw_off = inner_ip_off - ETH_HLEN;
+	skb_set_network_header(skb, nw_off);
+	iphdr_len = (tpa_info->flags2 & RX_TPA_START_CMP_FLAGS2_IP_TYPE) ?
+		     sizeof(struct ipv6hdr) : sizeof(struct iphdr);
+	skb_set_transport_header(skb, nw_off + iphdr_len);
+
+	if (inner_mac_off) { /* tunnel */
+		__be16 proto = *((__be16 *)(skb->data + outer_ip_off -
+					    ETH_HLEN - 2));
+
+		bnxt_gro_tunnel(skb, proto);
 	}
 #endif
 	return skb;
@@ -1213,28 +1784,8 @@ static struct sk_buff *bnxt_gro_func_5730x(struct bnxt_tpa_info *tpa_info,
 		return NULL;
 	}
 
-	if (nw_off) { /* tunnel */
-		struct udphdr *uh = NULL;
-
-		if (skb->protocol == htons(ETH_P_IP)) {
-			struct iphdr *iph = (struct iphdr *)skb->data;
-
-			if (iph->protocol == IPPROTO_UDP)
-				uh = (struct udphdr *)(iph + 1);
-		} else {
-			struct ipv6hdr *iph = (struct ipv6hdr *)skb->data;
-
-			if (iph->nexthdr == IPPROTO_UDP)
-				uh = (struct udphdr *)(iph + 1);
-		}
-		if (uh) {
-			if (uh->check)
-				skb_shinfo(skb)->gso_type |=
-					SKB_GSO_UDP_TUNNEL_CSUM;
-			else
-				skb_shinfo(skb)->gso_type |= SKB_GSO_UDP_TUNNEL;
-		}
-	}
+	if (nw_off) /* tunnel */
+		bnxt_gro_tunnel(skb, skb->protocol);
 #endif
 	return skb;
 }
@@ -1257,9 +1808,10 @@ static inline struct sk_buff *bnxt_gro_skb(struct bnxt *bp,
 	skb_shinfo(skb)->gso_size =
 		le32_to_cpu(tpa_end1->rx_tpa_end_cmp_seg_len);
 	skb_shinfo(skb)->gso_type = tpa_info->gso_type;
-	payload_off = (le32_to_cpu(tpa_end->rx_tpa_end_cmp_misc_v1) &
-		       RX_TPA_END_CMP_PAYLOAD_OFFSET) >>
-		      RX_TPA_END_CMP_PAYLOAD_OFFSET_SHIFT;
+	if (bp->flags & BNXT_FLAG_CHIP_P5_PLUS)
+		payload_off = TPA_END_PAYLOAD_OFF_P5(tpa_end1);
+	else
+		payload_off = TPA_END_PAYLOAD_OFF(tpa_end);
 	skb = bp->gro_func(tpa_info, payload_off, TPA_END_GRO_TS(tpa_end), skb);
 	if (likely(skb))
 		tcp_gro_complete(skb);
@@ -1267,70 +1819,103 @@ static inline struct sk_buff *bnxt_gro_skb(struct bnxt *bp,
 	return skb;
 }
 
+/* Given the cfa_code of a received packet determine which
+ * netdev (vf-rep or PF) the packet is destined to.
+ */
+static struct net_device *bnxt_get_pkt_dev(struct bnxt *bp, u16 cfa_code)
+{
+	struct net_device *dev = bnxt_get_vf_rep(bp, cfa_code);
+
+	/* if vf-rep dev is NULL, it must belong to the PF */
+	return dev ? dev : bp->dev;
+}
+
 static inline struct sk_buff *bnxt_tpa_end(struct bnxt *bp,
-					   struct bnxt_napi *bnapi,
+					   struct bnxt_cp_ring_info *cpr,
 					   u32 *raw_cons,
 					   struct rx_tpa_end_cmp *tpa_end,
 					   struct rx_tpa_end_cmp_ext *tpa_end1,
 					   u8 *event)
 {
-	struct bnxt_cp_ring_info *cpr = &bnapi->cp_ring;
+	struct bnxt_napi *bnapi = cpr->bnapi;
 	struct bnxt_rx_ring_info *rxr = bnapi->rx_ring;
-	u8 agg_id = TPA_END_AGG_ID(tpa_end);
+	struct net_device *dev = bp->dev;
 	u8 *data_ptr, agg_bufs;
-	u16 cp_cons = RING_CMP(*raw_cons);
 	unsigned int len;
 	struct bnxt_tpa_info *tpa_info;
 	dma_addr_t mapping;
 	struct sk_buff *skb;
+	u16 idx = 0, agg_id;
 	void *data;
+	bool gro;
 
 	if (unlikely(bnapi->in_reset)) {
-		int rc = bnxt_discard_rx(bp, bnapi, raw_cons, tpa_end);
+		int rc = bnxt_discard_rx(bp, cpr, raw_cons, tpa_end);
 
 		if (rc < 0)
 			return ERR_PTR(-EBUSY);
 		return NULL;
 	}
 
-	tpa_info = &rxr->rx_tpa[agg_id];
+	if (bp->flags & BNXT_FLAG_CHIP_P5_PLUS) {
+		agg_id = TPA_END_AGG_ID_P5(tpa_end);
+		agg_id = bnxt_lookup_agg_idx(rxr, agg_id);
+		agg_bufs = TPA_END_AGG_BUFS_P5(tpa_end1);
+		tpa_info = &rxr->rx_tpa[agg_id];
+		if (unlikely(agg_bufs != tpa_info->agg_count)) {
+			netdev_warn(bp->dev, "TPA end agg_buf %d != expected agg_bufs %d\n",
+				    agg_bufs, tpa_info->agg_count);
+			agg_bufs = tpa_info->agg_count;
+		}
+		tpa_info->agg_count = 0;
+		*event |= BNXT_AGG_EVENT;
+		bnxt_free_agg_idx(rxr, agg_id);
+		idx = agg_id;
+		gro = !!(bp->flags & BNXT_FLAG_GRO);
+	} else {
+		agg_id = TPA_END_AGG_ID(tpa_end);
+		agg_bufs = TPA_END_AGG_BUFS(tpa_end);
+		tpa_info = &rxr->rx_tpa[agg_id];
+		idx = RING_CMP(*raw_cons);
+		if (agg_bufs) {
+			if (!bnxt_agg_bufs_valid(bp, cpr, agg_bufs, raw_cons))
+				return ERR_PTR(-EBUSY);
+
+			*event |= BNXT_AGG_EVENT;
+			idx = NEXT_CMP(idx);
+		}
+		gro = !!TPA_END_GRO(tpa_end);
+	}
 	data = tpa_info->data;
 	data_ptr = tpa_info->data_ptr;
 	prefetch(data_ptr);
 	len = tpa_info->len;
 	mapping = tpa_info->mapping;
 
-	agg_bufs = (le32_to_cpu(tpa_end->rx_tpa_end_cmp_misc_v1) &
-		    RX_TPA_END_CMP_AGG_BUFS) >> RX_TPA_END_CMP_AGG_BUFS_SHIFT;
-
-	if (agg_bufs) {
-		if (!bnxt_agg_bufs_valid(bp, cpr, agg_bufs, raw_cons))
-			return ERR_PTR(-EBUSY);
-
-		*event |= BNXT_AGG_EVENT;
-		cp_cons = NEXT_CMP(cp_cons);
-	}
-
-	if (unlikely(agg_bufs > MAX_SKB_FRAGS)) {
-		bnxt_abort_tpa(bp, bnapi, cp_cons, agg_bufs);
-		netdev_warn(bp->dev, "TPA frags %d exceeded MAX_SKB_FRAGS %d\n",
-			    agg_bufs, (int)MAX_SKB_FRAGS);
+	if (unlikely(agg_bufs > MAX_SKB_FRAGS || TPA_END_ERRORS(tpa_end1))) {
+		bnxt_abort_tpa(cpr, idx, agg_bufs);
+		if (agg_bufs > MAX_SKB_FRAGS)
+			netdev_warn(bp->dev, "TPA frags %d exceeded MAX_SKB_FRAGS %d\n",
+				    agg_bufs, (int)MAX_SKB_FRAGS);
 		return NULL;
 	}
 
-	if (len <= bp->rx_copy_thresh) {
+	if (len <= bp->rx_copybreak) {
 		skb = bnxt_copy_skb(bnapi, data_ptr, len, mapping);
 		if (!skb) {
-			bnxt_abort_tpa(bp, bnapi, cp_cons, agg_bufs);
+			bnxt_abort_tpa(cpr, idx, agg_bufs);
+			cpr->sw_stats->rx.rx_oom_discards += 1;
 			return NULL;
 		}
 	} else {
 		u8 *new_data;
 		dma_addr_t new_mapping;
 
-		new_data = __bnxt_alloc_rx_data(bp, &new_mapping, GFP_ATOMIC);
+		new_data = __bnxt_alloc_rx_frag(bp, &new_mapping, rxr,
+						GFP_ATOMIC);
 		if (!new_data) {
-			bnxt_abort_tpa(bp, bnapi, cp_cons, agg_bufs);
+			bnxt_abort_tpa(cpr, idx, agg_bufs);
+			cpr->sw_stats->rx.rx_oom_discards += 1;
 			return NULL;
 		}
 
@@ -1338,39 +1923,50 @@ static inline struct sk_buff *bnxt_tpa_end(struct bnxt *bp,
 		tpa_info->data_ptr = new_data + bp->rx_offset;
 		tpa_info->mapping = new_mapping;
 
-		skb = build_skb(data, 0);
-		dma_unmap_single_attrs(&bp->pdev->dev, mapping,
-				       bp->rx_buf_use_size, bp->rx_dir,
-				       DMA_ATTR_WEAK_ORDERING);
+		skb = napi_build_skb(data, bp->rx_buf_size);
+		dma_sync_single_for_cpu(&bp->pdev->dev, mapping,
+					bp->rx_buf_use_size, bp->rx_dir);
 
 		if (!skb) {
-			kfree(data);
-			bnxt_abort_tpa(bp, bnapi, cp_cons, agg_bufs);
+			page_pool_free_va(rxr->head_pool, data, true);
+			bnxt_abort_tpa(cpr, idx, agg_bufs);
+			cpr->sw_stats->rx.rx_oom_discards += 1;
 			return NULL;
 		}
+		skb_mark_for_recycle(skb);
 		skb_reserve(skb, bp->rx_offset);
 		skb_put(skb, len);
 	}
 
 	if (agg_bufs) {
-		skb = bnxt_rx_pages(bp, bnapi, skb, cp_cons, agg_bufs);
+		skb = bnxt_rx_agg_netmems_skb(bp, cpr, skb, idx, agg_bufs,
+					      true);
 		if (!skb) {
 			/* Page reuse already handled by bnxt_rx_pages(). */
+			cpr->sw_stats->rx.rx_oom_discards += 1;
 			return NULL;
 		}
 	}
-	skb->protocol = eth_type_trans(skb, bp->dev);
+
+	if (tpa_info->cfa_code_valid)
+		dev = bnxt_get_pkt_dev(bp, tpa_info->cfa_code);
+	skb->protocol = eth_type_trans(skb, dev);
 
 	if (tpa_info->hash_type != PKT_HASH_TYPE_NONE)
 		skb_set_hash(skb, tpa_info->rss_hash, tpa_info->hash_type);
 
-	if ((tpa_info->flags2 & RX_CMP_FLAGS2_META_FORMAT_VLAN) &&
-	    (skb->dev->features & NETIF_F_HW_VLAN_CTAG_RX)) {
-		u16 vlan_proto = tpa_info->metadata >>
-			RX_CMP_FLAGS2_METADATA_TPID_SFT;
-		u16 vtag = tpa_info->metadata & RX_CMP_FLAGS2_METADATA_VID_MASK;
+	if (tpa_info->vlan_valid &&
+	    (dev->features & BNXT_HW_FEATURE_VLAN_ALL_RX)) {
+		__be16 vlan_proto = htons(tpa_info->metadata >>
+					  RX_CMP_FLAGS2_METADATA_TPID_SFT);
+		u16 vtag = tpa_info->metadata & RX_CMP_FLAGS2_METADATA_TCI_MASK;
 
-		__vlan_hwaccel_put_tag(skb, htons(vlan_proto), vtag);
+		if (eth_type_vlan(vlan_proto)) {
+			__vlan_hwaccel_put_tag(skb, vlan_proto, vtag);
+		} else {
+			dev_kfree_skb(skb);
+			return NULL;
+		}
 	}
 
 	skb_checksum_none_assert(skb);
@@ -1380,12 +1976,112 @@ static inline struct sk_buff *bnxt_tpa_end(struct bnxt *bp,
 			(tpa_info->flags2 & RX_CMP_FLAGS2_T_L4_CS_CALC) >> 3;
 	}
 
-	if (TPA_END_GRO(tpa_end))
+	if (gro)
 		skb = bnxt_gro_skb(bp, tpa_info, tpa_end, tpa_end1, skb);
 
 	return skb;
 }
 
+static void bnxt_tpa_agg(struct bnxt *bp, struct bnxt_rx_ring_info *rxr,
+			 struct rx_agg_cmp *rx_agg)
+{
+	u16 agg_id = TPA_AGG_AGG_ID(rx_agg);
+	struct bnxt_tpa_info *tpa_info;
+
+	agg_id = bnxt_lookup_agg_idx(rxr, agg_id);
+	tpa_info = &rxr->rx_tpa[agg_id];
+	BUG_ON(tpa_info->agg_count >= MAX_SKB_FRAGS);
+	tpa_info->agg_arr[tpa_info->agg_count++] = *rx_agg;
+}
+
+static void bnxt_deliver_skb(struct bnxt *bp, struct bnxt_napi *bnapi,
+			     struct sk_buff *skb)
+{
+	skb_mark_for_recycle(skb);
+
+	if (skb->dev != bp->dev) {
+		/* this packet belongs to a vf-rep */
+		bnxt_vf_rep_rx(bp, skb);
+		return;
+	}
+	skb_record_rx_queue(skb, bnapi->index);
+	napi_gro_receive(&bnapi->napi, skb);
+}
+
+static bool bnxt_rx_ts_valid(struct bnxt *bp, u32 flags,
+			     struct rx_cmp_ext *rxcmp1, u32 *cmpl_ts)
+{
+	u32 ts = le32_to_cpu(rxcmp1->rx_cmp_timestamp);
+
+	if (BNXT_PTP_RX_TS_VALID(flags))
+		goto ts_valid;
+	if (!bp->ptp_all_rx_tstamp || !ts || !BNXT_ALL_RX_TS_VALID(flags))
+		return false;
+
+ts_valid:
+	*cmpl_ts = ts;
+	return true;
+}
+
+static struct sk_buff *bnxt_rx_vlan(struct sk_buff *skb, u8 cmp_type,
+				    struct rx_cmp *rxcmp,
+				    struct rx_cmp_ext *rxcmp1)
+{
+	__be16 vlan_proto;
+	u16 vtag;
+
+	if (cmp_type == CMP_TYPE_RX_L2_CMP) {
+		__le32 flags2 = rxcmp1->rx_cmp_flags2;
+		u32 meta_data;
+
+		if (!(flags2 & cpu_to_le32(RX_CMP_FLAGS2_META_FORMAT_VLAN)))
+			return skb;
+
+		meta_data = le32_to_cpu(rxcmp1->rx_cmp_meta_data);
+		vtag = meta_data & RX_CMP_FLAGS2_METADATA_TCI_MASK;
+		vlan_proto = htons(meta_data >> RX_CMP_FLAGS2_METADATA_TPID_SFT);
+		if (eth_type_vlan(vlan_proto))
+			__vlan_hwaccel_put_tag(skb, vlan_proto, vtag);
+		else
+			goto vlan_err;
+	} else if (cmp_type == CMP_TYPE_RX_L2_V3_CMP) {
+		if (RX_CMP_VLAN_VALID(rxcmp)) {
+			u32 tpid_sel = RX_CMP_VLAN_TPID_SEL(rxcmp);
+
+			if (tpid_sel == RX_CMP_METADATA1_TPID_8021Q)
+				vlan_proto = htons(ETH_P_8021Q);
+			else if (tpid_sel == RX_CMP_METADATA1_TPID_8021AD)
+				vlan_proto = htons(ETH_P_8021AD);
+			else
+				goto vlan_err;
+			vtag = RX_CMP_METADATA0_TCI(rxcmp1);
+			__vlan_hwaccel_put_tag(skb, vlan_proto, vtag);
+		}
+	}
+	return skb;
+vlan_err:
+	skb_mark_for_recycle(skb);
+	dev_kfree_skb(skb);
+	return NULL;
+}
+
+static enum pkt_hash_types bnxt_rss_ext_op(struct bnxt *bp,
+					   struct rx_cmp *rxcmp)
+{
+	u8 ext_op;
+
+	ext_op = RX_CMP_V3_HASH_TYPE(bp, rxcmp);
+	switch (ext_op) {
+	case EXT_OP_INNER_4:
+	case EXT_OP_OUTER_4:
+	case EXT_OP_INNFL_3:
+	case EXT_OP_OUTFL_3:
+		return PKT_HASH_TYPE_L4;
+	default:
+		return PKT_HASH_TYPE_L3;
+	}
+}
+
 /* returns the following:
  * 1       - 1 packet successfully received
  * 0       - successful TPA_START, packet not completed yet
@@ -1393,28 +2089,39 @@ static inline struct sk_buff *bnxt_tpa_end(struct bnxt *bp,
  * -ENOMEM - packet aborted due to out of memory
  * -EIO    - packet aborted due to hw error indicated in BD
  */
-static int bnxt_rx_pkt(struct bnxt *bp, struct bnxt_napi *bnapi, u32 *raw_cons,
-		       u8 *event)
+static int bnxt_rx_pkt(struct bnxt *bp, struct bnxt_cp_ring_info *cpr,
+		       u32 *raw_cons, u8 *event)
 {
-	struct bnxt_cp_ring_info *cpr = &bnapi->cp_ring;
+	struct bnxt_napi *bnapi = cpr->bnapi;
 	struct bnxt_rx_ring_info *rxr = bnapi->rx_ring;
 	struct net_device *dev = bp->dev;
 	struct rx_cmp *rxcmp;
 	struct rx_cmp_ext *rxcmp1;
 	u32 tmp_raw_cons = *raw_cons;
 	u16 cons, prod, cp_cons = RING_CMP(tmp_raw_cons);
+	struct skb_shared_info *sinfo;
 	struct bnxt_sw_rx_bd *rx_buf;
 	unsigned int len;
 	u8 *data_ptr, agg_bufs, cmp_type;
+	bool xdp_active = false;
 	dma_addr_t dma_addr;
 	struct sk_buff *skb;
+	struct xdp_buff xdp;
+	u32 flags, misc;
+	u32 cmpl_ts;
 	void *data;
 	int rc = 0;
-	u32 misc;
 
 	rxcmp = (struct rx_cmp *)
 			&cpr->cp_desc_ring[CP_RING(cp_cons)][CP_IDX(cp_cons)];
 
+	cmp_type = RX_CMP_TYPE(rxcmp);
+
+	if (cmp_type == CMP_TYPE_RX_TPA_AGG_CMP) {
+		bnxt_tpa_agg(bp, rxr, (struct rx_agg_cmp *)rxcmp);
+		goto next_rx_no_prod_no_len;
+	}
+
 	tmp_raw_cons = NEXT_RAW_CMP(tmp_raw_cons);
 	cp_cons = RING_CMP(tmp_raw_cons);
 	rxcmp1 = (struct rx_cmp_ext *)
@@ -1423,45 +2130,54 @@ static int bnxt_rx_pkt(struct bnxt *bp, struct bnxt_napi *bnapi, u32 *raw_cons,
 	if (!RX_CMP_VALID(rxcmp1, tmp_raw_cons))
 		return -EBUSY;
 
-	cmp_type = RX_CMP_TYPE(rxcmp);
-
+	/* The valid test of the entry must be done first before
+	 * reading any further.
+	 */
+	dma_rmb();
 	prod = rxr->rx_prod;
 
-	if (cmp_type == CMP_TYPE_RX_L2_TPA_START_CMP) {
-		bnxt_tpa_start(bp, rxr, (struct rx_tpa_start_cmp *)rxcmp,
+	if (cmp_type == CMP_TYPE_RX_L2_TPA_START_CMP ||
+	    cmp_type == CMP_TYPE_RX_L2_TPA_START_V3_CMP) {
+		bnxt_tpa_start(bp, rxr, cmp_type,
+			       (struct rx_tpa_start_cmp *)rxcmp,
 			       (struct rx_tpa_start_cmp_ext *)rxcmp1);
 
 		*event |= BNXT_RX_EVENT;
-		goto next_rx_no_prod;
+		goto next_rx_no_prod_no_len;
 
 	} else if (cmp_type == CMP_TYPE_RX_L2_TPA_END_CMP) {
-		skb = bnxt_tpa_end(bp, bnapi, &tmp_raw_cons,
+		skb = bnxt_tpa_end(bp, cpr, &tmp_raw_cons,
 				   (struct rx_tpa_end_cmp *)rxcmp,
 				   (struct rx_tpa_end_cmp_ext *)rxcmp1, event);
 
-		if (unlikely(IS_ERR(skb)))
+		if (IS_ERR(skb))
 			return -EBUSY;
 
 		rc = -ENOMEM;
 		if (likely(skb)) {
-			skb_record_rx_queue(skb, bnapi->index);
-			napi_gro_receive(&bnapi->napi, skb);
+			bnxt_deliver_skb(bp, bnapi, skb);
 			rc = 1;
 		}
 		*event |= BNXT_RX_EVENT;
-		goto next_rx_no_prod;
+		goto next_rx_no_prod_no_len;
 	}
 
 	cons = rxcmp->rx_cmp_opaque;
-	rx_buf = &rxr->rx_buf_ring[cons];
-	data = rx_buf->data;
-	data_ptr = rx_buf->data_ptr;
 	if (unlikely(cons != rxr->rx_next_cons)) {
-		int rc1 = bnxt_discard_rx(bp, bnapi, raw_cons, rxcmp);
+		int rc1 = bnxt_discard_rx(bp, cpr, &tmp_raw_cons, rxcmp);
 
-		bnxt_sched_reset(bp, rxr);
-		return rc1;
+		/* 0xffff is forced error, don't print it */
+		if (rxr->rx_next_cons != 0xffff)
+			netdev_warn(bp->dev, "RX cons %x != expected cons %x\n",
+				    cons, rxr->rx_next_cons);
+		bnxt_sched_reset_rxr(bp, rxr);
+		if (rc1)
+			return rc1;
+		goto next_rx_no_prod_no_len;
 	}
+	rx_buf = &rxr->rx_buf_ring[cons];
+	data = rx_buf->data;
+	data_ptr = rx_buf->data_ptr;
 	prefetch(data_ptr);
 
 	misc = le32_to_cpu(rxcmp->rx_cmp_misc_v1);
@@ -1478,28 +2194,72 @@ static int bnxt_rx_pkt(struct bnxt *bp, struct bnxt_napi *bnapi, u32 *raw_cons,
 
 	rx_buf->data = NULL;
 	if (rxcmp1->rx_cmp_cfa_code_errors_v2 & RX_CMP_L2_ERRORS) {
+		u32 rx_err = le32_to_cpu(rxcmp1->rx_cmp_cfa_code_errors_v2);
+
 		bnxt_reuse_rx_data(rxr, cons, data);
 		if (agg_bufs)
-			bnxt_reuse_rx_agg_bufs(bnapi, cp_cons, agg_bufs);
+			bnxt_reuse_rx_agg_bufs(cpr, cp_cons, 0, agg_bufs,
+					       false);
 
 		rc = -EIO;
-		goto next_rx;
+		if (rx_err & RX_CMPL_ERRORS_BUFFER_ERROR_MASK) {
+			bnapi->cp_ring.sw_stats->rx.rx_buf_errors++;
+			if (!(bp->flags & BNXT_FLAG_CHIP_P5_PLUS) &&
+			    !(bp->fw_cap & BNXT_FW_CAP_RING_MONITOR)) {
+				netdev_warn_once(bp->dev, "RX buffer error %x\n",
+						 rx_err);
+				bnxt_sched_reset_rxr(bp, rxr);
+			}
+		}
+		goto next_rx_no_len;
 	}
 
-	len = le32_to_cpu(rxcmp->rx_cmp_len_flags_type) >> RX_CMP_LEN_SHIFT;
+	flags = le32_to_cpu(rxcmp->rx_cmp_len_flags_type);
+	len = flags >> RX_CMP_LEN_SHIFT;
 	dma_addr = rx_buf->mapping;
 
-	if (bnxt_rx_xdp(bp, rxr, cons, data, &data_ptr, &len, event)) {
-		rc = 1;
-		goto next_rx;
+	if (bnxt_xdp_attached(bp, rxr)) {
+		bnxt_xdp_buff_init(bp, rxr, cons, data_ptr, len, &xdp);
+		if (agg_bufs) {
+			u32 frag_len = bnxt_rx_agg_netmems_xdp(bp, cpr, &xdp,
+							       cp_cons,
+							       agg_bufs,
+							       false);
+			if (!frag_len)
+				goto oom_next_rx;
+
+		}
+		xdp_active = true;
 	}
 
-	if (len <= bp->rx_copy_thresh) {
-		skb = bnxt_copy_skb(bnapi, data_ptr, len, dma_addr);
+	if (xdp_active) {
+		if (bnxt_rx_xdp(bp, rxr, cons, &xdp, data, &data_ptr, &len, event)) {
+			rc = 1;
+			goto next_rx;
+		}
+		if (xdp_buff_has_frags(&xdp)) {
+			sinfo = xdp_get_shared_info_from_buff(&xdp);
+			agg_bufs = sinfo->nr_frags;
+		} else {
+			agg_bufs = 0;
+		}
+	}
+
+	if (len <= bp->rx_copybreak) {
+		if (!xdp_active)
+			skb = bnxt_copy_skb(bnapi, data_ptr, len, dma_addr);
+		else
+			skb = bnxt_copy_xdp(bnapi, &xdp, len, dma_addr);
 		bnxt_reuse_rx_data(rxr, cons, data);
 		if (!skb) {
-			rc = -ENOMEM;
-			goto next_rx;
+			if (agg_bufs) {
+				if (!xdp_active)
+					bnxt_reuse_rx_agg_bufs(cpr, cp_cons, 0,
+							       agg_bufs, false);
+				else
+					bnxt_xdp_buff_frags_free(rxr, &xdp);
+			}
+			goto oom_next_rx;
 		}
 	} else {
 		u32 payload;
@@ -1510,40 +2270,52 @@ static int bnxt_rx_pkt(struct bnxt *bp, struct bnxt_napi *bnapi, u32 *raw_cons,
 			payload = 0;
 		skb = bp->rx_skb_func(bp, rxr, cons, data, data_ptr, dma_addr,
 				      payload | len);
-		if (!skb) {
-			rc = -ENOMEM;
-			goto next_rx;
-		}
+		if (!skb)
+			goto oom_next_rx;
 	}
 
 	if (agg_bufs) {
-		skb = bnxt_rx_pages(bp, bnapi, skb, cp_cons, agg_bufs);
-		if (!skb) {
-			rc = -ENOMEM;
-			goto next_rx;
+		if (!xdp_active) {
+			skb = bnxt_rx_agg_netmems_skb(bp, cpr, skb, cp_cons,
+						      agg_bufs, false);
+			if (!skb)
+				goto oom_next_rx;
+		} else {
+			skb = bnxt_xdp_build_skb(bp, skb, agg_bufs,
+						 rxr->page_pool, &xdp);
+			if (!skb) {
+				/* we should be able to free the old skb here */
+				bnxt_xdp_buff_frags_free(rxr, &xdp);
+				goto oom_next_rx;
+			}
 		}
 	}
 
 	if (RX_CMP_HASH_VALID(rxcmp)) {
-		u32 hash_type = RX_CMP_HASH_TYPE(rxcmp);
-		enum pkt_hash_types type = PKT_HASH_TYPE_L4;
+		enum pkt_hash_types type;
 
-		/* RSS profiles 1 and 3 with extract code 0 for inner 4-tuple */
-		if (hash_type != 1 && hash_type != 3)
-			type = PKT_HASH_TYPE_L3;
+		if (cmp_type == CMP_TYPE_RX_L2_V3_CMP) {
+			type = bnxt_rss_ext_op(bp, rxcmp);
+		} else {
+			u32 itypes = RX_CMP_ITYPES(rxcmp);
+
+			if (itypes == RX_CMP_FLAGS_ITYPE_TCP ||
+			    itypes == RX_CMP_FLAGS_ITYPE_UDP)
+				type = PKT_HASH_TYPE_L4;
+			else
+				type = PKT_HASH_TYPE_L3;
+		}
 		skb_set_hash(skb, le32_to_cpu(rxcmp->rx_cmp_rss_hash), type);
 	}
 
+	if (cmp_type == CMP_TYPE_RX_L2_CMP)
+		dev = bnxt_get_pkt_dev(bp, RX_CMP_CFA_CODE(rxcmp1));
 	skb->protocol = eth_type_trans(skb, dev);
 
-	if ((rxcmp1->rx_cmp_flags2 &
-	     cpu_to_le32(RX_CMP_FLAGS2_META_FORMAT_VLAN)) &&
-	    (skb->dev->features & NETIF_F_HW_VLAN_CTAG_RX)) {
-		u32 meta_data = le32_to_cpu(rxcmp1->rx_cmp_meta_data);
-		u16 vtag = meta_data & RX_CMP_FLAGS2_METADATA_VID_MASK;
-		u16 vlan_proto = meta_data >> RX_CMP_FLAGS2_METADATA_TPID_SFT;
-
-		__vlan_hwaccel_put_tag(skb, htons(vlan_proto), vtag);
+	if (skb->dev->features & BNXT_HW_FEATURE_VLAN_ALL_RX) {
+		skb = bnxt_rx_vlan(skb, cmp_type, rxcmp, rxcmp1);
+		if (!skb)
+			goto next_rx;
 	}
 
 	skb_checksum_none_assert(skb);
@@ -1555,51 +2327,420 @@ static int bnxt_rx_pkt(struct bnxt *bp, struct bnxt_napi *bnapi, u32 *raw_cons,
 	} else {
 		if (rxcmp1->rx_cmp_cfa_code_errors_v2 & RX_CMP_L4_CS_ERR_BITS) {
 			if (dev->features & NETIF_F_RXCSUM)
-				cpr->rx_l4_csum_errors++;
+				bnapi->cp_ring.sw_stats->rx.rx_l4_csum_errors++;
 		}
 	}
 
-	skb_record_rx_queue(skb, bnapi->index);
-	napi_gro_receive(&bnapi->napi, skb);
+	if (bnxt_rx_ts_valid(bp, flags, rxcmp1, &cmpl_ts)) {
+		if (bp->flags & BNXT_FLAG_CHIP_P5_PLUS) {
+			u64 ns, ts;
+
+			if (!bnxt_get_rx_ts_p5(bp, &ts, cmpl_ts)) {
+				struct bnxt_ptp_cfg *ptp = bp->ptp_cfg;
+
+				ns = bnxt_timecounter_cyc2time(ptp, ts);
+				memset(skb_hwtstamps(skb), 0,
+				       sizeof(*skb_hwtstamps(skb)));
+				skb_hwtstamps(skb)->hwtstamp = ns_to_ktime(ns);
+			}
+		}
+	}
+	bnxt_deliver_skb(bp, bnapi, skb);
 	rc = 1;
 
 next_rx:
+	cpr->rx_packets += 1;
+	cpr->rx_bytes += len;
+
+next_rx_no_len:
 	rxr->rx_prod = NEXT_RX(prod);
-	rxr->rx_next_cons = NEXT_RX(cons);
+	rxr->rx_next_cons = RING_RX(bp, NEXT_RX(cons));
 
-next_rx_no_prod:
+next_rx_no_prod_no_len:
 	*raw_cons = tmp_raw_cons;
 
 	return rc;
+
+oom_next_rx:
+	cpr->sw_stats->rx.rx_oom_discards += 1;
+	rc = -ENOMEM;
+	goto next_rx;
+}
+
+/* In netpoll mode, if we are using a combined completion ring, we need to
+ * discard the rx packets and recycle the buffers.
+ */
+static int bnxt_force_rx_discard(struct bnxt *bp,
+				 struct bnxt_cp_ring_info *cpr,
+				 u32 *raw_cons, u8 *event)
+{
+	u32 tmp_raw_cons = *raw_cons;
+	struct rx_cmp_ext *rxcmp1;
+	struct rx_cmp *rxcmp;
+	u16 cp_cons;
+	u8 cmp_type;
+	int rc;
+
+	cp_cons = RING_CMP(tmp_raw_cons);
+	rxcmp = (struct rx_cmp *)
+			&cpr->cp_desc_ring[CP_RING(cp_cons)][CP_IDX(cp_cons)];
+
+	tmp_raw_cons = NEXT_RAW_CMP(tmp_raw_cons);
+	cp_cons = RING_CMP(tmp_raw_cons);
+	rxcmp1 = (struct rx_cmp_ext *)
+			&cpr->cp_desc_ring[CP_RING(cp_cons)][CP_IDX(cp_cons)];
+
+	if (!RX_CMP_VALID(rxcmp1, tmp_raw_cons))
+		return -EBUSY;
+
+	/* The valid test of the entry must be done first before
+	 * reading any further.
+	 */
+	dma_rmb();
+	cmp_type = RX_CMP_TYPE(rxcmp);
+	if (cmp_type == CMP_TYPE_RX_L2_CMP ||
+	    cmp_type == CMP_TYPE_RX_L2_V3_CMP) {
+		rxcmp1->rx_cmp_cfa_code_errors_v2 |=
+			cpu_to_le32(RX_CMPL_ERRORS_CRC_ERROR);
+	} else if (cmp_type == CMP_TYPE_RX_L2_TPA_END_CMP) {
+		struct rx_tpa_end_cmp_ext *tpa_end1;
+
+		tpa_end1 = (struct rx_tpa_end_cmp_ext *)rxcmp1;
+		tpa_end1->rx_tpa_end_cmp_errors_v2 |=
+			cpu_to_le32(RX_TPA_END_CMP_ERRORS);
+	}
+	rc = bnxt_rx_pkt(bp, cpr, raw_cons, event);
+	if (rc && rc != -EBUSY)
+		cpr->sw_stats->rx.rx_netpoll_discards += 1;
+	return rc;
+}
+
+u32 bnxt_fw_health_readl(struct bnxt *bp, int reg_idx)
+{
+	struct bnxt_fw_health *fw_health = bp->fw_health;
+	u32 reg = fw_health->regs[reg_idx];
+	u32 reg_type, reg_off, val = 0;
+
+	reg_type = BNXT_FW_HEALTH_REG_TYPE(reg);
+	reg_off = BNXT_FW_HEALTH_REG_OFF(reg);
+	switch (reg_type) {
+	case BNXT_FW_HEALTH_REG_TYPE_CFG:
+		pci_read_config_dword(bp->pdev, reg_off, &val);
+		break;
+	case BNXT_FW_HEALTH_REG_TYPE_GRC:
+		reg_off = fw_health->mapped_regs[reg_idx];
+		fallthrough;
+	case BNXT_FW_HEALTH_REG_TYPE_BAR0:
+		val = readl(bp->bar0 + reg_off);
+		break;
+	case BNXT_FW_HEALTH_REG_TYPE_BAR1:
+		val = readl(bp->bar1 + reg_off);
+		break;
+	}
+	if (reg_idx == BNXT_FW_RESET_INPROG_REG)
+		val &= fw_health->fw_reset_inprog_reg_mask;
+	return val;
+}
+
+static u16 bnxt_agg_ring_id_to_grp_idx(struct bnxt *bp, u16 ring_id)
+{
+	int i;
+
+	for (i = 0; i < bp->rx_nr_rings; i++) {
+		u16 grp_idx = bp->rx_ring[i].bnapi->index;
+		struct bnxt_ring_grp_info *grp_info;
+
+		grp_info = &bp->grp_info[grp_idx];
+		if (grp_info->agg_fw_ring_id == ring_id)
+			return grp_idx;
+	}
+	return INVALID_HW_RING_ID;
+}
+
+static u16 bnxt_get_force_speed(struct bnxt_link_info *link_info)
+{
+	struct bnxt *bp = container_of(link_info, struct bnxt, link_info);
+
+	if (bp->phy_flags & BNXT_PHY_FL_SPEEDS2)
+		return link_info->force_link_speed2;
+	if (link_info->req_signal_mode == BNXT_SIG_MODE_PAM4)
+		return link_info->force_pam4_link_speed;
+	return link_info->force_link_speed;
+}
+
+static void bnxt_set_force_speed(struct bnxt_link_info *link_info)
+{
+	struct bnxt *bp = container_of(link_info, struct bnxt, link_info);
+
+	if (bp->phy_flags & BNXT_PHY_FL_SPEEDS2) {
+		link_info->req_link_speed = link_info->force_link_speed2;
+		link_info->req_signal_mode = BNXT_SIG_MODE_NRZ;
+		switch (link_info->req_link_speed) {
+		case BNXT_LINK_SPEED_50GB_PAM4:
+		case BNXT_LINK_SPEED_100GB_PAM4:
+		case BNXT_LINK_SPEED_200GB_PAM4:
+		case BNXT_LINK_SPEED_400GB_PAM4:
+			link_info->req_signal_mode = BNXT_SIG_MODE_PAM4;
+			break;
+		case BNXT_LINK_SPEED_100GB_PAM4_112:
+		case BNXT_LINK_SPEED_200GB_PAM4_112:
+		case BNXT_LINK_SPEED_400GB_PAM4_112:
+			link_info->req_signal_mode = BNXT_SIG_MODE_PAM4_112;
+			break;
+		default:
+			link_info->req_signal_mode = BNXT_SIG_MODE_NRZ;
+		}
+		return;
+	}
+	link_info->req_link_speed = link_info->force_link_speed;
+	link_info->req_signal_mode = BNXT_SIG_MODE_NRZ;
+	if (link_info->force_pam4_link_speed) {
+		link_info->req_link_speed = link_info->force_pam4_link_speed;
+		link_info->req_signal_mode = BNXT_SIG_MODE_PAM4;
+	}
+}
+
+static void bnxt_set_auto_speed(struct bnxt_link_info *link_info)
+{
+	struct bnxt *bp = container_of(link_info, struct bnxt, link_info);
+
+	if (bp->phy_flags & BNXT_PHY_FL_SPEEDS2) {
+		link_info->advertising = link_info->auto_link_speeds2;
+		return;
+	}
+	link_info->advertising = link_info->auto_link_speeds;
+	link_info->advertising_pam4 = link_info->auto_pam4_link_speeds;
+}
+
+static bool bnxt_force_speed_updated(struct bnxt_link_info *link_info)
+{
+	struct bnxt *bp = container_of(link_info, struct bnxt, link_info);
+
+	if (bp->phy_flags & BNXT_PHY_FL_SPEEDS2) {
+		if (link_info->req_link_speed != link_info->force_link_speed2)
+			return true;
+		return false;
+	}
+	if (link_info->req_signal_mode == BNXT_SIG_MODE_NRZ &&
+	    link_info->req_link_speed != link_info->force_link_speed)
+		return true;
+	if (link_info->req_signal_mode == BNXT_SIG_MODE_PAM4 &&
+	    link_info->req_link_speed != link_info->force_pam4_link_speed)
+		return true;
+	return false;
+}
+
+static bool bnxt_auto_speed_updated(struct bnxt_link_info *link_info)
+{
+	struct bnxt *bp = container_of(link_info, struct bnxt, link_info);
+
+	if (bp->phy_flags & BNXT_PHY_FL_SPEEDS2) {
+		if (link_info->advertising != link_info->auto_link_speeds2)
+			return true;
+		return false;
+	}
+	if (link_info->advertising != link_info->auto_link_speeds ||
+	    link_info->advertising_pam4 != link_info->auto_pam4_link_speeds)
+		return true;
+	return false;
+}
+
+bool bnxt_bs_trace_avail(struct bnxt *bp, u16 type)
+{
+	u32 flags = bp->ctx->ctx_arr[type].flags;
+
+	return (flags & BNXT_CTX_MEM_TYPE_VALID) &&
+		((flags & BNXT_CTX_MEM_FW_TRACE) ||
+		 (flags & BNXT_CTX_MEM_FW_BIN_TRACE));
+}
+
+static void bnxt_bs_trace_init(struct bnxt *bp, struct bnxt_ctx_mem_type *ctxm)
+{
+	u32 mem_size, pages, rem_bytes, magic_byte_offset;
+	u16 trace_type = bnxt_bstore_to_trace[ctxm->type];
+	struct bnxt_ctx_pg_info *ctx_pg = ctxm->pg_info;
+	struct bnxt_ring_mem_info *rmem, *rmem_pg_tbl;
+	struct bnxt_bs_trace_info *bs_trace;
+	int last_pg;
+
+	if (ctxm->instance_bmap && ctxm->instance_bmap > 1)
+		return;
+
+	mem_size = ctxm->max_entries * ctxm->entry_size;
+	rem_bytes = mem_size % BNXT_PAGE_SIZE;
+	pages = DIV_ROUND_UP(mem_size, BNXT_PAGE_SIZE);
+
+	last_pg = (pages - 1) & (MAX_CTX_PAGES - 1);
+	magic_byte_offset = (rem_bytes ? rem_bytes : BNXT_PAGE_SIZE) - 1;
+
+	rmem = &ctx_pg[0].ring_mem;
+	bs_trace = &bp->bs_trace[trace_type];
+	bs_trace->ctx_type = ctxm->type;
+	bs_trace->trace_type = trace_type;
+	if (pages > MAX_CTX_PAGES) {
+		int last_pg_dir = rmem->nr_pages - 1;
+
+		rmem_pg_tbl = &ctx_pg[0].ctx_pg_tbl[last_pg_dir]->ring_mem;
+		bs_trace->magic_byte = rmem_pg_tbl->pg_arr[last_pg];
+	} else {
+		bs_trace->magic_byte = rmem->pg_arr[last_pg];
+	}
+	bs_trace->magic_byte += magic_byte_offset;
+	*bs_trace->magic_byte = BNXT_TRACE_BUF_MAGIC_BYTE;
+}
+
+#define BNXT_EVENT_BUF_PRODUCER_TYPE(data1)				\
+	(((data1) & ASYNC_EVENT_CMPL_DBG_BUF_PRODUCER_EVENT_DATA1_TYPE_MASK) >>\
+	 ASYNC_EVENT_CMPL_DBG_BUF_PRODUCER_EVENT_DATA1_TYPE_SFT)
+
+#define BNXT_EVENT_BUF_PRODUCER_OFFSET(data2)				\
+	(((data2) &							\
+	  ASYNC_EVENT_CMPL_DBG_BUF_PRODUCER_EVENT_DATA2_CURR_OFF_MASK) >>\
+	 ASYNC_EVENT_CMPL_DBG_BUF_PRODUCER_EVENT_DATA2_CURR_OFF_SFT)
+
+#define BNXT_EVENT_THERMAL_CURRENT_TEMP(data2)				\
+	((data2) &							\
+	  ASYNC_EVENT_CMPL_ERROR_REPORT_THERMAL_EVENT_DATA2_CURRENT_TEMP_MASK)
+
+#define BNXT_EVENT_THERMAL_THRESHOLD_TEMP(data2)			\
+	(((data2) &							\
+	  ASYNC_EVENT_CMPL_ERROR_REPORT_THERMAL_EVENT_DATA2_THRESHOLD_TEMP_MASK) >>\
+	 ASYNC_EVENT_CMPL_ERROR_REPORT_THERMAL_EVENT_DATA2_THRESHOLD_TEMP_SFT)
+
+#define EVENT_DATA1_THERMAL_THRESHOLD_TYPE(data1)			\
+	((data1) &							\
+	 ASYNC_EVENT_CMPL_ERROR_REPORT_THERMAL_EVENT_DATA1_THRESHOLD_TYPE_MASK)
+
+#define EVENT_DATA1_THERMAL_THRESHOLD_DIR_INCREASING(data1)		\
+	(((data1) &							\
+	  ASYNC_EVENT_CMPL_ERROR_REPORT_THERMAL_EVENT_DATA1_TRANSITION_DIR) ==\
+	 ASYNC_EVENT_CMPL_ERROR_REPORT_THERMAL_EVENT_DATA1_TRANSITION_DIR_INCREASING)
+
+/* Return true if the workqueue has to be scheduled */
+static bool bnxt_event_error_report(struct bnxt *bp, u32 data1, u32 data2)
+{
+	u32 err_type = BNXT_EVENT_ERROR_REPORT_TYPE(data1);
+
+	switch (err_type) {
+	case ASYNC_EVENT_CMPL_ERROR_REPORT_BASE_EVENT_DATA1_ERROR_TYPE_INVALID_SIGNAL:
+		netdev_err(bp->dev, "1PPS: Received invalid signal on pin%lu from the external source. Please fix the signal and reconfigure the pin\n",
+			   BNXT_EVENT_INVALID_SIGNAL_DATA(data2));
+		break;
+	case ASYNC_EVENT_CMPL_ERROR_REPORT_BASE_EVENT_DATA1_ERROR_TYPE_PAUSE_STORM:
+		netdev_warn(bp->dev, "Pause Storm detected!\n");
+		break;
+	case ASYNC_EVENT_CMPL_ERROR_REPORT_BASE_EVENT_DATA1_ERROR_TYPE_DOORBELL_DROP_THRESHOLD:
+		netdev_warn(bp->dev, "One or more MMIO doorbells dropped by the device!\n");
+		break;
+	case ASYNC_EVENT_CMPL_ERROR_REPORT_BASE_EVENT_DATA1_ERROR_TYPE_THERMAL_THRESHOLD: {
+		u32 type = EVENT_DATA1_THERMAL_THRESHOLD_TYPE(data1);
+		char *threshold_type;
+		bool notify = false;
+		char *dir_str;
+
+		switch (type) {
+		case ASYNC_EVENT_CMPL_ERROR_REPORT_THERMAL_EVENT_DATA1_THRESHOLD_TYPE_WARN:
+			threshold_type = "warning";
+			break;
+		case ASYNC_EVENT_CMPL_ERROR_REPORT_THERMAL_EVENT_DATA1_THRESHOLD_TYPE_CRITICAL:
+			threshold_type = "critical";
+			break;
+		case ASYNC_EVENT_CMPL_ERROR_REPORT_THERMAL_EVENT_DATA1_THRESHOLD_TYPE_FATAL:
+			threshold_type = "fatal";
+			break;
+		case ASYNC_EVENT_CMPL_ERROR_REPORT_THERMAL_EVENT_DATA1_THRESHOLD_TYPE_SHUTDOWN:
+			threshold_type = "shutdown";
+			break;
+		default:
+			netdev_err(bp->dev, "Unknown Thermal threshold type event\n");
+			return false;
+		}
+		if (EVENT_DATA1_THERMAL_THRESHOLD_DIR_INCREASING(data1)) {
+			dir_str = "above";
+			notify = true;
+		} else {
+			dir_str = "below";
+		}
+		netdev_warn(bp->dev, "Chip temperature has gone %s the %s thermal threshold!\n",
+			    dir_str, threshold_type);
+		netdev_warn(bp->dev, "Temperature (In Celsius), Current: %lu, threshold: %lu\n",
+			    BNXT_EVENT_THERMAL_CURRENT_TEMP(data2),
+			    BNXT_EVENT_THERMAL_THRESHOLD_TEMP(data2));
+		if (notify) {
+			bp->thermal_threshold_type = type;
+			set_bit(BNXT_THERMAL_THRESHOLD_SP_EVENT, &bp->sp_event);
+			return true;
+		}
+		return false;
+	}
+	case ASYNC_EVENT_CMPL_ERROR_REPORT_BASE_EVENT_DATA1_ERROR_TYPE_DUAL_DATA_RATE_NOT_SUPPORTED:
+		netdev_warn(bp->dev, "Speed change not supported with dual rate transceivers on this board\n");
+		break;
+	default:
+		netdev_err(bp->dev, "FW reported unknown error type %u\n",
+			   err_type);
+		break;
+	}
+	return false;
 }
 
 #define BNXT_GET_EVENT_PORT(data)	\
 	((data) &			\
 	 ASYNC_EVENT_CMPL_PORT_CONN_NOT_ALLOWED_EVENT_DATA1_PORT_ID_MASK)
 
+#define BNXT_EVENT_RING_TYPE(data2)	\
+	((data2) &			\
+	 ASYNC_EVENT_CMPL_RING_MONITOR_MSG_EVENT_DATA2_DISABLE_RING_TYPE_MASK)
+
+#define BNXT_EVENT_RING_TYPE_RX(data2)	\
+	(BNXT_EVENT_RING_TYPE(data2) ==	\
+	 ASYNC_EVENT_CMPL_RING_MONITOR_MSG_EVENT_DATA2_DISABLE_RING_TYPE_RX)
+
+#define BNXT_EVENT_PHC_EVENT_TYPE(data1)	\
+	(((data1) & ASYNC_EVENT_CMPL_PHC_UPDATE_EVENT_DATA1_FLAGS_MASK) >>\
+	 ASYNC_EVENT_CMPL_PHC_UPDATE_EVENT_DATA1_FLAGS_SFT)
+
+#define BNXT_EVENT_PHC_RTC_UPDATE(data1)	\
+	(((data1) & ASYNC_EVENT_CMPL_PHC_UPDATE_EVENT_DATA1_PHC_TIME_MSB_MASK) >>\
+	 ASYNC_EVENT_CMPL_PHC_UPDATE_EVENT_DATA1_PHC_TIME_MSB_SFT)
+
+#define BNXT_PHC_BITS	48
+
 static int bnxt_async_event_process(struct bnxt *bp,
 				    struct hwrm_async_event_cmpl *cmpl)
 {
 	u16 event_id = le16_to_cpu(cmpl->event_id);
+	u32 data1 = le32_to_cpu(cmpl->event_data1);
+	u32 data2 = le32_to_cpu(cmpl->event_data2);
+
+	netdev_dbg(bp->dev, "hwrm event 0x%x {0x%x, 0x%x}\n",
+		   event_id, data1, data2);
 
 	/* TODO CHIMP_FW: Define event id's for link change, error etc */
 	switch (event_id) {
 	case ASYNC_EVENT_CMPL_EVENT_ID_LINK_SPEED_CFG_CHANGE: {
-		u32 data1 = le32_to_cpu(cmpl->event_data1);
 		struct bnxt_link_info *link_info = &bp->link_info;
 
 		if (BNXT_VF(bp))
 			goto async_event_process_exit;
-		if (data1 & 0x20000) {
-			u16 fw_speed = link_info->force_link_speed;
+
+		/* print unsupported speed warning in forced speed mode only */
+		if (!(link_info->autoneg & BNXT_AUTONEG_SPEED) &&
+		    (data1 & 0x20000)) {
+			u16 fw_speed = bnxt_get_force_speed(link_info);
 			u32 speed = bnxt_fw_to_ethtool_speed(fw_speed);
 
-			netdev_warn(bp->dev, "Link speed %d no longer supported\n",
-				    speed);
+			if (speed != SPEED_UNKNOWN)
+				netdev_warn(bp->dev, "Link speed %d no longer supported\n",
+					    speed);
 		}
 		set_bit(BNXT_LINK_SPEED_CHNG_SP_EVENT, &bp->sp_event);
-		/* fall thru */
 	}
+		fallthrough;
+	case ASYNC_EVENT_CMPL_EVENT_ID_LINK_SPEED_CHANGE:
+	case ASYNC_EVENT_CMPL_EVENT_ID_PORT_PHY_CFG_CHANGE:
+		set_bit(BNXT_LINK_CFG_CHANGE_SP_EVENT, &bp->sp_event);
+		fallthrough;
 	case ASYNC_EVENT_CMPL_EVENT_ID_LINK_STATUS_CHANGE:
 		set_bit(BNXT_LINK_CHNG_SP_EVENT, &bp->sp_event);
 		break;
@@ -1607,7 +2748,6 @@ static int bnxt_async_event_process(struct bnxt *bp,
 		set_bit(BNXT_HWRM_PF_UNLOAD_SP_EVENT, &bp->sp_event);
 		break;
 	case ASYNC_EVENT_CMPL_EVENT_ID_PORT_CONN_NOT_ALLOWED: {
-		u32 data1 = le32_to_cpu(cmpl->event_data1);
 		u16 port_id = BNXT_GET_EVENT_PORT(data1);
 
 		if (BNXT_VF(bp))
@@ -1624,10 +2764,167 @@ static int bnxt_async_event_process(struct bnxt *bp,
 			goto async_event_process_exit;
 		set_bit(BNXT_RESET_TASK_SILENT_SP_EVENT, &bp->sp_event);
 		break;
+	case ASYNC_EVENT_CMPL_EVENT_ID_RESET_NOTIFY: {
+		char *type_str = "Solicited";
+
+		if (!bp->fw_health)
+			goto async_event_process_exit;
+
+		bp->fw_reset_timestamp = jiffies;
+		bp->fw_reset_min_dsecs = cmpl->timestamp_lo;
+		if (!bp->fw_reset_min_dsecs)
+			bp->fw_reset_min_dsecs = BNXT_DFLT_FW_RST_MIN_DSECS;
+		bp->fw_reset_max_dsecs = le16_to_cpu(cmpl->timestamp_hi);
+		if (!bp->fw_reset_max_dsecs)
+			bp->fw_reset_max_dsecs = BNXT_DFLT_FW_RST_MAX_DSECS;
+		if (EVENT_DATA1_RESET_NOTIFY_FW_ACTIVATION(data1)) {
+			set_bit(BNXT_STATE_FW_ACTIVATE_RESET, &bp->state);
+		} else if (EVENT_DATA1_RESET_NOTIFY_FATAL(data1)) {
+			type_str = "Fatal";
+			bp->fw_health->fatalities++;
+			set_bit(BNXT_STATE_FW_FATAL_COND, &bp->state);
+		} else if (data2 && BNXT_FW_STATUS_HEALTHY !=
+			   EVENT_DATA2_RESET_NOTIFY_FW_STATUS_CODE(data2)) {
+			type_str = "Non-fatal";
+			bp->fw_health->survivals++;
+			set_bit(BNXT_STATE_FW_NON_FATAL_COND, &bp->state);
+		}
+		netif_warn(bp, hw, bp->dev,
+			   "%s firmware reset event, data1: 0x%x, data2: 0x%x, min wait %u ms, max wait %u ms\n",
+			   type_str, data1, data2,
+			   bp->fw_reset_min_dsecs * 100,
+			   bp->fw_reset_max_dsecs * 100);
+		set_bit(BNXT_FW_RESET_NOTIFY_SP_EVENT, &bp->sp_event);
+		break;
+	}
+	case ASYNC_EVENT_CMPL_EVENT_ID_ERROR_RECOVERY: {
+		struct bnxt_fw_health *fw_health = bp->fw_health;
+		char *status_desc = "healthy";
+		u32 status;
+
+		if (!fw_health)
+			goto async_event_process_exit;
+
+		if (!EVENT_DATA1_RECOVERY_ENABLED(data1)) {
+			fw_health->enabled = false;
+			netif_info(bp, drv, bp->dev, "Driver recovery watchdog is disabled\n");
+			break;
+		}
+		fw_health->primary = EVENT_DATA1_RECOVERY_MASTER_FUNC(data1);
+		fw_health->tmr_multiplier =
+			DIV_ROUND_UP(fw_health->polling_dsecs * HZ,
+				     bp->current_interval * 10);
+		fw_health->tmr_counter = fw_health->tmr_multiplier;
+		if (!fw_health->enabled)
+			fw_health->last_fw_heartbeat =
+				bnxt_fw_health_readl(bp, BNXT_FW_HEARTBEAT_REG);
+		fw_health->last_fw_reset_cnt =
+			bnxt_fw_health_readl(bp, BNXT_FW_RESET_CNT_REG);
+		status = bnxt_fw_health_readl(bp, BNXT_FW_HEALTH_REG);
+		if (status != BNXT_FW_STATUS_HEALTHY)
+			status_desc = "unhealthy";
+		netif_info(bp, drv, bp->dev,
+			   "Driver recovery watchdog, role: %s, firmware status: 0x%x (%s), resets: %u\n",
+			   fw_health->primary ? "primary" : "backup", status,
+			   status_desc, fw_health->last_fw_reset_cnt);
+		if (!fw_health->enabled) {
+			/* Make sure tmr_counter is set and visible to
+			 * bnxt_health_check() before setting enabled to true.
+			 */
+			smp_wmb();
+			fw_health->enabled = true;
+		}
+		goto async_event_process_exit;
+	}
+	case ASYNC_EVENT_CMPL_EVENT_ID_DEBUG_NOTIFICATION:
+		netif_notice(bp, hw, bp->dev,
+			     "Received firmware debug notification, data1: 0x%x, data2: 0x%x\n",
+			     data1, data2);
+		goto async_event_process_exit;
+	case ASYNC_EVENT_CMPL_EVENT_ID_RING_MONITOR_MSG: {
+		struct bnxt_rx_ring_info *rxr;
+		u16 grp_idx;
+
+		if (bp->flags & BNXT_FLAG_CHIP_P5_PLUS)
+			goto async_event_process_exit;
+
+		netdev_warn(bp->dev, "Ring monitor event, ring type %lu id 0x%x\n",
+			    BNXT_EVENT_RING_TYPE(data2), data1);
+		if (!BNXT_EVENT_RING_TYPE_RX(data2))
+			goto async_event_process_exit;
+
+		grp_idx = bnxt_agg_ring_id_to_grp_idx(bp, data1);
+		if (grp_idx == INVALID_HW_RING_ID) {
+			netdev_warn(bp->dev, "Unknown RX agg ring id 0x%x\n",
+				    data1);
+			goto async_event_process_exit;
+		}
+		rxr = bp->bnapi[grp_idx]->rx_ring;
+		bnxt_sched_reset_rxr(bp, rxr);
+		goto async_event_process_exit;
+	}
+	case ASYNC_EVENT_CMPL_EVENT_ID_ECHO_REQUEST: {
+		struct bnxt_fw_health *fw_health = bp->fw_health;
+
+		netif_notice(bp, hw, bp->dev,
+			     "Received firmware echo request, data1: 0x%x, data2: 0x%x\n",
+			     data1, data2);
+		if (fw_health) {
+			fw_health->echo_req_data1 = data1;
+			fw_health->echo_req_data2 = data2;
+			set_bit(BNXT_FW_ECHO_REQUEST_SP_EVENT, &bp->sp_event);
+			break;
+		}
+		goto async_event_process_exit;
+	}
+	case ASYNC_EVENT_CMPL_EVENT_ID_PPS_TIMESTAMP: {
+		bnxt_ptp_pps_event(bp, data1, data2);
+		goto async_event_process_exit;
+	}
+	case ASYNC_EVENT_CMPL_EVENT_ID_ERROR_REPORT: {
+		if (bnxt_event_error_report(bp, data1, data2))
+			break;
+		goto async_event_process_exit;
+	}
+	case ASYNC_EVENT_CMPL_EVENT_ID_PHC_UPDATE: {
+		switch (BNXT_EVENT_PHC_EVENT_TYPE(data1)) {
+		case ASYNC_EVENT_CMPL_PHC_UPDATE_EVENT_DATA1_FLAGS_PHC_RTC_UPDATE:
+			if (BNXT_PTP_USE_RTC(bp)) {
+				struct bnxt_ptp_cfg *ptp = bp->ptp_cfg;
+				unsigned long flags;
+				u64 ns;
+
+				if (!ptp)
+					goto async_event_process_exit;
+
+				bnxt_ptp_update_current_time(bp);
+				ns = (((u64)BNXT_EVENT_PHC_RTC_UPDATE(data1) <<
+				       BNXT_PHC_BITS) | ptp->current_time);
+				write_seqlock_irqsave(&ptp->ptp_lock, flags);
+				bnxt_ptp_rtc_timecounter_init(ptp, ns);
+				write_sequnlock_irqrestore(&ptp->ptp_lock, flags);
+			}
+			break;
+		}
+		goto async_event_process_exit;
+	}
+	case ASYNC_EVENT_CMPL_EVENT_ID_DEFERRED_RESPONSE: {
+		u16 seq_id = le32_to_cpu(cmpl->event_data2) & 0xffff;
+
+		hwrm_update_token(bp, seq_id, BNXT_HWRM_DEFERRED);
+		goto async_event_process_exit;
+	}
+	case ASYNC_EVENT_CMPL_EVENT_ID_DBG_BUF_PRODUCER: {
+		u16 type = (u16)BNXT_EVENT_BUF_PRODUCER_TYPE(data1);
+		u32 offset =  BNXT_EVENT_BUF_PRODUCER_OFFSET(data2);
+
+		bnxt_bs_trace_check_wrap(&bp->bs_trace[type], offset);
+		goto async_event_process_exit;
+	}
 	default:
 		goto async_event_process_exit;
 	}
-	schedule_work(&bp->sp_task);
+	__bnxt_queue_sp_work(bp);
 async_event_process_exit:
 	bnxt_ulp_async_events(bp, cmpl);
 	return 0;
@@ -1643,10 +2940,7 @@ static int bnxt_hwrm_handler(struct bnxt *bp, struct tx_cmp *txcmp)
 	switch (cmpl_type) {
 	case CMPL_BASE_TYPE_HWRM_DONE:
 		seq_id = le16_to_cpu(h_cmpl->sequence_id);
-		if (seq_id == bp->hwrm_intr_seq_id)
-			bp->hwrm_intr_seq_id = HWRM_SEQ_ID_INVALID;
-		else
-			netdev_err(bp->dev, "Invalid hwrm seq id %d\n", seq_id);
+		hwrm_update_token(bp, seq_id, BNXT_HWRM_COMPLETE);
 		break;
 
 	case CMPL_BASE_TYPE_HWRM_FWD_REQ:
@@ -1660,13 +2954,13 @@ static int bnxt_hwrm_handler(struct bnxt *bp, struct tx_cmp *txcmp)
 		}
 
 		set_bit(vf_id - bp->pf.first_vf_id, bp->pf.vf_event_bmap);
-		set_bit(BNXT_HWRM_EXEC_FWD_REQ_SP_EVENT, &bp->sp_event);
-		schedule_work(&bp->sp_task);
+		bnxt_queue_sp_work(bp, BNXT_HWRM_EXEC_FWD_REQ_SP_EVENT);
 		break;
 
 	case CMPL_BASE_TYPE_HWRM_ASYNC_EVENT:
 		bnxt_async_event_process(bp,
 					 (struct hwrm_async_event_cmpl *)txcmp);
+		break;
 
 	default:
 		break;
@@ -1675,6 +2969,13 @@ static int bnxt_hwrm_handler(struct bnxt *bp, struct tx_cmp *txcmp)
 	return 0;
 }
 
+static bool bnxt_vnic_is_active(struct bnxt *bp)
+{
+	struct bnxt_vnic_info *vnic = &bp->vnic_info[0];
+
+	return vnic->fw_vnic_id != INVALID_HW_RING_ID && vnic->mru > 0;
+}
+
 static irqreturn_t bnxt_msix(int irq, void *dev_instance)
 {
 	struct bnxt_napi *bnapi = dev_instance;
@@ -1682,6 +2983,7 @@ static irqreturn_t bnxt_msix(int irq, void *dev_instance)
 	struct bnxt_cp_ring_info *cpr = &bnapi->cp_ring;
 	u32 cons = RING_CMP(cpr->cp_raw_cons);
 
+	cpr->event_ctr++;
 	prefetch(&cpr->cp_desc_ring[CP_RING(cons)][CP_IDX(cons)]);
 	napi_schedule(&bnapi->napi);
 	return IRQ_HANDLED;
@@ -1698,45 +3000,21 @@ static inline int bnxt_has_work(struct bnxt *bp, struct bnxt_cp_ring_info *cpr)
 	return TX_CMP_VALID(txcmp, raw_cons);
 }
 
-static irqreturn_t bnxt_inta(int irq, void *dev_instance)
-{
-	struct bnxt_napi *bnapi = dev_instance;
-	struct bnxt *bp = bnapi->bp;
-	struct bnxt_cp_ring_info *cpr = &bnapi->cp_ring;
-	u32 cons = RING_CMP(cpr->cp_raw_cons);
-	u32 int_status;
-
-	prefetch(&cpr->cp_desc_ring[CP_RING(cons)][CP_IDX(cons)]);
-
-	if (!bnxt_has_work(bp, cpr)) {
-		int_status = readl(bp->bar0 + BNXT_CAG_REG_LEGACY_INT_STATUS);
-		/* return if erroneous interrupt */
-		if (!(int_status & (0x10000 << cpr->cp_ring_struct.fw_ring_id)))
-			return IRQ_NONE;
-	}
-
-	/* disable ring IRQ */
-	BNXT_CP_DB_IRQ_DIS(cpr->cp_doorbell);
-
-	/* Return here if interrupt is shared and is disabled. */
-	if (unlikely(atomic_read(&bp->intr_sem) != 0))
-		return IRQ_HANDLED;
-
-	napi_schedule(&bnapi->napi);
-	return IRQ_HANDLED;
-}
-
-static int bnxt_poll_work(struct bnxt *bp, struct bnxt_napi *bnapi, int budget)
+static int __bnxt_poll_work(struct bnxt *bp, struct bnxt_cp_ring_info *cpr,
+			    int budget)
 {
-	struct bnxt_cp_ring_info *cpr = &bnapi->cp_ring;
+	struct bnxt_napi *bnapi = cpr->bnapi;
 	u32 raw_cons = cpr->cp_raw_cons;
+	bool flush_xdp = false;
 	u32 cons;
-	int tx_pkts = 0;
 	int rx_pkts = 0;
 	u8 event = 0;
 	struct tx_cmp *txcmp;
 
+	cpr->has_more_work = 0;
+	cpr->had_work_done = 1;
 	while (1) {
+		u8 cmp_type;
 		int rc;
 
 		cons = RING_CMP(raw_cons);
@@ -1749,60 +3027,120 @@ static int bnxt_poll_work(struct bnxt *bp, struct bnxt_napi *bnapi, int budget)
 		 * reading any further.
 		 */
 		dma_rmb();
-		if (TX_CMP_TYPE(txcmp) == CMP_TYPE_TX_L2_CMP) {
-			tx_pkts++;
+		cmp_type = TX_CMP_TYPE(txcmp);
+		if (cmp_type == CMP_TYPE_TX_L2_CMP ||
+		    cmp_type == CMP_TYPE_TX_L2_COAL_CMP) {
+			u32 opaque = txcmp->tx_cmp_opaque;
+			struct bnxt_tx_ring_info *txr;
+			u16 tx_freed;
+
+			txr = bnapi->tx_ring[TX_OPAQUE_RING(opaque)];
+			event |= BNXT_TX_CMP_EVENT;
+			if (cmp_type == CMP_TYPE_TX_L2_COAL_CMP)
+				txr->tx_hw_cons = TX_CMP_SQ_CONS_IDX(txcmp);
+			else
+				txr->tx_hw_cons = TX_OPAQUE_PROD(bp, opaque);
+			tx_freed = (txr->tx_hw_cons - txr->tx_cons) &
+				   bp->tx_ring_mask;
 			/* return full budget so NAPI will complete. */
-			if (unlikely(tx_pkts > bp->tx_wake_thresh))
+			if (unlikely(tx_freed >= bp->tx_wake_thresh)) {
 				rx_pkts = budget;
-		} else if ((TX_CMP_TYPE(txcmp) & 0x30) == 0x10) {
-			rc = bnxt_rx_pkt(bp, bnapi, &raw_cons, &event);
+				raw_cons = NEXT_RAW_CMP(raw_cons);
+				if (budget)
+					cpr->has_more_work = 1;
+				break;
+			}
+		} else if (cmp_type == CMP_TYPE_TX_L2_PKT_TS_CMP) {
+			bnxt_tx_ts_cmp(bp, bnapi, (struct tx_ts_cmp *)txcmp);
+		} else if (cmp_type >= CMP_TYPE_RX_L2_CMP &&
+			   cmp_type <= CMP_TYPE_RX_L2_TPA_START_V3_CMP) {
+			if (likely(budget))
+				rc = bnxt_rx_pkt(bp, cpr, &raw_cons, &event);
+			else
+				rc = bnxt_force_rx_discard(bp, cpr, &raw_cons,
+							   &event);
+			if (event & BNXT_REDIRECT_EVENT)
+				flush_xdp = true;
 			if (likely(rc >= 0))
 				rx_pkts += rc;
+			/* Increment rx_pkts when rc is -ENOMEM to count towards
+			 * the NAPI budget.  Otherwise, we may potentially loop
+			 * here forever if we consistently cannot allocate
+			 * buffers.
+			 */
+			else if (rc == -ENOMEM && budget)
+				rx_pkts++;
 			else if (rc == -EBUSY)	/* partial completion */
 				break;
-		} else if (unlikely((TX_CMP_TYPE(txcmp) ==
-				     CMPL_BASE_TYPE_HWRM_DONE) ||
-				    (TX_CMP_TYPE(txcmp) ==
-				     CMPL_BASE_TYPE_HWRM_FWD_REQ) ||
-				    (TX_CMP_TYPE(txcmp) ==
-				     CMPL_BASE_TYPE_HWRM_ASYNC_EVENT))) {
+		} else if (unlikely(cmp_type == CMPL_BASE_TYPE_HWRM_DONE ||
+				    cmp_type == CMPL_BASE_TYPE_HWRM_FWD_REQ ||
+				    cmp_type == CMPL_BASE_TYPE_HWRM_ASYNC_EVENT)) {
 			bnxt_hwrm_handler(bp, txcmp);
 		}
 		raw_cons = NEXT_RAW_CMP(raw_cons);
 
-		if (rx_pkts == budget)
+		if (rx_pkts && rx_pkts == budget) {
+			cpr->has_more_work = 1;
 			break;
+		}
+	}
+
+	if (flush_xdp) {
+		xdp_do_flush();
+		event &= ~BNXT_REDIRECT_EVENT;
 	}
 
 	if (event & BNXT_TX_EVENT) {
-		struct bnxt_tx_ring_info *txr = bnapi->tx_ring;
-		void __iomem *db = txr->tx_doorbell;
+		struct bnxt_tx_ring_info *txr = bnapi->tx_ring[0];
 		u16 prod = txr->tx_prod;
 
 		/* Sync BD data before updating doorbell */
 		wmb();
 
-		bnxt_db_write(bp, db, DB_KEY_TX | prod);
+		bnxt_db_write_relaxed(bp, &txr->tx_db, prod);
+		event &= ~BNXT_TX_EVENT;
 	}
 
 	cpr->cp_raw_cons = raw_cons;
-	/* ACK completion ring before freeing tx ring and producing new
-	 * buffers in rx/agg rings to prevent overflowing the completion
-	 * ring.
-	 */
-	BNXT_CP_DB(cpr->cp_doorbell, cpr->cp_raw_cons);
+	bnapi->events |= event;
+	return rx_pkts;
+}
 
-	if (tx_pkts)
-		bnapi->tx_int(bp, bnapi, tx_pkts);
+static void __bnxt_poll_work_done(struct bnxt *bp, struct bnxt_napi *bnapi,
+				  int budget)
+{
+	if ((bnapi->events & BNXT_TX_CMP_EVENT) && !bnapi->tx_fault)
+		bnapi->tx_int(bp, bnapi, budget);
 
-	if (event & BNXT_RX_EVENT) {
+	if ((bnapi->events & BNXT_RX_EVENT) && !(bnapi->in_reset)) {
+		struct bnxt_rx_ring_info *rxr = bnapi->rx_ring;
+
+		bnxt_db_write(bp, &rxr->rx_db, rxr->rx_prod);
+		bnapi->events &= ~BNXT_RX_EVENT;
+	}
+	if (bnapi->events & BNXT_AGG_EVENT) {
 		struct bnxt_rx_ring_info *rxr = bnapi->rx_ring;
 
-		bnxt_db_write(bp, rxr->rx_doorbell, DB_KEY_RX | rxr->rx_prod);
-		if (event & BNXT_AGG_EVENT)
-			bnxt_db_write(bp, rxr->rx_agg_doorbell,
-				      DB_KEY_RX | rxr->rx_agg_prod);
+		bnxt_db_write(bp, &rxr->rx_agg_db, rxr->rx_agg_prod);
+		bnapi->events &= ~BNXT_AGG_EVENT;
 	}
+}
+
+static int bnxt_poll_work(struct bnxt *bp, struct bnxt_cp_ring_info *cpr,
+			  int budget)
+{
+	struct bnxt_napi *bnapi = cpr->bnapi;
+	int rx_pkts;
+
+	rx_pkts = __bnxt_poll_work(bp, cpr, budget);
+
+	/* ACK completion ring before freeing tx ring and producing new
+	 * buffers in rx/agg rings to prevent overflowing the completion
+	 * ring.
+	 */
+	bnxt_db_cq(bp, &cpr->cp_db, cpr->cp_raw_cons);
+
+	__bnxt_poll_work_done(bp, bnapi, budget);
 	return rx_pkts;
 }
 
@@ -1816,6 +3154,7 @@ static int bnxt_poll_nitroa0(struct napi_struct *napi, int budget)
 	struct rx_cmp_ext *rxcmp1;
 	u32 cp_cons, tmp_raw_cons;
 	u32 raw_cons = cpr->cp_raw_cons;
+	bool flush_xdp = false;
 	u32 rx_pkts = 0;
 	u8 event = 0;
 
@@ -1828,6 +3167,10 @@ static int bnxt_poll_nitroa0(struct napi_struct *napi, int budget)
 		if (!TX_CMP_VALID(txcmp, raw_cons))
 			break;
 
+		/* The valid test of the entry must be done first before
+		 * reading any further.
+		 */
+		dma_rmb();
 		if ((TX_CMP_TYPE(txcmp) & 0x30) == 0x10) {
 			tmp_raw_cons = NEXT_RAW_CMP(raw_cons);
 			cp_cons = RING_CMP(tmp_raw_cons);
@@ -1841,11 +3184,13 @@ static int bnxt_poll_nitroa0(struct napi_struct *napi, int budget)
 			rxcmp1->rx_cmp_cfa_code_errors_v2 |=
 				cpu_to_le32(RX_CMPL_ERRORS_CRC_ERROR);
 
-			rc = bnxt_rx_pkt(bp, bnapi, &raw_cons, &event);
-			if (likely(rc == -EIO))
+			rc = bnxt_rx_pkt(bp, cpr, &raw_cons, &event);
+			if (likely(rc == -EIO) && budget)
 				rx_pkts++;
 			else if (rc == -EBUSY)	/* partial completion */
 				break;
+			if (event & BNXT_REDIRECT_EVENT)
+				flush_xdp = true;
 		} else if (unlikely(TX_CMP_TYPE(txcmp) ==
 				    CMPL_BASE_TYPE_HWRM_DONE)) {
 			bnxt_hwrm_handler(bp, txcmp);
@@ -1860,16 +3205,17 @@ static int bnxt_poll_nitroa0(struct napi_struct *napi, int budget)
 	}
 
 	cpr->cp_raw_cons = raw_cons;
-	BNXT_CP_DB(cpr->cp_doorbell, cpr->cp_raw_cons);
-	bnxt_db_write(bp, rxr->rx_doorbell, DB_KEY_RX | rxr->rx_prod);
+	BNXT_DB_CQ(&cpr->cp_db, cpr->cp_raw_cons);
+	bnxt_db_write(bp, &rxr->rx_db, rxr->rx_prod);
 
 	if (event & BNXT_AGG_EVENT)
-		bnxt_db_write(bp, rxr->rx_agg_doorbell,
-			      DB_KEY_RX | rxr->rx_agg_prod);
+		bnxt_db_write(bp, &rxr->rx_agg_db, rxr->rx_agg_prod);
+	if (flush_xdp)
+		xdp_do_flush();
 
 	if (!bnxt_has_work(bp, cpr) && rx_pkts < budget) {
 		napi_complete_done(napi, rx_pkts);
-		BNXT_CP_DB_REARM(cpr->cp_doorbell, cpr->cp_raw_cons);
+		BNXT_DB_CQ_ARM(&cpr->cp_db, cpr->cp_raw_cons);
 	}
 	return rx_pkts;
 }
@@ -1881,227 +3227,562 @@ static int bnxt_poll(struct napi_struct *napi, int budget)
 	struct bnxt_cp_ring_info *cpr = &bnapi->cp_ring;
 	int work_done = 0;
 
+	if (unlikely(test_bit(BNXT_STATE_FW_FATAL_COND, &bp->state))) {
+		napi_complete(napi);
+		return 0;
+	}
 	while (1) {
-		work_done += bnxt_poll_work(bp, bnapi, budget - work_done);
+		work_done += bnxt_poll_work(bp, cpr, budget - work_done);
 
-		if (work_done >= budget)
+		if (work_done >= budget) {
+			if (!budget)
+				BNXT_DB_CQ_ARM(&cpr->cp_db, cpr->cp_raw_cons);
 			break;
+		}
 
 		if (!bnxt_has_work(bp, cpr)) {
 			if (napi_complete_done(napi, work_done))
-				BNXT_CP_DB_REARM(cpr->cp_doorbell,
-						 cpr->cp_raw_cons);
+				BNXT_DB_CQ_ARM(&cpr->cp_db, cpr->cp_raw_cons);
 			break;
 		}
 	}
-	mmiowb();
+	if ((bp->flags & BNXT_FLAG_DIM) && bnxt_vnic_is_active(bp)) {
+		struct dim_sample dim_sample = {};
+
+		dim_update_sample(cpr->event_ctr,
+				  cpr->rx_packets,
+				  cpr->rx_bytes,
+				  &dim_sample);
+		net_dim(&cpr->dim, &dim_sample);
+	}
 	return work_done;
 }
 
-static void bnxt_free_tx_skbs(struct bnxt *bp)
+static int __bnxt_poll_cqs(struct bnxt *bp, struct bnxt_napi *bnapi, int budget)
 {
-	int i, max_idx;
-	struct pci_dev *pdev = bp->pdev;
+	struct bnxt_cp_ring_info *cpr = &bnapi->cp_ring;
+	int i, work_done = 0;
 
-	if (!bp->tx_ring)
-		return;
+	for (i = 0; i < cpr->cp_ring_count; i++) {
+		struct bnxt_cp_ring_info *cpr2 = &cpr->cp_ring_arr[i];
 
-	max_idx = bp->tx_nr_pages * TX_DESC_CNT;
-	for (i = 0; i < bp->tx_nr_rings; i++) {
-		struct bnxt_tx_ring_info *txr = &bp->tx_ring[i];
-		int j;
+		if (cpr2->had_nqe_notify) {
+			work_done += __bnxt_poll_work(bp, cpr2,
+						      budget - work_done);
+			cpr->has_more_work |= cpr2->has_more_work;
+		}
+	}
+	return work_done;
+}
 
-		for (j = 0; j < max_idx;) {
-			struct bnxt_sw_tx_bd *tx_buf = &txr->tx_buf_ring[j];
-			struct sk_buff *skb = tx_buf->skb;
-			int k, last;
+static void __bnxt_poll_cqs_done(struct bnxt *bp, struct bnxt_napi *bnapi,
+				 u64 dbr_type, int budget)
+{
+	struct bnxt_cp_ring_info *cpr = &bnapi->cp_ring;
+	int i;
 
-			if (!skb) {
-				j++;
-				continue;
-			}
+	for (i = 0; i < cpr->cp_ring_count; i++) {
+		struct bnxt_cp_ring_info *cpr2 = &cpr->cp_ring_arr[i];
+		struct bnxt_db_info *db;
 
-			tx_buf->skb = NULL;
+		if (cpr2->had_work_done) {
+			u32 tgl = 0;
 
-			if (tx_buf->is_push) {
-				dev_kfree_skb(skb);
-				j += 2;
-				continue;
+			if (dbr_type == DBR_TYPE_CQ_ARMALL) {
+				cpr2->had_nqe_notify = 0;
+				tgl = cpr2->toggle;
 			}
+			db = &cpr2->cp_db;
+			bnxt_writeq(bp,
+				    db->db_key64 | dbr_type | DB_TOGGLE(tgl) |
+				    DB_RING_IDX(db, cpr2->cp_raw_cons),
+				    db->doorbell);
+			cpr2->had_work_done = 0;
+		}
+	}
+	__bnxt_poll_work_done(bp, bnapi, budget);
+}
+
+static int bnxt_poll_p5(struct napi_struct *napi, int budget)
+{
+	struct bnxt_napi *bnapi = container_of(napi, struct bnxt_napi, napi);
+	struct bnxt_cp_ring_info *cpr = &bnapi->cp_ring;
+	struct bnxt_cp_ring_info *cpr_rx;
+	u32 raw_cons = cpr->cp_raw_cons;
+	struct bnxt *bp = bnapi->bp;
+	struct nqe_cn *nqcmp;
+	int work_done = 0;
+	u32 cons;
+
+	if (unlikely(test_bit(BNXT_STATE_FW_FATAL_COND, &bp->state))) {
+		napi_complete(napi);
+		return 0;
+	}
+	if (cpr->has_more_work) {
+		cpr->has_more_work = 0;
+		work_done = __bnxt_poll_cqs(bp, bnapi, budget);
+	}
+	while (1) {
+		u16 type;
+
+		cons = RING_CMP(raw_cons);
+		nqcmp = &cpr->nq_desc_ring[CP_RING(cons)][CP_IDX(cons)];
+
+		if (!NQ_CMP_VALID(nqcmp, raw_cons)) {
+			if (cpr->has_more_work)
+				break;
+
+			__bnxt_poll_cqs_done(bp, bnapi, DBR_TYPE_CQ_ARMALL,
+					     budget);
+			cpr->cp_raw_cons = raw_cons;
+			if (napi_complete_done(napi, work_done))
+				BNXT_DB_NQ_ARM_P5(&cpr->cp_db,
+						  cpr->cp_raw_cons);
+			goto poll_done;
+		}
 
+		/* The valid test of the entry must be done first before
+		 * reading any further.
+		 */
+		dma_rmb();
+
+		type = le16_to_cpu(nqcmp->type);
+		if (NQE_CN_TYPE(type) == NQ_CN_TYPE_CQ_NOTIFICATION) {
+			u32 idx = le32_to_cpu(nqcmp->cq_handle_low);
+			u32 cq_type = BNXT_NQ_HDL_TYPE(idx);
+			struct bnxt_cp_ring_info *cpr2;
+
+			/* No more budget for RX work */
+			if (budget && work_done >= budget &&
+			    cq_type == BNXT_NQ_HDL_TYPE_RX)
+				break;
+
+			idx = BNXT_NQ_HDL_IDX(idx);
+			cpr2 = &cpr->cp_ring_arr[idx];
+			cpr2->had_nqe_notify = 1;
+			cpr2->toggle = NQE_CN_TOGGLE(type);
+			work_done += __bnxt_poll_work(bp, cpr2,
+						      budget - work_done);
+			cpr->has_more_work |= cpr2->has_more_work;
+		} else {
+			bnxt_hwrm_handler(bp, (struct tx_cmp *)nqcmp);
+		}
+		raw_cons = NEXT_RAW_CMP(raw_cons);
+	}
+	__bnxt_poll_cqs_done(bp, bnapi, DBR_TYPE_CQ, budget);
+	if (raw_cons != cpr->cp_raw_cons) {
+		cpr->cp_raw_cons = raw_cons;
+		BNXT_DB_NQ_P5(&cpr->cp_db, raw_cons);
+	}
+poll_done:
+	cpr_rx = &cpr->cp_ring_arr[0];
+	if (cpr_rx->cp_ring_type == BNXT_NQ_HDL_TYPE_RX &&
+	    (bp->flags & BNXT_FLAG_DIM) && bnxt_vnic_is_active(bp)) {
+		struct dim_sample dim_sample = {};
+
+		dim_update_sample(cpr->event_ctr,
+				  cpr_rx->rx_packets,
+				  cpr_rx->rx_bytes,
+				  &dim_sample);
+		net_dim(&cpr->dim, &dim_sample);
+	}
+	return work_done;
+}
+
+static void bnxt_free_one_tx_ring_skbs(struct bnxt *bp,
+				       struct bnxt_tx_ring_info *txr, int idx)
+{
+	int i, max_idx;
+	struct pci_dev *pdev = bp->pdev;
+
+	max_idx = bp->tx_nr_pages * TX_DESC_CNT;
+
+	for (i = 0; i < max_idx;) {
+		struct bnxt_sw_tx_bd *tx_buf = &txr->tx_buf_ring[i];
+		struct sk_buff *skb;
+		int j, last;
+
+		if (idx  < bp->tx_nr_rings_xdp &&
+		    tx_buf->action == XDP_REDIRECT) {
 			dma_unmap_single(&pdev->dev,
 					 dma_unmap_addr(tx_buf, mapping),
-					 skb_headlen(skb),
-					 PCI_DMA_TODEVICE);
-
-			last = tx_buf->nr_frags;
-			j += 2;
-			for (k = 0; k < last; k++, j++) {
-				int ring_idx = j & bp->tx_ring_mask;
-				skb_frag_t *frag = &skb_shinfo(skb)->frags[k];
-
-				tx_buf = &txr->tx_buf_ring[ring_idx];
-				dma_unmap_page(
-					&pdev->dev,
-					dma_unmap_addr(tx_buf, mapping),
-					skb_frag_size(frag), PCI_DMA_TODEVICE);
-			}
+					 dma_unmap_len(tx_buf, len),
+					 DMA_TO_DEVICE);
+			xdp_return_frame(tx_buf->xdpf);
+			tx_buf->action = 0;
+			tx_buf->xdpf = NULL;
+			i++;
+			continue;
+		}
+
+		skb = tx_buf->skb;
+		if (!skb) {
+			i++;
+			continue;
+		}
+
+		tx_buf->skb = NULL;
+
+		if (tx_buf->is_push) {
 			dev_kfree_skb(skb);
+			i += 2;
+			continue;
 		}
-		netdev_tx_reset_queue(netdev_get_tx_queue(bp->dev, i));
+
+		dma_unmap_single(&pdev->dev,
+				 dma_unmap_addr(tx_buf, mapping),
+				 skb_headlen(skb),
+				 DMA_TO_DEVICE);
+
+		last = tx_buf->nr_frags;
+		i += 2;
+		for (j = 0; j < last; j++, i++) {
+			int ring_idx = i & bp->tx_ring_mask;
+			skb_frag_t *frag = &skb_shinfo(skb)->frags[j];
+
+			tx_buf = &txr->tx_buf_ring[ring_idx];
+			netmem_dma_unmap_page_attrs(&pdev->dev,
+						    dma_unmap_addr(tx_buf,
+								   mapping),
+						    skb_frag_size(frag),
+						    DMA_TO_DEVICE, 0);
+		}
+		dev_kfree_skb(skb);
 	}
+	netdev_tx_reset_queue(netdev_get_tx_queue(bp->dev, idx));
 }
 
-static void bnxt_free_rx_skbs(struct bnxt *bp)
+static void bnxt_free_tx_skbs(struct bnxt *bp)
 {
-	int i, max_idx, max_agg_idx;
-	struct pci_dev *pdev = bp->pdev;
+	int i;
 
-	if (!bp->rx_ring)
+	if (!bp->tx_ring)
 		return;
 
-	max_idx = bp->rx_nr_pages * RX_DESC_CNT;
-	max_agg_idx = bp->rx_agg_nr_pages * RX_DESC_CNT;
-	for (i = 0; i < bp->rx_nr_rings; i++) {
-		struct bnxt_rx_ring_info *rxr = &bp->rx_ring[i];
-		int j;
+	for (i = 0; i < bp->tx_nr_rings; i++) {
+		struct bnxt_tx_ring_info *txr = &bp->tx_ring[i];
 
-		if (rxr->rx_tpa) {
-			for (j = 0; j < MAX_TPA; j++) {
-				struct bnxt_tpa_info *tpa_info =
-							&rxr->rx_tpa[j];
-				u8 *data = tpa_info->data;
+		if (!txr->tx_buf_ring)
+			continue;
 
-				if (!data)
-					continue;
+		bnxt_free_one_tx_ring_skbs(bp, txr, i);
+	}
+
+	if (bp->ptp_cfg && !(bp->fw_cap & BNXT_FW_CAP_TX_TS_CMP))
+		bnxt_ptp_free_txts_skbs(bp->ptp_cfg);
+}
 
-				dma_unmap_single_attrs(&pdev->dev,
-						       tpa_info->mapping,
-						       bp->rx_buf_use_size,
-						       bp->rx_dir,
-						       DMA_ATTR_WEAK_ORDERING);
+static void bnxt_free_one_rx_ring(struct bnxt *bp, struct bnxt_rx_ring_info *rxr)
+{
+	int i, max_idx;
 
-				tpa_info->data = NULL;
+	max_idx = bp->rx_nr_pages * RX_DESC_CNT;
 
-				kfree(data);
-			}
-		}
+	for (i = 0; i < max_idx; i++) {
+		struct bnxt_sw_rx_bd *rx_buf = &rxr->rx_buf_ring[i];
+		void *data = rx_buf->data;
 
-		for (j = 0; j < max_idx; j++) {
-			struct bnxt_sw_rx_bd *rx_buf = &rxr->rx_buf_ring[j];
-			dma_addr_t mapping = rx_buf->mapping;
-			void *data = rx_buf->data;
+		if (!data)
+			continue;
 
-			if (!data)
-				continue;
+		rx_buf->data = NULL;
+		if (BNXT_RX_PAGE_MODE(bp))
+			page_pool_recycle_direct(rxr->page_pool, data);
+		else
+			page_pool_free_va(rxr->head_pool, data, true);
+	}
+}
 
-			rx_buf->data = NULL;
+static void bnxt_free_one_rx_agg_ring(struct bnxt *bp, struct bnxt_rx_ring_info *rxr)
+{
+	int i, max_idx;
 
-			if (BNXT_RX_PAGE_MODE(bp)) {
-				mapping -= bp->rx_dma_offset;
-				dma_unmap_page_attrs(&pdev->dev, mapping,
-						     PAGE_SIZE, bp->rx_dir,
-						     DMA_ATTR_WEAK_ORDERING);
-				__free_page(data);
-			} else {
-				dma_unmap_single_attrs(&pdev->dev, mapping,
-						       bp->rx_buf_use_size,
-						       bp->rx_dir,
-						       DMA_ATTR_WEAK_ORDERING);
-				kfree(data);
-			}
-		}
+	max_idx = bp->rx_agg_nr_pages * RX_DESC_CNT;
 
-		for (j = 0; j < max_agg_idx; j++) {
-			struct bnxt_sw_rx_agg_bd *rx_agg_buf =
-				&rxr->rx_agg_ring[j];
-			struct page *page = rx_agg_buf->page;
+	for (i = 0; i < max_idx; i++) {
+		struct bnxt_sw_rx_agg_bd *rx_agg_buf = &rxr->rx_agg_ring[i];
+		netmem_ref netmem = rx_agg_buf->netmem;
 
-			if (!page)
-				continue;
+		if (!netmem)
+			continue;
 
-			dma_unmap_page_attrs(&pdev->dev, rx_agg_buf->mapping,
-					     BNXT_RX_PAGE_SIZE,
-					     PCI_DMA_FROMDEVICE,
-					     DMA_ATTR_WEAK_ORDERING);
+		rx_agg_buf->netmem = 0;
+		__clear_bit(i, rxr->rx_agg_bmap);
 
-			rx_agg_buf->page = NULL;
-			__clear_bit(j, rxr->rx_agg_bmap);
+		page_pool_recycle_direct_netmem(rxr->page_pool, netmem);
+	}
+}
 
-			__free_page(page);
-		}
-		if (rxr->rx_page) {
-			__free_page(rxr->rx_page);
-			rxr->rx_page = NULL;
-		}
+static void bnxt_free_one_tpa_info_data(struct bnxt *bp,
+					struct bnxt_rx_ring_info *rxr)
+{
+	int i;
+
+	for (i = 0; i < bp->max_tpa; i++) {
+		struct bnxt_tpa_info *tpa_info = &rxr->rx_tpa[i];
+		u8 *data = tpa_info->data;
+
+		if (!data)
+			continue;
+
+		tpa_info->data = NULL;
+		page_pool_free_va(rxr->head_pool, data, false);
 	}
 }
 
+static void bnxt_free_one_rx_ring_skbs(struct bnxt *bp,
+				       struct bnxt_rx_ring_info *rxr)
+{
+	struct bnxt_tpa_idx_map *map;
+
+	if (!rxr->rx_tpa)
+		goto skip_rx_tpa_free;
+
+	bnxt_free_one_tpa_info_data(bp, rxr);
+
+skip_rx_tpa_free:
+	if (!rxr->rx_buf_ring)
+		goto skip_rx_buf_free;
+
+	bnxt_free_one_rx_ring(bp, rxr);
+
+skip_rx_buf_free:
+	if (!rxr->rx_agg_ring)
+		goto skip_rx_agg_free;
+
+	bnxt_free_one_rx_agg_ring(bp, rxr);
+
+skip_rx_agg_free:
+	map = rxr->rx_tpa_idx_map;
+	if (map)
+		memset(map->agg_idx_bmap, 0, sizeof(map->agg_idx_bmap));
+}
+
+static void bnxt_free_rx_skbs(struct bnxt *bp)
+{
+	int i;
+
+	if (!bp->rx_ring)
+		return;
+
+	for (i = 0; i < bp->rx_nr_rings; i++)
+		bnxt_free_one_rx_ring_skbs(bp, &bp->rx_ring[i]);
+}
+
 static void bnxt_free_skbs(struct bnxt *bp)
 {
 	bnxt_free_tx_skbs(bp);
 	bnxt_free_rx_skbs(bp);
 }
 
-static void bnxt_free_ring(struct bnxt *bp, struct bnxt_ring_struct *ring)
+static void bnxt_init_ctx_mem(struct bnxt_ctx_mem_type *ctxm, void *p, int len)
+{
+	u8 init_val = ctxm->init_value;
+	u16 offset = ctxm->init_offset;
+	u8 *p2 = p;
+	int i;
+
+	if (!init_val)
+		return;
+	if (offset == BNXT_CTX_INIT_INVALID_OFFSET) {
+		memset(p, init_val, len);
+		return;
+	}
+	for (i = 0; i < len; i += ctxm->entry_size)
+		*(p2 + i + offset) = init_val;
+}
+
+static size_t __bnxt_copy_ring(struct bnxt *bp, struct bnxt_ring_mem_info *rmem,
+			       void *buf, size_t offset, size_t head,
+			       size_t tail)
+{
+	int i, head_page, start_idx, source_offset;
+	size_t len, rem_len, total_len, max_bytes;
+
+	head_page = head / rmem->page_size;
+	source_offset = head % rmem->page_size;
+	total_len = (tail - head) & MAX_CTX_BYTES_MASK;
+	if (!total_len)
+		total_len = MAX_CTX_BYTES;
+	start_idx = head_page % MAX_CTX_PAGES;
+	max_bytes = (rmem->nr_pages - start_idx) * rmem->page_size -
+		    source_offset;
+	total_len = min(total_len, max_bytes);
+	rem_len = total_len;
+
+	for (i = start_idx; rem_len; i++, source_offset = 0) {
+		len = min((size_t)(rmem->page_size - source_offset), rem_len);
+		if (buf)
+			memcpy(buf + offset, rmem->pg_arr[i] + source_offset,
+			       len);
+		offset += len;
+		rem_len -= len;
+	}
+	return total_len;
+}
+
+static void bnxt_free_ring(struct bnxt *bp, struct bnxt_ring_mem_info *rmem)
 {
 	struct pci_dev *pdev = bp->pdev;
 	int i;
 
-	for (i = 0; i < ring->nr_pages; i++) {
-		if (!ring->pg_arr[i])
+	if (!rmem->pg_arr)
+		goto skip_pages;
+
+	for (i = 0; i < rmem->nr_pages; i++) {
+		if (!rmem->pg_arr[i])
 			continue;
 
-		dma_free_coherent(&pdev->dev, ring->page_size,
-				  ring->pg_arr[i], ring->dma_arr[i]);
+		dma_free_coherent(&pdev->dev, rmem->page_size,
+				  rmem->pg_arr[i], rmem->dma_arr[i]);
 
-		ring->pg_arr[i] = NULL;
+		rmem->pg_arr[i] = NULL;
 	}
-	if (ring->pg_tbl) {
-		dma_free_coherent(&pdev->dev, ring->nr_pages * 8,
-				  ring->pg_tbl, ring->pg_tbl_map);
-		ring->pg_tbl = NULL;
+skip_pages:
+	if (rmem->pg_tbl) {
+		size_t pg_tbl_size = rmem->nr_pages * 8;
+
+		if (rmem->flags & BNXT_RMEM_USE_FULL_PAGE_FLAG)
+			pg_tbl_size = rmem->page_size;
+		dma_free_coherent(&pdev->dev, pg_tbl_size,
+				  rmem->pg_tbl, rmem->pg_tbl_map);
+		rmem->pg_tbl = NULL;
 	}
-	if (ring->vmem_size && *ring->vmem) {
-		vfree(*ring->vmem);
-		*ring->vmem = NULL;
+	if (rmem->vmem_size && *rmem->vmem) {
+		vfree(*rmem->vmem);
+		*rmem->vmem = NULL;
 	}
 }
 
-static int bnxt_alloc_ring(struct bnxt *bp, struct bnxt_ring_struct *ring)
+static int bnxt_alloc_ring(struct bnxt *bp, struct bnxt_ring_mem_info *rmem)
 {
-	int i;
 	struct pci_dev *pdev = bp->pdev;
+	u64 valid_bit = 0;
+	int i;
 
-	if (ring->nr_pages > 1) {
-		ring->pg_tbl = dma_alloc_coherent(&pdev->dev,
-						  ring->nr_pages * 8,
-						  &ring->pg_tbl_map,
+	if (rmem->flags & (BNXT_RMEM_VALID_PTE_FLAG | BNXT_RMEM_RING_PTE_FLAG))
+		valid_bit = PTU_PTE_VALID;
+	if ((rmem->nr_pages > 1 || rmem->depth > 0) && !rmem->pg_tbl) {
+		size_t pg_tbl_size = rmem->nr_pages * 8;
+
+		if (rmem->flags & BNXT_RMEM_USE_FULL_PAGE_FLAG)
+			pg_tbl_size = rmem->page_size;
+		rmem->pg_tbl = dma_alloc_coherent(&pdev->dev, pg_tbl_size,
+						  &rmem->pg_tbl_map,
 						  GFP_KERNEL);
-		if (!ring->pg_tbl)
+		if (!rmem->pg_tbl)
 			return -ENOMEM;
 	}
 
-	for (i = 0; i < ring->nr_pages; i++) {
-		ring->pg_arr[i] = dma_alloc_coherent(&pdev->dev,
-						     ring->page_size,
-						     &ring->dma_arr[i],
+	for (i = 0; i < rmem->nr_pages; i++) {
+		u64 extra_bits = valid_bit;
+
+		rmem->pg_arr[i] = dma_alloc_coherent(&pdev->dev,
+						     rmem->page_size,
+						     &rmem->dma_arr[i],
 						     GFP_KERNEL);
-		if (!ring->pg_arr[i])
+		if (!rmem->pg_arr[i])
 			return -ENOMEM;
 
-		if (ring->nr_pages > 1)
-			ring->pg_tbl[i] = cpu_to_le64(ring->dma_arr[i]);
+		if (rmem->ctx_mem)
+			bnxt_init_ctx_mem(rmem->ctx_mem, rmem->pg_arr[i],
+					  rmem->page_size);
+		if (rmem->nr_pages > 1 || rmem->depth > 0) {
+			if (i == rmem->nr_pages - 2 &&
+			    (rmem->flags & BNXT_RMEM_RING_PTE_FLAG))
+				extra_bits |= PTU_PTE_NEXT_TO_LAST;
+			else if (i == rmem->nr_pages - 1 &&
+				 (rmem->flags & BNXT_RMEM_RING_PTE_FLAG))
+				extra_bits |= PTU_PTE_LAST;
+			rmem->pg_tbl[i] =
+				cpu_to_le64(rmem->dma_arr[i] | extra_bits);
+		}
 	}
 
-	if (ring->vmem_size) {
-		*ring->vmem = vzalloc(ring->vmem_size);
-		if (!(*ring->vmem))
+	if (rmem->vmem_size) {
+		*rmem->vmem = vzalloc(rmem->vmem_size);
+		if (!(*rmem->vmem))
 			return -ENOMEM;
 	}
 	return 0;
 }
 
+static void bnxt_free_one_tpa_info(struct bnxt *bp,
+				   struct bnxt_rx_ring_info *rxr)
+{
+	int i;
+
+	kfree(rxr->rx_tpa_idx_map);
+	rxr->rx_tpa_idx_map = NULL;
+	if (rxr->rx_tpa) {
+		for (i = 0; i < bp->max_tpa; i++) {
+			kfree(rxr->rx_tpa[i].agg_arr);
+			rxr->rx_tpa[i].agg_arr = NULL;
+		}
+	}
+	kfree(rxr->rx_tpa);
+	rxr->rx_tpa = NULL;
+}
+
+static void bnxt_free_tpa_info(struct bnxt *bp)
+{
+	int i;
+
+	for (i = 0; i < bp->rx_nr_rings; i++) {
+		struct bnxt_rx_ring_info *rxr = &bp->rx_ring[i];
+
+		bnxt_free_one_tpa_info(bp, rxr);
+	}
+}
+
+static int bnxt_alloc_one_tpa_info(struct bnxt *bp,
+				   struct bnxt_rx_ring_info *rxr)
+{
+	struct rx_agg_cmp *agg;
+	int i;
+
+	rxr->rx_tpa = kcalloc(bp->max_tpa, sizeof(struct bnxt_tpa_info),
+			      GFP_KERNEL);
+	if (!rxr->rx_tpa)
+		return -ENOMEM;
+
+	if (!(bp->flags & BNXT_FLAG_CHIP_P5_PLUS))
+		return 0;
+	for (i = 0; i < bp->max_tpa; i++) {
+		agg = kcalloc(MAX_SKB_FRAGS, sizeof(*agg), GFP_KERNEL);
+		if (!agg)
+			return -ENOMEM;
+		rxr->rx_tpa[i].agg_arr = agg;
+	}
+	rxr->rx_tpa_idx_map = kzalloc(sizeof(*rxr->rx_tpa_idx_map),
+				      GFP_KERNEL);
+	if (!rxr->rx_tpa_idx_map)
+		return -ENOMEM;
+
+	return 0;
+}
+
+static int bnxt_alloc_tpa_info(struct bnxt *bp)
+{
+	int i, rc;
+
+	bp->max_tpa = MAX_TPA;
+	if (bp->flags & BNXT_FLAG_CHIP_P5_PLUS) {
+		if (!bp->max_tpa_v2)
+			return 0;
+		bp->max_tpa = max_t(u16, bp->max_tpa_v2, MAX_TPA_P5);
+	}
+
+	for (i = 0; i < bp->rx_nr_rings; i++) {
+		struct bnxt_rx_ring_info *rxr = &bp->rx_ring[i];
+
+		rc = bnxt_alloc_one_tpa_info(bp, rxr);
+		if (rc)
+			return rc;
+	}
+	return 0;
+}
+
 static void bnxt_free_rx_rings(struct bnxt *bp)
 {
 	int i;
@@ -2109,6 +3790,7 @@ static void bnxt_free_rx_rings(struct bnxt *bp)
 	if (!bp->rx_ring)
 		return;
 
+	bnxt_free_tpa_info(bp);
 	for (i = 0; i < bp->rx_nr_rings; i++) {
 		struct bnxt_rx_ring_info *rxr = &bp->rx_ring[i];
 		struct bnxt_ring_struct *ring;
@@ -2116,23 +3798,93 @@ static void bnxt_free_rx_rings(struct bnxt *bp)
 		if (rxr->xdp_prog)
 			bpf_prog_put(rxr->xdp_prog);
 
-		kfree(rxr->rx_tpa);
-		rxr->rx_tpa = NULL;
+		if (xdp_rxq_info_is_reg(&rxr->xdp_rxq))
+			xdp_rxq_info_unreg(&rxr->xdp_rxq);
+
+		page_pool_destroy(rxr->page_pool);
+		page_pool_destroy(rxr->head_pool);
+		rxr->page_pool = rxr->head_pool = NULL;
 
 		kfree(rxr->rx_agg_bmap);
 		rxr->rx_agg_bmap = NULL;
 
 		ring = &rxr->rx_ring_struct;
-		bnxt_free_ring(bp, ring);
+		bnxt_free_ring(bp, &ring->ring_mem);
 
 		ring = &rxr->rx_agg_ring_struct;
-		bnxt_free_ring(bp, ring);
+		bnxt_free_ring(bp, &ring->ring_mem);
 	}
 }
 
+static int bnxt_alloc_rx_page_pool(struct bnxt *bp,
+				   struct bnxt_rx_ring_info *rxr,
+				   int numa_node)
+{
+	const unsigned int agg_size_fac = PAGE_SIZE / BNXT_RX_PAGE_SIZE;
+	const unsigned int rx_size_fac = PAGE_SIZE / SZ_4K;
+	struct page_pool_params pp = { 0 };
+	struct page_pool *pool;
+
+	pp.pool_size = bp->rx_agg_ring_size / agg_size_fac;
+	if (BNXT_RX_PAGE_MODE(bp))
+		pp.pool_size += bp->rx_ring_size / rx_size_fac;
+	pp.nid = numa_node;
+	pp.netdev = bp->dev;
+	pp.dev = &bp->pdev->dev;
+	pp.dma_dir = bp->rx_dir;
+	pp.max_len = PAGE_SIZE;
+	pp.flags = PP_FLAG_DMA_MAP | PP_FLAG_DMA_SYNC_DEV |
+		   PP_FLAG_ALLOW_UNREADABLE_NETMEM;
+	pp.queue_idx = rxr->bnapi->index;
+
+	pool = page_pool_create(&pp);
+	if (IS_ERR(pool))
+		return PTR_ERR(pool);
+	rxr->page_pool = pool;
+
+	rxr->need_head_pool = page_pool_is_unreadable(pool);
+	if (bnxt_separate_head_pool(rxr)) {
+		pp.pool_size = min(bp->rx_ring_size / rx_size_fac, 1024);
+		pp.flags = PP_FLAG_DMA_MAP | PP_FLAG_DMA_SYNC_DEV;
+		pool = page_pool_create(&pp);
+		if (IS_ERR(pool))
+			goto err_destroy_pp;
+	} else {
+		page_pool_get(pool);
+	}
+	rxr->head_pool = pool;
+
+	return 0;
+
+err_destroy_pp:
+	page_pool_destroy(rxr->page_pool);
+	rxr->page_pool = NULL;
+	return PTR_ERR(pool);
+}
+
+static void bnxt_enable_rx_page_pool(struct bnxt_rx_ring_info *rxr)
+{
+	page_pool_enable_direct_recycling(rxr->head_pool, &rxr->bnapi->napi);
+	page_pool_enable_direct_recycling(rxr->page_pool, &rxr->bnapi->napi);
+}
+
+static int bnxt_alloc_rx_agg_bmap(struct bnxt *bp, struct bnxt_rx_ring_info *rxr)
+{
+	u16 mem_size;
+
+	rxr->rx_agg_bmap_size = bp->rx_agg_ring_mask + 1;
+	mem_size = rxr->rx_agg_bmap_size / 8;
+	rxr->rx_agg_bmap = kzalloc(mem_size, GFP_KERNEL);
+	if (!rxr->rx_agg_bmap)
+		return -ENOMEM;
+
+	return 0;
+}
+
 static int bnxt_alloc_rx_rings(struct bnxt *bp)
 {
-	int i, rc, agg_rings = 0, tpa_rings = 0;
+	int numa_node = dev_to_node(&bp->pdev->dev);
+	int i, rc = 0, agg_rings = 0, cpu;
 
 	if (!bp->rx_ring)
 		return -ENOMEM;
@@ -2140,43 +3892,54 @@ static int bnxt_alloc_rx_rings(struct bnxt *bp)
 	if (bp->flags & BNXT_FLAG_AGG_RINGS)
 		agg_rings = 1;
 
-	if (bp->flags & BNXT_FLAG_TPA)
-		tpa_rings = 1;
-
 	for (i = 0; i < bp->rx_nr_rings; i++) {
 		struct bnxt_rx_ring_info *rxr = &bp->rx_ring[i];
 		struct bnxt_ring_struct *ring;
+		int cpu_node;
 
 		ring = &rxr->rx_ring_struct;
 
-		rc = bnxt_alloc_ring(bp, ring);
+		cpu = cpumask_local_spread(i, numa_node);
+		cpu_node = cpu_to_node(cpu);
+		netdev_dbg(bp->dev, "Allocating page pool for rx_ring[%d] on numa_node: %d\n",
+			   i, cpu_node);
+		rc = bnxt_alloc_rx_page_pool(bp, rxr, cpu_node);
 		if (rc)
 			return rc;
+		bnxt_enable_rx_page_pool(rxr);
 
-		if (agg_rings) {
-			u16 mem_size;
+		rc = xdp_rxq_info_reg(&rxr->xdp_rxq, bp->dev, i, 0);
+		if (rc < 0)
+			return rc;
+
+		rc = xdp_rxq_info_reg_mem_model(&rxr->xdp_rxq,
+						MEM_TYPE_PAGE_POOL,
+						rxr->page_pool);
+		if (rc) {
+			xdp_rxq_info_unreg(&rxr->xdp_rxq);
+			return rc;
+		}
 
+		rc = bnxt_alloc_ring(bp, &ring->ring_mem);
+		if (rc)
+			return rc;
+
+		ring->grp_idx = i;
+		if (agg_rings) {
 			ring = &rxr->rx_agg_ring_struct;
-			rc = bnxt_alloc_ring(bp, ring);
+			rc = bnxt_alloc_ring(bp, &ring->ring_mem);
 			if (rc)
 				return rc;
 
-			rxr->rx_agg_bmap_size = bp->rx_agg_ring_mask + 1;
-			mem_size = rxr->rx_agg_bmap_size / 8;
-			rxr->rx_agg_bmap = kzalloc(mem_size, GFP_KERNEL);
-			if (!rxr->rx_agg_bmap)
-				return -ENOMEM;
-
-			if (tpa_rings) {
-				rxr->rx_tpa = kcalloc(MAX_TPA,
-						sizeof(struct bnxt_tpa_info),
-						GFP_KERNEL);
-				if (!rxr->rx_tpa)
-					return -ENOMEM;
-			}
+			ring->grp_idx = i;
+			rc = bnxt_alloc_rx_agg_bmap(bp, rxr);
+			if (rc)
+				return rc;
 		}
 	}
-	return 0;
+	if (bp->flags & BNXT_FLAG_TPA)
+		rc = bnxt_alloc_tpa_info(bp);
+	return rc;
 }
 
 static void bnxt_free_tx_rings(struct bnxt *bp)
@@ -2199,10 +3962,19 @@ static void bnxt_free_tx_rings(struct bnxt *bp)
 
 		ring = &txr->tx_ring_struct;
 
-		bnxt_free_ring(bp, ring);
+		bnxt_free_ring(bp, &ring->ring_mem);
 	}
 }
 
+#define BNXT_TC_TO_RING_BASE(bp, tc)	\
+	((tc) * (bp)->tx_nr_rings_per_tc)
+
+#define BNXT_RING_TO_TC_OFF(bp, tx)	\
+	((tx) % (bp)->tx_nr_rings_per_tc)
+
+#define BNXT_RING_TO_TC(bp, tx)		\
+	((tx) / (bp)->tx_nr_rings_per_tc)
+
 static int bnxt_alloc_tx_rings(struct bnxt *bp)
 {
 	int i, j, rc;
@@ -2226,13 +3998,15 @@ static int bnxt_alloc_tx_rings(struct bnxt *bp)
 	for (i = 0, j = 0; i < bp->tx_nr_rings; i++) {
 		struct bnxt_tx_ring_info *txr = &bp->tx_ring[i];
 		struct bnxt_ring_struct *ring;
+		u8 qidx;
 
 		ring = &txr->tx_ring_struct;
 
-		rc = bnxt_alloc_ring(bp, ring);
+		rc = bnxt_alloc_ring(bp, &ring->ring_mem);
 		if (rc)
 			return rc;
 
+		ring->grp_idx = txr->bnapi->index;
 		if (bp->tx_push_size) {
 			dma_addr_t mapping;
 
@@ -2250,18 +4024,74 @@ static int bnxt_alloc_tx_rings(struct bnxt *bp)
 			mapping = txr->tx_push_mapping +
 				sizeof(struct tx_push_bd);
 			txr->data_mapping = cpu_to_le64(mapping);
-
-			memset(txr->tx_push, 0, sizeof(struct tx_push_bd));
 		}
-		ring->queue_id = bp->q_info[j].queue_id;
+		qidx = bp->tc_to_qidx[j];
+		ring->queue_id = bp->q_info[qidx].queue_id;
+		spin_lock_init(&txr->xdp_tx_lock);
 		if (i < bp->tx_nr_rings_xdp)
 			continue;
-		if (i % bp->tx_nr_rings_per_tc == (bp->tx_nr_rings_per_tc - 1))
+		if (BNXT_RING_TO_TC_OFF(bp, i) == (bp->tx_nr_rings_per_tc - 1))
 			j++;
 	}
 	return 0;
 }
 
+static void bnxt_free_cp_arrays(struct bnxt_cp_ring_info *cpr)
+{
+	struct bnxt_ring_struct *ring = &cpr->cp_ring_struct;
+
+	kfree(cpr->cp_desc_ring);
+	cpr->cp_desc_ring = NULL;
+	ring->ring_mem.pg_arr = NULL;
+	kfree(cpr->cp_desc_mapping);
+	cpr->cp_desc_mapping = NULL;
+	ring->ring_mem.dma_arr = NULL;
+}
+
+static int bnxt_alloc_cp_arrays(struct bnxt_cp_ring_info *cpr, int n)
+{
+	cpr->cp_desc_ring = kcalloc(n, sizeof(*cpr->cp_desc_ring), GFP_KERNEL);
+	if (!cpr->cp_desc_ring)
+		return -ENOMEM;
+	cpr->cp_desc_mapping = kcalloc(n, sizeof(*cpr->cp_desc_mapping),
+				       GFP_KERNEL);
+	if (!cpr->cp_desc_mapping)
+		return -ENOMEM;
+	return 0;
+}
+
+static void bnxt_free_all_cp_arrays(struct bnxt *bp)
+{
+	int i;
+
+	if (!bp->bnapi)
+		return;
+	for (i = 0; i < bp->cp_nr_rings; i++) {
+		struct bnxt_napi *bnapi = bp->bnapi[i];
+
+		if (!bnapi)
+			continue;
+		bnxt_free_cp_arrays(&bnapi->cp_ring);
+	}
+}
+
+static int bnxt_alloc_all_cp_arrays(struct bnxt *bp)
+{
+	int i, n = bp->cp_nr_pages;
+
+	for (i = 0; i < bp->cp_nr_rings; i++) {
+		struct bnxt_napi *bnapi = bp->bnapi[i];
+		int rc;
+
+		if (!bnapi)
+			continue;
+		rc = bnxt_alloc_cp_arrays(&bnapi->cp_ring, n);
+		if (rc)
+			return rc;
+	}
+	return 0;
+}
+
 static void bnxt_free_cp_rings(struct bnxt *bp)
 {
 	int i;
@@ -2273,6 +4103,7 @@ static void bnxt_free_cp_rings(struct bnxt *bp)
 		struct bnxt_napi *bnapi = bp->bnapi[i];
 		struct bnxt_cp_ring_info *cpr;
 		struct bnxt_ring_struct *ring;
+		int j;
 
 		if (!bnapi)
 			continue;
@@ -2280,38 +4111,193 @@ static void bnxt_free_cp_rings(struct bnxt *bp)
 		cpr = &bnapi->cp_ring;
 		ring = &cpr->cp_ring_struct;
 
-		bnxt_free_ring(bp, ring);
+		bnxt_free_ring(bp, &ring->ring_mem);
+
+		if (!cpr->cp_ring_arr)
+			continue;
+
+		for (j = 0; j < cpr->cp_ring_count; j++) {
+			struct bnxt_cp_ring_info *cpr2 = &cpr->cp_ring_arr[j];
+
+			ring = &cpr2->cp_ring_struct;
+			bnxt_free_ring(bp, &ring->ring_mem);
+			bnxt_free_cp_arrays(cpr2);
+		}
+		kfree(cpr->cp_ring_arr);
+		cpr->cp_ring_arr = NULL;
+		cpr->cp_ring_count = 0;
+	}
+}
+
+static int bnxt_alloc_cp_sub_ring(struct bnxt *bp,
+				  struct bnxt_cp_ring_info *cpr)
+{
+	struct bnxt_ring_mem_info *rmem;
+	struct bnxt_ring_struct *ring;
+	int rc;
+
+	rc = bnxt_alloc_cp_arrays(cpr, bp->cp_nr_pages);
+	if (rc) {
+		bnxt_free_cp_arrays(cpr);
+		return -ENOMEM;
+	}
+	ring = &cpr->cp_ring_struct;
+	rmem = &ring->ring_mem;
+	rmem->nr_pages = bp->cp_nr_pages;
+	rmem->page_size = HW_CMPD_RING_SIZE;
+	rmem->pg_arr = (void **)cpr->cp_desc_ring;
+	rmem->dma_arr = cpr->cp_desc_mapping;
+	rmem->flags = BNXT_RMEM_RING_PTE_FLAG;
+	rc = bnxt_alloc_ring(bp, rmem);
+	if (rc) {
+		bnxt_free_ring(bp, rmem);
+		bnxt_free_cp_arrays(cpr);
 	}
+	return rc;
 }
 
 static int bnxt_alloc_cp_rings(struct bnxt *bp)
 {
-	int i, rc;
+	bool sh = !!(bp->flags & BNXT_FLAG_SHARED_RINGS);
+	int i, j, rc, ulp_msix;
+	int tcs = bp->num_tc;
 
-	for (i = 0; i < bp->cp_nr_rings; i++) {
+	if (!tcs)
+		tcs = 1;
+	ulp_msix = bnxt_get_ulp_msix_num(bp);
+	for (i = 0, j = 0; i < bp->cp_nr_rings; i++) {
 		struct bnxt_napi *bnapi = bp->bnapi[i];
-		struct bnxt_cp_ring_info *cpr;
+		struct bnxt_cp_ring_info *cpr, *cpr2;
 		struct bnxt_ring_struct *ring;
+		int cp_count = 0, k;
+		int rx = 0, tx = 0;
 
 		if (!bnapi)
 			continue;
 
 		cpr = &bnapi->cp_ring;
+		cpr->bnapi = bnapi;
 		ring = &cpr->cp_ring_struct;
 
-		rc = bnxt_alloc_ring(bp, ring);
+		rc = bnxt_alloc_ring(bp, &ring->ring_mem);
 		if (rc)
 			return rc;
+
+		ring->map_idx = ulp_msix + i;
+
+		if (!(bp->flags & BNXT_FLAG_CHIP_P5_PLUS))
+			continue;
+
+		if (i < bp->rx_nr_rings) {
+			cp_count++;
+			rx = 1;
+		}
+		if (i < bp->tx_nr_rings_xdp) {
+			cp_count++;
+			tx = 1;
+		} else if ((sh && i < bp->tx_nr_rings) ||
+			 (!sh && i >= bp->rx_nr_rings)) {
+			cp_count += tcs;
+			tx = 1;
+		}
+
+		cpr->cp_ring_arr = kcalloc(cp_count, sizeof(*cpr),
+					   GFP_KERNEL);
+		if (!cpr->cp_ring_arr)
+			return -ENOMEM;
+		cpr->cp_ring_count = cp_count;
+
+		for (k = 0; k < cp_count; k++) {
+			cpr2 = &cpr->cp_ring_arr[k];
+			rc = bnxt_alloc_cp_sub_ring(bp, cpr2);
+			if (rc)
+				return rc;
+			cpr2->bnapi = bnapi;
+			cpr2->sw_stats = cpr->sw_stats;
+			cpr2->cp_idx = k;
+			if (!k && rx) {
+				bp->rx_ring[i].rx_cpr = cpr2;
+				cpr2->cp_ring_type = BNXT_NQ_HDL_TYPE_RX;
+			} else {
+				int n, tc = k - rx;
+
+				n = BNXT_TC_TO_RING_BASE(bp, tc) + j;
+				bp->tx_ring[n].tx_cpr = cpr2;
+				cpr2->cp_ring_type = BNXT_NQ_HDL_TYPE_TX;
+			}
+		}
+		if (tx)
+			j++;
 	}
 	return 0;
 }
 
-static void bnxt_init_ring_struct(struct bnxt *bp)
+static void bnxt_init_rx_ring_struct(struct bnxt *bp,
+				     struct bnxt_rx_ring_info *rxr)
+{
+	struct bnxt_ring_mem_info *rmem;
+	struct bnxt_ring_struct *ring;
+
+	ring = &rxr->rx_ring_struct;
+	rmem = &ring->ring_mem;
+	rmem->nr_pages = bp->rx_nr_pages;
+	rmem->page_size = HW_RXBD_RING_SIZE;
+	rmem->pg_arr = (void **)rxr->rx_desc_ring;
+	rmem->dma_arr = rxr->rx_desc_mapping;
+	rmem->vmem_size = SW_RXBD_RING_SIZE * bp->rx_nr_pages;
+	rmem->vmem = (void **)&rxr->rx_buf_ring;
+
+	ring = &rxr->rx_agg_ring_struct;
+	rmem = &ring->ring_mem;
+	rmem->nr_pages = bp->rx_agg_nr_pages;
+	rmem->page_size = HW_RXBD_RING_SIZE;
+	rmem->pg_arr = (void **)rxr->rx_agg_desc_ring;
+	rmem->dma_arr = rxr->rx_agg_desc_mapping;
+	rmem->vmem_size = SW_RXBD_AGG_RING_SIZE * bp->rx_agg_nr_pages;
+	rmem->vmem = (void **)&rxr->rx_agg_ring;
+}
+
+static void bnxt_reset_rx_ring_struct(struct bnxt *bp,
+				      struct bnxt_rx_ring_info *rxr)
 {
+	struct bnxt_ring_mem_info *rmem;
+	struct bnxt_ring_struct *ring;
 	int i;
 
+	rxr->page_pool->p.napi = NULL;
+	rxr->page_pool = NULL;
+	rxr->head_pool->p.napi = NULL;
+	rxr->head_pool = NULL;
+	memset(&rxr->xdp_rxq, 0, sizeof(struct xdp_rxq_info));
+
+	ring = &rxr->rx_ring_struct;
+	rmem = &ring->ring_mem;
+	rmem->pg_tbl = NULL;
+	rmem->pg_tbl_map = 0;
+	for (i = 0; i < rmem->nr_pages; i++) {
+		rmem->pg_arr[i] = NULL;
+		rmem->dma_arr[i] = 0;
+	}
+	*rmem->vmem = NULL;
+
+	ring = &rxr->rx_agg_ring_struct;
+	rmem = &ring->ring_mem;
+	rmem->pg_tbl = NULL;
+	rmem->pg_tbl_map = 0;
+	for (i = 0; i < rmem->nr_pages; i++) {
+		rmem->pg_arr[i] = NULL;
+		rmem->dma_arr[i] = 0;
+	}
+	*rmem->vmem = NULL;
+}
+
+static void bnxt_init_ring_struct(struct bnxt *bp)
+{
+	int i, j;
+
 	for (i = 0; i < bp->cp_nr_rings; i++) {
 		struct bnxt_napi *bnapi = bp->bnapi[i];
+		struct bnxt_ring_mem_info *rmem;
 		struct bnxt_cp_ring_info *cpr;
 		struct bnxt_rx_ring_info *rxr;
 		struct bnxt_tx_ring_info *txr;
@@ -2322,44 +4308,46 @@ static void bnxt_init_ring_struct(struct bnxt *bp)
 
 		cpr = &bnapi->cp_ring;
 		ring = &cpr->cp_ring_struct;
-		ring->nr_pages = bp->cp_nr_pages;
-		ring->page_size = HW_CMPD_RING_SIZE;
-		ring->pg_arr = (void **)cpr->cp_desc_ring;
-		ring->dma_arr = cpr->cp_desc_mapping;
-		ring->vmem_size = 0;
+		rmem = &ring->ring_mem;
+		rmem->nr_pages = bp->cp_nr_pages;
+		rmem->page_size = HW_CMPD_RING_SIZE;
+		rmem->pg_arr = (void **)cpr->cp_desc_ring;
+		rmem->dma_arr = cpr->cp_desc_mapping;
+		rmem->vmem_size = 0;
 
 		rxr = bnapi->rx_ring;
 		if (!rxr)
 			goto skip_rx;
 
 		ring = &rxr->rx_ring_struct;
-		ring->nr_pages = bp->rx_nr_pages;
-		ring->page_size = HW_RXBD_RING_SIZE;
-		ring->pg_arr = (void **)rxr->rx_desc_ring;
-		ring->dma_arr = rxr->rx_desc_mapping;
-		ring->vmem_size = SW_RXBD_RING_SIZE * bp->rx_nr_pages;
-		ring->vmem = (void **)&rxr->rx_buf_ring;
+		rmem = &ring->ring_mem;
+		rmem->nr_pages = bp->rx_nr_pages;
+		rmem->page_size = HW_RXBD_RING_SIZE;
+		rmem->pg_arr = (void **)rxr->rx_desc_ring;
+		rmem->dma_arr = rxr->rx_desc_mapping;
+		rmem->vmem_size = SW_RXBD_RING_SIZE * bp->rx_nr_pages;
+		rmem->vmem = (void **)&rxr->rx_buf_ring;
 
 		ring = &rxr->rx_agg_ring_struct;
-		ring->nr_pages = bp->rx_agg_nr_pages;
-		ring->page_size = HW_RXBD_RING_SIZE;
-		ring->pg_arr = (void **)rxr->rx_agg_desc_ring;
-		ring->dma_arr = rxr->rx_agg_desc_mapping;
-		ring->vmem_size = SW_RXBD_AGG_RING_SIZE * bp->rx_agg_nr_pages;
-		ring->vmem = (void **)&rxr->rx_agg_ring;
+		rmem = &ring->ring_mem;
+		rmem->nr_pages = bp->rx_agg_nr_pages;
+		rmem->page_size = HW_RXBD_RING_SIZE;
+		rmem->pg_arr = (void **)rxr->rx_agg_desc_ring;
+		rmem->dma_arr = rxr->rx_agg_desc_mapping;
+		rmem->vmem_size = SW_RXBD_AGG_RING_SIZE * bp->rx_agg_nr_pages;
+		rmem->vmem = (void **)&rxr->rx_agg_ring;
 
 skip_rx:
-		txr = bnapi->tx_ring;
-		if (!txr)
-			continue;
-
-		ring = &txr->tx_ring_struct;
-		ring->nr_pages = bp->tx_nr_pages;
-		ring->page_size = HW_RXBD_RING_SIZE;
-		ring->pg_arr = (void **)txr->tx_desc_ring;
-		ring->dma_arr = txr->tx_desc_mapping;
-		ring->vmem_size = SW_TXBD_RING_SIZE * bp->tx_nr_pages;
-		ring->vmem = (void **)&txr->tx_buf_ring;
+		bnxt_for_each_napi_tx(j, bnapi, txr) {
+			ring = &txr->tx_ring_struct;
+			rmem = &ring->ring_mem;
+			rmem->nr_pages = bp->tx_nr_pages;
+			rmem->page_size = HW_TXBD_RING_SIZE;
+			rmem->pg_arr = (void **)txr->tx_desc_ring;
+			rmem->dma_arr = txr->tx_desc_mapping;
+			rmem->vmem_size = SW_TXBD_RING_SIZE * bp->tx_nr_pages;
+			rmem->vmem = (void **)&txr->tx_buf_ring;
+		}
 	}
 }
 
@@ -2369,8 +4357,8 @@ static void bnxt_init_rxbd_pages(struct bnxt_ring_struct *ring, u32 type)
 	u32 prod;
 	struct rx_bd **rx_buf_ring;
 
-	rx_buf_ring = (struct rx_bd **)ring->pg_arr;
-	for (i = 0, prod = 0; i < ring->nr_pages; i++) {
+	rx_buf_ring = (struct rx_bd **)ring->ring_mem.pg_arr;
+	for (i = 0, prod = 0; i < ring->ring_mem.nr_pages; i++) {
 		int j;
 		struct rx_bd *rxbd;
 
@@ -2385,100 +4373,159 @@ static void bnxt_init_rxbd_pages(struct bnxt_ring_struct *ring, u32 type)
 	}
 }
 
-static int bnxt_init_one_rx_ring(struct bnxt *bp, int ring_nr)
+static void bnxt_alloc_one_rx_ring_skb(struct bnxt *bp,
+				       struct bnxt_rx_ring_info *rxr,
+				       int ring_nr)
 {
-	struct net_device *dev = bp->dev;
-	struct bnxt_rx_ring_info *rxr;
-	struct bnxt_ring_struct *ring;
-	u32 prod, type;
+	u32 prod;
 	int i;
 
-	type = (bp->rx_buf_use_size << RX_BD_LEN_SHIFT) |
-		RX_BD_TYPE_RX_PACKET_BD | RX_BD_FLAGS_EOP;
-
-	if (NET_IP_ALIGN == 2)
-		type |= RX_BD_FLAGS_SOP;
-
-	rxr = &bp->rx_ring[ring_nr];
-	ring = &rxr->rx_ring_struct;
-	bnxt_init_rxbd_pages(ring, type);
-
-	if (BNXT_RX_PAGE_MODE(bp) && bp->xdp_prog) {
-		rxr->xdp_prog = bpf_prog_add(bp->xdp_prog, 1);
-		if (IS_ERR(rxr->xdp_prog)) {
-			int rc = PTR_ERR(rxr->xdp_prog);
-
-			rxr->xdp_prog = NULL;
-			return rc;
-		}
-	}
 	prod = rxr->rx_prod;
 	for (i = 0; i < bp->rx_ring_size; i++) {
-		if (bnxt_alloc_rx_data(bp, rxr, prod, GFP_KERNEL) != 0) {
-			netdev_warn(dev, "init'ed rx ring %d with %d/%d skbs only\n",
+		if (bnxt_alloc_rx_data(bp, rxr, prod, GFP_KERNEL)) {
+			netdev_warn(bp->dev, "init'ed rx ring %d with %d/%d skbs only\n",
 				    ring_nr, i, bp->rx_ring_size);
 			break;
 		}
 		prod = NEXT_RX(prod);
 	}
 	rxr->rx_prod = prod;
-	ring->fw_ring_id = INVALID_HW_RING_ID;
-
-	ring = &rxr->rx_agg_ring_struct;
-	ring->fw_ring_id = INVALID_HW_RING_ID;
-
-	if (!(bp->flags & BNXT_FLAG_AGG_RINGS))
-		return 0;
-
-	type = ((u32)BNXT_RX_PAGE_SIZE << RX_BD_LEN_SHIFT) |
-		RX_BD_TYPE_RX_AGG_BD | RX_BD_FLAGS_SOP;
+}
 
-	bnxt_init_rxbd_pages(ring, type);
+static void bnxt_alloc_one_rx_ring_netmem(struct bnxt *bp,
+					  struct bnxt_rx_ring_info *rxr,
+					  int ring_nr)
+{
+	u32 prod;
+	int i;
 
 	prod = rxr->rx_agg_prod;
 	for (i = 0; i < bp->rx_agg_ring_size; i++) {
-		if (bnxt_alloc_rx_page(bp, rxr, prod, GFP_KERNEL) != 0) {
-			netdev_warn(dev, "init'ed rx ring %d with %d/%d pages only\n",
-				    ring_nr, i, bp->rx_ring_size);
+		if (bnxt_alloc_rx_netmem(bp, rxr, prod, GFP_KERNEL)) {
+			netdev_warn(bp->dev, "init'ed rx ring %d with %d/%d pages only\n",
+				    ring_nr, i, bp->rx_agg_ring_size);
 			break;
 		}
 		prod = NEXT_RX_AGG(prod);
 	}
 	rxr->rx_agg_prod = prod;
+}
 
-	if (bp->flags & BNXT_FLAG_TPA) {
-		if (rxr->rx_tpa) {
-			u8 *data;
-			dma_addr_t mapping;
-
-			for (i = 0; i < MAX_TPA; i++) {
-				data = __bnxt_alloc_rx_data(bp, &mapping,
-							    GFP_KERNEL);
-				if (!data)
-					return -ENOMEM;
+static int bnxt_alloc_one_tpa_info_data(struct bnxt *bp,
+					struct bnxt_rx_ring_info *rxr)
+{
+	dma_addr_t mapping;
+	u8 *data;
+	int i;
 
-				rxr->rx_tpa[i].data = data;
-				rxr->rx_tpa[i].data_ptr = data + bp->rx_offset;
-				rxr->rx_tpa[i].mapping = mapping;
-			}
-		} else {
-			netdev_err(bp->dev, "No resource allocated for LRO/GRO\n");
+	for (i = 0; i < bp->max_tpa; i++) {
+		data = __bnxt_alloc_rx_frag(bp, &mapping, rxr,
+					    GFP_KERNEL);
+		if (!data)
 			return -ENOMEM;
-		}
+
+		rxr->rx_tpa[i].data = data;
+		rxr->rx_tpa[i].data_ptr = data + bp->rx_offset;
+		rxr->rx_tpa[i].mapping = mapping;
 	}
 
 	return 0;
 }
 
+static int bnxt_alloc_one_rx_ring(struct bnxt *bp, int ring_nr)
+{
+	struct bnxt_rx_ring_info *rxr = &bp->rx_ring[ring_nr];
+	int rc;
+
+	bnxt_alloc_one_rx_ring_skb(bp, rxr, ring_nr);
+
+	if (!(bp->flags & BNXT_FLAG_AGG_RINGS))
+		return 0;
+
+	bnxt_alloc_one_rx_ring_netmem(bp, rxr, ring_nr);
+
+	if (rxr->rx_tpa) {
+		rc = bnxt_alloc_one_tpa_info_data(bp, rxr);
+		if (rc)
+			return rc;
+	}
+	return 0;
+}
+
+static void bnxt_init_one_rx_ring_rxbd(struct bnxt *bp,
+				       struct bnxt_rx_ring_info *rxr)
+{
+	struct bnxt_ring_struct *ring;
+	u32 type;
+
+	type = (bp->rx_buf_use_size << RX_BD_LEN_SHIFT) |
+		RX_BD_TYPE_RX_PACKET_BD | RX_BD_FLAGS_EOP;
+
+	if (NET_IP_ALIGN == 2)
+		type |= RX_BD_FLAGS_SOP;
+
+	ring = &rxr->rx_ring_struct;
+	bnxt_init_rxbd_pages(ring, type);
+	ring->fw_ring_id = INVALID_HW_RING_ID;
+}
+
+static void bnxt_init_one_rx_agg_ring_rxbd(struct bnxt *bp,
+					   struct bnxt_rx_ring_info *rxr)
+{
+	struct bnxt_ring_struct *ring;
+	u32 type;
+
+	ring = &rxr->rx_agg_ring_struct;
+	ring->fw_ring_id = INVALID_HW_RING_ID;
+	if ((bp->flags & BNXT_FLAG_AGG_RINGS)) {
+		type = ((u32)BNXT_RX_PAGE_SIZE << RX_BD_LEN_SHIFT) |
+			RX_BD_TYPE_RX_AGG_BD | RX_BD_FLAGS_SOP;
+
+		bnxt_init_rxbd_pages(ring, type);
+	}
+}
+
+static int bnxt_init_one_rx_ring(struct bnxt *bp, int ring_nr)
+{
+	struct bnxt_rx_ring_info *rxr;
+
+	rxr = &bp->rx_ring[ring_nr];
+	bnxt_init_one_rx_ring_rxbd(bp, rxr);
+
+	netif_queue_set_napi(bp->dev, ring_nr, NETDEV_QUEUE_TYPE_RX,
+			     &rxr->bnapi->napi);
+
+	if (BNXT_RX_PAGE_MODE(bp) && bp->xdp_prog) {
+		bpf_prog_add(bp->xdp_prog, 1);
+		rxr->xdp_prog = bp->xdp_prog;
+	}
+
+	bnxt_init_one_rx_agg_ring_rxbd(bp, rxr);
+
+	return bnxt_alloc_one_rx_ring(bp, ring_nr);
+}
+
 static void bnxt_init_cp_rings(struct bnxt *bp)
 {
-	int i;
+	int i, j;
 
 	for (i = 0; i < bp->cp_nr_rings; i++) {
 		struct bnxt_cp_ring_info *cpr = &bp->bnapi[i]->cp_ring;
 		struct bnxt_ring_struct *ring = &cpr->cp_ring_struct;
 
 		ring->fw_ring_id = INVALID_HW_RING_ID;
+		cpr->rx_ring_coal.coal_ticks = bp->rx_coal.coal_ticks;
+		cpr->rx_ring_coal.coal_bufs = bp->rx_coal.coal_bufs;
+		if (!cpr->cp_ring_arr)
+			continue;
+		for (j = 0; j < cpr->cp_ring_count; j++) {
+			struct bnxt_cp_ring_info *cpr2 = &cpr->cp_ring_arr[j];
+
+			ring = &cpr2->cp_ring_struct;
+			ring->fw_ring_id = INVALID_HW_RING_ID;
+			cpr2->rx_ring_coal.coal_ticks = bp->rx_coal.coal_ticks;
+			cpr2->rx_ring_coal.coal_bufs = bp->rx_coal.coal_bufs;
+		}
 	}
 }
 
@@ -2508,13 +4555,18 @@ static int bnxt_init_tx_rings(struct bnxt *bp)
 	u16 i;
 
 	bp->tx_wake_thresh = max_t(int, bp->tx_ring_size / 2,
-				   MAX_SKB_FRAGS + 1);
+				   BNXT_MIN_TX_DESC_CNT);
 
 	for (i = 0; i < bp->tx_nr_rings; i++) {
 		struct bnxt_tx_ring_info *txr = &bp->tx_ring[i];
 		struct bnxt_ring_struct *ring = &txr->tx_ring_struct;
 
 		ring->fw_ring_id = INVALID_HW_RING_ID;
+
+		if (i >= bp->tx_nr_rings_xdp)
+			netif_queue_set_napi(bp->dev, i - bp->tx_nr_rings_xdp,
+					     NETDEV_QUEUE_TYPE_TX,
+					     &txr->bnapi->napi);
 	}
 
 	return 0;
@@ -2560,8 +4612,12 @@ static int bnxt_alloc_vnics(struct bnxt *bp)
 	int num_vnics = 1;
 
 #ifdef CONFIG_RFS_ACCEL
-	if (bp->flags & BNXT_FLAG_RFS)
-		num_vnics += bp->rx_nr_rings;
+	if (bp->flags & BNXT_FLAG_RFS) {
+		if (BNXT_SUPPORTS_NTUPLE_VNIC(bp))
+			num_vnics++;
+		else if (!(bp->flags & BNXT_FLAG_CHIP_P5_PLUS))
+			num_vnics += bp->rx_nr_rings;
+	}
 #endif
 
 	if (BNXT_CHIP_TYPE_NITRO_A0(bp))
@@ -2578,24 +4634,50 @@ static int bnxt_alloc_vnics(struct bnxt *bp)
 
 static void bnxt_init_vnics(struct bnxt *bp)
 {
+	struct bnxt_vnic_info *vnic0 = &bp->vnic_info[BNXT_VNIC_DEFAULT];
 	int i;
 
 	for (i = 0; i < bp->nr_vnics; i++) {
 		struct bnxt_vnic_info *vnic = &bp->vnic_info[i];
+		int j;
 
 		vnic->fw_vnic_id = INVALID_HW_RING_ID;
-		vnic->fw_rss_cos_lb_ctx[0] = INVALID_HW_RING_ID;
-		vnic->fw_rss_cos_lb_ctx[1] = INVALID_HW_RING_ID;
+		vnic->vnic_id = i;
+		for (j = 0; j < BNXT_MAX_CTX_PER_VNIC; j++)
+			vnic->fw_rss_cos_lb_ctx[j] = INVALID_HW_RING_ID;
+
 		vnic->fw_l2_ctx_id = INVALID_HW_RING_ID;
 
 		if (bp->vnic_info[i].rss_hash_key) {
-			if (i == 0)
-				prandom_bytes(vnic->rss_hash_key,
-					      HW_HASH_KEY_SIZE);
-			else
-				memcpy(vnic->rss_hash_key,
-				       bp->vnic_info[0].rss_hash_key,
+			if (i == BNXT_VNIC_DEFAULT) {
+				u8 *key = (void *)vnic->rss_hash_key;
+				int k;
+
+				if (!bp->rss_hash_key_valid &&
+				    !bp->rss_hash_key_updated) {
+					get_random_bytes(bp->rss_hash_key,
+							 HW_HASH_KEY_SIZE);
+					bp->rss_hash_key_updated = true;
+				}
+
+				memcpy(vnic->rss_hash_key, bp->rss_hash_key,
+				       HW_HASH_KEY_SIZE);
+
+				if (!bp->rss_hash_key_updated)
+					continue;
+
+				bp->rss_hash_key_updated = false;
+				bp->rss_hash_key_valid = true;
+
+				bp->toeplitz_prefix = 0;
+				for (k = 0; k < 8; k++) {
+					bp->toeplitz_prefix <<= 8;
+					bp->toeplitz_prefix |= key[k];
+				}
+			} else {
+				memcpy(vnic->rss_hash_key, vnic0->rss_hash_key,
 				       HW_HASH_KEY_SIZE);
+			}
 		}
 	}
 }
@@ -2624,30 +4706,40 @@ void bnxt_set_tpa_flags(struct bnxt *bp)
 		return;
 	if (bp->dev->features & NETIF_F_LRO)
 		bp->flags |= BNXT_FLAG_LRO;
-	if (bp->dev->features & NETIF_F_GRO)
+	else if (bp->dev->features & NETIF_F_GRO_HW)
 		bp->flags |= BNXT_FLAG_GRO;
 }
 
+static void bnxt_init_ring_params(struct bnxt *bp)
+{
+	unsigned int rx_size;
+
+	bp->rx_copybreak = BNXT_DEFAULT_RX_COPYBREAK;
+	/* Try to fit 4 chunks into a 4k page */
+	rx_size = SZ_1K -
+		NET_SKB_PAD - SKB_DATA_ALIGN(sizeof(struct skb_shared_info));
+	bp->dev->cfg->hds_thresh = max(BNXT_DEFAULT_RX_COPYBREAK, rx_size);
+}
+
 /* bp->rx_ring_size, bp->tx_ring_size, dev->mtu, BNXT_FLAG_{G|L}RO flags must
  * be set on entry.
  */
 void bnxt_set_ring_params(struct bnxt *bp)
 {
-	u32 ring_size, rx_size, rx_space;
+	u32 ring_size, rx_size, rx_space, max_rx_cmpl;
 	u32 agg_factor = 0, agg_ring_size = 0;
 
 	/* 8 for CRC and VLAN */
 	rx_size = SKB_DATA_ALIGN(bp->dev->mtu + ETH_HLEN + NET_IP_ALIGN + 8);
 
-	rx_space = rx_size + NET_SKB_PAD +
+	rx_space = rx_size + ALIGN(max(NET_SKB_PAD, XDP_PACKET_HEADROOM), 8) +
 		SKB_DATA_ALIGN(sizeof(struct skb_shared_info));
 
-	bp->rx_copy_thresh = BNXT_RX_COPY_THRESH;
 	ring_size = bp->rx_ring_size;
 	bp->rx_agg_ring_size = 0;
 	bp->rx_agg_nr_pages = 0;
 
-	if (bp->flags & BNXT_FLAG_TPA)
+	if (bp->flags & BNXT_FLAG_TPA || bp->flags & BNXT_FLAG_HDS)
 		agg_factor = min_t(u32, 4, 65536 / BNXT_RX_PAGE_SIZE);
 
 	bp->flags &= ~BNXT_FLAG_JUMBO;
@@ -2659,9 +4751,15 @@ void bnxt_set_ring_params(struct bnxt *bp)
 		if (jumbo_factor > agg_factor)
 			agg_factor = jumbo_factor;
 	}
-	agg_ring_size = ring_size * agg_factor;
+	if (agg_factor) {
+		if (ring_size > BNXT_MAX_RX_DESC_CNT_JUM_ENA) {
+			ring_size = BNXT_MAX_RX_DESC_CNT_JUM_ENA;
+			netdev_warn(bp->dev, "RX ring size reduced from %d to %d because the jumbo ring is now enabled\n",
+				    bp->rx_ring_size, ring_size);
+			bp->rx_ring_size = ring_size;
+		}
+		agg_ring_size = ring_size * agg_factor;
 
-	if (agg_ring_size) {
 		bp->rx_agg_nr_pages = bnxt_calc_nr_ring_pages(agg_ring_size,
 							RX_DESC_CNT);
 		if (bp->rx_agg_nr_pages > MAX_RX_AGG_PAGES) {
@@ -2674,9 +4772,20 @@ void bnxt_set_ring_params(struct bnxt *bp)
 		}
 		bp->rx_agg_ring_size = agg_ring_size;
 		bp->rx_agg_ring_mask = (bp->rx_agg_nr_pages * RX_DESC_CNT) - 1;
-		rx_size = SKB_DATA_ALIGN(BNXT_RX_COPY_THRESH + NET_IP_ALIGN);
-		rx_space = rx_size + NET_SKB_PAD +
-			SKB_DATA_ALIGN(sizeof(struct skb_shared_info));
+
+		if (BNXT_RX_PAGE_MODE(bp)) {
+			rx_space = PAGE_SIZE;
+			rx_size = PAGE_SIZE -
+				  ALIGN(max(NET_SKB_PAD, XDP_PACKET_HEADROOM), 8) -
+				  SKB_DATA_ALIGN(sizeof(struct skb_shared_info));
+		} else {
+			rx_size = max3(BNXT_DEFAULT_RX_COPYBREAK,
+				       bp->rx_copybreak,
+				       bp->dev->cfg_pending->hds_thresh);
+			rx_size = SKB_DATA_ALIGN(rx_size + NET_IP_ALIGN);
+			rx_space = rx_size + NET_SKB_PAD +
+				SKB_DATA_ALIGN(sizeof(struct skb_shared_info));
+		}
 	}
 
 	bp->rx_buf_use_size = rx_size;
@@ -2689,7 +4798,15 @@ void bnxt_set_ring_params(struct bnxt *bp)
 	bp->tx_nr_pages = bnxt_calc_nr_ring_pages(ring_size, TX_DESC_CNT);
 	bp->tx_ring_mask = (bp->tx_nr_pages * TX_DESC_CNT) - 1;
 
-	ring_size = bp->rx_ring_size * (2 + agg_factor) + bp->tx_ring_size;
+	max_rx_cmpl = bp->rx_ring_size;
+	/* MAX TPA needs to be added because TPA_START completions are
+	 * immediately recycled, so the TPA completions are not bound by
+	 * the RX ring size.
+	 */
+	if (bp->flags & BNXT_FLAG_TPA)
+		max_rx_cmpl += bp->max_tpa;
+	/* RX and TPA completions are 32-byte, all others are 16-byte */
+	ring_size = max_rx_cmpl * 2 + agg_ring_size + bp->tx_ring_size;
 	bp->cp_ring_size = ring_size;
 
 	bp->cp_nr_pages = bnxt_calc_nr_ring_pages(ring_size, CP_DESC_CNT);
@@ -2703,25 +4820,54 @@ void bnxt_set_ring_params(struct bnxt *bp)
 	bp->cp_ring_mask = bp->cp_bit - 1;
 }
 
-int bnxt_set_rx_skb_mode(struct bnxt *bp, bool page_mode)
+/* Changing allocation mode of RX rings.
+ * TODO: Update when extending xdp_rxq_info to support allocation modes.
+ */
+static void __bnxt_set_rx_skb_mode(struct bnxt *bp, bool page_mode)
 {
+	struct net_device *dev = bp->dev;
+
 	if (page_mode) {
-		if (bp->dev->mtu > BNXT_MAX_PAGE_MODE_MTU)
-			return -EOPNOTSUPP;
-		bp->dev->max_mtu = BNXT_MAX_PAGE_MODE_MTU;
-		bp->flags &= ~BNXT_FLAG_AGG_RINGS;
-		bp->flags |= BNXT_FLAG_NO_AGG_RINGS | BNXT_FLAG_RX_PAGE_MODE;
-		bp->dev->hw_features &= ~NETIF_F_LRO;
-		bp->dev->features &= ~NETIF_F_LRO;
+		bp->flags &= ~(BNXT_FLAG_AGG_RINGS | BNXT_FLAG_NO_AGG_RINGS);
+		bp->flags |= BNXT_FLAG_RX_PAGE_MODE;
+
+		if (bp->xdp_prog->aux->xdp_has_frags)
+			dev->max_mtu = min_t(u16, bp->max_mtu, BNXT_MAX_MTU);
+		else
+			dev->max_mtu =
+				min_t(u16, bp->max_mtu, BNXT_MAX_PAGE_MODE_MTU);
+		if (dev->mtu > BNXT_MAX_PAGE_MODE_MTU) {
+			bp->flags |= BNXT_FLAG_JUMBO;
+			bp->rx_skb_func = bnxt_rx_multi_page_skb;
+		} else {
+			bp->flags |= BNXT_FLAG_NO_AGG_RINGS;
+			bp->rx_skb_func = bnxt_rx_page_skb;
+		}
 		bp->rx_dir = DMA_BIDIRECTIONAL;
-		bp->rx_skb_func = bnxt_rx_page_skb;
 	} else {
-		bp->dev->max_mtu = BNXT_MAX_MTU;
+		dev->max_mtu = bp->max_mtu;
 		bp->flags &= ~BNXT_FLAG_RX_PAGE_MODE;
 		bp->rx_dir = DMA_FROM_DEVICE;
 		bp->rx_skb_func = bnxt_rx_skb;
 	}
-	return 0;
+}
+
+void bnxt_set_rx_skb_mode(struct bnxt *bp, bool page_mode)
+{
+	__bnxt_set_rx_skb_mode(bp, page_mode);
+
+	if (!page_mode) {
+		int rx, tx;
+
+		bnxt_get_max_rings(bp, &rx, &tx, true);
+		if (rx > 1) {
+			bp->flags &= ~BNXT_FLAG_NO_AGG_RINGS;
+			bp->dev->hw_features |= NETIF_F_LRO;
+		}
+	}
+
+	/* Update LRO and GRO_HW availability */
+	netdev_update_features(bp->dev);
 }
 
 static void bnxt_free_vnic_attributes(struct bnxt *bp)
@@ -2749,7 +4895,7 @@ static void bnxt_free_vnic_attributes(struct bnxt *bp)
 		}
 
 		if (vnic->rss_table) {
-			dma_free_coherent(&pdev->dev, PAGE_SIZE,
+			dma_free_coherent(&pdev->dev, vnic->rss_table_size,
 					  vnic->rss_table,
 					  vnic->rss_table_dma_addr);
 			vnic->rss_table = NULL;
@@ -2795,6 +4941,9 @@ static int bnxt_alloc_vnic_attributes(struct bnxt *bp)
 			}
 		}
 
+		if (bp->flags & BNXT_FLAG_CHIP_P5_PLUS)
+			goto vnic_skip_grps;
+
 		if (vnic->flags & BNXT_VNIC_RSS_FLAG)
 			max_rings = bp->rx_nr_rings;
 		else
@@ -2805,13 +4954,19 @@ static int bnxt_alloc_vnic_attributes(struct bnxt *bp)
 			rc = -ENOMEM;
 			goto out;
 		}
-
-		if ((bp->flags & BNXT_FLAG_NEW_RSS_CAP) &&
+vnic_skip_grps:
+		if ((bp->rss_cap & BNXT_RSS_CAP_NEW_RSS_CAP) &&
 		    !(vnic->flags & BNXT_VNIC_RSS_FLAG))
 			continue;
 
 		/* Allocate rss table and hash key */
-		vnic->rss_table = dma_alloc_coherent(&pdev->dev, PAGE_SIZE,
+		size = L1_CACHE_ALIGN(HW_HASH_INDEX_SIZE * sizeof(u16));
+		if (bp->flags & BNXT_FLAG_CHIP_P5_PLUS)
+			size = L1_CACHE_ALIGN(BNXT_MAX_RSS_TABLE_SIZE_P5);
+
+		vnic->rss_table_size = size + HW_HASH_KEY_SIZE;
+		vnic->rss_table = dma_alloc_coherent(&pdev->dev,
+						     vnic->rss_table_size,
 						     &vnic->rss_table_dma_addr,
 						     GFP_KERNEL);
 		if (!vnic->rss_table) {
@@ -2819,8 +4974,6 @@ static int bnxt_alloc_vnic_attributes(struct bnxt *bp)
 			goto out;
 		}
 
-		size = L1_CACHE_ALIGN(HW_HASH_INDEX_SIZE * sizeof(u16));
-
 		vnic->rss_hash_key = ((void *)vnic->rss_table) + size;
 		vnic->rss_hash_key_dma_addr = vnic->rss_table_dma_addr + size;
 	}
@@ -2832,139 +4985,284 @@ out:
 
 static void bnxt_free_hwrm_resources(struct bnxt *bp)
 {
-	struct pci_dev *pdev = bp->pdev;
-
-	dma_free_coherent(&pdev->dev, PAGE_SIZE, bp->hwrm_cmd_resp_addr,
-			  bp->hwrm_cmd_resp_dma_addr);
+	struct bnxt_hwrm_wait_token *token;
 
-	bp->hwrm_cmd_resp_addr = NULL;
-	if (bp->hwrm_dbg_resp_addr) {
-		dma_free_coherent(&pdev->dev, HWRM_DBG_REG_BUF_SIZE,
-				  bp->hwrm_dbg_resp_addr,
-				  bp->hwrm_dbg_resp_dma_addr);
+	dma_pool_destroy(bp->hwrm_dma_pool);
+	bp->hwrm_dma_pool = NULL;
 
-		bp->hwrm_dbg_resp_addr = NULL;
-	}
+	rcu_read_lock();
+	hlist_for_each_entry_rcu(token, &bp->hwrm_pending_list, node)
+		WRITE_ONCE(token->state, BNXT_HWRM_CANCELLED);
+	rcu_read_unlock();
 }
 
 static int bnxt_alloc_hwrm_resources(struct bnxt *bp)
 {
-	struct pci_dev *pdev = bp->pdev;
-
-	bp->hwrm_cmd_resp_addr = dma_alloc_coherent(&pdev->dev, PAGE_SIZE,
-						   &bp->hwrm_cmd_resp_dma_addr,
-						   GFP_KERNEL);
-	if (!bp->hwrm_cmd_resp_addr)
+	bp->hwrm_dma_pool = dma_pool_create("bnxt_hwrm", &bp->pdev->dev,
+					    BNXT_HWRM_DMA_SIZE,
+					    BNXT_HWRM_DMA_ALIGN, 0);
+	if (!bp->hwrm_dma_pool)
 		return -ENOMEM;
-	bp->hwrm_dbg_resp_addr = dma_alloc_coherent(&pdev->dev,
-						    HWRM_DBG_REG_BUF_SIZE,
-						    &bp->hwrm_dbg_resp_dma_addr,
-						    GFP_KERNEL);
-	if (!bp->hwrm_dbg_resp_addr)
-		netdev_warn(bp->dev, "fail to alloc debug register dma mem\n");
+
+	INIT_HLIST_HEAD(&bp->hwrm_pending_list);
 
 	return 0;
 }
 
-static void bnxt_free_hwrm_short_cmd_req(struct bnxt *bp)
+static void bnxt_free_stats_mem(struct bnxt *bp, struct bnxt_stats_mem *stats)
 {
-	if (bp->hwrm_short_cmd_req_addr) {
-		struct pci_dev *pdev = bp->pdev;
-
-		dma_free_coherent(&pdev->dev, BNXT_HWRM_MAX_REQ_LEN,
-				  bp->hwrm_short_cmd_req_addr,
-				  bp->hwrm_short_cmd_req_dma_addr);
-		bp->hwrm_short_cmd_req_addr = NULL;
+	kfree(stats->hw_masks);
+	stats->hw_masks = NULL;
+	kfree(stats->sw_stats);
+	stats->sw_stats = NULL;
+	if (stats->hw_stats) {
+		dma_free_coherent(&bp->pdev->dev, stats->len, stats->hw_stats,
+				  stats->hw_stats_map);
+		stats->hw_stats = NULL;
 	}
 }
 
-static int bnxt_alloc_hwrm_short_cmd_req(struct bnxt *bp)
+static int bnxt_alloc_stats_mem(struct bnxt *bp, struct bnxt_stats_mem *stats,
+				bool alloc_masks)
 {
-	struct pci_dev *pdev = bp->pdev;
-
-	bp->hwrm_short_cmd_req_addr =
-		dma_alloc_coherent(&pdev->dev, BNXT_HWRM_MAX_REQ_LEN,
-				   &bp->hwrm_short_cmd_req_dma_addr,
-				   GFP_KERNEL);
-	if (!bp->hwrm_short_cmd_req_addr)
+	stats->hw_stats = dma_alloc_coherent(&bp->pdev->dev, stats->len,
+					     &stats->hw_stats_map, GFP_KERNEL);
+	if (!stats->hw_stats)
 		return -ENOMEM;
 
+	stats->sw_stats = kzalloc(stats->len, GFP_KERNEL);
+	if (!stats->sw_stats)
+		goto stats_mem_err;
+
+	if (alloc_masks) {
+		stats->hw_masks = kzalloc(stats->len, GFP_KERNEL);
+		if (!stats->hw_masks)
+			goto stats_mem_err;
+	}
 	return 0;
+
+stats_mem_err:
+	bnxt_free_stats_mem(bp, stats);
+	return -ENOMEM;
 }
 
-static void bnxt_free_stats(struct bnxt *bp)
+static void bnxt_fill_masks(u64 *mask_arr, u64 mask, int count)
 {
-	u32 size, i;
-	struct pci_dev *pdev = bp->pdev;
+	int i;
+
+	for (i = 0; i < count; i++)
+		mask_arr[i] = mask;
+}
+
+static void bnxt_copy_hw_masks(u64 *mask_arr, __le64 *hw_mask_arr, int count)
+{
+	int i;
+
+	for (i = 0; i < count; i++)
+		mask_arr[i] = le64_to_cpu(hw_mask_arr[i]);
+}
+
+static int bnxt_hwrm_func_qstat_ext(struct bnxt *bp,
+				    struct bnxt_stats_mem *stats)
+{
+	struct hwrm_func_qstats_ext_output *resp;
+	struct hwrm_func_qstats_ext_input *req;
+	__le64 *hw_masks;
+	int rc;
+
+	if (!(bp->fw_cap & BNXT_FW_CAP_EXT_HW_STATS_SUPPORTED) ||
+	    !(bp->flags & BNXT_FLAG_CHIP_P5_PLUS))
+		return -EOPNOTSUPP;
+
+	rc = hwrm_req_init(bp, req, HWRM_FUNC_QSTATS_EXT);
+	if (rc)
+		return rc;
+
+	req->fid = cpu_to_le16(0xffff);
+	req->flags = FUNC_QSTATS_EXT_REQ_FLAGS_COUNTER_MASK;
+
+	resp = hwrm_req_hold(bp, req);
+	rc = hwrm_req_send(bp, req);
+	if (!rc) {
+		hw_masks = &resp->rx_ucast_pkts;
+		bnxt_copy_hw_masks(stats->hw_masks, hw_masks, stats->len / 8);
+	}
+	hwrm_req_drop(bp, req);
+	return rc;
+}
+
+static int bnxt_hwrm_port_qstats(struct bnxt *bp, u8 flags);
+static int bnxt_hwrm_port_qstats_ext(struct bnxt *bp, u8 flags);
+
+static void bnxt_init_stats(struct bnxt *bp)
+{
+	struct bnxt_napi *bnapi = bp->bnapi[0];
+	struct bnxt_cp_ring_info *cpr;
+	struct bnxt_stats_mem *stats;
+	__le64 *rx_stats, *tx_stats;
+	int rc, rx_count, tx_count;
+	u64 *rx_masks, *tx_masks;
+	u64 mask;
+	u8 flags;
+
+	cpr = &bnapi->cp_ring;
+	stats = &cpr->stats;
+	rc = bnxt_hwrm_func_qstat_ext(bp, stats);
+	if (rc) {
+		if (bp->flags & BNXT_FLAG_CHIP_P5_PLUS)
+			mask = (1ULL << 48) - 1;
+		else
+			mask = -1ULL;
+		bnxt_fill_masks(stats->hw_masks, mask, stats->len / 8);
+	}
+	if (bp->flags & BNXT_FLAG_PORT_STATS) {
+		stats = &bp->port_stats;
+		rx_stats = stats->hw_stats;
+		rx_masks = stats->hw_masks;
+		rx_count = sizeof(struct rx_port_stats) / 8;
+		tx_stats = rx_stats + BNXT_TX_PORT_STATS_BYTE_OFFSET / 8;
+		tx_masks = rx_masks + BNXT_TX_PORT_STATS_BYTE_OFFSET / 8;
+		tx_count = sizeof(struct tx_port_stats) / 8;
+
+		flags = PORT_QSTATS_REQ_FLAGS_COUNTER_MASK;
+		rc = bnxt_hwrm_port_qstats(bp, flags);
+		if (rc) {
+			mask = (1ULL << 40) - 1;
+
+			bnxt_fill_masks(rx_masks, mask, rx_count);
+			bnxt_fill_masks(tx_masks, mask, tx_count);
+		} else {
+			bnxt_copy_hw_masks(rx_masks, rx_stats, rx_count);
+			bnxt_copy_hw_masks(tx_masks, tx_stats, tx_count);
+			bnxt_hwrm_port_qstats(bp, 0);
+		}
+	}
+	if (bp->flags & BNXT_FLAG_PORT_STATS_EXT) {
+		stats = &bp->rx_port_stats_ext;
+		rx_stats = stats->hw_stats;
+		rx_masks = stats->hw_masks;
+		rx_count = sizeof(struct rx_port_stats_ext) / 8;
+		stats = &bp->tx_port_stats_ext;
+		tx_stats = stats->hw_stats;
+		tx_masks = stats->hw_masks;
+		tx_count = sizeof(struct tx_port_stats_ext) / 8;
+
+		flags = PORT_QSTATS_EXT_REQ_FLAGS_COUNTER_MASK;
+		rc = bnxt_hwrm_port_qstats_ext(bp, flags);
+		if (rc) {
+			mask = (1ULL << 40) - 1;
 
-	if (bp->hw_rx_port_stats) {
-		dma_free_coherent(&pdev->dev, bp->hw_port_stats_size,
-				  bp->hw_rx_port_stats,
-				  bp->hw_rx_port_stats_map);
-		bp->hw_rx_port_stats = NULL;
-		bp->flags &= ~BNXT_FLAG_PORT_STATS;
+			bnxt_fill_masks(rx_masks, mask, rx_count);
+			if (tx_stats)
+				bnxt_fill_masks(tx_masks, mask, tx_count);
+		} else {
+			bnxt_copy_hw_masks(rx_masks, rx_stats, rx_count);
+			if (tx_stats)
+				bnxt_copy_hw_masks(tx_masks, tx_stats,
+						   tx_count);
+			bnxt_hwrm_port_qstats_ext(bp, 0);
+		}
 	}
+}
+
+static void bnxt_free_port_stats(struct bnxt *bp)
+{
+	bp->flags &= ~BNXT_FLAG_PORT_STATS;
+	bp->flags &= ~BNXT_FLAG_PORT_STATS_EXT;
+
+	bnxt_free_stats_mem(bp, &bp->port_stats);
+	bnxt_free_stats_mem(bp, &bp->rx_port_stats_ext);
+	bnxt_free_stats_mem(bp, &bp->tx_port_stats_ext);
+}
+
+static void bnxt_free_ring_stats(struct bnxt *bp)
+{
+	int i;
 
 	if (!bp->bnapi)
 		return;
 
-	size = sizeof(struct ctx_hw_stats);
-
 	for (i = 0; i < bp->cp_nr_rings; i++) {
 		struct bnxt_napi *bnapi = bp->bnapi[i];
 		struct bnxt_cp_ring_info *cpr = &bnapi->cp_ring;
 
-		if (cpr->hw_stats) {
-			dma_free_coherent(&pdev->dev, size, cpr->hw_stats,
-					  cpr->hw_stats_map);
-			cpr->hw_stats = NULL;
-		}
+		bnxt_free_stats_mem(bp, &cpr->stats);
+
+		kfree(cpr->sw_stats);
+		cpr->sw_stats = NULL;
 	}
 }
 
 static int bnxt_alloc_stats(struct bnxt *bp)
 {
 	u32 size, i;
-	struct pci_dev *pdev = bp->pdev;
+	int rc;
 
-	size = sizeof(struct ctx_hw_stats);
+	size = bp->hw_ring_stats_size;
 
 	for (i = 0; i < bp->cp_nr_rings; i++) {
 		struct bnxt_napi *bnapi = bp->bnapi[i];
 		struct bnxt_cp_ring_info *cpr = &bnapi->cp_ring;
 
-		cpr->hw_stats = dma_alloc_coherent(&pdev->dev, size,
-						   &cpr->hw_stats_map,
-						   GFP_KERNEL);
-		if (!cpr->hw_stats)
+		cpr->sw_stats = kzalloc(sizeof(*cpr->sw_stats), GFP_KERNEL);
+		if (!cpr->sw_stats)
 			return -ENOMEM;
 
+		cpr->stats.len = size;
+		rc = bnxt_alloc_stats_mem(bp, &cpr->stats, !i);
+		if (rc)
+			return rc;
+
 		cpr->hw_stats_ctx_id = INVALID_STATS_CTX_ID;
 	}
 
-	if (BNXT_PF(bp) && bp->chip_num != CHIP_NUM_58700) {
-		bp->hw_port_stats_size = sizeof(struct rx_port_stats) +
-					 sizeof(struct tx_port_stats) + 1024;
+	if (BNXT_VF(bp) || bp->chip_num == CHIP_NUM_58700)
+		return 0;
 
-		bp->hw_rx_port_stats =
-			dma_alloc_coherent(&pdev->dev, bp->hw_port_stats_size,
-					   &bp->hw_rx_port_stats_map,
-					   GFP_KERNEL);
-		if (!bp->hw_rx_port_stats)
-			return -ENOMEM;
+	if (bp->port_stats.hw_stats)
+		goto alloc_ext_stats;
+
+	bp->port_stats.len = BNXT_PORT_STATS_SIZE;
+	rc = bnxt_alloc_stats_mem(bp, &bp->port_stats, true);
+	if (rc)
+		return rc;
 
-		bp->hw_tx_port_stats = (void *)(bp->hw_rx_port_stats + 1) +
-				       512;
-		bp->hw_tx_port_stats_map = bp->hw_rx_port_stats_map +
-					   sizeof(struct rx_port_stats) + 512;
-		bp->flags |= BNXT_FLAG_PORT_STATS;
+	bp->flags |= BNXT_FLAG_PORT_STATS;
+
+alloc_ext_stats:
+	/* Display extended statistics only if FW supports it */
+	if (bp->hwrm_spec_code < 0x10804 || bp->hwrm_spec_code == 0x10900)
+		if (!(bp->fw_cap & BNXT_FW_CAP_EXT_STATS_SUPPORTED))
+			return 0;
+
+	if (bp->rx_port_stats_ext.hw_stats)
+		goto alloc_tx_ext_stats;
+
+	bp->rx_port_stats_ext.len = sizeof(struct rx_port_stats_ext);
+	rc = bnxt_alloc_stats_mem(bp, &bp->rx_port_stats_ext, true);
+	/* Extended stats are optional */
+	if (rc)
+		return 0;
+
+alloc_tx_ext_stats:
+	if (bp->tx_port_stats_ext.hw_stats)
+		return 0;
+
+	if (bp->hwrm_spec_code >= 0x10902 ||
+	    (bp->fw_cap & BNXT_FW_CAP_EXT_STATS_SUPPORTED)) {
+		bp->tx_port_stats_ext.len = sizeof(struct tx_port_stats_ext);
+		rc = bnxt_alloc_stats_mem(bp, &bp->tx_port_stats_ext, true);
+		/* Extended stats are optional */
+		if (rc)
+			return 0;
 	}
+	bp->flags |= BNXT_FLAG_PORT_STATS_EXT;
 	return 0;
 }
 
 static void bnxt_clear_ring_indices(struct bnxt *bp)
 {
-	int i;
+	int i, j;
 
 	if (!bp->bnapi)
 		return;
@@ -2981,10 +5279,10 @@ static void bnxt_clear_ring_indices(struct bnxt *bp)
 		cpr = &bnapi->cp_ring;
 		cpr->cp_raw_cons = 0;
 
-		txr = bnapi->tx_ring;
-		if (txr) {
+		bnxt_for_each_napi_tx(j, bnapi, txr) {
 			txr->tx_prod = 0;
 			txr->tx_cons = 0;
+			txr->tx_hw_cons = 0;
 		}
 
 		rxr = bnapi->rx_ring;
@@ -2994,16 +5292,56 @@ static void bnxt_clear_ring_indices(struct bnxt *bp)
 			rxr->rx_sw_agg_prod = 0;
 			rxr->rx_next_cons = 0;
 		}
+		bnapi->events = 0;
 	}
 }
 
-static void bnxt_free_ntp_fltrs(struct bnxt *bp, bool irq_reinit)
+void bnxt_insert_usr_fltr(struct bnxt *bp, struct bnxt_filter_base *fltr)
+{
+	u8 type = fltr->type, flags = fltr->flags;
+
+	INIT_LIST_HEAD(&fltr->list);
+	if ((type == BNXT_FLTR_TYPE_L2 && flags & BNXT_ACT_RING_DST) ||
+	    (type == BNXT_FLTR_TYPE_NTUPLE && flags & BNXT_ACT_NO_AGING))
+		list_add_tail(&fltr->list, &bp->usr_fltr_list);
+}
+
+void bnxt_del_one_usr_fltr(struct bnxt *bp, struct bnxt_filter_base *fltr)
+{
+	if (!list_empty(&fltr->list))
+		list_del_init(&fltr->list);
+}
+
+static void bnxt_clear_usr_fltrs(struct bnxt *bp, bool all)
+{
+	struct bnxt_filter_base *usr_fltr, *tmp;
+
+	list_for_each_entry_safe(usr_fltr, tmp, &bp->usr_fltr_list, list) {
+		if (!all && usr_fltr->type == BNXT_FLTR_TYPE_L2)
+			continue;
+		bnxt_del_one_usr_fltr(bp, usr_fltr);
+	}
+}
+
+static void bnxt_del_fltr(struct bnxt *bp, struct bnxt_filter_base *fltr)
+{
+	hlist_del(&fltr->hash);
+	bnxt_del_one_usr_fltr(bp, fltr);
+	if (fltr->flags) {
+		clear_bit(fltr->sw_id, bp->ntp_fltr_bmap);
+		bp->ntp_fltr_count--;
+	}
+	kfree(fltr);
+}
+
+static void bnxt_free_ntp_fltrs(struct bnxt *bp, bool all)
 {
-#ifdef CONFIG_RFS_ACCEL
 	int i;
 
-	/* Under rtnl_lock and all our NAPIs have been disabled.  It's
-	 * safe to delete the hash table.
+	netdev_assert_locked_or_invisible(bp->dev);
+
+	/* Under netdev instance lock and all our NAPIs have been disabled.
+	 * It's safe to delete the hash table.
 	 */
 	for (i = 0; i < BNXT_NTP_FLTR_HASH_SIZE; i++) {
 		struct hlist_head *head;
@@ -3011,42 +5349,67 @@ static void bnxt_free_ntp_fltrs(struct bnxt *bp, bool irq_reinit)
 		struct bnxt_ntuple_filter *fltr;
 
 		head = &bp->ntp_fltr_hash_tbl[i];
-		hlist_for_each_entry_safe(fltr, tmp, head, hash) {
-			hlist_del(&fltr->hash);
-			kfree(fltr);
+		hlist_for_each_entry_safe(fltr, tmp, head, base.hash) {
+			bnxt_del_l2_filter(bp, fltr->l2_fltr);
+			if (!all && ((fltr->base.flags & BNXT_ACT_FUNC_DST) ||
+				     !list_empty(&fltr->base.list)))
+				continue;
+			bnxt_del_fltr(bp, &fltr->base);
 		}
 	}
-	if (irq_reinit) {
-		kfree(bp->ntp_fltr_bmap);
-		bp->ntp_fltr_bmap = NULL;
-	}
+	if (!all)
+		return;
+
+	bitmap_free(bp->ntp_fltr_bmap);
+	bp->ntp_fltr_bmap = NULL;
 	bp->ntp_fltr_count = 0;
-#endif
 }
 
 static int bnxt_alloc_ntp_fltrs(struct bnxt *bp)
 {
-#ifdef CONFIG_RFS_ACCEL
 	int i, rc = 0;
 
-	if (!(bp->flags & BNXT_FLAG_RFS))
+	if (!(bp->flags & BNXT_FLAG_RFS) || bp->ntp_fltr_bmap)
 		return 0;
 
 	for (i = 0; i < BNXT_NTP_FLTR_HASH_SIZE; i++)
 		INIT_HLIST_HEAD(&bp->ntp_fltr_hash_tbl[i]);
 
 	bp->ntp_fltr_count = 0;
-	bp->ntp_fltr_bmap = kcalloc(BITS_TO_LONGS(BNXT_NTP_FLTR_MAX_FLTR),
-				    sizeof(long),
-				    GFP_KERNEL);
+	bp->ntp_fltr_bmap = bitmap_zalloc(bp->max_fltr, GFP_KERNEL);
 
 	if (!bp->ntp_fltr_bmap)
 		rc = -ENOMEM;
 
 	return rc;
-#else
-	return 0;
-#endif
+}
+
+static void bnxt_free_l2_filters(struct bnxt *bp, bool all)
+{
+	int i;
+
+	for (i = 0; i < BNXT_L2_FLTR_HASH_SIZE; i++) {
+		struct hlist_head *head;
+		struct hlist_node *tmp;
+		struct bnxt_l2_filter *fltr;
+
+		head = &bp->l2_fltr_hash_tbl[i];
+		hlist_for_each_entry_safe(fltr, tmp, head, base.hash) {
+			if (!all && ((fltr->base.flags & BNXT_ACT_FUNC_DST) ||
+				     !list_empty(&fltr->base.list)))
+				continue;
+			bnxt_del_fltr(bp, &fltr->base);
+		}
+	}
+}
+
+static void bnxt_init_l2_fltr_tbl(struct bnxt *bp)
+{
+	int i;
+
+	for (i = 0; i < BNXT_L2_FLTR_HASH_SIZE; i++)
+		INIT_HLIST_HEAD(&bp->l2_fltr_hash_tbl[i]);
+	get_random_bytes(&bp->hash_seed, sizeof(bp->hash_seed));
 }
 
 static void bnxt_free_mem(struct bnxt *bp, bool irq_re_init)
@@ -3055,9 +5418,14 @@ static void bnxt_free_mem(struct bnxt *bp, bool irq_re_init)
 	bnxt_free_tx_rings(bp);
 	bnxt_free_rx_rings(bp);
 	bnxt_free_cp_rings(bp);
-	bnxt_free_ntp_fltrs(bp, irq_re_init);
+	bnxt_free_all_cp_arrays(bp);
+	bnxt_free_ntp_fltrs(bp, false);
+	bnxt_free_l2_filters(bp, false);
 	if (irq_re_init) {
-		bnxt_free_stats(bp);
+		bnxt_free_ring_stats(bp);
+		if (!(bp->phy_flags & BNXT_PHY_FL_PORT_STATS_NO_RESET) ||
+		    test_bit(BNXT_STATE_IN_FW_RESET, &bp->state))
+			bnxt_free_port_stats(bp);
 		bnxt_free_ring_grps(bp);
 		bnxt_free_vnics(bp);
 		kfree(bp->tx_ring_map);
@@ -3095,6 +5463,13 @@ static int bnxt_alloc_mem(struct bnxt *bp, bool irq_re_init)
 			bp->bnapi[i] = bnapi;
 			bp->bnapi[i]->index = i;
 			bp->bnapi[i]->bp = bp;
+			if (bp->flags & BNXT_FLAG_CHIP_P5_PLUS) {
+				struct bnxt_cp_ring_info *cpr =
+					&bp->bnapi[i]->cp_ring;
+
+				cpr->cp_ring_struct.ring_mem.flags =
+					BNXT_RMEM_RING_PTE_FLAG;
+			}
 		}
 
 		bp->rx_ring = kcalloc(bp->rx_nr_rings,
@@ -3104,7 +5479,17 @@ static int bnxt_alloc_mem(struct bnxt *bp, bool irq_re_init)
 			return -ENOMEM;
 
 		for (i = 0; i < bp->rx_nr_rings; i++) {
-			bp->rx_ring[i].bnapi = bp->bnapi[i];
+			struct bnxt_rx_ring_info *rxr = &bp->rx_ring[i];
+
+			if (bp->flags & BNXT_FLAG_CHIP_P5_PLUS) {
+				rxr->rx_ring_struct.ring_mem.flags =
+					BNXT_RMEM_RING_PTE_FLAG;
+				rxr->rx_agg_ring_struct.ring_mem.flags =
+					BNXT_RMEM_RING_PTE_FLAG;
+			} else {
+				rxr->rx_cpr =  &bp->bnapi[i]->cp_ring;
+			}
+			rxr->bnapi = bp->bnapi[i];
 			bp->bnapi[i]->rx_ring = &bp->rx_ring[i];
 		}
 
@@ -3125,23 +5510,39 @@ static int bnxt_alloc_mem(struct bnxt *bp, bool irq_re_init)
 		else
 			j = bp->rx_nr_rings;
 
-		for (i = 0; i < bp->tx_nr_rings; i++, j++) {
-			bp->tx_ring[i].bnapi = bp->bnapi[j];
-			bp->bnapi[j]->tx_ring = &bp->tx_ring[i];
+		for (i = 0; i < bp->tx_nr_rings; i++) {
+			struct bnxt_tx_ring_info *txr = &bp->tx_ring[i];
+			struct bnxt_napi *bnapi2;
+
+			if (bp->flags & BNXT_FLAG_CHIP_P5_PLUS)
+				txr->tx_ring_struct.ring_mem.flags =
+					BNXT_RMEM_RING_PTE_FLAG;
 			bp->tx_ring_map[i] = bp->tx_nr_rings_xdp + i;
 			if (i >= bp->tx_nr_rings_xdp) {
-				bp->tx_ring[i].txq_index = i -
-					bp->tx_nr_rings_xdp;
-				bp->bnapi[j]->tx_int = bnxt_tx_int;
+				int k = j + BNXT_RING_TO_TC_OFF(bp, i);
+
+				bnapi2 = bp->bnapi[k];
+				txr->txq_index = i - bp->tx_nr_rings_xdp;
+				txr->tx_napi_idx =
+					BNXT_RING_TO_TC(bp, txr->txq_index);
+				bnapi2->tx_ring[txr->tx_napi_idx] = txr;
+				bnapi2->tx_int = bnxt_tx_int;
 			} else {
-				bp->bnapi[j]->flags |= BNXT_NAPI_FLAG_XDP;
-				bp->bnapi[j]->tx_int = bnxt_tx_int_xdp;
+				bnapi2 = bp->bnapi[j];
+				bnapi2->flags |= BNXT_NAPI_FLAG_XDP;
+				bnapi2->tx_ring[0] = txr;
+				bnapi2->tx_int = bnxt_tx_int_xdp;
+				j++;
 			}
+			txr->bnapi = bnapi2;
+			if (!(bp->flags & BNXT_FLAG_CHIP_P5_PLUS))
+				txr->tx_cpr = &bnapi2->cp_ring;
 		}
 
 		rc = bnxt_alloc_stats(bp);
 		if (rc)
 			goto alloc_mem_err;
+		bnxt_init_stats(bp);
 
 		rc = bnxt_alloc_ntp_fltrs(bp);
 		if (rc)
@@ -3152,6 +5553,10 @@ static int bnxt_alloc_mem(struct bnxt *bp, bool irq_re_init)
 			goto alloc_mem_err;
 	}
 
+	rc = bnxt_alloc_all_cp_arrays(bp);
+	if (rc)
+		goto alloc_mem_err;
+
 	bnxt_init_ring_struct(bp);
 
 	rc = bnxt_alloc_rx_rings(bp);
@@ -3166,8 +5571,13 @@ static int bnxt_alloc_mem(struct bnxt *bp, bool irq_re_init)
 	if (rc)
 		goto alloc_mem_err;
 
-	bp->vnic_info[0].flags |= BNXT_VNIC_RSS_FLAG | BNXT_VNIC_MCAST_FLAG |
-				  BNXT_VNIC_UCAST_FLAG;
+	bp->vnic_info[BNXT_VNIC_DEFAULT].flags |= BNXT_VNIC_RSS_FLAG |
+						  BNXT_VNIC_MCAST_FLAG |
+						  BNXT_VNIC_UCAST_FLAG;
+	if (BNXT_SUPPORTS_NTUPLE_VNIC(bp) && (bp->flags & BNXT_FLAG_RFS))
+		bp->vnic_info[BNXT_VNIC_NTUPLE].flags |=
+			BNXT_VNIC_RSS_FLAG | BNXT_VNIC_NTUPLE_FLAG;
+
 	rc = bnxt_alloc_vnic_attributes(bp);
 	if (rc)
 		goto alloc_mem_err;
@@ -3191,19 +5601,34 @@ static void bnxt_disable_int(struct bnxt *bp)
 		struct bnxt_ring_struct *ring = &cpr->cp_ring_struct;
 
 		if (ring->fw_ring_id != INVALID_HW_RING_ID)
-			BNXT_CP_DB(cpr->cp_doorbell, cpr->cp_raw_cons);
+			bnxt_db_nq(bp, &cpr->cp_db, cpr->cp_raw_cons);
 	}
 }
 
+static int bnxt_cp_num_to_irq_num(struct bnxt *bp, int n)
+{
+	struct bnxt_napi *bnapi = bp->bnapi[n];
+	struct bnxt_cp_ring_info *cpr;
+
+	cpr = &bnapi->cp_ring;
+	return cpr->cp_ring_struct.map_idx;
+}
+
 static void bnxt_disable_int_sync(struct bnxt *bp)
 {
 	int i;
 
+	if (!bp->irq_tbl)
+		return;
+
 	atomic_inc(&bp->intr_sem);
 
 	bnxt_disable_int(bp);
-	for (i = 0; i < bp->cp_nr_rings; i++)
-		synchronize_irq(bp->irq_tbl[i].vector);
+	for (i = 0; i < bp->cp_nr_rings; i++) {
+		int map_idx = bnxt_cp_num_to_irq_num(bp, i);
+
+		synchronize_irq(bp->irq_tbl[map_idx].vector);
+	}
 }
 
 static void bnxt_enable_int(struct bnxt *bp)
@@ -3215,322 +5640,569 @@ static void bnxt_enable_int(struct bnxt *bp)
 		struct bnxt_napi *bnapi = bp->bnapi[i];
 		struct bnxt_cp_ring_info *cpr = &bnapi->cp_ring;
 
-		BNXT_CP_DB_REARM(cpr->cp_doorbell, cpr->cp_raw_cons);
+		bnxt_db_nq_arm(bp, &cpr->cp_db, cpr->cp_raw_cons);
 	}
 }
 
-void bnxt_hwrm_cmd_hdr_init(struct bnxt *bp, void *request, u16 req_type,
-			    u16 cmpl_ring, u16 target_id)
+int bnxt_hwrm_func_drv_rgtr(struct bnxt *bp, unsigned long *bmap, int bmap_size,
+			    bool async_only)
 {
-	struct input *req = request;
+	DECLARE_BITMAP(async_events_bmap, 256);
+	u32 *events = (u32 *)async_events_bmap;
+	struct hwrm_func_drv_rgtr_output *resp;
+	struct hwrm_func_drv_rgtr_input *req;
+	u32 flags;
+	int rc, i;
 
-	req->req_type = cpu_to_le16(req_type);
-	req->cmpl_ring = cpu_to_le16(cmpl_ring);
-	req->target_id = cpu_to_le16(target_id);
-	req->resp_addr = cpu_to_le64(bp->hwrm_cmd_resp_dma_addr);
-}
+	rc = hwrm_req_init(bp, req, HWRM_FUNC_DRV_RGTR);
+	if (rc)
+		return rc;
 
-static int bnxt_hwrm_do_send_msg(struct bnxt *bp, void *msg, u32 msg_len,
-				 int timeout, bool silent)
-{
-	int i, intr_process, rc, tmo_count;
-	struct input *req = msg;
-	u32 *data = msg;
-	__le32 *resp_len, *valid;
-	u16 cp_ring_id, len = 0;
-	struct hwrm_err_output *resp = bp->hwrm_cmd_resp_addr;
-	u16 max_req_len = BNXT_HWRM_MAX_REQ_LEN;
+	req->enables = cpu_to_le32(FUNC_DRV_RGTR_REQ_ENABLES_OS_TYPE |
+				   FUNC_DRV_RGTR_REQ_ENABLES_VER |
+				   FUNC_DRV_RGTR_REQ_ENABLES_ASYNC_EVENT_FWD);
+
+	req->os_type = cpu_to_le16(FUNC_DRV_RGTR_REQ_OS_TYPE_LINUX);
+	flags = FUNC_DRV_RGTR_REQ_FLAGS_16BIT_VER_MODE;
+	if (bp->fw_cap & BNXT_FW_CAP_HOT_RESET)
+		flags |= FUNC_DRV_RGTR_REQ_FLAGS_HOT_RESET_SUPPORT;
+	if (bp->fw_cap & BNXT_FW_CAP_ERROR_RECOVERY)
+		flags |= FUNC_DRV_RGTR_REQ_FLAGS_ERROR_RECOVERY_SUPPORT |
+			 FUNC_DRV_RGTR_REQ_FLAGS_MASTER_SUPPORT;
+	if (bp->fw_cap & BNXT_FW_CAP_NPAR_1_2)
+		flags |= FUNC_DRV_RGTR_REQ_FLAGS_NPAR_1_2_SUPPORT;
+	req->flags = cpu_to_le32(flags);
+	req->ver_maj_8b = DRV_VER_MAJ;
+	req->ver_min_8b = DRV_VER_MIN;
+	req->ver_upd_8b = DRV_VER_UPD;
+	req->ver_maj = cpu_to_le16(DRV_VER_MAJ);
+	req->ver_min = cpu_to_le16(DRV_VER_MIN);
+	req->ver_upd = cpu_to_le16(DRV_VER_UPD);
 
-	req->seq_id = cpu_to_le16(bp->hwrm_cmd_seq++);
-	memset(resp, 0, PAGE_SIZE);
-	cp_ring_id = le16_to_cpu(req->cmpl_ring);
-	intr_process = (cp_ring_id == INVALID_HW_RING_ID) ? 0 : 1;
+	if (BNXT_PF(bp)) {
+		u32 data[8];
+		int i;
 
-	if (bp->flags & BNXT_FLAG_SHORT_CMD) {
-		void *short_cmd_req = bp->hwrm_short_cmd_req_addr;
-		struct hwrm_short_input short_input = {0};
+		memset(data, 0, sizeof(data));
+		for (i = 0; i < ARRAY_SIZE(bnxt_vf_req_snif); i++) {
+			u16 cmd = bnxt_vf_req_snif[i];
+			unsigned int bit, idx;
 
-		memcpy(short_cmd_req, req, msg_len);
-		memset(short_cmd_req + msg_len, 0, BNXT_HWRM_MAX_REQ_LEN -
-						   msg_len);
+			idx = cmd / 32;
+			bit = cmd % 32;
+			data[idx] |= 1 << bit;
+		}
+
+		for (i = 0; i < 8; i++)
+			req->vf_req_fwd[i] = cpu_to_le32(data[i]);
 
-		short_input.req_type = req->req_type;
-		short_input.signature =
-				cpu_to_le16(SHORT_REQ_SIGNATURE_SHORT_CMD);
-		short_input.size = cpu_to_le16(msg_len);
-		short_input.req_addr =
-			cpu_to_le64(bp->hwrm_short_cmd_req_dma_addr);
+		req->enables |=
+			cpu_to_le32(FUNC_DRV_RGTR_REQ_ENABLES_VF_REQ_FWD);
+	}
 
-		data = (u32 *)&short_input;
-		msg_len = sizeof(short_input);
+	if (bp->fw_cap & BNXT_FW_CAP_OVS_64BIT_HANDLE)
+		req->flags |= cpu_to_le32(
+			FUNC_DRV_RGTR_REQ_FLAGS_FLOW_HANDLE_64BIT_MODE);
 
-		/* Sync memory write before updating doorbell */
-		wmb();
+	memset(async_events_bmap, 0, sizeof(async_events_bmap));
+	for (i = 0; i < ARRAY_SIZE(bnxt_async_events_arr); i++) {
+		u16 event_id = bnxt_async_events_arr[i];
 
-		max_req_len = BNXT_HWRM_SHORT_REQ_LEN;
+		if (event_id == ASYNC_EVENT_CMPL_EVENT_ID_ERROR_RECOVERY &&
+		    !(bp->fw_cap & BNXT_FW_CAP_ERROR_RECOVERY))
+			continue;
+		if (event_id == ASYNC_EVENT_CMPL_EVENT_ID_PHC_UPDATE &&
+		    !bp->ptp_cfg)
+			continue;
+		__set_bit(bnxt_async_events_arr[i], async_events_bmap);
+	}
+	if (bmap && bmap_size) {
+		for (i = 0; i < bmap_size; i++) {
+			if (test_bit(i, bmap))
+				__set_bit(i, async_events_bmap);
+		}
 	}
+	for (i = 0; i < 8; i++)
+		req->async_event_fwd[i] |= cpu_to_le32(events[i]);
 
-	/* Write request msg to hwrm channel */
-	__iowrite32_copy(bp->bar0, data, msg_len / 4);
+	if (async_only)
+		req->enables =
+			cpu_to_le32(FUNC_DRV_RGTR_REQ_ENABLES_ASYNC_EVENT_FWD);
 
-	for (i = msg_len; i < max_req_len; i += 4)
-		writel(0, bp->bar0 + i);
+	resp = hwrm_req_hold(bp, req);
+	rc = hwrm_req_send(bp, req);
+	if (!rc) {
+		set_bit(BNXT_STATE_DRV_REGISTERED, &bp->state);
+		if (resp->flags &
+		    cpu_to_le32(FUNC_DRV_RGTR_RESP_FLAGS_IF_CHANGE_SUPPORTED))
+			bp->fw_cap |= BNXT_FW_CAP_IF_CHANGE;
+	}
+	hwrm_req_drop(bp, req);
+	return rc;
+}
 
-	/* currently supports only one outstanding message */
-	if (intr_process)
-		bp->hwrm_intr_seq_id = le16_to_cpu(req->seq_id);
+int bnxt_hwrm_func_drv_unrgtr(struct bnxt *bp)
+{
+	struct hwrm_func_drv_unrgtr_input *req;
+	int rc;
 
-	/* Ring channel doorbell */
-	writel(1, bp->bar0 + 0x100);
+	if (!test_and_clear_bit(BNXT_STATE_DRV_REGISTERED, &bp->state))
+		return 0;
 
-	if (!timeout)
-		timeout = DFLT_HWRM_CMD_TIMEOUT;
+	rc = hwrm_req_init(bp, req, HWRM_FUNC_DRV_UNRGTR);
+	if (rc)
+		return rc;
+	return hwrm_req_send(bp, req);
+}
 
-	i = 0;
-	tmo_count = timeout * 40;
-	if (intr_process) {
-		/* Wait until hwrm response cmpl interrupt is processed */
-		while (bp->hwrm_intr_seq_id != HWRM_SEQ_ID_INVALID &&
-		       i++ < tmo_count) {
-			usleep_range(25, 40);
-		}
+static int bnxt_set_tpa(struct bnxt *bp, bool set_tpa);
 
-		if (bp->hwrm_intr_seq_id != HWRM_SEQ_ID_INVALID) {
-			netdev_err(bp->dev, "Resp cmpl intr err msg: 0x%x\n",
-				   le16_to_cpu(req->req_type));
-			return -1;
-		}
-	} else {
-		/* Check if response len is updated */
-		resp_len = bp->hwrm_cmd_resp_addr + HWRM_RESP_LEN_OFFSET;
-		for (i = 0; i < tmo_count; i++) {
-			len = (le32_to_cpu(*resp_len) & HWRM_RESP_LEN_MASK) >>
-			      HWRM_RESP_LEN_SFT;
-			if (len)
-				break;
-			usleep_range(25, 40);
-		}
+static int bnxt_hwrm_tunnel_dst_port_free(struct bnxt *bp, u8 tunnel_type)
+{
+	struct hwrm_tunnel_dst_port_free_input *req;
+	int rc;
 
-		if (i >= tmo_count) {
-			netdev_err(bp->dev, "Error (timeout: %d) msg {0x%x 0x%x} len:%d\n",
-				   timeout, le16_to_cpu(req->req_type),
-				   le16_to_cpu(req->seq_id), len);
-			return -1;
-		}
+	if (tunnel_type == TUNNEL_DST_PORT_FREE_REQ_TUNNEL_TYPE_VXLAN &&
+	    bp->vxlan_fw_dst_port_id == INVALID_HW_RING_ID)
+		return 0;
+	if (tunnel_type == TUNNEL_DST_PORT_FREE_REQ_TUNNEL_TYPE_GENEVE &&
+	    bp->nge_fw_dst_port_id == INVALID_HW_RING_ID)
+		return 0;
 
-		/* Last word of resp contains valid bit */
-		valid = bp->hwrm_cmd_resp_addr + len - 4;
-		for (i = 0; i < 5; i++) {
-			if (le32_to_cpu(*valid) & HWRM_RESP_VALID_MASK)
-				break;
-			udelay(1);
-		}
+	rc = hwrm_req_init(bp, req, HWRM_TUNNEL_DST_PORT_FREE);
+	if (rc)
+		return rc;
 
-		if (i >= 5) {
-			netdev_err(bp->dev, "Error (timeout: %d) msg {0x%x 0x%x} len:%d v:%d\n",
-				   timeout, le16_to_cpu(req->req_type),
-				   le16_to_cpu(req->seq_id), len, *valid);
-			return -1;
-		}
+	req->tunnel_type = tunnel_type;
+
+	switch (tunnel_type) {
+	case TUNNEL_DST_PORT_FREE_REQ_TUNNEL_TYPE_VXLAN:
+		req->tunnel_dst_port_id = cpu_to_le16(bp->vxlan_fw_dst_port_id);
+		bp->vxlan_port = 0;
+		bp->vxlan_fw_dst_port_id = INVALID_HW_RING_ID;
+		break;
+	case TUNNEL_DST_PORT_FREE_REQ_TUNNEL_TYPE_GENEVE:
+		req->tunnel_dst_port_id = cpu_to_le16(bp->nge_fw_dst_port_id);
+		bp->nge_port = 0;
+		bp->nge_fw_dst_port_id = INVALID_HW_RING_ID;
+		break;
+	case TUNNEL_DST_PORT_FREE_REQ_TUNNEL_TYPE_VXLAN_GPE:
+		req->tunnel_dst_port_id = cpu_to_le16(bp->vxlan_gpe_fw_dst_port_id);
+		bp->vxlan_gpe_port = 0;
+		bp->vxlan_gpe_fw_dst_port_id = INVALID_HW_RING_ID;
+		break;
+	default:
+		break;
 	}
 
-	rc = le16_to_cpu(resp->error_code);
-	if (rc && !silent)
-		netdev_err(bp->dev, "hwrm req_type 0x%x seq id 0x%x error 0x%x\n",
-			   le16_to_cpu(resp->req_type),
-			   le16_to_cpu(resp->seq_id), rc);
+	rc = hwrm_req_send(bp, req);
+	if (rc)
+		netdev_err(bp->dev, "hwrm_tunnel_dst_port_free failed. rc:%d\n",
+			   rc);
+	if (bp->flags & BNXT_FLAG_TPA)
+		bnxt_set_tpa(bp, true);
 	return rc;
 }
 
-int _hwrm_send_message(struct bnxt *bp, void *msg, u32 msg_len, int timeout)
+static int bnxt_hwrm_tunnel_dst_port_alloc(struct bnxt *bp, __be16 port,
+					   u8 tunnel_type)
 {
-	return bnxt_hwrm_do_send_msg(bp, msg, msg_len, timeout, false);
+	struct hwrm_tunnel_dst_port_alloc_output *resp;
+	struct hwrm_tunnel_dst_port_alloc_input *req;
+	int rc;
+
+	rc = hwrm_req_init(bp, req, HWRM_TUNNEL_DST_PORT_ALLOC);
+	if (rc)
+		return rc;
+
+	req->tunnel_type = tunnel_type;
+	req->tunnel_dst_port_val = port;
+
+	resp = hwrm_req_hold(bp, req);
+	rc = hwrm_req_send(bp, req);
+	if (rc) {
+		netdev_err(bp->dev, "hwrm_tunnel_dst_port_alloc failed. rc:%d\n",
+			   rc);
+		goto err_out;
+	}
+
+	switch (tunnel_type) {
+	case TUNNEL_DST_PORT_ALLOC_REQ_TUNNEL_TYPE_VXLAN:
+		bp->vxlan_port = port;
+		bp->vxlan_fw_dst_port_id =
+			le16_to_cpu(resp->tunnel_dst_port_id);
+		break;
+	case TUNNEL_DST_PORT_ALLOC_REQ_TUNNEL_TYPE_GENEVE:
+		bp->nge_port = port;
+		bp->nge_fw_dst_port_id = le16_to_cpu(resp->tunnel_dst_port_id);
+		break;
+	case TUNNEL_DST_PORT_ALLOC_REQ_TUNNEL_TYPE_VXLAN_GPE:
+		bp->vxlan_gpe_port = port;
+		bp->vxlan_gpe_fw_dst_port_id =
+			le16_to_cpu(resp->tunnel_dst_port_id);
+		break;
+	default:
+		break;
+	}
+	if (bp->flags & BNXT_FLAG_TPA)
+		bnxt_set_tpa(bp, true);
+
+err_out:
+	hwrm_req_drop(bp, req);
+	return rc;
 }
 
-int hwrm_send_message(struct bnxt *bp, void *msg, u32 msg_len, int timeout)
+static int bnxt_hwrm_cfa_l2_set_rx_mask(struct bnxt *bp, u16 vnic_id)
 {
+	struct hwrm_cfa_l2_set_rx_mask_input *req;
+	struct bnxt_vnic_info *vnic = &bp->vnic_info[vnic_id];
 	int rc;
 
-	mutex_lock(&bp->hwrm_cmd_lock);
-	rc = _hwrm_send_message(bp, msg, msg_len, timeout);
-	mutex_unlock(&bp->hwrm_cmd_lock);
-	return rc;
+	rc = hwrm_req_init(bp, req, HWRM_CFA_L2_SET_RX_MASK);
+	if (rc)
+		return rc;
+
+	req->vnic_id = cpu_to_le32(vnic->fw_vnic_id);
+	if (vnic->rx_mask & CFA_L2_SET_RX_MASK_REQ_MASK_MCAST) {
+		req->num_mc_entries = cpu_to_le32(vnic->mc_list_count);
+		req->mc_tbl_addr = cpu_to_le64(vnic->mc_list_mapping);
+	}
+	req->mask = cpu_to_le32(vnic->rx_mask);
+	return hwrm_req_send_silent(bp, req);
 }
 
-int hwrm_send_message_silent(struct bnxt *bp, void *msg, u32 msg_len,
-			     int timeout)
+void bnxt_del_l2_filter(struct bnxt *bp, struct bnxt_l2_filter *fltr)
 {
-	int rc;
+	if (!atomic_dec_and_test(&fltr->refcnt))
+		return;
+	spin_lock_bh(&bp->ntp_fltr_lock);
+	if (!test_and_clear_bit(BNXT_FLTR_INSERTED, &fltr->base.state)) {
+		spin_unlock_bh(&bp->ntp_fltr_lock);
+		return;
+	}
+	hlist_del_rcu(&fltr->base.hash);
+	bnxt_del_one_usr_fltr(bp, &fltr->base);
+	if (fltr->base.flags) {
+		clear_bit(fltr->base.sw_id, bp->ntp_fltr_bmap);
+		bp->ntp_fltr_count--;
+	}
+	spin_unlock_bh(&bp->ntp_fltr_lock);
+	kfree_rcu(fltr, base.rcu);
+}
 
-	mutex_lock(&bp->hwrm_cmd_lock);
-	rc = bnxt_hwrm_do_send_msg(bp, msg, msg_len, timeout, true);
-	mutex_unlock(&bp->hwrm_cmd_lock);
-	return rc;
+static struct bnxt_l2_filter *__bnxt_lookup_l2_filter(struct bnxt *bp,
+						      struct bnxt_l2_key *key,
+						      u32 idx)
+{
+	struct hlist_head *head = &bp->l2_fltr_hash_tbl[idx];
+	struct bnxt_l2_filter *fltr;
+
+	hlist_for_each_entry_rcu(fltr, head, base.hash) {
+		struct bnxt_l2_key *l2_key = &fltr->l2_key;
+
+		if (ether_addr_equal(l2_key->dst_mac_addr, key->dst_mac_addr) &&
+		    l2_key->vlan == key->vlan)
+			return fltr;
+	}
+	return NULL;
 }
 
-int bnxt_hwrm_func_rgtr_async_events(struct bnxt *bp, unsigned long *bmap,
-				     int bmap_size)
+static struct bnxt_l2_filter *bnxt_lookup_l2_filter(struct bnxt *bp,
+						    struct bnxt_l2_key *key,
+						    u32 idx)
 {
-	struct hwrm_func_drv_rgtr_input req = {0};
-	DECLARE_BITMAP(async_events_bmap, 256);
-	u32 *events = (u32 *)async_events_bmap;
-	int i;
+	struct bnxt_l2_filter *fltr = NULL;
 
-	bnxt_hwrm_cmd_hdr_init(bp, &req, HWRM_FUNC_DRV_RGTR, -1, -1);
+	rcu_read_lock();
+	fltr = __bnxt_lookup_l2_filter(bp, key, idx);
+	if (fltr)
+		atomic_inc(&fltr->refcnt);
+	rcu_read_unlock();
+	return fltr;
+}
 
-	req.enables =
-		cpu_to_le32(FUNC_DRV_RGTR_REQ_ENABLES_ASYNC_EVENT_FWD);
+#define BNXT_IPV4_4TUPLE(bp, fkeys)					\
+	(((fkeys)->basic.ip_proto == IPPROTO_TCP &&			\
+	  (bp)->rss_hash_cfg & VNIC_RSS_CFG_REQ_HASH_TYPE_TCP_IPV4) ||	\
+	 ((fkeys)->basic.ip_proto == IPPROTO_UDP &&			\
+	  (bp)->rss_hash_cfg & VNIC_RSS_CFG_REQ_HASH_TYPE_UDP_IPV4))
 
-	memset(async_events_bmap, 0, sizeof(async_events_bmap));
-	for (i = 0; i < ARRAY_SIZE(bnxt_async_events_arr); i++)
-		__set_bit(bnxt_async_events_arr[i], async_events_bmap);
+#define BNXT_IPV6_4TUPLE(bp, fkeys)					\
+	(((fkeys)->basic.ip_proto == IPPROTO_TCP &&			\
+	  (bp)->rss_hash_cfg & VNIC_RSS_CFG_REQ_HASH_TYPE_TCP_IPV6) ||	\
+	 ((fkeys)->basic.ip_proto == IPPROTO_UDP &&			\
+	  (bp)->rss_hash_cfg & VNIC_RSS_CFG_REQ_HASH_TYPE_UDP_IPV6))
 
-	if (bmap && bmap_size) {
-		for (i = 0; i < bmap_size; i++) {
-			if (test_bit(i, bmap))
-				__set_bit(i, async_events_bmap);
+static u32 bnxt_get_rss_flow_tuple_len(struct bnxt *bp, struct flow_keys *fkeys)
+{
+	if (fkeys->basic.n_proto == htons(ETH_P_IP)) {
+		if (BNXT_IPV4_4TUPLE(bp, fkeys))
+			return sizeof(fkeys->addrs.v4addrs) +
+			       sizeof(fkeys->ports);
+
+		if (bp->rss_hash_cfg & VNIC_RSS_CFG_REQ_HASH_TYPE_IPV4)
+			return sizeof(fkeys->addrs.v4addrs);
+	}
+
+	if (fkeys->basic.n_proto == htons(ETH_P_IPV6)) {
+		if (BNXT_IPV6_4TUPLE(bp, fkeys))
+			return sizeof(fkeys->addrs.v6addrs) +
+			       sizeof(fkeys->ports);
+
+		if (bp->rss_hash_cfg & VNIC_RSS_CFG_REQ_HASH_TYPE_IPV6)
+			return sizeof(fkeys->addrs.v6addrs);
+	}
+
+	return 0;
+}
+
+static u32 bnxt_toeplitz(struct bnxt *bp, struct flow_keys *fkeys,
+			 const unsigned char *key)
+{
+	u64 prefix = bp->toeplitz_prefix, hash = 0;
+	struct bnxt_ipv4_tuple tuple4;
+	struct bnxt_ipv6_tuple tuple6;
+	int i, j, len = 0;
+	u8 *four_tuple;
+
+	len = bnxt_get_rss_flow_tuple_len(bp, fkeys);
+	if (!len)
+		return 0;
+
+	if (fkeys->basic.n_proto == htons(ETH_P_IP)) {
+		tuple4.v4addrs = fkeys->addrs.v4addrs;
+		tuple4.ports = fkeys->ports;
+		four_tuple = (unsigned char *)&tuple4;
+	} else {
+		tuple6.v6addrs = fkeys->addrs.v6addrs;
+		tuple6.ports = fkeys->ports;
+		four_tuple = (unsigned char *)&tuple6;
+	}
+
+	for (i = 0, j = 8; i < len; i++, j++) {
+		u8 byte = four_tuple[i];
+		int bit;
+
+		for (bit = 0; bit < 8; bit++, prefix <<= 1, byte <<= 1) {
+			if (byte & 0x80)
+				hash ^= prefix;
 		}
+		prefix |= (j < HW_HASH_KEY_SIZE) ? key[j] : 0;
 	}
 
-	for (i = 0; i < 8; i++)
-		req.async_event_fwd[i] |= cpu_to_le32(events[i]);
+	/* The valid part of the hash is in the upper 32 bits. */
+	return (hash >> 32) & BNXT_NTP_FLTR_HASH_MASK;
+}
 
-	return hwrm_send_message(bp, &req, sizeof(req), HWRM_CMD_TIMEOUT);
+#ifdef CONFIG_RFS_ACCEL
+static struct bnxt_l2_filter *
+bnxt_lookup_l2_filter_from_key(struct bnxt *bp, struct bnxt_l2_key *key)
+{
+	struct bnxt_l2_filter *fltr;
+	u32 idx;
+
+	idx = jhash2(&key->filter_key, BNXT_L2_KEY_SIZE, bp->hash_seed) &
+	      BNXT_L2_FLTR_HASH_MASK;
+	fltr = bnxt_lookup_l2_filter(bp, key, idx);
+	return fltr;
 }
+#endif
 
-static int bnxt_hwrm_func_drv_rgtr(struct bnxt *bp)
+static int bnxt_init_l2_filter(struct bnxt *bp, struct bnxt_l2_filter *fltr,
+			       struct bnxt_l2_key *key, u32 idx)
 {
-	struct hwrm_func_drv_rgtr_input req = {0};
+	struct hlist_head *head;
 
-	bnxt_hwrm_cmd_hdr_init(bp, &req, HWRM_FUNC_DRV_RGTR, -1, -1);
+	ether_addr_copy(fltr->l2_key.dst_mac_addr, key->dst_mac_addr);
+	fltr->l2_key.vlan = key->vlan;
+	fltr->base.type = BNXT_FLTR_TYPE_L2;
+	if (fltr->base.flags) {
+		int bit_id;
 
-	req.enables =
-		cpu_to_le32(FUNC_DRV_RGTR_REQ_ENABLES_OS_TYPE |
-			    FUNC_DRV_RGTR_REQ_ENABLES_VER);
+		bit_id = bitmap_find_free_region(bp->ntp_fltr_bmap,
+						 bp->max_fltr, 0);
+		if (bit_id < 0)
+			return -ENOMEM;
+		fltr->base.sw_id = (u16)bit_id;
+		bp->ntp_fltr_count++;
+	}
+	head = &bp->l2_fltr_hash_tbl[idx];
+	hlist_add_head_rcu(&fltr->base.hash, head);
+	bnxt_insert_usr_fltr(bp, &fltr->base);
+	set_bit(BNXT_FLTR_INSERTED, &fltr->base.state);
+	atomic_set(&fltr->refcnt, 1);
+	return 0;
+}
 
-	req.os_type = cpu_to_le16(FUNC_DRV_RGTR_REQ_OS_TYPE_LINUX);
-	req.ver_maj = DRV_VER_MAJ;
-	req.ver_min = DRV_VER_MIN;
-	req.ver_upd = DRV_VER_UPD;
+static struct bnxt_l2_filter *bnxt_alloc_l2_filter(struct bnxt *bp,
+						   struct bnxt_l2_key *key,
+						   gfp_t gfp)
+{
+	struct bnxt_l2_filter *fltr;
+	u32 idx;
+	int rc;
 
-	if (BNXT_PF(bp)) {
-		DECLARE_BITMAP(vf_req_snif_bmap, 256);
-		u32 *data = (u32 *)vf_req_snif_bmap;
-		int i;
+	idx = jhash2(&key->filter_key, BNXT_L2_KEY_SIZE, bp->hash_seed) &
+	      BNXT_L2_FLTR_HASH_MASK;
+	fltr = bnxt_lookup_l2_filter(bp, key, idx);
+	if (fltr)
+		return fltr;
 
-		memset(vf_req_snif_bmap, 0, sizeof(vf_req_snif_bmap));
-		for (i = 0; i < ARRAY_SIZE(bnxt_vf_req_snif); i++)
-			__set_bit(bnxt_vf_req_snif[i], vf_req_snif_bmap);
+	fltr = kzalloc(sizeof(*fltr), gfp);
+	if (!fltr)
+		return ERR_PTR(-ENOMEM);
+	spin_lock_bh(&bp->ntp_fltr_lock);
+	rc = bnxt_init_l2_filter(bp, fltr, key, idx);
+	spin_unlock_bh(&bp->ntp_fltr_lock);
+	if (rc) {
+		bnxt_del_l2_filter(bp, fltr);
+		fltr = ERR_PTR(rc);
+	}
+	return fltr;
+}
 
-		for (i = 0; i < 8; i++)
-			req.vf_req_fwd[i] = cpu_to_le32(data[i]);
+struct bnxt_l2_filter *bnxt_alloc_new_l2_filter(struct bnxt *bp,
+						struct bnxt_l2_key *key,
+						u16 flags)
+{
+	struct bnxt_l2_filter *fltr;
+	u32 idx;
+	int rc;
 
-		req.enables |=
-			cpu_to_le32(FUNC_DRV_RGTR_REQ_ENABLES_VF_REQ_FWD);
+	idx = jhash2(&key->filter_key, BNXT_L2_KEY_SIZE, bp->hash_seed) &
+	      BNXT_L2_FLTR_HASH_MASK;
+	spin_lock_bh(&bp->ntp_fltr_lock);
+	fltr = __bnxt_lookup_l2_filter(bp, key, idx);
+	if (fltr) {
+		fltr = ERR_PTR(-EEXIST);
+		goto l2_filter_exit;
+	}
+	fltr = kzalloc(sizeof(*fltr), GFP_ATOMIC);
+	if (!fltr) {
+		fltr = ERR_PTR(-ENOMEM);
+		goto l2_filter_exit;
+	}
+	fltr->base.flags = flags;
+	rc = bnxt_init_l2_filter(bp, fltr, key, idx);
+	if (rc) {
+		spin_unlock_bh(&bp->ntp_fltr_lock);
+		bnxt_del_l2_filter(bp, fltr);
+		return ERR_PTR(rc);
 	}
 
-	return hwrm_send_message(bp, &req, sizeof(req), HWRM_CMD_TIMEOUT);
+l2_filter_exit:
+	spin_unlock_bh(&bp->ntp_fltr_lock);
+	return fltr;
 }
 
-static int bnxt_hwrm_func_drv_unrgtr(struct bnxt *bp)
+static u16 bnxt_vf_target_id(struct bnxt_pf_info *pf, u16 vf_idx)
 {
-	struct hwrm_func_drv_unrgtr_input req = {0};
+#ifdef CONFIG_BNXT_SRIOV
+	struct bnxt_vf_info *vf = &pf->vf[vf_idx];
 
-	bnxt_hwrm_cmd_hdr_init(bp, &req, HWRM_FUNC_DRV_UNRGTR, -1, -1);
-	return hwrm_send_message(bp, &req, sizeof(req), HWRM_CMD_TIMEOUT);
+	return vf->fw_fid;
+#else
+	return INVALID_HW_RING_ID;
+#endif
 }
 
-static int bnxt_hwrm_tunnel_dst_port_free(struct bnxt *bp, u8 tunnel_type)
+int bnxt_hwrm_l2_filter_free(struct bnxt *bp, struct bnxt_l2_filter *fltr)
 {
-	u32 rc = 0;
-	struct hwrm_tunnel_dst_port_free_input req = {0};
+	struct hwrm_cfa_l2_filter_free_input *req;
+	u16 target_id = 0xffff;
+	int rc;
 
-	bnxt_hwrm_cmd_hdr_init(bp, &req, HWRM_TUNNEL_DST_PORT_FREE, -1, -1);
-	req.tunnel_type = tunnel_type;
+	if (fltr->base.flags & BNXT_ACT_FUNC_DST) {
+		struct bnxt_pf_info *pf = &bp->pf;
 
-	switch (tunnel_type) {
-	case TUNNEL_DST_PORT_FREE_REQ_TUNNEL_TYPE_VXLAN:
-		req.tunnel_dst_port_id = bp->vxlan_fw_dst_port_id;
-		break;
-	case TUNNEL_DST_PORT_FREE_REQ_TUNNEL_TYPE_GENEVE:
-		req.tunnel_dst_port_id = bp->nge_fw_dst_port_id;
-		break;
-	default:
-		break;
+		if (fltr->base.vf_idx >= pf->active_vfs)
+			return -EINVAL;
+
+		target_id = bnxt_vf_target_id(pf, fltr->base.vf_idx);
+		if (target_id == INVALID_HW_RING_ID)
+			return -EINVAL;
 	}
 
-	rc = hwrm_send_message(bp, &req, sizeof(req), HWRM_CMD_TIMEOUT);
+	rc = hwrm_req_init(bp, req, HWRM_CFA_L2_FILTER_FREE);
 	if (rc)
-		netdev_err(bp->dev, "hwrm_tunnel_dst_port_free failed. rc:%d\n",
-			   rc);
-	return rc;
+		return rc;
+
+	req->target_id = cpu_to_le16(target_id);
+	req->l2_filter_id = fltr->base.filter_id;
+	return hwrm_req_send(bp, req);
 }
 
-static int bnxt_hwrm_tunnel_dst_port_alloc(struct bnxt *bp, __be16 port,
-					   u8 tunnel_type)
+int bnxt_hwrm_l2_filter_alloc(struct bnxt *bp, struct bnxt_l2_filter *fltr)
 {
-	u32 rc = 0;
-	struct hwrm_tunnel_dst_port_alloc_input req = {0};
-	struct hwrm_tunnel_dst_port_alloc_output *resp = bp->hwrm_cmd_resp_addr;
+	struct hwrm_cfa_l2_filter_alloc_output *resp;
+	struct hwrm_cfa_l2_filter_alloc_input *req;
+	u16 target_id = 0xffff;
+	int rc;
 
-	bnxt_hwrm_cmd_hdr_init(bp, &req, HWRM_TUNNEL_DST_PORT_ALLOC, -1, -1);
+	if (fltr->base.flags & BNXT_ACT_FUNC_DST) {
+		struct bnxt_pf_info *pf = &bp->pf;
 
-	req.tunnel_type = tunnel_type;
-	req.tunnel_dst_port_val = port;
+		if (fltr->base.vf_idx >= pf->active_vfs)
+			return -EINVAL;
 
-	mutex_lock(&bp->hwrm_cmd_lock);
-	rc = _hwrm_send_message(bp, &req, sizeof(req), HWRM_CMD_TIMEOUT);
-	if (rc) {
-		netdev_err(bp->dev, "hwrm_tunnel_dst_port_alloc failed. rc:%d\n",
-			   rc);
-		goto err_out;
+		target_id = bnxt_vf_target_id(pf, fltr->base.vf_idx);
 	}
+	rc = hwrm_req_init(bp, req, HWRM_CFA_L2_FILTER_ALLOC);
+	if (rc)
+		return rc;
 
-	switch (tunnel_type) {
-	case TUNNEL_DST_PORT_ALLOC_REQ_TUNNEL_TYPE_VXLAN:
-		bp->vxlan_fw_dst_port_id = resp->tunnel_dst_port_id;
-		break;
-	case TUNNEL_DST_PORT_ALLOC_REQ_TUNNEL_TYPE_GENEVE:
-		bp->nge_fw_dst_port_id = resp->tunnel_dst_port_id;
-		break;
-	default:
-		break;
+	req->target_id = cpu_to_le16(target_id);
+	req->flags = cpu_to_le32(CFA_L2_FILTER_ALLOC_REQ_FLAGS_PATH_RX);
+
+	if (!BNXT_CHIP_TYPE_NITRO_A0(bp))
+		req->flags |=
+			cpu_to_le32(CFA_L2_FILTER_ALLOC_REQ_FLAGS_OUTERMOST);
+	req->dst_id = cpu_to_le16(fltr->base.fw_vnic_id);
+	req->enables =
+		cpu_to_le32(CFA_L2_FILTER_ALLOC_REQ_ENABLES_L2_ADDR |
+			    CFA_L2_FILTER_ALLOC_REQ_ENABLES_DST_ID |
+			    CFA_L2_FILTER_ALLOC_REQ_ENABLES_L2_ADDR_MASK);
+	ether_addr_copy(req->l2_addr, fltr->l2_key.dst_mac_addr);
+	eth_broadcast_addr(req->l2_addr_mask);
+
+	if (fltr->l2_key.vlan) {
+		req->enables |=
+			cpu_to_le32(CFA_L2_FILTER_ALLOC_REQ_ENABLES_L2_IVLAN |
+				CFA_L2_FILTER_ALLOC_REQ_ENABLES_L2_IVLAN_MASK |
+				CFA_L2_FILTER_ALLOC_REQ_ENABLES_NUM_VLANS);
+		req->num_vlans = 1;
+		req->l2_ivlan = cpu_to_le16(fltr->l2_key.vlan);
+		req->l2_ivlan_mask = cpu_to_le16(0xfff);
 	}
 
-err_out:
-	mutex_unlock(&bp->hwrm_cmd_lock);
+	resp = hwrm_req_hold(bp, req);
+	rc = hwrm_req_send(bp, req);
+	if (!rc) {
+		fltr->base.filter_id = resp->l2_filter_id;
+		set_bit(BNXT_FLTR_VALID, &fltr->base.state);
+	}
+	hwrm_req_drop(bp, req);
 	return rc;
 }
 
-static int bnxt_hwrm_cfa_l2_set_rx_mask(struct bnxt *bp, u16 vnic_id)
+int bnxt_hwrm_cfa_ntuple_filter_free(struct bnxt *bp,
+				     struct bnxt_ntuple_filter *fltr)
 {
-	struct hwrm_cfa_l2_set_rx_mask_input req = {0};
-	struct bnxt_vnic_info *vnic = &bp->vnic_info[vnic_id];
-
-	bnxt_hwrm_cmd_hdr_init(bp, &req, HWRM_CFA_L2_SET_RX_MASK, -1, -1);
-	req.vnic_id = cpu_to_le32(vnic->fw_vnic_id);
-
-	req.num_mc_entries = cpu_to_le32(vnic->mc_list_count);
-	req.mc_tbl_addr = cpu_to_le64(vnic->mc_list_mapping);
-	req.mask = cpu_to_le32(vnic->rx_mask);
-	return hwrm_send_message(bp, &req, sizeof(req), HWRM_CMD_TIMEOUT);
-}
+	struct hwrm_cfa_ntuple_filter_free_input *req;
+	int rc;
 
-#ifdef CONFIG_RFS_ACCEL
-static int bnxt_hwrm_cfa_ntuple_filter_free(struct bnxt *bp,
-					    struct bnxt_ntuple_filter *fltr)
-{
-	struct hwrm_cfa_ntuple_filter_free_input req = {0};
+	set_bit(BNXT_FLTR_FW_DELETED, &fltr->base.state);
+	rc = hwrm_req_init(bp, req, HWRM_CFA_NTUPLE_FILTER_FREE);
+	if (rc)
+		return rc;
 
-	bnxt_hwrm_cmd_hdr_init(bp, &req, HWRM_CFA_NTUPLE_FILTER_FREE, -1, -1);
-	req.ntuple_filter_id = fltr->filter_id;
-	return hwrm_send_message(bp, &req, sizeof(req), HWRM_CMD_TIMEOUT);
+	req->ntuple_filter_id = fltr->base.filter_id;
+	return hwrm_req_send(bp, req);
 }
 
 #define BNXT_NTP_FLTR_FLAGS					\
 	(CFA_NTUPLE_FILTER_ALLOC_REQ_ENABLES_L2_FILTER_ID |	\
 	 CFA_NTUPLE_FILTER_ALLOC_REQ_ENABLES_ETHERTYPE |	\
-	 CFA_NTUPLE_FILTER_ALLOC_REQ_ENABLES_SRC_MACADDR |	\
 	 CFA_NTUPLE_FILTER_ALLOC_REQ_ENABLES_IPADDR_TYPE |	\
 	 CFA_NTUPLE_FILTER_ALLOC_REQ_ENABLES_SRC_IPADDR |	\
 	 CFA_NTUPLE_FILTER_ALLOC_REQ_ENABLES_SRC_IPADDR_MASK |	\
@@ -3546,135 +6218,197 @@ static int bnxt_hwrm_cfa_ntuple_filter_free(struct bnxt *bp,
 #define BNXT_NTP_TUNNEL_FLTR_FLAG				\
 		CFA_NTUPLE_FILTER_ALLOC_REQ_ENABLES_TUNNEL_TYPE
 
-static int bnxt_hwrm_cfa_ntuple_filter_alloc(struct bnxt *bp,
-					     struct bnxt_ntuple_filter *fltr)
+void bnxt_fill_ipv6_mask(__be32 mask[4])
 {
-	int rc = 0;
-	struct hwrm_cfa_ntuple_filter_alloc_input req = {0};
-	struct hwrm_cfa_ntuple_filter_alloc_output *resp =
-		bp->hwrm_cmd_resp_addr;
+	int i;
+
+	for (i = 0; i < 4; i++)
+		mask[i] = cpu_to_be32(~0);
+}
+
+static void
+bnxt_cfg_rfs_ring_tbl_idx(struct bnxt *bp,
+			  struct hwrm_cfa_ntuple_filter_alloc_input *req,
+			  struct bnxt_ntuple_filter *fltr)
+{
+	u16 rxq = fltr->base.rxq;
+
+	if (fltr->base.flags & BNXT_ACT_RSS_CTX) {
+		struct ethtool_rxfh_context *ctx;
+		struct bnxt_rss_ctx *rss_ctx;
+		struct bnxt_vnic_info *vnic;
+
+		ctx = xa_load(&bp->dev->ethtool->rss_ctx,
+			      fltr->base.fw_vnic_id);
+		if (ctx) {
+			rss_ctx = ethtool_rxfh_context_priv(ctx);
+			vnic = &rss_ctx->vnic;
+
+			req->dst_id = cpu_to_le16(vnic->fw_vnic_id);
+		}
+		return;
+	}
+	if (BNXT_SUPPORTS_NTUPLE_VNIC(bp)) {
+		struct bnxt_vnic_info *vnic;
+		u32 enables;
+
+		vnic = &bp->vnic_info[BNXT_VNIC_NTUPLE];
+		req->dst_id = cpu_to_le16(vnic->fw_vnic_id);
+		enables = CFA_NTUPLE_FILTER_ALLOC_REQ_ENABLES_RFS_RING_TBL_IDX;
+		req->enables |= cpu_to_le32(enables);
+		req->rfs_ring_tbl_idx = cpu_to_le16(rxq);
+	} else {
+		u32 flags;
+
+		flags = CFA_NTUPLE_FILTER_ALLOC_REQ_FLAGS_DEST_RFS_RING_IDX;
+		req->flags |= cpu_to_le32(flags);
+		req->dst_id = cpu_to_le16(rxq);
+	}
+}
+
+int bnxt_hwrm_cfa_ntuple_filter_alloc(struct bnxt *bp,
+				      struct bnxt_ntuple_filter *fltr)
+{
+	struct hwrm_cfa_ntuple_filter_alloc_output *resp;
+	struct hwrm_cfa_ntuple_filter_alloc_input *req;
+	struct bnxt_flow_masks *masks = &fltr->fmasks;
 	struct flow_keys *keys = &fltr->fkeys;
-	struct bnxt_vnic_info *vnic = &bp->vnic_info[fltr->rxq + 1];
+	struct bnxt_l2_filter *l2_fltr;
+	struct bnxt_vnic_info *vnic;
+	int rc;
 
-	bnxt_hwrm_cmd_hdr_init(bp, &req, HWRM_CFA_NTUPLE_FILTER_ALLOC, -1, -1);
-	req.l2_filter_id = bp->vnic_info[0].fw_l2_filter_id[fltr->l2_fltr_idx];
+	rc = hwrm_req_init(bp, req, HWRM_CFA_NTUPLE_FILTER_ALLOC);
+	if (rc)
+		return rc;
 
-	req.enables = cpu_to_le32(BNXT_NTP_FLTR_FLAGS);
+	l2_fltr = fltr->l2_fltr;
+	req->l2_filter_id = l2_fltr->base.filter_id;
 
-	req.ethertype = htons(ETH_P_IP);
-	memcpy(req.src_macaddr, fltr->src_mac_addr, ETH_ALEN);
-	req.ip_addr_type = CFA_NTUPLE_FILTER_ALLOC_REQ_IP_ADDR_TYPE_IPV4;
-	req.ip_protocol = keys->basic.ip_proto;
+	if (fltr->base.flags & BNXT_ACT_DROP) {
+		req->flags =
+			cpu_to_le32(CFA_NTUPLE_FILTER_ALLOC_REQ_FLAGS_DROP);
+	} else if (bp->fw_cap & BNXT_FW_CAP_CFA_RFS_RING_TBL_IDX_V2) {
+		bnxt_cfg_rfs_ring_tbl_idx(bp, req, fltr);
+	} else {
+		vnic = &bp->vnic_info[fltr->base.rxq + 1];
+		req->dst_id = cpu_to_le16(vnic->fw_vnic_id);
+	}
+	req->enables |= cpu_to_le32(BNXT_NTP_FLTR_FLAGS);
 
-	if (keys->basic.n_proto == htons(ETH_P_IPV6)) {
-		int i;
+	req->ethertype = htons(ETH_P_IP);
+	req->ip_addr_type = CFA_NTUPLE_FILTER_ALLOC_REQ_IP_ADDR_TYPE_IPV4;
+	req->ip_protocol = keys->basic.ip_proto;
 
-		req.ethertype = htons(ETH_P_IPV6);
-		req.ip_addr_type =
+	if (keys->basic.n_proto == htons(ETH_P_IPV6)) {
+		req->ethertype = htons(ETH_P_IPV6);
+		req->ip_addr_type =
 			CFA_NTUPLE_FILTER_ALLOC_REQ_IP_ADDR_TYPE_IPV6;
-		*(struct in6_addr *)&req.src_ipaddr[0] =
-			keys->addrs.v6addrs.src;
-		*(struct in6_addr *)&req.dst_ipaddr[0] =
-			keys->addrs.v6addrs.dst;
-		for (i = 0; i < 4; i++) {
-			req.src_ipaddr_mask[i] = cpu_to_be32(0xffffffff);
-			req.dst_ipaddr_mask[i] = cpu_to_be32(0xffffffff);
-		}
+		*(struct in6_addr *)&req->src_ipaddr[0] = keys->addrs.v6addrs.src;
+		*(struct in6_addr *)&req->src_ipaddr_mask[0] = masks->addrs.v6addrs.src;
+		*(struct in6_addr *)&req->dst_ipaddr[0] = keys->addrs.v6addrs.dst;
+		*(struct in6_addr *)&req->dst_ipaddr_mask[0] = masks->addrs.v6addrs.dst;
 	} else {
-		req.src_ipaddr[0] = keys->addrs.v4addrs.src;
-		req.src_ipaddr_mask[0] = cpu_to_be32(0xffffffff);
-		req.dst_ipaddr[0] = keys->addrs.v4addrs.dst;
-		req.dst_ipaddr_mask[0] = cpu_to_be32(0xffffffff);
+		req->src_ipaddr[0] = keys->addrs.v4addrs.src;
+		req->src_ipaddr_mask[0] = masks->addrs.v4addrs.src;
+		req->dst_ipaddr[0] = keys->addrs.v4addrs.dst;
+		req->dst_ipaddr_mask[0] = masks->addrs.v4addrs.dst;
 	}
 	if (keys->control.flags & FLOW_DIS_ENCAPSULATION) {
-		req.enables |= cpu_to_le32(BNXT_NTP_TUNNEL_FLTR_FLAG);
-		req.tunnel_type =
+		req->enables |= cpu_to_le32(BNXT_NTP_TUNNEL_FLTR_FLAG);
+		req->tunnel_type =
 			CFA_NTUPLE_FILTER_ALLOC_REQ_TUNNEL_TYPE_ANYTUNNEL;
 	}
 
-	req.src_port = keys->ports.src;
-	req.src_port_mask = cpu_to_be16(0xffff);
-	req.dst_port = keys->ports.dst;
-	req.dst_port_mask = cpu_to_be16(0xffff);
+	req->src_port = keys->ports.src;
+	req->src_port_mask = masks->ports.src;
+	req->dst_port = keys->ports.dst;
+	req->dst_port_mask = masks->ports.dst;
 
-	req.dst_id = cpu_to_le16(vnic->fw_vnic_id);
-	mutex_lock(&bp->hwrm_cmd_lock);
-	rc = _hwrm_send_message(bp, &req, sizeof(req), HWRM_CMD_TIMEOUT);
+	resp = hwrm_req_hold(bp, req);
+	rc = hwrm_req_send(bp, req);
 	if (!rc)
-		fltr->filter_id = resp->ntuple_filter_id;
-	mutex_unlock(&bp->hwrm_cmd_lock);
+		fltr->base.filter_id = resp->ntuple_filter_id;
+	hwrm_req_drop(bp, req);
 	return rc;
 }
-#endif
 
 static int bnxt_hwrm_set_vnic_filter(struct bnxt *bp, u16 vnic_id, u16 idx,
-				     u8 *mac_addr)
+				     const u8 *mac_addr)
 {
-	u32 rc = 0;
-	struct hwrm_cfa_l2_filter_alloc_input req = {0};
-	struct hwrm_cfa_l2_filter_alloc_output *resp = bp->hwrm_cmd_resp_addr;
+	struct bnxt_l2_filter *fltr;
+	struct bnxt_l2_key key;
+	int rc;
 
-	bnxt_hwrm_cmd_hdr_init(bp, &req, HWRM_CFA_L2_FILTER_ALLOC, -1, -1);
-	req.flags = cpu_to_le32(CFA_L2_FILTER_ALLOC_REQ_FLAGS_PATH_RX);
-	if (!BNXT_CHIP_TYPE_NITRO_A0(bp))
-		req.flags |=
-			cpu_to_le32(CFA_L2_FILTER_ALLOC_REQ_FLAGS_OUTERMOST);
-	req.dst_id = cpu_to_le16(bp->vnic_info[vnic_id].fw_vnic_id);
-	req.enables =
-		cpu_to_le32(CFA_L2_FILTER_ALLOC_REQ_ENABLES_L2_ADDR |
-			    CFA_L2_FILTER_ALLOC_REQ_ENABLES_DST_ID |
-			    CFA_L2_FILTER_ALLOC_REQ_ENABLES_L2_ADDR_MASK);
-	memcpy(req.l2_addr, mac_addr, ETH_ALEN);
-	req.l2_addr_mask[0] = 0xff;
-	req.l2_addr_mask[1] = 0xff;
-	req.l2_addr_mask[2] = 0xff;
-	req.l2_addr_mask[3] = 0xff;
-	req.l2_addr_mask[4] = 0xff;
-	req.l2_addr_mask[5] = 0xff;
-
-	mutex_lock(&bp->hwrm_cmd_lock);
-	rc = _hwrm_send_message(bp, &req, sizeof(req), HWRM_CMD_TIMEOUT);
-	if (!rc)
-		bp->vnic_info[vnic_id].fw_l2_filter_id[idx] =
-							resp->l2_filter_id;
-	mutex_unlock(&bp->hwrm_cmd_lock);
+	ether_addr_copy(key.dst_mac_addr, mac_addr);
+	key.vlan = 0;
+	fltr = bnxt_alloc_l2_filter(bp, &key, GFP_KERNEL);
+	if (IS_ERR(fltr))
+		return PTR_ERR(fltr);
+
+	fltr->base.fw_vnic_id = bp->vnic_info[vnic_id].fw_vnic_id;
+	rc = bnxt_hwrm_l2_filter_alloc(bp, fltr);
+	if (rc)
+		bnxt_del_l2_filter(bp, fltr);
+	else
+		bp->vnic_info[vnic_id].l2_filters[idx] = fltr;
 	return rc;
 }
 
-static int bnxt_hwrm_clear_vnic_filter(struct bnxt *bp)
+static void bnxt_hwrm_clear_vnic_filter(struct bnxt *bp)
 {
 	u16 i, j, num_of_vnics = 1; /* only vnic 0 supported */
-	int rc = 0;
 
 	/* Any associated ntuple filters will also be cleared by firmware. */
-	mutex_lock(&bp->hwrm_cmd_lock);
 	for (i = 0; i < num_of_vnics; i++) {
 		struct bnxt_vnic_info *vnic = &bp->vnic_info[i];
 
 		for (j = 0; j < vnic->uc_filter_count; j++) {
-			struct hwrm_cfa_l2_filter_free_input req = {0};
-
-			bnxt_hwrm_cmd_hdr_init(bp, &req,
-					       HWRM_CFA_L2_FILTER_FREE, -1, -1);
-
-			req.l2_filter_id = vnic->fw_l2_filter_id[j];
+			struct bnxt_l2_filter *fltr = vnic->l2_filters[j];
 
-			rc = _hwrm_send_message(bp, &req, sizeof(req),
-						HWRM_CMD_TIMEOUT);
+			bnxt_hwrm_l2_filter_free(bp, fltr);
+			bnxt_del_l2_filter(bp, fltr);
 		}
 		vnic->uc_filter_count = 0;
 	}
-	mutex_unlock(&bp->hwrm_cmd_lock);
+}
 
-	return rc;
+#define BNXT_DFLT_TUNL_TPA_BMAP				\
+	(VNIC_TPA_CFG_REQ_TNL_TPA_EN_BITMAP_GRE |	\
+	 VNIC_TPA_CFG_REQ_TNL_TPA_EN_BITMAP_IPV4 |	\
+	 VNIC_TPA_CFG_REQ_TNL_TPA_EN_BITMAP_IPV6)
+
+static void bnxt_hwrm_vnic_update_tunl_tpa(struct bnxt *bp,
+					   struct hwrm_vnic_tpa_cfg_input *req)
+{
+	u32 tunl_tpa_bmap = BNXT_DFLT_TUNL_TPA_BMAP;
+
+	if (!(bp->fw_cap & BNXT_FW_CAP_VNIC_TUNNEL_TPA))
+		return;
+
+	if (bp->vxlan_port)
+		tunl_tpa_bmap |= VNIC_TPA_CFG_REQ_TNL_TPA_EN_BITMAP_VXLAN;
+	if (bp->vxlan_gpe_port)
+		tunl_tpa_bmap |= VNIC_TPA_CFG_REQ_TNL_TPA_EN_BITMAP_VXLAN_GPE;
+	if (bp->nge_port)
+		tunl_tpa_bmap |= VNIC_TPA_CFG_REQ_TNL_TPA_EN_BITMAP_GENEVE;
+
+	req->enables |= cpu_to_le32(VNIC_TPA_CFG_REQ_ENABLES_TNL_TPA_EN);
+	req->tnl_tpa_en_bitmap = cpu_to_le32(tunl_tpa_bmap);
 }
 
-static int bnxt_hwrm_vnic_set_tpa(struct bnxt *bp, u16 vnic_id, u32 tpa_flags)
+int bnxt_hwrm_vnic_set_tpa(struct bnxt *bp, struct bnxt_vnic_info *vnic,
+			   u32 tpa_flags)
 {
-	struct bnxt_vnic_info *vnic = &bp->vnic_info[vnic_id];
-	struct hwrm_vnic_tpa_cfg_input req = {0};
+	u16 max_aggs = VNIC_TPA_CFG_REQ_MAX_AGGS_MAX;
+	struct hwrm_vnic_tpa_cfg_input *req;
+	int rc;
 
-	bnxt_hwrm_cmd_hdr_init(bp, &req, HWRM_VNIC_TPA_CFG, -1, -1);
+	if (vnic->fw_vnic_id == INVALID_HW_RING_ID)
+		return 0;
+
+	rc = hwrm_req_init(bp, req, HWRM_VNIC_TPA_CFG);
+	if (rc)
+		return rc;
 
 	if (tpa_flags) {
 		u16 mss = bp->dev->mtu - 40;
@@ -3688,9 +6422,9 @@ static int bnxt_hwrm_vnic_set_tpa(struct bnxt *bp, u16 vnic_id, u32 tpa_flags)
 		if (tpa_flags & BNXT_FLAG_GRO)
 			flags |= VNIC_TPA_CFG_REQ_FLAGS_GRO;
 
-		req.flags = cpu_to_le32(flags);
+		req->flags = cpu_to_le32(flags);
 
-		req.enables =
+		req->enables =
 			cpu_to_le32(VNIC_TPA_CFG_REQ_ENABLES_MAX_AGG_SEGS |
 				    VNIC_TPA_CFG_REQ_ENABLES_MAX_AGGS |
 				    VNIC_TPA_CFG_REQ_ENABLES_MIN_AGG_LEN);
@@ -3708,83 +6442,299 @@ static int bnxt_hwrm_vnic_set_tpa(struct bnxt *bp, u16 vnic_id, u32 tpa_flags)
 			nsegs = (MAX_SKB_FRAGS - n) / n;
 		}
 
-		segs = ilog2(nsegs);
-		req.max_agg_segs = cpu_to_le16(segs);
-		req.max_aggs = cpu_to_le16(VNIC_TPA_CFG_REQ_MAX_AGGS_MAX);
+		if (bp->flags & BNXT_FLAG_CHIP_P5_PLUS) {
+			segs = MAX_TPA_SEGS_P5;
+			max_aggs = bp->max_tpa;
+		} else {
+			segs = ilog2(nsegs);
+		}
+		req->max_agg_segs = cpu_to_le16(segs);
+		req->max_aggs = cpu_to_le16(max_aggs);
 
-		req.min_agg_len = cpu_to_le32(512);
+		req->min_agg_len = cpu_to_le32(512);
+		bnxt_hwrm_vnic_update_tunl_tpa(bp, req);
 	}
-	req.vnic_id = cpu_to_le16(vnic->fw_vnic_id);
+	req->vnic_id = cpu_to_le16(vnic->fw_vnic_id);
 
-	return hwrm_send_message(bp, &req, sizeof(req), HWRM_CMD_TIMEOUT);
+	return hwrm_req_send(bp, req);
 }
 
-static int bnxt_hwrm_vnic_set_rss(struct bnxt *bp, u16 vnic_id, bool set_rss)
+static u16 bnxt_cp_ring_from_grp(struct bnxt *bp, struct bnxt_ring_struct *ring)
 {
-	u32 i, j, max_rings;
-	struct bnxt_vnic_info *vnic = &bp->vnic_info[vnic_id];
-	struct hwrm_vnic_rss_cfg_input req = {0};
+	struct bnxt_ring_grp_info *grp_info;
+
+	grp_info = &bp->grp_info[ring->grp_idx];
+	return grp_info->cp_fw_ring_id;
+}
+
+static u16 bnxt_cp_ring_for_rx(struct bnxt *bp, struct bnxt_rx_ring_info *rxr)
+{
+	if (bp->flags & BNXT_FLAG_CHIP_P5_PLUS)
+		return rxr->rx_cpr->cp_ring_struct.fw_ring_id;
+	else
+		return bnxt_cp_ring_from_grp(bp, &rxr->rx_ring_struct);
+}
+
+static u16 bnxt_cp_ring_for_tx(struct bnxt *bp, struct bnxt_tx_ring_info *txr)
+{
+	if (bp->flags & BNXT_FLAG_CHIP_P5_PLUS)
+		return txr->tx_cpr->cp_ring_struct.fw_ring_id;
+	else
+		return bnxt_cp_ring_from_grp(bp, &txr->tx_ring_struct);
+}
+
+static int bnxt_alloc_rss_indir_tbl(struct bnxt *bp)
+{
+	int entries;
+
+	if (bp->flags & BNXT_FLAG_CHIP_P5_PLUS)
+		entries = BNXT_MAX_RSS_TABLE_ENTRIES_P5;
+	else
+		entries = HW_HASH_INDEX_SIZE;
+
+	bp->rss_indir_tbl_entries = entries;
+	bp->rss_indir_tbl =
+		kmalloc_array(entries, sizeof(*bp->rss_indir_tbl), GFP_KERNEL);
+	if (!bp->rss_indir_tbl)
+		return -ENOMEM;
+
+	return 0;
+}
+
+void bnxt_set_dflt_rss_indir_tbl(struct bnxt *bp,
+				 struct ethtool_rxfh_context *rss_ctx)
+{
+	u16 max_rings, max_entries, pad, i;
+	u32 *rss_indir_tbl;
 
-	if (vnic->fw_rss_cos_lb_ctx[0] == INVALID_HW_RING_ID)
+	if (!bp->rx_nr_rings)
+		return;
+
+	if (BNXT_CHIP_TYPE_NITRO_A0(bp))
+		max_rings = bp->rx_nr_rings - 1;
+	else
+		max_rings = bp->rx_nr_rings;
+
+	max_entries = bnxt_get_rxfh_indir_size(bp->dev);
+	if (rss_ctx)
+		rss_indir_tbl = ethtool_rxfh_context_indir(rss_ctx);
+	else
+		rss_indir_tbl = &bp->rss_indir_tbl[0];
+
+	for (i = 0; i < max_entries; i++)
+		rss_indir_tbl[i] = ethtool_rxfh_indir_default(i, max_rings);
+
+	pad = bp->rss_indir_tbl_entries - max_entries;
+	if (pad)
+		memset(&rss_indir_tbl[i], 0, pad * sizeof(*rss_indir_tbl));
+}
+
+static u16 bnxt_get_max_rss_ring(struct bnxt *bp)
+{
+	u32 i, tbl_size, max_ring = 0;
+
+	if (!bp->rss_indir_tbl)
 		return 0;
 
-	bnxt_hwrm_cmd_hdr_init(bp, &req, HWRM_VNIC_RSS_CFG, -1, -1);
-	if (set_rss) {
-		req.hash_type = cpu_to_le32(bp->rss_hash_cfg);
-		if (vnic->flags & BNXT_VNIC_RSS_FLAG) {
-			if (BNXT_CHIP_TYPE_NITRO_A0(bp))
-				max_rings = bp->rx_nr_rings - 1;
-			else
-				max_rings = bp->rx_nr_rings;
-		} else {
-			max_rings = 1;
-		}
+	tbl_size = bnxt_get_rxfh_indir_size(bp->dev);
+	for (i = 0; i < tbl_size; i++)
+		max_ring = max(max_ring, bp->rss_indir_tbl[i]);
+	return max_ring;
+}
 
-		/* Fill the RSS indirection table with ring group ids */
-		for (i = 0, j = 0; i < HW_HASH_INDEX_SIZE; i++, j++) {
-			if (j == max_rings)
-				j = 0;
-			vnic->rss_table[i] = cpu_to_le16(vnic->fw_grp_ids[j]);
-		}
+int bnxt_get_nr_rss_ctxs(struct bnxt *bp, int rx_rings)
+{
+	if (bp->flags & BNXT_FLAG_CHIP_P5_PLUS) {
+		if (!rx_rings)
+			return 0;
+		return bnxt_calc_nr_ring_pages(rx_rings - 1,
+					       BNXT_RSS_TABLE_ENTRIES_P5);
+	}
+	if (BNXT_CHIP_TYPE_NITRO_A0(bp))
+		return 2;
+	return 1;
+}
+
+static void bnxt_fill_hw_rss_tbl(struct bnxt *bp, struct bnxt_vnic_info *vnic)
+{
+	bool no_rss = !(vnic->flags & BNXT_VNIC_RSS_FLAG);
+	u16 i, j;
 
-		req.ring_grp_tbl_addr = cpu_to_le64(vnic->rss_table_dma_addr);
-		req.hash_key_tbl_addr =
-			cpu_to_le64(vnic->rss_hash_key_dma_addr);
+	/* Fill the RSS indirection table with ring group ids */
+	for (i = 0, j = 0; i < HW_HASH_INDEX_SIZE; i++) {
+		if (!no_rss)
+			j = bp->rss_indir_tbl[i];
+		vnic->rss_table[i] = cpu_to_le16(vnic->fw_grp_ids[j]);
 	}
-	req.rss_ctx_idx = cpu_to_le16(vnic->fw_rss_cos_lb_ctx[0]);
-	return hwrm_send_message(bp, &req, sizeof(req), HWRM_CMD_TIMEOUT);
 }
 
-static int bnxt_hwrm_vnic_set_hds(struct bnxt *bp, u16 vnic_id)
+static void bnxt_fill_hw_rss_tbl_p5(struct bnxt *bp,
+				    struct bnxt_vnic_info *vnic)
 {
-	struct bnxt_vnic_info *vnic = &bp->vnic_info[vnic_id];
-	struct hwrm_vnic_plcmodes_cfg_input req = {0};
-
-	bnxt_hwrm_cmd_hdr_init(bp, &req, HWRM_VNIC_PLCMODES_CFG, -1, -1);
-	req.flags = cpu_to_le32(VNIC_PLCMODES_CFG_REQ_FLAGS_JUMBO_PLACEMENT |
-				VNIC_PLCMODES_CFG_REQ_FLAGS_HDS_IPV4 |
-				VNIC_PLCMODES_CFG_REQ_FLAGS_HDS_IPV6);
-	req.enables =
-		cpu_to_le32(VNIC_PLCMODES_CFG_REQ_ENABLES_JUMBO_THRESH_VALID |
-			    VNIC_PLCMODES_CFG_REQ_ENABLES_HDS_THRESHOLD_VALID);
-	/* thresholds not implemented in firmware yet */
-	req.jumbo_thresh = cpu_to_le16(bp->rx_copy_thresh);
-	req.hds_threshold = cpu_to_le16(bp->rx_copy_thresh);
-	req.vnic_id = cpu_to_le32(vnic->fw_vnic_id);
-	return hwrm_send_message(bp, &req, sizeof(req), HWRM_CMD_TIMEOUT);
-}
-
-static void bnxt_hwrm_vnic_ctx_free_one(struct bnxt *bp, u16 vnic_id,
+	__le16 *ring_tbl = vnic->rss_table;
+	struct bnxt_rx_ring_info *rxr;
+	u16 tbl_size, i;
+
+	tbl_size = bnxt_get_rxfh_indir_size(bp->dev);
+
+	for (i = 0; i < tbl_size; i++) {
+		u16 ring_id, j;
+
+		if (vnic->flags & BNXT_VNIC_NTUPLE_FLAG)
+			j = ethtool_rxfh_indir_default(i, bp->rx_nr_rings);
+		else if (vnic->flags & BNXT_VNIC_RSSCTX_FLAG)
+			j = ethtool_rxfh_context_indir(vnic->rss_ctx)[i];
+		else
+			j = bp->rss_indir_tbl[i];
+		rxr = &bp->rx_ring[j];
+
+		ring_id = rxr->rx_ring_struct.fw_ring_id;
+		*ring_tbl++ = cpu_to_le16(ring_id);
+		ring_id = bnxt_cp_ring_for_rx(bp, rxr);
+		*ring_tbl++ = cpu_to_le16(ring_id);
+	}
+}
+
+static void
+__bnxt_hwrm_vnic_set_rss(struct bnxt *bp, struct hwrm_vnic_rss_cfg_input *req,
+			 struct bnxt_vnic_info *vnic)
+{
+	if (bp->flags & BNXT_FLAG_CHIP_P5_PLUS) {
+		bnxt_fill_hw_rss_tbl_p5(bp, vnic);
+		if (bp->flags & BNXT_FLAG_CHIP_P7)
+			req->flags |= VNIC_RSS_CFG_REQ_FLAGS_IPSEC_HASH_TYPE_CFG_SUPPORT;
+	} else {
+		bnxt_fill_hw_rss_tbl(bp, vnic);
+	}
+
+	if (bp->rss_hash_delta) {
+		req->hash_type = cpu_to_le32(bp->rss_hash_delta);
+		if (bp->rss_hash_cfg & bp->rss_hash_delta)
+			req->flags |= VNIC_RSS_CFG_REQ_FLAGS_HASH_TYPE_INCLUDE;
+		else
+			req->flags |= VNIC_RSS_CFG_REQ_FLAGS_HASH_TYPE_EXCLUDE;
+	} else {
+		req->hash_type = cpu_to_le32(bp->rss_hash_cfg);
+	}
+	req->hash_mode_flags = VNIC_RSS_CFG_REQ_HASH_MODE_FLAGS_DEFAULT;
+	req->ring_grp_tbl_addr = cpu_to_le64(vnic->rss_table_dma_addr);
+	req->hash_key_tbl_addr = cpu_to_le64(vnic->rss_hash_key_dma_addr);
+}
+
+static int bnxt_hwrm_vnic_set_rss(struct bnxt *bp, struct bnxt_vnic_info *vnic,
+				  bool set_rss)
+{
+	struct hwrm_vnic_rss_cfg_input *req;
+	int rc;
+
+	if ((bp->flags & BNXT_FLAG_CHIP_P5_PLUS) ||
+	    vnic->fw_rss_cos_lb_ctx[0] == INVALID_HW_RING_ID)
+		return 0;
+
+	rc = hwrm_req_init(bp, req, HWRM_VNIC_RSS_CFG);
+	if (rc)
+		return rc;
+
+	if (set_rss)
+		__bnxt_hwrm_vnic_set_rss(bp, req, vnic);
+	req->rss_ctx_idx = cpu_to_le16(vnic->fw_rss_cos_lb_ctx[0]);
+	return hwrm_req_send(bp, req);
+}
+
+static int bnxt_hwrm_vnic_set_rss_p5(struct bnxt *bp,
+				     struct bnxt_vnic_info *vnic, bool set_rss)
+{
+	struct hwrm_vnic_rss_cfg_input *req;
+	dma_addr_t ring_tbl_map;
+	u32 i, nr_ctxs;
+	int rc;
+
+	rc = hwrm_req_init(bp, req, HWRM_VNIC_RSS_CFG);
+	if (rc)
+		return rc;
+
+	req->vnic_id = cpu_to_le16(vnic->fw_vnic_id);
+	if (!set_rss)
+		return hwrm_req_send(bp, req);
+
+	__bnxt_hwrm_vnic_set_rss(bp, req, vnic);
+	ring_tbl_map = vnic->rss_table_dma_addr;
+	nr_ctxs = bnxt_get_nr_rss_ctxs(bp, bp->rx_nr_rings);
+
+	hwrm_req_hold(bp, req);
+	for (i = 0; i < nr_ctxs; ring_tbl_map += BNXT_RSS_TABLE_SIZE_P5, i++) {
+		req->ring_grp_tbl_addr = cpu_to_le64(ring_tbl_map);
+		req->ring_table_pair_index = i;
+		req->rss_ctx_idx = cpu_to_le16(vnic->fw_rss_cos_lb_ctx[i]);
+		rc = hwrm_req_send(bp, req);
+		if (rc)
+			goto exit;
+	}
+
+exit:
+	hwrm_req_drop(bp, req);
+	return rc;
+}
+
+static void bnxt_hwrm_update_rss_hash_cfg(struct bnxt *bp)
+{
+	struct bnxt_vnic_info *vnic = &bp->vnic_info[BNXT_VNIC_DEFAULT];
+	struct hwrm_vnic_rss_qcfg_output *resp;
+	struct hwrm_vnic_rss_qcfg_input *req;
+
+	if (hwrm_req_init(bp, req, HWRM_VNIC_RSS_QCFG))
+		return;
+
+	req->vnic_id = cpu_to_le16(vnic->fw_vnic_id);
+	/* all contexts configured to same hash_type, zero always exists */
+	req->rss_ctx_idx = cpu_to_le16(vnic->fw_rss_cos_lb_ctx[0]);
+	resp = hwrm_req_hold(bp, req);
+	if (!hwrm_req_send(bp, req)) {
+		bp->rss_hash_cfg = le32_to_cpu(resp->hash_type) ?: bp->rss_hash_cfg;
+		bp->rss_hash_delta = 0;
+	}
+	hwrm_req_drop(bp, req);
+}
+
+static int bnxt_hwrm_vnic_set_hds(struct bnxt *bp, struct bnxt_vnic_info *vnic)
+{
+	u16 hds_thresh = (u16)bp->dev->cfg_pending->hds_thresh;
+	struct hwrm_vnic_plcmodes_cfg_input *req;
+	int rc;
+
+	rc = hwrm_req_init(bp, req, HWRM_VNIC_PLCMODES_CFG);
+	if (rc)
+		return rc;
+
+	req->flags = cpu_to_le32(VNIC_PLCMODES_CFG_REQ_FLAGS_JUMBO_PLACEMENT);
+	req->enables = cpu_to_le32(VNIC_PLCMODES_CFG_REQ_ENABLES_JUMBO_THRESH_VALID);
+	req->jumbo_thresh = cpu_to_le16(bp->rx_buf_use_size);
+
+	if (!BNXT_RX_PAGE_MODE(bp) && (bp->flags & BNXT_FLAG_AGG_RINGS)) {
+		req->flags |= cpu_to_le32(VNIC_PLCMODES_CFG_REQ_FLAGS_HDS_IPV4 |
+					  VNIC_PLCMODES_CFG_REQ_FLAGS_HDS_IPV6);
+		req->enables |=
+			cpu_to_le32(VNIC_PLCMODES_CFG_REQ_ENABLES_HDS_THRESHOLD_VALID);
+		req->hds_threshold = cpu_to_le16(hds_thresh);
+	}
+	req->vnic_id = cpu_to_le32(vnic->fw_vnic_id);
+	return hwrm_req_send(bp, req);
+}
+
+static void bnxt_hwrm_vnic_ctx_free_one(struct bnxt *bp,
+					struct bnxt_vnic_info *vnic,
 					u16 ctx_idx)
 {
-	struct hwrm_vnic_rss_cos_lb_ctx_free_input req = {0};
+	struct hwrm_vnic_rss_cos_lb_ctx_free_input *req;
+
+	if (hwrm_req_init(bp, req, HWRM_VNIC_RSS_COS_LB_CTX_FREE))
+		return;
 
-	bnxt_hwrm_cmd_hdr_init(bp, &req, HWRM_VNIC_RSS_COS_LB_CTX_FREE, -1, -1);
-	req.rss_cos_lb_ctx_id =
-		cpu_to_le16(bp->vnic_info[vnic_id].fw_rss_cos_lb_ctx[ctx_idx]);
+	req->rss_cos_lb_ctx_id =
+		cpu_to_le16(vnic->fw_rss_cos_lb_ctx[ctx_idx]);
 
-	hwrm_send_message(bp, &req, sizeof(req), HWRM_CMD_TIMEOUT);
-	bp->vnic_info[vnic_id].fw_rss_cos_lb_ctx[ctx_idx] = INVALID_HW_RING_ID;
+	hwrm_req_send(bp, req);
+	vnic->fw_rss_cos_lb_ctx[ctx_idx] = INVALID_HW_RING_ID;
 }
 
 static void bnxt_hwrm_vnic_ctx_free(struct bnxt *bp)
@@ -3796,110 +6746,128 @@ static void bnxt_hwrm_vnic_ctx_free(struct bnxt *bp)
 
 		for (j = 0; j < BNXT_MAX_CTX_PER_VNIC; j++) {
 			if (vnic->fw_rss_cos_lb_ctx[j] != INVALID_HW_RING_ID)
-				bnxt_hwrm_vnic_ctx_free_one(bp, i, j);
+				bnxt_hwrm_vnic_ctx_free_one(bp, vnic, j);
 		}
 	}
 	bp->rsscos_nr_ctxs = 0;
 }
 
-static int bnxt_hwrm_vnic_ctx_alloc(struct bnxt *bp, u16 vnic_id, u16 ctx_idx)
+static int bnxt_hwrm_vnic_ctx_alloc(struct bnxt *bp,
+				    struct bnxt_vnic_info *vnic, u16 ctx_idx)
 {
+	struct hwrm_vnic_rss_cos_lb_ctx_alloc_output *resp;
+	struct hwrm_vnic_rss_cos_lb_ctx_alloc_input *req;
 	int rc;
-	struct hwrm_vnic_rss_cos_lb_ctx_alloc_input req = {0};
-	struct hwrm_vnic_rss_cos_lb_ctx_alloc_output *resp =
-						bp->hwrm_cmd_resp_addr;
 
-	bnxt_hwrm_cmd_hdr_init(bp, &req, HWRM_VNIC_RSS_COS_LB_CTX_ALLOC, -1,
-			       -1);
+	rc = hwrm_req_init(bp, req, HWRM_VNIC_RSS_COS_LB_CTX_ALLOC);
+	if (rc)
+		return rc;
 
-	mutex_lock(&bp->hwrm_cmd_lock);
-	rc = _hwrm_send_message(bp, &req, sizeof(req), HWRM_CMD_TIMEOUT);
+	resp = hwrm_req_hold(bp, req);
+	rc = hwrm_req_send(bp, req);
 	if (!rc)
-		bp->vnic_info[vnic_id].fw_rss_cos_lb_ctx[ctx_idx] =
+		vnic->fw_rss_cos_lb_ctx[ctx_idx] =
 			le16_to_cpu(resp->rss_cos_lb_ctx_id);
-	mutex_unlock(&bp->hwrm_cmd_lock);
+	hwrm_req_drop(bp, req);
 
 	return rc;
 }
 
-int bnxt_hwrm_vnic_cfg(struct bnxt *bp, u16 vnic_id)
+static u32 bnxt_get_roce_vnic_mode(struct bnxt *bp)
+{
+	if (bp->flags & BNXT_FLAG_ROCE_MIRROR_CAP)
+		return VNIC_CFG_REQ_FLAGS_ROCE_MIRRORING_CAPABLE_VNIC_MODE;
+	return VNIC_CFG_REQ_FLAGS_ROCE_DUAL_VNIC_MODE;
+}
+
+int bnxt_hwrm_vnic_cfg(struct bnxt *bp, struct bnxt_vnic_info *vnic)
 {
+	struct bnxt_vnic_info *vnic0 = &bp->vnic_info[BNXT_VNIC_DEFAULT];
+	struct hwrm_vnic_cfg_input *req;
 	unsigned int ring = 0, grp_idx;
-	struct bnxt_vnic_info *vnic = &bp->vnic_info[vnic_id];
-	struct hwrm_vnic_cfg_input req = {0};
 	u16 def_vlan = 0;
+	int rc;
 
-	bnxt_hwrm_cmd_hdr_init(bp, &req, HWRM_VNIC_CFG, -1, -1);
+	rc = hwrm_req_init(bp, req, HWRM_VNIC_CFG);
+	if (rc)
+		return rc;
 
-	req.enables = cpu_to_le32(VNIC_CFG_REQ_ENABLES_DFLT_RING_GRP);
+	if (bp->flags & BNXT_FLAG_CHIP_P5_PLUS) {
+		struct bnxt_rx_ring_info *rxr = &bp->rx_ring[0];
+
+		req->default_rx_ring_id =
+			cpu_to_le16(rxr->rx_ring_struct.fw_ring_id);
+		req->default_cmpl_ring_id =
+			cpu_to_le16(bnxt_cp_ring_for_rx(bp, rxr));
+		req->enables =
+			cpu_to_le32(VNIC_CFG_REQ_ENABLES_DEFAULT_RX_RING_ID |
+				    VNIC_CFG_REQ_ENABLES_DEFAULT_CMPL_RING_ID);
+		goto vnic_mru;
+	}
+	req->enables = cpu_to_le32(VNIC_CFG_REQ_ENABLES_DFLT_RING_GRP);
 	/* Only RSS support for now TBD: COS & LB */
 	if (vnic->fw_rss_cos_lb_ctx[0] != INVALID_HW_RING_ID) {
-		req.rss_rule = cpu_to_le16(vnic->fw_rss_cos_lb_ctx[0]);
-		req.enables |= cpu_to_le32(VNIC_CFG_REQ_ENABLES_RSS_RULE |
+		req->rss_rule = cpu_to_le16(vnic->fw_rss_cos_lb_ctx[0]);
+		req->enables |= cpu_to_le32(VNIC_CFG_REQ_ENABLES_RSS_RULE |
 					   VNIC_CFG_REQ_ENABLES_MRU);
 	} else if (vnic->flags & BNXT_VNIC_RFS_NEW_RSS_FLAG) {
-		req.rss_rule =
-			cpu_to_le16(bp->vnic_info[0].fw_rss_cos_lb_ctx[0]);
-		req.enables |= cpu_to_le32(VNIC_CFG_REQ_ENABLES_RSS_RULE |
+		req->rss_rule = cpu_to_le16(vnic0->fw_rss_cos_lb_ctx[0]);
+		req->enables |= cpu_to_le32(VNIC_CFG_REQ_ENABLES_RSS_RULE |
 					   VNIC_CFG_REQ_ENABLES_MRU);
-		req.flags |= cpu_to_le32(VNIC_CFG_REQ_FLAGS_RSS_DFLT_CR_MODE);
+		req->flags |= cpu_to_le32(VNIC_CFG_REQ_FLAGS_RSS_DFLT_CR_MODE);
 	} else {
-		req.rss_rule = cpu_to_le16(0xffff);
+		req->rss_rule = cpu_to_le16(0xffff);
 	}
 
 	if (BNXT_CHIP_TYPE_NITRO_A0(bp) &&
 	    (vnic->fw_rss_cos_lb_ctx[0] != INVALID_HW_RING_ID)) {
-		req.cos_rule = cpu_to_le16(vnic->fw_rss_cos_lb_ctx[1]);
-		req.enables |= cpu_to_le32(VNIC_CFG_REQ_ENABLES_COS_RULE);
+		req->cos_rule = cpu_to_le16(vnic->fw_rss_cos_lb_ctx[1]);
+		req->enables |= cpu_to_le32(VNIC_CFG_REQ_ENABLES_COS_RULE);
 	} else {
-		req.cos_rule = cpu_to_le16(0xffff);
+		req->cos_rule = cpu_to_le16(0xffff);
 	}
 
 	if (vnic->flags & BNXT_VNIC_RSS_FLAG)
 		ring = 0;
 	else if (vnic->flags & BNXT_VNIC_RFS_FLAG)
-		ring = vnic_id - 1;
-	else if ((vnic_id == 1) && BNXT_CHIP_TYPE_NITRO_A0(bp))
+		ring = vnic->vnic_id - 1;
+	else if ((vnic->vnic_id == 1) && BNXT_CHIP_TYPE_NITRO_A0(bp))
 		ring = bp->rx_nr_rings - 1;
 
 	grp_idx = bp->rx_ring[ring].bnapi->index;
-	req.vnic_id = cpu_to_le16(vnic->fw_vnic_id);
-	req.dflt_ring_grp = cpu_to_le16(bp->grp_info[grp_idx].fw_grp_id);
-
-	req.lb_rule = cpu_to_le16(0xffff);
-	req.mru = cpu_to_le16(bp->dev->mtu + ETH_HLEN + ETH_FCS_LEN +
-			      VLAN_HLEN);
+	req->dflt_ring_grp = cpu_to_le16(bp->grp_info[grp_idx].fw_grp_id);
+	req->lb_rule = cpu_to_le16(0xffff);
+vnic_mru:
+	vnic->mru = bp->dev->mtu + VLAN_ETH_HLEN;
+	req->mru = cpu_to_le16(vnic->mru);
 
+	req->vnic_id = cpu_to_le16(vnic->fw_vnic_id);
 #ifdef CONFIG_BNXT_SRIOV
 	if (BNXT_VF(bp))
 		def_vlan = bp->vf.vlan;
 #endif
 	if ((bp->flags & BNXT_FLAG_STRIP_VLAN) || def_vlan)
-		req.flags |= cpu_to_le32(VNIC_CFG_REQ_FLAGS_VLAN_STRIP_MODE);
-	if (!vnic_id && bnxt_ulp_registered(bp->edev, BNXT_ROCE_ULP))
-		req.flags |=
-			cpu_to_le32(VNIC_CFG_REQ_FLAGS_ROCE_DUAL_VNIC_MODE);
+		req->flags |= cpu_to_le32(VNIC_CFG_REQ_FLAGS_VLAN_STRIP_MODE);
+	if (vnic->vnic_id == BNXT_VNIC_DEFAULT && bnxt_ulp_registered(bp->edev))
+		req->flags |= cpu_to_le32(bnxt_get_roce_vnic_mode(bp));
 
-	return hwrm_send_message(bp, &req, sizeof(req), HWRM_CMD_TIMEOUT);
+	return hwrm_req_send(bp, req);
 }
 
-static int bnxt_hwrm_vnic_free_one(struct bnxt *bp, u16 vnic_id)
+static void bnxt_hwrm_vnic_free_one(struct bnxt *bp,
+				    struct bnxt_vnic_info *vnic)
 {
-	u32 rc = 0;
+	if (vnic->fw_vnic_id != INVALID_HW_RING_ID) {
+		struct hwrm_vnic_free_input *req;
 
-	if (bp->vnic_info[vnic_id].fw_vnic_id != INVALID_HW_RING_ID) {
-		struct hwrm_vnic_free_input req = {0};
+		if (hwrm_req_init(bp, req, HWRM_VNIC_FREE))
+			return;
 
-		bnxt_hwrm_cmd_hdr_init(bp, &req, HWRM_VNIC_FREE, -1, -1);
-		req.vnic_id =
-			cpu_to_le32(bp->vnic_info[vnic_id].fw_vnic_id);
+		req->vnic_id = cpu_to_le32(vnic->fw_vnic_id);
 
-		rc = hwrm_send_message(bp, &req, sizeof(req), HWRM_CMD_TIMEOUT);
-		if (rc)
-			return rc;
-		bp->vnic_info[vnic_id].fw_vnic_id = INVALID_HW_RING_ID;
+		hwrm_req_send(bp, req);
+		vnic->fw_vnic_id = INVALID_HW_RING_ID;
 	}
-	return rc;
 }
 
 static void bnxt_hwrm_vnic_free(struct bnxt *bp)
@@ -3907,17 +6875,24 @@ static void bnxt_hwrm_vnic_free(struct bnxt *bp)
 	u16 i;
 
 	for (i = 0; i < bp->nr_vnics; i++)
-		bnxt_hwrm_vnic_free_one(bp, i);
+		bnxt_hwrm_vnic_free_one(bp, &bp->vnic_info[i]);
 }
 
-static int bnxt_hwrm_vnic_alloc(struct bnxt *bp, u16 vnic_id,
-				unsigned int start_rx_ring_idx,
-				unsigned int nr_rings)
+int bnxt_hwrm_vnic_alloc(struct bnxt *bp, struct bnxt_vnic_info *vnic,
+			 unsigned int start_rx_ring_idx,
+			 unsigned int nr_rings)
 {
-	int rc = 0;
 	unsigned int i, j, grp_idx, end_idx = start_rx_ring_idx + nr_rings;
-	struct hwrm_vnic_alloc_input req = {0};
-	struct hwrm_vnic_alloc_output *resp = bp->hwrm_cmd_resp_addr;
+	struct hwrm_vnic_alloc_output *resp;
+	struct hwrm_vnic_alloc_input *req;
+	int rc;
+
+	rc = hwrm_req_init(bp, req, HWRM_VNIC_ALLOC);
+	if (rc)
+		return rc;
+
+	if (bp->flags & BNXT_FLAG_CHIP_P5_PLUS)
+		goto vnic_no_ring_grps;
 
 	/* map ring groups to this vnic */
 	for (i = start_rx_ring_idx, j = 0; i < end_idx; i++, j++) {
@@ -3927,187 +6902,266 @@ static int bnxt_hwrm_vnic_alloc(struct bnxt *bp, u16 vnic_id,
 				   j, nr_rings);
 			break;
 		}
-		bp->vnic_info[vnic_id].fw_grp_ids[j] =
-					bp->grp_info[grp_idx].fw_grp_id;
+		vnic->fw_grp_ids[j] = bp->grp_info[grp_idx].fw_grp_id;
 	}
 
-	bp->vnic_info[vnic_id].fw_rss_cos_lb_ctx[0] = INVALID_HW_RING_ID;
-	bp->vnic_info[vnic_id].fw_rss_cos_lb_ctx[1] = INVALID_HW_RING_ID;
-	if (vnic_id == 0)
-		req.flags = cpu_to_le32(VNIC_ALLOC_REQ_FLAGS_DEFAULT);
-
-	bnxt_hwrm_cmd_hdr_init(bp, &req, HWRM_VNIC_ALLOC, -1, -1);
+vnic_no_ring_grps:
+	for (i = 0; i < BNXT_MAX_CTX_PER_VNIC; i++)
+		vnic->fw_rss_cos_lb_ctx[i] = INVALID_HW_RING_ID;
+	if (vnic->vnic_id == BNXT_VNIC_DEFAULT)
+		req->flags = cpu_to_le32(VNIC_ALLOC_REQ_FLAGS_DEFAULT);
 
-	mutex_lock(&bp->hwrm_cmd_lock);
-	rc = _hwrm_send_message(bp, &req, sizeof(req), HWRM_CMD_TIMEOUT);
+	resp = hwrm_req_hold(bp, req);
+	rc = hwrm_req_send(bp, req);
 	if (!rc)
-		bp->vnic_info[vnic_id].fw_vnic_id = le32_to_cpu(resp->vnic_id);
-	mutex_unlock(&bp->hwrm_cmd_lock);
+		vnic->fw_vnic_id = le32_to_cpu(resp->vnic_id);
+	hwrm_req_drop(bp, req);
 	return rc;
 }
 
 static int bnxt_hwrm_vnic_qcaps(struct bnxt *bp)
 {
-	struct hwrm_vnic_qcaps_output *resp = bp->hwrm_cmd_resp_addr;
-	struct hwrm_vnic_qcaps_input req = {0};
+	struct hwrm_vnic_qcaps_output *resp;
+	struct hwrm_vnic_qcaps_input *req;
 	int rc;
 
+	bp->hw_ring_stats_size = sizeof(struct ctx_hw_stats);
+	bp->flags &= ~BNXT_FLAG_ROCE_MIRROR_CAP;
+	bp->rss_cap &= ~BNXT_RSS_CAP_NEW_RSS_CAP;
 	if (bp->hwrm_spec_code < 0x10600)
 		return 0;
 
-	bnxt_hwrm_cmd_hdr_init(bp, &req, HWRM_VNIC_QCAPS, -1, -1);
-	mutex_lock(&bp->hwrm_cmd_lock);
-	rc = _hwrm_send_message(bp, &req, sizeof(req), HWRM_CMD_TIMEOUT);
+	rc = hwrm_req_init(bp, req, HWRM_VNIC_QCAPS);
+	if (rc)
+		return rc;
+
+	resp = hwrm_req_hold(bp, req);
+	rc = hwrm_req_send(bp, req);
 	if (!rc) {
-		if (resp->flags &
-		    cpu_to_le32(VNIC_QCAPS_RESP_FLAGS_RSS_DFLT_CR_CAP))
-			bp->flags |= BNXT_FLAG_NEW_RSS_CAP;
-	}
-	mutex_unlock(&bp->hwrm_cmd_lock);
+		u32 flags = le32_to_cpu(resp->flags);
+
+		if (!(bp->flags & BNXT_FLAG_CHIP_P5_PLUS) &&
+		    (flags & VNIC_QCAPS_RESP_FLAGS_RSS_DFLT_CR_CAP))
+			bp->rss_cap |= BNXT_RSS_CAP_NEW_RSS_CAP;
+		if (flags &
+		    VNIC_QCAPS_RESP_FLAGS_ROCE_MIRRORING_CAPABLE_VNIC_CAP)
+			bp->flags |= BNXT_FLAG_ROCE_MIRROR_CAP;
+
+		/* Older P5 fw before EXT_HW_STATS support did not set
+		 * VLAN_STRIP_CAP properly.
+		 */
+		if ((flags & VNIC_QCAPS_RESP_FLAGS_VLAN_STRIP_CAP) ||
+		    (BNXT_CHIP_P5(bp) &&
+		     !(bp->fw_cap & BNXT_FW_CAP_EXT_HW_STATS_SUPPORTED)))
+			bp->fw_cap |= BNXT_FW_CAP_VLAN_RX_STRIP;
+		if (flags & VNIC_QCAPS_RESP_FLAGS_RSS_HASH_TYPE_DELTA_CAP)
+			bp->rss_cap |= BNXT_RSS_CAP_RSS_HASH_TYPE_DELTA;
+		if (flags & VNIC_QCAPS_RESP_FLAGS_RSS_PROF_TCAM_MODE_ENABLED)
+			bp->rss_cap |= BNXT_RSS_CAP_RSS_TCAM;
+		bp->max_tpa_v2 = le16_to_cpu(resp->max_aggs_supported);
+		if (bp->max_tpa_v2) {
+			if (BNXT_CHIP_P5(bp))
+				bp->hw_ring_stats_size = BNXT_RING_STATS_SIZE_P5;
+			else
+				bp->hw_ring_stats_size = BNXT_RING_STATS_SIZE_P7;
+		}
+		if (flags & VNIC_QCAPS_RESP_FLAGS_HW_TUNNEL_TPA_CAP)
+			bp->fw_cap |= BNXT_FW_CAP_VNIC_TUNNEL_TPA;
+		if (flags & VNIC_QCAPS_RESP_FLAGS_RSS_IPSEC_AH_SPI_IPV4_CAP)
+			bp->rss_cap |= BNXT_RSS_CAP_AH_V4_RSS_CAP;
+		if (flags & VNIC_QCAPS_RESP_FLAGS_RSS_IPSEC_AH_SPI_IPV6_CAP)
+			bp->rss_cap |= BNXT_RSS_CAP_AH_V6_RSS_CAP;
+		if (flags & VNIC_QCAPS_RESP_FLAGS_RSS_IPSEC_ESP_SPI_IPV4_CAP)
+			bp->rss_cap |= BNXT_RSS_CAP_ESP_V4_RSS_CAP;
+		if (flags & VNIC_QCAPS_RESP_FLAGS_RSS_IPSEC_ESP_SPI_IPV6_CAP)
+			bp->rss_cap |= BNXT_RSS_CAP_ESP_V6_RSS_CAP;
+		if (flags & VNIC_QCAPS_RESP_FLAGS_RSS_IPV6_FLOW_LABEL_CAP)
+			bp->rss_cap |= BNXT_RSS_CAP_IPV6_FLOW_LABEL_RSS_CAP;
+		if (flags & VNIC_QCAPS_RESP_FLAGS_RE_FLUSH_CAP)
+			bp->fw_cap |= BNXT_FW_CAP_VNIC_RE_FLUSH;
+	}
+	hwrm_req_drop(bp, req);
 	return rc;
 }
 
 static int bnxt_hwrm_ring_grp_alloc(struct bnxt *bp)
 {
+	struct hwrm_ring_grp_alloc_output *resp;
+	struct hwrm_ring_grp_alloc_input *req;
+	int rc;
 	u16 i;
-	u32 rc = 0;
 
-	mutex_lock(&bp->hwrm_cmd_lock);
+	if (bp->flags & BNXT_FLAG_CHIP_P5_PLUS)
+		return 0;
+
+	rc = hwrm_req_init(bp, req, HWRM_RING_GRP_ALLOC);
+	if (rc)
+		return rc;
+
+	resp = hwrm_req_hold(bp, req);
 	for (i = 0; i < bp->rx_nr_rings; i++) {
-		struct hwrm_ring_grp_alloc_input req = {0};
-		struct hwrm_ring_grp_alloc_output *resp =
-					bp->hwrm_cmd_resp_addr;
 		unsigned int grp_idx = bp->rx_ring[i].bnapi->index;
 
-		bnxt_hwrm_cmd_hdr_init(bp, &req, HWRM_RING_GRP_ALLOC, -1, -1);
+		req->cr = cpu_to_le16(bp->grp_info[grp_idx].cp_fw_ring_id);
+		req->rr = cpu_to_le16(bp->grp_info[grp_idx].rx_fw_ring_id);
+		req->ar = cpu_to_le16(bp->grp_info[grp_idx].agg_fw_ring_id);
+		req->sc = cpu_to_le16(bp->grp_info[grp_idx].fw_stats_ctx);
 
-		req.cr = cpu_to_le16(bp->grp_info[grp_idx].cp_fw_ring_id);
-		req.rr = cpu_to_le16(bp->grp_info[grp_idx].rx_fw_ring_id);
-		req.ar = cpu_to_le16(bp->grp_info[grp_idx].agg_fw_ring_id);
-		req.sc = cpu_to_le16(bp->grp_info[grp_idx].fw_stats_ctx);
+		rc = hwrm_req_send(bp, req);
 
-		rc = _hwrm_send_message(bp, &req, sizeof(req),
-					HWRM_CMD_TIMEOUT);
 		if (rc)
 			break;
 
 		bp->grp_info[grp_idx].fw_grp_id =
 			le32_to_cpu(resp->ring_group_id);
 	}
-	mutex_unlock(&bp->hwrm_cmd_lock);
+	hwrm_req_drop(bp, req);
 	return rc;
 }
 
-static int bnxt_hwrm_ring_grp_free(struct bnxt *bp)
+static void bnxt_hwrm_ring_grp_free(struct bnxt *bp)
 {
+	struct hwrm_ring_grp_free_input *req;
 	u16 i;
-	u32 rc = 0;
-	struct hwrm_ring_grp_free_input req = {0};
 
-	if (!bp->grp_info)
-		return 0;
+	if (!bp->grp_info || (bp->flags & BNXT_FLAG_CHIP_P5_PLUS))
+		return;
 
-	bnxt_hwrm_cmd_hdr_init(bp, &req, HWRM_RING_GRP_FREE, -1, -1);
+	if (hwrm_req_init(bp, req, HWRM_RING_GRP_FREE))
+		return;
 
-	mutex_lock(&bp->hwrm_cmd_lock);
+	hwrm_req_hold(bp, req);
 	for (i = 0; i < bp->cp_nr_rings; i++) {
 		if (bp->grp_info[i].fw_grp_id == INVALID_HW_RING_ID)
 			continue;
-		req.ring_group_id =
+		req->ring_group_id =
 			cpu_to_le32(bp->grp_info[i].fw_grp_id);
 
-		rc = _hwrm_send_message(bp, &req, sizeof(req),
-					HWRM_CMD_TIMEOUT);
-		if (rc)
-			break;
+		hwrm_req_send(bp, req);
 		bp->grp_info[i].fw_grp_id = INVALID_HW_RING_ID;
 	}
-	mutex_unlock(&bp->hwrm_cmd_lock);
-	return rc;
+	hwrm_req_drop(bp, req);
+}
+
+static void bnxt_set_rx_ring_params_p5(struct bnxt *bp, u32 ring_type,
+				       struct hwrm_ring_alloc_input *req,
+				       struct bnxt_ring_struct *ring)
+{
+	struct bnxt_ring_grp_info *grp_info = &bp->grp_info[ring->grp_idx];
+	u32 enables = RING_ALLOC_REQ_ENABLES_RX_BUF_SIZE_VALID |
+		      RING_ALLOC_REQ_ENABLES_NQ_RING_ID_VALID;
+
+	if (ring_type == HWRM_RING_ALLOC_AGG) {
+		req->ring_type = RING_ALLOC_REQ_RING_TYPE_RX_AGG;
+		req->rx_ring_id = cpu_to_le16(grp_info->rx_fw_ring_id);
+		req->rx_buf_size = cpu_to_le16(BNXT_RX_PAGE_SIZE);
+		enables |= RING_ALLOC_REQ_ENABLES_RX_RING_ID_VALID;
+	} else {
+		req->rx_buf_size = cpu_to_le16(bp->rx_buf_use_size);
+		if (NET_IP_ALIGN == 2)
+			req->flags =
+				cpu_to_le16(RING_ALLOC_REQ_FLAGS_RX_SOP_PAD);
+	}
+	req->stat_ctx_id = cpu_to_le32(grp_info->fw_stats_ctx);
+	req->nq_ring_id = cpu_to_le16(grp_info->cp_fw_ring_id);
+	req->enables |= cpu_to_le32(enables);
 }
 
 static int hwrm_ring_alloc_send_msg(struct bnxt *bp,
 				    struct bnxt_ring_struct *ring,
-				    u32 ring_type, u32 map_index,
-				    u32 stats_ctx_id)
+				    u32 ring_type, u32 map_index)
 {
-	int rc = 0, err = 0;
-	struct hwrm_ring_alloc_input req = {0};
-	struct hwrm_ring_alloc_output *resp = bp->hwrm_cmd_resp_addr;
+	struct hwrm_ring_alloc_output *resp;
+	struct hwrm_ring_alloc_input *req;
+	struct bnxt_ring_mem_info *rmem = &ring->ring_mem;
+	struct bnxt_ring_grp_info *grp_info;
+	int rc, err = 0;
 	u16 ring_id;
 
-	bnxt_hwrm_cmd_hdr_init(bp, &req, HWRM_RING_ALLOC, -1, -1);
+	rc = hwrm_req_init(bp, req, HWRM_RING_ALLOC);
+	if (rc)
+		goto exit;
 
-	req.enables = 0;
-	if (ring->nr_pages > 1) {
-		req.page_tbl_addr = cpu_to_le64(ring->pg_tbl_map);
+	req->enables = 0;
+	if (rmem->nr_pages > 1) {
+		req->page_tbl_addr = cpu_to_le64(rmem->pg_tbl_map);
 		/* Page size is in log2 units */
-		req.page_size = BNXT_PAGE_SHIFT;
-		req.page_tbl_depth = 1;
+		req->page_size = BNXT_PAGE_SHIFT;
+		req->page_tbl_depth = 1;
 	} else {
-		req.page_tbl_addr =  cpu_to_le64(ring->dma_arr[0]);
+		req->page_tbl_addr =  cpu_to_le64(rmem->dma_arr[0]);
 	}
-	req.fbo = 0;
+	req->fbo = 0;
 	/* Association of ring index with doorbell index and MSIX number */
-	req.logical_id = cpu_to_le16(map_index);
+	req->logical_id = cpu_to_le16(map_index);
 
 	switch (ring_type) {
-	case HWRM_RING_ALLOC_TX:
-		req.ring_type = RING_ALLOC_REQ_RING_TYPE_TX;
+	case HWRM_RING_ALLOC_TX: {
+		struct bnxt_tx_ring_info *txr;
+		u16 flags = 0;
+
+		txr = container_of(ring, struct bnxt_tx_ring_info,
+				   tx_ring_struct);
+		req->ring_type = RING_ALLOC_REQ_RING_TYPE_TX;
 		/* Association of transmit ring with completion ring */
-		req.cmpl_ring_id =
-			cpu_to_le16(bp->grp_info[map_index].cp_fw_ring_id);
-		req.length = cpu_to_le32(bp->tx_ring_mask + 1);
-		req.stat_ctx_id = cpu_to_le32(stats_ctx_id);
-		req.queue_id = cpu_to_le16(ring->queue_id);
+		grp_info = &bp->grp_info[ring->grp_idx];
+		req->cmpl_ring_id = cpu_to_le16(bnxt_cp_ring_for_tx(bp, txr));
+		req->length = cpu_to_le32(bp->tx_ring_mask + 1);
+		req->stat_ctx_id = cpu_to_le32(grp_info->fw_stats_ctx);
+		req->queue_id = cpu_to_le16(ring->queue_id);
+		if (bp->flags & BNXT_FLAG_TX_COAL_CMPL)
+			req->cmpl_coal_cnt =
+				RING_ALLOC_REQ_CMPL_COAL_CNT_COAL_64;
+		if ((bp->fw_cap & BNXT_FW_CAP_TX_TS_CMP) && bp->ptp_cfg)
+			flags |= RING_ALLOC_REQ_FLAGS_TX_PKT_TS_CMPL_ENABLE;
+		req->flags = cpu_to_le16(flags);
 		break;
+	}
 	case HWRM_RING_ALLOC_RX:
-		req.ring_type = RING_ALLOC_REQ_RING_TYPE_RX;
-		req.length = cpu_to_le32(bp->rx_ring_mask + 1);
-		break;
 	case HWRM_RING_ALLOC_AGG:
-		req.ring_type = RING_ALLOC_REQ_RING_TYPE_RX;
-		req.length = cpu_to_le32(bp->rx_agg_ring_mask + 1);
+		req->ring_type = RING_ALLOC_REQ_RING_TYPE_RX;
+		req->length = (ring_type == HWRM_RING_ALLOC_RX) ?
+			      cpu_to_le32(bp->rx_ring_mask + 1) :
+			      cpu_to_le32(bp->rx_agg_ring_mask + 1);
+		if (bp->flags & BNXT_FLAG_CHIP_P5_PLUS)
+			bnxt_set_rx_ring_params_p5(bp, ring_type, req, ring);
 		break;
 	case HWRM_RING_ALLOC_CMPL:
-		req.ring_type = RING_ALLOC_REQ_RING_TYPE_L2_CMPL;
-		req.length = cpu_to_le32(bp->cp_ring_mask + 1);
-		if (bp->flags & BNXT_FLAG_USING_MSIX)
-			req.int_mode = RING_ALLOC_REQ_INT_MODE_MSIX;
+		req->ring_type = RING_ALLOC_REQ_RING_TYPE_L2_CMPL;
+		req->length = cpu_to_le32(bp->cp_ring_mask + 1);
+		if (bp->flags & BNXT_FLAG_CHIP_P5_PLUS) {
+			/* Association of cp ring with nq */
+			grp_info = &bp->grp_info[map_index];
+			req->nq_ring_id = cpu_to_le16(grp_info->cp_fw_ring_id);
+			req->cq_handle = cpu_to_le64(ring->handle);
+			req->enables |= cpu_to_le32(
+				RING_ALLOC_REQ_ENABLES_NQ_RING_ID_VALID);
+		} else {
+			req->int_mode = RING_ALLOC_REQ_INT_MODE_MSIX;
+		}
+		break;
+	case HWRM_RING_ALLOC_NQ:
+		req->ring_type = RING_ALLOC_REQ_RING_TYPE_NQ;
+		req->length = cpu_to_le32(bp->cp_ring_mask + 1);
+		req->int_mode = RING_ALLOC_REQ_INT_MODE_MSIX;
 		break;
 	default:
 		netdev_err(bp->dev, "hwrm alloc invalid ring type %d\n",
 			   ring_type);
-		return -1;
+		return -EINVAL;
 	}
 
-	mutex_lock(&bp->hwrm_cmd_lock);
-	rc = _hwrm_send_message(bp, &req, sizeof(req), HWRM_CMD_TIMEOUT);
+	resp = hwrm_req_hold(bp, req);
+	rc = hwrm_req_send(bp, req);
 	err = le16_to_cpu(resp->error_code);
 	ring_id = le16_to_cpu(resp->ring_id);
-	mutex_unlock(&bp->hwrm_cmd_lock);
+	hwrm_req_drop(bp, req);
 
+exit:
 	if (rc || err) {
-		switch (ring_type) {
-		case RING_FREE_REQ_RING_TYPE_L2_CMPL:
-			netdev_err(bp->dev, "hwrm_ring_alloc cp failed. rc:%x err:%x\n",
-				   rc, err);
-			return -1;
-
-		case RING_FREE_REQ_RING_TYPE_RX:
-			netdev_err(bp->dev, "hwrm_ring_alloc rx failed. rc:%x err:%x\n",
-				   rc, err);
-			return -1;
-
-		case RING_FREE_REQ_RING_TYPE_TX:
-			netdev_err(bp->dev, "hwrm_ring_alloc tx failed. rc:%x err:%x\n",
-				   rc, err);
-			return -1;
-
-		default:
-			netdev_err(bp->dev, "Invalid ring\n");
-			return -1;
-		}
+		netdev_err(bp->dev, "hwrm_ring_alloc type %d failed. rc:%x err:%x\n",
+			   ring_type, rc, err);
+		return -EIO;
 	}
 	ring->fw_ring_id = ring_id;
 	return rc;
@@ -4118,40 +7172,199 @@ static int bnxt_hwrm_set_async_event_cr(struct bnxt *bp, int idx)
 	int rc;
 
 	if (BNXT_PF(bp)) {
-		struct hwrm_func_cfg_input req = {0};
+		struct hwrm_func_cfg_input *req;
+
+		rc = bnxt_hwrm_func_cfg_short_req_init(bp, &req);
+		if (rc)
+			return rc;
 
-		bnxt_hwrm_cmd_hdr_init(bp, &req, HWRM_FUNC_CFG, -1, -1);
-		req.fid = cpu_to_le16(0xffff);
-		req.enables = cpu_to_le32(FUNC_CFG_REQ_ENABLES_ASYNC_EVENT_CR);
-		req.async_event_cr = cpu_to_le16(idx);
-		rc = hwrm_send_message(bp, &req, sizeof(req), HWRM_CMD_TIMEOUT);
+		req->fid = cpu_to_le16(0xffff);
+		req->enables = cpu_to_le32(FUNC_CFG_REQ_ENABLES_ASYNC_EVENT_CR);
+		req->async_event_cr = cpu_to_le16(idx);
+		return hwrm_req_send(bp, req);
 	} else {
-		struct hwrm_func_vf_cfg_input req = {0};
+		struct hwrm_func_vf_cfg_input *req;
 
-		bnxt_hwrm_cmd_hdr_init(bp, &req, HWRM_FUNC_VF_CFG, -1, -1);
-		req.enables =
+		rc = hwrm_req_init(bp, req, HWRM_FUNC_VF_CFG);
+		if (rc)
+			return rc;
+
+		req->enables =
 			cpu_to_le32(FUNC_VF_CFG_REQ_ENABLES_ASYNC_EVENT_CR);
-		req.async_event_cr = cpu_to_le16(idx);
-		rc = hwrm_send_message(bp, &req, sizeof(req), HWRM_CMD_TIMEOUT);
+		req->async_event_cr = cpu_to_le16(idx);
+		return hwrm_req_send(bp, req);
 	}
-	return rc;
+}
+
+static void bnxt_set_db_mask(struct bnxt *bp, struct bnxt_db_info *db,
+			     u32 ring_type)
+{
+	switch (ring_type) {
+	case HWRM_RING_ALLOC_TX:
+		db->db_ring_mask = bp->tx_ring_mask;
+		break;
+	case HWRM_RING_ALLOC_RX:
+		db->db_ring_mask = bp->rx_ring_mask;
+		break;
+	case HWRM_RING_ALLOC_AGG:
+		db->db_ring_mask = bp->rx_agg_ring_mask;
+		break;
+	case HWRM_RING_ALLOC_CMPL:
+	case HWRM_RING_ALLOC_NQ:
+		db->db_ring_mask = bp->cp_ring_mask;
+		break;
+	}
+	if (bp->flags & BNXT_FLAG_CHIP_P7) {
+		db->db_epoch_mask = db->db_ring_mask + 1;
+		db->db_epoch_shift = DBR_EPOCH_SFT - ilog2(db->db_epoch_mask);
+	}
+}
+
+static void bnxt_set_db(struct bnxt *bp, struct bnxt_db_info *db, u32 ring_type,
+			u32 map_idx, u32 xid)
+{
+	if (bp->flags & BNXT_FLAG_CHIP_P5_PLUS) {
+		switch (ring_type) {
+		case HWRM_RING_ALLOC_TX:
+			db->db_key64 = DBR_PATH_L2 | DBR_TYPE_SQ;
+			break;
+		case HWRM_RING_ALLOC_RX:
+		case HWRM_RING_ALLOC_AGG:
+			db->db_key64 = DBR_PATH_L2 | DBR_TYPE_SRQ;
+			break;
+		case HWRM_RING_ALLOC_CMPL:
+			db->db_key64 = DBR_PATH_L2;
+			break;
+		case HWRM_RING_ALLOC_NQ:
+			db->db_key64 = DBR_PATH_L2;
+			break;
+		}
+		db->db_key64 |= (u64)xid << DBR_XID_SFT;
+
+		if (bp->flags & BNXT_FLAG_CHIP_P7)
+			db->db_key64 |= DBR_VALID;
+
+		db->doorbell = bp->bar1 + bp->db_offset;
+	} else {
+		db->doorbell = bp->bar1 + map_idx * 0x80;
+		switch (ring_type) {
+		case HWRM_RING_ALLOC_TX:
+			db->db_key32 = DB_KEY_TX;
+			break;
+		case HWRM_RING_ALLOC_RX:
+		case HWRM_RING_ALLOC_AGG:
+			db->db_key32 = DB_KEY_RX;
+			break;
+		case HWRM_RING_ALLOC_CMPL:
+			db->db_key32 = DB_KEY_CP;
+			break;
+		}
+	}
+	bnxt_set_db_mask(bp, db, ring_type);
+}
+
+static int bnxt_hwrm_rx_ring_alloc(struct bnxt *bp,
+				   struct bnxt_rx_ring_info *rxr)
+{
+	struct bnxt_ring_struct *ring = &rxr->rx_ring_struct;
+	struct bnxt_napi *bnapi = rxr->bnapi;
+	u32 type = HWRM_RING_ALLOC_RX;
+	u32 map_idx = bnapi->index;
+	int rc;
+
+	rc = hwrm_ring_alloc_send_msg(bp, ring, type, map_idx);
+	if (rc)
+		return rc;
+
+	bnxt_set_db(bp, &rxr->rx_db, type, map_idx, ring->fw_ring_id);
+	bp->grp_info[map_idx].rx_fw_ring_id = ring->fw_ring_id;
+
+	return 0;
+}
+
+static int bnxt_hwrm_rx_agg_ring_alloc(struct bnxt *bp,
+				       struct bnxt_rx_ring_info *rxr)
+{
+	struct bnxt_ring_struct *ring = &rxr->rx_agg_ring_struct;
+	u32 type = HWRM_RING_ALLOC_AGG;
+	u32 grp_idx = ring->grp_idx;
+	u32 map_idx;
+	int rc;
+
+	map_idx = grp_idx + bp->rx_nr_rings;
+	rc = hwrm_ring_alloc_send_msg(bp, ring, type, map_idx);
+	if (rc)
+		return rc;
+
+	bnxt_set_db(bp, &rxr->rx_agg_db, type, map_idx,
+		    ring->fw_ring_id);
+	bnxt_db_write(bp, &rxr->rx_agg_db, rxr->rx_agg_prod);
+	bnxt_db_write(bp, &rxr->rx_db, rxr->rx_prod);
+	bp->grp_info[grp_idx].agg_fw_ring_id = ring->fw_ring_id;
+
+	return 0;
+}
+
+static int bnxt_hwrm_cp_ring_alloc_p5(struct bnxt *bp,
+				      struct bnxt_cp_ring_info *cpr)
+{
+	const u32 type = HWRM_RING_ALLOC_CMPL;
+	struct bnxt_napi *bnapi = cpr->bnapi;
+	struct bnxt_ring_struct *ring;
+	u32 map_idx = bnapi->index;
+	int rc;
+
+	ring = &cpr->cp_ring_struct;
+	ring->handle = BNXT_SET_NQ_HDL(cpr);
+	rc = hwrm_ring_alloc_send_msg(bp, ring, type, map_idx);
+	if (rc)
+		return rc;
+	bnxt_set_db(bp, &cpr->cp_db, type, map_idx, ring->fw_ring_id);
+	bnxt_db_cq(bp, &cpr->cp_db, cpr->cp_raw_cons);
+	return 0;
+}
+
+static int bnxt_hwrm_tx_ring_alloc(struct bnxt *bp,
+				   struct bnxt_tx_ring_info *txr, u32 tx_idx)
+{
+	struct bnxt_ring_struct *ring = &txr->tx_ring_struct;
+	const u32 type = HWRM_RING_ALLOC_TX;
+	int rc;
+
+	rc = hwrm_ring_alloc_send_msg(bp, ring, type, tx_idx);
+	if (rc)
+		return rc;
+	bnxt_set_db(bp, &txr->tx_db, type, tx_idx, ring->fw_ring_id);
+	return 0;
 }
 
 static int bnxt_hwrm_ring_alloc(struct bnxt *bp)
 {
+	bool agg_rings = !!(bp->flags & BNXT_FLAG_AGG_RINGS);
 	int i, rc = 0;
+	u32 type;
 
+	if (bp->flags & BNXT_FLAG_CHIP_P5_PLUS)
+		type = HWRM_RING_ALLOC_NQ;
+	else
+		type = HWRM_RING_ALLOC_CMPL;
 	for (i = 0; i < bp->cp_nr_rings; i++) {
 		struct bnxt_napi *bnapi = bp->bnapi[i];
 		struct bnxt_cp_ring_info *cpr = &bnapi->cp_ring;
 		struct bnxt_ring_struct *ring = &cpr->cp_ring_struct;
+		u32 map_idx = ring->map_idx;
+		unsigned int vector;
 
-		cpr->cp_doorbell = bp->bar1 + i * 0x80;
-		rc = hwrm_ring_alloc_send_msg(bp, ring, HWRM_RING_ALLOC_CMPL, i,
-					      INVALID_STATS_CTX_ID);
-		if (rc)
+		vector = bp->irq_tbl[map_idx].vector;
+		disable_irq_nosync(vector);
+		rc = hwrm_ring_alloc_send_msg(bp, ring, type, map_idx);
+		if (rc) {
+			enable_irq(vector);
 			goto err_out;
-		BNXT_CP_DB(cpr->cp_doorbell, cpr->cp_raw_cons);
+		}
+		bnxt_set_db(bp, &cpr->cp_db, type, map_idx, ring->fw_ring_id);
+		bnxt_db_nq(bp, &cpr->cp_db, cpr->cp_raw_cons);
+		enable_irq(vector);
 		bp->grp_info[i].cp_fw_ring_id = ring->fw_ring_id;
 
 		if (!i) {
@@ -4163,150 +7376,200 @@ static int bnxt_hwrm_ring_alloc(struct bnxt *bp)
 
 	for (i = 0; i < bp->tx_nr_rings; i++) {
 		struct bnxt_tx_ring_info *txr = &bp->tx_ring[i];
-		struct bnxt_ring_struct *ring = &txr->tx_ring_struct;
-		u32 map_idx = txr->bnapi->index;
-		u16 fw_stats_ctx = bp->grp_info[map_idx].fw_stats_ctx;
 
-		rc = hwrm_ring_alloc_send_msg(bp, ring, HWRM_RING_ALLOC_TX,
-					      map_idx, fw_stats_ctx);
+		if (bp->flags & BNXT_FLAG_CHIP_P5_PLUS) {
+			rc = bnxt_hwrm_cp_ring_alloc_p5(bp, txr->tx_cpr);
+			if (rc)
+				goto err_out;
+		}
+		rc = bnxt_hwrm_tx_ring_alloc(bp, txr, i);
 		if (rc)
 			goto err_out;
-		txr->tx_doorbell = bp->bar1 + map_idx * 0x80;
 	}
 
 	for (i = 0; i < bp->rx_nr_rings; i++) {
 		struct bnxt_rx_ring_info *rxr = &bp->rx_ring[i];
-		struct bnxt_ring_struct *ring = &rxr->rx_ring_struct;
-		u32 map_idx = rxr->bnapi->index;
 
-		rc = hwrm_ring_alloc_send_msg(bp, ring, HWRM_RING_ALLOC_RX,
-					      map_idx, INVALID_STATS_CTX_ID);
+		rc = bnxt_hwrm_rx_ring_alloc(bp, rxr);
 		if (rc)
 			goto err_out;
-		rxr->rx_doorbell = bp->bar1 + map_idx * 0x80;
-		writel(DB_KEY_RX | rxr->rx_prod, rxr->rx_doorbell);
-		bp->grp_info[map_idx].rx_fw_ring_id = ring->fw_ring_id;
+		/* If we have agg rings, post agg buffers first. */
+		if (!agg_rings)
+			bnxt_db_write(bp, &rxr->rx_db, rxr->rx_prod);
+		if (bp->flags & BNXT_FLAG_CHIP_P5_PLUS) {
+			rc = bnxt_hwrm_cp_ring_alloc_p5(bp, rxr->rx_cpr);
+			if (rc)
+				goto err_out;
+		}
 	}
 
-	if (bp->flags & BNXT_FLAG_AGG_RINGS) {
+	if (agg_rings) {
 		for (i = 0; i < bp->rx_nr_rings; i++) {
-			struct bnxt_rx_ring_info *rxr = &bp->rx_ring[i];
-			struct bnxt_ring_struct *ring =
-						&rxr->rx_agg_ring_struct;
-			u32 grp_idx = rxr->bnapi->index;
-			u32 map_idx = grp_idx + bp->rx_nr_rings;
-
-			rc = hwrm_ring_alloc_send_msg(bp, ring,
-						      HWRM_RING_ALLOC_AGG,
-						      map_idx,
-						      INVALID_STATS_CTX_ID);
+			rc = bnxt_hwrm_rx_agg_ring_alloc(bp, &bp->rx_ring[i]);
 			if (rc)
 				goto err_out;
-
-			rxr->rx_agg_doorbell = bp->bar1 + map_idx * 0x80;
-			writel(DB_KEY_RX | rxr->rx_agg_prod,
-			       rxr->rx_agg_doorbell);
-			bp->grp_info[grp_idx].agg_fw_ring_id = ring->fw_ring_id;
 		}
 	}
 err_out:
 	return rc;
 }
 
+static void bnxt_cancel_dim(struct bnxt *bp)
+{
+	int i;
+
+	/* DIM work is initialized in bnxt_enable_napi().  Proceed only
+	 * if NAPI is enabled.
+	 */
+	if (!bp->bnapi || test_bit(BNXT_STATE_NAPI_DISABLED, &bp->state))
+		return;
+
+	/* Make sure NAPI sees that the VNIC is disabled */
+	synchronize_net();
+	for (i = 0; i < bp->rx_nr_rings; i++) {
+		struct bnxt_rx_ring_info *rxr = &bp->rx_ring[i];
+		struct bnxt_napi *bnapi = rxr->bnapi;
+
+		cancel_work_sync(&bnapi->cp_ring.dim.work);
+	}
+}
+
 static int hwrm_ring_free_send_msg(struct bnxt *bp,
 				   struct bnxt_ring_struct *ring,
 				   u32 ring_type, int cmpl_ring_id)
 {
+	struct hwrm_ring_free_output *resp;
+	struct hwrm_ring_free_input *req;
+	u16 error_code = 0;
 	int rc;
-	struct hwrm_ring_free_input req = {0};
-	struct hwrm_ring_free_output *resp = bp->hwrm_cmd_resp_addr;
-	u16 error_code;
 
-	bnxt_hwrm_cmd_hdr_init(bp, &req, HWRM_RING_FREE, cmpl_ring_id, -1);
-	req.ring_type = ring_type;
-	req.ring_id = cpu_to_le16(ring->fw_ring_id);
+	if (BNXT_NO_FW_ACCESS(bp))
+		return 0;
 
-	mutex_lock(&bp->hwrm_cmd_lock);
-	rc = _hwrm_send_message(bp, &req, sizeof(req), HWRM_CMD_TIMEOUT);
-	error_code = le16_to_cpu(resp->error_code);
-	mutex_unlock(&bp->hwrm_cmd_lock);
+	rc = hwrm_req_init(bp, req, HWRM_RING_FREE);
+	if (rc)
+		goto exit;
 
+	req->cmpl_ring = cpu_to_le16(cmpl_ring_id);
+	req->ring_type = ring_type;
+	req->ring_id = cpu_to_le16(ring->fw_ring_id);
+
+	resp = hwrm_req_hold(bp, req);
+	rc = hwrm_req_send(bp, req);
+	error_code = le16_to_cpu(resp->error_code);
+	hwrm_req_drop(bp, req);
+exit:
 	if (rc || error_code) {
-		switch (ring_type) {
-		case RING_FREE_REQ_RING_TYPE_L2_CMPL:
-			netdev_err(bp->dev, "hwrm_ring_free cp failed. rc:%d\n",
-				   rc);
-			return rc;
-		case RING_FREE_REQ_RING_TYPE_RX:
-			netdev_err(bp->dev, "hwrm_ring_free rx failed. rc:%d\n",
-				   rc);
-			return rc;
-		case RING_FREE_REQ_RING_TYPE_TX:
-			netdev_err(bp->dev, "hwrm_ring_free tx failed. rc:%d\n",
-				   rc);
-			return rc;
-		default:
-			netdev_err(bp->dev, "Invalid ring\n");
-			return -1;
-		}
+		netdev_err(bp->dev, "hwrm_ring_free type %d failed. rc:%x err:%x\n",
+			   ring_type, rc, error_code);
+		return -EIO;
 	}
 	return 0;
 }
 
-static void bnxt_hwrm_ring_free(struct bnxt *bp, bool close_path)
+static void bnxt_hwrm_tx_ring_free(struct bnxt *bp,
+				   struct bnxt_tx_ring_info *txr,
+				   bool close_path)
 {
-	int i;
+	struct bnxt_ring_struct *ring = &txr->tx_ring_struct;
+	u32 cmpl_ring_id;
 
-	if (!bp->bnapi)
+	if (ring->fw_ring_id == INVALID_HW_RING_ID)
 		return;
 
-	for (i = 0; i < bp->tx_nr_rings; i++) {
-		struct bnxt_tx_ring_info *txr = &bp->tx_ring[i];
-		struct bnxt_ring_struct *ring = &txr->tx_ring_struct;
-		u32 grp_idx = txr->bnapi->index;
-		u32 cmpl_ring_id = bp->grp_info[grp_idx].cp_fw_ring_id;
+	cmpl_ring_id = close_path ? bnxt_cp_ring_for_tx(bp, txr) :
+		       INVALID_HW_RING_ID;
+	hwrm_ring_free_send_msg(bp, ring, RING_FREE_REQ_RING_TYPE_TX,
+				cmpl_ring_id);
+	ring->fw_ring_id = INVALID_HW_RING_ID;
+}
 
-		if (ring->fw_ring_id != INVALID_HW_RING_ID) {
-			hwrm_ring_free_send_msg(bp, ring,
-						RING_FREE_REQ_RING_TYPE_TX,
-						close_path ? cmpl_ring_id :
-						INVALID_HW_RING_ID);
-			ring->fw_ring_id = INVALID_HW_RING_ID;
-		}
-	}
+static void bnxt_hwrm_rx_ring_free(struct bnxt *bp,
+				   struct bnxt_rx_ring_info *rxr,
+				   bool close_path)
+{
+	struct bnxt_ring_struct *ring = &rxr->rx_ring_struct;
+	u32 grp_idx = rxr->bnapi->index;
+	u32 cmpl_ring_id;
 
-	for (i = 0; i < bp->rx_nr_rings; i++) {
-		struct bnxt_rx_ring_info *rxr = &bp->rx_ring[i];
-		struct bnxt_ring_struct *ring = &rxr->rx_ring_struct;
-		u32 grp_idx = rxr->bnapi->index;
-		u32 cmpl_ring_id = bp->grp_info[grp_idx].cp_fw_ring_id;
+	if (ring->fw_ring_id == INVALID_HW_RING_ID)
+		return;
 
-		if (ring->fw_ring_id != INVALID_HW_RING_ID) {
-			hwrm_ring_free_send_msg(bp, ring,
-						RING_FREE_REQ_RING_TYPE_RX,
-						close_path ? cmpl_ring_id :
-						INVALID_HW_RING_ID);
-			ring->fw_ring_id = INVALID_HW_RING_ID;
-			bp->grp_info[grp_idx].rx_fw_ring_id =
-				INVALID_HW_RING_ID;
-		}
-	}
+	cmpl_ring_id = bnxt_cp_ring_for_rx(bp, rxr);
+	hwrm_ring_free_send_msg(bp, ring,
+				RING_FREE_REQ_RING_TYPE_RX,
+				close_path ? cmpl_ring_id :
+				INVALID_HW_RING_ID);
+	ring->fw_ring_id = INVALID_HW_RING_ID;
+	bp->grp_info[grp_idx].rx_fw_ring_id = INVALID_HW_RING_ID;
+}
 
-	for (i = 0; i < bp->rx_nr_rings; i++) {
-		struct bnxt_rx_ring_info *rxr = &bp->rx_ring[i];
-		struct bnxt_ring_struct *ring = &rxr->rx_agg_ring_struct;
-		u32 grp_idx = rxr->bnapi->index;
-		u32 cmpl_ring_id = bp->grp_info[grp_idx].cp_fw_ring_id;
+static void bnxt_hwrm_rx_agg_ring_free(struct bnxt *bp,
+				       struct bnxt_rx_ring_info *rxr,
+				       bool close_path)
+{
+	struct bnxt_ring_struct *ring = &rxr->rx_agg_ring_struct;
+	u32 grp_idx = rxr->bnapi->index;
+	u32 type, cmpl_ring_id;
 
-		if (ring->fw_ring_id != INVALID_HW_RING_ID) {
-			hwrm_ring_free_send_msg(bp, ring,
-						RING_FREE_REQ_RING_TYPE_RX,
-						close_path ? cmpl_ring_id :
-						INVALID_HW_RING_ID);
-			ring->fw_ring_id = INVALID_HW_RING_ID;
-			bp->grp_info[grp_idx].agg_fw_ring_id =
-				INVALID_HW_RING_ID;
-		}
+	if (bp->flags & BNXT_FLAG_CHIP_P5_PLUS)
+		type = RING_FREE_REQ_RING_TYPE_RX_AGG;
+	else
+		type = RING_FREE_REQ_RING_TYPE_RX;
+
+	if (ring->fw_ring_id == INVALID_HW_RING_ID)
+		return;
+
+	cmpl_ring_id = bnxt_cp_ring_for_rx(bp, rxr);
+	hwrm_ring_free_send_msg(bp, ring, type,
+				close_path ? cmpl_ring_id :
+				INVALID_HW_RING_ID);
+	ring->fw_ring_id = INVALID_HW_RING_ID;
+	bp->grp_info[grp_idx].agg_fw_ring_id = INVALID_HW_RING_ID;
+}
+
+static void bnxt_hwrm_cp_ring_free(struct bnxt *bp,
+				   struct bnxt_cp_ring_info *cpr)
+{
+	struct bnxt_ring_struct *ring;
+
+	ring = &cpr->cp_ring_struct;
+	if (ring->fw_ring_id == INVALID_HW_RING_ID)
+		return;
+
+	hwrm_ring_free_send_msg(bp, ring, RING_FREE_REQ_RING_TYPE_L2_CMPL,
+				INVALID_HW_RING_ID);
+	ring->fw_ring_id = INVALID_HW_RING_ID;
+}
+
+static void bnxt_clear_one_cp_ring(struct bnxt *bp, struct bnxt_cp_ring_info *cpr)
+{
+	struct bnxt_ring_struct *ring = &cpr->cp_ring_struct;
+	int i, size = ring->ring_mem.page_size;
+
+	cpr->cp_raw_cons = 0;
+	cpr->toggle = 0;
+
+	for (i = 0; i < bp->cp_nr_pages; i++)
+		if (cpr->cp_desc_ring[i])
+			memset(cpr->cp_desc_ring[i], 0, size);
+}
+
+static void bnxt_hwrm_ring_free(struct bnxt *bp, bool close_path)
+{
+	u32 type;
+	int i;
+
+	if (!bp->bnapi)
+		return;
+
+	for (i = 0; i < bp->tx_nr_rings; i++)
+		bnxt_hwrm_tx_ring_free(bp, &bp->tx_ring[i], close_path);
+
+	bnxt_cancel_dim(bp);
+	for (i = 0; i < bp->rx_nr_rings; i++) {
+		bnxt_hwrm_rx_ring_free(bp, &bp->rx_ring[i], close_path);
+		bnxt_hwrm_rx_agg_ring_free(bp, &bp->rx_ring[i], close_path);
 	}
 
 	/* The completion rings are about to be freed.  After that the
@@ -4315,14 +7578,22 @@ static void bnxt_hwrm_ring_free(struct bnxt *bp, bool close_path)
 	 */
 	bnxt_disable_int_sync(bp);
 
+	if (bp->flags & BNXT_FLAG_CHIP_P5_PLUS)
+		type = RING_FREE_REQ_RING_TYPE_NQ;
+	else
+		type = RING_FREE_REQ_RING_TYPE_L2_CMPL;
 	for (i = 0; i < bp->cp_nr_rings; i++) {
 		struct bnxt_napi *bnapi = bp->bnapi[i];
 		struct bnxt_cp_ring_info *cpr = &bnapi->cp_ring;
-		struct bnxt_ring_struct *ring = &cpr->cp_ring_struct;
+		struct bnxt_ring_struct *ring;
+		int j;
+
+		for (j = 0; j < cpr->cp_ring_count && cpr->cp_ring_arr; j++)
+			bnxt_hwrm_cp_ring_free(bp, &cpr->cp_ring_arr[j]);
 
+		ring = &cpr->cp_ring_struct;
 		if (ring->fw_ring_id != INVALID_HW_RING_ID) {
-			hwrm_ring_free_send_msg(bp, ring,
-						RING_FREE_REQ_RING_TYPE_L2_CMPL,
+			hwrm_ring_free_send_msg(bp, ring, type,
 						INVALID_HW_RING_ID);
 			ring->fw_ring_id = INVALID_HW_RING_ID;
 			bp->grp_info[i].cp_fw_ring_id = INVALID_HW_RING_ID;
@@ -4330,189 +7601,855 @@ static void bnxt_hwrm_ring_free(struct bnxt *bp, bool close_path)
 	}
 }
 
-/* Caller must hold bp->hwrm_cmd_lock */
+static int __bnxt_trim_rings(struct bnxt *bp, int *rx, int *tx, int max,
+			     bool shared);
+static int bnxt_trim_rings(struct bnxt *bp, int *rx, int *tx, int max,
+			   bool shared);
+
+static int bnxt_hwrm_get_rings(struct bnxt *bp)
+{
+	struct bnxt_hw_resc *hw_resc = &bp->hw_resc;
+	struct hwrm_func_qcfg_output *resp;
+	struct hwrm_func_qcfg_input *req;
+	int rc;
+
+	if (bp->hwrm_spec_code < 0x10601)
+		return 0;
+
+	rc = hwrm_req_init(bp, req, HWRM_FUNC_QCFG);
+	if (rc)
+		return rc;
+
+	req->fid = cpu_to_le16(0xffff);
+	resp = hwrm_req_hold(bp, req);
+	rc = hwrm_req_send(bp, req);
+	if (rc) {
+		hwrm_req_drop(bp, req);
+		return rc;
+	}
+
+	hw_resc->resv_tx_rings = le16_to_cpu(resp->alloc_tx_rings);
+	if (BNXT_NEW_RM(bp)) {
+		u16 cp, stats;
+
+		hw_resc->resv_rx_rings = le16_to_cpu(resp->alloc_rx_rings);
+		hw_resc->resv_hw_ring_grps =
+			le32_to_cpu(resp->alloc_hw_ring_grps);
+		hw_resc->resv_vnics = le16_to_cpu(resp->alloc_vnics);
+		hw_resc->resv_rsscos_ctxs = le16_to_cpu(resp->alloc_rsscos_ctx);
+		cp = le16_to_cpu(resp->alloc_cmpl_rings);
+		stats = le16_to_cpu(resp->alloc_stat_ctx);
+		hw_resc->resv_irqs = cp;
+		if (bp->flags & BNXT_FLAG_CHIP_P5_PLUS) {
+			int rx = hw_resc->resv_rx_rings;
+			int tx = hw_resc->resv_tx_rings;
+
+			if (bp->flags & BNXT_FLAG_AGG_RINGS)
+				rx >>= 1;
+			if (cp < (rx + tx)) {
+				rc = __bnxt_trim_rings(bp, &rx, &tx, cp, false);
+				if (rc)
+					goto get_rings_exit;
+				if (bp->flags & BNXT_FLAG_AGG_RINGS)
+					rx <<= 1;
+				hw_resc->resv_rx_rings = rx;
+				hw_resc->resv_tx_rings = tx;
+			}
+			hw_resc->resv_irqs = le16_to_cpu(resp->alloc_msix);
+			hw_resc->resv_hw_ring_grps = rx;
+		}
+		hw_resc->resv_cp_rings = cp;
+		hw_resc->resv_stat_ctxs = stats;
+	}
+get_rings_exit:
+	hwrm_req_drop(bp, req);
+	return rc;
+}
+
 int __bnxt_hwrm_get_tx_rings(struct bnxt *bp, u16 fid, int *tx_rings)
 {
-	struct hwrm_func_qcfg_output *resp = bp->hwrm_cmd_resp_addr;
-	struct hwrm_func_qcfg_input req = {0};
+	struct hwrm_func_qcfg_output *resp;
+	struct hwrm_func_qcfg_input *req;
 	int rc;
 
 	if (bp->hwrm_spec_code < 0x10601)
 		return 0;
 
-	bnxt_hwrm_cmd_hdr_init(bp, &req, HWRM_FUNC_QCFG, -1, -1);
-	req.fid = cpu_to_le16(fid);
-	rc = _hwrm_send_message(bp, &req, sizeof(req), HWRM_CMD_TIMEOUT);
+	rc = hwrm_req_init(bp, req, HWRM_FUNC_QCFG);
+	if (rc)
+		return rc;
+
+	req->fid = cpu_to_le16(fid);
+	resp = hwrm_req_hold(bp, req);
+	rc = hwrm_req_send(bp, req);
 	if (!rc)
 		*tx_rings = le16_to_cpu(resp->alloc_tx_rings);
 
+	hwrm_req_drop(bp, req);
 	return rc;
 }
 
-static int bnxt_hwrm_reserve_tx_rings(struct bnxt *bp, int *tx_rings)
+static bool bnxt_rfs_supported(struct bnxt *bp);
+
+static struct hwrm_func_cfg_input *
+__bnxt_hwrm_reserve_pf_rings(struct bnxt *bp, struct bnxt_hw_rings *hwr)
+{
+	struct hwrm_func_cfg_input *req;
+	u32 enables = 0;
+
+	if (bnxt_hwrm_func_cfg_short_req_init(bp, &req))
+		return NULL;
+
+	req->fid = cpu_to_le16(0xffff);
+	enables |= hwr->tx ? FUNC_CFG_REQ_ENABLES_NUM_TX_RINGS : 0;
+	req->num_tx_rings = cpu_to_le16(hwr->tx);
+	if (BNXT_NEW_RM(bp)) {
+		enables |= hwr->rx ? FUNC_CFG_REQ_ENABLES_NUM_RX_RINGS : 0;
+		enables |= hwr->stat ? FUNC_CFG_REQ_ENABLES_NUM_STAT_CTXS : 0;
+		if (bp->flags & BNXT_FLAG_CHIP_P5_PLUS) {
+			enables |= hwr->cp ? FUNC_CFG_REQ_ENABLES_NUM_MSIX : 0;
+			enables |= hwr->cp_p5 ?
+				   FUNC_CFG_REQ_ENABLES_NUM_CMPL_RINGS : 0;
+		} else {
+			enables |= hwr->cp ?
+				   FUNC_CFG_REQ_ENABLES_NUM_CMPL_RINGS : 0;
+			enables |= hwr->grp ?
+				   FUNC_CFG_REQ_ENABLES_NUM_HW_RING_GRPS : 0;
+		}
+		enables |= hwr->vnic ? FUNC_CFG_REQ_ENABLES_NUM_VNICS : 0;
+		enables |= hwr->rss_ctx ? FUNC_CFG_REQ_ENABLES_NUM_RSSCOS_CTXS :
+					  0;
+		req->num_rx_rings = cpu_to_le16(hwr->rx);
+		req->num_rsscos_ctxs = cpu_to_le16(hwr->rss_ctx);
+		if (bp->flags & BNXT_FLAG_CHIP_P5_PLUS) {
+			req->num_cmpl_rings = cpu_to_le16(hwr->cp_p5);
+			req->num_msix = cpu_to_le16(hwr->cp);
+		} else {
+			req->num_cmpl_rings = cpu_to_le16(hwr->cp);
+			req->num_hw_ring_grps = cpu_to_le16(hwr->grp);
+		}
+		req->num_stat_ctxs = cpu_to_le16(hwr->stat);
+		req->num_vnics = cpu_to_le16(hwr->vnic);
+	}
+	req->enables = cpu_to_le32(enables);
+	return req;
+}
+
+static struct hwrm_func_vf_cfg_input *
+__bnxt_hwrm_reserve_vf_rings(struct bnxt *bp, struct bnxt_hw_rings *hwr)
+{
+	struct hwrm_func_vf_cfg_input *req;
+	u32 enables = 0;
+
+	if (hwrm_req_init(bp, req, HWRM_FUNC_VF_CFG))
+		return NULL;
+
+	enables |= hwr->tx ? FUNC_VF_CFG_REQ_ENABLES_NUM_TX_RINGS : 0;
+	enables |= hwr->rx ? FUNC_VF_CFG_REQ_ENABLES_NUM_RX_RINGS |
+			     FUNC_VF_CFG_REQ_ENABLES_NUM_RSSCOS_CTXS : 0;
+	enables |= hwr->stat ? FUNC_VF_CFG_REQ_ENABLES_NUM_STAT_CTXS : 0;
+	enables |= hwr->rss_ctx ? FUNC_VF_CFG_REQ_ENABLES_NUM_RSSCOS_CTXS : 0;
+	if (bp->flags & BNXT_FLAG_CHIP_P5_PLUS) {
+		enables |= hwr->cp_p5 ?
+			   FUNC_VF_CFG_REQ_ENABLES_NUM_CMPL_RINGS : 0;
+	} else {
+		enables |= hwr->cp ? FUNC_VF_CFG_REQ_ENABLES_NUM_CMPL_RINGS : 0;
+		enables |= hwr->grp ?
+			   FUNC_VF_CFG_REQ_ENABLES_NUM_HW_RING_GRPS : 0;
+	}
+	enables |= hwr->vnic ? FUNC_VF_CFG_REQ_ENABLES_NUM_VNICS : 0;
+	enables |= FUNC_VF_CFG_REQ_ENABLES_NUM_L2_CTXS;
+
+	req->num_l2_ctxs = cpu_to_le16(BNXT_VF_MAX_L2_CTX);
+	req->num_tx_rings = cpu_to_le16(hwr->tx);
+	req->num_rx_rings = cpu_to_le16(hwr->rx);
+	req->num_rsscos_ctxs = cpu_to_le16(hwr->rss_ctx);
+	if (bp->flags & BNXT_FLAG_CHIP_P5_PLUS) {
+		req->num_cmpl_rings = cpu_to_le16(hwr->cp_p5);
+	} else {
+		req->num_cmpl_rings = cpu_to_le16(hwr->cp);
+		req->num_hw_ring_grps = cpu_to_le16(hwr->grp);
+	}
+	req->num_stat_ctxs = cpu_to_le16(hwr->stat);
+	req->num_vnics = cpu_to_le16(hwr->vnic);
+
+	req->enables = cpu_to_le32(enables);
+	return req;
+}
+
+static int
+bnxt_hwrm_reserve_pf_rings(struct bnxt *bp, struct bnxt_hw_rings *hwr)
 {
-	struct hwrm_func_cfg_input req = {0};
+	struct hwrm_func_cfg_input *req;
 	int rc;
 
+	req = __bnxt_hwrm_reserve_pf_rings(bp, hwr);
+	if (!req)
+		return -ENOMEM;
+
+	if (!req->enables) {
+		hwrm_req_drop(bp, req);
+		return 0;
+	}
+
+	rc = hwrm_req_send(bp, req);
+	if (rc)
+		return rc;
+
 	if (bp->hwrm_spec_code < 0x10601)
+		bp->hw_resc.resv_tx_rings = hwr->tx;
+
+	return bnxt_hwrm_get_rings(bp);
+}
+
+static int
+bnxt_hwrm_reserve_vf_rings(struct bnxt *bp, struct bnxt_hw_rings *hwr)
+{
+	struct hwrm_func_vf_cfg_input *req;
+	int rc;
+
+	if (!BNXT_NEW_RM(bp)) {
+		bp->hw_resc.resv_tx_rings = hwr->tx;
 		return 0;
+	}
+
+	req = __bnxt_hwrm_reserve_vf_rings(bp, hwr);
+	if (!req)
+		return -ENOMEM;
+
+	rc = hwrm_req_send(bp, req);
+	if (rc)
+		return rc;
 
+	return bnxt_hwrm_get_rings(bp);
+}
+
+static int bnxt_hwrm_reserve_rings(struct bnxt *bp, struct bnxt_hw_rings *hwr)
+{
+	if (BNXT_PF(bp))
+		return bnxt_hwrm_reserve_pf_rings(bp, hwr);
+	else
+		return bnxt_hwrm_reserve_vf_rings(bp, hwr);
+}
+
+int bnxt_nq_rings_in_use(struct bnxt *bp)
+{
+	return bp->cp_nr_rings + bnxt_get_ulp_msix_num(bp);
+}
+
+static int bnxt_cp_rings_in_use(struct bnxt *bp)
+{
+	int cp;
+
+	if (!(bp->flags & BNXT_FLAG_CHIP_P5_PLUS))
+		return bnxt_nq_rings_in_use(bp);
+
+	cp = bp->tx_nr_rings + bp->rx_nr_rings;
+	return cp;
+}
+
+static int bnxt_get_func_stat_ctxs(struct bnxt *bp)
+{
+	return bp->cp_nr_rings + bnxt_get_ulp_stat_ctxs(bp);
+}
+
+static int bnxt_get_total_rss_ctxs(struct bnxt *bp, struct bnxt_hw_rings *hwr)
+{
+	if (!hwr->grp)
+		return 0;
+	if (bp->flags & BNXT_FLAG_CHIP_P5_PLUS) {
+		int rss_ctx = bnxt_get_nr_rss_ctxs(bp, hwr->grp);
+
+		if (BNXT_SUPPORTS_NTUPLE_VNIC(bp))
+			rss_ctx *= hwr->vnic;
+		return rss_ctx;
+	}
 	if (BNXT_VF(bp))
+		return BNXT_VF_MAX_RSS_CTX;
+	if (!(bp->rss_cap & BNXT_RSS_CAP_NEW_RSS_CAP) && bnxt_rfs_supported(bp))
+		return hwr->grp + 1;
+	return 1;
+}
+
+/* Check if a default RSS map needs to be setup.  This function is only
+ * used on older firmware that does not require reserving RX rings.
+ */
+static void bnxt_check_rss_tbl_no_rmgr(struct bnxt *bp)
+{
+	struct bnxt_hw_resc *hw_resc = &bp->hw_resc;
+
+	/* The RSS map is valid for RX rings set to resv_rx_rings */
+	if (hw_resc->resv_rx_rings != bp->rx_nr_rings) {
+		hw_resc->resv_rx_rings = bp->rx_nr_rings;
+		if (!netif_is_rxfh_configured(bp->dev))
+			bnxt_set_dflt_rss_indir_tbl(bp, NULL);
+	}
+}
+
+static int bnxt_get_total_vnics(struct bnxt *bp, int rx_rings)
+{
+	if (bp->flags & BNXT_FLAG_RFS) {
+		if (BNXT_SUPPORTS_NTUPLE_VNIC(bp))
+			return 2 + bp->num_rss_ctx;
+		if (!(bp->flags & BNXT_FLAG_CHIP_P5_PLUS))
+			return rx_rings + 1;
+	}
+	return 1;
+}
+
+static bool bnxt_need_reserve_rings(struct bnxt *bp)
+{
+	struct bnxt_hw_resc *hw_resc = &bp->hw_resc;
+	int cp = bnxt_cp_rings_in_use(bp);
+	int nq = bnxt_nq_rings_in_use(bp);
+	int rx = bp->rx_nr_rings, stat;
+	int vnic, grp = rx;
+
+	/* Old firmware does not need RX ring reservations but we still
+	 * need to setup a default RSS map when needed.  With new firmware
+	 * we go through RX ring reservations first and then set up the
+	 * RSS map for the successfully reserved RX rings when needed.
+	 */
+	if (!BNXT_NEW_RM(bp))
+		bnxt_check_rss_tbl_no_rmgr(bp);
+
+	if (hw_resc->resv_tx_rings != bp->tx_nr_rings &&
+	    bp->hwrm_spec_code >= 0x10601)
+		return true;
+
+	if (!BNXT_NEW_RM(bp))
+		return false;
+
+	vnic = bnxt_get_total_vnics(bp, rx);
+
+	if (bp->flags & BNXT_FLAG_AGG_RINGS)
+		rx <<= 1;
+	stat = bnxt_get_func_stat_ctxs(bp);
+	if (hw_resc->resv_rx_rings != rx || hw_resc->resv_cp_rings != cp ||
+	    hw_resc->resv_vnics != vnic || hw_resc->resv_stat_ctxs != stat ||
+	    (hw_resc->resv_hw_ring_grps != grp &&
+	     !(bp->flags & BNXT_FLAG_CHIP_P5_PLUS)))
+		return true;
+	if ((bp->flags & BNXT_FLAG_CHIP_P5_PLUS) && BNXT_PF(bp) &&
+	    hw_resc->resv_irqs != nq)
+		return true;
+	return false;
+}
+
+static void bnxt_copy_reserved_rings(struct bnxt *bp, struct bnxt_hw_rings *hwr)
+{
+	struct bnxt_hw_resc *hw_resc = &bp->hw_resc;
+
+	hwr->tx = hw_resc->resv_tx_rings;
+	if (BNXT_NEW_RM(bp)) {
+		hwr->rx = hw_resc->resv_rx_rings;
+		hwr->cp = hw_resc->resv_irqs;
+		if (bp->flags & BNXT_FLAG_CHIP_P5_PLUS)
+			hwr->cp_p5 = hw_resc->resv_cp_rings;
+		hwr->grp = hw_resc->resv_hw_ring_grps;
+		hwr->vnic = hw_resc->resv_vnics;
+		hwr->stat = hw_resc->resv_stat_ctxs;
+		hwr->rss_ctx = hw_resc->resv_rsscos_ctxs;
+	}
+}
+
+static bool bnxt_rings_ok(struct bnxt *bp, struct bnxt_hw_rings *hwr)
+{
+	return hwr->tx && hwr->rx && hwr->cp && hwr->grp && hwr->vnic &&
+	       hwr->stat && (hwr->cp_p5 || !(bp->flags & BNXT_FLAG_CHIP_P5_PLUS));
+}
+
+static int bnxt_get_avail_msix(struct bnxt *bp, int num);
+
+static int __bnxt_reserve_rings(struct bnxt *bp)
+{
+	struct bnxt_hw_rings hwr = {0};
+	int rx_rings, old_rx_rings, rc;
+	int cp = bp->cp_nr_rings;
+	int ulp_msix = 0;
+	bool sh = false;
+	int tx_cp;
+
+	if (!bnxt_need_reserve_rings(bp))
 		return 0;
 
-	bnxt_hwrm_cmd_hdr_init(bp, &req, HWRM_FUNC_CFG, -1, -1);
-	req.fid = cpu_to_le16(0xffff);
-	req.enables = cpu_to_le32(FUNC_CFG_REQ_ENABLES_NUM_TX_RINGS);
-	req.num_tx_rings = cpu_to_le16(*tx_rings);
-	rc = hwrm_send_message(bp, &req, sizeof(req), HWRM_CMD_TIMEOUT);
+	if (BNXT_NEW_RM(bp) && !bnxt_ulp_registered(bp->edev)) {
+		ulp_msix = bnxt_get_avail_msix(bp, bp->ulp_num_msix_want);
+		if (!ulp_msix)
+			bnxt_set_ulp_stat_ctxs(bp, 0);
+
+		if (ulp_msix > bp->ulp_num_msix_want)
+			ulp_msix = bp->ulp_num_msix_want;
+		hwr.cp = cp + ulp_msix;
+	} else {
+		hwr.cp = bnxt_nq_rings_in_use(bp);
+	}
+
+	hwr.tx = bp->tx_nr_rings;
+	hwr.rx = bp->rx_nr_rings;
+	if (bp->flags & BNXT_FLAG_SHARED_RINGS)
+		sh = true;
+	if (bp->flags & BNXT_FLAG_CHIP_P5_PLUS)
+		hwr.cp_p5 = hwr.rx + hwr.tx;
+
+	hwr.vnic = bnxt_get_total_vnics(bp, hwr.rx);
+
+	if (bp->flags & BNXT_FLAG_AGG_RINGS)
+		hwr.rx <<= 1;
+	hwr.grp = bp->rx_nr_rings;
+	hwr.rss_ctx = bnxt_get_total_rss_ctxs(bp, &hwr);
+	hwr.stat = bnxt_get_func_stat_ctxs(bp);
+	old_rx_rings = bp->hw_resc.resv_rx_rings;
+
+	rc = bnxt_hwrm_reserve_rings(bp, &hwr);
 	if (rc)
 		return rc;
 
-	mutex_lock(&bp->hwrm_cmd_lock);
-	rc = __bnxt_hwrm_get_tx_rings(bp, 0xffff, tx_rings);
-	mutex_unlock(&bp->hwrm_cmd_lock);
+	bnxt_copy_reserved_rings(bp, &hwr);
+
+	rx_rings = hwr.rx;
+	if (bp->flags & BNXT_FLAG_AGG_RINGS) {
+		if (hwr.rx >= 2) {
+			rx_rings = hwr.rx >> 1;
+		} else {
+			if (netif_running(bp->dev))
+				return -ENOMEM;
+
+			bp->flags &= ~BNXT_FLAG_AGG_RINGS;
+			bp->flags |= BNXT_FLAG_NO_AGG_RINGS;
+			bp->dev->hw_features &= ~NETIF_F_LRO;
+			bp->dev->features &= ~NETIF_F_LRO;
+			bnxt_set_ring_params(bp);
+		}
+	}
+	rx_rings = min_t(int, rx_rings, hwr.grp);
+	hwr.cp = min_t(int, hwr.cp, bp->cp_nr_rings);
+	if (bnxt_ulp_registered(bp->edev) &&
+	    hwr.stat > bnxt_get_ulp_stat_ctxs(bp))
+		hwr.stat -= bnxt_get_ulp_stat_ctxs(bp);
+	hwr.cp = min_t(int, hwr.cp, hwr.stat);
+	rc = bnxt_trim_rings(bp, &rx_rings, &hwr.tx, hwr.cp, sh);
+	if (bp->flags & BNXT_FLAG_AGG_RINGS)
+		hwr.rx = rx_rings << 1;
+	tx_cp = bnxt_num_tx_to_cp(bp, hwr.tx);
+	hwr.cp = sh ? max_t(int, tx_cp, rx_rings) : tx_cp + rx_rings;
+	if (hwr.tx != bp->tx_nr_rings) {
+		netdev_warn(bp->dev,
+			    "Able to reserve only %d out of %d requested TX rings\n",
+			    hwr.tx, bp->tx_nr_rings);
+	}
+	bp->tx_nr_rings = hwr.tx;
+
+	/* If we cannot reserve all the RX rings, reset the RSS map only
+	 * if absolutely necessary
+	 */
+	if (rx_rings != bp->rx_nr_rings) {
+		netdev_warn(bp->dev, "Able to reserve only %d out of %d requested RX rings\n",
+			    rx_rings, bp->rx_nr_rings);
+		if (netif_is_rxfh_configured(bp->dev) &&
+		    (bnxt_get_nr_rss_ctxs(bp, bp->rx_nr_rings) !=
+		     bnxt_get_nr_rss_ctxs(bp, rx_rings) ||
+		     bnxt_get_max_rss_ring(bp) >= rx_rings)) {
+			netdev_warn(bp->dev, "RSS table entries reverting to default\n");
+			bp->dev->priv_flags &= ~IFF_RXFH_CONFIGURED;
+		}
+	}
+	bp->rx_nr_rings = rx_rings;
+	bp->cp_nr_rings = hwr.cp;
+
+	if (!bnxt_rings_ok(bp, &hwr))
+		return -ENOMEM;
+
+	if (old_rx_rings != bp->hw_resc.resv_rx_rings &&
+	    !netif_is_rxfh_configured(bp->dev))
+		bnxt_set_dflt_rss_indir_tbl(bp, NULL);
+
+	if (!bnxt_ulp_registered(bp->edev) && BNXT_NEW_RM(bp)) {
+		int resv_msix, resv_ctx, ulp_ctxs;
+		struct bnxt_hw_resc *hw_resc;
+
+		hw_resc = &bp->hw_resc;
+		resv_msix = hw_resc->resv_irqs - bp->cp_nr_rings;
+		ulp_msix = min_t(int, resv_msix, ulp_msix);
+		bnxt_set_ulp_msix_num(bp, ulp_msix);
+		resv_ctx = hw_resc->resv_stat_ctxs  - bp->cp_nr_rings;
+		ulp_ctxs = min(resv_ctx, bnxt_get_ulp_stat_ctxs(bp));
+		bnxt_set_ulp_stat_ctxs(bp, ulp_ctxs);
+	}
+
 	return rc;
 }
 
-static void bnxt_hwrm_set_coal_params(struct bnxt *bp, u32 max_bufs,
-	u32 buf_tmrs, u16 flags,
-	struct hwrm_ring_cmpl_ring_cfg_aggint_params_input *req)
+static int bnxt_hwrm_check_vf_rings(struct bnxt *bp, struct bnxt_hw_rings *hwr)
 {
-	req->flags = cpu_to_le16(flags);
-	req->num_cmpl_dma_aggr = cpu_to_le16((u16)max_bufs);
-	req->num_cmpl_dma_aggr_during_int = cpu_to_le16(max_bufs >> 16);
-	req->cmpl_aggr_dma_tmr = cpu_to_le16((u16)buf_tmrs);
-	req->cmpl_aggr_dma_tmr_during_int = cpu_to_le16(buf_tmrs >> 16);
-	/* Minimum time between 2 interrupts set to buf_tmr x 2 */
-	req->int_lat_tmr_min = cpu_to_le16((u16)buf_tmrs * 2);
-	req->int_lat_tmr_max = cpu_to_le16((u16)buf_tmrs * 4);
-	req->num_cmpl_aggr_int = cpu_to_le16((u16)max_bufs * 4);
+	struct hwrm_func_vf_cfg_input *req;
+	u32 flags;
+
+	if (!BNXT_NEW_RM(bp))
+		return 0;
+
+	req = __bnxt_hwrm_reserve_vf_rings(bp, hwr);
+	flags = FUNC_VF_CFG_REQ_FLAGS_TX_ASSETS_TEST |
+		FUNC_VF_CFG_REQ_FLAGS_RX_ASSETS_TEST |
+		FUNC_VF_CFG_REQ_FLAGS_CMPL_ASSETS_TEST |
+		FUNC_VF_CFG_REQ_FLAGS_STAT_CTX_ASSETS_TEST |
+		FUNC_VF_CFG_REQ_FLAGS_VNIC_ASSETS_TEST |
+		FUNC_VF_CFG_REQ_FLAGS_RSSCOS_CTX_ASSETS_TEST;
+	if (!(bp->flags & BNXT_FLAG_CHIP_P5_PLUS))
+		flags |= FUNC_VF_CFG_REQ_FLAGS_RING_GRP_ASSETS_TEST;
+
+	req->flags = cpu_to_le32(flags);
+	return hwrm_req_send_silent(bp, req);
 }
 
-int bnxt_hwrm_set_coal(struct bnxt *bp)
+static int bnxt_hwrm_check_pf_rings(struct bnxt *bp, struct bnxt_hw_rings *hwr)
 {
-	int i, rc = 0;
-	struct hwrm_ring_cmpl_ring_cfg_aggint_params_input req_rx = {0},
-							   req_tx = {0}, *req;
-	u16 max_buf, max_buf_irq;
-	u16 buf_tmr, buf_tmr_irq;
+	struct hwrm_func_cfg_input *req;
 	u32 flags;
 
-	bnxt_hwrm_cmd_hdr_init(bp, &req_rx,
-			       HWRM_RING_CMPL_RING_CFG_AGGINT_PARAMS, -1, -1);
-	bnxt_hwrm_cmd_hdr_init(bp, &req_tx,
-			       HWRM_RING_CMPL_RING_CFG_AGGINT_PARAMS, -1, -1);
+	req = __bnxt_hwrm_reserve_pf_rings(bp, hwr);
+	flags = FUNC_CFG_REQ_FLAGS_TX_ASSETS_TEST;
+	if (BNXT_NEW_RM(bp)) {
+		flags |= FUNC_CFG_REQ_FLAGS_RX_ASSETS_TEST |
+			 FUNC_CFG_REQ_FLAGS_CMPL_ASSETS_TEST |
+			 FUNC_CFG_REQ_FLAGS_STAT_CTX_ASSETS_TEST |
+			 FUNC_CFG_REQ_FLAGS_VNIC_ASSETS_TEST;
+		if (bp->flags & BNXT_FLAG_CHIP_P5_PLUS)
+			flags |= FUNC_CFG_REQ_FLAGS_RSSCOS_CTX_ASSETS_TEST |
+				 FUNC_CFG_REQ_FLAGS_NQ_ASSETS_TEST;
+		else
+			flags |= FUNC_CFG_REQ_FLAGS_RING_GRP_ASSETS_TEST;
+	}
+
+	req->flags = cpu_to_le32(flags);
+	return hwrm_req_send_silent(bp, req);
+}
+
+static int bnxt_hwrm_check_rings(struct bnxt *bp, struct bnxt_hw_rings *hwr)
+{
+	if (bp->hwrm_spec_code < 0x10801)
+		return 0;
+
+	if (BNXT_PF(bp))
+		return bnxt_hwrm_check_pf_rings(bp, hwr);
+
+	return bnxt_hwrm_check_vf_rings(bp, hwr);
+}
+
+static void bnxt_hwrm_coal_params_qcaps(struct bnxt *bp)
+{
+	struct bnxt_coal_cap *coal_cap = &bp->coal_cap;
+	struct hwrm_ring_aggint_qcaps_output *resp;
+	struct hwrm_ring_aggint_qcaps_input *req;
+	int rc;
+
+	coal_cap->cmpl_params = BNXT_LEGACY_COAL_CMPL_PARAMS;
+	coal_cap->num_cmpl_dma_aggr_max = 63;
+	coal_cap->num_cmpl_dma_aggr_during_int_max = 63;
+	coal_cap->cmpl_aggr_dma_tmr_max = 65535;
+	coal_cap->cmpl_aggr_dma_tmr_during_int_max = 65535;
+	coal_cap->int_lat_tmr_min_max = 65535;
+	coal_cap->int_lat_tmr_max_max = 65535;
+	coal_cap->num_cmpl_aggr_int_max = 65535;
+	coal_cap->timer_units = 80;
+
+	if (bp->hwrm_spec_code < 0x10902)
+		return;
+
+	if (hwrm_req_init(bp, req, HWRM_RING_AGGINT_QCAPS))
+		return;
+
+	resp = hwrm_req_hold(bp, req);
+	rc = hwrm_req_send_silent(bp, req);
+	if (!rc) {
+		coal_cap->cmpl_params = le32_to_cpu(resp->cmpl_params);
+		coal_cap->nq_params = le32_to_cpu(resp->nq_params);
+		coal_cap->num_cmpl_dma_aggr_max =
+			le16_to_cpu(resp->num_cmpl_dma_aggr_max);
+		coal_cap->num_cmpl_dma_aggr_during_int_max =
+			le16_to_cpu(resp->num_cmpl_dma_aggr_during_int_max);
+		coal_cap->cmpl_aggr_dma_tmr_max =
+			le16_to_cpu(resp->cmpl_aggr_dma_tmr_max);
+		coal_cap->cmpl_aggr_dma_tmr_during_int_max =
+			le16_to_cpu(resp->cmpl_aggr_dma_tmr_during_int_max);
+		coal_cap->int_lat_tmr_min_max =
+			le16_to_cpu(resp->int_lat_tmr_min_max);
+		coal_cap->int_lat_tmr_max_max =
+			le16_to_cpu(resp->int_lat_tmr_max_max);
+		coal_cap->num_cmpl_aggr_int_max =
+			le16_to_cpu(resp->num_cmpl_aggr_int_max);
+		coal_cap->timer_units = le16_to_cpu(resp->timer_units);
+	}
+	hwrm_req_drop(bp, req);
+}
+
+static u16 bnxt_usec_to_coal_tmr(struct bnxt *bp, u16 usec)
+{
+	struct bnxt_coal_cap *coal_cap = &bp->coal_cap;
+
+	return usec * 1000 / coal_cap->timer_units;
+}
+
+static void bnxt_hwrm_set_coal_params(struct bnxt *bp,
+	struct bnxt_coal *hw_coal,
+	struct hwrm_ring_cmpl_ring_cfg_aggint_params_input *req)
+{
+	struct bnxt_coal_cap *coal_cap = &bp->coal_cap;
+	u16 val, tmr, max, flags = hw_coal->flags;
+	u32 cmpl_params = coal_cap->cmpl_params;
+
+	max = hw_coal->bufs_per_record * 128;
+	if (hw_coal->budget)
+		max = hw_coal->bufs_per_record * hw_coal->budget;
+	max = min_t(u16, max, coal_cap->num_cmpl_aggr_int_max);
+
+	val = clamp_t(u16, hw_coal->coal_bufs, 1, max);
+	req->num_cmpl_aggr_int = cpu_to_le16(val);
+
+	val = min_t(u16, val, coal_cap->num_cmpl_dma_aggr_max);
+	req->num_cmpl_dma_aggr = cpu_to_le16(val);
+
+	val = clamp_t(u16, hw_coal->coal_bufs_irq, 1,
+		      coal_cap->num_cmpl_dma_aggr_during_int_max);
+	req->num_cmpl_dma_aggr_during_int = cpu_to_le16(val);
+
+	tmr = bnxt_usec_to_coal_tmr(bp, hw_coal->coal_ticks);
+	tmr = clamp_t(u16, tmr, 1, coal_cap->int_lat_tmr_max_max);
+	req->int_lat_tmr_max = cpu_to_le16(tmr);
+
+	/* min timer set to 1/2 of interrupt timer */
+	if (cmpl_params & RING_AGGINT_QCAPS_RESP_CMPL_PARAMS_INT_LAT_TMR_MIN) {
+		val = tmr / 2;
+		val = clamp_t(u16, val, 1, coal_cap->int_lat_tmr_min_max);
+		req->int_lat_tmr_min = cpu_to_le16(val);
+		req->enables |= cpu_to_le16(BNXT_COAL_CMPL_MIN_TMR_ENABLE);
+	}
 
-	/* Each rx completion (2 records) should be DMAed immediately.
-	 * DMA 1/4 of the completion buffers at a time.
-	 */
-	max_buf = min_t(u16, bp->rx_coal_bufs / 4, 2);
-	/* max_buf must not be zero */
-	max_buf = clamp_t(u16, max_buf, 1, 63);
-	max_buf_irq = clamp_t(u16, bp->rx_coal_bufs_irq, 1, 63);
-	buf_tmr = BNXT_USEC_TO_COAL_TIMER(bp->rx_coal_ticks);
 	/* buf timer set to 1/4 of interrupt timer */
-	buf_tmr = max_t(u16, buf_tmr / 4, 1);
-	buf_tmr_irq = BNXT_USEC_TO_COAL_TIMER(bp->rx_coal_ticks_irq);
-	buf_tmr_irq = max_t(u16, buf_tmr_irq, 1);
+	val = clamp_t(u16, tmr / 4, 1, coal_cap->cmpl_aggr_dma_tmr_max);
+	req->cmpl_aggr_dma_tmr = cpu_to_le16(val);
+
+	if (cmpl_params &
+	    RING_AGGINT_QCAPS_RESP_CMPL_PARAMS_NUM_CMPL_DMA_AGGR_DURING_INT) {
+		tmr = bnxt_usec_to_coal_tmr(bp, hw_coal->coal_ticks_irq);
+		val = clamp_t(u16, tmr, 1,
+			      coal_cap->cmpl_aggr_dma_tmr_during_int_max);
+		req->cmpl_aggr_dma_tmr_during_int = cpu_to_le16(val);
+		req->enables |=
+			cpu_to_le16(BNXT_COAL_CMPL_AGGR_TMR_DURING_INT_ENABLE);
+	}
+
+	if ((cmpl_params & RING_AGGINT_QCAPS_RESP_CMPL_PARAMS_RING_IDLE) &&
+	    hw_coal->idle_thresh && hw_coal->coal_ticks < hw_coal->idle_thresh)
+		flags |= RING_CMPL_RING_CFG_AGGINT_PARAMS_REQ_FLAGS_RING_IDLE;
+	req->flags = cpu_to_le16(flags);
+	req->enables |= cpu_to_le16(BNXT_COAL_CMPL_ENABLES);
+}
+
+static int __bnxt_hwrm_set_coal_nq(struct bnxt *bp, struct bnxt_napi *bnapi,
+				   struct bnxt_coal *hw_coal)
+{
+	struct hwrm_ring_cmpl_ring_cfg_aggint_params_input *req;
+	struct bnxt_cp_ring_info *cpr = &bnapi->cp_ring;
+	struct bnxt_coal_cap *coal_cap = &bp->coal_cap;
+	u32 nq_params = coal_cap->nq_params;
+	u16 tmr;
+	int rc;
 
-	flags = RING_CMPL_RING_CFG_AGGINT_PARAMS_REQ_FLAGS_TIMER_RESET;
+	if (!(nq_params & RING_AGGINT_QCAPS_RESP_NQ_PARAMS_INT_LAT_TMR_MIN))
+		return 0;
+
+	rc = hwrm_req_init(bp, req, HWRM_RING_CMPL_RING_CFG_AGGINT_PARAMS);
+	if (rc)
+		return rc;
+
+	req->ring_id = cpu_to_le16(cpr->cp_ring_struct.fw_ring_id);
+	req->flags =
+		cpu_to_le16(RING_CMPL_RING_CFG_AGGINT_PARAMS_REQ_FLAGS_IS_NQ);
 
-	/* RING_IDLE generates more IRQs for lower latency.  Enable it only
-	 * if coal_ticks is less than 25 us.
+	tmr = bnxt_usec_to_coal_tmr(bp, hw_coal->coal_ticks) / 2;
+	tmr = clamp_t(u16, tmr, 1, coal_cap->int_lat_tmr_min_max);
+	req->int_lat_tmr_min = cpu_to_le16(tmr);
+	req->enables |= cpu_to_le16(BNXT_COAL_CMPL_MIN_TMR_ENABLE);
+	return hwrm_req_send(bp, req);
+}
+
+int bnxt_hwrm_set_ring_coal(struct bnxt *bp, struct bnxt_napi *bnapi)
+{
+	struct hwrm_ring_cmpl_ring_cfg_aggint_params_input *req_rx;
+	struct bnxt_cp_ring_info *cpr = &bnapi->cp_ring;
+	struct bnxt_coal coal;
+	int rc;
+
+	/* Tick values in micro seconds.
+	 * 1 coal_buf x bufs_per_record = 1 completion record.
 	 */
-	if (bp->rx_coal_ticks < 25)
-		flags |= RING_CMPL_RING_CFG_AGGINT_PARAMS_REQ_FLAGS_RING_IDLE;
+	memcpy(&coal, &bp->rx_coal, sizeof(struct bnxt_coal));
 
-	bnxt_hwrm_set_coal_params(bp, max_buf_irq << 16 | max_buf,
-				  buf_tmr_irq << 16 | buf_tmr, flags, &req_rx);
+	coal.coal_ticks = cpr->rx_ring_coal.coal_ticks;
+	coal.coal_bufs = cpr->rx_ring_coal.coal_bufs;
 
-	/* max_buf must not be zero */
-	max_buf = clamp_t(u16, bp->tx_coal_bufs, 1, 63);
-	max_buf_irq = clamp_t(u16, bp->tx_coal_bufs_irq, 1, 63);
-	buf_tmr = BNXT_USEC_TO_COAL_TIMER(bp->tx_coal_ticks);
-	/* buf timer set to 1/4 of interrupt timer */
-	buf_tmr = max_t(u16, buf_tmr / 4, 1);
-	buf_tmr_irq = BNXT_USEC_TO_COAL_TIMER(bp->tx_coal_ticks_irq);
-	buf_tmr_irq = max_t(u16, buf_tmr_irq, 1);
+	if (!bnapi->rx_ring)
+		return -ENODEV;
+
+	rc = hwrm_req_init(bp, req_rx, HWRM_RING_CMPL_RING_CFG_AGGINT_PARAMS);
+	if (rc)
+		return rc;
+
+	bnxt_hwrm_set_coal_params(bp, &coal, req_rx);
+
+	req_rx->ring_id = cpu_to_le16(bnxt_cp_ring_for_rx(bp, bnapi->rx_ring));
+
+	return hwrm_req_send(bp, req_rx);
+}
+
+static int
+bnxt_hwrm_set_rx_coal(struct bnxt *bp, struct bnxt_napi *bnapi,
+		      struct hwrm_ring_cmpl_ring_cfg_aggint_params_input *req)
+{
+	u16 ring_id = bnxt_cp_ring_for_rx(bp, bnapi->rx_ring);
+
+	req->ring_id = cpu_to_le16(ring_id);
+	return hwrm_req_send(bp, req);
+}
+
+static int
+bnxt_hwrm_set_tx_coal(struct bnxt *bp, struct bnxt_napi *bnapi,
+		      struct hwrm_ring_cmpl_ring_cfg_aggint_params_input *req)
+{
+	struct bnxt_tx_ring_info *txr;
+	int i, rc;
 
-	flags = RING_CMPL_RING_CFG_AGGINT_PARAMS_REQ_FLAGS_TIMER_RESET;
-	bnxt_hwrm_set_coal_params(bp, max_buf_irq << 16 | max_buf,
-				  buf_tmr_irq << 16 | buf_tmr, flags, &req_tx);
+	bnxt_for_each_napi_tx(i, bnapi, txr) {
+		u16 ring_id;
 
-	mutex_lock(&bp->hwrm_cmd_lock);
+		ring_id = bnxt_cp_ring_for_tx(bp, txr);
+		req->ring_id = cpu_to_le16(ring_id);
+		rc = hwrm_req_send(bp, req);
+		if (rc)
+			return rc;
+		if (!(bp->flags & BNXT_FLAG_CHIP_P5_PLUS))
+			return 0;
+	}
+	return 0;
+}
+
+int bnxt_hwrm_set_coal(struct bnxt *bp)
+{
+	struct hwrm_ring_cmpl_ring_cfg_aggint_params_input *req_rx, *req_tx;
+	int i, rc;
+
+	rc = hwrm_req_init(bp, req_rx, HWRM_RING_CMPL_RING_CFG_AGGINT_PARAMS);
+	if (rc)
+		return rc;
+
+	rc = hwrm_req_init(bp, req_tx, HWRM_RING_CMPL_RING_CFG_AGGINT_PARAMS);
+	if (rc) {
+		hwrm_req_drop(bp, req_rx);
+		return rc;
+	}
+
+	bnxt_hwrm_set_coal_params(bp, &bp->rx_coal, req_rx);
+	bnxt_hwrm_set_coal_params(bp, &bp->tx_coal, req_tx);
+
+	hwrm_req_hold(bp, req_rx);
+	hwrm_req_hold(bp, req_tx);
 	for (i = 0; i < bp->cp_nr_rings; i++) {
 		struct bnxt_napi *bnapi = bp->bnapi[i];
+		struct bnxt_coal *hw_coal;
 
-		req = &req_rx;
 		if (!bnapi->rx_ring)
-			req = &req_tx;
-		req->ring_id = cpu_to_le16(bp->grp_info[i].cp_fw_ring_id);
-
-		rc = _hwrm_send_message(bp, req, sizeof(*req),
-					HWRM_CMD_TIMEOUT);
+			rc = bnxt_hwrm_set_tx_coal(bp, bnapi, req_tx);
+		else
+			rc = bnxt_hwrm_set_rx_coal(bp, bnapi, req_rx);
 		if (rc)
 			break;
+
+		if (!(bp->flags & BNXT_FLAG_CHIP_P5_PLUS))
+			continue;
+
+		if (bnapi->rx_ring && bnapi->tx_ring[0]) {
+			rc = bnxt_hwrm_set_tx_coal(bp, bnapi, req_tx);
+			if (rc)
+				break;
+		}
+		if (bnapi->rx_ring)
+			hw_coal = &bp->rx_coal;
+		else
+			hw_coal = &bp->tx_coal;
+		__bnxt_hwrm_set_coal_nq(bp, bnapi, hw_coal);
 	}
-	mutex_unlock(&bp->hwrm_cmd_lock);
+	hwrm_req_drop(bp, req_rx);
+	hwrm_req_drop(bp, req_tx);
 	return rc;
 }
 
-static int bnxt_hwrm_stat_ctx_free(struct bnxt *bp)
+static void bnxt_hwrm_stat_ctx_free(struct bnxt *bp)
 {
-	int rc = 0, i;
-	struct hwrm_stat_ctx_free_input req = {0};
+	struct hwrm_stat_ctx_clr_stats_input *req0 = NULL;
+	struct hwrm_stat_ctx_free_input *req;
+	int i;
 
 	if (!bp->bnapi)
-		return 0;
+		return;
 
 	if (BNXT_CHIP_TYPE_NITRO_A0(bp))
-		return 0;
-
-	bnxt_hwrm_cmd_hdr_init(bp, &req, HWRM_STAT_CTX_FREE, -1, -1);
+		return;
 
-	mutex_lock(&bp->hwrm_cmd_lock);
+	if (hwrm_req_init(bp, req, HWRM_STAT_CTX_FREE))
+		return;
+	if (BNXT_FW_MAJ(bp) <= 20) {
+		if (hwrm_req_init(bp, req0, HWRM_STAT_CTX_CLR_STATS)) {
+			hwrm_req_drop(bp, req);
+			return;
+		}
+		hwrm_req_hold(bp, req0);
+	}
+	hwrm_req_hold(bp, req);
 	for (i = 0; i < bp->cp_nr_rings; i++) {
 		struct bnxt_napi *bnapi = bp->bnapi[i];
 		struct bnxt_cp_ring_info *cpr = &bnapi->cp_ring;
 
 		if (cpr->hw_stats_ctx_id != INVALID_STATS_CTX_ID) {
-			req.stat_ctx_id = cpu_to_le32(cpr->hw_stats_ctx_id);
-
-			rc = _hwrm_send_message(bp, &req, sizeof(req),
-						HWRM_CMD_TIMEOUT);
-			if (rc)
-				break;
+			req->stat_ctx_id = cpu_to_le32(cpr->hw_stats_ctx_id);
+			if (req0) {
+				req0->stat_ctx_id = req->stat_ctx_id;
+				hwrm_req_send(bp, req0);
+			}
+			hwrm_req_send(bp, req);
 
 			cpr->hw_stats_ctx_id = INVALID_STATS_CTX_ID;
 		}
 	}
-	mutex_unlock(&bp->hwrm_cmd_lock);
-	return rc;
+	hwrm_req_drop(bp, req);
+	if (req0)
+		hwrm_req_drop(bp, req0);
 }
 
 static int bnxt_hwrm_stat_ctx_alloc(struct bnxt *bp)
 {
-	int rc = 0, i;
-	struct hwrm_stat_ctx_alloc_input req = {0};
-	struct hwrm_stat_ctx_alloc_output *resp = bp->hwrm_cmd_resp_addr;
+	struct hwrm_stat_ctx_alloc_output *resp;
+	struct hwrm_stat_ctx_alloc_input *req;
+	int rc, i;
 
 	if (BNXT_CHIP_TYPE_NITRO_A0(bp))
 		return 0;
 
-	bnxt_hwrm_cmd_hdr_init(bp, &req, HWRM_STAT_CTX_ALLOC, -1, -1);
+	rc = hwrm_req_init(bp, req, HWRM_STAT_CTX_ALLOC);
+	if (rc)
+		return rc;
 
-	req.update_period_ms = cpu_to_le32(bp->stats_coal_ticks / 1000);
+	req->stats_dma_length = cpu_to_le16(bp->hw_ring_stats_size);
+	req->update_period_ms = cpu_to_le32(bp->stats_coal_ticks / 1000);
 
-	mutex_lock(&bp->hwrm_cmd_lock);
+	resp = hwrm_req_hold(bp, req);
 	for (i = 0; i < bp->cp_nr_rings; i++) {
 		struct bnxt_napi *bnapi = bp->bnapi[i];
 		struct bnxt_cp_ring_info *cpr = &bnapi->cp_ring;
 
-		req.stats_dma_addr = cpu_to_le64(cpr->hw_stats_map);
+		req->stats_dma_addr = cpu_to_le64(cpr->stats.hw_stats_map);
 
-		rc = _hwrm_send_message(bp, &req, sizeof(req),
-					HWRM_CMD_TIMEOUT);
+		rc = hwrm_req_send(bp, req);
 		if (rc)
 			break;
 
@@ -4520,163 +8457,1651 @@ static int bnxt_hwrm_stat_ctx_alloc(struct bnxt *bp)
 
 		bp->grp_info[i].fw_stats_ctx = cpr->hw_stats_ctx_id;
 	}
-	mutex_unlock(&bp->hwrm_cmd_lock);
+	hwrm_req_drop(bp, req);
 	return rc;
 }
 
 static int bnxt_hwrm_func_qcfg(struct bnxt *bp)
 {
-	struct hwrm_func_qcfg_input req = {0};
-	struct hwrm_func_qcfg_output *resp = bp->hwrm_cmd_resp_addr;
+	struct hwrm_func_qcfg_output *resp;
+	struct hwrm_func_qcfg_input *req;
+	u16 flags;
 	int rc;
 
-	bnxt_hwrm_cmd_hdr_init(bp, &req, HWRM_FUNC_QCFG, -1, -1);
-	req.fid = cpu_to_le16(0xffff);
-	mutex_lock(&bp->hwrm_cmd_lock);
-	rc = _hwrm_send_message(bp, &req, sizeof(req), HWRM_CMD_TIMEOUT);
+	rc = hwrm_req_init(bp, req, HWRM_FUNC_QCFG);
+	if (rc)
+		return rc;
+
+	req->fid = cpu_to_le16(0xffff);
+	resp = hwrm_req_hold(bp, req);
+	rc = hwrm_req_send(bp, req);
 	if (rc)
 		goto func_qcfg_exit;
 
+	flags = le16_to_cpu(resp->flags);
 #ifdef CONFIG_BNXT_SRIOV
 	if (BNXT_VF(bp)) {
 		struct bnxt_vf_info *vf = &bp->vf;
 
 		vf->vlan = le16_to_cpu(resp->vlan) & VLAN_VID_MASK;
+		if (flags & FUNC_QCFG_RESP_FLAGS_TRUSTED_VF)
+			vf->flags |= BNXT_VF_TRUST;
+		else
+			vf->flags &= ~BNXT_VF_TRUST;
+	} else {
+		bp->pf.registered_vfs = le16_to_cpu(resp->registered_vfs);
 	}
 #endif
-	if (BNXT_PF(bp)) {
-		u16 flags = le16_to_cpu(resp->flags);
-
-		if (flags & (FUNC_QCFG_RESP_FLAGS_FW_DCBX_AGENT_ENABLED |
-			     FUNC_QCFG_RESP_FLAGS_FW_LLDP_AGENT_ENABLED))
-			bp->flags |= BNXT_FLAG_FW_LLDP_AGENT;
-		if (flags & FUNC_QCFG_RESP_FLAGS_MULTI_HOST)
-			bp->flags |= BNXT_FLAG_MULTI_HOST;
+	if (flags & (FUNC_QCFG_RESP_FLAGS_FW_DCBX_AGENT_ENABLED |
+		     FUNC_QCFG_RESP_FLAGS_FW_LLDP_AGENT_ENABLED)) {
+		bp->fw_cap |= BNXT_FW_CAP_LLDP_AGENT;
+		if (flags & FUNC_QCFG_RESP_FLAGS_FW_DCBX_AGENT_ENABLED)
+			bp->fw_cap |= BNXT_FW_CAP_DCBX_AGENT;
 	}
+	if (BNXT_PF(bp) && (flags & FUNC_QCFG_RESP_FLAGS_MULTI_HOST))
+		bp->flags |= BNXT_FLAG_MULTI_HOST;
+
+	if (flags & FUNC_QCFG_RESP_FLAGS_RING_MONITOR_ENABLED)
+		bp->fw_cap |= BNXT_FW_CAP_RING_MONITOR;
+
+	if (flags & FUNC_QCFG_RESP_FLAGS_ENABLE_RDMA_SRIOV)
+		bp->fw_cap |= BNXT_FW_CAP_ENABLE_RDMA_SRIOV;
 
 	switch (resp->port_partition_type) {
 	case FUNC_QCFG_RESP_PORT_PARTITION_TYPE_NPAR1_0:
+	case FUNC_QCFG_RESP_PORT_PARTITION_TYPE_NPAR1_2:
 	case FUNC_QCFG_RESP_PORT_PARTITION_TYPE_NPAR1_5:
 	case FUNC_QCFG_RESP_PORT_PARTITION_TYPE_NPAR2_0:
 		bp->port_partition_type = resp->port_partition_type;
 		break;
 	}
+	if (bp->hwrm_spec_code < 0x10707 ||
+	    resp->evb_mode == FUNC_QCFG_RESP_EVB_MODE_VEB)
+		bp->br_mode = BRIDGE_MODE_VEB;
+	else if (resp->evb_mode == FUNC_QCFG_RESP_EVB_MODE_VEPA)
+		bp->br_mode = BRIDGE_MODE_VEPA;
+	else
+		bp->br_mode = BRIDGE_MODE_UNDEF;
+
+	bp->max_mtu = le16_to_cpu(resp->max_mtu_configured);
+	if (!bp->max_mtu)
+		bp->max_mtu = BNXT_MAX_MTU;
+
+	if (bp->db_size)
+		goto func_qcfg_exit;
+
+	bp->db_offset = le16_to_cpu(resp->legacy_l2_db_size_kb) * 1024;
+	if (BNXT_CHIP_P5(bp)) {
+		if (BNXT_PF(bp))
+			bp->db_offset = DB_PF_OFFSET_P5;
+		else
+			bp->db_offset = DB_VF_OFFSET_P5;
+	}
+	bp->db_size = PAGE_ALIGN(le16_to_cpu(resp->l2_doorbell_bar_size_kb) *
+				 1024);
+	if (!bp->db_size || bp->db_size > pci_resource_len(bp->pdev, 2) ||
+	    bp->db_size <= bp->db_offset)
+		bp->db_size = pci_resource_len(bp->pdev, 2);
 
 func_qcfg_exit:
-	mutex_unlock(&bp->hwrm_cmd_lock);
+	hwrm_req_drop(bp, req);
+	return rc;
+}
+
+static void bnxt_init_ctx_initializer(struct bnxt_ctx_mem_type *ctxm,
+				      u8 init_val, u8 init_offset,
+				      bool init_mask_set)
+{
+	ctxm->init_value = init_val;
+	ctxm->init_offset = BNXT_CTX_INIT_INVALID_OFFSET;
+	if (init_mask_set)
+		ctxm->init_offset = init_offset * 4;
+	else
+		ctxm->init_value = 0;
+}
+
+static int bnxt_alloc_all_ctx_pg_info(struct bnxt *bp, int ctx_max)
+{
+	struct bnxt_ctx_mem_info *ctx = bp->ctx;
+	u16 type;
+
+	for (type = 0; type < ctx_max; type++) {
+		struct bnxt_ctx_mem_type *ctxm = &ctx->ctx_arr[type];
+		int n = 1;
+
+		if (!ctxm->max_entries || ctxm->pg_info)
+			continue;
+
+		if (ctxm->instance_bmap)
+			n = hweight32(ctxm->instance_bmap);
+		ctxm->pg_info = kcalloc(n, sizeof(*ctxm->pg_info), GFP_KERNEL);
+		if (!ctxm->pg_info)
+			return -ENOMEM;
+	}
+	return 0;
+}
+
+static void bnxt_free_one_ctx_mem(struct bnxt *bp,
+				  struct bnxt_ctx_mem_type *ctxm, bool force);
+
+#define BNXT_CTX_INIT_VALID(flags)	\
+	(!!((flags) &			\
+	    FUNC_BACKING_STORE_QCAPS_V2_RESP_FLAGS_ENABLE_CTX_KIND_INIT))
+
+static int bnxt_hwrm_func_backing_store_qcaps_v2(struct bnxt *bp)
+{
+	struct hwrm_func_backing_store_qcaps_v2_output *resp;
+	struct hwrm_func_backing_store_qcaps_v2_input *req;
+	struct bnxt_ctx_mem_info *ctx = bp->ctx;
+	u16 type;
+	int rc;
+
+	rc = hwrm_req_init(bp, req, HWRM_FUNC_BACKING_STORE_QCAPS_V2);
+	if (rc)
+		return rc;
+
+	if (!ctx) {
+		ctx = kzalloc(sizeof(*ctx), GFP_KERNEL);
+		if (!ctx)
+			return -ENOMEM;
+		bp->ctx = ctx;
+	}
+
+	resp = hwrm_req_hold(bp, req);
+
+	for (type = 0; type < BNXT_CTX_V2_MAX; ) {
+		struct bnxt_ctx_mem_type *ctxm = &ctx->ctx_arr[type];
+		u8 init_val, init_off, i;
+		u32 max_entries;
+		u16 entry_size;
+		__le32 *p;
+		u32 flags;
+
+		req->type = cpu_to_le16(type);
+		rc = hwrm_req_send(bp, req);
+		if (rc)
+			goto ctx_done;
+		flags = le32_to_cpu(resp->flags);
+		type = le16_to_cpu(resp->next_valid_type);
+		if (!(flags & BNXT_CTX_MEM_TYPE_VALID)) {
+			bnxt_free_one_ctx_mem(bp, ctxm, true);
+			continue;
+		}
+		entry_size = le16_to_cpu(resp->entry_size);
+		max_entries = le32_to_cpu(resp->max_num_entries);
+		if (ctxm->mem_valid) {
+			if (!(flags & BNXT_CTX_MEM_PERSIST) ||
+			    ctxm->entry_size != entry_size ||
+			    ctxm->max_entries != max_entries)
+				bnxt_free_one_ctx_mem(bp, ctxm, true);
+			else
+				continue;
+		}
+		ctxm->type = le16_to_cpu(resp->type);
+		ctxm->entry_size = entry_size;
+		ctxm->flags = flags;
+		ctxm->instance_bmap = le32_to_cpu(resp->instance_bit_map);
+		ctxm->entry_multiple = resp->entry_multiple;
+		ctxm->max_entries = max_entries;
+		ctxm->min_entries = le32_to_cpu(resp->min_num_entries);
+		init_val = resp->ctx_init_value;
+		init_off = resp->ctx_init_offset;
+		bnxt_init_ctx_initializer(ctxm, init_val, init_off,
+					  BNXT_CTX_INIT_VALID(flags));
+		ctxm->split_entry_cnt = min_t(u8, resp->subtype_valid_cnt,
+					      BNXT_MAX_SPLIT_ENTRY);
+		for (i = 0, p = &resp->split_entry_0; i < ctxm->split_entry_cnt;
+		     i++, p++)
+			ctxm->split[i] = le32_to_cpu(*p);
+	}
+	rc = bnxt_alloc_all_ctx_pg_info(bp, BNXT_CTX_V2_MAX);
+
+ctx_done:
+	hwrm_req_drop(bp, req);
+	return rc;
+}
+
+static int bnxt_hwrm_func_backing_store_qcaps(struct bnxt *bp)
+{
+	struct hwrm_func_backing_store_qcaps_output *resp;
+	struct hwrm_func_backing_store_qcaps_input *req;
+	int rc;
+
+	if (bp->hwrm_spec_code < 0x10902 || BNXT_VF(bp) ||
+	    (bp->ctx && bp->ctx->flags & BNXT_CTX_FLAG_INITED))
+		return 0;
+
+	if (bp->fw_cap & BNXT_FW_CAP_BACKING_STORE_V2)
+		return bnxt_hwrm_func_backing_store_qcaps_v2(bp);
+
+	rc = hwrm_req_init(bp, req, HWRM_FUNC_BACKING_STORE_QCAPS);
+	if (rc)
+		return rc;
+
+	resp = hwrm_req_hold(bp, req);
+	rc = hwrm_req_send_silent(bp, req);
+	if (!rc) {
+		struct bnxt_ctx_mem_type *ctxm;
+		struct bnxt_ctx_mem_info *ctx;
+		u8 init_val, init_idx = 0;
+		u16 init_mask;
+
+		ctx = bp->ctx;
+		if (!ctx) {
+			ctx = kzalloc(sizeof(*ctx), GFP_KERNEL);
+			if (!ctx) {
+				rc = -ENOMEM;
+				goto ctx_err;
+			}
+			bp->ctx = ctx;
+		}
+		init_val = resp->ctx_kind_initializer;
+		init_mask = le16_to_cpu(resp->ctx_init_mask);
+
+		ctxm = &ctx->ctx_arr[BNXT_CTX_QP];
+		ctxm->max_entries = le32_to_cpu(resp->qp_max_entries);
+		ctxm->qp_qp1_entries = le16_to_cpu(resp->qp_min_qp1_entries);
+		ctxm->qp_l2_entries = le16_to_cpu(resp->qp_max_l2_entries);
+		ctxm->qp_fast_qpmd_entries = le16_to_cpu(resp->fast_qpmd_qp_num_entries);
+		ctxm->entry_size = le16_to_cpu(resp->qp_entry_size);
+		bnxt_init_ctx_initializer(ctxm, init_val, resp->qp_init_offset,
+					  (init_mask & (1 << init_idx++)) != 0);
+
+		ctxm = &ctx->ctx_arr[BNXT_CTX_SRQ];
+		ctxm->srq_l2_entries = le16_to_cpu(resp->srq_max_l2_entries);
+		ctxm->max_entries = le32_to_cpu(resp->srq_max_entries);
+		ctxm->entry_size = le16_to_cpu(resp->srq_entry_size);
+		bnxt_init_ctx_initializer(ctxm, init_val, resp->srq_init_offset,
+					  (init_mask & (1 << init_idx++)) != 0);
+
+		ctxm = &ctx->ctx_arr[BNXT_CTX_CQ];
+		ctxm->cq_l2_entries = le16_to_cpu(resp->cq_max_l2_entries);
+		ctxm->max_entries = le32_to_cpu(resp->cq_max_entries);
+		ctxm->entry_size = le16_to_cpu(resp->cq_entry_size);
+		bnxt_init_ctx_initializer(ctxm, init_val, resp->cq_init_offset,
+					  (init_mask & (1 << init_idx++)) != 0);
+
+		ctxm = &ctx->ctx_arr[BNXT_CTX_VNIC];
+		ctxm->vnic_entries = le16_to_cpu(resp->vnic_max_vnic_entries);
+		ctxm->max_entries = ctxm->vnic_entries +
+			le16_to_cpu(resp->vnic_max_ring_table_entries);
+		ctxm->entry_size = le16_to_cpu(resp->vnic_entry_size);
+		bnxt_init_ctx_initializer(ctxm, init_val,
+					  resp->vnic_init_offset,
+					  (init_mask & (1 << init_idx++)) != 0);
+
+		ctxm = &ctx->ctx_arr[BNXT_CTX_STAT];
+		ctxm->max_entries = le32_to_cpu(resp->stat_max_entries);
+		ctxm->entry_size = le16_to_cpu(resp->stat_entry_size);
+		bnxt_init_ctx_initializer(ctxm, init_val,
+					  resp->stat_init_offset,
+					  (init_mask & (1 << init_idx++)) != 0);
+
+		ctxm = &ctx->ctx_arr[BNXT_CTX_STQM];
+		ctxm->entry_size = le16_to_cpu(resp->tqm_entry_size);
+		ctxm->min_entries = le32_to_cpu(resp->tqm_min_entries_per_ring);
+		ctxm->max_entries = le32_to_cpu(resp->tqm_max_entries_per_ring);
+		ctxm->entry_multiple = resp->tqm_entries_multiple;
+		if (!ctxm->entry_multiple)
+			ctxm->entry_multiple = 1;
+
+		memcpy(&ctx->ctx_arr[BNXT_CTX_FTQM], ctxm, sizeof(*ctxm));
+
+		ctxm = &ctx->ctx_arr[BNXT_CTX_MRAV];
+		ctxm->max_entries = le32_to_cpu(resp->mrav_max_entries);
+		ctxm->entry_size = le16_to_cpu(resp->mrav_entry_size);
+		ctxm->mrav_num_entries_units =
+			le16_to_cpu(resp->mrav_num_entries_units);
+		bnxt_init_ctx_initializer(ctxm, init_val,
+					  resp->mrav_init_offset,
+					  (init_mask & (1 << init_idx++)) != 0);
+
+		ctxm = &ctx->ctx_arr[BNXT_CTX_TIM];
+		ctxm->entry_size = le16_to_cpu(resp->tim_entry_size);
+		ctxm->max_entries = le32_to_cpu(resp->tim_max_entries);
+
+		ctx->tqm_fp_rings_count = resp->tqm_fp_rings_count;
+		if (!ctx->tqm_fp_rings_count)
+			ctx->tqm_fp_rings_count = bp->max_q;
+		else if (ctx->tqm_fp_rings_count > BNXT_MAX_TQM_FP_RINGS)
+			ctx->tqm_fp_rings_count = BNXT_MAX_TQM_FP_RINGS;
+
+		ctxm = &ctx->ctx_arr[BNXT_CTX_FTQM];
+		memcpy(ctxm, &ctx->ctx_arr[BNXT_CTX_STQM], sizeof(*ctxm));
+		ctxm->instance_bmap = (1 << ctx->tqm_fp_rings_count) - 1;
+
+		rc = bnxt_alloc_all_ctx_pg_info(bp, BNXT_CTX_MAX);
+	} else {
+		rc = 0;
+	}
+ctx_err:
+	hwrm_req_drop(bp, req);
+	return rc;
+}
+
+static void bnxt_hwrm_set_pg_attr(struct bnxt_ring_mem_info *rmem, u8 *pg_attr,
+				  __le64 *pg_dir)
+{
+	if (!rmem->nr_pages)
+		return;
+
+	BNXT_SET_CTX_PAGE_ATTR(*pg_attr);
+	if (rmem->depth >= 1) {
+		if (rmem->depth == 2)
+			*pg_attr |= 2;
+		else
+			*pg_attr |= 1;
+		*pg_dir = cpu_to_le64(rmem->pg_tbl_map);
+	} else {
+		*pg_dir = cpu_to_le64(rmem->dma_arr[0]);
+	}
+}
+
+#define FUNC_BACKING_STORE_CFG_REQ_DFLT_ENABLES			\
+	(FUNC_BACKING_STORE_CFG_REQ_ENABLES_QP |		\
+	 FUNC_BACKING_STORE_CFG_REQ_ENABLES_SRQ |		\
+	 FUNC_BACKING_STORE_CFG_REQ_ENABLES_CQ |		\
+	 FUNC_BACKING_STORE_CFG_REQ_ENABLES_VNIC |		\
+	 FUNC_BACKING_STORE_CFG_REQ_ENABLES_STAT)
+
+static int bnxt_hwrm_func_backing_store_cfg(struct bnxt *bp, u32 enables)
+{
+	struct hwrm_func_backing_store_cfg_input *req;
+	struct bnxt_ctx_mem_info *ctx = bp->ctx;
+	struct bnxt_ctx_pg_info *ctx_pg;
+	struct bnxt_ctx_mem_type *ctxm;
+	void **__req = (void **)&req;
+	u32 req_len = sizeof(*req);
+	__le32 *num_entries;
+	__le64 *pg_dir;
+	u32 flags = 0;
+	u8 *pg_attr;
+	u32 ena;
+	int rc;
+	int i;
+
+	if (!ctx)
+		return 0;
+
+	if (req_len > bp->hwrm_max_ext_req_len)
+		req_len = BNXT_BACKING_STORE_CFG_LEGACY_LEN;
+	rc = __hwrm_req_init(bp, __req, HWRM_FUNC_BACKING_STORE_CFG, req_len);
+	if (rc)
+		return rc;
+
+	req->enables = cpu_to_le32(enables);
+	if (enables & FUNC_BACKING_STORE_CFG_REQ_ENABLES_QP) {
+		ctxm = &ctx->ctx_arr[BNXT_CTX_QP];
+		ctx_pg = ctxm->pg_info;
+		req->qp_num_entries = cpu_to_le32(ctx_pg->entries);
+		req->qp_num_qp1_entries = cpu_to_le16(ctxm->qp_qp1_entries);
+		req->qp_num_l2_entries = cpu_to_le16(ctxm->qp_l2_entries);
+		req->qp_entry_size = cpu_to_le16(ctxm->entry_size);
+		bnxt_hwrm_set_pg_attr(&ctx_pg->ring_mem,
+				      &req->qpc_pg_size_qpc_lvl,
+				      &req->qpc_page_dir);
+
+		if (enables & FUNC_BACKING_STORE_CFG_REQ_ENABLES_QP_FAST_QPMD)
+			req->qp_num_fast_qpmd_entries = cpu_to_le16(ctxm->qp_fast_qpmd_entries);
+	}
+	if (enables & FUNC_BACKING_STORE_CFG_REQ_ENABLES_SRQ) {
+		ctxm = &ctx->ctx_arr[BNXT_CTX_SRQ];
+		ctx_pg = ctxm->pg_info;
+		req->srq_num_entries = cpu_to_le32(ctx_pg->entries);
+		req->srq_num_l2_entries = cpu_to_le16(ctxm->srq_l2_entries);
+		req->srq_entry_size = cpu_to_le16(ctxm->entry_size);
+		bnxt_hwrm_set_pg_attr(&ctx_pg->ring_mem,
+				      &req->srq_pg_size_srq_lvl,
+				      &req->srq_page_dir);
+	}
+	if (enables & FUNC_BACKING_STORE_CFG_REQ_ENABLES_CQ) {
+		ctxm = &ctx->ctx_arr[BNXT_CTX_CQ];
+		ctx_pg = ctxm->pg_info;
+		req->cq_num_entries = cpu_to_le32(ctx_pg->entries);
+		req->cq_num_l2_entries = cpu_to_le16(ctxm->cq_l2_entries);
+		req->cq_entry_size = cpu_to_le16(ctxm->entry_size);
+		bnxt_hwrm_set_pg_attr(&ctx_pg->ring_mem,
+				      &req->cq_pg_size_cq_lvl,
+				      &req->cq_page_dir);
+	}
+	if (enables & FUNC_BACKING_STORE_CFG_REQ_ENABLES_VNIC) {
+		ctxm = &ctx->ctx_arr[BNXT_CTX_VNIC];
+		ctx_pg = ctxm->pg_info;
+		req->vnic_num_vnic_entries = cpu_to_le16(ctxm->vnic_entries);
+		req->vnic_num_ring_table_entries =
+			cpu_to_le16(ctxm->max_entries - ctxm->vnic_entries);
+		req->vnic_entry_size = cpu_to_le16(ctxm->entry_size);
+		bnxt_hwrm_set_pg_attr(&ctx_pg->ring_mem,
+				      &req->vnic_pg_size_vnic_lvl,
+				      &req->vnic_page_dir);
+	}
+	if (enables & FUNC_BACKING_STORE_CFG_REQ_ENABLES_STAT) {
+		ctxm = &ctx->ctx_arr[BNXT_CTX_STAT];
+		ctx_pg = ctxm->pg_info;
+		req->stat_num_entries = cpu_to_le32(ctxm->max_entries);
+		req->stat_entry_size = cpu_to_le16(ctxm->entry_size);
+		bnxt_hwrm_set_pg_attr(&ctx_pg->ring_mem,
+				      &req->stat_pg_size_stat_lvl,
+				      &req->stat_page_dir);
+	}
+	if (enables & FUNC_BACKING_STORE_CFG_REQ_ENABLES_MRAV) {
+		u32 units;
+
+		ctxm = &ctx->ctx_arr[BNXT_CTX_MRAV];
+		ctx_pg = ctxm->pg_info;
+		req->mrav_num_entries = cpu_to_le32(ctx_pg->entries);
+		units = ctxm->mrav_num_entries_units;
+		if (units) {
+			u32 num_mr, num_ah = ctxm->mrav_av_entries;
+			u32 entries;
+
+			num_mr = ctx_pg->entries - num_ah;
+			entries = ((num_mr / units) << 16) | (num_ah / units);
+			req->mrav_num_entries = cpu_to_le32(entries);
+			flags |= FUNC_BACKING_STORE_CFG_REQ_FLAGS_MRAV_RESERVATION_SPLIT;
+		}
+		req->mrav_entry_size = cpu_to_le16(ctxm->entry_size);
+		bnxt_hwrm_set_pg_attr(&ctx_pg->ring_mem,
+				      &req->mrav_pg_size_mrav_lvl,
+				      &req->mrav_page_dir);
+	}
+	if (enables & FUNC_BACKING_STORE_CFG_REQ_ENABLES_TIM) {
+		ctxm = &ctx->ctx_arr[BNXT_CTX_TIM];
+		ctx_pg = ctxm->pg_info;
+		req->tim_num_entries = cpu_to_le32(ctx_pg->entries);
+		req->tim_entry_size = cpu_to_le16(ctxm->entry_size);
+		bnxt_hwrm_set_pg_attr(&ctx_pg->ring_mem,
+				      &req->tim_pg_size_tim_lvl,
+				      &req->tim_page_dir);
+	}
+	ctxm = &ctx->ctx_arr[BNXT_CTX_STQM];
+	for (i = 0, num_entries = &req->tqm_sp_num_entries,
+	     pg_attr = &req->tqm_sp_pg_size_tqm_sp_lvl,
+	     pg_dir = &req->tqm_sp_page_dir,
+	     ena = FUNC_BACKING_STORE_CFG_REQ_ENABLES_TQM_SP,
+	     ctx_pg = ctxm->pg_info;
+	     i < BNXT_MAX_TQM_RINGS;
+	     ctx_pg = &ctx->ctx_arr[BNXT_CTX_FTQM].pg_info[i],
+	     i++, num_entries++, pg_attr++, pg_dir++, ena <<= 1) {
+		if (!(enables & ena))
+			continue;
+
+		req->tqm_entry_size = cpu_to_le16(ctxm->entry_size);
+		*num_entries = cpu_to_le32(ctx_pg->entries);
+		bnxt_hwrm_set_pg_attr(&ctx_pg->ring_mem, pg_attr, pg_dir);
+	}
+	req->flags = cpu_to_le32(flags);
+	return hwrm_req_send(bp, req);
+}
+
+static int bnxt_alloc_ctx_mem_blk(struct bnxt *bp,
+				  struct bnxt_ctx_pg_info *ctx_pg)
+{
+	struct bnxt_ring_mem_info *rmem = &ctx_pg->ring_mem;
+
+	rmem->page_size = BNXT_PAGE_SIZE;
+	rmem->pg_arr = ctx_pg->ctx_pg_arr;
+	rmem->dma_arr = ctx_pg->ctx_dma_arr;
+	rmem->flags = BNXT_RMEM_VALID_PTE_FLAG;
+	if (rmem->depth >= 1)
+		rmem->flags |= BNXT_RMEM_USE_FULL_PAGE_FLAG;
+	return bnxt_alloc_ring(bp, rmem);
+}
+
+static int bnxt_alloc_ctx_pg_tbls(struct bnxt *bp,
+				  struct bnxt_ctx_pg_info *ctx_pg, u32 mem_size,
+				  u8 depth, struct bnxt_ctx_mem_type *ctxm)
+{
+	struct bnxt_ring_mem_info *rmem = &ctx_pg->ring_mem;
+	int rc;
+
+	if (!mem_size)
+		return -EINVAL;
+
+	ctx_pg->nr_pages = DIV_ROUND_UP(mem_size, BNXT_PAGE_SIZE);
+	if (ctx_pg->nr_pages > MAX_CTX_TOTAL_PAGES) {
+		ctx_pg->nr_pages = 0;
+		return -EINVAL;
+	}
+	if (ctx_pg->nr_pages > MAX_CTX_PAGES || depth > 1) {
+		int nr_tbls, i;
+
+		rmem->depth = 2;
+		ctx_pg->ctx_pg_tbl = kcalloc(MAX_CTX_PAGES, sizeof(ctx_pg),
+					     GFP_KERNEL);
+		if (!ctx_pg->ctx_pg_tbl)
+			return -ENOMEM;
+		nr_tbls = DIV_ROUND_UP(ctx_pg->nr_pages, MAX_CTX_PAGES);
+		rmem->nr_pages = nr_tbls;
+		rc = bnxt_alloc_ctx_mem_blk(bp, ctx_pg);
+		if (rc)
+			return rc;
+		for (i = 0; i < nr_tbls; i++) {
+			struct bnxt_ctx_pg_info *pg_tbl;
+
+			pg_tbl = kzalloc(sizeof(*pg_tbl), GFP_KERNEL);
+			if (!pg_tbl)
+				return -ENOMEM;
+			ctx_pg->ctx_pg_tbl[i] = pg_tbl;
+			rmem = &pg_tbl->ring_mem;
+			rmem->pg_tbl = ctx_pg->ctx_pg_arr[i];
+			rmem->pg_tbl_map = ctx_pg->ctx_dma_arr[i];
+			rmem->depth = 1;
+			rmem->nr_pages = MAX_CTX_PAGES;
+			rmem->ctx_mem = ctxm;
+			if (i == (nr_tbls - 1)) {
+				int rem = ctx_pg->nr_pages % MAX_CTX_PAGES;
+
+				if (rem)
+					rmem->nr_pages = rem;
+			}
+			rc = bnxt_alloc_ctx_mem_blk(bp, pg_tbl);
+			if (rc)
+				break;
+		}
+	} else {
+		rmem->nr_pages = DIV_ROUND_UP(mem_size, BNXT_PAGE_SIZE);
+		if (rmem->nr_pages > 1 || depth)
+			rmem->depth = 1;
+		rmem->ctx_mem = ctxm;
+		rc = bnxt_alloc_ctx_mem_blk(bp, ctx_pg);
+	}
+	return rc;
+}
+
+static size_t bnxt_copy_ctx_pg_tbls(struct bnxt *bp,
+				    struct bnxt_ctx_pg_info *ctx_pg,
+				    void *buf, size_t offset, size_t head,
+				    size_t tail)
+{
+	struct bnxt_ring_mem_info *rmem = &ctx_pg->ring_mem;
+	size_t nr_pages = ctx_pg->nr_pages;
+	int page_size = rmem->page_size;
+	size_t len = 0, total_len = 0;
+	u16 depth = rmem->depth;
+
+	tail %= nr_pages * page_size;
+	do {
+		if (depth > 1) {
+			int i = head / (page_size * MAX_CTX_PAGES);
+			struct bnxt_ctx_pg_info *pg_tbl;
+
+			pg_tbl = ctx_pg->ctx_pg_tbl[i];
+			rmem = &pg_tbl->ring_mem;
+		}
+		len = __bnxt_copy_ring(bp, rmem, buf, offset, head, tail);
+		head += len;
+		offset += len;
+		total_len += len;
+		if (head >= nr_pages * page_size)
+			head = 0;
+	} while (head != tail);
+	return total_len;
+}
+
+static void bnxt_free_ctx_pg_tbls(struct bnxt *bp,
+				  struct bnxt_ctx_pg_info *ctx_pg)
+{
+	struct bnxt_ring_mem_info *rmem = &ctx_pg->ring_mem;
+
+	if (rmem->depth > 1 || ctx_pg->nr_pages > MAX_CTX_PAGES ||
+	    ctx_pg->ctx_pg_tbl) {
+		int i, nr_tbls = rmem->nr_pages;
+
+		for (i = 0; i < nr_tbls; i++) {
+			struct bnxt_ctx_pg_info *pg_tbl;
+			struct bnxt_ring_mem_info *rmem2;
+
+			pg_tbl = ctx_pg->ctx_pg_tbl[i];
+			if (!pg_tbl)
+				continue;
+			rmem2 = &pg_tbl->ring_mem;
+			bnxt_free_ring(bp, rmem2);
+			ctx_pg->ctx_pg_arr[i] = NULL;
+			kfree(pg_tbl);
+			ctx_pg->ctx_pg_tbl[i] = NULL;
+		}
+		kfree(ctx_pg->ctx_pg_tbl);
+		ctx_pg->ctx_pg_tbl = NULL;
+	}
+	bnxt_free_ring(bp, rmem);
+	ctx_pg->nr_pages = 0;
+}
+
+static int bnxt_setup_ctxm_pg_tbls(struct bnxt *bp,
+				   struct bnxt_ctx_mem_type *ctxm, u32 entries,
+				   u8 pg_lvl)
+{
+	struct bnxt_ctx_pg_info *ctx_pg = ctxm->pg_info;
+	int i, rc = 0, n = 1;
+	u32 mem_size;
+
+	if (!ctxm->entry_size || !ctx_pg)
+		return -EINVAL;
+	if (ctxm->instance_bmap)
+		n = hweight32(ctxm->instance_bmap);
+	if (ctxm->entry_multiple)
+		entries = roundup(entries, ctxm->entry_multiple);
+	entries = clamp_t(u32, entries, ctxm->min_entries, ctxm->max_entries);
+	mem_size = entries * ctxm->entry_size;
+	for (i = 0; i < n && !rc; i++) {
+		ctx_pg[i].entries = entries;
+		rc = bnxt_alloc_ctx_pg_tbls(bp, &ctx_pg[i], mem_size, pg_lvl,
+					    ctxm->init_value ? ctxm : NULL);
+	}
+	if (!rc)
+		ctxm->mem_valid = 1;
+	return rc;
+}
+
+static int bnxt_hwrm_func_backing_store_cfg_v2(struct bnxt *bp,
+					       struct bnxt_ctx_mem_type *ctxm,
+					       bool last)
+{
+	struct hwrm_func_backing_store_cfg_v2_input *req;
+	u32 instance_bmap = ctxm->instance_bmap;
+	int i, j, rc = 0, n = 1;
+	__le32 *p;
+
+	if (!(ctxm->flags & BNXT_CTX_MEM_TYPE_VALID) || !ctxm->pg_info)
+		return 0;
+
+	if (instance_bmap)
+		n = hweight32(ctxm->instance_bmap);
+	else
+		instance_bmap = 1;
+
+	rc = hwrm_req_init(bp, req, HWRM_FUNC_BACKING_STORE_CFG_V2);
+	if (rc)
+		return rc;
+	hwrm_req_hold(bp, req);
+	req->type = cpu_to_le16(ctxm->type);
+	req->entry_size = cpu_to_le16(ctxm->entry_size);
+	if ((ctxm->flags & BNXT_CTX_MEM_PERSIST) &&
+	    bnxt_bs_trace_avail(bp, ctxm->type)) {
+		struct bnxt_bs_trace_info *bs_trace;
+		u32 enables;
+
+		enables = FUNC_BACKING_STORE_CFG_V2_REQ_ENABLES_NEXT_BS_OFFSET;
+		req->enables = cpu_to_le32(enables);
+		bs_trace = &bp->bs_trace[bnxt_bstore_to_trace[ctxm->type]];
+		req->next_bs_offset = cpu_to_le32(bs_trace->last_offset);
+	}
+	req->subtype_valid_cnt = ctxm->split_entry_cnt;
+	for (i = 0, p = &req->split_entry_0; i < ctxm->split_entry_cnt; i++)
+		p[i] = cpu_to_le32(ctxm->split[i]);
+	for (i = 0, j = 0; j < n && !rc; i++) {
+		struct bnxt_ctx_pg_info *ctx_pg;
+
+		if (!(instance_bmap & (1 << i)))
+			continue;
+		req->instance = cpu_to_le16(i);
+		ctx_pg = &ctxm->pg_info[j++];
+		if (!ctx_pg->entries)
+			continue;
+		req->num_entries = cpu_to_le32(ctx_pg->entries);
+		bnxt_hwrm_set_pg_attr(&ctx_pg->ring_mem,
+				      &req->page_size_pbl_level,
+				      &req->page_dir);
+		if (last && j == n)
+			req->flags =
+				cpu_to_le32(FUNC_BACKING_STORE_CFG_V2_REQ_FLAGS_BS_CFG_ALL_DONE);
+		rc = hwrm_req_send(bp, req);
+	}
+	hwrm_req_drop(bp, req);
 	return rc;
 }
 
-static int bnxt_hwrm_func_qcaps(struct bnxt *bp)
+static int bnxt_backing_store_cfg_v2(struct bnxt *bp)
 {
+	struct bnxt_ctx_mem_info *ctx = bp->ctx;
+	struct bnxt_ctx_mem_type *ctxm;
+	u16 last_type = BNXT_CTX_INV;
 	int rc = 0;
-	struct hwrm_func_qcaps_input req = {0};
-	struct hwrm_func_qcaps_output *resp = bp->hwrm_cmd_resp_addr;
+	u16 type;
+
+	for (type = BNXT_CTX_SRT; type <= BNXT_CTX_QPC; type++) {
+		ctxm = &ctx->ctx_arr[type];
+		if (!bnxt_bs_trace_avail(bp, type))
+			continue;
+		if (!ctxm->mem_valid) {
+			rc = bnxt_setup_ctxm_pg_tbls(bp, ctxm,
+						     ctxm->max_entries, 1);
+			if (rc) {
+				netdev_warn(bp->dev, "Unable to setup ctx page for type:0x%x.\n",
+					    type);
+				continue;
+			}
+			bnxt_bs_trace_init(bp, ctxm);
+		}
+		last_type = type;
+	}
+
+	if (last_type == BNXT_CTX_INV) {
+		for (type = 0; type < BNXT_CTX_MAX; type++) {
+			ctxm = &ctx->ctx_arr[type];
+			if (ctxm->mem_valid)
+				last_type = type;
+		}
+		if (last_type == BNXT_CTX_INV)
+			return 0;
+	}
+	ctx->ctx_arr[last_type].last = 1;
+
+	for (type = 0 ; type < BNXT_CTX_V2_MAX; type++) {
+		ctxm = &ctx->ctx_arr[type];
+
+		if (!ctxm->mem_valid)
+			continue;
+		rc = bnxt_hwrm_func_backing_store_cfg_v2(bp, ctxm, ctxm->last);
+		if (rc)
+			return rc;
+	}
+	return 0;
+}
+
+/**
+ * __bnxt_copy_ctx_mem - copy host context memory
+ * @bp: The driver context
+ * @ctxm: The pointer to the context memory type
+ * @buf: The destination buffer or NULL to just obtain the length
+ * @offset: The buffer offset to copy the data to
+ * @head: The head offset of context memory to copy from
+ * @tail: The tail offset (last byte + 1) of context memory to end the copy
+ *
+ * This function is called for debugging purposes to dump the host context
+ * used by the chip.
+ *
+ * Return: Length of memory copied
+ */
+static size_t __bnxt_copy_ctx_mem(struct bnxt *bp,
+				  struct bnxt_ctx_mem_type *ctxm, void *buf,
+				  size_t offset, size_t head, size_t tail)
+{
+	struct bnxt_ctx_pg_info *ctx_pg = ctxm->pg_info;
+	size_t len = 0, total_len = 0;
+	int i, n = 1;
+
+	if (!ctx_pg)
+		return 0;
+
+	if (ctxm->instance_bmap)
+		n = hweight32(ctxm->instance_bmap);
+	for (i = 0; i < n; i++) {
+		len = bnxt_copy_ctx_pg_tbls(bp, &ctx_pg[i], buf, offset, head,
+					    tail);
+		offset += len;
+		total_len += len;
+	}
+	return total_len;
+}
+
+size_t bnxt_copy_ctx_mem(struct bnxt *bp, struct bnxt_ctx_mem_type *ctxm,
+			 void *buf, size_t offset)
+{
+	size_t tail = ctxm->max_entries * ctxm->entry_size;
+
+	return __bnxt_copy_ctx_mem(bp, ctxm, buf, offset, 0, tail);
+}
+
+static void bnxt_free_one_ctx_mem(struct bnxt *bp,
+				  struct bnxt_ctx_mem_type *ctxm, bool force)
+{
+	struct bnxt_ctx_pg_info *ctx_pg;
+	int i, n = 1;
+
+	ctxm->last = 0;
+
+	if (ctxm->mem_valid && !force && (ctxm->flags & BNXT_CTX_MEM_PERSIST))
+		return;
+
+	ctx_pg = ctxm->pg_info;
+	if (ctx_pg) {
+		if (ctxm->instance_bmap)
+			n = hweight32(ctxm->instance_bmap);
+		for (i = 0; i < n; i++)
+			bnxt_free_ctx_pg_tbls(bp, &ctx_pg[i]);
+
+		kfree(ctx_pg);
+		ctxm->pg_info = NULL;
+		ctxm->mem_valid = 0;
+	}
+	memset(ctxm, 0, sizeof(*ctxm));
+}
+
+void bnxt_free_ctx_mem(struct bnxt *bp, bool force)
+{
+	struct bnxt_ctx_mem_info *ctx = bp->ctx;
+	u16 type;
+
+	if (!ctx)
+		return;
+
+	for (type = 0; type < BNXT_CTX_V2_MAX; type++)
+		bnxt_free_one_ctx_mem(bp, &ctx->ctx_arr[type], force);
+
+	ctx->flags &= ~BNXT_CTX_FLAG_INITED;
+	if (force) {
+		kfree(ctx);
+		bp->ctx = NULL;
+	}
+}
+
+static int bnxt_alloc_ctx_mem(struct bnxt *bp)
+{
+	struct bnxt_ctx_mem_type *ctxm;
+	struct bnxt_ctx_mem_info *ctx;
+	u32 l2_qps, qp1_qps, max_qps;
+	u32 ena, entries_sp, entries;
+	u32 srqs, max_srqs, min;
+	u32 num_mr, num_ah;
+	u32 extra_srqs = 0;
+	u32 extra_qps = 0;
+	u32 fast_qpmd_qps;
+	u8 pg_lvl = 1;
+	int i, rc;
+
+	rc = bnxt_hwrm_func_backing_store_qcaps(bp);
+	if (rc) {
+		netdev_err(bp->dev, "Failed querying context mem capability, rc = %d.\n",
+			   rc);
+		return rc;
+	}
+	ctx = bp->ctx;
+	if (!ctx || (ctx->flags & BNXT_CTX_FLAG_INITED))
+		return 0;
+
+	ena = 0;
+	if (!(bp->flags & BNXT_FLAG_CHIP_P5_PLUS))
+		goto skip_legacy;
+
+	ctxm = &ctx->ctx_arr[BNXT_CTX_QP];
+	l2_qps = ctxm->qp_l2_entries;
+	qp1_qps = ctxm->qp_qp1_entries;
+	fast_qpmd_qps = ctxm->qp_fast_qpmd_entries;
+	max_qps = ctxm->max_entries;
+	ctxm = &ctx->ctx_arr[BNXT_CTX_SRQ];
+	srqs = ctxm->srq_l2_entries;
+	max_srqs = ctxm->max_entries;
+	if ((bp->flags & BNXT_FLAG_ROCE_CAP) && !is_kdump_kernel()) {
+		pg_lvl = 2;
+		if (BNXT_SW_RES_LMT(bp)) {
+			extra_qps = max_qps - l2_qps - qp1_qps;
+			extra_srqs = max_srqs - srqs;
+		} else {
+			extra_qps = min_t(u32, 65536,
+					  max_qps - l2_qps - qp1_qps);
+			/* allocate extra qps if fw supports RoCE fast qp
+			 * destroy feature
+			 */
+			extra_qps += fast_qpmd_qps;
+			extra_srqs = min_t(u32, 8192, max_srqs - srqs);
+		}
+		if (fast_qpmd_qps)
+			ena |= FUNC_BACKING_STORE_CFG_REQ_ENABLES_QP_FAST_QPMD;
+	}
+
+	ctxm = &ctx->ctx_arr[BNXT_CTX_QP];
+	rc = bnxt_setup_ctxm_pg_tbls(bp, ctxm, l2_qps + qp1_qps + extra_qps,
+				     pg_lvl);
+	if (rc)
+		return rc;
+
+	ctxm = &ctx->ctx_arr[BNXT_CTX_SRQ];
+	rc = bnxt_setup_ctxm_pg_tbls(bp, ctxm, srqs + extra_srqs, pg_lvl);
+	if (rc)
+		return rc;
+
+	ctxm = &ctx->ctx_arr[BNXT_CTX_CQ];
+	rc = bnxt_setup_ctxm_pg_tbls(bp, ctxm, ctxm->cq_l2_entries +
+				     extra_qps * 2, pg_lvl);
+	if (rc)
+		return rc;
+
+	ctxm = &ctx->ctx_arr[BNXT_CTX_VNIC];
+	rc = bnxt_setup_ctxm_pg_tbls(bp, ctxm, ctxm->max_entries, 1);
+	if (rc)
+		return rc;
+
+	ctxm = &ctx->ctx_arr[BNXT_CTX_STAT];
+	rc = bnxt_setup_ctxm_pg_tbls(bp, ctxm, ctxm->max_entries, 1);
+	if (rc)
+		return rc;
+
+	if (!(bp->flags & BNXT_FLAG_ROCE_CAP))
+		goto skip_rdma;
+
+	ctxm = &ctx->ctx_arr[BNXT_CTX_MRAV];
+	if (BNXT_SW_RES_LMT(bp) &&
+	    ctxm->split_entry_cnt == BNXT_CTX_MRAV_AV_SPLIT_ENTRY + 1) {
+		num_ah = ctxm->mrav_av_entries;
+		num_mr = ctxm->max_entries - num_ah;
+	} else {
+		/* 128K extra is needed to accommodate static AH context
+		 * allocation by f/w.
+		 */
+		num_mr = min_t(u32, ctxm->max_entries / 2, 1024 * 256);
+		num_ah = min_t(u32, num_mr, 1024 * 128);
+		ctxm->split_entry_cnt = BNXT_CTX_MRAV_AV_SPLIT_ENTRY + 1;
+		if (!ctxm->mrav_av_entries || ctxm->mrav_av_entries > num_ah)
+			ctxm->mrav_av_entries = num_ah;
+	}
+
+	rc = bnxt_setup_ctxm_pg_tbls(bp, ctxm, num_mr + num_ah, 2);
+	if (rc)
+		return rc;
+	ena |= FUNC_BACKING_STORE_CFG_REQ_ENABLES_MRAV;
+
+	ctxm = &ctx->ctx_arr[BNXT_CTX_TIM];
+	rc = bnxt_setup_ctxm_pg_tbls(bp, ctxm, l2_qps + qp1_qps + extra_qps, 1);
+	if (rc)
+		return rc;
+	ena |= FUNC_BACKING_STORE_CFG_REQ_ENABLES_TIM;
+
+skip_rdma:
+	ctxm = &ctx->ctx_arr[BNXT_CTX_STQM];
+	min = ctxm->min_entries;
+	entries_sp = ctx->ctx_arr[BNXT_CTX_VNIC].vnic_entries + l2_qps +
+		     2 * (extra_qps + qp1_qps) + min;
+	rc = bnxt_setup_ctxm_pg_tbls(bp, ctxm, entries_sp, 2);
+	if (rc)
+		return rc;
+
+	ctxm = &ctx->ctx_arr[BNXT_CTX_FTQM];
+	entries = l2_qps + 2 * (extra_qps + qp1_qps);
+	rc = bnxt_setup_ctxm_pg_tbls(bp, ctxm, entries, 2);
+	if (rc)
+		return rc;
+	for (i = 0; i < ctx->tqm_fp_rings_count + 1; i++)
+		ena |= FUNC_BACKING_STORE_CFG_REQ_ENABLES_TQM_SP << i;
+	ena |= FUNC_BACKING_STORE_CFG_REQ_DFLT_ENABLES;
+
+skip_legacy:
+	if (bp->fw_cap & BNXT_FW_CAP_BACKING_STORE_V2)
+		rc = bnxt_backing_store_cfg_v2(bp);
+	else
+		rc = bnxt_hwrm_func_backing_store_cfg(bp, ena);
+	if (rc) {
+		netdev_err(bp->dev, "Failed configuring context mem, rc = %d.\n",
+			   rc);
+		return rc;
+	}
+	ctx->flags |= BNXT_CTX_FLAG_INITED;
+	return 0;
+}
+
+static int bnxt_hwrm_crash_dump_mem_cfg(struct bnxt *bp)
+{
+	struct hwrm_dbg_crashdump_medium_cfg_input *req;
+	u16 page_attr;
+	int rc;
+
+	if (!(bp->fw_dbg_cap & DBG_QCAPS_RESP_FLAGS_CRASHDUMP_HOST_DDR))
+		return 0;
+
+	rc = hwrm_req_init(bp, req, HWRM_DBG_CRASHDUMP_MEDIUM_CFG);
+	if (rc)
+		return rc;
+
+	if (BNXT_PAGE_SIZE == 0x2000)
+		page_attr = DBG_CRASHDUMP_MEDIUM_CFG_REQ_PG_SIZE_PG_8K;
+	else if (BNXT_PAGE_SIZE == 0x10000)
+		page_attr = DBG_CRASHDUMP_MEDIUM_CFG_REQ_PG_SIZE_PG_64K;
+	else
+		page_attr = DBG_CRASHDUMP_MEDIUM_CFG_REQ_PG_SIZE_PG_4K;
+	req->pg_size_lvl = cpu_to_le16(page_attr |
+				       bp->fw_crash_mem->ring_mem.depth);
+	req->pbl = cpu_to_le64(bp->fw_crash_mem->ring_mem.pg_tbl_map);
+	req->size = cpu_to_le32(bp->fw_crash_len);
+	req->output_dest_flags = cpu_to_le16(BNXT_DBG_CR_DUMP_MDM_CFG_DDR);
+	return hwrm_req_send(bp, req);
+}
+
+static void bnxt_free_crash_dump_mem(struct bnxt *bp)
+{
+	if (bp->fw_crash_mem) {
+		bnxt_free_ctx_pg_tbls(bp, bp->fw_crash_mem);
+		kfree(bp->fw_crash_mem);
+		bp->fw_crash_mem = NULL;
+	}
+}
 
-	bnxt_hwrm_cmd_hdr_init(bp, &req, HWRM_FUNC_QCAPS, -1, -1);
-	req.fid = cpu_to_le16(0xffff);
+static int bnxt_alloc_crash_dump_mem(struct bnxt *bp)
+{
+	u32 mem_size = 0;
+	int rc;
+
+	if (!(bp->fw_dbg_cap & DBG_QCAPS_RESP_FLAGS_CRASHDUMP_HOST_DDR))
+		return 0;
 
-	mutex_lock(&bp->hwrm_cmd_lock);
-	rc = _hwrm_send_message(bp, &req, sizeof(req), HWRM_CMD_TIMEOUT);
+	rc = bnxt_hwrm_get_dump_len(bp, BNXT_DUMP_CRASH, &mem_size);
+	if (rc)
+		return rc;
+
+	mem_size = round_up(mem_size, 4);
+
+	/* keep and use the existing pages */
+	if (bp->fw_crash_mem &&
+	    mem_size <= bp->fw_crash_mem->nr_pages * BNXT_PAGE_SIZE)
+		goto alloc_done;
+
+	if (bp->fw_crash_mem)
+		bnxt_free_ctx_pg_tbls(bp, bp->fw_crash_mem);
+	else
+		bp->fw_crash_mem = kzalloc(sizeof(*bp->fw_crash_mem),
+					   GFP_KERNEL);
+	if (!bp->fw_crash_mem)
+		return -ENOMEM;
+
+	rc = bnxt_alloc_ctx_pg_tbls(bp, bp->fw_crash_mem, mem_size, 1, NULL);
+	if (rc) {
+		bnxt_free_crash_dump_mem(bp);
+		return rc;
+	}
+
+alloc_done:
+	bp->fw_crash_len = mem_size;
+	return 0;
+}
+
+int bnxt_hwrm_func_resc_qcaps(struct bnxt *bp, bool all)
+{
+	struct hwrm_func_resource_qcaps_output *resp;
+	struct hwrm_func_resource_qcaps_input *req;
+	struct bnxt_hw_resc *hw_resc = &bp->hw_resc;
+	int rc;
+
+	rc = hwrm_req_init(bp, req, HWRM_FUNC_RESOURCE_QCAPS);
+	if (rc)
+		return rc;
+
+	req->fid = cpu_to_le16(0xffff);
+	resp = hwrm_req_hold(bp, req);
+	rc = hwrm_req_send_silent(bp, req);
+	if (rc)
+		goto hwrm_func_resc_qcaps_exit;
+
+	hw_resc->max_tx_sch_inputs = le16_to_cpu(resp->max_tx_scheduler_inputs);
+	if (!all)
+		goto hwrm_func_resc_qcaps_exit;
+
+	hw_resc->min_rsscos_ctxs = le16_to_cpu(resp->min_rsscos_ctx);
+	hw_resc->max_rsscos_ctxs = le16_to_cpu(resp->max_rsscos_ctx);
+	hw_resc->min_cp_rings = le16_to_cpu(resp->min_cmpl_rings);
+	hw_resc->max_cp_rings = le16_to_cpu(resp->max_cmpl_rings);
+	hw_resc->min_tx_rings = le16_to_cpu(resp->min_tx_rings);
+	hw_resc->max_tx_rings = le16_to_cpu(resp->max_tx_rings);
+	hw_resc->min_rx_rings = le16_to_cpu(resp->min_rx_rings);
+	hw_resc->max_rx_rings = le16_to_cpu(resp->max_rx_rings);
+	hw_resc->min_hw_ring_grps = le16_to_cpu(resp->min_hw_ring_grps);
+	hw_resc->max_hw_ring_grps = le16_to_cpu(resp->max_hw_ring_grps);
+	hw_resc->min_l2_ctxs = le16_to_cpu(resp->min_l2_ctxs);
+	hw_resc->max_l2_ctxs = le16_to_cpu(resp->max_l2_ctxs);
+	hw_resc->min_vnics = le16_to_cpu(resp->min_vnics);
+	hw_resc->max_vnics = le16_to_cpu(resp->max_vnics);
+	hw_resc->min_stat_ctxs = le16_to_cpu(resp->min_stat_ctx);
+	hw_resc->max_stat_ctxs = le16_to_cpu(resp->max_stat_ctx);
+
+	if (bp->flags & BNXT_FLAG_CHIP_P5_PLUS) {
+		u16 max_msix = le16_to_cpu(resp->max_msix);
+
+		hw_resc->max_nqs = max_msix;
+		hw_resc->max_hw_ring_grps = hw_resc->max_rx_rings;
+	}
+
+	if (BNXT_PF(bp)) {
+		struct bnxt_pf_info *pf = &bp->pf;
+
+		pf->vf_resv_strategy =
+			le16_to_cpu(resp->vf_reservation_strategy);
+		if (pf->vf_resv_strategy > BNXT_VF_RESV_STRATEGY_MINIMAL_STATIC)
+			pf->vf_resv_strategy = BNXT_VF_RESV_STRATEGY_MAXIMAL;
+	}
+hwrm_func_resc_qcaps_exit:
+	hwrm_req_drop(bp, req);
+	return rc;
+}
+
+static int __bnxt_hwrm_ptp_qcfg(struct bnxt *bp)
+{
+	struct hwrm_port_mac_ptp_qcfg_output *resp;
+	struct hwrm_port_mac_ptp_qcfg_input *req;
+	struct bnxt_ptp_cfg *ptp = bp->ptp_cfg;
+	u8 flags;
+	int rc;
+
+	if (bp->hwrm_spec_code < 0x10801 || !BNXT_CHIP_P5_PLUS(bp)) {
+		rc = -ENODEV;
+		goto no_ptp;
+	}
+
+	rc = hwrm_req_init(bp, req, HWRM_PORT_MAC_PTP_QCFG);
+	if (rc)
+		goto no_ptp;
+
+	req->port_id = cpu_to_le16(bp->pf.port_id);
+	resp = hwrm_req_hold(bp, req);
+	rc = hwrm_req_send(bp, req);
+	if (rc)
+		goto exit;
+
+	flags = resp->flags;
+	if (BNXT_CHIP_P5_AND_MINUS(bp) &&
+	    !(flags & PORT_MAC_PTP_QCFG_RESP_FLAGS_HWRM_ACCESS)) {
+		rc = -ENODEV;
+		goto exit;
+	}
+	if (!ptp) {
+		ptp = kzalloc(sizeof(*ptp), GFP_KERNEL);
+		if (!ptp) {
+			rc = -ENOMEM;
+			goto exit;
+		}
+		ptp->bp = bp;
+		bp->ptp_cfg = ptp;
+	}
+
+	if (flags &
+	    (PORT_MAC_PTP_QCFG_RESP_FLAGS_PARTIAL_DIRECT_ACCESS_REF_CLOCK |
+	     PORT_MAC_PTP_QCFG_RESP_FLAGS_64B_PHC_TIME)) {
+		ptp->refclk_regs[0] = le32_to_cpu(resp->ts_ref_clock_reg_lower);
+		ptp->refclk_regs[1] = le32_to_cpu(resp->ts_ref_clock_reg_upper);
+	} else if (BNXT_CHIP_P5(bp)) {
+		ptp->refclk_regs[0] = BNXT_TS_REG_TIMESYNC_TS0_LOWER;
+		ptp->refclk_regs[1] = BNXT_TS_REG_TIMESYNC_TS0_UPPER;
+	} else {
+		rc = -ENODEV;
+		goto exit;
+	}
+	ptp->rtc_configured =
+		(flags & PORT_MAC_PTP_QCFG_RESP_FLAGS_RTC_CONFIGURED) != 0;
+	rc = bnxt_ptp_init(bp);
+	if (rc)
+		netdev_warn(bp->dev, "PTP initialization failed.\n");
+exit:
+	hwrm_req_drop(bp, req);
+	if (!rc)
+		return 0;
+
+no_ptp:
+	bnxt_ptp_clear(bp);
+	kfree(ptp);
+	bp->ptp_cfg = NULL;
+	return rc;
+}
+
+static int __bnxt_hwrm_func_qcaps(struct bnxt *bp)
+{
+	u32 flags, flags_ext, flags_ext2, flags_ext3;
+	struct bnxt_hw_resc *hw_resc = &bp->hw_resc;
+	struct hwrm_func_qcaps_output *resp;
+	struct hwrm_func_qcaps_input *req;
+	int rc;
+
+	rc = hwrm_req_init(bp, req, HWRM_FUNC_QCAPS);
+	if (rc)
+		return rc;
+
+	req->fid = cpu_to_le16(0xffff);
+	resp = hwrm_req_hold(bp, req);
+	rc = hwrm_req_send(bp, req);
 	if (rc)
 		goto hwrm_func_qcaps_exit;
 
-	if (resp->flags & cpu_to_le32(FUNC_QCAPS_RESP_FLAGS_ROCE_V1_SUPPORTED))
+	flags = le32_to_cpu(resp->flags);
+	if (flags & FUNC_QCAPS_RESP_FLAGS_ROCE_V1_SUPPORTED)
 		bp->flags |= BNXT_FLAG_ROCEV1_CAP;
-	if (resp->flags & cpu_to_le32(FUNC_QCAPS_RESP_FLAGS_ROCE_V2_SUPPORTED))
+	if (flags & FUNC_QCAPS_RESP_FLAGS_ROCE_V2_SUPPORTED)
 		bp->flags |= BNXT_FLAG_ROCEV2_CAP;
+	if (flags & FUNC_QCAPS_RESP_FLAGS_PCIE_STATS_SUPPORTED)
+		bp->fw_cap |= BNXT_FW_CAP_PCIE_STATS_SUPPORTED;
+	if (flags & FUNC_QCAPS_RESP_FLAGS_HOT_RESET_CAPABLE)
+		bp->fw_cap |= BNXT_FW_CAP_HOT_RESET;
+	if (flags & FUNC_QCAPS_RESP_FLAGS_EXT_STATS_SUPPORTED)
+		bp->fw_cap |= BNXT_FW_CAP_EXT_STATS_SUPPORTED;
+	if (flags &  FUNC_QCAPS_RESP_FLAGS_ERROR_RECOVERY_CAPABLE)
+		bp->fw_cap |= BNXT_FW_CAP_ERROR_RECOVERY;
+	if (flags & FUNC_QCAPS_RESP_FLAGS_ERR_RECOVER_RELOAD)
+		bp->fw_cap |= BNXT_FW_CAP_ERR_RECOVER_RELOAD;
+	if (!(flags & FUNC_QCAPS_RESP_FLAGS_VLAN_ACCELERATION_TX_DISABLED))
+		bp->fw_cap |= BNXT_FW_CAP_VLAN_TX_INSERT;
+	if (flags & FUNC_QCAPS_RESP_FLAGS_DBG_QCAPS_CMD_SUPPORTED)
+		bp->fw_cap |= BNXT_FW_CAP_DBG_QCAPS;
+
+	flags_ext = le32_to_cpu(resp->flags_ext);
+	if (flags_ext & FUNC_QCAPS_RESP_FLAGS_EXT_EXT_HW_STATS_SUPPORTED)
+		bp->fw_cap |= BNXT_FW_CAP_EXT_HW_STATS_SUPPORTED;
+	if (BNXT_PF(bp) && (flags_ext & FUNC_QCAPS_RESP_FLAGS_EXT_PTP_PPS_SUPPORTED))
+		bp->fw_cap |= BNXT_FW_CAP_PTP_PPS;
+	if (flags_ext & FUNC_QCAPS_RESP_FLAGS_EXT_PTP_64BIT_RTC_SUPPORTED)
+		bp->fw_cap |= BNXT_FW_CAP_PTP_RTC;
+	if (BNXT_PF(bp) && (flags_ext & FUNC_QCAPS_RESP_FLAGS_EXT_HOT_RESET_IF_SUPPORT))
+		bp->fw_cap |= BNXT_FW_CAP_HOT_RESET_IF;
+	if (BNXT_PF(bp) && (flags_ext & FUNC_QCAPS_RESP_FLAGS_EXT_FW_LIVEPATCH_SUPPORTED))
+		bp->fw_cap |= BNXT_FW_CAP_LIVEPATCH;
+	if (flags_ext & FUNC_QCAPS_RESP_FLAGS_EXT_NPAR_1_2_SUPPORTED)
+		bp->fw_cap |= BNXT_FW_CAP_NPAR_1_2;
+	if (BNXT_PF(bp) && (flags_ext & FUNC_QCAPS_RESP_FLAGS_EXT_DFLT_VLAN_TPID_PCP_SUPPORTED))
+		bp->fw_cap |= BNXT_FW_CAP_DFLT_VLAN_TPID_PCP;
+	if (flags_ext & FUNC_QCAPS_RESP_FLAGS_EXT_BS_V2_SUPPORTED)
+		bp->fw_cap |= BNXT_FW_CAP_BACKING_STORE_V2;
+	if (flags_ext & FUNC_QCAPS_RESP_FLAGS_EXT_TX_COAL_CMPL_CAP)
+		bp->flags |= BNXT_FLAG_TX_COAL_CMPL;
+
+	flags_ext2 = le32_to_cpu(resp->flags_ext2);
+	if (flags_ext2 & FUNC_QCAPS_RESP_FLAGS_EXT2_RX_ALL_PKTS_TIMESTAMPS_SUPPORTED)
+		bp->fw_cap |= BNXT_FW_CAP_RX_ALL_PKT_TS;
+	if (flags_ext2 & FUNC_QCAPS_RESP_FLAGS_EXT2_UDP_GSO_SUPPORTED)
+		bp->flags |= BNXT_FLAG_UDP_GSO_CAP;
+	if (flags_ext2 & FUNC_QCAPS_RESP_FLAGS_EXT2_TX_PKT_TS_CMPL_SUPPORTED)
+		bp->fw_cap |= BNXT_FW_CAP_TX_TS_CMP;
+	if (flags_ext2 &
+	    FUNC_QCAPS_RESP_FLAGS_EXT2_SW_MAX_RESOURCE_LIMITS_SUPPORTED)
+		bp->fw_cap |= BNXT_FW_CAP_SW_MAX_RESOURCE_LIMITS;
+	if (BNXT_PF(bp) &&
+	    (flags_ext2 & FUNC_QCAPS_RESP_FLAGS_EXT2_ROCE_VF_RESOURCE_MGMT_SUPPORTED))
+		bp->fw_cap |= BNXT_FW_CAP_ROCE_VF_RESC_MGMT_SUPPORTED;
+
+	flags_ext3 = le32_to_cpu(resp->flags_ext3);
+	if (flags_ext3 & FUNC_QCAPS_RESP_FLAGS_EXT3_ROCE_VF_DYN_ALLOC_SUPPORT)
+		bp->fw_cap |= BNXT_FW_CAP_ROCE_VF_DYN_ALLOC_SUPPORT;
+	if (flags_ext3 & FUNC_QCAPS_RESP_FLAGS_EXT3_MIRROR_ON_ROCE_SUPPORTED)
+		bp->fw_cap |= BNXT_FW_CAP_MIRROR_ON_ROCE;
 
 	bp->tx_push_thresh = 0;
-	if (resp->flags &
-	    cpu_to_le32(FUNC_QCAPS_RESP_FLAGS_PUSH_MODE_SUPPORTED))
+	if ((flags & FUNC_QCAPS_RESP_FLAGS_PUSH_MODE_SUPPORTED) &&
+	    BNXT_FW_MAJ(bp) > 217)
 		bp->tx_push_thresh = BNXT_TX_PUSH_THRESH;
 
+	hw_resc->max_rsscos_ctxs = le16_to_cpu(resp->max_rsscos_ctx);
+	hw_resc->max_cp_rings = le16_to_cpu(resp->max_cmpl_rings);
+	hw_resc->max_tx_rings = le16_to_cpu(resp->max_tx_rings);
+	hw_resc->max_rx_rings = le16_to_cpu(resp->max_rx_rings);
+	hw_resc->max_hw_ring_grps = le32_to_cpu(resp->max_hw_ring_grps);
+	if (!hw_resc->max_hw_ring_grps)
+		hw_resc->max_hw_ring_grps = hw_resc->max_tx_rings;
+	hw_resc->max_l2_ctxs = le16_to_cpu(resp->max_l2_ctxs);
+	hw_resc->max_vnics = le16_to_cpu(resp->max_vnics);
+	hw_resc->max_stat_ctxs = le16_to_cpu(resp->max_stat_ctx);
+
+	hw_resc->max_encap_records = le32_to_cpu(resp->max_encap_records);
+	hw_resc->max_decap_records = le32_to_cpu(resp->max_decap_records);
+	hw_resc->max_tx_em_flows = le32_to_cpu(resp->max_tx_em_flows);
+	hw_resc->max_tx_wm_flows = le32_to_cpu(resp->max_tx_wm_flows);
+	hw_resc->max_rx_em_flows = le32_to_cpu(resp->max_rx_em_flows);
+	hw_resc->max_rx_wm_flows = le32_to_cpu(resp->max_rx_wm_flows);
+
 	if (BNXT_PF(bp)) {
 		struct bnxt_pf_info *pf = &bp->pf;
 
 		pf->fw_fid = le16_to_cpu(resp->fid);
 		pf->port_id = le16_to_cpu(resp->port_id);
-		bp->dev->dev_port = pf->port_id;
 		memcpy(pf->mac_addr, resp->mac_address, ETH_ALEN);
-		memcpy(bp->dev->dev_addr, pf->mac_addr, ETH_ALEN);
-		pf->max_rsscos_ctxs = le16_to_cpu(resp->max_rsscos_ctx);
-		pf->max_cp_rings = le16_to_cpu(resp->max_cmpl_rings);
-		pf->max_tx_rings = le16_to_cpu(resp->max_tx_rings);
-		pf->max_rx_rings = le16_to_cpu(resp->max_rx_rings);
-		pf->max_hw_ring_grps = le32_to_cpu(resp->max_hw_ring_grps);
-		if (!pf->max_hw_ring_grps)
-			pf->max_hw_ring_grps = pf->max_tx_rings;
-		pf->max_l2_ctxs = le16_to_cpu(resp->max_l2_ctxs);
-		pf->max_vnics = le16_to_cpu(resp->max_vnics);
-		pf->max_stat_ctxs = le16_to_cpu(resp->max_stat_ctx);
 		pf->first_vf_id = le16_to_cpu(resp->first_vf_id);
 		pf->max_vfs = le16_to_cpu(resp->max_vfs);
-		pf->max_encap_records = le32_to_cpu(resp->max_encap_records);
-		pf->max_decap_records = le32_to_cpu(resp->max_decap_records);
-		pf->max_tx_em_flows = le32_to_cpu(resp->max_tx_em_flows);
-		pf->max_tx_wm_flows = le32_to_cpu(resp->max_tx_wm_flows);
-		pf->max_rx_em_flows = le32_to_cpu(resp->max_rx_em_flows);
-		pf->max_rx_wm_flows = le32_to_cpu(resp->max_rx_wm_flows);
-		if (resp->flags &
-		    cpu_to_le32(FUNC_QCAPS_RESP_FLAGS_WOL_MAGICPKT_SUPPORTED))
+		bp->flags &= ~BNXT_FLAG_WOL_CAP;
+		if (flags & FUNC_QCAPS_RESP_FLAGS_WOL_MAGICPKT_SUPPORTED)
 			bp->flags |= BNXT_FLAG_WOL_CAP;
+		if (flags & FUNC_QCAPS_RESP_FLAGS_PTP_SUPPORTED) {
+			bp->fw_cap |= BNXT_FW_CAP_PTP;
+		} else {
+			bnxt_ptp_clear(bp);
+			kfree(bp->ptp_cfg);
+			bp->ptp_cfg = NULL;
+		}
 	} else {
 #ifdef CONFIG_BNXT_SRIOV
 		struct bnxt_vf_info *vf = &bp->vf;
 
 		vf->fw_fid = le16_to_cpu(resp->fid);
+		memcpy(vf->mac_addr, resp->mac_address, ETH_ALEN);
+#endif
+	}
+	bp->tso_max_segs = le16_to_cpu(resp->max_tso_segs);
 
-		vf->max_rsscos_ctxs = le16_to_cpu(resp->max_rsscos_ctx);
-		vf->max_cp_rings = le16_to_cpu(resp->max_cmpl_rings);
-		vf->max_tx_rings = le16_to_cpu(resp->max_tx_rings);
-		vf->max_rx_rings = le16_to_cpu(resp->max_rx_rings);
-		vf->max_hw_ring_grps = le32_to_cpu(resp->max_hw_ring_grps);
-		if (!vf->max_hw_ring_grps)
-			vf->max_hw_ring_grps = vf->max_tx_rings;
-		vf->max_l2_ctxs = le16_to_cpu(resp->max_l2_ctxs);
-		vf->max_vnics = le16_to_cpu(resp->max_vnics);
-		vf->max_stat_ctxs = le16_to_cpu(resp->max_stat_ctx);
+hwrm_func_qcaps_exit:
+	hwrm_req_drop(bp, req);
+	return rc;
+}
 
-		memcpy(vf->mac_addr, resp->mac_address, ETH_ALEN);
-		mutex_unlock(&bp->hwrm_cmd_lock);
+static void bnxt_hwrm_dbg_qcaps(struct bnxt *bp)
+{
+	struct hwrm_dbg_qcaps_output *resp;
+	struct hwrm_dbg_qcaps_input *req;
+	int rc;
 
-		if (is_valid_ether_addr(vf->mac_addr)) {
-			/* overwrite netdev dev_adr with admin VF MAC */
-			memcpy(bp->dev->dev_addr, vf->mac_addr, ETH_ALEN);
-		} else {
-			eth_hw_addr_random(bp->dev);
-			rc = bnxt_approve_mac(bp, bp->dev->dev_addr);
-		}
+	bp->fw_dbg_cap = 0;
+	if (!(bp->fw_cap & BNXT_FW_CAP_DBG_QCAPS))
+		return;
+
+	rc = hwrm_req_init(bp, req, HWRM_DBG_QCAPS);
+	if (rc)
+		return;
+
+	req->fid = cpu_to_le16(0xffff);
+	resp = hwrm_req_hold(bp, req);
+	rc = hwrm_req_send(bp, req);
+	if (rc)
+		goto hwrm_dbg_qcaps_exit;
+
+	bp->fw_dbg_cap = le32_to_cpu(resp->flags);
+
+hwrm_dbg_qcaps_exit:
+	hwrm_req_drop(bp, req);
+}
+
+static int bnxt_hwrm_queue_qportcfg(struct bnxt *bp);
+
+int bnxt_hwrm_func_qcaps(struct bnxt *bp)
+{
+	int rc;
+
+	rc = __bnxt_hwrm_func_qcaps(bp);
+	if (rc)
+		return rc;
+
+	bnxt_hwrm_dbg_qcaps(bp);
+
+	rc = bnxt_hwrm_queue_qportcfg(bp);
+	if (rc) {
+		netdev_err(bp->dev, "hwrm query qportcfg failure rc: %d\n", rc);
 		return rc;
-#endif
 	}
+	if (bp->hwrm_spec_code >= 0x10803) {
+		rc = bnxt_alloc_ctx_mem(bp);
+		if (rc)
+			return rc;
+		rc = bnxt_hwrm_func_resc_qcaps(bp, true);
+		if (!rc)
+			bp->fw_cap |= BNXT_FW_CAP_NEW_RM;
+	}
+	return 0;
+}
 
-hwrm_func_qcaps_exit:
-	mutex_unlock(&bp->hwrm_cmd_lock);
+static int bnxt_hwrm_cfa_adv_flow_mgnt_qcaps(struct bnxt *bp)
+{
+	struct hwrm_cfa_adv_flow_mgnt_qcaps_output *resp;
+	struct hwrm_cfa_adv_flow_mgnt_qcaps_input *req;
+	u32 flags;
+	int rc;
+
+	if (!(bp->fw_cap & BNXT_FW_CAP_CFA_ADV_FLOW))
+		return 0;
+
+	rc = hwrm_req_init(bp, req, HWRM_CFA_ADV_FLOW_MGNT_QCAPS);
+	if (rc)
+		return rc;
+
+	resp = hwrm_req_hold(bp, req);
+	rc = hwrm_req_send(bp, req);
+	if (rc)
+		goto hwrm_cfa_adv_qcaps_exit;
+
+	flags = le32_to_cpu(resp->flags);
+	if (flags &
+	    CFA_ADV_FLOW_MGNT_QCAPS_RESP_FLAGS_RFS_RING_TBL_IDX_V2_SUPPORTED)
+		bp->fw_cap |= BNXT_FW_CAP_CFA_RFS_RING_TBL_IDX_V2;
+
+	if (flags &
+	    CFA_ADV_FLOW_MGNT_QCAPS_RESP_FLAGS_RFS_RING_TBL_IDX_V3_SUPPORTED)
+		bp->fw_cap |= BNXT_FW_CAP_CFA_RFS_RING_TBL_IDX_V3;
+
+	if (flags &
+	    CFA_ADV_FLOW_MGNT_QCAPS_RESP_FLAGS_NTUPLE_FLOW_RX_EXT_IP_PROTO_SUPPORTED)
+		bp->fw_cap |= BNXT_FW_CAP_CFA_NTUPLE_RX_EXT_IP_PROTO;
+
+hwrm_cfa_adv_qcaps_exit:
+	hwrm_req_drop(bp, req);
+	return rc;
+}
+
+static int __bnxt_alloc_fw_health(struct bnxt *bp)
+{
+	if (bp->fw_health)
+		return 0;
+
+	bp->fw_health = kzalloc(sizeof(*bp->fw_health), GFP_KERNEL);
+	if (!bp->fw_health)
+		return -ENOMEM;
+
+	mutex_init(&bp->fw_health->lock);
+	return 0;
+}
+
+static int bnxt_alloc_fw_health(struct bnxt *bp)
+{
+	int rc;
+
+	if (!(bp->fw_cap & BNXT_FW_CAP_HOT_RESET) &&
+	    !(bp->fw_cap & BNXT_FW_CAP_ERROR_RECOVERY))
+		return 0;
+
+	rc = __bnxt_alloc_fw_health(bp);
+	if (rc) {
+		bp->fw_cap &= ~BNXT_FW_CAP_HOT_RESET;
+		bp->fw_cap &= ~BNXT_FW_CAP_ERROR_RECOVERY;
+		return rc;
+	}
+
+	return 0;
+}
+
+static void __bnxt_map_fw_health_reg(struct bnxt *bp, u32 reg)
+{
+	writel(reg & BNXT_GRC_BASE_MASK, bp->bar0 +
+					 BNXT_GRCPF_REG_WINDOW_BASE_OUT +
+					 BNXT_FW_HEALTH_WIN_MAP_OFF);
+}
+
+static void bnxt_inv_fw_health_reg(struct bnxt *bp)
+{
+	struct bnxt_fw_health *fw_health = bp->fw_health;
+	u32 reg_type;
+
+	if (!fw_health)
+		return;
+
+	reg_type = BNXT_FW_HEALTH_REG_TYPE(fw_health->regs[BNXT_FW_HEALTH_REG]);
+	if (reg_type == BNXT_FW_HEALTH_REG_TYPE_GRC)
+		fw_health->status_reliable = false;
+
+	reg_type = BNXT_FW_HEALTH_REG_TYPE(fw_health->regs[BNXT_FW_RESET_CNT_REG]);
+	if (reg_type == BNXT_FW_HEALTH_REG_TYPE_GRC)
+		fw_health->resets_reliable = false;
+}
+
+static void bnxt_try_map_fw_health_reg(struct bnxt *bp)
+{
+	void __iomem *hs;
+	u32 status_loc;
+	u32 reg_type;
+	u32 sig;
+
+	if (bp->fw_health)
+		bp->fw_health->status_reliable = false;
+
+	__bnxt_map_fw_health_reg(bp, HCOMM_STATUS_STRUCT_LOC);
+	hs = bp->bar0 + BNXT_FW_HEALTH_WIN_OFF(HCOMM_STATUS_STRUCT_LOC);
+
+	sig = readl(hs + offsetof(struct hcomm_status, sig_ver));
+	if ((sig & HCOMM_STATUS_SIGNATURE_MASK) != HCOMM_STATUS_SIGNATURE_VAL) {
+		if (!bp->chip_num) {
+			__bnxt_map_fw_health_reg(bp, BNXT_GRC_REG_BASE);
+			bp->chip_num = readl(bp->bar0 +
+					     BNXT_FW_HEALTH_WIN_BASE +
+					     BNXT_GRC_REG_CHIP_NUM);
+		}
+		if (!BNXT_CHIP_P5_PLUS(bp))
+			return;
+
+		status_loc = BNXT_GRC_REG_STATUS_P5 |
+			     BNXT_FW_HEALTH_REG_TYPE_BAR0;
+	} else {
+		status_loc = readl(hs + offsetof(struct hcomm_status,
+						 fw_status_loc));
+	}
+
+	if (__bnxt_alloc_fw_health(bp)) {
+		netdev_warn(bp->dev, "no memory for firmware status checks\n");
+		return;
+	}
+
+	bp->fw_health->regs[BNXT_FW_HEALTH_REG] = status_loc;
+	reg_type = BNXT_FW_HEALTH_REG_TYPE(status_loc);
+	if (reg_type == BNXT_FW_HEALTH_REG_TYPE_GRC) {
+		__bnxt_map_fw_health_reg(bp, status_loc);
+		bp->fw_health->mapped_regs[BNXT_FW_HEALTH_REG] =
+			BNXT_FW_HEALTH_WIN_OFF(status_loc);
+	}
+
+	bp->fw_health->status_reliable = true;
+}
+
+static int bnxt_map_fw_health_regs(struct bnxt *bp)
+{
+	struct bnxt_fw_health *fw_health = bp->fw_health;
+	u32 reg_base = 0xffffffff;
+	int i;
+
+	bp->fw_health->status_reliable = false;
+	bp->fw_health->resets_reliable = false;
+	/* Only pre-map the monitoring GRC registers using window 3 */
+	for (i = 0; i < 4; i++) {
+		u32 reg = fw_health->regs[i];
+
+		if (BNXT_FW_HEALTH_REG_TYPE(reg) != BNXT_FW_HEALTH_REG_TYPE_GRC)
+			continue;
+		if (reg_base == 0xffffffff)
+			reg_base = reg & BNXT_GRC_BASE_MASK;
+		if ((reg & BNXT_GRC_BASE_MASK) != reg_base)
+			return -ERANGE;
+		fw_health->mapped_regs[i] = BNXT_FW_HEALTH_WIN_OFF(reg);
+	}
+	bp->fw_health->status_reliable = true;
+	bp->fw_health->resets_reliable = true;
+	if (reg_base == 0xffffffff)
+		return 0;
+
+	__bnxt_map_fw_health_reg(bp, reg_base);
+	return 0;
+}
+
+static void bnxt_remap_fw_health_regs(struct bnxt *bp)
+{
+	if (!bp->fw_health)
+		return;
+
+	if (bp->fw_cap & BNXT_FW_CAP_ERROR_RECOVERY) {
+		bp->fw_health->status_reliable = true;
+		bp->fw_health->resets_reliable = true;
+	} else {
+		bnxt_try_map_fw_health_reg(bp);
+	}
+}
+
+static int bnxt_hwrm_error_recovery_qcfg(struct bnxt *bp)
+{
+	struct bnxt_fw_health *fw_health = bp->fw_health;
+	struct hwrm_error_recovery_qcfg_output *resp;
+	struct hwrm_error_recovery_qcfg_input *req;
+	int rc, i;
+
+	if (!(bp->fw_cap & BNXT_FW_CAP_ERROR_RECOVERY))
+		return 0;
+
+	rc = hwrm_req_init(bp, req, HWRM_ERROR_RECOVERY_QCFG);
+	if (rc)
+		return rc;
+
+	resp = hwrm_req_hold(bp, req);
+	rc = hwrm_req_send(bp, req);
+	if (rc)
+		goto err_recovery_out;
+	fw_health->flags = le32_to_cpu(resp->flags);
+	if ((fw_health->flags & ERROR_RECOVERY_QCFG_RESP_FLAGS_CO_CPU) &&
+	    !(bp->fw_cap & BNXT_FW_CAP_KONG_MB_CHNL)) {
+		rc = -EINVAL;
+		goto err_recovery_out;
+	}
+	fw_health->polling_dsecs = le32_to_cpu(resp->driver_polling_freq);
+	fw_health->master_func_wait_dsecs =
+		le32_to_cpu(resp->master_func_wait_period);
+	fw_health->normal_func_wait_dsecs =
+		le32_to_cpu(resp->normal_func_wait_period);
+	fw_health->post_reset_wait_dsecs =
+		le32_to_cpu(resp->master_func_wait_period_after_reset);
+	fw_health->post_reset_max_wait_dsecs =
+		le32_to_cpu(resp->max_bailout_time_after_reset);
+	fw_health->regs[BNXT_FW_HEALTH_REG] =
+		le32_to_cpu(resp->fw_health_status_reg);
+	fw_health->regs[BNXT_FW_HEARTBEAT_REG] =
+		le32_to_cpu(resp->fw_heartbeat_reg);
+	fw_health->regs[BNXT_FW_RESET_CNT_REG] =
+		le32_to_cpu(resp->fw_reset_cnt_reg);
+	fw_health->regs[BNXT_FW_RESET_INPROG_REG] =
+		le32_to_cpu(resp->reset_inprogress_reg);
+	fw_health->fw_reset_inprog_reg_mask =
+		le32_to_cpu(resp->reset_inprogress_reg_mask);
+	fw_health->fw_reset_seq_cnt = resp->reg_array_cnt;
+	if (fw_health->fw_reset_seq_cnt >= 16) {
+		rc = -EINVAL;
+		goto err_recovery_out;
+	}
+	for (i = 0; i < fw_health->fw_reset_seq_cnt; i++) {
+		fw_health->fw_reset_seq_regs[i] =
+			le32_to_cpu(resp->reset_reg[i]);
+		fw_health->fw_reset_seq_vals[i] =
+			le32_to_cpu(resp->reset_reg_val[i]);
+		fw_health->fw_reset_seq_delay_msec[i] =
+			resp->delay_after_reset[i];
+	}
+err_recovery_out:
+	hwrm_req_drop(bp, req);
+	if (!rc)
+		rc = bnxt_map_fw_health_regs(bp);
+	if (rc)
+		bp->fw_cap &= ~BNXT_FW_CAP_ERROR_RECOVERY;
 	return rc;
 }
 
 static int bnxt_hwrm_func_reset(struct bnxt *bp)
 {
-	struct hwrm_func_reset_input req = {0};
+	struct hwrm_func_reset_input *req;
+	int rc;
 
-	bnxt_hwrm_cmd_hdr_init(bp, &req, HWRM_FUNC_RESET, -1, -1);
-	req.enables = 0;
+	rc = hwrm_req_init(bp, req, HWRM_FUNC_RESET);
+	if (rc)
+		return rc;
+
+	req->enables = 0;
+	hwrm_req_timeout(bp, req, HWRM_RESET_TIMEOUT);
+	return hwrm_req_send(bp, req);
+}
+
+static void bnxt_nvm_cfg_ver_get(struct bnxt *bp)
+{
+	struct hwrm_nvm_get_dev_info_output nvm_info;
 
-	return hwrm_send_message(bp, &req, sizeof(req), HWRM_RESET_TIMEOUT);
+	if (!bnxt_hwrm_nvm_get_dev_info(bp, &nvm_info))
+		snprintf(bp->nvm_cfg_ver, FW_VER_STR_LEN, "%d.%d.%d",
+			 nvm_info.nvm_cfg_ver_maj, nvm_info.nvm_cfg_ver_min,
+			 nvm_info.nvm_cfg_ver_upd);
 }
 
 static int bnxt_hwrm_queue_qportcfg(struct bnxt *bp)
 {
+	struct hwrm_queue_qportcfg_output *resp;
+	struct hwrm_queue_qportcfg_input *req;
+	u8 i, j, *qptr;
+	bool no_rdma;
 	int rc = 0;
-	struct hwrm_queue_qportcfg_input req = {0};
-	struct hwrm_queue_qportcfg_output *resp = bp->hwrm_cmd_resp_addr;
-	u8 i, *qptr;
 
-	bnxt_hwrm_cmd_hdr_init(bp, &req, HWRM_QUEUE_QPORTCFG, -1, -1);
+	rc = hwrm_req_init(bp, req, HWRM_QUEUE_QPORTCFG);
+	if (rc)
+		return rc;
 
-	mutex_lock(&bp->hwrm_cmd_lock);
-	rc = _hwrm_send_message(bp, &req, sizeof(req), HWRM_CMD_TIMEOUT);
+	resp = hwrm_req_hold(bp, req);
+	rc = hwrm_req_send(bp, req);
 	if (rc)
 		goto qportcfg_exit;
 
@@ -4689,62 +10114,146 @@ static int bnxt_hwrm_queue_qportcfg(struct bnxt *bp)
 	if (bp->max_tc > BNXT_MAX_QUEUE)
 		bp->max_tc = BNXT_MAX_QUEUE;
 
+	no_rdma = !(bp->flags & BNXT_FLAG_ROCE_CAP);
+	qptr = &resp->queue_id0;
+	for (i = 0, j = 0; i < bp->max_tc; i++) {
+		bp->q_info[j].queue_id = *qptr;
+		bp->q_ids[i] = *qptr++;
+		bp->q_info[j].queue_profile = *qptr++;
+		bp->tc_to_qidx[j] = j;
+		if (!BNXT_CNPQ(bp->q_info[j].queue_profile) ||
+		    (no_rdma && BNXT_PF(bp)))
+			j++;
+	}
+	bp->max_q = bp->max_tc;
+	bp->max_tc = max_t(u8, j, 1);
+
 	if (resp->queue_cfg_info & QUEUE_QPORTCFG_RESP_QUEUE_CFG_INFO_ASYM_CFG)
 		bp->max_tc = 1;
 
 	if (bp->max_lltc > bp->max_tc)
 		bp->max_lltc = bp->max_tc;
 
-	qptr = &resp->queue_id0;
-	for (i = 0; i < bp->max_tc; i++) {
-		bp->q_info[i].queue_id = *qptr++;
-		bp->q_info[i].queue_profile = *qptr++;
-	}
-
 qportcfg_exit:
-	mutex_unlock(&bp->hwrm_cmd_lock);
+	hwrm_req_drop(bp, req);
 	return rc;
 }
 
-static int bnxt_hwrm_ver_get(struct bnxt *bp)
+static int bnxt_hwrm_poll(struct bnxt *bp)
 {
+	struct hwrm_ver_get_input *req;
 	int rc;
-	struct hwrm_ver_get_input req = {0};
-	struct hwrm_ver_get_output *resp = bp->hwrm_cmd_resp_addr;
-	u32 dev_caps_cfg;
 
+	rc = hwrm_req_init(bp, req, HWRM_VER_GET);
+	if (rc)
+		return rc;
+
+	req->hwrm_intf_maj = HWRM_VERSION_MAJOR;
+	req->hwrm_intf_min = HWRM_VERSION_MINOR;
+	req->hwrm_intf_upd = HWRM_VERSION_UPDATE;
+
+	hwrm_req_flags(bp, req, BNXT_HWRM_CTX_SILENT | BNXT_HWRM_FULL_WAIT);
+	rc = hwrm_req_send(bp, req);
+	return rc;
+}
+
+static int bnxt_hwrm_ver_get(struct bnxt *bp)
+{
+	struct hwrm_ver_get_output *resp;
+	struct hwrm_ver_get_input *req;
+	u16 fw_maj, fw_min, fw_bld, fw_rsv;
+	u32 dev_caps_cfg, hwrm_ver;
+	int rc, len, max_tmo_secs;
+
+	rc = hwrm_req_init(bp, req, HWRM_VER_GET);
+	if (rc)
+		return rc;
+
+	hwrm_req_flags(bp, req, BNXT_HWRM_FULL_WAIT);
 	bp->hwrm_max_req_len = HWRM_MAX_REQ_LEN;
-	bnxt_hwrm_cmd_hdr_init(bp, &req, HWRM_VER_GET, -1, -1);
-	req.hwrm_intf_maj = HWRM_VERSION_MAJOR;
-	req.hwrm_intf_min = HWRM_VERSION_MINOR;
-	req.hwrm_intf_upd = HWRM_VERSION_UPDATE;
-	mutex_lock(&bp->hwrm_cmd_lock);
-	rc = _hwrm_send_message(bp, &req, sizeof(req), HWRM_CMD_TIMEOUT);
+	req->hwrm_intf_maj = HWRM_VERSION_MAJOR;
+	req->hwrm_intf_min = HWRM_VERSION_MINOR;
+	req->hwrm_intf_upd = HWRM_VERSION_UPDATE;
+
+	resp = hwrm_req_hold(bp, req);
+	rc = hwrm_req_send(bp, req);
 	if (rc)
 		goto hwrm_ver_get_exit;
 
 	memcpy(&bp->ver_resp, resp, sizeof(struct hwrm_ver_get_output));
 
-	bp->hwrm_spec_code = resp->hwrm_intf_maj << 16 |
-			     resp->hwrm_intf_min << 8 | resp->hwrm_intf_upd;
-	if (resp->hwrm_intf_maj < 1) {
+	bp->hwrm_spec_code = resp->hwrm_intf_maj_8b << 16 |
+			     resp->hwrm_intf_min_8b << 8 |
+			     resp->hwrm_intf_upd_8b;
+	if (resp->hwrm_intf_maj_8b < 1) {
 		netdev_warn(bp->dev, "HWRM interface %d.%d.%d is older than 1.0.0.\n",
-			    resp->hwrm_intf_maj, resp->hwrm_intf_min,
-			    resp->hwrm_intf_upd);
+			    resp->hwrm_intf_maj_8b, resp->hwrm_intf_min_8b,
+			    resp->hwrm_intf_upd_8b);
 		netdev_warn(bp->dev, "Please update firmware with HWRM interface 1.0.0 or newer.\n");
 	}
-	snprintf(bp->fw_ver_str, BC_HWRM_STR_LEN, "%d.%d.%d/%d.%d.%d",
-		 resp->hwrm_fw_maj, resp->hwrm_fw_min, resp->hwrm_fw_bld,
-		 resp->hwrm_intf_maj, resp->hwrm_intf_min, resp->hwrm_intf_upd);
+
+	hwrm_ver = HWRM_VERSION_MAJOR << 16 | HWRM_VERSION_MINOR << 8 |
+			HWRM_VERSION_UPDATE;
+
+	if (bp->hwrm_spec_code > hwrm_ver)
+		snprintf(bp->hwrm_ver_supp, FW_VER_STR_LEN, "%d.%d.%d",
+			 HWRM_VERSION_MAJOR, HWRM_VERSION_MINOR,
+			 HWRM_VERSION_UPDATE);
+	else
+		snprintf(bp->hwrm_ver_supp, FW_VER_STR_LEN, "%d.%d.%d",
+			 resp->hwrm_intf_maj_8b, resp->hwrm_intf_min_8b,
+			 resp->hwrm_intf_upd_8b);
+
+	fw_maj = le16_to_cpu(resp->hwrm_fw_major);
+	if (bp->hwrm_spec_code > 0x10803 && fw_maj) {
+		fw_min = le16_to_cpu(resp->hwrm_fw_minor);
+		fw_bld = le16_to_cpu(resp->hwrm_fw_build);
+		fw_rsv = le16_to_cpu(resp->hwrm_fw_patch);
+		len = FW_VER_STR_LEN;
+	} else {
+		fw_maj = resp->hwrm_fw_maj_8b;
+		fw_min = resp->hwrm_fw_min_8b;
+		fw_bld = resp->hwrm_fw_bld_8b;
+		fw_rsv = resp->hwrm_fw_rsvd_8b;
+		len = BC_HWRM_STR_LEN;
+	}
+	bp->fw_ver_code = BNXT_FW_VER_CODE(fw_maj, fw_min, fw_bld, fw_rsv);
+	snprintf(bp->fw_ver_str, len, "%d.%d.%d.%d", fw_maj, fw_min, fw_bld,
+		 fw_rsv);
+
+	if (strlen(resp->active_pkg_name)) {
+		int fw_ver_len = strlen(bp->fw_ver_str);
+
+		snprintf(bp->fw_ver_str + fw_ver_len,
+			 FW_VER_STR_LEN - fw_ver_len - 1, "/pkg %s",
+			 resp->active_pkg_name);
+		bp->fw_cap |= BNXT_FW_CAP_PKG_VER;
+	}
 
 	bp->hwrm_cmd_timeout = le16_to_cpu(resp->def_req_timeout);
 	if (!bp->hwrm_cmd_timeout)
 		bp->hwrm_cmd_timeout = DFLT_HWRM_CMD_TIMEOUT;
+	bp->hwrm_cmd_max_timeout = le16_to_cpu(resp->max_req_timeout) * 1000;
+	if (!bp->hwrm_cmd_max_timeout)
+		bp->hwrm_cmd_max_timeout = HWRM_CMD_MAX_TIMEOUT;
+	max_tmo_secs = bp->hwrm_cmd_max_timeout / 1000;
+#ifdef CONFIG_DETECT_HUNG_TASK
+	if (bp->hwrm_cmd_max_timeout > HWRM_CMD_MAX_TIMEOUT ||
+	    max_tmo_secs > CONFIG_DEFAULT_HUNG_TASK_TIMEOUT) {
+		netdev_warn(bp->dev, "Device requests max timeout of %d seconds, may trigger hung task watchdog (kernel default %ds)\n",
+			    max_tmo_secs, CONFIG_DEFAULT_HUNG_TASK_TIMEOUT);
+	}
+#endif
 
-	if (resp->hwrm_intf_maj >= 1)
+	if (resp->hwrm_intf_maj_8b >= 1) {
 		bp->hwrm_max_req_len = le16_to_cpu(resp->max_req_win_len);
+		bp->hwrm_max_ext_req_len = le16_to_cpu(resp->max_ext_req_len);
+	}
+	if (bp->hwrm_max_ext_req_len < HWRM_MAX_REQ_LEN)
+		bp->hwrm_max_ext_req_len = HWRM_MAX_REQ_LEN;
 
 	bp->chip_num = le16_to_cpu(resp->chip_num);
+	bp->chip_rev = resp->chip_rev;
 	if (bp->chip_num == CHIP_NUM_58700 && !resp->chip_rev &&
 	    !resp->chip_metal)
 		bp->flags |= BNXT_FLAG_CHIP_NITRO_A0;
@@ -4752,67 +10261,256 @@ static int bnxt_hwrm_ver_get(struct bnxt *bp)
 	dev_caps_cfg = le32_to_cpu(resp->dev_caps_cfg);
 	if ((dev_caps_cfg & VER_GET_RESP_DEV_CAPS_CFG_SHORT_CMD_SUPPORTED) &&
 	    (dev_caps_cfg & VER_GET_RESP_DEV_CAPS_CFG_SHORT_CMD_REQUIRED))
-		bp->flags |= BNXT_FLAG_SHORT_CMD;
+		bp->fw_cap |= BNXT_FW_CAP_SHORT_CMD;
+
+	if (dev_caps_cfg & VER_GET_RESP_DEV_CAPS_CFG_KONG_MB_CHNL_SUPPORTED)
+		bp->fw_cap |= BNXT_FW_CAP_KONG_MB_CHNL;
+
+	if (dev_caps_cfg &
+	    VER_GET_RESP_DEV_CAPS_CFG_FLOW_HANDLE_64BIT_SUPPORTED)
+		bp->fw_cap |= BNXT_FW_CAP_OVS_64BIT_HANDLE;
+
+	if (dev_caps_cfg &
+	    VER_GET_RESP_DEV_CAPS_CFG_TRUSTED_VF_SUPPORTED)
+		bp->fw_cap |= BNXT_FW_CAP_TRUSTED_VF;
+
+	if (dev_caps_cfg &
+	    VER_GET_RESP_DEV_CAPS_CFG_CFA_ADV_FLOW_MGNT_SUPPORTED)
+		bp->fw_cap |= BNXT_FW_CAP_CFA_ADV_FLOW;
 
 hwrm_ver_get_exit:
-	mutex_unlock(&bp->hwrm_cmd_lock);
+	hwrm_req_drop(bp, req);
 	return rc;
 }
 
 int bnxt_hwrm_fw_set_time(struct bnxt *bp)
 {
-#if IS_ENABLED(CONFIG_RTC_LIB)
-	struct hwrm_fw_set_time_input req = {0};
-	struct rtc_time tm;
-	struct timeval tv;
+	struct hwrm_fw_set_time_input *req;
+	struct tm tm;
+	time64_t now = ktime_get_real_seconds();
+	int rc;
 
-	if (bp->hwrm_spec_code < 0x10400)
+	if ((BNXT_VF(bp) && bp->hwrm_spec_code < 0x10901) ||
+	    bp->hwrm_spec_code < 0x10400)
 		return -EOPNOTSUPP;
 
-	do_gettimeofday(&tv);
-	rtc_time_to_tm(tv.tv_sec, &tm);
-	bnxt_hwrm_cmd_hdr_init(bp, &req, HWRM_FW_SET_TIME, -1, -1);
-	req.year = cpu_to_le16(1900 + tm.tm_year);
-	req.month = 1 + tm.tm_mon;
-	req.day = tm.tm_mday;
-	req.hour = tm.tm_hour;
-	req.minute = tm.tm_min;
-	req.second = tm.tm_sec;
-	return hwrm_send_message(bp, &req, sizeof(req), HWRM_CMD_TIMEOUT);
-#else
-	return -EOPNOTSUPP;
-#endif
+	time64_to_tm(now, 0, &tm);
+	rc = hwrm_req_init(bp, req, HWRM_FW_SET_TIME);
+	if (rc)
+		return rc;
+
+	req->year = cpu_to_le16(1900 + tm.tm_year);
+	req->month = 1 + tm.tm_mon;
+	req->day = tm.tm_mday;
+	req->hour = tm.tm_hour;
+	req->minute = tm.tm_min;
+	req->second = tm.tm_sec;
+	return hwrm_req_send(bp, req);
 }
 
-static int bnxt_hwrm_port_qstats(struct bnxt *bp)
+static void bnxt_add_one_ctr(u64 hw, u64 *sw, u64 mask)
 {
-	int rc;
+	u64 sw_tmp;
+
+	hw &= mask;
+	sw_tmp = (*sw & ~mask) | hw;
+	if (hw < (*sw & mask))
+		sw_tmp += mask + 1;
+	WRITE_ONCE(*sw, sw_tmp);
+}
+
+static void __bnxt_accumulate_stats(__le64 *hw_stats, u64 *sw_stats, u64 *masks,
+				    int count, bool ignore_zero)
+{
+	int i;
+
+	for (i = 0; i < count; i++) {
+		u64 hw = le64_to_cpu(READ_ONCE(hw_stats[i]));
+
+		if (ignore_zero && !hw)
+			continue;
+
+		if (masks[i] == -1ULL)
+			sw_stats[i] = hw;
+		else
+			bnxt_add_one_ctr(hw, &sw_stats[i], masks[i]);
+	}
+}
+
+static void bnxt_accumulate_stats(struct bnxt_stats_mem *stats)
+{
+	if (!stats->hw_stats)
+		return;
+
+	__bnxt_accumulate_stats(stats->hw_stats, stats->sw_stats,
+				stats->hw_masks, stats->len / 8, false);
+}
+
+static void bnxt_accumulate_all_stats(struct bnxt *bp)
+{
+	struct bnxt_stats_mem *ring0_stats;
+	bool ignore_zero = false;
+	int i;
+
+	/* Chip bug.  Counter intermittently becomes 0. */
+	if (bp->flags & BNXT_FLAG_CHIP_P5_PLUS)
+		ignore_zero = true;
+
+	for (i = 0; i < bp->cp_nr_rings; i++) {
+		struct bnxt_napi *bnapi = bp->bnapi[i];
+		struct bnxt_cp_ring_info *cpr;
+		struct bnxt_stats_mem *stats;
+
+		cpr = &bnapi->cp_ring;
+		stats = &cpr->stats;
+		if (!i)
+			ring0_stats = stats;
+		__bnxt_accumulate_stats(stats->hw_stats, stats->sw_stats,
+					ring0_stats->hw_masks,
+					ring0_stats->len / 8, ignore_zero);
+	}
+	if (bp->flags & BNXT_FLAG_PORT_STATS) {
+		struct bnxt_stats_mem *stats = &bp->port_stats;
+		__le64 *hw_stats = stats->hw_stats;
+		u64 *sw_stats = stats->sw_stats;
+		u64 *masks = stats->hw_masks;
+		int cnt;
+
+		cnt = sizeof(struct rx_port_stats) / 8;
+		__bnxt_accumulate_stats(hw_stats, sw_stats, masks, cnt, false);
+
+		hw_stats += BNXT_TX_PORT_STATS_BYTE_OFFSET / 8;
+		sw_stats += BNXT_TX_PORT_STATS_BYTE_OFFSET / 8;
+		masks += BNXT_TX_PORT_STATS_BYTE_OFFSET / 8;
+		cnt = sizeof(struct tx_port_stats) / 8;
+		__bnxt_accumulate_stats(hw_stats, sw_stats, masks, cnt, false);
+	}
+	if (bp->flags & BNXT_FLAG_PORT_STATS_EXT) {
+		bnxt_accumulate_stats(&bp->rx_port_stats_ext);
+		bnxt_accumulate_stats(&bp->tx_port_stats_ext);
+	}
+}
+
+static int bnxt_hwrm_port_qstats(struct bnxt *bp, u8 flags)
+{
+	struct hwrm_port_qstats_input *req;
 	struct bnxt_pf_info *pf = &bp->pf;
-	struct hwrm_port_qstats_input req = {0};
+	int rc;
 
 	if (!(bp->flags & BNXT_FLAG_PORT_STATS))
 		return 0;
 
-	bnxt_hwrm_cmd_hdr_init(bp, &req, HWRM_PORT_QSTATS, -1, -1);
-	req.port_id = cpu_to_le16(pf->port_id);
-	req.tx_stat_host_addr = cpu_to_le64(bp->hw_tx_port_stats_map);
-	req.rx_stat_host_addr = cpu_to_le64(bp->hw_rx_port_stats_map);
-	rc = hwrm_send_message(bp, &req, sizeof(req), HWRM_CMD_TIMEOUT);
-	return rc;
+	if (flags && !(bp->fw_cap & BNXT_FW_CAP_EXT_HW_STATS_SUPPORTED))
+		return -EOPNOTSUPP;
+
+	rc = hwrm_req_init(bp, req, HWRM_PORT_QSTATS);
+	if (rc)
+		return rc;
+
+	req->flags = flags;
+	req->port_id = cpu_to_le16(pf->port_id);
+	req->tx_stat_host_addr = cpu_to_le64(bp->port_stats.hw_stats_map +
+					    BNXT_TX_PORT_STATS_BYTE_OFFSET);
+	req->rx_stat_host_addr = cpu_to_le64(bp->port_stats.hw_stats_map);
+	return hwrm_req_send(bp, req);
 }
 
-static void bnxt_hwrm_free_tunnel_ports(struct bnxt *bp)
+static int bnxt_hwrm_port_qstats_ext(struct bnxt *bp, u8 flags)
 {
-	if (bp->vxlan_port_cnt) {
-		bnxt_hwrm_tunnel_dst_port_free(
-			bp, TUNNEL_DST_PORT_FREE_REQ_TUNNEL_TYPE_VXLAN);
+	struct hwrm_queue_pri2cos_qcfg_output *resp_qc;
+	struct hwrm_queue_pri2cos_qcfg_input *req_qc;
+	struct hwrm_port_qstats_ext_output *resp_qs;
+	struct hwrm_port_qstats_ext_input *req_qs;
+	struct bnxt_pf_info *pf = &bp->pf;
+	u32 tx_stat_size;
+	int rc;
+
+	if (!(bp->flags & BNXT_FLAG_PORT_STATS_EXT))
+		return 0;
+
+	if (flags && !(bp->fw_cap & BNXT_FW_CAP_EXT_HW_STATS_SUPPORTED))
+		return -EOPNOTSUPP;
+
+	rc = hwrm_req_init(bp, req_qs, HWRM_PORT_QSTATS_EXT);
+	if (rc)
+		return rc;
+
+	req_qs->flags = flags;
+	req_qs->port_id = cpu_to_le16(pf->port_id);
+	req_qs->rx_stat_size = cpu_to_le16(sizeof(struct rx_port_stats_ext));
+	req_qs->rx_stat_host_addr = cpu_to_le64(bp->rx_port_stats_ext.hw_stats_map);
+	tx_stat_size = bp->tx_port_stats_ext.hw_stats ?
+		       sizeof(struct tx_port_stats_ext) : 0;
+	req_qs->tx_stat_size = cpu_to_le16(tx_stat_size);
+	req_qs->tx_stat_host_addr = cpu_to_le64(bp->tx_port_stats_ext.hw_stats_map);
+	resp_qs = hwrm_req_hold(bp, req_qs);
+	rc = hwrm_req_send(bp, req_qs);
+	if (!rc) {
+		bp->fw_rx_stats_ext_size =
+			le16_to_cpu(resp_qs->rx_stat_size) / 8;
+		if (BNXT_FW_MAJ(bp) < 220 &&
+		    bp->fw_rx_stats_ext_size > BNXT_RX_STATS_EXT_NUM_LEGACY)
+			bp->fw_rx_stats_ext_size = BNXT_RX_STATS_EXT_NUM_LEGACY;
+
+		bp->fw_tx_stats_ext_size = tx_stat_size ?
+			le16_to_cpu(resp_qs->tx_stat_size) / 8 : 0;
+	} else {
+		bp->fw_rx_stats_ext_size = 0;
+		bp->fw_tx_stats_ext_size = 0;
 	}
-	bp->vxlan_port_cnt = 0;
-	if (bp->nge_port_cnt) {
-		bnxt_hwrm_tunnel_dst_port_free(
-			bp, TUNNEL_DST_PORT_FREE_REQ_TUNNEL_TYPE_GENEVE);
+	hwrm_req_drop(bp, req_qs);
+
+	if (flags)
+		return rc;
+
+	if (bp->fw_tx_stats_ext_size <=
+	    offsetof(struct tx_port_stats_ext, pfc_pri0_tx_duration_us) / 8) {
+		bp->pri2cos_valid = 0;
+		return rc;
+	}
+
+	rc = hwrm_req_init(bp, req_qc, HWRM_QUEUE_PRI2COS_QCFG);
+	if (rc)
+		return rc;
+
+	req_qc->flags = cpu_to_le32(QUEUE_PRI2COS_QCFG_REQ_FLAGS_IVLAN);
+
+	resp_qc = hwrm_req_hold(bp, req_qc);
+	rc = hwrm_req_send(bp, req_qc);
+	if (!rc) {
+		u8 *pri2cos;
+		int i, j;
+
+		pri2cos = &resp_qc->pri0_cos_queue_id;
+		for (i = 0; i < 8; i++) {
+			u8 queue_id = pri2cos[i];
+			u8 queue_idx;
+
+			/* Per port queue IDs start from 0, 10, 20, etc */
+			queue_idx = queue_id % 10;
+			if (queue_idx > BNXT_MAX_QUEUE) {
+				bp->pri2cos_valid = false;
+				hwrm_req_drop(bp, req_qc);
+				return rc;
+			}
+			for (j = 0; j < bp->max_q; j++) {
+				if (bp->q_ids[j] == queue_id)
+					bp->pri2cos_idx[i] = queue_idx;
+			}
+		}
+		bp->pri2cos_valid = true;
 	}
-	bp->nge_port_cnt = 0;
+	hwrm_req_drop(bp, req_qc);
+
+	return rc;
+}
+
+static void bnxt_hwrm_free_tunnel_ports(struct bnxt *bp)
+{
+	bnxt_hwrm_tunnel_dst_port_free(bp,
+		TUNNEL_DST_PORT_FREE_REQ_TUNNEL_TYPE_VXLAN);
+	bnxt_hwrm_tunnel_dst_port_free(bp,
+		TUNNEL_DST_PORT_FREE_REQ_TUNNEL_TYPE_GENEVE);
 }
 
 static int bnxt_set_tpa(struct bnxt *bp, bool set_tpa)
@@ -4822,8 +10520,10 @@ static int bnxt_set_tpa(struct bnxt *bp, bool set_tpa)
 
 	if (set_tpa)
 		tpa_flags = bp->flags & BNXT_FLAG_TPA;
+	else if (BNXT_NO_FW_ACCESS(bp))
+		return 0;
 	for (i = 0; i < bp->nr_vnics; i++) {
-		rc = bnxt_hwrm_vnic_set_tpa(bp, i, tpa_flags);
+		rc = bnxt_hwrm_vnic_set_tpa(bp, &bp->vnic_info[i], tpa_flags);
 		if (rc) {
 			netdev_err(bp->dev, "hwrm vnic set tpa failure rc for vnic %d: %x\n",
 				   i, rc);
@@ -4838,22 +10538,32 @@ static void bnxt_hwrm_clear_vnic_rss(struct bnxt *bp)
 	int i;
 
 	for (i = 0; i < bp->nr_vnics; i++)
-		bnxt_hwrm_vnic_set_rss(bp, i, false);
+		bnxt_hwrm_vnic_set_rss(bp, &bp->vnic_info[i], false);
 }
 
-static void bnxt_hwrm_resource_free(struct bnxt *bp, bool close_path,
-				    bool irq_re_init)
+static void bnxt_clear_vnic(struct bnxt *bp)
 {
-	if (bp->vnic_info) {
-		bnxt_hwrm_clear_vnic_filter(bp);
+	if (!bp->vnic_info)
+		return;
+
+	bnxt_hwrm_clear_vnic_filter(bp);
+	if (!(bp->flags & BNXT_FLAG_CHIP_P5_PLUS)) {
 		/* clear all RSS setting before free vnic ctx */
 		bnxt_hwrm_clear_vnic_rss(bp);
 		bnxt_hwrm_vnic_ctx_free(bp);
-		/* before free the vnic, undo the vnic tpa settings */
-		if (bp->flags & BNXT_FLAG_TPA)
-			bnxt_set_tpa(bp, false);
-		bnxt_hwrm_vnic_free(bp);
 	}
+	/* before free the vnic, undo the vnic tpa settings */
+	if (bp->flags & BNXT_FLAG_TPA)
+		bnxt_set_tpa(bp, false);
+	bnxt_hwrm_vnic_free(bp);
+	if (bp->flags & BNXT_FLAG_CHIP_P5_PLUS)
+		bnxt_hwrm_vnic_ctx_free(bp);
+}
+
+static void bnxt_hwrm_resource_free(struct bnxt *bp, bool close_path,
+				    bool irq_re_init)
+{
+	bnxt_clear_vnic(bp);
 	bnxt_hwrm_ring_free(bp, close_path);
 	bnxt_hwrm_ring_grp_free(bp);
 	if (irq_re_init) {
@@ -4862,28 +10572,71 @@ static void bnxt_hwrm_resource_free(struct bnxt *bp, bool close_path,
 	}
 }
 
-static int bnxt_setup_vnic(struct bnxt *bp, u16 vnic_id)
+static int bnxt_hwrm_set_br_mode(struct bnxt *bp, u16 br_mode)
+{
+	struct hwrm_func_cfg_input *req;
+	u8 evb_mode;
+	int rc;
+
+	if (br_mode == BRIDGE_MODE_VEB)
+		evb_mode = FUNC_CFG_REQ_EVB_MODE_VEB;
+	else if (br_mode == BRIDGE_MODE_VEPA)
+		evb_mode = FUNC_CFG_REQ_EVB_MODE_VEPA;
+	else
+		return -EINVAL;
+
+	rc = bnxt_hwrm_func_cfg_short_req_init(bp, &req);
+	if (rc)
+		return rc;
+
+	req->fid = cpu_to_le16(0xffff);
+	req->enables = cpu_to_le32(FUNC_CFG_REQ_ENABLES_EVB_MODE);
+	req->evb_mode = evb_mode;
+	return hwrm_req_send(bp, req);
+}
+
+static int bnxt_hwrm_set_cache_line_size(struct bnxt *bp, int size)
+{
+	struct hwrm_func_cfg_input *req;
+	int rc;
+
+	if (BNXT_VF(bp) || bp->hwrm_spec_code < 0x10803)
+		return 0;
+
+	rc = bnxt_hwrm_func_cfg_short_req_init(bp, &req);
+	if (rc)
+		return rc;
+
+	req->fid = cpu_to_le16(0xffff);
+	req->enables = cpu_to_le32(FUNC_CFG_REQ_ENABLES_CACHE_LINESIZE);
+	req->options = FUNC_CFG_REQ_OPTIONS_CACHE_LINESIZE_SIZE_64;
+	if (size == 128)
+		req->options = FUNC_CFG_REQ_OPTIONS_CACHE_LINESIZE_SIZE_128;
+
+	return hwrm_req_send(bp, req);
+}
+
+static int __bnxt_setup_vnic(struct bnxt *bp, struct bnxt_vnic_info *vnic)
 {
-	struct bnxt_vnic_info *vnic = &bp->vnic_info[vnic_id];
 	int rc;
 
 	if (vnic->flags & BNXT_VNIC_RFS_NEW_RSS_FLAG)
 		goto skip_rss_ctx;
 
 	/* allocate context for vnic */
-	rc = bnxt_hwrm_vnic_ctx_alloc(bp, vnic_id, 0);
+	rc = bnxt_hwrm_vnic_ctx_alloc(bp, vnic, 0);
 	if (rc) {
 		netdev_err(bp->dev, "hwrm vnic %d alloc failure rc: %x\n",
-			   vnic_id, rc);
+			   vnic->vnic_id, rc);
 		goto vnic_setup_err;
 	}
 	bp->rsscos_nr_ctxs++;
 
 	if (BNXT_CHIP_TYPE_NITRO_A0(bp)) {
-		rc = bnxt_hwrm_vnic_ctx_alloc(bp, vnic_id, 1);
+		rc = bnxt_hwrm_vnic_ctx_alloc(bp, vnic, 1);
 		if (rc) {
 			netdev_err(bp->dev, "hwrm vnic %d cos ctx alloc failure rc: %x\n",
-				   vnic_id, rc);
+				   vnic->vnic_id, rc);
 			goto vnic_setup_err;
 		}
 		bp->rsscos_nr_ctxs++;
@@ -4891,26 +10644,26 @@ static int bnxt_setup_vnic(struct bnxt *bp, u16 vnic_id)
 
 skip_rss_ctx:
 	/* configure default vnic, ring grp */
-	rc = bnxt_hwrm_vnic_cfg(bp, vnic_id);
+	rc = bnxt_hwrm_vnic_cfg(bp, vnic);
 	if (rc) {
 		netdev_err(bp->dev, "hwrm vnic %d cfg failure rc: %x\n",
-			   vnic_id, rc);
+			   vnic->vnic_id, rc);
 		goto vnic_setup_err;
 	}
 
 	/* Enable RSS hashing on vnic */
-	rc = bnxt_hwrm_vnic_set_rss(bp, vnic_id, true);
+	rc = bnxt_hwrm_vnic_set_rss(bp, vnic, true);
 	if (rc) {
 		netdev_err(bp->dev, "hwrm vnic %d set rss failure rc: %x\n",
-			   vnic_id, rc);
+			   vnic->vnic_id, rc);
 		goto vnic_setup_err;
 	}
 
 	if (bp->flags & BNXT_FLAG_AGG_RINGS) {
-		rc = bnxt_hwrm_vnic_set_hds(bp, vnic_id);
+		rc = bnxt_hwrm_vnic_set_hds(bp, vnic);
 		if (rc) {
 			netdev_err(bp->dev, "hwrm vnic %d set hds failure rc: %x\n",
-				   vnic_id, rc);
+				   vnic->vnic_id, rc);
 		}
 	}
 
@@ -4918,13 +10671,111 @@ vnic_setup_err:
 	return rc;
 }
 
+int bnxt_hwrm_vnic_update(struct bnxt *bp, struct bnxt_vnic_info *vnic,
+			  u8 valid)
+{
+	struct hwrm_vnic_update_input *req;
+	int rc;
+
+	rc = hwrm_req_init(bp, req, HWRM_VNIC_UPDATE);
+	if (rc)
+		return rc;
+
+	req->vnic_id = cpu_to_le32(vnic->fw_vnic_id);
+
+	if (valid & VNIC_UPDATE_REQ_ENABLES_MRU_VALID)
+		req->mru = cpu_to_le16(vnic->mru);
+
+	req->enables = cpu_to_le32(valid);
+
+	return hwrm_req_send(bp, req);
+}
+
+int bnxt_hwrm_vnic_rss_cfg_p5(struct bnxt *bp, struct bnxt_vnic_info *vnic)
+{
+	int rc;
+
+	rc = bnxt_hwrm_vnic_set_rss_p5(bp, vnic, true);
+	if (rc) {
+		netdev_err(bp->dev, "hwrm vnic %d set rss failure rc: %d\n",
+			   vnic->vnic_id, rc);
+		return rc;
+	}
+	rc = bnxt_hwrm_vnic_cfg(bp, vnic);
+	if (rc)
+		netdev_err(bp->dev, "hwrm vnic %d cfg failure rc: %x\n",
+			   vnic->vnic_id, rc);
+	return rc;
+}
+
+int __bnxt_setup_vnic_p5(struct bnxt *bp, struct bnxt_vnic_info *vnic)
+{
+	int rc, i, nr_ctxs;
+
+	nr_ctxs = bnxt_get_nr_rss_ctxs(bp, bp->rx_nr_rings);
+	for (i = 0; i < nr_ctxs; i++) {
+		rc = bnxt_hwrm_vnic_ctx_alloc(bp, vnic, i);
+		if (rc) {
+			netdev_err(bp->dev, "hwrm vnic %d ctx %d alloc failure rc: %x\n",
+				   vnic->vnic_id, i, rc);
+			break;
+		}
+		bp->rsscos_nr_ctxs++;
+	}
+	if (i < nr_ctxs)
+		return -ENOMEM;
+
+	rc = bnxt_hwrm_vnic_rss_cfg_p5(bp, vnic);
+	if (rc)
+		return rc;
+
+	if (bp->flags & BNXT_FLAG_AGG_RINGS) {
+		rc = bnxt_hwrm_vnic_set_hds(bp, vnic);
+		if (rc) {
+			netdev_err(bp->dev, "hwrm vnic %d set hds failure rc: %x\n",
+				   vnic->vnic_id, rc);
+		}
+	}
+	return rc;
+}
+
+static int bnxt_setup_vnic(struct bnxt *bp, struct bnxt_vnic_info *vnic)
+{
+	if (bp->flags & BNXT_FLAG_CHIP_P5_PLUS)
+		return __bnxt_setup_vnic_p5(bp, vnic);
+	else
+		return __bnxt_setup_vnic(bp, vnic);
+}
+
+static int bnxt_alloc_and_setup_vnic(struct bnxt *bp,
+				     struct bnxt_vnic_info *vnic,
+				     u16 start_rx_ring_idx, int rx_rings)
+{
+	int rc;
+
+	rc = bnxt_hwrm_vnic_alloc(bp, vnic, start_rx_ring_idx, rx_rings);
+	if (rc) {
+		netdev_err(bp->dev, "hwrm vnic %d alloc failure rc: %x\n",
+			   vnic->vnic_id, rc);
+		return rc;
+	}
+	return bnxt_setup_vnic(bp, vnic);
+}
+
 static int bnxt_alloc_rfs_vnics(struct bnxt *bp)
 {
-#ifdef CONFIG_RFS_ACCEL
+	struct bnxt_vnic_info *vnic;
 	int i, rc = 0;
 
+	if (BNXT_SUPPORTS_NTUPLE_VNIC(bp)) {
+		vnic = &bp->vnic_info[BNXT_VNIC_NTUPLE];
+		return bnxt_alloc_and_setup_vnic(bp, vnic, 0, bp->rx_nr_rings);
+	}
+
+	if (bp->flags & BNXT_FLAG_CHIP_P5_PLUS)
+		return 0;
+
 	for (i = 0; i < bp->rx_nr_rings; i++) {
-		struct bnxt_vnic_info *vnic;
 		u16 vnic_id = i + 1;
 		u16 ring_id = i;
 
@@ -4933,29 +10784,155 @@ static int bnxt_alloc_rfs_vnics(struct bnxt *bp)
 
 		vnic = &bp->vnic_info[vnic_id];
 		vnic->flags |= BNXT_VNIC_RFS_FLAG;
-		if (bp->flags & BNXT_FLAG_NEW_RSS_CAP)
+		if (bp->rss_cap & BNXT_RSS_CAP_NEW_RSS_CAP)
 			vnic->flags |= BNXT_VNIC_RFS_NEW_RSS_FLAG;
-		rc = bnxt_hwrm_vnic_alloc(bp, vnic_id, ring_id, 1);
-		if (rc) {
-			netdev_err(bp->dev, "hwrm vnic %d alloc failure rc: %x\n",
-				   vnic_id, rc);
+		if (bnxt_alloc_and_setup_vnic(bp, &bp->vnic_info[vnic_id], ring_id, 1))
 			break;
+	}
+	return rc;
+}
+
+void bnxt_del_one_rss_ctx(struct bnxt *bp, struct bnxt_rss_ctx *rss_ctx,
+			  bool all)
+{
+	struct bnxt_vnic_info *vnic = &rss_ctx->vnic;
+	struct bnxt_filter_base *usr_fltr, *tmp;
+	struct bnxt_ntuple_filter *ntp_fltr;
+	int i;
+
+	if (netif_running(bp->dev)) {
+		bnxt_hwrm_vnic_free_one(bp, &rss_ctx->vnic);
+		for (i = 0; i < BNXT_MAX_CTX_PER_VNIC; i++) {
+			if (vnic->fw_rss_cos_lb_ctx[i] != INVALID_HW_RING_ID)
+				bnxt_hwrm_vnic_ctx_free_one(bp, vnic, i);
+		}
+	}
+	if (!all)
+		return;
+
+	list_for_each_entry_safe(usr_fltr, tmp, &bp->usr_fltr_list, list) {
+		if ((usr_fltr->flags & BNXT_ACT_RSS_CTX) &&
+		    usr_fltr->fw_vnic_id == rss_ctx->index) {
+			ntp_fltr = container_of(usr_fltr,
+						struct bnxt_ntuple_filter,
+						base);
+			bnxt_hwrm_cfa_ntuple_filter_free(bp, ntp_fltr);
+			bnxt_del_ntp_filter(bp, ntp_fltr);
+			bnxt_del_one_usr_fltr(bp, usr_fltr);
 		}
-		rc = bnxt_setup_vnic(bp, vnic_id);
+	}
+
+	if (vnic->rss_table)
+		dma_free_coherent(&bp->pdev->dev, vnic->rss_table_size,
+				  vnic->rss_table,
+				  vnic->rss_table_dma_addr);
+	bp->num_rss_ctx--;
+}
+
+static bool bnxt_vnic_has_rx_ring(struct bnxt *bp, struct bnxt_vnic_info *vnic,
+				  int rxr_id)
+{
+	u16 tbl_size = bnxt_get_rxfh_indir_size(bp->dev);
+	int i, vnic_rx;
+
+	/* Ntuple VNIC always has all the rx rings. Any change of ring id
+	 * must be updated because a future filter may use it.
+	 */
+	if (vnic->flags & BNXT_VNIC_NTUPLE_FLAG)
+		return true;
+
+	for (i = 0; i < tbl_size; i++) {
+		if (vnic->flags & BNXT_VNIC_RSSCTX_FLAG)
+			vnic_rx = ethtool_rxfh_context_indir(vnic->rss_ctx)[i];
+		else
+			vnic_rx = bp->rss_indir_tbl[i];
+
+		if (rxr_id == vnic_rx)
+			return true;
+	}
+
+	return false;
+}
+
+static int bnxt_set_vnic_mru_p5(struct bnxt *bp, struct bnxt_vnic_info *vnic,
+				u16 mru, int rxr_id)
+{
+	int rc;
+
+	if (!bnxt_vnic_has_rx_ring(bp, vnic, rxr_id))
+		return 0;
+
+	if (mru) {
+		rc = bnxt_hwrm_vnic_set_rss_p5(bp, vnic, true);
+		if (rc) {
+			netdev_err(bp->dev, "hwrm vnic %d set rss failure rc: %d\n",
+				   vnic->vnic_id, rc);
+			return rc;
+		}
+	}
+	vnic->mru = mru;
+	bnxt_hwrm_vnic_update(bp, vnic,
+			      VNIC_UPDATE_REQ_ENABLES_MRU_VALID);
+
+	return 0;
+}
+
+static int bnxt_set_rss_ctx_vnic_mru(struct bnxt *bp, u16 mru, int rxr_id)
+{
+	struct ethtool_rxfh_context *ctx;
+	unsigned long context;
+	int rc;
+
+	xa_for_each(&bp->dev->ethtool->rss_ctx, context, ctx) {
+		struct bnxt_rss_ctx *rss_ctx = ethtool_rxfh_context_priv(ctx);
+		struct bnxt_vnic_info *vnic = &rss_ctx->vnic;
+
+		rc = bnxt_set_vnic_mru_p5(bp, vnic, mru, rxr_id);
 		if (rc)
-			break;
+			return rc;
 	}
-	return rc;
-#else
+
 	return 0;
-#endif
 }
 
-/* Allow PF and VF with default VLAN to be in promiscuous mode */
+static void bnxt_hwrm_realloc_rss_ctx_vnic(struct bnxt *bp)
+{
+	bool set_tpa = !!(bp->flags & BNXT_FLAG_TPA);
+	struct ethtool_rxfh_context *ctx;
+	unsigned long context;
+
+	xa_for_each(&bp->dev->ethtool->rss_ctx, context, ctx) {
+		struct bnxt_rss_ctx *rss_ctx = ethtool_rxfh_context_priv(ctx);
+		struct bnxt_vnic_info *vnic = &rss_ctx->vnic;
+
+		if (bnxt_hwrm_vnic_alloc(bp, vnic, 0, bp->rx_nr_rings) ||
+		    bnxt_hwrm_vnic_set_tpa(bp, vnic, set_tpa) ||
+		    __bnxt_setup_vnic_p5(bp, vnic)) {
+			netdev_err(bp->dev, "Failed to restore RSS ctx %d\n",
+				   rss_ctx->index);
+			bnxt_del_one_rss_ctx(bp, rss_ctx, true);
+			ethtool_rxfh_context_lost(bp->dev, rss_ctx->index);
+		}
+	}
+}
+
+static void bnxt_clear_rss_ctxs(struct bnxt *bp)
+{
+	struct ethtool_rxfh_context *ctx;
+	unsigned long context;
+
+	xa_for_each(&bp->dev->ethtool->rss_ctx, context, ctx) {
+		struct bnxt_rss_ctx *rss_ctx = ethtool_rxfh_context_priv(ctx);
+
+		bnxt_del_one_rss_ctx(bp, rss_ctx, false);
+	}
+}
+
+/* Allow PF, trusted VFs and VFs with default VLAN to be in promiscuous mode */
 static bool bnxt_promisc_ok(struct bnxt *bp)
 {
 #ifdef CONFIG_BNXT_SRIOV
-	if (BNXT_VF(bp) && !bp->vf.vlan)
+	if (BNXT_VF(bp) && !bp->vf.vlan && !bnxt_is_trusted_vf(bp, &bp->vf))
 		return false;
 #endif
 	return true;
@@ -4963,16 +10940,17 @@ static bool bnxt_promisc_ok(struct bnxt *bp)
 
 static int bnxt_setup_nitroa0_vnic(struct bnxt *bp)
 {
+	struct bnxt_vnic_info *vnic = &bp->vnic_info[1];
 	unsigned int rc = 0;
 
-	rc = bnxt_hwrm_vnic_alloc(bp, 1, bp->rx_nr_rings - 1, 1);
+	rc = bnxt_hwrm_vnic_alloc(bp, vnic, bp->rx_nr_rings - 1, 1);
 	if (rc) {
 		netdev_err(bp->dev, "Cannot allocate special vnic for NS2 A0: %x\n",
 			   rc);
 		return rc;
 	}
 
-	rc = bnxt_hwrm_vnic_cfg(bp, 1);
+	rc = bnxt_hwrm_vnic_cfg(bp, vnic);
 	if (rc) {
 		netdev_err(bp->dev, "Cannot allocate special vnic for NS2 A0: %x\n",
 			   rc);
@@ -4986,7 +10964,7 @@ static bool bnxt_mc_list_updated(struct bnxt *, u32 *);
 
 static int bnxt_init_chip(struct bnxt *bp, bool irq_re_init)
 {
-	struct bnxt_vnic_info *vnic = &bp->vnic_info[0];
+	struct bnxt_vnic_info *vnic = &bp->vnic_info[BNXT_VNIC_DEFAULT];
 	int rc = 0;
 	unsigned int rx_nr_rings = bp->rx_nr_rings;
 
@@ -5015,15 +10993,20 @@ static int bnxt_init_chip(struct bnxt *bp, bool irq_re_init)
 		rx_nr_rings--;
 
 	/* default vnic 0 */
-	rc = bnxt_hwrm_vnic_alloc(bp, 0, 0, rx_nr_rings);
+	rc = bnxt_hwrm_vnic_alloc(bp, vnic, 0, rx_nr_rings);
 	if (rc) {
 		netdev_err(bp->dev, "hwrm vnic alloc failure rc: %x\n", rc);
 		goto err_out;
 	}
 
-	rc = bnxt_setup_vnic(bp, 0);
+	if (BNXT_VF(bp))
+		bnxt_hwrm_func_qcfg(bp);
+
+	rc = bnxt_setup_vnic(bp, vnic);
 	if (rc)
 		goto err_out;
+	if (bp->rss_cap & BNXT_RSS_CAP_RSS_HASH_TYPE_DELTA)
+		bnxt_hwrm_update_rss_hash_cfg(bp);
 
 	if (bp->flags & BNXT_FLAG_RFS) {
 		rc = bnxt_alloc_rfs_vnics(bp);
@@ -5043,20 +11026,28 @@ static int bnxt_init_chip(struct bnxt *bp, bool irq_re_init)
 	/* Filter for default vnic 0 */
 	rc = bnxt_hwrm_set_vnic_filter(bp, 0, 0, bp->dev->dev_addr);
 	if (rc) {
-		netdev_err(bp->dev, "HWRM vnic filter failure rc: %x\n", rc);
+		if (BNXT_VF(bp) && rc == -ENODEV)
+			netdev_err(bp->dev, "Cannot configure L2 filter while PF is unavailable\n");
+		else
+			netdev_err(bp->dev, "HWRM vnic filter failure rc: %x\n", rc);
 		goto err_out;
 	}
 	vnic->uc_filter_count = 1;
 
-	vnic->rx_mask = CFA_L2_SET_RX_MASK_REQ_MASK_BCAST;
+	vnic->rx_mask = 0;
+	if (test_bit(BNXT_STATE_HALF_OPEN, &bp->state))
+		goto skip_rx_mask;
+
+	if (bp->dev->flags & IFF_BROADCAST)
+		vnic->rx_mask |= CFA_L2_SET_RX_MASK_REQ_MASK_BCAST;
 
-	if ((bp->dev->flags & IFF_PROMISC) && bnxt_promisc_ok(bp))
+	if (bp->dev->flags & IFF_PROMISC)
 		vnic->rx_mask |= CFA_L2_SET_RX_MASK_REQ_MASK_PROMISCUOUS;
 
 	if (bp->dev->flags & IFF_ALLMULTI) {
 		vnic->rx_mask |= CFA_L2_SET_RX_MASK_REQ_MASK_ALL_MCAST;
 		vnic->mc_list_count = 0;
-	} else {
+	} else if (bp->dev->flags & IFF_MULTICAST) {
 		u32 mask = 0;
 
 		bnxt_mc_list_updated(bp, &mask);
@@ -5067,6 +11058,7 @@ static int bnxt_init_chip(struct bnxt *bp, bool irq_re_init)
 	if (rc)
 		goto err_out;
 
+skip_rx_mask:
 	rc = bnxt_hwrm_set_coal(bp);
 	if (rc)
 		netdev_warn(bp->dev, "HWRM set coalescing failure rc: %x\n",
@@ -5131,8 +11123,8 @@ static int bnxt_set_real_num_queues(struct bnxt *bp)
 	return rc;
 }
 
-static int bnxt_trim_rings(struct bnxt *bp, int *rx, int *tx, int max,
-			   bool shared)
+static int __bnxt_trim_rings(struct bnxt *bp, int *rx, int *tx, int max,
+			     bool shared)
 {
 	int _rx = *rx, _tx = *tx;
 
@@ -5155,24 +11147,65 @@ static int bnxt_trim_rings(struct bnxt *bp, int *rx, int *tx, int max,
 	return 0;
 }
 
+static int __bnxt_num_tx_to_cp(struct bnxt *bp, int tx, int tx_sets, int tx_xdp)
+{
+	return (tx - tx_xdp) / tx_sets + tx_xdp;
+}
+
+int bnxt_num_tx_to_cp(struct bnxt *bp, int tx)
+{
+	int tcs = bp->num_tc;
+
+	if (!tcs)
+		tcs = 1;
+	return __bnxt_num_tx_to_cp(bp, tx, tcs, bp->tx_nr_rings_xdp);
+}
+
+static int bnxt_num_cp_to_tx(struct bnxt *bp, int tx_cp)
+{
+	int tcs = bp->num_tc;
+
+	return (tx_cp - bp->tx_nr_rings_xdp) * tcs +
+	       bp->tx_nr_rings_xdp;
+}
+
+static int bnxt_trim_rings(struct bnxt *bp, int *rx, int *tx, int max,
+			   bool sh)
+{
+	int tx_cp = bnxt_num_tx_to_cp(bp, *tx);
+
+	if (tx_cp != *tx) {
+		int tx_saved = tx_cp, rc;
+
+		rc = __bnxt_trim_rings(bp, rx, &tx_cp, max, sh);
+		if (rc)
+			return rc;
+		if (tx_cp != tx_saved)
+			*tx = bnxt_num_cp_to_tx(bp, tx_cp);
+		return 0;
+	}
+	return __bnxt_trim_rings(bp, rx, tx, max, sh);
+}
+
 static void bnxt_setup_msix(struct bnxt *bp)
 {
 	const int len = sizeof(bp->irq_tbl[0].name);
 	struct net_device *dev = bp->dev;
 	int tcs, i;
 
-	tcs = netdev_get_num_tc(dev);
-	if (tcs > 1) {
+	tcs = bp->num_tc;
+	if (tcs) {
 		int i, off, count;
 
 		for (i = 0; i < tcs; i++) {
 			count = bp->tx_nr_rings_per_tc;
-			off = i * count;
+			off = BNXT_TC_TO_RING_BASE(bp, i);
 			netdev_set_tc_queue(dev, i, count, off);
 		}
 	}
 
 	for (i = 0; i < bp->cp_nr_rings; i++) {
+		int map_idx = bnxt_cp_num_to_irq_num(bp, i);
 		char *attr;
 
 		if (bp->flags & BNXT_FLAG_SHARED_RINGS)
@@ -5182,211 +11215,404 @@ static void bnxt_setup_msix(struct bnxt *bp)
 		else
 			attr = "tx";
 
-		snprintf(bp->irq_tbl[i].name, len, "%s-%s-%d", dev->name, attr,
-			 i);
-		bp->irq_tbl[i].handler = bnxt_msix;
+		snprintf(bp->irq_tbl[map_idx].name, len, "%s-%s-%d", dev->name,
+			 attr, i);
+		bp->irq_tbl[map_idx].handler = bnxt_msix;
 	}
 }
 
-static void bnxt_setup_inta(struct bnxt *bp)
+static int bnxt_init_int_mode(struct bnxt *bp);
+
+static int bnxt_change_msix(struct bnxt *bp, int total)
 {
-	const int len = sizeof(bp->irq_tbl[0].name);
+	struct msi_map map;
+	int i;
 
-	if (netdev_get_num_tc(bp->dev))
-		netdev_reset_tc(bp->dev);
+	/* add MSIX to the end if needed */
+	for (i = bp->total_irqs; i < total; i++) {
+		map = pci_msix_alloc_irq_at(bp->pdev, i, NULL);
+		if (map.index < 0)
+			return bp->total_irqs;
+		bp->irq_tbl[i].vector = map.virq;
+		bp->total_irqs++;
+	}
 
-	snprintf(bp->irq_tbl[0].name, len, "%s-%s-%d", bp->dev->name, "TxRx",
-		 0);
-	bp->irq_tbl[0].handler = bnxt_inta;
+	/* trim MSIX from the end if needed */
+	for (i = bp->total_irqs; i > total; i--) {
+		map.index = i - 1;
+		map.virq = bp->irq_tbl[i - 1].vector;
+		pci_msix_free_irq(bp->pdev, map);
+		bp->total_irqs--;
+	}
+	return bp->total_irqs;
 }
 
 static int bnxt_setup_int_mode(struct bnxt *bp)
 {
 	int rc;
 
-	if (bp->flags & BNXT_FLAG_USING_MSIX)
-		bnxt_setup_msix(bp);
-	else
-		bnxt_setup_inta(bp);
+	if (!bp->irq_tbl) {
+		rc = bnxt_init_int_mode(bp);
+		if (rc || !bp->irq_tbl)
+			return rc ?: -ENODEV;
+	}
+
+	bnxt_setup_msix(bp);
 
 	rc = bnxt_set_real_num_queues(bp);
 	return rc;
 }
 
-#ifdef CONFIG_RFS_ACCEL
 static unsigned int bnxt_get_max_func_rss_ctxs(struct bnxt *bp)
 {
-#if defined(CONFIG_BNXT_SRIOV)
-	if (BNXT_VF(bp))
-		return bp->vf.max_rsscos_ctxs;
-#endif
-	return bp->pf.max_rsscos_ctxs;
+	return bp->hw_resc.max_rsscos_ctxs;
 }
 
 static unsigned int bnxt_get_max_func_vnics(struct bnxt *bp)
 {
-#if defined(CONFIG_BNXT_SRIOV)
-	if (BNXT_VF(bp))
-		return bp->vf.max_vnics;
-#endif
-	return bp->pf.max_vnics;
+	return bp->hw_resc.max_vnics;
 }
-#endif
 
 unsigned int bnxt_get_max_func_stat_ctxs(struct bnxt *bp)
 {
-#if defined(CONFIG_BNXT_SRIOV)
-	if (BNXT_VF(bp))
-		return bp->vf.max_stat_ctxs;
-#endif
-	return bp->pf.max_stat_ctxs;
+	return bp->hw_resc.max_stat_ctxs;
 }
 
-void bnxt_set_max_func_stat_ctxs(struct bnxt *bp, unsigned int max)
+unsigned int bnxt_get_max_func_cp_rings(struct bnxt *bp)
 {
-#if defined(CONFIG_BNXT_SRIOV)
-	if (BNXT_VF(bp))
-		bp->vf.max_stat_ctxs = max;
-	else
-#endif
-		bp->pf.max_stat_ctxs = max;
+	return bp->hw_resc.max_cp_rings;
 }
 
-unsigned int bnxt_get_max_func_cp_rings(struct bnxt *bp)
+static unsigned int bnxt_get_max_func_cp_rings_for_en(struct bnxt *bp)
 {
-#if defined(CONFIG_BNXT_SRIOV)
-	if (BNXT_VF(bp))
-		return bp->vf.max_cp_rings;
-#endif
-	return bp->pf.max_cp_rings;
+	unsigned int cp = bp->hw_resc.max_cp_rings;
+
+	if (!(bp->flags & BNXT_FLAG_CHIP_P5_PLUS))
+		cp -= bnxt_get_ulp_msix_num(bp);
+
+	return cp;
 }
 
-void bnxt_set_max_func_cp_rings(struct bnxt *bp, unsigned int max)
+static unsigned int bnxt_get_max_func_irqs(struct bnxt *bp)
 {
-#if defined(CONFIG_BNXT_SRIOV)
-	if (BNXT_VF(bp))
-		bp->vf.max_cp_rings = max;
-	else
-#endif
-		bp->pf.max_cp_rings = max;
+	struct bnxt_hw_resc *hw_resc = &bp->hw_resc;
+
+	if (bp->flags & BNXT_FLAG_CHIP_P5_PLUS)
+		return min_t(unsigned int, hw_resc->max_irqs, hw_resc->max_nqs);
+
+	return min_t(unsigned int, hw_resc->max_irqs, hw_resc->max_cp_rings);
 }
 
-static unsigned int bnxt_get_max_func_irqs(struct bnxt *bp)
+static void bnxt_set_max_func_irqs(struct bnxt *bp, unsigned int max_irqs)
 {
-#if defined(CONFIG_BNXT_SRIOV)
-	if (BNXT_VF(bp))
-		return min_t(unsigned int, bp->vf.max_irqs,
-			     bp->vf.max_cp_rings);
-#endif
-	return min_t(unsigned int, bp->pf.max_irqs, bp->pf.max_cp_rings);
+	bp->hw_resc.max_irqs = max_irqs;
 }
 
-void bnxt_set_max_func_irqs(struct bnxt *bp, unsigned int max_irqs)
+unsigned int bnxt_get_avail_cp_rings_for_en(struct bnxt *bp)
 {
-#if defined(CONFIG_BNXT_SRIOV)
-	if (BNXT_VF(bp))
-		bp->vf.max_irqs = max_irqs;
+	unsigned int cp;
+
+	cp = bnxt_get_max_func_cp_rings_for_en(bp);
+	if (bp->flags & BNXT_FLAG_CHIP_P5_PLUS)
+		return cp - bp->rx_nr_rings - bp->tx_nr_rings;
 	else
-#endif
-		bp->pf.max_irqs = max_irqs;
+		return cp - bp->cp_nr_rings;
 }
 
-static int bnxt_init_msix(struct bnxt *bp)
+unsigned int bnxt_get_avail_stat_ctxs_for_en(struct bnxt *bp)
 {
-	int i, total_vecs, rc = 0, min = 1;
-	struct msix_entry *msix_ent;
+	return bnxt_get_max_func_stat_ctxs(bp) - bnxt_get_func_stat_ctxs(bp);
+}
 
-	total_vecs = bnxt_get_max_func_irqs(bp);
-	msix_ent = kcalloc(total_vecs, sizeof(struct msix_entry), GFP_KERNEL);
-	if (!msix_ent)
-		return -ENOMEM;
+static int bnxt_get_avail_msix(struct bnxt *bp, int num)
+{
+	int max_irq = bnxt_get_max_func_irqs(bp);
+	int total_req = bp->cp_nr_rings + num;
 
-	for (i = 0; i < total_vecs; i++) {
-		msix_ent[i].entry = i;
-		msix_ent[i].vector = 0;
+	if (max_irq < total_req) {
+		num = max_irq - bp->cp_nr_rings;
+		if (num <= 0)
+			return 0;
 	}
+	return num;
+}
+
+static int bnxt_get_num_msix(struct bnxt *bp)
+{
+	if (!BNXT_NEW_RM(bp))
+		return bnxt_get_max_func_irqs(bp);
+
+	return bnxt_nq_rings_in_use(bp);
+}
+
+static int bnxt_init_int_mode(struct bnxt *bp)
+{
+	int i, total_vecs, max, rc = 0, min = 1, ulp_msix, tx_cp, tbl_size;
+
+	total_vecs = bnxt_get_num_msix(bp);
+	max = bnxt_get_max_func_irqs(bp);
+	if (total_vecs > max)
+		total_vecs = max;
+
+	if (!total_vecs)
+		return 0;
 
 	if (!(bp->flags & BNXT_FLAG_SHARED_RINGS))
 		min = 2;
 
-	total_vecs = pci_enable_msix_range(bp->pdev, msix_ent, min, total_vecs);
-	if (total_vecs < 0) {
+	total_vecs = pci_alloc_irq_vectors(bp->pdev, min, total_vecs,
+					   PCI_IRQ_MSIX);
+	ulp_msix = bnxt_get_ulp_msix_num(bp);
+	if (total_vecs < 0 || total_vecs < ulp_msix) {
 		rc = -ENODEV;
 		goto msix_setup_exit;
 	}
 
-	bp->irq_tbl = kcalloc(total_vecs, sizeof(struct bnxt_irq), GFP_KERNEL);
+	tbl_size = total_vecs;
+	if (pci_msix_can_alloc_dyn(bp->pdev))
+		tbl_size = max;
+	bp->irq_tbl = kcalloc(tbl_size, sizeof(*bp->irq_tbl), GFP_KERNEL);
 	if (bp->irq_tbl) {
 		for (i = 0; i < total_vecs; i++)
-			bp->irq_tbl[i].vector = msix_ent[i].vector;
+			bp->irq_tbl[i].vector = pci_irq_vector(bp->pdev, i);
 
 		bp->total_irqs = total_vecs;
 		/* Trim rings based upon num of vectors allocated */
 		rc = bnxt_trim_rings(bp, &bp->rx_nr_rings, &bp->tx_nr_rings,
-				     total_vecs, min == 1);
+				     total_vecs - ulp_msix, min == 1);
 		if (rc)
 			goto msix_setup_exit;
 
-		bp->tx_nr_rings_per_tc = bp->tx_nr_rings;
+		tx_cp = bnxt_num_tx_to_cp(bp, bp->tx_nr_rings);
 		bp->cp_nr_rings = (min == 1) ?
-				  max_t(int, bp->tx_nr_rings, bp->rx_nr_rings) :
-				  bp->tx_nr_rings + bp->rx_nr_rings;
+				  max_t(int, tx_cp, bp->rx_nr_rings) :
+				  tx_cp + bp->rx_nr_rings;
 
 	} else {
 		rc = -ENOMEM;
 		goto msix_setup_exit;
 	}
-	bp->flags |= BNXT_FLAG_USING_MSIX;
-	kfree(msix_ent);
 	return 0;
 
 msix_setup_exit:
-	netdev_err(bp->dev, "bnxt_init_msix err: %x\n", rc);
+	netdev_err(bp->dev, "bnxt_init_int_mode err: %x\n", rc);
 	kfree(bp->irq_tbl);
 	bp->irq_tbl = NULL;
-	pci_disable_msix(bp->pdev);
-	kfree(msix_ent);
+	pci_free_irq_vectors(bp->pdev);
 	return rc;
 }
 
-static int bnxt_init_inta(struct bnxt *bp)
+static void bnxt_clear_int_mode(struct bnxt *bp)
 {
-	bp->irq_tbl = kcalloc(1, sizeof(struct bnxt_irq), GFP_KERNEL);
-	if (!bp->irq_tbl)
+	pci_free_irq_vectors(bp->pdev);
+
+	kfree(bp->irq_tbl);
+	bp->irq_tbl = NULL;
+}
+
+int bnxt_reserve_rings(struct bnxt *bp, bool irq_re_init)
+{
+	bool irq_cleared = false;
+	bool irq_change = false;
+	int tcs = bp->num_tc;
+	int irqs_required;
+	int rc;
+
+	if (!bnxt_need_reserve_rings(bp))
+		return 0;
+
+	if (BNXT_NEW_RM(bp) && !bnxt_ulp_registered(bp->edev)) {
+		int ulp_msix = bnxt_get_avail_msix(bp, bp->ulp_num_msix_want);
+
+		if (ulp_msix > bp->ulp_num_msix_want)
+			ulp_msix = bp->ulp_num_msix_want;
+		irqs_required = ulp_msix + bp->cp_nr_rings;
+	} else {
+		irqs_required = bnxt_get_num_msix(bp);
+	}
+
+	if (irq_re_init && BNXT_NEW_RM(bp) && irqs_required != bp->total_irqs) {
+		irq_change = true;
+		if (!pci_msix_can_alloc_dyn(bp->pdev)) {
+			bnxt_ulp_irq_stop(bp);
+			bnxt_clear_int_mode(bp);
+			irq_cleared = true;
+		}
+	}
+	rc = __bnxt_reserve_rings(bp);
+	if (irq_cleared) {
+		if (!rc)
+			rc = bnxt_init_int_mode(bp);
+		bnxt_ulp_irq_restart(bp, rc);
+	} else if (irq_change && !rc) {
+		if (bnxt_change_msix(bp, irqs_required) != irqs_required)
+			rc = -ENOSPC;
+	}
+	if (rc) {
+		netdev_err(bp->dev, "ring reservation/IRQ init failure rc: %d\n", rc);
+		return rc;
+	}
+	if (tcs && (bp->tx_nr_rings_per_tc * tcs !=
+		    bp->tx_nr_rings - bp->tx_nr_rings_xdp)) {
+		netdev_err(bp->dev, "tx ring reservation failure\n");
+		netdev_reset_tc(bp->dev);
+		bp->num_tc = 0;
+		if (bp->tx_nr_rings_xdp)
+			bp->tx_nr_rings_per_tc = bp->tx_nr_rings_xdp;
+		else
+			bp->tx_nr_rings_per_tc = bp->tx_nr_rings;
 		return -ENOMEM;
+	}
+	return 0;
+}
+
+static void bnxt_tx_queue_stop(struct bnxt *bp, int idx)
+{
+	struct bnxt_tx_ring_info *txr;
+	struct netdev_queue *txq;
+	struct bnxt_napi *bnapi;
+	int i;
+
+	bnapi = bp->bnapi[idx];
+	bnxt_for_each_napi_tx(i, bnapi, txr) {
+		WRITE_ONCE(txr->dev_state, BNXT_DEV_STATE_CLOSING);
+		synchronize_net();
+
+		if (!(bnapi->flags & BNXT_NAPI_FLAG_XDP)) {
+			txq = netdev_get_tx_queue(bp->dev, txr->txq_index);
+			if (txq) {
+				__netif_tx_lock_bh(txq);
+				netif_tx_stop_queue(txq);
+				__netif_tx_unlock_bh(txq);
+			}
+		}
+
+		if (!bp->tph_mode)
+			continue;
+
+		bnxt_hwrm_tx_ring_free(bp, txr, true);
+		bnxt_hwrm_cp_ring_free(bp, txr->tx_cpr);
+		bnxt_free_one_tx_ring_skbs(bp, txr, txr->txq_index);
+		bnxt_clear_one_cp_ring(bp, txr->tx_cpr);
+	}
+}
+
+static int bnxt_tx_queue_start(struct bnxt *bp, int idx)
+{
+	struct bnxt_tx_ring_info *txr;
+	struct netdev_queue *txq;
+	struct bnxt_napi *bnapi;
+	int rc, i;
+
+	bnapi = bp->bnapi[idx];
+	/* All rings have been reserved and previously allocated.
+	 * Reallocating with the same parameters should never fail.
+	 */
+	bnxt_for_each_napi_tx(i, bnapi, txr) {
+		if (!bp->tph_mode)
+			goto start_tx;
+
+		rc = bnxt_hwrm_cp_ring_alloc_p5(bp, txr->tx_cpr);
+		if (rc)
+			return rc;
+
+		rc = bnxt_hwrm_tx_ring_alloc(bp, txr, false);
+		if (rc)
+			return rc;
+
+		txr->tx_prod = 0;
+		txr->tx_cons = 0;
+		txr->tx_hw_cons = 0;
+start_tx:
+		WRITE_ONCE(txr->dev_state, 0);
+		synchronize_net();
+
+		if (bnapi->flags & BNXT_NAPI_FLAG_XDP)
+			continue;
+
+		txq = netdev_get_tx_queue(bp->dev, txr->txq_index);
+		if (txq)
+			netif_tx_start_queue(txq);
+	}
 
-	bp->total_irqs = 1;
-	bp->rx_nr_rings = 1;
-	bp->tx_nr_rings = 1;
-	bp->cp_nr_rings = 1;
-	bp->tx_nr_rings_per_tc = bp->tx_nr_rings;
-	bp->flags |= BNXT_FLAG_SHARED_RINGS;
-	bp->irq_tbl[0].vector = bp->pdev->irq;
 	return 0;
 }
 
-static int bnxt_init_int_mode(struct bnxt *bp)
+static void bnxt_irq_affinity_notify(struct irq_affinity_notify *notify,
+				     const cpumask_t *mask)
 {
-	int rc = 0;
+	struct bnxt_irq *irq;
+	u16 tag;
+	int err;
+
+	irq = container_of(notify, struct bnxt_irq, affinity_notify);
+
+	if (!irq->bp->tph_mode)
+		return;
+
+	cpumask_copy(irq->cpu_mask, mask);
+
+	if (irq->ring_nr >= irq->bp->rx_nr_rings)
+		return;
+
+	if (pcie_tph_get_cpu_st(irq->bp->pdev, TPH_MEM_TYPE_VM,
+				cpumask_first(irq->cpu_mask), &tag))
+		return;
 
-	if (bp->flags & BNXT_FLAG_MSIX_CAP)
-		rc = bnxt_init_msix(bp);
+	if (pcie_tph_set_st_entry(irq->bp->pdev, irq->msix_nr, tag))
+		return;
 
-	if (!(bp->flags & BNXT_FLAG_USING_MSIX) && BNXT_PF(bp)) {
-		/* fallback to INTA */
-		rc = bnxt_init_inta(bp);
+	netdev_lock(irq->bp->dev);
+	if (netif_running(irq->bp->dev)) {
+		err = netdev_rx_queue_restart(irq->bp->dev, irq->ring_nr);
+		if (err)
+			netdev_err(irq->bp->dev,
+				   "RX queue restart failed: err=%d\n", err);
 	}
-	return rc;
+	netdev_unlock(irq->bp->dev);
 }
 
-static void bnxt_clear_int_mode(struct bnxt *bp)
+static void bnxt_irq_affinity_release(struct kref *ref)
 {
-	if (bp->flags & BNXT_FLAG_USING_MSIX)
-		pci_disable_msix(bp->pdev);
+	struct irq_affinity_notify *notify =
+		container_of(ref, struct irq_affinity_notify, kref);
+	struct bnxt_irq *irq;
 
-	kfree(bp->irq_tbl);
-	bp->irq_tbl = NULL;
-	bp->flags &= ~BNXT_FLAG_USING_MSIX;
+	irq = container_of(notify, struct bnxt_irq, affinity_notify);
+
+	if (!irq->bp->tph_mode)
+		return;
+
+	if (pcie_tph_set_st_entry(irq->bp->pdev, irq->msix_nr, 0)) {
+		netdev_err(irq->bp->dev,
+			   "Setting ST=0 for MSIX entry %d failed\n",
+			   irq->msix_nr);
+		return;
+	}
+}
+
+static void bnxt_release_irq_notifier(struct bnxt_irq *irq)
+{
+	irq_set_affinity_notifier(irq->vector, NULL);
+}
+
+static void bnxt_register_irq_notifier(struct bnxt *bp, struct bnxt_irq *irq)
+{
+	struct irq_affinity_notify *notify;
+
+	irq->bp = bp;
+
+	/* Nothing to do if TPH is not enabled */
+	if (!bp->tph_mode)
+		return;
+
+	/* Register IRQ affinity notifier */
+	notify = &irq->affinity_notify;
+	notify->irq = irq->vector;
+	notify->notify = bnxt_irq_affinity_notify;
+	notify->release = bnxt_irq_affinity_release;
+
+	irq_set_affinity_notifier(irq->vector, notify);
 }
 
 static void bnxt_free_irq(struct bnxt *bp)
@@ -5398,45 +11624,102 @@ static void bnxt_free_irq(struct bnxt *bp)
 	free_irq_cpu_rmap(bp->dev->rx_cpu_rmap);
 	bp->dev->rx_cpu_rmap = NULL;
 #endif
-	if (!bp->irq_tbl)
+	if (!bp->irq_tbl || !bp->bnapi)
 		return;
 
 	for (i = 0; i < bp->cp_nr_rings; i++) {
-		irq = &bp->irq_tbl[i];
-		if (irq->requested)
+		int map_idx = bnxt_cp_num_to_irq_num(bp, i);
+
+		irq = &bp->irq_tbl[map_idx];
+		if (irq->requested) {
+			if (irq->have_cpumask) {
+				irq_update_affinity_hint(irq->vector, NULL);
+				free_cpumask_var(irq->cpu_mask);
+				irq->have_cpumask = 0;
+			}
+
+			bnxt_release_irq_notifier(irq);
+
 			free_irq(irq->vector, bp->bnapi[i]);
+		}
+
 		irq->requested = 0;
 	}
+
+	/* Disable TPH support */
+	pcie_disable_tph(bp->pdev);
+	bp->tph_mode = 0;
 }
 
 static int bnxt_request_irq(struct bnxt *bp)
 {
+	struct cpu_rmap *rmap = NULL;
 	int i, j, rc = 0;
 	unsigned long flags = 0;
+
+	rc = bnxt_setup_int_mode(bp);
+	if (rc) {
+		netdev_err(bp->dev, "bnxt_setup_int_mode err: %x\n",
+			   rc);
+		return rc;
+	}
 #ifdef CONFIG_RFS_ACCEL
-	struct cpu_rmap *rmap = bp->dev->rx_cpu_rmap;
+	rmap = bp->dev->rx_cpu_rmap;
 #endif
 
-	if (!(bp->flags & BNXT_FLAG_USING_MSIX))
-		flags = IRQF_SHARED;
+	/* Enable TPH support as part of IRQ request */
+	rc = pcie_enable_tph(bp->pdev, PCI_TPH_ST_IV_MODE);
+	if (!rc)
+		bp->tph_mode = PCI_TPH_ST_IV_MODE;
 
 	for (i = 0, j = 0; i < bp->cp_nr_rings; i++) {
-		struct bnxt_irq *irq = &bp->irq_tbl[i];
-#ifdef CONFIG_RFS_ACCEL
-		if (rmap && bp->bnapi[i]->rx_ring) {
+		int map_idx = bnxt_cp_num_to_irq_num(bp, i);
+		struct bnxt_irq *irq = &bp->irq_tbl[map_idx];
+
+		if (IS_ENABLED(CONFIG_RFS_ACCEL) &&
+		    rmap && bp->bnapi[i]->rx_ring) {
 			rc = irq_cpu_rmap_add(rmap, irq->vector);
 			if (rc)
 				netdev_warn(bp->dev, "failed adding irq rmap for ring %d\n",
 					    j);
 			j++;
 		}
-#endif
+
 		rc = request_irq(irq->vector, irq->handler, flags, irq->name,
 				 bp->bnapi[i]);
 		if (rc)
 			break;
 
+		netif_napi_set_irq_locked(&bp->bnapi[i]->napi, irq->vector);
 		irq->requested = 1;
+
+		if (zalloc_cpumask_var(&irq->cpu_mask, GFP_KERNEL)) {
+			int numa_node = dev_to_node(&bp->pdev->dev);
+			u16 tag;
+
+			irq->have_cpumask = 1;
+			irq->msix_nr = map_idx;
+			irq->ring_nr = i;
+			cpumask_set_cpu(cpumask_local_spread(i, numa_node),
+					irq->cpu_mask);
+			rc = irq_update_affinity_hint(irq->vector, irq->cpu_mask);
+			if (rc) {
+				netdev_warn(bp->dev,
+					    "Update affinity hint failed, IRQ = %d\n",
+					    irq->vector);
+				break;
+			}
+
+			bnxt_register_irq_notifier(bp, irq);
+
+			/* Init ST table entry */
+			if (pcie_tph_get_cpu_st(irq->bp->pdev, TPH_MEM_TYPE_VM,
+						cpumask_first(irq->cpu_mask),
+						&tag))
+				continue;
+
+			pcie_tph_set_st_entry(irq->bp->pdev, irq->msix_nr, tag);
+		}
 	}
 	return rc;
 }
@@ -5448,13 +11731,17 @@ static void bnxt_del_napi(struct bnxt *bp)
 	if (!bp->bnapi)
 		return;
 
+	for (i = 0; i < bp->rx_nr_rings; i++)
+		netif_queue_set_napi(bp->dev, i, NETDEV_QUEUE_TYPE_RX, NULL);
+	for (i = 0; i < bp->tx_nr_rings - bp->tx_nr_rings_xdp; i++)
+		netif_queue_set_napi(bp->dev, i, NETDEV_QUEUE_TYPE_TX, NULL);
+
 	for (i = 0; i < bp->cp_nr_rings; i++) {
 		struct bnxt_napi *bnapi = bp->bnapi[i];
 
-		napi_hash_del(&bnapi->napi);
-		netif_napi_del(&bnapi->napi);
+		__netif_napi_del_locked(&bnapi->napi);
 	}
-	/* We called napi_hash_del() before netif_napi_del(), we need
+	/* We called __netif_napi_del_locked(), we need
 	 * to respect an RCU grace period before freeing napi structures.
 	 */
 	synchronize_net();
@@ -5462,26 +11749,26 @@ static void bnxt_del_napi(struct bnxt *bp)
 
 static void bnxt_init_napi(struct bnxt *bp)
 {
-	int i;
+	int (*poll_fn)(struct napi_struct *, int) = bnxt_poll;
 	unsigned int cp_nr_rings = bp->cp_nr_rings;
 	struct bnxt_napi *bnapi;
+	int i;
 
-	if (bp->flags & BNXT_FLAG_USING_MSIX) {
-		if (BNXT_CHIP_TYPE_NITRO_A0(bp))
-			cp_nr_rings--;
-		for (i = 0; i < cp_nr_rings; i++) {
-			bnapi = bp->bnapi[i];
-			netif_napi_add(bp->dev, &bnapi->napi,
-				       bnxt_poll, 64);
-		}
-		if (BNXT_CHIP_TYPE_NITRO_A0(bp)) {
-			bnapi = bp->bnapi[cp_nr_rings];
-			netif_napi_add(bp->dev, &bnapi->napi,
-				       bnxt_poll_nitroa0, 64);
-		}
-	} else {
-		bnapi = bp->bnapi[0];
-		netif_napi_add(bp->dev, &bnapi->napi, bnxt_poll, 64);
+	if (bp->flags & BNXT_FLAG_CHIP_P5_PLUS)
+		poll_fn = bnxt_poll_p5;
+	else if (BNXT_CHIP_TYPE_NITRO_A0(bp))
+		cp_nr_rings--;
+
+	set_bit(BNXT_STATE_NAPI_DISABLED, &bp->state);
+
+	for (i = 0; i < cp_nr_rings; i++) {
+		bnapi = bp->bnapi[i];
+		netif_napi_add_config_locked(bp->dev, &bnapi->napi, poll_fn,
+					     bnapi->index);
+	}
+	if (BNXT_CHIP_TYPE_NITRO_A0(bp)) {
+		bnapi = bp->bnapi[cp_nr_rings];
+		netif_napi_add_locked(bp->dev, &bnapi->napi, bnxt_poll_nitroa0);
 	}
 }
 
@@ -5489,20 +11776,42 @@ static void bnxt_disable_napi(struct bnxt *bp)
 {
 	int i;
 
-	if (!bp->bnapi)
+	if (!bp->bnapi ||
+	    test_and_set_bit(BNXT_STATE_NAPI_DISABLED, &bp->state))
 		return;
 
-	for (i = 0; i < bp->cp_nr_rings; i++)
-		napi_disable(&bp->bnapi[i]->napi);
+	for (i = 0; i < bp->cp_nr_rings; i++) {
+		struct bnxt_napi *bnapi = bp->bnapi[i];
+		struct bnxt_cp_ring_info *cpr;
+
+		cpr = &bnapi->cp_ring;
+		if (bnapi->tx_fault)
+			cpr->sw_stats->tx.tx_resets++;
+		if (bnapi->in_reset)
+			cpr->sw_stats->rx.rx_resets++;
+		napi_disable_locked(&bnapi->napi);
+	}
 }
 
 static void bnxt_enable_napi(struct bnxt *bp)
 {
 	int i;
 
+	clear_bit(BNXT_STATE_NAPI_DISABLED, &bp->state);
 	for (i = 0; i < bp->cp_nr_rings; i++) {
-		bp->bnapi[i]->in_reset = false;
-		napi_enable(&bp->bnapi[i]->napi);
+		struct bnxt_napi *bnapi = bp->bnapi[i];
+		struct bnxt_cp_ring_info *cpr;
+
+		bnapi->tx_fault = 0;
+
+		cpr = &bnapi->cp_ring;
+		bnapi->in_reset = false;
+
+		if (bnapi->rx_ring) {
+			INIT_WORK(&cpr->dim.work, bnxt_dim_work);
+			cpr->dim.mode = DIM_CQ_PERIOD_MODE_START_FROM_EQE;
+		}
+		napi_enable_locked(&bnapi->napi);
 	}
 }
 
@@ -5510,45 +11819,76 @@ void bnxt_tx_disable(struct bnxt *bp)
 {
 	int i;
 	struct bnxt_tx_ring_info *txr;
-	struct netdev_queue *txq;
 
 	if (bp->tx_ring) {
 		for (i = 0; i < bp->tx_nr_rings; i++) {
 			txr = &bp->tx_ring[i];
-			txq = netdev_get_tx_queue(bp->dev, i);
-			txr->dev_state = BNXT_DEV_STATE_CLOSING;
+			WRITE_ONCE(txr->dev_state, BNXT_DEV_STATE_CLOSING);
 		}
 	}
+	/* Make sure napi polls see @dev_state change */
+	synchronize_net();
+	/* Drop carrier first to prevent TX timeout */
+	netif_carrier_off(bp->dev);
 	/* Stop all TX queues */
 	netif_tx_disable(bp->dev);
-	netif_carrier_off(bp->dev);
 }
 
 void bnxt_tx_enable(struct bnxt *bp)
 {
 	int i;
 	struct bnxt_tx_ring_info *txr;
-	struct netdev_queue *txq;
 
 	for (i = 0; i < bp->tx_nr_rings; i++) {
 		txr = &bp->tx_ring[i];
-		txq = netdev_get_tx_queue(bp->dev, i);
-		txr->dev_state = 0;
+		WRITE_ONCE(txr->dev_state, 0);
 	}
+	/* Make sure napi polls see @dev_state change */
+	synchronize_net();
 	netif_tx_wake_all_queues(bp->dev);
-	if (bp->link_info.link_up)
+	if (BNXT_LINK_IS_UP(bp))
 		netif_carrier_on(bp->dev);
 }
 
-static void bnxt_report_link(struct bnxt *bp)
+static char *bnxt_report_fec(struct bnxt_link_info *link_info)
 {
-	if (bp->link_info.link_up) {
-		const char *duplex;
+	u8 active_fec = link_info->active_fec_sig_mode &
+			PORT_PHY_QCFG_RESP_ACTIVE_FEC_MASK;
+
+	switch (active_fec) {
+	default:
+	case PORT_PHY_QCFG_RESP_ACTIVE_FEC_FEC_NONE_ACTIVE:
+		return "None";
+	case PORT_PHY_QCFG_RESP_ACTIVE_FEC_FEC_CLAUSE74_ACTIVE:
+		return "Clause 74 BaseR";
+	case PORT_PHY_QCFG_RESP_ACTIVE_FEC_FEC_CLAUSE91_ACTIVE:
+		return "Clause 91 RS(528,514)";
+	case PORT_PHY_QCFG_RESP_ACTIVE_FEC_FEC_RS544_1XN_ACTIVE:
+		return "Clause 91 RS544_1XN";
+	case PORT_PHY_QCFG_RESP_ACTIVE_FEC_FEC_RS544_IEEE_ACTIVE:
+		return "Clause 91 RS(544,514)";
+	case PORT_PHY_QCFG_RESP_ACTIVE_FEC_FEC_RS272_1XN_ACTIVE:
+		return "Clause 91 RS272_1XN";
+	case PORT_PHY_QCFG_RESP_ACTIVE_FEC_FEC_RS272_IEEE_ACTIVE:
+		return "Clause 91 RS(272,257)";
+	}
+}
+
+void bnxt_report_link(struct bnxt *bp)
+{
+	if (BNXT_LINK_IS_UP(bp)) {
+		const char *signal = "";
 		const char *flow_ctrl;
+		const char *duplex;
 		u32 speed;
 		u16 fec;
 
 		netif_carrier_on(bp->dev);
+		speed = bnxt_fw_to_ethtool_speed(bp->link_info.link_speed);
+		if (speed == SPEED_UNKNOWN) {
+			netdev_info(bp->dev, "NIC Link is Up, speed unknown\n");
+			return;
+		}
 		if (bp->link_info.duplex == BNXT_LINK_DUPLEX_FULL)
 			duplex = "full";
 		else
@@ -5561,98 +11901,227 @@ static void bnxt_report_link(struct bnxt *bp)
 			flow_ctrl = "ON - receive";
 		else
 			flow_ctrl = "none";
-		speed = bnxt_fw_to_ethtool_speed(bp->link_info.link_speed);
-		netdev_info(bp->dev, "NIC Link is Up, %u Mbps %s duplex, Flow control: %s\n",
-			    speed, duplex, flow_ctrl);
-		if (bp->flags & BNXT_FLAG_EEE_CAP)
+		if (bp->link_info.phy_qcfg_resp.option_flags &
+		    PORT_PHY_QCFG_RESP_OPTION_FLAGS_SIGNAL_MODE_KNOWN) {
+			u8 sig_mode = bp->link_info.active_fec_sig_mode &
+				      PORT_PHY_QCFG_RESP_SIGNAL_MODE_MASK;
+			switch (sig_mode) {
+			case PORT_PHY_QCFG_RESP_SIGNAL_MODE_NRZ:
+				signal = "(NRZ) ";
+				break;
+			case PORT_PHY_QCFG_RESP_SIGNAL_MODE_PAM4:
+				signal = "(PAM4 56Gbps) ";
+				break;
+			case PORT_PHY_QCFG_RESP_SIGNAL_MODE_PAM4_112:
+				signal = "(PAM4 112Gbps) ";
+				break;
+			default:
+				break;
+			}
+		}
+		netdev_info(bp->dev, "NIC Link is Up, %u Mbps %s%s duplex, Flow control: %s\n",
+			    speed, signal, duplex, flow_ctrl);
+		if (bp->phy_flags & BNXT_PHY_FL_EEE_CAP)
 			netdev_info(bp->dev, "EEE is %s\n",
 				    bp->eee.eee_active ? "active" :
 							 "not active");
 		fec = bp->link_info.fec_cfg;
 		if (!(fec & PORT_PHY_QCFG_RESP_FEC_CFG_FEC_NONE_SUPPORTED))
-			netdev_info(bp->dev, "FEC autoneg %s encodings: %s\n",
+			netdev_info(bp->dev, "FEC autoneg %s encoding: %s\n",
 				    (fec & BNXT_FEC_AUTONEG) ? "on" : "off",
-				    (fec & BNXT_FEC_ENC_BASE_R) ? "BaseR" :
-				     (fec & BNXT_FEC_ENC_RS) ? "RS" : "None");
+				    bnxt_report_fec(&bp->link_info));
 	} else {
 		netif_carrier_off(bp->dev);
 		netdev_err(bp->dev, "NIC Link is Down\n");
 	}
 }
 
+static bool bnxt_phy_qcaps_no_speed(struct hwrm_port_phy_qcaps_output *resp)
+{
+	if (!resp->supported_speeds_auto_mode &&
+	    !resp->supported_speeds_force_mode &&
+	    !resp->supported_pam4_speeds_auto_mode &&
+	    !resp->supported_pam4_speeds_force_mode &&
+	    !resp->supported_speeds2_auto_mode &&
+	    !resp->supported_speeds2_force_mode)
+		return true;
+	return false;
+}
+
 static int bnxt_hwrm_phy_qcaps(struct bnxt *bp)
 {
-	int rc = 0;
-	struct hwrm_port_phy_qcaps_input req = {0};
-	struct hwrm_port_phy_qcaps_output *resp = bp->hwrm_cmd_resp_addr;
 	struct bnxt_link_info *link_info = &bp->link_info;
+	struct hwrm_port_phy_qcaps_output *resp;
+	struct hwrm_port_phy_qcaps_input *req;
+	int rc = 0;
 
 	if (bp->hwrm_spec_code < 0x10201)
 		return 0;
 
-	bnxt_hwrm_cmd_hdr_init(bp, &req, HWRM_PORT_PHY_QCAPS, -1, -1);
+	rc = hwrm_req_init(bp, req, HWRM_PORT_PHY_QCAPS);
+	if (rc)
+		return rc;
 
-	mutex_lock(&bp->hwrm_cmd_lock);
-	rc = _hwrm_send_message(bp, &req, sizeof(req), HWRM_CMD_TIMEOUT);
+	resp = hwrm_req_hold(bp, req);
+	rc = hwrm_req_send(bp, req);
 	if (rc)
 		goto hwrm_phy_qcaps_exit;
 
-	if (resp->eee_supported & PORT_PHY_QCAPS_RESP_EEE_SUPPORTED) {
-		struct ethtool_eee *eee = &bp->eee;
+	bp->phy_flags = resp->flags | (le16_to_cpu(resp->flags2) << 8);
+	if (resp->flags & PORT_PHY_QCAPS_RESP_FLAGS_EEE_SUPPORTED) {
+		struct ethtool_keee *eee = &bp->eee;
 		u16 fw_speeds = le16_to_cpu(resp->supported_speeds_eee_mode);
 
-		bp->flags |= BNXT_FLAG_EEE_CAP;
-		eee->supported = _bnxt_fw_to_ethtool_adv_spds(fw_speeds, 0);
+		_bnxt_fw_to_linkmode(eee->supported, fw_speeds);
 		bp->lpi_tmr_lo = le32_to_cpu(resp->tx_lpi_timer_low) &
 				 PORT_PHY_QCAPS_RESP_TX_LPI_TIMER_LOW_MASK;
 		bp->lpi_tmr_hi = le32_to_cpu(resp->valid_tx_lpi_timer_high) &
 				 PORT_PHY_QCAPS_RESP_TX_LPI_TIMER_HIGH_MASK;
 	}
+
+	if (bp->hwrm_spec_code >= 0x10a01) {
+		if (bnxt_phy_qcaps_no_speed(resp)) {
+			link_info->phy_state = BNXT_PHY_STATE_DISABLED;
+			netdev_warn(bp->dev, "Ethernet link disabled\n");
+		} else if (link_info->phy_state == BNXT_PHY_STATE_DISABLED) {
+			link_info->phy_state = BNXT_PHY_STATE_ENABLED;
+			netdev_info(bp->dev, "Ethernet link enabled\n");
+			/* Phy re-enabled, reprobe the speeds */
+			link_info->support_auto_speeds = 0;
+			link_info->support_pam4_auto_speeds = 0;
+			link_info->support_auto_speeds2 = 0;
+		}
+	}
 	if (resp->supported_speeds_auto_mode)
 		link_info->support_auto_speeds =
 			le16_to_cpu(resp->supported_speeds_auto_mode);
+	if (resp->supported_pam4_speeds_auto_mode)
+		link_info->support_pam4_auto_speeds =
+			le16_to_cpu(resp->supported_pam4_speeds_auto_mode);
+	if (resp->supported_speeds2_auto_mode)
+		link_info->support_auto_speeds2 =
+			le16_to_cpu(resp->supported_speeds2_auto_mode);
+
+	bp->port_count = resp->port_cnt;
 
 hwrm_phy_qcaps_exit:
-	mutex_unlock(&bp->hwrm_cmd_lock);
+	hwrm_req_drop(bp, req);
 	return rc;
 }
 
-static int bnxt_update_link(struct bnxt *bp, bool chng_link_state)
+static void bnxt_hwrm_mac_qcaps(struct bnxt *bp)
+{
+	struct hwrm_port_mac_qcaps_output *resp;
+	struct hwrm_port_mac_qcaps_input *req;
+	int rc;
+
+	if (bp->hwrm_spec_code < 0x10a03)
+		return;
+
+	rc = hwrm_req_init(bp, req, HWRM_PORT_MAC_QCAPS);
+	if (rc)
+		return;
+
+	resp = hwrm_req_hold(bp, req);
+	rc = hwrm_req_send_silent(bp, req);
+	if (!rc)
+		bp->mac_flags = resp->flags;
+	hwrm_req_drop(bp, req);
+}
+
+static bool bnxt_support_dropped(u16 advertising, u16 supported)
+{
+	u16 diff = advertising ^ supported;
+
+	return ((supported | diff) != supported);
+}
+
+static bool bnxt_support_speed_dropped(struct bnxt_link_info *link_info)
+{
+	struct bnxt *bp = container_of(link_info, struct bnxt, link_info);
+
+	/* Check if any advertised speeds are no longer supported. The caller
+	 * holds the link_lock mutex, so we can modify link_info settings.
+	 */
+	if (bp->phy_flags & BNXT_PHY_FL_SPEEDS2) {
+		if (bnxt_support_dropped(link_info->advertising,
+					 link_info->support_auto_speeds2)) {
+			link_info->advertising = link_info->support_auto_speeds2;
+			return true;
+		}
+		return false;
+	}
+	if (bnxt_support_dropped(link_info->advertising,
+				 link_info->support_auto_speeds)) {
+		link_info->advertising = link_info->support_auto_speeds;
+		return true;
+	}
+	if (bnxt_support_dropped(link_info->advertising_pam4,
+				 link_info->support_pam4_auto_speeds)) {
+		link_info->advertising_pam4 = link_info->support_pam4_auto_speeds;
+		return true;
+	}
+	return false;
+}
+
+int bnxt_update_link(struct bnxt *bp, bool chng_link_state)
 {
-	int rc = 0;
 	struct bnxt_link_info *link_info = &bp->link_info;
-	struct hwrm_port_phy_qcfg_input req = {0};
-	struct hwrm_port_phy_qcfg_output *resp = bp->hwrm_cmd_resp_addr;
-	u8 link_up = link_info->link_up;
-	u16 diff;
+	struct hwrm_port_phy_qcfg_output *resp;
+	struct hwrm_port_phy_qcfg_input *req;
+	u8 link_state = link_info->link_state;
+	bool support_changed;
+	int rc;
 
-	bnxt_hwrm_cmd_hdr_init(bp, &req, HWRM_PORT_PHY_QCFG, -1, -1);
+	rc = hwrm_req_init(bp, req, HWRM_PORT_PHY_QCFG);
+	if (rc)
+		return rc;
 
-	mutex_lock(&bp->hwrm_cmd_lock);
-	rc = _hwrm_send_message(bp, &req, sizeof(req), HWRM_CMD_TIMEOUT);
+	resp = hwrm_req_hold(bp, req);
+	rc = hwrm_req_send(bp, req);
 	if (rc) {
-		mutex_unlock(&bp->hwrm_cmd_lock);
+		hwrm_req_drop(bp, req);
+		if (BNXT_VF(bp) && rc == -ENODEV) {
+			netdev_warn(bp->dev, "Cannot obtain link state while PF unavailable.\n");
+			rc = 0;
+		}
 		return rc;
 	}
 
 	memcpy(&link_info->phy_qcfg_resp, resp, sizeof(*resp));
 	link_info->phy_link_status = resp->link;
-	link_info->duplex =  resp->duplex;
+	link_info->duplex = resp->duplex_cfg;
+	if (bp->hwrm_spec_code >= 0x10800)
+		link_info->duplex = resp->duplex_state;
 	link_info->pause = resp->pause;
 	link_info->auto_mode = resp->auto_mode;
 	link_info->auto_pause_setting = resp->auto_pause;
 	link_info->lp_pause = resp->link_partner_adv_pause;
 	link_info->force_pause_setting = resp->force_pause;
-	link_info->duplex_setting = resp->duplex;
-	if (link_info->phy_link_status == BNXT_LINK_LINK)
+	link_info->duplex_setting = resp->duplex_cfg;
+	if (link_info->phy_link_status == BNXT_LINK_LINK) {
 		link_info->link_speed = le16_to_cpu(resp->link_speed);
-	else
+		if (bp->phy_flags & BNXT_PHY_FL_SPEEDS2)
+			link_info->active_lanes = resp->active_lanes;
+	} else {
 		link_info->link_speed = 0;
+		link_info->active_lanes = 0;
+	}
 	link_info->force_link_speed = le16_to_cpu(resp->force_link_speed);
+	link_info->force_pam4_link_speed =
+		le16_to_cpu(resp->force_pam4_link_speed);
+	link_info->force_link_speed2 = le16_to_cpu(resp->force_link_speeds2);
 	link_info->support_speeds = le16_to_cpu(resp->support_speeds);
+	link_info->support_pam4_speeds = le16_to_cpu(resp->support_pam4_speeds);
+	link_info->support_speeds2 = le16_to_cpu(resp->support_speeds2);
 	link_info->auto_link_speeds = le16_to_cpu(resp->auto_link_speed_mask);
+	link_info->auto_pam4_link_speeds =
+		le16_to_cpu(resp->auto_pam4_link_speed_mask);
+	link_info->auto_link_speeds2 = le16_to_cpu(resp->auto_link_speeds2);
 	link_info->lp_auto_link_speeds =
 		le16_to_cpu(resp->link_partner_adv_speeds);
+	link_info->lp_auto_pam4_link_speeds =
+		resp->link_partner_pam4_adv_speeds;
 	link_info->preemphasis = le32_to_cpu(resp->preemphasis);
 	link_info->phy_ver[0] = resp->phy_maj;
 	link_info->phy_ver[1] = resp->phy_min;
@@ -5664,8 +12133,8 @@ static int bnxt_update_link(struct bnxt *bp, bool chng_link_state)
 			      PORT_PHY_QCFG_RESP_PHY_ADDR_MASK;
 	link_info->module_status = resp->module_status;
 
-	if (bp->flags & BNXT_FLAG_EEE_CAP) {
-		struct ethtool_eee *eee = &bp->eee;
+	if (bp->phy_flags & BNXT_PHY_FL_EEE_CAP) {
+		struct ethtool_keee *eee = &bp->eee;
 		u16 fw_speeds;
 
 		eee->eee_active = 0;
@@ -5674,8 +12143,7 @@ static int bnxt_update_link(struct bnxt *bp, bool chng_link_state)
 			eee->eee_active = 1;
 			fw_speeds = le16_to_cpu(
 				resp->link_partner_adv_eee_link_speed_mask);
-			eee->lp_advertised =
-				_bnxt_fw_to_ethtool_adv_spds(fw_speeds, 0);
+			_bnxt_fw_to_linkmode(eee->lp_advertised, fw_speeds);
 		}
 
 		/* Pull initial EEE config */
@@ -5685,8 +12153,7 @@ static int bnxt_update_link(struct bnxt *bp, bool chng_link_state)
 				eee->eee_enabled = 1;
 
 			fw_speeds = le16_to_cpu(resp->adv_eee_link_speed_mask);
-			eee->advertised =
-				_bnxt_fw_to_ethtool_adv_spds(fw_speeds, 0);
+			_bnxt_fw_to_linkmode(eee->advertised, fw_speeds);
 
 			if (resp->eee_config_phy_addr &
 			    PORT_PHY_QCFG_RESP_EEE_CONFIG_EEE_TX_LPI) {
@@ -5701,34 +12168,30 @@ static int bnxt_update_link(struct bnxt *bp, bool chng_link_state)
 	}
 
 	link_info->fec_cfg = PORT_PHY_QCFG_RESP_FEC_CFG_FEC_NONE_SUPPORTED;
-	if (bp->hwrm_spec_code >= 0x10504)
+	if (bp->hwrm_spec_code >= 0x10504) {
 		link_info->fec_cfg = le16_to_cpu(resp->fec_cfg);
-
+		link_info->active_fec_sig_mode = resp->active_fec_signal_mode;
+	}
 	/* TODO: need to add more logic to report VF link */
 	if (chng_link_state) {
 		if (link_info->phy_link_status == BNXT_LINK_LINK)
-			link_info->link_up = 1;
+			link_info->link_state = BNXT_LINK_STATE_UP;
 		else
-			link_info->link_up = 0;
-		if (link_up != link_info->link_up)
+			link_info->link_state = BNXT_LINK_STATE_DOWN;
+		if (link_state != link_info->link_state)
 			bnxt_report_link(bp);
 	} else {
-		/* alwasy link down if not require to update link state */
-		link_info->link_up = 0;
+		/* always link down if not require to update link state */
+		link_info->link_state = BNXT_LINK_STATE_DOWN;
 	}
-	mutex_unlock(&bp->hwrm_cmd_lock);
+	hwrm_req_drop(bp, req);
 
-	diff = link_info->support_auto_speeds ^ link_info->advertising;
-	if ((link_info->support_auto_speeds | diff) !=
-	    link_info->support_auto_speeds) {
-		/* An advertised speed is no longer supported, so we need to
-		 * update the advertisement settings.  Caller holds RTNL
-		 * so we can modify link settings.
-		 */
-		link_info->advertising = link_info->support_auto_speeds;
-		if (link_info->autoneg & BNXT_AUTONEG_SPEED)
-			bnxt_hwrm_set_link_setting(bp, true, false);
-	}
+	if (!BNXT_PHY_CFG_ABLE(bp))
+		return 0;
+
+	support_changed = bnxt_support_speed_dropped(link_info);
+	if (support_changed && (link_info->autoneg & BNXT_AUTONEG_SPEED))
+		bnxt_hwrm_set_link_setting(bp, true, false);
 	return 0;
 }
 
@@ -5787,27 +12250,39 @@ bnxt_hwrm_set_pause_common(struct bnxt *bp, struct hwrm_port_phy_cfg_input *req)
 	}
 }
 
-static void bnxt_hwrm_set_link_common(struct bnxt *bp,
-				      struct hwrm_port_phy_cfg_input *req)
+static void bnxt_hwrm_set_link_common(struct bnxt *bp, struct hwrm_port_phy_cfg_input *req)
 {
-	u8 autoneg = bp->link_info.autoneg;
-	u16 fw_link_speed = bp->link_info.req_link_speed;
-	u16 advertising = bp->link_info.advertising;
-
-	if (autoneg & BNXT_AUTONEG_SPEED) {
-		req->auto_mode |=
-			PORT_PHY_CFG_REQ_AUTO_MODE_SPEED_MASK;
-
-		req->enables |= cpu_to_le32(
-			PORT_PHY_CFG_REQ_ENABLES_AUTO_LINK_SPEED_MASK);
-		req->auto_link_speed_mask = cpu_to_le16(advertising);
-
+	if (bp->link_info.autoneg & BNXT_AUTONEG_SPEED) {
+		req->auto_mode |= PORT_PHY_CFG_REQ_AUTO_MODE_SPEED_MASK;
+		if (bp->phy_flags & BNXT_PHY_FL_SPEEDS2) {
+			req->enables |=
+				cpu_to_le32(PORT_PHY_CFG_REQ_ENABLES_AUTO_LINK_SPEEDS2_MASK);
+			req->auto_link_speeds2_mask = cpu_to_le16(bp->link_info.advertising);
+		} else if (bp->link_info.advertising) {
+			req->enables |= cpu_to_le32(PORT_PHY_CFG_REQ_ENABLES_AUTO_LINK_SPEED_MASK);
+			req->auto_link_speed_mask = cpu_to_le16(bp->link_info.advertising);
+		}
+		if (bp->link_info.advertising_pam4) {
+			req->enables |=
+				cpu_to_le32(PORT_PHY_CFG_REQ_ENABLES_AUTO_PAM4_LINK_SPEED_MASK);
+			req->auto_link_pam4_speed_mask =
+				cpu_to_le16(bp->link_info.advertising_pam4);
+		}
 		req->enables |= cpu_to_le32(PORT_PHY_CFG_REQ_ENABLES_AUTO_MODE);
-		req->flags |=
-			cpu_to_le32(PORT_PHY_CFG_REQ_FLAGS_RESTART_AUTONEG);
+		req->flags |= cpu_to_le32(PORT_PHY_CFG_REQ_FLAGS_RESTART_AUTONEG);
 	} else {
-		req->force_link_speed = cpu_to_le16(fw_link_speed);
 		req->flags |= cpu_to_le32(PORT_PHY_CFG_REQ_FLAGS_FORCE);
+		if (bp->phy_flags & BNXT_PHY_FL_SPEEDS2) {
+			req->force_link_speeds2 = cpu_to_le16(bp->link_info.req_link_speed);
+			req->enables |= cpu_to_le32(PORT_PHY_CFG_REQ_ENABLES_FORCE_LINK_SPEEDS2);
+			netif_info(bp, link, bp->dev, "Forcing FW speed2: %d\n",
+				   (u32)bp->link_info.req_link_speed);
+		} else if (bp->link_info.req_signal_mode == BNXT_SIG_MODE_PAM4) {
+			req->force_pam4_link_speed = cpu_to_le16(bp->link_info.req_link_speed);
+			req->enables |= cpu_to_le32(PORT_PHY_CFG_REQ_ENABLES_FORCE_PAM4_LINK_SPEED);
+		} else {
+			req->force_link_speed = cpu_to_le16(bp->link_info.req_link_speed);
+		}
 	}
 
 	/* tell chimp that the setting takes effect immediately */
@@ -5816,18 +12291,20 @@ static void bnxt_hwrm_set_link_common(struct bnxt *bp,
 
 int bnxt_hwrm_set_pause(struct bnxt *bp)
 {
-	struct hwrm_port_phy_cfg_input req = {0};
+	struct hwrm_port_phy_cfg_input *req;
 	int rc;
 
-	bnxt_hwrm_cmd_hdr_init(bp, &req, HWRM_PORT_PHY_CFG, -1, -1);
-	bnxt_hwrm_set_pause_common(bp, &req);
+	rc = hwrm_req_init(bp, req, HWRM_PORT_PHY_CFG);
+	if (rc)
+		return rc;
+
+	bnxt_hwrm_set_pause_common(bp, req);
 
 	if ((bp->link_info.autoneg & BNXT_AUTONEG_FLOW_CTRL) ||
 	    bp->link_info.force_link_chng)
-		bnxt_hwrm_set_link_common(bp, &req);
+		bnxt_hwrm_set_link_common(bp, req);
 
-	mutex_lock(&bp->hwrm_cmd_lock);
-	rc = _hwrm_send_message(bp, &req, sizeof(req), HWRM_CMD_TIMEOUT);
+	rc = hwrm_req_send(bp, req);
 	if (!rc && !(bp->link_info.autoneg & BNXT_AUTONEG_FLOW_CTRL)) {
 		/* since changing of pause setting doesn't trigger any link
 		 * change event, the driver needs to update the current pause
@@ -5840,14 +12317,13 @@ int bnxt_hwrm_set_pause(struct bnxt *bp)
 			bnxt_report_link(bp);
 	}
 	bp->link_info.force_link_chng = false;
-	mutex_unlock(&bp->hwrm_cmd_lock);
 	return rc;
 }
 
 static void bnxt_hwrm_set_eee(struct bnxt *bp,
 			      struct hwrm_port_phy_cfg_input *req)
 {
-	struct ethtool_eee *eee = &bp->eee;
+	struct ethtool_keee *eee = &bp->eee;
 
 	if (eee->eee_enabled) {
 		u16 eee_speeds;
@@ -5869,50 +12345,246 @@ static void bnxt_hwrm_set_eee(struct bnxt *bp,
 
 int bnxt_hwrm_set_link_setting(struct bnxt *bp, bool set_pause, bool set_eee)
 {
-	struct hwrm_port_phy_cfg_input req = {0};
+	struct hwrm_port_phy_cfg_input *req;
+	int rc;
+
+	rc = hwrm_req_init(bp, req, HWRM_PORT_PHY_CFG);
+	if (rc)
+		return rc;
 
-	bnxt_hwrm_cmd_hdr_init(bp, &req, HWRM_PORT_PHY_CFG, -1, -1);
 	if (set_pause)
-		bnxt_hwrm_set_pause_common(bp, &req);
+		bnxt_hwrm_set_pause_common(bp, req);
 
-	bnxt_hwrm_set_link_common(bp, &req);
+	bnxt_hwrm_set_link_common(bp, req);
 
 	if (set_eee)
-		bnxt_hwrm_set_eee(bp, &req);
-	return hwrm_send_message(bp, &req, sizeof(req), HWRM_CMD_TIMEOUT);
+		bnxt_hwrm_set_eee(bp, req);
+	return hwrm_req_send(bp, req);
 }
 
 static int bnxt_hwrm_shutdown_link(struct bnxt *bp)
 {
-	struct hwrm_port_phy_cfg_input req = {0};
+	struct hwrm_port_phy_cfg_input *req;
+	int rc;
 
 	if (!BNXT_SINGLE_PF(bp))
 		return 0;
 
-	if (pci_num_vf(bp->pdev))
+	if (pci_num_vf(bp->pdev) &&
+	    !(bp->phy_flags & BNXT_PHY_FL_FW_MANAGED_LKDN))
 		return 0;
 
-	bnxt_hwrm_cmd_hdr_init(bp, &req, HWRM_PORT_PHY_CFG, -1, -1);
-	req.flags = cpu_to_le32(PORT_PHY_CFG_REQ_FLAGS_FORCE_LINK_DWN);
-	return hwrm_send_message(bp, &req, sizeof(req), HWRM_CMD_TIMEOUT);
+	rc = hwrm_req_init(bp, req, HWRM_PORT_PHY_CFG);
+	if (rc)
+		return rc;
+
+	req->flags = cpu_to_le32(PORT_PHY_CFG_REQ_FLAGS_FORCE_LINK_DWN);
+	rc = hwrm_req_send(bp, req);
+	if (!rc) {
+		mutex_lock(&bp->link_lock);
+		/* Device is not obliged link down in certain scenarios, even
+		 * when forced. Setting the state unknown is consistent with
+		 * driver startup and will force link state to be reported
+		 * during subsequent open based on PORT_PHY_QCFG.
+		 */
+		bp->link_info.link_state = BNXT_LINK_STATE_UNKNOWN;
+		mutex_unlock(&bp->link_lock);
+	}
+	return rc;
+}
+
+static int bnxt_fw_reset_via_optee(struct bnxt *bp)
+{
+#ifdef CONFIG_TEE_BNXT_FW
+	int rc = tee_bnxt_fw_load();
+
+	if (rc)
+		netdev_err(bp->dev, "Failed FW reset via OP-TEE, rc=%d\n", rc);
+
+	return rc;
+#else
+	netdev_err(bp->dev, "OP-TEE not supported\n");
+	return -ENODEV;
+#endif
+}
+
+static int bnxt_try_recover_fw(struct bnxt *bp)
+{
+	if (bp->fw_health && bp->fw_health->status_reliable) {
+		int retry = 0, rc;
+		u32 sts;
+
+		do {
+			sts = bnxt_fw_health_readl(bp, BNXT_FW_HEALTH_REG);
+			rc = bnxt_hwrm_poll(bp);
+			if (!BNXT_FW_IS_BOOTING(sts) &&
+			    !BNXT_FW_IS_RECOVERING(sts))
+				break;
+			retry++;
+		} while (rc == -EBUSY && retry < BNXT_FW_RETRY);
+
+		if (!BNXT_FW_IS_HEALTHY(sts)) {
+			netdev_err(bp->dev,
+				   "Firmware not responding, status: 0x%x\n",
+				   sts);
+			rc = -ENODEV;
+		}
+		if (sts & FW_STATUS_REG_CRASHED_NO_MASTER) {
+			netdev_warn(bp->dev, "Firmware recover via OP-TEE requested\n");
+			return bnxt_fw_reset_via_optee(bp);
+		}
+		return rc;
+	}
+
+	return -ENODEV;
+}
+
+static void bnxt_clear_reservations(struct bnxt *bp, bool fw_reset)
+{
+	struct bnxt_hw_resc *hw_resc = &bp->hw_resc;
+
+	if (!BNXT_NEW_RM(bp))
+		return; /* no resource reservations required */
+
+	hw_resc->resv_cp_rings = 0;
+	hw_resc->resv_stat_ctxs = 0;
+	hw_resc->resv_irqs = 0;
+	hw_resc->resv_tx_rings = 0;
+	hw_resc->resv_rx_rings = 0;
+	hw_resc->resv_hw_ring_grps = 0;
+	hw_resc->resv_vnics = 0;
+	hw_resc->resv_rsscos_ctxs = 0;
+	if (!fw_reset) {
+		bp->tx_nr_rings = 0;
+		bp->rx_nr_rings = 0;
+	}
+}
+
+int bnxt_cancel_reservations(struct bnxt *bp, bool fw_reset)
+{
+	int rc;
+
+	if (!BNXT_NEW_RM(bp))
+		return 0; /* no resource reservations required */
+
+	rc = bnxt_hwrm_func_resc_qcaps(bp, true);
+	if (rc)
+		netdev_err(bp->dev, "resc_qcaps failed\n");
+
+	bnxt_clear_reservations(bp, fw_reset);
+
+	return rc;
+}
+
+static int bnxt_hwrm_if_change(struct bnxt *bp, bool up)
+{
+	struct hwrm_func_drv_if_change_output *resp;
+	struct hwrm_func_drv_if_change_input *req;
+	bool resc_reinit = false;
+	bool caps_change = false;
+	int rc, retry = 0;
+	bool fw_reset;
+	u32 flags = 0;
+
+	fw_reset = (bp->fw_reset_state == BNXT_FW_RESET_STATE_ABORT);
+	bp->fw_reset_state = 0;
+
+	if (!(bp->fw_cap & BNXT_FW_CAP_IF_CHANGE))
+		return 0;
+
+	rc = hwrm_req_init(bp, req, HWRM_FUNC_DRV_IF_CHANGE);
+	if (rc)
+		return rc;
+
+	if (up)
+		req->flags = cpu_to_le32(FUNC_DRV_IF_CHANGE_REQ_FLAGS_UP);
+	resp = hwrm_req_hold(bp, req);
+
+	hwrm_req_flags(bp, req, BNXT_HWRM_FULL_WAIT);
+	while (retry < BNXT_FW_IF_RETRY) {
+		rc = hwrm_req_send(bp, req);
+		if (rc != -EAGAIN)
+			break;
+
+		msleep(50);
+		retry++;
+	}
+
+	if (rc == -EAGAIN) {
+		hwrm_req_drop(bp, req);
+		return rc;
+	} else if (!rc) {
+		flags = le32_to_cpu(resp->flags);
+	} else if (up) {
+		rc = bnxt_try_recover_fw(bp);
+		fw_reset = true;
+	}
+	hwrm_req_drop(bp, req);
+	if (rc)
+		return rc;
+
+	if (!up) {
+		bnxt_inv_fw_health_reg(bp);
+		return 0;
+	}
+
+	if (flags & FUNC_DRV_IF_CHANGE_RESP_FLAGS_RESC_CHANGE)
+		resc_reinit = true;
+	if (flags & FUNC_DRV_IF_CHANGE_RESP_FLAGS_HOT_FW_RESET_DONE ||
+	    test_bit(BNXT_STATE_FW_RESET_DET, &bp->state))
+		fw_reset = true;
+	else
+		bnxt_remap_fw_health_regs(bp);
+
+	if (test_bit(BNXT_STATE_IN_FW_RESET, &bp->state) && !fw_reset) {
+		netdev_err(bp->dev, "RESET_DONE not set during FW reset.\n");
+		set_bit(BNXT_STATE_ABORT_ERR, &bp->state);
+		return -ENODEV;
+	}
+	if (flags & FUNC_DRV_IF_CHANGE_RESP_FLAGS_CAPS_CHANGE)
+		caps_change = true;
+
+	if (resc_reinit || fw_reset || caps_change) {
+		if (fw_reset || caps_change) {
+			set_bit(BNXT_STATE_FW_RESET_DET, &bp->state);
+			if (!test_bit(BNXT_STATE_IN_FW_RESET, &bp->state))
+				bnxt_ulp_irq_stop(bp);
+			bnxt_free_ctx_mem(bp, false);
+			bnxt_dcb_free(bp);
+			rc = bnxt_fw_init_one(bp);
+			if (rc) {
+				clear_bit(BNXT_STATE_FW_RESET_DET, &bp->state);
+				set_bit(BNXT_STATE_ABORT_ERR, &bp->state);
+				return rc;
+			}
+			/* IRQ will be initialized later in bnxt_request_irq()*/
+			bnxt_clear_int_mode(bp);
+		}
+		rc = bnxt_cancel_reservations(bp, fw_reset);
+	}
+	return rc;
 }
 
 static int bnxt_hwrm_port_led_qcaps(struct bnxt *bp)
 {
-	struct hwrm_port_led_qcaps_output *resp = bp->hwrm_cmd_resp_addr;
-	struct hwrm_port_led_qcaps_input req = {0};
+	struct hwrm_port_led_qcaps_output *resp;
+	struct hwrm_port_led_qcaps_input *req;
 	struct bnxt_pf_info *pf = &bp->pf;
 	int rc;
 
+	bp->num_leds = 0;
 	if (BNXT_VF(bp) || bp->hwrm_spec_code < 0x10601)
 		return 0;
 
-	bnxt_hwrm_cmd_hdr_init(bp, &req, HWRM_PORT_LED_QCAPS, -1, -1);
-	req.port_id = cpu_to_le16(pf->port_id);
-	mutex_lock(&bp->hwrm_cmd_lock);
-	rc = _hwrm_send_message(bp, &req, sizeof(req), HWRM_CMD_TIMEOUT);
+	rc = hwrm_req_init(bp, req, HWRM_PORT_LED_QCAPS);
+	if (rc)
+		return rc;
+
+	req->port_id = cpu_to_le16(pf->port_id);
+	resp = hwrm_req_hold(bp, req);
+	rc = hwrm_req_send(bp, req);
 	if (rc) {
-		mutex_unlock(&bp->hwrm_cmd_lock);
+		hwrm_req_drop(bp, req);
 		return rc;
 	}
 	if (resp->num_leds > 0 && resp->num_leds < BNXT_MAX_LED) {
@@ -5932,54 +12604,64 @@ static int bnxt_hwrm_port_led_qcaps(struct bnxt *bp)
 			}
 		}
 	}
-	mutex_unlock(&bp->hwrm_cmd_lock);
+	hwrm_req_drop(bp, req);
 	return 0;
 }
 
 int bnxt_hwrm_alloc_wol_fltr(struct bnxt *bp)
 {
-	struct hwrm_wol_filter_alloc_input req = {0};
-	struct hwrm_wol_filter_alloc_output *resp = bp->hwrm_cmd_resp_addr;
+	struct hwrm_wol_filter_alloc_output *resp;
+	struct hwrm_wol_filter_alloc_input *req;
 	int rc;
 
-	bnxt_hwrm_cmd_hdr_init(bp, &req, HWRM_WOL_FILTER_ALLOC, -1, -1);
-	req.port_id = cpu_to_le16(bp->pf.port_id);
-	req.wol_type = WOL_FILTER_ALLOC_REQ_WOL_TYPE_MAGICPKT;
-	req.enables = cpu_to_le32(WOL_FILTER_ALLOC_REQ_ENABLES_MAC_ADDRESS);
-	memcpy(req.mac_address, bp->dev->dev_addr, ETH_ALEN);
-	mutex_lock(&bp->hwrm_cmd_lock);
-	rc = _hwrm_send_message(bp, &req, sizeof(req), HWRM_CMD_TIMEOUT);
+	rc = hwrm_req_init(bp, req, HWRM_WOL_FILTER_ALLOC);
+	if (rc)
+		return rc;
+
+	req->port_id = cpu_to_le16(bp->pf.port_id);
+	req->wol_type = WOL_FILTER_ALLOC_REQ_WOL_TYPE_MAGICPKT;
+	req->enables = cpu_to_le32(WOL_FILTER_ALLOC_REQ_ENABLES_MAC_ADDRESS);
+	memcpy(req->mac_address, bp->dev->dev_addr, ETH_ALEN);
+
+	resp = hwrm_req_hold(bp, req);
+	rc = hwrm_req_send(bp, req);
 	if (!rc)
 		bp->wol_filter_id = resp->wol_filter_id;
-	mutex_unlock(&bp->hwrm_cmd_lock);
+	hwrm_req_drop(bp, req);
 	return rc;
 }
 
 int bnxt_hwrm_free_wol_fltr(struct bnxt *bp)
 {
-	struct hwrm_wol_filter_free_input req = {0};
+	struct hwrm_wol_filter_free_input *req;
 	int rc;
 
-	bnxt_hwrm_cmd_hdr_init(bp, &req, HWRM_WOL_FILTER_FREE, -1, -1);
-	req.port_id = cpu_to_le16(bp->pf.port_id);
-	req.enables = cpu_to_le32(WOL_FILTER_FREE_REQ_ENABLES_WOL_FILTER_ID);
-	req.wol_filter_id = bp->wol_filter_id;
-	rc = hwrm_send_message(bp, &req, sizeof(req), HWRM_CMD_TIMEOUT);
-	return rc;
+	rc = hwrm_req_init(bp, req, HWRM_WOL_FILTER_FREE);
+	if (rc)
+		return rc;
+
+	req->port_id = cpu_to_le16(bp->pf.port_id);
+	req->enables = cpu_to_le32(WOL_FILTER_FREE_REQ_ENABLES_WOL_FILTER_ID);
+	req->wol_filter_id = bp->wol_filter_id;
+
+	return hwrm_req_send(bp, req);
 }
 
 static u16 bnxt_hwrm_get_wol_fltrs(struct bnxt *bp, u16 handle)
 {
-	struct hwrm_wol_filter_qcfg_input req = {0};
-	struct hwrm_wol_filter_qcfg_output *resp = bp->hwrm_cmd_resp_addr;
+	struct hwrm_wol_filter_qcfg_output *resp;
+	struct hwrm_wol_filter_qcfg_input *req;
 	u16 next_handle = 0;
 	int rc;
 
-	bnxt_hwrm_cmd_hdr_init(bp, &req, HWRM_WOL_FILTER_QCFG, -1, -1);
-	req.port_id = cpu_to_le16(bp->pf.port_id);
-	req.handle = cpu_to_le16(handle);
-	mutex_lock(&bp->hwrm_cmd_lock);
-	rc = _hwrm_send_message(bp, &req, sizeof(req), HWRM_CMD_TIMEOUT);
+	rc = hwrm_req_init(bp, req, HWRM_WOL_FILTER_QCFG);
+	if (rc)
+		return rc;
+
+	req->port_id = cpu_to_le16(bp->pf.port_id);
+	req->handle = cpu_to_le16(handle);
+	resp = hwrm_req_hold(bp, req);
+	rc = hwrm_req_send(bp, req);
 	if (!rc) {
 		next_handle = le16_to_cpu(resp->next_handle);
 		if (next_handle != 0) {
@@ -5990,7 +12672,7 @@ static u16 bnxt_hwrm_get_wol_fltrs(struct bnxt *bp, u16 handle)
 			}
 		}
 	}
-	mutex_unlock(&bp->hwrm_cmd_lock);
+	hwrm_req_drop(bp, req);
 	return next_handle;
 }
 
@@ -5998,6 +12680,7 @@ static void bnxt_get_wol_settings(struct bnxt *bp)
 {
 	u16 handle = 0;
 
+	bp->wol = 0;
 	if (!BNXT_PF(bp) || !(bp->flags & BNXT_FLAG_WOL_CAP))
 		return;
 
@@ -6008,22 +12691,25 @@ static void bnxt_get_wol_settings(struct bnxt *bp)
 
 static bool bnxt_eee_config_ok(struct bnxt *bp)
 {
-	struct ethtool_eee *eee = &bp->eee;
+	struct ethtool_keee *eee = &bp->eee;
 	struct bnxt_link_info *link_info = &bp->link_info;
 
-	if (!(bp->flags & BNXT_FLAG_EEE_CAP))
+	if (!(bp->phy_flags & BNXT_PHY_FL_EEE_CAP))
 		return true;
 
 	if (eee->eee_enabled) {
-		u32 advertising =
-			_bnxt_fw_to_ethtool_adv_spds(link_info->advertising, 0);
+		__ETHTOOL_DECLARE_LINK_MODE_MASK(advertising);
+		__ETHTOOL_DECLARE_LINK_MODE_MASK(tmp);
+
+		_bnxt_fw_to_linkmode(advertising, link_info->advertising);
 
 		if (!(link_info->autoneg & BNXT_AUTONEG_SPEED)) {
 			eee->eee_enabled = 0;
 			return false;
 		}
-		if (eee->advertised & ~advertising) {
-			eee->advertised = advertising & eee->supported;
+		if (linkmode_andnot(tmp, eee->advertised, advertising)) {
+			linkmode_and(eee->advertised, advertising,
+				     eee->supported);
 			return false;
 		}
 	}
@@ -6057,21 +12743,21 @@ static int bnxt_update_phy_setting(struct bnxt *bp)
 	if (!(link_info->autoneg & BNXT_AUTONEG_SPEED)) {
 		if (BNXT_AUTO_MODE(link_info->auto_mode))
 			update_link = true;
-		if (link_info->req_link_speed != link_info->force_link_speed)
+		if (bnxt_force_speed_updated(link_info))
 			update_link = true;
 		if (link_info->req_duplex != link_info->duplex_setting)
 			update_link = true;
 	} else {
 		if (link_info->auto_mode == BNXT_LINK_AUTO_NONE)
 			update_link = true;
-		if (link_info->advertising != link_info->auto_link_speeds)
+		if (bnxt_auto_speed_updated(link_info))
 			update_link = true;
 	}
 
 	/* The last close may have shutdown the link, so need to call
 	 * PHY_CFG to bring it back up.
 	 */
-	if (!netif_carrier_ok(bp->dev))
+	if (!BNXT_LINK_IS_UP(bp))
 		update_link = true;
 
 	if (!bnxt_eee_config_ok(bp))
@@ -6090,41 +12776,141 @@ static int bnxt_update_phy_setting(struct bnxt *bp)
 	return rc;
 }
 
-/* Common routine to pre-map certain register block to different GRC window.
- * A PF has 16 4K windows and a VF has 4 4K windows. However, only 15 windows
- * in PF and 3 windows in VF that can be customized to map in different
- * register blocks.
- */
-static void bnxt_preset_reg_win(struct bnxt *bp)
+static int bnxt_init_dflt_ring_mode(struct bnxt *bp);
+
+static int bnxt_reinit_after_abort(struct bnxt *bp)
 {
-	if (BNXT_PF(bp)) {
-		/* CAG registers map to GRC window #4 */
-		writel(BNXT_CAG_REG_BASE,
-		       bp->bar0 + BNXT_GRCPF_REG_WINDOW_BASE_OUT + 12);
+	int rc;
+
+	if (test_bit(BNXT_STATE_IN_FW_RESET, &bp->state))
+		return -EBUSY;
+
+	if (bp->dev->reg_state == NETREG_UNREGISTERED)
+		return -ENODEV;
+
+	rc = bnxt_fw_init_one(bp);
+	if (!rc) {
+		bnxt_clear_int_mode(bp);
+		rc = bnxt_init_int_mode(bp);
+		if (!rc) {
+			clear_bit(BNXT_STATE_ABORT_ERR, &bp->state);
+			set_bit(BNXT_STATE_FW_RESET_DET, &bp->state);
+		}
+	}
+	return rc;
+}
+
+static void bnxt_cfg_one_usr_fltr(struct bnxt *bp, struct bnxt_filter_base *fltr)
+{
+	struct bnxt_ntuple_filter *ntp_fltr;
+	struct bnxt_l2_filter *l2_fltr;
+
+	if (list_empty(&fltr->list))
+		return;
+
+	if (fltr->type == BNXT_FLTR_TYPE_NTUPLE) {
+		ntp_fltr = container_of(fltr, struct bnxt_ntuple_filter, base);
+		l2_fltr = bp->vnic_info[BNXT_VNIC_DEFAULT].l2_filters[0];
+		atomic_inc(&l2_fltr->refcnt);
+		ntp_fltr->l2_fltr = l2_fltr;
+		if (bnxt_hwrm_cfa_ntuple_filter_alloc(bp, ntp_fltr)) {
+			bnxt_del_ntp_filter(bp, ntp_fltr);
+			netdev_err(bp->dev, "restoring previously configured ntuple filter id %d failed\n",
+				   fltr->sw_id);
+		}
+	} else if (fltr->type == BNXT_FLTR_TYPE_L2) {
+		l2_fltr = container_of(fltr, struct bnxt_l2_filter, base);
+		if (bnxt_hwrm_l2_filter_alloc(bp, l2_fltr)) {
+			bnxt_del_l2_filter(bp, l2_fltr);
+			netdev_err(bp->dev, "restoring previously configured l2 filter id %d failed\n",
+				   fltr->sw_id);
+		}
 	}
 }
 
+static void bnxt_cfg_usr_fltrs(struct bnxt *bp)
+{
+	struct bnxt_filter_base *usr_fltr, *tmp;
+
+	list_for_each_entry_safe(usr_fltr, tmp, &bp->usr_fltr_list, list)
+		bnxt_cfg_one_usr_fltr(bp, usr_fltr);
+}
+
+static int bnxt_set_xps_mapping(struct bnxt *bp)
+{
+	int numa_node = dev_to_node(&bp->pdev->dev);
+	unsigned int q_idx, map_idx, cpu, i;
+	const struct cpumask *cpu_mask_ptr;
+	int nr_cpus = num_online_cpus();
+	cpumask_t *q_map;
+	int rc = 0;
+
+	q_map = kcalloc(bp->tx_nr_rings_per_tc, sizeof(*q_map), GFP_KERNEL);
+	if (!q_map)
+		return -ENOMEM;
+
+	/* Create CPU mask for all TX queues across MQPRIO traffic classes.
+	 * Each TC has the same number of TX queues. The nth TX queue for each
+	 * TC will have the same CPU mask.
+	 */
+	for (i = 0; i < nr_cpus; i++) {
+		map_idx = i % bp->tx_nr_rings_per_tc;
+		cpu = cpumask_local_spread(i, numa_node);
+		cpu_mask_ptr = get_cpu_mask(cpu);
+		cpumask_or(&q_map[map_idx], &q_map[map_idx], cpu_mask_ptr);
+	}
+
+	/* Register CPU mask for each TX queue except the ones marked for XDP */
+	for (q_idx = 0; q_idx < bp->dev->real_num_tx_queues; q_idx++) {
+		map_idx = q_idx % bp->tx_nr_rings_per_tc;
+		rc = netif_set_xps_queue(bp->dev, &q_map[map_idx], q_idx);
+		if (rc) {
+			netdev_warn(bp->dev, "Error setting XPS for q:%d\n",
+				    q_idx);
+			break;
+		}
+	}
+
+	kfree(q_map);
+
+	return rc;
+}
+
+static int bnxt_tx_nr_rings(struct bnxt *bp)
+{
+	return bp->num_tc ? bp->tx_nr_rings_per_tc * bp->num_tc :
+			    bp->tx_nr_rings_per_tc;
+}
+
+static int bnxt_tx_nr_rings_per_tc(struct bnxt *bp)
+{
+	return bp->num_tc ? bp->tx_nr_rings / bp->num_tc : bp->tx_nr_rings;
+}
+
 static int __bnxt_open_nic(struct bnxt *bp, bool irq_re_init, bool link_re_init)
 {
 	int rc = 0;
 
-	bnxt_preset_reg_win(bp);
 	netif_carrier_off(bp->dev);
 	if (irq_re_init) {
-		rc = bnxt_setup_int_mode(bp);
+		/* Reserve rings now if none were reserved at driver probe. */
+		rc = bnxt_init_dflt_ring_mode(bp);
 		if (rc) {
-			netdev_err(bp->dev, "bnxt_setup_int_mode err: %x\n",
-				   rc);
+			netdev_err(bp->dev, "Failed to reserve default rings at open\n");
 			return rc;
 		}
 	}
-	if ((bp->flags & BNXT_FLAG_RFS) &&
-	    !(bp->flags & BNXT_FLAG_USING_MSIX)) {
-		/* disable RFS if falling back to INTA */
-		bp->dev->hw_features &= ~NETIF_F_NTUPLE;
-		bp->flags &= ~BNXT_FLAG_RFS;
-	}
+	rc = bnxt_reserve_rings(bp, irq_re_init);
+	if (rc)
+		return rc;
 
+	/* Make adjustments if reserved TX rings are less than requested */
+	bp->tx_nr_rings -= bp->tx_nr_rings_xdp;
+	bp->tx_nr_rings_per_tc = bnxt_tx_nr_rings_per_tc(bp);
+	if (bp->tx_nr_rings_xdp) {
+		bp->tx_nr_rings_xdp = bp->tx_nr_rings_per_tc;
+		bp->tx_nr_rings += bp->tx_nr_rings_xdp;
+	}
 	rc = bnxt_alloc_mem(bp, irq_re_init);
 	if (rc) {
 		netdev_err(bp->dev, "bnxt_alloc_mem err: %x\n", rc);
@@ -6136,39 +12922,67 @@ static int __bnxt_open_nic(struct bnxt *bp, bool irq_re_init, bool link_re_init)
 		rc = bnxt_request_irq(bp);
 		if (rc) {
 			netdev_err(bp->dev, "bnxt_request_irq err: %x\n", rc);
-			goto open_err;
+			goto open_err_irq;
 		}
 	}
 
-	bnxt_enable_napi(bp);
-
 	rc = bnxt_init_nic(bp, irq_re_init);
 	if (rc) {
 		netdev_err(bp->dev, "bnxt_init_nic err: %x\n", rc);
-		goto open_err;
+		goto open_err_irq;
 	}
 
+	bnxt_enable_napi(bp);
+	bnxt_debug_dev_init(bp);
+
 	if (link_re_init) {
+		mutex_lock(&bp->link_lock);
 		rc = bnxt_update_phy_setting(bp);
-		if (rc)
+		mutex_unlock(&bp->link_lock);
+		if (rc) {
 			netdev_warn(bp->dev, "failed to update phy settings\n");
+			if (BNXT_SINGLE_PF(bp)) {
+				bp->link_info.phy_retry = true;
+				bp->link_info.phy_retry_expires =
+					jiffies + 5 * HZ;
+			}
+		}
 	}
 
-	if (irq_re_init)
-		udp_tunnel_get_rx_info(bp->dev);
+	if (irq_re_init) {
+		udp_tunnel_nic_reset_ntf(bp->dev);
+		rc = bnxt_set_xps_mapping(bp);
+		if (rc)
+			netdev_warn(bp->dev, "failed to set xps mapping\n");
+	}
 
+	if (bp->tx_nr_rings_xdp < num_possible_cpus()) {
+		if (!static_key_enabled(&bnxt_xdp_locking_key))
+			static_branch_enable(&bnxt_xdp_locking_key);
+	} else if (static_key_enabled(&bnxt_xdp_locking_key)) {
+		static_branch_disable(&bnxt_xdp_locking_key);
+	}
 	set_bit(BNXT_STATE_OPEN, &bp->state);
 	bnxt_enable_int(bp);
 	/* Enable TX queues */
 	bnxt_tx_enable(bp);
 	mod_timer(&bp->timer, jiffies + bp->current_interval);
 	/* Poll link status and check for SFP+ module status */
+	mutex_lock(&bp->link_lock);
 	bnxt_get_port_module_status(bp);
+	mutex_unlock(&bp->link_lock);
 
+	/* VF-reps may need to be re-opened after the PF is re-opened */
+	if (BNXT_PF(bp))
+		bnxt_vf_reps_open(bp);
+	bnxt_ptp_init_rtc(bp, true);
+	bnxt_ptp_cfg_tstamp_filters(bp);
+	if (BNXT_SUPPORTS_MULTI_RSS_CTX(bp))
+		bnxt_hwrm_realloc_rss_ctx_vnic(bp);
+	bnxt_cfg_usr_fltrs(bp);
 	return 0;
 
-open_err:
-	bnxt_disable_napi(bp);
+open_err_irq:
 	bnxt_del_napi(bp);
 
 open_err_free_mem:
@@ -6178,34 +12992,46 @@ open_err_free_mem:
 	return rc;
 }
 
-/* rtnl_lock held */
 int bnxt_open_nic(struct bnxt *bp, bool irq_re_init, bool link_re_init)
 {
 	int rc = 0;
 
-	rc = __bnxt_open_nic(bp, irq_re_init, link_re_init);
+	if (test_bit(BNXT_STATE_ABORT_ERR, &bp->state))
+		rc = -EIO;
+	if (!rc)
+		rc = __bnxt_open_nic(bp, irq_re_init, link_re_init);
 	if (rc) {
 		netdev_err(bp->dev, "nic open fail (rc: %x)\n", rc);
-		dev_close(bp->dev);
+		netif_close(bp->dev);
 	}
 	return rc;
 }
 
-/* rtnl_lock held, open the NIC half way by allocating all resources, but
- * NAPI, IRQ, and TX are not enabled.  This is mainly used for offline
- * self tests.
+/* netdev instance lock held, open the NIC half way by allocating all
+ * resources, but NAPI, IRQ, and TX are not enabled.  This is mainly used
+ * for offline self tests.
  */
 int bnxt_half_open_nic(struct bnxt *bp)
 {
 	int rc = 0;
 
-	rc = bnxt_alloc_mem(bp, false);
+	if (test_bit(BNXT_STATE_ABORT_ERR, &bp->state)) {
+		netdev_err(bp->dev, "A previous firmware reset has not completed, aborting half open\n");
+		rc = -ENODEV;
+		goto half_open_err;
+	}
+
+	rc = bnxt_alloc_mem(bp, true);
 	if (rc) {
 		netdev_err(bp->dev, "bnxt_alloc_mem err: %x\n", rc);
 		goto half_open_err;
 	}
-	rc = bnxt_init_nic(bp, false);
+	bnxt_init_napi(bp);
+	set_bit(BNXT_STATE_HALF_OPEN, &bp->state);
+	rc = bnxt_init_nic(bp, true);
 	if (rc) {
+		clear_bit(BNXT_STATE_HALF_OPEN, &bp->state);
+		bnxt_del_napi(bp);
 		netdev_err(bp->dev, "bnxt_init_nic err: %x\n", rc);
 		goto half_open_err;
 	}
@@ -6213,64 +13039,145 @@ int bnxt_half_open_nic(struct bnxt *bp)
 
 half_open_err:
 	bnxt_free_skbs(bp);
-	bnxt_free_mem(bp, false);
-	dev_close(bp->dev);
+	bnxt_free_mem(bp, true);
+	netif_close(bp->dev);
 	return rc;
 }
 
-/* rtnl_lock held, this call can only be made after a previous successful
- * call to bnxt_half_open_nic().
+/* netdev instance lock held, this call can only be made after a previous
+ * successful call to bnxt_half_open_nic().
  */
 void bnxt_half_close_nic(struct bnxt *bp)
 {
-	bnxt_hwrm_resource_free(bp, false, false);
+	bnxt_hwrm_resource_free(bp, false, true);
+	bnxt_del_napi(bp);
 	bnxt_free_skbs(bp);
-	bnxt_free_mem(bp, false);
+	bnxt_free_mem(bp, true);
+	clear_bit(BNXT_STATE_HALF_OPEN, &bp->state);
+}
+
+void bnxt_reenable_sriov(struct bnxt *bp)
+{
+	if (BNXT_PF(bp)) {
+		struct bnxt_pf_info *pf = &bp->pf;
+		int n = pf->active_vfs;
+
+		if (n)
+			bnxt_cfg_hw_sriov(bp, &n, true);
+	}
 }
 
 static int bnxt_open(struct net_device *dev)
 {
 	struct bnxt *bp = netdev_priv(dev);
+	int rc;
+
+	if (test_bit(BNXT_STATE_ABORT_ERR, &bp->state)) {
+		rc = bnxt_reinit_after_abort(bp);
+		if (rc) {
+			if (rc == -EBUSY)
+				netdev_err(bp->dev, "A previous firmware reset has not completed, aborting\n");
+			else
+				netdev_err(bp->dev, "Failed to reinitialize after aborted firmware reset\n");
+			return -ENODEV;
+		}
+	}
+
+	rc = bnxt_hwrm_if_change(bp, true);
+	if (rc)
+		return rc;
+
+	rc = __bnxt_open_nic(bp, true, true);
+	if (rc) {
+		bnxt_hwrm_if_change(bp, false);
+	} else {
+		if (test_and_clear_bit(BNXT_STATE_FW_RESET_DET, &bp->state)) {
+			if (!test_bit(BNXT_STATE_IN_FW_RESET, &bp->state))
+				bnxt_queue_sp_work(bp,
+						   BNXT_RESTART_ULP_SP_EVENT);
+		}
+	}
 
-	return __bnxt_open_nic(bp, true, true);
+	return rc;
 }
 
-int bnxt_close_nic(struct bnxt *bp, bool irq_re_init, bool link_re_init)
+static bool bnxt_drv_busy(struct bnxt *bp)
 {
-	int rc = 0;
+	return (test_bit(BNXT_STATE_IN_SP_TASK, &bp->state) ||
+		test_bit(BNXT_STATE_READ_STATS, &bp->state));
+}
+
+static void bnxt_get_ring_stats(struct bnxt *bp,
+				struct rtnl_link_stats64 *stats);
+
+static void __bnxt_close_nic(struct bnxt *bp, bool irq_re_init,
+			     bool link_re_init)
+{
+	/* Close the VF-reps before closing PF */
+	if (BNXT_PF(bp))
+		bnxt_vf_reps_close(bp);
 
-#ifdef CONFIG_BNXT_SRIOV
-	if (bp->sriov_cfg) {
-		rc = wait_event_interruptible_timeout(bp->sriov_cfg_wait,
-						      !bp->sriov_cfg,
-						      BNXT_SRIOV_CFG_WAIT_TMO);
-		if (rc)
-			netdev_warn(bp->dev, "timeout waiting for SRIOV config operation to complete!\n");
-	}
-#endif
 	/* Change device state to avoid TX queue wake up's */
 	bnxt_tx_disable(bp);
 
 	clear_bit(BNXT_STATE_OPEN, &bp->state);
 	smp_mb__after_atomic();
-	while (test_bit(BNXT_STATE_IN_SP_TASK, &bp->state))
+	while (bnxt_drv_busy(bp))
 		msleep(20);
 
-	/* Flush rings and and disable interrupts */
+	if (BNXT_SUPPORTS_MULTI_RSS_CTX(bp))
+		bnxt_clear_rss_ctxs(bp);
+	/* Flush rings and disable interrupts */
 	bnxt_shutdown_nic(bp, irq_re_init);
 
 	/* TODO CHIMP_FW: Link/PHY related cleanup if (link_re_init) */
 
+	bnxt_debug_dev_exit(bp);
 	bnxt_disable_napi(bp);
-	del_timer_sync(&bp->timer);
+	timer_delete_sync(&bp->timer);
 	bnxt_free_skbs(bp);
 
+	/* Save ring stats before shutdown */
+	if (bp->bnapi && irq_re_init) {
+		bnxt_get_ring_stats(bp, &bp->net_stats_prev);
+		bnxt_get_ring_err_stats(bp, &bp->ring_err_stats_prev);
+	}
 	if (irq_re_init) {
 		bnxt_free_irq(bp);
 		bnxt_del_napi(bp);
 	}
 	bnxt_free_mem(bp, irq_re_init);
-	return rc;
+}
+
+void bnxt_close_nic(struct bnxt *bp, bool irq_re_init, bool link_re_init)
+{
+	if (test_bit(BNXT_STATE_IN_FW_RESET, &bp->state)) {
+		/* If we get here, it means firmware reset is in progress
+		 * while we are trying to close.  We can safely proceed with
+		 * the close because we are holding netdev instance lock.
+		 * Some firmware messages may fail as we proceed to close.
+		 * We set the ABORT_ERR flag here so that the FW reset thread
+		 * will later abort when it gets the netdev instance lock
+		 * and sees the flag.
+		 */
+		netdev_warn(bp->dev, "FW reset in progress during close, FW reset will be aborted\n");
+		set_bit(BNXT_STATE_ABORT_ERR, &bp->state);
+	}
+
+#ifdef CONFIG_BNXT_SRIOV
+	if (bp->sriov_cfg) {
+		int rc;
+
+		rc = wait_event_interruptible_timeout(bp->sriov_cfg_wait,
+						      !bp->sriov_cfg,
+						      BNXT_SRIOV_CFG_WAIT_TMO);
+		if (!rc)
+			netdev_warn(bp->dev, "timeout waiting for SRIOV config operation to complete, proceeding to close!\n");
+		else if (rc < 0)
+			netdev_warn(bp->dev, "SRIOV config operation interrupted, proceeding to close!\n");
+	}
+#endif
+	__bnxt_close_nic(bp, irq_re_init, link_re_init);
 }
 
 static int bnxt_close(struct net_device *dev)
@@ -6279,27 +13186,99 @@ static int bnxt_close(struct net_device *dev)
 
 	bnxt_close_nic(bp, true, true);
 	bnxt_hwrm_shutdown_link(bp);
+	bnxt_hwrm_if_change(bp, false);
 	return 0;
 }
 
-/* rtnl_lock held */
+static int bnxt_hwrm_port_phy_read(struct bnxt *bp, u16 phy_addr, u16 reg,
+				   u16 *val)
+{
+	struct hwrm_port_phy_mdio_read_output *resp;
+	struct hwrm_port_phy_mdio_read_input *req;
+	int rc;
+
+	if (bp->hwrm_spec_code < 0x10a00)
+		return -EOPNOTSUPP;
+
+	rc = hwrm_req_init(bp, req, HWRM_PORT_PHY_MDIO_READ);
+	if (rc)
+		return rc;
+
+	req->port_id = cpu_to_le16(bp->pf.port_id);
+	req->phy_addr = phy_addr;
+	req->reg_addr = cpu_to_le16(reg & 0x1f);
+	if (mdio_phy_id_is_c45(phy_addr)) {
+		req->cl45_mdio = 1;
+		req->phy_addr = mdio_phy_id_prtad(phy_addr);
+		req->dev_addr = mdio_phy_id_devad(phy_addr);
+		req->reg_addr = cpu_to_le16(reg);
+	}
+
+	resp = hwrm_req_hold(bp, req);
+	rc = hwrm_req_send(bp, req);
+	if (!rc)
+		*val = le16_to_cpu(resp->reg_data);
+	hwrm_req_drop(bp, req);
+	return rc;
+}
+
+static int bnxt_hwrm_port_phy_write(struct bnxt *bp, u16 phy_addr, u16 reg,
+				    u16 val)
+{
+	struct hwrm_port_phy_mdio_write_input *req;
+	int rc;
+
+	if (bp->hwrm_spec_code < 0x10a00)
+		return -EOPNOTSUPP;
+
+	rc = hwrm_req_init(bp, req, HWRM_PORT_PHY_MDIO_WRITE);
+	if (rc)
+		return rc;
+
+	req->port_id = cpu_to_le16(bp->pf.port_id);
+	req->phy_addr = phy_addr;
+	req->reg_addr = cpu_to_le16(reg & 0x1f);
+	if (mdio_phy_id_is_c45(phy_addr)) {
+		req->cl45_mdio = 1;
+		req->phy_addr = mdio_phy_id_prtad(phy_addr);
+		req->dev_addr = mdio_phy_id_devad(phy_addr);
+		req->reg_addr = cpu_to_le16(reg);
+	}
+	req->reg_data = cpu_to_le16(val);
+
+	return hwrm_req_send(bp, req);
+}
+
+/* netdev instance lock held */
 static int bnxt_ioctl(struct net_device *dev, struct ifreq *ifr, int cmd)
 {
+	struct mii_ioctl_data *mdio = if_mii(ifr);
+	struct bnxt *bp = netdev_priv(dev);
+	int rc;
+
 	switch (cmd) {
 	case SIOCGMIIPHY:
-		/* fallthru */
+		mdio->phy_id = bp->link_info.phy_addr;
+
+		fallthrough;
 	case SIOCGMIIREG: {
+		u16 mii_regval = 0;
+
 		if (!netif_running(dev))
 			return -EAGAIN;
 
-		return 0;
+		rc = bnxt_hwrm_port_phy_read(bp, mdio->phy_id, mdio->reg_num,
+					     &mii_regval);
+		mdio->val_out = mii_regval;
+		return rc;
 	}
 
 	case SIOCSMIIREG:
 		if (!netif_running(dev))
 			return -EAGAIN;
 
-		return 0;
+		return bnxt_hwrm_port_phy_write(bp, mdio->phy_id, mdio->reg_num,
+						mdio->val_in);
 
 	default:
 		/* do nothing */
@@ -6308,66 +13287,137 @@ static int bnxt_ioctl(struct net_device *dev, struct ifreq *ifr, int cmd)
 	return -EOPNOTSUPP;
 }
 
-static void
-bnxt_get_stats64(struct net_device *dev, struct rtnl_link_stats64 *stats)
+static void bnxt_get_ring_stats(struct bnxt *bp,
+				struct rtnl_link_stats64 *stats)
 {
-	u32 i;
-	struct bnxt *bp = netdev_priv(dev);
-
-	if (!bp->bnapi)
-		return;
+	int i;
 
-	/* TODO check if we need to synchronize with bnxt_close path */
 	for (i = 0; i < bp->cp_nr_rings; i++) {
 		struct bnxt_napi *bnapi = bp->bnapi[i];
 		struct bnxt_cp_ring_info *cpr = &bnapi->cp_ring;
-		struct ctx_hw_stats *hw_stats = cpr->hw_stats;
+		u64 *sw = cpr->stats.sw_stats;
 
-		stats->rx_packets += le64_to_cpu(hw_stats->rx_ucast_pkts);
-		stats->rx_packets += le64_to_cpu(hw_stats->rx_mcast_pkts);
-		stats->rx_packets += le64_to_cpu(hw_stats->rx_bcast_pkts);
+		stats->rx_packets += BNXT_GET_RING_STATS64(sw, rx_ucast_pkts);
+		stats->rx_packets += BNXT_GET_RING_STATS64(sw, rx_mcast_pkts);
+		stats->rx_packets += BNXT_GET_RING_STATS64(sw, rx_bcast_pkts);
 
-		stats->tx_packets += le64_to_cpu(hw_stats->tx_ucast_pkts);
-		stats->tx_packets += le64_to_cpu(hw_stats->tx_mcast_pkts);
-		stats->tx_packets += le64_to_cpu(hw_stats->tx_bcast_pkts);
+		stats->tx_packets += BNXT_GET_RING_STATS64(sw, tx_ucast_pkts);
+		stats->tx_packets += BNXT_GET_RING_STATS64(sw, tx_mcast_pkts);
+		stats->tx_packets += BNXT_GET_RING_STATS64(sw, tx_bcast_pkts);
 
-		stats->rx_bytes += le64_to_cpu(hw_stats->rx_ucast_bytes);
-		stats->rx_bytes += le64_to_cpu(hw_stats->rx_mcast_bytes);
-		stats->rx_bytes += le64_to_cpu(hw_stats->rx_bcast_bytes);
+		stats->rx_bytes += BNXT_GET_RING_STATS64(sw, rx_ucast_bytes);
+		stats->rx_bytes += BNXT_GET_RING_STATS64(sw, rx_mcast_bytes);
+		stats->rx_bytes += BNXT_GET_RING_STATS64(sw, rx_bcast_bytes);
 
-		stats->tx_bytes += le64_to_cpu(hw_stats->tx_ucast_bytes);
-		stats->tx_bytes += le64_to_cpu(hw_stats->tx_mcast_bytes);
-		stats->tx_bytes += le64_to_cpu(hw_stats->tx_bcast_bytes);
+		stats->tx_bytes += BNXT_GET_RING_STATS64(sw, tx_ucast_bytes);
+		stats->tx_bytes += BNXT_GET_RING_STATS64(sw, tx_mcast_bytes);
+		stats->tx_bytes += BNXT_GET_RING_STATS64(sw, tx_bcast_bytes);
 
 		stats->rx_missed_errors +=
-			le64_to_cpu(hw_stats->rx_discard_pkts);
+			BNXT_GET_RING_STATS64(sw, rx_discard_pkts);
+
+		stats->multicast += BNXT_GET_RING_STATS64(sw, rx_mcast_pkts);
 
-		stats->multicast += le64_to_cpu(hw_stats->rx_mcast_pkts);
+		stats->tx_dropped += BNXT_GET_RING_STATS64(sw, tx_error_pkts);
 
-		stats->tx_dropped += le64_to_cpu(hw_stats->tx_drop_pkts);
+		stats->rx_dropped +=
+			cpr->sw_stats->rx.rx_netpoll_discards +
+			cpr->sw_stats->rx.rx_oom_discards;
 	}
+}
 
-	if (bp->flags & BNXT_FLAG_PORT_STATS) {
-		struct rx_port_stats *rx = bp->hw_rx_port_stats;
-		struct tx_port_stats *tx = bp->hw_tx_port_stats;
+static void bnxt_add_prev_stats(struct bnxt *bp,
+				struct rtnl_link_stats64 *stats)
+{
+	struct rtnl_link_stats64 *prev_stats = &bp->net_stats_prev;
+
+	stats->rx_packets += prev_stats->rx_packets;
+	stats->tx_packets += prev_stats->tx_packets;
+	stats->rx_bytes += prev_stats->rx_bytes;
+	stats->tx_bytes += prev_stats->tx_bytes;
+	stats->rx_missed_errors += prev_stats->rx_missed_errors;
+	stats->multicast += prev_stats->multicast;
+	stats->rx_dropped += prev_stats->rx_dropped;
+	stats->tx_dropped += prev_stats->tx_dropped;
+}
+
+static void
+bnxt_get_stats64(struct net_device *dev, struct rtnl_link_stats64 *stats)
+{
+	struct bnxt *bp = netdev_priv(dev);
 
-		stats->rx_crc_errors = le64_to_cpu(rx->rx_fcs_err_frames);
-		stats->rx_frame_errors = le64_to_cpu(rx->rx_align_err_frames);
-		stats->rx_length_errors = le64_to_cpu(rx->rx_undrsz_frames) +
-					  le64_to_cpu(rx->rx_ovrsz_frames) +
-					  le64_to_cpu(rx->rx_runt_frames);
-		stats->rx_errors = le64_to_cpu(rx->rx_false_carrier_frames) +
-				   le64_to_cpu(rx->rx_jbr_frames);
-		stats->collisions = le64_to_cpu(tx->tx_total_collisions);
-		stats->tx_fifo_errors = le64_to_cpu(tx->tx_fifo_underruns);
-		stats->tx_errors = le64_to_cpu(tx->tx_err);
+	set_bit(BNXT_STATE_READ_STATS, &bp->state);
+	/* Make sure bnxt_close_nic() sees that we are reading stats before
+	 * we check the BNXT_STATE_OPEN flag.
+	 */
+	smp_mb__after_atomic();
+	if (!test_bit(BNXT_STATE_OPEN, &bp->state)) {
+		clear_bit(BNXT_STATE_READ_STATS, &bp->state);
+		*stats = bp->net_stats_prev;
+		return;
 	}
+
+	bnxt_get_ring_stats(bp, stats);
+	bnxt_add_prev_stats(bp, stats);
+
+	if (bp->flags & BNXT_FLAG_PORT_STATS) {
+		u64 *rx = bp->port_stats.sw_stats;
+		u64 *tx = bp->port_stats.sw_stats +
+			  BNXT_TX_PORT_STATS_BYTE_OFFSET / 8;
+
+		stats->rx_crc_errors =
+			BNXT_GET_RX_PORT_STATS64(rx, rx_fcs_err_frames);
+		stats->rx_frame_errors =
+			BNXT_GET_RX_PORT_STATS64(rx, rx_align_err_frames);
+		stats->rx_length_errors =
+			BNXT_GET_RX_PORT_STATS64(rx, rx_undrsz_frames) +
+			BNXT_GET_RX_PORT_STATS64(rx, rx_ovrsz_frames) +
+			BNXT_GET_RX_PORT_STATS64(rx, rx_runt_frames);
+		stats->rx_errors =
+			BNXT_GET_RX_PORT_STATS64(rx, rx_false_carrier_frames) +
+			BNXT_GET_RX_PORT_STATS64(rx, rx_jbr_frames);
+		stats->collisions =
+			BNXT_GET_TX_PORT_STATS64(tx, tx_total_collisions);
+		stats->tx_fifo_errors =
+			BNXT_GET_TX_PORT_STATS64(tx, tx_fifo_underruns);
+		stats->tx_errors = BNXT_GET_TX_PORT_STATS64(tx, tx_err);
+	}
+	clear_bit(BNXT_STATE_READ_STATS, &bp->state);
+}
+
+static void bnxt_get_one_ring_err_stats(struct bnxt *bp,
+					struct bnxt_total_ring_err_stats *stats,
+					struct bnxt_cp_ring_info *cpr)
+{
+	struct bnxt_sw_stats *sw_stats = cpr->sw_stats;
+	u64 *hw_stats = cpr->stats.sw_stats;
+
+	stats->rx_total_l4_csum_errors += sw_stats->rx.rx_l4_csum_errors;
+	stats->rx_total_resets += sw_stats->rx.rx_resets;
+	stats->rx_total_buf_errors += sw_stats->rx.rx_buf_errors;
+	stats->rx_total_oom_discards += sw_stats->rx.rx_oom_discards;
+	stats->rx_total_netpoll_discards += sw_stats->rx.rx_netpoll_discards;
+	stats->rx_total_ring_discards +=
+		BNXT_GET_RING_STATS64(hw_stats, rx_discard_pkts);
+	stats->tx_total_resets += sw_stats->tx.tx_resets;
+	stats->tx_total_ring_discards +=
+		BNXT_GET_RING_STATS64(hw_stats, tx_discard_pkts);
+	stats->total_missed_irqs += sw_stats->cmn.missed_irqs;
+}
+
+void bnxt_get_ring_err_stats(struct bnxt *bp,
+			     struct bnxt_total_ring_err_stats *stats)
+{
+	int i;
+
+	for (i = 0; i < bp->cp_nr_rings; i++)
+		bnxt_get_one_ring_err_stats(bp, stats, &bp->bnapi[i]->cp_ring);
 }
 
 static bool bnxt_mc_list_updated(struct bnxt *bp, u32 *rx_mask)
 {
+	struct bnxt_vnic_info *vnic = &bp->vnic_info[BNXT_VNIC_DEFAULT];
 	struct net_device *dev = bp->dev;
-	struct bnxt_vnic_info *vnic = &bp->vnic_info[0];
 	struct netdev_hw_addr *ha;
 	u8 *haddr;
 	int mc_count = 0;
@@ -6401,7 +13451,7 @@ static bool bnxt_mc_list_updated(struct bnxt *bp, u32 *rx_mask)
 static bool bnxt_uc_list_updated(struct bnxt *bp)
 {
 	struct net_device *dev = bp->dev;
-	struct bnxt_vnic_info *vnic = &bp->vnic_info[0];
+	struct bnxt_vnic_info *vnic = &bp->vnic_info[BNXT_VNIC_DEFAULT];
 	struct netdev_hw_addr *ha;
 	int off = 0;
 
@@ -6420,42 +13470,46 @@ static bool bnxt_uc_list_updated(struct bnxt *bp)
 static void bnxt_set_rx_mode(struct net_device *dev)
 {
 	struct bnxt *bp = netdev_priv(dev);
-	struct bnxt_vnic_info *vnic = &bp->vnic_info[0];
-	u32 mask = vnic->rx_mask;
+	struct bnxt_vnic_info *vnic;
 	bool mc_update = false;
 	bool uc_update;
+	u32 mask;
 
-	if (!netif_running(dev))
+	if (!test_bit(BNXT_STATE_OPEN, &bp->state))
 		return;
 
+	vnic = &bp->vnic_info[BNXT_VNIC_DEFAULT];
+	mask = vnic->rx_mask;
 	mask &= ~(CFA_L2_SET_RX_MASK_REQ_MASK_PROMISCUOUS |
 		  CFA_L2_SET_RX_MASK_REQ_MASK_MCAST |
-		  CFA_L2_SET_RX_MASK_REQ_MASK_ALL_MCAST);
+		  CFA_L2_SET_RX_MASK_REQ_MASK_ALL_MCAST |
+		  CFA_L2_SET_RX_MASK_REQ_MASK_BCAST);
 
-	if ((dev->flags & IFF_PROMISC) && bnxt_promisc_ok(bp))
+	if (dev->flags & IFF_PROMISC)
 		mask |= CFA_L2_SET_RX_MASK_REQ_MASK_PROMISCUOUS;
 
 	uc_update = bnxt_uc_list_updated(bp);
 
+	if (dev->flags & IFF_BROADCAST)
+		mask |= CFA_L2_SET_RX_MASK_REQ_MASK_BCAST;
 	if (dev->flags & IFF_ALLMULTI) {
 		mask |= CFA_L2_SET_RX_MASK_REQ_MASK_ALL_MCAST;
 		vnic->mc_list_count = 0;
-	} else {
+	} else if (dev->flags & IFF_MULTICAST) {
 		mc_update = bnxt_mc_list_updated(bp, &mask);
 	}
 
 	if (mask != vnic->rx_mask || uc_update || mc_update) {
 		vnic->rx_mask = mask;
 
-		set_bit(BNXT_RX_MASK_SP_EVENT, &bp->sp_event);
-		schedule_work(&bp->sp_task);
+		bnxt_queue_sp_work(bp, BNXT_RX_MASK_SP_EVENT);
 	}
 }
 
 static int bnxt_cfg_rx_mode(struct bnxt *bp)
 {
 	struct net_device *dev = bp->dev;
-	struct bnxt_vnic_info *vnic = &bp->vnic_info[0];
+	struct bnxt_vnic_info *vnic = &bp->vnic_info[BNXT_VNIC_DEFAULT];
 	struct netdev_hw_addr *ha;
 	int i, off = 0, rc;
 	bool uc_update;
@@ -6467,19 +13521,12 @@ static int bnxt_cfg_rx_mode(struct bnxt *bp)
 	if (!uc_update)
 		goto skip_uc;
 
-	mutex_lock(&bp->hwrm_cmd_lock);
 	for (i = 1; i < vnic->uc_filter_count; i++) {
-		struct hwrm_cfa_l2_filter_free_input req = {0};
-
-		bnxt_hwrm_cmd_hdr_init(bp, &req, HWRM_CFA_L2_FILTER_FREE, -1,
-				       -1);
-
-		req.l2_filter_id = vnic->fw_l2_filter_id[i];
+		struct bnxt_l2_filter *fltr = vnic->l2_filters[i];
 
-		rc = _hwrm_send_message(bp, &req, sizeof(req),
-					HWRM_CMD_TIMEOUT);
+		bnxt_hwrm_l2_filter_free(bp, fltr);
+		bnxt_del_l2_filter(bp, fltr);
 	}
-	mutex_unlock(&bp->hwrm_cmd_lock);
 
 	vnic->uc_filter_count = 1;
 
@@ -6498,146 +13545,235 @@ static int bnxt_cfg_rx_mode(struct bnxt *bp)
 	for (i = 1, off = 0; i < vnic->uc_filter_count; i++, off += ETH_ALEN) {
 		rc = bnxt_hwrm_set_vnic_filter(bp, 0, i, vnic->uc_list + off);
 		if (rc) {
-			netdev_err(bp->dev, "HWRM vnic filter failure rc: %x\n",
-				   rc);
+			if (BNXT_VF(bp) && rc == -ENODEV) {
+				if (!test_and_set_bit(BNXT_STATE_L2_FILTER_RETRY, &bp->state))
+					netdev_warn(bp->dev, "Cannot configure L2 filters while PF is unavailable, will retry\n");
+				else
+					netdev_dbg(bp->dev, "PF still unavailable while configuring L2 filters.\n");
+				rc = 0;
+			} else {
+				netdev_err(bp->dev, "HWRM vnic filter failure rc: %x\n", rc);
+			}
 			vnic->uc_filter_count = i;
 			return rc;
 		}
 	}
+	if (test_and_clear_bit(BNXT_STATE_L2_FILTER_RETRY, &bp->state))
+		netdev_notice(bp->dev, "Retry of L2 filter configuration successful.\n");
 
 skip_uc:
+	if ((vnic->rx_mask & CFA_L2_SET_RX_MASK_REQ_MASK_PROMISCUOUS) &&
+	    !bnxt_promisc_ok(bp))
+		vnic->rx_mask &= ~CFA_L2_SET_RX_MASK_REQ_MASK_PROMISCUOUS;
 	rc = bnxt_hwrm_cfa_l2_set_rx_mask(bp, 0);
+	if (rc && (vnic->rx_mask & CFA_L2_SET_RX_MASK_REQ_MASK_MCAST)) {
+		netdev_info(bp->dev, "Failed setting MC filters rc: %d, turning on ALL_MCAST mode\n",
+			    rc);
+		vnic->rx_mask &= ~CFA_L2_SET_RX_MASK_REQ_MASK_MCAST;
+		vnic->rx_mask |= CFA_L2_SET_RX_MASK_REQ_MASK_ALL_MCAST;
+		vnic->mc_list_count = 0;
+		rc = bnxt_hwrm_cfa_l2_set_rx_mask(bp, 0);
+	}
 	if (rc)
-		netdev_err(bp->dev, "HWRM cfa l2 rx mask failure rc: %x\n",
+		netdev_err(bp->dev, "HWRM cfa l2 rx mask failure rc: %d\n",
 			   rc);
 
 	return rc;
 }
 
+static bool bnxt_can_reserve_rings(struct bnxt *bp)
+{
+#ifdef CONFIG_BNXT_SRIOV
+	if (BNXT_NEW_RM(bp) && BNXT_VF(bp)) {
+		struct bnxt_hw_resc *hw_resc = &bp->hw_resc;
+
+		/* No minimum rings were provisioned by the PF.  Don't
+		 * reserve rings by default when device is down.
+		 */
+		if (hw_resc->min_tx_rings || hw_resc->resv_tx_rings)
+			return true;
+
+		if (!netif_running(bp->dev))
+			return false;
+	}
+#endif
+	return true;
+}
+
 /* If the chip and firmware supports RFS */
 static bool bnxt_rfs_supported(struct bnxt *bp)
 {
+	if (bp->flags & BNXT_FLAG_CHIP_P5_PLUS) {
+		if (bp->fw_cap & BNXT_FW_CAP_CFA_RFS_RING_TBL_IDX_V2)
+			return true;
+		return false;
+	}
+	/* 212 firmware is broken for aRFS */
+	if (BNXT_FW_MAJ(bp) == 212)
+		return false;
 	if (BNXT_PF(bp) && !BNXT_CHIP_TYPE_NITRO_A0(bp))
 		return true;
-	if (bp->flags & BNXT_FLAG_NEW_RSS_CAP)
+	if (bp->rss_cap & BNXT_RSS_CAP_NEW_RSS_CAP)
 		return true;
 	return false;
 }
 
 /* If runtime conditions support RFS */
-static bool bnxt_rfs_capable(struct bnxt *bp)
+bool bnxt_rfs_capable(struct bnxt *bp, bool new_rss_ctx)
 {
-#ifdef CONFIG_RFS_ACCEL
-	int vnics, max_vnics, max_rss_ctxs;
+	struct bnxt_hw_rings hwr = {0};
+	int max_vnics, max_rss_ctxs;
 
-	if (!(bp->flags & BNXT_FLAG_MSIX_CAP))
+	if ((bp->flags & BNXT_FLAG_CHIP_P5_PLUS) &&
+	    !BNXT_SUPPORTS_NTUPLE_VNIC(bp))
+		return bnxt_rfs_supported(bp);
+
+	if (!bnxt_can_reserve_rings(bp) || !bp->rx_nr_rings)
 		return false;
 
-	vnics = 1 + bp->rx_nr_rings;
+	hwr.grp = bp->rx_nr_rings;
+	hwr.vnic = bnxt_get_total_vnics(bp, bp->rx_nr_rings);
+	if (new_rss_ctx)
+		hwr.vnic++;
+	hwr.rss_ctx = bnxt_get_total_rss_ctxs(bp, &hwr);
 	max_vnics = bnxt_get_max_func_vnics(bp);
 	max_rss_ctxs = bnxt_get_max_func_rss_ctxs(bp);
 
-	/* RSS contexts not a limiting factor */
-	if (bp->flags & BNXT_FLAG_NEW_RSS_CAP)
-		max_rss_ctxs = max_vnics;
-	if (vnics > max_vnics || vnics > max_rss_ctxs) {
-		netdev_warn(bp->dev,
-			    "Not enough resources to support NTUPLE filters, enough resources for up to %d rx rings\n",
-			    min(max_rss_ctxs - 1, max_vnics - 1));
+	if (hwr.vnic > max_vnics || hwr.rss_ctx > max_rss_ctxs) {
+		if (bp->rx_nr_rings > 1)
+			netdev_warn(bp->dev,
+				    "Not enough resources to support NTUPLE filters, enough resources for up to %d rx rings\n",
+				    min(max_rss_ctxs - 1, max_vnics - 1));
 		return false;
 	}
 
-	return true;
-#else
+	if (!BNXT_NEW_RM(bp))
+		return true;
+
+	/* Do not reduce VNIC and RSS ctx reservations.  There is a FW
+	 * issue that will mess up the default VNIC if we reduce the
+	 * reservations.
+	 */
+	if (hwr.vnic <= bp->hw_resc.resv_vnics &&
+	    hwr.rss_ctx <= bp->hw_resc.resv_rsscos_ctxs)
+		return true;
+
+	bnxt_hwrm_reserve_rings(bp, &hwr);
+	if (hwr.vnic <= bp->hw_resc.resv_vnics &&
+	    hwr.rss_ctx <= bp->hw_resc.resv_rsscos_ctxs)
+		return true;
+
+	netdev_warn(bp->dev, "Unable to reserve resources to support NTUPLE filters.\n");
+	hwr.vnic = 1;
+	hwr.rss_ctx = 0;
+	bnxt_hwrm_reserve_rings(bp, &hwr);
 	return false;
-#endif
 }
 
 static netdev_features_t bnxt_fix_features(struct net_device *dev,
 					   netdev_features_t features)
 {
 	struct bnxt *bp = netdev_priv(dev);
+	netdev_features_t vlan_features;
 
-	if ((features & NETIF_F_NTUPLE) && !bnxt_rfs_capable(bp))
+	if ((features & NETIF_F_NTUPLE) && !bnxt_rfs_capable(bp, false))
 		features &= ~NETIF_F_NTUPLE;
 
-	/* Both CTAG and STAG VLAN accelaration on the RX side have to be
+	if ((bp->flags & BNXT_FLAG_NO_AGG_RINGS) || bp->xdp_prog)
+		features &= ~(NETIF_F_LRO | NETIF_F_GRO_HW);
+
+	if (!(features & NETIF_F_GRO))
+		features &= ~NETIF_F_GRO_HW;
+
+	if (features & NETIF_F_GRO_HW)
+		features &= ~NETIF_F_LRO;
+
+	/* Both CTAG and STAG VLAN acceleration on the RX side have to be
 	 * turned on or off together.
 	 */
-	if ((features & (NETIF_F_HW_VLAN_CTAG_RX | NETIF_F_HW_VLAN_STAG_RX)) !=
-	    (NETIF_F_HW_VLAN_CTAG_RX | NETIF_F_HW_VLAN_STAG_RX)) {
-		if (dev->features & NETIF_F_HW_VLAN_CTAG_RX)
-			features &= ~(NETIF_F_HW_VLAN_CTAG_RX |
-				      NETIF_F_HW_VLAN_STAG_RX);
-		else
-			features |= NETIF_F_HW_VLAN_CTAG_RX |
-				    NETIF_F_HW_VLAN_STAG_RX;
+	vlan_features = features & BNXT_HW_FEATURE_VLAN_ALL_RX;
+	if (vlan_features != BNXT_HW_FEATURE_VLAN_ALL_RX) {
+		if (dev->features & BNXT_HW_FEATURE_VLAN_ALL_RX)
+			features &= ~BNXT_HW_FEATURE_VLAN_ALL_RX;
+		else if (vlan_features)
+			features |= BNXT_HW_FEATURE_VLAN_ALL_RX;
 	}
 #ifdef CONFIG_BNXT_SRIOV
-	if (BNXT_VF(bp)) {
-		if (bp->vf.vlan) {
-			features &= ~(NETIF_F_HW_VLAN_CTAG_RX |
-				      NETIF_F_HW_VLAN_STAG_RX);
-		}
-	}
+	if (BNXT_VF(bp) && bp->vf.vlan)
+		features &= ~BNXT_HW_FEATURE_VLAN_ALL_RX;
 #endif
 	return features;
 }
 
+static int bnxt_reinit_features(struct bnxt *bp, bool irq_re_init,
+				bool link_re_init, u32 flags, bool update_tpa)
+{
+	bnxt_close_nic(bp, irq_re_init, link_re_init);
+	bp->flags = flags;
+	if (update_tpa)
+		bnxt_set_ring_params(bp);
+	return bnxt_open_nic(bp, irq_re_init, link_re_init);
+}
+
 static int bnxt_set_features(struct net_device *dev, netdev_features_t features)
 {
+	bool update_tpa = false, update_ntuple = false;
 	struct bnxt *bp = netdev_priv(dev);
 	u32 flags = bp->flags;
 	u32 changes;
 	int rc = 0;
 	bool re_init = false;
-	bool update_tpa = false;
 
 	flags &= ~BNXT_FLAG_ALL_CONFIG_FEATS;
-	if ((features & NETIF_F_GRO) && !BNXT_CHIP_TYPE_NITRO_A0(bp))
+	if (features & NETIF_F_GRO_HW)
 		flags |= BNXT_FLAG_GRO;
-	if (features & NETIF_F_LRO)
+	else if (features & NETIF_F_LRO)
 		flags |= BNXT_FLAG_LRO;
 
 	if (bp->flags & BNXT_FLAG_NO_AGG_RINGS)
 		flags &= ~BNXT_FLAG_TPA;
 
-	if (features & NETIF_F_HW_VLAN_CTAG_RX)
+	if (features & BNXT_HW_FEATURE_VLAN_ALL_RX)
 		flags |= BNXT_FLAG_STRIP_VLAN;
 
 	if (features & NETIF_F_NTUPLE)
 		flags |= BNXT_FLAG_RFS;
+	else
+		bnxt_clear_usr_fltrs(bp, true);
 
 	changes = flags ^ bp->flags;
 	if (changes & BNXT_FLAG_TPA) {
 		update_tpa = true;
 		if ((bp->flags & BNXT_FLAG_TPA) == 0 ||
-		    (flags & BNXT_FLAG_TPA) == 0)
+		    (flags & BNXT_FLAG_TPA) == 0 ||
+		    (bp->flags & BNXT_FLAG_CHIP_P5_PLUS))
 			re_init = true;
 	}
 
 	if (changes & ~BNXT_FLAG_TPA)
 		re_init = true;
 
+	if (changes & BNXT_FLAG_RFS)
+		update_ntuple = true;
+
 	if (flags != bp->flags) {
 		u32 old_flags = bp->flags;
 
-		bp->flags = flags;
-
 		if (!test_bit(BNXT_STATE_OPEN, &bp->state)) {
+			bp->flags = flags;
 			if (update_tpa)
 				bnxt_set_ring_params(bp);
 			return rc;
 		}
 
-		if (re_init) {
-			bnxt_close_nic(bp, false, false);
-			if (update_tpa)
-				bnxt_set_ring_params(bp);
+		if (update_ntuple)
+			return bnxt_reinit_features(bp, true, false, flags, update_tpa);
+
+		if (re_init)
+			return bnxt_reinit_features(bp, false, false, flags, update_tpa);
 
-			return bnxt_open_nic(bp, false, false);
-		}
 		if (update_tpa) {
+			bp->flags = flags;
 			rc = bnxt_set_tpa(bp,
 					  (flags & BNXT_FLAG_TPA) ?
 					  true : false);
@@ -6648,17 +13784,224 @@ static int bnxt_set_features(struct net_device *dev, netdev_features_t features)
 	return rc;
 }
 
-static void bnxt_dump_tx_sw_state(struct bnxt_napi *bnapi)
+static bool bnxt_exthdr_check(struct bnxt *bp, struct sk_buff *skb, int nw_off,
+			      u8 **nextp)
 {
-	struct bnxt_tx_ring_info *txr = bnapi->tx_ring;
-	int i = bnapi->index;
+	struct ipv6hdr *ip6h = (struct ipv6hdr *)(skb->data + nw_off);
+	struct hop_jumbo_hdr *jhdr;
+	int hdr_count = 0;
+	u8 *nexthdr;
+	int start;
+
+	/* Check that there are at most 2 IPv6 extension headers, no
+	 * fragment header, and each is <= 64 bytes.
+	 */
+	start = nw_off + sizeof(*ip6h);
+	nexthdr = &ip6h->nexthdr;
+	while (ipv6_ext_hdr(*nexthdr)) {
+		struct ipv6_opt_hdr *hp;
+		int hdrlen;
+
+		if (hdr_count >= 3 || *nexthdr == NEXTHDR_NONE ||
+		    *nexthdr == NEXTHDR_FRAGMENT)
+			return false;
+		hp = __skb_header_pointer(NULL, start, sizeof(*hp), skb->data,
+					  skb_headlen(skb), NULL);
+		if (!hp)
+			return false;
+		if (*nexthdr == NEXTHDR_AUTH)
+			hdrlen = ipv6_authlen(hp);
+		else
+			hdrlen = ipv6_optlen(hp);
 
-	if (!txr)
-		return;
+		if (hdrlen > 64)
+			return false;
+
+		/* The ext header may be a hop-by-hop header inserted for
+		 * big TCP purposes. This will be removed before sending
+		 * from NIC, so do not count it.
+		 */
+		if (*nexthdr == NEXTHDR_HOP) {
+			if (likely(skb->len <= GRO_LEGACY_MAX_SIZE))
+				goto increment_hdr;
+
+			jhdr = (struct hop_jumbo_hdr *)hp;
+			if (jhdr->tlv_type != IPV6_TLV_JUMBO || jhdr->hdrlen != 0 ||
+			    jhdr->nexthdr != IPPROTO_TCP)
+				goto increment_hdr;
 
-	netdev_info(bnapi->bp->dev, "[%d]: tx{fw_ring: %d prod: %x cons: %x}\n",
-		    i, txr->tx_ring_struct.fw_ring_id, txr->tx_prod,
-		    txr->tx_cons);
+			goto next_hdr;
+		}
+increment_hdr:
+		hdr_count++;
+next_hdr:
+		nexthdr = &hp->nexthdr;
+		start += hdrlen;
+	}
+	if (nextp) {
+		/* Caller will check inner protocol */
+		if (skb->encapsulation) {
+			*nextp = nexthdr;
+			return true;
+		}
+		*nextp = NULL;
+	}
+	/* Only support TCP/UDP for non-tunneled ipv6 and inner ipv6 */
+	return *nexthdr == IPPROTO_TCP || *nexthdr == IPPROTO_UDP;
+}
+
+/* For UDP, we can only handle 1 Vxlan port and 1 Geneve port. */
+static bool bnxt_udp_tunl_check(struct bnxt *bp, struct sk_buff *skb)
+{
+	struct udphdr *uh = udp_hdr(skb);
+	__be16 udp_port = uh->dest;
+
+	if (udp_port != bp->vxlan_port && udp_port != bp->nge_port &&
+	    udp_port != bp->vxlan_gpe_port)
+		return false;
+	if (skb->inner_protocol == htons(ETH_P_TEB)) {
+		struct ethhdr *eh = inner_eth_hdr(skb);
+
+		switch (eh->h_proto) {
+		case htons(ETH_P_IP):
+			return true;
+		case htons(ETH_P_IPV6):
+			return bnxt_exthdr_check(bp, skb,
+						 skb_inner_network_offset(skb),
+						 NULL);
+		}
+	} else if (skb->inner_protocol == htons(ETH_P_IP)) {
+		return true;
+	} else if (skb->inner_protocol == htons(ETH_P_IPV6)) {
+		return bnxt_exthdr_check(bp, skb, skb_inner_network_offset(skb),
+					 NULL);
+	}
+	return false;
+}
+
+static bool bnxt_tunl_check(struct bnxt *bp, struct sk_buff *skb, u8 l4_proto)
+{
+	switch (l4_proto) {
+	case IPPROTO_UDP:
+		return bnxt_udp_tunl_check(bp, skb);
+	case IPPROTO_IPIP:
+		return true;
+	case IPPROTO_GRE: {
+		switch (skb->inner_protocol) {
+		default:
+			return false;
+		case htons(ETH_P_IP):
+			return true;
+		case htons(ETH_P_IPV6):
+			fallthrough;
+		}
+	}
+	case IPPROTO_IPV6:
+		/* Check ext headers of inner ipv6 */
+		return bnxt_exthdr_check(bp, skb, skb_inner_network_offset(skb),
+					 NULL);
+	}
+	return false;
+}
+
+static netdev_features_t bnxt_features_check(struct sk_buff *skb,
+					     struct net_device *dev,
+					     netdev_features_t features)
+{
+	struct bnxt *bp = netdev_priv(dev);
+	u8 *l4_proto;
+
+	features = vlan_features_check(skb, features);
+	switch (vlan_get_protocol(skb)) {
+	case htons(ETH_P_IP):
+		if (!skb->encapsulation)
+			return features;
+		l4_proto = &ip_hdr(skb)->protocol;
+		if (bnxt_tunl_check(bp, skb, *l4_proto))
+			return features;
+		break;
+	case htons(ETH_P_IPV6):
+		if (!bnxt_exthdr_check(bp, skb, skb_network_offset(skb),
+				       &l4_proto))
+			break;
+		if (!l4_proto || bnxt_tunl_check(bp, skb, *l4_proto))
+			return features;
+		break;
+	}
+	return features & ~(NETIF_F_CSUM_MASK | NETIF_F_GSO_MASK);
+}
+
+int bnxt_dbg_hwrm_rd_reg(struct bnxt *bp, u32 reg_off, u16 num_words,
+			 u32 *reg_buf)
+{
+	struct hwrm_dbg_read_direct_output *resp;
+	struct hwrm_dbg_read_direct_input *req;
+	__le32 *dbg_reg_buf;
+	dma_addr_t mapping;
+	int rc, i;
+
+	rc = hwrm_req_init(bp, req, HWRM_DBG_READ_DIRECT);
+	if (rc)
+		return rc;
+
+	dbg_reg_buf = hwrm_req_dma_slice(bp, req, num_words * 4,
+					 &mapping);
+	if (!dbg_reg_buf) {
+		rc = -ENOMEM;
+		goto dbg_rd_reg_exit;
+	}
+
+	req->host_dest_addr = cpu_to_le64(mapping);
+
+	resp = hwrm_req_hold(bp, req);
+	req->read_addr = cpu_to_le32(reg_off + CHIMP_REG_VIEW_ADDR);
+	req->read_len32 = cpu_to_le32(num_words);
+
+	rc = hwrm_req_send(bp, req);
+	if (rc || resp->error_code) {
+		rc = -EIO;
+		goto dbg_rd_reg_exit;
+	}
+	for (i = 0; i < num_words; i++)
+		reg_buf[i] = le32_to_cpu(dbg_reg_buf[i]);
+
+dbg_rd_reg_exit:
+	hwrm_req_drop(bp, req);
+	return rc;
+}
+
+static int bnxt_dbg_hwrm_ring_info_get(struct bnxt *bp, u8 ring_type,
+				       u32 ring_id, u32 *prod, u32 *cons)
+{
+	struct hwrm_dbg_ring_info_get_output *resp;
+	struct hwrm_dbg_ring_info_get_input *req;
+	int rc;
+
+	rc = hwrm_req_init(bp, req, HWRM_DBG_RING_INFO_GET);
+	if (rc)
+		return rc;
+
+	req->ring_type = ring_type;
+	req->fw_ring_id = cpu_to_le32(ring_id);
+	resp = hwrm_req_hold(bp, req);
+	rc = hwrm_req_send(bp, req);
+	if (!rc) {
+		*prod = le32_to_cpu(resp->producer_index);
+		*cons = le32_to_cpu(resp->consumer_index);
+	}
+	hwrm_req_drop(bp, req);
+	return rc;
+}
+
+static void bnxt_dump_tx_sw_state(struct bnxt_napi *bnapi)
+{
+	struct bnxt_tx_ring_info *txr;
+	int i = bnapi->index, j;
+
+	bnxt_for_each_napi_tx(j, bnapi, txr)
+		netdev_info(bnapi->bp->dev, "[%d.%d]: tx{fw_ring: %d prod: %x cons: %x}\n",
+			    i, j, txr->tx_ring_struct.fw_ring_id, txr->tx_prod,
+			    txr->tx_cons);
 }
 
 static void bnxt_dump_rx_sw_state(struct bnxt_napi *bnapi)
@@ -6699,94 +14042,463 @@ static void bnxt_dbg_dump_states(struct bnxt *bp)
 	}
 }
 
+static int bnxt_hwrm_rx_ring_reset(struct bnxt *bp, int ring_nr)
+{
+	struct bnxt_rx_ring_info *rxr = &bp->rx_ring[ring_nr];
+	struct hwrm_ring_reset_input *req;
+	struct bnxt_napi *bnapi = rxr->bnapi;
+	struct bnxt_cp_ring_info *cpr;
+	u16 cp_ring_id;
+	int rc;
+
+	rc = hwrm_req_init(bp, req, HWRM_RING_RESET);
+	if (rc)
+		return rc;
+
+	cpr = &bnapi->cp_ring;
+	cp_ring_id = cpr->cp_ring_struct.fw_ring_id;
+	req->cmpl_ring = cpu_to_le16(cp_ring_id);
+	req->ring_type = RING_RESET_REQ_RING_TYPE_RX_RING_GRP;
+	req->ring_id = cpu_to_le16(bp->grp_info[bnapi->index].fw_grp_id);
+	return hwrm_req_send_silent(bp, req);
+}
+
 static void bnxt_reset_task(struct bnxt *bp, bool silent)
 {
 	if (!silent)
 		bnxt_dbg_dump_states(bp);
 	if (netif_running(bp->dev)) {
-		int rc;
-
-		if (!silent)
-			bnxt_ulp_stop(bp);
-		bnxt_close_nic(bp, false, false);
-		rc = bnxt_open_nic(bp, false, false);
-		if (!silent && !rc)
-			bnxt_ulp_start(bp);
+		bnxt_close_nic(bp, !silent, false);
+		bnxt_open_nic(bp, !silent, false);
 	}
 }
 
-static void bnxt_tx_timeout(struct net_device *dev)
+static void bnxt_tx_timeout(struct net_device *dev, unsigned int txqueue)
 {
 	struct bnxt *bp = netdev_priv(dev);
 
 	netdev_err(bp->dev,  "TX timeout detected, starting reset task!\n");
-	set_bit(BNXT_RESET_TASK_SP_EVENT, &bp->sp_event);
-	schedule_work(&bp->sp_task);
+	bnxt_queue_sp_work(bp, BNXT_RESET_TASK_SP_EVENT);
 }
 
-#ifdef CONFIG_NET_POLL_CONTROLLER
-static void bnxt_poll_controller(struct net_device *dev)
+static void bnxt_fw_health_check(struct bnxt *bp)
 {
-	struct bnxt *bp = netdev_priv(dev);
-	int i;
+	struct bnxt_fw_health *fw_health = bp->fw_health;
+	struct pci_dev *pdev = bp->pdev;
+	u32 val;
 
-	for (i = 0; i < bp->cp_nr_rings; i++) {
-		struct bnxt_irq *irq = &bp->irq_tbl[i];
+	if (!fw_health->enabled || test_bit(BNXT_STATE_IN_FW_RESET, &bp->state))
+		return;
+
+	/* Make sure it is enabled before checking the tmr_counter. */
+	smp_rmb();
+	if (fw_health->tmr_counter) {
+		fw_health->tmr_counter--;
+		return;
+	}
 
-		disable_irq(irq->vector);
-		irq->handler(irq->vector, bp->bnapi[i]);
-		enable_irq(irq->vector);
+	val = bnxt_fw_health_readl(bp, BNXT_FW_HEARTBEAT_REG);
+	if (val == fw_health->last_fw_heartbeat && pci_device_is_present(pdev)) {
+		fw_health->arrests++;
+		goto fw_reset;
 	}
+
+	fw_health->last_fw_heartbeat = val;
+
+	val = bnxt_fw_health_readl(bp, BNXT_FW_RESET_CNT_REG);
+	if (val != fw_health->last_fw_reset_cnt && pci_device_is_present(pdev)) {
+		fw_health->discoveries++;
+		goto fw_reset;
+	}
+
+	fw_health->tmr_counter = fw_health->tmr_multiplier;
+	return;
+
+fw_reset:
+	bnxt_queue_sp_work(bp, BNXT_FW_EXCEPTION_SP_EVENT);
 }
-#endif
 
-static void bnxt_timer(unsigned long data)
+static void bnxt_timer(struct timer_list *t)
 {
-	struct bnxt *bp = (struct bnxt *)data;
+	struct bnxt *bp = timer_container_of(bp, t, timer);
 	struct net_device *dev = bp->dev;
 
-	if (!netif_running(dev))
+	if (!netif_running(dev) || !test_bit(BNXT_STATE_OPEN, &bp->state))
 		return;
 
 	if (atomic_read(&bp->intr_sem) != 0)
 		goto bnxt_restart_timer;
 
-	if (bp->link_info.link_up && (bp->flags & BNXT_FLAG_PORT_STATS)) {
-		set_bit(BNXT_PERIODIC_STATS_SP_EVENT, &bp->sp_event);
-		schedule_work(&bp->sp_task);
+	if (bp->fw_cap & BNXT_FW_CAP_ERROR_RECOVERY)
+		bnxt_fw_health_check(bp);
+
+	if (BNXT_LINK_IS_UP(bp) && bp->stats_coal_ticks)
+		bnxt_queue_sp_work(bp, BNXT_PERIODIC_STATS_SP_EVENT);
+
+	if (bnxt_tc_flower_enabled(bp))
+		bnxt_queue_sp_work(bp, BNXT_FLOW_STATS_SP_EVENT);
+
+#ifdef CONFIG_RFS_ACCEL
+	if ((bp->flags & BNXT_FLAG_RFS) && bp->ntp_fltr_count)
+		bnxt_queue_sp_work(bp, BNXT_RX_NTP_FLTR_SP_EVENT);
+#endif /*CONFIG_RFS_ACCEL*/
+
+	if (bp->link_info.phy_retry) {
+		if (time_after(jiffies, bp->link_info.phy_retry_expires)) {
+			bp->link_info.phy_retry = false;
+			netdev_warn(bp->dev, "failed to update phy settings after maximum retries.\n");
+		} else {
+			bnxt_queue_sp_work(bp, BNXT_UPDATE_PHY_SP_EVENT);
+		}
 	}
+
+	if (test_bit(BNXT_STATE_L2_FILTER_RETRY, &bp->state))
+		bnxt_queue_sp_work(bp, BNXT_RX_MASK_SP_EVENT);
+
+	if ((BNXT_CHIP_P5(bp)) && !bp->chip_rev && netif_carrier_ok(dev))
+		bnxt_queue_sp_work(bp, BNXT_RING_COAL_NOW_SP_EVENT);
+
 bnxt_restart_timer:
 	mod_timer(&bp->timer, jiffies + bp->current_interval);
 }
 
-static void bnxt_rtnl_lock_sp(struct bnxt *bp)
+static void bnxt_lock_sp(struct bnxt *bp)
 {
 	/* We are called from bnxt_sp_task which has BNXT_STATE_IN_SP_TASK
 	 * set.  If the device is being closed, bnxt_close() may be holding
-	 * rtnl() and waiting for BNXT_STATE_IN_SP_TASK to clear.  So we
-	 * must clear BNXT_STATE_IN_SP_TASK before holding rtnl().
+	 * netdev instance lock and waiting for BNXT_STATE_IN_SP_TASK to clear.
+	 * So we must clear BNXT_STATE_IN_SP_TASK before holding netdev
+	 * instance lock.
 	 */
 	clear_bit(BNXT_STATE_IN_SP_TASK, &bp->state);
-	rtnl_lock();
+	netdev_lock(bp->dev);
 }
 
-static void bnxt_rtnl_unlock_sp(struct bnxt *bp)
+static void bnxt_unlock_sp(struct bnxt *bp)
 {
 	set_bit(BNXT_STATE_IN_SP_TASK, &bp->state);
-	rtnl_unlock();
+	netdev_unlock(bp->dev);
 }
 
 /* Only called from bnxt_sp_task() */
 static void bnxt_reset(struct bnxt *bp, bool silent)
 {
-	bnxt_rtnl_lock_sp(bp);
+	bnxt_lock_sp(bp);
 	if (test_bit(BNXT_STATE_OPEN, &bp->state))
 		bnxt_reset_task(bp, silent);
-	bnxt_rtnl_unlock_sp(bp);
+	bnxt_unlock_sp(bp);
+}
+
+/* Only called from bnxt_sp_task() */
+static void bnxt_rx_ring_reset(struct bnxt *bp)
+{
+	int i;
+
+	bnxt_lock_sp(bp);
+	if (!test_bit(BNXT_STATE_OPEN, &bp->state)) {
+		bnxt_unlock_sp(bp);
+		return;
+	}
+	/* Disable and flush TPA before resetting the RX ring */
+	if (bp->flags & BNXT_FLAG_TPA)
+		bnxt_set_tpa(bp, false);
+	for (i = 0; i < bp->rx_nr_rings; i++) {
+		struct bnxt_rx_ring_info *rxr = &bp->rx_ring[i];
+		struct bnxt_cp_ring_info *cpr;
+		int rc;
+
+		if (!rxr->bnapi->in_reset)
+			continue;
+
+		rc = bnxt_hwrm_rx_ring_reset(bp, i);
+		if (rc) {
+			if (rc == -EINVAL || rc == -EOPNOTSUPP)
+				netdev_info_once(bp->dev, "RX ring reset not supported by firmware, falling back to global reset\n");
+			else
+				netdev_warn(bp->dev, "RX ring reset failed, rc = %d, falling back to global reset\n",
+					    rc);
+			bnxt_reset_task(bp, true);
+			break;
+		}
+		bnxt_free_one_rx_ring_skbs(bp, rxr);
+		rxr->rx_prod = 0;
+		rxr->rx_agg_prod = 0;
+		rxr->rx_sw_agg_prod = 0;
+		rxr->rx_next_cons = 0;
+		rxr->bnapi->in_reset = false;
+		bnxt_alloc_one_rx_ring(bp, i);
+		cpr = &rxr->bnapi->cp_ring;
+		cpr->sw_stats->rx.rx_resets++;
+		if (bp->flags & BNXT_FLAG_AGG_RINGS)
+			bnxt_db_write(bp, &rxr->rx_agg_db, rxr->rx_agg_prod);
+		bnxt_db_write(bp, &rxr->rx_db, rxr->rx_prod);
+	}
+	if (bp->flags & BNXT_FLAG_TPA)
+		bnxt_set_tpa(bp, true);
+	bnxt_unlock_sp(bp);
+}
+
+static void bnxt_fw_fatal_close(struct bnxt *bp)
+{
+	bnxt_tx_disable(bp);
+	bnxt_disable_napi(bp);
+	bnxt_disable_int_sync(bp);
+	bnxt_free_irq(bp);
+	bnxt_clear_int_mode(bp);
+	pci_disable_device(bp->pdev);
+}
+
+static void bnxt_fw_reset_close(struct bnxt *bp)
+{
+	/* When firmware is in fatal state, quiesce device and disable
+	 * bus master to prevent any potential bad DMAs before freeing
+	 * kernel memory.
+	 */
+	if (test_bit(BNXT_STATE_FW_FATAL_COND, &bp->state)) {
+		u16 val = 0;
+
+		pci_read_config_word(bp->pdev, PCI_SUBSYSTEM_ID, &val);
+		if (val == 0xffff)
+			bp->fw_reset_min_dsecs = 0;
+		bnxt_fw_fatal_close(bp);
+	}
+	__bnxt_close_nic(bp, true, false);
+	bnxt_vf_reps_free(bp);
+	bnxt_clear_int_mode(bp);
+	bnxt_hwrm_func_drv_unrgtr(bp);
+	if (pci_is_enabled(bp->pdev))
+		pci_disable_device(bp->pdev);
+	bnxt_free_ctx_mem(bp, false);
+}
+
+static bool is_bnxt_fw_ok(struct bnxt *bp)
+{
+	struct bnxt_fw_health *fw_health = bp->fw_health;
+	bool no_heartbeat = false, has_reset = false;
+	u32 val;
+
+	val = bnxt_fw_health_readl(bp, BNXT_FW_HEARTBEAT_REG);
+	if (val == fw_health->last_fw_heartbeat)
+		no_heartbeat = true;
+
+	val = bnxt_fw_health_readl(bp, BNXT_FW_RESET_CNT_REG);
+	if (val != fw_health->last_fw_reset_cnt)
+		has_reset = true;
+
+	if (!no_heartbeat && has_reset)
+		return true;
+
+	return false;
+}
+
+/* netdev instance lock is acquired before calling this function */
+static void bnxt_force_fw_reset(struct bnxt *bp)
+{
+	struct bnxt_fw_health *fw_health = bp->fw_health;
+	struct bnxt_ptp_cfg *ptp = bp->ptp_cfg;
+	u32 wait_dsecs;
+
+	if (!test_bit(BNXT_STATE_OPEN, &bp->state) ||
+	    test_bit(BNXT_STATE_IN_FW_RESET, &bp->state))
+		return;
+
+	/* we have to serialize with bnxt_refclk_read()*/
+	if (ptp) {
+		unsigned long flags;
+
+		write_seqlock_irqsave(&ptp->ptp_lock, flags);
+		set_bit(BNXT_STATE_IN_FW_RESET, &bp->state);
+		write_sequnlock_irqrestore(&ptp->ptp_lock, flags);
+	} else {
+		set_bit(BNXT_STATE_IN_FW_RESET, &bp->state);
+	}
+	bnxt_fw_reset_close(bp);
+	wait_dsecs = fw_health->master_func_wait_dsecs;
+	if (fw_health->primary) {
+		if (fw_health->flags & ERROR_RECOVERY_QCFG_RESP_FLAGS_CO_CPU)
+			wait_dsecs = 0;
+		bp->fw_reset_state = BNXT_FW_RESET_STATE_RESET_FW;
+	} else {
+		bp->fw_reset_timestamp = jiffies + wait_dsecs * HZ / 10;
+		wait_dsecs = fw_health->normal_func_wait_dsecs;
+		bp->fw_reset_state = BNXT_FW_RESET_STATE_ENABLE_DEV;
+	}
+
+	bp->fw_reset_min_dsecs = fw_health->post_reset_wait_dsecs;
+	bp->fw_reset_max_dsecs = fw_health->post_reset_max_wait_dsecs;
+	bnxt_queue_fw_reset_work(bp, wait_dsecs * HZ / 10);
+}
+
+void bnxt_fw_exception(struct bnxt *bp)
+{
+	netdev_warn(bp->dev, "Detected firmware fatal condition, initiating reset\n");
+	set_bit(BNXT_STATE_FW_FATAL_COND, &bp->state);
+	bnxt_ulp_stop(bp);
+	bnxt_lock_sp(bp);
+	bnxt_force_fw_reset(bp);
+	bnxt_unlock_sp(bp);
+}
+
+/* Returns the number of registered VFs, or 1 if VF configuration is pending, or
+ * < 0 on error.
+ */
+static int bnxt_get_registered_vfs(struct bnxt *bp)
+{
+#ifdef CONFIG_BNXT_SRIOV
+	int rc;
+
+	if (!BNXT_PF(bp))
+		return 0;
+
+	rc = bnxt_hwrm_func_qcfg(bp);
+	if (rc) {
+		netdev_err(bp->dev, "func_qcfg cmd failed, rc = %d\n", rc);
+		return rc;
+	}
+	if (bp->pf.registered_vfs)
+		return bp->pf.registered_vfs;
+	if (bp->sriov_cfg)
+		return 1;
+#endif
+	return 0;
+}
+
+void bnxt_fw_reset(struct bnxt *bp)
+{
+	bnxt_ulp_stop(bp);
+	bnxt_lock_sp(bp);
+	if (test_bit(BNXT_STATE_OPEN, &bp->state) &&
+	    !test_bit(BNXT_STATE_IN_FW_RESET, &bp->state)) {
+		struct bnxt_ptp_cfg *ptp = bp->ptp_cfg;
+		int n = 0, tmo;
+
+		/* we have to serialize with bnxt_refclk_read()*/
+		if (ptp) {
+			unsigned long flags;
+
+			write_seqlock_irqsave(&ptp->ptp_lock, flags);
+			set_bit(BNXT_STATE_IN_FW_RESET, &bp->state);
+			write_sequnlock_irqrestore(&ptp->ptp_lock, flags);
+		} else {
+			set_bit(BNXT_STATE_IN_FW_RESET, &bp->state);
+		}
+		if (bp->pf.active_vfs &&
+		    !test_bit(BNXT_STATE_FW_FATAL_COND, &bp->state))
+			n = bnxt_get_registered_vfs(bp);
+		if (n < 0) {
+			netdev_err(bp->dev, "Firmware reset aborted, rc = %d\n",
+				   n);
+			clear_bit(BNXT_STATE_IN_FW_RESET, &bp->state);
+			netif_close(bp->dev);
+			goto fw_reset_exit;
+		} else if (n > 0) {
+			u16 vf_tmo_dsecs = n * 10;
+
+			if (bp->fw_reset_max_dsecs < vf_tmo_dsecs)
+				bp->fw_reset_max_dsecs = vf_tmo_dsecs;
+			bp->fw_reset_state =
+				BNXT_FW_RESET_STATE_POLL_VF;
+			bnxt_queue_fw_reset_work(bp, HZ / 10);
+			goto fw_reset_exit;
+		}
+		bnxt_fw_reset_close(bp);
+		if (bp->fw_cap & BNXT_FW_CAP_ERR_RECOVER_RELOAD) {
+			bp->fw_reset_state = BNXT_FW_RESET_STATE_POLL_FW_DOWN;
+			tmo = HZ / 10;
+		} else {
+			bp->fw_reset_state = BNXT_FW_RESET_STATE_ENABLE_DEV;
+			tmo = bp->fw_reset_min_dsecs * HZ / 10;
+		}
+		bnxt_queue_fw_reset_work(bp, tmo);
+	}
+fw_reset_exit:
+	bnxt_unlock_sp(bp);
+}
+
+static void bnxt_chk_missed_irq(struct bnxt *bp)
+{
+	int i;
+
+	if (!(bp->flags & BNXT_FLAG_CHIP_P5_PLUS))
+		return;
+
+	for (i = 0; i < bp->cp_nr_rings; i++) {
+		struct bnxt_napi *bnapi = bp->bnapi[i];
+		struct bnxt_cp_ring_info *cpr;
+		u32 fw_ring_id;
+		int j;
+
+		if (!bnapi)
+			continue;
+
+		cpr = &bnapi->cp_ring;
+		for (j = 0; j < cpr->cp_ring_count; j++) {
+			struct bnxt_cp_ring_info *cpr2 = &cpr->cp_ring_arr[j];
+			u32 val[2];
+
+			if (cpr2->has_more_work || !bnxt_has_work(bp, cpr2))
+				continue;
+
+			if (cpr2->cp_raw_cons != cpr2->last_cp_raw_cons) {
+				cpr2->last_cp_raw_cons = cpr2->cp_raw_cons;
+				continue;
+			}
+			fw_ring_id = cpr2->cp_ring_struct.fw_ring_id;
+			bnxt_dbg_hwrm_ring_info_get(bp,
+				DBG_RING_INFO_GET_REQ_RING_TYPE_L2_CMPL,
+				fw_ring_id, &val[0], &val[1]);
+			cpr->sw_stats->cmn.missed_irqs++;
+		}
+	}
 }
 
 static void bnxt_cfg_ntp_filters(struct bnxt *);
 
+static void bnxt_init_ethtool_link_settings(struct bnxt *bp)
+{
+	struct bnxt_link_info *link_info = &bp->link_info;
+
+	if (BNXT_AUTO_MODE(link_info->auto_mode)) {
+		link_info->autoneg = BNXT_AUTONEG_SPEED;
+		if (bp->hwrm_spec_code >= 0x10201) {
+			if (link_info->auto_pause_setting &
+			    PORT_PHY_CFG_REQ_AUTO_PAUSE_AUTONEG_PAUSE)
+				link_info->autoneg |= BNXT_AUTONEG_FLOW_CTRL;
+		} else {
+			link_info->autoneg |= BNXT_AUTONEG_FLOW_CTRL;
+		}
+		bnxt_set_auto_speed(link_info);
+	} else {
+		bnxt_set_force_speed(link_info);
+		link_info->req_duplex = link_info->duplex_setting;
+	}
+	if (link_info->autoneg & BNXT_AUTONEG_FLOW_CTRL)
+		link_info->req_flow_ctrl =
+			link_info->auto_pause_setting & BNXT_LINK_PAUSE_BOTH;
+	else
+		link_info->req_flow_ctrl = link_info->force_pause_setting;
+}
+
+static void bnxt_fw_echo_reply(struct bnxt *bp)
+{
+	struct bnxt_fw_health *fw_health = bp->fw_health;
+	struct hwrm_func_echo_response_input *req;
+	int rc;
+
+	rc = hwrm_req_init(bp, req, HWRM_FUNC_ECHO_RESPONSE);
+	if (rc)
+		return;
+	req->event_data1 = cpu_to_le32(fw_health->echo_req_data1);
+	req->event_data2 = cpu_to_le32(fw_health->echo_req_data2);
+	hwrm_req_send(bp, req);
+}
+
+static void bnxt_ulp_restart(struct bnxt *bp)
+{
+	bnxt_ulp_stop(bp);
+	bnxt_ulp_start(bp, 0);
+}
+
 static void bnxt_sp_task(struct work_struct *work)
 {
 	struct bnxt *bp = container_of(work, struct bnxt, sp_task);
@@ -6798,6 +14510,11 @@ static void bnxt_sp_task(struct work_struct *work)
 		return;
 	}
 
+	if (test_and_clear_bit(BNXT_RESTART_ULP_SP_EVENT, &bp->sp_event)) {
+		bnxt_ulp_restart(bp);
+		bnxt_reenable_sriov(bp);
+	}
+
 	if (test_and_clear_bit(BNXT_RX_MASK_SP_EVENT, &bp->sp_event))
 		bnxt_cfg_rx_mode(bp);
 
@@ -6805,90 +14522,152 @@ static void bnxt_sp_task(struct work_struct *work)
 		bnxt_cfg_ntp_filters(bp);
 	if (test_and_clear_bit(BNXT_HWRM_EXEC_FWD_REQ_SP_EVENT, &bp->sp_event))
 		bnxt_hwrm_exec_fwd_req(bp);
-	if (test_and_clear_bit(BNXT_VXLAN_ADD_PORT_SP_EVENT, &bp->sp_event)) {
-		bnxt_hwrm_tunnel_dst_port_alloc(
-			bp, bp->vxlan_port,
-			TUNNEL_DST_PORT_FREE_REQ_TUNNEL_TYPE_VXLAN);
-	}
-	if (test_and_clear_bit(BNXT_VXLAN_DEL_PORT_SP_EVENT, &bp->sp_event)) {
-		bnxt_hwrm_tunnel_dst_port_free(
-			bp, TUNNEL_DST_PORT_FREE_REQ_TUNNEL_TYPE_VXLAN);
-	}
-	if (test_and_clear_bit(BNXT_GENEVE_ADD_PORT_SP_EVENT, &bp->sp_event)) {
-		bnxt_hwrm_tunnel_dst_port_alloc(
-			bp, bp->nge_port,
-			TUNNEL_DST_PORT_FREE_REQ_TUNNEL_TYPE_GENEVE);
-	}
-	if (test_and_clear_bit(BNXT_GENEVE_DEL_PORT_SP_EVENT, &bp->sp_event)) {
-		bnxt_hwrm_tunnel_dst_port_free(
-			bp, TUNNEL_DST_PORT_FREE_REQ_TUNNEL_TYPE_GENEVE);
+	if (test_and_clear_bit(BNXT_HWRM_PF_UNLOAD_SP_EVENT, &bp->sp_event))
+		netdev_info(bp->dev, "Receive PF driver unload event!\n");
+	if (test_and_clear_bit(BNXT_PERIODIC_STATS_SP_EVENT, &bp->sp_event)) {
+		bnxt_hwrm_port_qstats(bp, 0);
+		bnxt_hwrm_port_qstats_ext(bp, 0);
+		bnxt_accumulate_all_stats(bp);
 	}
-	if (test_and_clear_bit(BNXT_PERIODIC_STATS_SP_EVENT, &bp->sp_event))
-		bnxt_hwrm_port_qstats(bp);
 
-	/* These functions below will clear BNXT_STATE_IN_SP_TASK.  They
-	 * must be the last functions to be called before exiting.
-	 */
 	if (test_and_clear_bit(BNXT_LINK_CHNG_SP_EVENT, &bp->sp_event)) {
-		int rc = 0;
+		int rc;
 
+		mutex_lock(&bp->link_lock);
 		if (test_and_clear_bit(BNXT_LINK_SPEED_CHNG_SP_EVENT,
 				       &bp->sp_event))
 			bnxt_hwrm_phy_qcaps(bp);
 
-		bnxt_rtnl_lock_sp(bp);
-		if (test_bit(BNXT_STATE_OPEN, &bp->state))
-			rc = bnxt_update_link(bp, true);
-		bnxt_rtnl_unlock_sp(bp);
+		rc = bnxt_update_link(bp, true);
 		if (rc)
 			netdev_err(bp->dev, "SP task can't update link (rc: %x)\n",
 				   rc);
+
+		if (test_and_clear_bit(BNXT_LINK_CFG_CHANGE_SP_EVENT,
+				       &bp->sp_event))
+			bnxt_init_ethtool_link_settings(bp);
+		mutex_unlock(&bp->link_lock);
+	}
+	if (test_and_clear_bit(BNXT_UPDATE_PHY_SP_EVENT, &bp->sp_event)) {
+		int rc;
+
+		mutex_lock(&bp->link_lock);
+		rc = bnxt_update_phy_setting(bp);
+		mutex_unlock(&bp->link_lock);
+		if (rc) {
+			netdev_warn(bp->dev, "update phy settings retry failed\n");
+		} else {
+			bp->link_info.phy_retry = false;
+			netdev_info(bp->dev, "update phy settings retry succeeded\n");
+		}
 	}
 	if (test_and_clear_bit(BNXT_HWRM_PORT_MODULE_SP_EVENT, &bp->sp_event)) {
-		bnxt_rtnl_lock_sp(bp);
-		if (test_bit(BNXT_STATE_OPEN, &bp->state))
-			bnxt_get_port_module_status(bp);
-		bnxt_rtnl_unlock_sp(bp);
+		mutex_lock(&bp->link_lock);
+		bnxt_get_port_module_status(bp);
+		mutex_unlock(&bp->link_lock);
 	}
+
+	if (test_and_clear_bit(BNXT_FLOW_STATS_SP_EVENT, &bp->sp_event))
+		bnxt_tc_flow_stats_work(bp);
+
+	if (test_and_clear_bit(BNXT_RING_COAL_NOW_SP_EVENT, &bp->sp_event))
+		bnxt_chk_missed_irq(bp);
+
+	if (test_and_clear_bit(BNXT_FW_ECHO_REQUEST_SP_EVENT, &bp->sp_event))
+		bnxt_fw_echo_reply(bp);
+
+	if (test_and_clear_bit(BNXT_THERMAL_THRESHOLD_SP_EVENT, &bp->sp_event))
+		bnxt_hwmon_notify_event(bp);
+
+	/* These functions below will clear BNXT_STATE_IN_SP_TASK.  They
+	 * must be the last functions to be called before exiting.
+	 */
 	if (test_and_clear_bit(BNXT_RESET_TASK_SP_EVENT, &bp->sp_event))
 		bnxt_reset(bp, false);
 
 	if (test_and_clear_bit(BNXT_RESET_TASK_SILENT_SP_EVENT, &bp->sp_event))
 		bnxt_reset(bp, true);
 
+	if (test_and_clear_bit(BNXT_RST_RING_SP_EVENT, &bp->sp_event))
+		bnxt_rx_ring_reset(bp);
+
+	if (test_and_clear_bit(BNXT_FW_RESET_NOTIFY_SP_EVENT, &bp->sp_event)) {
+		if (test_bit(BNXT_STATE_FW_FATAL_COND, &bp->state) ||
+		    test_bit(BNXT_STATE_FW_NON_FATAL_COND, &bp->state))
+			bnxt_devlink_health_fw_report(bp);
+		else
+			bnxt_fw_reset(bp);
+	}
+
+	if (test_and_clear_bit(BNXT_FW_EXCEPTION_SP_EVENT, &bp->sp_event)) {
+		if (!is_bnxt_fw_ok(bp))
+			bnxt_devlink_health_fw_report(bp);
+	}
+
 	smp_mb__before_atomic();
 	clear_bit(BNXT_STATE_IN_SP_TASK, &bp->state);
 }
 
-/* Under rtnl_lock */
-int bnxt_reserve_rings(struct bnxt *bp, int tx, int rx, int tcs, int tx_xdp)
+static void _bnxt_get_max_rings(struct bnxt *bp, int *max_rx, int *max_tx,
+				int *max_cp);
+
+/* Under netdev instance lock */
+int bnxt_check_rings(struct bnxt *bp, int tx, int rx, bool sh, int tcs,
+		     int tx_xdp)
 {
-	int max_rx, max_tx, tx_sets = 1;
-	int tx_rings_needed;
-	bool sh = true;
+	int max_rx, max_tx, max_cp, tx_sets = 1, tx_cp;
+	struct bnxt_hw_rings hwr = {0};
+	int rx_rings = rx;
 	int rc;
 
-	if (!(bp->flags & BNXT_FLAG_SHARED_RINGS))
-		sh = false;
-
 	if (tcs)
 		tx_sets = tcs;
 
-	rc = bnxt_get_max_rings(bp, &max_rx, &max_tx, sh);
-	if (rc)
-		return rc;
+	_bnxt_get_max_rings(bp, &max_rx, &max_tx, &max_cp);
 
-	if (max_rx < rx)
+	if (max_rx < rx_rings)
 		return -ENOMEM;
 
-	tx_rings_needed = tx * tx_sets + tx_xdp;
-	if (max_tx < tx_rings_needed)
+	if (bp->flags & BNXT_FLAG_AGG_RINGS)
+		rx_rings <<= 1;
+
+	hwr.rx = rx_rings;
+	hwr.tx = tx * tx_sets + tx_xdp;
+	if (max_tx < hwr.tx)
 		return -ENOMEM;
 
-	if (bnxt_hwrm_reserve_tx_rings(bp, &tx_rings_needed) ||
-	    tx_rings_needed < (tx * tx_sets + tx_xdp))
+	hwr.vnic = bnxt_get_total_vnics(bp, rx);
+
+	tx_cp = __bnxt_num_tx_to_cp(bp, hwr.tx, tx_sets, tx_xdp);
+	hwr.cp = sh ? max_t(int, tx_cp, rx) : tx_cp + rx;
+	if (max_cp < hwr.cp)
 		return -ENOMEM;
-	return 0;
+	hwr.stat = hwr.cp;
+	if (BNXT_NEW_RM(bp)) {
+		hwr.cp += bnxt_get_ulp_msix_num_in_use(bp);
+		hwr.stat += bnxt_get_ulp_stat_ctxs_in_use(bp);
+		hwr.grp = rx;
+		hwr.rss_ctx = bnxt_get_total_rss_ctxs(bp, &hwr);
+	}
+	if (bp->flags & BNXT_FLAG_CHIP_P5_PLUS)
+		hwr.cp_p5 = hwr.tx + rx;
+	rc = bnxt_hwrm_check_rings(bp, &hwr);
+	if (!rc && pci_msix_can_alloc_dyn(bp->pdev)) {
+		if (!bnxt_ulp_registered(bp->edev)) {
+			hwr.cp += bnxt_get_ulp_msix_num(bp);
+			hwr.cp = min_t(int, hwr.cp, bnxt_get_max_func_irqs(bp));
+		}
+		if (hwr.cp > bp->total_irqs) {
+			int total_msix = bnxt_change_msix(bp, hwr.cp);
+
+			if (total_msix < hwr.cp) {
+				netdev_warn(bp->dev, "Unable to allocate %d MSIX vectors, maximum available %d\n",
+					    hwr.cp, total_msix);
+				rc = -ENOSPC;
+			}
+		}
+	}
+	return rc;
 }
 
 static void bnxt_unmap_bars(struct bnxt *bp, struct pci_dev *pdev)
@@ -6913,7 +14692,510 @@ static void bnxt_cleanup_pci(struct bnxt *bp)
 {
 	bnxt_unmap_bars(bp, bp->pdev);
 	pci_release_regions(bp->pdev);
-	pci_disable_device(bp->pdev);
+	if (pci_is_enabled(bp->pdev))
+		pci_disable_device(bp->pdev);
+}
+
+static void bnxt_init_dflt_coal(struct bnxt *bp)
+{
+	struct bnxt_coal_cap *coal_cap = &bp->coal_cap;
+	struct bnxt_coal *coal;
+	u16 flags = 0;
+
+	if (coal_cap->cmpl_params &
+	    RING_AGGINT_QCAPS_RESP_CMPL_PARAMS_TIMER_RESET)
+		flags |= RING_CMPL_RING_CFG_AGGINT_PARAMS_REQ_FLAGS_TIMER_RESET;
+
+	/* Tick values in micro seconds.
+	 * 1 coal_buf x bufs_per_record = 1 completion record.
+	 */
+	coal = &bp->rx_coal;
+	coal->coal_ticks = 10;
+	coal->coal_bufs = 30;
+	coal->coal_ticks_irq = 1;
+	coal->coal_bufs_irq = 2;
+	coal->idle_thresh = 50;
+	coal->bufs_per_record = 2;
+	coal->budget = 64;		/* NAPI budget */
+	coal->flags = flags;
+
+	coal = &bp->tx_coal;
+	coal->coal_ticks = 28;
+	coal->coal_bufs = 30;
+	coal->coal_ticks_irq = 2;
+	coal->coal_bufs_irq = 2;
+	coal->bufs_per_record = 1;
+	coal->flags = flags;
+
+	bp->stats_coal_ticks = BNXT_DEF_STATS_COAL_TICKS;
+}
+
+/* FW that pre-reserves 1 VNIC per function */
+static bool bnxt_fw_pre_resv_vnics(struct bnxt *bp)
+{
+	u16 fw_maj = BNXT_FW_MAJ(bp), fw_bld = BNXT_FW_BLD(bp);
+
+	if (!(bp->flags & BNXT_FLAG_CHIP_P5_PLUS) &&
+	    (fw_maj > 218 || (fw_maj == 218 && fw_bld >= 18)))
+		return true;
+	if ((bp->flags & BNXT_FLAG_CHIP_P5_PLUS) &&
+	    (fw_maj > 216 || (fw_maj == 216 && fw_bld >= 172)))
+		return true;
+	return false;
+}
+
+static void bnxt_hwrm_pfcwd_qcaps(struct bnxt *bp)
+{
+	struct hwrm_queue_pfcwd_timeout_qcaps_output *resp;
+	struct hwrm_queue_pfcwd_timeout_qcaps_input *req;
+	int rc;
+
+	bp->max_pfcwd_tmo_ms = 0;
+	rc = hwrm_req_init(bp, req, HWRM_QUEUE_PFCWD_TIMEOUT_QCAPS);
+	if (rc)
+		return;
+	resp = hwrm_req_hold(bp, req);
+	rc = hwrm_req_send_silent(bp, req);
+	if (!rc)
+		bp->max_pfcwd_tmo_ms = le16_to_cpu(resp->max_pfcwd_timeout);
+	hwrm_req_drop(bp, req);
+}
+
+static int bnxt_fw_init_one_p1(struct bnxt *bp)
+{
+	int rc;
+
+	bp->fw_cap = 0;
+	rc = bnxt_hwrm_ver_get(bp);
+	/* FW may be unresponsive after FLR. FLR must complete within 100 msec
+	 * so wait before continuing with recovery.
+	 */
+	if (rc)
+		msleep(100);
+	bnxt_try_map_fw_health_reg(bp);
+	if (rc) {
+		rc = bnxt_try_recover_fw(bp);
+		if (rc)
+			return rc;
+		rc = bnxt_hwrm_ver_get(bp);
+		if (rc)
+			return rc;
+	}
+
+	bnxt_nvm_cfg_ver_get(bp);
+
+	rc = bnxt_hwrm_func_reset(bp);
+	if (rc)
+		return -ENODEV;
+
+	bnxt_hwrm_fw_set_time(bp);
+	return 0;
+}
+
+static int bnxt_fw_init_one_p2(struct bnxt *bp)
+{
+	int rc;
+
+	/* Get the MAX capabilities for this function */
+	rc = bnxt_hwrm_func_qcaps(bp);
+	if (rc) {
+		netdev_err(bp->dev, "hwrm query capability failure rc: %x\n",
+			   rc);
+		return -ENODEV;
+	}
+
+	rc = bnxt_hwrm_cfa_adv_flow_mgnt_qcaps(bp);
+	if (rc)
+		netdev_warn(bp->dev, "hwrm query adv flow mgnt failure rc: %d\n",
+			    rc);
+
+	if (bnxt_alloc_fw_health(bp)) {
+		netdev_warn(bp->dev, "no memory for firmware error recovery\n");
+	} else {
+		rc = bnxt_hwrm_error_recovery_qcfg(bp);
+		if (rc)
+			netdev_warn(bp->dev, "hwrm query error recovery failure rc: %d\n",
+				    rc);
+	}
+
+	rc = bnxt_hwrm_func_drv_rgtr(bp, NULL, 0, false);
+	if (rc)
+		return -ENODEV;
+
+	rc = bnxt_alloc_crash_dump_mem(bp);
+	if (rc)
+		netdev_warn(bp->dev, "crash dump mem alloc failure rc: %d\n",
+			    rc);
+	if (!rc) {
+		rc = bnxt_hwrm_crash_dump_mem_cfg(bp);
+		if (rc) {
+			bnxt_free_crash_dump_mem(bp);
+			netdev_warn(bp->dev,
+				    "hwrm crash dump mem failure rc: %d\n", rc);
+		}
+	}
+
+	if (bnxt_fw_pre_resv_vnics(bp))
+		bp->fw_cap |= BNXT_FW_CAP_PRE_RESV_VNICS;
+
+	bnxt_hwrm_pfcwd_qcaps(bp);
+	bnxt_hwrm_func_qcfg(bp);
+	bnxt_hwrm_vnic_qcaps(bp);
+	bnxt_hwrm_port_led_qcaps(bp);
+	bnxt_ethtool_init(bp);
+	if (bp->fw_cap & BNXT_FW_CAP_PTP)
+		__bnxt_hwrm_ptp_qcfg(bp);
+	bnxt_dcb_init(bp);
+	bnxt_hwmon_init(bp);
+	return 0;
+}
+
+static void bnxt_set_dflt_rss_hash_type(struct bnxt *bp)
+{
+	bp->rss_cap &= ~BNXT_RSS_CAP_UDP_RSS_CAP;
+	bp->rss_hash_cfg = VNIC_RSS_CFG_REQ_HASH_TYPE_IPV4 |
+			   VNIC_RSS_CFG_REQ_HASH_TYPE_TCP_IPV4 |
+			   VNIC_RSS_CFG_REQ_HASH_TYPE_IPV6 |
+			   VNIC_RSS_CFG_REQ_HASH_TYPE_TCP_IPV6;
+	if (bp->rss_cap & BNXT_RSS_CAP_RSS_HASH_TYPE_DELTA)
+		bp->rss_hash_delta = bp->rss_hash_cfg;
+	if (BNXT_CHIP_P4_PLUS(bp) && bp->hwrm_spec_code >= 0x10501) {
+		bp->rss_cap |= BNXT_RSS_CAP_UDP_RSS_CAP;
+		bp->rss_hash_cfg |= VNIC_RSS_CFG_REQ_HASH_TYPE_UDP_IPV4 |
+				    VNIC_RSS_CFG_REQ_HASH_TYPE_UDP_IPV6;
+	}
+}
+
+static void bnxt_set_dflt_rfs(struct bnxt *bp)
+{
+	struct net_device *dev = bp->dev;
+
+	dev->hw_features &= ~NETIF_F_NTUPLE;
+	dev->features &= ~NETIF_F_NTUPLE;
+	bp->flags &= ~BNXT_FLAG_RFS;
+	if (bnxt_rfs_supported(bp)) {
+		dev->hw_features |= NETIF_F_NTUPLE;
+		if (bnxt_rfs_capable(bp, false)) {
+			bp->flags |= BNXT_FLAG_RFS;
+			dev->features |= NETIF_F_NTUPLE;
+		}
+	}
+}
+
+static void bnxt_fw_init_one_p3(struct bnxt *bp)
+{
+	struct pci_dev *pdev = bp->pdev;
+
+	bnxt_set_dflt_rss_hash_type(bp);
+	bnxt_set_dflt_rfs(bp);
+
+	bnxt_get_wol_settings(bp);
+	if (bp->flags & BNXT_FLAG_WOL_CAP)
+		device_set_wakeup_enable(&pdev->dev, bp->wol);
+	else
+		device_set_wakeup_capable(&pdev->dev, false);
+
+	bnxt_hwrm_set_cache_line_size(bp, cache_line_size());
+	bnxt_hwrm_coal_params_qcaps(bp);
+}
+
+static int bnxt_probe_phy(struct bnxt *bp, bool fw_dflt);
+
+int bnxt_fw_init_one(struct bnxt *bp)
+{
+	int rc;
+
+	rc = bnxt_fw_init_one_p1(bp);
+	if (rc) {
+		netdev_err(bp->dev, "Firmware init phase 1 failed\n");
+		return rc;
+	}
+	rc = bnxt_fw_init_one_p2(bp);
+	if (rc) {
+		netdev_err(bp->dev, "Firmware init phase 2 failed\n");
+		return rc;
+	}
+	rc = bnxt_probe_phy(bp, false);
+	if (rc)
+		return rc;
+	rc = bnxt_approve_mac(bp, bp->dev->dev_addr, false);
+	if (rc)
+		return rc;
+
+	bnxt_fw_init_one_p3(bp);
+	return 0;
+}
+
+static void bnxt_fw_reset_writel(struct bnxt *bp, int reg_idx)
+{
+	struct bnxt_fw_health *fw_health = bp->fw_health;
+	u32 reg = fw_health->fw_reset_seq_regs[reg_idx];
+	u32 val = fw_health->fw_reset_seq_vals[reg_idx];
+	u32 reg_type, reg_off, delay_msecs;
+
+	delay_msecs = fw_health->fw_reset_seq_delay_msec[reg_idx];
+	reg_type = BNXT_FW_HEALTH_REG_TYPE(reg);
+	reg_off = BNXT_FW_HEALTH_REG_OFF(reg);
+	switch (reg_type) {
+	case BNXT_FW_HEALTH_REG_TYPE_CFG:
+		pci_write_config_dword(bp->pdev, reg_off, val);
+		break;
+	case BNXT_FW_HEALTH_REG_TYPE_GRC:
+		writel(reg_off & BNXT_GRC_BASE_MASK,
+		       bp->bar0 + BNXT_GRCPF_REG_WINDOW_BASE_OUT + 4);
+		reg_off = (reg_off & BNXT_GRC_OFFSET_MASK) + 0x2000;
+		fallthrough;
+	case BNXT_FW_HEALTH_REG_TYPE_BAR0:
+		writel(val, bp->bar0 + reg_off);
+		break;
+	case BNXT_FW_HEALTH_REG_TYPE_BAR1:
+		writel(val, bp->bar1 + reg_off);
+		break;
+	}
+	if (delay_msecs) {
+		pci_read_config_dword(bp->pdev, 0, &val);
+		msleep(delay_msecs);
+	}
+}
+
+bool bnxt_hwrm_reset_permitted(struct bnxt *bp)
+{
+	struct hwrm_func_qcfg_output *resp;
+	struct hwrm_func_qcfg_input *req;
+	bool result = true; /* firmware will enforce if unknown */
+
+	if (~bp->fw_cap & BNXT_FW_CAP_HOT_RESET_IF)
+		return result;
+
+	if (hwrm_req_init(bp, req, HWRM_FUNC_QCFG))
+		return result;
+
+	req->fid = cpu_to_le16(0xffff);
+	resp = hwrm_req_hold(bp, req);
+	if (!hwrm_req_send(bp, req))
+		result = !!(le16_to_cpu(resp->flags) &
+			    FUNC_QCFG_RESP_FLAGS_HOT_RESET_ALLOWED);
+	hwrm_req_drop(bp, req);
+	return result;
+}
+
+static void bnxt_reset_all(struct bnxt *bp)
+{
+	struct bnxt_fw_health *fw_health = bp->fw_health;
+	int i, rc;
+
+	if (bp->fw_cap & BNXT_FW_CAP_ERR_RECOVER_RELOAD) {
+		bnxt_fw_reset_via_optee(bp);
+		bp->fw_reset_timestamp = jiffies;
+		return;
+	}
+
+	if (fw_health->flags & ERROR_RECOVERY_QCFG_RESP_FLAGS_HOST) {
+		for (i = 0; i < fw_health->fw_reset_seq_cnt; i++)
+			bnxt_fw_reset_writel(bp, i);
+	} else if (fw_health->flags & ERROR_RECOVERY_QCFG_RESP_FLAGS_CO_CPU) {
+		struct hwrm_fw_reset_input *req;
+
+		rc = hwrm_req_init(bp, req, HWRM_FW_RESET);
+		if (!rc) {
+			req->target_id = cpu_to_le16(HWRM_TARGET_ID_KONG);
+			req->embedded_proc_type = FW_RESET_REQ_EMBEDDED_PROC_TYPE_CHIP;
+			req->selfrst_status = FW_RESET_REQ_SELFRST_STATUS_SELFRSTASAP;
+			req->flags = FW_RESET_REQ_FLAGS_RESET_GRACEFUL;
+			rc = hwrm_req_send(bp, req);
+		}
+		if (rc != -ENODEV)
+			netdev_warn(bp->dev, "Unable to reset FW rc=%d\n", rc);
+	}
+	bp->fw_reset_timestamp = jiffies;
+}
+
+static bool bnxt_fw_reset_timeout(struct bnxt *bp)
+{
+	return time_after(jiffies, bp->fw_reset_timestamp +
+			  (bp->fw_reset_max_dsecs * HZ / 10));
+}
+
+static void bnxt_fw_reset_abort(struct bnxt *bp, int rc)
+{
+	clear_bit(BNXT_STATE_IN_FW_RESET, &bp->state);
+	if (bp->fw_reset_state != BNXT_FW_RESET_STATE_POLL_VF)
+		bnxt_dl_health_fw_status_update(bp, false);
+	bp->fw_reset_state = BNXT_FW_RESET_STATE_ABORT;
+	netif_close(bp->dev);
+}
+
+static void bnxt_fw_reset_task(struct work_struct *work)
+{
+	struct bnxt *bp = container_of(work, struct bnxt, fw_reset_task.work);
+	int rc = 0;
+
+	if (!test_bit(BNXT_STATE_IN_FW_RESET, &bp->state)) {
+		netdev_err(bp->dev, "bnxt_fw_reset_task() called when not in fw reset mode!\n");
+		return;
+	}
+
+	switch (bp->fw_reset_state) {
+	case BNXT_FW_RESET_STATE_POLL_VF: {
+		int n = bnxt_get_registered_vfs(bp);
+		int tmo;
+
+		if (n < 0) {
+			netdev_err(bp->dev, "Firmware reset aborted, subsequent func_qcfg cmd failed, rc = %d, %d msecs since reset timestamp\n",
+				   n, jiffies_to_msecs(jiffies -
+				   bp->fw_reset_timestamp));
+			goto fw_reset_abort;
+		} else if (n > 0) {
+			if (bnxt_fw_reset_timeout(bp)) {
+				clear_bit(BNXT_STATE_IN_FW_RESET, &bp->state);
+				bp->fw_reset_state = 0;
+				netdev_err(bp->dev, "Firmware reset aborted, bnxt_get_registered_vfs() returns %d\n",
+					   n);
+				goto ulp_start;
+			}
+			bnxt_queue_fw_reset_work(bp, HZ / 10);
+			return;
+		}
+		bp->fw_reset_timestamp = jiffies;
+		netdev_lock(bp->dev);
+		if (test_bit(BNXT_STATE_ABORT_ERR, &bp->state)) {
+			bnxt_fw_reset_abort(bp, rc);
+			netdev_unlock(bp->dev);
+			goto ulp_start;
+		}
+		bnxt_fw_reset_close(bp);
+		if (bp->fw_cap & BNXT_FW_CAP_ERR_RECOVER_RELOAD) {
+			bp->fw_reset_state = BNXT_FW_RESET_STATE_POLL_FW_DOWN;
+			tmo = HZ / 10;
+		} else {
+			bp->fw_reset_state = BNXT_FW_RESET_STATE_ENABLE_DEV;
+			tmo = bp->fw_reset_min_dsecs * HZ / 10;
+		}
+		netdev_unlock(bp->dev);
+		bnxt_queue_fw_reset_work(bp, tmo);
+		return;
+	}
+	case BNXT_FW_RESET_STATE_POLL_FW_DOWN: {
+		u32 val;
+
+		val = bnxt_fw_health_readl(bp, BNXT_FW_HEALTH_REG);
+		if (!(val & BNXT_FW_STATUS_SHUTDOWN) &&
+		    !bnxt_fw_reset_timeout(bp)) {
+			bnxt_queue_fw_reset_work(bp, HZ / 5);
+			return;
+		}
+
+		if (!bp->fw_health->primary) {
+			u32 wait_dsecs = bp->fw_health->normal_func_wait_dsecs;
+
+			bp->fw_reset_state = BNXT_FW_RESET_STATE_ENABLE_DEV;
+			bnxt_queue_fw_reset_work(bp, wait_dsecs * HZ / 10);
+			return;
+		}
+		bp->fw_reset_state = BNXT_FW_RESET_STATE_RESET_FW;
+	}
+		fallthrough;
+	case BNXT_FW_RESET_STATE_RESET_FW:
+		bnxt_reset_all(bp);
+		bp->fw_reset_state = BNXT_FW_RESET_STATE_ENABLE_DEV;
+		bnxt_queue_fw_reset_work(bp, bp->fw_reset_min_dsecs * HZ / 10);
+		return;
+	case BNXT_FW_RESET_STATE_ENABLE_DEV:
+		bnxt_inv_fw_health_reg(bp);
+		if (test_bit(BNXT_STATE_FW_FATAL_COND, &bp->state) &&
+		    !bp->fw_reset_min_dsecs) {
+			u16 val;
+
+			pci_read_config_word(bp->pdev, PCI_SUBSYSTEM_ID, &val);
+			if (val == 0xffff) {
+				if (bnxt_fw_reset_timeout(bp)) {
+					netdev_err(bp->dev, "Firmware reset aborted, PCI config space invalid\n");
+					rc = -ETIMEDOUT;
+					goto fw_reset_abort;
+				}
+				bnxt_queue_fw_reset_work(bp, HZ / 1000);
+				return;
+			}
+		}
+		clear_bit(BNXT_STATE_FW_FATAL_COND, &bp->state);
+		clear_bit(BNXT_STATE_FW_NON_FATAL_COND, &bp->state);
+		if (test_and_clear_bit(BNXT_STATE_FW_ACTIVATE_RESET, &bp->state) &&
+		    !test_bit(BNXT_STATE_FW_ACTIVATE, &bp->state))
+			bnxt_dl_remote_reload(bp);
+		if (pci_enable_device(bp->pdev)) {
+			netdev_err(bp->dev, "Cannot re-enable PCI device\n");
+			rc = -ENODEV;
+			goto fw_reset_abort;
+		}
+		pci_set_master(bp->pdev);
+		bp->fw_reset_state = BNXT_FW_RESET_STATE_POLL_FW;
+		fallthrough;
+	case BNXT_FW_RESET_STATE_POLL_FW:
+		bp->hwrm_cmd_timeout = SHORT_HWRM_CMD_TIMEOUT;
+		rc = bnxt_hwrm_poll(bp);
+		if (rc) {
+			if (bnxt_fw_reset_timeout(bp)) {
+				netdev_err(bp->dev, "Firmware reset aborted\n");
+				goto fw_reset_abort_status;
+			}
+			bnxt_queue_fw_reset_work(bp, HZ / 5);
+			return;
+		}
+		bp->hwrm_cmd_timeout = DFLT_HWRM_CMD_TIMEOUT;
+		bp->fw_reset_state = BNXT_FW_RESET_STATE_OPENING;
+		fallthrough;
+	case BNXT_FW_RESET_STATE_OPENING:
+		while (!netdev_trylock(bp->dev)) {
+			bnxt_queue_fw_reset_work(bp, HZ / 10);
+			return;
+		}
+		rc = bnxt_open(bp->dev);
+		if (rc) {
+			netdev_err(bp->dev, "bnxt_open() failed during FW reset\n");
+			bnxt_fw_reset_abort(bp, rc);
+			netdev_unlock(bp->dev);
+			goto ulp_start;
+		}
+
+		if ((bp->fw_cap & BNXT_FW_CAP_ERROR_RECOVERY) &&
+		    bp->fw_health->enabled) {
+			bp->fw_health->last_fw_reset_cnt =
+				bnxt_fw_health_readl(bp, BNXT_FW_RESET_CNT_REG);
+		}
+		bp->fw_reset_state = 0;
+		/* Make sure fw_reset_state is 0 before clearing the flag */
+		smp_mb__before_atomic();
+		clear_bit(BNXT_STATE_IN_FW_RESET, &bp->state);
+		bnxt_ptp_reapply_pps(bp);
+		clear_bit(BNXT_STATE_FW_ACTIVATE, &bp->state);
+		if (test_and_clear_bit(BNXT_STATE_RECOVER, &bp->state)) {
+			bnxt_dl_health_fw_recovery_done(bp);
+			bnxt_dl_health_fw_status_update(bp, true);
+		}
+		netdev_unlock(bp->dev);
+		bnxt_ulp_start(bp, 0);
+		bnxt_reenable_sriov(bp);
+		netdev_lock(bp->dev);
+		bnxt_vf_reps_alloc(bp);
+		bnxt_vf_reps_open(bp);
+		netdev_unlock(bp->dev);
+		break;
+	}
+	return;
+
+fw_reset_abort_status:
+	if (bp->fw_health->status_reliable ||
+	    (bp->fw_cap & BNXT_FW_CAP_ERROR_RECOVERY)) {
+		u32 sts = bnxt_fw_health_readl(bp, BNXT_FW_HEALTH_REG);
+
+		netdev_err(bp->dev, "fw_health_status 0x%x\n", sts);
+	}
+fw_reset_abort:
+	netdev_lock(bp->dev);
+	bnxt_fw_reset_abort(bp, rc);
+	netdev_unlock(bp->dev);
+ulp_start:
+	bnxt_ulp_start(bp, rc);
 }
 
 static int bnxt_init_board(struct pci_dev *pdev, struct net_device *dev)
@@ -6946,7 +15228,8 @@ static int bnxt_init_board(struct pci_dev *pdev, struct net_device *dev)
 	if (dma_set_mask_and_coherent(&pdev->dev, DMA_BIT_MASK(64)) != 0 &&
 	    dma_set_mask_and_coherent(&pdev->dev, DMA_BIT_MASK(32)) != 0) {
 		dev_err(&pdev->dev, "System does not support DMA, aborting\n");
-		goto init_err_disable;
+		rc = -EIO;
+		goto init_err_release;
 	}
 
 	pci_set_master(pdev);
@@ -6954,6 +15237,9 @@ static int bnxt_init_board(struct pci_dev *pdev, struct net_device *dev)
 	bp->dev = dev;
 	bp->pdev = pdev;
 
+	/* Doorbell BAR bp->bar1 is mapped after bnxt_fw_init_one_p2()
+	 * determines the BAR size.
+	 */
 	bp->bar0 = pci_ioremap_bar(pdev, 0);
 	if (!bp->bar0) {
 		dev_err(&pdev->dev, "Cannot map device registers, aborting\n");
@@ -6961,13 +15247,6 @@ static int bnxt_init_board(struct pci_dev *pdev, struct net_device *dev)
 		goto init_err_release;
 	}
 
-	bp->bar1 = pci_ioremap_bar(pdev, 2);
-	if (!bp->bar1) {
-		dev_err(&pdev->dev, "Cannot map doorbell registers, aborting\n");
-		rc = -ENOMEM;
-		goto init_err_release;
-	}
-
 	bp->bar2 = pci_ioremap_bar(pdev, 4);
 	if (!bp->bar2) {
 		dev_err(&pdev->dev, "Cannot map bar4 registers, aborting\n");
@@ -6975,33 +15254,23 @@ static int bnxt_init_board(struct pci_dev *pdev, struct net_device *dev)
 		goto init_err_release;
 	}
 
-	pci_enable_pcie_error_reporting(pdev);
-
 	INIT_WORK(&bp->sp_task, bnxt_sp_task);
+	INIT_DELAYED_WORK(&bp->fw_reset_task, bnxt_fw_reset_task);
 
 	spin_lock_init(&bp->ntp_fltr_lock);
+#if BITS_PER_LONG == 32
+	spin_lock_init(&bp->db_lock);
+#endif
 
 	bp->rx_ring_size = BNXT_DEFAULT_RX_RING_SIZE;
 	bp->tx_ring_size = BNXT_DEFAULT_TX_RING_SIZE;
 
-	/* tick values in micro seconds */
-	bp->rx_coal_ticks = 12;
-	bp->rx_coal_bufs = 30;
-	bp->rx_coal_ticks_irq = 1;
-	bp->rx_coal_bufs_irq = 2;
-
-	bp->tx_coal_ticks = 25;
-	bp->tx_coal_bufs = 30;
-	bp->tx_coal_ticks_irq = 2;
-	bp->tx_coal_bufs_irq = 2;
-
-	bp->stats_coal_ticks = BNXT_DEF_STATS_COAL_TICKS;
-
-	init_timer(&bp->timer);
-	bp->timer.data = (unsigned long)bp;
-	bp->timer.function = bnxt_timer;
+	timer_setup(&bp->timer, bnxt_timer, 0);
 	bp->current_interval = BNXT_TIMER_INTERVAL;
 
+	bp->vxlan_fw_dst_port_id = INVALID_HW_RING_ID;
+	bp->nge_fw_dst_port_id = INVALID_HW_RING_ID;
+
 	clear_bit(BNXT_STATE_OPEN, &bp->state);
 	return 0;
 
@@ -7016,24 +15285,26 @@ init_err:
 	return rc;
 }
 
-/* rtnl_lock held */
 static int bnxt_change_mac_addr(struct net_device *dev, void *p)
 {
 	struct sockaddr *addr = p;
 	struct bnxt *bp = netdev_priv(dev);
 	int rc = 0;
 
+	netdev_assert_locked(dev);
+
 	if (!is_valid_ether_addr(addr->sa_data))
 		return -EADDRNOTAVAIL;
 
-	rc = bnxt_approve_mac(bp, addr->sa_data);
-	if (rc)
-		return rc;
-
 	if (ether_addr_equal(addr->sa_data, dev->dev_addr))
 		return 0;
 
-	memcpy(dev->dev_addr, addr->sa_data, dev->addr_len);
+	rc = bnxt_approve_mac(bp, addr->sa_data, true);
+	if (rc)
+		return rc;
+
+	eth_hw_addr_set(dev, addr->sa_data);
+	bnxt_clear_usr_fltrs(bp, true);
 	if (netif_running(dev)) {
 		bnxt_close_nic(bp, false, false);
 		rc = bnxt_open_nic(bp, false, false);
@@ -7042,19 +15313,28 @@ static int bnxt_change_mac_addr(struct net_device *dev, void *p)
 	return rc;
 }
 
-/* rtnl_lock held */
 static int bnxt_change_mtu(struct net_device *dev, int new_mtu)
 {
 	struct bnxt *bp = netdev_priv(dev);
 
+	netdev_assert_locked(dev);
+
 	if (netif_running(dev))
-		bnxt_close_nic(bp, false, false);
+		bnxt_close_nic(bp, true, false);
+
+	WRITE_ONCE(dev->mtu, new_mtu);
+
+	/* MTU change may change the AGG ring settings if an XDP multi-buffer
+	 * program is attached.  We need to set the AGG rings settings and
+	 * rx_skb_func accordingly.
+	 */
+	if (READ_ONCE(bp->xdp_prog))
+		bnxt_set_rx_skb_mode(bp, true);
 
-	dev->mtu = new_mtu;
 	bnxt_set_ring_params(bp);
 
 	if (netif_running(dev))
-		return bnxt_open_nic(bp, false, false);
+		return bnxt_open_nic(bp, true, false);
 
 	return 0;
 }
@@ -7063,7 +15343,7 @@ int bnxt_setup_mq_tc(struct net_device *dev, u8 tc)
 {
 	struct bnxt *bp = netdev_priv(dev);
 	bool sh = false;
-	int rc;
+	int rc, tx_cp;
 
 	if (tc > bp->max_tc) {
 		netdev_err(dev, "Too many traffic classes requested: %d. Max supported is %d.\n",
@@ -7071,14 +15351,14 @@ int bnxt_setup_mq_tc(struct net_device *dev, u8 tc)
 		return -EINVAL;
 	}
 
-	if (netdev_get_num_tc(dev) == tc)
+	if (bp->num_tc == tc)
 		return 0;
 
 	if (bp->flags & BNXT_FLAG_SHARED_RINGS)
 		sh = true;
 
-	rc = bnxt_reserve_rings(bp, bp->tx_nr_rings_per_tc, bp->rx_nr_rings,
-				tc, bp->tx_nr_rings_xdp);
+	rc = bnxt_check_rings(bp, bp->tx_nr_rings_per_tc, bp->rx_nr_rings,
+			      sh, tc, bp->tx_nr_rings_xdp);
 	if (rc)
 		return rc;
 
@@ -7089,13 +15369,16 @@ int bnxt_setup_mq_tc(struct net_device *dev, u8 tc)
 	if (tc) {
 		bp->tx_nr_rings = bp->tx_nr_rings_per_tc * tc;
 		netdev_set_num_tc(dev, tc);
+		bp->num_tc = tc;
 	} else {
 		bp->tx_nr_rings = bp->tx_nr_rings_per_tc;
 		netdev_reset_tc(dev);
+		bp->num_tc = 0;
 	}
-	bp->cp_nr_rings = sh ? max_t(int, bp->tx_nr_rings, bp->rx_nr_rings) :
-			       bp->tx_nr_rings + bp->rx_nr_rings;
-	bp->num_stat_ctxs = bp->cp_nr_rings;
+	bp->tx_nr_rings += bp->tx_nr_rings_xdp;
+	tx_cp = bnxt_num_tx_to_cp(bp, bp->tx_nr_rings);
+	bp->cp_nr_rings = sh ? max_t(int, tx_cp, bp->rx_nr_rings) :
+			       tx_cp + bp->rx_nr_rings;
 
 	if (netif_running(bp->dev))
 		return bnxt_open_nic(bp, true, false);
@@ -7103,37 +15386,139 @@ int bnxt_setup_mq_tc(struct net_device *dev, u8 tc)
 	return 0;
 }
 
-static int bnxt_setup_tc(struct net_device *dev, u32 handle, __be16 proto,
-			 struct tc_to_netdev *ntc)
+static int bnxt_setup_tc_block_cb(enum tc_setup_type type, void *type_data,
+				  void *cb_priv)
 {
-	if (ntc->type != TC_SETUP_MQPRIO)
-		return -EINVAL;
+	struct bnxt *bp = cb_priv;
 
-	ntc->mqprio->hw = TC_MQPRIO_HW_OFFLOAD_TCS;
+	if (!bnxt_tc_flower_enabled(bp) ||
+	    !tc_cls_can_offload_and_chain0(bp->dev, type_data))
+		return -EOPNOTSUPP;
 
-	return bnxt_setup_mq_tc(dev, ntc->mqprio->num_tc);
+	switch (type) {
+	case TC_SETUP_CLSFLOWER:
+		return bnxt_tc_setup_flower(bp, bp->pf.fw_fid, type_data);
+	default:
+		return -EOPNOTSUPP;
+	}
+}
+
+LIST_HEAD(bnxt_block_cb_list);
+
+static int bnxt_setup_tc(struct net_device *dev, enum tc_setup_type type,
+			 void *type_data)
+{
+	struct bnxt *bp = netdev_priv(dev);
+
+	switch (type) {
+	case TC_SETUP_BLOCK:
+		return flow_block_cb_setup_simple(type_data,
+						  &bnxt_block_cb_list,
+						  bnxt_setup_tc_block_cb,
+						  bp, bp, true);
+	case TC_SETUP_QDISC_MQPRIO: {
+		struct tc_mqprio_qopt *mqprio = type_data;
+
+		mqprio->hw = TC_MQPRIO_HW_OFFLOAD_TCS;
+
+		return bnxt_setup_mq_tc(dev, mqprio->num_tc);
+	}
+	default:
+		return -EOPNOTSUPP;
+	}
+}
+
+u32 bnxt_get_ntp_filter_idx(struct bnxt *bp, struct flow_keys *fkeys,
+			    const struct sk_buff *skb)
+{
+	struct bnxt_vnic_info *vnic;
+
+	if (skb)
+		return skb_get_hash_raw(skb) & BNXT_NTP_FLTR_HASH_MASK;
+
+	vnic = &bp->vnic_info[BNXT_VNIC_DEFAULT];
+	return bnxt_toeplitz(bp, fkeys, (void *)vnic->rss_hash_key);
+}
+
+int bnxt_insert_ntp_filter(struct bnxt *bp, struct bnxt_ntuple_filter *fltr,
+			   u32 idx)
+{
+	struct hlist_head *head;
+	int bit_id;
+
+	spin_lock_bh(&bp->ntp_fltr_lock);
+	bit_id = bitmap_find_free_region(bp->ntp_fltr_bmap, bp->max_fltr, 0);
+	if (bit_id < 0) {
+		spin_unlock_bh(&bp->ntp_fltr_lock);
+		return -ENOMEM;
+	}
+
+	fltr->base.sw_id = (u16)bit_id;
+	fltr->base.type = BNXT_FLTR_TYPE_NTUPLE;
+	fltr->base.flags |= BNXT_ACT_RING_DST;
+	head = &bp->ntp_fltr_hash_tbl[idx];
+	hlist_add_head_rcu(&fltr->base.hash, head);
+	set_bit(BNXT_FLTR_INSERTED, &fltr->base.state);
+	bnxt_insert_usr_fltr(bp, &fltr->base);
+	bp->ntp_fltr_count++;
+	spin_unlock_bh(&bp->ntp_fltr_lock);
+	return 0;
 }
 
-#ifdef CONFIG_RFS_ACCEL
 static bool bnxt_fltr_match(struct bnxt_ntuple_filter *f1,
 			    struct bnxt_ntuple_filter *f2)
 {
+	struct bnxt_flow_masks *masks1 = &f1->fmasks;
+	struct bnxt_flow_masks *masks2 = &f2->fmasks;
 	struct flow_keys *keys1 = &f1->fkeys;
 	struct flow_keys *keys2 = &f2->fkeys;
 
-	if (keys1->addrs.v4addrs.src == keys2->addrs.v4addrs.src &&
-	    keys1->addrs.v4addrs.dst == keys2->addrs.v4addrs.dst &&
-	    keys1->ports.ports == keys2->ports.ports &&
-	    keys1->basic.ip_proto == keys2->basic.ip_proto &&
-	    keys1->basic.n_proto == keys2->basic.n_proto &&
-	    keys1->control.flags == keys2->control.flags &&
-	    ether_addr_equal(f1->src_mac_addr, f2->src_mac_addr) &&
-	    ether_addr_equal(f1->dst_mac_addr, f2->dst_mac_addr))
-		return true;
+	if (keys1->basic.n_proto != keys2->basic.n_proto ||
+	    keys1->basic.ip_proto != keys2->basic.ip_proto)
+		return false;
 
-	return false;
+	if (keys1->basic.n_proto == htons(ETH_P_IP)) {
+		if (keys1->addrs.v4addrs.src != keys2->addrs.v4addrs.src ||
+		    masks1->addrs.v4addrs.src != masks2->addrs.v4addrs.src ||
+		    keys1->addrs.v4addrs.dst != keys2->addrs.v4addrs.dst ||
+		    masks1->addrs.v4addrs.dst != masks2->addrs.v4addrs.dst)
+			return false;
+	} else {
+		if (!ipv6_addr_equal(&keys1->addrs.v6addrs.src,
+				     &keys2->addrs.v6addrs.src) ||
+		    !ipv6_addr_equal(&masks1->addrs.v6addrs.src,
+				     &masks2->addrs.v6addrs.src) ||
+		    !ipv6_addr_equal(&keys1->addrs.v6addrs.dst,
+				     &keys2->addrs.v6addrs.dst) ||
+		    !ipv6_addr_equal(&masks1->addrs.v6addrs.dst,
+				     &masks2->addrs.v6addrs.dst))
+			return false;
+	}
+
+	return keys1->ports.src == keys2->ports.src &&
+	       masks1->ports.src == masks2->ports.src &&
+	       keys1->ports.dst == keys2->ports.dst &&
+	       masks1->ports.dst == masks2->ports.dst &&
+	       keys1->control.flags == keys2->control.flags &&
+	       f1->l2_fltr == f2->l2_fltr;
+}
+
+struct bnxt_ntuple_filter *
+bnxt_lookup_ntp_filter_from_idx(struct bnxt *bp,
+				struct bnxt_ntuple_filter *fltr, u32 idx)
+{
+	struct bnxt_ntuple_filter *f;
+	struct hlist_head *head;
+
+	head = &bp->ntp_fltr_hash_tbl[idx];
+	hlist_for_each_entry_rcu(f, head, base.hash) {
+		if (bnxt_fltr_match(f, fltr))
+			return f;
+	}
+	return NULL;
 }
 
+#ifdef CONFIG_RFS_ACCEL
 static int bnxt_rx_flow_steer(struct net_device *dev, const struct sk_buff *skb,
 			      u16 rxq_index, u32 flow_id)
 {
@@ -7141,28 +15526,31 @@ static int bnxt_rx_flow_steer(struct net_device *dev, const struct sk_buff *skb,
 	struct bnxt_ntuple_filter *fltr, *new_fltr;
 	struct flow_keys *fkeys;
 	struct ethhdr *eth = (struct ethhdr *)skb_mac_header(skb);
-	int rc = 0, idx, bit_id, l2_idx = 0;
-	struct hlist_head *head;
+	struct bnxt_l2_filter *l2_fltr;
+	int rc = 0, idx;
+	u32 flags;
 
-	if (!ether_addr_equal(dev->dev_addr, eth->h_dest)) {
-		struct bnxt_vnic_info *vnic = &bp->vnic_info[0];
-		int off = 0, j;
+	if (ether_addr_equal(dev->dev_addr, eth->h_dest)) {
+		l2_fltr = bp->vnic_info[BNXT_VNIC_DEFAULT].l2_filters[0];
+		atomic_inc(&l2_fltr->refcnt);
+	} else {
+		struct bnxt_l2_key key;
 
-		netif_addr_lock_bh(dev);
-		for (j = 0; j < vnic->uc_filter_count; j++, off += ETH_ALEN) {
-			if (ether_addr_equal(eth->h_dest,
-					     vnic->uc_list + off)) {
-				l2_idx = j + 1;
-				break;
-			}
-		}
-		netif_addr_unlock_bh(dev);
-		if (!l2_idx)
+		ether_addr_copy(key.dst_mac_addr, eth->h_dest);
+		key.vlan = 0;
+		l2_fltr = bnxt_lookup_l2_filter_from_key(bp, &key);
+		if (!l2_fltr)
+			return -EINVAL;
+		if (l2_fltr->base.flags & BNXT_ACT_FUNC_DST) {
+			bnxt_del_l2_filter(bp, l2_fltr);
 			return -EINVAL;
+		}
 	}
 	new_fltr = kzalloc(sizeof(*new_fltr), GFP_ATOMIC);
-	if (!new_fltr)
+	if (!new_fltr) {
+		bnxt_del_l2_filter(bp, l2_fltr);
 		return -ENOMEM;
+	}
 
 	fkeys = &new_fltr->fkeys;
 	if (!skb_flow_dissect_flow_keys(skb, fkeys, 0)) {
@@ -7177,61 +15565,66 @@ static int bnxt_rx_flow_steer(struct net_device *dev, const struct sk_buff *skb,
 		rc = -EPROTONOSUPPORT;
 		goto err_free;
 	}
-	if (fkeys->basic.n_proto == htons(ETH_P_IPV6) &&
-	    bp->hwrm_spec_code < 0x10601) {
-		rc = -EPROTONOSUPPORT;
-		goto err_free;
+	new_fltr->fmasks = BNXT_FLOW_IPV4_MASK_ALL;
+	if (fkeys->basic.n_proto == htons(ETH_P_IPV6)) {
+		if (bp->hwrm_spec_code < 0x10601) {
+			rc = -EPROTONOSUPPORT;
+			goto err_free;
+		}
+		new_fltr->fmasks = BNXT_FLOW_IPV6_MASK_ALL;
 	}
-	if ((fkeys->control.flags & FLOW_DIS_ENCAPSULATION) &&
-	    bp->hwrm_spec_code < 0x10601) {
+	flags = fkeys->control.flags;
+	if (((flags & FLOW_DIS_ENCAPSULATION) &&
+	     bp->hwrm_spec_code < 0x10601) || (flags & FLOW_DIS_IS_FRAGMENT)) {
 		rc = -EPROTONOSUPPORT;
 		goto err_free;
 	}
+	new_fltr->l2_fltr = l2_fltr;
 
-	memcpy(new_fltr->dst_mac_addr, eth->h_dest, ETH_ALEN);
-	memcpy(new_fltr->src_mac_addr, eth->h_source, ETH_ALEN);
-
-	idx = skb_get_hash_raw(skb) & BNXT_NTP_FLTR_HASH_MASK;
-	head = &bp->ntp_fltr_hash_tbl[idx];
+	idx = bnxt_get_ntp_filter_idx(bp, fkeys, skb);
 	rcu_read_lock();
-	hlist_for_each_entry_rcu(fltr, head, hash) {
-		if (bnxt_fltr_match(fltr, new_fltr)) {
-			rcu_read_unlock();
-			rc = 0;
-			goto err_free;
-		}
-	}
-	rcu_read_unlock();
-
-	spin_lock_bh(&bp->ntp_fltr_lock);
-	bit_id = bitmap_find_free_region(bp->ntp_fltr_bmap,
-					 BNXT_NTP_FLTR_MAX_FLTR, 0);
-	if (bit_id < 0) {
-		spin_unlock_bh(&bp->ntp_fltr_lock);
-		rc = -ENOMEM;
+	fltr = bnxt_lookup_ntp_filter_from_idx(bp, new_fltr, idx);
+	if (fltr) {
+		rc = fltr->base.sw_id;
+		rcu_read_unlock();
 		goto err_free;
 	}
+	rcu_read_unlock();
 
-	new_fltr->sw_id = (u16)bit_id;
 	new_fltr->flow_id = flow_id;
-	new_fltr->l2_fltr_idx = l2_idx;
-	new_fltr->rxq = rxq_index;
-	hlist_add_head_rcu(&new_fltr->hash, head);
-	bp->ntp_fltr_count++;
-	spin_unlock_bh(&bp->ntp_fltr_lock);
-
-	set_bit(BNXT_RX_NTP_FLTR_SP_EVENT, &bp->sp_event);
-	schedule_work(&bp->sp_task);
-
-	return new_fltr->sw_id;
+	new_fltr->base.rxq = rxq_index;
+	rc = bnxt_insert_ntp_filter(bp, new_fltr, idx);
+	if (!rc) {
+		bnxt_queue_sp_work(bp, BNXT_RX_NTP_FLTR_SP_EVENT);
+		return new_fltr->base.sw_id;
+	}
 
 err_free:
+	bnxt_del_l2_filter(bp, l2_fltr);
 	kfree(new_fltr);
 	return rc;
 }
+#endif
+
+void bnxt_del_ntp_filter(struct bnxt *bp, struct bnxt_ntuple_filter *fltr)
+{
+	spin_lock_bh(&bp->ntp_fltr_lock);
+	if (!test_and_clear_bit(BNXT_FLTR_INSERTED, &fltr->base.state)) {
+		spin_unlock_bh(&bp->ntp_fltr_lock);
+		return;
+	}
+	hlist_del_rcu(&fltr->base.hash);
+	bnxt_del_one_usr_fltr(bp, &fltr->base);
+	bp->ntp_fltr_count--;
+	spin_unlock_bh(&bp->ntp_fltr_lock);
+	bnxt_del_l2_filter(bp, fltr->l2_fltr);
+	clear_bit(fltr->base.sw_id, bp->ntp_fltr_bmap);
+	kfree_rcu(fltr, base.rcu);
+}
 
 static void bnxt_cfg_ntp_filters(struct bnxt *bp)
 {
+#ifdef CONFIG_RFS_ACCEL
 	int i;
 
 	for (i = 0; i < BNXT_NTP_FLTR_HASH_SIZE; i++) {
@@ -7241,13 +15634,15 @@ static void bnxt_cfg_ntp_filters(struct bnxt *bp)
 		int rc;
 
 		head = &bp->ntp_fltr_hash_tbl[i];
-		hlist_for_each_entry_safe(fltr, tmp, head, hash) {
+		hlist_for_each_entry_safe(fltr, tmp, head, base.hash) {
 			bool del = false;
 
-			if (test_bit(BNXT_FLTR_VALID, &fltr->state)) {
-				if (rps_may_expire_flow(bp->dev, fltr->rxq,
+			if (test_bit(BNXT_FLTR_VALID, &fltr->base.state)) {
+				if (fltr->base.flags & BNXT_ACT_NO_AGING)
+					continue;
+				if (rps_may_expire_flow(bp->dev, fltr->base.rxq,
 							fltr->flow_id,
-							fltr->sw_id)) {
+							fltr->base.sw_id)) {
 					bnxt_hwrm_cfa_ntuple_filter_free(bp,
 									 fltr);
 					del = true;
@@ -7258,109 +15653,122 @@ static void bnxt_cfg_ntp_filters(struct bnxt *bp)
 				if (rc)
 					del = true;
 				else
-					set_bit(BNXT_FLTR_VALID, &fltr->state);
+					set_bit(BNXT_FLTR_VALID, &fltr->base.state);
 			}
 
-			if (del) {
-				spin_lock_bh(&bp->ntp_fltr_lock);
-				hlist_del_rcu(&fltr->hash);
-				bp->ntp_fltr_count--;
-				spin_unlock_bh(&bp->ntp_fltr_lock);
-				synchronize_rcu();
-				clear_bit(fltr->sw_id, bp->ntp_fltr_bmap);
-				kfree(fltr);
-			}
+			if (del)
+				bnxt_del_ntp_filter(bp, fltr);
 		}
 	}
-	if (test_and_clear_bit(BNXT_HWRM_PF_UNLOAD_SP_EVENT, &bp->sp_event))
-		netdev_info(bp->dev, "Receive PF driver unload event!");
+#endif
 }
 
-#else
-
-static void bnxt_cfg_ntp_filters(struct bnxt *bp)
+static int bnxt_udp_tunnel_set_port(struct net_device *netdev, unsigned int table,
+				    unsigned int entry, struct udp_tunnel_info *ti)
 {
-}
+	struct bnxt *bp = netdev_priv(netdev);
+	unsigned int cmd;
 
-#endif /* CONFIG_RFS_ACCEL */
+	if (ti->type == UDP_TUNNEL_TYPE_VXLAN)
+		cmd = TUNNEL_DST_PORT_ALLOC_REQ_TUNNEL_TYPE_VXLAN;
+	else if (ti->type == UDP_TUNNEL_TYPE_GENEVE)
+		cmd = TUNNEL_DST_PORT_ALLOC_REQ_TUNNEL_TYPE_GENEVE;
+	else
+		cmd = TUNNEL_DST_PORT_ALLOC_REQ_TUNNEL_TYPE_VXLAN_GPE;
 
-static void bnxt_udp_tunnel_add(struct net_device *dev,
-				struct udp_tunnel_info *ti)
-{
-	struct bnxt *bp = netdev_priv(dev);
+	return bnxt_hwrm_tunnel_dst_port_alloc(bp, ti->port, cmd);
+}
 
-	if (ti->sa_family != AF_INET6 && ti->sa_family != AF_INET)
-		return;
+static int bnxt_udp_tunnel_unset_port(struct net_device *netdev, unsigned int table,
+				      unsigned int entry, struct udp_tunnel_info *ti)
+{
+	struct bnxt *bp = netdev_priv(netdev);
+	unsigned int cmd;
 
-	if (!netif_running(dev))
-		return;
+	if (ti->type == UDP_TUNNEL_TYPE_VXLAN)
+		cmd = TUNNEL_DST_PORT_FREE_REQ_TUNNEL_TYPE_VXLAN;
+	else if (ti->type == UDP_TUNNEL_TYPE_GENEVE)
+		cmd = TUNNEL_DST_PORT_FREE_REQ_TUNNEL_TYPE_GENEVE;
+	else
+		cmd = TUNNEL_DST_PORT_FREE_REQ_TUNNEL_TYPE_VXLAN_GPE;
 
-	switch (ti->type) {
-	case UDP_TUNNEL_TYPE_VXLAN:
-		if (bp->vxlan_port_cnt && bp->vxlan_port != ti->port)
-			return;
+	return bnxt_hwrm_tunnel_dst_port_free(bp, cmd);
+}
 
-		bp->vxlan_port_cnt++;
-		if (bp->vxlan_port_cnt == 1) {
-			bp->vxlan_port = ti->port;
-			set_bit(BNXT_VXLAN_ADD_PORT_SP_EVENT, &bp->sp_event);
-			schedule_work(&bp->sp_task);
-		}
-		break;
-	case UDP_TUNNEL_TYPE_GENEVE:
-		if (bp->nge_port_cnt && bp->nge_port != ti->port)
-			return;
+static const struct udp_tunnel_nic_info bnxt_udp_tunnels = {
+	.set_port	= bnxt_udp_tunnel_set_port,
+	.unset_port	= bnxt_udp_tunnel_unset_port,
+	.flags		= UDP_TUNNEL_NIC_INFO_OPEN_ONLY,
+	.tables		= {
+		{ .n_entries = 1, .tunnel_types = UDP_TUNNEL_TYPE_VXLAN,  },
+		{ .n_entries = 1, .tunnel_types = UDP_TUNNEL_TYPE_GENEVE, },
+	},
+}, bnxt_udp_tunnels_p7 = {
+	.set_port	= bnxt_udp_tunnel_set_port,
+	.unset_port	= bnxt_udp_tunnel_unset_port,
+	.flags		= UDP_TUNNEL_NIC_INFO_OPEN_ONLY,
+	.tables		= {
+		{ .n_entries = 1, .tunnel_types = UDP_TUNNEL_TYPE_VXLAN,  },
+		{ .n_entries = 1, .tunnel_types = UDP_TUNNEL_TYPE_GENEVE, },
+		{ .n_entries = 1, .tunnel_types = UDP_TUNNEL_TYPE_VXLAN_GPE, },
+	},
+};
 
-		bp->nge_port_cnt++;
-		if (bp->nge_port_cnt == 1) {
-			bp->nge_port = ti->port;
-			set_bit(BNXT_GENEVE_ADD_PORT_SP_EVENT, &bp->sp_event);
-		}
-		break;
-	default:
-		return;
-	}
+static int bnxt_bridge_getlink(struct sk_buff *skb, u32 pid, u32 seq,
+			       struct net_device *dev, u32 filter_mask,
+			       int nlflags)
+{
+	struct bnxt *bp = netdev_priv(dev);
 
-	schedule_work(&bp->sp_task);
+	return ndo_dflt_bridge_getlink(skb, pid, seq, dev, bp->br_mode, 0, 0,
+				       nlflags, filter_mask, NULL);
 }
 
-static void bnxt_udp_tunnel_del(struct net_device *dev,
-				struct udp_tunnel_info *ti)
+static int bnxt_bridge_setlink(struct net_device *dev, struct nlmsghdr *nlh,
+			       u16 flags, struct netlink_ext_ack *extack)
 {
 	struct bnxt *bp = netdev_priv(dev);
+	struct nlattr *attr, *br_spec;
+	int rem, rc = 0;
 
-	if (ti->sa_family != AF_INET6 && ti->sa_family != AF_INET)
-		return;
+	if (bp->hwrm_spec_code < 0x10708 || !BNXT_SINGLE_PF(bp))
+		return -EOPNOTSUPP;
 
-	if (!netif_running(dev))
-		return;
+	br_spec = nlmsg_find_attr(nlh, sizeof(struct ifinfomsg), IFLA_AF_SPEC);
+	if (!br_spec)
+		return -EINVAL;
 
-	switch (ti->type) {
-	case UDP_TUNNEL_TYPE_VXLAN:
-		if (!bp->vxlan_port_cnt || bp->vxlan_port != ti->port)
-			return;
-		bp->vxlan_port_cnt--;
+	nla_for_each_nested_type(attr, IFLA_BRIDGE_MODE, br_spec, rem) {
+		u16 mode;
 
-		if (bp->vxlan_port_cnt != 0)
-			return;
+		mode = nla_get_u16(attr);
+		if (mode == bp->br_mode)
+			break;
 
-		set_bit(BNXT_VXLAN_DEL_PORT_SP_EVENT, &bp->sp_event);
+		rc = bnxt_hwrm_set_br_mode(bp, mode);
+		if (!rc)
+			bp->br_mode = mode;
 		break;
-	case UDP_TUNNEL_TYPE_GENEVE:
-		if (!bp->nge_port_cnt || bp->nge_port != ti->port)
-			return;
-		bp->nge_port_cnt--;
+	}
+	return rc;
+}
 
-		if (bp->nge_port_cnt != 0)
-			return;
+int bnxt_get_port_parent_id(struct net_device *dev,
+			    struct netdev_phys_item_id *ppid)
+{
+	struct bnxt *bp = netdev_priv(dev);
 
-		set_bit(BNXT_GENEVE_DEL_PORT_SP_EVENT, &bp->sp_event);
-		break;
-	default:
-		return;
-	}
+	if (bp->eswitch_mode != DEVLINK_ESWITCH_MODE_SWITCHDEV)
+		return -EOPNOTSUPP;
+
+	/* The PF and it's VF-reps only support the switchdev framework */
+	if (!BNXT_PF(bp) || !(bp->flags & BNXT_FLAG_DSN_VALID))
+		return -EOPNOTSUPP;
+
+	ppid->id_len = sizeof(bp->dsn);
+	memcpy(ppid->id, bp->dsn, ppid->id_len);
 
-	schedule_work(&bp->sp_task);
+	return 0;
 }
 
 static const struct net_device_ops bnxt_netdev_ops = {
@@ -7369,12 +15777,13 @@ static const struct net_device_ops bnxt_netdev_ops = {
 	.ndo_stop		= bnxt_close,
 	.ndo_get_stats64	= bnxt_get_stats64,
 	.ndo_set_rx_mode	= bnxt_set_rx_mode,
-	.ndo_do_ioctl		= bnxt_ioctl,
+	.ndo_eth_ioctl		= bnxt_ioctl,
 	.ndo_validate_addr	= eth_validate_addr,
 	.ndo_set_mac_address	= bnxt_change_mac_addr,
 	.ndo_change_mtu		= bnxt_change_mtu,
 	.ndo_fix_features	= bnxt_fix_features,
 	.ndo_set_features	= bnxt_set_features,
+	.ndo_features_check	= bnxt_features_check,
 	.ndo_tx_timeout		= bnxt_tx_timeout,
 #ifdef CONFIG_BNXT_SRIOV
 	.ndo_get_vf_config	= bnxt_get_vf_config,
@@ -7383,17 +15792,389 @@ static const struct net_device_ops bnxt_netdev_ops = {
 	.ndo_set_vf_rate	= bnxt_set_vf_bw,
 	.ndo_set_vf_link_state	= bnxt_set_vf_link_state,
 	.ndo_set_vf_spoofchk	= bnxt_set_vf_spoofchk,
-#endif
-#ifdef CONFIG_NET_POLL_CONTROLLER
-	.ndo_poll_controller	= bnxt_poll_controller,
+	.ndo_set_vf_trust	= bnxt_set_vf_trust,
 #endif
 	.ndo_setup_tc           = bnxt_setup_tc,
 #ifdef CONFIG_RFS_ACCEL
 	.ndo_rx_flow_steer	= bnxt_rx_flow_steer,
 #endif
-	.ndo_udp_tunnel_add	= bnxt_udp_tunnel_add,
-	.ndo_udp_tunnel_del	= bnxt_udp_tunnel_del,
-	.ndo_xdp		= bnxt_xdp,
+	.ndo_bpf		= bnxt_xdp,
+	.ndo_xdp_xmit		= bnxt_xdp_xmit,
+	.ndo_bridge_getlink	= bnxt_bridge_getlink,
+	.ndo_bridge_setlink	= bnxt_bridge_setlink,
+	.ndo_hwtstamp_get	= bnxt_hwtstamp_get,
+	.ndo_hwtstamp_set	= bnxt_hwtstamp_set,
+};
+
+static void bnxt_get_queue_stats_rx(struct net_device *dev, int i,
+				    struct netdev_queue_stats_rx *stats)
+{
+	struct bnxt *bp = netdev_priv(dev);
+	struct bnxt_cp_ring_info *cpr;
+	u64 *sw;
+
+	if (!bp->bnapi)
+		return;
+
+	cpr = &bp->bnapi[i]->cp_ring;
+	sw = cpr->stats.sw_stats;
+
+	stats->packets = 0;
+	stats->packets += BNXT_GET_RING_STATS64(sw, rx_ucast_pkts);
+	stats->packets += BNXT_GET_RING_STATS64(sw, rx_mcast_pkts);
+	stats->packets += BNXT_GET_RING_STATS64(sw, rx_bcast_pkts);
+
+	stats->bytes = 0;
+	stats->bytes += BNXT_GET_RING_STATS64(sw, rx_ucast_bytes);
+	stats->bytes += BNXT_GET_RING_STATS64(sw, rx_mcast_bytes);
+	stats->bytes += BNXT_GET_RING_STATS64(sw, rx_bcast_bytes);
+
+	stats->alloc_fail = cpr->sw_stats->rx.rx_oom_discards;
+}
+
+static void bnxt_get_queue_stats_tx(struct net_device *dev, int i,
+				    struct netdev_queue_stats_tx *stats)
+{
+	struct bnxt *bp = netdev_priv(dev);
+	struct bnxt_napi *bnapi;
+	u64 *sw;
+
+	if (!bp->tx_ring)
+		return;
+
+	bnapi = bp->tx_ring[bp->tx_ring_map[i]].bnapi;
+	sw = bnapi->cp_ring.stats.sw_stats;
+
+	stats->packets = 0;
+	stats->packets += BNXT_GET_RING_STATS64(sw, tx_ucast_pkts);
+	stats->packets += BNXT_GET_RING_STATS64(sw, tx_mcast_pkts);
+	stats->packets += BNXT_GET_RING_STATS64(sw, tx_bcast_pkts);
+
+	stats->bytes = 0;
+	stats->bytes += BNXT_GET_RING_STATS64(sw, tx_ucast_bytes);
+	stats->bytes += BNXT_GET_RING_STATS64(sw, tx_mcast_bytes);
+	stats->bytes += BNXT_GET_RING_STATS64(sw, tx_bcast_bytes);
+}
+
+static void bnxt_get_base_stats(struct net_device *dev,
+				struct netdev_queue_stats_rx *rx,
+				struct netdev_queue_stats_tx *tx)
+{
+	struct bnxt *bp = netdev_priv(dev);
+
+	rx->packets = bp->net_stats_prev.rx_packets;
+	rx->bytes = bp->net_stats_prev.rx_bytes;
+	rx->alloc_fail = bp->ring_err_stats_prev.rx_total_oom_discards;
+
+	tx->packets = bp->net_stats_prev.tx_packets;
+	tx->bytes = bp->net_stats_prev.tx_bytes;
+}
+
+static const struct netdev_stat_ops bnxt_stat_ops = {
+	.get_queue_stats_rx	= bnxt_get_queue_stats_rx,
+	.get_queue_stats_tx	= bnxt_get_queue_stats_tx,
+	.get_base_stats		= bnxt_get_base_stats,
+};
+
+static int bnxt_queue_mem_alloc(struct net_device *dev, void *qmem, int idx)
+{
+	struct bnxt_rx_ring_info *rxr, *clone;
+	struct bnxt *bp = netdev_priv(dev);
+	struct bnxt_ring_struct *ring;
+	int rc;
+
+	if (!bp->rx_ring)
+		return -ENETDOWN;
+
+	rxr = &bp->rx_ring[idx];
+	clone = qmem;
+	memcpy(clone, rxr, sizeof(*rxr));
+	bnxt_init_rx_ring_struct(bp, clone);
+	bnxt_reset_rx_ring_struct(bp, clone);
+
+	clone->rx_prod = 0;
+	clone->rx_agg_prod = 0;
+	clone->rx_sw_agg_prod = 0;
+	clone->rx_next_cons = 0;
+	clone->need_head_pool = false;
+
+	rc = bnxt_alloc_rx_page_pool(bp, clone, rxr->page_pool->p.nid);
+	if (rc)
+		return rc;
+
+	rc = xdp_rxq_info_reg(&clone->xdp_rxq, bp->dev, idx, 0);
+	if (rc < 0)
+		goto err_page_pool_destroy;
+
+	rc = xdp_rxq_info_reg_mem_model(&clone->xdp_rxq,
+					MEM_TYPE_PAGE_POOL,
+					clone->page_pool);
+	if (rc)
+		goto err_rxq_info_unreg;
+
+	ring = &clone->rx_ring_struct;
+	rc = bnxt_alloc_ring(bp, &ring->ring_mem);
+	if (rc)
+		goto err_free_rx_ring;
+
+	if (bp->flags & BNXT_FLAG_AGG_RINGS) {
+		ring = &clone->rx_agg_ring_struct;
+		rc = bnxt_alloc_ring(bp, &ring->ring_mem);
+		if (rc)
+			goto err_free_rx_agg_ring;
+
+		rc = bnxt_alloc_rx_agg_bmap(bp, clone);
+		if (rc)
+			goto err_free_rx_agg_ring;
+	}
+
+	if (bp->flags & BNXT_FLAG_TPA) {
+		rc = bnxt_alloc_one_tpa_info(bp, clone);
+		if (rc)
+			goto err_free_tpa_info;
+	}
+
+	bnxt_init_one_rx_ring_rxbd(bp, clone);
+	bnxt_init_one_rx_agg_ring_rxbd(bp, clone);
+
+	bnxt_alloc_one_rx_ring_skb(bp, clone, idx);
+	if (bp->flags & BNXT_FLAG_AGG_RINGS)
+		bnxt_alloc_one_rx_ring_netmem(bp, clone, idx);
+	if (bp->flags & BNXT_FLAG_TPA)
+		bnxt_alloc_one_tpa_info_data(bp, clone);
+
+	return 0;
+
+err_free_tpa_info:
+	bnxt_free_one_tpa_info(bp, clone);
+err_free_rx_agg_ring:
+	bnxt_free_ring(bp, &clone->rx_agg_ring_struct.ring_mem);
+err_free_rx_ring:
+	bnxt_free_ring(bp, &clone->rx_ring_struct.ring_mem);
+err_rxq_info_unreg:
+	xdp_rxq_info_unreg(&clone->xdp_rxq);
+err_page_pool_destroy:
+	page_pool_destroy(clone->page_pool);
+	page_pool_destroy(clone->head_pool);
+	clone->page_pool = NULL;
+	clone->head_pool = NULL;
+	return rc;
+}
+
+static void bnxt_queue_mem_free(struct net_device *dev, void *qmem)
+{
+	struct bnxt_rx_ring_info *rxr = qmem;
+	struct bnxt *bp = netdev_priv(dev);
+	struct bnxt_ring_struct *ring;
+
+	bnxt_free_one_rx_ring_skbs(bp, rxr);
+	bnxt_free_one_tpa_info(bp, rxr);
+
+	xdp_rxq_info_unreg(&rxr->xdp_rxq);
+
+	page_pool_destroy(rxr->page_pool);
+	page_pool_destroy(rxr->head_pool);
+	rxr->page_pool = NULL;
+	rxr->head_pool = NULL;
+
+	ring = &rxr->rx_ring_struct;
+	bnxt_free_ring(bp, &ring->ring_mem);
+
+	ring = &rxr->rx_agg_ring_struct;
+	bnxt_free_ring(bp, &ring->ring_mem);
+
+	kfree(rxr->rx_agg_bmap);
+	rxr->rx_agg_bmap = NULL;
+}
+
+static void bnxt_copy_rx_ring(struct bnxt *bp,
+			      struct bnxt_rx_ring_info *dst,
+			      struct bnxt_rx_ring_info *src)
+{
+	struct bnxt_ring_mem_info *dst_rmem, *src_rmem;
+	struct bnxt_ring_struct *dst_ring, *src_ring;
+	int i;
+
+	dst_ring = &dst->rx_ring_struct;
+	dst_rmem = &dst_ring->ring_mem;
+	src_ring = &src->rx_ring_struct;
+	src_rmem = &src_ring->ring_mem;
+
+	WARN_ON(dst_rmem->nr_pages != src_rmem->nr_pages);
+	WARN_ON(dst_rmem->page_size != src_rmem->page_size);
+	WARN_ON(dst_rmem->flags != src_rmem->flags);
+	WARN_ON(dst_rmem->depth != src_rmem->depth);
+	WARN_ON(dst_rmem->vmem_size != src_rmem->vmem_size);
+	WARN_ON(dst_rmem->ctx_mem != src_rmem->ctx_mem);
+
+	dst_rmem->pg_tbl = src_rmem->pg_tbl;
+	dst_rmem->pg_tbl_map = src_rmem->pg_tbl_map;
+	*dst_rmem->vmem = *src_rmem->vmem;
+	for (i = 0; i < dst_rmem->nr_pages; i++) {
+		dst_rmem->pg_arr[i] = src_rmem->pg_arr[i];
+		dst_rmem->dma_arr[i] = src_rmem->dma_arr[i];
+	}
+
+	if (!(bp->flags & BNXT_FLAG_AGG_RINGS))
+		return;
+
+	dst_ring = &dst->rx_agg_ring_struct;
+	dst_rmem = &dst_ring->ring_mem;
+	src_ring = &src->rx_agg_ring_struct;
+	src_rmem = &src_ring->ring_mem;
+
+	WARN_ON(dst_rmem->nr_pages != src_rmem->nr_pages);
+	WARN_ON(dst_rmem->page_size != src_rmem->page_size);
+	WARN_ON(dst_rmem->flags != src_rmem->flags);
+	WARN_ON(dst_rmem->depth != src_rmem->depth);
+	WARN_ON(dst_rmem->vmem_size != src_rmem->vmem_size);
+	WARN_ON(dst_rmem->ctx_mem != src_rmem->ctx_mem);
+	WARN_ON(dst->rx_agg_bmap_size != src->rx_agg_bmap_size);
+
+	dst_rmem->pg_tbl = src_rmem->pg_tbl;
+	dst_rmem->pg_tbl_map = src_rmem->pg_tbl_map;
+	*dst_rmem->vmem = *src_rmem->vmem;
+	for (i = 0; i < dst_rmem->nr_pages; i++) {
+		dst_rmem->pg_arr[i] = src_rmem->pg_arr[i];
+		dst_rmem->dma_arr[i] = src_rmem->dma_arr[i];
+	}
+
+	dst->rx_agg_bmap = src->rx_agg_bmap;
+}
+
+static int bnxt_queue_start(struct net_device *dev, void *qmem, int idx)
+{
+	struct bnxt *bp = netdev_priv(dev);
+	struct bnxt_rx_ring_info *rxr, *clone;
+	struct bnxt_cp_ring_info *cpr;
+	struct bnxt_vnic_info *vnic;
+	struct bnxt_napi *bnapi;
+	int i, rc;
+	u16 mru;
+
+	rxr = &bp->rx_ring[idx];
+	clone = qmem;
+
+	rxr->rx_prod = clone->rx_prod;
+	rxr->rx_agg_prod = clone->rx_agg_prod;
+	rxr->rx_sw_agg_prod = clone->rx_sw_agg_prod;
+	rxr->rx_next_cons = clone->rx_next_cons;
+	rxr->rx_tpa = clone->rx_tpa;
+	rxr->rx_tpa_idx_map = clone->rx_tpa_idx_map;
+	rxr->page_pool = clone->page_pool;
+	rxr->head_pool = clone->head_pool;
+	rxr->xdp_rxq = clone->xdp_rxq;
+	rxr->need_head_pool = clone->need_head_pool;
+
+	bnxt_copy_rx_ring(bp, rxr, clone);
+
+	bnapi = rxr->bnapi;
+	cpr = &bnapi->cp_ring;
+
+	/* All rings have been reserved and previously allocated.
+	 * Reallocating with the same parameters should never fail.
+	 */
+	rc = bnxt_hwrm_rx_ring_alloc(bp, rxr);
+	if (rc)
+		goto err_reset;
+
+	if (bp->tph_mode) {
+		rc = bnxt_hwrm_cp_ring_alloc_p5(bp, rxr->rx_cpr);
+		if (rc)
+			goto err_reset;
+	}
+
+	rc = bnxt_hwrm_rx_agg_ring_alloc(bp, rxr);
+	if (rc)
+		goto err_reset;
+
+	bnxt_db_write(bp, &rxr->rx_db, rxr->rx_prod);
+	if (bp->flags & BNXT_FLAG_AGG_RINGS)
+		bnxt_db_write(bp, &rxr->rx_agg_db, rxr->rx_agg_prod);
+
+	if (bp->flags & BNXT_FLAG_SHARED_RINGS) {
+		rc = bnxt_tx_queue_start(bp, idx);
+		if (rc)
+			goto err_reset;
+	}
+
+	bnxt_enable_rx_page_pool(rxr);
+	napi_enable_locked(&bnapi->napi);
+	bnxt_db_nq_arm(bp, &cpr->cp_db, cpr->cp_raw_cons);
+
+	mru = bp->dev->mtu + VLAN_ETH_HLEN;
+	for (i = 0; i < bp->nr_vnics; i++) {
+		vnic = &bp->vnic_info[i];
+
+		rc = bnxt_set_vnic_mru_p5(bp, vnic, mru, idx);
+		if (rc)
+			return rc;
+	}
+	return bnxt_set_rss_ctx_vnic_mru(bp, mru, idx);
+
+err_reset:
+	netdev_err(bp->dev, "Unexpected HWRM error during queue start rc: %d\n",
+		   rc);
+	napi_enable_locked(&bnapi->napi);
+	bnxt_db_nq_arm(bp, &cpr->cp_db, cpr->cp_raw_cons);
+	bnxt_reset_task(bp, true);
+	return rc;
+}
+
+static int bnxt_queue_stop(struct net_device *dev, void *qmem, int idx)
+{
+	struct bnxt *bp = netdev_priv(dev);
+	struct bnxt_rx_ring_info *rxr;
+	struct bnxt_cp_ring_info *cpr;
+	struct bnxt_vnic_info *vnic;
+	struct bnxt_napi *bnapi;
+	int i;
+
+	for (i = 0; i < bp->nr_vnics; i++) {
+		vnic = &bp->vnic_info[i];
+
+		bnxt_set_vnic_mru_p5(bp, vnic, 0, idx);
+	}
+	bnxt_set_rss_ctx_vnic_mru(bp, 0, idx);
+	/* Make sure NAPI sees that the VNIC is disabled */
+	synchronize_net();
+	rxr = &bp->rx_ring[idx];
+	bnapi = rxr->bnapi;
+	cpr = &bnapi->cp_ring;
+	cancel_work_sync(&cpr->dim.work);
+	bnxt_hwrm_rx_ring_free(bp, rxr, false);
+	bnxt_hwrm_rx_agg_ring_free(bp, rxr, false);
+	page_pool_disable_direct_recycling(rxr->page_pool);
+	if (bnxt_separate_head_pool(rxr))
+		page_pool_disable_direct_recycling(rxr->head_pool);
+
+	if (bp->flags & BNXT_FLAG_SHARED_RINGS)
+		bnxt_tx_queue_stop(bp, idx);
+
+	/* Disable NAPI now after freeing the rings because HWRM_RING_FREE
+	 * completion is handled in NAPI to guarantee no more DMA on that ring
+	 * after seeing the completion.
+	 */
+	napi_disable_locked(&bnapi->napi);
+
+	if (bp->tph_mode) {
+		bnxt_hwrm_cp_ring_free(bp, rxr->rx_cpr);
+		bnxt_clear_one_cp_ring(bp, rxr->rx_cpr);
+	}
+	bnxt_db_nq(bp, &cpr->cp_db, cpr->cp_raw_cons);
+
+	memcpy(qmem, rxr, sizeof(*rxr));
+	bnxt_init_rx_ring_struct(bp, qmem);
+
+	return 0;
+}
+
+static const struct netdev_queue_mgmt_ops bnxt_queue_mgmt_ops = {
+	.ndo_queue_mem_size	= sizeof(struct bnxt_rx_ring_info),
+	.ndo_queue_mem_alloc	= bnxt_queue_mem_alloc,
+	.ndo_queue_mem_free	= bnxt_queue_mem_free,
+	.ndo_queue_start	= bnxt_queue_start,
+	.ndo_queue_stop		= bnxt_queue_stop,
 };
 
 static void bnxt_remove_one(struct pci_dev *pdev)
@@ -7402,41 +16183,74 @@ static void bnxt_remove_one(struct pci_dev *pdev)
 	struct bnxt *bp = netdev_priv(dev);
 
 	if (BNXT_PF(bp))
-		bnxt_sriov_disable(bp);
+		__bnxt_sriov_disable(bp);
+
+	bnxt_rdma_aux_device_del(bp);
 
-	pci_disable_pcie_error_reporting(pdev);
 	unregister_netdev(dev);
+	bnxt_ptp_clear(bp);
+
+	bnxt_rdma_aux_device_uninit(bp);
+
+	bnxt_free_l2_filters(bp, true);
+	bnxt_free_ntp_fltrs(bp, true);
+	WARN_ON(bp->num_rss_ctx);
+	clear_bit(BNXT_STATE_IN_FW_RESET, &bp->state);
+	/* Flush any pending tasks */
 	cancel_work_sync(&bp->sp_task);
+	cancel_delayed_work_sync(&bp->fw_reset_task);
 	bp->sp_event = 0;
 
+	bnxt_dl_fw_reporters_destroy(bp);
+	bnxt_dl_unregister(bp);
+	bnxt_shutdown_tc(bp);
+
 	bnxt_clear_int_mode(bp);
 	bnxt_hwrm_func_drv_unrgtr(bp);
 	bnxt_free_hwrm_resources(bp);
-	bnxt_free_hwrm_short_cmd_req(bp);
+	bnxt_hwmon_uninit(bp);
 	bnxt_ethtool_free(bp);
 	bnxt_dcb_free(bp);
-	kfree(bp->edev);
-	bp->edev = NULL;
-	if (bp->xdp_prog)
-		bpf_prog_put(bp->xdp_prog);
+	kfree(bp->ptp_cfg);
+	bp->ptp_cfg = NULL;
+	kfree(bp->fw_health);
+	bp->fw_health = NULL;
 	bnxt_cleanup_pci(bp);
+	bnxt_free_ctx_mem(bp, true);
+	bnxt_free_crash_dump_mem(bp);
+	kfree(bp->rss_indir_tbl);
+	bp->rss_indir_tbl = NULL;
+	bnxt_free_port_stats(bp);
 	free_netdev(dev);
 }
 
-static int bnxt_probe_phy(struct bnxt *bp)
+static int bnxt_probe_phy(struct bnxt *bp, bool fw_dflt)
 {
 	int rc = 0;
 	struct bnxt_link_info *link_info = &bp->link_info;
 
+	bp->phy_flags = 0;
 	rc = bnxt_hwrm_phy_qcaps(bp);
 	if (rc) {
 		netdev_err(bp->dev, "Probe phy can't get phy capabilities (rc: %x)\n",
 			   rc);
 		return rc;
 	}
+	if (bp->phy_flags & BNXT_PHY_FL_NO_FCS)
+		bp->dev->priv_flags |= IFF_SUPP_NOFCS;
+	else
+		bp->dev->priv_flags &= ~IFF_SUPP_NOFCS;
+
+	bp->mac_flags = 0;
+	bnxt_hwrm_mac_qcaps(bp);
+
+	if (!fw_dflt)
+		return 0;
 
+	mutex_lock(&bp->link_lock);
 	rc = bnxt_update_link(bp, false);
 	if (rc) {
+		mutex_unlock(&bp->link_lock);
 		netdev_err(bp->dev, "Probe phy can't update link (rc: %x)\n",
 			   rc);
 		return rc;
@@ -7448,27 +16262,9 @@ static int bnxt_probe_phy(struct bnxt *bp)
 	if (link_info->auto_link_speeds && !link_info->support_auto_speeds)
 		link_info->support_auto_speeds = link_info->support_speeds;
 
-	/*initialize the ethool setting copy with NVM settings */
-	if (BNXT_AUTO_MODE(link_info->auto_mode)) {
-		link_info->autoneg = BNXT_AUTONEG_SPEED;
-		if (bp->hwrm_spec_code >= 0x10201) {
-			if (link_info->auto_pause_setting &
-			    PORT_PHY_CFG_REQ_AUTO_PAUSE_AUTONEG_PAUSE)
-				link_info->autoneg |= BNXT_AUTONEG_FLOW_CTRL;
-		} else {
-			link_info->autoneg |= BNXT_AUTONEG_FLOW_CTRL;
-		}
-		link_info->advertising = link_info->auto_link_speeds;
-	} else {
-		link_info->req_link_speed = link_info->force_link_speed;
-		link_info->req_duplex = link_info->duplex_setting;
-	}
-	if (link_info->autoneg & BNXT_AUTONEG_FLOW_CTRL)
-		link_info->req_flow_ctrl =
-			link_info->auto_pause_setting & BNXT_LINK_PAUSE_BOTH;
-	else
-		link_info->req_flow_ctrl = link_info->force_pause_setting;
-	return rc;
+	bnxt_init_ethtool_link_settings(bp);
+	mutex_unlock(&bp->link_lock);
+	return 0;
 }
 
 static int bnxt_get_max_irq(struct pci_dev *pdev)
@@ -7485,30 +16281,36 @@ static int bnxt_get_max_irq(struct pci_dev *pdev)
 static void _bnxt_get_max_rings(struct bnxt *bp, int *max_rx, int *max_tx,
 				int *max_cp)
 {
-	int max_ring_grps = 0;
-
-#ifdef CONFIG_BNXT_SRIOV
-	if (!BNXT_PF(bp)) {
-		*max_tx = bp->vf.max_tx_rings;
-		*max_rx = bp->vf.max_rx_rings;
-		*max_cp = min_t(int, bp->vf.max_irqs, bp->vf.max_cp_rings);
-		*max_cp = min_t(int, *max_cp, bp->vf.max_stat_ctxs);
-		max_ring_grps = bp->vf.max_hw_ring_grps;
-	} else
-#endif
-	{
-		*max_tx = bp->pf.max_tx_rings;
-		*max_rx = bp->pf.max_rx_rings;
-		*max_cp = min_t(int, bp->pf.max_irqs, bp->pf.max_cp_rings);
-		*max_cp = min_t(int, *max_cp, bp->pf.max_stat_ctxs);
-		max_ring_grps = bp->pf.max_hw_ring_grps;
-	}
+	struct bnxt_hw_resc *hw_resc = &bp->hw_resc;
+	int max_ring_grps = 0, max_irq;
+
+	*max_tx = hw_resc->max_tx_rings;
+	*max_rx = hw_resc->max_rx_rings;
+	*max_cp = bnxt_get_max_func_cp_rings_for_en(bp);
+	max_irq = min_t(int, bnxt_get_max_func_irqs(bp) -
+			bnxt_get_ulp_msix_num_in_use(bp),
+			hw_resc->max_stat_ctxs -
+			bnxt_get_ulp_stat_ctxs_in_use(bp));
+	if (!(bp->flags & BNXT_FLAG_CHIP_P5_PLUS))
+		*max_cp = min_t(int, *max_cp, max_irq);
+	max_ring_grps = hw_resc->max_hw_ring_grps;
 	if (BNXT_CHIP_TYPE_NITRO_A0(bp) && BNXT_PF(bp)) {
 		*max_cp -= 1;
 		*max_rx -= 2;
 	}
 	if (bp->flags & BNXT_FLAG_AGG_RINGS)
 		*max_rx >>= 1;
+	if (bp->flags & BNXT_FLAG_CHIP_P5_PLUS) {
+		int rc;
+
+		rc = __bnxt_trim_rings(bp, max_rx, max_tx, *max_cp, false);
+		if (rc) {
+			*max_rx = 0;
+			*max_tx = 0;
+		}
+		/* On P5 chips, max_cp output param should be available NQs */
+		*max_cp = max_irq;
+	}
 	*max_rx = min_t(int, *max_rx, max_ring_grps);
 }
 
@@ -7517,11 +16319,11 @@ int bnxt_get_max_rings(struct bnxt *bp, int *max_rx, int *max_tx, bool shared)
 	int rx, tx, cp;
 
 	_bnxt_get_max_rings(bp, &rx, &tx, &cp);
+	*max_rx = rx;
+	*max_tx = tx;
 	if (!rx || !tx || !cp)
 		return -ENOMEM;
 
-	*max_rx = rx;
-	*max_tx = tx;
 	return bnxt_trim_rings(bp, max_rx, max_tx, cp, shared);
 }
 
@@ -7535,11 +16337,14 @@ static int bnxt_get_dflt_rings(struct bnxt *bp, int *max_rx, int *max_tx,
 		/* Not enough rings, try disabling agg rings. */
 		bp->flags &= ~BNXT_FLAG_AGG_RINGS;
 		rc = bnxt_get_max_rings(bp, max_rx, max_tx, shared);
-		if (rc)
+		if (rc) {
+			/* set BNXT_FLAG_AGG_RINGS back for consistency */
+			bp->flags |= BNXT_FLAG_AGG_RINGS;
 			return rc;
+		}
 		bp->flags |= BNXT_FLAG_NO_AGG_RINGS;
-		bp->dev->hw_features &= ~NETIF_F_LRO;
-		bp->dev->features &= ~NETIF_F_LRO;
+		bp->dev->hw_features &= ~(NETIF_F_LRO | NETIF_F_GRO_HW);
+		bp->dev->features &= ~(NETIF_F_LRO | NETIF_F_GRO_HW);
 		bnxt_set_ring_params(bp);
 	}
 
@@ -7567,60 +16372,236 @@ static int bnxt_get_dflt_rings(struct bnxt *bp, int *max_rx, int *max_tx,
 	return rc;
 }
 
+/* In initial default shared ring setting, each shared ring must have a
+ * RX/TX ring pair.
+ */
+static void bnxt_trim_dflt_sh_rings(struct bnxt *bp)
+{
+	bp->cp_nr_rings = min_t(int, bp->tx_nr_rings_per_tc, bp->rx_nr_rings);
+	bp->rx_nr_rings = bp->cp_nr_rings;
+	bp->tx_nr_rings_per_tc = bp->cp_nr_rings;
+	bp->tx_nr_rings = bnxt_tx_nr_rings(bp);
+}
+
 static int bnxt_set_dflt_rings(struct bnxt *bp, bool sh)
 {
 	int dflt_rings, max_rx_rings, max_tx_rings, rc;
+	int avail_msix;
+
+	if (!bnxt_can_reserve_rings(bp))
+		return 0;
 
 	if (sh)
 		bp->flags |= BNXT_FLAG_SHARED_RINGS;
-	dflt_rings = netif_get_num_default_rss_queues();
+	dflt_rings = is_kdump_kernel() ? 1 : netif_get_num_default_rss_queues();
+	/* Reduce default rings on multi-port cards so that total default
+	 * rings do not exceed CPU count.
+	 */
+	if (bp->port_count > 1) {
+		int max_rings =
+			max_t(int, num_online_cpus() / bp->port_count, 1);
+
+		dflt_rings = min_t(int, dflt_rings, max_rings);
+	}
 	rc = bnxt_get_dflt_rings(bp, &max_rx_rings, &max_tx_rings, sh);
 	if (rc)
 		return rc;
 	bp->rx_nr_rings = min_t(int, dflt_rings, max_rx_rings);
 	bp->tx_nr_rings_per_tc = min_t(int, dflt_rings, max_tx_rings);
+	if (sh)
+		bnxt_trim_dflt_sh_rings(bp);
+	else
+		bp->cp_nr_rings = bp->tx_nr_rings_per_tc + bp->rx_nr_rings;
+	bp->tx_nr_rings = bnxt_tx_nr_rings(bp);
 
-	rc = bnxt_hwrm_reserve_tx_rings(bp, &bp->tx_nr_rings_per_tc);
-	if (rc)
+	avail_msix = bnxt_get_max_func_irqs(bp) - bp->cp_nr_rings;
+	if (avail_msix >= BNXT_MIN_ROCE_CP_RINGS) {
+		int ulp_num_msix = min(avail_msix, bp->ulp_num_msix_want);
+
+		bnxt_set_ulp_msix_num(bp, ulp_num_msix);
+		bnxt_set_dflt_ulp_stat_ctxs(bp);
+	}
+
+	rc = __bnxt_reserve_rings(bp);
+	if (rc && rc != -ENODEV)
 		netdev_warn(bp->dev, "Unable to reserve tx rings\n");
+	bp->tx_nr_rings_per_tc = bnxt_tx_nr_rings_per_tc(bp);
+	if (sh)
+		bnxt_trim_dflt_sh_rings(bp);
 
-	bp->tx_nr_rings = bp->tx_nr_rings_per_tc;
-	bp->cp_nr_rings = sh ? max_t(int, bp->tx_nr_rings, bp->rx_nr_rings) :
-			       bp->tx_nr_rings + bp->rx_nr_rings;
-	bp->num_stat_ctxs = bp->cp_nr_rings;
+	/* Rings may have been trimmed, re-reserve the trimmed rings. */
+	if (bnxt_need_reserve_rings(bp)) {
+		rc = __bnxt_reserve_rings(bp);
+		if (rc && rc != -ENODEV)
+			netdev_warn(bp->dev, "2nd rings reservation failed.\n");
+		bp->tx_nr_rings_per_tc = bnxt_tx_nr_rings_per_tc(bp);
+	}
 	if (BNXT_CHIP_TYPE_NITRO_A0(bp)) {
 		bp->rx_nr_rings++;
 		bp->cp_nr_rings++;
 	}
+	if (rc) {
+		bp->tx_nr_rings = 0;
+		bp->rx_nr_rings = 0;
+	}
 	return rc;
 }
 
-void bnxt_restore_pf_fw_resources(struct bnxt *bp)
+static int bnxt_init_dflt_ring_mode(struct bnxt *bp)
 {
-	ASSERT_RTNL();
+	int rc;
+
+	if (bp->tx_nr_rings)
+		return 0;
+
+	bnxt_ulp_irq_stop(bp);
+	bnxt_clear_int_mode(bp);
+	rc = bnxt_set_dflt_rings(bp, true);
+	if (rc) {
+		if (BNXT_VF(bp) && rc == -ENODEV)
+			netdev_err(bp->dev, "Cannot configure VF rings while PF is unavailable.\n");
+		else
+			netdev_err(bp->dev, "Not enough rings available.\n");
+		goto init_dflt_ring_err;
+	}
+	rc = bnxt_init_int_mode(bp);
+	if (rc)
+		goto init_dflt_ring_err;
+
+	bp->tx_nr_rings_per_tc = bnxt_tx_nr_rings_per_tc(bp);
+
+	bnxt_set_dflt_rfs(bp);
+
+init_dflt_ring_err:
+	bnxt_ulp_irq_restart(bp, rc);
+	return rc;
+}
+
+int bnxt_restore_pf_fw_resources(struct bnxt *bp)
+{
+	int rc;
+
+	netdev_ops_assert_locked(bp->dev);
 	bnxt_hwrm_func_qcaps(bp);
-	bnxt_subtract_ulp_resources(bp, BNXT_ROCE_ULP);
+
+	if (netif_running(bp->dev))
+		__bnxt_close_nic(bp, true, false);
+
+	bnxt_ulp_irq_stop(bp);
+	bnxt_clear_int_mode(bp);
+	rc = bnxt_init_int_mode(bp);
+	bnxt_ulp_irq_restart(bp, rc);
+
+	if (netif_running(bp->dev)) {
+		if (rc)
+			netif_close(bp->dev);
+		else
+			rc = bnxt_open_nic(bp, true, false);
+	}
+
+	return rc;
 }
 
-static void bnxt_parse_log_pcie_link(struct bnxt *bp)
+static int bnxt_init_mac_addr(struct bnxt *bp)
 {
-	enum pcie_link_width width = PCIE_LNK_WIDTH_UNKNOWN;
-	enum pci_bus_speed speed = PCI_SPEED_UNKNOWN;
+	int rc = 0;
 
-	if (pcie_get_minimum_link(bp->pdev, &speed, &width) ||
-	    speed == PCI_SPEED_UNKNOWN || width == PCIE_LNK_WIDTH_UNKNOWN)
-		netdev_info(bp->dev, "Failed to determine PCIe Link Info\n");
-	else
-		netdev_info(bp->dev, "PCIe: Speed %s Width x%d\n",
-			    speed == PCIE_SPEED_2_5GT ? "2.5GT/s" :
-			    speed == PCIE_SPEED_5_0GT ? "5.0GT/s" :
-			    speed == PCIE_SPEED_8_0GT ? "8.0GT/s" :
-			    "Unknown", width);
+	if (BNXT_PF(bp)) {
+		eth_hw_addr_set(bp->dev, bp->pf.mac_addr);
+	} else {
+#ifdef CONFIG_BNXT_SRIOV
+		struct bnxt_vf_info *vf = &bp->vf;
+		bool strict_approval = true;
+
+		if (is_valid_ether_addr(vf->mac_addr)) {
+			/* overwrite netdev dev_addr with admin VF MAC */
+			eth_hw_addr_set(bp->dev, vf->mac_addr);
+			/* Older PF driver or firmware may not approve this
+			 * correctly.
+			 */
+			strict_approval = false;
+		} else {
+			eth_hw_addr_random(bp->dev);
+		}
+		rc = bnxt_approve_mac(bp, bp->dev->dev_addr, strict_approval);
+#endif
+	}
+	return rc;
+}
+
+static void bnxt_vpd_read_info(struct bnxt *bp)
+{
+	struct pci_dev *pdev = bp->pdev;
+	unsigned int vpd_size, kw_len;
+	int pos, size;
+	u8 *vpd_data;
+
+	vpd_data = pci_vpd_alloc(pdev, &vpd_size);
+	if (IS_ERR(vpd_data)) {
+		pci_warn(pdev, "Unable to read VPD\n");
+		return;
+	}
+
+	pos = pci_vpd_find_ro_info_keyword(vpd_data, vpd_size,
+					   PCI_VPD_RO_KEYWORD_PARTNO, &kw_len);
+	if (pos < 0)
+		goto read_sn;
+
+	size = min_t(int, kw_len, BNXT_VPD_FLD_LEN - 1);
+	memcpy(bp->board_partno, &vpd_data[pos], size);
+
+read_sn:
+	pos = pci_vpd_find_ro_info_keyword(vpd_data, vpd_size,
+					   PCI_VPD_RO_KEYWORD_SERIALNO,
+					   &kw_len);
+	if (pos < 0)
+		goto exit;
+
+	size = min_t(int, kw_len, BNXT_VPD_FLD_LEN - 1);
+	memcpy(bp->board_serialno, &vpd_data[pos], size);
+exit:
+	kfree(vpd_data);
+}
+
+static int bnxt_pcie_dsn_get(struct bnxt *bp, u8 dsn[])
+{
+	struct pci_dev *pdev = bp->pdev;
+	u64 qword;
+
+	qword = pci_get_dsn(pdev);
+	if (!qword) {
+		netdev_info(bp->dev, "Unable to read adapter's DSN\n");
+		return -EOPNOTSUPP;
+	}
+
+	put_unaligned_le64(qword, dsn);
+
+	bp->flags |= BNXT_FLAG_DSN_VALID;
+	return 0;
+}
+
+static int bnxt_map_db_bar(struct bnxt *bp)
+{
+	if (!bp->db_size)
+		return -ENODEV;
+	bp->bar1 = pci_iomap(bp->pdev, 2, bp->db_size);
+	if (!bp->bar1)
+		return -ENOMEM;
+	return 0;
+}
+
+void bnxt_print_device_info(struct bnxt *bp)
+{
+	netdev_info(bp->dev, "%s found at mem %lx, node addr %pM\n",
+		    board_info[bp->board_idx].name,
+		    (long)pci_resource_start(bp->pdev, 0), bp->dev->dev_addr);
+
+	pcie_print_link_status(bp->pdev);
 }
 
 static int bnxt_init_one(struct pci_dev *pdev, const struct pci_device_id *ent)
 {
-	static int version_printed;
+	struct bnxt_hw_resc *hw_resc;
 	struct net_device *dev;
 	struct bnxt *bp;
 	int rc, max_irqs;
@@ -7628,27 +16609,43 @@ static int bnxt_init_one(struct pci_dev *pdev, const struct pci_device_id *ent)
 	if (pci_is_bridge(pdev))
 		return -ENODEV;
 
-	if (version_printed++ == 0)
-		pr_info("%s", version);
+	if (!pdev->msix_cap) {
+		dev_err(&pdev->dev, "MSIX capability not found, aborting\n");
+		return -ENODEV;
+	}
+
+	/* Clear any pending DMA transactions from crash kernel
+	 * while loading driver in capture kernel.
+	 */
+	if (is_kdump_kernel()) {
+		pci_clear_master(pdev);
+		pcie_flr(pdev);
+	}
 
 	max_irqs = bnxt_get_max_irq(pdev);
-	dev = alloc_etherdev_mq(sizeof(*bp), max_irqs);
+	dev = alloc_etherdev_mqs(sizeof(*bp), max_irqs * BNXT_MAX_QUEUE,
+				 max_irqs);
 	if (!dev)
 		return -ENOMEM;
 
 	bp = netdev_priv(dev);
+	bp->board_idx = ent->driver_data;
+	bp->msg_enable = BNXT_DEF_MSG_ENABLE;
+	bnxt_set_max_func_irqs(bp, max_irqs);
 
-	if (bnxt_vf_pciid(ent->driver_data))
+	if (bnxt_vf_pciid(bp->board_idx))
 		bp->flags |= BNXT_FLAG_VF;
 
-	if (pdev->msix_cap)
-		bp->flags |= BNXT_FLAG_MSIX_CAP;
+	/* No devlink port registration in case of a VF */
+	if (BNXT_PF(bp))
+		SET_NETDEV_DEVLINK_PORT(dev, &bp->dl_port);
 
 	rc = bnxt_init_board(pdev, dev);
 	if (rc < 0)
 		goto init_err_free;
 
 	dev->netdev_ops = &bnxt_netdev_ops;
+	dev->stat_ops = &bnxt_stat_ops;
 	dev->watchdog_timeo = BNXT_TX_TIMEOUT;
 	dev->ethtool_ops = &bnxt_ethtool_ops;
 	pci_set_drvdata(pdev, dev);
@@ -7658,21 +16655,35 @@ static int bnxt_init_one(struct pci_dev *pdev, const struct pci_device_id *ent)
 		goto init_err_pci_clean;
 
 	mutex_init(&bp->hwrm_cmd_lock);
-	rc = bnxt_hwrm_ver_get(bp);
+	mutex_init(&bp->link_lock);
+
+	rc = bnxt_fw_init_one_p1(bp);
 	if (rc)
 		goto init_err_pci_clean;
 
-	if (bp->flags & BNXT_FLAG_SHORT_CMD) {
-		rc = bnxt_alloc_hwrm_short_cmd_req(bp);
-		if (rc)
-			goto init_err_pci_clean;
+	if (BNXT_PF(bp))
+		bnxt_vpd_read_info(bp);
+
+	if (BNXT_CHIP_P5_PLUS(bp)) {
+		bp->flags |= BNXT_FLAG_CHIP_P5_PLUS;
+		if (BNXT_CHIP_P7(bp))
+			bp->flags |= BNXT_FLAG_CHIP_P7;
 	}
 
-	rc = bnxt_hwrm_func_reset(bp);
+	rc = bnxt_alloc_rss_indir_tbl(bp);
 	if (rc)
 		goto init_err_pci_clean;
 
-	bnxt_hwrm_fw_set_time(bp);
+	rc = bnxt_fw_init_one_p2(bp);
+	if (rc)
+		goto init_err_pci_clean;
+
+	rc = bnxt_map_db_bar(bp);
+	if (rc) {
+		dev_err(&pdev->dev, "Cannot map doorbell BAR rc = %d, aborting\n",
+			rc);
+		goto init_err_pci_clean;
+	}
 
 	dev->hw_features = NETIF_F_IP_CSUM | NETIF_F_IPV6_CSUM | NETIF_F_SG |
 			   NETIF_F_TSO | NETIF_F_TSO6 |
@@ -7681,8 +16692,10 @@ static int bnxt_init_one(struct pci_dev *pdev, const struct pci_device_id *ent)
 			   NETIF_F_GSO_UDP_TUNNEL_CSUM | NETIF_F_GSO_GRE_CSUM |
 			   NETIF_F_GSO_PARTIAL | NETIF_F_RXHASH |
 			   NETIF_F_RXCSUM | NETIF_F_GRO;
+	if (bp->flags & BNXT_FLAG_UDP_GSO_CAP)
+		dev->hw_features |= NETIF_F_GSO_UDP_L4;
 
-	if (!BNXT_CHIP_TYPE_NITRO_A0(bp))
+	if (BNXT_SUPPORTS_TPA(bp))
 		dev->hw_features |= NETIF_F_LRO;
 
 	dev->hw_enc_features =
@@ -7691,124 +16704,171 @@ static int bnxt_init_one(struct pci_dev *pdev, const struct pci_device_id *ent)
 			NETIF_F_GSO_UDP_TUNNEL | NETIF_F_GSO_GRE |
 			NETIF_F_GSO_UDP_TUNNEL_CSUM | NETIF_F_GSO_GRE_CSUM |
 			NETIF_F_GSO_IPXIP4 | NETIF_F_GSO_PARTIAL;
+	if (bp->flags & BNXT_FLAG_UDP_GSO_CAP)
+		dev->hw_enc_features |= NETIF_F_GSO_UDP_L4;
+	if (bp->flags & BNXT_FLAG_CHIP_P7)
+		dev->udp_tunnel_nic_info = &bnxt_udp_tunnels_p7;
+	else
+		dev->udp_tunnel_nic_info = &bnxt_udp_tunnels;
+
 	dev->gso_partial_features = NETIF_F_GSO_UDP_TUNNEL_CSUM |
 				    NETIF_F_GSO_GRE_CSUM;
 	dev->vlan_features = dev->hw_features | NETIF_F_HIGHDMA;
-	dev->hw_features |= NETIF_F_HW_VLAN_CTAG_RX | NETIF_F_HW_VLAN_CTAG_TX |
-			    NETIF_F_HW_VLAN_STAG_RX | NETIF_F_HW_VLAN_STAG_TX;
+	if (bp->fw_cap & BNXT_FW_CAP_VLAN_RX_STRIP)
+		dev->hw_features |= BNXT_HW_FEATURE_VLAN_ALL_RX;
+	if (bp->fw_cap & BNXT_FW_CAP_VLAN_TX_INSERT)
+		dev->hw_features |= BNXT_HW_FEATURE_VLAN_ALL_TX;
+	if (BNXT_SUPPORTS_TPA(bp))
+		dev->hw_features |= NETIF_F_GRO_HW;
 	dev->features |= dev->hw_features | NETIF_F_HIGHDMA;
+	if (dev->features & NETIF_F_GRO_HW)
+		dev->features &= ~NETIF_F_LRO;
 	dev->priv_flags |= IFF_UNICAST_FLT;
 
-	/* MTU range: 60 - 9500 */
-	dev->min_mtu = ETH_ZLEN;
-	dev->max_mtu = BNXT_MAX_MTU;
+	netif_set_tso_max_size(dev, GSO_MAX_SIZE);
+	if (bp->tso_max_segs)
+		netif_set_tso_max_segs(dev, bp->tso_max_segs);
+
+	dev->xdp_features = NETDEV_XDP_ACT_BASIC | NETDEV_XDP_ACT_REDIRECT |
+			    NETDEV_XDP_ACT_RX_SG;
 
 #ifdef CONFIG_BNXT_SRIOV
 	init_waitqueue_head(&bp->sriov_cfg_wait);
 #endif
-	bp->gro_func = bnxt_gro_func_5730x;
-	if (BNXT_CHIP_P4_PLUS(bp))
-		bp->gro_func = bnxt_gro_func_5731x;
-	else
+	if (BNXT_SUPPORTS_TPA(bp)) {
+		bp->gro_func = bnxt_gro_func_5730x;
+		if (BNXT_CHIP_P4(bp))
+			bp->gro_func = bnxt_gro_func_5731x;
+		else if (BNXT_CHIP_P5_PLUS(bp))
+			bp->gro_func = bnxt_gro_func_5750x;
+	}
+	if (!BNXT_CHIP_P4_PLUS(bp))
 		bp->flags |= BNXT_FLAG_DOUBLE_DB;
 
-	rc = bnxt_hwrm_func_drv_rgtr(bp);
-	if (rc)
-		goto init_err_pci_clean;
-
-	rc = bnxt_hwrm_func_rgtr_async_events(bp, NULL, 0);
-	if (rc)
-		goto init_err_pci_clean;
-
-	bp->ulp_probe = bnxt_ulp_probe;
-
-	/* Get the MAX capabilities for this function */
-	rc = bnxt_hwrm_func_qcaps(bp);
+	rc = bnxt_init_mac_addr(bp);
 	if (rc) {
-		netdev_err(bp->dev, "hwrm query capability failure rc: %x\n",
-			   rc);
-		rc = -1;
+		dev_err(&pdev->dev, "Unable to initialize mac address.\n");
+		rc = -EADDRNOTAVAIL;
 		goto init_err_pci_clean;
 	}
 
-	rc = bnxt_hwrm_queue_qportcfg(bp);
-	if (rc) {
-		netdev_err(bp->dev, "hwrm query qportcfg failure rc: %x\n",
-			   rc);
-		rc = -1;
-		goto init_err_pci_clean;
+	if (BNXT_PF(bp)) {
+		/* Read the adapter's DSN to use as the eswitch switch_id */
+		rc = bnxt_pcie_dsn_get(bp, bp->dsn);
 	}
 
-	bnxt_hwrm_func_qcfg(bp);
-	bnxt_hwrm_port_led_qcaps(bp);
-	bnxt_ethtool_init(bp);
-	bnxt_dcb_init(bp);
+	/* MTU range: 60 - FW defined max */
+	dev->min_mtu = ETH_ZLEN;
+	dev->max_mtu = bp->max_mtu;
 
-	bnxt_set_rx_skb_mode(bp, false);
+	rc = bnxt_probe_phy(bp, true);
+	if (rc)
+		goto init_err_pci_clean;
+
+	hw_resc = &bp->hw_resc;
+	bp->max_fltr = hw_resc->max_rx_em_flows + hw_resc->max_rx_wm_flows +
+		       BNXT_L2_FLTR_MAX_FLTR;
+	/* Older firmware may not report these filters properly */
+	if (bp->max_fltr < BNXT_MAX_FLTR)
+		bp->max_fltr = BNXT_MAX_FLTR;
+	bnxt_init_l2_fltr_tbl(bp);
+	__bnxt_set_rx_skb_mode(bp, false);
 	bnxt_set_tpa_flags(bp);
+	bnxt_init_ring_params(bp);
 	bnxt_set_ring_params(bp);
-	bnxt_set_max_func_irqs(bp, max_irqs);
+	bnxt_rdma_aux_device_init(bp);
 	rc = bnxt_set_dflt_rings(bp, true);
 	if (rc) {
-		netdev_err(bp->dev, "Not enough rings available.\n");
-		rc = -ENOMEM;
+		if (BNXT_VF(bp) && rc == -ENODEV) {
+			netdev_err(bp->dev, "Cannot configure VF rings while PF is unavailable.\n");
+		} else {
+			netdev_err(bp->dev, "Not enough rings available.\n");
+			rc = -ENOMEM;
+		}
 		goto init_err_pci_clean;
 	}
 
-	/* Default RSS hash cfg. */
-	bp->rss_hash_cfg = VNIC_RSS_CFG_REQ_HASH_TYPE_IPV4 |
-			   VNIC_RSS_CFG_REQ_HASH_TYPE_TCP_IPV4 |
-			   VNIC_RSS_CFG_REQ_HASH_TYPE_IPV6 |
-			   VNIC_RSS_CFG_REQ_HASH_TYPE_TCP_IPV6;
-	if (BNXT_CHIP_P4_PLUS(bp) && bp->hwrm_spec_code >= 0x10501) {
-		bp->flags |= BNXT_FLAG_UDP_RSS_CAP;
-		bp->rss_hash_cfg |= VNIC_RSS_CFG_REQ_HASH_TYPE_UDP_IPV4 |
-				    VNIC_RSS_CFG_REQ_HASH_TYPE_UDP_IPV6;
-	}
+	bnxt_fw_init_one_p3(bp);
 
-	bnxt_hwrm_vnic_qcaps(bp);
-	if (bnxt_rfs_supported(bp)) {
-		dev->hw_features |= NETIF_F_NTUPLE;
-		if (bnxt_rfs_capable(bp)) {
-			bp->flags |= BNXT_FLAG_RFS;
-			dev->features |= NETIF_F_NTUPLE;
-		}
-	}
+	bnxt_init_dflt_coal(bp);
 
-	if (dev->hw_features & NETIF_F_HW_VLAN_CTAG_RX)
+	if (dev->hw_features & BNXT_HW_FEATURE_VLAN_ALL_RX)
 		bp->flags |= BNXT_FLAG_STRIP_VLAN;
 
-	rc = bnxt_probe_phy(bp);
+	rc = bnxt_init_int_mode(bp);
 	if (rc)
 		goto init_err_pci_clean;
 
-	rc = bnxt_init_int_mode(bp);
+	/* No TC has been set yet and rings may have been trimmed due to
+	 * limited MSIX, so we re-initialize the TX rings per TC.
+	 */
+	bp->tx_nr_rings_per_tc = bp->tx_nr_rings;
+
+	if (BNXT_PF(bp)) {
+		if (!bnxt_pf_wq) {
+			bnxt_pf_wq =
+				create_singlethread_workqueue("bnxt_pf_wq");
+			if (!bnxt_pf_wq) {
+				dev_err(&pdev->dev, "Unable to create workqueue.\n");
+				rc = -ENOMEM;
+				goto init_err_pci_clean;
+			}
+		}
+		rc = bnxt_init_tc(bp);
+		if (rc)
+			netdev_err(dev, "Failed to initialize TC flower offload, err = %d.\n",
+				   rc);
+	}
+
+	bnxt_inv_fw_health_reg(bp);
+	rc = bnxt_dl_register(bp);
 	if (rc)
-		goto init_err_pci_clean;
+		goto init_err_dl;
 
-	bnxt_get_wol_settings(bp);
-	if (bp->flags & BNXT_FLAG_WOL_CAP)
-		device_set_wakeup_enable(&pdev->dev, bp->wol);
-	else
-		device_set_wakeup_capable(&pdev->dev, false);
+	INIT_LIST_HEAD(&bp->usr_fltr_list);
+
+	if (BNXT_SUPPORTS_NTUPLE_VNIC(bp))
+		bp->rss_cap |= BNXT_RSS_CAP_MULTI_RSS_CTX;
+	if (BNXT_SUPPORTS_QUEUE_API(bp))
+		dev->queue_mgmt_ops = &bnxt_queue_mgmt_ops;
+	dev->request_ops_lock = true;
+	dev->netmem_tx = true;
 
 	rc = register_netdev(dev);
 	if (rc)
-		goto init_err_clr_int;
+		goto init_err_cleanup;
 
-	netdev_info(dev, "%s found at mem %lx, node addr %pM\n",
-		    board_info[ent->driver_data].name,
-		    (long)pci_resource_start(pdev, 0), dev->dev_addr);
+	bnxt_dl_fw_reporters_create(bp);
 
-	bnxt_parse_log_pcie_link(bp);
+	bnxt_rdma_aux_device_add(bp);
 
-	return 0;
+	bnxt_print_device_info(bp);
 
-init_err_clr_int:
+	pci_save_state(pdev);
+
+	return 0;
+init_err_cleanup:
+	bnxt_rdma_aux_device_uninit(bp);
+	bnxt_dl_unregister(bp);
+init_err_dl:
+	bnxt_shutdown_tc(bp);
 	bnxt_clear_int_mode(bp);
 
 init_err_pci_clean:
+	bnxt_hwrm_func_drv_unrgtr(bp);
+	bnxt_free_hwrm_resources(bp);
+	bnxt_hwmon_uninit(bp);
+	bnxt_ethtool_free(bp);
+	bnxt_ptp_clear(bp);
+	kfree(bp->ptp_cfg);
+	bp->ptp_cfg = NULL;
+	kfree(bp->fw_health);
+	bp->fw_health = NULL;
 	bnxt_cleanup_pci(bp);
+	bnxt_free_ctx_mem(bp, true);
+	bnxt_free_crash_dump_mem(bp);
+	kfree(bp->rss_indir_tbl);
+	bp->rss_indir_tbl = NULL;
 
 init_err_free:
 	free_netdev(dev);
@@ -7824,51 +16884,65 @@ static void bnxt_shutdown(struct pci_dev *pdev)
 		return;
 
 	rtnl_lock();
+	netdev_lock(dev);
 	bp = netdev_priv(dev);
 	if (!bp)
 		goto shutdown_exit;
 
 	if (netif_running(dev))
-		dev_close(dev);
+		netif_close(dev);
+
+	bnxt_ptp_clear(bp);
+	bnxt_clear_int_mode(bp);
+	pci_disable_device(pdev);
 
 	if (system_state == SYSTEM_POWER_OFF) {
-		bnxt_ulp_shutdown(bp);
-		bnxt_clear_int_mode(bp);
 		pci_wake_from_d3(pdev, bp->wol);
 		pci_set_power_state(pdev, PCI_D3hot);
 	}
 
 shutdown_exit:
+	netdev_unlock(dev);
 	rtnl_unlock();
 }
 
 #ifdef CONFIG_PM_SLEEP
 static int bnxt_suspend(struct device *device)
 {
-	struct pci_dev *pdev = to_pci_dev(device);
-	struct net_device *dev = pci_get_drvdata(pdev);
+	struct net_device *dev = dev_get_drvdata(device);
 	struct bnxt *bp = netdev_priv(dev);
 	int rc = 0;
 
-	rtnl_lock();
+	bnxt_ulp_stop(bp);
+
+	netdev_lock(dev);
 	if (netif_running(dev)) {
 		netif_device_detach(dev);
 		rc = bnxt_close(dev);
 	}
 	bnxt_hwrm_func_drv_unrgtr(bp);
-	rtnl_unlock();
+	bnxt_ptp_clear(bp);
+	pci_disable_device(bp->pdev);
+	bnxt_free_ctx_mem(bp, false);
+	netdev_unlock(dev);
 	return rc;
 }
 
 static int bnxt_resume(struct device *device)
 {
-	struct pci_dev *pdev = to_pci_dev(device);
-	struct net_device *dev = pci_get_drvdata(pdev);
+	struct net_device *dev = dev_get_drvdata(device);
 	struct bnxt *bp = netdev_priv(dev);
 	int rc = 0;
 
-	rtnl_lock();
-	if (bnxt_hwrm_ver_get(bp) || bnxt_hwrm_func_drv_rgtr(bp)) {
+	netdev_lock(dev);
+	rc = pci_enable_device(bp->pdev);
+	if (rc) {
+		netdev_err(dev, "Cannot re-enable PCI device during resume, err = %d\n",
+			   rc);
+		goto resume_exit;
+	}
+	pci_set_master(bp->pdev);
+	if (bnxt_hwrm_ver_get(bp)) {
 		rc = -ENODEV;
 		goto resume_exit;
 	}
@@ -7877,6 +16951,24 @@ static int bnxt_resume(struct device *device)
 		rc = -EBUSY;
 		goto resume_exit;
 	}
+
+	rc = bnxt_hwrm_func_qcaps(bp);
+	if (rc)
+		goto resume_exit;
+
+	bnxt_clear_reservations(bp, true);
+
+	if (bnxt_hwrm_func_drv_rgtr(bp, NULL, 0, false)) {
+		rc = -ENODEV;
+		goto resume_exit;
+	}
+	if (bp->fw_crash_mem)
+		bnxt_hwrm_crash_dump_mem_cfg(bp);
+
+	if (bnxt_ptp_init(bp)) {
+		kfree(bp->ptp_cfg);
+		bp->ptp_cfg = NULL;
+	}
 	bnxt_get_wol_settings(bp);
 	if (netif_running(dev)) {
 		rc = bnxt_open(dev);
@@ -7885,7 +16977,10 @@ static int bnxt_resume(struct device *device)
 	}
 
 resume_exit:
-	rtnl_unlock();
+	netdev_unlock(bp->dev);
+	bnxt_ulp_start(bp, rc);
+	if (!rc)
+		bnxt_reenable_sriov(bp);
 	return rc;
 }
 
@@ -7911,26 +17006,43 @@ static pci_ers_result_t bnxt_io_error_detected(struct pci_dev *pdev,
 {
 	struct net_device *netdev = pci_get_drvdata(pdev);
 	struct bnxt *bp = netdev_priv(netdev);
+	bool abort = false;
 
 	netdev_info(netdev, "PCI I/O error detected\n");
 
-	rtnl_lock();
+	bnxt_ulp_stop(bp);
+
+	netdev_lock(netdev);
 	netif_device_detach(netdev);
 
-	bnxt_ulp_stop(bp);
+	if (test_and_set_bit(BNXT_STATE_IN_FW_RESET, &bp->state)) {
+		netdev_err(bp->dev, "Firmware reset already in progress\n");
+		abort = true;
+	}
 
-	if (state == pci_channel_io_perm_failure) {
-		rtnl_unlock();
+	if (abort || state == pci_channel_io_perm_failure) {
+		netdev_unlock(netdev);
 		return PCI_ERS_RESULT_DISCONNECT;
 	}
 
+	/* Link is not reliable anymore if state is pci_channel_io_frozen
+	 * so we disable bus master to prevent any potential bad DMAs before
+	 * freeing kernel memory.
+	 */
+	if (state == pci_channel_io_frozen) {
+		set_bit(BNXT_STATE_PCI_CHANNEL_IO_FROZEN, &bp->state);
+		bnxt_fw_fatal_close(bp);
+	}
+
 	if (netif_running(netdev))
-		bnxt_close(netdev);
+		__bnxt_close_nic(bp, true, true);
 
-	pci_disable_device(pdev);
-	rtnl_unlock();
+	if (pci_is_enabled(pdev))
+		pci_disable_device(pdev);
+	bnxt_free_ctx_mem(bp, false);
+	netdev_unlock(netdev);
 
-	/* Request a slot slot reset. */
+	/* Request a slot reset. */
 	return PCI_ERS_RESULT_NEED_RESET;
 }
 
@@ -7939,50 +17051,82 @@ static pci_ers_result_t bnxt_io_error_detected(struct pci_dev *pdev,
  * @pdev: Pointer to PCI device
  *
  * Restart the card from scratch, as if from a cold-boot.
- * At this point, the card has exprienced a hard reset,
+ * At this point, the card has experienced a hard reset,
  * followed by fixups by BIOS, and has its config space
  * set up identically to what it was at cold boot.
  */
 static pci_ers_result_t bnxt_io_slot_reset(struct pci_dev *pdev)
 {
+	pci_ers_result_t result = PCI_ERS_RESULT_DISCONNECT;
 	struct net_device *netdev = pci_get_drvdata(pdev);
 	struct bnxt *bp = netdev_priv(netdev);
+	int retry = 0;
 	int err = 0;
-	pci_ers_result_t result = PCI_ERS_RESULT_DISCONNECT;
+	int off;
 
 	netdev_info(bp->dev, "PCI Slot Reset\n");
 
-	rtnl_lock();
+	if (!(bp->flags & BNXT_FLAG_CHIP_P5_PLUS) &&
+	    test_bit(BNXT_STATE_PCI_CHANNEL_IO_FROZEN, &bp->state))
+		msleep(900);
+
+	netdev_lock(netdev);
 
 	if (pci_enable_device(pdev)) {
 		dev_err(&pdev->dev,
 			"Cannot re-enable PCI device after reset.\n");
 	} else {
 		pci_set_master(pdev);
+		/* Upon fatal error, our device internal logic that latches to
+		 * BAR value is getting reset and will restore only upon
+		 * rewriting the BARs.
+		 *
+		 * As pci_restore_state() does not re-write the BARs if the
+		 * value is same as saved value earlier, driver needs to
+		 * write the BARs to 0 to force restore, in case of fatal error.
+		 */
+		if (test_and_clear_bit(BNXT_STATE_PCI_CHANNEL_IO_FROZEN,
+				       &bp->state)) {
+			for (off = PCI_BASE_ADDRESS_0;
+			     off <= PCI_BASE_ADDRESS_5; off += 4)
+				pci_write_config_dword(bp->pdev, off, 0);
+		}
+		pci_restore_state(pdev);
+		pci_save_state(pdev);
 
-		err = bnxt_hwrm_func_reset(bp);
-		if (!err && netif_running(netdev))
-			err = bnxt_open(netdev);
+		bnxt_inv_fw_health_reg(bp);
+		bnxt_try_map_fw_health_reg(bp);
 
-		if (!err) {
-			result = PCI_ERS_RESULT_RECOVERED;
-			bnxt_ulp_start(bp);
-		}
-	}
+		/* In some PCIe AER scenarios, firmware may take up to
+		 * 10 seconds to become ready in the worst case.
+		 */
+		do {
+			err = bnxt_try_recover_fw(bp);
+			if (!err)
+				break;
+			retry++;
+		} while (retry < BNXT_FW_SLOT_RESET_RETRY);
 
-	if (result != PCI_ERS_RESULT_RECOVERED && netif_running(netdev))
-		dev_close(netdev);
+		if (err) {
+			dev_err(&pdev->dev, "Firmware not ready\n");
+			goto reset_exit;
+		}
 
-	rtnl_unlock();
+		err = bnxt_hwrm_func_reset(bp);
+		if (!err)
+			result = PCI_ERS_RESULT_RECOVERED;
 
-	err = pci_cleanup_aer_uncorrect_error_status(pdev);
-	if (err) {
-		dev_err(&pdev->dev,
-			"pci_cleanup_aer_uncorrect_error_status failed 0x%0x\n",
-			 err); /* non-fatal, continue */
+		/* IRQ will be initialized later in bnxt_io_resume */
+		bnxt_ulp_irq_stop(bp);
+		bnxt_clear_int_mode(bp);
 	}
 
-	return PCI_ERS_RESULT_RECOVERED;
+reset_exit:
+	clear_bit(BNXT_STATE_IN_FW_RESET, &bp->state);
+	bnxt_clear_reservations(bp, true);
+	netdev_unlock(netdev);
+
+	return result;
 }
 
 /**
@@ -7995,12 +17139,30 @@ static pci_ers_result_t bnxt_io_slot_reset(struct pci_dev *pdev)
 static void bnxt_io_resume(struct pci_dev *pdev)
 {
 	struct net_device *netdev = pci_get_drvdata(pdev);
+	struct bnxt *bp = netdev_priv(netdev);
+	int err;
 
-	rtnl_lock();
+	netdev_info(bp->dev, "PCI Slot Resume\n");
+	netdev_lock(netdev);
+
+	err = bnxt_hwrm_func_qcaps(bp);
+	if (!err) {
+		if (netif_running(netdev)) {
+			err = bnxt_open(netdev);
+		} else {
+			err = bnxt_reserve_rings(bp, true);
+			if (!err)
+				err = bnxt_init_int_mode(bp);
+		}
+	}
 
-	netif_device_attach(netdev);
+	if (!err)
+		netif_device_attach(netdev);
 
-	rtnl_unlock();
+	netdev_unlock(netdev);
+	bnxt_ulp_start(bp, err);
+	if (!err)
+		bnxt_reenable_sriov(bp);
 }
 
 static const struct pci_error_handlers bnxt_err_handler = {
@@ -8022,4 +17184,27 @@ static struct pci_driver bnxt_pci_driver = {
 #endif
 };
 
-module_pci_driver(bnxt_pci_driver);
+static int __init bnxt_init(void)
+{
+	int err;
+
+	bnxt_debug_init();
+	err = pci_register_driver(&bnxt_pci_driver);
+	if (err) {
+		bnxt_debug_exit();
+		return err;
+	}
+
+	return 0;
+}
+
+static void __exit bnxt_exit(void)
+{
+	pci_unregister_driver(&bnxt_pci_driver);
+	if (bnxt_pf_wq)
+		destroy_workqueue(bnxt_pf_wq);
+	bnxt_debug_exit();
+}
+
+module_init(bnxt_init);
+module_exit(bnxt_exit);
diff --git a/drivers/net/ethernet/broadcom/bnxt/bnxt.h b/drivers/net/ethernet/broadcom/bnxt/bnxt.h
index 5984423499e6..741b2d854789 100644
--- a/drivers/net/ethernet/broadcom/bnxt/bnxt.h
+++ b/drivers/net/ethernet/broadcom/bnxt/bnxt.h
@@ -1,7 +1,7 @@
 /* Broadcom NetXtreme-C/E network driver.
  *
  * Copyright (c) 2014-2016 Broadcom Corporation
- * Copyright (c) 2016-2017 Broadcom Limited
+ * Copyright (c) 2016-2018 Broadcom Limited
  *
  * This program is free software; you can redistribute it and/or modify
  * it under the terms of the GNU General Public License as published by
@@ -12,13 +12,34 @@
 #define BNXT_H
 
 #define DRV_MODULE_NAME		"bnxt_en"
-#define DRV_MODULE_VERSION	"1.7.0"
 
+/* DO NOT CHANGE DRV_VER_* defines
+ * FIXME: Delete them
+ */
 #define DRV_VER_MAJ	1
-#define DRV_VER_MIN	7
-#define DRV_VER_UPD	0
+#define DRV_VER_MIN	10
+#define DRV_VER_UPD	3
 
+#include <linux/ethtool.h>
 #include <linux/interrupt.h>
+#include <linux/rhashtable.h>
+#include <linux/crash_dump.h>
+#include <linux/auxiliary_bus.h>
+#include <net/devlink.h>
+#include <net/dst_metadata.h>
+#include <net/xdp.h>
+#include <linux/dim.h>
+#include <linux/io-64-nonatomic-lo-hi.h>
+#ifdef CONFIG_TEE_BNXT_FW
+#include <linux/firmware/broadcom/tee_bnxt_fw.h>
+#endif
+
+#define BNXT_DEFAULT_RX_COPYBREAK 256
+#define BNXT_MAX_RX_COPYBREAK 1024
+
+extern struct list_head bnxt_block_cb_list;
+
+struct page_pool;
 
 struct tx_bd {
 	__le32 tx_bd_len_flags_type;
@@ -43,6 +64,30 @@ struct tx_bd {
 	__le64 tx_bd_haddr;
 } __packed;
 
+#define TX_OPAQUE_IDX_MASK	0x0000ffff
+#define TX_OPAQUE_BDS_MASK	0x00ff0000
+#define TX_OPAQUE_BDS_SHIFT	16
+#define TX_OPAQUE_RING_MASK	0xff000000
+#define TX_OPAQUE_RING_SHIFT	24
+
+#define SET_TX_OPAQUE(bp, txr, idx, bds)				\
+	(((txr)->tx_napi_idx << TX_OPAQUE_RING_SHIFT) |			\
+	 ((bds) << TX_OPAQUE_BDS_SHIFT) | ((idx) & (bp)->tx_ring_mask))
+
+#define TX_OPAQUE_IDX(opq)	((opq) & TX_OPAQUE_IDX_MASK)
+#define TX_OPAQUE_RING(opq)	(((opq) & TX_OPAQUE_RING_MASK) >>	\
+				 TX_OPAQUE_RING_SHIFT)
+#define TX_OPAQUE_BDS(opq)	(((opq) & TX_OPAQUE_BDS_MASK) >>	\
+				 TX_OPAQUE_BDS_SHIFT)
+#define TX_OPAQUE_PROD(bp, opq)	((TX_OPAQUE_IDX(opq) + TX_OPAQUE_BDS(opq)) &\
+				 (bp)->tx_ring_mask)
+
+#define TX_BD_CNT(n)	(((n) << TX_BD_FLAGS_BD_CNT_SHIFT) & TX_BD_FLAGS_BD_CNT)
+
+#define TX_MAX_BD_CNT	32
+
+#define TX_MAX_FRAGS		(TX_MAX_BD_CNT - 2)
+
 struct tx_bd_ext {
 	__le32 tx_bd_hsize_lflags;
 	#define TX_BD_FLAGS_TCP_UDP_CHKSUM			(1 << 0)
@@ -73,6 +118,8 @@ struct tx_bd_ext {
 	#define TX_BD_CFA_META_KEY_VLAN                         (1 << 28)
 };
 
+#define BNXT_TX_PTP_IS_SET(lflags) ((lflags) & cpu_to_le32(TX_BD_FLAGS_STAMP))
+
 struct rx_bd {
 	__le32 rx_bd_len_flags_type;
 	#define RX_BD_TYPE					(0x3f << 0)
@@ -101,10 +148,15 @@ struct tx_cmp {
 	__le32 tx_cmp_flags_type;
 	#define CMP_TYPE					(0x3f << 0)
 	 #define CMP_TYPE_TX_L2_CMP				 0
+	 #define CMP_TYPE_TX_L2_COAL_CMP			 2
+	 #define CMP_TYPE_TX_L2_PKT_TS_CMP			 4
 	 #define CMP_TYPE_RX_L2_CMP				 17
 	 #define CMP_TYPE_RX_AGG_CMP				 18
 	 #define CMP_TYPE_RX_L2_TPA_START_CMP			 19
 	 #define CMP_TYPE_RX_L2_TPA_END_CMP			 21
+	 #define CMP_TYPE_RX_TPA_AGG_CMP			 22
+	 #define CMP_TYPE_RX_L2_V3_CMP				 23
+	 #define CMP_TYPE_RX_L2_TPA_START_V3_CMP		 25
 	 #define CMP_TYPE_STATUS_CMP				 32
 	 #define CMP_TYPE_REMOTE_DRIVER_REQ			 34
 	 #define CMP_TYPE_REMOTE_DRIVER_RESP			 36
@@ -131,17 +183,48 @@ struct tx_cmp {
 	 #define TX_CMP_ERRORS_DMA_ERROR			 (1 << 6)
 	 #define TX_CMP_ERRORS_HINT_TOO_SHORT			 (1 << 7)
 
-	__le32 tx_cmp_unsed_3;
+	__le32 sq_cons_idx;
+	#define TX_CMP_SQ_CONS_IDX_MASK				0x00ffffff
 };
 
+#define TX_CMP_SQ_CONS_IDX(txcmp)					\
+	(le32_to_cpu((txcmp)->sq_cons_idx) & TX_CMP_SQ_CONS_IDX_MASK)
+
+struct tx_ts_cmp {
+	__le32 tx_ts_cmp_flags_type;
+	#define TX_TS_CMP_FLAGS_ERROR				(1 << 6)
+	#define TX_TS_CMP_FLAGS_TS_TYPE				(1 << 7)
+	 #define TX_TS_CMP_FLAGS_TS_TYPE_PM			 (0 << 7)
+	 #define TX_TS_CMP_FLAGS_TS_TYPE_PA			 (1 << 7)
+	#define TX_TS_CMP_FLAGS_TS_FALLBACK			(1 << 8)
+	#define TX_TS_CMP_TS_SUB_NS				(0xf << 12)
+	#define TX_TS_CMP_TS_NS_MID				(0xffff << 16)
+	#define TX_TS_CMP_TS_NS_MID_SFT				16
+	u32 tx_ts_cmp_opaque;
+	__le32 tx_ts_cmp_errors_v;
+	#define TX_TS_CMP_V					(1 << 0)
+	#define TX_TS_CMP_TS_INVALID_ERR			(1 << 10)
+	__le32 tx_ts_cmp_ts_ns_lo;
+};
+
+#define BNXT_GET_TX_TS_48B_NS(tscmp)					\
+	(le32_to_cpu((tscmp)->tx_ts_cmp_ts_ns_lo) |			\
+	 ((u64)(le32_to_cpu((tscmp)->tx_ts_cmp_flags_type) &		\
+	  TX_TS_CMP_TS_NS_MID) << TX_TS_CMP_TS_NS_MID_SFT))
+
+#define BNXT_TX_TS_ERR(tscmp)						\
+	(((tscmp)->tx_ts_cmp_flags_type & cpu_to_le32(TX_TS_CMP_FLAGS_ERROR)) &&\
+	 ((tscmp)->tx_ts_cmp_errors_v & cpu_to_le32(TX_TS_CMP_TS_INVALID_ERR)))
+
 struct rx_cmp {
 	__le32 rx_cmp_len_flags_type;
 	#define RX_CMP_CMP_TYPE					(0x3f << 0)
 	#define RX_CMP_FLAGS_ERROR				(1 << 6)
 	#define RX_CMP_FLAGS_PLACEMENT				(7 << 7)
 	#define RX_CMP_FLAGS_RSS_VALID				(1 << 10)
-	#define RX_CMP_FLAGS_UNUSED				(1 << 11)
+	#define RX_CMP_FLAGS_PKT_METADATA_PRESENT		(1 << 11)
 	 #define RX_CMP_FLAGS_ITYPES_SHIFT			 12
+	 #define RX_CMP_FLAGS_ITYPES_MASK			 0xf000
 	 #define RX_CMP_FLAGS_ITYPE_UNKNOWN			 (0 << 12)
 	 #define RX_CMP_FLAGS_ITYPE_IP				 (1 << 12)
 	 #define RX_CMP_FLAGS_ITYPE_TCP				 (2 << 12)
@@ -160,12 +243,30 @@ struct rx_cmp {
 	 #define RX_CMP_AGG_BUFS_SHIFT				 1
 	#define RX_CMP_RSS_HASH_TYPE				(0x7f << 9)
 	 #define RX_CMP_RSS_HASH_TYPE_SHIFT			 9
+	#define RX_CMP_V3_RSS_EXT_OP_LEGACY			(0xf << 12)
+	 #define RX_CMP_V3_RSS_EXT_OP_LEGACY_SHIFT		 12
+	#define RX_CMP_V3_RSS_EXT_OP_NEW			(0xf << 8)
+	 #define RX_CMP_V3_RSS_EXT_OP_NEW_SHIFT			 8
 	#define RX_CMP_PAYLOAD_OFFSET				(0xff << 16)
 	 #define RX_CMP_PAYLOAD_OFFSET_SHIFT			 16
+	#define RX_CMP_SUB_NS_TS				(0xf << 16)
+	 #define RX_CMP_SUB_NS_TS_SHIFT				 16
+	#define RX_CMP_METADATA1				(0xf << 28)
+	 #define RX_CMP_METADATA1_SHIFT				 28
+	#define RX_CMP_METADATA1_TPID_SEL			(0x7 << 28)
+	#define RX_CMP_METADATA1_TPID_8021Q			(0x1 << 28)
+	#define RX_CMP_METADATA1_TPID_8021AD			(0x0 << 28)
+	#define RX_CMP_METADATA1_VALID				(0x8 << 28)
 
 	__le32 rx_cmp_rss_hash;
 };
 
+#define BNXT_PTP_RX_TS_VALID(flags)				\
+	(((flags) & RX_CMP_FLAGS_ITYPES_MASK) == RX_CMP_FLAGS_ITYPE_PTP_W_TS)
+
+#define BNXT_ALL_RX_TS_VALID(flags)				\
+	!((flags) & RX_CMP_FLAGS_PKT_METADATA_PRESENT)
+
 #define RX_CMP_HASH_VALID(rxcmp)				\
 	((rxcmp)->rx_cmp_len_flags_type & cpu_to_le32(RX_CMP_FLAGS_RSS_VALID))
 
@@ -175,6 +276,33 @@ struct rx_cmp {
 	(((le32_to_cpu((rxcmp)->rx_cmp_misc_v1) & RX_CMP_RSS_HASH_TYPE) >>\
 	  RX_CMP_RSS_HASH_TYPE_SHIFT) & RSS_PROFILE_ID_MASK)
 
+#define RX_CMP_ITYPES(rxcmp)					\
+	(le32_to_cpu((rxcmp)->rx_cmp_len_flags_type) & RX_CMP_FLAGS_ITYPES_MASK)
+
+#define RX_CMP_V3_HASH_TYPE_LEGACY(rxcmp)				\
+	((le32_to_cpu((rxcmp)->rx_cmp_misc_v1) & RX_CMP_V3_RSS_EXT_OP_LEGACY) >>\
+	 RX_CMP_V3_RSS_EXT_OP_LEGACY_SHIFT)
+
+#define RX_CMP_V3_HASH_TYPE_NEW(rxcmp)				\
+	((le32_to_cpu((rxcmp)->rx_cmp_misc_v1) & RX_CMP_V3_RSS_EXT_OP_NEW) >>\
+	 RX_CMP_V3_RSS_EXT_OP_NEW_SHIFT)
+
+#define RX_CMP_V3_HASH_TYPE(bp, rxcmp)				\
+	(((bp)->rss_cap & BNXT_RSS_CAP_RSS_TCAM) ?		\
+	  RX_CMP_V3_HASH_TYPE_NEW(rxcmp) :			\
+	  RX_CMP_V3_HASH_TYPE_LEGACY(rxcmp))
+
+#define EXT_OP_INNER_4		0x0
+#define EXT_OP_OUTER_4		0x2
+#define EXT_OP_INNFL_3		0x8
+#define EXT_OP_OUTFL_3		0xa
+
+#define RX_CMP_VLAN_VALID(rxcmp)				\
+	((rxcmp)->rx_cmp_misc_v1 & cpu_to_le32(RX_CMP_METADATA1_VALID))
+
+#define RX_CMP_VLAN_TPID_SEL(rxcmp)				\
+	(le32_to_cpu((rxcmp)->rx_cmp_misc_v1) & RX_CMP_METADATA1_TPID_SEL)
+
 struct rx_cmp_ext {
 	__le32 rx_cmp_flags2;
 	#define RX_CMP_FLAGS2_IP_CS_CALC			0x1
@@ -183,6 +311,7 @@ struct rx_cmp_ext {
 	#define RX_CMP_FLAGS2_T_L4_CS_CALC			(0x1 << 3)
 	#define RX_CMP_FLAGS2_META_FORMAT_VLAN			(0x1 << 4)
 	__le32 rx_cmp_meta_data;
+	#define RX_CMP_FLAGS2_METADATA_TCI_MASK			0xffff
 	#define RX_CMP_FLAGS2_METADATA_VID_MASK			0xfff
 	#define RX_CMP_FLAGS2_METADATA_TPID_MASK		0xffff0000
 	 #define RX_CMP_FLAGS2_METADATA_TPID_SFT		 16
@@ -221,8 +350,11 @@ struct rx_cmp_ext {
 
 	#define RX_CMPL_CFA_CODE_MASK				(0xffff << 16)
 	 #define RX_CMPL_CFA_CODE_SFT				 16
+	#define RX_CMPL_METADATA0_TCI_MASK			(0xffff << 16)
+	#define RX_CMPL_METADATA0_VID_MASK			(0x0fff << 16)
+	 #define RX_CMPL_METADATA0_SFT				 16
 
-	__le32 rx_cmp_unused3;
+	__le32 rx_cmp_timestamp;
 };
 
 #define RX_CMP_L2_ERRORS						\
@@ -242,6 +374,14 @@ struct rx_cmp_ext {
 	    ((le32_to_cpu((rxcmp1)->rx_cmp_flags2) &			\
 	     RX_CMP_FLAGS2_T_L4_CS_CALC) >> 3)
 
+#define RX_CMP_CFA_CODE(rxcmpl1)					\
+	((le32_to_cpu((rxcmpl1)->rx_cmp_cfa_code_errors_v2) &		\
+	  RX_CMPL_CFA_CODE_MASK) >> RX_CMPL_CFA_CODE_SFT)
+
+#define RX_CMP_METADATA0_TCI(rxcmp1)					\
+	((le32_to_cpu((rxcmp1)->rx_cmp_cfa_code_errors_v2) &		\
+	  RX_CMPL_METADATA0_TCI_MASK) >> RX_CMPL_METADATA0_SFT)
+
 struct rx_agg_cmp {
 	__le32 rx_agg_cmp_len_flags_type;
 	#define RX_AGG_CMP_TYPE					(0x3f << 0)
@@ -250,14 +390,21 @@ struct rx_agg_cmp {
 	u32 rx_agg_cmp_opaque;
 	__le32 rx_agg_cmp_v;
 	#define RX_AGG_CMP_V					(1 << 0)
+	#define RX_AGG_CMP_AGG_ID				(0x0fff << 16)
+	 #define RX_AGG_CMP_AGG_ID_SHIFT			 16
 	__le32 rx_agg_cmp_unused;
 };
 
+#define TPA_AGG_AGG_ID(rx_agg)				\
+	((le32_to_cpu((rx_agg)->rx_agg_cmp_v) &		\
+	 RX_AGG_CMP_AGG_ID) >> RX_AGG_CMP_AGG_ID_SHIFT)
+
 struct rx_tpa_start_cmp {
 	__le32 rx_tpa_start_cmp_len_flags_type;
 	#define RX_TPA_START_CMP_TYPE				(0x3f << 0)
 	#define RX_TPA_START_CMP_FLAGS				(0x3ff << 6)
 	 #define RX_TPA_START_CMP_FLAGS_SHIFT			 6
+	#define RX_TPA_START_CMP_FLAGS_ERROR			(0x1 << 6)
 	#define RX_TPA_START_CMP_FLAGS_PLACEMENT		(0x7 << 7)
 	 #define RX_TPA_START_CMP_FLAGS_PLACEMENT_SHIFT		 7
 	 #define RX_TPA_START_CMP_FLAGS_PLACEMENT_JUMBO		 (0x1 << 7)
@@ -265,6 +412,7 @@ struct rx_tpa_start_cmp {
 	 #define RX_TPA_START_CMP_FLAGS_PLACEMENT_GRO_JUMBO	 (0x5 << 7)
 	 #define RX_TPA_START_CMP_FLAGS_PLACEMENT_GRO_HDS	 (0x6 << 7)
 	#define RX_TPA_START_CMP_FLAGS_RSS_VALID		(0x1 << 10)
+	#define RX_TPA_START_CMP_FLAGS_TIMESTAMP		(0x1 << 11)
 	#define RX_TPA_START_CMP_FLAGS_ITYPES			(0xf << 12)
 	 #define RX_TPA_START_CMP_FLAGS_ITYPES_SHIFT		 12
 	 #define RX_TPA_START_CMP_FLAGS_ITYPE_TCP		 (0x2 << 12)
@@ -276,8 +424,18 @@ struct rx_tpa_start_cmp {
 	#define RX_TPA_START_CMP_V1				(0x1 << 0)
 	#define RX_TPA_START_CMP_RSS_HASH_TYPE			(0x7f << 9)
 	 #define RX_TPA_START_CMP_RSS_HASH_TYPE_SHIFT		 9
+	#define RX_TPA_START_CMP_V3_RSS_HASH_TYPE		(0x1ff << 7)
+	 #define RX_TPA_START_CMP_V3_RSS_HASH_TYPE_SHIFT	 7
 	#define RX_TPA_START_CMP_AGG_ID				(0x7f << 25)
 	 #define RX_TPA_START_CMP_AGG_ID_SHIFT			 25
+	#define RX_TPA_START_CMP_AGG_ID_P5			(0x0fff << 16)
+	 #define RX_TPA_START_CMP_AGG_ID_SHIFT_P5		 16
+	#define RX_TPA_START_CMP_METADATA1			(0xf << 28)
+	 #define RX_TPA_START_CMP_METADATA1_SHIFT		 28
+	#define RX_TPA_START_METADATA1_TPID_SEL			(0x7 << 28)
+	#define RX_TPA_START_METADATA1_TPID_8021Q		(0x1 << 28)
+	#define RX_TPA_START_METADATA1_TPID_8021AD		(0x0 << 28)
+	#define RX_TPA_START_METADATA1_VALID			(0x8 << 28)
 
 	__le32 rx_tpa_start_cmp_rss_hash;
 };
@@ -291,10 +449,31 @@ struct rx_tpa_start_cmp {
 	   RX_TPA_START_CMP_RSS_HASH_TYPE) >>				\
 	  RX_TPA_START_CMP_RSS_HASH_TYPE_SHIFT) & RSS_PROFILE_ID_MASK)
 
+#define TPA_START_V3_HASH_TYPE(rx_tpa_start)				\
+	(((le32_to_cpu((rx_tpa_start)->rx_tpa_start_cmp_misc_v1) &	\
+	   RX_TPA_START_CMP_V3_RSS_HASH_TYPE) >>			\
+	  RX_TPA_START_CMP_V3_RSS_HASH_TYPE_SHIFT) & RSS_PROFILE_ID_MASK)
+
 #define TPA_START_AGG_ID(rx_tpa_start)					\
 	((le32_to_cpu((rx_tpa_start)->rx_tpa_start_cmp_misc_v1) &	\
 	 RX_TPA_START_CMP_AGG_ID) >> RX_TPA_START_CMP_AGG_ID_SHIFT)
 
+#define TPA_START_AGG_ID_P5(rx_tpa_start)				\
+	((le32_to_cpu((rx_tpa_start)->rx_tpa_start_cmp_misc_v1) &	\
+	 RX_TPA_START_CMP_AGG_ID_P5) >> RX_TPA_START_CMP_AGG_ID_SHIFT_P5)
+
+#define TPA_START_ERROR(rx_tpa_start)					\
+	((rx_tpa_start)->rx_tpa_start_cmp_len_flags_type &		\
+	 cpu_to_le32(RX_TPA_START_CMP_FLAGS_ERROR))
+
+#define TPA_START_VLAN_VALID(rx_tpa_start)				\
+	((rx_tpa_start)->rx_tpa_start_cmp_misc_v1 &			\
+	 cpu_to_le32(RX_TPA_START_METADATA1_VALID))
+
+#define TPA_START_VLAN_TPID_SEL(rx_tpa_start)				\
+	(le32_to_cpu((rx_tpa_start)->rx_tpa_start_cmp_misc_v1) &	\
+	 RX_TPA_START_METADATA1_TPID_SEL)
+
 struct rx_tpa_start_cmp_ext {
 	__le32 rx_tpa_start_cmp_flags2;
 	#define RX_TPA_START_CMP_FLAGS2_IP_CS_CALC		(0x1 << 0)
@@ -302,15 +481,48 @@ struct rx_tpa_start_cmp_ext {
 	#define RX_TPA_START_CMP_FLAGS2_T_IP_CS_CALC		(0x1 << 2)
 	#define RX_TPA_START_CMP_FLAGS2_T_L4_CS_CALC		(0x1 << 3)
 	#define RX_TPA_START_CMP_FLAGS2_IP_TYPE			(0x1 << 8)
+	#define RX_TPA_START_CMP_FLAGS2_CSUM_CMPL_VALID		(0x1 << 9)
+	#define RX_TPA_START_CMP_FLAGS2_EXT_META_FORMAT		(0x3 << 10)
+	 #define RX_TPA_START_CMP_FLAGS2_EXT_META_FORMAT_SHIFT	 10
+	#define RX_TPA_START_CMP_V3_FLAGS2_T_IP_TYPE		(0x1 << 10)
+	#define RX_TPA_START_CMP_V3_FLAGS2_AGG_GRO		(0x1 << 11)
+	#define RX_TPA_START_CMP_FLAGS2_CSUM_CMPL		(0xffff << 16)
+	 #define RX_TPA_START_CMP_FLAGS2_CSUM_CMPL_SHIFT	 16
 
 	__le32 rx_tpa_start_cmp_metadata;
 	__le32 rx_tpa_start_cmp_cfa_code_v2;
 	#define RX_TPA_START_CMP_V2				(0x1 << 0)
+	#define RX_TPA_START_CMP_ERRORS_BUFFER_ERROR_MASK	(0x7 << 1)
+	 #define RX_TPA_START_CMP_ERRORS_BUFFER_ERROR_SHIFT	 1
+	 #define RX_TPA_START_CMP_ERRORS_BUFFER_ERROR_NO_BUFFER	 (0x0 << 1)
+	 #define RX_TPA_START_CMP_ERRORS_BUFFER_ERROR_BAD_FORMAT (0x3 << 1)
+	 #define RX_TPA_START_CMP_ERRORS_BUFFER_ERROR_FLUSH	 (0x5 << 1)
 	#define RX_TPA_START_CMP_CFA_CODE			(0xffff << 16)
 	 #define RX_TPA_START_CMPL_CFA_CODE_SHIFT		 16
+	#define RX_TPA_START_CMP_METADATA0_TCI_MASK		(0xffff << 16)
+	#define RX_TPA_START_CMP_METADATA0_VID_MASK		(0x0fff << 16)
+	 #define RX_TPA_START_CMP_METADATA0_SFT			 16
 	__le32 rx_tpa_start_cmp_hdr_info;
 };
 
+#define TPA_START_CFA_CODE(rx_tpa_start)				\
+	((le32_to_cpu((rx_tpa_start)->rx_tpa_start_cmp_cfa_code_v2) &	\
+	 RX_TPA_START_CMP_CFA_CODE) >> RX_TPA_START_CMPL_CFA_CODE_SHIFT)
+
+#define TPA_START_IS_IPV6(rx_tpa_start)				\
+	(!!((rx_tpa_start)->rx_tpa_start_cmp_flags2 &		\
+	    cpu_to_le32(RX_TPA_START_CMP_FLAGS2_IP_TYPE)))
+
+#define TPA_START_ERROR_CODE(rx_tpa_start)				\
+	((le32_to_cpu((rx_tpa_start)->rx_tpa_start_cmp_cfa_code_v2) &	\
+	  RX_TPA_START_CMP_ERRORS_BUFFER_ERROR_MASK) >>			\
+	 RX_TPA_START_CMP_ERRORS_BUFFER_ERROR_SHIFT)
+
+#define TPA_START_METADATA0_TCI(rx_tpa_start)				\
+	((le32_to_cpu((rx_tpa_start)->rx_tpa_start_cmp_cfa_code_v2) &	\
+	  RX_TPA_START_CMP_METADATA0_TCI_MASK) >>			\
+	 RX_TPA_START_CMP_METADATA0_SFT)
+
 struct rx_tpa_end_cmp {
 	__le32 rx_tpa_end_cmp_len_flags_type;
 	#define RX_TPA_END_CMP_TYPE				(0x3f << 0)
@@ -340,6 +552,8 @@ struct rx_tpa_end_cmp {
 	 #define RX_TPA_END_CMP_PAYLOAD_OFFSET_SHIFT		 16
 	#define RX_TPA_END_CMP_AGG_ID				(0x7f << 25)
 	 #define RX_TPA_END_CMP_AGG_ID_SHIFT			 25
+	#define RX_TPA_END_CMP_AGG_ID_P5			(0x0fff << 16)
+	 #define RX_TPA_END_CMP_AGG_ID_SHIFT_P5			 16
 
 	__le32 rx_tpa_end_cmp_tsdelta;
 	#define RX_TPA_END_GRO_TS				(0x1 << 31)
@@ -349,6 +563,18 @@ struct rx_tpa_end_cmp {
 	((le32_to_cpu((rx_tpa_end)->rx_tpa_end_cmp_misc_v1) &		\
 	 RX_TPA_END_CMP_AGG_ID) >> RX_TPA_END_CMP_AGG_ID_SHIFT)
 
+#define TPA_END_AGG_ID_P5(rx_tpa_end)					\
+	((le32_to_cpu((rx_tpa_end)->rx_tpa_end_cmp_misc_v1) &		\
+	 RX_TPA_END_CMP_AGG_ID_P5) >> RX_TPA_END_CMP_AGG_ID_SHIFT_P5)
+
+#define TPA_END_PAYLOAD_OFF(rx_tpa_end)					\
+	((le32_to_cpu((rx_tpa_end)->rx_tpa_end_cmp_misc_v1) &		\
+	 RX_TPA_END_CMP_PAYLOAD_OFFSET) >> RX_TPA_END_CMP_PAYLOAD_OFFSET_SHIFT)
+
+#define TPA_END_AGG_BUFS(rx_tpa_end)					\
+	((le32_to_cpu((rx_tpa_end)->rx_tpa_end_cmp_misc_v1) &		\
+	 RX_TPA_END_CMP_AGG_BUFS) >> RX_TPA_END_CMP_AGG_BUFS_SHIFT)
+
 #define TPA_END_TPA_SEGS(rx_tpa_end)					\
 	((le32_to_cpu((rx_tpa_end)->rx_tpa_end_cmp_misc_v1) &		\
 	 RX_TPA_END_CMP_TPA_SEGS) >> RX_TPA_END_CMP_TPA_SEGS_SHIFT)
@@ -368,18 +594,105 @@ struct rx_tpa_end_cmp {
 struct rx_tpa_end_cmp_ext {
 	__le32 rx_tpa_end_cmp_dup_acks;
 	#define RX_TPA_END_CMP_TPA_DUP_ACKS			(0xf << 0)
+	#define RX_TPA_END_CMP_PAYLOAD_OFFSET_P5		(0xff << 16)
+	 #define RX_TPA_END_CMP_PAYLOAD_OFFSET_SHIFT_P5		 16
+	#define RX_TPA_END_CMP_AGG_BUFS_P5			(0xff << 24)
+	 #define RX_TPA_END_CMP_AGG_BUFS_SHIFT_P5		 24
 
 	__le32 rx_tpa_end_cmp_seg_len;
 	#define RX_TPA_END_CMP_TPA_SEG_LEN			(0xffff << 0)
 
 	__le32 rx_tpa_end_cmp_errors_v2;
 	#define RX_TPA_END_CMP_V2				(0x1 << 0)
-	#define RX_TPA_END_CMP_ERRORS				(0x7fff << 1)
+	#define RX_TPA_END_CMP_ERRORS				(0x3 << 1)
+	#define RX_TPA_END_CMP_ERRORS_P5			(0x7 << 1)
 	#define RX_TPA_END_CMPL_ERRORS_SHIFT			 1
+	 #define RX_TPA_END_CMP_ERRORS_BUFFER_ERROR_NO_BUFFER	 (0x0 << 1)
+	 #define RX_TPA_END_CMP_ERRORS_BUFFER_ERROR_NOT_ON_CHIP	 (0x2 << 1)
+	 #define RX_TPA_END_CMP_ERRORS_BUFFER_ERROR_BAD_FORMAT	 (0x3 << 1)
+	 #define RX_TPA_END_CMP_ERRORS_BUFFER_ERROR_RSV_ERROR	 (0x4 << 1)
+	 #define RX_TPA_END_CMP_ERRORS_BUFFER_ERROR_FLUSH	 (0x5 << 1)
 
 	u32 rx_tpa_end_cmp_start_opaque;
 };
 
+#define TPA_END_ERRORS(rx_tpa_end_ext)					\
+	((rx_tpa_end_ext)->rx_tpa_end_cmp_errors_v2 &			\
+	 cpu_to_le32(RX_TPA_END_CMP_ERRORS))
+
+#define TPA_END_PAYLOAD_OFF_P5(rx_tpa_end_ext)				\
+	((le32_to_cpu((rx_tpa_end_ext)->rx_tpa_end_cmp_dup_acks) &	\
+	 RX_TPA_END_CMP_PAYLOAD_OFFSET_P5) >>				\
+	RX_TPA_END_CMP_PAYLOAD_OFFSET_SHIFT_P5)
+
+#define TPA_END_AGG_BUFS_P5(rx_tpa_end_ext)				\
+	((le32_to_cpu((rx_tpa_end_ext)->rx_tpa_end_cmp_dup_acks) &	\
+	 RX_TPA_END_CMP_AGG_BUFS_P5) >> RX_TPA_END_CMP_AGG_BUFS_SHIFT_P5)
+
+#define EVENT_DATA1_RESET_NOTIFY_FATAL(data1)				\
+	(((data1) &							\
+	  ASYNC_EVENT_CMPL_RESET_NOTIFY_EVENT_DATA1_REASON_CODE_MASK) ==\
+	 ASYNC_EVENT_CMPL_RESET_NOTIFY_EVENT_DATA1_REASON_CODE_FW_EXCEPTION_FATAL)
+
+#define EVENT_DATA1_RESET_NOTIFY_FW_ACTIVATION(data1)			\
+	(((data1) &							\
+	  ASYNC_EVENT_CMPL_RESET_NOTIFY_EVENT_DATA1_REASON_CODE_MASK) ==\
+	ASYNC_EVENT_CMPL_RESET_NOTIFY_EVENT_DATA1_REASON_CODE_FW_ACTIVATION)
+
+#define EVENT_DATA2_RESET_NOTIFY_FW_STATUS_CODE(data2)			\
+	((data2) &							\
+	ASYNC_EVENT_CMPL_RESET_NOTIFY_EVENT_DATA2_FW_STATUS_CODE_MASK)
+
+#define EVENT_DATA1_RECOVERY_MASTER_FUNC(data1)				\
+	!!((data1) &							\
+	   ASYNC_EVENT_CMPL_ERROR_RECOVERY_EVENT_DATA1_FLAGS_MASTER_FUNC)
+
+#define EVENT_DATA1_RECOVERY_ENABLED(data1)				\
+	!!((data1) &							\
+	   ASYNC_EVENT_CMPL_ERROR_RECOVERY_EVENT_DATA1_FLAGS_RECOVERY_ENABLED)
+
+#define BNXT_EVENT_ERROR_REPORT_TYPE(data1)				\
+	(((data1) &							\
+	  ASYNC_EVENT_CMPL_ERROR_REPORT_BASE_EVENT_DATA1_ERROR_TYPE_MASK) >>\
+	 ASYNC_EVENT_CMPL_ERROR_REPORT_BASE_EVENT_DATA1_ERROR_TYPE_SFT)
+
+#define BNXT_EVENT_INVALID_SIGNAL_DATA(data2)				\
+	(((data2) &							\
+	  ASYNC_EVENT_CMPL_ERROR_REPORT_INVALID_SIGNAL_EVENT_DATA2_PIN_ID_MASK) >>\
+	 ASYNC_EVENT_CMPL_ERROR_REPORT_INVALID_SIGNAL_EVENT_DATA2_PIN_ID_SFT)
+
+struct nqe_cn {
+	__le16	type;
+	#define NQ_CN_TYPE_MASK           0x3fUL
+	#define NQ_CN_TYPE_SFT            0
+	#define NQ_CN_TYPE_CQ_NOTIFICATION  0x30UL
+	#define NQ_CN_TYPE_LAST            NQ_CN_TYPE_CQ_NOTIFICATION
+	#define NQ_CN_TOGGLE_MASK         0xc0UL
+	#define NQ_CN_TOGGLE_SFT          6
+	__le16	reserved16;
+	__le32	cq_handle_low;
+	__le32	v;
+	#define NQ_CN_V     0x1UL
+	__le32	cq_handle_high;
+};
+
+#define BNXT_NQ_HDL_IDX_MASK	0x00ffffff
+#define BNXT_NQ_HDL_TYPE_MASK	0xff000000
+#define BNXT_NQ_HDL_TYPE_SHIFT	24
+#define BNXT_NQ_HDL_TYPE_RX	0x00
+#define BNXT_NQ_HDL_TYPE_TX	0x01
+
+#define BNXT_NQ_HDL_IDX(hdl)	((hdl) & BNXT_NQ_HDL_IDX_MASK)
+#define BNXT_NQ_HDL_TYPE(hdl)	(((hdl) & BNXT_NQ_HDL_TYPE_MASK) >>	\
+				 BNXT_NQ_HDL_TYPE_SHIFT)
+
+#define BNXT_SET_NQ_HDL(cpr)						\
+	(((cpr)->cp_ring_type << BNXT_NQ_HDL_TYPE_SHIFT) | (cpr)->cp_idx)
+
+#define NQE_CN_TYPE(type)	((type) & NQ_CN_TYPE_MASK)
+#define NQE_CN_TOGGLE(type)	(((type) & NQ_CN_TOGGLE_MASK) >>	\
+				 NQ_CN_TOGGLE_SFT)
+
 #define DB_IDX_MASK						0xffffff
 #define DB_IDX_VALID						(0x1 << 26)
 #define DB_IRQ_DIS						(0x1 << 27)
@@ -393,6 +706,34 @@ struct rx_tpa_end_cmp_ext {
 #define BNXT_MIN_ROCE_CP_RINGS	2
 #define BNXT_MIN_ROCE_STAT_CTXS	1
 
+/* 64-bit doorbell */
+#define DBR_INDEX_MASK					0x0000000000ffffffULL
+#define DBR_EPOCH_MASK					0x01000000UL
+#define DBR_EPOCH_SFT					24
+#define DBR_TOGGLE_MASK					0x06000000UL
+#define DBR_TOGGLE_SFT					25
+#define DBR_XID_MASK					0x000fffff00000000ULL
+#define DBR_XID_SFT					32
+#define DBR_PATH_L2					(0x1ULL << 56)
+#define DBR_VALID					(0x1ULL << 58)
+#define DBR_TYPE_SQ					(0x0ULL << 60)
+#define DBR_TYPE_RQ					(0x1ULL << 60)
+#define DBR_TYPE_SRQ					(0x2ULL << 60)
+#define DBR_TYPE_SRQ_ARM				(0x3ULL << 60)
+#define DBR_TYPE_CQ					(0x4ULL << 60)
+#define DBR_TYPE_CQ_ARMSE				(0x5ULL << 60)
+#define DBR_TYPE_CQ_ARMALL				(0x6ULL << 60)
+#define DBR_TYPE_CQ_ARMENA				(0x7ULL << 60)
+#define DBR_TYPE_SRQ_ARMENA				(0x8ULL << 60)
+#define DBR_TYPE_CQ_CUTOFF_ACK				(0x9ULL << 60)
+#define DBR_TYPE_NQ					(0xaULL << 60)
+#define DBR_TYPE_NQ_ARM					(0xbULL << 60)
+#define DBR_TYPE_NQ_MASK				(0xeULL << 60)
+#define DBR_TYPE_NULL					(0xfULL << 60)
+
+#define DB_PF_OFFSET_P5					0x10000
+#define DB_VF_OFFSET_P5					0x4000
+
 #define INVALID_HW_RING_ID	((u16)-1)
 
 /* The hardware supports certain page sizes.  Use the supported page sizes
@@ -420,9 +761,20 @@ struct rx_tpa_end_cmp_ext {
 #define BNXT_RX_PAGE_SIZE (1 << BNXT_RX_PAGE_SHIFT)
 
 #define BNXT_MAX_MTU		9500
-#define BNXT_MAX_PAGE_MODE_MTU	\
+
+/* First RX buffer page in XDP multi-buf mode
+ *
+ * +-------------------------------------------------------------------------+
+ * | XDP_PACKET_HEADROOM | bp->rx_buf_use_size              | skb_shared_info|
+ * | (bp->rx_dma_offset) |                                  |                |
+ * +-------------------------------------------------------------------------+
+ */
+#define BNXT_MAX_PAGE_MODE_MTU_SBUF \
 	((unsigned int)PAGE_SIZE - VLAN_ETH_HLEN - NET_IP_ALIGN -	\
 	 XDP_PACKET_HEADROOM)
+#define BNXT_MAX_PAGE_MODE_MTU	\
+	(BNXT_MAX_PAGE_MODE_MTU_SBUF - \
+	 SKB_DATA_ALIGN((unsigned int)sizeof(struct skb_shared_info)))
 
 #define BNXT_MIN_PKT_SIZE	52
 
@@ -430,17 +782,22 @@ struct rx_tpa_end_cmp_ext {
 #define BNXT_DEFAULT_TX_RING_SIZE	511
 
 #define MAX_TPA		64
+#define MAX_TPA_P5	256
+#define MAX_TPA_P5_MASK	(MAX_TPA_P5 - 1)
+#define MAX_TPA_SEGS_P5	0x3f
 
 #if (BNXT_PAGE_SHIFT == 16)
-#define MAX_RX_PAGES	1
+#define MAX_RX_PAGES_AGG_ENA	1
+#define MAX_RX_PAGES	4
 #define MAX_RX_AGG_PAGES	4
 #define MAX_TX_PAGES	1
-#define MAX_CP_PAGES	8
+#define MAX_CP_PAGES	16
 #else
-#define MAX_RX_PAGES	8
+#define MAX_RX_PAGES_AGG_ENA	8
+#define MAX_RX_PAGES	32
 #define MAX_RX_AGG_PAGES	32
 #define MAX_TX_PAGES	8
-#define MAX_CP_PAGES	64
+#define MAX_CP_PAGES	128
 #endif
 
 #define RX_DESC_CNT (BNXT_PAGE_SIZE / sizeof(struct rx_bd))
@@ -458,13 +815,21 @@ struct rx_tpa_end_cmp_ext {
 #define HW_CMPD_RING_SIZE (sizeof(struct tx_cmp) * CP_DESC_CNT)
 
 #define BNXT_MAX_RX_DESC_CNT		(RX_DESC_CNT * MAX_RX_PAGES - 1)
+#define BNXT_MAX_RX_DESC_CNT_JUM_ENA	(RX_DESC_CNT * MAX_RX_PAGES_AGG_ENA - 1)
 #define BNXT_MAX_RX_JUM_DESC_CNT	(RX_DESC_CNT * MAX_RX_AGG_PAGES - 1)
 #define BNXT_MAX_TX_DESC_CNT		(TX_DESC_CNT * MAX_TX_PAGES - 1)
 
-#define RX_RING(x)	(((x) & ~(RX_DESC_CNT - 1)) >> (BNXT_PAGE_SHIFT - 4))
+/* Minimum TX BDs for a TX packet with MAX_SKB_FRAGS + 1.  We need one extra
+ * BD because the first TX BD is always a long BD.
+ */
+#define BNXT_MIN_TX_DESC_CNT		(MAX_SKB_FRAGS + 2)
+
+#define RX_RING(bp, x)	(((x) & (bp)->rx_ring_mask) >> (BNXT_PAGE_SHIFT - 4))
+#define RX_AGG_RING(bp, x)	(((x) & (bp)->rx_agg_ring_mask) >>	\
+				 (BNXT_PAGE_SHIFT - 4))
 #define RX_IDX(x)	((x) & (RX_DESC_CNT - 1))
 
-#define TX_RING(x)	(((x) & ~(TX_DESC_CNT - 1)) >> (BNXT_PAGE_SHIFT - 4))
+#define TX_RING(bp, x)	(((x) & (bp)->tx_ring_mask) >> (BNXT_PAGE_SHIFT - 4))
 #define TX_IDX(x)	((x) & (TX_DESC_CNT - 1))
 
 #define CP_RING(x)	(((x) & ~(CP_DESC_CNT - 1)) >> (BNXT_PAGE_SHIFT - 4))
@@ -482,50 +847,52 @@ struct rx_tpa_end_cmp_ext {
 	(!!((agg)->rx_agg_cmp_v & cpu_to_le32(RX_AGG_CMP_V)) ==	\
 	 !((raw_cons) & bp->cp_bit))
 
+#define NQ_CMP_VALID(nqcmp, raw_cons)				\
+	(!!((nqcmp)->v & cpu_to_le32(NQ_CN_V)) == !((raw_cons) & bp->cp_bit))
+
 #define TX_CMP_TYPE(txcmp)					\
 	(le32_to_cpu((txcmp)->tx_cmp_flags_type) & CMP_TYPE)
 
 #define RX_CMP_TYPE(rxcmp)					\
 	(le32_to_cpu((rxcmp)->rx_cmp_len_flags_type) & RX_CMP_CMP_TYPE)
 
-#define NEXT_RX(idx)		(((idx) + 1) & bp->rx_ring_mask)
+#define RING_RX(bp, idx)	((idx) & (bp)->rx_ring_mask)
+#define NEXT_RX(idx)		((idx) + 1)
 
-#define NEXT_RX_AGG(idx)	(((idx) + 1) & bp->rx_agg_ring_mask)
+#define RING_RX_AGG(bp, idx)	((idx) & (bp)->rx_agg_ring_mask)
+#define NEXT_RX_AGG(idx)	((idx) + 1)
 
-#define NEXT_TX(idx)		(((idx) + 1) & bp->tx_ring_mask)
+#define RING_TX(bp, idx)	((idx) & (bp)->tx_ring_mask)
+#define NEXT_TX(idx)		((idx) + 1)
 
 #define ADV_RAW_CMP(idx, n)	((idx) + (n))
 #define NEXT_RAW_CMP(idx)	ADV_RAW_CMP(idx, 1)
 #define RING_CMP(idx)		((idx) & bp->cp_ring_mask)
 #define NEXT_CMP(idx)		RING_CMP(ADV_RAW_CMP(idx, 1))
 
-#define BNXT_HWRM_MAX_REQ_LEN		(bp->hwrm_max_req_len)
-#define BNXT_HWRM_SHORT_REQ_LEN		sizeof(struct hwrm_short_input)
 #define DFLT_HWRM_CMD_TIMEOUT		500
-#define HWRM_CMD_TIMEOUT		(bp->hwrm_cmd_timeout)
-#define HWRM_RESET_TIMEOUT		((HWRM_CMD_TIMEOUT) * 4)
-#define HWRM_RESP_ERR_CODE_MASK		0xffff
-#define HWRM_RESP_LEN_OFFSET		4
-#define HWRM_RESP_LEN_MASK		0xffff0000
-#define HWRM_RESP_LEN_SFT		16
-#define HWRM_RESP_VALID_MASK		0xff000000
-#define HWRM_SEQ_ID_INVALID		-1
-#define BNXT_HWRM_REQ_MAX_SIZE		128
-#define BNXT_HWRM_REQS_PER_PAGE		(BNXT_PAGE_SIZE /	\
-					 BNXT_HWRM_REQ_MAX_SIZE)
-
-#define BNXT_RX_EVENT	1
-#define BNXT_AGG_EVENT	2
-#define BNXT_TX_EVENT	4
+
+#define BNXT_RX_EVENT		1
+#define BNXT_AGG_EVENT		2
+#define BNXT_TX_EVENT		4
+#define BNXT_REDIRECT_EVENT	8
+#define BNXT_TX_CMP_EVENT	0x10
 
 struct bnxt_sw_tx_bd {
-	struct sk_buff		*skb;
+	union {
+		struct sk_buff		*skb;
+		struct xdp_frame	*xdpf;
+	};
 	DEFINE_DMA_UNMAP_ADDR(mapping);
-	u8			is_gso;
+	DEFINE_DMA_UNMAP_LEN(len);
+	struct page		*page;
+	u8			is_ts_pkt;
 	u8			is_push;
+	u8			action;
+	unsigned short		nr_frags;
 	union {
-		unsigned short		nr_frags;
 		u16			rx_prod;
+		u16			txts_prod;
 	};
 };
 
@@ -536,14 +903,22 @@ struct bnxt_sw_rx_bd {
 };
 
 struct bnxt_sw_rx_agg_bd {
-	struct page		*page;
+	netmem_ref		netmem;
 	unsigned int		offset;
 	dma_addr_t		mapping;
 };
 
-struct bnxt_ring_struct {
+struct bnxt_ring_mem_info {
 	int			nr_pages;
 	int			page_size;
+	u16			flags;
+#define BNXT_RMEM_VALID_PTE_FLAG	1
+#define BNXT_RMEM_RING_PTE_FLAG		2
+#define BNXT_RMEM_USE_FULL_PAGE_FLAG	4
+
+	u16			depth;
+	struct bnxt_ctx_mem_type	*ctx_mem;
+
 	void			**pg_arr;
 	dma_addr_t		*dma_arr;
 
@@ -552,8 +927,17 @@ struct bnxt_ring_struct {
 
 	int			vmem_size;
 	void			**vmem;
+};
+
+struct bnxt_ring_struct {
+	struct bnxt_ring_mem_info	ring_mem;
 
 	u16			fw_ring_id; /* Ring id filled by Chimp FW */
+	union {
+		u16		grp_idx;
+		u16		map_idx; /* Used by cmpl rings */
+	};
+	u32			handle;
 	u8			queue_id;
 };
 
@@ -569,12 +953,35 @@ struct tx_push_buffer {
 	u32			data[25];
 };
 
+struct bnxt_db_info {
+	void __iomem		*doorbell;
+	union {
+		u64		db_key64;
+		u32		db_key32;
+	};
+	u32			db_ring_mask;
+	u32			db_epoch_mask;
+	u8			db_epoch_shift;
+};
+
+#define DB_EPOCH(db, idx)	(((idx) & (db)->db_epoch_mask) <<	\
+				 ((db)->db_epoch_shift))
+
+#define DB_TOGGLE(tgl)		((tgl) << DBR_TOGGLE_SFT)
+
+#define DB_RING_IDX(db, idx)	(((idx) & (db)->db_ring_mask) |		\
+				 DB_EPOCH(db, idx))
+
 struct bnxt_tx_ring_info {
 	struct bnxt_napi	*bnapi;
+	struct bnxt_cp_ring_info	*tx_cpr;
 	u16			tx_prod;
 	u16			tx_cons;
+	u16			tx_hw_cons;
 	u16			txq_index;
-	void __iomem		*tx_doorbell;
+	u8			tx_napi_idx;
+	u8			kick_pending;
+	struct bnxt_db_info	tx_db;
 
 	struct tx_bd		*tx_desc_ring[MAX_TX_PAGES];
 	struct bnxt_sw_tx_bd	*tx_buf_ring;
@@ -589,6 +996,56 @@ struct bnxt_tx_ring_info {
 	u32			dev_state;
 
 	struct bnxt_ring_struct	tx_ring_struct;
+	/* Synchronize simultaneous xdp_xmit on same ring */
+	spinlock_t		xdp_tx_lock;
+};
+
+#define BNXT_LEGACY_COAL_CMPL_PARAMS					\
+	(RING_AGGINT_QCAPS_RESP_CMPL_PARAMS_INT_LAT_TMR_MIN |		\
+	 RING_AGGINT_QCAPS_RESP_CMPL_PARAMS_INT_LAT_TMR_MAX |		\
+	 RING_AGGINT_QCAPS_RESP_CMPL_PARAMS_TIMER_RESET |		\
+	 RING_AGGINT_QCAPS_RESP_CMPL_PARAMS_RING_IDLE |			\
+	 RING_AGGINT_QCAPS_RESP_CMPL_PARAMS_NUM_CMPL_DMA_AGGR |		\
+	 RING_AGGINT_QCAPS_RESP_CMPL_PARAMS_NUM_CMPL_DMA_AGGR_DURING_INT | \
+	 RING_AGGINT_QCAPS_RESP_CMPL_PARAMS_CMPL_AGGR_DMA_TMR |		\
+	 RING_AGGINT_QCAPS_RESP_CMPL_PARAMS_CMPL_AGGR_DMA_TMR_DURING_INT | \
+	 RING_AGGINT_QCAPS_RESP_CMPL_PARAMS_NUM_CMPL_AGGR_INT)
+
+#define BNXT_COAL_CMPL_ENABLES						\
+	(RING_CMPL_RING_CFG_AGGINT_PARAMS_REQ_ENABLES_NUM_CMPL_DMA_AGGR | \
+	 RING_CMPL_RING_CFG_AGGINT_PARAMS_REQ_ENABLES_CMPL_AGGR_DMA_TMR | \
+	 RING_CMPL_RING_CFG_AGGINT_PARAMS_REQ_ENABLES_INT_LAT_TMR_MAX | \
+	 RING_CMPL_RING_CFG_AGGINT_PARAMS_REQ_ENABLES_NUM_CMPL_AGGR_INT)
+
+#define BNXT_COAL_CMPL_MIN_TMR_ENABLE					\
+	RING_CMPL_RING_CFG_AGGINT_PARAMS_REQ_ENABLES_INT_LAT_TMR_MIN
+
+#define BNXT_COAL_CMPL_AGGR_TMR_DURING_INT_ENABLE			\
+	RING_CMPL_RING_CFG_AGGINT_PARAMS_REQ_ENABLES_NUM_CMPL_DMA_AGGR_DURING_INT
+
+struct bnxt_coal_cap {
+	u32			cmpl_params;
+	u32			nq_params;
+	u16			num_cmpl_dma_aggr_max;
+	u16			num_cmpl_dma_aggr_during_int_max;
+	u16			cmpl_aggr_dma_tmr_max;
+	u16			cmpl_aggr_dma_tmr_during_int_max;
+	u16			int_lat_tmr_min_max;
+	u16			int_lat_tmr_max_max;
+	u16			num_cmpl_aggr_int_max;
+	u16			timer_units;
+};
+
+struct bnxt_coal {
+	u16			coal_ticks;
+	u16			coal_ticks_irq;
+	u16			coal_bufs;
+	u16			coal_bufs_irq;
+			/* RING_IDLE enabled when coal ticks < idle_thresh  */
+	u16			idle_thresh;
+	u8			bufs_per_record;
+	u8			budget;
+	u16			flags;
 };
 
 struct bnxt_tpa_info {
@@ -614,16 +1071,30 @@ struct bnxt_tpa_info {
 
 #define BNXT_TPA_OUTER_L3_OFF(hdr_info)	\
 	((hdr_info) & 0x1ff)
+
+	u16			cfa_code; /* cfa_code in TPA start compl */
+	u8			agg_count;
+	u8			vlan_valid:1;
+	u8			cfa_code_valid:1;
+	struct rx_agg_cmp	*agg_arr;
+};
+
+#define BNXT_AGG_IDX_BMAP_SIZE	(MAX_TPA_P5 / BITS_PER_LONG)
+
+struct bnxt_tpa_idx_map {
+	u16		agg_id_tbl[1024];
+	unsigned long	agg_idx_bmap[BNXT_AGG_IDX_BMAP_SIZE];
 };
 
 struct bnxt_rx_ring_info {
 	struct bnxt_napi	*bnapi;
+	struct bnxt_cp_ring_info	*rx_cpr;
 	u16			rx_prod;
 	u16			rx_agg_prod;
 	u16			rx_sw_agg_prod;
 	u16			rx_next_cons;
-	void __iomem		*rx_doorbell;
-	void __iomem		*rx_agg_doorbell;
+	struct bnxt_db_info	rx_db;
+	struct bnxt_db_info	rx_agg_db;
 
 	struct bpf_prog		*xdp_prog;
 
@@ -635,35 +1106,111 @@ struct bnxt_rx_ring_info {
 
 	unsigned long		*rx_agg_bmap;
 	u16			rx_agg_bmap_size;
-
-	struct page		*rx_page;
-	unsigned int		rx_page_offset;
+	bool                    need_head_pool;
 
 	dma_addr_t		rx_desc_mapping[MAX_RX_PAGES];
 	dma_addr_t		rx_agg_desc_mapping[MAX_RX_AGG_PAGES];
 
 	struct bnxt_tpa_info	*rx_tpa;
+	struct bnxt_tpa_idx_map *rx_tpa_idx_map;
 
 	struct bnxt_ring_struct	rx_ring_struct;
 	struct bnxt_ring_struct	rx_agg_ring_struct;
+	struct xdp_rxq_info	xdp_rxq;
+	struct page_pool	*page_pool;
+	struct page_pool	*head_pool;
+};
+
+struct bnxt_rx_sw_stats {
+	u64			rx_l4_csum_errors;
+	u64			rx_resets;
+	u64			rx_buf_errors;
+	u64			rx_oom_discards;
+	u64			rx_netpoll_discards;
+};
+
+struct bnxt_tx_sw_stats {
+	u64			tx_resets;
+};
+
+struct bnxt_cmn_sw_stats {
+	u64			missed_irqs;
+};
+
+struct bnxt_sw_stats {
+	struct bnxt_rx_sw_stats rx;
+	struct bnxt_tx_sw_stats tx;
+	struct bnxt_cmn_sw_stats cmn;
+};
+
+struct bnxt_total_ring_err_stats {
+	u64			rx_total_l4_csum_errors;
+	u64			rx_total_resets;
+	u64			rx_total_buf_errors;
+	u64			rx_total_oom_discards;
+	u64			rx_total_netpoll_discards;
+	u64			rx_total_ring_discards;
+	u64			tx_total_resets;
+	u64			tx_total_ring_discards;
+	u64			total_missed_irqs;
+};
+
+struct bnxt_stats_mem {
+	u64		*sw_stats;
+	u64		*hw_masks;
+	void		*hw_stats;
+	dma_addr_t	hw_stats_map;
+	int		len;
 };
 
 struct bnxt_cp_ring_info {
+	struct bnxt_napi	*bnapi;
 	u32			cp_raw_cons;
-	void __iomem		*cp_doorbell;
+	struct bnxt_db_info	cp_db;
+
+	u8			had_work_done:1;
+	u8			has_more_work:1;
+	u8			had_nqe_notify:1;
+	u8			toggle;
 
-	struct tx_cmp		*cp_desc_ring[MAX_CP_PAGES];
+	u8			cp_ring_type;
+	u8			cp_idx;
+
+	u32			last_cp_raw_cons;
+
+	struct bnxt_coal	rx_ring_coal;
+	u64			rx_packets;
+	u64			rx_bytes;
+	u64			event_ctr;
+
+	struct dim		dim;
+
+	union {
+		struct tx_cmp	**cp_desc_ring;
+		struct nqe_cn	**nq_desc_ring;
+	};
 
-	dma_addr_t		cp_desc_mapping[MAX_CP_PAGES];
+	dma_addr_t		*cp_desc_mapping;
 
-	struct ctx_hw_stats	*hw_stats;
-	dma_addr_t		hw_stats_map;
+	struct bnxt_stats_mem	stats;
 	u32			hw_stats_ctx_id;
-	u64			rx_l4_csum_errors;
+
+	struct bnxt_sw_stats	*sw_stats;
 
 	struct bnxt_ring_struct	cp_ring_struct;
+
+	int			cp_ring_count;
+	struct bnxt_cp_ring_info *cp_ring_arr;
 };
 
+#define BNXT_MAX_QUEUE		8
+#define BNXT_MAX_TXR_PER_NAPI	BNXT_MAX_QUEUE
+
+#define bnxt_for_each_napi_tx(iter, bnapi, txr)		\
+	for (iter = 0, txr = (bnapi)->tx_ring[0]; txr;	\
+	     txr = (iter < BNXT_MAX_TXR_PER_NAPI - 1) ?	\
+	     (bnapi)->tx_ring[++iter] : NULL)
+
 struct bnxt_napi {
 	struct napi_struct	napi;
 	struct bnxt		*bp;
@@ -671,27 +1218,41 @@ struct bnxt_napi {
 	int			index;
 	struct bnxt_cp_ring_info	cp_ring;
 	struct bnxt_rx_ring_info	*rx_ring;
-	struct bnxt_tx_ring_info	*tx_ring;
+	struct bnxt_tx_ring_info	*tx_ring[BNXT_MAX_TXR_PER_NAPI];
 
 	void			(*tx_int)(struct bnxt *, struct bnxt_napi *,
-					  int);
+					  int budget);
+	u8			events;
+	u8			tx_fault:1;
+
 	u32			flags;
 #define BNXT_NAPI_FLAG_XDP	0x1
 
 	bool			in_reset;
 };
 
+/* "TxRx", 2 hypens, plus maximum integer */
+#define BNXT_IRQ_NAME_EXTRA	17
+
 struct bnxt_irq {
 	irq_handler_t	handler;
 	unsigned int	vector;
-	u8		requested;
-	char		name[IFNAMSIZ + 2];
+	u8		requested:1;
+	u8		have_cpumask:1;
+	char		name[IFNAMSIZ + BNXT_IRQ_NAME_EXTRA];
+	cpumask_var_t	cpu_mask;
+
+	struct bnxt	*bp;
+	int		msix_nr;
+	int		ring_nr;
+	struct irq_affinity_notify affinity_notify;
 };
 
 #define HWRM_RING_ALLOC_TX	0x1
 #define HWRM_RING_ALLOC_RX	0x2
 #define HWRM_RING_ALLOC_AGG	0x4
 #define HWRM_RING_ALLOC_CMPL	0x8
+#define HWRM_RING_ALLOC_NQ	0x10
 
 #define INVALID_STATS_CTX_ID	-1
 
@@ -703,13 +1264,17 @@ struct bnxt_ring_grp_info {
 	u16	cp_fw_ring_id;
 };
 
+#define BNXT_VNIC_DEFAULT	0
+#define BNXT_VNIC_NTUPLE	1
+
 struct bnxt_vnic_info {
 	u16		fw_vnic_id; /* returned by Chimp during alloc */
-#define BNXT_MAX_CTX_PER_VNIC	2
+#define BNXT_MAX_CTX_PER_VNIC	8
 	u16		fw_rss_cos_lb_ctx[BNXT_MAX_CTX_PER_VNIC];
 	u16		fw_l2_ctx_id;
+	u16		mru;
 #define BNXT_MAX_UC_ADDRS	4
-	__le64		fw_l2_filter_id[BNXT_MAX_UC_ADDRS];
+	struct bnxt_l2_filter *l2_filters[BNXT_MAX_UC_ADDRS];
 				/* index 0 always dev_addr */
 	u16		uc_filter_count;
 	u8		*uc_list;
@@ -719,6 +1284,15 @@ struct bnxt_vnic_info {
 	__le16		*rss_table;
 	dma_addr_t	rss_hash_key_dma_addr;
 	u64		*rss_hash_key;
+	int		rss_table_size;
+#define BNXT_RSS_TABLE_ENTRIES_P5	64
+#define BNXT_RSS_TABLE_SIZE_P5		(BNXT_RSS_TABLE_ENTRIES_P5 * 4)
+#define BNXT_RSS_TABLE_MAX_TBL_P5	8
+#define BNXT_MAX_RSS_TABLE_SIZE_P5				\
+	(BNXT_RSS_TABLE_SIZE_P5 * BNXT_RSS_TABLE_MAX_TBL_P5)
+#define BNXT_MAX_RSS_TABLE_ENTRIES_P5				\
+	(BNXT_RSS_TABLE_ENTRIES_P5 * BNXT_RSS_TABLE_MAX_TBL_P5)
+
 	u32		rx_mask;
 
 	u8		*mc_list;
@@ -733,28 +1307,81 @@ struct bnxt_vnic_info {
 #define BNXT_VNIC_MCAST_FLAG	4
 #define BNXT_VNIC_UCAST_FLAG	8
 #define BNXT_VNIC_RFS_NEW_RSS_FLAG	0x10
+#define BNXT_VNIC_NTUPLE_FLAG		0x20
+#define BNXT_VNIC_RSSCTX_FLAG		0x40
+	struct ethtool_rxfh_context *rss_ctx;
+	u32		vnic_id;
 };
 
-#if defined(CONFIG_BNXT_SRIOV)
-struct bnxt_vf_info {
-	u16	fw_fid;
-	u8	mac_addr[ETH_ALEN];
+struct bnxt_rss_ctx {
+	struct bnxt_vnic_info vnic;
+	u8	index;
+};
+
+#define BNXT_MAX_ETH_RSS_CTX	32
+#define BNXT_VNIC_ID_INVALID	0xffffffff
+
+struct bnxt_hw_rings {
+	int tx;
+	int rx;
+	int grp;
+	int cp;
+	int cp_p5;
+	int stat;
+	int vnic;
+	int rss_ctx;
+};
+
+struct bnxt_hw_resc {
+	u16	min_rsscos_ctxs;
 	u16	max_rsscos_ctxs;
+	u16	resv_rsscos_ctxs;
+	u16	min_cp_rings;
 	u16	max_cp_rings;
+	u16	resv_cp_rings;
+	u16	min_tx_rings;
 	u16	max_tx_rings;
+	u16	resv_tx_rings;
+	u16	max_tx_sch_inputs;
+	u16	min_rx_rings;
 	u16	max_rx_rings;
+	u16	resv_rx_rings;
+	u16	min_hw_ring_grps;
 	u16	max_hw_ring_grps;
+	u16	resv_hw_ring_grps;
+	u16	min_l2_ctxs;
 	u16	max_l2_ctxs;
-	u16	max_irqs;
+	u16	min_vnics;
 	u16	max_vnics;
+	u16	resv_vnics;
+	u16	min_stat_ctxs;
 	u16	max_stat_ctxs;
+	u16	resv_stat_ctxs;
+	u16	max_nqs;
+	u16	max_irqs;
+	u16	resv_irqs;
+	u32	max_encap_records;
+	u32	max_decap_records;
+	u32	max_tx_em_flows;
+	u32	max_tx_wm_flows;
+	u32	max_rx_em_flows;
+	u32	max_rx_wm_flows;
+};
+
+#if defined(CONFIG_BNXT_SRIOV)
+struct bnxt_vf_info {
+	u16	fw_fid;
+	u8	mac_addr[ETH_ALEN];	/* PF assigned MAC Address */
+	u8	vf_mac_addr[ETH_ALEN];	/* VF assigned MAC address, only
+					 * stored by PF.
+					 */
 	u16	vlan;
+	u16	func_qcfg_flags;
 	u32	flags;
-#define BNXT_VF_QOS		0x1
 #define BNXT_VF_SPOOFCHK	0x2
 #define BNXT_VF_LINK_FORCED	0x4
 #define BNXT_VF_LINK_UP		0x8
-	u32	func_flags; /* func cfg flags */
+#define BNXT_VF_TRUST		0x10
 	u32	min_tx_rate;
 	u32	max_tx_rate;
 	void	*hwrm_cmd_req_addr;
@@ -768,44 +1395,142 @@ struct bnxt_pf_info {
 	u16	fw_fid;
 	u16	port_id;
 	u8	mac_addr[ETH_ALEN];
-	u16	max_rsscos_ctxs;
-	u16	max_cp_rings;
-	u16	max_tx_rings; /* HW assigned max tx rings for this PF */
-	u16	max_rx_rings; /* HW assigned max rx rings for this PF */
-	u16	max_hw_ring_grps;
-	u16	max_irqs;
-	u16	max_l2_ctxs;
-	u16	max_vnics;
-	u16	max_stat_ctxs;
 	u32	first_vf_id;
 	u16	active_vfs;
+	u16	registered_vfs;
 	u16	max_vfs;
-	u32	max_encap_records;
-	u32	max_decap_records;
-	u32	max_tx_em_flows;
-	u32	max_tx_wm_flows;
-	u32	max_rx_em_flows;
-	u32	max_rx_wm_flows;
 	unsigned long	*vf_event_bmap;
 	u16	hwrm_cmd_req_pages;
+	u8	vf_resv_strategy;
+#define BNXT_VF_RESV_STRATEGY_MAXIMAL	0
+#define BNXT_VF_RESV_STRATEGY_MINIMAL	1
+#define BNXT_VF_RESV_STRATEGY_MINIMAL_STATIC	2
 	void			*hwrm_cmd_req_addr[4];
 	dma_addr_t		hwrm_cmd_req_dma_addr[4];
 	struct bnxt_vf_info	*vf;
 };
 
-struct bnxt_ntuple_filter {
+struct bnxt_filter_base {
 	struct hlist_node	hash;
-	u8			dst_mac_addr[ETH_ALEN];
-	u8			src_mac_addr[ETH_ALEN];
-	struct flow_keys	fkeys;
+	struct list_head	list;
 	__le64			filter_id;
+	u8			type;
+#define BNXT_FLTR_TYPE_NTUPLE	1
+#define BNXT_FLTR_TYPE_L2	2
+	u8			flags;
+#define BNXT_ACT_DROP		1
+#define BNXT_ACT_RING_DST	2
+#define BNXT_ACT_FUNC_DST	4
+#define BNXT_ACT_NO_AGING	8
+#define BNXT_ACT_RSS_CTX	0x10
 	u16			sw_id;
-	u8			l2_fltr_idx;
 	u16			rxq;
-	u32			flow_id;
+	u16			fw_vnic_id;
+	u16			vf_idx;
 	unsigned long		state;
 #define BNXT_FLTR_VALID		0
-#define BNXT_FLTR_UPDATE	1
+#define BNXT_FLTR_INSERTED	1
+#define BNXT_FLTR_FW_DELETED	2
+
+	struct rcu_head         rcu;
+};
+
+struct bnxt_flow_masks {
+	struct flow_dissector_key_ports ports;
+	struct flow_dissector_key_addrs addrs;
+};
+
+extern const struct bnxt_flow_masks BNXT_FLOW_MASK_NONE;
+extern const struct bnxt_flow_masks BNXT_FLOW_IPV6_MASK_ALL;
+extern const struct bnxt_flow_masks BNXT_FLOW_IPV4_MASK_ALL;
+
+struct bnxt_ntuple_filter {
+	/* base filter must be the first member */
+	struct bnxt_filter_base	base;
+	struct flow_keys	fkeys;
+	struct bnxt_flow_masks	fmasks;
+	struct bnxt_l2_filter	*l2_fltr;
+	u32			flow_id;
+};
+
+struct bnxt_l2_key {
+	union {
+		struct {
+			u8	dst_mac_addr[ETH_ALEN];
+			u16	vlan;
+		};
+		u32	filter_key;
+	};
+};
+
+struct bnxt_ipv4_tuple {
+	struct flow_dissector_key_ipv4_addrs v4addrs;
+	struct flow_dissector_key_ports ports;
+};
+
+struct bnxt_ipv6_tuple {
+	struct flow_dissector_key_ipv6_addrs v6addrs;
+	struct flow_dissector_key_ports ports;
+};
+
+#define BNXT_L2_KEY_SIZE	(sizeof(struct bnxt_l2_key) / 4)
+
+struct bnxt_l2_filter {
+	/* base filter must be the first member */
+	struct bnxt_filter_base	base;
+	struct bnxt_l2_key	l2_key;
+	atomic_t		refcnt;
+};
+
+/* Compat version of hwrm_port_phy_qcfg_output capped at 96 bytes.  The
+ * first 95 bytes are identical to hwrm_port_phy_qcfg_output in bnxt_hsi.h.
+ * The last valid byte in the compat version is different.
+ */
+struct hwrm_port_phy_qcfg_output_compat {
+	__le16	error_code;
+	__le16	req_type;
+	__le16	seq_id;
+	__le16	resp_len;
+	u8	link;
+	u8	active_fec_signal_mode;
+	__le16	link_speed;
+	u8	duplex_cfg;
+	u8	pause;
+	__le16	support_speeds;
+	__le16	force_link_speed;
+	u8	auto_mode;
+	u8	auto_pause;
+	__le16	auto_link_speed;
+	__le16	auto_link_speed_mask;
+	u8	wirespeed;
+	u8	lpbk;
+	u8	force_pause;
+	u8	module_status;
+	__le32	preemphasis;
+	u8	phy_maj;
+	u8	phy_min;
+	u8	phy_bld;
+	u8	phy_type;
+	u8	media_type;
+	u8	xcvr_pkg_type;
+	u8	eee_config_phy_addr;
+	u8	parallel_detect;
+	__le16	link_partner_adv_speeds;
+	u8	link_partner_adv_auto_mode;
+	u8	link_partner_adv_pause;
+	__le16	adv_eee_link_speed_mask;
+	__le16	link_partner_adv_eee_link_speed_mask;
+	__le32	xcvr_identifier_type_tx_lpi_timer;
+	__le16	fec_cfg;
+	u8	duplex_state;
+	u8	option_flags;
+	char	phy_vendor_name[16];
+	char	phy_vendor_partnumber[16];
+	__le16	support_pam4_speeds;
+	__le16	force_pam4_link_speed;
+	__le16	auto_pam4_link_speed_mask;
+	u8	link_partner_pam4_adv_speeds;
+	u8	valid;
 };
 
 struct bnxt_link_info {
@@ -818,11 +1543,19 @@ struct bnxt_link_info {
 #define BNXT_LINK_SIGNAL	PORT_PHY_QCFG_RESP_LINK_SIGNAL
 #define BNXT_LINK_LINK		PORT_PHY_QCFG_RESP_LINK_LINK
 	u8			wire_speed;
-	u8			loop_back;
-	u8			link_up;
+	u8			phy_state;
+#define BNXT_PHY_STATE_ENABLED		0
+#define BNXT_PHY_STATE_DISABLED		1
+
+	u8			link_state;
+#define BNXT_LINK_STATE_UNKNOWN	0
+#define BNXT_LINK_STATE_DOWN	1
+#define BNXT_LINK_STATE_UP	2
+#define BNXT_LINK_IS_UP(bp)	((bp)->link_info.link_state == BNXT_LINK_STATE_UP)
+	u8			active_lanes;
 	u8			duplex;
-#define BNXT_LINK_DUPLEX_HALF	PORT_PHY_QCFG_RESP_DUPLEX_HALF
-#define BNXT_LINK_DUPLEX_FULL	PORT_PHY_QCFG_RESP_DUPLEX_FULL
+#define BNXT_LINK_DUPLEX_HALF	PORT_PHY_QCFG_RESP_DUPLEX_STATE_HALF
+#define BNXT_LINK_DUPLEX_FULL	PORT_PHY_QCFG_RESP_DUPLEX_STATE_FULL
 	u8			pause;
 #define BNXT_LINK_PAUSE_TX	PORT_PHY_QCFG_RESP_PAUSE_TX
 #define BNXT_LINK_PAUSE_RX	PORT_PHY_QCFG_RESP_PAUSE_RX
@@ -853,7 +1586,12 @@ struct bnxt_link_info {
 #define BNXT_LINK_SPEED_40GB	PORT_PHY_QCFG_RESP_LINK_SPEED_40GB
 #define BNXT_LINK_SPEED_50GB	PORT_PHY_QCFG_RESP_LINK_SPEED_50GB
 #define BNXT_LINK_SPEED_100GB	PORT_PHY_QCFG_RESP_LINK_SPEED_100GB
+#define BNXT_LINK_SPEED_200GB	PORT_PHY_QCFG_RESP_LINK_SPEED_200GB
+#define BNXT_LINK_SPEED_400GB	PORT_PHY_QCFG_RESP_LINK_SPEED_400GB
 	u16			support_speeds;
+	u16			support_pam4_speeds;
+	u16			support_speeds2;
+
 	u16			auto_link_speeds;	/* fw adv setting */
 #define BNXT_LINK_SPEED_MSK_100MB PORT_PHY_QCFG_RESP_SUPPORT_SPEEDS_100MB
 #define BNXT_LINK_SPEED_MSK_1GB PORT_PHY_QCFG_RESP_SUPPORT_SPEEDS_1GB
@@ -865,33 +1603,146 @@ struct bnxt_link_info {
 #define BNXT_LINK_SPEED_MSK_40GB PORT_PHY_QCFG_RESP_SUPPORT_SPEEDS_40GB
 #define BNXT_LINK_SPEED_MSK_50GB PORT_PHY_QCFG_RESP_SUPPORT_SPEEDS_50GB
 #define BNXT_LINK_SPEED_MSK_100GB PORT_PHY_QCFG_RESP_SUPPORT_SPEEDS_100GB
+	u16			auto_pam4_link_speeds;
+#define BNXT_LINK_PAM4_SPEED_MSK_50GB PORT_PHY_QCFG_RESP_SUPPORT_PAM4_SPEEDS_50G
+#define BNXT_LINK_PAM4_SPEED_MSK_100GB PORT_PHY_QCFG_RESP_SUPPORT_PAM4_SPEEDS_100G
+#define BNXT_LINK_PAM4_SPEED_MSK_200GB PORT_PHY_QCFG_RESP_SUPPORT_PAM4_SPEEDS_200G
+	u16			auto_link_speeds2;
+#define BNXT_LINK_SPEEDS2_MSK_1GB PORT_PHY_QCFG_RESP_SUPPORT_SPEEDS2_1GB
+#define BNXT_LINK_SPEEDS2_MSK_10GB PORT_PHY_QCFG_RESP_SUPPORT_SPEEDS2_10GB
+#define BNXT_LINK_SPEEDS2_MSK_25GB PORT_PHY_QCFG_RESP_SUPPORT_SPEEDS2_25GB
+#define BNXT_LINK_SPEEDS2_MSK_40GB PORT_PHY_QCFG_RESP_SUPPORT_SPEEDS2_40GB
+#define BNXT_LINK_SPEEDS2_MSK_50GB PORT_PHY_QCFG_RESP_SUPPORT_SPEEDS2_50GB
+#define BNXT_LINK_SPEEDS2_MSK_100GB PORT_PHY_QCFG_RESP_SUPPORT_SPEEDS2_100GB
+#define BNXT_LINK_SPEEDS2_MSK_50GB_PAM4	\
+	PORT_PHY_QCFG_RESP_SUPPORT_SPEEDS2_50GB_PAM4_56
+#define BNXT_LINK_SPEEDS2_MSK_100GB_PAM4	\
+	PORT_PHY_QCFG_RESP_SUPPORT_SPEEDS2_100GB_PAM4_56
+#define BNXT_LINK_SPEEDS2_MSK_200GB_PAM4	\
+	PORT_PHY_QCFG_RESP_SUPPORT_SPEEDS2_200GB_PAM4_56
+#define BNXT_LINK_SPEEDS2_MSK_400GB_PAM4	\
+	PORT_PHY_QCFG_RESP_SUPPORT_SPEEDS2_400GB_PAM4_56
+#define BNXT_LINK_SPEEDS2_MSK_100GB_PAM4_112	\
+	PORT_PHY_QCFG_RESP_SUPPORT_SPEEDS2_100GB_PAM4_112
+#define BNXT_LINK_SPEEDS2_MSK_200GB_PAM4_112	\
+	PORT_PHY_QCFG_RESP_SUPPORT_SPEEDS2_200GB_PAM4_112
+#define BNXT_LINK_SPEEDS2_MSK_400GB_PAM4_112	\
+	PORT_PHY_QCFG_RESP_SUPPORT_SPEEDS2_400GB_PAM4_112
+
 	u16			support_auto_speeds;
+	u16			support_pam4_auto_speeds;
+	u16			support_auto_speeds2;
+
 	u16			lp_auto_link_speeds;
+	u16			lp_auto_pam4_link_speeds;
 	u16			force_link_speed;
+	u16			force_pam4_link_speed;
+	u16			force_link_speed2;
+#define BNXT_LINK_SPEED_50GB_PAM4	\
+	PORT_PHY_CFG_REQ_FORCE_LINK_SPEEDS2_50GB_PAM4_56
+#define BNXT_LINK_SPEED_100GB_PAM4	\
+	PORT_PHY_CFG_REQ_FORCE_LINK_SPEEDS2_100GB_PAM4_56
+#define BNXT_LINK_SPEED_200GB_PAM4	\
+	PORT_PHY_CFG_REQ_FORCE_LINK_SPEEDS2_200GB_PAM4_56
+#define BNXT_LINK_SPEED_400GB_PAM4	\
+	PORT_PHY_CFG_REQ_FORCE_LINK_SPEEDS2_400GB_PAM4_56
+#define BNXT_LINK_SPEED_100GB_PAM4_112	\
+	PORT_PHY_CFG_REQ_FORCE_LINK_SPEEDS2_100GB_PAM4_112
+#define BNXT_LINK_SPEED_200GB_PAM4_112	\
+	PORT_PHY_CFG_REQ_FORCE_LINK_SPEEDS2_200GB_PAM4_112
+#define BNXT_LINK_SPEED_400GB_PAM4_112	\
+	PORT_PHY_CFG_REQ_FORCE_LINK_SPEEDS2_400GB_PAM4_112
+
 	u32			preemphasis;
 	u8			module_status;
+	u8			active_fec_sig_mode;
 	u16			fec_cfg;
+#define BNXT_FEC_NONE		PORT_PHY_QCFG_RESP_FEC_CFG_FEC_NONE_SUPPORTED
+#define BNXT_FEC_AUTONEG_CAP	PORT_PHY_QCFG_RESP_FEC_CFG_FEC_AUTONEG_SUPPORTED
 #define BNXT_FEC_AUTONEG	PORT_PHY_QCFG_RESP_FEC_CFG_FEC_AUTONEG_ENABLED
+#define BNXT_FEC_ENC_BASE_R_CAP	\
+	PORT_PHY_QCFG_RESP_FEC_CFG_FEC_CLAUSE74_SUPPORTED
 #define BNXT_FEC_ENC_BASE_R	PORT_PHY_QCFG_RESP_FEC_CFG_FEC_CLAUSE74_ENABLED
-#define BNXT_FEC_ENC_RS		PORT_PHY_QCFG_RESP_FEC_CFG_FEC_CLAUSE91_ENABLED
+#define BNXT_FEC_ENC_RS_CAP	\
+	PORT_PHY_QCFG_RESP_FEC_CFG_FEC_CLAUSE91_SUPPORTED
+#define BNXT_FEC_ENC_LLRS_CAP	\
+	(PORT_PHY_QCFG_RESP_FEC_CFG_FEC_RS272_1XN_SUPPORTED |	\
+	 PORT_PHY_QCFG_RESP_FEC_CFG_FEC_RS272_IEEE_SUPPORTED)
+#define BNXT_FEC_ENC_RS		\
+	(PORT_PHY_QCFG_RESP_FEC_CFG_FEC_CLAUSE91_ENABLED |	\
+	 PORT_PHY_QCFG_RESP_FEC_CFG_FEC_RS544_1XN_ENABLED |	\
+	 PORT_PHY_QCFG_RESP_FEC_CFG_FEC_RS544_IEEE_ENABLED)
+#define BNXT_FEC_ENC_LLRS	\
+	(PORT_PHY_QCFG_RESP_FEC_CFG_FEC_RS272_1XN_ENABLED |	\
+	 PORT_PHY_QCFG_RESP_FEC_CFG_FEC_RS272_IEEE_ENABLED)
 
 	/* copy of requested setting from ethtool cmd */
 	u8			autoneg;
 #define BNXT_AUTONEG_SPEED		1
 #define BNXT_AUTONEG_FLOW_CTRL		2
+	u8			req_signal_mode;
+#define BNXT_SIG_MODE_NRZ	PORT_PHY_QCFG_RESP_SIGNAL_MODE_NRZ
+#define BNXT_SIG_MODE_PAM4	PORT_PHY_QCFG_RESP_SIGNAL_MODE_PAM4
+#define BNXT_SIG_MODE_PAM4_112	PORT_PHY_QCFG_RESP_SIGNAL_MODE_PAM4_112
+#define BNXT_SIG_MODE_MAX	(PORT_PHY_QCFG_RESP_SIGNAL_MODE_LAST + 1)
 	u8			req_duplex;
 	u8			req_flow_ctrl;
 	u16			req_link_speed;
 	u16			advertising;	/* user adv setting */
+	u16			advertising_pam4;
 	bool			force_link_chng;
 
+	bool			phy_retry;
+	unsigned long		phy_retry_expires;
+
 	/* a copy of phy_qcfg output used to report link
 	 * info to VF
 	 */
 	struct hwrm_port_phy_qcfg_output phy_qcfg_resp;
 };
 
-#define BNXT_MAX_QUEUE	8
+#define BNXT_FEC_RS544_ON					\
+	 (PORT_PHY_CFG_REQ_FLAGS_FEC_RS544_1XN_ENABLE |		\
+	  PORT_PHY_CFG_REQ_FLAGS_FEC_RS544_IEEE_ENABLE)
+
+#define BNXT_FEC_RS544_OFF					\
+	 (PORT_PHY_CFG_REQ_FLAGS_FEC_RS544_1XN_DISABLE |	\
+	  PORT_PHY_CFG_REQ_FLAGS_FEC_RS544_IEEE_DISABLE)
+
+#define BNXT_FEC_RS272_ON					\
+	 (PORT_PHY_CFG_REQ_FLAGS_FEC_RS272_1XN_ENABLE |		\
+	  PORT_PHY_CFG_REQ_FLAGS_FEC_RS272_IEEE_ENABLE)
+
+#define BNXT_FEC_RS272_OFF					\
+	 (PORT_PHY_CFG_REQ_FLAGS_FEC_RS272_1XN_DISABLE |	\
+	  PORT_PHY_CFG_REQ_FLAGS_FEC_RS272_IEEE_DISABLE)
+
+#define BNXT_PAM4_SUPPORTED(link_info)				\
+	((link_info)->support_pam4_speeds)
+
+#define BNXT_FEC_RS_ON(link_info)				\
+	(PORT_PHY_CFG_REQ_FLAGS_FEC_CLAUSE91_ENABLE |		\
+	 PORT_PHY_CFG_REQ_FLAGS_FEC_CLAUSE74_DISABLE |		\
+	 (BNXT_PAM4_SUPPORTED(link_info) ?			\
+	  (BNXT_FEC_RS544_ON | BNXT_FEC_RS272_OFF) : 0))
+
+#define BNXT_FEC_LLRS_ON					\
+	(PORT_PHY_CFG_REQ_FLAGS_FEC_CLAUSE91_ENABLE |		\
+	 PORT_PHY_CFG_REQ_FLAGS_FEC_CLAUSE74_DISABLE |		\
+	 BNXT_FEC_RS272_ON | BNXT_FEC_RS544_OFF)
+
+#define BNXT_FEC_RS_OFF(link_info)				\
+	(PORT_PHY_CFG_REQ_FLAGS_FEC_CLAUSE91_DISABLE |		\
+	 (BNXT_PAM4_SUPPORTED(link_info) ?			\
+	  (BNXT_FEC_RS544_OFF | BNXT_FEC_RS272_OFF) : 0))
+
+#define BNXT_FEC_BASE_R_ON(link_info)				\
+	(PORT_PHY_CFG_REQ_FLAGS_FEC_CLAUSE74_ENABLE |		\
+	 BNXT_FEC_RS_OFF(link_info))
+
+#define BNXT_FEC_ALL_OFF(link_info)				\
+	(PORT_PHY_CFG_REQ_FLAGS_FEC_CLAUSE74_DISABLE |		\
+	 BNXT_FEC_RS_OFF(link_info))
 
 struct bnxt_queue_info {
 	u8	queue_id;
@@ -920,9 +1771,389 @@ struct bnxt_test_info {
 	char string[BNXT_MAX_TEST][ETH_GSTRING_LEN];
 };
 
-#define BNXT_GRCPF_REG_WINDOW_BASE_OUT	0x400
-#define BNXT_CAG_REG_LEGACY_INT_STATUS	0x4014
-#define BNXT_CAG_REG_BASE		0x300000
+#define CHIMP_REG_VIEW_ADDR				\
+	((bp->flags & BNXT_FLAG_CHIP_P5_PLUS) ? 0x80000000 : 0xb1000000)
+
+#define BNXT_GRCPF_REG_CHIMP_COMM		0x0
+#define BNXT_GRCPF_REG_CHIMP_COMM_TRIGGER	0x100
+#define BNXT_GRCPF_REG_WINDOW_BASE_OUT		0x400
+
+#define BNXT_GRC_REG_STATUS_P5			0x520
+
+#define BNXT_GRCPF_REG_KONG_COMM		0xA00
+#define BNXT_GRCPF_REG_KONG_COMM_TRIGGER	0xB00
+
+#define BNXT_GRC_REG_CHIP_NUM			0x48
+#define BNXT_GRC_REG_BASE			0x260000
+
+#define BNXT_TS_REG_TIMESYNC_TS0_LOWER		0x640180c
+#define BNXT_TS_REG_TIMESYNC_TS0_UPPER		0x6401810
+
+#define BNXT_GRC_BASE_MASK			0xfffff000
+#define BNXT_GRC_OFFSET_MASK			0x00000ffc
+
+struct bnxt_tc_flow_stats {
+	u64		packets;
+	u64		bytes;
+};
+
+#ifdef CONFIG_BNXT_FLOWER_OFFLOAD
+struct bnxt_flower_indr_block_cb_priv {
+	struct net_device *tunnel_netdev;
+	struct bnxt *bp;
+	struct list_head list;
+};
+#endif
+
+struct bnxt_tc_info {
+	bool				enabled;
+
+	/* hash table to store TC offloaded flows */
+	struct rhashtable		flow_table;
+	struct rhashtable_params	flow_ht_params;
+
+	/* hash table to store L2 keys of TC flows */
+	struct rhashtable		l2_table;
+	struct rhashtable_params	l2_ht_params;
+	/* hash table to store L2 keys for TC tunnel decap */
+	struct rhashtable		decap_l2_table;
+	struct rhashtable_params	decap_l2_ht_params;
+	/* hash table to store tunnel decap entries */
+	struct rhashtable		decap_table;
+	struct rhashtable_params	decap_ht_params;
+	/* hash table to store tunnel encap entries */
+	struct rhashtable		encap_table;
+	struct rhashtable_params	encap_ht_params;
+
+	/* lock to atomically add/del an l2 node when a flow is
+	 * added or deleted.
+	 */
+	struct mutex			lock;
+
+	/* Fields used for batching stats query */
+	struct rhashtable_iter		iter;
+#define BNXT_FLOW_STATS_BATCH_MAX	10
+	struct bnxt_tc_stats_batch {
+		void			  *flow_node;
+		struct bnxt_tc_flow_stats hw_stats;
+	} stats_batch[BNXT_FLOW_STATS_BATCH_MAX];
+
+	/* Stat counter mask (width) */
+	u64				bytes_mask;
+	u64				packets_mask;
+};
+
+struct bnxt_vf_rep_stats {
+	u64			packets;
+	u64			bytes;
+	u64			dropped;
+};
+
+struct bnxt_vf_rep {
+	struct bnxt			*bp;
+	struct net_device		*dev;
+	struct metadata_dst		*dst;
+	u16				vf_idx;
+	u16				tx_cfa_action;
+	u16				rx_cfa_code;
+
+	struct bnxt_vf_rep_stats	rx_stats;
+	struct bnxt_vf_rep_stats	tx_stats;
+};
+
+#define PTU_PTE_VALID             0x1UL
+#define PTU_PTE_LAST              0x2UL
+#define PTU_PTE_NEXT_TO_LAST      0x4UL
+
+#define MAX_CTX_PAGES	(BNXT_PAGE_SIZE / 8)
+#define MAX_CTX_TOTAL_PAGES	(MAX_CTX_PAGES * MAX_CTX_PAGES)
+#define MAX_CTX_BYTES		((size_t)MAX_CTX_TOTAL_PAGES * BNXT_PAGE_SIZE)
+#define MAX_CTX_BYTES_MASK	(MAX_CTX_BYTES - 1)
+
+struct bnxt_ctx_pg_info {
+	u32		entries;
+	u32		nr_pages;
+	void		*ctx_pg_arr[MAX_CTX_PAGES];
+	dma_addr_t	ctx_dma_arr[MAX_CTX_PAGES];
+	struct bnxt_ring_mem_info ring_mem;
+	struct bnxt_ctx_pg_info **ctx_pg_tbl;
+};
+
+#define BNXT_MAX_TQM_SP_RINGS		1
+#define BNXT_MAX_TQM_FP_RINGS		8
+#define BNXT_MAX_TQM_RINGS		\
+	(BNXT_MAX_TQM_SP_RINGS + BNXT_MAX_TQM_FP_RINGS)
+
+#define BNXT_BACKING_STORE_CFG_LEGACY_LEN	256
+
+#define BNXT_SET_CTX_PAGE_ATTR(attr)					\
+do {									\
+	if (BNXT_PAGE_SIZE == 0x2000)					\
+		attr = FUNC_BACKING_STORE_CFG_REQ_SRQ_PG_SIZE_PG_8K;	\
+	else if (BNXT_PAGE_SIZE == 0x10000)				\
+		attr = FUNC_BACKING_STORE_CFG_REQ_QPC_PG_SIZE_PG_64K;	\
+	else								\
+		attr = FUNC_BACKING_STORE_CFG_REQ_QPC_PG_SIZE_PG_4K;	\
+} while (0)
+
+struct bnxt_ctx_mem_type {
+	u16	type;
+	u16	entry_size;
+	u32	flags;
+#define BNXT_CTX_MEM_TYPE_VALID FUNC_BACKING_STORE_QCAPS_V2_RESP_FLAGS_TYPE_VALID
+#define BNXT_CTX_MEM_FW_TRACE		\
+	FUNC_BACKING_STORE_QCAPS_V2_RESP_FLAGS_FW_DBG_TRACE
+#define BNXT_CTX_MEM_FW_BIN_TRACE	\
+	FUNC_BACKING_STORE_QCAPS_V2_RESP_FLAGS_FW_BIN_DBG_TRACE
+#define BNXT_CTX_MEM_PERSIST		\
+	FUNC_BACKING_STORE_QCAPS_V2_RESP_FLAGS_NEXT_BS_OFFSET
+
+	u32	instance_bmap;
+	u8	init_value;
+	u8	entry_multiple;
+	u16	init_offset;
+#define	BNXT_CTX_INIT_INVALID_OFFSET	0xffff
+	u32	max_entries;
+	u32	min_entries;
+	u8	last:1;
+	u8	mem_valid:1;
+	u8	split_entry_cnt;
+#define BNXT_MAX_SPLIT_ENTRY	4
+	union {
+		struct {
+			u32	qp_l2_entries;
+			u32	qp_qp1_entries;
+			u32	qp_fast_qpmd_entries;
+		};
+		u32	srq_l2_entries;
+		u32	cq_l2_entries;
+		u32	vnic_entries;
+		struct {
+			u32	mrav_av_entries;
+			u32	mrav_num_entries_units;
+		};
+		u32	split[BNXT_MAX_SPLIT_ENTRY];
+	};
+	struct bnxt_ctx_pg_info	*pg_info;
+};
+
+#define BNXT_CTX_MRAV_AV_SPLIT_ENTRY	0
+
+#define BNXT_CTX_QP	FUNC_BACKING_STORE_QCAPS_V2_REQ_TYPE_QP
+#define BNXT_CTX_SRQ	FUNC_BACKING_STORE_QCAPS_V2_REQ_TYPE_SRQ
+#define BNXT_CTX_CQ	FUNC_BACKING_STORE_QCAPS_V2_REQ_TYPE_CQ
+#define BNXT_CTX_VNIC	FUNC_BACKING_STORE_QCAPS_V2_REQ_TYPE_VNIC
+#define BNXT_CTX_STAT	FUNC_BACKING_STORE_QCAPS_V2_REQ_TYPE_STAT
+#define BNXT_CTX_STQM	FUNC_BACKING_STORE_QCAPS_V2_REQ_TYPE_SP_TQM_RING
+#define BNXT_CTX_FTQM	FUNC_BACKING_STORE_QCAPS_V2_REQ_TYPE_FP_TQM_RING
+#define BNXT_CTX_MRAV	FUNC_BACKING_STORE_QCAPS_V2_REQ_TYPE_MRAV
+#define BNXT_CTX_TIM	FUNC_BACKING_STORE_QCAPS_V2_REQ_TYPE_TIM
+#define BNXT_CTX_TCK	FUNC_BACKING_STORE_QCAPS_V2_REQ_TYPE_TX_CK
+#define BNXT_CTX_RCK	FUNC_BACKING_STORE_QCAPS_V2_REQ_TYPE_RX_CK
+#define BNXT_CTX_MTQM	FUNC_BACKING_STORE_QCAPS_V2_REQ_TYPE_MP_TQM_RING
+#define BNXT_CTX_SQDBS	FUNC_BACKING_STORE_QCAPS_V2_REQ_TYPE_SQ_DB_SHADOW
+#define BNXT_CTX_RQDBS	FUNC_BACKING_STORE_QCAPS_V2_REQ_TYPE_RQ_DB_SHADOW
+#define BNXT_CTX_SRQDBS	FUNC_BACKING_STORE_QCAPS_V2_REQ_TYPE_SRQ_DB_SHADOW
+#define BNXT_CTX_CQDBS	FUNC_BACKING_STORE_QCAPS_V2_REQ_TYPE_CQ_DB_SHADOW
+#define BNXT_CTX_TBLSC	FUNC_BACKING_STORE_QCAPS_V2_REQ_TYPE_TBL_SCOPE
+#define BNXT_CTX_XPAR	FUNC_BACKING_STORE_QCAPS_V2_REQ_TYPE_XID_PARTITION
+#define BNXT_CTX_SRT	FUNC_BACKING_STORE_QCAPS_V2_REQ_TYPE_SRT_TRACE
+#define BNXT_CTX_SRT2	FUNC_BACKING_STORE_QCAPS_V2_REQ_TYPE_SRT2_TRACE
+#define BNXT_CTX_CRT	FUNC_BACKING_STORE_QCAPS_V2_REQ_TYPE_CRT_TRACE
+#define BNXT_CTX_CRT2	FUNC_BACKING_STORE_QCAPS_V2_REQ_TYPE_CRT2_TRACE
+#define BNXT_CTX_RIGP0	FUNC_BACKING_STORE_QCAPS_V2_REQ_TYPE_RIGP0_TRACE
+#define BNXT_CTX_L2HWRM	FUNC_BACKING_STORE_QCAPS_V2_REQ_TYPE_L2_HWRM_TRACE
+#define BNXT_CTX_REHWRM	FUNC_BACKING_STORE_QCAPS_V2_REQ_TYPE_ROCE_HWRM_TRACE
+#define BNXT_CTX_CA0	FUNC_BACKING_STORE_QCAPS_V2_REQ_TYPE_CA0_TRACE
+#define BNXT_CTX_CA1	FUNC_BACKING_STORE_QCAPS_V2_REQ_TYPE_CA1_TRACE
+#define BNXT_CTX_CA2	FUNC_BACKING_STORE_QCAPS_V2_REQ_TYPE_CA2_TRACE
+#define BNXT_CTX_RIGP1	FUNC_BACKING_STORE_QCAPS_V2_REQ_TYPE_RIGP1_TRACE
+#define BNXT_CTX_KONG	FUNC_BACKING_STORE_QCAPS_V2_REQ_TYPE_AFM_KONG_HWRM_TRACE
+#define BNXT_CTX_QPC	FUNC_BACKING_STORE_QCAPS_V2_REQ_TYPE_ERR_QPC_TRACE
+
+#define BNXT_CTX_MAX	(BNXT_CTX_TIM + 1)
+#define BNXT_CTX_L2_MAX	(BNXT_CTX_FTQM + 1)
+#define BNXT_CTX_V2_MAX (BNXT_CTX_QPC + 1)
+#define BNXT_CTX_INV	((u16)-1)
+
+struct bnxt_ctx_mem_info {
+	u8	tqm_fp_rings_count;
+
+	u32	flags;
+	#define BNXT_CTX_FLAG_INITED	0x01
+	struct bnxt_ctx_mem_type	ctx_arr[BNXT_CTX_V2_MAX];
+};
+
+enum bnxt_health_severity {
+	SEVERITY_NORMAL = 0,
+	SEVERITY_WARNING,
+	SEVERITY_RECOVERABLE,
+	SEVERITY_FATAL,
+};
+
+enum bnxt_health_remedy {
+	REMEDY_DEVLINK_RECOVER,
+	REMEDY_POWER_CYCLE_DEVICE,
+	REMEDY_POWER_CYCLE_HOST,
+	REMEDY_FW_UPDATE,
+	REMEDY_HW_REPLACE,
+};
+
+struct bnxt_fw_health {
+	u32 flags;
+	u32 polling_dsecs;
+	u32 master_func_wait_dsecs;
+	u32 normal_func_wait_dsecs;
+	u32 post_reset_wait_dsecs;
+	u32 post_reset_max_wait_dsecs;
+	u32 regs[4];
+	u32 mapped_regs[4];
+#define BNXT_FW_HEALTH_REG		0
+#define BNXT_FW_HEARTBEAT_REG		1
+#define BNXT_FW_RESET_CNT_REG		2
+#define BNXT_FW_RESET_INPROG_REG	3
+	u32 fw_reset_inprog_reg_mask;
+	u32 last_fw_heartbeat;
+	u32 last_fw_reset_cnt;
+	u8 enabled:1;
+	u8 primary:1;
+	u8 status_reliable:1;
+	u8 resets_reliable:1;
+	u8 tmr_multiplier;
+	u8 tmr_counter;
+	u8 fw_reset_seq_cnt;
+	u32 fw_reset_seq_regs[16];
+	u32 fw_reset_seq_vals[16];
+	u32 fw_reset_seq_delay_msec[16];
+	u32 echo_req_data1;
+	u32 echo_req_data2;
+	struct devlink_health_reporter	*fw_reporter;
+	/* Protects severity and remedy */
+	struct mutex lock;
+	enum bnxt_health_severity severity;
+	enum bnxt_health_remedy remedy;
+	u32 arrests;
+	u32 discoveries;
+	u32 survivals;
+	u32 fatalities;
+	u32 diagnoses;
+};
+
+#define BNXT_FW_HEALTH_REG_TYPE_MASK	3
+#define BNXT_FW_HEALTH_REG_TYPE_CFG	0
+#define BNXT_FW_HEALTH_REG_TYPE_GRC	1
+#define BNXT_FW_HEALTH_REG_TYPE_BAR0	2
+#define BNXT_FW_HEALTH_REG_TYPE_BAR1	3
+
+#define BNXT_FW_HEALTH_REG_TYPE(reg)	((reg) & BNXT_FW_HEALTH_REG_TYPE_MASK)
+#define BNXT_FW_HEALTH_REG_OFF(reg)	((reg) & ~BNXT_FW_HEALTH_REG_TYPE_MASK)
+
+#define BNXT_FW_HEALTH_WIN_BASE		0x3000
+#define BNXT_FW_HEALTH_WIN_MAP_OFF	8
+
+#define BNXT_FW_HEALTH_WIN_OFF(reg)	(BNXT_FW_HEALTH_WIN_BASE +	\
+					 ((reg) & BNXT_GRC_OFFSET_MASK))
+
+#define BNXT_FW_STATUS_HEALTH_MSK	0xffff
+#define BNXT_FW_STATUS_HEALTHY		0x8000
+#define BNXT_FW_STATUS_SHUTDOWN		0x100000
+#define BNXT_FW_STATUS_RECOVERING	0x400000
+
+#define BNXT_FW_IS_HEALTHY(sts)		(((sts) & BNXT_FW_STATUS_HEALTH_MSK) ==\
+					 BNXT_FW_STATUS_HEALTHY)
+
+#define BNXT_FW_IS_BOOTING(sts)		(((sts) & BNXT_FW_STATUS_HEALTH_MSK) < \
+					 BNXT_FW_STATUS_HEALTHY)
+
+#define BNXT_FW_IS_ERR(sts)		(((sts) & BNXT_FW_STATUS_HEALTH_MSK) > \
+					 BNXT_FW_STATUS_HEALTHY)
+
+#define BNXT_FW_IS_RECOVERING(sts)	(BNXT_FW_IS_ERR(sts) &&		       \
+					 ((sts) & BNXT_FW_STATUS_RECOVERING))
+
+#define BNXT_FW_RETRY			5
+#define BNXT_FW_IF_RETRY		10
+#define BNXT_FW_SLOT_RESET_RETRY	4
+
+struct bnxt_aux_priv {
+	struct auxiliary_device aux_dev;
+	struct bnxt_en_dev *edev;
+	int id;
+};
+
+enum board_idx {
+	BCM57301,
+	BCM57302,
+	BCM57304,
+	BCM57417_NPAR,
+	BCM58700,
+	BCM57311,
+	BCM57312,
+	BCM57402,
+	BCM57404,
+	BCM57406,
+	BCM57402_NPAR,
+	BCM57407,
+	BCM57412,
+	BCM57414,
+	BCM57416,
+	BCM57417,
+	BCM57412_NPAR,
+	BCM57314,
+	BCM57417_SFP,
+	BCM57416_SFP,
+	BCM57404_NPAR,
+	BCM57406_NPAR,
+	BCM57407_SFP,
+	BCM57407_NPAR,
+	BCM57414_NPAR,
+	BCM57416_NPAR,
+	BCM57452,
+	BCM57454,
+	BCM5745x_NPAR,
+	BCM57508,
+	BCM57504,
+	BCM57502,
+	BCM57508_NPAR,
+	BCM57504_NPAR,
+	BCM57502_NPAR,
+	BCM57608,
+	BCM57604,
+	BCM57602,
+	BCM57601,
+	BCM58802,
+	BCM58804,
+	BCM58808,
+	NETXTREME_E_VF,
+	NETXTREME_C_VF,
+	NETXTREME_S_VF,
+	NETXTREME_C_VF_HV,
+	NETXTREME_E_VF_HV,
+	NETXTREME_E_P5_VF,
+	NETXTREME_E_P5_VF_HV,
+	NETXTREME_E_P7_VF,
+	NETXTREME_E_P7_VF_HV,
+};
+
+#define BNXT_TRACE_BUF_MAGIC_BYTE ((u8)0xbc)
+#define BNXT_TRACE_MAX 11
+
+struct bnxt_bs_trace_info {
+	u8 *magic_byte;
+	u32 last_offset;
+	u8 wrapped:1;
+	u16 ctx_type;
+	u16 trace_type;
+};
+
+static inline void bnxt_bs_trace_check_wrap(struct bnxt_bs_trace_info *bs_trace,
+					    u32 offset)
+{
+	if (!bs_trace->wrapped &&
+	    *bs_trace->magic_byte != BNXT_TRACE_BUF_MAGIC_BYTE)
+		bs_trace->wrapped = 1;
+	bs_trace->last_offset = offset;
+}
 
 struct bnxt {
 	void __iomem		*bar0;
@@ -952,6 +2183,20 @@ struct bnxt {
 #define CHIP_NUM_57414L		0x16db
 
 #define CHIP_NUM_5745X		0xd730
+#define CHIP_NUM_57452		0xc452
+#define CHIP_NUM_57454		0xc454
+
+#define CHIP_NUM_57508		0x1750
+#define CHIP_NUM_57504		0x1751
+#define CHIP_NUM_57502		0x1752
+
+#define CHIP_NUM_57608		0x1760
+
+#define CHIP_NUM_58802		0xd802
+#define CHIP_NUM_58804		0xd804
+#define CHIP_NUM_58808		0xd808
+
+	u8			chip_rev;
 
 #define BNXT_CHIP_NUM_5730X(chip_num)		\
 	((chip_num) >= CHIP_NUM_57301 &&	\
@@ -976,7 +2221,10 @@ struct bnxt {
 	 ((chip_num) == CHIP_NUM_58700)
 
 #define BNXT_CHIP_NUM_5745X(chip_num)		\
-	 ((chip_num) == CHIP_NUM_5745X)
+	((chip_num) == CHIP_NUM_5745X ||	\
+	 (chip_num) == CHIP_NUM_57452 ||	\
+	 (chip_num) == CHIP_NUM_57454)
+
 
 #define BNXT_CHIP_NUM_57X0X(chip_num)		\
 	(BNXT_CHIP_NUM_5730X(chip_num) || BNXT_CHIP_NUM_5740X(chip_num))
@@ -984,13 +2232,22 @@ struct bnxt {
 #define BNXT_CHIP_NUM_57X1X(chip_num)		\
 	(BNXT_CHIP_NUM_5731X(chip_num) || BNXT_CHIP_NUM_5741X(chip_num))
 
+#define BNXT_CHIP_NUM_588XX(chip_num)		\
+	((chip_num) == CHIP_NUM_58802 ||	\
+	 (chip_num) == CHIP_NUM_58804 ||        \
+	 (chip_num) == CHIP_NUM_58808)
+
+#define BNXT_VPD_FLD_LEN	32
+	char			board_partno[BNXT_VPD_FLD_LEN];
+	char			board_serialno[BNXT_VPD_FLD_LEN];
+
 	struct net_device	*dev;
 	struct pci_dev		*pdev;
 
 	atomic_t		intr_sem;
 
 	u32			flags;
-	#define BNXT_FLAG_DCB_ENABLED	0x1
+	#define BNXT_FLAG_CHIP_P5_PLUS	0x1
 	#define BNXT_FLAG_VF		0x2
 	#define BNXT_FLAG_LRO		0x4
 #ifdef CONFIG_INET
@@ -1002,16 +2259,9 @@ struct bnxt {
 	#define BNXT_FLAG_TPA		(BNXT_FLAG_LRO | BNXT_FLAG_GRO)
 	#define BNXT_FLAG_JUMBO		0x10
 	#define BNXT_FLAG_STRIP_VLAN	0x20
-	#define BNXT_FLAG_AGG_RINGS	(BNXT_FLAG_JUMBO | BNXT_FLAG_GRO | \
-					 BNXT_FLAG_LRO)
-	#define BNXT_FLAG_USING_MSIX	0x40
-	#define BNXT_FLAG_MSIX_CAP	0x80
 	#define BNXT_FLAG_RFS		0x100
 	#define BNXT_FLAG_SHARED_RINGS	0x200
 	#define BNXT_FLAG_PORT_STATS	0x400
-	#define BNXT_FLAG_UDP_RSS_CAP	0x800
-	#define BNXT_FLAG_EEE_CAP	0x1000
-	#define BNXT_FLAG_NEW_RSS_CAP	0x2000
 	#define BNXT_FLAG_WOL_CAP	0x4000
 	#define BNXT_FLAG_ROCEV1_CAP	0x8000
 	#define BNXT_FLAG_ROCEV2_CAP	0x10000
@@ -1019,11 +2269,19 @@ struct bnxt {
 					 BNXT_FLAG_ROCEV2_CAP)
 	#define BNXT_FLAG_NO_AGG_RINGS	0x20000
 	#define BNXT_FLAG_RX_PAGE_MODE	0x40000
-	#define BNXT_FLAG_FW_LLDP_AGENT	0x80000
+	#define BNXT_FLAG_CHIP_P7	0x80000
 	#define BNXT_FLAG_MULTI_HOST	0x100000
-	#define BNXT_FLAG_SHORT_CMD	0x200000
+	#define BNXT_FLAG_DSN_VALID	0x200000
 	#define BNXT_FLAG_DOUBLE_DB	0x400000
+	#define BNXT_FLAG_UDP_GSO_CAP	0x800000
 	#define BNXT_FLAG_CHIP_NITRO_A0	0x1000000
+	#define BNXT_FLAG_DIM		0x2000000
+	#define BNXT_FLAG_ROCE_MIRROR_CAP	0x4000000
+	#define BNXT_FLAG_TX_COAL_CMPL	0x8000000
+	#define BNXT_FLAG_PORT_STATS_EXT	0x10000000
+	#define BNXT_FLAG_HDS		0x20000000
+	#define BNXT_FLAG_AGG_RINGS	(BNXT_FLAG_JUMBO | BNXT_FLAG_GRO | \
+					 BNXT_FLAG_LRO | BNXT_FLAG_HDS)
 
 	#define BNXT_FLAG_ALL_CONFIG_FEATS (BNXT_FLAG_TPA |		\
 					    BNXT_FLAG_RFS |		\
@@ -1031,21 +2289,59 @@ struct bnxt {
 
 #define BNXT_PF(bp)		(!((bp)->flags & BNXT_FLAG_VF))
 #define BNXT_VF(bp)		((bp)->flags & BNXT_FLAG_VF)
+#ifdef CONFIG_BNXT_SRIOV
+#define	BNXT_VF_IS_TRUSTED(bp)	((bp)->vf.flags & BNXT_VF_TRUST)
+#else
+#define	BNXT_VF_IS_TRUSTED(bp)	0
+#endif
 #define BNXT_NPAR(bp)		((bp)->port_partition_type)
 #define BNXT_MH(bp)		((bp)->flags & BNXT_FLAG_MULTI_HOST)
 #define BNXT_SINGLE_PF(bp)	(BNXT_PF(bp) && !BNXT_NPAR(bp) && !BNXT_MH(bp))
+#define BNXT_SH_PORT_CFG_OK(bp)	(BNXT_PF(bp) &&				\
+				 ((bp)->phy_flags & BNXT_PHY_FL_SHARED_PORT_CFG))
+#define BNXT_PHY_CFG_ABLE(bp)	((BNXT_SINGLE_PF(bp) ||			\
+				  BNXT_SH_PORT_CFG_OK(bp)) &&		\
+				 (bp)->link_info.phy_state == BNXT_PHY_STATE_ENABLED)
 #define BNXT_CHIP_TYPE_NITRO_A0(bp) ((bp)->flags & BNXT_FLAG_CHIP_NITRO_A0)
 #define BNXT_RX_PAGE_MODE(bp)	((bp)->flags & BNXT_FLAG_RX_PAGE_MODE)
+#define BNXT_SUPPORTS_TPA(bp)	(!BNXT_CHIP_TYPE_NITRO_A0(bp) &&	\
+				 (!((bp)->flags & BNXT_FLAG_CHIP_P5_PLUS) ||\
+				  (bp)->max_tpa_v2) && !is_kdump_kernel())
+#define BNXT_RX_JUMBO_MODE(bp)	((bp)->flags & BNXT_FLAG_JUMBO)
 
-/* Chip class phase 4 and later */
-#define BNXT_CHIP_P4_PLUS(bp)			\
+#define BNXT_CHIP_P7(bp)			\
+	((bp)->chip_num == CHIP_NUM_57608)
+
+#define BNXT_CHIP_P5(bp)			\
+	((bp)->chip_num == CHIP_NUM_57508 ||	\
+	 (bp)->chip_num == CHIP_NUM_57504 ||	\
+	 (bp)->chip_num == CHIP_NUM_57502)
+
+/* Chip class phase 5 */
+#define BNXT_CHIP_P5_PLUS(bp)			\
+	(BNXT_CHIP_P5(bp) || BNXT_CHIP_P7(bp))
+
+/* Chip class phase 4.x */
+#define BNXT_CHIP_P4(bp)			\
 	(BNXT_CHIP_NUM_57X1X((bp)->chip_num) ||	\
 	 BNXT_CHIP_NUM_5745X((bp)->chip_num) ||	\
+	 BNXT_CHIP_NUM_588XX((bp)->chip_num) ||	\
 	 (BNXT_CHIP_NUM_58700((bp)->chip_num) &&	\
 	  !BNXT_CHIP_TYPE_NITRO_A0(bp)))
 
+/* Chip class phase 3.x */
+#define BNXT_CHIP_P3(bp)			\
+	(BNXT_CHIP_NUM_57X0X((bp)->chip_num) ||	\
+	 BNXT_CHIP_TYPE_NITRO_A0(bp))
+
+#define BNXT_CHIP_P4_PLUS(bp)			\
+	(BNXT_CHIP_P4(bp) || BNXT_CHIP_P5_PLUS(bp))
+
+#define BNXT_CHIP_P5_AND_MINUS(bp)		\
+	(BNXT_CHIP_P3(bp) || BNXT_CHIP_P4(bp) || BNXT_CHIP_P5(bp))
+
+	struct bnxt_aux_priv	*aux_priv;
 	struct bnxt_en_dev	*edev;
-	struct bnxt_en_dev *	(*ulp_probe)(struct net_device *);
 
 	struct bnxt_napi	**bnapi;
 
@@ -1061,6 +2357,8 @@ struct bnxt {
 					       u16, void *, u8 *, dma_addr_t,
 					       unsigned int);
 
+	u16			max_tpa_v2;
+	u16			max_tpa;
 	u32			rx_buf_size;
 	u32			rx_buf_use_size;	/* useable size */
 	u16			rx_offset;
@@ -1068,7 +2366,7 @@ struct bnxt {
 	enum dma_data_direction	rx_dir;
 	u32			rx_ring_size;
 	u32			rx_agg_ring_size;
-	u32			rx_copy_thresh;
+	u32			rx_copybreak;
 	u32			rx_ring_mask;
 	u32			rx_agg_ring_mask;
 	int			rx_nr_pages;
@@ -1093,17 +2391,44 @@ struct bnxt {
 	int			cp_nr_pages;
 	int			cp_nr_rings;
 
-	int			num_stat_ctxs;
-
 	/* grp_info indexed by completion ring index */
 	struct bnxt_ring_grp_info	*grp_info;
 	struct bnxt_vnic_info	*vnic_info;
+	u32			num_rss_ctx;
 	int			nr_vnics;
+	u32			*rss_indir_tbl;
+	u16			rss_indir_tbl_entries;
 	u32			rss_hash_cfg;
-
+	u32			rss_hash_delta;
+	u32			rss_cap;
+#define BNXT_RSS_CAP_RSS_HASH_TYPE_DELTA	BIT(0)
+#define BNXT_RSS_CAP_UDP_RSS_CAP		BIT(1)
+#define BNXT_RSS_CAP_NEW_RSS_CAP		BIT(2)
+#define BNXT_RSS_CAP_RSS_TCAM			BIT(3)
+#define BNXT_RSS_CAP_AH_V4_RSS_CAP		BIT(4)
+#define BNXT_RSS_CAP_AH_V6_RSS_CAP		BIT(5)
+#define BNXT_RSS_CAP_ESP_V4_RSS_CAP		BIT(6)
+#define BNXT_RSS_CAP_ESP_V6_RSS_CAP		BIT(7)
+#define BNXT_RSS_CAP_MULTI_RSS_CTX		BIT(8)
+#define BNXT_RSS_CAP_IPV6_FLOW_LABEL_RSS_CAP	BIT(9)
+
+	u8			rss_hash_key[HW_HASH_KEY_SIZE];
+	u8			rss_hash_key_valid:1;
+	u8			rss_hash_key_updated:1;
+
+	u16			max_mtu;
+	u16			tso_max_segs;
 	u8			max_tc;
 	u8			max_lltc;	/* lossless TCs */
 	struct bnxt_queue_info	q_info[BNXT_MAX_QUEUE];
+	u8			tc_to_qidx[BNXT_MAX_QUEUE];
+	u8			q_ids[BNXT_MAX_QUEUE];
+	u8			max_q;
+	u8			num_tc;
+
+	u16			max_pfcwd_tmo_ms;
+
+	u8			tph_mode;
 
 	unsigned int		current_interval;
 #define BNXT_TIMER_INTERVAL	HZ
@@ -1113,9 +2438,28 @@ struct bnxt {
 	unsigned long		state;
 #define BNXT_STATE_OPEN		0
 #define BNXT_STATE_IN_SP_TASK	1
+#define BNXT_STATE_READ_STATS	2
+#define BNXT_STATE_FW_RESET_DET 3
+#define BNXT_STATE_IN_FW_RESET	4
+#define BNXT_STATE_ABORT_ERR	5
+#define BNXT_STATE_FW_FATAL_COND	6
+#define BNXT_STATE_DRV_REGISTERED	7
+#define BNXT_STATE_PCI_CHANNEL_IO_FROZEN	8
+#define BNXT_STATE_NAPI_DISABLED	9
+#define BNXT_STATE_L2_FILTER_RETRY	10
+#define BNXT_STATE_FW_ACTIVATE		11
+#define BNXT_STATE_RECOVER		12
+#define BNXT_STATE_FW_NON_FATAL_COND	13
+#define BNXT_STATE_FW_ACTIVATE_RESET	14
+#define BNXT_STATE_HALF_OPEN		15	/* For offline ethtool tests */
+
+#define BNXT_NO_FW_ACCESS(bp)					\
+	(test_bit(BNXT_STATE_FW_FATAL_COND, &(bp)->state) ||	\
+	 pci_channel_offline((bp)->pdev))
 
 	struct bnxt_irq	*irq_tbl;
 	int			total_irqs;
+	int			ulp_num_msix_want;
 	u8			mac_addr[ETH_ALEN];
 
 #ifdef CONFIG_BNXT_DCB
@@ -1123,53 +2467,130 @@ struct bnxt {
 	struct ieee_ets		*ieee_ets;
 	u8			dcbx_cap;
 	u8			default_pri;
+	u8			max_dscp_value;
 #endif /* CONFIG_BNXT_DCB */
 
 	u32			msg_enable;
 
+	u64			fw_cap;
+	#define BNXT_FW_CAP_SHORT_CMD			BIT_ULL(0)
+	#define BNXT_FW_CAP_LLDP_AGENT			BIT_ULL(1)
+	#define BNXT_FW_CAP_DCBX_AGENT			BIT_ULL(2)
+	#define BNXT_FW_CAP_NEW_RM			BIT_ULL(3)
+	#define BNXT_FW_CAP_IF_CHANGE			BIT_ULL(4)
+	#define BNXT_FW_CAP_ENABLE_RDMA_SRIOV           BIT_ULL(5)
+	#define BNXT_FW_CAP_ROCE_VF_RESC_MGMT_SUPPORTED	BIT_ULL(6)
+	#define BNXT_FW_CAP_KONG_MB_CHNL		BIT_ULL(7)
+	#define BNXT_FW_CAP_ROCE_VF_DYN_ALLOC_SUPPORT	BIT_ULL(8)
+	#define BNXT_FW_CAP_OVS_64BIT_HANDLE		BIT_ULL(10)
+	#define BNXT_FW_CAP_TRUSTED_VF			BIT_ULL(11)
+	#define BNXT_FW_CAP_ERROR_RECOVERY		BIT_ULL(13)
+	#define BNXT_FW_CAP_PKG_VER			BIT_ULL(14)
+	#define BNXT_FW_CAP_CFA_ADV_FLOW		BIT_ULL(15)
+	#define BNXT_FW_CAP_CFA_RFS_RING_TBL_IDX_V2	BIT_ULL(16)
+	#define BNXT_FW_CAP_PCIE_STATS_SUPPORTED	BIT_ULL(17)
+	#define BNXT_FW_CAP_EXT_STATS_SUPPORTED		BIT_ULL(18)
+	#define BNXT_FW_CAP_TX_TS_CMP			BIT_ULL(19)
+	#define BNXT_FW_CAP_ERR_RECOVER_RELOAD		BIT_ULL(20)
+	#define BNXT_FW_CAP_HOT_RESET			BIT_ULL(21)
+	#define BNXT_FW_CAP_PTP_RTC			BIT_ULL(22)
+	#define BNXT_FW_CAP_RX_ALL_PKT_TS		BIT_ULL(23)
+	#define BNXT_FW_CAP_VLAN_RX_STRIP		BIT_ULL(24)
+	#define BNXT_FW_CAP_VLAN_TX_INSERT		BIT_ULL(25)
+	#define BNXT_FW_CAP_EXT_HW_STATS_SUPPORTED	BIT_ULL(26)
+	#define BNXT_FW_CAP_LIVEPATCH			BIT_ULL(27)
+	#define BNXT_FW_CAP_PTP_PPS			BIT_ULL(28)
+	#define BNXT_FW_CAP_HOT_RESET_IF		BIT_ULL(29)
+	#define BNXT_FW_CAP_RING_MONITOR		BIT_ULL(30)
+	#define BNXT_FW_CAP_DBG_QCAPS			BIT_ULL(31)
+	#define BNXT_FW_CAP_PTP				BIT_ULL(32)
+	#define BNXT_FW_CAP_THRESHOLD_TEMP_SUPPORTED	BIT_ULL(33)
+	#define BNXT_FW_CAP_DFLT_VLAN_TPID_PCP		BIT_ULL(34)
+	#define BNXT_FW_CAP_PRE_RESV_VNICS		BIT_ULL(35)
+	#define BNXT_FW_CAP_BACKING_STORE_V2		BIT_ULL(36)
+	#define BNXT_FW_CAP_VNIC_TUNNEL_TPA		BIT_ULL(37)
+	#define BNXT_FW_CAP_CFA_NTUPLE_RX_EXT_IP_PROTO	BIT_ULL(38)
+	#define BNXT_FW_CAP_CFA_RFS_RING_TBL_IDX_V3	BIT_ULL(39)
+	#define BNXT_FW_CAP_VNIC_RE_FLUSH		BIT_ULL(40)
+	#define BNXT_FW_CAP_SW_MAX_RESOURCE_LIMITS	BIT_ULL(41)
+	#define BNXT_FW_CAP_NPAR_1_2			BIT_ULL(42)
+	#define BNXT_FW_CAP_MIRROR_ON_ROCE		BIT_ULL(43)
+
+	u32			fw_dbg_cap;
+
+#define BNXT_NEW_RM(bp)		((bp)->fw_cap & BNXT_FW_CAP_NEW_RM)
+#define BNXT_PTP_USE_RTC(bp)	(!BNXT_MH(bp) && \
+				 ((bp)->fw_cap & BNXT_FW_CAP_PTP_RTC))
+#define BNXT_SUPPORTS_NTUPLE_VNIC(bp)	\
+	(BNXT_PF(bp) && ((bp)->fw_cap & BNXT_FW_CAP_CFA_RFS_RING_TBL_IDX_V3))
+
+#define BNXT_SUPPORTS_MULTI_RSS_CTX(bp)				\
+	(BNXT_PF(bp) && BNXT_SUPPORTS_NTUPLE_VNIC(bp) &&	\
+	 ((bp)->rss_cap & BNXT_RSS_CAP_MULTI_RSS_CTX))
+#define BNXT_ROCE_VF_DYN_ALLOC_CAP(bp)				\
+	((bp)->fw_cap & BNXT_FW_CAP_ROCE_VF_DYN_ALLOC_SUPPORT)
+#define BNXT_SUPPORTS_QUEUE_API(bp)				\
+	(BNXT_PF(bp) && BNXT_SUPPORTS_NTUPLE_VNIC(bp) &&	\
+	 ((bp)->fw_cap & BNXT_FW_CAP_VNIC_RE_FLUSH))
+#define BNXT_RDMA_SRIOV_EN(bp)		\
+	((bp)->fw_cap & BNXT_FW_CAP_ENABLE_RDMA_SRIOV)
+#define BNXT_ROCE_VF_RESC_CAP(bp)	\
+	((bp)->fw_cap & BNXT_FW_CAP_ROCE_VF_RESC_MGMT_SUPPORTED)
+#define BNXT_SW_RES_LMT(bp)		\
+	((bp)->fw_cap & BNXT_FW_CAP_SW_MAX_RESOURCE_LIMITS)
+#define BNXT_MIRROR_ON_ROCE_CAP(bp)	\
+	((bp)->fw_cap & BNXT_FW_CAP_MIRROR_ON_ROCE)
+
 	u32			hwrm_spec_code;
 	u16			hwrm_cmd_seq;
-	u32			hwrm_intr_seq_id;
-	void			*hwrm_short_cmd_req_addr;
-	dma_addr_t		hwrm_short_cmd_req_dma_addr;
-	void			*hwrm_cmd_resp_addr;
-	dma_addr_t		hwrm_cmd_resp_dma_addr;
-	void			*hwrm_dbg_resp_addr;
-	dma_addr_t		hwrm_dbg_resp_dma_addr;
-#define HWRM_DBG_REG_BUF_SIZE	128
-
-	struct rx_port_stats	*hw_rx_port_stats;
-	struct tx_port_stats	*hw_tx_port_stats;
-	dma_addr_t		hw_rx_port_stats_map;
-	dma_addr_t		hw_tx_port_stats_map;
-	int			hw_port_stats_size;
+	u16                     hwrm_cmd_kong_seq;
+	struct dma_pool		*hwrm_dma_pool;
+	struct hlist_head	hwrm_pending_list;
+
+	struct rtnl_link_stats64	net_stats_prev;
+	struct bnxt_stats_mem	port_stats;
+	struct bnxt_stats_mem	rx_port_stats_ext;
+	struct bnxt_stats_mem	tx_port_stats_ext;
+	u16			fw_rx_stats_ext_size;
+	u16			fw_tx_stats_ext_size;
+	u16			hw_ring_stats_size;
+	u16			pcie_stat_len;
+	u8			pri2cos_idx[8];
+	u8			pri2cos_valid;
+
+	struct bnxt_total_ring_err_stats ring_err_stats_prev;
 
 	u16			hwrm_max_req_len;
-	int			hwrm_cmd_timeout;
+	u16			hwrm_max_ext_req_len;
+	unsigned int		hwrm_cmd_timeout;
+	unsigned int		hwrm_cmd_max_timeout;
 	struct mutex		hwrm_cmd_lock;	/* serialize hwrm messages */
 	struct hwrm_ver_get_output	ver_resp;
 #define FW_VER_STR_LEN		32
 #define BC_HWRM_STR_LEN		21
 #define PHY_VER_STR_LEN         (FW_VER_STR_LEN - BC_HWRM_STR_LEN)
 	char			fw_ver_str[FW_VER_STR_LEN];
+	char			hwrm_ver_supp[FW_VER_STR_LEN];
+	char			nvm_cfg_ver[FW_VER_STR_LEN];
+	u64			fw_ver_code;
+#define BNXT_FW_VER_CODE(maj, min, bld, rsv)			\
+	((u64)(maj) << 48 | (u64)(min) << 32 | (u64)(bld) << 16 | (rsv))
+#define BNXT_FW_MAJ(bp)		((bp)->fw_ver_code >> 48)
+#define BNXT_FW_BLD(bp)		(((bp)->fw_ver_code >> 16) & 0xffff)
+
+	u16			vxlan_fw_dst_port_id;
+	u16			nge_fw_dst_port_id;
+	u16			vxlan_gpe_fw_dst_port_id;
 	__be16			vxlan_port;
-	u8			vxlan_port_cnt;
-	__le16			vxlan_fw_dst_port_id;
 	__be16			nge_port;
-	u8			nge_port_cnt;
-	__le16			nge_fw_dst_port_id;
+	__be16			vxlan_gpe_port;
 	u8			port_partition_type;
+	u8			port_count;
+	u16			br_mode;
 
-	u16			rx_coal_ticks;
-	u16			rx_coal_ticks_irq;
-	u16			rx_coal_bufs;
-	u16			rx_coal_bufs_irq;
-	u16			tx_coal_ticks;
-	u16			tx_coal_ticks_irq;
-	u16			tx_coal_bufs;
-	u16			tx_coal_bufs_irq;
-
-#define BNXT_USEC_TO_COAL_TIMER(x)	((x) * 25 / 2)
+	struct bnxt_coal_cap	coal_cap;
+	struct bnxt_coal	rx_coal;
+	struct bnxt_coal	tx_coal;
 
 	u32			stats_coal_ticks;
 #define BNXT_DEF_STATS_COAL_TICKS	 1000000
@@ -1182,19 +2603,44 @@ struct bnxt {
 #define BNXT_RX_NTP_FLTR_SP_EVENT	1
 #define BNXT_LINK_CHNG_SP_EVENT		2
 #define BNXT_HWRM_EXEC_FWD_REQ_SP_EVENT	3
-#define BNXT_VXLAN_ADD_PORT_SP_EVENT	4
-#define BNXT_VXLAN_DEL_PORT_SP_EVENT	5
 #define BNXT_RESET_TASK_SP_EVENT	6
 #define BNXT_RST_RING_SP_EVENT		7
 #define BNXT_HWRM_PF_UNLOAD_SP_EVENT	8
 #define BNXT_PERIODIC_STATS_SP_EVENT	9
 #define BNXT_HWRM_PORT_MODULE_SP_EVENT	10
 #define BNXT_RESET_TASK_SILENT_SP_EVENT	11
-#define BNXT_GENEVE_ADD_PORT_SP_EVENT	12
-#define BNXT_GENEVE_DEL_PORT_SP_EVENT	13
 #define BNXT_LINK_SPEED_CHNG_SP_EVENT	14
-
+#define BNXT_FLOW_STATS_SP_EVENT	15
+#define BNXT_UPDATE_PHY_SP_EVENT	16
+#define BNXT_RING_COAL_NOW_SP_EVENT	17
+#define BNXT_FW_RESET_NOTIFY_SP_EVENT	18
+#define BNXT_FW_EXCEPTION_SP_EVENT	19
+#define BNXT_LINK_CFG_CHANGE_SP_EVENT	21
+#define BNXT_THERMAL_THRESHOLD_SP_EVENT	22
+#define BNXT_FW_ECHO_REQUEST_SP_EVENT	23
+#define BNXT_RESTART_ULP_SP_EVENT	24
+
+	struct delayed_work	fw_reset_task;
+	int			fw_reset_state;
+#define BNXT_FW_RESET_STATE_POLL_VF	1
+#define BNXT_FW_RESET_STATE_RESET_FW	2
+#define BNXT_FW_RESET_STATE_ENABLE_DEV	3
+#define BNXT_FW_RESET_STATE_POLL_FW	4
+#define BNXT_FW_RESET_STATE_OPENING	5
+#define BNXT_FW_RESET_STATE_POLL_FW_DOWN	6
+#define BNXT_FW_RESET_STATE_ABORT	7
+
+	u16			fw_reset_min_dsecs;
+#define BNXT_DFLT_FW_RST_MIN_DSECS	20
+	u16			fw_reset_max_dsecs;
+#define BNXT_DFLT_FW_RST_MAX_DSECS	60
+	unsigned long		fw_reset_timestamp;
+
+	struct bnxt_fw_health	*fw_health;
+
+	struct bnxt_hw_resc	hw_resc;
 	struct bnxt_pf_info	pf;
+	struct bnxt_ctx_mem_info	*ctx;
 #ifdef CONFIG_BNXT_SRIOV
 	int			nr_vfs;
 	struct bnxt_vf_info	vf;
@@ -1203,7 +2649,15 @@ struct bnxt {
 #define BNXT_SRIOV_CFG_WAIT_TMO	msecs_to_jiffies(10000)
 #endif
 
+#if BITS_PER_LONG == 32
+	/* ensure atomic 64-bit doorbell writes on 32-bit systems. */
+	spinlock_t		db_lock;
+#endif
+	int			db_offset;	/* db_offset within db_size */
+	int			db_size;
+
 #define BNXT_NTP_FLTR_MAX_FLTR	4096
+#define BNXT_MAX_FLTR		(BNXT_NTP_FLTR_MAX_FLTR + BNXT_L2_FLTR_MAX_FLTR)
 #define BNXT_NTP_FLTR_HASH_SIZE	512
 #define BNXT_NTP_FLTR_HASH_MASK	(BNXT_NTP_FLTR_HASH_SIZE - 1)
 	struct hlist_head	ntp_fltr_hash_tbl[BNXT_NTP_FLTR_HASH_SIZE];
@@ -1211,12 +2665,47 @@ struct bnxt {
 
 	unsigned long		*ntp_fltr_bmap;
 	int			ntp_fltr_count;
+	int			max_fltr;
 
+#define BNXT_L2_FLTR_MAX_FLTR	1024
+#define BNXT_L2_FLTR_HASH_SIZE	32
+#define BNXT_L2_FLTR_HASH_MASK	(BNXT_L2_FLTR_HASH_SIZE - 1)
+	struct hlist_head	l2_fltr_hash_tbl[BNXT_L2_FLTR_HASH_SIZE];
+
+	u32			hash_seed;
+	u64			toeplitz_prefix;
+
+	struct list_head	usr_fltr_list;
+
+	/* To protect link related settings during link changes and
+	 * ethtool settings changes.
+	 */
+	struct mutex		link_lock;
 	struct bnxt_link_info	link_info;
-	struct ethtool_eee	eee;
+	struct ethtool_keee	eee;
 	u32			lpi_tmr_lo;
 	u32			lpi_tmr_hi;
 
+	/* copied from flags and flags2 in hwrm_port_phy_qcaps_output */
+	u32			phy_flags;
+#define BNXT_PHY_FL_EEE_CAP		PORT_PHY_QCAPS_RESP_FLAGS_EEE_SUPPORTED
+#define BNXT_PHY_FL_EXT_LPBK		PORT_PHY_QCAPS_RESP_FLAGS_EXTERNAL_LPBK_SUPPORTED
+#define BNXT_PHY_FL_AN_PHY_LPBK		PORT_PHY_QCAPS_RESP_FLAGS_AUTONEG_LPBK_SUPPORTED
+#define BNXT_PHY_FL_SHARED_PORT_CFG	PORT_PHY_QCAPS_RESP_FLAGS_SHARED_PHY_CFG_SUPPORTED
+#define BNXT_PHY_FL_PORT_STATS_NO_RESET	PORT_PHY_QCAPS_RESP_FLAGS_CUMULATIVE_COUNTERS_ON_RESET
+#define BNXT_PHY_FL_NO_PHY_LPBK		PORT_PHY_QCAPS_RESP_FLAGS_LOCAL_LPBK_NOT_SUPPORTED
+#define BNXT_PHY_FL_FW_MANAGED_LKDN	PORT_PHY_QCAPS_RESP_FLAGS_FW_MANAGED_LINK_DOWN
+#define BNXT_PHY_FL_NO_FCS		PORT_PHY_QCAPS_RESP_FLAGS_NO_FCS
+#define BNXT_PHY_FL_NO_PAUSE		(PORT_PHY_QCAPS_RESP_FLAGS2_PAUSE_UNSUPPORTED << 8)
+#define BNXT_PHY_FL_NO_PFC		(PORT_PHY_QCAPS_RESP_FLAGS2_PFC_UNSUPPORTED << 8)
+#define BNXT_PHY_FL_BANK_SEL		(PORT_PHY_QCAPS_RESP_FLAGS2_BANK_ADDR_SUPPORTED << 8)
+#define BNXT_PHY_FL_SPEEDS2		(PORT_PHY_QCAPS_RESP_FLAGS2_SPEEDS2_SUPPORTED << 8)
+
+	/* copied from flags in hwrm_port_mac_qcaps_output */
+	u8			mac_flags;
+#define BNXT_MAC_FL_NO_MAC_LPBK		\
+	PORT_MAC_QCAPS_RESP_FLAGS_LOCAL_LPBK_NOT_SUPPORTED
+
 	u8			num_tests;
 	struct bnxt_test_info	*test_info;
 
@@ -1225,79 +2714,274 @@ struct bnxt {
 
 	u8			num_leds;
 	struct bnxt_led_info	leds[BNXT_MAX_LED];
+	u16			dump_flag;
+#define BNXT_DUMP_LIVE		0
+#define BNXT_DUMP_CRASH		1
+#define BNXT_DUMP_DRIVER	2
+#define BNXT_DUMP_LIVE_WITH_CTX_L1_CACHE	3
 
 	struct bpf_prog		*xdp_prog;
+
+	struct bnxt_ptp_cfg	*ptp_cfg;
+	u8			ptp_all_rx_tstamp;
+
+	/* devlink interface and vf-rep structs */
+	struct devlink		*dl;
+	struct devlink_port	dl_port;
+	enum devlink_eswitch_mode eswitch_mode;
+	struct bnxt_vf_rep	**vf_reps; /* array of vf-rep ptrs */
+	u16			*cfa_code_map; /* cfa_code -> vf_idx map */
+	u8			dsn[8];
+	struct bnxt_tc_info	*tc_info;
+	struct list_head	tc_indr_block_list;
+	struct dentry		*debugfs_pdev;
+#ifdef CONFIG_BNXT_HWMON
+	struct device		*hwmon_dev;
+	u8			warn_thresh_temp;
+	u8			crit_thresh_temp;
+	u8			fatal_thresh_temp;
+	u8			shutdown_thresh_temp;
+#endif
+	u32			thermal_threshold_type;
+	enum board_idx		board_idx;
+
+	struct bnxt_ctx_pg_info	*fw_crash_mem;
+	u32			fw_crash_len;
+	struct bnxt_bs_trace_info bs_trace[BNXT_TRACE_MAX];
 };
 
+#define BNXT_NUM_RX_RING_STATS			8
+#define BNXT_NUM_TX_RING_STATS			8
+#define BNXT_NUM_TPA_RING_STATS			4
+#define BNXT_NUM_TPA_RING_STATS_P5		5
+#define BNXT_NUM_TPA_RING_STATS_P7		6
+
+#define BNXT_RING_STATS_SIZE_P5					\
+	((BNXT_NUM_RX_RING_STATS + BNXT_NUM_TX_RING_STATS +	\
+	  BNXT_NUM_TPA_RING_STATS_P5) * 8)
+
+#define BNXT_RING_STATS_SIZE_P7					\
+	((BNXT_NUM_RX_RING_STATS + BNXT_NUM_TX_RING_STATS +	\
+	  BNXT_NUM_TPA_RING_STATS_P7) * 8)
+
+#define BNXT_GET_RING_STATS64(sw, counter)		\
+	(*((sw) + offsetof(struct ctx_hw_stats, counter) / 8))
+
+#define BNXT_GET_RX_PORT_STATS64(sw, counter)		\
+	(*((sw) + offsetof(struct rx_port_stats, counter) / 8))
+
+#define BNXT_GET_TX_PORT_STATS64(sw, counter)		\
+	(*((sw) + offsetof(struct tx_port_stats, counter) / 8))
+
+#define BNXT_PORT_STATS_SIZE				\
+	(sizeof(struct rx_port_stats) + sizeof(struct tx_port_stats) + 1024)
+
+#define BNXT_TX_PORT_STATS_BYTE_OFFSET			\
+	(sizeof(struct rx_port_stats) + 512)
+
 #define BNXT_RX_STATS_OFFSET(counter)			\
 	(offsetof(struct rx_port_stats, counter) / 8)
 
 #define BNXT_TX_STATS_OFFSET(counter)			\
 	((offsetof(struct tx_port_stats, counter) +	\
-	  sizeof(struct rx_port_stats) + 512) / 8)
+	  BNXT_TX_PORT_STATS_BYTE_OFFSET) / 8)
+
+#define BNXT_RX_STATS_EXT_OFFSET(counter)		\
+	(offsetof(struct rx_port_stats_ext, counter) / 8)
+
+#define BNXT_RX_STATS_EXT_NUM_LEGACY                   \
+	BNXT_RX_STATS_EXT_OFFSET(rx_fec_corrected_blocks)
+
+#define BNXT_TX_STATS_EXT_OFFSET(counter)		\
+	(offsetof(struct tx_port_stats_ext, counter) / 8)
+
+#define BNXT_HW_FEATURE_VLAN_ALL_RX				\
+	(NETIF_F_HW_VLAN_CTAG_RX | NETIF_F_HW_VLAN_STAG_RX)
+#define BNXT_HW_FEATURE_VLAN_ALL_TX				\
+	(NETIF_F_HW_VLAN_CTAG_TX | NETIF_F_HW_VLAN_STAG_TX)
 
 #define I2C_DEV_ADDR_A0				0xa0
 #define I2C_DEV_ADDR_A2				0xa2
-#define SFP_EEPROM_SFF_8472_COMP_ADDR		0x5e
-#define SFP_EEPROM_SFF_8472_COMP_SIZE		1
+#define SFF_DIAG_SUPPORT_OFFSET			0x5c
 #define SFF_MODULE_ID_SFP			0x3
 #define SFF_MODULE_ID_QSFP			0xc
 #define SFF_MODULE_ID_QSFP_PLUS			0xd
 #define SFF_MODULE_ID_QSFP28			0x11
 #define BNXT_MAX_PHY_I2C_RESP_SIZE		64
 
-static inline u32 bnxt_tx_avail(struct bnxt *bp, struct bnxt_tx_ring_info *txr)
+#define BNXT_HDS_THRESHOLD_MAX			1023
+
+static inline u32 bnxt_tx_avail(struct bnxt *bp,
+				const struct bnxt_tx_ring_info *txr)
+{
+	u32 used = READ_ONCE(txr->tx_prod) - READ_ONCE(txr->tx_cons);
+
+	return bp->tx_ring_size - (used & bp->tx_ring_mask);
+}
+
+static inline void bnxt_writeq(struct bnxt *bp, u64 val,
+			       volatile void __iomem *addr)
+{
+#if BITS_PER_LONG == 32
+	spin_lock(&bp->db_lock);
+	lo_hi_writeq(val, addr);
+	spin_unlock(&bp->db_lock);
+#else
+	writeq(val, addr);
+#endif
+}
+
+static inline void bnxt_writeq_relaxed(struct bnxt *bp, u64 val,
+				       volatile void __iomem *addr)
 {
-	/* Tell compiler to fetch tx indices from memory. */
-	barrier();
+#if BITS_PER_LONG == 32
+	spin_lock(&bp->db_lock);
+	lo_hi_writeq_relaxed(val, addr);
+	spin_unlock(&bp->db_lock);
+#else
+	writeq_relaxed(val, addr);
+#endif
+}
 
-	return bp->tx_ring_size -
-		((txr->tx_prod - txr->tx_cons) & bp->tx_ring_mask);
+/* For TX and RX ring doorbells with no ordering guarantee*/
+static inline void bnxt_db_write_relaxed(struct bnxt *bp,
+					 struct bnxt_db_info *db, u32 idx)
+{
+	if (bp->flags & BNXT_FLAG_CHIP_P5_PLUS) {
+		bnxt_writeq_relaxed(bp, db->db_key64 | DB_RING_IDX(db, idx),
+				    db->doorbell);
+	} else {
+		u32 db_val = db->db_key32 | DB_RING_IDX(db, idx);
+
+		writel_relaxed(db_val, db->doorbell);
+		if (bp->flags & BNXT_FLAG_DOUBLE_DB)
+			writel_relaxed(db_val, db->doorbell);
+	}
 }
 
 /* For TX and RX ring doorbells */
-static inline void bnxt_db_write(struct bnxt *bp, void __iomem *db, u32 val)
+static inline void bnxt_db_write(struct bnxt *bp, struct bnxt_db_info *db,
+				 u32 idx)
+{
+	if (bp->flags & BNXT_FLAG_CHIP_P5_PLUS) {
+		bnxt_writeq(bp, db->db_key64 | DB_RING_IDX(db, idx),
+			    db->doorbell);
+	} else {
+		u32 db_val = db->db_key32 | DB_RING_IDX(db, idx);
+
+		writel(db_val, db->doorbell);
+		if (bp->flags & BNXT_FLAG_DOUBLE_DB)
+			writel(db_val, db->doorbell);
+	}
+}
+
+/* Must hold rtnl_lock */
+static inline bool bnxt_sriov_cfg(struct bnxt *bp)
 {
-	writel(val, db);
-	if (bp->flags & BNXT_FLAG_DOUBLE_DB)
-		writel(val, db);
+#if defined(CONFIG_BNXT_SRIOV)
+	return BNXT_PF(bp) && (bp->pf.active_vfs || bp->sriov_cfg);
+#else
+	return false;
+#endif
 }
 
+extern const u16 bnxt_bstore_to_trace[];
 extern const u16 bnxt_lhint_arr[];
 
 int bnxt_alloc_rx_data(struct bnxt *bp, struct bnxt_rx_ring_info *rxr,
 		       u16 prod, gfp_t gfp);
 void bnxt_reuse_rx_data(struct bnxt_rx_ring_info *rxr, u16 cons, void *data);
+u32 bnxt_fw_health_readl(struct bnxt *bp, int reg_idx);
+bool bnxt_bs_trace_avail(struct bnxt *bp, u16 type);
 void bnxt_set_tpa_flags(struct bnxt *bp);
 void bnxt_set_ring_params(struct bnxt *);
-int bnxt_set_rx_skb_mode(struct bnxt *bp, bool page_mode);
-void bnxt_hwrm_cmd_hdr_init(struct bnxt *, void *, u16, u16, u16);
-int _hwrm_send_message(struct bnxt *, void *, u32, int);
-int hwrm_send_message(struct bnxt *, void *, u32, int);
-int hwrm_send_message_silent(struct bnxt *, void *, u32, int);
-int bnxt_hwrm_func_rgtr_async_events(struct bnxt *bp, unsigned long *bmap,
-				     int bmap_size);
-int bnxt_hwrm_vnic_cfg(struct bnxt *bp, u16 vnic_id);
+void bnxt_set_rx_skb_mode(struct bnxt *bp, bool page_mode);
+void bnxt_insert_usr_fltr(struct bnxt *bp, struct bnxt_filter_base *fltr);
+void bnxt_del_one_usr_fltr(struct bnxt *bp, struct bnxt_filter_base *fltr);
+int bnxt_hwrm_func_drv_rgtr(struct bnxt *bp, unsigned long *bmap,
+			    int bmap_size, bool async_only);
+int bnxt_hwrm_func_drv_unrgtr(struct bnxt *bp);
+void bnxt_del_l2_filter(struct bnxt *bp, struct bnxt_l2_filter *fltr);
+struct bnxt_l2_filter *bnxt_alloc_new_l2_filter(struct bnxt *bp,
+						struct bnxt_l2_key *key,
+						u16 flags);
+int bnxt_hwrm_l2_filter_free(struct bnxt *bp, struct bnxt_l2_filter *fltr);
+int bnxt_hwrm_l2_filter_alloc(struct bnxt *bp, struct bnxt_l2_filter *fltr);
+int bnxt_hwrm_cfa_ntuple_filter_free(struct bnxt *bp,
+				     struct bnxt_ntuple_filter *fltr);
+int bnxt_hwrm_cfa_ntuple_filter_alloc(struct bnxt *bp,
+				      struct bnxt_ntuple_filter *fltr);
+int bnxt_hwrm_vnic_set_tpa(struct bnxt *bp, struct bnxt_vnic_info *vnic,
+			   u32 tpa_flags);
+void bnxt_fill_ipv6_mask(__be32 mask[4]);
+void bnxt_set_dflt_rss_indir_tbl(struct bnxt *bp,
+				 struct ethtool_rxfh_context *rss_ctx);
+int bnxt_get_nr_rss_ctxs(struct bnxt *bp, int rx_rings);
+int bnxt_hwrm_vnic_cfg(struct bnxt *bp, struct bnxt_vnic_info *vnic);
+int bnxt_hwrm_vnic_alloc(struct bnxt *bp, struct bnxt_vnic_info *vnic,
+			 unsigned int start_rx_ring_idx,
+			 unsigned int nr_rings);
 int __bnxt_hwrm_get_tx_rings(struct bnxt *bp, u16 fid, int *tx_rings);
+int bnxt_nq_rings_in_use(struct bnxt *bp);
 int bnxt_hwrm_set_coal(struct bnxt *);
+size_t bnxt_copy_ctx_mem(struct bnxt *bp, struct bnxt_ctx_mem_type *ctxm,
+			 void *buf, size_t offset);
+void bnxt_free_ctx_mem(struct bnxt *bp, bool force);
+int bnxt_num_tx_to_cp(struct bnxt *bp, int tx);
 unsigned int bnxt_get_max_func_stat_ctxs(struct bnxt *bp);
-void bnxt_set_max_func_stat_ctxs(struct bnxt *bp, unsigned int max);
+unsigned int bnxt_get_avail_stat_ctxs_for_en(struct bnxt *bp);
 unsigned int bnxt_get_max_func_cp_rings(struct bnxt *bp);
-void bnxt_set_max_func_cp_rings(struct bnxt *bp, unsigned int max);
-void bnxt_set_max_func_irqs(struct bnxt *bp, unsigned int max);
+unsigned int bnxt_get_avail_cp_rings_for_en(struct bnxt *bp);
+int bnxt_reserve_rings(struct bnxt *bp, bool irq_re_init);
 void bnxt_tx_disable(struct bnxt *bp);
 void bnxt_tx_enable(struct bnxt *bp);
+void bnxt_sched_reset_txr(struct bnxt *bp, struct bnxt_tx_ring_info *txr,
+			  u16 curr);
+void bnxt_report_link(struct bnxt *bp);
+int bnxt_update_link(struct bnxt *bp, bool chng_link_state);
 int bnxt_hwrm_set_pause(struct bnxt *);
 int bnxt_hwrm_set_link_setting(struct bnxt *, bool, bool);
+int bnxt_cancel_reservations(struct bnxt *bp, bool fw_reset);
 int bnxt_hwrm_alloc_wol_fltr(struct bnxt *bp);
 int bnxt_hwrm_free_wol_fltr(struct bnxt *bp);
+int bnxt_hwrm_func_resc_qcaps(struct bnxt *bp, bool all);
+int bnxt_hwrm_func_qcaps(struct bnxt *bp);
 int bnxt_hwrm_fw_set_time(struct bnxt *);
+int bnxt_hwrm_vnic_update(struct bnxt *bp, struct bnxt_vnic_info *vnic,
+			  u8 valid);
+int bnxt_hwrm_vnic_rss_cfg_p5(struct bnxt *bp, struct bnxt_vnic_info *vnic);
+int __bnxt_setup_vnic_p5(struct bnxt *bp, struct bnxt_vnic_info *vnic);
+void bnxt_del_one_rss_ctx(struct bnxt *bp, struct bnxt_rss_ctx *rss_ctx,
+			  bool all);
 int bnxt_open_nic(struct bnxt *, bool, bool);
 int bnxt_half_open_nic(struct bnxt *bp);
 void bnxt_half_close_nic(struct bnxt *bp);
-int bnxt_close_nic(struct bnxt *, bool, bool);
-int bnxt_reserve_rings(struct bnxt *bp, int tx, int rx, int tcs, int tx_xdp);
+void bnxt_reenable_sriov(struct bnxt *bp);
+void bnxt_close_nic(struct bnxt *, bool, bool);
+void bnxt_get_ring_err_stats(struct bnxt *bp,
+			     struct bnxt_total_ring_err_stats *stats);
+bool bnxt_rfs_capable(struct bnxt *bp, bool new_rss_ctx);
+int bnxt_dbg_hwrm_rd_reg(struct bnxt *bp, u32 reg_off, u16 num_words,
+			 u32 *reg_buf);
+void bnxt_fw_exception(struct bnxt *bp);
+void bnxt_fw_reset(struct bnxt *bp);
+int bnxt_check_rings(struct bnxt *bp, int tx, int rx, bool sh, int tcs,
+		     int tx_xdp);
+int bnxt_fw_init_one(struct bnxt *bp);
+bool bnxt_hwrm_reset_permitted(struct bnxt *bp);
 int bnxt_setup_mq_tc(struct net_device *dev, u8 tc);
+struct bnxt_ntuple_filter *bnxt_lookup_ntp_filter_from_idx(struct bnxt *bp,
+				struct bnxt_ntuple_filter *fltr, u32 idx);
+u32 bnxt_get_ntp_filter_idx(struct bnxt *bp, struct flow_keys *fkeys,
+			    const struct sk_buff *skb);
+int bnxt_insert_ntp_filter(struct bnxt *bp, struct bnxt_ntuple_filter *fltr,
+			   u32 idx);
+void bnxt_del_ntp_filter(struct bnxt *bp, struct bnxt_ntuple_filter *fltr);
 int bnxt_get_max_rings(struct bnxt *, int *, int *, bool);
-void bnxt_restore_pf_fw_resources(struct bnxt *bp);
+int bnxt_restore_pf_fw_resources(struct bnxt *bp);
+int bnxt_get_port_parent_id(struct net_device *dev,
+			    struct netdev_phys_item_id *ppid);
+void bnxt_dim_work(struct work_struct *work);
+int bnxt_hwrm_set_ring_coal(struct bnxt *bp, struct bnxt_napi *bnapi);
+void bnxt_print_device_info(struct bnxt *bp);
 #endif
diff --git a/drivers/net/ethernet/broadcom/bnxt/bnxt_coredump.c b/drivers/net/ethernet/broadcom/bnxt/bnxt_coredump.c
new file mode 100644
index 000000000000..0181ab1f2dfd
--- /dev/null
+++ b/drivers/net/ethernet/broadcom/bnxt/bnxt_coredump.c
@@ -0,0 +1,676 @@
+/* Broadcom NetXtreme-C/E network driver.
+ *
+ * Copyright (c) 2021 Broadcom Limited
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation.
+ */
+
+#include <linux/types.h>
+#include <linux/errno.h>
+#include <linux/pci.h>
+#include <linux/bnxt/hsi.h>
+#include "bnxt.h"
+#include "bnxt_hwrm.h"
+#include "bnxt_coredump.h"
+
+static const u16 bnxt_bstore_to_seg_id[] = {
+	[BNXT_CTX_QP]			= BNXT_CTX_MEM_SEG_QP,
+	[BNXT_CTX_SRQ]			= BNXT_CTX_MEM_SEG_SRQ,
+	[BNXT_CTX_CQ]			= BNXT_CTX_MEM_SEG_CQ,
+	[BNXT_CTX_VNIC]			= BNXT_CTX_MEM_SEG_VNIC,
+	[BNXT_CTX_STAT]			= BNXT_CTX_MEM_SEG_STAT,
+	[BNXT_CTX_STQM]			= BNXT_CTX_MEM_SEG_STQM,
+	[BNXT_CTX_FTQM]			= BNXT_CTX_MEM_SEG_FTQM,
+	[BNXT_CTX_MRAV]			= BNXT_CTX_MEM_SEG_MRAV,
+	[BNXT_CTX_TIM]			= BNXT_CTX_MEM_SEG_TIM,
+	[BNXT_CTX_SRT]			= BNXT_CTX_MEM_SEG_SRT,
+	[BNXT_CTX_SRT2]			= BNXT_CTX_MEM_SEG_SRT2,
+	[BNXT_CTX_CRT]			= BNXT_CTX_MEM_SEG_CRT,
+	[BNXT_CTX_CRT2]			= BNXT_CTX_MEM_SEG_CRT2,
+	[BNXT_CTX_RIGP0]		= BNXT_CTX_MEM_SEG_RIGP0,
+	[BNXT_CTX_L2HWRM]		= BNXT_CTX_MEM_SEG_L2HWRM,
+	[BNXT_CTX_REHWRM]		= BNXT_CTX_MEM_SEG_REHWRM,
+	[BNXT_CTX_CA0]			= BNXT_CTX_MEM_SEG_CA0,
+	[BNXT_CTX_CA1]			= BNXT_CTX_MEM_SEG_CA1,
+	[BNXT_CTX_CA2]			= BNXT_CTX_MEM_SEG_CA2,
+	[BNXT_CTX_RIGP1]		= BNXT_CTX_MEM_SEG_RIGP1,
+	[BNXT_CTX_KONG]			= BNXT_CTX_MEM_SEG_KONG,
+	[BNXT_CTX_QPC]			= BNXT_CTX_MEM_SEG_QPC,
+};
+
+static int bnxt_dbg_hwrm_log_buffer_flush(struct bnxt *bp, u16 type, u32 flags,
+					  u32 *offset)
+{
+	struct hwrm_dbg_log_buffer_flush_output *resp;
+	struct hwrm_dbg_log_buffer_flush_input *req;
+	int rc;
+
+	rc = hwrm_req_init(bp, req, HWRM_DBG_LOG_BUFFER_FLUSH);
+	if (rc)
+		return rc;
+
+	req->flags = cpu_to_le32(flags);
+	req->type = cpu_to_le16(type);
+	resp = hwrm_req_hold(bp, req);
+	rc = hwrm_req_send(bp, req);
+	if (!rc)
+		*offset = le32_to_cpu(resp->current_buffer_offset);
+	hwrm_req_drop(bp, req);
+	return rc;
+}
+
+static int bnxt_hwrm_dbg_dma_data(struct bnxt *bp, void *msg,
+				  struct bnxt_hwrm_dbg_dma_info *info)
+{
+	struct hwrm_dbg_cmn_input *cmn_req = msg;
+	__le16 *seq_ptr = msg + info->seq_off;
+	struct hwrm_dbg_cmn_output *cmn_resp;
+	u16 seq = 0, len, segs_off;
+	dma_addr_t dma_handle;
+	void *dma_buf, *resp;
+	int rc, off = 0;
+
+	dma_buf = hwrm_req_dma_slice(bp, msg, info->dma_len, &dma_handle);
+	if (!dma_buf) {
+		hwrm_req_drop(bp, msg);
+		return -ENOMEM;
+	}
+
+	hwrm_req_timeout(bp, msg, bp->hwrm_cmd_max_timeout);
+	cmn_resp = hwrm_req_hold(bp, msg);
+	resp = cmn_resp;
+
+	segs_off = offsetof(struct hwrm_dbg_coredump_list_output,
+			    total_segments);
+	cmn_req->host_dest_addr = cpu_to_le64(dma_handle);
+	cmn_req->host_buf_len = cpu_to_le32(info->dma_len);
+	while (1) {
+		*seq_ptr = cpu_to_le16(seq);
+		rc = hwrm_req_send(bp, msg);
+		if (rc)
+			break;
+
+		len = le16_to_cpu(*((__le16 *)(resp + info->data_len_off)));
+		if (!seq &&
+		    cmn_req->req_type == cpu_to_le16(HWRM_DBG_COREDUMP_LIST)) {
+			info->segs = le16_to_cpu(*((__le16 *)(resp +
+							      segs_off)));
+			if (!info->segs) {
+				rc = -EIO;
+				break;
+			}
+
+			info->dest_buf_size = info->segs *
+					sizeof(struct coredump_segment_record);
+			info->dest_buf = kmalloc(info->dest_buf_size,
+						 GFP_KERNEL);
+			if (!info->dest_buf) {
+				rc = -ENOMEM;
+				break;
+			}
+		}
+
+		if (cmn_req->req_type ==
+				cpu_to_le16(HWRM_DBG_COREDUMP_RETRIEVE))
+			info->dest_buf_size += len;
+
+		if (info->dest_buf) {
+			if ((info->seg_start + off + len) <=
+			    BNXT_COREDUMP_BUF_LEN(info->buf_len)) {
+				u16 copylen = min_t(u16, len,
+						    info->dest_buf_size - off);
+
+				memcpy(info->dest_buf + off, dma_buf, copylen);
+				if (copylen < len)
+					break;
+			} else {
+				rc = -ENOBUFS;
+				if (cmn_req->req_type ==
+				    cpu_to_le16(HWRM_DBG_COREDUMP_LIST)) {
+					kfree(info->dest_buf);
+					info->dest_buf = NULL;
+				}
+				break;
+			}
+		}
+
+		if (!(cmn_resp->flags & HWRM_DBG_CMN_FLAGS_MORE))
+			break;
+
+		seq++;
+		off += len;
+	}
+	hwrm_req_drop(bp, msg);
+	return rc;
+}
+
+static int bnxt_hwrm_dbg_coredump_list(struct bnxt *bp,
+				       struct bnxt_coredump *coredump)
+{
+	struct bnxt_hwrm_dbg_dma_info info = {NULL};
+	struct hwrm_dbg_coredump_list_input *req;
+	int rc;
+
+	rc = hwrm_req_init(bp, req, HWRM_DBG_COREDUMP_LIST);
+	if (rc)
+		return rc;
+
+	info.dma_len = COREDUMP_LIST_BUF_LEN;
+	info.seq_off = offsetof(struct hwrm_dbg_coredump_list_input, seq_no);
+	info.data_len_off = offsetof(struct hwrm_dbg_coredump_list_output,
+				     data_len);
+
+	rc = bnxt_hwrm_dbg_dma_data(bp, req, &info);
+	if (!rc) {
+		coredump->data = info.dest_buf;
+		coredump->data_size = info.dest_buf_size;
+		coredump->total_segs = info.segs;
+	}
+	return rc;
+}
+
+static int bnxt_hwrm_dbg_coredump_initiate(struct bnxt *bp, u16 dump_type,
+					   u16 component_id, u16 segment_id)
+{
+	struct hwrm_dbg_coredump_initiate_input *req;
+	int rc;
+
+	rc = hwrm_req_init(bp, req, HWRM_DBG_COREDUMP_INITIATE);
+	if (rc)
+		return rc;
+
+	hwrm_req_timeout(bp, req, bp->hwrm_cmd_max_timeout);
+	req->component_id = cpu_to_le16(component_id);
+	req->segment_id = cpu_to_le16(segment_id);
+	if (dump_type == BNXT_DUMP_LIVE_WITH_CTX_L1_CACHE)
+		req->seg_flags = DBG_COREDUMP_INITIATE_REQ_SEG_FLAGS_COLLECT_CTX_L1_CACHE;
+
+	return hwrm_req_send(bp, req);
+}
+
+static int bnxt_hwrm_dbg_coredump_retrieve(struct bnxt *bp, u16 component_id,
+					   u16 segment_id, u32 *seg_len,
+					   void *buf, u32 buf_len, u32 offset)
+{
+	struct hwrm_dbg_coredump_retrieve_input *req;
+	struct bnxt_hwrm_dbg_dma_info info = {NULL};
+	int rc;
+
+	rc = hwrm_req_init(bp, req, HWRM_DBG_COREDUMP_RETRIEVE);
+	if (rc)
+		return rc;
+
+	req->component_id = cpu_to_le16(component_id);
+	req->segment_id = cpu_to_le16(segment_id);
+
+	info.dma_len = COREDUMP_RETRIEVE_BUF_LEN;
+	info.seq_off = offsetof(struct hwrm_dbg_coredump_retrieve_input,
+				seq_no);
+	info.data_len_off = offsetof(struct hwrm_dbg_coredump_retrieve_output,
+				     data_len);
+	if (buf) {
+		info.dest_buf = buf + offset;
+		info.buf_len = buf_len;
+		info.seg_start = offset;
+	}
+
+	rc = bnxt_hwrm_dbg_dma_data(bp, req, &info);
+	if (!rc)
+		*seg_len = info.dest_buf_size;
+
+	return rc;
+}
+
+void
+bnxt_fill_coredump_seg_hdr(struct bnxt *bp,
+			   struct bnxt_coredump_segment_hdr *seg_hdr,
+			   struct coredump_segment_record *seg_rec, u32 seg_len,
+			   int status, u32 duration, u32 instance, u32 comp_id,
+			   u32 seg_id)
+{
+	memset(seg_hdr, 0, sizeof(*seg_hdr));
+	memcpy(seg_hdr->signature, "sEgM", 4);
+	if (seg_rec) {
+		seg_hdr->component_id = (__force __le32)seg_rec->component_id;
+		seg_hdr->segment_id = (__force __le32)seg_rec->segment_id;
+		seg_hdr->low_version = seg_rec->version_low;
+		seg_hdr->high_version = seg_rec->version_hi;
+		seg_hdr->flags = cpu_to_le32(seg_rec->compress_flags);
+	} else {
+		seg_hdr->component_id = cpu_to_le32(comp_id);
+		seg_hdr->segment_id = cpu_to_le32(seg_id);
+	}
+	seg_hdr->function_id = cpu_to_le16(bp->pdev->devfn);
+	seg_hdr->length = cpu_to_le32(seg_len);
+	seg_hdr->status = cpu_to_le32(status);
+	seg_hdr->duration = cpu_to_le32(duration);
+	seg_hdr->data_offset = cpu_to_le32(sizeof(*seg_hdr));
+	seg_hdr->instance = cpu_to_le32(instance);
+}
+
+static void bnxt_fill_cmdline(struct bnxt_coredump_record *record)
+{
+	struct mm_struct *mm = current->mm;
+	int i, len, last = 0;
+
+	if (mm) {
+		len = min_t(int, mm->arg_end - mm->arg_start,
+			    sizeof(record->commandline) - 1);
+		if (len && !copy_from_user(record->commandline,
+					   (char __user *)mm->arg_start, len)) {
+			for (i = 0; i < len; i++) {
+				if (record->commandline[i])
+					last = i;
+				else
+					record->commandline[i] = ' ';
+			}
+			record->commandline[last + 1] = 0;
+			return;
+		}
+	}
+
+	strscpy(record->commandline, current->comm, TASK_COMM_LEN);
+}
+
+static void
+bnxt_fill_coredump_record(struct bnxt *bp, struct bnxt_coredump_record *record,
+			  time64_t start, s16 start_utc, u16 total_segs,
+			  int status)
+{
+	time64_t end = ktime_get_real_seconds();
+	u32 os_ver_major = 0, os_ver_minor = 0;
+	struct tm tm;
+
+	time64_to_tm(start, 0, &tm);
+	memset(record, 0, sizeof(*record));
+	memcpy(record->signature, "cOrE", 4);
+	record->flags = 0;
+	record->low_version = 0;
+	record->high_version = 1;
+	record->asic_state = 0;
+	strscpy(record->system_name, utsname()->nodename,
+		sizeof(record->system_name));
+	record->year = cpu_to_le16(tm.tm_year + 1900);
+	record->month = cpu_to_le16(tm.tm_mon + 1);
+	record->day = cpu_to_le16(tm.tm_mday);
+	record->hour = cpu_to_le16(tm.tm_hour);
+	record->minute = cpu_to_le16(tm.tm_min);
+	record->second = cpu_to_le16(tm.tm_sec);
+	record->utc_bias = cpu_to_le16(start_utc);
+	bnxt_fill_cmdline(record);
+	record->total_segments = cpu_to_le32(total_segs);
+
+	if (sscanf(utsname()->release, "%u.%u", &os_ver_major, &os_ver_minor) != 2)
+		netdev_warn(bp->dev, "Unknown OS release in coredump\n");
+	record->os_ver_major = cpu_to_le32(os_ver_major);
+	record->os_ver_minor = cpu_to_le32(os_ver_minor);
+
+	strscpy(record->os_name, utsname()->sysname, sizeof(record->os_name));
+	time64_to_tm(end, 0, &tm);
+	record->end_year = cpu_to_le16(tm.tm_year + 1900);
+	record->end_month = cpu_to_le16(tm.tm_mon + 1);
+	record->end_day = cpu_to_le16(tm.tm_mday);
+	record->end_hour = cpu_to_le16(tm.tm_hour);
+	record->end_minute = cpu_to_le16(tm.tm_min);
+	record->end_second = cpu_to_le16(tm.tm_sec);
+	record->end_utc_bias = cpu_to_le16(sys_tz.tz_minuteswest * 60);
+	record->asic_id1 = cpu_to_le32(bp->chip_num << 16 |
+				       bp->ver_resp.chip_rev << 8 |
+				       bp->ver_resp.chip_metal);
+	record->asic_id2 = 0;
+	record->coredump_status = cpu_to_le32(status);
+	record->ioctl_low_version = 0;
+	record->ioctl_high_version = 0;
+}
+
+static void bnxt_fill_drv_seg_record(struct bnxt *bp,
+				     struct bnxt_driver_segment_record *record,
+				     struct bnxt_ctx_mem_type *ctxm, u16 type)
+{
+	struct bnxt_bs_trace_info *bs_trace = &bp->bs_trace[type];
+	u32 offset = 0;
+	int rc = 0;
+
+	rc = bnxt_dbg_hwrm_log_buffer_flush(bp, type, 0, &offset);
+	if (rc)
+		return;
+
+	bnxt_bs_trace_check_wrap(bs_trace, offset);
+	record->max_entries = cpu_to_le32(ctxm->max_entries);
+	record->entry_size = cpu_to_le32(ctxm->entry_size);
+	record->offset = cpu_to_le32(bs_trace->last_offset);
+	record->wrapped = bs_trace->wrapped;
+}
+
+static u32 bnxt_get_ctx_coredump(struct bnxt *bp, void *buf, u32 offset,
+				 u32 *segs)
+{
+	struct bnxt_driver_segment_record record = {};
+	struct bnxt_coredump_segment_hdr seg_hdr;
+	struct bnxt_ctx_mem_info *ctx = bp->ctx;
+	u32 comp_id = BNXT_DRV_COMP_ID;
+	void *data = NULL;
+	size_t len = 0;
+	u16 type;
+
+	*segs = 0;
+	if (!ctx)
+		return 0;
+
+	if (buf)
+		buf += offset;
+	for (type = 0; type < BNXT_CTX_V2_MAX; type++) {
+		struct bnxt_ctx_mem_type *ctxm = &ctx->ctx_arr[type];
+		bool trace = bnxt_bs_trace_avail(bp, type);
+		u32 seg_id = bnxt_bstore_to_seg_id[type];
+		size_t seg_len, extra_hlen = 0;
+
+		if (!ctxm->mem_valid || !seg_id)
+			continue;
+
+		if (trace) {
+			extra_hlen = BNXT_SEG_RCD_LEN;
+			if (buf) {
+				u16 trace_type = bnxt_bstore_to_trace[type];
+
+				bnxt_fill_drv_seg_record(bp, &record, ctxm,
+							 trace_type);
+			}
+		}
+
+		if (buf)
+			data = buf + BNXT_SEG_HDR_LEN + extra_hlen;
+
+		seg_len = bnxt_copy_ctx_mem(bp, ctxm, data, 0) + extra_hlen;
+		if (buf) {
+			bnxt_fill_coredump_seg_hdr(bp, &seg_hdr, NULL, seg_len,
+						   0, 0, 0, comp_id, seg_id);
+			memcpy(buf, &seg_hdr, BNXT_SEG_HDR_LEN);
+			buf += BNXT_SEG_HDR_LEN;
+			if (trace)
+				memcpy(buf, &record, BNXT_SEG_RCD_LEN);
+			buf += seg_len;
+		}
+		len += BNXT_SEG_HDR_LEN + seg_len;
+		*segs += 1;
+	}
+	return len;
+}
+
+static int __bnxt_get_coredump(struct bnxt *bp, u16 dump_type, void *buf,
+			       u32 *dump_len)
+{
+	u32 ver_get_resp_len = sizeof(struct hwrm_ver_get_output);
+	u32 offset = 0, seg_hdr_len, seg_record_len, buf_len = 0;
+	struct coredump_segment_record *seg_record = NULL;
+	struct bnxt_coredump_segment_hdr seg_hdr;
+	struct bnxt_coredump coredump = {NULL};
+	time64_t start_time;
+	u16 start_utc;
+	int rc = 0, i;
+
+	if (buf)
+		buf_len = *dump_len;
+
+	start_time = ktime_get_real_seconds();
+	start_utc = sys_tz.tz_minuteswest * 60;
+	seg_hdr_len = sizeof(seg_hdr);
+
+	/* First segment should be hwrm_ver_get response.
+	 * For hwrm_ver_get response Component id = 2 and Segment id = 0.
+	 */
+	*dump_len = seg_hdr_len + ver_get_resp_len;
+	if (buf) {
+		bnxt_fill_coredump_seg_hdr(bp, &seg_hdr, NULL, ver_get_resp_len,
+					   0, 0, 0, BNXT_VER_GET_COMP_ID, 0);
+		memcpy(buf + offset, &seg_hdr, seg_hdr_len);
+		offset += seg_hdr_len;
+		memcpy(buf + offset, &bp->ver_resp, ver_get_resp_len);
+		offset += ver_get_resp_len;
+	}
+
+	if (dump_type == BNXT_DUMP_DRIVER) {
+		u32 drv_len, segs = 0;
+
+		drv_len = bnxt_get_ctx_coredump(bp, buf, offset, &segs);
+		*dump_len += drv_len;
+		offset += drv_len;
+		if (buf)
+			coredump.total_segs += segs;
+		goto err;
+	}
+
+	seg_record_len = sizeof(*seg_record);
+	rc = bnxt_hwrm_dbg_coredump_list(bp, &coredump);
+	if (rc) {
+		netdev_err(bp->dev, "Failed to get coredump segment list\n");
+		goto err;
+	}
+
+	*dump_len += seg_hdr_len * coredump.total_segs;
+
+	seg_record = (struct coredump_segment_record *)coredump.data;
+	seg_record_len = sizeof(*seg_record);
+
+	for (i = 0; i < coredump.total_segs; i++) {
+		u16 comp_id = le16_to_cpu(seg_record->component_id);
+		u16 seg_id = le16_to_cpu(seg_record->segment_id);
+		u32 duration = 0, seg_len = 0;
+		unsigned long start, end;
+
+		if (buf && ((offset + seg_hdr_len) >
+			    BNXT_COREDUMP_BUF_LEN(buf_len))) {
+			rc = -ENOBUFS;
+			goto err;
+		}
+
+		start = jiffies;
+
+		rc = bnxt_hwrm_dbg_coredump_initiate(bp, dump_type, comp_id,
+						     seg_id);
+		if (rc) {
+			netdev_err(bp->dev,
+				   "Failed to initiate coredump for seg = %d\n",
+				   seg_record->segment_id);
+			goto next_seg;
+		}
+
+		/* Write segment data into the buffer */
+		rc = bnxt_hwrm_dbg_coredump_retrieve(bp, comp_id, seg_id,
+						     &seg_len, buf, buf_len,
+						     offset + seg_hdr_len);
+		if (rc && rc == -ENOBUFS)
+			goto err;
+		else if (rc)
+			netdev_err(bp->dev,
+				   "Failed to retrieve coredump for seg = %d\n",
+				   seg_record->segment_id);
+
+next_seg:
+		end = jiffies;
+		duration = jiffies_to_msecs(end - start);
+		bnxt_fill_coredump_seg_hdr(bp, &seg_hdr, seg_record, seg_len,
+					   rc, duration, 0, 0, 0);
+
+		if (buf) {
+			/* Write segment header into the buffer */
+			memcpy(buf + offset, &seg_hdr, seg_hdr_len);
+			offset += seg_hdr_len + seg_len;
+		}
+
+		*dump_len += seg_len;
+		seg_record =
+			(struct coredump_segment_record *)((u8 *)seg_record +
+							   seg_record_len);
+	}
+
+err:
+	if (buf)
+		bnxt_fill_coredump_record(bp, buf + offset, start_time,
+					  start_utc, coredump.total_segs + 1,
+					  rc);
+	kfree(coredump.data);
+	if (!rc) {
+		*dump_len += sizeof(struct bnxt_coredump_record);
+		/* The actual coredump length can be smaller than the FW
+		 * reported length earlier.  Use the ethtool provided length.
+		 */
+		if (buf_len)
+			*dump_len = buf_len;
+	} else if (rc == -ENOBUFS) {
+		netdev_err(bp->dev, "Firmware returned large coredump buffer\n");
+	}
+	return rc;
+}
+
+static u32 bnxt_copy_crash_data(struct bnxt_ring_mem_info *rmem, void *buf,
+				u32 dump_len)
+{
+	u32 data_copied = 0;
+	u32 data_len;
+	int i;
+
+	for (i = 0; i < rmem->nr_pages; i++) {
+		data_len = rmem->page_size;
+		if (data_copied + data_len > dump_len)
+			data_len = dump_len - data_copied;
+		memcpy(buf + data_copied, rmem->pg_arr[i], data_len);
+		data_copied += data_len;
+		if (data_copied >= dump_len)
+			break;
+	}
+	return data_copied;
+}
+
+static int bnxt_copy_crash_dump(struct bnxt *bp, void *buf, u32 dump_len)
+{
+	struct bnxt_ring_mem_info *rmem;
+	u32 offset = 0;
+
+	if (!bp->fw_crash_mem)
+		return -ENOENT;
+
+	rmem = &bp->fw_crash_mem->ring_mem;
+
+	if (rmem->depth > 1) {
+		int i;
+
+		for (i = 0; i < rmem->nr_pages; i++) {
+			struct bnxt_ctx_pg_info *pg_tbl;
+
+			pg_tbl = bp->fw_crash_mem->ctx_pg_tbl[i];
+			offset += bnxt_copy_crash_data(&pg_tbl->ring_mem,
+						       buf + offset,
+						       dump_len - offset);
+			if (offset >= dump_len)
+				break;
+		}
+	} else {
+		bnxt_copy_crash_data(rmem, buf, dump_len);
+	}
+
+	return 0;
+}
+
+static bool bnxt_crash_dump_avail(struct bnxt *bp)
+{
+	u32 sig = 0;
+
+	/* First 4 bytes(signature) of crash dump is always non-zero */
+	bnxt_copy_crash_dump(bp, &sig, sizeof(sig));
+	return !!sig;
+}
+
+int bnxt_get_coredump(struct bnxt *bp, u16 dump_type, void *buf, u32 *dump_len)
+{
+	if (dump_type == BNXT_DUMP_CRASH) {
+		if (bp->fw_dbg_cap & DBG_QCAPS_RESP_FLAGS_CRASHDUMP_HOST_DDR)
+			return bnxt_copy_crash_dump(bp, buf, *dump_len);
+#ifdef CONFIG_TEE_BNXT_FW
+		else if (bp->fw_dbg_cap & DBG_QCAPS_RESP_FLAGS_CRASHDUMP_SOC_DDR)
+			return tee_bnxt_copy_coredump(buf, 0, *dump_len);
+#endif
+		else
+			return -EOPNOTSUPP;
+	} else {
+		return __bnxt_get_coredump(bp, dump_type, buf, dump_len);
+	}
+}
+
+int bnxt_hwrm_get_dump_len(struct bnxt *bp, u16 dump_type, u32 *dump_len)
+{
+	struct hwrm_dbg_qcfg_output *resp;
+	struct hwrm_dbg_qcfg_input *req;
+	int rc, hdr_len = 0;
+
+	if (!(bp->fw_cap & BNXT_FW_CAP_DBG_QCAPS))
+		return -EOPNOTSUPP;
+
+	if (dump_type == BNXT_DUMP_CRASH &&
+	    !(bp->fw_dbg_cap & DBG_QCAPS_RESP_FLAGS_CRASHDUMP_SOC_DDR ||
+	     (bp->fw_dbg_cap & DBG_QCAPS_RESP_FLAGS_CRASHDUMP_HOST_DDR)))
+		return -EOPNOTSUPP;
+
+	rc = hwrm_req_init(bp, req, HWRM_DBG_QCFG);
+	if (rc)
+		return rc;
+
+	req->fid = cpu_to_le16(0xffff);
+	if (dump_type == BNXT_DUMP_CRASH) {
+		if (bp->fw_dbg_cap & DBG_QCAPS_RESP_FLAGS_CRASHDUMP_SOC_DDR)
+			req->flags = cpu_to_le16(BNXT_DBG_FL_CR_DUMP_SIZE_SOC);
+		else
+			req->flags = cpu_to_le16(BNXT_DBG_FL_CR_DUMP_SIZE_HOST);
+	}
+
+	resp = hwrm_req_hold(bp, req);
+	rc = hwrm_req_send(bp, req);
+	if (rc)
+		goto get_dump_len_exit;
+
+	if (dump_type == BNXT_DUMP_CRASH) {
+		if (bp->fw_dbg_cap & DBG_QCAPS_RESP_FLAGS_CRASHDUMP_SOC_DDR)
+			*dump_len = BNXT_CRASH_DUMP_LEN;
+		else
+			*dump_len = le32_to_cpu(resp->crashdump_size);
+	} else {
+		/* Driver adds coredump header and "HWRM_VER_GET response"
+		 * segment additionally to coredump.
+		 */
+		hdr_len = sizeof(struct bnxt_coredump_segment_hdr) +
+		sizeof(struct hwrm_ver_get_output) +
+		sizeof(struct bnxt_coredump_record);
+		*dump_len = le32_to_cpu(resp->coredump_size) + hdr_len;
+	}
+	if (*dump_len <= hdr_len)
+		rc = -EINVAL;
+
+get_dump_len_exit:
+	hwrm_req_drop(bp, req);
+	return rc;
+}
+
+u32 bnxt_get_coredump_length(struct bnxt *bp, u16 dump_type)
+{
+	u32 len = 0;
+
+	if (dump_type == BNXT_DUMP_CRASH &&
+	    bp->fw_dbg_cap & DBG_QCAPS_RESP_FLAGS_CRASHDUMP_HOST_DDR &&
+	    bp->fw_crash_mem) {
+		if (!bnxt_crash_dump_avail(bp))
+			return 0;
+
+		return bp->fw_crash_len;
+	}
+
+	if (dump_type != BNXT_DUMP_DRIVER) {
+		if (!bnxt_hwrm_get_dump_len(bp, dump_type, &len))
+			return len;
+	}
+	if (dump_type != BNXT_DUMP_CRASH)
+		__bnxt_get_coredump(bp, dump_type, NULL, &len);
+
+	return len;
+}
diff --git a/drivers/net/ethernet/broadcom/bnxt/bnxt_coredump.h b/drivers/net/ethernet/broadcom/bnxt/bnxt_coredump.h
new file mode 100644
index 000000000000..c087df88154a
--- /dev/null
+++ b/drivers/net/ethernet/broadcom/bnxt/bnxt_coredump.h
@@ -0,0 +1,170 @@
+/* Broadcom NetXtreme-C/E network driver.
+ *
+ * Copyright (c) 2018 Broadcom Inc
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation.
+ */
+
+#ifndef BNXT_COREDUMP_H
+#define BNXT_COREDUMP_H
+
+#include <linux/utsname.h>
+#include <linux/time.h>
+#include <linux/rtc.h>
+
+struct bnxt_coredump_segment_hdr {
+	__u8 signature[4];
+	__le32 component_id;
+	__le32 segment_id;
+	__le32 flags;
+	__u8 low_version;
+	__u8 high_version;
+	__le16 function_id;
+	__le32 offset;
+	__le32 length;
+	__le32 status;
+	__le32 duration;
+	__le32 data_offset;
+	__le32 instance;
+	__le32 rsvd[5];
+};
+
+struct bnxt_coredump_record {
+	__u8 signature[4];
+	__le32 flags;
+	__u8 low_version;
+	__u8 high_version;
+	__u8 asic_state;
+	__u8 rsvd0[5];
+	char system_name[32];
+	__le16 year;
+	__le16 month;
+	__le16 day;
+	__le16 hour;
+	__le16 minute;
+	__le16 second;
+	__le16 utc_bias;
+	__le16 rsvd1;
+	char commandline[256];
+	__le32 total_segments;
+	__le32 os_ver_major;
+	__le32 os_ver_minor;
+	__le32 rsvd2;
+	char os_name[32];
+	__le16 end_year;
+	__le16 end_month;
+	__le16 end_day;
+	__le16 end_hour;
+	__le16 end_minute;
+	__le16 end_second;
+	__le16 end_utc_bias;
+	__le32 asic_id1;
+	__le32 asic_id2;
+	__le32 coredump_status;
+	__u8 ioctl_low_version;
+	__u8 ioctl_high_version;
+	__le16 rsvd3[313];
+};
+
+struct bnxt_driver_segment_record {
+	__le32 max_entries;
+	__le32 entry_size;
+	__le32 offset;
+	__u8 wrapped:1;
+	__u8 unused[3];
+};
+
+#define BNXT_VER_GET_COMP_ID	2
+#define BNXT_DRV_COMP_ID	0xd
+
+#define BNXT_CTX_MEM_SEG_ID_START  0x200
+
+#define BNXT_CTX_MEM_SEG_QP	(BNXT_CTX_MEM_SEG_ID_START + BNXT_CTX_QP)
+#define BNXT_CTX_MEM_SEG_SRQ	(BNXT_CTX_MEM_SEG_ID_START + BNXT_CTX_SRQ)
+#define BNXT_CTX_MEM_SEG_CQ	(BNXT_CTX_MEM_SEG_ID_START + BNXT_CTX_CQ)
+#define BNXT_CTX_MEM_SEG_VNIC	(BNXT_CTX_MEM_SEG_ID_START + BNXT_CTX_VNIC)
+#define BNXT_CTX_MEM_SEG_STAT	(BNXT_CTX_MEM_SEG_ID_START + BNXT_CTX_STAT)
+#define BNXT_CTX_MEM_SEG_STQM	(BNXT_CTX_MEM_SEG_ID_START + BNXT_CTX_STQM)
+#define BNXT_CTX_MEM_SEG_FTQM	(BNXT_CTX_MEM_SEG_ID_START + BNXT_CTX_FTQM)
+#define BNXT_CTX_MEM_SEG_MRAV	(BNXT_CTX_MEM_SEG_ID_START + BNXT_CTX_MRAV)
+#define BNXT_CTX_MEM_SEG_TIM	(BNXT_CTX_MEM_SEG_ID_START + BNXT_CTX_TIM)
+
+#define BNXT_CTX_MEM_SEG_SRT	0x1
+#define BNXT_CTX_MEM_SEG_SRT2	0x2
+#define BNXT_CTX_MEM_SEG_CRT	0x3
+#define BNXT_CTX_MEM_SEG_CRT2	0x4
+#define BNXT_CTX_MEM_SEG_RIGP0	0x5
+#define BNXT_CTX_MEM_SEG_L2HWRM	0x6
+#define BNXT_CTX_MEM_SEG_REHWRM	0x7
+#define BNXT_CTX_MEM_SEG_CA0	0x8
+#define BNXT_CTX_MEM_SEG_CA1	0x9
+#define BNXT_CTX_MEM_SEG_CA2	0xa
+#define BNXT_CTX_MEM_SEG_RIGP1	0xb
+#define BNXT_CTX_MEM_SEG_QPC	0xc
+#define BNXT_CTX_MEM_SEG_KONG	0xd
+
+#define BNXT_CRASH_DUMP_LEN	(8 << 20)
+
+#define COREDUMP_LIST_BUF_LEN		2048
+#define COREDUMP_RETRIEVE_BUF_LEN	4096
+
+#define BNXT_SEG_HDR_LEN	sizeof(struct bnxt_coredump_segment_hdr)
+#define BNXT_SEG_RCD_LEN	sizeof(struct bnxt_driver_segment_record)
+
+struct bnxt_coredump {
+	void		*data;
+	int		data_size;
+	u16		total_segs;
+};
+
+#define BNXT_COREDUMP_BUF_LEN(len) ((len) - sizeof(struct bnxt_coredump_record))
+
+struct bnxt_hwrm_dbg_dma_info {
+	void *dest_buf;
+	int dest_buf_size;
+	u16 dma_len;
+	u16 seq_off;
+	u16 data_len_off;
+	u16 segs;
+	u32 seg_start;
+	u32 buf_len;
+};
+
+struct hwrm_dbg_cmn_input {
+	__le16 req_type;
+	__le16 cmpl_ring;
+	__le16 seq_id;
+	__le16 target_id;
+	__le64 resp_addr;
+	__le64 host_dest_addr;
+	__le32 host_buf_len;
+};
+
+struct hwrm_dbg_cmn_output {
+	__le16 error_code;
+	__le16 req_type;
+	__le16 seq_id;
+	__le16 resp_len;
+	u8 flags;
+	#define HWRM_DBG_CMN_FLAGS_MORE	1
+};
+
+#define BNXT_DBG_FL_CR_DUMP_SIZE_SOC	\
+	DBG_QCFG_REQ_FLAGS_CRASHDUMP_SIZE_FOR_DEST_DEST_SOC_DDR
+#define BNXT_DBG_FL_CR_DUMP_SIZE_HOST	\
+	DBG_QCFG_REQ_FLAGS_CRASHDUMP_SIZE_FOR_DEST_DEST_HOST_DDR
+#define BNXT_DBG_CR_DUMP_MDM_CFG_DDR	\
+	DBG_CRASHDUMP_MEDIUM_CFG_REQ_TYPE_DDR
+
+void bnxt_fill_coredump_seg_hdr(struct bnxt *bp,
+				struct bnxt_coredump_segment_hdr *seg_hdr,
+				struct coredump_segment_record *seg_rec,
+				u32 seg_len, int status, u32 duration,
+				u32 instance, u32 comp_id, u32 seg_id);
+int bnxt_get_coredump(struct bnxt *bp, u16 dump_type, void *buf, u32 *dump_len);
+int bnxt_hwrm_get_dump_len(struct bnxt *bp, u16 dump_type, u32 *dump_len);
+u32 bnxt_get_coredump_length(struct bnxt *bp, u16 dump_type);
+
+#endif
diff --git a/drivers/net/ethernet/broadcom/bnxt/bnxt_dcb.c b/drivers/net/ethernet/broadcom/bnxt/bnxt_dcb.c
index 5c6dd0ce209f..a00b67334f9b 100644
--- a/drivers/net/ethernet/broadcom/bnxt/bnxt_dcb.c
+++ b/drivers/net/ethernet/broadcom/bnxt/bnxt_dcb.c
@@ -16,75 +16,103 @@
 #include <linux/pci.h>
 #include <linux/etherdevice.h>
 #include <rdma/ib_verbs.h>
-#include "bnxt_hsi.h"
+#include <linux/bnxt/hsi.h>
 #include "bnxt.h"
+#include "bnxt_hwrm.h"
 #include "bnxt_dcb.h"
 
 #ifdef CONFIG_BNXT_DCB
+static int bnxt_queue_to_tc(struct bnxt *bp, u8 queue_id)
+{
+	int i, j;
+
+	for (i = 0; i < bp->max_tc; i++) {
+		if (bp->q_info[i].queue_id == queue_id) {
+			for (j = 0; j < bp->max_tc; j++) {
+				if (bp->tc_to_qidx[j] == i)
+					return j;
+			}
+		}
+	}
+	return -EINVAL;
+}
+
 static int bnxt_hwrm_queue_pri2cos_cfg(struct bnxt *bp, struct ieee_ets *ets)
 {
-	struct hwrm_queue_pri2cos_cfg_input req = {0};
-	int rc = 0, i;
+	struct hwrm_queue_pri2cos_cfg_input *req;
 	u8 *pri2cos;
+	int rc, i;
 
-	bnxt_hwrm_cmd_hdr_init(bp, &req, HWRM_QUEUE_PRI2COS_CFG, -1, -1);
-	req.flags = cpu_to_le32(QUEUE_PRI2COS_CFG_REQ_FLAGS_PATH_BIDIR |
-				QUEUE_PRI2COS_CFG_REQ_FLAGS_IVLAN);
+	rc = hwrm_req_init(bp, req, HWRM_QUEUE_PRI2COS_CFG);
+	if (rc)
+		return rc;
 
-	pri2cos = &req.pri0_cos_queue_id;
+	req->flags = cpu_to_le32(QUEUE_PRI2COS_CFG_REQ_FLAGS_PATH_BIDIR |
+				 QUEUE_PRI2COS_CFG_REQ_FLAGS_IVLAN);
+
+	pri2cos = &req->pri0_cos_queue_id;
 	for (i = 0; i < IEEE_8021QAZ_MAX_TCS; i++) {
-		req.enables |= cpu_to_le32(
+		u8 qidx;
+
+		req->enables |= cpu_to_le32(
 			QUEUE_PRI2COS_CFG_REQ_ENABLES_PRI0_COS_QUEUE_ID << i);
 
-		pri2cos[i] = bp->q_info[ets->prio_tc[i]].queue_id;
+		qidx = bp->tc_to_qidx[ets->prio_tc[i]];
+		pri2cos[i] = bp->q_info[qidx].queue_id;
 	}
-	rc = hwrm_send_message(bp, &req, sizeof(req), HWRM_CMD_TIMEOUT);
-	return rc;
+	return hwrm_req_send(bp, req);
 }
 
 static int bnxt_hwrm_queue_pri2cos_qcfg(struct bnxt *bp, struct ieee_ets *ets)
 {
-	struct hwrm_queue_pri2cos_qcfg_output *resp = bp->hwrm_cmd_resp_addr;
-	struct hwrm_queue_pri2cos_qcfg_input req = {0};
-	int rc = 0;
+	struct hwrm_queue_pri2cos_qcfg_output *resp;
+	struct hwrm_queue_pri2cos_qcfg_input *req;
+	int rc;
 
-	bnxt_hwrm_cmd_hdr_init(bp, &req, HWRM_QUEUE_PRI2COS_QCFG, -1, -1);
-	req.flags = cpu_to_le32(QUEUE_PRI2COS_QCFG_REQ_FLAGS_IVLAN);
-	rc = hwrm_send_message(bp, &req, sizeof(req), HWRM_CMD_TIMEOUT);
+	rc = hwrm_req_init(bp, req, HWRM_QUEUE_PRI2COS_QCFG);
+	if (rc)
+		return rc;
+
+	req->flags = cpu_to_le32(QUEUE_PRI2COS_QCFG_REQ_FLAGS_IVLAN);
+	resp = hwrm_req_hold(bp, req);
+	rc = hwrm_req_send(bp, req);
 	if (!rc) {
 		u8 *pri2cos = &resp->pri0_cos_queue_id;
-		int i, j;
+		int i;
 
 		for (i = 0; i < IEEE_8021QAZ_MAX_TCS; i++) {
 			u8 queue_id = pri2cos[i];
+			int tc;
 
-			for (j = 0; j < bp->max_tc; j++) {
-				if (bp->q_info[j].queue_id == queue_id) {
-					ets->prio_tc[i] = j;
-					break;
-				}
-			}
+			tc = bnxt_queue_to_tc(bp, queue_id);
+			if (tc >= 0)
+				ets->prio_tc[i] = tc;
 		}
 	}
+	hwrm_req_drop(bp, req);
 	return rc;
 }
 
 static int bnxt_hwrm_queue_cos2bw_cfg(struct bnxt *bp, struct ieee_ets *ets,
 				      u8 max_tc)
 {
-	struct hwrm_queue_cos2bw_cfg_input req = {0};
+	struct hwrm_queue_cos2bw_cfg_input *req;
 	struct bnxt_cos2bw_cfg cos2bw;
-	int rc = 0, i;
-	void *data;
+	int rc, i;
 
-	bnxt_hwrm_cmd_hdr_init(bp, &req, HWRM_QUEUE_COS2BW_CFG, -1, -1);
-	data = &req.unused_0;
-	for (i = 0; i < max_tc; i++, data += sizeof(cos2bw) - 4) {
-		req.enables |= cpu_to_le32(
-			QUEUE_COS2BW_CFG_REQ_ENABLES_COS_QUEUE_ID0_VALID << i);
+	rc = hwrm_req_init(bp, req, HWRM_QUEUE_COS2BW_CFG);
+	if (rc)
+		return rc;
+
+	for (i = 0; i < max_tc; i++) {
+		u8 qidx = bp->tc_to_qidx[i];
+
+		req->enables |= cpu_to_le32(
+			QUEUE_COS2BW_CFG_REQ_ENABLES_COS_QUEUE_ID0_VALID <<
+			qidx);
 
 		memset(&cos2bw, 0, sizeof(cos2bw));
-		cos2bw.queue_id = bp->q_info[i].queue_id;
+		cos2bw.queue_id = bp->q_info[qidx].queue_id;
 		if (ets->tc_tsa[i] == IEEE_8021QAZ_TSA_STRICT) {
 			cos2bw.tsa =
 				QUEUE_COS2BW_QCFG_RESP_QUEUE_ID0_TSA_ASSIGN_SP;
@@ -93,101 +121,138 @@ static int bnxt_hwrm_queue_cos2bw_cfg(struct bnxt *bp, struct ieee_ets *ets,
 			cos2bw.tsa =
 				QUEUE_COS2BW_QCFG_RESP_QUEUE_ID0_TSA_ASSIGN_ETS;
 			cos2bw.bw_weight = ets->tc_tx_bw[i];
+			/* older firmware requires min_bw to be set to the
+			 * same weight value in percent.
+			 */
+			cos2bw.min_bw =
+				cpu_to_le32((ets->tc_tx_bw[i] * 100) |
+					    BW_VALUE_UNIT_PERCENT1_100);
 		}
-		memcpy(data, &cos2bw.queue_id, sizeof(cos2bw) - 4);
-		if (i == 0) {
-			req.queue_id0 = cos2bw.queue_id;
-			req.unused_0 = 0;
+		if (qidx == 0) {
+			req->queue_id0 = cos2bw.queue_id;
+			req->queue_id0_min_bw = cos2bw.min_bw;
+			req->queue_id0_max_bw = cos2bw.max_bw;
+			req->queue_id0_tsa_assign = cos2bw.tsa;
+			req->queue_id0_pri_lvl = cos2bw.pri_lvl;
+			req->queue_id0_bw_weight = cos2bw.bw_weight;
+		} else {
+			memcpy(&req->cfg[i - 1], &cos2bw.cfg, sizeof(cos2bw.cfg));
 		}
 	}
-	rc = hwrm_send_message(bp, &req, sizeof(req), HWRM_CMD_TIMEOUT);
-	return rc;
+	return hwrm_req_send(bp, req);
 }
 
 static int bnxt_hwrm_queue_cos2bw_qcfg(struct bnxt *bp, struct ieee_ets *ets)
 {
-	struct hwrm_queue_cos2bw_qcfg_output *resp = bp->hwrm_cmd_resp_addr;
-	struct hwrm_queue_cos2bw_qcfg_input req = {0};
+	struct hwrm_queue_cos2bw_qcfg_output *resp;
+	struct hwrm_queue_cos2bw_qcfg_input *req;
 	struct bnxt_cos2bw_cfg cos2bw;
-	void *data;
 	int rc, i;
 
-	bnxt_hwrm_cmd_hdr_init(bp, &req, HWRM_QUEUE_COS2BW_QCFG, -1, -1);
-	rc = hwrm_send_message(bp, &req, sizeof(req), HWRM_CMD_TIMEOUT);
+	rc = hwrm_req_init(bp, req, HWRM_QUEUE_COS2BW_QCFG);
 	if (rc)
 		return rc;
 
-	data = &resp->queue_id0 + offsetof(struct bnxt_cos2bw_cfg, queue_id);
-	for (i = 0; i < bp->max_tc; i++, data += sizeof(cos2bw) - 4) {
-		int j;
+	resp = hwrm_req_hold(bp, req);
+	rc = hwrm_req_send(bp, req);
+	if (rc) {
+		hwrm_req_drop(bp, req);
+		return rc;
+	}
+
+	for (i = 0; i < bp->max_tc; i++) {
+		int tc;
 
-		memcpy(&cos2bw.queue_id, data, sizeof(cos2bw) - 4);
-		if (i == 0)
+		if (i == 0) {
 			cos2bw.queue_id = resp->queue_id0;
+			cos2bw.min_bw = resp->queue_id0_min_bw;
+			cos2bw.max_bw = resp->queue_id0_max_bw;
+			cos2bw.tsa = resp->queue_id0_tsa_assign;
+			cos2bw.pri_lvl = resp->queue_id0_pri_lvl;
+			cos2bw.bw_weight = resp->queue_id0_bw_weight;
+		} else {
+			memcpy(&cos2bw.cfg, &resp->cfg[i - 1], sizeof(cos2bw.cfg));
+		}
 
-		for (j = 0; j < bp->max_tc; j++) {
-			if (bp->q_info[j].queue_id != cos2bw.queue_id)
-				continue;
-			if (cos2bw.tsa ==
-			    QUEUE_COS2BW_QCFG_RESP_QUEUE_ID0_TSA_ASSIGN_SP) {
-				ets->tc_tsa[j] = IEEE_8021QAZ_TSA_STRICT;
-			} else {
-				ets->tc_tsa[j] = IEEE_8021QAZ_TSA_ETS;
-				ets->tc_tx_bw[j] = cos2bw.bw_weight;
-			}
+		tc = bnxt_queue_to_tc(bp, cos2bw.queue_id);
+		if (tc < 0)
+			continue;
+
+		if (cos2bw.tsa ==
+		    QUEUE_COS2BW_QCFG_RESP_QUEUE_ID0_TSA_ASSIGN_SP) {
+			ets->tc_tsa[tc] = IEEE_8021QAZ_TSA_STRICT;
+		} else {
+			ets->tc_tsa[tc] = IEEE_8021QAZ_TSA_ETS;
+			ets->tc_tx_bw[tc] = cos2bw.bw_weight;
 		}
 	}
+	hwrm_req_drop(bp, req);
 	return 0;
 }
 
-static int bnxt_hwrm_queue_cfg(struct bnxt *bp, unsigned int lltc_mask)
+static int bnxt_queue_remap(struct bnxt *bp, unsigned int lltc_mask)
 {
-	struct hwrm_queue_cfg_input req = {0};
-	int i;
+	unsigned long qmap = 0;
+	int max = bp->max_tc;
+	int i, j, rc;
 
-	if (netif_running(bp->dev))
-		bnxt_tx_disable(bp);
+	/* Assign lossless TCs first */
+	for (i = 0, j = 0; i < max; ) {
+		if (lltc_mask & (1 << i)) {
+			if (BNXT_LLQ(bp->q_info[j].queue_profile)) {
+				bp->tc_to_qidx[i] = j;
+				__set_bit(j, &qmap);
+				i++;
+			}
+			j++;
+			continue;
+		}
+		i++;
+	}
 
-	bnxt_hwrm_cmd_hdr_init(bp, &req, HWRM_QUEUE_CFG, -1, -1);
-	req.flags = cpu_to_le32(QUEUE_CFG_REQ_FLAGS_PATH_BIDIR);
-	req.enables = cpu_to_le32(QUEUE_CFG_REQ_ENABLES_SERVICE_PROFILE);
+	for (i = 0, j = 0; i < max; i++) {
+		if (lltc_mask & (1 << i))
+			continue;
+		j = find_next_zero_bit(&qmap, max, j);
+		bp->tc_to_qidx[i] = j;
+		__set_bit(j, &qmap);
+		j++;
+	}
 
-	/* Configure lossless queues to lossy first */
-	req.service_profile = QUEUE_CFG_REQ_SERVICE_PROFILE_LOSSY;
-	for (i = 0; i < bp->max_tc; i++) {
-		if (BNXT_LLQ(bp->q_info[i].queue_profile)) {
-			req.queue_id = cpu_to_le32(bp->q_info[i].queue_id);
-			hwrm_send_message(bp, &req, sizeof(req),
-					  HWRM_CMD_TIMEOUT);
-			bp->q_info[i].queue_profile =
-				QUEUE_CFG_REQ_SERVICE_PROFILE_LOSSY;
+	if (netif_running(bp->dev)) {
+		bnxt_close_nic(bp, false, false);
+		rc = bnxt_open_nic(bp, false, false);
+		if (rc) {
+			netdev_warn(bp->dev, "failed to open NIC, rc = %d\n", rc);
+			return rc;
 		}
 	}
-
-	/* Now configure desired queues to lossless */
-	req.service_profile = QUEUE_CFG_REQ_SERVICE_PROFILE_LOSSLESS;
-	for (i = 0; i < bp->max_tc; i++) {
-		if (lltc_mask & (1 << i)) {
-			req.queue_id = cpu_to_le32(bp->q_info[i].queue_id);
-			hwrm_send_message(bp, &req, sizeof(req),
-					  HWRM_CMD_TIMEOUT);
-			bp->q_info[i].queue_profile =
-				QUEUE_CFG_REQ_SERVICE_PROFILE_LOSSLESS;
+	if (bp->ieee_ets) {
+		int tc = bp->num_tc;
+
+		if (!tc)
+			tc = 1;
+		rc = bnxt_hwrm_queue_cos2bw_cfg(bp, bp->ieee_ets, tc);
+		if (rc) {
+			netdev_warn(bp->dev, "failed to config BW, rc = %d\n", rc);
+			return rc;
+		}
+		rc = bnxt_hwrm_queue_pri2cos_cfg(bp, bp->ieee_ets);
+		if (rc) {
+			netdev_warn(bp->dev, "failed to config prio, rc = %d\n", rc);
+			return rc;
 		}
 	}
-	if (netif_running(bp->dev))
-		bnxt_tx_enable(bp);
-
 	return 0;
 }
 
 static int bnxt_hwrm_queue_pfc_cfg(struct bnxt *bp, struct ieee_pfc *pfc)
 {
-	struct hwrm_queue_pfcenable_cfg_input req = {0};
+	struct hwrm_queue_pfcenable_cfg_input *req;
 	struct ieee_ets *my_ets = bp->ieee_ets;
 	unsigned int tc_mask = 0, pri_mask = 0;
 	u8 i, pri, lltc_count = 0;
-	bool need_q_recfg = false;
+	bool need_q_remap = false;
 	int rc;
 
 	if (!my_ets)
@@ -207,47 +272,57 @@ static int bnxt_hwrm_queue_pfc_cfg(struct bnxt *bp, struct ieee_pfc *pfc)
 	if (lltc_count > bp->max_lltc)
 		return -EINVAL;
 
-	bnxt_hwrm_cmd_hdr_init(bp, &req, HWRM_QUEUE_PFCENABLE_CFG, -1, -1);
-	req.flags = cpu_to_le32(pri_mask);
-	rc = hwrm_send_message(bp, &req, sizeof(req), HWRM_CMD_TIMEOUT);
-	if (rc)
-		return rc;
-
 	for (i = 0; i < bp->max_tc; i++) {
 		if (tc_mask & (1 << i)) {
-			if (!BNXT_LLQ(bp->q_info[i].queue_profile))
-				need_q_recfg = true;
+			u8 qidx = bp->tc_to_qidx[i];
+
+			if (!BNXT_LLQ(bp->q_info[qidx].queue_profile)) {
+				need_q_remap = true;
+				break;
+			}
 		}
 	}
 
-	if (need_q_recfg)
-		rc = bnxt_hwrm_queue_cfg(bp, tc_mask);
+	if (need_q_remap)
+		bnxt_queue_remap(bp, tc_mask);
 
-	return rc;
+	rc = hwrm_req_init(bp, req, HWRM_QUEUE_PFCENABLE_CFG);
+	if (rc)
+		return rc;
+
+	req->flags = cpu_to_le32(pri_mask);
+	return hwrm_req_send(bp, req);
 }
 
 static int bnxt_hwrm_queue_pfc_qcfg(struct bnxt *bp, struct ieee_pfc *pfc)
 {
-	struct hwrm_queue_pfcenable_qcfg_output *resp = bp->hwrm_cmd_resp_addr;
-	struct hwrm_queue_pfcenable_qcfg_input req = {0};
+	struct hwrm_queue_pfcenable_qcfg_output *resp;
+	struct hwrm_queue_pfcenable_qcfg_input *req;
 	u8 pri_mask;
 	int rc;
 
-	bnxt_hwrm_cmd_hdr_init(bp, &req, HWRM_QUEUE_PFCENABLE_QCFG, -1, -1);
-	rc = hwrm_send_message(bp, &req, sizeof(req), HWRM_CMD_TIMEOUT);
+	rc = hwrm_req_init(bp, req, HWRM_QUEUE_PFCENABLE_QCFG);
 	if (rc)
 		return rc;
 
+	resp = hwrm_req_hold(bp, req);
+	rc = hwrm_req_send(bp, req);
+	if (rc) {
+		hwrm_req_drop(bp, req);
+		return rc;
+	}
+
 	pri_mask = le32_to_cpu(resp->flags);
 	pfc->pfc_en = pri_mask;
+	hwrm_req_drop(bp, req);
 	return 0;
 }
 
 static int bnxt_hwrm_set_dcbx_app(struct bnxt *bp, struct dcb_app *app,
 				  bool add)
 {
-	struct hwrm_fw_set_structured_data_input set = {0};
-	struct hwrm_fw_get_structured_data_input get = {0};
+	struct hwrm_fw_set_structured_data_input *set;
+	struct hwrm_fw_get_structured_data_input *get;
 	struct hwrm_struct_data_dcbx_app *fw_app;
 	struct hwrm_struct_hdr *data;
 	dma_addr_t mapping;
@@ -257,20 +332,26 @@ static int bnxt_hwrm_set_dcbx_app(struct bnxt *bp, struct dcb_app *app,
 	if (bp->hwrm_spec_code < 0x10601)
 		return 0;
 
+	rc = hwrm_req_init(bp, get, HWRM_FW_GET_STRUCTURED_DATA);
+	if (rc)
+		return rc;
+
+	hwrm_req_hold(bp, get);
+	hwrm_req_alloc_flags(bp, get, GFP_KERNEL | __GFP_ZERO);
+
 	n = IEEE_8021QAZ_MAX_TCS;
 	data_len = sizeof(*data) + sizeof(*fw_app) * n;
-	data = dma_alloc_coherent(&bp->pdev->dev, data_len, &mapping,
-				  GFP_KERNEL);
-	if (!data)
-		return -ENOMEM;
+	data = hwrm_req_dma_slice(bp, get, data_len, &mapping);
+	if (!data) {
+		rc = -ENOMEM;
+		goto set_app_exit;
+	}
 
-	memset(data, 0, data_len);
-	bnxt_hwrm_cmd_hdr_init(bp, &get, HWRM_FW_GET_STRUCTURED_DATA, -1, -1);
-	get.dest_data_addr = cpu_to_le64(mapping);
-	get.structure_id = cpu_to_le16(STRUCT_HDR_STRUCT_ID_DCBX_APP);
-	get.subtype = cpu_to_le16(HWRM_STRUCT_DATA_SUBTYPE_HOST_OPERATIONAL);
-	get.count = 0;
-	rc = hwrm_send_message(bp, &get, sizeof(get), HWRM_CMD_TIMEOUT);
+	get->dest_data_addr = cpu_to_le64(mapping);
+	get->structure_id = cpu_to_le16(STRUCT_HDR_STRUCT_ID_DCBX_APP);
+	get->subtype = cpu_to_le16(HWRM_STRUCT_DATA_SUBTYPE_HOST_OPERATIONAL);
+	get->count = 0;
+	rc = hwrm_req_send(bp, get);
 	if (rc)
 		goto set_app_exit;
 
@@ -316,22 +397,82 @@ static int bnxt_hwrm_set_dcbx_app(struct bnxt *bp, struct dcb_app *app,
 	data->len = cpu_to_le16(sizeof(*fw_app) * n);
 	data->subtype = cpu_to_le16(HWRM_STRUCT_DATA_SUBTYPE_HOST_OPERATIONAL);
 
-	bnxt_hwrm_cmd_hdr_init(bp, &set, HWRM_FW_SET_STRUCTURED_DATA, -1, -1);
-	set.src_data_addr = cpu_to_le64(mapping);
-	set.data_len = cpu_to_le16(sizeof(*data) + sizeof(*fw_app) * n);
-	set.hdr_cnt = 1;
-	rc = hwrm_send_message(bp, &set, sizeof(set), HWRM_CMD_TIMEOUT);
+	rc = hwrm_req_init(bp, set, HWRM_FW_SET_STRUCTURED_DATA);
 	if (rc)
-		rc = -EIO;
+		goto set_app_exit;
+
+	set->src_data_addr = cpu_to_le64(mapping);
+	set->data_len = cpu_to_le16(sizeof(*data) + sizeof(*fw_app) * n);
+	set->hdr_cnt = 1;
+	rc = hwrm_req_send(bp, set);
 
 set_app_exit:
-	dma_free_coherent(&bp->pdev->dev, data_len, data, mapping);
+	hwrm_req_drop(bp, get); /* dropping get request and associated slice */
+	return rc;
+}
+
+static int bnxt_hwrm_queue_dscp_qcaps(struct bnxt *bp)
+{
+	struct hwrm_queue_dscp_qcaps_output *resp;
+	struct hwrm_queue_dscp_qcaps_input *req;
+	int rc;
+
+	bp->max_dscp_value = 0;
+	if (bp->hwrm_spec_code < 0x10800 || BNXT_VF(bp))
+		return 0;
+
+	rc = hwrm_req_init(bp, req, HWRM_QUEUE_DSCP_QCAPS);
+	if (rc)
+		return rc;
+
+	resp = hwrm_req_hold(bp, req);
+	rc = hwrm_req_send_silent(bp, req);
+	if (!rc) {
+		bp->max_dscp_value = (1 << resp->num_dscp_bits) - 1;
+		if (bp->max_dscp_value < 0x3f)
+			bp->max_dscp_value = 0;
+	}
+	hwrm_req_drop(bp, req);
+	return rc;
+}
+
+static int bnxt_hwrm_queue_dscp2pri_cfg(struct bnxt *bp, struct dcb_app *app,
+					bool add)
+{
+	struct hwrm_queue_dscp2pri_cfg_input *req;
+	struct bnxt_dscp2pri_entry *dscp2pri;
+	dma_addr_t mapping;
+	int rc;
+
+	if (bp->hwrm_spec_code < 0x10800)
+		return 0;
+
+	rc = hwrm_req_init(bp, req, HWRM_QUEUE_DSCP2PRI_CFG);
+	if (rc)
+		return rc;
+
+	dscp2pri = hwrm_req_dma_slice(bp, req, sizeof(*dscp2pri), &mapping);
+	if (!dscp2pri) {
+		hwrm_req_drop(bp, req);
+		return -ENOMEM;
+	}
+
+	req->src_data_addr = cpu_to_le64(mapping);
+	dscp2pri->dscp = app->protocol;
+	if (add)
+		dscp2pri->mask = 0x3f;
+	else
+		dscp2pri->mask = 0;
+	dscp2pri->pri = app->priority;
+	req->entry_cnt = cpu_to_le16(1);
+	rc = hwrm_req_send(bp, req);
 	return rc;
 }
 
 static int bnxt_ets_validate(struct bnxt *bp, struct ieee_ets *ets, u8 *tc)
 {
 	int total_ets_bw = 0;
+	bool zero = false;
 	u8 max_tc = 0;
 	int i;
 
@@ -346,21 +487,33 @@ static int bnxt_ets_validate(struct bnxt *bp, struct ieee_ets *ets, u8 *tc)
 
 		if ((ets->tc_tx_bw[i] || ets->tc_tsa[i]) && i > bp->max_tc)
 			return -EINVAL;
+	}
 
+	for (i = 0; i < max_tc; i++) {
 		switch (ets->tc_tsa[i]) {
 		case IEEE_8021QAZ_TSA_STRICT:
 			break;
 		case IEEE_8021QAZ_TSA_ETS:
 			total_ets_bw += ets->tc_tx_bw[i];
+			zero = zero || !ets->tc_tx_bw[i];
 			break;
 		default:
 			return -ENOTSUPP;
 		}
 	}
-	if (total_ets_bw > 100)
+	if (total_ets_bw > 100) {
+		netdev_warn(bp->dev, "rejecting ETS config exceeding available bandwidth\n");
 		return -EINVAL;
+	}
+	if (zero && total_ets_bw == 100) {
+		netdev_warn(bp->dev, "rejecting ETS config starving a TC\n");
+		return -EINVAL;
+	}
 
-	*tc = max_tc + 1;
+	if (max_tc >= bp->max_tc)
+		*tc = bp->max_tc;
+	else
+		*tc = max_tc + 1;
 	return 0;
 }
 
@@ -368,24 +521,26 @@ static int bnxt_dcbnl_ieee_getets(struct net_device *dev, struct ieee_ets *ets)
 {
 	struct bnxt *bp = netdev_priv(dev);
 	struct ieee_ets *my_ets = bp->ieee_ets;
+	int rc;
 
 	ets->ets_cap = bp->max_tc;
 
 	if (!my_ets) {
-		int rc;
-
 		if (bp->dcbx_cap & DCB_CAP_DCBX_HOST)
 			return 0;
 
 		my_ets = kzalloc(sizeof(*my_ets), GFP_KERNEL);
 		if (!my_ets)
-			return 0;
+			return -ENOMEM;
 		rc = bnxt_hwrm_queue_cos2bw_qcfg(bp, my_ets);
 		if (rc)
-			return 0;
+			goto error;
 		rc = bnxt_hwrm_queue_pri2cos_qcfg(bp, my_ets);
 		if (rc)
-			return 0;
+			goto error;
+
+		/* cache result */
+		bp->ieee_ets = my_ets;
 	}
 
 	ets->cbs = my_ets->cbs;
@@ -394,6 +549,9 @@ static int bnxt_dcbnl_ieee_getets(struct net_device *dev, struct ieee_ets *ets)
 	memcpy(ets->tc_tsa, my_ets->tc_tsa, sizeof(ets->tc_tsa));
 	memcpy(ets->prio_tc, my_ets->prio_tc, sizeof(ets->prio_tc));
 	return 0;
+error:
+	kfree(my_ets);
+	return rc;
 }
 
 static int bnxt_dcbnl_ieee_setets(struct net_device *dev, struct ieee_ets *ets)
@@ -435,7 +593,7 @@ static int bnxt_dcbnl_ieee_setets(struct net_device *dev, struct ieee_ets *ets)
 static int bnxt_dcbnl_ieee_getpfc(struct net_device *dev, struct ieee_pfc *pfc)
 {
 	struct bnxt *bp = netdev_priv(dev);
-	__le64 *stats = (__le64 *)bp->hw_rx_port_stats;
+	__le64 *stats = bp->port_stats.hw_stats;
 	struct ieee_pfc *my_pfc = bp->ieee_pfc;
 	long rx_off, tx_off;
 	int i, rc;
@@ -479,7 +637,8 @@ static int bnxt_dcbnl_ieee_setpfc(struct net_device *dev, struct ieee_pfc *pfc)
 	int rc;
 
 	if (!(bp->dcbx_cap & DCB_CAP_DCBX_VER_IEEE) ||
-	    !(bp->dcbx_cap & DCB_CAP_DCBX_HOST))
+	    !(bp->dcbx_cap & DCB_CAP_DCBX_HOST) ||
+	    (bp->phy_flags & BNXT_PHY_FL_NO_PAUSE))
 		return -EINVAL;
 
 	if (!my_pfc) {
@@ -495,15 +654,30 @@ static int bnxt_dcbnl_ieee_setpfc(struct net_device *dev, struct ieee_pfc *pfc)
 	return rc;
 }
 
+static int bnxt_dcbnl_ieee_dscp_app_prep(struct bnxt *bp, struct dcb_app *app)
+{
+	if (app->selector == IEEE_8021QAZ_APP_SEL_DSCP) {
+		if (!bp->max_dscp_value)
+			return -ENOTSUPP;
+		if (app->protocol > bp->max_dscp_value)
+			return -EINVAL;
+	}
+	return 0;
+}
+
 static int bnxt_dcbnl_ieee_setapp(struct net_device *dev, struct dcb_app *app)
 {
 	struct bnxt *bp = netdev_priv(dev);
-	int rc = -EINVAL;
+	int rc;
 
 	if (!(bp->dcbx_cap & DCB_CAP_DCBX_VER_IEEE) ||
 	    !(bp->dcbx_cap & DCB_CAP_DCBX_HOST))
 		return -EINVAL;
 
+	rc = bnxt_dcbnl_ieee_dscp_app_prep(bp, app);
+	if (rc)
+		return rc;
+
 	rc = dcb_ieee_setapp(dev, app);
 	if (rc)
 		return rc;
@@ -514,6 +688,9 @@ static int bnxt_dcbnl_ieee_setapp(struct net_device *dev, struct dcb_app *app)
 	     app->protocol == ROCE_V2_UDP_DPORT))
 		rc = bnxt_hwrm_set_dcbx_app(bp, app, true);
 
+	if (app->selector == IEEE_8021QAZ_APP_SEL_DSCP)
+		rc = bnxt_hwrm_queue_dscp2pri_cfg(bp, app, true);
+
 	return rc;
 }
 
@@ -526,6 +703,10 @@ static int bnxt_dcbnl_ieee_delapp(struct net_device *dev, struct dcb_app *app)
 	    !(bp->dcbx_cap & DCB_CAP_DCBX_HOST))
 		return -EINVAL;
 
+	rc = bnxt_dcbnl_ieee_dscp_app_prep(bp, app);
+	if (rc)
+		return rc;
+
 	rc = dcb_ieee_delapp(dev, app);
 	if (rc)
 		return rc;
@@ -535,6 +716,9 @@ static int bnxt_dcbnl_ieee_delapp(struct net_device *dev, struct dcb_app *app)
 	     app->protocol == ROCE_V2_UDP_DPORT))
 		rc = bnxt_hwrm_set_dcbx_app(bp, app, false);
 
+	if (app->selector == IEEE_8021QAZ_APP_SEL_DSCP)
+		rc = bnxt_hwrm_queue_dscp2pri_cfg(bp, app, false);
+
 	return rc;
 }
 
@@ -549,12 +733,17 @@ static u8 bnxt_dcbnl_setdcbx(struct net_device *dev, u8 mode)
 {
 	struct bnxt *bp = netdev_priv(dev);
 
-	/* only support IEEE */
-	if ((mode & DCB_CAP_DCBX_VER_CEE) || !(mode & DCB_CAP_DCBX_VER_IEEE))
+	/* All firmware DCBX settings are set in NVRAM */
+	if (bp->dcbx_cap & DCB_CAP_DCBX_LLD_MANAGED)
 		return 1;
 
 	if (mode & DCB_CAP_DCBX_HOST) {
-		if (BNXT_VF(bp) || (bp->flags & BNXT_FLAG_FW_LLDP_AGENT))
+		if (BNXT_VF(bp) || (bp->fw_cap & BNXT_FW_CAP_LLDP_AGENT))
+			return 1;
+
+		/* only support IEEE */
+		if ((mode & DCB_CAP_DCBX_VER_CEE) ||
+		    !(mode & DCB_CAP_DCBX_VER_IEEE))
 			return 1;
 	}
 
@@ -578,13 +767,15 @@ static const struct dcbnl_rtnl_ops dcbnl_ops = {
 
 void bnxt_dcb_init(struct bnxt *bp)
 {
+	bp->dcbx_cap = 0;
 	if (bp->hwrm_spec_code < 0x10501)
 		return;
 
+	bnxt_hwrm_queue_dscp_qcaps(bp);
 	bp->dcbx_cap = DCB_CAP_DCBX_VER_IEEE;
-	if (BNXT_PF(bp) && !(bp->flags & BNXT_FLAG_FW_LLDP_AGENT))
+	if (BNXT_PF(bp) && !(bp->fw_cap & BNXT_FW_CAP_LLDP_AGENT))
 		bp->dcbx_cap |= DCB_CAP_DCBX_HOST;
-	else
+	else if (bp->fw_cap & BNXT_FW_CAP_DCBX_AGENT)
 		bp->dcbx_cap |= DCB_CAP_DCBX_LLD_MANAGED;
 	bp->dev->dcbnl_ops = &dcbnl_ops;
 }
diff --git a/drivers/net/ethernet/broadcom/bnxt/bnxt_dcb.h b/drivers/net/ethernet/broadcom/bnxt/bnxt_dcb.h
index ecd0a5e46a49..5b2a6f678244 100644
--- a/drivers/net/ethernet/broadcom/bnxt/bnxt_dcb.h
+++ b/drivers/net/ethernet/broadcom/bnxt/bnxt_dcb.h
@@ -1,7 +1,7 @@
 /* Broadcom NetXtreme-C/E network driver.
  *
  * Copyright (c) 2014-2016 Broadcom Corporation
- * Copyright (c) 2016-2017 Broadcom Limited
+ * Copyright (c) 2016-2018 Broadcom Limited
  *
  * This program is free software; you can redistribute it and/or modify
  * it under the terms of the GNU General Public License as published by
@@ -23,17 +23,32 @@ struct bnxt_dcb {
 
 struct bnxt_cos2bw_cfg {
 	u8			pad[3];
-	u8			queue_id;
-	__le32			min_bw;
-	__le32			max_bw;
-	u8			tsa;
-	u8			pri_lvl;
-	u8			bw_weight;
+	struct_group_attr(cfg, __packed,
+		u8		queue_id;
+		__le32		min_bw;
+		__le32		max_bw;
+		u8		tsa;
+		u8		pri_lvl;
+		u8		bw_weight;
+	);
+/* for min_bw / max_bw */
+#define BW_VALUE_UNIT_PERCENT1_100		(0x1UL << 29)
 	u8			unused;
 };
 
+struct bnxt_dscp2pri_entry {
+	u8	dscp;
+	u8	mask;
+	u8	pri;
+};
+
 #define BNXT_LLQ(q_profile)	\
-	((q_profile) == QUEUE_QPORTCFG_RESP_QUEUE_ID0_SERVICE_PROFILE_LOSSLESS)
+	((q_profile) ==		\
+	 QUEUE_QPORTCFG_RESP_QUEUE_ID0_SERVICE_PROFILE_LOSSLESS_ROCE)
+
+#define BNXT_CNPQ(q_profile)	\
+	((q_profile) ==		\
+	 QUEUE_QPORTCFG_RESP_QUEUE_ID0_SERVICE_PROFILE_LOSSY_ROCE_CNP)
 
 #define HWRM_STRUCT_DATA_SUBTYPE_HOST_OPERATIONAL	0x0300
 
diff --git a/drivers/net/ethernet/broadcom/bnxt/bnxt_debugfs.c b/drivers/net/ethernet/broadcom/bnxt/bnxt_debugfs.c
new file mode 100644
index 000000000000..3324afbb3bec
--- /dev/null
+++ b/drivers/net/ethernet/broadcom/bnxt/bnxt_debugfs.c
@@ -0,0 +1,107 @@
+/* Broadcom NetXtreme-C/E network driver.
+ *
+ * Copyright (c) 2017-2018 Broadcom Limited
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation.
+ */
+
+#include <linux/debugfs.h>
+#include <linux/module.h>
+#include <linux/pci.h>
+#include <linux/bnxt/hsi.h>
+#include <linux/dim.h>
+#include "bnxt.h"
+#include "bnxt_debugfs.h"
+
+static struct dentry *bnxt_debug_mnt;
+
+static ssize_t debugfs_dim_read(struct file *filep,
+				char __user *buffer,
+				size_t count, loff_t *ppos)
+{
+	struct dim *dim = filep->private_data;
+	int len;
+	char *buf;
+
+	if (*ppos)
+		return 0;
+	if (!dim)
+		return -ENODEV;
+	buf = kasprintf(GFP_KERNEL,
+			"state = %d\n" \
+			"profile_ix = %d\n" \
+			"mode = %d\n" \
+			"tune_state = %d\n" \
+			"steps_right = %d\n" \
+			"steps_left = %d\n" \
+			"tired = %d\n",
+			dim->state,
+			dim->profile_ix,
+			dim->mode,
+			dim->tune_state,
+			dim->steps_right,
+			dim->steps_left,
+			dim->tired);
+	if (!buf)
+		return -ENOMEM;
+	if (count < strlen(buf)) {
+		kfree(buf);
+		return -ENOSPC;
+	}
+	len = simple_read_from_buffer(buffer, count, ppos, buf, strlen(buf));
+	kfree(buf);
+	return len;
+}
+
+static const struct file_operations debugfs_dim_fops = {
+	.owner = THIS_MODULE,
+	.open = simple_open,
+	.read = debugfs_dim_read,
+};
+
+static void debugfs_dim_ring_init(struct dim *dim, int ring_idx,
+				  struct dentry *dd)
+{
+	static char qname[12];
+
+	snprintf(qname, sizeof(qname), "%d", ring_idx);
+	debugfs_create_file(qname, 0600, dd, dim, &debugfs_dim_fops);
+}
+
+void bnxt_debug_dev_init(struct bnxt *bp)
+{
+	const char *pname = pci_name(bp->pdev);
+	struct dentry *dir;
+	int i;
+
+	bp->debugfs_pdev = debugfs_create_dir(pname, bnxt_debug_mnt);
+	dir = debugfs_create_dir("dim", bp->debugfs_pdev);
+
+	/* create files for each rx ring */
+	for (i = 0; i < bp->cp_nr_rings; i++) {
+		struct bnxt_cp_ring_info *cpr = &bp->bnapi[i]->cp_ring;
+
+		if (cpr && bp->bnapi[i]->rx_ring)
+			debugfs_dim_ring_init(&cpr->dim, i, dir);
+	}
+}
+
+void bnxt_debug_dev_exit(struct bnxt *bp)
+{
+	if (bp) {
+		debugfs_remove_recursive(bp->debugfs_pdev);
+		bp->debugfs_pdev = NULL;
+	}
+}
+
+void bnxt_debug_init(void)
+{
+	bnxt_debug_mnt = debugfs_create_dir("bnxt_en", NULL);
+}
+
+void bnxt_debug_exit(void)
+{
+	debugfs_remove_recursive(bnxt_debug_mnt);
+}
diff --git a/drivers/net/ethernet/broadcom/bnxt/bnxt_debugfs.h b/drivers/net/ethernet/broadcom/bnxt/bnxt_debugfs.h
new file mode 100644
index 000000000000..a0a8d687dd99
--- /dev/null
+++ b/drivers/net/ethernet/broadcom/bnxt/bnxt_debugfs.h
@@ -0,0 +1,23 @@
+/* Broadcom NetXtreme-C/E network driver.
+ *
+ * Copyright (c) 2017-2018 Broadcom Limited
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation.
+ */
+
+#include <linux/bnxt/hsi.h>
+#include "bnxt.h"
+
+#ifdef CONFIG_DEBUG_FS
+void bnxt_debug_init(void);
+void bnxt_debug_exit(void);
+void bnxt_debug_dev_init(struct bnxt *bp);
+void bnxt_debug_dev_exit(struct bnxt *bp);
+#else
+static inline void bnxt_debug_init(void) {}
+static inline void bnxt_debug_exit(void) {}
+static inline void bnxt_debug_dev_init(struct bnxt *bp) {}
+static inline void bnxt_debug_dev_exit(struct bnxt *bp) {}
+#endif
diff --git a/drivers/net/ethernet/broadcom/bnxt/bnxt_devlink.c b/drivers/net/ethernet/broadcom/bnxt/bnxt_devlink.c
new file mode 100644
index 000000000000..02961d93ed35
--- /dev/null
+++ b/drivers/net/ethernet/broadcom/bnxt/bnxt_devlink.c
@@ -0,0 +1,1329 @@
+/* Broadcom NetXtreme-C/E network driver.
+ *
+ * Copyright (c) 2017 Broadcom Limited
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation.
+ */
+
+#include <linux/pci.h>
+#include <linux/netdevice.h>
+#include <linux/vmalloc.h>
+#include <net/devlink.h>
+#include <net/netdev_lock.h>
+#include <linux/bnxt/hsi.h>
+#include "bnxt.h"
+#include "bnxt_hwrm.h"
+#include "bnxt_vfr.h"
+#include "bnxt_devlink.h"
+#include "bnxt_ethtool.h"
+#include "bnxt_ulp.h"
+#include "bnxt_ptp.h"
+#include "bnxt_coredump.h"
+#include "bnxt_nvm_defs.h"
+
+static void __bnxt_fw_recover(struct bnxt *bp)
+{
+	if (test_bit(BNXT_STATE_FW_FATAL_COND, &bp->state) ||
+	    test_bit(BNXT_STATE_FW_NON_FATAL_COND, &bp->state))
+		bnxt_fw_reset(bp);
+	else
+		bnxt_fw_exception(bp);
+}
+
+static int
+bnxt_dl_flash_update(struct devlink *dl,
+		     struct devlink_flash_update_params *params,
+		     struct netlink_ext_ack *extack)
+{
+	struct bnxt *bp = bnxt_get_bp_from_dl(dl);
+	int rc;
+
+	devlink_flash_update_status_notify(dl, "Preparing to flash", NULL, 0, 0);
+	rc = bnxt_flash_package_from_fw_obj(bp->dev, params->fw, 0, extack);
+	if (!rc)
+		devlink_flash_update_status_notify(dl, "Flashing done", NULL, 0, 0);
+	else
+		devlink_flash_update_status_notify(dl, "Flashing failed", NULL, 0, 0);
+	return rc;
+}
+
+static int bnxt_hwrm_remote_dev_reset_set(struct bnxt *bp, bool remote_reset)
+{
+	struct hwrm_func_cfg_input *req;
+	int rc;
+
+	if (~bp->fw_cap & BNXT_FW_CAP_HOT_RESET_IF)
+		return -EOPNOTSUPP;
+
+	rc = bnxt_hwrm_func_cfg_short_req_init(bp, &req);
+	if (rc)
+		return rc;
+
+	req->fid = cpu_to_le16(0xffff);
+	req->enables = cpu_to_le32(FUNC_CFG_REQ_ENABLES_HOT_RESET_IF_SUPPORT);
+	if (remote_reset)
+		req->flags = cpu_to_le32(FUNC_CFG_REQ_FLAGS_HOT_RESET_IF_EN_DIS);
+
+	return hwrm_req_send(bp, req);
+}
+
+static char *bnxt_health_severity_str(enum bnxt_health_severity severity)
+{
+	switch (severity) {
+	case SEVERITY_NORMAL: return "normal";
+	case SEVERITY_WARNING: return "warning";
+	case SEVERITY_RECOVERABLE: return "recoverable";
+	case SEVERITY_FATAL: return "fatal";
+	default: return "unknown";
+	}
+}
+
+static char *bnxt_health_remedy_str(enum bnxt_health_remedy remedy)
+{
+	switch (remedy) {
+	case REMEDY_DEVLINK_RECOVER: return "devlink recover";
+	case REMEDY_POWER_CYCLE_DEVICE: return "device power cycle";
+	case REMEDY_POWER_CYCLE_HOST: return "host power cycle";
+	case REMEDY_FW_UPDATE: return "update firmware";
+	case REMEDY_HW_REPLACE: return "replace hardware";
+	default: return "unknown";
+	}
+}
+
+static int bnxt_fw_diagnose(struct devlink_health_reporter *reporter,
+			    struct devlink_fmsg *fmsg,
+			    struct netlink_ext_ack *extack)
+{
+	struct bnxt *bp = devlink_health_reporter_priv(reporter);
+	struct bnxt_fw_health *h = bp->fw_health;
+	u32 fw_status, fw_resets;
+
+	if (test_bit(BNXT_STATE_IN_FW_RESET, &bp->state)) {
+		devlink_fmsg_string_pair_put(fmsg, "Status", "recovering");
+		return 0;
+	}
+
+	if (!h->status_reliable) {
+		devlink_fmsg_string_pair_put(fmsg, "Status", "unknown");
+		return 0;
+	}
+
+	mutex_lock(&h->lock);
+	fw_status = bnxt_fw_health_readl(bp, BNXT_FW_HEALTH_REG);
+	if (BNXT_FW_IS_BOOTING(fw_status)) {
+		devlink_fmsg_string_pair_put(fmsg, "Status", "initializing");
+	} else if (h->severity || fw_status != BNXT_FW_STATUS_HEALTHY) {
+		if (!h->severity) {
+			h->severity = SEVERITY_FATAL;
+			h->remedy = REMEDY_POWER_CYCLE_DEVICE;
+			h->diagnoses++;
+			devlink_health_report(h->fw_reporter,
+					      "FW error diagnosed", h);
+		}
+		devlink_fmsg_string_pair_put(fmsg, "Status", "error");
+		devlink_fmsg_u32_pair_put(fmsg, "Syndrome", fw_status);
+	} else {
+		devlink_fmsg_string_pair_put(fmsg, "Status", "healthy");
+	}
+
+	devlink_fmsg_string_pair_put(fmsg, "Severity",
+				     bnxt_health_severity_str(h->severity));
+
+	if (h->severity) {
+		devlink_fmsg_string_pair_put(fmsg, "Remedy",
+					     bnxt_health_remedy_str(h->remedy));
+		if (h->remedy == REMEDY_DEVLINK_RECOVER)
+			devlink_fmsg_string_pair_put(fmsg, "Impact",
+						     "traffic+ntuple_cfg");
+	}
+
+	mutex_unlock(&h->lock);
+	if (!h->resets_reliable)
+		return 0;
+
+	fw_resets = bnxt_fw_health_readl(bp, BNXT_FW_RESET_CNT_REG);
+	devlink_fmsg_u32_pair_put(fmsg, "Resets", fw_resets);
+	devlink_fmsg_u32_pair_put(fmsg, "Arrests", h->arrests);
+	devlink_fmsg_u32_pair_put(fmsg, "Survivals", h->survivals);
+	devlink_fmsg_u32_pair_put(fmsg, "Discoveries", h->discoveries);
+	devlink_fmsg_u32_pair_put(fmsg, "Fatalities", h->fatalities);
+	devlink_fmsg_u32_pair_put(fmsg, "Diagnoses", h->diagnoses);
+	return 0;
+}
+
+static int bnxt_fw_dump(struct devlink_health_reporter *reporter,
+			struct devlink_fmsg *fmsg, void *priv_ctx,
+			struct netlink_ext_ack *extack)
+{
+	struct bnxt *bp = devlink_health_reporter_priv(reporter);
+	u32 dump_len;
+	void *data;
+	int rc;
+
+	/* TODO: no firmware dump support in devlink_health_report() context */
+	if (priv_ctx)
+		return -EOPNOTSUPP;
+
+	dump_len = bnxt_get_coredump_length(bp, BNXT_DUMP_LIVE);
+	if (!dump_len)
+		return -EIO;
+
+	data = vmalloc(dump_len);
+	if (!data)
+		return -ENOMEM;
+
+	rc = bnxt_get_coredump(bp, BNXT_DUMP_LIVE, data, &dump_len);
+	if (!rc) {
+		devlink_fmsg_pair_nest_start(fmsg, "core");
+		devlink_fmsg_binary_pair_put(fmsg, "data", data, dump_len);
+		devlink_fmsg_u32_pair_put(fmsg, "size", dump_len);
+		devlink_fmsg_pair_nest_end(fmsg);
+	}
+
+	vfree(data);
+	return rc;
+}
+
+static int bnxt_fw_recover(struct devlink_health_reporter *reporter,
+			   void *priv_ctx,
+			   struct netlink_ext_ack *extack)
+{
+	struct bnxt *bp = devlink_health_reporter_priv(reporter);
+
+	if (bp->fw_health->severity == SEVERITY_FATAL)
+		return -ENODEV;
+
+	set_bit(BNXT_STATE_RECOVER, &bp->state);
+	__bnxt_fw_recover(bp);
+
+	return -EINPROGRESS;
+}
+
+static const struct devlink_health_reporter_ops bnxt_dl_fw_reporter_ops = {
+	.name = "fw",
+	.diagnose = bnxt_fw_diagnose,
+	.dump = bnxt_fw_dump,
+	.recover = bnxt_fw_recover,
+};
+
+static struct devlink_health_reporter *
+__bnxt_dl_reporter_create(struct bnxt *bp,
+			  const struct devlink_health_reporter_ops *ops)
+{
+	struct devlink_health_reporter *reporter;
+
+	reporter = devlink_health_reporter_create(bp->dl, ops, bp);
+	if (IS_ERR(reporter)) {
+		netdev_warn(bp->dev, "Failed to create %s health reporter, rc = %ld\n",
+			    ops->name, PTR_ERR(reporter));
+		return NULL;
+	}
+
+	return reporter;
+}
+
+void bnxt_dl_fw_reporters_create(struct bnxt *bp)
+{
+	struct bnxt_fw_health *fw_health = bp->fw_health;
+
+	if (fw_health && !fw_health->fw_reporter)
+		fw_health->fw_reporter = __bnxt_dl_reporter_create(bp, &bnxt_dl_fw_reporter_ops);
+}
+
+void bnxt_dl_fw_reporters_destroy(struct bnxt *bp)
+{
+	struct bnxt_fw_health *fw_health = bp->fw_health;
+
+	if (fw_health && fw_health->fw_reporter) {
+		devlink_health_reporter_destroy(fw_health->fw_reporter);
+		fw_health->fw_reporter = NULL;
+	}
+}
+
+void bnxt_devlink_health_fw_report(struct bnxt *bp)
+{
+	struct bnxt_fw_health *fw_health = bp->fw_health;
+	int rc;
+
+	if (!fw_health)
+		return;
+
+	if (!fw_health->fw_reporter) {
+		__bnxt_fw_recover(bp);
+		return;
+	}
+
+	mutex_lock(&fw_health->lock);
+	fw_health->severity = SEVERITY_RECOVERABLE;
+	fw_health->remedy = REMEDY_DEVLINK_RECOVER;
+	mutex_unlock(&fw_health->lock);
+	rc = devlink_health_report(fw_health->fw_reporter, "FW error reported",
+				   fw_health);
+	if (rc == -ECANCELED)
+		__bnxt_fw_recover(bp);
+}
+
+void bnxt_dl_health_fw_status_update(struct bnxt *bp, bool healthy)
+{
+	struct bnxt_fw_health *fw_health = bp->fw_health;
+	u8 state;
+
+	mutex_lock(&fw_health->lock);
+	if (healthy) {
+		fw_health->severity = SEVERITY_NORMAL;
+		state = DEVLINK_HEALTH_REPORTER_STATE_HEALTHY;
+	} else {
+		fw_health->severity = SEVERITY_FATAL;
+		fw_health->remedy = REMEDY_POWER_CYCLE_DEVICE;
+		state = DEVLINK_HEALTH_REPORTER_STATE_ERROR;
+	}
+	mutex_unlock(&fw_health->lock);
+	devlink_health_reporter_state_update(fw_health->fw_reporter, state);
+}
+
+void bnxt_dl_health_fw_recovery_done(struct bnxt *bp)
+{
+	struct bnxt_dl *dl = devlink_priv(bp->dl);
+
+	devlink_health_reporter_recovery_done(bp->fw_health->fw_reporter);
+	bnxt_hwrm_remote_dev_reset_set(bp, dl->remote_reset);
+}
+
+static int bnxt_dl_info_get(struct devlink *dl, struct devlink_info_req *req,
+			    struct netlink_ext_ack *extack);
+
+static void
+bnxt_dl_livepatch_report_err(struct bnxt *bp, struct netlink_ext_ack *extack,
+			     struct hwrm_fw_livepatch_output *resp)
+{
+	int err = ((struct hwrm_err_output *)resp)->cmd_err;
+
+	switch (err) {
+	case FW_LIVEPATCH_CMD_ERR_CODE_INVALID_OPCODE:
+		netdev_err(bp->dev, "Illegal live patch opcode");
+		NL_SET_ERR_MSG_MOD(extack, "Invalid opcode");
+		break;
+	case FW_LIVEPATCH_CMD_ERR_CODE_NOT_SUPPORTED:
+		NL_SET_ERR_MSG_MOD(extack, "Live patch operation not supported");
+		break;
+	case FW_LIVEPATCH_CMD_ERR_CODE_NOT_INSTALLED:
+		NL_SET_ERR_MSG_MOD(extack, "Live patch not found");
+		break;
+	case FW_LIVEPATCH_CMD_ERR_CODE_NOT_PATCHED:
+		NL_SET_ERR_MSG_MOD(extack,
+				   "Live patch deactivation failed. Firmware not patched.");
+		break;
+	case FW_LIVEPATCH_CMD_ERR_CODE_AUTH_FAIL:
+		NL_SET_ERR_MSG_MOD(extack, "Live patch not authenticated");
+		break;
+	case FW_LIVEPATCH_CMD_ERR_CODE_INVALID_HEADER:
+		NL_SET_ERR_MSG_MOD(extack, "Incompatible live patch");
+		break;
+	case FW_LIVEPATCH_CMD_ERR_CODE_INVALID_SIZE:
+		NL_SET_ERR_MSG_MOD(extack, "Live patch has invalid size");
+		break;
+	case FW_LIVEPATCH_CMD_ERR_CODE_ALREADY_PATCHED:
+		NL_SET_ERR_MSG_MOD(extack, "Live patch already applied");
+		break;
+	default:
+		netdev_err(bp->dev, "Unexpected live patch error: %d\n", err);
+		NL_SET_ERR_MSG_MOD(extack, "Failed to activate live patch");
+		break;
+	}
+}
+
+/* Live patch status in NVM */
+#define BNXT_LIVEPATCH_NOT_INSTALLED	0
+#define BNXT_LIVEPATCH_INSTALLED	FW_LIVEPATCH_QUERY_RESP_STATUS_FLAGS_INSTALL
+#define BNXT_LIVEPATCH_REMOVED		FW_LIVEPATCH_QUERY_RESP_STATUS_FLAGS_ACTIVE
+#define BNXT_LIVEPATCH_MASK		(FW_LIVEPATCH_QUERY_RESP_STATUS_FLAGS_INSTALL | \
+					 FW_LIVEPATCH_QUERY_RESP_STATUS_FLAGS_ACTIVE)
+#define BNXT_LIVEPATCH_ACTIVATED	BNXT_LIVEPATCH_MASK
+
+#define BNXT_LIVEPATCH_STATE(flags)	((flags) & BNXT_LIVEPATCH_MASK)
+
+static int
+bnxt_dl_livepatch_activate(struct bnxt *bp, struct netlink_ext_ack *extack)
+{
+	struct hwrm_fw_livepatch_query_output *query_resp;
+	struct hwrm_fw_livepatch_query_input *query_req;
+	struct hwrm_fw_livepatch_output *patch_resp;
+	struct hwrm_fw_livepatch_input *patch_req;
+	u16 flags, live_patch_state;
+	bool activated = false;
+	u32 installed = 0;
+	u8 target;
+	int rc;
+
+	if (~bp->fw_cap & BNXT_FW_CAP_LIVEPATCH) {
+		NL_SET_ERR_MSG_MOD(extack, "Device does not support live patch");
+		return -EOPNOTSUPP;
+	}
+
+	rc = hwrm_req_init(bp, query_req, HWRM_FW_LIVEPATCH_QUERY);
+	if (rc)
+		return rc;
+	query_resp = hwrm_req_hold(bp, query_req);
+
+	rc = hwrm_req_init(bp, patch_req, HWRM_FW_LIVEPATCH);
+	if (rc) {
+		hwrm_req_drop(bp, query_req);
+		return rc;
+	}
+	patch_req->loadtype = FW_LIVEPATCH_REQ_LOADTYPE_NVM_INSTALL;
+	patch_resp = hwrm_req_hold(bp, patch_req);
+
+	for (target = 1; target <= FW_LIVEPATCH_REQ_FW_TARGET_LAST; target++) {
+		query_req->fw_target = target;
+		rc = hwrm_req_send(bp, query_req);
+		if (rc) {
+			NL_SET_ERR_MSG_MOD(extack, "Failed to query packages");
+			break;
+		}
+
+		flags = le16_to_cpu(query_resp->status_flags);
+		live_patch_state = BNXT_LIVEPATCH_STATE(flags);
+
+		if (live_patch_state == BNXT_LIVEPATCH_NOT_INSTALLED)
+			continue;
+
+		if (live_patch_state == BNXT_LIVEPATCH_ACTIVATED) {
+			activated = true;
+			continue;
+		}
+
+		if (live_patch_state == BNXT_LIVEPATCH_INSTALLED)
+			patch_req->opcode = FW_LIVEPATCH_REQ_OPCODE_ACTIVATE;
+		else if (live_patch_state == BNXT_LIVEPATCH_REMOVED)
+			patch_req->opcode = FW_LIVEPATCH_REQ_OPCODE_DEACTIVATE;
+
+		patch_req->fw_target = target;
+		rc = hwrm_req_send(bp, patch_req);
+		if (rc) {
+			bnxt_dl_livepatch_report_err(bp, extack, patch_resp);
+			break;
+		}
+		installed++;
+	}
+
+	if (!rc && !installed) {
+		if (activated) {
+			NL_SET_ERR_MSG_MOD(extack, "Live patch already activated");
+			rc = -EEXIST;
+		} else {
+			NL_SET_ERR_MSG_MOD(extack, "No live patches found");
+			rc = -ENOENT;
+		}
+	}
+	hwrm_req_drop(bp, query_req);
+	hwrm_req_drop(bp, patch_req);
+	return rc;
+}
+
+static int bnxt_dl_reload_down(struct devlink *dl, bool netns_change,
+			       enum devlink_reload_action action,
+			       enum devlink_reload_limit limit,
+			       struct netlink_ext_ack *extack)
+{
+	struct bnxt *bp = bnxt_get_bp_from_dl(dl);
+	int rc = 0;
+
+	switch (action) {
+	case DEVLINK_RELOAD_ACTION_DRIVER_REINIT: {
+		bnxt_ulp_stop(bp);
+		rtnl_lock();
+		netdev_lock(bp->dev);
+		if (bnxt_sriov_cfg(bp)) {
+			NL_SET_ERR_MSG_MOD(extack,
+					   "reload is unsupported while VFs are allocated or being configured");
+			netdev_unlock(bp->dev);
+			rtnl_unlock();
+			bnxt_ulp_start(bp, 0);
+			return -EOPNOTSUPP;
+		}
+		if (bp->dev->reg_state == NETREG_UNREGISTERED) {
+			netdev_unlock(bp->dev);
+			rtnl_unlock();
+			bnxt_ulp_start(bp, 0);
+			return -ENODEV;
+		}
+		if (netif_running(bp->dev))
+			bnxt_close_nic(bp, true, true);
+		bnxt_vf_reps_free(bp);
+		rc = bnxt_hwrm_func_drv_unrgtr(bp);
+		if (rc) {
+			NL_SET_ERR_MSG_MOD(extack, "Failed to deregister");
+			if (netif_running(bp->dev))
+				netif_close(bp->dev);
+			netdev_unlock(bp->dev);
+			rtnl_unlock();
+			break;
+		}
+		bnxt_cancel_reservations(bp, false);
+		bnxt_free_ctx_mem(bp, false);
+		break;
+	}
+	case DEVLINK_RELOAD_ACTION_FW_ACTIVATE: {
+		if (limit == DEVLINK_RELOAD_LIMIT_NO_RESET)
+			return bnxt_dl_livepatch_activate(bp, extack);
+		if (~bp->fw_cap & BNXT_FW_CAP_HOT_RESET) {
+			NL_SET_ERR_MSG_MOD(extack, "Device not capable, requires reboot");
+			return -EOPNOTSUPP;
+		}
+		if (!bnxt_hwrm_reset_permitted(bp)) {
+			NL_SET_ERR_MSG_MOD(extack,
+					   "Reset denied by firmware, it may be inhibited by remote driver");
+			return -EPERM;
+		}
+		rtnl_lock();
+		netdev_lock(bp->dev);
+		if (bp->dev->reg_state == NETREG_UNREGISTERED) {
+			netdev_unlock(bp->dev);
+			rtnl_unlock();
+			return -ENODEV;
+		}
+		if (netif_running(bp->dev))
+			set_bit(BNXT_STATE_FW_ACTIVATE, &bp->state);
+		rc = bnxt_hwrm_firmware_reset(bp->dev,
+					      FW_RESET_REQ_EMBEDDED_PROC_TYPE_CHIP,
+					      FW_RESET_REQ_SELFRST_STATUS_SELFRSTASAP,
+					      FW_RESET_REQ_FLAGS_RESET_GRACEFUL |
+					      FW_RESET_REQ_FLAGS_FW_ACTIVATION);
+		if (rc) {
+			NL_SET_ERR_MSG_MOD(extack, "Failed to activate firmware");
+			clear_bit(BNXT_STATE_FW_ACTIVATE, &bp->state);
+			netdev_unlock(bp->dev);
+			rtnl_unlock();
+		}
+		break;
+	}
+	default:
+		rc = -EOPNOTSUPP;
+	}
+
+	return rc;
+}
+
+static int bnxt_dl_reload_up(struct devlink *dl, enum devlink_reload_action action,
+			     enum devlink_reload_limit limit, u32 *actions_performed,
+			     struct netlink_ext_ack *extack)
+{
+	struct bnxt *bp = bnxt_get_bp_from_dl(dl);
+	int rc = 0;
+
+	netdev_assert_locked(bp->dev);
+
+	*actions_performed = 0;
+	switch (action) {
+	case DEVLINK_RELOAD_ACTION_DRIVER_REINIT: {
+		bnxt_fw_init_one(bp);
+		bnxt_vf_reps_alloc(bp);
+		if (netif_running(bp->dev))
+			rc = bnxt_open_nic(bp, true, true);
+		if (!rc) {
+			bnxt_reenable_sriov(bp);
+			bnxt_ptp_reapply_pps(bp);
+		}
+		break;
+	}
+	case DEVLINK_RELOAD_ACTION_FW_ACTIVATE: {
+		unsigned long start = jiffies;
+		unsigned long timeout = start + BNXT_DFLT_FW_RST_MAX_DSECS * HZ / 10;
+
+		if (limit == DEVLINK_RELOAD_LIMIT_NO_RESET)
+			break;
+		if (bp->fw_cap & BNXT_FW_CAP_ERROR_RECOVERY)
+			timeout = start + bp->fw_health->normal_func_wait_dsecs * HZ / 10;
+		if (!netif_running(bp->dev))
+			NL_SET_ERR_MSG_MOD(extack,
+					   "Device is closed, not waiting for reset notice that will never come");
+		netdev_unlock(bp->dev);
+		rtnl_unlock();
+		while (test_bit(BNXT_STATE_FW_ACTIVATE, &bp->state)) {
+			if (time_after(jiffies, timeout)) {
+				NL_SET_ERR_MSG_MOD(extack, "Activation incomplete");
+				rc = -ETIMEDOUT;
+				break;
+			}
+			if (test_bit(BNXT_STATE_ABORT_ERR, &bp->state)) {
+				NL_SET_ERR_MSG_MOD(extack, "Activation aborted");
+				rc = -ENODEV;
+				break;
+			}
+			msleep(50);
+		}
+		rtnl_lock();
+		netdev_lock(bp->dev);
+		if (!rc)
+			*actions_performed |= BIT(DEVLINK_RELOAD_ACTION_DRIVER_REINIT);
+		clear_bit(BNXT_STATE_FW_ACTIVATE, &bp->state);
+		break;
+	}
+	default:
+		return -EOPNOTSUPP;
+	}
+
+	if (!rc) {
+		bnxt_print_device_info(bp);
+		if (netif_running(bp->dev)) {
+			mutex_lock(&bp->link_lock);
+			bnxt_report_link(bp);
+			mutex_unlock(&bp->link_lock);
+		}
+		*actions_performed |= BIT(action);
+	} else if (netif_running(bp->dev)) {
+		netif_close(bp->dev);
+	}
+	netdev_unlock(bp->dev);
+	rtnl_unlock();
+	if (action == DEVLINK_RELOAD_ACTION_DRIVER_REINIT)
+		bnxt_ulp_start(bp, rc);
+	return rc;
+}
+
+static bool bnxt_nvm_test(struct bnxt *bp, struct netlink_ext_ack *extack)
+{
+	bool rc = false;
+	u32 datalen;
+	u16 index;
+	u8 *buf;
+
+	if (bnxt_find_nvram_item(bp->dev, BNX_DIR_TYPE_VPD,
+				 BNX_DIR_ORDINAL_FIRST, BNX_DIR_EXT_NONE,
+				 &index, NULL, &datalen) || !datalen) {
+		NL_SET_ERR_MSG_MOD(extack, "nvm test vpd entry error");
+		return false;
+	}
+
+	buf = kzalloc(datalen, GFP_KERNEL);
+	if (!buf) {
+		NL_SET_ERR_MSG_MOD(extack, "insufficient memory for nvm test");
+		return false;
+	}
+
+	if (bnxt_get_nvram_item(bp->dev, index, 0, datalen, buf)) {
+		NL_SET_ERR_MSG_MOD(extack, "nvm test vpd read error");
+		goto done;
+	}
+
+	if (bnxt_flash_nvram(bp->dev, BNX_DIR_TYPE_VPD, BNX_DIR_ORDINAL_FIRST,
+			     BNX_DIR_EXT_NONE, 0, 0, buf, datalen)) {
+		NL_SET_ERR_MSG_MOD(extack, "nvm test vpd write error");
+		goto done;
+	}
+
+	rc = true;
+
+done:
+	kfree(buf);
+	return rc;
+}
+
+static bool bnxt_dl_selftest_check(struct devlink *dl, unsigned int id,
+				   struct netlink_ext_ack *extack)
+{
+	return id == DEVLINK_ATTR_SELFTEST_ID_FLASH;
+}
+
+static enum devlink_selftest_status bnxt_dl_selftest_run(struct devlink *dl,
+							 unsigned int id,
+							 struct netlink_ext_ack *extack)
+{
+	struct bnxt *bp = bnxt_get_bp_from_dl(dl);
+
+	if (id == DEVLINK_ATTR_SELFTEST_ID_FLASH)
+		return bnxt_nvm_test(bp, extack) ?
+				DEVLINK_SELFTEST_STATUS_PASS :
+				DEVLINK_SELFTEST_STATUS_FAIL;
+
+	return DEVLINK_SELFTEST_STATUS_SKIP;
+}
+
+static const struct devlink_ops bnxt_dl_ops = {
+#ifdef CONFIG_BNXT_SRIOV
+	.eswitch_mode_set = bnxt_dl_eswitch_mode_set,
+	.eswitch_mode_get = bnxt_dl_eswitch_mode_get,
+#endif /* CONFIG_BNXT_SRIOV */
+	.info_get	  = bnxt_dl_info_get,
+	.flash_update	  = bnxt_dl_flash_update,
+	.reload_actions	  = BIT(DEVLINK_RELOAD_ACTION_DRIVER_REINIT) |
+			    BIT(DEVLINK_RELOAD_ACTION_FW_ACTIVATE),
+	.reload_limits	  = BIT(DEVLINK_RELOAD_LIMIT_NO_RESET),
+	.reload_down	  = bnxt_dl_reload_down,
+	.reload_up	  = bnxt_dl_reload_up,
+	.selftest_check	  = bnxt_dl_selftest_check,
+	.selftest_run	  = bnxt_dl_selftest_run,
+};
+
+static const struct devlink_ops bnxt_vf_dl_ops;
+
+enum bnxt_dl_param_id {
+	BNXT_DEVLINK_PARAM_ID_BASE = DEVLINK_PARAM_GENERIC_ID_MAX,
+	BNXT_DEVLINK_PARAM_ID_GRE_VER_CHECK,
+};
+
+static const struct bnxt_dl_nvm_param nvm_params[] = {
+	{DEVLINK_PARAM_GENERIC_ID_ENABLE_SRIOV, NVM_OFF_ENABLE_SRIOV,
+	 BNXT_NVM_SHARED_CFG, 1, 1},
+	{DEVLINK_PARAM_GENERIC_ID_IGNORE_ARI, NVM_OFF_IGNORE_ARI,
+	 BNXT_NVM_SHARED_CFG, 1, 1},
+	{DEVLINK_PARAM_GENERIC_ID_MSIX_VEC_PER_PF_MAX,
+	 NVM_OFF_MSIX_VEC_PER_PF_MAX, BNXT_NVM_SHARED_CFG, 10, 4},
+	{DEVLINK_PARAM_GENERIC_ID_MSIX_VEC_PER_PF_MIN,
+	 NVM_OFF_MSIX_VEC_PER_PF_MIN, BNXT_NVM_SHARED_CFG, 7, 4},
+	{DEVLINK_PARAM_GENERIC_ID_ENABLE_ROCE, NVM_OFF_SUPPORT_RDMA,
+	 BNXT_NVM_FUNC_CFG, 1, 1},
+	{BNXT_DEVLINK_PARAM_ID_GRE_VER_CHECK, NVM_OFF_DIS_GRE_VER_CHECK,
+	 BNXT_NVM_SHARED_CFG, 1, 1},
+};
+
+union bnxt_nvm_data {
+	u8	val8;
+	__le32	val32;
+};
+
+static void bnxt_copy_to_nvm_data(union bnxt_nvm_data *dst,
+				  union devlink_param_value *src,
+				  int nvm_num_bits, int dl_num_bytes)
+{
+	u32 val32 = 0;
+
+	if (nvm_num_bits == 1) {
+		dst->val8 = src->vbool;
+		return;
+	}
+	if (dl_num_bytes == 4)
+		val32 = src->vu32;
+	else if (dl_num_bytes == 2)
+		val32 = (u32)src->vu16;
+	else if (dl_num_bytes == 1)
+		val32 = (u32)src->vu8;
+	dst->val32 = cpu_to_le32(val32);
+}
+
+static void bnxt_copy_from_nvm_data(union devlink_param_value *dst,
+				    union bnxt_nvm_data *src,
+				    int nvm_num_bits, int dl_num_bytes)
+{
+	u32 val32;
+
+	if (nvm_num_bits == 1) {
+		dst->vbool = src->val8;
+		return;
+	}
+	val32 = le32_to_cpu(src->val32);
+	if (dl_num_bytes == 4)
+		dst->vu32 = val32;
+	else if (dl_num_bytes == 2)
+		dst->vu16 = (u16)val32;
+	else if (dl_num_bytes == 1)
+		dst->vu8 = (u8)val32;
+}
+
+static int bnxt_hwrm_get_nvm_cfg_ver(struct bnxt *bp, u32 *nvm_cfg_ver)
+{
+	struct hwrm_nvm_get_variable_input *req;
+	u16 bytes = BNXT_NVM_CFG_VER_BYTES;
+	u16 bits = BNXT_NVM_CFG_VER_BITS;
+	union devlink_param_value ver;
+	union bnxt_nvm_data *data;
+	dma_addr_t data_dma_addr;
+	int rc, i = 2;
+	u16 dim = 1;
+
+	rc = hwrm_req_init(bp, req, HWRM_NVM_GET_VARIABLE);
+	if (rc)
+		return rc;
+
+	data = hwrm_req_dma_slice(bp, req, sizeof(*data), &data_dma_addr);
+	if (!data) {
+		rc = -ENOMEM;
+		goto exit;
+	}
+
+	/* earlier devices present as an array of raw bytes */
+	if (!BNXT_CHIP_P5_PLUS(bp)) {
+		dim = 0;
+		i = 0;
+		bits *= 3;  /* array of 3 version components */
+		bytes *= 4; /* copy whole word */
+	}
+
+	hwrm_req_hold(bp, req);
+	req->dest_data_addr = cpu_to_le64(data_dma_addr);
+	req->data_len = cpu_to_le16(bits);
+	req->option_num = cpu_to_le16(NVM_OFF_NVM_CFG_VER);
+	req->dimensions = cpu_to_le16(dim);
+
+	while (i >= 0) {
+		req->index_0 = cpu_to_le16(i--);
+		rc = hwrm_req_send_silent(bp, req);
+		if (rc)
+			goto exit;
+		bnxt_copy_from_nvm_data(&ver, data, bits, bytes);
+
+		if (BNXT_CHIP_P5_PLUS(bp)) {
+			*nvm_cfg_ver <<= 8;
+			*nvm_cfg_ver |= ver.vu8;
+		} else {
+			*nvm_cfg_ver = ver.vu32;
+		}
+	}
+
+exit:
+	hwrm_req_drop(bp, req);
+	return rc;
+}
+
+static int bnxt_dl_info_put(struct bnxt *bp, struct devlink_info_req *req,
+			    enum bnxt_dl_version_type type, const char *key,
+			    char *buf)
+{
+	if (!strlen(buf))
+		return 0;
+
+	if ((bp->flags & BNXT_FLAG_CHIP_P5_PLUS) &&
+	    (!strcmp(key, DEVLINK_INFO_VERSION_GENERIC_FW_NCSI) ||
+	     !strcmp(key, DEVLINK_INFO_VERSION_GENERIC_FW_ROCE)))
+		return 0;
+
+	switch (type) {
+	case BNXT_VERSION_FIXED:
+		return devlink_info_version_fixed_put(req, key, buf);
+	case BNXT_VERSION_RUNNING:
+		return devlink_info_version_running_put(req, key, buf);
+	case BNXT_VERSION_STORED:
+		return devlink_info_version_stored_put(req, key, buf);
+	}
+	return 0;
+}
+
+#define BNXT_FW_SRT_PATCH	"fw.srt.patch"
+#define BNXT_FW_CRT_PATCH	"fw.crt.patch"
+
+static int bnxt_dl_livepatch_info_put(struct bnxt *bp,
+				      struct devlink_info_req *req,
+				      const char *key)
+{
+	struct hwrm_fw_livepatch_query_input *query;
+	struct hwrm_fw_livepatch_query_output *resp;
+	u16 flags;
+	int rc;
+
+	if (~bp->fw_cap & BNXT_FW_CAP_LIVEPATCH)
+		return 0;
+
+	rc = hwrm_req_init(bp, query, HWRM_FW_LIVEPATCH_QUERY);
+	if (rc)
+		return rc;
+
+	if (!strcmp(key, BNXT_FW_SRT_PATCH))
+		query->fw_target = FW_LIVEPATCH_QUERY_REQ_FW_TARGET_SECURE_FW;
+	else if (!strcmp(key, BNXT_FW_CRT_PATCH))
+		query->fw_target = FW_LIVEPATCH_QUERY_REQ_FW_TARGET_COMMON_FW;
+	else
+		goto exit;
+
+	resp = hwrm_req_hold(bp, query);
+	rc = hwrm_req_send(bp, query);
+	if (rc)
+		goto exit;
+
+	flags = le16_to_cpu(resp->status_flags);
+	if (flags & FW_LIVEPATCH_QUERY_RESP_STATUS_FLAGS_ACTIVE) {
+		resp->active_ver[sizeof(resp->active_ver) - 1] = '\0';
+		rc = devlink_info_version_running_put(req, key, resp->active_ver);
+		if (rc)
+			goto exit;
+	}
+
+	if (flags & FW_LIVEPATCH_QUERY_RESP_STATUS_FLAGS_INSTALL) {
+		resp->install_ver[sizeof(resp->install_ver) - 1] = '\0';
+		rc = devlink_info_version_stored_put(req, key, resp->install_ver);
+		if (rc)
+			goto exit;
+	}
+
+exit:
+	hwrm_req_drop(bp, query);
+	return rc;
+}
+
+#define HWRM_FW_VER_STR_LEN	16
+
+static int bnxt_dl_info_get(struct devlink *dl, struct devlink_info_req *req,
+			    struct netlink_ext_ack *extack)
+{
+	struct hwrm_nvm_get_dev_info_output nvm_dev_info;
+	struct bnxt *bp = bnxt_get_bp_from_dl(dl);
+	struct hwrm_ver_get_output *ver_resp;
+	char mgmt_ver[FW_VER_STR_LEN];
+	char roce_ver[FW_VER_STR_LEN];
+	char ncsi_ver[FW_VER_STR_LEN];
+	char buf[32];
+	u32 ver = 0;
+	int rc;
+
+	if (BNXT_PF(bp) && (bp->flags & BNXT_FLAG_DSN_VALID)) {
+		sprintf(buf, "%02X-%02X-%02X-%02X-%02X-%02X-%02X-%02X",
+			bp->dsn[7], bp->dsn[6], bp->dsn[5], bp->dsn[4],
+			bp->dsn[3], bp->dsn[2], bp->dsn[1], bp->dsn[0]);
+		rc = devlink_info_serial_number_put(req, buf);
+		if (rc)
+			return rc;
+	}
+
+	if (strlen(bp->board_serialno)) {
+		rc = devlink_info_board_serial_number_put(req, bp->board_serialno);
+		if (rc)
+			return rc;
+	}
+
+	rc = bnxt_dl_info_put(bp, req, BNXT_VERSION_FIXED,
+			      DEVLINK_INFO_VERSION_GENERIC_BOARD_ID,
+			      bp->board_partno);
+	if (rc)
+		return rc;
+
+	sprintf(buf, "%X", bp->chip_num);
+	rc = bnxt_dl_info_put(bp, req, BNXT_VERSION_FIXED,
+			      DEVLINK_INFO_VERSION_GENERIC_ASIC_ID, buf);
+	if (rc)
+		return rc;
+
+	ver_resp = &bp->ver_resp;
+	sprintf(buf, "%c%d", 'A' + ver_resp->chip_rev, ver_resp->chip_metal);
+	rc = bnxt_dl_info_put(bp, req, BNXT_VERSION_FIXED,
+			      DEVLINK_INFO_VERSION_GENERIC_ASIC_REV, buf);
+	if (rc)
+		return rc;
+
+	rc = bnxt_dl_info_put(bp, req, BNXT_VERSION_RUNNING,
+			      DEVLINK_INFO_VERSION_GENERIC_FW_PSID,
+			      bp->nvm_cfg_ver);
+	if (rc)
+		return rc;
+
+	buf[0] = 0;
+	strncat(buf, ver_resp->active_pkg_name, HWRM_FW_VER_STR_LEN);
+	rc = bnxt_dl_info_put(bp, req, BNXT_VERSION_RUNNING,
+			      DEVLINK_INFO_VERSION_GENERIC_FW, buf);
+	if (rc)
+		return rc;
+
+	if (BNXT_PF(bp) && !bnxt_hwrm_get_nvm_cfg_ver(bp, &ver)) {
+		sprintf(buf, "%d.%d.%d", (ver >> 16) & 0xff, (ver >> 8) & 0xff,
+			ver & 0xff);
+		rc = bnxt_dl_info_put(bp, req, BNXT_VERSION_STORED,
+				      DEVLINK_INFO_VERSION_GENERIC_FW_PSID,
+				      buf);
+		if (rc)
+			return rc;
+	}
+
+	if (ver_resp->flags & VER_GET_RESP_FLAGS_EXT_VER_AVAIL) {
+		snprintf(mgmt_ver, FW_VER_STR_LEN, "%d.%d.%d.%d",
+			 ver_resp->hwrm_fw_major, ver_resp->hwrm_fw_minor,
+			 ver_resp->hwrm_fw_build, ver_resp->hwrm_fw_patch);
+
+		snprintf(ncsi_ver, FW_VER_STR_LEN, "%d.%d.%d.%d",
+			 ver_resp->mgmt_fw_major, ver_resp->mgmt_fw_minor,
+			 ver_resp->mgmt_fw_build, ver_resp->mgmt_fw_patch);
+
+		snprintf(roce_ver, FW_VER_STR_LEN, "%d.%d.%d.%d",
+			 ver_resp->roce_fw_major, ver_resp->roce_fw_minor,
+			 ver_resp->roce_fw_build, ver_resp->roce_fw_patch);
+	} else {
+		snprintf(mgmt_ver, FW_VER_STR_LEN, "%d.%d.%d.%d",
+			 ver_resp->hwrm_fw_maj_8b, ver_resp->hwrm_fw_min_8b,
+			 ver_resp->hwrm_fw_bld_8b, ver_resp->hwrm_fw_rsvd_8b);
+
+		snprintf(ncsi_ver, FW_VER_STR_LEN, "%d.%d.%d.%d",
+			 ver_resp->mgmt_fw_maj_8b, ver_resp->mgmt_fw_min_8b,
+			 ver_resp->mgmt_fw_bld_8b, ver_resp->mgmt_fw_rsvd_8b);
+
+		snprintf(roce_ver, FW_VER_STR_LEN, "%d.%d.%d.%d",
+			 ver_resp->roce_fw_maj_8b, ver_resp->roce_fw_min_8b,
+			 ver_resp->roce_fw_bld_8b, ver_resp->roce_fw_rsvd_8b);
+	}
+	rc = bnxt_dl_info_put(bp, req, BNXT_VERSION_RUNNING,
+			      DEVLINK_INFO_VERSION_GENERIC_FW_MGMT, mgmt_ver);
+	if (rc)
+		return rc;
+
+	rc = bnxt_dl_info_put(bp, req, BNXT_VERSION_RUNNING,
+			      DEVLINK_INFO_VERSION_GENERIC_FW_MGMT_API,
+			      bp->hwrm_ver_supp);
+	if (rc)
+		return rc;
+
+	rc = bnxt_dl_info_put(bp, req, BNXT_VERSION_RUNNING,
+			      DEVLINK_INFO_VERSION_GENERIC_FW_NCSI, ncsi_ver);
+	if (rc)
+		return rc;
+
+	rc = bnxt_dl_info_put(bp, req, BNXT_VERSION_RUNNING,
+			      DEVLINK_INFO_VERSION_GENERIC_FW_ROCE, roce_ver);
+	if (rc)
+		return rc;
+
+	rc = bnxt_hwrm_nvm_get_dev_info(bp, &nvm_dev_info);
+	if (rc ||
+	    !(nvm_dev_info.flags & NVM_GET_DEV_INFO_RESP_FLAGS_FW_VER_VALID)) {
+		if (!bnxt_get_pkginfo(bp->dev, buf, sizeof(buf)))
+			return bnxt_dl_info_put(bp, req, BNXT_VERSION_STORED,
+						DEVLINK_INFO_VERSION_GENERIC_FW,
+						buf);
+		return 0;
+	}
+
+	buf[0] = 0;
+	strncat(buf, nvm_dev_info.pkg_name, HWRM_FW_VER_STR_LEN);
+	rc = bnxt_dl_info_put(bp, req, BNXT_VERSION_STORED,
+			      DEVLINK_INFO_VERSION_GENERIC_FW, buf);
+	if (rc)
+		return rc;
+
+	snprintf(mgmt_ver, FW_VER_STR_LEN, "%d.%d.%d.%d",
+		 nvm_dev_info.hwrm_fw_major, nvm_dev_info.hwrm_fw_minor,
+		 nvm_dev_info.hwrm_fw_build, nvm_dev_info.hwrm_fw_patch);
+	rc = bnxt_dl_info_put(bp, req, BNXT_VERSION_STORED,
+			      DEVLINK_INFO_VERSION_GENERIC_FW_MGMT, mgmt_ver);
+	if (rc)
+		return rc;
+
+	snprintf(ncsi_ver, FW_VER_STR_LEN, "%d.%d.%d.%d",
+		 nvm_dev_info.mgmt_fw_major, nvm_dev_info.mgmt_fw_minor,
+		 nvm_dev_info.mgmt_fw_build, nvm_dev_info.mgmt_fw_patch);
+	rc = bnxt_dl_info_put(bp, req, BNXT_VERSION_STORED,
+			      DEVLINK_INFO_VERSION_GENERIC_FW_NCSI, ncsi_ver);
+	if (rc)
+		return rc;
+
+	snprintf(roce_ver, FW_VER_STR_LEN, "%d.%d.%d.%d",
+		 nvm_dev_info.roce_fw_major, nvm_dev_info.roce_fw_minor,
+		 nvm_dev_info.roce_fw_build, nvm_dev_info.roce_fw_patch);
+	rc = bnxt_dl_info_put(bp, req, BNXT_VERSION_STORED,
+			      DEVLINK_INFO_VERSION_GENERIC_FW_ROCE, roce_ver);
+	if (rc)
+		return rc;
+
+	if (BNXT_CHIP_P5_PLUS(bp)) {
+		rc = bnxt_dl_livepatch_info_put(bp, req, BNXT_FW_SRT_PATCH);
+		if (rc)
+			return rc;
+	}
+	return bnxt_dl_livepatch_info_put(bp, req, BNXT_FW_CRT_PATCH);
+
+}
+
+static int __bnxt_hwrm_nvm_req(struct bnxt *bp,
+			       const struct bnxt_dl_nvm_param *nvm, void *msg,
+			       union devlink_param_value *val)
+{
+	struct hwrm_nvm_get_variable_input *req = msg;
+	struct hwrm_err_output *resp;
+	union bnxt_nvm_data *data;
+	dma_addr_t data_dma_addr;
+	int idx = 0, rc;
+
+	if (nvm->dir_type == BNXT_NVM_PORT_CFG)
+		idx = bp->pf.port_id;
+	else if (nvm->dir_type == BNXT_NVM_FUNC_CFG)
+		idx = bp->pf.fw_fid - BNXT_FIRST_PF_FID;
+
+	data = hwrm_req_dma_slice(bp, req, sizeof(*data), &data_dma_addr);
+
+	if (!data) {
+		hwrm_req_drop(bp, req);
+		return -ENOMEM;
+	}
+
+	req->dest_data_addr = cpu_to_le64(data_dma_addr);
+	req->data_len = cpu_to_le16(nvm->nvm_num_bits);
+	req->option_num = cpu_to_le16(nvm->offset);
+	req->index_0 = cpu_to_le16(idx);
+	if (idx)
+		req->dimensions = cpu_to_le16(1);
+
+	resp = hwrm_req_hold(bp, req);
+	if (req->req_type == cpu_to_le16(HWRM_NVM_SET_VARIABLE)) {
+		bnxt_copy_to_nvm_data(data, val, nvm->nvm_num_bits,
+				      nvm->dl_num_bytes);
+		rc = hwrm_req_send(bp, msg);
+	} else {
+		rc = hwrm_req_send_silent(bp, msg);
+		if (!rc) {
+			bnxt_copy_from_nvm_data(val, data,
+						nvm->nvm_num_bits,
+						nvm->dl_num_bytes);
+		} else {
+			if (resp->cmd_err ==
+				NVM_GET_VARIABLE_CMD_ERR_CODE_VAR_NOT_EXIST)
+				rc = -EOPNOTSUPP;
+		}
+	}
+	hwrm_req_drop(bp, req);
+	if (rc == -EACCES)
+		netdev_err(bp->dev, "PF does not have admin privileges to modify NVM config\n");
+	return rc;
+}
+
+static int bnxt_hwrm_nvm_req(struct bnxt *bp, u32 param_id, void *msg,
+			     union devlink_param_value *val)
+{
+	const struct bnxt_dl_nvm_param *nvm_param;
+	int i;
+
+	for (i = 0; i < ARRAY_SIZE(nvm_params); i++) {
+		nvm_param = &nvm_params[i];
+		if (nvm_param->id == param_id)
+			return __bnxt_hwrm_nvm_req(bp, nvm_param, msg, val);
+	}
+	return -EOPNOTSUPP;
+}
+
+static int bnxt_dl_nvm_param_get(struct devlink *dl, u32 id,
+				 struct devlink_param_gset_ctx *ctx)
+{
+	struct bnxt *bp = bnxt_get_bp_from_dl(dl);
+	struct hwrm_nvm_get_variable_input *req;
+	int rc;
+
+	rc = hwrm_req_init(bp, req, HWRM_NVM_GET_VARIABLE);
+	if (rc)
+		return rc;
+
+	rc = bnxt_hwrm_nvm_req(bp, id, req, &ctx->val);
+	if (!rc && id == BNXT_DEVLINK_PARAM_ID_GRE_VER_CHECK)
+		ctx->val.vbool = !ctx->val.vbool;
+
+	return rc;
+}
+
+static int bnxt_dl_nvm_param_set(struct devlink *dl, u32 id,
+				 struct devlink_param_gset_ctx *ctx,
+				 struct netlink_ext_ack *extack)
+{
+	struct bnxt *bp = bnxt_get_bp_from_dl(dl);
+	struct hwrm_nvm_set_variable_input *req;
+	int rc;
+
+	rc = hwrm_req_init(bp, req, HWRM_NVM_SET_VARIABLE);
+	if (rc)
+		return rc;
+
+	if (id == BNXT_DEVLINK_PARAM_ID_GRE_VER_CHECK)
+		ctx->val.vbool = !ctx->val.vbool;
+
+	return bnxt_hwrm_nvm_req(bp, id, req, &ctx->val);
+}
+
+static int bnxt_dl_roce_validate(struct devlink *dl, u32 id,
+				 union devlink_param_value val,
+				 struct netlink_ext_ack *extack)
+{
+	const struct bnxt_dl_nvm_param nvm_roce_cap = {0, NVM_OFF_RDMA_CAPABLE,
+		BNXT_NVM_SHARED_CFG, 1, 1};
+	struct bnxt *bp = bnxt_get_bp_from_dl(dl);
+	struct hwrm_nvm_get_variable_input *req;
+	union devlink_param_value roce_cap;
+	int rc;
+
+	rc = hwrm_req_init(bp, req, HWRM_NVM_GET_VARIABLE);
+	if (rc)
+		return rc;
+
+	if (__bnxt_hwrm_nvm_req(bp, &nvm_roce_cap, req, &roce_cap)) {
+		NL_SET_ERR_MSG_MOD(extack, "Unable to verify if device is RDMA Capable");
+		return -EINVAL;
+	}
+	if (!roce_cap.vbool) {
+		NL_SET_ERR_MSG_MOD(extack, "Device does not support RDMA");
+		return -EINVAL;
+	}
+	return 0;
+}
+
+static int bnxt_dl_msix_validate(struct devlink *dl, u32 id,
+				 union devlink_param_value val,
+				 struct netlink_ext_ack *extack)
+{
+	int max_val = -1;
+
+	if (id == DEVLINK_PARAM_GENERIC_ID_MSIX_VEC_PER_PF_MAX)
+		max_val = BNXT_MSIX_VEC_MAX;
+
+	if (id == DEVLINK_PARAM_GENERIC_ID_MSIX_VEC_PER_PF_MIN)
+		max_val = BNXT_MSIX_VEC_MIN_MAX;
+
+	if (val.vu32 > max_val) {
+		NL_SET_ERR_MSG_MOD(extack, "MSIX value is exceeding the range");
+		return -EINVAL;
+	}
+
+	return 0;
+}
+
+static int bnxt_remote_dev_reset_get(struct devlink *dl, u32 id,
+				     struct devlink_param_gset_ctx *ctx)
+{
+	struct bnxt *bp = bnxt_get_bp_from_dl(dl);
+
+	if (~bp->fw_cap & BNXT_FW_CAP_HOT_RESET_IF)
+		return -EOPNOTSUPP;
+
+	ctx->val.vbool = bnxt_dl_get_remote_reset(dl);
+	return 0;
+}
+
+static int bnxt_remote_dev_reset_set(struct devlink *dl, u32 id,
+				     struct devlink_param_gset_ctx *ctx,
+				     struct netlink_ext_ack *extack)
+{
+	struct bnxt *bp = bnxt_get_bp_from_dl(dl);
+	int rc;
+
+	rc = bnxt_hwrm_remote_dev_reset_set(bp, ctx->val.vbool);
+	if (rc)
+		return rc;
+
+	bnxt_dl_set_remote_reset(dl, ctx->val.vbool);
+	return rc;
+}
+
+static const struct devlink_param bnxt_dl_params[] = {
+	DEVLINK_PARAM_GENERIC(ENABLE_SRIOV,
+			      BIT(DEVLINK_PARAM_CMODE_PERMANENT),
+			      bnxt_dl_nvm_param_get, bnxt_dl_nvm_param_set,
+			      NULL),
+	DEVLINK_PARAM_GENERIC(IGNORE_ARI,
+			      BIT(DEVLINK_PARAM_CMODE_PERMANENT),
+			      bnxt_dl_nvm_param_get, bnxt_dl_nvm_param_set,
+			      NULL),
+	DEVLINK_PARAM_GENERIC(MSIX_VEC_PER_PF_MAX,
+			      BIT(DEVLINK_PARAM_CMODE_PERMANENT),
+			      bnxt_dl_nvm_param_get, bnxt_dl_nvm_param_set,
+			      bnxt_dl_msix_validate),
+	DEVLINK_PARAM_GENERIC(MSIX_VEC_PER_PF_MIN,
+			      BIT(DEVLINK_PARAM_CMODE_PERMANENT),
+			      bnxt_dl_nvm_param_get, bnxt_dl_nvm_param_set,
+			      bnxt_dl_msix_validate),
+	DEVLINK_PARAM_GENERIC(ENABLE_ROCE,
+			      BIT(DEVLINK_PARAM_CMODE_PERMANENT),
+			      bnxt_dl_nvm_param_get, bnxt_dl_nvm_param_set,
+			      bnxt_dl_roce_validate),
+	DEVLINK_PARAM_DRIVER(BNXT_DEVLINK_PARAM_ID_GRE_VER_CHECK,
+			     "gre_ver_check", DEVLINK_PARAM_TYPE_BOOL,
+			     BIT(DEVLINK_PARAM_CMODE_PERMANENT),
+			     bnxt_dl_nvm_param_get, bnxt_dl_nvm_param_set,
+			     NULL),
+	/* keep REMOTE_DEV_RESET last, it is excluded based on caps */
+	DEVLINK_PARAM_GENERIC(ENABLE_REMOTE_DEV_RESET,
+			      BIT(DEVLINK_PARAM_CMODE_RUNTIME),
+			      bnxt_remote_dev_reset_get,
+			      bnxt_remote_dev_reset_set, NULL),
+};
+
+static int bnxt_dl_params_register(struct bnxt *bp)
+{
+	int num_params = ARRAY_SIZE(bnxt_dl_params);
+	int rc;
+
+	if (bp->hwrm_spec_code < 0x10600)
+		return 0;
+
+	if (~bp->fw_cap & BNXT_FW_CAP_HOT_RESET_IF)
+		num_params--;
+
+	rc = devlink_params_register(bp->dl, bnxt_dl_params, num_params);
+	if (rc)
+		netdev_warn(bp->dev, "devlink_params_register failed. rc=%d\n",
+			    rc);
+	return rc;
+}
+
+static void bnxt_dl_params_unregister(struct bnxt *bp)
+{
+	int num_params = ARRAY_SIZE(bnxt_dl_params);
+
+	if (bp->hwrm_spec_code < 0x10600)
+		return;
+
+	if (~bp->fw_cap & BNXT_FW_CAP_HOT_RESET_IF)
+		num_params--;
+
+	devlink_params_unregister(bp->dl, bnxt_dl_params, num_params);
+}
+
+int bnxt_dl_register(struct bnxt *bp)
+{
+	const struct devlink_ops *devlink_ops;
+	struct devlink_port_attrs attrs = {};
+	struct bnxt_dl *bp_dl;
+	struct devlink *dl;
+	int rc;
+
+	if (BNXT_PF(bp))
+		devlink_ops = &bnxt_dl_ops;
+	else
+		devlink_ops = &bnxt_vf_dl_ops;
+
+	dl = devlink_alloc(devlink_ops, sizeof(struct bnxt_dl), &bp->pdev->dev);
+	if (!dl) {
+		netdev_warn(bp->dev, "devlink_alloc failed\n");
+		return -ENOMEM;
+	}
+
+	bp->dl = dl;
+	bp_dl = devlink_priv(dl);
+	bp_dl->bp = bp;
+	bnxt_dl_set_remote_reset(dl, true);
+
+	/* Add switchdev eswitch mode setting, if SRIOV supported */
+	if (pci_find_ext_capability(bp->pdev, PCI_EXT_CAP_ID_SRIOV) &&
+	    bp->hwrm_spec_code > 0x10803)
+		bp->eswitch_mode = DEVLINK_ESWITCH_MODE_LEGACY;
+
+	if (!BNXT_PF(bp))
+		goto out;
+
+	attrs.flavour = DEVLINK_PORT_FLAVOUR_PHYSICAL;
+	attrs.phys.port_number = bp->pf.port_id;
+	memcpy(attrs.switch_id.id, bp->dsn, sizeof(bp->dsn));
+	attrs.switch_id.id_len = sizeof(bp->dsn);
+	devlink_port_attrs_set(&bp->dl_port, &attrs);
+	rc = devlink_port_register(dl, &bp->dl_port, bp->pf.port_id);
+	if (rc) {
+		netdev_err(bp->dev, "devlink_port_register failed\n");
+		goto err_dl_free;
+	}
+
+	rc = bnxt_dl_params_register(bp);
+	if (rc)
+		goto err_dl_port_unreg;
+
+out:
+	devlink_register(dl);
+	return 0;
+
+err_dl_port_unreg:
+	devlink_port_unregister(&bp->dl_port);
+err_dl_free:
+	devlink_free(dl);
+	return rc;
+}
+
+void bnxt_dl_unregister(struct bnxt *bp)
+{
+	struct devlink *dl = bp->dl;
+
+	devlink_unregister(dl);
+	if (BNXT_PF(bp)) {
+		bnxt_dl_params_unregister(bp);
+		devlink_port_unregister(&bp->dl_port);
+	}
+	devlink_free(dl);
+}
diff --git a/drivers/net/ethernet/broadcom/bnxt/bnxt_devlink.h b/drivers/net/ethernet/broadcom/bnxt/bnxt_devlink.h
new file mode 100644
index 000000000000..7f45dcd7b287
--- /dev/null
+++ b/drivers/net/ethernet/broadcom/bnxt/bnxt_devlink.h
@@ -0,0 +1,84 @@
+/* Broadcom NetXtreme-C/E network driver.
+ *
+ * Copyright (c) 2017 Broadcom Limited
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation.
+ */
+
+#ifndef BNXT_DEVLINK_H
+#define BNXT_DEVLINK_H
+
+/* Struct to hold housekeeping info needed by devlink interface */
+struct bnxt_dl {
+	struct bnxt *bp;	/* back ptr to the controlling dev */
+	bool remote_reset;
+};
+
+static inline struct bnxt *bnxt_get_bp_from_dl(struct devlink *dl)
+{
+	return ((struct bnxt_dl *)devlink_priv(dl))->bp;
+}
+
+static inline void bnxt_dl_remote_reload(struct bnxt *bp)
+{
+	devlink_remote_reload_actions_performed(bp->dl, 0,
+						BIT(DEVLINK_RELOAD_ACTION_DRIVER_REINIT) |
+						BIT(DEVLINK_RELOAD_ACTION_FW_ACTIVATE));
+}
+
+static inline bool bnxt_dl_get_remote_reset(struct devlink *dl)
+{
+	return ((struct bnxt_dl *)devlink_priv(dl))->remote_reset;
+}
+
+static inline void bnxt_dl_set_remote_reset(struct devlink *dl, bool value)
+{
+	((struct bnxt_dl *)devlink_priv(dl))->remote_reset = value;
+}
+
+#define NVM_OFF_MSIX_VEC_PER_PF_MAX	108
+#define NVM_OFF_MSIX_VEC_PER_PF_MIN	114
+#define NVM_OFF_IGNORE_ARI		164
+#define NVM_OFF_RDMA_CAPABLE		161
+#define NVM_OFF_DIS_GRE_VER_CHECK	171
+#define NVM_OFF_ENABLE_SRIOV		401
+#define NVM_OFF_SUPPORT_RDMA		506
+#define NVM_OFF_NVM_CFG_VER		602
+
+#define BNXT_NVM_CFG_VER_BITS		8
+#define BNXT_NVM_CFG_VER_BYTES		1
+
+#define BNXT_MSIX_VEC_MAX	512
+#define BNXT_MSIX_VEC_MIN_MAX	128
+
+enum bnxt_nvm_dir_type {
+	BNXT_NVM_SHARED_CFG = 40,
+	BNXT_NVM_PORT_CFG,
+	BNXT_NVM_FUNC_CFG,
+};
+
+struct bnxt_dl_nvm_param {
+	u16 id;
+	u16 offset;
+	u16 dir_type;
+	u16 nvm_num_bits;
+	u8 dl_num_bytes;
+};
+
+enum bnxt_dl_version_type {
+	BNXT_VERSION_FIXED,
+	BNXT_VERSION_RUNNING,
+	BNXT_VERSION_STORED,
+};
+
+void bnxt_devlink_health_fw_report(struct bnxt *bp);
+void bnxt_dl_health_fw_status_update(struct bnxt *bp, bool healthy);
+void bnxt_dl_health_fw_recovery_done(struct bnxt *bp);
+void bnxt_dl_fw_reporters_create(struct bnxt *bp);
+void bnxt_dl_fw_reporters_destroy(struct bnxt *bp);
+int bnxt_dl_register(struct bnxt *bp);
+void bnxt_dl_unregister(struct bnxt *bp);
+
+#endif /* BNXT_DEVLINK_H */
diff --git a/drivers/net/ethernet/broadcom/bnxt/bnxt_dim.c b/drivers/net/ethernet/broadcom/bnxt/bnxt_dim.c
new file mode 100644
index 000000000000..53a3bcb0efe0
--- /dev/null
+++ b/drivers/net/ethernet/broadcom/bnxt/bnxt_dim.c
@@ -0,0 +1,31 @@
+/* Broadcom NetXtreme-C/E network driver.
+ *
+ * Copyright (c) 2017-2018 Broadcom Limited
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation.
+ */
+
+#include <linux/dim.h>
+#include <linux/bnxt/hsi.h>
+#include "bnxt.h"
+
+void bnxt_dim_work(struct work_struct *work)
+{
+	struct dim *dim = container_of(work, struct dim, work);
+	struct bnxt_cp_ring_info *cpr = container_of(dim,
+						     struct bnxt_cp_ring_info,
+						     dim);
+	struct bnxt_napi *bnapi = container_of(cpr,
+					       struct bnxt_napi,
+					       cp_ring);
+	struct dim_cq_moder cur_moder =
+		net_dim_get_rx_moderation(dim->mode, dim->profile_ix);
+
+	cpr->rx_ring_coal.coal_ticks = cur_moder.usec;
+	cpr->rx_ring_coal.coal_bufs = cur_moder.pkts;
+
+	bnxt_hwrm_set_ring_coal(bnapi->bp, bnapi);
+	dim->state = DIM_START_MEASURE;
+}
diff --git a/drivers/net/ethernet/broadcom/bnxt/bnxt_ethtool.c b/drivers/net/ethernet/broadcom/bnxt/bnxt_ethtool.c
index fd1181510b65..41686a6f84b5 100644
--- a/drivers/net/ethernet/broadcom/bnxt/bnxt_ethtool.c
+++ b/drivers/net/ethernet/broadcom/bnxt/bnxt_ethtool.c
@@ -8,25 +8,41 @@
  * the Free Software Foundation.
  */
 
+#include <linux/bitops.h>
 #include <linux/ctype.h>
 #include <linux/stringify.h>
 #include <linux/ethtool.h>
+#include <linux/ethtool_netlink.h>
+#include <linux/linkmode.h>
 #include <linux/interrupt.h>
 #include <linux/pci.h>
 #include <linux/etherdevice.h>
 #include <linux/crc32.h>
 #include <linux/firmware.h>
-#include "bnxt_hsi.h"
+#include <linux/utsname.h>
+#include <linux/time.h>
+#include <linux/ptp_clock_kernel.h>
+#include <linux/net_tstamp.h>
+#include <linux/timecounter.h>
+#include <net/netdev_queues.h>
+#include <net/netlink.h>
+#include <linux/bnxt/hsi.h>
 #include "bnxt.h"
+#include "bnxt_hwrm.h"
+#include "bnxt_ulp.h"
 #include "bnxt_xdp.h"
+#include "bnxt_ptp.h"
 #include "bnxt_ethtool.h"
 #include "bnxt_nvm_defs.h"	/* NVRAM content constant and structure defs */
 #include "bnxt_fw_hdr.h"	/* Firmware hdr constant and structure defs */
-#define FLASH_NVRAM_TIMEOUT	((HWRM_CMD_TIMEOUT) * 100)
-#define FLASH_PACKAGE_TIMEOUT	((HWRM_CMD_TIMEOUT) * 200)
-#define INSTALL_PACKAGE_TIMEOUT	((HWRM_CMD_TIMEOUT) * 200)
+#include "bnxt_coredump.h"
 
-static char *bnxt_get_pkgver(struct net_device *dev, char *buf, size_t buflen);
+#define BNXT_NVM_ERR_MSG(dev, extack, msg)			\
+	do {							\
+		if (extack)					\
+			NL_SET_ERR_MSG_MOD(extack, msg);	\
+		netdev_err(dev, "%s\n", msg);			\
+	} while (0)
 
 static u32 bnxt_get_msglevel(struct net_device *dev)
 {
@@ -43,22 +59,37 @@ static void bnxt_set_msglevel(struct net_device *dev, u32 value)
 }
 
 static int bnxt_get_coalesce(struct net_device *dev,
-			     struct ethtool_coalesce *coal)
+			     struct ethtool_coalesce *coal,
+			     struct kernel_ethtool_coalesce *kernel_coal,
+			     struct netlink_ext_ack *extack)
 {
 	struct bnxt *bp = netdev_priv(dev);
+	struct bnxt_coal *hw_coal;
+	u16 mult;
 
 	memset(coal, 0, sizeof(*coal));
 
-	coal->rx_coalesce_usecs = bp->rx_coal_ticks;
-	/* 2 completion records per rx packet */
-	coal->rx_max_coalesced_frames = bp->rx_coal_bufs / 2;
-	coal->rx_coalesce_usecs_irq = bp->rx_coal_ticks_irq;
-	coal->rx_max_coalesced_frames_irq = bp->rx_coal_bufs_irq / 2;
-
-	coal->tx_coalesce_usecs = bp->tx_coal_ticks;
-	coal->tx_max_coalesced_frames = bp->tx_coal_bufs;
-	coal->tx_coalesce_usecs_irq = bp->tx_coal_ticks_irq;
-	coal->tx_max_coalesced_frames_irq = bp->tx_coal_bufs_irq;
+	coal->use_adaptive_rx_coalesce = bp->flags & BNXT_FLAG_DIM;
+
+	hw_coal = &bp->rx_coal;
+	mult = hw_coal->bufs_per_record;
+	coal->rx_coalesce_usecs = hw_coal->coal_ticks;
+	coal->rx_max_coalesced_frames = hw_coal->coal_bufs / mult;
+	coal->rx_coalesce_usecs_irq = hw_coal->coal_ticks_irq;
+	coal->rx_max_coalesced_frames_irq = hw_coal->coal_bufs_irq / mult;
+	if (hw_coal->flags &
+	    RING_CMPL_RING_CFG_AGGINT_PARAMS_REQ_FLAGS_TIMER_RESET)
+		kernel_coal->use_cqe_mode_rx = true;
+
+	hw_coal = &bp->tx_coal;
+	mult = hw_coal->bufs_per_record;
+	coal->tx_coalesce_usecs = hw_coal->coal_ticks;
+	coal->tx_max_coalesced_frames = hw_coal->coal_bufs / mult;
+	coal->tx_coalesce_usecs_irq = hw_coal->coal_ticks_irq;
+	coal->tx_max_coalesced_frames_irq = hw_coal->coal_bufs_irq / mult;
+	if (hw_coal->flags &
+	    RING_CMPL_RING_CFG_AGGINT_PARAMS_REQ_FLAGS_TIMER_RESET)
+		kernel_coal->use_cqe_mode_tx = true;
 
 	coal->stats_block_coalesce_usecs = bp->stats_coal_ticks;
 
@@ -66,39 +97,77 @@ static int bnxt_get_coalesce(struct net_device *dev,
 }
 
 static int bnxt_set_coalesce(struct net_device *dev,
-			     struct ethtool_coalesce *coal)
+			     struct ethtool_coalesce *coal,
+			     struct kernel_ethtool_coalesce *kernel_coal,
+			     struct netlink_ext_ack *extack)
 {
 	struct bnxt *bp = netdev_priv(dev);
 	bool update_stats = false;
+	struct bnxt_coal *hw_coal;
 	int rc = 0;
+	u16 mult;
+
+	if (coal->use_adaptive_rx_coalesce) {
+		bp->flags |= BNXT_FLAG_DIM;
+	} else {
+		if (bp->flags & BNXT_FLAG_DIM) {
+			bp->flags &= ~(BNXT_FLAG_DIM);
+			goto reset_coalesce;
+		}
+	}
 
-	bp->rx_coal_ticks = coal->rx_coalesce_usecs;
-	/* 2 completion records per rx packet */
-	bp->rx_coal_bufs = coal->rx_max_coalesced_frames * 2;
-	bp->rx_coal_ticks_irq = coal->rx_coalesce_usecs_irq;
-	bp->rx_coal_bufs_irq = coal->rx_max_coalesced_frames_irq * 2;
+	if ((kernel_coal->use_cqe_mode_rx || kernel_coal->use_cqe_mode_tx) &&
+	    !(bp->coal_cap.cmpl_params &
+	      RING_AGGINT_QCAPS_RESP_CMPL_PARAMS_TIMER_RESET))
+		return -EOPNOTSUPP;
 
-	bp->tx_coal_ticks = coal->tx_coalesce_usecs;
-	bp->tx_coal_bufs = coal->tx_max_coalesced_frames;
-	bp->tx_coal_ticks_irq = coal->tx_coalesce_usecs_irq;
-	bp->tx_coal_bufs_irq = coal->tx_max_coalesced_frames_irq;
+	hw_coal = &bp->rx_coal;
+	mult = hw_coal->bufs_per_record;
+	hw_coal->coal_ticks = coal->rx_coalesce_usecs;
+	hw_coal->coal_bufs = coal->rx_max_coalesced_frames * mult;
+	hw_coal->coal_ticks_irq = coal->rx_coalesce_usecs_irq;
+	hw_coal->coal_bufs_irq = coal->rx_max_coalesced_frames_irq * mult;
+	hw_coal->flags &=
+		~RING_CMPL_RING_CFG_AGGINT_PARAMS_REQ_FLAGS_TIMER_RESET;
+	if (kernel_coal->use_cqe_mode_rx)
+		hw_coal->flags |=
+			RING_CMPL_RING_CFG_AGGINT_PARAMS_REQ_FLAGS_TIMER_RESET;
+
+	hw_coal = &bp->tx_coal;
+	mult = hw_coal->bufs_per_record;
+	hw_coal->coal_ticks = coal->tx_coalesce_usecs;
+	hw_coal->coal_bufs = coal->tx_max_coalesced_frames * mult;
+	hw_coal->coal_ticks_irq = coal->tx_coalesce_usecs_irq;
+	hw_coal->coal_bufs_irq = coal->tx_max_coalesced_frames_irq * mult;
+	hw_coal->flags &=
+		~RING_CMPL_RING_CFG_AGGINT_PARAMS_REQ_FLAGS_TIMER_RESET;
+	if (kernel_coal->use_cqe_mode_tx)
+		hw_coal->flags |=
+			RING_CMPL_RING_CFG_AGGINT_PARAMS_REQ_FLAGS_TIMER_RESET;
 
 	if (bp->stats_coal_ticks != coal->stats_block_coalesce_usecs) {
 		u32 stats_ticks = coal->stats_block_coalesce_usecs;
 
-		stats_ticks = clamp_t(u32, stats_ticks,
-				      BNXT_MIN_STATS_COAL_TICKS,
-				      BNXT_MAX_STATS_COAL_TICKS);
+		/* Allow 0, which means disable. */
+		if (stats_ticks)
+			stats_ticks = clamp_t(u32, stats_ticks,
+					      BNXT_MIN_STATS_COAL_TICKS,
+					      BNXT_MAX_STATS_COAL_TICKS);
 		stats_ticks = rounddown(stats_ticks, BNXT_MIN_STATS_COAL_TICKS);
 		bp->stats_coal_ticks = stats_ticks;
+		if (bp->stats_coal_ticks)
+			bp->current_interval =
+				bp->stats_coal_ticks * HZ / 1000000;
+		else
+			bp->current_interval = BNXT_TIMER_INTERVAL;
 		update_stats = true;
 	}
 
-	if (netif_running(dev)) {
+reset_coalesce:
+	if (test_bit(BNXT_STATE_OPEN, &bp->state)) {
 		if (update_stats) {
-			rc = bnxt_close_nic(bp, true, false);
-			if (!rc)
-				rc = bnxt_open_nic(bp, true, false);
+			bnxt_close_nic(bp, true, false);
+			rc = bnxt_open_nic(bp, true, false);
 		} else {
 			rc = bnxt_hwrm_set_coal(bp);
 		}
@@ -107,7 +176,53 @@ static int bnxt_set_coalesce(struct net_device *dev,
 	return rc;
 }
 
-#define BNXT_NUM_STATS	21
+static const char * const bnxt_ring_rx_stats_str[] = {
+	"rx_ucast_packets",
+	"rx_mcast_packets",
+	"rx_bcast_packets",
+	"rx_discards",
+	"rx_errors",
+	"rx_ucast_bytes",
+	"rx_mcast_bytes",
+	"rx_bcast_bytes",
+};
+
+static const char * const bnxt_ring_tx_stats_str[] = {
+	"tx_ucast_packets",
+	"tx_mcast_packets",
+	"tx_bcast_packets",
+	"tx_errors",
+	"tx_discards",
+	"tx_ucast_bytes",
+	"tx_mcast_bytes",
+	"tx_bcast_bytes",
+};
+
+static const char * const bnxt_ring_tpa_stats_str[] = {
+	"tpa_packets",
+	"tpa_bytes",
+	"tpa_events",
+	"tpa_aborts",
+};
+
+static const char * const bnxt_ring_tpa2_stats_str[] = {
+	"rx_tpa_eligible_pkt",
+	"rx_tpa_eligible_bytes",
+	"rx_tpa_pkt",
+	"rx_tpa_bytes",
+	"rx_tpa_errors",
+	"rx_tpa_events",
+};
+
+static const char * const bnxt_rx_sw_stats_str[] = {
+	"rx_l4_csum_errors",
+	"rx_resets",
+	"rx_buf_errors",
+};
+
+static const char * const bnxt_cmn_sw_stats_str[] = {
+	"missed_irqs",
+};
 
 #define BNXT_RX_STATS_ENTRY(counter)	\
 	{ BNXT_RX_STATS_OFFSET(counter), __stringify(counter) }
@@ -115,6 +230,133 @@ static int bnxt_set_coalesce(struct net_device *dev,
 #define BNXT_TX_STATS_ENTRY(counter)	\
 	{ BNXT_TX_STATS_OFFSET(counter), __stringify(counter) }
 
+#define BNXT_RX_STATS_EXT_ENTRY(counter)	\
+	{ BNXT_RX_STATS_EXT_OFFSET(counter), __stringify(counter) }
+
+#define BNXT_TX_STATS_EXT_ENTRY(counter)	\
+	{ BNXT_TX_STATS_EXT_OFFSET(counter), __stringify(counter) }
+
+#define BNXT_RX_STATS_EXT_PFC_ENTRY(n)				\
+	BNXT_RX_STATS_EXT_ENTRY(pfc_pri##n##_rx_duration_us),	\
+	BNXT_RX_STATS_EXT_ENTRY(pfc_pri##n##_rx_transitions)
+
+#define BNXT_TX_STATS_EXT_PFC_ENTRY(n)				\
+	BNXT_TX_STATS_EXT_ENTRY(pfc_pri##n##_tx_duration_us),	\
+	BNXT_TX_STATS_EXT_ENTRY(pfc_pri##n##_tx_transitions)
+
+#define BNXT_RX_STATS_EXT_PFC_ENTRIES				\
+	BNXT_RX_STATS_EXT_PFC_ENTRY(0),				\
+	BNXT_RX_STATS_EXT_PFC_ENTRY(1),				\
+	BNXT_RX_STATS_EXT_PFC_ENTRY(2),				\
+	BNXT_RX_STATS_EXT_PFC_ENTRY(3),				\
+	BNXT_RX_STATS_EXT_PFC_ENTRY(4),				\
+	BNXT_RX_STATS_EXT_PFC_ENTRY(5),				\
+	BNXT_RX_STATS_EXT_PFC_ENTRY(6),				\
+	BNXT_RX_STATS_EXT_PFC_ENTRY(7)
+
+#define BNXT_TX_STATS_EXT_PFC_ENTRIES				\
+	BNXT_TX_STATS_EXT_PFC_ENTRY(0),				\
+	BNXT_TX_STATS_EXT_PFC_ENTRY(1),				\
+	BNXT_TX_STATS_EXT_PFC_ENTRY(2),				\
+	BNXT_TX_STATS_EXT_PFC_ENTRY(3),				\
+	BNXT_TX_STATS_EXT_PFC_ENTRY(4),				\
+	BNXT_TX_STATS_EXT_PFC_ENTRY(5),				\
+	BNXT_TX_STATS_EXT_PFC_ENTRY(6),				\
+	BNXT_TX_STATS_EXT_PFC_ENTRY(7)
+
+#define BNXT_RX_STATS_EXT_COS_ENTRY(n)				\
+	BNXT_RX_STATS_EXT_ENTRY(rx_bytes_cos##n),		\
+	BNXT_RX_STATS_EXT_ENTRY(rx_packets_cos##n)
+
+#define BNXT_TX_STATS_EXT_COS_ENTRY(n)				\
+	BNXT_TX_STATS_EXT_ENTRY(tx_bytes_cos##n),		\
+	BNXT_TX_STATS_EXT_ENTRY(tx_packets_cos##n)
+
+#define BNXT_RX_STATS_EXT_COS_ENTRIES				\
+	BNXT_RX_STATS_EXT_COS_ENTRY(0),				\
+	BNXT_RX_STATS_EXT_COS_ENTRY(1),				\
+	BNXT_RX_STATS_EXT_COS_ENTRY(2),				\
+	BNXT_RX_STATS_EXT_COS_ENTRY(3),				\
+	BNXT_RX_STATS_EXT_COS_ENTRY(4),				\
+	BNXT_RX_STATS_EXT_COS_ENTRY(5),				\
+	BNXT_RX_STATS_EXT_COS_ENTRY(6),				\
+	BNXT_RX_STATS_EXT_COS_ENTRY(7)				\
+
+#define BNXT_TX_STATS_EXT_COS_ENTRIES				\
+	BNXT_TX_STATS_EXT_COS_ENTRY(0),				\
+	BNXT_TX_STATS_EXT_COS_ENTRY(1),				\
+	BNXT_TX_STATS_EXT_COS_ENTRY(2),				\
+	BNXT_TX_STATS_EXT_COS_ENTRY(3),				\
+	BNXT_TX_STATS_EXT_COS_ENTRY(4),				\
+	BNXT_TX_STATS_EXT_COS_ENTRY(5),				\
+	BNXT_TX_STATS_EXT_COS_ENTRY(6),				\
+	BNXT_TX_STATS_EXT_COS_ENTRY(7)				\
+
+#define BNXT_RX_STATS_EXT_DISCARD_COS_ENTRY(n)			\
+	BNXT_RX_STATS_EXT_ENTRY(rx_discard_bytes_cos##n),	\
+	BNXT_RX_STATS_EXT_ENTRY(rx_discard_packets_cos##n)
+
+#define BNXT_RX_STATS_EXT_DISCARD_COS_ENTRIES				\
+	BNXT_RX_STATS_EXT_DISCARD_COS_ENTRY(0),				\
+	BNXT_RX_STATS_EXT_DISCARD_COS_ENTRY(1),				\
+	BNXT_RX_STATS_EXT_DISCARD_COS_ENTRY(2),				\
+	BNXT_RX_STATS_EXT_DISCARD_COS_ENTRY(3),				\
+	BNXT_RX_STATS_EXT_DISCARD_COS_ENTRY(4),				\
+	BNXT_RX_STATS_EXT_DISCARD_COS_ENTRY(5),				\
+	BNXT_RX_STATS_EXT_DISCARD_COS_ENTRY(6),				\
+	BNXT_RX_STATS_EXT_DISCARD_COS_ENTRY(7)
+
+#define BNXT_RX_STATS_PRI_ENTRY(counter, n)		\
+	{ BNXT_RX_STATS_EXT_OFFSET(counter##_cos0),	\
+	  __stringify(counter##_pri##n) }
+
+#define BNXT_TX_STATS_PRI_ENTRY(counter, n)		\
+	{ BNXT_TX_STATS_EXT_OFFSET(counter##_cos0),	\
+	  __stringify(counter##_pri##n) }
+
+#define BNXT_RX_STATS_PRI_ENTRIES(counter)		\
+	BNXT_RX_STATS_PRI_ENTRY(counter, 0),		\
+	BNXT_RX_STATS_PRI_ENTRY(counter, 1),		\
+	BNXT_RX_STATS_PRI_ENTRY(counter, 2),		\
+	BNXT_RX_STATS_PRI_ENTRY(counter, 3),		\
+	BNXT_RX_STATS_PRI_ENTRY(counter, 4),		\
+	BNXT_RX_STATS_PRI_ENTRY(counter, 5),		\
+	BNXT_RX_STATS_PRI_ENTRY(counter, 6),		\
+	BNXT_RX_STATS_PRI_ENTRY(counter, 7)
+
+#define BNXT_TX_STATS_PRI_ENTRIES(counter)		\
+	BNXT_TX_STATS_PRI_ENTRY(counter, 0),		\
+	BNXT_TX_STATS_PRI_ENTRY(counter, 1),		\
+	BNXT_TX_STATS_PRI_ENTRY(counter, 2),		\
+	BNXT_TX_STATS_PRI_ENTRY(counter, 3),		\
+	BNXT_TX_STATS_PRI_ENTRY(counter, 4),		\
+	BNXT_TX_STATS_PRI_ENTRY(counter, 5),		\
+	BNXT_TX_STATS_PRI_ENTRY(counter, 6),		\
+	BNXT_TX_STATS_PRI_ENTRY(counter, 7)
+
+enum {
+	RX_TOTAL_DISCARDS,
+	TX_TOTAL_DISCARDS,
+	RX_NETPOLL_DISCARDS,
+};
+
+static const char *const bnxt_ring_err_stats_arr[] = {
+	"rx_total_l4_csum_errors",
+	"rx_total_resets",
+	"rx_total_buf_errors",
+	"rx_total_oom_discards",
+	"rx_total_netpoll_discards",
+	"rx_total_ring_discards",
+	"tx_total_resets",
+	"tx_total_ring_discards",
+	"total_missed_irqs",
+};
+
+#define NUM_RING_RX_SW_STATS		ARRAY_SIZE(bnxt_rx_sw_stats_str)
+#define NUM_RING_CMN_SW_STATS		ARRAY_SIZE(bnxt_cmn_sw_stats_str)
+#define NUM_RING_RX_HW_STATS		ARRAY_SIZE(bnxt_ring_rx_stats_str)
+#define NUM_RING_TX_HW_STATS		ARRAY_SIZE(bnxt_ring_tx_stats_str)
+
 static const struct {
 	long offset;
 	char string[ETH_GSTRING_LEN];
@@ -124,7 +366,7 @@ static const struct {
 	BNXT_RX_STATS_ENTRY(rx_128b_255b_frames),
 	BNXT_RX_STATS_ENTRY(rx_256b_511b_frames),
 	BNXT_RX_STATS_ENTRY(rx_512b_1023b_frames),
-	BNXT_RX_STATS_ENTRY(rx_1024b_1518_frames),
+	BNXT_RX_STATS_ENTRY(rx_1024b_1518b_frames),
 	BNXT_RX_STATS_ENTRY(rx_good_vlan_frames),
 	BNXT_RX_STATS_ENTRY(rx_1519b_2047b_frames),
 	BNXT_RX_STATS_ENTRY(rx_2048b_4095b_frames),
@@ -159,15 +401,17 @@ static const struct {
 	BNXT_RX_STATS_ENTRY(rx_bytes),
 	BNXT_RX_STATS_ENTRY(rx_runt_bytes),
 	BNXT_RX_STATS_ENTRY(rx_runt_frames),
+	BNXT_RX_STATS_ENTRY(rx_stat_discard),
+	BNXT_RX_STATS_ENTRY(rx_stat_err),
 
 	BNXT_TX_STATS_ENTRY(tx_64b_frames),
 	BNXT_TX_STATS_ENTRY(tx_65b_127b_frames),
 	BNXT_TX_STATS_ENTRY(tx_128b_255b_frames),
 	BNXT_TX_STATS_ENTRY(tx_256b_511b_frames),
 	BNXT_TX_STATS_ENTRY(tx_512b_1023b_frames),
-	BNXT_TX_STATS_ENTRY(tx_1024b_1518_frames),
+	BNXT_TX_STATS_ENTRY(tx_1024b_1518b_frames),
 	BNXT_TX_STATS_ENTRY(tx_good_vlan_frames),
-	BNXT_TX_STATS_ENTRY(tx_1519b_2047_frames),
+	BNXT_TX_STATS_ENTRY(tx_1519b_2047b_frames),
 	BNXT_TX_STATS_ENTRY(tx_2048b_4095b_frames),
 	BNXT_TX_STATS_ENTRY(tx_4096b_9216b_frames),
 	BNXT_TX_STATS_ENTRY(tx_9217b_16383b_frames),
@@ -194,23 +438,133 @@ static const struct {
 	BNXT_TX_STATS_ENTRY(tx_eee_lpi_duration),
 	BNXT_TX_STATS_ENTRY(tx_total_collisions),
 	BNXT_TX_STATS_ENTRY(tx_bytes),
+	BNXT_TX_STATS_ENTRY(tx_xthol_frames),
+	BNXT_TX_STATS_ENTRY(tx_stat_discard),
+	BNXT_TX_STATS_ENTRY(tx_stat_error),
+};
+
+static const struct {
+	long offset;
+	char string[ETH_GSTRING_LEN];
+} bnxt_port_stats_ext_arr[] = {
+	BNXT_RX_STATS_EXT_ENTRY(link_down_events),
+	BNXT_RX_STATS_EXT_ENTRY(continuous_pause_events),
+	BNXT_RX_STATS_EXT_ENTRY(resume_pause_events),
+	BNXT_RX_STATS_EXT_ENTRY(continuous_roce_pause_events),
+	BNXT_RX_STATS_EXT_ENTRY(resume_roce_pause_events),
+	BNXT_RX_STATS_EXT_COS_ENTRIES,
+	BNXT_RX_STATS_EXT_PFC_ENTRIES,
+	BNXT_RX_STATS_EXT_ENTRY(rx_bits),
+	BNXT_RX_STATS_EXT_ENTRY(rx_buffer_passed_threshold),
+	BNXT_RX_STATS_EXT_ENTRY(rx_pcs_symbol_err),
+	BNXT_RX_STATS_EXT_ENTRY(rx_corrected_bits),
+	BNXT_RX_STATS_EXT_DISCARD_COS_ENTRIES,
+	BNXT_RX_STATS_EXT_ENTRY(rx_fec_corrected_blocks),
+	BNXT_RX_STATS_EXT_ENTRY(rx_fec_uncorrectable_blocks),
+	BNXT_RX_STATS_EXT_ENTRY(rx_filter_miss),
+};
+
+static const struct {
+	long offset;
+	char string[ETH_GSTRING_LEN];
+} bnxt_tx_port_stats_ext_arr[] = {
+	BNXT_TX_STATS_EXT_COS_ENTRIES,
+	BNXT_TX_STATS_EXT_PFC_ENTRIES,
+};
+
+static const struct {
+	long base_off;
+	char string[ETH_GSTRING_LEN];
+} bnxt_rx_bytes_pri_arr[] = {
+	BNXT_RX_STATS_PRI_ENTRIES(rx_bytes),
+};
+
+static const struct {
+	long base_off;
+	char string[ETH_GSTRING_LEN];
+} bnxt_rx_pkts_pri_arr[] = {
+	BNXT_RX_STATS_PRI_ENTRIES(rx_packets),
+};
+
+static const struct {
+	long base_off;
+	char string[ETH_GSTRING_LEN];
+} bnxt_tx_bytes_pri_arr[] = {
+	BNXT_TX_STATS_PRI_ENTRIES(tx_bytes),
 };
 
+static const struct {
+	long base_off;
+	char string[ETH_GSTRING_LEN];
+} bnxt_tx_pkts_pri_arr[] = {
+	BNXT_TX_STATS_PRI_ENTRIES(tx_packets),
+};
+
+#define BNXT_NUM_RING_ERR_STATS	ARRAY_SIZE(bnxt_ring_err_stats_arr)
 #define BNXT_NUM_PORT_STATS ARRAY_SIZE(bnxt_port_stats_arr)
+#define BNXT_NUM_STATS_PRI			\
+	(ARRAY_SIZE(bnxt_rx_bytes_pri_arr) +	\
+	 ARRAY_SIZE(bnxt_rx_pkts_pri_arr) +	\
+	 ARRAY_SIZE(bnxt_tx_bytes_pri_arr) +	\
+	 ARRAY_SIZE(bnxt_tx_pkts_pri_arr))
+
+static int bnxt_get_num_tpa_ring_stats(struct bnxt *bp)
+{
+	if (BNXT_SUPPORTS_TPA(bp)) {
+		if (bp->max_tpa_v2) {
+			if (BNXT_CHIP_P5(bp))
+				return BNXT_NUM_TPA_RING_STATS_P5;
+			return BNXT_NUM_TPA_RING_STATS_P7;
+		}
+		return BNXT_NUM_TPA_RING_STATS;
+	}
+	return 0;
+}
+
+static int bnxt_get_num_ring_stats(struct bnxt *bp)
+{
+	int rx, tx, cmn;
+
+	rx = NUM_RING_RX_HW_STATS + NUM_RING_RX_SW_STATS +
+	     bnxt_get_num_tpa_ring_stats(bp);
+	tx = NUM_RING_TX_HW_STATS;
+	cmn = NUM_RING_CMN_SW_STATS;
+	return rx * bp->rx_nr_rings +
+	       tx * (bp->tx_nr_rings_xdp + bp->tx_nr_rings_per_tc) +
+	       cmn * bp->cp_nr_rings;
+}
+
+static int bnxt_get_num_stats(struct bnxt *bp)
+{
+	int num_stats = bnxt_get_num_ring_stats(bp);
+	int len;
+
+	num_stats += BNXT_NUM_RING_ERR_STATS;
+
+	if (bp->flags & BNXT_FLAG_PORT_STATS)
+		num_stats += BNXT_NUM_PORT_STATS;
+
+	if (bp->flags & BNXT_FLAG_PORT_STATS_EXT) {
+		len = min_t(int, bp->fw_rx_stats_ext_size,
+			    ARRAY_SIZE(bnxt_port_stats_ext_arr));
+		num_stats += len;
+		len = min_t(int, bp->fw_tx_stats_ext_size,
+			    ARRAY_SIZE(bnxt_tx_port_stats_ext_arr));
+		num_stats += len;
+		if (bp->pri2cos_valid)
+			num_stats += BNXT_NUM_STATS_PRI;
+	}
+
+	return num_stats;
+}
 
 static int bnxt_get_sset_count(struct net_device *dev, int sset)
 {
 	struct bnxt *bp = netdev_priv(dev);
 
 	switch (sset) {
-	case ETH_SS_STATS: {
-		int num_stats = BNXT_NUM_STATS * bp->cp_nr_rings;
-
-		if (bp->flags & BNXT_FLAG_PORT_STATS)
-			num_stats += BNXT_NUM_PORT_STATS;
-
-		return num_stats;
-	}
+	case ETH_SS_STATS:
+		return bnxt_get_num_stats(bp);
 	case ETH_SS_TEST:
 		if (!bp->num_tests)
 			return -EOPNOTSUPP;
@@ -220,35 +574,131 @@ static int bnxt_get_sset_count(struct net_device *dev, int sset)
 	}
 }
 
+static bool is_rx_ring(struct bnxt *bp, int ring_num)
+{
+	return ring_num < bp->rx_nr_rings;
+}
+
+static bool is_tx_ring(struct bnxt *bp, int ring_num)
+{
+	int tx_base = 0;
+
+	if (!(bp->flags & BNXT_FLAG_SHARED_RINGS))
+		tx_base = bp->rx_nr_rings;
+
+	if (ring_num >= tx_base && ring_num < (tx_base + bp->tx_nr_rings))
+		return true;
+	return false;
+}
+
 static void bnxt_get_ethtool_stats(struct net_device *dev,
 				   struct ethtool_stats *stats, u64 *buf)
 {
-	u32 i, j = 0;
+	struct bnxt_total_ring_err_stats ring_err_stats = {0};
 	struct bnxt *bp = netdev_priv(dev);
-	u32 buf_size = sizeof(struct ctx_hw_stats) * bp->cp_nr_rings;
-	u32 stat_fields = sizeof(struct ctx_hw_stats) / 8;
-
-	memset(buf, 0, buf_size);
+	u64 *curr, *prev;
+	u32 tpa_stats;
+	u32 i, j = 0;
 
-	if (!bp->bnapi)
-		return;
+	if (!bp->bnapi) {
+		j += bnxt_get_num_ring_stats(bp);
+		goto skip_ring_stats;
+	}
 
+	tpa_stats = bnxt_get_num_tpa_ring_stats(bp);
 	for (i = 0; i < bp->cp_nr_rings; i++) {
 		struct bnxt_napi *bnapi = bp->bnapi[i];
 		struct bnxt_cp_ring_info *cpr = &bnapi->cp_ring;
-		__le64 *hw_stats = (__le64 *)cpr->hw_stats;
+		u64 *sw_stats = cpr->stats.sw_stats;
+		u64 *sw;
 		int k;
 
-		for (k = 0; k < stat_fields; j++, k++)
-			buf[j] = le64_to_cpu(hw_stats[k]);
-		buf[j++] = cpr->rx_l4_csum_errors;
+		if (is_rx_ring(bp, i)) {
+			for (k = 0; k < NUM_RING_RX_HW_STATS; j++, k++)
+				buf[j] = sw_stats[k];
+		}
+		if (is_tx_ring(bp, i)) {
+			k = NUM_RING_RX_HW_STATS;
+			for (; k < NUM_RING_RX_HW_STATS + NUM_RING_TX_HW_STATS;
+			       j++, k++)
+				buf[j] = sw_stats[k];
+		}
+		if (!tpa_stats || !is_rx_ring(bp, i))
+			goto skip_tpa_ring_stats;
+
+		k = NUM_RING_RX_HW_STATS + NUM_RING_TX_HW_STATS;
+		for (; k < NUM_RING_RX_HW_STATS + NUM_RING_TX_HW_STATS +
+			   tpa_stats; j++, k++)
+			buf[j] = sw_stats[k];
+
+skip_tpa_ring_stats:
+		sw = (u64 *)&cpr->sw_stats->rx;
+		if (is_rx_ring(bp, i)) {
+			for (k = 0; k < NUM_RING_RX_SW_STATS; j++, k++)
+				buf[j] = sw[k];
+		}
+
+		sw = (u64 *)&cpr->sw_stats->cmn;
+		for (k = 0; k < NUM_RING_CMN_SW_STATS; j++, k++)
+			buf[j] = sw[k];
 	}
+
+	bnxt_get_ring_err_stats(bp, &ring_err_stats);
+
+skip_ring_stats:
+	curr = &ring_err_stats.rx_total_l4_csum_errors;
+	prev = &bp->ring_err_stats_prev.rx_total_l4_csum_errors;
+	for (i = 0; i < BNXT_NUM_RING_ERR_STATS; i++, j++, curr++, prev++)
+		buf[j] = *curr + *prev;
+
 	if (bp->flags & BNXT_FLAG_PORT_STATS) {
-		__le64 *port_stats = (__le64 *)bp->hw_rx_port_stats;
+		u64 *port_stats = bp->port_stats.sw_stats;
 
-		for (i = 0; i < BNXT_NUM_PORT_STATS; i++, j++) {
-			buf[j] = le64_to_cpu(*(port_stats +
-					       bnxt_port_stats_arr[i].offset));
+		for (i = 0; i < BNXT_NUM_PORT_STATS; i++, j++)
+			buf[j] = *(port_stats + bnxt_port_stats_arr[i].offset);
+	}
+	if (bp->flags & BNXT_FLAG_PORT_STATS_EXT) {
+		u64 *rx_port_stats_ext = bp->rx_port_stats_ext.sw_stats;
+		u64 *tx_port_stats_ext = bp->tx_port_stats_ext.sw_stats;
+		u32 len;
+
+		len = min_t(u32, bp->fw_rx_stats_ext_size,
+			    ARRAY_SIZE(bnxt_port_stats_ext_arr));
+		for (i = 0; i < len; i++, j++) {
+			buf[j] = *(rx_port_stats_ext +
+				   bnxt_port_stats_ext_arr[i].offset);
+		}
+		len = min_t(u32, bp->fw_tx_stats_ext_size,
+			    ARRAY_SIZE(bnxt_tx_port_stats_ext_arr));
+		for (i = 0; i < len; i++, j++) {
+			buf[j] = *(tx_port_stats_ext +
+				   bnxt_tx_port_stats_ext_arr[i].offset);
+		}
+		if (bp->pri2cos_valid) {
+			for (i = 0; i < 8; i++, j++) {
+				long n = bnxt_rx_bytes_pri_arr[i].base_off +
+					 bp->pri2cos_idx[i];
+
+				buf[j] = *(rx_port_stats_ext + n);
+			}
+			for (i = 0; i < 8; i++, j++) {
+				long n = bnxt_rx_pkts_pri_arr[i].base_off +
+					 bp->pri2cos_idx[i];
+
+				buf[j] = *(rx_port_stats_ext + n);
+			}
+			for (i = 0; i < 8; i++, j++) {
+				long n = bnxt_tx_bytes_pri_arr[i].base_off +
+					 bp->pri2cos_idx[i];
+
+				buf[j] = *(tx_port_stats_ext + n);
+			}
+			for (i = 0; i < 8; i++, j++) {
+				long n = bnxt_tx_pkts_pri_arr[i].base_off +
+					 bp->pri2cos_idx[i];
+
+				buf[j] = *(tx_port_stats_ext + n);
+			}
 		}
 	}
 }
@@ -256,66 +706,105 @@ static void bnxt_get_ethtool_stats(struct net_device *dev,
 static void bnxt_get_strings(struct net_device *dev, u32 stringset, u8 *buf)
 {
 	struct bnxt *bp = netdev_priv(dev);
-	u32 i;
+	u32 i, j, num_str;
+	const char *str;
 
 	switch (stringset) {
-	/* The number of strings must match BNXT_NUM_STATS defined above. */
 	case ETH_SS_STATS:
 		for (i = 0; i < bp->cp_nr_rings; i++) {
-			sprintf(buf, "[%d]: rx_ucast_packets", i);
-			buf += ETH_GSTRING_LEN;
-			sprintf(buf, "[%d]: rx_mcast_packets", i);
-			buf += ETH_GSTRING_LEN;
-			sprintf(buf, "[%d]: rx_bcast_packets", i);
-			buf += ETH_GSTRING_LEN;
-			sprintf(buf, "[%d]: rx_discards", i);
-			buf += ETH_GSTRING_LEN;
-			sprintf(buf, "[%d]: rx_drops", i);
-			buf += ETH_GSTRING_LEN;
-			sprintf(buf, "[%d]: rx_ucast_bytes", i);
-			buf += ETH_GSTRING_LEN;
-			sprintf(buf, "[%d]: rx_mcast_bytes", i);
-			buf += ETH_GSTRING_LEN;
-			sprintf(buf, "[%d]: rx_bcast_bytes", i);
-			buf += ETH_GSTRING_LEN;
-			sprintf(buf, "[%d]: tx_ucast_packets", i);
-			buf += ETH_GSTRING_LEN;
-			sprintf(buf, "[%d]: tx_mcast_packets", i);
-			buf += ETH_GSTRING_LEN;
-			sprintf(buf, "[%d]: tx_bcast_packets", i);
-			buf += ETH_GSTRING_LEN;
-			sprintf(buf, "[%d]: tx_discards", i);
-			buf += ETH_GSTRING_LEN;
-			sprintf(buf, "[%d]: tx_drops", i);
-			buf += ETH_GSTRING_LEN;
-			sprintf(buf, "[%d]: tx_ucast_bytes", i);
-			buf += ETH_GSTRING_LEN;
-			sprintf(buf, "[%d]: tx_mcast_bytes", i);
-			buf += ETH_GSTRING_LEN;
-			sprintf(buf, "[%d]: tx_bcast_bytes", i);
-			buf += ETH_GSTRING_LEN;
-			sprintf(buf, "[%d]: tpa_packets", i);
-			buf += ETH_GSTRING_LEN;
-			sprintf(buf, "[%d]: tpa_bytes", i);
-			buf += ETH_GSTRING_LEN;
-			sprintf(buf, "[%d]: tpa_events", i);
-			buf += ETH_GSTRING_LEN;
-			sprintf(buf, "[%d]: tpa_aborts", i);
-			buf += ETH_GSTRING_LEN;
-			sprintf(buf, "[%d]: rx_l4_csum_errors", i);
-			buf += ETH_GSTRING_LEN;
-		}
-		if (bp->flags & BNXT_FLAG_PORT_STATS) {
+			if (is_rx_ring(bp, i))
+				for (j = 0; j < NUM_RING_RX_HW_STATS; j++) {
+					str = bnxt_ring_rx_stats_str[j];
+					ethtool_sprintf(&buf, "[%d]: %s", i,
+							str);
+				}
+			if (is_tx_ring(bp, i))
+				for (j = 0; j < NUM_RING_TX_HW_STATS; j++) {
+					str = bnxt_ring_tx_stats_str[j];
+					ethtool_sprintf(&buf, "[%d]: %s", i,
+							str);
+				}
+			num_str = bnxt_get_num_tpa_ring_stats(bp);
+			if (!num_str || !is_rx_ring(bp, i))
+				goto skip_tpa_stats;
+
+			if (bp->max_tpa_v2)
+				for (j = 0; j < num_str; j++) {
+					str = bnxt_ring_tpa2_stats_str[j];
+					ethtool_sprintf(&buf, "[%d]: %s", i,
+							str);
+				}
+			else
+				for (j = 0; j < num_str; j++) {
+					str = bnxt_ring_tpa_stats_str[j];
+					ethtool_sprintf(&buf, "[%d]: %s", i,
+							str);
+				}
+skip_tpa_stats:
+			if (is_rx_ring(bp, i))
+				for (j = 0; j < NUM_RING_RX_SW_STATS; j++) {
+					str = bnxt_rx_sw_stats_str[j];
+					ethtool_sprintf(&buf, "[%d]: %s", i,
+							str);
+				}
+			for (j = 0; j < NUM_RING_CMN_SW_STATS; j++) {
+				str = bnxt_cmn_sw_stats_str[j];
+				ethtool_sprintf(&buf, "[%d]: %s", i, str);
+			}
+		}
+		for (i = 0; i < BNXT_NUM_RING_ERR_STATS; i++)
+			ethtool_puts(&buf, bnxt_ring_err_stats_arr[i]);
+
+		if (bp->flags & BNXT_FLAG_PORT_STATS)
 			for (i = 0; i < BNXT_NUM_PORT_STATS; i++) {
-				strcpy(buf, bnxt_port_stats_arr[i].string);
-				buf += ETH_GSTRING_LEN;
+				str = bnxt_port_stats_arr[i].string;
+				ethtool_puts(&buf, str);
+			}
+
+		if (bp->flags & BNXT_FLAG_PORT_STATS_EXT) {
+			u32 len;
+
+			len = min_t(u32, bp->fw_rx_stats_ext_size,
+				    ARRAY_SIZE(bnxt_port_stats_ext_arr));
+			for (i = 0; i < len; i++) {
+				str = bnxt_port_stats_ext_arr[i].string;
+				ethtool_puts(&buf, str);
+			}
+
+			len = min_t(u32, bp->fw_tx_stats_ext_size,
+				    ARRAY_SIZE(bnxt_tx_port_stats_ext_arr));
+			for (i = 0; i < len; i++) {
+				str = bnxt_tx_port_stats_ext_arr[i].string;
+				ethtool_puts(&buf, str);
+			}
+
+			if (bp->pri2cos_valid) {
+				for (i = 0; i < 8; i++) {
+					str = bnxt_rx_bytes_pri_arr[i].string;
+					ethtool_puts(&buf, str);
+				}
+
+				for (i = 0; i < 8; i++) {
+					str = bnxt_rx_pkts_pri_arr[i].string;
+					ethtool_puts(&buf, str);
+				}
+
+				for (i = 0; i < 8; i++) {
+					str = bnxt_tx_bytes_pri_arr[i].string;
+					ethtool_puts(&buf, str);
+				}
+
+				for (i = 0; i < 8; i++) {
+					str = bnxt_tx_pkts_pri_arr[i].string;
+					ethtool_puts(&buf, str);
+				}
 			}
 		}
 		break;
 	case ETH_SS_TEST:
 		if (bp->num_tests)
-			memcpy(buf, bp->test_info->string,
-			       bp->num_tests * ETH_GSTRING_LEN);
+			for (i = 0; i < bp->num_tests; i++)
+				ethtool_puts(&buf, bp->test_info->string[i]);
 		break;
 	default:
 		netdev_err(bp->dev, "bnxt_get_strings invalid request %x\n",
@@ -325,32 +814,64 @@ static void bnxt_get_strings(struct net_device *dev, u32 stringset, u8 *buf)
 }
 
 static void bnxt_get_ringparam(struct net_device *dev,
-			       struct ethtool_ringparam *ering)
+			       struct ethtool_ringparam *ering,
+			       struct kernel_ethtool_ringparam *kernel_ering,
+			       struct netlink_ext_ack *extack)
 {
 	struct bnxt *bp = netdev_priv(dev);
 
-	ering->rx_max_pending = BNXT_MAX_RX_DESC_CNT;
-	ering->rx_jumbo_max_pending = BNXT_MAX_RX_JUM_DESC_CNT;
+	if (bp->flags & BNXT_FLAG_AGG_RINGS) {
+		ering->rx_max_pending = BNXT_MAX_RX_DESC_CNT_JUM_ENA;
+		ering->rx_jumbo_max_pending = BNXT_MAX_RX_JUM_DESC_CNT;
+		kernel_ering->tcp_data_split = ETHTOOL_TCP_DATA_SPLIT_ENABLED;
+	} else {
+		ering->rx_max_pending = BNXT_MAX_RX_DESC_CNT;
+		ering->rx_jumbo_max_pending = 0;
+		kernel_ering->tcp_data_split = ETHTOOL_TCP_DATA_SPLIT_DISABLED;
+	}
 	ering->tx_max_pending = BNXT_MAX_TX_DESC_CNT;
 
 	ering->rx_pending = bp->rx_ring_size;
 	ering->rx_jumbo_pending = bp->rx_agg_ring_size;
 	ering->tx_pending = bp->tx_ring_size;
+
+	kernel_ering->hds_thresh_max = BNXT_HDS_THRESHOLD_MAX;
 }
 
 static int bnxt_set_ringparam(struct net_device *dev,
-			      struct ethtool_ringparam *ering)
+			      struct ethtool_ringparam *ering,
+			      struct kernel_ethtool_ringparam *kernel_ering,
+			      struct netlink_ext_ack *extack)
 {
+	u8 tcp_data_split = kernel_ering->tcp_data_split;
 	struct bnxt *bp = netdev_priv(dev);
+	u8 hds_config_mod;
 
 	if ((ering->rx_pending > BNXT_MAX_RX_DESC_CNT) ||
 	    (ering->tx_pending > BNXT_MAX_TX_DESC_CNT) ||
-	    (ering->tx_pending <= MAX_SKB_FRAGS))
+	    (ering->tx_pending < BNXT_MIN_TX_DESC_CNT))
+		return -EINVAL;
+
+	hds_config_mod = tcp_data_split != dev->cfg->hds_config;
+	if (tcp_data_split == ETHTOOL_TCP_DATA_SPLIT_DISABLED && hds_config_mod)
 		return -EINVAL;
 
+	if (tcp_data_split == ETHTOOL_TCP_DATA_SPLIT_ENABLED &&
+	    hds_config_mod && BNXT_RX_PAGE_MODE(bp)) {
+		NL_SET_ERR_MSG_MOD(extack, "tcp-data-split is disallowed when XDP is attached");
+		return -EINVAL;
+	}
+
 	if (netif_running(dev))
 		bnxt_close_nic(bp, false, false);
 
+	if (hds_config_mod) {
+		if (tcp_data_split == ETHTOOL_TCP_DATA_SPLIT_ENABLED)
+			bp->flags |= BNXT_FLAG_HDS;
+		else if (tcp_data_split == ETHTOOL_TCP_DATA_SPLIT_UNKNOWN)
+			bp->flags &= ~BNXT_FLAG_HDS;
+	}
+
 	bp->rx_ring_size = ering->rx_pending;
 	bp->tx_ring_size = ering->tx_pending;
 	bnxt_set_ring_params(bp);
@@ -365,17 +886,33 @@ static void bnxt_get_channels(struct net_device *dev,
 			      struct ethtool_channels *channel)
 {
 	struct bnxt *bp = netdev_priv(dev);
+	struct bnxt_hw_resc *hw_resc = &bp->hw_resc;
 	int max_rx_rings, max_tx_rings, tcs;
+	int max_tx_sch_inputs, tx_grps;
+
+	/* Get the most up-to-date max_tx_sch_inputs. */
+	if (netif_running(dev) && BNXT_NEW_RM(bp))
+		bnxt_hwrm_func_resc_qcaps(bp, false);
+	max_tx_sch_inputs = hw_resc->max_tx_sch_inputs;
 
 	bnxt_get_max_rings(bp, &max_rx_rings, &max_tx_rings, true);
+	if (max_tx_sch_inputs)
+		max_tx_rings = min_t(int, max_tx_rings, max_tx_sch_inputs);
+
+	tcs = bp->num_tc;
+	tx_grps = max(tcs, 1);
+	if (bp->tx_nr_rings_xdp)
+		tx_grps++;
+	max_tx_rings /= tx_grps;
 	channel->max_combined = min_t(int, max_rx_rings, max_tx_rings);
 
 	if (bnxt_get_max_rings(bp, &max_rx_rings, &max_tx_rings, false)) {
 		max_rx_rings = 0;
 		max_tx_rings = 0;
 	}
+	if (max_tx_sch_inputs)
+		max_tx_rings = min_t(int, max_tx_rings, max_tx_sch_inputs);
 
-	tcs = netdev_get_num_tc(dev);
 	if (tcs > 1)
 		max_tx_rings /= tcs;
 
@@ -402,6 +939,7 @@ static int bnxt_set_channels(struct net_device *dev,
 	bool sh = false;
 	int tx_xdp = 0;
 	int rc = 0;
+	int tx_cp;
 
 	if (channel->other_count)
 		return -EINVAL;
@@ -421,7 +959,7 @@ static int bnxt_set_channels(struct net_device *dev,
 	if (channel->combined_count)
 		sh = true;
 
-	tcs = netdev_get_num_tc(dev);
+	tcs = bp->num_tc;
 
 	req_tx_rings = sh ? channel->combined_count : channel->tx_count;
 	req_rx_rings = sh ? channel->combined_count : channel->rx_count;
@@ -432,7 +970,15 @@ static int bnxt_set_channels(struct net_device *dev,
 		}
 		tx_xdp = req_rx_rings;
 	}
-	rc = bnxt_reserve_rings(bp, req_tx_rings, req_rx_rings, tcs, tx_xdp);
+
+	if (bnxt_get_nr_rss_ctxs(bp, req_rx_rings) !=
+	    bnxt_get_nr_rss_ctxs(bp, bp->rx_nr_rings) &&
+	    netif_is_rxfh_configured(dev)) {
+		netdev_warn(dev, "RSS table size change required, RSS table entries must be default to proceed\n");
+		return -EINVAL;
+	}
+
+	rc = bnxt_check_rings(bp, req_tx_rings, req_rx_rings, sh, tcs, tx_xdp);
 	if (rc) {
 		netdev_warn(dev, "Unable to allocate the requested rings\n");
 		return rc;
@@ -444,12 +990,7 @@ static int bnxt_set_channels(struct net_device *dev,
 			 * before PF unload
 			 */
 		}
-		rc = bnxt_close_nic(bp, true, false);
-		if (rc) {
-			netdev_err(bp->dev, "Set channel failure rc :%x\n",
-				   rc);
-			return rc;
-		}
+		bnxt_close_nic(bp, true, false);
 	}
 
 	if (sh) {
@@ -466,10 +1007,9 @@ static int bnxt_set_channels(struct net_device *dev,
 	if (tcs > 1)
 		bp->tx_nr_rings = bp->tx_nr_rings_per_tc * tcs + tx_xdp;
 
-	bp->cp_nr_rings = sh ? max_t(int, bp->tx_nr_rings, bp->rx_nr_rings) :
-			       bp->tx_nr_rings + bp->rx_nr_rings;
-
-	bp->num_stat_ctxs = bp->cp_nr_rings;
+	tx_cp = bnxt_num_tx_to_cp(bp, bp->tx_nr_rings);
+	bp->cp_nr_rings = sh ? max_t(int, tx_cp, bp->rx_nr_rings) :
+			       tx_cp + bp->rx_nr_rings;
 
 	/* After changing number of rx channels, update NTUPLE feature. */
 	netdev_update_features(dev);
@@ -480,34 +1020,73 @@ static int bnxt_set_channels(struct net_device *dev,
 			 * to renable
 			 */
 		}
+	} else {
+		rc = bnxt_reserve_rings(bp, true);
 	}
 
 	return rc;
 }
 
-#ifdef CONFIG_RFS_ACCEL
-static int bnxt_grxclsrlall(struct bnxt *bp, struct ethtool_rxnfc *cmd,
-			    u32 *rule_locs)
+static u32 bnxt_get_all_fltr_ids_rcu(struct bnxt *bp, struct hlist_head tbl[],
+				     int tbl_size, u32 *ids, u32 start,
+				     u32 id_cnt)
 {
-	int i, j = 0;
+	int i, j = start;
 
-	cmd->data = bp->ntp_fltr_count;
-	for (i = 0; i < BNXT_NTP_FLTR_HASH_SIZE; i++) {
+	if (j >= id_cnt)
+		return j;
+	for (i = 0; i < tbl_size; i++) {
 		struct hlist_head *head;
-		struct bnxt_ntuple_filter *fltr;
+		struct bnxt_filter_base *fltr;
 
-		head = &bp->ntp_fltr_hash_tbl[i];
-		rcu_read_lock();
+		head = &tbl[i];
 		hlist_for_each_entry_rcu(fltr, head, hash) {
-			if (j == cmd->rule_cnt)
-				break;
-			rule_locs[j++] = fltr->sw_id;
+			if (!fltr->flags ||
+			    test_bit(BNXT_FLTR_FW_DELETED, &fltr->state))
+				continue;
+			ids[j++] = fltr->sw_id;
+			if (j == id_cnt)
+				return j;
+		}
+	}
+	return j;
+}
+
+static struct bnxt_filter_base *bnxt_get_one_fltr_rcu(struct bnxt *bp,
+						      struct hlist_head tbl[],
+						      int tbl_size, u32 id)
+{
+	int i;
+
+	for (i = 0; i < tbl_size; i++) {
+		struct hlist_head *head;
+		struct bnxt_filter_base *fltr;
+
+		head = &tbl[i];
+		hlist_for_each_entry_rcu(fltr, head, hash) {
+			if (fltr->flags && fltr->sw_id == id)
+				return fltr;
 		}
-		rcu_read_unlock();
-		if (j == cmd->rule_cnt)
-			break;
 	}
-	cmd->rule_cnt = j;
+	return NULL;
+}
+
+static int bnxt_grxclsrlall(struct bnxt *bp, struct ethtool_rxnfc *cmd,
+			    u32 *rule_locs)
+{
+	u32 count;
+
+	cmd->data = bp->ntp_fltr_count;
+	rcu_read_lock();
+	count = bnxt_get_all_fltr_ids_rcu(bp, bp->l2_fltr_hash_tbl,
+					  BNXT_L2_FLTR_HASH_SIZE, rule_locs, 0,
+					  cmd->rule_cnt);
+	cmd->rule_cnt = bnxt_get_all_fltr_ids_rcu(bp, bp->ntp_fltr_hash_tbl,
+						  BNXT_NTP_FLTR_HASH_SIZE,
+						  rule_locs, count,
+						  cmd->rule_cnt);
+	rcu_read_unlock();
+
 	return 0;
 }
 
@@ -515,73 +1094,129 @@ static int bnxt_grxclsrule(struct bnxt *bp, struct ethtool_rxnfc *cmd)
 {
 	struct ethtool_rx_flow_spec *fs =
 		(struct ethtool_rx_flow_spec *)&cmd->fs;
+	struct bnxt_filter_base *fltr_base;
 	struct bnxt_ntuple_filter *fltr;
+	struct bnxt_flow_masks *fmasks;
 	struct flow_keys *fkeys;
-	int i, rc = -EINVAL;
+	int rc = -EINVAL;
 
-	if (fs->location < 0 || fs->location >= BNXT_NTP_FLTR_MAX_FLTR)
+	if (fs->location >= bp->max_fltr)
 		return rc;
 
-	for (i = 0; i < BNXT_NTP_FLTR_HASH_SIZE; i++) {
-		struct hlist_head *head;
-
-		head = &bp->ntp_fltr_hash_tbl[i];
-		rcu_read_lock();
-		hlist_for_each_entry_rcu(fltr, head, hash) {
-			if (fltr->sw_id == fs->location)
-				goto fltr_found;
+	rcu_read_lock();
+	fltr_base = bnxt_get_one_fltr_rcu(bp, bp->l2_fltr_hash_tbl,
+					  BNXT_L2_FLTR_HASH_SIZE,
+					  fs->location);
+	if (fltr_base) {
+		struct ethhdr *h_ether = &fs->h_u.ether_spec;
+		struct ethhdr *m_ether = &fs->m_u.ether_spec;
+		struct bnxt_l2_filter *l2_fltr;
+		struct bnxt_l2_key *l2_key;
+
+		l2_fltr = container_of(fltr_base, struct bnxt_l2_filter, base);
+		l2_key = &l2_fltr->l2_key;
+		fs->flow_type = ETHER_FLOW;
+		ether_addr_copy(h_ether->h_dest, l2_key->dst_mac_addr);
+		eth_broadcast_addr(m_ether->h_dest);
+		if (l2_key->vlan) {
+			struct ethtool_flow_ext *m_ext = &fs->m_ext;
+			struct ethtool_flow_ext *h_ext = &fs->h_ext;
+
+			fs->flow_type |= FLOW_EXT;
+			m_ext->vlan_tci = htons(0xfff);
+			h_ext->vlan_tci = htons(l2_key->vlan);
 		}
+		if (fltr_base->flags & BNXT_ACT_RING_DST)
+			fs->ring_cookie = fltr_base->rxq;
+		if (fltr_base->flags & BNXT_ACT_FUNC_DST)
+			fs->ring_cookie = (u64)(fltr_base->vf_idx + 1) <<
+					  ETHTOOL_RX_FLOW_SPEC_RING_VF_OFF;
 		rcu_read_unlock();
+		return 0;
 	}
-	return rc;
+	fltr_base = bnxt_get_one_fltr_rcu(bp, bp->ntp_fltr_hash_tbl,
+					  BNXT_NTP_FLTR_HASH_SIZE,
+					  fs->location);
+	if (!fltr_base) {
+		rcu_read_unlock();
+		return rc;
+	}
+	fltr = container_of(fltr_base, struct bnxt_ntuple_filter, base);
 
-fltr_found:
 	fkeys = &fltr->fkeys;
+	fmasks = &fltr->fmasks;
 	if (fkeys->basic.n_proto == htons(ETH_P_IP)) {
-		if (fkeys->basic.ip_proto == IPPROTO_TCP)
+		if (fkeys->basic.ip_proto == BNXT_IP_PROTO_WILDCARD) {
+			fs->flow_type = IP_USER_FLOW;
+			fs->h_u.usr_ip4_spec.ip_ver = ETH_RX_NFC_IP4;
+			fs->h_u.usr_ip4_spec.proto = BNXT_IP_PROTO_WILDCARD;
+			fs->m_u.usr_ip4_spec.proto = 0;
+		} else if (fkeys->basic.ip_proto == IPPROTO_ICMP) {
+			fs->flow_type = IP_USER_FLOW;
+			fs->h_u.usr_ip4_spec.ip_ver = ETH_RX_NFC_IP4;
+			fs->h_u.usr_ip4_spec.proto = IPPROTO_ICMP;
+			fs->m_u.usr_ip4_spec.proto = BNXT_IP_PROTO_FULL_MASK;
+		} else if (fkeys->basic.ip_proto == IPPROTO_TCP) {
 			fs->flow_type = TCP_V4_FLOW;
-		else if (fkeys->basic.ip_proto == IPPROTO_UDP)
+		} else if (fkeys->basic.ip_proto == IPPROTO_UDP) {
 			fs->flow_type = UDP_V4_FLOW;
-		else
+		} else {
 			goto fltr_err;
+		}
 
 		fs->h_u.tcp_ip4_spec.ip4src = fkeys->addrs.v4addrs.src;
-		fs->m_u.tcp_ip4_spec.ip4src = cpu_to_be32(~0);
-
+		fs->m_u.tcp_ip4_spec.ip4src = fmasks->addrs.v4addrs.src;
 		fs->h_u.tcp_ip4_spec.ip4dst = fkeys->addrs.v4addrs.dst;
-		fs->m_u.tcp_ip4_spec.ip4dst = cpu_to_be32(~0);
-
-		fs->h_u.tcp_ip4_spec.psrc = fkeys->ports.src;
-		fs->m_u.tcp_ip4_spec.psrc = cpu_to_be16(~0);
-
-		fs->h_u.tcp_ip4_spec.pdst = fkeys->ports.dst;
-		fs->m_u.tcp_ip4_spec.pdst = cpu_to_be16(~0);
+		fs->m_u.tcp_ip4_spec.ip4dst = fmasks->addrs.v4addrs.dst;
+		if (fs->flow_type == TCP_V4_FLOW ||
+		    fs->flow_type == UDP_V4_FLOW) {
+			fs->h_u.tcp_ip4_spec.psrc = fkeys->ports.src;
+			fs->m_u.tcp_ip4_spec.psrc = fmasks->ports.src;
+			fs->h_u.tcp_ip4_spec.pdst = fkeys->ports.dst;
+			fs->m_u.tcp_ip4_spec.pdst = fmasks->ports.dst;
+		}
 	} else {
-		int i;
-
-		if (fkeys->basic.ip_proto == IPPROTO_TCP)
+		if (fkeys->basic.ip_proto == BNXT_IP_PROTO_WILDCARD) {
+			fs->flow_type = IPV6_USER_FLOW;
+			fs->h_u.usr_ip6_spec.l4_proto = BNXT_IP_PROTO_WILDCARD;
+			fs->m_u.usr_ip6_spec.l4_proto = 0;
+		} else if (fkeys->basic.ip_proto == IPPROTO_ICMPV6) {
+			fs->flow_type = IPV6_USER_FLOW;
+			fs->h_u.usr_ip6_spec.l4_proto = IPPROTO_ICMPV6;
+			fs->m_u.usr_ip6_spec.l4_proto = BNXT_IP_PROTO_FULL_MASK;
+		} else if (fkeys->basic.ip_proto == IPPROTO_TCP) {
 			fs->flow_type = TCP_V6_FLOW;
-		else if (fkeys->basic.ip_proto == IPPROTO_UDP)
+		} else if (fkeys->basic.ip_proto == IPPROTO_UDP) {
 			fs->flow_type = UDP_V6_FLOW;
-		else
+		} else {
 			goto fltr_err;
+		}
 
 		*(struct in6_addr *)&fs->h_u.tcp_ip6_spec.ip6src[0] =
 			fkeys->addrs.v6addrs.src;
+		*(struct in6_addr *)&fs->m_u.tcp_ip6_spec.ip6src[0] =
+			fmasks->addrs.v6addrs.src;
 		*(struct in6_addr *)&fs->h_u.tcp_ip6_spec.ip6dst[0] =
 			fkeys->addrs.v6addrs.dst;
-		for (i = 0; i < 4; i++) {
-			fs->m_u.tcp_ip6_spec.ip6src[i] = cpu_to_be32(~0);
-			fs->m_u.tcp_ip6_spec.ip6dst[i] = cpu_to_be32(~0);
+		*(struct in6_addr *)&fs->m_u.tcp_ip6_spec.ip6dst[0] =
+			fmasks->addrs.v6addrs.dst;
+		if (fs->flow_type == TCP_V6_FLOW ||
+		    fs->flow_type == UDP_V6_FLOW) {
+			fs->h_u.tcp_ip6_spec.psrc = fkeys->ports.src;
+			fs->m_u.tcp_ip6_spec.psrc = fmasks->ports.src;
+			fs->h_u.tcp_ip6_spec.pdst = fkeys->ports.dst;
+			fs->m_u.tcp_ip6_spec.pdst = fmasks->ports.dst;
 		}
-		fs->h_u.tcp_ip6_spec.psrc = fkeys->ports.src;
-		fs->m_u.tcp_ip6_spec.psrc = cpu_to_be16(~0);
-
-		fs->h_u.tcp_ip6_spec.pdst = fkeys->ports.dst;
-		fs->m_u.tcp_ip6_spec.pdst = cpu_to_be16(~0);
 	}
 
-	fs->ring_cookie = fltr->rxq;
+	if (fltr->base.flags & BNXT_ACT_DROP) {
+		fs->ring_cookie = RX_CLS_FLOW_DISC;
+	} else if (fltr->base.flags & BNXT_ACT_RSS_CTX) {
+		fs->flow_type |= FLOW_RSS;
+		cmd->rss_context = fltr->base.fw_vnic_id;
+	} else {
+		fs->ring_cookie = fltr->base.rxq;
+	}
 	rc = 0;
 
 fltr_err:
@@ -589,7 +1224,354 @@ fltr_err:
 
 	return rc;
 }
-#endif
+
+static struct bnxt_rss_ctx *bnxt_get_rss_ctx_from_index(struct bnxt *bp,
+							u32 index)
+{
+	struct ethtool_rxfh_context *ctx;
+
+	ctx = xa_load(&bp->dev->ethtool->rss_ctx, index);
+	if (!ctx)
+		return NULL;
+	return ethtool_rxfh_context_priv(ctx);
+}
+
+static int bnxt_alloc_vnic_rss_table(struct bnxt *bp,
+				     struct bnxt_vnic_info *vnic)
+{
+	int size = L1_CACHE_ALIGN(BNXT_MAX_RSS_TABLE_SIZE_P5);
+
+	vnic->rss_table_size = size + HW_HASH_KEY_SIZE;
+	vnic->rss_table = dma_alloc_coherent(&bp->pdev->dev,
+					     vnic->rss_table_size,
+					     &vnic->rss_table_dma_addr,
+					     GFP_KERNEL);
+	if (!vnic->rss_table)
+		return -ENOMEM;
+
+	vnic->rss_hash_key = ((void *)vnic->rss_table) + size;
+	vnic->rss_hash_key_dma_addr = vnic->rss_table_dma_addr + size;
+	return 0;
+}
+
+static int bnxt_add_l2_cls_rule(struct bnxt *bp,
+				struct ethtool_rx_flow_spec *fs)
+{
+	u32 ring = ethtool_get_flow_spec_ring(fs->ring_cookie);
+	u8 vf = ethtool_get_flow_spec_ring_vf(fs->ring_cookie);
+	struct ethhdr *h_ether = &fs->h_u.ether_spec;
+	struct ethhdr *m_ether = &fs->m_u.ether_spec;
+	struct bnxt_l2_filter *fltr;
+	struct bnxt_l2_key key;
+	u16 vnic_id;
+	u8 flags;
+	int rc;
+
+	if (BNXT_CHIP_P5_PLUS(bp))
+		return -EOPNOTSUPP;
+
+	if (!is_broadcast_ether_addr(m_ether->h_dest))
+		return -EINVAL;
+	ether_addr_copy(key.dst_mac_addr, h_ether->h_dest);
+	key.vlan = 0;
+	if (fs->flow_type & FLOW_EXT) {
+		struct ethtool_flow_ext *m_ext = &fs->m_ext;
+		struct ethtool_flow_ext *h_ext = &fs->h_ext;
+
+		if (m_ext->vlan_tci != htons(0xfff) || !h_ext->vlan_tci)
+			return -EINVAL;
+		key.vlan = ntohs(h_ext->vlan_tci);
+	}
+
+	if (vf) {
+		flags = BNXT_ACT_FUNC_DST;
+		vnic_id = 0xffff;
+		vf--;
+	} else {
+		flags = BNXT_ACT_RING_DST;
+		vnic_id = bp->vnic_info[ring + 1].fw_vnic_id;
+	}
+	fltr = bnxt_alloc_new_l2_filter(bp, &key, flags);
+	if (IS_ERR(fltr))
+		return PTR_ERR(fltr);
+
+	fltr->base.fw_vnic_id = vnic_id;
+	fltr->base.rxq = ring;
+	fltr->base.vf_idx = vf;
+	rc = bnxt_hwrm_l2_filter_alloc(bp, fltr);
+	if (rc)
+		bnxt_del_l2_filter(bp, fltr);
+	else
+		fs->location = fltr->base.sw_id;
+	return rc;
+}
+
+static bool bnxt_verify_ntuple_ip4_flow(struct ethtool_usrip4_spec *ip_spec,
+					struct ethtool_usrip4_spec *ip_mask)
+{
+	u8 mproto = ip_mask->proto;
+	u8 sproto = ip_spec->proto;
+
+	if (ip_mask->l4_4_bytes || ip_mask->tos ||
+	    ip_spec->ip_ver != ETH_RX_NFC_IP4 ||
+	    (mproto && (mproto != BNXT_IP_PROTO_FULL_MASK || sproto != IPPROTO_ICMP)))
+		return false;
+	return true;
+}
+
+static bool bnxt_verify_ntuple_ip6_flow(struct ethtool_usrip6_spec *ip_spec,
+					struct ethtool_usrip6_spec *ip_mask)
+{
+	u8 mproto = ip_mask->l4_proto;
+	u8 sproto = ip_spec->l4_proto;
+
+	if (ip_mask->l4_4_bytes || ip_mask->tclass ||
+	    (mproto && (mproto != BNXT_IP_PROTO_FULL_MASK || sproto != IPPROTO_ICMPV6)))
+		return false;
+	return true;
+}
+
+static int bnxt_add_ntuple_cls_rule(struct bnxt *bp,
+				    struct ethtool_rxnfc *cmd)
+{
+	struct ethtool_rx_flow_spec *fs = &cmd->fs;
+	struct bnxt_ntuple_filter *new_fltr, *fltr;
+	u32 flow_type = fs->flow_type & 0xff;
+	struct bnxt_l2_filter *l2_fltr;
+	struct bnxt_flow_masks *fmasks;
+	struct flow_keys *fkeys;
+	u32 idx, ring;
+	int rc;
+	u8 vf;
+
+	if (!bp->vnic_info)
+		return -EAGAIN;
+
+	vf = ethtool_get_flow_spec_ring_vf(fs->ring_cookie);
+	ring = ethtool_get_flow_spec_ring(fs->ring_cookie);
+	if ((fs->flow_type & (FLOW_MAC_EXT | FLOW_EXT)) || vf)
+		return -EOPNOTSUPP;
+
+	if (flow_type == IP_USER_FLOW) {
+		if (!bnxt_verify_ntuple_ip4_flow(&fs->h_u.usr_ip4_spec,
+						 &fs->m_u.usr_ip4_spec))
+			return -EOPNOTSUPP;
+	}
+
+	if (flow_type == IPV6_USER_FLOW) {
+		if (!bnxt_verify_ntuple_ip6_flow(&fs->h_u.usr_ip6_spec,
+						 &fs->m_u.usr_ip6_spec))
+			return -EOPNOTSUPP;
+	}
+
+	new_fltr = kzalloc(sizeof(*new_fltr), GFP_KERNEL);
+	if (!new_fltr)
+		return -ENOMEM;
+
+	l2_fltr = bp->vnic_info[BNXT_VNIC_DEFAULT].l2_filters[0];
+	atomic_inc(&l2_fltr->refcnt);
+	new_fltr->l2_fltr = l2_fltr;
+	fmasks = &new_fltr->fmasks;
+	fkeys = &new_fltr->fkeys;
+
+	rc = -EOPNOTSUPP;
+	switch (flow_type) {
+	case IP_USER_FLOW: {
+		struct ethtool_usrip4_spec *ip_spec = &fs->h_u.usr_ip4_spec;
+		struct ethtool_usrip4_spec *ip_mask = &fs->m_u.usr_ip4_spec;
+
+		fkeys->basic.ip_proto = ip_mask->proto ? ip_spec->proto
+						       : BNXT_IP_PROTO_WILDCARD;
+		fkeys->basic.n_proto = htons(ETH_P_IP);
+		fkeys->addrs.v4addrs.src = ip_spec->ip4src;
+		fmasks->addrs.v4addrs.src = ip_mask->ip4src;
+		fkeys->addrs.v4addrs.dst = ip_spec->ip4dst;
+		fmasks->addrs.v4addrs.dst = ip_mask->ip4dst;
+		break;
+	}
+	case TCP_V4_FLOW:
+	case UDP_V4_FLOW: {
+		struct ethtool_tcpip4_spec *ip_spec = &fs->h_u.tcp_ip4_spec;
+		struct ethtool_tcpip4_spec *ip_mask = &fs->m_u.tcp_ip4_spec;
+
+		fkeys->basic.ip_proto = IPPROTO_TCP;
+		if (flow_type == UDP_V4_FLOW)
+			fkeys->basic.ip_proto = IPPROTO_UDP;
+		fkeys->basic.n_proto = htons(ETH_P_IP);
+		fkeys->addrs.v4addrs.src = ip_spec->ip4src;
+		fmasks->addrs.v4addrs.src = ip_mask->ip4src;
+		fkeys->addrs.v4addrs.dst = ip_spec->ip4dst;
+		fmasks->addrs.v4addrs.dst = ip_mask->ip4dst;
+		fkeys->ports.src = ip_spec->psrc;
+		fmasks->ports.src = ip_mask->psrc;
+		fkeys->ports.dst = ip_spec->pdst;
+		fmasks->ports.dst = ip_mask->pdst;
+		break;
+	}
+	case IPV6_USER_FLOW: {
+		struct ethtool_usrip6_spec *ip_spec = &fs->h_u.usr_ip6_spec;
+		struct ethtool_usrip6_spec *ip_mask = &fs->m_u.usr_ip6_spec;
+
+		fkeys->basic.ip_proto = ip_mask->l4_proto ? ip_spec->l4_proto
+							  : BNXT_IP_PROTO_WILDCARD;
+		fkeys->basic.n_proto = htons(ETH_P_IPV6);
+		fkeys->addrs.v6addrs.src = *(struct in6_addr *)&ip_spec->ip6src;
+		fmasks->addrs.v6addrs.src = *(struct in6_addr *)&ip_mask->ip6src;
+		fkeys->addrs.v6addrs.dst = *(struct in6_addr *)&ip_spec->ip6dst;
+		fmasks->addrs.v6addrs.dst = *(struct in6_addr *)&ip_mask->ip6dst;
+		break;
+	}
+	case TCP_V6_FLOW:
+	case UDP_V6_FLOW: {
+		struct ethtool_tcpip6_spec *ip_spec = &fs->h_u.tcp_ip6_spec;
+		struct ethtool_tcpip6_spec *ip_mask = &fs->m_u.tcp_ip6_spec;
+
+		fkeys->basic.ip_proto = IPPROTO_TCP;
+		if (flow_type == UDP_V6_FLOW)
+			fkeys->basic.ip_proto = IPPROTO_UDP;
+		fkeys->basic.n_proto = htons(ETH_P_IPV6);
+
+		fkeys->addrs.v6addrs.src = *(struct in6_addr *)&ip_spec->ip6src;
+		fmasks->addrs.v6addrs.src = *(struct in6_addr *)&ip_mask->ip6src;
+		fkeys->addrs.v6addrs.dst = *(struct in6_addr *)&ip_spec->ip6dst;
+		fmasks->addrs.v6addrs.dst = *(struct in6_addr *)&ip_mask->ip6dst;
+		fkeys->ports.src = ip_spec->psrc;
+		fmasks->ports.src = ip_mask->psrc;
+		fkeys->ports.dst = ip_spec->pdst;
+		fmasks->ports.dst = ip_mask->pdst;
+		break;
+	}
+	default:
+		rc = -EOPNOTSUPP;
+		goto ntuple_err;
+	}
+	if (!memcmp(&BNXT_FLOW_MASK_NONE, fmasks, sizeof(*fmasks)))
+		goto ntuple_err;
+
+	idx = bnxt_get_ntp_filter_idx(bp, fkeys, NULL);
+	rcu_read_lock();
+	fltr = bnxt_lookup_ntp_filter_from_idx(bp, new_fltr, idx);
+	if (fltr) {
+		rcu_read_unlock();
+		rc = -EEXIST;
+		goto ntuple_err;
+	}
+	rcu_read_unlock();
+
+	new_fltr->base.flags = BNXT_ACT_NO_AGING;
+	if (fs->flow_type & FLOW_RSS) {
+		struct bnxt_rss_ctx *rss_ctx;
+
+		new_fltr->base.fw_vnic_id = 0;
+		new_fltr->base.flags |= BNXT_ACT_RSS_CTX;
+		rss_ctx = bnxt_get_rss_ctx_from_index(bp, cmd->rss_context);
+		if (rss_ctx) {
+			new_fltr->base.fw_vnic_id = rss_ctx->index;
+		} else {
+			rc = -EINVAL;
+			goto ntuple_err;
+		}
+	}
+	if (fs->ring_cookie == RX_CLS_FLOW_DISC)
+		new_fltr->base.flags |= BNXT_ACT_DROP;
+	else
+		new_fltr->base.rxq = ring;
+	__set_bit(BNXT_FLTR_VALID, &new_fltr->base.state);
+	rc = bnxt_insert_ntp_filter(bp, new_fltr, idx);
+	if (!rc) {
+		rc = bnxt_hwrm_cfa_ntuple_filter_alloc(bp, new_fltr);
+		if (rc) {
+			bnxt_del_ntp_filter(bp, new_fltr);
+			return rc;
+		}
+		fs->location = new_fltr->base.sw_id;
+		return 0;
+	}
+
+ntuple_err:
+	atomic_dec(&l2_fltr->refcnt);
+	kfree(new_fltr);
+	return rc;
+}
+
+static int bnxt_srxclsrlins(struct bnxt *bp, struct ethtool_rxnfc *cmd)
+{
+	struct ethtool_rx_flow_spec *fs = &cmd->fs;
+	u32 ring, flow_type;
+	int rc;
+	u8 vf;
+
+	if (!netif_running(bp->dev))
+		return -EAGAIN;
+	if (!(bp->flags & BNXT_FLAG_RFS))
+		return -EPERM;
+	if (fs->location != RX_CLS_LOC_ANY)
+		return -EINVAL;
+
+	flow_type = fs->flow_type;
+	if ((flow_type == IP_USER_FLOW ||
+	     flow_type == IPV6_USER_FLOW) &&
+	    !(bp->fw_cap & BNXT_FW_CAP_CFA_NTUPLE_RX_EXT_IP_PROTO))
+		return -EOPNOTSUPP;
+	if (flow_type & FLOW_MAC_EXT)
+		return -EINVAL;
+	flow_type &= ~FLOW_EXT;
+
+	if (fs->ring_cookie == RX_CLS_FLOW_DISC && flow_type != ETHER_FLOW)
+		return bnxt_add_ntuple_cls_rule(bp, cmd);
+
+	ring = ethtool_get_flow_spec_ring(fs->ring_cookie);
+	vf = ethtool_get_flow_spec_ring_vf(fs->ring_cookie);
+	if (BNXT_VF(bp) && vf)
+		return -EINVAL;
+	if (BNXT_PF(bp) && vf > bp->pf.active_vfs)
+		return -EINVAL;
+	if (!vf && ring >= bp->rx_nr_rings)
+		return -EINVAL;
+
+	if (flow_type == ETHER_FLOW)
+		rc = bnxt_add_l2_cls_rule(bp, fs);
+	else
+		rc = bnxt_add_ntuple_cls_rule(bp, cmd);
+	return rc;
+}
+
+static int bnxt_srxclsrldel(struct bnxt *bp, struct ethtool_rxnfc *cmd)
+{
+	struct ethtool_rx_flow_spec *fs = &cmd->fs;
+	struct bnxt_filter_base *fltr_base;
+	struct bnxt_ntuple_filter *fltr;
+	u32 id = fs->location;
+
+	rcu_read_lock();
+	fltr_base = bnxt_get_one_fltr_rcu(bp, bp->l2_fltr_hash_tbl,
+					  BNXT_L2_FLTR_HASH_SIZE, id);
+	if (fltr_base) {
+		struct bnxt_l2_filter *l2_fltr;
+
+		l2_fltr = container_of(fltr_base, struct bnxt_l2_filter, base);
+		rcu_read_unlock();
+		bnxt_hwrm_l2_filter_free(bp, l2_fltr);
+		bnxt_del_l2_filter(bp, l2_fltr);
+		return 0;
+	}
+	fltr_base = bnxt_get_one_fltr_rcu(bp, bp->ntp_fltr_hash_tbl,
+					  BNXT_NTP_FLTR_HASH_SIZE, id);
+	if (!fltr_base) {
+		rcu_read_unlock();
+		return -ENOENT;
+	}
+
+	fltr = container_of(fltr_base, struct bnxt_ntuple_filter, base);
+	if (!(fltr->base.flags & BNXT_ACT_NO_AGING)) {
+		rcu_read_unlock();
+		return -EINVAL;
+	}
+	rcu_read_unlock();
+	bnxt_hwrm_cfa_ntuple_filter_free(bp, fltr);
+	bnxt_del_ntp_filter(bp, fltr);
+	return 0;
+}
 
 static u64 get_ethtool_ipv4_rss(struct bnxt *bp)
 {
@@ -602,11 +1584,16 @@ static u64 get_ethtool_ipv6_rss(struct bnxt *bp)
 {
 	if (bp->rss_hash_cfg & VNIC_RSS_CFG_REQ_HASH_TYPE_IPV6)
 		return RXH_IP_SRC | RXH_IP_DST;
+	if (bp->rss_hash_cfg & VNIC_RSS_CFG_REQ_HASH_TYPE_IPV6_FLOW_LABEL)
+		return RXH_IP_SRC | RXH_IP_DST | RXH_IP6_FL;
 	return 0;
 }
 
-static int bnxt_grxfh(struct bnxt *bp, struct ethtool_rxnfc *cmd)
+static int bnxt_get_rxfh_fields(struct net_device *dev,
+				struct ethtool_rxfh_fields *cmd)
 {
+	struct bnxt *bp = netdev_priv(dev);
+
 	cmd->data = 0;
 	switch (cmd->flow_type) {
 	case TCP_V4_FLOW:
@@ -619,9 +1606,15 @@ static int bnxt_grxfh(struct bnxt *bp, struct ethtool_rxnfc *cmd)
 		if (bp->rss_hash_cfg & VNIC_RSS_CFG_REQ_HASH_TYPE_UDP_IPV4)
 			cmd->data |= RXH_IP_SRC | RXH_IP_DST |
 				     RXH_L4_B_0_1 | RXH_L4_B_2_3;
-		/* fall through */
-	case SCTP_V4_FLOW:
+		fallthrough;
 	case AH_ESP_V4_FLOW:
+		if (bp->rss_hash_cfg &
+		    (VNIC_RSS_CFG_REQ_HASH_TYPE_AH_SPI_IPV4 |
+		     VNIC_RSS_CFG_REQ_HASH_TYPE_ESP_SPI_IPV4))
+			cmd->data |= RXH_IP_SRC | RXH_IP_DST |
+				     RXH_L4_B_0_1 | RXH_L4_B_2_3;
+		fallthrough;
+	case SCTP_V4_FLOW:
 	case AH_V4_FLOW:
 	case ESP_V4_FLOW:
 	case IPV4_FLOW:
@@ -638,9 +1631,15 @@ static int bnxt_grxfh(struct bnxt *bp, struct ethtool_rxnfc *cmd)
 		if (bp->rss_hash_cfg & VNIC_RSS_CFG_REQ_HASH_TYPE_UDP_IPV6)
 			cmd->data |= RXH_IP_SRC | RXH_IP_DST |
 				     RXH_L4_B_0_1 | RXH_L4_B_2_3;
-		/* fall through */
-	case SCTP_V6_FLOW:
+		fallthrough;
 	case AH_ESP_V6_FLOW:
+		if (bp->rss_hash_cfg &
+		    (VNIC_RSS_CFG_REQ_HASH_TYPE_AH_SPI_IPV6 |
+		     VNIC_RSS_CFG_REQ_HASH_TYPE_ESP_SPI_IPV6))
+			cmd->data |= RXH_IP_SRC | RXH_IP_DST |
+				     RXH_L4_B_0_1 | RXH_L4_B_2_3;
+		fallthrough;
+	case SCTP_V6_FLOW:
 	case AH_V6_FLOW:
 	case ESP_V6_FLOW:
 	case IPV6_FLOW:
@@ -653,26 +1652,36 @@ static int bnxt_grxfh(struct bnxt *bp, struct ethtool_rxnfc *cmd)
 #define RXH_4TUPLE (RXH_IP_SRC | RXH_IP_DST | RXH_L4_B_0_1 | RXH_L4_B_2_3)
 #define RXH_2TUPLE (RXH_IP_SRC | RXH_IP_DST)
 
-static int bnxt_srxfh(struct bnxt *bp, struct ethtool_rxnfc *cmd)
+static int bnxt_set_rxfh_fields(struct net_device *dev,
+				const struct ethtool_rxfh_fields *cmd,
+				struct netlink_ext_ack *extack)
 {
-	u32 rss_hash_cfg = bp->rss_hash_cfg;
+	struct bnxt *bp = netdev_priv(dev);
 	int tuple, rc = 0;
+	u32 rss_hash_cfg;
+
+	rss_hash_cfg = bp->rss_hash_cfg;
 
 	if (cmd->data == RXH_4TUPLE)
 		tuple = 4;
-	else if (cmd->data == RXH_2TUPLE)
+	else if (cmd->data == RXH_2TUPLE ||
+		 cmd->data == (RXH_2TUPLE | RXH_IP6_FL))
 		tuple = 2;
 	else if (!cmd->data)
 		tuple = 0;
 	else
 		return -EINVAL;
 
+	if (cmd->data & RXH_IP6_FL &&
+	    !(bp->rss_cap & BNXT_RSS_CAP_IPV6_FLOW_LABEL_RSS_CAP))
+		return -EINVAL;
+
 	if (cmd->flow_type == TCP_V4_FLOW) {
 		rss_hash_cfg &= ~VNIC_RSS_CFG_REQ_HASH_TYPE_TCP_IPV4;
 		if (tuple == 4)
 			rss_hash_cfg |= VNIC_RSS_CFG_REQ_HASH_TYPE_TCP_IPV4;
 	} else if (cmd->flow_type == UDP_V4_FLOW) {
-		if (tuple == 4 && !(bp->flags & BNXT_FLAG_UDP_RSS_CAP))
+		if (tuple == 4 && !(bp->rss_cap & BNXT_RSS_CAP_UDP_RSS_CAP))
 			return -EINVAL;
 		rss_hash_cfg &= ~VNIC_RSS_CFG_REQ_HASH_TYPE_UDP_IPV4;
 		if (tuple == 4)
@@ -682,11 +1691,29 @@ static int bnxt_srxfh(struct bnxt *bp, struct ethtool_rxnfc *cmd)
 		if (tuple == 4)
 			rss_hash_cfg |= VNIC_RSS_CFG_REQ_HASH_TYPE_TCP_IPV6;
 	} else if (cmd->flow_type == UDP_V6_FLOW) {
-		if (tuple == 4 && !(bp->flags & BNXT_FLAG_UDP_RSS_CAP))
+		if (tuple == 4 && !(bp->rss_cap & BNXT_RSS_CAP_UDP_RSS_CAP))
 			return -EINVAL;
 		rss_hash_cfg &= ~VNIC_RSS_CFG_REQ_HASH_TYPE_UDP_IPV6;
 		if (tuple == 4)
 			rss_hash_cfg |= VNIC_RSS_CFG_REQ_HASH_TYPE_UDP_IPV6;
+	} else if (cmd->flow_type == AH_ESP_V4_FLOW) {
+		if (tuple == 4 && (!(bp->rss_cap & BNXT_RSS_CAP_AH_V4_RSS_CAP) ||
+				   !(bp->rss_cap & BNXT_RSS_CAP_ESP_V4_RSS_CAP)))
+			return -EINVAL;
+		rss_hash_cfg &= ~(VNIC_RSS_CFG_REQ_HASH_TYPE_AH_SPI_IPV4 |
+				  VNIC_RSS_CFG_REQ_HASH_TYPE_ESP_SPI_IPV4);
+		if (tuple == 4)
+			rss_hash_cfg |= VNIC_RSS_CFG_REQ_HASH_TYPE_AH_SPI_IPV4 |
+					VNIC_RSS_CFG_REQ_HASH_TYPE_ESP_SPI_IPV4;
+	} else if (cmd->flow_type == AH_ESP_V6_FLOW) {
+		if (tuple == 4 && (!(bp->rss_cap & BNXT_RSS_CAP_AH_V6_RSS_CAP) ||
+				   !(bp->rss_cap & BNXT_RSS_CAP_ESP_V6_RSS_CAP)))
+			return -EINVAL;
+		rss_hash_cfg &= ~(VNIC_RSS_CFG_REQ_HASH_TYPE_AH_SPI_IPV6 |
+				  VNIC_RSS_CFG_REQ_HASH_TYPE_ESP_SPI_IPV6);
+		if (tuple == 4)
+			rss_hash_cfg |= VNIC_RSS_CFG_REQ_HASH_TYPE_AH_SPI_IPV6 |
+					VNIC_RSS_CFG_REQ_HASH_TYPE_ESP_SPI_IPV6;
 	} else if (tuple == 4) {
 		return -EINVAL;
 	}
@@ -712,16 +1739,23 @@ static int bnxt_srxfh(struct bnxt *bp, struct ethtool_rxnfc *cmd)
 	case AH_V6_FLOW:
 	case ESP_V6_FLOW:
 	case IPV6_FLOW:
-		if (tuple == 2)
+		rss_hash_cfg &= ~(VNIC_RSS_CFG_REQ_HASH_TYPE_IPV6 |
+				  VNIC_RSS_CFG_REQ_HASH_TYPE_IPV6_FLOW_LABEL);
+		if (!tuple)
+			break;
+		if (cmd->data & RXH_IP6_FL)
+			rss_hash_cfg |=
+				VNIC_RSS_CFG_REQ_HASH_TYPE_IPV6_FLOW_LABEL;
+		else if (tuple == 2)
 			rss_hash_cfg |= VNIC_RSS_CFG_REQ_HASH_TYPE_IPV6;
-		else if (!tuple)
-			rss_hash_cfg &= ~VNIC_RSS_CFG_REQ_HASH_TYPE_IPV6;
 		break;
 	}
 
 	if (bp->rss_hash_cfg == rss_hash_cfg)
 		return 0;
 
+	if (bp->rss_cap & BNXT_RSS_CAP_RSS_HASH_TYPE_DELTA)
+		bp->rss_hash_delta = bp->rss_hash_cfg ^ rss_hash_cfg;
 	bp->rss_hash_cfg = rss_hash_cfg;
 	if (netif_running(bp->dev)) {
 		bnxt_close_nic(bp, false, false);
@@ -737,14 +1771,13 @@ static int bnxt_get_rxnfc(struct net_device *dev, struct ethtool_rxnfc *cmd,
 	int rc = 0;
 
 	switch (cmd->cmd) {
-#ifdef CONFIG_RFS_ACCEL
 	case ETHTOOL_GRXRINGS:
 		cmd->data = bp->rx_nr_rings;
 		break;
 
 	case ETHTOOL_GRXCLSRLCNT:
 		cmd->rule_cnt = bp->ntp_fltr_count;
-		cmd->data = BNXT_NTP_FLTR_MAX_FLTR;
+		cmd->data = bp->max_fltr | RX_CLS_LOC_SPECIAL;
 		break;
 
 	case ETHTOOL_GRXCLSRLALL:
@@ -754,11 +1787,6 @@ static int bnxt_get_rxnfc(struct net_device *dev, struct ethtool_rxnfc *cmd,
 	case ETHTOOL_GRXCLSRULE:
 		rc = bnxt_grxclsrule(bp, cmd);
 		break;
-#endif
-
-	case ETHTOOL_GRXFH:
-		rc = bnxt_grxfh(bp, cmd);
-		break;
 
 	default:
 		rc = -EOPNOTSUPP;
@@ -774,8 +1802,12 @@ static int bnxt_set_rxnfc(struct net_device *dev, struct ethtool_rxnfc *cmd)
 	int rc;
 
 	switch (cmd->cmd) {
-	case ETHTOOL_SRXFH:
-		rc = bnxt_srxfh(bp, cmd);
+	case ETHTOOL_SRXCLSRLINS:
+		rc = bnxt_srxclsrlins(bp, cmd);
+		break;
+
+	case ETHTOOL_SRXCLSRLDEL:
+		rc = bnxt_srxclsrldel(bp, cmd);
 		break;
 
 	default:
@@ -785,8 +1817,13 @@ static int bnxt_set_rxnfc(struct net_device *dev, struct ethtool_rxnfc *cmd)
 	return rc;
 }
 
-static u32 bnxt_get_rxfh_indir_size(struct net_device *dev)
+u32 bnxt_get_rxfh_indir_size(struct net_device *dev)
 {
+	struct bnxt *bp = netdev_priv(dev);
+
+	if (bp->flags & BNXT_FLAG_CHIP_P5_PLUS)
+		return bnxt_get_nr_rss_ctxs(bp, bp->rx_nr_rings) *
+		       BNXT_RSS_TABLE_ENTRIES_P5;
 	return HW_HASH_INDEX_SIZE;
 }
 
@@ -795,52 +1832,325 @@ static u32 bnxt_get_rxfh_key_size(struct net_device *dev)
 	return HW_HASH_KEY_SIZE;
 }
 
-static int bnxt_get_rxfh(struct net_device *dev, u32 *indir, u8 *key,
-			 u8 *hfunc)
+static int bnxt_get_rxfh(struct net_device *dev,
+			 struct ethtool_rxfh_param *rxfh)
 {
+	struct bnxt_rss_ctx *rss_ctx = NULL;
 	struct bnxt *bp = netdev_priv(dev);
-	struct bnxt_vnic_info *vnic = &bp->vnic_info[0];
-	int i = 0;
+	u32 *indir_tbl = bp->rss_indir_tbl;
+	struct bnxt_vnic_info *vnic;
+	u32 i, tbl_size;
+
+	rxfh->hfunc = ETH_RSS_HASH_TOP;
+
+	if (!bp->vnic_info)
+		return 0;
+
+	vnic = &bp->vnic_info[BNXT_VNIC_DEFAULT];
+	if (rxfh->rss_context) {
+		struct ethtool_rxfh_context *ctx;
 
-	if (hfunc)
-		*hfunc = ETH_RSS_HASH_TOP;
+		ctx = xa_load(&bp->dev->ethtool->rss_ctx, rxfh->rss_context);
+		if (!ctx)
+			return -EINVAL;
+		indir_tbl = ethtool_rxfh_context_indir(ctx);
+		rss_ctx = ethtool_rxfh_context_priv(ctx);
+		vnic = &rss_ctx->vnic;
+	}
 
-	if (indir)
-		for (i = 0; i < HW_HASH_INDEX_SIZE; i++)
-			indir[i] = le16_to_cpu(vnic->rss_table[i]);
+	if (rxfh->indir && indir_tbl) {
+		tbl_size = bnxt_get_rxfh_indir_size(dev);
+		for (i = 0; i < tbl_size; i++)
+			rxfh->indir[i] = indir_tbl[i];
+	}
 
-	if (key)
-		memcpy(key, vnic->rss_hash_key, HW_HASH_KEY_SIZE);
+	if (rxfh->key && vnic->rss_hash_key)
+		memcpy(rxfh->key, vnic->rss_hash_key, HW_HASH_KEY_SIZE);
 
 	return 0;
 }
 
+static void bnxt_modify_rss(struct bnxt *bp, struct ethtool_rxfh_context *ctx,
+			    struct bnxt_rss_ctx *rss_ctx,
+			    const struct ethtool_rxfh_param *rxfh)
+{
+	if (rxfh->key) {
+		if (rss_ctx) {
+			memcpy(rss_ctx->vnic.rss_hash_key, rxfh->key,
+			       HW_HASH_KEY_SIZE);
+		} else {
+			memcpy(bp->rss_hash_key, rxfh->key, HW_HASH_KEY_SIZE);
+			bp->rss_hash_key_updated = true;
+		}
+	}
+	if (rxfh->indir) {
+		u32 i, pad, tbl_size = bnxt_get_rxfh_indir_size(bp->dev);
+		u32 *indir_tbl = bp->rss_indir_tbl;
+
+		if (rss_ctx)
+			indir_tbl = ethtool_rxfh_context_indir(ctx);
+		for (i = 0; i < tbl_size; i++)
+			indir_tbl[i] = rxfh->indir[i];
+		pad = bp->rss_indir_tbl_entries - tbl_size;
+		if (pad)
+			memset(&indir_tbl[i], 0, pad * sizeof(*indir_tbl));
+	}
+}
+
+static int bnxt_rxfh_context_check(struct bnxt *bp,
+				   const struct ethtool_rxfh_param *rxfh,
+				   struct netlink_ext_ack *extack)
+{
+	if (rxfh->hfunc && rxfh->hfunc != ETH_RSS_HASH_TOP) {
+		NL_SET_ERR_MSG_MOD(extack, "RSS hash function not supported");
+		return -EOPNOTSUPP;
+	}
+
+	if (!BNXT_SUPPORTS_MULTI_RSS_CTX(bp)) {
+		NL_SET_ERR_MSG_MOD(extack, "RSS contexts not supported");
+		return -EOPNOTSUPP;
+	}
+
+	if (!netif_running(bp->dev)) {
+		NL_SET_ERR_MSG_MOD(extack, "Unable to set RSS contexts when interface is down");
+		return -EAGAIN;
+	}
+
+	return 0;
+}
+
+static int bnxt_create_rxfh_context(struct net_device *dev,
+				    struct ethtool_rxfh_context *ctx,
+				    const struct ethtool_rxfh_param *rxfh,
+				    struct netlink_ext_ack *extack)
+{
+	struct bnxt *bp = netdev_priv(dev);
+	struct bnxt_rss_ctx *rss_ctx;
+	struct bnxt_vnic_info *vnic;
+	int rc;
+
+	rc = bnxt_rxfh_context_check(bp, rxfh, extack);
+	if (rc)
+		return rc;
+
+	if (bp->num_rss_ctx >= BNXT_MAX_ETH_RSS_CTX) {
+		NL_SET_ERR_MSG_FMT_MOD(extack, "Out of RSS contexts, maximum %u",
+				       BNXT_MAX_ETH_RSS_CTX);
+		return -EINVAL;
+	}
+
+	if (!bnxt_rfs_capable(bp, true)) {
+		NL_SET_ERR_MSG_MOD(extack, "Out hardware resources");
+		return -ENOMEM;
+	}
+
+	rss_ctx = ethtool_rxfh_context_priv(ctx);
+
+	bp->num_rss_ctx++;
+
+	vnic = &rss_ctx->vnic;
+	vnic->rss_ctx = ctx;
+	vnic->flags |= BNXT_VNIC_RSSCTX_FLAG;
+	vnic->vnic_id = BNXT_VNIC_ID_INVALID;
+	rc = bnxt_alloc_vnic_rss_table(bp, vnic);
+	if (rc)
+		goto out;
+
+	/* Populate defaults in the context */
+	bnxt_set_dflt_rss_indir_tbl(bp, ctx);
+	ctx->hfunc = ETH_RSS_HASH_TOP;
+	memcpy(vnic->rss_hash_key, bp->rss_hash_key, HW_HASH_KEY_SIZE);
+	memcpy(ethtool_rxfh_context_key(ctx),
+	       bp->rss_hash_key, HW_HASH_KEY_SIZE);
+
+	rc = bnxt_hwrm_vnic_alloc(bp, vnic, 0, bp->rx_nr_rings);
+	if (rc) {
+		NL_SET_ERR_MSG_MOD(extack, "Unable to allocate VNIC");
+		goto out;
+	}
+
+	rc = bnxt_hwrm_vnic_set_tpa(bp, vnic, bp->flags & BNXT_FLAG_TPA);
+	if (rc) {
+		NL_SET_ERR_MSG_MOD(extack, "Unable to setup TPA");
+		goto out;
+	}
+	bnxt_modify_rss(bp, ctx, rss_ctx, rxfh);
+
+	rc = __bnxt_setup_vnic_p5(bp, vnic);
+	if (rc) {
+		NL_SET_ERR_MSG_MOD(extack, "Unable to setup TPA");
+		goto out;
+	}
+
+	rss_ctx->index = rxfh->rss_context;
+	return 0;
+out:
+	bnxt_del_one_rss_ctx(bp, rss_ctx, true);
+	return rc;
+}
+
+static int bnxt_modify_rxfh_context(struct net_device *dev,
+				    struct ethtool_rxfh_context *ctx,
+				    const struct ethtool_rxfh_param *rxfh,
+				    struct netlink_ext_ack *extack)
+{
+	struct bnxt *bp = netdev_priv(dev);
+	struct bnxt_rss_ctx *rss_ctx;
+	int rc;
+
+	rc = bnxt_rxfh_context_check(bp, rxfh, extack);
+	if (rc)
+		return rc;
+
+	rss_ctx = ethtool_rxfh_context_priv(ctx);
+
+	bnxt_modify_rss(bp, ctx, rss_ctx, rxfh);
+
+	return bnxt_hwrm_vnic_rss_cfg_p5(bp, &rss_ctx->vnic);
+}
+
+static int bnxt_remove_rxfh_context(struct net_device *dev,
+				    struct ethtool_rxfh_context *ctx,
+				    u32 rss_context,
+				    struct netlink_ext_ack *extack)
+{
+	struct bnxt *bp = netdev_priv(dev);
+	struct bnxt_rss_ctx *rss_ctx;
+
+	rss_ctx = ethtool_rxfh_context_priv(ctx);
+
+	bnxt_del_one_rss_ctx(bp, rss_ctx, true);
+	return 0;
+}
+
+static int bnxt_set_rxfh(struct net_device *dev,
+			 struct ethtool_rxfh_param *rxfh,
+			 struct netlink_ext_ack *extack)
+{
+	struct bnxt *bp = netdev_priv(dev);
+	int rc = 0;
+
+	if (rxfh->hfunc && rxfh->hfunc != ETH_RSS_HASH_TOP)
+		return -EOPNOTSUPP;
+
+	bnxt_modify_rss(bp, NULL, NULL, rxfh);
+
+	if (netif_running(bp->dev)) {
+		bnxt_close_nic(bp, false, false);
+		rc = bnxt_open_nic(bp, false, false);
+	}
+	return rc;
+}
+
 static void bnxt_get_drvinfo(struct net_device *dev,
 			     struct ethtool_drvinfo *info)
 {
 	struct bnxt *bp = netdev_priv(dev);
-	char *pkglog;
-	char *pkgver = NULL;
 
-	pkglog = kmalloc(BNX_PKG_LOG_MAX_LENGTH, GFP_KERNEL);
-	if (pkglog)
-		pkgver = bnxt_get_pkgver(dev, pkglog, BNX_PKG_LOG_MAX_LENGTH);
-	strlcpy(info->driver, DRV_MODULE_NAME, sizeof(info->driver));
-	strlcpy(info->version, DRV_MODULE_VERSION, sizeof(info->version));
-	if (pkgver && *pkgver != 0 && isdigit(*pkgver))
-		snprintf(info->fw_version, sizeof(info->fw_version) - 1,
-			 "%s pkg %s", bp->fw_ver_str, pkgver);
-	else
-		strlcpy(info->fw_version, bp->fw_ver_str,
-			sizeof(info->fw_version));
-	strlcpy(info->bus_info, pci_name(bp->pdev), sizeof(info->bus_info));
-	info->n_stats = BNXT_NUM_STATS * bp->cp_nr_rings;
+	strscpy(info->driver, DRV_MODULE_NAME, sizeof(info->driver));
+	strscpy(info->fw_version, bp->fw_ver_str, sizeof(info->fw_version));
+	strscpy(info->bus_info, pci_name(bp->pdev), sizeof(info->bus_info));
+	info->n_stats = bnxt_get_num_stats(bp);
 	info->testinfo_len = bp->num_tests;
 	/* TODO CHIMP_FW: eeprom dump details */
 	info->eedump_len = 0;
 	/* TODO CHIMP FW: reg dump details */
 	info->regdump_len = 0;
-	kfree(pkglog);
+}
+
+static int bnxt_get_regs_len(struct net_device *dev)
+{
+	struct bnxt *bp = netdev_priv(dev);
+
+	if (!BNXT_PF(bp))
+		return -EOPNOTSUPP;
+
+	return BNXT_PXP_REG_LEN + bp->pcie_stat_len;
+}
+
+static void *
+__bnxt_hwrm_pcie_qstats(struct bnxt *bp, struct hwrm_pcie_qstats_input *req)
+{
+	struct pcie_ctx_hw_stats_v2 *hw_pcie_stats;
+	dma_addr_t hw_pcie_stats_addr;
+	int rc;
+
+	hw_pcie_stats = hwrm_req_dma_slice(bp, req, sizeof(*hw_pcie_stats),
+					   &hw_pcie_stats_addr);
+	if (!hw_pcie_stats)
+		return NULL;
+
+	req->pcie_stat_size = cpu_to_le16(sizeof(*hw_pcie_stats));
+	req->pcie_stat_host_addr = cpu_to_le64(hw_pcie_stats_addr);
+	rc = hwrm_req_send(bp, req);
+
+	return rc ? NULL : hw_pcie_stats;
+}
+
+#define BNXT_PCIE_32B_ENTRY(start, end)			\
+	 { offsetof(struct pcie_ctx_hw_stats_v2, start),\
+	   offsetof(struct pcie_ctx_hw_stats_v2, end) }
+
+static const struct {
+	u16 start;
+	u16 end;
+} bnxt_pcie_32b_entries[] = {
+	BNXT_PCIE_32B_ENTRY(pcie_ltssm_histogram[0], pcie_ltssm_histogram[3]),
+	BNXT_PCIE_32B_ENTRY(pcie_tl_credit_nph_histogram[0], unused_1),
+	BNXT_PCIE_32B_ENTRY(pcie_rd_latency_histogram[0], unused_2),
+};
+
+static void bnxt_get_regs(struct net_device *dev, struct ethtool_regs *regs,
+			  void *_p)
+{
+	struct hwrm_pcie_qstats_output *resp;
+	struct hwrm_pcie_qstats_input *req;
+	struct bnxt *bp = netdev_priv(dev);
+	u8 *src;
+
+	regs->version = 0;
+	if (!(bp->fw_dbg_cap & DBG_QCAPS_RESP_FLAGS_REG_ACCESS_RESTRICTED))
+		bnxt_dbg_hwrm_rd_reg(bp, 0, BNXT_PXP_REG_LEN / 4, _p);
+
+	if (!(bp->fw_cap & BNXT_FW_CAP_PCIE_STATS_SUPPORTED))
+		return;
+
+	if (hwrm_req_init(bp, req, HWRM_PCIE_QSTATS))
+		return;
+
+	resp = hwrm_req_hold(bp, req);
+	src = __bnxt_hwrm_pcie_qstats(bp, req);
+	if (src) {
+		u8 *dst = (u8 *)(_p + BNXT_PXP_REG_LEN);
+		int i, j, len;
+
+		len = min(bp->pcie_stat_len, le16_to_cpu(resp->pcie_stat_size));
+		if (len <= sizeof(struct pcie_ctx_hw_stats))
+			regs->version = 1;
+		else if (len < sizeof(struct pcie_ctx_hw_stats_v2))
+			regs->version = 2;
+		else
+			regs->version = 3;
+
+		for (i = 0, j = 0; i < len; ) {
+			if (i >= bnxt_pcie_32b_entries[j].start &&
+			    i <= bnxt_pcie_32b_entries[j].end) {
+				u32 *dst32 = (u32 *)(dst + i);
+
+				*dst32 = le32_to_cpu(*(__le32 *)(src + i));
+				i += 4;
+				if (i > bnxt_pcie_32b_entries[j].end &&
+				    j < ARRAY_SIZE(bnxt_pcie_32b_entries) - 1)
+					j++;
+			} else {
+				u64 *dst64 = (u64 *)(dst + i);
+
+				*dst64 = le64_to_cpu(*(__le64 *)(src + i));
+				i += 8;
+			}
+		}
+	}
+	hwrm_req_drop(bp, req);
 }
 
 static void bnxt_get_wol(struct net_device *dev, struct ethtool_wolinfo *wol)
@@ -882,136 +2192,628 @@ static int bnxt_set_wol(struct net_device *dev, struct ethtool_wolinfo *wol)
 	return 0;
 }
 
-u32 _bnxt_fw_to_ethtool_adv_spds(u16 fw_speeds, u8 fw_pause)
+/* TODO: support 25GB, 40GB, 50GB with different cable type */
+void _bnxt_fw_to_linkmode(unsigned long *mode, u16 fw_speeds)
 {
-	u32 speed_mask = 0;
+	linkmode_zero(mode);
 
-	/* TODO: support 25GB, 40GB, 50GB with different cable type */
-	/* set the advertised speeds */
 	if (fw_speeds & BNXT_LINK_SPEED_MSK_100MB)
-		speed_mask |= ADVERTISED_100baseT_Full;
+		linkmode_set_bit(ETHTOOL_LINK_MODE_100baseT_Full_BIT, mode);
 	if (fw_speeds & BNXT_LINK_SPEED_MSK_1GB)
-		speed_mask |= ADVERTISED_1000baseT_Full;
+		linkmode_set_bit(ETHTOOL_LINK_MODE_1000baseT_Full_BIT, mode);
 	if (fw_speeds & BNXT_LINK_SPEED_MSK_2_5GB)
-		speed_mask |= ADVERTISED_2500baseX_Full;
+		linkmode_set_bit(ETHTOOL_LINK_MODE_2500baseX_Full_BIT, mode);
 	if (fw_speeds & BNXT_LINK_SPEED_MSK_10GB)
-		speed_mask |= ADVERTISED_10000baseT_Full;
+		linkmode_set_bit(ETHTOOL_LINK_MODE_10000baseT_Full_BIT, mode);
 	if (fw_speeds & BNXT_LINK_SPEED_MSK_40GB)
-		speed_mask |= ADVERTISED_40000baseCR4_Full;
-
-	if ((fw_pause & BNXT_LINK_PAUSE_BOTH) == BNXT_LINK_PAUSE_BOTH)
-		speed_mask |= ADVERTISED_Pause;
-	else if (fw_pause & BNXT_LINK_PAUSE_TX)
-		speed_mask |= ADVERTISED_Asym_Pause;
-	else if (fw_pause & BNXT_LINK_PAUSE_RX)
-		speed_mask |= ADVERTISED_Pause | ADVERTISED_Asym_Pause;
-
-	return speed_mask;
-}
-
-#define BNXT_FW_TO_ETHTOOL_SPDS(fw_speeds, fw_pause, lk_ksettings, name)\
-{									\
-	if ((fw_speeds) & BNXT_LINK_SPEED_MSK_100MB)			\
-		ethtool_link_ksettings_add_link_mode(lk_ksettings, name,\
-						     100baseT_Full);	\
-	if ((fw_speeds) & BNXT_LINK_SPEED_MSK_1GB)			\
-		ethtool_link_ksettings_add_link_mode(lk_ksettings, name,\
-						     1000baseT_Full);	\
-	if ((fw_speeds) & BNXT_LINK_SPEED_MSK_10GB)			\
-		ethtool_link_ksettings_add_link_mode(lk_ksettings, name,\
-						     10000baseT_Full);	\
-	if ((fw_speeds) & BNXT_LINK_SPEED_MSK_25GB)			\
-		ethtool_link_ksettings_add_link_mode(lk_ksettings, name,\
-						     25000baseCR_Full);	\
-	if ((fw_speeds) & BNXT_LINK_SPEED_MSK_40GB)			\
-		ethtool_link_ksettings_add_link_mode(lk_ksettings, name,\
-						     40000baseCR4_Full);\
-	if ((fw_speeds) & BNXT_LINK_SPEED_MSK_50GB)			\
-		ethtool_link_ksettings_add_link_mode(lk_ksettings, name,\
-						     50000baseCR2_Full);\
-	if ((fw_speeds) & BNXT_LINK_SPEED_MSK_100GB)			\
-		ethtool_link_ksettings_add_link_mode(lk_ksettings, name,\
-						     100000baseCR4_Full);\
-	if ((fw_pause) & BNXT_LINK_PAUSE_RX) {				\
-		ethtool_link_ksettings_add_link_mode(lk_ksettings, name,\
-						     Pause);		\
-		if (!((fw_pause) & BNXT_LINK_PAUSE_TX))			\
-			ethtool_link_ksettings_add_link_mode(		\
-					lk_ksettings, name, Asym_Pause);\
-	} else if ((fw_pause) & BNXT_LINK_PAUSE_TX) {			\
-		ethtool_link_ksettings_add_link_mode(lk_ksettings, name,\
-						     Asym_Pause);	\
-	}								\
-}
-
-#define BNXT_ETHTOOL_TO_FW_SPDS(fw_speeds, lk_ksettings, name)		\
-{									\
-	if (ethtool_link_ksettings_test_link_mode(lk_ksettings, name,	\
-						  100baseT_Full) ||	\
-	    ethtool_link_ksettings_test_link_mode(lk_ksettings, name,	\
-						  100baseT_Half))	\
-		(fw_speeds) |= BNXT_LINK_SPEED_MSK_100MB;		\
-	if (ethtool_link_ksettings_test_link_mode(lk_ksettings, name,	\
-						  1000baseT_Full) ||	\
-	    ethtool_link_ksettings_test_link_mode(lk_ksettings, name,	\
-						  1000baseT_Half))	\
-		(fw_speeds) |= BNXT_LINK_SPEED_MSK_1GB;			\
-	if (ethtool_link_ksettings_test_link_mode(lk_ksettings, name,	\
-						  10000baseT_Full))	\
-		(fw_speeds) |= BNXT_LINK_SPEED_MSK_10GB;		\
-	if (ethtool_link_ksettings_test_link_mode(lk_ksettings, name,	\
-						  25000baseCR_Full))	\
-		(fw_speeds) |= BNXT_LINK_SPEED_MSK_25GB;		\
-	if (ethtool_link_ksettings_test_link_mode(lk_ksettings, name,	\
-						  40000baseCR4_Full))	\
-		(fw_speeds) |= BNXT_LINK_SPEED_MSK_40GB;		\
-	if (ethtool_link_ksettings_test_link_mode(lk_ksettings, name,	\
-						  50000baseCR2_Full))	\
-		(fw_speeds) |= BNXT_LINK_SPEED_MSK_50GB;		\
-	if (ethtool_link_ksettings_test_link_mode(lk_ksettings, name,	\
-						  100000baseCR4_Full))	\
-		(fw_speeds) |= BNXT_LINK_SPEED_MSK_100GB;		\
-}
-
-static void bnxt_fw_to_ethtool_advertised_spds(struct bnxt_link_info *link_info,
-				struct ethtool_link_ksettings *lk_ksettings)
+		linkmode_set_bit(ETHTOOL_LINK_MODE_40000baseCR4_Full_BIT, mode);
+}
+
+enum bnxt_media_type {
+	BNXT_MEDIA_UNKNOWN = 0,
+	BNXT_MEDIA_TP,
+	BNXT_MEDIA_CR,
+	BNXT_MEDIA_SR,
+	BNXT_MEDIA_LR_ER_FR,
+	BNXT_MEDIA_KR,
+	BNXT_MEDIA_KX,
+	BNXT_MEDIA_X,
+	__BNXT_MEDIA_END,
+};
+
+static const enum bnxt_media_type bnxt_phy_types[] = {
+	[PORT_PHY_QCFG_RESP_PHY_TYPE_BASECR] = BNXT_MEDIA_CR,
+	[PORT_PHY_QCFG_RESP_PHY_TYPE_BASEKR4] =  BNXT_MEDIA_KR,
+	[PORT_PHY_QCFG_RESP_PHY_TYPE_BASELR] = BNXT_MEDIA_LR_ER_FR,
+	[PORT_PHY_QCFG_RESP_PHY_TYPE_BASESR] = BNXT_MEDIA_SR,
+	[PORT_PHY_QCFG_RESP_PHY_TYPE_BASEKR2] = BNXT_MEDIA_KR,
+	[PORT_PHY_QCFG_RESP_PHY_TYPE_BASEKX] = BNXT_MEDIA_KX,
+	[PORT_PHY_QCFG_RESP_PHY_TYPE_BASEKR] = BNXT_MEDIA_KR,
+	[PORT_PHY_QCFG_RESP_PHY_TYPE_BASET] = BNXT_MEDIA_TP,
+	[PORT_PHY_QCFG_RESP_PHY_TYPE_BASETE] = BNXT_MEDIA_TP,
+	[PORT_PHY_QCFG_RESP_PHY_TYPE_25G_BASECR_CA_L] = BNXT_MEDIA_CR,
+	[PORT_PHY_QCFG_RESP_PHY_TYPE_25G_BASECR_CA_S] = BNXT_MEDIA_CR,
+	[PORT_PHY_QCFG_RESP_PHY_TYPE_25G_BASECR_CA_N] = BNXT_MEDIA_CR,
+	[PORT_PHY_QCFG_RESP_PHY_TYPE_25G_BASESR] = BNXT_MEDIA_SR,
+	[PORT_PHY_QCFG_RESP_PHY_TYPE_100G_BASECR4] = BNXT_MEDIA_CR,
+	[PORT_PHY_QCFG_RESP_PHY_TYPE_100G_BASESR4] = BNXT_MEDIA_SR,
+	[PORT_PHY_QCFG_RESP_PHY_TYPE_100G_BASELR4] = BNXT_MEDIA_LR_ER_FR,
+	[PORT_PHY_QCFG_RESP_PHY_TYPE_100G_BASEER4] = BNXT_MEDIA_LR_ER_FR,
+	[PORT_PHY_QCFG_RESP_PHY_TYPE_100G_BASESR10] = BNXT_MEDIA_SR,
+	[PORT_PHY_QCFG_RESP_PHY_TYPE_40G_BASECR4] = BNXT_MEDIA_CR,
+	[PORT_PHY_QCFG_RESP_PHY_TYPE_40G_BASESR4] = BNXT_MEDIA_SR,
+	[PORT_PHY_QCFG_RESP_PHY_TYPE_40G_BASELR4] = BNXT_MEDIA_LR_ER_FR,
+	[PORT_PHY_QCFG_RESP_PHY_TYPE_40G_BASEER4] = BNXT_MEDIA_LR_ER_FR,
+	[PORT_PHY_QCFG_RESP_PHY_TYPE_40G_ACTIVE_CABLE] = BNXT_MEDIA_SR,
+	[PORT_PHY_QCFG_RESP_PHY_TYPE_1G_BASET] = BNXT_MEDIA_TP,
+	[PORT_PHY_QCFG_RESP_PHY_TYPE_1G_BASESX] = BNXT_MEDIA_X,
+	[PORT_PHY_QCFG_RESP_PHY_TYPE_1G_BASECX] = BNXT_MEDIA_X,
+	[PORT_PHY_QCFG_RESP_PHY_TYPE_200G_BASECR4] = BNXT_MEDIA_CR,
+	[PORT_PHY_QCFG_RESP_PHY_TYPE_200G_BASESR4] = BNXT_MEDIA_SR,
+	[PORT_PHY_QCFG_RESP_PHY_TYPE_200G_BASELR4] = BNXT_MEDIA_LR_ER_FR,
+	[PORT_PHY_QCFG_RESP_PHY_TYPE_200G_BASEER4] = BNXT_MEDIA_LR_ER_FR,
+	[PORT_PHY_QCFG_RESP_PHY_TYPE_50G_BASECR] = BNXT_MEDIA_CR,
+	[PORT_PHY_QCFG_RESP_PHY_TYPE_50G_BASESR] = BNXT_MEDIA_SR,
+	[PORT_PHY_QCFG_RESP_PHY_TYPE_50G_BASELR] = BNXT_MEDIA_LR_ER_FR,
+	[PORT_PHY_QCFG_RESP_PHY_TYPE_50G_BASEER] = BNXT_MEDIA_LR_ER_FR,
+	[PORT_PHY_QCFG_RESP_PHY_TYPE_100G_BASECR2] = BNXT_MEDIA_CR,
+	[PORT_PHY_QCFG_RESP_PHY_TYPE_100G_BASESR2] = BNXT_MEDIA_SR,
+	[PORT_PHY_QCFG_RESP_PHY_TYPE_100G_BASELR2] = BNXT_MEDIA_LR_ER_FR,
+	[PORT_PHY_QCFG_RESP_PHY_TYPE_100G_BASEER2] = BNXT_MEDIA_LR_ER_FR,
+	[PORT_PHY_QCFG_RESP_PHY_TYPE_100G_BASECR] = BNXT_MEDIA_CR,
+	[PORT_PHY_QCFG_RESP_PHY_TYPE_100G_BASESR] = BNXT_MEDIA_SR,
+	[PORT_PHY_QCFG_RESP_PHY_TYPE_100G_BASELR] = BNXT_MEDIA_LR_ER_FR,
+	[PORT_PHY_QCFG_RESP_PHY_TYPE_100G_BASEER] = BNXT_MEDIA_LR_ER_FR,
+	[PORT_PHY_QCFG_RESP_PHY_TYPE_200G_BASECR2] = BNXT_MEDIA_CR,
+	[PORT_PHY_QCFG_RESP_PHY_TYPE_200G_BASESR2] = BNXT_MEDIA_SR,
+	[PORT_PHY_QCFG_RESP_PHY_TYPE_200G_BASELR2] = BNXT_MEDIA_LR_ER_FR,
+	[PORT_PHY_QCFG_RESP_PHY_TYPE_200G_BASEER2] = BNXT_MEDIA_LR_ER_FR,
+	[PORT_PHY_QCFG_RESP_PHY_TYPE_400G_BASECR8] = BNXT_MEDIA_CR,
+	[PORT_PHY_QCFG_RESP_PHY_TYPE_400G_BASESR8] = BNXT_MEDIA_SR,
+	[PORT_PHY_QCFG_RESP_PHY_TYPE_400G_BASELR8] = BNXT_MEDIA_LR_ER_FR,
+	[PORT_PHY_QCFG_RESP_PHY_TYPE_400G_BASEER8] = BNXT_MEDIA_LR_ER_FR,
+	[PORT_PHY_QCFG_RESP_PHY_TYPE_400G_BASECR4] = BNXT_MEDIA_CR,
+	[PORT_PHY_QCFG_RESP_PHY_TYPE_400G_BASESR4] = BNXT_MEDIA_SR,
+	[PORT_PHY_QCFG_RESP_PHY_TYPE_400G_BASELR4] = BNXT_MEDIA_LR_ER_FR,
+	[PORT_PHY_QCFG_RESP_PHY_TYPE_400G_BASEER4] = BNXT_MEDIA_LR_ER_FR,
+};
+
+static enum bnxt_media_type
+bnxt_get_media(struct bnxt_link_info *link_info)
+{
+	switch (link_info->media_type) {
+	case PORT_PHY_QCFG_RESP_MEDIA_TYPE_TP:
+		return BNXT_MEDIA_TP;
+	case PORT_PHY_QCFG_RESP_MEDIA_TYPE_DAC:
+		return BNXT_MEDIA_CR;
+	default:
+		if (link_info->phy_type < ARRAY_SIZE(bnxt_phy_types))
+			return bnxt_phy_types[link_info->phy_type];
+		return BNXT_MEDIA_UNKNOWN;
+	}
+}
+
+enum bnxt_link_speed_indices {
+	BNXT_LINK_SPEED_UNKNOWN = 0,
+	BNXT_LINK_SPEED_100MB_IDX,
+	BNXT_LINK_SPEED_1GB_IDX,
+	BNXT_LINK_SPEED_10GB_IDX,
+	BNXT_LINK_SPEED_25GB_IDX,
+	BNXT_LINK_SPEED_40GB_IDX,
+	BNXT_LINK_SPEED_50GB_IDX,
+	BNXT_LINK_SPEED_100GB_IDX,
+	BNXT_LINK_SPEED_200GB_IDX,
+	BNXT_LINK_SPEED_400GB_IDX,
+	__BNXT_LINK_SPEED_END
+};
+
+static enum bnxt_link_speed_indices bnxt_fw_speed_idx(u16 speed)
+{
+	switch (speed) {
+	case BNXT_LINK_SPEED_100MB: return BNXT_LINK_SPEED_100MB_IDX;
+	case BNXT_LINK_SPEED_1GB: return BNXT_LINK_SPEED_1GB_IDX;
+	case BNXT_LINK_SPEED_10GB: return BNXT_LINK_SPEED_10GB_IDX;
+	case BNXT_LINK_SPEED_25GB: return BNXT_LINK_SPEED_25GB_IDX;
+	case BNXT_LINK_SPEED_40GB: return BNXT_LINK_SPEED_40GB_IDX;
+	case BNXT_LINK_SPEED_50GB:
+	case BNXT_LINK_SPEED_50GB_PAM4:
+		return BNXT_LINK_SPEED_50GB_IDX;
+	case BNXT_LINK_SPEED_100GB:
+	case BNXT_LINK_SPEED_100GB_PAM4:
+	case BNXT_LINK_SPEED_100GB_PAM4_112:
+		return BNXT_LINK_SPEED_100GB_IDX;
+	case BNXT_LINK_SPEED_200GB:
+	case BNXT_LINK_SPEED_200GB_PAM4:
+	case BNXT_LINK_SPEED_200GB_PAM4_112:
+		return BNXT_LINK_SPEED_200GB_IDX;
+	case BNXT_LINK_SPEED_400GB:
+	case BNXT_LINK_SPEED_400GB_PAM4:
+	case BNXT_LINK_SPEED_400GB_PAM4_112:
+		return BNXT_LINK_SPEED_400GB_IDX;
+	default: return BNXT_LINK_SPEED_UNKNOWN;
+	}
+}
+
+static const enum ethtool_link_mode_bit_indices
+bnxt_link_modes[__BNXT_LINK_SPEED_END][BNXT_SIG_MODE_MAX][__BNXT_MEDIA_END] = {
+	[BNXT_LINK_SPEED_100MB_IDX] = {
+		{
+			[BNXT_MEDIA_TP] = ETHTOOL_LINK_MODE_100baseT_Full_BIT,
+		},
+	},
+	[BNXT_LINK_SPEED_1GB_IDX] = {
+		{
+			[BNXT_MEDIA_TP] = ETHTOOL_LINK_MODE_1000baseT_Full_BIT,
+			/* historically baseT, but DAC is more correctly baseX */
+			[BNXT_MEDIA_CR] = ETHTOOL_LINK_MODE_1000baseX_Full_BIT,
+			[BNXT_MEDIA_KX] = ETHTOOL_LINK_MODE_1000baseKX_Full_BIT,
+			[BNXT_MEDIA_X] = ETHTOOL_LINK_MODE_1000baseX_Full_BIT,
+			[BNXT_MEDIA_KR] = ETHTOOL_LINK_MODE_1000baseKX_Full_BIT,
+		},
+	},
+	[BNXT_LINK_SPEED_10GB_IDX] = {
+		{
+			[BNXT_MEDIA_TP] = ETHTOOL_LINK_MODE_10000baseT_Full_BIT,
+			[BNXT_MEDIA_CR] = ETHTOOL_LINK_MODE_10000baseCR_Full_BIT,
+			[BNXT_MEDIA_SR] = ETHTOOL_LINK_MODE_10000baseSR_Full_BIT,
+			[BNXT_MEDIA_LR_ER_FR] = ETHTOOL_LINK_MODE_10000baseLR_Full_BIT,
+			[BNXT_MEDIA_KR] = ETHTOOL_LINK_MODE_10000baseKR_Full_BIT,
+			[BNXT_MEDIA_KX] = ETHTOOL_LINK_MODE_10000baseKX4_Full_BIT,
+		},
+	},
+	[BNXT_LINK_SPEED_25GB_IDX] = {
+		{
+			[BNXT_MEDIA_CR] = ETHTOOL_LINK_MODE_25000baseCR_Full_BIT,
+			[BNXT_MEDIA_SR] = ETHTOOL_LINK_MODE_25000baseSR_Full_BIT,
+			[BNXT_MEDIA_KR] = ETHTOOL_LINK_MODE_25000baseKR_Full_BIT,
+		},
+	},
+	[BNXT_LINK_SPEED_40GB_IDX] = {
+		{
+			[BNXT_MEDIA_CR] = ETHTOOL_LINK_MODE_40000baseCR4_Full_BIT,
+			[BNXT_MEDIA_SR] = ETHTOOL_LINK_MODE_40000baseSR4_Full_BIT,
+			[BNXT_MEDIA_LR_ER_FR] = ETHTOOL_LINK_MODE_40000baseLR4_Full_BIT,
+			[BNXT_MEDIA_KR] = ETHTOOL_LINK_MODE_40000baseKR4_Full_BIT,
+		},
+	},
+	[BNXT_LINK_SPEED_50GB_IDX] = {
+		[BNXT_SIG_MODE_NRZ] = {
+			[BNXT_MEDIA_CR] = ETHTOOL_LINK_MODE_50000baseCR2_Full_BIT,
+			[BNXT_MEDIA_SR] = ETHTOOL_LINK_MODE_50000baseSR2_Full_BIT,
+			[BNXT_MEDIA_KR] = ETHTOOL_LINK_MODE_50000baseKR2_Full_BIT,
+		},
+		[BNXT_SIG_MODE_PAM4] = {
+			[BNXT_MEDIA_CR] = ETHTOOL_LINK_MODE_50000baseCR_Full_BIT,
+			[BNXT_MEDIA_SR] = ETHTOOL_LINK_MODE_50000baseSR_Full_BIT,
+			[BNXT_MEDIA_LR_ER_FR] = ETHTOOL_LINK_MODE_50000baseLR_ER_FR_Full_BIT,
+			[BNXT_MEDIA_KR] = ETHTOOL_LINK_MODE_50000baseKR_Full_BIT,
+		},
+	},
+	[BNXT_LINK_SPEED_100GB_IDX] = {
+		[BNXT_SIG_MODE_NRZ] = {
+			[BNXT_MEDIA_CR] = ETHTOOL_LINK_MODE_100000baseCR4_Full_BIT,
+			[BNXT_MEDIA_SR] = ETHTOOL_LINK_MODE_100000baseSR4_Full_BIT,
+			[BNXT_MEDIA_LR_ER_FR] = ETHTOOL_LINK_MODE_100000baseLR4_ER4_Full_BIT,
+			[BNXT_MEDIA_KR] = ETHTOOL_LINK_MODE_100000baseKR4_Full_BIT,
+		},
+		[BNXT_SIG_MODE_PAM4] = {
+			[BNXT_MEDIA_CR] = ETHTOOL_LINK_MODE_100000baseCR2_Full_BIT,
+			[BNXT_MEDIA_SR] = ETHTOOL_LINK_MODE_100000baseSR2_Full_BIT,
+			[BNXT_MEDIA_LR_ER_FR] = ETHTOOL_LINK_MODE_100000baseLR2_ER2_FR2_Full_BIT,
+			[BNXT_MEDIA_KR] = ETHTOOL_LINK_MODE_100000baseKR2_Full_BIT,
+		},
+		[BNXT_SIG_MODE_PAM4_112] = {
+			[BNXT_MEDIA_CR] = ETHTOOL_LINK_MODE_100000baseCR_Full_BIT,
+			[BNXT_MEDIA_SR] = ETHTOOL_LINK_MODE_100000baseSR_Full_BIT,
+			[BNXT_MEDIA_KR] = ETHTOOL_LINK_MODE_100000baseKR_Full_BIT,
+			[BNXT_MEDIA_LR_ER_FR] = ETHTOOL_LINK_MODE_100000baseLR_ER_FR_Full_BIT,
+		},
+	},
+	[BNXT_LINK_SPEED_200GB_IDX] = {
+		[BNXT_SIG_MODE_PAM4] = {
+			[BNXT_MEDIA_CR] = ETHTOOL_LINK_MODE_200000baseCR4_Full_BIT,
+			[BNXT_MEDIA_SR] = ETHTOOL_LINK_MODE_200000baseSR4_Full_BIT,
+			[BNXT_MEDIA_LR_ER_FR] = ETHTOOL_LINK_MODE_200000baseLR4_ER4_FR4_Full_BIT,
+			[BNXT_MEDIA_KR] = ETHTOOL_LINK_MODE_200000baseKR4_Full_BIT,
+		},
+		[BNXT_SIG_MODE_PAM4_112] = {
+			[BNXT_MEDIA_CR] = ETHTOOL_LINK_MODE_200000baseCR2_Full_BIT,
+			[BNXT_MEDIA_KR] = ETHTOOL_LINK_MODE_200000baseKR2_Full_BIT,
+			[BNXT_MEDIA_SR] = ETHTOOL_LINK_MODE_200000baseSR2_Full_BIT,
+			[BNXT_MEDIA_LR_ER_FR] = ETHTOOL_LINK_MODE_200000baseLR2_ER2_FR2_Full_BIT,
+		},
+	},
+	[BNXT_LINK_SPEED_400GB_IDX] = {
+		[BNXT_SIG_MODE_PAM4] = {
+			[BNXT_MEDIA_CR] = ETHTOOL_LINK_MODE_400000baseCR8_Full_BIT,
+			[BNXT_MEDIA_KR] = ETHTOOL_LINK_MODE_400000baseKR8_Full_BIT,
+			[BNXT_MEDIA_SR] = ETHTOOL_LINK_MODE_400000baseSR8_Full_BIT,
+			[BNXT_MEDIA_LR_ER_FR] = ETHTOOL_LINK_MODE_400000baseLR8_ER8_FR8_Full_BIT,
+		},
+		[BNXT_SIG_MODE_PAM4_112] = {
+			[BNXT_MEDIA_CR] = ETHTOOL_LINK_MODE_400000baseCR4_Full_BIT,
+			[BNXT_MEDIA_KR] = ETHTOOL_LINK_MODE_400000baseKR4_Full_BIT,
+			[BNXT_MEDIA_SR] = ETHTOOL_LINK_MODE_400000baseSR4_Full_BIT,
+			[BNXT_MEDIA_LR_ER_FR] = ETHTOOL_LINK_MODE_400000baseLR4_ER4_FR4_Full_BIT,
+		},
+	},
+};
+
+#define BNXT_LINK_MODE_UNKNOWN -1
+
+static enum ethtool_link_mode_bit_indices
+bnxt_get_link_mode(struct bnxt_link_info *link_info)
+{
+	enum ethtool_link_mode_bit_indices link_mode;
+	enum bnxt_link_speed_indices speed;
+	enum bnxt_media_type media;
+	u8 sig_mode;
+
+	if (link_info->phy_link_status != BNXT_LINK_LINK)
+		return BNXT_LINK_MODE_UNKNOWN;
+
+	media = bnxt_get_media(link_info);
+	if (BNXT_AUTO_MODE(link_info->auto_mode)) {
+		speed = bnxt_fw_speed_idx(link_info->link_speed);
+		sig_mode = link_info->active_fec_sig_mode &
+			PORT_PHY_QCFG_RESP_SIGNAL_MODE_MASK;
+	} else {
+		speed = bnxt_fw_speed_idx(link_info->req_link_speed);
+		sig_mode = link_info->req_signal_mode;
+	}
+	if (sig_mode >= BNXT_SIG_MODE_MAX)
+		return BNXT_LINK_MODE_UNKNOWN;
+
+	/* Note ETHTOOL_LINK_MODE_10baseT_Half_BIT == 0 is a legal Linux
+	 * link mode, but since no such devices exist, the zeroes in the
+	 * map can be conveniently used to represent unknown link modes.
+	 */
+	link_mode = bnxt_link_modes[speed][sig_mode][media];
+	if (!link_mode)
+		return BNXT_LINK_MODE_UNKNOWN;
+
+	switch (link_mode) {
+	case ETHTOOL_LINK_MODE_100baseT_Full_BIT:
+		if (~link_info->duplex & BNXT_LINK_DUPLEX_FULL)
+			link_mode = ETHTOOL_LINK_MODE_100baseT_Half_BIT;
+		break;
+	case ETHTOOL_LINK_MODE_1000baseT_Full_BIT:
+		if (~link_info->duplex & BNXT_LINK_DUPLEX_FULL)
+			link_mode = ETHTOOL_LINK_MODE_1000baseT_Half_BIT;
+		break;
+	default:
+		break;
+	}
+
+	return link_mode;
+}
+
+static void bnxt_get_ethtool_modes(struct bnxt_link_info *link_info,
+				   struct ethtool_link_ksettings *lk_ksettings)
 {
-	u16 fw_speeds = link_info->advertising;
-	u8 fw_pause = 0;
+	struct bnxt *bp = container_of(link_info, struct bnxt, link_info);
 
-	if (link_info->autoneg & BNXT_AUTONEG_FLOW_CTRL)
-		fw_pause = link_info->auto_pause_setting;
+	if (!(bp->phy_flags & BNXT_PHY_FL_NO_PAUSE)) {
+		linkmode_set_bit(ETHTOOL_LINK_MODE_Pause_BIT,
+				 lk_ksettings->link_modes.supported);
+		linkmode_set_bit(ETHTOOL_LINK_MODE_Asym_Pause_BIT,
+				 lk_ksettings->link_modes.supported);
+	}
+
+	if (link_info->support_auto_speeds || link_info->support_auto_speeds2 ||
+	    link_info->support_pam4_auto_speeds)
+		linkmode_set_bit(ETHTOOL_LINK_MODE_Autoneg_BIT,
+				 lk_ksettings->link_modes.supported);
 
-	BNXT_FW_TO_ETHTOOL_SPDS(fw_speeds, fw_pause, lk_ksettings, advertising);
+	if (~link_info->autoneg & BNXT_AUTONEG_FLOW_CTRL)
+		return;
+
+	if (link_info->auto_pause_setting & BNXT_LINK_PAUSE_RX)
+		linkmode_set_bit(ETHTOOL_LINK_MODE_Pause_BIT,
+				 lk_ksettings->link_modes.advertising);
+	if (hweight8(link_info->auto_pause_setting & BNXT_LINK_PAUSE_BOTH) == 1)
+		linkmode_set_bit(ETHTOOL_LINK_MODE_Asym_Pause_BIT,
+				 lk_ksettings->link_modes.advertising);
+	if (link_info->lp_pause & BNXT_LINK_PAUSE_RX)
+		linkmode_set_bit(ETHTOOL_LINK_MODE_Pause_BIT,
+				 lk_ksettings->link_modes.lp_advertising);
+	if (hweight8(link_info->lp_pause & BNXT_LINK_PAUSE_BOTH) == 1)
+		linkmode_set_bit(ETHTOOL_LINK_MODE_Asym_Pause_BIT,
+				 lk_ksettings->link_modes.lp_advertising);
 }
 
-static void bnxt_fw_to_ethtool_lp_adv(struct bnxt_link_info *link_info,
+static const u16 bnxt_nrz_speed_masks[] = {
+	[BNXT_LINK_SPEED_100MB_IDX] = BNXT_LINK_SPEED_MSK_100MB,
+	[BNXT_LINK_SPEED_1GB_IDX] = BNXT_LINK_SPEED_MSK_1GB,
+	[BNXT_LINK_SPEED_10GB_IDX] = BNXT_LINK_SPEED_MSK_10GB,
+	[BNXT_LINK_SPEED_25GB_IDX] = BNXT_LINK_SPEED_MSK_25GB,
+	[BNXT_LINK_SPEED_40GB_IDX] = BNXT_LINK_SPEED_MSK_40GB,
+	[BNXT_LINK_SPEED_50GB_IDX] = BNXT_LINK_SPEED_MSK_50GB,
+	[BNXT_LINK_SPEED_100GB_IDX] = BNXT_LINK_SPEED_MSK_100GB,
+	[__BNXT_LINK_SPEED_END - 1] = 0 /* make any legal speed a valid index */
+};
+
+static const u16 bnxt_pam4_speed_masks[] = {
+	[BNXT_LINK_SPEED_50GB_IDX] = BNXT_LINK_PAM4_SPEED_MSK_50GB,
+	[BNXT_LINK_SPEED_100GB_IDX] = BNXT_LINK_PAM4_SPEED_MSK_100GB,
+	[BNXT_LINK_SPEED_200GB_IDX] = BNXT_LINK_PAM4_SPEED_MSK_200GB,
+	[__BNXT_LINK_SPEED_END - 1] = 0 /* make any legal speed a valid index */
+};
+
+static const u16 bnxt_nrz_speeds2_masks[] = {
+	[BNXT_LINK_SPEED_1GB_IDX] = BNXT_LINK_SPEEDS2_MSK_1GB,
+	[BNXT_LINK_SPEED_10GB_IDX] = BNXT_LINK_SPEEDS2_MSK_10GB,
+	[BNXT_LINK_SPEED_25GB_IDX] = BNXT_LINK_SPEEDS2_MSK_25GB,
+	[BNXT_LINK_SPEED_40GB_IDX] = BNXT_LINK_SPEEDS2_MSK_40GB,
+	[BNXT_LINK_SPEED_50GB_IDX] = BNXT_LINK_SPEEDS2_MSK_50GB,
+	[BNXT_LINK_SPEED_100GB_IDX] = BNXT_LINK_SPEEDS2_MSK_100GB,
+	[__BNXT_LINK_SPEED_END - 1] = 0 /* make any legal speed a valid index */
+};
+
+static const u16 bnxt_pam4_speeds2_masks[] = {
+	[BNXT_LINK_SPEED_50GB_IDX] = BNXT_LINK_SPEEDS2_MSK_50GB_PAM4,
+	[BNXT_LINK_SPEED_100GB_IDX] = BNXT_LINK_SPEEDS2_MSK_100GB_PAM4,
+	[BNXT_LINK_SPEED_200GB_IDX] = BNXT_LINK_SPEEDS2_MSK_200GB_PAM4,
+	[BNXT_LINK_SPEED_400GB_IDX] = BNXT_LINK_SPEEDS2_MSK_400GB_PAM4,
+};
+
+static const u16 bnxt_pam4_112_speeds2_masks[] = {
+	[BNXT_LINK_SPEED_100GB_IDX] = BNXT_LINK_SPEEDS2_MSK_100GB_PAM4_112,
+	[BNXT_LINK_SPEED_200GB_IDX] = BNXT_LINK_SPEEDS2_MSK_200GB_PAM4_112,
+	[BNXT_LINK_SPEED_400GB_IDX] = BNXT_LINK_SPEEDS2_MSK_400GB_PAM4_112,
+};
+
+static enum bnxt_link_speed_indices
+bnxt_encoding_speed_idx(u8 sig_mode, u16 phy_flags, u16 speed_msk)
+{
+	const u16 *speeds;
+	int idx, len;
+
+	switch (sig_mode) {
+	case BNXT_SIG_MODE_NRZ:
+		if (phy_flags & BNXT_PHY_FL_SPEEDS2) {
+			speeds = bnxt_nrz_speeds2_masks;
+			len = ARRAY_SIZE(bnxt_nrz_speeds2_masks);
+		} else {
+			speeds = bnxt_nrz_speed_masks;
+			len = ARRAY_SIZE(bnxt_nrz_speed_masks);
+		}
+		break;
+	case BNXT_SIG_MODE_PAM4:
+		if (phy_flags & BNXT_PHY_FL_SPEEDS2) {
+			speeds = bnxt_pam4_speeds2_masks;
+			len = ARRAY_SIZE(bnxt_pam4_speeds2_masks);
+		} else {
+			speeds = bnxt_pam4_speed_masks;
+			len = ARRAY_SIZE(bnxt_pam4_speed_masks);
+		}
+		break;
+	case BNXT_SIG_MODE_PAM4_112:
+		speeds = bnxt_pam4_112_speeds2_masks;
+		len = ARRAY_SIZE(bnxt_pam4_112_speeds2_masks);
+		break;
+	default:
+		return BNXT_LINK_SPEED_UNKNOWN;
+	}
+
+	for (idx = 0; idx < len; idx++) {
+		if (speeds[idx] == speed_msk)
+			return idx;
+	}
+
+	return BNXT_LINK_SPEED_UNKNOWN;
+}
+
+#define BNXT_FW_SPEED_MSK_BITS 16
+
+static void
+__bnxt_get_ethtool_speeds(unsigned long fw_mask, enum bnxt_media_type media,
+			  u8 sig_mode, u16 phy_flags, unsigned long *et_mask)
+{
+	enum ethtool_link_mode_bit_indices link_mode;
+	enum bnxt_link_speed_indices speed;
+	u8 bit;
+
+	for_each_set_bit(bit, &fw_mask, BNXT_FW_SPEED_MSK_BITS) {
+		speed = bnxt_encoding_speed_idx(sig_mode, phy_flags, 1 << bit);
+		if (!speed)
+			continue;
+
+		link_mode = bnxt_link_modes[speed][sig_mode][media];
+		if (!link_mode)
+			continue;
+
+		linkmode_set_bit(link_mode, et_mask);
+	}
+}
+
+static void
+bnxt_get_ethtool_speeds(unsigned long fw_mask, enum bnxt_media_type media,
+			u8 sig_mode, u16 phy_flags, unsigned long *et_mask)
+{
+	if (media) {
+		__bnxt_get_ethtool_speeds(fw_mask, media, sig_mode, phy_flags,
+					  et_mask);
+		return;
+	}
+
+	/* list speeds for all media if unknown */
+	for (media = 1; media < __BNXT_MEDIA_END; media++)
+		__bnxt_get_ethtool_speeds(fw_mask, media, sig_mode, phy_flags,
+					  et_mask);
+}
+
+static void
+bnxt_get_all_ethtool_support_speeds(struct bnxt_link_info *link_info,
+				    enum bnxt_media_type media,
+				    struct ethtool_link_ksettings *lk_ksettings)
+{
+	struct bnxt *bp = container_of(link_info, struct bnxt, link_info);
+	u16 sp_nrz, sp_pam4, sp_pam4_112 = 0;
+	u16 phy_flags = bp->phy_flags;
+
+	if (phy_flags & BNXT_PHY_FL_SPEEDS2) {
+		sp_nrz = link_info->support_speeds2;
+		sp_pam4 = link_info->support_speeds2;
+		sp_pam4_112 = link_info->support_speeds2;
+	} else {
+		sp_nrz = link_info->support_speeds;
+		sp_pam4 = link_info->support_pam4_speeds;
+	}
+	bnxt_get_ethtool_speeds(sp_nrz, media, BNXT_SIG_MODE_NRZ, phy_flags,
+				lk_ksettings->link_modes.supported);
+	bnxt_get_ethtool_speeds(sp_pam4, media, BNXT_SIG_MODE_PAM4, phy_flags,
+				lk_ksettings->link_modes.supported);
+	bnxt_get_ethtool_speeds(sp_pam4_112, media, BNXT_SIG_MODE_PAM4_112,
+				phy_flags, lk_ksettings->link_modes.supported);
+}
+
+static void
+bnxt_get_all_ethtool_adv_speeds(struct bnxt_link_info *link_info,
+				enum bnxt_media_type media,
 				struct ethtool_link_ksettings *lk_ksettings)
 {
-	u16 fw_speeds = link_info->lp_auto_link_speeds;
-	u8 fw_pause = 0;
+	struct bnxt *bp = container_of(link_info, struct bnxt, link_info);
+	u16 sp_nrz, sp_pam4, sp_pam4_112 = 0;
+	u16 phy_flags = bp->phy_flags;
+
+	sp_nrz = link_info->advertising;
+	if (phy_flags & BNXT_PHY_FL_SPEEDS2) {
+		sp_pam4 = link_info->advertising;
+		sp_pam4_112 = link_info->advertising;
+	} else {
+		sp_pam4 = link_info->advertising_pam4;
+	}
+	bnxt_get_ethtool_speeds(sp_nrz, media, BNXT_SIG_MODE_NRZ, phy_flags,
+				lk_ksettings->link_modes.advertising);
+	bnxt_get_ethtool_speeds(sp_pam4, media, BNXT_SIG_MODE_PAM4, phy_flags,
+				lk_ksettings->link_modes.advertising);
+	bnxt_get_ethtool_speeds(sp_pam4_112, media, BNXT_SIG_MODE_PAM4_112,
+				phy_flags, lk_ksettings->link_modes.advertising);
+}
+
+static void
+bnxt_get_all_ethtool_lp_speeds(struct bnxt_link_info *link_info,
+			       enum bnxt_media_type media,
+			       struct ethtool_link_ksettings *lk_ksettings)
+{
+	struct bnxt *bp = container_of(link_info, struct bnxt, link_info);
+	u16 phy_flags = bp->phy_flags;
+
+	bnxt_get_ethtool_speeds(link_info->lp_auto_link_speeds, media,
+				BNXT_SIG_MODE_NRZ, phy_flags,
+				lk_ksettings->link_modes.lp_advertising);
+	bnxt_get_ethtool_speeds(link_info->lp_auto_pam4_link_speeds, media,
+				BNXT_SIG_MODE_PAM4, phy_flags,
+				lk_ksettings->link_modes.lp_advertising);
+}
 
-	if (link_info->autoneg & BNXT_AUTONEG_FLOW_CTRL)
-		fw_pause = link_info->lp_pause;
+static void bnxt_update_speed(u32 *delta, bool installed_media, u16 *speeds,
+			      u16 speed_msk, const unsigned long *et_mask,
+			      enum ethtool_link_mode_bit_indices mode)
+{
+	bool mode_desired = linkmode_test_bit(mode, et_mask);
 
-	BNXT_FW_TO_ETHTOOL_SPDS(fw_speeds, fw_pause, lk_ksettings,
-				lp_advertising);
+	if (!mode)
+		return;
+
+	/* enabled speeds for installed media should override */
+	if (installed_media && mode_desired) {
+		*speeds |= speed_msk;
+		*delta |= speed_msk;
+		return;
+	}
+
+	/* many to one mapping, only allow one change per fw_speed bit */
+	if (!(*delta & speed_msk) && (mode_desired == !(*speeds & speed_msk))) {
+		*speeds ^= speed_msk;
+		*delta |= speed_msk;
+	}
 }
 
-static void bnxt_fw_to_ethtool_support_spds(struct bnxt_link_info *link_info,
+static void bnxt_set_ethtool_speeds(struct bnxt_link_info *link_info,
+				    const unsigned long *et_mask)
+{
+	struct bnxt *bp = container_of(link_info, struct bnxt, link_info);
+	u16 const *sp_msks, *sp_pam4_msks, *sp_pam4_112_msks;
+	enum bnxt_media_type media = bnxt_get_media(link_info);
+	u16 *adv, *adv_pam4, *adv_pam4_112 = NULL;
+	u32 delta_pam4_112 = 0;
+	u32 delta_pam4 = 0;
+	u32 delta_nrz = 0;
+	int i, m;
+
+	adv = &link_info->advertising;
+	if (bp->phy_flags & BNXT_PHY_FL_SPEEDS2) {
+		adv_pam4 = &link_info->advertising;
+		adv_pam4_112 = &link_info->advertising;
+		sp_msks = bnxt_nrz_speeds2_masks;
+		sp_pam4_msks = bnxt_pam4_speeds2_masks;
+		sp_pam4_112_msks = bnxt_pam4_112_speeds2_masks;
+	} else {
+		adv_pam4 = &link_info->advertising_pam4;
+		sp_msks = bnxt_nrz_speed_masks;
+		sp_pam4_msks = bnxt_pam4_speed_masks;
+	}
+	for (i = 1; i < __BNXT_LINK_SPEED_END; i++) {
+		/* accept any legal media from user */
+		for (m = 1; m < __BNXT_MEDIA_END; m++) {
+			bnxt_update_speed(&delta_nrz, m == media,
+					  adv, sp_msks[i], et_mask,
+					  bnxt_link_modes[i][BNXT_SIG_MODE_NRZ][m]);
+			bnxt_update_speed(&delta_pam4, m == media,
+					  adv_pam4, sp_pam4_msks[i], et_mask,
+					  bnxt_link_modes[i][BNXT_SIG_MODE_PAM4][m]);
+			if (!adv_pam4_112)
+				continue;
+
+			bnxt_update_speed(&delta_pam4_112, m == media,
+					  adv_pam4_112, sp_pam4_112_msks[i], et_mask,
+					  bnxt_link_modes[i][BNXT_SIG_MODE_PAM4_112][m]);
+		}
+	}
+}
+
+static void bnxt_fw_to_ethtool_advertised_fec(struct bnxt_link_info *link_info,
 				struct ethtool_link_ksettings *lk_ksettings)
 {
-	u16 fw_speeds = link_info->support_speeds;
+	u16 fec_cfg = link_info->fec_cfg;
 
-	BNXT_FW_TO_ETHTOOL_SPDS(fw_speeds, 0, lk_ksettings, supported);
+	if ((fec_cfg & BNXT_FEC_NONE) || !(fec_cfg & BNXT_FEC_AUTONEG)) {
+		linkmode_set_bit(ETHTOOL_LINK_MODE_FEC_NONE_BIT,
+				 lk_ksettings->link_modes.advertising);
+		return;
+	}
+	if (fec_cfg & BNXT_FEC_ENC_BASE_R)
+		linkmode_set_bit(ETHTOOL_LINK_MODE_FEC_BASER_BIT,
+				 lk_ksettings->link_modes.advertising);
+	if (fec_cfg & BNXT_FEC_ENC_RS)
+		linkmode_set_bit(ETHTOOL_LINK_MODE_FEC_RS_BIT,
+				 lk_ksettings->link_modes.advertising);
+	if (fec_cfg & BNXT_FEC_ENC_LLRS)
+		linkmode_set_bit(ETHTOOL_LINK_MODE_FEC_LLRS_BIT,
+				 lk_ksettings->link_modes.advertising);
+}
 
-	ethtool_link_ksettings_add_link_mode(lk_ksettings, supported, Pause);
-	ethtool_link_ksettings_add_link_mode(lk_ksettings, supported,
-					     Asym_Pause);
+static void bnxt_fw_to_ethtool_support_fec(struct bnxt_link_info *link_info,
+				struct ethtool_link_ksettings *lk_ksettings)
+{
+	u16 fec_cfg = link_info->fec_cfg;
 
-	if (link_info->support_auto_speeds)
-		ethtool_link_ksettings_add_link_mode(lk_ksettings, supported,
-						     Autoneg);
+	if (fec_cfg & BNXT_FEC_NONE) {
+		linkmode_set_bit(ETHTOOL_LINK_MODE_FEC_NONE_BIT,
+				 lk_ksettings->link_modes.supported);
+		return;
+	}
+	if (fec_cfg & BNXT_FEC_ENC_BASE_R_CAP)
+		linkmode_set_bit(ETHTOOL_LINK_MODE_FEC_BASER_BIT,
+				 lk_ksettings->link_modes.supported);
+	if (fec_cfg & BNXT_FEC_ENC_RS_CAP)
+		linkmode_set_bit(ETHTOOL_LINK_MODE_FEC_RS_BIT,
+				 lk_ksettings->link_modes.supported);
+	if (fec_cfg & BNXT_FEC_ENC_LLRS_CAP)
+		linkmode_set_bit(ETHTOOL_LINK_MODE_FEC_LLRS_BIT,
+				 lk_ksettings->link_modes.supported);
 }
 
 u32 bnxt_fw_to_ethtool_speed(u16 fw_link_speed)
@@ -1032,142 +2834,267 @@ u32 bnxt_fw_to_ethtool_speed(u16 fw_link_speed)
 	case BNXT_LINK_SPEED_40GB:
 		return SPEED_40000;
 	case BNXT_LINK_SPEED_50GB:
+	case BNXT_LINK_SPEED_50GB_PAM4:
 		return SPEED_50000;
 	case BNXT_LINK_SPEED_100GB:
+	case BNXT_LINK_SPEED_100GB_PAM4:
+	case BNXT_LINK_SPEED_100GB_PAM4_112:
 		return SPEED_100000;
+	case BNXT_LINK_SPEED_200GB:
+	case BNXT_LINK_SPEED_200GB_PAM4:
+	case BNXT_LINK_SPEED_200GB_PAM4_112:
+		return SPEED_200000;
+	case BNXT_LINK_SPEED_400GB:
+	case BNXT_LINK_SPEED_400GB_PAM4:
+	case BNXT_LINK_SPEED_400GB_PAM4_112:
+		return SPEED_400000;
 	default:
 		return SPEED_UNKNOWN;
 	}
 }
 
+static void bnxt_get_default_speeds(struct ethtool_link_ksettings *lk_ksettings,
+				    struct bnxt_link_info *link_info)
+{
+	struct ethtool_link_settings *base = &lk_ksettings->base;
+
+	if (link_info->link_state == BNXT_LINK_STATE_UP) {
+		base->speed = bnxt_fw_to_ethtool_speed(link_info->link_speed);
+		base->duplex = DUPLEX_HALF;
+		if (link_info->duplex & BNXT_LINK_DUPLEX_FULL)
+			base->duplex = DUPLEX_FULL;
+		lk_ksettings->lanes = link_info->active_lanes;
+	} else if (!link_info->autoneg) {
+		base->speed = bnxt_fw_to_ethtool_speed(link_info->req_link_speed);
+		base->duplex = DUPLEX_HALF;
+		if (link_info->req_duplex == BNXT_LINK_DUPLEX_FULL)
+			base->duplex = DUPLEX_FULL;
+	}
+}
+
 static int bnxt_get_link_ksettings(struct net_device *dev,
 				   struct ethtool_link_ksettings *lk_ksettings)
 {
-	struct bnxt *bp = netdev_priv(dev);
-	struct bnxt_link_info *link_info = &bp->link_info;
 	struct ethtool_link_settings *base = &lk_ksettings->base;
-	u32 ethtool_speed;
+	enum ethtool_link_mode_bit_indices link_mode;
+	struct bnxt *bp = netdev_priv(dev);
+	struct bnxt_link_info *link_info;
+	enum bnxt_media_type media;
 
+	ethtool_link_ksettings_zero_link_mode(lk_ksettings, lp_advertising);
+	ethtool_link_ksettings_zero_link_mode(lk_ksettings, advertising);
 	ethtool_link_ksettings_zero_link_mode(lk_ksettings, supported);
-	bnxt_fw_to_ethtool_support_spds(link_info, lk_ksettings);
+	base->duplex = DUPLEX_UNKNOWN;
+	base->speed = SPEED_UNKNOWN;
+	link_info = &bp->link_info;
+
+	mutex_lock(&bp->link_lock);
+	bnxt_get_ethtool_modes(link_info, lk_ksettings);
+	media = bnxt_get_media(link_info);
+	bnxt_get_all_ethtool_support_speeds(link_info, media, lk_ksettings);
+	bnxt_fw_to_ethtool_support_fec(link_info, lk_ksettings);
+	link_mode = bnxt_get_link_mode(link_info);
+	if (link_mode != BNXT_LINK_MODE_UNKNOWN)
+		ethtool_params_from_link_mode(lk_ksettings, link_mode);
+	else
+		bnxt_get_default_speeds(lk_ksettings, link_info);
 
-	ethtool_link_ksettings_zero_link_mode(lk_ksettings, advertising);
 	if (link_info->autoneg) {
-		bnxt_fw_to_ethtool_advertised_spds(link_info, lk_ksettings);
-		ethtool_link_ksettings_add_link_mode(lk_ksettings,
-						     advertising, Autoneg);
+		bnxt_fw_to_ethtool_advertised_fec(link_info, lk_ksettings);
+		linkmode_set_bit(ETHTOOL_LINK_MODE_Autoneg_BIT,
+				 lk_ksettings->link_modes.advertising);
 		base->autoneg = AUTONEG_ENABLE;
+		bnxt_get_all_ethtool_adv_speeds(link_info, media, lk_ksettings);
 		if (link_info->phy_link_status == BNXT_LINK_LINK)
-			bnxt_fw_to_ethtool_lp_adv(link_info, lk_ksettings);
-		ethtool_speed = bnxt_fw_to_ethtool_speed(link_info->link_speed);
-		if (!netif_carrier_ok(dev))
-			base->duplex = DUPLEX_UNKNOWN;
-		else if (link_info->duplex & BNXT_LINK_DUPLEX_FULL)
-			base->duplex = DUPLEX_FULL;
-		else
-			base->duplex = DUPLEX_HALF;
+			bnxt_get_all_ethtool_lp_speeds(link_info, media,
+						       lk_ksettings);
 	} else {
 		base->autoneg = AUTONEG_DISABLE;
-		ethtool_speed =
-			bnxt_fw_to_ethtool_speed(link_info->req_link_speed);
-		base->duplex = DUPLEX_HALF;
-		if (link_info->req_duplex == BNXT_LINK_DUPLEX_FULL)
-			base->duplex = DUPLEX_FULL;
 	}
-	base->speed = ethtool_speed;
 
 	base->port = PORT_NONE;
-	if (link_info->media_type == PORT_PHY_QCFG_RESP_MEDIA_TYPE_TP) {
+	if (media == BNXT_MEDIA_TP) {
 		base->port = PORT_TP;
-		ethtool_link_ksettings_add_link_mode(lk_ksettings, supported,
-						     TP);
-		ethtool_link_ksettings_add_link_mode(lk_ksettings, advertising,
-						     TP);
+		linkmode_set_bit(ETHTOOL_LINK_MODE_TP_BIT,
+				 lk_ksettings->link_modes.supported);
+		linkmode_set_bit(ETHTOOL_LINK_MODE_TP_BIT,
+				 lk_ksettings->link_modes.advertising);
+	} else if (media == BNXT_MEDIA_KR) {
+		linkmode_set_bit(ETHTOOL_LINK_MODE_Backplane_BIT,
+				 lk_ksettings->link_modes.supported);
+		linkmode_set_bit(ETHTOOL_LINK_MODE_Backplane_BIT,
+				 lk_ksettings->link_modes.advertising);
 	} else {
-		ethtool_link_ksettings_add_link_mode(lk_ksettings, supported,
-						     FIBRE);
-		ethtool_link_ksettings_add_link_mode(lk_ksettings, advertising,
-						     FIBRE);
+		linkmode_set_bit(ETHTOOL_LINK_MODE_FIBRE_BIT,
+				 lk_ksettings->link_modes.supported);
+		linkmode_set_bit(ETHTOOL_LINK_MODE_FIBRE_BIT,
+				 lk_ksettings->link_modes.advertising);
 
-		if (link_info->media_type == PORT_PHY_QCFG_RESP_MEDIA_TYPE_DAC)
+		if (media == BNXT_MEDIA_CR)
 			base->port = PORT_DA;
-		else if (link_info->media_type ==
-			 PORT_PHY_QCFG_RESP_MEDIA_TYPE_FIBRE)
+		else
 			base->port = PORT_FIBRE;
 	}
 	base->phy_address = link_info->phy_addr;
+	mutex_unlock(&bp->link_lock);
 
 	return 0;
 }
 
-static u32 bnxt_get_fw_speed(struct net_device *dev, u32 ethtool_speed)
+static int
+bnxt_force_link_speed(struct net_device *dev, u32 ethtool_speed, u32 lanes)
 {
 	struct bnxt *bp = netdev_priv(dev);
 	struct bnxt_link_info *link_info = &bp->link_info;
+	u16 support_pam4_spds = link_info->support_pam4_speeds;
+	u16 support_spds2 = link_info->support_speeds2;
 	u16 support_spds = link_info->support_speeds;
-	u32 fw_speed = 0;
+	u8 sig_mode = BNXT_SIG_MODE_NRZ;
+	u32 lanes_needed = 1;
+	u16 fw_speed = 0;
 
 	switch (ethtool_speed) {
 	case SPEED_100:
 		if (support_spds & BNXT_LINK_SPEED_MSK_100MB)
-			fw_speed = PORT_PHY_CFG_REQ_AUTO_LINK_SPEED_100MB;
+			fw_speed = PORT_PHY_CFG_REQ_FORCE_LINK_SPEED_100MB;
 		break;
 	case SPEED_1000:
-		if (support_spds & BNXT_LINK_SPEED_MSK_1GB)
-			fw_speed = PORT_PHY_CFG_REQ_AUTO_LINK_SPEED_1GB;
+		if ((support_spds & BNXT_LINK_SPEED_MSK_1GB) ||
+		    (support_spds2 & BNXT_LINK_SPEEDS2_MSK_1GB))
+			fw_speed = PORT_PHY_CFG_REQ_FORCE_LINK_SPEED_1GB;
 		break;
 	case SPEED_2500:
 		if (support_spds & BNXT_LINK_SPEED_MSK_2_5GB)
-			fw_speed = PORT_PHY_CFG_REQ_AUTO_LINK_SPEED_2_5GB;
+			fw_speed = PORT_PHY_CFG_REQ_FORCE_LINK_SPEED_2_5GB;
 		break;
 	case SPEED_10000:
-		if (support_spds & BNXT_LINK_SPEED_MSK_10GB)
-			fw_speed = PORT_PHY_CFG_REQ_AUTO_LINK_SPEED_10GB;
+		if ((support_spds & BNXT_LINK_SPEED_MSK_10GB) ||
+		    (support_spds2 & BNXT_LINK_SPEEDS2_MSK_10GB))
+			fw_speed = PORT_PHY_CFG_REQ_FORCE_LINK_SPEED_10GB;
 		break;
 	case SPEED_20000:
-		if (support_spds & BNXT_LINK_SPEED_MSK_20GB)
-			fw_speed = PORT_PHY_CFG_REQ_AUTO_LINK_SPEED_20GB;
+		if (support_spds & BNXT_LINK_SPEED_MSK_20GB) {
+			fw_speed = PORT_PHY_CFG_REQ_FORCE_LINK_SPEED_20GB;
+			lanes_needed = 2;
+		}
 		break;
 	case SPEED_25000:
-		if (support_spds & BNXT_LINK_SPEED_MSK_25GB)
-			fw_speed = PORT_PHY_CFG_REQ_AUTO_LINK_SPEED_25GB;
+		if ((support_spds & BNXT_LINK_SPEED_MSK_25GB) ||
+		    (support_spds2 & BNXT_LINK_SPEEDS2_MSK_25GB))
+			fw_speed = PORT_PHY_CFG_REQ_FORCE_LINK_SPEED_25GB;
 		break;
 	case SPEED_40000:
-		if (support_spds & BNXT_LINK_SPEED_MSK_40GB)
-			fw_speed = PORT_PHY_CFG_REQ_AUTO_LINK_SPEED_40GB;
+		if ((support_spds & BNXT_LINK_SPEED_MSK_40GB) ||
+		    (support_spds2 & BNXT_LINK_SPEEDS2_MSK_40GB)) {
+			fw_speed = PORT_PHY_CFG_REQ_FORCE_LINK_SPEED_40GB;
+			lanes_needed = 4;
+		}
 		break;
 	case SPEED_50000:
-		if (support_spds & BNXT_LINK_SPEED_MSK_50GB)
-			fw_speed = PORT_PHY_CFG_REQ_AUTO_LINK_SPEED_50GB;
+		if (((support_spds & BNXT_LINK_SPEED_MSK_50GB) ||
+		     (support_spds2 & BNXT_LINK_SPEEDS2_MSK_50GB)) &&
+		    lanes != 1) {
+			fw_speed = PORT_PHY_CFG_REQ_FORCE_LINK_SPEED_50GB;
+			lanes_needed = 2;
+		} else if (support_pam4_spds & BNXT_LINK_PAM4_SPEED_MSK_50GB) {
+			fw_speed = PORT_PHY_CFG_REQ_FORCE_PAM4_LINK_SPEED_50GB;
+			sig_mode = BNXT_SIG_MODE_PAM4;
+		} else if (support_spds2 & BNXT_LINK_SPEEDS2_MSK_50GB_PAM4) {
+			fw_speed = BNXT_LINK_SPEED_50GB_PAM4;
+			sig_mode = BNXT_SIG_MODE_PAM4;
+		}
 		break;
 	case SPEED_100000:
-		if (support_spds & BNXT_LINK_SPEED_MSK_100GB)
-			fw_speed = PORT_PHY_CFG_REQ_AUTO_LINK_SPEED_100GB;
+		if (((support_spds & BNXT_LINK_SPEED_MSK_100GB) ||
+		     (support_spds2 & BNXT_LINK_SPEEDS2_MSK_100GB)) &&
+		    lanes != 2 && lanes != 1) {
+			fw_speed = PORT_PHY_CFG_REQ_FORCE_LINK_SPEED_100GB;
+			lanes_needed = 4;
+		} else if (support_pam4_spds & BNXT_LINK_PAM4_SPEED_MSK_100GB) {
+			fw_speed = PORT_PHY_CFG_REQ_FORCE_PAM4_LINK_SPEED_100GB;
+			sig_mode = BNXT_SIG_MODE_PAM4;
+			lanes_needed = 2;
+		} else if ((support_spds2 & BNXT_LINK_SPEEDS2_MSK_100GB_PAM4) &&
+			   lanes != 1) {
+			fw_speed = BNXT_LINK_SPEED_100GB_PAM4;
+			sig_mode = BNXT_SIG_MODE_PAM4;
+			lanes_needed = 2;
+		} else if (support_spds2 & BNXT_LINK_SPEEDS2_MSK_100GB_PAM4_112) {
+			fw_speed = BNXT_LINK_SPEED_100GB_PAM4_112;
+			sig_mode = BNXT_SIG_MODE_PAM4_112;
+		}
 		break;
-	default:
-		netdev_err(dev, "unsupported speed!\n");
+	case SPEED_200000:
+		if (support_pam4_spds & BNXT_LINK_PAM4_SPEED_MSK_200GB) {
+			fw_speed = PORT_PHY_CFG_REQ_FORCE_PAM4_LINK_SPEED_200GB;
+			sig_mode = BNXT_SIG_MODE_PAM4;
+			lanes_needed = 4;
+		} else if ((support_spds2 & BNXT_LINK_SPEEDS2_MSK_200GB_PAM4) &&
+			   lanes != 2) {
+			fw_speed = BNXT_LINK_SPEED_200GB_PAM4;
+			sig_mode = BNXT_SIG_MODE_PAM4;
+			lanes_needed = 4;
+		} else if (support_spds2 & BNXT_LINK_SPEEDS2_MSK_200GB_PAM4_112) {
+			fw_speed = BNXT_LINK_SPEED_200GB_PAM4_112;
+			sig_mode = BNXT_SIG_MODE_PAM4_112;
+			lanes_needed = 2;
+		}
+		break;
+	case SPEED_400000:
+		if ((support_spds2 & BNXT_LINK_SPEEDS2_MSK_400GB_PAM4) &&
+		    lanes != 4) {
+			fw_speed = BNXT_LINK_SPEED_400GB_PAM4;
+			sig_mode = BNXT_SIG_MODE_PAM4;
+			lanes_needed = 8;
+		} else if (support_spds2 & BNXT_LINK_SPEEDS2_MSK_400GB_PAM4_112) {
+			fw_speed = BNXT_LINK_SPEED_400GB_PAM4_112;
+			sig_mode = BNXT_SIG_MODE_PAM4_112;
+			lanes_needed = 4;
+		}
 		break;
 	}
-	return fw_speed;
+
+	if (!fw_speed) {
+		netdev_err(dev, "unsupported speed!\n");
+		return -EINVAL;
+	}
+
+	if (lanes && lanes != lanes_needed) {
+		netdev_err(dev, "unsupported number of lanes for speed\n");
+		return -EINVAL;
+	}
+
+	if (link_info->req_link_speed == fw_speed &&
+	    link_info->req_signal_mode == sig_mode &&
+	    link_info->autoneg == 0)
+		return -EALREADY;
+
+	link_info->req_link_speed = fw_speed;
+	link_info->req_signal_mode = sig_mode;
+	link_info->req_duplex = BNXT_LINK_DUPLEX_FULL;
+	link_info->autoneg = 0;
+	link_info->advertising = 0;
+	link_info->advertising_pam4 = 0;
+
+	return 0;
 }
 
-u16 bnxt_get_fw_auto_link_speeds(u32 advertising)
+u16 bnxt_get_fw_auto_link_speeds(const unsigned long *mode)
 {
 	u16 fw_speed_mask = 0;
 
-	/* only support autoneg at speed 100, 1000, and 10000 */
-	if (advertising & (ADVERTISED_100baseT_Full |
-			   ADVERTISED_100baseT_Half)) {
+	if (linkmode_test_bit(ETHTOOL_LINK_MODE_100baseT_Full_BIT, mode) ||
+	    linkmode_test_bit(ETHTOOL_LINK_MODE_100baseT_Half_BIT, mode))
 		fw_speed_mask |= BNXT_LINK_SPEED_MSK_100MB;
-	}
-	if (advertising & (ADVERTISED_1000baseT_Full |
-			   ADVERTISED_1000baseT_Half)) {
+
+	if (linkmode_test_bit(ETHTOOL_LINK_MODE_1000baseT_Full_BIT, mode) ||
+	    linkmode_test_bit(ETHTOOL_LINK_MODE_1000baseT_Half_BIT, mode))
 		fw_speed_mask |= BNXT_LINK_SPEED_MSK_1GB;
-	}
-	if (advertising & ADVERTISED_10000baseT_Full)
+
+	if (linkmode_test_bit(ETHTOOL_LINK_MODE_10000baseT_Full_BIT, mode))
 		fw_speed_mask |= BNXT_LINK_SPEED_MSK_10GB;
 
-	if (advertising & ADVERTISED_40000baseCR4_Full)
+	if (linkmode_test_bit(ETHTOOL_LINK_MODE_40000baseCR4_Full_BIT, mode))
 		fw_speed_mask |= BNXT_LINK_SPEED_MSK_40GB;
 
 	return fw_speed_mask;
@@ -1180,27 +3107,28 @@ static int bnxt_set_link_ksettings(struct net_device *dev,
 	struct bnxt_link_info *link_info = &bp->link_info;
 	const struct ethtool_link_settings *base = &lk_ksettings->base;
 	bool set_pause = false;
-	u16 fw_advertising = 0;
-	u32 speed;
+	u32 speed, lanes = 0;
 	int rc = 0;
 
-	if (!BNXT_SINGLE_PF(bp))
+	if (!BNXT_PHY_CFG_ABLE(bp))
 		return -EOPNOTSUPP;
 
+	mutex_lock(&bp->link_lock);
 	if (base->autoneg == AUTONEG_ENABLE) {
-		BNXT_ETHTOOL_TO_FW_SPDS(fw_advertising, lk_ksettings,
-					advertising);
+		bnxt_set_ethtool_speeds(link_info,
+					lk_ksettings->link_modes.advertising);
 		link_info->autoneg |= BNXT_AUTONEG_SPEED;
-		if (!fw_advertising)
+		if (!link_info->advertising && !link_info->advertising_pam4) {
 			link_info->advertising = link_info->support_auto_speeds;
-		else
-			link_info->advertising = fw_advertising;
+			link_info->advertising_pam4 =
+				link_info->support_pam4_auto_speeds;
+		}
 		/* any change to autoneg will cause link change, therefore the
 		 * driver should put back the original pause setting in autoneg
 		 */
-		set_pause = true;
+		if (!(bp->phy_flags & BNXT_PHY_FL_NO_PAUSE))
+			set_pause = true;
 	} else {
-		u16 fw_speed;
 		u8 phy_type = link_info->phy_type;
 
 		if (phy_type == PORT_PHY_QCFG_RESP_PHY_TYPE_BASET  ||
@@ -1216,21 +3144,152 @@ static int bnxt_set_link_ksettings(struct net_device *dev,
 			goto set_setting_exit;
 		}
 		speed = base->speed;
-		fw_speed = bnxt_get_fw_speed(dev, speed);
-		if (!fw_speed) {
-			rc = -EINVAL;
+		lanes = lk_ksettings->lanes;
+		rc = bnxt_force_link_speed(dev, speed, lanes);
+		if (rc) {
+			if (rc == -EALREADY)
+				rc = 0;
 			goto set_setting_exit;
 		}
-		link_info->req_link_speed = fw_speed;
-		link_info->req_duplex = BNXT_LINK_DUPLEX_FULL;
-		link_info->autoneg = 0;
-		link_info->advertising = 0;
 	}
 
 	if (netif_running(dev))
 		rc = bnxt_hwrm_set_link_setting(bp, set_pause, false);
 
 set_setting_exit:
+	mutex_unlock(&bp->link_lock);
+	return rc;
+}
+
+static int bnxt_get_fecparam(struct net_device *dev,
+			     struct ethtool_fecparam *fec)
+{
+	struct bnxt *bp = netdev_priv(dev);
+	struct bnxt_link_info *link_info;
+	u8 active_fec;
+	u16 fec_cfg;
+
+	link_info = &bp->link_info;
+	fec_cfg = link_info->fec_cfg;
+	active_fec = link_info->active_fec_sig_mode &
+		     PORT_PHY_QCFG_RESP_ACTIVE_FEC_MASK;
+	if (fec_cfg & BNXT_FEC_NONE) {
+		fec->fec = ETHTOOL_FEC_NONE;
+		fec->active_fec = ETHTOOL_FEC_NONE;
+		return 0;
+	}
+	if (fec_cfg & BNXT_FEC_AUTONEG)
+		fec->fec |= ETHTOOL_FEC_AUTO;
+	if (fec_cfg & BNXT_FEC_ENC_BASE_R)
+		fec->fec |= ETHTOOL_FEC_BASER;
+	if (fec_cfg & BNXT_FEC_ENC_RS)
+		fec->fec |= ETHTOOL_FEC_RS;
+	if (fec_cfg & BNXT_FEC_ENC_LLRS)
+		fec->fec |= ETHTOOL_FEC_LLRS;
+
+	switch (active_fec) {
+	case PORT_PHY_QCFG_RESP_ACTIVE_FEC_FEC_CLAUSE74_ACTIVE:
+		fec->active_fec |= ETHTOOL_FEC_BASER;
+		break;
+	case PORT_PHY_QCFG_RESP_ACTIVE_FEC_FEC_CLAUSE91_ACTIVE:
+	case PORT_PHY_QCFG_RESP_ACTIVE_FEC_FEC_RS544_1XN_ACTIVE:
+	case PORT_PHY_QCFG_RESP_ACTIVE_FEC_FEC_RS544_IEEE_ACTIVE:
+		fec->active_fec |= ETHTOOL_FEC_RS;
+		break;
+	case PORT_PHY_QCFG_RESP_ACTIVE_FEC_FEC_RS272_1XN_ACTIVE:
+	case PORT_PHY_QCFG_RESP_ACTIVE_FEC_FEC_RS272_IEEE_ACTIVE:
+		fec->active_fec |= ETHTOOL_FEC_LLRS;
+		break;
+	case PORT_PHY_QCFG_RESP_ACTIVE_FEC_FEC_NONE_ACTIVE:
+		fec->active_fec |= ETHTOOL_FEC_OFF;
+		break;
+	}
+	return 0;
+}
+
+static void bnxt_get_fec_stats(struct net_device *dev,
+			       struct ethtool_fec_stats *fec_stats,
+			       struct ethtool_fec_hist *hist)
+{
+	struct bnxt *bp = netdev_priv(dev);
+	u64 *rx;
+
+	if (BNXT_VF(bp) || !(bp->flags & BNXT_FLAG_PORT_STATS_EXT))
+		return;
+
+	rx = bp->rx_port_stats_ext.sw_stats;
+	fec_stats->corrected_bits.total =
+		*(rx + BNXT_RX_STATS_EXT_OFFSET(rx_corrected_bits));
+
+	if (bp->fw_rx_stats_ext_size <= BNXT_RX_STATS_EXT_NUM_LEGACY)
+		return;
+
+	fec_stats->corrected_blocks.total =
+		*(rx + BNXT_RX_STATS_EXT_OFFSET(rx_fec_corrected_blocks));
+	fec_stats->uncorrectable_blocks.total =
+		*(rx + BNXT_RX_STATS_EXT_OFFSET(rx_fec_uncorrectable_blocks));
+}
+
+static u32 bnxt_ethtool_forced_fec_to_fw(struct bnxt_link_info *link_info,
+					 u32 fec)
+{
+	u32 fw_fec = PORT_PHY_CFG_REQ_FLAGS_FEC_AUTONEG_DISABLE;
+
+	if (fec & ETHTOOL_FEC_BASER)
+		fw_fec |= BNXT_FEC_BASE_R_ON(link_info);
+	else if (fec & ETHTOOL_FEC_RS)
+		fw_fec |= BNXT_FEC_RS_ON(link_info);
+	else if (fec & ETHTOOL_FEC_LLRS)
+		fw_fec |= BNXT_FEC_LLRS_ON;
+	return fw_fec;
+}
+
+static int bnxt_set_fecparam(struct net_device *dev,
+			     struct ethtool_fecparam *fecparam)
+{
+	struct hwrm_port_phy_cfg_input *req;
+	struct bnxt *bp = netdev_priv(dev);
+	struct bnxt_link_info *link_info;
+	u32 new_cfg, fec = fecparam->fec;
+	u16 fec_cfg;
+	int rc;
+
+	link_info = &bp->link_info;
+	fec_cfg = link_info->fec_cfg;
+	if (fec_cfg & BNXT_FEC_NONE)
+		return -EOPNOTSUPP;
+
+	if (fec & ETHTOOL_FEC_OFF) {
+		new_cfg = PORT_PHY_CFG_REQ_FLAGS_FEC_AUTONEG_DISABLE |
+			  BNXT_FEC_ALL_OFF(link_info);
+		goto apply_fec;
+	}
+	if (((fec & ETHTOOL_FEC_AUTO) && !(fec_cfg & BNXT_FEC_AUTONEG_CAP)) ||
+	    ((fec & ETHTOOL_FEC_RS) && !(fec_cfg & BNXT_FEC_ENC_RS_CAP)) ||
+	    ((fec & ETHTOOL_FEC_LLRS) && !(fec_cfg & BNXT_FEC_ENC_LLRS_CAP)) ||
+	    ((fec & ETHTOOL_FEC_BASER) && !(fec_cfg & BNXT_FEC_ENC_BASE_R_CAP)))
+		return -EINVAL;
+
+	if (fec & ETHTOOL_FEC_AUTO) {
+		if (!link_info->autoneg)
+			return -EINVAL;
+		new_cfg = PORT_PHY_CFG_REQ_FLAGS_FEC_AUTONEG_ENABLE;
+	} else {
+		new_cfg = bnxt_ethtool_forced_fec_to_fw(link_info, fec);
+	}
+
+apply_fec:
+	rc = hwrm_req_init(bp, req, HWRM_PORT_PHY_CFG);
+	if (rc)
+		return rc;
+	req->flags = cpu_to_le32(new_cfg | PORT_PHY_CFG_REQ_FLAGS_RESET_PHY);
+	rc = hwrm_req_send(bp, req);
+	/* update current settings */
+	if (!rc) {
+		mutex_lock(&bp->link_lock);
+		bnxt_update_link(bp, false);
+		mutex_unlock(&bp->link_lock);
+	}
 	return rc;
 }
 
@@ -1247,6 +3306,22 @@ static void bnxt_get_pauseparam(struct net_device *dev,
 	epause->tx_pause = !!(link_info->req_flow_ctrl & BNXT_LINK_PAUSE_TX);
 }
 
+static void bnxt_get_pause_stats(struct net_device *dev,
+				 struct ethtool_pause_stats *epstat)
+{
+	struct bnxt *bp = netdev_priv(dev);
+	u64 *rx, *tx;
+
+	if (BNXT_VF(bp) || !(bp->flags & BNXT_FLAG_PORT_STATS))
+		return;
+
+	rx = bp->port_stats.sw_stats;
+	tx = bp->port_stats.sw_stats + BNXT_TX_PORT_STATS_BYTE_OFFSET / 8;
+
+	epstat->rx_pause_frames = BNXT_GET_RX_PORT_STATS64(rx, rx_pause_frames);
+	epstat->tx_pause_frames = BNXT_GET_TX_PORT_STATS64(tx, tx_pause_frames);
+}
+
 static int bnxt_set_pauseparam(struct net_device *dev,
 			       struct ethtool_pauseparam *epause)
 {
@@ -1254,17 +3329,18 @@ static int bnxt_set_pauseparam(struct net_device *dev,
 	struct bnxt *bp = netdev_priv(dev);
 	struct bnxt_link_info *link_info = &bp->link_info;
 
-	if (!BNXT_SINGLE_PF(bp))
+	if (!BNXT_PHY_CFG_ABLE(bp) || (bp->phy_flags & BNXT_PHY_FL_NO_PAUSE))
 		return -EOPNOTSUPP;
 
+	mutex_lock(&bp->link_lock);
 	if (epause->autoneg) {
-		if (!(link_info->autoneg & BNXT_AUTONEG_SPEED))
-			return -EINVAL;
+		if (!(link_info->autoneg & BNXT_AUTONEG_SPEED)) {
+			rc = -EINVAL;
+			goto pause_exit;
+		}
 
 		link_info->autoneg |= BNXT_AUTONEG_FLOW_CTRL;
-		if (bp->hwrm_spec_code >= 0x10201)
-			link_info->req_flow_ctrl =
-				PORT_PHY_CFG_REQ_AUTO_PAUSE_AUTONEG_PAUSE;
+		link_info->req_flow_ctrl = 0;
 	} else {
 		/* when transition from auto pause to force pause,
 		 * force a link change
@@ -1282,6 +3358,9 @@ static int bnxt_set_pauseparam(struct net_device *dev,
 
 	if (netif_running(dev))
 		rc = bnxt_hwrm_set_pause(bp);
+
+pause_exit:
+	mutex_unlock(&bp->link_lock);
 	return rc;
 }
 
@@ -1290,90 +3369,168 @@ static u32 bnxt_get_link(struct net_device *dev)
 	struct bnxt *bp = netdev_priv(dev);
 
 	/* TODO: handle MF, VF, driver close case */
-	return bp->link_info.link_up;
+	return BNXT_LINK_IS_UP(bp);
 }
 
-static int bnxt_find_nvram_item(struct net_device *dev, u16 type, u16 ordinal,
-				u16 ext, u16 *index, u32 *item_length,
-				u32 *data_length);
+int bnxt_hwrm_nvm_get_dev_info(struct bnxt *bp,
+			       struct hwrm_nvm_get_dev_info_output *nvm_dev_info)
+{
+	struct hwrm_nvm_get_dev_info_output *resp;
+	struct hwrm_nvm_get_dev_info_input *req;
+	int rc;
 
-static int bnxt_flash_nvram(struct net_device *dev,
-			    u16 dir_type,
-			    u16 dir_ordinal,
-			    u16 dir_ext,
-			    u16 dir_attr,
-			    const u8 *data,
-			    size_t data_len)
+	if (BNXT_VF(bp))
+		return -EOPNOTSUPP;
+
+	rc = hwrm_req_init(bp, req, HWRM_NVM_GET_DEV_INFO);
+	if (rc)
+		return rc;
+
+	resp = hwrm_req_hold(bp, req);
+	rc = hwrm_req_send(bp, req);
+	if (!rc)
+		memcpy(nvm_dev_info, resp, sizeof(*resp));
+	hwrm_req_drop(bp, req);
+	return rc;
+}
+
+static void bnxt_print_admin_err(struct bnxt *bp)
+{
+	netdev_info(bp->dev, "PF does not have admin privileges to flash or reset the device\n");
+}
+
+int bnxt_find_nvram_item(struct net_device *dev, u16 type, u16 ordinal,
+			 u16 ext, u16 *index, u32 *item_length,
+			 u32 *data_length);
+
+int bnxt_flash_nvram(struct net_device *dev, u16 dir_type,
+		     u16 dir_ordinal, u16 dir_ext, u16 dir_attr,
+		     u32 dir_item_len, const u8 *data,
+		     size_t data_len)
 {
 	struct bnxt *bp = netdev_priv(dev);
+	struct hwrm_nvm_write_input *req;
 	int rc;
-	struct hwrm_nvm_write_input req = {0};
-	dma_addr_t dma_handle;
-	u8 *kmem;
 
-	bnxt_hwrm_cmd_hdr_init(bp, &req, HWRM_NVM_WRITE, -1, -1);
+	rc = hwrm_req_init(bp, req, HWRM_NVM_WRITE);
+	if (rc)
+		return rc;
 
-	req.dir_type = cpu_to_le16(dir_type);
-	req.dir_ordinal = cpu_to_le16(dir_ordinal);
-	req.dir_ext = cpu_to_le16(dir_ext);
-	req.dir_attr = cpu_to_le16(dir_attr);
-	req.dir_data_length = cpu_to_le32(data_len);
+	if (data_len && data) {
+		dma_addr_t dma_handle;
+		u8 *kmem;
 
-	kmem = dma_alloc_coherent(&bp->pdev->dev, data_len, &dma_handle,
-				  GFP_KERNEL);
-	if (!kmem) {
-		netdev_err(dev, "dma_alloc_coherent failure, length = %u\n",
-			   (unsigned)data_len);
-		return -ENOMEM;
+		kmem = hwrm_req_dma_slice(bp, req, data_len, &dma_handle);
+		if (!kmem) {
+			hwrm_req_drop(bp, req);
+			return -ENOMEM;
+		}
+
+		req->dir_data_length = cpu_to_le32(data_len);
+
+		memcpy(kmem, data, data_len);
+		req->host_src_addr = cpu_to_le64(dma_handle);
 	}
-	memcpy(kmem, data, data_len);
-	req.host_src_addr = cpu_to_le64(dma_handle);
 
-	rc = hwrm_send_message(bp, &req, sizeof(req), FLASH_NVRAM_TIMEOUT);
-	dma_free_coherent(&bp->pdev->dev, data_len, kmem, dma_handle);
+	hwrm_req_timeout(bp, req, bp->hwrm_cmd_max_timeout);
+	req->dir_type = cpu_to_le16(dir_type);
+	req->dir_ordinal = cpu_to_le16(dir_ordinal);
+	req->dir_ext = cpu_to_le16(dir_ext);
+	req->dir_attr = cpu_to_le16(dir_attr);
+	req->dir_item_length = cpu_to_le32(dir_item_len);
+	rc = hwrm_req_send(bp, req);
 
+	if (rc == -EACCES)
+		bnxt_print_admin_err(bp);
 	return rc;
 }
 
-static int bnxt_firmware_reset(struct net_device *dev,
-			       u16 dir_type)
+int bnxt_hwrm_firmware_reset(struct net_device *dev, u8 proc_type,
+			     u8 self_reset, u8 flags)
 {
 	struct bnxt *bp = netdev_priv(dev);
-	struct hwrm_fw_reset_input req = {0};
+	struct hwrm_fw_reset_input *req;
+	int rc;
+
+	if (!bnxt_hwrm_reset_permitted(bp)) {
+		netdev_warn(bp->dev, "Reset denied by firmware, it may be inhibited by remote driver");
+		return -EPERM;
+	}
+
+	rc = hwrm_req_init(bp, req, HWRM_FW_RESET);
+	if (rc)
+		return rc;
+
+	req->embedded_proc_type = proc_type;
+	req->selfrst_status = self_reset;
+	req->flags = flags;
 
-	bnxt_hwrm_cmd_hdr_init(bp, &req, HWRM_FW_RESET, -1, -1);
+	if (proc_type == FW_RESET_REQ_EMBEDDED_PROC_TYPE_AP) {
+		rc = hwrm_req_send_silent(bp, req);
+	} else {
+		rc = hwrm_req_send(bp, req);
+		if (rc == -EACCES)
+			bnxt_print_admin_err(bp);
+	}
+	return rc;
+}
+
+static int bnxt_firmware_reset(struct net_device *dev,
+			       enum bnxt_nvm_directory_type dir_type)
+{
+	u8 self_reset = FW_RESET_REQ_SELFRST_STATUS_SELFRSTNONE;
+	u8 proc_type, flags = 0;
 
-	/* TODO: Support ASAP ChiMP self-reset (e.g. upon PF driver unload) */
 	/* TODO: Address self-reset of APE/KONG/BONO/TANG or ungraceful reset */
 	/*       (e.g. when firmware isn't already running) */
 	switch (dir_type) {
 	case BNX_DIR_TYPE_CHIMP_PATCH:
 	case BNX_DIR_TYPE_BOOTCODE:
 	case BNX_DIR_TYPE_BOOTCODE_2:
-		req.embedded_proc_type = FW_RESET_REQ_EMBEDDED_PROC_TYPE_BOOT;
+		proc_type = FW_RESET_REQ_EMBEDDED_PROC_TYPE_BOOT;
 		/* Self-reset ChiMP upon next PCIe reset: */
-		req.selfrst_status = FW_RESET_REQ_SELFRST_STATUS_SELFRSTPCIERST;
+		self_reset = FW_RESET_REQ_SELFRST_STATUS_SELFRSTPCIERST;
 		break;
 	case BNX_DIR_TYPE_APE_FW:
 	case BNX_DIR_TYPE_APE_PATCH:
-		req.embedded_proc_type = FW_RESET_REQ_EMBEDDED_PROC_TYPE_MGMT;
+		proc_type = FW_RESET_REQ_EMBEDDED_PROC_TYPE_MGMT;
 		/* Self-reset APE upon next PCIe reset: */
-		req.selfrst_status = FW_RESET_REQ_SELFRST_STATUS_SELFRSTPCIERST;
+		self_reset = FW_RESET_REQ_SELFRST_STATUS_SELFRSTPCIERST;
 		break;
 	case BNX_DIR_TYPE_KONG_FW:
 	case BNX_DIR_TYPE_KONG_PATCH:
-		req.embedded_proc_type =
-			FW_RESET_REQ_EMBEDDED_PROC_TYPE_NETCTRL;
+		proc_type = FW_RESET_REQ_EMBEDDED_PROC_TYPE_NETCTRL;
 		break;
 	case BNX_DIR_TYPE_BONO_FW:
 	case BNX_DIR_TYPE_BONO_PATCH:
-		req.embedded_proc_type = FW_RESET_REQ_EMBEDDED_PROC_TYPE_ROCE;
+		proc_type = FW_RESET_REQ_EMBEDDED_PROC_TYPE_ROCE;
 		break;
 	default:
 		return -EINVAL;
 	}
 
-	return hwrm_send_message(bp, &req, sizeof(req), HWRM_CMD_TIMEOUT);
+	return bnxt_hwrm_firmware_reset(dev, proc_type, self_reset, flags);
+}
+
+static int bnxt_firmware_reset_chip(struct net_device *dev)
+{
+	struct bnxt *bp = netdev_priv(dev);
+	u8 flags = 0;
+
+	if (bp->fw_cap & BNXT_FW_CAP_HOT_RESET)
+		flags = FW_RESET_REQ_FLAGS_RESET_GRACEFUL;
+
+	return bnxt_hwrm_firmware_reset(dev,
+					FW_RESET_REQ_EMBEDDED_PROC_TYPE_CHIP,
+					FW_RESET_REQ_SELFRST_STATUS_SELFRSTASAP,
+					flags);
+}
+
+static int bnxt_firmware_reset_ap(struct net_device *dev)
+{
+	return bnxt_hwrm_firmware_reset(dev, FW_RESET_REQ_EMBEDDED_PROC_TYPE_AP,
+					FW_RESET_REQ_SELFRST_STATUS_SELFRSTNONE,
+					0);
 }
 
 static int bnxt_flash_firmware(struct net_device *dev,
@@ -1449,7 +3606,7 @@ static int bnxt_flash_firmware(struct net_device *dev,
 		return -EINVAL;
 	}
 	rc = bnxt_flash_nvram(dev, dir_type, BNX_DIR_ORDINAL_FIRST,
-			      0, 0, fw_data, fw_size);
+			      0, 0, 0, fw_data, fw_size);
 	if (rc == 0)	/* Firmware update successful */
 		rc = bnxt_firmware_reset(dev, dir_type);
 
@@ -1502,7 +3659,7 @@ static int bnxt_flash_microcode(struct net_device *dev,
 		return -EINVAL;
 	}
 	rc = bnxt_flash_nvram(dev, dir_type, BNX_DIR_ORDINAL_FIRST,
-			      0, 0, fw_data, fw_size);
+			      0, 0, 0, fw_data, fw_size);
 
 	return rc;
 }
@@ -1562,114 +3719,278 @@ static int bnxt_flash_firmware_from_file(struct net_device *dev,
 			   rc, filename);
 		return rc;
 	}
-	if (bnxt_dir_type_is_ape_bin_format(dir_type) == true)
+	if (bnxt_dir_type_is_ape_bin_format(dir_type))
 		rc = bnxt_flash_firmware(dev, dir_type, fw->data, fw->size);
-	else if (bnxt_dir_type_is_other_exec_format(dir_type) == true)
+	else if (bnxt_dir_type_is_other_exec_format(dir_type))
 		rc = bnxt_flash_microcode(dev, dir_type, fw->data, fw->size);
 	else
 		rc = bnxt_flash_nvram(dev, dir_type, BNX_DIR_ORDINAL_FIRST,
-				      0, 0, fw->data, fw->size);
+				      0, 0, 0, fw->data, fw->size);
 	release_firmware(fw);
 	return rc;
 }
 
-static int bnxt_flash_package_from_file(struct net_device *dev,
-					char *filename, u32 install_type)
+#define MSG_INTEGRITY_ERR "PKG install error : Data integrity on NVM"
+#define MSG_INVALID_PKG "PKG install error : Invalid package"
+#define MSG_AUTHENTICATION_ERR "PKG install error : Authentication error"
+#define MSG_INVALID_DEV "PKG install error : Invalid device"
+#define MSG_INTERNAL_ERR "PKG install error : Internal error"
+#define MSG_NO_PKG_UPDATE_AREA_ERR "PKG update area not created in nvram"
+#define MSG_NO_SPACE_ERR "PKG insufficient update area in nvram"
+#define MSG_RESIZE_UPDATE_ERR "Resize UPDATE entry error"
+#define MSG_ANTI_ROLLBACK_ERR "HWRM_NVM_INSTALL_UPDATE failure due to Anti-rollback detected"
+#define MSG_GENERIC_FAILURE_ERR "HWRM_NVM_INSTALL_UPDATE failure"
+
+static int nvm_update_err_to_stderr(struct net_device *dev, u8 result,
+				    struct netlink_ext_ack *extack)
+{
+	switch (result) {
+	case NVM_INSTALL_UPDATE_RESP_RESULT_INVALID_TYPE_PARAMETER:
+	case NVM_INSTALL_UPDATE_RESP_RESULT_INVALID_INDEX_PARAMETER:
+	case NVM_INSTALL_UPDATE_RESP_RESULT_INSTALL_DATA_ERROR:
+	case NVM_INSTALL_UPDATE_RESP_RESULT_INSTALL_CHECKSUM_ERROR:
+	case NVM_INSTALL_UPDATE_RESP_RESULT_ITEM_NOT_FOUND:
+	case NVM_INSTALL_UPDATE_RESP_RESULT_ITEM_LOCKED:
+		BNXT_NVM_ERR_MSG(dev, extack, MSG_INTEGRITY_ERR);
+		return -EINVAL;
+	case NVM_INSTALL_UPDATE_RESP_RESULT_INVALID_PREREQUISITE:
+	case NVM_INSTALL_UPDATE_RESP_RESULT_INVALID_FILE_HEADER:
+	case NVM_INSTALL_UPDATE_RESP_RESULT_INVALID_SIGNATURE:
+	case NVM_INSTALL_UPDATE_RESP_RESULT_INVALID_PROP_STREAM:
+	case NVM_INSTALL_UPDATE_RESP_RESULT_INVALID_PROP_LENGTH:
+	case NVM_INSTALL_UPDATE_RESP_RESULT_INVALID_MANIFEST:
+	case NVM_INSTALL_UPDATE_RESP_RESULT_INVALID_TRAILER:
+	case NVM_INSTALL_UPDATE_RESP_RESULT_INVALID_CHECKSUM:
+	case NVM_INSTALL_UPDATE_RESP_RESULT_INVALID_ITEM_CHECKSUM:
+	case NVM_INSTALL_UPDATE_RESP_RESULT_INVALID_DATA_LENGTH:
+	case NVM_INSTALL_UPDATE_RESP_RESULT_INVALID_DIRECTIVE:
+	case NVM_INSTALL_UPDATE_RESP_RESULT_DUPLICATE_ITEM:
+	case NVM_INSTALL_UPDATE_RESP_RESULT_ZERO_LENGTH_ITEM:
+		BNXT_NVM_ERR_MSG(dev, extack, MSG_INVALID_PKG);
+		return -ENOPKG;
+	case NVM_INSTALL_UPDATE_RESP_RESULT_INSTALL_AUTHENTICATION_ERROR:
+		BNXT_NVM_ERR_MSG(dev, extack, MSG_AUTHENTICATION_ERR);
+		return -EPERM;
+	case NVM_INSTALL_UPDATE_RESP_RESULT_UNSUPPORTED_CHIP_REV:
+	case NVM_INSTALL_UPDATE_RESP_RESULT_UNSUPPORTED_DEVICE_ID:
+	case NVM_INSTALL_UPDATE_RESP_RESULT_UNSUPPORTED_SUBSYS_VENDOR:
+	case NVM_INSTALL_UPDATE_RESP_RESULT_UNSUPPORTED_SUBSYS_ID:
+	case NVM_INSTALL_UPDATE_RESP_RESULT_UNSUPPORTED_PLATFORM:
+		BNXT_NVM_ERR_MSG(dev, extack, MSG_INVALID_DEV);
+		return -EOPNOTSUPP;
+	default:
+		BNXT_NVM_ERR_MSG(dev, extack, MSG_INTERNAL_ERR);
+		return -EIO;
+	}
+}
+
+#define BNXT_PKG_DMA_SIZE	0x40000
+#define BNXT_NVM_MORE_FLAG	(cpu_to_le16(NVM_MODIFY_REQ_FLAGS_BATCH_MODE))
+#define BNXT_NVM_LAST_FLAG	(cpu_to_le16(NVM_MODIFY_REQ_FLAGS_BATCH_LAST))
+
+static int bnxt_resize_update_entry(struct net_device *dev, size_t fw_size,
+				    struct netlink_ext_ack *extack)
+{
+	u32 item_len;
+	int rc;
+
+	rc = bnxt_find_nvram_item(dev, BNX_DIR_TYPE_UPDATE,
+				  BNX_DIR_ORDINAL_FIRST, BNX_DIR_EXT_NONE, NULL,
+				  &item_len, NULL);
+	if (rc) {
+		BNXT_NVM_ERR_MSG(dev, extack, MSG_NO_PKG_UPDATE_AREA_ERR);
+		return rc;
+	}
+
+	if (fw_size > item_len) {
+		rc = bnxt_flash_nvram(dev, BNX_DIR_TYPE_UPDATE,
+				      BNX_DIR_ORDINAL_FIRST, 0, 1,
+				      round_up(fw_size, 4096), NULL, 0);
+		if (rc) {
+			BNXT_NVM_ERR_MSG(dev, extack, MSG_RESIZE_UPDATE_ERR);
+			return rc;
+		}
+	}
+	return 0;
+}
+
+int bnxt_flash_package_from_fw_obj(struct net_device *dev, const struct firmware *fw,
+				   u32 install_type, struct netlink_ext_ack *extack)
 {
+	struct hwrm_nvm_install_update_input *install;
+	struct hwrm_nvm_install_update_output *resp;
+	struct hwrm_nvm_modify_input *modify;
 	struct bnxt *bp = netdev_priv(dev);
-	struct hwrm_nvm_install_update_output *resp = bp->hwrm_cmd_resp_addr;
-	struct hwrm_nvm_install_update_input install = {0};
-	const struct firmware *fw;
+	bool defrag_attempted = false;
+	dma_addr_t dma_handle;
+	u8 *kmem = NULL;
+	u32 modify_len;
 	u32 item_len;
+	u8 cmd_err;
 	u16 index;
 	int rc;
 
+	/* resize before flashing larger image than available space */
+	rc = bnxt_resize_update_entry(dev, fw->size, extack);
+	if (rc)
+		return rc;
+
 	bnxt_hwrm_fw_set_time(bp);
 
-	if (bnxt_find_nvram_item(dev, BNX_DIR_TYPE_UPDATE,
-				 BNX_DIR_ORDINAL_FIRST, BNX_DIR_EXT_NONE,
-				 &index, &item_len, NULL) != 0) {
-		netdev_err(dev, "PKG update area not created in nvram\n");
-		return -ENOBUFS;
+	rc = hwrm_req_init(bp, modify, HWRM_NVM_MODIFY);
+	if (rc)
+		return rc;
+
+	/* Try allocating a large DMA buffer first.  Older fw will
+	 * cause excessive NVRAM erases when using small blocks.
+	 */
+	modify_len = roundup_pow_of_two(fw->size);
+	modify_len = min_t(u32, modify_len, BNXT_PKG_DMA_SIZE);
+	while (1) {
+		kmem = hwrm_req_dma_slice(bp, modify, modify_len, &dma_handle);
+		if (!kmem && modify_len > PAGE_SIZE)
+			modify_len /= 2;
+		else
+			break;
+	}
+	if (!kmem) {
+		hwrm_req_drop(bp, modify);
+		return -ENOMEM;
 	}
 
-	rc = request_firmware(&fw, filename, &dev->dev);
-	if (rc != 0) {
-		netdev_err(dev, "PKG error %d requesting file: %s\n",
-			   rc, filename);
+	rc = hwrm_req_init(bp, install, HWRM_NVM_INSTALL_UPDATE);
+	if (rc) {
+		hwrm_req_drop(bp, modify);
 		return rc;
 	}
 
-	if (fw->size > item_len) {
-		netdev_err(dev, "PKG insufficient update area in nvram: %lu",
-			   (unsigned long)fw->size);
-		rc = -EFBIG;
-	} else {
-		dma_addr_t dma_handle;
-		u8 *kmem;
-		struct hwrm_nvm_modify_input modify = {0};
+	hwrm_req_timeout(bp, modify, bp->hwrm_cmd_max_timeout);
+	hwrm_req_timeout(bp, install, bp->hwrm_cmd_max_timeout);
+
+	hwrm_req_hold(bp, modify);
+	modify->host_src_addr = cpu_to_le64(dma_handle);
+
+	resp = hwrm_req_hold(bp, install);
+	if ((install_type & 0xffff) == 0)
+		install_type >>= 16;
+	install->install_type = cpu_to_le32(install_type);
 
-		bnxt_hwrm_cmd_hdr_init(bp, &modify, HWRM_NVM_MODIFY, -1, -1);
+	do {
+		u32 copied = 0, len = modify_len;
 
-		modify.dir_idx = cpu_to_le16(index);
-		modify.len = cpu_to_le32(fw->size);
+		rc = bnxt_find_nvram_item(dev, BNX_DIR_TYPE_UPDATE,
+					  BNX_DIR_ORDINAL_FIRST,
+					  BNX_DIR_EXT_NONE,
+					  &index, &item_len, NULL);
+		if (rc) {
+			BNXT_NVM_ERR_MSG(dev, extack, MSG_NO_PKG_UPDATE_AREA_ERR);
+			break;
+		}
+		if (fw->size > item_len) {
+			BNXT_NVM_ERR_MSG(dev, extack, MSG_NO_SPACE_ERR);
+			rc = -EFBIG;
+			break;
+		}
 
-		kmem = dma_alloc_coherent(&bp->pdev->dev, fw->size,
-					  &dma_handle, GFP_KERNEL);
-		if (!kmem) {
-			netdev_err(dev,
-				   "dma_alloc_coherent failure, length = %u\n",
-				   (unsigned int)fw->size);
-			rc = -ENOMEM;
-		} else {
-			memcpy(kmem, fw->data, fw->size);
-			modify.host_src_addr = cpu_to_le64(dma_handle);
+		modify->dir_idx = cpu_to_le16(index);
 
-			rc = hwrm_send_message(bp, &modify, sizeof(modify),
-					       FLASH_PACKAGE_TIMEOUT);
-			dma_free_coherent(&bp->pdev->dev, fw->size, kmem,
-					  dma_handle);
+		if (fw->size > modify_len)
+			modify->flags = BNXT_NVM_MORE_FLAG;
+		while (copied < fw->size) {
+			u32 balance = fw->size - copied;
+
+			if (balance <= modify_len) {
+				len = balance;
+				if (copied)
+					modify->flags |= BNXT_NVM_LAST_FLAG;
+			}
+			memcpy(kmem, fw->data + copied, len);
+			modify->len = cpu_to_le32(len);
+			modify->offset = cpu_to_le32(copied);
+			rc = hwrm_req_send(bp, modify);
+			if (rc)
+				goto pkg_abort;
+			copied += len;
 		}
-	}
-	release_firmware(fw);
-	if (rc)
-		return rc;
 
-	if ((install_type & 0xffff) == 0)
-		install_type >>= 16;
-	bnxt_hwrm_cmd_hdr_init(bp, &install, HWRM_NVM_INSTALL_UPDATE, -1, -1);
-	install.install_type = cpu_to_le32(install_type);
+		rc = hwrm_req_send_silent(bp, install);
+		if (!rc)
+			break;
 
-	mutex_lock(&bp->hwrm_cmd_lock);
-	rc = _hwrm_send_message(bp, &install, sizeof(install),
-				INSTALL_PACKAGE_TIMEOUT);
-	if (rc) {
-		rc = -EOPNOTSUPP;
-		goto flash_pkg_exit;
-	}
+		if (defrag_attempted) {
+			/* We have tried to defragment already in the previous
+			 * iteration. Return with the result for INSTALL_UPDATE
+			 */
+			break;
+		}
+
+		cmd_err = ((struct hwrm_err_output *)resp)->cmd_err;
+
+		switch (cmd_err) {
+		case NVM_INSTALL_UPDATE_CMD_ERR_CODE_ANTI_ROLLBACK:
+			BNXT_NVM_ERR_MSG(dev, extack, MSG_ANTI_ROLLBACK_ERR);
+			rc = -EALREADY;
+			break;
+		case NVM_INSTALL_UPDATE_CMD_ERR_CODE_FRAG_ERR:
+			install->flags =
+				cpu_to_le16(NVM_INSTALL_UPDATE_REQ_FLAGS_ALLOWED_TO_DEFRAG);
 
-	if (resp->error_code) {
-		u8 error_code = ((struct hwrm_err_output *)resp)->cmd_err;
+			rc = hwrm_req_send_silent(bp, install);
+			if (!rc)
+				break;
 
-		if (error_code == NVM_INSTALL_UPDATE_CMD_ERR_CODE_FRAG_ERR) {
-			install.flags |= cpu_to_le16(
-			       NVM_INSTALL_UPDATE_REQ_FLAGS_ALLOWED_TO_DEFRAG);
-			rc = _hwrm_send_message(bp, &install, sizeof(install),
-						INSTALL_PACKAGE_TIMEOUT);
-			if (rc) {
-				rc = -EOPNOTSUPP;
-				goto flash_pkg_exit;
+			cmd_err = ((struct hwrm_err_output *)resp)->cmd_err;
+
+			if (cmd_err == NVM_INSTALL_UPDATE_CMD_ERR_CODE_NO_SPACE) {
+				/* FW has cleared NVM area, driver will create
+				 * UPDATE directory and try the flash again
+				 */
+				defrag_attempted = true;
+				install->flags = 0;
+				rc = bnxt_flash_nvram(bp->dev,
+						      BNX_DIR_TYPE_UPDATE,
+						      BNX_DIR_ORDINAL_FIRST,
+						      0, 0, item_len, NULL, 0);
+				if (!rc)
+					break;
 			}
+			fallthrough;
+		default:
+			BNXT_NVM_ERR_MSG(dev, extack, MSG_GENERIC_FAILURE_ERR);
 		}
-	}
+	} while (defrag_attempted && !rc);
+
+pkg_abort:
+	hwrm_req_drop(bp, modify);
+	hwrm_req_drop(bp, install);
 
 	if (resp->result) {
 		netdev_err(dev, "PKG install error = %d, problem_item = %d\n",
 			   (s8)resp->result, (int)resp->problem_item);
-		rc = -ENOPKG;
+		rc = nvm_update_err_to_stderr(dev, resp->result, extack);
 	}
-flash_pkg_exit:
-	mutex_unlock(&bp->hwrm_cmd_lock);
+	if (rc == -EACCES)
+		bnxt_print_admin_err(bp);
+	return rc;
+}
+
+static int bnxt_flash_package_from_file(struct net_device *dev, const char *filename,
+					u32 install_type, struct netlink_ext_ack *extack)
+{
+	const struct firmware *fw;
+	int rc;
+
+	rc = request_firmware(&fw, filename, &dev->dev);
+	if (rc != 0) {
+		netdev_err(dev, "PKG error %d requesting file: %s\n",
+			   rc, filename);
+		return rc;
+	}
+
+	rc = bnxt_flash_package_from_fw_obj(dev, fw, install_type, extack);
+
+	release_firmware(fw);
+
 	return rc;
 }
 
@@ -1684,32 +4005,39 @@ static int bnxt_flash_device(struct net_device *dev,
 	if (flash->region == ETHTOOL_FLASH_ALL_REGIONS ||
 	    flash->region > 0xffff)
 		return bnxt_flash_package_from_file(dev, flash->data,
-						    flash->region);
+						    flash->region, NULL);
 
 	return bnxt_flash_firmware_from_file(dev, flash->region, flash->data);
 }
 
 static int nvm_get_dir_info(struct net_device *dev, u32 *entries, u32 *length)
 {
+	struct hwrm_nvm_get_dir_info_output *output;
+	struct hwrm_nvm_get_dir_info_input *req;
 	struct bnxt *bp = netdev_priv(dev);
 	int rc;
-	struct hwrm_nvm_get_dir_info_input req = {0};
-	struct hwrm_nvm_get_dir_info_output *output = bp->hwrm_cmd_resp_addr;
 
-	bnxt_hwrm_cmd_hdr_init(bp, &req, HWRM_NVM_GET_DIR_INFO, -1, -1);
+	rc = hwrm_req_init(bp, req, HWRM_NVM_GET_DIR_INFO);
+	if (rc)
+		return rc;
 
-	mutex_lock(&bp->hwrm_cmd_lock);
-	rc = _hwrm_send_message(bp, &req, sizeof(req), HWRM_CMD_TIMEOUT);
+	output = hwrm_req_hold(bp, req);
+	rc = hwrm_req_send(bp, req);
 	if (!rc) {
 		*entries = le32_to_cpu(output->entries);
 		*length = le32_to_cpu(output->entry_length);
 	}
-	mutex_unlock(&bp->hwrm_cmd_lock);
+	hwrm_req_drop(bp, req);
 	return rc;
 }
 
 static int bnxt_get_eeprom_len(struct net_device *dev)
 {
+	struct bnxt *bp = netdev_priv(dev);
+
+	if (BNXT_VF(bp))
+		return 0;
+
 	/* The -1 return value allows the entire 32-bit range of offsets to be
 	 * passed via the ethtool command-line utility.
 	 */
@@ -1725,12 +4053,15 @@ static int bnxt_get_nvram_directory(struct net_device *dev, u32 len, u8 *data)
 	u8 *buf;
 	size_t buflen;
 	dma_addr_t dma_handle;
-	struct hwrm_nvm_get_dir_entries_input req = {0};
+	struct hwrm_nvm_get_dir_entries_input *req;
 
 	rc = nvm_get_dir_info(dev, &dir_entries, &entry_length);
 	if (rc != 0)
 		return rc;
 
+	if (!dir_entries || !entry_length)
+		return -EIO;
+
 	/* Insert 2 bytes of directory info (count and size of entries) */
 	if (len < 2)
 		return -EINVAL;
@@ -1740,69 +4071,82 @@ static int bnxt_get_nvram_directory(struct net_device *dev, u32 len, u8 *data)
 	len -= 2;
 	memset(data, 0xff, len);
 
-	buflen = dir_entries * entry_length;
-	buf = dma_alloc_coherent(&bp->pdev->dev, buflen, &dma_handle,
-				 GFP_KERNEL);
+	rc = hwrm_req_init(bp, req, HWRM_NVM_GET_DIR_ENTRIES);
+	if (rc)
+		return rc;
+
+	buflen = mul_u32_u32(dir_entries, entry_length);
+	buf = hwrm_req_dma_slice(bp, req, buflen, &dma_handle);
 	if (!buf) {
-		netdev_err(dev, "dma_alloc_coherent failure, length = %u\n",
-			   (unsigned)buflen);
+		hwrm_req_drop(bp, req);
 		return -ENOMEM;
 	}
-	bnxt_hwrm_cmd_hdr_init(bp, &req, HWRM_NVM_GET_DIR_ENTRIES, -1, -1);
-	req.host_dest_addr = cpu_to_le64(dma_handle);
-	rc = hwrm_send_message(bp, &req, sizeof(req), HWRM_CMD_TIMEOUT);
+	req->host_dest_addr = cpu_to_le64(dma_handle);
+
+	hwrm_req_hold(bp, req); /* hold the slice */
+	rc = hwrm_req_send(bp, req);
 	if (rc == 0)
 		memcpy(data, buf, len > buflen ? buflen : len);
-	dma_free_coherent(&bp->pdev->dev, buflen, buf, dma_handle);
+	hwrm_req_drop(bp, req);
 	return rc;
 }
 
-static int bnxt_get_nvram_item(struct net_device *dev, u32 index, u32 offset,
-			       u32 length, u8 *data)
+int bnxt_get_nvram_item(struct net_device *dev, u32 index, u32 offset,
+			u32 length, u8 *data)
 {
 	struct bnxt *bp = netdev_priv(dev);
 	int rc;
 	u8 *buf;
 	dma_addr_t dma_handle;
-	struct hwrm_nvm_read_input req = {0};
+	struct hwrm_nvm_read_input *req;
+
+	if (!length)
+		return -EINVAL;
 
-	buf = dma_alloc_coherent(&bp->pdev->dev, length, &dma_handle,
-				 GFP_KERNEL);
+	rc = hwrm_req_init(bp, req, HWRM_NVM_READ);
+	if (rc)
+		return rc;
+
+	buf = hwrm_req_dma_slice(bp, req, length, &dma_handle);
 	if (!buf) {
-		netdev_err(dev, "dma_alloc_coherent failure, length = %u\n",
-			   (unsigned)length);
+		hwrm_req_drop(bp, req);
 		return -ENOMEM;
 	}
-	bnxt_hwrm_cmd_hdr_init(bp, &req, HWRM_NVM_READ, -1, -1);
-	req.host_dest_addr = cpu_to_le64(dma_handle);
-	req.dir_idx = cpu_to_le16(index);
-	req.offset = cpu_to_le32(offset);
-	req.len = cpu_to_le32(length);
 
-	rc = hwrm_send_message(bp, &req, sizeof(req), HWRM_CMD_TIMEOUT);
+	req->host_dest_addr = cpu_to_le64(dma_handle);
+	req->dir_idx = cpu_to_le16(index);
+	req->offset = cpu_to_le32(offset);
+	req->len = cpu_to_le32(length);
+
+	hwrm_req_hold(bp, req); /* hold the slice */
+	rc = hwrm_req_send(bp, req);
 	if (rc == 0)
 		memcpy(data, buf, length);
-	dma_free_coherent(&bp->pdev->dev, length, buf, dma_handle);
+	hwrm_req_drop(bp, req);
 	return rc;
 }
 
-static int bnxt_find_nvram_item(struct net_device *dev, u16 type, u16 ordinal,
-				u16 ext, u16 *index, u32 *item_length,
-				u32 *data_length)
+int bnxt_find_nvram_item(struct net_device *dev, u16 type, u16 ordinal,
+			 u16 ext, u16 *index, u32 *item_length,
+			 u32 *data_length)
 {
+	struct hwrm_nvm_find_dir_entry_output *output;
+	struct hwrm_nvm_find_dir_entry_input *req;
 	struct bnxt *bp = netdev_priv(dev);
 	int rc;
-	struct hwrm_nvm_find_dir_entry_input req = {0};
-	struct hwrm_nvm_find_dir_entry_output *output = bp->hwrm_cmd_resp_addr;
-
-	bnxt_hwrm_cmd_hdr_init(bp, &req, HWRM_NVM_FIND_DIR_ENTRY, -1, -1);
-	req.enables = 0;
-	req.dir_idx = 0;
-	req.dir_type = cpu_to_le16(type);
-	req.dir_ordinal = cpu_to_le16(ordinal);
-	req.dir_ext = cpu_to_le16(ext);
-	req.opt_ordinal = NVM_FIND_DIR_ENTRY_REQ_OPT_ORDINAL_EQ;
-	rc = hwrm_send_message_silent(bp, &req, sizeof(req), HWRM_CMD_TIMEOUT);
+
+	rc = hwrm_req_init(bp, req, HWRM_NVM_FIND_DIR_ENTRY);
+	if (rc)
+		return rc;
+
+	req->enables = 0;
+	req->dir_idx = 0;
+	req->dir_type = cpu_to_le16(type);
+	req->dir_ordinal = cpu_to_le16(ordinal);
+	req->dir_ext = cpu_to_le16(ext);
+	req->opt_ordinal = NVM_FIND_DIR_ENTRY_REQ_OPT_ORDINAL_EQ;
+	output = hwrm_req_hold(bp, req);
+	rc = hwrm_req_send_silent(bp, req);
 	if (rc == 0) {
 		if (index)
 			*index = le16_to_cpu(output->dir_idx);
@@ -1811,6 +4155,7 @@ static int bnxt_find_nvram_item(struct net_device *dev, u16 type, u16 ordinal,
 		if (data_length)
 			*data_length = le32_to_cpu(output->dir_data_length);
 	}
+	hwrm_req_drop(bp, req);
 	return rc;
 }
 
@@ -1847,22 +4192,56 @@ static char *bnxt_parse_pkglog(int desired_field, u8 *data, size_t datalen)
 	return retval;
 }
 
-static char *bnxt_get_pkgver(struct net_device *dev, char *buf, size_t buflen)
+int bnxt_get_pkginfo(struct net_device *dev, char *ver, int size)
 {
+	struct bnxt *bp = netdev_priv(dev);
 	u16 index = 0;
-	u32 datalen;
+	char *pkgver;
+	u32 pkglen;
+	u8 *pkgbuf;
+	int rc;
 
-	if (bnxt_find_nvram_item(dev, BNX_DIR_TYPE_PKG_LOG,
-				 BNX_DIR_ORDINAL_FIRST, BNX_DIR_EXT_NONE,
-				 &index, NULL, &datalen) != 0)
-		return NULL;
+	rc = bnxt_find_nvram_item(dev, BNX_DIR_TYPE_PKG_LOG,
+				  BNX_DIR_ORDINAL_FIRST, BNX_DIR_EXT_NONE,
+				  &index, NULL, &pkglen);
+	if (rc)
+		return rc;
 
-	memset(buf, 0, buflen);
-	if (bnxt_get_nvram_item(dev, index, 0, datalen, buf) != 0)
-		return NULL;
+	pkgbuf = kzalloc(pkglen, GFP_KERNEL);
+	if (!pkgbuf) {
+		dev_err(&bp->pdev->dev, "Unable to allocate memory for pkg version, length = %u\n",
+			pkglen);
+		return -ENOMEM;
+	}
+
+	rc = bnxt_get_nvram_item(dev, index, 0, pkglen, pkgbuf);
+	if (rc)
+		goto err;
+
+	pkgver = bnxt_parse_pkglog(BNX_PKG_LOG_FIELD_IDX_PKG_VERSION, pkgbuf,
+				   pkglen);
+	if (pkgver && *pkgver != 0 && isdigit(*pkgver))
+		strscpy(ver, pkgver, size);
+	else
+		rc = -ENOENT;
+
+err:
+	kfree(pkgbuf);
+
+	return rc;
+}
+
+static void bnxt_get_pkgver(struct net_device *dev)
+{
+	struct bnxt *bp = netdev_priv(dev);
+	char buf[FW_VER_STR_LEN - 5];
+	int len;
 
-	return bnxt_parse_pkglog(BNX_PKG_LOG_FIELD_IDX_PKG_VERSION, buf,
-		datalen);
+	if (!bnxt_get_pkginfo(dev, buf, sizeof(buf))) {
+		len = strlen(bp->fw_ver_str);
+		snprintf(bp->fw_ver_str + len, FW_VER_STR_LEN - len,
+			 "/pkg %s", buf);
+	}
 }
 
 static int bnxt_get_eeprom(struct net_device *dev,
@@ -1888,12 +4267,16 @@ static int bnxt_get_eeprom(struct net_device *dev,
 
 static int bnxt_erase_nvram_directory(struct net_device *dev, u8 index)
 {
+	struct hwrm_nvm_erase_dir_entry_input *req;
 	struct bnxt *bp = netdev_priv(dev);
-	struct hwrm_nvm_erase_dir_entry_input req = {0};
+	int rc;
 
-	bnxt_hwrm_cmd_hdr_init(bp, &req, HWRM_NVM_ERASE_DIR_ENTRY, -1, -1);
-	req.dir_idx = cpu_to_le16(index);
-	return hwrm_send_message(bp, &req, sizeof(req), HWRM_CMD_TIMEOUT);
+	rc = hwrm_req_init(bp, req, HWRM_NVM_ERASE_DIR_ENTRY);
+	if (rc)
+		return rc;
+
+	req->dir_idx = cpu_to_le16(index);
+	return hwrm_req_send(bp, req);
 }
 
 static int bnxt_set_eeprom(struct net_device *dev,
@@ -1927,57 +4310,61 @@ static int bnxt_set_eeprom(struct net_device *dev,
 	}
 
 	/* Create or re-write an NVM item: */
-	if (bnxt_dir_type_is_executable(type) == true)
+	if (bnxt_dir_type_is_executable(type))
 		return -EOPNOTSUPP;
 	ext = eeprom->magic & 0xffff;
 	ordinal = eeprom->offset >> 16;
 	attr = eeprom->offset & 0xffff;
 
-	return bnxt_flash_nvram(dev, type, ordinal, ext, attr, data,
+	return bnxt_flash_nvram(dev, type, ordinal, ext, attr, 0, data,
 				eeprom->len);
 }
 
-static int bnxt_set_eee(struct net_device *dev, struct ethtool_eee *edata)
+static int bnxt_set_eee(struct net_device *dev, struct ethtool_keee *edata)
 {
+	__ETHTOOL_DECLARE_LINK_MODE_MASK(advertising);
+	__ETHTOOL_DECLARE_LINK_MODE_MASK(tmp);
 	struct bnxt *bp = netdev_priv(dev);
-	struct ethtool_eee *eee = &bp->eee;
+	struct ethtool_keee *eee = &bp->eee;
 	struct bnxt_link_info *link_info = &bp->link_info;
-	u32 advertising =
-		 _bnxt_fw_to_ethtool_adv_spds(link_info->advertising, 0);
 	int rc = 0;
 
-	if (!BNXT_SINGLE_PF(bp))
+	if (!BNXT_PHY_CFG_ABLE(bp))
 		return -EOPNOTSUPP;
 
-	if (!(bp->flags & BNXT_FLAG_EEE_CAP))
+	if (!(bp->phy_flags & BNXT_PHY_FL_EEE_CAP))
 		return -EOPNOTSUPP;
 
+	mutex_lock(&bp->link_lock);
+	_bnxt_fw_to_linkmode(advertising, link_info->advertising);
 	if (!edata->eee_enabled)
 		goto eee_ok;
 
 	if (!(link_info->autoneg & BNXT_AUTONEG_SPEED)) {
 		netdev_warn(dev, "EEE requires autoneg\n");
-		return -EINVAL;
+		rc = -EINVAL;
+		goto eee_exit;
 	}
 	if (edata->tx_lpi_enabled) {
 		if (bp->lpi_tmr_hi && (edata->tx_lpi_timer > bp->lpi_tmr_hi ||
 				       edata->tx_lpi_timer < bp->lpi_tmr_lo)) {
 			netdev_warn(dev, "Valid LPI timer range is %d and %d microsecs\n",
 				    bp->lpi_tmr_lo, bp->lpi_tmr_hi);
-			return -EINVAL;
+			rc = -EINVAL;
+			goto eee_exit;
 		} else if (!bp->lpi_tmr_hi) {
 			edata->tx_lpi_timer = eee->tx_lpi_timer;
 		}
 	}
-	if (!edata->advertised) {
-		edata->advertised = advertising & eee->supported;
-	} else if (edata->advertised & ~advertising) {
-		netdev_warn(dev, "EEE advertised %x must be a subset of autoneg advertised speeds %x\n",
-			    edata->advertised, advertising);
-		return -EINVAL;
+	if (linkmode_empty(edata->advertised)) {
+		linkmode_and(edata->advertised, advertising, eee->supported);
+	} else if (linkmode_andnot(tmp, edata->advertised, advertising)) {
+		netdev_warn(dev, "EEE advertised must be a subset of autoneg advertised speeds\n");
+		rc = -EINVAL;
+		goto eee_exit;
 	}
 
-	eee->advertised = edata->advertised;
+	linkmode_copy(eee->advertised, edata->advertised);
 	eee->tx_lpi_enabled = edata->tx_lpi_enabled;
 	eee->tx_lpi_timer = edata->tx_lpi_timer;
 eee_ok:
@@ -1986,14 +4373,16 @@ eee_ok:
 	if (netif_running(dev))
 		rc = bnxt_hwrm_set_link_setting(bp, false, true);
 
+eee_exit:
+	mutex_unlock(&bp->link_lock);
 	return rc;
 }
 
-static int bnxt_get_eee(struct net_device *dev, struct ethtool_eee *edata)
+static int bnxt_get_eee(struct net_device *dev, struct ethtool_keee *edata)
 {
 	struct bnxt *bp = netdev_priv(dev);
 
-	if (!(bp->flags & BNXT_FLAG_EEE_CAP))
+	if (!(bp->phy_flags & BNXT_PHY_FL_EEE_CAP))
 		return -EOPNOTSUPP;
 
 	*edata = bp->eee;
@@ -2001,45 +4390,135 @@ static int bnxt_get_eee(struct net_device *dev, struct ethtool_eee *edata)
 		/* Preserve tx_lpi_timer so that the last value will be used
 		 * by default when it is re-enabled.
 		 */
-		edata->advertised = 0;
+		linkmode_zero(edata->advertised);
 		edata->tx_lpi_enabled = 0;
 	}
 
 	if (!bp->eee.eee_active)
-		edata->lp_advertised = 0;
+		linkmode_zero(edata->lp_advertised);
+
+	return 0;
+}
+
+static int bnxt_hwrm_pfcwd_qcfg(struct bnxt *bp, u16 *val)
+{
+	struct hwrm_queue_pfcwd_timeout_qcfg_output *resp;
+	struct hwrm_queue_pfcwd_timeout_qcfg_input *req;
+	int rc;
+
+	rc = hwrm_req_init(bp, req, HWRM_QUEUE_PFCWD_TIMEOUT_QCFG);
+	if (rc)
+		return rc;
+	resp = hwrm_req_hold(bp, req);
+	rc = hwrm_req_send(bp, req);
+	if (!rc)
+		*val = le16_to_cpu(resp->pfcwd_timeout_value);
+	hwrm_req_drop(bp, req);
+	return rc;
+}
+
+static int bnxt_hwrm_pfcwd_cfg(struct bnxt *bp, u16 val)
+{
+	struct hwrm_queue_pfcwd_timeout_cfg_input *req;
+	int rc;
+
+	rc = hwrm_req_init(bp, req, HWRM_QUEUE_PFCWD_TIMEOUT_CFG);
+	if (rc)
+		return rc;
+	req->pfcwd_timeout_value = cpu_to_le16(val);
+	rc = hwrm_req_send(bp, req);
+	return rc;
+}
+
+static int bnxt_set_tunable(struct net_device *dev,
+			    const struct ethtool_tunable *tuna,
+			    const void *data)
+{
+	struct bnxt *bp = netdev_priv(dev);
+	u32 rx_copybreak, val;
+
+	switch (tuna->id) {
+	case ETHTOOL_RX_COPYBREAK:
+		rx_copybreak = *(u32 *)data;
+		if (rx_copybreak > BNXT_MAX_RX_COPYBREAK)
+			return -ERANGE;
+		if (rx_copybreak != bp->rx_copybreak) {
+			if (netif_running(dev))
+				return -EBUSY;
+			bp->rx_copybreak = rx_copybreak;
+		}
+		return 0;
+	case ETHTOOL_PFC_PREVENTION_TOUT:
+		if (BNXT_VF(bp) || !bp->max_pfcwd_tmo_ms)
+			return -EOPNOTSUPP;
+
+		val = *(u16 *)data;
+		if (val > bp->max_pfcwd_tmo_ms &&
+		    val != PFC_STORM_PREVENTION_AUTO)
+			return -EINVAL;
+		return bnxt_hwrm_pfcwd_cfg(bp, val);
+	default:
+		return -EOPNOTSUPP;
+	}
+}
+
+static int bnxt_get_tunable(struct net_device *dev,
+			    const struct ethtool_tunable *tuna, void *data)
+{
+	struct bnxt *bp = netdev_priv(dev);
+
+	switch (tuna->id) {
+	case ETHTOOL_RX_COPYBREAK:
+		*(u32 *)data = bp->rx_copybreak;
+		break;
+	case ETHTOOL_PFC_PREVENTION_TOUT:
+		if (!bp->max_pfcwd_tmo_ms)
+			return -EOPNOTSUPP;
+		return bnxt_hwrm_pfcwd_qcfg(bp, data);
+	default:
+		return -EOPNOTSUPP;
+	}
 
 	return 0;
 }
 
 static int bnxt_read_sfp_module_eeprom_info(struct bnxt *bp, u16 i2c_addr,
-					    u16 page_number, u16 start_addr,
-					    u16 data_length, u8 *buf)
+					    u16 page_number, u8 bank,
+					    u16 start_addr, u16 data_length,
+					    u8 *buf)
 {
-	struct hwrm_port_phy_i2c_read_input req = {0};
-	struct hwrm_port_phy_i2c_read_output *output = bp->hwrm_cmd_resp_addr;
+	struct hwrm_port_phy_i2c_read_output *output;
+	struct hwrm_port_phy_i2c_read_input *req;
 	int rc, byte_offset = 0;
 
-	bnxt_hwrm_cmd_hdr_init(bp, &req, HWRM_PORT_PHY_I2C_READ, -1, -1);
-	req.i2c_slave_addr = i2c_addr;
-	req.page_number = cpu_to_le16(page_number);
-	req.port_id = cpu_to_le16(bp->pf.port_id);
+	rc = hwrm_req_init(bp, req, HWRM_PORT_PHY_I2C_READ);
+	if (rc)
+		return rc;
+
+	output = hwrm_req_hold(bp, req);
+	req->i2c_slave_addr = i2c_addr;
+	req->page_number = cpu_to_le16(page_number);
+	req->port_id = cpu_to_le16(bp->pf.port_id);
 	do {
 		u16 xfer_size;
 
 		xfer_size = min_t(u16, data_length, BNXT_MAX_PHY_I2C_RESP_SIZE);
 		data_length -= xfer_size;
-		req.page_offset = cpu_to_le16(start_addr + byte_offset);
-		req.data_length = xfer_size;
-		req.enables = cpu_to_le32(start_addr + byte_offset ?
-				 PORT_PHY_I2C_READ_REQ_ENABLES_PAGE_OFFSET : 0);
-		mutex_lock(&bp->hwrm_cmd_lock);
-		rc = _hwrm_send_message(bp, &req, sizeof(req),
-					HWRM_CMD_TIMEOUT);
+		req->page_offset = cpu_to_le16(start_addr + byte_offset);
+		req->data_length = xfer_size;
+		req->enables =
+			cpu_to_le32((start_addr + byte_offset ?
+				     PORT_PHY_I2C_READ_REQ_ENABLES_PAGE_OFFSET :
+				     0) |
+				    (bank ?
+				     PORT_PHY_I2C_READ_REQ_ENABLES_BANK_NUMBER :
+				     0));
+		rc = hwrm_req_send(bp, req);
 		if (!rc)
 			memcpy(buf + byte_offset, output->data, xfer_size);
-		mutex_unlock(&bp->hwrm_cmd_lock);
 		byte_offset += xfer_size;
 	} while (!rc && data_length > 0);
+	hwrm_req_drop(bp, req);
 
 	return rc;
 }
@@ -2047,11 +4526,13 @@ static int bnxt_read_sfp_module_eeprom_info(struct bnxt *bp, u16 i2c_addr,
 static int bnxt_get_module_info(struct net_device *dev,
 				struct ethtool_modinfo *modinfo)
 {
+	u8 data[SFF_DIAG_SUPPORT_OFFSET + 1];
 	struct bnxt *bp = netdev_priv(dev);
-	struct hwrm_port_phy_i2c_read_input req = {0};
-	struct hwrm_port_phy_i2c_read_output *output = bp->hwrm_cmd_resp_addr;
 	int rc;
 
+	if (BNXT_VF(bp) && !BNXT_VF_IS_TRUSTED(bp))
+		return -EPERM;
+
 	/* No point in going further if phy status indicates
 	 * module is not inserted or if it is powered down or
 	 * if it is of type 10GBase-T
@@ -2064,21 +4545,19 @@ static int bnxt_get_module_info(struct net_device *dev,
 	if (bp->hwrm_spec_code < 0x10202)
 		return -EOPNOTSUPP;
 
-	bnxt_hwrm_cmd_hdr_init(bp, &req, HWRM_PORT_PHY_I2C_READ, -1, -1);
-	req.i2c_slave_addr = I2C_DEV_ADDR_A0;
-	req.page_number = 0;
-	req.page_offset = cpu_to_le16(SFP_EEPROM_SFF_8472_COMP_ADDR);
-	req.data_length = SFP_EEPROM_SFF_8472_COMP_SIZE;
-	req.port_id = cpu_to_le16(bp->pf.port_id);
-	mutex_lock(&bp->hwrm_cmd_lock);
-	rc = _hwrm_send_message(bp, &req, sizeof(req), HWRM_CMD_TIMEOUT);
+	rc = bnxt_read_sfp_module_eeprom_info(bp, I2C_DEV_ADDR_A0, 0, 0, 0,
+					      SFF_DIAG_SUPPORT_OFFSET + 1,
+					      data);
 	if (!rc) {
-		u32 module_id = le32_to_cpu(output->data[0]);
+		u8 module_id = data[0];
+		u8 diag_supported = data[SFF_DIAG_SUPPORT_OFFSET];
 
 		switch (module_id) {
 		case SFF_MODULE_ID_SFP:
 			modinfo->type = ETH_MODULE_SFF_8472;
 			modinfo->eeprom_len = ETH_MODULE_SFF_8472_LEN;
+			if (!diag_supported)
+				modinfo->eeprom_len = ETH_MODULE_SFF_8436_LEN;
 			break;
 		case SFF_MODULE_ID_QSFP:
 		case SFF_MODULE_ID_QSFP_PLUS:
@@ -2094,7 +4573,6 @@ static int bnxt_get_module_info(struct net_device *dev,
 			break;
 		}
 	}
-	mutex_unlock(&bp->hwrm_cmd_lock);
 	return rc;
 }
 
@@ -2106,13 +4584,16 @@ static int bnxt_get_module_eeprom(struct net_device *dev,
 	u16  start = eeprom->offset, length = eeprom->len;
 	int rc = 0;
 
+	if (BNXT_VF(bp) && !BNXT_VF_IS_TRUSTED(bp))
+		return -EPERM;
+
 	memset(data, 0, eeprom->len);
 
 	/* Read A0 portion of the EEPROM */
 	if (start < ETH_MODULE_SFF_8436_LEN) {
 		if (start + eeprom->len > ETH_MODULE_SFF_8436_LEN)
 			length = ETH_MODULE_SFF_8436_LEN - start;
-		rc = bnxt_read_sfp_module_eeprom_info(bp, I2C_DEV_ADDR_A0, 0,
+		rc = bnxt_read_sfp_module_eeprom_info(bp, I2C_DEV_ADDR_A0, 0, 0,
 						      start, length, data);
 		if (rc)
 			return rc;
@@ -2124,12 +4605,143 @@ static int bnxt_get_module_eeprom(struct net_device *dev,
 	/* Read A2 portion of the EEPROM */
 	if (length) {
 		start -= ETH_MODULE_SFF_8436_LEN;
-		bnxt_read_sfp_module_eeprom_info(bp, I2C_DEV_ADDR_A2, 1, start,
-						 length, data);
+		rc = bnxt_read_sfp_module_eeprom_info(bp, I2C_DEV_ADDR_A2, 0, 0,
+						      start, length, data);
 	}
 	return rc;
 }
 
+static int bnxt_get_module_status(struct bnxt *bp, struct netlink_ext_ack *extack)
+{
+	if (bp->link_info.module_status <=
+	    PORT_PHY_QCFG_RESP_MODULE_STATUS_WARNINGMSG)
+		return 0;
+
+	switch (bp->link_info.module_status) {
+	case PORT_PHY_QCFG_RESP_MODULE_STATUS_PWRDOWN:
+		NL_SET_ERR_MSG_MOD(extack, "Transceiver module is powering down");
+		break;
+	case PORT_PHY_QCFG_RESP_MODULE_STATUS_NOTINSERTED:
+		NL_SET_ERR_MSG_MOD(extack, "Transceiver module not inserted");
+		break;
+	case PORT_PHY_QCFG_RESP_MODULE_STATUS_CURRENTFAULT:
+		NL_SET_ERR_MSG_MOD(extack, "Transceiver module disabled due to current fault");
+		break;
+	default:
+		NL_SET_ERR_MSG_MOD(extack, "Unknown error");
+		break;
+	}
+	return -EINVAL;
+}
+
+static int
+bnxt_mod_eeprom_by_page_precheck(struct bnxt *bp,
+				 const struct ethtool_module_eeprom *page_data,
+				 struct netlink_ext_ack *extack)
+{
+	int rc;
+
+	if (BNXT_VF(bp) && !BNXT_VF_IS_TRUSTED(bp)) {
+		NL_SET_ERR_MSG_MOD(extack,
+				   "Module read/write not permitted on untrusted VF");
+		return -EPERM;
+	}
+
+	rc = bnxt_get_module_status(bp, extack);
+	if (rc)
+		return rc;
+
+	if (bp->hwrm_spec_code < 0x10202) {
+		NL_SET_ERR_MSG_MOD(extack, "Firmware version too old");
+		return -EINVAL;
+	}
+
+	if (page_data->bank && !(bp->phy_flags & BNXT_PHY_FL_BANK_SEL)) {
+		NL_SET_ERR_MSG_MOD(extack, "Firmware not capable for bank selection");
+		return -EINVAL;
+	}
+	return 0;
+}
+
+static int bnxt_get_module_eeprom_by_page(struct net_device *dev,
+					  const struct ethtool_module_eeprom *page_data,
+					  struct netlink_ext_ack *extack)
+{
+	struct bnxt *bp = netdev_priv(dev);
+	int rc;
+
+	rc = bnxt_mod_eeprom_by_page_precheck(bp, page_data, extack);
+	if (rc)
+		return rc;
+
+	rc = bnxt_read_sfp_module_eeprom_info(bp, page_data->i2c_address << 1,
+					      page_data->page, page_data->bank,
+					      page_data->offset,
+					      page_data->length,
+					      page_data->data);
+	if (rc) {
+		NL_SET_ERR_MSG_MOD(extack, "Module`s eeprom read failed");
+		return rc;
+	}
+	return page_data->length;
+}
+
+static int bnxt_write_sfp_module_eeprom_info(struct bnxt *bp,
+					     const struct ethtool_module_eeprom *page)
+{
+	struct hwrm_port_phy_i2c_write_input *req;
+	int bytes_written = 0;
+	int rc;
+
+	rc = hwrm_req_init(bp, req, HWRM_PORT_PHY_I2C_WRITE);
+	if (rc)
+		return rc;
+
+	hwrm_req_hold(bp, req);
+	req->i2c_slave_addr = page->i2c_address << 1;
+	req->page_number = cpu_to_le16(page->page);
+	req->bank_number = page->bank;
+	req->port_id = cpu_to_le16(bp->pf.port_id);
+	req->enables = cpu_to_le32(PORT_PHY_I2C_WRITE_REQ_ENABLES_PAGE_OFFSET |
+				   PORT_PHY_I2C_WRITE_REQ_ENABLES_BANK_NUMBER);
+
+	while (bytes_written < page->length) {
+		u16 xfer_size;
+
+		xfer_size = min_t(u16, page->length - bytes_written,
+				  BNXT_MAX_PHY_I2C_RESP_SIZE);
+		req->page_offset = cpu_to_le16(page->offset + bytes_written);
+		req->data_length = xfer_size;
+		memcpy(req->data, page->data + bytes_written, xfer_size);
+		rc = hwrm_req_send(bp, req);
+		if (rc)
+			break;
+		bytes_written += xfer_size;
+	}
+
+	hwrm_req_drop(bp, req);
+	return rc;
+}
+
+static int bnxt_set_module_eeprom_by_page(struct net_device *dev,
+					  const struct ethtool_module_eeprom *page_data,
+					  struct netlink_ext_ack *extack)
+{
+	struct bnxt *bp = netdev_priv(dev);
+	int rc;
+
+	rc = bnxt_mod_eeprom_by_page_precheck(bp, page_data, extack);
+	if (rc)
+		return rc;
+
+	rc = bnxt_write_sfp_module_eeprom_info(bp, page_data);
+	if (rc) {
+		NL_SET_ERR_MSG_MOD(extack, "Module`s eeprom write failed");
+		return rc;
+	}
+	return page_data->length;
+}
+
 static int bnxt_nway_reset(struct net_device *dev)
 {
 	int rc = 0;
@@ -2137,7 +4749,7 @@ static int bnxt_nway_reset(struct net_device *dev)
 	struct bnxt *bp = netdev_priv(dev);
 	struct bnxt_link_info *link_info = &bp->link_info;
 
-	if (!BNXT_SINGLE_PF(bp))
+	if (!BNXT_PHY_CFG_ABLE(bp))
 		return -EOPNOTSUPP;
 
 	if (!(link_info->autoneg & BNXT_AUTONEG_SPEED))
@@ -2152,13 +4764,13 @@ static int bnxt_nway_reset(struct net_device *dev)
 static int bnxt_set_phys_id(struct net_device *dev,
 			    enum ethtool_phys_id_state state)
 {
-	struct hwrm_port_led_cfg_input req = {0};
+	struct hwrm_port_led_cfg_input *req;
 	struct bnxt *bp = netdev_priv(dev);
 	struct bnxt_pf_info *pf = &bp->pf;
 	struct bnxt_led_cfg *led_cfg;
 	u8 led_state;
 	__le16 duration;
-	int i, rc;
+	int rc, i;
 
 	if (!bp->num_leds || BNXT_VF(bp))
 		return -EOPNOTSUPP;
@@ -2172,30 +4784,35 @@ static int bnxt_set_phys_id(struct net_device *dev,
 	} else {
 		return -EINVAL;
 	}
-	bnxt_hwrm_cmd_hdr_init(bp, &req, HWRM_PORT_LED_CFG, -1, -1);
-	req.port_id = cpu_to_le16(pf->port_id);
-	req.num_leds = bp->num_leds;
-	led_cfg = (struct bnxt_led_cfg *)&req.led0_id;
+	rc = hwrm_req_init(bp, req, HWRM_PORT_LED_CFG);
+	if (rc)
+		return rc;
+
+	req->port_id = cpu_to_le16(pf->port_id);
+	req->num_leds = bp->num_leds;
+	led_cfg = (struct bnxt_led_cfg *)&req->led0_id;
 	for (i = 0; i < bp->num_leds; i++, led_cfg++) {
-		req.enables |= BNXT_LED_DFLT_ENABLES(i);
+		req->enables |= BNXT_LED_DFLT_ENABLES(i);
 		led_cfg->led_id = bp->leds[i].led_id;
 		led_cfg->led_state = led_state;
 		led_cfg->led_blink_on = duration;
 		led_cfg->led_blink_off = duration;
 		led_cfg->led_group_id = bp->leds[i].led_group_id;
 	}
-	rc = hwrm_send_message(bp, &req, sizeof(req), HWRM_CMD_TIMEOUT);
-	if (rc)
-		rc = -EIO;
-	return rc;
+	return hwrm_req_send(bp, req);
 }
 
 static int bnxt_hwrm_selftest_irq(struct bnxt *bp, u16 cmpl_ring)
 {
-	struct hwrm_selftest_irq_input req = {0};
+	struct hwrm_selftest_irq_input *req;
+	int rc;
 
-	bnxt_hwrm_cmd_hdr_init(bp, &req, HWRM_SELFTEST_IRQ, cmpl_ring, -1);
-	return hwrm_send_message(bp, &req, sizeof(req), HWRM_CMD_TIMEOUT);
+	rc = hwrm_req_init(bp, req, HWRM_SELFTEST_IRQ);
+	if (rc)
+		return rc;
+
+	req->cmpl_ring = cpu_to_le16(cmpl_ring);
+	return hwrm_req_send(bp, req);
 }
 
 static int bnxt_test_irq(struct bnxt *bp)
@@ -2215,31 +4832,58 @@ static int bnxt_test_irq(struct bnxt *bp)
 
 static int bnxt_hwrm_mac_loopback(struct bnxt *bp, bool enable)
 {
-	struct hwrm_port_mac_cfg_input req = {0};
+	struct hwrm_port_mac_cfg_input *req;
+	int rc;
 
-	bnxt_hwrm_cmd_hdr_init(bp, &req, HWRM_PORT_MAC_CFG, -1, -1);
+	rc = hwrm_req_init(bp, req, HWRM_PORT_MAC_CFG);
+	if (rc)
+		return rc;
 
-	req.enables = cpu_to_le32(PORT_MAC_CFG_REQ_ENABLES_LPBK);
+	req->enables = cpu_to_le32(PORT_MAC_CFG_REQ_ENABLES_LPBK);
 	if (enable)
-		req.lpbk = PORT_MAC_CFG_REQ_LPBK_LOCAL;
+		req->lpbk = PORT_MAC_CFG_REQ_LPBK_LOCAL;
 	else
-		req.lpbk = PORT_MAC_CFG_REQ_LPBK_NONE;
-	return hwrm_send_message(bp, &req, sizeof(req), HWRM_CMD_TIMEOUT);
+		req->lpbk = PORT_MAC_CFG_REQ_LPBK_NONE;
+	return hwrm_req_send(bp, req);
+}
+
+static int bnxt_query_force_speeds(struct bnxt *bp, u16 *force_speeds)
+{
+	struct hwrm_port_phy_qcaps_output *resp;
+	struct hwrm_port_phy_qcaps_input *req;
+	int rc;
+
+	rc = hwrm_req_init(bp, req, HWRM_PORT_PHY_QCAPS);
+	if (rc)
+		return rc;
+
+	resp = hwrm_req_hold(bp, req);
+	rc = hwrm_req_send(bp, req);
+	if (!rc)
+		*force_speeds = le16_to_cpu(resp->supported_speeds_force_mode);
+
+	hwrm_req_drop(bp, req);
+	return rc;
 }
 
 static int bnxt_disable_an_for_lpbk(struct bnxt *bp,
 				    struct hwrm_port_phy_cfg_input *req)
 {
 	struct bnxt_link_info *link_info = &bp->link_info;
-	u16 fw_advertising = link_info->advertising;
+	u16 fw_advertising;
 	u16 fw_speed;
 	int rc;
 
-	if (!link_info->autoneg)
+	if (!link_info->autoneg ||
+	    (bp->phy_flags & BNXT_PHY_FL_AN_PHY_LPBK))
 		return 0;
 
+	rc = bnxt_query_force_speeds(bp, &fw_advertising);
+	if (rc)
+		return rc;
+
 	fw_speed = PORT_PHY_CFG_REQ_FORCE_LINK_SPEED_1GB;
-	if (netif_carrier_ok(bp->dev))
+	if (BNXT_LINK_IS_UP(bp))
 		fw_speed = bp->link_info.link_speed;
 	else if (fw_advertising & BNXT_LINK_SPEED_MSK_10GB)
 		fw_speed = PORT_PHY_CFG_REQ_FORCE_LINK_SPEED_10GB;
@@ -2253,33 +4897,44 @@ static int bnxt_disable_an_for_lpbk(struct bnxt *bp,
 	req->force_link_speed = cpu_to_le16(fw_speed);
 	req->flags |= cpu_to_le32(PORT_PHY_CFG_REQ_FLAGS_FORCE |
 				  PORT_PHY_CFG_REQ_FLAGS_RESET_PHY);
-	rc = hwrm_send_message(bp, req, sizeof(*req), HWRM_CMD_TIMEOUT);
+	rc = hwrm_req_send(bp, req);
 	req->flags = 0;
 	req->force_link_speed = cpu_to_le16(0);
 	return rc;
 }
 
-static int bnxt_hwrm_phy_loopback(struct bnxt *bp, bool enable)
+static int bnxt_hwrm_phy_loopback(struct bnxt *bp, bool enable, bool ext)
 {
-	struct hwrm_port_phy_cfg_input req = {0};
+	struct hwrm_port_phy_cfg_input *req;
+	int rc;
+
+	rc = hwrm_req_init(bp, req, HWRM_PORT_PHY_CFG);
+	if (rc)
+		return rc;
 
-	bnxt_hwrm_cmd_hdr_init(bp, &req, HWRM_PORT_PHY_CFG, -1, -1);
+	/* prevent bnxt_disable_an_for_lpbk() from consuming the request */
+	hwrm_req_hold(bp, req);
 
 	if (enable) {
-		bnxt_disable_an_for_lpbk(bp, &req);
-		req.lpbk = PORT_PHY_CFG_REQ_LPBK_LOCAL;
+		bnxt_disable_an_for_lpbk(bp, req);
+		if (ext)
+			req->lpbk = PORT_PHY_CFG_REQ_LPBK_EXTERNAL;
+		else
+			req->lpbk = PORT_PHY_CFG_REQ_LPBK_LOCAL;
 	} else {
-		req.lpbk = PORT_PHY_CFG_REQ_LPBK_NONE;
+		req->lpbk = PORT_PHY_CFG_REQ_LPBK_NONE;
 	}
-	req.enables = cpu_to_le32(PORT_PHY_CFG_REQ_ENABLES_LPBK);
-	return hwrm_send_message(bp, &req, sizeof(req), HWRM_CMD_TIMEOUT);
+	req->enables = cpu_to_le32(PORT_PHY_CFG_REQ_ENABLES_LPBK);
+	rc = hwrm_req_send(bp, req);
+	hwrm_req_drop(bp, req);
+	return rc;
 }
 
-static int bnxt_rx_loopback(struct bnxt *bp, struct bnxt_napi *bnapi,
+static int bnxt_rx_loopback(struct bnxt *bp, struct bnxt_cp_ring_info *cpr,
 			    u32 raw_cons, int pkt_size)
 {
-	struct bnxt_cp_ring_info *cpr = &bnapi->cp_ring;
-	struct bnxt_rx_ring_info *rxr = bnapi->rx_ring;
+	struct bnxt_napi *bnapi = cpr->bnapi;
+	struct bnxt_rx_ring_info *rxr;
 	struct bnxt_sw_rx_bd *rx_buf;
 	struct rx_cmp *rxcmp;
 	u16 cp_cons, cons;
@@ -2287,6 +4942,7 @@ static int bnxt_rx_loopback(struct bnxt *bp, struct bnxt_napi *bnapi,
 	u32 len;
 	int i;
 
+	rxr = bnapi->rx_ring;
 	cp_cons = RING_CMP(raw_cons);
 	rxcmp = (struct rx_cmp *)
 		&cpr->cp_desc_ring[CP_RING(cp_cons)][CP_IDX(cp_cons)];
@@ -2307,17 +4963,15 @@ static int bnxt_rx_loopback(struct bnxt *bp, struct bnxt_napi *bnapi,
 	return 0;
 }
 
-static int bnxt_poll_loopback(struct bnxt *bp, int pkt_size)
+static int bnxt_poll_loopback(struct bnxt *bp, struct bnxt_cp_ring_info *cpr,
+			      int pkt_size)
 {
-	struct bnxt_napi *bnapi = bp->bnapi[0];
-	struct bnxt_cp_ring_info *cpr;
 	struct tx_cmp *txcmp;
 	int rc = -EIO;
 	u32 raw_cons;
 	u32 cons;
 	int i;
 
-	cpr = &bnapi->cp_ring;
 	raw_cons = cpr->cp_raw_cons;
 	for (i = 0; i < 200; i++) {
 		cons = RING_CMP(raw_cons);
@@ -2332,8 +4986,9 @@ static int bnxt_poll_loopback(struct bnxt *bp, int pkt_size)
 		 * reading any further.
 		 */
 		dma_rmb();
-		if (TX_CMP_TYPE(txcmp) == CMP_TYPE_RX_L2_CMP) {
-			rc = bnxt_rx_loopback(bp, bnapi, raw_cons, pkt_size);
+		if (TX_CMP_TYPE(txcmp) == CMP_TYPE_RX_L2_CMP ||
+		    TX_CMP_TYPE(txcmp) == CMP_TYPE_RX_L2_V3_CMP) {
+			rc = bnxt_rx_loopback(bp, cpr, raw_cons, pkt_size);
 			raw_cons = NEXT_RAW_CMP(raw_cons);
 			raw_cons = NEXT_RAW_CMP(raw_cons);
 			break;
@@ -2347,18 +5002,24 @@ static int bnxt_poll_loopback(struct bnxt *bp, int pkt_size)
 static int bnxt_run_loopback(struct bnxt *bp)
 {
 	struct bnxt_tx_ring_info *txr = &bp->tx_ring[0];
+	struct bnxt_rx_ring_info *rxr = &bp->rx_ring[0];
+	struct bnxt_cp_ring_info *cpr;
 	int pkt_size, i = 0;
 	struct sk_buff *skb;
 	dma_addr_t map;
 	u8 *data;
 	int rc;
 
-	pkt_size = min(bp->dev->mtu + ETH_HLEN, bp->rx_copy_thresh);
+	cpr = &rxr->bnapi->cp_ring;
+	if (bp->flags & BNXT_FLAG_CHIP_P5_PLUS)
+		cpr = rxr->rx_cpr;
+	pkt_size = min(bp->dev->mtu + ETH_HLEN, max(BNXT_DEFAULT_RX_COPYBREAK,
+						    bp->rx_copybreak));
 	skb = netdev_alloc_skb(bp->dev, pkt_size);
 	if (!skb)
 		return -ENOMEM;
 	data = skb_put(skb, pkt_size);
-	eth_broadcast_addr(data);
+	ether_addr_copy(&data[i], bp->dev->dev_addr);
 	i += ETH_ALEN;
 	ether_addr_copy(&data[i], bp->dev->dev_addr);
 	i += ETH_ALEN;
@@ -2366,66 +5027,84 @@ static int bnxt_run_loopback(struct bnxt *bp)
 		data[i] = (u8)(i & 0xff);
 
 	map = dma_map_single(&bp->pdev->dev, skb->data, pkt_size,
-			     PCI_DMA_TODEVICE);
+			     DMA_TO_DEVICE);
 	if (dma_mapping_error(&bp->pdev->dev, map)) {
 		dev_kfree_skb(skb);
 		return -EIO;
 	}
-	bnxt_xmit_xdp(bp, txr, map, pkt_size, 0);
+	bnxt_xmit_bd(bp, txr, map, pkt_size, NULL);
 
 	/* Sync BD data before updating doorbell */
 	wmb();
 
-	bnxt_db_write(bp, txr->tx_doorbell, DB_KEY_TX | txr->tx_prod);
-	rc = bnxt_poll_loopback(bp, pkt_size);
+	bnxt_db_write(bp, &txr->tx_db, txr->tx_prod);
+	rc = bnxt_poll_loopback(bp, cpr, pkt_size);
 
-	dma_unmap_single(&bp->pdev->dev, map, pkt_size, PCI_DMA_TODEVICE);
+	dma_unmap_single(&bp->pdev->dev, map, pkt_size, DMA_TO_DEVICE);
 	dev_kfree_skb(skb);
 	return rc;
 }
 
 static int bnxt_run_fw_tests(struct bnxt *bp, u8 test_mask, u8 *test_results)
 {
-	struct hwrm_selftest_exec_output *resp = bp->hwrm_cmd_resp_addr;
-	struct hwrm_selftest_exec_input req = {0};
+	struct hwrm_selftest_exec_output *resp;
+	struct hwrm_selftest_exec_input *req;
 	int rc;
 
-	bnxt_hwrm_cmd_hdr_init(bp, &req, HWRM_SELFTEST_EXEC, -1, -1);
-	mutex_lock(&bp->hwrm_cmd_lock);
-	resp->test_success = 0;
-	req.flags = test_mask;
-	rc = _hwrm_send_message(bp, &req, sizeof(req), bp->test_info->timeout);
+	rc = hwrm_req_init(bp, req, HWRM_SELFTEST_EXEC);
+	if (rc)
+		return rc;
+
+	hwrm_req_timeout(bp, req, bp->test_info->timeout);
+	req->flags = test_mask;
+
+	resp = hwrm_req_hold(bp, req);
+	rc = hwrm_req_send(bp, req);
 	*test_results = resp->test_success;
-	mutex_unlock(&bp->hwrm_cmd_lock);
+	hwrm_req_drop(bp, req);
 	return rc;
 }
 
-#define BNXT_DRV_TESTS			3
+#define BNXT_DRV_TESTS			4
 #define BNXT_MACLPBK_TEST_IDX		(bp->num_tests - BNXT_DRV_TESTS)
 #define BNXT_PHYLPBK_TEST_IDX		(BNXT_MACLPBK_TEST_IDX + 1)
-#define BNXT_IRQ_TEST_IDX		(BNXT_MACLPBK_TEST_IDX + 2)
+#define BNXT_EXTLPBK_TEST_IDX		(BNXT_MACLPBK_TEST_IDX + 2)
+#define BNXT_IRQ_TEST_IDX		(BNXT_MACLPBK_TEST_IDX + 3)
 
 static void bnxt_self_test(struct net_device *dev, struct ethtool_test *etest,
 			   u64 *buf)
 {
 	struct bnxt *bp = netdev_priv(dev);
+	bool do_ext_lpbk = false;
 	bool offline = false;
 	u8 test_results = 0;
 	u8 test_mask = 0;
-	int rc, i;
+	int rc = 0, i;
 
-	if (!bp->num_tests || !BNXT_SINGLE_PF(bp))
+	if (!bp->num_tests || !BNXT_PF(bp))
 		return;
+
 	memset(buf, 0, sizeof(u64) * bp->num_tests);
+	if (etest->flags & ETH_TEST_FL_OFFLINE &&
+	    bnxt_ulp_registered(bp->edev)) {
+		etest->flags |= ETH_TEST_FL_FAILED;
+		netdev_warn(dev, "Offline tests cannot be run with RoCE driver loaded\n");
+		return;
+	}
+
 	if (!netif_running(dev)) {
 		etest->flags |= ETH_TEST_FL_FAILED;
 		return;
 	}
 
+	if ((etest->flags & ETH_TEST_FL_EXTERNAL_LB) &&
+	    (bp->phy_flags & BNXT_PHY_FL_EXT_LPBK))
+		do_ext_lpbk = true;
+
 	if (etest->flags & ETH_TEST_FL_OFFLINE) {
-		if (bp->pf.active_vfs) {
+		if (bp->pf.active_vfs || !BNXT_SINGLE_PF(bp)) {
 			etest->flags |= ETH_TEST_FL_FAILED;
-			netdev_warn(dev, "Offline tests cannot be run with active VFs\n");
+			netdev_warn(dev, "Offline tests cannot be run with active VFs or on shared PF\n");
 			return;
 		}
 		offline = true;
@@ -2442,37 +5121,53 @@ static void bnxt_self_test(struct net_device *dev, struct ethtool_test *etest,
 	if (!offline) {
 		bnxt_run_fw_tests(bp, test_mask, &test_results);
 	} else {
-		rc = bnxt_close_nic(bp, false, false);
-		if (rc)
-			return;
+		bnxt_close_nic(bp, true, false);
 		bnxt_run_fw_tests(bp, test_mask, &test_results);
 
-		buf[BNXT_MACLPBK_TEST_IDX] = 1;
-		bnxt_hwrm_mac_loopback(bp, true);
-		msleep(250);
 		rc = bnxt_half_open_nic(bp);
 		if (rc) {
-			bnxt_hwrm_mac_loopback(bp, false);
 			etest->flags |= ETH_TEST_FL_FAILED;
 			return;
 		}
+		buf[BNXT_MACLPBK_TEST_IDX] = 1;
+		if (bp->mac_flags & BNXT_MAC_FL_NO_MAC_LPBK)
+			goto skip_mac_loopback;
+
+		bnxt_hwrm_mac_loopback(bp, true);
+		msleep(250);
 		if (bnxt_run_loopback(bp))
 			etest->flags |= ETH_TEST_FL_FAILED;
 		else
 			buf[BNXT_MACLPBK_TEST_IDX] = 0;
 
 		bnxt_hwrm_mac_loopback(bp, false);
-		bnxt_hwrm_phy_loopback(bp, true);
+skip_mac_loopback:
+		buf[BNXT_PHYLPBK_TEST_IDX] = 1;
+		if (bp->phy_flags & BNXT_PHY_FL_NO_PHY_LPBK)
+			goto skip_phy_loopback;
+
+		bnxt_hwrm_phy_loopback(bp, true, false);
 		msleep(1000);
-		if (bnxt_run_loopback(bp)) {
-			buf[BNXT_PHYLPBK_TEST_IDX] = 1;
+		if (bnxt_run_loopback(bp))
 			etest->flags |= ETH_TEST_FL_FAILED;
+		else
+			buf[BNXT_PHYLPBK_TEST_IDX] = 0;
+skip_phy_loopback:
+		buf[BNXT_EXTLPBK_TEST_IDX] = 1;
+		if (do_ext_lpbk) {
+			etest->flags |= ETH_TEST_FL_EXTERNAL_LB_DONE;
+			bnxt_hwrm_phy_loopback(bp, true, true);
+			msleep(1000);
+			if (bnxt_run_loopback(bp))
+				etest->flags |= ETH_TEST_FL_FAILED;
+			else
+				buf[BNXT_EXTLPBK_TEST_IDX] = 0;
 		}
-		bnxt_hwrm_phy_loopback(bp, false);
+		bnxt_hwrm_phy_loopback(bp, false, false);
 		bnxt_half_close_nic(bp);
-		bnxt_open_nic(bp, false, true);
+		rc = bnxt_open_nic(bp, true, true);
 	}
-	if (bnxt_test_irq(bp)) {
+	if (rc || bnxt_test_irq(bp)) {
 		buf[BNXT_IRQ_TEST_IDX] = 1;
 		etest->flags |= ETH_TEST_FL_FAILED;
 	}
@@ -2486,27 +5181,198 @@ static void bnxt_self_test(struct net_device *dev, struct ethtool_test *etest,
 	}
 }
 
+static int bnxt_reset(struct net_device *dev, u32 *flags)
+{
+	struct bnxt *bp = netdev_priv(dev);
+	bool reload = false;
+	u32 req = *flags;
+
+	if (!req)
+		return -EINVAL;
+
+	if (!BNXT_PF(bp)) {
+		netdev_err(dev, "Reset is not supported from a VF\n");
+		return -EOPNOTSUPP;
+	}
+
+	if (pci_vfs_assigned(bp->pdev) &&
+	    !(bp->fw_cap & BNXT_FW_CAP_HOT_RESET)) {
+		netdev_err(dev,
+			   "Reset not allowed when VFs are assigned to VMs\n");
+		return -EBUSY;
+	}
+
+	if ((req & BNXT_FW_RESET_CHIP) == BNXT_FW_RESET_CHIP) {
+		/* This feature is not supported in older firmware versions */
+		if (bp->hwrm_spec_code >= 0x10803) {
+			if (!bnxt_firmware_reset_chip(dev)) {
+				netdev_info(dev, "Firmware reset request successful.\n");
+				if (!(bp->fw_cap & BNXT_FW_CAP_HOT_RESET))
+					reload = true;
+				*flags &= ~BNXT_FW_RESET_CHIP;
+			}
+		} else if (req == BNXT_FW_RESET_CHIP) {
+			return -EOPNOTSUPP; /* only request, fail hard */
+		}
+	}
+
+	if (!BNXT_CHIP_P4_PLUS(bp) && (req & BNXT_FW_RESET_AP)) {
+		/* This feature is not supported in older firmware versions */
+		if (bp->hwrm_spec_code >= 0x10803) {
+			if (!bnxt_firmware_reset_ap(dev)) {
+				netdev_info(dev, "Reset application processor successful.\n");
+				reload = true;
+				*flags &= ~BNXT_FW_RESET_AP;
+			}
+		} else if (req == BNXT_FW_RESET_AP) {
+			return -EOPNOTSUPP; /* only request, fail hard */
+		}
+	}
+
+	if (reload)
+		netdev_info(dev, "Reload driver to complete reset\n");
+
+	return 0;
+}
+
+static int bnxt_set_dump(struct net_device *dev, struct ethtool_dump *dump)
+{
+	struct bnxt *bp = netdev_priv(dev);
+
+	if (dump->flag > BNXT_DUMP_LIVE_WITH_CTX_L1_CACHE) {
+		netdev_info(dev,
+			    "Supports only Live(0), Crash(1), Driver(2), Live with cached context(3) dumps.\n");
+		return -EINVAL;
+	}
+
+	if (dump->flag == BNXT_DUMP_CRASH) {
+		if (bp->fw_dbg_cap & DBG_QCAPS_RESP_FLAGS_CRASHDUMP_SOC_DDR &&
+		    (!IS_ENABLED(CONFIG_TEE_BNXT_FW))) {
+			netdev_info(dev,
+				    "Cannot collect crash dump as TEE_BNXT_FW config option is not enabled.\n");
+			return -EOPNOTSUPP;
+		} else if (!(bp->fw_dbg_cap & DBG_QCAPS_RESP_FLAGS_CRASHDUMP_HOST_DDR)) {
+			netdev_info(dev, "Crash dump collection from host memory is not supported on this interface.\n");
+			return -EOPNOTSUPP;
+		}
+	}
+
+	bp->dump_flag = dump->flag;
+	return 0;
+}
+
+static int bnxt_get_dump_flag(struct net_device *dev, struct ethtool_dump *dump)
+{
+	struct bnxt *bp = netdev_priv(dev);
+
+	if (bp->hwrm_spec_code < 0x10801)
+		return -EOPNOTSUPP;
+
+	dump->version = bp->ver_resp.hwrm_fw_maj_8b << 24 |
+			bp->ver_resp.hwrm_fw_min_8b << 16 |
+			bp->ver_resp.hwrm_fw_bld_8b << 8 |
+			bp->ver_resp.hwrm_fw_rsvd_8b;
+
+	dump->flag = bp->dump_flag;
+	dump->len = bnxt_get_coredump_length(bp, bp->dump_flag);
+	return 0;
+}
+
+static int bnxt_get_dump_data(struct net_device *dev, struct ethtool_dump *dump,
+			      void *buf)
+{
+	struct bnxt *bp = netdev_priv(dev);
+
+	if (bp->hwrm_spec_code < 0x10801)
+		return -EOPNOTSUPP;
+
+	memset(buf, 0, dump->len);
+
+	dump->flag = bp->dump_flag;
+	return bnxt_get_coredump(bp, dump->flag, buf, &dump->len);
+}
+
+static int bnxt_get_ts_info(struct net_device *dev,
+			    struct kernel_ethtool_ts_info *info)
+{
+	struct bnxt *bp = netdev_priv(dev);
+	struct bnxt_ptp_cfg *ptp;
+
+	ptp = bp->ptp_cfg;
+	info->so_timestamping = SOF_TIMESTAMPING_TX_SOFTWARE;
+
+	if (!ptp)
+		return 0;
+
+	info->so_timestamping |= SOF_TIMESTAMPING_TX_HARDWARE |
+				 SOF_TIMESTAMPING_RX_HARDWARE |
+				 SOF_TIMESTAMPING_RAW_HARDWARE;
+	if (ptp->ptp_clock)
+		info->phc_index = ptp_clock_index(ptp->ptp_clock);
+
+	info->tx_types = (1 << HWTSTAMP_TX_OFF) | (1 << HWTSTAMP_TX_ON);
+
+	info->rx_filters = (1 << HWTSTAMP_FILTER_NONE) |
+			   (1 << HWTSTAMP_FILTER_PTP_V2_L2_EVENT) |
+			   (1 << HWTSTAMP_FILTER_PTP_V2_L4_EVENT);
+
+	if (bp->fw_cap & BNXT_FW_CAP_RX_ALL_PKT_TS)
+		info->rx_filters |= (1 << HWTSTAMP_FILTER_ALL);
+	return 0;
+}
+
+static void bnxt_hwrm_pcie_qstats(struct bnxt *bp)
+{
+	struct hwrm_pcie_qstats_output *resp;
+	struct hwrm_pcie_qstats_input *req;
+
+	bp->pcie_stat_len = 0;
+	if (!(bp->fw_cap & BNXT_FW_CAP_PCIE_STATS_SUPPORTED))
+		return;
+
+	if (hwrm_req_init(bp, req, HWRM_PCIE_QSTATS))
+		return;
+
+	resp = hwrm_req_hold(bp, req);
+	if (__bnxt_hwrm_pcie_qstats(bp, req))
+		bp->pcie_stat_len = min_t(u16,
+					  le16_to_cpu(resp->pcie_stat_size),
+					  sizeof(struct pcie_ctx_hw_stats_v2));
+	hwrm_req_drop(bp, req);
+}
+
 void bnxt_ethtool_init(struct bnxt *bp)
 {
-	struct hwrm_selftest_qlist_output *resp = bp->hwrm_cmd_resp_addr;
-	struct hwrm_selftest_qlist_input req = {0};
+	struct hwrm_selftest_qlist_output *resp;
+	struct hwrm_selftest_qlist_input *req;
 	struct bnxt_test_info *test_info;
+	struct net_device *dev = bp->dev;
 	int i, rc;
 
-	if (bp->hwrm_spec_code < 0x10704 || !BNXT_SINGLE_PF(bp))
+	bnxt_hwrm_pcie_qstats(bp);
+	if (!(bp->fw_cap & BNXT_FW_CAP_PKG_VER))
+		bnxt_get_pkgver(dev);
+
+	bp->num_tests = 0;
+	if (bp->hwrm_spec_code < 0x10704 || !BNXT_PF(bp))
 		return;
 
-	bnxt_hwrm_cmd_hdr_init(bp, &req, HWRM_SELFTEST_QLIST, -1, -1);
-	mutex_lock(&bp->hwrm_cmd_lock);
-	rc = _hwrm_send_message(bp, &req, sizeof(req), HWRM_CMD_TIMEOUT);
-	if (rc)
-		goto ethtool_init_exit;
+	test_info = bp->test_info;
+	if (!test_info) {
+		test_info = kzalloc(sizeof(*bp->test_info), GFP_KERNEL);
+		if (!test_info)
+			return;
+		bp->test_info = test_info;
+	}
+
+	if (hwrm_req_init(bp, req, HWRM_SELFTEST_QLIST))
+		return;
 
-	test_info = kzalloc(sizeof(*bp->test_info), GFP_KERNEL);
-	if (!test_info)
+	resp = hwrm_req_hold(bp, req);
+	rc = hwrm_req_send_silent(bp, req);
+	if (rc)
 		goto ethtool_init_exit;
 
-	bp->test_info = test_info;
 	bp->num_tests = resp->num_tests + BNXT_DRV_TESTS;
 	if (bp->num_tests > BNXT_MAX_TEST)
 		bp->num_tests = BNXT_MAX_TEST;
@@ -2517,28 +5383,179 @@ void bnxt_ethtool_init(struct bnxt *bp)
 		test_info->timeout = HWRM_CMD_TIMEOUT;
 	for (i = 0; i < bp->num_tests; i++) {
 		char *str = test_info->string[i];
-		char *fw_str = resp->test0_name + i * 32;
+		char *fw_str = resp->test_name[i];
 
 		if (i == BNXT_MACLPBK_TEST_IDX) {
 			strcpy(str, "Mac loopback test (offline)");
 		} else if (i == BNXT_PHYLPBK_TEST_IDX) {
 			strcpy(str, "Phy loopback test (offline)");
+		} else if (i == BNXT_EXTLPBK_TEST_IDX) {
+			strcpy(str, "Ext loopback test (offline)");
 		} else if (i == BNXT_IRQ_TEST_IDX) {
 			strcpy(str, "Interrupt_test (offline)");
 		} else {
-			strlcpy(str, fw_str, ETH_GSTRING_LEN);
-			strncat(str, " test", ETH_GSTRING_LEN - strlen(str));
-			if (test_info->offline_mask & (1 << i))
-				strncat(str, " (offline)",
-					ETH_GSTRING_LEN - strlen(str));
-			else
-				strncat(str, " (online)",
-					ETH_GSTRING_LEN - strlen(str));
+			snprintf(str, ETH_GSTRING_LEN, "%s test (%s)",
+				 fw_str, test_info->offline_mask & (1 << i) ?
+					"offline" : "online");
 		}
 	}
 
 ethtool_init_exit:
-	mutex_unlock(&bp->hwrm_cmd_lock);
+	hwrm_req_drop(bp, req);
+}
+
+static void bnxt_get_eth_phy_stats(struct net_device *dev,
+				   struct ethtool_eth_phy_stats *phy_stats)
+{
+	struct bnxt *bp = netdev_priv(dev);
+	u64 *rx;
+
+	if (BNXT_VF(bp) || !(bp->flags & BNXT_FLAG_PORT_STATS_EXT))
+		return;
+
+	rx = bp->rx_port_stats_ext.sw_stats;
+	phy_stats->SymbolErrorDuringCarrier =
+		*(rx + BNXT_RX_STATS_EXT_OFFSET(rx_pcs_symbol_err));
+}
+
+static void bnxt_get_eth_mac_stats(struct net_device *dev,
+				   struct ethtool_eth_mac_stats *mac_stats)
+{
+	struct bnxt *bp = netdev_priv(dev);
+	u64 *rx, *tx;
+
+	if (BNXT_VF(bp) || !(bp->flags & BNXT_FLAG_PORT_STATS))
+		return;
+
+	rx = bp->port_stats.sw_stats;
+	tx = bp->port_stats.sw_stats + BNXT_TX_PORT_STATS_BYTE_OFFSET / 8;
+
+	mac_stats->FramesReceivedOK =
+		BNXT_GET_RX_PORT_STATS64(rx, rx_good_frames);
+	mac_stats->FramesTransmittedOK =
+		BNXT_GET_TX_PORT_STATS64(tx, tx_good_frames);
+	mac_stats->FrameCheckSequenceErrors =
+		BNXT_GET_RX_PORT_STATS64(rx, rx_fcs_err_frames);
+	mac_stats->AlignmentErrors =
+		BNXT_GET_RX_PORT_STATS64(rx, rx_align_err_frames);
+	mac_stats->OutOfRangeLengthField =
+		BNXT_GET_RX_PORT_STATS64(rx, rx_oor_len_frames);
+}
+
+static void bnxt_get_eth_ctrl_stats(struct net_device *dev,
+				    struct ethtool_eth_ctrl_stats *ctrl_stats)
+{
+	struct bnxt *bp = netdev_priv(dev);
+	u64 *rx;
+
+	if (BNXT_VF(bp) || !(bp->flags & BNXT_FLAG_PORT_STATS))
+		return;
+
+	rx = bp->port_stats.sw_stats;
+	ctrl_stats->MACControlFramesReceived =
+		BNXT_GET_RX_PORT_STATS64(rx, rx_ctrl_frames);
+}
+
+static const struct ethtool_rmon_hist_range bnxt_rmon_ranges[] = {
+	{    0,    64 },
+	{   65,   127 },
+	{  128,   255 },
+	{  256,   511 },
+	{  512,  1023 },
+	{ 1024,  1518 },
+	{ 1519,  2047 },
+	{ 2048,  4095 },
+	{ 4096,  9216 },
+	{ 9217, 16383 },
+	{}
+};
+
+static void bnxt_get_rmon_stats(struct net_device *dev,
+				struct ethtool_rmon_stats *rmon_stats,
+				const struct ethtool_rmon_hist_range **ranges)
+{
+	struct bnxt *bp = netdev_priv(dev);
+	u64 *rx, *tx;
+
+	if (BNXT_VF(bp) || !(bp->flags & BNXT_FLAG_PORT_STATS))
+		return;
+
+	rx = bp->port_stats.sw_stats;
+	tx = bp->port_stats.sw_stats + BNXT_TX_PORT_STATS_BYTE_OFFSET / 8;
+
+	rmon_stats->jabbers =
+		BNXT_GET_RX_PORT_STATS64(rx, rx_jbr_frames);
+	rmon_stats->oversize_pkts =
+		BNXT_GET_RX_PORT_STATS64(rx, rx_ovrsz_frames);
+	rmon_stats->undersize_pkts =
+		BNXT_GET_RX_PORT_STATS64(rx, rx_undrsz_frames);
+
+	rmon_stats->hist[0] = BNXT_GET_RX_PORT_STATS64(rx, rx_64b_frames);
+	rmon_stats->hist[1] = BNXT_GET_RX_PORT_STATS64(rx, rx_65b_127b_frames);
+	rmon_stats->hist[2] = BNXT_GET_RX_PORT_STATS64(rx, rx_128b_255b_frames);
+	rmon_stats->hist[3] = BNXT_GET_RX_PORT_STATS64(rx, rx_256b_511b_frames);
+	rmon_stats->hist[4] =
+		BNXT_GET_RX_PORT_STATS64(rx, rx_512b_1023b_frames);
+	rmon_stats->hist[5] =
+		BNXT_GET_RX_PORT_STATS64(rx, rx_1024b_1518b_frames);
+	rmon_stats->hist[6] =
+		BNXT_GET_RX_PORT_STATS64(rx, rx_1519b_2047b_frames);
+	rmon_stats->hist[7] =
+		BNXT_GET_RX_PORT_STATS64(rx, rx_2048b_4095b_frames);
+	rmon_stats->hist[8] =
+		BNXT_GET_RX_PORT_STATS64(rx, rx_4096b_9216b_frames);
+	rmon_stats->hist[9] =
+		BNXT_GET_RX_PORT_STATS64(rx, rx_9217b_16383b_frames);
+
+	rmon_stats->hist_tx[0] =
+		BNXT_GET_TX_PORT_STATS64(tx, tx_64b_frames);
+	rmon_stats->hist_tx[1] =
+		BNXT_GET_TX_PORT_STATS64(tx, tx_65b_127b_frames);
+	rmon_stats->hist_tx[2] =
+		BNXT_GET_TX_PORT_STATS64(tx, tx_128b_255b_frames);
+	rmon_stats->hist_tx[3] =
+		BNXT_GET_TX_PORT_STATS64(tx, tx_256b_511b_frames);
+	rmon_stats->hist_tx[4] =
+		BNXT_GET_TX_PORT_STATS64(tx, tx_512b_1023b_frames);
+	rmon_stats->hist_tx[5] =
+		BNXT_GET_TX_PORT_STATS64(tx, tx_1024b_1518b_frames);
+	rmon_stats->hist_tx[6] =
+		BNXT_GET_TX_PORT_STATS64(tx, tx_1519b_2047b_frames);
+	rmon_stats->hist_tx[7] =
+		BNXT_GET_TX_PORT_STATS64(tx, tx_2048b_4095b_frames);
+	rmon_stats->hist_tx[8] =
+		BNXT_GET_TX_PORT_STATS64(tx, tx_4096b_9216b_frames);
+	rmon_stats->hist_tx[9] =
+		BNXT_GET_TX_PORT_STATS64(tx, tx_9217b_16383b_frames);
+
+	*ranges = bnxt_rmon_ranges;
+}
+
+static void bnxt_get_ptp_stats(struct net_device *dev,
+			       struct ethtool_ts_stats *ts_stats)
+{
+	struct bnxt *bp = netdev_priv(dev);
+	struct bnxt_ptp_cfg *ptp = bp->ptp_cfg;
+
+	if (ptp) {
+		ts_stats->pkts = ptp->stats.ts_pkts;
+		ts_stats->lost = ptp->stats.ts_lost;
+		ts_stats->err = atomic64_read(&ptp->stats.ts_err);
+	}
+}
+
+static void bnxt_get_link_ext_stats(struct net_device *dev,
+				    struct ethtool_link_ext_stats *stats)
+{
+	struct bnxt *bp = netdev_priv(dev);
+	u64 *rx;
+
+	if (BNXT_VF(bp) || !(bp->flags & BNXT_FLAG_PORT_STATS_EXT))
+		return;
+
+	rx = bp->rx_port_stats_ext.sw_stats;
+	stats->link_down_events =
+		*(rx + BNXT_RX_STATS_EXT_OFFSET(link_down_events));
 }
 
 void bnxt_ethtool_free(struct bnxt *bp)
@@ -2548,11 +5565,31 @@ void bnxt_ethtool_free(struct bnxt *bp)
 }
 
 const struct ethtool_ops bnxt_ethtool_ops = {
+	.cap_link_lanes_supported	= 1,
+	.rxfh_per_ctx_key		= 1,
+	.rxfh_max_num_contexts		= BNXT_MAX_ETH_RSS_CTX + 1,
+	.rxfh_indir_space		= BNXT_MAX_RSS_TABLE_ENTRIES_P5,
+	.rxfh_priv_size			= sizeof(struct bnxt_rss_ctx),
+	.supported_coalesce_params = ETHTOOL_COALESCE_USECS |
+				     ETHTOOL_COALESCE_MAX_FRAMES |
+				     ETHTOOL_COALESCE_USECS_IRQ |
+				     ETHTOOL_COALESCE_MAX_FRAMES_IRQ |
+				     ETHTOOL_COALESCE_STATS_BLOCK_USECS |
+				     ETHTOOL_COALESCE_USE_ADAPTIVE_RX |
+				     ETHTOOL_COALESCE_USE_CQE,
+	.supported_ring_params	= ETHTOOL_RING_USE_TCP_DATA_SPLIT |
+				  ETHTOOL_RING_USE_HDS_THRS,
 	.get_link_ksettings	= bnxt_get_link_ksettings,
 	.set_link_ksettings	= bnxt_set_link_ksettings,
+	.get_fec_stats		= bnxt_get_fec_stats,
+	.get_fecparam		= bnxt_get_fecparam,
+	.set_fecparam		= bnxt_set_fecparam,
+	.get_pause_stats	= bnxt_get_pause_stats,
 	.get_pauseparam		= bnxt_get_pauseparam,
 	.set_pauseparam		= bnxt_set_pauseparam,
 	.get_drvinfo		= bnxt_get_drvinfo,
+	.get_regs_len		= bnxt_get_regs_len,
+	.get_regs		= bnxt_get_regs,
 	.get_wol		= bnxt_get_wol,
 	.set_wol		= bnxt_set_wol,
 	.get_coalesce		= bnxt_get_coalesce,
@@ -2571,16 +5608,37 @@ const struct ethtool_ops bnxt_ethtool_ops = {
 	.get_rxfh_indir_size    = bnxt_get_rxfh_indir_size,
 	.get_rxfh_key_size      = bnxt_get_rxfh_key_size,
 	.get_rxfh               = bnxt_get_rxfh,
+	.set_rxfh		= bnxt_set_rxfh,
+	.get_rxfh_fields        = bnxt_get_rxfh_fields,
+	.set_rxfh_fields        = bnxt_set_rxfh_fields,
+	.create_rxfh_context	= bnxt_create_rxfh_context,
+	.modify_rxfh_context	= bnxt_modify_rxfh_context,
+	.remove_rxfh_context	= bnxt_remove_rxfh_context,
 	.flash_device		= bnxt_flash_device,
 	.get_eeprom_len         = bnxt_get_eeprom_len,
 	.get_eeprom             = bnxt_get_eeprom,
 	.set_eeprom		= bnxt_set_eeprom,
 	.get_link		= bnxt_get_link,
+	.get_link_ext_stats	= bnxt_get_link_ext_stats,
 	.get_eee		= bnxt_get_eee,
 	.set_eee		= bnxt_set_eee,
+	.get_tunable		= bnxt_get_tunable,
+	.set_tunable		= bnxt_set_tunable,
 	.get_module_info	= bnxt_get_module_info,
 	.get_module_eeprom	= bnxt_get_module_eeprom,
+	.get_module_eeprom_by_page = bnxt_get_module_eeprom_by_page,
+	.set_module_eeprom_by_page = bnxt_set_module_eeprom_by_page,
 	.nway_reset		= bnxt_nway_reset,
 	.set_phys_id		= bnxt_set_phys_id,
 	.self_test		= bnxt_self_test,
+	.get_ts_info		= bnxt_get_ts_info,
+	.reset			= bnxt_reset,
+	.set_dump		= bnxt_set_dump,
+	.get_dump_flag		= bnxt_get_dump_flag,
+	.get_dump_data		= bnxt_get_dump_data,
+	.get_eth_phy_stats	= bnxt_get_eth_phy_stats,
+	.get_eth_mac_stats	= bnxt_get_eth_mac_stats,
+	.get_eth_ctrl_stats	= bnxt_get_eth_ctrl_stats,
+	.get_rmon_stats		= bnxt_get_rmon_stats,
+	.get_ts_stats		= bnxt_get_ptp_stats,
 };
diff --git a/drivers/net/ethernet/broadcom/bnxt/bnxt_ethtool.h b/drivers/net/ethernet/broadcom/bnxt/bnxt_ethtool.h
index f1bc90b6fb5b..33b86ede1ce5 100644
--- a/drivers/net/ethernet/broadcom/bnxt/bnxt_ethtool.h
+++ b/drivers/net/ethernet/broadcom/bnxt/bnxt_ethtool.h
@@ -34,12 +34,44 @@ struct bnxt_led_cfg {
 #define BNXT_LED_DFLT_ENABLES(x)			\
 	cpu_to_le32(BNXT_LED_DFLT_ENA << (BNXT_LED_DFLT_ENA_SHIFT * (x)))
 
+#define BNXT_FW_RESET_AP	(ETH_RESET_AP << ETH_RESET_SHARED_SHIFT)
+#define BNXT_FW_RESET_CHIP	((ETH_RESET_MGMT | ETH_RESET_IRQ |	\
+				  ETH_RESET_DMA | ETH_RESET_FILTER |	\
+				  ETH_RESET_OFFLOAD | ETH_RESET_MAC |	\
+				  ETH_RESET_PHY | ETH_RESET_RAM)	\
+				 << ETH_RESET_SHARED_SHIFT)
+
+#define BNXT_PXP_REG_LEN	0x3110
+
+#define BNXT_IP_PROTO_FULL_MASK	0xFF
+#define BNXT_IP_PROTO_WILDCARD	0x0
+
 extern const struct ethtool_ops bnxt_ethtool_ops;
 
-u32 _bnxt_fw_to_ethtool_adv_spds(u16, u8);
+u32 bnxt_get_rxfh_indir_size(struct net_device *dev);
+void _bnxt_fw_to_linkmode(unsigned long *mode, u16 fw_speeds);
 u32 bnxt_fw_to_ethtool_speed(u16);
-u16 bnxt_get_fw_auto_link_speeds(u32);
+u16 bnxt_get_fw_auto_link_speeds(const unsigned long *mode);
+int bnxt_hwrm_nvm_get_dev_info(struct bnxt *bp,
+			       struct hwrm_nvm_get_dev_info_output *nvm_dev_info);
+int bnxt_hwrm_firmware_reset(struct net_device *dev, u8 proc_type,
+			     u8 self_reset, u8 flags);
+int bnxt_flash_package_from_fw_obj(struct net_device *dev, const struct firmware *fw,
+				   u32 install_type, struct netlink_ext_ack *extack);
+int bnxt_get_pkginfo(struct net_device *dev, char *ver, int size);
 void bnxt_ethtool_init(struct bnxt *bp);
 void bnxt_ethtool_free(struct bnxt *bp);
+int bnxt_find_nvram_item(struct net_device *dev, u16 type, u16 ordinal,
+			 u16 ext, u16 *index, u32 *item_length,
+			 u32 *data_length);
+int bnxt_find_nvram_item(struct net_device *dev, u16 type, u16 ordinal,
+			 u16 ext, u16 *index, u32 *item_length,
+			 u32 *data_length);
+int bnxt_flash_nvram(struct net_device *dev, u16 dir_type,
+		     u16 dir_ordinal, u16 dir_ext, u16 dir_attr,
+		     u32 dir_item_len, const u8 *data,
+		     size_t data_len);
+int bnxt_get_nvram_item(struct net_device *dev, u32 index, u32 offset,
+			u32 length, u8 *data);
 
 #endif
diff --git a/drivers/net/ethernet/broadcom/bnxt/bnxt_hsi.h b/drivers/net/ethernet/broadcom/bnxt/bnxt_hsi.h
deleted file mode 100644
index 7dc71bb95837..000000000000
--- a/drivers/net/ethernet/broadcom/bnxt/bnxt_hsi.h
+++ /dev/null
@@ -1,5745 +0,0 @@
-/* Broadcom NetXtreme-C/E network driver.
- *
- * Copyright (c) 2014-2016 Broadcom Corporation
- * Copyright (c) 2016-2017 Broadcom Limited
- *
- * This program is free software; you can redistribute it and/or modify
- * it under the terms of the GNU General Public License as published by
- * the Free Software Foundation.
- */
-
-#ifndef BNXT_HSI_H
-#define BNXT_HSI_H
-
-/* HSI and HWRM Specification 1.7.6 */
-#define HWRM_VERSION_MAJOR	1
-#define HWRM_VERSION_MINOR	7
-#define HWRM_VERSION_UPDATE	6
-
-#define HWRM_VERSION_RSVD	2 /* non-zero means beta version */
-
-#define HWRM_VERSION_STR	"1.7.6.2"
-/*
- * Following is the signature for HWRM message field that indicates not
- * applicable (All F's). Need to cast it the size of the field if needed.
- */
-#define HWRM_NA_SIGNATURE	((__le32)(-1))
-#define HWRM_MAX_REQ_LEN    (128)  /* hwrm_func_buf_rgtr */
-#define HWRM_MAX_RESP_LEN    (248)  /* hwrm_selftest_qlist */
-#define HW_HASH_INDEX_SIZE      0x80    /* 7 bit indirection table index. */
-#define HW_HASH_KEY_SIZE	40
-#define HWRM_RESP_VALID_KEY      1 /* valid key for HWRM response */
-
-/* Statistics Ejection Buffer Completion Record (16 bytes) */
-struct eject_cmpl {
-	__le16 type;
-	#define EJECT_CMPL_TYPE_MASK				    0x3fUL
-	#define EJECT_CMPL_TYPE_SFT				    0
-	#define EJECT_CMPL_TYPE_STAT_EJECT			   0x1aUL
-	__le16 len;
-	__le32 opaque;
-	__le32 v;
-	#define EJECT_CMPL_V					    0x1UL
-	__le32 unused_2;
-};
-
-/* HWRM Completion Record (16 bytes) */
-struct hwrm_cmpl {
-	__le16 type;
-	#define CMPL_TYPE_MASK					    0x3fUL
-	#define CMPL_TYPE_SFT					    0
-	#define CMPL_TYPE_HWRM_DONE				   0x20UL
-	__le16 sequence_id;
-	__le32 unused_1;
-	__le32 v;
-	#define CMPL_V						    0x1UL
-	__le32 unused_3;
-};
-
-/* HWRM Forwarded Request (16 bytes) */
-struct hwrm_fwd_req_cmpl {
-	__le16 req_len_type;
-	#define FWD_REQ_CMPL_TYPE_MASK				    0x3fUL
-	#define FWD_REQ_CMPL_TYPE_SFT				    0
-	#define FWD_REQ_CMPL_TYPE_HWRM_FWD_REQ			   0x22UL
-	#define FWD_REQ_CMPL_REQ_LEN_MASK			    0xffc0UL
-	#define FWD_REQ_CMPL_REQ_LEN_SFT			    6
-	__le16 source_id;
-	__le32 unused_0;
-	__le32 req_buf_addr_v[2];
-	#define FWD_REQ_CMPL_V					    0x1UL
-	#define FWD_REQ_CMPL_REQ_BUF_ADDR_MASK			    0xfffffffeUL
-	#define FWD_REQ_CMPL_REQ_BUF_ADDR_SFT			    1
-};
-
-/* HWRM Forwarded Response (16 bytes) */
-struct hwrm_fwd_resp_cmpl {
-	__le16 type;
-	#define FWD_RESP_CMPL_TYPE_MASK			    0x3fUL
-	#define FWD_RESP_CMPL_TYPE_SFT				    0
-	#define FWD_RESP_CMPL_TYPE_HWRM_FWD_RESP		   0x24UL
-	__le16 source_id;
-	__le16 resp_len;
-	__le16 unused_1;
-	__le32 resp_buf_addr_v[2];
-	#define FWD_RESP_CMPL_V				    0x1UL
-	#define FWD_RESP_CMPL_RESP_BUF_ADDR_MASK		    0xfffffffeUL
-	#define FWD_RESP_CMPL_RESP_BUF_ADDR_SFT		    1
-};
-
-/* HWRM Asynchronous Event Completion Record (16 bytes) */
-struct hwrm_async_event_cmpl {
-	__le16 type;
-	#define ASYNC_EVENT_CMPL_TYPE_MASK			    0x3fUL
-	#define ASYNC_EVENT_CMPL_TYPE_SFT			    0
-	#define ASYNC_EVENT_CMPL_TYPE_HWRM_ASYNC_EVENT		   0x2eUL
-	__le16 event_id;
-	#define ASYNC_EVENT_CMPL_EVENT_ID_LINK_STATUS_CHANGE      0x0UL
-	#define ASYNC_EVENT_CMPL_EVENT_ID_LINK_MTU_CHANGE	   0x1UL
-	#define ASYNC_EVENT_CMPL_EVENT_ID_LINK_SPEED_CHANGE       0x2UL
-	#define ASYNC_EVENT_CMPL_EVENT_ID_DCB_CONFIG_CHANGE       0x3UL
-	#define ASYNC_EVENT_CMPL_EVENT_ID_PORT_CONN_NOT_ALLOWED   0x4UL
-	#define ASYNC_EVENT_CMPL_EVENT_ID_LINK_SPEED_CFG_NOT_ALLOWED 0x5UL
-	#define ASYNC_EVENT_CMPL_EVENT_ID_LINK_SPEED_CFG_CHANGE   0x6UL
-	#define ASYNC_EVENT_CMPL_EVENT_ID_PORT_PHY_CFG_CHANGE     0x7UL
-	#define ASYNC_EVENT_CMPL_EVENT_ID_FUNC_DRVR_UNLOAD	   0x10UL
-	#define ASYNC_EVENT_CMPL_EVENT_ID_FUNC_DRVR_LOAD	   0x11UL
-	#define ASYNC_EVENT_CMPL_EVENT_ID_FUNC_FLR_PROC_CMPLT     0x12UL
-	#define ASYNC_EVENT_CMPL_EVENT_ID_PF_DRVR_UNLOAD	   0x20UL
-	#define ASYNC_EVENT_CMPL_EVENT_ID_PF_DRVR_LOAD		   0x21UL
-	#define ASYNC_EVENT_CMPL_EVENT_ID_VF_FLR		   0x30UL
-	#define ASYNC_EVENT_CMPL_EVENT_ID_VF_MAC_ADDR_CHANGE      0x31UL
-	#define ASYNC_EVENT_CMPL_EVENT_ID_PF_VF_COMM_STATUS_CHANGE 0x32UL
-	#define ASYNC_EVENT_CMPL_EVENT_ID_VF_CFG_CHANGE	   0x33UL
-	#define ASYNC_EVENT_CMPL_EVENT_ID_HWRM_ERROR		   0xffUL
-	__le32 event_data2;
-	u8 opaque_v;
-	#define ASYNC_EVENT_CMPL_V				    0x1UL
-	#define ASYNC_EVENT_CMPL_OPAQUE_MASK			    0xfeUL
-	#define ASYNC_EVENT_CMPL_OPAQUE_SFT			    1
-	u8 timestamp_lo;
-	__le16 timestamp_hi;
-	__le32 event_data1;
-};
-
-/* HWRM Asynchronous Event Completion Record for link status change (16 bytes) */
-struct hwrm_async_event_cmpl_link_status_change {
-	__le16 type;
-	#define ASYNC_EVENT_CMPL_LINK_STATUS_CHANGE_TYPE_MASK      0x3fUL
-	#define ASYNC_EVENT_CMPL_LINK_STATUS_CHANGE_TYPE_SFT       0
-	#define ASYNC_EVENT_CMPL_LINK_STATUS_CHANGE_TYPE_HWRM_ASYNC_EVENT 0x2eUL
-	__le16 event_id;
-	#define ASYNC_EVENT_CMPL_LINK_STATUS_CHANGE_EVENT_ID_LINK_STATUS_CHANGE 0x0UL
-	__le32 event_data2;
-	u8 opaque_v;
-	#define ASYNC_EVENT_CMPL_LINK_STATUS_CHANGE_V		    0x1UL
-	#define ASYNC_EVENT_CMPL_LINK_STATUS_CHANGE_OPAQUE_MASK    0xfeUL
-	#define ASYNC_EVENT_CMPL_LINK_STATUS_CHANGE_OPAQUE_SFT     1
-	u8 timestamp_lo;
-	__le16 timestamp_hi;
-	__le32 event_data1;
-	#define ASYNC_EVENT_CMPL_LINK_STATUS_CHANGE_EVENT_DATA1_LINK_CHANGE 0x1UL
-	#define ASYNC_EVENT_CMPL_LINK_STATUS_CHANGE_EVENT_DATA1_LINK_CHANGE_DOWN (0x0UL << 0)
-	#define ASYNC_EVENT_CMPL_LINK_STATUS_CHANGE_EVENT_DATA1_LINK_CHANGE_UP (0x1UL << 0)
-	#define ASYNC_EVENT_CMPL_LINK_STATUS_CHANGE_EVENT_DATA1_LINK_CHANGE_LAST    ASYNC_EVENT_CMPL_LINK_STATUS_CHANGE_EVENT_DATA1_LINK_CHANGE_UP
-	#define ASYNC_EVENT_CMPL_LINK_STATUS_CHANGE_EVENT_DATA1_PORT_MASK 0xeUL
-	#define ASYNC_EVENT_CMPL_LINK_STATUS_CHANGE_EVENT_DATA1_PORT_SFT 1
-	#define ASYNC_EVENT_CMPL_LINK_STATUS_CHANGE_EVENT_DATA1_PORT_ID_MASK 0xffff0UL
-	#define ASYNC_EVENT_CMPL_LINK_STATUS_CHANGE_EVENT_DATA1_PORT_ID_SFT 4
-};
-
-/* HWRM Asynchronous Event Completion Record for link MTU change (16 bytes) */
-struct hwrm_async_event_cmpl_link_mtu_change {
-	__le16 type;
-	#define ASYNC_EVENT_CMPL_LINK_MTU_CHANGE_TYPE_MASK	    0x3fUL
-	#define ASYNC_EVENT_CMPL_LINK_MTU_CHANGE_TYPE_SFT	    0
-	#define ASYNC_EVENT_CMPL_LINK_MTU_CHANGE_TYPE_HWRM_ASYNC_EVENT 0x2eUL
-	__le16 event_id;
-	#define ASYNC_EVENT_CMPL_LINK_MTU_CHANGE_EVENT_ID_LINK_MTU_CHANGE 0x1UL
-	__le32 event_data2;
-	u8 opaque_v;
-	#define ASYNC_EVENT_CMPL_LINK_MTU_CHANGE_V		    0x1UL
-	#define ASYNC_EVENT_CMPL_LINK_MTU_CHANGE_OPAQUE_MASK       0xfeUL
-	#define ASYNC_EVENT_CMPL_LINK_MTU_CHANGE_OPAQUE_SFT	    1
-	u8 timestamp_lo;
-	__le16 timestamp_hi;
-	__le32 event_data1;
-	#define ASYNC_EVENT_CMPL_LINK_MTU_CHANGE_EVENT_DATA1_NEW_MTU_MASK 0xffffUL
-	#define ASYNC_EVENT_CMPL_LINK_MTU_CHANGE_EVENT_DATA1_NEW_MTU_SFT 0
-};
-
-/* HWRM Asynchronous Event Completion Record for link speed change (16 bytes) */
-struct hwrm_async_event_cmpl_link_speed_change {
-	__le16 type;
-	#define ASYNC_EVENT_CMPL_LINK_SPEED_CHANGE_TYPE_MASK       0x3fUL
-	#define ASYNC_EVENT_CMPL_LINK_SPEED_CHANGE_TYPE_SFT	    0
-	#define ASYNC_EVENT_CMPL_LINK_SPEED_CHANGE_TYPE_HWRM_ASYNC_EVENT 0x2eUL
-	__le16 event_id;
-	#define ASYNC_EVENT_CMPL_LINK_SPEED_CHANGE_EVENT_ID_LINK_SPEED_CHANGE 0x2UL
-	__le32 event_data2;
-	u8 opaque_v;
-	#define ASYNC_EVENT_CMPL_LINK_SPEED_CHANGE_V		    0x1UL
-	#define ASYNC_EVENT_CMPL_LINK_SPEED_CHANGE_OPAQUE_MASK     0xfeUL
-	#define ASYNC_EVENT_CMPL_LINK_SPEED_CHANGE_OPAQUE_SFT      1
-	u8 timestamp_lo;
-	__le16 timestamp_hi;
-	__le32 event_data1;
-	#define ASYNC_EVENT_CMPL_LINK_SPEED_CHANGE_EVENT_DATA1_FORCE 0x1UL
-	#define ASYNC_EVENT_CMPL_LINK_SPEED_CHANGE_EVENT_DATA1_NEW_LINK_SPEED_100MBPS_MASK 0xfffeUL
-	#define ASYNC_EVENT_CMPL_LINK_SPEED_CHANGE_EVENT_DATA1_NEW_LINK_SPEED_100MBPS_SFT 1
-	#define ASYNC_EVENT_CMPL_LINK_SPEED_CHANGE_EVENT_DATA1_NEW_LINK_SPEED_100MBPS_100MB (0x1UL << 1)
-	#define ASYNC_EVENT_CMPL_LINK_SPEED_CHANGE_EVENT_DATA1_NEW_LINK_SPEED_100MBPS_1GB (0xaUL << 1)
-	#define ASYNC_EVENT_CMPL_LINK_SPEED_CHANGE_EVENT_DATA1_NEW_LINK_SPEED_100MBPS_2GB (0x14UL << 1)
-	#define ASYNC_EVENT_CMPL_LINK_SPEED_CHANGE_EVENT_DATA1_NEW_LINK_SPEED_100MBPS_2_5GB (0x19UL << 1)
-	#define ASYNC_EVENT_CMPL_LINK_SPEED_CHANGE_EVENT_DATA1_NEW_LINK_SPEED_100MBPS_10GB (0x64UL << 1)
-	#define ASYNC_EVENT_CMPL_LINK_SPEED_CHANGE_EVENT_DATA1_NEW_LINK_SPEED_100MBPS_20GB (0xc8UL << 1)
-	#define ASYNC_EVENT_CMPL_LINK_SPEED_CHANGE_EVENT_DATA1_NEW_LINK_SPEED_100MBPS_25GB (0xfaUL << 1)
-	#define ASYNC_EVENT_CMPL_LINK_SPEED_CHANGE_EVENT_DATA1_NEW_LINK_SPEED_100MBPS_40GB (0x190UL << 1)
-	#define ASYNC_EVENT_CMPL_LINK_SPEED_CHANGE_EVENT_DATA1_NEW_LINK_SPEED_100MBPS_50GB (0x1f4UL << 1)
-	#define ASYNC_EVENT_CMPL_LINK_SPEED_CHANGE_EVENT_DATA1_NEW_LINK_SPEED_100MBPS_100GB (0x3e8UL << 1)
-	#define ASYNC_EVENT_CMPL_LINK_SPEED_CHANGE_EVENT_DATA1_NEW_LINK_SPEED_100MBPS_LAST    ASYNC_EVENT_CMPL_LINK_SPEED_CHANGE_EVENT_DATA1_NEW_LINK_SPEED_100MBPS_100GB
-	#define ASYNC_EVENT_CMPL_LINK_SPEED_CHANGE_EVENT_DATA1_PORT_ID_MASK 0xffff0000UL
-	#define ASYNC_EVENT_CMPL_LINK_SPEED_CHANGE_EVENT_DATA1_PORT_ID_SFT 16
-};
-
-/* HWRM Asynchronous Event Completion Record for DCB Config change (16 bytes) */
-struct hwrm_async_event_cmpl_dcb_config_change {
-	__le16 type;
-	#define ASYNC_EVENT_CMPL_DCB_CONFIG_CHANGE_TYPE_MASK       0x3fUL
-	#define ASYNC_EVENT_CMPL_DCB_CONFIG_CHANGE_TYPE_SFT	    0
-	#define ASYNC_EVENT_CMPL_DCB_CONFIG_CHANGE_TYPE_HWRM_ASYNC_EVENT 0x2eUL
-	__le16 event_id;
-	#define ASYNC_EVENT_CMPL_DCB_CONFIG_CHANGE_EVENT_ID_DCB_CONFIG_CHANGE 0x3UL
-	__le32 event_data2;
-	#define ASYNC_EVENT_CMPL_DCB_CONFIG_CHANGE_EVENT_DATA2_ETS 0x1UL
-	#define ASYNC_EVENT_CMPL_DCB_CONFIG_CHANGE_EVENT_DATA2_PFC 0x2UL
-	#define ASYNC_EVENT_CMPL_DCB_CONFIG_CHANGE_EVENT_DATA2_APP 0x4UL
-	u8 opaque_v;
-	#define ASYNC_EVENT_CMPL_DCB_CONFIG_CHANGE_V		    0x1UL
-	#define ASYNC_EVENT_CMPL_DCB_CONFIG_CHANGE_OPAQUE_MASK     0xfeUL
-	#define ASYNC_EVENT_CMPL_DCB_CONFIG_CHANGE_OPAQUE_SFT      1
-	u8 timestamp_lo;
-	__le16 timestamp_hi;
-	__le32 event_data1;
-	#define ASYNC_EVENT_CMPL_DCB_CONFIG_CHANGE_EVENT_DATA1_PORT_ID_MASK 0xffffUL
-	#define ASYNC_EVENT_CMPL_DCB_CONFIG_CHANGE_EVENT_DATA1_PORT_ID_SFT 0
-	#define ASYNC_EVENT_CMPL_DCB_CONFIG_CHANGE_EVENT_DATA1_RECOMMEND_ROCE_PRIORITY_MASK 0xff0000UL
-	#define ASYNC_EVENT_CMPL_DCB_CONFIG_CHANGE_EVENT_DATA1_RECOMMEND_ROCE_PRIORITY_SFT 16
-	#define ASYNC_EVENT_CMPL_DCB_CONFIG_CHANGE_EVENT_DATA1_RECOMMEND_ROCE_PRIORITY_NONE (0xffUL << 16)
-	#define ASYNC_EVENT_CMPL_DCB_CONFIG_CHANGE_EVENT_DATA1_RECOMMEND_ROCE_PRIORITY_LAST    ASYNC_EVENT_CMPL_DCB_CONFIG_CHANGE_EVENT_DATA1_RECOMMEND_ROCE_PRIORITY_NONE
-	#define ASYNC_EVENT_CMPL_DCB_CONFIG_CHANGE_EVENT_DATA1_RECOMMEND_L2_PRIORITY_MASK 0xff000000UL
-	#define ASYNC_EVENT_CMPL_DCB_CONFIG_CHANGE_EVENT_DATA1_RECOMMEND_L2_PRIORITY_SFT 24
-	#define ASYNC_EVENT_CMPL_DCB_CONFIG_CHANGE_EVENT_DATA1_RECOMMEND_L2_PRIORITY_NONE (0xffUL << 24)
-	#define ASYNC_EVENT_CMPL_DCB_CONFIG_CHANGE_EVENT_DATA1_RECOMMEND_L2_PRIORITY_LAST    ASYNC_EVENT_CMPL_DCB_CONFIG_CHANGE_EVENT_DATA1_RECOMMEND_L2_PRIORITY_NONE
-};
-
-/* HWRM Asynchronous Event Completion Record for port connection not allowed (16 bytes) */
-struct hwrm_async_event_cmpl_port_conn_not_allowed {
-	__le16 type;
-	#define ASYNC_EVENT_CMPL_PORT_CONN_NOT_ALLOWED_TYPE_MASK   0x3fUL
-	#define ASYNC_EVENT_CMPL_PORT_CONN_NOT_ALLOWED_TYPE_SFT    0
-	#define ASYNC_EVENT_CMPL_PORT_CONN_NOT_ALLOWED_TYPE_HWRM_ASYNC_EVENT 0x2eUL
-	__le16 event_id;
-	#define ASYNC_EVENT_CMPL_PORT_CONN_NOT_ALLOWED_EVENT_ID_PORT_CONN_NOT_ALLOWED 0x4UL
-	__le32 event_data2;
-	u8 opaque_v;
-	#define ASYNC_EVENT_CMPL_PORT_CONN_NOT_ALLOWED_V	    0x1UL
-	#define ASYNC_EVENT_CMPL_PORT_CONN_NOT_ALLOWED_OPAQUE_MASK 0xfeUL
-	#define ASYNC_EVENT_CMPL_PORT_CONN_NOT_ALLOWED_OPAQUE_SFT  1
-	u8 timestamp_lo;
-	__le16 timestamp_hi;
-	__le32 event_data1;
-	#define ASYNC_EVENT_CMPL_PORT_CONN_NOT_ALLOWED_EVENT_DATA1_PORT_ID_MASK 0xffffUL
-	#define ASYNC_EVENT_CMPL_PORT_CONN_NOT_ALLOWED_EVENT_DATA1_PORT_ID_SFT 0
-	#define ASYNC_EVENT_CMPL_PORT_CONN_NOT_ALLOWED_EVENT_DATA1_ENFORCEMENT_POLICY_MASK 0xff0000UL
-	#define ASYNC_EVENT_CMPL_PORT_CONN_NOT_ALLOWED_EVENT_DATA1_ENFORCEMENT_POLICY_SFT 16
-	#define ASYNC_EVENT_CMPL_PORT_CONN_NOT_ALLOWED_EVENT_DATA1_ENFORCEMENT_POLICY_NONE (0x0UL << 16)
-	#define ASYNC_EVENT_CMPL_PORT_CONN_NOT_ALLOWED_EVENT_DATA1_ENFORCEMENT_POLICY_DISABLETX (0x1UL << 16)
-	#define ASYNC_EVENT_CMPL_PORT_CONN_NOT_ALLOWED_EVENT_DATA1_ENFORCEMENT_POLICY_WARNINGMSG (0x2UL << 16)
-	#define ASYNC_EVENT_CMPL_PORT_CONN_NOT_ALLOWED_EVENT_DATA1_ENFORCEMENT_POLICY_PWRDOWN (0x3UL << 16)
-	#define ASYNC_EVENT_CMPL_PORT_CONN_NOT_ALLOWED_EVENT_DATA1_ENFORCEMENT_POLICY_LAST    ASYNC_EVENT_CMPL_PORT_CONN_NOT_ALLOWED_EVENT_DATA1_ENFORCEMENT_POLICY_PWRDOWN
-};
-
-/* HWRM Asynchronous Event Completion Record for link speed config not allowed (16 bytes) */
-struct hwrm_async_event_cmpl_link_speed_cfg_not_allowed {
-	__le16 type;
-	#define ASYNC_EVENT_CMPL_LINK_SPEED_CFG_NOT_ALLOWED_TYPE_MASK 0x3fUL
-	#define ASYNC_EVENT_CMPL_LINK_SPEED_CFG_NOT_ALLOWED_TYPE_SFT 0
-	#define ASYNC_EVENT_CMPL_LINK_SPEED_CFG_NOT_ALLOWED_TYPE_HWRM_ASYNC_EVENT 0x2eUL
-	__le16 event_id;
-	#define ASYNC_EVENT_CMPL_LINK_SPEED_CFG_NOT_ALLOWED_EVENT_ID_LINK_SPEED_CFG_NOT_ALLOWED 0x5UL
-	__le32 event_data2;
-	u8 opaque_v;
-	#define ASYNC_EVENT_CMPL_LINK_SPEED_CFG_NOT_ALLOWED_V      0x1UL
-	#define ASYNC_EVENT_CMPL_LINK_SPEED_CFG_NOT_ALLOWED_OPAQUE_MASK 0xfeUL
-	#define ASYNC_EVENT_CMPL_LINK_SPEED_CFG_NOT_ALLOWED_OPAQUE_SFT 1
-	u8 timestamp_lo;
-	__le16 timestamp_hi;
-	__le32 event_data1;
-	#define ASYNC_EVENT_CMPL_LINK_SPEED_CFG_NOT_ALLOWED_EVENT_DATA1_PORT_ID_MASK 0xffffUL
-	#define ASYNC_EVENT_CMPL_LINK_SPEED_CFG_NOT_ALLOWED_EVENT_DATA1_PORT_ID_SFT 0
-};
-
-/* HWRM Asynchronous Event Completion Record for link speed configuration change (16 bytes) */
-struct hwrm_async_event_cmpl_link_speed_cfg_change {
-	__le16 type;
-	#define ASYNC_EVENT_CMPL_LINK_SPEED_CFG_CHANGE_TYPE_MASK   0x3fUL
-	#define ASYNC_EVENT_CMPL_LINK_SPEED_CFG_CHANGE_TYPE_SFT    0
-	#define ASYNC_EVENT_CMPL_LINK_SPEED_CFG_CHANGE_TYPE_HWRM_ASYNC_EVENT 0x2eUL
-	__le16 event_id;
-	#define ASYNC_EVENT_CMPL_LINK_SPEED_CFG_CHANGE_EVENT_ID_LINK_SPEED_CFG_CHANGE 0x6UL
-	__le32 event_data2;
-	u8 opaque_v;
-	#define ASYNC_EVENT_CMPL_LINK_SPEED_CFG_CHANGE_V	    0x1UL
-	#define ASYNC_EVENT_CMPL_LINK_SPEED_CFG_CHANGE_OPAQUE_MASK 0xfeUL
-	#define ASYNC_EVENT_CMPL_LINK_SPEED_CFG_CHANGE_OPAQUE_SFT  1
-	u8 timestamp_lo;
-	__le16 timestamp_hi;
-	__le32 event_data1;
-	#define ASYNC_EVENT_CMPL_LINK_SPEED_CFG_CHANGE_EVENT_DATA1_PORT_ID_MASK 0xffffUL
-	#define ASYNC_EVENT_CMPL_LINK_SPEED_CFG_CHANGE_EVENT_DATA1_PORT_ID_SFT 0
-	#define ASYNC_EVENT_CMPL_LINK_SPEED_CFG_CHANGE_EVENT_DATA1_SUPPORTED_LINK_SPEEDS_CHANGE 0x10000UL
-	#define ASYNC_EVENT_CMPL_LINK_SPEED_CFG_CHANGE_EVENT_DATA1_ILLEGAL_LINK_SPEED_CFG 0x20000UL
-};
-
-/* HWRM Asynchronous Event Completion Record for Function Driver Unload (16 bytes) */
-struct hwrm_async_event_cmpl_func_drvr_unload {
-	__le16 type;
-	#define ASYNC_EVENT_CMPL_FUNC_DRVR_UNLOAD_TYPE_MASK	    0x3fUL
-	#define ASYNC_EVENT_CMPL_FUNC_DRVR_UNLOAD_TYPE_SFT	    0
-	#define ASYNC_EVENT_CMPL_FUNC_DRVR_UNLOAD_TYPE_HWRM_ASYNC_EVENT 0x2eUL
-	__le16 event_id;
-	#define ASYNC_EVENT_CMPL_FUNC_DRVR_UNLOAD_EVENT_ID_FUNC_DRVR_UNLOAD 0x10UL
-	__le32 event_data2;
-	u8 opaque_v;
-	#define ASYNC_EVENT_CMPL_FUNC_DRVR_UNLOAD_V		    0x1UL
-	#define ASYNC_EVENT_CMPL_FUNC_DRVR_UNLOAD_OPAQUE_MASK      0xfeUL
-	#define ASYNC_EVENT_CMPL_FUNC_DRVR_UNLOAD_OPAQUE_SFT       1
-	u8 timestamp_lo;
-	__le16 timestamp_hi;
-	__le32 event_data1;
-	#define ASYNC_EVENT_CMPL_FUNC_DRVR_UNLOAD_EVENT_DATA1_FUNC_ID_MASK 0xffffUL
-	#define ASYNC_EVENT_CMPL_FUNC_DRVR_UNLOAD_EVENT_DATA1_FUNC_ID_SFT 0
-};
-
-/* HWRM Asynchronous Event Completion Record for Function Driver load (16 bytes) */
-struct hwrm_async_event_cmpl_func_drvr_load {
-	__le16 type;
-	#define ASYNC_EVENT_CMPL_FUNC_DRVR_LOAD_TYPE_MASK	    0x3fUL
-	#define ASYNC_EVENT_CMPL_FUNC_DRVR_LOAD_TYPE_SFT	    0
-	#define ASYNC_EVENT_CMPL_FUNC_DRVR_LOAD_TYPE_HWRM_ASYNC_EVENT 0x2eUL
-	__le16 event_id;
-	#define ASYNC_EVENT_CMPL_FUNC_DRVR_LOAD_EVENT_ID_FUNC_DRVR_LOAD 0x11UL
-	__le32 event_data2;
-	u8 opaque_v;
-	#define ASYNC_EVENT_CMPL_FUNC_DRVR_LOAD_V		    0x1UL
-	#define ASYNC_EVENT_CMPL_FUNC_DRVR_LOAD_OPAQUE_MASK	    0xfeUL
-	#define ASYNC_EVENT_CMPL_FUNC_DRVR_LOAD_OPAQUE_SFT	    1
-	u8 timestamp_lo;
-	__le16 timestamp_hi;
-	__le32 event_data1;
-	#define ASYNC_EVENT_CMPL_FUNC_DRVR_LOAD_EVENT_DATA1_FUNC_ID_MASK 0xffffUL
-	#define ASYNC_EVENT_CMPL_FUNC_DRVR_LOAD_EVENT_DATA1_FUNC_ID_SFT 0
-};
-
-/* HWRM Asynchronous Event Completion Record to indicate completion of FLR related processing (16 bytes) */
-struct hwrm_async_event_cmpl_func_flr_proc_cmplt {
-	__le16 type;
-	#define ASYNC_EVENT_CMPL_FUNC_FLR_PROC_CMPLT_TYPE_MASK     0x3fUL
-	#define ASYNC_EVENT_CMPL_FUNC_FLR_PROC_CMPLT_TYPE_SFT      0
-	#define ASYNC_EVENT_CMPL_FUNC_FLR_PROC_CMPLT_TYPE_HWRM_ASYNC_EVENT 0x2eUL
-	__le16 event_id;
-	#define ASYNC_EVENT_CMPL_FUNC_FLR_PROC_CMPLT_EVENT_ID_FUNC_FLR_PROC_CMPLT 0x12UL
-	__le32 event_data2;
-	u8 opaque_v;
-	#define ASYNC_EVENT_CMPL_FUNC_FLR_PROC_CMPLT_V		    0x1UL
-	#define ASYNC_EVENT_CMPL_FUNC_FLR_PROC_CMPLT_OPAQUE_MASK   0xfeUL
-	#define ASYNC_EVENT_CMPL_FUNC_FLR_PROC_CMPLT_OPAQUE_SFT    1
-	u8 timestamp_lo;
-	__le16 timestamp_hi;
-	__le32 event_data1;
-	#define ASYNC_EVENT_CMPL_FUNC_FLR_PROC_CMPLT_EVENT_DATA1_FUNC_ID_MASK 0xffffUL
-	#define ASYNC_EVENT_CMPL_FUNC_FLR_PROC_CMPLT_EVENT_DATA1_FUNC_ID_SFT 0
-};
-
-/* HWRM Asynchronous Event Completion Record for PF Driver Unload (16 bytes) */
-struct hwrm_async_event_cmpl_pf_drvr_unload {
-	__le16 type;
-	#define ASYNC_EVENT_CMPL_PF_DRVR_UNLOAD_TYPE_MASK	    0x3fUL
-	#define ASYNC_EVENT_CMPL_PF_DRVR_UNLOAD_TYPE_SFT	    0
-	#define ASYNC_EVENT_CMPL_PF_DRVR_UNLOAD_TYPE_HWRM_ASYNC_EVENT 0x2eUL
-	__le16 event_id;
-	#define ASYNC_EVENT_CMPL_PF_DRVR_UNLOAD_EVENT_ID_PF_DRVR_UNLOAD 0x20UL
-	__le32 event_data2;
-	u8 opaque_v;
-	#define ASYNC_EVENT_CMPL_PF_DRVR_UNLOAD_V		    0x1UL
-	#define ASYNC_EVENT_CMPL_PF_DRVR_UNLOAD_OPAQUE_MASK	    0xfeUL
-	#define ASYNC_EVENT_CMPL_PF_DRVR_UNLOAD_OPAQUE_SFT	    1
-	u8 timestamp_lo;
-	__le16 timestamp_hi;
-	__le32 event_data1;
-	#define ASYNC_EVENT_CMPL_PF_DRVR_UNLOAD_EVENT_DATA1_FUNC_ID_MASK 0xffffUL
-	#define ASYNC_EVENT_CMPL_PF_DRVR_UNLOAD_EVENT_DATA1_FUNC_ID_SFT 0
-	#define ASYNC_EVENT_CMPL_PF_DRVR_UNLOAD_EVENT_DATA1_PORT_MASK 0x70000UL
-	#define ASYNC_EVENT_CMPL_PF_DRVR_UNLOAD_EVENT_DATA1_PORT_SFT 16
-};
-
-/* HWRM Asynchronous Event Completion Record for PF Driver load (16 bytes) */
-struct hwrm_async_event_cmpl_pf_drvr_load {
-	__le16 type;
-	#define ASYNC_EVENT_CMPL_PF_DRVR_LOAD_TYPE_MASK	    0x3fUL
-	#define ASYNC_EVENT_CMPL_PF_DRVR_LOAD_TYPE_SFT		    0
-	#define ASYNC_EVENT_CMPL_PF_DRVR_LOAD_TYPE_HWRM_ASYNC_EVENT 0x2eUL
-	__le16 event_id;
-	#define ASYNC_EVENT_CMPL_PF_DRVR_LOAD_EVENT_ID_PF_DRVR_LOAD 0x21UL
-	__le32 event_data2;
-	u8 opaque_v;
-	#define ASYNC_EVENT_CMPL_PF_DRVR_LOAD_V		    0x1UL
-	#define ASYNC_EVENT_CMPL_PF_DRVR_LOAD_OPAQUE_MASK	    0xfeUL
-	#define ASYNC_EVENT_CMPL_PF_DRVR_LOAD_OPAQUE_SFT	    1
-	u8 timestamp_lo;
-	__le16 timestamp_hi;
-	__le32 event_data1;
-	#define ASYNC_EVENT_CMPL_PF_DRVR_LOAD_EVENT_DATA1_FUNC_ID_MASK 0xffffUL
-	#define ASYNC_EVENT_CMPL_PF_DRVR_LOAD_EVENT_DATA1_FUNC_ID_SFT 0
-	#define ASYNC_EVENT_CMPL_PF_DRVR_LOAD_EVENT_DATA1_PORT_MASK 0x70000UL
-	#define ASYNC_EVENT_CMPL_PF_DRVR_LOAD_EVENT_DATA1_PORT_SFT 16
-};
-
-/* HWRM Asynchronous Event Completion Record for VF FLR (16 bytes) */
-struct hwrm_async_event_cmpl_vf_flr {
-	__le16 type;
-	#define ASYNC_EVENT_CMPL_VF_FLR_TYPE_MASK		    0x3fUL
-	#define ASYNC_EVENT_CMPL_VF_FLR_TYPE_SFT		    0
-	#define ASYNC_EVENT_CMPL_VF_FLR_TYPE_HWRM_ASYNC_EVENT     0x2eUL
-	__le16 event_id;
-	#define ASYNC_EVENT_CMPL_VF_FLR_EVENT_ID_VF_FLR	   0x30UL
-	__le32 event_data2;
-	u8 opaque_v;
-	#define ASYNC_EVENT_CMPL_VF_FLR_V			    0x1UL
-	#define ASYNC_EVENT_CMPL_VF_FLR_OPAQUE_MASK		    0xfeUL
-	#define ASYNC_EVENT_CMPL_VF_FLR_OPAQUE_SFT		    1
-	u8 timestamp_lo;
-	__le16 timestamp_hi;
-	__le32 event_data1;
-	#define ASYNC_EVENT_CMPL_VF_FLR_EVENT_DATA1_VF_ID_MASK     0xffffUL
-	#define ASYNC_EVENT_CMPL_VF_FLR_EVENT_DATA1_VF_ID_SFT      0
-};
-
-/* HWRM Asynchronous Event Completion Record for VF MAC Addr change (16 bytes) */
-struct hwrm_async_event_cmpl_vf_mac_addr_change {
-	__le16 type;
-	#define ASYNC_EVENT_CMPL_VF_MAC_ADDR_CHANGE_TYPE_MASK      0x3fUL
-	#define ASYNC_EVENT_CMPL_VF_MAC_ADDR_CHANGE_TYPE_SFT       0
-	#define ASYNC_EVENT_CMPL_VF_MAC_ADDR_CHANGE_TYPE_HWRM_ASYNC_EVENT 0x2eUL
-	__le16 event_id;
-	#define ASYNC_EVENT_CMPL_VF_MAC_ADDR_CHANGE_EVENT_ID_VF_MAC_ADDR_CHANGE 0x31UL
-	__le32 event_data2;
-	u8 opaque_v;
-	#define ASYNC_EVENT_CMPL_VF_MAC_ADDR_CHANGE_V		    0x1UL
-	#define ASYNC_EVENT_CMPL_VF_MAC_ADDR_CHANGE_OPAQUE_MASK    0xfeUL
-	#define ASYNC_EVENT_CMPL_VF_MAC_ADDR_CHANGE_OPAQUE_SFT     1
-	u8 timestamp_lo;
-	__le16 timestamp_hi;
-	__le32 event_data1;
-	#define ASYNC_EVENT_CMPL_VF_MAC_ADDR_CHANGE_EVENT_DATA1_VF_ID_MASK 0xffffUL
-	#define ASYNC_EVENT_CMPL_VF_MAC_ADDR_CHANGE_EVENT_DATA1_VF_ID_SFT 0
-};
-
-/* HWRM Asynchronous Event Completion Record for PF-VF communication status change (16 bytes) */
-struct hwrm_async_event_cmpl_pf_vf_comm_status_change {
-	__le16 type;
-	#define ASYNC_EVENT_CMPL_PF_VF_COMM_STATUS_CHANGE_TYPE_MASK 0x3fUL
-	#define ASYNC_EVENT_CMPL_PF_VF_COMM_STATUS_CHANGE_TYPE_SFT 0
-	#define ASYNC_EVENT_CMPL_PF_VF_COMM_STATUS_CHANGE_TYPE_HWRM_ASYNC_EVENT 0x2eUL
-	__le16 event_id;
-	#define ASYNC_EVENT_CMPL_PF_VF_COMM_STATUS_CHANGE_EVENT_ID_PF_VF_COMM_STATUS_CHANGE 0x32UL
-	__le32 event_data2;
-	u8 opaque_v;
-	#define ASYNC_EVENT_CMPL_PF_VF_COMM_STATUS_CHANGE_V	    0x1UL
-	#define ASYNC_EVENT_CMPL_PF_VF_COMM_STATUS_CHANGE_OPAQUE_MASK 0xfeUL
-	#define ASYNC_EVENT_CMPL_PF_VF_COMM_STATUS_CHANGE_OPAQUE_SFT 1
-	u8 timestamp_lo;
-	__le16 timestamp_hi;
-	__le32 event_data1;
-	#define ASYNC_EVENT_CMPL_PF_VF_COMM_STATUS_CHANGE_EVENT_DATA1_COMM_ESTABLISHED 0x1UL
-};
-
-/* HWRM Asynchronous Event Completion Record for VF configuration change (16 bytes) */
-struct hwrm_async_event_cmpl_vf_cfg_change {
-	__le16 type;
-	#define ASYNC_EVENT_CMPL_VF_CFG_CHANGE_TYPE_MASK	    0x3fUL
-	#define ASYNC_EVENT_CMPL_VF_CFG_CHANGE_TYPE_SFT	    0
-	#define ASYNC_EVENT_CMPL_VF_CFG_CHANGE_TYPE_HWRM_ASYNC_EVENT 0x2eUL
-	__le16 event_id;
-	#define ASYNC_EVENT_CMPL_VF_CFG_CHANGE_EVENT_ID_VF_CFG_CHANGE 0x33UL
-	__le32 event_data2;
-	u8 opaque_v;
-	#define ASYNC_EVENT_CMPL_VF_CFG_CHANGE_V		    0x1UL
-	#define ASYNC_EVENT_CMPL_VF_CFG_CHANGE_OPAQUE_MASK	    0xfeUL
-	#define ASYNC_EVENT_CMPL_VF_CFG_CHANGE_OPAQUE_SFT	    1
-	u8 timestamp_lo;
-	__le16 timestamp_hi;
-	__le32 event_data1;
-	#define ASYNC_EVENT_CMPL_VF_CFG_CHANGE_EVENT_DATA1_MTU_CHANGE 0x1UL
-	#define ASYNC_EVENT_CMPL_VF_CFG_CHANGE_EVENT_DATA1_MRU_CHANGE 0x2UL
-	#define ASYNC_EVENT_CMPL_VF_CFG_CHANGE_EVENT_DATA1_DFLT_MAC_ADDR_CHANGE 0x4UL
-	#define ASYNC_EVENT_CMPL_VF_CFG_CHANGE_EVENT_DATA1_DFLT_VLAN_CHANGE 0x8UL
-};
-
-/* HWRM Asynchronous Event Completion Record for HWRM Error (16 bytes) */
-struct hwrm_async_event_cmpl_hwrm_error {
-	__le16 type;
-	#define ASYNC_EVENT_CMPL_HWRM_ERROR_TYPE_MASK		    0x3fUL
-	#define ASYNC_EVENT_CMPL_HWRM_ERROR_TYPE_SFT		    0
-	#define ASYNC_EVENT_CMPL_HWRM_ERROR_TYPE_HWRM_ASYNC_EVENT 0x2eUL
-	__le16 event_id;
-	#define ASYNC_EVENT_CMPL_HWRM_ERROR_EVENT_ID_HWRM_ERROR   0xffUL
-	__le32 event_data2;
-	#define ASYNC_EVENT_CMPL_HWRM_ERROR_EVENT_DATA2_SEVERITY_MASK 0xffUL
-	#define ASYNC_EVENT_CMPL_HWRM_ERROR_EVENT_DATA2_SEVERITY_SFT 0
-	#define ASYNC_EVENT_CMPL_HWRM_ERROR_EVENT_DATA2_SEVERITY_WARNING 0x0UL
-	#define ASYNC_EVENT_CMPL_HWRM_ERROR_EVENT_DATA2_SEVERITY_NONFATAL 0x1UL
-	#define ASYNC_EVENT_CMPL_HWRM_ERROR_EVENT_DATA2_SEVERITY_FATAL 0x2UL
-	#define ASYNC_EVENT_CMPL_HWRM_ERROR_EVENT_DATA2_SEVERITY_LAST    ASYNC_EVENT_CMPL_HWRM_ERROR_EVENT_DATA2_SEVERITY_FATAL
-	u8 opaque_v;
-	#define ASYNC_EVENT_CMPL_HWRM_ERROR_V			    0x1UL
-	#define ASYNC_EVENT_CMPL_HWRM_ERROR_OPAQUE_MASK	    0xfeUL
-	#define ASYNC_EVENT_CMPL_HWRM_ERROR_OPAQUE_SFT		    1
-	u8 timestamp_lo;
-	__le16 timestamp_hi;
-	__le32 event_data1;
-	#define ASYNC_EVENT_CMPL_HWRM_ERROR_EVENT_DATA1_TIMESTAMP  0x1UL
-};
-
-/* hwrm_ver_get */
-/* Input (24 bytes) */
-struct hwrm_ver_get_input {
-	__le16 req_type;
-	__le16 cmpl_ring;
-	__le16 seq_id;
-	__le16 target_id;
-	__le64 resp_addr;
-	u8 hwrm_intf_maj;
-	u8 hwrm_intf_min;
-	u8 hwrm_intf_upd;
-	u8 unused_0[5];
-};
-
-/* Output (128 bytes) */
-struct hwrm_ver_get_output {
-	__le16 error_code;
-	__le16 req_type;
-	__le16 seq_id;
-	__le16 resp_len;
-	u8 hwrm_intf_maj;
-	u8 hwrm_intf_min;
-	u8 hwrm_intf_upd;
-	u8 hwrm_intf_rsvd;
-	u8 hwrm_fw_maj;
-	u8 hwrm_fw_min;
-	u8 hwrm_fw_bld;
-	u8 hwrm_fw_rsvd;
-	u8 mgmt_fw_maj;
-	u8 mgmt_fw_min;
-	u8 mgmt_fw_bld;
-	u8 mgmt_fw_rsvd;
-	u8 netctrl_fw_maj;
-	u8 netctrl_fw_min;
-	u8 netctrl_fw_bld;
-	u8 netctrl_fw_rsvd;
-	__le32 dev_caps_cfg;
-	#define VER_GET_RESP_DEV_CAPS_CFG_SECURE_FW_UPD_SUPPORTED  0x1UL
-	#define VER_GET_RESP_DEV_CAPS_CFG_FW_DCBX_AGENT_SUPPORTED  0x2UL
-	#define VER_GET_RESP_DEV_CAPS_CFG_SHORT_CMD_SUPPORTED      0x4UL
-	#define VER_GET_RESP_DEV_CAPS_CFG_SHORT_CMD_REQUIRED       0x8UL
-	u8 roce_fw_maj;
-	u8 roce_fw_min;
-	u8 roce_fw_bld;
-	u8 roce_fw_rsvd;
-	char hwrm_fw_name[16];
-	char mgmt_fw_name[16];
-	char netctrl_fw_name[16];
-	__le32 reserved2[4];
-	char roce_fw_name[16];
-	__le16 chip_num;
-	u8 chip_rev;
-	u8 chip_metal;
-	u8 chip_bond_id;
-	u8 chip_platform_type;
-	#define VER_GET_RESP_CHIP_PLATFORM_TYPE_ASIC		   0x0UL
-	#define VER_GET_RESP_CHIP_PLATFORM_TYPE_FPGA		   0x1UL
-	#define VER_GET_RESP_CHIP_PLATFORM_TYPE_PALLADIUM	   0x2UL
-	__le16 max_req_win_len;
-	__le16 max_resp_len;
-	__le16 def_req_timeout;
-	u8 init_pending;
-	#define VER_GET_RESP_INIT_PENDING_DEV_NOT_RDY		    0x1UL
-	u8 unused_0;
-	u8 unused_1;
-	u8 valid;
-};
-
-/* hwrm_func_reset */
-/* Input (24 bytes) */
-struct hwrm_func_reset_input {
-	__le16 req_type;
-	__le16 cmpl_ring;
-	__le16 seq_id;
-	__le16 target_id;
-	__le64 resp_addr;
-	__le32 enables;
-	#define FUNC_RESET_REQ_ENABLES_VF_ID_VALID		    0x1UL
-	__le16 vf_id;
-	u8 func_reset_level;
-	#define FUNC_RESET_REQ_FUNC_RESET_LEVEL_RESETALL	   0x0UL
-	#define FUNC_RESET_REQ_FUNC_RESET_LEVEL_RESETME	   0x1UL
-	#define FUNC_RESET_REQ_FUNC_RESET_LEVEL_RESETCHILDREN     0x2UL
-	#define FUNC_RESET_REQ_FUNC_RESET_LEVEL_RESETVF	   0x3UL
-	u8 unused_0;
-};
-
-/* Output (16 bytes) */
-struct hwrm_func_reset_output {
-	__le16 error_code;
-	__le16 req_type;
-	__le16 seq_id;
-	__le16 resp_len;
-	__le32 unused_0;
-	u8 unused_1;
-	u8 unused_2;
-	u8 unused_3;
-	u8 valid;
-};
-
-/* hwrm_func_getfid */
-/* Input (24 bytes) */
-struct hwrm_func_getfid_input {
-	__le16 req_type;
-	__le16 cmpl_ring;
-	__le16 seq_id;
-	__le16 target_id;
-	__le64 resp_addr;
-	__le32 enables;
-	#define FUNC_GETFID_REQ_ENABLES_PCI_ID			    0x1UL
-	__le16 pci_id;
-	__le16 unused_0;
-};
-
-/* Output (16 bytes) */
-struct hwrm_func_getfid_output {
-	__le16 error_code;
-	__le16 req_type;
-	__le16 seq_id;
-	__le16 resp_len;
-	__le16 fid;
-	u8 unused_0;
-	u8 unused_1;
-	u8 unused_2;
-	u8 unused_3;
-	u8 unused_4;
-	u8 valid;
-};
-
-/* hwrm_func_vf_alloc */
-/* Input (24 bytes) */
-struct hwrm_func_vf_alloc_input {
-	__le16 req_type;
-	__le16 cmpl_ring;
-	__le16 seq_id;
-	__le16 target_id;
-	__le64 resp_addr;
-	__le32 enables;
-	#define FUNC_VF_ALLOC_REQ_ENABLES_FIRST_VF_ID		    0x1UL
-	__le16 first_vf_id;
-	__le16 num_vfs;
-};
-
-/* Output (16 bytes) */
-struct hwrm_func_vf_alloc_output {
-	__le16 error_code;
-	__le16 req_type;
-	__le16 seq_id;
-	__le16 resp_len;
-	__le16 first_vf_id;
-	u8 unused_0;
-	u8 unused_1;
-	u8 unused_2;
-	u8 unused_3;
-	u8 unused_4;
-	u8 valid;
-};
-
-/* hwrm_func_vf_free */
-/* Input (24 bytes) */
-struct hwrm_func_vf_free_input {
-	__le16 req_type;
-	__le16 cmpl_ring;
-	__le16 seq_id;
-	__le16 target_id;
-	__le64 resp_addr;
-	__le32 enables;
-	#define FUNC_VF_FREE_REQ_ENABLES_FIRST_VF_ID		    0x1UL
-	__le16 first_vf_id;
-	__le16 num_vfs;
-};
-
-/* Output (16 bytes) */
-struct hwrm_func_vf_free_output {
-	__le16 error_code;
-	__le16 req_type;
-	__le16 seq_id;
-	__le16 resp_len;
-	__le32 unused_0;
-	u8 unused_1;
-	u8 unused_2;
-	u8 unused_3;
-	u8 valid;
-};
-
-/* hwrm_func_vf_cfg */
-/* Input (32 bytes) */
-struct hwrm_func_vf_cfg_input {
-	__le16 req_type;
-	__le16 cmpl_ring;
-	__le16 seq_id;
-	__le16 target_id;
-	__le64 resp_addr;
-	__le32 enables;
-	#define FUNC_VF_CFG_REQ_ENABLES_MTU			    0x1UL
-	#define FUNC_VF_CFG_REQ_ENABLES_GUEST_VLAN		    0x2UL
-	#define FUNC_VF_CFG_REQ_ENABLES_ASYNC_EVENT_CR		    0x4UL
-	#define FUNC_VF_CFG_REQ_ENABLES_DFLT_MAC_ADDR		    0x8UL
-	__le16 mtu;
-	__le16 guest_vlan;
-	__le16 async_event_cr;
-	u8 dflt_mac_addr[6];
-};
-
-/* Output (16 bytes) */
-struct hwrm_func_vf_cfg_output {
-	__le16 error_code;
-	__le16 req_type;
-	__le16 seq_id;
-	__le16 resp_len;
-	__le32 unused_0;
-	u8 unused_1;
-	u8 unused_2;
-	u8 unused_3;
-	u8 valid;
-};
-
-/* hwrm_func_qcaps */
-/* Input (24 bytes) */
-struct hwrm_func_qcaps_input {
-	__le16 req_type;
-	__le16 cmpl_ring;
-	__le16 seq_id;
-	__le16 target_id;
-	__le64 resp_addr;
-	__le16 fid;
-	__le16 unused_0[3];
-};
-
-/* Output (80 bytes) */
-struct hwrm_func_qcaps_output {
-	__le16 error_code;
-	__le16 req_type;
-	__le16 seq_id;
-	__le16 resp_len;
-	__le16 fid;
-	__le16 port_id;
-	__le32 flags;
-	#define FUNC_QCAPS_RESP_FLAGS_PUSH_MODE_SUPPORTED	    0x1UL
-	#define FUNC_QCAPS_RESP_FLAGS_GLOBAL_MSIX_AUTOMASKING      0x2UL
-	#define FUNC_QCAPS_RESP_FLAGS_PTP_SUPPORTED		    0x4UL
-	#define FUNC_QCAPS_RESP_FLAGS_ROCE_V1_SUPPORTED	    0x8UL
-	#define FUNC_QCAPS_RESP_FLAGS_ROCE_V2_SUPPORTED	    0x10UL
-	#define FUNC_QCAPS_RESP_FLAGS_WOL_MAGICPKT_SUPPORTED       0x20UL
-	#define FUNC_QCAPS_RESP_FLAGS_WOL_BMP_SUPPORTED	    0x40UL
-	#define FUNC_QCAPS_RESP_FLAGS_TX_RING_RL_SUPPORTED	    0x80UL
-	#define FUNC_QCAPS_RESP_FLAGS_TX_BW_CFG_SUPPORTED	    0x100UL
-	#define FUNC_QCAPS_RESP_FLAGS_VF_TX_RING_RL_SUPPORTED      0x200UL
-	#define FUNC_QCAPS_RESP_FLAGS_VF_BW_CFG_SUPPORTED	    0x400UL
-	#define FUNC_QCAPS_RESP_FLAGS_STD_TX_RING_MODE_SUPPORTED   0x800UL
-	u8 mac_address[6];
-	__le16 max_rsscos_ctx;
-	__le16 max_cmpl_rings;
-	__le16 max_tx_rings;
-	__le16 max_rx_rings;
-	__le16 max_l2_ctxs;
-	__le16 max_vnics;
-	__le16 first_vf_id;
-	__le16 max_vfs;
-	__le16 max_stat_ctx;
-	__le32 max_encap_records;
-	__le32 max_decap_records;
-	__le32 max_tx_em_flows;
-	__le32 max_tx_wm_flows;
-	__le32 max_rx_em_flows;
-	__le32 max_rx_wm_flows;
-	__le32 max_mcast_filters;
-	__le32 max_flow_id;
-	__le32 max_hw_ring_grps;
-	__le16 max_sp_tx_rings;
-	u8 unused_0;
-	u8 valid;
-};
-
-/* hwrm_func_qcfg */
-/* Input (24 bytes) */
-struct hwrm_func_qcfg_input {
-	__le16 req_type;
-	__le16 cmpl_ring;
-	__le16 seq_id;
-	__le16 target_id;
-	__le64 resp_addr;
-	__le16 fid;
-	__le16 unused_0[3];
-};
-
-/* Output (72 bytes) */
-struct hwrm_func_qcfg_output {
-	__le16 error_code;
-	__le16 req_type;
-	__le16 seq_id;
-	__le16 resp_len;
-	__le16 fid;
-	__le16 port_id;
-	__le16 vlan;
-	__le16 flags;
-	#define FUNC_QCFG_RESP_FLAGS_OOB_WOL_MAGICPKT_ENABLED      0x1UL
-	#define FUNC_QCFG_RESP_FLAGS_OOB_WOL_BMP_ENABLED	    0x2UL
-	#define FUNC_QCFG_RESP_FLAGS_FW_DCBX_AGENT_ENABLED	    0x4UL
-	#define FUNC_QCFG_RESP_FLAGS_STD_TX_RING_MODE_ENABLED      0x8UL
-	#define FUNC_QCFG_RESP_FLAGS_FW_LLDP_AGENT_ENABLED	    0x10UL
-	#define FUNC_QCFG_RESP_FLAGS_MULTI_HOST			    0x20UL
-	u8 mac_address[6];
-	__le16 pci_id;
-	__le16 alloc_rsscos_ctx;
-	__le16 alloc_cmpl_rings;
-	__le16 alloc_tx_rings;
-	__le16 alloc_rx_rings;
-	__le16 alloc_l2_ctx;
-	__le16 alloc_vnics;
-	__le16 mtu;
-	__le16 mru;
-	__le16 stat_ctx_id;
-	u8 port_partition_type;
-	#define FUNC_QCFG_RESP_PORT_PARTITION_TYPE_SPF		   0x0UL
-	#define FUNC_QCFG_RESP_PORT_PARTITION_TYPE_MPFS	   0x1UL
-	#define FUNC_QCFG_RESP_PORT_PARTITION_TYPE_NPAR1_0	   0x2UL
-	#define FUNC_QCFG_RESP_PORT_PARTITION_TYPE_NPAR1_5	   0x3UL
-	#define FUNC_QCFG_RESP_PORT_PARTITION_TYPE_NPAR2_0	   0x4UL
-	#define FUNC_QCFG_RESP_PORT_PARTITION_TYPE_UNKNOWN	   0xffUL
-	u8 port_pf_cnt;
-	#define FUNC_QCFG_RESP_PORT_PF_CNT_UNAVAIL		   0x0UL
-	__le16 dflt_vnic_id;
-	u8 host_cnt;
-	#define FUNC_QCFG_RESP_HOST_CNT_UNAVAIL		   0x0UL
-	u8 unused_0;
-	__le32 min_bw;
-	#define FUNC_QCFG_RESP_MIN_BW_BW_VALUE_MASK		    0xfffffffUL
-	#define FUNC_QCFG_RESP_MIN_BW_BW_VALUE_SFT		    0
-	#define FUNC_QCFG_RESP_MIN_BW_SCALE			    0x10000000UL
-	#define FUNC_QCFG_RESP_MIN_BW_SCALE_BITS		   (0x0UL << 28)
-	#define FUNC_QCFG_RESP_MIN_BW_SCALE_BYTES		   (0x1UL << 28)
-	#define FUNC_QCFG_RESP_MIN_BW_SCALE_LAST    FUNC_QCFG_RESP_MIN_BW_SCALE_BYTES
-	#define FUNC_QCFG_RESP_MIN_BW_BW_VALUE_UNIT_MASK	    0xe0000000UL
-	#define FUNC_QCFG_RESP_MIN_BW_BW_VALUE_UNIT_SFT	    29
-	#define FUNC_QCFG_RESP_MIN_BW_BW_VALUE_UNIT_MEGA	   (0x0UL << 29)
-	#define FUNC_QCFG_RESP_MIN_BW_BW_VALUE_UNIT_KILO	   (0x2UL << 29)
-	#define FUNC_QCFG_RESP_MIN_BW_BW_VALUE_UNIT_BASE	   (0x4UL << 29)
-	#define FUNC_QCFG_RESP_MIN_BW_BW_VALUE_UNIT_GIGA	   (0x6UL << 29)
-	#define FUNC_QCFG_RESP_MIN_BW_BW_VALUE_UNIT_PERCENT1_100  (0x1UL << 29)
-	#define FUNC_QCFG_RESP_MIN_BW_BW_VALUE_UNIT_INVALID       (0x7UL << 29)
-	#define FUNC_QCFG_RESP_MIN_BW_BW_VALUE_UNIT_LAST    FUNC_QCFG_RESP_MIN_BW_BW_VALUE_UNIT_INVALID
-	__le32 max_bw;
-	#define FUNC_QCFG_RESP_MAX_BW_BW_VALUE_MASK		    0xfffffffUL
-	#define FUNC_QCFG_RESP_MAX_BW_BW_VALUE_SFT		    0
-	#define FUNC_QCFG_RESP_MAX_BW_SCALE			    0x10000000UL
-	#define FUNC_QCFG_RESP_MAX_BW_SCALE_BITS		   (0x0UL << 28)
-	#define FUNC_QCFG_RESP_MAX_BW_SCALE_BYTES		   (0x1UL << 28)
-	#define FUNC_QCFG_RESP_MAX_BW_SCALE_LAST    FUNC_QCFG_RESP_MAX_BW_SCALE_BYTES
-	#define FUNC_QCFG_RESP_MAX_BW_BW_VALUE_UNIT_MASK	    0xe0000000UL
-	#define FUNC_QCFG_RESP_MAX_BW_BW_VALUE_UNIT_SFT	    29
-	#define FUNC_QCFG_RESP_MAX_BW_BW_VALUE_UNIT_MEGA	   (0x0UL << 29)
-	#define FUNC_QCFG_RESP_MAX_BW_BW_VALUE_UNIT_KILO	   (0x2UL << 29)
-	#define FUNC_QCFG_RESP_MAX_BW_BW_VALUE_UNIT_BASE	   (0x4UL << 29)
-	#define FUNC_QCFG_RESP_MAX_BW_BW_VALUE_UNIT_GIGA	   (0x6UL << 29)
-	#define FUNC_QCFG_RESP_MAX_BW_BW_VALUE_UNIT_PERCENT1_100  (0x1UL << 29)
-	#define FUNC_QCFG_RESP_MAX_BW_BW_VALUE_UNIT_INVALID       (0x7UL << 29)
-	#define FUNC_QCFG_RESP_MAX_BW_BW_VALUE_UNIT_LAST    FUNC_QCFG_RESP_MAX_BW_BW_VALUE_UNIT_INVALID
-	u8 evb_mode;
-	#define FUNC_QCFG_RESP_EVB_MODE_NO_EVB			   0x0UL
-	#define FUNC_QCFG_RESP_EVB_MODE_VEB			   0x1UL
-	#define FUNC_QCFG_RESP_EVB_MODE_VEPA			   0x2UL
-	u8 unused_1;
-	__le16 alloc_vfs;
-	__le32 alloc_mcast_filters;
-	__le32 alloc_hw_ring_grps;
-	__le16 alloc_sp_tx_rings;
-	u8 unused_2;
-	u8 valid;
-};
-
-/* hwrm_func_cfg */
-/* Input (88 bytes) */
-struct hwrm_func_cfg_input {
-	__le16 req_type;
-	__le16 cmpl_ring;
-	__le16 seq_id;
-	__le16 target_id;
-	__le64 resp_addr;
-	__le16 fid;
-	u8 unused_0;
-	u8 unused_1;
-	__le32 flags;
-	#define FUNC_CFG_REQ_FLAGS_SRC_MAC_ADDR_CHECK_DISABLE      0x1UL
-	#define FUNC_CFG_REQ_FLAGS_SRC_MAC_ADDR_CHECK_ENABLE       0x2UL
-	#define FUNC_CFG_REQ_FLAGS_RSVD_MASK			    0x1fcUL
-	#define FUNC_CFG_REQ_FLAGS_RSVD_SFT			    2
-	#define FUNC_CFG_REQ_FLAGS_STD_TX_RING_MODE_ENABLE	    0x200UL
-	#define FUNC_CFG_REQ_FLAGS_STD_TX_RING_MODE_DISABLE	    0x400UL
-	#define FUNC_CFG_REQ_FLAGS_VIRT_MAC_PERSIST		    0x800UL
-	__le32 enables;
-	#define FUNC_CFG_REQ_ENABLES_MTU			    0x1UL
-	#define FUNC_CFG_REQ_ENABLES_MRU			    0x2UL
-	#define FUNC_CFG_REQ_ENABLES_NUM_RSSCOS_CTXS		    0x4UL
-	#define FUNC_CFG_REQ_ENABLES_NUM_CMPL_RINGS		    0x8UL
-	#define FUNC_CFG_REQ_ENABLES_NUM_TX_RINGS		    0x10UL
-	#define FUNC_CFG_REQ_ENABLES_NUM_RX_RINGS		    0x20UL
-	#define FUNC_CFG_REQ_ENABLES_NUM_L2_CTXS		    0x40UL
-	#define FUNC_CFG_REQ_ENABLES_NUM_VNICS			    0x80UL
-	#define FUNC_CFG_REQ_ENABLES_NUM_STAT_CTXS		    0x100UL
-	#define FUNC_CFG_REQ_ENABLES_DFLT_MAC_ADDR		    0x200UL
-	#define FUNC_CFG_REQ_ENABLES_DFLT_VLAN			    0x400UL
-	#define FUNC_CFG_REQ_ENABLES_DFLT_IP_ADDR		    0x800UL
-	#define FUNC_CFG_REQ_ENABLES_MIN_BW			    0x1000UL
-	#define FUNC_CFG_REQ_ENABLES_MAX_BW			    0x2000UL
-	#define FUNC_CFG_REQ_ENABLES_ASYNC_EVENT_CR		    0x4000UL
-	#define FUNC_CFG_REQ_ENABLES_VLAN_ANTISPOOF_MODE	    0x8000UL
-	#define FUNC_CFG_REQ_ENABLES_ALLOWED_VLAN_PRIS		    0x10000UL
-	#define FUNC_CFG_REQ_ENABLES_EVB_MODE			    0x20000UL
-	#define FUNC_CFG_REQ_ENABLES_NUM_MCAST_FILTERS		    0x40000UL
-	#define FUNC_CFG_REQ_ENABLES_NUM_HW_RING_GRPS		    0x80000UL
-	__le16 mtu;
-	__le16 mru;
-	__le16 num_rsscos_ctxs;
-	__le16 num_cmpl_rings;
-	__le16 num_tx_rings;
-	__le16 num_rx_rings;
-	__le16 num_l2_ctxs;
-	__le16 num_vnics;
-	__le16 num_stat_ctxs;
-	__le16 num_hw_ring_grps;
-	u8 dflt_mac_addr[6];
-	__le16 dflt_vlan;
-	__be32 dflt_ip_addr[4];
-	__le32 min_bw;
-	#define FUNC_CFG_REQ_MIN_BW_BW_VALUE_MASK		    0xfffffffUL
-	#define FUNC_CFG_REQ_MIN_BW_BW_VALUE_SFT		    0
-	#define FUNC_CFG_REQ_MIN_BW_SCALE			    0x10000000UL
-	#define FUNC_CFG_REQ_MIN_BW_SCALE_BITS			   (0x0UL << 28)
-	#define FUNC_CFG_REQ_MIN_BW_SCALE_BYTES		   (0x1UL << 28)
-	#define FUNC_CFG_REQ_MIN_BW_SCALE_LAST    FUNC_CFG_REQ_MIN_BW_SCALE_BYTES
-	#define FUNC_CFG_REQ_MIN_BW_BW_VALUE_UNIT_MASK		    0xe0000000UL
-	#define FUNC_CFG_REQ_MIN_BW_BW_VALUE_UNIT_SFT		    29
-	#define FUNC_CFG_REQ_MIN_BW_BW_VALUE_UNIT_MEGA		   (0x0UL << 29)
-	#define FUNC_CFG_REQ_MIN_BW_BW_VALUE_UNIT_KILO		   (0x2UL << 29)
-	#define FUNC_CFG_REQ_MIN_BW_BW_VALUE_UNIT_BASE		   (0x4UL << 29)
-	#define FUNC_CFG_REQ_MIN_BW_BW_VALUE_UNIT_GIGA		   (0x6UL << 29)
-	#define FUNC_CFG_REQ_MIN_BW_BW_VALUE_UNIT_PERCENT1_100    (0x1UL << 29)
-	#define FUNC_CFG_REQ_MIN_BW_BW_VALUE_UNIT_INVALID	   (0x7UL << 29)
-	#define FUNC_CFG_REQ_MIN_BW_BW_VALUE_UNIT_LAST    FUNC_CFG_REQ_MIN_BW_BW_VALUE_UNIT_INVALID
-	__le32 max_bw;
-	#define FUNC_CFG_REQ_MAX_BW_BW_VALUE_MASK		    0xfffffffUL
-	#define FUNC_CFG_REQ_MAX_BW_BW_VALUE_SFT		    0
-	#define FUNC_CFG_REQ_MAX_BW_SCALE			    0x10000000UL
-	#define FUNC_CFG_REQ_MAX_BW_SCALE_BITS			   (0x0UL << 28)
-	#define FUNC_CFG_REQ_MAX_BW_SCALE_BYTES		   (0x1UL << 28)
-	#define FUNC_CFG_REQ_MAX_BW_SCALE_LAST    FUNC_CFG_REQ_MAX_BW_SCALE_BYTES
-	#define FUNC_CFG_REQ_MAX_BW_BW_VALUE_UNIT_MASK		    0xe0000000UL
-	#define FUNC_CFG_REQ_MAX_BW_BW_VALUE_UNIT_SFT		    29
-	#define FUNC_CFG_REQ_MAX_BW_BW_VALUE_UNIT_MEGA		   (0x0UL << 29)
-	#define FUNC_CFG_REQ_MAX_BW_BW_VALUE_UNIT_KILO		   (0x2UL << 29)
-	#define FUNC_CFG_REQ_MAX_BW_BW_VALUE_UNIT_BASE		   (0x4UL << 29)
-	#define FUNC_CFG_REQ_MAX_BW_BW_VALUE_UNIT_GIGA		   (0x6UL << 29)
-	#define FUNC_CFG_REQ_MAX_BW_BW_VALUE_UNIT_PERCENT1_100    (0x1UL << 29)
-	#define FUNC_CFG_REQ_MAX_BW_BW_VALUE_UNIT_INVALID	   (0x7UL << 29)
-	#define FUNC_CFG_REQ_MAX_BW_BW_VALUE_UNIT_LAST    FUNC_CFG_REQ_MAX_BW_BW_VALUE_UNIT_INVALID
-	__le16 async_event_cr;
-	u8 vlan_antispoof_mode;
-	#define FUNC_CFG_REQ_VLAN_ANTISPOOF_MODE_NOCHECK	   0x0UL
-	#define FUNC_CFG_REQ_VLAN_ANTISPOOF_MODE_VALIDATE_VLAN    0x1UL
-	#define FUNC_CFG_REQ_VLAN_ANTISPOOF_MODE_INSERT_IF_VLANDNE 0x2UL
-	#define FUNC_CFG_REQ_VLAN_ANTISPOOF_MODE_INSERT_OR_OVERRIDE_VLAN 0x3UL
-	u8 allowed_vlan_pris;
-	u8 evb_mode;
-	#define FUNC_CFG_REQ_EVB_MODE_NO_EVB			   0x0UL
-	#define FUNC_CFG_REQ_EVB_MODE_VEB			   0x1UL
-	#define FUNC_CFG_REQ_EVB_MODE_VEPA			   0x2UL
-	u8 unused_2;
-	__le16 num_mcast_filters;
-};
-
-/* Output (16 bytes) */
-struct hwrm_func_cfg_output {
-	__le16 error_code;
-	__le16 req_type;
-	__le16 seq_id;
-	__le16 resp_len;
-	__le32 unused_0;
-	u8 unused_1;
-	u8 unused_2;
-	u8 unused_3;
-	u8 valid;
-};
-
-/* hwrm_func_qstats */
-/* Input (24 bytes) */
-struct hwrm_func_qstats_input {
-	__le16 req_type;
-	__le16 cmpl_ring;
-	__le16 seq_id;
-	__le16 target_id;
-	__le64 resp_addr;
-	__le16 fid;
-	__le16 unused_0[3];
-};
-
-/* Output (176 bytes) */
-struct hwrm_func_qstats_output {
-	__le16 error_code;
-	__le16 req_type;
-	__le16 seq_id;
-	__le16 resp_len;
-	__le64 tx_ucast_pkts;
-	__le64 tx_mcast_pkts;
-	__le64 tx_bcast_pkts;
-	__le64 tx_discard_pkts;
-	__le64 tx_drop_pkts;
-	__le64 tx_ucast_bytes;
-	__le64 tx_mcast_bytes;
-	__le64 tx_bcast_bytes;
-	__le64 rx_ucast_pkts;
-	__le64 rx_mcast_pkts;
-	__le64 rx_bcast_pkts;
-	__le64 rx_discard_pkts;
-	__le64 rx_drop_pkts;
-	__le64 rx_ucast_bytes;
-	__le64 rx_mcast_bytes;
-	__le64 rx_bcast_bytes;
-	__le64 rx_agg_pkts;
-	__le64 rx_agg_bytes;
-	__le64 rx_agg_events;
-	__le64 rx_agg_aborts;
-	__le32 unused_0;
-	u8 unused_1;
-	u8 unused_2;
-	u8 unused_3;
-	u8 valid;
-};
-
-/* hwrm_func_clr_stats */
-/* Input (24 bytes) */
-struct hwrm_func_clr_stats_input {
-	__le16 req_type;
-	__le16 cmpl_ring;
-	__le16 seq_id;
-	__le16 target_id;
-	__le64 resp_addr;
-	__le16 fid;
-	__le16 unused_0[3];
-};
-
-/* Output (16 bytes) */
-struct hwrm_func_clr_stats_output {
-	__le16 error_code;
-	__le16 req_type;
-	__le16 seq_id;
-	__le16 resp_len;
-	__le32 unused_0;
-	u8 unused_1;
-	u8 unused_2;
-	u8 unused_3;
-	u8 valid;
-};
-
-/* hwrm_func_vf_resc_free */
-/* Input (24 bytes) */
-struct hwrm_func_vf_resc_free_input {
-	__le16 req_type;
-	__le16 cmpl_ring;
-	__le16 seq_id;
-	__le16 target_id;
-	__le64 resp_addr;
-	__le16 vf_id;
-	__le16 unused_0[3];
-};
-
-/* Output (16 bytes) */
-struct hwrm_func_vf_resc_free_output {
-	__le16 error_code;
-	__le16 req_type;
-	__le16 seq_id;
-	__le16 resp_len;
-	__le32 unused_0;
-	u8 unused_1;
-	u8 unused_2;
-	u8 unused_3;
-	u8 valid;
-};
-
-/* hwrm_func_vf_vnic_ids_query */
-/* Input (32 bytes) */
-struct hwrm_func_vf_vnic_ids_query_input {
-	__le16 req_type;
-	__le16 cmpl_ring;
-	__le16 seq_id;
-	__le16 target_id;
-	__le64 resp_addr;
-	__le16 vf_id;
-	u8 unused_0;
-	u8 unused_1;
-	__le32 max_vnic_id_cnt;
-	__le64 vnic_id_tbl_addr;
-};
-
-/* Output (16 bytes) */
-struct hwrm_func_vf_vnic_ids_query_output {
-	__le16 error_code;
-	__le16 req_type;
-	__le16 seq_id;
-	__le16 resp_len;
-	__le32 vnic_id_cnt;
-	u8 unused_0;
-	u8 unused_1;
-	u8 unused_2;
-	u8 valid;
-};
-
-/* hwrm_func_drv_rgtr */
-/* Input (80 bytes) */
-struct hwrm_func_drv_rgtr_input {
-	__le16 req_type;
-	__le16 cmpl_ring;
-	__le16 seq_id;
-	__le16 target_id;
-	__le64 resp_addr;
-	__le32 flags;
-	#define FUNC_DRV_RGTR_REQ_FLAGS_FWD_ALL_MODE		    0x1UL
-	#define FUNC_DRV_RGTR_REQ_FLAGS_FWD_NONE_MODE		    0x2UL
-	__le32 enables;
-	#define FUNC_DRV_RGTR_REQ_ENABLES_OS_TYPE		    0x1UL
-	#define FUNC_DRV_RGTR_REQ_ENABLES_VER			    0x2UL
-	#define FUNC_DRV_RGTR_REQ_ENABLES_TIMESTAMP		    0x4UL
-	#define FUNC_DRV_RGTR_REQ_ENABLES_VF_REQ_FWD		    0x8UL
-	#define FUNC_DRV_RGTR_REQ_ENABLES_ASYNC_EVENT_FWD	    0x10UL
-	__le16 os_type;
-	#define FUNC_DRV_RGTR_REQ_OS_TYPE_UNKNOWN		   0x0UL
-	#define FUNC_DRV_RGTR_REQ_OS_TYPE_OTHER		   0x1UL
-	#define FUNC_DRV_RGTR_REQ_OS_TYPE_MSDOS		   0xeUL
-	#define FUNC_DRV_RGTR_REQ_OS_TYPE_WINDOWS		   0x12UL
-	#define FUNC_DRV_RGTR_REQ_OS_TYPE_SOLARIS		   0x1dUL
-	#define FUNC_DRV_RGTR_REQ_OS_TYPE_LINUX		   0x24UL
-	#define FUNC_DRV_RGTR_REQ_OS_TYPE_FREEBSD		   0x2aUL
-	#define FUNC_DRV_RGTR_REQ_OS_TYPE_ESXI			   0x68UL
-	#define FUNC_DRV_RGTR_REQ_OS_TYPE_WIN864		   0x73UL
-	#define FUNC_DRV_RGTR_REQ_OS_TYPE_WIN2012R2		   0x74UL
-	#define FUNC_DRV_RGTR_REQ_OS_TYPE_UEFI			   0x8000UL
-	u8 ver_maj;
-	u8 ver_min;
-	u8 ver_upd;
-	u8 unused_0;
-	__le16 unused_1;
-	__le32 timestamp;
-	__le32 unused_2;
-	__le32 vf_req_fwd[8];
-	__le32 async_event_fwd[8];
-};
-
-/* Output (16 bytes) */
-struct hwrm_func_drv_rgtr_output {
-	__le16 error_code;
-	__le16 req_type;
-	__le16 seq_id;
-	__le16 resp_len;
-	__le32 unused_0;
-	u8 unused_1;
-	u8 unused_2;
-	u8 unused_3;
-	u8 valid;
-};
-
-/* hwrm_func_drv_unrgtr */
-/* Input (24 bytes) */
-struct hwrm_func_drv_unrgtr_input {
-	__le16 req_type;
-	__le16 cmpl_ring;
-	__le16 seq_id;
-	__le16 target_id;
-	__le64 resp_addr;
-	__le32 flags;
-	#define FUNC_DRV_UNRGTR_REQ_FLAGS_PREPARE_FOR_SHUTDOWN     0x1UL
-	__le32 unused_0;
-};
-
-/* Output (16 bytes) */
-struct hwrm_func_drv_unrgtr_output {
-	__le16 error_code;
-	__le16 req_type;
-	__le16 seq_id;
-	__le16 resp_len;
-	__le32 unused_0;
-	u8 unused_1;
-	u8 unused_2;
-	u8 unused_3;
-	u8 valid;
-};
-
-/* hwrm_func_buf_rgtr */
-/* Input (128 bytes) */
-struct hwrm_func_buf_rgtr_input {
-	__le16 req_type;
-	__le16 cmpl_ring;
-	__le16 seq_id;
-	__le16 target_id;
-	__le64 resp_addr;
-	__le32 enables;
-	#define FUNC_BUF_RGTR_REQ_ENABLES_VF_ID		    0x1UL
-	#define FUNC_BUF_RGTR_REQ_ENABLES_ERR_BUF_ADDR		    0x2UL
-	__le16 vf_id;
-	__le16 req_buf_num_pages;
-	__le16 req_buf_page_size;
-	#define FUNC_BUF_RGTR_REQ_REQ_BUF_PAGE_SIZE_16B	   0x4UL
-	#define FUNC_BUF_RGTR_REQ_REQ_BUF_PAGE_SIZE_4K		   0xcUL
-	#define FUNC_BUF_RGTR_REQ_REQ_BUF_PAGE_SIZE_8K		   0xdUL
-	#define FUNC_BUF_RGTR_REQ_REQ_BUF_PAGE_SIZE_64K	   0x10UL
-	#define FUNC_BUF_RGTR_REQ_REQ_BUF_PAGE_SIZE_2M		   0x15UL
-	#define FUNC_BUF_RGTR_REQ_REQ_BUF_PAGE_SIZE_4M		   0x16UL
-	#define FUNC_BUF_RGTR_REQ_REQ_BUF_PAGE_SIZE_1G		   0x1eUL
-	__le16 req_buf_len;
-	__le16 resp_buf_len;
-	u8 unused_0;
-	u8 unused_1;
-	__le64 req_buf_page_addr0;
-	__le64 req_buf_page_addr1;
-	__le64 req_buf_page_addr2;
-	__le64 req_buf_page_addr3;
-	__le64 req_buf_page_addr4;
-	__le64 req_buf_page_addr5;
-	__le64 req_buf_page_addr6;
-	__le64 req_buf_page_addr7;
-	__le64 req_buf_page_addr8;
-	__le64 req_buf_page_addr9;
-	__le64 error_buf_addr;
-	__le64 resp_buf_addr;
-};
-
-/* Output (16 bytes) */
-struct hwrm_func_buf_rgtr_output {
-	__le16 error_code;
-	__le16 req_type;
-	__le16 seq_id;
-	__le16 resp_len;
-	__le32 unused_0;
-	u8 unused_1;
-	u8 unused_2;
-	u8 unused_3;
-	u8 valid;
-};
-
-/* hwrm_func_drv_qver */
-/* Input (24 bytes) */
-struct hwrm_func_drv_qver_input {
-	__le16 req_type;
-	__le16 cmpl_ring;
-	__le16 seq_id;
-	__le16 target_id;
-	__le64 resp_addr;
-	__le32 reserved;
-	__le16 fid;
-	__le16 unused_0;
-};
-
-/* Output (16 bytes) */
-struct hwrm_func_drv_qver_output {
-	__le16 error_code;
-	__le16 req_type;
-	__le16 seq_id;
-	__le16 resp_len;
-	__le16 os_type;
-	#define FUNC_DRV_QVER_RESP_OS_TYPE_UNKNOWN		   0x0UL
-	#define FUNC_DRV_QVER_RESP_OS_TYPE_OTHER		   0x1UL
-	#define FUNC_DRV_QVER_RESP_OS_TYPE_MSDOS		   0xeUL
-	#define FUNC_DRV_QVER_RESP_OS_TYPE_WINDOWS		   0x12UL
-	#define FUNC_DRV_QVER_RESP_OS_TYPE_SOLARIS		   0x1dUL
-	#define FUNC_DRV_QVER_RESP_OS_TYPE_LINUX		   0x24UL
-	#define FUNC_DRV_QVER_RESP_OS_TYPE_FREEBSD		   0x2aUL
-	#define FUNC_DRV_QVER_RESP_OS_TYPE_ESXI		   0x68UL
-	#define FUNC_DRV_QVER_RESP_OS_TYPE_WIN864		   0x73UL
-	#define FUNC_DRV_QVER_RESP_OS_TYPE_WIN2012R2		   0x74UL
-	#define FUNC_DRV_QVER_RESP_OS_TYPE_UEFI		   0x8000UL
-	u8 ver_maj;
-	u8 ver_min;
-	u8 ver_upd;
-	u8 unused_0;
-	u8 unused_1;
-	u8 valid;
-};
-
-/* hwrm_port_phy_cfg */
-/* Input (56 bytes) */
-struct hwrm_port_phy_cfg_input {
-	__le16 req_type;
-	__le16 cmpl_ring;
-	__le16 seq_id;
-	__le16 target_id;
-	__le64 resp_addr;
-	__le32 flags;
-	#define PORT_PHY_CFG_REQ_FLAGS_RESET_PHY		    0x1UL
-	#define PORT_PHY_CFG_REQ_FLAGS_DEPRECATED		    0x2UL
-	#define PORT_PHY_CFG_REQ_FLAGS_FORCE			    0x4UL
-	#define PORT_PHY_CFG_REQ_FLAGS_RESTART_AUTONEG		    0x8UL
-	#define PORT_PHY_CFG_REQ_FLAGS_EEE_ENABLE		    0x10UL
-	#define PORT_PHY_CFG_REQ_FLAGS_EEE_DISABLE		    0x20UL
-	#define PORT_PHY_CFG_REQ_FLAGS_EEE_TX_LPI_ENABLE	    0x40UL
-	#define PORT_PHY_CFG_REQ_FLAGS_EEE_TX_LPI_DISABLE	    0x80UL
-	#define PORT_PHY_CFG_REQ_FLAGS_FEC_AUTONEG_ENABLE	    0x100UL
-	#define PORT_PHY_CFG_REQ_FLAGS_FEC_AUTONEG_DISABLE	    0x200UL
-	#define PORT_PHY_CFG_REQ_FLAGS_FEC_CLAUSE74_ENABLE	    0x400UL
-	#define PORT_PHY_CFG_REQ_FLAGS_FEC_CLAUSE74_DISABLE	    0x800UL
-	#define PORT_PHY_CFG_REQ_FLAGS_FEC_CLAUSE91_ENABLE	    0x1000UL
-	#define PORT_PHY_CFG_REQ_FLAGS_FEC_CLAUSE91_DISABLE	    0x2000UL
-	#define PORT_PHY_CFG_REQ_FLAGS_FORCE_LINK_DWN		    0x4000UL
-	__le32 enables;
-	#define PORT_PHY_CFG_REQ_ENABLES_AUTO_MODE		    0x1UL
-	#define PORT_PHY_CFG_REQ_ENABLES_AUTO_DUPLEX		    0x2UL
-	#define PORT_PHY_CFG_REQ_ENABLES_AUTO_PAUSE		    0x4UL
-	#define PORT_PHY_CFG_REQ_ENABLES_AUTO_LINK_SPEED	    0x8UL
-	#define PORT_PHY_CFG_REQ_ENABLES_AUTO_LINK_SPEED_MASK      0x10UL
-	#define PORT_PHY_CFG_REQ_ENABLES_WIRESPEED		    0x20UL
-	#define PORT_PHY_CFG_REQ_ENABLES_LPBK			    0x40UL
-	#define PORT_PHY_CFG_REQ_ENABLES_PREEMPHASIS		    0x80UL
-	#define PORT_PHY_CFG_REQ_ENABLES_FORCE_PAUSE		    0x100UL
-	#define PORT_PHY_CFG_REQ_ENABLES_EEE_LINK_SPEED_MASK       0x200UL
-	#define PORT_PHY_CFG_REQ_ENABLES_TX_LPI_TIMER		    0x400UL
-	__le16 port_id;
-	__le16 force_link_speed;
-	#define PORT_PHY_CFG_REQ_FORCE_LINK_SPEED_100MB	   0x1UL
-	#define PORT_PHY_CFG_REQ_FORCE_LINK_SPEED_1GB		   0xaUL
-	#define PORT_PHY_CFG_REQ_FORCE_LINK_SPEED_2GB		   0x14UL
-	#define PORT_PHY_CFG_REQ_FORCE_LINK_SPEED_2_5GB	   0x19UL
-	#define PORT_PHY_CFG_REQ_FORCE_LINK_SPEED_10GB		   0x64UL
-	#define PORT_PHY_CFG_REQ_FORCE_LINK_SPEED_20GB		   0xc8UL
-	#define PORT_PHY_CFG_REQ_FORCE_LINK_SPEED_25GB		   0xfaUL
-	#define PORT_PHY_CFG_REQ_FORCE_LINK_SPEED_40GB		   0x190UL
-	#define PORT_PHY_CFG_REQ_FORCE_LINK_SPEED_50GB		   0x1f4UL
-	#define PORT_PHY_CFG_REQ_FORCE_LINK_SPEED_100GB	   0x3e8UL
-	#define PORT_PHY_CFG_REQ_FORCE_LINK_SPEED_10MB		   0xffffUL
-	u8 auto_mode;
-	#define PORT_PHY_CFG_REQ_AUTO_MODE_NONE		   0x0UL
-	#define PORT_PHY_CFG_REQ_AUTO_MODE_ALL_SPEEDS		   0x1UL
-	#define PORT_PHY_CFG_REQ_AUTO_MODE_ONE_SPEED		   0x2UL
-	#define PORT_PHY_CFG_REQ_AUTO_MODE_ONE_OR_BELOW	   0x3UL
-	#define PORT_PHY_CFG_REQ_AUTO_MODE_SPEED_MASK		   0x4UL
-	u8 auto_duplex;
-	#define PORT_PHY_CFG_REQ_AUTO_DUPLEX_HALF		   0x0UL
-	#define PORT_PHY_CFG_REQ_AUTO_DUPLEX_FULL		   0x1UL
-	#define PORT_PHY_CFG_REQ_AUTO_DUPLEX_BOTH		   0x2UL
-	u8 auto_pause;
-	#define PORT_PHY_CFG_REQ_AUTO_PAUSE_TX			    0x1UL
-	#define PORT_PHY_CFG_REQ_AUTO_PAUSE_RX			    0x2UL
-	#define PORT_PHY_CFG_REQ_AUTO_PAUSE_AUTONEG_PAUSE	    0x4UL
-	u8 unused_0;
-	__le16 auto_link_speed;
-	#define PORT_PHY_CFG_REQ_AUTO_LINK_SPEED_100MB		   0x1UL
-	#define PORT_PHY_CFG_REQ_AUTO_LINK_SPEED_1GB		   0xaUL
-	#define PORT_PHY_CFG_REQ_AUTO_LINK_SPEED_2GB		   0x14UL
-	#define PORT_PHY_CFG_REQ_AUTO_LINK_SPEED_2_5GB		   0x19UL
-	#define PORT_PHY_CFG_REQ_AUTO_LINK_SPEED_10GB		   0x64UL
-	#define PORT_PHY_CFG_REQ_AUTO_LINK_SPEED_20GB		   0xc8UL
-	#define PORT_PHY_CFG_REQ_AUTO_LINK_SPEED_25GB		   0xfaUL
-	#define PORT_PHY_CFG_REQ_AUTO_LINK_SPEED_40GB		   0x190UL
-	#define PORT_PHY_CFG_REQ_AUTO_LINK_SPEED_50GB		   0x1f4UL
-	#define PORT_PHY_CFG_REQ_AUTO_LINK_SPEED_100GB		   0x3e8UL
-	#define PORT_PHY_CFG_REQ_AUTO_LINK_SPEED_10MB		   0xffffUL
-	__le16 auto_link_speed_mask;
-	#define PORT_PHY_CFG_REQ_AUTO_LINK_SPEED_MASK_100MBHD      0x1UL
-	#define PORT_PHY_CFG_REQ_AUTO_LINK_SPEED_MASK_100MB	    0x2UL
-	#define PORT_PHY_CFG_REQ_AUTO_LINK_SPEED_MASK_1GBHD	    0x4UL
-	#define PORT_PHY_CFG_REQ_AUTO_LINK_SPEED_MASK_1GB	    0x8UL
-	#define PORT_PHY_CFG_REQ_AUTO_LINK_SPEED_MASK_2GB	    0x10UL
-	#define PORT_PHY_CFG_REQ_AUTO_LINK_SPEED_MASK_2_5GB	    0x20UL
-	#define PORT_PHY_CFG_REQ_AUTO_LINK_SPEED_MASK_10GB	    0x40UL
-	#define PORT_PHY_CFG_REQ_AUTO_LINK_SPEED_MASK_20GB	    0x80UL
-	#define PORT_PHY_CFG_REQ_AUTO_LINK_SPEED_MASK_25GB	    0x100UL
-	#define PORT_PHY_CFG_REQ_AUTO_LINK_SPEED_MASK_40GB	    0x200UL
-	#define PORT_PHY_CFG_REQ_AUTO_LINK_SPEED_MASK_50GB	    0x400UL
-	#define PORT_PHY_CFG_REQ_AUTO_LINK_SPEED_MASK_100GB	    0x800UL
-	#define PORT_PHY_CFG_REQ_AUTO_LINK_SPEED_MASK_10MBHD       0x1000UL
-	#define PORT_PHY_CFG_REQ_AUTO_LINK_SPEED_MASK_10MB	    0x2000UL
-	u8 wirespeed;
-	#define PORT_PHY_CFG_REQ_WIRESPEED_OFF			   0x0UL
-	#define PORT_PHY_CFG_REQ_WIRESPEED_ON			   0x1UL
-	u8 lpbk;
-	#define PORT_PHY_CFG_REQ_LPBK_NONE			   0x0UL
-	#define PORT_PHY_CFG_REQ_LPBK_LOCAL			   0x1UL
-	#define PORT_PHY_CFG_REQ_LPBK_REMOTE			   0x2UL
-	u8 force_pause;
-	#define PORT_PHY_CFG_REQ_FORCE_PAUSE_TX		    0x1UL
-	#define PORT_PHY_CFG_REQ_FORCE_PAUSE_RX		    0x2UL
-	u8 unused_1;
-	__le32 preemphasis;
-	__le16 eee_link_speed_mask;
-	#define PORT_PHY_CFG_REQ_EEE_LINK_SPEED_MASK_RSVD1	    0x1UL
-	#define PORT_PHY_CFG_REQ_EEE_LINK_SPEED_MASK_100MB	    0x2UL
-	#define PORT_PHY_CFG_REQ_EEE_LINK_SPEED_MASK_RSVD2	    0x4UL
-	#define PORT_PHY_CFG_REQ_EEE_LINK_SPEED_MASK_1GB	    0x8UL
-	#define PORT_PHY_CFG_REQ_EEE_LINK_SPEED_MASK_RSVD3	    0x10UL
-	#define PORT_PHY_CFG_REQ_EEE_LINK_SPEED_MASK_RSVD4	    0x20UL
-	#define PORT_PHY_CFG_REQ_EEE_LINK_SPEED_MASK_10GB	    0x40UL
-	u8 unused_2;
-	u8 unused_3;
-	__le32 tx_lpi_timer;
-	__le32 unused_4;
-	#define PORT_PHY_CFG_REQ_TX_LPI_TIMER_MASK		    0xffffffUL
-	#define PORT_PHY_CFG_REQ_TX_LPI_TIMER_SFT		    0
-};
-
-/* Output (16 bytes) */
-struct hwrm_port_phy_cfg_output {
-	__le16 error_code;
-	__le16 req_type;
-	__le16 seq_id;
-	__le16 resp_len;
-	__le32 unused_0;
-	u8 unused_1;
-	u8 unused_2;
-	u8 unused_3;
-	u8 valid;
-};
-
-/* hwrm_port_phy_qcfg */
-/* Input (24 bytes) */
-struct hwrm_port_phy_qcfg_input {
-	__le16 req_type;
-	__le16 cmpl_ring;
-	__le16 seq_id;
-	__le16 target_id;
-	__le64 resp_addr;
-	__le16 port_id;
-	__le16 unused_0[3];
-};
-
-/* Output (96 bytes) */
-struct hwrm_port_phy_qcfg_output {
-	__le16 error_code;
-	__le16 req_type;
-	__le16 seq_id;
-	__le16 resp_len;
-	u8 link;
-	#define PORT_PHY_QCFG_RESP_LINK_NO_LINK		   0x0UL
-	#define PORT_PHY_QCFG_RESP_LINK_SIGNAL			   0x1UL
-	#define PORT_PHY_QCFG_RESP_LINK_LINK			   0x2UL
-	u8 unused_0;
-	__le16 link_speed;
-	#define PORT_PHY_QCFG_RESP_LINK_SPEED_100MB		   0x1UL
-	#define PORT_PHY_QCFG_RESP_LINK_SPEED_1GB		   0xaUL
-	#define PORT_PHY_QCFG_RESP_LINK_SPEED_2GB		   0x14UL
-	#define PORT_PHY_QCFG_RESP_LINK_SPEED_2_5GB		   0x19UL
-	#define PORT_PHY_QCFG_RESP_LINK_SPEED_10GB		   0x64UL
-	#define PORT_PHY_QCFG_RESP_LINK_SPEED_20GB		   0xc8UL
-	#define PORT_PHY_QCFG_RESP_LINK_SPEED_25GB		   0xfaUL
-	#define PORT_PHY_QCFG_RESP_LINK_SPEED_40GB		   0x190UL
-	#define PORT_PHY_QCFG_RESP_LINK_SPEED_50GB		   0x1f4UL
-	#define PORT_PHY_QCFG_RESP_LINK_SPEED_100GB		   0x3e8UL
-	#define PORT_PHY_QCFG_RESP_LINK_SPEED_10MB		   0xffffUL
-	u8 duplex;
-	#define PORT_PHY_QCFG_RESP_DUPLEX_HALF			   0x0UL
-	#define PORT_PHY_QCFG_RESP_DUPLEX_FULL			   0x1UL
-	u8 pause;
-	#define PORT_PHY_QCFG_RESP_PAUSE_TX			    0x1UL
-	#define PORT_PHY_QCFG_RESP_PAUSE_RX			    0x2UL
-	__le16 support_speeds;
-	#define PORT_PHY_QCFG_RESP_SUPPORT_SPEEDS_100MBHD	    0x1UL
-	#define PORT_PHY_QCFG_RESP_SUPPORT_SPEEDS_100MB	    0x2UL
-	#define PORT_PHY_QCFG_RESP_SUPPORT_SPEEDS_1GBHD	    0x4UL
-	#define PORT_PHY_QCFG_RESP_SUPPORT_SPEEDS_1GB		    0x8UL
-	#define PORT_PHY_QCFG_RESP_SUPPORT_SPEEDS_2GB		    0x10UL
-	#define PORT_PHY_QCFG_RESP_SUPPORT_SPEEDS_2_5GB	    0x20UL
-	#define PORT_PHY_QCFG_RESP_SUPPORT_SPEEDS_10GB		    0x40UL
-	#define PORT_PHY_QCFG_RESP_SUPPORT_SPEEDS_20GB		    0x80UL
-	#define PORT_PHY_QCFG_RESP_SUPPORT_SPEEDS_25GB		    0x100UL
-	#define PORT_PHY_QCFG_RESP_SUPPORT_SPEEDS_40GB		    0x200UL
-	#define PORT_PHY_QCFG_RESP_SUPPORT_SPEEDS_50GB		    0x400UL
-	#define PORT_PHY_QCFG_RESP_SUPPORT_SPEEDS_100GB	    0x800UL
-	#define PORT_PHY_QCFG_RESP_SUPPORT_SPEEDS_10MBHD	    0x1000UL
-	#define PORT_PHY_QCFG_RESP_SUPPORT_SPEEDS_10MB		    0x2000UL
-	__le16 force_link_speed;
-	#define PORT_PHY_QCFG_RESP_FORCE_LINK_SPEED_100MB	   0x1UL
-	#define PORT_PHY_QCFG_RESP_FORCE_LINK_SPEED_1GB	   0xaUL
-	#define PORT_PHY_QCFG_RESP_FORCE_LINK_SPEED_2GB	   0x14UL
-	#define PORT_PHY_QCFG_RESP_FORCE_LINK_SPEED_2_5GB	   0x19UL
-	#define PORT_PHY_QCFG_RESP_FORCE_LINK_SPEED_10GB	   0x64UL
-	#define PORT_PHY_QCFG_RESP_FORCE_LINK_SPEED_20GB	   0xc8UL
-	#define PORT_PHY_QCFG_RESP_FORCE_LINK_SPEED_25GB	   0xfaUL
-	#define PORT_PHY_QCFG_RESP_FORCE_LINK_SPEED_40GB	   0x190UL
-	#define PORT_PHY_QCFG_RESP_FORCE_LINK_SPEED_50GB	   0x1f4UL
-	#define PORT_PHY_QCFG_RESP_FORCE_LINK_SPEED_100GB	   0x3e8UL
-	#define PORT_PHY_QCFG_RESP_FORCE_LINK_SPEED_10MB	   0xffffUL
-	u8 auto_mode;
-	#define PORT_PHY_QCFG_RESP_AUTO_MODE_NONE		   0x0UL
-	#define PORT_PHY_QCFG_RESP_AUTO_MODE_ALL_SPEEDS	   0x1UL
-	#define PORT_PHY_QCFG_RESP_AUTO_MODE_ONE_SPEED		   0x2UL
-	#define PORT_PHY_QCFG_RESP_AUTO_MODE_ONE_OR_BELOW	   0x3UL
-	#define PORT_PHY_QCFG_RESP_AUTO_MODE_SPEED_MASK	   0x4UL
-	u8 auto_pause;
-	#define PORT_PHY_QCFG_RESP_AUTO_PAUSE_TX		    0x1UL
-	#define PORT_PHY_QCFG_RESP_AUTO_PAUSE_RX		    0x2UL
-	#define PORT_PHY_QCFG_RESP_AUTO_PAUSE_AUTONEG_PAUSE	    0x4UL
-	__le16 auto_link_speed;
-	#define PORT_PHY_QCFG_RESP_AUTO_LINK_SPEED_100MB	   0x1UL
-	#define PORT_PHY_QCFG_RESP_AUTO_LINK_SPEED_1GB		   0xaUL
-	#define PORT_PHY_QCFG_RESP_AUTO_LINK_SPEED_2GB		   0x14UL
-	#define PORT_PHY_QCFG_RESP_AUTO_LINK_SPEED_2_5GB	   0x19UL
-	#define PORT_PHY_QCFG_RESP_AUTO_LINK_SPEED_10GB	   0x64UL
-	#define PORT_PHY_QCFG_RESP_AUTO_LINK_SPEED_20GB	   0xc8UL
-	#define PORT_PHY_QCFG_RESP_AUTO_LINK_SPEED_25GB	   0xfaUL
-	#define PORT_PHY_QCFG_RESP_AUTO_LINK_SPEED_40GB	   0x190UL
-	#define PORT_PHY_QCFG_RESP_AUTO_LINK_SPEED_50GB	   0x1f4UL
-	#define PORT_PHY_QCFG_RESP_AUTO_LINK_SPEED_100GB	   0x3e8UL
-	#define PORT_PHY_QCFG_RESP_AUTO_LINK_SPEED_10MB	   0xffffUL
-	__le16 auto_link_speed_mask;
-	#define PORT_PHY_QCFG_RESP_AUTO_LINK_SPEED_MASK_100MBHD    0x1UL
-	#define PORT_PHY_QCFG_RESP_AUTO_LINK_SPEED_MASK_100MB      0x2UL
-	#define PORT_PHY_QCFG_RESP_AUTO_LINK_SPEED_MASK_1GBHD      0x4UL
-	#define PORT_PHY_QCFG_RESP_AUTO_LINK_SPEED_MASK_1GB	    0x8UL
-	#define PORT_PHY_QCFG_RESP_AUTO_LINK_SPEED_MASK_2GB	    0x10UL
-	#define PORT_PHY_QCFG_RESP_AUTO_LINK_SPEED_MASK_2_5GB      0x20UL
-	#define PORT_PHY_QCFG_RESP_AUTO_LINK_SPEED_MASK_10GB       0x40UL
-	#define PORT_PHY_QCFG_RESP_AUTO_LINK_SPEED_MASK_20GB       0x80UL
-	#define PORT_PHY_QCFG_RESP_AUTO_LINK_SPEED_MASK_25GB       0x100UL
-	#define PORT_PHY_QCFG_RESP_AUTO_LINK_SPEED_MASK_40GB       0x200UL
-	#define PORT_PHY_QCFG_RESP_AUTO_LINK_SPEED_MASK_50GB       0x400UL
-	#define PORT_PHY_QCFG_RESP_AUTO_LINK_SPEED_MASK_100GB      0x800UL
-	#define PORT_PHY_QCFG_RESP_AUTO_LINK_SPEED_MASK_10MBHD     0x1000UL
-	#define PORT_PHY_QCFG_RESP_AUTO_LINK_SPEED_MASK_10MB       0x2000UL
-	u8 wirespeed;
-	#define PORT_PHY_QCFG_RESP_WIRESPEED_OFF		   0x0UL
-	#define PORT_PHY_QCFG_RESP_WIRESPEED_ON		   0x1UL
-	u8 lpbk;
-	#define PORT_PHY_QCFG_RESP_LPBK_NONE			   0x0UL
-	#define PORT_PHY_QCFG_RESP_LPBK_LOCAL			   0x1UL
-	#define PORT_PHY_QCFG_RESP_LPBK_REMOTE			   0x2UL
-	u8 force_pause;
-	#define PORT_PHY_QCFG_RESP_FORCE_PAUSE_TX		    0x1UL
-	#define PORT_PHY_QCFG_RESP_FORCE_PAUSE_RX		    0x2UL
-	u8 module_status;
-	#define PORT_PHY_QCFG_RESP_MODULE_STATUS_NONE		   0x0UL
-	#define PORT_PHY_QCFG_RESP_MODULE_STATUS_DISABLETX	   0x1UL
-	#define PORT_PHY_QCFG_RESP_MODULE_STATUS_WARNINGMSG       0x2UL
-	#define PORT_PHY_QCFG_RESP_MODULE_STATUS_PWRDOWN	   0x3UL
-	#define PORT_PHY_QCFG_RESP_MODULE_STATUS_NOTINSERTED      0x4UL
-	#define PORT_PHY_QCFG_RESP_MODULE_STATUS_NOTAPPLICABLE    0xffUL
-	__le32 preemphasis;
-	u8 phy_maj;
-	u8 phy_min;
-	u8 phy_bld;
-	u8 phy_type;
-	#define PORT_PHY_QCFG_RESP_PHY_TYPE_UNKNOWN		   0x0UL
-	#define PORT_PHY_QCFG_RESP_PHY_TYPE_BASECR		   0x1UL
-	#define PORT_PHY_QCFG_RESP_PHY_TYPE_BASEKR4		   0x2UL
-	#define PORT_PHY_QCFG_RESP_PHY_TYPE_BASELR		   0x3UL
-	#define PORT_PHY_QCFG_RESP_PHY_TYPE_BASESR		   0x4UL
-	#define PORT_PHY_QCFG_RESP_PHY_TYPE_BASEKR2		   0x5UL
-	#define PORT_PHY_QCFG_RESP_PHY_TYPE_BASEKX		   0x6UL
-	#define PORT_PHY_QCFG_RESP_PHY_TYPE_BASEKR		   0x7UL
-	#define PORT_PHY_QCFG_RESP_PHY_TYPE_BASET		   0x8UL
-	#define PORT_PHY_QCFG_RESP_PHY_TYPE_BASETE		   0x9UL
-	#define PORT_PHY_QCFG_RESP_PHY_TYPE_SGMIIEXTPHY	   0xaUL
-	#define PORT_PHY_QCFG_RESP_PHY_TYPE_25G_BASECR_CA_L       0xbUL
-	#define PORT_PHY_QCFG_RESP_PHY_TYPE_25G_BASECR_CA_S       0xcUL
-	#define PORT_PHY_QCFG_RESP_PHY_TYPE_25G_BASECR_CA_N       0xdUL
-	#define PORT_PHY_QCFG_RESP_PHY_TYPE_25G_BASESR		   0xeUL
-	#define PORT_PHY_QCFG_RESP_PHY_TYPE_100G_BASECR4	   0xfUL
-	#define PORT_PHY_QCFG_RESP_PHY_TYPE_100G_BASESR4	   0x10UL
-	#define PORT_PHY_QCFG_RESP_PHY_TYPE_100G_BASELR4	   0x11UL
-	#define PORT_PHY_QCFG_RESP_PHY_TYPE_100G_BASEER4	   0x12UL
-	#define PORT_PHY_QCFG_RESP_PHY_TYPE_100G_BASESR10	   0x13UL
-	#define PORT_PHY_QCFG_RESP_PHY_TYPE_40G_BASECR4	   0x14UL
-	#define PORT_PHY_QCFG_RESP_PHY_TYPE_40G_BASESR4	   0x15UL
-	#define PORT_PHY_QCFG_RESP_PHY_TYPE_40G_BASELR4	   0x16UL
-	#define PORT_PHY_QCFG_RESP_PHY_TYPE_40G_BASEER4	   0x17UL
-	#define PORT_PHY_QCFG_RESP_PHY_TYPE_40G_ACTIVE_CABLE      0x18UL
-	u8 media_type;
-	#define PORT_PHY_QCFG_RESP_MEDIA_TYPE_UNKNOWN		   0x0UL
-	#define PORT_PHY_QCFG_RESP_MEDIA_TYPE_TP		   0x1UL
-	#define PORT_PHY_QCFG_RESP_MEDIA_TYPE_DAC		   0x2UL
-	#define PORT_PHY_QCFG_RESP_MEDIA_TYPE_FIBRE		   0x3UL
-	u8 xcvr_pkg_type;
-	#define PORT_PHY_QCFG_RESP_XCVR_PKG_TYPE_XCVR_INTERNAL    0x1UL
-	#define PORT_PHY_QCFG_RESP_XCVR_PKG_TYPE_XCVR_EXTERNAL    0x2UL
-	u8 eee_config_phy_addr;
-	#define PORT_PHY_QCFG_RESP_PHY_ADDR_MASK		    0x1fUL
-	#define PORT_PHY_QCFG_RESP_PHY_ADDR_SFT		    0
-	#define PORT_PHY_QCFG_RESP_EEE_CONFIG_EEE_ENABLED	    0x20UL
-	#define PORT_PHY_QCFG_RESP_EEE_CONFIG_EEE_ACTIVE	    0x40UL
-	#define PORT_PHY_QCFG_RESP_EEE_CONFIG_EEE_TX_LPI	    0x80UL
-	#define PORT_PHY_QCFG_RESP_EEE_CONFIG_MASK		    0xe0UL
-	#define PORT_PHY_QCFG_RESP_EEE_CONFIG_SFT		    5
-	u8 parallel_detect;
-	#define PORT_PHY_QCFG_RESP_PARALLEL_DETECT		    0x1UL
-	#define PORT_PHY_QCFG_RESP_RESERVED_MASK		    0xfeUL
-	#define PORT_PHY_QCFG_RESP_RESERVED_SFT		    1
-	__le16 link_partner_adv_speeds;
-	#define PORT_PHY_QCFG_RESP_LINK_PARTNER_ADV_SPEEDS_100MBHD 0x1UL
-	#define PORT_PHY_QCFG_RESP_LINK_PARTNER_ADV_SPEEDS_100MB   0x2UL
-	#define PORT_PHY_QCFG_RESP_LINK_PARTNER_ADV_SPEEDS_1GBHD   0x4UL
-	#define PORT_PHY_QCFG_RESP_LINK_PARTNER_ADV_SPEEDS_1GB     0x8UL
-	#define PORT_PHY_QCFG_RESP_LINK_PARTNER_ADV_SPEEDS_2GB     0x10UL
-	#define PORT_PHY_QCFG_RESP_LINK_PARTNER_ADV_SPEEDS_2_5GB   0x20UL
-	#define PORT_PHY_QCFG_RESP_LINK_PARTNER_ADV_SPEEDS_10GB    0x40UL
-	#define PORT_PHY_QCFG_RESP_LINK_PARTNER_ADV_SPEEDS_20GB    0x80UL
-	#define PORT_PHY_QCFG_RESP_LINK_PARTNER_ADV_SPEEDS_25GB    0x100UL
-	#define PORT_PHY_QCFG_RESP_LINK_PARTNER_ADV_SPEEDS_40GB    0x200UL
-	#define PORT_PHY_QCFG_RESP_LINK_PARTNER_ADV_SPEEDS_50GB    0x400UL
-	#define PORT_PHY_QCFG_RESP_LINK_PARTNER_ADV_SPEEDS_100GB   0x800UL
-	#define PORT_PHY_QCFG_RESP_LINK_PARTNER_ADV_SPEEDS_10MBHD  0x1000UL
-	#define PORT_PHY_QCFG_RESP_LINK_PARTNER_ADV_SPEEDS_10MB    0x2000UL
-	u8 link_partner_adv_auto_mode;
-	#define PORT_PHY_QCFG_RESP_LINK_PARTNER_ADV_AUTO_MODE_NONE 0x0UL
-	#define PORT_PHY_QCFG_RESP_LINK_PARTNER_ADV_AUTO_MODE_ALL_SPEEDS 0x1UL
-	#define PORT_PHY_QCFG_RESP_LINK_PARTNER_ADV_AUTO_MODE_ONE_SPEED 0x2UL
-	#define PORT_PHY_QCFG_RESP_LINK_PARTNER_ADV_AUTO_MODE_ONE_OR_BELOW 0x3UL
-	#define PORT_PHY_QCFG_RESP_LINK_PARTNER_ADV_AUTO_MODE_SPEED_MASK 0x4UL
-	u8 link_partner_adv_pause;
-	#define PORT_PHY_QCFG_RESP_LINK_PARTNER_ADV_PAUSE_TX       0x1UL
-	#define PORT_PHY_QCFG_RESP_LINK_PARTNER_ADV_PAUSE_RX       0x2UL
-	__le16 adv_eee_link_speed_mask;
-	#define PORT_PHY_QCFG_RESP_ADV_EEE_LINK_SPEED_MASK_RSVD1   0x1UL
-	#define PORT_PHY_QCFG_RESP_ADV_EEE_LINK_SPEED_MASK_100MB   0x2UL
-	#define PORT_PHY_QCFG_RESP_ADV_EEE_LINK_SPEED_MASK_RSVD2   0x4UL
-	#define PORT_PHY_QCFG_RESP_ADV_EEE_LINK_SPEED_MASK_1GB     0x8UL
-	#define PORT_PHY_QCFG_RESP_ADV_EEE_LINK_SPEED_MASK_RSVD3   0x10UL
-	#define PORT_PHY_QCFG_RESP_ADV_EEE_LINK_SPEED_MASK_RSVD4   0x20UL
-	#define PORT_PHY_QCFG_RESP_ADV_EEE_LINK_SPEED_MASK_10GB    0x40UL
-	__le16 link_partner_adv_eee_link_speed_mask;
-	#define PORT_PHY_QCFG_RESP_LINK_PARTNER_ADV_EEE_LINK_SPEED_MASK_RSVD1 0x1UL
-	#define PORT_PHY_QCFG_RESP_LINK_PARTNER_ADV_EEE_LINK_SPEED_MASK_100MB 0x2UL
-	#define PORT_PHY_QCFG_RESP_LINK_PARTNER_ADV_EEE_LINK_SPEED_MASK_RSVD2 0x4UL
-	#define PORT_PHY_QCFG_RESP_LINK_PARTNER_ADV_EEE_LINK_SPEED_MASK_1GB 0x8UL
-	#define PORT_PHY_QCFG_RESP_LINK_PARTNER_ADV_EEE_LINK_SPEED_MASK_RSVD3 0x10UL
-	#define PORT_PHY_QCFG_RESP_LINK_PARTNER_ADV_EEE_LINK_SPEED_MASK_RSVD4 0x20UL
-	#define PORT_PHY_QCFG_RESP_LINK_PARTNER_ADV_EEE_LINK_SPEED_MASK_10GB 0x40UL
-	__le32 xcvr_identifier_type_tx_lpi_timer;
-	#define PORT_PHY_QCFG_RESP_TX_LPI_TIMER_MASK		    0xffffffUL
-	#define PORT_PHY_QCFG_RESP_TX_LPI_TIMER_SFT		    0
-	#define PORT_PHY_QCFG_RESP_XCVR_IDENTIFIER_TYPE_MASK       0xff000000UL
-	#define PORT_PHY_QCFG_RESP_XCVR_IDENTIFIER_TYPE_SFT	    24
-	#define PORT_PHY_QCFG_RESP_XCVR_IDENTIFIER_TYPE_UNKNOWN   (0x0UL << 24)
-	#define PORT_PHY_QCFG_RESP_XCVR_IDENTIFIER_TYPE_SFP       (0x3UL << 24)
-	#define PORT_PHY_QCFG_RESP_XCVR_IDENTIFIER_TYPE_QSFP      (0xcUL << 24)
-	#define PORT_PHY_QCFG_RESP_XCVR_IDENTIFIER_TYPE_QSFPPLUS  (0xdUL << 24)
-	#define PORT_PHY_QCFG_RESP_XCVR_IDENTIFIER_TYPE_QSFP28    (0x11UL << 24)
-	__le16 fec_cfg;
-	#define PORT_PHY_QCFG_RESP_FEC_CFG_FEC_NONE_SUPPORTED      0x1UL
-	#define PORT_PHY_QCFG_RESP_FEC_CFG_FEC_AUTONEG_SUPPORTED   0x2UL
-	#define PORT_PHY_QCFG_RESP_FEC_CFG_FEC_AUTONEG_ENABLED     0x4UL
-	#define PORT_PHY_QCFG_RESP_FEC_CFG_FEC_CLAUSE74_SUPPORTED  0x8UL
-	#define PORT_PHY_QCFG_RESP_FEC_CFG_FEC_CLAUSE74_ENABLED    0x10UL
-	#define PORT_PHY_QCFG_RESP_FEC_CFG_FEC_CLAUSE91_SUPPORTED  0x20UL
-	#define PORT_PHY_QCFG_RESP_FEC_CFG_FEC_CLAUSE91_ENABLED    0x40UL
-	u8 unused_1;
-	u8 unused_2;
-	char phy_vendor_name[16];
-	char phy_vendor_partnumber[16];
-	__le32 unused_3;
-	u8 unused_4;
-	u8 unused_5;
-	u8 unused_6;
-	u8 valid;
-};
-
-/* hwrm_port_mac_cfg */
-/* Input (40 bytes) */
-struct hwrm_port_mac_cfg_input {
-	__le16 req_type;
-	__le16 cmpl_ring;
-	__le16 seq_id;
-	__le16 target_id;
-	__le64 resp_addr;
-	__le32 flags;
-	#define PORT_MAC_CFG_REQ_FLAGS_MATCH_LINK		    0x1UL
-	#define PORT_MAC_CFG_REQ_FLAGS_VLAN_PRI2COS_ENABLE	    0x2UL
-	#define PORT_MAC_CFG_REQ_FLAGS_TUNNEL_PRI2COS_ENABLE       0x4UL
-	#define PORT_MAC_CFG_REQ_FLAGS_IP_DSCP2COS_ENABLE	    0x8UL
-	#define PORT_MAC_CFG_REQ_FLAGS_PTP_RX_TS_CAPTURE_ENABLE    0x10UL
-	#define PORT_MAC_CFG_REQ_FLAGS_PTP_RX_TS_CAPTURE_DISABLE   0x20UL
-	#define PORT_MAC_CFG_REQ_FLAGS_PTP_TX_TS_CAPTURE_ENABLE    0x40UL
-	#define PORT_MAC_CFG_REQ_FLAGS_PTP_TX_TS_CAPTURE_DISABLE   0x80UL
-	#define PORT_MAC_CFG_REQ_FLAGS_OOB_WOL_ENABLE		    0x100UL
-	#define PORT_MAC_CFG_REQ_FLAGS_OOB_WOL_DISABLE		    0x200UL
-	#define PORT_MAC_CFG_REQ_FLAGS_VLAN_PRI2COS_DISABLE	    0x400UL
-	#define PORT_MAC_CFG_REQ_FLAGS_TUNNEL_PRI2COS_DISABLE      0x800UL
-	#define PORT_MAC_CFG_REQ_FLAGS_IP_DSCP2COS_DISABLE	    0x1000UL
-	__le32 enables;
-	#define PORT_MAC_CFG_REQ_ENABLES_IPG			    0x1UL
-	#define PORT_MAC_CFG_REQ_ENABLES_LPBK			    0x2UL
-	#define PORT_MAC_CFG_REQ_ENABLES_VLAN_PRI2COS_MAP_PRI      0x4UL
-	#define PORT_MAC_CFG_REQ_ENABLES_RESERVED1		    0x8UL
-	#define PORT_MAC_CFG_REQ_ENABLES_TUNNEL_PRI2COS_MAP_PRI    0x10UL
-	#define PORT_MAC_CFG_REQ_ENABLES_DSCP2COS_MAP_PRI	    0x20UL
-	#define PORT_MAC_CFG_REQ_ENABLES_RX_TS_CAPTURE_PTP_MSG_TYPE 0x40UL
-	#define PORT_MAC_CFG_REQ_ENABLES_TX_TS_CAPTURE_PTP_MSG_TYPE 0x80UL
-	#define PORT_MAC_CFG_REQ_ENABLES_COS_FIELD_CFG		    0x100UL
-	__le16 port_id;
-	u8 ipg;
-	u8 lpbk;
-	#define PORT_MAC_CFG_REQ_LPBK_NONE			   0x0UL
-	#define PORT_MAC_CFG_REQ_LPBK_LOCAL			   0x1UL
-	#define PORT_MAC_CFG_REQ_LPBK_REMOTE			   0x2UL
-	u8 vlan_pri2cos_map_pri;
-	u8 reserved1;
-	u8 tunnel_pri2cos_map_pri;
-	u8 dscp2pri_map_pri;
-	__le16 rx_ts_capture_ptp_msg_type;
-	__le16 tx_ts_capture_ptp_msg_type;
-	u8 cos_field_cfg;
-	#define PORT_MAC_CFG_REQ_COS_FIELD_CFG_RSVD1		    0x1UL
-	#define PORT_MAC_CFG_REQ_COS_FIELD_CFG_VLAN_PRI_SEL_MASK   0x6UL
-	#define PORT_MAC_CFG_REQ_COS_FIELD_CFG_VLAN_PRI_SEL_SFT    1
-	#define PORT_MAC_CFG_REQ_COS_FIELD_CFG_VLAN_PRI_SEL_INNERMOST (0x0UL << 1)
-	#define PORT_MAC_CFG_REQ_COS_FIELD_CFG_VLAN_PRI_SEL_OUTER (0x1UL << 1)
-	#define PORT_MAC_CFG_REQ_COS_FIELD_CFG_VLAN_PRI_SEL_OUTERMOST (0x2UL << 1)
-	#define PORT_MAC_CFG_REQ_COS_FIELD_CFG_VLAN_PRI_SEL_UNSPECIFIED (0x3UL << 1)
-	#define PORT_MAC_CFG_REQ_COS_FIELD_CFG_VLAN_PRI_SEL_LAST    PORT_MAC_CFG_REQ_COS_FIELD_CFG_VLAN_PRI_SEL_UNSPECIFIED
-	#define PORT_MAC_CFG_REQ_COS_FIELD_CFG_T_VLAN_PRI_SEL_MASK 0x18UL
-	#define PORT_MAC_CFG_REQ_COS_FIELD_CFG_T_VLAN_PRI_SEL_SFT  3
-	#define PORT_MAC_CFG_REQ_COS_FIELD_CFG_T_VLAN_PRI_SEL_INNERMOST (0x0UL << 3)
-	#define PORT_MAC_CFG_REQ_COS_FIELD_CFG_T_VLAN_PRI_SEL_OUTER (0x1UL << 3)
-	#define PORT_MAC_CFG_REQ_COS_FIELD_CFG_T_VLAN_PRI_SEL_OUTERMOST (0x2UL << 3)
-	#define PORT_MAC_CFG_REQ_COS_FIELD_CFG_T_VLAN_PRI_SEL_UNSPECIFIED (0x3UL << 3)
-	#define PORT_MAC_CFG_REQ_COS_FIELD_CFG_T_VLAN_PRI_SEL_LAST    PORT_MAC_CFG_REQ_COS_FIELD_CFG_T_VLAN_PRI_SEL_UNSPECIFIED
-	#define PORT_MAC_CFG_REQ_COS_FIELD_CFG_DEFAULT_COS_MASK    0xe0UL
-	#define PORT_MAC_CFG_REQ_COS_FIELD_CFG_DEFAULT_COS_SFT     5
-	u8 unused_0[3];
-};
-
-/* Output (16 bytes) */
-struct hwrm_port_mac_cfg_output {
-	__le16 error_code;
-	__le16 req_type;
-	__le16 seq_id;
-	__le16 resp_len;
-	__le16 mru;
-	__le16 mtu;
-	u8 ipg;
-	u8 lpbk;
-	#define PORT_MAC_CFG_RESP_LPBK_NONE			   0x0UL
-	#define PORT_MAC_CFG_RESP_LPBK_LOCAL			   0x1UL
-	#define PORT_MAC_CFG_RESP_LPBK_REMOTE			   0x2UL
-	u8 unused_0;
-	u8 valid;
-};
-
-/* hwrm_port_qstats */
-/* Input (40 bytes) */
-struct hwrm_port_qstats_input {
-	__le16 req_type;
-	__le16 cmpl_ring;
-	__le16 seq_id;
-	__le16 target_id;
-	__le64 resp_addr;
-	__le16 port_id;
-	u8 unused_0;
-	u8 unused_1;
-	u8 unused_2[3];
-	u8 unused_3;
-	__le64 tx_stat_host_addr;
-	__le64 rx_stat_host_addr;
-};
-
-/* Output (16 bytes) */
-struct hwrm_port_qstats_output {
-	__le16 error_code;
-	__le16 req_type;
-	__le16 seq_id;
-	__le16 resp_len;
-	__le16 tx_stat_size;
-	__le16 rx_stat_size;
-	u8 unused_0;
-	u8 unused_1;
-	u8 unused_2;
-	u8 valid;
-};
-
-/* hwrm_port_lpbk_qstats */
-/* Input (16 bytes) */
-struct hwrm_port_lpbk_qstats_input {
-	__le16 req_type;
-	__le16 cmpl_ring;
-	__le16 seq_id;
-	__le16 target_id;
-	__le64 resp_addr;
-};
-
-/* Output (96 bytes) */
-struct hwrm_port_lpbk_qstats_output {
-	__le16 error_code;
-	__le16 req_type;
-	__le16 seq_id;
-	__le16 resp_len;
-	__le64 lpbk_ucast_frames;
-	__le64 lpbk_mcast_frames;
-	__le64 lpbk_bcast_frames;
-	__le64 lpbk_ucast_bytes;
-	__le64 lpbk_mcast_bytes;
-	__le64 lpbk_bcast_bytes;
-	__le64 tx_stat_discard;
-	__le64 tx_stat_error;
-	__le64 rx_stat_discard;
-	__le64 rx_stat_error;
-	__le32 unused_0;
-	u8 unused_1;
-	u8 unused_2;
-	u8 unused_3;
-	u8 valid;
-};
-
-/* hwrm_port_clr_stats */
-/* Input (24 bytes) */
-struct hwrm_port_clr_stats_input {
-	__le16 req_type;
-	__le16 cmpl_ring;
-	__le16 seq_id;
-	__le16 target_id;
-	__le64 resp_addr;
-	__le16 port_id;
-	__le16 unused_0[3];
-};
-
-/* Output (16 bytes) */
-struct hwrm_port_clr_stats_output {
-	__le16 error_code;
-	__le16 req_type;
-	__le16 seq_id;
-	__le16 resp_len;
-	__le32 unused_0;
-	u8 unused_1;
-	u8 unused_2;
-	u8 unused_3;
-	u8 valid;
-};
-
-/* hwrm_port_lpbk_clr_stats */
-/* Input (16 bytes) */
-struct hwrm_port_lpbk_clr_stats_input {
-	__le16 req_type;
-	__le16 cmpl_ring;
-	__le16 seq_id;
-	__le16 target_id;
-	__le64 resp_addr;
-};
-
-/* Output (16 bytes) */
-struct hwrm_port_lpbk_clr_stats_output {
-	__le16 error_code;
-	__le16 req_type;
-	__le16 seq_id;
-	__le16 resp_len;
-	__le32 unused_0;
-	u8 unused_1;
-	u8 unused_2;
-	u8 unused_3;
-	u8 valid;
-};
-
-/* hwrm_port_phy_qcaps */
-/* Input (24 bytes) */
-struct hwrm_port_phy_qcaps_input {
-	__le16 req_type;
-	__le16 cmpl_ring;
-	__le16 seq_id;
-	__le16 target_id;
-	__le64 resp_addr;
-	__le16 port_id;
-	__le16 unused_0[3];
-};
-
-/* Output (24 bytes) */
-struct hwrm_port_phy_qcaps_output {
-	__le16 error_code;
-	__le16 req_type;
-	__le16 seq_id;
-	__le16 resp_len;
-	u8 eee_supported;
-	#define PORT_PHY_QCAPS_RESP_EEE_SUPPORTED		    0x1UL
-	#define PORT_PHY_QCAPS_RESP_RSVD1_MASK			    0xfeUL
-	#define PORT_PHY_QCAPS_RESP_RSVD1_SFT			    1
-	u8 unused_0;
-	__le16 supported_speeds_force_mode;
-	#define PORT_PHY_QCAPS_RESP_SUPPORTED_SPEEDS_FORCE_MODE_100MBHD 0x1UL
-	#define PORT_PHY_QCAPS_RESP_SUPPORTED_SPEEDS_FORCE_MODE_100MB 0x2UL
-	#define PORT_PHY_QCAPS_RESP_SUPPORTED_SPEEDS_FORCE_MODE_1GBHD 0x4UL
-	#define PORT_PHY_QCAPS_RESP_SUPPORTED_SPEEDS_FORCE_MODE_1GB 0x8UL
-	#define PORT_PHY_QCAPS_RESP_SUPPORTED_SPEEDS_FORCE_MODE_2GB 0x10UL
-	#define PORT_PHY_QCAPS_RESP_SUPPORTED_SPEEDS_FORCE_MODE_2_5GB 0x20UL
-	#define PORT_PHY_QCAPS_RESP_SUPPORTED_SPEEDS_FORCE_MODE_10GB 0x40UL
-	#define PORT_PHY_QCAPS_RESP_SUPPORTED_SPEEDS_FORCE_MODE_20GB 0x80UL
-	#define PORT_PHY_QCAPS_RESP_SUPPORTED_SPEEDS_FORCE_MODE_25GB 0x100UL
-	#define PORT_PHY_QCAPS_RESP_SUPPORTED_SPEEDS_FORCE_MODE_40GB 0x200UL
-	#define PORT_PHY_QCAPS_RESP_SUPPORTED_SPEEDS_FORCE_MODE_50GB 0x400UL
-	#define PORT_PHY_QCAPS_RESP_SUPPORTED_SPEEDS_FORCE_MODE_100GB 0x800UL
-	#define PORT_PHY_QCAPS_RESP_SUPPORTED_SPEEDS_FORCE_MODE_10MBHD 0x1000UL
-	#define PORT_PHY_QCAPS_RESP_SUPPORTED_SPEEDS_FORCE_MODE_10MB 0x2000UL
-	__le16 supported_speeds_auto_mode;
-	#define PORT_PHY_QCAPS_RESP_SUPPORTED_SPEEDS_AUTO_MODE_100MBHD 0x1UL
-	#define PORT_PHY_QCAPS_RESP_SUPPORTED_SPEEDS_AUTO_MODE_100MB 0x2UL
-	#define PORT_PHY_QCAPS_RESP_SUPPORTED_SPEEDS_AUTO_MODE_1GBHD 0x4UL
-	#define PORT_PHY_QCAPS_RESP_SUPPORTED_SPEEDS_AUTO_MODE_1GB 0x8UL
-	#define PORT_PHY_QCAPS_RESP_SUPPORTED_SPEEDS_AUTO_MODE_2GB 0x10UL
-	#define PORT_PHY_QCAPS_RESP_SUPPORTED_SPEEDS_AUTO_MODE_2_5GB 0x20UL
-	#define PORT_PHY_QCAPS_RESP_SUPPORTED_SPEEDS_AUTO_MODE_10GB 0x40UL
-	#define PORT_PHY_QCAPS_RESP_SUPPORTED_SPEEDS_AUTO_MODE_20GB 0x80UL
-	#define PORT_PHY_QCAPS_RESP_SUPPORTED_SPEEDS_AUTO_MODE_25GB 0x100UL
-	#define PORT_PHY_QCAPS_RESP_SUPPORTED_SPEEDS_AUTO_MODE_40GB 0x200UL
-	#define PORT_PHY_QCAPS_RESP_SUPPORTED_SPEEDS_AUTO_MODE_50GB 0x400UL
-	#define PORT_PHY_QCAPS_RESP_SUPPORTED_SPEEDS_AUTO_MODE_100GB 0x800UL
-	#define PORT_PHY_QCAPS_RESP_SUPPORTED_SPEEDS_AUTO_MODE_10MBHD 0x1000UL
-	#define PORT_PHY_QCAPS_RESP_SUPPORTED_SPEEDS_AUTO_MODE_10MB 0x2000UL
-	__le16 supported_speeds_eee_mode;
-	#define PORT_PHY_QCAPS_RESP_SUPPORTED_SPEEDS_EEE_MODE_RSVD1 0x1UL
-	#define PORT_PHY_QCAPS_RESP_SUPPORTED_SPEEDS_EEE_MODE_100MB 0x2UL
-	#define PORT_PHY_QCAPS_RESP_SUPPORTED_SPEEDS_EEE_MODE_RSVD2 0x4UL
-	#define PORT_PHY_QCAPS_RESP_SUPPORTED_SPEEDS_EEE_MODE_1GB  0x8UL
-	#define PORT_PHY_QCAPS_RESP_SUPPORTED_SPEEDS_EEE_MODE_RSVD3 0x10UL
-	#define PORT_PHY_QCAPS_RESP_SUPPORTED_SPEEDS_EEE_MODE_RSVD4 0x20UL
-	#define PORT_PHY_QCAPS_RESP_SUPPORTED_SPEEDS_EEE_MODE_10GB 0x40UL
-	__le32 tx_lpi_timer_low;
-	#define PORT_PHY_QCAPS_RESP_TX_LPI_TIMER_LOW_MASK	    0xffffffUL
-	#define PORT_PHY_QCAPS_RESP_TX_LPI_TIMER_LOW_SFT	    0
-	#define PORT_PHY_QCAPS_RESP_RSVD2_MASK			    0xff000000UL
-	#define PORT_PHY_QCAPS_RESP_RSVD2_SFT			    24
-	__le32 valid_tx_lpi_timer_high;
-	#define PORT_PHY_QCAPS_RESP_TX_LPI_TIMER_HIGH_MASK	    0xffffffUL
-	#define PORT_PHY_QCAPS_RESP_TX_LPI_TIMER_HIGH_SFT	    0
-	#define PORT_PHY_QCAPS_RESP_VALID_MASK			    0xff000000UL
-	#define PORT_PHY_QCAPS_RESP_VALID_SFT			    24
-};
-
-/* hwrm_port_phy_i2c_read */
-/* Input (40 bytes) */
-struct hwrm_port_phy_i2c_read_input {
-	__le16 req_type;
-	__le16 cmpl_ring;
-	__le16 seq_id;
-	__le16 target_id;
-	__le64 resp_addr;
-	__le32 flags;
-	__le32 enables;
-	#define PORT_PHY_I2C_READ_REQ_ENABLES_PAGE_OFFSET	    0x1UL
-	__le16 port_id;
-	u8 i2c_slave_addr;
-	u8 unused_0;
-	__le16 page_number;
-	__le16 page_offset;
-	u8 data_length;
-	u8 unused_1[7];
-};
-
-/* Output (80 bytes) */
-struct hwrm_port_phy_i2c_read_output {
-	__le16 error_code;
-	__le16 req_type;
-	__le16 seq_id;
-	__le16 resp_len;
-	__le32 data[16];
-	__le32 unused_0;
-	u8 unused_1;
-	u8 unused_2;
-	u8 unused_3;
-	u8 valid;
-};
-
-/* hwrm_port_led_cfg */
-/* Input (64 bytes) */
-struct hwrm_port_led_cfg_input {
-	__le16 req_type;
-	__le16 cmpl_ring;
-	__le16 seq_id;
-	__le16 target_id;
-	__le64 resp_addr;
-	__le32 enables;
-	#define PORT_LED_CFG_REQ_ENABLES_LED0_ID		    0x1UL
-	#define PORT_LED_CFG_REQ_ENABLES_LED0_STATE		    0x2UL
-	#define PORT_LED_CFG_REQ_ENABLES_LED0_COLOR		    0x4UL
-	#define PORT_LED_CFG_REQ_ENABLES_LED0_BLINK_ON		    0x8UL
-	#define PORT_LED_CFG_REQ_ENABLES_LED0_BLINK_OFF	    0x10UL
-	#define PORT_LED_CFG_REQ_ENABLES_LED0_GROUP_ID		    0x20UL
-	#define PORT_LED_CFG_REQ_ENABLES_LED1_ID		    0x40UL
-	#define PORT_LED_CFG_REQ_ENABLES_LED1_STATE		    0x80UL
-	#define PORT_LED_CFG_REQ_ENABLES_LED1_COLOR		    0x100UL
-	#define PORT_LED_CFG_REQ_ENABLES_LED1_BLINK_ON		    0x200UL
-	#define PORT_LED_CFG_REQ_ENABLES_LED1_BLINK_OFF	    0x400UL
-	#define PORT_LED_CFG_REQ_ENABLES_LED1_GROUP_ID		    0x800UL
-	#define PORT_LED_CFG_REQ_ENABLES_LED2_ID		    0x1000UL
-	#define PORT_LED_CFG_REQ_ENABLES_LED2_STATE		    0x2000UL
-	#define PORT_LED_CFG_REQ_ENABLES_LED2_COLOR		    0x4000UL
-	#define PORT_LED_CFG_REQ_ENABLES_LED2_BLINK_ON		    0x8000UL
-	#define PORT_LED_CFG_REQ_ENABLES_LED2_BLINK_OFF	    0x10000UL
-	#define PORT_LED_CFG_REQ_ENABLES_LED2_GROUP_ID		    0x20000UL
-	#define PORT_LED_CFG_REQ_ENABLES_LED3_ID		    0x40000UL
-	#define PORT_LED_CFG_REQ_ENABLES_LED3_STATE		    0x80000UL
-	#define PORT_LED_CFG_REQ_ENABLES_LED3_COLOR		    0x100000UL
-	#define PORT_LED_CFG_REQ_ENABLES_LED3_BLINK_ON		    0x200000UL
-	#define PORT_LED_CFG_REQ_ENABLES_LED3_BLINK_OFF	    0x400000UL
-	#define PORT_LED_CFG_REQ_ENABLES_LED3_GROUP_ID		    0x800000UL
-	__le16 port_id;
-	u8 num_leds;
-	u8 rsvd;
-	u8 led0_id;
-	u8 led0_state;
-	#define PORT_LED_CFG_REQ_LED0_STATE_DEFAULT		   0x0UL
-	#define PORT_LED_CFG_REQ_LED0_STATE_OFF		   0x1UL
-	#define PORT_LED_CFG_REQ_LED0_STATE_ON			   0x2UL
-	#define PORT_LED_CFG_REQ_LED0_STATE_BLINK		   0x3UL
-	#define PORT_LED_CFG_REQ_LED0_STATE_BLINKALT		   0x4UL
-	u8 led0_color;
-	#define PORT_LED_CFG_REQ_LED0_COLOR_DEFAULT		   0x0UL
-	#define PORT_LED_CFG_REQ_LED0_COLOR_AMBER		   0x1UL
-	#define PORT_LED_CFG_REQ_LED0_COLOR_GREEN		   0x2UL
-	#define PORT_LED_CFG_REQ_LED0_COLOR_GREENAMBER		   0x3UL
-	u8 unused_0;
-	__le16 led0_blink_on;
-	__le16 led0_blink_off;
-	u8 led0_group_id;
-	u8 rsvd0;
-	u8 led1_id;
-	u8 led1_state;
-	#define PORT_LED_CFG_REQ_LED1_STATE_DEFAULT		   0x0UL
-	#define PORT_LED_CFG_REQ_LED1_STATE_OFF		   0x1UL
-	#define PORT_LED_CFG_REQ_LED1_STATE_ON			   0x2UL
-	#define PORT_LED_CFG_REQ_LED1_STATE_BLINK		   0x3UL
-	#define PORT_LED_CFG_REQ_LED1_STATE_BLINKALT		   0x4UL
-	u8 led1_color;
-	#define PORT_LED_CFG_REQ_LED1_COLOR_DEFAULT		   0x0UL
-	#define PORT_LED_CFG_REQ_LED1_COLOR_AMBER		   0x1UL
-	#define PORT_LED_CFG_REQ_LED1_COLOR_GREEN		   0x2UL
-	#define PORT_LED_CFG_REQ_LED1_COLOR_GREENAMBER		   0x3UL
-	u8 unused_1;
-	__le16 led1_blink_on;
-	__le16 led1_blink_off;
-	u8 led1_group_id;
-	u8 rsvd1;
-	u8 led2_id;
-	u8 led2_state;
-	#define PORT_LED_CFG_REQ_LED2_STATE_DEFAULT		   0x0UL
-	#define PORT_LED_CFG_REQ_LED2_STATE_OFF		   0x1UL
-	#define PORT_LED_CFG_REQ_LED2_STATE_ON			   0x2UL
-	#define PORT_LED_CFG_REQ_LED2_STATE_BLINK		   0x3UL
-	#define PORT_LED_CFG_REQ_LED2_STATE_BLINKALT		   0x4UL
-	u8 led2_color;
-	#define PORT_LED_CFG_REQ_LED2_COLOR_DEFAULT		   0x0UL
-	#define PORT_LED_CFG_REQ_LED2_COLOR_AMBER		   0x1UL
-	#define PORT_LED_CFG_REQ_LED2_COLOR_GREEN		   0x2UL
-	#define PORT_LED_CFG_REQ_LED2_COLOR_GREENAMBER		   0x3UL
-	u8 unused_2;
-	__le16 led2_blink_on;
-	__le16 led2_blink_off;
-	u8 led2_group_id;
-	u8 rsvd2;
-	u8 led3_id;
-	u8 led3_state;
-	#define PORT_LED_CFG_REQ_LED3_STATE_DEFAULT		   0x0UL
-	#define PORT_LED_CFG_REQ_LED3_STATE_OFF		   0x1UL
-	#define PORT_LED_CFG_REQ_LED3_STATE_ON			   0x2UL
-	#define PORT_LED_CFG_REQ_LED3_STATE_BLINK		   0x3UL
-	#define PORT_LED_CFG_REQ_LED3_STATE_BLINKALT		   0x4UL
-	u8 led3_color;
-	#define PORT_LED_CFG_REQ_LED3_COLOR_DEFAULT		   0x0UL
-	#define PORT_LED_CFG_REQ_LED3_COLOR_AMBER		   0x1UL
-	#define PORT_LED_CFG_REQ_LED3_COLOR_GREEN		   0x2UL
-	#define PORT_LED_CFG_REQ_LED3_COLOR_GREENAMBER		   0x3UL
-	u8 unused_3;
-	__le16 led3_blink_on;
-	__le16 led3_blink_off;
-	u8 led3_group_id;
-	u8 rsvd3;
-};
-
-/* Output (16 bytes) */
-struct hwrm_port_led_cfg_output {
-	__le16 error_code;
-	__le16 req_type;
-	__le16 seq_id;
-	__le16 resp_len;
-	__le32 unused_0;
-	u8 unused_1;
-	u8 unused_2;
-	u8 unused_3;
-	u8 valid;
-};
-
-/* hwrm_port_led_qcaps */
-/* Input (24 bytes) */
-struct hwrm_port_led_qcaps_input {
-	__le16 req_type;
-	__le16 cmpl_ring;
-	__le16 seq_id;
-	__le16 target_id;
-	__le64 resp_addr;
-	__le16 port_id;
-	__le16 unused_0[3];
-};
-
-/* Output (48 bytes) */
-struct hwrm_port_led_qcaps_output {
-	__le16 error_code;
-	__le16 req_type;
-	__le16 seq_id;
-	__le16 resp_len;
-	u8 num_leds;
-	u8 unused_0[3];
-	u8 led0_id;
-	u8 led0_type;
-	#define PORT_LED_QCAPS_RESP_LED0_TYPE_SPEED		   0x0UL
-	#define PORT_LED_QCAPS_RESP_LED0_TYPE_ACTIVITY		   0x1UL
-	#define PORT_LED_QCAPS_RESP_LED0_TYPE_INVALID		   0xffUL
-	u8 led0_group_id;
-	u8 unused_1;
-	__le16 led0_state_caps;
-	#define PORT_LED_QCAPS_RESP_LED0_STATE_CAPS_ENABLED	    0x1UL
-	#define PORT_LED_QCAPS_RESP_LED0_STATE_CAPS_OFF_SUPPORTED  0x2UL
-	#define PORT_LED_QCAPS_RESP_LED0_STATE_CAPS_ON_SUPPORTED   0x4UL
-	#define PORT_LED_QCAPS_RESP_LED0_STATE_CAPS_BLINK_SUPPORTED 0x8UL
-	#define PORT_LED_QCAPS_RESP_LED0_STATE_CAPS_BLINK_ALT_SUPPORTED 0x10UL
-	__le16 led0_color_caps;
-	#define PORT_LED_QCAPS_RESP_LED0_COLOR_CAPS_RSVD	    0x1UL
-	#define PORT_LED_QCAPS_RESP_LED0_COLOR_CAPS_AMBER_SUPPORTED 0x2UL
-	#define PORT_LED_QCAPS_RESP_LED0_COLOR_CAPS_GREEN_SUPPORTED 0x4UL
-	u8 led1_id;
-	u8 led1_type;
-	#define PORT_LED_QCAPS_RESP_LED1_TYPE_SPEED		   0x0UL
-	#define PORT_LED_QCAPS_RESP_LED1_TYPE_ACTIVITY		   0x1UL
-	#define PORT_LED_QCAPS_RESP_LED1_TYPE_INVALID		   0xffUL
-	u8 led1_group_id;
-	u8 unused_2;
-	__le16 led1_state_caps;
-	#define PORT_LED_QCAPS_RESP_LED1_STATE_CAPS_ENABLED	    0x1UL
-	#define PORT_LED_QCAPS_RESP_LED1_STATE_CAPS_OFF_SUPPORTED  0x2UL
-	#define PORT_LED_QCAPS_RESP_LED1_STATE_CAPS_ON_SUPPORTED   0x4UL
-	#define PORT_LED_QCAPS_RESP_LED1_STATE_CAPS_BLINK_SUPPORTED 0x8UL
-	#define PORT_LED_QCAPS_RESP_LED1_STATE_CAPS_BLINK_ALT_SUPPORTED 0x10UL
-	__le16 led1_color_caps;
-	#define PORT_LED_QCAPS_RESP_LED1_COLOR_CAPS_RSVD	    0x1UL
-	#define PORT_LED_QCAPS_RESP_LED1_COLOR_CAPS_AMBER_SUPPORTED 0x2UL
-	#define PORT_LED_QCAPS_RESP_LED1_COLOR_CAPS_GREEN_SUPPORTED 0x4UL
-	u8 led2_id;
-	u8 led2_type;
-	#define PORT_LED_QCAPS_RESP_LED2_TYPE_SPEED		   0x0UL
-	#define PORT_LED_QCAPS_RESP_LED2_TYPE_ACTIVITY		   0x1UL
-	#define PORT_LED_QCAPS_RESP_LED2_TYPE_INVALID		   0xffUL
-	u8 led2_group_id;
-	u8 unused_3;
-	__le16 led2_state_caps;
-	#define PORT_LED_QCAPS_RESP_LED2_STATE_CAPS_ENABLED	    0x1UL
-	#define PORT_LED_QCAPS_RESP_LED2_STATE_CAPS_OFF_SUPPORTED  0x2UL
-	#define PORT_LED_QCAPS_RESP_LED2_STATE_CAPS_ON_SUPPORTED   0x4UL
-	#define PORT_LED_QCAPS_RESP_LED2_STATE_CAPS_BLINK_SUPPORTED 0x8UL
-	#define PORT_LED_QCAPS_RESP_LED2_STATE_CAPS_BLINK_ALT_SUPPORTED 0x10UL
-	__le16 led2_color_caps;
-	#define PORT_LED_QCAPS_RESP_LED2_COLOR_CAPS_RSVD	    0x1UL
-	#define PORT_LED_QCAPS_RESP_LED2_COLOR_CAPS_AMBER_SUPPORTED 0x2UL
-	#define PORT_LED_QCAPS_RESP_LED2_COLOR_CAPS_GREEN_SUPPORTED 0x4UL
-	u8 led3_id;
-	u8 led3_type;
-	#define PORT_LED_QCAPS_RESP_LED3_TYPE_SPEED		   0x0UL
-	#define PORT_LED_QCAPS_RESP_LED3_TYPE_ACTIVITY		   0x1UL
-	#define PORT_LED_QCAPS_RESP_LED3_TYPE_INVALID		   0xffUL
-	u8 led3_group_id;
-	u8 unused_4;
-	__le16 led3_state_caps;
-	#define PORT_LED_QCAPS_RESP_LED3_STATE_CAPS_ENABLED	    0x1UL
-	#define PORT_LED_QCAPS_RESP_LED3_STATE_CAPS_OFF_SUPPORTED  0x2UL
-	#define PORT_LED_QCAPS_RESP_LED3_STATE_CAPS_ON_SUPPORTED   0x4UL
-	#define PORT_LED_QCAPS_RESP_LED3_STATE_CAPS_BLINK_SUPPORTED 0x8UL
-	#define PORT_LED_QCAPS_RESP_LED3_STATE_CAPS_BLINK_ALT_SUPPORTED 0x10UL
-	__le16 led3_color_caps;
-	#define PORT_LED_QCAPS_RESP_LED3_COLOR_CAPS_RSVD	    0x1UL
-	#define PORT_LED_QCAPS_RESP_LED3_COLOR_CAPS_AMBER_SUPPORTED 0x2UL
-	#define PORT_LED_QCAPS_RESP_LED3_COLOR_CAPS_GREEN_SUPPORTED 0x4UL
-	u8 unused_5;
-	u8 unused_6;
-	u8 unused_7;
-	u8 valid;
-};
-
-/* hwrm_queue_qportcfg */
-/* Input (24 bytes) */
-struct hwrm_queue_qportcfg_input {
-	__le16 req_type;
-	__le16 cmpl_ring;
-	__le16 seq_id;
-	__le16 target_id;
-	__le64 resp_addr;
-	__le32 flags;
-	#define QUEUE_QPORTCFG_REQ_FLAGS_PATH			    0x1UL
-	#define QUEUE_QPORTCFG_REQ_FLAGS_PATH_TX		   0x0UL
-	#define QUEUE_QPORTCFG_REQ_FLAGS_PATH_RX		   0x1UL
-	#define QUEUE_QPORTCFG_REQ_FLAGS_PATH_LAST    QUEUE_QPORTCFG_REQ_FLAGS_PATH_RX
-	__le16 port_id;
-	__le16 unused_0;
-};
-
-/* Output (32 bytes) */
-struct hwrm_queue_qportcfg_output {
-	__le16 error_code;
-	__le16 req_type;
-	__le16 seq_id;
-	__le16 resp_len;
-	u8 max_configurable_queues;
-	u8 max_configurable_lossless_queues;
-	u8 queue_cfg_allowed;
-	u8 queue_cfg_info;
-	#define QUEUE_QPORTCFG_RESP_QUEUE_CFG_INFO_ASYM_CFG	    0x1UL
-	u8 queue_pfcenable_cfg_allowed;
-	u8 queue_pri2cos_cfg_allowed;
-	u8 queue_cos2bw_cfg_allowed;
-	u8 queue_id0;
-	u8 queue_id0_service_profile;
-	#define QUEUE_QPORTCFG_RESP_QUEUE_ID0_SERVICE_PROFILE_LOSSY 0x0UL
-	#define QUEUE_QPORTCFG_RESP_QUEUE_ID0_SERVICE_PROFILE_LOSSLESS 0x1UL
-	#define QUEUE_QPORTCFG_RESP_QUEUE_ID0_SERVICE_PROFILE_UNKNOWN 0xffUL
-	u8 queue_id1;
-	u8 queue_id1_service_profile;
-	#define QUEUE_QPORTCFG_RESP_QUEUE_ID1_SERVICE_PROFILE_LOSSY 0x0UL
-	#define QUEUE_QPORTCFG_RESP_QUEUE_ID1_SERVICE_PROFILE_LOSSLESS 0x1UL
-	#define QUEUE_QPORTCFG_RESP_QUEUE_ID1_SERVICE_PROFILE_UNKNOWN 0xffUL
-	u8 queue_id2;
-	u8 queue_id2_service_profile;
-	#define QUEUE_QPORTCFG_RESP_QUEUE_ID2_SERVICE_PROFILE_LOSSY 0x0UL
-	#define QUEUE_QPORTCFG_RESP_QUEUE_ID2_SERVICE_PROFILE_LOSSLESS 0x1UL
-	#define QUEUE_QPORTCFG_RESP_QUEUE_ID2_SERVICE_PROFILE_UNKNOWN 0xffUL
-	u8 queue_id3;
-	u8 queue_id3_service_profile;
-	#define QUEUE_QPORTCFG_RESP_QUEUE_ID3_SERVICE_PROFILE_LOSSY 0x0UL
-	#define QUEUE_QPORTCFG_RESP_QUEUE_ID3_SERVICE_PROFILE_LOSSLESS 0x1UL
-	#define QUEUE_QPORTCFG_RESP_QUEUE_ID3_SERVICE_PROFILE_UNKNOWN 0xffUL
-	u8 queue_id4;
-	u8 queue_id4_service_profile;
-	#define QUEUE_QPORTCFG_RESP_QUEUE_ID4_SERVICE_PROFILE_LOSSY 0x0UL
-	#define QUEUE_QPORTCFG_RESP_QUEUE_ID4_SERVICE_PROFILE_LOSSLESS 0x1UL
-	#define QUEUE_QPORTCFG_RESP_QUEUE_ID4_SERVICE_PROFILE_UNKNOWN 0xffUL
-	u8 queue_id5;
-	u8 queue_id5_service_profile;
-	#define QUEUE_QPORTCFG_RESP_QUEUE_ID5_SERVICE_PROFILE_LOSSY 0x0UL
-	#define QUEUE_QPORTCFG_RESP_QUEUE_ID5_SERVICE_PROFILE_LOSSLESS 0x1UL
-	#define QUEUE_QPORTCFG_RESP_QUEUE_ID5_SERVICE_PROFILE_UNKNOWN 0xffUL
-	u8 queue_id6;
-	u8 queue_id6_service_profile;
-	#define QUEUE_QPORTCFG_RESP_QUEUE_ID6_SERVICE_PROFILE_LOSSY 0x0UL
-	#define QUEUE_QPORTCFG_RESP_QUEUE_ID6_SERVICE_PROFILE_LOSSLESS 0x1UL
-	#define QUEUE_QPORTCFG_RESP_QUEUE_ID6_SERVICE_PROFILE_UNKNOWN 0xffUL
-	u8 queue_id7;
-	u8 queue_id7_service_profile;
-	#define QUEUE_QPORTCFG_RESP_QUEUE_ID7_SERVICE_PROFILE_LOSSY 0x0UL
-	#define QUEUE_QPORTCFG_RESP_QUEUE_ID7_SERVICE_PROFILE_LOSSLESS 0x1UL
-	#define QUEUE_QPORTCFG_RESP_QUEUE_ID7_SERVICE_PROFILE_UNKNOWN 0xffUL
-	u8 valid;
-};
-
-/* hwrm_queue_cfg */
-/* Input (40 bytes) */
-struct hwrm_queue_cfg_input {
-	__le16 req_type;
-	__le16 cmpl_ring;
-	__le16 seq_id;
-	__le16 target_id;
-	__le64 resp_addr;
-	__le32 flags;
-	#define QUEUE_CFG_REQ_FLAGS_PATH_MASK			    0x3UL
-	#define QUEUE_CFG_REQ_FLAGS_PATH_SFT			    0
-	#define QUEUE_CFG_REQ_FLAGS_PATH_TX			   0x0UL
-	#define QUEUE_CFG_REQ_FLAGS_PATH_RX			   0x1UL
-	#define QUEUE_CFG_REQ_FLAGS_PATH_BIDIR			   0x2UL
-	#define QUEUE_CFG_REQ_FLAGS_PATH_LAST    QUEUE_CFG_REQ_FLAGS_PATH_BIDIR
-	__le32 enables;
-	#define QUEUE_CFG_REQ_ENABLES_DFLT_LEN			    0x1UL
-	#define QUEUE_CFG_REQ_ENABLES_SERVICE_PROFILE		    0x2UL
-	__le32 queue_id;
-	__le32 dflt_len;
-	u8 service_profile;
-	#define QUEUE_CFG_REQ_SERVICE_PROFILE_LOSSY		   0x0UL
-	#define QUEUE_CFG_REQ_SERVICE_PROFILE_LOSSLESS		   0x1UL
-	#define QUEUE_CFG_REQ_SERVICE_PROFILE_UNKNOWN		   0xffUL
-	u8 unused_0[7];
-};
-
-/* Output (16 bytes) */
-struct hwrm_queue_cfg_output {
-	__le16 error_code;
-	__le16 req_type;
-	__le16 seq_id;
-	__le16 resp_len;
-	__le32 unused_0;
-	u8 unused_1;
-	u8 unused_2;
-	u8 unused_3;
-	u8 valid;
-};
-
-/* hwrm_queue_pfcenable_qcfg */
-/* Input (24 bytes) */
-struct hwrm_queue_pfcenable_qcfg_input {
-	__le16 req_type;
-	__le16 cmpl_ring;
-	__le16 seq_id;
-	__le16 target_id;
-	__le64 resp_addr;
-	__le16 port_id;
-	__le16 unused_0[3];
-};
-
-/* Output (16 bytes) */
-struct hwrm_queue_pfcenable_qcfg_output {
-	__le16 error_code;
-	__le16 req_type;
-	__le16 seq_id;
-	__le16 resp_len;
-	__le32 flags;
-	#define QUEUE_PFCENABLE_QCFG_RESP_FLAGS_PRI0_PFC_ENABLED   0x1UL
-	#define QUEUE_PFCENABLE_QCFG_RESP_FLAGS_PRI1_PFC_ENABLED   0x2UL
-	#define QUEUE_PFCENABLE_QCFG_RESP_FLAGS_PRI2_PFC_ENABLED   0x4UL
-	#define QUEUE_PFCENABLE_QCFG_RESP_FLAGS_PRI3_PFC_ENABLED   0x8UL
-	#define QUEUE_PFCENABLE_QCFG_RESP_FLAGS_PRI4_PFC_ENABLED   0x10UL
-	#define QUEUE_PFCENABLE_QCFG_RESP_FLAGS_PRI5_PFC_ENABLED   0x20UL
-	#define QUEUE_PFCENABLE_QCFG_RESP_FLAGS_PRI6_PFC_ENABLED   0x40UL
-	#define QUEUE_PFCENABLE_QCFG_RESP_FLAGS_PRI7_PFC_ENABLED   0x80UL
-	u8 unused_0;
-	u8 unused_1;
-	u8 unused_2;
-	u8 valid;
-};
-
-/* hwrm_queue_pfcenable_cfg */
-/* Input (24 bytes) */
-struct hwrm_queue_pfcenable_cfg_input {
-	__le16 req_type;
-	__le16 cmpl_ring;
-	__le16 seq_id;
-	__le16 target_id;
-	__le64 resp_addr;
-	__le32 flags;
-	#define QUEUE_PFCENABLE_CFG_REQ_FLAGS_PRI0_PFC_ENABLED     0x1UL
-	#define QUEUE_PFCENABLE_CFG_REQ_FLAGS_PRI1_PFC_ENABLED     0x2UL
-	#define QUEUE_PFCENABLE_CFG_REQ_FLAGS_PRI2_PFC_ENABLED     0x4UL
-	#define QUEUE_PFCENABLE_CFG_REQ_FLAGS_PRI3_PFC_ENABLED     0x8UL
-	#define QUEUE_PFCENABLE_CFG_REQ_FLAGS_PRI4_PFC_ENABLED     0x10UL
-	#define QUEUE_PFCENABLE_CFG_REQ_FLAGS_PRI5_PFC_ENABLED     0x20UL
-	#define QUEUE_PFCENABLE_CFG_REQ_FLAGS_PRI6_PFC_ENABLED     0x40UL
-	#define QUEUE_PFCENABLE_CFG_REQ_FLAGS_PRI7_PFC_ENABLED     0x80UL
-	__le16 port_id;
-	__le16 unused_0;
-};
-
-/* Output (16 bytes) */
-struct hwrm_queue_pfcenable_cfg_output {
-	__le16 error_code;
-	__le16 req_type;
-	__le16 seq_id;
-	__le16 resp_len;
-	__le32 unused_0;
-	u8 unused_1;
-	u8 unused_2;
-	u8 unused_3;
-	u8 valid;
-};
-
-/* hwrm_queue_pri2cos_qcfg */
-/* Input (24 bytes) */
-struct hwrm_queue_pri2cos_qcfg_input {
-	__le16 req_type;
-	__le16 cmpl_ring;
-	__le16 seq_id;
-	__le16 target_id;
-	__le64 resp_addr;
-	__le32 flags;
-	#define QUEUE_PRI2COS_QCFG_REQ_FLAGS_PATH		    0x1UL
-	#define QUEUE_PRI2COS_QCFG_REQ_FLAGS_PATH_TX		   (0x0UL << 0)
-	#define QUEUE_PRI2COS_QCFG_REQ_FLAGS_PATH_RX		   (0x1UL << 0)
-	#define QUEUE_PRI2COS_QCFG_REQ_FLAGS_PATH_LAST    QUEUE_PRI2COS_QCFG_REQ_FLAGS_PATH_RX
-	#define QUEUE_PRI2COS_QCFG_REQ_FLAGS_IVLAN		    0x2UL
-	u8 port_id;
-	u8 unused_0[3];
-};
-
-/* Output (24 bytes) */
-struct hwrm_queue_pri2cos_qcfg_output {
-	__le16 error_code;
-	__le16 req_type;
-	__le16 seq_id;
-	__le16 resp_len;
-	u8 pri0_cos_queue_id;
-	u8 pri1_cos_queue_id;
-	u8 pri2_cos_queue_id;
-	u8 pri3_cos_queue_id;
-	u8 pri4_cos_queue_id;
-	u8 pri5_cos_queue_id;
-	u8 pri6_cos_queue_id;
-	u8 pri7_cos_queue_id;
-	u8 queue_cfg_info;
-	#define QUEUE_PRI2COS_QCFG_RESP_QUEUE_CFG_INFO_ASYM_CFG    0x1UL
-	u8 unused_0;
-	__le16 unused_1;
-	u8 unused_2;
-	u8 unused_3;
-	u8 unused_4;
-	u8 valid;
-};
-
-/* hwrm_queue_pri2cos_cfg */
-/* Input (40 bytes) */
-struct hwrm_queue_pri2cos_cfg_input {
-	__le16 req_type;
-	__le16 cmpl_ring;
-	__le16 seq_id;
-	__le16 target_id;
-	__le64 resp_addr;
-	__le32 flags;
-	#define QUEUE_PRI2COS_CFG_REQ_FLAGS_PATH_MASK		    0x3UL
-	#define QUEUE_PRI2COS_CFG_REQ_FLAGS_PATH_SFT		    0
-	#define QUEUE_PRI2COS_CFG_REQ_FLAGS_PATH_TX		   (0x0UL << 0)
-	#define QUEUE_PRI2COS_CFG_REQ_FLAGS_PATH_RX		   (0x1UL << 0)
-	#define QUEUE_PRI2COS_CFG_REQ_FLAGS_PATH_BIDIR		   (0x2UL << 0)
-	#define QUEUE_PRI2COS_CFG_REQ_FLAGS_PATH_LAST    QUEUE_PRI2COS_CFG_REQ_FLAGS_PATH_BIDIR
-	#define QUEUE_PRI2COS_CFG_REQ_FLAGS_IVLAN		    0x4UL
-	__le32 enables;
-	#define QUEUE_PRI2COS_CFG_REQ_ENABLES_PRI0_COS_QUEUE_ID    0x1UL
-	#define QUEUE_PRI2COS_CFG_REQ_ENABLES_PRI1_COS_QUEUE_ID    0x2UL
-	#define QUEUE_PRI2COS_CFG_REQ_ENABLES_PRI2_COS_QUEUE_ID    0x4UL
-	#define QUEUE_PRI2COS_CFG_REQ_ENABLES_PRI3_COS_QUEUE_ID    0x8UL
-	#define QUEUE_PRI2COS_CFG_REQ_ENABLES_PRI4_COS_QUEUE_ID    0x10UL
-	#define QUEUE_PRI2COS_CFG_REQ_ENABLES_PRI5_COS_QUEUE_ID    0x20UL
-	#define QUEUE_PRI2COS_CFG_REQ_ENABLES_PRI6_COS_QUEUE_ID    0x40UL
-	#define QUEUE_PRI2COS_CFG_REQ_ENABLES_PRI7_COS_QUEUE_ID    0x80UL
-	u8 port_id;
-	u8 pri0_cos_queue_id;
-	u8 pri1_cos_queue_id;
-	u8 pri2_cos_queue_id;
-	u8 pri3_cos_queue_id;
-	u8 pri4_cos_queue_id;
-	u8 pri5_cos_queue_id;
-	u8 pri6_cos_queue_id;
-	u8 pri7_cos_queue_id;
-	u8 unused_0[7];
-};
-
-/* Output (16 bytes) */
-struct hwrm_queue_pri2cos_cfg_output {
-	__le16 error_code;
-	__le16 req_type;
-	__le16 seq_id;
-	__le16 resp_len;
-	__le32 unused_0;
-	u8 unused_1;
-	u8 unused_2;
-	u8 unused_3;
-	u8 valid;
-};
-
-/* hwrm_queue_cos2bw_qcfg */
-/* Input (24 bytes) */
-struct hwrm_queue_cos2bw_qcfg_input {
-	__le16 req_type;
-	__le16 cmpl_ring;
-	__le16 seq_id;
-	__le16 target_id;
-	__le64 resp_addr;
-	__le16 port_id;
-	__le16 unused_0[3];
-};
-
-/* Output (112 bytes) */
-struct hwrm_queue_cos2bw_qcfg_output {
-	__le16 error_code;
-	__le16 req_type;
-	__le16 seq_id;
-	__le16 resp_len;
-	u8 queue_id0;
-	u8 unused_0;
-	__le16 unused_1;
-	__le32 queue_id0_min_bw;
-	#define QUEUE_COS2BW_QCFG_RESP_QUEUE_ID0_MIN_BW_BW_VALUE_MASK 0xfffffffUL
-	#define QUEUE_COS2BW_QCFG_RESP_QUEUE_ID0_MIN_BW_BW_VALUE_SFT 0
-	#define QUEUE_COS2BW_QCFG_RESP_QUEUE_ID0_MIN_BW_SCALE      0x10000000UL
-	#define QUEUE_COS2BW_QCFG_RESP_QUEUE_ID0_MIN_BW_SCALE_BITS (0x0UL << 28)
-	#define QUEUE_COS2BW_QCFG_RESP_QUEUE_ID0_MIN_BW_SCALE_BYTES (0x1UL << 28)
-	#define QUEUE_COS2BW_QCFG_RESP_QUEUE_ID0_MIN_BW_SCALE_LAST    QUEUE_COS2BW_QCFG_RESP_QUEUE_ID0_MIN_BW_SCALE_BYTES
-	#define QUEUE_COS2BW_QCFG_RESP_QUEUE_ID0_MIN_BW_BW_VALUE_UNIT_MASK 0xe0000000UL
-	#define QUEUE_COS2BW_QCFG_RESP_QUEUE_ID0_MIN_BW_BW_VALUE_UNIT_SFT 29
-	#define QUEUE_COS2BW_QCFG_RESP_QUEUE_ID0_MIN_BW_BW_VALUE_UNIT_MEGA (0x0UL << 29)
-	#define QUEUE_COS2BW_QCFG_RESP_QUEUE_ID0_MIN_BW_BW_VALUE_UNIT_KILO (0x2UL << 29)
-	#define QUEUE_COS2BW_QCFG_RESP_QUEUE_ID0_MIN_BW_BW_VALUE_UNIT_BASE (0x4UL << 29)
-	#define QUEUE_COS2BW_QCFG_RESP_QUEUE_ID0_MIN_BW_BW_VALUE_UNIT_GIGA (0x6UL << 29)
-	#define QUEUE_COS2BW_QCFG_RESP_QUEUE_ID0_MIN_BW_BW_VALUE_UNIT_PERCENT1_100 (0x1UL << 29)
-	#define QUEUE_COS2BW_QCFG_RESP_QUEUE_ID0_MIN_BW_BW_VALUE_UNIT_INVALID (0x7UL << 29)
-	#define QUEUE_COS2BW_QCFG_RESP_QUEUE_ID0_MIN_BW_BW_VALUE_UNIT_LAST    QUEUE_COS2BW_QCFG_RESP_QUEUE_ID0_MIN_BW_BW_VALUE_UNIT_INVALID
-	__le32 queue_id0_max_bw;
-	#define QUEUE_COS2BW_QCFG_RESP_QUEUE_ID0_MAX_BW_BW_VALUE_MASK 0xfffffffUL
-	#define QUEUE_COS2BW_QCFG_RESP_QUEUE_ID0_MAX_BW_BW_VALUE_SFT 0
-	#define QUEUE_COS2BW_QCFG_RESP_QUEUE_ID0_MAX_BW_SCALE      0x10000000UL
-	#define QUEUE_COS2BW_QCFG_RESP_QUEUE_ID0_MAX_BW_SCALE_BITS (0x0UL << 28)
-	#define QUEUE_COS2BW_QCFG_RESP_QUEUE_ID0_MAX_BW_SCALE_BYTES (0x1UL << 28)
-	#define QUEUE_COS2BW_QCFG_RESP_QUEUE_ID0_MAX_BW_SCALE_LAST    QUEUE_COS2BW_QCFG_RESP_QUEUE_ID0_MAX_BW_SCALE_BYTES
-	#define QUEUE_COS2BW_QCFG_RESP_QUEUE_ID0_MAX_BW_BW_VALUE_UNIT_MASK 0xe0000000UL
-	#define QUEUE_COS2BW_QCFG_RESP_QUEUE_ID0_MAX_BW_BW_VALUE_UNIT_SFT 29
-	#define QUEUE_COS2BW_QCFG_RESP_QUEUE_ID0_MAX_BW_BW_VALUE_UNIT_MEGA (0x0UL << 29)
-	#define QUEUE_COS2BW_QCFG_RESP_QUEUE_ID0_MAX_BW_BW_VALUE_UNIT_KILO (0x2UL << 29)
-	#define QUEUE_COS2BW_QCFG_RESP_QUEUE_ID0_MAX_BW_BW_VALUE_UNIT_BASE (0x4UL << 29)
-	#define QUEUE_COS2BW_QCFG_RESP_QUEUE_ID0_MAX_BW_BW_VALUE_UNIT_GIGA (0x6UL << 29)
-	#define QUEUE_COS2BW_QCFG_RESP_QUEUE_ID0_MAX_BW_BW_VALUE_UNIT_PERCENT1_100 (0x1UL << 29)
-	#define QUEUE_COS2BW_QCFG_RESP_QUEUE_ID0_MAX_BW_BW_VALUE_UNIT_INVALID (0x7UL << 29)
-	#define QUEUE_COS2BW_QCFG_RESP_QUEUE_ID0_MAX_BW_BW_VALUE_UNIT_LAST    QUEUE_COS2BW_QCFG_RESP_QUEUE_ID0_MAX_BW_BW_VALUE_UNIT_INVALID
-	u8 queue_id0_tsa_assign;
-	#define QUEUE_COS2BW_QCFG_RESP_QUEUE_ID0_TSA_ASSIGN_SP    0x0UL
-	#define QUEUE_COS2BW_QCFG_RESP_QUEUE_ID0_TSA_ASSIGN_ETS   0x1UL
-	#define QUEUE_COS2BW_QCFG_RESP_QUEUE_ID0_TSA_ASSIGN_RESERVED_FIRST 0x2UL
-	#define QUEUE_COS2BW_QCFG_RESP_QUEUE_ID0_TSA_ASSIGN_RESERVED_LAST 0xffUL
-	u8 queue_id0_pri_lvl;
-	u8 queue_id0_bw_weight;
-	u8 queue_id1;
-	__le32 queue_id1_min_bw;
-	#define QUEUE_COS2BW_QCFG_RESP_QUEUE_ID1_MIN_BW_BW_VALUE_MASK 0xfffffffUL
-	#define QUEUE_COS2BW_QCFG_RESP_QUEUE_ID1_MIN_BW_BW_VALUE_SFT 0
-	#define QUEUE_COS2BW_QCFG_RESP_QUEUE_ID1_MIN_BW_SCALE      0x10000000UL
-	#define QUEUE_COS2BW_QCFG_RESP_QUEUE_ID1_MIN_BW_SCALE_BITS (0x0UL << 28)
-	#define QUEUE_COS2BW_QCFG_RESP_QUEUE_ID1_MIN_BW_SCALE_BYTES (0x1UL << 28)
-	#define QUEUE_COS2BW_QCFG_RESP_QUEUE_ID1_MIN_BW_SCALE_LAST    QUEUE_COS2BW_QCFG_RESP_QUEUE_ID1_MIN_BW_SCALE_BYTES
-	#define QUEUE_COS2BW_QCFG_RESP_QUEUE_ID1_MIN_BW_BW_VALUE_UNIT_MASK 0xe0000000UL
-	#define QUEUE_COS2BW_QCFG_RESP_QUEUE_ID1_MIN_BW_BW_VALUE_UNIT_SFT 29
-	#define QUEUE_COS2BW_QCFG_RESP_QUEUE_ID1_MIN_BW_BW_VALUE_UNIT_MEGA (0x0UL << 29)
-	#define QUEUE_COS2BW_QCFG_RESP_QUEUE_ID1_MIN_BW_BW_VALUE_UNIT_KILO (0x2UL << 29)
-	#define QUEUE_COS2BW_QCFG_RESP_QUEUE_ID1_MIN_BW_BW_VALUE_UNIT_BASE (0x4UL << 29)
-	#define QUEUE_COS2BW_QCFG_RESP_QUEUE_ID1_MIN_BW_BW_VALUE_UNIT_GIGA (0x6UL << 29)
-	#define QUEUE_COS2BW_QCFG_RESP_QUEUE_ID1_MIN_BW_BW_VALUE_UNIT_PERCENT1_100 (0x1UL << 29)
-	#define QUEUE_COS2BW_QCFG_RESP_QUEUE_ID1_MIN_BW_BW_VALUE_UNIT_INVALID (0x7UL << 29)
-	#define QUEUE_COS2BW_QCFG_RESP_QUEUE_ID1_MIN_BW_BW_VALUE_UNIT_LAST    QUEUE_COS2BW_QCFG_RESP_QUEUE_ID1_MIN_BW_BW_VALUE_UNIT_INVALID
-	__le32 queue_id1_max_bw;
-	#define QUEUE_COS2BW_QCFG_RESP_QUEUE_ID1_MAX_BW_BW_VALUE_MASK 0xfffffffUL
-	#define QUEUE_COS2BW_QCFG_RESP_QUEUE_ID1_MAX_BW_BW_VALUE_SFT 0
-	#define QUEUE_COS2BW_QCFG_RESP_QUEUE_ID1_MAX_BW_SCALE      0x10000000UL
-	#define QUEUE_COS2BW_QCFG_RESP_QUEUE_ID1_MAX_BW_SCALE_BITS (0x0UL << 28)
-	#define QUEUE_COS2BW_QCFG_RESP_QUEUE_ID1_MAX_BW_SCALE_BYTES (0x1UL << 28)
-	#define QUEUE_COS2BW_QCFG_RESP_QUEUE_ID1_MAX_BW_SCALE_LAST    QUEUE_COS2BW_QCFG_RESP_QUEUE_ID1_MAX_BW_SCALE_BYTES
-	#define QUEUE_COS2BW_QCFG_RESP_QUEUE_ID1_MAX_BW_BW_VALUE_UNIT_MASK 0xe0000000UL
-	#define QUEUE_COS2BW_QCFG_RESP_QUEUE_ID1_MAX_BW_BW_VALUE_UNIT_SFT 29
-	#define QUEUE_COS2BW_QCFG_RESP_QUEUE_ID1_MAX_BW_BW_VALUE_UNIT_MEGA (0x0UL << 29)
-	#define QUEUE_COS2BW_QCFG_RESP_QUEUE_ID1_MAX_BW_BW_VALUE_UNIT_KILO (0x2UL << 29)
-	#define QUEUE_COS2BW_QCFG_RESP_QUEUE_ID1_MAX_BW_BW_VALUE_UNIT_BASE (0x4UL << 29)
-	#define QUEUE_COS2BW_QCFG_RESP_QUEUE_ID1_MAX_BW_BW_VALUE_UNIT_GIGA (0x6UL << 29)
-	#define QUEUE_COS2BW_QCFG_RESP_QUEUE_ID1_MAX_BW_BW_VALUE_UNIT_PERCENT1_100 (0x1UL << 29)
-	#define QUEUE_COS2BW_QCFG_RESP_QUEUE_ID1_MAX_BW_BW_VALUE_UNIT_INVALID (0x7UL << 29)
-	#define QUEUE_COS2BW_QCFG_RESP_QUEUE_ID1_MAX_BW_BW_VALUE_UNIT_LAST    QUEUE_COS2BW_QCFG_RESP_QUEUE_ID1_MAX_BW_BW_VALUE_UNIT_INVALID
-	u8 queue_id1_tsa_assign;
-	#define QUEUE_COS2BW_QCFG_RESP_QUEUE_ID1_TSA_ASSIGN_SP    0x0UL
-	#define QUEUE_COS2BW_QCFG_RESP_QUEUE_ID1_TSA_ASSIGN_ETS   0x1UL
-	#define QUEUE_COS2BW_QCFG_RESP_QUEUE_ID1_TSA_ASSIGN_RESERVED_FIRST 0x2UL
-	#define QUEUE_COS2BW_QCFG_RESP_QUEUE_ID1_TSA_ASSIGN_RESERVED_LAST 0xffUL
-	u8 queue_id1_pri_lvl;
-	u8 queue_id1_bw_weight;
-	u8 queue_id2;
-	__le32 queue_id2_min_bw;
-	#define QUEUE_COS2BW_QCFG_RESP_QUEUE_ID2_MIN_BW_BW_VALUE_MASK 0xfffffffUL
-	#define QUEUE_COS2BW_QCFG_RESP_QUEUE_ID2_MIN_BW_BW_VALUE_SFT 0
-	#define QUEUE_COS2BW_QCFG_RESP_QUEUE_ID2_MIN_BW_SCALE      0x10000000UL
-	#define QUEUE_COS2BW_QCFG_RESP_QUEUE_ID2_MIN_BW_SCALE_BITS (0x0UL << 28)
-	#define QUEUE_COS2BW_QCFG_RESP_QUEUE_ID2_MIN_BW_SCALE_BYTES (0x1UL << 28)
-	#define QUEUE_COS2BW_QCFG_RESP_QUEUE_ID2_MIN_BW_SCALE_LAST    QUEUE_COS2BW_QCFG_RESP_QUEUE_ID2_MIN_BW_SCALE_BYTES
-	#define QUEUE_COS2BW_QCFG_RESP_QUEUE_ID2_MIN_BW_BW_VALUE_UNIT_MASK 0xe0000000UL
-	#define QUEUE_COS2BW_QCFG_RESP_QUEUE_ID2_MIN_BW_BW_VALUE_UNIT_SFT 29
-	#define QUEUE_COS2BW_QCFG_RESP_QUEUE_ID2_MIN_BW_BW_VALUE_UNIT_MEGA (0x0UL << 29)
-	#define QUEUE_COS2BW_QCFG_RESP_QUEUE_ID2_MIN_BW_BW_VALUE_UNIT_KILO (0x2UL << 29)
-	#define QUEUE_COS2BW_QCFG_RESP_QUEUE_ID2_MIN_BW_BW_VALUE_UNIT_BASE (0x4UL << 29)
-	#define QUEUE_COS2BW_QCFG_RESP_QUEUE_ID2_MIN_BW_BW_VALUE_UNIT_GIGA (0x6UL << 29)
-	#define QUEUE_COS2BW_QCFG_RESP_QUEUE_ID2_MIN_BW_BW_VALUE_UNIT_PERCENT1_100 (0x1UL << 29)
-	#define QUEUE_COS2BW_QCFG_RESP_QUEUE_ID2_MIN_BW_BW_VALUE_UNIT_INVALID (0x7UL << 29)
-	#define QUEUE_COS2BW_QCFG_RESP_QUEUE_ID2_MIN_BW_BW_VALUE_UNIT_LAST    QUEUE_COS2BW_QCFG_RESP_QUEUE_ID2_MIN_BW_BW_VALUE_UNIT_INVALID
-	__le32 queue_id2_max_bw;
-	#define QUEUE_COS2BW_QCFG_RESP_QUEUE_ID2_MAX_BW_BW_VALUE_MASK 0xfffffffUL
-	#define QUEUE_COS2BW_QCFG_RESP_QUEUE_ID2_MAX_BW_BW_VALUE_SFT 0
-	#define QUEUE_COS2BW_QCFG_RESP_QUEUE_ID2_MAX_BW_SCALE      0x10000000UL
-	#define QUEUE_COS2BW_QCFG_RESP_QUEUE_ID2_MAX_BW_SCALE_BITS (0x0UL << 28)
-	#define QUEUE_COS2BW_QCFG_RESP_QUEUE_ID2_MAX_BW_SCALE_BYTES (0x1UL << 28)
-	#define QUEUE_COS2BW_QCFG_RESP_QUEUE_ID2_MAX_BW_SCALE_LAST    QUEUE_COS2BW_QCFG_RESP_QUEUE_ID2_MAX_BW_SCALE_BYTES
-	#define QUEUE_COS2BW_QCFG_RESP_QUEUE_ID2_MAX_BW_BW_VALUE_UNIT_MASK 0xe0000000UL
-	#define QUEUE_COS2BW_QCFG_RESP_QUEUE_ID2_MAX_BW_BW_VALUE_UNIT_SFT 29
-	#define QUEUE_COS2BW_QCFG_RESP_QUEUE_ID2_MAX_BW_BW_VALUE_UNIT_MEGA (0x0UL << 29)
-	#define QUEUE_COS2BW_QCFG_RESP_QUEUE_ID2_MAX_BW_BW_VALUE_UNIT_KILO (0x2UL << 29)
-	#define QUEUE_COS2BW_QCFG_RESP_QUEUE_ID2_MAX_BW_BW_VALUE_UNIT_BASE (0x4UL << 29)
-	#define QUEUE_COS2BW_QCFG_RESP_QUEUE_ID2_MAX_BW_BW_VALUE_UNIT_GIGA (0x6UL << 29)
-	#define QUEUE_COS2BW_QCFG_RESP_QUEUE_ID2_MAX_BW_BW_VALUE_UNIT_PERCENT1_100 (0x1UL << 29)
-	#define QUEUE_COS2BW_QCFG_RESP_QUEUE_ID2_MAX_BW_BW_VALUE_UNIT_INVALID (0x7UL << 29)
-	#define QUEUE_COS2BW_QCFG_RESP_QUEUE_ID2_MAX_BW_BW_VALUE_UNIT_LAST    QUEUE_COS2BW_QCFG_RESP_QUEUE_ID2_MAX_BW_BW_VALUE_UNIT_INVALID
-	u8 queue_id2_tsa_assign;
-	#define QUEUE_COS2BW_QCFG_RESP_QUEUE_ID2_TSA_ASSIGN_SP    0x0UL
-	#define QUEUE_COS2BW_QCFG_RESP_QUEUE_ID2_TSA_ASSIGN_ETS   0x1UL
-	#define QUEUE_COS2BW_QCFG_RESP_QUEUE_ID2_TSA_ASSIGN_RESERVED_FIRST 0x2UL
-	#define QUEUE_COS2BW_QCFG_RESP_QUEUE_ID2_TSA_ASSIGN_RESERVED_LAST 0xffUL
-	u8 queue_id2_pri_lvl;
-	u8 queue_id2_bw_weight;
-	u8 queue_id3;
-	__le32 queue_id3_min_bw;
-	#define QUEUE_COS2BW_QCFG_RESP_QUEUE_ID3_MIN_BW_BW_VALUE_MASK 0xfffffffUL
-	#define QUEUE_COS2BW_QCFG_RESP_QUEUE_ID3_MIN_BW_BW_VALUE_SFT 0
-	#define QUEUE_COS2BW_QCFG_RESP_QUEUE_ID3_MIN_BW_SCALE      0x10000000UL
-	#define QUEUE_COS2BW_QCFG_RESP_QUEUE_ID3_MIN_BW_SCALE_BITS (0x0UL << 28)
-	#define QUEUE_COS2BW_QCFG_RESP_QUEUE_ID3_MIN_BW_SCALE_BYTES (0x1UL << 28)
-	#define QUEUE_COS2BW_QCFG_RESP_QUEUE_ID3_MIN_BW_SCALE_LAST    QUEUE_COS2BW_QCFG_RESP_QUEUE_ID3_MIN_BW_SCALE_BYTES
-	#define QUEUE_COS2BW_QCFG_RESP_QUEUE_ID3_MIN_BW_BW_VALUE_UNIT_MASK 0xe0000000UL
-	#define QUEUE_COS2BW_QCFG_RESP_QUEUE_ID3_MIN_BW_BW_VALUE_UNIT_SFT 29
-	#define QUEUE_COS2BW_QCFG_RESP_QUEUE_ID3_MIN_BW_BW_VALUE_UNIT_MEGA (0x0UL << 29)
-	#define QUEUE_COS2BW_QCFG_RESP_QUEUE_ID3_MIN_BW_BW_VALUE_UNIT_KILO (0x2UL << 29)
-	#define QUEUE_COS2BW_QCFG_RESP_QUEUE_ID3_MIN_BW_BW_VALUE_UNIT_BASE (0x4UL << 29)
-	#define QUEUE_COS2BW_QCFG_RESP_QUEUE_ID3_MIN_BW_BW_VALUE_UNIT_GIGA (0x6UL << 29)
-	#define QUEUE_COS2BW_QCFG_RESP_QUEUE_ID3_MIN_BW_BW_VALUE_UNIT_PERCENT1_100 (0x1UL << 29)
-	#define QUEUE_COS2BW_QCFG_RESP_QUEUE_ID3_MIN_BW_BW_VALUE_UNIT_INVALID (0x7UL << 29)
-	#define QUEUE_COS2BW_QCFG_RESP_QUEUE_ID3_MIN_BW_BW_VALUE_UNIT_LAST    QUEUE_COS2BW_QCFG_RESP_QUEUE_ID3_MIN_BW_BW_VALUE_UNIT_INVALID
-	__le32 queue_id3_max_bw;
-	#define QUEUE_COS2BW_QCFG_RESP_QUEUE_ID3_MAX_BW_BW_VALUE_MASK 0xfffffffUL
-	#define QUEUE_COS2BW_QCFG_RESP_QUEUE_ID3_MAX_BW_BW_VALUE_SFT 0
-	#define QUEUE_COS2BW_QCFG_RESP_QUEUE_ID3_MAX_BW_SCALE      0x10000000UL
-	#define QUEUE_COS2BW_QCFG_RESP_QUEUE_ID3_MAX_BW_SCALE_BITS (0x0UL << 28)
-	#define QUEUE_COS2BW_QCFG_RESP_QUEUE_ID3_MAX_BW_SCALE_BYTES (0x1UL << 28)
-	#define QUEUE_COS2BW_QCFG_RESP_QUEUE_ID3_MAX_BW_SCALE_LAST    QUEUE_COS2BW_QCFG_RESP_QUEUE_ID3_MAX_BW_SCALE_BYTES
-	#define QUEUE_COS2BW_QCFG_RESP_QUEUE_ID3_MAX_BW_BW_VALUE_UNIT_MASK 0xe0000000UL
-	#define QUEUE_COS2BW_QCFG_RESP_QUEUE_ID3_MAX_BW_BW_VALUE_UNIT_SFT 29
-	#define QUEUE_COS2BW_QCFG_RESP_QUEUE_ID3_MAX_BW_BW_VALUE_UNIT_MEGA (0x0UL << 29)
-	#define QUEUE_COS2BW_QCFG_RESP_QUEUE_ID3_MAX_BW_BW_VALUE_UNIT_KILO (0x2UL << 29)
-	#define QUEUE_COS2BW_QCFG_RESP_QUEUE_ID3_MAX_BW_BW_VALUE_UNIT_BASE (0x4UL << 29)
-	#define QUEUE_COS2BW_QCFG_RESP_QUEUE_ID3_MAX_BW_BW_VALUE_UNIT_GIGA (0x6UL << 29)
-	#define QUEUE_COS2BW_QCFG_RESP_QUEUE_ID3_MAX_BW_BW_VALUE_UNIT_PERCENT1_100 (0x1UL << 29)
-	#define QUEUE_COS2BW_QCFG_RESP_QUEUE_ID3_MAX_BW_BW_VALUE_UNIT_INVALID (0x7UL << 29)
-	#define QUEUE_COS2BW_QCFG_RESP_QUEUE_ID3_MAX_BW_BW_VALUE_UNIT_LAST    QUEUE_COS2BW_QCFG_RESP_QUEUE_ID3_MAX_BW_BW_VALUE_UNIT_INVALID
-	u8 queue_id3_tsa_assign;
-	#define QUEUE_COS2BW_QCFG_RESP_QUEUE_ID3_TSA_ASSIGN_SP    0x0UL
-	#define QUEUE_COS2BW_QCFG_RESP_QUEUE_ID3_TSA_ASSIGN_ETS   0x1UL
-	#define QUEUE_COS2BW_QCFG_RESP_QUEUE_ID3_TSA_ASSIGN_RESERVED_FIRST 0x2UL
-	#define QUEUE_COS2BW_QCFG_RESP_QUEUE_ID3_TSA_ASSIGN_RESERVED_LAST 0xffUL
-	u8 queue_id3_pri_lvl;
-	u8 queue_id3_bw_weight;
-	u8 queue_id4;
-	__le32 queue_id4_min_bw;
-	#define QUEUE_COS2BW_QCFG_RESP_QUEUE_ID4_MIN_BW_BW_VALUE_MASK 0xfffffffUL
-	#define QUEUE_COS2BW_QCFG_RESP_QUEUE_ID4_MIN_BW_BW_VALUE_SFT 0
-	#define QUEUE_COS2BW_QCFG_RESP_QUEUE_ID4_MIN_BW_SCALE      0x10000000UL
-	#define QUEUE_COS2BW_QCFG_RESP_QUEUE_ID4_MIN_BW_SCALE_BITS (0x0UL << 28)
-	#define QUEUE_COS2BW_QCFG_RESP_QUEUE_ID4_MIN_BW_SCALE_BYTES (0x1UL << 28)
-	#define QUEUE_COS2BW_QCFG_RESP_QUEUE_ID4_MIN_BW_SCALE_LAST    QUEUE_COS2BW_QCFG_RESP_QUEUE_ID4_MIN_BW_SCALE_BYTES
-	#define QUEUE_COS2BW_QCFG_RESP_QUEUE_ID4_MIN_BW_BW_VALUE_UNIT_MASK 0xe0000000UL
-	#define QUEUE_COS2BW_QCFG_RESP_QUEUE_ID4_MIN_BW_BW_VALUE_UNIT_SFT 29
-	#define QUEUE_COS2BW_QCFG_RESP_QUEUE_ID4_MIN_BW_BW_VALUE_UNIT_MEGA (0x0UL << 29)
-	#define QUEUE_COS2BW_QCFG_RESP_QUEUE_ID4_MIN_BW_BW_VALUE_UNIT_KILO (0x2UL << 29)
-	#define QUEUE_COS2BW_QCFG_RESP_QUEUE_ID4_MIN_BW_BW_VALUE_UNIT_BASE (0x4UL << 29)
-	#define QUEUE_COS2BW_QCFG_RESP_QUEUE_ID4_MIN_BW_BW_VALUE_UNIT_GIGA (0x6UL << 29)
-	#define QUEUE_COS2BW_QCFG_RESP_QUEUE_ID4_MIN_BW_BW_VALUE_UNIT_PERCENT1_100 (0x1UL << 29)
-	#define QUEUE_COS2BW_QCFG_RESP_QUEUE_ID4_MIN_BW_BW_VALUE_UNIT_INVALID (0x7UL << 29)
-	#define QUEUE_COS2BW_QCFG_RESP_QUEUE_ID4_MIN_BW_BW_VALUE_UNIT_LAST    QUEUE_COS2BW_QCFG_RESP_QUEUE_ID4_MIN_BW_BW_VALUE_UNIT_INVALID
-	__le32 queue_id4_max_bw;
-	#define QUEUE_COS2BW_QCFG_RESP_QUEUE_ID4_MAX_BW_BW_VALUE_MASK 0xfffffffUL
-	#define QUEUE_COS2BW_QCFG_RESP_QUEUE_ID4_MAX_BW_BW_VALUE_SFT 0
-	#define QUEUE_COS2BW_QCFG_RESP_QUEUE_ID4_MAX_BW_SCALE      0x10000000UL
-	#define QUEUE_COS2BW_QCFG_RESP_QUEUE_ID4_MAX_BW_SCALE_BITS (0x0UL << 28)
-	#define QUEUE_COS2BW_QCFG_RESP_QUEUE_ID4_MAX_BW_SCALE_BYTES (0x1UL << 28)
-	#define QUEUE_COS2BW_QCFG_RESP_QUEUE_ID4_MAX_BW_SCALE_LAST    QUEUE_COS2BW_QCFG_RESP_QUEUE_ID4_MAX_BW_SCALE_BYTES
-	#define QUEUE_COS2BW_QCFG_RESP_QUEUE_ID4_MAX_BW_BW_VALUE_UNIT_MASK 0xe0000000UL
-	#define QUEUE_COS2BW_QCFG_RESP_QUEUE_ID4_MAX_BW_BW_VALUE_UNIT_SFT 29
-	#define QUEUE_COS2BW_QCFG_RESP_QUEUE_ID4_MAX_BW_BW_VALUE_UNIT_MEGA (0x0UL << 29)
-	#define QUEUE_COS2BW_QCFG_RESP_QUEUE_ID4_MAX_BW_BW_VALUE_UNIT_KILO (0x2UL << 29)
-	#define QUEUE_COS2BW_QCFG_RESP_QUEUE_ID4_MAX_BW_BW_VALUE_UNIT_BASE (0x4UL << 29)
-	#define QUEUE_COS2BW_QCFG_RESP_QUEUE_ID4_MAX_BW_BW_VALUE_UNIT_GIGA (0x6UL << 29)
-	#define QUEUE_COS2BW_QCFG_RESP_QUEUE_ID4_MAX_BW_BW_VALUE_UNIT_PERCENT1_100 (0x1UL << 29)
-	#define QUEUE_COS2BW_QCFG_RESP_QUEUE_ID4_MAX_BW_BW_VALUE_UNIT_INVALID (0x7UL << 29)
-	#define QUEUE_COS2BW_QCFG_RESP_QUEUE_ID4_MAX_BW_BW_VALUE_UNIT_LAST    QUEUE_COS2BW_QCFG_RESP_QUEUE_ID4_MAX_BW_BW_VALUE_UNIT_INVALID
-	u8 queue_id4_tsa_assign;
-	#define QUEUE_COS2BW_QCFG_RESP_QUEUE_ID4_TSA_ASSIGN_SP    0x0UL
-	#define QUEUE_COS2BW_QCFG_RESP_QUEUE_ID4_TSA_ASSIGN_ETS   0x1UL
-	#define QUEUE_COS2BW_QCFG_RESP_QUEUE_ID4_TSA_ASSIGN_RESERVED_FIRST 0x2UL
-	#define QUEUE_COS2BW_QCFG_RESP_QUEUE_ID4_TSA_ASSIGN_RESERVED_LAST 0xffUL
-	u8 queue_id4_pri_lvl;
-	u8 queue_id4_bw_weight;
-	u8 queue_id5;
-	__le32 queue_id5_min_bw;
-	#define QUEUE_COS2BW_QCFG_RESP_QUEUE_ID5_MIN_BW_BW_VALUE_MASK 0xfffffffUL
-	#define QUEUE_COS2BW_QCFG_RESP_QUEUE_ID5_MIN_BW_BW_VALUE_SFT 0
-	#define QUEUE_COS2BW_QCFG_RESP_QUEUE_ID5_MIN_BW_SCALE      0x10000000UL
-	#define QUEUE_COS2BW_QCFG_RESP_QUEUE_ID5_MIN_BW_SCALE_BITS (0x0UL << 28)
-	#define QUEUE_COS2BW_QCFG_RESP_QUEUE_ID5_MIN_BW_SCALE_BYTES (0x1UL << 28)
-	#define QUEUE_COS2BW_QCFG_RESP_QUEUE_ID5_MIN_BW_SCALE_LAST    QUEUE_COS2BW_QCFG_RESP_QUEUE_ID5_MIN_BW_SCALE_BYTES
-	#define QUEUE_COS2BW_QCFG_RESP_QUEUE_ID5_MIN_BW_BW_VALUE_UNIT_MASK 0xe0000000UL
-	#define QUEUE_COS2BW_QCFG_RESP_QUEUE_ID5_MIN_BW_BW_VALUE_UNIT_SFT 29
-	#define QUEUE_COS2BW_QCFG_RESP_QUEUE_ID5_MIN_BW_BW_VALUE_UNIT_MEGA (0x0UL << 29)
-	#define QUEUE_COS2BW_QCFG_RESP_QUEUE_ID5_MIN_BW_BW_VALUE_UNIT_KILO (0x2UL << 29)
-	#define QUEUE_COS2BW_QCFG_RESP_QUEUE_ID5_MIN_BW_BW_VALUE_UNIT_BASE (0x4UL << 29)
-	#define QUEUE_COS2BW_QCFG_RESP_QUEUE_ID5_MIN_BW_BW_VALUE_UNIT_GIGA (0x6UL << 29)
-	#define QUEUE_COS2BW_QCFG_RESP_QUEUE_ID5_MIN_BW_BW_VALUE_UNIT_PERCENT1_100 (0x1UL << 29)
-	#define QUEUE_COS2BW_QCFG_RESP_QUEUE_ID5_MIN_BW_BW_VALUE_UNIT_INVALID (0x7UL << 29)
-	#define QUEUE_COS2BW_QCFG_RESP_QUEUE_ID5_MIN_BW_BW_VALUE_UNIT_LAST    QUEUE_COS2BW_QCFG_RESP_QUEUE_ID5_MIN_BW_BW_VALUE_UNIT_INVALID
-	__le32 queue_id5_max_bw;
-	#define QUEUE_COS2BW_QCFG_RESP_QUEUE_ID5_MAX_BW_BW_VALUE_MASK 0xfffffffUL
-	#define QUEUE_COS2BW_QCFG_RESP_QUEUE_ID5_MAX_BW_BW_VALUE_SFT 0
-	#define QUEUE_COS2BW_QCFG_RESP_QUEUE_ID5_MAX_BW_SCALE      0x10000000UL
-	#define QUEUE_COS2BW_QCFG_RESP_QUEUE_ID5_MAX_BW_SCALE_BITS (0x0UL << 28)
-	#define QUEUE_COS2BW_QCFG_RESP_QUEUE_ID5_MAX_BW_SCALE_BYTES (0x1UL << 28)
-	#define QUEUE_COS2BW_QCFG_RESP_QUEUE_ID5_MAX_BW_SCALE_LAST    QUEUE_COS2BW_QCFG_RESP_QUEUE_ID5_MAX_BW_SCALE_BYTES
-	#define QUEUE_COS2BW_QCFG_RESP_QUEUE_ID5_MAX_BW_BW_VALUE_UNIT_MASK 0xe0000000UL
-	#define QUEUE_COS2BW_QCFG_RESP_QUEUE_ID5_MAX_BW_BW_VALUE_UNIT_SFT 29
-	#define QUEUE_COS2BW_QCFG_RESP_QUEUE_ID5_MAX_BW_BW_VALUE_UNIT_MEGA (0x0UL << 29)
-	#define QUEUE_COS2BW_QCFG_RESP_QUEUE_ID5_MAX_BW_BW_VALUE_UNIT_KILO (0x2UL << 29)
-	#define QUEUE_COS2BW_QCFG_RESP_QUEUE_ID5_MAX_BW_BW_VALUE_UNIT_BASE (0x4UL << 29)
-	#define QUEUE_COS2BW_QCFG_RESP_QUEUE_ID5_MAX_BW_BW_VALUE_UNIT_GIGA (0x6UL << 29)
-	#define QUEUE_COS2BW_QCFG_RESP_QUEUE_ID5_MAX_BW_BW_VALUE_UNIT_PERCENT1_100 (0x1UL << 29)
-	#define QUEUE_COS2BW_QCFG_RESP_QUEUE_ID5_MAX_BW_BW_VALUE_UNIT_INVALID (0x7UL << 29)
-	#define QUEUE_COS2BW_QCFG_RESP_QUEUE_ID5_MAX_BW_BW_VALUE_UNIT_LAST    QUEUE_COS2BW_QCFG_RESP_QUEUE_ID5_MAX_BW_BW_VALUE_UNIT_INVALID
-	u8 queue_id5_tsa_assign;
-	#define QUEUE_COS2BW_QCFG_RESP_QUEUE_ID5_TSA_ASSIGN_SP    0x0UL
-	#define QUEUE_COS2BW_QCFG_RESP_QUEUE_ID5_TSA_ASSIGN_ETS   0x1UL
-	#define QUEUE_COS2BW_QCFG_RESP_QUEUE_ID5_TSA_ASSIGN_RESERVED_FIRST 0x2UL
-	#define QUEUE_COS2BW_QCFG_RESP_QUEUE_ID5_TSA_ASSIGN_RESERVED_LAST 0xffUL
-	u8 queue_id5_pri_lvl;
-	u8 queue_id5_bw_weight;
-	u8 queue_id6;
-	__le32 queue_id6_min_bw;
-	#define QUEUE_COS2BW_QCFG_RESP_QUEUE_ID6_MIN_BW_BW_VALUE_MASK 0xfffffffUL
-	#define QUEUE_COS2BW_QCFG_RESP_QUEUE_ID6_MIN_BW_BW_VALUE_SFT 0
-	#define QUEUE_COS2BW_QCFG_RESP_QUEUE_ID6_MIN_BW_SCALE      0x10000000UL
-	#define QUEUE_COS2BW_QCFG_RESP_QUEUE_ID6_MIN_BW_SCALE_BITS (0x0UL << 28)
-	#define QUEUE_COS2BW_QCFG_RESP_QUEUE_ID6_MIN_BW_SCALE_BYTES (0x1UL << 28)
-	#define QUEUE_COS2BW_QCFG_RESP_QUEUE_ID6_MIN_BW_SCALE_LAST    QUEUE_COS2BW_QCFG_RESP_QUEUE_ID6_MIN_BW_SCALE_BYTES
-	#define QUEUE_COS2BW_QCFG_RESP_QUEUE_ID6_MIN_BW_BW_VALUE_UNIT_MASK 0xe0000000UL
-	#define QUEUE_COS2BW_QCFG_RESP_QUEUE_ID6_MIN_BW_BW_VALUE_UNIT_SFT 29
-	#define QUEUE_COS2BW_QCFG_RESP_QUEUE_ID6_MIN_BW_BW_VALUE_UNIT_MEGA (0x0UL << 29)
-	#define QUEUE_COS2BW_QCFG_RESP_QUEUE_ID6_MIN_BW_BW_VALUE_UNIT_KILO (0x2UL << 29)
-	#define QUEUE_COS2BW_QCFG_RESP_QUEUE_ID6_MIN_BW_BW_VALUE_UNIT_BASE (0x4UL << 29)
-	#define QUEUE_COS2BW_QCFG_RESP_QUEUE_ID6_MIN_BW_BW_VALUE_UNIT_GIGA (0x6UL << 29)
-	#define QUEUE_COS2BW_QCFG_RESP_QUEUE_ID6_MIN_BW_BW_VALUE_UNIT_PERCENT1_100 (0x1UL << 29)
-	#define QUEUE_COS2BW_QCFG_RESP_QUEUE_ID6_MIN_BW_BW_VALUE_UNIT_INVALID (0x7UL << 29)
-	#define QUEUE_COS2BW_QCFG_RESP_QUEUE_ID6_MIN_BW_BW_VALUE_UNIT_LAST    QUEUE_COS2BW_QCFG_RESP_QUEUE_ID6_MIN_BW_BW_VALUE_UNIT_INVALID
-	__le32 queue_id6_max_bw;
-	#define QUEUE_COS2BW_QCFG_RESP_QUEUE_ID6_MAX_BW_BW_VALUE_MASK 0xfffffffUL
-	#define QUEUE_COS2BW_QCFG_RESP_QUEUE_ID6_MAX_BW_BW_VALUE_SFT 0
-	#define QUEUE_COS2BW_QCFG_RESP_QUEUE_ID6_MAX_BW_SCALE      0x10000000UL
-	#define QUEUE_COS2BW_QCFG_RESP_QUEUE_ID6_MAX_BW_SCALE_BITS (0x0UL << 28)
-	#define QUEUE_COS2BW_QCFG_RESP_QUEUE_ID6_MAX_BW_SCALE_BYTES (0x1UL << 28)
-	#define QUEUE_COS2BW_QCFG_RESP_QUEUE_ID6_MAX_BW_SCALE_LAST    QUEUE_COS2BW_QCFG_RESP_QUEUE_ID6_MAX_BW_SCALE_BYTES
-	#define QUEUE_COS2BW_QCFG_RESP_QUEUE_ID6_MAX_BW_BW_VALUE_UNIT_MASK 0xe0000000UL
-	#define QUEUE_COS2BW_QCFG_RESP_QUEUE_ID6_MAX_BW_BW_VALUE_UNIT_SFT 29
-	#define QUEUE_COS2BW_QCFG_RESP_QUEUE_ID6_MAX_BW_BW_VALUE_UNIT_MEGA (0x0UL << 29)
-	#define QUEUE_COS2BW_QCFG_RESP_QUEUE_ID6_MAX_BW_BW_VALUE_UNIT_KILO (0x2UL << 29)
-	#define QUEUE_COS2BW_QCFG_RESP_QUEUE_ID6_MAX_BW_BW_VALUE_UNIT_BASE (0x4UL << 29)
-	#define QUEUE_COS2BW_QCFG_RESP_QUEUE_ID6_MAX_BW_BW_VALUE_UNIT_GIGA (0x6UL << 29)
-	#define QUEUE_COS2BW_QCFG_RESP_QUEUE_ID6_MAX_BW_BW_VALUE_UNIT_PERCENT1_100 (0x1UL << 29)
-	#define QUEUE_COS2BW_QCFG_RESP_QUEUE_ID6_MAX_BW_BW_VALUE_UNIT_INVALID (0x7UL << 29)
-	#define QUEUE_COS2BW_QCFG_RESP_QUEUE_ID6_MAX_BW_BW_VALUE_UNIT_LAST    QUEUE_COS2BW_QCFG_RESP_QUEUE_ID6_MAX_BW_BW_VALUE_UNIT_INVALID
-	u8 queue_id6_tsa_assign;
-	#define QUEUE_COS2BW_QCFG_RESP_QUEUE_ID6_TSA_ASSIGN_SP    0x0UL
-	#define QUEUE_COS2BW_QCFG_RESP_QUEUE_ID6_TSA_ASSIGN_ETS   0x1UL
-	#define QUEUE_COS2BW_QCFG_RESP_QUEUE_ID6_TSA_ASSIGN_RESERVED_FIRST 0x2UL
-	#define QUEUE_COS2BW_QCFG_RESP_QUEUE_ID6_TSA_ASSIGN_RESERVED_LAST 0xffUL
-	u8 queue_id6_pri_lvl;
-	u8 queue_id6_bw_weight;
-	u8 queue_id7;
-	__le32 queue_id7_min_bw;
-	#define QUEUE_COS2BW_QCFG_RESP_QUEUE_ID7_MIN_BW_BW_VALUE_MASK 0xfffffffUL
-	#define QUEUE_COS2BW_QCFG_RESP_QUEUE_ID7_MIN_BW_BW_VALUE_SFT 0
-	#define QUEUE_COS2BW_QCFG_RESP_QUEUE_ID7_MIN_BW_SCALE      0x10000000UL
-	#define QUEUE_COS2BW_QCFG_RESP_QUEUE_ID7_MIN_BW_SCALE_BITS (0x0UL << 28)
-	#define QUEUE_COS2BW_QCFG_RESP_QUEUE_ID7_MIN_BW_SCALE_BYTES (0x1UL << 28)
-	#define QUEUE_COS2BW_QCFG_RESP_QUEUE_ID7_MIN_BW_SCALE_LAST    QUEUE_COS2BW_QCFG_RESP_QUEUE_ID7_MIN_BW_SCALE_BYTES
-	#define QUEUE_COS2BW_QCFG_RESP_QUEUE_ID7_MIN_BW_BW_VALUE_UNIT_MASK 0xe0000000UL
-	#define QUEUE_COS2BW_QCFG_RESP_QUEUE_ID7_MIN_BW_BW_VALUE_UNIT_SFT 29
-	#define QUEUE_COS2BW_QCFG_RESP_QUEUE_ID7_MIN_BW_BW_VALUE_UNIT_MEGA (0x0UL << 29)
-	#define QUEUE_COS2BW_QCFG_RESP_QUEUE_ID7_MIN_BW_BW_VALUE_UNIT_KILO (0x2UL << 29)
-	#define QUEUE_COS2BW_QCFG_RESP_QUEUE_ID7_MIN_BW_BW_VALUE_UNIT_BASE (0x4UL << 29)
-	#define QUEUE_COS2BW_QCFG_RESP_QUEUE_ID7_MIN_BW_BW_VALUE_UNIT_GIGA (0x6UL << 29)
-	#define QUEUE_COS2BW_QCFG_RESP_QUEUE_ID7_MIN_BW_BW_VALUE_UNIT_PERCENT1_100 (0x1UL << 29)
-	#define QUEUE_COS2BW_QCFG_RESP_QUEUE_ID7_MIN_BW_BW_VALUE_UNIT_INVALID (0x7UL << 29)
-	#define QUEUE_COS2BW_QCFG_RESP_QUEUE_ID7_MIN_BW_BW_VALUE_UNIT_LAST    QUEUE_COS2BW_QCFG_RESP_QUEUE_ID7_MIN_BW_BW_VALUE_UNIT_INVALID
-	__le32 queue_id7_max_bw;
-	#define QUEUE_COS2BW_QCFG_RESP_QUEUE_ID7_MAX_BW_BW_VALUE_MASK 0xfffffffUL
-	#define QUEUE_COS2BW_QCFG_RESP_QUEUE_ID7_MAX_BW_BW_VALUE_SFT 0
-	#define QUEUE_COS2BW_QCFG_RESP_QUEUE_ID7_MAX_BW_SCALE      0x10000000UL
-	#define QUEUE_COS2BW_QCFG_RESP_QUEUE_ID7_MAX_BW_SCALE_BITS (0x0UL << 28)
-	#define QUEUE_COS2BW_QCFG_RESP_QUEUE_ID7_MAX_BW_SCALE_BYTES (0x1UL << 28)
-	#define QUEUE_COS2BW_QCFG_RESP_QUEUE_ID7_MAX_BW_SCALE_LAST    QUEUE_COS2BW_QCFG_RESP_QUEUE_ID7_MAX_BW_SCALE_BYTES
-	#define QUEUE_COS2BW_QCFG_RESP_QUEUE_ID7_MAX_BW_BW_VALUE_UNIT_MASK 0xe0000000UL
-	#define QUEUE_COS2BW_QCFG_RESP_QUEUE_ID7_MAX_BW_BW_VALUE_UNIT_SFT 29
-	#define QUEUE_COS2BW_QCFG_RESP_QUEUE_ID7_MAX_BW_BW_VALUE_UNIT_MEGA (0x0UL << 29)
-	#define QUEUE_COS2BW_QCFG_RESP_QUEUE_ID7_MAX_BW_BW_VALUE_UNIT_KILO (0x2UL << 29)
-	#define QUEUE_COS2BW_QCFG_RESP_QUEUE_ID7_MAX_BW_BW_VALUE_UNIT_BASE (0x4UL << 29)
-	#define QUEUE_COS2BW_QCFG_RESP_QUEUE_ID7_MAX_BW_BW_VALUE_UNIT_GIGA (0x6UL << 29)
-	#define QUEUE_COS2BW_QCFG_RESP_QUEUE_ID7_MAX_BW_BW_VALUE_UNIT_PERCENT1_100 (0x1UL << 29)
-	#define QUEUE_COS2BW_QCFG_RESP_QUEUE_ID7_MAX_BW_BW_VALUE_UNIT_INVALID (0x7UL << 29)
-	#define QUEUE_COS2BW_QCFG_RESP_QUEUE_ID7_MAX_BW_BW_VALUE_UNIT_LAST    QUEUE_COS2BW_QCFG_RESP_QUEUE_ID7_MAX_BW_BW_VALUE_UNIT_INVALID
-	u8 queue_id7_tsa_assign;
-	#define QUEUE_COS2BW_QCFG_RESP_QUEUE_ID7_TSA_ASSIGN_SP    0x0UL
-	#define QUEUE_COS2BW_QCFG_RESP_QUEUE_ID7_TSA_ASSIGN_ETS   0x1UL
-	#define QUEUE_COS2BW_QCFG_RESP_QUEUE_ID7_TSA_ASSIGN_RESERVED_FIRST 0x2UL
-	#define QUEUE_COS2BW_QCFG_RESP_QUEUE_ID7_TSA_ASSIGN_RESERVED_LAST 0xffUL
-	u8 queue_id7_pri_lvl;
-	u8 queue_id7_bw_weight;
-	u8 unused_2;
-	u8 unused_3;
-	u8 unused_4;
-	u8 unused_5;
-	u8 valid;
-};
-
-/* hwrm_queue_cos2bw_cfg */
-/* Input (128 bytes) */
-struct hwrm_queue_cos2bw_cfg_input {
-	__le16 req_type;
-	__le16 cmpl_ring;
-	__le16 seq_id;
-	__le16 target_id;
-	__le64 resp_addr;
-	__le32 flags;
-	__le32 enables;
-	#define QUEUE_COS2BW_CFG_REQ_ENABLES_COS_QUEUE_ID0_VALID   0x1UL
-	#define QUEUE_COS2BW_CFG_REQ_ENABLES_COS_QUEUE_ID1_VALID   0x2UL
-	#define QUEUE_COS2BW_CFG_REQ_ENABLES_COS_QUEUE_ID2_VALID   0x4UL
-	#define QUEUE_COS2BW_CFG_REQ_ENABLES_COS_QUEUE_ID3_VALID   0x8UL
-	#define QUEUE_COS2BW_CFG_REQ_ENABLES_COS_QUEUE_ID4_VALID   0x10UL
-	#define QUEUE_COS2BW_CFG_REQ_ENABLES_COS_QUEUE_ID5_VALID   0x20UL
-	#define QUEUE_COS2BW_CFG_REQ_ENABLES_COS_QUEUE_ID6_VALID   0x40UL
-	#define QUEUE_COS2BW_CFG_REQ_ENABLES_COS_QUEUE_ID7_VALID   0x80UL
-	__le16 port_id;
-	u8 queue_id0;
-	u8 unused_0;
-	__le32 queue_id0_min_bw;
-	#define QUEUE_COS2BW_CFG_REQ_QUEUE_ID0_MIN_BW_BW_VALUE_MASK 0xfffffffUL
-	#define QUEUE_COS2BW_CFG_REQ_QUEUE_ID0_MIN_BW_BW_VALUE_SFT 0
-	#define QUEUE_COS2BW_CFG_REQ_QUEUE_ID0_MIN_BW_SCALE	    0x10000000UL
-	#define QUEUE_COS2BW_CFG_REQ_QUEUE_ID0_MIN_BW_SCALE_BITS  (0x0UL << 28)
-	#define QUEUE_COS2BW_CFG_REQ_QUEUE_ID0_MIN_BW_SCALE_BYTES (0x1UL << 28)
-	#define QUEUE_COS2BW_CFG_REQ_QUEUE_ID0_MIN_BW_SCALE_LAST    QUEUE_COS2BW_CFG_REQ_QUEUE_ID0_MIN_BW_SCALE_BYTES
-	#define QUEUE_COS2BW_CFG_REQ_QUEUE_ID0_MIN_BW_BW_VALUE_UNIT_MASK 0xe0000000UL
-	#define QUEUE_COS2BW_CFG_REQ_QUEUE_ID0_MIN_BW_BW_VALUE_UNIT_SFT 29
-	#define QUEUE_COS2BW_CFG_REQ_QUEUE_ID0_MIN_BW_BW_VALUE_UNIT_MEGA (0x0UL << 29)
-	#define QUEUE_COS2BW_CFG_REQ_QUEUE_ID0_MIN_BW_BW_VALUE_UNIT_KILO (0x2UL << 29)
-	#define QUEUE_COS2BW_CFG_REQ_QUEUE_ID0_MIN_BW_BW_VALUE_UNIT_BASE (0x4UL << 29)
-	#define QUEUE_COS2BW_CFG_REQ_QUEUE_ID0_MIN_BW_BW_VALUE_UNIT_GIGA (0x6UL << 29)
-	#define QUEUE_COS2BW_CFG_REQ_QUEUE_ID0_MIN_BW_BW_VALUE_UNIT_PERCENT1_100 (0x1UL << 29)
-	#define QUEUE_COS2BW_CFG_REQ_QUEUE_ID0_MIN_BW_BW_VALUE_UNIT_INVALID (0x7UL << 29)
-	#define QUEUE_COS2BW_CFG_REQ_QUEUE_ID0_MIN_BW_BW_VALUE_UNIT_LAST    QUEUE_COS2BW_CFG_REQ_QUEUE_ID0_MIN_BW_BW_VALUE_UNIT_INVALID
-	__le32 queue_id0_max_bw;
-	#define QUEUE_COS2BW_CFG_REQ_QUEUE_ID0_MAX_BW_BW_VALUE_MASK 0xfffffffUL
-	#define QUEUE_COS2BW_CFG_REQ_QUEUE_ID0_MAX_BW_BW_VALUE_SFT 0
-	#define QUEUE_COS2BW_CFG_REQ_QUEUE_ID0_MAX_BW_SCALE	    0x10000000UL
-	#define QUEUE_COS2BW_CFG_REQ_QUEUE_ID0_MAX_BW_SCALE_BITS  (0x0UL << 28)
-	#define QUEUE_COS2BW_CFG_REQ_QUEUE_ID0_MAX_BW_SCALE_BYTES (0x1UL << 28)
-	#define QUEUE_COS2BW_CFG_REQ_QUEUE_ID0_MAX_BW_SCALE_LAST    QUEUE_COS2BW_CFG_REQ_QUEUE_ID0_MAX_BW_SCALE_BYTES
-	#define QUEUE_COS2BW_CFG_REQ_QUEUE_ID0_MAX_BW_BW_VALUE_UNIT_MASK 0xe0000000UL
-	#define QUEUE_COS2BW_CFG_REQ_QUEUE_ID0_MAX_BW_BW_VALUE_UNIT_SFT 29
-	#define QUEUE_COS2BW_CFG_REQ_QUEUE_ID0_MAX_BW_BW_VALUE_UNIT_MEGA (0x0UL << 29)
-	#define QUEUE_COS2BW_CFG_REQ_QUEUE_ID0_MAX_BW_BW_VALUE_UNIT_KILO (0x2UL << 29)
-	#define QUEUE_COS2BW_CFG_REQ_QUEUE_ID0_MAX_BW_BW_VALUE_UNIT_BASE (0x4UL << 29)
-	#define QUEUE_COS2BW_CFG_REQ_QUEUE_ID0_MAX_BW_BW_VALUE_UNIT_GIGA (0x6UL << 29)
-	#define QUEUE_COS2BW_CFG_REQ_QUEUE_ID0_MAX_BW_BW_VALUE_UNIT_PERCENT1_100 (0x1UL << 29)
-	#define QUEUE_COS2BW_CFG_REQ_QUEUE_ID0_MAX_BW_BW_VALUE_UNIT_INVALID (0x7UL << 29)
-	#define QUEUE_COS2BW_CFG_REQ_QUEUE_ID0_MAX_BW_BW_VALUE_UNIT_LAST    QUEUE_COS2BW_CFG_REQ_QUEUE_ID0_MAX_BW_BW_VALUE_UNIT_INVALID
-	u8 queue_id0_tsa_assign;
-	#define QUEUE_COS2BW_CFG_REQ_QUEUE_ID0_TSA_ASSIGN_SP      0x0UL
-	#define QUEUE_COS2BW_CFG_REQ_QUEUE_ID0_TSA_ASSIGN_ETS     0x1UL
-	#define QUEUE_COS2BW_CFG_REQ_QUEUE_ID0_TSA_ASSIGN_RESERVED_FIRST 0x2UL
-	#define QUEUE_COS2BW_CFG_REQ_QUEUE_ID0_TSA_ASSIGN_RESERVED_LAST 0xffUL
-	u8 queue_id0_pri_lvl;
-	u8 queue_id0_bw_weight;
-	u8 queue_id1;
-	__le32 queue_id1_min_bw;
-	#define QUEUE_COS2BW_CFG_REQ_QUEUE_ID1_MIN_BW_BW_VALUE_MASK 0xfffffffUL
-	#define QUEUE_COS2BW_CFG_REQ_QUEUE_ID1_MIN_BW_BW_VALUE_SFT 0
-	#define QUEUE_COS2BW_CFG_REQ_QUEUE_ID1_MIN_BW_SCALE	    0x10000000UL
-	#define QUEUE_COS2BW_CFG_REQ_QUEUE_ID1_MIN_BW_SCALE_BITS  (0x0UL << 28)
-	#define QUEUE_COS2BW_CFG_REQ_QUEUE_ID1_MIN_BW_SCALE_BYTES (0x1UL << 28)
-	#define QUEUE_COS2BW_CFG_REQ_QUEUE_ID1_MIN_BW_SCALE_LAST    QUEUE_COS2BW_CFG_REQ_QUEUE_ID1_MIN_BW_SCALE_BYTES
-	#define QUEUE_COS2BW_CFG_REQ_QUEUE_ID1_MIN_BW_BW_VALUE_UNIT_MASK 0xe0000000UL
-	#define QUEUE_COS2BW_CFG_REQ_QUEUE_ID1_MIN_BW_BW_VALUE_UNIT_SFT 29
-	#define QUEUE_COS2BW_CFG_REQ_QUEUE_ID1_MIN_BW_BW_VALUE_UNIT_MEGA (0x0UL << 29)
-	#define QUEUE_COS2BW_CFG_REQ_QUEUE_ID1_MIN_BW_BW_VALUE_UNIT_KILO (0x2UL << 29)
-	#define QUEUE_COS2BW_CFG_REQ_QUEUE_ID1_MIN_BW_BW_VALUE_UNIT_BASE (0x4UL << 29)
-	#define QUEUE_COS2BW_CFG_REQ_QUEUE_ID1_MIN_BW_BW_VALUE_UNIT_GIGA (0x6UL << 29)
-	#define QUEUE_COS2BW_CFG_REQ_QUEUE_ID1_MIN_BW_BW_VALUE_UNIT_PERCENT1_100 (0x1UL << 29)
-	#define QUEUE_COS2BW_CFG_REQ_QUEUE_ID1_MIN_BW_BW_VALUE_UNIT_INVALID (0x7UL << 29)
-	#define QUEUE_COS2BW_CFG_REQ_QUEUE_ID1_MIN_BW_BW_VALUE_UNIT_LAST    QUEUE_COS2BW_CFG_REQ_QUEUE_ID1_MIN_BW_BW_VALUE_UNIT_INVALID
-	__le32 queue_id1_max_bw;
-	#define QUEUE_COS2BW_CFG_REQ_QUEUE_ID1_MAX_BW_BW_VALUE_MASK 0xfffffffUL
-	#define QUEUE_COS2BW_CFG_REQ_QUEUE_ID1_MAX_BW_BW_VALUE_SFT 0
-	#define QUEUE_COS2BW_CFG_REQ_QUEUE_ID1_MAX_BW_SCALE	    0x10000000UL
-	#define QUEUE_COS2BW_CFG_REQ_QUEUE_ID1_MAX_BW_SCALE_BITS  (0x0UL << 28)
-	#define QUEUE_COS2BW_CFG_REQ_QUEUE_ID1_MAX_BW_SCALE_BYTES (0x1UL << 28)
-	#define QUEUE_COS2BW_CFG_REQ_QUEUE_ID1_MAX_BW_SCALE_LAST    QUEUE_COS2BW_CFG_REQ_QUEUE_ID1_MAX_BW_SCALE_BYTES
-	#define QUEUE_COS2BW_CFG_REQ_QUEUE_ID1_MAX_BW_BW_VALUE_UNIT_MASK 0xe0000000UL
-	#define QUEUE_COS2BW_CFG_REQ_QUEUE_ID1_MAX_BW_BW_VALUE_UNIT_SFT 29
-	#define QUEUE_COS2BW_CFG_REQ_QUEUE_ID1_MAX_BW_BW_VALUE_UNIT_MEGA (0x0UL << 29)
-	#define QUEUE_COS2BW_CFG_REQ_QUEUE_ID1_MAX_BW_BW_VALUE_UNIT_KILO (0x2UL << 29)
-	#define QUEUE_COS2BW_CFG_REQ_QUEUE_ID1_MAX_BW_BW_VALUE_UNIT_BASE (0x4UL << 29)
-	#define QUEUE_COS2BW_CFG_REQ_QUEUE_ID1_MAX_BW_BW_VALUE_UNIT_GIGA (0x6UL << 29)
-	#define QUEUE_COS2BW_CFG_REQ_QUEUE_ID1_MAX_BW_BW_VALUE_UNIT_PERCENT1_100 (0x1UL << 29)
-	#define QUEUE_COS2BW_CFG_REQ_QUEUE_ID1_MAX_BW_BW_VALUE_UNIT_INVALID (0x7UL << 29)
-	#define QUEUE_COS2BW_CFG_REQ_QUEUE_ID1_MAX_BW_BW_VALUE_UNIT_LAST    QUEUE_COS2BW_CFG_REQ_QUEUE_ID1_MAX_BW_BW_VALUE_UNIT_INVALID
-	u8 queue_id1_tsa_assign;
-	#define QUEUE_COS2BW_CFG_REQ_QUEUE_ID1_TSA_ASSIGN_SP      0x0UL
-	#define QUEUE_COS2BW_CFG_REQ_QUEUE_ID1_TSA_ASSIGN_ETS     0x1UL
-	#define QUEUE_COS2BW_CFG_REQ_QUEUE_ID1_TSA_ASSIGN_RESERVED_FIRST 0x2UL
-	#define QUEUE_COS2BW_CFG_REQ_QUEUE_ID1_TSA_ASSIGN_RESERVED_LAST 0xffUL
-	u8 queue_id1_pri_lvl;
-	u8 queue_id1_bw_weight;
-	u8 queue_id2;
-	__le32 queue_id2_min_bw;
-	#define QUEUE_COS2BW_CFG_REQ_QUEUE_ID2_MIN_BW_BW_VALUE_MASK 0xfffffffUL
-	#define QUEUE_COS2BW_CFG_REQ_QUEUE_ID2_MIN_BW_BW_VALUE_SFT 0
-	#define QUEUE_COS2BW_CFG_REQ_QUEUE_ID2_MIN_BW_SCALE	    0x10000000UL
-	#define QUEUE_COS2BW_CFG_REQ_QUEUE_ID2_MIN_BW_SCALE_BITS  (0x0UL << 28)
-	#define QUEUE_COS2BW_CFG_REQ_QUEUE_ID2_MIN_BW_SCALE_BYTES (0x1UL << 28)
-	#define QUEUE_COS2BW_CFG_REQ_QUEUE_ID2_MIN_BW_SCALE_LAST    QUEUE_COS2BW_CFG_REQ_QUEUE_ID2_MIN_BW_SCALE_BYTES
-	#define QUEUE_COS2BW_CFG_REQ_QUEUE_ID2_MIN_BW_BW_VALUE_UNIT_MASK 0xe0000000UL
-	#define QUEUE_COS2BW_CFG_REQ_QUEUE_ID2_MIN_BW_BW_VALUE_UNIT_SFT 29
-	#define QUEUE_COS2BW_CFG_REQ_QUEUE_ID2_MIN_BW_BW_VALUE_UNIT_MEGA (0x0UL << 29)
-	#define QUEUE_COS2BW_CFG_REQ_QUEUE_ID2_MIN_BW_BW_VALUE_UNIT_KILO (0x2UL << 29)
-	#define QUEUE_COS2BW_CFG_REQ_QUEUE_ID2_MIN_BW_BW_VALUE_UNIT_BASE (0x4UL << 29)
-	#define QUEUE_COS2BW_CFG_REQ_QUEUE_ID2_MIN_BW_BW_VALUE_UNIT_GIGA (0x6UL << 29)
-	#define QUEUE_COS2BW_CFG_REQ_QUEUE_ID2_MIN_BW_BW_VALUE_UNIT_PERCENT1_100 (0x1UL << 29)
-	#define QUEUE_COS2BW_CFG_REQ_QUEUE_ID2_MIN_BW_BW_VALUE_UNIT_INVALID (0x7UL << 29)
-	#define QUEUE_COS2BW_CFG_REQ_QUEUE_ID2_MIN_BW_BW_VALUE_UNIT_LAST    QUEUE_COS2BW_CFG_REQ_QUEUE_ID2_MIN_BW_BW_VALUE_UNIT_INVALID
-	__le32 queue_id2_max_bw;
-	#define QUEUE_COS2BW_CFG_REQ_QUEUE_ID2_MAX_BW_BW_VALUE_MASK 0xfffffffUL
-	#define QUEUE_COS2BW_CFG_REQ_QUEUE_ID2_MAX_BW_BW_VALUE_SFT 0
-	#define QUEUE_COS2BW_CFG_REQ_QUEUE_ID2_MAX_BW_SCALE	    0x10000000UL
-	#define QUEUE_COS2BW_CFG_REQ_QUEUE_ID2_MAX_BW_SCALE_BITS  (0x0UL << 28)
-	#define QUEUE_COS2BW_CFG_REQ_QUEUE_ID2_MAX_BW_SCALE_BYTES (0x1UL << 28)
-	#define QUEUE_COS2BW_CFG_REQ_QUEUE_ID2_MAX_BW_SCALE_LAST    QUEUE_COS2BW_CFG_REQ_QUEUE_ID2_MAX_BW_SCALE_BYTES
-	#define QUEUE_COS2BW_CFG_REQ_QUEUE_ID2_MAX_BW_BW_VALUE_UNIT_MASK 0xe0000000UL
-	#define QUEUE_COS2BW_CFG_REQ_QUEUE_ID2_MAX_BW_BW_VALUE_UNIT_SFT 29
-	#define QUEUE_COS2BW_CFG_REQ_QUEUE_ID2_MAX_BW_BW_VALUE_UNIT_MEGA (0x0UL << 29)
-	#define QUEUE_COS2BW_CFG_REQ_QUEUE_ID2_MAX_BW_BW_VALUE_UNIT_KILO (0x2UL << 29)
-	#define QUEUE_COS2BW_CFG_REQ_QUEUE_ID2_MAX_BW_BW_VALUE_UNIT_BASE (0x4UL << 29)
-	#define QUEUE_COS2BW_CFG_REQ_QUEUE_ID2_MAX_BW_BW_VALUE_UNIT_GIGA (0x6UL << 29)
-	#define QUEUE_COS2BW_CFG_REQ_QUEUE_ID2_MAX_BW_BW_VALUE_UNIT_PERCENT1_100 (0x1UL << 29)
-	#define QUEUE_COS2BW_CFG_REQ_QUEUE_ID2_MAX_BW_BW_VALUE_UNIT_INVALID (0x7UL << 29)
-	#define QUEUE_COS2BW_CFG_REQ_QUEUE_ID2_MAX_BW_BW_VALUE_UNIT_LAST    QUEUE_COS2BW_CFG_REQ_QUEUE_ID2_MAX_BW_BW_VALUE_UNIT_INVALID
-	u8 queue_id2_tsa_assign;
-	#define QUEUE_COS2BW_CFG_REQ_QUEUE_ID2_TSA_ASSIGN_SP      0x0UL
-	#define QUEUE_COS2BW_CFG_REQ_QUEUE_ID2_TSA_ASSIGN_ETS     0x1UL
-	#define QUEUE_COS2BW_CFG_REQ_QUEUE_ID2_TSA_ASSIGN_RESERVED_FIRST 0x2UL
-	#define QUEUE_COS2BW_CFG_REQ_QUEUE_ID2_TSA_ASSIGN_RESERVED_LAST 0xffUL
-	u8 queue_id2_pri_lvl;
-	u8 queue_id2_bw_weight;
-	u8 queue_id3;
-	__le32 queue_id3_min_bw;
-	#define QUEUE_COS2BW_CFG_REQ_QUEUE_ID3_MIN_BW_BW_VALUE_MASK 0xfffffffUL
-	#define QUEUE_COS2BW_CFG_REQ_QUEUE_ID3_MIN_BW_BW_VALUE_SFT 0
-	#define QUEUE_COS2BW_CFG_REQ_QUEUE_ID3_MIN_BW_SCALE	    0x10000000UL
-	#define QUEUE_COS2BW_CFG_REQ_QUEUE_ID3_MIN_BW_SCALE_BITS  (0x0UL << 28)
-	#define QUEUE_COS2BW_CFG_REQ_QUEUE_ID3_MIN_BW_SCALE_BYTES (0x1UL << 28)
-	#define QUEUE_COS2BW_CFG_REQ_QUEUE_ID3_MIN_BW_SCALE_LAST    QUEUE_COS2BW_CFG_REQ_QUEUE_ID3_MIN_BW_SCALE_BYTES
-	#define QUEUE_COS2BW_CFG_REQ_QUEUE_ID3_MIN_BW_BW_VALUE_UNIT_MASK 0xe0000000UL
-	#define QUEUE_COS2BW_CFG_REQ_QUEUE_ID3_MIN_BW_BW_VALUE_UNIT_SFT 29
-	#define QUEUE_COS2BW_CFG_REQ_QUEUE_ID3_MIN_BW_BW_VALUE_UNIT_MEGA (0x0UL << 29)
-	#define QUEUE_COS2BW_CFG_REQ_QUEUE_ID3_MIN_BW_BW_VALUE_UNIT_KILO (0x2UL << 29)
-	#define QUEUE_COS2BW_CFG_REQ_QUEUE_ID3_MIN_BW_BW_VALUE_UNIT_BASE (0x4UL << 29)
-	#define QUEUE_COS2BW_CFG_REQ_QUEUE_ID3_MIN_BW_BW_VALUE_UNIT_GIGA (0x6UL << 29)
-	#define QUEUE_COS2BW_CFG_REQ_QUEUE_ID3_MIN_BW_BW_VALUE_UNIT_PERCENT1_100 (0x1UL << 29)
-	#define QUEUE_COS2BW_CFG_REQ_QUEUE_ID3_MIN_BW_BW_VALUE_UNIT_INVALID (0x7UL << 29)
-	#define QUEUE_COS2BW_CFG_REQ_QUEUE_ID3_MIN_BW_BW_VALUE_UNIT_LAST    QUEUE_COS2BW_CFG_REQ_QUEUE_ID3_MIN_BW_BW_VALUE_UNIT_INVALID
-	__le32 queue_id3_max_bw;
-	#define QUEUE_COS2BW_CFG_REQ_QUEUE_ID3_MAX_BW_BW_VALUE_MASK 0xfffffffUL
-	#define QUEUE_COS2BW_CFG_REQ_QUEUE_ID3_MAX_BW_BW_VALUE_SFT 0
-	#define QUEUE_COS2BW_CFG_REQ_QUEUE_ID3_MAX_BW_SCALE	    0x10000000UL
-	#define QUEUE_COS2BW_CFG_REQ_QUEUE_ID3_MAX_BW_SCALE_BITS  (0x0UL << 28)
-	#define QUEUE_COS2BW_CFG_REQ_QUEUE_ID3_MAX_BW_SCALE_BYTES (0x1UL << 28)
-	#define QUEUE_COS2BW_CFG_REQ_QUEUE_ID3_MAX_BW_SCALE_LAST    QUEUE_COS2BW_CFG_REQ_QUEUE_ID3_MAX_BW_SCALE_BYTES
-	#define QUEUE_COS2BW_CFG_REQ_QUEUE_ID3_MAX_BW_BW_VALUE_UNIT_MASK 0xe0000000UL
-	#define QUEUE_COS2BW_CFG_REQ_QUEUE_ID3_MAX_BW_BW_VALUE_UNIT_SFT 29
-	#define QUEUE_COS2BW_CFG_REQ_QUEUE_ID3_MAX_BW_BW_VALUE_UNIT_MEGA (0x0UL << 29)
-	#define QUEUE_COS2BW_CFG_REQ_QUEUE_ID3_MAX_BW_BW_VALUE_UNIT_KILO (0x2UL << 29)
-	#define QUEUE_COS2BW_CFG_REQ_QUEUE_ID3_MAX_BW_BW_VALUE_UNIT_BASE (0x4UL << 29)
-	#define QUEUE_COS2BW_CFG_REQ_QUEUE_ID3_MAX_BW_BW_VALUE_UNIT_GIGA (0x6UL << 29)
-	#define QUEUE_COS2BW_CFG_REQ_QUEUE_ID3_MAX_BW_BW_VALUE_UNIT_PERCENT1_100 (0x1UL << 29)
-	#define QUEUE_COS2BW_CFG_REQ_QUEUE_ID3_MAX_BW_BW_VALUE_UNIT_INVALID (0x7UL << 29)
-	#define QUEUE_COS2BW_CFG_REQ_QUEUE_ID3_MAX_BW_BW_VALUE_UNIT_LAST    QUEUE_COS2BW_CFG_REQ_QUEUE_ID3_MAX_BW_BW_VALUE_UNIT_INVALID
-	u8 queue_id3_tsa_assign;
-	#define QUEUE_COS2BW_CFG_REQ_QUEUE_ID3_TSA_ASSIGN_SP      0x0UL
-	#define QUEUE_COS2BW_CFG_REQ_QUEUE_ID3_TSA_ASSIGN_ETS     0x1UL
-	#define QUEUE_COS2BW_CFG_REQ_QUEUE_ID3_TSA_ASSIGN_RESERVED_FIRST 0x2UL
-	#define QUEUE_COS2BW_CFG_REQ_QUEUE_ID3_TSA_ASSIGN_RESERVED_LAST 0xffUL
-	u8 queue_id3_pri_lvl;
-	u8 queue_id3_bw_weight;
-	u8 queue_id4;
-	__le32 queue_id4_min_bw;
-	#define QUEUE_COS2BW_CFG_REQ_QUEUE_ID4_MIN_BW_BW_VALUE_MASK 0xfffffffUL
-	#define QUEUE_COS2BW_CFG_REQ_QUEUE_ID4_MIN_BW_BW_VALUE_SFT 0
-	#define QUEUE_COS2BW_CFG_REQ_QUEUE_ID4_MIN_BW_SCALE	    0x10000000UL
-	#define QUEUE_COS2BW_CFG_REQ_QUEUE_ID4_MIN_BW_SCALE_BITS  (0x0UL << 28)
-	#define QUEUE_COS2BW_CFG_REQ_QUEUE_ID4_MIN_BW_SCALE_BYTES (0x1UL << 28)
-	#define QUEUE_COS2BW_CFG_REQ_QUEUE_ID4_MIN_BW_SCALE_LAST    QUEUE_COS2BW_CFG_REQ_QUEUE_ID4_MIN_BW_SCALE_BYTES
-	#define QUEUE_COS2BW_CFG_REQ_QUEUE_ID4_MIN_BW_BW_VALUE_UNIT_MASK 0xe0000000UL
-	#define QUEUE_COS2BW_CFG_REQ_QUEUE_ID4_MIN_BW_BW_VALUE_UNIT_SFT 29
-	#define QUEUE_COS2BW_CFG_REQ_QUEUE_ID4_MIN_BW_BW_VALUE_UNIT_MEGA (0x0UL << 29)
-	#define QUEUE_COS2BW_CFG_REQ_QUEUE_ID4_MIN_BW_BW_VALUE_UNIT_KILO (0x2UL << 29)
-	#define QUEUE_COS2BW_CFG_REQ_QUEUE_ID4_MIN_BW_BW_VALUE_UNIT_BASE (0x4UL << 29)
-	#define QUEUE_COS2BW_CFG_REQ_QUEUE_ID4_MIN_BW_BW_VALUE_UNIT_GIGA (0x6UL << 29)
-	#define QUEUE_COS2BW_CFG_REQ_QUEUE_ID4_MIN_BW_BW_VALUE_UNIT_PERCENT1_100 (0x1UL << 29)
-	#define QUEUE_COS2BW_CFG_REQ_QUEUE_ID4_MIN_BW_BW_VALUE_UNIT_INVALID (0x7UL << 29)
-	#define QUEUE_COS2BW_CFG_REQ_QUEUE_ID4_MIN_BW_BW_VALUE_UNIT_LAST    QUEUE_COS2BW_CFG_REQ_QUEUE_ID4_MIN_BW_BW_VALUE_UNIT_INVALID
-	__le32 queue_id4_max_bw;
-	#define QUEUE_COS2BW_CFG_REQ_QUEUE_ID4_MAX_BW_BW_VALUE_MASK 0xfffffffUL
-	#define QUEUE_COS2BW_CFG_REQ_QUEUE_ID4_MAX_BW_BW_VALUE_SFT 0
-	#define QUEUE_COS2BW_CFG_REQ_QUEUE_ID4_MAX_BW_SCALE	    0x10000000UL
-	#define QUEUE_COS2BW_CFG_REQ_QUEUE_ID4_MAX_BW_SCALE_BITS  (0x0UL << 28)
-	#define QUEUE_COS2BW_CFG_REQ_QUEUE_ID4_MAX_BW_SCALE_BYTES (0x1UL << 28)
-	#define QUEUE_COS2BW_CFG_REQ_QUEUE_ID4_MAX_BW_SCALE_LAST    QUEUE_COS2BW_CFG_REQ_QUEUE_ID4_MAX_BW_SCALE_BYTES
-	#define QUEUE_COS2BW_CFG_REQ_QUEUE_ID4_MAX_BW_BW_VALUE_UNIT_MASK 0xe0000000UL
-	#define QUEUE_COS2BW_CFG_REQ_QUEUE_ID4_MAX_BW_BW_VALUE_UNIT_SFT 29
-	#define QUEUE_COS2BW_CFG_REQ_QUEUE_ID4_MAX_BW_BW_VALUE_UNIT_MEGA (0x0UL << 29)
-	#define QUEUE_COS2BW_CFG_REQ_QUEUE_ID4_MAX_BW_BW_VALUE_UNIT_KILO (0x2UL << 29)
-	#define QUEUE_COS2BW_CFG_REQ_QUEUE_ID4_MAX_BW_BW_VALUE_UNIT_BASE (0x4UL << 29)
-	#define QUEUE_COS2BW_CFG_REQ_QUEUE_ID4_MAX_BW_BW_VALUE_UNIT_GIGA (0x6UL << 29)
-	#define QUEUE_COS2BW_CFG_REQ_QUEUE_ID4_MAX_BW_BW_VALUE_UNIT_PERCENT1_100 (0x1UL << 29)
-	#define QUEUE_COS2BW_CFG_REQ_QUEUE_ID4_MAX_BW_BW_VALUE_UNIT_INVALID (0x7UL << 29)
-	#define QUEUE_COS2BW_CFG_REQ_QUEUE_ID4_MAX_BW_BW_VALUE_UNIT_LAST    QUEUE_COS2BW_CFG_REQ_QUEUE_ID4_MAX_BW_BW_VALUE_UNIT_INVALID
-	u8 queue_id4_tsa_assign;
-	#define QUEUE_COS2BW_CFG_REQ_QUEUE_ID4_TSA_ASSIGN_SP      0x0UL
-	#define QUEUE_COS2BW_CFG_REQ_QUEUE_ID4_TSA_ASSIGN_ETS     0x1UL
-	#define QUEUE_COS2BW_CFG_REQ_QUEUE_ID4_TSA_ASSIGN_RESERVED_FIRST 0x2UL
-	#define QUEUE_COS2BW_CFG_REQ_QUEUE_ID4_TSA_ASSIGN_RESERVED_LAST 0xffUL
-	u8 queue_id4_pri_lvl;
-	u8 queue_id4_bw_weight;
-	u8 queue_id5;
-	__le32 queue_id5_min_bw;
-	#define QUEUE_COS2BW_CFG_REQ_QUEUE_ID5_MIN_BW_BW_VALUE_MASK 0xfffffffUL
-	#define QUEUE_COS2BW_CFG_REQ_QUEUE_ID5_MIN_BW_BW_VALUE_SFT 0
-	#define QUEUE_COS2BW_CFG_REQ_QUEUE_ID5_MIN_BW_SCALE	    0x10000000UL
-	#define QUEUE_COS2BW_CFG_REQ_QUEUE_ID5_MIN_BW_SCALE_BITS  (0x0UL << 28)
-	#define QUEUE_COS2BW_CFG_REQ_QUEUE_ID5_MIN_BW_SCALE_BYTES (0x1UL << 28)
-	#define QUEUE_COS2BW_CFG_REQ_QUEUE_ID5_MIN_BW_SCALE_LAST    QUEUE_COS2BW_CFG_REQ_QUEUE_ID5_MIN_BW_SCALE_BYTES
-	#define QUEUE_COS2BW_CFG_REQ_QUEUE_ID5_MIN_BW_BW_VALUE_UNIT_MASK 0xe0000000UL
-	#define QUEUE_COS2BW_CFG_REQ_QUEUE_ID5_MIN_BW_BW_VALUE_UNIT_SFT 29
-	#define QUEUE_COS2BW_CFG_REQ_QUEUE_ID5_MIN_BW_BW_VALUE_UNIT_MEGA (0x0UL << 29)
-	#define QUEUE_COS2BW_CFG_REQ_QUEUE_ID5_MIN_BW_BW_VALUE_UNIT_KILO (0x2UL << 29)
-	#define QUEUE_COS2BW_CFG_REQ_QUEUE_ID5_MIN_BW_BW_VALUE_UNIT_BASE (0x4UL << 29)
-	#define QUEUE_COS2BW_CFG_REQ_QUEUE_ID5_MIN_BW_BW_VALUE_UNIT_GIGA (0x6UL << 29)
-	#define QUEUE_COS2BW_CFG_REQ_QUEUE_ID5_MIN_BW_BW_VALUE_UNIT_PERCENT1_100 (0x1UL << 29)
-	#define QUEUE_COS2BW_CFG_REQ_QUEUE_ID5_MIN_BW_BW_VALUE_UNIT_INVALID (0x7UL << 29)
-	#define QUEUE_COS2BW_CFG_REQ_QUEUE_ID5_MIN_BW_BW_VALUE_UNIT_LAST    QUEUE_COS2BW_CFG_REQ_QUEUE_ID5_MIN_BW_BW_VALUE_UNIT_INVALID
-	__le32 queue_id5_max_bw;
-	#define QUEUE_COS2BW_CFG_REQ_QUEUE_ID5_MAX_BW_BW_VALUE_MASK 0xfffffffUL
-	#define QUEUE_COS2BW_CFG_REQ_QUEUE_ID5_MAX_BW_BW_VALUE_SFT 0
-	#define QUEUE_COS2BW_CFG_REQ_QUEUE_ID5_MAX_BW_SCALE	    0x10000000UL
-	#define QUEUE_COS2BW_CFG_REQ_QUEUE_ID5_MAX_BW_SCALE_BITS  (0x0UL << 28)
-	#define QUEUE_COS2BW_CFG_REQ_QUEUE_ID5_MAX_BW_SCALE_BYTES (0x1UL << 28)
-	#define QUEUE_COS2BW_CFG_REQ_QUEUE_ID5_MAX_BW_SCALE_LAST    QUEUE_COS2BW_CFG_REQ_QUEUE_ID5_MAX_BW_SCALE_BYTES
-	#define QUEUE_COS2BW_CFG_REQ_QUEUE_ID5_MAX_BW_BW_VALUE_UNIT_MASK 0xe0000000UL
-	#define QUEUE_COS2BW_CFG_REQ_QUEUE_ID5_MAX_BW_BW_VALUE_UNIT_SFT 29
-	#define QUEUE_COS2BW_CFG_REQ_QUEUE_ID5_MAX_BW_BW_VALUE_UNIT_MEGA (0x0UL << 29)
-	#define QUEUE_COS2BW_CFG_REQ_QUEUE_ID5_MAX_BW_BW_VALUE_UNIT_KILO (0x2UL << 29)
-	#define QUEUE_COS2BW_CFG_REQ_QUEUE_ID5_MAX_BW_BW_VALUE_UNIT_BASE (0x4UL << 29)
-	#define QUEUE_COS2BW_CFG_REQ_QUEUE_ID5_MAX_BW_BW_VALUE_UNIT_GIGA (0x6UL << 29)
-	#define QUEUE_COS2BW_CFG_REQ_QUEUE_ID5_MAX_BW_BW_VALUE_UNIT_PERCENT1_100 (0x1UL << 29)
-	#define QUEUE_COS2BW_CFG_REQ_QUEUE_ID5_MAX_BW_BW_VALUE_UNIT_INVALID (0x7UL << 29)
-	#define QUEUE_COS2BW_CFG_REQ_QUEUE_ID5_MAX_BW_BW_VALUE_UNIT_LAST    QUEUE_COS2BW_CFG_REQ_QUEUE_ID5_MAX_BW_BW_VALUE_UNIT_INVALID
-	u8 queue_id5_tsa_assign;
-	#define QUEUE_COS2BW_CFG_REQ_QUEUE_ID5_TSA_ASSIGN_SP      0x0UL
-	#define QUEUE_COS2BW_CFG_REQ_QUEUE_ID5_TSA_ASSIGN_ETS     0x1UL
-	#define QUEUE_COS2BW_CFG_REQ_QUEUE_ID5_TSA_ASSIGN_RESERVED_FIRST 0x2UL
-	#define QUEUE_COS2BW_CFG_REQ_QUEUE_ID5_TSA_ASSIGN_RESERVED_LAST 0xffUL
-	u8 queue_id5_pri_lvl;
-	u8 queue_id5_bw_weight;
-	u8 queue_id6;
-	__le32 queue_id6_min_bw;
-	#define QUEUE_COS2BW_CFG_REQ_QUEUE_ID6_MIN_BW_BW_VALUE_MASK 0xfffffffUL
-	#define QUEUE_COS2BW_CFG_REQ_QUEUE_ID6_MIN_BW_BW_VALUE_SFT 0
-	#define QUEUE_COS2BW_CFG_REQ_QUEUE_ID6_MIN_BW_SCALE	    0x10000000UL
-	#define QUEUE_COS2BW_CFG_REQ_QUEUE_ID6_MIN_BW_SCALE_BITS  (0x0UL << 28)
-	#define QUEUE_COS2BW_CFG_REQ_QUEUE_ID6_MIN_BW_SCALE_BYTES (0x1UL << 28)
-	#define QUEUE_COS2BW_CFG_REQ_QUEUE_ID6_MIN_BW_SCALE_LAST    QUEUE_COS2BW_CFG_REQ_QUEUE_ID6_MIN_BW_SCALE_BYTES
-	#define QUEUE_COS2BW_CFG_REQ_QUEUE_ID6_MIN_BW_BW_VALUE_UNIT_MASK 0xe0000000UL
-	#define QUEUE_COS2BW_CFG_REQ_QUEUE_ID6_MIN_BW_BW_VALUE_UNIT_SFT 29
-	#define QUEUE_COS2BW_CFG_REQ_QUEUE_ID6_MIN_BW_BW_VALUE_UNIT_MEGA (0x0UL << 29)
-	#define QUEUE_COS2BW_CFG_REQ_QUEUE_ID6_MIN_BW_BW_VALUE_UNIT_KILO (0x2UL << 29)
-	#define QUEUE_COS2BW_CFG_REQ_QUEUE_ID6_MIN_BW_BW_VALUE_UNIT_BASE (0x4UL << 29)
-	#define QUEUE_COS2BW_CFG_REQ_QUEUE_ID6_MIN_BW_BW_VALUE_UNIT_GIGA (0x6UL << 29)
-	#define QUEUE_COS2BW_CFG_REQ_QUEUE_ID6_MIN_BW_BW_VALUE_UNIT_PERCENT1_100 (0x1UL << 29)
-	#define QUEUE_COS2BW_CFG_REQ_QUEUE_ID6_MIN_BW_BW_VALUE_UNIT_INVALID (0x7UL << 29)
-	#define QUEUE_COS2BW_CFG_REQ_QUEUE_ID6_MIN_BW_BW_VALUE_UNIT_LAST    QUEUE_COS2BW_CFG_REQ_QUEUE_ID6_MIN_BW_BW_VALUE_UNIT_INVALID
-	__le32 queue_id6_max_bw;
-	#define QUEUE_COS2BW_CFG_REQ_QUEUE_ID6_MAX_BW_BW_VALUE_MASK 0xfffffffUL
-	#define QUEUE_COS2BW_CFG_REQ_QUEUE_ID6_MAX_BW_BW_VALUE_SFT 0
-	#define QUEUE_COS2BW_CFG_REQ_QUEUE_ID6_MAX_BW_SCALE	    0x10000000UL
-	#define QUEUE_COS2BW_CFG_REQ_QUEUE_ID6_MAX_BW_SCALE_BITS  (0x0UL << 28)
-	#define QUEUE_COS2BW_CFG_REQ_QUEUE_ID6_MAX_BW_SCALE_BYTES (0x1UL << 28)
-	#define QUEUE_COS2BW_CFG_REQ_QUEUE_ID6_MAX_BW_SCALE_LAST    QUEUE_COS2BW_CFG_REQ_QUEUE_ID6_MAX_BW_SCALE_BYTES
-	#define QUEUE_COS2BW_CFG_REQ_QUEUE_ID6_MAX_BW_BW_VALUE_UNIT_MASK 0xe0000000UL
-	#define QUEUE_COS2BW_CFG_REQ_QUEUE_ID6_MAX_BW_BW_VALUE_UNIT_SFT 29
-	#define QUEUE_COS2BW_CFG_REQ_QUEUE_ID6_MAX_BW_BW_VALUE_UNIT_MEGA (0x0UL << 29)
-	#define QUEUE_COS2BW_CFG_REQ_QUEUE_ID6_MAX_BW_BW_VALUE_UNIT_KILO (0x2UL << 29)
-	#define QUEUE_COS2BW_CFG_REQ_QUEUE_ID6_MAX_BW_BW_VALUE_UNIT_BASE (0x4UL << 29)
-	#define QUEUE_COS2BW_CFG_REQ_QUEUE_ID6_MAX_BW_BW_VALUE_UNIT_GIGA (0x6UL << 29)
-	#define QUEUE_COS2BW_CFG_REQ_QUEUE_ID6_MAX_BW_BW_VALUE_UNIT_PERCENT1_100 (0x1UL << 29)
-	#define QUEUE_COS2BW_CFG_REQ_QUEUE_ID6_MAX_BW_BW_VALUE_UNIT_INVALID (0x7UL << 29)
-	#define QUEUE_COS2BW_CFG_REQ_QUEUE_ID6_MAX_BW_BW_VALUE_UNIT_LAST    QUEUE_COS2BW_CFG_REQ_QUEUE_ID6_MAX_BW_BW_VALUE_UNIT_INVALID
-	u8 queue_id6_tsa_assign;
-	#define QUEUE_COS2BW_CFG_REQ_QUEUE_ID6_TSA_ASSIGN_SP      0x0UL
-	#define QUEUE_COS2BW_CFG_REQ_QUEUE_ID6_TSA_ASSIGN_ETS     0x1UL
-	#define QUEUE_COS2BW_CFG_REQ_QUEUE_ID6_TSA_ASSIGN_RESERVED_FIRST 0x2UL
-	#define QUEUE_COS2BW_CFG_REQ_QUEUE_ID6_TSA_ASSIGN_RESERVED_LAST 0xffUL
-	u8 queue_id6_pri_lvl;
-	u8 queue_id6_bw_weight;
-	u8 queue_id7;
-	__le32 queue_id7_min_bw;
-	#define QUEUE_COS2BW_CFG_REQ_QUEUE_ID7_MIN_BW_BW_VALUE_MASK 0xfffffffUL
-	#define QUEUE_COS2BW_CFG_REQ_QUEUE_ID7_MIN_BW_BW_VALUE_SFT 0
-	#define QUEUE_COS2BW_CFG_REQ_QUEUE_ID7_MIN_BW_SCALE	    0x10000000UL
-	#define QUEUE_COS2BW_CFG_REQ_QUEUE_ID7_MIN_BW_SCALE_BITS  (0x0UL << 28)
-	#define QUEUE_COS2BW_CFG_REQ_QUEUE_ID7_MIN_BW_SCALE_BYTES (0x1UL << 28)
-	#define QUEUE_COS2BW_CFG_REQ_QUEUE_ID7_MIN_BW_SCALE_LAST    QUEUE_COS2BW_CFG_REQ_QUEUE_ID7_MIN_BW_SCALE_BYTES
-	#define QUEUE_COS2BW_CFG_REQ_QUEUE_ID7_MIN_BW_BW_VALUE_UNIT_MASK 0xe0000000UL
-	#define QUEUE_COS2BW_CFG_REQ_QUEUE_ID7_MIN_BW_BW_VALUE_UNIT_SFT 29
-	#define QUEUE_COS2BW_CFG_REQ_QUEUE_ID7_MIN_BW_BW_VALUE_UNIT_MEGA (0x0UL << 29)
-	#define QUEUE_COS2BW_CFG_REQ_QUEUE_ID7_MIN_BW_BW_VALUE_UNIT_KILO (0x2UL << 29)
-	#define QUEUE_COS2BW_CFG_REQ_QUEUE_ID7_MIN_BW_BW_VALUE_UNIT_BASE (0x4UL << 29)
-	#define QUEUE_COS2BW_CFG_REQ_QUEUE_ID7_MIN_BW_BW_VALUE_UNIT_GIGA (0x6UL << 29)
-	#define QUEUE_COS2BW_CFG_REQ_QUEUE_ID7_MIN_BW_BW_VALUE_UNIT_PERCENT1_100 (0x1UL << 29)
-	#define QUEUE_COS2BW_CFG_REQ_QUEUE_ID7_MIN_BW_BW_VALUE_UNIT_INVALID (0x7UL << 29)
-	#define QUEUE_COS2BW_CFG_REQ_QUEUE_ID7_MIN_BW_BW_VALUE_UNIT_LAST    QUEUE_COS2BW_CFG_REQ_QUEUE_ID7_MIN_BW_BW_VALUE_UNIT_INVALID
-	__le32 queue_id7_max_bw;
-	#define QUEUE_COS2BW_CFG_REQ_QUEUE_ID7_MAX_BW_BW_VALUE_MASK 0xfffffffUL
-	#define QUEUE_COS2BW_CFG_REQ_QUEUE_ID7_MAX_BW_BW_VALUE_SFT 0
-	#define QUEUE_COS2BW_CFG_REQ_QUEUE_ID7_MAX_BW_SCALE	    0x10000000UL
-	#define QUEUE_COS2BW_CFG_REQ_QUEUE_ID7_MAX_BW_SCALE_BITS  (0x0UL << 28)
-	#define QUEUE_COS2BW_CFG_REQ_QUEUE_ID7_MAX_BW_SCALE_BYTES (0x1UL << 28)
-	#define QUEUE_COS2BW_CFG_REQ_QUEUE_ID7_MAX_BW_SCALE_LAST    QUEUE_COS2BW_CFG_REQ_QUEUE_ID7_MAX_BW_SCALE_BYTES
-	#define QUEUE_COS2BW_CFG_REQ_QUEUE_ID7_MAX_BW_BW_VALUE_UNIT_MASK 0xe0000000UL
-	#define QUEUE_COS2BW_CFG_REQ_QUEUE_ID7_MAX_BW_BW_VALUE_UNIT_SFT 29
-	#define QUEUE_COS2BW_CFG_REQ_QUEUE_ID7_MAX_BW_BW_VALUE_UNIT_MEGA (0x0UL << 29)
-	#define QUEUE_COS2BW_CFG_REQ_QUEUE_ID7_MAX_BW_BW_VALUE_UNIT_KILO (0x2UL << 29)
-	#define QUEUE_COS2BW_CFG_REQ_QUEUE_ID7_MAX_BW_BW_VALUE_UNIT_BASE (0x4UL << 29)
-	#define QUEUE_COS2BW_CFG_REQ_QUEUE_ID7_MAX_BW_BW_VALUE_UNIT_GIGA (0x6UL << 29)
-	#define QUEUE_COS2BW_CFG_REQ_QUEUE_ID7_MAX_BW_BW_VALUE_UNIT_PERCENT1_100 (0x1UL << 29)
-	#define QUEUE_COS2BW_CFG_REQ_QUEUE_ID7_MAX_BW_BW_VALUE_UNIT_INVALID (0x7UL << 29)
-	#define QUEUE_COS2BW_CFG_REQ_QUEUE_ID7_MAX_BW_BW_VALUE_UNIT_LAST    QUEUE_COS2BW_CFG_REQ_QUEUE_ID7_MAX_BW_BW_VALUE_UNIT_INVALID
-	u8 queue_id7_tsa_assign;
-	#define QUEUE_COS2BW_CFG_REQ_QUEUE_ID7_TSA_ASSIGN_SP      0x0UL
-	#define QUEUE_COS2BW_CFG_REQ_QUEUE_ID7_TSA_ASSIGN_ETS     0x1UL
-	#define QUEUE_COS2BW_CFG_REQ_QUEUE_ID7_TSA_ASSIGN_RESERVED_FIRST 0x2UL
-	#define QUEUE_COS2BW_CFG_REQ_QUEUE_ID7_TSA_ASSIGN_RESERVED_LAST 0xffUL
-	u8 queue_id7_pri_lvl;
-	u8 queue_id7_bw_weight;
-	u8 unused_1[5];
-};
-
-/* Output (16 bytes) */
-struct hwrm_queue_cos2bw_cfg_output {
-	__le16 error_code;
-	__le16 req_type;
-	__le16 seq_id;
-	__le16 resp_len;
-	__le32 unused_0;
-	u8 unused_1;
-	u8 unused_2;
-	u8 unused_3;
-	u8 valid;
-};
-
-/* hwrm_vnic_alloc */
-/* Input (24 bytes) */
-struct hwrm_vnic_alloc_input {
-	__le16 req_type;
-	__le16 cmpl_ring;
-	__le16 seq_id;
-	__le16 target_id;
-	__le64 resp_addr;
-	__le32 flags;
-	#define VNIC_ALLOC_REQ_FLAGS_DEFAULT			    0x1UL
-	__le32 unused_0;
-};
-
-/* Output (16 bytes) */
-struct hwrm_vnic_alloc_output {
-	__le16 error_code;
-	__le16 req_type;
-	__le16 seq_id;
-	__le16 resp_len;
-	__le32 vnic_id;
-	u8 unused_0;
-	u8 unused_1;
-	u8 unused_2;
-	u8 valid;
-};
-
-/* hwrm_vnic_free */
-/* Input (24 bytes) */
-struct hwrm_vnic_free_input {
-	__le16 req_type;
-	__le16 cmpl_ring;
-	__le16 seq_id;
-	__le16 target_id;
-	__le64 resp_addr;
-	__le32 vnic_id;
-	__le32 unused_0;
-};
-
-/* Output (16 bytes) */
-struct hwrm_vnic_free_output {
-	__le16 error_code;
-	__le16 req_type;
-	__le16 seq_id;
-	__le16 resp_len;
-	__le32 unused_0;
-	u8 unused_1;
-	u8 unused_2;
-	u8 unused_3;
-	u8 valid;
-};
-
-/* hwrm_vnic_cfg */
-/* Input (40 bytes) */
-struct hwrm_vnic_cfg_input {
-	__le16 req_type;
-	__le16 cmpl_ring;
-	__le16 seq_id;
-	__le16 target_id;
-	__le64 resp_addr;
-	__le32 flags;
-	#define VNIC_CFG_REQ_FLAGS_DEFAULT			    0x1UL
-	#define VNIC_CFG_REQ_FLAGS_VLAN_STRIP_MODE		    0x2UL
-	#define VNIC_CFG_REQ_FLAGS_BD_STALL_MODE		    0x4UL
-	#define VNIC_CFG_REQ_FLAGS_ROCE_DUAL_VNIC_MODE		    0x8UL
-	#define VNIC_CFG_REQ_FLAGS_ROCE_ONLY_VNIC_MODE		    0x10UL
-	#define VNIC_CFG_REQ_FLAGS_RSS_DFLT_CR_MODE		    0x20UL
-	__le32 enables;
-	#define VNIC_CFG_REQ_ENABLES_DFLT_RING_GRP		    0x1UL
-	#define VNIC_CFG_REQ_ENABLES_RSS_RULE			    0x2UL
-	#define VNIC_CFG_REQ_ENABLES_COS_RULE			    0x4UL
-	#define VNIC_CFG_REQ_ENABLES_LB_RULE			    0x8UL
-	#define VNIC_CFG_REQ_ENABLES_MRU			    0x10UL
-	__le16 vnic_id;
-	__le16 dflt_ring_grp;
-	__le16 rss_rule;
-	__le16 cos_rule;
-	__le16 lb_rule;
-	__le16 mru;
-	__le32 unused_0;
-};
-
-/* Output (16 bytes) */
-struct hwrm_vnic_cfg_output {
-	__le16 error_code;
-	__le16 req_type;
-	__le16 seq_id;
-	__le16 resp_len;
-	__le32 unused_0;
-	u8 unused_1;
-	u8 unused_2;
-	u8 unused_3;
-	u8 valid;
-};
-
-/* hwrm_vnic_qcaps */
-/* Input (24 bytes) */
-struct hwrm_vnic_qcaps_input {
-	__le16 req_type;
-	__le16 cmpl_ring;
-	__le16 seq_id;
-	__le16 target_id;
-	__le64 resp_addr;
-	__le32 enables;
-	__le32 unused_0;
-};
-
-/* Output (24 bytes) */
-struct hwrm_vnic_qcaps_output {
-	__le16 error_code;
-	__le16 req_type;
-	__le16 seq_id;
-	__le16 resp_len;
-	__le16 mru;
-	u8 unused_0;
-	u8 unused_1;
-	__le32 flags;
-	#define VNIC_QCAPS_RESP_FLAGS_UNUSED			    0x1UL
-	#define VNIC_QCAPS_RESP_FLAGS_VLAN_STRIP_CAP		    0x2UL
-	#define VNIC_QCAPS_RESP_FLAGS_BD_STALL_CAP		    0x4UL
-	#define VNIC_QCAPS_RESP_FLAGS_ROCE_DUAL_VNIC_CAP	    0x8UL
-	#define VNIC_QCAPS_RESP_FLAGS_ROCE_ONLY_VNIC_CAP	    0x10UL
-	#define VNIC_QCAPS_RESP_FLAGS_RSS_DFLT_CR_CAP		    0x20UL
-	__le32 unused_2;
-	u8 unused_3;
-	u8 unused_4;
-	u8 unused_5;
-	u8 valid;
-};
-
-/* hwrm_vnic_tpa_cfg */
-/* Input (40 bytes) */
-struct hwrm_vnic_tpa_cfg_input {
-	__le16 req_type;
-	__le16 cmpl_ring;
-	__le16 seq_id;
-	__le16 target_id;
-	__le64 resp_addr;
-	__le32 flags;
-	#define VNIC_TPA_CFG_REQ_FLAGS_TPA			    0x1UL
-	#define VNIC_TPA_CFG_REQ_FLAGS_ENCAP_TPA		    0x2UL
-	#define VNIC_TPA_CFG_REQ_FLAGS_RSC_WND_UPDATE		    0x4UL
-	#define VNIC_TPA_CFG_REQ_FLAGS_GRO			    0x8UL
-	#define VNIC_TPA_CFG_REQ_FLAGS_AGG_WITH_ECN		    0x10UL
-	#define VNIC_TPA_CFG_REQ_FLAGS_AGG_WITH_SAME_GRE_SEQ       0x20UL
-	#define VNIC_TPA_CFG_REQ_FLAGS_GRO_IPID_CHECK		    0x40UL
-	#define VNIC_TPA_CFG_REQ_FLAGS_GRO_TTL_CHECK		    0x80UL
-	__le32 enables;
-	#define VNIC_TPA_CFG_REQ_ENABLES_MAX_AGG_SEGS		    0x1UL
-	#define VNIC_TPA_CFG_REQ_ENABLES_MAX_AGGS		    0x2UL
-	#define VNIC_TPA_CFG_REQ_ENABLES_MAX_AGG_TIMER		    0x4UL
-	#define VNIC_TPA_CFG_REQ_ENABLES_MIN_AGG_LEN		    0x8UL
-	__le16 vnic_id;
-	__le16 max_agg_segs;
-	#define VNIC_TPA_CFG_REQ_MAX_AGG_SEGS_1		   0x0UL
-	#define VNIC_TPA_CFG_REQ_MAX_AGG_SEGS_2		   0x1UL
-	#define VNIC_TPA_CFG_REQ_MAX_AGG_SEGS_4		   0x2UL
-	#define VNIC_TPA_CFG_REQ_MAX_AGG_SEGS_8		   0x3UL
-	#define VNIC_TPA_CFG_REQ_MAX_AGG_SEGS_MAX		   0x1fUL
-	__le16 max_aggs;
-	#define VNIC_TPA_CFG_REQ_MAX_AGGS_1			   0x0UL
-	#define VNIC_TPA_CFG_REQ_MAX_AGGS_2			   0x1UL
-	#define VNIC_TPA_CFG_REQ_MAX_AGGS_4			   0x2UL
-	#define VNIC_TPA_CFG_REQ_MAX_AGGS_8			   0x3UL
-	#define VNIC_TPA_CFG_REQ_MAX_AGGS_16			   0x4UL
-	#define VNIC_TPA_CFG_REQ_MAX_AGGS_MAX			   0x7UL
-	u8 unused_0;
-	u8 unused_1;
-	__le32 max_agg_timer;
-	__le32 min_agg_len;
-};
-
-/* Output (16 bytes) */
-struct hwrm_vnic_tpa_cfg_output {
-	__le16 error_code;
-	__le16 req_type;
-	__le16 seq_id;
-	__le16 resp_len;
-	__le32 unused_0;
-	u8 unused_1;
-	u8 unused_2;
-	u8 unused_3;
-	u8 valid;
-};
-
-/* hwrm_vnic_rss_cfg */
-/* Input (48 bytes) */
-struct hwrm_vnic_rss_cfg_input {
-	__le16 req_type;
-	__le16 cmpl_ring;
-	__le16 seq_id;
-	__le16 target_id;
-	__le64 resp_addr;
-	__le32 hash_type;
-	#define VNIC_RSS_CFG_REQ_HASH_TYPE_IPV4		    0x1UL
-	#define VNIC_RSS_CFG_REQ_HASH_TYPE_TCP_IPV4		    0x2UL
-	#define VNIC_RSS_CFG_REQ_HASH_TYPE_UDP_IPV4		    0x4UL
-	#define VNIC_RSS_CFG_REQ_HASH_TYPE_IPV6		    0x8UL
-	#define VNIC_RSS_CFG_REQ_HASH_TYPE_TCP_IPV6		    0x10UL
-	#define VNIC_RSS_CFG_REQ_HASH_TYPE_UDP_IPV6		    0x20UL
-	__le32 unused_0;
-	__le64 ring_grp_tbl_addr;
-	__le64 hash_key_tbl_addr;
-	__le16 rss_ctx_idx;
-	__le16 unused_1[3];
-};
-
-/* Output (16 bytes) */
-struct hwrm_vnic_rss_cfg_output {
-	__le16 error_code;
-	__le16 req_type;
-	__le16 seq_id;
-	__le16 resp_len;
-	__le32 unused_0;
-	u8 unused_1;
-	u8 unused_2;
-	u8 unused_3;
-	u8 valid;
-};
-
-/* hwrm_vnic_plcmodes_cfg */
-/* Input (40 bytes) */
-struct hwrm_vnic_plcmodes_cfg_input {
-	__le16 req_type;
-	__le16 cmpl_ring;
-	__le16 seq_id;
-	__le16 target_id;
-	__le64 resp_addr;
-	__le32 flags;
-	#define VNIC_PLCMODES_CFG_REQ_FLAGS_REGULAR_PLACEMENT      0x1UL
-	#define VNIC_PLCMODES_CFG_REQ_FLAGS_JUMBO_PLACEMENT	    0x2UL
-	#define VNIC_PLCMODES_CFG_REQ_FLAGS_HDS_IPV4		    0x4UL
-	#define VNIC_PLCMODES_CFG_REQ_FLAGS_HDS_IPV6		    0x8UL
-	#define VNIC_PLCMODES_CFG_REQ_FLAGS_HDS_FCOE		    0x10UL
-	#define VNIC_PLCMODES_CFG_REQ_FLAGS_HDS_ROCE		    0x20UL
-	__le32 enables;
-	#define VNIC_PLCMODES_CFG_REQ_ENABLES_JUMBO_THRESH_VALID   0x1UL
-	#define VNIC_PLCMODES_CFG_REQ_ENABLES_HDS_OFFSET_VALID     0x2UL
-	#define VNIC_PLCMODES_CFG_REQ_ENABLES_HDS_THRESHOLD_VALID  0x4UL
-	__le32 vnic_id;
-	__le16 jumbo_thresh;
-	__le16 hds_offset;
-	__le16 hds_threshold;
-	__le16 unused_0[3];
-};
-
-/* Output (16 bytes) */
-struct hwrm_vnic_plcmodes_cfg_output {
-	__le16 error_code;
-	__le16 req_type;
-	__le16 seq_id;
-	__le16 resp_len;
-	__le32 unused_0;
-	u8 unused_1;
-	u8 unused_2;
-	u8 unused_3;
-	u8 valid;
-};
-
-/* hwrm_vnic_rss_cos_lb_ctx_alloc */
-/* Input (16 bytes) */
-struct hwrm_vnic_rss_cos_lb_ctx_alloc_input {
-	__le16 req_type;
-	__le16 cmpl_ring;
-	__le16 seq_id;
-	__le16 target_id;
-	__le64 resp_addr;
-};
-
-/* Output (16 bytes) */
-struct hwrm_vnic_rss_cos_lb_ctx_alloc_output {
-	__le16 error_code;
-	__le16 req_type;
-	__le16 seq_id;
-	__le16 resp_len;
-	__le16 rss_cos_lb_ctx_id;
-	u8 unused_0;
-	u8 unused_1;
-	u8 unused_2;
-	u8 unused_3;
-	u8 unused_4;
-	u8 valid;
-};
-
-/* hwrm_vnic_rss_cos_lb_ctx_free */
-/* Input (24 bytes) */
-struct hwrm_vnic_rss_cos_lb_ctx_free_input {
-	__le16 req_type;
-	__le16 cmpl_ring;
-	__le16 seq_id;
-	__le16 target_id;
-	__le64 resp_addr;
-	__le16 rss_cos_lb_ctx_id;
-	__le16 unused_0[3];
-};
-
-/* Output (16 bytes) */
-struct hwrm_vnic_rss_cos_lb_ctx_free_output {
-	__le16 error_code;
-	__le16 req_type;
-	__le16 seq_id;
-	__le16 resp_len;
-	__le32 unused_0;
-	u8 unused_1;
-	u8 unused_2;
-	u8 unused_3;
-	u8 valid;
-};
-
-/* hwrm_ring_alloc */
-/* Input (80 bytes) */
-struct hwrm_ring_alloc_input {
-	__le16 req_type;
-	__le16 cmpl_ring;
-	__le16 seq_id;
-	__le16 target_id;
-	__le64 resp_addr;
-	__le32 enables;
-	#define RING_ALLOC_REQ_ENABLES_RESERVED1		    0x1UL
-	#define RING_ALLOC_REQ_ENABLES_RING_ARB_CFG		    0x2UL
-	#define RING_ALLOC_REQ_ENABLES_RESERVED3		    0x4UL
-	#define RING_ALLOC_REQ_ENABLES_STAT_CTX_ID_VALID	    0x8UL
-	#define RING_ALLOC_REQ_ENABLES_RESERVED4		    0x10UL
-	#define RING_ALLOC_REQ_ENABLES_MAX_BW_VALID		    0x20UL
-	u8 ring_type;
-	#define RING_ALLOC_REQ_RING_TYPE_L2_CMPL		   0x0UL
-	#define RING_ALLOC_REQ_RING_TYPE_TX			   0x1UL
-	#define RING_ALLOC_REQ_RING_TYPE_RX			   0x2UL
-	#define RING_ALLOC_REQ_RING_TYPE_ROCE_CMPL		   0x3UL
-	u8 unused_0;
-	__le16 unused_1;
-	__le64 page_tbl_addr;
-	__le32 fbo;
-	u8 page_size;
-	u8 page_tbl_depth;
-	u8 unused_2;
-	u8 unused_3;
-	__le32 length;
-	__le16 logical_id;
-	__le16 cmpl_ring_id;
-	__le16 queue_id;
-	u8 unused_4;
-	u8 unused_5;
-	__le32 reserved1;
-	__le16 ring_arb_cfg;
-	#define RING_ALLOC_REQ_RING_ARB_CFG_ARB_POLICY_MASK	    0xfUL
-	#define RING_ALLOC_REQ_RING_ARB_CFG_ARB_POLICY_SFT	    0
-	#define RING_ALLOC_REQ_RING_ARB_CFG_ARB_POLICY_SP	   (0x1UL << 0)
-	#define RING_ALLOC_REQ_RING_ARB_CFG_ARB_POLICY_WFQ	   (0x2UL << 0)
-	#define RING_ALLOC_REQ_RING_ARB_CFG_ARB_POLICY_LAST    RING_ALLOC_REQ_RING_ARB_CFG_ARB_POLICY_WFQ
-	#define RING_ALLOC_REQ_RING_ARB_CFG_RSVD_MASK		    0xf0UL
-	#define RING_ALLOC_REQ_RING_ARB_CFG_RSVD_SFT		    4
-	#define RING_ALLOC_REQ_RING_ARB_CFG_ARB_POLICY_PARAM_MASK  0xff00UL
-	#define RING_ALLOC_REQ_RING_ARB_CFG_ARB_POLICY_PARAM_SFT   8
-	u8 unused_6;
-	u8 unused_7;
-	__le32 reserved3;
-	__le32 stat_ctx_id;
-	__le32 reserved4;
-	__le32 max_bw;
-	#define RING_ALLOC_REQ_MAX_BW_BW_VALUE_MASK		    0xfffffffUL
-	#define RING_ALLOC_REQ_MAX_BW_BW_VALUE_SFT		    0
-	#define RING_ALLOC_REQ_MAX_BW_SCALE			    0x10000000UL
-	#define RING_ALLOC_REQ_MAX_BW_SCALE_BITS		   (0x0UL << 28)
-	#define RING_ALLOC_REQ_MAX_BW_SCALE_BYTES		   (0x1UL << 28)
-	#define RING_ALLOC_REQ_MAX_BW_SCALE_LAST    RING_ALLOC_REQ_MAX_BW_SCALE_BYTES
-	#define RING_ALLOC_REQ_MAX_BW_BW_VALUE_UNIT_MASK	    0xe0000000UL
-	#define RING_ALLOC_REQ_MAX_BW_BW_VALUE_UNIT_SFT	    29
-	#define RING_ALLOC_REQ_MAX_BW_BW_VALUE_UNIT_MEGA	   (0x0UL << 29)
-	#define RING_ALLOC_REQ_MAX_BW_BW_VALUE_UNIT_KILO	   (0x2UL << 29)
-	#define RING_ALLOC_REQ_MAX_BW_BW_VALUE_UNIT_BASE	   (0x4UL << 29)
-	#define RING_ALLOC_REQ_MAX_BW_BW_VALUE_UNIT_GIGA	   (0x6UL << 29)
-	#define RING_ALLOC_REQ_MAX_BW_BW_VALUE_UNIT_PERCENT1_100  (0x1UL << 29)
-	#define RING_ALLOC_REQ_MAX_BW_BW_VALUE_UNIT_INVALID       (0x7UL << 29)
-	#define RING_ALLOC_REQ_MAX_BW_BW_VALUE_UNIT_LAST    RING_ALLOC_REQ_MAX_BW_BW_VALUE_UNIT_INVALID
-	u8 int_mode;
-	#define RING_ALLOC_REQ_INT_MODE_LEGACY			   0x0UL
-	#define RING_ALLOC_REQ_INT_MODE_RSVD			   0x1UL
-	#define RING_ALLOC_REQ_INT_MODE_MSIX			   0x2UL
-	#define RING_ALLOC_REQ_INT_MODE_POLL			   0x3UL
-	u8 unused_8[3];
-};
-
-/* Output (16 bytes) */
-struct hwrm_ring_alloc_output {
-	__le16 error_code;
-	__le16 req_type;
-	__le16 seq_id;
-	__le16 resp_len;
-	__le16 ring_id;
-	__le16 logical_ring_id;
-	u8 unused_0;
-	u8 unused_1;
-	u8 unused_2;
-	u8 valid;
-};
-
-/* hwrm_ring_free */
-/* Input (24 bytes) */
-struct hwrm_ring_free_input {
-	__le16 req_type;
-	__le16 cmpl_ring;
-	__le16 seq_id;
-	__le16 target_id;
-	__le64 resp_addr;
-	u8 ring_type;
-	#define RING_FREE_REQ_RING_TYPE_L2_CMPL		   0x0UL
-	#define RING_FREE_REQ_RING_TYPE_TX			   0x1UL
-	#define RING_FREE_REQ_RING_TYPE_RX			   0x2UL
-	#define RING_FREE_REQ_RING_TYPE_ROCE_CMPL		   0x3UL
-	u8 unused_0;
-	__le16 ring_id;
-	__le32 unused_1;
-};
-
-/* Output (16 bytes) */
-struct hwrm_ring_free_output {
-	__le16 error_code;
-	__le16 req_type;
-	__le16 seq_id;
-	__le16 resp_len;
-	__le32 unused_0;
-	u8 unused_1;
-	u8 unused_2;
-	u8 unused_3;
-	u8 valid;
-};
-
-/* hwrm_ring_cmpl_ring_qaggint_params */
-/* Input (24 bytes) */
-struct hwrm_ring_cmpl_ring_qaggint_params_input {
-	__le16 req_type;
-	__le16 cmpl_ring;
-	__le16 seq_id;
-	__le16 target_id;
-	__le64 resp_addr;
-	__le16 ring_id;
-	__le16 unused_0[3];
-};
-
-/* Output (32 bytes) */
-struct hwrm_ring_cmpl_ring_qaggint_params_output {
-	__le16 error_code;
-	__le16 req_type;
-	__le16 seq_id;
-	__le16 resp_len;
-	__le16 flags;
-	#define RING_CMPL_RING_QAGGINT_PARAMS_RESP_FLAGS_TIMER_RESET 0x1UL
-	#define RING_CMPL_RING_QAGGINT_PARAMS_RESP_FLAGS_RING_IDLE 0x2UL
-	__le16 num_cmpl_dma_aggr;
-	__le16 num_cmpl_dma_aggr_during_int;
-	__le16 cmpl_aggr_dma_tmr;
-	__le16 cmpl_aggr_dma_tmr_during_int;
-	__le16 int_lat_tmr_min;
-	__le16 int_lat_tmr_max;
-	__le16 num_cmpl_aggr_int;
-	__le32 unused_0;
-	u8 unused_1;
-	u8 unused_2;
-	u8 unused_3;
-	u8 valid;
-};
-
-/* hwrm_ring_cmpl_ring_cfg_aggint_params */
-/* Input (40 bytes) */
-struct hwrm_ring_cmpl_ring_cfg_aggint_params_input {
-	__le16 req_type;
-	__le16 cmpl_ring;
-	__le16 seq_id;
-	__le16 target_id;
-	__le64 resp_addr;
-	__le16 ring_id;
-	__le16 flags;
-	#define RING_CMPL_RING_CFG_AGGINT_PARAMS_REQ_FLAGS_TIMER_RESET 0x1UL
-	#define RING_CMPL_RING_CFG_AGGINT_PARAMS_REQ_FLAGS_RING_IDLE 0x2UL
-	__le16 num_cmpl_dma_aggr;
-	__le16 num_cmpl_dma_aggr_during_int;
-	__le16 cmpl_aggr_dma_tmr;
-	__le16 cmpl_aggr_dma_tmr_during_int;
-	__le16 int_lat_tmr_min;
-	__le16 int_lat_tmr_max;
-	__le16 num_cmpl_aggr_int;
-	__le16 unused_0[3];
-};
-
-/* Output (16 bytes) */
-struct hwrm_ring_cmpl_ring_cfg_aggint_params_output {
-	__le16 error_code;
-	__le16 req_type;
-	__le16 seq_id;
-	__le16 resp_len;
-	__le32 unused_0;
-	u8 unused_1;
-	u8 unused_2;
-	u8 unused_3;
-	u8 valid;
-};
-
-/* hwrm_ring_reset */
-/* Input (24 bytes) */
-struct hwrm_ring_reset_input {
-	__le16 req_type;
-	__le16 cmpl_ring;
-	__le16 seq_id;
-	__le16 target_id;
-	__le64 resp_addr;
-	u8 ring_type;
-	#define RING_RESET_REQ_RING_TYPE_L2_CMPL		   0x0UL
-	#define RING_RESET_REQ_RING_TYPE_TX			   0x1UL
-	#define RING_RESET_REQ_RING_TYPE_RX			   0x2UL
-	#define RING_RESET_REQ_RING_TYPE_ROCE_CMPL		   0x3UL
-	u8 unused_0;
-	__le16 ring_id;
-	__le32 unused_1;
-};
-
-/* Output (16 bytes) */
-struct hwrm_ring_reset_output {
-	__le16 error_code;
-	__le16 req_type;
-	__le16 seq_id;
-	__le16 resp_len;
-	__le32 unused_0;
-	u8 unused_1;
-	u8 unused_2;
-	u8 unused_3;
-	u8 valid;
-};
-
-/* hwrm_ring_grp_alloc */
-/* Input (24 bytes) */
-struct hwrm_ring_grp_alloc_input {
-	__le16 req_type;
-	__le16 cmpl_ring;
-	__le16 seq_id;
-	__le16 target_id;
-	__le64 resp_addr;
-	__le16 cr;
-	__le16 rr;
-	__le16 ar;
-	__le16 sc;
-};
-
-/* Output (16 bytes) */
-struct hwrm_ring_grp_alloc_output {
-	__le16 error_code;
-	__le16 req_type;
-	__le16 seq_id;
-	__le16 resp_len;
-	__le32 ring_group_id;
-	u8 unused_0;
-	u8 unused_1;
-	u8 unused_2;
-	u8 valid;
-};
-
-/* hwrm_ring_grp_free */
-/* Input (24 bytes) */
-struct hwrm_ring_grp_free_input {
-	__le16 req_type;
-	__le16 cmpl_ring;
-	__le16 seq_id;
-	__le16 target_id;
-	__le64 resp_addr;
-	__le32 ring_group_id;
-	__le32 unused_0;
-};
-
-/* Output (16 bytes) */
-struct hwrm_ring_grp_free_output {
-	__le16 error_code;
-	__le16 req_type;
-	__le16 seq_id;
-	__le16 resp_len;
-	__le32 unused_0;
-	u8 unused_1;
-	u8 unused_2;
-	u8 unused_3;
-	u8 valid;
-};
-
-/* hwrm_cfa_l2_filter_alloc */
-/* Input (96 bytes) */
-struct hwrm_cfa_l2_filter_alloc_input {
-	__le16 req_type;
-	__le16 cmpl_ring;
-	__le16 seq_id;
-	__le16 target_id;
-	__le64 resp_addr;
-	__le32 flags;
-	#define CFA_L2_FILTER_ALLOC_REQ_FLAGS_PATH		    0x1UL
-	#define CFA_L2_FILTER_ALLOC_REQ_FLAGS_PATH_TX		   (0x0UL << 0)
-	#define CFA_L2_FILTER_ALLOC_REQ_FLAGS_PATH_RX		   (0x1UL << 0)
-	#define CFA_L2_FILTER_ALLOC_REQ_FLAGS_PATH_LAST    CFA_L2_FILTER_ALLOC_REQ_FLAGS_PATH_RX
-	#define CFA_L2_FILTER_ALLOC_REQ_FLAGS_LOOPBACK		    0x2UL
-	#define CFA_L2_FILTER_ALLOC_REQ_FLAGS_DROP		    0x4UL
-	#define CFA_L2_FILTER_ALLOC_REQ_FLAGS_OUTERMOST	    0x8UL
-	__le32 enables;
-	#define CFA_L2_FILTER_ALLOC_REQ_ENABLES_L2_ADDR	    0x1UL
-	#define CFA_L2_FILTER_ALLOC_REQ_ENABLES_L2_ADDR_MASK       0x2UL
-	#define CFA_L2_FILTER_ALLOC_REQ_ENABLES_L2_OVLAN	    0x4UL
-	#define CFA_L2_FILTER_ALLOC_REQ_ENABLES_L2_OVLAN_MASK      0x8UL
-	#define CFA_L2_FILTER_ALLOC_REQ_ENABLES_L2_IVLAN	    0x10UL
-	#define CFA_L2_FILTER_ALLOC_REQ_ENABLES_L2_IVLAN_MASK      0x20UL
-	#define CFA_L2_FILTER_ALLOC_REQ_ENABLES_T_L2_ADDR	    0x40UL
-	#define CFA_L2_FILTER_ALLOC_REQ_ENABLES_T_L2_ADDR_MASK     0x80UL
-	#define CFA_L2_FILTER_ALLOC_REQ_ENABLES_T_L2_OVLAN	    0x100UL
-	#define CFA_L2_FILTER_ALLOC_REQ_ENABLES_T_L2_OVLAN_MASK    0x200UL
-	#define CFA_L2_FILTER_ALLOC_REQ_ENABLES_T_L2_IVLAN	    0x400UL
-	#define CFA_L2_FILTER_ALLOC_REQ_ENABLES_T_L2_IVLAN_MASK    0x800UL
-	#define CFA_L2_FILTER_ALLOC_REQ_ENABLES_SRC_TYPE	    0x1000UL
-	#define CFA_L2_FILTER_ALLOC_REQ_ENABLES_SRC_ID		    0x2000UL
-	#define CFA_L2_FILTER_ALLOC_REQ_ENABLES_TUNNEL_TYPE	    0x4000UL
-	#define CFA_L2_FILTER_ALLOC_REQ_ENABLES_DST_ID		    0x8000UL
-	#define CFA_L2_FILTER_ALLOC_REQ_ENABLES_MIRROR_VNIC_ID     0x10000UL
-	u8 l2_addr[6];
-	u8 unused_0;
-	u8 unused_1;
-	u8 l2_addr_mask[6];
-	__le16 l2_ovlan;
-	__le16 l2_ovlan_mask;
-	__le16 l2_ivlan;
-	__le16 l2_ivlan_mask;
-	u8 unused_2;
-	u8 unused_3;
-	u8 t_l2_addr[6];
-	u8 unused_4;
-	u8 unused_5;
-	u8 t_l2_addr_mask[6];
-	__le16 t_l2_ovlan;
-	__le16 t_l2_ovlan_mask;
-	__le16 t_l2_ivlan;
-	__le16 t_l2_ivlan_mask;
-	u8 src_type;
-	#define CFA_L2_FILTER_ALLOC_REQ_SRC_TYPE_NPORT		   0x0UL
-	#define CFA_L2_FILTER_ALLOC_REQ_SRC_TYPE_PF		   0x1UL
-	#define CFA_L2_FILTER_ALLOC_REQ_SRC_TYPE_VF		   0x2UL
-	#define CFA_L2_FILTER_ALLOC_REQ_SRC_TYPE_VNIC		   0x3UL
-	#define CFA_L2_FILTER_ALLOC_REQ_SRC_TYPE_KONG		   0x4UL
-	#define CFA_L2_FILTER_ALLOC_REQ_SRC_TYPE_APE		   0x5UL
-	#define CFA_L2_FILTER_ALLOC_REQ_SRC_TYPE_BONO		   0x6UL
-	#define CFA_L2_FILTER_ALLOC_REQ_SRC_TYPE_TANG		   0x7UL
-	u8 unused_6;
-	__le32 src_id;
-	u8 tunnel_type;
-	#define CFA_L2_FILTER_ALLOC_REQ_TUNNEL_TYPE_NONTUNNEL     0x0UL
-	#define CFA_L2_FILTER_ALLOC_REQ_TUNNEL_TYPE_VXLAN	   0x1UL
-	#define CFA_L2_FILTER_ALLOC_REQ_TUNNEL_TYPE_NVGRE	   0x2UL
-	#define CFA_L2_FILTER_ALLOC_REQ_TUNNEL_TYPE_L2GRE	   0x3UL
-	#define CFA_L2_FILTER_ALLOC_REQ_TUNNEL_TYPE_IPIP	   0x4UL
-	#define CFA_L2_FILTER_ALLOC_REQ_TUNNEL_TYPE_GENEVE	   0x5UL
-	#define CFA_L2_FILTER_ALLOC_REQ_TUNNEL_TYPE_MPLS	   0x6UL
-	#define CFA_L2_FILTER_ALLOC_REQ_TUNNEL_TYPE_STT	   0x7UL
-	#define CFA_L2_FILTER_ALLOC_REQ_TUNNEL_TYPE_IPGRE	   0x8UL
-	#define CFA_L2_FILTER_ALLOC_REQ_TUNNEL_TYPE_ANYTUNNEL     0xffUL
-	u8 unused_7;
-	__le16 dst_id;
-	__le16 mirror_vnic_id;
-	u8 pri_hint;
-	#define CFA_L2_FILTER_ALLOC_REQ_PRI_HINT_NO_PREFER	   0x0UL
-	#define CFA_L2_FILTER_ALLOC_REQ_PRI_HINT_ABOVE_FILTER     0x1UL
-	#define CFA_L2_FILTER_ALLOC_REQ_PRI_HINT_BELOW_FILTER     0x2UL
-	#define CFA_L2_FILTER_ALLOC_REQ_PRI_HINT_MAX		   0x3UL
-	#define CFA_L2_FILTER_ALLOC_REQ_PRI_HINT_MIN		   0x4UL
-	u8 unused_8;
-	__le32 unused_9;
-	__le64 l2_filter_id_hint;
-};
-
-/* Output (24 bytes) */
-struct hwrm_cfa_l2_filter_alloc_output {
-	__le16 error_code;
-	__le16 req_type;
-	__le16 seq_id;
-	__le16 resp_len;
-	__le64 l2_filter_id;
-	__le32 flow_id;
-	u8 unused_0;
-	u8 unused_1;
-	u8 unused_2;
-	u8 valid;
-};
-
-/* hwrm_cfa_l2_filter_free */
-/* Input (24 bytes) */
-struct hwrm_cfa_l2_filter_free_input {
-	__le16 req_type;
-	__le16 cmpl_ring;
-	__le16 seq_id;
-	__le16 target_id;
-	__le64 resp_addr;
-	__le64 l2_filter_id;
-};
-
-/* Output (16 bytes) */
-struct hwrm_cfa_l2_filter_free_output {
-	__le16 error_code;
-	__le16 req_type;
-	__le16 seq_id;
-	__le16 resp_len;
-	__le32 unused_0;
-	u8 unused_1;
-	u8 unused_2;
-	u8 unused_3;
-	u8 valid;
-};
-
-/* hwrm_cfa_l2_filter_cfg */
-/* Input (40 bytes) */
-struct hwrm_cfa_l2_filter_cfg_input {
-	__le16 req_type;
-	__le16 cmpl_ring;
-	__le16 seq_id;
-	__le16 target_id;
-	__le64 resp_addr;
-	__le32 flags;
-	#define CFA_L2_FILTER_CFG_REQ_FLAGS_PATH		    0x1UL
-	#define CFA_L2_FILTER_CFG_REQ_FLAGS_PATH_TX		   (0x0UL << 0)
-	#define CFA_L2_FILTER_CFG_REQ_FLAGS_PATH_RX		   (0x1UL << 0)
-	#define CFA_L2_FILTER_CFG_REQ_FLAGS_PATH_LAST    CFA_L2_FILTER_CFG_REQ_FLAGS_PATH_RX
-	#define CFA_L2_FILTER_CFG_REQ_FLAGS_DROP		    0x2UL
-	__le32 enables;
-	#define CFA_L2_FILTER_CFG_REQ_ENABLES_DST_ID		    0x1UL
-	#define CFA_L2_FILTER_CFG_REQ_ENABLES_NEW_MIRROR_VNIC_ID   0x2UL
-	__le64 l2_filter_id;
-	__le32 dst_id;
-	__le32 new_mirror_vnic_id;
-};
-
-/* Output (16 bytes) */
-struct hwrm_cfa_l2_filter_cfg_output {
-	__le16 error_code;
-	__le16 req_type;
-	__le16 seq_id;
-	__le16 resp_len;
-	__le32 unused_0;
-	u8 unused_1;
-	u8 unused_2;
-	u8 unused_3;
-	u8 valid;
-};
-
-/* hwrm_cfa_l2_set_rx_mask */
-/* Input (56 bytes) */
-struct hwrm_cfa_l2_set_rx_mask_input {
-	__le16 req_type;
-	__le16 cmpl_ring;
-	__le16 seq_id;
-	__le16 target_id;
-	__le64 resp_addr;
-	__le32 vnic_id;
-	__le32 mask;
-	#define CFA_L2_SET_RX_MASK_REQ_MASK_RESERVED		    0x1UL
-	#define CFA_L2_SET_RX_MASK_REQ_MASK_MCAST		    0x2UL
-	#define CFA_L2_SET_RX_MASK_REQ_MASK_ALL_MCAST		    0x4UL
-	#define CFA_L2_SET_RX_MASK_REQ_MASK_BCAST		    0x8UL
-	#define CFA_L2_SET_RX_MASK_REQ_MASK_PROMISCUOUS	    0x10UL
-	#define CFA_L2_SET_RX_MASK_REQ_MASK_OUTERMOST		    0x20UL
-	#define CFA_L2_SET_RX_MASK_REQ_MASK_VLANONLY		    0x40UL
-	#define CFA_L2_SET_RX_MASK_REQ_MASK_VLAN_NONVLAN	    0x80UL
-	#define CFA_L2_SET_RX_MASK_REQ_MASK_ANYVLAN_NONVLAN	    0x100UL
-	__le64 mc_tbl_addr;
-	__le32 num_mc_entries;
-	__le32 unused_0;
-	__le64 vlan_tag_tbl_addr;
-	__le32 num_vlan_tags;
-	__le32 unused_1;
-};
-
-/* Output (16 bytes) */
-struct hwrm_cfa_l2_set_rx_mask_output {
-	__le16 error_code;
-	__le16 req_type;
-	__le16 seq_id;
-	__le16 resp_len;
-	__le32 unused_0;
-	u8 unused_1;
-	u8 unused_2;
-	u8 unused_3;
-	u8 valid;
-};
-
-/* hwrm_cfa_tunnel_filter_alloc */
-/* Input (88 bytes) */
-struct hwrm_cfa_tunnel_filter_alloc_input {
-	__le16 req_type;
-	__le16 cmpl_ring;
-	__le16 seq_id;
-	__le16 target_id;
-	__le64 resp_addr;
-	__le32 flags;
-	#define CFA_TUNNEL_FILTER_ALLOC_REQ_FLAGS_LOOPBACK	    0x1UL
-	__le32 enables;
-	#define CFA_TUNNEL_FILTER_ALLOC_REQ_ENABLES_L2_FILTER_ID   0x1UL
-	#define CFA_TUNNEL_FILTER_ALLOC_REQ_ENABLES_L2_ADDR	    0x2UL
-	#define CFA_TUNNEL_FILTER_ALLOC_REQ_ENABLES_L2_IVLAN       0x4UL
-	#define CFA_TUNNEL_FILTER_ALLOC_REQ_ENABLES_L3_ADDR	    0x8UL
-	#define CFA_TUNNEL_FILTER_ALLOC_REQ_ENABLES_L3_ADDR_TYPE   0x10UL
-	#define CFA_TUNNEL_FILTER_ALLOC_REQ_ENABLES_T_L3_ADDR_TYPE 0x20UL
-	#define CFA_TUNNEL_FILTER_ALLOC_REQ_ENABLES_T_L3_ADDR      0x40UL
-	#define CFA_TUNNEL_FILTER_ALLOC_REQ_ENABLES_TUNNEL_TYPE    0x80UL
-	#define CFA_TUNNEL_FILTER_ALLOC_REQ_ENABLES_VNI	    0x100UL
-	#define CFA_TUNNEL_FILTER_ALLOC_REQ_ENABLES_DST_VNIC_ID    0x200UL
-	#define CFA_TUNNEL_FILTER_ALLOC_REQ_ENABLES_MIRROR_VNIC_ID 0x400UL
-	__le64 l2_filter_id;
-	u8 l2_addr[6];
-	__le16 l2_ivlan;
-	__le32 l3_addr[4];
-	__le32 t_l3_addr[4];
-	u8 l3_addr_type;
-	u8 t_l3_addr_type;
-	u8 tunnel_type;
-	#define CFA_TUNNEL_FILTER_ALLOC_REQ_TUNNEL_TYPE_NONTUNNEL 0x0UL
-	#define CFA_TUNNEL_FILTER_ALLOC_REQ_TUNNEL_TYPE_VXLAN     0x1UL
-	#define CFA_TUNNEL_FILTER_ALLOC_REQ_TUNNEL_TYPE_NVGRE     0x2UL
-	#define CFA_TUNNEL_FILTER_ALLOC_REQ_TUNNEL_TYPE_L2GRE     0x3UL
-	#define CFA_TUNNEL_FILTER_ALLOC_REQ_TUNNEL_TYPE_IPIP      0x4UL
-	#define CFA_TUNNEL_FILTER_ALLOC_REQ_TUNNEL_TYPE_GENEVE    0x5UL
-	#define CFA_TUNNEL_FILTER_ALLOC_REQ_TUNNEL_TYPE_MPLS      0x6UL
-	#define CFA_TUNNEL_FILTER_ALLOC_REQ_TUNNEL_TYPE_STT       0x7UL
-	#define CFA_TUNNEL_FILTER_ALLOC_REQ_TUNNEL_TYPE_IPGRE     0x8UL
-	#define CFA_TUNNEL_FILTER_ALLOC_REQ_TUNNEL_TYPE_ANYTUNNEL 0xffUL
-	u8 unused_0;
-	__le32 vni;
-	__le32 dst_vnic_id;
-	__le32 mirror_vnic_id;
-};
-
-/* Output (24 bytes) */
-struct hwrm_cfa_tunnel_filter_alloc_output {
-	__le16 error_code;
-	__le16 req_type;
-	__le16 seq_id;
-	__le16 resp_len;
-	__le64 tunnel_filter_id;
-	__le32 flow_id;
-	u8 unused_0;
-	u8 unused_1;
-	u8 unused_2;
-	u8 valid;
-};
-
-/* hwrm_cfa_tunnel_filter_free */
-/* Input (24 bytes) */
-struct hwrm_cfa_tunnel_filter_free_input {
-	__le16 req_type;
-	__le16 cmpl_ring;
-	__le16 seq_id;
-	__le16 target_id;
-	__le64 resp_addr;
-	__le64 tunnel_filter_id;
-};
-
-/* Output (16 bytes) */
-struct hwrm_cfa_tunnel_filter_free_output {
-	__le16 error_code;
-	__le16 req_type;
-	__le16 seq_id;
-	__le16 resp_len;
-	__le32 unused_0;
-	u8 unused_1;
-	u8 unused_2;
-	u8 unused_3;
-	u8 valid;
-};
-
-/* hwrm_cfa_encap_record_alloc */
-/* Input (32 bytes) */
-struct hwrm_cfa_encap_record_alloc_input {
-	__le16 req_type;
-	__le16 cmpl_ring;
-	__le16 seq_id;
-	__le16 target_id;
-	__le64 resp_addr;
-	__le32 flags;
-	#define CFA_ENCAP_RECORD_ALLOC_REQ_FLAGS_LOOPBACK	    0x1UL
-	u8 encap_type;
-	#define CFA_ENCAP_RECORD_ALLOC_REQ_ENCAP_TYPE_VXLAN       0x1UL
-	#define CFA_ENCAP_RECORD_ALLOC_REQ_ENCAP_TYPE_NVGRE       0x2UL
-	#define CFA_ENCAP_RECORD_ALLOC_REQ_ENCAP_TYPE_L2GRE       0x3UL
-	#define CFA_ENCAP_RECORD_ALLOC_REQ_ENCAP_TYPE_IPIP	   0x4UL
-	#define CFA_ENCAP_RECORD_ALLOC_REQ_ENCAP_TYPE_GENEVE      0x5UL
-	#define CFA_ENCAP_RECORD_ALLOC_REQ_ENCAP_TYPE_MPLS	   0x6UL
-	#define CFA_ENCAP_RECORD_ALLOC_REQ_ENCAP_TYPE_VLAN	   0x7UL
-	#define CFA_ENCAP_RECORD_ALLOC_REQ_ENCAP_TYPE_IPGRE       0x8UL
-	u8 unused_0;
-	__le16 unused_1;
-	__le32 encap_data[16];
-};
-
-/* Output (16 bytes) */
-struct hwrm_cfa_encap_record_alloc_output {
-	__le16 error_code;
-	__le16 req_type;
-	__le16 seq_id;
-	__le16 resp_len;
-	__le32 encap_record_id;
-	u8 unused_0;
-	u8 unused_1;
-	u8 unused_2;
-	u8 valid;
-};
-
-/* hwrm_cfa_encap_record_free */
-/* Input (24 bytes) */
-struct hwrm_cfa_encap_record_free_input {
-	__le16 req_type;
-	__le16 cmpl_ring;
-	__le16 seq_id;
-	__le16 target_id;
-	__le64 resp_addr;
-	__le32 encap_record_id;
-	__le32 unused_0;
-};
-
-/* Output (16 bytes) */
-struct hwrm_cfa_encap_record_free_output {
-	__le16 error_code;
-	__le16 req_type;
-	__le16 seq_id;
-	__le16 resp_len;
-	__le32 unused_0;
-	u8 unused_1;
-	u8 unused_2;
-	u8 unused_3;
-	u8 valid;
-};
-
-/* hwrm_cfa_ntuple_filter_alloc */
-/* Input (128 bytes) */
-struct hwrm_cfa_ntuple_filter_alloc_input {
-	__le16 req_type;
-	__le16 cmpl_ring;
-	__le16 seq_id;
-	__le16 target_id;
-	__le64 resp_addr;
-	__le32 flags;
-	#define CFA_NTUPLE_FILTER_ALLOC_REQ_FLAGS_LOOPBACK	    0x1UL
-	#define CFA_NTUPLE_FILTER_ALLOC_REQ_FLAGS_DROP		    0x2UL
-	#define CFA_NTUPLE_FILTER_ALLOC_REQ_FLAGS_METER	    0x4UL
-	__le32 enables;
-	#define CFA_NTUPLE_FILTER_ALLOC_REQ_ENABLES_L2_FILTER_ID   0x1UL
-	#define CFA_NTUPLE_FILTER_ALLOC_REQ_ENABLES_ETHERTYPE      0x2UL
-	#define CFA_NTUPLE_FILTER_ALLOC_REQ_ENABLES_TUNNEL_TYPE    0x4UL
-	#define CFA_NTUPLE_FILTER_ALLOC_REQ_ENABLES_SRC_MACADDR    0x8UL
-	#define CFA_NTUPLE_FILTER_ALLOC_REQ_ENABLES_IPADDR_TYPE    0x10UL
-	#define CFA_NTUPLE_FILTER_ALLOC_REQ_ENABLES_SRC_IPADDR     0x20UL
-	#define CFA_NTUPLE_FILTER_ALLOC_REQ_ENABLES_SRC_IPADDR_MASK 0x40UL
-	#define CFA_NTUPLE_FILTER_ALLOC_REQ_ENABLES_DST_IPADDR     0x80UL
-	#define CFA_NTUPLE_FILTER_ALLOC_REQ_ENABLES_DST_IPADDR_MASK 0x100UL
-	#define CFA_NTUPLE_FILTER_ALLOC_REQ_ENABLES_IP_PROTOCOL    0x200UL
-	#define CFA_NTUPLE_FILTER_ALLOC_REQ_ENABLES_SRC_PORT       0x400UL
-	#define CFA_NTUPLE_FILTER_ALLOC_REQ_ENABLES_SRC_PORT_MASK  0x800UL
-	#define CFA_NTUPLE_FILTER_ALLOC_REQ_ENABLES_DST_PORT       0x1000UL
-	#define CFA_NTUPLE_FILTER_ALLOC_REQ_ENABLES_DST_PORT_MASK  0x2000UL
-	#define CFA_NTUPLE_FILTER_ALLOC_REQ_ENABLES_PRI_HINT       0x4000UL
-	#define CFA_NTUPLE_FILTER_ALLOC_REQ_ENABLES_NTUPLE_FILTER_ID 0x8000UL
-	#define CFA_NTUPLE_FILTER_ALLOC_REQ_ENABLES_DST_ID	    0x10000UL
-	#define CFA_NTUPLE_FILTER_ALLOC_REQ_ENABLES_MIRROR_VNIC_ID 0x20000UL
-	#define CFA_NTUPLE_FILTER_ALLOC_REQ_ENABLES_DST_MACADDR    0x40000UL
-	__le64 l2_filter_id;
-	u8 src_macaddr[6];
-	__be16 ethertype;
-	u8 ip_addr_type;
-	#define CFA_NTUPLE_FILTER_ALLOC_REQ_IP_ADDR_TYPE_UNKNOWN  0x0UL
-	#define CFA_NTUPLE_FILTER_ALLOC_REQ_IP_ADDR_TYPE_IPV4     0x4UL
-	#define CFA_NTUPLE_FILTER_ALLOC_REQ_IP_ADDR_TYPE_IPV6     0x6UL
-	u8 ip_protocol;
-	#define CFA_NTUPLE_FILTER_ALLOC_REQ_IP_PROTOCOL_UNKNOWN   0x0UL
-	#define CFA_NTUPLE_FILTER_ALLOC_REQ_IP_PROTOCOL_UDP       0x6UL
-	#define CFA_NTUPLE_FILTER_ALLOC_REQ_IP_PROTOCOL_TCP       0x11UL
-	__le16 dst_id;
-	__le16 mirror_vnic_id;
-	u8 tunnel_type;
-	#define CFA_NTUPLE_FILTER_ALLOC_REQ_TUNNEL_TYPE_NONTUNNEL 0x0UL
-	#define CFA_NTUPLE_FILTER_ALLOC_REQ_TUNNEL_TYPE_VXLAN     0x1UL
-	#define CFA_NTUPLE_FILTER_ALLOC_REQ_TUNNEL_TYPE_NVGRE     0x2UL
-	#define CFA_NTUPLE_FILTER_ALLOC_REQ_TUNNEL_TYPE_L2GRE     0x3UL
-	#define CFA_NTUPLE_FILTER_ALLOC_REQ_TUNNEL_TYPE_IPIP      0x4UL
-	#define CFA_NTUPLE_FILTER_ALLOC_REQ_TUNNEL_TYPE_GENEVE    0x5UL
-	#define CFA_NTUPLE_FILTER_ALLOC_REQ_TUNNEL_TYPE_MPLS      0x6UL
-	#define CFA_NTUPLE_FILTER_ALLOC_REQ_TUNNEL_TYPE_STT       0x7UL
-	#define CFA_NTUPLE_FILTER_ALLOC_REQ_TUNNEL_TYPE_IPGRE     0x8UL
-	#define CFA_NTUPLE_FILTER_ALLOC_REQ_TUNNEL_TYPE_ANYTUNNEL 0xffUL
-	u8 pri_hint;
-	#define CFA_NTUPLE_FILTER_ALLOC_REQ_PRI_HINT_NO_PREFER    0x0UL
-	#define CFA_NTUPLE_FILTER_ALLOC_REQ_PRI_HINT_ABOVE	   0x1UL
-	#define CFA_NTUPLE_FILTER_ALLOC_REQ_PRI_HINT_BELOW	   0x2UL
-	#define CFA_NTUPLE_FILTER_ALLOC_REQ_PRI_HINT_HIGHEST      0x3UL
-	#define CFA_NTUPLE_FILTER_ALLOC_REQ_PRI_HINT_LOWEST       0x4UL
-	__be32 src_ipaddr[4];
-	__be32 src_ipaddr_mask[4];
-	__be32 dst_ipaddr[4];
-	__be32 dst_ipaddr_mask[4];
-	__be16 src_port;
-	__be16 src_port_mask;
-	__be16 dst_port;
-	__be16 dst_port_mask;
-	__le64 ntuple_filter_id_hint;
-};
-
-/* Output (24 bytes) */
-struct hwrm_cfa_ntuple_filter_alloc_output {
-	__le16 error_code;
-	__le16 req_type;
-	__le16 seq_id;
-	__le16 resp_len;
-	__le64 ntuple_filter_id;
-	__le32 flow_id;
-	u8 unused_0;
-	u8 unused_1;
-	u8 unused_2;
-	u8 valid;
-};
-
-/* hwrm_cfa_ntuple_filter_free */
-/* Input (24 bytes) */
-struct hwrm_cfa_ntuple_filter_free_input {
-	__le16 req_type;
-	__le16 cmpl_ring;
-	__le16 seq_id;
-	__le16 target_id;
-	__le64 resp_addr;
-	__le64 ntuple_filter_id;
-};
-
-/* Output (16 bytes) */
-struct hwrm_cfa_ntuple_filter_free_output {
-	__le16 error_code;
-	__le16 req_type;
-	__le16 seq_id;
-	__le16 resp_len;
-	__le32 unused_0;
-	u8 unused_1;
-	u8 unused_2;
-	u8 unused_3;
-	u8 valid;
-};
-
-/* hwrm_cfa_ntuple_filter_cfg */
-/* Input (48 bytes) */
-struct hwrm_cfa_ntuple_filter_cfg_input {
-	__le16 req_type;
-	__le16 cmpl_ring;
-	__le16 seq_id;
-	__le16 target_id;
-	__le64 resp_addr;
-	__le32 enables;
-	#define CFA_NTUPLE_FILTER_CFG_REQ_ENABLES_NEW_DST_ID       0x1UL
-	#define CFA_NTUPLE_FILTER_CFG_REQ_ENABLES_NEW_MIRROR_VNIC_ID 0x2UL
-	#define CFA_NTUPLE_FILTER_CFG_REQ_ENABLES_NEW_METER_INSTANCE_ID 0x4UL
-	__le32 unused_0;
-	__le64 ntuple_filter_id;
-	__le32 new_dst_id;
-	__le32 new_mirror_vnic_id;
-	__le16 new_meter_instance_id;
-	#define CFA_NTUPLE_FILTER_CFG_REQ_NEW_METER_INSTANCE_ID_INVALID 0xffffUL
-	__le16 unused_1[3];
-};
-
-/* Output (16 bytes) */
-struct hwrm_cfa_ntuple_filter_cfg_output {
-	__le16 error_code;
-	__le16 req_type;
-	__le16 seq_id;
-	__le16 resp_len;
-	__le32 unused_0;
-	u8 unused_1;
-	u8 unused_2;
-	u8 unused_3;
-	u8 valid;
-};
-
-/* hwrm_tunnel_dst_port_query */
-/* Input (24 bytes) */
-struct hwrm_tunnel_dst_port_query_input {
-	__le16 req_type;
-	__le16 cmpl_ring;
-	__le16 seq_id;
-	__le16 target_id;
-	__le64 resp_addr;
-	u8 tunnel_type;
-	#define TUNNEL_DST_PORT_QUERY_REQ_TUNNEL_TYPE_VXLAN       0x1UL
-	#define TUNNEL_DST_PORT_QUERY_REQ_TUNNEL_TYPE_GENEVE      0x5UL
-	u8 unused_0[7];
-};
-
-/* Output (16 bytes) */
-struct hwrm_tunnel_dst_port_query_output {
-	__le16 error_code;
-	__le16 req_type;
-	__le16 seq_id;
-	__le16 resp_len;
-	__le16 tunnel_dst_port_id;
-	__be16 tunnel_dst_port_val;
-	u8 unused_0;
-	u8 unused_1;
-	u8 unused_2;
-	u8 valid;
-};
-
-/* hwrm_tunnel_dst_port_alloc */
-/* Input (24 bytes) */
-struct hwrm_tunnel_dst_port_alloc_input {
-	__le16 req_type;
-	__le16 cmpl_ring;
-	__le16 seq_id;
-	__le16 target_id;
-	__le64 resp_addr;
-	u8 tunnel_type;
-	#define TUNNEL_DST_PORT_ALLOC_REQ_TUNNEL_TYPE_VXLAN       0x1UL
-	#define TUNNEL_DST_PORT_ALLOC_REQ_TUNNEL_TYPE_GENEVE      0x5UL
-	u8 unused_0;
-	__be16 tunnel_dst_port_val;
-	__le32 unused_1;
-};
-
-/* Output (16 bytes) */
-struct hwrm_tunnel_dst_port_alloc_output {
-	__le16 error_code;
-	__le16 req_type;
-	__le16 seq_id;
-	__le16 resp_len;
-	__le16 tunnel_dst_port_id;
-	u8 unused_0;
-	u8 unused_1;
-	u8 unused_2;
-	u8 unused_3;
-	u8 unused_4;
-	u8 valid;
-};
-
-/* hwrm_tunnel_dst_port_free */
-/* Input (24 bytes) */
-struct hwrm_tunnel_dst_port_free_input {
-	__le16 req_type;
-	__le16 cmpl_ring;
-	__le16 seq_id;
-	__le16 target_id;
-	__le64 resp_addr;
-	u8 tunnel_type;
-	#define TUNNEL_DST_PORT_FREE_REQ_TUNNEL_TYPE_VXLAN	   0x1UL
-	#define TUNNEL_DST_PORT_FREE_REQ_TUNNEL_TYPE_GENEVE       0x5UL
-	u8 unused_0;
-	__le16 tunnel_dst_port_id;
-	__le32 unused_1;
-};
-
-/* Output (16 bytes) */
-struct hwrm_tunnel_dst_port_free_output {
-	__le16 error_code;
-	__le16 req_type;
-	__le16 seq_id;
-	__le16 resp_len;
-	__le32 unused_0;
-	u8 unused_1;
-	u8 unused_2;
-	u8 unused_3;
-	u8 valid;
-};
-
-/* hwrm_stat_ctx_alloc */
-/* Input (32 bytes) */
-struct hwrm_stat_ctx_alloc_input {
-	__le16 req_type;
-	__le16 cmpl_ring;
-	__le16 seq_id;
-	__le16 target_id;
-	__le64 resp_addr;
-	__le64 stats_dma_addr;
-	__le32 update_period_ms;
-	u8 stat_ctx_flags;
-	#define STAT_CTX_ALLOC_REQ_STAT_CTX_FLAGS_ROCE		    0x1UL
-	u8 unused_0[3];
-};
-
-/* Output (16 bytes) */
-struct hwrm_stat_ctx_alloc_output {
-	__le16 error_code;
-	__le16 req_type;
-	__le16 seq_id;
-	__le16 resp_len;
-	__le32 stat_ctx_id;
-	u8 unused_0;
-	u8 unused_1;
-	u8 unused_2;
-	u8 valid;
-};
-
-/* hwrm_stat_ctx_free */
-/* Input (24 bytes) */
-struct hwrm_stat_ctx_free_input {
-	__le16 req_type;
-	__le16 cmpl_ring;
-	__le16 seq_id;
-	__le16 target_id;
-	__le64 resp_addr;
-	__le32 stat_ctx_id;
-	__le32 unused_0;
-};
-
-/* Output (16 bytes) */
-struct hwrm_stat_ctx_free_output {
-	__le16 error_code;
-	__le16 req_type;
-	__le16 seq_id;
-	__le16 resp_len;
-	__le32 stat_ctx_id;
-	u8 unused_0;
-	u8 unused_1;
-	u8 unused_2;
-	u8 valid;
-};
-
-/* hwrm_stat_ctx_query */
-/* Input (24 bytes) */
-struct hwrm_stat_ctx_query_input {
-	__le16 req_type;
-	__le16 cmpl_ring;
-	__le16 seq_id;
-	__le16 target_id;
-	__le64 resp_addr;
-	__le32 stat_ctx_id;
-	__le32 unused_0;
-};
-
-/* Output (176 bytes) */
-struct hwrm_stat_ctx_query_output {
-	__le16 error_code;
-	__le16 req_type;
-	__le16 seq_id;
-	__le16 resp_len;
-	__le64 tx_ucast_pkts;
-	__le64 tx_mcast_pkts;
-	__le64 tx_bcast_pkts;
-	__le64 tx_err_pkts;
-	__le64 tx_drop_pkts;
-	__le64 tx_ucast_bytes;
-	__le64 tx_mcast_bytes;
-	__le64 tx_bcast_bytes;
-	__le64 rx_ucast_pkts;
-	__le64 rx_mcast_pkts;
-	__le64 rx_bcast_pkts;
-	__le64 rx_err_pkts;
-	__le64 rx_drop_pkts;
-	__le64 rx_ucast_bytes;
-	__le64 rx_mcast_bytes;
-	__le64 rx_bcast_bytes;
-	__le64 rx_agg_pkts;
-	__le64 rx_agg_bytes;
-	__le64 rx_agg_events;
-	__le64 rx_agg_aborts;
-	__le32 unused_0;
-	u8 unused_1;
-	u8 unused_2;
-	u8 unused_3;
-	u8 valid;
-};
-
-/* hwrm_stat_ctx_clr_stats */
-/* Input (24 bytes) */
-struct hwrm_stat_ctx_clr_stats_input {
-	__le16 req_type;
-	__le16 cmpl_ring;
-	__le16 seq_id;
-	__le16 target_id;
-	__le64 resp_addr;
-	__le32 stat_ctx_id;
-	__le32 unused_0;
-};
-
-/* Output (16 bytes) */
-struct hwrm_stat_ctx_clr_stats_output {
-	__le16 error_code;
-	__le16 req_type;
-	__le16 seq_id;
-	__le16 resp_len;
-	__le32 unused_0;
-	u8 unused_1;
-	u8 unused_2;
-	u8 unused_3;
-	u8 valid;
-};
-
-/* hwrm_fw_reset */
-/* Input (24 bytes) */
-struct hwrm_fw_reset_input {
-	__le16 req_type;
-	__le16 cmpl_ring;
-	__le16 seq_id;
-	__le16 target_id;
-	__le64 resp_addr;
-	u8 embedded_proc_type;
-	#define FW_RESET_REQ_EMBEDDED_PROC_TYPE_BOOT		   0x0UL
-	#define FW_RESET_REQ_EMBEDDED_PROC_TYPE_MGMT		   0x1UL
-	#define FW_RESET_REQ_EMBEDDED_PROC_TYPE_NETCTRL	   0x2UL
-	#define FW_RESET_REQ_EMBEDDED_PROC_TYPE_ROCE		   0x3UL
-	#define FW_RESET_REQ_EMBEDDED_PROC_TYPE_RSVD		   0x4UL
-	u8 selfrst_status;
-	#define FW_RESET_REQ_SELFRST_STATUS_SELFRSTNONE	   0x0UL
-	#define FW_RESET_REQ_SELFRST_STATUS_SELFRSTASAP	   0x1UL
-	#define FW_RESET_REQ_SELFRST_STATUS_SELFRSTPCIERST	   0x2UL
-	__le16 unused_0[3];
-};
-
-/* Output (16 bytes) */
-struct hwrm_fw_reset_output {
-	__le16 error_code;
-	__le16 req_type;
-	__le16 seq_id;
-	__le16 resp_len;
-	u8 selfrst_status;
-	#define FW_RESET_RESP_SELFRST_STATUS_SELFRSTNONE	   0x0UL
-	#define FW_RESET_RESP_SELFRST_STATUS_SELFRSTASAP	   0x1UL
-	#define FW_RESET_RESP_SELFRST_STATUS_SELFRSTPCIERST       0x2UL
-	u8 unused_0;
-	__le16 unused_1;
-	u8 unused_2;
-	u8 unused_3;
-	u8 unused_4;
-	u8 valid;
-};
-
-/* hwrm_fw_qstatus */
-/* Input (24 bytes) */
-struct hwrm_fw_qstatus_input {
-	__le16 req_type;
-	__le16 cmpl_ring;
-	__le16 seq_id;
-	__le16 target_id;
-	__le64 resp_addr;
-	u8 embedded_proc_type;
-	#define FW_QSTATUS_REQ_EMBEDDED_PROC_TYPE_BOOT		   0x0UL
-	#define FW_QSTATUS_REQ_EMBEDDED_PROC_TYPE_MGMT		   0x1UL
-	#define FW_QSTATUS_REQ_EMBEDDED_PROC_TYPE_NETCTRL	   0x2UL
-	#define FW_QSTATUS_REQ_EMBEDDED_PROC_TYPE_ROCE		   0x3UL
-	#define FW_QSTATUS_REQ_EMBEDDED_PROC_TYPE_RSVD		   0x4UL
-	u8 unused_0[7];
-};
-
-/* Output (16 bytes) */
-struct hwrm_fw_qstatus_output {
-	__le16 error_code;
-	__le16 req_type;
-	__le16 seq_id;
-	__le16 resp_len;
-	u8 selfrst_status;
-	#define FW_QSTATUS_RESP_SELFRST_STATUS_SELFRSTNONE	   0x0UL
-	#define FW_QSTATUS_RESP_SELFRST_STATUS_SELFRSTASAP	   0x1UL
-	#define FW_QSTATUS_RESP_SELFRST_STATUS_SELFRSTPCIERST     0x2UL
-	u8 unused_0;
-	__le16 unused_1;
-	u8 unused_2;
-	u8 unused_3;
-	u8 unused_4;
-	u8 valid;
-};
-
-/* hwrm_fw_set_time */
-/* Input (32 bytes) */
-struct hwrm_fw_set_time_input {
-	__le16 req_type;
-	__le16 cmpl_ring;
-	__le16 seq_id;
-	__le16 target_id;
-	__le64 resp_addr;
-	__le16 year;
-	#define FW_SET_TIME_REQ_YEAR_UNKNOWN			   0x0UL
-	u8 month;
-	u8 day;
-	u8 hour;
-	u8 minute;
-	u8 second;
-	u8 unused_0;
-	__le16 millisecond;
-	__le16 zone;
-	#define FW_SET_TIME_REQ_ZONE_UTC			   0x0UL
-	#define FW_SET_TIME_REQ_ZONE_UNKNOWN			   0xffffUL
-	__le32 unused_1;
-};
-
-/* Output (16 bytes) */
-struct hwrm_fw_set_time_output {
-	__le16 error_code;
-	__le16 req_type;
-	__le16 seq_id;
-	__le16 resp_len;
-	__le32 unused_0;
-	u8 unused_1;
-	u8 unused_2;
-	u8 unused_3;
-	u8 valid;
-};
-
-/* hwrm_fw_set_structured_data */
-/* Input (32 bytes) */
-struct hwrm_fw_set_structured_data_input {
-	__le16 req_type;
-	__le16 cmpl_ring;
-	__le16 seq_id;
-	__le16 target_id;
-	__le64 resp_addr;
-	__le64 src_data_addr;
-	__le16 data_len;
-	u8 hdr_cnt;
-	u8 unused_0[5];
-};
-
-/* Output (16 bytes) */
-struct hwrm_fw_set_structured_data_output {
-	__le16 error_code;
-	__le16 req_type;
-	__le16 seq_id;
-	__le16 resp_len;
-	__le32 unused_0;
-	u8 unused_1;
-	u8 unused_2;
-	u8 unused_3;
-	u8 valid;
-};
-
-/* hwrm_fw_get_structured_data */
-/* Input (32 bytes) */
-struct hwrm_fw_get_structured_data_input {
-	__le16 req_type;
-	__le16 cmpl_ring;
-	__le16 seq_id;
-	__le16 target_id;
-	__le64 resp_addr;
-	__le64 dest_data_addr;
-	__le16 data_len;
-	__le16 structure_id;
-	__le16 subtype;
-	#define FW_GET_STRUCTURED_DATA_REQ_SUBTYPE_ALL		   0xffffUL
-	#define FW_GET_STRUCTURED_DATA_REQ_SUBTYPE_NEAR_BRIDGE_ADMIN 0x100UL
-	#define FW_GET_STRUCTURED_DATA_REQ_SUBTYPE_NEAR_BRIDGE_PEER 0x101UL
-	#define FW_GET_STRUCTURED_DATA_REQ_SUBTYPE_NEAR_BRIDGE_OPERATIONAL 0x102UL
-	#define FW_GET_STRUCTURED_DATA_REQ_SUBTYPE_NON_TPMR_ADMIN 0x200UL
-	#define FW_GET_STRUCTURED_DATA_REQ_SUBTYPE_NON_TPMR_PEER  0x201UL
-	#define FW_GET_STRUCTURED_DATA_REQ_SUBTYPE_NON_TPMR_OPERATIONAL 0x202UL
-	#define FW_GET_STRUCTURED_DATA_REQ_SUBTYPE_HOST_OPERATIONAL 0x300UL
-	u8 count;
-	u8 unused_0;
-};
-
-/* Output (16 bytes) */
-struct hwrm_fw_get_structured_data_output {
-	__le16 error_code;
-	__le16 req_type;
-	__le16 seq_id;
-	__le16 resp_len;
-	u8 hdr_cnt;
-	u8 unused_0;
-	__le16 unused_1;
-	u8 unused_2;
-	u8 unused_3;
-	u8 unused_4;
-	u8 valid;
-};
-
-/* hwrm_exec_fwd_resp */
-/* Input (128 bytes) */
-struct hwrm_exec_fwd_resp_input {
-	__le16 req_type;
-	__le16 cmpl_ring;
-	__le16 seq_id;
-	__le16 target_id;
-	__le64 resp_addr;
-	__le32 encap_request[26];
-	__le16 encap_resp_target_id;
-	__le16 unused_0[3];
-};
-
-/* Output (16 bytes) */
-struct hwrm_exec_fwd_resp_output {
-	__le16 error_code;
-	__le16 req_type;
-	__le16 seq_id;
-	__le16 resp_len;
-	__le32 unused_0;
-	u8 unused_1;
-	u8 unused_2;
-	u8 unused_3;
-	u8 valid;
-};
-
-/* hwrm_reject_fwd_resp */
-/* Input (128 bytes) */
-struct hwrm_reject_fwd_resp_input {
-	__le16 req_type;
-	__le16 cmpl_ring;
-	__le16 seq_id;
-	__le16 target_id;
-	__le64 resp_addr;
-	__le32 encap_request[26];
-	__le16 encap_resp_target_id;
-	__le16 unused_0[3];
-};
-
-/* Output (16 bytes) */
-struct hwrm_reject_fwd_resp_output {
-	__le16 error_code;
-	__le16 req_type;
-	__le16 seq_id;
-	__le16 resp_len;
-	__le32 unused_0;
-	u8 unused_1;
-	u8 unused_2;
-	u8 unused_3;
-	u8 valid;
-};
-
-/* hwrm_fwd_resp */
-/* Input (40 bytes) */
-struct hwrm_fwd_resp_input {
-	__le16 req_type;
-	__le16 cmpl_ring;
-	__le16 seq_id;
-	__le16 target_id;
-	__le64 resp_addr;
-	__le16 encap_resp_target_id;
-	__le16 encap_resp_cmpl_ring;
-	__le16 encap_resp_len;
-	u8 unused_0;
-	u8 unused_1;
-	__le64 encap_resp_addr;
-	__le32 encap_resp[24];
-};
-
-/* Output (16 bytes) */
-struct hwrm_fwd_resp_output {
-	__le16 error_code;
-	__le16 req_type;
-	__le16 seq_id;
-	__le16 resp_len;
-	__le32 unused_0;
-	u8 unused_1;
-	u8 unused_2;
-	u8 unused_3;
-	u8 valid;
-};
-
-/* hwrm_fwd_async_event_cmpl */
-/* Input (32 bytes) */
-struct hwrm_fwd_async_event_cmpl_input {
-	__le16 req_type;
-	__le16 cmpl_ring;
-	__le16 seq_id;
-	__le16 target_id;
-	__le64 resp_addr;
-	__le16 encap_async_event_target_id;
-	u8 unused_0;
-	u8 unused_1;
-	u8 unused_2[3];
-	u8 unused_3;
-	__le32 encap_async_event_cmpl[4];
-};
-
-/* Output (16 bytes) */
-struct hwrm_fwd_async_event_cmpl_output {
-	__le16 error_code;
-	__le16 req_type;
-	__le16 seq_id;
-	__le16 resp_len;
-	__le32 unused_0;
-	u8 unused_1;
-	u8 unused_2;
-	u8 unused_3;
-	u8 valid;
-};
-
-/* hwrm_temp_monitor_query */
-/* Input (16 bytes) */
-struct hwrm_temp_monitor_query_input {
-	__le16 req_type;
-	__le16 cmpl_ring;
-	__le16 seq_id;
-	__le16 target_id;
-	__le64 resp_addr;
-};
-
-/* Output (16 bytes) */
-struct hwrm_temp_monitor_query_output {
-	__le16 error_code;
-	__le16 req_type;
-	__le16 seq_id;
-	__le16 resp_len;
-	u8 temp;
-	u8 unused_0;
-	__le16 unused_1;
-	u8 unused_2;
-	u8 unused_3;
-	u8 unused_4;
-	u8 valid;
-};
-
-/* hwrm_wol_filter_alloc */
-/* Input (64 bytes) */
-struct hwrm_wol_filter_alloc_input {
-	__le16 req_type;
-	__le16 cmpl_ring;
-	__le16 seq_id;
-	__le16 target_id;
-	__le64 resp_addr;
-	__le32 flags;
-	__le32 enables;
-	#define WOL_FILTER_ALLOC_REQ_ENABLES_MAC_ADDRESS	    0x1UL
-	#define WOL_FILTER_ALLOC_REQ_ENABLES_PATTERN_OFFSET	    0x2UL
-	#define WOL_FILTER_ALLOC_REQ_ENABLES_PATTERN_BUF_SIZE      0x4UL
-	#define WOL_FILTER_ALLOC_REQ_ENABLES_PATTERN_BUF_ADDR      0x8UL
-	#define WOL_FILTER_ALLOC_REQ_ENABLES_PATTERN_MASK_ADDR     0x10UL
-	#define WOL_FILTER_ALLOC_REQ_ENABLES_PATTERN_MASK_SIZE     0x20UL
-	__le16 port_id;
-	u8 wol_type;
-	#define WOL_FILTER_ALLOC_REQ_WOL_TYPE_MAGICPKT		   0x0UL
-	#define WOL_FILTER_ALLOC_REQ_WOL_TYPE_BMP		   0x1UL
-	#define WOL_FILTER_ALLOC_REQ_WOL_TYPE_INVALID		   0xffUL
-	u8 unused_0;
-	__le32 unused_1;
-	u8 mac_address[6];
-	__le16 pattern_offset;
-	__le16 pattern_buf_size;
-	__le16 pattern_mask_size;
-	__le32 unused_2;
-	__le64 pattern_buf_addr;
-	__le64 pattern_mask_addr;
-};
-
-/* Output (16 bytes) */
-struct hwrm_wol_filter_alloc_output {
-	__le16 error_code;
-	__le16 req_type;
-	__le16 seq_id;
-	__le16 resp_len;
-	u8 wol_filter_id;
-	u8 unused_0;
-	__le16 unused_1;
-	u8 unused_2;
-	u8 unused_3;
-	u8 unused_4;
-	u8 valid;
-};
-
-/* hwrm_wol_filter_free */
-/* Input (32 bytes) */
-struct hwrm_wol_filter_free_input {
-	__le16 req_type;
-	__le16 cmpl_ring;
-	__le16 seq_id;
-	__le16 target_id;
-	__le64 resp_addr;
-	__le32 flags;
-	#define WOL_FILTER_FREE_REQ_FLAGS_FREE_ALL_WOL_FILTERS     0x1UL
-	__le32 enables;
-	#define WOL_FILTER_FREE_REQ_ENABLES_WOL_FILTER_ID	    0x1UL
-	__le16 port_id;
-	u8 wol_filter_id;
-	u8 unused_0[5];
-};
-
-/* Output (16 bytes) */
-struct hwrm_wol_filter_free_output {
-	__le16 error_code;
-	__le16 req_type;
-	__le16 seq_id;
-	__le16 resp_len;
-	__le32 unused_0;
-	u8 unused_1;
-	u8 unused_2;
-	u8 unused_3;
-	u8 valid;
-};
-
-/* hwrm_wol_filter_qcfg */
-/* Input (56 bytes) */
-struct hwrm_wol_filter_qcfg_input {
-	__le16 req_type;
-	__le16 cmpl_ring;
-	__le16 seq_id;
-	__le16 target_id;
-	__le64 resp_addr;
-	__le16 port_id;
-	__le16 handle;
-	__le32 unused_0;
-	__le64 pattern_buf_addr;
-	__le16 pattern_buf_size;
-	u8 unused_1;
-	u8 unused_2;
-	u8 unused_3[3];
-	u8 unused_4;
-	__le64 pattern_mask_addr;
-	__le16 pattern_mask_size;
-	__le16 unused_5[3];
-};
-
-/* Output (32 bytes) */
-struct hwrm_wol_filter_qcfg_output {
-	__le16 error_code;
-	__le16 req_type;
-	__le16 seq_id;
-	__le16 resp_len;
-	__le16 next_handle;
-	u8 wol_filter_id;
-	u8 wol_type;
-	#define WOL_FILTER_QCFG_RESP_WOL_TYPE_MAGICPKT		   0x0UL
-	#define WOL_FILTER_QCFG_RESP_WOL_TYPE_BMP		   0x1UL
-	#define WOL_FILTER_QCFG_RESP_WOL_TYPE_INVALID		   0xffUL
-	__le32 unused_0;
-	u8 mac_address[6];
-	__le16 pattern_offset;
-	__le16 pattern_size;
-	__le16 pattern_mask_size;
-	u8 unused_1;
-	u8 unused_2;
-	u8 unused_3;
-	u8 valid;
-};
-
-/* hwrm_wol_reason_qcfg */
-/* Input (40 bytes) */
-struct hwrm_wol_reason_qcfg_input {
-	__le16 req_type;
-	__le16 cmpl_ring;
-	__le16 seq_id;
-	__le16 target_id;
-	__le64 resp_addr;
-	__le16 port_id;
-	u8 unused_0;
-	u8 unused_1;
-	u8 unused_2[3];
-	u8 unused_3;
-	__le64 wol_pkt_buf_addr;
-	__le16 wol_pkt_buf_size;
-	__le16 unused_4[3];
-};
-
-/* Output (16 bytes) */
-struct hwrm_wol_reason_qcfg_output {
-	__le16 error_code;
-	__le16 req_type;
-	__le16 seq_id;
-	__le16 resp_len;
-	u8 wol_filter_id;
-	u8 wol_reason;
-	#define WOL_REASON_QCFG_RESP_WOL_REASON_MAGICPKT	   0x0UL
-	#define WOL_REASON_QCFG_RESP_WOL_REASON_BMP		   0x1UL
-	#define WOL_REASON_QCFG_RESP_WOL_REASON_INVALID	   0xffUL
-	u8 wol_pkt_len;
-	u8 unused_0;
-	u8 unused_1;
-	u8 unused_2;
-	u8 unused_3;
-	u8 valid;
-};
-
-/* hwrm_nvm_read */
-/* Input (40 bytes) */
-struct hwrm_nvm_read_input {
-	__le16 req_type;
-	__le16 cmpl_ring;
-	__le16 seq_id;
-	__le16 target_id;
-	__le64 resp_addr;
-	__le64 host_dest_addr;
-	__le16 dir_idx;
-	u8 unused_0;
-	u8 unused_1;
-	__le32 offset;
-	__le32 len;
-	__le32 unused_2;
-};
-
-/* Output (16 bytes) */
-struct hwrm_nvm_read_output {
-	__le16 error_code;
-	__le16 req_type;
-	__le16 seq_id;
-	__le16 resp_len;
-	__le32 unused_0;
-	u8 unused_1;
-	u8 unused_2;
-	u8 unused_3;
-	u8 valid;
-};
-
-/* hwrm_nvm_get_dir_entries */
-/* Input (24 bytes) */
-struct hwrm_nvm_get_dir_entries_input {
-	__le16 req_type;
-	__le16 cmpl_ring;
-	__le16 seq_id;
-	__le16 target_id;
-	__le64 resp_addr;
-	__le64 host_dest_addr;
-};
-
-/* Output (16 bytes) */
-struct hwrm_nvm_get_dir_entries_output {
-	__le16 error_code;
-	__le16 req_type;
-	__le16 seq_id;
-	__le16 resp_len;
-	__le32 unused_0;
-	u8 unused_1;
-	u8 unused_2;
-	u8 unused_3;
-	u8 valid;
-};
-
-/* hwrm_nvm_get_dir_info */
-/* Input (16 bytes) */
-struct hwrm_nvm_get_dir_info_input {
-	__le16 req_type;
-	__le16 cmpl_ring;
-	__le16 seq_id;
-	__le16 target_id;
-	__le64 resp_addr;
-};
-
-/* Output (24 bytes) */
-struct hwrm_nvm_get_dir_info_output {
-	__le16 error_code;
-	__le16 req_type;
-	__le16 seq_id;
-	__le16 resp_len;
-	__le32 entries;
-	__le32 entry_length;
-	__le32 unused_0;
-	u8 unused_1;
-	u8 unused_2;
-	u8 unused_3;
-	u8 valid;
-};
-
-/* hwrm_nvm_write */
-/* Input (48 bytes) */
-struct hwrm_nvm_write_input {
-	__le16 req_type;
-	__le16 cmpl_ring;
-	__le16 seq_id;
-	__le16 target_id;
-	__le64 resp_addr;
-	__le64 host_src_addr;
-	__le16 dir_type;
-	__le16 dir_ordinal;
-	__le16 dir_ext;
-	__le16 dir_attr;
-	__le32 dir_data_length;
-	__le16 option;
-	__le16 flags;
-	#define NVM_WRITE_REQ_FLAGS_KEEP_ORIG_ACTIVE_IMG	    0x1UL
-	__le32 dir_item_length;
-	__le32 unused_0;
-};
-
-/* Output (16 bytes) */
-struct hwrm_nvm_write_output {
-	__le16 error_code;
-	__le16 req_type;
-	__le16 seq_id;
-	__le16 resp_len;
-	__le32 dir_item_length;
-	__le16 dir_idx;
-	u8 unused_0;
-	u8 valid;
-};
-
-/* Command specific Error Codes (8 bytes) */
-struct hwrm_nvm_write_cmd_err {
-	u8 code;
-	#define NVM_WRITE_CMD_ERR_CODE_UNKNOWN			   0x0UL
-	#define NVM_WRITE_CMD_ERR_CODE_FRAG_ERR		   0x1UL
-	#define NVM_WRITE_CMD_ERR_CODE_NO_SPACE		   0x2UL
-	u8 unused_0[7];
-};
-
-/* hwrm_nvm_modify */
-/* Input (40 bytes) */
-struct hwrm_nvm_modify_input {
-	__le16 req_type;
-	__le16 cmpl_ring;
-	__le16 seq_id;
-	__le16 target_id;
-	__le64 resp_addr;
-	__le64 host_src_addr;
-	__le16 dir_idx;
-	u8 unused_0;
-	u8 unused_1;
-	__le32 offset;
-	__le32 len;
-	__le32 unused_2;
-};
-
-/* Output (16 bytes) */
-struct hwrm_nvm_modify_output {
-	__le16 error_code;
-	__le16 req_type;
-	__le16 seq_id;
-	__le16 resp_len;
-	__le32 unused_0;
-	u8 unused_1;
-	u8 unused_2;
-	u8 unused_3;
-	u8 valid;
-};
-
-/* hwrm_nvm_find_dir_entry */
-/* Input (32 bytes) */
-struct hwrm_nvm_find_dir_entry_input {
-	__le16 req_type;
-	__le16 cmpl_ring;
-	__le16 seq_id;
-	__le16 target_id;
-	__le64 resp_addr;
-	__le32 enables;
-	#define NVM_FIND_DIR_ENTRY_REQ_ENABLES_DIR_IDX_VALID       0x1UL
-	__le16 dir_idx;
-	__le16 dir_type;
-	__le16 dir_ordinal;
-	__le16 dir_ext;
-	u8 opt_ordinal;
-	#define NVM_FIND_DIR_ENTRY_REQ_OPT_ORDINAL_MASK	    0x3UL
-	#define NVM_FIND_DIR_ENTRY_REQ_OPT_ORDINAL_SFT		    0
-	#define NVM_FIND_DIR_ENTRY_REQ_OPT_ORDINAL_EQ		   0x0UL
-	#define NVM_FIND_DIR_ENTRY_REQ_OPT_ORDINAL_GE		   0x1UL
-	#define NVM_FIND_DIR_ENTRY_REQ_OPT_ORDINAL_GT		   0x2UL
-	u8 unused_1[3];
-};
-
-/* Output (32 bytes) */
-struct hwrm_nvm_find_dir_entry_output {
-	__le16 error_code;
-	__le16 req_type;
-	__le16 seq_id;
-	__le16 resp_len;
-	__le32 dir_item_length;
-	__le32 dir_data_length;
-	__le32 fw_ver;
-	__le16 dir_ordinal;
-	__le16 dir_idx;
-	__le32 unused_0;
-	u8 unused_1;
-	u8 unused_2;
-	u8 unused_3;
-	u8 valid;
-};
-
-/* hwrm_nvm_erase_dir_entry */
-/* Input (24 bytes) */
-struct hwrm_nvm_erase_dir_entry_input {
-	__le16 req_type;
-	__le16 cmpl_ring;
-	__le16 seq_id;
-	__le16 target_id;
-	__le64 resp_addr;
-	__le16 dir_idx;
-	__le16 unused_0[3];
-};
-
-/* Output (16 bytes) */
-struct hwrm_nvm_erase_dir_entry_output {
-	__le16 error_code;
-	__le16 req_type;
-	__le16 seq_id;
-	__le16 resp_len;
-	__le32 unused_0;
-	u8 unused_1;
-	u8 unused_2;
-	u8 unused_3;
-	u8 valid;
-};
-
-/* hwrm_nvm_get_dev_info */
-/* Input (16 bytes) */
-struct hwrm_nvm_get_dev_info_input {
-	__le16 req_type;
-	__le16 cmpl_ring;
-	__le16 seq_id;
-	__le16 target_id;
-	__le64 resp_addr;
-};
-
-/* Output (32 bytes) */
-struct hwrm_nvm_get_dev_info_output {
-	__le16 error_code;
-	__le16 req_type;
-	__le16 seq_id;
-	__le16 resp_len;
-	__le16 manufacturer_id;
-	__le16 device_id;
-	__le32 sector_size;
-	__le32 nvram_size;
-	__le32 reserved_size;
-	__le32 available_size;
-	u8 unused_0;
-	u8 unused_1;
-	u8 unused_2;
-	u8 valid;
-};
-
-/* hwrm_nvm_mod_dir_entry */
-/* Input (32 bytes) */
-struct hwrm_nvm_mod_dir_entry_input {
-	__le16 req_type;
-	__le16 cmpl_ring;
-	__le16 seq_id;
-	__le16 target_id;
-	__le64 resp_addr;
-	__le32 enables;
-	#define NVM_MOD_DIR_ENTRY_REQ_ENABLES_CHECKSUM		    0x1UL
-	__le16 dir_idx;
-	__le16 dir_ordinal;
-	__le16 dir_ext;
-	__le16 dir_attr;
-	__le32 checksum;
-};
-
-/* Output (16 bytes) */
-struct hwrm_nvm_mod_dir_entry_output {
-	__le16 error_code;
-	__le16 req_type;
-	__le16 seq_id;
-	__le16 resp_len;
-	__le32 unused_0;
-	u8 unused_1;
-	u8 unused_2;
-	u8 unused_3;
-	u8 valid;
-};
-
-/* hwrm_nvm_verify_update */
-/* Input (24 bytes) */
-struct hwrm_nvm_verify_update_input {
-	__le16 req_type;
-	__le16 cmpl_ring;
-	__le16 seq_id;
-	__le16 target_id;
-	__le64 resp_addr;
-	__le16 dir_type;
-	__le16 dir_ordinal;
-	__le16 dir_ext;
-	__le16 unused_0;
-};
-
-/* Output (16 bytes) */
-struct hwrm_nvm_verify_update_output {
-	__le16 error_code;
-	__le16 req_type;
-	__le16 seq_id;
-	__le16 resp_len;
-	__le32 unused_0;
-	u8 unused_1;
-	u8 unused_2;
-	u8 unused_3;
-	u8 valid;
-};
-
-/* hwrm_nvm_install_update */
-/* Input (24 bytes) */
-struct hwrm_nvm_install_update_input {
-	__le16 req_type;
-	__le16 cmpl_ring;
-	__le16 seq_id;
-	__le16 target_id;
-	__le64 resp_addr;
-	__le32 install_type;
-	#define NVM_INSTALL_UPDATE_REQ_INSTALL_TYPE_NORMAL	   0x0UL
-	#define NVM_INSTALL_UPDATE_REQ_INSTALL_TYPE_ALL	   0xffffffffUL
-	__le16 flags;
-	#define NVM_INSTALL_UPDATE_REQ_FLAGS_ERASE_UNUSED_SPACE    0x1UL
-	#define NVM_INSTALL_UPDATE_REQ_FLAGS_REMOVE_UNUSED_PKG     0x2UL
-	#define NVM_INSTALL_UPDATE_REQ_FLAGS_ALLOWED_TO_DEFRAG     0x4UL
-	__le16 unused_0;
-};
-
-/* Output (24 bytes) */
-struct hwrm_nvm_install_update_output {
-	__le16 error_code;
-	__le16 req_type;
-	__le16 seq_id;
-	__le16 resp_len;
-	__le64 installed_items;
-	u8 result;
-	#define NVM_INSTALL_UPDATE_RESP_RESULT_SUCCESS		   0x0UL
-	u8 problem_item;
-	#define NVM_INSTALL_UPDATE_RESP_PROBLEM_ITEM_NONE	   0x0UL
-	#define NVM_INSTALL_UPDATE_RESP_PROBLEM_ITEM_PACKAGE      0xffUL
-	u8 reset_required;
-	#define NVM_INSTALL_UPDATE_RESP_RESET_REQUIRED_NONE       0x0UL
-	#define NVM_INSTALL_UPDATE_RESP_RESET_REQUIRED_PCI	   0x1UL
-	#define NVM_INSTALL_UPDATE_RESP_RESET_REQUIRED_POWER      0x2UL
-	u8 unused_0;
-	u8 unused_1;
-	u8 unused_2;
-	u8 unused_3;
-	u8 valid;
-};
-
-/* Command specific Error Codes (8 bytes) */
-struct hwrm_nvm_install_update_cmd_err {
-	u8 code;
-	#define NVM_INSTALL_UPDATE_CMD_ERR_CODE_UNKNOWN	   0x0UL
-	#define NVM_INSTALL_UPDATE_CMD_ERR_CODE_FRAG_ERR	   0x1UL
-	#define NVM_INSTALL_UPDATE_CMD_ERR_CODE_NO_SPACE	   0x2UL
-	u8 unused_0[7];
-};
-
-/* hwrm_selftest_qlist */
-/* Input (16 bytes) */
-struct hwrm_selftest_qlist_input {
-	__le16 req_type;
-	__le16 cmpl_ring;
-	__le16 seq_id;
-	__le16 target_id;
-	__le64 resp_addr;
-};
-
-/* Output (248 bytes) */
-struct hwrm_selftest_qlist_output {
-	__le16 error_code;
-	__le16 req_type;
-	__le16 seq_id;
-	__le16 resp_len;
-	u8 num_tests;
-	u8 available_tests;
-	#define SELFTEST_QLIST_RESP_AVAILABLE_TESTS_NVM_TEST       0x1UL
-	#define SELFTEST_QLIST_RESP_AVAILABLE_TESTS_LINK_TEST      0x2UL
-	#define SELFTEST_QLIST_RESP_AVAILABLE_TESTS_REGISTER_TEST  0x4UL
-	#define SELFTEST_QLIST_RESP_AVAILABLE_TESTS_MEMORY_TEST    0x8UL
-	u8 offline_tests;
-	#define SELFTEST_QLIST_RESP_OFFLINE_TESTS_NVM_TEST	    0x1UL
-	#define SELFTEST_QLIST_RESP_OFFLINE_TESTS_LINK_TEST	    0x2UL
-	#define SELFTEST_QLIST_RESP_OFFLINE_TESTS_REGISTER_TEST    0x4UL
-	#define SELFTEST_QLIST_RESP_OFFLINE_TESTS_MEMORY_TEST      0x8UL
-	u8 unused_0;
-	__le16 test_timeout;
-	u8 unused_1;
-	u8 unused_2;
-	char test0_name[32];
-	char test1_name[32];
-	char test2_name[32];
-	char test3_name[32];
-	char test4_name[32];
-	char test5_name[32];
-	char test6_name[32];
-	char test7_name[32];
-};
-
-/* hwrm_selftest_exec */
-/* Input (24 bytes) */
-struct hwrm_selftest_exec_input {
-	__le16 req_type;
-	__le16 cmpl_ring;
-	__le16 seq_id;
-	__le16 target_id;
-	__le64 resp_addr;
-	u8 flags;
-	#define SELFTEST_EXEC_REQ_FLAGS_NVM_TEST		    0x1UL
-	#define SELFTEST_EXEC_REQ_FLAGS_LINK_TEST		    0x2UL
-	#define SELFTEST_EXEC_REQ_FLAGS_REGISTER_TEST		    0x4UL
-	#define SELFTEST_EXEC_REQ_FLAGS_MEMORY_TEST		    0x8UL
-	u8 unused_0[7];
-};
-
-/* Output (16 bytes) */
-struct hwrm_selftest_exec_output {
-	__le16 error_code;
-	__le16 req_type;
-	__le16 seq_id;
-	__le16 resp_len;
-	u8 requested_tests;
-	#define SELFTEST_EXEC_RESP_REQUESTED_TESTS_NVM_TEST	    0x1UL
-	#define SELFTEST_EXEC_RESP_REQUESTED_TESTS_LINK_TEST       0x2UL
-	#define SELFTEST_EXEC_RESP_REQUESTED_TESTS_REGISTER_TEST   0x4UL
-	#define SELFTEST_EXEC_RESP_REQUESTED_TESTS_MEMORY_TEST     0x8UL
-	u8 test_success;
-	#define SELFTEST_EXEC_RESP_TEST_SUCCESS_NVM_TEST	    0x1UL
-	#define SELFTEST_EXEC_RESP_TEST_SUCCESS_LINK_TEST	    0x2UL
-	#define SELFTEST_EXEC_RESP_TEST_SUCCESS_REGISTER_TEST      0x4UL
-	#define SELFTEST_EXEC_RESP_TEST_SUCCESS_MEMORY_TEST	    0x8UL
-	__le16 unused_0[3];
-};
-
-/* hwrm_selftest_irq */
-/* Input (16 bytes) */
-struct hwrm_selftest_irq_input {
-	__le16 req_type;
-	__le16 cmpl_ring;
-	__le16 seq_id;
-	__le16 target_id;
-	__le64 resp_addr;
-};
-
-/* Output (8 bytes) */
-struct hwrm_selftest_irq_output {
-	__le16 error_code;
-	__le16 req_type;
-	__le16 seq_id;
-	__le16 resp_len;
-};
-
-/* Hardware Resource Manager Specification */
-/* Input (16 bytes) */
-struct input {
-	__le16 req_type;
-	__le16 cmpl_ring;
-	__le16 seq_id;
-	__le16 target_id;
-	__le64 resp_addr;
-};
-
-/* Output (8 bytes) */
-struct output {
-	__le16 error_code;
-	__le16 req_type;
-	__le16 seq_id;
-	__le16 resp_len;
-};
-
-/* Short Command Structure (16 bytes) */
-struct hwrm_short_input {
-	__le16 req_type;
-	__le16 signature;
-	#define SHORT_REQ_SIGNATURE_SHORT_CMD			   0x4321UL
-	__le16 unused_0;
-	__le16 size;
-	__le64 req_addr;
-};
-
-/* Command numbering (8 bytes) */
-struct cmd_nums {
-	__le16 req_type;
-	#define HWRM_VER_GET					   (0x0UL)
-	#define HWRM_FUNC_BUF_UNRGTR				   (0xeUL)
-	#define HWRM_FUNC_VF_CFG				   (0xfUL)
-	#define RESERVED1					   (0x10UL)
-	#define HWRM_FUNC_RESET				   (0x11UL)
-	#define HWRM_FUNC_GETFID				   (0x12UL)
-	#define HWRM_FUNC_VF_ALLOC				   (0x13UL)
-	#define HWRM_FUNC_VF_FREE				   (0x14UL)
-	#define HWRM_FUNC_QCAPS				   (0x15UL)
-	#define HWRM_FUNC_QCFG					   (0x16UL)
-	#define HWRM_FUNC_CFG					   (0x17UL)
-	#define HWRM_FUNC_QSTATS				   (0x18UL)
-	#define HWRM_FUNC_CLR_STATS				   (0x19UL)
-	#define HWRM_FUNC_DRV_UNRGTR				   (0x1aUL)
-	#define HWRM_FUNC_VF_RESC_FREE				   (0x1bUL)
-	#define HWRM_FUNC_VF_VNIC_IDS_QUERY			   (0x1cUL)
-	#define HWRM_FUNC_DRV_RGTR				   (0x1dUL)
-	#define HWRM_FUNC_DRV_QVER				   (0x1eUL)
-	#define HWRM_FUNC_BUF_RGTR				   (0x1fUL)
-	#define HWRM_PORT_PHY_CFG				   (0x20UL)
-	#define HWRM_PORT_MAC_CFG				   (0x21UL)
-	#define HWRM_PORT_TS_QUERY				   (0x22UL)
-	#define HWRM_PORT_QSTATS				   (0x23UL)
-	#define HWRM_PORT_LPBK_QSTATS				   (0x24UL)
-	#define HWRM_PORT_CLR_STATS				   (0x25UL)
-	#define HWRM_PORT_LPBK_CLR_STATS			   (0x26UL)
-	#define HWRM_PORT_PHY_QCFG				   (0x27UL)
-	#define HWRM_PORT_MAC_QCFG				   (0x28UL)
-	#define RESERVED7					   (0x29UL)
-	#define HWRM_PORT_PHY_QCAPS				   (0x2aUL)
-	#define HWRM_PORT_PHY_I2C_WRITE			   (0x2bUL)
-	#define HWRM_PORT_PHY_I2C_READ				   (0x2cUL)
-	#define HWRM_PORT_LED_CFG				   (0x2dUL)
-	#define HWRM_PORT_LED_QCFG				   (0x2eUL)
-	#define HWRM_PORT_LED_QCAPS				   (0x2fUL)
-	#define HWRM_QUEUE_QPORTCFG				   (0x30UL)
-	#define HWRM_QUEUE_QCFG				   (0x31UL)
-	#define HWRM_QUEUE_CFG					   (0x32UL)
-	#define RESERVED2					   (0x33UL)
-	#define RESERVED3					   (0x34UL)
-	#define HWRM_QUEUE_PFCENABLE_QCFG			   (0x35UL)
-	#define HWRM_QUEUE_PFCENABLE_CFG			   (0x36UL)
-	#define HWRM_QUEUE_PRI2COS_QCFG			   (0x37UL)
-	#define HWRM_QUEUE_PRI2COS_CFG				   (0x38UL)
-	#define HWRM_QUEUE_COS2BW_QCFG				   (0x39UL)
-	#define HWRM_QUEUE_COS2BW_CFG				   (0x3aUL)
-	#define HWRM_VNIC_ALLOC				   (0x40UL)
-	#define HWRM_VNIC_FREE					   (0x41UL)
-	#define HWRM_VNIC_CFG					   (0x42UL)
-	#define HWRM_VNIC_QCFG					   (0x43UL)
-	#define HWRM_VNIC_TPA_CFG				   (0x44UL)
-	#define HWRM_VNIC_TPA_QCFG				   (0x45UL)
-	#define HWRM_VNIC_RSS_CFG				   (0x46UL)
-	#define HWRM_VNIC_RSS_QCFG				   (0x47UL)
-	#define HWRM_VNIC_PLCMODES_CFG				   (0x48UL)
-	#define HWRM_VNIC_PLCMODES_QCFG			   (0x49UL)
-	#define HWRM_VNIC_QCAPS				   (0x4aUL)
-	#define HWRM_RING_ALLOC				   (0x50UL)
-	#define HWRM_RING_FREE					   (0x51UL)
-	#define HWRM_RING_CMPL_RING_QAGGINT_PARAMS		   (0x52UL)
-	#define HWRM_RING_CMPL_RING_CFG_AGGINT_PARAMS		   (0x53UL)
-	#define HWRM_RING_RESET				   (0x5eUL)
-	#define HWRM_RING_GRP_ALLOC				   (0x60UL)
-	#define HWRM_RING_GRP_FREE				   (0x61UL)
-	#define RESERVED5					   (0x64UL)
-	#define RESERVED6					   (0x65UL)
-	#define HWRM_VNIC_RSS_COS_LB_CTX_ALLOC			   (0x70UL)
-	#define HWRM_VNIC_RSS_COS_LB_CTX_FREE			   (0x71UL)
-	#define HWRM_CFA_L2_FILTER_ALLOC			   (0x90UL)
-	#define HWRM_CFA_L2_FILTER_FREE			   (0x91UL)
-	#define HWRM_CFA_L2_FILTER_CFG				   (0x92UL)
-	#define HWRM_CFA_L2_SET_RX_MASK			   (0x93UL)
-	#define RESERVED4					   (0x94UL)
-	#define HWRM_CFA_TUNNEL_FILTER_ALLOC			   (0x95UL)
-	#define HWRM_CFA_TUNNEL_FILTER_FREE			   (0x96UL)
-	#define HWRM_CFA_ENCAP_RECORD_ALLOC			   (0x97UL)
-	#define HWRM_CFA_ENCAP_RECORD_FREE			   (0x98UL)
-	#define HWRM_CFA_NTUPLE_FILTER_ALLOC			   (0x99UL)
-	#define HWRM_CFA_NTUPLE_FILTER_FREE			   (0x9aUL)
-	#define HWRM_CFA_NTUPLE_FILTER_CFG			   (0x9bUL)
-	#define HWRM_CFA_EM_FLOW_ALLOC				   (0x9cUL)
-	#define HWRM_CFA_EM_FLOW_FREE				   (0x9dUL)
-	#define HWRM_CFA_EM_FLOW_CFG				   (0x9eUL)
-	#define HWRM_TUNNEL_DST_PORT_QUERY			   (0xa0UL)
-	#define HWRM_TUNNEL_DST_PORT_ALLOC			   (0xa1UL)
-	#define HWRM_TUNNEL_DST_PORT_FREE			   (0xa2UL)
-	#define HWRM_STAT_CTX_ALLOC				   (0xb0UL)
-	#define HWRM_STAT_CTX_FREE				   (0xb1UL)
-	#define HWRM_STAT_CTX_QUERY				   (0xb2UL)
-	#define HWRM_STAT_CTX_CLR_STATS			   (0xb3UL)
-	#define HWRM_FW_RESET					   (0xc0UL)
-	#define HWRM_FW_QSTATUS				   (0xc1UL)
-	#define HWRM_FW_SET_TIME				   (0xc8UL)
-	#define HWRM_FW_GET_TIME				   (0xc9UL)
-	#define HWRM_FW_SET_STRUCTURED_DATA			   (0xcaUL)
-	#define HWRM_FW_GET_STRUCTURED_DATA			   (0xcbUL)
-	#define HWRM_FW_IPC_MAILBOX				   (0xccUL)
-	#define HWRM_EXEC_FWD_RESP				   (0xd0UL)
-	#define HWRM_REJECT_FWD_RESP				   (0xd1UL)
-	#define HWRM_FWD_RESP					   (0xd2UL)
-	#define HWRM_FWD_ASYNC_EVENT_CMPL			   (0xd3UL)
-	#define HWRM_TEMP_MONITOR_QUERY			   (0xe0UL)
-	#define HWRM_WOL_FILTER_ALLOC				   (0xf0UL)
-	#define HWRM_WOL_FILTER_FREE				   (0xf1UL)
-	#define HWRM_WOL_FILTER_QCFG				   (0xf2UL)
-	#define HWRM_WOL_REASON_QCFG				   (0xf3UL)
-	#define HWRM_CFA_METER_PROFILE_ALLOC			   (0xf5UL)
-	#define HWRM_CFA_METER_PROFILE_FREE			   (0xf6UL)
-	#define HWRM_CFA_METER_PROFILE_CFG			   (0xf7UL)
-	#define HWRM_CFA_METER_INSTANCE_ALLOC			   (0xf8UL)
-	#define HWRM_CFA_METER_INSTANCE_FREE			   (0xf9UL)
-	#define HWRM_CFA_VF_PAIR_ALLOC				   (0x100UL)
-	#define HWRM_CFA_VF_PAIR_FREE				   (0x101UL)
-	#define HWRM_CFA_VF_PAIR_INFO				   (0x102UL)
-	#define HWRM_CFA_FLOW_ALLOC				   (0x103UL)
-	#define HWRM_CFA_FLOW_FREE				   (0x104UL)
-	#define HWRM_CFA_FLOW_FLUSH				   (0x105UL)
-	#define HWRM_CFA_FLOW_STATS				   (0x106UL)
-	#define HWRM_CFA_FLOW_INFO				   (0x107UL)
-	#define HWRM_SELFTEST_QLIST				   (0x200UL)
-	#define HWRM_SELFTEST_EXEC				   (0x201UL)
-	#define HWRM_SELFTEST_IRQ				   (0x202UL)
-	#define HWRM_DBG_READ_DIRECT				   (0xff10UL)
-	#define HWRM_DBG_READ_INDIRECT				   (0xff11UL)
-	#define HWRM_DBG_WRITE_DIRECT				   (0xff12UL)
-	#define HWRM_DBG_WRITE_INDIRECT			   (0xff13UL)
-	#define HWRM_DBG_DUMP					   (0xff14UL)
-	#define HWRM_NVM_FACTORY_DEFAULTS			   (0xffeeUL)
-	#define HWRM_NVM_VALIDATE_OPTION			   (0xffefUL)
-	#define HWRM_NVM_FLUSH					   (0xfff0UL)
-	#define HWRM_NVM_GET_VARIABLE				   (0xfff1UL)
-	#define HWRM_NVM_SET_VARIABLE				   (0xfff2UL)
-	#define HWRM_NVM_INSTALL_UPDATE			   (0xfff3UL)
-	#define HWRM_NVM_MODIFY				   (0xfff4UL)
-	#define HWRM_NVM_VERIFY_UPDATE				   (0xfff5UL)
-	#define HWRM_NVM_GET_DEV_INFO				   (0xfff6UL)
-	#define HWRM_NVM_ERASE_DIR_ENTRY			   (0xfff7UL)
-	#define HWRM_NVM_MOD_DIR_ENTRY				   (0xfff8UL)
-	#define HWRM_NVM_FIND_DIR_ENTRY			   (0xfff9UL)
-	#define HWRM_NVM_GET_DIR_ENTRIES			   (0xfffaUL)
-	#define HWRM_NVM_GET_DIR_INFO				   (0xfffbUL)
-	#define HWRM_NVM_RAW_DUMP				   (0xfffcUL)
-	#define HWRM_NVM_READ					   (0xfffdUL)
-	#define HWRM_NVM_WRITE					   (0xfffeUL)
-	#define HWRM_NVM_RAW_WRITE_BLK				   (0xffffUL)
-	__le16 unused_0[3];
-};
-
-/* Return Codes (8 bytes) */
-struct ret_codes {
-	__le16 error_code;
-	#define HWRM_ERR_CODE_SUCCESS				   (0x0UL)
-	#define HWRM_ERR_CODE_FAIL				   (0x1UL)
-	#define HWRM_ERR_CODE_INVALID_PARAMS			   (0x2UL)
-	#define HWRM_ERR_CODE_RESOURCE_ACCESS_DENIED		   (0x3UL)
-	#define HWRM_ERR_CODE_RESOURCE_ALLOC_ERROR		   (0x4UL)
-	#define HWRM_ERR_CODE_INVALID_FLAGS			   (0x5UL)
-	#define HWRM_ERR_CODE_INVALID_ENABLES			   (0x6UL)
-	#define HWRM_ERR_CODE_HWRM_ERROR			   (0xfUL)
-	#define HWRM_ERR_CODE_UNKNOWN_ERR			   (0xfffeUL)
-	#define HWRM_ERR_CODE_CMD_NOT_SUPPORTED		   (0xffffUL)
-	__le16 unused_0[3];
-};
-
-/* Output (16 bytes) */
-struct hwrm_err_output {
-	__le16 error_code;
-	__le16 req_type;
-	__le16 seq_id;
-	__le16 resp_len;
-	__le32 opaque_0;
-	__le16 opaque_1;
-	u8 cmd_err;
-	u8 valid;
-};
-
-/* Port Tx Statistics Formats (408 bytes) */
-struct tx_port_stats {
-	__le64 tx_64b_frames;
-	__le64 tx_65b_127b_frames;
-	__le64 tx_128b_255b_frames;
-	__le64 tx_256b_511b_frames;
-	__le64 tx_512b_1023b_frames;
-	__le64 tx_1024b_1518_frames;
-	__le64 tx_good_vlan_frames;
-	__le64 tx_1519b_2047_frames;
-	__le64 tx_2048b_4095b_frames;
-	__le64 tx_4096b_9216b_frames;
-	__le64 tx_9217b_16383b_frames;
-	__le64 tx_good_frames;
-	__le64 tx_total_frames;
-	__le64 tx_ucast_frames;
-	__le64 tx_mcast_frames;
-	__le64 tx_bcast_frames;
-	__le64 tx_pause_frames;
-	__le64 tx_pfc_frames;
-	__le64 tx_jabber_frames;
-	__le64 tx_fcs_err_frames;
-	__le64 tx_control_frames;
-	__le64 tx_oversz_frames;
-	__le64 tx_single_dfrl_frames;
-	__le64 tx_multi_dfrl_frames;
-	__le64 tx_single_coll_frames;
-	__le64 tx_multi_coll_frames;
-	__le64 tx_late_coll_frames;
-	__le64 tx_excessive_coll_frames;
-	__le64 tx_frag_frames;
-	__le64 tx_err;
-	__le64 tx_tagged_frames;
-	__le64 tx_dbl_tagged_frames;
-	__le64 tx_runt_frames;
-	__le64 tx_fifo_underruns;
-	__le64 tx_pfc_ena_frames_pri0;
-	__le64 tx_pfc_ena_frames_pri1;
-	__le64 tx_pfc_ena_frames_pri2;
-	__le64 tx_pfc_ena_frames_pri3;
-	__le64 tx_pfc_ena_frames_pri4;
-	__le64 tx_pfc_ena_frames_pri5;
-	__le64 tx_pfc_ena_frames_pri6;
-	__le64 tx_pfc_ena_frames_pri7;
-	__le64 tx_eee_lpi_events;
-	__le64 tx_eee_lpi_duration;
-	__le64 tx_llfc_logical_msgs;
-	__le64 tx_hcfc_msgs;
-	__le64 tx_total_collisions;
-	__le64 tx_bytes;
-	__le64 tx_xthol_frames;
-	__le64 tx_stat_discard;
-	__le64 tx_stat_error;
-};
-
-/* Port Rx Statistics Formats (528 bytes) */
-struct rx_port_stats {
-	__le64 rx_64b_frames;
-	__le64 rx_65b_127b_frames;
-	__le64 rx_128b_255b_frames;
-	__le64 rx_256b_511b_frames;
-	__le64 rx_512b_1023b_frames;
-	__le64 rx_1024b_1518_frames;
-	__le64 rx_good_vlan_frames;
-	__le64 rx_1519b_2047b_frames;
-	__le64 rx_2048b_4095b_frames;
-	__le64 rx_4096b_9216b_frames;
-	__le64 rx_9217b_16383b_frames;
-	__le64 rx_total_frames;
-	__le64 rx_ucast_frames;
-	__le64 rx_mcast_frames;
-	__le64 rx_bcast_frames;
-	__le64 rx_fcs_err_frames;
-	__le64 rx_ctrl_frames;
-	__le64 rx_pause_frames;
-	__le64 rx_pfc_frames;
-	__le64 rx_unsupported_opcode_frames;
-	__le64 rx_unsupported_da_pausepfc_frames;
-	__le64 rx_wrong_sa_frames;
-	__le64 rx_align_err_frames;
-	__le64 rx_oor_len_frames;
-	__le64 rx_code_err_frames;
-	__le64 rx_false_carrier_frames;
-	__le64 rx_ovrsz_frames;
-	__le64 rx_jbr_frames;
-	__le64 rx_mtu_err_frames;
-	__le64 rx_match_crc_frames;
-	__le64 rx_promiscuous_frames;
-	__le64 rx_tagged_frames;
-	__le64 rx_double_tagged_frames;
-	__le64 rx_trunc_frames;
-	__le64 rx_good_frames;
-	__le64 rx_pfc_xon2xoff_frames_pri0;
-	__le64 rx_pfc_xon2xoff_frames_pri1;
-	__le64 rx_pfc_xon2xoff_frames_pri2;
-	__le64 rx_pfc_xon2xoff_frames_pri3;
-	__le64 rx_pfc_xon2xoff_frames_pri4;
-	__le64 rx_pfc_xon2xoff_frames_pri5;
-	__le64 rx_pfc_xon2xoff_frames_pri6;
-	__le64 rx_pfc_xon2xoff_frames_pri7;
-	__le64 rx_pfc_ena_frames_pri0;
-	__le64 rx_pfc_ena_frames_pri1;
-	__le64 rx_pfc_ena_frames_pri2;
-	__le64 rx_pfc_ena_frames_pri3;
-	__le64 rx_pfc_ena_frames_pri4;
-	__le64 rx_pfc_ena_frames_pri5;
-	__le64 rx_pfc_ena_frames_pri6;
-	__le64 rx_pfc_ena_frames_pri7;
-	__le64 rx_sch_crc_err_frames;
-	__le64 rx_undrsz_frames;
-	__le64 rx_frag_frames;
-	__le64 rx_eee_lpi_events;
-	__le64 rx_eee_lpi_duration;
-	__le64 rx_llfc_physical_msgs;
-	__le64 rx_llfc_logical_msgs;
-	__le64 rx_llfc_msgs_with_crc_err;
-	__le64 rx_hcfc_msgs;
-	__le64 rx_hcfc_msgs_with_crc_err;
-	__le64 rx_bytes;
-	__le64 rx_runt_bytes;
-	__le64 rx_runt_frames;
-	__le64 rx_stat_discard;
-	__le64 rx_stat_err;
-};
-
-/* Periodic Statistics Context DMA to host (160 bytes) */
-struct ctx_hw_stats {
-	__le64 rx_ucast_pkts;
-	__le64 rx_mcast_pkts;
-	__le64 rx_bcast_pkts;
-	__le64 rx_discard_pkts;
-	__le64 rx_drop_pkts;
-	__le64 rx_ucast_bytes;
-	__le64 rx_mcast_bytes;
-	__le64 rx_bcast_bytes;
-	__le64 tx_ucast_pkts;
-	__le64 tx_mcast_pkts;
-	__le64 tx_bcast_pkts;
-	__le64 tx_discard_pkts;
-	__le64 tx_drop_pkts;
-	__le64 tx_ucast_bytes;
-	__le64 tx_mcast_bytes;
-	__le64 tx_bcast_bytes;
-	__le64 tpa_pkts;
-	__le64 tpa_bytes;
-	__le64 tpa_events;
-	__le64 tpa_aborts;
-};
-
-/* Structure data header (16 bytes) */
-struct hwrm_struct_hdr {
-	__le16 struct_id;
-	#define STRUCT_HDR_STRUCT_ID_LLDP_CFG			   0x41bUL
-	#define STRUCT_HDR_STRUCT_ID_DCBX_ETS			   0x41dUL
-	#define STRUCT_HDR_STRUCT_ID_DCBX_PFC			   0x41fUL
-	#define STRUCT_HDR_STRUCT_ID_DCBX_APP			   0x421UL
-	#define STRUCT_HDR_STRUCT_ID_DCBX_FEATURE_STATE	   0x422UL
-	#define STRUCT_HDR_STRUCT_ID_LLDP_GENERIC		   0x424UL
-	#define STRUCT_HDR_STRUCT_ID_LLDP_DEVICE		   0x426UL
-	#define STRUCT_HDR_STRUCT_ID_AFM_OPAQUE		   0x1UL
-	#define STRUCT_HDR_STRUCT_ID_PORT_DESCRIPTION		   0xaUL
-	__le16 len;
-	u8 version;
-	u8 count;
-	__le16 subtype;
-	__le16 next_offset;
-	#define STRUCT_HDR_NEXT_OFFSET_LAST			   0x0UL
-	__le16 unused_0[3];
-};
-
-/* DCBX Application configuration structure (1057) (8 bytes) */
-struct hwrm_struct_data_dcbx_app {
-	__be16 protocol_id;
-	u8 protocol_selector;
-	#define STRUCT_DATA_DCBX_APP_PROTOCOL_SELECTOR_ETHER_TYPE 0x1UL
-	#define STRUCT_DATA_DCBX_APP_PROTOCOL_SELECTOR_TCP_PORT   0x2UL
-	#define STRUCT_DATA_DCBX_APP_PROTOCOL_SELECTOR_UDP_PORT   0x3UL
-	#define STRUCT_DATA_DCBX_APP_PROTOCOL_SELECTOR_TCP_UDP_PORT 0x4UL
-	u8 priority;
-	u8 valid;
-	u8 unused_0[3];
-};
-
-#endif
diff --git a/drivers/net/ethernet/broadcom/bnxt/bnxt_hwmon.c b/drivers/net/ethernet/broadcom/bnxt/bnxt_hwmon.c
new file mode 100644
index 000000000000..de3427c6c6aa
--- /dev/null
+++ b/drivers/net/ethernet/broadcom/bnxt/bnxt_hwmon.c
@@ -0,0 +1,241 @@
+/* Broadcom NetXtreme-C/E network driver.
+ *
+ * Copyright (c) 2023 Broadcom Limited
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation.
+ */
+
+#include <linux/dev_printk.h>
+#include <linux/errno.h>
+#include <linux/hwmon.h>
+#include <linux/hwmon-sysfs.h>
+#include <linux/pci.h>
+#include <linux/bnxt/hsi.h>
+
+#include "bnxt.h"
+#include "bnxt_hwrm.h"
+#include "bnxt_hwmon.h"
+
+void bnxt_hwmon_notify_event(struct bnxt *bp)
+{
+	u32 attr;
+
+	if (!bp->hwmon_dev)
+		return;
+
+	switch (bp->thermal_threshold_type) {
+	case ASYNC_EVENT_CMPL_ERROR_REPORT_THERMAL_EVENT_DATA1_THRESHOLD_TYPE_WARN:
+		attr = hwmon_temp_max_alarm;
+		break;
+	case ASYNC_EVENT_CMPL_ERROR_REPORT_THERMAL_EVENT_DATA1_THRESHOLD_TYPE_CRITICAL:
+		attr = hwmon_temp_crit_alarm;
+		break;
+	case ASYNC_EVENT_CMPL_ERROR_REPORT_THERMAL_EVENT_DATA1_THRESHOLD_TYPE_FATAL:
+	case ASYNC_EVENT_CMPL_ERROR_REPORT_THERMAL_EVENT_DATA1_THRESHOLD_TYPE_SHUTDOWN:
+		attr = hwmon_temp_emergency_alarm;
+		break;
+	default:
+		return;
+	}
+
+	hwmon_notify_event(&bp->pdev->dev, hwmon_temp, attr, 0);
+}
+
+static int bnxt_hwrm_temp_query(struct bnxt *bp, u8 *temp)
+{
+	struct hwrm_temp_monitor_query_output *resp;
+	struct hwrm_temp_monitor_query_input *req;
+	int rc;
+
+	rc = hwrm_req_init(bp, req, HWRM_TEMP_MONITOR_QUERY);
+	if (rc)
+		return rc;
+	resp = hwrm_req_hold(bp, req);
+	rc = hwrm_req_send_silent(bp, req);
+	if (rc)
+		goto drop_req;
+
+	if (temp) {
+		*temp = resp->temp;
+	} else if (resp->flags &
+		   TEMP_MONITOR_QUERY_RESP_FLAGS_THRESHOLD_VALUES_AVAILABLE) {
+		bp->fw_cap |= BNXT_FW_CAP_THRESHOLD_TEMP_SUPPORTED;
+		bp->warn_thresh_temp = resp->warn_threshold;
+		bp->crit_thresh_temp = resp->critical_threshold;
+		bp->fatal_thresh_temp = resp->fatal_threshold;
+		bp->shutdown_thresh_temp = resp->shutdown_threshold;
+	}
+drop_req:
+	hwrm_req_drop(bp, req);
+	return rc;
+}
+
+static umode_t bnxt_hwmon_is_visible(const void *_data, enum hwmon_sensor_types type,
+				     u32 attr, int channel)
+{
+	const struct bnxt *bp = _data;
+
+	if (type != hwmon_temp)
+		return 0;
+
+	switch (attr) {
+	case hwmon_temp_input:
+		return 0444;
+	case hwmon_temp_max:
+	case hwmon_temp_crit:
+	case hwmon_temp_emergency:
+	case hwmon_temp_max_alarm:
+	case hwmon_temp_crit_alarm:
+	case hwmon_temp_emergency_alarm:
+		if (!(bp->fw_cap & BNXT_FW_CAP_THRESHOLD_TEMP_SUPPORTED))
+			return 0;
+		return 0444;
+	default:
+		return 0;
+	}
+}
+
+static int bnxt_hwmon_read(struct device *dev, enum hwmon_sensor_types type, u32 attr,
+			   int channel, long *val)
+{
+	struct bnxt *bp = dev_get_drvdata(dev);
+	u8 temp = 0;
+	int rc;
+
+	switch (attr) {
+	case hwmon_temp_input:
+		rc = bnxt_hwrm_temp_query(bp, &temp);
+		if (!rc)
+			*val = temp * 1000;
+		return rc;
+	case hwmon_temp_max:
+		*val = bp->warn_thresh_temp * 1000;
+		return 0;
+	case hwmon_temp_crit:
+		*val = bp->crit_thresh_temp * 1000;
+		return 0;
+	case hwmon_temp_emergency:
+		*val = bp->fatal_thresh_temp * 1000;
+		return 0;
+	case hwmon_temp_max_alarm:
+		rc = bnxt_hwrm_temp_query(bp, &temp);
+		if (!rc)
+			*val = temp >= bp->warn_thresh_temp;
+		return rc;
+	case hwmon_temp_crit_alarm:
+		rc = bnxt_hwrm_temp_query(bp, &temp);
+		if (!rc)
+			*val = temp >= bp->crit_thresh_temp;
+		return rc;
+	case hwmon_temp_emergency_alarm:
+		rc = bnxt_hwrm_temp_query(bp, &temp);
+		if (!rc)
+			*val = temp >= bp->fatal_thresh_temp;
+		return rc;
+	default:
+		return -EOPNOTSUPP;
+	}
+}
+
+static const struct hwmon_channel_info *bnxt_hwmon_info[] = {
+	HWMON_CHANNEL_INFO(temp, HWMON_T_INPUT | HWMON_T_MAX | HWMON_T_CRIT |
+			   HWMON_T_EMERGENCY | HWMON_T_MAX_ALARM |
+			   HWMON_T_CRIT_ALARM | HWMON_T_EMERGENCY_ALARM),
+	NULL
+};
+
+static const struct hwmon_ops bnxt_hwmon_ops = {
+	.is_visible     = bnxt_hwmon_is_visible,
+	.read           = bnxt_hwmon_read,
+};
+
+static const struct hwmon_chip_info bnxt_hwmon_chip_info = {
+	.ops    = &bnxt_hwmon_ops,
+	.info   = bnxt_hwmon_info,
+};
+
+static ssize_t temp1_shutdown_show(struct device *dev,
+				   struct device_attribute *attr, char *buf)
+{
+	struct bnxt *bp = dev_get_drvdata(dev);
+
+	return sysfs_emit(buf, "%u\n", bp->shutdown_thresh_temp * 1000);
+}
+
+static ssize_t temp1_shutdown_alarm_show(struct device *dev,
+					 struct device_attribute *attr, char *buf)
+{
+	struct bnxt *bp = dev_get_drvdata(dev);
+	u8 temp;
+	int rc;
+
+	rc = bnxt_hwrm_temp_query(bp, &temp);
+	if (rc)
+		return -EIO;
+
+	return sysfs_emit(buf, "%u\n", temp >= bp->shutdown_thresh_temp);
+}
+
+static DEVICE_ATTR_RO(temp1_shutdown);
+static DEVICE_ATTR_RO(temp1_shutdown_alarm);
+
+static struct attribute *bnxt_temp_extra_attrs[] = {
+	&dev_attr_temp1_shutdown.attr,
+	&dev_attr_temp1_shutdown_alarm.attr,
+	NULL,
+};
+
+static umode_t bnxt_temp_extra_attrs_visible(struct kobject *kobj,
+					     struct attribute *attr, int index)
+{
+	struct device *dev = kobj_to_dev(kobj);
+	struct bnxt *bp = dev_get_drvdata(dev);
+
+	/* Shutdown temperature setting in NVM is optional */
+	if (!(bp->fw_cap & BNXT_FW_CAP_THRESHOLD_TEMP_SUPPORTED) ||
+	    !bp->shutdown_thresh_temp)
+		return 0;
+
+	return attr->mode;
+}
+
+static const struct attribute_group bnxt_temp_extra_group = {
+	.attrs		= bnxt_temp_extra_attrs,
+	.is_visible	= bnxt_temp_extra_attrs_visible,
+};
+__ATTRIBUTE_GROUPS(bnxt_temp_extra);
+
+void bnxt_hwmon_uninit(struct bnxt *bp)
+{
+	if (bp->hwmon_dev) {
+		hwmon_device_unregister(bp->hwmon_dev);
+		bp->hwmon_dev = NULL;
+	}
+}
+
+void bnxt_hwmon_init(struct bnxt *bp)
+{
+	struct pci_dev *pdev = bp->pdev;
+	int rc;
+
+	/* temp1_xxx is only sensor, ensure not registered if it will fail */
+	rc = bnxt_hwrm_temp_query(bp, NULL);
+	if (rc == -EACCES || rc == -EOPNOTSUPP) {
+		bnxt_hwmon_uninit(bp);
+		return;
+	}
+
+	if (bp->hwmon_dev)
+		return;
+
+	bp->hwmon_dev = hwmon_device_register_with_info(&pdev->dev,
+							DRV_MODULE_NAME, bp,
+							&bnxt_hwmon_chip_info,
+							bnxt_temp_extra_groups);
+	if (IS_ERR(bp->hwmon_dev)) {
+		bp->hwmon_dev = NULL;
+		dev_warn(&pdev->dev, "Cannot register hwmon device\n");
+	}
+}
diff --git a/drivers/net/ethernet/broadcom/bnxt/bnxt_hwmon.h b/drivers/net/ethernet/broadcom/bnxt/bnxt_hwmon.h
new file mode 100644
index 000000000000..de54a562e06a
--- /dev/null
+++ b/drivers/net/ethernet/broadcom/bnxt/bnxt_hwmon.h
@@ -0,0 +1,30 @@
+/* Broadcom NetXtreme-C/E network driver.
+ *
+ * Copyright (c) 2023 Broadcom Limited
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation.
+ */
+
+#ifndef BNXT_HWMON_H
+#define BNXT_HWMON_H
+
+#ifdef CONFIG_BNXT_HWMON
+void bnxt_hwmon_notify_event(struct bnxt *bp);
+void bnxt_hwmon_uninit(struct bnxt *bp);
+void bnxt_hwmon_init(struct bnxt *bp);
+#else
+static inline void bnxt_hwmon_notify_event(struct bnxt *bp)
+{
+}
+
+static inline void bnxt_hwmon_uninit(struct bnxt *bp)
+{
+}
+
+static inline void bnxt_hwmon_init(struct bnxt *bp)
+{
+}
+#endif
+#endif
diff --git a/drivers/net/ethernet/broadcom/bnxt/bnxt_hwrm.c b/drivers/net/ethernet/broadcom/bnxt/bnxt_hwrm.c
new file mode 100644
index 000000000000..5ce190f50120
--- /dev/null
+++ b/drivers/net/ethernet/broadcom/bnxt/bnxt_hwrm.c
@@ -0,0 +1,819 @@
+/* Broadcom NetXtreme-C/E network driver.
+ *
+ * Copyright (c) 2020 Broadcom Limited
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation.
+ */
+
+#include <asm/byteorder.h>
+#include <linux/dma-mapping.h>
+#include <linux/dmapool.h>
+#include <linux/errno.h>
+#include <linux/ethtool.h>
+#include <linux/if_ether.h>
+#include <linux/io.h>
+#include <linux/irq.h>
+#include <linux/kernel.h>
+#include <linux/list.h>
+#include <linux/netdevice.h>
+#include <linux/pci.h>
+#include <linux/skbuff.h>
+#include <linux/bnxt/hsi.h>
+
+#include "bnxt.h"
+#include "bnxt_hwrm.h"
+
+static u64 hwrm_calc_sentinel(struct bnxt_hwrm_ctx *ctx, u16 req_type)
+{
+	return (((uintptr_t)ctx) + req_type) ^ BNXT_HWRM_SENTINEL;
+}
+
+/**
+ * __hwrm_req_init() - Initialize an HWRM request.
+ * @bp: The driver context.
+ * @req: A pointer to the request pointer to initialize.
+ * @req_type: The request type. This will be converted to the little endian
+ *	before being written to the req_type field of the returned request.
+ * @req_len: The length of the request to be allocated.
+ *
+ * Allocate DMA resources and initialize a new HWRM request object of the
+ * given type. The response address field in the request is configured with
+ * the DMA bus address that has been mapped for the response and the passed
+ * request is pointed to kernel virtual memory mapped for the request (such
+ * that short_input indirection can be accomplished without copying). The
+ * request’s target and completion ring are initialized to default values and
+ * can be overridden by writing to the returned request object directly.
+ *
+ * The initialized request can be further customized by writing to its fields
+ * directly, taking care to covert such fields to little endian. The request
+ * object will be consumed (and all its associated resources release) upon
+ * passing it to hwrm_req_send() unless ownership of the request has been
+ * claimed by the caller via a call to hwrm_req_hold(). If the request is not
+ * consumed, either because it is never sent or because ownership has been
+ * claimed, then it must be released by a call to hwrm_req_drop().
+ *
+ * Return: zero on success, negative error code otherwise:
+ *	E2BIG: the type of request pointer is too large to fit.
+ *	ENOMEM: an allocation failure occurred.
+ */
+int __hwrm_req_init(struct bnxt *bp, void **req, u16 req_type, u32 req_len)
+{
+	struct bnxt_hwrm_ctx *ctx;
+	dma_addr_t dma_handle;
+	u8 *req_addr;
+
+	if (req_len > BNXT_HWRM_CTX_OFFSET)
+		return -E2BIG;
+
+	req_addr = dma_pool_alloc(bp->hwrm_dma_pool, GFP_KERNEL | __GFP_ZERO,
+				  &dma_handle);
+	if (!req_addr)
+		return -ENOMEM;
+
+	ctx = (struct bnxt_hwrm_ctx *)(req_addr + BNXT_HWRM_CTX_OFFSET);
+	/* safety first, sentinel used to check for invalid requests */
+	ctx->sentinel = hwrm_calc_sentinel(ctx, req_type);
+	ctx->req_len = req_len;
+	ctx->req = (struct input *)req_addr;
+	ctx->resp = (struct output *)(req_addr + BNXT_HWRM_RESP_OFFSET);
+	ctx->dma_handle = dma_handle;
+	ctx->flags = 0; /* __GFP_ZERO, but be explicit regarding ownership */
+	ctx->timeout = bp->hwrm_cmd_timeout ?: DFLT_HWRM_CMD_TIMEOUT;
+	ctx->allocated = BNXT_HWRM_DMA_SIZE - BNXT_HWRM_CTX_OFFSET;
+	ctx->gfp = GFP_KERNEL;
+	ctx->slice_addr = NULL;
+
+	/* initialize common request fields */
+	ctx->req->req_type = cpu_to_le16(req_type);
+	ctx->req->resp_addr = cpu_to_le64(dma_handle + BNXT_HWRM_RESP_OFFSET);
+	ctx->req->cmpl_ring = cpu_to_le16(BNXT_HWRM_NO_CMPL_RING);
+	ctx->req->target_id = cpu_to_le16(BNXT_HWRM_TARGET);
+	*req = ctx->req;
+
+	return 0;
+}
+
+static struct bnxt_hwrm_ctx *__hwrm_ctx(struct bnxt *bp, u8 *req_addr)
+{
+	void *ctx_addr = req_addr + BNXT_HWRM_CTX_OFFSET;
+	struct input *req = (struct input *)req_addr;
+	struct bnxt_hwrm_ctx *ctx = ctx_addr;
+	u64 sentinel;
+
+	if (!req) {
+		/* can only be due to software bug, be loud */
+		netdev_err(bp->dev, "null HWRM request");
+		dump_stack();
+		return NULL;
+	}
+
+	/* HWRM API has no type safety, verify sentinel to validate address */
+	sentinel = hwrm_calc_sentinel(ctx, le16_to_cpu(req->req_type));
+	if (ctx->sentinel != sentinel) {
+		/* can only be due to software bug, be loud */
+		netdev_err(bp->dev, "HWRM sentinel mismatch, req_type = %u\n",
+			   (u32)le16_to_cpu(req->req_type));
+		dump_stack();
+		return NULL;
+	}
+
+	return ctx;
+}
+
+/**
+ * hwrm_req_timeout() - Set the completion timeout for the request.
+ * @bp: The driver context.
+ * @req: The request to set the timeout.
+ * @timeout: The timeout in milliseconds.
+ *
+ * Set the timeout associated with the request for subsequent calls to
+ * hwrm_req_send(). Some requests are long running and require a different
+ * timeout than the default.
+ */
+void hwrm_req_timeout(struct bnxt *bp, void *req, unsigned int timeout)
+{
+	struct bnxt_hwrm_ctx *ctx = __hwrm_ctx(bp, req);
+
+	if (ctx)
+		ctx->timeout = timeout;
+}
+
+/**
+ * hwrm_req_alloc_flags() - Sets GFP allocation flags for slices.
+ * @bp: The driver context.
+ * @req: The request for which calls to hwrm_req_dma_slice() will have altered
+ *	allocation flags.
+ * @gfp: A bitmask of GFP flags. These flags are passed to dma_alloc_coherent()
+ *	whenever it is used to allocate backing memory for slices. Note that
+ *	calls to hwrm_req_dma_slice() will not always result in new allocations,
+ *	however, memory suballocated from the request buffer is already
+ *	__GFP_ZERO.
+ *
+ * Sets the GFP allocation flags associated with the request for subsequent
+ * calls to hwrm_req_dma_slice(). This can be useful for specifying __GFP_ZERO
+ * for slice allocations.
+ */
+void hwrm_req_alloc_flags(struct bnxt *bp, void *req, gfp_t gfp)
+{
+	struct bnxt_hwrm_ctx *ctx = __hwrm_ctx(bp, req);
+
+	if (ctx)
+		ctx->gfp = gfp;
+}
+
+/**
+ * hwrm_req_replace() - Replace request data.
+ * @bp: The driver context.
+ * @req: The request to modify. A call to hwrm_req_replace() is conceptually
+ *	an assignment of new_req to req. Subsequent calls to HWRM API functions,
+ *	such as hwrm_req_send(), should thus use req and not new_req (in fact,
+ *	calls to HWRM API functions will fail if non-managed request objects
+ *	are passed).
+ * @len: The length of new_req.
+ * @new_req: The pre-built request to copy or reference.
+ *
+ * Replaces the request data in req with that of new_req. This is useful in
+ * scenarios where a request object has already been constructed by a third
+ * party prior to creating a resource managed request using hwrm_req_init().
+ * Depending on the length, hwrm_req_replace() will either copy the new
+ * request data into the DMA memory allocated for req, or it will simply
+ * reference the new request and use it in lieu of req during subsequent
+ * calls to hwrm_req_send(). The resource management is associated with
+ * req and is independent of and does not apply to new_req. The caller must
+ * ensure that the lifetime of new_req is least as long as req. Any slices
+ * that may have been associated with the original request are released.
+ *
+ * Return: zero on success, negative error code otherwise:
+ *     E2BIG: Request is too large.
+ *     EINVAL: Invalid request to modify.
+ */
+int hwrm_req_replace(struct bnxt *bp, void *req, void *new_req, u32 len)
+{
+	struct bnxt_hwrm_ctx *ctx = __hwrm_ctx(bp, req);
+	struct input *internal_req = req;
+	u16 req_type;
+
+	if (!ctx)
+		return -EINVAL;
+
+	if (len > BNXT_HWRM_CTX_OFFSET)
+		return -E2BIG;
+
+	/* free any existing slices */
+	ctx->allocated = BNXT_HWRM_DMA_SIZE - BNXT_HWRM_CTX_OFFSET;
+	if (ctx->slice_addr) {
+		dma_free_coherent(&bp->pdev->dev, ctx->slice_size,
+				  ctx->slice_addr, ctx->slice_handle);
+		ctx->slice_addr = NULL;
+	}
+	ctx->gfp = GFP_KERNEL;
+
+	if ((bp->fw_cap & BNXT_FW_CAP_SHORT_CMD) || len > BNXT_HWRM_MAX_REQ_LEN) {
+		memcpy(internal_req, new_req, len);
+	} else {
+		internal_req->req_type = ((struct input *)new_req)->req_type;
+		ctx->req = new_req;
+	}
+
+	ctx->req_len = len;
+	ctx->req->resp_addr = cpu_to_le64(ctx->dma_handle +
+					  BNXT_HWRM_RESP_OFFSET);
+
+	/* update sentinel for potentially new request type */
+	req_type = le16_to_cpu(internal_req->req_type);
+	ctx->sentinel = hwrm_calc_sentinel(ctx, req_type);
+
+	return 0;
+}
+
+/**
+ * hwrm_req_flags() - Set non internal flags of the ctx
+ * @bp: The driver context.
+ * @req: The request containing the HWRM command
+ * @flags: ctx flags that don't have BNXT_HWRM_INTERNAL_FLAG set
+ *
+ * ctx flags can be used by the callers to instruct how the subsequent
+ * hwrm_req_send() should behave. Example: callers can use hwrm_req_flags
+ * with BNXT_HWRM_CTX_SILENT to omit kernel prints of errors of hwrm_req_send()
+ * or with BNXT_HWRM_FULL_WAIT enforce hwrm_req_send() to wait for full timeout
+ * even if FW is not responding.
+ * This generic function can be used to set any flag that is not an internal flag
+ * of the HWRM module.
+ */
+void hwrm_req_flags(struct bnxt *bp, void *req, enum bnxt_hwrm_ctx_flags flags)
+{
+	struct bnxt_hwrm_ctx *ctx = __hwrm_ctx(bp, req);
+
+	if (ctx)
+		ctx->flags |= (flags & HWRM_API_FLAGS);
+}
+
+/**
+ * hwrm_req_hold() - Claim ownership of the request's resources.
+ * @bp: The driver context.
+ * @req: A pointer to the request to own. The request will no longer be
+ *	consumed by calls to hwrm_req_send().
+ *
+ * Take ownership of the request. Ownership places responsibility on the
+ * caller to free the resources associated with the request via a call to
+ * hwrm_req_drop(). The caller taking ownership implies that a subsequent
+ * call to hwrm_req_send() will not consume the request (ie. sending will
+ * not free the associated resources if the request is owned by the caller).
+ * Taking ownership returns a reference to the response. Retaining and
+ * accessing the response data is the most common reason to take ownership
+ * of the request. Ownership can also be acquired in order to reuse the same
+ * request object across multiple invocations of hwrm_req_send().
+ *
+ * Return: A pointer to the response object.
+ *
+ * The resources associated with the response will remain available to the
+ * caller until ownership of the request is relinquished via a call to
+ * hwrm_req_drop(). It is not possible for hwrm_req_hold() to return NULL if
+ * a valid request is provided. A returned NULL value would imply a driver
+ * bug and the implementation will complain loudly in the logs to aid in
+ * detection. It should not be necessary to check the result for NULL.
+ */
+void *hwrm_req_hold(struct bnxt *bp, void *req)
+{
+	struct bnxt_hwrm_ctx *ctx = __hwrm_ctx(bp, req);
+	struct input *input = (struct input *)req;
+
+	if (!ctx)
+		return NULL;
+
+	if (ctx->flags & BNXT_HWRM_INTERNAL_CTX_OWNED) {
+		/* can only be due to software bug, be loud */
+		netdev_err(bp->dev, "HWRM context already owned, req_type = %u\n",
+			   (u32)le16_to_cpu(input->req_type));
+		dump_stack();
+		return NULL;
+	}
+
+	ctx->flags |= BNXT_HWRM_INTERNAL_CTX_OWNED;
+	return ((u8 *)req) + BNXT_HWRM_RESP_OFFSET;
+}
+
+static void __hwrm_ctx_drop(struct bnxt *bp, struct bnxt_hwrm_ctx *ctx)
+{
+	void *addr = ((u8 *)ctx) - BNXT_HWRM_CTX_OFFSET;
+	dma_addr_t dma_handle = ctx->dma_handle; /* save before invalidate */
+
+	/* unmap any auxiliary DMA slice */
+	if (ctx->slice_addr)
+		dma_free_coherent(&bp->pdev->dev, ctx->slice_size,
+				  ctx->slice_addr, ctx->slice_handle);
+
+	/* invalidate, ensure ownership, sentinel and dma_handle are cleared */
+	memset(ctx, 0, sizeof(struct bnxt_hwrm_ctx));
+
+	/* return the buffer to the DMA pool */
+	if (dma_handle)
+		dma_pool_free(bp->hwrm_dma_pool, addr, dma_handle);
+}
+
+/**
+ * hwrm_req_drop() - Release all resources associated with the request.
+ * @bp: The driver context.
+ * @req: The request to consume, releasing the associated resources. The
+ *	request object, any slices, and its associated response are no
+ *	longer valid.
+ *
+ * It is legal to call hwrm_req_drop() on an unowned request, provided it
+ * has not already been consumed by hwrm_req_send() (for example, to release
+ * an aborted request). A given request should not be dropped more than once,
+ * nor should it be dropped after having been consumed by hwrm_req_send(). To
+ * do so is an error (the context will not be found and a stack trace will be
+ * rendered in the kernel log).
+ */
+void hwrm_req_drop(struct bnxt *bp, void *req)
+{
+	struct bnxt_hwrm_ctx *ctx = __hwrm_ctx(bp, req);
+
+	if (ctx)
+		__hwrm_ctx_drop(bp, ctx);
+}
+
+static int __hwrm_to_stderr(u32 hwrm_err)
+{
+	switch (hwrm_err) {
+	case HWRM_ERR_CODE_SUCCESS:
+		return 0;
+	case HWRM_ERR_CODE_RESOURCE_LOCKED:
+		return -EROFS;
+	case HWRM_ERR_CODE_RESOURCE_ACCESS_DENIED:
+		return -EACCES;
+	case HWRM_ERR_CODE_RESOURCE_ALLOC_ERROR:
+		return -ENOSPC;
+	case HWRM_ERR_CODE_INVALID_PARAMS:
+	case HWRM_ERR_CODE_INVALID_FLAGS:
+	case HWRM_ERR_CODE_INVALID_ENABLES:
+	case HWRM_ERR_CODE_UNSUPPORTED_TLV:
+	case HWRM_ERR_CODE_UNSUPPORTED_OPTION_ERR:
+		return -EINVAL;
+	case HWRM_ERR_CODE_NO_BUFFER:
+		return -ENOMEM;
+	case HWRM_ERR_CODE_HOT_RESET_PROGRESS:
+	case HWRM_ERR_CODE_BUSY:
+		return -EAGAIN;
+	case HWRM_ERR_CODE_CMD_NOT_SUPPORTED:
+		return -EOPNOTSUPP;
+	case HWRM_ERR_CODE_PF_UNAVAILABLE:
+		return -ENODEV;
+	default:
+		return -EIO;
+	}
+}
+
+static struct bnxt_hwrm_wait_token *
+__hwrm_acquire_token(struct bnxt *bp, enum bnxt_hwrm_chnl dst)
+{
+	struct bnxt_hwrm_wait_token *token;
+
+	token = kzalloc(sizeof(*token), GFP_KERNEL);
+	if (!token)
+		return NULL;
+
+	mutex_lock(&bp->hwrm_cmd_lock);
+
+	token->dst = dst;
+	token->state = BNXT_HWRM_PENDING;
+	if (dst == BNXT_HWRM_CHNL_CHIMP) {
+		token->seq_id = bp->hwrm_cmd_seq++;
+		hlist_add_head_rcu(&token->node, &bp->hwrm_pending_list);
+	} else {
+		token->seq_id = bp->hwrm_cmd_kong_seq++;
+	}
+
+	return token;
+}
+
+static void
+__hwrm_release_token(struct bnxt *bp, struct bnxt_hwrm_wait_token *token)
+{
+	if (token->dst == BNXT_HWRM_CHNL_CHIMP) {
+		hlist_del_rcu(&token->node);
+		kfree_rcu(token, rcu);
+	} else {
+		kfree(token);
+	}
+	mutex_unlock(&bp->hwrm_cmd_lock);
+}
+
+void
+hwrm_update_token(struct bnxt *bp, u16 seq_id, enum bnxt_hwrm_wait_state state)
+{
+	struct bnxt_hwrm_wait_token *token;
+
+	rcu_read_lock();
+	hlist_for_each_entry_rcu(token, &bp->hwrm_pending_list, node) {
+		if (token->seq_id == seq_id) {
+			WRITE_ONCE(token->state, state);
+			rcu_read_unlock();
+			return;
+		}
+	}
+	rcu_read_unlock();
+	netdev_err(bp->dev, "Invalid hwrm seq id %d\n", seq_id);
+}
+
+static void hwrm_req_dbg(struct bnxt *bp, struct input *req)
+{
+	u32 ring = le16_to_cpu(req->cmpl_ring);
+	u32 type = le16_to_cpu(req->req_type);
+	u32 tgt = le16_to_cpu(req->target_id);
+	u32 seq = le16_to_cpu(req->seq_id);
+	char opt[32] = "\n";
+
+	if (unlikely(ring != (u16)BNXT_HWRM_NO_CMPL_RING))
+		snprintf(opt, 16, " ring %d\n", ring);
+
+	if (unlikely(tgt != BNXT_HWRM_TARGET))
+		snprintf(opt + strlen(opt) - 1, 16, " tgt 0x%x\n", tgt);
+
+	netdev_dbg(bp->dev, "sent hwrm req_type 0x%x seq id 0x%x%s",
+		   type, seq, opt);
+}
+
+#define hwrm_err(bp, ctx, fmt, ...)				       \
+	do {							       \
+		if ((ctx)->flags & BNXT_HWRM_CTX_SILENT)	       \
+			netdev_dbg((bp)->dev, fmt, __VA_ARGS__);       \
+		else						       \
+			netdev_err((bp)->dev, fmt, __VA_ARGS__);       \
+	} while (0)
+
+static bool hwrm_wait_must_abort(struct bnxt *bp, u32 req_type, u32 *fw_status)
+{
+	if (req_type == HWRM_VER_GET)
+		return false;
+
+	if (!bp->fw_health || !bp->fw_health->status_reliable)
+		return false;
+
+	*fw_status = bnxt_fw_health_readl(bp, BNXT_FW_HEALTH_REG);
+	return *fw_status && !BNXT_FW_IS_HEALTHY(*fw_status);
+}
+
+static int __hwrm_send(struct bnxt *bp, struct bnxt_hwrm_ctx *ctx)
+{
+	u32 doorbell_offset = BNXT_GRCPF_REG_CHIMP_COMM_TRIGGER;
+	enum bnxt_hwrm_chnl dst = BNXT_HWRM_CHNL_CHIMP;
+	u32 bar_offset = BNXT_GRCPF_REG_CHIMP_COMM;
+	struct bnxt_hwrm_wait_token *token = NULL;
+	struct hwrm_short_input short_input = {0};
+	u16 max_req_len = BNXT_HWRM_MAX_REQ_LEN;
+	unsigned int i, timeout, tmo_count;
+	u32 *data = (u32 *)ctx->req;
+	u32 msg_len = ctx->req_len;
+	u32 req_type, sts;
+	int rc = -EBUSY;
+	u16 len = 0;
+	u8 *valid;
+
+	if (ctx->flags & BNXT_HWRM_INTERNAL_RESP_DIRTY)
+		memset(ctx->resp, 0, PAGE_SIZE);
+
+	req_type = le16_to_cpu(ctx->req->req_type);
+	if (BNXT_NO_FW_ACCESS(bp) &&
+	    (req_type != HWRM_FUNC_RESET && req_type != HWRM_VER_GET)) {
+		netdev_dbg(bp->dev, "hwrm req_type 0x%x skipped, FW channel down\n",
+			   req_type);
+		goto exit;
+	}
+
+	if (msg_len > BNXT_HWRM_MAX_REQ_LEN &&
+	    msg_len > bp->hwrm_max_ext_req_len) {
+		netdev_warn(bp->dev, "oversized hwrm request, req_type 0x%x",
+			    req_type);
+		rc = -E2BIG;
+		goto exit;
+	}
+
+	if (bnxt_kong_hwrm_message(bp, ctx->req)) {
+		dst = BNXT_HWRM_CHNL_KONG;
+		bar_offset = BNXT_GRCPF_REG_KONG_COMM;
+		doorbell_offset = BNXT_GRCPF_REG_KONG_COMM_TRIGGER;
+		if (le16_to_cpu(ctx->req->cmpl_ring) != INVALID_HW_RING_ID) {
+			netdev_err(bp->dev, "Ring completions not supported for KONG commands, req_type = %d\n",
+				   req_type);
+			rc = -EINVAL;
+			goto exit;
+		}
+	}
+
+	token = __hwrm_acquire_token(bp, dst);
+	if (!token) {
+		rc = -ENOMEM;
+		goto exit;
+	}
+	ctx->req->seq_id = cpu_to_le16(token->seq_id);
+
+	if ((bp->fw_cap & BNXT_FW_CAP_SHORT_CMD) ||
+	    msg_len > BNXT_HWRM_MAX_REQ_LEN) {
+		short_input.req_type = ctx->req->req_type;
+		short_input.signature =
+				cpu_to_le16(SHORT_REQ_SIGNATURE_SHORT_CMD);
+		short_input.size = cpu_to_le16(msg_len);
+		short_input.req_addr = cpu_to_le64(ctx->dma_handle);
+
+		data = (u32 *)&short_input;
+		msg_len = sizeof(short_input);
+
+		max_req_len = BNXT_HWRM_SHORT_REQ_LEN;
+	}
+
+	/* Ensure any associated DMA buffers are written before doorbell */
+	wmb();
+
+	/* Write request msg to hwrm channel */
+	__iowrite32_copy(bp->bar0 + bar_offset, data, msg_len / 4);
+
+	for (i = msg_len; i < max_req_len; i += 4)
+		writel(0, bp->bar0 + bar_offset + i);
+
+	/* Ring channel doorbell */
+	writel(1, bp->bar0 + doorbell_offset);
+
+	hwrm_req_dbg(bp, ctx->req);
+
+	if (!pci_is_enabled(bp->pdev)) {
+		rc = -ENODEV;
+		goto exit;
+	}
+
+	/* Limit timeout to an upper limit */
+	timeout = min(ctx->timeout, bp->hwrm_cmd_max_timeout ?: HWRM_CMD_MAX_TIMEOUT);
+	/* convert timeout to usec */
+	timeout *= 1000;
+
+	i = 0;
+	/* Short timeout for the first few iterations:
+	 * number of loops = number of loops for short timeout +
+	 * number of loops for standard timeout.
+	 */
+	tmo_count = HWRM_SHORT_TIMEOUT_COUNTER;
+	timeout = timeout - HWRM_SHORT_MIN_TIMEOUT * HWRM_SHORT_TIMEOUT_COUNTER;
+	tmo_count += DIV_ROUND_UP(timeout, HWRM_MIN_TIMEOUT);
+
+	if (le16_to_cpu(ctx->req->cmpl_ring) != INVALID_HW_RING_ID) {
+		/* Wait until hwrm response cmpl interrupt is processed */
+		while (READ_ONCE(token->state) < BNXT_HWRM_COMPLETE &&
+		       i++ < tmo_count) {
+			/* Abort the wait for completion if the FW health
+			 * check has failed.
+			 */
+			if (test_bit(BNXT_STATE_FW_FATAL_COND, &bp->state))
+				goto exit;
+			/* on first few passes, just barely sleep */
+			if (i < HWRM_SHORT_TIMEOUT_COUNTER) {
+				usleep_range(HWRM_SHORT_MIN_TIMEOUT,
+					     HWRM_SHORT_MAX_TIMEOUT);
+			} else {
+				if (hwrm_wait_must_abort(bp, req_type, &sts)) {
+					hwrm_err(bp, ctx, "Resp cmpl intr abandoning msg: 0x%x due to firmware status: 0x%x\n",
+						 req_type, sts);
+					goto exit;
+				}
+				usleep_range(HWRM_MIN_TIMEOUT,
+					     HWRM_MAX_TIMEOUT);
+			}
+		}
+
+		if (READ_ONCE(token->state) != BNXT_HWRM_COMPLETE) {
+			hwrm_err(bp, ctx, "Resp cmpl intr err msg: 0x%x\n",
+				 req_type);
+			goto exit;
+		}
+		len = le16_to_cpu(READ_ONCE(ctx->resp->resp_len));
+		valid = ((u8 *)ctx->resp) + len - 1;
+	} else {
+		__le16 seen_out_of_seq = ctx->req->seq_id; /* will never see */
+		int j;
+
+		/* Check if response len is updated */
+		for (i = 0; i < tmo_count; i++) {
+			/* Abort the wait for completion if the FW health
+			 * check has failed.
+			 */
+			if (test_bit(BNXT_STATE_FW_FATAL_COND, &bp->state))
+				goto exit;
+
+			if (token &&
+			    READ_ONCE(token->state) == BNXT_HWRM_DEFERRED) {
+				__hwrm_release_token(bp, token);
+				token = NULL;
+			}
+
+			len = le16_to_cpu(READ_ONCE(ctx->resp->resp_len));
+			if (len) {
+				__le16 resp_seq = READ_ONCE(ctx->resp->seq_id);
+
+				if (resp_seq == ctx->req->seq_id)
+					break;
+				if (resp_seq != seen_out_of_seq) {
+					netdev_warn(bp->dev, "Discarding out of seq response: 0x%x for msg {0x%x 0x%x}\n",
+						    le16_to_cpu(resp_seq),
+						    req_type,
+						    le16_to_cpu(ctx->req->seq_id));
+					seen_out_of_seq = resp_seq;
+				}
+			}
+
+			/* on first few passes, just barely sleep */
+			if (i < HWRM_SHORT_TIMEOUT_COUNTER) {
+				usleep_range(HWRM_SHORT_MIN_TIMEOUT,
+					     HWRM_SHORT_MAX_TIMEOUT);
+			} else {
+				if (hwrm_wait_must_abort(bp, req_type, &sts)) {
+					hwrm_err(bp, ctx, "Abandoning msg {0x%x 0x%x} len: %d due to firmware status: 0x%x\n",
+						 req_type,
+						 le16_to_cpu(ctx->req->seq_id),
+						 len, sts);
+					goto exit;
+				}
+				usleep_range(HWRM_MIN_TIMEOUT,
+					     HWRM_MAX_TIMEOUT);
+			}
+		}
+
+		if (i >= tmo_count) {
+			hwrm_err(bp, ctx, "Error (timeout: %u) msg {0x%x 0x%x} len:%d\n",
+				 hwrm_total_timeout(i), req_type,
+				 le16_to_cpu(ctx->req->seq_id), len);
+			goto exit;
+		}
+
+		/* Last byte of resp contains valid bit */
+		valid = ((u8 *)ctx->resp) + len - 1;
+		for (j = 0; j < HWRM_VALID_BIT_DELAY_USEC; ) {
+			/* make sure we read from updated DMA memory */
+			dma_rmb();
+			if (*valid)
+				break;
+			if (j < 10) {
+				udelay(1);
+				j++;
+			} else {
+				usleep_range(20, 30);
+				j += 20;
+			}
+		}
+
+		if (j >= HWRM_VALID_BIT_DELAY_USEC) {
+			hwrm_err(bp, ctx, "Error (timeout: %u) msg {0x%x 0x%x} len:%d v:%d\n",
+				 hwrm_total_timeout(i) + j, req_type,
+				 le16_to_cpu(ctx->req->seq_id), len, *valid);
+			goto exit;
+		}
+	}
+
+	/* Zero valid bit for compatibility.  Valid bit in an older spec
+	 * may become a new field in a newer spec.  We must make sure that
+	 * a new field not implemented by old spec will read zero.
+	 */
+	*valid = 0;
+	rc = le16_to_cpu(ctx->resp->error_code);
+	if (rc == HWRM_ERR_CODE_BUSY && !(ctx->flags & BNXT_HWRM_CTX_SILENT))
+		netdev_warn(bp->dev, "FW returned busy, hwrm req_type 0x%x\n",
+			    req_type);
+	else if (rc && rc != HWRM_ERR_CODE_PF_UNAVAILABLE)
+		hwrm_err(bp, ctx, "hwrm req_type 0x%x seq id 0x%x error 0x%x\n",
+			 req_type, le16_to_cpu(ctx->req->seq_id), rc);
+	rc = __hwrm_to_stderr(rc);
+exit:
+	if (token)
+		__hwrm_release_token(bp, token);
+	if (ctx->flags & BNXT_HWRM_INTERNAL_CTX_OWNED)
+		ctx->flags |= BNXT_HWRM_INTERNAL_RESP_DIRTY;
+	else
+		__hwrm_ctx_drop(bp, ctx);
+	return rc;
+}
+
+/**
+ * hwrm_req_send() - Execute an HWRM command.
+ * @bp: The driver context.
+ * @req: A pointer to the request to send. The DMA resources associated with
+ *	the request will be released (ie. the request will be consumed) unless
+ *	ownership of the request has been assumed by the caller via a call to
+ *	hwrm_req_hold().
+ *
+ * Send an HWRM request to the device and wait for a response. The request is
+ * consumed if it is not owned by the caller. This function will block until
+ * the request has either completed or times out due to an error.
+ *
+ * Return: A result code.
+ *
+ * The result is zero on success, otherwise the negative error code indicates
+ * one of the following errors:
+ *	E2BIG: The request was too large.
+ *	EBUSY: The firmware is in a fatal state or the request timed out
+ *	EACCESS: HWRM access denied.
+ *	ENOSPC: HWRM resource allocation error.
+ *	EINVAL: Request parameters are invalid.
+ *	ENOMEM: HWRM has no buffers.
+ *	EAGAIN: HWRM busy or reset in progress.
+ *	EOPNOTSUPP: Invalid request type.
+ *	EIO: Any other error.
+ * Error handling is orthogonal to request ownership. An unowned request will
+ * still be consumed on error. If the caller owns the request, then the caller
+ * is responsible for releasing the resources. Otherwise, hwrm_req_send() will
+ * always consume the request.
+ */
+int hwrm_req_send(struct bnxt *bp, void *req)
+{
+	struct bnxt_hwrm_ctx *ctx = __hwrm_ctx(bp, req);
+
+	if (!ctx)
+		return -EINVAL;
+
+	return __hwrm_send(bp, ctx);
+}
+
+/**
+ * hwrm_req_send_silent() - A silent version of hwrm_req_send().
+ * @bp: The driver context.
+ * @req: The request to send without logging.
+ *
+ * The same as hwrm_req_send(), except that the request is silenced using
+ * hwrm_req_silence() prior the call. This version of the function is
+ * provided solely to preserve the legacy API’s flavor for this functionality.
+ *
+ * Return: A result code, see hwrm_req_send().
+ */
+int hwrm_req_send_silent(struct bnxt *bp, void *req)
+{
+	hwrm_req_flags(bp, req, BNXT_HWRM_CTX_SILENT);
+	return hwrm_req_send(bp, req);
+}
+
+/**
+ * hwrm_req_dma_slice() - Allocate a slice of DMA mapped memory.
+ * @bp: The driver context.
+ * @req: The request for which indirect data will be associated.
+ * @size: The size of the allocation.
+ * @dma_handle: The bus address associated with the allocation. The HWRM API has
+ *	no knowledge about the type of the request and so cannot infer how the
+ *	caller intends to use the indirect data. Thus, the caller is
+ *	responsible for configuring the request object appropriately to
+ *	point to the associated indirect memory. Note, DMA handle has the
+ *	same definition as it does in dma_alloc_coherent(), the caller is
+ *	responsible for endian conversions via cpu_to_le64() before assigning
+ *	this address.
+ *
+ * Allocates DMA mapped memory for indirect data related to a request. The
+ * lifetime of the DMA resources will be bound to that of the request (ie.
+ * they will be automatically released when the request is either consumed by
+ * hwrm_req_send() or dropped by hwrm_req_drop()). Small allocations are
+ * efficiently suballocated out of the request buffer space, hence the name
+ * slice, while larger requests are satisfied via an underlying call to
+ * dma_alloc_coherent(). Multiple suballocations are supported, however, only
+ * one externally mapped region is.
+ *
+ * Return: The kernel virtual address of the DMA mapping.
+ */
+void *
+hwrm_req_dma_slice(struct bnxt *bp, void *req, u32 size, dma_addr_t *dma_handle)
+{
+	struct bnxt_hwrm_ctx *ctx = __hwrm_ctx(bp, req);
+	u8 *end = ((u8 *)req) + BNXT_HWRM_DMA_SIZE;
+	struct input *input = req;
+	u8 *addr, *req_addr = req;
+	u32 max_offset, offset;
+
+	if (!ctx)
+		return NULL;
+
+	max_offset = BNXT_HWRM_DMA_SIZE - ctx->allocated;
+	offset = max_offset - size;
+	offset = ALIGN_DOWN(offset, BNXT_HWRM_DMA_ALIGN);
+	addr = req_addr + offset;
+
+	if (addr < req_addr + max_offset && req_addr + ctx->req_len <= addr) {
+		ctx->allocated = end - addr;
+		*dma_handle = ctx->dma_handle + offset;
+		return addr;
+	}
+
+	/* could not suballocate from ctx buffer, try create a new mapping */
+	if (ctx->slice_addr) {
+		/* if one exists, can only be due to software bug, be loud */
+		netdev_err(bp->dev, "HWRM refusing to reallocate DMA slice, req_type = %u\n",
+			   (u32)le16_to_cpu(input->req_type));
+		dump_stack();
+		return NULL;
+	}
+
+	addr = dma_alloc_coherent(&bp->pdev->dev, size, dma_handle, ctx->gfp);
+
+	if (!addr)
+		return NULL;
+
+	ctx->slice_addr = addr;
+	ctx->slice_size = size;
+	ctx->slice_handle = *dma_handle;
+
+	return addr;
+}
diff --git a/drivers/net/ethernet/broadcom/bnxt/bnxt_hwrm.h b/drivers/net/ethernet/broadcom/bnxt/bnxt_hwrm.h
new file mode 100644
index 000000000000..791b3a0cdb83
--- /dev/null
+++ b/drivers/net/ethernet/broadcom/bnxt/bnxt_hwrm.h
@@ -0,0 +1,154 @@
+/* Broadcom NetXtreme-C/E network driver.
+ *
+ * Copyright (c) 2020 Broadcom Limited
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation.
+ */
+
+#ifndef BNXT_HWRM_H
+#define BNXT_HWRM_H
+
+#include <linux/bnxt/hsi.h>
+
+enum bnxt_hwrm_ctx_flags {
+	/* Update the HWRM_API_FLAGS right below for any new non-internal bit added here */
+	BNXT_HWRM_INTERNAL_CTX_OWNED	= BIT(0), /* caller owns the context */
+	BNXT_HWRM_INTERNAL_RESP_DIRTY	= BIT(1), /* response contains data */
+	BNXT_HWRM_CTX_SILENT		= BIT(2), /* squelch firmware errors */
+	BNXT_HWRM_FULL_WAIT		= BIT(3), /* wait for full timeout of HWRM command */
+};
+
+#define HWRM_API_FLAGS (BNXT_HWRM_CTX_SILENT | BNXT_HWRM_FULL_WAIT)
+
+struct bnxt_hwrm_ctx {
+	u64 sentinel;
+	dma_addr_t dma_handle;
+	struct output *resp;
+	struct input *req;
+	dma_addr_t slice_handle;
+	void *slice_addr;
+	u32 slice_size;
+	u32 req_len;
+	enum bnxt_hwrm_ctx_flags flags;
+	unsigned int timeout;
+	u32 allocated;
+	gfp_t gfp;
+};
+
+enum bnxt_hwrm_wait_state {
+	BNXT_HWRM_PENDING,
+	BNXT_HWRM_DEFERRED,
+	BNXT_HWRM_COMPLETE,
+	BNXT_HWRM_CANCELLED,
+};
+
+enum bnxt_hwrm_chnl { BNXT_HWRM_CHNL_CHIMP, BNXT_HWRM_CHNL_KONG };
+
+struct bnxt_hwrm_wait_token {
+	struct rcu_head rcu;
+	struct hlist_node node;
+	enum bnxt_hwrm_wait_state state;
+	enum bnxt_hwrm_chnl dst;
+	u16 seq_id;
+};
+
+void hwrm_update_token(struct bnxt *bp, u16 seq, enum bnxt_hwrm_wait_state s);
+
+#define BNXT_HWRM_MAX_REQ_LEN		(bp->hwrm_max_req_len)
+#define BNXT_HWRM_SHORT_REQ_LEN		sizeof(struct hwrm_short_input)
+#define HWRM_CMD_MAX_TIMEOUT		60000U
+#define SHORT_HWRM_CMD_TIMEOUT		20
+#define HWRM_CMD_TIMEOUT		(bp->hwrm_cmd_timeout)
+#define HWRM_RESET_TIMEOUT		((HWRM_CMD_TIMEOUT) * 4)
+#define BNXT_HWRM_TARGET		0xffff
+#define BNXT_HWRM_NO_CMPL_RING		-1
+#define BNXT_HWRM_REQ_MAX_SIZE		128
+#define BNXT_HWRM_DMA_SIZE		(2 * PAGE_SIZE) /* space for req+resp */
+#define BNXT_HWRM_RESP_RESERVED		PAGE_SIZE
+#define BNXT_HWRM_RESP_OFFSET		(BNXT_HWRM_DMA_SIZE -		\
+					 BNXT_HWRM_RESP_RESERVED)
+#define BNXT_HWRM_CTX_OFFSET		(BNXT_HWRM_RESP_OFFSET -	\
+					 sizeof(struct bnxt_hwrm_ctx))
+#define BNXT_HWRM_DMA_ALIGN		16
+#define BNXT_HWRM_SENTINEL		0xb6e1f68a12e9a7eb /* arbitrary value */
+#define BNXT_HWRM_REQS_PER_PAGE		(BNXT_PAGE_SIZE /	\
+					 BNXT_HWRM_REQ_MAX_SIZE)
+#define HWRM_SHORT_MIN_TIMEOUT		3
+#define HWRM_SHORT_MAX_TIMEOUT		10
+#define HWRM_SHORT_TIMEOUT_COUNTER	5
+
+#define HWRM_MIN_TIMEOUT		25
+#define HWRM_MAX_TIMEOUT		40
+
+static inline unsigned int hwrm_total_timeout(unsigned int n)
+{
+	return n <= HWRM_SHORT_TIMEOUT_COUNTER ? n * HWRM_SHORT_MIN_TIMEOUT :
+		HWRM_SHORT_TIMEOUT_COUNTER * HWRM_SHORT_MIN_TIMEOUT +
+		(n - HWRM_SHORT_TIMEOUT_COUNTER) * HWRM_MIN_TIMEOUT;
+}
+
+
+#define HWRM_VALID_BIT_DELAY_USEC	50000
+
+static inline bool bnxt_cfa_hwrm_message(u16 req_type)
+{
+	switch (req_type) {
+	case HWRM_CFA_ENCAP_RECORD_ALLOC:
+	case HWRM_CFA_ENCAP_RECORD_FREE:
+	case HWRM_CFA_DECAP_FILTER_ALLOC:
+	case HWRM_CFA_DECAP_FILTER_FREE:
+	case HWRM_CFA_EM_FLOW_ALLOC:
+	case HWRM_CFA_EM_FLOW_FREE:
+	case HWRM_CFA_EM_FLOW_CFG:
+	case HWRM_CFA_FLOW_ALLOC:
+	case HWRM_CFA_FLOW_FREE:
+	case HWRM_CFA_FLOW_INFO:
+	case HWRM_CFA_FLOW_FLUSH:
+	case HWRM_CFA_FLOW_STATS:
+	case HWRM_CFA_METER_PROFILE_ALLOC:
+	case HWRM_CFA_METER_PROFILE_FREE:
+	case HWRM_CFA_METER_PROFILE_CFG:
+	case HWRM_CFA_METER_INSTANCE_ALLOC:
+	case HWRM_CFA_METER_INSTANCE_FREE:
+		return true;
+	default:
+		return false;
+	}
+}
+
+static inline bool bnxt_kong_hwrm_message(struct bnxt *bp, struct input *req)
+{
+	return (bp->fw_cap & BNXT_FW_CAP_KONG_MB_CHNL &&
+		(bnxt_cfa_hwrm_message(le16_to_cpu(req->req_type)) ||
+		 le16_to_cpu(req->target_id) == HWRM_TARGET_ID_KONG));
+}
+
+int __hwrm_req_init(struct bnxt *bp, void **req, u16 req_type, u32 req_len);
+#define hwrm_req_init(bp, req, req_type) \
+	__hwrm_req_init((bp), (void **)&(req), (req_type), sizeof(*(req)))
+void *hwrm_req_hold(struct bnxt *bp, void *req);
+void hwrm_req_drop(struct bnxt *bp, void *req);
+void hwrm_req_flags(struct bnxt *bp, void *req, enum bnxt_hwrm_ctx_flags flags);
+void hwrm_req_timeout(struct bnxt *bp, void *req, unsigned int timeout);
+int hwrm_req_send(struct bnxt *bp, void *req);
+int hwrm_req_send_silent(struct bnxt *bp, void *req);
+int hwrm_req_replace(struct bnxt *bp, void *req, void *new_req, u32 len);
+void hwrm_req_alloc_flags(struct bnxt *bp, void *req, gfp_t flags);
+void *hwrm_req_dma_slice(struct bnxt *bp, void *req, u32 size, dma_addr_t *dma);
+
+/* Older devices can only support req length of 128.
+ * HWRM_FUNC_CFG requests which don't need fields starting at
+ * num_quic_tx_key_ctxs can use this helper to avoid getting -E2BIG.
+ */
+static inline int
+bnxt_hwrm_func_cfg_short_req_init(struct bnxt *bp,
+				  struct hwrm_func_cfg_input **req)
+{
+	u32 req_len;
+
+	req_len = min_t(u32, sizeof(**req), bp->hwrm_max_ext_req_len);
+	return __hwrm_req_init(bp, (void **)req, HWRM_FUNC_CFG, req_len);
+}
+#endif
diff --git a/drivers/net/ethernet/broadcom/bnxt/bnxt_nvm_defs.h b/drivers/net/ethernet/broadcom/bnxt/bnxt_nvm_defs.h
index 73f2249555b5..83444811d3c6 100644
--- a/drivers/net/ethernet/broadcom/bnxt/bnxt_nvm_defs.h
+++ b/drivers/net/ethernet/broadcom/bnxt/bnxt_nvm_defs.h
@@ -59,8 +59,6 @@ enum bnxt_nvm_directory_type {
 #define BNX_DIR_ATTR_NO_CHKSUM			(1 << 0)
 #define BNX_DIR_ATTR_PROP_STREAM		(1 << 1)
 
-#define BNX_PKG_LOG_MAX_LENGTH			4096
-
 enum bnxnvm_pkglog_field_index {
 	BNX_PKG_LOG_FIELD_IDX_INSTALLED_TIMESTAMP	= 0,
 	BNX_PKG_LOG_FIELD_IDX_PKG_DESCRIPTION		= 1,
diff --git a/drivers/net/ethernet/broadcom/bnxt/bnxt_ptp.c b/drivers/net/ethernet/broadcom/bnxt/bnxt_ptp.c
new file mode 100644
index 000000000000..db81cf6d5289
--- /dev/null
+++ b/drivers/net/ethernet/broadcom/bnxt/bnxt_ptp.c
@@ -0,0 +1,1138 @@
+/* Broadcom NetXtreme-C/E network driver.
+ *
+ * Copyright (c) 2021 Broadcom Inc.
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation.
+ */
+#include <linux/kernel.h>
+#include <linux/errno.h>
+#include <linux/pci.h>
+#include <linux/netdevice.h>
+#include <linux/etherdevice.h>
+#include <linux/net_tstamp.h>
+#include <linux/timekeeping.h>
+#include <linux/ptp_classify.h>
+#include <linux/clocksource.h>
+#include <linux/bnxt/hsi.h>
+#include "bnxt.h"
+#include "bnxt_hwrm.h"
+#include "bnxt_ptp.h"
+
+static int bnxt_ptp_cfg_settime(struct bnxt *bp, u64 time)
+{
+	struct hwrm_func_ptp_cfg_input *req;
+	int rc;
+
+	rc = hwrm_req_init(bp, req, HWRM_FUNC_PTP_CFG);
+	if (rc)
+		return rc;
+
+	req->enables = cpu_to_le16(FUNC_PTP_CFG_REQ_ENABLES_PTP_SET_TIME);
+	req->ptp_set_time = cpu_to_le64(time);
+	return hwrm_req_send(bp, req);
+}
+
+int bnxt_ptp_parse(struct sk_buff *skb, u16 *seq_id, u16 *hdr_off)
+{
+	unsigned int ptp_class;
+	struct ptp_header *hdr;
+
+	ptp_class = ptp_classify_raw(skb);
+
+	switch (ptp_class & PTP_CLASS_VMASK) {
+	case PTP_CLASS_V1:
+	case PTP_CLASS_V2:
+		hdr = ptp_parse_header(skb, ptp_class);
+		if (!hdr)
+			return -EINVAL;
+
+		*hdr_off = (u8 *)hdr - skb->data;
+		*seq_id	 = ntohs(hdr->sequence_id);
+		return 0;
+	default:
+		return -ERANGE;
+	}
+}
+
+static int bnxt_ptp_settime(struct ptp_clock_info *ptp_info,
+			    const struct timespec64 *ts)
+{
+	struct bnxt_ptp_cfg *ptp = container_of(ptp_info, struct bnxt_ptp_cfg,
+						ptp_info);
+	u64 ns = timespec64_to_ns(ts);
+	unsigned long flags;
+
+	if (BNXT_PTP_USE_RTC(ptp->bp))
+		return bnxt_ptp_cfg_settime(ptp->bp, ns);
+
+	write_seqlock_irqsave(&ptp->ptp_lock, flags);
+	timecounter_init(&ptp->tc, &ptp->cc, ns);
+	write_sequnlock_irqrestore(&ptp->ptp_lock, flags);
+	return 0;
+}
+
+/* Caller holds ptp_lock */
+static int __bnxt_refclk_read(struct bnxt *bp, struct ptp_system_timestamp *sts,
+			      u64 *ns)
+{
+	struct bnxt_ptp_cfg *ptp = bp->ptp_cfg;
+	u32 high_before, high_now, low;
+
+	if (test_bit(BNXT_STATE_IN_FW_RESET, &bp->state))
+		return -EIO;
+
+	high_before = readl(bp->bar0 + ptp->refclk_mapped_regs[1]);
+	ptp_read_system_prets(sts);
+	low = readl(bp->bar0 + ptp->refclk_mapped_regs[0]);
+	ptp_read_system_postts(sts);
+	high_now = readl(bp->bar0 + ptp->refclk_mapped_regs[1]);
+	if (high_now != high_before) {
+		ptp_read_system_prets(sts);
+		low = readl(bp->bar0 + ptp->refclk_mapped_regs[0]);
+		ptp_read_system_postts(sts);
+	}
+	*ns = ((u64)high_now << 32) | low;
+
+	return 0;
+}
+
+static int bnxt_refclk_read(struct bnxt *bp, struct ptp_system_timestamp *sts,
+			    u64 *ns)
+{
+	struct bnxt_ptp_cfg *ptp = bp->ptp_cfg;
+	unsigned long flags;
+	int rc;
+
+	/* We have to serialize reg access and FW reset */
+	read_seqlock_excl_irqsave(&ptp->ptp_lock, flags);
+	rc = __bnxt_refclk_read(bp, sts, ns);
+	read_sequnlock_excl_irqrestore(&ptp->ptp_lock, flags);
+	return rc;
+}
+
+static int bnxt_refclk_read_low(struct bnxt *bp, struct ptp_system_timestamp *sts,
+				u32 *low)
+{
+	struct bnxt_ptp_cfg *ptp = bp->ptp_cfg;
+	unsigned long flags;
+
+	/* We have to serialize reg access and FW reset */
+	read_seqlock_excl_irqsave(&ptp->ptp_lock, flags);
+
+	if (test_bit(BNXT_STATE_IN_FW_RESET, &bp->state)) {
+		read_sequnlock_excl_irqrestore(&ptp->ptp_lock, flags);
+		return -EIO;
+	}
+
+	ptp_read_system_prets(sts);
+	*low = readl(bp->bar0 + ptp->refclk_mapped_regs[0]);
+	ptp_read_system_postts(sts);
+
+	read_sequnlock_excl_irqrestore(&ptp->ptp_lock, flags);
+	return 0;
+}
+
+static void bnxt_ptp_get_current_time(struct bnxt *bp)
+{
+	struct bnxt_ptp_cfg *ptp = bp->ptp_cfg;
+
+	if (!ptp)
+		return;
+	WRITE_ONCE(ptp->old_time, ptp->current_time >> BNXT_HI_TIMER_SHIFT);
+	bnxt_refclk_read(bp, NULL, &ptp->current_time);
+}
+
+static int bnxt_hwrm_port_ts_query(struct bnxt *bp, u32 flags, u64 *ts,
+				   u32 txts_tmo, int slot)
+{
+	struct hwrm_port_ts_query_output *resp;
+	struct hwrm_port_ts_query_input *req;
+	int rc;
+
+	rc = hwrm_req_init(bp, req, HWRM_PORT_TS_QUERY);
+	if (rc)
+		return rc;
+
+	req->flags = cpu_to_le32(flags);
+	if ((flags & PORT_TS_QUERY_REQ_FLAGS_PATH) ==
+	    PORT_TS_QUERY_REQ_FLAGS_PATH_TX) {
+		struct bnxt_ptp_tx_req *txts_req = &bp->ptp_cfg->txts_req[slot];
+		u32 tmo_us = txts_tmo * 1000;
+
+		req->enables = cpu_to_le16(BNXT_PTP_QTS_TX_ENABLES);
+		req->ptp_seq_id = cpu_to_le32(txts_req->tx_seqid);
+		req->ptp_hdr_offset = cpu_to_le16(txts_req->tx_hdr_off);
+		if (!tmo_us)
+			tmo_us = BNXT_PTP_QTS_TIMEOUT;
+		tmo_us = min(tmo_us, BNXT_PTP_QTS_MAX_TMO_US);
+		req->ts_req_timeout = cpu_to_le16(tmo_us);
+	}
+	resp = hwrm_req_hold(bp, req);
+
+	rc = hwrm_req_send_silent(bp, req);
+	if (!rc)
+		*ts = le64_to_cpu(resp->ptp_msg_ts);
+	hwrm_req_drop(bp, req);
+	return rc;
+}
+
+static int bnxt_ptp_gettimex(struct ptp_clock_info *ptp_info,
+			     struct timespec64 *ts,
+			     struct ptp_system_timestamp *sts)
+{
+	struct bnxt_ptp_cfg *ptp = container_of(ptp_info, struct bnxt_ptp_cfg,
+						ptp_info);
+	u64 ns, cycles;
+	u32 low;
+	int rc;
+
+	rc = bnxt_refclk_read_low(ptp->bp, sts, &low);
+	if (rc)
+		return rc;
+
+	cycles = bnxt_extend_cycles_32b_to_48b(ptp, low);
+	ns = bnxt_timecounter_cyc2time(ptp, cycles);
+	*ts = ns_to_timespec64(ns);
+
+	return 0;
+}
+
+void bnxt_ptp_update_current_time(struct bnxt *bp)
+{
+	struct bnxt_ptp_cfg *ptp = bp->ptp_cfg;
+
+	bnxt_refclk_read(ptp->bp, NULL, &ptp->current_time);
+	WRITE_ONCE(ptp->old_time, ptp->current_time >> BNXT_HI_TIMER_SHIFT);
+}
+
+static int bnxt_ptp_adjphc(struct bnxt_ptp_cfg *ptp, s64 delta)
+{
+	struct hwrm_port_mac_cfg_input *req;
+	int rc;
+
+	rc = hwrm_req_init(ptp->bp, req, HWRM_PORT_MAC_CFG);
+	if (rc)
+		return rc;
+
+	req->enables = cpu_to_le32(PORT_MAC_CFG_REQ_ENABLES_PTP_ADJ_PHASE);
+	req->ptp_adj_phase = cpu_to_le64(delta);
+
+	rc = hwrm_req_send(ptp->bp, req);
+	if (rc) {
+		netdev_err(ptp->bp->dev, "ptp adjphc failed. rc = %x\n", rc);
+	} else {
+		bnxt_ptp_update_current_time(ptp->bp);
+	}
+
+	return rc;
+}
+
+static int bnxt_ptp_adjtime(struct ptp_clock_info *ptp_info, s64 delta)
+{
+	struct bnxt_ptp_cfg *ptp = container_of(ptp_info, struct bnxt_ptp_cfg,
+						ptp_info);
+	unsigned long flags;
+
+	if (BNXT_PTP_USE_RTC(ptp->bp))
+		return bnxt_ptp_adjphc(ptp, delta);
+
+	write_seqlock_irqsave(&ptp->ptp_lock, flags);
+	timecounter_adjtime(&ptp->tc, delta);
+	write_sequnlock_irqrestore(&ptp->ptp_lock, flags);
+	return 0;
+}
+
+static int bnxt_ptp_adjfine_rtc(struct bnxt *bp, long scaled_ppm)
+{
+	s32 ppb = scaled_ppm_to_ppb(scaled_ppm);
+	struct hwrm_port_mac_cfg_input *req;
+	int rc;
+
+	rc = hwrm_req_init(bp, req, HWRM_PORT_MAC_CFG);
+	if (rc)
+		return rc;
+
+	req->ptp_freq_adj_ppb = cpu_to_le32(ppb);
+	req->enables = cpu_to_le32(PORT_MAC_CFG_REQ_ENABLES_PTP_FREQ_ADJ_PPB);
+	rc = hwrm_req_send(bp, req);
+	if (rc)
+		netdev_err(bp->dev,
+			   "ptp adjfine failed. rc = %d\n", rc);
+	return rc;
+}
+
+static int bnxt_ptp_adjfine(struct ptp_clock_info *ptp_info, long scaled_ppm)
+{
+	struct bnxt_ptp_cfg *ptp = container_of(ptp_info, struct bnxt_ptp_cfg,
+						ptp_info);
+	struct bnxt *bp = ptp->bp;
+	unsigned long flags;
+
+	if (!BNXT_MH(bp))
+		return bnxt_ptp_adjfine_rtc(bp, scaled_ppm);
+
+	write_seqlock_irqsave(&ptp->ptp_lock, flags);
+	timecounter_read(&ptp->tc);
+	ptp->cc.mult = adjust_by_scaled_ppm(ptp->cmult, scaled_ppm);
+	write_sequnlock_irqrestore(&ptp->ptp_lock, flags);
+	return 0;
+}
+
+void bnxt_ptp_pps_event(struct bnxt *bp, u32 data1, u32 data2)
+{
+	struct bnxt_ptp_cfg *ptp = bp->ptp_cfg;
+	struct ptp_clock_event event;
+	u64 ns, pps_ts;
+
+	pps_ts = EVENT_PPS_TS(data2, data1);
+	ns = bnxt_timecounter_cyc2time(ptp, pps_ts);
+
+	switch (EVENT_DATA2_PPS_EVENT_TYPE(data2)) {
+	case ASYNC_EVENT_CMPL_PPS_TIMESTAMP_EVENT_DATA2_EVENT_TYPE_INTERNAL:
+		event.pps_times.ts_real = ns_to_timespec64(ns);
+		event.type = PTP_CLOCK_PPSUSR;
+		event.index = EVENT_DATA2_PPS_PIN_NUM(data2);
+		break;
+	case ASYNC_EVENT_CMPL_PPS_TIMESTAMP_EVENT_DATA2_EVENT_TYPE_EXTERNAL:
+		event.timestamp = ns;
+		event.type = PTP_CLOCK_EXTTS;
+		event.index = EVENT_DATA2_PPS_PIN_NUM(data2);
+		break;
+	}
+
+	ptp_clock_event(bp->ptp_cfg->ptp_clock, &event);
+}
+
+static int bnxt_ptp_cfg_pin(struct bnxt *bp, u8 pin, u8 usage)
+{
+	struct hwrm_func_ptp_pin_cfg_input *req;
+	struct bnxt_ptp_cfg *ptp = bp->ptp_cfg;
+	u8 state = usage != BNXT_PPS_PIN_NONE;
+	u8 *pin_state, *pin_usg;
+	u32 enables;
+	int rc;
+
+	if (!TSIO_PIN_VALID(pin)) {
+		netdev_err(ptp->bp->dev, "1PPS: Invalid pin. Check pin-function configuration\n");
+		return -EOPNOTSUPP;
+	}
+
+	rc = hwrm_req_init(ptp->bp, req, HWRM_FUNC_PTP_PIN_CFG);
+	if (rc)
+		return rc;
+
+	enables = (FUNC_PTP_PIN_CFG_REQ_ENABLES_PIN0_STATE |
+		   FUNC_PTP_PIN_CFG_REQ_ENABLES_PIN0_USAGE) << (pin * 2);
+	req->enables = cpu_to_le32(enables);
+
+	pin_state = &req->pin0_state;
+	pin_usg = &req->pin0_usage;
+
+	*(pin_state + (pin * 2)) = state;
+	*(pin_usg + (pin * 2)) = usage;
+
+	rc = hwrm_req_send(ptp->bp, req);
+	if (rc)
+		return rc;
+
+	ptp->pps_info.pins[pin].usage = usage;
+	ptp->pps_info.pins[pin].state = state;
+
+	return 0;
+}
+
+static int bnxt_ptp_cfg_event(struct bnxt *bp, u8 event)
+{
+	struct hwrm_func_ptp_cfg_input *req;
+	int rc;
+
+	rc = hwrm_req_init(bp, req, HWRM_FUNC_PTP_CFG);
+	if (rc)
+		return rc;
+
+	req->enables = cpu_to_le16(FUNC_PTP_CFG_REQ_ENABLES_PTP_PPS_EVENT);
+	req->ptp_pps_event = event;
+	return hwrm_req_send(bp, req);
+}
+
+int bnxt_ptp_cfg_tstamp_filters(struct bnxt *bp)
+{
+	struct bnxt_ptp_cfg *ptp = bp->ptp_cfg;
+	struct hwrm_port_mac_cfg_input *req;
+	int rc;
+
+	if (!ptp || !ptp->tstamp_filters)
+		return -EIO;
+
+	rc = hwrm_req_init(bp, req, HWRM_PORT_MAC_CFG);
+	if (rc)
+		goto out;
+
+	if (!(bp->fw_cap & BNXT_FW_CAP_RX_ALL_PKT_TS) && (ptp->tstamp_filters &
+	    (PORT_MAC_CFG_REQ_FLAGS_ALL_RX_TS_CAPTURE_ENABLE |
+	     PORT_MAC_CFG_REQ_FLAGS_ALL_RX_TS_CAPTURE_DISABLE))) {
+		ptp->tstamp_filters &= ~(PORT_MAC_CFG_REQ_FLAGS_ALL_RX_TS_CAPTURE_ENABLE |
+					 PORT_MAC_CFG_REQ_FLAGS_ALL_RX_TS_CAPTURE_DISABLE);
+		netdev_warn(bp->dev, "Unsupported FW for all RX pkts timestamp filter\n");
+	}
+
+	req->flags = cpu_to_le32(ptp->tstamp_filters);
+	req->enables = cpu_to_le32(PORT_MAC_CFG_REQ_ENABLES_RX_TS_CAPTURE_PTP_MSG_TYPE);
+	req->rx_ts_capture_ptp_msg_type = cpu_to_le16(ptp->rxctl);
+
+	rc = hwrm_req_send(bp, req);
+	if (!rc) {
+		bp->ptp_all_rx_tstamp = !!(ptp->tstamp_filters &
+					   PORT_MAC_CFG_REQ_FLAGS_ALL_RX_TS_CAPTURE_ENABLE);
+		return 0;
+	}
+	ptp->tstamp_filters = 0;
+out:
+	bp->ptp_all_rx_tstamp = 0;
+	netdev_warn(bp->dev, "Failed to configure HW packet timestamp filters\n");
+	return rc;
+}
+
+void bnxt_ptp_reapply_pps(struct bnxt *bp)
+{
+	struct bnxt_ptp_cfg *ptp = bp->ptp_cfg;
+	struct bnxt_pps *pps;
+	u32 pin = 0;
+	int rc;
+
+	if (!ptp || !(bp->fw_cap & BNXT_FW_CAP_PTP_PPS) ||
+	    !(ptp->ptp_info.pin_config))
+		return;
+	pps = &ptp->pps_info;
+	for (pin = 0; pin < BNXT_MAX_TSIO_PINS; pin++) {
+		if (pps->pins[pin].state) {
+			rc = bnxt_ptp_cfg_pin(bp, pin, pps->pins[pin].usage);
+			if (!rc && pps->pins[pin].event)
+				rc = bnxt_ptp_cfg_event(bp,
+							pps->pins[pin].event);
+			if (rc)
+				netdev_err(bp->dev, "1PPS: Failed to configure pin%d\n",
+					   pin);
+		}
+	}
+}
+
+static int bnxt_get_target_cycles(struct bnxt_ptp_cfg *ptp, u64 target_ns,
+				  u64 *cycles_delta)
+{
+	u64 cycles_now;
+	u64 nsec_now, nsec_delta;
+	int rc;
+
+	rc = bnxt_refclk_read(ptp->bp, NULL, &cycles_now);
+	if (rc)
+		return rc;
+
+	nsec_now = bnxt_timecounter_cyc2time(ptp, cycles_now);
+
+	nsec_delta = target_ns - nsec_now;
+	*cycles_delta = div64_u64(nsec_delta << ptp->cc.shift, ptp->cc.mult);
+	return 0;
+}
+
+static int bnxt_ptp_perout_cfg(struct bnxt_ptp_cfg *ptp,
+			       struct ptp_clock_request *rq)
+{
+	struct hwrm_func_ptp_cfg_input *req;
+	struct bnxt *bp = ptp->bp;
+	struct timespec64 ts;
+	u64 target_ns, delta;
+	u16 enables;
+	int rc;
+
+	ts.tv_sec = rq->perout.start.sec;
+	ts.tv_nsec = rq->perout.start.nsec;
+	target_ns = timespec64_to_ns(&ts);
+
+	rc = bnxt_get_target_cycles(ptp, target_ns, &delta);
+	if (rc)
+		return rc;
+
+	rc = hwrm_req_init(bp, req, HWRM_FUNC_PTP_CFG);
+	if (rc)
+		return rc;
+
+	enables = FUNC_PTP_CFG_REQ_ENABLES_PTP_FREQ_ADJ_EXT_PERIOD |
+		  FUNC_PTP_CFG_REQ_ENABLES_PTP_FREQ_ADJ_EXT_UP |
+		  FUNC_PTP_CFG_REQ_ENABLES_PTP_FREQ_ADJ_EXT_PHASE;
+	req->enables = cpu_to_le16(enables);
+	req->ptp_pps_event = 0;
+	req->ptp_freq_adj_dll_source = 0;
+	req->ptp_freq_adj_dll_phase = 0;
+	req->ptp_freq_adj_ext_period = cpu_to_le32(NSEC_PER_SEC);
+	req->ptp_freq_adj_ext_up = 0;
+	req->ptp_freq_adj_ext_phase_lower = cpu_to_le32(delta);
+
+	return hwrm_req_send(bp, req);
+}
+
+static int bnxt_ptp_enable(struct ptp_clock_info *ptp_info,
+			   struct ptp_clock_request *rq, int on)
+{
+	struct bnxt_ptp_cfg *ptp = container_of(ptp_info, struct bnxt_ptp_cfg,
+						ptp_info);
+	struct bnxt *bp = ptp->bp;
+	int pin_id;
+	int rc;
+
+	switch (rq->type) {
+	case PTP_CLK_REQ_EXTTS:
+		/* Configure an External PPS IN */
+		pin_id = ptp_find_pin(ptp->ptp_clock, PTP_PF_EXTTS,
+				      rq->extts.index);
+		if (!TSIO_PIN_VALID(pin_id))
+			return -EOPNOTSUPP;
+		if (!on)
+			break;
+		rc = bnxt_ptp_cfg_pin(bp, pin_id, BNXT_PPS_PIN_PPS_IN);
+		if (rc)
+			return rc;
+		rc = bnxt_ptp_cfg_event(bp, BNXT_PPS_EVENT_EXTERNAL);
+		if (!rc)
+			ptp->pps_info.pins[pin_id].event = BNXT_PPS_EVENT_EXTERNAL;
+		return rc;
+	case PTP_CLK_REQ_PEROUT:
+		/* Configure a Periodic PPS OUT */
+		pin_id = ptp_find_pin(ptp->ptp_clock, PTP_PF_PEROUT,
+				      rq->perout.index);
+		if (!TSIO_PIN_VALID(pin_id))
+			return -EOPNOTSUPP;
+		if (!on)
+			break;
+
+		rc = bnxt_ptp_cfg_pin(bp, pin_id, BNXT_PPS_PIN_PPS_OUT);
+		if (!rc)
+			rc = bnxt_ptp_perout_cfg(ptp, rq);
+
+		return rc;
+	case PTP_CLK_REQ_PPS:
+		/* Configure PHC PPS IN */
+		rc = bnxt_ptp_cfg_pin(bp, 0, BNXT_PPS_PIN_PPS_IN);
+		if (rc)
+			return rc;
+		rc = bnxt_ptp_cfg_event(bp, BNXT_PPS_EVENT_INTERNAL);
+		if (!rc)
+			ptp->pps_info.pins[0].event = BNXT_PPS_EVENT_INTERNAL;
+		return rc;
+	default:
+		netdev_err(ptp->bp->dev, "Unrecognized PIN function\n");
+		return -EOPNOTSUPP;
+	}
+
+	return bnxt_ptp_cfg_pin(bp, pin_id, BNXT_PPS_PIN_NONE);
+}
+
+static int bnxt_hwrm_ptp_cfg(struct bnxt *bp)
+{
+	struct bnxt_ptp_cfg *ptp = bp->ptp_cfg;
+	u32 flags = 0;
+
+	switch (ptp->rx_filter) {
+	case HWTSTAMP_FILTER_ALL:
+		flags = PORT_MAC_CFG_REQ_FLAGS_ALL_RX_TS_CAPTURE_ENABLE;
+		break;
+	case HWTSTAMP_FILTER_NONE:
+		flags = PORT_MAC_CFG_REQ_FLAGS_PTP_RX_TS_CAPTURE_DISABLE;
+		if (bp->fw_cap & BNXT_FW_CAP_RX_ALL_PKT_TS)
+			flags |= PORT_MAC_CFG_REQ_FLAGS_ALL_RX_TS_CAPTURE_DISABLE;
+		break;
+	case HWTSTAMP_FILTER_PTP_V2_EVENT:
+	case HWTSTAMP_FILTER_PTP_V2_SYNC:
+	case HWTSTAMP_FILTER_PTP_V2_DELAY_REQ:
+		flags = PORT_MAC_CFG_REQ_FLAGS_PTP_RX_TS_CAPTURE_ENABLE;
+		break;
+	}
+
+	if (ptp->tx_tstamp_en)
+		flags |= PORT_MAC_CFG_REQ_FLAGS_PTP_TX_TS_CAPTURE_ENABLE;
+	else
+		flags |= PORT_MAC_CFG_REQ_FLAGS_PTP_TX_TS_CAPTURE_DISABLE;
+
+	ptp->tstamp_filters = flags;
+
+	return bnxt_ptp_cfg_tstamp_filters(bp);
+}
+
+int bnxt_hwtstamp_set(struct net_device *dev,
+		      struct kernel_hwtstamp_config *stmpconf,
+		      struct netlink_ext_ack *extack)
+{
+	struct bnxt *bp = netdev_priv(dev);
+	struct bnxt_ptp_cfg *ptp;
+	u16 old_rxctl;
+	int old_rx_filter, rc;
+	u8 old_tx_tstamp_en;
+
+	ptp = bp->ptp_cfg;
+	if (!ptp)
+		return -EOPNOTSUPP;
+
+	if (stmpconf->tx_type != HWTSTAMP_TX_ON &&
+	    stmpconf->tx_type != HWTSTAMP_TX_OFF)
+		return -ERANGE;
+
+	old_rx_filter = ptp->rx_filter;
+	old_rxctl = ptp->rxctl;
+	old_tx_tstamp_en = ptp->tx_tstamp_en;
+	switch (stmpconf->rx_filter) {
+	case HWTSTAMP_FILTER_NONE:
+		ptp->rxctl = 0;
+		ptp->rx_filter = HWTSTAMP_FILTER_NONE;
+		break;
+	case HWTSTAMP_FILTER_ALL:
+		if (bp->fw_cap & BNXT_FW_CAP_RX_ALL_PKT_TS) {
+			ptp->rx_filter = HWTSTAMP_FILTER_ALL;
+			break;
+		}
+		return -EOPNOTSUPP;
+	case HWTSTAMP_FILTER_PTP_V2_EVENT:
+	case HWTSTAMP_FILTER_PTP_V2_L2_EVENT:
+	case HWTSTAMP_FILTER_PTP_V2_L4_EVENT:
+		ptp->rxctl = BNXT_PTP_MSG_EVENTS;
+		ptp->rx_filter = HWTSTAMP_FILTER_PTP_V2_EVENT;
+		break;
+	case HWTSTAMP_FILTER_PTP_V2_SYNC:
+	case HWTSTAMP_FILTER_PTP_V2_L2_SYNC:
+	case HWTSTAMP_FILTER_PTP_V2_L4_SYNC:
+		ptp->rxctl = BNXT_PTP_MSG_SYNC;
+		ptp->rx_filter = HWTSTAMP_FILTER_PTP_V2_SYNC;
+		break;
+	case HWTSTAMP_FILTER_PTP_V2_DELAY_REQ:
+	case HWTSTAMP_FILTER_PTP_V2_L2_DELAY_REQ:
+	case HWTSTAMP_FILTER_PTP_V2_L4_DELAY_REQ:
+		ptp->rxctl = BNXT_PTP_MSG_DELAY_REQ;
+		ptp->rx_filter = HWTSTAMP_FILTER_PTP_V2_DELAY_REQ;
+		break;
+	default:
+		return -ERANGE;
+	}
+
+	if (stmpconf->tx_type == HWTSTAMP_TX_ON)
+		ptp->tx_tstamp_en = 1;
+	else
+		ptp->tx_tstamp_en = 0;
+
+	rc = bnxt_hwrm_ptp_cfg(bp);
+	if (rc)
+		goto ts_set_err;
+
+	stmpconf->rx_filter = ptp->rx_filter;
+	return 0;
+
+ts_set_err:
+	ptp->rx_filter = old_rx_filter;
+	ptp->rxctl = old_rxctl;
+	ptp->tx_tstamp_en = old_tx_tstamp_en;
+	return rc;
+}
+
+int bnxt_hwtstamp_get(struct net_device *dev,
+		      struct kernel_hwtstamp_config *stmpconf)
+{
+	struct bnxt *bp = netdev_priv(dev);
+	struct bnxt_ptp_cfg *ptp;
+
+	ptp = bp->ptp_cfg;
+	if (!ptp)
+		return -EOPNOTSUPP;
+
+	stmpconf->flags = 0;
+	stmpconf->tx_type = ptp->tx_tstamp_en ? HWTSTAMP_TX_ON
+					      : HWTSTAMP_TX_OFF;
+
+	stmpconf->rx_filter = ptp->rx_filter;
+	return 0;
+}
+
+static int bnxt_map_regs(struct bnxt *bp, u32 *reg_arr, int count, int reg_win)
+{
+	u32 reg_base = *reg_arr & BNXT_GRC_BASE_MASK;
+	u32 win_off;
+	int i;
+
+	for (i = 0; i < count; i++) {
+		if ((reg_arr[i] & BNXT_GRC_BASE_MASK) != reg_base)
+			return -ERANGE;
+	}
+	win_off = BNXT_GRCPF_REG_WINDOW_BASE_OUT + (reg_win - 1) * 4;
+	writel(reg_base, bp->bar0 + win_off);
+	return 0;
+}
+
+static int bnxt_map_ptp_regs(struct bnxt *bp)
+{
+	struct bnxt_ptp_cfg *ptp = bp->ptp_cfg;
+	u32 *reg_arr;
+	int rc, i;
+
+	reg_arr = ptp->refclk_regs;
+	if (BNXT_CHIP_P5(bp)) {
+		rc = bnxt_map_regs(bp, reg_arr, 2, BNXT_PTP_GRC_WIN);
+		if (rc)
+			return rc;
+		for (i = 0; i < 2; i++)
+			ptp->refclk_mapped_regs[i] = BNXT_PTP_GRC_WIN_BASE +
+				(ptp->refclk_regs[i] & BNXT_GRC_OFFSET_MASK);
+		return 0;
+	}
+	if (bp->flags & BNXT_FLAG_CHIP_P7) {
+		for (i = 0; i < 2; i++) {
+			if (reg_arr[i] & BNXT_GRC_BASE_MASK)
+				return -EINVAL;
+			ptp->refclk_mapped_regs[i] = reg_arr[i];
+		}
+		return 0;
+	}
+	return -ENODEV;
+}
+
+static void bnxt_unmap_ptp_regs(struct bnxt *bp)
+{
+	writel(0, bp->bar0 + BNXT_GRCPF_REG_WINDOW_BASE_OUT +
+		  (BNXT_PTP_GRC_WIN - 1) * 4);
+}
+
+static u64 bnxt_cc_read(struct cyclecounter *cc)
+{
+	struct bnxt_ptp_cfg *ptp = container_of(cc, struct bnxt_ptp_cfg, cc);
+	u64 ns = 0;
+
+	__bnxt_refclk_read(ptp->bp, NULL, &ns);
+	return ns;
+}
+
+static int bnxt_stamp_tx_skb(struct bnxt *bp, int slot)
+{
+	struct bnxt_ptp_cfg *ptp = bp->ptp_cfg;
+	struct skb_shared_hwtstamps timestamp;
+	struct bnxt_ptp_tx_req *txts_req;
+	unsigned long now = jiffies;
+	u64 ts = 0, ns = 0;
+	u32 tmo = 0;
+	int rc;
+
+	txts_req = &ptp->txts_req[slot];
+	/* make sure bnxt_get_tx_ts_p5() has updated abs_txts_tmo */
+	smp_rmb();
+	if (!time_after_eq(now, txts_req->abs_txts_tmo))
+		tmo = jiffies_to_msecs(txts_req->abs_txts_tmo - now);
+	rc = bnxt_hwrm_port_ts_query(bp, PORT_TS_QUERY_REQ_FLAGS_PATH_TX, &ts,
+				     tmo, slot);
+	if (!rc) {
+		memset(&timestamp, 0, sizeof(timestamp));
+		ns = bnxt_timecounter_cyc2time(ptp, ts);
+		timestamp.hwtstamp = ns_to_ktime(ns);
+		skb_tstamp_tx(txts_req->tx_skb, &timestamp);
+		ptp->stats.ts_pkts++;
+	} else {
+		if (!time_after_eq(jiffies, txts_req->abs_txts_tmo))
+			return -EAGAIN;
+
+		ptp->stats.ts_lost++;
+		netdev_warn_once(bp->dev,
+				 "TS query for TX timer failed rc = %x\n", rc);
+	}
+
+	dev_kfree_skb_any(txts_req->tx_skb);
+	txts_req->tx_skb = NULL;
+
+	return 0;
+}
+
+static long bnxt_ptp_ts_aux_work(struct ptp_clock_info *ptp_info)
+{
+	struct bnxt_ptp_cfg *ptp = container_of(ptp_info, struct bnxt_ptp_cfg,
+						ptp_info);
+	unsigned long now = jiffies;
+	struct bnxt *bp = ptp->bp;
+	u16 cons = ptp->txts_cons;
+	unsigned long flags;
+	u32 num_requests;
+	int rc = 0;
+
+	num_requests = BNXT_MAX_TX_TS - READ_ONCE(ptp->tx_avail);
+	while (num_requests--) {
+		if (IS_ERR(ptp->txts_req[cons].tx_skb))
+			goto next_slot;
+		if (!ptp->txts_req[cons].tx_skb)
+			break;
+		rc = bnxt_stamp_tx_skb(bp, cons);
+		if (rc == -EAGAIN)
+			break;
+next_slot:
+		BNXT_PTP_INC_TX_AVAIL(ptp);
+		cons = NEXT_TXTS(cons);
+	}
+	ptp->txts_cons = cons;
+
+	if (!time_after_eq(now, ptp->next_period)) {
+		if (rc == -EAGAIN)
+			return 0;
+		return ptp->next_period - now;
+	}
+
+	bnxt_ptp_get_current_time(bp);
+	ptp->next_period = now + HZ;
+	if (time_after_eq(now, ptp->next_overflow_check)) {
+		write_seqlock_irqsave(&ptp->ptp_lock, flags);
+		timecounter_read(&ptp->tc);
+		write_sequnlock_irqrestore(&ptp->ptp_lock, flags);
+		ptp->next_overflow_check = now + BNXT_PHC_OVERFLOW_PERIOD;
+	}
+	if (rc == -EAGAIN)
+		return 0;
+	return HZ;
+}
+
+void bnxt_ptp_free_txts_skbs(struct bnxt_ptp_cfg *ptp)
+{
+	struct bnxt_ptp_tx_req *txts_req;
+	u16 cons = ptp->txts_cons;
+
+	/* make sure ptp aux worker finished with
+	 * possible BNXT_STATE_OPEN set
+	 */
+	ptp_cancel_worker_sync(ptp->ptp_clock);
+
+	ptp->tx_avail = BNXT_MAX_TX_TS;
+	while (cons != ptp->txts_prod) {
+		txts_req = &ptp->txts_req[cons];
+		if (!IS_ERR_OR_NULL(txts_req->tx_skb))
+			dev_kfree_skb_any(txts_req->tx_skb);
+		cons = NEXT_TXTS(cons);
+	}
+	ptp->txts_cons = cons;
+	ptp_schedule_worker(ptp->ptp_clock, 0);
+}
+
+int bnxt_ptp_get_txts_prod(struct bnxt_ptp_cfg *ptp, u16 *prod)
+{
+	spin_lock_bh(&ptp->ptp_tx_lock);
+	if (ptp->tx_avail) {
+		*prod = ptp->txts_prod;
+		ptp->txts_prod = NEXT_TXTS(*prod);
+		ptp->tx_avail--;
+		spin_unlock_bh(&ptp->ptp_tx_lock);
+		return 0;
+	}
+	spin_unlock_bh(&ptp->ptp_tx_lock);
+	atomic64_inc(&ptp->stats.ts_err);
+	return -ENOSPC;
+}
+
+void bnxt_get_tx_ts_p5(struct bnxt *bp, struct sk_buff *skb, u16 prod)
+{
+	struct bnxt_ptp_cfg *ptp = bp->ptp_cfg;
+	struct bnxt_ptp_tx_req *txts_req;
+
+	txts_req = &ptp->txts_req[prod];
+	txts_req->abs_txts_tmo = jiffies + msecs_to_jiffies(ptp->txts_tmo);
+	/* make sure abs_txts_tmo is written first */
+	smp_wmb();
+	txts_req->tx_skb = skb;
+	ptp_schedule_worker(ptp->ptp_clock, 0);
+}
+
+int bnxt_get_rx_ts_p5(struct bnxt *bp, u64 *ts, u32 pkt_ts)
+{
+	struct bnxt_ptp_cfg *ptp = bp->ptp_cfg;
+
+	if (!ptp)
+		return -ENODEV;
+
+	*ts = bnxt_extend_cycles_32b_to_48b(ptp, pkt_ts);
+
+	return 0;
+}
+
+void bnxt_tx_ts_cmp(struct bnxt *bp, struct bnxt_napi *bnapi,
+		    struct tx_ts_cmp *tscmp)
+{
+	struct skb_shared_hwtstamps timestamp = {};
+	struct bnxt_ptp_cfg *ptp = bp->ptp_cfg;
+	u32 opaque = tscmp->tx_ts_cmp_opaque;
+	struct bnxt_tx_ring_info *txr;
+	struct bnxt_sw_tx_bd *tx_buf;
+	u64 ts, ns;
+	u16 cons;
+
+	txr = bnapi->tx_ring[TX_OPAQUE_RING(opaque)];
+	ts = BNXT_GET_TX_TS_48B_NS(tscmp);
+	cons = TX_OPAQUE_IDX(opaque);
+	tx_buf = &txr->tx_buf_ring[RING_TX(bp, cons)];
+	if (tx_buf->is_ts_pkt) {
+		if (BNXT_TX_TS_ERR(tscmp)) {
+			netdev_err(bp->dev,
+				   "timestamp completion error 0x%x 0x%x\n",
+				   le32_to_cpu(tscmp->tx_ts_cmp_flags_type),
+				   le32_to_cpu(tscmp->tx_ts_cmp_errors_v));
+		} else {
+			ns = bnxt_timecounter_cyc2time(ptp, ts);
+			timestamp.hwtstamp = ns_to_ktime(ns);
+			skb_tstamp_tx(tx_buf->skb, &timestamp);
+		}
+		tx_buf->is_ts_pkt = 0;
+	}
+}
+
+static const struct ptp_clock_info bnxt_ptp_caps = {
+	.owner		= THIS_MODULE,
+	.name		= "bnxt clock",
+	.max_adj	= BNXT_MAX_PHC_DRIFT,
+	.n_alarm	= 0,
+	.n_ext_ts	= 0,
+	.n_per_out	= 0,
+	.n_pins		= 0,
+	.pps		= 0,
+	.adjfine	= bnxt_ptp_adjfine,
+	.adjtime	= bnxt_ptp_adjtime,
+	.do_aux_work	= bnxt_ptp_ts_aux_work,
+	.gettimex64	= bnxt_ptp_gettimex,
+	.settime64	= bnxt_ptp_settime,
+	.enable		= bnxt_ptp_enable,
+};
+
+static int bnxt_ptp_verify(struct ptp_clock_info *ptp_info, unsigned int pin,
+			   enum ptp_pin_function func, unsigned int chan)
+{
+	struct bnxt_ptp_cfg *ptp = container_of(ptp_info, struct bnxt_ptp_cfg,
+						ptp_info);
+	/* Allow only PPS pin function configuration */
+	if (ptp->pps_info.pins[pin].usage <= BNXT_PPS_PIN_PPS_OUT &&
+	    func != PTP_PF_PHYSYNC)
+		return 0;
+	else
+		return -EOPNOTSUPP;
+}
+
+static int bnxt_ptp_pps_init(struct bnxt *bp)
+{
+	struct hwrm_func_ptp_pin_qcfg_output *resp;
+	struct hwrm_func_ptp_pin_qcfg_input *req;
+	struct bnxt_ptp_cfg *ptp = bp->ptp_cfg;
+	struct ptp_clock_info *ptp_info;
+	struct bnxt_pps *pps_info;
+	u8 *pin_usg;
+	u32 i, rc;
+
+	/* Query current/default PIN CFG */
+	rc = hwrm_req_init(bp, req, HWRM_FUNC_PTP_PIN_QCFG);
+	if (rc)
+		return rc;
+
+	resp = hwrm_req_hold(bp, req);
+	rc = hwrm_req_send(bp, req);
+	if (rc || !resp->num_pins) {
+		hwrm_req_drop(bp, req);
+		return -EOPNOTSUPP;
+	}
+
+	ptp_info = &ptp->ptp_info;
+	pps_info = &ptp->pps_info;
+	pps_info->num_pins = resp->num_pins;
+	ptp_info->n_pins = pps_info->num_pins;
+	ptp_info->pin_config = kcalloc(ptp_info->n_pins,
+				       sizeof(*ptp_info->pin_config),
+				       GFP_KERNEL);
+	if (!ptp_info->pin_config) {
+		hwrm_req_drop(bp, req);
+		return -ENOMEM;
+	}
+
+	/* Report the TSIO capability to kernel */
+	pin_usg = &resp->pin0_usage;
+	for (i = 0; i < pps_info->num_pins; i++, pin_usg++) {
+		snprintf(ptp_info->pin_config[i].name,
+			 sizeof(ptp_info->pin_config[i].name), "bnxt_pps%d", i);
+		ptp_info->pin_config[i].index = i;
+		ptp_info->pin_config[i].chan = i;
+		if (*pin_usg == BNXT_PPS_PIN_PPS_IN)
+			ptp_info->pin_config[i].func = PTP_PF_EXTTS;
+		else if (*pin_usg == BNXT_PPS_PIN_PPS_OUT)
+			ptp_info->pin_config[i].func = PTP_PF_PEROUT;
+		else
+			ptp_info->pin_config[i].func = PTP_PF_NONE;
+
+		pps_info->pins[i].usage = *pin_usg;
+	}
+	hwrm_req_drop(bp, req);
+
+	/* Only 1 each of ext_ts and per_out pins is available in HW */
+	ptp_info->n_ext_ts = 1;
+	ptp_info->n_per_out = 1;
+	ptp_info->pps = 1;
+	ptp_info->verify = bnxt_ptp_verify;
+
+	return 0;
+}
+
+static bool bnxt_pps_config_ok(struct bnxt *bp)
+{
+	struct bnxt_ptp_cfg *ptp = bp->ptp_cfg;
+
+	return !(bp->fw_cap & BNXT_FW_CAP_PTP_PPS) == !ptp->ptp_info.pin_config;
+}
+
+static void bnxt_ptp_timecounter_init(struct bnxt *bp, bool init_tc)
+{
+	struct bnxt_ptp_cfg *ptp = bp->ptp_cfg;
+	unsigned long flags;
+
+	if (!ptp->ptp_clock) {
+		memset(&ptp->cc, 0, sizeof(ptp->cc));
+		ptp->cc.read = bnxt_cc_read;
+		ptp->cc.mask = CYCLECOUNTER_MASK(48);
+		if (BNXT_MH(bp)) {
+			/* Use timecounter based non-real time mode */
+			ptp->cc.shift = BNXT_CYCLES_SHIFT;
+			ptp->cc.mult = clocksource_khz2mult(BNXT_DEVCLK_FREQ, ptp->cc.shift);
+			ptp->cmult = ptp->cc.mult;
+		} else {
+			ptp->cc.shift = 0;
+			ptp->cc.mult = 1;
+		}
+		ptp->next_overflow_check = jiffies + BNXT_PHC_OVERFLOW_PERIOD;
+	}
+	if (init_tc) {
+		write_seqlock_irqsave(&ptp->ptp_lock, flags);
+		timecounter_init(&ptp->tc, &ptp->cc, ktime_to_ns(ktime_get_real()));
+		write_sequnlock_irqrestore(&ptp->ptp_lock, flags);
+	}
+}
+
+/* Caller holds ptp_lock */
+void bnxt_ptp_rtc_timecounter_init(struct bnxt_ptp_cfg *ptp, u64 ns)
+{
+	timecounter_init(&ptp->tc, &ptp->cc, ns);
+	/* For RTC, cycle_last must be in sync with the timecounter value. */
+	ptp->tc.cycle_last = ns & ptp->cc.mask;
+}
+
+int bnxt_ptp_init_rtc(struct bnxt *bp, bool phc_cfg)
+{
+	struct timespec64 tsp;
+	unsigned long flags;
+	u64 ns;
+	int rc;
+
+	if (!bp->ptp_cfg || !BNXT_PTP_USE_RTC(bp))
+		return -ENODEV;
+
+	if (!phc_cfg) {
+		ktime_get_real_ts64(&tsp);
+		ns = timespec64_to_ns(&tsp);
+		rc = bnxt_ptp_cfg_settime(bp, ns);
+		if (rc)
+			return rc;
+	} else {
+		rc = bnxt_hwrm_port_ts_query(bp, PORT_TS_QUERY_REQ_FLAGS_CURRENT_TIME,
+					     &ns, 0, 0);
+		if (rc)
+			return rc;
+	}
+	write_seqlock_irqsave(&bp->ptp_cfg->ptp_lock, flags);
+	bnxt_ptp_rtc_timecounter_init(bp->ptp_cfg, ns);
+	write_sequnlock_irqrestore(&bp->ptp_cfg->ptp_lock, flags);
+
+	return 0;
+}
+
+static void bnxt_ptp_free(struct bnxt *bp)
+{
+	struct bnxt_ptp_cfg *ptp = bp->ptp_cfg;
+
+	if (ptp->ptp_clock) {
+		ptp_clock_unregister(ptp->ptp_clock);
+		ptp->ptp_clock = NULL;
+		kfree(ptp->ptp_info.pin_config);
+		ptp->ptp_info.pin_config = NULL;
+	}
+}
+
+int bnxt_ptp_init(struct bnxt *bp)
+{
+	struct bnxt_ptp_cfg *ptp = bp->ptp_cfg;
+	int rc;
+
+	if (!ptp)
+		return 0;
+
+	rc = bnxt_map_ptp_regs(bp);
+	if (rc)
+		return rc;
+
+	if (ptp->ptp_clock && bnxt_pps_config_ok(bp))
+		return 0;
+
+	bnxt_ptp_free(bp);
+
+	WRITE_ONCE(ptp->tx_avail, BNXT_MAX_TX_TS);
+	seqlock_init(&ptp->ptp_lock);
+	spin_lock_init(&ptp->ptp_tx_lock);
+
+	if (BNXT_PTP_USE_RTC(bp)) {
+		bnxt_ptp_timecounter_init(bp, false);
+		rc = bnxt_ptp_init_rtc(bp, ptp->rtc_configured);
+		if (rc)
+			goto out;
+	} else {
+		bnxt_ptp_timecounter_init(bp, true);
+		bnxt_ptp_adjfine_rtc(bp, 0);
+	}
+	bnxt_hwrm_func_drv_rgtr(bp, NULL, 0, true);
+
+	ptp->ptp_info = bnxt_ptp_caps;
+	if ((bp->fw_cap & BNXT_FW_CAP_PTP_PPS)) {
+		if (bnxt_ptp_pps_init(bp))
+			netdev_err(bp->dev, "1pps not initialized, continuing without 1pps support\n");
+	}
+	ptp->ptp_clock = ptp_clock_register(&ptp->ptp_info, &bp->pdev->dev);
+	if (IS_ERR(ptp->ptp_clock)) {
+		int err = PTR_ERR(ptp->ptp_clock);
+
+		ptp->ptp_clock = NULL;
+		rc = err;
+		goto out;
+	}
+
+	ptp->stats.ts_pkts = 0;
+	ptp->stats.ts_lost = 0;
+	atomic64_set(&ptp->stats.ts_err, 0);
+
+	if (bp->flags & BNXT_FLAG_CHIP_P5_PLUS) {
+		bnxt_refclk_read(bp, NULL, &ptp->current_time);
+		WRITE_ONCE(ptp->old_time, ptp->current_time >> BNXT_HI_TIMER_SHIFT);
+		ptp_schedule_worker(ptp->ptp_clock, 0);
+	}
+	ptp->txts_tmo = BNXT_PTP_DFLT_TX_TMO;
+	return 0;
+
+out:
+	bnxt_ptp_free(bp);
+	bnxt_unmap_ptp_regs(bp);
+	return rc;
+}
+
+void bnxt_ptp_clear(struct bnxt *bp)
+{
+	struct bnxt_ptp_cfg *ptp = bp->ptp_cfg;
+
+	if (!ptp)
+		return;
+
+	if (ptp->ptp_clock)
+		ptp_clock_unregister(ptp->ptp_clock);
+
+	ptp->ptp_clock = NULL;
+	kfree(ptp->ptp_info.pin_config);
+	ptp->ptp_info.pin_config = NULL;
+
+	bnxt_unmap_ptp_regs(bp);
+}
diff --git a/drivers/net/ethernet/broadcom/bnxt/bnxt_ptp.h b/drivers/net/ethernet/broadcom/bnxt/bnxt_ptp.h
new file mode 100644
index 000000000000..8cc2fae47644
--- /dev/null
+++ b/drivers/net/ethernet/broadcom/bnxt/bnxt_ptp.h
@@ -0,0 +1,201 @@
+/* Broadcom NetXtreme-C/E network driver.
+ *
+ * Copyright (c) 2021 Broadcom Inc.
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation.
+ */
+
+#ifndef BNXT_PTP_H
+#define BNXT_PTP_H
+
+#include <linux/ptp_clock_kernel.h>
+#include <linux/timecounter.h>
+
+#define BNXT_PTP_GRC_WIN	6
+#define BNXT_PTP_GRC_WIN_BASE	0x6000
+
+#define BNXT_MAX_PHC_DRIFT	31000000
+#define BNXT_CYCLES_SHIFT	23
+#define BNXT_DEVCLK_FREQ	1000000
+#define BNXT_LO_TIMER_MASK	0x0000ffffffffUL
+#define BNXT_HI_TIMER_MASK	0xffff00000000UL
+#define BNXT_HI_TIMER_SHIFT	24
+
+#define BNXT_PTP_DFLT_TX_TMO	1000 /* ms */
+#define BNXT_PTP_QTS_TIMEOUT	1000
+#define BNXT_PTP_QTS_MAX_TMO_US	65535U
+#define BNXT_PTP_QTS_TX_ENABLES	(PORT_TS_QUERY_REQ_ENABLES_PTP_SEQ_ID |	\
+				 PORT_TS_QUERY_REQ_ENABLES_TS_REQ_TIMEOUT | \
+				 PORT_TS_QUERY_REQ_ENABLES_PTP_HDR_OFFSET)
+
+struct pps_pin {
+	u8 event;
+	u8 usage;
+	u8 state;
+};
+
+#define TSIO_PIN_VALID(pin) ((pin) >= 0 && (pin) < (BNXT_MAX_TSIO_PINS))
+
+#define EVENT_DATA2_PPS_EVENT_TYPE(data2)				\
+	((data2) & ASYNC_EVENT_CMPL_PPS_TIMESTAMP_EVENT_DATA2_EVENT_TYPE)
+
+#define EVENT_DATA2_PPS_PIN_NUM(data2)					\
+	(((data2) &							\
+	  ASYNC_EVENT_CMPL_PPS_TIMESTAMP_EVENT_DATA2_PIN_NUMBER_MASK) >>\
+	 ASYNC_EVENT_CMPL_PPS_TIMESTAMP_EVENT_DATA2_PIN_NUMBER_SFT)
+
+#define BNXT_DATA2_UPPER_MSK						\
+	ASYNC_EVENT_CMPL_PPS_TIMESTAMP_EVENT_DATA2_PPS_TIMESTAMP_UPPER_MASK
+
+#define BNXT_DATA2_UPPER_SFT						\
+	(32 -								\
+	 ASYNC_EVENT_CMPL_PPS_TIMESTAMP_EVENT_DATA2_PPS_TIMESTAMP_UPPER_SFT)
+
+#define BNXT_DATA1_LOWER_MSK						\
+	ASYNC_EVENT_CMPL_PPS_TIMESTAMP_EVENT_DATA1_PPS_TIMESTAMP_LOWER_MASK
+
+#define BNXT_DATA1_LOWER_SFT						\
+	  ASYNC_EVENT_CMPL_PPS_TIMESTAMP_EVENT_DATA1_PPS_TIMESTAMP_LOWER_SFT
+
+#define EVENT_PPS_TS(data2, data1)					\
+	(((u64)((data2) & BNXT_DATA2_UPPER_MSK) << BNXT_DATA2_UPPER_SFT) |\
+	 (((data1) & BNXT_DATA1_LOWER_MSK) >> BNXT_DATA1_LOWER_SFT))
+
+#define BNXT_PPS_PIN_DISABLE	0
+#define BNXT_PPS_PIN_ENABLE	1
+#define BNXT_PPS_PIN_NONE	0
+#define BNXT_PPS_PIN_PPS_IN	1
+#define BNXT_PPS_PIN_PPS_OUT	2
+#define BNXT_PPS_PIN_SYNC_IN	3
+#define BNXT_PPS_PIN_SYNC_OUT	4
+
+#define BNXT_PPS_EVENT_INTERNAL	1
+#define BNXT_PPS_EVENT_EXTERNAL	2
+
+struct bnxt_pps {
+	u8 num_pins;
+#define BNXT_MAX_TSIO_PINS	4
+	struct pps_pin pins[BNXT_MAX_TSIO_PINS];
+};
+
+struct bnxt_ptp_stats {
+	u64		ts_pkts;
+	u64		ts_lost;
+	atomic64_t	ts_err;
+};
+
+#define BNXT_MAX_TX_TS		4
+#define NEXT_TXTS(idx)		(((idx) + 1) & (BNXT_MAX_TX_TS - 1))
+
+struct bnxt_ptp_tx_req {
+	struct sk_buff		*tx_skb;
+	u16			tx_seqid;
+	u16			tx_hdr_off;
+	unsigned long		abs_txts_tmo;
+};
+
+struct bnxt_ptp_cfg {
+	struct ptp_clock_info	ptp_info;
+	struct ptp_clock	*ptp_clock;
+	struct cyclecounter	cc;
+	struct timecounter	tc;
+	struct bnxt_pps		pps_info;
+	/* serialize timecounter access */
+	seqlock_t		ptp_lock;
+	/* serialize ts tx request queuing */
+	spinlock_t		ptp_tx_lock;
+	u64			current_time;
+	unsigned long		next_period;
+	unsigned long		next_overflow_check;
+	u32			cmult;
+	/* cache of upper 24 bits of cyclecoutner. 8 bits are used to check for roll-over */
+	u32			old_time;
+	/* a 23b shift cyclecounter will overflow in ~36 mins.  Check overflow every 18 mins. */
+	#define BNXT_PHC_OVERFLOW_PERIOD	(18 * 60 * HZ)
+
+	struct bnxt_ptp_tx_req	txts_req[BNXT_MAX_TX_TS];
+
+	struct bnxt		*bp;
+	u32			tx_avail;
+	u16			rxctl;
+#define BNXT_PTP_MSG_SYNC			(1 << 0)
+#define BNXT_PTP_MSG_DELAY_REQ			(1 << 1)
+#define BNXT_PTP_MSG_PDELAY_REQ			(1 << 2)
+#define BNXT_PTP_MSG_PDELAY_RESP		(1 << 3)
+#define BNXT_PTP_MSG_FOLLOW_UP			(1 << 8)
+#define BNXT_PTP_MSG_DELAY_RESP			(1 << 9)
+#define BNXT_PTP_MSG_PDELAY_RESP_FOLLOW_UP	(1 << 10)
+#define BNXT_PTP_MSG_ANNOUNCE			(1 << 11)
+#define BNXT_PTP_MSG_SIGNALING			(1 << 12)
+#define BNXT_PTP_MSG_MANAGEMENT			(1 << 13)
+#define BNXT_PTP_MSG_EVENTS		(BNXT_PTP_MSG_SYNC |		\
+					 BNXT_PTP_MSG_DELAY_REQ |	\
+					 BNXT_PTP_MSG_PDELAY_REQ |	\
+					 BNXT_PTP_MSG_PDELAY_RESP)
+	u8			tx_tstamp_en:1;
+	u8			rtc_configured:1;
+	int			rx_filter;
+	u32			tstamp_filters;
+
+	u32			refclk_regs[2];
+	u32			refclk_mapped_regs[2];
+	u32			txts_tmo;
+	u16			txts_prod;
+	u16			txts_cons;
+
+	struct bnxt_ptp_stats	stats;
+};
+
+#define BNXT_PTP_INC_TX_AVAIL(ptp)		\
+do {						\
+	spin_lock_bh(&(ptp)->ptp_tx_lock);	\
+	(ptp)->tx_avail++;			\
+	spin_unlock_bh(&(ptp)->ptp_tx_lock);	\
+} while (0)
+
+int bnxt_ptp_parse(struct sk_buff *skb, u16 *seq_id, u16 *hdr_off);
+void bnxt_ptp_update_current_time(struct bnxt *bp);
+void bnxt_ptp_pps_event(struct bnxt *bp, u32 data1, u32 data2);
+int bnxt_ptp_cfg_tstamp_filters(struct bnxt *bp);
+void bnxt_ptp_reapply_pps(struct bnxt *bp);
+int bnxt_hwtstamp_set(struct net_device *dev,
+		      struct kernel_hwtstamp_config *stmpconf,
+		      struct netlink_ext_ack *extack);
+int bnxt_hwtstamp_get(struct net_device *dev,
+		      struct kernel_hwtstamp_config *stmpconf);
+void bnxt_ptp_free_txts_skbs(struct bnxt_ptp_cfg *ptp);
+int bnxt_ptp_get_txts_prod(struct bnxt_ptp_cfg *ptp, u16 *prod);
+void bnxt_get_tx_ts_p5(struct bnxt *bp, struct sk_buff *skb, u16 prod);
+int bnxt_get_rx_ts_p5(struct bnxt *bp, u64 *ts, u32 pkt_ts);
+void bnxt_tx_ts_cmp(struct bnxt *bp, struct bnxt_napi *bnapi,
+		    struct tx_ts_cmp *tscmp);
+void bnxt_ptp_rtc_timecounter_init(struct bnxt_ptp_cfg *ptp, u64 ns);
+int bnxt_ptp_init_rtc(struct bnxt *bp, bool phc_cfg);
+int bnxt_ptp_init(struct bnxt *bp);
+void bnxt_ptp_clear(struct bnxt *bp);
+static inline u64 bnxt_timecounter_cyc2time(struct bnxt_ptp_cfg *ptp, u64 ts)
+{
+	unsigned int seq;
+	u64 ns;
+
+	do {
+		seq = read_seqbegin(&ptp->ptp_lock);
+		ns = timecounter_cyc2time(&ptp->tc, ts);
+	} while (read_seqretry(&ptp->ptp_lock, seq));
+
+	return ns;
+}
+
+static inline u64 bnxt_extend_cycles_32b_to_48b(struct bnxt_ptp_cfg *ptp, u32 ts)
+{
+	u64 time, cycles;
+
+	time = (u64)READ_ONCE(ptp->old_time) << BNXT_HI_TIMER_SHIFT;
+	cycles = (time & BNXT_HI_TIMER_MASK) | ts;
+	if (ts < (time & BNXT_LO_TIMER_MASK))
+		cycles += BNXT_LO_TIMER_MASK + 1;
+	return cycles;
+}
+#endif
diff --git a/drivers/net/ethernet/broadcom/bnxt/bnxt_sriov.c b/drivers/net/ethernet/broadcom/bnxt/bnxt_sriov.c
index b8e7248294d9..80fed2c07b9e 100644
--- a/drivers/net/ethernet/broadcom/bnxt/bnxt_sriov.c
+++ b/drivers/net/ethernet/broadcom/bnxt/bnxt_sriov.c
@@ -1,75 +1,66 @@
 /* Broadcom NetXtreme-C/E network driver.
  *
  * Copyright (c) 2014-2016 Broadcom Corporation
- * Copyright (c) 2016-2017 Broadcom Limited
+ * Copyright (c) 2016-2018 Broadcom Limited
  *
  * This program is free software; you can redistribute it and/or modify
  * it under the terms of the GNU General Public License as published by
  * the Free Software Foundation.
  */
 
+#include <linux/ethtool.h>
 #include <linux/module.h>
 #include <linux/pci.h>
 #include <linux/netdevice.h>
 #include <linux/if_vlan.h>
 #include <linux/interrupt.h>
 #include <linux/etherdevice.h>
-#include "bnxt_hsi.h"
+#include <net/dcbnl.h>
+#include <linux/bnxt/hsi.h>
 #include "bnxt.h"
+#include "bnxt_hwrm.h"
 #include "bnxt_ulp.h"
 #include "bnxt_sriov.h"
+#include "bnxt_vfr.h"
 #include "bnxt_ethtool.h"
 
 #ifdef CONFIG_BNXT_SRIOV
 static int bnxt_hwrm_fwd_async_event_cmpl(struct bnxt *bp,
 					  struct bnxt_vf_info *vf, u16 event_id)
 {
-	struct hwrm_fwd_async_event_cmpl_output *resp = bp->hwrm_cmd_resp_addr;
-	struct hwrm_fwd_async_event_cmpl_input req = {0};
+	struct hwrm_fwd_async_event_cmpl_input *req;
 	struct hwrm_async_event_cmpl *async_cmpl;
 	int rc = 0;
 
-	bnxt_hwrm_cmd_hdr_init(bp, &req, HWRM_FWD_ASYNC_EVENT_CMPL, -1, -1);
+	rc = hwrm_req_init(bp, req, HWRM_FWD_ASYNC_EVENT_CMPL);
+	if (rc)
+		goto exit;
+
 	if (vf)
-		req.encap_async_event_target_id = cpu_to_le16(vf->fw_fid);
+		req->encap_async_event_target_id = cpu_to_le16(vf->fw_fid);
 	else
 		/* broadcast this async event to all VFs */
-		req.encap_async_event_target_id = cpu_to_le16(0xffff);
-	async_cmpl = (struct hwrm_async_event_cmpl *)req.encap_async_event_cmpl;
+		req->encap_async_event_target_id = cpu_to_le16(0xffff);
+	async_cmpl =
+		(struct hwrm_async_event_cmpl *)req->encap_async_event_cmpl;
 	async_cmpl->type = cpu_to_le16(ASYNC_EVENT_CMPL_TYPE_HWRM_ASYNC_EVENT);
 	async_cmpl->event_id = cpu_to_le16(event_id);
 
-	mutex_lock(&bp->hwrm_cmd_lock);
-	rc = _hwrm_send_message(bp, &req, sizeof(req), HWRM_CMD_TIMEOUT);
-
-	if (rc) {
+	rc = hwrm_req_send(bp, req);
+exit:
+	if (rc)
 		netdev_err(bp->dev, "hwrm_fwd_async_event_cmpl failed. rc:%d\n",
 			   rc);
-		goto fwd_async_event_cmpl_exit;
-	}
-
-	if (resp->error_code) {
-		netdev_err(bp->dev, "hwrm_fwd_async_event_cmpl error %d\n",
-			   resp->error_code);
-		rc = -1;
-	}
-
-fwd_async_event_cmpl_exit:
-	mutex_unlock(&bp->hwrm_cmd_lock);
 	return rc;
 }
 
 static int bnxt_vf_ndo_prep(struct bnxt *bp, int vf_id)
 {
-	if (!test_bit(BNXT_STATE_OPEN, &bp->state)) {
-		netdev_err(bp->dev, "vf ndo called though PF is down\n");
-		return -EINVAL;
-	}
 	if (!bp->pf.active_vfs) {
 		netdev_err(bp->dev, "vf ndo called though sriov is disabled\n");
 		return -EINVAL;
 	}
-	if (vf_id >= bp->pf.max_vfs) {
+	if (vf_id >= bp->pf.active_vfs) {
 		netdev_err(bp->dev, "Invalid VF id %d\n", vf_id);
 		return -EINVAL;
 	}
@@ -78,10 +69,10 @@ static int bnxt_vf_ndo_prep(struct bnxt *bp, int vf_id)
 
 int bnxt_set_vf_spoofchk(struct net_device *dev, int vf_id, bool setting)
 {
-	struct hwrm_func_cfg_input req = {0};
 	struct bnxt *bp = netdev_priv(dev);
-	struct bnxt_vf_info *vf;
+	struct hwrm_func_cfg_input *req;
 	bool old_setting = false;
+	struct bnxt_vf_info *vf;
 	u32 func_flags;
 	int rc;
 
@@ -98,28 +89,94 @@ int bnxt_set_vf_spoofchk(struct net_device *dev, int vf_id, bool setting)
 	if (old_setting == setting)
 		return 0;
 
-	func_flags = vf->func_flags;
 	if (setting)
-		func_flags |= FUNC_CFG_REQ_FLAGS_SRC_MAC_ADDR_CHECK_ENABLE;
+		func_flags = FUNC_CFG_REQ_FLAGS_SRC_MAC_ADDR_CHECK_ENABLE;
 	else
-		func_flags |= FUNC_CFG_REQ_FLAGS_SRC_MAC_ADDR_CHECK_DISABLE;
+		func_flags = FUNC_CFG_REQ_FLAGS_SRC_MAC_ADDR_CHECK_DISABLE;
 	/*TODO: if the driver supports VLAN filter on guest VLAN,
 	 * the spoof check should also include vlan anti-spoofing
 	 */
-	bnxt_hwrm_cmd_hdr_init(bp, &req, HWRM_FUNC_CFG, -1, -1);
-	req.fid = cpu_to_le16(vf->fw_fid);
-	req.flags = cpu_to_le32(func_flags);
-	rc = hwrm_send_message(bp, &req, sizeof(req), HWRM_CMD_TIMEOUT);
+	rc = bnxt_hwrm_func_cfg_short_req_init(bp, &req);
 	if (!rc) {
-		vf->func_flags = func_flags;
-		if (setting)
-			vf->flags |= BNXT_VF_SPOOFCHK;
-		else
-			vf->flags &= ~BNXT_VF_SPOOFCHK;
+		req->fid = cpu_to_le16(vf->fw_fid);
+		req->flags = cpu_to_le32(func_flags);
+		rc = hwrm_req_send(bp, req);
+		if (!rc) {
+			if (setting)
+				vf->flags |= BNXT_VF_SPOOFCHK;
+			else
+				vf->flags &= ~BNXT_VF_SPOOFCHK;
+		}
 	}
 	return rc;
 }
 
+static int bnxt_hwrm_func_qcfg_flags(struct bnxt *bp, struct bnxt_vf_info *vf)
+{
+	struct hwrm_func_qcfg_output *resp;
+	struct hwrm_func_qcfg_input *req;
+	int rc;
+
+	rc = hwrm_req_init(bp, req, HWRM_FUNC_QCFG);
+	if (rc)
+		return rc;
+
+	req->fid = cpu_to_le16(BNXT_PF(bp) ? vf->fw_fid : 0xffff);
+	resp = hwrm_req_hold(bp, req);
+	rc = hwrm_req_send(bp, req);
+	if (!rc)
+		vf->func_qcfg_flags = le16_to_cpu(resp->flags);
+	hwrm_req_drop(bp, req);
+	return rc;
+}
+
+bool bnxt_is_trusted_vf(struct bnxt *bp, struct bnxt_vf_info *vf)
+{
+	if (BNXT_PF(bp) && !(bp->fw_cap & BNXT_FW_CAP_TRUSTED_VF))
+		return !!(vf->flags & BNXT_VF_TRUST);
+
+	bnxt_hwrm_func_qcfg_flags(bp, vf);
+	return !!(vf->func_qcfg_flags & FUNC_QCFG_RESP_FLAGS_TRUSTED_VF);
+}
+
+static int bnxt_hwrm_set_trusted_vf(struct bnxt *bp, struct bnxt_vf_info *vf)
+{
+	struct hwrm_func_cfg_input *req;
+	int rc;
+
+	if (!(bp->fw_cap & BNXT_FW_CAP_TRUSTED_VF))
+		return 0;
+
+	rc = bnxt_hwrm_func_cfg_short_req_init(bp, &req);
+	if (rc)
+		return rc;
+
+	req->fid = cpu_to_le16(vf->fw_fid);
+	if (vf->flags & BNXT_VF_TRUST)
+		req->flags = cpu_to_le32(FUNC_CFG_REQ_FLAGS_TRUSTED_VF_ENABLE);
+	else
+		req->flags = cpu_to_le32(FUNC_CFG_REQ_FLAGS_TRUSTED_VF_DISABLE);
+	return hwrm_req_send(bp, req);
+}
+
+int bnxt_set_vf_trust(struct net_device *dev, int vf_id, bool trusted)
+{
+	struct bnxt *bp = netdev_priv(dev);
+	struct bnxt_vf_info *vf;
+
+	if (bnxt_vf_ndo_prep(bp, vf_id))
+		return -EINVAL;
+
+	vf = &bp->pf.vf[vf_id];
+	if (trusted)
+		vf->flags |= BNXT_VF_TRUST;
+	else
+		vf->flags &= ~BNXT_VF_TRUST;
+
+	bnxt_hwrm_set_trusted_vf(bp, vf);
+	return 0;
+}
+
 int bnxt_get_vf_config(struct net_device *dev, int vf_id,
 		       struct ifla_vf_info *ivi)
 {
@@ -134,15 +191,16 @@ int bnxt_get_vf_config(struct net_device *dev, int vf_id,
 	ivi->vf = vf_id;
 	vf = &bp->pf.vf[vf_id];
 
-	memcpy(&ivi->mac, vf->mac_addr, ETH_ALEN);
+	if (is_valid_ether_addr(vf->mac_addr))
+		memcpy(&ivi->mac, vf->mac_addr, ETH_ALEN);
+	else
+		memcpy(&ivi->mac, vf->vf_mac_addr, ETH_ALEN);
 	ivi->max_tx_rate = vf->max_tx_rate;
 	ivi->min_tx_rate = vf->min_tx_rate;
-	ivi->vlan = vf->vlan;
-	if (vf->flags & BNXT_VF_QOS)
-		ivi->qos = vf->vlan >> VLAN_PRIO_SHIFT;
-	else
-		ivi->qos = 0;
+	ivi->vlan = vf->vlan & VLAN_VID_MASK;
+	ivi->qos = vf->vlan >> VLAN_PRIO_SHIFT;
 	ivi->spoofchk = !!(vf->flags & BNXT_VF_SPOOFCHK);
+	ivi->trusted = bnxt_is_trusted_vf(bp, vf);
 	if (!(vf->flags & BNXT_VF_LINK_FORCED))
 		ivi->linkstate = IFLA_VF_LINK_STATE_AUTO;
 	else if (vf->flags & BNXT_VF_LINK_UP)
@@ -155,8 +213,8 @@ int bnxt_get_vf_config(struct net_device *dev, int vf_id,
 
 int bnxt_set_vf_mac(struct net_device *dev, int vf_id, u8 *mac)
 {
-	struct hwrm_func_cfg_input req = {0};
 	struct bnxt *bp = netdev_priv(dev);
+	struct hwrm_func_cfg_input *req;
 	struct bnxt_vf_info *vf;
 	int rc;
 
@@ -172,20 +230,23 @@ int bnxt_set_vf_mac(struct net_device *dev, int vf_id, u8 *mac)
 	}
 	vf = &bp->pf.vf[vf_id];
 
+	rc = bnxt_hwrm_func_cfg_short_req_init(bp, &req);
+	if (rc)
+		return rc;
+
 	memcpy(vf->mac_addr, mac, ETH_ALEN);
-	bnxt_hwrm_cmd_hdr_init(bp, &req, HWRM_FUNC_CFG, -1, -1);
-	req.fid = cpu_to_le16(vf->fw_fid);
-	req.flags = cpu_to_le32(vf->func_flags);
-	req.enables = cpu_to_le32(FUNC_CFG_REQ_ENABLES_DFLT_MAC_ADDR);
-	memcpy(req.dflt_mac_addr, mac, ETH_ALEN);
-	return hwrm_send_message(bp, &req, sizeof(req), HWRM_CMD_TIMEOUT);
+
+	req->fid = cpu_to_le16(vf->fw_fid);
+	req->enables = cpu_to_le32(FUNC_CFG_REQ_ENABLES_DFLT_MAC_ADDR);
+	memcpy(req->dflt_mac_addr, mac, ETH_ALEN);
+	return hwrm_req_send(bp, req);
 }
 
 int bnxt_set_vf_vlan(struct net_device *dev, int vf_id, u16 vlan_id, u8 qos,
 		     __be16 vlan_proto)
 {
-	struct hwrm_func_cfg_input req = {0};
 	struct bnxt *bp = netdev_priv(dev);
+	struct hwrm_func_cfg_input *req;
 	struct bnxt_vf_info *vf;
 	u16 vlan_tag;
 	int rc;
@@ -193,40 +254,45 @@ int bnxt_set_vf_vlan(struct net_device *dev, int vf_id, u16 vlan_id, u8 qos,
 	if (bp->hwrm_spec_code < 0x10201)
 		return -ENOTSUPP;
 
-	if (vlan_proto != htons(ETH_P_8021Q))
+	if (vlan_proto != htons(ETH_P_8021Q) &&
+	    (vlan_proto != htons(ETH_P_8021AD) ||
+	     !(bp->fw_cap & BNXT_FW_CAP_DFLT_VLAN_TPID_PCP)))
 		return -EPROTONOSUPPORT;
 
 	rc = bnxt_vf_ndo_prep(bp, vf_id);
 	if (rc)
 		return rc;
 
-	/* TODO: needed to implement proper handling of user priority,
-	 * currently fail the command if there is valid priority
-	 */
-	if (vlan_id > 4095 || qos)
+	if (vlan_id >= VLAN_N_VID || qos >= IEEE_8021Q_MAX_PRIORITIES ||
+	    (!vlan_id && qos))
 		return -EINVAL;
 
 	vf = &bp->pf.vf[vf_id];
-	vlan_tag = vlan_id;
+	vlan_tag = vlan_id | (u16)qos << VLAN_PRIO_SHIFT;
 	if (vlan_tag == vf->vlan)
 		return 0;
 
-	bnxt_hwrm_cmd_hdr_init(bp, &req, HWRM_FUNC_CFG, -1, -1);
-	req.fid = cpu_to_le16(vf->fw_fid);
-	req.flags = cpu_to_le32(vf->func_flags);
-	req.dflt_vlan = cpu_to_le16(vlan_tag);
-	req.enables = cpu_to_le32(FUNC_CFG_REQ_ENABLES_DFLT_VLAN);
-	rc = hwrm_send_message(bp, &req, sizeof(req), HWRM_CMD_TIMEOUT);
-	if (!rc)
-		vf->vlan = vlan_tag;
+	rc = bnxt_hwrm_func_cfg_short_req_init(bp, &req);
+	if (!rc) {
+		req->fid = cpu_to_le16(vf->fw_fid);
+		req->dflt_vlan = cpu_to_le16(vlan_tag);
+		req->enables = cpu_to_le32(FUNC_CFG_REQ_ENABLES_DFLT_VLAN);
+		if (bp->fw_cap & BNXT_FW_CAP_DFLT_VLAN_TPID_PCP) {
+			req->enables |= cpu_to_le32(FUNC_CFG_REQ_ENABLES_TPID);
+			req->tpid = vlan_proto;
+		}
+		rc = hwrm_req_send(bp, req);
+		if (!rc)
+			vf->vlan = vlan_tag;
+	}
 	return rc;
 }
 
 int bnxt_set_vf_bw(struct net_device *dev, int vf_id, int min_tx_rate,
 		   int max_tx_rate)
 {
-	struct hwrm_func_cfg_input req = {0};
 	struct bnxt *bp = netdev_priv(dev);
+	struct hwrm_func_cfg_input *req;
 	struct bnxt_vf_info *vf;
 	u32 pf_link_speed;
 	int rc;
@@ -243,24 +309,25 @@ int bnxt_set_vf_bw(struct net_device *dev, int vf_id, int min_tx_rate,
 		return -EINVAL;
 	}
 
-	if (min_tx_rate > pf_link_speed || min_tx_rate > max_tx_rate) {
+	if (min_tx_rate > pf_link_speed) {
 		netdev_info(bp->dev, "min tx rate %d is invalid for VF %d\n",
 			    min_tx_rate, vf_id);
 		return -EINVAL;
 	}
 	if (min_tx_rate == vf->min_tx_rate && max_tx_rate == vf->max_tx_rate)
 		return 0;
-	bnxt_hwrm_cmd_hdr_init(bp, &req, HWRM_FUNC_CFG, -1, -1);
-	req.fid = cpu_to_le16(vf->fw_fid);
-	req.flags = cpu_to_le32(vf->func_flags);
-	req.enables = cpu_to_le32(FUNC_CFG_REQ_ENABLES_MAX_BW);
-	req.max_bw = cpu_to_le32(max_tx_rate);
-	req.enables |= cpu_to_le32(FUNC_CFG_REQ_ENABLES_MIN_BW);
-	req.min_bw = cpu_to_le32(min_tx_rate);
-	rc = hwrm_send_message(bp, &req, sizeof(req), HWRM_CMD_TIMEOUT);
+	rc = bnxt_hwrm_func_cfg_short_req_init(bp, &req);
 	if (!rc) {
-		vf->min_tx_rate = min_tx_rate;
-		vf->max_tx_rate = max_tx_rate;
+		req->fid = cpu_to_le16(vf->fw_fid);
+		req->enables = cpu_to_le32(FUNC_CFG_REQ_ENABLES_MAX_BW |
+					   FUNC_CFG_REQ_ENABLES_MIN_BW);
+		req->max_bw = cpu_to_le32(max_tx_rate);
+		req->min_bw = cpu_to_le32(min_tx_rate);
+		rc = hwrm_req_send(bp, req);
+		if (!rc) {
+			vf->min_tx_rate = min_tx_rate;
+			vf->max_tx_rate = max_tx_rate;
+		}
 	}
 	return rc;
 }
@@ -313,21 +380,22 @@ static int bnxt_set_vf_attr(struct bnxt *bp, int num_vfs)
 
 static int bnxt_hwrm_func_vf_resource_free(struct bnxt *bp, int num_vfs)
 {
-	int i, rc = 0;
+	struct hwrm_func_vf_resc_free_input *req;
 	struct bnxt_pf_info *pf = &bp->pf;
-	struct hwrm_func_vf_resc_free_input req = {0};
+	int i, rc;
 
-	bnxt_hwrm_cmd_hdr_init(bp, &req, HWRM_FUNC_VF_RESC_FREE, -1, -1);
+	rc = hwrm_req_init(bp, req, HWRM_FUNC_VF_RESC_FREE);
+	if (rc)
+		return rc;
 
-	mutex_lock(&bp->hwrm_cmd_lock);
+	hwrm_req_hold(bp, req);
 	for (i = pf->first_vf_id; i < pf->first_vf_id + num_vfs; i++) {
-		req.vf_id = cpu_to_le16(i);
-		rc = _hwrm_send_message(bp, &req, sizeof(req),
-					HWRM_CMD_TIMEOUT);
+		req->vf_id = cpu_to_le16(i);
+		rc = hwrm_req_send(bp, req);
 		if (rc)
 			break;
 	}
-	mutex_unlock(&bp->hwrm_cmd_lock);
+	hwrm_req_drop(bp, req);
 	return rc;
 }
 
@@ -348,6 +416,7 @@ static void bnxt_free_vf_resources(struct bnxt *bp)
 		}
 	}
 
+	bp->pf.active_vfs = 0;
 	kfree(bp->pf.vf);
 	bp->pf.vf = NULL;
 }
@@ -400,135 +469,401 @@ static int bnxt_alloc_vf_resources(struct bnxt *bp, int num_vfs)
 
 static int bnxt_hwrm_func_buf_rgtr(struct bnxt *bp)
 {
-	struct hwrm_func_buf_rgtr_input req = {0};
+	struct hwrm_func_buf_rgtr_input *req;
+	int rc;
+
+	rc = hwrm_req_init(bp, req, HWRM_FUNC_BUF_RGTR);
+	if (rc)
+		return rc;
+
+	req->req_buf_num_pages = cpu_to_le16(bp->pf.hwrm_cmd_req_pages);
+	req->req_buf_page_size = cpu_to_le16(BNXT_PAGE_SHIFT);
+	req->req_buf_len = cpu_to_le16(BNXT_HWRM_REQ_MAX_SIZE);
+	req->req_buf_page_addr0 = cpu_to_le64(bp->pf.hwrm_cmd_req_dma_addr[0]);
+	req->req_buf_page_addr1 = cpu_to_le64(bp->pf.hwrm_cmd_req_dma_addr[1]);
+	req->req_buf_page_addr2 = cpu_to_le64(bp->pf.hwrm_cmd_req_dma_addr[2]);
+	req->req_buf_page_addr3 = cpu_to_le64(bp->pf.hwrm_cmd_req_dma_addr[3]);
+
+	return hwrm_req_send(bp, req);
+}
+
+static int __bnxt_set_vf_params(struct bnxt *bp, int vf_id)
+{
+	struct hwrm_func_cfg_input *req;
+	struct bnxt_vf_info *vf;
+	int rc;
+
+	rc = bnxt_hwrm_func_cfg_short_req_init(bp, &req);
+	if (rc)
+		return rc;
+
+	vf = &bp->pf.vf[vf_id];
+	req->fid = cpu_to_le16(vf->fw_fid);
+
+	if (is_valid_ether_addr(vf->mac_addr)) {
+		req->enables |= cpu_to_le32(FUNC_CFG_REQ_ENABLES_DFLT_MAC_ADDR);
+		memcpy(req->dflt_mac_addr, vf->mac_addr, ETH_ALEN);
+	}
+	if (vf->vlan) {
+		req->enables |= cpu_to_le32(FUNC_CFG_REQ_ENABLES_DFLT_VLAN);
+		req->dflt_vlan = cpu_to_le16(vf->vlan);
+	}
+	if (vf->max_tx_rate) {
+		req->enables |= cpu_to_le32(FUNC_CFG_REQ_ENABLES_MAX_BW |
+					    FUNC_CFG_REQ_ENABLES_MIN_BW);
+		req->max_bw = cpu_to_le32(vf->max_tx_rate);
+		req->min_bw = cpu_to_le32(vf->min_tx_rate);
+	}
+	if (vf->flags & BNXT_VF_TRUST)
+		req->flags |= cpu_to_le32(FUNC_CFG_REQ_FLAGS_TRUSTED_VF_ENABLE);
+
+	return hwrm_req_send(bp, req);
+}
+
+static void bnxt_hwrm_roce_sriov_cfg(struct bnxt *bp, int num_vfs)
+{
+	struct hwrm_func_qcaps_output *resp;
+	struct hwrm_func_cfg_input *cfg_req;
+	struct hwrm_func_qcaps_input *req;
+	int rc;
+
+	rc = hwrm_req_init(bp, req, HWRM_FUNC_QCAPS);
+	if (rc)
+		return;
+
+	req->fid = cpu_to_le16(0xffff);
+	resp = hwrm_req_hold(bp, req);
+	rc = hwrm_req_send(bp, req);
+	if (rc)
+		goto err;
+
+	rc = hwrm_req_init(bp, cfg_req, HWRM_FUNC_CFG);
+	if (rc)
+		goto err;
+
+	/* In case of VF Dynamic resource allocation, driver will provision
+	 * maximum resources to all the VFs. FW will dynamically allocate
+	 * resources to VFs on the fly, so always divide the resources by 1.
+	 */
+	if (BNXT_ROCE_VF_DYN_ALLOC_CAP(bp))
+		num_vfs = 1;
+
+	cfg_req->fid = cpu_to_le16(0xffff);
+	cfg_req->enables2 =
+		cpu_to_le32(FUNC_CFG_REQ_ENABLES2_ROCE_MAX_AV_PER_VF |
+			    FUNC_CFG_REQ_ENABLES2_ROCE_MAX_CQ_PER_VF |
+			    FUNC_CFG_REQ_ENABLES2_ROCE_MAX_MRW_PER_VF |
+			    FUNC_CFG_REQ_ENABLES2_ROCE_MAX_QP_PER_VF |
+			    FUNC_CFG_REQ_ENABLES2_ROCE_MAX_SRQ_PER_VF |
+			    FUNC_CFG_REQ_ENABLES2_ROCE_MAX_GID_PER_VF);
+	cfg_req->roce_max_av_per_vf =
+		cpu_to_le32(le32_to_cpu(resp->roce_vf_max_av) / num_vfs);
+	cfg_req->roce_max_cq_per_vf =
+		cpu_to_le32(le32_to_cpu(resp->roce_vf_max_cq) / num_vfs);
+	cfg_req->roce_max_mrw_per_vf =
+		cpu_to_le32(le32_to_cpu(resp->roce_vf_max_mrw) / num_vfs);
+	cfg_req->roce_max_qp_per_vf =
+		cpu_to_le32(le32_to_cpu(resp->roce_vf_max_qp) / num_vfs);
+	cfg_req->roce_max_srq_per_vf =
+		cpu_to_le32(le32_to_cpu(resp->roce_vf_max_srq) / num_vfs);
+	cfg_req->roce_max_gid_per_vf =
+		cpu_to_le32(le32_to_cpu(resp->roce_vf_max_gid) / num_vfs);
+
+	rc = hwrm_req_send(bp, cfg_req);
+
+err:
+	hwrm_req_drop(bp, req);
+	if (rc)
+		netdev_err(bp->dev, "RoCE sriov configuration failed\n");
+}
+
+/* Only called by PF to reserve resources for VFs, returns actual number of
+ * VFs configured, or < 0 on error.
+ */
+static int bnxt_hwrm_func_vf_resc_cfg(struct bnxt *bp, int num_vfs, bool reset)
+{
+	struct hwrm_func_vf_resource_cfg_input *req;
+	struct bnxt_hw_resc *hw_resc = &bp->hw_resc;
+	u16 vf_tx_rings, vf_rx_rings, vf_cp_rings;
+	u16 vf_stat_ctx, vf_vnics, vf_ring_grps;
+	struct bnxt_pf_info *pf = &bp->pf;
+	int i, rc = 0, min = 1;
+	u16 vf_msix = 0;
+	u16 vf_rss;
 
-	bnxt_hwrm_cmd_hdr_init(bp, &req, HWRM_FUNC_BUF_RGTR, -1, -1);
+	rc = hwrm_req_init(bp, req, HWRM_FUNC_VF_RESOURCE_CFG);
+	if (rc)
+		return rc;
+
+	if (bp->flags & BNXT_FLAG_CHIP_P5_PLUS) {
+		vf_msix = hw_resc->max_nqs - bnxt_nq_rings_in_use(bp);
+		vf_ring_grps = 0;
+	} else {
+		vf_ring_grps = hw_resc->max_hw_ring_grps - bp->rx_nr_rings;
+	}
+	vf_cp_rings = bnxt_get_avail_cp_rings_for_en(bp);
+	vf_stat_ctx = bnxt_get_avail_stat_ctxs_for_en(bp);
+	if (bp->flags & BNXT_FLAG_AGG_RINGS)
+		vf_rx_rings = hw_resc->max_rx_rings - bp->rx_nr_rings * 2;
+	else
+		vf_rx_rings = hw_resc->max_rx_rings - bp->rx_nr_rings;
+	vf_tx_rings = hw_resc->max_tx_rings - bp->tx_nr_rings;
+	vf_vnics = hw_resc->max_vnics - bp->nr_vnics;
+	vf_rss = hw_resc->max_rsscos_ctxs - bp->rsscos_nr_ctxs;
+
+	req->min_rsscos_ctx = cpu_to_le16(BNXT_VF_MIN_RSS_CTX);
+	if (pf->vf_resv_strategy == BNXT_VF_RESV_STRATEGY_MINIMAL_STATIC) {
+		min = 0;
+		req->min_rsscos_ctx = cpu_to_le16(min);
+	}
+	if (pf->vf_resv_strategy == BNXT_VF_RESV_STRATEGY_MINIMAL ||
+	    pf->vf_resv_strategy == BNXT_VF_RESV_STRATEGY_MINIMAL_STATIC) {
+		req->min_cmpl_rings = cpu_to_le16(min);
+		req->min_tx_rings = cpu_to_le16(min);
+		req->min_rx_rings = cpu_to_le16(min);
+		req->min_l2_ctxs = cpu_to_le16(min);
+		req->min_vnics = cpu_to_le16(min);
+		req->min_stat_ctx = cpu_to_le16(min);
+		if (!(bp->flags & BNXT_FLAG_CHIP_P5_PLUS))
+			req->min_hw_ring_grps = cpu_to_le16(min);
+	} else {
+		vf_cp_rings /= num_vfs;
+		vf_tx_rings /= num_vfs;
+		vf_rx_rings /= num_vfs;
+		if ((bp->fw_cap & BNXT_FW_CAP_PRE_RESV_VNICS) &&
+		    vf_vnics >= pf->max_vfs) {
+			/* Take into account that FW has pre-reserved 1 VNIC for
+			 * each pf->max_vfs.
+			 */
+			vf_vnics = (vf_vnics - pf->max_vfs + num_vfs) / num_vfs;
+		} else {
+			vf_vnics /= num_vfs;
+		}
+		vf_stat_ctx /= num_vfs;
+		vf_ring_grps /= num_vfs;
+		vf_rss /= num_vfs;
+
+		vf_vnics = min_t(u16, vf_vnics, vf_rx_rings);
+		req->min_cmpl_rings = cpu_to_le16(vf_cp_rings);
+		req->min_tx_rings = cpu_to_le16(vf_tx_rings);
+		req->min_rx_rings = cpu_to_le16(vf_rx_rings);
+		req->min_l2_ctxs = cpu_to_le16(BNXT_VF_MAX_L2_CTX);
+		req->min_vnics = cpu_to_le16(vf_vnics);
+		req->min_stat_ctx = cpu_to_le16(vf_stat_ctx);
+		req->min_hw_ring_grps = cpu_to_le16(vf_ring_grps);
+		req->min_rsscos_ctx = cpu_to_le16(vf_rss);
+	}
+	req->max_cmpl_rings = cpu_to_le16(vf_cp_rings);
+	req->max_tx_rings = cpu_to_le16(vf_tx_rings);
+	req->max_rx_rings = cpu_to_le16(vf_rx_rings);
+	req->max_l2_ctxs = cpu_to_le16(BNXT_VF_MAX_L2_CTX);
+	req->max_vnics = cpu_to_le16(vf_vnics);
+	req->max_stat_ctx = cpu_to_le16(vf_stat_ctx);
+	req->max_hw_ring_grps = cpu_to_le16(vf_ring_grps);
+	req->max_rsscos_ctx = cpu_to_le16(vf_rss);
+	if (bp->flags & BNXT_FLAG_CHIP_P5_PLUS)
+		req->max_msix = cpu_to_le16(vf_msix / num_vfs);
+
+	hwrm_req_hold(bp, req);
+	for (i = 0; i < num_vfs; i++) {
+		if (reset)
+			__bnxt_set_vf_params(bp, i);
 
-	req.req_buf_num_pages = cpu_to_le16(bp->pf.hwrm_cmd_req_pages);
-	req.req_buf_page_size = cpu_to_le16(BNXT_PAGE_SHIFT);
-	req.req_buf_len = cpu_to_le16(BNXT_HWRM_REQ_MAX_SIZE);
-	req.req_buf_page_addr0 = cpu_to_le64(bp->pf.hwrm_cmd_req_dma_addr[0]);
-	req.req_buf_page_addr1 = cpu_to_le64(bp->pf.hwrm_cmd_req_dma_addr[1]);
-	req.req_buf_page_addr2 = cpu_to_le64(bp->pf.hwrm_cmd_req_dma_addr[2]);
-	req.req_buf_page_addr3 = cpu_to_le64(bp->pf.hwrm_cmd_req_dma_addr[3]);
+		req->vf_id = cpu_to_le16(pf->first_vf_id + i);
+		rc = hwrm_req_send(bp, req);
+		if (rc)
+			break;
+		pf->active_vfs = i + 1;
+		pf->vf[i].fw_fid = pf->first_vf_id + i;
+	}
 
-	return hwrm_send_message(bp, &req, sizeof(req), HWRM_CMD_TIMEOUT);
+	if (pf->active_vfs) {
+		u16 n = pf->active_vfs;
+
+		hw_resc->max_tx_rings -= le16_to_cpu(req->min_tx_rings) * n;
+		hw_resc->max_rx_rings -= le16_to_cpu(req->min_rx_rings) * n;
+		hw_resc->max_hw_ring_grps -=
+			le16_to_cpu(req->min_hw_ring_grps) * n;
+		hw_resc->max_cp_rings -= le16_to_cpu(req->min_cmpl_rings) * n;
+		hw_resc->max_rsscos_ctxs -=
+			le16_to_cpu(req->min_rsscos_ctx) * n;
+		hw_resc->max_stat_ctxs -= le16_to_cpu(req->min_stat_ctx) * n;
+		hw_resc->max_vnics -= le16_to_cpu(req->min_vnics) * n;
+		if (bp->flags & BNXT_FLAG_CHIP_P5_PLUS)
+			hw_resc->max_nqs -= vf_msix;
+
+		rc = pf->active_vfs;
+	}
+	hwrm_req_drop(bp, req);
+	return rc;
 }
 
-/* only call by PF to reserve resources for VF */
+/* Only called by PF to reserve resources for VFs, returns actual number of
+ * VFs configured, or < 0 on error.
+ */
 static int bnxt_hwrm_func_cfg(struct bnxt *bp, int num_vfs)
 {
-	u32 rc = 0, mtu, i;
 	u16 vf_tx_rings, vf_rx_rings, vf_cp_rings, vf_stat_ctx, vf_vnics;
-	u16 vf_ring_grps;
-	struct hwrm_func_cfg_input req = {0};
+	struct bnxt_hw_resc *hw_resc = &bp->hw_resc;
 	struct bnxt_pf_info *pf = &bp->pf;
+	struct hwrm_func_cfg_input *req;
 	int total_vf_tx_rings = 0;
+	u16 vf_ring_grps;
+	u32 mtu, i;
+	int rc;
 
-	bnxt_hwrm_cmd_hdr_init(bp, &req, HWRM_FUNC_CFG, -1, -1);
+	rc = bnxt_hwrm_func_cfg_short_req_init(bp, &req);
+	if (rc)
+		return rc;
 
 	/* Remaining rings are distributed equally amongs VF's for now */
-	vf_cp_rings = (pf->max_cp_rings - bp->cp_nr_rings) / num_vfs;
-	vf_stat_ctx = (pf->max_stat_ctxs - bp->num_stat_ctxs) / num_vfs;
+	vf_cp_rings = bnxt_get_avail_cp_rings_for_en(bp) / num_vfs;
+	vf_stat_ctx = bnxt_get_avail_stat_ctxs_for_en(bp) / num_vfs;
 	if (bp->flags & BNXT_FLAG_AGG_RINGS)
-		vf_rx_rings = (pf->max_rx_rings - bp->rx_nr_rings * 2) /
+		vf_rx_rings = (hw_resc->max_rx_rings - bp->rx_nr_rings * 2) /
 			      num_vfs;
 	else
-		vf_rx_rings = (pf->max_rx_rings - bp->rx_nr_rings) / num_vfs;
-	vf_ring_grps = (bp->pf.max_hw_ring_grps - bp->rx_nr_rings) / num_vfs;
-	vf_tx_rings = (pf->max_tx_rings - bp->tx_nr_rings) / num_vfs;
-	vf_vnics = (pf->max_vnics - bp->nr_vnics) / num_vfs;
+		vf_rx_rings = (hw_resc->max_rx_rings - bp->rx_nr_rings) /
+			      num_vfs;
+	vf_ring_grps = (hw_resc->max_hw_ring_grps - bp->rx_nr_rings) / num_vfs;
+	vf_tx_rings = (hw_resc->max_tx_rings - bp->tx_nr_rings) / num_vfs;
+	vf_vnics = (hw_resc->max_vnics - bp->nr_vnics) / num_vfs;
 	vf_vnics = min_t(u16, vf_vnics, vf_rx_rings);
 
-	req.enables = cpu_to_le32(FUNC_CFG_REQ_ENABLES_MTU |
-				  FUNC_CFG_REQ_ENABLES_MRU |
-				  FUNC_CFG_REQ_ENABLES_NUM_RSSCOS_CTXS |
-				  FUNC_CFG_REQ_ENABLES_NUM_STAT_CTXS |
-				  FUNC_CFG_REQ_ENABLES_NUM_CMPL_RINGS |
-				  FUNC_CFG_REQ_ENABLES_NUM_TX_RINGS |
-				  FUNC_CFG_REQ_ENABLES_NUM_RX_RINGS |
-				  FUNC_CFG_REQ_ENABLES_NUM_L2_CTXS |
-				  FUNC_CFG_REQ_ENABLES_NUM_VNICS |
-				  FUNC_CFG_REQ_ENABLES_NUM_HW_RING_GRPS);
-
-	mtu = bp->dev->mtu + ETH_HLEN + ETH_FCS_LEN + VLAN_HLEN;
-	req.mru = cpu_to_le16(mtu);
-	req.mtu = cpu_to_le16(mtu);
-
-	req.num_rsscos_ctxs = cpu_to_le16(1);
-	req.num_cmpl_rings = cpu_to_le16(vf_cp_rings);
-	req.num_tx_rings = cpu_to_le16(vf_tx_rings);
-	req.num_rx_rings = cpu_to_le16(vf_rx_rings);
-	req.num_hw_ring_grps = cpu_to_le16(vf_ring_grps);
-	req.num_l2_ctxs = cpu_to_le16(4);
-
-	req.num_vnics = cpu_to_le16(vf_vnics);
+	req->enables = cpu_to_le32(FUNC_CFG_REQ_ENABLES_ADMIN_MTU |
+				   FUNC_CFG_REQ_ENABLES_MRU |
+				   FUNC_CFG_REQ_ENABLES_NUM_RSSCOS_CTXS |
+				   FUNC_CFG_REQ_ENABLES_NUM_STAT_CTXS |
+				   FUNC_CFG_REQ_ENABLES_NUM_CMPL_RINGS |
+				   FUNC_CFG_REQ_ENABLES_NUM_TX_RINGS |
+				   FUNC_CFG_REQ_ENABLES_NUM_RX_RINGS |
+				   FUNC_CFG_REQ_ENABLES_NUM_L2_CTXS |
+				   FUNC_CFG_REQ_ENABLES_NUM_VNICS |
+				   FUNC_CFG_REQ_ENABLES_NUM_HW_RING_GRPS);
+
+	mtu = bp->dev->mtu + VLAN_ETH_HLEN;
+	req->mru = cpu_to_le16(mtu);
+	req->admin_mtu = cpu_to_le16(mtu);
+
+	req->num_rsscos_ctxs = cpu_to_le16(1);
+	req->num_cmpl_rings = cpu_to_le16(vf_cp_rings);
+	req->num_tx_rings = cpu_to_le16(vf_tx_rings);
+	req->num_rx_rings = cpu_to_le16(vf_rx_rings);
+	req->num_hw_ring_grps = cpu_to_le16(vf_ring_grps);
+	req->num_l2_ctxs = cpu_to_le16(4);
+
+	req->num_vnics = cpu_to_le16(vf_vnics);
 	/* FIXME spec currently uses 1 bit for stats ctx */
-	req.num_stat_ctxs = cpu_to_le16(vf_stat_ctx);
+	req->num_stat_ctxs = cpu_to_le16(vf_stat_ctx);
 
-	mutex_lock(&bp->hwrm_cmd_lock);
+	hwrm_req_hold(bp, req);
 	for (i = 0; i < num_vfs; i++) {
 		int vf_tx_rsvd = vf_tx_rings;
 
-		req.fid = cpu_to_le16(pf->first_vf_id + i);
-		rc = _hwrm_send_message(bp, &req, sizeof(req),
-					HWRM_CMD_TIMEOUT);
+		req->fid = cpu_to_le16(pf->first_vf_id + i);
+		rc = hwrm_req_send(bp, req);
 		if (rc)
 			break;
 		pf->active_vfs = i + 1;
-		pf->vf[i].fw_fid = le16_to_cpu(req.fid);
+		pf->vf[i].fw_fid = le16_to_cpu(req->fid);
 		rc = __bnxt_hwrm_get_tx_rings(bp, pf->vf[i].fw_fid,
 					      &vf_tx_rsvd);
 		if (rc)
 			break;
 		total_vf_tx_rings += vf_tx_rsvd;
 	}
-	mutex_unlock(&bp->hwrm_cmd_lock);
-	if (!rc) {
-		pf->max_tx_rings -= total_vf_tx_rings;
-		pf->max_rx_rings -= vf_rx_rings * num_vfs;
-		pf->max_hw_ring_grps -= vf_ring_grps * num_vfs;
-		pf->max_cp_rings -= vf_cp_rings * num_vfs;
-		pf->max_rsscos_ctxs -= num_vfs;
-		pf->max_stat_ctxs -= vf_stat_ctx * num_vfs;
-		pf->max_vnics -= vf_vnics * num_vfs;
+	hwrm_req_drop(bp, req);
+	if (pf->active_vfs) {
+		hw_resc->max_tx_rings -= total_vf_tx_rings;
+		hw_resc->max_rx_rings -= vf_rx_rings * num_vfs;
+		hw_resc->max_hw_ring_grps -= vf_ring_grps * num_vfs;
+		hw_resc->max_cp_rings -= vf_cp_rings * num_vfs;
+		hw_resc->max_rsscos_ctxs -= num_vfs;
+		hw_resc->max_stat_ctxs -= vf_stat_ctx * num_vfs;
+		hw_resc->max_vnics -= vf_vnics * num_vfs;
+		rc = pf->active_vfs;
 	}
 	return rc;
 }
 
+static int bnxt_func_cfg(struct bnxt *bp, int num_vfs, bool reset)
+{
+	if (BNXT_NEW_RM(bp))
+		return bnxt_hwrm_func_vf_resc_cfg(bp, num_vfs, reset);
+	else
+		return bnxt_hwrm_func_cfg(bp, num_vfs);
+}
+
+int bnxt_cfg_hw_sriov(struct bnxt *bp, int *num_vfs, bool reset)
+{
+	int rc;
+
+	/* Register buffers for VFs */
+	rc = bnxt_hwrm_func_buf_rgtr(bp);
+	if (rc)
+		return rc;
+
+	/* Reserve resources for VFs */
+	rc = bnxt_func_cfg(bp, *num_vfs, reset);
+	if (rc != *num_vfs) {
+		if (rc <= 0) {
+			netdev_warn(bp->dev, "Unable to reserve resources for SRIOV.\n");
+			*num_vfs = 0;
+			return rc;
+		}
+		netdev_warn(bp->dev, "Only able to reserve resources for %d VFs.\n",
+			    rc);
+		*num_vfs = rc;
+	}
+
+	if (BNXT_RDMA_SRIOV_EN(bp) && BNXT_ROCE_VF_RESC_CAP(bp))
+		bnxt_hwrm_roce_sriov_cfg(bp, *num_vfs);
+
+	return 0;
+}
+
 static int bnxt_sriov_enable(struct bnxt *bp, int *num_vfs)
 {
 	int rc = 0, vfs_supported;
 	int min_rx_rings, min_tx_rings, min_rss_ctxs;
+	struct bnxt_hw_resc *hw_resc = &bp->hw_resc;
 	int tx_ok = 0, rx_ok = 0, rss_ok = 0;
+	int avail_cp, avail_stat;
 
-	/* Check if we can enable requested num of vf's. At a mininum
+	/* Check if we can enable requested num of vf's. At a minimum
 	 * we require 1 RX 1 TX rings for each VF. In this minimum conf
 	 * features like TPA will not be available.
 	 */
 	vfs_supported = *num_vfs;
 
+	avail_cp = bnxt_get_avail_cp_rings_for_en(bp);
+	avail_stat = bnxt_get_avail_stat_ctxs_for_en(bp);
+	avail_cp = min_t(int, avail_cp, avail_stat);
+
 	while (vfs_supported) {
 		min_rx_rings = vfs_supported;
 		min_tx_rings = vfs_supported;
 		min_rss_ctxs = vfs_supported;
 
 		if (bp->flags & BNXT_FLAG_AGG_RINGS) {
-			if (bp->pf.max_rx_rings - bp->rx_nr_rings * 2 >=
+			if (hw_resc->max_rx_rings - bp->rx_nr_rings * 2 >=
 			    min_rx_rings)
 				rx_ok = 1;
 		} else {
-			if (bp->pf.max_rx_rings - bp->rx_nr_rings >=
+			if (hw_resc->max_rx_rings - bp->rx_nr_rings >=
 			    min_rx_rings)
 				rx_ok = 1;
 		}
-		if (bp->pf.max_vnics - bp->nr_vnics < min_rx_rings)
+		if (hw_resc->max_vnics - bp->nr_vnics < min_rx_rings ||
+		    avail_cp < min_rx_rings)
 			rx_ok = 0;
 
-		if (bp->pf.max_tx_rings - bp->tx_nr_rings >= min_tx_rings)
+		if (hw_resc->max_tx_rings - bp->tx_nr_rings >= min_tx_rings &&
+		    avail_cp >= min_tx_rings)
 			tx_ok = 1;
 
-		if (bp->pf.max_rsscos_ctxs - bp->rsscos_nr_ctxs >= min_rss_ctxs)
+		if (hw_resc->max_rsscos_ctxs - bp->rsscos_nr_ctxs >=
+		    min_rss_ctxs)
 			rss_ok = 1;
 
 		if (tx_ok && rx_ok && rss_ok)
@@ -552,41 +887,56 @@ static int bnxt_sriov_enable(struct bnxt *bp, int *num_vfs)
 	if (rc)
 		goto err_out1;
 
-	/* Reserve resources for VFs */
-	rc = bnxt_hwrm_func_cfg(bp, *num_vfs);
+	rc = bnxt_cfg_hw_sriov(bp, num_vfs, false);
 	if (rc)
 		goto err_out2;
 
-	/* Register buffers for VFs */
-	rc = bnxt_hwrm_func_buf_rgtr(bp);
+	rc = pci_enable_sriov(bp->pdev, *num_vfs);
 	if (rc)
 		goto err_out2;
 
-	bnxt_ulp_sriov_cfg(bp, *num_vfs);
+	if (bp->eswitch_mode != DEVLINK_ESWITCH_MODE_SWITCHDEV)
+		return 0;
 
-	rc = pci_enable_sriov(bp->pdev, *num_vfs);
-	if (rc)
-		goto err_out2;
+	/* Create representors for VFs in switchdev mode */
+	devl_lock(bp->dl);
+	rc = bnxt_vf_reps_create(bp);
+	devl_unlock(bp->dl);
+	if (rc) {
+		netdev_info(bp->dev, "Cannot enable VFS as representors cannot be created\n");
+		goto err_out3;
+	}
 
 	return 0;
 
+err_out3:
+	/* Disable SR-IOV */
+	pci_disable_sriov(bp->pdev);
+
 err_out2:
 	/* Free the resources reserved for various VF's */
 	bnxt_hwrm_func_vf_resource_free(bp, *num_vfs);
 
+	/* Restore the max resources */
+	bnxt_hwrm_func_qcaps(bp);
+
 err_out1:
 	bnxt_free_vf_resources(bp);
 
 	return rc;
 }
 
-void bnxt_sriov_disable(struct bnxt *bp)
+void __bnxt_sriov_disable(struct bnxt *bp)
 {
 	u16 num_vfs = pci_num_vf(bp->pdev);
 
 	if (!num_vfs)
 		return;
 
+	/* synchronize VF and VF-rep create and destroy */
+	devl_lock(bp->dl);
+	bnxt_vf_reps_destroy(bp);
+
 	if (pci_vfs_assigned(bp->pdev)) {
 		bnxt_hwrm_fwd_async_event_cmpl(
 			bp, NULL, ASYNC_EVENT_CMPL_EVENT_ID_PF_DRVR_UNLOAD);
@@ -597,16 +947,24 @@ void bnxt_sriov_disable(struct bnxt *bp)
 		/* Free the HW resources reserved for various VF's */
 		bnxt_hwrm_func_vf_resource_free(bp, num_vfs);
 	}
+	devl_unlock(bp->dl);
 
 	bnxt_free_vf_resources(bp);
+}
+
+static void bnxt_sriov_disable(struct bnxt *bp)
+{
+	if (!pci_num_vf(bp->pdev))
+		return;
+
+	__bnxt_sriov_disable(bp);
 
-	bp->pf.active_vfs = 0;
 	/* Reclaim all resources for the PF. */
 	rtnl_lock();
+	netdev_lock(bp->dev);
 	bnxt_restore_pf_fw_resources(bp);
+	netdev_unlock(bp->dev);
 	rtnl_unlock();
-
-	bnxt_ulp_sriov_cfg(bp, 0);
 }
 
 int bnxt_sriov_configure(struct pci_dev *pdev, int num_vfs)
@@ -614,18 +972,22 @@ int bnxt_sriov_configure(struct pci_dev *pdev, int num_vfs)
 	struct net_device *dev = pci_get_drvdata(pdev);
 	struct bnxt *bp = netdev_priv(dev);
 
-	if (!(bp->flags & BNXT_FLAG_USING_MSIX)) {
-		netdev_warn(dev, "Not allow SRIOV if the irq mode is not MSIX\n");
-		return 0;
-	}
-
 	rtnl_lock();
+	netdev_lock(dev);
 	if (!netif_running(dev)) {
 		netdev_warn(dev, "Reject SRIOV config request since if is down!\n");
+		netdev_unlock(dev);
+		rtnl_unlock();
+		return 0;
+	}
+	if (test_bit(BNXT_STATE_IN_FW_RESET, &bp->state)) {
+		netdev_warn(dev, "Reject SRIOV config request when FW reset is in progress\n");
+		netdev_unlock(dev);
 		rtnl_unlock();
 		return 0;
 	}
 	bp->sriov_cfg = true;
+	netdev_unlock(dev);
 	rtnl_unlock();
 
 	if (pci_vfs_assigned(bp->pdev)) {
@@ -656,101 +1018,99 @@ static int bnxt_hwrm_fwd_resp(struct bnxt *bp, struct bnxt_vf_info *vf,
 			      void *encap_resp, __le64 encap_resp_addr,
 			      __le16 encap_resp_cpr, u32 msg_size)
 {
-	int rc = 0;
-	struct hwrm_fwd_resp_input req = {0};
-	struct hwrm_fwd_resp_output *resp = bp->hwrm_cmd_resp_addr;
-
-	bnxt_hwrm_cmd_hdr_init(bp, &req, HWRM_FWD_RESP, -1, -1);
-
-	/* Set the new target id */
-	req.target_id = cpu_to_le16(vf->fw_fid);
-	req.encap_resp_target_id = cpu_to_le16(vf->fw_fid);
-	req.encap_resp_len = cpu_to_le16(msg_size);
-	req.encap_resp_addr = encap_resp_addr;
-	req.encap_resp_cmpl_ring = encap_resp_cpr;
-	memcpy(req.encap_resp, encap_resp, msg_size);
-
-	mutex_lock(&bp->hwrm_cmd_lock);
-	rc = _hwrm_send_message(bp, &req, sizeof(req), HWRM_CMD_TIMEOUT);
+	struct hwrm_fwd_resp_input *req;
+	int rc;
 
-	if (rc) {
-		netdev_err(bp->dev, "hwrm_fwd_resp failed. rc:%d\n", rc);
-		goto fwd_resp_exit;
+	if (BNXT_FWD_RESP_SIZE_ERR(msg_size)) {
+		netdev_warn_once(bp->dev, "HWRM fwd response too big (%d bytes)\n",
+				 msg_size);
+		return -EINVAL;
 	}
 
-	if (resp->error_code) {
-		netdev_err(bp->dev, "hwrm_fwd_resp error %d\n",
-			   resp->error_code);
-		rc = -1;
+	rc = hwrm_req_init(bp, req, HWRM_FWD_RESP);
+	if (!rc) {
+		/* Set the new target id */
+		req->target_id = cpu_to_le16(vf->fw_fid);
+		req->encap_resp_target_id = cpu_to_le16(vf->fw_fid);
+		req->encap_resp_len = cpu_to_le16(msg_size);
+		req->encap_resp_addr = encap_resp_addr;
+		req->encap_resp_cmpl_ring = encap_resp_cpr;
+		memcpy(req->encap_resp, encap_resp, msg_size);
+
+		rc = hwrm_req_send(bp, req);
 	}
-
-fwd_resp_exit:
-	mutex_unlock(&bp->hwrm_cmd_lock);
+	if (rc)
+		netdev_err(bp->dev, "hwrm_fwd_resp failed. rc:%d\n", rc);
 	return rc;
 }
 
 static int bnxt_hwrm_fwd_err_resp(struct bnxt *bp, struct bnxt_vf_info *vf,
 				  u32 msg_size)
 {
-	int rc = 0;
-	struct hwrm_reject_fwd_resp_input req = {0};
-	struct hwrm_reject_fwd_resp_output *resp = bp->hwrm_cmd_resp_addr;
-
-	bnxt_hwrm_cmd_hdr_init(bp, &req, HWRM_REJECT_FWD_RESP, -1, -1);
-	/* Set the new target id */
-	req.target_id = cpu_to_le16(vf->fw_fid);
-	req.encap_resp_target_id = cpu_to_le16(vf->fw_fid);
-	memcpy(req.encap_request, vf->hwrm_cmd_req_addr, msg_size);
+	struct hwrm_reject_fwd_resp_input *req;
+	int rc;
 
-	mutex_lock(&bp->hwrm_cmd_lock);
-	rc = _hwrm_send_message(bp, &req, sizeof(req), HWRM_CMD_TIMEOUT);
+	if (BNXT_REJ_FWD_RESP_SIZE_ERR(msg_size))
+		return -EINVAL;
 
-	if (rc) {
-		netdev_err(bp->dev, "hwrm_fwd_err_resp failed. rc:%d\n", rc);
-		goto fwd_err_resp_exit;
-	}
+	rc = hwrm_req_init(bp, req, HWRM_REJECT_FWD_RESP);
+	if (!rc) {
+		/* Set the new target id */
+		req->target_id = cpu_to_le16(vf->fw_fid);
+		req->encap_resp_target_id = cpu_to_le16(vf->fw_fid);
+		memcpy(req->encap_request, vf->hwrm_cmd_req_addr, msg_size);
 
-	if (resp->error_code) {
-		netdev_err(bp->dev, "hwrm_fwd_err_resp error %d\n",
-			   resp->error_code);
-		rc = -1;
+		rc = hwrm_req_send(bp, req);
 	}
-
-fwd_err_resp_exit:
-	mutex_unlock(&bp->hwrm_cmd_lock);
+	if (rc)
+		netdev_err(bp->dev, "hwrm_fwd_err_resp failed. rc:%d\n", rc);
 	return rc;
 }
 
 static int bnxt_hwrm_exec_fwd_resp(struct bnxt *bp, struct bnxt_vf_info *vf,
 				   u32 msg_size)
 {
-	int rc = 0;
-	struct hwrm_exec_fwd_resp_input req = {0};
-	struct hwrm_exec_fwd_resp_output *resp = bp->hwrm_cmd_resp_addr;
+	struct hwrm_exec_fwd_resp_input *req;
+	int rc;
 
-	bnxt_hwrm_cmd_hdr_init(bp, &req, HWRM_EXEC_FWD_RESP, -1, -1);
-	/* Set the new target id */
-	req.target_id = cpu_to_le16(vf->fw_fid);
-	req.encap_resp_target_id = cpu_to_le16(vf->fw_fid);
-	memcpy(req.encap_request, vf->hwrm_cmd_req_addr, msg_size);
+	if (BNXT_EXEC_FWD_RESP_SIZE_ERR(msg_size))
+		return -EINVAL;
 
-	mutex_lock(&bp->hwrm_cmd_lock);
-	rc = _hwrm_send_message(bp, &req, sizeof(req), HWRM_CMD_TIMEOUT);
+	rc = hwrm_req_init(bp, req, HWRM_EXEC_FWD_RESP);
+	if (!rc) {
+		/* Set the new target id */
+		req->target_id = cpu_to_le16(vf->fw_fid);
+		req->encap_resp_target_id = cpu_to_le16(vf->fw_fid);
+		memcpy(req->encap_request, vf->hwrm_cmd_req_addr, msg_size);
 
-	if (rc) {
-		netdev_err(bp->dev, "hwrm_exec_fw_resp failed. rc:%d\n", rc);
-		goto exec_fwd_resp_exit;
+		rc = hwrm_req_send(bp, req);
 	}
+	if (rc)
+		netdev_err(bp->dev, "hwrm_exec_fw_resp failed. rc:%d\n", rc);
+	return rc;
+}
 
-	if (resp->error_code) {
-		netdev_err(bp->dev, "hwrm_exec_fw_resp error %d\n",
-			   resp->error_code);
-		rc = -1;
-	}
+static int bnxt_vf_configure_mac(struct bnxt *bp, struct bnxt_vf_info *vf)
+{
+	u32 msg_size = sizeof(struct hwrm_func_vf_cfg_input);
+	struct hwrm_func_vf_cfg_input *req =
+		(struct hwrm_func_vf_cfg_input *)vf->hwrm_cmd_req_addr;
 
-exec_fwd_resp_exit:
-	mutex_unlock(&bp->hwrm_cmd_lock);
-	return rc;
+	/* Allow VF to set a valid MAC address, if trust is set to on or
+	 * if the PF assigned MAC address is zero
+	 */
+	if (req->enables & cpu_to_le32(FUNC_VF_CFG_REQ_ENABLES_DFLT_MAC_ADDR)) {
+		bool trust = bnxt_is_trusted_vf(bp, vf);
+
+		if (is_valid_ether_addr(req->dflt_mac_addr) &&
+		    (trust || !is_valid_ether_addr(vf->mac_addr) ||
+		     ether_addr_equal(req->dflt_mac_addr, vf->mac_addr))) {
+			ether_addr_copy(vf->vf_mac_addr, req->dflt_mac_addr);
+			return bnxt_hwrm_exec_fwd_resp(bp, vf, msg_size);
+		}
+		return bnxt_hwrm_fwd_err_resp(bp, vf, msg_size);
+	}
+	return bnxt_hwrm_exec_fwd_resp(bp, vf, msg_size);
 }
 
 static int bnxt_vf_validate_set_mac(struct bnxt *bp, struct bnxt_vf_info *vf)
@@ -758,12 +1118,36 @@ static int bnxt_vf_validate_set_mac(struct bnxt *bp, struct bnxt_vf_info *vf)
 	u32 msg_size = sizeof(struct hwrm_cfa_l2_filter_alloc_input);
 	struct hwrm_cfa_l2_filter_alloc_input *req =
 		(struct hwrm_cfa_l2_filter_alloc_input *)vf->hwrm_cmd_req_addr;
+	bool mac_ok = false;
 
-	if (!is_valid_ether_addr(vf->mac_addr) ||
-	    ether_addr_equal((const u8 *)req->l2_addr, vf->mac_addr))
-		return bnxt_hwrm_exec_fwd_resp(bp, vf, msg_size);
-	else
+	if (!is_valid_ether_addr((const u8 *)req->l2_addr))
 		return bnxt_hwrm_fwd_err_resp(bp, vf, msg_size);
+
+	/* Allow VF to set a valid MAC address, if trust is set to on.
+	 * Or VF MAC address must first match MAC address in PF's context.
+	 * Otherwise, it must match the VF MAC address if firmware spec >=
+	 * 1.2.2
+	 */
+	if (bnxt_is_trusted_vf(bp, vf)) {
+		mac_ok = true;
+	} else if (is_valid_ether_addr(vf->mac_addr)) {
+		if (ether_addr_equal((const u8 *)req->l2_addr, vf->mac_addr))
+			mac_ok = true;
+	} else if (is_valid_ether_addr(vf->vf_mac_addr)) {
+		if (ether_addr_equal((const u8 *)req->l2_addr, vf->vf_mac_addr))
+			mac_ok = true;
+	} else {
+		/* There are two cases:
+		 * 1.If firmware spec < 0x10202,VF MAC address is not forwarded
+		 *   to the PF and so it doesn't have to match
+		 * 2.Allow VF to modify its own MAC when PF has not assigned a
+		 *   valid MAC address and firmware spec >= 0x10202
+		 */
+		mac_ok = true;
+	}
+	if (mac_ok)
+		return bnxt_hwrm_exec_fwd_resp(bp, vf, msg_size);
+	return bnxt_hwrm_fwd_err_resp(bp, vf, msg_size);
 }
 
 static int bnxt_vf_set_link(struct bnxt *bp, struct bnxt_vf_info *vf)
@@ -775,16 +1159,23 @@ static int bnxt_vf_set_link(struct bnxt *bp, struct bnxt_vf_info *vf)
 		rc = bnxt_hwrm_exec_fwd_resp(
 			bp, vf, sizeof(struct hwrm_port_phy_qcfg_input));
 	} else {
-		struct hwrm_port_phy_qcfg_output phy_qcfg_resp;
+		struct hwrm_port_phy_qcfg_output_compat phy_qcfg_resp = {};
 		struct hwrm_port_phy_qcfg_input *phy_qcfg_req;
 
 		phy_qcfg_req =
 		(struct hwrm_port_phy_qcfg_input *)vf->hwrm_cmd_req_addr;
-		mutex_lock(&bp->hwrm_cmd_lock);
+		mutex_lock(&bp->link_lock);
 		memcpy(&phy_qcfg_resp, &bp->link_info.phy_qcfg_resp,
 		       sizeof(phy_qcfg_resp));
-		mutex_unlock(&bp->hwrm_cmd_lock);
+		mutex_unlock(&bp->link_lock);
+		phy_qcfg_resp.resp_len = cpu_to_le16(sizeof(phy_qcfg_resp));
 		phy_qcfg_resp.seq_id = phy_qcfg_req->seq_id;
+		/* New SPEEDS2 fields are beyond the legacy structure, so
+		 * clear the SPEEDS2_SUPPORTED flag.
+		 */
+		phy_qcfg_resp.option_flags &=
+			~PORT_PHY_QCAPS_RESP_FLAGS2_SPEEDS2_SUPPORTED;
+		phy_qcfg_resp.valid = 1;
 
 		if (vf->flags & BNXT_VF_LINK_UP) {
 			/* if physical link is down, force link up on VF */
@@ -794,8 +1185,10 @@ static int bnxt_vf_set_link(struct bnxt *bp, struct bnxt_vf_info *vf)
 					PORT_PHY_QCFG_RESP_LINK_LINK;
 				phy_qcfg_resp.link_speed = cpu_to_le16(
 					PORT_PHY_QCFG_RESP_LINK_SPEED_10GB);
-				phy_qcfg_resp.duplex =
-					PORT_PHY_QCFG_RESP_DUPLEX_FULL;
+				phy_qcfg_resp.duplex_cfg =
+					PORT_PHY_QCFG_RESP_DUPLEX_CFG_FULL;
+				phy_qcfg_resp.duplex_state =
+					PORT_PHY_QCFG_RESP_DUPLEX_STATE_FULL;
 				phy_qcfg_resp.pause =
 					(PORT_PHY_QCFG_RESP_PAUSE_TX |
 					 PORT_PHY_QCFG_RESP_PAUSE_RX);
@@ -804,7 +1197,8 @@ static int bnxt_vf_set_link(struct bnxt *bp, struct bnxt_vf_info *vf)
 			/* force link down */
 			phy_qcfg_resp.link = PORT_PHY_QCFG_RESP_LINK_NO_LINK;
 			phy_qcfg_resp.link_speed = 0;
-			phy_qcfg_resp.duplex = PORT_PHY_QCFG_RESP_DUPLEX_HALF;
+			phy_qcfg_resp.duplex_state =
+				PORT_PHY_QCFG_RESP_DUPLEX_STATE_HALF;
 			phy_qcfg_resp.pause = 0;
 		}
 		rc = bnxt_hwrm_fwd_resp(bp, vf, &phy_qcfg_resp,
@@ -822,6 +1216,9 @@ static int bnxt_vf_req_validate_snd(struct bnxt *bp, struct bnxt_vf_info *vf)
 	u32 req_type = le16_to_cpu(encap_req->req_type);
 
 	switch (req_type) {
+	case HWRM_FUNC_VF_CFG:
+		rc = bnxt_vf_configure_mac(bp, vf);
+		break;
 	case HWRM_CFA_L2_FILTER_ALLOC:
 		rc = bnxt_vf_validate_set_mac(bp, vf);
 		break;
@@ -857,16 +1254,52 @@ void bnxt_hwrm_exec_fwd_req(struct bnxt *bp)
 	}
 }
 
+int bnxt_approve_mac(struct bnxt *bp, const u8 *mac, bool strict)
+{
+	struct hwrm_func_vf_cfg_input *req;
+	int rc = 0;
+
+	if (!BNXT_VF(bp))
+		return 0;
+
+	if (bp->hwrm_spec_code < 0x10202) {
+		if (is_valid_ether_addr(bp->vf.mac_addr))
+			rc = -EADDRNOTAVAIL;
+		goto mac_done;
+	}
+
+	rc = hwrm_req_init(bp, req, HWRM_FUNC_VF_CFG);
+	if (rc)
+		goto mac_done;
+
+	req->enables = cpu_to_le32(FUNC_VF_CFG_REQ_ENABLES_DFLT_MAC_ADDR);
+	memcpy(req->dflt_mac_addr, mac, ETH_ALEN);
+	if (!strict)
+		hwrm_req_flags(bp, req, BNXT_HWRM_CTX_SILENT);
+	rc = hwrm_req_send(bp, req);
+mac_done:
+	if (rc && strict) {
+		rc = -EADDRNOTAVAIL;
+		netdev_warn(bp->dev, "VF MAC address %pM not approved by the PF\n",
+			    mac);
+		return rc;
+	}
+	return 0;
+}
+
 void bnxt_update_vf_mac(struct bnxt *bp)
 {
-	struct hwrm_func_qcaps_input req = {0};
-	struct hwrm_func_qcaps_output *resp = bp->hwrm_cmd_resp_addr;
+	struct hwrm_func_qcaps_output *resp;
+	struct hwrm_func_qcaps_input *req;
+	bool inform_pf = false;
 
-	bnxt_hwrm_cmd_hdr_init(bp, &req, HWRM_FUNC_QCAPS, -1, -1);
-	req.fid = cpu_to_le16(0xffff);
+	if (hwrm_req_init(bp, req, HWRM_FUNC_QCAPS))
+		return;
+
+	req->fid = cpu_to_le16(0xffff);
 
-	mutex_lock(&bp->hwrm_cmd_lock);
-	if (_hwrm_send_message(bp, &req, sizeof(req), HWRM_CMD_TIMEOUT))
+	resp = hwrm_req_hold(bp, req);
+	if (hwrm_req_send(bp, req))
 		goto update_vf_mac_exit;
 
 	/* Store MAC address from the firmware.  There are 2 cases:
@@ -876,44 +1309,34 @@ void bnxt_update_vf_mac(struct bnxt *bp)
 	 *    default but the stored zero MAC will allow the VF user to change
 	 *    the random MAC address using ndo_set_mac_address() if he wants.
 	 */
-	if (!ether_addr_equal(resp->mac_address, bp->vf.mac_addr))
+	if (!ether_addr_equal(resp->mac_address, bp->vf.mac_addr)) {
 		memcpy(bp->vf.mac_addr, resp->mac_address, ETH_ALEN);
+		/* This means we are now using our own MAC address, let
+		 * the PF know about this MAC address.
+		 */
+		if (!is_valid_ether_addr(bp->vf.mac_addr))
+			inform_pf = true;
+	}
 
 	/* overwrite netdev dev_addr with admin VF MAC */
 	if (is_valid_ether_addr(bp->vf.mac_addr))
-		memcpy(bp->dev->dev_addr, bp->vf.mac_addr, ETH_ALEN);
+		eth_hw_addr_set(bp->dev, bp->vf.mac_addr);
 update_vf_mac_exit:
-	mutex_unlock(&bp->hwrm_cmd_lock);
+	hwrm_req_drop(bp, req);
+	if (inform_pf)
+		bnxt_approve_mac(bp, bp->dev->dev_addr, false);
 }
 
-int bnxt_approve_mac(struct bnxt *bp, u8 *mac)
-{
-	struct hwrm_func_vf_cfg_input req = {0};
-	int rc = 0;
-
-	if (!BNXT_VF(bp))
-		return 0;
+#else
 
-	if (bp->hwrm_spec_code < 0x10202) {
-		if (is_valid_ether_addr(bp->vf.mac_addr))
-			rc = -EADDRNOTAVAIL;
-		goto mac_done;
-	}
-	bnxt_hwrm_cmd_hdr_init(bp, &req, HWRM_FUNC_VF_CFG, -1, -1);
-	req.enables = cpu_to_le32(FUNC_VF_CFG_REQ_ENABLES_DFLT_MAC_ADDR);
-	memcpy(req.dflt_mac_addr, mac, ETH_ALEN);
-	rc = hwrm_send_message(bp, &req, sizeof(req), HWRM_CMD_TIMEOUT);
-mac_done:
-	if (rc) {
-		rc = -EADDRNOTAVAIL;
-		netdev_warn(bp->dev, "VF MAC address %pM not approved by the PF\n",
-			    mac);
-	}
-	return rc;
+int bnxt_cfg_hw_sriov(struct bnxt *bp, int *num_vfs, bool reset)
+{
+	if (*num_vfs)
+		return -EOPNOTSUPP;
+	return 0;
 }
-#else
 
-void bnxt_sriov_disable(struct bnxt *bp)
+void __bnxt_sriov_disable(struct bnxt *bp)
 {
 }
 
@@ -926,7 +1349,7 @@ void bnxt_update_vf_mac(struct bnxt *bp)
 {
 }
 
-int bnxt_approve_mac(struct bnxt *bp, u8 *mac)
+int bnxt_approve_mac(struct bnxt *bp, const u8 *mac, bool strict)
 {
 	return 0;
 }
diff --git a/drivers/net/ethernet/broadcom/bnxt/bnxt_sriov.h b/drivers/net/ethernet/broadcom/bnxt/bnxt_sriov.h
index dbc8d977fc5a..e4979d729312 100644
--- a/drivers/net/ethernet/broadcom/bnxt/bnxt_sriov.h
+++ b/drivers/net/ethernet/broadcom/bnxt/bnxt_sriov.h
@@ -1,7 +1,7 @@
 /* Broadcom NetXtreme-C/E network driver.
  *
  * Copyright (c) 2014-2016 Broadcom Corporation
- * Copyright (c) 2016-2017 Broadcom Limited
+ * Copyright (c) 2016-2018 Broadcom Limited
  *
  * This program is free software; you can redistribute it and/or modify
  * it under the terms of the GNU General Public License as published by
@@ -11,15 +11,35 @@
 #ifndef BNXT_SRIOV_H
 #define BNXT_SRIOV_H
 
+#define BNXT_FWD_RESP_SIZE_ERR(n)					\
+	((offsetof(struct hwrm_fwd_resp_input, encap_resp) + n) >	\
+	 sizeof(struct hwrm_fwd_resp_input))
+
+#define BNXT_EXEC_FWD_RESP_SIZE_ERR(n)					\
+	((offsetof(struct hwrm_exec_fwd_resp_input, encap_request) + n) >\
+	 offsetof(struct hwrm_exec_fwd_resp_input, encap_resp_target_id))
+
+#define BNXT_REJ_FWD_RESP_SIZE_ERR(n)					\
+	((offsetof(struct hwrm_reject_fwd_resp_input, encap_request) + n) >\
+	 offsetof(struct hwrm_reject_fwd_resp_input, encap_resp_target_id))
+
+#define BNXT_VF_MIN_RSS_CTX	1
+#define BNXT_VF_MAX_RSS_CTX	1
+#define BNXT_VF_MIN_L2_CTX	1
+#define BNXT_VF_MAX_L2_CTX	4
+
 int bnxt_get_vf_config(struct net_device *, int, struct ifla_vf_info *);
 int bnxt_set_vf_mac(struct net_device *, int, u8 *);
 int bnxt_set_vf_vlan(struct net_device *, int, u16, u8, __be16);
 int bnxt_set_vf_bw(struct net_device *, int, int, int);
 int bnxt_set_vf_link_state(struct net_device *, int, int);
 int bnxt_set_vf_spoofchk(struct net_device *, int, bool);
+bool bnxt_is_trusted_vf(struct bnxt *bp, struct bnxt_vf_info *vf);
+int bnxt_set_vf_trust(struct net_device *dev, int vf_id, bool trust);
 int bnxt_sriov_configure(struct pci_dev *pdev, int num_vfs);
-void bnxt_sriov_disable(struct bnxt *);
+int bnxt_cfg_hw_sriov(struct bnxt *bp, int *num_vfs, bool reset);
+void __bnxt_sriov_disable(struct bnxt *bp);
 void bnxt_hwrm_exec_fwd_req(struct bnxt *);
 void bnxt_update_vf_mac(struct bnxt *);
-int bnxt_approve_mac(struct bnxt *, u8 *);
+int bnxt_approve_mac(struct bnxt *, const u8 *, bool);
 #endif
diff --git a/drivers/net/ethernet/broadcom/bnxt/bnxt_tc.c b/drivers/net/ethernet/broadcom/bnxt/bnxt_tc.c
new file mode 100644
index 000000000000..2d66bf59cd64
--- /dev/null
+++ b/drivers/net/ethernet/broadcom/bnxt/bnxt_tc.c
@@ -0,0 +1,2102 @@
+/* Broadcom NetXtreme-C/E network driver.
+ *
+ * Copyright (c) 2017 Broadcom Limited
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation.
+ */
+
+#include <linux/netdevice.h>
+#include <linux/inetdevice.h>
+#include <linux/if_vlan.h>
+#include <net/flow_dissector.h>
+#include <net/pkt_cls.h>
+#include <net/tc_act/tc_gact.h>
+#include <net/tc_act/tc_skbedit.h>
+#include <net/tc_act/tc_mirred.h>
+#include <net/tc_act/tc_vlan.h>
+#include <net/tc_act/tc_pedit.h>
+#include <net/tc_act/tc_tunnel_key.h>
+#include <net/vxlan.h>
+#include <linux/bnxt/hsi.h>
+
+#include "bnxt.h"
+#include "bnxt_hwrm.h"
+#include "bnxt_sriov.h"
+#include "bnxt_tc.h"
+#include "bnxt_vfr.h"
+
+#define BNXT_FID_INVALID			0xffff
+#define VLAN_TCI(vid, prio)	((vid) | ((prio) << VLAN_PRIO_SHIFT))
+
+#define is_vlan_pcp_wildcarded(vlan_tci_mask)	\
+	((ntohs(vlan_tci_mask) & VLAN_PRIO_MASK) == 0x0000)
+#define is_vlan_pcp_exactmatch(vlan_tci_mask)	\
+	((ntohs(vlan_tci_mask) & VLAN_PRIO_MASK) == VLAN_PRIO_MASK)
+#define is_vlan_pcp_zero(vlan_tci)	\
+	((ntohs(vlan_tci) & VLAN_PRIO_MASK) == 0x0000)
+#define is_vid_exactmatch(vlan_tci_mask)	\
+	((ntohs(vlan_tci_mask) & VLAN_VID_MASK) == VLAN_VID_MASK)
+
+static bool is_wildcard(void *mask, int len);
+static bool is_exactmatch(void *mask, int len);
+/* Return the dst fid of the func for flow forwarding
+ * For PFs: src_fid is the fid of the PF
+ * For VF-reps: src_fid the fid of the VF
+ */
+static u16 bnxt_flow_get_dst_fid(struct bnxt *pf_bp, struct net_device *dev)
+{
+	struct bnxt *bp;
+
+	/* check if dev belongs to the same switch */
+	if (!netdev_port_same_parent_id(pf_bp->dev, dev)) {
+		netdev_info(pf_bp->dev, "dev(ifindex=%d) not on same switch\n",
+			    dev->ifindex);
+		return BNXT_FID_INVALID;
+	}
+
+	/* Is dev a VF-rep? */
+	if (bnxt_dev_is_vf_rep(dev))
+		return bnxt_vf_rep_get_fid(dev);
+
+	bp = netdev_priv(dev);
+	return bp->pf.fw_fid;
+}
+
+static int bnxt_tc_parse_redir(struct bnxt *bp,
+			       struct bnxt_tc_actions *actions,
+			       const struct flow_action_entry *act)
+{
+	struct net_device *dev = act->dev;
+
+	if (!dev) {
+		netdev_info(bp->dev, "no dev in mirred action\n");
+		return -EINVAL;
+	}
+
+	actions->flags |= BNXT_TC_ACTION_FLAG_FWD;
+	actions->dst_dev = dev;
+	return 0;
+}
+
+static int bnxt_tc_parse_vlan(struct bnxt *bp,
+			      struct bnxt_tc_actions *actions,
+			      const struct flow_action_entry *act)
+{
+	switch (act->id) {
+	case FLOW_ACTION_VLAN_POP:
+		actions->flags |= BNXT_TC_ACTION_FLAG_POP_VLAN;
+		break;
+	case FLOW_ACTION_VLAN_PUSH:
+		actions->flags |= BNXT_TC_ACTION_FLAG_PUSH_VLAN;
+		actions->push_vlan_tci = htons(act->vlan.vid);
+		actions->push_vlan_tpid = act->vlan.proto;
+		break;
+	default:
+		return -EOPNOTSUPP;
+	}
+	return 0;
+}
+
+static int bnxt_tc_parse_tunnel_set(struct bnxt *bp,
+				    struct bnxt_tc_actions *actions,
+				    const struct flow_action_entry *act)
+{
+	const struct ip_tunnel_info *tun_info = act->tunnel;
+	const struct ip_tunnel_key *tun_key = &tun_info->key;
+
+	if (ip_tunnel_info_af(tun_info) != AF_INET) {
+		netdev_info(bp->dev, "only IPv4 tunnel-encap is supported\n");
+		return -EOPNOTSUPP;
+	}
+
+	actions->tun_encap_key = *tun_key;
+	actions->flags |= BNXT_TC_ACTION_FLAG_TUNNEL_ENCAP;
+	return 0;
+}
+
+/* Key & Mask from the stack comes unaligned in multiple iterations of 4 bytes
+ * each(u32).
+ * This routine consolidates such multiple unaligned values into one
+ * field each for Key & Mask (for src and dst macs separately)
+ * For example,
+ *			Mask/Key	Offset	Iteration
+ *			==========	======	=========
+ *	dst mac		0xffffffff	0	1
+ *	dst mac		0x0000ffff	4	2
+ *
+ *	src mac		0xffff0000	4	1
+ *	src mac		0xffffffff	8	2
+ *
+ * The above combination coming from the stack will be consolidated as
+ *			Mask/Key
+ *			==============
+ *	src mac:	0xffffffffffff
+ *	dst mac:	0xffffffffffff
+ */
+static void bnxt_set_l2_key_mask(u32 part_key, u32 part_mask,
+				 u8 *actual_key, u8 *actual_mask)
+{
+	u32 key = get_unaligned((u32 *)actual_key);
+	u32 mask = get_unaligned((u32 *)actual_mask);
+
+	part_key &= part_mask;
+	part_key |= key & ~part_mask;
+
+	put_unaligned(mask | part_mask, (u32 *)actual_mask);
+	put_unaligned(part_key, (u32 *)actual_key);
+}
+
+static int
+bnxt_fill_l2_rewrite_fields(struct bnxt_tc_actions *actions,
+			    u16 *eth_addr, u16 *eth_addr_mask)
+{
+	u16 *p;
+	int j;
+
+	if (unlikely(bnxt_eth_addr_key_mask_invalid(eth_addr, eth_addr_mask)))
+		return -EINVAL;
+
+	if (!is_wildcard(&eth_addr_mask[0], ETH_ALEN)) {
+		if (!is_exactmatch(&eth_addr_mask[0], ETH_ALEN))
+			return -EINVAL;
+		/* FW expects dmac to be in u16 array format */
+		p = eth_addr;
+		for (j = 0; j < 3; j++)
+			actions->l2_rewrite_dmac[j] = cpu_to_be16(*(p + j));
+	}
+
+	if (!is_wildcard(&eth_addr_mask[ETH_ALEN / 2], ETH_ALEN)) {
+		if (!is_exactmatch(&eth_addr_mask[ETH_ALEN / 2], ETH_ALEN))
+			return -EINVAL;
+		/* FW expects smac to be in u16 array format */
+		p = &eth_addr[ETH_ALEN / 2];
+		for (j = 0; j < 3; j++)
+			actions->l2_rewrite_smac[j] = cpu_to_be16(*(p + j));
+	}
+
+	return 0;
+}
+
+static int
+bnxt_tc_parse_pedit(struct bnxt *bp, struct bnxt_tc_actions *actions,
+		    struct flow_action_entry *act, int act_idx, u8 *eth_addr,
+		    u8 *eth_addr_mask)
+{
+	size_t offset_of_ip6_daddr = offsetof(struct ipv6hdr, daddr);
+	size_t offset_of_ip6_saddr = offsetof(struct ipv6hdr, saddr);
+	u32 mask, val, offset, idx;
+	u8 htype;
+
+	offset = act->mangle.offset;
+	htype = act->mangle.htype;
+	mask = ~act->mangle.mask;
+	val = act->mangle.val;
+
+	switch (htype) {
+	case FLOW_ACT_MANGLE_HDR_TYPE_ETH:
+		if (offset > PEDIT_OFFSET_SMAC_LAST_4_BYTES) {
+			netdev_err(bp->dev,
+				   "%s: eth_hdr: Invalid pedit field\n",
+				   __func__);
+			return -EINVAL;
+		}
+		actions->flags |= BNXT_TC_ACTION_FLAG_L2_REWRITE;
+
+		bnxt_set_l2_key_mask(val, mask, &eth_addr[offset],
+				     &eth_addr_mask[offset]);
+		break;
+	case FLOW_ACT_MANGLE_HDR_TYPE_IP4:
+		actions->flags |= BNXT_TC_ACTION_FLAG_NAT_XLATE;
+		actions->nat.l3_is_ipv4 = true;
+		if (offset ==  offsetof(struct iphdr, saddr)) {
+			actions->nat.src_xlate = true;
+			actions->nat.l3.ipv4.saddr.s_addr = htonl(val);
+		} else if (offset ==  offsetof(struct iphdr, daddr)) {
+			actions->nat.src_xlate = false;
+			actions->nat.l3.ipv4.daddr.s_addr = htonl(val);
+		} else {
+			netdev_err(bp->dev,
+				   "%s: IPv4_hdr: Invalid pedit field\n",
+				   __func__);
+			return -EINVAL;
+		}
+
+		netdev_dbg(bp->dev, "nat.src_xlate = %d src IP: %pI4 dst ip : %pI4\n",
+			   actions->nat.src_xlate, &actions->nat.l3.ipv4.saddr,
+			   &actions->nat.l3.ipv4.daddr);
+		break;
+
+	case FLOW_ACT_MANGLE_HDR_TYPE_IP6:
+		actions->flags |= BNXT_TC_ACTION_FLAG_NAT_XLATE;
+		actions->nat.l3_is_ipv4 = false;
+		if (offset >= offsetof(struct ipv6hdr, saddr) &&
+		    offset < offset_of_ip6_daddr) {
+			/* 16 byte IPv6 address comes in 4 iterations of
+			 * 4byte chunks each
+			 */
+			actions->nat.src_xlate = true;
+			idx = (offset - offset_of_ip6_saddr) / 4;
+			/* First 4bytes will be copied to idx 0 and so on */
+			actions->nat.l3.ipv6.saddr.s6_addr32[idx] = htonl(val);
+		} else if (offset >= offset_of_ip6_daddr &&
+			   offset < offset_of_ip6_daddr + 16) {
+			actions->nat.src_xlate = false;
+			idx = (offset - offset_of_ip6_daddr) / 4;
+			actions->nat.l3.ipv6.daddr.s6_addr32[idx] = htonl(val);
+		} else {
+			netdev_err(bp->dev,
+				   "%s: IPv6_hdr: Invalid pedit field\n",
+				   __func__);
+			return -EINVAL;
+		}
+		break;
+	case FLOW_ACT_MANGLE_HDR_TYPE_TCP:
+	case FLOW_ACT_MANGLE_HDR_TYPE_UDP:
+		/* HW does not support L4 rewrite alone without L3
+		 * rewrite
+		 */
+		if (!(actions->flags & BNXT_TC_ACTION_FLAG_NAT_XLATE)) {
+			netdev_err(bp->dev,
+				   "Need to specify L3 rewrite as well\n");
+			return -EINVAL;
+		}
+		if (actions->nat.src_xlate)
+			actions->nat.l4.ports.sport = htons(val);
+		else
+			actions->nat.l4.ports.dport = htons(val);
+		netdev_dbg(bp->dev, "actions->nat.sport = %d dport = %d\n",
+			   actions->nat.l4.ports.sport,
+			   actions->nat.l4.ports.dport);
+		break;
+	default:
+		netdev_err(bp->dev, "%s: Unsupported pedit hdr type\n",
+			   __func__);
+		return -EINVAL;
+	}
+	return 0;
+}
+
+static int bnxt_tc_parse_actions(struct bnxt *bp,
+				 struct bnxt_tc_actions *actions,
+				 struct flow_action *flow_action,
+				 struct netlink_ext_ack *extack)
+{
+	/* Used to store the L2 rewrite mask for dmac (6 bytes) followed by
+	 * smac (6 bytes) if rewrite of both is specified, otherwise either
+	 * dmac or smac
+	 */
+	u16 eth_addr_mask[ETH_ALEN] = { 0 };
+	/* Used to store the L2 rewrite key for dmac (6 bytes) followed by
+	 * smac (6 bytes) if rewrite of both is specified, otherwise either
+	 * dmac or smac
+	 */
+	u16 eth_addr[ETH_ALEN] = { 0 };
+	struct flow_action_entry *act;
+	int i, rc;
+
+	if (!flow_action_has_entries(flow_action)) {
+		netdev_info(bp->dev, "no actions\n");
+		return -EINVAL;
+	}
+
+	if (!flow_action_basic_hw_stats_check(flow_action, extack))
+		return -EOPNOTSUPP;
+
+	flow_action_for_each(i, act, flow_action) {
+		switch (act->id) {
+		case FLOW_ACTION_DROP:
+			actions->flags |= BNXT_TC_ACTION_FLAG_DROP;
+			return 0; /* don't bother with other actions */
+		case FLOW_ACTION_REDIRECT:
+			rc = bnxt_tc_parse_redir(bp, actions, act);
+			if (rc)
+				return rc;
+			break;
+		case FLOW_ACTION_VLAN_POP:
+		case FLOW_ACTION_VLAN_PUSH:
+		case FLOW_ACTION_VLAN_MANGLE:
+			rc = bnxt_tc_parse_vlan(bp, actions, act);
+			if (rc)
+				return rc;
+			break;
+		case FLOW_ACTION_TUNNEL_ENCAP:
+			rc = bnxt_tc_parse_tunnel_set(bp, actions, act);
+			if (rc)
+				return rc;
+			break;
+		case FLOW_ACTION_TUNNEL_DECAP:
+			actions->flags |= BNXT_TC_ACTION_FLAG_TUNNEL_DECAP;
+			break;
+		/* Packet edit: L2 rewrite, NAT, NAPT */
+		case FLOW_ACTION_MANGLE:
+			rc = bnxt_tc_parse_pedit(bp, actions, act, i,
+						 (u8 *)eth_addr,
+						 (u8 *)eth_addr_mask);
+			if (rc)
+				return rc;
+			break;
+		default:
+			break;
+		}
+	}
+
+	if (actions->flags & BNXT_TC_ACTION_FLAG_L2_REWRITE) {
+		rc = bnxt_fill_l2_rewrite_fields(actions, eth_addr,
+						 eth_addr_mask);
+		if (rc)
+			return rc;
+	}
+
+	if (actions->flags & BNXT_TC_ACTION_FLAG_FWD) {
+		if (actions->flags & BNXT_TC_ACTION_FLAG_TUNNEL_ENCAP) {
+			/* dst_fid is PF's fid */
+			actions->dst_fid = bp->pf.fw_fid;
+		} else {
+			/* find the FID from dst_dev */
+			actions->dst_fid =
+				bnxt_flow_get_dst_fid(bp, actions->dst_dev);
+			if (actions->dst_fid == BNXT_FID_INVALID)
+				return -EINVAL;
+		}
+	}
+
+	return 0;
+}
+
+static int bnxt_tc_parse_flow(struct bnxt *bp,
+			      struct flow_cls_offload *tc_flow_cmd,
+			      struct bnxt_tc_flow *flow)
+{
+	struct flow_rule *rule = flow_cls_offload_flow_rule(tc_flow_cmd);
+	struct netlink_ext_ack *extack = tc_flow_cmd->common.extack;
+	struct flow_dissector *dissector = rule->match.dissector;
+
+	/* KEY_CONTROL and KEY_BASIC are needed for forming a meaningful key */
+	if ((dissector->used_keys & BIT_ULL(FLOW_DISSECTOR_KEY_CONTROL)) == 0 ||
+	    (dissector->used_keys & BIT_ULL(FLOW_DISSECTOR_KEY_BASIC)) == 0) {
+		netdev_info(bp->dev, "cannot form TC key: used_keys = 0x%llx\n",
+			    dissector->used_keys);
+		return -EOPNOTSUPP;
+	}
+
+	if (flow_rule_match_has_control_flags(rule, extack))
+		return -EOPNOTSUPP;
+
+	if (flow_rule_match_key(rule, FLOW_DISSECTOR_KEY_BASIC)) {
+		struct flow_match_basic match;
+
+		flow_rule_match_basic(rule, &match);
+		flow->l2_key.ether_type = match.key->n_proto;
+		flow->l2_mask.ether_type = match.mask->n_proto;
+
+		if (match.key->n_proto == htons(ETH_P_IP) ||
+		    match.key->n_proto == htons(ETH_P_IPV6)) {
+			flow->l4_key.ip_proto = match.key->ip_proto;
+			flow->l4_mask.ip_proto = match.mask->ip_proto;
+		}
+	}
+
+	if (flow_rule_match_key(rule, FLOW_DISSECTOR_KEY_ETH_ADDRS)) {
+		struct flow_match_eth_addrs match;
+
+		flow_rule_match_eth_addrs(rule, &match);
+		flow->flags |= BNXT_TC_FLOW_FLAGS_ETH_ADDRS;
+		ether_addr_copy(flow->l2_key.dmac, match.key->dst);
+		ether_addr_copy(flow->l2_mask.dmac, match.mask->dst);
+		ether_addr_copy(flow->l2_key.smac, match.key->src);
+		ether_addr_copy(flow->l2_mask.smac, match.mask->src);
+	}
+
+	if (flow_rule_match_key(rule, FLOW_DISSECTOR_KEY_VLAN)) {
+		struct flow_match_vlan match;
+
+		flow_rule_match_vlan(rule, &match);
+		flow->l2_key.inner_vlan_tci =
+			cpu_to_be16(VLAN_TCI(match.key->vlan_id,
+					     match.key->vlan_priority));
+		flow->l2_mask.inner_vlan_tci =
+			cpu_to_be16((VLAN_TCI(match.mask->vlan_id,
+					      match.mask->vlan_priority)));
+		flow->l2_key.inner_vlan_tpid = htons(ETH_P_8021Q);
+		flow->l2_mask.inner_vlan_tpid = htons(0xffff);
+		flow->l2_key.num_vlans = 1;
+	}
+
+	if (flow_rule_match_key(rule, FLOW_DISSECTOR_KEY_IPV4_ADDRS)) {
+		struct flow_match_ipv4_addrs match;
+
+		flow_rule_match_ipv4_addrs(rule, &match);
+		flow->flags |= BNXT_TC_FLOW_FLAGS_IPV4_ADDRS;
+		flow->l3_key.ipv4.daddr.s_addr = match.key->dst;
+		flow->l3_mask.ipv4.daddr.s_addr = match.mask->dst;
+		flow->l3_key.ipv4.saddr.s_addr = match.key->src;
+		flow->l3_mask.ipv4.saddr.s_addr = match.mask->src;
+	} else if (flow_rule_match_key(rule, FLOW_DISSECTOR_KEY_IPV6_ADDRS)) {
+		struct flow_match_ipv6_addrs match;
+
+		flow_rule_match_ipv6_addrs(rule, &match);
+		flow->flags |= BNXT_TC_FLOW_FLAGS_IPV6_ADDRS;
+		flow->l3_key.ipv6.daddr = match.key->dst;
+		flow->l3_mask.ipv6.daddr = match.mask->dst;
+		flow->l3_key.ipv6.saddr = match.key->src;
+		flow->l3_mask.ipv6.saddr = match.mask->src;
+	}
+
+	if (flow_rule_match_key(rule, FLOW_DISSECTOR_KEY_PORTS)) {
+		struct flow_match_ports match;
+
+		flow_rule_match_ports(rule, &match);
+		flow->flags |= BNXT_TC_FLOW_FLAGS_PORTS;
+		flow->l4_key.ports.dport = match.key->dst;
+		flow->l4_mask.ports.dport = match.mask->dst;
+		flow->l4_key.ports.sport = match.key->src;
+		flow->l4_mask.ports.sport = match.mask->src;
+	}
+
+	if (flow_rule_match_key(rule, FLOW_DISSECTOR_KEY_ICMP)) {
+		struct flow_match_icmp match;
+
+		flow_rule_match_icmp(rule, &match);
+		flow->flags |= BNXT_TC_FLOW_FLAGS_ICMP;
+		flow->l4_key.icmp.type = match.key->type;
+		flow->l4_key.icmp.code = match.key->code;
+		flow->l4_mask.icmp.type = match.mask->type;
+		flow->l4_mask.icmp.code = match.mask->code;
+	}
+
+	if (flow_rule_match_key(rule, FLOW_DISSECTOR_KEY_ENC_IPV4_ADDRS)) {
+		struct flow_match_ipv4_addrs match;
+
+		flow_rule_match_enc_ipv4_addrs(rule, &match);
+		flow->flags |= BNXT_TC_FLOW_FLAGS_TUNL_IPV4_ADDRS;
+		flow->tun_key.u.ipv4.dst = match.key->dst;
+		flow->tun_mask.u.ipv4.dst = match.mask->dst;
+		flow->tun_key.u.ipv4.src = match.key->src;
+		flow->tun_mask.u.ipv4.src = match.mask->src;
+	} else if (flow_rule_match_key(rule,
+				      FLOW_DISSECTOR_KEY_ENC_IPV6_ADDRS)) {
+		return -EOPNOTSUPP;
+	}
+
+	if (flow_rule_match_key(rule, FLOW_DISSECTOR_KEY_ENC_KEYID)) {
+		struct flow_match_enc_keyid match;
+
+		flow_rule_match_enc_keyid(rule, &match);
+		flow->flags |= BNXT_TC_FLOW_FLAGS_TUNL_ID;
+		flow->tun_key.tun_id = key32_to_tunnel_id(match.key->keyid);
+		flow->tun_mask.tun_id = key32_to_tunnel_id(match.mask->keyid);
+	}
+
+	if (flow_rule_match_key(rule, FLOW_DISSECTOR_KEY_ENC_PORTS)) {
+		struct flow_match_ports match;
+
+		flow_rule_match_enc_ports(rule, &match);
+		flow->flags |= BNXT_TC_FLOW_FLAGS_TUNL_PORTS;
+		flow->tun_key.tp_dst = match.key->dst;
+		flow->tun_mask.tp_dst = match.mask->dst;
+		flow->tun_key.tp_src = match.key->src;
+		flow->tun_mask.tp_src = match.mask->src;
+	}
+
+	return bnxt_tc_parse_actions(bp, &flow->actions, &rule->action,
+				     tc_flow_cmd->common.extack);
+}
+
+static int bnxt_hwrm_cfa_flow_free(struct bnxt *bp,
+				   struct bnxt_tc_flow_node *flow_node)
+{
+	struct hwrm_cfa_flow_free_input *req;
+	int rc;
+
+	rc = hwrm_req_init(bp, req, HWRM_CFA_FLOW_FREE);
+	if (!rc) {
+		if (bp->fw_cap & BNXT_FW_CAP_OVS_64BIT_HANDLE)
+			req->ext_flow_handle = flow_node->ext_flow_handle;
+		else
+			req->flow_handle = flow_node->flow_handle;
+
+		rc = hwrm_req_send(bp, req);
+	}
+	if (rc)
+		netdev_info(bp->dev, "%s: Error rc=%d\n", __func__, rc);
+
+	return rc;
+}
+
+static int ipv6_mask_len(struct in6_addr *mask)
+{
+	int mask_len = 0, i;
+
+	for (i = 0; i < 4; i++)
+		mask_len += inet_mask_len(mask->s6_addr32[i]);
+
+	return mask_len;
+}
+
+static bool is_wildcard(void *mask, int len)
+{
+	const u8 *p = mask;
+	int i;
+
+	for (i = 0; i < len; i++) {
+		if (p[i] != 0)
+			return false;
+	}
+	return true;
+}
+
+static bool is_exactmatch(void *mask, int len)
+{
+	const u8 *p = mask;
+	int i;
+
+	for (i = 0; i < len; i++)
+		if (p[i] != 0xff)
+			return false;
+
+	return true;
+}
+
+static bool is_vlan_tci_allowed(__be16  vlan_tci_mask,
+				__be16  vlan_tci)
+{
+	/* VLAN priority must be either exactly zero or fully wildcarded and
+	 * VLAN id must be exact match.
+	 */
+	if (is_vid_exactmatch(vlan_tci_mask) &&
+	    ((is_vlan_pcp_exactmatch(vlan_tci_mask) &&
+	      is_vlan_pcp_zero(vlan_tci)) ||
+	     is_vlan_pcp_wildcarded(vlan_tci_mask)))
+		return true;
+
+	return false;
+}
+
+static bool bits_set(void *key, int len)
+{
+	const u8 *p = key;
+	int i;
+
+	for (i = 0; i < len; i++)
+		if (p[i] != 0)
+			return true;
+
+	return false;
+}
+
+static int bnxt_hwrm_cfa_flow_alloc(struct bnxt *bp, struct bnxt_tc_flow *flow,
+				    __le16 ref_flow_handle,
+				    __le32 tunnel_handle,
+				    struct bnxt_tc_flow_node *flow_node)
+{
+	struct bnxt_tc_actions *actions = &flow->actions;
+	struct bnxt_tc_l3_key *l3_mask = &flow->l3_mask;
+	struct bnxt_tc_l3_key *l3_key = &flow->l3_key;
+	struct hwrm_cfa_flow_alloc_output *resp;
+	struct hwrm_cfa_flow_alloc_input *req;
+	u16 flow_flags = 0, action_flags = 0;
+	int rc;
+
+	rc = hwrm_req_init(bp, req, HWRM_CFA_FLOW_ALLOC);
+	if (rc)
+		return rc;
+
+	req->src_fid = cpu_to_le16(flow->src_fid);
+	req->ref_flow_handle = ref_flow_handle;
+
+	if (actions->flags & BNXT_TC_ACTION_FLAG_L2_REWRITE) {
+		memcpy(req->l2_rewrite_dmac, actions->l2_rewrite_dmac,
+		       ETH_ALEN);
+		memcpy(req->l2_rewrite_smac, actions->l2_rewrite_smac,
+		       ETH_ALEN);
+		action_flags |=
+			CFA_FLOW_ALLOC_REQ_ACTION_FLAGS_L2_HEADER_REWRITE;
+	}
+
+	if (actions->flags & BNXT_TC_ACTION_FLAG_NAT_XLATE) {
+		if (actions->nat.l3_is_ipv4) {
+			action_flags |=
+				CFA_FLOW_ALLOC_REQ_ACTION_FLAGS_NAT_IPV4_ADDRESS;
+
+			if (actions->nat.src_xlate) {
+				action_flags |=
+					CFA_FLOW_ALLOC_REQ_ACTION_FLAGS_NAT_SRC;
+				/* L3 source rewrite */
+				req->nat_ip_address[0] =
+					actions->nat.l3.ipv4.saddr.s_addr;
+				/* L4 source port */
+				if (actions->nat.l4.ports.sport)
+					req->nat_port =
+						actions->nat.l4.ports.sport;
+			} else {
+				action_flags |=
+					CFA_FLOW_ALLOC_REQ_ACTION_FLAGS_NAT_DEST;
+				/* L3 destination rewrite */
+				req->nat_ip_address[0] =
+					actions->nat.l3.ipv4.daddr.s_addr;
+				/* L4 destination port */
+				if (actions->nat.l4.ports.dport)
+					req->nat_port =
+						actions->nat.l4.ports.dport;
+			}
+			netdev_dbg(bp->dev,
+				   "req->nat_ip_address: %pI4 src_xlate: %d req->nat_port: %x\n",
+				   req->nat_ip_address, actions->nat.src_xlate,
+				   req->nat_port);
+		} else {
+			if (actions->nat.src_xlate) {
+				action_flags |=
+					CFA_FLOW_ALLOC_REQ_ACTION_FLAGS_NAT_SRC;
+				/* L3 source rewrite */
+				memcpy(req->nat_ip_address,
+				       actions->nat.l3.ipv6.saddr.s6_addr32,
+				       sizeof(req->nat_ip_address));
+				/* L4 source port */
+				if (actions->nat.l4.ports.sport)
+					req->nat_port =
+						actions->nat.l4.ports.sport;
+			} else {
+				action_flags |=
+					CFA_FLOW_ALLOC_REQ_ACTION_FLAGS_NAT_DEST;
+				/* L3 destination rewrite */
+				memcpy(req->nat_ip_address,
+				       actions->nat.l3.ipv6.daddr.s6_addr32,
+				       sizeof(req->nat_ip_address));
+				/* L4 destination port */
+				if (actions->nat.l4.ports.dport)
+					req->nat_port =
+						actions->nat.l4.ports.dport;
+			}
+			netdev_dbg(bp->dev,
+				   "req->nat_ip_address: %pI6 src_xlate: %d req->nat_port: %x\n",
+				   req->nat_ip_address, actions->nat.src_xlate,
+				   req->nat_port);
+		}
+	}
+
+	if (actions->flags & BNXT_TC_ACTION_FLAG_TUNNEL_DECAP ||
+	    actions->flags & BNXT_TC_ACTION_FLAG_TUNNEL_ENCAP) {
+		req->tunnel_handle = tunnel_handle;
+		flow_flags |= CFA_FLOW_ALLOC_REQ_FLAGS_TUNNEL;
+		action_flags |= CFA_FLOW_ALLOC_REQ_ACTION_FLAGS_TUNNEL;
+	}
+
+	req->ethertype = flow->l2_key.ether_type;
+	req->ip_proto = flow->l4_key.ip_proto;
+
+	if (flow->flags & BNXT_TC_FLOW_FLAGS_ETH_ADDRS) {
+		memcpy(req->dmac, flow->l2_key.dmac, ETH_ALEN);
+		memcpy(req->smac, flow->l2_key.smac, ETH_ALEN);
+	}
+
+	if (flow->l2_key.num_vlans > 0) {
+		flow_flags |= CFA_FLOW_ALLOC_REQ_FLAGS_NUM_VLAN_ONE;
+		/* FW expects the inner_vlan_tci value to be set
+		 * in outer_vlan_tci when num_vlans is 1 (which is
+		 * always the case in TC.)
+		 */
+		req->outer_vlan_tci = flow->l2_key.inner_vlan_tci;
+	}
+
+	/* If all IP and L4 fields are wildcarded then this is an L2 flow */
+	if (is_wildcard(l3_mask, sizeof(*l3_mask)) &&
+	    is_wildcard(&flow->l4_mask, sizeof(flow->l4_mask))) {
+		flow_flags |= CFA_FLOW_ALLOC_REQ_FLAGS_FLOWTYPE_L2;
+	} else {
+		flow_flags |= flow->l2_key.ether_type == htons(ETH_P_IP) ?
+				CFA_FLOW_ALLOC_REQ_FLAGS_FLOWTYPE_IPV4 :
+				CFA_FLOW_ALLOC_REQ_FLAGS_FLOWTYPE_IPV6;
+
+		if (flow->flags & BNXT_TC_FLOW_FLAGS_IPV4_ADDRS) {
+			req->ip_dst[0] = l3_key->ipv4.daddr.s_addr;
+			req->ip_dst_mask_len =
+				inet_mask_len(l3_mask->ipv4.daddr.s_addr);
+			req->ip_src[0] = l3_key->ipv4.saddr.s_addr;
+			req->ip_src_mask_len =
+				inet_mask_len(l3_mask->ipv4.saddr.s_addr);
+		} else if (flow->flags & BNXT_TC_FLOW_FLAGS_IPV6_ADDRS) {
+			memcpy(req->ip_dst, l3_key->ipv6.daddr.s6_addr32,
+			       sizeof(req->ip_dst));
+			req->ip_dst_mask_len =
+					ipv6_mask_len(&l3_mask->ipv6.daddr);
+			memcpy(req->ip_src, l3_key->ipv6.saddr.s6_addr32,
+			       sizeof(req->ip_src));
+			req->ip_src_mask_len =
+					ipv6_mask_len(&l3_mask->ipv6.saddr);
+		}
+	}
+
+	if (flow->flags & BNXT_TC_FLOW_FLAGS_PORTS) {
+		req->l4_src_port = flow->l4_key.ports.sport;
+		req->l4_src_port_mask = flow->l4_mask.ports.sport;
+		req->l4_dst_port = flow->l4_key.ports.dport;
+		req->l4_dst_port_mask = flow->l4_mask.ports.dport;
+	} else if (flow->flags & BNXT_TC_FLOW_FLAGS_ICMP) {
+		/* l4 ports serve as type/code when ip_proto is ICMP */
+		req->l4_src_port = htons(flow->l4_key.icmp.type);
+		req->l4_src_port_mask = htons(flow->l4_mask.icmp.type);
+		req->l4_dst_port = htons(flow->l4_key.icmp.code);
+		req->l4_dst_port_mask = htons(flow->l4_mask.icmp.code);
+	}
+	req->flags = cpu_to_le16(flow_flags);
+
+	if (actions->flags & BNXT_TC_ACTION_FLAG_DROP) {
+		action_flags |= CFA_FLOW_ALLOC_REQ_ACTION_FLAGS_DROP;
+	} else {
+		if (actions->flags & BNXT_TC_ACTION_FLAG_FWD) {
+			action_flags |= CFA_FLOW_ALLOC_REQ_ACTION_FLAGS_FWD;
+			req->dst_fid = cpu_to_le16(actions->dst_fid);
+		}
+		if (actions->flags & BNXT_TC_ACTION_FLAG_PUSH_VLAN) {
+			action_flags |=
+			    CFA_FLOW_ALLOC_REQ_ACTION_FLAGS_L2_HEADER_REWRITE;
+			req->l2_rewrite_vlan_tpid = actions->push_vlan_tpid;
+			req->l2_rewrite_vlan_tci = actions->push_vlan_tci;
+			memcpy(&req->l2_rewrite_dmac, &req->dmac, ETH_ALEN);
+			memcpy(&req->l2_rewrite_smac, &req->smac, ETH_ALEN);
+		}
+		if (actions->flags & BNXT_TC_ACTION_FLAG_POP_VLAN) {
+			action_flags |=
+			    CFA_FLOW_ALLOC_REQ_ACTION_FLAGS_L2_HEADER_REWRITE;
+			/* Rewrite config with tpid = 0 implies vlan pop */
+			req->l2_rewrite_vlan_tpid = 0;
+			memcpy(&req->l2_rewrite_dmac, &req->dmac, ETH_ALEN);
+			memcpy(&req->l2_rewrite_smac, &req->smac, ETH_ALEN);
+		}
+	}
+	req->action_flags = cpu_to_le16(action_flags);
+
+	resp = hwrm_req_hold(bp, req);
+	rc = hwrm_req_send_silent(bp, req);
+	if (!rc) {
+		/* CFA_FLOW_ALLOC response interpretation:
+		 *		    fw with	     fw with
+		 *		    16-bit	     64-bit
+		 *		    flow handle      flow handle
+		 *		    ===========	     ===========
+		 * flow_handle      flow handle      flow context id
+		 * ext_flow_handle  INVALID	     flow handle
+		 * flow_id	    INVALID	     flow counter id
+		 */
+		flow_node->flow_handle = resp->flow_handle;
+		if (bp->fw_cap & BNXT_FW_CAP_OVS_64BIT_HANDLE) {
+			flow_node->ext_flow_handle = resp->ext_flow_handle;
+			flow_node->flow_id = resp->flow_id;
+		}
+	}
+	hwrm_req_drop(bp, req);
+	return rc;
+}
+
+static int hwrm_cfa_decap_filter_alloc(struct bnxt *bp,
+				       struct bnxt_tc_flow *flow,
+				       struct bnxt_tc_l2_key *l2_info,
+				       __le32 ref_decap_handle,
+				       __le32 *decap_filter_handle)
+{
+	struct hwrm_cfa_decap_filter_alloc_output *resp;
+	struct ip_tunnel_key *tun_key = &flow->tun_key;
+	struct hwrm_cfa_decap_filter_alloc_input *req;
+	u32 enables = 0;
+	int rc;
+
+	rc = hwrm_req_init(bp, req, HWRM_CFA_DECAP_FILTER_ALLOC);
+	if (rc)
+		goto exit;
+
+	req->flags = cpu_to_le32(CFA_DECAP_FILTER_ALLOC_REQ_FLAGS_OVS_TUNNEL);
+	enables |= CFA_DECAP_FILTER_ALLOC_REQ_ENABLES_TUNNEL_TYPE |
+		   CFA_DECAP_FILTER_ALLOC_REQ_ENABLES_IP_PROTOCOL;
+	req->tunnel_type = CFA_DECAP_FILTER_ALLOC_REQ_TUNNEL_TYPE_VXLAN;
+	req->ip_protocol = CFA_DECAP_FILTER_ALLOC_REQ_IP_PROTOCOL_UDP;
+
+	if (flow->flags & BNXT_TC_FLOW_FLAGS_TUNL_ID) {
+		enables |= CFA_DECAP_FILTER_ALLOC_REQ_ENABLES_TUNNEL_ID;
+		/* tunnel_id is wrongly defined in hsi defn. as __le32 */
+		req->tunnel_id = tunnel_id_to_key32(tun_key->tun_id);
+	}
+
+	if (flow->flags & BNXT_TC_FLOW_FLAGS_TUNL_ETH_ADDRS) {
+		enables |= CFA_DECAP_FILTER_ALLOC_REQ_ENABLES_DST_MACADDR;
+		ether_addr_copy(req->dst_macaddr, l2_info->dmac);
+	}
+	if (l2_info->num_vlans) {
+		enables |= CFA_DECAP_FILTER_ALLOC_REQ_ENABLES_T_IVLAN_VID;
+		req->t_ivlan_vid = l2_info->inner_vlan_tci;
+	}
+
+	enables |= CFA_DECAP_FILTER_ALLOC_REQ_ENABLES_ETHERTYPE;
+	req->ethertype = htons(ETH_P_IP);
+
+	if (flow->flags & BNXT_TC_FLOW_FLAGS_TUNL_IPV4_ADDRS) {
+		enables |= CFA_DECAP_FILTER_ALLOC_REQ_ENABLES_SRC_IPADDR |
+			   CFA_DECAP_FILTER_ALLOC_REQ_ENABLES_DST_IPADDR |
+			   CFA_DECAP_FILTER_ALLOC_REQ_ENABLES_IPADDR_TYPE;
+		req->ip_addr_type =
+			CFA_DECAP_FILTER_ALLOC_REQ_IP_ADDR_TYPE_IPV4;
+		req->dst_ipaddr[0] = tun_key->u.ipv4.dst;
+		req->src_ipaddr[0] = tun_key->u.ipv4.src;
+	}
+
+	if (flow->flags & BNXT_TC_FLOW_FLAGS_TUNL_PORTS) {
+		enables |= CFA_DECAP_FILTER_ALLOC_REQ_ENABLES_DST_PORT;
+		req->dst_port = tun_key->tp_dst;
+	}
+
+	/* Eventhough the decap_handle returned by hwrm_cfa_decap_filter_alloc
+	 * is defined as __le32, l2_ctxt_ref_id is defined in HSI as __le16.
+	 */
+	req->l2_ctxt_ref_id = (__force __le16)ref_decap_handle;
+	req->enables = cpu_to_le32(enables);
+
+	resp = hwrm_req_hold(bp, req);
+	rc = hwrm_req_send_silent(bp, req);
+	if (!rc)
+		*decap_filter_handle = resp->decap_filter_id;
+	hwrm_req_drop(bp, req);
+exit:
+	if (rc)
+		netdev_info(bp->dev, "%s: Error rc=%d\n", __func__, rc);
+
+	return rc;
+}
+
+static int hwrm_cfa_decap_filter_free(struct bnxt *bp,
+				      __le32 decap_filter_handle)
+{
+	struct hwrm_cfa_decap_filter_free_input *req;
+	int rc;
+
+	rc = hwrm_req_init(bp, req, HWRM_CFA_DECAP_FILTER_FREE);
+	if (!rc) {
+		req->decap_filter_id = decap_filter_handle;
+		rc = hwrm_req_send(bp, req);
+	}
+	if (rc)
+		netdev_info(bp->dev, "%s: Error rc=%d\n", __func__, rc);
+
+	return rc;
+}
+
+static int hwrm_cfa_encap_record_alloc(struct bnxt *bp,
+				       struct ip_tunnel_key *encap_key,
+				       struct bnxt_tc_l2_key *l2_info,
+				       __le32 *encap_record_handle)
+{
+	struct hwrm_cfa_encap_record_alloc_output *resp;
+	struct hwrm_cfa_encap_record_alloc_input *req;
+	struct hwrm_cfa_encap_data_vxlan *encap;
+	struct hwrm_vxlan_ipv4_hdr *encap_ipv4;
+	int rc;
+
+	rc = hwrm_req_init(bp, req, HWRM_CFA_ENCAP_RECORD_ALLOC);
+	if (rc)
+		goto exit;
+
+	encap = (struct hwrm_cfa_encap_data_vxlan *)&req->encap_data;
+	req->encap_type = CFA_ENCAP_RECORD_ALLOC_REQ_ENCAP_TYPE_VXLAN;
+	ether_addr_copy(encap->dst_mac_addr, l2_info->dmac);
+	ether_addr_copy(encap->src_mac_addr, l2_info->smac);
+	if (l2_info->num_vlans) {
+		encap->num_vlan_tags = l2_info->num_vlans;
+		encap->ovlan_tci = l2_info->inner_vlan_tci;
+		encap->ovlan_tpid = l2_info->inner_vlan_tpid;
+	}
+
+	encap_ipv4 = (struct hwrm_vxlan_ipv4_hdr *)encap->l3;
+	encap_ipv4->ver_hlen = 4 << VXLAN_IPV4_HDR_VER_HLEN_VERSION_SFT;
+	encap_ipv4->ver_hlen |= 5 << VXLAN_IPV4_HDR_VER_HLEN_HEADER_LENGTH_SFT;
+	encap_ipv4->ttl = encap_key->ttl;
+
+	encap_ipv4->dest_ip_addr = encap_key->u.ipv4.dst;
+	encap_ipv4->src_ip_addr = encap_key->u.ipv4.src;
+	encap_ipv4->protocol = IPPROTO_UDP;
+
+	encap->dst_port = encap_key->tp_dst;
+	encap->vni = tunnel_id_to_key32(encap_key->tun_id);
+
+	resp = hwrm_req_hold(bp, req);
+	rc = hwrm_req_send_silent(bp, req);
+	if (!rc)
+		*encap_record_handle = resp->encap_record_id;
+	hwrm_req_drop(bp, req);
+exit:
+	if (rc)
+		netdev_info(bp->dev, "%s: Error rc=%d\n", __func__, rc);
+
+	return rc;
+}
+
+static int hwrm_cfa_encap_record_free(struct bnxt *bp,
+				      __le32 encap_record_handle)
+{
+	struct hwrm_cfa_encap_record_free_input *req;
+	int rc;
+
+	rc = hwrm_req_init(bp, req, HWRM_CFA_ENCAP_RECORD_FREE);
+	if (!rc) {
+		req->encap_record_id = encap_record_handle;
+		rc = hwrm_req_send(bp, req);
+	}
+	if (rc)
+		netdev_info(bp->dev, "%s: Error rc=%d\n", __func__, rc);
+
+	return rc;
+}
+
+static int bnxt_tc_put_l2_node(struct bnxt *bp,
+			       struct bnxt_tc_flow_node *flow_node)
+{
+	struct bnxt_tc_l2_node *l2_node = flow_node->l2_node;
+	struct bnxt_tc_info *tc_info = bp->tc_info;
+	int rc;
+
+	/* remove flow_node from the L2 shared flow list */
+	list_del(&flow_node->l2_list_node);
+	if (--l2_node->refcount == 0) {
+		rc =  rhashtable_remove_fast(&tc_info->l2_table, &l2_node->node,
+					     tc_info->l2_ht_params);
+		if (rc)
+			netdev_err(bp->dev,
+				   "Error: %s: rhashtable_remove_fast: %d\n",
+				   __func__, rc);
+		kfree_rcu(l2_node, rcu);
+	}
+	return 0;
+}
+
+static struct bnxt_tc_l2_node *
+bnxt_tc_get_l2_node(struct bnxt *bp, struct rhashtable *l2_table,
+		    struct rhashtable_params ht_params,
+		    struct bnxt_tc_l2_key *l2_key)
+{
+	struct bnxt_tc_l2_node *l2_node;
+	int rc;
+
+	l2_node = rhashtable_lookup_fast(l2_table, l2_key, ht_params);
+	if (!l2_node) {
+		l2_node = kzalloc(sizeof(*l2_node), GFP_KERNEL);
+		if (!l2_node) {
+			rc = -ENOMEM;
+			return NULL;
+		}
+
+		l2_node->key = *l2_key;
+		rc = rhashtable_insert_fast(l2_table, &l2_node->node,
+					    ht_params);
+		if (rc) {
+			kfree_rcu(l2_node, rcu);
+			netdev_err(bp->dev,
+				   "Error: %s: rhashtable_insert_fast: %d\n",
+				   __func__, rc);
+			return NULL;
+		}
+		INIT_LIST_HEAD(&l2_node->common_l2_flows);
+	}
+	return l2_node;
+}
+
+/* Get the ref_flow_handle for a flow by checking if there are any other
+ * flows that share the same L2 key as this flow.
+ */
+static int
+bnxt_tc_get_ref_flow_handle(struct bnxt *bp, struct bnxt_tc_flow *flow,
+			    struct bnxt_tc_flow_node *flow_node,
+			    __le16 *ref_flow_handle)
+{
+	struct bnxt_tc_info *tc_info = bp->tc_info;
+	struct bnxt_tc_flow_node *ref_flow_node;
+	struct bnxt_tc_l2_node *l2_node;
+
+	l2_node = bnxt_tc_get_l2_node(bp, &tc_info->l2_table,
+				      tc_info->l2_ht_params,
+				      &flow->l2_key);
+	if (!l2_node)
+		return -1;
+
+	/* If any other flow is using this l2_node, use it's flow_handle
+	 * as the ref_flow_handle
+	 */
+	if (l2_node->refcount > 0) {
+		ref_flow_node = list_first_entry(&l2_node->common_l2_flows,
+						 struct bnxt_tc_flow_node,
+						 l2_list_node);
+		*ref_flow_handle = ref_flow_node->flow_handle;
+	} else {
+		*ref_flow_handle = cpu_to_le16(0xffff);
+	}
+
+	/* Insert the l2_node into the flow_node so that subsequent flows
+	 * with a matching l2 key can use the flow_handle of this flow
+	 * as their ref_flow_handle
+	 */
+	flow_node->l2_node = l2_node;
+	list_add(&flow_node->l2_list_node, &l2_node->common_l2_flows);
+	l2_node->refcount++;
+	return 0;
+}
+
+/* After the flow parsing is done, this routine is used for checking
+ * if there are any aspects of the flow that prevent it from being
+ * offloaded.
+ */
+static bool bnxt_tc_can_offload(struct bnxt *bp, struct bnxt_tc_flow *flow)
+{
+	/* If L4 ports are specified then ip_proto must be TCP or UDP */
+	if ((flow->flags & BNXT_TC_FLOW_FLAGS_PORTS) &&
+	    (flow->l4_key.ip_proto != IPPROTO_TCP &&
+	     flow->l4_key.ip_proto != IPPROTO_UDP)) {
+		netdev_info(bp->dev, "Cannot offload non-TCP/UDP (%d) ports\n",
+			    flow->l4_key.ip_proto);
+		return false;
+	}
+
+	/* Currently source/dest MAC cannot be partial wildcard  */
+	if (bits_set(&flow->l2_key.smac, sizeof(flow->l2_key.smac)) &&
+	    !is_exactmatch(flow->l2_mask.smac, sizeof(flow->l2_mask.smac))) {
+		netdev_info(bp->dev, "Wildcard match unsupported for Source MAC\n");
+		return false;
+	}
+	if (bits_set(&flow->l2_key.dmac, sizeof(flow->l2_key.dmac)) &&
+	    !is_exactmatch(&flow->l2_mask.dmac, sizeof(flow->l2_mask.dmac))) {
+		netdev_info(bp->dev, "Wildcard match unsupported for Dest MAC\n");
+		return false;
+	}
+
+	/* Currently VLAN fields cannot be partial wildcard */
+	if (bits_set(&flow->l2_key.inner_vlan_tci,
+		     sizeof(flow->l2_key.inner_vlan_tci)) &&
+	    !is_vlan_tci_allowed(flow->l2_mask.inner_vlan_tci,
+				 flow->l2_key.inner_vlan_tci)) {
+		netdev_info(bp->dev, "Unsupported VLAN TCI\n");
+		return false;
+	}
+	if (bits_set(&flow->l2_key.inner_vlan_tpid,
+		     sizeof(flow->l2_key.inner_vlan_tpid)) &&
+	    !is_exactmatch(&flow->l2_mask.inner_vlan_tpid,
+			   sizeof(flow->l2_mask.inner_vlan_tpid))) {
+		netdev_info(bp->dev, "Wildcard match unsupported for VLAN TPID\n");
+		return false;
+	}
+
+	/* Currently Ethertype must be set */
+	if (!is_exactmatch(&flow->l2_mask.ether_type,
+			   sizeof(flow->l2_mask.ether_type))) {
+		netdev_info(bp->dev, "Wildcard match unsupported for Ethertype\n");
+		return false;
+	}
+
+	return true;
+}
+
+/* Returns the final refcount of the node on success
+ * or a -ve error code on failure
+ */
+static int bnxt_tc_put_tunnel_node(struct bnxt *bp,
+				   struct rhashtable *tunnel_table,
+				   struct rhashtable_params *ht_params,
+				   struct bnxt_tc_tunnel_node *tunnel_node)
+{
+	int rc;
+
+	if (--tunnel_node->refcount == 0) {
+		rc =  rhashtable_remove_fast(tunnel_table, &tunnel_node->node,
+					     *ht_params);
+		if (rc) {
+			netdev_err(bp->dev, "rhashtable_remove_fast rc=%d\n", rc);
+			rc = -1;
+		}
+		kfree_rcu(tunnel_node, rcu);
+		return rc;
+	} else {
+		return tunnel_node->refcount;
+	}
+}
+
+/* Get (or add) either encap or decap tunnel node from/to the supplied
+ * hash table.
+ */
+static struct bnxt_tc_tunnel_node *
+bnxt_tc_get_tunnel_node(struct bnxt *bp, struct rhashtable *tunnel_table,
+			struct rhashtable_params *ht_params,
+			struct ip_tunnel_key *tun_key)
+{
+	struct bnxt_tc_tunnel_node *tunnel_node;
+	int rc;
+
+	tunnel_node = rhashtable_lookup_fast(tunnel_table, tun_key, *ht_params);
+	if (!tunnel_node) {
+		tunnel_node = kzalloc(sizeof(*tunnel_node), GFP_KERNEL);
+		if (!tunnel_node) {
+			rc = -ENOMEM;
+			goto err;
+		}
+
+		tunnel_node->key = *tun_key;
+		tunnel_node->tunnel_handle = INVALID_TUNNEL_HANDLE;
+		rc = rhashtable_insert_fast(tunnel_table, &tunnel_node->node,
+					    *ht_params);
+		if (rc) {
+			kfree_rcu(tunnel_node, rcu);
+			goto err;
+		}
+	}
+	tunnel_node->refcount++;
+	return tunnel_node;
+err:
+	netdev_info(bp->dev, "error rc=%d\n", rc);
+	return NULL;
+}
+
+static int bnxt_tc_get_ref_decap_handle(struct bnxt *bp,
+					struct bnxt_tc_flow *flow,
+					struct bnxt_tc_l2_key *l2_key,
+					struct bnxt_tc_flow_node *flow_node,
+					__le32 *ref_decap_handle)
+{
+	struct bnxt_tc_info *tc_info = bp->tc_info;
+	struct bnxt_tc_flow_node *ref_flow_node;
+	struct bnxt_tc_l2_node *decap_l2_node;
+
+	decap_l2_node = bnxt_tc_get_l2_node(bp, &tc_info->decap_l2_table,
+					    tc_info->decap_l2_ht_params,
+					    l2_key);
+	if (!decap_l2_node)
+		return -1;
+
+	/* If any other flow is using this decap_l2_node, use it's decap_handle
+	 * as the ref_decap_handle
+	 */
+	if (decap_l2_node->refcount > 0) {
+		ref_flow_node =
+			list_first_entry(&decap_l2_node->common_l2_flows,
+					 struct bnxt_tc_flow_node,
+					 decap_l2_list_node);
+		*ref_decap_handle = ref_flow_node->decap_node->tunnel_handle;
+	} else {
+		*ref_decap_handle = INVALID_TUNNEL_HANDLE;
+	}
+
+	/* Insert the l2_node into the flow_node so that subsequent flows
+	 * with a matching decap l2 key can use the decap_filter_handle of
+	 * this flow as their ref_decap_handle
+	 */
+	flow_node->decap_l2_node = decap_l2_node;
+	list_add(&flow_node->decap_l2_list_node,
+		 &decap_l2_node->common_l2_flows);
+	decap_l2_node->refcount++;
+	return 0;
+}
+
+static void bnxt_tc_put_decap_l2_node(struct bnxt *bp,
+				      struct bnxt_tc_flow_node *flow_node)
+{
+	struct bnxt_tc_l2_node *decap_l2_node = flow_node->decap_l2_node;
+	struct bnxt_tc_info *tc_info = bp->tc_info;
+	int rc;
+
+	/* remove flow_node from the decap L2 sharing flow list */
+	list_del(&flow_node->decap_l2_list_node);
+	if (--decap_l2_node->refcount == 0) {
+		rc =  rhashtable_remove_fast(&tc_info->decap_l2_table,
+					     &decap_l2_node->node,
+					     tc_info->decap_l2_ht_params);
+		if (rc)
+			netdev_err(bp->dev, "rhashtable_remove_fast rc=%d\n", rc);
+		kfree_rcu(decap_l2_node, rcu);
+	}
+}
+
+static void bnxt_tc_put_decap_handle(struct bnxt *bp,
+				     struct bnxt_tc_flow_node *flow_node)
+{
+	__le32 decap_handle = flow_node->decap_node->tunnel_handle;
+	struct bnxt_tc_info *tc_info = bp->tc_info;
+	int rc;
+
+	if (flow_node->decap_l2_node)
+		bnxt_tc_put_decap_l2_node(bp, flow_node);
+
+	rc = bnxt_tc_put_tunnel_node(bp, &tc_info->decap_table,
+				     &tc_info->decap_ht_params,
+				     flow_node->decap_node);
+	if (!rc && decap_handle != INVALID_TUNNEL_HANDLE)
+		hwrm_cfa_decap_filter_free(bp, decap_handle);
+}
+
+static int bnxt_tc_resolve_tunnel_hdrs(struct bnxt *bp,
+				       struct ip_tunnel_key *tun_key,
+				       struct bnxt_tc_l2_key *l2_info)
+{
+#ifdef CONFIG_INET
+	struct net_device *real_dst_dev = bp->dev;
+	struct flowi4 flow = { {0} };
+	struct net_device *dst_dev;
+	struct neighbour *nbr;
+	struct rtable *rt;
+	int rc;
+
+	flow.flowi4_proto = IPPROTO_UDP;
+	flow.fl4_dport = tun_key->tp_dst;
+	flow.daddr = tun_key->u.ipv4.dst;
+
+	rt = ip_route_output_key(dev_net(real_dst_dev), &flow);
+	if (IS_ERR(rt)) {
+		netdev_info(bp->dev, "no route to %pI4b\n", &flow.daddr);
+		return -EOPNOTSUPP;
+	}
+
+	/* The route must either point to the real_dst_dev or a dst_dev that
+	 * uses the real_dst_dev.
+	 */
+	dst_dev = rt->dst.dev;
+	if (is_vlan_dev(dst_dev)) {
+#if IS_ENABLED(CONFIG_VLAN_8021Q)
+		struct vlan_dev_priv *vlan = vlan_dev_priv(dst_dev);
+
+		if (vlan->real_dev != real_dst_dev) {
+			netdev_info(bp->dev,
+				    "dst_dev(%s) doesn't use PF-if(%s)\n",
+				    netdev_name(dst_dev),
+				    netdev_name(real_dst_dev));
+			rc = -EOPNOTSUPP;
+			goto put_rt;
+		}
+		l2_info->inner_vlan_tci = htons(vlan->vlan_id);
+		l2_info->inner_vlan_tpid = vlan->vlan_proto;
+		l2_info->num_vlans = 1;
+#endif
+	} else if (dst_dev != real_dst_dev) {
+		netdev_info(bp->dev,
+			    "dst_dev(%s) for %pI4b is not PF-if(%s)\n",
+			    netdev_name(dst_dev), &flow.daddr,
+			    netdev_name(real_dst_dev));
+		rc = -EOPNOTSUPP;
+		goto put_rt;
+	}
+
+	nbr = dst_neigh_lookup(&rt->dst, &flow.daddr);
+	if (!nbr) {
+		netdev_info(bp->dev, "can't lookup neighbor for %pI4b\n",
+			    &flow.daddr);
+		rc = -EOPNOTSUPP;
+		goto put_rt;
+	}
+
+	tun_key->u.ipv4.src = flow.saddr;
+	tun_key->ttl = ip4_dst_hoplimit(&rt->dst);
+	neigh_ha_snapshot(l2_info->dmac, nbr, dst_dev);
+	ether_addr_copy(l2_info->smac, dst_dev->dev_addr);
+	neigh_release(nbr);
+	ip_rt_put(rt);
+
+	return 0;
+put_rt:
+	ip_rt_put(rt);
+	return rc;
+#else
+	return -EOPNOTSUPP;
+#endif
+}
+
+static int bnxt_tc_get_decap_handle(struct bnxt *bp, struct bnxt_tc_flow *flow,
+				    struct bnxt_tc_flow_node *flow_node,
+				    __le32 *decap_filter_handle)
+{
+	struct ip_tunnel_key *decap_key = &flow->tun_key;
+	struct bnxt_tc_info *tc_info = bp->tc_info;
+	struct bnxt_tc_l2_key l2_info = { {0} };
+	struct bnxt_tc_tunnel_node *decap_node;
+	struct ip_tunnel_key tun_key = { 0 };
+	struct bnxt_tc_l2_key *decap_l2_info;
+	__le32 ref_decap_handle;
+	int rc;
+
+	/* Check if there's another flow using the same tunnel decap.
+	 * If not, add this tunnel to the table and resolve the other
+	 * tunnel header fields. Ignore src_port in the tunnel_key,
+	 * since it is not required for decap filters.
+	 */
+	decap_key->tp_src = 0;
+	decap_node = bnxt_tc_get_tunnel_node(bp, &tc_info->decap_table,
+					     &tc_info->decap_ht_params,
+					     decap_key);
+	if (!decap_node)
+		return -ENOMEM;
+
+	flow_node->decap_node = decap_node;
+
+	if (decap_node->tunnel_handle != INVALID_TUNNEL_HANDLE)
+		goto done;
+
+	/* Resolve the L2 fields for tunnel decap
+	 * Resolve the route for remote vtep (saddr) of the decap key
+	 * Find it's next-hop mac addrs
+	 */
+	tun_key.u.ipv4.dst = flow->tun_key.u.ipv4.src;
+	tun_key.tp_dst = flow->tun_key.tp_dst;
+	rc = bnxt_tc_resolve_tunnel_hdrs(bp, &tun_key, &l2_info);
+	if (rc)
+		goto put_decap;
+
+	decap_l2_info = &decap_node->l2_info;
+	/* decap smac is wildcarded */
+	ether_addr_copy(decap_l2_info->dmac, l2_info.smac);
+	if (l2_info.num_vlans) {
+		decap_l2_info->num_vlans = l2_info.num_vlans;
+		decap_l2_info->inner_vlan_tpid = l2_info.inner_vlan_tpid;
+		decap_l2_info->inner_vlan_tci = l2_info.inner_vlan_tci;
+	}
+	flow->flags |= BNXT_TC_FLOW_FLAGS_TUNL_ETH_ADDRS;
+
+	/* For getting a decap_filter_handle we first need to check if
+	 * there are any other decap flows that share the same tunnel L2
+	 * key and if so, pass that flow's decap_filter_handle as the
+	 * ref_decap_handle for this flow.
+	 */
+	rc = bnxt_tc_get_ref_decap_handle(bp, flow, decap_l2_info, flow_node,
+					  &ref_decap_handle);
+	if (rc)
+		goto put_decap;
+
+	/* Issue the hwrm cmd to allocate a decap filter handle */
+	rc = hwrm_cfa_decap_filter_alloc(bp, flow, decap_l2_info,
+					 ref_decap_handle,
+					 &decap_node->tunnel_handle);
+	if (rc)
+		goto put_decap_l2;
+
+done:
+	*decap_filter_handle = decap_node->tunnel_handle;
+	return 0;
+
+put_decap_l2:
+	bnxt_tc_put_decap_l2_node(bp, flow_node);
+put_decap:
+	bnxt_tc_put_tunnel_node(bp, &tc_info->decap_table,
+				&tc_info->decap_ht_params,
+				flow_node->decap_node);
+	return rc;
+}
+
+static void bnxt_tc_put_encap_handle(struct bnxt *bp,
+				     struct bnxt_tc_tunnel_node *encap_node)
+{
+	__le32 encap_handle = encap_node->tunnel_handle;
+	struct bnxt_tc_info *tc_info = bp->tc_info;
+	int rc;
+
+	rc = bnxt_tc_put_tunnel_node(bp, &tc_info->encap_table,
+				     &tc_info->encap_ht_params, encap_node);
+	if (!rc && encap_handle != INVALID_TUNNEL_HANDLE)
+		hwrm_cfa_encap_record_free(bp, encap_handle);
+}
+
+/* Lookup the tunnel encap table and check if there's an encap_handle
+ * alloc'd already.
+ * If not, query L2 info via a route lookup and issue an encap_record_alloc
+ * cmd to FW.
+ */
+static int bnxt_tc_get_encap_handle(struct bnxt *bp, struct bnxt_tc_flow *flow,
+				    struct bnxt_tc_flow_node *flow_node,
+				    __le32 *encap_handle)
+{
+	struct ip_tunnel_key *encap_key = &flow->actions.tun_encap_key;
+	struct bnxt_tc_info *tc_info = bp->tc_info;
+	struct bnxt_tc_tunnel_node *encap_node;
+	int rc;
+
+	/* Check if there's another flow using the same tunnel encap.
+	 * If not, add this tunnel to the table and resolve the other
+	 * tunnel header fields
+	 */
+	encap_node = bnxt_tc_get_tunnel_node(bp, &tc_info->encap_table,
+					     &tc_info->encap_ht_params,
+					     encap_key);
+	if (!encap_node)
+		return -ENOMEM;
+
+	flow_node->encap_node = encap_node;
+
+	if (encap_node->tunnel_handle != INVALID_TUNNEL_HANDLE)
+		goto done;
+
+	rc = bnxt_tc_resolve_tunnel_hdrs(bp, encap_key, &encap_node->l2_info);
+	if (rc)
+		goto put_encap;
+
+	/* Allocate a new tunnel encap record */
+	rc = hwrm_cfa_encap_record_alloc(bp, encap_key, &encap_node->l2_info,
+					 &encap_node->tunnel_handle);
+	if (rc)
+		goto put_encap;
+
+done:
+	*encap_handle = encap_node->tunnel_handle;
+	return 0;
+
+put_encap:
+	bnxt_tc_put_tunnel_node(bp, &tc_info->encap_table,
+				&tc_info->encap_ht_params, encap_node);
+	return rc;
+}
+
+static void bnxt_tc_put_tunnel_handle(struct bnxt *bp,
+				      struct bnxt_tc_flow *flow,
+				      struct bnxt_tc_flow_node *flow_node)
+{
+	if (flow->actions.flags & BNXT_TC_ACTION_FLAG_TUNNEL_DECAP)
+		bnxt_tc_put_decap_handle(bp, flow_node);
+	else if (flow->actions.flags & BNXT_TC_ACTION_FLAG_TUNNEL_ENCAP)
+		bnxt_tc_put_encap_handle(bp, flow_node->encap_node);
+}
+
+static int bnxt_tc_get_tunnel_handle(struct bnxt *bp,
+				     struct bnxt_tc_flow *flow,
+				     struct bnxt_tc_flow_node *flow_node,
+				     __le32 *tunnel_handle)
+{
+	if (flow->actions.flags & BNXT_TC_ACTION_FLAG_TUNNEL_DECAP)
+		return bnxt_tc_get_decap_handle(bp, flow, flow_node,
+						tunnel_handle);
+	else if (flow->actions.flags & BNXT_TC_ACTION_FLAG_TUNNEL_ENCAP)
+		return bnxt_tc_get_encap_handle(bp, flow, flow_node,
+						tunnel_handle);
+	else
+		return 0;
+}
+static int __bnxt_tc_del_flow(struct bnxt *bp,
+			      struct bnxt_tc_flow_node *flow_node)
+{
+	struct bnxt_tc_info *tc_info = bp->tc_info;
+	int rc;
+
+	/* send HWRM cmd to free the flow-id */
+	bnxt_hwrm_cfa_flow_free(bp, flow_node);
+
+	mutex_lock(&tc_info->lock);
+
+	/* release references to any tunnel encap/decap nodes */
+	bnxt_tc_put_tunnel_handle(bp, &flow_node->flow, flow_node);
+
+	/* release reference to l2 node */
+	bnxt_tc_put_l2_node(bp, flow_node);
+
+	mutex_unlock(&tc_info->lock);
+
+	rc = rhashtable_remove_fast(&tc_info->flow_table, &flow_node->node,
+				    tc_info->flow_ht_params);
+	if (rc)
+		netdev_err(bp->dev, "Error: %s: rhashtable_remove_fast rc=%d\n",
+			   __func__, rc);
+
+	kfree_rcu(flow_node, rcu);
+	return 0;
+}
+
+static void bnxt_tc_set_flow_dir(struct bnxt *bp, struct bnxt_tc_flow *flow,
+				 u16 src_fid)
+{
+	flow->l2_key.dir = (bp->pf.fw_fid == src_fid) ? BNXT_DIR_RX : BNXT_DIR_TX;
+}
+
+static void bnxt_tc_set_src_fid(struct bnxt *bp, struct bnxt_tc_flow *flow,
+				u16 src_fid)
+{
+	if (flow->actions.flags & BNXT_TC_ACTION_FLAG_TUNNEL_DECAP)
+		flow->src_fid = bp->pf.fw_fid;
+	else
+		flow->src_fid = src_fid;
+}
+
+/* Add a new flow or replace an existing flow.
+ * Notes on locking:
+ * There are essentially two critical sections here.
+ * 1. while adding a new flow
+ *    a) lookup l2-key
+ *    b) issue HWRM cmd and get flow_handle
+ *    c) link l2-key with flow
+ * 2. while deleting a flow
+ *    a) unlinking l2-key from flow
+ * A lock is needed to protect these two critical sections.
+ *
+ * The hash-tables are already protected by the rhashtable API.
+ */
+static int bnxt_tc_add_flow(struct bnxt *bp, u16 src_fid,
+			    struct flow_cls_offload *tc_flow_cmd)
+{
+	struct bnxt_tc_flow_node *new_node, *old_node;
+	struct bnxt_tc_info *tc_info = bp->tc_info;
+	struct bnxt_tc_flow *flow;
+	__le32 tunnel_handle = 0;
+	__le16 ref_flow_handle;
+	int rc;
+
+	/* allocate memory for the new flow and it's node */
+	new_node = kzalloc(sizeof(*new_node), GFP_KERNEL);
+	if (!new_node) {
+		rc = -ENOMEM;
+		goto done;
+	}
+	new_node->cookie = tc_flow_cmd->cookie;
+	flow = &new_node->flow;
+
+	rc = bnxt_tc_parse_flow(bp, tc_flow_cmd, flow);
+	if (rc)
+		goto free_node;
+
+	bnxt_tc_set_src_fid(bp, flow, src_fid);
+	bnxt_tc_set_flow_dir(bp, flow, flow->src_fid);
+
+	if (!bnxt_tc_can_offload(bp, flow)) {
+		rc = -EOPNOTSUPP;
+		kfree_rcu(new_node, rcu);
+		return rc;
+	}
+
+	/* If a flow exists with the same cookie, delete it */
+	old_node = rhashtable_lookup_fast(&tc_info->flow_table,
+					  &tc_flow_cmd->cookie,
+					  tc_info->flow_ht_params);
+	if (old_node)
+		__bnxt_tc_del_flow(bp, old_node);
+
+	/* Check if the L2 part of the flow has been offloaded already.
+	 * If so, bump up it's refcnt and get it's reference handle.
+	 */
+	mutex_lock(&tc_info->lock);
+	rc = bnxt_tc_get_ref_flow_handle(bp, flow, new_node, &ref_flow_handle);
+	if (rc)
+		goto unlock;
+
+	/* If the flow involves tunnel encap/decap, get tunnel_handle */
+	rc = bnxt_tc_get_tunnel_handle(bp, flow, new_node, &tunnel_handle);
+	if (rc)
+		goto put_l2;
+
+	/* send HWRM cmd to alloc the flow */
+	rc = bnxt_hwrm_cfa_flow_alloc(bp, flow, ref_flow_handle,
+				      tunnel_handle, new_node);
+	if (rc)
+		goto put_tunnel;
+
+	flow->lastused = jiffies;
+	spin_lock_init(&flow->stats_lock);
+	/* add new flow to flow-table */
+	rc = rhashtable_insert_fast(&tc_info->flow_table, &new_node->node,
+				    tc_info->flow_ht_params);
+	if (rc)
+		goto hwrm_flow_free;
+
+	mutex_unlock(&tc_info->lock);
+	return 0;
+
+hwrm_flow_free:
+	bnxt_hwrm_cfa_flow_free(bp, new_node);
+put_tunnel:
+	bnxt_tc_put_tunnel_handle(bp, flow, new_node);
+put_l2:
+	bnxt_tc_put_l2_node(bp, new_node);
+unlock:
+	mutex_unlock(&tc_info->lock);
+free_node:
+	kfree_rcu(new_node, rcu);
+done:
+	netdev_err(bp->dev, "Error: %s: cookie=0x%lx error=%d\n",
+		   __func__, tc_flow_cmd->cookie, rc);
+	return rc;
+}
+
+static int bnxt_tc_del_flow(struct bnxt *bp,
+			    struct flow_cls_offload *tc_flow_cmd)
+{
+	struct bnxt_tc_info *tc_info = bp->tc_info;
+	struct bnxt_tc_flow_node *flow_node;
+
+	flow_node = rhashtable_lookup_fast(&tc_info->flow_table,
+					   &tc_flow_cmd->cookie,
+					   tc_info->flow_ht_params);
+	if (!flow_node)
+		return -EINVAL;
+
+	return __bnxt_tc_del_flow(bp, flow_node);
+}
+
+static int bnxt_tc_get_flow_stats(struct bnxt *bp,
+				  struct flow_cls_offload *tc_flow_cmd)
+{
+	struct bnxt_tc_flow_stats stats, *curr_stats, *prev_stats;
+	struct bnxt_tc_info *tc_info = bp->tc_info;
+	struct bnxt_tc_flow_node *flow_node;
+	struct bnxt_tc_flow *flow;
+	unsigned long lastused;
+
+	flow_node = rhashtable_lookup_fast(&tc_info->flow_table,
+					   &tc_flow_cmd->cookie,
+					   tc_info->flow_ht_params);
+	if (!flow_node)
+		return -1;
+
+	flow = &flow_node->flow;
+	curr_stats = &flow->stats;
+	prev_stats = &flow->prev_stats;
+
+	spin_lock(&flow->stats_lock);
+	stats.packets = curr_stats->packets - prev_stats->packets;
+	stats.bytes = curr_stats->bytes - prev_stats->bytes;
+	*prev_stats = *curr_stats;
+	lastused = flow->lastused;
+	spin_unlock(&flow->stats_lock);
+
+	flow_stats_update(&tc_flow_cmd->stats, stats.bytes, stats.packets, 0,
+			  lastused, FLOW_ACTION_HW_STATS_DELAYED);
+	return 0;
+}
+
+static void bnxt_fill_cfa_stats_req(struct bnxt *bp,
+				    struct bnxt_tc_flow_node *flow_node,
+				    __le16 *flow_handle, __le32 *flow_id)
+{
+	u16 handle;
+
+	if (bp->fw_cap & BNXT_FW_CAP_OVS_64BIT_HANDLE) {
+		*flow_id = flow_node->flow_id;
+
+		/* If flow_id is used to fetch flow stats then:
+		 * 1. lower 12 bits of flow_handle must be set to all 1s.
+		 * 2. 15th bit of flow_handle must specify the flow
+		 *    direction (TX/RX).
+		 */
+		if (flow_node->flow.l2_key.dir == BNXT_DIR_RX)
+			handle = CFA_FLOW_INFO_REQ_FLOW_HANDLE_DIR_RX |
+				 CFA_FLOW_INFO_REQ_FLOW_HANDLE_MAX_MASK;
+		else
+			handle = CFA_FLOW_INFO_REQ_FLOW_HANDLE_MAX_MASK;
+
+		*flow_handle = cpu_to_le16(handle);
+	} else {
+		*flow_handle = flow_node->flow_handle;
+	}
+}
+
+static int
+bnxt_hwrm_cfa_flow_stats_get(struct bnxt *bp, int num_flows,
+			     struct bnxt_tc_stats_batch stats_batch[])
+{
+	struct hwrm_cfa_flow_stats_output *resp;
+	struct hwrm_cfa_flow_stats_input *req;
+	__le16 *req_flow_handles;
+	__le32 *req_flow_ids;
+	int rc, i;
+
+	rc = hwrm_req_init(bp, req, HWRM_CFA_FLOW_STATS);
+	if (rc)
+		goto exit;
+
+	req_flow_handles = &req->flow_handle_0;
+	req_flow_ids = &req->flow_id_0;
+
+	req->num_flows = cpu_to_le16(num_flows);
+	for (i = 0; i < num_flows; i++) {
+		struct bnxt_tc_flow_node *flow_node = stats_batch[i].flow_node;
+
+		bnxt_fill_cfa_stats_req(bp, flow_node,
+					&req_flow_handles[i], &req_flow_ids[i]);
+	}
+
+	resp = hwrm_req_hold(bp, req);
+	rc = hwrm_req_send(bp, req);
+	if (!rc) {
+		__le64 *resp_packets;
+		__le64 *resp_bytes;
+
+		resp_packets = &resp->packet_0;
+		resp_bytes = &resp->byte_0;
+
+		for (i = 0; i < num_flows; i++) {
+			stats_batch[i].hw_stats.packets =
+						le64_to_cpu(resp_packets[i]);
+			stats_batch[i].hw_stats.bytes =
+						le64_to_cpu(resp_bytes[i]);
+		}
+	}
+	hwrm_req_drop(bp, req);
+exit:
+	if (rc)
+		netdev_info(bp->dev, "error rc=%d\n", rc);
+
+	return rc;
+}
+
+/* Add val to accum while handling a possible wraparound
+ * of val. Eventhough val is of type u64, its actual width
+ * is denoted by mask and will wrap-around beyond that width.
+ */
+static void accumulate_val(u64 *accum, u64 val, u64 mask)
+{
+#define low_bits(x, mask)		((x) & (mask))
+#define high_bits(x, mask)		((x) & ~(mask))
+	bool wrapped = val < low_bits(*accum, mask);
+
+	*accum = high_bits(*accum, mask) + val;
+	if (wrapped)
+		*accum += (mask + 1);
+}
+
+/* The HW counters' width is much less than 64bits.
+ * Handle possible wrap-around while updating the stat counters
+ */
+static void bnxt_flow_stats_accum(struct bnxt_tc_info *tc_info,
+				  struct bnxt_tc_flow_stats *acc_stats,
+				  struct bnxt_tc_flow_stats *hw_stats)
+{
+	accumulate_val(&acc_stats->bytes, hw_stats->bytes, tc_info->bytes_mask);
+	accumulate_val(&acc_stats->packets, hw_stats->packets,
+		       tc_info->packets_mask);
+}
+
+static int
+bnxt_tc_flow_stats_batch_update(struct bnxt *bp, int num_flows,
+				struct bnxt_tc_stats_batch stats_batch[])
+{
+	struct bnxt_tc_info *tc_info = bp->tc_info;
+	int rc, i;
+
+	rc = bnxt_hwrm_cfa_flow_stats_get(bp, num_flows, stats_batch);
+	if (rc)
+		return rc;
+
+	for (i = 0; i < num_flows; i++) {
+		struct bnxt_tc_flow_node *flow_node = stats_batch[i].flow_node;
+		struct bnxt_tc_flow *flow = &flow_node->flow;
+
+		spin_lock(&flow->stats_lock);
+		bnxt_flow_stats_accum(tc_info, &flow->stats,
+				      &stats_batch[i].hw_stats);
+		if (flow->stats.packets != flow->prev_stats.packets)
+			flow->lastused = jiffies;
+		spin_unlock(&flow->stats_lock);
+	}
+
+	return 0;
+}
+
+static int
+bnxt_tc_flow_stats_batch_prep(struct bnxt *bp,
+			      struct bnxt_tc_stats_batch stats_batch[],
+			      int *num_flows)
+{
+	struct bnxt_tc_info *tc_info = bp->tc_info;
+	struct rhashtable_iter *iter = &tc_info->iter;
+	void *flow_node;
+	int rc, i;
+
+	rhashtable_walk_start(iter);
+
+	rc = 0;
+	for (i = 0; i < BNXT_FLOW_STATS_BATCH_MAX; i++) {
+		flow_node = rhashtable_walk_next(iter);
+		if (IS_ERR(flow_node)) {
+			i = 0;
+			if (PTR_ERR(flow_node) == -EAGAIN) {
+				continue;
+			} else {
+				rc = PTR_ERR(flow_node);
+				goto done;
+			}
+		}
+
+		/* No more flows */
+		if (!flow_node)
+			goto done;
+
+		stats_batch[i].flow_node = flow_node;
+	}
+done:
+	rhashtable_walk_stop(iter);
+	*num_flows = i;
+	return rc;
+}
+
+void bnxt_tc_flow_stats_work(struct bnxt *bp)
+{
+	struct bnxt_tc_info *tc_info = bp->tc_info;
+	int num_flows, rc;
+
+	num_flows = atomic_read(&tc_info->flow_table.nelems);
+	if (!num_flows)
+		return;
+
+	rhashtable_walk_enter(&tc_info->flow_table, &tc_info->iter);
+
+	for (;;) {
+		rc = bnxt_tc_flow_stats_batch_prep(bp, tc_info->stats_batch,
+						   &num_flows);
+		if (rc) {
+			if (rc == -EAGAIN)
+				continue;
+			break;
+		}
+
+		if (!num_flows)
+			break;
+
+		bnxt_tc_flow_stats_batch_update(bp, num_flows,
+						tc_info->stats_batch);
+	}
+
+	rhashtable_walk_exit(&tc_info->iter);
+}
+
+int bnxt_tc_setup_flower(struct bnxt *bp, u16 src_fid,
+			 struct flow_cls_offload *cls_flower)
+{
+	switch (cls_flower->command) {
+	case FLOW_CLS_REPLACE:
+		return bnxt_tc_add_flow(bp, src_fid, cls_flower);
+	case FLOW_CLS_DESTROY:
+		return bnxt_tc_del_flow(bp, cls_flower);
+	case FLOW_CLS_STATS:
+		return bnxt_tc_get_flow_stats(bp, cls_flower);
+	default:
+		return -EOPNOTSUPP;
+	}
+}
+
+static int bnxt_tc_setup_indr_block_cb(enum tc_setup_type type,
+				       void *type_data, void *cb_priv)
+{
+	struct bnxt_flower_indr_block_cb_priv *priv = cb_priv;
+	struct flow_cls_offload *flower = type_data;
+	struct bnxt *bp = priv->bp;
+
+	if (!tc_cls_can_offload_and_chain0(bp->dev, type_data))
+		return -EOPNOTSUPP;
+
+	switch (type) {
+	case TC_SETUP_CLSFLOWER:
+		return bnxt_tc_setup_flower(bp, bp->pf.fw_fid, flower);
+	default:
+		return -EOPNOTSUPP;
+	}
+}
+
+static struct bnxt_flower_indr_block_cb_priv *
+bnxt_tc_indr_block_cb_lookup(struct bnxt *bp, struct net_device *netdev)
+{
+	struct bnxt_flower_indr_block_cb_priv *cb_priv;
+
+	list_for_each_entry(cb_priv, &bp->tc_indr_block_list, list)
+		if (cb_priv->tunnel_netdev == netdev)
+			return cb_priv;
+
+	return NULL;
+}
+
+static void bnxt_tc_setup_indr_rel(void *cb_priv)
+{
+	struct bnxt_flower_indr_block_cb_priv *priv = cb_priv;
+
+	list_del(&priv->list);
+	kfree(priv);
+}
+
+static int bnxt_tc_setup_indr_block(struct net_device *netdev, struct Qdisc *sch, struct bnxt *bp,
+				    struct flow_block_offload *f, void *data,
+				    void (*cleanup)(struct flow_block_cb *block_cb))
+{
+	struct bnxt_flower_indr_block_cb_priv *cb_priv;
+	struct flow_block_cb *block_cb;
+
+	if (f->binder_type != FLOW_BLOCK_BINDER_TYPE_CLSACT_INGRESS)
+		return -EOPNOTSUPP;
+
+	switch (f->command) {
+	case FLOW_BLOCK_BIND:
+		cb_priv = kmalloc(sizeof(*cb_priv), GFP_KERNEL);
+		if (!cb_priv)
+			return -ENOMEM;
+
+		cb_priv->tunnel_netdev = netdev;
+		cb_priv->bp = bp;
+		list_add(&cb_priv->list, &bp->tc_indr_block_list);
+
+		block_cb = flow_indr_block_cb_alloc(bnxt_tc_setup_indr_block_cb,
+						    cb_priv, cb_priv,
+						    bnxt_tc_setup_indr_rel, f,
+						    netdev, sch, data, bp, cleanup);
+		if (IS_ERR(block_cb)) {
+			list_del(&cb_priv->list);
+			kfree(cb_priv);
+			return PTR_ERR(block_cb);
+		}
+
+		flow_block_cb_add(block_cb, f);
+		list_add_tail(&block_cb->driver_list, &bnxt_block_cb_list);
+		break;
+	case FLOW_BLOCK_UNBIND:
+		cb_priv = bnxt_tc_indr_block_cb_lookup(bp, netdev);
+		if (!cb_priv)
+			return -ENOENT;
+
+		block_cb = flow_block_cb_lookup(f->block,
+						bnxt_tc_setup_indr_block_cb,
+						cb_priv);
+		if (!block_cb)
+			return -ENOENT;
+
+		flow_indr_block_cb_remove(block_cb, f);
+		list_del(&block_cb->driver_list);
+		break;
+	default:
+		return -EOPNOTSUPP;
+	}
+	return 0;
+}
+
+static bool bnxt_is_netdev_indr_offload(struct net_device *netdev)
+{
+	return netif_is_vxlan(netdev);
+}
+
+static int bnxt_tc_setup_indr_cb(struct net_device *netdev, struct Qdisc *sch, void *cb_priv,
+				 enum tc_setup_type type, void *type_data,
+				 void *data,
+				 void (*cleanup)(struct flow_block_cb *block_cb))
+{
+	if (!netdev || !bnxt_is_netdev_indr_offload(netdev))
+		return -EOPNOTSUPP;
+
+	switch (type) {
+	case TC_SETUP_BLOCK:
+		return bnxt_tc_setup_indr_block(netdev, sch, cb_priv, type_data, data, cleanup);
+	default:
+		break;
+	}
+
+	return -EOPNOTSUPP;
+}
+
+static const struct rhashtable_params bnxt_tc_flow_ht_params = {
+	.head_offset = offsetof(struct bnxt_tc_flow_node, node),
+	.key_offset = offsetof(struct bnxt_tc_flow_node, cookie),
+	.key_len = sizeof(((struct bnxt_tc_flow_node *)0)->cookie),
+	.automatic_shrinking = true
+};
+
+static const struct rhashtable_params bnxt_tc_l2_ht_params = {
+	.head_offset = offsetof(struct bnxt_tc_l2_node, node),
+	.key_offset = offsetof(struct bnxt_tc_l2_node, key),
+	.key_len = BNXT_TC_L2_KEY_LEN,
+	.automatic_shrinking = true
+};
+
+static const struct rhashtable_params bnxt_tc_decap_l2_ht_params = {
+	.head_offset = offsetof(struct bnxt_tc_l2_node, node),
+	.key_offset = offsetof(struct bnxt_tc_l2_node, key),
+	.key_len = BNXT_TC_L2_KEY_LEN,
+	.automatic_shrinking = true
+};
+
+static const struct rhashtable_params bnxt_tc_tunnel_ht_params = {
+	.head_offset = offsetof(struct bnxt_tc_tunnel_node, node),
+	.key_offset = offsetof(struct bnxt_tc_tunnel_node, key),
+	.key_len = sizeof(struct ip_tunnel_key),
+	.automatic_shrinking = true
+};
+
+/* convert counter width in bits to a mask */
+#define mask(width)		((u64)~0 >> (64 - (width)))
+
+int bnxt_init_tc(struct bnxt *bp)
+{
+	struct bnxt_tc_info *tc_info;
+	int rc;
+
+	if (bp->hwrm_spec_code < 0x10803)
+		return 0;
+
+	tc_info = kzalloc(sizeof(*tc_info), GFP_KERNEL);
+	if (!tc_info)
+		return -ENOMEM;
+	mutex_init(&tc_info->lock);
+
+	/* Counter widths are programmed by FW */
+	tc_info->bytes_mask = mask(36);
+	tc_info->packets_mask = mask(28);
+
+	tc_info->flow_ht_params = bnxt_tc_flow_ht_params;
+	rc = rhashtable_init(&tc_info->flow_table, &tc_info->flow_ht_params);
+	if (rc)
+		goto free_tc_info;
+
+	tc_info->l2_ht_params = bnxt_tc_l2_ht_params;
+	rc = rhashtable_init(&tc_info->l2_table, &tc_info->l2_ht_params);
+	if (rc)
+		goto destroy_flow_table;
+
+	tc_info->decap_l2_ht_params = bnxt_tc_decap_l2_ht_params;
+	rc = rhashtable_init(&tc_info->decap_l2_table,
+			     &tc_info->decap_l2_ht_params);
+	if (rc)
+		goto destroy_l2_table;
+
+	tc_info->decap_ht_params = bnxt_tc_tunnel_ht_params;
+	rc = rhashtable_init(&tc_info->decap_table,
+			     &tc_info->decap_ht_params);
+	if (rc)
+		goto destroy_decap_l2_table;
+
+	tc_info->encap_ht_params = bnxt_tc_tunnel_ht_params;
+	rc = rhashtable_init(&tc_info->encap_table,
+			     &tc_info->encap_ht_params);
+	if (rc)
+		goto destroy_decap_table;
+
+	tc_info->enabled = true;
+	bp->dev->hw_features |= NETIF_F_HW_TC;
+	bp->dev->features |= NETIF_F_HW_TC;
+	bp->tc_info = tc_info;
+
+	/* init indirect block notifications */
+	INIT_LIST_HEAD(&bp->tc_indr_block_list);
+
+	rc = flow_indr_dev_register(bnxt_tc_setup_indr_cb, bp);
+	if (!rc)
+		return 0;
+
+	rhashtable_destroy(&tc_info->encap_table);
+
+destroy_decap_table:
+	rhashtable_destroy(&tc_info->decap_table);
+destroy_decap_l2_table:
+	rhashtable_destroy(&tc_info->decap_l2_table);
+destroy_l2_table:
+	rhashtable_destroy(&tc_info->l2_table);
+destroy_flow_table:
+	rhashtable_destroy(&tc_info->flow_table);
+free_tc_info:
+	kfree(tc_info);
+	bp->tc_info = NULL;
+	return rc;
+}
+
+void bnxt_shutdown_tc(struct bnxt *bp)
+{
+	struct bnxt_tc_info *tc_info = bp->tc_info;
+
+	if (!bnxt_tc_flower_enabled(bp))
+		return;
+
+	flow_indr_dev_unregister(bnxt_tc_setup_indr_cb, bp,
+				 bnxt_tc_setup_indr_rel);
+	rhashtable_destroy(&tc_info->flow_table);
+	rhashtable_destroy(&tc_info->l2_table);
+	rhashtable_destroy(&tc_info->decap_l2_table);
+	rhashtable_destroy(&tc_info->decap_table);
+	rhashtable_destroy(&tc_info->encap_table);
+	kfree(tc_info);
+	bp->tc_info = NULL;
+}
diff --git a/drivers/net/ethernet/broadcom/bnxt/bnxt_tc.h b/drivers/net/ethernet/broadcom/bnxt/bnxt_tc.h
new file mode 100644
index 000000000000..10c62b094914
--- /dev/null
+++ b/drivers/net/ethernet/broadcom/bnxt/bnxt_tc.h
@@ -0,0 +1,255 @@
+/* Broadcom NetXtreme-C/E network driver.
+ *
+ * Copyright (c) 2017 Broadcom Limited
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation.
+ */
+
+#ifndef BNXT_TC_H
+#define BNXT_TC_H
+
+#ifdef CONFIG_BNXT_FLOWER_OFFLOAD
+
+#include <net/ip_tunnels.h>
+
+/* Structs used for storing the filter/actions of the TC cmd.
+ */
+struct bnxt_tc_l2_key {
+	u8		dmac[ETH_ALEN];
+	u8		smac[ETH_ALEN];
+	__be16		inner_vlan_tpid;
+	__be16		inner_vlan_tci;
+	__be16		ether_type;
+	u8		num_vlans;
+	u8		dir;
+#define BNXT_DIR_RX	1
+#define BNXT_DIR_TX	0
+};
+
+struct bnxt_tc_l3_key {
+	union {
+		struct {
+			struct in_addr daddr;
+			struct in_addr saddr;
+		} ipv4;
+		struct {
+			struct in6_addr daddr;
+			struct in6_addr saddr;
+		} ipv6;
+	};
+};
+
+struct bnxt_tc_l4_key {
+	u8  ip_proto;
+	union {
+		struct {
+			__be16 sport;
+			__be16 dport;
+		} ports;
+		struct {
+			u8 type;
+			u8 code;
+		} icmp;
+	};
+};
+
+struct bnxt_tc_tunnel_key {
+	struct bnxt_tc_l2_key	l2;
+	struct bnxt_tc_l3_key	l3;
+	struct bnxt_tc_l4_key	l4;
+	__be32			id;
+};
+
+#define bnxt_eth_addr_key_mask_invalid(eth_addr, eth_addr_mask)		\
+	((is_wildcard(&(eth_addr)[0], ETH_ALEN) &&			\
+	 is_wildcard(&(eth_addr)[ETH_ALEN / 2], ETH_ALEN)) ||		\
+	(is_wildcard(&(eth_addr_mask)[0], ETH_ALEN) &&			\
+	 is_wildcard(&(eth_addr_mask)[ETH_ALEN / 2], ETH_ALEN)))
+
+struct bnxt_tc_actions {
+	u32				flags;
+#define BNXT_TC_ACTION_FLAG_FWD			BIT(0)
+#define BNXT_TC_ACTION_FLAG_FWD_VXLAN		BIT(1)
+#define BNXT_TC_ACTION_FLAG_PUSH_VLAN		BIT(3)
+#define BNXT_TC_ACTION_FLAG_POP_VLAN		BIT(4)
+#define BNXT_TC_ACTION_FLAG_DROP		BIT(5)
+#define BNXT_TC_ACTION_FLAG_TUNNEL_ENCAP	BIT(6)
+#define BNXT_TC_ACTION_FLAG_TUNNEL_DECAP	BIT(7)
+#define BNXT_TC_ACTION_FLAG_L2_REWRITE		BIT(8)
+#define BNXT_TC_ACTION_FLAG_NAT_XLATE		BIT(9)
+
+	u16				dst_fid;
+	struct net_device		*dst_dev;
+	__be16				push_vlan_tpid;
+	__be16				push_vlan_tci;
+
+	/* tunnel encap */
+	struct ip_tunnel_key		tun_encap_key;
+#define	PEDIT_OFFSET_SMAC_LAST_4_BYTES		0x8
+	__be16				l2_rewrite_dmac[3];
+	__be16				l2_rewrite_smac[3];
+	struct {
+		bool src_xlate;  /* true => translate src,
+				  * false => translate dst
+				  * Mutually exclusive, i.e cannot set both
+				  */
+		bool l3_is_ipv4; /* false means L3 is ipv6 */
+		struct bnxt_tc_l3_key l3;
+		struct bnxt_tc_l4_key l4;
+	} nat;
+};
+
+struct bnxt_tc_flow {
+	u32				flags;
+#define BNXT_TC_FLOW_FLAGS_ETH_ADDRS		BIT(1)
+#define BNXT_TC_FLOW_FLAGS_IPV4_ADDRS		BIT(2)
+#define BNXT_TC_FLOW_FLAGS_IPV6_ADDRS		BIT(3)
+#define BNXT_TC_FLOW_FLAGS_PORTS		BIT(4)
+#define BNXT_TC_FLOW_FLAGS_ICMP			BIT(5)
+#define BNXT_TC_FLOW_FLAGS_TUNL_ETH_ADDRS	BIT(6)
+#define BNXT_TC_FLOW_FLAGS_TUNL_IPV4_ADDRS	BIT(7)
+#define BNXT_TC_FLOW_FLAGS_TUNL_IPV6_ADDRS	BIT(8)
+#define BNXT_TC_FLOW_FLAGS_TUNL_PORTS		BIT(9)
+#define BNXT_TC_FLOW_FLAGS_TUNL_ID		BIT(10)
+#define BNXT_TC_FLOW_FLAGS_TUNNEL	(BNXT_TC_FLOW_FLAGS_TUNL_ETH_ADDRS | \
+					 BNXT_TC_FLOW_FLAGS_TUNL_IPV4_ADDRS | \
+					 BNXT_TC_FLOW_FLAGS_TUNL_IPV6_ADDRS |\
+					 BNXT_TC_FLOW_FLAGS_TUNL_PORTS |\
+					 BNXT_TC_FLOW_FLAGS_TUNL_ID)
+
+	/* flow applicable to pkts ingressing on this fid */
+	u16				src_fid;
+	struct bnxt_tc_l2_key		l2_key;
+	struct bnxt_tc_l2_key		l2_mask;
+	struct bnxt_tc_l3_key		l3_key;
+	struct bnxt_tc_l3_key		l3_mask;
+	struct bnxt_tc_l4_key		l4_key;
+	struct bnxt_tc_l4_key		l4_mask;
+	struct ip_tunnel_key		tun_key;
+	struct ip_tunnel_key		tun_mask;
+
+	struct bnxt_tc_actions		actions;
+
+	/* updated stats accounting for hw-counter wrap-around */
+	struct bnxt_tc_flow_stats	stats;
+	/* previous snap-shot of stats */
+	struct bnxt_tc_flow_stats	prev_stats;
+	unsigned long			lastused; /* jiffies */
+	/* for calculating delta from prev_stats and
+	 * updating prev_stats atomically.
+	 */
+	spinlock_t			stats_lock;
+};
+
+/* Tunnel encap/decap hash table
+ * This table is used to maintain a list of flows that use
+ * the same tunnel encap/decap params (ip_daddrs, vni, udp_dport)
+ * and the FW returned handle.
+ * A separate table is maintained for encap and decap
+ */
+struct bnxt_tc_tunnel_node {
+	struct ip_tunnel_key		key;
+	struct rhash_head		node;
+
+	/* tunnel l2 info */
+	struct bnxt_tc_l2_key		l2_info;
+
+#define	INVALID_TUNNEL_HANDLE		cpu_to_le32(0xffffffff)
+	/* tunnel handle returned by FW */
+	__le32				tunnel_handle;
+
+	u32				refcount;
+	struct rcu_head			rcu;
+};
+
+/* L2 hash table
+ * The same data-struct is used for L2-flow table and L2-tunnel table.
+ * The L2 part of a flow or tunnel is stored in a hash table.
+ * A flow that shares the same L2 key/mask with an
+ * already existing flow/tunnel must refer to it's flow handle or
+ * decap_filter_id respectively.
+ */
+struct bnxt_tc_l2_node {
+	/* hash key: first 16b of key */
+#define BNXT_TC_L2_KEY_LEN			16
+	struct bnxt_tc_l2_key	key;
+	struct rhash_head	node;
+
+	/* a linked list of flows that share the same l2 key */
+	struct list_head	common_l2_flows;
+
+	/* number of flows/tunnels sharing the l2 key */
+	u16			refcount;
+
+	struct rcu_head		rcu;
+};
+
+struct bnxt_tc_flow_node {
+	/* hash key: provided by TC */
+	unsigned long			cookie;
+	struct rhash_head		node;
+
+	struct bnxt_tc_flow		flow;
+
+	__le64				ext_flow_handle;
+	__le16				flow_handle;
+	__le32				flow_id;
+
+	/* L2 node in l2 hashtable that shares flow's l2 key */
+	struct bnxt_tc_l2_node		*l2_node;
+	/* for the shared_flows list maintained in l2_node */
+	struct list_head		l2_list_node;
+
+	/* tunnel encap related */
+	struct bnxt_tc_tunnel_node	*encap_node;
+
+	/* tunnel decap related */
+	struct bnxt_tc_tunnel_node	*decap_node;
+	/* L2 node in tunnel-l2 hashtable that shares flow's tunnel l2 key */
+	struct bnxt_tc_l2_node		*decap_l2_node;
+	/* for the shared_flows list maintained in tunnel decap l2_node */
+	struct list_head		decap_l2_list_node;
+
+	struct rcu_head			rcu;
+};
+
+int bnxt_tc_setup_flower(struct bnxt *bp, u16 src_fid,
+			 struct flow_cls_offload *cls_flower);
+int bnxt_init_tc(struct bnxt *bp);
+void bnxt_shutdown_tc(struct bnxt *bp);
+void bnxt_tc_flow_stats_work(struct bnxt *bp);
+
+static inline bool bnxt_tc_flower_enabled(struct bnxt *bp)
+{
+	return bp->tc_info && bp->tc_info->enabled;
+}
+
+#else /* CONFIG_BNXT_FLOWER_OFFLOAD */
+
+static inline int bnxt_tc_setup_flower(struct bnxt *bp, u16 src_fid,
+				       struct flow_cls_offload *cls_flower)
+{
+	return -EOPNOTSUPP;
+}
+
+static inline int bnxt_init_tc(struct bnxt *bp)
+{
+	return 0;
+}
+
+static inline void bnxt_shutdown_tc(struct bnxt *bp)
+{
+}
+
+static inline void bnxt_tc_flow_stats_work(struct bnxt *bp)
+{
+}
+
+static inline bool bnxt_tc_flower_enabled(struct bnxt *bp)
+{
+	return false;
+}
+#endif /* CONFIG_BNXT_FLOWER_OFFLOAD */
+#endif /* BNXT_TC_H */
diff --git a/drivers/net/ethernet/broadcom/bnxt/bnxt_ulp.c b/drivers/net/ethernet/broadcom/bnxt/bnxt_ulp.c
index 77da75a55c02..f8c2c72b382d 100644
--- a/drivers/net/ethernet/broadcom/bnxt/bnxt_ulp.c
+++ b/drivers/net/ethernet/broadcom/bnxt/bnxt_ulp.c
@@ -1,6 +1,6 @@
 /* Broadcom NetXtreme-C/E network driver.
  *
- * Copyright (c) 2016 Broadcom Limited
+ * Copyright (c) 2016-2018 Broadcom Limited
  *
  * This program is free software; you can redistribute it and/or modify
  * it under the terms of the GNU General Public License as published by
@@ -19,269 +19,331 @@
 #include <linux/irq.h>
 #include <asm/byteorder.h>
 #include <linux/bitmap.h>
+#include <linux/auxiliary_bus.h>
+#include <net/netdev_lock.h>
+#include <linux/bnxt/hsi.h>
 
-#include "bnxt_hsi.h"
 #include "bnxt.h"
+#include "bnxt_hwrm.h"
 #include "bnxt_ulp.h"
 
-static int bnxt_register_dev(struct bnxt_en_dev *edev, int ulp_id,
-			     struct bnxt_ulp_ops *ulp_ops, void *handle)
-{
-	struct net_device *dev = edev->net;
-	struct bnxt *bp = netdev_priv(dev);
-	struct bnxt_ulp *ulp;
+static DEFINE_IDA(bnxt_aux_dev_ids);
 
-	ASSERT_RTNL();
-	if (ulp_id >= BNXT_MAX_ULP)
-		return -EINVAL;
+static void bnxt_fill_msix_vecs(struct bnxt *bp, struct bnxt_msix_entry *ent)
+{
+	struct bnxt_en_dev *edev = bp->edev;
+	int num_msix, i;
 
-	ulp = &edev->ulp_tbl[ulp_id];
-	if (rcu_access_pointer(ulp->ulp_ops)) {
-		netdev_err(bp->dev, "ulp id %d already registered\n", ulp_id);
-		return -EBUSY;
+	if (!edev->ulp_tbl->msix_requested) {
+		netdev_warn(bp->dev, "Requested MSI-X vectors insufficient\n");
+		return;
 	}
-	if (ulp_id == BNXT_ROCE_ULP) {
-		unsigned int max_stat_ctxs;
-
-		max_stat_ctxs = bnxt_get_max_func_stat_ctxs(bp);
-		if (max_stat_ctxs <= BNXT_MIN_ROCE_STAT_CTXS ||
-		    bp->num_stat_ctxs == max_stat_ctxs)
-			return -ENOMEM;
-		bnxt_set_max_func_stat_ctxs(bp, max_stat_ctxs -
-					    BNXT_MIN_ROCE_STAT_CTXS);
+	num_msix = edev->ulp_tbl->msix_requested;
+	for (i = 0; i < num_msix; i++) {
+		ent[i].vector = bp->irq_tbl[i].vector;
+		ent[i].ring_idx = i;
+		if (bp->flags & BNXT_FLAG_CHIP_P5_PLUS)
+			ent[i].db_offset = bp->db_offset;
+		else
+			ent[i].db_offset = i * 0x80;
 	}
+}
 
-	atomic_set(&ulp->ref_count, 0);
-	ulp->handle = handle;
-	rcu_assign_pointer(ulp->ulp_ops, ulp_ops);
+int bnxt_get_ulp_msix_num(struct bnxt *bp)
+{
+	if (bp->edev)
+		return bp->edev->ulp_num_msix_vec;
+	return 0;
+}
 
-	if (ulp_id == BNXT_ROCE_ULP) {
-		if (test_bit(BNXT_STATE_OPEN, &bp->state))
-			bnxt_hwrm_vnic_cfg(bp, 0);
-	}
+void bnxt_set_ulp_msix_num(struct bnxt *bp, int num)
+{
+	if (bp->edev)
+		bp->edev->ulp_num_msix_vec = num;
+}
 
+int bnxt_get_ulp_msix_num_in_use(struct bnxt *bp)
+{
+	if (bnxt_ulp_registered(bp->edev))
+		return bp->edev->ulp_num_msix_vec;
 	return 0;
 }
 
-static int bnxt_unregister_dev(struct bnxt_en_dev *edev, int ulp_id)
+int bnxt_get_ulp_stat_ctxs(struct bnxt *bp)
 {
-	struct net_device *dev = edev->net;
-	struct bnxt *bp = netdev_priv(dev);
-	struct bnxt_ulp *ulp;
-	int i = 0;
-
-	ASSERT_RTNL();
-	if (ulp_id >= BNXT_MAX_ULP)
-		return -EINVAL;
+	if (bp->edev)
+		return bp->edev->ulp_num_ctxs;
+	return 0;
+}
 
-	ulp = &edev->ulp_tbl[ulp_id];
-	if (!rcu_access_pointer(ulp->ulp_ops)) {
-		netdev_err(bp->dev, "ulp id %d not registered\n", ulp_id);
-		return -EINVAL;
-	}
-	if (ulp_id == BNXT_ROCE_ULP) {
-		unsigned int max_stat_ctxs;
+void bnxt_set_ulp_stat_ctxs(struct bnxt *bp, int num_ulp_ctx)
+{
+	if (bp->edev)
+		bp->edev->ulp_num_ctxs = num_ulp_ctx;
+}
 
-		max_stat_ctxs = bnxt_get_max_func_stat_ctxs(bp);
-		bnxt_set_max_func_stat_ctxs(bp, max_stat_ctxs + 1);
-	}
-	if (ulp->max_async_event_id)
-		bnxt_hwrm_func_rgtr_async_events(bp, NULL, 0);
+int bnxt_get_ulp_stat_ctxs_in_use(struct bnxt *bp)
+{
+	if (bnxt_ulp_registered(bp->edev))
+		return bp->edev->ulp_num_ctxs;
+	return 0;
+}
 
-	RCU_INIT_POINTER(ulp->ulp_ops, NULL);
-	synchronize_rcu();
-	ulp->max_async_event_id = 0;
-	ulp->async_events_bmap = NULL;
-	while (atomic_read(&ulp->ref_count) != 0 && i < 10) {
-		msleep(100);
-		i++;
+void bnxt_set_dflt_ulp_stat_ctxs(struct bnxt *bp)
+{
+	if (bp->edev) {
+		bp->edev->ulp_num_ctxs = BNXT_MIN_ROCE_STAT_CTXS;
+		/* Reserve one additional stat_ctx for PF0 (except
+		 * on 1-port NICs) as it also creates one stat_ctx
+		 * for PF1 in case of RoCE bonding.
+		 */
+		if (BNXT_PF(bp) && !bp->pf.port_id &&
+		    bp->port_count > 1)
+			bp->edev->ulp_num_ctxs++;
+
+		/* Reserve one additional stat_ctx when the device is capable
+		 * of supporting port mirroring on RDMA device.
+		 */
+		if (BNXT_MIRROR_ON_ROCE_CAP(bp))
+			bp->edev->ulp_num_ctxs++;
 	}
-	return 0;
 }
 
-static int bnxt_req_msix_vecs(struct bnxt_en_dev *edev, int ulp_id,
-			      struct bnxt_msix_entry *ent, int num_msix)
+int bnxt_register_dev(struct bnxt_en_dev *edev,
+		      struct bnxt_ulp_ops *ulp_ops,
+		      void *handle)
 {
 	struct net_device *dev = edev->net;
 	struct bnxt *bp = netdev_priv(dev);
-	int max_idx, max_cp_rings;
-	int avail_msix, i, idx;
-
-	ASSERT_RTNL();
-	if (ulp_id != BNXT_ROCE_ULP)
-		return -EINVAL;
-
-	if (!(bp->flags & BNXT_FLAG_USING_MSIX))
-		return -ENODEV;
-
-	max_cp_rings = bnxt_get_max_func_cp_rings(bp);
-	max_idx = min_t(int, bp->total_irqs, max_cp_rings);
-	avail_msix = max_idx - bp->cp_nr_rings;
-	if (!avail_msix)
-		return -ENOMEM;
-	if (avail_msix > num_msix)
-		avail_msix = num_msix;
-
-	idx = max_idx - avail_msix;
-	for (i = 0; i < avail_msix; i++) {
-		ent[i].vector = bp->irq_tbl[idx + i].vector;
-		ent[i].ring_idx = idx + i;
-		ent[i].db_offset = (idx + i) * 0x80;
+	unsigned int max_stat_ctxs;
+	struct bnxt_ulp *ulp;
+	int rc = 0;
+
+	netdev_lock(dev);
+	mutex_lock(&edev->en_dev_lock);
+	if (!bp->irq_tbl) {
+		rc = -ENODEV;
+		goto exit;
 	}
-	bnxt_set_max_func_irqs(bp, max_idx - avail_msix);
-	bnxt_set_max_func_cp_rings(bp, max_cp_rings - avail_msix);
-	edev->ulp_tbl[ulp_id].msix_requested = avail_msix;
-	return avail_msix;
+	max_stat_ctxs = bnxt_get_max_func_stat_ctxs(bp);
+	if (max_stat_ctxs <= BNXT_MIN_ROCE_STAT_CTXS ||
+	    bp->cp_nr_rings == max_stat_ctxs) {
+		rc = -ENOMEM;
+		goto exit;
+	}
+
+	ulp = edev->ulp_tbl;
+	ulp->handle = handle;
+	rcu_assign_pointer(ulp->ulp_ops, ulp_ops);
+
+	if (test_bit(BNXT_STATE_OPEN, &bp->state))
+		bnxt_hwrm_vnic_cfg(bp, &bp->vnic_info[BNXT_VNIC_DEFAULT]);
+
+	edev->ulp_tbl->msix_requested = bnxt_get_ulp_msix_num(bp);
+
+	bnxt_fill_msix_vecs(bp, bp->edev->msix_entries);
+	edev->flags |= BNXT_EN_FLAG_MSIX_REQUESTED;
+exit:
+	mutex_unlock(&edev->en_dev_lock);
+	netdev_unlock(dev);
+	return rc;
 }
+EXPORT_SYMBOL(bnxt_register_dev);
 
-static int bnxt_free_msix_vecs(struct bnxt_en_dev *edev, int ulp_id)
+void bnxt_unregister_dev(struct bnxt_en_dev *edev)
 {
 	struct net_device *dev = edev->net;
 	struct bnxt *bp = netdev_priv(dev);
-	int max_cp_rings, msix_requested;
+	struct bnxt_ulp *ulp;
 
-	ASSERT_RTNL();
-	if (ulp_id != BNXT_ROCE_ULP)
-		return -EINVAL;
+	ulp = edev->ulp_tbl;
+	netdev_lock(dev);
+	mutex_lock(&edev->en_dev_lock);
+	if (ulp->msix_requested)
+		edev->flags &= ~BNXT_EN_FLAG_MSIX_REQUESTED;
+	edev->ulp_tbl->msix_requested = 0;
 
-	max_cp_rings = bnxt_get_max_func_cp_rings(bp);
-	msix_requested = edev->ulp_tbl[ulp_id].msix_requested;
-	bnxt_set_max_func_cp_rings(bp, max_cp_rings + msix_requested);
-	edev->ulp_tbl[ulp_id].msix_requested = 0;
-	bnxt_set_max_func_irqs(bp, bp->total_irqs);
-	return 0;
+	if (ulp->max_async_event_id)
+		bnxt_hwrm_func_drv_rgtr(bp, NULL, 0, true);
+
+	RCU_INIT_POINTER(ulp->ulp_ops, NULL);
+	synchronize_rcu();
+	ulp->max_async_event_id = 0;
+	ulp->async_events_bmap = NULL;
+	mutex_unlock(&edev->en_dev_lock);
+	netdev_unlock(dev);
+	return;
 }
+EXPORT_SYMBOL(bnxt_unregister_dev);
 
-void bnxt_subtract_ulp_resources(struct bnxt *bp, int ulp_id)
+static int bnxt_set_dflt_ulp_msix(struct bnxt *bp)
 {
-	ASSERT_RTNL();
-	if (bnxt_ulp_registered(bp->edev, ulp_id)) {
-		struct bnxt_en_dev *edev = bp->edev;
-		unsigned int msix_req, max;
-
-		msix_req = edev->ulp_tbl[ulp_id].msix_requested;
-		max = bnxt_get_max_func_cp_rings(bp);
-		bnxt_set_max_func_cp_rings(bp, max - msix_req);
-		max = bnxt_get_max_func_stat_ctxs(bp);
-		bnxt_set_max_func_stat_ctxs(bp, max - 1);
-	}
+	int roce_msix = BNXT_MAX_ROCE_MSIX;
+
+	if (BNXT_VF(bp))
+		roce_msix = BNXT_MAX_ROCE_MSIX_VF;
+	else if (bp->port_partition_type)
+		roce_msix = BNXT_MAX_ROCE_MSIX_NPAR_PF;
+
+	/* NQ MSIX vectors should match the number of CPUs plus 1 more for
+	 * the CREQ MSIX, up to the default.
+	 */
+	return min_t(int, roce_msix, num_online_cpus() + 1);
 }
 
-static int bnxt_send_msg(struct bnxt_en_dev *edev, int ulp_id,
+int bnxt_send_msg(struct bnxt_en_dev *edev,
 			 struct bnxt_fw_msg *fw_msg)
 {
 	struct net_device *dev = edev->net;
 	struct bnxt *bp = netdev_priv(dev);
+	struct output *resp;
 	struct input *req;
+	u32 resp_len;
 	int rc;
 
-	mutex_lock(&bp->hwrm_cmd_lock);
-	req = fw_msg->msg;
-	req->resp_addr = cpu_to_le64(bp->hwrm_cmd_resp_dma_addr);
-	rc = _hwrm_send_message(bp, fw_msg->msg, fw_msg->msg_len,
-				fw_msg->timeout);
-	if (!rc) {
-		struct output *resp = bp->hwrm_cmd_resp_addr;
-		u32 len = le16_to_cpu(resp->resp_len);
+	if (bp->fw_reset_state)
+		return -EBUSY;
 
-		if (fw_msg->resp_max_len < len)
-			len = fw_msg->resp_max_len;
+	rc = hwrm_req_init(bp, req, 0 /* don't care */);
+	if (rc)
+		return rc;
 
-		memcpy(fw_msg->resp, resp, len);
-	}
-	mutex_unlock(&bp->hwrm_cmd_lock);
-	return rc;
-}
+	rc = hwrm_req_replace(bp, req, fw_msg->msg, fw_msg->msg_len);
+	if (rc)
+		goto drop_req;
 
-static void bnxt_ulp_get(struct bnxt_ulp *ulp)
-{
-	atomic_inc(&ulp->ref_count);
-}
+	hwrm_req_timeout(bp, req, fw_msg->timeout);
+	resp = hwrm_req_hold(bp, req);
+	rc = hwrm_req_send(bp, req);
+	resp_len = le16_to_cpu(resp->resp_len);
+	if (resp_len) {
+		if (fw_msg->resp_max_len < resp_len)
+			resp_len = fw_msg->resp_max_len;
 
-static void bnxt_ulp_put(struct bnxt_ulp *ulp)
-{
-	atomic_dec(&ulp->ref_count);
+		memcpy(fw_msg->resp, resp, resp_len);
+	}
+drop_req:
+	hwrm_req_drop(bp, req);
+	return rc;
 }
+EXPORT_SYMBOL(bnxt_send_msg);
 
 void bnxt_ulp_stop(struct bnxt *bp)
 {
+	struct bnxt_aux_priv *aux_priv = bp->aux_priv;
 	struct bnxt_en_dev *edev = bp->edev;
-	struct bnxt_ulp_ops *ops;
-	int i;
 
 	if (!edev)
 		return;
 
-	for (i = 0; i < BNXT_MAX_ULP; i++) {
-		struct bnxt_ulp *ulp = &edev->ulp_tbl[i];
+	mutex_lock(&edev->en_dev_lock);
+	if (!bnxt_ulp_registered(edev) ||
+	    (edev->flags & BNXT_EN_FLAG_ULP_STOPPED))
+		goto ulp_stop_exit;
+
+	edev->flags |= BNXT_EN_FLAG_ULP_STOPPED;
+	if (aux_priv) {
+		struct auxiliary_device *adev;
+
+		adev = &aux_priv->aux_dev;
+		if (adev->dev.driver) {
+			const struct auxiliary_driver *adrv;
+			pm_message_t pm = {};
 
-		ops = rtnl_dereference(ulp->ulp_ops);
-		if (!ops || !ops->ulp_stop)
-			continue;
-		ops->ulp_stop(ulp->handle);
+			adrv = to_auxiliary_drv(adev->dev.driver);
+			edev->en_state = bp->state;
+			adrv->suspend(adev, pm);
+		}
 	}
+ulp_stop_exit:
+	mutex_unlock(&edev->en_dev_lock);
 }
 
-void bnxt_ulp_start(struct bnxt *bp)
+void bnxt_ulp_start(struct bnxt *bp, int err)
 {
+	struct bnxt_aux_priv *aux_priv = bp->aux_priv;
 	struct bnxt_en_dev *edev = bp->edev;
-	struct bnxt_ulp_ops *ops;
-	int i;
 
-	if (!edev)
+	if (!edev || err)
 		return;
 
-	for (i = 0; i < BNXT_MAX_ULP; i++) {
-		struct bnxt_ulp *ulp = &edev->ulp_tbl[i];
+	mutex_lock(&edev->en_dev_lock);
+	if (!bnxt_ulp_registered(edev) ||
+	    !(edev->flags & BNXT_EN_FLAG_ULP_STOPPED))
+		goto ulp_start_exit;
+
+	if (edev->ulp_tbl->msix_requested)
+		bnxt_fill_msix_vecs(bp, edev->msix_entries);
 
-		ops = rtnl_dereference(ulp->ulp_ops);
-		if (!ops || !ops->ulp_start)
-			continue;
-		ops->ulp_start(ulp->handle);
+	if (aux_priv) {
+		struct auxiliary_device *adev;
+
+		adev = &aux_priv->aux_dev;
+		if (adev->dev.driver) {
+			const struct auxiliary_driver *adrv;
+
+			adrv = to_auxiliary_drv(adev->dev.driver);
+			edev->en_state = bp->state;
+			adrv->resume(adev);
+		}
 	}
+ulp_start_exit:
+	edev->flags &= ~BNXT_EN_FLAG_ULP_STOPPED;
+	mutex_unlock(&edev->en_dev_lock);
 }
 
-void bnxt_ulp_sriov_cfg(struct bnxt *bp, int num_vfs)
+void bnxt_ulp_irq_stop(struct bnxt *bp)
 {
 	struct bnxt_en_dev *edev = bp->edev;
 	struct bnxt_ulp_ops *ops;
-	int i;
+	bool reset = false;
 
-	if (!edev)
+	if (!edev || !(edev->flags & BNXT_EN_FLAG_MSIX_REQUESTED))
 		return;
 
-	for (i = 0; i < BNXT_MAX_ULP; i++) {
-		struct bnxt_ulp *ulp = &edev->ulp_tbl[i];
+	if (bnxt_ulp_registered(bp->edev)) {
+		struct bnxt_ulp *ulp = edev->ulp_tbl;
 
-		rcu_read_lock();
-		ops = rcu_dereference(ulp->ulp_ops);
-		if (!ops || !ops->ulp_sriov_config) {
-			rcu_read_unlock();
-			continue;
-		}
-		bnxt_ulp_get(ulp);
-		rcu_read_unlock();
-		ops->ulp_sriov_config(ulp->handle, num_vfs);
-		bnxt_ulp_put(ulp);
+		if (!ulp->msix_requested)
+			return;
+
+		ops = netdev_lock_dereference(ulp->ulp_ops, bp->dev);
+		if (!ops || !ops->ulp_irq_stop)
+			return;
+		if (test_bit(BNXT_STATE_FW_RESET_DET, &bp->state))
+			reset = true;
+		ops->ulp_irq_stop(ulp->handle, reset);
 	}
 }
 
-void bnxt_ulp_shutdown(struct bnxt *bp)
+void bnxt_ulp_irq_restart(struct bnxt *bp, int err)
 {
 	struct bnxt_en_dev *edev = bp->edev;
 	struct bnxt_ulp_ops *ops;
-	int i;
 
-	if (!edev)
+	if (!edev || !(edev->flags & BNXT_EN_FLAG_MSIX_REQUESTED))
 		return;
 
-	for (i = 0; i < BNXT_MAX_ULP; i++) {
-		struct bnxt_ulp *ulp = &edev->ulp_tbl[i];
+	if (bnxt_ulp_registered(bp->edev)) {
+		struct bnxt_ulp *ulp = edev->ulp_tbl;
+		struct bnxt_msix_entry *ent = NULL;
+
+		if (!ulp->msix_requested)
+			return;
 
-		ops = rtnl_dereference(ulp->ulp_ops);
-		if (!ops || !ops->ulp_shutdown)
-			continue;
-		ops->ulp_shutdown(ulp->handle);
+		ops = netdev_lock_dereference(ulp->ulp_ops, bp->dev);
+		if (!ops || !ops->ulp_irq_restart)
+			return;
+
+		if (!err) {
+			ent = kcalloc(ulp->msix_requested, sizeof(*ent),
+				      GFP_KERNEL);
+			if (!ent)
+				return;
+			bnxt_fill_msix_vecs(bp, ent);
+		}
+		ops->ulp_irq_restart(ulp->handle, ent);
+		kfree(ent);
 	}
 }
 
@@ -290,76 +352,185 @@ void bnxt_ulp_async_events(struct bnxt *bp, struct hwrm_async_event_cmpl *cmpl)
 	u16 event_id = le16_to_cpu(cmpl->event_id);
 	struct bnxt_en_dev *edev = bp->edev;
 	struct bnxt_ulp_ops *ops;
-	int i;
+	struct bnxt_ulp *ulp;
 
-	if (!edev)
+	if (!bnxt_ulp_registered(edev))
 		return;
+	ulp = edev->ulp_tbl;
 
 	rcu_read_lock();
-	for (i = 0; i < BNXT_MAX_ULP; i++) {
-		struct bnxt_ulp *ulp = &edev->ulp_tbl[i];
-
-		ops = rcu_dereference(ulp->ulp_ops);
-		if (!ops || !ops->ulp_async_notifier)
-			continue;
-		if (!ulp->async_events_bmap ||
-		    event_id > ulp->max_async_event_id)
-			continue;
-
-		/* Read max_async_event_id first before testing the bitmap. */
-		smp_rmb();
-		if (test_bit(event_id, ulp->async_events_bmap))
-			ops->ulp_async_notifier(ulp->handle, cmpl);
-	}
+
+	ops = rcu_dereference(ulp->ulp_ops);
+	if (!ops || !ops->ulp_async_notifier)
+		goto exit_unlock_rcu;
+	if (!ulp->async_events_bmap || event_id > ulp->max_async_event_id)
+		goto exit_unlock_rcu;
+
+	/* Read max_async_event_id first before testing the bitmap. */
+	smp_rmb();
+
+	if (test_bit(event_id, ulp->async_events_bmap))
+		ops->ulp_async_notifier(ulp->handle, cmpl);
+exit_unlock_rcu:
 	rcu_read_unlock();
 }
 
-static int bnxt_register_async_events(struct bnxt_en_dev *edev, int ulp_id,
-				      unsigned long *events_bmap, u16 max_id)
+void bnxt_register_async_events(struct bnxt_en_dev *edev,
+				unsigned long *events_bmap, u16 max_id)
 {
 	struct net_device *dev = edev->net;
 	struct bnxt *bp = netdev_priv(dev);
 	struct bnxt_ulp *ulp;
 
-	if (ulp_id >= BNXT_MAX_ULP)
-		return -EINVAL;
-
-	ulp = &edev->ulp_tbl[ulp_id];
+	ulp = edev->ulp_tbl;
 	ulp->async_events_bmap = events_bmap;
 	/* Make sure bnxt_ulp_async_events() sees this order */
 	smp_wmb();
 	ulp->max_async_event_id = max_id;
-	bnxt_hwrm_func_rgtr_async_events(bp, events_bmap, max_id + 1);
-	return 0;
+	bnxt_hwrm_func_drv_rgtr(bp, events_bmap, max_id + 1, true);
 }
+EXPORT_SYMBOL(bnxt_register_async_events);
+
+void bnxt_rdma_aux_device_uninit(struct bnxt *bp)
+{
+	struct bnxt_aux_priv *aux_priv;
+	struct auxiliary_device *adev;
 
-static const struct bnxt_en_ops bnxt_en_ops_tbl = {
-	.bnxt_register_device	= bnxt_register_dev,
-	.bnxt_unregister_device	= bnxt_unregister_dev,
-	.bnxt_request_msix	= bnxt_req_msix_vecs,
-	.bnxt_free_msix		= bnxt_free_msix_vecs,
-	.bnxt_send_fw_msg	= bnxt_send_msg,
-	.bnxt_register_fw_async_events	= bnxt_register_async_events,
-};
+	/* Skip if no auxiliary device init was done. */
+	if (!bp->aux_priv)
+		return;
+
+	aux_priv = bp->aux_priv;
+	adev = &aux_priv->aux_dev;
+	auxiliary_device_uninit(adev);
+}
 
-struct bnxt_en_dev *bnxt_ulp_probe(struct net_device *dev)
+static void bnxt_aux_dev_release(struct device *dev)
 {
-	struct bnxt *bp = netdev_priv(dev);
+	struct bnxt_aux_priv *aux_priv =
+		container_of(dev, struct bnxt_aux_priv, aux_dev.dev);
+	struct bnxt *bp = netdev_priv(aux_priv->edev->net);
+
+	ida_free(&bnxt_aux_dev_ids, aux_priv->id);
+	kfree(aux_priv->edev->ulp_tbl);
+	bp->edev = NULL;
+	kfree(aux_priv->edev);
+	kfree(aux_priv);
+	bp->aux_priv = NULL;
+}
+
+void bnxt_rdma_aux_device_del(struct bnxt *bp)
+{
+	if (!bp->edev)
+		return;
+
+	auxiliary_device_delete(&bp->aux_priv->aux_dev);
+}
+
+static void bnxt_set_edev_info(struct bnxt_en_dev *edev, struct bnxt *bp)
+{
+	edev->net = bp->dev;
+	edev->pdev = bp->pdev;
+	edev->l2_db_size = bp->db_size;
+	edev->l2_db_size_nc = bp->db_size;
+	edev->l2_db_offset = bp->db_offset;
+	mutex_init(&edev->en_dev_lock);
+
+	if (bp->flags & BNXT_FLAG_ROCEV1_CAP)
+		edev->flags |= BNXT_EN_FLAG_ROCEV1_CAP;
+	if (bp->flags & BNXT_FLAG_ROCEV2_CAP)
+		edev->flags |= BNXT_EN_FLAG_ROCEV2_CAP;
+	if (bp->flags & BNXT_FLAG_VF)
+		edev->flags |= BNXT_EN_FLAG_VF;
+	if (BNXT_ROCE_VF_RESC_CAP(bp))
+		edev->flags |= BNXT_EN_FLAG_ROCE_VF_RES_MGMT;
+	if (BNXT_SW_RES_LMT(bp))
+		edev->flags |= BNXT_EN_FLAG_SW_RES_LMT;
+
+	edev->chip_num = bp->chip_num;
+	edev->hw_ring_stats_size = bp->hw_ring_stats_size;
+	edev->pf_port_id = bp->pf.port_id;
+	edev->en_state = bp->state;
+	edev->bar0 = bp->bar0;
+}
+
+void bnxt_rdma_aux_device_add(struct bnxt *bp)
+{
+	struct auxiliary_device *aux_dev;
+	int rc;
+
+	if (!bp->edev)
+		return;
+
+	aux_dev = &bp->aux_priv->aux_dev;
+	rc = auxiliary_device_add(aux_dev);
+	if (rc) {
+		netdev_warn(bp->dev, "Failed to add auxiliary device for ROCE\n");
+		auxiliary_device_uninit(aux_dev);
+		bp->flags &= ~BNXT_FLAG_ROCE_CAP;
+	}
+}
+
+void bnxt_rdma_aux_device_init(struct bnxt *bp)
+{
+	struct auxiliary_device *aux_dev;
+	struct bnxt_aux_priv *aux_priv;
 	struct bnxt_en_dev *edev;
+	struct bnxt_ulp *ulp;
+	int rc;
+
+	if (!(bp->flags & BNXT_FLAG_ROCE_CAP))
+		return;
+
+	aux_priv = kzalloc(sizeof(*bp->aux_priv), GFP_KERNEL);
+	if (!aux_priv)
+		goto exit;
+
+	aux_priv->id = ida_alloc(&bnxt_aux_dev_ids, GFP_KERNEL);
+	if (aux_priv->id < 0) {
+		netdev_warn(bp->dev,
+			    "ida alloc failed for ROCE auxiliary device\n");
+		kfree(aux_priv);
+		goto exit;
+	}
 
-	edev = bp->edev;
-	if (!edev) {
-		edev = kzalloc(sizeof(*edev), GFP_KERNEL);
-		if (!edev)
-			return ERR_PTR(-ENOMEM);
-		edev->en_ops = &bnxt_en_ops_tbl;
-		if (bp->flags & BNXT_FLAG_ROCEV1_CAP)
-			edev->flags |= BNXT_EN_FLAG_ROCEV1_CAP;
-		if (bp->flags & BNXT_FLAG_ROCEV2_CAP)
-			edev->flags |= BNXT_EN_FLAG_ROCEV2_CAP;
-		edev->net = dev;
-		edev->pdev = bp->pdev;
-		bp->edev = edev;
+	aux_dev = &aux_priv->aux_dev;
+	aux_dev->id = aux_priv->id;
+	aux_dev->name = "rdma";
+	aux_dev->dev.parent = &bp->pdev->dev;
+	aux_dev->dev.release = bnxt_aux_dev_release;
+
+	rc = auxiliary_device_init(aux_dev);
+	if (rc) {
+		ida_free(&bnxt_aux_dev_ids, aux_priv->id);
+		kfree(aux_priv);
+		goto exit;
 	}
-	return bp->edev;
+	bp->aux_priv = aux_priv;
+
+	/* From this point, all cleanup will happen via the .release callback &
+	 * any error unwinding will need to include a call to
+	 * auxiliary_device_uninit.
+	 */
+	edev = kzalloc(sizeof(*edev), GFP_KERNEL);
+	if (!edev)
+		goto aux_dev_uninit;
+
+	aux_priv->edev = edev;
+
+	ulp = kzalloc(sizeof(*ulp), GFP_KERNEL);
+	if (!ulp)
+		goto aux_dev_uninit;
+
+	edev->ulp_tbl = ulp;
+	bp->edev = edev;
+	bnxt_set_edev_info(edev, bp);
+	bp->ulp_num_msix_want = bnxt_set_dflt_ulp_msix(bp);
+
+	return;
+
+aux_dev_uninit:
+	auxiliary_device_uninit(aux_dev);
+exit:
+	bp->flags &= ~BNXT_FLAG_ROCE_CAP;
 }
diff --git a/drivers/net/ethernet/broadcom/bnxt/bnxt_ulp.h b/drivers/net/ethernet/broadcom/bnxt/bnxt_ulp.h
index d2471067dc37..7b9dd8ebe4bc 100644
--- a/drivers/net/ethernet/broadcom/bnxt/bnxt_ulp.h
+++ b/drivers/net/ethernet/broadcom/bnxt/bnxt_ulp.h
@@ -1,6 +1,6 @@
 /* Broadcom NetXtreme-C/E network driver.
  *
- * Copyright (c) 2016 Broadcom Limited
+ * Copyright (c) 2016-2018 Broadcom Limited
  *
  * This program is free software; you can redistribute it and/or modify
  * it under the terms of the GNU General Public License as published by
@@ -10,31 +10,29 @@
 #ifndef BNXT_ULP_H
 #define BNXT_ULP_H
 
-#define BNXT_ROCE_ULP	0
-#define BNXT_OTHER_ULP	1
-#define BNXT_MAX_ULP	2
-
 #define BNXT_MIN_ROCE_CP_RINGS	2
 #define BNXT_MIN_ROCE_STAT_CTXS	1
 
+#define BNXT_MAX_ROCE_MSIX_VF		2
+#define BNXT_MAX_ROCE_MSIX_NPAR_PF	5
+#define BNXT_MAX_ROCE_MSIX		64
+
 struct hwrm_async_event_cmpl;
 struct bnxt;
 
-struct bnxt_ulp_ops {
-	/* async_notifier() cannot sleep (in BH context) */
-	void (*ulp_async_notifier)(void *, struct hwrm_async_event_cmpl *);
-	void (*ulp_stop)(void *);
-	void (*ulp_start)(void *);
-	void (*ulp_sriov_config)(void *, int);
-	void (*ulp_shutdown)(void *);
-};
-
 struct bnxt_msix_entry {
 	u32	vector;
 	u32	ring_idx;
 	u32	db_offset;
 };
 
+struct bnxt_ulp_ops {
+	/* async_notifier() cannot sleep (in BH context) */
+	void (*ulp_async_notifier)(void *, struct hwrm_async_event_cmpl *);
+	void (*ulp_irq_stop)(void *, bool);
+	void (*ulp_irq_restart)(void *, struct bnxt_msix_entry *);
+};
+
 struct bnxt_fw_msg {
 	void	*msg;
 	int	msg_len;
@@ -49,47 +47,83 @@ struct bnxt_ulp {
 	unsigned long	*async_events_bmap;
 	u16		max_async_event_id;
 	u16		msix_requested;
-	atomic_t	ref_count;
 };
 
 struct bnxt_en_dev {
 	struct net_device *net;
 	struct pci_dev *pdev;
+	struct bnxt_msix_entry			msix_entries[BNXT_MAX_ROCE_MSIX];
 	u32 flags;
 	#define BNXT_EN_FLAG_ROCEV1_CAP		0x1
 	#define BNXT_EN_FLAG_ROCEV2_CAP		0x2
 	#define BNXT_EN_FLAG_ROCE_CAP		(BNXT_EN_FLAG_ROCEV1_CAP | \
 						 BNXT_EN_FLAG_ROCEV2_CAP)
-	const struct bnxt_en_ops	*en_ops;
-	struct bnxt_ulp			ulp_tbl[BNXT_MAX_ULP];
-};
+	#define BNXT_EN_FLAG_MSIX_REQUESTED	0x4
+	#define BNXT_EN_FLAG_ULP_STOPPED	0x8
+	#define BNXT_EN_FLAG_VF			0x10
+#define BNXT_EN_VF(edev)	((edev)->flags & BNXT_EN_FLAG_VF)
+	#define BNXT_EN_FLAG_ROCE_VF_RES_MGMT	0x20
+	#define BNXT_EN_FLAG_SW_RES_LMT		0x40
+#define BNXT_EN_SW_RES_LMT(edev) ((edev)->flags & BNXT_EN_FLAG_SW_RES_LMT)
+
+	struct bnxt_ulp			*ulp_tbl;
+	int				l2_db_size;	/* Doorbell BAR size in
+							 * bytes mapped by L2
+							 * driver.
+							 */
+	int				l2_db_size_nc;	/* Doorbell BAR size in
+							 * bytes mapped as non-
+							 * cacheable.
+							 */
+	int				l2_db_offset;	/* Doorbell offset in
+							 * bytes within
+							 * l2_db_size_nc.
+							 */
+	u16				chip_num;
+	u16				hw_ring_stats_size;
+	u16				pf_port_id;
+	unsigned long			en_state;	/* Could be checked in
+							 * RoCE driver suspend
+							 * mode only. Will be
+							 * updated in resume.
+							 */
+	void __iomem                    *bar0;
 
-struct bnxt_en_ops {
-	int (*bnxt_register_device)(struct bnxt_en_dev *, int,
-				    struct bnxt_ulp_ops *, void *);
-	int (*bnxt_unregister_device)(struct bnxt_en_dev *, int);
-	int (*bnxt_request_msix)(struct bnxt_en_dev *, int,
-				 struct bnxt_msix_entry *, int);
-	int (*bnxt_free_msix)(struct bnxt_en_dev *, int);
-	int (*bnxt_send_fw_msg)(struct bnxt_en_dev *, int,
-				struct bnxt_fw_msg *);
-	int (*bnxt_register_fw_async_events)(struct bnxt_en_dev *, int,
-					     unsigned long *, u16);
+	u16				ulp_num_msix_vec;
+	u16				ulp_num_ctxs;
+
+					/* serialize ulp operations */
+	struct mutex			en_dev_lock;
 };
 
-static inline bool bnxt_ulp_registered(struct bnxt_en_dev *edev, int ulp_id)
+static inline bool bnxt_ulp_registered(struct bnxt_en_dev *edev)
 {
-	if (edev && rcu_access_pointer(edev->ulp_tbl[ulp_id].ulp_ops))
+	if (edev && rcu_access_pointer(edev->ulp_tbl->ulp_ops))
 		return true;
 	return false;
 }
 
-void bnxt_subtract_ulp_resources(struct bnxt *bp, int ulp_id);
+int bnxt_get_ulp_msix_num(struct bnxt *bp);
+int bnxt_get_ulp_msix_num_in_use(struct bnxt *bp);
+void bnxt_set_ulp_msix_num(struct bnxt *bp, int num);
+int bnxt_get_ulp_stat_ctxs(struct bnxt *bp);
+void bnxt_set_ulp_stat_ctxs(struct bnxt *bp, int num_ctxs);
+int bnxt_get_ulp_stat_ctxs_in_use(struct bnxt *bp);
+void bnxt_set_dflt_ulp_stat_ctxs(struct bnxt *bp);
 void bnxt_ulp_stop(struct bnxt *bp);
-void bnxt_ulp_start(struct bnxt *bp);
+void bnxt_ulp_start(struct bnxt *bp, int err);
 void bnxt_ulp_sriov_cfg(struct bnxt *bp, int num_vfs);
-void bnxt_ulp_shutdown(struct bnxt *bp);
+void bnxt_ulp_irq_stop(struct bnxt *bp);
+void bnxt_ulp_irq_restart(struct bnxt *bp, int err);
 void bnxt_ulp_async_events(struct bnxt *bp, struct hwrm_async_event_cmpl *cmpl);
-struct bnxt_en_dev *bnxt_ulp_probe(struct net_device *dev);
-
+void bnxt_rdma_aux_device_uninit(struct bnxt *bp);
+void bnxt_rdma_aux_device_del(struct bnxt *bp);
+void bnxt_rdma_aux_device_add(struct bnxt *bp);
+void bnxt_rdma_aux_device_init(struct bnxt *bp);
+int bnxt_register_dev(struct bnxt_en_dev *edev, struct bnxt_ulp_ops *ulp_ops,
+		      void *handle);
+void bnxt_unregister_dev(struct bnxt_en_dev *edev);
+int bnxt_send_msg(struct bnxt_en_dev *edev, struct bnxt_fw_msg *fw_msg);
+void bnxt_register_async_events(struct bnxt_en_dev *edev,
+				unsigned long *events_bmap, u16 max_id);
 #endif
diff --git a/drivers/net/ethernet/broadcom/bnxt/bnxt_vfr.c b/drivers/net/ethernet/broadcom/bnxt/bnxt_vfr.c
new file mode 100644
index 000000000000..bd116fd578d8
--- /dev/null
+++ b/drivers/net/ethernet/broadcom/bnxt/bnxt_vfr.c
@@ -0,0 +1,600 @@
+/* Broadcom NetXtreme-C/E network driver.
+ *
+ * Copyright (c) 2016-2017 Broadcom Limited
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation.
+ */
+#include <linux/pci.h>
+#include <linux/netdevice.h>
+#include <linux/etherdevice.h>
+#include <linux/rtnetlink.h>
+#include <linux/jhash.h>
+#include <net/pkt_cls.h>
+#include <linux/bnxt/hsi.h>
+
+#include "bnxt.h"
+#include "bnxt_hwrm.h"
+#include "bnxt_vfr.h"
+#include "bnxt_devlink.h"
+#include "bnxt_tc.h"
+
+#ifdef CONFIG_BNXT_SRIOV
+
+#define CFA_HANDLE_INVALID		0xffff
+#define VF_IDX_INVALID			0xffff
+
+static int hwrm_cfa_vfr_alloc(struct bnxt *bp, u16 vf_idx,
+			      u16 *tx_cfa_action, u16 *rx_cfa_code)
+{
+	struct hwrm_cfa_vfr_alloc_output *resp;
+	struct hwrm_cfa_vfr_alloc_input *req;
+	int rc;
+
+	rc = hwrm_req_init(bp, req, HWRM_CFA_VFR_ALLOC);
+	if (!rc) {
+		req->vf_id = cpu_to_le16(vf_idx);
+		sprintf(req->vfr_name, "vfr%d", vf_idx);
+
+		resp = hwrm_req_hold(bp, req);
+		rc = hwrm_req_send(bp, req);
+		if (!rc) {
+			*tx_cfa_action = le16_to_cpu(resp->tx_cfa_action);
+			*rx_cfa_code = le16_to_cpu(resp->rx_cfa_code);
+			netdev_dbg(bp->dev, "tx_cfa_action=0x%x, rx_cfa_code=0x%x",
+				   *tx_cfa_action, *rx_cfa_code);
+		}
+		hwrm_req_drop(bp, req);
+	}
+	if (rc)
+		netdev_info(bp->dev, "%s error rc=%d\n", __func__, rc);
+	return rc;
+}
+
+static int hwrm_cfa_vfr_free(struct bnxt *bp, u16 vf_idx)
+{
+	struct hwrm_cfa_vfr_free_input *req;
+	int rc;
+
+	rc = hwrm_req_init(bp, req, HWRM_CFA_VFR_FREE);
+	if (!rc) {
+		sprintf(req->vfr_name, "vfr%d", vf_idx);
+		rc = hwrm_req_send(bp, req);
+	}
+	if (rc)
+		netdev_info(bp->dev, "%s error rc=%d\n", __func__, rc);
+	return rc;
+}
+
+static int bnxt_hwrm_vfr_qcfg(struct bnxt *bp, struct bnxt_vf_rep *vf_rep,
+			      u16 *max_mtu)
+{
+	struct hwrm_func_qcfg_output *resp;
+	struct hwrm_func_qcfg_input *req;
+	u16 mtu;
+	int rc;
+
+	rc = hwrm_req_init(bp, req, HWRM_FUNC_QCFG);
+	if (rc)
+		return rc;
+
+	req->fid = cpu_to_le16(bp->pf.vf[vf_rep->vf_idx].fw_fid);
+	resp = hwrm_req_hold(bp, req);
+	rc = hwrm_req_send(bp, req);
+	if (!rc) {
+		mtu = le16_to_cpu(resp->max_mtu_configured);
+		if (!mtu)
+			*max_mtu = BNXT_MAX_MTU;
+		else
+			*max_mtu = mtu;
+	}
+	hwrm_req_drop(bp, req);
+	return rc;
+}
+
+static int bnxt_vf_rep_open(struct net_device *dev)
+{
+	struct bnxt_vf_rep *vf_rep = netdev_priv(dev);
+	struct bnxt *bp = vf_rep->bp;
+
+	/* Enable link and TX only if the parent PF is open. */
+	if (netif_running(bp->dev)) {
+		netif_carrier_on(dev);
+		netif_tx_start_all_queues(dev);
+	}
+	return 0;
+}
+
+static int bnxt_vf_rep_close(struct net_device *dev)
+{
+	netif_carrier_off(dev);
+	netif_tx_disable(dev);
+
+	return 0;
+}
+
+static netdev_tx_t bnxt_vf_rep_xmit(struct sk_buff *skb,
+				    struct net_device *dev)
+{
+	struct bnxt_vf_rep *vf_rep = netdev_priv(dev);
+	int rc, len = skb->len;
+
+	skb_dst_drop(skb);
+	dst_hold((struct dst_entry *)vf_rep->dst);
+	skb_dst_set(skb, (struct dst_entry *)vf_rep->dst);
+	skb->dev = vf_rep->dst->u.port_info.lower_dev;
+
+	rc = dev_queue_xmit(skb);
+	if (!rc) {
+		vf_rep->tx_stats.packets++;
+		vf_rep->tx_stats.bytes += len;
+	}
+	return rc;
+}
+
+static void
+bnxt_vf_rep_get_stats64(struct net_device *dev,
+			struct rtnl_link_stats64 *stats)
+{
+	struct bnxt_vf_rep *vf_rep = netdev_priv(dev);
+
+	stats->rx_packets = vf_rep->rx_stats.packets;
+	stats->rx_bytes = vf_rep->rx_stats.bytes;
+	stats->tx_packets = vf_rep->tx_stats.packets;
+	stats->tx_bytes = vf_rep->tx_stats.bytes;
+}
+
+static int bnxt_vf_rep_setup_tc_block_cb(enum tc_setup_type type,
+					 void *type_data,
+					 void *cb_priv)
+{
+	struct bnxt_vf_rep *vf_rep = cb_priv;
+	struct bnxt *bp = vf_rep->bp;
+	int vf_fid = bp->pf.vf[vf_rep->vf_idx].fw_fid;
+
+	if (!bnxt_tc_flower_enabled(vf_rep->bp) ||
+	    !tc_cls_can_offload_and_chain0(bp->dev, type_data))
+		return -EOPNOTSUPP;
+
+	switch (type) {
+	case TC_SETUP_CLSFLOWER:
+		return bnxt_tc_setup_flower(bp, vf_fid, type_data);
+	default:
+		return -EOPNOTSUPP;
+	}
+}
+
+static LIST_HEAD(bnxt_vf_block_cb_list);
+
+static int bnxt_vf_rep_setup_tc(struct net_device *dev, enum tc_setup_type type,
+				void *type_data)
+{
+	struct bnxt_vf_rep *vf_rep = netdev_priv(dev);
+
+	switch (type) {
+	case TC_SETUP_BLOCK:
+		return flow_block_cb_setup_simple(type_data,
+						  &bnxt_vf_block_cb_list,
+						  bnxt_vf_rep_setup_tc_block_cb,
+						  vf_rep, vf_rep, true);
+	default:
+		return -EOPNOTSUPP;
+	}
+}
+
+struct net_device *bnxt_get_vf_rep(struct bnxt *bp, u16 cfa_code)
+{
+	u16 vf_idx;
+
+	if (cfa_code && bp->cfa_code_map && BNXT_PF(bp)) {
+		vf_idx = bp->cfa_code_map[cfa_code];
+		if (vf_idx != VF_IDX_INVALID)
+			return bp->vf_reps[vf_idx]->dev;
+	}
+	return NULL;
+}
+
+void bnxt_vf_rep_rx(struct bnxt *bp, struct sk_buff *skb)
+{
+	struct bnxt_vf_rep *vf_rep = netdev_priv(skb->dev);
+
+	vf_rep->rx_stats.bytes += skb->len;
+	vf_rep->rx_stats.packets++;
+
+	netif_receive_skb(skb);
+}
+
+static int bnxt_vf_rep_get_phys_port_name(struct net_device *dev, char *buf,
+					  size_t len)
+{
+	struct bnxt_vf_rep *vf_rep = netdev_priv(dev);
+	struct pci_dev *pf_pdev = vf_rep->bp->pdev;
+	int rc;
+
+	rc = snprintf(buf, len, "pf%dvf%d", PCI_FUNC(pf_pdev->devfn),
+		      vf_rep->vf_idx);
+	if (rc >= len)
+		return -EOPNOTSUPP;
+	return 0;
+}
+
+static void bnxt_vf_rep_get_drvinfo(struct net_device *dev,
+				    struct ethtool_drvinfo *info)
+{
+	strscpy(info->driver, DRV_MODULE_NAME, sizeof(info->driver));
+}
+
+static int bnxt_vf_rep_get_port_parent_id(struct net_device *dev,
+					  struct netdev_phys_item_id *ppid)
+{
+	struct bnxt_vf_rep *vf_rep = netdev_priv(dev);
+
+	/* as only PORT_PARENT_ID is supported currently use common code
+	 * between PF and VF-rep for now.
+	 */
+	return bnxt_get_port_parent_id(vf_rep->bp->dev, ppid);
+}
+
+static const struct ethtool_ops bnxt_vf_rep_ethtool_ops = {
+	.get_drvinfo		= bnxt_vf_rep_get_drvinfo
+};
+
+static const struct net_device_ops bnxt_vf_rep_netdev_ops = {
+	.ndo_open		= bnxt_vf_rep_open,
+	.ndo_stop		= bnxt_vf_rep_close,
+	.ndo_start_xmit		= bnxt_vf_rep_xmit,
+	.ndo_get_stats64	= bnxt_vf_rep_get_stats64,
+	.ndo_setup_tc		= bnxt_vf_rep_setup_tc,
+	.ndo_get_port_parent_id	= bnxt_vf_rep_get_port_parent_id,
+	.ndo_get_phys_port_name = bnxt_vf_rep_get_phys_port_name
+};
+
+bool bnxt_dev_is_vf_rep(struct net_device *dev)
+{
+	return dev->netdev_ops == &bnxt_vf_rep_netdev_ops;
+}
+
+/* Called when the parent PF interface is closed:
+ * As the mode transition from SWITCHDEV to LEGACY
+ * happens under the netdev instance lock this routine is safe
+ */
+void bnxt_vf_reps_close(struct bnxt *bp)
+{
+	struct bnxt_vf_rep *vf_rep;
+	u16 num_vfs, i;
+
+	if (bp->eswitch_mode != DEVLINK_ESWITCH_MODE_SWITCHDEV)
+		return;
+
+	num_vfs = pci_num_vf(bp->pdev);
+	for (i = 0; i < num_vfs; i++) {
+		vf_rep = bp->vf_reps[i];
+		if (netif_running(vf_rep->dev))
+			bnxt_vf_rep_close(vf_rep->dev);
+	}
+}
+
+/* Called when the parent PF interface is opened (re-opened):
+ * As the mode transition from SWITCHDEV to LEGACY
+ * happen under the netdev instance lock this routine is safe
+ */
+void bnxt_vf_reps_open(struct bnxt *bp)
+{
+	int i;
+
+	if (bp->eswitch_mode != DEVLINK_ESWITCH_MODE_SWITCHDEV)
+		return;
+
+	for (i = 0; i < pci_num_vf(bp->pdev); i++) {
+		/* Open the VF-Rep only if it is allocated in the FW */
+		if (bp->vf_reps[i]->tx_cfa_action != CFA_HANDLE_INVALID)
+			bnxt_vf_rep_open(bp->vf_reps[i]->dev);
+	}
+}
+
+static void __bnxt_free_one_vf_rep(struct bnxt *bp, struct bnxt_vf_rep *vf_rep)
+{
+	if (!vf_rep)
+		return;
+
+	if (vf_rep->dst) {
+		dst_release((struct dst_entry *)vf_rep->dst);
+		vf_rep->dst = NULL;
+	}
+	if (vf_rep->tx_cfa_action != CFA_HANDLE_INVALID) {
+		hwrm_cfa_vfr_free(bp, vf_rep->vf_idx);
+		vf_rep->tx_cfa_action = CFA_HANDLE_INVALID;
+	}
+}
+
+static void __bnxt_vf_reps_destroy(struct bnxt *bp)
+{
+	u16 num_vfs = pci_num_vf(bp->pdev);
+	struct bnxt_vf_rep *vf_rep;
+	int i;
+
+	for (i = 0; i < num_vfs; i++) {
+		vf_rep = bp->vf_reps[i];
+		if (vf_rep) {
+			__bnxt_free_one_vf_rep(bp, vf_rep);
+			if (vf_rep->dev) {
+				/* if register_netdev failed, then netdev_ops
+				 * would have been set to NULL
+				 */
+				if (vf_rep->dev->netdev_ops)
+					unregister_netdev(vf_rep->dev);
+				free_netdev(vf_rep->dev);
+			}
+		}
+	}
+
+	kfree(bp->vf_reps);
+	bp->vf_reps = NULL;
+}
+
+void bnxt_vf_reps_destroy(struct bnxt *bp)
+{
+	bool closed = false;
+
+	if (bp->eswitch_mode != DEVLINK_ESWITCH_MODE_SWITCHDEV)
+		return;
+
+	if (!bp->vf_reps)
+		return;
+
+	/* Ensure that parent PF's and VF-reps' RX/TX has been quiesced
+	 * before proceeding with VF-rep cleanup.
+	 */
+	netdev_lock(bp->dev);
+	if (netif_running(bp->dev)) {
+		bnxt_close_nic(bp, false, false);
+		closed = true;
+	}
+	/* un-publish cfa_code_map so that RX path can't see it anymore */
+	kfree(bp->cfa_code_map);
+	bp->cfa_code_map = NULL;
+
+	if (closed) {
+		/* Temporarily set legacy mode to avoid re-opening
+		 * representors and restore switchdev mode after that.
+		 */
+		bp->eswitch_mode = DEVLINK_ESWITCH_MODE_LEGACY;
+		bnxt_open_nic(bp, false, false);
+		bp->eswitch_mode = DEVLINK_ESWITCH_MODE_SWITCHDEV;
+	}
+	netdev_unlock(bp->dev);
+
+	/* Need to call vf_reps_destroy() outside of netdev instance lock
+	 * as unregister_netdev takes it
+	 */
+	__bnxt_vf_reps_destroy(bp);
+}
+
+/* Free the VF-Reps in firmware, during firmware hot-reset processing.
+ * Note that the VF-Rep netdevs are still active (not unregistered) during
+ * this process. As the mode transition from SWITCHDEV to LEGACY happens
+ * under the netdev instance lock this routine is safe.
+ */
+void bnxt_vf_reps_free(struct bnxt *bp)
+{
+	u16 num_vfs = pci_num_vf(bp->pdev);
+	int i;
+
+	if (bp->eswitch_mode != DEVLINK_ESWITCH_MODE_SWITCHDEV)
+		return;
+
+	for (i = 0; i < num_vfs; i++)
+		__bnxt_free_one_vf_rep(bp, bp->vf_reps[i]);
+}
+
+static int bnxt_alloc_vf_rep(struct bnxt *bp, struct bnxt_vf_rep *vf_rep,
+			     u16 *cfa_code_map)
+{
+	/* get cfa handles from FW */
+	if (hwrm_cfa_vfr_alloc(bp, vf_rep->vf_idx, &vf_rep->tx_cfa_action,
+			       &vf_rep->rx_cfa_code))
+		return -ENOLINK;
+
+	cfa_code_map[vf_rep->rx_cfa_code] = vf_rep->vf_idx;
+	vf_rep->dst = metadata_dst_alloc(0, METADATA_HW_PORT_MUX, GFP_KERNEL);
+	if (!vf_rep->dst)
+		return -ENOMEM;
+
+	/* only cfa_action is needed to mux a packet while TXing */
+	vf_rep->dst->u.port_info.port_id = vf_rep->tx_cfa_action;
+	vf_rep->dst->u.port_info.lower_dev = bp->dev;
+
+	return 0;
+}
+
+/* Allocate the VF-Reps in firmware, during firmware hot-reset processing.
+ * Note that the VF-Rep netdevs are still active (not unregistered) during
+ * this process. As the mode transition from SWITCHDEV to LEGACY happens
+ * under the netdev instance lock this routine is safe.
+ */
+int bnxt_vf_reps_alloc(struct bnxt *bp)
+{
+	u16 *cfa_code_map = bp->cfa_code_map, num_vfs = pci_num_vf(bp->pdev);
+	struct bnxt_vf_rep *vf_rep;
+	int rc, i;
+
+	if (bp->eswitch_mode != DEVLINK_ESWITCH_MODE_SWITCHDEV)
+		return 0;
+
+	if (!cfa_code_map)
+		return -EINVAL;
+
+	for (i = 0; i < MAX_CFA_CODE; i++)
+		cfa_code_map[i] = VF_IDX_INVALID;
+
+	for (i = 0; i < num_vfs; i++) {
+		vf_rep = bp->vf_reps[i];
+		vf_rep->vf_idx = i;
+
+		rc = bnxt_alloc_vf_rep(bp, vf_rep, cfa_code_map);
+		if (rc)
+			goto err;
+	}
+
+	return 0;
+
+err:
+	netdev_info(bp->dev, "%s error=%d\n", __func__, rc);
+	bnxt_vf_reps_free(bp);
+	return rc;
+}
+
+/* Use the OUI of the PF's perm addr and report the same mac addr
+ * for the same VF-rep each time
+ */
+static void bnxt_vf_rep_eth_addr_gen(u8 *src_mac, u16 vf_idx, u8 *mac)
+{
+	u32 addr;
+
+	ether_addr_copy(mac, src_mac);
+
+	addr = jhash(src_mac, ETH_ALEN, 0) + vf_idx;
+	mac[3] = (u8)(addr & 0xFF);
+	mac[4] = (u8)((addr >> 8) & 0xFF);
+	mac[5] = (u8)((addr >> 16) & 0xFF);
+}
+
+static void bnxt_vf_rep_netdev_init(struct bnxt *bp, struct bnxt_vf_rep *vf_rep,
+				    struct net_device *dev)
+{
+	struct net_device *pf_dev = bp->dev;
+	u16 max_mtu;
+
+	SET_NETDEV_DEV(dev, &bp->pdev->dev);
+	dev->netdev_ops = &bnxt_vf_rep_netdev_ops;
+	dev->ethtool_ops = &bnxt_vf_rep_ethtool_ops;
+	/* Just inherit all the featues of the parent PF as the VF-R
+	 * uses the RX/TX rings of the parent PF
+	 */
+	dev->hw_features = pf_dev->hw_features;
+	dev->gso_partial_features = pf_dev->gso_partial_features;
+	dev->vlan_features = pf_dev->vlan_features;
+	dev->hw_enc_features = pf_dev->hw_enc_features;
+	dev->features |= pf_dev->features;
+	bnxt_vf_rep_eth_addr_gen(bp->pf.mac_addr, vf_rep->vf_idx,
+				 dev->perm_addr);
+	eth_hw_addr_set(dev, dev->perm_addr);
+	/* Set VF-Rep's max-mtu to the corresponding VF's max-mtu */
+	if (!bnxt_hwrm_vfr_qcfg(bp, vf_rep, &max_mtu))
+		dev->max_mtu = max_mtu;
+	dev->min_mtu = ETH_ZLEN;
+}
+
+int bnxt_vf_reps_create(struct bnxt *bp)
+{
+	u16 *cfa_code_map = NULL, num_vfs = pci_num_vf(bp->pdev);
+	struct bnxt_vf_rep *vf_rep;
+	struct net_device *dev;
+	int rc, i;
+
+	if (!(bp->flags & BNXT_FLAG_DSN_VALID))
+		return -ENODEV;
+
+	bp->vf_reps = kcalloc(num_vfs, sizeof(vf_rep), GFP_KERNEL);
+	if (!bp->vf_reps)
+		return -ENOMEM;
+
+	/* storage for cfa_code to vf-idx mapping */
+	cfa_code_map = kmalloc_array(MAX_CFA_CODE, sizeof(*bp->cfa_code_map),
+				     GFP_KERNEL);
+	if (!cfa_code_map) {
+		rc = -ENOMEM;
+		goto err;
+	}
+	for (i = 0; i < MAX_CFA_CODE; i++)
+		cfa_code_map[i] = VF_IDX_INVALID;
+
+	for (i = 0; i < num_vfs; i++) {
+		dev = alloc_etherdev(sizeof(*vf_rep));
+		if (!dev) {
+			rc = -ENOMEM;
+			goto err;
+		}
+
+		vf_rep = netdev_priv(dev);
+		bp->vf_reps[i] = vf_rep;
+		vf_rep->dev = dev;
+		vf_rep->bp = bp;
+		vf_rep->vf_idx = i;
+		vf_rep->tx_cfa_action = CFA_HANDLE_INVALID;
+
+		rc = bnxt_alloc_vf_rep(bp, vf_rep, cfa_code_map);
+		if (rc)
+			goto err;
+
+		bnxt_vf_rep_netdev_init(bp, vf_rep, dev);
+		rc = register_netdev(dev);
+		if (rc) {
+			/* no need for unregister_netdev in cleanup */
+			dev->netdev_ops = NULL;
+			goto err;
+		}
+	}
+
+	/* publish cfa_code_map only after all VF-reps have been initialized */
+	bp->cfa_code_map = cfa_code_map;
+	netif_keep_dst(bp->dev);
+	return 0;
+
+err:
+	netdev_info(bp->dev, "%s error=%d\n", __func__, rc);
+	kfree(cfa_code_map);
+	__bnxt_vf_reps_destroy(bp);
+	return rc;
+}
+
+/* Devlink related routines */
+int bnxt_dl_eswitch_mode_get(struct devlink *devlink, u16 *mode)
+{
+	struct bnxt *bp = bnxt_get_bp_from_dl(devlink);
+
+	*mode = bp->eswitch_mode;
+	return 0;
+}
+
+int bnxt_dl_eswitch_mode_set(struct devlink *devlink, u16 mode,
+			     struct netlink_ext_ack *extack)
+{
+	struct bnxt *bp = bnxt_get_bp_from_dl(devlink);
+	int ret = 0;
+
+	if (bp->eswitch_mode == mode) {
+		netdev_info(bp->dev, "already in %s eswitch mode\n",
+			    mode == DEVLINK_ESWITCH_MODE_LEGACY ?
+			    "legacy" : "switchdev");
+		return -EINVAL;
+	}
+
+	switch (mode) {
+	case DEVLINK_ESWITCH_MODE_LEGACY:
+		bnxt_vf_reps_destroy(bp);
+		break;
+
+	case DEVLINK_ESWITCH_MODE_SWITCHDEV:
+		if (bp->hwrm_spec_code < 0x10803) {
+			netdev_warn(bp->dev, "FW does not support SRIOV E-Switch SWITCHDEV mode\n");
+			return -ENOTSUPP;
+		}
+
+		/* Create representors for existing VFs */
+		if (pci_num_vf(bp->pdev) > 0)
+			ret = bnxt_vf_reps_create(bp);
+		break;
+
+	default:
+		return -EINVAL;
+	}
+
+	if (!ret)
+		bp->eswitch_mode = mode;
+
+	return ret;
+}
+
+#endif
diff --git a/drivers/net/ethernet/broadcom/bnxt/bnxt_vfr.h b/drivers/net/ethernet/broadcom/bnxt/bnxt_vfr.h
new file mode 100644
index 000000000000..33a965631d0b
--- /dev/null
+++ b/drivers/net/ethernet/broadcom/bnxt/bnxt_vfr.h
@@ -0,0 +1,83 @@
+/* Broadcom NetXtreme-C/E network driver.
+ *
+ * Copyright (c) 2016-2017 Broadcom Limited
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation.
+ */
+
+#ifndef BNXT_VFR_H
+#define BNXT_VFR_H
+
+#ifdef CONFIG_BNXT_SRIOV
+
+#define	MAX_CFA_CODE			65536
+
+int bnxt_vf_reps_create(struct bnxt *bp);
+void bnxt_vf_reps_destroy(struct bnxt *bp);
+void bnxt_vf_reps_close(struct bnxt *bp);
+void bnxt_vf_reps_open(struct bnxt *bp);
+void bnxt_vf_rep_rx(struct bnxt *bp, struct sk_buff *skb);
+struct net_device *bnxt_get_vf_rep(struct bnxt *bp, u16 cfa_code);
+int bnxt_vf_reps_alloc(struct bnxt *bp);
+void bnxt_vf_reps_free(struct bnxt *bp);
+
+static inline u16 bnxt_vf_rep_get_fid(struct net_device *dev)
+{
+	struct bnxt_vf_rep *vf_rep = netdev_priv(dev);
+	struct bnxt *bp = vf_rep->bp;
+
+	return bp->pf.vf[vf_rep->vf_idx].fw_fid;
+}
+
+bool bnxt_dev_is_vf_rep(struct net_device *dev);
+int bnxt_dl_eswitch_mode_get(struct devlink *devlink, u16 *mode);
+int bnxt_dl_eswitch_mode_set(struct devlink *devlink, u16 mode,
+			     struct netlink_ext_ack *extack);
+
+#else
+
+static inline int bnxt_vf_reps_create(struct bnxt *bp)
+{
+	return 0;
+}
+
+static inline void bnxt_vf_reps_close(struct bnxt *bp)
+{
+}
+
+static inline void bnxt_vf_reps_open(struct bnxt *bp)
+{
+}
+
+static inline void bnxt_vf_rep_rx(struct bnxt *bp, struct sk_buff *skb)
+{
+}
+
+static inline struct net_device *bnxt_get_vf_rep(struct bnxt *bp, u16 cfa_code)
+{
+	return NULL;
+}
+
+static inline u16 bnxt_vf_rep_get_fid(struct net_device *dev)
+{
+	return 0;
+}
+
+static inline bool bnxt_dev_is_vf_rep(struct net_device *dev)
+{
+	return false;
+}
+
+static inline int bnxt_vf_reps_alloc(struct bnxt *bp)
+{
+	return 0;
+}
+
+static inline void bnxt_vf_reps_free(struct bnxt *bp)
+{
+}
+
+#endif /* CONFIG_BNXT_SRIOV */
+#endif /* BNXT_VFR_H */
diff --git a/drivers/net/ethernet/broadcom/bnxt/bnxt_xdp.c b/drivers/net/ethernet/broadcom/bnxt/bnxt_xdp.c
index 8ce793a0d030..3e77a96e5a3e 100644
--- a/drivers/net/ethernet/broadcom/bnxt/bnxt_xdp.c
+++ b/drivers/net/ethernet/broadcom/bnxt/bnxt_xdp.c
@@ -15,55 +15,206 @@
 #include <linux/bpf.h>
 #include <linux/bpf_trace.h>
 #include <linux/filter.h>
-#include "bnxt_hsi.h"
+#include <net/netdev_lock.h>
+#include <net/page_pool/helpers.h>
+#include <linux/bnxt/hsi.h>
 #include "bnxt.h"
 #include "bnxt_xdp.h"
 
-void bnxt_xmit_xdp(struct bnxt *bp, struct bnxt_tx_ring_info *txr,
-		   dma_addr_t mapping, u32 len, u16 rx_prod)
+DEFINE_STATIC_KEY_FALSE(bnxt_xdp_locking_key);
+
+struct bnxt_sw_tx_bd *bnxt_xmit_bd(struct bnxt *bp,
+				   struct bnxt_tx_ring_info *txr,
+				   dma_addr_t mapping, u32 len,
+				   struct xdp_buff *xdp)
 {
+	struct skb_shared_info *sinfo;
 	struct bnxt_sw_tx_bd *tx_buf;
 	struct tx_bd *txbd;
+	int num_frags = 0;
 	u32 flags;
 	u16 prod;
+	int i;
+
+	if (xdp && xdp_buff_has_frags(xdp)) {
+		sinfo = xdp_get_shared_info_from_buff(xdp);
+		num_frags = sinfo->nr_frags;
+	}
 
+	/* fill up the first buffer */
 	prod = txr->tx_prod;
-	tx_buf = &txr->tx_buf_ring[prod];
-	tx_buf->rx_prod = rx_prod;
+	tx_buf = &txr->tx_buf_ring[RING_TX(bp, prod)];
+	tx_buf->nr_frags = num_frags;
+	if (xdp)
+		tx_buf->page = virt_to_head_page(xdp->data);
 
-	txbd = &txr->tx_desc_ring[TX_RING(prod)][TX_IDX(prod)];
-	flags = (len << TX_BD_LEN_SHIFT) | (1 << TX_BD_FLAGS_BD_CNT_SHIFT) |
-		TX_BD_FLAGS_PACKET_END | bnxt_lhint_arr[len >> 9];
+	txbd = &txr->tx_desc_ring[TX_RING(bp, prod)][TX_IDX(prod)];
+	flags = (len << TX_BD_LEN_SHIFT) | TX_BD_CNT(num_frags + 1) |
+		bnxt_lhint_arr[len >> 9];
 	txbd->tx_bd_len_flags_type = cpu_to_le32(flags);
-	txbd->tx_bd_opaque = prod;
+	txbd->tx_bd_opaque = SET_TX_OPAQUE(bp, txr, prod, 1 + num_frags);
 	txbd->tx_bd_haddr = cpu_to_le64(mapping);
 
+	/* now let us fill up the frags into the next buffers */
+	for (i = 0; i < num_frags ; i++) {
+		skb_frag_t *frag = &sinfo->frags[i];
+		struct bnxt_sw_tx_bd *frag_tx_buf;
+		dma_addr_t frag_mapping;
+		int frag_len;
+
+		prod = NEXT_TX(prod);
+		WRITE_ONCE(txr->tx_prod, prod);
+
+		/* first fill up the first buffer */
+		frag_tx_buf = &txr->tx_buf_ring[RING_TX(bp, prod)];
+		frag_tx_buf->page = skb_frag_page(frag);
+
+		txbd = &txr->tx_desc_ring[TX_RING(bp, prod)][TX_IDX(prod)];
+
+		frag_len = skb_frag_size(frag);
+		flags = frag_len << TX_BD_LEN_SHIFT;
+		txbd->tx_bd_len_flags_type = cpu_to_le32(flags);
+		frag_mapping = page_pool_get_dma_addr(skb_frag_page(frag)) +
+			       skb_frag_off(frag);
+		txbd->tx_bd_haddr = cpu_to_le64(frag_mapping);
+
+		len = frag_len;
+	}
+
+	flags &= ~TX_BD_LEN;
+	txbd->tx_bd_len_flags_type = cpu_to_le32(((len) << TX_BD_LEN_SHIFT) | flags |
+			TX_BD_FLAGS_PACKET_END);
+	/* Sync TX BD */
+	wmb();
 	prod = NEXT_TX(prod);
-	txr->tx_prod = prod;
+	WRITE_ONCE(txr->tx_prod, prod);
+
+	return tx_buf;
+}
+
+static void __bnxt_xmit_xdp(struct bnxt *bp, struct bnxt_tx_ring_info *txr,
+			    dma_addr_t mapping, u32 len, u16 rx_prod,
+			    struct xdp_buff *xdp)
+{
+	struct bnxt_sw_tx_bd *tx_buf;
+
+	tx_buf = bnxt_xmit_bd(bp, txr, mapping, len, xdp);
+	tx_buf->rx_prod = rx_prod;
+	tx_buf->action = XDP_TX;
+
+}
+
+static void __bnxt_xmit_xdp_redirect(struct bnxt *bp,
+				     struct bnxt_tx_ring_info *txr,
+				     dma_addr_t mapping, u32 len,
+				     struct xdp_frame *xdpf)
+{
+	struct bnxt_sw_tx_bd *tx_buf;
+
+	tx_buf = bnxt_xmit_bd(bp, txr, mapping, len, NULL);
+	tx_buf->action = XDP_REDIRECT;
+	tx_buf->xdpf = xdpf;
+	dma_unmap_addr_set(tx_buf, mapping, mapping);
+	dma_unmap_len_set(tx_buf, len, len);
 }
 
-void bnxt_tx_int_xdp(struct bnxt *bp, struct bnxt_napi *bnapi, int nr_pkts)
+void bnxt_tx_int_xdp(struct bnxt *bp, struct bnxt_napi *bnapi, int budget)
 {
-	struct bnxt_tx_ring_info *txr = bnapi->tx_ring;
+	struct bnxt_tx_ring_info *txr = bnapi->tx_ring[0];
 	struct bnxt_rx_ring_info *rxr = bnapi->rx_ring;
+	u16 tx_hw_cons = txr->tx_hw_cons;
+	bool rx_doorbell_needed = false;
 	struct bnxt_sw_tx_bd *tx_buf;
 	u16 tx_cons = txr->tx_cons;
 	u16 last_tx_cons = tx_cons;
-	u16 rx_prod;
-	int i;
+	int j, frags;
+
+	if (!budget)
+		return;
+
+	while (RING_TX(bp, tx_cons) != tx_hw_cons) {
+		tx_buf = &txr->tx_buf_ring[RING_TX(bp, tx_cons)];
 
-	for (i = 0; i < nr_pkts; i++) {
-		last_tx_cons = tx_cons;
+		if (tx_buf->action == XDP_REDIRECT) {
+			struct pci_dev *pdev = bp->pdev;
+
+			dma_unmap_single(&pdev->dev,
+					 dma_unmap_addr(tx_buf, mapping),
+					 dma_unmap_len(tx_buf, len),
+					 DMA_TO_DEVICE);
+			xdp_return_frame(tx_buf->xdpf);
+			tx_buf->action = 0;
+			tx_buf->xdpf = NULL;
+		} else if (tx_buf->action == XDP_TX) {
+			tx_buf->action = 0;
+			rx_doorbell_needed = true;
+			last_tx_cons = tx_cons;
+
+			frags = tx_buf->nr_frags;
+			for (j = 0; j < frags; j++) {
+				tx_cons = NEXT_TX(tx_cons);
+				tx_buf = &txr->tx_buf_ring[RING_TX(bp, tx_cons)];
+				page_pool_recycle_direct(rxr->page_pool, tx_buf->page);
+			}
+		} else {
+			bnxt_sched_reset_txr(bp, txr, tx_cons);
+			return;
+		}
 		tx_cons = NEXT_TX(tx_cons);
 	}
-	txr->tx_cons = tx_cons;
-	if (bnxt_tx_avail(bp, txr) == bp->tx_ring_size) {
-		rx_prod = rxr->rx_prod;
-	} else {
-		tx_buf = &txr->tx_buf_ring[last_tx_cons];
-		rx_prod = tx_buf->rx_prod;
+
+	bnapi->events &= ~BNXT_TX_CMP_EVENT;
+	WRITE_ONCE(txr->tx_cons, tx_cons);
+	if (rx_doorbell_needed) {
+		tx_buf = &txr->tx_buf_ring[RING_TX(bp, last_tx_cons)];
+		bnxt_db_write(bp, &rxr->rx_db, tx_buf->rx_prod);
+
+	}
+}
+
+bool bnxt_xdp_attached(struct bnxt *bp, struct bnxt_rx_ring_info *rxr)
+{
+	struct bpf_prog *xdp_prog = READ_ONCE(rxr->xdp_prog);
+
+	return !!xdp_prog;
+}
+
+void bnxt_xdp_buff_init(struct bnxt *bp, struct bnxt_rx_ring_info *rxr,
+			u16 cons, u8 *data_ptr, unsigned int len,
+			struct xdp_buff *xdp)
+{
+	u32 buflen = BNXT_RX_PAGE_SIZE;
+	struct bnxt_sw_rx_bd *rx_buf;
+	struct pci_dev *pdev;
+	dma_addr_t mapping;
+	u32 offset;
+
+	pdev = bp->pdev;
+	rx_buf = &rxr->rx_buf_ring[cons];
+	offset = bp->rx_offset;
+
+	mapping = rx_buf->mapping - bp->rx_dma_offset;
+	dma_sync_single_for_cpu(&pdev->dev, mapping + offset, len, bp->rx_dir);
+
+	xdp_init_buff(xdp, buflen, &rxr->xdp_rxq);
+	xdp_prepare_buff(xdp, data_ptr - offset, offset, len, true);
+}
+
+void bnxt_xdp_buff_frags_free(struct bnxt_rx_ring_info *rxr,
+			      struct xdp_buff *xdp)
+{
+	struct skb_shared_info *shinfo;
+	int i;
+
+	if (!xdp || !xdp_buff_has_frags(xdp))
+		return;
+	shinfo = xdp_get_shared_info_from_buff(xdp);
+	for (i = 0; i < shinfo->nr_frags; i++) {
+		struct page *page = skb_frag_page(&shinfo->frags[i]);
+
+		page_pool_recycle_direct(rxr->page_pool, page);
 	}
-	bnxt_db_write(bp, rxr->rx_doorbell, DB_KEY_RX | rx_prod);
+	shinfo->nr_frags = 0;
 }
 
 /* returns the following:
@@ -71,14 +222,15 @@ void bnxt_tx_int_xdp(struct bnxt *bp, struct bnxt_napi *bnapi, int nr_pkts)
  * false   - packet should be passed to the stack.
  */
 bool bnxt_rx_xdp(struct bnxt *bp, struct bnxt_rx_ring_info *rxr, u16 cons,
-		 struct page *page, u8 **data_ptr, unsigned int *len, u8 *event)
+		 struct xdp_buff *xdp, struct page *page, u8 **data_ptr,
+		 unsigned int *len, u8 *event)
 {
 	struct bpf_prog *xdp_prog = READ_ONCE(rxr->xdp_prog);
 	struct bnxt_tx_ring_info *txr;
 	struct bnxt_sw_rx_bd *rx_buf;
 	struct pci_dev *pdev;
-	struct xdp_buff xdp;
 	dma_addr_t mapping;
+	u32 tx_needed = 1;
 	void *orig_data;
 	u32 tx_avail;
 	u32 offset;
@@ -88,21 +240,13 @@ bool bnxt_rx_xdp(struct bnxt *bp, struct bnxt_rx_ring_info *rxr, u16 cons,
 		return false;
 
 	pdev = bp->pdev;
-	txr = rxr->bnapi->tx_ring;
-	rx_buf = &rxr->rx_buf_ring[cons];
 	offset = bp->rx_offset;
 
-	xdp.data_hard_start = *data_ptr - offset;
-	xdp.data = *data_ptr;
-	xdp.data_end = *data_ptr + *len;
-	orig_data = xdp.data;
-	mapping = rx_buf->mapping - bp->rx_dma_offset;
-
-	dma_sync_single_for_cpu(&pdev->dev, mapping + offset, *len, bp->rx_dir);
+	txr = rxr->bnapi->tx_ring[0];
+	/* BNXT_RX_PAGE_MODE(bp) when XDP enabled */
+	orig_data = xdp->data;
 
-	rcu_read_lock();
-	act = bpf_prog_run_xdp(xdp_prog, &xdp);
-	rcu_read_unlock();
+	act = bpf_prog_run_xdp(xdp_prog, xdp);
 
 	tx_avail = bnxt_tx_avail(bp, txr);
 	/* If the tx ring is not full, we must not update the rx producer yet
@@ -111,54 +255,151 @@ bool bnxt_rx_xdp(struct bnxt *bp, struct bnxt_rx_ring_info *rxr, u16 cons,
 	if (tx_avail != bp->tx_ring_size)
 		*event &= ~BNXT_RX_EVENT;
 
-	if (orig_data != xdp.data) {
-		offset = xdp.data - xdp.data_hard_start;
-		*data_ptr = xdp.data_hard_start + offset;
-		*len = xdp.data_end - xdp.data;
+	*len = xdp->data_end - xdp->data;
+	if (orig_data != xdp->data) {
+		offset = xdp->data - xdp->data_hard_start;
+		*data_ptr = xdp->data_hard_start + offset;
 	}
+
 	switch (act) {
 	case XDP_PASS:
 		return false;
 
 	case XDP_TX:
-		if (tx_avail < 1) {
+		rx_buf = &rxr->rx_buf_ring[cons];
+		mapping = rx_buf->mapping - bp->rx_dma_offset;
+		*event &= BNXT_TX_CMP_EVENT;
+
+		if (unlikely(xdp_buff_has_frags(xdp))) {
+			struct skb_shared_info *sinfo = xdp_get_shared_info_from_buff(xdp);
+
+			tx_needed += sinfo->nr_frags;
+			*event = BNXT_AGG_EVENT;
+		}
+
+		if (tx_avail < tx_needed) {
 			trace_xdp_exception(bp->dev, xdp_prog, act);
+			bnxt_xdp_buff_frags_free(rxr, xdp);
 			bnxt_reuse_rx_data(rxr, cons, page);
 			return true;
 		}
 
-		*event = BNXT_TX_EVENT;
 		dma_sync_single_for_device(&pdev->dev, mapping + offset, *len,
 					   bp->rx_dir);
-		bnxt_xmit_xdp(bp, txr, mapping + offset, *len,
-			      NEXT_RX(rxr->rx_prod));
+
+		*event |= BNXT_TX_EVENT;
+		__bnxt_xmit_xdp(bp, txr, mapping + offset, *len,
+				NEXT_RX(rxr->rx_prod), xdp);
 		bnxt_reuse_rx_data(rxr, cons, page);
 		return true;
+	case XDP_REDIRECT:
+		/* if we are calling this here then we know that the
+		 * redirect is coming from a frame received by the
+		 * bnxt_en driver.
+		 */
+
+		/* if we are unable to allocate a new buffer, abort and reuse */
+		if (bnxt_alloc_rx_data(bp, rxr, rxr->rx_prod, GFP_ATOMIC)) {
+			trace_xdp_exception(bp->dev, xdp_prog, act);
+			bnxt_xdp_buff_frags_free(rxr, xdp);
+			bnxt_reuse_rx_data(rxr, cons, page);
+			return true;
+		}
+
+		if (xdp_do_redirect(bp->dev, xdp, xdp_prog)) {
+			trace_xdp_exception(bp->dev, xdp_prog, act);
+			page_pool_recycle_direct(rxr->page_pool, page);
+			return true;
+		}
+
+		*event |= BNXT_REDIRECT_EVENT;
+		break;
 	default:
-		bpf_warn_invalid_xdp_action(act);
-		/* Fall thru */
+		bpf_warn_invalid_xdp_action(bp->dev, xdp_prog, act);
+		fallthrough;
 	case XDP_ABORTED:
 		trace_xdp_exception(bp->dev, xdp_prog, act);
-		/* Fall thru */
+		fallthrough;
 	case XDP_DROP:
+		bnxt_xdp_buff_frags_free(rxr, xdp);
 		bnxt_reuse_rx_data(rxr, cons, page);
 		break;
 	}
 	return true;
 }
 
-/* Under rtnl_lock */
+int bnxt_xdp_xmit(struct net_device *dev, int num_frames,
+		  struct xdp_frame **frames, u32 flags)
+{
+	struct bnxt *bp = netdev_priv(dev);
+	struct bpf_prog *xdp_prog = READ_ONCE(bp->xdp_prog);
+	struct pci_dev *pdev = bp->pdev;
+	struct bnxt_tx_ring_info *txr;
+	dma_addr_t mapping;
+	int nxmit = 0;
+	int ring;
+	int i;
+
+	if (!test_bit(BNXT_STATE_OPEN, &bp->state) ||
+	    !bp->tx_nr_rings_xdp ||
+	    !xdp_prog)
+		return -EINVAL;
+
+	ring = smp_processor_id() % bp->tx_nr_rings_xdp;
+	txr = &bp->tx_ring[ring];
+
+	if (READ_ONCE(txr->dev_state) == BNXT_DEV_STATE_CLOSING)
+		return -EINVAL;
+
+	if (static_branch_unlikely(&bnxt_xdp_locking_key))
+		spin_lock(&txr->xdp_tx_lock);
+
+	for (i = 0; i < num_frames; i++) {
+		struct xdp_frame *xdp = frames[i];
+
+		if (!bnxt_tx_avail(bp, txr))
+			break;
+
+		mapping = dma_map_single(&pdev->dev, xdp->data, xdp->len,
+					 DMA_TO_DEVICE);
+
+		if (dma_mapping_error(&pdev->dev, mapping))
+			break;
+
+		__bnxt_xmit_xdp_redirect(bp, txr, mapping, xdp->len, xdp);
+		nxmit++;
+	}
+
+	if (flags & XDP_XMIT_FLUSH) {
+		/* Sync BD data before updating doorbell */
+		wmb();
+		bnxt_db_write(bp, &txr->tx_db, txr->tx_prod);
+	}
+
+	if (static_branch_unlikely(&bnxt_xdp_locking_key))
+		spin_unlock(&txr->xdp_tx_lock);
+
+	return nxmit;
+}
+
 static int bnxt_xdp_set(struct bnxt *bp, struct bpf_prog *prog)
 {
 	struct net_device *dev = bp->dev;
-	int tx_xdp = 0, rc, tc;
+	int tx_xdp = 0, tx_cp, rc, tc;
 	struct bpf_prog *old;
 
-	if (prog && bp->dev->mtu > BNXT_MAX_PAGE_MODE_MTU) {
-		netdev_warn(dev, "MTU %d larger than largest XDP supported MTU %d.\n",
+	netdev_assert_locked(dev);
+
+	if (prog && !prog->aux->xdp_has_frags &&
+	    bp->dev->mtu > BNXT_MAX_PAGE_MODE_MTU) {
+		netdev_warn(dev, "MTU %d larger than %d without XDP frag support.\n",
 			    bp->dev->mtu, BNXT_MAX_PAGE_MODE_MTU);
 		return -EOPNOTSUPP;
 	}
+	if (prog && bp->flags & BNXT_FLAG_HDS) {
+		netdev_warn(dev, "XDP is disallowed when HDS is enabled.\n");
+		return -EOPNOTSUPP;
+	}
 	if (!(bp->flags & BNXT_FLAG_SHARED_RINGS)) {
 		netdev_warn(dev, "ethtool rx/tx channels must be combined to support XDP.\n");
 		return -EOPNOTSUPP;
@@ -166,11 +407,11 @@ static int bnxt_xdp_set(struct bnxt *bp, struct bpf_prog *prog)
 	if (prog)
 		tx_xdp = bp->rx_nr_rings;
 
-	tc = netdev_get_num_tc(dev);
+	tc = bp->num_tc;
 	if (!tc)
 		tc = 1;
-	rc = bnxt_reserve_rings(bp, bp->tx_nr_rings_per_tc, bp->rx_nr_rings,
-				tc, tx_xdp);
+	rc = bnxt_check_rings(bp, bp->tx_nr_rings_per_tc, bp->rx_nr_rings,
+			      true, tc, tx_xdp);
 	if (rc) {
 		netdev_warn(dev, "Unable to reserve enough TX rings to support XDP.\n");
 		return rc;
@@ -184,20 +425,15 @@ static int bnxt_xdp_set(struct bnxt *bp, struct bpf_prog *prog)
 
 	if (prog) {
 		bnxt_set_rx_skb_mode(bp, true);
+		xdp_features_set_redirect_target_locked(dev, true);
 	} else {
-		int rx, tx;
-
+		xdp_features_clear_redirect_target_locked(dev);
 		bnxt_set_rx_skb_mode(bp, false);
-		bnxt_get_max_rings(bp, &rx, &tx, true);
-		if (rx > 1) {
-			bp->flags &= ~BNXT_FLAG_NO_AGG_RINGS;
-			bp->dev->hw_features |= NETIF_F_LRO;
-		}
 	}
 	bp->tx_nr_rings_xdp = tx_xdp;
 	bp->tx_nr_rings = bp->tx_nr_rings_per_tc * tc + tx_xdp;
-	bp->cp_nr_rings = max_t(int, bp->tx_nr_rings, bp->rx_nr_rings);
-	bp->num_stat_ctxs = bp->cp_nr_rings;
+	tx_cp = bnxt_num_tx_to_cp(bp, bp->tx_nr_rings);
+	bp->cp_nr_rings = max_t(int, tx_cp, bp->rx_nr_rings);
 	bnxt_set_tpa_flags(bp);
 	bnxt_set_ring_params(bp);
 
@@ -207,7 +443,7 @@ static int bnxt_xdp_set(struct bnxt *bp, struct bpf_prog *prog)
 	return 0;
 }
 
-int bnxt_xdp(struct net_device *dev, struct netdev_xdp *xdp)
+int bnxt_xdp(struct net_device *dev, struct netdev_bpf *xdp)
 {
 	struct bnxt *bp = netdev_priv(dev);
 	int rc;
@@ -216,13 +452,24 @@ int bnxt_xdp(struct net_device *dev, struct netdev_xdp *xdp)
 	case XDP_SETUP_PROG:
 		rc = bnxt_xdp_set(bp, xdp->prog);
 		break;
-	case XDP_QUERY_PROG:
-		xdp->prog_attached = !!bp->xdp_prog;
-		rc = 0;
-		break;
 	default:
 		rc = -EINVAL;
 		break;
 	}
 	return rc;
 }
+
+struct sk_buff *
+bnxt_xdp_build_skb(struct bnxt *bp, struct sk_buff *skb, u8 num_frags,
+		   struct page_pool *pool, struct xdp_buff *xdp)
+{
+	struct skb_shared_info *sinfo = xdp_get_shared_info_from_buff(xdp);
+
+	if (!skb)
+		return NULL;
+
+	xdp_update_skb_frags_info(skb, num_frags, sinfo->xdp_frags_size,
+				  BNXT_RX_PAGE_SIZE * num_frags,
+				  xdp_buff_get_skb_flags(xdp));
+	return skb;
+}
diff --git a/drivers/net/ethernet/broadcom/bnxt/bnxt_xdp.h b/drivers/net/ethernet/broadcom/bnxt/bnxt_xdp.h
index 12a5ad66b564..220285e190fc 100644
--- a/drivers/net/ethernet/broadcom/bnxt/bnxt_xdp.h
+++ b/drivers/net/ethernet/broadcom/bnxt/bnxt_xdp.h
@@ -10,12 +10,28 @@
 #ifndef BNXT_XDP_H
 #define BNXT_XDP_H
 
-void bnxt_xmit_xdp(struct bnxt *bp, struct bnxt_tx_ring_info *txr,
-		   dma_addr_t mapping, u32 len, u16 rx_prod);
-void bnxt_tx_int_xdp(struct bnxt *bp, struct bnxt_napi *bnapi, int nr_pkts);
+DECLARE_STATIC_KEY_FALSE(bnxt_xdp_locking_key);
+
+struct bnxt_sw_tx_bd *bnxt_xmit_bd(struct bnxt *bp,
+				   struct bnxt_tx_ring_info *txr,
+				   dma_addr_t mapping, u32 len,
+				   struct xdp_buff *xdp);
+void bnxt_tx_int_xdp(struct bnxt *bp, struct bnxt_napi *bnapi, int budget);
 bool bnxt_rx_xdp(struct bnxt *bp, struct bnxt_rx_ring_info *rxr, u16 cons,
-		 struct page *page, u8 **data_ptr, unsigned int *len,
-		 u8 *event);
-int bnxt_xdp(struct net_device *dev, struct netdev_xdp *xdp);
+		 struct xdp_buff *xdp, struct page *page, u8 **data_ptr,
+		 unsigned int *len, u8 *event);
+int bnxt_xdp(struct net_device *dev, struct netdev_bpf *xdp);
+int bnxt_xdp_xmit(struct net_device *dev, int num_frames,
+		  struct xdp_frame **frames, u32 flags);
+
+bool bnxt_xdp_attached(struct bnxt *bp, struct bnxt_rx_ring_info *rxr);
 
+void bnxt_xdp_buff_init(struct bnxt *bp, struct bnxt_rx_ring_info *rxr,
+			u16 cons, u8 *data_ptr, unsigned int len,
+			struct xdp_buff *xdp);
+void bnxt_xdp_buff_frags_free(struct bnxt_rx_ring_info *rxr,
+			      struct xdp_buff *xdp);
+struct sk_buff *bnxt_xdp_build_skb(struct bnxt *bp, struct sk_buff *skb,
+				   u8 num_frags, struct page_pool *pool,
+				   struct xdp_buff *xdp);
 #endif
diff --git a/drivers/net/ethernet/broadcom/cnic.c b/drivers/net/ethernet/broadcom/cnic.c
index cec94bbb2ea5..6e97a5a7daaf 100644
--- a/drivers/net/ethernet/broadcom/cnic.c
+++ b/drivers/net/ethernet/broadcom/cnic.c
@@ -31,6 +31,7 @@
 #include <linux/if_vlan.h>
 #include <linux/prefetch.h>
 #include <linux/random.h>
+#include <linux/workqueue.h>
 #if IS_ENABLED(CONFIG_VLAN_8021Q)
 #define BCM_VLAN 1
 #endif
@@ -660,7 +661,7 @@ static int cnic_init_id_tbl(struct cnic_id_tbl *id_tbl, u32 size, u32 start_id,
 	id_tbl->max = size;
 	id_tbl->next = next;
 	spin_lock_init(&id_tbl->lock);
-	id_tbl->table = kzalloc(DIV_ROUND_UP(size, 32) * 4, GFP_KERNEL);
+	id_tbl->table = bitmap_zalloc(size, GFP_KERNEL);
 	if (!id_tbl->table)
 		return -ENOMEM;
 
@@ -669,7 +670,7 @@ static int cnic_init_id_tbl(struct cnic_id_tbl *id_tbl, u32 size, u32 start_id,
 
 static void cnic_free_id_tbl(struct cnic_id_tbl *id_tbl)
 {
-	kfree(id_tbl->table);
+	bitmap_free(id_tbl->table);
 	id_tbl->table = NULL;
 }
 
@@ -1027,16 +1028,14 @@ static int __cnic_alloc_uio_rings(struct cnic_uio_dev *udev, int pages)
 
 	udev->l2_ring_size = pages * CNIC_PAGE_SIZE;
 	udev->l2_ring = dma_alloc_coherent(&udev->pdev->dev, udev->l2_ring_size,
-					   &udev->l2_ring_map,
-					   GFP_KERNEL | __GFP_COMP);
+					   &udev->l2_ring_map, GFP_KERNEL);
 	if (!udev->l2_ring)
 		return -ENOMEM;
 
 	udev->l2_buf_size = (cp->l2_rx_ring_size + 1) * cp->l2_single_buf_size;
 	udev->l2_buf_size = CNIC_PAGE_ALIGN(udev->l2_buf_size);
 	udev->l2_buf = dma_alloc_coherent(&udev->pdev->dev, udev->l2_buf_size,
-					  &udev->l2_buf_map,
-					  GFP_KERNEL | __GFP_COMP);
+					  &udev->l2_buf_map, GFP_KERNEL);
 	if (!udev->l2_buf) {
 		__cnic_free_uio_rings(udev);
 		return -ENOMEM;
@@ -1109,10 +1108,11 @@ static int cnic_init_uio(struct cnic_dev *dev)
 						     TX_MAX_TSS_RINGS + 1);
 		uinfo->mem[1].addr = (unsigned long) cp->status_blk.gen &
 					CNIC_PAGE_MASK;
+		uinfo->mem[1].dma_addr = cp->status_blk_map;
 		if (cp->ethdev->drv_state & CNIC_DRV_STATE_USING_MSIX)
-			uinfo->mem[1].size = BNX2_SBLK_MSIX_ALIGN_SIZE * 9;
+			uinfo->mem[1].size = PAGE_ALIGN(BNX2_SBLK_MSIX_ALIGN_SIZE * 9);
 		else
-			uinfo->mem[1].size = BNX2_SBLK_MSIX_ALIGN_SIZE;
+			uinfo->mem[1].size = PAGE_ALIGN(BNX2_SBLK_MSIX_ALIGN_SIZE);
 
 		uinfo->name = "bnx2_cnic";
 	} else if (test_bit(CNIC_F_BNX2X_CLASS, &dev->flags)) {
@@ -1120,20 +1120,26 @@ static int cnic_init_uio(struct cnic_dev *dev)
 
 		uinfo->mem[1].addr = (unsigned long) cp->bnx2x_def_status_blk &
 			CNIC_PAGE_MASK;
-		uinfo->mem[1].size = sizeof(*cp->bnx2x_def_status_blk);
+		uinfo->mem[1].dma_addr = cp->status_blk_map;
+		uinfo->mem[1].size = PAGE_ALIGN(sizeof(*cp->bnx2x_def_status_blk));
 
 		uinfo->name = "bnx2x_cnic";
 	}
 
-	uinfo->mem[1].memtype = UIO_MEM_LOGICAL;
+	uinfo->mem[1].dma_device = &dev->pcidev->dev;
+	uinfo->mem[1].memtype = UIO_MEM_DMA_COHERENT;
 
 	uinfo->mem[2].addr = (unsigned long) udev->l2_ring;
-	uinfo->mem[2].size = udev->l2_ring_size;
-	uinfo->mem[2].memtype = UIO_MEM_LOGICAL;
+	uinfo->mem[2].dma_addr = udev->l2_ring_map;
+	uinfo->mem[2].size = PAGE_ALIGN(udev->l2_ring_size);
+	uinfo->mem[2].dma_device = &dev->pcidev->dev;
+	uinfo->mem[2].memtype = UIO_MEM_DMA_COHERENT;
 
 	uinfo->mem[3].addr = (unsigned long) udev->l2_buf;
-	uinfo->mem[3].size = udev->l2_buf_size;
-	uinfo->mem[3].memtype = UIO_MEM_LOGICAL;
+	uinfo->mem[3].dma_addr = udev->l2_buf_map;
+	uinfo->mem[3].size = PAGE_ALIGN(udev->l2_buf_size);
+	uinfo->mem[3].dma_device = &dev->pcidev->dev;
+	uinfo->mem[3].memtype = UIO_MEM_DMA_COHERENT;
 
 	uinfo->version = CNIC_MODULE_VERSION;
 	uinfo->irq = UIO_IRQ_CUSTOM;
@@ -1255,13 +1261,13 @@ static int cnic_alloc_bnx2x_resc(struct cnic_dev *dev)
 			cp->fcoe_init_cid = 0x10;
 	}
 
-	cp->iscsi_tbl = kzalloc(sizeof(struct cnic_iscsi) * MAX_ISCSI_TBL_SZ,
+	cp->iscsi_tbl = kcalloc(MAX_ISCSI_TBL_SZ, sizeof(struct cnic_iscsi),
 				GFP_KERNEL);
 	if (!cp->iscsi_tbl)
 		goto error;
 
-	cp->ctx_tbl = kzalloc(sizeof(struct cnic_context) *
-				cp->max_cid_space, GFP_KERNEL);
+	cp->ctx_tbl = kcalloc(cp->max_cid_space, sizeof(struct cnic_context),
+			      GFP_KERNEL);
 	if (!cp->ctx_tbl)
 		goto error;
 
@@ -1278,7 +1284,7 @@ static int cnic_alloc_bnx2x_resc(struct cnic_dev *dev)
 
 	ret = cnic_alloc_dma(dev, kwq_16_dma, pages, 0);
 	if (ret)
-		return -ENOMEM;
+		goto error;
 
 	n = CNIC_PAGE_SIZE / CNIC_KWQ16_DATA_SIZE;
 	for (i = 0, j = 0; i < cp->max_cid_space; i++) {
@@ -1315,6 +1321,7 @@ static int cnic_alloc_bnx2x_resc(struct cnic_dev *dev)
 		return 0;
 
 	cp->bnx2x_def_status_blk = cp->ethdev->irq_arr[1].status_blk;
+	cp->status_blk_map = cp->ethdev->irq_arr[1].status_blk_map;
 
 	cp->l2_rx_ring_size = 15;
 
@@ -1918,7 +1925,6 @@ static int cnic_bnx2x_iscsi_ofld1(struct cnic_dev *dev, struct kwqe *wqes[],
 	ret = cnic_alloc_bnx2x_conn_resc(dev, l5_cid);
 	if (ret) {
 		atomic_dec(&cp->iscsi_conn);
-		ret = 0;
 		goto done;
 	}
 	ret = cnic_setup_bnx2x_ctx(dev, wqes, num);
@@ -2562,7 +2568,6 @@ static void cnic_bnx2x_delete_wait(struct cnic_dev *dev, u32 start_cid)
 
 static int cnic_bnx2x_fcoe_fw_destroy(struct cnic_dev *dev, struct kwqe *kwqe)
 {
-	struct fcoe_kwqe_destroy *req;
 	union l5cm_specific_data l5_data;
 	struct cnic_local *cp = dev->cnic_priv;
 	struct bnx2x *bp = netdev_priv(dev->netdev);
@@ -2571,7 +2576,6 @@ static int cnic_bnx2x_fcoe_fw_destroy(struct cnic_dev *dev, struct kwqe *kwqe)
 
 	cnic_bnx2x_delete_wait(dev, MAX_ISCSI_TBL_SZ);
 
-	req = (struct fcoe_kwqe_destroy *) kwqe;
 	cid = BNX2X_HW_CID(bp, cp->fcoe_init_cid);
 
 	memset(&l5_data, 0, sizeof(l5_data));
@@ -3012,10 +3016,10 @@ static int cnic_service_bnx2(void *data, void *status_blk)
 	return cnic_service_bnx2_queues(dev);
 }
 
-static void cnic_service_bnx2_msix(unsigned long data)
+static void cnic_service_bnx2_msix(struct work_struct *work)
 {
-	struct cnic_dev *dev = (struct cnic_dev *) data;
-	struct cnic_local *cp = dev->cnic_priv;
+	struct cnic_local *cp = from_work(cp, work, cnic_irq_bh_work);
+	struct cnic_dev *dev = cp->dev;
 
 	cp->last_status_idx = cnic_service_bnx2_queues(dev);
 
@@ -3033,7 +3037,7 @@ static void cnic_doirq(struct cnic_dev *dev)
 		prefetch(cp->status_blk.gen);
 		prefetch(&cp->kcq1.kcq[KCQ_PG(prod)][KCQ_IDX(prod)]);
 
-		tasklet_schedule(&cp->cnic_irq_task);
+		queue_work(system_bh_wq, &cp->cnic_irq_bh_work);
 	}
 }
 
@@ -3137,10 +3141,10 @@ static u32 cnic_service_bnx2x_kcq(struct cnic_dev *dev, struct kcq_info *info)
 	return last_status;
 }
 
-static void cnic_service_bnx2x_bh(unsigned long data)
+static void cnic_service_bnx2x_bh_work(struct work_struct *work)
 {
-	struct cnic_dev *dev = (struct cnic_dev *) data;
-	struct cnic_local *cp = dev->cnic_priv;
+	struct cnic_local *cp = from_work(cp, work, cnic_irq_bh_work);
+	struct cnic_dev *dev = cp->dev;
 	struct bnx2x *bp = netdev_priv(dev->netdev);
 	u32 status_idx, new_status_idx;
 
@@ -3314,7 +3318,7 @@ static int cnic_ctl(void *data, struct cnic_ctl_info *info)
 	}
 	case CNIC_CTL_FCOE_STATS_GET_CMD:
 		ulp_type = CNIC_ULP_FCOE;
-		/* fall through */
+		fallthrough;
 	case CNIC_CTL_ISCSI_STATS_GET_CMD:
 		cnic_hold(dev);
 		cnic_copy_ulp_stats(dev, ulp_type);
@@ -3679,7 +3683,8 @@ static int cnic_get_v4_route(struct sockaddr_in *dst_addr,
 #if defined(CONFIG_INET)
 	struct rtable *rt;
 
-	rt = ip_route_output(&init_net, dst_addr->sin_addr.s_addr, 0, 0, 0);
+	rt = ip_route_output(&init_net, dst_addr->sin_addr.s_addr, 0, 0, 0,
+			     RT_SCOPE_UNIVERSE);
 	if (!IS_ERR(rt)) {
 		*dst = &rt->dst;
 		return 0;
@@ -4040,15 +4045,14 @@ static void cnic_cm_process_kcqe(struct cnic_dev *dev, struct kcqe *kcqe)
 	case L5CM_RAMROD_CMD_ID_CLOSE: {
 		struct iscsi_kcqe *l5kcqe = (struct iscsi_kcqe *) kcqe;
 
-		if (l4kcqe->status != 0 || l5kcqe->completion_status != 0) {
-			netdev_warn(dev->netdev, "RAMROD CLOSE compl with status 0x%x completion status 0x%x\n",
-				    l4kcqe->status, l5kcqe->completion_status);
-			opcode = L4_KCQE_OPCODE_VALUE_CLOSE_COMP;
-			/* Fall through */
-		} else {
+		if (l4kcqe->status == 0 && l5kcqe->completion_status == 0)
 			break;
-		}
+
+		netdev_warn(dev->netdev, "RAMROD CLOSE compl with status 0x%x completion status 0x%x\n",
+			    l4kcqe->status, l5kcqe->completion_status);
+		opcode = L4_KCQE_OPCODE_VALUE_CLOSE_COMP;
 	}
+		fallthrough;
 	case L4_KCQE_OPCODE_VALUE_RESET_RECEIVED:
 	case L4_KCQE_OPCODE_VALUE_CLOSE_COMP:
 	case L4_KCQE_OPCODE_VALUE_RESET_COMP:
@@ -4090,7 +4094,7 @@ static void cnic_cm_free_mem(struct cnic_dev *dev)
 {
 	struct cnic_local *cp = dev->cnic_priv;
 
-	kfree(cp->csk_tbl);
+	kvfree(cp->csk_tbl);
 	cp->csk_tbl = NULL;
 	cnic_free_id_tbl(&cp->csk_port_tbl);
 }
@@ -4099,14 +4103,17 @@ static int cnic_cm_alloc_mem(struct cnic_dev *dev)
 {
 	struct cnic_local *cp = dev->cnic_priv;
 	u32 port_id;
+	int i;
 
-	cp->csk_tbl = kzalloc(sizeof(struct cnic_sock) * MAX_CM_SK_TBL_SZ,
-			      GFP_KERNEL);
+	cp->csk_tbl = kvcalloc(MAX_CM_SK_TBL_SZ, sizeof(struct cnic_sock),
+			       GFP_KERNEL);
 	if (!cp->csk_tbl)
 		return -ENOMEM;
 
-	port_id = prandom_u32();
-	port_id %= CNIC_LOCAL_PORT_RANGE;
+	for (i = 0; i < MAX_CM_SK_TBL_SZ; i++)
+		atomic_set(&cp->csk_tbl[i].ref_count, 0);
+
+	port_id = get_random_u32_below(CNIC_LOCAL_PORT_RANGE);
 	if (cnic_init_id_tbl(&cp->csk_port_tbl, CNIC_LOCAL_PORT_RANGE,
 			     CNIC_LOCAL_PORT_MIN, port_id)) {
 		cnic_cm_free_mem(dev);
@@ -4165,7 +4172,7 @@ static int cnic_cm_init_bnx2_hw(struct cnic_dev *dev)
 {
 	u32 seed;
 
-	seed = prandom_u32();
+	seed = get_random_u32();
 	cnic_ctx_wr(dev, 45, 0, seed);
 	return 0;
 }
@@ -4223,8 +4230,7 @@ static void cnic_cm_stop_bnx2x_hw(struct cnic_dev *dev)
 
 	cnic_bnx2x_delete_wait(dev, 0);
 
-	cancel_delayed_work(&cp->delete_task);
-	flush_workqueue(cnic_wq);
+	cancel_delayed_work_sync(&cp->delete_task);
 
 	if (atomic_read(&cp->iscsi_conn) != 0)
 		netdev_warn(dev->netdev, "%d iSCSI connections not destroyed\n",
@@ -4422,7 +4428,7 @@ static void cnic_free_irq(struct cnic_dev *dev)
 
 	if (ethdev->drv_state & CNIC_DRV_STATE_USING_MSIX) {
 		cp->disable_int_sync(dev);
-		tasklet_kill(&cp->cnic_irq_task);
+		cancel_work_sync(&cp->cnic_irq_bh_work);
 		free_irq(ethdev->irq_arr[0].vector, dev);
 	}
 }
@@ -4435,7 +4441,7 @@ static int cnic_request_irq(struct cnic_dev *dev)
 
 	err = request_irq(ethdev->irq_arr[0].vector, cnic_irq, 0, "cnic", dev);
 	if (err)
-		tasklet_disable(&cp->cnic_irq_task);
+		disable_work_sync(&cp->cnic_irq_bh_work);
 
 	return err;
 }
@@ -4458,8 +4464,7 @@ static int cnic_init_bnx2_irq(struct cnic_dev *dev)
 		CNIC_WR(dev, base + BNX2_HC_CMD_TICKS_OFF, (64 << 16) | 220);
 
 		cp->last_status_idx = cp->status_blk.bnx2->status_idx;
-		tasklet_init(&cp->cnic_irq_task, cnic_service_bnx2_msix,
-			     (unsigned long) dev);
+		INIT_WORK(&cp->cnic_irq_bh_work, cnic_service_bnx2_msix);
 		err = cnic_request_irq(dev);
 		if (err)
 			return err;
@@ -4868,8 +4873,7 @@ static int cnic_init_bnx2x_irq(struct cnic_dev *dev)
 	struct cnic_eth_dev *ethdev = cp->ethdev;
 	int err = 0;
 
-	tasklet_init(&cp->cnic_irq_task, cnic_service_bnx2x_bh,
-		     (unsigned long) dev);
+	INIT_WORK(&cp->cnic_irq_bh_work, cnic_service_bnx2x_bh_work);
 	if (ethdev->drv_state & CNIC_DRV_STATE_USING_MSIX)
 		err = cnic_request_irq(dev);
 
@@ -5091,13 +5095,12 @@ static int cnic_start_bnx2x_hw(struct cnic_dev *dev)
 	struct cnic_local *cp = dev->cnic_priv;
 	struct bnx2x *bp = netdev_priv(dev->netdev);
 	struct cnic_eth_dev *ethdev = cp->ethdev;
-	int func, ret;
+	int ret;
 	u32 pfid;
 
 	dev->stats_addr = ethdev->addr_drv_info_to_mcp;
 	cp->func = bp->pf_num;
 
-	func = CNIC_FUNC(cp);
 	pfid = bp->pfid;
 
 	ret = cnic_init_id_tbl(&cp->cid_tbl, MAX_ISCSI_TBL_SZ,
@@ -5208,6 +5211,8 @@ static void cnic_init_rings(struct cnic_dev *dev)
 		cnic_init_bnx2x_tx_ring(dev, data);
 		cnic_init_bnx2x_rx_ring(dev, data);
 
+		data->general.fp_hsi_ver =  ETH_FP_HSI_VERSION;
+
 		l5_data.phy_address.lo = udev->l2_buf_map & 0xffffffff;
 		l5_data.phy_address.hi = (u64) udev->l2_buf_map >> 32;
 
@@ -5327,6 +5332,7 @@ static int cnic_start_hw(struct cnic_dev *dev)
 	pci_dev_get(dev->pcidev);
 	cp->func = PCI_FUNC(dev->pcidev->devfn);
 	cp->status_blk.gen = ethdev->irq_arr[0].status_blk;
+	cp->status_blk_map = ethdev->irq_arr[0].status_blk_map;
 	cp->status_blk_num = ethdev->irq_arr[0].status_blk_num;
 
 	err = cp->alloc_resc(dev);
@@ -5484,6 +5490,7 @@ static struct cnic_dev *cnic_alloc_dev(struct net_device *dev,
 	cdev->unregister_device = cnic_unregister_device;
 	cdev->iscsi_nl_msg_recv = cnic_iscsi_nl_msg_recv;
 	cdev->get_fc_npiv_tbl = cnic_get_fc_npiv_tbl;
+	atomic_set(&cdev->ref_count, 0);
 
 	cp = cdev->cnic_priv;
 	cp->dev = cdev;
@@ -5734,7 +5741,7 @@ static int cnic_netdev_event(struct notifier_block *this, unsigned long event,
 		if (realdev) {
 			dev = cnic_from_netdev(realdev);
 			if (dev) {
-				vid |= VLAN_TAG_PRESENT;
+				vid |= VLAN_CFI_MASK;	/* make non-zero */
 				cnic_rcv_netevent(dev->cnic_priv, event, vid);
 				cnic_put(dev);
 			}
diff --git a/drivers/net/ethernet/broadcom/cnic.h b/drivers/net/ethernet/broadcom/cnic.h
index 4baea81bae7a..1a314a75d2d2 100644
--- a/drivers/net/ethernet/broadcom/cnic.h
+++ b/drivers/net/ethernet/broadcom/cnic.h
@@ -260,6 +260,7 @@ struct cnic_local {
 		#define SM_RX_ID		0
 		#define SM_TX_ID		1
 	} status_blk;
+	dma_addr_t status_blk_map;
 
 	struct host_sp_status_block	*bnx2x_def_status_blk;
 
@@ -267,7 +268,7 @@ struct cnic_local {
 	u32				bnx2x_igu_sb_id;
 	u32				int_num;
 	u32				last_status_idx;
-	struct tasklet_struct		cnic_irq_task;
+	struct work_struct		cnic_irq_bh_work;
 
 	struct kcqe		*completed_kcq[MAX_COMPLETED_KCQE];
 
diff --git a/drivers/net/ethernet/broadcom/cnic_defs.h b/drivers/net/ethernet/broadcom/cnic_defs.h
index b38499774071..99e2c6d4d8c3 100644
--- a/drivers/net/ethernet/broadcom/cnic_defs.h
+++ b/drivers/net/ethernet/broadcom/cnic_defs.h
@@ -543,13 +543,13 @@ struct l4_kwq_update_pg {
 #define L4_KWQ_UPDATE_PG_RESERVERD2_SHIFT 2
 #endif
 #if defined(__BIG_ENDIAN)
-	u16 reserverd3;
+	u16 reserved3;
 	u8 da0;
 	u8 da1;
 #elif defined(__LITTLE_ENDIAN)
 	u8 da1;
 	u8 da0;
-	u16 reserverd3;
+	u16 reserved3;
 #endif
 #if defined(__BIG_ENDIAN)
 	u8 da2;
diff --git a/drivers/net/ethernet/broadcom/cnic_if.h b/drivers/net/ethernet/broadcom/cnic_if.h
index 789e5c7e9311..49a11ec80b36 100644
--- a/drivers/net/ethernet/broadcom/cnic_if.h
+++ b/drivers/net/ethernet/broadcom/cnic_if.h
@@ -190,6 +190,7 @@ struct cnic_ops {
 struct cnic_irq {
 	unsigned int	vector;
 	void		*status_blk;
+	dma_addr_t	status_blk_map;
 	u32		status_blk_num;
 	u32		status_blk_num2;
 	u32		irq_flags;
diff --git a/drivers/net/ethernet/broadcom/genet/Makefile b/drivers/net/ethernet/broadcom/genet/Makefile
index 9b6885efa9e7..edfc26a46948 100644
--- a/drivers/net/ethernet/broadcom/genet/Makefile
+++ b/drivers/net/ethernet/broadcom/genet/Makefile
@@ -1,2 +1,3 @@
+# SPDX-License-Identifier: GPL-2.0-only
 obj-$(CONFIG_BCMGENET) += genet.o
 genet-objs := bcmgenet.o bcmmii.o bcmgenet_wol.o
diff --git a/drivers/net/ethernet/broadcom/genet/bcmgenet.c b/drivers/net/ethernet/broadcom/genet/bcmgenet.c
index a205a9ff9e17..98971ae4f87d 100644
--- a/drivers/net/ethernet/broadcom/genet/bcmgenet.c
+++ b/drivers/net/ethernet/broadcom/genet/bcmgenet.c
@@ -1,15 +1,13 @@
+// SPDX-License-Identifier: GPL-2.0-only
 /*
  * Broadcom GENET (Gigabit Ethernet) controller driver
  *
- * Copyright (c) 2014-2017 Broadcom
- *
- * This program is free software; you can redistribute it and/or modify
- * it under the terms of the GNU General Public License version 2 as
- * published by the Free Software Foundation.
+ * Copyright (c) 2014-2025 Broadcom
  */
 
 #define pr_fmt(fmt)				"bcmgenet: " fmt
 
+#include <linux/acpi.h>
 #include <linux/kernel.h>
 #include <linux/module.h>
 #include <linux/sched.h>
@@ -25,11 +23,6 @@
 #include <linux/dma-mapping.h>
 #include <linux/pm.h>
 #include <linux/clk.h>
-#include <linux/of.h>
-#include <linux/of_address.h>
-#include <linux/of_irq.h>
-#include <linux/of_net.h>
-#include <linux/of_platform.h>
 #include <net/arp.h>
 
 #include <linux/mii.h>
@@ -44,19 +37,17 @@
 #include <linux/phy.h>
 #include <linux/platform_data/bcmgenet.h>
 
-#include <asm/unaligned.h>
+#include <linux/unaligned.h>
 
 #include "bcmgenet.h"
 
-/* Maximum number of hardware queues, downsized if needed */
-#define GENET_MAX_MQ_CNT	4
-
 /* Default highest priority queue for multi queue support */
-#define GENET_Q0_PRIORITY	0
+#define GENET_Q1_PRIORITY	0
+#define GENET_Q0_PRIORITY	1
 
-#define GENET_Q16_RX_BD_CNT	\
+#define GENET_Q0_RX_BD_CNT	\
 	(TOTAL_DESC - priv->hw_params->rx_queues * priv->hw_params->rx_bds_per_q)
-#define GENET_Q16_TX_BD_CNT	\
+#define GENET_Q0_TX_BD_CNT	\
 	(TOTAL_DESC - priv->hw_params->tx_queues * priv->hw_params->tx_bds_per_q)
 
 #define RX_BUF_LENGTH		2048
@@ -72,31 +63,47 @@
 #define GENET_RDMA_REG_OFF	(priv->hw_params->rdma_offset + \
 				TOTAL_DESC * DMA_DESC_SIZE)
 
-static inline void dmadesc_set_length_status(struct bcmgenet_priv *priv,
-					     void __iomem *d, u32 value)
+/* Forward declarations */
+static void bcmgenet_set_rx_mode(struct net_device *dev);
+
+static inline void bcmgenet_writel(u32 value, void __iomem *offset)
 {
-	__raw_writel(value, d + DMA_DESC_LENGTH_STATUS);
+	/* MIPS chips strapped for BE will automagically configure the
+	 * peripheral registers for CPU-native byte order.
+	 */
+	if (IS_ENABLED(CONFIG_MIPS) && IS_ENABLED(CONFIG_CPU_BIG_ENDIAN))
+		__raw_writel(value, offset);
+	else
+		writel_relaxed(value, offset);
 }
 
-static inline u32 dmadesc_get_length_status(struct bcmgenet_priv *priv,
-					    void __iomem *d)
+static inline u32 bcmgenet_readl(void __iomem *offset)
 {
-	return __raw_readl(d + DMA_DESC_LENGTH_STATUS);
+	if (IS_ENABLED(CONFIG_MIPS) && IS_ENABLED(CONFIG_CPU_BIG_ENDIAN))
+		return __raw_readl(offset);
+	else
+		return readl_relaxed(offset);
+}
+
+static inline void dmadesc_set_length_status(struct bcmgenet_priv *priv,
+					     void __iomem *d, u32 value)
+{
+	bcmgenet_writel(value, d + DMA_DESC_LENGTH_STATUS);
 }
 
 static inline void dmadesc_set_addr(struct bcmgenet_priv *priv,
 				    void __iomem *d,
 				    dma_addr_t addr)
 {
-	__raw_writel(lower_32_bits(addr), d + DMA_DESC_ADDRESS_LO);
+	bcmgenet_writel(lower_32_bits(addr), d + DMA_DESC_ADDRESS_LO);
 
 	/* Register writes to GISB bus can take couple hundred nanoseconds
 	 * and are done for each packet, save these expensive writes unless
 	 * the platform is explicitly configured for 64-bits/LPAE.
 	 */
 #ifdef CONFIG_PHYS_ADDR_T_64BIT
-	if (priv->hw_params->flags & GENET_HAS_40BITS)
-		__raw_writel(upper_32_bits(addr), d + DMA_DESC_ADDRESS_HI);
+	if (bcmgenet_has_40bits(priv))
+		bcmgenet_writel(upper_32_bits(addr), d + DMA_DESC_ADDRESS_HI);
 #endif
 }
 
@@ -108,24 +115,6 @@ static inline void dmadesc_set(struct bcmgenet_priv *priv,
 	dmadesc_set_length_status(priv, d, val);
 }
 
-static inline dma_addr_t dmadesc_get_addr(struct bcmgenet_priv *priv,
-					  void __iomem *d)
-{
-	dma_addr_t addr;
-
-	addr = __raw_readl(d + DMA_DESC_ADDRESS_LO);
-
-	/* Register writes to GISB bus can take couple hundred nanoseconds
-	 * and are done for each packet, save these expensive writes unless
-	 * the platform is explicitly configured for 64-bits/LPAE.
-	 */
-#ifdef CONFIG_PHYS_ADDR_T_64BIT
-	if (priv->hw_params->flags & GENET_HAS_40BITS)
-		addr |= (u64)__raw_readl(d + DMA_DESC_ADDRESS_HI) << 32;
-#endif
-	return addr;
-}
-
 #define GENET_VER_FMT	"%1d.%1d EPHY: 0x%04x"
 
 #define GENET_MSG_DEFAULT	(NETIF_MSG_DRV | NETIF_MSG_PROBE | \
@@ -156,8 +145,8 @@ static inline u32 bcmgenet_tbuf_ctrl_get(struct bcmgenet_priv *priv)
 	if (GENET_IS_V1(priv))
 		return bcmgenet_rbuf_readl(priv, TBUF_CTRL_V1);
 	else
-		return __raw_readl(priv->base +
-				priv->hw_params->tbuf_offset + TBUF_CTRL);
+		return bcmgenet_readl(priv->base +
+				      priv->hw_params->tbuf_offset + TBUF_CTRL);
 }
 
 static inline void bcmgenet_tbuf_ctrl_set(struct bcmgenet_priv *priv, u32 val)
@@ -165,7 +154,7 @@ static inline void bcmgenet_tbuf_ctrl_set(struct bcmgenet_priv *priv, u32 val)
 	if (GENET_IS_V1(priv))
 		bcmgenet_rbuf_writel(priv, val, TBUF_CTRL_V1);
 	else
-		__raw_writel(val, priv->base +
+		bcmgenet_writel(val, priv->base +
 				priv->hw_params->tbuf_offset + TBUF_CTRL);
 }
 
@@ -174,8 +163,8 @@ static inline u32 bcmgenet_bp_mc_get(struct bcmgenet_priv *priv)
 	if (GENET_IS_V1(priv))
 		return bcmgenet_rbuf_readl(priv, TBUF_BP_MC_V1);
 	else
-		return __raw_readl(priv->base +
-				priv->hw_params->tbuf_offset + TBUF_BP_MC);
+		return bcmgenet_readl(priv->base +
+				      priv->hw_params->tbuf_offset + TBUF_BP_MC);
 }
 
 static inline void bcmgenet_bp_mc_set(struct bcmgenet_priv *priv, u32 val)
@@ -183,7 +172,7 @@ static inline void bcmgenet_bp_mc_set(struct bcmgenet_priv *priv, u32 val)
 	if (GENET_IS_V1(priv))
 		bcmgenet_rbuf_writel(priv, val, TBUF_BP_MC_V1);
 	else
-		__raw_writel(val, priv->base +
+		bcmgenet_writel(val, priv->base +
 				priv->hw_params->tbuf_offset + TBUF_BP_MC);
 }
 
@@ -326,28 +315,28 @@ static inline struct bcmgenet_priv *dev_to_priv(struct device *dev)
 static inline u32 bcmgenet_tdma_readl(struct bcmgenet_priv *priv,
 				      enum dma_reg r)
 {
-	return __raw_readl(priv->base + GENET_TDMA_REG_OFF +
-			DMA_RINGS_SIZE + bcmgenet_dma_regs[r]);
+	return bcmgenet_readl(priv->base + GENET_TDMA_REG_OFF +
+			      DMA_RINGS_SIZE + bcmgenet_dma_regs[r]);
 }
 
 static inline void bcmgenet_tdma_writel(struct bcmgenet_priv *priv,
 					u32 val, enum dma_reg r)
 {
-	__raw_writel(val, priv->base + GENET_TDMA_REG_OFF +
+	bcmgenet_writel(val, priv->base + GENET_TDMA_REG_OFF +
 			DMA_RINGS_SIZE + bcmgenet_dma_regs[r]);
 }
 
 static inline u32 bcmgenet_rdma_readl(struct bcmgenet_priv *priv,
 				      enum dma_reg r)
 {
-	return __raw_readl(priv->base + GENET_RDMA_REG_OFF +
-			DMA_RINGS_SIZE + bcmgenet_dma_regs[r]);
+	return bcmgenet_readl(priv->base + GENET_RDMA_REG_OFF +
+			      DMA_RINGS_SIZE + bcmgenet_dma_regs[r]);
 }
 
 static inline void bcmgenet_rdma_writel(struct bcmgenet_priv *priv,
 					u32 val, enum dma_reg r)
 {
-	__raw_writel(val, priv->base + GENET_RDMA_REG_OFF +
+	bcmgenet_writel(val, priv->base + GENET_RDMA_REG_OFF +
 			DMA_RINGS_SIZE + bcmgenet_dma_regs[r]);
 }
 
@@ -418,16 +407,16 @@ static inline u32 bcmgenet_tdma_ring_readl(struct bcmgenet_priv *priv,
 					   unsigned int ring,
 					   enum dma_ring_reg r)
 {
-	return __raw_readl(priv->base + GENET_TDMA_REG_OFF +
-			(DMA_RING_SIZE * ring) +
-			genet_dma_ring_regs[r]);
+	return bcmgenet_readl(priv->base + GENET_TDMA_REG_OFF +
+			      (DMA_RING_SIZE * ring) +
+			      genet_dma_ring_regs[r]);
 }
 
 static inline void bcmgenet_tdma_ring_writel(struct bcmgenet_priv *priv,
 					     unsigned int ring, u32 val,
 					     enum dma_ring_reg r)
 {
-	__raw_writel(val, priv->base + GENET_TDMA_REG_OFF +
+	bcmgenet_writel(val, priv->base + GENET_TDMA_REG_OFF +
 			(DMA_RING_SIZE * ring) +
 			genet_dma_ring_regs[r]);
 }
@@ -436,134 +425,388 @@ static inline u32 bcmgenet_rdma_ring_readl(struct bcmgenet_priv *priv,
 					   unsigned int ring,
 					   enum dma_ring_reg r)
 {
-	return __raw_readl(priv->base + GENET_RDMA_REG_OFF +
-			(DMA_RING_SIZE * ring) +
-			genet_dma_ring_regs[r]);
+	return bcmgenet_readl(priv->base + GENET_RDMA_REG_OFF +
+			      (DMA_RING_SIZE * ring) +
+			      genet_dma_ring_regs[r]);
 }
 
 static inline void bcmgenet_rdma_ring_writel(struct bcmgenet_priv *priv,
 					     unsigned int ring, u32 val,
 					     enum dma_ring_reg r)
 {
-	__raw_writel(val, priv->base + GENET_RDMA_REG_OFF +
+	bcmgenet_writel(val, priv->base + GENET_RDMA_REG_OFF +
 			(DMA_RING_SIZE * ring) +
 			genet_dma_ring_regs[r]);
 }
 
-static int bcmgenet_begin(struct net_device *dev)
+static void bcmgenet_hfb_enable_filter(struct bcmgenet_priv *priv, u32 f_index)
 {
-	struct bcmgenet_priv *priv = netdev_priv(dev);
+	u32 offset;
+	u32 reg;
 
-	/* Turn on the clock */
-	return clk_prepare_enable(priv->clk);
+	if (GENET_IS_V1(priv) || GENET_IS_V2(priv)) {
+		reg = bcmgenet_hfb_reg_readl(priv, HFB_CTRL);
+		reg |= (1 << ((f_index % 32) + RBUF_HFB_FILTER_EN_SHIFT)) |
+			RBUF_HFB_EN;
+		bcmgenet_hfb_reg_writel(priv, reg, HFB_CTRL);
+	} else {
+		offset = HFB_FLT_ENABLE_V3PLUS + (f_index < 32) * sizeof(u32);
+		reg = bcmgenet_hfb_reg_readl(priv, offset);
+		reg |= (1 << (f_index % 32));
+		bcmgenet_hfb_reg_writel(priv, reg, offset);
+		reg = bcmgenet_hfb_reg_readl(priv, HFB_CTRL);
+		reg |= RBUF_HFB_EN;
+		bcmgenet_hfb_reg_writel(priv, reg, HFB_CTRL);
+	}
 }
 
-static void bcmgenet_complete(struct net_device *dev)
+static void bcmgenet_hfb_disable_filter(struct bcmgenet_priv *priv, u32 f_index)
 {
-	struct bcmgenet_priv *priv = netdev_priv(dev);
-
-	/* Turn off the clock */
-	clk_disable_unprepare(priv->clk);
+	u32 offset, reg, reg1;
+
+	if (GENET_IS_V1(priv) || GENET_IS_V2(priv)) {
+		reg = bcmgenet_hfb_reg_readl(priv, HFB_CTRL);
+		reg &= ~(1 << ((f_index % 32) + RBUF_HFB_FILTER_EN_SHIFT));
+		if (!(reg & RBUF_HFB_FILTER_EN_MASK))
+			reg &= ~RBUF_HFB_EN;
+		bcmgenet_hfb_reg_writel(priv, reg, HFB_CTRL);
+	} else {
+		offset = HFB_FLT_ENABLE_V3PLUS;
+		reg = bcmgenet_hfb_reg_readl(priv, offset);
+		reg1 = bcmgenet_hfb_reg_readl(priv, offset + sizeof(u32));
+		if  (f_index < 32) {
+			reg1 &= ~(1 << (f_index % 32));
+			bcmgenet_hfb_reg_writel(priv, reg1, offset + sizeof(u32));
+		} else {
+			reg &= ~(1 << (f_index % 32));
+			bcmgenet_hfb_reg_writel(priv, reg, offset);
+		}
+		if (!reg && !reg1) {
+			reg = bcmgenet_hfb_reg_readl(priv, HFB_CTRL);
+			reg &= ~RBUF_HFB_EN;
+			bcmgenet_hfb_reg_writel(priv, reg, HFB_CTRL);
+		}
+	}
 }
 
-static int bcmgenet_get_link_ksettings(struct net_device *dev,
-				       struct ethtool_link_ksettings *cmd)
+static void bcmgenet_hfb_set_filter_rx_queue_mapping(struct bcmgenet_priv *priv,
+						     u32 f_index, u32 rx_queue)
 {
-	struct bcmgenet_priv *priv = netdev_priv(dev);
+	u32 offset;
+	u32 reg;
 
-	if (!netif_running(dev))
-		return -EINVAL;
+	if (GENET_IS_V1(priv) || GENET_IS_V2(priv))
+		return;
 
-	if (!priv->phydev)
-		return -ENODEV;
+	offset = f_index / 8;
+	reg = bcmgenet_rdma_readl(priv, DMA_INDEX2RING_0 + offset);
+	reg &= ~(0xF << (4 * (f_index % 8)));
+	reg |= ((rx_queue & 0xF) << (4 * (f_index % 8)));
+	bcmgenet_rdma_writel(priv, reg, DMA_INDEX2RING_0 + offset);
+}
 
-	return phy_ethtool_ksettings_get(priv->phydev, cmd);
+static void bcmgenet_hfb_set_filter_length(struct bcmgenet_priv *priv,
+					   u32 f_index, u32 f_length)
+{
+	u32 offset;
+	u32 reg;
+
+	if (GENET_IS_V1(priv) || GENET_IS_V2(priv))
+		offset = HFB_FLT_LEN_V2;
+	else
+		offset = HFB_FLT_LEN_V3PLUS;
+
+	offset += sizeof(u32) *
+		  ((priv->hw_params->hfb_filter_cnt - 1 - f_index) / 4);
+	reg = bcmgenet_hfb_reg_readl(priv, offset);
+	reg &= ~(0xFF << (8 * (f_index % 4)));
+	reg |= ((f_length & 0xFF) << (8 * (f_index % 4)));
+	bcmgenet_hfb_reg_writel(priv, reg, offset);
 }
 
-static int bcmgenet_set_link_ksettings(struct net_device *dev,
-				       const struct ethtool_link_ksettings *cmd)
+static int bcmgenet_hfb_validate_mask(void *mask, size_t size)
 {
-	struct bcmgenet_priv *priv = netdev_priv(dev);
+	while (size) {
+		switch (*(unsigned char *)mask++) {
+		case 0x00:
+		case 0x0f:
+		case 0xf0:
+		case 0xff:
+			size--;
+			continue;
+		default:
+			return -EINVAL;
+		}
+	}
 
-	if (!netif_running(dev))
-		return -EINVAL;
+	return 0;
+}
 
-	if (!priv->phydev)
-		return -ENODEV;
+#define VALIDATE_MASK(x) \
+	bcmgenet_hfb_validate_mask(&(x), sizeof(x))
+
+static int bcmgenet_hfb_insert_data(struct bcmgenet_priv *priv, u32 f_index,
+				    u32 offset, void *val, void *mask,
+				    size_t size)
+{
+	u32 index, tmp;
+
+	index = f_index * priv->hw_params->hfb_filter_size + offset / 2;
+	tmp = bcmgenet_hfb_readl(priv, index * sizeof(u32));
+
+	while (size--) {
+		if (offset++ & 1) {
+			tmp &= ~0x300FF;
+			tmp |= (*(unsigned char *)val++);
+			switch ((*(unsigned char *)mask++)) {
+			case 0xFF:
+				tmp |= 0x30000;
+				break;
+			case 0xF0:
+				tmp |= 0x20000;
+				break;
+			case 0x0F:
+				tmp |= 0x10000;
+				break;
+			}
+			bcmgenet_hfb_writel(priv, tmp, index++ * sizeof(u32));
+			if (size)
+				tmp = bcmgenet_hfb_readl(priv,
+							 index * sizeof(u32));
+		} else {
+			tmp &= ~0xCFF00;
+			tmp |= (*(unsigned char *)val++) << 8;
+			switch ((*(unsigned char *)mask++)) {
+			case 0xFF:
+				tmp |= 0xC0000;
+				break;
+			case 0xF0:
+				tmp |= 0x80000;
+				break;
+			case 0x0F:
+				tmp |= 0x40000;
+				break;
+			}
+			if (!size)
+				bcmgenet_hfb_writel(priv, tmp, index * sizeof(u32));
+		}
+	}
 
-	return phy_ethtool_ksettings_set(priv->phydev, cmd);
+	return 0;
 }
 
-static int bcmgenet_set_rx_csum(struct net_device *dev,
-				netdev_features_t wanted)
+static void bcmgenet_hfb_create_rxnfc_filter(struct bcmgenet_priv *priv,
+					     struct bcmgenet_rxnfc_rule *rule)
 {
-	struct bcmgenet_priv *priv = netdev_priv(dev);
-	u32 rbuf_chk_ctrl;
-	bool rx_csum_en;
+	struct ethtool_rx_flow_spec *fs = &rule->fs;
+	u32 offset = 0, f_length = 0, f, q;
+	u8 val_8, mask_8;
+	__be16 val_16;
+	u16 mask_16;
+	size_t size;
+
+	f = fs->location + 1;
+	if (fs->flow_type & FLOW_MAC_EXT) {
+		bcmgenet_hfb_insert_data(priv, f, 0,
+					 &fs->h_ext.h_dest, &fs->m_ext.h_dest,
+					 sizeof(fs->h_ext.h_dest));
+	}
 
-	rx_csum_en = !!(wanted & NETIF_F_RXCSUM);
+	if (fs->flow_type & FLOW_EXT) {
+		if (fs->m_ext.vlan_etype ||
+		    fs->m_ext.vlan_tci) {
+			bcmgenet_hfb_insert_data(priv, f, 12,
+						 &fs->h_ext.vlan_etype,
+						 &fs->m_ext.vlan_etype,
+						 sizeof(fs->h_ext.vlan_etype));
+			bcmgenet_hfb_insert_data(priv, f, 14,
+						 &fs->h_ext.vlan_tci,
+						 &fs->m_ext.vlan_tci,
+						 sizeof(fs->h_ext.vlan_tci));
+			offset += VLAN_HLEN;
+			f_length += DIV_ROUND_UP(VLAN_HLEN, 2);
+		}
+	}
 
-	rbuf_chk_ctrl = bcmgenet_rbuf_readl(priv, RBUF_CHK_CTRL);
+	switch (fs->flow_type & ~(FLOW_EXT | FLOW_MAC_EXT)) {
+	case ETHER_FLOW:
+		f_length += DIV_ROUND_UP(ETH_HLEN, 2);
+		bcmgenet_hfb_insert_data(priv, f, 0,
+					 &fs->h_u.ether_spec.h_dest,
+					 &fs->m_u.ether_spec.h_dest,
+					 sizeof(fs->h_u.ether_spec.h_dest));
+		bcmgenet_hfb_insert_data(priv, f, ETH_ALEN,
+					 &fs->h_u.ether_spec.h_source,
+					 &fs->m_u.ether_spec.h_source,
+					 sizeof(fs->h_u.ether_spec.h_source));
+		bcmgenet_hfb_insert_data(priv, f, (2 * ETH_ALEN) + offset,
+					 &fs->h_u.ether_spec.h_proto,
+					 &fs->m_u.ether_spec.h_proto,
+					 sizeof(fs->h_u.ether_spec.h_proto));
+		break;
+	case IP_USER_FLOW:
+		f_length += DIV_ROUND_UP(ETH_HLEN + 20, 2);
+		/* Specify IP Ether Type */
+		val_16 = htons(ETH_P_IP);
+		mask_16 = 0xFFFF;
+		bcmgenet_hfb_insert_data(priv, f, (2 * ETH_ALEN) + offset,
+					 &val_16, &mask_16, sizeof(val_16));
+		bcmgenet_hfb_insert_data(priv, f, 15 + offset,
+					 &fs->h_u.usr_ip4_spec.tos,
+					 &fs->m_u.usr_ip4_spec.tos,
+					 sizeof(fs->h_u.usr_ip4_spec.tos));
+		bcmgenet_hfb_insert_data(priv, f, 23 + offset,
+					 &fs->h_u.usr_ip4_spec.proto,
+					 &fs->m_u.usr_ip4_spec.proto,
+					 sizeof(fs->h_u.usr_ip4_spec.proto));
+		bcmgenet_hfb_insert_data(priv, f, 26 + offset,
+					 &fs->h_u.usr_ip4_spec.ip4src,
+					 &fs->m_u.usr_ip4_spec.ip4src,
+					 sizeof(fs->h_u.usr_ip4_spec.ip4src));
+		bcmgenet_hfb_insert_data(priv, f, 30 + offset,
+					 &fs->h_u.usr_ip4_spec.ip4dst,
+					 &fs->m_u.usr_ip4_spec.ip4dst,
+					 sizeof(fs->h_u.usr_ip4_spec.ip4dst));
+		if (!fs->m_u.usr_ip4_spec.l4_4_bytes)
+			break;
 
-	/* enable rx checksumming */
-	if (rx_csum_en)
-		rbuf_chk_ctrl |= RBUF_RXCHK_EN;
-	else
-		rbuf_chk_ctrl &= ~RBUF_RXCHK_EN;
-	priv->desc_rxchk_en = rx_csum_en;
+		/* Only supports 20 byte IPv4 header */
+		val_8 = 0x45;
+		mask_8 = 0xFF;
+		bcmgenet_hfb_insert_data(priv, f, ETH_HLEN + offset,
+					 &val_8, &mask_8,
+					 sizeof(val_8));
+		size = sizeof(fs->h_u.usr_ip4_spec.l4_4_bytes);
+		bcmgenet_hfb_insert_data(priv, f,
+					 ETH_HLEN + 20 + offset,
+					 &fs->h_u.usr_ip4_spec.l4_4_bytes,
+					 &fs->m_u.usr_ip4_spec.l4_4_bytes,
+					 size);
+		f_length += DIV_ROUND_UP(size, 2);
+		break;
+	}
 
-	/* If UniMAC forwards CRC, we need to skip over it to get
-	 * a valid CHK bit to be set in the per-packet status word
-	*/
-	if (rx_csum_en && priv->crc_fwd_en)
-		rbuf_chk_ctrl |= RBUF_SKIP_FCS;
+	bcmgenet_hfb_set_filter_length(priv, f, 2 * f_length);
+	if (fs->ring_cookie == RX_CLS_FLOW_WAKE)
+		q = 0;
+	else if (fs->ring_cookie == RX_CLS_FLOW_DISC)
+		q = priv->hw_params->rx_queues + 1;
 	else
-		rbuf_chk_ctrl &= ~RBUF_SKIP_FCS;
+		/* Other Rx rings are direct mapped here */
+		q = fs->ring_cookie;
+	bcmgenet_hfb_set_filter_rx_queue_mapping(priv, f, q);
+	bcmgenet_hfb_enable_filter(priv, f);
+	rule->state = BCMGENET_RXNFC_STATE_ENABLED;
+}
 
-	bcmgenet_rbuf_writel(priv, rbuf_chk_ctrl, RBUF_CHK_CTRL);
+/* bcmgenet_hfb_clear
+ *
+ * Clear Hardware Filter Block and disable all filtering.
+ */
+static void bcmgenet_hfb_clear_filter(struct bcmgenet_priv *priv, u32 f_index)
+{
+	u32 base, i;
 
-	return 0;
+	bcmgenet_hfb_set_filter_length(priv, f_index, 0);
+	base = f_index * priv->hw_params->hfb_filter_size;
+	for (i = 0; i < priv->hw_params->hfb_filter_size; i++)
+		bcmgenet_hfb_writel(priv, 0x0, (base + i) * sizeof(u32));
 }
 
-static int bcmgenet_set_tx_csum(struct net_device *dev,
-				netdev_features_t wanted)
+static void bcmgenet_hfb_clear(struct bcmgenet_priv *priv)
 {
-	struct bcmgenet_priv *priv = netdev_priv(dev);
-	bool desc_64b_en;
-	u32 tbuf_ctrl, rbuf_ctrl;
+	u32 i;
 
-	tbuf_ctrl = bcmgenet_tbuf_ctrl_get(priv);
-	rbuf_ctrl = bcmgenet_rbuf_readl(priv, RBUF_CTRL);
+	bcmgenet_hfb_reg_writel(priv, 0, HFB_CTRL);
 
-	desc_64b_en = !!(wanted & (NETIF_F_IP_CSUM | NETIF_F_IPV6_CSUM));
+	if (!GENET_IS_V1(priv) && !GENET_IS_V2(priv)) {
+		bcmgenet_hfb_reg_writel(priv, 0,
+					HFB_FLT_ENABLE_V3PLUS);
+		bcmgenet_hfb_reg_writel(priv, 0,
+					HFB_FLT_ENABLE_V3PLUS + 4);
+		for (i = DMA_INDEX2RING_0; i <= DMA_INDEX2RING_7; i++)
+			bcmgenet_rdma_writel(priv, 0, i);
+	}
 
-	/* enable 64 bytes descriptor in both directions (RBUF and TBUF) */
-	if (desc_64b_en) {
-		tbuf_ctrl |= RBUF_64B_EN;
-		rbuf_ctrl |= RBUF_64B_EN;
-	} else {
-		tbuf_ctrl &= ~RBUF_64B_EN;
-		rbuf_ctrl &= ~RBUF_64B_EN;
+	for (i = 0; i < priv->hw_params->hfb_filter_cnt; i++)
+		bcmgenet_hfb_clear_filter(priv, i);
+
+	/* Enable filter 0 to send default flow to ring 0 */
+	bcmgenet_hfb_set_filter_length(priv, 0, 4);
+	bcmgenet_hfb_enable_filter(priv, 0);
+}
+
+static void bcmgenet_hfb_init(struct bcmgenet_priv *priv)
+{
+	int i;
+
+	INIT_LIST_HEAD(&priv->rxnfc_list);
+	for (i = 0; i < MAX_NUM_OF_FS_RULES; i++) {
+		INIT_LIST_HEAD(&priv->rxnfc_rules[i].list);
+		priv->rxnfc_rules[i].state = BCMGENET_RXNFC_STATE_UNUSED;
 	}
-	priv->desc_64b_en = desc_64b_en;
 
-	bcmgenet_tbuf_ctrl_set(priv, tbuf_ctrl);
-	bcmgenet_rbuf_writel(priv, rbuf_ctrl, RBUF_CTRL);
+	bcmgenet_hfb_clear(priv);
+}
+
+static int bcmgenet_begin(struct net_device *dev)
+{
+	struct bcmgenet_priv *priv = netdev_priv(dev);
+
+	/* Turn on the clock */
+	return clk_prepare_enable(priv->clk);
+}
+
+static void bcmgenet_complete(struct net_device *dev)
+{
+	struct bcmgenet_priv *priv = netdev_priv(dev);
+
+	/* Turn off the clock */
+	clk_disable_unprepare(priv->clk);
+}
+
+static int bcmgenet_get_link_ksettings(struct net_device *dev,
+				       struct ethtool_link_ksettings *cmd)
+{
+	if (!netif_running(dev))
+		return -EINVAL;
+
+	if (!dev->phydev)
+		return -ENODEV;
+
+	phy_ethtool_ksettings_get(dev->phydev, cmd);
 
 	return 0;
 }
 
+static int bcmgenet_set_link_ksettings(struct net_device *dev,
+				       const struct ethtool_link_ksettings *cmd)
+{
+	if (!netif_running(dev))
+		return -EINVAL;
+
+	if (!dev->phydev)
+		return -ENODEV;
+
+	return phy_ethtool_ksettings_set(dev->phydev, cmd);
+}
+
 static int bcmgenet_set_features(struct net_device *dev,
 				 netdev_features_t features)
 {
-	netdev_features_t changed = features ^ dev->features;
-	netdev_features_t wanted = dev->wanted_features;
-	int ret = 0;
+	struct bcmgenet_priv *priv = netdev_priv(dev);
+	u32 reg;
+	int ret;
 
-	if (changed & (NETIF_F_IP_CSUM | NETIF_F_IPV6_CSUM))
-		ret = bcmgenet_set_tx_csum(dev, wanted);
-	if (changed & (NETIF_F_RXCSUM))
-		ret = bcmgenet_set_rx_csum(dev, wanted);
+	ret = clk_prepare_enable(priv->clk);
+	if (ret)
+		return ret;
+
+	/* Make sure we reflect the value of CRC_CMD_FWD */
+	reg = bcmgenet_umac_readl(priv, UMAC_CMD);
+	priv->crc_fwd_en = !!(reg & CMD_CRC_FWD);
+
+	clk_disable_unprepare(priv->clk);
 
 	return ret;
 }
@@ -583,28 +826,72 @@ static void bcmgenet_set_msglevel(struct net_device *dev, u32 level)
 }
 
 static int bcmgenet_get_coalesce(struct net_device *dev,
-				 struct ethtool_coalesce *ec)
+				 struct ethtool_coalesce *ec,
+				 struct kernel_ethtool_coalesce *kernel_coal,
+				 struct netlink_ext_ack *extack)
 {
 	struct bcmgenet_priv *priv = netdev_priv(dev);
+	struct bcmgenet_rx_ring *ring;
+	unsigned int i;
 
 	ec->tx_max_coalesced_frames =
-		bcmgenet_tdma_ring_readl(priv, DESC_INDEX,
-					 DMA_MBUF_DONE_THRESH);
+		bcmgenet_tdma_ring_readl(priv, 0, DMA_MBUF_DONE_THRESH);
 	ec->rx_max_coalesced_frames =
-		bcmgenet_rdma_ring_readl(priv, DESC_INDEX,
-					 DMA_MBUF_DONE_THRESH);
+		bcmgenet_rdma_ring_readl(priv, 0, DMA_MBUF_DONE_THRESH);
 	ec->rx_coalesce_usecs =
-		bcmgenet_rdma_readl(priv, DMA_RING16_TIMEOUT) * 8192 / 1000;
+		bcmgenet_rdma_readl(priv, DMA_RING0_TIMEOUT) * 8192 / 1000;
+
+	for (i = 0; i <= priv->hw_params->rx_queues; i++) {
+		ring = &priv->rx_rings[i];
+		ec->use_adaptive_rx_coalesce |= ring->dim.use_dim;
+	}
 
 	return 0;
 }
 
+static void bcmgenet_set_rx_coalesce(struct bcmgenet_rx_ring *ring,
+				     u32 usecs, u32 pkts)
+{
+	struct bcmgenet_priv *priv = ring->priv;
+	unsigned int i = ring->index;
+	u32 reg;
+
+	bcmgenet_rdma_ring_writel(priv, i, pkts, DMA_MBUF_DONE_THRESH);
+
+	reg = bcmgenet_rdma_readl(priv, DMA_RING0_TIMEOUT + i);
+	reg &= ~DMA_TIMEOUT_MASK;
+	reg |= DIV_ROUND_UP(usecs * 1000, 8192);
+	bcmgenet_rdma_writel(priv, reg, DMA_RING0_TIMEOUT + i);
+}
+
+static void bcmgenet_set_ring_rx_coalesce(struct bcmgenet_rx_ring *ring,
+					  struct ethtool_coalesce *ec)
+{
+	struct dim_cq_moder moder;
+	u32 usecs, pkts;
+
+	ring->rx_coalesce_usecs = ec->rx_coalesce_usecs;
+	ring->rx_max_coalesced_frames = ec->rx_max_coalesced_frames;
+	usecs = ring->rx_coalesce_usecs;
+	pkts = ring->rx_max_coalesced_frames;
+
+	if (ec->use_adaptive_rx_coalesce && !ring->dim.use_dim) {
+		moder = net_dim_get_def_rx_moderation(ring->dim.dim.mode);
+		usecs = moder.usec;
+		pkts = moder.pkts;
+	}
+
+	ring->dim.use_dim = ec->use_adaptive_rx_coalesce;
+	bcmgenet_set_rx_coalesce(ring, usecs, pkts);
+}
+
 static int bcmgenet_set_coalesce(struct net_device *dev,
-				 struct ethtool_coalesce *ec)
+				 struct ethtool_coalesce *ec,
+				 struct kernel_ethtool_coalesce *kernel_coal,
+				 struct netlink_ext_ack *extack)
 {
 	struct bcmgenet_priv *priv = netdev_priv(dev);
 	unsigned int i;
-	u32 reg;
 
 	/* Base system clock is 125Mhz, DMA timeout is this reference clock
 	 * divided by 1024, which yields roughly 8.192us, our maximum value
@@ -623,52 +910,72 @@ static int bcmgenet_set_coalesce(struct net_device *dev,
 	 * always generate an interrupt either after MBDONE packets have been
 	 * transmitted, or when the ring is empty.
 	 */
-	if (ec->tx_coalesce_usecs || ec->tx_coalesce_usecs_high ||
-	    ec->tx_coalesce_usecs_irq || ec->tx_coalesce_usecs_low)
-		return -EOPNOTSUPP;
 
 	/* Program all TX queues with the same values, as there is no
 	 * ethtool knob to do coalescing on a per-queue basis
 	 */
-	for (i = 0; i < priv->hw_params->tx_queues; i++)
+	for (i = 0; i <= priv->hw_params->tx_queues; i++)
 		bcmgenet_tdma_ring_writel(priv, i,
 					  ec->tx_max_coalesced_frames,
 					  DMA_MBUF_DONE_THRESH);
-	bcmgenet_tdma_ring_writel(priv, DESC_INDEX,
-				  ec->tx_max_coalesced_frames,
-				  DMA_MBUF_DONE_THRESH);
 
-	for (i = 0; i < priv->hw_params->rx_queues; i++) {
-		bcmgenet_rdma_ring_writel(priv, i,
-					  ec->rx_max_coalesced_frames,
-					  DMA_MBUF_DONE_THRESH);
+	for (i = 0; i <= priv->hw_params->rx_queues; i++)
+		bcmgenet_set_ring_rx_coalesce(&priv->rx_rings[i], ec);
+
+	return 0;
+}
+
+static void bcmgenet_get_pauseparam(struct net_device *dev,
+				    struct ethtool_pauseparam *epause)
+{
+	struct bcmgenet_priv *priv;
+	u32 umac_cmd;
+
+	priv = netdev_priv(dev);
+
+	epause->autoneg = priv->autoneg_pause;
 
-		reg = bcmgenet_rdma_readl(priv, DMA_RING0_TIMEOUT + i);
-		reg &= ~DMA_TIMEOUT_MASK;
-		reg |= DIV_ROUND_UP(ec->rx_coalesce_usecs * 1000, 8192);
-		bcmgenet_rdma_writel(priv, reg, DMA_RING0_TIMEOUT + i);
+	if (netif_carrier_ok(dev)) {
+		/* report active state when link is up */
+		umac_cmd = bcmgenet_umac_readl(priv, UMAC_CMD);
+		epause->tx_pause = !(umac_cmd & CMD_TX_PAUSE_IGNORE);
+		epause->rx_pause = !(umac_cmd & CMD_RX_PAUSE_IGNORE);
+	} else {
+		/* otherwise report stored settings */
+		epause->tx_pause = priv->tx_pause;
+		epause->rx_pause = priv->rx_pause;
 	}
+}
 
-	bcmgenet_rdma_ring_writel(priv, DESC_INDEX,
-				  ec->rx_max_coalesced_frames,
-				  DMA_MBUF_DONE_THRESH);
+static int bcmgenet_set_pauseparam(struct net_device *dev,
+				   struct ethtool_pauseparam *epause)
+{
+	struct bcmgenet_priv *priv = netdev_priv(dev);
 
-	reg = bcmgenet_rdma_readl(priv, DMA_RING16_TIMEOUT);
-	reg &= ~DMA_TIMEOUT_MASK;
-	reg |= DIV_ROUND_UP(ec->rx_coalesce_usecs * 1000, 8192);
-	bcmgenet_rdma_writel(priv, reg, DMA_RING16_TIMEOUT);
+	if (!dev->phydev)
+		return -ENODEV;
+
+	if (!phy_validate_pause(dev->phydev, epause))
+		return -EINVAL;
+
+	priv->autoneg_pause = !!epause->autoneg;
+	priv->tx_pause = !!epause->tx_pause;
+	priv->rx_pause = !!epause->rx_pause;
+
+	bcmgenet_phy_pause_set(dev, priv->rx_pause, priv->tx_pause);
 
 	return 0;
 }
 
 /* standard ethtool support functions. */
 enum bcmgenet_stat_type {
-	BCMGENET_STAT_NETDEV = -1,
+	BCMGENET_STAT_RTNL = -1,
 	BCMGENET_STAT_MIB_RX,
 	BCMGENET_STAT_MIB_TX,
 	BCMGENET_STAT_RUNT,
 	BCMGENET_STAT_MISC,
 	BCMGENET_STAT_SOFT,
+	BCMGENET_STAT_SOFT64,
 };
 
 struct bcmgenet_stats {
@@ -678,13 +985,15 @@ struct bcmgenet_stats {
 	enum bcmgenet_stat_type type;
 	/* reg offset from UMAC base for misc counters */
 	u16 reg_offset;
+	/* sync for u64 stats counters */
+	int syncp_offset;
 };
 
-#define STAT_NETDEV(m) { \
+#define STAT_RTNL(m) { \
 	.stat_string = __stringify(m), \
-	.stat_sizeof = sizeof(((struct net_device_stats *)0)->m), \
-	.stat_offset = offsetof(struct net_device_stats, m), \
-	.type = BCMGENET_STAT_NETDEV, \
+	.stat_sizeof = sizeof(((struct rtnl_link_stats64 *)0)->m), \
+	.stat_offset = offsetof(struct rtnl_link_stats64, m), \
+	.type = BCMGENET_STAT_RTNL, \
 }
 
 #define STAT_GENET_MIB(str, m, _type) { \
@@ -694,6 +1003,14 @@ struct bcmgenet_stats {
 	.type = _type, \
 }
 
+#define STAT_GENET_SOFT_MIB64(str, s, m) { \
+	.stat_string = str, \
+	.stat_sizeof = sizeof(((struct bcmgenet_priv *)0)->s.m), \
+	.stat_offset = offsetof(struct bcmgenet_priv, s.m), \
+	.type = BCMGENET_STAT_SOFT64, \
+	.syncp_offset = offsetof(struct bcmgenet_priv, s.syncp), \
+}
+
 #define STAT_GENET_MIB_RX(str, m) STAT_GENET_MIB(str, m, BCMGENET_STAT_MIB_RX)
 #define STAT_GENET_MIB_TX(str, m) STAT_GENET_MIB(str, m, BCMGENET_STAT_MIB_TX)
 #define STAT_GENET_RUNT(str, m) STAT_GENET_MIB(str, m, BCMGENET_STAT_RUNT)
@@ -708,18 +1025,38 @@ struct bcmgenet_stats {
 }
 
 #define STAT_GENET_Q(num) \
-	STAT_GENET_SOFT_MIB("txq" __stringify(num) "_packets", \
-			tx_rings[num].packets), \
-	STAT_GENET_SOFT_MIB("txq" __stringify(num) "_bytes", \
-			tx_rings[num].bytes), \
-	STAT_GENET_SOFT_MIB("rxq" __stringify(num) "_bytes", \
-			rx_rings[num].bytes),	 \
-	STAT_GENET_SOFT_MIB("rxq" __stringify(num) "_packets", \
-			rx_rings[num].packets), \
-	STAT_GENET_SOFT_MIB("rxq" __stringify(num) "_errors", \
-			rx_rings[num].errors), \
-	STAT_GENET_SOFT_MIB("rxq" __stringify(num) "_dropped", \
-			rx_rings[num].dropped)
+	STAT_GENET_SOFT_MIB64("txq" __stringify(num) "_packets", \
+			tx_rings[num].stats64, packets), \
+	STAT_GENET_SOFT_MIB64("txq" __stringify(num) "_bytes", \
+			tx_rings[num].stats64, bytes), \
+	STAT_GENET_SOFT_MIB64("txq" __stringify(num) "_errors", \
+			tx_rings[num].stats64, errors), \
+	STAT_GENET_SOFT_MIB64("txq" __stringify(num) "_dropped", \
+			tx_rings[num].stats64, dropped), \
+	STAT_GENET_SOFT_MIB64("rxq" __stringify(num) "_bytes", \
+			rx_rings[num].stats64, bytes),	 \
+	STAT_GENET_SOFT_MIB64("rxq" __stringify(num) "_packets", \
+			rx_rings[num].stats64, packets), \
+	STAT_GENET_SOFT_MIB64("rxq" __stringify(num) "_errors", \
+			rx_rings[num].stats64, errors), \
+	STAT_GENET_SOFT_MIB64("rxq" __stringify(num) "_dropped", \
+			rx_rings[num].stats64, dropped), \
+	STAT_GENET_SOFT_MIB64("rxq" __stringify(num) "_multicast", \
+			rx_rings[num].stats64, multicast), \
+	STAT_GENET_SOFT_MIB64("rxq" __stringify(num) "_missed", \
+			rx_rings[num].stats64, missed), \
+	STAT_GENET_SOFT_MIB64("rxq" __stringify(num) "_length_errors", \
+			rx_rings[num].stats64, length_errors), \
+	STAT_GENET_SOFT_MIB64("rxq" __stringify(num) "_over_errors", \
+			rx_rings[num].stats64, over_errors), \
+	STAT_GENET_SOFT_MIB64("rxq" __stringify(num) "_crc_errors", \
+			rx_rings[num].stats64, crc_errors), \
+	STAT_GENET_SOFT_MIB64("rxq" __stringify(num) "_frame_errors", \
+			rx_rings[num].stats64, frame_errors), \
+	STAT_GENET_SOFT_MIB64("rxq" __stringify(num) "_fragmented_errors", \
+			rx_rings[num].stats64, fragmented_errors), \
+	STAT_GENET_SOFT_MIB64("rxq" __stringify(num) "_broadcast", \
+			rx_rings[num].stats64, broadcast)
 
 /* There is a 0xC gap between the end of RX and beginning of TX stats and then
  * between the end of TX stats and the beginning of the RX RUNT
@@ -731,15 +1068,20 @@ struct bcmgenet_stats {
  */
 static const struct bcmgenet_stats bcmgenet_gstrings_stats[] = {
 	/* general stats */
-	STAT_NETDEV(rx_packets),
-	STAT_NETDEV(tx_packets),
-	STAT_NETDEV(rx_bytes),
-	STAT_NETDEV(tx_bytes),
-	STAT_NETDEV(rx_errors),
-	STAT_NETDEV(tx_errors),
-	STAT_NETDEV(rx_dropped),
-	STAT_NETDEV(tx_dropped),
-	STAT_NETDEV(multicast),
+	STAT_RTNL(rx_packets),
+	STAT_RTNL(tx_packets),
+	STAT_RTNL(rx_bytes),
+	STAT_RTNL(tx_bytes),
+	STAT_RTNL(rx_errors),
+	STAT_RTNL(tx_errors),
+	STAT_RTNL(rx_dropped),
+	STAT_RTNL(tx_dropped),
+	STAT_RTNL(multicast),
+	STAT_RTNL(rx_missed_errors),
+	STAT_RTNL(rx_length_errors),
+	STAT_RTNL(rx_over_errors),
+	STAT_RTNL(rx_crc_errors),
+	STAT_RTNL(rx_frame_errors),
 	/* UniMAC RSV counters */
 	STAT_GENET_MIB_RX("rx_64_octets", mib.rx.pkt_cnt.cnt_64),
 	STAT_GENET_MIB_RX("rx_65_127_oct", mib.rx.pkt_cnt.cnt_127),
@@ -814,21 +1156,37 @@ static const struct bcmgenet_stats bcmgenet_gstrings_stats[] = {
 	STAT_GENET_SOFT_MIB("alloc_rx_buff_failed", mib.alloc_rx_buff_failed),
 	STAT_GENET_SOFT_MIB("rx_dma_failed", mib.rx_dma_failed),
 	STAT_GENET_SOFT_MIB("tx_dma_failed", mib.tx_dma_failed),
+	STAT_GENET_SOFT_MIB("tx_realloc_tsb", mib.tx_realloc_tsb),
+	STAT_GENET_SOFT_MIB("tx_realloc_tsb_failed",
+			    mib.tx_realloc_tsb_failed),
 	/* Per TX queues */
 	STAT_GENET_Q(0),
 	STAT_GENET_Q(1),
 	STAT_GENET_Q(2),
 	STAT_GENET_Q(3),
-	STAT_GENET_Q(16),
+	STAT_GENET_Q(4),
 };
 
 #define BCMGENET_STATS_LEN	ARRAY_SIZE(bcmgenet_gstrings_stats)
 
+#define BCMGENET_STATS64_ADD(stats, m, v) \
+	do { \
+		u64_stats_update_begin(&stats->syncp); \
+		u64_stats_add(&stats->m, v); \
+		u64_stats_update_end(&stats->syncp); \
+	} while (0)
+
+#define BCMGENET_STATS64_INC(stats, m) \
+	do { \
+		u64_stats_update_begin(&stats->syncp); \
+		u64_stats_inc(&stats->m); \
+		u64_stats_update_end(&stats->syncp); \
+	} while (0)
+
 static void bcmgenet_get_drvinfo(struct net_device *dev,
 				 struct ethtool_drvinfo *info)
 {
-	strlcpy(info->driver, "bcmgenet", sizeof(info->driver));
-	strlcpy(info->version, "v2.0", sizeof(info->version));
+	strscpy(info->driver, "bcmgenet", sizeof(info->driver));
 }
 
 static int bcmgenet_get_sset_count(struct net_device *dev, int string_set)
@@ -844,14 +1202,14 @@ static int bcmgenet_get_sset_count(struct net_device *dev, int string_set)
 static void bcmgenet_get_strings(struct net_device *dev, u32 stringset,
 				 u8 *data)
 {
+	const char *str;
 	int i;
 
 	switch (stringset) {
 	case ETH_SS_STATS:
 		for (i = 0; i < BCMGENET_STATS_LEN; i++) {
-			memcpy(data + i * ETH_GSTRING_LEN,
-			       bcmgenet_gstrings_stats[i].stat_string,
-			       ETH_GSTRING_LEN);
+			str = bcmgenet_gstrings_stats[i].stat_string;
+			ethtool_puts(&data, str);
 		}
 		break;
 	}
@@ -908,15 +1266,16 @@ static void bcmgenet_update_mib_counters(struct bcmgenet_priv *priv)
 
 		s = &bcmgenet_gstrings_stats[i];
 		switch (s->type) {
-		case BCMGENET_STAT_NETDEV:
+		case BCMGENET_STAT_RTNL:
 		case BCMGENET_STAT_SOFT:
+		case BCMGENET_STAT_SOFT64:
 			continue;
 		case BCMGENET_STAT_RUNT:
 			offset += BCMGENET_STAT_OFFSET;
-			/* fall through */
+			fallthrough;
 		case BCMGENET_STAT_MIB_TX:
 			offset += BCMGENET_STAT_OFFSET;
-			/* fall through */
+			fallthrough;
 		case BCMGENET_STAT_MIB_RX:
 			val = bcmgenet_umac_readl(priv,
 						  UMAC_MIB_START + j + offset);
@@ -947,30 +1306,45 @@ static void bcmgenet_get_ethtool_stats(struct net_device *dev,
 				       u64 *data)
 {
 	struct bcmgenet_priv *priv = netdev_priv(dev);
+	struct rtnl_link_stats64 stats64;
+	struct u64_stats_sync *syncp;
+	unsigned int start;
 	int i;
 
 	if (netif_running(dev))
 		bcmgenet_update_mib_counters(priv);
 
+	dev_get_stats(dev, &stats64);
+
 	for (i = 0; i < BCMGENET_STATS_LEN; i++) {
 		const struct bcmgenet_stats *s;
 		char *p;
 
 		s = &bcmgenet_gstrings_stats[i];
-		if (s->type == BCMGENET_STAT_NETDEV)
-			p = (char *)&dev->stats;
-		else
-			p = (char *)priv;
-		p += s->stat_offset;
-		if (sizeof(unsigned long) != sizeof(u32) &&
-		    s->stat_sizeof == sizeof(unsigned long))
-			data[i] = *(unsigned long *)p;
-		else
-			data[i] = *(u32 *)p;
+		p = (char *)priv;
+
+		if (s->type == BCMGENET_STAT_SOFT64) {
+			syncp = (struct u64_stats_sync *)(p + s->syncp_offset);
+			do {
+				start = u64_stats_fetch_begin(syncp);
+				data[i] = u64_stats_read((u64_stats_t *)(p + s->stat_offset));
+			} while (u64_stats_fetch_retry(syncp, start));
+		} else {
+			if (s->type == BCMGENET_STAT_RTNL)
+				p = (char *)&stats64;
+
+			p += s->stat_offset;
+			if (sizeof(unsigned long) != sizeof(u32) &&
+				s->stat_sizeof == sizeof(unsigned long))
+				data[i] = *(unsigned long *)p;
+			else
+				data[i] = *(u32 *)p;
+		}
 	}
 }
 
-static void bcmgenet_eee_enable_set(struct net_device *dev, bool enable)
+void bcmgenet_eee_enable_set(struct net_device *dev, bool enable,
+			     bool tx_lpi_enabled)
 {
 	struct bcmgenet_priv *priv = netdev_priv(dev);
 	u32 off = priv->hw_params->tbuf_offset + TBUF_ENERGY_CTRL;
@@ -989,12 +1363,12 @@ static void bcmgenet_eee_enable_set(struct net_device *dev, bool enable)
 	bcmgenet_umac_writel(priv, reg, UMAC_EEE_CTRL);
 
 	/* Enable EEE and switch to a 27Mhz clock automatically */
-	reg = __raw_readl(priv->base + off);
-	if (enable)
+	reg = bcmgenet_readl(priv->base + off);
+	if (tx_lpi_enabled)
 		reg |= TBUF_EEE_EN | TBUF_PM_EN;
 	else
 		reg &= ~(TBUF_EEE_EN | TBUF_PM_EN);
-	__raw_writel(reg, priv->base + off);
+	bcmgenet_writel(reg, priv->base + off);
 
 	/* Do the same for thing for RBUF */
 	reg = bcmgenet_rbuf_readl(priv, RBUF_ENERGY_CTRL);
@@ -1010,53 +1384,302 @@ static void bcmgenet_eee_enable_set(struct net_device *dev, bool enable)
 	}
 
 	priv->eee.eee_enabled = enable;
-	priv->eee.eee_active = enable;
+	priv->eee.tx_lpi_enabled = tx_lpi_enabled;
 }
 
-static int bcmgenet_get_eee(struct net_device *dev, struct ethtool_eee *e)
+static int bcmgenet_get_eee(struct net_device *dev, struct ethtool_keee *e)
 {
 	struct bcmgenet_priv *priv = netdev_priv(dev);
-	struct ethtool_eee *p = &priv->eee;
+	struct ethtool_keee *p = &priv->eee;
 
 	if (GENET_IS_V1(priv))
 		return -EOPNOTSUPP;
 
-	e->eee_enabled = p->eee_enabled;
-	e->eee_active = p->eee_active;
+	if (!dev->phydev)
+		return -ENODEV;
+
+	e->tx_lpi_enabled = p->tx_lpi_enabled;
 	e->tx_lpi_timer = bcmgenet_umac_readl(priv, UMAC_EEE_LPI_TIMER);
 
-	return phy_ethtool_get_eee(priv->phydev, e);
+	return phy_ethtool_get_eee(dev->phydev, e);
 }
 
-static int bcmgenet_set_eee(struct net_device *dev, struct ethtool_eee *e)
+static int bcmgenet_set_eee(struct net_device *dev, struct ethtool_keee *e)
 {
 	struct bcmgenet_priv *priv = netdev_priv(dev);
-	struct ethtool_eee *p = &priv->eee;
-	int ret = 0;
+	struct ethtool_keee *p = &priv->eee;
+	bool active;
 
 	if (GENET_IS_V1(priv))
 		return -EOPNOTSUPP;
 
+	if (!dev->phydev)
+		return -ENODEV;
+
 	p->eee_enabled = e->eee_enabled;
 
 	if (!p->eee_enabled) {
-		bcmgenet_eee_enable_set(dev, false);
+		bcmgenet_eee_enable_set(dev, false, false);
 	} else {
-		ret = phy_init_eee(priv->phydev, 0);
-		if (ret) {
-			netif_err(priv, hw, dev, "EEE initialization failed\n");
-			return ret;
+		active = phy_init_eee(dev->phydev, false) >= 0;
+		bcmgenet_umac_writel(priv, e->tx_lpi_timer, UMAC_EEE_LPI_TIMER);
+		bcmgenet_eee_enable_set(dev, active, e->tx_lpi_enabled);
+	}
+
+	return phy_ethtool_set_eee(dev->phydev, e);
+}
+
+static int bcmgenet_validate_flow(struct net_device *dev,
+				  struct ethtool_rxnfc *cmd)
+{
+	struct ethtool_usrip4_spec *l4_mask;
+	struct ethhdr *eth_mask;
+
+	if (cmd->fs.location >= MAX_NUM_OF_FS_RULES &&
+	    cmd->fs.location != RX_CLS_LOC_ANY) {
+		netdev_err(dev, "rxnfc: Invalid location (%d)\n",
+			   cmd->fs.location);
+		return -EINVAL;
+	}
+
+	switch (cmd->fs.flow_type & ~(FLOW_EXT | FLOW_MAC_EXT)) {
+	case IP_USER_FLOW:
+		l4_mask = &cmd->fs.m_u.usr_ip4_spec;
+		/* don't allow mask which isn't valid */
+		if (VALIDATE_MASK(l4_mask->ip4src) ||
+		    VALIDATE_MASK(l4_mask->ip4dst) ||
+		    VALIDATE_MASK(l4_mask->l4_4_bytes) ||
+		    VALIDATE_MASK(l4_mask->proto) ||
+		    VALIDATE_MASK(l4_mask->ip_ver) ||
+		    VALIDATE_MASK(l4_mask->tos)) {
+			netdev_err(dev, "rxnfc: Unsupported mask\n");
+			return -EINVAL;
 		}
+		break;
+	case ETHER_FLOW:
+		eth_mask = &cmd->fs.m_u.ether_spec;
+		/* don't allow mask which isn't valid */
+		if (VALIDATE_MASK(eth_mask->h_dest) ||
+		    VALIDATE_MASK(eth_mask->h_source) ||
+		    VALIDATE_MASK(eth_mask->h_proto)) {
+			netdev_err(dev, "rxnfc: Unsupported mask\n");
+			return -EINVAL;
+		}
+		break;
+	default:
+		netdev_err(dev, "rxnfc: Unsupported flow type (0x%x)\n",
+			   cmd->fs.flow_type);
+		return -EINVAL;
+	}
 
-		bcmgenet_umac_writel(priv, e->tx_lpi_timer, UMAC_EEE_LPI_TIMER);
-		bcmgenet_eee_enable_set(dev, true);
+	if ((cmd->fs.flow_type & FLOW_EXT)) {
+		/* don't allow mask which isn't valid */
+		if (VALIDATE_MASK(cmd->fs.m_ext.vlan_etype) ||
+		    VALIDATE_MASK(cmd->fs.m_ext.vlan_tci)) {
+			netdev_err(dev, "rxnfc: Unsupported mask\n");
+			return -EINVAL;
+		}
+		if (cmd->fs.m_ext.data[0] || cmd->fs.m_ext.data[1]) {
+			netdev_err(dev, "rxnfc: user-def not supported\n");
+			return -EINVAL;
+		}
+	}
+
+	if ((cmd->fs.flow_type & FLOW_MAC_EXT)) {
+		/* don't allow mask which isn't valid */
+		if (VALIDATE_MASK(cmd->fs.m_ext.h_dest)) {
+			netdev_err(dev, "rxnfc: Unsupported mask\n");
+			return -EINVAL;
+		}
+	}
+
+	return 0;
+}
+
+static int bcmgenet_insert_flow(struct net_device *dev,
+				struct ethtool_rxnfc *cmd)
+{
+	struct bcmgenet_priv *priv = netdev_priv(dev);
+	struct bcmgenet_rxnfc_rule *loc_rule;
+	int err, i;
+
+	if (priv->hw_params->hfb_filter_size < 128) {
+		netdev_err(dev, "rxnfc: Not supported by this device\n");
+		return -EINVAL;
+	}
+
+	if (cmd->fs.ring_cookie > priv->hw_params->rx_queues &&
+	    cmd->fs.ring_cookie != RX_CLS_FLOW_WAKE &&
+	    cmd->fs.ring_cookie != RX_CLS_FLOW_DISC) {
+		netdev_err(dev, "rxnfc: Unsupported action (%llu)\n",
+			   cmd->fs.ring_cookie);
+		return -EINVAL;
+	}
+
+	err = bcmgenet_validate_flow(dev, cmd);
+	if (err)
+		return err;
+
+	if (cmd->fs.location == RX_CLS_LOC_ANY) {
+		list_for_each_entry(loc_rule, &priv->rxnfc_list, list) {
+			cmd->fs.location = loc_rule->fs.location;
+			err = memcmp(&loc_rule->fs, &cmd->fs,
+				     sizeof(struct ethtool_rx_flow_spec));
+			if (!err)
+				/* rule exists so return current location */
+				return 0;
+		}
+		for (i = 0; i < MAX_NUM_OF_FS_RULES; i++) {
+			loc_rule = &priv->rxnfc_rules[i];
+			if (loc_rule->state == BCMGENET_RXNFC_STATE_UNUSED) {
+				cmd->fs.location = i;
+				break;
+			}
+		}
+		if (i == MAX_NUM_OF_FS_RULES) {
+			cmd->fs.location = RX_CLS_LOC_ANY;
+			return -ENOSPC;
+		}
+	} else {
+		loc_rule = &priv->rxnfc_rules[cmd->fs.location];
+	}
+	if (loc_rule->state == BCMGENET_RXNFC_STATE_ENABLED)
+		bcmgenet_hfb_disable_filter(priv, cmd->fs.location + 1);
+	if (loc_rule->state != BCMGENET_RXNFC_STATE_UNUSED) {
+		list_del(&loc_rule->list);
+		bcmgenet_hfb_clear_filter(priv, cmd->fs.location + 1);
+	}
+	loc_rule->state = BCMGENET_RXNFC_STATE_UNUSED;
+	memcpy(&loc_rule->fs, &cmd->fs,
+	       sizeof(struct ethtool_rx_flow_spec));
+
+	bcmgenet_hfb_create_rxnfc_filter(priv, loc_rule);
+
+	list_add_tail(&loc_rule->list, &priv->rxnfc_list);
+
+	return 0;
+}
+
+static int bcmgenet_delete_flow(struct net_device *dev,
+				struct ethtool_rxnfc *cmd)
+{
+	struct bcmgenet_priv *priv = netdev_priv(dev);
+	struct bcmgenet_rxnfc_rule *rule;
+	int err = 0;
+
+	if (cmd->fs.location >= MAX_NUM_OF_FS_RULES)
+		return -EINVAL;
+
+	rule = &priv->rxnfc_rules[cmd->fs.location];
+	if (rule->state == BCMGENET_RXNFC_STATE_UNUSED) {
+		err =  -ENOENT;
+		goto out;
 	}
 
-	return phy_ethtool_set_eee(priv->phydev, e);
+	if (rule->state == BCMGENET_RXNFC_STATE_ENABLED)
+		bcmgenet_hfb_disable_filter(priv, cmd->fs.location + 1);
+	if (rule->state != BCMGENET_RXNFC_STATE_UNUSED) {
+		list_del(&rule->list);
+		bcmgenet_hfb_clear_filter(priv, cmd->fs.location + 1);
+	}
+	rule->state = BCMGENET_RXNFC_STATE_UNUSED;
+	memset(&rule->fs, 0, sizeof(struct ethtool_rx_flow_spec));
+
+out:
+	return err;
+}
+
+static int bcmgenet_set_rxnfc(struct net_device *dev, struct ethtool_rxnfc *cmd)
+{
+	struct bcmgenet_priv *priv = netdev_priv(dev);
+	int err = 0;
+
+	switch (cmd->cmd) {
+	case ETHTOOL_SRXCLSRLINS:
+		err = bcmgenet_insert_flow(dev, cmd);
+		break;
+	case ETHTOOL_SRXCLSRLDEL:
+		err = bcmgenet_delete_flow(dev, cmd);
+		break;
+	default:
+		netdev_warn(priv->dev, "Unsupported ethtool command. (%d)\n",
+			    cmd->cmd);
+		return -EINVAL;
+	}
+
+	return err;
+}
+
+static int bcmgenet_get_flow(struct net_device *dev, struct ethtool_rxnfc *cmd,
+			     int loc)
+{
+	struct bcmgenet_priv *priv = netdev_priv(dev);
+	struct bcmgenet_rxnfc_rule *rule;
+	int err = 0;
+
+	if (loc < 0 || loc >= MAX_NUM_OF_FS_RULES)
+		return -EINVAL;
+
+	rule = &priv->rxnfc_rules[loc];
+	if (rule->state == BCMGENET_RXNFC_STATE_UNUSED)
+		err = -ENOENT;
+	else
+		memcpy(&cmd->fs, &rule->fs,
+		       sizeof(struct ethtool_rx_flow_spec));
+
+	return err;
+}
+
+static int bcmgenet_get_num_flows(struct bcmgenet_priv *priv)
+{
+	struct list_head *pos;
+	int res = 0;
+
+	list_for_each(pos, &priv->rxnfc_list)
+		res++;
+
+	return res;
+}
+
+static int bcmgenet_get_rxnfc(struct net_device *dev, struct ethtool_rxnfc *cmd,
+			      u32 *rule_locs)
+{
+	struct bcmgenet_priv *priv = netdev_priv(dev);
+	struct bcmgenet_rxnfc_rule *rule;
+	int err = 0;
+	int i = 0;
+
+	switch (cmd->cmd) {
+	case ETHTOOL_GRXRINGS:
+		cmd->data = priv->hw_params->rx_queues ?: 1;
+		break;
+	case ETHTOOL_GRXCLSRLCNT:
+		cmd->rule_cnt = bcmgenet_get_num_flows(priv);
+		cmd->data = MAX_NUM_OF_FS_RULES | RX_CLS_LOC_SPECIAL;
+		break;
+	case ETHTOOL_GRXCLSRULE:
+		err = bcmgenet_get_flow(dev, cmd, cmd->fs.location);
+		break;
+	case ETHTOOL_GRXCLSRLALL:
+		list_for_each_entry(rule, &priv->rxnfc_list, list)
+			if (i < cmd->rule_cnt)
+				rule_locs[i++] = rule->fs.location;
+		cmd->rule_cnt = i;
+		cmd->data = MAX_NUM_OF_FS_RULES;
+		break;
+	default:
+		err = -EOPNOTSUPP;
+		break;
+	}
+
+	return err;
 }
 
 /* standard ethtool support functions. */
 static const struct ethtool_ops bcmgenet_ethtool_ops = {
+	.supported_coalesce_params = ETHTOOL_COALESCE_RX_USECS |
+				     ETHTOOL_COALESCE_MAX_FRAMES |
+				     ETHTOOL_COALESCE_USE_ADAPTIVE_RX,
 	.begin			= bcmgenet_begin,
 	.complete		= bcmgenet_complete,
 	.get_strings		= bcmgenet_get_strings,
@@ -1075,6 +1698,11 @@ static const struct ethtool_ops bcmgenet_ethtool_ops = {
 	.set_coalesce		= bcmgenet_set_coalesce,
 	.get_link_ksettings	= bcmgenet_get_link_ksettings,
 	.set_link_ksettings	= bcmgenet_set_link_ksettings,
+	.get_ts_info		= ethtool_op_get_ts_info,
+	.get_rxnfc		= bcmgenet_get_rxnfc,
+	.set_rxnfc		= bcmgenet_set_rxnfc,
+	.get_pauseparam		= bcmgenet_get_pauseparam,
+	.set_pauseparam		= bcmgenet_set_pauseparam,
 };
 
 /* Power down the unimac, based on mode. */
@@ -1086,7 +1714,7 @@ static int bcmgenet_power_down(struct bcmgenet_priv *priv,
 
 	switch (mode) {
 	case GENET_POWER_CABLE_SENSE:
-		phy_detach(priv->phydev);
+		phy_detach(priv->dev->phydev);
 		break;
 
 	case GENET_POWER_WOL_MAGIC:
@@ -1095,9 +1723,9 @@ static int bcmgenet_power_down(struct bcmgenet_priv *priv,
 
 	case GENET_POWER_PASSIVE:
 		/* Power down LED */
-		if (priv->hw_params->flags & GENET_HAS_EXT) {
+		if (bcmgenet_has_ext(priv)) {
 			reg = bcmgenet_ext_readl(priv, EXT_EXT_PWR_MGMT);
-			if (GENET_IS_V5(priv))
+			if (GENET_IS_V5(priv) && !bcmgenet_has_ephy_16nm(priv))
 				reg |= EXT_PWR_DOWN_PHY_EN |
 				       EXT_PWR_DOWN_PHY_RD |
 				       EXT_PWR_DOWN_PHY_SD |
@@ -1117,23 +1745,25 @@ static int bcmgenet_power_down(struct bcmgenet_priv *priv,
 		break;
 	}
 
-	return 0;
+	return ret;
 }
 
-static void bcmgenet_power_up(struct bcmgenet_priv *priv,
-			      enum bcmgenet_power_mode mode)
+static int bcmgenet_power_up(struct bcmgenet_priv *priv,
+			     enum bcmgenet_power_mode mode)
 {
+	int ret = 0;
 	u32 reg;
 
-	if (!(priv->hw_params->flags & GENET_HAS_EXT))
-		return;
+	if (!bcmgenet_has_ext(priv))
+		return ret;
 
 	reg = bcmgenet_ext_readl(priv, EXT_EXT_PWR_MGMT);
 
 	switch (mode) {
 	case GENET_POWER_PASSIVE:
-		reg &= ~(EXT_PWR_DOWN_DLL | EXT_PWR_DOWN_BIAS);
-		if (GENET_IS_V5(priv)) {
+		reg &= ~(EXT_PWR_DOWN_DLL | EXT_PWR_DOWN_BIAS |
+			 EXT_ENERGY_DET_MASK);
+		if (GENET_IS_V5(priv) && !bcmgenet_has_ephy_16nm(priv)) {
 			reg &= ~(EXT_PWR_DOWN_PHY_EN |
 				 EXT_PWR_DOWN_PHY_RD |
 				 EXT_PWR_DOWN_PHY_SD |
@@ -1151,7 +1781,6 @@ static void bcmgenet_power_up(struct bcmgenet_priv *priv,
 		}
 		bcmgenet_ext_writel(priv, reg, EXT_EXT_PWR_MGMT);
 		bcmgenet_phy_power_set(priv->dev, true);
-		bcmgenet_mii_reset(priv->dev);
 		break;
 
 	case GENET_POWER_CABLE_SENSE:
@@ -1162,25 +1791,13 @@ static void bcmgenet_power_up(struct bcmgenet_priv *priv,
 		}
 		break;
 	case GENET_POWER_WOL_MAGIC:
-		bcmgenet_wol_power_up_cfg(priv, mode);
-		return;
+		ret = bcmgenet_wol_power_up_cfg(priv, mode);
+		break;
 	default:
 		break;
 	}
-}
 
-/* ioctl handle special commands that are not present in ethtool. */
-static int bcmgenet_ioctl(struct net_device *dev, struct ifreq *rq, int cmd)
-{
-	struct bcmgenet_priv *priv = netdev_priv(dev);
-
-	if (!netif_running(dev))
-		return -EINVAL;
-
-	if (!priv->phydev)
-		return -ENODEV;
-
-	return phy_mii_ioctl(priv->phydev, rq, cmd);
+	return ret;
 }
 
 static struct enet_cb *bcmgenet_get_txcb(struct bcmgenet_priv *priv,
@@ -1200,24 +1817,21 @@ static struct enet_cb *bcmgenet_get_txcb(struct bcmgenet_priv *priv,
 	return tx_cb_ptr;
 }
 
-/* Simple helper to free a control block's resources */
-static void bcmgenet_free_cb(struct enet_cb *cb)
+static struct enet_cb *bcmgenet_put_txcb(struct bcmgenet_priv *priv,
+					 struct bcmgenet_tx_ring *ring)
 {
-	dev_kfree_skb_any(cb->skb);
-	cb->skb = NULL;
-	dma_unmap_addr_set(cb, dma_addr, 0);
-}
+	struct enet_cb *tx_cb_ptr;
 
-static inline void bcmgenet_rx_ring16_int_disable(struct bcmgenet_rx_ring *ring)
-{
-	bcmgenet_intrl2_0_writel(ring->priv, UMAC_IRQ_RXDMA_DONE,
-				 INTRL2_CPU_MASK_SET);
-}
+	tx_cb_ptr = ring->cbs;
+	tx_cb_ptr += ring->write_ptr - ring->cb_ptr;
 
-static inline void bcmgenet_rx_ring16_int_enable(struct bcmgenet_rx_ring *ring)
-{
-	bcmgenet_intrl2_0_writel(ring->priv, UMAC_IRQ_RXDMA_DONE,
-				 INTRL2_CPU_MASK_CLEAR);
+	/* Rewinding local write pointer */
+	if (ring->write_ptr == ring->cb_ptr)
+		ring->write_ptr = ring->end_ptr;
+	else
+		ring->write_ptr--;
+
+	return tx_cb_ptr;
 }
 
 static inline void bcmgenet_rx_ring_int_disable(struct bcmgenet_rx_ring *ring)
@@ -1234,18 +1848,6 @@ static inline void bcmgenet_rx_ring_int_enable(struct bcmgenet_rx_ring *ring)
 				 INTRL2_CPU_MASK_CLEAR);
 }
 
-static inline void bcmgenet_tx_ring16_int_disable(struct bcmgenet_tx_ring *ring)
-{
-	bcmgenet_intrl2_0_writel(ring->priv, UMAC_IRQ_TXDMA_DONE,
-				 INTRL2_CPU_MASK_SET);
-}
-
-static inline void bcmgenet_tx_ring16_int_enable(struct bcmgenet_tx_ring *ring)
-{
-	bcmgenet_intrl2_0_writel(ring->priv, UMAC_IRQ_TXDMA_DONE,
-				 INTRL2_CPU_MASK_CLEAR);
-}
-
 static inline void bcmgenet_tx_ring_int_enable(struct bcmgenet_tx_ring *ring)
 {
 	bcmgenet_intrl2_1_writel(ring->priv, 1 << ring->index,
@@ -1258,26 +1860,76 @@ static inline void bcmgenet_tx_ring_int_disable(struct bcmgenet_tx_ring *ring)
 				 INTRL2_CPU_MASK_SET);
 }
 
+/* Simple helper to free a transmit control block's resources
+ * Returns an skb when the last transmit control block associated with the
+ * skb is freed.  The skb should be freed by the caller if necessary.
+ */
+static struct sk_buff *bcmgenet_free_tx_cb(struct device *dev,
+					   struct enet_cb *cb)
+{
+	struct sk_buff *skb;
+
+	skb = cb->skb;
+
+	if (skb) {
+		cb->skb = NULL;
+		if (cb == GENET_CB(skb)->first_cb)
+			dma_unmap_single(dev, dma_unmap_addr(cb, dma_addr),
+					 dma_unmap_len(cb, dma_len),
+					 DMA_TO_DEVICE);
+		else
+			dma_unmap_page(dev, dma_unmap_addr(cb, dma_addr),
+				       dma_unmap_len(cb, dma_len),
+				       DMA_TO_DEVICE);
+		dma_unmap_addr_set(cb, dma_addr, 0);
+
+		if (cb == GENET_CB(skb)->last_cb)
+			return skb;
+
+	} else if (dma_unmap_addr(cb, dma_addr)) {
+		dma_unmap_page(dev,
+			       dma_unmap_addr(cb, dma_addr),
+			       dma_unmap_len(cb, dma_len),
+			       DMA_TO_DEVICE);
+		dma_unmap_addr_set(cb, dma_addr, 0);
+	}
+
+	return NULL;
+}
+
+/* Simple helper to free a receive control block's resources */
+static struct sk_buff *bcmgenet_free_rx_cb(struct device *dev,
+					   struct enet_cb *cb)
+{
+	struct sk_buff *skb;
+
+	skb = cb->skb;
+	cb->skb = NULL;
+
+	if (dma_unmap_addr(cb, dma_addr)) {
+		dma_unmap_single(dev, dma_unmap_addr(cb, dma_addr),
+				 dma_unmap_len(cb, dma_len), DMA_FROM_DEVICE);
+		dma_unmap_addr_set(cb, dma_addr, 0);
+	}
+
+	return skb;
+}
+
 /* Unlocked version of the reclaim routine */
 static unsigned int __bcmgenet_tx_reclaim(struct net_device *dev,
 					  struct bcmgenet_tx_ring *ring)
 {
+	struct bcmgenet_tx_stats64 *stats = &ring->stats64;
 	struct bcmgenet_priv *priv = netdev_priv(dev);
-	struct device *kdev = &priv->pdev->dev;
-	struct enet_cb *tx_cb_ptr;
-	unsigned int pkts_compl = 0;
+	unsigned int txbds_processed = 0;
 	unsigned int bytes_compl = 0;
-	unsigned int c_index;
+	unsigned int pkts_compl = 0;
 	unsigned int txbds_ready;
-	unsigned int txbds_processed = 0;
+	unsigned int c_index;
+	struct sk_buff *skb;
 
 	/* Clear status before servicing to reduce spurious interrupts */
-	if (ring->index == DESC_INDEX)
-		bcmgenet_intrl2_0_writel(priv, UMAC_IRQ_TXDMA_DONE,
-					 INTRL2_CPU_CLEAR);
-	else
-		bcmgenet_intrl2_1_writel(priv, (1 << ring->index),
-					 INTRL2_CPU_CLEAR);
+	bcmgenet_intrl2_1_writel(priv, (1 << ring->index), INTRL2_CPU_CLEAR);
 
 	/* Compute how many buffers are transmitted since last xmit call */
 	c_index = bcmgenet_tdma_ring_readl(priv, ring->index, TDMA_CONS_INDEX)
@@ -1290,21 +1942,12 @@ static unsigned int __bcmgenet_tx_reclaim(struct net_device *dev,
 
 	/* Reclaim transmitted buffers */
 	while (txbds_processed < txbds_ready) {
-		tx_cb_ptr = &priv->tx_cbs[ring->clean_ptr];
-		if (tx_cb_ptr->skb) {
+		skb = bcmgenet_free_tx_cb(&priv->pdev->dev,
+					  &priv->tx_cbs[ring->clean_ptr]);
+		if (skb) {
 			pkts_compl++;
-			bytes_compl += GENET_CB(tx_cb_ptr->skb)->bytes_sent;
-			dma_unmap_single(kdev,
-					 dma_unmap_addr(tx_cb_ptr, dma_addr),
-					 dma_unmap_len(tx_cb_ptr, dma_len),
-					 DMA_TO_DEVICE);
-			bcmgenet_free_cb(tx_cb_ptr);
-		} else if (dma_unmap_addr(tx_cb_ptr, dma_addr)) {
-			dma_unmap_page(kdev,
-				       dma_unmap_addr(tx_cb_ptr, dma_addr),
-				       dma_unmap_len(tx_cb_ptr, dma_len),
-				       DMA_TO_DEVICE);
-			dma_unmap_addr_set(tx_cb_ptr, dma_addr, 0);
+			bytes_compl += GENET_CB(skb)->bytes_sent;
+			dev_consume_skb_any(skb);
 		}
 
 		txbds_processed++;
@@ -1317,24 +1960,52 @@ static unsigned int __bcmgenet_tx_reclaim(struct net_device *dev,
 	ring->free_bds += txbds_processed;
 	ring->c_index = c_index;
 
-	ring->packets += pkts_compl;
-	ring->bytes += bytes_compl;
+	u64_stats_update_begin(&stats->syncp);
+	u64_stats_add(&stats->packets, pkts_compl);
+	u64_stats_add(&stats->bytes, bytes_compl);
+	u64_stats_update_end(&stats->syncp);
 
-	netdev_tx_completed_queue(netdev_get_tx_queue(dev, ring->queue),
+	netdev_tx_completed_queue(netdev_get_tx_queue(dev, ring->index),
 				  pkts_compl, bytes_compl);
 
 	return txbds_processed;
 }
 
 static unsigned int bcmgenet_tx_reclaim(struct net_device *dev,
-				struct bcmgenet_tx_ring *ring)
+				struct bcmgenet_tx_ring *ring,
+				bool all)
 {
-	unsigned int released;
-	unsigned long flags;
+	struct bcmgenet_priv *priv = netdev_priv(dev);
+	struct device *kdev = &priv->pdev->dev;
+	unsigned int released, drop, wr_ptr;
+	struct enet_cb *cb_ptr;
+	struct sk_buff *skb;
 
-	spin_lock_irqsave(&ring->lock, flags);
+	spin_lock_bh(&ring->lock);
 	released = __bcmgenet_tx_reclaim(dev, ring);
-	spin_unlock_irqrestore(&ring->lock, flags);
+	if (all) {
+		skb = NULL;
+		drop = (ring->prod_index - ring->c_index) & DMA_C_INDEX_MASK;
+		released += drop;
+		ring->prod_index = ring->c_index & DMA_C_INDEX_MASK;
+		while (drop--) {
+			cb_ptr = bcmgenet_put_txcb(priv, ring);
+			skb = cb_ptr->skb;
+			bcmgenet_free_tx_cb(kdev, cb_ptr);
+			if (skb && cb_ptr == GENET_CB(skb)->first_cb) {
+				dev_consume_skb_any(skb);
+				skb = NULL;
+			}
+		}
+		if (skb)
+			dev_consume_skb_any(skb);
+		bcmgenet_tdma_ring_writel(priv, ring->index,
+					  ring->prod_index, TDMA_PROD_INDEX);
+		wr_ptr = ring->write_ptr * WORDS_PER_BD(priv);
+		bcmgenet_tdma_ring_writel(priv, ring->index, wr_ptr,
+					  TDMA_WRITE_PTR);
+	}
+	spin_unlock_bh(&ring->lock);
 
 	return released;
 }
@@ -1345,19 +2016,18 @@ static int bcmgenet_tx_poll(struct napi_struct *napi, int budget)
 		container_of(napi, struct bcmgenet_tx_ring, napi);
 	unsigned int work_done = 0;
 	struct netdev_queue *txq;
-	unsigned long flags;
 
-	spin_lock_irqsave(&ring->lock, flags);
+	spin_lock(&ring->lock);
 	work_done = __bcmgenet_tx_reclaim(ring->priv->dev, ring);
 	if (ring->free_bds > (MAX_SKB_FRAGS + 1)) {
-		txq = netdev_get_tx_queue(ring->priv->dev, ring->queue);
+		txq = netdev_get_tx_queue(ring->priv->dev, ring->index);
 		netif_tx_wake_queue(txq);
 	}
-	spin_unlock_irqrestore(&ring->lock, flags);
+	spin_unlock(&ring->lock);
 
 	if (work_done == 0) {
 		napi_complete(napi);
-		ring->int_enable(ring);
+		bcmgenet_tx_ring_int_enable(ring);
 
 		return 0;
 	}
@@ -1368,116 +2038,27 @@ static int bcmgenet_tx_poll(struct napi_struct *napi, int budget)
 static void bcmgenet_tx_reclaim_all(struct net_device *dev)
 {
 	struct bcmgenet_priv *priv = netdev_priv(dev);
-	int i;
-
-	if (netif_is_multiqueue(dev)) {
-		for (i = 0; i < priv->hw_params->tx_queues; i++)
-			bcmgenet_tx_reclaim(dev, &priv->tx_rings[i]);
-	}
+	int i = 0;
 
-	bcmgenet_tx_reclaim(dev, &priv->tx_rings[DESC_INDEX]);
-}
-
-/* Transmits a single SKB (either head of a fragment or a single SKB)
- * caller must hold priv->lock
- */
-static int bcmgenet_xmit_single(struct net_device *dev,
-				struct sk_buff *skb,
-				u16 dma_desc_flags,
-				struct bcmgenet_tx_ring *ring)
-{
-	struct bcmgenet_priv *priv = netdev_priv(dev);
-	struct device *kdev = &priv->pdev->dev;
-	struct enet_cb *tx_cb_ptr;
-	unsigned int skb_len;
-	dma_addr_t mapping;
-	u32 length_status;
-	int ret;
-
-	tx_cb_ptr = bcmgenet_get_txcb(priv, ring);
-
-	if (unlikely(!tx_cb_ptr))
-		BUG();
-
-	tx_cb_ptr->skb = skb;
-
-	skb_len = skb_headlen(skb);
-
-	mapping = dma_map_single(kdev, skb->data, skb_len, DMA_TO_DEVICE);
-	ret = dma_mapping_error(kdev, mapping);
-	if (ret) {
-		priv->mib.tx_dma_failed++;
-		netif_err(priv, tx_err, dev, "Tx DMA map failed\n");
-		dev_kfree_skb(skb);
-		return ret;
-	}
-
-	dma_unmap_addr_set(tx_cb_ptr, dma_addr, mapping);
-	dma_unmap_len_set(tx_cb_ptr, dma_len, skb_len);
-	length_status = (skb_len << DMA_BUFLENGTH_SHIFT) | dma_desc_flags |
-			(priv->hw_params->qtag_mask << DMA_TX_QTAG_SHIFT) |
-			DMA_TX_APPEND_CRC;
-
-	if (skb->ip_summed == CHECKSUM_PARTIAL)
-		length_status |= DMA_TX_DO_CSUM;
-
-	dmadesc_set(priv, tx_cb_ptr->bd_addr, mapping, length_status);
-
-	return 0;
-}
-
-/* Transmit a SKB fragment */
-static int bcmgenet_xmit_frag(struct net_device *dev,
-			      skb_frag_t *frag,
-			      u16 dma_desc_flags,
-			      struct bcmgenet_tx_ring *ring)
-{
-	struct bcmgenet_priv *priv = netdev_priv(dev);
-	struct device *kdev = &priv->pdev->dev;
-	struct enet_cb *tx_cb_ptr;
-	unsigned int frag_size;
-	dma_addr_t mapping;
-	int ret;
-
-	tx_cb_ptr = bcmgenet_get_txcb(priv, ring);
-
-	if (unlikely(!tx_cb_ptr))
-		BUG();
-
-	tx_cb_ptr->skb = NULL;
-
-	frag_size = skb_frag_size(frag);
-
-	mapping = skb_frag_dma_map(kdev, frag, 0, frag_size, DMA_TO_DEVICE);
-	ret = dma_mapping_error(kdev, mapping);
-	if (ret) {
-		priv->mib.tx_dma_failed++;
-		netif_err(priv, tx_err, dev, "%s: Tx DMA map failed\n",
-			  __func__);
-		return ret;
-	}
-
-	dma_unmap_addr_set(tx_cb_ptr, dma_addr, mapping);
-	dma_unmap_len_set(tx_cb_ptr, dma_len, frag_size);
-
-	dmadesc_set(priv, tx_cb_ptr->bd_addr, mapping,
-		    (frag_size << DMA_BUFLENGTH_SHIFT) | dma_desc_flags |
-		    (priv->hw_params->qtag_mask << DMA_TX_QTAG_SHIFT));
-
-	return 0;
+	do {
+		bcmgenet_tx_reclaim(dev, &priv->tx_rings[i++], true);
+	} while (i <= priv->hw_params->tx_queues && netif_is_multiqueue(dev));
 }
 
 /* Reallocate the SKB to put enough headroom in front of it and insert
  * the transmit checksum offsets in the descriptors
  */
-static struct sk_buff *bcmgenet_put_tx_csum(struct net_device *dev,
-					    struct sk_buff *skb)
+static struct sk_buff *bcmgenet_add_tsb(struct net_device *dev,
+					struct sk_buff *skb,
+					struct bcmgenet_tx_ring *ring)
 {
+	struct bcmgenet_tx_stats64 *stats = &ring->stats64;
+	struct bcmgenet_priv *priv = netdev_priv(dev);
 	struct status_64 *status = NULL;
 	struct sk_buff *new_skb;
 	u16 offset;
 	u8 ip_proto;
-	u16 ip_ver;
+	__be16 ip_ver;
 	u32 tx_csum_info;
 
 	if (unlikely(skb_headroom(skb) < sizeof(*status))) {
@@ -1485,44 +2066,43 @@ static struct sk_buff *bcmgenet_put_tx_csum(struct net_device *dev,
 		 * enough headroom for us to insert 64B status block.
 		 */
 		new_skb = skb_realloc_headroom(skb, sizeof(*status));
-		dev_kfree_skb(skb);
 		if (!new_skb) {
-			dev->stats.tx_dropped++;
+			dev_kfree_skb_any(skb);
+			priv->mib.tx_realloc_tsb_failed++;
+			BCMGENET_STATS64_INC(stats, dropped);
 			return NULL;
 		}
+		dev_consume_skb_any(skb);
 		skb = new_skb;
+		priv->mib.tx_realloc_tsb++;
 	}
 
 	skb_push(skb, sizeof(*status));
 	status = (struct status_64 *)skb->data;
 
 	if (skb->ip_summed  == CHECKSUM_PARTIAL) {
-		ip_ver = htons(skb->protocol);
+		ip_ver = skb->protocol;
 		switch (ip_ver) {
-		case ETH_P_IP:
+		case htons(ETH_P_IP):
 			ip_proto = ip_hdr(skb)->protocol;
 			break;
-		case ETH_P_IPV6:
+		case htons(ETH_P_IPV6):
 			ip_proto = ipv6_hdr(skb)->nexthdr;
 			break;
 		default:
-			return skb;
+			/* don't use UDP flag */
+			ip_proto = 0;
+			break;
 		}
 
 		offset = skb_checksum_start_offset(skb) - sizeof(*status);
 		tx_csum_info = (offset << STATUS_TX_CSUM_START_SHIFT) |
-				(offset + skb->csum_offset);
+				(offset + skb->csum_offset) |
+				STATUS_TX_CSUM_LV;
 
-		/* Set the length valid bit for TCP and UDP and just set
-		 * the special UDP flag for IPv4, else just set to 0.
-		 */
-		if (ip_proto == IPPROTO_TCP || ip_proto == IPPROTO_UDP) {
-			tx_csum_info |= STATUS_TX_CSUM_LV;
-			if (ip_proto == IPPROTO_UDP && ip_ver == ETH_P_IP)
-				tx_csum_info |= STATUS_TX_CSUM_PROTO_UDP;
-		} else {
-			tx_csum_info = 0;
-		}
+		/* Set the special UDP flag for UDP */
+		if (ip_proto == IPPROTO_UDP)
+			tx_csum_info |= STATUS_TX_CSUM_PROTO_UDP;
 
 		status->tx_csum_info = tx_csum_info;
 	}
@@ -1530,89 +2110,112 @@ static struct sk_buff *bcmgenet_put_tx_csum(struct net_device *dev,
 	return skb;
 }
 
+static void bcmgenet_hide_tsb(struct sk_buff *skb)
+{
+	__skb_pull(skb, sizeof(struct status_64));
+}
+
 static netdev_tx_t bcmgenet_xmit(struct sk_buff *skb, struct net_device *dev)
 {
 	struct bcmgenet_priv *priv = netdev_priv(dev);
+	struct device *kdev = &priv->pdev->dev;
 	struct bcmgenet_tx_ring *ring = NULL;
+	struct enet_cb *tx_cb_ptr;
 	struct netdev_queue *txq;
-	unsigned long flags = 0;
 	int nr_frags, index;
-	u16 dma_desc_flags;
+	dma_addr_t mapping;
+	unsigned int size;
+	skb_frag_t *frag;
+	u32 len_stat;
 	int ret;
 	int i;
 
 	index = skb_get_queue_mapping(skb);
 	/* Mapping strategy:
-	 * queue_mapping = 0, unclassified, packet xmited through ring16
-	 * queue_mapping = 1, goes to ring 0. (highest priority queue
-	 * queue_mapping = 2, goes to ring 1.
-	 * queue_mapping = 3, goes to ring 2.
-	 * queue_mapping = 4, goes to ring 3.
+	 * queue_mapping = 0, unclassified, packet xmited through ring 0
+	 * queue_mapping = 1, goes to ring 1. (highest priority queue)
+	 * queue_mapping = 2, goes to ring 2.
+	 * queue_mapping = 3, goes to ring 3.
+	 * queue_mapping = 4, goes to ring 4.
 	 */
-	if (index == 0)
-		index = DESC_INDEX;
-	else
-		index -= 1;
-
 	ring = &priv->tx_rings[index];
-	txq = netdev_get_tx_queue(dev, ring->queue);
+	txq = netdev_get_tx_queue(dev, index);
 
 	nr_frags = skb_shinfo(skb)->nr_frags;
 
-	spin_lock_irqsave(&ring->lock, flags);
+	spin_lock(&ring->lock);
 	if (ring->free_bds <= (nr_frags + 1)) {
-		if (!netif_tx_queue_stopped(txq)) {
+		if (!netif_tx_queue_stopped(txq))
 			netif_tx_stop_queue(txq);
-			netdev_err(dev,
-				   "%s: tx ring %d full when queue %d awake\n",
-				   __func__, index, ring->queue);
-		}
 		ret = NETDEV_TX_BUSY;
 		goto out;
 	}
 
-	if (skb_padto(skb, ETH_ZLEN)) {
-		ret = NETDEV_TX_OK;
-		goto out;
-	}
-
 	/* Retain how many bytes will be sent on the wire, without TSB inserted
 	 * by transmit checksum offload
 	 */
 	GENET_CB(skb)->bytes_sent = skb->len;
 
-	/* set the SKB transmit checksum */
-	if (priv->desc_64b_en) {
-		skb = bcmgenet_put_tx_csum(dev, skb);
-		if (!skb) {
-			ret = NETDEV_TX_OK;
-			goto out;
-		}
-	}
-
-	dma_desc_flags = DMA_SOP;
-	if (nr_frags == 0)
-		dma_desc_flags |= DMA_EOP;
-
-	/* Transmit single SKB or head of fragment list */
-	ret = bcmgenet_xmit_single(dev, skb, dma_desc_flags, ring);
-	if (ret) {
+	/* add the Transmit Status Block */
+	skb = bcmgenet_add_tsb(dev, skb, ring);
+	if (!skb) {
 		ret = NETDEV_TX_OK;
 		goto out;
 	}
 
-	/* xmit fragment */
-	for (i = 0; i < nr_frags; i++) {
-		ret = bcmgenet_xmit_frag(dev,
-					 &skb_shinfo(skb)->frags[i],
-					 (i == nr_frags - 1) ? DMA_EOP : 0,
-					 ring);
+	for (i = 0; i <= nr_frags; i++) {
+		tx_cb_ptr = bcmgenet_get_txcb(priv, ring);
+
+		BUG_ON(!tx_cb_ptr);
+
+		if (!i) {
+			/* Transmit single SKB or head of fragment list */
+			GENET_CB(skb)->first_cb = tx_cb_ptr;
+			size = skb_headlen(skb);
+			mapping = dma_map_single(kdev, skb->data, size,
+						 DMA_TO_DEVICE);
+		} else {
+			/* xmit fragment */
+			frag = &skb_shinfo(skb)->frags[i - 1];
+			size = skb_frag_size(frag);
+			mapping = skb_frag_dma_map(kdev, frag, 0, size,
+						   DMA_TO_DEVICE);
+		}
+
+		ret = dma_mapping_error(kdev, mapping);
 		if (ret) {
+			priv->mib.tx_dma_failed++;
+			netif_err(priv, tx_err, dev, "Tx DMA map failed\n");
 			ret = NETDEV_TX_OK;
-			goto out;
+			goto out_unmap_frags;
+		}
+		dma_unmap_addr_set(tx_cb_ptr, dma_addr, mapping);
+		dma_unmap_len_set(tx_cb_ptr, dma_len, size);
+
+		tx_cb_ptr->skb = skb;
+
+		len_stat = (size << DMA_BUFLENGTH_SHIFT) |
+			   (priv->hw_params->qtag_mask << DMA_TX_QTAG_SHIFT);
+
+		/* Note: if we ever change from DMA_TX_APPEND_CRC below we
+		 * will need to restore software padding of "runt" packets
+		 */
+		len_stat |= DMA_TX_APPEND_CRC;
+
+		if (!i) {
+			len_stat |= DMA_SOP;
+			if (skb->ip_summed == CHECKSUM_PARTIAL)
+				len_stat |= DMA_TX_DO_CSUM;
 		}
+		if (i == nr_frags)
+			len_stat |= DMA_EOP;
+
+		dmadesc_set(priv, tx_cb_ptr->bd_addr, mapping, len_stat);
 	}
 
+	GENET_CB(skb)->last_cb = tx_cb_ptr;
+
+	bcmgenet_hide_tsb(skb);
 	skb_tx_timestamp(skb);
 
 	/* Decrement total BD count and advance our write pointer */
@@ -1625,14 +2228,27 @@ static netdev_tx_t bcmgenet_xmit(struct sk_buff *skb, struct net_device *dev)
 	if (ring->free_bds <= (MAX_SKB_FRAGS + 1))
 		netif_tx_stop_queue(txq);
 
-	if (!skb->xmit_more || netif_xmit_stopped(txq))
+	if (!netdev_xmit_more() || netif_xmit_stopped(txq))
 		/* Packets are ready, update producer index */
 		bcmgenet_tdma_ring_writel(priv, ring->index,
 					  ring->prod_index, TDMA_PROD_INDEX);
 out:
-	spin_unlock_irqrestore(&ring->lock, flags);
+	spin_unlock(&ring->lock);
 
 	return ret;
+
+out_unmap_frags:
+	/* Back up for failed control block mapping */
+	bcmgenet_put_txcb(priv, ring);
+
+	/* Unmap successfully mapped control blocks */
+	while (i-- > 0) {
+		tx_cb_ptr = bcmgenet_put_txcb(priv, ring);
+		bcmgenet_free_tx_cb(kdev, tx_cb_ptr);
+	}
+
+	dev_kfree_skb(skb);
+	goto out;
 }
 
 static struct sk_buff *bcmgenet_rx_refill(struct bcmgenet_priv *priv,
@@ -1644,7 +2260,8 @@ static struct sk_buff *bcmgenet_rx_refill(struct bcmgenet_priv *priv,
 	dma_addr_t mapping;
 
 	/* Allocate a new Rx skb */
-	skb = netdev_alloc_skb(priv->dev, priv->rx_buf_len + SKB_ALIGNMENT);
+	skb = __netdev_alloc_skb(priv->dev, priv->rx_buf_len + SKB_ALIGNMENT,
+				 GFP_ATOMIC | __GFP_NOWARN);
 	if (!skb) {
 		priv->mib.alloc_rx_buff_failed++;
 		netif_err(priv, rx_err, priv->dev,
@@ -1664,14 +2281,12 @@ static struct sk_buff *bcmgenet_rx_refill(struct bcmgenet_priv *priv,
 	}
 
 	/* Grab the current Rx skb from the ring and DMA-unmap it */
-	rx_skb = cb->skb;
-	if (likely(rx_skb))
-		dma_unmap_single(kdev, dma_unmap_addr(cb, dma_addr),
-				 priv->rx_buf_len, DMA_FROM_DEVICE);
+	rx_skb = bcmgenet_free_rx_cb(kdev, cb);
 
 	/* Put the new Rx skb on the ring */
 	cb->skb = skb;
 	dma_unmap_addr_set(cb, dma_addr, mapping);
+	dma_unmap_len_set(cb, dma_len, priv->rx_buf_len);
 	dmadesc_set_addr(priv, cb->bd_addr, mapping);
 
 	/* Return the current Rx skb to caller */
@@ -1684,6 +2299,7 @@ static struct sk_buff *bcmgenet_rx_refill(struct bcmgenet_priv *priv,
 static unsigned int bcmgenet_desc_rx(struct bcmgenet_rx_ring *ring,
 				     unsigned int budget)
 {
+	struct bcmgenet_rx_stats64 *stats = &ring->stats64;
 	struct bcmgenet_priv *priv = ring->priv;
 	struct net_device *dev = priv->dev;
 	struct enet_cb *cb;
@@ -1692,20 +2308,13 @@ static unsigned int bcmgenet_desc_rx(struct bcmgenet_rx_ring *ring,
 	unsigned long dma_flag;
 	int len;
 	unsigned int rxpktprocessed = 0, rxpkttoprocess;
+	unsigned int bytes_processed = 0;
 	unsigned int p_index, mask;
 	unsigned int discards;
-	unsigned int chksum_ok = 0;
 
 	/* Clear status before servicing to reduce spurious interrupts */
-	if (ring->index == DESC_INDEX) {
-		bcmgenet_intrl2_0_writel(priv, UMAC_IRQ_RXDMA_DONE,
-					 INTRL2_CPU_CLEAR);
-	} else {
-		mask = 1 << (UMAC_IRQ1_RX_INTR_SHIFT + ring->index);
-		bcmgenet_intrl2_1_writel(priv,
-					 mask,
-					 INTRL2_CPU_CLEAR);
-	}
+	mask = 1 << (UMAC_IRQ1_RX_INTR_SHIFT + ring->index);
+	bcmgenet_intrl2_1_writel(priv, mask, INTRL2_CPU_CLEAR);
 
 	p_index = bcmgenet_rdma_ring_readl(priv, ring->index, RDMA_PROD_INDEX);
 
@@ -1713,7 +2322,7 @@ static unsigned int bcmgenet_desc_rx(struct bcmgenet_rx_ring *ring,
 		   DMA_P_INDEX_DISCARD_CNT_MASK;
 	if (discards > ring->old_discards) {
 		discards = discards - ring->old_discards;
-		ring->errors += discards;
+		BCMGENET_STATS64_ADD(stats, missed, discards);
 		ring->old_discards += discards;
 
 		/* Clear HW register when we reach 75% of maximum 0xFFFF */
@@ -1732,22 +2341,25 @@ static unsigned int bcmgenet_desc_rx(struct bcmgenet_rx_ring *ring,
 
 	while ((rxpktprocessed < rxpkttoprocess) &&
 	       (rxpktprocessed < budget)) {
+		struct status_64 *status;
+		__be16 rx_csum;
+
 		cb = &priv->rx_cbs[ring->read_ptr];
 		skb = bcmgenet_rx_refill(priv, cb);
 
 		if (unlikely(!skb)) {
-			ring->dropped++;
+			BCMGENET_STATS64_INC(stats, dropped);
 			goto next;
 		}
 
-		if (!priv->desc_64b_en) {
-			dma_length_status =
-				dmadesc_get_length_status(priv, cb->bd_addr);
-		} else {
-			struct status_64 *status;
-
-			status = (struct status_64 *)skb->data;
-			dma_length_status = status->length_status;
+		status = (struct status_64 *)skb->data;
+		dma_length_status = status->length_status;
+		if (dev->features & NETIF_F_RXCSUM) {
+			rx_csum = (__force __be16)(status->rx_csum & 0xffff);
+			if (rx_csum) {
+				skb->csum = (__force __wsum)ntohs(rx_csum);
+				skb->ip_summed = CHECKSUM_COMPLETE;
+			}
 		}
 
 		/* DMA flags and length are still valid no matter how
@@ -1761,10 +2373,17 @@ static unsigned int bcmgenet_desc_rx(struct bcmgenet_rx_ring *ring,
 			  __func__, p_index, ring->c_index,
 			  ring->read_ptr, dma_length_status);
 
+		if (unlikely(len > RX_BUF_LENGTH)) {
+			netif_err(priv, rx_status, dev, "oversized packet\n");
+			BCMGENET_STATS64_INC(stats, length_errors);
+			dev_kfree_skb_any(skb);
+			goto next;
+		}
+
 		if (unlikely(!(dma_flag & DMA_EOP) || !(dma_flag & DMA_SOP))) {
 			netif_err(priv, rx_status, dev,
 				  "dropping fragmented packet!\n");
-			ring->errors++;
+			BCMGENET_STATS64_INC(stats, fragmented_errors);
 			dev_kfree_skb_any(skb);
 			goto next;
 		}
@@ -1777,46 +2396,50 @@ static unsigned int bcmgenet_desc_rx(struct bcmgenet_rx_ring *ring,
 						DMA_RX_RXER))) {
 			netif_err(priv, rx_status, dev, "dma_flag=0x%x\n",
 				  (unsigned int)dma_flag);
+			u64_stats_update_begin(&stats->syncp);
 			if (dma_flag & DMA_RX_CRC_ERROR)
-				dev->stats.rx_crc_errors++;
+				u64_stats_inc(&stats->crc_errors);
 			if (dma_flag & DMA_RX_OV)
-				dev->stats.rx_over_errors++;
+				u64_stats_inc(&stats->over_errors);
 			if (dma_flag & DMA_RX_NO)
-				dev->stats.rx_frame_errors++;
+				u64_stats_inc(&stats->frame_errors);
 			if (dma_flag & DMA_RX_LG)
-				dev->stats.rx_length_errors++;
-			dev->stats.rx_errors++;
+				u64_stats_inc(&stats->length_errors);
+			if ((dma_flag & (DMA_RX_CRC_ERROR |
+						DMA_RX_OV |
+						DMA_RX_NO |
+						DMA_RX_LG |
+						DMA_RX_RXER)) == DMA_RX_RXER)
+				u64_stats_inc(&stats->errors);
+			u64_stats_update_end(&stats->syncp);
 			dev_kfree_skb_any(skb);
 			goto next;
 		} /* error packet */
 
-		chksum_ok = (dma_flag & priv->dma_rx_chk_bit) &&
-			     priv->desc_rxchk_en;
-
 		skb_put(skb, len);
-		if (priv->desc_64b_en) {
-			skb_pull(skb, 64);
-			len -= 64;
-		}
 
-		if (likely(chksum_ok))
-			skb->ip_summed = CHECKSUM_UNNECESSARY;
-
-		/* remove hardware 2bytes added for IP alignment */
-		skb_pull(skb, 2);
-		len -= 2;
+		/* remove RSB and hardware 2bytes added for IP alignment */
+		skb_pull(skb, 66);
+		len -= 66;
 
 		if (priv->crc_fwd_en) {
 			skb_trim(skb, len - ETH_FCS_LEN);
 			len -= ETH_FCS_LEN;
 		}
 
+		bytes_processed += len;
+
 		/*Finish setting up the received SKB and send it to the kernel*/
 		skb->protocol = eth_type_trans(skb, priv->dev);
-		ring->packets++;
-		ring->bytes += len;
+
+		u64_stats_update_begin(&stats->syncp);
+		u64_stats_inc(&stats->packets);
+		u64_stats_add(&stats->bytes, len);
 		if (dma_flag & DMA_RX_MULT)
-			dev->stats.multicast++;
+			u64_stats_inc(&stats->multicast);
+		else if (dma_flag & DMA_RX_BRDCAST)
+			u64_stats_inc(&stats->broadcast);
+		u64_stats_update_end(&stats->syncp);
 
 		/* Notify kernel */
 		napi_gro_receive(&ring->napi, skb);
@@ -1833,6 +2456,9 @@ next:
 		bcmgenet_rdma_ring_writel(priv, ring->index, ring->c_index, RDMA_CONS_INDEX);
 	}
 
+	ring->dim.bytes = bytes_processed;
+	ring->dim.packets = rxpktprocessed;
+
 	return rxpktprocessed;
 }
 
@@ -1841,18 +2467,37 @@ static int bcmgenet_rx_poll(struct napi_struct *napi, int budget)
 {
 	struct bcmgenet_rx_ring *ring = container_of(napi,
 			struct bcmgenet_rx_ring, napi);
+	struct dim_sample dim_sample = {};
 	unsigned int work_done;
 
 	work_done = bcmgenet_desc_rx(ring, budget);
 
-	if (work_done < budget) {
-		napi_complete_done(napi, work_done);
-		ring->int_enable(ring);
+	if (work_done < budget && napi_complete_done(napi, work_done))
+		bcmgenet_rx_ring_int_enable(ring);
+
+	if (ring->dim.use_dim) {
+		dim_update_sample(ring->dim.event_ctr, ring->dim.packets,
+				  ring->dim.bytes, &dim_sample);
+		net_dim(&ring->dim.dim, &dim_sample);
 	}
 
 	return work_done;
 }
 
+static void bcmgenet_dim_work(struct work_struct *work)
+{
+	struct dim *dim = container_of(work, struct dim, work);
+	struct bcmgenet_net_dim *ndim =
+			container_of(dim, struct bcmgenet_net_dim, dim);
+	struct bcmgenet_rx_ring *ring =
+			container_of(ndim, struct bcmgenet_rx_ring, dim);
+	struct dim_cq_moder cur_profile =
+			net_dim_get_rx_moderation(dim->mode, dim->profile_ix);
+
+	bcmgenet_set_rx_coalesce(ring, cur_profile.usec, cur_profile.pkts);
+	dim->state = DIM_START_MEASURE;
+}
+
 /* Assign skb to RX DMA descriptor. */
 static int bcmgenet_alloc_rx_buffers(struct bcmgenet_priv *priv,
 				     struct bcmgenet_rx_ring *ring)
@@ -1868,7 +2513,7 @@ static int bcmgenet_alloc_rx_buffers(struct bcmgenet_priv *priv,
 		cb = ring->cbs + i;
 		skb = bcmgenet_rx_refill(priv, cb);
 		if (skb)
-			dev_kfree_skb_any(skb);
+			dev_consume_skb_any(skb);
 		if (!cb->skb)
 			return -ENOMEM;
 	}
@@ -1878,22 +2523,16 @@ static int bcmgenet_alloc_rx_buffers(struct bcmgenet_priv *priv,
 
 static void bcmgenet_free_rx_buffers(struct bcmgenet_priv *priv)
 {
-	struct device *kdev = &priv->pdev->dev;
+	struct sk_buff *skb;
 	struct enet_cb *cb;
 	int i;
 
 	for (i = 0; i < priv->num_rx_bds; i++) {
 		cb = &priv->rx_cbs[i];
 
-		if (dma_unmap_addr(cb, dma_addr)) {
-			dma_unmap_single(kdev,
-					 dma_unmap_addr(cb, dma_addr),
-					 priv->rx_buf_len, DMA_FROM_DEVICE);
-			dma_unmap_addr_set(cb, dma_addr, 0);
-		}
-
-		if (cb->skb)
-			bcmgenet_free_cb(cb);
+		skb = bcmgenet_free_rx_cb(&priv->pdev->dev, cb);
+		if (skb)
+			dev_consume_skb_any(skb);
 	}
 }
 
@@ -1901,12 +2540,18 @@ static void umac_enable_set(struct bcmgenet_priv *priv, u32 mask, bool enable)
 {
 	u32 reg;
 
+	spin_lock_bh(&priv->reg_lock);
 	reg = bcmgenet_umac_readl(priv, UMAC_CMD);
+	if (reg & CMD_SW_RESET) {
+		spin_unlock_bh(&priv->reg_lock);
+		return;
+	}
 	if (enable)
 		reg |= mask;
 	else
 		reg &= ~mask;
 	bcmgenet_umac_writel(priv, reg, UMAC_CMD);
+	spin_unlock_bh(&priv->reg_lock);
 
 	/* UniMAC stops on a packet boundary, wait for a full-size packet
 	 * to be processed
@@ -1915,36 +2560,17 @@ static void umac_enable_set(struct bcmgenet_priv *priv, u32 mask, bool enable)
 		usleep_range(1000, 2000);
 }
 
-static int reset_umac(struct bcmgenet_priv *priv)
+static void reset_umac(struct bcmgenet_priv *priv)
 {
-	struct device *kdev = &priv->pdev->dev;
-	unsigned int timeout = 0;
-	u32 reg;
-
 	/* 7358a0/7552a0: bad default in RBUF_FLUSH_CTRL.umac_sw_rst */
 	bcmgenet_rbuf_ctrl_set(priv, 0);
 	udelay(10);
 
-	/* disable MAC while updating its registers */
-	bcmgenet_umac_writel(priv, 0, UMAC_CMD);
-
-	/* issue soft reset, wait for it to complete */
+	/* issue soft reset and disable MAC while updating its registers */
+	spin_lock_bh(&priv->reg_lock);
 	bcmgenet_umac_writel(priv, CMD_SW_RESET, UMAC_CMD);
-	while (timeout++ < 1000) {
-		reg = bcmgenet_umac_readl(priv, UMAC_CMD);
-		if (!(reg & CMD_SW_RESET))
-			return 0;
-
-		udelay(1);
-	}
-
-	if (timeout == 1000) {
-		dev_err(kdev,
-			"timeout waiting for MAC to come out of reset\n");
-		return -ETIMEDOUT;
-	}
-
-	return 0;
+	udelay(2);
+	spin_unlock_bh(&priv->reg_lock);
 }
 
 static void bcmgenet_intr_disable(struct bcmgenet_priv *priv)
@@ -1965,29 +2591,27 @@ static void bcmgenet_link_intr_enable(struct bcmgenet_priv *priv)
 	 */
 	if (priv->internal_phy) {
 		int0_enable |= UMAC_IRQ_LINK_EVENT;
+		if (GENET_IS_V1(priv) || GENET_IS_V2(priv) || GENET_IS_V3(priv))
+			int0_enable |= UMAC_IRQ_PHY_DET_R;
 	} else if (priv->ext_phy) {
 		int0_enable |= UMAC_IRQ_LINK_EVENT;
 	} else if (priv->phy_interface == PHY_INTERFACE_MODE_MOCA) {
-		if (priv->hw_params->flags & GENET_HAS_MOCA_LINK_DET)
+		if (bcmgenet_has_moca_link_det(priv))
 			int0_enable |= UMAC_IRQ_LINK_EVENT;
 	}
 	bcmgenet_intrl2_0_writel(priv, int0_enable, INTRL2_CPU_MASK_CLEAR);
 }
 
-static int init_umac(struct bcmgenet_priv *priv)
+static void init_umac(struct bcmgenet_priv *priv)
 {
 	struct device *kdev = &priv->pdev->dev;
-	int ret;
 	u32 reg;
 	u32 int0_enable = 0;
 
 	dev_dbg(&priv->pdev->dev, "bcmgenet: init_umac\n");
 
-	ret = reset_umac(priv);
-	if (ret)
-		return ret;
+	reset_umac(priv);
 
-	bcmgenet_umac_writel(priv, 0, UMAC_CMD);
 	/* clear tx/rx counter */
 	bcmgenet_umac_writel(priv,
 			     MIB_RESET_RX | MIB_RESET_TX | MIB_RESET_RUNT,
@@ -1996,11 +2620,28 @@ static int init_umac(struct bcmgenet_priv *priv)
 
 	bcmgenet_umac_writel(priv, ENET_MAX_MTU_SIZE, UMAC_MAX_FRAME_LEN);
 
-	/* init rx registers, enable ip header optimization */
+	/* init tx registers, enable TSB */
+	reg = bcmgenet_tbuf_ctrl_get(priv);
+	reg |= TBUF_64B_EN;
+	bcmgenet_tbuf_ctrl_set(priv, reg);
+
+	/* init rx registers, enable ip header optimization and RSB */
 	reg = bcmgenet_rbuf_readl(priv, RBUF_CTRL);
-	reg |= RBUF_ALIGN_2B;
+	reg |= RBUF_ALIGN_2B | RBUF_64B_EN;
 	bcmgenet_rbuf_writel(priv, reg, RBUF_CTRL);
 
+	/* enable rx checksumming */
+	reg = bcmgenet_rbuf_readl(priv, RBUF_CHK_CTRL);
+	reg |= RBUF_RXCHK_EN | RBUF_L3_PARSE_DIS;
+	/* If UniMAC forwards CRC, we need to skip over it to get
+	 * a valid CHK bit to be set in the per-packet status word
+	 */
+	if (priv->crc_fwd_en)
+		reg |= RBUF_SKIP_FCS;
+	else
+		reg &= ~RBUF_SKIP_FCS;
+	bcmgenet_rbuf_writel(priv, reg, RBUF_CHK_CTRL);
+
 	if (!GENET_IS_V1(priv) && !GENET_IS_V2(priv))
 		bcmgenet_rbuf_writel(priv, 1, RBUF_TBUF_SIZE_CTRL);
 
@@ -2020,14 +2661,43 @@ static int init_umac(struct bcmgenet_priv *priv)
 	}
 
 	/* Enable MDIO interrupts on GENET v3+ */
-	if (priv->hw_params->flags & GENET_HAS_MDIO_INTR)
-		int0_enable |= (UMAC_IRQ_MDIO_DONE | UMAC_IRQ_MDIO_ERROR);
+	if (bcmgenet_has_mdio_intr(priv))
+		int0_enable |= UMAC_IRQ_MDIO_EVENT;
 
 	bcmgenet_intrl2_0_writel(priv, int0_enable, INTRL2_CPU_MASK_CLEAR);
 
 	dev_dbg(kdev, "done init umac\n");
+}
 
-	return 0;
+static void bcmgenet_init_dim(struct bcmgenet_rx_ring *ring,
+			      void (*cb)(struct work_struct *work))
+{
+	struct bcmgenet_net_dim *dim = &ring->dim;
+
+	INIT_WORK(&dim->dim.work, cb);
+	dim->dim.mode = DIM_CQ_PERIOD_MODE_START_FROM_EQE;
+	dim->event_ctr = 0;
+	dim->packets = 0;
+	dim->bytes = 0;
+}
+
+static void bcmgenet_init_rx_coalesce(struct bcmgenet_rx_ring *ring)
+{
+	struct bcmgenet_net_dim *dim = &ring->dim;
+	struct dim_cq_moder moder;
+	u32 usecs, pkts;
+
+	usecs = ring->rx_coalesce_usecs;
+	pkts = ring->rx_max_coalesced_frames;
+
+	/* If DIM was enabled, re-apply default parameters */
+	if (dim->use_dim) {
+		moder = net_dim_get_def_rx_moderation(dim->dim.mode);
+		usecs = moder.usec;
+		pkts = moder.pkts;
+	}
+
+	bcmgenet_set_rx_coalesce(ring, usecs, pkts);
 }
 
 /* Initialize a Tx ring along with corresponding hardware registers */
@@ -2042,15 +2712,6 @@ static void bcmgenet_init_tx_ring(struct bcmgenet_priv *priv,
 	spin_lock_init(&ring->lock);
 	ring->priv = priv;
 	ring->index = index;
-	if (index == DESC_INDEX) {
-		ring->queue = 0;
-		ring->int_enable = bcmgenet_tx_ring16_int_enable;
-		ring->int_disable = bcmgenet_tx_ring16_int_disable;
-	} else {
-		ring->queue = index + 1;
-		ring->int_enable = bcmgenet_tx_ring_int_enable;
-		ring->int_disable = bcmgenet_tx_ring_int_disable;
-	}
 	ring->cbs = priv->tx_cbs + start_ptr;
 	ring->size = size;
 	ring->clean_ptr = start_ptr;
@@ -2061,8 +2722,8 @@ static void bcmgenet_init_tx_ring(struct bcmgenet_priv *priv,
 	ring->end_ptr = end_ptr - 1;
 	ring->prod_index = 0;
 
-	/* Set flow period for ring != 16 */
-	if (index != DESC_INDEX)
+	/* Set flow period for ring != 0 */
+	if (index)
 		flow_period_val = ENET_MAX_MTU_SIZE << 16;
 
 	bcmgenet_tdma_ring_writel(priv, index, 0, TDMA_PROD_INDEX);
@@ -2084,6 +2745,9 @@ static void bcmgenet_init_tx_ring(struct bcmgenet_priv *priv,
 				  TDMA_WRITE_PTR);
 	bcmgenet_tdma_ring_writel(priv, index, end_ptr * words_per_bd - 1,
 				  DMA_END_ADDR);
+
+	/* Initialize Tx NAPI */
+	netif_napi_add_tx(priv->dev, &ring->napi, bcmgenet_tx_poll);
 }
 
 /* Initialize a RDMA ring */
@@ -2097,13 +2761,6 @@ static int bcmgenet_init_rx_ring(struct bcmgenet_priv *priv,
 
 	ring->priv = priv;
 	ring->index = index;
-	if (index == DESC_INDEX) {
-		ring->int_enable = bcmgenet_rx_ring16_int_enable;
-		ring->int_disable = bcmgenet_rx_ring16_int_disable;
-	} else {
-		ring->int_enable = bcmgenet_rx_ring_int_enable;
-		ring->int_disable = bcmgenet_rx_ring_int_disable;
-	}
 	ring->cbs = priv->rx_cbs + start_ptr;
 	ring->size = size;
 	ring->c_index = 0;
@@ -2115,9 +2772,14 @@ static int bcmgenet_init_rx_ring(struct bcmgenet_priv *priv,
 	if (ret)
 		return ret;
 
+	bcmgenet_init_dim(ring, bcmgenet_dim_work);
+	bcmgenet_init_rx_coalesce(ring);
+
+	/* Initialize Rx NAPI */
+	netif_napi_add(priv->dev, &ring->napi, bcmgenet_rx_poll);
+
 	bcmgenet_rdma_ring_writel(priv, index, 0, RDMA_PROD_INDEX);
 	bcmgenet_rdma_ring_writel(priv, index, 0, RDMA_CONS_INDEX);
-	bcmgenet_rdma_ring_writel(priv, index, 1, DMA_MBUF_DONE_THRESH);
 	bcmgenet_rdma_ring_writel(priv, index,
 				  ((size << DMA_RING_SIZE_SHIFT) |
 				   RX_BUF_LENGTH), DMA_RING_BUF_SIZE);
@@ -2139,57 +2801,27 @@ static int bcmgenet_init_rx_ring(struct bcmgenet_priv *priv,
 	return ret;
 }
 
-static void bcmgenet_init_tx_napi(struct bcmgenet_priv *priv)
-{
-	unsigned int i;
-	struct bcmgenet_tx_ring *ring;
-
-	for (i = 0; i < priv->hw_params->tx_queues; ++i) {
-		ring = &priv->tx_rings[i];
-		netif_tx_napi_add(priv->dev, &ring->napi, bcmgenet_tx_poll, 64);
-	}
-
-	ring = &priv->tx_rings[DESC_INDEX];
-	netif_tx_napi_add(priv->dev, &ring->napi, bcmgenet_tx_poll, 64);
-}
-
 static void bcmgenet_enable_tx_napi(struct bcmgenet_priv *priv)
 {
 	unsigned int i;
-	u32 int0_enable = UMAC_IRQ_TXDMA_DONE;
-	u32 int1_enable = 0;
 	struct bcmgenet_tx_ring *ring;
 
-	for (i = 0; i < priv->hw_params->tx_queues; ++i) {
+	for (i = 0; i <= priv->hw_params->tx_queues; ++i) {
 		ring = &priv->tx_rings[i];
 		napi_enable(&ring->napi);
-		int1_enable |= (1 << i);
+		bcmgenet_tx_ring_int_enable(ring);
 	}
-
-	ring = &priv->tx_rings[DESC_INDEX];
-	napi_enable(&ring->napi);
-
-	bcmgenet_intrl2_0_writel(priv, int0_enable, INTRL2_CPU_MASK_CLEAR);
-	bcmgenet_intrl2_1_writel(priv, int1_enable, INTRL2_CPU_MASK_CLEAR);
 }
 
 static void bcmgenet_disable_tx_napi(struct bcmgenet_priv *priv)
 {
 	unsigned int i;
-	u32 int0_disable = UMAC_IRQ_TXDMA_DONE;
-	u32 int1_disable = 0xffff;
 	struct bcmgenet_tx_ring *ring;
 
-	bcmgenet_intrl2_0_writel(priv, int0_disable, INTRL2_CPU_MASK_SET);
-	bcmgenet_intrl2_1_writel(priv, int1_disable, INTRL2_CPU_MASK_SET);
-
-	for (i = 0; i < priv->hw_params->tx_queues; ++i) {
+	for (i = 0; i <= priv->hw_params->tx_queues; ++i) {
 		ring = &priv->tx_rings[i];
 		napi_disable(&ring->napi);
 	}
-
-	ring = &priv->tx_rings[DESC_INDEX];
-	napi_disable(&ring->napi);
 }
 
 static void bcmgenet_fini_tx_napi(struct bcmgenet_priv *priv)
@@ -2197,138 +2829,128 @@ static void bcmgenet_fini_tx_napi(struct bcmgenet_priv *priv)
 	unsigned int i;
 	struct bcmgenet_tx_ring *ring;
 
-	for (i = 0; i < priv->hw_params->tx_queues; ++i) {
+	for (i = 0; i <= priv->hw_params->tx_queues; ++i) {
 		ring = &priv->tx_rings[i];
 		netif_napi_del(&ring->napi);
 	}
+}
+
+static int bcmgenet_tdma_disable(struct bcmgenet_priv *priv)
+{
+	int timeout = 0;
+	u32 reg, mask;
+
+	reg = bcmgenet_tdma_readl(priv, DMA_CTRL);
+	mask = (1 << (priv->hw_params->tx_queues + 1)) - 1;
+	mask = (mask << DMA_RING_BUF_EN_SHIFT) | DMA_EN;
+	reg &= ~mask;
+	bcmgenet_tdma_writel(priv, reg, DMA_CTRL);
+
+	/* Check DMA status register to confirm DMA is disabled */
+	while (timeout++ < DMA_TIMEOUT_VAL) {
+		reg = bcmgenet_tdma_readl(priv, DMA_STATUS);
+		if ((reg & mask) == mask)
+			return 0;
+
+		udelay(1);
+	}
+
+	return -ETIMEDOUT;
+}
+
+static int bcmgenet_rdma_disable(struct bcmgenet_priv *priv)
+{
+	int timeout = 0;
+	u32 reg, mask;
+
+	reg = bcmgenet_rdma_readl(priv, DMA_CTRL);
+	mask = (1 << (priv->hw_params->rx_queues + 1)) - 1;
+	mask = (mask << DMA_RING_BUF_EN_SHIFT) | DMA_EN;
+	reg &= ~mask;
+	bcmgenet_rdma_writel(priv, reg, DMA_CTRL);
+
+	/* Check DMA status register to confirm DMA is disabled */
+	while (timeout++ < DMA_TIMEOUT_VAL) {
+		reg = bcmgenet_rdma_readl(priv, DMA_STATUS);
+		if ((reg & mask) == mask)
+			return 0;
+
+		udelay(1);
+	}
 
-	ring = &priv->tx_rings[DESC_INDEX];
-	netif_napi_del(&ring->napi);
+	return -ETIMEDOUT;
 }
 
 /* Initialize Tx queues
  *
- * Queues 0-3 are priority-based, each one has 32 descriptors,
- * with queue 0 being the highest priority queue.
+ * Queues 1-4 are priority-based, each one has 32 descriptors,
+ * with queue 1 being the highest priority queue.
  *
- * Queue 16 is the default Tx queue with
- * GENET_Q16_TX_BD_CNT = 256 - 4 * 32 = 128 descriptors.
+ * Queue 0 is the default Tx queue with
+ * GENET_Q0_TX_BD_CNT = 256 - 4 * 32 = 128 descriptors.
  *
  * The transmit control block pool is then partitioned as follows:
- * - Tx queue 0 uses tx_cbs[0..31]
- * - Tx queue 1 uses tx_cbs[32..63]
- * - Tx queue 2 uses tx_cbs[64..95]
- * - Tx queue 3 uses tx_cbs[96..127]
- * - Tx queue 16 uses tx_cbs[128..255]
+ * - Tx queue 0 uses tx_cbs[0..127]
+ * - Tx queue 1 uses tx_cbs[128..159]
+ * - Tx queue 2 uses tx_cbs[160..191]
+ * - Tx queue 3 uses tx_cbs[192..223]
+ * - Tx queue 4 uses tx_cbs[224..255]
  */
 static void bcmgenet_init_tx_queues(struct net_device *dev)
 {
 	struct bcmgenet_priv *priv = netdev_priv(dev);
-	u32 i, dma_enable;
-	u32 dma_ctrl, ring_cfg;
-	u32 dma_priority[3] = {0, 0, 0};
-
-	dma_ctrl = bcmgenet_tdma_readl(priv, DMA_CTRL);
-	dma_enable = dma_ctrl & DMA_EN;
-	dma_ctrl &= ~DMA_EN;
-	bcmgenet_tdma_writel(priv, dma_ctrl, DMA_CTRL);
-
-	dma_ctrl = 0;
-	ring_cfg = 0;
+	unsigned int start = 0, end = GENET_Q0_TX_BD_CNT;
+	u32 i, ring_mask, dma_priority[3] = {0, 0, 0};
 
 	/* Enable strict priority arbiter mode */
 	bcmgenet_tdma_writel(priv, DMA_ARBITER_SP, DMA_ARB_CTRL);
 
 	/* Initialize Tx priority queues */
-	for (i = 0; i < priv->hw_params->tx_queues; i++) {
-		bcmgenet_init_tx_ring(priv, i, priv->hw_params->tx_bds_per_q,
-				      i * priv->hw_params->tx_bds_per_q,
-				      (i + 1) * priv->hw_params->tx_bds_per_q);
-		ring_cfg |= (1 << i);
-		dma_ctrl |= (1 << (i + DMA_RING_BUF_EN_SHIFT));
+	for (i = 0; i <= priv->hw_params->tx_queues; i++) {
+		bcmgenet_init_tx_ring(priv, i, end - start, start, end);
+		start = end;
+		end += priv->hw_params->tx_bds_per_q;
 		dma_priority[DMA_PRIO_REG_INDEX(i)] |=
-			((GENET_Q0_PRIORITY + i) << DMA_PRIO_REG_SHIFT(i));
+			(i ? GENET_Q1_PRIORITY : GENET_Q0_PRIORITY)
+			<< DMA_PRIO_REG_SHIFT(i);
 	}
 
-	/* Initialize Tx default queue 16 */
-	bcmgenet_init_tx_ring(priv, DESC_INDEX, GENET_Q16_TX_BD_CNT,
-			      priv->hw_params->tx_queues *
-			      priv->hw_params->tx_bds_per_q,
-			      TOTAL_DESC);
-	ring_cfg |= (1 << DESC_INDEX);
-	dma_ctrl |= (1 << (DESC_INDEX + DMA_RING_BUF_EN_SHIFT));
-	dma_priority[DMA_PRIO_REG_INDEX(DESC_INDEX)] |=
-		((GENET_Q0_PRIORITY + priv->hw_params->tx_queues) <<
-		 DMA_PRIO_REG_SHIFT(DESC_INDEX));
-
 	/* Set Tx queue priorities */
 	bcmgenet_tdma_writel(priv, dma_priority[0], DMA_PRIORITY_0);
 	bcmgenet_tdma_writel(priv, dma_priority[1], DMA_PRIORITY_1);
 	bcmgenet_tdma_writel(priv, dma_priority[2], DMA_PRIORITY_2);
 
-	/* Initialize Tx NAPI */
-	bcmgenet_init_tx_napi(priv);
-
-	/* Enable Tx queues */
-	bcmgenet_tdma_writel(priv, ring_cfg, DMA_RING_CFG);
+	/* Configure Tx queues as descriptor rings */
+	ring_mask = (1 << (priv->hw_params->tx_queues + 1)) - 1;
+	bcmgenet_tdma_writel(priv, ring_mask, DMA_RING_CFG);
 
-	/* Enable Tx DMA */
-	if (dma_enable)
-		dma_ctrl |= DMA_EN;
-	bcmgenet_tdma_writel(priv, dma_ctrl, DMA_CTRL);
-}
-
-static void bcmgenet_init_rx_napi(struct bcmgenet_priv *priv)
-{
-	unsigned int i;
-	struct bcmgenet_rx_ring *ring;
-
-	for (i = 0; i < priv->hw_params->rx_queues; ++i) {
-		ring = &priv->rx_rings[i];
-		netif_napi_add(priv->dev, &ring->napi, bcmgenet_rx_poll, 64);
-	}
-
-	ring = &priv->rx_rings[DESC_INDEX];
-	netif_napi_add(priv->dev, &ring->napi, bcmgenet_rx_poll, 64);
+	/* Enable Tx rings */
+	ring_mask <<= DMA_RING_BUF_EN_SHIFT;
+	bcmgenet_tdma_writel(priv, ring_mask, DMA_CTRL);
 }
 
 static void bcmgenet_enable_rx_napi(struct bcmgenet_priv *priv)
 {
 	unsigned int i;
-	u32 int0_enable = UMAC_IRQ_RXDMA_DONE;
-	u32 int1_enable = 0;
 	struct bcmgenet_rx_ring *ring;
 
-	for (i = 0; i < priv->hw_params->rx_queues; ++i) {
+	for (i = 0; i <= priv->hw_params->rx_queues; ++i) {
 		ring = &priv->rx_rings[i];
 		napi_enable(&ring->napi);
-		int1_enable |= (1 << (UMAC_IRQ1_RX_INTR_SHIFT + i));
+		bcmgenet_rx_ring_int_enable(ring);
 	}
-
-	ring = &priv->rx_rings[DESC_INDEX];
-	napi_enable(&ring->napi);
-
-	bcmgenet_intrl2_0_writel(priv, int0_enable, INTRL2_CPU_MASK_CLEAR);
-	bcmgenet_intrl2_1_writel(priv, int1_enable, INTRL2_CPU_MASK_CLEAR);
 }
 
 static void bcmgenet_disable_rx_napi(struct bcmgenet_priv *priv)
 {
 	unsigned int i;
-	u32 int0_disable = UMAC_IRQ_RXDMA_DONE;
-	u32 int1_disable = 0xffff << UMAC_IRQ1_RX_INTR_SHIFT;
 	struct bcmgenet_rx_ring *ring;
 
-	bcmgenet_intrl2_0_writel(priv, int0_disable, INTRL2_CPU_MASK_SET);
-	bcmgenet_intrl2_1_writel(priv, int1_disable, INTRL2_CPU_MASK_SET);
-
-	for (i = 0; i < priv->hw_params->rx_queues; ++i) {
+	for (i = 0; i <= priv->hw_params->rx_queues; ++i) {
 		ring = &priv->rx_rings[i];
 		napi_disable(&ring->napi);
+		cancel_work_sync(&ring->dim.dim.work);
 	}
-
-	ring = &priv->rx_rings[DESC_INDEX];
-	napi_disable(&ring->napi);
 }
 
 static void bcmgenet_fini_rx_napi(struct bcmgenet_priv *priv)
@@ -2336,13 +2958,10 @@ static void bcmgenet_fini_rx_napi(struct bcmgenet_priv *priv)
 	unsigned int i;
 	struct bcmgenet_rx_ring *ring;
 
-	for (i = 0; i < priv->hw_params->rx_queues; ++i) {
+	for (i = 0; i <= priv->hw_params->rx_queues; ++i) {
 		ring = &priv->rx_rings[i];
 		netif_napi_del(&ring->napi);
 	}
-
-	ring = &priv->rx_rings[DESC_INDEX];
-	netif_napi_del(&ring->napi);
 }
 
 /* Initialize Rx queues
@@ -2350,60 +2969,32 @@ static void bcmgenet_fini_rx_napi(struct bcmgenet_priv *priv)
  * Queues 0-15 are priority queues. Hardware Filtering Block (HFB) can be
  * used to direct traffic to these queues.
  *
- * Queue 16 is the default Rx queue with GENET_Q16_RX_BD_CNT descriptors.
+ * Queue 0 is also the default Rx queue with GENET_Q0_RX_BD_CNT descriptors.
  */
 static int bcmgenet_init_rx_queues(struct net_device *dev)
 {
 	struct bcmgenet_priv *priv = netdev_priv(dev);
-	u32 i;
-	u32 dma_enable;
-	u32 dma_ctrl;
-	u32 ring_cfg;
+	unsigned int start = 0, end = GENET_Q0_RX_BD_CNT;
+	u32 i, ring_mask;
 	int ret;
 
-	dma_ctrl = bcmgenet_rdma_readl(priv, DMA_CTRL);
-	dma_enable = dma_ctrl & DMA_EN;
-	dma_ctrl &= ~DMA_EN;
-	bcmgenet_rdma_writel(priv, dma_ctrl, DMA_CTRL);
-
-	dma_ctrl = 0;
-	ring_cfg = 0;
-
 	/* Initialize Rx priority queues */
-	for (i = 0; i < priv->hw_params->rx_queues; i++) {
-		ret = bcmgenet_init_rx_ring(priv, i,
-					    priv->hw_params->rx_bds_per_q,
-					    i * priv->hw_params->rx_bds_per_q,
-					    (i + 1) *
-					    priv->hw_params->rx_bds_per_q);
+	for (i = 0; i <= priv->hw_params->rx_queues; i++) {
+		ret = bcmgenet_init_rx_ring(priv, i, end - start, start, end);
 		if (ret)
 			return ret;
 
-		ring_cfg |= (1 << i);
-		dma_ctrl |= (1 << (i + DMA_RING_BUF_EN_SHIFT));
+		start = end;
+		end += priv->hw_params->rx_bds_per_q;
 	}
 
-	/* Initialize Rx default queue 16 */
-	ret = bcmgenet_init_rx_ring(priv, DESC_INDEX, GENET_Q16_RX_BD_CNT,
-				    priv->hw_params->rx_queues *
-				    priv->hw_params->rx_bds_per_q,
-				    TOTAL_DESC);
-	if (ret)
-		return ret;
-
-	ring_cfg |= (1 << DESC_INDEX);
-	dma_ctrl |= (1 << (DESC_INDEX + DMA_RING_BUF_EN_SHIFT));
-
-	/* Initialize Rx NAPI */
-	bcmgenet_init_rx_napi(priv);
+	/* Configure Rx queues as descriptor rings */
+	ring_mask = (1 << (priv->hw_params->rx_queues + 1)) - 1;
+	bcmgenet_rdma_writel(priv, ring_mask, DMA_RING_CFG);
 
-	/* Enable rings */
-	bcmgenet_rdma_writel(priv, ring_cfg, DMA_RING_CFG);
-
-	/* Configure ring as descriptor ring and re-enable DMA if enabled */
-	if (dma_enable)
-		dma_ctrl |= DMA_EN;
-	bcmgenet_rdma_writel(priv, dma_ctrl, DMA_CTRL);
+	/* Enable Rx rings */
+	ring_mask <<= DMA_RING_BUF_EN_SHIFT;
+	bcmgenet_rdma_writel(priv, ring_mask, DMA_CTRL);
 
 	return 0;
 }
@@ -2411,26 +3002,9 @@ static int bcmgenet_init_rx_queues(struct net_device *dev)
 static int bcmgenet_dma_teardown(struct bcmgenet_priv *priv)
 {
 	int ret = 0;
-	int timeout = 0;
-	u32 reg;
-	u32 dma_ctrl;
-	int i;
 
 	/* Disable TDMA to stop add more frames in TX DMA */
-	reg = bcmgenet_tdma_readl(priv, DMA_CTRL);
-	reg &= ~DMA_EN;
-	bcmgenet_tdma_writel(priv, reg, DMA_CTRL);
-
-	/* Check TDMA status register to confirm TDMA is disabled */
-	while (timeout++ < DMA_TIMEOUT_VAL) {
-		reg = bcmgenet_tdma_readl(priv, DMA_STATUS);
-		if (reg & DMA_DISABLED)
-			break;
-
-		udelay(1);
-	}
-
-	if (timeout == DMA_TIMEOUT_VAL) {
+	if (-ETIMEDOUT == bcmgenet_tdma_disable(priv)) {
 		netdev_warn(priv->dev, "Timed out while disabling TX DMA\n");
 		ret = -ETIMEDOUT;
 	}
@@ -2439,82 +3013,69 @@ static int bcmgenet_dma_teardown(struct bcmgenet_priv *priv)
 	usleep_range(10000, 20000);
 
 	/* Disable RDMA */
-	reg = bcmgenet_rdma_readl(priv, DMA_CTRL);
-	reg &= ~DMA_EN;
-	bcmgenet_rdma_writel(priv, reg, DMA_CTRL);
-
-	timeout = 0;
-	/* Check RDMA status register to confirm RDMA is disabled */
-	while (timeout++ < DMA_TIMEOUT_VAL) {
-		reg = bcmgenet_rdma_readl(priv, DMA_STATUS);
-		if (reg & DMA_DISABLED)
-			break;
-
-		udelay(1);
-	}
-
-	if (timeout == DMA_TIMEOUT_VAL) {
+	if (-ETIMEDOUT == bcmgenet_rdma_disable(priv)) {
 		netdev_warn(priv->dev, "Timed out while disabling RX DMA\n");
 		ret = -ETIMEDOUT;
 	}
 
-	dma_ctrl = 0;
-	for (i = 0; i < priv->hw_params->rx_queues; i++)
-		dma_ctrl |= (1 << (i + DMA_RING_BUF_EN_SHIFT));
-	reg = bcmgenet_rdma_readl(priv, DMA_CTRL);
-	reg &= ~dma_ctrl;
-	bcmgenet_rdma_writel(priv, reg, DMA_CTRL);
-
-	dma_ctrl = 0;
-	for (i = 0; i < priv->hw_params->tx_queues; i++)
-		dma_ctrl |= (1 << (i + DMA_RING_BUF_EN_SHIFT));
-	reg = bcmgenet_tdma_readl(priv, DMA_CTRL);
-	reg &= ~dma_ctrl;
-	bcmgenet_tdma_writel(priv, reg, DMA_CTRL);
-
 	return ret;
 }
 
 static void bcmgenet_fini_dma(struct bcmgenet_priv *priv)
 {
-	int i;
 	struct netdev_queue *txq;
+	int i;
 
 	bcmgenet_fini_rx_napi(priv);
 	bcmgenet_fini_tx_napi(priv);
 
-	/* disable DMA */
-	bcmgenet_dma_teardown(priv);
-
-	for (i = 0; i < priv->num_tx_bds; i++) {
-		if (priv->tx_cbs[i].skb != NULL) {
-			dev_kfree_skb(priv->tx_cbs[i].skb);
-			priv->tx_cbs[i].skb = NULL;
-		}
-	}
-
-	for (i = 0; i < priv->hw_params->tx_queues; i++) {
-		txq = netdev_get_tx_queue(priv->dev, priv->tx_rings[i].queue);
+	for (i = 0; i <= priv->hw_params->tx_queues; i++) {
+		txq = netdev_get_tx_queue(priv->dev, i);
 		netdev_tx_reset_queue(txq);
 	}
 
-	txq = netdev_get_tx_queue(priv->dev, priv->tx_rings[DESC_INDEX].queue);
-	netdev_tx_reset_queue(txq);
-
 	bcmgenet_free_rx_buffers(priv);
 	kfree(priv->rx_cbs);
 	kfree(priv->tx_cbs);
 }
 
 /* init_edma: Initialize DMA control register */
-static int bcmgenet_init_dma(struct bcmgenet_priv *priv)
+static int bcmgenet_init_dma(struct bcmgenet_priv *priv, bool flush_rx)
 {
-	int ret;
-	unsigned int i;
 	struct enet_cb *cb;
+	unsigned int i;
+	int ret;
+	u32 reg;
 
 	netif_dbg(priv, hw, priv->dev, "%s\n", __func__);
 
+	/* Disable TX DMA */
+	ret = bcmgenet_tdma_disable(priv);
+	if (ret) {
+		netdev_err(priv->dev, "failed to halt Tx DMA\n");
+		return ret;
+	}
+
+	/* Disable RX DMA */
+	ret = bcmgenet_rdma_disable(priv);
+	if (ret) {
+		netdev_err(priv->dev, "failed to halt Rx DMA\n");
+		return ret;
+	}
+
+	/* Flush TX queues */
+	bcmgenet_umac_writel(priv, 1, UMAC_TX_FLUSH);
+	udelay(10);
+	bcmgenet_umac_writel(priv, 0, UMAC_TX_FLUSH);
+
+	if (flush_rx) {
+		reg = bcmgenet_rbuf_ctrl_get(priv);
+		bcmgenet_rbuf_ctrl_set(priv, reg | BIT(0));
+		udelay(10);
+		bcmgenet_rbuf_ctrl_set(priv, reg);
+		udelay(10);
+	}
+
 	/* Initialize common Rx ring structures */
 	priv->rx_bds = priv->base + priv->hw_params->rdma_offset;
 	priv->num_rx_bds = TOTAL_DESC;
@@ -2544,7 +3105,8 @@ static int bcmgenet_init_dma(struct bcmgenet_priv *priv)
 	}
 
 	/* Init rDma */
-	bcmgenet_rdma_writel(priv, DMA_MAX_BURST_LENGTH, DMA_SCB_BURST_SIZE);
+	bcmgenet_rdma_writel(priv, priv->dma_max_burst_length,
+			     DMA_SCB_BURST_SIZE);
 
 	/* Initialize Rx queues */
 	ret = bcmgenet_init_rx_queues(priv->dev);
@@ -2557,42 +3119,51 @@ static int bcmgenet_init_dma(struct bcmgenet_priv *priv)
 	}
 
 	/* Init tDma */
-	bcmgenet_tdma_writel(priv, DMA_MAX_BURST_LENGTH, DMA_SCB_BURST_SIZE);
+	bcmgenet_tdma_writel(priv, priv->dma_max_burst_length,
+			     DMA_SCB_BURST_SIZE);
 
 	/* Initialize Tx queues */
 	bcmgenet_init_tx_queues(priv->dev);
 
+	/* Enable RX/TX DMA */
+	reg = bcmgenet_rdma_readl(priv, DMA_CTRL);
+	reg |= DMA_EN;
+	bcmgenet_rdma_writel(priv, reg, DMA_CTRL);
+
+	reg = bcmgenet_tdma_readl(priv, DMA_CTRL);
+	reg |= DMA_EN;
+	bcmgenet_tdma_writel(priv, reg, DMA_CTRL);
+
 	return 0;
 }
 
 /* Interrupt bottom half */
 static void bcmgenet_irq_task(struct work_struct *work)
 {
-	unsigned long flags;
 	unsigned int status;
 	struct bcmgenet_priv *priv = container_of(
 			work, struct bcmgenet_priv, bcmgenet_irq_work);
 
 	netif_dbg(priv, intr, priv->dev, "%s\n", __func__);
 
-	spin_lock_irqsave(&priv->lock, flags);
+	spin_lock_irq(&priv->lock);
 	status = priv->irq0_stat;
 	priv->irq0_stat = 0;
-	spin_unlock_irqrestore(&priv->lock, flags);
+	spin_unlock_irq(&priv->lock);
 
-	if (status & UMAC_IRQ_MPD_R) {
-		netif_dbg(priv, wol, priv->dev,
-			  "magic packet detected, waking up\n");
-		bcmgenet_power_up(priv, GENET_POWER_WOL_MAGIC);
+	if (status & UMAC_IRQ_PHY_DET_R &&
+	    priv->dev->phydev->autoneg != AUTONEG_ENABLE) {
+		phy_init_hw(priv->dev->phydev);
+		genphy_config_aneg(priv->dev->phydev);
 	}
 
 	/* Link UP/DOWN event */
 	if (status & UMAC_IRQ_LINK_EVENT)
-		phy_mac_interrupt(priv->phydev,
-				  !!(status & UMAC_IRQ_LINK_UP));
+		phy_mac_interrupt(priv->dev->phydev);
+
 }
 
-/* bcmgenet_isr1: handle Rx and Tx priority queues */
+/* bcmgenet_isr1: handle Rx and Tx queues */
 static irqreturn_t bcmgenet_isr1(int irq, void *dev_id)
 {
 	struct bcmgenet_priv *priv = dev_id;
@@ -2611,27 +3182,28 @@ static irqreturn_t bcmgenet_isr1(int irq, void *dev_id)
 		  "%s: IRQ=0x%x\n", __func__, status);
 
 	/* Check Rx priority queue interrupts */
-	for (index = 0; index < priv->hw_params->rx_queues; index++) {
+	for (index = 0; index <= priv->hw_params->rx_queues; index++) {
 		if (!(status & BIT(UMAC_IRQ1_RX_INTR_SHIFT + index)))
 			continue;
 
 		rx_ring = &priv->rx_rings[index];
+		rx_ring->dim.event_ctr++;
 
 		if (likely(napi_schedule_prep(&rx_ring->napi))) {
-			rx_ring->int_disable(rx_ring);
+			bcmgenet_rx_ring_int_disable(rx_ring);
 			__napi_schedule_irqoff(&rx_ring->napi);
 		}
 	}
 
 	/* Check Tx priority queue interrupts */
-	for (index = 0; index < priv->hw_params->tx_queues; index++) {
+	for (index = 0; index <= priv->hw_params->tx_queues; index++) {
 		if (!(status & BIT(index)))
 			continue;
 
 		tx_ring = &priv->tx_rings[index];
 
 		if (likely(napi_schedule_prep(&tx_ring->napi))) {
-			tx_ring->int_disable(tx_ring);
+			bcmgenet_tx_ring_int_disable(tx_ring);
 			__napi_schedule_irqoff(&tx_ring->napi);
 		}
 	}
@@ -2639,12 +3211,10 @@ static irqreturn_t bcmgenet_isr1(int irq, void *dev_id)
 	return IRQ_HANDLED;
 }
 
-/* bcmgenet_isr0: handle Rx and Tx default queues + other stuff */
+/* bcmgenet_isr0: handle other stuff */
 static irqreturn_t bcmgenet_isr0(int irq, void *dev_id)
 {
 	struct bcmgenet_priv *priv = dev_id;
-	struct bcmgenet_rx_ring *rx_ring;
-	struct bcmgenet_tx_ring *tx_ring;
 	unsigned int status;
 	unsigned long flags;
 
@@ -2658,41 +3228,11 @@ static irqreturn_t bcmgenet_isr0(int irq, void *dev_id)
 	netif_dbg(priv, intr, priv->dev,
 		  "IRQ=0x%x\n", status);
 
-	if (status & UMAC_IRQ_RXDMA_DONE) {
-		rx_ring = &priv->rx_rings[DESC_INDEX];
-
-		if (likely(napi_schedule_prep(&rx_ring->napi))) {
-			rx_ring->int_disable(rx_ring);
-			__napi_schedule_irqoff(&rx_ring->napi);
-		}
-	}
-
-	if (status & UMAC_IRQ_TXDMA_DONE) {
-		tx_ring = &priv->tx_rings[DESC_INDEX];
-
-		if (likely(napi_schedule_prep(&tx_ring->napi))) {
-			tx_ring->int_disable(tx_ring);
-			__napi_schedule_irqoff(&tx_ring->napi);
-		}
-	}
-
-	if (priv->irq0_stat & (UMAC_IRQ_PHY_DET_R |
-				UMAC_IRQ_PHY_DET_F |
-				UMAC_IRQ_LINK_EVENT |
-				UMAC_IRQ_HFB_SM |
-				UMAC_IRQ_HFB_MM)) {
-		/* all other interested interrupts handled in bottom half */
-		schedule_work(&priv->bcmgenet_irq_work);
-	}
-
-	if ((priv->hw_params->flags & GENET_HAS_MDIO_INTR) &&
-		status & (UMAC_IRQ_MDIO_DONE | UMAC_IRQ_MDIO_ERROR)) {
+	if (bcmgenet_has_mdio_intr(priv) && status & UMAC_IRQ_MDIO_EVENT)
 		wake_up(&priv->wq);
-	}
 
 	/* all other interested interrupts handled in bottom half */
-	status &= (UMAC_IRQ_LINK_EVENT |
-		   UMAC_IRQ_MPD_R);
+	status &= (UMAC_IRQ_LINK_EVENT | UMAC_IRQ_PHY_DET_R);
 	if (status) {
 		/* Save irq status for bottom-half processing. */
 		spin_lock_irqsave(&priv->lock, flags);
@@ -2707,30 +3247,10 @@ static irqreturn_t bcmgenet_isr0(int irq, void *dev_id)
 
 static irqreturn_t bcmgenet_wol_isr(int irq, void *dev_id)
 {
-	struct bcmgenet_priv *priv = dev_id;
-
-	pm_wakeup_event(&priv->pdev->dev, 0);
-
+	/* Acknowledge the interrupt */
 	return IRQ_HANDLED;
 }
 
-#ifdef CONFIG_NET_POLL_CONTROLLER
-static void bcmgenet_poll_controller(struct net_device *dev)
-{
-	struct bcmgenet_priv *priv = netdev_priv(dev);
-
-	/* Invoke the main RX/TX interrupt handler */
-	disable_irq(priv->irq0);
-	bcmgenet_isr0(priv->irq0, priv);
-	enable_irq(priv->irq0);
-
-	/* And the interrupt handler for RX/TX priority queues */
-	disable_irq(priv->irq1);
-	bcmgenet_isr1(priv->irq1, priv);
-	enable_irq(priv->irq1);
-}
-#endif
-
 static void bcmgenet_umac_reset(struct bcmgenet_priv *priv)
 {
 	u32 reg;
@@ -2746,79 +3266,21 @@ static void bcmgenet_umac_reset(struct bcmgenet_priv *priv)
 }
 
 static void bcmgenet_set_hw_addr(struct bcmgenet_priv *priv,
-				 unsigned char *addr)
-{
-	bcmgenet_umac_writel(priv, (addr[0] << 24) | (addr[1] << 16) |
-			(addr[2] << 8) | addr[3], UMAC_MAC0);
-	bcmgenet_umac_writel(priv, (addr[4] << 8) | addr[5], UMAC_MAC1);
-}
-
-/* Returns a reusable dma control register value */
-static u32 bcmgenet_dma_disable(struct bcmgenet_priv *priv)
-{
-	u32 reg;
-	u32 dma_ctrl;
-
-	/* disable DMA */
-	dma_ctrl = 1 << (DESC_INDEX + DMA_RING_BUF_EN_SHIFT) | DMA_EN;
-	reg = bcmgenet_tdma_readl(priv, DMA_CTRL);
-	reg &= ~dma_ctrl;
-	bcmgenet_tdma_writel(priv, reg, DMA_CTRL);
-
-	reg = bcmgenet_rdma_readl(priv, DMA_CTRL);
-	reg &= ~dma_ctrl;
-	bcmgenet_rdma_writel(priv, reg, DMA_CTRL);
-
-	bcmgenet_umac_writel(priv, 1, UMAC_TX_FLUSH);
-	udelay(10);
-	bcmgenet_umac_writel(priv, 0, UMAC_TX_FLUSH);
-
-	return dma_ctrl;
-}
-
-static void bcmgenet_enable_dma(struct bcmgenet_priv *priv, u32 dma_ctrl)
+				 const unsigned char *addr)
 {
-	u32 reg;
-
-	reg = bcmgenet_rdma_readl(priv, DMA_CTRL);
-	reg |= dma_ctrl;
-	bcmgenet_rdma_writel(priv, reg, DMA_CTRL);
-
-	reg = bcmgenet_tdma_readl(priv, DMA_CTRL);
-	reg |= dma_ctrl;
-	bcmgenet_tdma_writel(priv, reg, DMA_CTRL);
+	bcmgenet_umac_writel(priv, get_unaligned_be32(&addr[0]), UMAC_MAC0);
+	bcmgenet_umac_writel(priv, get_unaligned_be16(&addr[4]), UMAC_MAC1);
 }
 
-/* bcmgenet_hfb_clear
- *
- * Clear Hardware Filter Block and disable all filtering.
- */
-static void bcmgenet_hfb_clear(struct bcmgenet_priv *priv)
-{
-	u32 i;
-
-	bcmgenet_hfb_reg_writel(priv, 0x0, HFB_CTRL);
-	bcmgenet_hfb_reg_writel(priv, 0x0, HFB_FLT_ENABLE_V3PLUS);
-	bcmgenet_hfb_reg_writel(priv, 0x0, HFB_FLT_ENABLE_V3PLUS + 4);
-
-	for (i = DMA_INDEX2RING_0; i <= DMA_INDEX2RING_7; i++)
-		bcmgenet_rdma_writel(priv, 0x0, i);
-
-	for (i = 0; i < (priv->hw_params->hfb_filter_cnt / 4); i++)
-		bcmgenet_hfb_reg_writel(priv, 0x0,
-					HFB_FLT_LEN_V3PLUS + i * sizeof(u32));
-
-	for (i = 0; i < priv->hw_params->hfb_filter_cnt *
-			priv->hw_params->hfb_filter_size; i++)
-		bcmgenet_hfb_writel(priv, 0x0, i * sizeof(u32));
-}
-
-static void bcmgenet_hfb_init(struct bcmgenet_priv *priv)
+static void bcmgenet_get_hw_addr(struct bcmgenet_priv *priv,
+				 unsigned char *addr)
 {
-	if (GENET_IS_V1(priv) || GENET_IS_V2(priv))
-		return;
+	u32 addr_tmp;
 
-	bcmgenet_hfb_clear(priv);
+	addr_tmp = bcmgenet_umac_readl(priv, UMAC_MAC0);
+	put_unaligned_be32(addr_tmp, &addr[0]);
+	addr_tmp = bcmgenet_umac_readl(priv, UMAC_MAC1);
+	put_unaligned_be16(addr_tmp, &addr[4]);
 }
 
 static void bcmgenet_netif_start(struct net_device *dev)
@@ -2826,24 +3288,24 @@ static void bcmgenet_netif_start(struct net_device *dev)
 	struct bcmgenet_priv *priv = netdev_priv(dev);
 
 	/* Start the network engine */
+	netif_addr_lock_bh(dev);
+	bcmgenet_set_rx_mode(dev);
+	netif_addr_unlock_bh(dev);
 	bcmgenet_enable_rx_napi(priv);
-	bcmgenet_enable_tx_napi(priv);
 
 	umac_enable_set(priv, CMD_TX_EN | CMD_RX_EN, true);
 
-	netif_tx_start_all_queues(dev);
+	bcmgenet_enable_tx_napi(priv);
 
 	/* Monitor link interrupts now */
 	bcmgenet_link_intr_enable(priv);
 
-	phy_start(priv->phydev);
+	phy_start(dev->phydev);
 }
 
 static int bcmgenet_open(struct net_device *dev)
 {
 	struct bcmgenet_priv *priv = netdev_priv(dev);
-	unsigned long dma_ctrl;
-	u32 reg;
 	int ret;
 
 	netif_dbg(priv, ifup, dev, "bcmgenet_open\n");
@@ -2860,41 +3322,25 @@ static int bcmgenet_open(struct net_device *dev)
 	/* take MAC out of reset */
 	bcmgenet_umac_reset(priv);
 
-	ret = init_umac(priv);
-	if (ret)
-		goto err_clk_disable;
-
-	/* disable ethernet MAC while updating its registers */
-	umac_enable_set(priv, CMD_TX_EN | CMD_RX_EN, false);
+	init_umac(priv);
 
-	/* Make sure we reflect the value of CRC_CMD_FWD */
-	reg = bcmgenet_umac_readl(priv, UMAC_CMD);
-	priv->crc_fwd_en = !!(reg & CMD_CRC_FWD);
+	/* Apply features again in case we changed them while interface was
+	 * down
+	 */
+	bcmgenet_set_features(dev, dev->features);
 
 	bcmgenet_set_hw_addr(priv, dev->dev_addr);
 
-	if (priv->internal_phy) {
-		reg = bcmgenet_ext_readl(priv, EXT_EXT_PWR_MGMT);
-		reg |= EXT_ENERGY_DET_MASK;
-		bcmgenet_ext_writel(priv, reg, EXT_EXT_PWR_MGMT);
-	}
-
-	/* Disable RX/TX DMA and flush TX queues */
-	dma_ctrl = bcmgenet_dma_disable(priv);
+	/* HFB init */
+	bcmgenet_hfb_init(priv);
 
 	/* Reinitialize TDMA and RDMA and SW housekeeping */
-	ret = bcmgenet_init_dma(priv);
+	ret = bcmgenet_init_dma(priv, true);
 	if (ret) {
 		netdev_err(dev, "failed to initialize DMA\n");
 		goto err_clk_disable;
 	}
 
-	/* Always enable ring 16 - descriptor ring */
-	bcmgenet_enable_dma(priv, dma_ctrl);
-
-	/* HFB init */
-	bcmgenet_hfb_init(priv);
-
 	ret = request_irq(priv->irq0, bcmgenet_isr0, IRQF_SHARED,
 			  dev->name, priv);
 	if (ret < 0) {
@@ -2915,8 +3361,12 @@ static int bcmgenet_open(struct net_device *dev)
 		goto err_irq1;
 	}
 
+	bcmgenet_phy_pause_set(dev, priv->rx_pause, priv->tx_pause);
+
 	bcmgenet_netif_start(dev);
 
+	netif_tx_start_all_queues(dev);
+
 	return 0;
 
 err_irq1:
@@ -2924,6 +3374,7 @@ err_irq1:
 err_irq0:
 	free_irq(priv->irq0, priv);
 err_fini_dma:
+	bcmgenet_dma_teardown(priv);
 	bcmgenet_fini_dma(priv);
 err_clk_disable:
 	if (priv->internal_phy)
@@ -2932,52 +3383,49 @@ err_clk_disable:
 	return ret;
 }
 
-static void bcmgenet_netif_stop(struct net_device *dev)
+static void bcmgenet_netif_stop(struct net_device *dev, bool stop_phy)
 {
 	struct bcmgenet_priv *priv = netdev_priv(dev);
 
-	netif_tx_stop_all_queues(dev);
-	phy_stop(priv->phydev);
-	bcmgenet_intr_disable(priv);
-	bcmgenet_disable_rx_napi(priv);
+	netif_tx_disable(dev);
+
+	/* Disable MAC receive */
+	bcmgenet_hfb_reg_writel(priv, 0, HFB_CTRL);
+	umac_enable_set(priv, CMD_RX_EN, false);
+
+	if (stop_phy)
+		phy_stop(dev->phydev);
+
+	bcmgenet_dma_teardown(priv);
+
+	/* Disable MAC transmit. TX DMA disabled must be done before this */
+	umac_enable_set(priv, CMD_TX_EN, false);
+
 	bcmgenet_disable_tx_napi(priv);
+	bcmgenet_disable_rx_napi(priv);
+	bcmgenet_intr_disable(priv);
 
 	/* Wait for pending work items to complete. Since interrupts are
 	 * disabled no new work will be scheduled.
 	 */
 	cancel_work_sync(&priv->bcmgenet_irq_work);
 
-	priv->old_link = -1;
-	priv->old_speed = -1;
-	priv->old_duplex = -1;
-	priv->old_pause = -1;
+	/* tx reclaim */
+	bcmgenet_tx_reclaim_all(dev);
+	bcmgenet_fini_dma(priv);
 }
 
 static int bcmgenet_close(struct net_device *dev)
 {
 	struct bcmgenet_priv *priv = netdev_priv(dev);
-	int ret;
+	int ret = 0;
 
 	netif_dbg(priv, ifdown, dev, "bcmgenet_close\n");
 
-	bcmgenet_netif_stop(dev);
+	bcmgenet_netif_stop(dev, false);
 
 	/* Really kill the PHY state machine and disconnect from it */
-	phy_disconnect(priv->phydev);
-
-	/* Disable MAC receive */
-	umac_enable_set(priv, CMD_RX_EN, false);
-
-	ret = bcmgenet_dma_teardown(priv);
-	if (ret)
-		return ret;
-
-	/* Disable MAC transmit. TX DMA disabled must be done before this */
-	umac_enable_set(priv, CMD_TX_EN, false);
-
-	/* tx reclaim */
-	bcmgenet_tx_reclaim_all(dev);
-	bcmgenet_fini_dma(priv);
+	phy_disconnect(dev->phydev);
 
 	free_irq(priv->irq0, priv);
 	free_irq(priv->irq1, priv);
@@ -2996,27 +3444,21 @@ static void bcmgenet_dump_tx_queue(struct bcmgenet_tx_ring *ring)
 	u32 p_index, c_index, intsts, intmsk;
 	struct netdev_queue *txq;
 	unsigned int free_bds;
-	unsigned long flags;
 	bool txq_stopped;
 
 	if (!netif_msg_tx_err(priv))
 		return;
 
-	txq = netdev_get_tx_queue(priv->dev, ring->queue);
+	txq = netdev_get_tx_queue(priv->dev, ring->index);
 
-	spin_lock_irqsave(&ring->lock, flags);
-	if (ring->index == DESC_INDEX) {
-		intsts = ~bcmgenet_intrl2_0_readl(priv, INTRL2_CPU_MASK_STATUS);
-		intmsk = UMAC_IRQ_TXDMA_DONE | UMAC_IRQ_TXDMA_MBDONE;
-	} else {
-		intsts = ~bcmgenet_intrl2_1_readl(priv, INTRL2_CPU_MASK_STATUS);
-		intmsk = 1 << ring->index;
-	}
+	spin_lock(&ring->lock);
+	intsts = ~bcmgenet_intrl2_1_readl(priv, INTRL2_CPU_MASK_STATUS);
+	intmsk = 1 << ring->index;
 	c_index = bcmgenet_tdma_ring_readl(priv, ring->index, TDMA_CONS_INDEX);
 	p_index = bcmgenet_tdma_ring_readl(priv, ring->index, TDMA_PROD_INDEX);
 	txq_stopped = netif_tx_queue_stopped(txq);
 	free_bds = ring->free_bds;
-	spin_unlock_irqrestore(&ring->lock, flags);
+	spin_unlock(&ring->lock);
 
 	netif_err(priv, tx_err, priv->dev, "Ring %d queue %d status summary\n"
 		  "TX queue status: %s, interrupts: %s\n"
@@ -3025,7 +3467,7 @@ static void bcmgenet_dump_tx_queue(struct bcmgenet_tx_ring *ring)
 		  "(sw)c_index: %d (hw)c_index: %d\n"
 		  "(sw)clean_p: %d (sw)write_p: %d\n"
 		  "(sw)cb_ptr: %d (sw)end_ptr: %d\n",
-		  ring->index, ring->queue,
+		  ring->index, ring->index,
 		  txq_stopped ? "stopped" : "active",
 		  intsts & intmsk ? "enabled" : "disabled",
 		  free_bds, ring->size,
@@ -3035,105 +3477,98 @@ static void bcmgenet_dump_tx_queue(struct bcmgenet_tx_ring *ring)
 		  ring->cb_ptr, ring->end_ptr);
 }
 
-static void bcmgenet_timeout(struct net_device *dev)
+static void bcmgenet_timeout(struct net_device *dev, unsigned int txqueue)
 {
 	struct bcmgenet_priv *priv = netdev_priv(dev);
-	u32 int0_enable = 0;
 	u32 int1_enable = 0;
 	unsigned int q;
 
 	netif_dbg(priv, tx_err, dev, "bcmgenet_timeout\n");
 
-	for (q = 0; q < priv->hw_params->tx_queues; q++)
+	for (q = 0; q <= priv->hw_params->tx_queues; q++)
 		bcmgenet_dump_tx_queue(&priv->tx_rings[q]);
-	bcmgenet_dump_tx_queue(&priv->tx_rings[DESC_INDEX]);
 
 	bcmgenet_tx_reclaim_all(dev);
 
-	for (q = 0; q < priv->hw_params->tx_queues; q++)
+	for (q = 0; q <= priv->hw_params->tx_queues; q++)
 		int1_enable |= (1 << q);
 
-	int0_enable = UMAC_IRQ_TXDMA_DONE;
-
 	/* Re-enable TX interrupts if disabled */
-	bcmgenet_intrl2_0_writel(priv, int0_enable, INTRL2_CPU_MASK_CLEAR);
 	bcmgenet_intrl2_1_writel(priv, int1_enable, INTRL2_CPU_MASK_CLEAR);
 
 	netif_trans_update(dev);
 
-	dev->stats.tx_errors++;
+	BCMGENET_STATS64_INC((&priv->tx_rings[txqueue].stats64), errors);
 
 	netif_tx_wake_all_queues(dev);
 }
 
-#define MAX_MC_COUNT	16
+#define MAX_MDF_FILTER	17
 
 static inline void bcmgenet_set_mdf_addr(struct bcmgenet_priv *priv,
-					 unsigned char *addr,
-					 int *i,
-					 int *mc)
+					 const unsigned char *addr,
+					 int *i)
 {
-	u32 reg;
-
 	bcmgenet_umac_writel(priv, addr[0] << 8 | addr[1],
 			     UMAC_MDF_ADDR + (*i * 4));
 	bcmgenet_umac_writel(priv, addr[2] << 24 | addr[3] << 16 |
 			     addr[4] << 8 | addr[5],
 			     UMAC_MDF_ADDR + ((*i + 1) * 4));
-	reg = bcmgenet_umac_readl(priv, UMAC_MDF_CTRL);
-	reg |= (1 << (MAX_MC_COUNT - *mc));
-	bcmgenet_umac_writel(priv, reg, UMAC_MDF_CTRL);
 	*i += 2;
-	(*mc)++;
 }
 
 static void bcmgenet_set_rx_mode(struct net_device *dev)
 {
 	struct bcmgenet_priv *priv = netdev_priv(dev);
 	struct netdev_hw_addr *ha;
-	int i, mc;
+	int i, nfilter;
 	u32 reg;
 
 	netif_dbg(priv, hw, dev, "%s: %08X\n", __func__, dev->flags);
 
-	/* Promiscuous mode */
+	/* Number of filters needed */
+	nfilter = netdev_uc_count(dev) + netdev_mc_count(dev) + 2;
+
+	/*
+	 * Turn on promicuous mode for three scenarios
+	 * 1. IFF_PROMISC flag is set
+	 * 2. IFF_ALLMULTI flag is set
+	 * 3. The number of filters needed exceeds the number filters
+	 *    supported by the hardware.
+	*/
+	spin_lock(&priv->reg_lock);
 	reg = bcmgenet_umac_readl(priv, UMAC_CMD);
-	if (dev->flags & IFF_PROMISC) {
+	if ((dev->flags & (IFF_PROMISC | IFF_ALLMULTI)) ||
+	    (nfilter > MAX_MDF_FILTER)) {
 		reg |= CMD_PROMISC;
 		bcmgenet_umac_writel(priv, reg, UMAC_CMD);
+		spin_unlock(&priv->reg_lock);
 		bcmgenet_umac_writel(priv, 0, UMAC_MDF_CTRL);
 		return;
 	} else {
 		reg &= ~CMD_PROMISC;
 		bcmgenet_umac_writel(priv, reg, UMAC_CMD);
-	}
-
-	/* UniMac doesn't support ALLMULTI */
-	if (dev->flags & IFF_ALLMULTI) {
-		netdev_warn(dev, "ALLMULTI is not supported\n");
-		return;
+		spin_unlock(&priv->reg_lock);
 	}
 
 	/* update MDF filter */
 	i = 0;
-	mc = 0;
 	/* Broadcast */
-	bcmgenet_set_mdf_addr(priv, dev->broadcast, &i, &mc);
+	bcmgenet_set_mdf_addr(priv, dev->broadcast, &i);
 	/* my own address.*/
-	bcmgenet_set_mdf_addr(priv, dev->dev_addr, &i, &mc);
-	/* Unicast list*/
-	if (netdev_uc_count(dev) > (MAX_MC_COUNT - mc))
-		return;
+	bcmgenet_set_mdf_addr(priv, dev->dev_addr, &i);
 
-	if (!netdev_uc_empty(dev))
-		netdev_for_each_uc_addr(ha, dev)
-			bcmgenet_set_mdf_addr(priv, ha->addr, &i, &mc);
-	/* Multicast */
-	if (netdev_mc_empty(dev) || netdev_mc_count(dev) >= (MAX_MC_COUNT - mc))
-		return;
+	/* Unicast */
+	netdev_for_each_uc_addr(ha, dev)
+		bcmgenet_set_mdf_addr(priv, ha->addr, &i);
 
+	/* Multicast */
 	netdev_for_each_mc_addr(ha, dev)
-		bcmgenet_set_mdf_addr(priv, ha->addr, &i, &mc);
+		bcmgenet_set_mdf_addr(priv, ha->addr, &i);
+
+	/* Enable filters */
+	reg = GENMASK(MAX_MDF_FILTER - 1, MAX_MDF_FILTER - nfilter);
+	bcmgenet_umac_writel(priv, reg, UMAC_MDF_CTRL);
 }
 
 /* Set the hardware MAC address. */
@@ -3147,51 +3582,93 @@ static int bcmgenet_set_mac_addr(struct net_device *dev, void *p)
 	if (netif_running(dev))
 		return -EBUSY;
 
-	ether_addr_copy(dev->dev_addr, addr->sa_data);
+	eth_hw_addr_set(dev, addr->sa_data);
 
 	return 0;
 }
 
-static struct net_device_stats *bcmgenet_get_stats(struct net_device *dev)
+static void bcmgenet_get_stats64(struct net_device *dev,
+				 struct rtnl_link_stats64 *stats)
 {
 	struct bcmgenet_priv *priv = netdev_priv(dev);
-	unsigned long tx_bytes = 0, tx_packets = 0;
-	unsigned long rx_bytes = 0, rx_packets = 0;
-	unsigned long rx_errors = 0, rx_dropped = 0;
-	struct bcmgenet_tx_ring *tx_ring;
-	struct bcmgenet_rx_ring *rx_ring;
+	struct bcmgenet_tx_stats64 *tx_stats;
+	struct bcmgenet_rx_stats64 *rx_stats;
+	u64 rx_length_errors, rx_over_errors;
+	u64 rx_missed, rx_fragmented_errors;
+	u64 rx_crc_errors, rx_frame_errors;
+	u64 tx_errors, tx_dropped;
+	u64 rx_errors, rx_dropped;
+	u64 tx_bytes, tx_packets;
+	u64 rx_bytes, rx_packets;
+	unsigned int start;
 	unsigned int q;
+	u64 multicast;
+
+	for (q = 0; q <= priv->hw_params->tx_queues; q++) {
+		tx_stats = &priv->tx_rings[q].stats64;
+		do {
+			start = u64_stats_fetch_begin(&tx_stats->syncp);
+			tx_bytes = u64_stats_read(&tx_stats->bytes);
+			tx_packets = u64_stats_read(&tx_stats->packets);
+			tx_errors = u64_stats_read(&tx_stats->errors);
+			tx_dropped = u64_stats_read(&tx_stats->dropped);
+		} while (u64_stats_fetch_retry(&tx_stats->syncp, start));
+
+		stats->tx_bytes += tx_bytes;
+		stats->tx_packets += tx_packets;
+		stats->tx_errors += tx_errors;
+		stats->tx_dropped += tx_dropped;
+	}
 
-	for (q = 0; q < priv->hw_params->tx_queues; q++) {
-		tx_ring = &priv->tx_rings[q];
-		tx_bytes += tx_ring->bytes;
-		tx_packets += tx_ring->packets;
+	for (q = 0; q <= priv->hw_params->rx_queues; q++) {
+		rx_stats = &priv->rx_rings[q].stats64;
+		do {
+			start = u64_stats_fetch_begin(&rx_stats->syncp);
+			rx_bytes = u64_stats_read(&rx_stats->bytes);
+			rx_packets = u64_stats_read(&rx_stats->packets);
+			rx_errors = u64_stats_read(&rx_stats->errors);
+			rx_dropped = u64_stats_read(&rx_stats->dropped);
+			rx_missed = u64_stats_read(&rx_stats->missed);
+			rx_length_errors = u64_stats_read(&rx_stats->length_errors);
+			rx_over_errors = u64_stats_read(&rx_stats->over_errors);
+			rx_crc_errors = u64_stats_read(&rx_stats->crc_errors);
+			rx_frame_errors = u64_stats_read(&rx_stats->frame_errors);
+			rx_fragmented_errors = u64_stats_read(&rx_stats->fragmented_errors);
+			multicast = u64_stats_read(&rx_stats->multicast);
+		} while (u64_stats_fetch_retry(&rx_stats->syncp, start));
+
+		rx_errors += rx_length_errors;
+		rx_errors += rx_crc_errors;
+		rx_errors += rx_frame_errors;
+		rx_errors += rx_fragmented_errors;
+
+		stats->rx_bytes += rx_bytes;
+		stats->rx_packets += rx_packets;
+		stats->rx_errors += rx_errors;
+		stats->rx_dropped += rx_dropped;
+		stats->rx_missed_errors += rx_missed;
+		stats->rx_length_errors += rx_length_errors;
+		stats->rx_over_errors += rx_over_errors;
+		stats->rx_crc_errors += rx_crc_errors;
+		stats->rx_frame_errors += rx_frame_errors;
+		stats->multicast += multicast;
 	}
-	tx_ring = &priv->tx_rings[DESC_INDEX];
-	tx_bytes += tx_ring->bytes;
-	tx_packets += tx_ring->packets;
+}
 
-	for (q = 0; q < priv->hw_params->rx_queues; q++) {
-		rx_ring = &priv->rx_rings[q];
+static int bcmgenet_change_carrier(struct net_device *dev, bool new_carrier)
+{
+	struct bcmgenet_priv *priv = netdev_priv(dev);
 
-		rx_bytes += rx_ring->bytes;
-		rx_packets += rx_ring->packets;
-		rx_errors += rx_ring->errors;
-		rx_dropped += rx_ring->dropped;
-	}
-	rx_ring = &priv->rx_rings[DESC_INDEX];
-	rx_bytes += rx_ring->bytes;
-	rx_packets += rx_ring->packets;
-	rx_errors += rx_ring->errors;
-	rx_dropped += rx_ring->dropped;
+	if (!dev->phydev || !phy_is_pseudo_fixed_link(dev->phydev) ||
+	    priv->phy_interface != PHY_INTERFACE_MODE_MOCA)
+		return -EOPNOTSUPP;
+
+	if (new_carrier)
+		netif_carrier_on(dev);
+	else
+		netif_carrier_off(dev);
 
-	dev->stats.tx_bytes = tx_bytes;
-	dev->stats.tx_packets = tx_packets;
-	dev->stats.rx_bytes = rx_bytes;
-	dev->stats.rx_packets = rx_packets;
-	dev->stats.rx_errors = rx_errors;
-	dev->stats.rx_missed_errors = rx_errors;
-	return &dev->stats;
+	return 0;
 }
 
 static const struct net_device_ops bcmgenet_netdev_ops = {
@@ -3201,140 +3678,115 @@ static const struct net_device_ops bcmgenet_netdev_ops = {
 	.ndo_tx_timeout		= bcmgenet_timeout,
 	.ndo_set_rx_mode	= bcmgenet_set_rx_mode,
 	.ndo_set_mac_address	= bcmgenet_set_mac_addr,
-	.ndo_do_ioctl		= bcmgenet_ioctl,
+	.ndo_eth_ioctl		= phy_do_ioctl_running,
 	.ndo_set_features	= bcmgenet_set_features,
-#ifdef CONFIG_NET_POLL_CONTROLLER
-	.ndo_poll_controller	= bcmgenet_poll_controller,
-#endif
-	.ndo_get_stats		= bcmgenet_get_stats,
+	.ndo_get_stats64	= bcmgenet_get_stats64,
+	.ndo_change_carrier	= bcmgenet_change_carrier,
 };
 
-/* Array of GENET hardware parameters/characteristics */
-static struct bcmgenet_hw_params bcmgenet_hw_params[] = {
-	[GENET_V1] = {
-		.tx_queues = 0,
-		.tx_bds_per_q = 0,
-		.rx_queues = 0,
-		.rx_bds_per_q = 0,
-		.bp_in_en_shift = 16,
-		.bp_in_mask = 0xffff,
-		.hfb_filter_cnt = 16,
-		.qtag_mask = 0x1F,
-		.hfb_offset = 0x1000,
-		.rdma_offset = 0x2000,
-		.tdma_offset = 0x3000,
-		.words_per_bd = 2,
-	},
-	[GENET_V2] = {
-		.tx_queues = 4,
-		.tx_bds_per_q = 32,
-		.rx_queues = 0,
-		.rx_bds_per_q = 0,
-		.bp_in_en_shift = 16,
-		.bp_in_mask = 0xffff,
-		.hfb_filter_cnt = 16,
-		.qtag_mask = 0x1F,
-		.tbuf_offset = 0x0600,
-		.hfb_offset = 0x1000,
-		.hfb_reg_offset = 0x2000,
-		.rdma_offset = 0x3000,
-		.tdma_offset = 0x4000,
-		.words_per_bd = 2,
-		.flags = GENET_HAS_EXT,
-	},
-	[GENET_V3] = {
-		.tx_queues = 4,
-		.tx_bds_per_q = 32,
-		.rx_queues = 0,
-		.rx_bds_per_q = 0,
-		.bp_in_en_shift = 17,
-		.bp_in_mask = 0x1ffff,
-		.hfb_filter_cnt = 48,
-		.hfb_filter_size = 128,
-		.qtag_mask = 0x3F,
-		.tbuf_offset = 0x0600,
-		.hfb_offset = 0x8000,
-		.hfb_reg_offset = 0xfc00,
-		.rdma_offset = 0x10000,
-		.tdma_offset = 0x11000,
-		.words_per_bd = 2,
-		.flags = GENET_HAS_EXT | GENET_HAS_MDIO_INTR |
-			 GENET_HAS_MOCA_LINK_DET,
-	},
-	[GENET_V4] = {
-		.tx_queues = 4,
-		.tx_bds_per_q = 32,
-		.rx_queues = 0,
-		.rx_bds_per_q = 0,
-		.bp_in_en_shift = 17,
-		.bp_in_mask = 0x1ffff,
-		.hfb_filter_cnt = 48,
-		.hfb_filter_size = 128,
-		.qtag_mask = 0x3F,
-		.tbuf_offset = 0x0600,
-		.hfb_offset = 0x8000,
-		.hfb_reg_offset = 0xfc00,
-		.rdma_offset = 0x2000,
-		.tdma_offset = 0x4000,
-		.words_per_bd = 3,
-		.flags = GENET_HAS_40BITS | GENET_HAS_EXT |
-			 GENET_HAS_MDIO_INTR | GENET_HAS_MOCA_LINK_DET,
-	},
-	[GENET_V5] = {
-		.tx_queues = 4,
-		.tx_bds_per_q = 32,
-		.rx_queues = 0,
-		.rx_bds_per_q = 0,
-		.bp_in_en_shift = 17,
-		.bp_in_mask = 0x1ffff,
-		.hfb_filter_cnt = 48,
-		.hfb_filter_size = 128,
-		.qtag_mask = 0x3F,
-		.tbuf_offset = 0x0600,
-		.hfb_offset = 0x8000,
-		.hfb_reg_offset = 0xfc00,
-		.rdma_offset = 0x2000,
-		.tdma_offset = 0x4000,
-		.words_per_bd = 3,
-		.flags = GENET_HAS_40BITS | GENET_HAS_EXT |
-			 GENET_HAS_MDIO_INTR | GENET_HAS_MOCA_LINK_DET,
-	},
+/* GENET hardware parameters/characteristics */
+static const struct bcmgenet_hw_params bcmgenet_hw_params_v1 = {
+	.tx_queues = 0,
+	.tx_bds_per_q = 0,
+	.rx_queues = 0,
+	.rx_bds_per_q = 0,
+	.bp_in_en_shift = 16,
+	.bp_in_mask = 0xffff,
+	.hfb_filter_cnt = 16,
+	.hfb_filter_size = 64,
+	.qtag_mask = 0x1F,
+	.hfb_offset = 0x1000,
+	.hfb_reg_offset = GENET_RBUF_OFF + RBUF_HFB_CTRL_V1,
+	.rdma_offset = 0x2000,
+	.tdma_offset = 0x3000,
+	.words_per_bd = 2,
+};
+
+static const struct bcmgenet_hw_params bcmgenet_hw_params_v2 = {
+	.tx_queues = 4,
+	.tx_bds_per_q = 32,
+	.rx_queues = 0,
+	.rx_bds_per_q = 0,
+	.bp_in_en_shift = 16,
+	.bp_in_mask = 0xffff,
+	.hfb_filter_cnt = 16,
+	.hfb_filter_size = 64,
+	.qtag_mask = 0x1F,
+	.tbuf_offset = 0x0600,
+	.hfb_offset = 0x1000,
+	.hfb_reg_offset = 0x2000,
+	.rdma_offset = 0x3000,
+	.tdma_offset = 0x4000,
+	.words_per_bd = 2,
+};
+
+static const struct bcmgenet_hw_params bcmgenet_hw_params_v3 = {
+	.tx_queues = 4,
+	.tx_bds_per_q = 32,
+	.rx_queues = 0,
+	.rx_bds_per_q = 0,
+	.bp_in_en_shift = 17,
+	.bp_in_mask = 0x1ffff,
+	.hfb_filter_cnt = 48,
+	.hfb_filter_size = 128,
+	.qtag_mask = 0x3F,
+	.tbuf_offset = 0x0600,
+	.hfb_offset = 0x8000,
+	.hfb_reg_offset = 0xfc00,
+	.rdma_offset = 0x10000,
+	.tdma_offset = 0x11000,
+	.words_per_bd = 2,
+};
+
+static const struct bcmgenet_hw_params bcmgenet_hw_params_v4 = {
+	.tx_queues = 4,
+	.tx_bds_per_q = 32,
+	.rx_queues = 0,
+	.rx_bds_per_q = 0,
+	.bp_in_en_shift = 17,
+	.bp_in_mask = 0x1ffff,
+	.hfb_filter_cnt = 48,
+	.hfb_filter_size = 128,
+	.qtag_mask = 0x3F,
+	.tbuf_offset = 0x0600,
+	.hfb_offset = 0x8000,
+	.hfb_reg_offset = 0xfc00,
+	.rdma_offset = 0x2000,
+	.tdma_offset = 0x4000,
+	.words_per_bd = 3,
 };
 
 /* Infer hardware parameters from the detected GENET version */
 static void bcmgenet_set_hw_params(struct bcmgenet_priv *priv)
 {
-	struct bcmgenet_hw_params *params;
+	const struct bcmgenet_hw_params *params;
 	u32 reg;
 	u8 major;
 	u16 gphy_rev;
 
-	if (GENET_IS_V5(priv) || GENET_IS_V4(priv)) {
-		bcmgenet_dma_regs = bcmgenet_dma_regs_v3plus;
-		genet_dma_ring_regs = genet_dma_ring_regs_v4;
-		priv->dma_rx_chk_bit = DMA_RX_CHK_V3PLUS;
-	} else if (GENET_IS_V3(priv)) {
+	/* default to latest values */
+	params = &bcmgenet_hw_params_v4;
+	bcmgenet_dma_regs = bcmgenet_dma_regs_v3plus;
+	genet_dma_ring_regs = genet_dma_ring_regs_v4;
+	if (GENET_IS_V3(priv)) {
+		params = &bcmgenet_hw_params_v3;
 		bcmgenet_dma_regs = bcmgenet_dma_regs_v3plus;
 		genet_dma_ring_regs = genet_dma_ring_regs_v123;
-		priv->dma_rx_chk_bit = DMA_RX_CHK_V3PLUS;
 	} else if (GENET_IS_V2(priv)) {
+		params = &bcmgenet_hw_params_v2;
 		bcmgenet_dma_regs = bcmgenet_dma_regs_v2;
 		genet_dma_ring_regs = genet_dma_ring_regs_v123;
-		priv->dma_rx_chk_bit = DMA_RX_CHK_V12;
 	} else if (GENET_IS_V1(priv)) {
+		params = &bcmgenet_hw_params_v1;
 		bcmgenet_dma_regs = bcmgenet_dma_regs_v1;
 		genet_dma_ring_regs = genet_dma_ring_regs_v123;
-		priv->dma_rx_chk_bit = DMA_RX_CHK_V12;
 	}
-
-	/* enum genet_version starts at 1 */
-	priv->hw_params = &bcmgenet_hw_params[priv->version];
-	params = priv->hw_params;
+	priv->hw_params = params;
 
 	/* Read GENET HW version */
 	reg = bcmgenet_sys_readl(priv, SYS_REV_CTRL);
 	major = (reg >> 24 & 0x0f);
-	if (major == 6)
+	if (major == 6 || major == 7)
 		major = 5;
 	else if (major == 5)
 		major = 4;
@@ -3385,7 +3837,7 @@ static void bcmgenet_set_hw_params(struct bcmgenet_priv *priv)
 	}
 
 #ifdef CONFIG_PHYS_ADDR_T_64BIT
-	if (!(params->flags & GENET_HAS_40BITS))
+	if (!bcmgenet_has_40bits(priv))
 		pr_warn("GENET does not support 40-bits PA\n");
 #endif
 
@@ -3407,12 +3859,67 @@ static void bcmgenet_set_hw_params(struct bcmgenet_priv *priv)
 		params->words_per_bd);
 }
 
+struct bcmgenet_plat_data {
+	enum bcmgenet_version version;
+	u32 dma_max_burst_length;
+	u32 flags;
+};
+
+static const struct bcmgenet_plat_data v1_plat_data = {
+	.version = GENET_V1,
+	.dma_max_burst_length = DMA_MAX_BURST_LENGTH,
+};
+
+static const struct bcmgenet_plat_data v2_plat_data = {
+	.version = GENET_V2,
+	.dma_max_burst_length = DMA_MAX_BURST_LENGTH,
+	.flags = GENET_HAS_EXT,
+};
+
+static const struct bcmgenet_plat_data v3_plat_data = {
+	.version = GENET_V3,
+	.dma_max_burst_length = DMA_MAX_BURST_LENGTH,
+	.flags = GENET_HAS_EXT | GENET_HAS_MDIO_INTR |
+		 GENET_HAS_MOCA_LINK_DET,
+};
+
+static const struct bcmgenet_plat_data v4_plat_data = {
+	.version = GENET_V4,
+	.dma_max_burst_length = DMA_MAX_BURST_LENGTH,
+	.flags = GENET_HAS_40BITS | GENET_HAS_EXT |
+		 GENET_HAS_MDIO_INTR | GENET_HAS_MOCA_LINK_DET,
+};
+
+static const struct bcmgenet_plat_data v5_plat_data = {
+	.version = GENET_V5,
+	.dma_max_burst_length = DMA_MAX_BURST_LENGTH,
+	.flags = GENET_HAS_40BITS | GENET_HAS_EXT |
+		 GENET_HAS_MDIO_INTR | GENET_HAS_MOCA_LINK_DET,
+};
+
+static const struct bcmgenet_plat_data bcm2711_plat_data = {
+	.version = GENET_V5,
+	.dma_max_burst_length = 0x08,
+	.flags = GENET_HAS_40BITS | GENET_HAS_EXT |
+		 GENET_HAS_MDIO_INTR | GENET_HAS_MOCA_LINK_DET,
+};
+
+static const struct bcmgenet_plat_data bcm7712_plat_data = {
+	.version = GENET_V5,
+	.dma_max_burst_length = DMA_MAX_BURST_LENGTH,
+	.flags = GENET_HAS_40BITS | GENET_HAS_EXT |
+		 GENET_HAS_MDIO_INTR | GENET_HAS_MOCA_LINK_DET |
+		 GENET_HAS_EPHY_16NM,
+};
+
 static const struct of_device_id bcmgenet_match[] = {
-	{ .compatible = "brcm,genet-v1", .data = (void *)GENET_V1 },
-	{ .compatible = "brcm,genet-v2", .data = (void *)GENET_V2 },
-	{ .compatible = "brcm,genet-v3", .data = (void *)GENET_V3 },
-	{ .compatible = "brcm,genet-v4", .data = (void *)GENET_V4 },
-	{ .compatible = "brcm,genet-v5", .data = (void *)GENET_V5 },
+	{ .compatible = "brcm,genet-v1", .data = &v1_plat_data },
+	{ .compatible = "brcm,genet-v2", .data = &v2_plat_data },
+	{ .compatible = "brcm,genet-v3", .data = &v3_plat_data },
+	{ .compatible = "brcm,genet-v4", .data = &v4_plat_data },
+	{ .compatible = "brcm,genet-v5", .data = &v5_plat_data },
+	{ .compatible = "brcm,bcm2711-genet-v5", .data = &bcm2711_plat_data },
+	{ .compatible = "brcm,bcm7712-genet-v5", .data = &bcm7712_plat_data },
 	{ },
 };
 MODULE_DEVICE_TABLE(of, bcmgenet_match);
@@ -3420,14 +3927,11 @@ MODULE_DEVICE_TABLE(of, bcmgenet_match);
 static int bcmgenet_probe(struct platform_device *pdev)
 {
 	struct bcmgenet_platform_data *pd = pdev->dev.platform_data;
-	struct device_node *dn = pdev->dev.of_node;
-	const struct of_device_id *of_id = NULL;
+	const struct bcmgenet_plat_data *pdata;
 	struct bcmgenet_priv *priv;
 	struct net_device *dev;
-	const void *macaddr;
-	struct resource *r;
+	unsigned int i;
 	int err = -EIO;
-	const char *phy_mode_str;
 
 	/* Up to GENET_MAX_MQ_CNT + 1 TX queues and RX queues */
 	dev = alloc_etherdev_mqs(sizeof(*priv), GENET_MAX_MQ_CNT + 1,
@@ -3437,125 +3941,163 @@ static int bcmgenet_probe(struct platform_device *pdev)
 		return -ENOMEM;
 	}
 
-	if (dn) {
-		of_id = of_match_node(bcmgenet_match, dn);
-		if (!of_id)
-			return -EINVAL;
-	}
-
 	priv = netdev_priv(dev);
 	priv->irq0 = platform_get_irq(pdev, 0);
+	if (priv->irq0 < 0) {
+		err = priv->irq0;
+		goto err;
+	}
 	priv->irq1 = platform_get_irq(pdev, 1);
-	priv->wol_irq = platform_get_irq(pdev, 2);
-	if (!priv->irq0 || !priv->irq1) {
-		dev_err(&pdev->dev, "can't find IRQs\n");
-		err = -EINVAL;
+	if (priv->irq1 < 0) {
+		err = priv->irq1;
 		goto err;
 	}
-
-	if (dn) {
-		macaddr = of_get_mac_address(dn);
-		if (!macaddr) {
-			dev_err(&pdev->dev, "can't find MAC address\n");
-			err = -EINVAL;
-			goto err;
-		}
-	} else {
-		macaddr = pd->mac_address;
+	priv->wol_irq = platform_get_irq_optional(pdev, 2);
+	if (priv->wol_irq == -EPROBE_DEFER) {
+		err = priv->wol_irq;
+		goto err;
 	}
 
-	r = platform_get_resource(pdev, IORESOURCE_MEM, 0);
-	priv->base = devm_ioremap_resource(&pdev->dev, r);
+	priv->base = devm_platform_ioremap_resource(pdev, 0);
 	if (IS_ERR(priv->base)) {
 		err = PTR_ERR(priv->base);
 		goto err;
 	}
 
+	spin_lock_init(&priv->reg_lock);
 	spin_lock_init(&priv->lock);
 
+	/* Set default pause parameters */
+	priv->autoneg_pause = 1;
+	priv->tx_pause = 1;
+	priv->rx_pause = 1;
+
 	SET_NETDEV_DEV(dev, &pdev->dev);
 	dev_set_drvdata(&pdev->dev, dev);
-	ether_addr_copy(dev->dev_addr, macaddr);
 	dev->watchdog_timeo = 2 * HZ;
 	dev->ethtool_ops = &bcmgenet_ethtool_ops;
 	dev->netdev_ops = &bcmgenet_netdev_ops;
 
 	priv->msg_enable = netif_msg_init(-1, GENET_MSG_DEFAULT);
 
-	/* Set hardware features */
-	dev->hw_features |= NETIF_F_SG | NETIF_F_IP_CSUM |
-		NETIF_F_IPV6_CSUM | NETIF_F_RXCSUM;
+	/* Set default features */
+	dev->features |= NETIF_F_SG | NETIF_F_HIGHDMA | NETIF_F_HW_CSUM |
+			 NETIF_F_RXCSUM;
+	dev->hw_features |= dev->features;
+	dev->vlan_features |= dev->features;
+
+	netdev_sw_irq_coalesce_default_on(dev);
 
 	/* Request the WOL interrupt and advertise suspend if available */
 	priv->wol_irq_disabled = true;
-	err = devm_request_irq(&pdev->dev, priv->wol_irq, bcmgenet_wol_isr, 0,
-			       dev->name, priv);
-	if (!err)
-		device_set_wakeup_capable(&pdev->dev, 1);
+	if (priv->wol_irq > 0) {
+		err = devm_request_irq(&pdev->dev, priv->wol_irq,
+				       bcmgenet_wol_isr, 0, dev->name, priv);
+		if (!err)
+			device_set_wakeup_capable(&pdev->dev, 1);
+	}
 
 	/* Set the needed headroom to account for any possible
 	 * features enabling/disabling at runtime
 	 */
 	dev->needed_headroom += 64;
 
-	netdev_boot_setup_check(dev);
-
 	priv->dev = dev;
 	priv->pdev = pdev;
-	if (of_id)
-		priv->version = (enum bcmgenet_version)of_id->data;
-	else
+
+	pdata = device_get_match_data(&pdev->dev);
+	if (pdata) {
+		priv->version = pdata->version;
+		priv->dma_max_burst_length = pdata->dma_max_burst_length;
+		priv->flags = pdata->flags;
+	} else {
 		priv->version = pd->genet_version;
+		priv->dma_max_burst_length = DMA_MAX_BURST_LENGTH;
+	}
 
-	priv->clk = devm_clk_get(&priv->pdev->dev, "enet");
+	priv->clk = devm_clk_get_optional(&priv->pdev->dev, "enet");
 	if (IS_ERR(priv->clk)) {
-		dev_warn(&priv->pdev->dev, "failed to get enet clock\n");
-		priv->clk = NULL;
+		dev_dbg(&priv->pdev->dev, "failed to get enet clock\n");
+		err = PTR_ERR(priv->clk);
+		goto err;
 	}
 
-	clk_prepare_enable(priv->clk);
+	err = clk_prepare_enable(priv->clk);
+	if (err)
+		goto err;
 
 	bcmgenet_set_hw_params(priv);
 
+	err = -EIO;
+	if (bcmgenet_has_40bits(priv))
+		err = dma_set_mask_and_coherent(&pdev->dev, DMA_BIT_MASK(40));
+	if (err)
+		err = dma_set_mask_and_coherent(&pdev->dev, DMA_BIT_MASK(32));
+	if (err)
+		goto err_clk_disable;
+
 	/* Mii wait queue */
 	init_waitqueue_head(&priv->wq);
 	/* Always use RX_BUF_LENGTH (2KB) buffer for all chips */
 	priv->rx_buf_len = RX_BUF_LENGTH;
 	INIT_WORK(&priv->bcmgenet_irq_work, bcmgenet_irq_task);
 
-	priv->clk_wol = devm_clk_get(&priv->pdev->dev, "enet-wol");
+	priv->clk_wol = devm_clk_get_optional(&priv->pdev->dev, "enet-wol");
 	if (IS_ERR(priv->clk_wol)) {
-		dev_warn(&priv->pdev->dev, "failed to get enet-wol clock\n");
-		priv->clk_wol = NULL;
+		dev_dbg(&priv->pdev->dev, "failed to get enet-wol clock\n");
+		err = PTR_ERR(priv->clk_wol);
+		goto err_clk_disable;
 	}
 
-	priv->clk_eee = devm_clk_get(&priv->pdev->dev, "enet-eee");
+	priv->clk_eee = devm_clk_get_optional(&priv->pdev->dev, "enet-eee");
 	if (IS_ERR(priv->clk_eee)) {
-		dev_warn(&priv->pdev->dev, "failed to get enet-eee clock\n");
-		priv->clk_eee = NULL;
+		dev_dbg(&priv->pdev->dev, "failed to get enet-eee clock\n");
+		err = PTR_ERR(priv->clk_eee);
+		goto err_clk_disable;
 	}
 
 	/* If this is an internal GPHY, power it on now, before UniMAC is
 	 * brought out of reset as absolutely no UniMAC activity is allowed
 	 */
-	if (dn && !of_property_read_string(dn, "phy-mode", &phy_mode_str) &&
-	    !strcasecmp(phy_mode_str, "internal"))
+	if (device_get_phy_mode(&pdev->dev) == PHY_INTERFACE_MODE_INTERNAL)
 		bcmgenet_power_up(priv, GENET_POWER_PASSIVE);
 
-	err = reset_umac(priv);
-	if (err)
-		goto err_clk_disable;
+	if (pd && !IS_ERR_OR_NULL(pd->mac_address))
+		eth_hw_addr_set(dev, pd->mac_address);
+	else
+		if (device_get_ethdev_address(&pdev->dev, dev))
+			if (has_acpi_companion(&pdev->dev)) {
+				u8 addr[ETH_ALEN];
+
+				bcmgenet_get_hw_addr(priv, addr);
+				eth_hw_addr_set(dev, addr);
+			}
+
+	if (!is_valid_ether_addr(dev->dev_addr)) {
+		dev_warn(&pdev->dev, "using random Ethernet MAC\n");
+		eth_hw_addr_random(dev);
+	}
+
+	reset_umac(priv);
 
 	err = bcmgenet_mii_init(dev);
 	if (err)
 		goto err_clk_disable;
 
-	/* setup number of real queues  + 1 (GENET_V1 has 0 hardware queues
-	 * just the ring 16 descriptor based TX
-	 */
+	/* setup number of real queues + 1 */
 	netif_set_real_num_tx_queues(priv->dev, priv->hw_params->tx_queues + 1);
 	netif_set_real_num_rx_queues(priv->dev, priv->hw_params->rx_queues + 1);
 
+	/* Set default coalescing parameters */
+	for (i = 0; i <= priv->hw_params->rx_queues; i++)
+		priv->rx_rings[i].rx_max_coalesced_frames = 1;
+
+	/* Initialize u64 stats seq counter for 32bit machines */
+	for (i = 0; i <= priv->hw_params->rx_queues; i++)
+		u64_stats_init(&priv->rx_rings[i].stats64.syncp);
+	for (i = 0; i <= priv->hw_params->tx_queues; i++)
+		u64_stats_init(&priv->tx_rings[i].stats64.syncp);
+
 	/* libphy will determine the link state */
 	netif_carrier_off(dev);
 
@@ -3563,8 +4105,10 @@ static int bcmgenet_probe(struct platform_device *pdev)
 	clk_disable_unprepare(priv->clk);
 
 	err = register_netdev(dev);
-	if (err)
+	if (err) {
+		bcmgenet_mii_exit(dev);
 		goto err;
+	}
 
 	return err;
 
@@ -3575,7 +4119,7 @@ err:
 	return err;
 }
 
-static int bcmgenet_remove(struct platform_device *pdev)
+static void bcmgenet_remove(struct platform_device *pdev)
 {
 	struct bcmgenet_priv *priv = dev_to_priv(&pdev->dev);
 
@@ -3583,128 +4127,138 @@ static int bcmgenet_remove(struct platform_device *pdev)
 	unregister_netdev(priv->dev);
 	bcmgenet_mii_exit(priv->dev);
 	free_netdev(priv->dev);
+}
 
-	return 0;
+static void bcmgenet_shutdown(struct platform_device *pdev)
+{
+	bcmgenet_remove(pdev);
 }
 
 #ifdef CONFIG_PM_SLEEP
-static int bcmgenet_suspend(struct device *d)
+static int bcmgenet_resume_noirq(struct device *d)
 {
 	struct net_device *dev = dev_get_drvdata(d);
 	struct bcmgenet_priv *priv = netdev_priv(dev);
 	int ret;
+	u32 reg;
 
 	if (!netif_running(dev))
 		return 0;
 
-	bcmgenet_netif_stop(dev);
-
-	if (!device_may_wakeup(d))
-		phy_suspend(priv->phydev);
-
-	netif_device_detach(dev);
-
-	/* Disable MAC receive */
-	umac_enable_set(priv, CMD_RX_EN, false);
-
-	ret = bcmgenet_dma_teardown(priv);
+	/* Turn on the clock */
+	ret = clk_prepare_enable(priv->clk);
 	if (ret)
 		return ret;
 
-	/* Disable MAC transmit. TX DMA disabled must be done before this */
-	umac_enable_set(priv, CMD_TX_EN, false);
+	if (device_may_wakeup(d) && priv->wolopts) {
+		/* Account for Wake-on-LAN events and clear those events
+		 * (Some devices need more time between enabling the clocks
+		 *  and the interrupt register reflecting the wake event so
+		 *  read the register twice)
+		 */
+		reg = bcmgenet_intrl2_0_readl(priv, INTRL2_CPU_STAT);
+		reg = bcmgenet_intrl2_0_readl(priv, INTRL2_CPU_STAT);
+		if (reg & UMAC_IRQ_WAKE_EVENT)
+			pm_wakeup_event(&priv->pdev->dev, 0);
 
-	/* tx reclaim */
-	bcmgenet_tx_reclaim_all(dev);
-	bcmgenet_fini_dma(priv);
+		/* From WOL-enabled suspend, switch to regular clock */
+		if (!bcmgenet_power_up(priv, GENET_POWER_WOL_MAGIC))
+			return 0;
 
-	/* Prepare the device for Wake-on-LAN and switch to the slow clock */
-	if (device_may_wakeup(d) && priv->wolopts) {
-		ret = bcmgenet_power_down(priv, GENET_POWER_WOL_MAGIC);
-		clk_prepare_enable(priv->clk_wol);
-	} else if (priv->internal_phy) {
-		ret = bcmgenet_power_down(priv, GENET_POWER_PASSIVE);
+		/* Failed so fall through to reset MAC */
 	}
 
-	/* Turn off the clocks */
-	clk_disable_unprepare(priv->clk);
+	/* If this is an internal GPHY, power it back on now, before UniMAC is
+	 * brought out of reset as absolutely no UniMAC activity is allowed
+	 */
+	if (priv->internal_phy)
+		bcmgenet_power_up(priv, GENET_POWER_PASSIVE);
 
-	return ret;
+	/* take MAC out of reset */
+	bcmgenet_umac_reset(priv);
+
+	return 0;
 }
 
 static int bcmgenet_resume(struct device *d)
 {
 	struct net_device *dev = dev_get_drvdata(d);
 	struct bcmgenet_priv *priv = netdev_priv(dev);
-	unsigned long dma_ctrl;
+	struct bcmgenet_rxnfc_rule *rule;
 	int ret;
 	u32 reg;
 
 	if (!netif_running(dev))
 		return 0;
 
-	/* Turn on the clock */
-	ret = clk_prepare_enable(priv->clk);
-	if (ret)
-		return ret;
-
-	/* If this is an internal GPHY, power it back on now, before UniMAC is
-	 * brought out of reset as absolutely no UniMAC activity is allowed
-	 */
-	if (priv->internal_phy)
-		bcmgenet_power_up(priv, GENET_POWER_PASSIVE);
+	if (device_may_wakeup(d) && priv->wolopts) {
+		reg = bcmgenet_umac_readl(priv, UMAC_CMD);
+		if (reg & CMD_RX_EN) {
+			/* Successfully exited WoL, just resume data flows */
+			list_for_each_entry(rule, &priv->rxnfc_list, list)
+				if (rule->state == BCMGENET_RXNFC_STATE_ENABLED)
+					bcmgenet_hfb_enable_filter(priv,
+							rule->fs.location + 1);
+			bcmgenet_hfb_enable_filter(priv, 0);
+			bcmgenet_set_rx_mode(dev);
+			bcmgenet_enable_rx_napi(priv);
+
+			/* Reinitialize Tx flows */
+			bcmgenet_tdma_disable(priv);
+			bcmgenet_init_tx_queues(priv->dev);
+			reg = bcmgenet_tdma_readl(priv, DMA_CTRL);
+			reg |= DMA_EN;
+			bcmgenet_tdma_writel(priv, reg, DMA_CTRL);
+			bcmgenet_enable_tx_napi(priv);
+
+			bcmgenet_link_intr_enable(priv);
+			phy_start_machine(dev->phydev);
+
+			netif_device_attach(dev);
+			enable_irq(priv->irq1);
+			return 0;
+		}
+		/* MAC was reset so complete bcmgenet_netif_stop() */
+		umac_enable_set(priv, CMD_RX_EN | CMD_TX_EN, false);
+		bcmgenet_rdma_disable(priv);
+		bcmgenet_intr_disable(priv);
+		bcmgenet_fini_dma(priv);
+		enable_irq(priv->irq1);
+	}
 
-	bcmgenet_umac_reset(priv);
+	init_umac(priv);
 
-	ret = init_umac(priv);
-	if (ret)
-		goto out_clk_disable;
+	phy_init_hw(dev->phydev);
 
-	/* From WOL-enabled suspend, switch to regular clock */
-	if (priv->wolopts)
-		clk_disable_unprepare(priv->clk_wol);
-
-	phy_init_hw(priv->phydev);
 	/* Speed settings must be restored */
-	bcmgenet_mii_config(priv->dev);
+	genphy_config_aneg(dev->phydev);
+	bcmgenet_mii_config(priv->dev, false);
 
-	/* disable ethernet MAC while updating its registers */
-	umac_enable_set(priv, CMD_TX_EN | CMD_RX_EN, false);
+	/* Restore enabled features */
+	bcmgenet_set_features(dev, dev->features);
 
 	bcmgenet_set_hw_addr(priv, dev->dev_addr);
 
-	if (priv->internal_phy) {
-		reg = bcmgenet_ext_readl(priv, EXT_EXT_PWR_MGMT);
-		reg |= EXT_ENERGY_DET_MASK;
-		bcmgenet_ext_writel(priv, reg, EXT_EXT_PWR_MGMT);
-	}
-
-	if (priv->wolopts)
-		bcmgenet_power_up(priv, GENET_POWER_WOL_MAGIC);
-
-	/* Disable RX/TX DMA and flush TX queues */
-	dma_ctrl = bcmgenet_dma_disable(priv);
+	/* Restore hardware filters */
+	bcmgenet_hfb_clear(priv);
+	list_for_each_entry(rule, &priv->rxnfc_list, list)
+		if (rule->state != BCMGENET_RXNFC_STATE_UNUSED)
+			bcmgenet_hfb_create_rxnfc_filter(priv, rule);
 
 	/* Reinitialize TDMA and RDMA and SW housekeeping */
-	ret = bcmgenet_init_dma(priv);
+	ret = bcmgenet_init_dma(priv, false);
 	if (ret) {
 		netdev_err(dev, "failed to initialize DMA\n");
 		goto out_clk_disable;
 	}
 
-	/* Always enable ring 16 - descriptor ring */
-	bcmgenet_enable_dma(priv, dma_ctrl);
-
-	netif_device_attach(dev);
-
 	if (!device_may_wakeup(d))
-		phy_resume(priv->phydev);
-
-	if (priv->eee.eee_enabled)
-		bcmgenet_eee_enable_set(dev, true);
+		phy_resume(dev->phydev);
 
 	bcmgenet_netif_start(dev);
 
+	netif_device_attach(dev);
+
 	return 0;
 
 out_clk_disable:
@@ -3713,17 +4267,114 @@ out_clk_disable:
 	clk_disable_unprepare(priv->clk);
 	return ret;
 }
+
+static int bcmgenet_suspend(struct device *d)
+{
+	struct net_device *dev = dev_get_drvdata(d);
+	struct bcmgenet_priv *priv = netdev_priv(dev);
+	struct bcmgenet_rxnfc_rule *rule;
+	u32 reg, hfb_enable = 0;
+
+	if (!netif_running(dev))
+		return 0;
+
+	netif_device_detach(dev);
+
+	if (device_may_wakeup(d) && priv->wolopts) {
+		netif_tx_disable(dev);
+
+		/* Suspend non-wake Rx data flows */
+		if (priv->wolopts & WAKE_FILTER)
+			list_for_each_entry(rule, &priv->rxnfc_list, list)
+				if (rule->fs.ring_cookie == RX_CLS_FLOW_WAKE &&
+				    rule->state == BCMGENET_RXNFC_STATE_ENABLED)
+					hfb_enable |= 1 << rule->fs.location;
+		reg = bcmgenet_hfb_reg_readl(priv, HFB_CTRL);
+		if (GENET_IS_V1(priv) || GENET_IS_V2(priv)) {
+			reg &= ~RBUF_HFB_FILTER_EN_MASK;
+			reg |= hfb_enable << (RBUF_HFB_FILTER_EN_SHIFT + 1);
+		} else {
+			bcmgenet_hfb_reg_writel(priv, hfb_enable << 1,
+						HFB_FLT_ENABLE_V3PLUS + 4);
+		}
+		if (!hfb_enable)
+			reg &= ~RBUF_HFB_EN;
+		bcmgenet_hfb_reg_writel(priv, reg, HFB_CTRL);
+
+		/* Clear any old filter matches so only new matches wake */
+		bcmgenet_intrl2_0_writel(priv, 0xFFFFFFFF, INTRL2_CPU_MASK_SET);
+		bcmgenet_intrl2_0_writel(priv, 0xFFFFFFFF, INTRL2_CPU_CLEAR);
+
+		if (-ETIMEDOUT == bcmgenet_tdma_disable(priv))
+			netdev_warn(priv->dev,
+				    "Timed out while disabling TX DMA\n");
+
+		bcmgenet_disable_tx_napi(priv);
+		bcmgenet_disable_rx_napi(priv);
+		disable_irq(priv->irq1);
+		bcmgenet_tx_reclaim_all(dev);
+		bcmgenet_fini_tx_napi(priv);
+	} else {
+		/* Teardown the interface */
+		bcmgenet_netif_stop(dev, true);
+	}
+
+	return 0;
+}
+
+static int bcmgenet_suspend_noirq(struct device *d)
+{
+	struct net_device *dev = dev_get_drvdata(d);
+	struct bcmgenet_priv *priv = netdev_priv(dev);
+	int ret = 0;
+
+	if (!netif_running(dev))
+		return 0;
+
+	/* Prepare the device for Wake-on-LAN and switch to the slow clock */
+	if (device_may_wakeup(d) && priv->wolopts)
+		ret = bcmgenet_power_down(priv, GENET_POWER_WOL_MAGIC);
+	else if (priv->internal_phy)
+		ret = bcmgenet_power_down(priv, GENET_POWER_PASSIVE);
+
+	/* Let the framework handle resumption and leave the clocks on */
+	if (ret)
+		return ret;
+
+	/* Turn off the clocks */
+	clk_disable_unprepare(priv->clk);
+
+	return 0;
+}
+#else
+#define bcmgenet_suspend	NULL
+#define bcmgenet_suspend_noirq	NULL
+#define bcmgenet_resume		NULL
+#define bcmgenet_resume_noirq	NULL
 #endif /* CONFIG_PM_SLEEP */
 
-static SIMPLE_DEV_PM_OPS(bcmgenet_pm_ops, bcmgenet_suspend, bcmgenet_resume);
+static const struct dev_pm_ops bcmgenet_pm_ops = {
+	.suspend	= bcmgenet_suspend,
+	.suspend_noirq	= bcmgenet_suspend_noirq,
+	.resume		= bcmgenet_resume,
+	.resume_noirq	= bcmgenet_resume_noirq,
+};
+
+static const struct acpi_device_id genet_acpi_match[] = {
+	{ "BCM6E4E", (kernel_ulong_t)&bcm2711_plat_data },
+	{ },
+};
+MODULE_DEVICE_TABLE(acpi, genet_acpi_match);
 
 static struct platform_driver bcmgenet_driver = {
 	.probe	= bcmgenet_probe,
-	.remove	= bcmgenet_remove,
+	.remove = bcmgenet_remove,
+	.shutdown = bcmgenet_shutdown,
 	.driver	= {
 		.name	= "bcmgenet",
 		.of_match_table = bcmgenet_match,
 		.pm	= &bcmgenet_pm_ops,
+		.acpi_match_table = genet_acpi_match,
 	},
 };
 module_platform_driver(bcmgenet_driver);
@@ -3732,3 +4383,4 @@ MODULE_AUTHOR("Broadcom Corporation");
 MODULE_DESCRIPTION("Broadcom GENET Ethernet controller driver");
 MODULE_ALIAS("platform:bcmgenet");
 MODULE_LICENSE("GPL");
+MODULE_SOFTDEP("pre: mdio-bcm-unimac");
diff --git a/drivers/net/ethernet/broadcom/genet/bcmgenet.h b/drivers/net/ethernet/broadcom/genet/bcmgenet.h
index efd07020b89f..5ec3979779ec 100644
--- a/drivers/net/ethernet/broadcom/genet/bcmgenet.h
+++ b/drivers/net/ethernet/broadcom/genet/bcmgenet.h
@@ -1,9 +1,6 @@
+/* SPDX-License-Identifier: GPL-2.0-only */
 /*
- * Copyright (c) 2014-2017 Broadcom
- *
- * This program is free software; you can redistribute it and/or modify
- * it under the terms of the GNU General Public License version 2 as
- * published by the Free Software Foundation.
+ * Copyright (c) 2014-2025 Broadcom
  */
 
 #ifndef __BCMGENET_H__
@@ -16,6 +13,13 @@
 #include <linux/mii.h>
 #include <linux/if_vlan.h>
 #include <linux/phy.h>
+#include <linux/dim.h>
+#include <linux/ethtool.h>
+
+#include "../unimac.h"
+
+/* Maximum number of hardware queues, downsized if needed */
+#define GENET_MAX_MQ_CNT	4
 
 /* total number of Buffer Descriptors, same for Rx/Tx */
 #define TOTAL_DESC				256
@@ -33,6 +37,7 @@
 #define DMA_MAX_BURST_LENGTH    0x10
 
 /* misc. configuration */
+#define MAX_NUM_OF_FS_RULES		16
 #define CLEAR_ALL_HFB			0xFF
 #define DMA_FC_THRESH_HI		(TOTAL_DESC >> 4)
 #define DMA_FC_THRESH_LO		5
@@ -146,61 +151,33 @@ struct bcmgenet_mib_counters {
 	u32	alloc_rx_buff_failed;
 	u32	rx_dma_failed;
 	u32	tx_dma_failed;
+	u32	tx_realloc_tsb;
+	u32	tx_realloc_tsb_failed;
 };
 
-#define UMAC_HD_BKP_CTRL		0x004
-#define	 HD_FC_EN			(1 << 0)
-#define  HD_FC_BKOFF_OK			(1 << 1)
-#define  IPG_CONFIG_RX_SHIFT		2
-#define  IPG_CONFIG_RX_MASK		0x1F
-
-#define UMAC_CMD			0x008
-#define  CMD_TX_EN			(1 << 0)
-#define  CMD_RX_EN			(1 << 1)
-#define  UMAC_SPEED_10			0
-#define  UMAC_SPEED_100			1
-#define  UMAC_SPEED_1000		2
-#define  UMAC_SPEED_2500		3
-#define  CMD_SPEED_SHIFT		2
-#define  CMD_SPEED_MASK			3
-#define  CMD_PROMISC			(1 << 4)
-#define  CMD_PAD_EN			(1 << 5)
-#define  CMD_CRC_FWD			(1 << 6)
-#define  CMD_PAUSE_FWD			(1 << 7)
-#define  CMD_RX_PAUSE_IGNORE		(1 << 8)
-#define  CMD_TX_ADDR_INS		(1 << 9)
-#define  CMD_HD_EN			(1 << 10)
-#define  CMD_SW_RESET			(1 << 13)
-#define  CMD_LCL_LOOP_EN		(1 << 15)
-#define  CMD_AUTO_CONFIG		(1 << 22)
-#define  CMD_CNTL_FRM_EN		(1 << 23)
-#define  CMD_NO_LEN_CHK			(1 << 24)
-#define  CMD_RMT_LOOP_EN		(1 << 25)
-#define  CMD_PRBL_EN			(1 << 27)
-#define  CMD_TX_PAUSE_IGNORE		(1 << 28)
-#define  CMD_TX_RX_EN			(1 << 29)
-#define  CMD_RUNT_FILTER_DIS		(1 << 30)
-
-#define UMAC_MAC0			0x00C
-#define UMAC_MAC1			0x010
-#define UMAC_MAX_FRAME_LEN		0x014
-
-#define UMAC_EEE_CTRL			0x064
-#define  EN_LPI_RX_PAUSE		(1 << 0)
-#define  EN_LPI_TX_PFC			(1 << 1)
-#define  EN_LPI_TX_PAUSE		(1 << 2)
-#define  EEE_EN				(1 << 3)
-#define  RX_FIFO_CHECK			(1 << 4)
-#define  EEE_TX_CLK_DIS			(1 << 5)
-#define  DIS_EEE_10M			(1 << 6)
-#define  LP_IDLE_PREDICTION_MODE	(1 << 7)
-
-#define UMAC_EEE_LPI_TIMER		0x068
-#define UMAC_EEE_WAKE_TIMER		0x06C
-#define UMAC_EEE_REF_COUNT		0x070
-#define  EEE_REFERENCE_COUNT_MASK	0xffff
-
-#define UMAC_TX_FLUSH			0x334
+struct bcmgenet_tx_stats64 {
+	struct u64_stats_sync syncp;
+	u64_stats_t	packets;
+	u64_stats_t	bytes;
+	u64_stats_t	errors;
+	u64_stats_t	dropped;
+};
+
+struct bcmgenet_rx_stats64 {
+	struct u64_stats_sync syncp;
+	u64_stats_t	bytes;
+	u64_stats_t	packets;
+	u64_stats_t	errors;
+	u64_stats_t	dropped;
+	u64_stats_t	multicast;
+	u64_stats_t	broadcast;
+	u64_stats_t	missed;
+	u64_stats_t	length_errors;
+	u64_stats_t	over_errors;
+	u64_stats_t	crc_errors;
+	u64_stats_t	frame_errors;
+	u64_stats_t	fragmented_errors;
+};
 
 #define UMAC_MIB_START			0x400
 
@@ -250,6 +227,7 @@ struct bcmgenet_mib_counters {
 #define RBUF_CHK_CTRL			0x14
 #define  RBUF_RXCHK_EN			(1 << 0)
 #define  RBUF_SKIP_FCS			(1 << 4)
+#define  RBUF_L3_PARSE_DIS		(1 << 5)
 
 #define RBUF_ENERGY_CTRL		0x9c
 #define  RBUF_EEE_EN			(1 << 0)
@@ -269,6 +247,7 @@ struct bcmgenet_mib_counters {
 #define  RBUF_FLTR_LEN_SHIFT		8
 
 #define TBUF_CTRL			0x00
+#define  TBUF_64B_EN			(1 << 0)
 #define TBUF_BP_MC			0x0C
 #define TBUF_ENERGY_CTRL		0x14
 #define  TBUF_EEE_EN			(1 << 0)
@@ -305,6 +284,8 @@ struct bcmgenet_mib_counters {
 #define UMAC_IRQ_HFB_SM			(1 << 10)
 #define UMAC_IRQ_HFB_MM			(1 << 11)
 #define UMAC_IRQ_MPD_R			(1 << 12)
+#define UMAC_IRQ_WAKE_EVENT		(UMAC_IRQ_HFB_SM | UMAC_IRQ_HFB_MM | \
+					 UMAC_IRQ_MPD_R)
 #define UMAC_IRQ_RXDMA_MBDONE		(1 << 13)
 #define UMAC_IRQ_RXDMA_PDONE		(1 << 14)
 #define UMAC_IRQ_RXDMA_BDONE		(1 << 15)
@@ -317,6 +298,8 @@ struct bcmgenet_mib_counters {
 /* Only valid for GENETv3+ */
 #define UMAC_IRQ_MDIO_DONE		(1 << 23)
 #define UMAC_IRQ_MDIO_ERROR		(1 << 24)
+#define UMAC_IRQ_MDIO_EVENT		(UMAC_IRQ_MDIO_DONE | \
+					 UMAC_IRQ_MDIO_ERROR)
 
 /* INTRL2 instance 1 definitions */
 #define UMAC_IRQ1_TX_INTR_MASK		0xFFFF
@@ -365,6 +348,7 @@ struct bcmgenet_mib_counters {
 #define  EXT_PWR_DOWN_PHY_EN		(1 << 20)
 
 #define EXT_RGMII_OOB_CTRL		0x0C
+#define  RGMII_MODE_EN_V123		(1 << 0)
 #define  RGMII_LINK			(1 << 4)
 #define  OOB_DISABLE			(1 << 5)
 #define  RGMII_MODE_EN			(1 << 6)
@@ -374,6 +358,7 @@ struct bcmgenet_mib_counters {
 #define  EXT_CFG_IDDQ_BIAS		(1 << 0)
 #define  EXT_CFG_PWR_DOWN		(1 << 1)
 #define  EXT_CK25_DIS			(1 << 4)
+#define  EXT_CFG_IDDQ_GLOBAL_PWR	(1 << 3)
 #define  EXT_GPHY_RESET			(1 << 5)
 
 /* DMA rings size */
@@ -520,6 +505,7 @@ enum bcmgenet_version {
 #define GENET_HAS_EXT		(1 << 1)
 #define GENET_HAS_MDIO_INTR	(1 << 2)
 #define GENET_HAS_MOCA_LINK_DET	(1 << 3)
+#define GENET_HAS_EPHY_16NM	(1 << 4)
 
 /* BCMGENET hardware parameters, keep this structure nicely aligned
  * since it is going to be used in hot paths
@@ -540,10 +526,11 @@ struct bcmgenet_hw_params {
 	u32		rdma_offset;
 	u32		tdma_offset;
 	u32		words_per_bd;
-	u32		flags;
 };
 
 struct bcmgenet_skb_cb {
+	struct enet_cb *first_cb;	/* First control block of SKB */
+	struct enet_cb *last_cb;	/* Last control block of SKB */
 	unsigned int bytes_sent;	/* bytes on the wire (no TSB) */
 };
 
@@ -552,10 +539,8 @@ struct bcmgenet_skb_cb {
 struct bcmgenet_tx_ring {
 	spinlock_t	lock;		/* ring lock */
 	struct napi_struct napi;	/* NAPI per tx queue */
-	unsigned long	packets;
-	unsigned long	bytes;
+	struct bcmgenet_tx_stats64 stats64;
 	unsigned int	index;		/* ring index */
-	unsigned int	queue;		/* queue index */
 	struct enet_cb	*cbs;		/* tx ring buffer control block*/
 	unsigned int	size;		/* size of each tx ring */
 	unsigned int    clean_ptr;      /* Tx ring clean pointer */
@@ -565,17 +550,20 @@ struct bcmgenet_tx_ring {
 	unsigned int	prod_index;	/* Tx ring producer index SW copy */
 	unsigned int	cb_ptr;		/* Tx ring initial CB ptr */
 	unsigned int	end_ptr;	/* Tx ring end CB ptr */
-	void (*int_enable)(struct bcmgenet_tx_ring *);
-	void (*int_disable)(struct bcmgenet_tx_ring *);
 	struct bcmgenet_priv *priv;
 };
 
+struct bcmgenet_net_dim {
+	u16		use_dim;
+	u16		event_ctr;
+	unsigned long	packets;
+	unsigned long	bytes;
+	struct dim	dim;
+};
+
 struct bcmgenet_rx_ring {
 	struct napi_struct napi;	/* Rx NAPI struct */
-	unsigned long	bytes;
-	unsigned long	packets;
-	unsigned long	errors;
-	unsigned long	dropped;
+	struct bcmgenet_rx_stats64 stats64;
 	unsigned int	index;		/* Rx ring index */
 	struct enet_cb	*cbs;		/* Rx ring buffer control block */
 	unsigned int	size;		/* Rx ring size */
@@ -584,14 +572,29 @@ struct bcmgenet_rx_ring {
 	unsigned int	cb_ptr;		/* Rx ring initial CB ptr */
 	unsigned int	end_ptr;	/* Rx ring end CB ptr */
 	unsigned int	old_discards;
-	void (*int_enable)(struct bcmgenet_rx_ring *);
-	void (*int_disable)(struct bcmgenet_rx_ring *);
+	struct bcmgenet_net_dim dim;
+	u32		rx_max_coalesced_frames;
+	u32		rx_coalesce_usecs;
 	struct bcmgenet_priv *priv;
 };
 
+enum bcmgenet_rxnfc_state {
+	BCMGENET_RXNFC_STATE_UNUSED = 0,
+	BCMGENET_RXNFC_STATE_DISABLED,
+	BCMGENET_RXNFC_STATE_ENABLED
+};
+
+struct bcmgenet_rxnfc_rule {
+	struct	list_head list;
+	struct ethtool_rx_flow_spec	fs;
+	enum bcmgenet_rxnfc_state state;
+};
+
 /* device context */
 struct bcmgenet_priv {
 	void __iomem *base;
+	/* reg_lock: lock to serialize access to shared registers */
+	spinlock_t reg_lock;
 	enum bcmgenet_version version;
 	struct net_device *dev;
 
@@ -600,22 +603,27 @@ struct bcmgenet_priv {
 	struct enet_cb *tx_cbs;
 	unsigned int num_tx_bds;
 
-	struct bcmgenet_tx_ring tx_rings[DESC_INDEX + 1];
+	struct bcmgenet_tx_ring tx_rings[GENET_MAX_MQ_CNT + 1];
 
 	/* receive variables */
 	void __iomem *rx_bds;
 	struct enet_cb *rx_cbs;
 	unsigned int num_rx_bds;
 	unsigned int rx_buf_len;
+	struct bcmgenet_rxnfc_rule rxnfc_rules[MAX_NUM_OF_FS_RULES];
+	struct list_head rxnfc_list;
 
-	struct bcmgenet_rx_ring rx_rings[DESC_INDEX + 1];
+	struct bcmgenet_rx_ring rx_rings[GENET_MAX_MQ_CNT + 1];
 
 	/* other misc variables */
-	struct bcmgenet_hw_params *hw_params;
+	const struct bcmgenet_hw_params *hw_params;
+	u32 flags;
+	unsigned autoneg_pause:1;
+	unsigned tx_pause:1;
+	unsigned rx_pause:1;
 
 	/* MDIO bus variables */
 	wait_queue_head_t wq;
-	struct phy_device *phydev;
 	bool internal_phy;
 	struct device_node *phy_dn;
 	struct device_node *mdio_dn;
@@ -625,10 +633,6 @@ struct bcmgenet_priv {
 	bool clk_eee_enabled;
 
 	/* PHY device variables */
-	int old_link;
-	int old_speed;
-	int old_duplex;
-	int old_pause;
 	phy_interface_t phy_interface;
 	int phy_addr;
 	int ext_phy;
@@ -645,36 +649,70 @@ struct bcmgenet_priv {
 	unsigned int irq0_stat;
 
 	/* HW descriptors/checksum variables */
-	bool desc_64b_en;
-	bool desc_rxchk_en;
 	bool crc_fwd_en;
 
-	unsigned int dma_rx_chk_bit;
+	u32 dma_max_burst_length;
 
 	u32 msg_enable;
 
 	struct clk *clk;
 	struct platform_device *pdev;
+	struct platform_device *mii_pdev;
 
 	/* WOL */
 	struct clk *clk_wol;
 	u32 wolopts;
+	u8 sopass[SOPASS_MAX];
 
 	struct bcmgenet_mib_counters mib;
 
-	struct ethtool_eee eee;
+	struct ethtool_keee eee;
 };
 
+static inline bool bcmgenet_has_40bits(struct bcmgenet_priv *priv)
+{
+	return !!(priv->flags & GENET_HAS_40BITS);
+}
+
+static inline bool bcmgenet_has_ext(struct bcmgenet_priv *priv)
+{
+	return !!(priv->flags & GENET_HAS_EXT);
+}
+
+static inline bool bcmgenet_has_mdio_intr(struct bcmgenet_priv *priv)
+{
+	return !!(priv->flags & GENET_HAS_MDIO_INTR);
+}
+
+static inline bool bcmgenet_has_moca_link_det(struct bcmgenet_priv *priv)
+{
+	return !!(priv->flags & GENET_HAS_MOCA_LINK_DET);
+}
+
+static inline bool bcmgenet_has_ephy_16nm(struct bcmgenet_priv *priv)
+{
+	return !!(priv->flags & GENET_HAS_EPHY_16NM);
+}
+
 #define GENET_IO_MACRO(name, offset)					\
 static inline u32 bcmgenet_##name##_readl(struct bcmgenet_priv *priv,	\
 					u32 off)			\
 {									\
-	return __raw_readl(priv->base + offset + off);			\
+	/* MIPS chips strapped for BE will automagically configure the	\
+	 * peripheral registers for CPU-native byte order.		\
+	 */								\
+	if (IS_ENABLED(CONFIG_MIPS) && IS_ENABLED(CONFIG_CPU_BIG_ENDIAN)) \
+		return __raw_readl(priv->base + offset + off);		\
+	else								\
+		return readl_relaxed(priv->base + offset + off);	\
 }									\
 static inline void bcmgenet_##name##_writel(struct bcmgenet_priv *priv,	\
 					u32 val, u32 off)		\
 {									\
-	__raw_writel(val, priv->base + offset + off);			\
+	if (IS_ENABLED(CONFIG_MIPS) && IS_ENABLED(CONFIG_CPU_BIG_ENDIAN)) \
+		__raw_writel(val, priv->base + offset + off);		\
+	else								\
+		writel_relaxed(val, priv->base + offset + off);		\
 }
 
 GENET_IO_MACRO(ext, GENET_EXT_OFF);
@@ -696,10 +734,10 @@ GENET_IO_MACRO(rbuf, GENET_RBUF_OFF);
 
 /* MDIO routines */
 int bcmgenet_mii_init(struct net_device *dev);
-int bcmgenet_mii_config(struct net_device *dev);
+int bcmgenet_mii_config(struct net_device *dev, bool init);
 int bcmgenet_mii_probe(struct net_device *dev);
 void bcmgenet_mii_exit(struct net_device *dev);
-void bcmgenet_mii_reset(struct net_device *dev);
+void bcmgenet_phy_pause_set(struct net_device *dev, bool rx, bool tx);
 void bcmgenet_phy_power_set(struct net_device *dev, bool enable);
 void bcmgenet_mii_setup(struct net_device *dev);
 
@@ -708,7 +746,10 @@ void bcmgenet_get_wol(struct net_device *dev, struct ethtool_wolinfo *wol);
 int bcmgenet_set_wol(struct net_device *dev, struct ethtool_wolinfo *wol);
 int bcmgenet_wol_power_down_cfg(struct bcmgenet_priv *priv,
 				enum bcmgenet_power_mode mode);
-void bcmgenet_wol_power_up_cfg(struct bcmgenet_priv *priv,
-			       enum bcmgenet_power_mode mode);
+int bcmgenet_wol_power_up_cfg(struct bcmgenet_priv *priv,
+			      enum bcmgenet_power_mode mode);
+
+void bcmgenet_eee_enable_set(struct net_device *dev, bool enable,
+			     bool tx_lpi_enabled);
 
 #endif /* __BCMGENET_H__ */
diff --git a/drivers/net/ethernet/broadcom/genet/bcmgenet_wol.c b/drivers/net/ethernet/broadcom/genet/bcmgenet_wol.c
index 2fbd027f0148..8fb551288298 100644
--- a/drivers/net/ethernet/broadcom/genet/bcmgenet_wol.c
+++ b/drivers/net/ethernet/broadcom/genet/bcmgenet_wol.c
@@ -1,11 +1,8 @@
+// SPDX-License-Identifier: GPL-2.0-only
 /*
  * Broadcom GENET (Gigabit Ethernet) Wake-on-LAN support
  *
- * Copyright (c) 2014-2017 Broadcom
- *
- * This program is free software; you can redistribute it and/or modify
- * it under the terms of the GNU General Public License version 2 as
- * published by the Free Software Foundation.
+ * Copyright (c) 2014-2025 Broadcom
  */
 
 #define pr_fmt(fmt)				"bcmgenet_wol: " fmt
@@ -21,7 +18,6 @@
 #include <linux/delay.h>
 #include <linux/pm.h>
 #include <linux/clk.h>
-#include <linux/version.h>
 #include <linux/platform_device.h>
 #include <net/arp.h>
 
@@ -44,18 +40,30 @@
 void bcmgenet_get_wol(struct net_device *dev, struct ethtool_wolinfo *wol)
 {
 	struct bcmgenet_priv *priv = netdev_priv(dev);
-	u32 reg;
-
-	wol->supported = WAKE_MAGIC | WAKE_MAGICSECURE;
-	wol->wolopts = priv->wolopts;
-	memset(wol->sopass, 0, sizeof(wol->sopass));
+	struct device *kdev = &priv->pdev->dev;
+	u32 phy_wolopts = 0;
 
-	if (wol->wolopts & WAKE_MAGICSECURE) {
-		reg = bcmgenet_umac_readl(priv, UMAC_MPD_PW_MS);
-		put_unaligned_be16(reg, &wol->sopass[0]);
-		reg = bcmgenet_umac_readl(priv, UMAC_MPD_PW_LS);
-		put_unaligned_be32(reg, &wol->sopass[2]);
+	if (dev->phydev) {
+		phy_ethtool_get_wol(dev->phydev, wol);
+		phy_wolopts = wol->wolopts;
 	}
+
+	/* MAC is not wake-up capable, return what the PHY does */
+	if (!device_can_wakeup(kdev))
+		return;
+
+	/* Overlay MAC capabilities with that of the PHY queried before */
+	wol->supported |= WAKE_MAGIC | WAKE_MAGICSECURE | WAKE_FILTER;
+	wol->wolopts |= priv->wolopts;
+
+	/* Return the PHY configured magic password */
+	if (phy_wolopts & WAKE_MAGICSECURE)
+		return;
+
+	/* Otherwise the MAC one */
+	memset(wol->sopass, 0, sizeof(wol->sopass));
+	if (wol->wolopts & WAKE_MAGICSECURE)
+		memcpy(wol->sopass, priv->sopass, sizeof(priv->sopass));
 }
 
 /* ethtool function - set WOL (Wake on LAN) settings.
@@ -65,38 +73,40 @@ int bcmgenet_set_wol(struct net_device *dev, struct ethtool_wolinfo *wol)
 {
 	struct bcmgenet_priv *priv = netdev_priv(dev);
 	struct device *kdev = &priv->pdev->dev;
-	u32 reg;
+	int ret;
+
+	/* Try Wake-on-LAN from the PHY first */
+	if (dev->phydev) {
+		ret = phy_ethtool_set_wol(dev->phydev, wol);
+		if (ret != -EOPNOTSUPP && wol->wolopts)
+			return ret;
+	}
 
 	if (!device_can_wakeup(kdev))
 		return -ENOTSUPP;
 
-	if (wol->wolopts & ~(WAKE_MAGIC | WAKE_MAGICSECURE))
+	if (wol->wolopts & ~(WAKE_MAGIC | WAKE_MAGICSECURE | WAKE_FILTER))
 		return -EINVAL;
 
-	reg = bcmgenet_umac_readl(priv, UMAC_MPD_CTRL);
-	if (wol->wolopts & WAKE_MAGICSECURE) {
-		bcmgenet_umac_writel(priv, get_unaligned_be16(&wol->sopass[0]),
-				     UMAC_MPD_PW_MS);
-		bcmgenet_umac_writel(priv, get_unaligned_be32(&wol->sopass[2]),
-				     UMAC_MPD_PW_LS);
-		reg |= MPD_PW_EN;
-	} else {
-		reg &= ~MPD_PW_EN;
-	}
-	bcmgenet_umac_writel(priv, reg, UMAC_MPD_CTRL);
+	if (wol->wolopts & WAKE_MAGICSECURE)
+		memcpy(priv->sopass, wol->sopass, sizeof(priv->sopass));
 
 	/* Flag the device and relevant IRQ as wakeup capable */
 	if (wol->wolopts) {
 		device_set_wakeup_enable(kdev, 1);
 		/* Avoid unbalanced enable_irq_wake calls */
-		if (priv->wol_irq_disabled)
+		if (priv->wol_irq_disabled) {
 			enable_irq_wake(priv->wol_irq);
+			enable_irq_wake(priv->irq0);
+		}
 		priv->wol_irq_disabled = false;
 	} else {
 		device_set_wakeup_enable(kdev, 0);
 		/* Avoid unbalanced disable_irq_wake calls */
-		if (!priv->wol_irq_disabled)
+		if (!priv->wol_irq_disabled) {
 			disable_irq_wake(priv->wol_irq);
+			disable_irq_wake(priv->irq0);
+		}
 		priv->wol_irq_disabled = true;
 	}
 
@@ -123,75 +133,135 @@ static int bcmgenet_poll_wol_status(struct bcmgenet_priv *priv)
 	return retries;
 }
 
+static void bcmgenet_set_mpd_password(struct bcmgenet_priv *priv)
+{
+	bcmgenet_umac_writel(priv, get_unaligned_be16(&priv->sopass[0]),
+			     UMAC_MPD_PW_MS);
+	bcmgenet_umac_writel(priv, get_unaligned_be32(&priv->sopass[2]),
+			     UMAC_MPD_PW_LS);
+}
+
 int bcmgenet_wol_power_down_cfg(struct bcmgenet_priv *priv,
 				enum bcmgenet_power_mode mode)
 {
 	struct net_device *dev = priv->dev;
+	u32 reg, hfb_ctrl_reg;
 	int retries = 0;
-	u32 reg;
 
 	if (mode != GENET_POWER_WOL_MAGIC) {
 		netif_err(priv, wol, dev, "unsupported mode: %d\n", mode);
 		return -EINVAL;
 	}
 
-	/* disable RX */
-	reg = bcmgenet_umac_readl(priv, UMAC_CMD);
-	reg &= ~CMD_RX_EN;
-	bcmgenet_umac_writel(priv, reg, UMAC_CMD);
-	mdelay(10);
+	if (priv->wolopts & (WAKE_MAGIC | WAKE_MAGICSECURE)) {
+		reg = bcmgenet_umac_readl(priv, UMAC_MPD_CTRL);
+		reg |= MPD_EN;
+		if (priv->wolopts & WAKE_MAGICSECURE) {
+			bcmgenet_set_mpd_password(priv);
+			reg |= MPD_PW_EN;
+		}
+		bcmgenet_umac_writel(priv, reg, UMAC_MPD_CTRL);
+	}
 
-	reg = bcmgenet_umac_readl(priv, UMAC_MPD_CTRL);
-	reg |= MPD_EN;
-	bcmgenet_umac_writel(priv, reg, UMAC_MPD_CTRL);
+	hfb_ctrl_reg = bcmgenet_hfb_reg_readl(priv, HFB_CTRL);
+	reg = hfb_ctrl_reg | RBUF_ACPI_EN;
+	bcmgenet_hfb_reg_writel(priv, reg, HFB_CTRL);
 
 	/* Do not leave UniMAC in MPD mode only */
 	retries = bcmgenet_poll_wol_status(priv);
 	if (retries < 0) {
 		reg = bcmgenet_umac_readl(priv, UMAC_MPD_CTRL);
-		reg &= ~MPD_EN;
+		reg &= ~(MPD_EN | MPD_PW_EN);
 		bcmgenet_umac_writel(priv, reg, UMAC_MPD_CTRL);
+		bcmgenet_hfb_reg_writel(priv, hfb_ctrl_reg, HFB_CTRL);
 		return retries;
 	}
 
 	netif_dbg(priv, wol, dev, "MPD WOL-ready status set after %d msec\n",
 		  retries);
 
+	/* Disable phy status updates while suspending */
+	mutex_lock(&dev->phydev->lock);
+	dev->phydev->state = PHY_READY;
+	mutex_unlock(&dev->phydev->lock);
+
+	clk_prepare_enable(priv->clk_wol);
+
 	/* Enable CRC forward */
+	spin_lock_bh(&priv->reg_lock);
 	reg = bcmgenet_umac_readl(priv, UMAC_CMD);
 	priv->crc_fwd_en = 1;
 	reg |= CMD_CRC_FWD;
 
+	/* Can't suspend with WoL if MAC is still in reset */
+	if (reg & CMD_SW_RESET)
+		reg &= ~CMD_SW_RESET;
+
 	/* Receiver must be enabled for WOL MP detection */
 	reg |= CMD_RX_EN;
 	bcmgenet_umac_writel(priv, reg, UMAC_CMD);
+	spin_unlock_bh(&priv->reg_lock);
 
-	if (priv->hw_params->flags & GENET_HAS_EXT) {
-		reg = bcmgenet_ext_readl(priv, EXT_EXT_PWR_MGMT);
-		reg &= ~EXT_ENERGY_DET_MASK;
-		bcmgenet_ext_writel(priv, reg, EXT_EXT_PWR_MGMT);
-	}
+	reg = UMAC_IRQ_MPD_R;
+	if (hfb_ctrl_reg & RBUF_HFB_EN)
+		reg |=  UMAC_IRQ_HFB_SM | UMAC_IRQ_HFB_MM;
+
+	bcmgenet_intrl2_0_writel(priv, reg, INTRL2_CPU_MASK_CLEAR);
 
 	return 0;
 }
 
-void bcmgenet_wol_power_up_cfg(struct bcmgenet_priv *priv,
-			       enum bcmgenet_power_mode mode)
+int bcmgenet_wol_power_up_cfg(struct bcmgenet_priv *priv,
+			      enum bcmgenet_power_mode mode)
 {
+	struct net_device *dev = priv->dev;
 	u32 reg;
 
 	if (mode != GENET_POWER_WOL_MAGIC) {
 		netif_err(priv, wol, priv->dev, "invalid mode: %d\n", mode);
-		return;
+		return -EINVAL;
+	}
+
+	clk_disable_unprepare(priv->clk_wol);
+	priv->crc_fwd_en = 0;
+
+	bcmgenet_intrl2_0_writel(priv, UMAC_IRQ_WAKE_EVENT,
+				 INTRL2_CPU_MASK_SET);
+	if (bcmgenet_has_mdio_intr(priv))
+		bcmgenet_intrl2_0_writel(priv,
+					 UMAC_IRQ_MDIO_EVENT,
+					 INTRL2_CPU_MASK_CLEAR);
+
+	/* Disable Magic Packet Detection */
+	if (priv->wolopts & (WAKE_MAGIC | WAKE_MAGICSECURE)) {
+		reg = bcmgenet_umac_readl(priv, UMAC_MPD_CTRL);
+		if (!(reg & MPD_EN))
+			return -EPERM;	/* already reset so skip the rest */
+		reg &= ~(MPD_EN | MPD_PW_EN);
+		bcmgenet_umac_writel(priv, reg, UMAC_MPD_CTRL);
 	}
 
-	reg = bcmgenet_umac_readl(priv, UMAC_MPD_CTRL);
-	reg &= ~MPD_EN;
-	bcmgenet_umac_writel(priv, reg, UMAC_MPD_CTRL);
+	/* Disable ACPI mode */
+	reg = bcmgenet_hfb_reg_readl(priv, HFB_CTRL);
+	if (!(reg & RBUF_ACPI_EN))
+		return -EPERM;	/* already reset so skip the rest */
+	reg &= ~RBUF_ACPI_EN;
+	bcmgenet_hfb_reg_writel(priv, reg, HFB_CTRL);
 
 	/* Disable CRC Forward */
+	spin_lock_bh(&priv->reg_lock);
 	reg = bcmgenet_umac_readl(priv, UMAC_CMD);
 	reg &= ~CMD_CRC_FWD;
 	bcmgenet_umac_writel(priv, reg, UMAC_CMD);
-	priv->crc_fwd_en = 0;
+	spin_unlock_bh(&priv->reg_lock);
+
+	/* Resume link status tracking */
+	mutex_lock(&dev->phydev->lock);
+	if (dev->phydev->link)
+		dev->phydev->state = PHY_RUNNING;
+	else
+		dev->phydev->state = PHY_NOLINK;
+	mutex_unlock(&dev->phydev->lock);
+
+	return 0;
 }
diff --git a/drivers/net/ethernet/broadcom/genet/bcmmii.c b/drivers/net/ethernet/broadcom/genet/bcmmii.c
index 285676f8da6b..573e8b279e52 100644
--- a/drivers/net/ethernet/broadcom/genet/bcmmii.c
+++ b/drivers/net/ethernet/broadcom/genet/bcmmii.c
@@ -1,14 +1,11 @@
+// SPDX-License-Identifier: GPL-2.0-only
 /*
  * Broadcom GENET MDIO routines
  *
- * Copyright (c) 2014-2017 Broadcom
- *
- * This program is free software; you can redistribute it and/or modify
- * it under the terms of the GNU General Public License version 2 as
- * published by the Free Software Foundation.
+ * Copyright (c) 2014-2025 Broadcom
  */
 
-
+#include <linux/acpi.h>
 #include <linux/types.h>
 #include <linux/delay.h>
 #include <linux/wait.h>
@@ -24,60 +21,80 @@
 #include <linux/of_net.h>
 #include <linux/of_mdio.h>
 #include <linux/platform_data/bcmgenet.h>
+#include <linux/platform_data/mdio-bcm-unimac.h>
 
 #include "bcmgenet.h"
 
-/* read a value from the MII */
-static int bcmgenet_mii_read(struct mii_bus *bus, int phy_id, int location)
+static void bcmgenet_mac_config(struct net_device *dev)
 {
-	int ret;
-	struct net_device *dev = bus->priv;
 	struct bcmgenet_priv *priv = netdev_priv(dev);
-	u32 reg;
+	struct phy_device *phydev = dev->phydev;
+	u32 reg, cmd_bits = 0;
+	bool active;
 
-	bcmgenet_umac_writel(priv, (MDIO_RD | (phy_id << MDIO_PMD_SHIFT) |
-			     (location << MDIO_REG_SHIFT)), UMAC_MDIO_CMD);
-	/* Start MDIO transaction*/
-	reg = bcmgenet_umac_readl(priv, UMAC_MDIO_CMD);
-	reg |= MDIO_START_BUSY;
-	bcmgenet_umac_writel(priv, reg, UMAC_MDIO_CMD);
-	wait_event_timeout(priv->wq,
-			   !(bcmgenet_umac_readl(priv, UMAC_MDIO_CMD)
-			   & MDIO_START_BUSY),
-			   HZ / 100);
-	ret = bcmgenet_umac_readl(priv, UMAC_MDIO_CMD);
+	/* speed */
+	if (phydev->speed == SPEED_1000)
+		cmd_bits = CMD_SPEED_1000;
+	else if (phydev->speed == SPEED_100)
+		cmd_bits = CMD_SPEED_100;
+	else
+		cmd_bits = CMD_SPEED_10;
+	cmd_bits <<= CMD_SPEED_SHIFT;
 
-	/* Some broken devices are known not to release the line during
-	 * turn-around, e.g: Broadcom BCM53125 external switches, so check for
-	 * that condition here and ignore the MDIO controller read failure
-	 * indication.
-	 */
-	if (!(bus->phy_ignore_ta_mask & 1 << phy_id) && (ret & MDIO_READ_FAIL))
-		return -EIO;
+	/* duplex */
+	if (phydev->duplex != DUPLEX_FULL) {
+		cmd_bits |= CMD_HD_EN |
+			CMD_RX_PAUSE_IGNORE | CMD_TX_PAUSE_IGNORE;
+	} else {
+		/* pause capability defaults to Symmetric */
+		if (priv->autoneg_pause) {
+			bool tx_pause = 0, rx_pause = 0;
 
-	return ret & 0xffff;
-}
+			if (phydev->autoneg)
+				phy_get_pause(phydev, &tx_pause, &rx_pause);
 
-/* write a value to the MII */
-static int bcmgenet_mii_write(struct mii_bus *bus, int phy_id,
-			      int location, u16 val)
-{
-	struct net_device *dev = bus->priv;
-	struct bcmgenet_priv *priv = netdev_priv(dev);
-	u32 reg;
+			if (!tx_pause)
+				cmd_bits |= CMD_TX_PAUSE_IGNORE;
+			if (!rx_pause)
+				cmd_bits |= CMD_RX_PAUSE_IGNORE;
+		}
 
-	bcmgenet_umac_writel(priv, (MDIO_WR | (phy_id << MDIO_PMD_SHIFT) |
-			     (location << MDIO_REG_SHIFT) | (0xffff & val)),
-			     UMAC_MDIO_CMD);
-	reg = bcmgenet_umac_readl(priv, UMAC_MDIO_CMD);
-	reg |= MDIO_START_BUSY;
-	bcmgenet_umac_writel(priv, reg, UMAC_MDIO_CMD);
-	wait_event_timeout(priv->wq,
-			   !(bcmgenet_umac_readl(priv, UMAC_MDIO_CMD) &
-			   MDIO_START_BUSY),
-			   HZ / 100);
+		/* Manual override */
+		if (!priv->rx_pause)
+			cmd_bits |= CMD_RX_PAUSE_IGNORE;
+		if (!priv->tx_pause)
+			cmd_bits |= CMD_TX_PAUSE_IGNORE;
+	}
 
-	return 0;
+	/* Program UMAC and RGMII block based on established
+	 * link speed, duplex, and pause. The speed set in
+	 * umac->cmd tell RGMII block which clock to use for
+	 * transmit -- 25MHz(100Mbps) or 125MHz(1Gbps).
+	 * Receive clock is provided by the PHY.
+	 */
+	reg = bcmgenet_ext_readl(priv, EXT_RGMII_OOB_CTRL);
+	reg |= RGMII_LINK;
+	bcmgenet_ext_writel(priv, reg, EXT_RGMII_OOB_CTRL);
+
+	spin_lock_bh(&priv->reg_lock);
+	reg = bcmgenet_umac_readl(priv, UMAC_CMD);
+	reg &= ~((CMD_SPEED_MASK << CMD_SPEED_SHIFT) |
+		       CMD_HD_EN |
+		       CMD_RX_PAUSE_IGNORE | CMD_TX_PAUSE_IGNORE);
+	reg |= cmd_bits;
+	if (reg & CMD_SW_RESET) {
+		reg &= ~CMD_SW_RESET;
+		bcmgenet_umac_writel(priv, reg, UMAC_CMD);
+		udelay(2);
+		reg |= CMD_TX_EN | CMD_RX_EN;
+	}
+	bcmgenet_umac_writel(priv, reg, UMAC_CMD);
+	spin_unlock_bh(&priv->reg_lock);
+
+	active = phy_init_eee(phydev, 0) >= 0;
+	bcmgenet_eee_enable_set(dev,
+				priv->eee.eee_enabled && active,
+				priv->eee.tx_lpi_enabled);
 }
 
 /* setup netdev link state when PHY link status change and
@@ -86,78 +103,15 @@ static int bcmgenet_mii_write(struct mii_bus *bus, int phy_id,
 void bcmgenet_mii_setup(struct net_device *dev)
 {
 	struct bcmgenet_priv *priv = netdev_priv(dev);
-	struct phy_device *phydev = priv->phydev;
-	u32 reg, cmd_bits = 0;
-	bool status_changed = false;
-
-	if (priv->old_link != phydev->link) {
-		status_changed = true;
-		priv->old_link = phydev->link;
-	}
+	struct phy_device *phydev = dev->phydev;
+	u32 reg;
 
 	if (phydev->link) {
-		/* check speed/duplex/pause changes */
-		if (priv->old_speed != phydev->speed) {
-			status_changed = true;
-			priv->old_speed = phydev->speed;
-		}
-
-		if (priv->old_duplex != phydev->duplex) {
-			status_changed = true;
-			priv->old_duplex = phydev->duplex;
-		}
-
-		if (priv->old_pause != phydev->pause) {
-			status_changed = true;
-			priv->old_pause = phydev->pause;
-		}
-
-		/* done if nothing has changed */
-		if (!status_changed)
-			return;
-
-		/* speed */
-		if (phydev->speed == SPEED_1000)
-			cmd_bits = UMAC_SPEED_1000;
-		else if (phydev->speed == SPEED_100)
-			cmd_bits = UMAC_SPEED_100;
-		else
-			cmd_bits = UMAC_SPEED_10;
-		cmd_bits <<= CMD_SPEED_SHIFT;
-
-		/* duplex */
-		if (phydev->duplex != DUPLEX_FULL)
-			cmd_bits |= CMD_HD_EN;
-
-		/* pause capability */
-		if (!phydev->pause)
-			cmd_bits |= CMD_RX_PAUSE_IGNORE | CMD_TX_PAUSE_IGNORE;
-
-		/*
-		 * Program UMAC and RGMII block based on established
-		 * link speed, duplex, and pause. The speed set in
-		 * umac->cmd tell RGMII block which clock to use for
-		 * transmit -- 25MHz(100Mbps) or 125MHz(1Gbps).
-		 * Receive clock is provided by the PHY.
-		 */
+		bcmgenet_mac_config(dev);
+	} else {
 		reg = bcmgenet_ext_readl(priv, EXT_RGMII_OOB_CTRL);
-		reg &= ~OOB_DISABLE;
-		reg |= RGMII_LINK;
+		reg &= ~RGMII_LINK;
 		bcmgenet_ext_writel(priv, reg, EXT_RGMII_OOB_CTRL);
-
-		reg = bcmgenet_umac_readl(priv, UMAC_CMD);
-		reg &= ~((CMD_SPEED_MASK << CMD_SPEED_SHIFT) |
-			       CMD_HD_EN |
-			       CMD_RX_PAUSE_IGNORE | CMD_TX_PAUSE_IGNORE);
-		reg |= cmd_bits;
-		bcmgenet_umac_writel(priv, reg, UMAC_CMD);
-	} else {
-		/* done if nothing has changed */
-		if (!status_changed)
-			return;
-
-		/* needed for MoCA fixed PHY to reflect correct link status */
-		netif_carrier_off(dev);
 	}
 
 	phy_print_status(phydev);
@@ -167,26 +121,31 @@ void bcmgenet_mii_setup(struct net_device *dev)
 static int bcmgenet_fixed_phy_link_update(struct net_device *dev,
 					  struct fixed_phy_status *status)
 {
-	if (dev && dev->phydev && status)
-		status->link = dev->phydev->link;
+	struct bcmgenet_priv *priv;
+	u32 reg;
+
+	if (dev && dev->phydev && status) {
+		priv = netdev_priv(dev);
+		reg = bcmgenet_umac_readl(priv, UMAC_MODE);
+		status->link = !!(reg & MODE_LINK_STATUS);
+	}
 
 	return 0;
 }
 
-/* Perform a voluntary PHY software reset, since the EPHY is very finicky about
- * not doing it and will start corrupting packets
- */
-void bcmgenet_mii_reset(struct net_device *dev)
+void bcmgenet_phy_pause_set(struct net_device *dev, bool rx, bool tx)
 {
-	struct bcmgenet_priv *priv = netdev_priv(dev);
+	struct phy_device *phydev = dev->phydev;
 
-	if (GENET_IS_V4(priv))
-		return;
+	linkmode_mod_bit(ETHTOOL_LINK_MODE_Pause_BIT, phydev->advertising, rx);
+	linkmode_mod_bit(ETHTOOL_LINK_MODE_Asym_Pause_BIT, phydev->advertising,
+			 rx | tx);
+	phy_start_aneg(phydev);
 
-	if (priv->phydev) {
-		phy_init_hw(priv->phydev);
-		phy_start_aneg(priv->phydev);
-	}
+	mutex_lock(&phydev->lock);
+	if (phydev->link)
+		bcmgenet_mac_config(dev);
+	mutex_unlock(&phydev->lock);
 }
 
 void bcmgenet_phy_power_set(struct net_device *dev, bool enable)
@@ -195,24 +154,30 @@ void bcmgenet_phy_power_set(struct net_device *dev, bool enable)
 	u32 reg = 0;
 
 	/* EXT_GPHY_CTRL is only valid for GENETv4 and onward */
-	if (GENET_IS_V4(priv)) {
+	if (GENET_IS_V4(priv) || bcmgenet_has_ephy_16nm(priv)) {
 		reg = bcmgenet_ext_readl(priv, EXT_GPHY_CTRL);
 		if (enable) {
 			reg &= ~EXT_CK25_DIS;
 			bcmgenet_ext_writel(priv, reg, EXT_GPHY_CTRL);
 			mdelay(1);
 
-			reg &= ~(EXT_CFG_IDDQ_BIAS | EXT_CFG_PWR_DOWN);
+			reg &= ~(EXT_CFG_IDDQ_BIAS | EXT_CFG_PWR_DOWN |
+				 EXT_CFG_IDDQ_GLOBAL_PWR);
 			reg |= EXT_GPHY_RESET;
 			bcmgenet_ext_writel(priv, reg, EXT_GPHY_CTRL);
 			mdelay(1);
 
 			reg &= ~EXT_GPHY_RESET;
 		} else {
+			reg |= EXT_GPHY_RESET;
+			bcmgenet_ext_writel(priv, reg, EXT_GPHY_CTRL);
+			mdelay(1);
+
 			reg |= EXT_CFG_IDDQ_BIAS | EXT_CFG_PWR_DOWN |
-			       EXT_GPHY_RESET;
+			       EXT_CFG_IDDQ_GLOBAL_PWR;
 			bcmgenet_ext_writel(priv, reg, EXT_GPHY_CTRL);
 			mdelay(1);
+
 			reg |= EXT_CK25_DIS;
 		}
 		bcmgenet_ext_writel(priv, reg, EXT_GPHY_CTRL);
@@ -224,38 +189,25 @@ void bcmgenet_phy_power_set(struct net_device *dev, bool enable)
 
 static void bcmgenet_moca_phy_setup(struct bcmgenet_priv *priv)
 {
-	u32 reg;
-
-	if (!GENET_IS_V5(priv)) {
-		/* Speed settings are set in bcmgenet_mii_setup() */
-		reg = bcmgenet_sys_readl(priv, SYS_PORT_CTRL);
-		reg |= LED_ACT_SOURCE_MAC;
-		bcmgenet_sys_writel(priv, reg, SYS_PORT_CTRL);
-	}
-
-	if (priv->hw_params->flags & GENET_HAS_MOCA_LINK_DET)
-		fixed_phy_set_link_update(priv->phydev,
+	if (bcmgenet_has_moca_link_det(priv))
+		fixed_phy_set_link_update(priv->dev->phydev,
 					  bcmgenet_fixed_phy_link_update);
 }
 
-int bcmgenet_mii_config(struct net_device *dev)
+int bcmgenet_mii_config(struct net_device *dev, bool init)
 {
 	struct bcmgenet_priv *priv = netdev_priv(dev);
-	struct phy_device *phydev = priv->phydev;
+	struct phy_device *phydev = dev->phydev;
 	struct device *kdev = &priv->pdev->dev;
 	const char *phy_name = NULL;
 	u32 id_mode_dis = 0;
 	u32 port_ctrl;
 	u32 reg;
 
-	priv->ext_phy = !priv->internal_phy &&
-			(priv->phy_interface != PHY_INTERFACE_MODE_MOCA);
-
-	if (priv->internal_phy)
-		priv->phy_interface = PHY_INTERFACE_MODE_NA;
-
 	switch (priv->phy_interface) {
-	case PHY_INTERFACE_MODE_NA:
+	case PHY_INTERFACE_MODE_INTERNAL:
+		phy_name = "internal PHY";
+		fallthrough;
 	case PHY_INTERFACE_MODE_MOCA:
 		/* Irrespective of the actually configured PHY speed (100 or
 		 * 1000) GENETv4 only has an internal GPHY so we will just end
@@ -267,21 +219,18 @@ int bcmgenet_mii_config(struct net_device *dev)
 		else
 			port_ctrl = PORT_MODE_INT_EPHY;
 
-		bcmgenet_sys_writel(priv, port_ctrl, SYS_PORT_CTRL);
-
-		if (priv->internal_phy) {
-			phy_name = "internal PHY";
-		} else if (priv->phy_interface == PHY_INTERFACE_MODE_MOCA) {
+		if (!phy_name) {
 			phy_name = "MoCA";
+			if (!GENET_IS_V5(priv))
+				port_ctrl |= LED_ACT_SOURCE_MAC;
 			bcmgenet_moca_phy_setup(priv);
 		}
 		break;
 
 	case PHY_INTERFACE_MODE_MII:
 		phy_name = "external MII";
-		phydev->supported &= PHY_BASIC_FEATURES;
-		bcmgenet_sys_writel(priv,
-				    PORT_MODE_EXT_EPHY, SYS_PORT_CTRL);
+		phy_set_max_speed(phydev, SPEED_100);
+		port_ctrl = PORT_MODE_EXT_EPHY;
 		break;
 
 	case PHY_INTERFACE_MODE_REVMII:
@@ -291,46 +240,64 @@ int bcmgenet_mii_config(struct net_device *dev)
 		 * capabilities, use that knowledge to also configure the
 		 * Reverse MII interface correctly.
 		 */
-		if ((priv->phydev->supported & PHY_BASIC_FEATURES) ==
-				PHY_BASIC_FEATURES)
-			port_ctrl = PORT_MODE_EXT_RVMII_25;
-		else
+		if (linkmode_test_bit(ETHTOOL_LINK_MODE_1000baseT_Full_BIT,
+				      dev->phydev->supported))
 			port_ctrl = PORT_MODE_EXT_RVMII_50;
-		bcmgenet_sys_writel(priv, port_ctrl, SYS_PORT_CTRL);
+		else
+			port_ctrl = PORT_MODE_EXT_RVMII_25;
 		break;
 
 	case PHY_INTERFACE_MODE_RGMII:
 		/* RGMII_NO_ID: TXC transitions at the same time as TXD
 		 *		(requires PCB or receiver-side delay)
-		 * RGMII:	Add 2ns delay on TXC (90 degree shift)
 		 *
 		 * ID is implicitly disabled for 100Mbps (RG)MII operation.
 		 */
+		phy_name = "external RGMII (no delay)";
 		id_mode_dis = BIT(16);
-		/* fall through */
+		port_ctrl = PORT_MODE_EXT_GPHY;
+		break;
+
 	case PHY_INTERFACE_MODE_RGMII_TXID:
-		if (id_mode_dis)
-			phy_name = "external RGMII (no delay)";
-		else
-			phy_name = "external RGMII (TX delay)";
-		bcmgenet_sys_writel(priv,
-				    PORT_MODE_EXT_GPHY, SYS_PORT_CTRL);
+		/* RGMII_TXID:	Add 2ns delay on TXC (90 degree shift) */
+		phy_name = "external RGMII (TX delay)";
+		port_ctrl = PORT_MODE_EXT_GPHY;
+		break;
+
+	case PHY_INTERFACE_MODE_RGMII_RXID:
+		phy_name = "external RGMII (RX delay)";
+		port_ctrl = PORT_MODE_EXT_GPHY;
 		break;
 	default:
 		dev_err(kdev, "unknown phy mode: %d\n", priv->phy_interface);
 		return -EINVAL;
 	}
 
+	bcmgenet_sys_writel(priv, port_ctrl, SYS_PORT_CTRL);
+
+	priv->ext_phy = !priv->internal_phy &&
+			(priv->phy_interface != PHY_INTERFACE_MODE_MOCA);
+
 	/* This is an external PHY (xMII), so we need to enable the RGMII
-	 * block for the interface to work
+	 * block for the interface to work, unconditionally clear the
+	 * Out-of-band disable since we do not need it.
 	 */
+	mutex_lock(&phydev->lock);
+	reg = bcmgenet_ext_readl(priv, EXT_RGMII_OOB_CTRL);
+	reg &= ~OOB_DISABLE;
 	if (priv->ext_phy) {
-		reg = bcmgenet_ext_readl(priv, EXT_RGMII_OOB_CTRL);
-		reg |= RGMII_MODE_EN | id_mode_dis;
-		bcmgenet_ext_writel(priv, reg, EXT_RGMII_OOB_CTRL);
+		reg &= ~ID_MODE_DIS;
+		reg |= id_mode_dis;
+		if (GENET_IS_V1(priv) || GENET_IS_V2(priv) || GENET_IS_V3(priv))
+			reg |= RGMII_MODE_EN_V123;
+		else
+			reg |= RGMII_MODE_EN;
 	}
+	bcmgenet_ext_writel(priv, reg, EXT_RGMII_OOB_CTRL);
+	mutex_unlock(&phydev->lock);
 
-	dev_info_once(kdev, "configuring instance for %s\n", phy_name);
+	if (init)
+		dev_info(kdev, "configuring instance for %s\n", phy_name);
 
 	return 0;
 }
@@ -338,161 +305,263 @@ int bcmgenet_mii_config(struct net_device *dev)
 int bcmgenet_mii_probe(struct net_device *dev)
 {
 	struct bcmgenet_priv *priv = netdev_priv(dev);
-	struct device_node *dn = priv->pdev->dev.of_node;
+	struct device *kdev = &priv->pdev->dev;
+	struct device_node *dn = kdev->of_node;
+	phy_interface_t phy_iface = priv->phy_interface;
 	struct phy_device *phydev;
-	u32 phy_flags;
+	u32 phy_flags = PHY_BRCM_AUTO_PWRDWN_ENABLE |
+			PHY_BRCM_DIS_TXCRXC_NOENRGY |
+			PHY_BRCM_IDDQ_SUSPEND;
 	int ret;
 
 	/* Communicate the integrated PHY revision */
-	phy_flags = priv->gphy_rev;
-
-	/* Initialize link state variables that bcmgenet_mii_setup() uses */
-	priv->old_link = -1;
-	priv->old_speed = -1;
-	priv->old_duplex = -1;
-	priv->old_pause = -1;
+	if (priv->internal_phy)
+		phy_flags = priv->gphy_rev;
+
+	/* This is an ugly quirk but we have not been correctly interpreting
+	 * the phy_interface values and we have done that across different
+	 * drivers, so at least we are consistent in our mistakes.
+	 *
+	 * When the Generic PHY driver is in use either the PHY has been
+	 * strapped or programmed correctly by the boot loader so we should
+	 * stick to our incorrect interpretation since we have validated it.
+	 *
+	 * Now when a dedicated PHY driver is in use, we need to reverse the
+	 * meaning of the phy_interface_mode values to something that the PHY
+	 * driver will interpret and act on such that we have two mistakes
+	 * canceling themselves so to speak. We only do this for the two
+	 * modes that GENET driver officially supports on Broadcom STB chips:
+	 * PHY_INTERFACE_MODE_RGMII and PHY_INTERFACE_MODE_RGMII_TXID. Other
+	 * modes are not *officially* supported with the boot loader and the
+	 * scripted environment generating Device Tree blobs for those
+	 * platforms.
+	 *
+	 * Note that internal PHY, MoCA and fixed-link configurations are not
+	 * affected because they use different phy_interface_t values or the
+	 * Generic PHY driver.
+	 */
+	switch (priv->phy_interface) {
+	case PHY_INTERFACE_MODE_RGMII:
+		phy_iface = PHY_INTERFACE_MODE_RGMII_ID;
+		break;
+	case PHY_INTERFACE_MODE_RGMII_TXID:
+		phy_iface = PHY_INTERFACE_MODE_RGMII_RXID;
+		break;
+	default:
+		break;
+	}
 
 	if (dn) {
 		phydev = of_phy_connect(dev, priv->phy_dn, bcmgenet_mii_setup,
-					phy_flags, priv->phy_interface);
+					phy_flags, phy_iface);
 		if (!phydev) {
 			pr_err("could not attach to PHY\n");
 			return -ENODEV;
 		}
 	} else {
-		phydev = priv->phydev;
+		if (has_acpi_companion(kdev)) {
+			char mdio_bus_id[MII_BUS_ID_SIZE];
+			struct mii_bus *unimacbus;
+
+			snprintf(mdio_bus_id, MII_BUS_ID_SIZE, "%s-%d",
+				 UNIMAC_MDIO_DRV_NAME, priv->pdev->id);
+
+			unimacbus = mdio_find_bus(mdio_bus_id);
+			if (!unimacbus) {
+				pr_err("Unable to find mii\n");
+				return -ENODEV;
+			}
+			phydev = phy_find_first(unimacbus);
+			put_device(&unimacbus->dev);
+			if (!phydev) {
+				pr_err("Unable to find PHY\n");
+				return -ENODEV;
+			}
+		} else {
+			phydev = dev->phydev;
+		}
 		phydev->dev_flags = phy_flags;
 
 		ret = phy_connect_direct(dev, phydev, bcmgenet_mii_setup,
-					 priv->phy_interface);
+					 phy_iface);
 		if (ret) {
 			pr_err("could not attach to PHY\n");
 			return -ENODEV;
 		}
 	}
 
-	priv->phydev = phydev;
-
 	/* Configure port multiplexer based on what the probed PHY device since
 	 * reading the 'max-speed' property determines the maximum supported
 	 * PHY speed which is needed for bcmgenet_mii_config() to configure
 	 * things appropriately.
 	 */
-	ret = bcmgenet_mii_config(dev);
+	ret = bcmgenet_mii_config(dev, true);
 	if (ret) {
-		phy_disconnect(priv->phydev);
+		phy_disconnect(dev->phydev);
 		return ret;
 	}
 
-	phydev->advertising = phydev->supported;
-
 	/* The internal PHY has its link interrupts routed to the
-	 * Ethernet MAC ISRs
+	 * Ethernet MAC ISRs. On GENETv5 there is a hardware issue
+	 * that prevents the signaling of link UP interrupts when
+	 * the link operates at 10Mbps, so fallback to polling for
+	 * those versions of GENET.
 	 */
-	if (priv->internal_phy)
-		priv->phydev->irq = PHY_IGNORE_INTERRUPT;
+	if (priv->internal_phy && !GENET_IS_V5(priv))
+		dev->phydev->irq = PHY_MAC_INTERRUPT;
+
+	/* Indicate that the MAC is responsible for PHY PM */
+	dev->phydev->mac_managed_pm = true;
 
 	return 0;
 }
 
-/* Workaround for integrated BCM7xxx Gigabit PHYs which have a problem with
- * their internal MDIO management controller making them fail to successfully
- * be read from or written to for the first transaction.  We insert a dummy
- * BMSR read here to make sure that phy_get_device() and get_phy_id() can
- * correctly read the PHY MII_PHYSID1/2 registers and successfully register a
- * PHY device for this peripheral.
- *
- * Once the PHY driver is registered, we can workaround subsequent reads from
- * there (e.g: during system-wide power management).
- *
- * bus->reset is invoked before mdiobus_scan during mdiobus_register and is
- * therefore the right location to stick that workaround. Since we do not want
- * to read from non-existing PHYs, we either use bus->phy_mask or do a manual
- * Device Tree scan to limit the search area.
- */
-static int bcmgenet_mii_bus_reset(struct mii_bus *bus)
+static struct device_node *bcmgenet_mii_of_find_mdio(struct bcmgenet_priv *priv)
 {
-	struct net_device *dev = bus->priv;
-	struct bcmgenet_priv *priv = netdev_priv(dev);
-	struct device_node *np = priv->mdio_dn;
-	struct device_node *child = NULL;
-	u32 read_mask = 0;
-	int addr = 0;
+	struct device_node *dn = priv->pdev->dev.of_node;
+	struct device *kdev = &priv->pdev->dev;
+	char *compat;
 
-	if (!np) {
-		read_mask = 1 << priv->phy_addr;
-	} else {
-		for_each_available_child_of_node(np, child) {
-			addr = of_mdio_parse_addr(&dev->dev, child);
-			if (addr < 0)
-				continue;
+	compat = kasprintf(GFP_KERNEL, "brcm,genet-mdio-v%d", priv->version);
+	if (!compat)
+		return NULL;
 
-			read_mask |= 1 << addr;
-		}
+	priv->mdio_dn = of_get_compatible_child(dn, compat);
+	kfree(compat);
+	if (!priv->mdio_dn) {
+		dev_err(kdev, "unable to find MDIO bus node\n");
+		return NULL;
 	}
 
-	for (addr = 0; addr < PHY_MAX_ADDR; addr++) {
-		if (read_mask & 1 << addr) {
-			dev_dbg(&dev->dev, "Workaround for PHY @ %d\n", addr);
-			mdiobus_read(bus, addr, MII_BMSR);
-		}
+	return priv->mdio_dn;
+}
+
+static void bcmgenet_mii_pdata_init(struct bcmgenet_priv *priv,
+				    struct unimac_mdio_pdata *ppd)
+{
+	struct device *kdev = &priv->pdev->dev;
+	struct bcmgenet_platform_data *pd = kdev->platform_data;
+
+	if (pd->phy_interface != PHY_INTERFACE_MODE_MOCA && pd->mdio_enabled) {
+		/*
+		 * Internal or external PHY with MDIO access
+		 */
+		if (pd->phy_address >= 0 && pd->phy_address < PHY_MAX_ADDR)
+			ppd->phy_mask = 1 << pd->phy_address;
+		else
+			ppd->phy_mask = 0;
 	}
+}
+
+static int bcmgenet_mii_wait(void *wait_func_data)
+{
+	struct bcmgenet_priv *priv = wait_func_data;
 
+	wait_event_timeout(priv->wq,
+			   !(bcmgenet_umac_readl(priv, UMAC_MDIO_CMD)
+			   & MDIO_START_BUSY),
+			   HZ / 100);
 	return 0;
 }
 
-static int bcmgenet_mii_alloc(struct bcmgenet_priv *priv)
+static int bcmgenet_mii_register(struct bcmgenet_priv *priv)
 {
-	struct mii_bus *bus;
+	struct platform_device *pdev = priv->pdev;
+	struct bcmgenet_platform_data *pdata = pdev->dev.platform_data;
+	struct device_node *dn = pdev->dev.of_node;
+	struct unimac_mdio_pdata ppd;
+	struct platform_device *ppdev;
+	struct resource *pres, res;
+	int id, ret;
+
+	pres = platform_get_resource(pdev, IORESOURCE_MEM, 0);
+	if (!pres) {
+		dev_err(&pdev->dev, "Invalid resource\n");
+		return -EINVAL;
+	}
+	memset(&res, 0, sizeof(res));
+	memset(&ppd, 0, sizeof(ppd));
+
+	ppd.wait_func = bcmgenet_mii_wait;
+	ppd.wait_func_data = priv;
+	ppd.bus_name = "bcmgenet MII bus";
+	/* Pass a reference to our "main" clock which is used for MDIO
+	 * transfers
+	 */
+	ppd.clk = priv->clk;
 
-	if (priv->mii_bus)
-		return 0;
+	/* Unimac MDIO bus controller starts at UniMAC offset + MDIO_CMD
+	 * and is 2 * 32-bits word long, 8 bytes total.
+	 */
+	res.start = pres->start + GENET_UMAC_OFF + UMAC_MDIO_CMD;
+	res.end = res.start + 8;
+	res.flags = IORESOURCE_MEM;
+
+	if (dn)
+		id = of_alias_get_id(dn, "eth");
+	else
+		id = pdev->id;
 
-	priv->mii_bus = mdiobus_alloc();
-	if (!priv->mii_bus) {
-		pr_err("failed to allocate\n");
+	ppdev = platform_device_alloc(UNIMAC_MDIO_DRV_NAME, id);
+	if (!ppdev)
 		return -ENOMEM;
-	}
 
-	bus = priv->mii_bus;
-	bus->priv = priv->dev;
-	bus->name = "bcmgenet MII bus";
-	bus->parent = &priv->pdev->dev;
-	bus->read = bcmgenet_mii_read;
-	bus->write = bcmgenet_mii_write;
-	bus->reset = bcmgenet_mii_bus_reset;
-	snprintf(bus->id, MII_BUS_ID_SIZE, "%s-%d",
-		 priv->pdev->name, priv->pdev->id);
+	/* Retain this platform_device pointer for later cleanup */
+	priv->mii_pdev = ppdev;
+	ppdev->dev.parent = &pdev->dev;
+	if (dn)
+		ppdev->dev.of_node = bcmgenet_mii_of_find_mdio(priv);
+	else if (pdata)
+		bcmgenet_mii_pdata_init(priv, &ppd);
+	else
+		ppd.phy_mask = ~0;
+
+	ret = platform_device_add_resources(ppdev, &res, 1);
+	if (ret)
+		goto out;
+
+	ret = platform_device_add_data(ppdev, &ppd, sizeof(ppd));
+	if (ret)
+		goto out;
+
+	ret = platform_device_add(ppdev);
+	if (ret)
+		goto out;
 
 	return 0;
+out:
+	platform_device_put(ppdev);
+	return ret;
 }
 
-static int bcmgenet_mii_of_init(struct bcmgenet_priv *priv)
+static int bcmgenet_phy_interface_init(struct bcmgenet_priv *priv)
 {
-	struct device_node *dn = priv->pdev->dev.of_node;
 	struct device *kdev = &priv->pdev->dev;
-	const char *phy_mode_str = NULL;
-	struct phy_device *phydev = NULL;
-	char *compat;
-	int phy_mode;
-	int ret;
-
-	compat = kasprintf(GFP_KERNEL, "brcm,genet-mdio-v%d", priv->version);
-	if (!compat)
-		return -ENOMEM;
+	int phy_mode = device_get_phy_mode(kdev);
 
-	priv->mdio_dn = of_find_compatible_node(dn, NULL, compat);
-	kfree(compat);
-	if (!priv->mdio_dn) {
-		dev_err(kdev, "unable to find MDIO bus node\n");
-		return -ENODEV;
+	if (phy_mode < 0) {
+		dev_err(kdev, "invalid PHY mode property\n");
+		return phy_mode;
 	}
 
-	ret = of_mdiobus_register(priv->mii_bus, priv->mdio_dn);
-	if (ret) {
-		dev_err(kdev, "failed to register MDIO bus\n");
-		return ret;
-	}
+	priv->phy_interface = phy_mode;
+
+	/* We need to specifically look up whether this PHY interface is
+	 * internal or not *before* we even try to probe the PHY driver
+	 * over MDIO as we may have shut down the internal PHY for power
+	 * saving purposes.
+	 */
+	if (priv->phy_interface == PHY_INTERFACE_MODE_INTERNAL)
+		priv->internal_phy = true;
+
+	return 0;
+}
+
+static int bcmgenet_mii_of_init(struct bcmgenet_priv *priv)
+{
+	struct device_node *dn = priv->pdev->dev.of_node;
+	struct phy_device *phydev;
+	int ret;
 
 	/* Fetch the PHY phandle */
 	priv->phy_dn = of_parse_phandle(dn, "phy-handle", 0);
@@ -509,27 +578,12 @@ static int bcmgenet_mii_of_init(struct bcmgenet_priv *priv)
 	}
 
 	/* Get the link mode */
-	phy_mode = of_get_phy_mode(dn);
-	priv->phy_interface = phy_mode;
-
-	/* We need to specifically look up whether this PHY interface is internal
-	 * or not *before* we even try to probe the PHY driver over MDIO as we
-	 * may have shut down the internal PHY for power saving purposes.
-	 */
-	if (phy_mode < 0) {
-		ret = of_property_read_string(dn, "phy-mode", &phy_mode_str);
-		if (ret < 0) {
-			dev_err(kdev, "invalid PHY mode property\n");
-			return ret;
-		}
-
-		priv->phy_interface = PHY_INTERFACE_MODE_NA;
-		if (!strcasecmp(phy_mode_str, "internal"))
-			priv->internal_phy = true;
-	}
+	ret = bcmgenet_phy_interface_init(priv);
+	if (ret)
+		return ret;
 
 	/* Make sure we initialize MoCA PHYs with a link down */
-	if (phy_mode == PHY_INTERFACE_MODE_MOCA) {
+	if (priv->phy_interface == PHY_INTERFACE_MODE_MOCA) {
 		phydev = of_phy_find_device(dn);
 		if (phydev) {
 			phydev->link = 0;
@@ -544,34 +598,24 @@ static int bcmgenet_mii_pd_init(struct bcmgenet_priv *priv)
 {
 	struct device *kdev = &priv->pdev->dev;
 	struct bcmgenet_platform_data *pd = kdev->platform_data;
-	struct mii_bus *mdio = priv->mii_bus;
+	char phy_name[MII_BUS_ID_SIZE + 3];
+	char mdio_bus_id[MII_BUS_ID_SIZE];
 	struct phy_device *phydev;
-	int ret;
+
+	snprintf(mdio_bus_id, MII_BUS_ID_SIZE, "%s-%d",
+		 UNIMAC_MDIO_DRV_NAME, priv->pdev->id);
 
 	if (pd->phy_interface != PHY_INTERFACE_MODE_MOCA && pd->mdio_enabled) {
+		snprintf(phy_name, MII_BUS_ID_SIZE, PHY_ID_FMT,
+			 mdio_bus_id, pd->phy_address);
+
 		/*
 		 * Internal or external PHY with MDIO access
 		 */
-		if (pd->phy_address >= 0 && pd->phy_address < PHY_MAX_ADDR)
-			mdio->phy_mask = ~(1 << pd->phy_address);
-		else
-			mdio->phy_mask = 0;
-
-		ret = mdiobus_register(mdio);
-		if (ret) {
-			dev_err(kdev, "failed to register MDIO bus\n");
-			return ret;
-		}
-
-		if (pd->phy_address >= 0 && pd->phy_address < PHY_MAX_ADDR)
-			phydev = mdiobus_get_phy(mdio, pd->phy_address);
-		else
-			phydev = phy_find_first(mdio);
-
-		if (!phydev) {
+		phydev = phy_attach(priv->dev, phy_name, pd->phy_interface);
+		if (IS_ERR(phydev)) {
 			dev_err(kdev, "failed to register PHY device\n");
-			mdiobus_unregister(mdio);
-			return -ENODEV;
+			return PTR_ERR(phydev);
 		}
 	} else {
 		/*
@@ -586,10 +630,10 @@ static int bcmgenet_mii_pd_init(struct bcmgenet_priv *priv)
 			.asym_pause = 0,
 		};
 
-		phydev = fixed_phy_register(PHY_POLL, &fphy_status, -1, NULL);
-		if (!phydev || IS_ERR(phydev)) {
+		phydev = fixed_phy_register(&fphy_status, NULL);
+		if (IS_ERR(phydev)) {
 			dev_err(kdev, "failed to register fixed PHY device\n");
-			return -ENODEV;
+			return PTR_ERR(phydev);
 		}
 
 		/* Make sure we initialize MoCA PHYs with a link down */
@@ -597,7 +641,6 @@ static int bcmgenet_mii_pd_init(struct bcmgenet_priv *priv)
 
 	}
 
-	priv->phydev = phydev;
 	priv->phy_interface = pd->phy_interface;
 
 	return 0;
@@ -605,10 +648,13 @@ static int bcmgenet_mii_pd_init(struct bcmgenet_priv *priv)
 
 static int bcmgenet_mii_bus_init(struct bcmgenet_priv *priv)
 {
-	struct device_node *dn = priv->pdev->dev.of_node;
+	struct device *kdev = &priv->pdev->dev;
+	struct device_node *dn = kdev->of_node;
 
 	if (dn)
 		return bcmgenet_mii_of_init(priv);
+	else if (has_acpi_companion(kdev))
+		return bcmgenet_phy_interface_init(priv);
 	else
 		return bcmgenet_mii_pd_init(priv);
 }
@@ -616,10 +662,9 @@ static int bcmgenet_mii_bus_init(struct bcmgenet_priv *priv)
 int bcmgenet_mii_init(struct net_device *dev)
 {
 	struct bcmgenet_priv *priv = netdev_priv(dev);
-	struct device_node *dn = priv->pdev->dev.of_node;
 	int ret;
 
-	ret = bcmgenet_mii_alloc(priv);
+	ret = bcmgenet_mii_register(priv);
 	if (ret)
 		return ret;
 
@@ -630,11 +675,7 @@ int bcmgenet_mii_init(struct net_device *dev)
 	return 0;
 
 out:
-	if (of_phy_is_fixed_link(dn))
-		of_phy_deregister_fixed_link(dn);
-	of_node_put(priv->phy_dn);
-	mdiobus_unregister(priv->mii_bus);
-	mdiobus_free(priv->mii_bus);
+	bcmgenet_mii_exit(dev);
 	return ret;
 }
 
@@ -646,6 +687,5 @@ void bcmgenet_mii_exit(struct net_device *dev)
 	if (of_phy_is_fixed_link(dn))
 		of_phy_deregister_fixed_link(dn);
 	of_node_put(priv->phy_dn);
-	mdiobus_unregister(priv->mii_bus);
-	mdiobus_free(priv->mii_bus);
+	platform_device_unregister(priv->mii_pdev);
 }
diff --git a/drivers/net/ethernet/broadcom/sb1250-mac.c b/drivers/net/ethernet/broadcom/sb1250-mac.c
index 16a0f192daec..30865fe03eeb 100644
--- a/drivers/net/ethernet/broadcom/sb1250-mac.c
+++ b/drivers/net/ethernet/broadcom/sb1250-mac.c
@@ -1,21 +1,8 @@
+// SPDX-License-Identifier: GPL-2.0-or-later
 /*
  * Copyright (C) 2001,2002,2003,2004 Broadcom Corporation
  * Copyright (c) 2006, 2007  Maciej W. Rozycki
  *
- * This program is free software; you can redistribute it and/or
- * modify it under the terms of the GNU General Public License
- * as published by the Free Software Foundation; either version 2
- * of the License, or (at your option) any later version.
- *
- * This program is distributed in the hope that it will be useful,
- * but WITHOUT ANY WARRANTY; without even the implied warranty of
- * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
- * GNU General Public License for more details.
- *
- * You should have received a copy of the GNU General Public License
- * along with this program; if not, see <http://www.gnu.org/licenses/>.
- *
- *
  * This driver is designed for the Broadcom SiByte SOC built-in
  * Ethernet controllers. Written by Mitch Lichtenberg at Broadcom Corp.
  *
@@ -63,30 +50,30 @@ MODULE_DESCRIPTION("Broadcom SiByte SOC GB Ethernet driver");
 
 /* 1 normal messages, 0 quiet .. 7 verbose. */
 static int debug = 1;
-module_param(debug, int, S_IRUGO);
+module_param(debug, int, 0444);
 MODULE_PARM_DESC(debug, "Debug messages");
 
 #ifdef CONFIG_SBMAC_COALESCE
 static int int_pktcnt_tx = 255;
-module_param(int_pktcnt_tx, int, S_IRUGO);
+module_param(int_pktcnt_tx, int, 0444);
 MODULE_PARM_DESC(int_pktcnt_tx, "TX packet count");
 
 static int int_timeout_tx = 255;
-module_param(int_timeout_tx, int, S_IRUGO);
+module_param(int_timeout_tx, int, 0444);
 MODULE_PARM_DESC(int_timeout_tx, "TX timeout value");
 
 static int int_pktcnt_rx = 64;
-module_param(int_pktcnt_rx, int, S_IRUGO);
+module_param(int_pktcnt_rx, int, 0444);
 MODULE_PARM_DESC(int_pktcnt_rx, "RX packet count");
 
 static int int_timeout_rx = 64;
-module_param(int_timeout_rx, int, S_IRUGO);
+module_param(int_timeout_rx, int, 0444);
 MODULE_PARM_DESC(int_timeout_rx, "RX timeout value");
 #endif
 
 #include <asm/sibyte/board.h>
 #include <asm/sibyte/sb1250.h>
-#if defined(CONFIG_SIBYTE_BCM1x55) || defined(CONFIG_SIBYTE_BCM1x80)
+#if defined(CONFIG_SIBYTE_BCM1x80)
 #include <asm/sibyte/bcm1480_regs.h>
 #include <asm/sibyte/bcm1480_int.h>
 #define R_MAC_DMA_OODPKTLOST_RX	R_MAC_DMA_OODPKTLOST
@@ -100,7 +87,7 @@ MODULE_PARM_DESC(int_timeout_rx, "RX timeout value");
 #include <asm/sibyte/sb1250_mac.h>
 #include <asm/sibyte/sb1250_dma.h>
 
-#if defined(CONFIG_SIBYTE_BCM1x55) || defined(CONFIG_SIBYTE_BCM1x80)
+#if defined(CONFIG_SIBYTE_BCM1x80)
 #define UNIT_INT(n)		(K_BCM1480_INT_MAC_0 + ((n) * 2))
 #elif defined(CONFIG_SIBYTE_SB1250) || defined(CONFIG_SIBYTE_BCM112X)
 #define UNIT_INT(n)		(K_INT_MAC_0 + (n))
@@ -156,7 +143,7 @@ enum sbmac_state {
 			  (d)->sbdma_dscrtable : (d)->f+1)
 
 
-#define NUMCACHEBLKS(x) (((x)+SMP_CACHE_BYTES-1)/SMP_CACHE_BYTES)
+#define NUMCACHEBLKS(x) DIV_ROUND_UP(x, SMP_CACHE_BYTES)
 
 #define SBMAC_MAX_TXDESCR	256
 #define SBMAC_MAX_RXDESCR	256
@@ -299,7 +286,7 @@ static enum sbmac_state sbmac_set_channel_state(struct sbmac_softc *,
 static void sbmac_promiscuous_mode(struct sbmac_softc *sc, int onoff);
 static uint64_t sbmac_addr2reg(unsigned char *ptr);
 static irqreturn_t sbmac_intr(int irq, void *dev_instance);
-static int sbmac_start_tx(struct sk_buff *skb, struct net_device *dev);
+static netdev_tx_t sbmac_start_tx(struct sk_buff *skb, struct net_device *dev);
 static void sbmac_setmulti(struct sbmac_softc *sc);
 static int sbmac_init(struct platform_device *pldev, long long base);
 static int sbmac_set_speed(struct sbmac_softc *s, enum sbmac_speed speed);
@@ -307,7 +294,7 @@ static int sbmac_set_duplex(struct sbmac_softc *s, enum sbmac_duplex duplex,
 			    enum sbmac_fc fc);
 
 static int sbmac_open(struct net_device *dev);
-static void sbmac_tx_timeout (struct net_device *dev);
+static void sbmac_tx_timeout (struct net_device *dev, unsigned int txqueue);
 static void sbmac_set_rx_mode(struct net_device *dev);
 static int sbmac_mii_ioctl(struct net_device *dev, struct ifreq *rq, int cmd);
 static int sbmac_close(struct net_device *dev);
@@ -1288,7 +1275,7 @@ static void sbdma_tx_process(struct sbmac_softc *sc, struct sbmacdma *d,
 		 * for transmits, we just free buffers.
 		 */
 
-		dev_kfree_skb_irq(sb);
+		dev_consume_skb_irq(sb);
 
 		/*
 		 * .. and advance to the next buffer.
@@ -1367,15 +1354,11 @@ static int sbmac_initctx(struct sbmac_softc *s)
 
 static void sbdma_uninitctx(struct sbmacdma *d)
 {
-	if (d->sbdma_dscrtable_unaligned) {
-		kfree(d->sbdma_dscrtable_unaligned);
-		d->sbdma_dscrtable_unaligned = d->sbdma_dscrtable = NULL;
-	}
+	kfree(d->sbdma_dscrtable_unaligned);
+	d->sbdma_dscrtable_unaligned = d->sbdma_dscrtable = NULL;
 
-	if (d->sbdma_ctxtable) {
-		kfree(d->sbdma_ctxtable);
-		d->sbdma_ctxtable = NULL;
-	}
+	kfree(d->sbdma_ctxtable);
+	d->sbdma_ctxtable = NULL;
 }
 
 
@@ -1544,7 +1527,7 @@ static void sbmac_channel_start(struct sbmac_softc *s)
 	 * Turn on the rest of the bits in the enable register
 	 */
 
-#if defined(CONFIG_SIBYTE_BCM1x55) || defined(CONFIG_SIBYTE_BCM1x80)
+#if defined(CONFIG_SIBYTE_BCM1x80)
 	__raw_writeq(M_MAC_RXDMA_EN0 |
 		       M_MAC_TXDMA_EN0, s->sbm_macenable);
 #elif defined(CONFIG_SIBYTE_SB1250) || defined(CONFIG_SIBYTE_BCM112X)
@@ -2032,7 +2015,7 @@ static irqreturn_t sbmac_intr(int irq,void *dev_instance)
  *  Return value:
  *  	   nothing
  ********************************************************************* */
-static int sbmac_start_tx(struct sk_buff *skb, struct net_device *dev)
+static netdev_tx_t sbmac_start_tx(struct sk_buff *skb, struct net_device *dev)
 {
 	struct sbmac_softc *sc = netdev_priv(dev);
 	unsigned long flags;
@@ -2153,7 +2136,7 @@ static const struct net_device_ops sbmac_netdev_ops = {
 	.ndo_start_xmit		= sbmac_start_tx,
 	.ndo_set_rx_mode	= sbmac_set_rx_mode,
 	.ndo_tx_timeout		= sbmac_tx_timeout,
-	.ndo_do_ioctl		= sbmac_mii_ioctl,
+	.ndo_eth_ioctl		= sbmac_mii_ioctl,
 	.ndo_validate_addr	= eth_validate_addr,
 	.ndo_set_mac_address	= eth_mac_addr,
 #ifdef CONFIG_NET_POLL_CONTROLLER
@@ -2200,9 +2183,7 @@ static int sbmac_init(struct platform_device *pldev, long long base)
 		ea_reg >>= 8;
 	}
 
-	for (i = 0; i < 6; i++) {
-		dev->dev_addr[i] = eaddr[i];
-	}
+	eth_hw_addr_set(dev, eaddr);
 
 	/*
 	 * Initialize context (get pointers to registers and stuff), then
@@ -2222,7 +2203,7 @@ static int sbmac_init(struct platform_device *pldev, long long base)
 	dev->min_mtu = 0;
 	dev->max_mtu = ENET_PACKET_SIZE;
 
-	netif_napi_add(dev, &sc->napi, sbmac_poll, 16);
+	netif_napi_add_weight(dev, &sc->napi, sbmac_poll, 16);
 
 	dev->irq		= UNIT_INT(idx);
 
@@ -2361,21 +2342,11 @@ static int sbmac_mii_probe(struct net_device *dev)
 	}
 
 	/* Remove any features not supported by the controller */
-	phy_dev->supported &= SUPPORTED_10baseT_Half |
-			      SUPPORTED_10baseT_Full |
-			      SUPPORTED_100baseT_Half |
-			      SUPPORTED_100baseT_Full |
-			      SUPPORTED_1000baseT_Half |
-			      SUPPORTED_1000baseT_Full |
-			      SUPPORTED_Autoneg |
-			      SUPPORTED_MII |
-			      SUPPORTED_Pause |
-			      SUPPORTED_Asym_Pause;
+	phy_set_max_speed(phy_dev, SPEED_1000);
+	phy_support_asym_pause(phy_dev);
 
 	phy_attached_info(phy_dev);
 
-	phy_dev->advertising = phy_dev->supported;
-
 	sc->phy_dev = phy_dev;
 
 	return 0;
@@ -2446,7 +2417,7 @@ static void sbmac_mii_poll(struct net_device *dev)
 }
 
 
-static void sbmac_tx_timeout (struct net_device *dev)
+static void sbmac_tx_timeout (struct net_device *dev, unsigned int txqueue)
 {
 	struct sbmac_softc *sc = netdev_priv(dev);
 	unsigned long flags;
@@ -2563,8 +2534,13 @@ static int sbmac_probe(struct platform_device *pldev)
 	int err;
 
 	res = platform_get_resource(pldev, IORESOURCE_MEM, 0);
-	BUG_ON(!res);
-	sbm_base = ioremap_nocache(res->start, resource_size(res));
+	if (!res) {
+		printk(KERN_ERR "%s: failed to get resource\n",
+		       dev_name(&pldev->dev));
+		err = -EINVAL;
+		goto out_out;
+	}
+	sbm_base = ioremap(res->start, resource_size(res));
 	if (!sbm_base) {
 		printk(KERN_ERR "%s: unable to map device registers\n",
 		       dev_name(&pldev->dev));
@@ -2617,7 +2593,7 @@ out_out:
 	return err;
 }
 
-static int sbmac_remove(struct platform_device *pldev)
+static void sbmac_remove(struct platform_device *pldev)
 {
 	struct net_device *dev = platform_get_drvdata(pldev);
 	struct sbmac_softc *sc = netdev_priv(dev);
@@ -2628,8 +2604,6 @@ static int sbmac_remove(struct platform_device *pldev)
 	mdiobus_free(sc->mii_bus);
 	iounmap(sc->sbm_base);
 	free_netdev(dev);
-
-	return 0;
 }
 
 static struct platform_driver sbmac_driver = {
diff --git a/drivers/net/ethernet/broadcom/tg3.c b/drivers/net/ethernet/broadcom/tg3.c
index 537d571ee601..7f00ec7fd7b9 100644
--- a/drivers/net/ethernet/broadcom/tg3.c
+++ b/drivers/net/ethernet/broadcom/tg3.c
@@ -4,11 +4,17 @@
  * Copyright (C) 2001, 2002, 2003, 2004 David S. Miller (davem@redhat.com)
  * Copyright (C) 2001, 2002, 2003 Jeff Garzik (jgarzik@pobox.com)
  * Copyright (C) 2004 Sun Microsystems Inc.
- * Copyright (C) 2005-2014 Broadcom Corporation.
+ * Copyright (C) 2005-2016 Broadcom Corporation.
+ * Copyright (C) 2016-2017 Broadcom Limited.
+ * Copyright (C) 2018 Broadcom. All Rights Reserved. The term "Broadcom"
+ * refers to Broadcom Inc. and/or its subsidiaries.
  *
  * Firmware is:
  *	Derived from proprietary unpublished source code,
- *	Copyright (C) 2000-2003 Broadcom Corporation.
+ *	Copyright (C) 2000-2016 Broadcom Corporation.
+ *	Copyright (C) 2016-2017 Broadcom Ltd.
+ *	Copyright (C) 2018 Broadcom. All Rights Reserved. The term "Broadcom"
+ *	refers to Broadcom Inc. and/or its subsidiaries.
  *
  *	Permission is hereby granted for the distribution of this firmware
  *	data in hexadecimal or equivalent format, provided this copyright
@@ -48,8 +54,11 @@
 #include <linux/ssb/ssb_driver_gige.h>
 #include <linux/hwmon.h>
 #include <linux/hwmon-sysfs.h>
+#include <linux/crc32.h>
+#include <linux/dmi.h>
 
 #include <net/checksum.h>
+#include <net/gso.h>
 #include <net/ip.h>
 
 #include <linux/io.h>
@@ -59,11 +68,6 @@
 #include <uapi/linux/net_tstamp.h>
 #include <linux/ptp_clock_kernel.h>
 
-#ifdef CONFIG_SPARC
-#include <asm/idprom.h>
-#include <asm/prom.h>
-#endif
-
 #define BAR_0	0
 #define BAR_2	2
 
@@ -94,11 +98,9 @@ static inline void _tg3_flag_clear(enum TG3_FLAGS flag, unsigned long *bits)
 	_tg3_flag_clear(TG3_FLAG_##flag, (tp)->tg3_flags)
 
 #define DRV_MODULE_NAME		"tg3"
+/* DO NOT UPDATE TG3_*_NUM defines */
 #define TG3_MAJ_NUM			3
 #define TG3_MIN_NUM			137
-#define DRV_MODULE_VERSION	\
-	__stringify(TG3_MAJ_NUM) "." __stringify(TG3_MIN_NUM)
-#define DRV_MODULE_RELDATE	"May 11, 2014"
 
 #define RESET_KIND_SHUTDOWN	0
 #define RESET_KIND_INIT		1
@@ -220,14 +222,11 @@ static inline void _tg3_flag_clear(enum TG3_FLAGS flag, unsigned long *bits)
 #define FIRMWARE_TG3TSO		"tigon/tg3_tso.bin"
 #define FIRMWARE_TG3TSO5	"tigon/tg3_tso5.bin"
 
-static char version[] =
-	DRV_MODULE_NAME ".c:v" DRV_MODULE_VERSION " (" DRV_MODULE_RELDATE ")";
-
-MODULE_AUTHOR("David S. Miller (davem@redhat.com) and Jeff Garzik (jgarzik@pobox.com)");
+MODULE_AUTHOR("David S. Miller <davem@redhat.com> and Jeff Garzik <jgarzik@pobox.com>");
 MODULE_DESCRIPTION("Broadcom Tigon3 ethernet driver");
 MODULE_LICENSE("GPL");
-MODULE_VERSION(DRV_MODULE_VERSION);
 MODULE_FIRMWARE(FIRMWARE_TG3);
+MODULE_FIRMWARE(FIRMWARE_TG357766);
 MODULE_FIRMWARE(FIRMWARE_TG3TSO);
 MODULE_FIRMWARE(FIRMWARE_TG3TSO5);
 
@@ -719,6 +718,7 @@ static int tg3_ape_lock(struct tg3 *tp, int locknum)
 	case TG3_APE_LOCK_GPIO:
 		if (tg3_asic_rev(tp) == ASIC_REV_5761)
 			return 0;
+		fallthrough;
 	case TG3_APE_LOCK_GRC:
 	case TG3_APE_LOCK_MEM:
 		if (!tp->pci_fn)
@@ -779,6 +779,7 @@ static void tg3_ape_unlock(struct tg3 *tp, int locknum)
 	case TG3_APE_LOCK_GPIO:
 		if (tg3_asic_rev(tp) == ASIC_REV_5761)
 			return;
+		fallthrough;
 	case TG3_APE_LOCK_GRC:
 	case TG3_APE_LOCK_MEM:
 		if (!tp->pci_fn)
@@ -920,8 +921,8 @@ static int tg3_ape_send_event(struct tg3 *tp, u32 event)
 	if (!(apedata & APE_FW_STATUS_READY))
 		return -EAGAIN;
 
-	/* Wait for up to 1 millisecond for APE to service previous event. */
-	err = tg3_ape_event_lock(tp, 1000);
+	/* Wait for up to 20 millisecond for APE to service previous event. */
+	err = tg3_ape_event_lock(tp, 20000);
 	if (err)
 		return err;
 
@@ -944,6 +945,7 @@ static void tg3_ape_driver_state_change(struct tg3 *tp, int kind)
 
 	switch (kind) {
 	case RESET_KIND_INIT:
+		tg3_ape_write32(tp, TG3_APE_HOST_HEARTBEAT_COUNT, tp->ape_hb++);
 		tg3_ape_write32(tp, TG3_APE_HOST_SEG_SIG,
 				APE_HOST_SEG_SIG_MAGIC);
 		tg3_ape_write32(tp, TG3_APE_HOST_SEG_LEN,
@@ -960,13 +962,6 @@ static void tg3_ape_driver_state_change(struct tg3 *tp, int kind)
 		event = APE_EVENT_STATUS_STATE_START;
 		break;
 	case RESET_KIND_SHUTDOWN:
-		/* With the interface we are currently using,
-		 * APE does not track driver state.  Wiping
-		 * out the HOST SEGMENT SIGNATURE forces
-		 * the APE to assume OS absent status.
-		 */
-		tg3_ape_write32(tp, TG3_APE_HOST_SEG_SIG, 0x0);
-
 		if (device_may_wakeup(&tp->pdev->dev) &&
 		    tg3_flag(tp, WOL_ENABLE)) {
 			tg3_ape_write32(tp, TG3_APE_HOST_WOL_SPEED,
@@ -988,6 +983,18 @@ static void tg3_ape_driver_state_change(struct tg3 *tp, int kind)
 	tg3_ape_send_event(tp, event);
 }
 
+static void tg3_send_ape_heartbeat(struct tg3 *tp,
+				   unsigned long interval)
+{
+	/* Check if hb interval has exceeded */
+	if (!tg3_flag(tp, ENABLE_APE) ||
+	    time_before(jiffies, tp->ape_hb_jiffies + interval))
+		return;
+
+	tg3_ape_write32(tp, TG3_APE_HOST_HEARTBEAT_COUNT, tp->ape_hb++);
+	tp->ape_hb_jiffies = jiffies;
+}
+
 static void tg3_disable_ints(struct tg3 *tp)
 {
 	int i;
@@ -1063,7 +1070,6 @@ static void tg3_int_reenable(struct tg3_napi *tnapi)
 	struct tg3 *tp = tnapi->tp;
 
 	tw32_mailbox(tnapi->int_mbox, tnapi->last_tag << 24);
-	mmiowb();
 
 	/* When doing tagged status, this work check is unnecessary.
 	 * The last_tag we write above tells the chip which piece of
@@ -1534,8 +1540,7 @@ static int tg3_mdio_init(struct tg3 *tp)
 		return -ENOMEM;
 
 	tp->mdio_bus->name     = "tg3 mdio bus";
-	snprintf(tp->mdio_bus->id, MII_BUS_ID_SIZE, "%x",
-		 (tp->pdev->bus->number << 8) | tp->pdev->devfn);
+	snprintf(tp->mdio_bus->id, MII_BUS_ID_SIZE, "%x", pci_dev_id(tp->pdev));
 	tp->mdio_bus->priv     = tp;
 	tp->mdio_bus->parent   = &tp->pdev->dev;
 	tp->mdio_bus->read     = &tg3_mdio_read;
@@ -1577,13 +1582,7 @@ static int tg3_mdio_init(struct tg3 *tp)
 				     PHY_BRCM_RX_REFCLK_UNUSED |
 				     PHY_BRCM_DIS_TXCRXC_NOENRGY |
 				     PHY_BRCM_AUTO_PWRDWN_ENABLE;
-		if (tg3_flag(tp, RGMII_INBAND_DISABLE))
-			phydev->dev_flags |= PHY_BRCM_STD_IBND_DISABLE;
-		if (tg3_flag(tp, RGMII_EXT_IBND_RX_EN))
-			phydev->dev_flags |= PHY_BRCM_EXT_IBND_RX_ENABLE;
-		if (tg3_flag(tp, RGMII_EXT_IBND_TX_EN))
-			phydev->dev_flags |= PHY_BRCM_EXT_IBND_TX_ENABLE;
-		/* fallthru */
+		fallthrough;
 	case PHY_ID_RTL8211C:
 		phydev->interface = PHY_INTERFACE_MODE_RGMII;
 		break;
@@ -2107,16 +2106,14 @@ static int tg3_phy_init(struct tg3 *tp)
 	case PHY_INTERFACE_MODE_GMII:
 	case PHY_INTERFACE_MODE_RGMII:
 		if (!(tp->phy_flags & TG3_PHYFLG_10_100_ONLY)) {
-			phydev->supported &= (PHY_GBIT_FEATURES |
-					      SUPPORTED_Pause |
-					      SUPPORTED_Asym_Pause);
+			phy_set_max_speed(phydev, SPEED_1000);
+			phy_support_asym_pause(phydev);
 			break;
 		}
-		/* fallthru */
+		fallthrough;
 	case PHY_INTERFACE_MODE_MII:
-		phydev->supported &= (PHY_BASIC_FEATURES |
-				      SUPPORTED_Pause |
-				      SUPPORTED_Asym_Pause);
+		phy_set_max_speed(phydev, SPEED_100);
+		phy_support_asym_pause(phydev);
 		break;
 	default:
 		phy_disconnect(mdiobus_get_phy(tp->mdio_bus, tp->phy_addr));
@@ -2125,8 +2122,6 @@ static int tg3_phy_init(struct tg3 *tp)
 
 	tp->phy_flags |= TG3_PHYFLG_IS_CONNECTED;
 
-	phydev->advertising = phydev->supported;
-
 	phy_attached_info(phydev);
 
 	return 0;
@@ -2146,7 +2141,8 @@ static void tg3_phy_start(struct tg3 *tp)
 		phydev->speed = tp->link_config.speed;
 		phydev->duplex = tp->link_config.duplex;
 		phydev->autoneg = tp->link_config.autoneg;
-		phydev->advertising = tp->link_config.advertising;
+		ethtool_convert_legacy_u32_to_link_mode(
+			phydev->advertising, tp->link_config.advertising);
 	}
 
 	phy_start(phydev);
@@ -2343,10 +2339,10 @@ static void tg3_phy_apply_otp(struct tg3 *tp)
 	tg3_phy_toggle_auxctl_smdsp(tp, false);
 }
 
-static void tg3_eee_pull_config(struct tg3 *tp, struct ethtool_eee *eee)
+static void tg3_eee_pull_config(struct tg3 *tp, struct ethtool_keee *eee)
 {
 	u32 val;
-	struct ethtool_eee *dest = &tp->eee;
+	struct ethtool_keee *dest = &tp->eee;
 
 	if (!(tp->phy_flags & TG3_PHYFLG_EEE_CAP))
 		return;
@@ -2367,13 +2363,13 @@ static void tg3_eee_pull_config(struct tg3 *tp, struct ethtool_eee *eee)
 	/* Pull lp advertised settings */
 	if (tg3_phy_cl45_read(tp, MDIO_MMD_AN, MDIO_AN_EEE_LPABLE, &val))
 		return;
-	dest->lp_advertised = mmd_eee_adv_to_ethtool_adv_t(val);
+	mii_eee_cap1_mod_linkmode_t(dest->lp_advertised, val);
 
 	/* Pull advertised and eee_enabled settings */
 	if (tg3_phy_cl45_read(tp, MDIO_MMD_AN, MDIO_AN_EEE_ADV, &val))
 		return;
 	dest->eee_enabled = !!val;
-	dest->advertised = mmd_eee_adv_to_ethtool_adv_t(val);
+	mii_eee_cap1_mod_linkmode_t(dest->advertised, val);
 
 	/* Pull tx_lpi_enabled */
 	val = tr32(TG3_CPMU_EEE_MODE);
@@ -3225,7 +3221,7 @@ static int tg3_nvram_read_using_eeprom(struct tg3 *tp,
 	return 0;
 }
 
-#define NVRAM_CMD_TIMEOUT 5000
+#define NVRAM_CMD_TIMEOUT 10000
 
 static int tg3_nvram_exec_cmd(struct tg3 *tp, u32 nvram_cmd)
 {
@@ -3742,7 +3738,7 @@ static int tg3_load_firmware_cpu(struct tg3 *tp, u32 cpu_base,
 	}
 
 	do {
-		u32 *fw_data = (u32 *)(fw_hdr + 1);
+		__be32 *fw_data = (__be32 *)(fw_hdr + 1);
 		for (i = 0; i < tg3_fw_data_len(tp, fw_hdr); i++)
 			write_op(tp, cpu_scratch_base +
 				     (be32_to_cpu(fw_hdr->base_addr) & 0xffff) +
@@ -3948,7 +3944,8 @@ static int tg3_load_tso_firmware(struct tg3 *tp)
 }
 
 /* tp->lock is held. */
-static void __tg3_set_one_mac_addr(struct tg3 *tp, u8 *mac_addr, int index)
+static void __tg3_set_one_mac_addr(struct tg3 *tp, const u8 *mac_addr,
+				   int index)
 {
 	u32 addr_high, addr_low;
 
@@ -4023,7 +4020,7 @@ static int tg3_power_up(struct tg3 *tp)
 
 static int tg3_setup_phy(struct tg3 *, bool);
 
-static int tg3_power_down_prepare(struct tg3 *tp)
+static void tg3_power_down_prepare(struct tg3 *tp)
 {
 	u32 misc_host_ctrl;
 	bool device_should_wake, do_low_power;
@@ -4046,8 +4043,9 @@ static int tg3_power_down_prepare(struct tg3 *tp)
 		do_low_power = false;
 		if ((tp->phy_flags & TG3_PHYFLG_IS_CONNECTED) &&
 		    !(tp->phy_flags & TG3_PHYFLG_IS_LOW_POWER)) {
+			__ETHTOOL_DECLARE_LINK_MODE_MASK(advertising) = { 0, };
 			struct phy_device *phydev;
-			u32 phyid, advertising;
+			u32 phyid;
 
 			phydev = mdiobus_get_phy(tp->mdio_bus, tp->phy_addr);
 
@@ -4056,25 +4054,33 @@ static int tg3_power_down_prepare(struct tg3 *tp)
 			tp->link_config.speed = phydev->speed;
 			tp->link_config.duplex = phydev->duplex;
 			tp->link_config.autoneg = phydev->autoneg;
-			tp->link_config.advertising = phydev->advertising;
-
-			advertising = ADVERTISED_TP |
-				      ADVERTISED_Pause |
-				      ADVERTISED_Autoneg |
-				      ADVERTISED_10baseT_Half;
+			ethtool_convert_link_mode_to_legacy_u32(
+				&tp->link_config.advertising,
+				phydev->advertising);
+
+			linkmode_set_bit(ETHTOOL_LINK_MODE_TP_BIT, advertising);
+			linkmode_set_bit(ETHTOOL_LINK_MODE_Pause_BIT,
+					 advertising);
+			linkmode_set_bit(ETHTOOL_LINK_MODE_Autoneg_BIT,
+					 advertising);
+			linkmode_set_bit(ETHTOOL_LINK_MODE_10baseT_Half_BIT,
+					 advertising);
 
 			if (tg3_flag(tp, ENABLE_ASF) || device_should_wake) {
-				if (tg3_flag(tp, WOL_SPEED_100MB))
-					advertising |=
-						ADVERTISED_100baseT_Half |
-						ADVERTISED_100baseT_Full |
-						ADVERTISED_10baseT_Full;
-				else
-					advertising |= ADVERTISED_10baseT_Full;
+				if (tg3_flag(tp, WOL_SPEED_100MB)) {
+					linkmode_set_bit(ETHTOOL_LINK_MODE_100baseT_Half_BIT,
+							 advertising);
+					linkmode_set_bit(ETHTOOL_LINK_MODE_100baseT_Full_BIT,
+							 advertising);
+					linkmode_set_bit(ETHTOOL_LINK_MODE_10baseT_Full_BIT,
+							 advertising);
+				} else {
+					linkmode_set_bit(ETHTOOL_LINK_MODE_10baseT_Full_BIT,
+							 advertising);
+				}
 			}
 
-			phydev->advertising = advertising;
-
+			linkmode_copy(phydev->advertising, advertising);
 			phy_start_aneg(phydev);
 
 			phyid = phydev->drv->phy_id & phydev->drv->phy_id_mask;
@@ -4258,7 +4264,7 @@ static int tg3_power_down_prepare(struct tg3 *tp)
 
 	tg3_ape_driver_state_change(tp, RESET_KIND_SHUTDOWN);
 
-	return 0;
+	return;
 }
 
 static void tg3_power_down(struct tg3 *tp)
@@ -4267,7 +4273,7 @@ static void tg3_power_down(struct tg3 *tp)
 	pci_set_power_state(tp->pdev, PCI_D3hot);
 }
 
-static void tg3_aux_stat_to_speed_duplex(struct tg3 *tp, u32 val, u16 *speed, u8 *duplex)
+static void tg3_aux_stat_to_speed_duplex(struct tg3 *tp, u32 val, u32 *speed, u8 *duplex)
 {
 	switch (val & MII_TG3_AUX_STAT_SPDMASK) {
 	case MII_TG3_AUX_STAT_10HALF:
@@ -4349,23 +4355,12 @@ static int tg3_phy_autoneg_cfg(struct tg3 *tp, u32 advertise, u32 flowctrl)
 	if (!err) {
 		u32 err2;
 
-		val = 0;
-		/* Advertise 100-BaseTX EEE ability */
-		if (advertise & ADVERTISED_100baseT_Full)
-			val |= MDIO_AN_EEE_ADV_100TX;
-		/* Advertise 1000-BaseT EEE ability */
-		if (advertise & ADVERTISED_1000baseT_Full)
-			val |= MDIO_AN_EEE_ADV_1000T;
-
-		if (!tp->eee.eee_enabled) {
+		if (!tp->eee.eee_enabled)
 			val = 0;
-			tp->eee.advertised = 0;
-		} else {
-			tp->eee.advertised = advertise &
-					     (ADVERTISED_100baseT_Full |
-					      ADVERTISED_1000baseT_Full);
-		}
+		else
+			val = ethtool_adv_to_mmd_eee_adv_t(advertise);
 
+		mii_eee_cap1_mod_linkmode_t(tp->eee.advertised, val);
 		err = tg3_phy_cl45_write(tp, MDIO_MMD_AN, MDIO_AN_EEE_ADV, val);
 		if (err)
 			val = 0;
@@ -4381,7 +4376,7 @@ static int tg3_phy_autoneg_cfg(struct tg3 *tp, u32 advertise, u32 flowctrl)
 				      MII_TG3_DSP_TAP26_RMRXSTO |
 				      MII_TG3_DSP_TAP26_OPCSINPT;
 			tg3_phydsp_write(tp, MII_TG3_DSP_TAP26, val);
-			/* Fall through */
+			fallthrough;
 		case ASIC_REV_5720:
 		case ASIC_REV_5762:
 			if (!tg3_phydsp_read(tp, MII_TG3_DSP_CH34TP2, &val))
@@ -4529,7 +4524,7 @@ static int tg3_phy_pull_config(struct tg3 *tp)
 				tp->link_config.speed = SPEED_1000;
 				break;
 			}
-			/* Fall through */
+			fallthrough;
 		default:
 			goto done;
 		}
@@ -4613,7 +4608,7 @@ static int tg3_init_5401phy_dsp(struct tg3 *tp)
 
 static bool tg3_phy_eee_config_ok(struct tg3 *tp)
 {
-	struct ethtool_eee eee;
+	struct ethtool_keee eee = {};
 
 	if (!(tp->phy_flags & TG3_PHYFLG_EEE_CAP))
 		return true;
@@ -4621,13 +4616,13 @@ static bool tg3_phy_eee_config_ok(struct tg3 *tp)
 	tg3_eee_pull_config(tp, &eee);
 
 	if (tp->eee.eee_enabled) {
-		if (tp->eee.advertised != eee.advertised ||
+		if (!linkmode_equal(tp->eee.advertised, eee.advertised) ||
 		    tp->eee.tx_lpi_timer != eee.tx_lpi_timer ||
 		    tp->eee.tx_lpi_enabled != eee.tx_lpi_enabled)
 			return false;
 	} else {
 		/* EEE is disabled but we're advertising */
-		if (eee.advertised)
+		if (!linkmode_empty(eee.advertised))
 			return false;
 	}
 
@@ -4771,7 +4766,7 @@ static int tg3_setup_copper_phy(struct tg3 *tp, bool force_reset)
 	bool current_link_up;
 	u32 bmsr, val;
 	u32 lcl_adv, rmt_adv;
-	u16 current_speed;
+	u32 current_speed;
 	u8 current_duplex;
 	int i, err;
 
@@ -5200,7 +5195,7 @@ static int tg3_fiber_aneg_smachine(struct tg3 *tp,
 		if (ap->flags & (MR_AN_ENABLE | MR_RESTART_AN))
 			ap->state = ANEG_STATE_AN_ENABLE;
 
-		/* fallthru */
+		fallthrough;
 	case ANEG_STATE_AN_ENABLE:
 		ap->flags &= ~(MR_AN_COMPLETE | MR_PAGE_RX);
 		if (ap->flags & MR_AN_ENABLE) {
@@ -5230,7 +5225,7 @@ static int tg3_fiber_aneg_smachine(struct tg3 *tp,
 		ret = ANEG_TIMER_ENAB;
 		ap->state = ANEG_STATE_RESTART;
 
-		/* fallthru */
+		fallthrough;
 	case ANEG_STATE_RESTART:
 		delta = ap->cur_time - ap->link_time;
 		if (delta > ANEG_STATE_SETTLE_TIME)
@@ -5273,7 +5268,7 @@ static int tg3_fiber_aneg_smachine(struct tg3 *tp,
 
 		ap->state = ANEG_STATE_ACK_DETECT;
 
-		/* fallthru */
+		fallthrough;
 	case ANEG_STATE_ACK_DETECT:
 		if (ap->ack_match != 0) {
 			if ((ap->rxconfig & ~ANEG_CFG_ACK) ==
@@ -5499,7 +5494,6 @@ static bool tg3_setup_fiber_hw_autoneg(struct tg3 *tp, u32 mac_status)
 	int workaround, port_a;
 
 	serdes_cfg = 0;
-	expected_sg_dig_ctrl = 0;
 	workaround = 0;
 	port_a = 1;
 	current_link_up = false;
@@ -5703,7 +5697,7 @@ out:
 static int tg3_setup_fiber_phy(struct tg3 *tp, bool force_reset)
 {
 	u32 orig_pause_cfg;
-	u16 orig_active_speed;
+	u32 orig_active_speed;
 	u8 orig_active_duplex;
 	u32 mac_status;
 	bool current_link_up;
@@ -5743,7 +5737,6 @@ static int tg3_setup_fiber_phy(struct tg3 *tp, bool force_reset)
 	tw32_f(MAC_EVENT, MAC_EVENT_LNKSTATE_CHANGED);
 	udelay(40);
 
-	current_link_up = false;
 	tp->link_config.rmt_adv = 0;
 	mac_status = tr32(MAC_STATUS);
 
@@ -5807,7 +5800,7 @@ static int tg3_setup_fiber_mii_phy(struct tg3 *tp, bool force_reset)
 {
 	int err = 0;
 	u32 bmsr, bmcr;
-	u16 current_speed = SPEED_UNKNOWN;
+	u32 current_speed = SPEED_UNKNOWN;
 	u8 current_duplex = DUPLEX_UNKNOWN;
 	bool current_link_up = false;
 	u32 local_adv, remote_adv, sgsr;
@@ -6124,10 +6117,16 @@ static int tg3_setup_phy(struct tg3 *tp, bool force_reset)
 }
 
 /* tp->lock must be held */
-static u64 tg3_refclk_read(struct tg3 *tp)
+static u64 tg3_refclk_read(struct tg3 *tp, struct ptp_system_timestamp *sts)
 {
-	u64 stamp = tr32(TG3_EAV_REF_CLCK_LSB);
-	return stamp | (u64)tr32(TG3_EAV_REF_CLCK_MSB) << 32;
+	u64 stamp;
+
+	ptp_read_system_prets(sts);
+	stamp = tr32(TG3_EAV_REF_CLCK_LSB);
+	ptp_read_system_postts(sts);
+	stamp |= (u64)tr32(TG3_EAV_REF_CLCK_MSB) << 32;
+
+	return stamp;
 }
 
 /* tp->lock must be held */
@@ -6143,13 +6142,11 @@ static void tg3_refclk_write(struct tg3 *tp, u64 newval)
 
 static inline void tg3_full_lock(struct tg3 *tp, int irq_sync);
 static inline void tg3_full_unlock(struct tg3 *tp);
-static int tg3_get_ts_info(struct net_device *dev, struct ethtool_ts_info *info)
+static int tg3_get_ts_info(struct net_device *dev, struct kernel_ethtool_ts_info *info)
 {
 	struct tg3 *tp = netdev_priv(dev);
 
-	info->so_timestamping = SOF_TIMESTAMPING_TX_SOFTWARE |
-				SOF_TIMESTAMPING_RX_SOFTWARE |
-				SOF_TIMESTAMPING_SOFTWARE;
+	info->so_timestamping = SOF_TIMESTAMPING_TX_SOFTWARE;
 
 	if (tg3_flag(tp, PTP_CAPABLE)) {
 		info->so_timestamping |= SOF_TIMESTAMPING_TX_HARDWARE |
@@ -6159,8 +6156,6 @@ static int tg3_get_ts_info(struct net_device *dev, struct ethtool_ts_info *info)
 
 	if (tp->ptp_clock)
 		info->phc_index = ptp_clock_index(tp->ptp_clock);
-	else
-		info->phc_index = -1;
 
 	info->tx_types = (1 << HWTSTAMP_TX_OFF) | (1 << HWTSTAMP_TX_ON);
 
@@ -6171,34 +6166,26 @@ static int tg3_get_ts_info(struct net_device *dev, struct ethtool_ts_info *info)
 	return 0;
 }
 
-static int tg3_ptp_adjfreq(struct ptp_clock_info *ptp, s32 ppb)
+static int tg3_ptp_adjfine(struct ptp_clock_info *ptp, long scaled_ppm)
 {
 	struct tg3 *tp = container_of(ptp, struct tg3, ptp_info);
-	bool neg_adj = false;
-	u32 correction = 0;
-
-	if (ppb < 0) {
-		neg_adj = true;
-		ppb = -ppb;
-	}
+	u64 correction;
+	bool neg_adj;
 
 	/* Frequency adjustment is performed using hardware with a 24 bit
 	 * accumulator and a programmable correction value. On each clk, the
 	 * correction value gets added to the accumulator and when it
 	 * overflows, the time counter is incremented/decremented.
-	 *
-	 * So conversion from ppb to correction value is
-	 *		ppb * (1 << 24) / 1000000000
 	 */
-	correction = div_u64((u64)ppb * (1 << 24), 1000000000ULL) &
-		     TG3_EAV_REF_CLK_CORRECT_MASK;
+	neg_adj = diff_by_scaled_ppm(1 << 24, scaled_ppm, &correction);
 
 	tg3_full_lock(tp, 0);
 
 	if (correction)
 		tw32(TG3_EAV_REF_CLK_CORRECT_CTL,
 		     TG3_EAV_REF_CLK_CORRECT_EN |
-		     (neg_adj ? TG3_EAV_REF_CLK_CORRECT_NEG : 0) | correction);
+		     (neg_adj ? TG3_EAV_REF_CLK_CORRECT_NEG : 0) |
+		     ((u32)correction & TG3_EAV_REF_CLK_CORRECT_MASK));
 	else
 		tw32(TG3_EAV_REF_CLK_CORRECT_CTL, 0);
 
@@ -6218,13 +6205,14 @@ static int tg3_ptp_adjtime(struct ptp_clock_info *ptp, s64 delta)
 	return 0;
 }
 
-static int tg3_ptp_gettime(struct ptp_clock_info *ptp, struct timespec64 *ts)
+static int tg3_ptp_gettimex(struct ptp_clock_info *ptp, struct timespec64 *ts,
+			    struct ptp_system_timestamp *sts)
 {
 	u64 ns;
 	struct tg3 *tp = container_of(ptp, struct tg3, ptp_info);
 
 	tg3_full_lock(tp, 0);
-	ns = tg3_refclk_read(tp);
+	ns = tg3_refclk_read(tp, sts);
 	ns += tp->ptp_adjust;
 	tg3_full_unlock(tp);
 
@@ -6258,6 +6246,10 @@ static int tg3_ptp_enable(struct ptp_clock_info *ptp,
 
 	switch (rq->type) {
 	case PTP_CLK_REQ_PEROUT:
+		/* Reject requests with unsupported flags */
+		if (rq->perout.flags)
+			return -EOPNOTSUPP;
+
 		if (rq->perout.index != 0)
 			return -EINVAL;
 
@@ -6308,6 +6300,46 @@ err_out:
 	return -EOPNOTSUPP;
 }
 
+static void tg3_hwclock_to_timestamp(struct tg3 *tp, u64 hwclock,
+				     struct skb_shared_hwtstamps *timestamp)
+{
+	memset(timestamp, 0, sizeof(struct skb_shared_hwtstamps));
+	timestamp->hwtstamp  = ns_to_ktime((hwclock & TG3_TSTAMP_MASK) +
+					   tp->ptp_adjust);
+}
+
+static void tg3_read_tx_tstamp(struct tg3 *tp, u64 *hwclock)
+{
+	*hwclock = tr32(TG3_TX_TSTAMP_LSB);
+	*hwclock |= (u64)tr32(TG3_TX_TSTAMP_MSB) << 32;
+}
+
+static long tg3_ptp_ts_aux_work(struct ptp_clock_info *ptp)
+{
+	struct tg3 *tp = container_of(ptp, struct tg3, ptp_info);
+	struct skb_shared_hwtstamps timestamp;
+	u64 hwclock;
+
+	if (tp->ptp_txts_retrycnt > 2)
+		goto done;
+
+	tg3_read_tx_tstamp(tp, &hwclock);
+
+	if (hwclock != tp->pre_tx_ts) {
+		tg3_hwclock_to_timestamp(tp, hwclock, &timestamp);
+		skb_tstamp_tx(tp->tx_tstamp_skb, &timestamp);
+		goto done;
+	}
+	tp->ptp_txts_retrycnt++;
+	return HZ / 10;
+done:
+	dev_consume_skb_any(tp->tx_tstamp_skb);
+	tp->tx_tstamp_skb = NULL;
+	tp->ptp_txts_retrycnt = 0;
+	tp->pre_tx_ts = 0;
+	return -1;
+}
+
 static const struct ptp_clock_info tg3_ptp_caps = {
 	.owner		= THIS_MODULE,
 	.name		= "tg3 clock",
@@ -6317,21 +6349,14 @@ static const struct ptp_clock_info tg3_ptp_caps = {
 	.n_per_out	= 1,
 	.n_pins		= 0,
 	.pps		= 0,
-	.adjfreq	= tg3_ptp_adjfreq,
+	.adjfine	= tg3_ptp_adjfine,
 	.adjtime	= tg3_ptp_adjtime,
-	.gettime64	= tg3_ptp_gettime,
+	.do_aux_work	= tg3_ptp_ts_aux_work,
+	.gettimex64	= tg3_ptp_gettimex,
 	.settime64	= tg3_ptp_settime,
 	.enable		= tg3_ptp_enable,
 };
 
-static void tg3_hwclock_to_timestamp(struct tg3 *tp, u64 hwclock,
-				     struct skb_shared_hwtstamps *timestamp)
-{
-	memset(timestamp, 0, sizeof(struct skb_shared_hwtstamps));
-	timestamp->hwtstamp  = ns_to_ktime((hwclock & TG3_TSTAMP_MASK) +
-					   tp->ptp_adjust);
-}
-
 /* tp->lock must be held */
 static void tg3_ptp_init(struct tg3 *tp)
 {
@@ -6362,6 +6387,8 @@ static void tg3_ptp_fini(struct tg3 *tp)
 	ptp_clock_unregister(tp->ptp_clock);
 	tp->ptp_clock = NULL;
 	tp->ptp_adjust = 0;
+	dev_consume_skb_any(tp->tx_tstamp_skb);
+	tp->tx_tstamp_skb = NULL;
 }
 
 static inline int tg3_irq_sync(struct tg3 *tp)
@@ -6433,6 +6460,14 @@ static void tg3_dump_state(struct tg3 *tp)
 	int i;
 	u32 *regs;
 
+	/* If it is a PCI error, all registers will be 0xffff,
+	 * we don't dump them out, just report the error and return
+	 */
+	if (tp->pdev->error_state != pci_channel_io_normal) {
+		netdev_err(tp->dev, "PCI channel ERROR!\n");
+		return;
+	}
+
 	regs = kzalloc(TG3_REG_BLK_SIZE, GFP_ATOMIC);
 	if (!regs)
 		return;
@@ -6532,6 +6567,7 @@ static void tg3_tx(struct tg3_napi *tnapi)
 
 	while (sw_idx != hw_idx) {
 		struct tg3_tx_ring_info *ri = &tnapi->tx_buffers[sw_idx];
+		bool complete_skb_later = false;
 		struct sk_buff *skb = ri->skb;
 		int i, tx_bug = 0;
 
@@ -6542,18 +6578,21 @@ static void tg3_tx(struct tg3_napi *tnapi)
 
 		if (tnapi->tx_ring[sw_idx].len_flags & TXD_FLAG_HWTSTAMP) {
 			struct skb_shared_hwtstamps timestamp;
-			u64 hwclock = tr32(TG3_TX_TSTAMP_LSB);
-			hwclock |= (u64)tr32(TG3_TX_TSTAMP_MSB) << 32;
+			u64 hwclock;
 
-			tg3_hwclock_to_timestamp(tp, hwclock, &timestamp);
-
-			skb_tstamp_tx(skb, &timestamp);
+			tg3_read_tx_tstamp(tp, &hwclock);
+			if (hwclock != tp->pre_tx_ts) {
+				tg3_hwclock_to_timestamp(tp, hwclock, &timestamp);
+				skb_tstamp_tx(skb, &timestamp);
+				tp->pre_tx_ts = 0;
+			} else {
+				tp->tx_tstamp_skb = skb;
+				complete_skb_later = true;
+			}
 		}
 
-		pci_unmap_single(tp->pdev,
-				 dma_unmap_addr(ri, mapping),
-				 skb_headlen(skb),
-				 PCI_DMA_TODEVICE);
+		dma_unmap_single(&tp->pdev->dev, dma_unmap_addr(ri, mapping),
+				 skb_headlen(skb), DMA_TO_DEVICE);
 
 		ri->skb = NULL;
 
@@ -6570,10 +6609,10 @@ static void tg3_tx(struct tg3_napi *tnapi)
 			if (unlikely(ri->skb != NULL || sw_idx == hw_idx))
 				tx_bug = 1;
 
-			pci_unmap_page(tp->pdev,
+			dma_unmap_page(&tp->pdev->dev,
 				       dma_unmap_addr(ri, mapping),
 				       skb_frag_size(&skb_shinfo(skb)->frags[i]),
-				       PCI_DMA_TODEVICE);
+				       DMA_TO_DEVICE);
 
 			while (ri->fragmented) {
 				ri->fragmented = false;
@@ -6587,7 +6626,10 @@ static void tg3_tx(struct tg3_napi *tnapi)
 		pkts_compl++;
 		bytes_compl += skb->len;
 
-		dev_kfree_skb_any(skb);
+		if (!complete_skb_later)
+			dev_consume_skb_any(skb);
+		else
+			ptp_schedule_worker(tp->ptp_clock, 0);
 
 		if (unlikely(tx_bug)) {
 			tg3_tx_recover(tp);
@@ -6599,9 +6641,9 @@ static void tg3_tx(struct tg3_napi *tnapi)
 
 	tnapi->tx_cons = sw_idx;
 
-	/* Need to make the tx_cons update visible to tg3_start_xmit()
+	/* Need to make the tx_cons update visible to __tg3_start_xmit()
 	 * before checking for netif_queue_stopped().  Without the
-	 * memory barrier, there is a small possibility that tg3_start_xmit()
+	 * memory barrier, there is a small possibility that __tg3_start_xmit()
 	 * will miss it and cause the queue to be stopped forever.
 	 */
 	smp_mb();
@@ -6632,8 +6674,8 @@ static void tg3_rx_data_free(struct tg3 *tp, struct ring_info *ri, u32 map_sz)
 	if (!ri->data)
 		return;
 
-	pci_unmap_single(tp->pdev, dma_unmap_addr(ri, mapping),
-			 map_sz, PCI_DMA_FROMDEVICE);
+	dma_unmap_single(&tp->pdev->dev, dma_unmap_addr(ri, mapping), map_sz,
+			 DMA_FROM_DEVICE);
 	tg3_frag_free(skb_size <= PAGE_SIZE, ri->data);
 	ri->data = NULL;
 }
@@ -6644,7 +6686,7 @@ static void tg3_rx_data_free(struct tg3 *tp, struct ring_info *ri, u32 map_sz)
  * We only need to fill in the address because the other members
  * of the RX descriptor are invariant, see tg3_init_rings.
  *
- * Note the purposeful assymetry of cpu vs. chip accesses.  For
+ * Note the purposeful asymmetry of cpu vs. chip accesses.  For
  * posting buffers we only dirty the first cache line of the RX
  * descriptor (containing the address).  Whereas for the RX status
  * buffers the cpu only reads the last cacheline of the RX descriptor
@@ -6688,7 +6730,7 @@ static int tg3_alloc_rx_data(struct tg3 *tp, struct tg3_rx_prodring_set *tpr,
 	skb_size = SKB_DATA_ALIGN(data_size + TG3_RX_OFFSET(tp)) +
 		   SKB_DATA_ALIGN(sizeof(struct skb_shared_info));
 	if (skb_size <= PAGE_SIZE) {
-		data = netdev_alloc_frag(skb_size);
+		data = napi_alloc_frag(skb_size);
 		*frag_size = skb_size;
 	} else {
 		data = kmalloc(skb_size, GFP_ATOMIC);
@@ -6697,11 +6739,9 @@ static int tg3_alloc_rx_data(struct tg3 *tp, struct tg3_rx_prodring_set *tpr,
 	if (!data)
 		return -ENOMEM;
 
-	mapping = pci_map_single(tp->pdev,
-				 data + TG3_RX_OFFSET(tp),
-				 data_size,
-				 PCI_DMA_FROMDEVICE);
-	if (unlikely(pci_dma_mapping_error(tp->pdev, mapping))) {
+	mapping = dma_map_single(&tp->pdev->dev, data + TG3_RX_OFFSET(tp),
+				 data_size, DMA_FROM_DEVICE);
+	if (unlikely(dma_mapping_error(&tp->pdev->dev, mapping))) {
 		tg3_frag_free(skb_size <= PAGE_SIZE, data);
 		return -EIO;
 	}
@@ -6843,7 +6883,7 @@ static int tg3_rx(struct tg3_napi *tnapi, int budget)
 				       desc_idx, *post_ptr);
 		drop_it_no_recycle:
 			/* Other statistics kept track of by card. */
-			tp->rx_dropped++;
+			tnapi->rx_dropped++;
 			goto next_pkt;
 		}
 
@@ -6868,8 +6908,8 @@ static int tg3_rx(struct tg3_napi *tnapi, int budget)
 			if (skb_size < 0)
 				goto drop_it;
 
-			pci_unmap_single(tp->pdev, dma_addr, skb_size,
-					 PCI_DMA_FROMDEVICE);
+			dma_unmap_single(&tp->pdev->dev, dma_addr, skb_size,
+					 DMA_FROM_DEVICE);
 
 			/* Ensure that the update to the data happens
 			 * after the usage of the old DMA mapping.
@@ -6878,7 +6918,10 @@ static int tg3_rx(struct tg3_napi *tnapi, int budget)
 
 			ri->data = NULL;
 
-			skb = build_skb(data, frag_size);
+			if (frag_size)
+				skb = build_skb(data, frag_size);
+			else
+				skb = slab_build_skb(data);
 			if (!skb) {
 				tg3_frag_free(frag_size != 0, data);
 				goto drop_it_no_recycle;
@@ -6894,11 +6937,13 @@ static int tg3_rx(struct tg3_napi *tnapi, int budget)
 				goto drop_it_no_recycle;
 
 			skb_reserve(skb, TG3_RAW_IP_ALIGN);
-			pci_dma_sync_single_for_cpu(tp->pdev, dma_addr, len, PCI_DMA_FROMDEVICE);
+			dma_sync_single_for_cpu(&tp->pdev->dev, dma_addr, len,
+						DMA_FROM_DEVICE);
 			memcpy(skb->data,
 			       data + TG3_RX_OFFSET(tp),
 			       len);
-			pci_dma_sync_single_for_device(tp->pdev, dma_addr, len, PCI_DMA_FROMDEVICE);
+			dma_sync_single_for_device(&tp->pdev->dev, dma_addr,
+						   len, DMA_FROM_DEVICE);
 		}
 
 		skb_put(skb, len);
@@ -6976,7 +7021,6 @@ next_pkt_nopost:
 			tw32_rx_mbox(TG3_RX_JMB_PROD_IDX_REG,
 				     tpr->rx_jmb_prod_idx);
 		}
-		mmiowb();
 	} else if (work_mask) {
 		/* rx_std_buffers[] and rx_jmb_buffers[] entries must be
 		 * updated before the producer indices can be updated.
@@ -7187,8 +7231,6 @@ static int tg3_poll_work(struct tg3_napi *tnapi, int work_done, int budget)
 			tw32_rx_mbox(TG3_RX_JMB_PROD_IDX_REG,
 				     dpr->rx_jmb_prod_idx);
 
-		mmiowb();
-
 		if (err)
 			tw32_f(HOSTCC_MODE, tp->coal_now);
 	}
@@ -7204,8 +7246,8 @@ static inline void tg3_reset_task_schedule(struct tg3 *tp)
 
 static inline void tg3_reset_task_cancel(struct tg3 *tp)
 {
-	cancel_work_sync(&tp->reset_task);
-	tg3_flag_clear(tp, RESET_TASK_PENDING);
+	if (test_and_clear_bit(TG3_FLAG_RESET_TASK_PENDING, tp->tg3_flags))
+		cancel_work_sync(&tp->reset_task);
 	tg3_flag_clear(tp, TX_RECOVERY_PENDING);
 }
 
@@ -7255,11 +7297,11 @@ static int tg3_poll_msix(struct napi_struct *napi, int budget)
 						  HOSTCC_MODE_ENABLE |
 						  tnapi->coal_now);
 			}
-			mmiowb();
 			break;
 		}
 	}
 
+	tg3_send_ape_heartbeat(tp, TG3_APE_HB_INTERVAL << 1);
 	return work_done;
 
 tx_recovery:
@@ -7342,6 +7384,7 @@ static int tg3_poll(struct napi_struct *napi, int budget)
 		}
 	}
 
+	tg3_send_ape_heartbeat(tp, TG3_APE_HB_INTERVAL << 1);
 	return work_done;
 
 tx_recovery:
@@ -7353,27 +7396,61 @@ tx_recovery:
 
 static void tg3_napi_disable(struct tg3 *tp)
 {
+	int txq_idx = tp->txq_cnt - 1;
+	int rxq_idx = tp->rxq_cnt - 1;
+	struct tg3_napi *tnapi;
 	int i;
 
-	for (i = tp->irq_cnt - 1; i >= 0; i--)
-		napi_disable(&tp->napi[i].napi);
+	for (i = tp->irq_cnt - 1; i >= 0; i--) {
+		tnapi = &tp->napi[i];
+		if (tnapi->tx_buffers) {
+			netif_queue_set_napi(tp->dev, txq_idx,
+					     NETDEV_QUEUE_TYPE_TX, NULL);
+			txq_idx--;
+		}
+		if (tnapi->rx_rcb) {
+			netif_queue_set_napi(tp->dev, rxq_idx,
+					     NETDEV_QUEUE_TYPE_RX, NULL);
+			rxq_idx--;
+		}
+		napi_disable(&tnapi->napi);
+	}
 }
 
 static void tg3_napi_enable(struct tg3 *tp)
 {
+	int txq_idx = 0, rxq_idx = 0;
+	struct tg3_napi *tnapi;
 	int i;
 
-	for (i = 0; i < tp->irq_cnt; i++)
-		napi_enable(&tp->napi[i].napi);
+	for (i = 0; i < tp->irq_cnt; i++) {
+		tnapi = &tp->napi[i];
+		napi_enable_locked(&tnapi->napi);
+		if (tnapi->tx_buffers) {
+			netif_queue_set_napi(tp->dev, txq_idx,
+					     NETDEV_QUEUE_TYPE_TX,
+					     &tnapi->napi);
+			txq_idx++;
+		}
+		if (tnapi->rx_rcb) {
+			netif_queue_set_napi(tp->dev, rxq_idx,
+					     NETDEV_QUEUE_TYPE_RX,
+					     &tnapi->napi);
+			rxq_idx++;
+		}
+	}
 }
 
 static void tg3_napi_init(struct tg3 *tp)
 {
 	int i;
 
-	netif_napi_add(tp->dev, &tp->napi[0].napi, tg3_poll, 64);
-	for (i = 1; i < tp->irq_cnt; i++)
-		netif_napi_add(tp->dev, &tp->napi[i].napi, tg3_poll_msix, 64);
+	for (i = 0; i < tp->irq_cnt; i++) {
+		netif_napi_add_locked(tp->dev, &tp->napi[i].napi,
+				      i ? tg3_poll_msix : tg3_poll);
+		netif_napi_set_irq_locked(&tp->napi[i].napi,
+					  tp->napi[i].irq_vec);
+	}
 }
 
 static void tg3_napi_fini(struct tg3 *tp)
@@ -7621,7 +7698,7 @@ static void tg3_poll_controller(struct net_device *dev)
 }
 #endif
 
-static void tg3_tx_timeout(struct net_device *dev)
+static void tg3_tx_timeout(struct net_device *dev, unsigned int txqueue)
 {
 	struct tg3 *tp = netdev_priv(dev);
 
@@ -7750,10 +7827,8 @@ static void tg3_tx_skb_unmap(struct tg3_napi *tnapi, u32 entry, int last)
 	skb = txb->skb;
 	txb->skb = NULL;
 
-	pci_unmap_single(tnapi->tp->pdev,
-			 dma_unmap_addr(txb, mapping),
-			 skb_headlen(skb),
-			 PCI_DMA_TODEVICE);
+	dma_unmap_single(&tnapi->tp->pdev->dev, dma_unmap_addr(txb, mapping),
+			 skb_headlen(skb), DMA_TO_DEVICE);
 
 	while (txb->fragmented) {
 		txb->fragmented = false;
@@ -7767,9 +7842,9 @@ static void tg3_tx_skb_unmap(struct tg3_napi *tnapi, u32 entry, int last)
 		entry = NEXT_TX(entry);
 		txb = &tnapi->tx_buffers[entry];
 
-		pci_unmap_page(tnapi->tp->pdev,
+		dma_unmap_page(&tnapi->tp->pdev->dev,
 			       dma_unmap_addr(txb, mapping),
-			       skb_frag_size(frag), PCI_DMA_TODEVICE);
+			       skb_frag_size(frag), DMA_TO_DEVICE);
 
 		while (txb->fragmented) {
 			txb->fragmented = false;
@@ -7804,10 +7879,10 @@ static int tigon3_dma_hwbug_workaround(struct tg3_napi *tnapi,
 		ret = -1;
 	} else {
 		/* New SKB is guaranteed to be linear. */
-		new_addr = pci_map_single(tp->pdev, new_skb->data, new_skb->len,
-					  PCI_DMA_TODEVICE);
+		new_addr = dma_map_single(&tp->pdev->dev, new_skb->data,
+					  new_skb->len, DMA_TO_DEVICE);
 		/* Make sure the mapping succeeded */
-		if (pci_dma_mapping_error(tp->pdev, new_addr)) {
+		if (dma_mapping_error(&tp->pdev->dev, new_addr)) {
 			dev_kfree_skb_any(new_skb);
 			ret = -1;
 		} else {
@@ -7829,7 +7904,7 @@ static int tigon3_dma_hwbug_workaround(struct tg3_napi *tnapi,
 		}
 	}
 
-	dev_kfree_skb_any(skb);
+	dev_consume_skb_any(skb);
 	*pskb = new_skb;
 	return ret;
 }
@@ -7842,7 +7917,7 @@ static bool tg3_tso_bug_gso_check(struct tg3_napi *tnapi, struct sk_buff *skb)
 	return skb_shinfo(skb)->gso_segs < tnapi->tx_pending / 3;
 }
 
-static netdev_tx_t tg3_start_xmit(struct sk_buff *, struct net_device *);
+static netdev_tx_t __tg3_start_xmit(struct sk_buff *, struct net_device *);
 
 /* Use GSO to workaround all TSO packets that meet HW bug conditions
  * indicated in tg3_tx_frag_set()
@@ -7850,8 +7925,8 @@ static netdev_tx_t tg3_start_xmit(struct sk_buff *, struct net_device *);
 static int tg3_tso_bug(struct tg3 *tp, struct tg3_napi *tnapi,
 		       struct netdev_queue *txq, struct sk_buff *skb)
 {
-	struct sk_buff *segs, *nskb;
 	u32 frag_cnt_est = skb_shinfo(skb)->gso_segs * 3;
+	struct sk_buff *segs, *seg, *next;
 
 	/* Estimate the number of fragments in the worst case */
 	if (unlikely(tg3_tx_avail(tnapi) <= frag_cnt_est)) {
@@ -7871,24 +7946,24 @@ static int tg3_tso_bug(struct tg3 *tp, struct tg3_napi *tnapi,
 
 	segs = skb_gso_segment(skb, tp->dev->features &
 				    ~(NETIF_F_TSO | NETIF_F_TSO6));
-	if (IS_ERR(segs) || !segs)
+	if (IS_ERR(segs) || !segs) {
+		tnapi->tx_dropped++;
 		goto tg3_tso_bug_end;
+	}
 
-	do {
-		nskb = segs;
-		segs = segs->next;
-		nskb->next = NULL;
-		tg3_start_xmit(nskb, tp->dev);
-	} while (segs);
+	skb_list_walk_safe(segs, seg, next) {
+		skb_mark_not_on_list(seg);
+		__tg3_start_xmit(seg, tp->dev);
+	}
 
 tg3_tso_bug_end:
-	dev_kfree_skb_any(skb);
+	dev_consume_skb_any(skb);
 
 	return NETDEV_TX_OK;
 }
 
 /* hard_start_xmit for all devices */
-static netdev_tx_t tg3_start_xmit(struct sk_buff *skb, struct net_device *dev)
+static netdev_tx_t __tg3_start_xmit(struct sk_buff *skb, struct net_device *dev)
 {
 	struct tg3 *tp = netdev_priv(dev);
 	u32 len, entry, base_flags, mss, vlan = 0;
@@ -7939,7 +8014,7 @@ static netdev_tx_t tg3_start_xmit(struct sk_buff *skb, struct net_device *dev)
 		iph = ip_hdr(skb);
 		tcp_opt_len = tcp_optlen(skb);
 
-		hdr_len = skb_transport_offset(skb) + tcp_hdrlen(skb) - ETH_HLEN;
+		hdr_len = skb_tcp_all_headers(skb) - ETH_HLEN;
 
 		/* HW/FW can not correctly segment packets that have been
 		 * vlan encapsulated.
@@ -8027,14 +8102,20 @@ static netdev_tx_t tg3_start_xmit(struct sk_buff *skb, struct net_device *dev)
 
 	if ((unlikely(skb_shinfo(skb)->tx_flags & SKBTX_HW_TSTAMP)) &&
 	    tg3_flag(tp, TX_TSTAMP_EN)) {
-		skb_shinfo(skb)->tx_flags |= SKBTX_IN_PROGRESS;
-		base_flags |= TXD_FLAG_HWTSTAMP;
+		tg3_full_lock(tp, 0);
+		if (!tp->pre_tx_ts) {
+			skb_shinfo(skb)->tx_flags |= SKBTX_IN_PROGRESS;
+			base_flags |= TXD_FLAG_HWTSTAMP;
+			tg3_read_tx_tstamp(tp, &tp->pre_tx_ts);
+		}
+		tg3_full_unlock(tp);
 	}
 
 	len = skb_headlen(skb);
 
-	mapping = pci_map_single(tp->pdev, skb->data, len, PCI_DMA_TODEVICE);
-	if (pci_dma_mapping_error(tp->pdev, mapping))
+	mapping = dma_map_single(&tp->pdev->dev, skb->data, len,
+				 DMA_TO_DEVICE);
+	if (dma_mapping_error(&tp->pdev->dev, mapping))
 		goto drop;
 
 
@@ -8131,12 +8212,6 @@ static netdev_tx_t tg3_start_xmit(struct sk_buff *skb, struct net_device *dev)
 			netif_tx_wake_queue(txq);
 	}
 
-	if (!skb->xmit_more || netif_xmit_stopped(txq)) {
-		/* Packets are ready, update Tx producer idx on card. */
-		tw32_tx_mbox(tnapi->prodmbox, entry);
-		mmiowb();
-	}
-
 	return NETDEV_TX_OK;
 
 dma_error:
@@ -8145,10 +8220,46 @@ dma_error:
 drop:
 	dev_kfree_skb_any(skb);
 drop_nofree:
-	tp->tx_dropped++;
+	tnapi->tx_dropped++;
 	return NETDEV_TX_OK;
 }
 
+static netdev_tx_t tg3_start_xmit(struct sk_buff *skb, struct net_device *dev)
+{
+	struct netdev_queue *txq;
+	u16 skb_queue_mapping;
+	netdev_tx_t ret;
+
+	skb_queue_mapping = skb_get_queue_mapping(skb);
+	txq = netdev_get_tx_queue(dev, skb_queue_mapping);
+
+	ret = __tg3_start_xmit(skb, dev);
+
+	/* Notify the hardware that packets are ready by updating the TX ring
+	 * tail pointer. We respect netdev_xmit_more() thus avoiding poking
+	 * the hardware for every packet. To guarantee forward progress the TX
+	 * ring must be drained when it is full as indicated by
+	 * netif_xmit_stopped(). This needs to happen even when the current
+	 * skb was dropped or rejected with NETDEV_TX_BUSY. Otherwise packets
+	 * queued by previous __tg3_start_xmit() calls might get stuck in
+	 * the queue forever.
+	 */
+	if (!netdev_xmit_more() || netif_xmit_stopped(txq)) {
+		struct tg3_napi *tnapi;
+		struct tg3 *tp;
+
+		tp = netdev_priv(dev);
+		tnapi = &tp->napi[skb_queue_mapping];
+
+		if (tg3_flag(tp, ENABLE_TSS))
+			tnapi++;
+
+		tw32_tx_mbox(tnapi->prodmbox, tnapi->tx_prod);
+	}
+
+	return ret;
+}
+
 static void tg3_mac_loopback(struct tg3 *tp, bool enable)
 {
 	if (enable) {
@@ -8543,7 +8654,7 @@ static void tg3_free_rings(struct tg3 *tp)
 			tg3_tx_skb_unmap(tnapi, i,
 					 skb_shinfo(skb)->nr_frags - 1);
 
-			dev_kfree_skb_any(skb);
+			dev_consume_skb_any(skb);
 		}
 		netdev_tx_reset_queue(netdev_get_tx_queue(tp->dev, j));
 	}
@@ -8621,8 +8732,9 @@ static int tg3_mem_tx_acquire(struct tg3 *tp)
 		tnapi++;
 
 	for (i = 0; i < tp->txq_cnt; i++, tnapi++) {
-		tnapi->tx_buffers = kzalloc(sizeof(struct tg3_tx_ring_info) *
-					    TG3_TX_RING_SIZE, GFP_KERNEL);
+		tnapi->tx_buffers = kcalloc(TG3_TX_RING_SIZE,
+					    sizeof(struct tg3_tx_ring_info),
+					    GFP_KERNEL);
 		if (!tnapi->tx_buffers)
 			goto err_out;
 
@@ -8686,10 +8798,10 @@ static int tg3_mem_rx_acquire(struct tg3 *tp)
 		if (!i && tg3_flag(tp, ENABLE_RSS))
 			continue;
 
-		tnapi->rx_rcb = dma_zalloc_coherent(&tp->pdev->dev,
-						    TG3_RX_RCB_RING_BYTES(tp),
-						    &tnapi->rx_rcb_mapping,
-						    GFP_KERNEL);
+		tnapi->rx_rcb = dma_alloc_coherent(&tp->pdev->dev,
+						   TG3_RX_RCB_RING_BYTES(tp),
+						   &tnapi->rx_rcb_mapping,
+						   GFP_KERNEL);
 		if (!tnapi->rx_rcb)
 			goto err_out;
 	}
@@ -8723,14 +8835,15 @@ static void tg3_free_consistent(struct tg3 *tp)
 	tg3_mem_rx_release(tp);
 	tg3_mem_tx_release(tp);
 
-	/* Protect tg3_get_stats64() from reading freed tp->hw_stats. */
-	tg3_full_lock(tp, 0);
+	/* tp->hw_stats can be referenced safely:
+	 *     1. under rtnl_lock
+	 *     2. or under tp->lock if TG3_FLAG_INIT_COMPLETE is set.
+	 */
 	if (tp->hw_stats) {
 		dma_free_coherent(&tp->pdev->dev, sizeof(struct tg3_hw_stats),
 				  tp->hw_stats, tp->stats_mapping);
 		tp->hw_stats = NULL;
 	}
-	tg3_full_unlock(tp);
 }
 
 /*
@@ -8741,9 +8854,9 @@ static int tg3_alloc_consistent(struct tg3 *tp)
 {
 	int i;
 
-	tp->hw_stats = dma_zalloc_coherent(&tp->pdev->dev,
-					   sizeof(struct tg3_hw_stats),
-					   &tp->stats_mapping, GFP_KERNEL);
+	tp->hw_stats = dma_alloc_coherent(&tp->pdev->dev,
+					  sizeof(struct tg3_hw_stats),
+					  &tp->stats_mapping, GFP_KERNEL);
 	if (!tp->hw_stats)
 		goto err_out;
 
@@ -8751,10 +8864,10 @@ static int tg3_alloc_consistent(struct tg3 *tp)
 		struct tg3_napi *tnapi = &tp->napi[i];
 		struct tg3_hw_status *sblk;
 
-		tnapi->hw_status = dma_zalloc_coherent(&tp->pdev->dev,
-						       TG3_HW_STATUS_SIZE,
-						       &tnapi->status_mapping,
-						       GFP_KERNEL);
+		tnapi->hw_status = dma_alloc_coherent(&tp->pdev->dev,
+						      TG3_HW_STATUS_SIZE,
+						      &tnapi->status_mapping,
+						      GFP_KERNEL);
 		if (!tnapi->hw_status)
 			goto err_out;
 
@@ -9278,6 +9391,15 @@ static int tg3_chip_reset(struct tg3 *tp)
 
 	tg3_restore_clk(tp);
 
+	/* Increase the core clock speed to fix tx timeout issue for 5762
+	 * with 100Mbps link speed.
+	 */
+	if (tg3_asic_rev(tp) == ASIC_REV_5762) {
+		val = tr32(TG3_CPMU_CLCK_ORIDE_ENABLE);
+		tw32(TG3_CPMU_CLCK_ORIDE_ENABLE, val |
+		     TG3_CPMU_MAC_ORIDE_ENABLE);
+	}
+
 	/* Reprobe ASF enable state.  */
 	tg3_flag_clear(tp, ENABLE_ASF);
 	tp->phy_flags &= ~(TG3_PHYFLG_1G_ON_VAUX_OK |
@@ -9313,7 +9435,7 @@ static void __tg3_set_rx_mode(struct net_device *);
 /* tp->lock is held. */
 static int tg3_halt(struct tg3 *tp, int kind, bool silent)
 {
-	int err;
+	int err, i;
 
 	tg3_stop_fw(tp);
 
@@ -9334,6 +9456,13 @@ static int tg3_halt(struct tg3 *tp, int kind, bool silent)
 
 		/* And make sure the next sample is new data */
 		memset(tp->hw_stats, 0, sizeof(struct tg3_hw_stats));
+
+		for (i = 0; i < TG3_IRQ_MAX_VECS; ++i) {
+			struct tg3_napi *tnapi = &tp->napi[i];
+
+			tnapi->rx_dropped = 0;
+			tnapi->tx_dropped = 0;
+		}
 	}
 
 	return err;
@@ -9349,7 +9478,7 @@ static int tg3_set_mac_addr(struct net_device *dev, void *p)
 	if (!is_valid_ether_addr(addr->sa_data))
 		return -EADDRNOTAVAIL;
 
-	memcpy(dev->dev_addr, addr->sa_data, dev->addr_len);
+	eth_hw_addr_set(dev, addr->sa_data);
 
 	if (!netif_running(dev))
 		return 0;
@@ -9680,26 +9809,7 @@ static void tg3_setup_rxbd_thresholds(struct tg3 *tp)
 
 static inline u32 calc_crc(unsigned char *buf, int len)
 {
-	u32 reg;
-	u32 tmp;
-	int j, k;
-
-	reg = 0xffffffff;
-
-	for (j = 0; j < len; j++) {
-		reg ^= buf[j];
-
-		for (k = 0; k < 8; k++) {
-			tmp = reg & 0x01;
-
-			reg >>= 1;
-
-			if (tmp)
-				reg ^= 0xedb88320;
-		}
-	}
-
-	return ~reg;
+	return ~crc32(~0, buf, len);
 }
 
 static void tg3_set_multi(struct tg3 *tp, unsigned int accept_all)
@@ -10035,7 +10145,7 @@ static int tg3_reset_hw(struct tg3 *tp, bool reset_phy)
 	tp->grc_mode |= GRC_MODE_HOST_SENDBDS;
 
 	/* Pseudo-header checksum is done by hardware logic and not
-	 * the offload processers, so make the chip do the pseudo-
+	 * the offload processors, so make the chip do the pseudo-
 	 * header checksums on receive.  For transmit it is more
 	 * convenient to do the pseudo-header checksum in software
 	 * as Linux does that on transmit for us in all cases.
@@ -10052,6 +10162,16 @@ static int tg3_reset_hw(struct tg3 *tp, bool reset_phy)
 
 	tw32(GRC_MODE, tp->grc_mode | val);
 
+	/* On one of the AMD platform, MRRS is restricted to 4000 because of
+	 * south bridge limitation. As a workaround, Driver is setting MRRS
+	 * to 2048 instead of default 4096.
+	 */
+	if (tp->pdev->subsystem_vendor == PCI_VENDOR_ID_DELL &&
+	    tp->pdev->subsystem_device == TG3PCI_SUBDEVICE_ID_DELL_5762) {
+		val = tr32(TG3PCI_DEV_STATUS_CTRL) & ~MAX_READ_REQ_MASK;
+		tw32(TG3PCI_DEV_STATUS_CTRL, val | MAX_READ_REQ_SIZE_2048);
+	}
+
 	/* Setup the timer prescalar register.  Clock is always 66Mhz. */
 	val = tr32(GRC_MISC_CFG);
 	val &= ~0xff;
@@ -10246,8 +10366,7 @@ static int tg3_reset_hw(struct tg3 *tp, bool reset_phy)
 
 	if (tg3_asic_rev(tp) == ASIC_REV_5705 &&
 	    tg3_chip_rev_id(tp) != CHIPREV_ID_5705_A0) {
-		if (tg3_flag(tp, TSO_CAPABLE) &&
-		    tg3_asic_rev(tp) == ASIC_REV_5705) {
+		if (tg3_flag(tp, TSO_CAPABLE)) {
 			rdmac_mode |= RDMAC_MODE_FIFO_SIZE_128;
 		} else if (!(tr32(TG3PCI_PCISTATE) & PCISTATE_BUS_SPEED_HIGH) &&
 			   !tg3_flag(tp, IS_5788)) {
@@ -10684,28 +10803,40 @@ static int tg3_reset_hw(struct tg3 *tp, bool reset_phy)
 	switch (limit) {
 	case 16:
 		tw32(MAC_RCV_RULE_15,  0); tw32(MAC_RCV_VALUE_15,  0);
+		fallthrough;
 	case 15:
 		tw32(MAC_RCV_RULE_14,  0); tw32(MAC_RCV_VALUE_14,  0);
+		fallthrough;
 	case 14:
 		tw32(MAC_RCV_RULE_13,  0); tw32(MAC_RCV_VALUE_13,  0);
+		fallthrough;
 	case 13:
 		tw32(MAC_RCV_RULE_12,  0); tw32(MAC_RCV_VALUE_12,  0);
+		fallthrough;
 	case 12:
 		tw32(MAC_RCV_RULE_11,  0); tw32(MAC_RCV_VALUE_11,  0);
+		fallthrough;
 	case 11:
 		tw32(MAC_RCV_RULE_10,  0); tw32(MAC_RCV_VALUE_10,  0);
+		fallthrough;
 	case 10:
 		tw32(MAC_RCV_RULE_9,  0); tw32(MAC_RCV_VALUE_9,  0);
+		fallthrough;
 	case 9:
 		tw32(MAC_RCV_RULE_8,  0); tw32(MAC_RCV_VALUE_8,  0);
+		fallthrough;
 	case 8:
 		tw32(MAC_RCV_RULE_7,  0); tw32(MAC_RCV_VALUE_7,  0);
+		fallthrough;
 	case 7:
 		tw32(MAC_RCV_RULE_6,  0); tw32(MAC_RCV_VALUE_6,  0);
+		fallthrough;
 	case 6:
 		tw32(MAC_RCV_RULE_5,  0); tw32(MAC_RCV_VALUE_5,  0);
+		fallthrough;
 	case 5:
 		tw32(MAC_RCV_RULE_4,  0); tw32(MAC_RCV_VALUE_4,  0);
+		fallthrough;
 	case 4:
 		/* tw32(MAC_RCV_RULE_3,  0); tw32(MAC_RCV_VALUE_3,  0); */
 	case 3:
@@ -10720,7 +10851,7 @@ static int tg3_reset_hw(struct tg3 *tp, bool reset_phy)
 	if (tg3_flag(tp, ENABLE_APE))
 		/* Write our heartbeat update interval to APE. */
 		tg3_ape_write32(tp, TG3_APE_HOST_HEARTBEAT_INT_MS,
-				APE_HOST_HEARTBEAT_INT_DISABLE);
+				APE_HOST_HEARTBEAT_INT_5SEC);
 
 	tg3_write_sig_post_reset(tp, RESET_KIND_INIT);
 
@@ -10749,17 +10880,15 @@ static int tg3_init_hw(struct tg3 *tp, bool reset_phy)
 #ifdef CONFIG_TIGON3_HWMON
 static void tg3_sd_scan_scratchpad(struct tg3 *tp, struct tg3_ocir *ocir)
 {
+	u32 off, len = TG3_OCIR_LEN;
 	int i;
 
-	for (i = 0; i < TG3_SD_NUM_RECS; i++, ocir++) {
-		u32 off = i * TG3_OCIR_LEN, len = TG3_OCIR_LEN;
-
+	for (i = 0, off = 0; i < TG3_SD_NUM_RECS; i++, ocir++, off += len) {
 		tg3_ape_scratchpad_read(tp, (u32 *) ocir, off, len);
-		off += len;
 
 		if (ocir->signature != TG3_OCIR_SIG_MAGIC ||
 		    !(ocir->version_flags & TG3_OCIR_FLAG_ACTIVE))
-			memset(ocir, 0, TG3_OCIR_LEN);
+			memset(ocir, 0, len);
 	}
 }
 
@@ -10779,11 +10908,11 @@ static ssize_t tg3_show_temp(struct device *dev,
 }
 
 
-static SENSOR_DEVICE_ATTR(temp1_input, S_IRUGO, tg3_show_temp, NULL,
+static SENSOR_DEVICE_ATTR(temp1_input, 0444, tg3_show_temp, NULL,
 			  TG3_TEMP_SENSOR_OFFSET);
-static SENSOR_DEVICE_ATTR(temp1_crit, S_IRUGO, tg3_show_temp, NULL,
+static SENSOR_DEVICE_ATTR(temp1_crit, 0444, tg3_show_temp, NULL,
 			  TG3_TEMP_CAUTION_OFFSET);
-static SENSOR_DEVICE_ATTR(temp1_max, S_IRUGO, tg3_show_temp, NULL,
+static SENSOR_DEVICE_ATTR(temp1_max, 0444, tg3_show_temp, NULL,
 			  TG3_TEMP_MAX_OFFSET);
 
 static struct attribute *tg3_attrs[] = {
@@ -10931,9 +11060,9 @@ static void tg3_chk_missed_msi(struct tg3 *tp)
 	}
 }
 
-static void tg3_timer(unsigned long __opaque)
+static void tg3_timer(struct timer_list *t)
 {
-	struct tg3 *tp = (struct tg3 *) __opaque;
+	struct tg3 *tp = timer_container_of(tp, t, timer);
 
 	spin_lock(&tp->lock);
 
@@ -11065,6 +11194,9 @@ static void tg3_timer(unsigned long __opaque)
 		tp->asf_counter = tp->asf_multiplier;
 	}
 
+	/* Update the APE heartbeat every 5 seconds.*/
+	tg3_send_ape_heartbeat(tp, TG3_APE_HB_INTERVAL);
+
 	spin_unlock(&tp->lock);
 
 restart_timer:
@@ -11087,9 +11219,7 @@ static void tg3_timer_init(struct tg3 *tp)
 	tp->asf_multiplier = (HZ / tp->timer_offset) *
 			     TG3_FW_UPDATE_FREQ_SEC;
 
-	init_timer(&tp->timer);
-	tp->timer.data = (unsigned long) tp;
-	tp->timer.function = tg3_timer;
+	timer_setup(&tp->timer, tg3_timer, 0);
 }
 
 static void tg3_timer_start(struct tg3 *tp)
@@ -11103,7 +11233,7 @@ static void tg3_timer_start(struct tg3 *tp)
 
 static void tg3_timer_stop(struct tg3 *tp)
 {
-	del_timer_sync(&tp->timer);
+	timer_delete_sync(&tp->timer);
 }
 
 /* Restart hardware after configuration changes, self-test, etc.
@@ -11112,6 +11242,8 @@ static void tg3_timer_stop(struct tg3 *tp)
 static int tg3_restart_hw(struct tg3 *tp, bool reset_phy)
 	__releases(tp->lock)
 	__acquires(tp->lock)
+	__releases(tp->dev->lock)
+	__acquires(tp->dev->lock)
 {
 	int err;
 
@@ -11124,7 +11256,9 @@ static int tg3_restart_hw(struct tg3 *tp, bool reset_phy)
 		tg3_timer_stop(tp);
 		tp->irq_sync = 0;
 		tg3_napi_enable(tp);
+		netdev_unlock(tp->dev);
 		dev_close(tp->dev);
+		netdev_lock(tp->dev);
 		tg3_full_lock(tp, 0);
 	}
 	return err;
@@ -11138,7 +11272,8 @@ static void tg3_reset_task(struct work_struct *work)
 	rtnl_lock();
 	tg3_full_lock(tp, 0);
 
-	if (!netif_running(tp->dev)) {
+	if (tp->pcierr_recovery || !netif_running(tp->dev) ||
+	    tp->pdev->error_state != pci_channel_io_normal) {
 		tg3_flag_clear(tp, RESET_TASK_PENDING);
 		tg3_full_unlock(tp);
 		rtnl_unlock();
@@ -11151,6 +11286,7 @@ static void tg3_reset_task(struct work_struct *work)
 
 	tg3_netif_stop(tp);
 
+	netdev_lock(tp->dev);
 	tg3_full_lock(tp, 1);
 
 	if (tg3_flag(tp, TX_RECOVERY_PENDING)) {
@@ -11162,18 +11298,25 @@ static void tg3_reset_task(struct work_struct *work)
 
 	tg3_halt(tp, RESET_KIND_SHUTDOWN, 0);
 	err = tg3_init_hw(tp, true);
-	if (err)
+	if (err) {
+		tg3_full_unlock(tp);
+		tp->irq_sync = 0;
+		tg3_napi_enable(tp);
+		/* Clear this flag so that tg3_reset_task_cancel() will not
+		 * call cancel_work_sync() and wait forever.
+		 */
+		tg3_flag_clear(tp, RESET_TASK_PENDING);
+		netdev_unlock(tp->dev);
+		dev_close(tp->dev);
 		goto out;
+	}
 
 	tg3_netif_start(tp);
-
-out:
 	tg3_full_unlock(tp);
-
-	if (!err)
-		tg3_phy_start(tp);
-
+	netdev_unlock(tp->dev);
+	tg3_phy_start(tp);
 	tg3_flag_clear(tp, RESET_TASK_PENDING);
+out:
 	rtnl_unlock();
 }
 
@@ -11189,18 +11332,17 @@ static int tg3_request_irq(struct tg3 *tp, int irq_num)
 	else {
 		name = &tnapi->irq_lbl[0];
 		if (tnapi->tx_buffers && tnapi->rx_rcb)
-			snprintf(name, IFNAMSIZ,
+			snprintf(name, sizeof(tnapi->irq_lbl),
 				 "%s-txrx-%d", tp->dev->name, irq_num);
 		else if (tnapi->tx_buffers)
-			snprintf(name, IFNAMSIZ,
+			snprintf(name, sizeof(tnapi->irq_lbl),
 				 "%s-tx-%d", tp->dev->name, irq_num);
 		else if (tnapi->rx_rcb)
-			snprintf(name, IFNAMSIZ,
+			snprintf(name, sizeof(tnapi->irq_lbl),
 				 "%s-rx-%d", tp->dev->name, irq_num);
 		else
-			snprintf(name, IFNAMSIZ,
+			snprintf(name, sizeof(tnapi->irq_lbl),
 				 "%s-%d", tp->dev->name, irq_num);
-		name[IFNAMSIZ-1] = 0;
 	}
 
 	if (tg3_flag(tp, USING_MSI) || tg3_flag(tp, USING_MSIX)) {
@@ -11531,16 +11673,18 @@ static int tg3_start(struct tg3 *tp, bool reset_phy, bool test_irq,
 	if (err)
 		goto out_ints_fini;
 
+	netdev_lock(dev);
 	tg3_napi_init(tp);
 
 	tg3_napi_enable(tp);
+	netdev_unlock(dev);
 
 	for (i = 0; i < tp->irq_cnt; i++) {
-		struct tg3_napi *tnapi = &tp->napi[i];
 		err = tg3_request_irq(tp, i);
 		if (err) {
 			for (i--; i >= 0; i--) {
-				tnapi = &tp->napi[i];
+				struct tg3_napi *tnapi = &tp->napi[i];
+
 				free_irq(tnapi->irq_vec, tnapi);
 			}
 			goto out_napi_fini;
@@ -11858,6 +12002,9 @@ static void tg3_get_nstats(struct tg3 *tp, struct rtnl_link_stats64 *stats)
 {
 	struct rtnl_link_stats64 *old_stats = &tp->net_stats_prev;
 	struct tg3_hw_stats *hw_stats = tp->hw_stats;
+	unsigned long rx_dropped;
+	unsigned long tx_dropped;
+	int i;
 
 	stats->rx_packets = old_stats->rx_packets +
 		get_stat64(&hw_stats->rx_ucast_packets) +
@@ -11904,8 +12051,26 @@ static void tg3_get_nstats(struct tg3 *tp, struct rtnl_link_stats64 *stats)
 	stats->rx_missed_errors = old_stats->rx_missed_errors +
 		get_stat64(&hw_stats->rx_discards);
 
-	stats->rx_dropped = tp->rx_dropped;
-	stats->tx_dropped = tp->tx_dropped;
+	/* Aggregate per-queue counters. The per-queue counters are updated
+	 * by a single writer, race-free. The result computed by this loop
+	 * might not be 100% accurate (counters can be updated in the middle of
+	 * the loop) but the next tg3_get_nstats() will recompute the current
+	 * value so it is acceptable.
+	 *
+	 * Note that these counters wrap around at 4G on 32bit machines.
+	 */
+	rx_dropped = (unsigned long)(old_stats->rx_dropped);
+	tx_dropped = (unsigned long)(old_stats->tx_dropped);
+
+	for (i = 0; i < tp->irq_cnt; i++) {
+		struct tg3_napi *tnapi = &tp->napi[i];
+
+		rx_dropped += tnapi->rx_dropped;
+		tx_dropped += tnapi->tx_dropped;
+	}
+
+	stats->rx_dropped = rx_dropped;
+	stats->tx_dropped = tx_dropped;
 }
 
 static int tg3_get_regs_len(struct net_device *dev)
@@ -12097,7 +12262,9 @@ static int tg3_get_link_ksettings(struct net_device *dev,
 		if (!(tp->phy_flags & TG3_PHYFLG_IS_CONNECTED))
 			return -EAGAIN;
 		phydev = mdiobus_get_phy(tp->mdio_bus, tp->phy_addr);
-		return phy_ethtool_ksettings_get(phydev, cmd);
+		phy_ethtool_ksettings_get(phydev, cmd);
+
+		return 0;
 	}
 
 	supported = (SUPPORTED_Autoneg);
@@ -12259,10 +12426,9 @@ static void tg3_get_drvinfo(struct net_device *dev, struct ethtool_drvinfo *info
 {
 	struct tg3 *tp = netdev_priv(dev);
 
-	strlcpy(info->driver, DRV_MODULE_NAME, sizeof(info->driver));
-	strlcpy(info->version, DRV_MODULE_VERSION, sizeof(info->version));
-	strlcpy(info->fw_version, tp->fw_ver, sizeof(info->fw_version));
-	strlcpy(info->bus_info, pci_name(tp->pdev), sizeof(info->bus_info));
+	strscpy(info->driver, DRV_MODULE_NAME, sizeof(info->driver));
+	strscpy(info->fw_version, tp->fw_ver, sizeof(info->fw_version));
+	strscpy(info->bus_info, pci_name(tp->pdev), sizeof(info->bus_info));
 }
 
 static void tg3_get_wol(struct net_device *dev, struct ethtool_wolinfo *wol)
@@ -12348,7 +12514,10 @@ static int tg3_nway_reset(struct net_device *dev)
 	return r;
 }
 
-static void tg3_get_ringparam(struct net_device *dev, struct ethtool_ringparam *ering)
+static void tg3_get_ringparam(struct net_device *dev,
+			      struct ethtool_ringparam *ering,
+			      struct kernel_ethtool_ringparam *kernel_ering,
+			      struct netlink_ext_ack *extack)
 {
 	struct tg3 *tp = netdev_priv(dev);
 
@@ -12369,10 +12538,14 @@ static void tg3_get_ringparam(struct net_device *dev, struct ethtool_ringparam *
 	ering->tx_pending = tp->napi[0].tx_pending;
 }
 
-static int tg3_set_ringparam(struct net_device *dev, struct ethtool_ringparam *ering)
+static int tg3_set_ringparam(struct net_device *dev,
+			     struct ethtool_ringparam *ering,
+			     struct kernel_ethtool_ringparam *kernel_ering,
+			     struct netlink_ext_ack *extack)
 {
 	struct tg3 *tp = netdev_priv(dev);
 	int i, irq_sync = 0, err = 0;
+	bool reset_phy = false;
 
 	if ((ering->rx_pending > tp->rx_std_ring_mask) ||
 	    (ering->rx_jumbo_pending > tp->rx_jmb_ring_mask) ||
@@ -12388,6 +12561,7 @@ static int tg3_set_ringparam(struct net_device *dev, struct ethtool_ringparam *e
 		irq_sync = 1;
 	}
 
+	netdev_lock(dev);
 	tg3_full_lock(tp, irq_sync);
 
 	tp->rx_pending = ering->rx_pending;
@@ -12404,12 +12578,19 @@ static int tg3_set_ringparam(struct net_device *dev, struct ethtool_ringparam *e
 
 	if (netif_running(dev)) {
 		tg3_halt(tp, RESET_KIND_SHUTDOWN, 1);
-		err = tg3_restart_hw(tp, false);
+		/* Reset PHY to avoid PHY lock up */
+		if (tg3_asic_rev(tp) == ASIC_REV_5717 ||
+		    tg3_asic_rev(tp) == ASIC_REV_5719 ||
+		    tg3_asic_rev(tp) == ASIC_REV_5720)
+			reset_phy = true;
+
+		err = tg3_restart_hw(tp, reset_phy);
 		if (!err)
 			tg3_netif_start(tp);
 	}
 
 	tg3_full_unlock(tp);
+	netdev_unlock(dev);
 
 	if (irq_sync && !err)
 		tg3_phy_start(tp);
@@ -12438,36 +12619,30 @@ static int tg3_set_pauseparam(struct net_device *dev, struct ethtool_pauseparam
 {
 	struct tg3 *tp = netdev_priv(dev);
 	int err = 0;
+	bool reset_phy = false;
 
 	if (tp->link_config.autoneg == AUTONEG_ENABLE)
 		tg3_warn_mgmt_link_flap(tp);
 
 	if (tg3_flag(tp, USE_PHYLIB)) {
-		u32 newadv;
 		struct phy_device *phydev;
 
 		phydev = mdiobus_get_phy(tp->mdio_bus, tp->phy_addr);
 
-		if (!(phydev->supported & SUPPORTED_Pause) ||
-		    (!(phydev->supported & SUPPORTED_Asym_Pause) &&
-		     (epause->rx_pause != epause->tx_pause)))
+		if (!phy_validate_pause(phydev, epause))
 			return -EINVAL;
 
 		tp->link_config.flowctrl = 0;
+		phy_set_asym_pause(phydev, epause->rx_pause, epause->tx_pause);
 		if (epause->rx_pause) {
 			tp->link_config.flowctrl |= FLOW_CTRL_RX;
 
 			if (epause->tx_pause) {
 				tp->link_config.flowctrl |= FLOW_CTRL_TX;
-				newadv = ADVERTISED_Pause;
-			} else
-				newadv = ADVERTISED_Pause |
-					 ADVERTISED_Asym_Pause;
+			}
 		} else if (epause->tx_pause) {
 			tp->link_config.flowctrl |= FLOW_CTRL_TX;
-			newadv = ADVERTISED_Asym_Pause;
-		} else
-			newadv = 0;
+		}
 
 		if (epause->autoneg)
 			tg3_flag_set(tp, PAUSE_AUTONEG);
@@ -12475,33 +12650,19 @@ static int tg3_set_pauseparam(struct net_device *dev, struct ethtool_pauseparam
 			tg3_flag_clear(tp, PAUSE_AUTONEG);
 
 		if (tp->phy_flags & TG3_PHYFLG_IS_CONNECTED) {
-			u32 oldadv = phydev->advertising &
-				     (ADVERTISED_Pause | ADVERTISED_Asym_Pause);
-			if (oldadv != newadv) {
-				phydev->advertising &=
-					~(ADVERTISED_Pause |
-					  ADVERTISED_Asym_Pause);
-				phydev->advertising |= newadv;
-				if (phydev->autoneg) {
-					/*
-					 * Always renegotiate the link to
-					 * inform our link partner of our
-					 * flow control settings, even if the
-					 * flow control is forced.  Let
-					 * tg3_adjust_link() do the final
-					 * flow control setup.
-					 */
-					return phy_start_aneg(phydev);
-				}
+			if (phydev->autoneg) {
+				/* phy_set_asym_pause() will
+				 * renegotiate the link to inform our
+				 * link partner of our flow control
+				 * settings, even if the flow control
+				 * is forced.  Let tg3_adjust_link()
+				 * do the final flow control setup.
+				 */
+				return 0;
 			}
 
 			if (!epause->autoneg)
 				tg3_setup_flow_control(tp, 0, 0);
-		} else {
-			tp->link_config.advertising &=
-					~(ADVERTISED_Pause |
-					  ADVERTISED_Asym_Pause);
-			tp->link_config.advertising |= newadv;
 		}
 	} else {
 		int irq_sync = 0;
@@ -12511,6 +12672,7 @@ static int tg3_set_pauseparam(struct net_device *dev, struct ethtool_pauseparam
 			irq_sync = 1;
 		}
 
+		netdev_lock(dev);
 		tg3_full_lock(tp, irq_sync);
 
 		if (epause->autoneg)
@@ -12528,12 +12690,19 @@ static int tg3_set_pauseparam(struct net_device *dev, struct ethtool_pauseparam
 
 		if (netif_running(dev)) {
 			tg3_halt(tp, RESET_KIND_SHUTDOWN, 1);
-			err = tg3_restart_hw(tp, false);
+			/* Reset PHY to avoid PHY lock up */
+			if (tg3_asic_rev(tp) == ASIC_REV_5717 ||
+			    tg3_asic_rev(tp) == ASIC_REV_5719 ||
+			    tg3_asic_rev(tp) == ASIC_REV_5720)
+				reset_phy = true;
+
+			err = tg3_restart_hw(tp, reset_phy);
 			if (!err)
 				tg3_netif_start(tp);
 		}
 
 		tg3_full_unlock(tp);
+		netdev_unlock(dev);
 	}
 
 	tp->phy_flags |= TG3_PHYFLG_USER_CONFIGURED;
@@ -12589,24 +12758,23 @@ static u32 tg3_get_rxfh_indir_size(struct net_device *dev)
 	return size;
 }
 
-static int tg3_get_rxfh(struct net_device *dev, u32 *indir, u8 *key, u8 *hfunc)
+static int tg3_get_rxfh(struct net_device *dev, struct ethtool_rxfh_param *rxfh)
 {
 	struct tg3 *tp = netdev_priv(dev);
 	int i;
 
-	if (hfunc)
-		*hfunc = ETH_RSS_HASH_TOP;
-	if (!indir)
+	rxfh->hfunc = ETH_RSS_HASH_TOP;
+	if (!rxfh->indir)
 		return 0;
 
 	for (i = 0; i < TG3_RSS_INDIR_TBL_SIZE; i++)
-		indir[i] = tp->rss_ind_tbl[i];
+		rxfh->indir[i] = tp->rss_ind_tbl[i];
 
 	return 0;
 }
 
-static int tg3_set_rxfh(struct net_device *dev, const u32 *indir, const u8 *key,
-			const u8 hfunc)
+static int tg3_set_rxfh(struct net_device *dev, struct ethtool_rxfh_param *rxfh,
+			struct netlink_ext_ack *extack)
 {
 	struct tg3 *tp = netdev_priv(dev);
 	size_t i;
@@ -12614,15 +12782,16 @@ static int tg3_set_rxfh(struct net_device *dev, const u32 *indir, const u8 *key,
 	/* We require at least one supported parameter to be changed and no
 	 * change in any of the unsupported parameters
 	 */
-	if (key ||
-	    (hfunc != ETH_RSS_HASH_NO_CHANGE && hfunc != ETH_RSS_HASH_TOP))
+	if (rxfh->key ||
+	    (rxfh->hfunc != ETH_RSS_HASH_NO_CHANGE &&
+	     rxfh->hfunc != ETH_RSS_HASH_TOP))
 		return -EOPNOTSUPP;
 
-	if (!indir)
+	if (!rxfh->indir)
 		return 0;
 
 	for (i = 0; i < TG3_RSS_INDIR_TBL_SIZE; i++)
-		tp->rss_ind_tbl[i] = indir[i];
+		tp->rss_ind_tbl[i] = rxfh->indir[i];
 
 	if (!netif_running(dev) || !tg3_flag(tp, ENABLE_RSS))
 		return 0;
@@ -12709,9 +12878,6 @@ static int tg3_set_phys_id(struct net_device *dev,
 {
 	struct tg3 *tp = netdev_priv(dev);
 
-	if (!netif_running(tp->dev))
-		return -EAGAIN;
-
 	switch (state) {
 	case ETHTOOL_ID_ACTIVE:
 		return 1;	/* cycle on/off once per second */
@@ -12750,7 +12916,7 @@ static void tg3_get_ethtool_stats(struct net_device *dev,
 		memset(tmp_stats, 0, sizeof(struct tg3_ethtool_stats));
 }
 
-static __be32 *tg3_vpd_readblock(struct tg3 *tp, u32 *vpdlen)
+static __be32 *tg3_vpd_readblock(struct tg3 *tp, unsigned int *vpdlen)
 {
 	int i;
 	__be32 *buf;
@@ -12779,18 +12945,16 @@ static __be32 *tg3_vpd_readblock(struct tg3 *tp, u32 *vpdlen)
 
 			offset = tg3_nvram_logical_addr(tp, offset);
 		}
-	}
 
-	if (!offset || !len) {
-		offset = TG3_NVM_VPD_OFF;
-		len = TG3_NVM_VPD_LEN;
-	}
+		if (!offset || !len) {
+			offset = TG3_NVM_VPD_OFF;
+			len = TG3_NVM_VPD_LEN;
+		}
 
-	buf = kmalloc(len, GFP_KERNEL);
-	if (buf == NULL)
-		return NULL;
+		buf = kmalloc(len, GFP_KERNEL);
+		if (!buf)
+			return NULL;
 
-	if (magic == TG3_EEPROM_MAGIC) {
 		for (i = 0; i < len; i += 4) {
 			/* The data is in little-endian format in NVRAM.
 			 * Use the big-endian read routines to preserve
@@ -12799,26 +12963,13 @@ static __be32 *tg3_vpd_readblock(struct tg3 *tp, u32 *vpdlen)
 			if (tg3_nvram_read_be32(tp, offset + i, &buf[i/4]))
 				goto error;
 		}
+		*vpdlen = len;
 	} else {
-		u8 *ptr;
-		ssize_t cnt;
-		unsigned int pos = 0;
-
-		ptr = (u8 *)&buf[0];
-		for (i = 0; pos < len && i < 3; i++, pos += cnt, ptr += cnt) {
-			cnt = pci_read_vpd(tp->pdev, pos,
-					   len - pos, ptr);
-			if (cnt == -ETIMEDOUT || cnt == -EINTR)
-				cnt = 0;
-			else if (cnt < 0)
-				goto error;
-		}
-		if (pos != len)
-			goto error;
+		buf = pci_vpd_alloc(tp->pdev, vpdlen);
+		if (IS_ERR(buf))
+			return NULL;
 	}
 
-	*vpdlen = len;
-
 	return buf;
 
 error:
@@ -12838,9 +12989,10 @@ error:
 
 static int tg3_test_nvram(struct tg3 *tp)
 {
-	u32 csum, magic, len;
+	u32 csum, magic;
 	__be32 *buf;
 	int i, j, k, err = 0, size;
+	unsigned int len;
 
 	if (tg3_flag(tp, NO_NVRAM))
 		return 0;
@@ -12969,12 +13121,16 @@ static int tg3_test_nvram(struct tg3 *tp)
 
 	/* Bootstrap checksum at offset 0x10 */
 	csum = calc_crc((unsigned char *) buf, 0x10);
-	if (csum != le32_to_cpu(buf[0x10/4]))
+
+	/* The type of buf is __be32 *, but this value is __le32 */
+	if (csum != le32_to_cpu((__force __le32)buf[0x10 / 4]))
 		goto out;
 
 	/* Manufacturing block starts at offset 0x74, checksum at 0xfc */
-	csum = calc_crc((unsigned char *) &buf[0x74/4], 0x88);
-	if (csum != le32_to_cpu(buf[0xfc/4]))
+	csum = calc_crc((unsigned char *)&buf[0x74 / 4], 0x88);
+
+	/* The type of buf is __be32 *, but this value is __le32 */
+	if (csum != le32_to_cpu((__force __le32)buf[0xfc / 4]))
 		goto out;
 
 	kfree(buf);
@@ -12983,33 +13139,10 @@ static int tg3_test_nvram(struct tg3 *tp)
 	if (!buf)
 		return -ENOMEM;
 
-	i = pci_vpd_find_tag((u8 *)buf, 0, len, PCI_VPD_LRDT_RO_DATA);
-	if (i > 0) {
-		j = pci_vpd_lrdt_size(&((u8 *)buf)[i]);
-		if (j < 0)
-			goto out;
-
-		if (i + PCI_VPD_LRDT_TAG_SIZE + j > len)
-			goto out;
-
-		i += PCI_VPD_LRDT_TAG_SIZE;
-		j = pci_vpd_find_info_keyword((u8 *)buf, i, j,
-					      PCI_VPD_RO_KEYWORD_CHKSUM);
-		if (j > 0) {
-			u8 csum8 = 0;
-
-			j += PCI_VPD_INFO_FLD_HDR_SIZE;
-
-			for (i = 0; i <= j; i++)
-				csum8 += ((u8 *)buf)[i];
-
-			if (csum8)
-				goto out;
-		}
-	}
-
-	err = 0;
-
+	err = pci_vpd_check_csum(buf, len);
+	/* go on if no checksum found */
+	if (err == 1)
+		err = 0;
 out:
 	kfree(buf);
 	return err;
@@ -13466,8 +13599,8 @@ static int tg3_run_loopback(struct tg3 *tp, u32 pktsz, bool tso_loopback)
 	for (i = data_off; i < tx_len; i++)
 		tx_data[i] = (u8) (i & 0xff);
 
-	map = pci_map_single(tp->pdev, skb->data, tx_len, PCI_DMA_TODEVICE);
-	if (pci_dma_mapping_error(tp->pdev, map)) {
+	map = dma_map_single(&tp->pdev->dev, skb->data, tx_len, DMA_TO_DEVICE);
+	if (dma_mapping_error(&tp->pdev->dev, map)) {
 		dev_kfree_skb(skb);
 		return -EIO;
 	}
@@ -13565,8 +13698,8 @@ static int tg3_run_loopback(struct tg3 *tp, u32 pktsz, bool tso_loopback)
 		} else
 			goto out;
 
-		pci_dma_sync_single_for_cpu(tp->pdev, map, rx_len,
-					    PCI_DMA_FROMDEVICE);
+		dma_sync_single_for_cpu(&tp->pdev->dev, map, rx_len,
+					DMA_FROM_DEVICE);
 
 		rx_data += TG3_RX_OFFSET(tp);
 		for (i = data_off; i < rx_len; i++, val++) {
@@ -13774,6 +13907,7 @@ static void tg3_self_test(struct net_device *dev, struct ethtool_test *etest,
 			data[TG3_INTERRUPT_TEST] = 1;
 		}
 
+		netdev_lock(dev);
 		tg3_full_lock(tp, 0);
 
 		tg3_halt(tp, RESET_KIND_SHUTDOWN, 1);
@@ -13785,6 +13919,7 @@ static void tg3_self_test(struct net_device *dev, struct ethtool_test *etest,
 		}
 
 		tg3_full_unlock(tp);
+		netdev_unlock(dev);
 
 		if (irq_sync && !err2)
 			tg3_phy_start(tp);
@@ -13794,25 +13929,20 @@ static void tg3_self_test(struct net_device *dev, struct ethtool_test *etest,
 
 }
 
-static int tg3_hwtstamp_set(struct net_device *dev, struct ifreq *ifr)
+static int tg3_hwtstamp_set(struct net_device *dev,
+			    struct kernel_hwtstamp_config *stmpconf,
+			    struct netlink_ext_ack *extack)
 {
 	struct tg3 *tp = netdev_priv(dev);
-	struct hwtstamp_config stmpconf;
 
 	if (!tg3_flag(tp, PTP_CAPABLE))
 		return -EOPNOTSUPP;
 
-	if (copy_from_user(&stmpconf, ifr->ifr_data, sizeof(stmpconf)))
-		return -EFAULT;
-
-	if (stmpconf.flags)
-		return -EINVAL;
-
-	if (stmpconf.tx_type != HWTSTAMP_TX_ON &&
-	    stmpconf.tx_type != HWTSTAMP_TX_OFF)
+	if (stmpconf->tx_type != HWTSTAMP_TX_ON &&
+	    stmpconf->tx_type != HWTSTAMP_TX_OFF)
 		return -ERANGE;
 
-	switch (stmpconf.rx_filter) {
+	switch (stmpconf->rx_filter) {
 	case HWTSTAMP_FILTER_NONE:
 		tp->rxptpctl = 0;
 		break;
@@ -13872,74 +14002,72 @@ static int tg3_hwtstamp_set(struct net_device *dev, struct ifreq *ifr)
 		tw32(TG3_RX_PTP_CTL,
 		     tp->rxptpctl | TG3_RX_PTP_CTL_HWTS_INTERLOCK);
 
-	if (stmpconf.tx_type == HWTSTAMP_TX_ON)
+	if (stmpconf->tx_type == HWTSTAMP_TX_ON)
 		tg3_flag_set(tp, TX_TSTAMP_EN);
 	else
 		tg3_flag_clear(tp, TX_TSTAMP_EN);
 
-	return copy_to_user(ifr->ifr_data, &stmpconf, sizeof(stmpconf)) ?
-		-EFAULT : 0;
+	return 0;
 }
 
-static int tg3_hwtstamp_get(struct net_device *dev, struct ifreq *ifr)
+static int tg3_hwtstamp_get(struct net_device *dev,
+			    struct kernel_hwtstamp_config *stmpconf)
 {
 	struct tg3 *tp = netdev_priv(dev);
-	struct hwtstamp_config stmpconf;
 
 	if (!tg3_flag(tp, PTP_CAPABLE))
 		return -EOPNOTSUPP;
 
-	stmpconf.flags = 0;
-	stmpconf.tx_type = (tg3_flag(tp, TX_TSTAMP_EN) ?
-			    HWTSTAMP_TX_ON : HWTSTAMP_TX_OFF);
+	stmpconf->flags = 0;
+	stmpconf->tx_type = tg3_flag(tp, TX_TSTAMP_EN) ?
+			    HWTSTAMP_TX_ON : HWTSTAMP_TX_OFF;
 
 	switch (tp->rxptpctl) {
 	case 0:
-		stmpconf.rx_filter = HWTSTAMP_FILTER_NONE;
+		stmpconf->rx_filter = HWTSTAMP_FILTER_NONE;
 		break;
 	case TG3_RX_PTP_CTL_RX_PTP_V1_EN | TG3_RX_PTP_CTL_ALL_V1_EVENTS:
-		stmpconf.rx_filter = HWTSTAMP_FILTER_PTP_V1_L4_EVENT;
+		stmpconf->rx_filter = HWTSTAMP_FILTER_PTP_V1_L4_EVENT;
 		break;
 	case TG3_RX_PTP_CTL_RX_PTP_V1_EN | TG3_RX_PTP_CTL_SYNC_EVNT:
-		stmpconf.rx_filter = HWTSTAMP_FILTER_PTP_V1_L4_SYNC;
+		stmpconf->rx_filter = HWTSTAMP_FILTER_PTP_V1_L4_SYNC;
 		break;
 	case TG3_RX_PTP_CTL_RX_PTP_V1_EN | TG3_RX_PTP_CTL_DELAY_REQ:
-		stmpconf.rx_filter = HWTSTAMP_FILTER_PTP_V1_L4_DELAY_REQ;
+		stmpconf->rx_filter = HWTSTAMP_FILTER_PTP_V1_L4_DELAY_REQ;
 		break;
 	case TG3_RX_PTP_CTL_RX_PTP_V2_EN | TG3_RX_PTP_CTL_ALL_V2_EVENTS:
-		stmpconf.rx_filter = HWTSTAMP_FILTER_PTP_V2_EVENT;
+		stmpconf->rx_filter = HWTSTAMP_FILTER_PTP_V2_EVENT;
 		break;
 	case TG3_RX_PTP_CTL_RX_PTP_V2_L2_EN | TG3_RX_PTP_CTL_ALL_V2_EVENTS:
-		stmpconf.rx_filter = HWTSTAMP_FILTER_PTP_V2_L2_EVENT;
+		stmpconf->rx_filter = HWTSTAMP_FILTER_PTP_V2_L2_EVENT;
 		break;
 	case TG3_RX_PTP_CTL_RX_PTP_V2_L4_EN | TG3_RX_PTP_CTL_ALL_V2_EVENTS:
-		stmpconf.rx_filter = HWTSTAMP_FILTER_PTP_V2_L4_EVENT;
+		stmpconf->rx_filter = HWTSTAMP_FILTER_PTP_V2_L4_EVENT;
 		break;
 	case TG3_RX_PTP_CTL_RX_PTP_V2_EN | TG3_RX_PTP_CTL_SYNC_EVNT:
-		stmpconf.rx_filter = HWTSTAMP_FILTER_PTP_V2_SYNC;
+		stmpconf->rx_filter = HWTSTAMP_FILTER_PTP_V2_SYNC;
 		break;
 	case TG3_RX_PTP_CTL_RX_PTP_V2_L2_EN | TG3_RX_PTP_CTL_SYNC_EVNT:
-		stmpconf.rx_filter = HWTSTAMP_FILTER_PTP_V2_L2_SYNC;
+		stmpconf->rx_filter = HWTSTAMP_FILTER_PTP_V2_L2_SYNC;
 		break;
 	case TG3_RX_PTP_CTL_RX_PTP_V2_L4_EN | TG3_RX_PTP_CTL_SYNC_EVNT:
-		stmpconf.rx_filter = HWTSTAMP_FILTER_PTP_V2_L4_SYNC;
+		stmpconf->rx_filter = HWTSTAMP_FILTER_PTP_V2_L4_SYNC;
 		break;
 	case TG3_RX_PTP_CTL_RX_PTP_V2_EN | TG3_RX_PTP_CTL_DELAY_REQ:
-		stmpconf.rx_filter = HWTSTAMP_FILTER_PTP_V2_DELAY_REQ;
+		stmpconf->rx_filter = HWTSTAMP_FILTER_PTP_V2_DELAY_REQ;
 		break;
 	case TG3_RX_PTP_CTL_RX_PTP_V2_L2_EN | TG3_RX_PTP_CTL_DELAY_REQ:
-		stmpconf.rx_filter = HWTSTAMP_FILTER_PTP_V2_L2_DELAY_REQ;
+		stmpconf->rx_filter = HWTSTAMP_FILTER_PTP_V2_L2_DELAY_REQ;
 		break;
 	case TG3_RX_PTP_CTL_RX_PTP_V2_L4_EN | TG3_RX_PTP_CTL_DELAY_REQ:
-		stmpconf.rx_filter = HWTSTAMP_FILTER_PTP_V2_L4_DELAY_REQ;
+		stmpconf->rx_filter = HWTSTAMP_FILTER_PTP_V2_L4_DELAY_REQ;
 		break;
 	default:
 		WARN_ON_ONCE(1);
 		return -ERANGE;
 	}
 
-	return copy_to_user(ifr->ifr_data, &stmpconf, sizeof(stmpconf)) ?
-		-EFAULT : 0;
+	return 0;
 }
 
 static int tg3_ioctl(struct net_device *dev, struct ifreq *ifr, int cmd)
@@ -13960,7 +14088,7 @@ static int tg3_ioctl(struct net_device *dev, struct ifreq *ifr, int cmd)
 	case SIOCGMIIPHY:
 		data->phy_id = tp->phy_addr;
 
-		/* fallthru */
+		fallthrough;
 	case SIOCGMIIREG: {
 		u32 mii_regval;
 
@@ -13994,12 +14122,6 @@ static int tg3_ioctl(struct net_device *dev, struct ifreq *ifr, int cmd)
 
 		return err;
 
-	case SIOCSHWTSTAMP:
-		return tg3_hwtstamp_set(dev, ifr);
-
-	case SIOCGHWTSTAMP:
-		return tg3_hwtstamp_get(dev, ifr);
-
 	default:
 		/* do nothing */
 		break;
@@ -14007,7 +14129,10 @@ static int tg3_ioctl(struct net_device *dev, struct ifreq *ifr, int cmd)
 	return -EOPNOTSUPP;
 }
 
-static int tg3_get_coalesce(struct net_device *dev, struct ethtool_coalesce *ec)
+static int tg3_get_coalesce(struct net_device *dev,
+			    struct ethtool_coalesce *ec,
+			    struct kernel_ethtool_coalesce *kernel_coal,
+			    struct netlink_ext_ack *extack)
 {
 	struct tg3 *tp = netdev_priv(dev);
 
@@ -14015,7 +14140,10 @@ static int tg3_get_coalesce(struct net_device *dev, struct ethtool_coalesce *ec)
 	return 0;
 }
 
-static int tg3_set_coalesce(struct net_device *dev, struct ethtool_coalesce *ec)
+static int tg3_set_coalesce(struct net_device *dev,
+			    struct ethtool_coalesce *ec,
+			    struct kernel_ethtool_coalesce *kernel_coal,
+			    struct netlink_ext_ack *extack)
 {
 	struct tg3 *tp = netdev_priv(dev);
 	u32 max_rxcoal_tick_int = 0, max_txcoal_tick_int = 0;
@@ -14061,7 +14189,7 @@ static int tg3_set_coalesce(struct net_device *dev, struct ethtool_coalesce *ec)
 	return 0;
 }
 
-static int tg3_set_eee(struct net_device *dev, struct ethtool_eee *edata)
+static int tg3_set_eee(struct net_device *dev, struct ethtool_keee *edata)
 {
 	struct tg3 *tp = netdev_priv(dev);
 
@@ -14070,7 +14198,7 @@ static int tg3_set_eee(struct net_device *dev, struct ethtool_eee *edata)
 		return -EOPNOTSUPP;
 	}
 
-	if (edata->advertised != tp->eee.advertised) {
+	if (!linkmode_equal(edata->advertised, tp->eee.advertised)) {
 		netdev_warn(tp->dev,
 			    "Direct manipulation of EEE advertisement is not supported\n");
 		return -EINVAL;
@@ -14083,7 +14211,9 @@ static int tg3_set_eee(struct net_device *dev, struct ethtool_eee *edata)
 		return -EINVAL;
 	}
 
-	tp->eee = *edata;
+	tp->eee.eee_enabled = edata->eee_enabled;
+	tp->eee.tx_lpi_enabled = edata->tx_lpi_enabled;
+	tp->eee.tx_lpi_timer = edata->tx_lpi_timer;
 
 	tp->phy_flags |= TG3_PHYFLG_USER_CONFIGURED;
 	tg3_warn_mgmt_link_flap(tp);
@@ -14098,7 +14228,7 @@ static int tg3_set_eee(struct net_device *dev, struct ethtool_eee *edata)
 	return 0;
 }
 
-static int tg3_get_eee(struct net_device *dev, struct ethtool_eee *edata)
+static int tg3_get_eee(struct net_device *dev, struct ethtool_keee *edata)
 {
 	struct tg3 *tp = netdev_priv(dev);
 
@@ -14113,6 +14243,11 @@ static int tg3_get_eee(struct net_device *dev, struct ethtool_eee *edata)
 }
 
 static const struct ethtool_ops tg3_ethtool_ops = {
+	.supported_coalesce_params = ETHTOOL_COALESCE_USECS |
+				     ETHTOOL_COALESCE_MAX_FRAMES |
+				     ETHTOOL_COALESCE_USECS_IRQ |
+				     ETHTOOL_COALESCE_MAX_FRAMES_IRQ |
+				     ETHTOOL_COALESCE_STATS_BLOCK_USECS,
 	.get_drvinfo		= tg3_get_drvinfo,
 	.get_regs_len		= tg3_get_regs_len,
 	.get_regs		= tg3_get_regs,
@@ -14155,7 +14290,7 @@ static void tg3_get_stats64(struct net_device *dev,
 	struct tg3 *tp = netdev_priv(dev);
 
 	spin_lock_bh(&tp->lock);
-	if (!tp->hw_stats) {
+	if (!tp->hw_stats || !tg3_flag(tp, INIT_COMPLETE)) {
 		*stats = tp->net_stats_prev;
 		spin_unlock_bh(&tp->lock);
 		return;
@@ -14180,7 +14315,7 @@ static void tg3_set_rx_mode(struct net_device *dev)
 static inline void tg3_set_mtu(struct net_device *dev, struct tg3 *tp,
 			       int new_mtu)
 {
-	dev->mtu = new_mtu;
+	WRITE_ONCE(dev->mtu, new_mtu);
 
 	if (new_mtu > ETH_DATA_LEN) {
 		if (tg3_flag(tp, 5780_CLASS)) {
@@ -14218,6 +14353,7 @@ static int tg3_change_mtu(struct net_device *dev, int new_mtu)
 
 	tg3_set_mtu(dev, tp, new_mtu);
 
+	netdev_lock(dev);
 	tg3_full_lock(tp, 1);
 
 	tg3_halt(tp, RESET_KIND_SHUTDOWN, 1);
@@ -14225,7 +14361,10 @@ static int tg3_change_mtu(struct net_device *dev, int new_mtu)
 	/* Reset PHY, otherwise the read DMA engine will be in a mode that
 	 * breaks all requests to 256 bytes.
 	 */
-	if (tg3_asic_rev(tp) == ASIC_REV_57766)
+	if (tg3_asic_rev(tp) == ASIC_REV_57766 ||
+	    tg3_asic_rev(tp) == ASIC_REV_5717 ||
+	    tg3_asic_rev(tp) == ASIC_REV_5719 ||
+	    tg3_asic_rev(tp) == ASIC_REV_5720)
 		reset_phy = true;
 
 	err = tg3_restart_hw(tp, reset_phy);
@@ -14234,6 +14373,7 @@ static int tg3_change_mtu(struct net_device *dev, int new_mtu)
 		tg3_netif_start(tp);
 
 	tg3_full_unlock(tp);
+	netdev_unlock(dev);
 
 	if (!err)
 		tg3_phy_start(tp);
@@ -14249,7 +14389,7 @@ static const struct net_device_ops tg3_netdev_ops = {
 	.ndo_validate_addr	= eth_validate_addr,
 	.ndo_set_rx_mode	= tg3_set_rx_mode,
 	.ndo_set_mac_address	= tg3_set_mac_addr,
-	.ndo_do_ioctl		= tg3_ioctl,
+	.ndo_eth_ioctl		= tg3_ioctl,
 	.ndo_tx_timeout		= tg3_tx_timeout,
 	.ndo_change_mtu		= tg3_change_mtu,
 	.ndo_fix_features	= tg3_fix_features,
@@ -14257,6 +14397,8 @@ static const struct net_device_ops tg3_netdev_ops = {
 #ifdef CONFIG_NET_POLL_CONTROLLER
 	.ndo_poll_controller	= tg3_poll_controller,
 #endif
+	.ndo_hwtstamp_get	= tg3_hwtstamp_get,
+	.ndo_hwtstamp_set	= tg3_hwtstamp_set,
 };
 
 static void tg3_get_eeprom_size(struct tg3 *tp)
@@ -14774,7 +14916,7 @@ static void tg3_get_5717_nvram_info(struct tg3 *tp)
 
 static void tg3_get_5720_nvram_info(struct tg3 *tp)
 {
-	u32 nvcfg1, nvmpinstrp;
+	u32 nvcfg1, nvmpinstrp, nv_status;
 
 	nvcfg1 = tr32(NVRAM_CFG1);
 	nvmpinstrp = nvcfg1 & NVRAM_CFG1_5752VENDOR_MASK;
@@ -14786,6 +14928,23 @@ static void tg3_get_5720_nvram_info(struct tg3 *tp)
 		}
 
 		switch (nvmpinstrp) {
+		case FLASH_5762_MX25L_100:
+		case FLASH_5762_MX25L_200:
+		case FLASH_5762_MX25L_400:
+		case FLASH_5762_MX25L_800:
+		case FLASH_5762_MX25L_160_320:
+			tp->nvram_pagesize = 4096;
+			tp->nvram_jedecnum = JEDEC_MACRONIX;
+			tg3_flag_set(tp, NVRAM_BUFFERED);
+			tg3_flag_set(tp, NO_NVRAM_ADDR_TRANS);
+			tg3_flag_set(tp, FLASH);
+			nv_status = tr32(NVRAM_AUTOSENSE_STATUS);
+			tp->nvram_size =
+				(1 << (nv_status >> AUTOSENSE_DEVID &
+						AUTOSENSE_DEVID_MASK)
+					<< AUTOSENSE_SIZE_IN_MB);
+			return;
+
 		case FLASH_5762_EEPROM_HD:
 			nvmpinstrp = FLASH_5720_EEPROM_HD;
 			break;
@@ -15511,10 +15670,13 @@ static int tg3_phy_probe(struct tg3 *tp)
 	      tg3_chip_rev_id(tp) != CHIPREV_ID_57765_A0))) {
 		tp->phy_flags |= TG3_PHYFLG_EEE_CAP;
 
-		tp->eee.supported = SUPPORTED_100baseT_Full |
-				    SUPPORTED_1000baseT_Full;
-		tp->eee.advertised = ADVERTISED_100baseT_Full |
-				     ADVERTISED_1000baseT_Full;
+		linkmode_zero(tp->eee.supported);
+		linkmode_set_bit(ETHTOOL_LINK_MODE_100baseT_Full_BIT,
+				 tp->eee.supported);
+		linkmode_set_bit(ETHTOOL_LINK_MODE_1000baseT_Full_BIT,
+				 tp->eee.supported);
+		linkmode_copy(tp->eee.advertised, tp->eee.supported);
+
 		tp->eee.eee_enabled = 1;
 		tp->eee.tx_lpi_enabled = 1;
 		tp->eee.tx_lpi_timer = TG3_CPMU_DBTMR1_LNKIDLE_2047US;
@@ -15563,64 +15725,36 @@ skip_phy_reset:
 static void tg3_read_vpd(struct tg3 *tp)
 {
 	u8 *vpd_data;
-	unsigned int block_end, rosize, len;
-	u32 vpdlen;
-	int j, i = 0;
+	unsigned int len, vpdlen;
+	int i;
 
 	vpd_data = (u8 *)tg3_vpd_readblock(tp, &vpdlen);
 	if (!vpd_data)
 		goto out_no_vpd;
 
-	i = pci_vpd_find_tag(vpd_data, 0, vpdlen, PCI_VPD_LRDT_RO_DATA);
+	i = pci_vpd_find_ro_info_keyword(vpd_data, vpdlen,
+					 PCI_VPD_RO_KEYWORD_MFR_ID, &len);
 	if (i < 0)
-		goto out_not_found;
-
-	rosize = pci_vpd_lrdt_size(&vpd_data[i]);
-	block_end = i + PCI_VPD_LRDT_TAG_SIZE + rosize;
-	i += PCI_VPD_LRDT_TAG_SIZE;
-
-	if (block_end > vpdlen)
-		goto out_not_found;
-
-	j = pci_vpd_find_info_keyword(vpd_data, i, rosize,
-				      PCI_VPD_RO_KEYWORD_MFR_ID);
-	if (j > 0) {
-		len = pci_vpd_info_field_size(&vpd_data[j]);
+		goto partno;
 
-		j += PCI_VPD_INFO_FLD_HDR_SIZE;
-		if (j + len > block_end || len != 4 ||
-		    memcmp(&vpd_data[j], "1028", 4))
-			goto partno;
+	if (len != 4 || memcmp(vpd_data + i, "1028", 4))
+		goto partno;
 
-		j = pci_vpd_find_info_keyword(vpd_data, i, rosize,
-					      PCI_VPD_RO_KEYWORD_VENDOR0);
-		if (j < 0)
-			goto partno;
-
-		len = pci_vpd_info_field_size(&vpd_data[j]);
+	i = pci_vpd_find_ro_info_keyword(vpd_data, vpdlen,
+					 PCI_VPD_RO_KEYWORD_VENDOR0, &len);
+	if (i < 0)
+		goto partno;
 
-		j += PCI_VPD_INFO_FLD_HDR_SIZE;
-		if (j + len > block_end)
-			goto partno;
-
-		if (len >= sizeof(tp->fw_ver))
-			len = sizeof(tp->fw_ver) - 1;
-		memset(tp->fw_ver, 0, sizeof(tp->fw_ver));
-		snprintf(tp->fw_ver, sizeof(tp->fw_ver), "%.*s bc ", len,
-			 &vpd_data[j]);
-	}
+	memset(tp->fw_ver, 0, sizeof(tp->fw_ver));
+	snprintf(tp->fw_ver, sizeof(tp->fw_ver), "%.*s bc ", len, vpd_data + i);
 
 partno:
-	i = pci_vpd_find_info_keyword(vpd_data, i, rosize,
-				      PCI_VPD_RO_KEYWORD_PARTNO);
+	i = pci_vpd_find_ro_info_keyword(vpd_data, vpdlen,
+					 PCI_VPD_RO_KEYWORD_PARTNO, &len);
 	if (i < 0)
 		goto out_not_found;
 
-	len = pci_vpd_info_field_size(&vpd_data[i]);
-
-	i += PCI_VPD_INFO_FLD_HDR_SIZE;
-	if (len > TG3_BPN_SIZE ||
-	    (len + i) > vpdlen)
+	if (len > TG3_BPN_SIZE)
 		goto out_not_found;
 
 	memcpy(tp->board_part_number, &vpd_data[i], len);
@@ -16468,7 +16602,7 @@ static int tg3_get_invariants(struct tg3 *tp, const struct pci_device_id *ent)
 
 			tg3_flag_set(tp, PCIX_TARGET_HWBUG);
 
-			/* The chip can have it's power management PCI config
+			/* The chip can have its power management PCI config
 			 * space registers clobbered due to this bug.
 			 * So explicitly force the chip into D0 here.
 			 */
@@ -16621,6 +16755,8 @@ static int tg3_get_invariants(struct tg3 *tp, const struct pci_device_id *ent)
 				       pci_state_reg);
 
 		tg3_ape_lock_init(tp);
+		tp->ape_hb_interval =
+			msecs_to_jiffies(APE_HOST_HEARTBEAT_INT_5SEC);
 	}
 
 	/* Set up tp->grc_local_ctrl before calling
@@ -16909,47 +17045,18 @@ static int tg3_get_invariants(struct tg3 *tp, const struct pci_device_id *ent)
 	return err;
 }
 
-#ifdef CONFIG_SPARC
-static int tg3_get_macaddr_sparc(struct tg3 *tp)
-{
-	struct net_device *dev = tp->dev;
-	struct pci_dev *pdev = tp->pdev;
-	struct device_node *dp = pci_device_to_OF_node(pdev);
-	const unsigned char *addr;
-	int len;
-
-	addr = of_get_property(dp, "local-mac-address", &len);
-	if (addr && len == ETH_ALEN) {
-		memcpy(dev->dev_addr, addr, ETH_ALEN);
-		return 0;
-	}
-	return -ENODEV;
-}
-
-static int tg3_get_default_macaddr_sparc(struct tg3 *tp)
+static int tg3_get_device_address(struct tg3 *tp, u8 *addr)
 {
-	struct net_device *dev = tp->dev;
-
-	memcpy(dev->dev_addr, idprom->id_ethaddr, ETH_ALEN);
-	return 0;
-}
-#endif
-
-static int tg3_get_device_address(struct tg3 *tp)
-{
-	struct net_device *dev = tp->dev;
 	u32 hi, lo, mac_offset;
 	int addr_ok = 0;
 	int err;
 
-#ifdef CONFIG_SPARC
-	if (!tg3_get_macaddr_sparc(tp))
+	if (!eth_platform_get_mac_address(&tp->pdev->dev, addr))
 		return 0;
-#endif
 
 	if (tg3_flag(tp, IS_SSB_CORE)) {
-		err = ssb_gige_get_macaddr(tp->pdev, &dev->dev_addr[0]);
-		if (!err && is_valid_ether_addr(&dev->dev_addr[0]))
+		err = ssb_gige_get_macaddr(tp->pdev, addr);
+		if (!err && is_valid_ether_addr(addr))
 			return 0;
 	}
 
@@ -16973,47 +17080,44 @@ static int tg3_get_device_address(struct tg3 *tp)
 	/* First try to get it from MAC address mailbox. */
 	tg3_read_mem(tp, NIC_SRAM_MAC_ADDR_HIGH_MBOX, &hi);
 	if ((hi >> 16) == 0x484b) {
-		dev->dev_addr[0] = (hi >>  8) & 0xff;
-		dev->dev_addr[1] = (hi >>  0) & 0xff;
+		addr[0] = (hi >>  8) & 0xff;
+		addr[1] = (hi >>  0) & 0xff;
 
 		tg3_read_mem(tp, NIC_SRAM_MAC_ADDR_LOW_MBOX, &lo);
-		dev->dev_addr[2] = (lo >> 24) & 0xff;
-		dev->dev_addr[3] = (lo >> 16) & 0xff;
-		dev->dev_addr[4] = (lo >>  8) & 0xff;
-		dev->dev_addr[5] = (lo >>  0) & 0xff;
+		addr[2] = (lo >> 24) & 0xff;
+		addr[3] = (lo >> 16) & 0xff;
+		addr[4] = (lo >>  8) & 0xff;
+		addr[5] = (lo >>  0) & 0xff;
 
 		/* Some old bootcode may report a 0 MAC address in SRAM */
-		addr_ok = is_valid_ether_addr(&dev->dev_addr[0]);
+		addr_ok = is_valid_ether_addr(addr);
 	}
 	if (!addr_ok) {
+		__be32 be_hi, be_lo;
+
 		/* Next, try NVRAM. */
 		if (!tg3_flag(tp, NO_NVRAM) &&
-		    !tg3_nvram_read_be32(tp, mac_offset + 0, &hi) &&
-		    !tg3_nvram_read_be32(tp, mac_offset + 4, &lo)) {
-			memcpy(&dev->dev_addr[0], ((char *)&hi) + 2, 2);
-			memcpy(&dev->dev_addr[2], (char *)&lo, sizeof(lo));
+		    !tg3_nvram_read_be32(tp, mac_offset + 0, &be_hi) &&
+		    !tg3_nvram_read_be32(tp, mac_offset + 4, &be_lo)) {
+			memcpy(&addr[0], ((char *)&be_hi) + 2, 2);
+			memcpy(&addr[2], (char *)&be_lo, sizeof(be_lo));
 		}
 		/* Finally just fetch it out of the MAC control regs. */
 		else {
 			hi = tr32(MAC_ADDR_0_HIGH);
 			lo = tr32(MAC_ADDR_0_LOW);
 
-			dev->dev_addr[5] = lo & 0xff;
-			dev->dev_addr[4] = (lo >> 8) & 0xff;
-			dev->dev_addr[3] = (lo >> 16) & 0xff;
-			dev->dev_addr[2] = (lo >> 24) & 0xff;
-			dev->dev_addr[1] = hi & 0xff;
-			dev->dev_addr[0] = (hi >> 8) & 0xff;
+			addr[5] = lo & 0xff;
+			addr[4] = (lo >> 8) & 0xff;
+			addr[3] = (lo >> 16) & 0xff;
+			addr[2] = (lo >> 24) & 0xff;
+			addr[1] = hi & 0xff;
+			addr[0] = (hi >> 8) & 0xff;
 		}
 	}
 
-	if (!is_valid_ether_addr(&dev->dev_addr[0])) {
-#ifdef CONFIG_SPARC
-		if (!tg3_get_default_macaddr_sparc(tp))
-			return 0;
-#endif
+	if (!is_valid_ether_addr(addr))
 		return -EINVAL;
-	}
 	return 0;
 }
 
@@ -17040,7 +17144,7 @@ static u32 tg3_calc_dma_bndry(struct tg3 *tp, u32 val)
 	    !tg3_flag(tp, PCI_EXPRESS))
 		goto out;
 
-#if defined(CONFIG_PPC64) || defined(CONFIG_IA64) || defined(CONFIG_PARISC)
+#if defined(CONFIG_PPC64) || defined(CONFIG_PARISC)
 	goal = BOUNDARY_MULTI_CACHELINE;
 #else
 #if defined(CONFIG_SPARC64) || defined(CONFIG_ALPHA)
@@ -17104,7 +17208,7 @@ static u32 tg3_calc_dma_bndry(struct tg3 *tp, u32 val)
 				val |= DMA_RWCTRL_WRITE_BNDRY_64_PCIE;
 				break;
 			}
-			/* fallthrough */
+			fallthrough;
 		case 128:
 		default:
 			val &= ~DMA_RWCTRL_WRITE_BNDRY_DISAB_PCIE;
@@ -17119,28 +17223,28 @@ static u32 tg3_calc_dma_bndry(struct tg3 *tp, u32 val)
 					DMA_RWCTRL_WRITE_BNDRY_16);
 				break;
 			}
-			/* fallthrough */
+			fallthrough;
 		case 32:
 			if (goal == BOUNDARY_SINGLE_CACHELINE) {
 				val |= (DMA_RWCTRL_READ_BNDRY_32 |
 					DMA_RWCTRL_WRITE_BNDRY_32);
 				break;
 			}
-			/* fallthrough */
+			fallthrough;
 		case 64:
 			if (goal == BOUNDARY_SINGLE_CACHELINE) {
 				val |= (DMA_RWCTRL_READ_BNDRY_64 |
 					DMA_RWCTRL_WRITE_BNDRY_64);
 				break;
 			}
-			/* fallthrough */
+			fallthrough;
 		case 128:
 			if (goal == BOUNDARY_SINGLE_CACHELINE) {
 				val |= (DMA_RWCTRL_READ_BNDRY_128 |
 					DMA_RWCTRL_WRITE_BNDRY_128);
 				break;
 			}
-			/* fallthrough */
+			fallthrough;
 		case 256:
 			val |= (DMA_RWCTRL_READ_BNDRY_256 |
 				DMA_RWCTRL_WRITE_BNDRY_256);
@@ -17588,8 +17692,7 @@ static int tg3_init_one(struct pci_dev *pdev,
 	char str[40];
 	u64 dma_mask, persist_dma_mask;
 	netdev_features_t features = 0;
-
-	printk_once(KERN_INFO "%s\n", version);
+	u8 addr[ETH_ALEN] __aligned(2);
 
 	err = pci_enable_device(pdev);
 	if (err) {
@@ -17716,7 +17819,7 @@ static int tg3_init_one(struct pci_dev *pdev,
 	 * device behind the EPB cannot support DMA addresses > 40-bit.
 	 * On 64-bit systems with IOMMU, use 40-bit dma_mask.
 	 * On 64-bit systems without IOMMU, use 64-bit dma_mask and
-	 * do DMA address check in tg3_start_xmit().
+	 * do DMA address check in __tg3_start_xmit().
 	 */
 	if (tg3_flag(tp, IS_5788))
 		persist_dma_mask = dma_mask = DMA_BIT_MASK(32);
@@ -17728,13 +17831,16 @@ static int tg3_init_one(struct pci_dev *pdev,
 	} else
 		persist_dma_mask = dma_mask = DMA_BIT_MASK(64);
 
+	if (tg3_asic_rev(tp) == ASIC_REV_57766)
+		persist_dma_mask = DMA_BIT_MASK(31);
+
 	/* Configure DMA attributes. */
 	if (dma_mask > DMA_BIT_MASK(32)) {
-		err = pci_set_dma_mask(pdev, dma_mask);
+		err = dma_set_mask(&pdev->dev, dma_mask);
 		if (!err) {
 			features |= NETIF_F_HIGHDMA;
-			err = pci_set_consistent_dma_mask(pdev,
-							  persist_dma_mask);
+			err = dma_set_coherent_mask(&pdev->dev,
+						    persist_dma_mask);
 			if (err < 0) {
 				dev_err(&pdev->dev, "Unable to obtain 64 bit "
 					"DMA for consistent allocations\n");
@@ -17743,7 +17849,7 @@ static int tg3_init_one(struct pci_dev *pdev,
 		}
 	}
 	if (err || dma_mask == DMA_BIT_MASK(32)) {
-		err = pci_set_dma_mask(pdev, DMA_BIT_MASK(32));
+		err = dma_set_mask(&pdev->dev, DMA_BIT_MASK(32));
 		if (err) {
 			dev_err(&pdev->dev,
 				"No usable DMA configuration, aborting\n");
@@ -17812,12 +17918,13 @@ static int tg3_init_one(struct pci_dev *pdev,
 		tp->rx_pending = 63;
 	}
 
-	err = tg3_get_device_address(tp);
+	err = tg3_get_device_address(tp, addr);
 	if (err) {
 		dev_err(&pdev->dev,
 			"Could not obtain valid ethernet address, aborting\n");
 		goto err_out_apeunmap;
 	}
+	eth_hw_addr_set(dev, addr);
 
 	intmbx = MAILBOX_INTERRUPT_0 + TG3_64BIT_REG_LOW;
 	rcvmbx = MAILBOX_RCVRET_CON_IDX_0 + TG3_64BIT_REG_LOW;
@@ -17829,10 +17936,7 @@ static int tg3_init_one(struct pci_dev *pdev,
 		tnapi->tx_pending = TG3_DEF_TX_RING_PENDING;
 
 		tnapi->int_mbox = intmbx;
-		if (i <= 4)
-			intmbx += 0x8;
-		else
-			intmbx += 0x4;
+		intmbx += 0x8;
 
 		tnapi->consmbox = rcvmbx;
 		tnapi->prodmbox = sndmbx;
@@ -18007,10 +18111,8 @@ static void tg3_remove_one(struct pci_dev *pdev)
 #ifdef CONFIG_PM_SLEEP
 static int tg3_suspend(struct device *device)
 {
-	struct pci_dev *pdev = to_pci_dev(device);
-	struct net_device *dev = pci_get_drvdata(pdev);
+	struct net_device *dev = dev_get_drvdata(device);
 	struct tg3 *tp = netdev_priv(dev);
-	int err = 0;
 
 	rtnl_lock();
 
@@ -18034,38 +18136,16 @@ static int tg3_suspend(struct device *device)
 	tg3_flag_clear(tp, INIT_COMPLETE);
 	tg3_full_unlock(tp);
 
-	err = tg3_power_down_prepare(tp);
-	if (err) {
-		int err2;
-
-		tg3_full_lock(tp, 0);
-
-		tg3_flag_set(tp, INIT_COMPLETE);
-		err2 = tg3_restart_hw(tp, true);
-		if (err2)
-			goto out;
-
-		tg3_timer_start(tp);
-
-		netif_device_attach(dev);
-		tg3_netif_start(tp);
-
-out:
-		tg3_full_unlock(tp);
-
-		if (!err2)
-			tg3_phy_start(tp);
-	}
+	tg3_power_down_prepare(tp);
 
 unlock:
 	rtnl_unlock();
-	return err;
+	return 0;
 }
 
 static int tg3_resume(struct device *device)
 {
-	struct pci_dev *pdev = to_pci_dev(device);
-	struct net_device *dev = pci_get_drvdata(pdev);
+	struct net_device *dev = dev_get_drvdata(device);
 	struct tg3 *tp = netdev_priv(dev);
 	int err = 0;
 
@@ -18076,6 +18156,7 @@ static int tg3_resume(struct device *device)
 
 	netif_device_attach(dev);
 
+	netdev_lock(dev);
 	tg3_full_lock(tp, 0);
 
 	tg3_ape_driver_state_change(tp, RESET_KIND_INIT);
@@ -18092,6 +18173,7 @@ static int tg3_resume(struct device *device)
 
 out:
 	tg3_full_unlock(tp);
+	netdev_unlock(dev);
 
 	if (!err)
 		tg3_phy_start(tp);
@@ -18104,12 +18186,59 @@ unlock:
 
 static SIMPLE_DEV_PM_OPS(tg3_pm_ops, tg3_suspend, tg3_resume);
 
+/* Systems where ACPI _PTS (Prepare To Sleep) S5 will result in a fatal
+ * PCIe AER event on the tg3 device if the tg3 device is not, or cannot
+ * be, powered down.
+ */
+static const struct dmi_system_id tg3_restart_aer_quirk_table[] = {
+	{
+		.matches = {
+			DMI_MATCH(DMI_SYS_VENDOR, "Dell Inc."),
+			DMI_MATCH(DMI_PRODUCT_NAME, "PowerEdge R440"),
+		},
+	},
+	{
+		.matches = {
+			DMI_MATCH(DMI_SYS_VENDOR, "Dell Inc."),
+			DMI_MATCH(DMI_PRODUCT_NAME, "PowerEdge R540"),
+		},
+	},
+	{
+		.matches = {
+			DMI_MATCH(DMI_SYS_VENDOR, "Dell Inc."),
+			DMI_MATCH(DMI_PRODUCT_NAME, "PowerEdge R640"),
+		},
+	},
+	{
+		.matches = {
+			DMI_MATCH(DMI_SYS_VENDOR, "Dell Inc."),
+			DMI_MATCH(DMI_PRODUCT_NAME, "PowerEdge R650"),
+		},
+	},
+	{
+		.matches = {
+			DMI_MATCH(DMI_SYS_VENDOR, "Dell Inc."),
+			DMI_MATCH(DMI_PRODUCT_NAME, "PowerEdge R740"),
+		},
+	},
+	{
+		.matches = {
+			DMI_MATCH(DMI_SYS_VENDOR, "Dell Inc."),
+			DMI_MATCH(DMI_PRODUCT_NAME, "PowerEdge R750"),
+		},
+	},
+	{}
+};
+
 static void tg3_shutdown(struct pci_dev *pdev)
 {
 	struct net_device *dev = pci_get_drvdata(pdev);
 	struct tg3 *tp = netdev_priv(dev);
 
+	tg3_reset_task_cancel(tp);
+
 	rtnl_lock();
+
 	netif_device_detach(dev);
 
 	if (netif_running(dev))
@@ -18117,8 +18246,23 @@ static void tg3_shutdown(struct pci_dev *pdev)
 
 	if (system_state == SYSTEM_POWER_OFF)
 		tg3_power_down(tp);
+	else if (system_state == SYSTEM_RESTART &&
+		 dmi_first_match(tg3_restart_aer_quirk_table) &&
+		 pdev->current_state != PCI_D3cold &&
+		 pdev->current_state != PCI_UNKNOWN) {
+		/* Disable PCIe AER on the tg3 to avoid a fatal
+		 * error during this system restart.
+		 */
+		pcie_capability_clear_word(pdev, PCI_EXP_DEVCTL,
+					   PCI_EXP_DEVCTL_CERE |
+					   PCI_EXP_DEVCTL_NFERE |
+					   PCI_EXP_DEVCTL_FERE |
+					   PCI_EXP_DEVCTL_URRE);
+	}
 
 	rtnl_unlock();
+
+	pci_disable_device(pdev);
 }
 
 /**
@@ -18138,10 +18282,13 @@ static pci_ers_result_t tg3_io_error_detected(struct pci_dev *pdev,
 
 	netdev_info(netdev, "PCI I/O error detected\n");
 
+	/* Want to make sure that the reset task doesn't run */
+	tg3_reset_task_cancel(tp);
+
 	rtnl_lock();
 
-	/* We probably don't have netdev yet */
-	if (!netdev || !netif_running(netdev))
+	/* Could be second call or maybe we don't have netdev yet */
+	if (!netdev || tp->pcierr_recovery || !netif_running(netdev))
 		goto done;
 
 	/* We needn't recover from permanent error */
@@ -18154,9 +18301,6 @@ static pci_ers_result_t tg3_io_error_detected(struct pci_dev *pdev,
 
 	tg3_timer_stop(tp);
 
-	/* Want to make sure that the reset task doesn't run */
-	tg3_reset_task_cancel(tp);
-
 	netif_device_detach(netdev);
 
 	/* Clean up software state, even if MMIO is blocked */
@@ -18167,7 +18311,9 @@ static pci_ers_result_t tg3_io_error_detected(struct pci_dev *pdev,
 done:
 	if (state == pci_channel_io_perm_failure) {
 		if (netdev) {
+			netdev_lock(netdev);
 			tg3_napi_enable(tp);
+			netdev_unlock(netdev);
 			dev_close(netdev);
 		}
 		err = PCI_ERS_RESULT_DISCONNECT;
@@ -18185,7 +18331,7 @@ done:
  * @pdev: Pointer to PCI device
  *
  * Restart the card from scratch, as if from a cold-boot.
- * At this point, the card has exprienced a hard reset,
+ * At this point, the card has experienced a hard reset,
  * followed by fixups by BIOS, and has its config space
  * set up identically to what it was at cold boot.
  */
@@ -18221,7 +18367,9 @@ static pci_ers_result_t tg3_io_slot_reset(struct pci_dev *pdev)
 
 done:
 	if (rc != PCI_ERS_RESULT_RECOVERED && netdev && netif_running(netdev)) {
+		netdev_lock(netdev);
 		tg3_napi_enable(tp);
+		netdev_unlock(netdev);
 		dev_close(netdev);
 	}
 	rtnl_unlock();
@@ -18247,12 +18395,14 @@ static void tg3_io_resume(struct pci_dev *pdev)
 	if (!netdev || !netif_running(netdev))
 		goto done;
 
+	netdev_lock(netdev);
 	tg3_full_lock(tp, 0);
 	tg3_ape_driver_state_change(tp, RESET_KIND_INIT);
 	tg3_flag_set(tp, INIT_COMPLETE);
 	err = tg3_restart_hw(tp, true);
 	if (err) {
 		tg3_full_unlock(tp);
+		netdev_unlock(netdev);
 		netdev_err(netdev, "Cannot restart hardware after reset.\n");
 		goto done;
 	}
@@ -18264,6 +18414,7 @@ static void tg3_io_resume(struct pci_dev *pdev)
 	tg3_netif_start(tp);
 
 	tg3_full_unlock(tp);
+	netdev_unlock(netdev);
 
 	tg3_phy_start(tp);
 
diff --git a/drivers/net/ethernet/broadcom/tg3.h b/drivers/net/ethernet/broadcom/tg3.h
index 3b5e98ecba00..a9e7f88fa26d 100644
--- a/drivers/net/ethernet/broadcom/tg3.h
+++ b/drivers/net/ethernet/broadcom/tg3.h
@@ -1,10 +1,14 @@
+/* SPDX-License-Identifier: GPL-2.0 */
 /* $Id: tg3.h,v 1.37.2.32 2002/03/11 12:18:18 davem Exp $
  * tg3.h: Definitions for Broadcom Tigon3 ethernet driver.
  *
  * Copyright (C) 2001, 2002, 2003, 2004 David S. Miller (davem@redhat.com)
  * Copyright (C) 2001 Jeff Garzik (jgarzik@pobox.com)
  * Copyright (C) 2004 Sun Microsystems Inc.
- * Copyright (C) 2007-2014 Broadcom Corporation.
+ * Copyright (C) 2007-2016 Broadcom Corporation.
+ * Copyright (C) 2016-2017 Broadcom Limited.
+ * Copyright (C) 2018 Broadcom. All Rights Reserved. The term "Broadcom"
+ * refers to Broadcom Inc. and/or its subsidiaries.
  */
 
 #ifndef _T3_H
@@ -95,6 +99,7 @@
 #define TG3PCI_SUBDEVICE_ID_DELL_JAGUAR		0x0106
 #define TG3PCI_SUBDEVICE_ID_DELL_MERLOT		0x0109
 #define TG3PCI_SUBDEVICE_ID_DELL_SLIM_MERLOT	0x010a
+#define TG3PCI_SUBDEVICE_ID_DELL_5762		0x07f0
 #define TG3PCI_SUBVENDOR_ID_COMPAQ		PCI_VENDOR_ID_COMPAQ
 #define TG3PCI_SUBDEVICE_ID_COMPAQ_BANSHEE	0x007c
 #define TG3PCI_SUBDEVICE_ID_COMPAQ_BANSHEE_2	0x009a
@@ -280,6 +285,9 @@
 #define TG3PCI_STD_RING_PROD_IDX	0x00000098 /* 64-bit */
 #define TG3PCI_RCV_RET_RING_CON_IDX	0x000000a0 /* 64-bit */
 /* 0xa8 --> 0xb8 unused */
+#define TG3PCI_DEV_STATUS_CTRL		0x000000b4
+#define  MAX_READ_REQ_SIZE_2048		 0x00004000
+#define  MAX_READ_REQ_MASK		 0x00007000
 #define TG3PCI_DUAL_MAC_CTRL		0x000000b8
 #define  DUAL_MAC_CTRL_CH_MASK		 0x00000003
 #define  DUAL_MAC_CTRL_ID		 0x00000004
@@ -1857,7 +1865,7 @@
 #define NVRAM_STAT			0x00007004
 #define NVRAM_WRDATA			0x00007008
 #define NVRAM_ADDR			0x0000700c
-#define  NVRAM_ADDR_MSK			0x00ffffff
+#define  NVRAM_ADDR_MSK			0x07ffffff
 #define NVRAM_RDDATA			0x00007010
 #define NVRAM_CFG1			0x00007014
 #define  NVRAM_CFG1_FLASHIF_ENAB	 0x00000001
@@ -1939,6 +1947,11 @@
 #define  FLASH_5720_EEPROM_LD		 0x00000003
 #define  FLASH_5762_EEPROM_HD		 0x02000001
 #define  FLASH_5762_EEPROM_LD		 0x02000003
+#define  FLASH_5762_MX25L_100           0x00800000
+#define  FLASH_5762_MX25L_200           0x00800002
+#define  FLASH_5762_MX25L_400           0x00800001
+#define  FLASH_5762_MX25L_800           0x00800003
+#define  FLASH_5762_MX25L_160_320       0x03800002
 #define  FLASH_5720VENDOR_M_ATMEL_DB011D 0x01000000
 #define  FLASH_5720VENDOR_M_ATMEL_DB021D 0x01000002
 #define  FLASH_5720VENDOR_M_ATMEL_DB041D 0x01000001
@@ -2003,7 +2016,11 @@
 /* 0x702c unused */
 
 #define NVRAM_ADDR_LOCKOUT		0x00007030
-/* 0x7034 --> 0x7500 unused */
+#define NVRAM_AUTOSENSE_STATUS         0x00007038
+#define AUTOSENSE_DEVID                        0x00000010
+#define AUTOSENSE_DEVID_MASK           0x00000007
+#define AUTOSENSE_SIZE_IN_MB           17
+/* 0x703c --> 0x7500 unused */
 
 #define OTP_MODE			0x00007500
 #define OTP_MODE_OTP_THRU_GRC		 0x00000001
@@ -2373,7 +2390,7 @@
 #define TG3_CL45_D7_EEERES_STAT_LP_1000T	0x0004
 
 
-/* Fast Ethernet Tranceiver definitions */
+/* Fast Ethernet Transceiver definitions */
 #define MII_TG3_FET_PTEST		0x17
 #define  MII_TG3_FET_PTEST_TRIM_SEL	0x0010
 #define  MII_TG3_FET_PTEST_TRIM_2	0x0002
@@ -2493,6 +2510,7 @@
 #define TG3_APE_LOCK_PHY3		5
 #define TG3_APE_LOCK_GPIO		7
 
+#define TG3_APE_HB_INTERVAL             (tp->ape_hb_interval)
 #define TG3_EEPROM_SB_F1R2_MBA_OFF	0x10
 
 
@@ -2829,7 +2847,7 @@ struct tg3_ocir {
 	u32				port1_flags;
 	u32				port2_flags;
 	u32				port3_flags;
-	u32				reserved2[1];
+	u32				reserved2;
 };
 
 
@@ -2855,7 +2873,7 @@ struct tg3_tx_ring_info {
 struct tg3_link_config {
 	/* Describes what we're trying to get. */
 	u32				advertising;
-	u16				speed;
+	u32				speed;
 	u8				duplex;
 	u8				autoneg;
 	u8				flowctrl;
@@ -2864,7 +2882,7 @@ struct tg3_link_config {
 	u8				active_flowctrl;
 
 	u8				active_duplex;
-	u16				active_speed;
+	u32				active_speed;
 	u32				rmt_adv;
 };
 
@@ -3000,6 +3018,7 @@ struct tg3_napi {
 	u16				*rx_rcb_prod_idx;
 	struct tg3_rx_prodring_set	prodring;
 	struct tg3_rx_buffer_desc	*rx_rcb;
+	unsigned long			rx_dropped;
 
 	u32				tx_prod	____cacheline_aligned;
 	u32				tx_cons;
@@ -3008,12 +3027,13 @@ struct tg3_napi {
 	u32				prodmbox;
 	struct tg3_tx_buffer_desc	*tx_ring;
 	struct tg3_tx_ring_info		*tx_buffers;
+	unsigned long			tx_dropped;
 
 	dma_addr_t			status_mapping;
 	dma_addr_t			rx_rcb_mapping;
 	dma_addr_t			tx_desc_mapping;
 
-	char				irq_lbl[IFNAMSIZ];
+	char				irq_lbl[IFNAMSIZ + 6 + 10]; /* name + "-txrx-" + %d */
 	unsigned int			irq_vec;
 };
 
@@ -3172,6 +3192,7 @@ struct tg3 {
 	struct ptp_clock_info		ptp_info;
 	struct ptp_clock		*ptp_clock;
 	s64				ptp_adjust;
+	u8				ptp_txts_retrycnt;
 
 	/* begin "tx thread" cacheline section */
 	void				(*write32_tx_mbox) (struct tg3 *, u32,
@@ -3201,8 +3222,6 @@ struct tg3 {
 
 
 	/* begin "everything else" cacheline(s) section */
-	unsigned long			rx_dropped;
-	unsigned long			tx_dropped;
 	struct rtnl_link_stats64	net_stats_prev;
 	struct tg3_ethtool_stats	estats_prev;
 
@@ -3354,6 +3373,8 @@ struct tg3 {
 	struct tg3_hw_stats		*hw_stats;
 	dma_addr_t			stats_mapping;
 	struct work_struct		reset_task;
+	struct sk_buff			*tx_tstamp_skb;
+	u64				pre_tx_ts;
 
 	int				nvram_lock_cnt;
 	u32				nvram_size;
@@ -3372,6 +3393,7 @@ struct tg3 {
 #define JEDEC_ST			0x20
 #define JEDEC_SAIFUN			0x4f
 #define JEDEC_SST			0xbf
+#define JEDEC_MACRONIX                 0xc2
 
 #define ATMEL_AT24C02_CHIP_SIZE		TG3_NVRAM_SIZE_2KB
 #define ATMEL_AT24C02_PAGE_SIZE		(8)
@@ -3397,7 +3419,7 @@ struct tg3 {
 	unsigned int			irq_cnt;
 
 	struct ethtool_coalesce		coal;
-	struct ethtool_eee		eee;
+	struct ethtool_keee		eee;
 
 	/* firmware info */
 	const char			*fw_needed;
@@ -3407,6 +3429,10 @@ struct tg3 {
 	struct device			*hwmon_dev;
 	bool				link_up;
 	bool				pcierr_recovery;
+
+	u32                             ape_hb;
+	unsigned long                   ape_hb_interval;
+	unsigned long                   ape_hb_jiffies;
 };
 
 /* Accessor macros for chip and asic attributes
diff --git a/drivers/net/ethernet/broadcom/unimac.h b/drivers/net/ethernet/broadcom/unimac.h
new file mode 100644
index 000000000000..585a85286257
--- /dev/null
+++ b/drivers/net/ethernet/broadcom/unimac.h
@@ -0,0 +1,68 @@
+/* SPDX-License-Identifier: GPL-2.0-only */
+#ifndef __UNIMAC_H
+#define __UNIMAC_H
+
+#define UMAC_HD_BKP_CTRL		0x004
+#define  HD_FC_EN			(1 << 0)
+#define  HD_FC_BKOFF_OK			(1 << 1)
+#define  IPG_CONFIG_RX_SHIFT		2
+#define  IPG_CONFIG_RX_MASK		0x1F
+#define UMAC_CMD			0x008
+#define  CMD_TX_EN			(1 << 0)
+#define  CMD_RX_EN			(1 << 1)
+#define  CMD_SPEED_10			0
+#define  CMD_SPEED_100			1
+#define  CMD_SPEED_1000			2
+#define  CMD_SPEED_2500			3
+#define  CMD_SPEED_SHIFT		2
+#define  CMD_SPEED_MASK			3
+#define  CMD_PROMISC			(1 << 4)
+#define  CMD_PAD_EN			(1 << 5)
+#define  CMD_CRC_FWD			(1 << 6)
+#define  CMD_PAUSE_FWD			(1 << 7)
+#define  CMD_RX_PAUSE_IGNORE		(1 << 8)
+#define  CMD_TX_ADDR_INS		(1 << 9)
+#define  CMD_HD_EN			(1 << 10)
+#define  CMD_SW_RESET_OLD		(1 << 11)
+#define  CMD_SW_RESET			(1 << 13)
+#define  CMD_LCL_LOOP_EN		(1 << 15)
+#define  CMD_AUTO_CONFIG		(1 << 22)
+#define  CMD_CNTL_FRM_EN		(1 << 23)
+#define  CMD_NO_LEN_CHK			(1 << 24)
+#define  CMD_RMT_LOOP_EN		(1 << 25)
+#define  CMD_RX_ERR_DISC		(1 << 26)
+#define  CMD_PRBL_EN			(1 << 27)
+#define  CMD_TX_PAUSE_IGNORE		(1 << 28)
+#define  CMD_TX_RX_EN			(1 << 29)
+#define  CMD_RUNT_FILTER_DIS		(1 << 30)
+#define UMAC_MAC0			0x00c
+#define UMAC_MAC1			0x010
+#define UMAC_MAX_FRAME_LEN		0x014
+#define UMAC_PAUSE_QUANTA		0x018
+#define UMAC_MODE			0x044
+#define  MODE_LINK_STATUS		(1 << 5)
+#define UMAC_FRM_TAG0			0x048		/* outer tag */
+#define UMAC_FRM_TAG1			0x04c		/* inner tag */
+#define UMAC_TX_IPG_LEN			0x05c
+#define UMAC_EEE_CTRL			0x064
+#define  EN_LPI_RX_PAUSE		(1 << 0)
+#define  EN_LPI_TX_PFC			(1 << 1)
+#define  EN_LPI_TX_PAUSE		(1 << 2)
+#define  EEE_EN				(1 << 3)
+#define  RX_FIFO_CHECK			(1 << 4)
+#define  EEE_TX_CLK_DIS			(1 << 5)
+#define  DIS_EEE_10M			(1 << 6)
+#define  LP_IDLE_PREDICTION_MODE	(1 << 7)
+#define UMAC_EEE_LPI_TIMER		0x068
+#define UMAC_EEE_WAKE_TIMER		0x06C
+#define UMAC_EEE_REF_COUNT		0x070
+#define  EEE_REFERENCE_COUNT_MASK	0xffff
+#define UMAC_RX_IPG_INV			0x078
+#define UMAC_MACSEC_PROG_TX_CRC		0x310
+#define UMAC_MACSEC_CTRL		0x314
+#define UMAC_PAUSE_CTRL			0x330
+#define UMAC_TX_FLUSH			0x334
+#define UMAC_RX_FIFO_STATUS		0x338
+#define UMAC_TX_FIFO_STATUS		0x33c
+
+#endif
diff --git a/drivers/net/ethernet/brocade/Kconfig b/drivers/net/ethernet/brocade/Kconfig
index c4bbe54e2cad..fb4c3cdf7233 100644
--- a/drivers/net/ethernet/brocade/Kconfig
+++ b/drivers/net/ethernet/brocade/Kconfig
@@ -1,3 +1,4 @@
+# SPDX-License-Identifier: GPL-2.0-only
 #
 # QLogic BR-series device configuration
 #
@@ -6,7 +7,7 @@ config NET_VENDOR_BROCADE
 	bool "QLogic BR-series devices"
 	default y
 	depends on PCI
-	---help---
+	help
 	  If you have a network (Ethernet) card belonging to this class, say Y.
 
 	  Note that the answer to this question doesn't directly affect the
diff --git a/drivers/net/ethernet/brocade/Makefile b/drivers/net/ethernet/brocade/Makefile
index fec10f9b4558..88b2f402675f 100644
--- a/drivers/net/ethernet/brocade/Makefile
+++ b/drivers/net/ethernet/brocade/Makefile
@@ -1,3 +1,4 @@
+# SPDX-License-Identifier: GPL-2.0-only
 #
 # Makefile for the QLogic BR-series device drivers.
 #
diff --git a/drivers/net/ethernet/brocade/bna/Kconfig b/drivers/net/ethernet/brocade/bna/Kconfig
index fe01279a8843..0a48ad93ed6c 100644
--- a/drivers/net/ethernet/brocade/bna/Kconfig
+++ b/drivers/net/ethernet/brocade/bna/Kconfig
@@ -1,3 +1,4 @@
+# SPDX-License-Identifier: GPL-2.0-only
 #
 # QLogic BR-series network device configuration
 #
@@ -5,7 +6,7 @@
 config BNA
 	tristate "QLogic BR-series 1010/1020/1860 10Gb Ethernet Driver support"
 	depends on PCI
-	---help---
+	help
 	  This driver supports QLogic BR-series 1010/1020/1860 10Gb CEE capable
 	  Ethernet cards.
 	  To compile this driver as a module, choose M here: the module
diff --git a/drivers/net/ethernet/brocade/bna/Makefile b/drivers/net/ethernet/brocade/bna/Makefile
index 8584abcf5366..d804b30c33eb 100644
--- a/drivers/net/ethernet/brocade/bna/Makefile
+++ b/drivers/net/ethernet/brocade/bna/Makefile
@@ -1,3 +1,4 @@
+# SPDX-License-Identifier: GPL-2.0-only
 #
 # Copyright (c) 2005-2014 Brocade Communications Systems, Inc.
 # Copyright (c) 2014-2015 QLogic Corporation.
diff --git a/drivers/net/ethernet/brocade/bna/bfa_cee.c b/drivers/net/ethernet/brocade/bna/bfa_cee.c
index 95bc8b644a5d..eeb05e31713f 100644
--- a/drivers/net/ethernet/brocade/bna/bfa_cee.c
+++ b/drivers/net/ethernet/brocade/bna/bfa_cee.c
@@ -1,14 +1,6 @@
+// SPDX-License-Identifier: GPL-2.0-only
 /*
  * Linux network driver for QLogic BR-series Converged Network Adapter.
- *
- * This program is free software; you can redistribute it and/or modify it
- * under the terms of the GNU General Public License (GPL) Version 2 as
- * published by the Free Software Foundation
- *
- * This program is distributed in the hope that it will be useful, but
- * WITHOUT ANY WARRANTY; without even the implied warranty of
- * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU
- * General Public License for more details.
  */
 /*
  * Copyright (c) 2005-2014 Brocade Communications Systems, Inc.
@@ -90,7 +82,7 @@ bfa_cee_get_attr_isr(struct bfa_cee *cee, enum bfa_status status)
 }
 
 /**
- * bfa_cee_get_attr_isr - CEE ISR for get-stats responses from f/w
+ * bfa_cee_get_stats_isr - CEE ISR for get-stats responses from f/w
  *
  * @cee: Pointer to the CEE module
  * @status: Return status from the f/w
@@ -110,14 +102,10 @@ bfa_cee_get_stats_isr(struct bfa_cee *cee, enum bfa_status status)
 }
 
 /**
- * bfa_cee_get_attr_isr()
- *
- * @brief CEE ISR for reset-stats responses from f/w
+ * bfa_cee_reset_stats_isr - CEE ISR for reset-stats responses from f/w
  *
- * @param[in] cee - Pointer to the CEE module
- *            status - Return status from the f/w
- *
- * @return void
+ * @cee: Input Pointer to the CEE module
+ * @status: Return status from the f/w
  */
 static void
 bfa_cee_reset_stats_isr(struct bfa_cee *cee, enum bfa_status status)
@@ -156,9 +144,12 @@ bfa_nw_cee_mem_claim(struct bfa_cee *cee, u8 *dma_kva, u64 dma_pa)
 }
 
 /**
- * bfa_cee_get_attr - Send the request to the f/w to fetch CEE attributes.
+ * bfa_nw_cee_get_attr - Send the request to the f/w to fetch CEE attributes.
  *
  * @cee: Pointer to the CEE module data structure.
+ * @attr: attribute requested
+ * @cbfn: function pointer
+ * @cbarg: function pointer arguments
  *
  * Return: status
  */
@@ -189,7 +180,9 @@ bfa_nw_cee_get_attr(struct bfa_cee *cee, struct bfa_cee_attr *attr,
 }
 
 /**
- * bfa_cee_isrs - Handles Mail-box interrupts for CEE module.
+ * bfa_cee_isr - Handles Mail-box interrupts for CEE module.
+ * @cbarg: argument passed containing pointer to the CEE module data structure.
+ * @m: message pointer
  */
 
 static void
@@ -218,6 +211,7 @@ bfa_cee_isr(void *cbarg, struct bfi_mbmsg *m)
 /**
  * bfa_cee_notify - CEE module heart-beat failure handler.
  *
+ * @arg: argument passed containing pointer to the CEE module data structure.
  * @event: IOC event type
  */
 
diff --git a/drivers/net/ethernet/brocade/bna/bfa_cee.h b/drivers/net/ethernet/brocade/bna/bfa_cee.h
index d04eef5d5a77..8e628bb54ba0 100644
--- a/drivers/net/ethernet/brocade/bna/bfa_cee.h
+++ b/drivers/net/ethernet/brocade/bna/bfa_cee.h
@@ -1,14 +1,6 @@
+/* SPDX-License-Identifier: GPL-2.0-only */
 /*
  * Linux network driver for QLogic BR-series Converged Network Adapter.
- *
- * This program is free software; you can redistribute it and/or modify it
- * under the terms of the GNU General Public License (GPL) Version 2 as
- * published by the Free Software Foundation
- *
- * This program is distributed in the hope that it will be useful, but
- * WITHOUT ANY WARRANTY; without even the implied warranty of
- * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU
- * General Public License for more details.
  */
 /*
  * Copyright (c) 2005-2014 Brocade Communications Systems, Inc.
diff --git a/drivers/net/ethernet/brocade/bna/bfa_cs.h b/drivers/net/ethernet/brocade/bna/bfa_cs.h
index 1d11d666d408..858c92129451 100644
--- a/drivers/net/ethernet/brocade/bna/bfa_cs.h
+++ b/drivers/net/ethernet/brocade/bna/bfa_cs.h
@@ -1,14 +1,6 @@
+/* SPDX-License-Identifier: GPL-2.0-only */
 /*
  * Linux network driver for QLogic BR-series Converged Network Adapter.
- *
- * This program is free software; you can redistribute it and/or modify it
- * under the terms of the GNU General Public License (GPL) Version 2 as
- * published by the Free Software Foundation
- *
- * This program is distributed in the hope that it will be useful, but
- * WITHOUT ANY WARRANTY; without even the implied warranty of
- * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU
- * General Public License for more details.
  */
 /*
  * Copyright (c) 2005-2014 Brocade Communications Systems, Inc.
@@ -26,15 +18,43 @@
 
 /* BFA state machine interfaces */
 
-typedef void (*bfa_sm_t)(void *sm, int event);
-
 /* For converting from state machine function to state encoding. */
-struct bfa_sm_table {
-	bfa_sm_t	sm;	/*!< state machine function	*/
-	int		state;	/*!< state machine encoding	*/
-	char		*name;	/*!< state name for display	*/
-};
-#define BFA_SM(_sm)		((bfa_sm_t)(_sm))
+#define BFA_SM_TABLE(n, s, e, t)				\
+struct s;							\
+enum e;								\
+typedef void (*t)(struct s *, enum e);				\
+								\
+struct n ## _sm_table_s {					\
+	t		sm;	/* state machine function */	\
+	int		state;	/* state machine encoding */	\
+	char		*name;	/* state name for display */	\
+};								\
+								\
+static inline int						\
+n ## _sm_to_state(struct n ## _sm_table_s *smt, t sm)		\
+{								\
+	int	i = 0;						\
+								\
+	while (smt[i].sm && smt[i].sm != sm)			\
+		i++;						\
+	return smt[i].state;					\
+}
+
+BFA_SM_TABLE(iocpf,	bfa_iocpf,	iocpf_event,	bfa_fsm_iocpf_t)
+BFA_SM_TABLE(ioc,	bfa_ioc,	ioc_event,	bfa_fsm_ioc_t)
+BFA_SM_TABLE(cmdq,	bfa_msgq_cmdq,	cmdq_event,	bfa_fsm_msgq_cmdq_t)
+BFA_SM_TABLE(rspq,	bfa_msgq_rspq,	rspq_event,	bfa_fsm_msgq_rspq_t)
+
+BFA_SM_TABLE(ioceth,	bna_ioceth,	bna_ioceth_event, bna_fsm_ioceth_t)
+BFA_SM_TABLE(enet,	bna_enet,	bna_enet_event, bna_fsm_enet_t)
+BFA_SM_TABLE(ethport,	bna_ethport,	bna_ethport_event, bna_fsm_ethport_t)
+BFA_SM_TABLE(tx,	bna_tx,		bna_tx_event,	bna_fsm_tx_t)
+BFA_SM_TABLE(rxf,	bna_rxf,	bna_rxf_event, bna_fsm_rxf_t)
+BFA_SM_TABLE(rx,	bna_rx,		bna_rx_event,	bna_fsm_rx_t)
+
+#undef BFA_SM_TABLE
+
+#define BFA_SM(_sm)	(_sm)
 
 /* State machine with entry actions. */
 typedef void (*bfa_fsm_t)(void *fsm, int event);
@@ -49,24 +69,12 @@ typedef void (*bfa_fsm_t)(void *fsm, int event);
 	static void oc ## _sm_ ## st ## _entry(otype * fsm)
 
 #define bfa_fsm_set_state(_fsm, _state) do {				\
-	(_fsm)->fsm = (bfa_fsm_t)(_state);				\
+	(_fsm)->fsm = (_state);						\
 	_state ## _entry(_fsm);						\
 } while (0)
 
 #define bfa_fsm_send_event(_fsm, _event)	((_fsm)->fsm((_fsm), (_event)))
-#define bfa_fsm_cmp_state(_fsm, _state)					\
-	((_fsm)->fsm == (bfa_fsm_t)(_state))
-
-static inline int
-bfa_sm_to_state(const struct bfa_sm_table *smt, bfa_sm_t sm)
-{
-	int	i = 0;
-
-	while (smt[i].sm && smt[i].sm != sm)
-		i++;
-	return smt[i].state;
-}
-
+#define bfa_fsm_cmp_state(_fsm, _state)		((_fsm)->fsm == (_state))
 /* Generic wait counter. */
 
 typedef void (*bfa_wc_resume_t) (void *cbarg);
diff --git a/drivers/net/ethernet/brocade/bna/bfa_defs.h b/drivers/net/ethernet/brocade/bna/bfa_defs.h
index d152b3fa6c54..b08b16864bb0 100644
--- a/drivers/net/ethernet/brocade/bna/bfa_defs.h
+++ b/drivers/net/ethernet/brocade/bna/bfa_defs.h
@@ -1,14 +1,6 @@
+/* SPDX-License-Identifier: GPL-2.0-only */
 /*
  * Linux network driver for QLogic BR-series Converged Network Adapter.
- *
- * This program is free software; you can redistribute it and/or modify it
- * under the terms of the GNU General Public License (GPL) Version 2 as
- * published by the Free Software Foundation
- *
- * This program is distributed in the hope that it will be useful, but
- * WITHOUT ANY WARRANTY; without even the implied warranty of
- * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU
- * General Public License for more details.
  */
 /*
  * Copyright (c) 2005-2014 Brocade Communications Systems, Inc.
diff --git a/drivers/net/ethernet/brocade/bna/bfa_defs_cna.h b/drivers/net/ethernet/brocade/bna/bfa_defs_cna.h
index f048887cbb81..50d3562f7d32 100644
--- a/drivers/net/ethernet/brocade/bna/bfa_defs_cna.h
+++ b/drivers/net/ethernet/brocade/bna/bfa_defs_cna.h
@@ -1,14 +1,6 @@
+/* SPDX-License-Identifier: GPL-2.0-only */
 /*
  * Linux network driver for QLogic BR-series Converged Network Adapter.
- *
- * This program is free software; you can redistribute it and/or modify it
- * under the terms of the GNU General Public License (GPL) Version 2 as
- * published by the Free Software Foundation
- *
- * This program is distributed in the hope that it will be useful, but
- * WITHOUT ANY WARRANTY; without even the implied warranty of
- * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU
- * General Public License for more details.
  */
 /*
  * Copyright (c) 2005-2014 Brocade Communications Systems, Inc.
diff --git a/drivers/net/ethernet/brocade/bna/bfa_defs_mfg_comm.h b/drivers/net/ethernet/brocade/bna/bfa_defs_mfg_comm.h
index 7e17451c94d1..0478f35ae685 100644
--- a/drivers/net/ethernet/brocade/bna/bfa_defs_mfg_comm.h
+++ b/drivers/net/ethernet/brocade/bna/bfa_defs_mfg_comm.h
@@ -1,14 +1,6 @@
+/* SPDX-License-Identifier: GPL-2.0-only */
 /*
  * Linux network driver for QLogic BR-series Converged Network Adapter.
- *
- * This program is free software; you can redistribute it and/or modify it
- * under the terms of the GNU General Public License (GPL) Version 2 as
- * published by the Free Software Foundation
- *
- * This program is distributed in the hope that it will be useful, but
- * WITHOUT ANY WARRANTY; without even the implied warranty of
- * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU
- * General Public License for more details.
  */
 /*
  * Copyright (c) 2005-2014 Brocade Communications Systems, Inc.
diff --git a/drivers/net/ethernet/brocade/bna/bfa_defs_status.h b/drivers/net/ethernet/brocade/bna/bfa_defs_status.h
index a43b56002752..0ed9ec2e68a6 100644
--- a/drivers/net/ethernet/brocade/bna/bfa_defs_status.h
+++ b/drivers/net/ethernet/brocade/bna/bfa_defs_status.h
@@ -1,14 +1,6 @@
+/* SPDX-License-Identifier: GPL-2.0-only */
 /*
  * Linux network driver for QLogic BR-series Converged Network Adapter.
- *
- * This program is free software; you can redistribute it and/or modify it
- * under the terms of the GNU General Public License (GPL) Version 2 as
- * published by the Free Software Foundation
- *
- * This program is distributed in the hope that it will be useful, but
- * WITHOUT ANY WARRANTY; without even the implied warranty of
- * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU
- * General Public License for more details.
  */
 /*
  * Copyright (c) 2005-2014 Brocade Communications Systems, Inc.
diff --git a/drivers/net/ethernet/brocade/bna/bfa_ioc.c b/drivers/net/ethernet/brocade/bna/bfa_ioc.c
index a36e38676640..92c7639d1fc7 100644
--- a/drivers/net/ethernet/brocade/bna/bfa_ioc.c
+++ b/drivers/net/ethernet/brocade/bna/bfa_ioc.c
@@ -1,14 +1,6 @@
+// SPDX-License-Identifier: GPL-2.0-only
 /*
  * Linux network driver for QLogic BR-series Converged Network Adapter.
- *
- * This program is free software; you can redistribute it and/or modify it
- * under the terms of the GNU General Public License (GPL) Version 2 as
- * published by the Free Software Foundation
- *
- * This program is distributed in the hope that it will be useful, but
- * WITHOUT ANY WARRANTY; without even the implied warranty of
- * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU
- * General Public License for more details.
  */
 /*
  * Copyright (c) 2005-2014 Brocade Communications Systems, Inc.
@@ -122,7 +114,7 @@ bfa_fsm_state_decl(bfa_ioc, disabling, struct bfa_ioc, enum ioc_event);
 bfa_fsm_state_decl(bfa_ioc, disabled, struct bfa_ioc, enum ioc_event);
 bfa_fsm_state_decl(bfa_ioc, hwfail, struct bfa_ioc, enum ioc_event);
 
-static struct bfa_sm_table ioc_sm_table[] = {
+static struct ioc_sm_table_s ioc_sm_table[] = {
 	{BFA_SM(bfa_ioc_sm_uninit), BFA_IOC_UNINIT},
 	{BFA_SM(bfa_ioc_sm_reset), BFA_IOC_RESET},
 	{BFA_SM(bfa_ioc_sm_enabling), BFA_IOC_ENABLING},
@@ -191,7 +183,7 @@ bfa_fsm_state_decl(bfa_iocpf, disabling_sync, struct bfa_iocpf,
 						enum iocpf_event);
 bfa_fsm_state_decl(bfa_iocpf, disabled, struct bfa_iocpf, enum iocpf_event);
 
-static struct bfa_sm_table iocpf_sm_table[] = {
+static struct iocpf_sm_table_s iocpf_sm_table[] = {
 	{BFA_SM(bfa_iocpf_sm_reset), BFA_IOCPF_RESET},
 	{BFA_SM(bfa_iocpf_sm_fwcheck), BFA_IOCPF_FWMISMATCH},
 	{BFA_SM(bfa_iocpf_sm_mismatch), BFA_IOCPF_FWMISMATCH},
@@ -277,7 +269,7 @@ bfa_ioc_sm_enabling(struct bfa_ioc *ioc, enum ioc_event event)
 		break;
 
 	case IOC_E_PFFAILED:
-		/* !!! fall through !!! */
+		fallthrough;
 	case IOC_E_HWERROR:
 		ioc->cbfn->enable_cbfn(ioc->bfa, BFA_STATUS_IOC_FAILURE);
 		bfa_fsm_set_state(ioc, bfa_ioc_sm_fail);
@@ -322,14 +314,14 @@ bfa_ioc_sm_getattr(struct bfa_ioc *ioc, enum ioc_event event)
 {
 	switch (event) {
 	case IOC_E_FWRSP_GETATTR:
-		del_timer(&ioc->ioc_timer);
+		timer_delete(&ioc->ioc_timer);
 		bfa_fsm_set_state(ioc, bfa_ioc_sm_op);
 		break;
 
 	case IOC_E_PFFAILED:
 	case IOC_E_HWERROR:
-		del_timer(&ioc->ioc_timer);
-		/* fall through */
+		timer_delete(&ioc->ioc_timer);
+		fallthrough;
 	case IOC_E_TIMEOUT:
 		ioc->cbfn->enable_cbfn(ioc->bfa, BFA_STATUS_IOC_FAILURE);
 		bfa_fsm_set_state(ioc, bfa_ioc_sm_fail);
@@ -338,7 +330,7 @@ bfa_ioc_sm_getattr(struct bfa_ioc *ioc, enum ioc_event event)
 		break;
 
 	case IOC_E_DISABLE:
-		del_timer(&ioc->ioc_timer);
+		timer_delete(&ioc->ioc_timer);
 		bfa_fsm_set_state(ioc, bfa_ioc_sm_disabling);
 		break;
 
@@ -373,7 +365,8 @@ bfa_ioc_sm_op(struct bfa_ioc *ioc, enum ioc_event event)
 	case IOC_E_PFFAILED:
 	case IOC_E_HWERROR:
 		bfa_ioc_hb_stop(ioc);
-		/* !!! fall through !!! */
+		fallthrough;
+
 	case IOC_E_HBFAIL:
 		if (ioc->iocpf.auto_recover)
 			bfa_fsm_set_state(ioc, bfa_ioc_sm_fail_retry);
@@ -666,13 +659,13 @@ bfa_iocpf_sm_mismatch(struct bfa_iocpf *iocpf, enum iocpf_event event)
 		break;
 
 	case IOCPF_E_DISABLE:
-		del_timer(&ioc->iocpf_timer);
+		timer_delete(&ioc->iocpf_timer);
 		bfa_fsm_set_state(iocpf, bfa_iocpf_sm_reset);
 		bfa_ioc_pf_disabled(ioc);
 		break;
 
 	case IOCPF_E_STOP:
-		del_timer(&ioc->iocpf_timer);
+		timer_delete(&ioc->iocpf_timer);
 		bfa_fsm_set_state(iocpf, bfa_iocpf_sm_reset);
 		break;
 
@@ -743,12 +736,12 @@ bfa_iocpf_sm_hwinit(struct bfa_iocpf *iocpf, enum iocpf_event event)
 
 	case IOCPF_E_TIMEOUT:
 		bfa_nw_ioc_hw_sem_release(ioc);
-			bfa_ioc_pf_failed(ioc);
+		bfa_ioc_pf_failed(ioc);
 		bfa_fsm_set_state(iocpf, bfa_iocpf_sm_initfail_sync);
 		break;
 
 	case IOCPF_E_DISABLE:
-		del_timer(&ioc->iocpf_timer);
+		timer_delete(&ioc->iocpf_timer);
 		bfa_ioc_sync_leave(ioc);
 		bfa_nw_ioc_hw_sem_release(ioc);
 		bfa_fsm_set_state(iocpf, bfa_iocpf_sm_disabled);
@@ -781,16 +774,15 @@ bfa_iocpf_sm_enabling(struct bfa_iocpf *iocpf, enum iocpf_event event)
 
 	switch (event) {
 	case IOCPF_E_FWRSP_ENABLE:
-		del_timer(&ioc->iocpf_timer);
+		timer_delete(&ioc->iocpf_timer);
 		bfa_nw_ioc_hw_sem_release(ioc);
 		bfa_fsm_set_state(iocpf, bfa_iocpf_sm_ready);
 		break;
 
 	case IOCPF_E_INITFAIL:
-		del_timer(&ioc->iocpf_timer);
-		/*
-		 * !!! fall through !!!
-		 */
+		timer_delete(&ioc->iocpf_timer);
+		fallthrough;
+
 	case IOCPF_E_TIMEOUT:
 		bfa_nw_ioc_hw_sem_release(ioc);
 		if (event == IOCPF_E_TIMEOUT)
@@ -799,7 +791,7 @@ bfa_iocpf_sm_enabling(struct bfa_iocpf *iocpf, enum iocpf_event event)
 		break;
 
 	case IOCPF_E_DISABLE:
-		del_timer(&ioc->iocpf_timer);
+		timer_delete(&ioc->iocpf_timer);
 		bfa_nw_ioc_hw_sem_release(ioc);
 		bfa_fsm_set_state(iocpf, bfa_iocpf_sm_disabling);
 		break;
@@ -852,15 +844,13 @@ bfa_iocpf_sm_disabling(struct bfa_iocpf *iocpf, enum iocpf_event event)
 
 	switch (event) {
 	case IOCPF_E_FWRSP_DISABLE:
-		del_timer(&ioc->iocpf_timer);
+		timer_delete(&ioc->iocpf_timer);
 		bfa_fsm_set_state(iocpf, bfa_iocpf_sm_disabling_sync);
 		break;
 
 	case IOCPF_E_FAIL:
-		del_timer(&ioc->iocpf_timer);
-		/*
-		 * !!! fall through !!!
-		 */
+		timer_delete(&ioc->iocpf_timer);
+		fallthrough;
 
 	case IOCPF_E_TIMEOUT:
 		bfa_ioc_set_cur_ioc_fwstate(ioc, BFI_IOC_FAIL);
@@ -1135,11 +1125,10 @@ bfa_nw_ioc_sem_release(void __iomem *sem_reg)
 static void
 bfa_ioc_fwver_clear(struct bfa_ioc *ioc)
 {
-	u32 pgnum, pgoff, loff = 0;
+	u32 pgnum, loff = 0;
 	int i;
 
 	pgnum = PSS_SMEM_PGNUM(ioc->ioc_regs.smem_pg0, loff);
-	pgoff = PSS_SMEM_PGOFF(loff);
 	writel(pgnum, ioc->ioc_regs.host_page_num_fn);
 
 	for (i = 0; i < (sizeof(struct bfi_ioc_image_hdr) / sizeof(u32)); i++) {
@@ -1221,7 +1210,7 @@ bfa_nw_ioc_hw_sem_release(struct bfa_ioc *ioc)
 static void
 bfa_ioc_hw_sem_get_cancel(struct bfa_ioc *ioc)
 {
-	del_timer(&ioc->sem_timer);
+	timer_delete(&ioc->sem_timer);
 }
 
 /* Initialize LPU local memory (aka secondary memory / SRAM) */
@@ -1547,7 +1536,6 @@ static int
 bfa_flash_fifo_flush(void __iomem *pci_bar)
 {
 	u32 i;
-	u32 t;
 	union bfa_flash_dev_status_reg dev_status;
 
 	dev_status.i = readl(pci_bar + FLI_DEV_STATUS_REG);
@@ -1557,7 +1545,7 @@ bfa_flash_fifo_flush(void __iomem *pci_bar)
 
 	/* fifo counter in terms of words */
 	for (i = 0; i < dev_status.r.fifo_cnt; i++)
-		t = readl(pci_bar + FLI_RDDATA_REG);
+		readl(pci_bar + FLI_RDDATA_REG);
 
 	/* Check the device status. It may take some time. */
 	for (i = 0; i < BFA_FLASH_CHECK_MAX; i++) {
@@ -1776,7 +1764,7 @@ bfa_ioc_flash_fwver_cmp(struct bfa_ioc *ioc,
 		return BFI_IOC_IMG_VER_INCOMP;
 }
 
-/**
+/*
  * Returns TRUE if driver is willing to work with current smem f/w version.
  */
 bool
@@ -1994,7 +1982,7 @@ bfa_ioc_hb_monitor(struct bfa_ioc *ioc)
 static void
 bfa_ioc_hb_stop(struct bfa_ioc *ioc)
 {
-	del_timer(&ioc->hb_timer);
+	timer_delete(&ioc->hb_timer);
 }
 
 /* Initiate a full firmware download. */
@@ -2482,6 +2470,7 @@ bfa_ioc_isr(struct bfa_ioc *ioc, struct bfi_mbmsg *m)
  *
  * @ioc:	memory for IOC
  * @bfa:	driver instance structure
+ * @cbfn:	callback function
  */
 void
 bfa_nw_ioc_attach(struct bfa_ioc *ioc, void *bfa, struct bfa_ioc_cbfn *cbfn)
@@ -2513,7 +2502,9 @@ bfa_nw_ioc_detach(struct bfa_ioc *ioc)
 /**
  * bfa_nw_ioc_pci_init - Setup IOC PCI properties.
  *
+ * @ioc:	memory for IOC
  * @pcidev:	PCI device information for this IOC
+ * @clscode:	class code
  */
 void
 bfa_nw_ioc_pci_init(struct bfa_ioc *ioc, struct bfa_pcidev *pcidev,
@@ -2582,6 +2573,7 @@ bfa_nw_ioc_pci_init(struct bfa_ioc *ioc, struct bfa_pcidev *pcidev,
 /**
  * bfa_nw_ioc_mem_claim - Initialize IOC dma memory
  *
+ * @ioc:	memory for IOC
  * @dm_kva:	kernel virtual address of IOC dma memory
  * @dm_pa:	physical address of IOC dma memory
  */
@@ -2649,6 +2641,8 @@ bfa_nw_ioc_mbox_regisr(struct bfa_ioc *ioc, enum bfi_mclass mc,
  *
  * @ioc:	IOC instance
  * @cmd:	Mailbox command
+ * @cbfn:	callback function
+ * @cbarg:	arguments to callback
  *
  * Waits if mailbox is busy. Responsibility of caller to serialize
  */
@@ -2845,7 +2839,7 @@ bfa_ioc_get_adapter_optrom_ver(struct bfa_ioc *ioc, char *optrom_ver)
 static void
 bfa_ioc_get_adapter_manufacturer(struct bfa_ioc *ioc, char *manufacturer)
 {
-	strncpy(manufacturer, BFA_MFG_NAME, BFA_ADAPTER_MFG_NAME_LEN);
+	strscpy_pad(manufacturer, BFA_MFG_NAME, BFA_ADAPTER_MFG_NAME_LEN);
 }
 
 static void
@@ -2866,12 +2860,12 @@ static enum bfa_ioc_state
 bfa_ioc_get_state(struct bfa_ioc *ioc)
 {
 	enum bfa_iocpf_state iocpf_st;
-	enum bfa_ioc_state ioc_st = bfa_sm_to_state(ioc_sm_table, ioc->fsm);
+	enum bfa_ioc_state ioc_st = ioc_sm_to_state(ioc_sm_table, ioc->fsm);
 
 	if (ioc_st == BFA_IOC_ENABLING ||
 		ioc_st == BFA_IOC_FAIL || ioc_st == BFA_IOC_INITFAIL) {
 
-		iocpf_st = bfa_sm_to_state(iocpf_sm_table, ioc->iocpf.fsm);
+		iocpf_st = iocpf_sm_to_state(iocpf_sm_table, ioc->iocpf.fsm);
 
 		switch (iocpf_st) {
 		case BFA_IOCPF_SEMWAIT:
@@ -2989,7 +2983,7 @@ bfa_nw_iocpf_timeout(struct bfa_ioc *ioc)
 {
 	enum bfa_iocpf_state iocpf_st;
 
-	iocpf_st = bfa_sm_to_state(iocpf_sm_table, ioc->iocpf.fsm);
+	iocpf_st = iocpf_sm_to_state(iocpf_sm_table, ioc->iocpf.fsm);
 
 	if (iocpf_st == BFA_IOCPF_HWINIT)
 		bfa_ioc_poll_fwinit(ioc);
diff --git a/drivers/net/ethernet/brocade/bna/bfa_ioc.h b/drivers/net/ethernet/brocade/bna/bfa_ioc.h
index 2c0b4c076355..f30d06ec4ffe 100644
--- a/drivers/net/ethernet/brocade/bna/bfa_ioc.h
+++ b/drivers/net/ethernet/brocade/bna/bfa_ioc.h
@@ -1,14 +1,6 @@
+/* SPDX-License-Identifier: GPL-2.0-only */
 /*
  * Linux network driver for QLogic BR-series Converged Network Adapter.
- *
- * This program is free software; you can redistribute it and/or modify it
- * under the terms of the GNU General Public License (GPL) Version 2 as
- * published by the Free Software Foundation
- *
- * This program is distributed in the hope that it will be useful, but
- * WITHOUT ANY WARRANTY; without even the implied warranty of
- * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU
- * General Public License for more details.
  */
 /*
  * Copyright (c) 2005-2014 Brocade Communications Systems, Inc.
@@ -155,16 +147,20 @@ struct bfa_ioc_notify {
 	(__notify)->cbarg = (__cbarg);				\
 } while (0)
 
+enum iocpf_event;
+
 struct bfa_iocpf {
-	bfa_fsm_t		fsm;
+	void (*fsm)(struct bfa_iocpf *s, enum iocpf_event e);
 	struct bfa_ioc		*ioc;
 	bool			fw_mismatch_notified;
 	bool			auto_recover;
 	u32			poll_time;
 };
 
+enum ioc_event;
+
 struct bfa_ioc {
-	bfa_fsm_t		fsm;
+	void (*fsm)(struct bfa_ioc *s, enum ioc_event e);
 	struct bfa		*bfa;
 	struct bfa_pcidev	pcidev;
 	struct timer_list	ioc_timer;
diff --git a/drivers/net/ethernet/brocade/bna/bfa_ioc_ct.c b/drivers/net/ethernet/brocade/bna/bfa_ioc_ct.c
index 74e5ed55ac01..bd643d841180 100644
--- a/drivers/net/ethernet/brocade/bna/bfa_ioc_ct.c
+++ b/drivers/net/ethernet/brocade/bna/bfa_ioc_ct.c
@@ -1,14 +1,6 @@
+// SPDX-License-Identifier: GPL-2.0-only
 /*
  * Linux network driver for QLogic BR-series Converged Network Adapter.
- *
- * This program is free software; you can redistribute it and/or modify it
- * under the terms of the GNU General Public License (GPL) Version 2 as
- * published by the Free Software Foundation
- *
- * This program is distributed in the hope that it will be useful, but
- * WITHOUT ANY WARRANTY; without even the implied warranty of
- * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU
- * General Public License for more details.
  */
 /*
  * Copyright (c) 2005-2014 Brocade Communications Systems, Inc.
diff --git a/drivers/net/ethernet/brocade/bna/bfa_msgq.c b/drivers/net/ethernet/brocade/bna/bfa_msgq.c
index 9c5bb24e8abb..fa40d5ec6f1c 100644
--- a/drivers/net/ethernet/brocade/bna/bfa_msgq.c
+++ b/drivers/net/ethernet/brocade/bna/bfa_msgq.c
@@ -1,14 +1,6 @@
+// SPDX-License-Identifier: GPL-2.0-only
 /*
  * Linux network driver for QLogic BR-series Converged Network Adapter.
- *
- * This program is free software; you can redistribute it and/or modify it
- * under the terms of the GNU General Public License (GPL) Version 2 as
- * published by the Free Software Foundation
- *
- * This program is distributed in the hope that it will be useful, but
- * WITHOUT ANY WARRANTY; without even the implied warranty of
- * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU
- * General Public License for more details.
  */
 /*
  * Copyright (c) 2005-2014 Brocade Communications Systems, Inc.
@@ -210,7 +202,6 @@ static void
 __cmd_copy(struct bfa_msgq_cmdq *cmdq, struct bfa_msgq_cmd_entry *cmd)
 {
 	size_t len = cmd->msg_size;
-	int num_entries = 0;
 	size_t to_copy;
 	u8 *src, *dst;
 
@@ -227,7 +218,6 @@ __cmd_copy(struct bfa_msgq_cmdq *cmdq, struct bfa_msgq_cmd_entry *cmd)
 		BFA_MSGQ_INDX_ADD(cmdq->producer_index, 1, cmdq->depth);
 		dst = (u8 *)cmdq->addr.kva;
 		dst += (cmdq->producer_index * BFI_MSGQ_CMD_ENTRY_SIZE);
-		num_entries++;
 	}
 
 }
diff --git a/drivers/net/ethernet/brocade/bna/bfa_msgq.h b/drivers/net/ethernet/brocade/bna/bfa_msgq.h
index 66bc8b5acd57..170a4b4bed96 100644
--- a/drivers/net/ethernet/brocade/bna/bfa_msgq.h
+++ b/drivers/net/ethernet/brocade/bna/bfa_msgq.h
@@ -1,14 +1,6 @@
+/* SPDX-License-Identifier: GPL-2.0-only */
 /*
  * Linux network driver for QLogic BR-series Converged Network Adapter.
- *
- * This program is free software; you can redistribute it and/or modify it
- * under the terms of the GNU General Public License (GPL) Version 2 as
- * published by the Free Software Foundation
- *
- * This program is distributed in the hope that it will be useful, but
- * WITHOUT ANY WARRANTY; without even the implied warranty of
- * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU
- * General Public License for more details.
  */
 /*
  * Copyright (c) 2005-2014 Brocade Communications Systems, Inc.
@@ -63,8 +55,10 @@ enum bfa_msgq_cmdq_flags {
 	BFA_MSGQ_CMDQ_F_DB_UPDATE	= 1,
 };
 
+enum cmdq_event;
+
 struct bfa_msgq_cmdq {
-	bfa_fsm_t			fsm;
+	void (*fsm)(struct bfa_msgq_cmdq *s, enum cmdq_event e);
 	enum bfa_msgq_cmdq_flags flags;
 
 	u16			producer_index;
@@ -89,8 +83,10 @@ enum bfa_msgq_rspq_flags {
 
 typedef void (*bfa_msgq_mcfunc_t)(void *cbarg, struct bfi_msgq_mhdr *mhdr);
 
+enum rspq_event;
+
 struct bfa_msgq_rspq {
-	bfa_fsm_t			fsm;
+	void (*fsm)(struct bfa_msgq_rspq *s, enum rspq_event e);
 	enum bfa_msgq_rspq_flags flags;
 
 	u16			producer_index;
diff --git a/drivers/net/ethernet/brocade/bna/bfi.h b/drivers/net/ethernet/brocade/bna/bfi.h
index 81e59ea8b4f2..f780d42c946d 100644
--- a/drivers/net/ethernet/brocade/bna/bfi.h
+++ b/drivers/net/ethernet/brocade/bna/bfi.h
@@ -1,14 +1,6 @@
+/* SPDX-License-Identifier: GPL-2.0-only */
 /*
  * Linux network driver for QLogic BR-series Converged Network Adapter.
- *
- * This program is free software; you can redistribute it and/or modify it
- * under the terms of the GNU General Public License (GPL) Version 2 as
- * published by the Free Software Foundation
- *
- * This program is distributed in the hope that it will be useful, but
- * WITHOUT ANY WARRANTY; without even the implied warranty of
- * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU
- * General Public License for more details.
  */
 /*
  * Copyright (c) 2005-2014 Brocade Communications Systems, Inc.
@@ -397,7 +389,7 @@ struct bfi_msgq_mhdr {
 	u16	msg_token;
 	u16	num_entries;
 	u8	enet_id;
-	u8	rsvd[1];
+	u8	rsvd;
 } __packed;
 
 #define bfi_msgq_mhdr_set(_mh, _mc, _mid, _tok, _enet_id) do {	\
diff --git a/drivers/net/ethernet/brocade/bna/bfi_cna.h b/drivers/net/ethernet/brocade/bna/bfi_cna.h
index fad651101c48..fb78bfd0bcf2 100644
--- a/drivers/net/ethernet/brocade/bna/bfi_cna.h
+++ b/drivers/net/ethernet/brocade/bna/bfi_cna.h
@@ -1,14 +1,6 @@
+/* SPDX-License-Identifier: GPL-2.0-only */
 /*
  * Linux network driver for QLogic BR-series Converged Network Adapter.
- *
- * This program is free software; you can redistribute it and/or modify it
- * under the terms of the GNU General Public License (GPL) Version 2 as
- * published by the Free Software Foundation
- *
- * This program is distributed in the hope that it will be useful, but
- * WITHOUT ANY WARRANTY; without even the implied warranty of
- * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU
- * General Public License for more details.
  */
 /*
  * Copyright (c) 2005-2014 Brocade Communications Systems, Inc.
diff --git a/drivers/net/ethernet/brocade/bna/bfi_enet.h b/drivers/net/ethernet/brocade/bna/bfi_enet.h
index d7be7ea8c7f5..112aadf493b1 100644
--- a/drivers/net/ethernet/brocade/bna/bfi_enet.h
+++ b/drivers/net/ethernet/brocade/bna/bfi_enet.h
@@ -1,14 +1,6 @@
+/* SPDX-License-Identifier: GPL-2.0-only */
 /*
  * Linux network driver for QLogic BR-series Converged Network Adapter.
- *
- * This program is free software; you can redistribute it and/or modify it
- * under the terms of the GNU General Public License (GPL) Version 2 as
- * published by the Free Software Foundation
- *
- * This program is distributed in the hope that it will be useful, but
- * WITHOUT ANY WARRANTY; without even the implied warranty of
- * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU
- * General Public License for more details.
  */
 /*
  * Copyright (c) 2005-2014 Brocade Communications Systems, Inc.
diff --git a/drivers/net/ethernet/brocade/bna/bfi_reg.h b/drivers/net/ethernet/brocade/bna/bfi_reg.h
index 2835b51eabec..b15ae9e341af 100644
--- a/drivers/net/ethernet/brocade/bna/bfi_reg.h
+++ b/drivers/net/ethernet/brocade/bna/bfi_reg.h
@@ -1,14 +1,6 @@
+/* SPDX-License-Identifier: GPL-2.0-only */
 /*
  * Linux network driver for QLogic BR-series Converged Network Adapter.
- *
- * This program is free software; you can redistribute it and/or modify it
- * under the terms of the GNU General Public License (GPL) Version 2 as
- * published by the Free Software Foundation
- *
- * This program is distributed in the hope that it will be useful, but
- * WITHOUT ANY WARRANTY; without even the implied warranty of
- * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU
- * General Public License for more details.
  */
 /*
  * Copyright (c) 2005-2014 Brocade Communications Systems, Inc.
diff --git a/drivers/net/ethernet/brocade/bna/bna.h b/drivers/net/ethernet/brocade/bna/bna.h
index 006dcad9a260..50de1532a807 100644
--- a/drivers/net/ethernet/brocade/bna/bna.h
+++ b/drivers/net/ethernet/brocade/bna/bna.h
@@ -1,14 +1,6 @@
+/* SPDX-License-Identifier: GPL-2.0-only */
 /*
  * Linux network driver for QLogic BR-series Converged Network Adapter.
- *
- * This program is free software; you can redistribute it and/or modify it
- * under the terms of the GNU General Public License (GPL) Version 2 as
- * published by the Free Software Foundation
- *
- * This program is distributed in the hope that it will be useful, but
- * WITHOUT ANY WARRANTY; without even the implied warranty of
- * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU
- * General Public License for more details.
  */
 /*
  * Copyright (c) 2005-2014 Brocade Communications Systems, Inc.
diff --git a/drivers/net/ethernet/brocade/bna/bna_enet.c b/drivers/net/ethernet/brocade/bna/bna_enet.c
index bba81735ce87..883de0ac8de4 100644
--- a/drivers/net/ethernet/brocade/bna/bna_enet.c
+++ b/drivers/net/ethernet/brocade/bna/bna_enet.c
@@ -1,14 +1,6 @@
+// SPDX-License-Identifier: GPL-2.0-only
 /*
  * Linux network driver for QLogic BR-series Converged Network Adapter.
- *
- * This program is free software; you can redistribute it and/or modify it
- * under the terms of the GNU General Public License (GPL) Version 2 as
- * published by the Free Software Foundation
- *
- * This program is distributed in the hope that it will be useful, but
- * WITHOUT ANY WARRANTY; without even the implied warranty of
- * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU
- * General Public License for more details.
  */
 /*
  * Copyright (c) 2005-2014 Brocade Communications Systems, Inc.
@@ -1092,7 +1084,7 @@ bna_enet_sm_cfg_wait(struct bna_enet *enet,
 
 	case ENET_E_CHLD_STOPPED:
 		bna_enet_rx_start(enet);
-		/* Fall through */
+		fallthrough;
 	case ENET_E_FWRESP_PAUSE:
 		if (enet->flags & BNA_ENET_F_PAUSE_CHANGED) {
 			enet->flags &= ~BNA_ENET_F_PAUSE_CHANGED;
@@ -1265,7 +1257,7 @@ bna_enet_mtu_get(struct bna_enet *enet)
 void
 bna_enet_enable(struct bna_enet *enet)
 {
-	if (enet->fsm != (bfa_sm_t)bna_enet_sm_stopped)
+	if (enet->fsm != bna_enet_sm_stopped)
 		return;
 
 	enet->flags |= BNA_ENET_F_ENABLED;
@@ -1759,12 +1751,12 @@ bna_ioceth_uninit(struct bna_ioceth *ioceth)
 void
 bna_ioceth_enable(struct bna_ioceth *ioceth)
 {
-	if (ioceth->fsm == (bfa_fsm_t)bna_ioceth_sm_ready) {
+	if (ioceth->fsm == bna_ioceth_sm_ready) {
 		bnad_cb_ioceth_ready(ioceth->bna->bnad);
 		return;
 	}
 
-	if (ioceth->fsm == (bfa_fsm_t)bna_ioceth_sm_stopped)
+	if (ioceth->fsm == bna_ioceth_sm_stopped)
 		bfa_fsm_send_event(ioceth, IOCETH_E_ENABLE);
 }
 
@@ -1797,7 +1789,7 @@ bna_ucam_mod_init(struct bna_ucam_mod *ucam_mod, struct bna *bna,
 
 	/* A separate queue to allow synchronous setting of a list of MACs */
 	INIT_LIST_HEAD(&ucam_mod->del_q);
-	for (i = i; i < (bna->ioceth.attr.num_ucmac * 2); i++)
+	for (; i < (bna->ioceth.attr.num_ucmac * 2); i++)
 		list_add_tail(&ucam_mod->ucmac[i].qe, &ucam_mod->del_q);
 
 	ucam_mod->bna = bna;
@@ -1832,7 +1824,7 @@ bna_mcam_mod_init(struct bna_mcam_mod *mcam_mod, struct bna *bna,
 
 	/* A separate queue to allow synchronous setting of a list of MACs */
 	INIT_LIST_HEAD(&mcam_mod->del_q);
-	for (i = i; i < (bna->ioceth.attr.num_mcmac * 2); i++)
+	for (; i < (bna->ioceth.attr.num_mcmac * 2); i++)
 		list_add_tail(&mcam_mod->mcmac[i].qe, &mcam_mod->del_q);
 
 	mcam_mod->bna = bna;
diff --git a/drivers/net/ethernet/brocade/bna/bna_hw_defs.h b/drivers/net/ethernet/brocade/bna/bna_hw_defs.h
index 52b45c9935aa..dc34e38f97c7 100644
--- a/drivers/net/ethernet/brocade/bna/bna_hw_defs.h
+++ b/drivers/net/ethernet/brocade/bna/bna_hw_defs.h
@@ -1,14 +1,6 @@
+/* SPDX-License-Identifier: GPL-2.0-only */
 /*
  * Linux network driver for QLogic BR-series Converged Network Adapter.
- *
- * This program is free software; you can redistribute it and/or modify it
- * under the terms of the GNU General Public License (GPL) Version 2 as
- * published by the Free Software Foundation
- *
- * This program is distributed in the hope that it will be useful, but
- * WITHOUT ANY WARRANTY; without even the implied warranty of
- * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU
- * General Public License for more details.
  */
 /*
  * Copyright (c) 2005-2014 Brocade Communications Systems, Inc.
@@ -226,17 +218,17 @@ do {									\
 
 /* Set the coalescing timer for the given ib */
 #define bna_ib_coalescing_timer_set(_i_dbell, _cls_timer)		\
-	((_i_dbell)->doorbell_ack = BNA_DOORBELL_IB_INT_ACK((_cls_timer), 0));
+	((_i_dbell)->doorbell_ack = BNA_DOORBELL_IB_INT_ACK((_cls_timer), 0))
 
 /* Acks 'events' # of events for a given ib while disabling interrupts */
 #define bna_ib_ack_disable_irq(_i_dbell, _events)			\
-	(writel(BNA_DOORBELL_IB_INT_ACK(0, (_events)), \
-		(_i_dbell)->doorbell_addr));
+	(writel(BNA_DOORBELL_IB_INT_ACK(0, (_events)),			\
+		(_i_dbell)->doorbell_addr))
 
 /* Acks 'events' # of events for a given ib */
 #define bna_ib_ack(_i_dbell, _events)					\
-	(writel(((_i_dbell)->doorbell_ack | (_events)), \
-		(_i_dbell)->doorbell_addr));
+	(writel(((_i_dbell)->doorbell_ack | (_events)),		\
+		(_i_dbell)->doorbell_addr))
 
 #define bna_ib_start(_bna, _ib, _is_regular)				\
 {									\
@@ -267,12 +259,12 @@ do {									\
 }
 
 #define bna_txq_prod_indx_doorbell(_tcb)				\
-	(writel(BNA_DOORBELL_Q_PRD_IDX((_tcb)->producer_index), \
-		(_tcb)->q_dbell));
+	(writel(BNA_DOORBELL_Q_PRD_IDX((_tcb)->producer_index),		\
+		(_tcb)->q_dbell))
 
 #define bna_rxq_prod_indx_doorbell(_rcb)				\
-	(writel(BNA_DOORBELL_Q_PRD_IDX((_rcb)->producer_index), \
-		(_rcb)->q_dbell));
+	(writel(BNA_DOORBELL_Q_PRD_IDX((_rcb)->producer_index),		\
+		(_rcb)->q_dbell))
 
 /* TxQ, RxQ, CQ related bits, offsets, macros */
 
diff --git a/drivers/net/ethernet/brocade/bna/bna_tx_rx.c b/drivers/net/ethernet/brocade/bna/bna_tx_rx.c
index 95bc470ae441..c05dc7a1c4a1 100644
--- a/drivers/net/ethernet/brocade/bna/bna_tx_rx.c
+++ b/drivers/net/ethernet/brocade/bna/bna_tx_rx.c
@@ -1,14 +1,6 @@
+// SPDX-License-Identifier: GPL-2.0-only
 /*
  * Linux network driver for QLogic BR-series Converged Network Adapter.
- *
- * This program is free software; you can redistribute it and/or modify it
- * under the terms of the GNU General Public License (GPL) Version 2 as
- * published by the Free Software Foundation
- *
- * This program is distributed in the hope that it will be useful, but
- * WITHOUT ANY WARRANTY; without even the implied warranty of
- * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU
- * General Public License for more details.
   */
 /*
  * Copyright (c) 2005-2014 Brocade Communications Systems, Inc.
@@ -1644,7 +1636,7 @@ bna_bfi_rx_enet_start(struct bna_rx *rx)
 						&q1->qpt);
 			cfg_req->q_cfg[i].qs.rx_buffer_size =
 				htons((u16)q1->buffer_size);
-			/* Fall through */
+			fallthrough;
 
 		case BNA_RXP_SINGLE:
 			/* Large/Single RxQ */
@@ -1964,7 +1956,7 @@ static void
 bna_rx_stop(struct bna_rx *rx)
 {
 	rx->rx_flags &= ~BNA_RX_F_ENET_STARTED;
-	if (rx->fsm == (bfa_fsm_t) bna_rx_sm_stopped)
+	if (rx->fsm == bna_rx_sm_stopped)
 		bna_rx_mod_cb_rx_stopped(&rx->bna->rx_mod, rx);
 	else {
 		rx->stop_cbfn = bna_rx_mod_cb_rx_stopped;
@@ -2543,7 +2535,7 @@ bna_rx_destroy(struct bna_rx *rx)
 void
 bna_rx_enable(struct bna_rx *rx)
 {
-	if (rx->fsm != (bfa_sm_t)bna_rx_sm_stopped)
+	if (rx->fsm != bna_rx_sm_stopped)
 		return;
 
 	rx->rx_flags |= BNA_RX_F_ENABLED;
@@ -3531,7 +3523,7 @@ bna_tx_destroy(struct bna_tx *tx)
 void
 bna_tx_enable(struct bna_tx *tx)
 {
-	if (tx->fsm != (bfa_sm_t)bna_tx_sm_stopped)
+	if (tx->fsm != bna_tx_sm_stopped)
 		return;
 
 	tx->flags |= BNA_TX_F_ENABLED;
diff --git a/drivers/net/ethernet/brocade/bna/bna_types.h b/drivers/net/ethernet/brocade/bna/bna_types.h
index c438d032e8bf..986f43d27711 100644
--- a/drivers/net/ethernet/brocade/bna/bna_types.h
+++ b/drivers/net/ethernet/brocade/bna/bna_types.h
@@ -1,14 +1,6 @@
+/* SPDX-License-Identifier: GPL-2.0-only */
 /*
  * Linux network driver for QLogic BR-series Converged Network Adapter.
- *
- * This program is free software; you can redistribute it and/or modify it
- * under the terms of the GNU General Public License (GPL) Version 2 as
- * published by the Free Software Foundation
- *
- * This program is distributed in the hope that it will be useful, but
- * WITHOUT ANY WARRANTY; without even the implied warranty of
- * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU
- * General Public License for more details.
  */
 /*
  * Copyright (c) 2005-2014 Brocade Communications Systems, Inc.
@@ -320,8 +312,10 @@ struct bna_attr {
 
 /* IOCEth */
 
+enum bna_ioceth_event;
+
 struct bna_ioceth {
-	bfa_fsm_t		fsm;
+	void (*fsm)(struct bna_ioceth *s, enum bna_ioceth_event e);
 	struct bfa_ioc ioc;
 
 	struct bna_attr attr;
@@ -342,8 +336,10 @@ struct bna_pause_config {
 	enum bna_status rx_pause;
 };
 
+enum bna_enet_event;
+
 struct bna_enet {
-	bfa_fsm_t		fsm;
+	void (*fsm)(struct bna_enet *s, enum bna_enet_event e);
 	enum bna_enet_flags flags;
 
 	enum bna_enet_type type;
@@ -368,8 +364,10 @@ struct bna_enet {
 
 /* Ethport */
 
+enum bna_ethport_event;
+
 struct bna_ethport {
-	bfa_fsm_t		fsm;
+	void (*fsm)(struct bna_ethport *s, enum bna_ethport_event e);
 	enum bna_ethport_flags flags;
 
 	enum bna_link_status link_status;
@@ -418,7 +416,7 @@ struct bna_ib {
 /* Tx object */
 
 /* Tx datapath control structure */
-#define BNA_Q_NAME_SIZE		16
+#define BNA_Q_NAME_SIZE		(IFNAMSIZ + 6)
 struct bna_tcb {
 	/* Fast path */
 	void			**sw_qpt;
@@ -462,13 +460,16 @@ struct bna_txq {
 };
 
 /* Tx object */
+
+enum bna_tx_event;
+
 struct bna_tx {
 	/* This should be the first one */
 	struct list_head			qe;
 	int			rid;
 	int			hw_id;
 
-	bfa_fsm_t		fsm;
+	void (*fsm)(struct bna_tx *s, enum bna_tx_event e);
 	enum bna_tx_flags flags;
 
 	enum bna_tx_type type;
@@ -706,8 +707,11 @@ struct bna_rxp {
 };
 
 /* RxF structure (hardware Rx Function) */
+
+enum bna_rxf_event;
+
 struct bna_rxf {
-	bfa_fsm_t		fsm;
+	void (*fsm)(struct bna_rxf *s, enum bna_rxf_event e);
 
 	struct bfa_msgq_cmd_entry msgq_cmd;
 	union {
@@ -777,13 +781,16 @@ struct bna_rxf {
 };
 
 /* Rx object */
+
+enum bna_rx_event;
+
 struct bna_rx {
 	/* This should be the first one */
 	struct list_head			qe;
 	int			rid;
 	int			hw_id;
 
-	bfa_fsm_t		fsm;
+	void (*fsm)(struct bna_rx *s, enum bna_rx_event e);
 
 	enum bna_rx_type type;
 
diff --git a/drivers/net/ethernet/brocade/bna/bnad.c b/drivers/net/ethernet/brocade/bna/bnad.c
index 6e13c937d715..9bed33295839 100644
--- a/drivers/net/ethernet/brocade/bna/bnad.c
+++ b/drivers/net/ethernet/brocade/bna/bnad.c
@@ -1,14 +1,6 @@
+// SPDX-License-Identifier: GPL-2.0-only
 /*
  * Linux network driver for QLogic BR-series Converged Network Adapter.
- *
- * This program is free software; you can redistribute it and/or modify it
- * under the terms of the GNU General Public License (GPL) Version 2 as
- * published by the Free Software Foundation
- *
- * This program is distributed in the hope that it will be useful, but
- * WITHOUT ANY WARRANTY; without even the implied warranty of
- * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU
- * General Public License for more details.
  */
 /*
  * Copyright (c) 2005-2014 Brocade Communications Systems, Inc.
@@ -27,6 +19,7 @@
 #include <linux/ip.h>
 #include <linux/prefetch.h>
 #include <linux/module.h>
+#include <net/gro.h>
 
 #include "bnad.h"
 #include "bna.h"
@@ -46,7 +39,7 @@ module_param(bnad_ioc_auto_recover, uint, 0444);
 MODULE_PARM_DESC(bnad_ioc_auto_recover, "Enable / Disable auto recovery");
 
 static uint bna_debugfs_enable = 1;
-module_param(bna_debugfs_enable, uint, S_IRUGO | S_IWUSR);
+module_param(bna_debugfs_enable, uint, 0644);
 MODULE_PARM_DESC(bna_debugfs_enable, "Enables debugfs feature, default=1,"
 		 " Range[false:0|true:1]");
 
@@ -883,7 +876,7 @@ bnad_set_netdev_perm_addr(struct bnad *bnad)
 
 	ether_addr_copy(netdev->perm_addr, bnad->perm_addr);
 	if (is_zero_ether_addr(netdev->dev_addr))
-		ether_addr_copy(netdev->dev_addr, bnad->perm_addr);
+		eth_hw_addr_set(netdev, bnad->perm_addr);
 }
 
 /* Control Path Handlers */
@@ -1045,8 +1038,7 @@ bnad_cb_ccb_destroy(struct bnad *bnad, struct bna_ccb *ccb)
 static void
 bnad_cb_tx_stall(struct bnad *bnad, struct bna_tx *tx)
 {
-	struct bnad_tx_info *tx_info =
-			(struct bnad_tx_info *)tx->priv;
+	struct bnad_tx_info *tx_info = tx->priv;
 	struct bna_tcb *tcb;
 	u32 txq_id;
 	int i;
@@ -1064,7 +1056,7 @@ bnad_cb_tx_stall(struct bnad *bnad, struct bna_tx *tx)
 static void
 bnad_cb_tx_resume(struct bnad *bnad, struct bna_tx *tx)
 {
-	struct bnad_tx_info *tx_info = (struct bnad_tx_info *)tx->priv;
+	struct bnad_tx_info *tx_info = tx->priv;
 	struct bna_tcb *tcb;
 	u32 txq_id;
 	int i;
@@ -1100,10 +1092,10 @@ bnad_cb_tx_resume(struct bnad *bnad, struct bna_tx *tx)
  * Free all TxQs buffers and then notify TX_E_CLEANUP_DONE to Tx fsm.
  */
 static void
-bnad_tx_cleanup(struct delayed_work *work)
+bnad_tx_cleanup(struct work_struct *work)
 {
 	struct bnad_tx_info *tx_info =
-		container_of(work, struct bnad_tx_info, tx_cleanup_work);
+		container_of(work, struct bnad_tx_info, tx_cleanup_work.work);
 	struct bnad *bnad = NULL;
 	struct bna_tcb *tcb;
 	unsigned long flags;
@@ -1141,7 +1133,7 @@ bnad_tx_cleanup(struct delayed_work *work)
 static void
 bnad_cb_tx_cleanup(struct bnad *bnad, struct bna_tx *tx)
 {
-	struct bnad_tx_info *tx_info = (struct bnad_tx_info *)tx->priv;
+	struct bnad_tx_info *tx_info = tx->priv;
 	struct bna_tcb *tcb;
 	int i;
 
@@ -1157,7 +1149,7 @@ bnad_cb_tx_cleanup(struct bnad *bnad, struct bna_tx *tx)
 static void
 bnad_cb_rx_stall(struct bnad *bnad, struct bna_rx *rx)
 {
-	struct bnad_rx_info *rx_info = (struct bnad_rx_info *)rx->priv;
+	struct bnad_rx_info *rx_info = rx->priv;
 	struct bna_ccb *ccb;
 	struct bnad_rx_ctrl *rx_ctrl;
 	int i;
@@ -1179,7 +1171,7 @@ bnad_cb_rx_stall(struct bnad *bnad, struct bna_rx *rx)
  * Free all RxQs buffers and then notify RX_E_CLEANUP_DONE to Rx fsm.
  */
 static void
-bnad_rx_cleanup(void *work)
+bnad_rx_cleanup(struct work_struct *work)
 {
 	struct bnad_rx_info *rx_info =
 		container_of(work, struct bnad_rx_info, rx_cleanup_work);
@@ -1216,7 +1208,7 @@ bnad_rx_cleanup(void *work)
 static void
 bnad_cb_rx_cleanup(struct bnad *bnad, struct bna_rx *rx)
 {
-	struct bnad_rx_info *rx_info = (struct bnad_rx_info *)rx->priv;
+	struct bnad_rx_info *rx_info = rx->priv;
 	struct bna_ccb *ccb;
 	struct bnad_rx_ctrl *rx_ctrl;
 	int i;
@@ -1239,7 +1231,7 @@ bnad_cb_rx_cleanup(struct bnad *bnad, struct bna_rx *rx)
 static void
 bnad_cb_rx_post(struct bnad *bnad, struct bna_rx *rx)
 {
-	struct bnad_rx_info *rx_info = (struct bnad_rx_info *)rx->priv;
+	struct bnad_rx_info *rx_info = rx->priv;
 	struct bna_ccb *ccb;
 	struct bna_rcb *rcb;
 	struct bnad_rx_ctrl *rx_ctrl;
@@ -1543,8 +1535,9 @@ bnad_tx_msix_register(struct bnad *bnad, struct bnad_tx_info *tx_info,
 
 	for (i = 0; i < num_txqs; i++) {
 		vector_num = tx_info->tcb[i]->intr_vector;
-		sprintf(tx_info->tcb[i]->name, "%s TXQ %d", bnad->netdev->name,
-				tx_id + tx_info->tcb[i]->id);
+		snprintf(tx_info->tcb[i]->name, BNA_Q_NAME_SIZE, "%s TXQ %d",
+			 bnad->netdev->name,
+			 tx_id + tx_info->tcb[i]->id);
 		err = request_irq(bnad->msix_table[vector_num].vector,
 				  (irq_handler_t)bnad_msix_tx, 0,
 				  tx_info->tcb[i]->name,
@@ -1594,9 +1587,9 @@ bnad_rx_msix_register(struct bnad *bnad, struct bnad_rx_info *rx_info,
 
 	for (i = 0; i < num_rxps; i++) {
 		vector_num = rx_info->rx_ctrl[i].ccb->intr_vector;
-		sprintf(rx_info->rx_ctrl[i].ccb->name, "%s CQ %d",
-			bnad->netdev->name,
-			rx_id + rx_info->rx_ctrl[i].ccb->id);
+		snprintf(rx_info->rx_ctrl[i].ccb->name, BNA_Q_NAME_SIZE,
+			 "%s CQ %d", bnad->netdev->name,
+			 rx_id + rx_info->rx_ctrl[i].ccb->id);
 		err = request_irq(bnad->msix_table[vector_num].vector,
 				  (irq_handler_t)bnad_msix_rx, 0,
 				  rx_info->rx_ctrl[i].ccb->name,
@@ -1693,9 +1686,10 @@ err_return:
 /* Timer callbacks */
 /* a) IOC timer */
 static void
-bnad_ioc_timeout(unsigned long data)
+bnad_ioc_timeout(struct timer_list *t)
 {
-	struct bnad *bnad = (struct bnad *)data;
+	struct bnad *bnad = timer_container_of(bnad, t,
+					       bna.ioceth.ioc.ioc_timer);
 	unsigned long flags;
 
 	spin_lock_irqsave(&bnad->bna_lock, flags);
@@ -1704,9 +1698,10 @@ bnad_ioc_timeout(unsigned long data)
 }
 
 static void
-bnad_ioc_hb_check(unsigned long data)
+bnad_ioc_hb_check(struct timer_list *t)
 {
-	struct bnad *bnad = (struct bnad *)data;
+	struct bnad *bnad = timer_container_of(bnad, t,
+					       bna.ioceth.ioc.hb_timer);
 	unsigned long flags;
 
 	spin_lock_irqsave(&bnad->bna_lock, flags);
@@ -1715,9 +1710,10 @@ bnad_ioc_hb_check(unsigned long data)
 }
 
 static void
-bnad_iocpf_timeout(unsigned long data)
+bnad_iocpf_timeout(struct timer_list *t)
 {
-	struct bnad *bnad = (struct bnad *)data;
+	struct bnad *bnad = timer_container_of(bnad, t,
+					       bna.ioceth.ioc.iocpf_timer);
 	unsigned long flags;
 
 	spin_lock_irqsave(&bnad->bna_lock, flags);
@@ -1726,9 +1722,10 @@ bnad_iocpf_timeout(unsigned long data)
 }
 
 static void
-bnad_iocpf_sem_timeout(unsigned long data)
+bnad_iocpf_sem_timeout(struct timer_list *t)
 {
-	struct bnad *bnad = (struct bnad *)data;
+	struct bnad *bnad = timer_container_of(bnad, t,
+					       bna.ioceth.ioc.sem_timer);
 	unsigned long flags;
 
 	spin_lock_irqsave(&bnad->bna_lock, flags);
@@ -1742,15 +1739,15 @@ bnad_iocpf_sem_timeout(unsigned long data)
  *	Time	CPU m	CPU n
  *	0       1 = test_bit
  *	1			clear_bit
- *	2			del_timer_sync
+ *	2			timer_delete_sync
  *	3	mod_timer
  */
 
 /* b) Dynamic Interrupt Moderation Timer */
 static void
-bnad_dim_timeout(unsigned long data)
+bnad_dim_timeout(struct timer_list *t)
 {
-	struct bnad *bnad = (struct bnad *)data;
+	struct bnad *bnad = timer_container_of(bnad, t, dim_timer);
 	struct bnad_rx_info *rx_info;
 	struct bnad_rx_ctrl *rx_ctrl;
 	int i, j;
@@ -1772,7 +1769,7 @@ bnad_dim_timeout(unsigned long data)
 		}
 	}
 
-	/* Check for BNAD_CF_DIM_ENABLED, does not eleminate a race */
+	/* Check for BNAD_CF_DIM_ENABLED, does not eliminate a race */
 	if (test_bit(BNAD_RF_DIM_TIMER_RUNNING, &bnad->run_flags))
 		mod_timer(&bnad->dim_timer,
 			  jiffies + msecs_to_jiffies(BNAD_DIM_TIMER_FREQ));
@@ -1781,9 +1778,9 @@ bnad_dim_timeout(unsigned long data)
 
 /* c)  Statistics Timer */
 static void
-bnad_stats_timeout(unsigned long data)
+bnad_stats_timeout(struct timer_list *t)
 {
-	struct bnad *bnad = (struct bnad *)data;
+	struct bnad *bnad = timer_container_of(bnad, t, stats_timer);
 	unsigned long flags;
 
 	if (!netif_running(bnad->netdev) ||
@@ -1804,8 +1801,7 @@ bnad_dim_timer_start(struct bnad *bnad)
 {
 	if (bnad->cfg_flags & BNAD_CF_DIM_ENABLED &&
 	    !test_bit(BNAD_RF_DIM_TIMER_RUNNING, &bnad->run_flags)) {
-		setup_timer(&bnad->dim_timer, bnad_dim_timeout,
-			    (unsigned long)bnad);
+		timer_setup(&bnad->dim_timer, bnad_dim_timeout, 0);
 		set_bit(BNAD_RF_DIM_TIMER_RUNNING, &bnad->run_flags);
 		mod_timer(&bnad->dim_timer,
 			  jiffies + msecs_to_jiffies(BNAD_DIM_TIMER_FREQ));
@@ -1823,8 +1819,7 @@ bnad_stats_timer_start(struct bnad *bnad)
 
 	spin_lock_irqsave(&bnad->bna_lock, flags);
 	if (!test_and_set_bit(BNAD_RF_STATS_TIMER_RUNNING, &bnad->run_flags)) {
-		setup_timer(&bnad->stats_timer, bnad_stats_timeout,
-			    (unsigned long)bnad);
+		timer_setup(&bnad->stats_timer, bnad_stats_timeout, 0);
 		mod_timer(&bnad->stats_timer,
 			  jiffies + msecs_to_jiffies(BNAD_STATS_TIMER_FREQ));
 	}
@@ -1846,7 +1841,7 @@ bnad_stats_timer_stop(struct bnad *bnad)
 		to_del = 1;
 	spin_unlock_irqrestore(&bnad->bna_lock, flags);
 	if (to_del)
-		del_timer_sync(&bnad->stats_timer);
+		timer_delete_sync(&bnad->stats_timer);
 }
 
 /* Utilities */
@@ -1891,7 +1886,6 @@ poll_exit:
 	return rcvd;
 }
 
-#define BNAD_NAPI_POLL_QUOTA		64
 static void
 bnad_napi_add(struct bnad *bnad, u32 rx_id)
 {
@@ -1902,7 +1896,7 @@ bnad_napi_add(struct bnad *bnad, u32 rx_id)
 	for (i = 0; i <	bnad->num_rxp_per_rx; i++) {
 		rx_ctrl = &bnad->rx_info[rx_id].rx_ctrl[i];
 		netif_napi_add(bnad->netdev, &rx_ctrl->napi,
-			       bnad_napi_poll_rx, BNAD_NAPI_POLL_QUOTA);
+			       bnad_napi_poll_rx);
 	}
 }
 
@@ -2003,8 +1997,7 @@ bnad_setup_tx(struct bnad *bnad, u32 tx_id)
 	}
 	tx_info->tx = tx;
 
-	INIT_DELAYED_WORK(&tx_info->tx_cleanup_work,
-			(work_func_t)bnad_tx_cleanup);
+	INIT_DELAYED_WORK(&tx_info->tx_cleanup_work, bnad_tx_cleanup);
 
 	/* Register ISR for the Tx object */
 	if (intr_info->intr_type == BNA_INTR_T_MSIX) {
@@ -2171,7 +2164,7 @@ bnad_destroy_rx(struct bnad *bnad, u32 rx_id)
 		}
 		spin_unlock_irqrestore(&bnad->bna_lock, flags);
 		if (to_del)
-			del_timer_sync(&bnad->dim_timer);
+			timer_delete_sync(&bnad->dim_timer);
 	}
 
 	init_completion(&bnad->bnad_completions.rx_comp);
@@ -2260,8 +2253,7 @@ bnad_setup_rx(struct bnad *bnad, u32 rx_id)
 	rx_info->rx = rx;
 	spin_unlock_irqrestore(&bnad->bna_lock, flags);
 
-	INIT_WORK(&rx_info->rx_cleanup_work,
-			(work_func_t)(bnad_rx_cleanup));
+	INIT_WORK(&rx_info->rx_cleanup_work, bnad_rx_cleanup);
 
 	/*
 	 * Init NAPI, so that state is set to NAPI_STATE_SCHED,
@@ -2514,12 +2506,7 @@ bnad_tso_prepare(struct bnad *bnad, struct sk_buff *skb)
 					   IPPROTO_TCP, 0);
 		BNAD_UPDATE_CTR(bnad, tso4);
 	} else {
-		struct ipv6hdr *ipv6h = ipv6_hdr(skb);
-
-		ipv6h->payload_len = 0;
-		tcp_hdr(skb)->check =
-			~csum_ipv6_magic(&ipv6h->saddr, &ipv6h->daddr, 0,
-					 IPPROTO_TCP, 0);
+		tcp_v6_gso_csum_prep(skb);
 		BNAD_UPDATE_CTR(bnad, tso6);
 	}
 
@@ -2839,8 +2826,7 @@ bnad_txq_wi_prepare(struct bnad *bnad, struct bna_tcb *tcb,
 			BNAD_UPDATE_CTR(bnad, tx_skb_mss_too_long);
 			return -EINVAL;
 		}
-		if (unlikely((gso_size + skb_transport_offset(skb) +
-			      tcp_hdrlen(skb)) >= skb->len)) {
+		if (unlikely((gso_size + skb_tcp_all_headers(skb)) >= skb->len)) {
 			txqent->hdr.wi.opcode = htons(BNA_TXQ_WI_SEND);
 			txqent->hdr.wi.lso_mss = 0;
 			BNAD_UPDATE_CTR(bnad, tx_skb_tso_too_short);
@@ -2888,8 +2874,7 @@ bnad_txq_wi_prepare(struct bnad *bnad, struct bna_tcb *tcb,
 				BNAD_UPDATE_CTR(bnad, tcpcsum_offload);
 
 				if (unlikely(skb_headlen(skb) <
-					    skb_transport_offset(skb) +
-				    tcp_hdrlen(skb))) {
+					    skb_tcp_all_headers(skb))) {
 					BNAD_UPDATE_CTR(bnad, tx_skb_tcp_hdr);
 					return -EINVAL;
 				}
@@ -3042,7 +3027,7 @@ bnad_start_xmit(struct sk_buff *skb, struct net_device *netdev)
 	head_unmap->nvecs++;
 
 	for (i = 0, vect_id = 0; i < vectors - 1; i++) {
-		const struct skb_frag_struct *frag = &skb_shinfo(skb)->frags[i];
+		const skb_frag_t *frag = &skb_shinfo(skb)->frags[i];
 		u32		size = skb_frag_size(frag);
 
 		if (unlikely(size == 0)) {
@@ -3143,7 +3128,7 @@ bnad_set_rx_ucast_fltr(struct bnad *bnad)
 	if (uc_count > bna_attr(&bnad->bna)->num_ucmac)
 		goto mode_default;
 
-	mac_list = kzalloc(uc_count * ETH_ALEN, GFP_ATOMIC);
+	mac_list = kcalloc(ETH_ALEN, uc_count, GFP_ATOMIC);
 	if (mac_list == NULL)
 		goto mode_default;
 
@@ -3184,7 +3169,7 @@ bnad_set_rx_mcast_fltr(struct bnad *bnad)
 	if (mc_count > bna_attr(&bnad->bna)->num_mcmac)
 		goto mode_allmulti;
 
-	mac_list = kzalloc((mc_count + 1) * ETH_ALEN, GFP_ATOMIC);
+	mac_list = kcalloc(mc_count + 1, ETH_ALEN, GFP_ATOMIC);
 
 	if (mac_list == NULL)
 		goto mode_allmulti;
@@ -3264,7 +3249,7 @@ bnad_set_mac_address(struct net_device *netdev, void *addr)
 
 	err = bnad_mac_addr_set_locked(bnad, sa->sa_data);
 	if (!err)
-		ether_addr_copy(netdev->dev_addr, sa->sa_data);
+		eth_hw_addr_set(netdev, sa->sa_data);
 
 	spin_unlock_irqrestore(&bnad->bna_lock, flags);
 
@@ -3292,12 +3277,12 @@ bnad_change_mtu(struct net_device *netdev, int new_mtu)
 {
 	int err, mtu;
 	struct bnad *bnad = netdev_priv(netdev);
-	u32 rx_count = 0, frame, new_frame;
+	u32 frame, new_frame;
 
 	mutex_lock(&bnad->conf_mutex);
 
 	mtu = netdev->mtu;
-	netdev->mtu = new_mtu;
+	WRITE_ONCE(netdev->mtu, new_mtu);
 
 	frame = BNAD_FRAME_SIZE(mtu);
 	new_frame = BNAD_FRAME_SIZE(new_mtu);
@@ -3308,12 +3293,9 @@ bnad_change_mtu(struct net_device *netdev, int new_mtu)
 		/* only when transition is over 4K */
 		if ((frame <= 4096 && new_frame > 4096) ||
 		    (frame > 4096 && new_frame <= 4096))
-			rx_count = bnad_reinit_rx(bnad);
+			bnad_reinit_rx(bnad);
 	}
 
-	/* rx_count > 0 - new rx created
-	 *	- Linux set err = 0 and return
-	 */
 	err = bnad_mtu_set(bnad, new_frame);
 	if (err)
 		err = -EBUSY;
@@ -3439,7 +3421,7 @@ static const struct net_device_ops bnad_netdev_ops = {
 };
 
 static void
-bnad_netdev_init(struct bnad *bnad, bool using_dac)
+bnad_netdev_init(struct bnad *bnad)
 {
 	struct net_device *netdev = bnad->netdev;
 
@@ -3452,10 +3434,8 @@ bnad_netdev_init(struct bnad *bnad, bool using_dac)
 		NETIF_F_IP_CSUM | NETIF_F_IPV6_CSUM |
 		NETIF_F_TSO | NETIF_F_TSO6;
 
-	netdev->features |= netdev->hw_features | NETIF_F_HW_VLAN_CTAG_FILTER;
-
-	if (using_dac)
-		netdev->features |= NETIF_F_HIGHDMA;
+	netdev->features |= netdev->hw_features | NETIF_F_HW_VLAN_CTAG_FILTER |
+			    NETIF_F_HIGHDMA;
 
 	netdev->mem_start = bnad->mmio_start;
 	netdev->mem_end = bnad->mmio_start + bnad->mmio_len - 1;
@@ -3487,7 +3467,7 @@ bnad_init(struct bnad *bnad,
 	bnad->pcidev = pdev;
 	bnad->mmio_start = pci_resource_start(pdev, 0);
 	bnad->mmio_len = pci_resource_len(pdev, 0);
-	bnad->bar0 = ioremap_nocache(bnad->mmio_start, bnad->mmio_len);
+	bnad->bar0 = ioremap(bnad->mmio_start, bnad->mmio_len);
 	if (!bnad->bar0) {
 		dev_err(&pdev->dev, "ioremap for bar0 failed\n");
 		return -ENOMEM;
@@ -3533,7 +3513,6 @@ static void
 bnad_uninit(struct bnad *bnad)
 {
 	if (bnad->work_q) {
-		flush_workqueue(bnad->work_q);
 		destroy_workqueue(bnad->work_q);
 		bnad->work_q = NULL;
 	}
@@ -3563,8 +3542,7 @@ bnad_lock_uninit(struct bnad *bnad)
 
 /* PCI Initialization */
 static int
-bnad_pci_init(struct bnad *bnad,
-	      struct pci_dev *pdev, bool *using_dac)
+bnad_pci_init(struct bnad *bnad, struct pci_dev *pdev)
 {
 	int err;
 
@@ -3574,14 +3552,9 @@ bnad_pci_init(struct bnad *bnad,
 	err = pci_request_regions(pdev, BNAD_NAME);
 	if (err)
 		goto disable_device;
-	if (!dma_set_mask_and_coherent(&pdev->dev, DMA_BIT_MASK(64))) {
-		*using_dac = true;
-	} else {
-		err = dma_set_mask_and_coherent(&pdev->dev, DMA_BIT_MASK(32));
-		if (err)
-			goto release_regions;
-		*using_dac = false;
-	}
+	err = dma_set_mask_and_coherent(&pdev->dev, DMA_BIT_MASK(64));
+	if (err)
+		goto release_regions;
 	pci_set_master(pdev);
 	return 0;
 
@@ -3604,7 +3577,6 @@ static int
 bnad_pci_probe(struct pci_dev *pdev,
 		const struct pci_device_id *pcidev_id)
 {
-	bool	using_dac;
 	int	err;
 	struct bnad *bnad;
 	struct bna *bna;
@@ -3634,13 +3606,8 @@ bnad_pci_probe(struct pci_dev *pdev,
 	bnad->id = atomic_inc_return(&bna_id) - 1;
 
 	mutex_lock(&bnad->conf_mutex);
-	/*
-	 * PCI initialization
-	 *	Output : using_dac = 1 for 64 bit DMA
-	 *			   = 0 for 32 bit DMA
-	 */
-	using_dac = false;
-	err = bnad_pci_init(bnad, pdev, &using_dac);
+	/* PCI initialization */
+	err = bnad_pci_init(bnad, pdev);
 	if (err)
 		goto unlock_mutex;
 
@@ -3653,7 +3620,7 @@ bnad_pci_probe(struct pci_dev *pdev,
 		goto pci_uninit;
 
 	/* Initialize netdev structure, set up ethtool ops */
-	bnad_netdev_init(bnad, using_dac);
+	bnad_netdev_init(bnad);
 
 	/* Set link to down state */
 	netif_carrier_off(netdev);
@@ -3692,14 +3659,11 @@ bnad_pci_probe(struct pci_dev *pdev,
 		goto res_free;
 
 	/* Set up timers */
-	setup_timer(&bnad->bna.ioceth.ioc.ioc_timer, bnad_ioc_timeout,
-		    (unsigned long)bnad);
-	setup_timer(&bnad->bna.ioceth.ioc.hb_timer, bnad_ioc_hb_check,
-		    (unsigned long)bnad);
-	setup_timer(&bnad->bna.ioceth.ioc.iocpf_timer, bnad_iocpf_timeout,
-		    (unsigned long)bnad);
-	setup_timer(&bnad->bna.ioceth.ioc.sem_timer, bnad_iocpf_sem_timeout,
-		    (unsigned long)bnad);
+	timer_setup(&bnad->bna.ioceth.ioc.ioc_timer, bnad_ioc_timeout, 0);
+	timer_setup(&bnad->bna.ioceth.ioc.hb_timer, bnad_ioc_hb_check, 0);
+	timer_setup(&bnad->bna.ioceth.ioc.iocpf_timer, bnad_iocpf_timeout, 0);
+	timer_setup(&bnad->bna.ioceth.ioc.sem_timer, bnad_iocpf_sem_timeout,
+		    0);
 
 	/*
 	 * Start the chip
@@ -3766,9 +3730,9 @@ probe_uninit:
 	bnad_res_free(bnad, &bnad->mod_res_info[0], BNA_MOD_RES_T_MAX);
 disable_ioceth:
 	bnad_ioceth_disable(bnad);
-	del_timer_sync(&bnad->bna.ioceth.ioc.ioc_timer);
-	del_timer_sync(&bnad->bna.ioceth.ioc.sem_timer);
-	del_timer_sync(&bnad->bna.ioceth.ioc.hb_timer);
+	timer_delete_sync(&bnad->bna.ioceth.ioc.ioc_timer);
+	timer_delete_sync(&bnad->bna.ioceth.ioc.sem_timer);
+	timer_delete_sync(&bnad->bna.ioceth.ioc.hb_timer);
 	spin_lock_irqsave(&bnad->bna_lock, flags);
 	bna_uninit(bna);
 	spin_unlock_irqrestore(&bnad->bna_lock, flags);
@@ -3809,9 +3773,9 @@ bnad_pci_remove(struct pci_dev *pdev)
 
 	mutex_lock(&bnad->conf_mutex);
 	bnad_ioceth_disable(bnad);
-	del_timer_sync(&bnad->bna.ioceth.ioc.ioc_timer);
-	del_timer_sync(&bnad->bna.ioceth.ioc.sem_timer);
-	del_timer_sync(&bnad->bna.ioceth.ioc.hb_timer);
+	timer_delete_sync(&bnad->bna.ioceth.ioc.ioc_timer);
+	timer_delete_sync(&bnad->bna.ioceth.ioc.sem_timer);
+	timer_delete_sync(&bnad->bna.ioceth.ioc.hb_timer);
 	spin_lock_irqsave(&bnad->bna_lock, flags);
 	bna_uninit(bna);
 	spin_unlock_irqrestore(&bnad->bna_lock, flags);
@@ -3860,9 +3824,6 @@ bnad_module_init(void)
 {
 	int err;
 
-	pr_info("bna: QLogic BR-series 10G Ethernet driver - version: %s\n",
-		BNAD_VERSION);
-
 	bfa_nw_ioc_auto_recover(bnad_ioc_auto_recover);
 
 	err = pci_register_driver(&bnad_pci_driver);
@@ -3887,6 +3848,5 @@ module_exit(bnad_module_exit);
 MODULE_AUTHOR("Brocade");
 MODULE_LICENSE("GPL");
 MODULE_DESCRIPTION("QLogic BR-series 10G PCIe Ethernet driver");
-MODULE_VERSION(BNAD_VERSION);
 MODULE_FIRMWARE(CNA_FW_FILE_CT);
 MODULE_FIRMWARE(CNA_FW_FILE_CT2);
diff --git a/drivers/net/ethernet/brocade/bna/bnad.h b/drivers/net/ethernet/brocade/bna/bnad.h
index 46f7b842b39c..4396997c59d0 100644
--- a/drivers/net/ethernet/brocade/bna/bnad.h
+++ b/drivers/net/ethernet/brocade/bna/bnad.h
@@ -1,14 +1,6 @@
+/* SPDX-License-Identifier: GPL-2.0-only */
 /*
  * Linux network driver for QLogic BR-series Converged Network Adapter.
- *
- * This program is free software; you can redistribute it and/or modify it
- * under the terms of the GNU General Public License (GPL) Version 2 as
- * published by the Free Software Foundation
- *
- * This program is distributed in the hope that it will be useful, but
- * WITHOUT ANY WARRANTY; without even the implied warranty of
- * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU
- * General Public License for more details.
  */
 /*
  * Copyright (c) 2005-2014 Brocade Communications Systems, Inc.
@@ -27,7 +19,6 @@
 #include <linux/firmware.h>
 #include <linux/if_vlan.h>
 
-/* Fix for IA64 */
 #include <asm/checksum.h>
 #include <net/ip6_checksum.h>
 
@@ -72,8 +63,6 @@ struct bnad_rx_ctrl {
 #define BNAD_NAME			"bna"
 #define BNAD_NAME_LEN			64
 
-#define BNAD_VERSION			"3.2.25.1"
-
 #define BNAD_MAILBOX_MSIX_INDEX		0
 #define BNAD_MAILBOX_MSIX_VECTORS	1
 #define BNAD_INTX_TX_IB_BITMASK		0x1
@@ -261,7 +250,7 @@ struct bnad_rx_unmap_q {
 	int			alloc_order;
 	u32			map_size;
 	enum bnad_rxbuf_type	type;
-	struct bnad_rx_unmap	unmap[0] ____cacheline_aligned;
+	struct bnad_rx_unmap	unmap[] ____cacheline_aligned;
 };
 
 #define BNAD_PCI_DEV_IS_CAT2(_bnad) \
@@ -362,7 +351,6 @@ struct bnad {
 	/* debugfs specific data */
 	char	*regdata;
 	u32	reglen;
-	struct dentry *bnad_dentry_files[5];
 	struct dentry *port_debugfs_root;
 };
 
diff --git a/drivers/net/ethernet/brocade/bna/bnad_debugfs.c b/drivers/net/ethernet/brocade/bna/bnad_debugfs.c
index cebfe3bd086e..8f0972e6737c 100644
--- a/drivers/net/ethernet/brocade/bna/bnad_debugfs.c
+++ b/drivers/net/ethernet/brocade/bna/bnad_debugfs.c
@@ -1,14 +1,6 @@
+// SPDX-License-Identifier: GPL-2.0-only
 /*
  * Linux network driver for QLogic BR-series Converged Network Adapter.
- *
- * This program is free software; you can redistribute it and/or modify it
- * under the terms of the GNU General Public License (GPL) Version 2 as
- * published by the Free Software Foundation
- *
- * This program is distributed in the hope that it will be useful, but
- * WITHOUT ANY WARRANTY; without even the implied warranty of
- * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU
- * General Public License for more details.
  */
 /*
  * Copyright (c) 2005-2014 Brocade Communications Systems, Inc.
@@ -320,7 +312,7 @@ bnad_debugfs_write_regrd(struct file *file, const char __user *buf,
 	void *kern_buf;
 
 	/* Copy the user space buf */
-	kern_buf = memdup_user(buf, nbytes);
+	kern_buf = memdup_user_nul(buf, nbytes);
 	if (IS_ERR(kern_buf))
 		return PTR_ERR(kern_buf);
 
@@ -380,7 +372,7 @@ bnad_debugfs_write_regwr(struct file *file, const char __user *buf,
 	void *kern_buf;
 
 	/* Copy the user space buf */
-	kern_buf = memdup_user(buf, nbytes);
+	kern_buf = memdup_user_nul(buf, nbytes);
 	if (IS_ERR(kern_buf))
 		return PTR_ERR(kern_buf);
 
@@ -486,11 +478,11 @@ struct bnad_debugfs_entry {
 };
 
 static const struct bnad_debugfs_entry bnad_debugfs_files[] = {
-	{ "fwtrc",  S_IFREG|S_IRUGO, &bnad_debugfs_op_fwtrc, },
-	{ "fwsave", S_IFREG|S_IRUGO, &bnad_debugfs_op_fwsave, },
-	{ "regrd",  S_IFREG|S_IRUGO|S_IWUSR, &bnad_debugfs_op_regrd, },
-	{ "regwr",  S_IFREG|S_IWUSR, &bnad_debugfs_op_regwr, },
-	{ "drvinfo", S_IFREG|S_IRUGO, &bnad_debugfs_op_drvinfo, },
+	{ "fwtrc",  S_IFREG | 0444, &bnad_debugfs_op_fwtrc, },
+	{ "fwsave", S_IFREG | 0444, &bnad_debugfs_op_fwsave, },
+	{ "regrd",  S_IFREG | 0644, &bnad_debugfs_op_regrd, },
+	{ "regwr",  S_IFREG | 0200, &bnad_debugfs_op_regwr, },
+	{ "drvinfo", S_IFREG | 0444, &bnad_debugfs_op_drvinfo, },
 };
 
 static struct dentry *bna_debugfs_root;
@@ -508,11 +500,6 @@ bnad_debugfs_init(struct bnad *bnad)
 	if (!bna_debugfs_root) {
 		bna_debugfs_root = debugfs_create_dir("bna", NULL);
 		atomic_set(&bna_debugfs_port_count, 0);
-		if (!bna_debugfs_root) {
-			netdev_warn(bnad->netdev,
-				    "debugfs root dir creation failed\n");
-			return;
-		}
 	}
 
 	/* Setup the pci_dev debugfs directory for the port */
@@ -520,28 +507,16 @@ bnad_debugfs_init(struct bnad *bnad)
 	if (!bnad->port_debugfs_root) {
 		bnad->port_debugfs_root =
 			debugfs_create_dir(name, bna_debugfs_root);
-		if (!bnad->port_debugfs_root) {
-			netdev_warn(bnad->netdev,
-				    "debugfs root dir creation failed\n");
-			return;
-		}
 
 		atomic_inc(&bna_debugfs_port_count);
 
 		for (i = 0; i < ARRAY_SIZE(bnad_debugfs_files); i++) {
 			file = &bnad_debugfs_files[i];
-			bnad->bnad_dentry_files[i] =
-					debugfs_create_file(file->name,
-							file->mode,
-							bnad->port_debugfs_root,
-							bnad,
-							file->fops);
-			if (!bnad->bnad_dentry_files[i]) {
-				netdev_warn(bnad->netdev,
-					    "create %s entry failed\n",
-					    file->name);
-				return;
-			}
+			debugfs_create_file(file->name,
+					    file->mode,
+					    bnad->port_debugfs_root,
+					    bnad,
+					    file->fops);
 		}
 	}
 }
@@ -550,15 +525,6 @@ bnad_debugfs_init(struct bnad *bnad)
 void
 bnad_debugfs_uninit(struct bnad *bnad)
 {
-	int i;
-
-	for (i = 0; i < ARRAY_SIZE(bnad_debugfs_files); i++) {
-		if (bnad->bnad_dentry_files[i]) {
-			debugfs_remove(bnad->bnad_dentry_files[i]);
-			bnad->bnad_dentry_files[i] = NULL;
-		}
-	}
-
 	/* Remove the pci_dev debugfs directory for the port */
 	if (bnad->port_debugfs_root) {
 		debugfs_remove(bnad->port_debugfs_root);
diff --git a/drivers/net/ethernet/brocade/bna/bnad_ethtool.c b/drivers/net/ethernet/brocade/bna/bnad_ethtool.c
index 31032de5843b..216e25f26dbb 100644
--- a/drivers/net/ethernet/brocade/bna/bnad_ethtool.c
+++ b/drivers/net/ethernet/brocade/bna/bnad_ethtool.c
@@ -1,14 +1,6 @@
+// SPDX-License-Identifier: GPL-2.0-only
 /*
  * Linux network driver for QLogic BR-series Converged Network Adapter.
- *
- * This program is free software; you can redistribute it and/or modify it
- * under the terms of the GNU General Public License (GPL) Version 2 as
- * published by the Free Software Foundation
- *
- * This program is distributed in the hope that it will be useful, but
- * WITHOUT ANY WARRANTY; without even the implied warranty of
- * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU
- * General Public License for more details.
  */
 /*
  * Copyright (c) 2005-2014 Brocade Communications Systems, Inc.
@@ -122,7 +114,7 @@ static const char *bnad_net_stats_strings[] = {
 	"mac_tx_deferral",
 	"mac_tx_excessive_deferral",
 	"mac_tx_single_collision",
-	"mac_tx_muliple_collision",
+	"mac_tx_multiple_collision",
 	"mac_tx_late_collision",
 	"mac_tx_excessive_collision",
 	"mac_tx_total_collision",
@@ -243,13 +235,18 @@ static int
 bnad_get_link_ksettings(struct net_device *netdev,
 			struct ethtool_link_ksettings *cmd)
 {
-	u32 supported, advertising;
-
-	supported = SUPPORTED_10000baseT_Full;
-	advertising = ADVERTISED_10000baseT_Full;
+	ethtool_link_ksettings_zero_link_mode(cmd, supported);
+	ethtool_link_ksettings_zero_link_mode(cmd, advertising);
+
+	ethtool_link_ksettings_add_link_mode(cmd, supported, 10000baseCR_Full);
+	ethtool_link_ksettings_add_link_mode(cmd, supported, 10000baseSR_Full);
+	ethtool_link_ksettings_add_link_mode(cmd, supported, 10000baseLR_Full);
+	ethtool_link_ksettings_add_link_mode(cmd, advertising, 10000baseCR_Full);
+	ethtool_link_ksettings_add_link_mode(cmd, advertising, 10000baseSR_Full);
+	ethtool_link_ksettings_add_link_mode(cmd, advertising, 10000baseLR_Full);
 	cmd->base.autoneg = AUTONEG_DISABLE;
-	supported |= SUPPORTED_FIBRE;
-	advertising |= ADVERTISED_FIBRE;
+	ethtool_link_ksettings_add_link_mode(cmd, supported, FIBRE);
+	ethtool_link_ksettings_add_link_mode(cmd, advertising, FIBRE);
 	cmd->base.port = PORT_FIBRE;
 	cmd->base.phy_address = 0;
 
@@ -261,11 +258,6 @@ bnad_get_link_ksettings(struct net_device *netdev,
 		cmd->base.duplex = DUPLEX_UNKNOWN;
 	}
 
-	ethtool_convert_legacy_u32_to_link_mode(cmd->link_modes.supported,
-						supported);
-	ethtool_convert_legacy_u32_to_link_mode(cmd->link_modes.advertising,
-						advertising);
-
 	return 0;
 }
 
@@ -291,8 +283,7 @@ bnad_get_drvinfo(struct net_device *netdev, struct ethtool_drvinfo *drvinfo)
 	struct bfa_ioc_attr *ioc_attr;
 	unsigned long flags;
 
-	strlcpy(drvinfo->driver, BNAD_NAME, sizeof(drvinfo->driver));
-	strlcpy(drvinfo->version, BNAD_VERSION, sizeof(drvinfo->version));
+	strscpy(drvinfo->driver, BNAD_NAME, sizeof(drvinfo->driver));
 
 	ioc_attr = kzalloc(sizeof(*ioc_attr), GFP_KERNEL);
 	if (ioc_attr) {
@@ -300,12 +291,12 @@ bnad_get_drvinfo(struct net_device *netdev, struct ethtool_drvinfo *drvinfo)
 		bfa_nw_ioc_get_attr(&bnad->bna.ioceth.ioc, ioc_attr);
 		spin_unlock_irqrestore(&bnad->bna_lock, flags);
 
-		strlcpy(drvinfo->fw_version, ioc_attr->adapter_attr.fw_ver,
+		strscpy(drvinfo->fw_version, ioc_attr->adapter_attr.fw_ver,
 			sizeof(drvinfo->fw_version));
 		kfree(ioc_attr);
 	}
 
-	strlcpy(drvinfo->bus_info, pci_name(bnad->pcidev),
+	strscpy(drvinfo->bus_info, pci_name(bnad->pcidev),
 		sizeof(drvinfo->bus_info));
 }
 
@@ -316,8 +307,10 @@ bnad_get_wol(struct net_device *netdev, struct ethtool_wolinfo *wolinfo)
 	wolinfo->wolopts = 0;
 }
 
-static int
-bnad_get_coalesce(struct net_device *netdev, struct ethtool_coalesce *coalesce)
+static int bnad_get_coalesce(struct net_device *netdev,
+			     struct ethtool_coalesce *coalesce,
+			     struct kernel_ethtool_coalesce *kernel_coal,
+			     struct netlink_ext_ack *extack)
 {
 	struct bnad *bnad = netdev_priv(netdev);
 	unsigned long flags;
@@ -337,8 +330,10 @@ bnad_get_coalesce(struct net_device *netdev, struct ethtool_coalesce *coalesce)
 	return 0;
 }
 
-static int
-bnad_set_coalesce(struct net_device *netdev, struct ethtool_coalesce *coalesce)
+static int bnad_set_coalesce(struct net_device *netdev,
+			     struct ethtool_coalesce *coalesce,
+			     struct kernel_ethtool_coalesce *kernel_coal,
+			     struct netlink_ext_ack *extack)
 {
 	struct bnad *bnad = netdev_priv(netdev);
 	unsigned long flags;
@@ -378,7 +373,7 @@ bnad_set_coalesce(struct net_device *netdev, struct ethtool_coalesce *coalesce)
 			}
 			spin_unlock_irqrestore(&bnad->bna_lock, flags);
 			if (to_del)
-				del_timer_sync(&bnad->dim_timer);
+				timer_delete_sync(&bnad->dim_timer);
 			spin_lock_irqsave(&bnad->bna_lock, flags);
 			bnad_rx_coalescing_timeo_set(bnad);
 		}
@@ -410,7 +405,9 @@ bnad_set_coalesce(struct net_device *netdev, struct ethtool_coalesce *coalesce)
 
 static void
 bnad_get_ringparam(struct net_device *netdev,
-		   struct ethtool_ringparam *ringparam)
+		   struct ethtool_ringparam *ringparam,
+		   struct kernel_ethtool_ringparam *kernel_ringparam,
+		   struct netlink_ext_ack *extack)
 {
 	struct bnad *bnad = netdev_priv(netdev);
 
@@ -423,7 +420,9 @@ bnad_get_ringparam(struct net_device *netdev,
 
 static int
 bnad_set_ringparam(struct net_device *netdev,
-		   struct ethtool_ringparam *ringparam)
+		   struct ethtool_ringparam *ringparam,
+		   struct kernel_ethtool_ringparam *kernel_ringparam,
+		   struct netlink_ext_ack *extack)
 {
 	int i, current_err, err = 0;
 	struct bnad *bnad = netdev_priv(netdev);
@@ -533,6 +532,68 @@ bnad_set_pauseparam(struct net_device *netdev,
 	return 0;
 }
 
+static void bnad_get_txf_strings(u8 **string, int f_num)
+{
+	ethtool_sprintf(string, "txf%d_ucast_octets", f_num);
+	ethtool_sprintf(string, "txf%d_ucast", f_num);
+	ethtool_sprintf(string, "txf%d_ucast_vlan", f_num);
+	ethtool_sprintf(string, "txf%d_mcast_octets", f_num);
+	ethtool_sprintf(string, "txf%d_mcast", f_num);
+	ethtool_sprintf(string, "txf%d_mcast_vlan", f_num);
+	ethtool_sprintf(string, "txf%d_bcast_octets", f_num);
+	ethtool_sprintf(string, "txf%d_bcast", f_num);
+	ethtool_sprintf(string, "txf%d_bcast_vlan", f_num);
+	ethtool_sprintf(string, "txf%d_errors", f_num);
+	ethtool_sprintf(string, "txf%d_filter_vlan", f_num);
+	ethtool_sprintf(string, "txf%d_filter_mac_sa", f_num);
+}
+
+static void bnad_get_rxf_strings(u8 **string, int f_num)
+{
+	ethtool_sprintf(string, "rxf%d_ucast_octets", f_num);
+	ethtool_sprintf(string, "rxf%d_ucast", f_num);
+	ethtool_sprintf(string, "rxf%d_ucast_vlan", f_num);
+	ethtool_sprintf(string, "rxf%d_mcast_octets", f_num);
+	ethtool_sprintf(string, "rxf%d_mcast", f_num);
+	ethtool_sprintf(string, "rxf%d_mcast_vlan", f_num);
+	ethtool_sprintf(string, "rxf%d_bcast_octets", f_num);
+	ethtool_sprintf(string, "rxf%d_bcast", f_num);
+	ethtool_sprintf(string, "rxf%d_bcast_vlan", f_num);
+	ethtool_sprintf(string, "rxf%d_frame_drops", f_num);
+}
+
+static void bnad_get_cq_strings(u8 **string, int q_num)
+{
+	ethtool_sprintf(string, "cq%d_producer_index", q_num);
+	ethtool_sprintf(string, "cq%d_consumer_index", q_num);
+	ethtool_sprintf(string, "cq%d_hw_producer_index", q_num);
+	ethtool_sprintf(string, "cq%d_intr", q_num);
+	ethtool_sprintf(string, "cq%d_poll", q_num);
+	ethtool_sprintf(string, "cq%d_schedule", q_num);
+	ethtool_sprintf(string, "cq%d_keep_poll", q_num);
+	ethtool_sprintf(string, "cq%d_complete", q_num);
+}
+
+static void bnad_get_rxq_strings(u8 **string, int q_num)
+{
+	ethtool_sprintf(string, "rxq%d_packets", q_num);
+	ethtool_sprintf(string, "rxq%d_bytes", q_num);
+	ethtool_sprintf(string, "rxq%d_packets_with_error", q_num);
+	ethtool_sprintf(string, "rxq%d_allocbuf_failed", q_num);
+	ethtool_sprintf(string, "rxq%d_mapbuf_failed", q_num);
+	ethtool_sprintf(string, "rxq%d_producer_index", q_num);
+	ethtool_sprintf(string, "rxq%d_consumer_index", q_num);
+}
+
+static void bnad_get_txq_strings(u8 **string, int q_num)
+{
+	ethtool_sprintf(string, "txq%d_packets", q_num);
+	ethtool_sprintf(string, "txq%d_bytes", q_num);
+	ethtool_sprintf(string, "txq%d_producer_index", q_num);
+	ethtool_sprintf(string, "txq%d_consumer_index", q_num);
+	ethtool_sprintf(string, "txq%d_hw_consumer_index", q_num);
+}
+
 static void
 bnad_get_strings(struct net_device *netdev, u32 stringset, u8 *string)
 {
@@ -540,175 +601,57 @@ bnad_get_strings(struct net_device *netdev, u32 stringset, u8 *string)
 	int i, j, q_num;
 	u32 bmap;
 
+	if (stringset != ETH_SS_STATS)
+		return;
+
 	mutex_lock(&bnad->conf_mutex);
 
-	switch (stringset) {
-	case ETH_SS_STATS:
-		for (i = 0; i < BNAD_ETHTOOL_STATS_NUM; i++) {
-			BUG_ON(!(strlen(bnad_net_stats_strings[i]) <
-				   ETH_GSTRING_LEN));
-			strncpy(string, bnad_net_stats_strings[i],
-				ETH_GSTRING_LEN);
-			string += ETH_GSTRING_LEN;
-		}
-		bmap = bna_tx_rid_mask(&bnad->bna);
-		for (i = 0; bmap; i++) {
-			if (bmap & 1) {
-				sprintf(string, "txf%d_ucast_octets", i);
-				string += ETH_GSTRING_LEN;
-				sprintf(string, "txf%d_ucast", i);
-				string += ETH_GSTRING_LEN;
-				sprintf(string, "txf%d_ucast_vlan", i);
-				string += ETH_GSTRING_LEN;
-				sprintf(string, "txf%d_mcast_octets", i);
-				string += ETH_GSTRING_LEN;
-				sprintf(string, "txf%d_mcast", i);
-				string += ETH_GSTRING_LEN;
-				sprintf(string, "txf%d_mcast_vlan", i);
-				string += ETH_GSTRING_LEN;
-				sprintf(string, "txf%d_bcast_octets", i);
-				string += ETH_GSTRING_LEN;
-				sprintf(string, "txf%d_bcast", i);
-				string += ETH_GSTRING_LEN;
-				sprintf(string, "txf%d_bcast_vlan", i);
-				string += ETH_GSTRING_LEN;
-				sprintf(string, "txf%d_errors", i);
-				string += ETH_GSTRING_LEN;
-				sprintf(string, "txf%d_filter_vlan", i);
-				string += ETH_GSTRING_LEN;
-				sprintf(string, "txf%d_filter_mac_sa", i);
-				string += ETH_GSTRING_LEN;
-			}
-			bmap >>= 1;
-		}
+	for (i = 0; i < BNAD_ETHTOOL_STATS_NUM; i++) {
+		BUG_ON(!(strlen(bnad_net_stats_strings[i]) < ETH_GSTRING_LEN));
+		ethtool_puts(&string, bnad_net_stats_strings[i]);
+	}
 
-		bmap = bna_rx_rid_mask(&bnad->bna);
-		for (i = 0; bmap; i++) {
-			if (bmap & 1) {
-				sprintf(string, "rxf%d_ucast_octets", i);
-				string += ETH_GSTRING_LEN;
-				sprintf(string, "rxf%d_ucast", i);
-				string += ETH_GSTRING_LEN;
-				sprintf(string, "rxf%d_ucast_vlan", i);
-				string += ETH_GSTRING_LEN;
-				sprintf(string, "rxf%d_mcast_octets", i);
-				string += ETH_GSTRING_LEN;
-				sprintf(string, "rxf%d_mcast", i);
-				string += ETH_GSTRING_LEN;
-				sprintf(string, "rxf%d_mcast_vlan", i);
-				string += ETH_GSTRING_LEN;
-				sprintf(string, "rxf%d_bcast_octets", i);
-				string += ETH_GSTRING_LEN;
-				sprintf(string, "rxf%d_bcast", i);
-				string += ETH_GSTRING_LEN;
-				sprintf(string, "rxf%d_bcast_vlan", i);
-				string += ETH_GSTRING_LEN;
-				sprintf(string, "rxf%d_frame_drops", i);
-				string += ETH_GSTRING_LEN;
-			}
-			bmap >>= 1;
-		}
+	bmap = bna_tx_rid_mask(&bnad->bna);
+	for (i = 0; bmap; i++) {
+		if (bmap & 1)
+			bnad_get_txf_strings(&string, i);
+		bmap >>= 1;
+	}
 
-		q_num = 0;
-		for (i = 0; i < bnad->num_rx; i++) {
-			if (!bnad->rx_info[i].rx)
-				continue;
-			for (j = 0; j < bnad->num_rxp_per_rx; j++) {
-				sprintf(string, "cq%d_producer_index", q_num);
-				string += ETH_GSTRING_LEN;
-				sprintf(string, "cq%d_consumer_index", q_num);
-				string += ETH_GSTRING_LEN;
-				sprintf(string, "cq%d_hw_producer_index",
-					q_num);
-				string += ETH_GSTRING_LEN;
-				sprintf(string, "cq%d_intr", q_num);
-				string += ETH_GSTRING_LEN;
-				sprintf(string, "cq%d_poll", q_num);
-				string += ETH_GSTRING_LEN;
-				sprintf(string, "cq%d_schedule", q_num);
-				string += ETH_GSTRING_LEN;
-				sprintf(string, "cq%d_keep_poll", q_num);
-				string += ETH_GSTRING_LEN;
-				sprintf(string, "cq%d_complete", q_num);
-				string += ETH_GSTRING_LEN;
-				q_num++;
-			}
-		}
+	bmap = bna_rx_rid_mask(&bnad->bna);
+	for (i = 0; bmap; i++, bmap >>= 1) {
+		if (bmap & 1)
+			bnad_get_rxf_strings(&string, i);
+		bmap >>= 1;
+	}
 
-		q_num = 0;
-		for (i = 0; i < bnad->num_rx; i++) {
-			if (!bnad->rx_info[i].rx)
-				continue;
-			for (j = 0; j < bnad->num_rxp_per_rx; j++) {
-				sprintf(string, "rxq%d_packets", q_num);
-				string += ETH_GSTRING_LEN;
-				sprintf(string, "rxq%d_bytes", q_num);
-				string += ETH_GSTRING_LEN;
-				sprintf(string, "rxq%d_packets_with_error",
-								q_num);
-				string += ETH_GSTRING_LEN;
-				sprintf(string, "rxq%d_allocbuf_failed", q_num);
-				string += ETH_GSTRING_LEN;
-				sprintf(string, "rxq%d_mapbuf_failed", q_num);
-				string += ETH_GSTRING_LEN;
-				sprintf(string, "rxq%d_producer_index", q_num);
-				string += ETH_GSTRING_LEN;
-				sprintf(string, "rxq%d_consumer_index", q_num);
-				string += ETH_GSTRING_LEN;
-				q_num++;
-				if (bnad->rx_info[i].rx_ctrl[j].ccb &&
-					bnad->rx_info[i].rx_ctrl[j].ccb->
-					rcb[1] &&
-					bnad->rx_info[i].rx_ctrl[j].ccb->
-					rcb[1]->rxq) {
-					sprintf(string, "rxq%d_packets", q_num);
-					string += ETH_GSTRING_LEN;
-					sprintf(string, "rxq%d_bytes", q_num);
-					string += ETH_GSTRING_LEN;
-					sprintf(string,
-					"rxq%d_packets_with_error", q_num);
-					string += ETH_GSTRING_LEN;
-					sprintf(string, "rxq%d_allocbuf_failed",
-								q_num);
-					string += ETH_GSTRING_LEN;
-					sprintf(string, "rxq%d_mapbuf_failed",
-						q_num);
-					string += ETH_GSTRING_LEN;
-					sprintf(string, "rxq%d_producer_index",
-								q_num);
-					string += ETH_GSTRING_LEN;
-					sprintf(string, "rxq%d_consumer_index",
-								q_num);
-					string += ETH_GSTRING_LEN;
-					q_num++;
-				}
-			}
-		}
+	q_num = 0;
+	for (i = 0; i < bnad->num_rx; i++) {
+		if (!bnad->rx_info[i].rx)
+			continue;
+		for (j = 0; j < bnad->num_rxp_per_rx; j++)
+			bnad_get_cq_strings(&string, q_num++);
+	}
 
-		q_num = 0;
-		for (i = 0; i < bnad->num_tx; i++) {
-			if (!bnad->tx_info[i].tx)
-				continue;
-			for (j = 0; j < bnad->num_txq_per_tx; j++) {
-				sprintf(string, "txq%d_packets", q_num);
-				string += ETH_GSTRING_LEN;
-				sprintf(string, "txq%d_bytes", q_num);
-				string += ETH_GSTRING_LEN;
-				sprintf(string, "txq%d_producer_index", q_num);
-				string += ETH_GSTRING_LEN;
-				sprintf(string, "txq%d_consumer_index", q_num);
-				string += ETH_GSTRING_LEN;
-				sprintf(string, "txq%d_hw_consumer_index",
-									q_num);
-				string += ETH_GSTRING_LEN;
-				q_num++;
-			}
+	q_num = 0;
+	for (i = 0; i < bnad->num_rx; i++) {
+		if (!bnad->rx_info[i].rx)
+			continue;
+		for (j = 0; j < bnad->num_rxp_per_rx; j++) {
+			bnad_get_rxq_strings(&string, q_num++);
+			if (bnad->rx_info[i].rx_ctrl[j].ccb &&
+			    bnad->rx_info[i].rx_ctrl[j].ccb->rcb[1] &&
+			    bnad->rx_info[i].rx_ctrl[j].ccb->rcb[1]->rxq)
+				bnad_get_rxq_strings(&string, q_num++);
 		}
+	}
 
-		break;
-
-	default:
-		break;
+	q_num = 0;
+	for (i = 0; i < bnad->num_tx; i++) {
+		if (!bnad->tx_info[i].tx)
+			continue;
+		for (j = 0; j < bnad->num_txq_per_tx; j++)
+			bnad_get_txq_strings(&string, q_num++);
 	}
 
 	mutex_unlock(&bnad->conf_mutex);
@@ -1124,6 +1067,9 @@ out:
 }
 
 static const struct ethtool_ops bnad_ethtool_ops = {
+	.supported_coalesce_params = ETHTOOL_COALESCE_USECS |
+				     ETHTOOL_COALESCE_TX_MAX_FRAMES |
+				     ETHTOOL_COALESCE_USE_ADAPTIVE_RX,
 	.get_drvinfo = bnad_get_drvinfo,
 	.get_wol = bnad_get_wol,
 	.get_link = ethtool_op_get_link,
diff --git a/drivers/net/ethernet/brocade/bna/cna.h b/drivers/net/ethernet/brocade/bna/cna.h
index 75f8f1ac9fb7..28d89d0c2297 100644
--- a/drivers/net/ethernet/brocade/bna/cna.h
+++ b/drivers/net/ethernet/brocade/bna/cna.h
@@ -1,14 +1,6 @@
+/* SPDX-License-Identifier: GPL-2.0-only */
 /*
  * Linux network driver for QLogic BR-series Converged Network Adapter.
- *
- * This program is free software; you can redistribute it and/or modify it
- * under the terms of the GNU General Public License (GPL) Version 2 as
- * published by the Free Software Foundation
- *
- * This program is distributed in the hope that it will be useful, but
- * WITHOUT ANY WARRANTY; without even the implied warranty of
- * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU
- * General Public License for more details.
  */
 /*
  * Copyright (c) 2006-2014 Brocade Communications Systems, Inc.
diff --git a/drivers/net/ethernet/brocade/bna/cna_fwimg.c b/drivers/net/ethernet/brocade/bna/cna_fwimg.c
index 2e7fb97883dc..824eaef30704 100644
--- a/drivers/net/ethernet/brocade/bna/cna_fwimg.c
+++ b/drivers/net/ethernet/brocade/bna/cna_fwimg.c
@@ -1,14 +1,6 @@
+// SPDX-License-Identifier: GPL-2.0-only
 /*
  * Linux network driver for QLogic BR-series Converged Network Adapter.
- *
- * This program is free software; you can redistribute it and/or modify it
- * under the terms of the GNU General Public License (GPL) Version 2 as
- * published by the Free Software Foundation
- *
- * This program is distributed in the hope that it will be useful, but
- * WITHOUT ANY WARRANTY; without even the implied warranty of
- * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU
- * General Public License for more details.
  */
 /*
  * Copyright (c) 2005-2014 Brocade Communications Systems, Inc.
diff --git a/drivers/net/ethernet/cadence/Kconfig b/drivers/net/ethernet/cadence/Kconfig
index 608bea171956..5b2a461dfd28 100644
--- a/drivers/net/ethernet/cadence/Kconfig
+++ b/drivers/net/ethernet/cadence/Kconfig
@@ -1,43 +1,54 @@
+# SPDX-License-Identifier: GPL-2.0-only
 #
-# Atmel device configuration
+# Cadence device configuration
 #
 
-config NET_CADENCE
+config NET_VENDOR_CADENCE
 	bool "Cadence devices"
 	depends on HAS_IOMEM
 	default y
-	---help---
+	help
 	  If you have a network (Ethernet) card belonging to this class, say Y.
 
 	  If unsure, say Y.
 
 	  Note that the answer to this question doesn't directly affect the
-	  kernel: saying N will just cause the configurator to skip all
-	  the remaining Atmel network card questions. If you say Y, you will be
+	  kernel: saying N will just cause the configurator to skip all the
+	  remaining Cadence network card questions. If you say Y, you will be
 	  asked for your specific card in the following questions.
 
-if NET_CADENCE
+if NET_VENDOR_CADENCE
 
 config MACB
 	tristate "Cadence MACB/GEM support"
-	depends on HAS_DMA
-	select PHYLIB
-	---help---
+	depends on HAS_DMA && COMMON_CLK
+	depends on PTP_1588_CLOCK_OPTIONAL
+	select PHYLINK
+	select CRC32
+	help
 	  The Cadence MACB ethernet interface is found on many Atmel AT32 and
 	  AT91 parts.  This driver also supports the Cadence GEM (Gigabit
 	  Ethernet MAC found in some ARM SoC devices).  Say Y to include
 	  support for the MACB/GEM chip.
 
 	  To compile this driver as a module, choose M here: the module
-	  will be called macb.
+	  will be macb.
+
+config MACB_USE_HWSTAMP
+	bool "Use IEEE 1588 hwstamp"
+	depends on MACB
+	depends on PTP_1588_CLOCK
+	default y
+	help
+	  Enable IEEE 1588 Precision Time Protocol (PTP) support for MACB.
 
 config MACB_PCI
 	tristate "Cadence PCI MACB/GEM support"
-	depends on MACB && PCI && COMMON_CLK
-	---help---
+	depends on MACB && PCI
+	help
 	  This is PCI wrapper for MACB driver.
 
 	  To compile this driver as a module, choose M here: the module
 	  will be called macb_pci.
 
-endif # NET_CADENCE
+endif # NET_VENDOR_CADENCE
diff --git a/drivers/net/ethernet/cadence/Makefile b/drivers/net/ethernet/cadence/Makefile
index 4ba75594d5c5..1f33cdca9a3c 100644
--- a/drivers/net/ethernet/cadence/Makefile
+++ b/drivers/net/ethernet/cadence/Makefile
@@ -1,6 +1,12 @@
+# SPDX-License-Identifier: GPL-2.0
 #
 # Makefile for the Atmel network device drivers.
 #
+macb-y	:= macb_main.o
+
+ifeq ($(CONFIG_MACB_USE_HWSTAMP),y)
+macb-y	+= macb_ptp.o
+endif
 
 obj-$(CONFIG_MACB) += macb.o
 obj-$(CONFIG_MACB_PCI) += macb_pci.o
diff --git a/drivers/net/ethernet/cadence/macb.c b/drivers/net/ethernet/cadence/macb.c
deleted file mode 100644
index 91f7492623d3..000000000000
--- a/drivers/net/ethernet/cadence/macb.c
+++ /dev/null
@@ -1,3509 +0,0 @@
-/*
- * Cadence MACB/GEM Ethernet Controller driver
- *
- * Copyright (C) 2004-2006 Atmel Corporation
- *
- * This program is free software; you can redistribute it and/or modify
- * it under the terms of the GNU General Public License version 2 as
- * published by the Free Software Foundation.
- */
-
-#define pr_fmt(fmt) KBUILD_MODNAME ": " fmt
-#include <linux/clk.h>
-#include <linux/module.h>
-#include <linux/moduleparam.h>
-#include <linux/kernel.h>
-#include <linux/types.h>
-#include <linux/circ_buf.h>
-#include <linux/slab.h>
-#include <linux/init.h>
-#include <linux/io.h>
-#include <linux/gpio.h>
-#include <linux/gpio/consumer.h>
-#include <linux/interrupt.h>
-#include <linux/netdevice.h>
-#include <linux/etherdevice.h>
-#include <linux/dma-mapping.h>
-#include <linux/platform_data/macb.h>
-#include <linux/platform_device.h>
-#include <linux/phy.h>
-#include <linux/of.h>
-#include <linux/of_device.h>
-#include <linux/of_gpio.h>
-#include <linux/of_mdio.h>
-#include <linux/of_net.h>
-#include <linux/ip.h>
-#include <linux/udp.h>
-#include <linux/tcp.h>
-#include "macb.h"
-
-#define MACB_RX_BUFFER_SIZE	128
-#define RX_BUFFER_MULTIPLE	64  /* bytes */
-
-#define DEFAULT_RX_RING_SIZE	512 /* must be power of 2 */
-#define MIN_RX_RING_SIZE	64
-#define MAX_RX_RING_SIZE	8192
-#define RX_RING_BYTES(bp)	(macb_dma_desc_get_size(bp)	\
-				 * (bp)->rx_ring_size)
-
-#define DEFAULT_TX_RING_SIZE	512 /* must be power of 2 */
-#define MIN_TX_RING_SIZE	64
-#define MAX_TX_RING_SIZE	4096
-#define TX_RING_BYTES(bp)	(macb_dma_desc_get_size(bp)	\
-				 * (bp)->tx_ring_size)
-
-/* level of occupied TX descriptors under which we wake up TX process */
-#define MACB_TX_WAKEUP_THRESH(bp)	(3 * (bp)->tx_ring_size / 4)
-
-#define MACB_RX_INT_FLAGS	(MACB_BIT(RCOMP) | MACB_BIT(RXUBR)	\
-				 | MACB_BIT(ISR_ROVR))
-#define MACB_TX_ERR_FLAGS	(MACB_BIT(ISR_TUND)			\
-					| MACB_BIT(ISR_RLE)		\
-					| MACB_BIT(TXERR))
-#define MACB_TX_INT_FLAGS	(MACB_TX_ERR_FLAGS | MACB_BIT(TCOMP))
-
-/* Max length of transmit frame must be a multiple of 8 bytes */
-#define MACB_TX_LEN_ALIGN	8
-#define MACB_MAX_TX_LEN		((unsigned int)((1 << MACB_TX_FRMLEN_SIZE) - 1) & ~((unsigned int)(MACB_TX_LEN_ALIGN - 1)))
-#define GEM_MAX_TX_LEN		((unsigned int)((1 << GEM_TX_FRMLEN_SIZE) - 1) & ~((unsigned int)(MACB_TX_LEN_ALIGN - 1)))
-
-#define GEM_MTU_MIN_SIZE	ETH_MIN_MTU
-#define MACB_NETIF_LSO		(NETIF_F_TSO | NETIF_F_UFO)
-
-#define MACB_WOL_HAS_MAGIC_PACKET	(0x1 << 0)
-#define MACB_WOL_ENABLED		(0x1 << 1)
-
-/* Graceful stop timeouts in us. We should allow up to
- * 1 frame time (10 Mbits/s, full-duplex, ignoring collisions)
- */
-#define MACB_HALT_TIMEOUT	1230
-
-/* DMA buffer descriptor might be different size
- * depends on hardware configuration.
- */
-static unsigned int macb_dma_desc_get_size(struct macb *bp)
-{
-#ifdef CONFIG_ARCH_DMA_ADDR_T_64BIT
-	if (bp->hw_dma_cap == HW_DMA_CAP_64B)
-		return sizeof(struct macb_dma_desc) + sizeof(struct macb_dma_desc_64);
-#endif
-	return sizeof(struct macb_dma_desc);
-}
-
-static unsigned int macb_adj_dma_desc_idx(struct macb *bp, unsigned int idx)
-{
-#ifdef CONFIG_ARCH_DMA_ADDR_T_64BIT
-	/* Dma buffer descriptor is 4 words length (instead of 2 words)
-	 * for 64b GEM.
-	 */
-	if (bp->hw_dma_cap == HW_DMA_CAP_64B)
-		idx <<= 1;
-#endif
-	return idx;
-}
-
-#ifdef CONFIG_ARCH_DMA_ADDR_T_64BIT
-static struct macb_dma_desc_64 *macb_64b_desc(struct macb *bp, struct macb_dma_desc *desc)
-{
-	return (struct macb_dma_desc_64 *)((void *)desc + sizeof(struct macb_dma_desc));
-}
-#endif
-
-/* Ring buffer accessors */
-static unsigned int macb_tx_ring_wrap(struct macb *bp, unsigned int index)
-{
-	return index & (bp->tx_ring_size - 1);
-}
-
-static struct macb_dma_desc *macb_tx_desc(struct macb_queue *queue,
-					  unsigned int index)
-{
-	index = macb_tx_ring_wrap(queue->bp, index);
-	index = macb_adj_dma_desc_idx(queue->bp, index);
-	return &queue->tx_ring[index];
-}
-
-static struct macb_tx_skb *macb_tx_skb(struct macb_queue *queue,
-				       unsigned int index)
-{
-	return &queue->tx_skb[macb_tx_ring_wrap(queue->bp, index)];
-}
-
-static dma_addr_t macb_tx_dma(struct macb_queue *queue, unsigned int index)
-{
-	dma_addr_t offset;
-
-	offset = macb_tx_ring_wrap(queue->bp, index) *
-			macb_dma_desc_get_size(queue->bp);
-
-	return queue->tx_ring_dma + offset;
-}
-
-static unsigned int macb_rx_ring_wrap(struct macb *bp, unsigned int index)
-{
-	return index & (bp->rx_ring_size - 1);
-}
-
-static struct macb_dma_desc *macb_rx_desc(struct macb *bp, unsigned int index)
-{
-	index = macb_rx_ring_wrap(bp, index);
-	index = macb_adj_dma_desc_idx(bp, index);
-	return &bp->rx_ring[index];
-}
-
-static void *macb_rx_buffer(struct macb *bp, unsigned int index)
-{
-	return bp->rx_buffers + bp->rx_buffer_size *
-	       macb_rx_ring_wrap(bp, index);
-}
-
-/* I/O accessors */
-static u32 hw_readl_native(struct macb *bp, int offset)
-{
-	return __raw_readl(bp->regs + offset);
-}
-
-static void hw_writel_native(struct macb *bp, int offset, u32 value)
-{
-	__raw_writel(value, bp->regs + offset);
-}
-
-static u32 hw_readl(struct macb *bp, int offset)
-{
-	return readl_relaxed(bp->regs + offset);
-}
-
-static void hw_writel(struct macb *bp, int offset, u32 value)
-{
-	writel_relaxed(value, bp->regs + offset);
-}
-
-/* Find the CPU endianness by using the loopback bit of NCR register. When the
- * CPU is in big endian we need to program swapped mode for management
- * descriptor access.
- */
-static bool hw_is_native_io(void __iomem *addr)
-{
-	u32 value = MACB_BIT(LLB);
-
-	__raw_writel(value, addr + MACB_NCR);
-	value = __raw_readl(addr + MACB_NCR);
-
-	/* Write 0 back to disable everything */
-	__raw_writel(0, addr + MACB_NCR);
-
-	return value == MACB_BIT(LLB);
-}
-
-static bool hw_is_gem(void __iomem *addr, bool native_io)
-{
-	u32 id;
-
-	if (native_io)
-		id = __raw_readl(addr + MACB_MID);
-	else
-		id = readl_relaxed(addr + MACB_MID);
-
-	return MACB_BFEXT(IDNUM, id) >= 0x2;
-}
-
-static void macb_set_hwaddr(struct macb *bp)
-{
-	u32 bottom;
-	u16 top;
-
-	bottom = cpu_to_le32(*((u32 *)bp->dev->dev_addr));
-	macb_or_gem_writel(bp, SA1B, bottom);
-	top = cpu_to_le16(*((u16 *)(bp->dev->dev_addr + 4)));
-	macb_or_gem_writel(bp, SA1T, top);
-
-	/* Clear unused address register sets */
-	macb_or_gem_writel(bp, SA2B, 0);
-	macb_or_gem_writel(bp, SA2T, 0);
-	macb_or_gem_writel(bp, SA3B, 0);
-	macb_or_gem_writel(bp, SA3T, 0);
-	macb_or_gem_writel(bp, SA4B, 0);
-	macb_or_gem_writel(bp, SA4T, 0);
-}
-
-static void macb_get_hwaddr(struct macb *bp)
-{
-	struct macb_platform_data *pdata;
-	u32 bottom;
-	u16 top;
-	u8 addr[6];
-	int i;
-
-	pdata = dev_get_platdata(&bp->pdev->dev);
-
-	/* Check all 4 address register for valid address */
-	for (i = 0; i < 4; i++) {
-		bottom = macb_or_gem_readl(bp, SA1B + i * 8);
-		top = macb_or_gem_readl(bp, SA1T + i * 8);
-
-		if (pdata && pdata->rev_eth_addr) {
-			addr[5] = bottom & 0xff;
-			addr[4] = (bottom >> 8) & 0xff;
-			addr[3] = (bottom >> 16) & 0xff;
-			addr[2] = (bottom >> 24) & 0xff;
-			addr[1] = top & 0xff;
-			addr[0] = (top & 0xff00) >> 8;
-		} else {
-			addr[0] = bottom & 0xff;
-			addr[1] = (bottom >> 8) & 0xff;
-			addr[2] = (bottom >> 16) & 0xff;
-			addr[3] = (bottom >> 24) & 0xff;
-			addr[4] = top & 0xff;
-			addr[5] = (top >> 8) & 0xff;
-		}
-
-		if (is_valid_ether_addr(addr)) {
-			memcpy(bp->dev->dev_addr, addr, sizeof(addr));
-			return;
-		}
-	}
-
-	dev_info(&bp->pdev->dev, "invalid hw address, using random\n");
-	eth_hw_addr_random(bp->dev);
-}
-
-static int macb_mdio_read(struct mii_bus *bus, int mii_id, int regnum)
-{
-	struct macb *bp = bus->priv;
-	int value;
-
-	macb_writel(bp, MAN, (MACB_BF(SOF, MACB_MAN_SOF)
-			      | MACB_BF(RW, MACB_MAN_READ)
-			      | MACB_BF(PHYA, mii_id)
-			      | MACB_BF(REGA, regnum)
-			      | MACB_BF(CODE, MACB_MAN_CODE)));
-
-	/* wait for end of transfer */
-	while (!MACB_BFEXT(IDLE, macb_readl(bp, NSR)))
-		cpu_relax();
-
-	value = MACB_BFEXT(DATA, macb_readl(bp, MAN));
-
-	return value;
-}
-
-static int macb_mdio_write(struct mii_bus *bus, int mii_id, int regnum,
-			   u16 value)
-{
-	struct macb *bp = bus->priv;
-
-	macb_writel(bp, MAN, (MACB_BF(SOF, MACB_MAN_SOF)
-			      | MACB_BF(RW, MACB_MAN_WRITE)
-			      | MACB_BF(PHYA, mii_id)
-			      | MACB_BF(REGA, regnum)
-			      | MACB_BF(CODE, MACB_MAN_CODE)
-			      | MACB_BF(DATA, value)));
-
-	/* wait for end of transfer */
-	while (!MACB_BFEXT(IDLE, macb_readl(bp, NSR)))
-		cpu_relax();
-
-	return 0;
-}
-
-/**
- * macb_set_tx_clk() - Set a clock to a new frequency
- * @clk		Pointer to the clock to change
- * @rate	New frequency in Hz
- * @dev		Pointer to the struct net_device
- */
-static void macb_set_tx_clk(struct clk *clk, int speed, struct net_device *dev)
-{
-	long ferr, rate, rate_rounded;
-
-	if (!clk)
-		return;
-
-	switch (speed) {
-	case SPEED_10:
-		rate = 2500000;
-		break;
-	case SPEED_100:
-		rate = 25000000;
-		break;
-	case SPEED_1000:
-		rate = 125000000;
-		break;
-	default:
-		return;
-	}
-
-	rate_rounded = clk_round_rate(clk, rate);
-	if (rate_rounded < 0)
-		return;
-
-	/* RGMII allows 50 ppm frequency error. Test and warn if this limit
-	 * is not satisfied.
-	 */
-	ferr = abs(rate_rounded - rate);
-	ferr = DIV_ROUND_UP(ferr, rate / 100000);
-	if (ferr > 5)
-		netdev_warn(dev, "unable to generate target frequency: %ld Hz\n",
-			    rate);
-
-	if (clk_set_rate(clk, rate_rounded))
-		netdev_err(dev, "adjusting tx_clk failed.\n");
-}
-
-static void macb_handle_link_change(struct net_device *dev)
-{
-	struct macb *bp = netdev_priv(dev);
-	struct phy_device *phydev = dev->phydev;
-	unsigned long flags;
-	int status_change = 0;
-
-	spin_lock_irqsave(&bp->lock, flags);
-
-	if (phydev->link) {
-		if ((bp->speed != phydev->speed) ||
-		    (bp->duplex != phydev->duplex)) {
-			u32 reg;
-
-			reg = macb_readl(bp, NCFGR);
-			reg &= ~(MACB_BIT(SPD) | MACB_BIT(FD));
-			if (macb_is_gem(bp))
-				reg &= ~GEM_BIT(GBE);
-
-			if (phydev->duplex)
-				reg |= MACB_BIT(FD);
-			if (phydev->speed == SPEED_100)
-				reg |= MACB_BIT(SPD);
-			if (phydev->speed == SPEED_1000 &&
-			    bp->caps & MACB_CAPS_GIGABIT_MODE_AVAILABLE)
-				reg |= GEM_BIT(GBE);
-
-			macb_or_gem_writel(bp, NCFGR, reg);
-
-			bp->speed = phydev->speed;
-			bp->duplex = phydev->duplex;
-			status_change = 1;
-		}
-	}
-
-	if (phydev->link != bp->link) {
-		if (!phydev->link) {
-			bp->speed = 0;
-			bp->duplex = -1;
-		}
-		bp->link = phydev->link;
-
-		status_change = 1;
-	}
-
-	spin_unlock_irqrestore(&bp->lock, flags);
-
-	if (status_change) {
-		if (phydev->link) {
-			/* Update the TX clock rate if and only if the link is
-			 * up and there has been a link change.
-			 */
-			macb_set_tx_clk(bp->tx_clk, phydev->speed, dev);
-
-			netif_carrier_on(dev);
-			netdev_info(dev, "link up (%d/%s)\n",
-				    phydev->speed,
-				    phydev->duplex == DUPLEX_FULL ?
-				    "Full" : "Half");
-		} else {
-			netif_carrier_off(dev);
-			netdev_info(dev, "link down\n");
-		}
-	}
-}
-
-/* based on au1000_eth. c*/
-static int macb_mii_probe(struct net_device *dev)
-{
-	struct macb *bp = netdev_priv(dev);
-	struct macb_platform_data *pdata;
-	struct phy_device *phydev;
-	int phy_irq;
-	int ret;
-
-	phydev = phy_find_first(bp->mii_bus);
-	if (!phydev) {
-		netdev_err(dev, "no PHY found\n");
-		return -ENXIO;
-	}
-
-	pdata = dev_get_platdata(&bp->pdev->dev);
-	if (pdata) {
-		if (gpio_is_valid(pdata->phy_irq_pin)) {
-			ret = devm_gpio_request(&bp->pdev->dev,
-						pdata->phy_irq_pin, "phy int");
-			if (!ret) {
-				phy_irq = gpio_to_irq(pdata->phy_irq_pin);
-				phydev->irq = (phy_irq < 0) ? PHY_POLL : phy_irq;
-			}
-		} else {
-			phydev->irq = PHY_POLL;
-		}
-	}
-
-	/* attach the mac to the phy */
-	ret = phy_connect_direct(dev, phydev, &macb_handle_link_change,
-				 bp->phy_interface);
-	if (ret) {
-		netdev_err(dev, "Could not attach to PHY\n");
-		return ret;
-	}
-
-	/* mask with MAC supported features */
-	if (macb_is_gem(bp) && bp->caps & MACB_CAPS_GIGABIT_MODE_AVAILABLE)
-		phydev->supported &= PHY_GBIT_FEATURES;
-	else
-		phydev->supported &= PHY_BASIC_FEATURES;
-
-	if (bp->caps & MACB_CAPS_NO_GIGABIT_HALF)
-		phydev->supported &= ~SUPPORTED_1000baseT_Half;
-
-	phydev->advertising = phydev->supported;
-
-	bp->link = 0;
-	bp->speed = 0;
-	bp->duplex = -1;
-
-	return 0;
-}
-
-static int macb_mii_init(struct macb *bp)
-{
-	struct macb_platform_data *pdata;
-	struct device_node *np;
-	int err = -ENXIO, i;
-
-	/* Enable management port */
-	macb_writel(bp, NCR, MACB_BIT(MPE));
-
-	bp->mii_bus = mdiobus_alloc();
-	if (!bp->mii_bus) {
-		err = -ENOMEM;
-		goto err_out;
-	}
-
-	bp->mii_bus->name = "MACB_mii_bus";
-	bp->mii_bus->read = &macb_mdio_read;
-	bp->mii_bus->write = &macb_mdio_write;
-	snprintf(bp->mii_bus->id, MII_BUS_ID_SIZE, "%s-%x",
-		 bp->pdev->name, bp->pdev->id);
-	bp->mii_bus->priv = bp;
-	bp->mii_bus->parent = &bp->pdev->dev;
-	pdata = dev_get_platdata(&bp->pdev->dev);
-
-	dev_set_drvdata(&bp->dev->dev, bp->mii_bus);
-
-	np = bp->pdev->dev.of_node;
-	if (np) {
-		/* try dt phy registration */
-		err = of_mdiobus_register(bp->mii_bus, np);
-
-		/* fallback to standard phy registration if no phy were
-		 * found during dt phy registration
-		 */
-		if (!err && !phy_find_first(bp->mii_bus)) {
-			for (i = 0; i < PHY_MAX_ADDR; i++) {
-				struct phy_device *phydev;
-
-				phydev = mdiobus_scan(bp->mii_bus, i);
-				if (IS_ERR(phydev) &&
-				    PTR_ERR(phydev) != -ENODEV) {
-					err = PTR_ERR(phydev);
-					break;
-				}
-			}
-
-			if (err)
-				goto err_out_unregister_bus;
-		}
-	} else {
-		for (i = 0; i < PHY_MAX_ADDR; i++)
-			bp->mii_bus->irq[i] = PHY_POLL;
-
-		if (pdata)
-			bp->mii_bus->phy_mask = pdata->phy_mask;
-
-		err = mdiobus_register(bp->mii_bus);
-	}
-
-	if (err)
-		goto err_out_free_mdiobus;
-
-	err = macb_mii_probe(bp->dev);
-	if (err)
-		goto err_out_unregister_bus;
-
-	return 0;
-
-err_out_unregister_bus:
-	mdiobus_unregister(bp->mii_bus);
-err_out_free_mdiobus:
-	mdiobus_free(bp->mii_bus);
-err_out:
-	return err;
-}
-
-static void macb_update_stats(struct macb *bp)
-{
-	u32 *p = &bp->hw_stats.macb.rx_pause_frames;
-	u32 *end = &bp->hw_stats.macb.tx_pause_frames + 1;
-	int offset = MACB_PFR;
-
-	WARN_ON((unsigned long)(end - p - 1) != (MACB_TPF - MACB_PFR) / 4);
-
-	for (; p < end; p++, offset += 4)
-		*p += bp->macb_reg_readl(bp, offset);
-}
-
-static int macb_halt_tx(struct macb *bp)
-{
-	unsigned long	halt_time, timeout;
-	u32		status;
-
-	macb_writel(bp, NCR, macb_readl(bp, NCR) | MACB_BIT(THALT));
-
-	timeout = jiffies + usecs_to_jiffies(MACB_HALT_TIMEOUT);
-	do {
-		halt_time = jiffies;
-		status = macb_readl(bp, TSR);
-		if (!(status & MACB_BIT(TGO)))
-			return 0;
-
-		usleep_range(10, 250);
-	} while (time_before(halt_time, timeout));
-
-	return -ETIMEDOUT;
-}
-
-static void macb_tx_unmap(struct macb *bp, struct macb_tx_skb *tx_skb)
-{
-	if (tx_skb->mapping) {
-		if (tx_skb->mapped_as_page)
-			dma_unmap_page(&bp->pdev->dev, tx_skb->mapping,
-				       tx_skb->size, DMA_TO_DEVICE);
-		else
-			dma_unmap_single(&bp->pdev->dev, tx_skb->mapping,
-					 tx_skb->size, DMA_TO_DEVICE);
-		tx_skb->mapping = 0;
-	}
-
-	if (tx_skb->skb) {
-		dev_kfree_skb_any(tx_skb->skb);
-		tx_skb->skb = NULL;
-	}
-}
-
-static void macb_set_addr(struct macb *bp, struct macb_dma_desc *desc, dma_addr_t addr)
-{
-#ifdef CONFIG_ARCH_DMA_ADDR_T_64BIT
-	struct macb_dma_desc_64 *desc_64;
-
-	if (bp->hw_dma_cap == HW_DMA_CAP_64B) {
-		desc_64 = macb_64b_desc(bp, desc);
-		desc_64->addrh = upper_32_bits(addr);
-	}
-#endif
-	desc->addr = lower_32_bits(addr);
-}
-
-static dma_addr_t macb_get_addr(struct macb *bp, struct macb_dma_desc *desc)
-{
-	dma_addr_t addr = 0;
-#ifdef CONFIG_ARCH_DMA_ADDR_T_64BIT
-	struct macb_dma_desc_64 *desc_64;
-
-	if (bp->hw_dma_cap == HW_DMA_CAP_64B) {
-		desc_64 = macb_64b_desc(bp, desc);
-		addr = ((u64)(desc_64->addrh) << 32);
-	}
-#endif
-	addr |= MACB_BF(RX_WADDR, MACB_BFEXT(RX_WADDR, desc->addr));
-	return addr;
-}
-
-static void macb_tx_error_task(struct work_struct *work)
-{
-	struct macb_queue	*queue = container_of(work, struct macb_queue,
-						      tx_error_task);
-	struct macb		*bp = queue->bp;
-	struct macb_tx_skb	*tx_skb;
-	struct macb_dma_desc	*desc;
-	struct sk_buff		*skb;
-	unsigned int		tail;
-	unsigned long		flags;
-
-	netdev_vdbg(bp->dev, "macb_tx_error_task: q = %u, t = %u, h = %u\n",
-		    (unsigned int)(queue - bp->queues),
-		    queue->tx_tail, queue->tx_head);
-
-	/* Prevent the queue IRQ handlers from running: each of them may call
-	 * macb_tx_interrupt(), which in turn may call netif_wake_subqueue().
-	 * As explained below, we have to halt the transmission before updating
-	 * TBQP registers so we call netif_tx_stop_all_queues() to notify the
-	 * network engine about the macb/gem being halted.
-	 */
-	spin_lock_irqsave(&bp->lock, flags);
-
-	/* Make sure nobody is trying to queue up new packets */
-	netif_tx_stop_all_queues(bp->dev);
-
-	/* Stop transmission now
-	 * (in case we have just queued new packets)
-	 * macb/gem must be halted to write TBQP register
-	 */
-	if (macb_halt_tx(bp))
-		/* Just complain for now, reinitializing TX path can be good */
-		netdev_err(bp->dev, "BUG: halt tx timed out\n");
-
-	/* Treat frames in TX queue including the ones that caused the error.
-	 * Free transmit buffers in upper layer.
-	 */
-	for (tail = queue->tx_tail; tail != queue->tx_head; tail++) {
-		u32	ctrl;
-
-		desc = macb_tx_desc(queue, tail);
-		ctrl = desc->ctrl;
-		tx_skb = macb_tx_skb(queue, tail);
-		skb = tx_skb->skb;
-
-		if (ctrl & MACB_BIT(TX_USED)) {
-			/* skb is set for the last buffer of the frame */
-			while (!skb) {
-				macb_tx_unmap(bp, tx_skb);
-				tail++;
-				tx_skb = macb_tx_skb(queue, tail);
-				skb = tx_skb->skb;
-			}
-
-			/* ctrl still refers to the first buffer descriptor
-			 * since it's the only one written back by the hardware
-			 */
-			if (!(ctrl & MACB_BIT(TX_BUF_EXHAUSTED))) {
-				netdev_vdbg(bp->dev, "txerr skb %u (data %p) TX complete\n",
-					    macb_tx_ring_wrap(bp, tail),
-					    skb->data);
-				bp->dev->stats.tx_packets++;
-				bp->dev->stats.tx_bytes += skb->len;
-			}
-		} else {
-			/* "Buffers exhausted mid-frame" errors may only happen
-			 * if the driver is buggy, so complain loudly about
-			 * those. Statistics are updated by hardware.
-			 */
-			if (ctrl & MACB_BIT(TX_BUF_EXHAUSTED))
-				netdev_err(bp->dev,
-					   "BUG: TX buffers exhausted mid-frame\n");
-
-			desc->ctrl = ctrl | MACB_BIT(TX_USED);
-		}
-
-		macb_tx_unmap(bp, tx_skb);
-	}
-
-	/* Set end of TX queue */
-	desc = macb_tx_desc(queue, 0);
-	macb_set_addr(bp, desc, 0);
-	desc->ctrl = MACB_BIT(TX_USED);
-
-	/* Make descriptor updates visible to hardware */
-	wmb();
-
-	/* Reinitialize the TX desc queue */
-	queue_writel(queue, TBQP, lower_32_bits(queue->tx_ring_dma));
-#ifdef CONFIG_ARCH_DMA_ADDR_T_64BIT
-	if (bp->hw_dma_cap == HW_DMA_CAP_64B)
-		queue_writel(queue, TBQPH, upper_32_bits(queue->tx_ring_dma));
-#endif
-	/* Make TX ring reflect state of hardware */
-	queue->tx_head = 0;
-	queue->tx_tail = 0;
-
-	/* Housework before enabling TX IRQ */
-	macb_writel(bp, TSR, macb_readl(bp, TSR));
-	queue_writel(queue, IER, MACB_TX_INT_FLAGS);
-
-	/* Now we are ready to start transmission again */
-	netif_tx_start_all_queues(bp->dev);
-	macb_writel(bp, NCR, macb_readl(bp, NCR) | MACB_BIT(TSTART));
-
-	spin_unlock_irqrestore(&bp->lock, flags);
-}
-
-static void macb_tx_interrupt(struct macb_queue *queue)
-{
-	unsigned int tail;
-	unsigned int head;
-	u32 status;
-	struct macb *bp = queue->bp;
-	u16 queue_index = queue - bp->queues;
-
-	status = macb_readl(bp, TSR);
-	macb_writel(bp, TSR, status);
-
-	if (bp->caps & MACB_CAPS_ISR_CLEAR_ON_WRITE)
-		queue_writel(queue, ISR, MACB_BIT(TCOMP));
-
-	netdev_vdbg(bp->dev, "macb_tx_interrupt status = 0x%03lx\n",
-		    (unsigned long)status);
-
-	head = queue->tx_head;
-	for (tail = queue->tx_tail; tail != head; tail++) {
-		struct macb_tx_skb	*tx_skb;
-		struct sk_buff		*skb;
-		struct macb_dma_desc	*desc;
-		u32			ctrl;
-
-		desc = macb_tx_desc(queue, tail);
-
-		/* Make hw descriptor updates visible to CPU */
-		rmb();
-
-		ctrl = desc->ctrl;
-
-		/* TX_USED bit is only set by hardware on the very first buffer
-		 * descriptor of the transmitted frame.
-		 */
-		if (!(ctrl & MACB_BIT(TX_USED)))
-			break;
-
-		/* Process all buffers of the current transmitted frame */
-		for (;; tail++) {
-			tx_skb = macb_tx_skb(queue, tail);
-			skb = tx_skb->skb;
-
-			/* First, update TX stats if needed */
-			if (skb) {
-				netdev_vdbg(bp->dev, "skb %u (data %p) TX complete\n",
-					    macb_tx_ring_wrap(bp, tail),
-					    skb->data);
-				bp->dev->stats.tx_packets++;
-				bp->dev->stats.tx_bytes += skb->len;
-			}
-
-			/* Now we can safely release resources */
-			macb_tx_unmap(bp, tx_skb);
-
-			/* skb is set only for the last buffer of the frame.
-			 * WARNING: at this point skb has been freed by
-			 * macb_tx_unmap().
-			 */
-			if (skb)
-				break;
-		}
-	}
-
-	queue->tx_tail = tail;
-	if (__netif_subqueue_stopped(bp->dev, queue_index) &&
-	    CIRC_CNT(queue->tx_head, queue->tx_tail,
-		     bp->tx_ring_size) <= MACB_TX_WAKEUP_THRESH(bp))
-		netif_wake_subqueue(bp->dev, queue_index);
-}
-
-static void gem_rx_refill(struct macb *bp)
-{
-	unsigned int		entry;
-	struct sk_buff		*skb;
-	dma_addr_t		paddr;
-	struct macb_dma_desc *desc;
-
-	while (CIRC_SPACE(bp->rx_prepared_head, bp->rx_tail,
-			  bp->rx_ring_size) > 0) {
-		entry = macb_rx_ring_wrap(bp, bp->rx_prepared_head);
-
-		/* Make hw descriptor updates visible to CPU */
-		rmb();
-
-		bp->rx_prepared_head++;
-		desc = macb_rx_desc(bp, entry);
-
-		if (!bp->rx_skbuff[entry]) {
-			/* allocate sk_buff for this free entry in ring */
-			skb = netdev_alloc_skb(bp->dev, bp->rx_buffer_size);
-			if (unlikely(!skb)) {
-				netdev_err(bp->dev,
-					   "Unable to allocate sk_buff\n");
-				break;
-			}
-
-			/* now fill corresponding descriptor entry */
-			paddr = dma_map_single(&bp->pdev->dev, skb->data,
-					       bp->rx_buffer_size,
-					       DMA_FROM_DEVICE);
-			if (dma_mapping_error(&bp->pdev->dev, paddr)) {
-				dev_kfree_skb(skb);
-				break;
-			}
-
-			bp->rx_skbuff[entry] = skb;
-
-			if (entry == bp->rx_ring_size - 1)
-				paddr |= MACB_BIT(RX_WRAP);
-			macb_set_addr(bp, desc, paddr);
-			desc->ctrl = 0;
-
-			/* properly align Ethernet header */
-			skb_reserve(skb, NET_IP_ALIGN);
-		} else {
-			desc->addr &= ~MACB_BIT(RX_USED);
-			desc->ctrl = 0;
-		}
-	}
-
-	/* Make descriptor updates visible to hardware */
-	wmb();
-
-	netdev_vdbg(bp->dev, "rx ring: prepared head %d, tail %d\n",
-		    bp->rx_prepared_head, bp->rx_tail);
-}
-
-/* Mark DMA descriptors from begin up to and not including end as unused */
-static void discard_partial_frame(struct macb *bp, unsigned int begin,
-				  unsigned int end)
-{
-	unsigned int frag;
-
-	for (frag = begin; frag != end; frag++) {
-		struct macb_dma_desc *desc = macb_rx_desc(bp, frag);
-
-		desc->addr &= ~MACB_BIT(RX_USED);
-	}
-
-	/* Make descriptor updates visible to hardware */
-	wmb();
-
-	/* When this happens, the hardware stats registers for
-	 * whatever caused this is updated, so we don't have to record
-	 * anything.
-	 */
-}
-
-static int gem_rx(struct macb *bp, int budget)
-{
-	unsigned int		len;
-	unsigned int		entry;
-	struct sk_buff		*skb;
-	struct macb_dma_desc	*desc;
-	int			count = 0;
-
-	while (count < budget) {
-		u32 ctrl;
-		dma_addr_t addr;
-		bool rxused;
-
-		entry = macb_rx_ring_wrap(bp, bp->rx_tail);
-		desc = macb_rx_desc(bp, entry);
-
-		/* Make hw descriptor updates visible to CPU */
-		rmb();
-
-		rxused = (desc->addr & MACB_BIT(RX_USED)) ? true : false;
-		addr = macb_get_addr(bp, desc);
-		ctrl = desc->ctrl;
-
-		if (!rxused)
-			break;
-
-		bp->rx_tail++;
-		count++;
-
-		if (!(ctrl & MACB_BIT(RX_SOF) && ctrl & MACB_BIT(RX_EOF))) {
-			netdev_err(bp->dev,
-				   "not whole frame pointed by descriptor\n");
-			bp->dev->stats.rx_dropped++;
-			break;
-		}
-		skb = bp->rx_skbuff[entry];
-		if (unlikely(!skb)) {
-			netdev_err(bp->dev,
-				   "inconsistent Rx descriptor chain\n");
-			bp->dev->stats.rx_dropped++;
-			break;
-		}
-		/* now everything is ready for receiving packet */
-		bp->rx_skbuff[entry] = NULL;
-		len = ctrl & bp->rx_frm_len_mask;
-
-		netdev_vdbg(bp->dev, "gem_rx %u (len %u)\n", entry, len);
-
-		skb_put(skb, len);
-		dma_unmap_single(&bp->pdev->dev, addr,
-				 bp->rx_buffer_size, DMA_FROM_DEVICE);
-
-		skb->protocol = eth_type_trans(skb, bp->dev);
-		skb_checksum_none_assert(skb);
-		if (bp->dev->features & NETIF_F_RXCSUM &&
-		    !(bp->dev->flags & IFF_PROMISC) &&
-		    GEM_BFEXT(RX_CSUM, ctrl) & GEM_RX_CSUM_CHECKED_MASK)
-			skb->ip_summed = CHECKSUM_UNNECESSARY;
-
-		bp->dev->stats.rx_packets++;
-		bp->dev->stats.rx_bytes += skb->len;
-
-#if defined(DEBUG) && defined(VERBOSE_DEBUG)
-		netdev_vdbg(bp->dev, "received skb of length %u, csum: %08x\n",
-			    skb->len, skb->csum);
-		print_hex_dump(KERN_DEBUG, " mac: ", DUMP_PREFIX_ADDRESS, 16, 1,
-			       skb_mac_header(skb), 16, true);
-		print_hex_dump(KERN_DEBUG, "data: ", DUMP_PREFIX_ADDRESS, 16, 1,
-			       skb->data, 32, true);
-#endif
-
-		netif_receive_skb(skb);
-	}
-
-	gem_rx_refill(bp);
-
-	return count;
-}
-
-static int macb_rx_frame(struct macb *bp, unsigned int first_frag,
-			 unsigned int last_frag)
-{
-	unsigned int len;
-	unsigned int frag;
-	unsigned int offset;
-	struct sk_buff *skb;
-	struct macb_dma_desc *desc;
-
-	desc = macb_rx_desc(bp, last_frag);
-	len = desc->ctrl & bp->rx_frm_len_mask;
-
-	netdev_vdbg(bp->dev, "macb_rx_frame frags %u - %u (len %u)\n",
-		macb_rx_ring_wrap(bp, first_frag),
-		macb_rx_ring_wrap(bp, last_frag), len);
-
-	/* The ethernet header starts NET_IP_ALIGN bytes into the
-	 * first buffer. Since the header is 14 bytes, this makes the
-	 * payload word-aligned.
-	 *
-	 * Instead of calling skb_reserve(NET_IP_ALIGN), we just copy
-	 * the two padding bytes into the skb so that we avoid hitting
-	 * the slowpath in memcpy(), and pull them off afterwards.
-	 */
-	skb = netdev_alloc_skb(bp->dev, len + NET_IP_ALIGN);
-	if (!skb) {
-		bp->dev->stats.rx_dropped++;
-		for (frag = first_frag; ; frag++) {
-			desc = macb_rx_desc(bp, frag);
-			desc->addr &= ~MACB_BIT(RX_USED);
-			if (frag == last_frag)
-				break;
-		}
-
-		/* Make descriptor updates visible to hardware */
-		wmb();
-
-		return 1;
-	}
-
-	offset = 0;
-	len += NET_IP_ALIGN;
-	skb_checksum_none_assert(skb);
-	skb_put(skb, len);
-
-	for (frag = first_frag; ; frag++) {
-		unsigned int frag_len = bp->rx_buffer_size;
-
-		if (offset + frag_len > len) {
-			if (unlikely(frag != last_frag)) {
-				dev_kfree_skb_any(skb);
-				return -1;
-			}
-			frag_len = len - offset;
-		}
-		skb_copy_to_linear_data_offset(skb, offset,
-					       macb_rx_buffer(bp, frag),
-					       frag_len);
-		offset += bp->rx_buffer_size;
-		desc = macb_rx_desc(bp, frag);
-		desc->addr &= ~MACB_BIT(RX_USED);
-
-		if (frag == last_frag)
-			break;
-	}
-
-	/* Make descriptor updates visible to hardware */
-	wmb();
-
-	__skb_pull(skb, NET_IP_ALIGN);
-	skb->protocol = eth_type_trans(skb, bp->dev);
-
-	bp->dev->stats.rx_packets++;
-	bp->dev->stats.rx_bytes += skb->len;
-	netdev_vdbg(bp->dev, "received skb of length %u, csum: %08x\n",
-		    skb->len, skb->csum);
-	netif_receive_skb(skb);
-
-	return 0;
-}
-
-static inline void macb_init_rx_ring(struct macb *bp)
-{
-	dma_addr_t addr;
-	struct macb_dma_desc *desc = NULL;
-	int i;
-
-	addr = bp->rx_buffers_dma;
-	for (i = 0; i < bp->rx_ring_size; i++) {
-		desc = macb_rx_desc(bp, i);
-		macb_set_addr(bp, desc, addr);
-		desc->ctrl = 0;
-		addr += bp->rx_buffer_size;
-	}
-	desc->addr |= MACB_BIT(RX_WRAP);
-	bp->rx_tail = 0;
-}
-
-static int macb_rx(struct macb *bp, int budget)
-{
-	bool reset_rx_queue = false;
-	int received = 0;
-	unsigned int tail;
-	int first_frag = -1;
-
-	for (tail = bp->rx_tail; budget > 0; tail++) {
-		struct macb_dma_desc *desc = macb_rx_desc(bp, tail);
-		u32 ctrl;
-
-		/* Make hw descriptor updates visible to CPU */
-		rmb();
-
-		ctrl = desc->ctrl;
-
-		if (!(desc->addr & MACB_BIT(RX_USED)))
-			break;
-
-		if (ctrl & MACB_BIT(RX_SOF)) {
-			if (first_frag != -1)
-				discard_partial_frame(bp, first_frag, tail);
-			first_frag = tail;
-		}
-
-		if (ctrl & MACB_BIT(RX_EOF)) {
-			int dropped;
-
-			if (unlikely(first_frag == -1)) {
-				reset_rx_queue = true;
-				continue;
-			}
-
-			dropped = macb_rx_frame(bp, first_frag, tail);
-			first_frag = -1;
-			if (unlikely(dropped < 0)) {
-				reset_rx_queue = true;
-				continue;
-			}
-			if (!dropped) {
-				received++;
-				budget--;
-			}
-		}
-	}
-
-	if (unlikely(reset_rx_queue)) {
-		unsigned long flags;
-		u32 ctrl;
-
-		netdev_err(bp->dev, "RX queue corruption: reset it\n");
-
-		spin_lock_irqsave(&bp->lock, flags);
-
-		ctrl = macb_readl(bp, NCR);
-		macb_writel(bp, NCR, ctrl & ~MACB_BIT(RE));
-
-		macb_init_rx_ring(bp);
-		macb_writel(bp, RBQP, bp->rx_ring_dma);
-
-		macb_writel(bp, NCR, ctrl | MACB_BIT(RE));
-
-		spin_unlock_irqrestore(&bp->lock, flags);
-		return received;
-	}
-
-	if (first_frag != -1)
-		bp->rx_tail = first_frag;
-	else
-		bp->rx_tail = tail;
-
-	return received;
-}
-
-static int macb_poll(struct napi_struct *napi, int budget)
-{
-	struct macb *bp = container_of(napi, struct macb, napi);
-	int work_done;
-	u32 status;
-
-	status = macb_readl(bp, RSR);
-	macb_writel(bp, RSR, status);
-
-	work_done = 0;
-
-	netdev_vdbg(bp->dev, "poll: status = %08lx, budget = %d\n",
-		    (unsigned long)status, budget);
-
-	work_done = bp->macbgem_ops.mog_rx(bp, budget);
-	if (work_done < budget) {
-		napi_complete_done(napi, work_done);
-
-		/* Packets received while interrupts were disabled */
-		status = macb_readl(bp, RSR);
-		if (status) {
-			if (bp->caps & MACB_CAPS_ISR_CLEAR_ON_WRITE)
-				macb_writel(bp, ISR, MACB_BIT(RCOMP));
-			napi_reschedule(napi);
-		} else {
-			macb_writel(bp, IER, MACB_RX_INT_FLAGS);
-		}
-	}
-
-	/* TODO: Handle errors */
-
-	return work_done;
-}
-
-static irqreturn_t macb_interrupt(int irq, void *dev_id)
-{
-	struct macb_queue *queue = dev_id;
-	struct macb *bp = queue->bp;
-	struct net_device *dev = bp->dev;
-	u32 status, ctrl;
-
-	status = queue_readl(queue, ISR);
-
-	if (unlikely(!status))
-		return IRQ_NONE;
-
-	spin_lock(&bp->lock);
-
-	while (status) {
-		/* close possible race with dev_close */
-		if (unlikely(!netif_running(dev))) {
-			queue_writel(queue, IDR, -1);
-			if (bp->caps & MACB_CAPS_ISR_CLEAR_ON_WRITE)
-				queue_writel(queue, ISR, -1);
-			break;
-		}
-
-		netdev_vdbg(bp->dev, "queue = %u, isr = 0x%08lx\n",
-			    (unsigned int)(queue - bp->queues),
-			    (unsigned long)status);
-
-		if (status & MACB_RX_INT_FLAGS) {
-			/* There's no point taking any more interrupts
-			 * until we have processed the buffers. The
-			 * scheduling call may fail if the poll routine
-			 * is already scheduled, so disable interrupts
-			 * now.
-			 */
-			queue_writel(queue, IDR, MACB_RX_INT_FLAGS);
-			if (bp->caps & MACB_CAPS_ISR_CLEAR_ON_WRITE)
-				queue_writel(queue, ISR, MACB_BIT(RCOMP));
-
-			if (napi_schedule_prep(&bp->napi)) {
-				netdev_vdbg(bp->dev, "scheduling RX softirq\n");
-				__napi_schedule(&bp->napi);
-			}
-		}
-
-		if (unlikely(status & (MACB_TX_ERR_FLAGS))) {
-			queue_writel(queue, IDR, MACB_TX_INT_FLAGS);
-			schedule_work(&queue->tx_error_task);
-
-			if (bp->caps & MACB_CAPS_ISR_CLEAR_ON_WRITE)
-				queue_writel(queue, ISR, MACB_TX_ERR_FLAGS);
-
-			break;
-		}
-
-		if (status & MACB_BIT(TCOMP))
-			macb_tx_interrupt(queue);
-
-		/* Link change detection isn't possible with RMII, so we'll
-		 * add that if/when we get our hands on a full-blown MII PHY.
-		 */
-
-		/* There is a hardware issue under heavy load where DMA can
-		 * stop, this causes endless "used buffer descriptor read"
-		 * interrupts but it can be cleared by re-enabling RX. See
-		 * the at91 manual, section 41.3.1 or the Zynq manual
-		 * section 16.7.4 for details.
-		 */
-		if (status & MACB_BIT(RXUBR)) {
-			ctrl = macb_readl(bp, NCR);
-			macb_writel(bp, NCR, ctrl & ~MACB_BIT(RE));
-			wmb();
-			macb_writel(bp, NCR, ctrl | MACB_BIT(RE));
-
-			if (bp->caps & MACB_CAPS_ISR_CLEAR_ON_WRITE)
-				queue_writel(queue, ISR, MACB_BIT(RXUBR));
-		}
-
-		if (status & MACB_BIT(ISR_ROVR)) {
-			/* We missed at least one packet */
-			if (macb_is_gem(bp))
-				bp->hw_stats.gem.rx_overruns++;
-			else
-				bp->hw_stats.macb.rx_overruns++;
-
-			if (bp->caps & MACB_CAPS_ISR_CLEAR_ON_WRITE)
-				queue_writel(queue, ISR, MACB_BIT(ISR_ROVR));
-		}
-
-		if (status & MACB_BIT(HRESP)) {
-			/* TODO: Reset the hardware, and maybe move the
-			 * netdev_err to a lower-priority context as well
-			 * (work queue?)
-			 */
-			netdev_err(dev, "DMA bus error: HRESP not OK\n");
-
-			if (bp->caps & MACB_CAPS_ISR_CLEAR_ON_WRITE)
-				queue_writel(queue, ISR, MACB_BIT(HRESP));
-		}
-
-		status = queue_readl(queue, ISR);
-	}
-
-	spin_unlock(&bp->lock);
-
-	return IRQ_HANDLED;
-}
-
-#ifdef CONFIG_NET_POLL_CONTROLLER
-/* Polling receive - used by netconsole and other diagnostic tools
- * to allow network i/o with interrupts disabled.
- */
-static void macb_poll_controller(struct net_device *dev)
-{
-	struct macb *bp = netdev_priv(dev);
-	struct macb_queue *queue;
-	unsigned long flags;
-	unsigned int q;
-
-	local_irq_save(flags);
-	for (q = 0, queue = bp->queues; q < bp->num_queues; ++q, ++queue)
-		macb_interrupt(dev->irq, queue);
-	local_irq_restore(flags);
-}
-#endif
-
-static unsigned int macb_tx_map(struct macb *bp,
-				struct macb_queue *queue,
-				struct sk_buff *skb,
-				unsigned int hdrlen)
-{
-	dma_addr_t mapping;
-	unsigned int len, entry, i, tx_head = queue->tx_head;
-	struct macb_tx_skb *tx_skb = NULL;
-	struct macb_dma_desc *desc;
-	unsigned int offset, size, count = 0;
-	unsigned int f, nr_frags = skb_shinfo(skb)->nr_frags;
-	unsigned int eof = 1, mss_mfs = 0;
-	u32 ctrl, lso_ctrl = 0, seq_ctrl = 0;
-
-	/* LSO */
-	if (skb_shinfo(skb)->gso_size != 0) {
-		if (ip_hdr(skb)->protocol == IPPROTO_UDP)
-			/* UDP - UFO */
-			lso_ctrl = MACB_LSO_UFO_ENABLE;
-		else
-			/* TCP - TSO */
-			lso_ctrl = MACB_LSO_TSO_ENABLE;
-	}
-
-	/* First, map non-paged data */
-	len = skb_headlen(skb);
-
-	/* first buffer length */
-	size = hdrlen;
-
-	offset = 0;
-	while (len) {
-		entry = macb_tx_ring_wrap(bp, tx_head);
-		tx_skb = &queue->tx_skb[entry];
-
-		mapping = dma_map_single(&bp->pdev->dev,
-					 skb->data + offset,
-					 size, DMA_TO_DEVICE);
-		if (dma_mapping_error(&bp->pdev->dev, mapping))
-			goto dma_error;
-
-		/* Save info to properly release resources */
-		tx_skb->skb = NULL;
-		tx_skb->mapping = mapping;
-		tx_skb->size = size;
-		tx_skb->mapped_as_page = false;
-
-		len -= size;
-		offset += size;
-		count++;
-		tx_head++;
-
-		size = min(len, bp->max_tx_length);
-	}
-
-	/* Then, map paged data from fragments */
-	for (f = 0; f < nr_frags; f++) {
-		const skb_frag_t *frag = &skb_shinfo(skb)->frags[f];
-
-		len = skb_frag_size(frag);
-		offset = 0;
-		while (len) {
-			size = min(len, bp->max_tx_length);
-			entry = macb_tx_ring_wrap(bp, tx_head);
-			tx_skb = &queue->tx_skb[entry];
-
-			mapping = skb_frag_dma_map(&bp->pdev->dev, frag,
-						   offset, size, DMA_TO_DEVICE);
-			if (dma_mapping_error(&bp->pdev->dev, mapping))
-				goto dma_error;
-
-			/* Save info to properly release resources */
-			tx_skb->skb = NULL;
-			tx_skb->mapping = mapping;
-			tx_skb->size = size;
-			tx_skb->mapped_as_page = true;
-
-			len -= size;
-			offset += size;
-			count++;
-			tx_head++;
-		}
-	}
-
-	/* Should never happen */
-	if (unlikely(!tx_skb)) {
-		netdev_err(bp->dev, "BUG! empty skb!\n");
-		return 0;
-	}
-
-	/* This is the last buffer of the frame: save socket buffer */
-	tx_skb->skb = skb;
-
-	/* Update TX ring: update buffer descriptors in reverse order
-	 * to avoid race condition
-	 */
-
-	/* Set 'TX_USED' bit in buffer descriptor at tx_head position
-	 * to set the end of TX queue
-	 */
-	i = tx_head;
-	entry = macb_tx_ring_wrap(bp, i);
-	ctrl = MACB_BIT(TX_USED);
-	desc = macb_tx_desc(queue, entry);
-	desc->ctrl = ctrl;
-
-	if (lso_ctrl) {
-		if (lso_ctrl == MACB_LSO_UFO_ENABLE)
-			/* include header and FCS in value given to h/w */
-			mss_mfs = skb_shinfo(skb)->gso_size +
-					skb_transport_offset(skb) +
-					ETH_FCS_LEN;
-		else /* TSO */ {
-			mss_mfs = skb_shinfo(skb)->gso_size;
-			/* TCP Sequence Number Source Select
-			 * can be set only for TSO
-			 */
-			seq_ctrl = 0;
-		}
-	}
-
-	do {
-		i--;
-		entry = macb_tx_ring_wrap(bp, i);
-		tx_skb = &queue->tx_skb[entry];
-		desc = macb_tx_desc(queue, entry);
-
-		ctrl = (u32)tx_skb->size;
-		if (eof) {
-			ctrl |= MACB_BIT(TX_LAST);
-			eof = 0;
-		}
-		if (unlikely(entry == (bp->tx_ring_size - 1)))
-			ctrl |= MACB_BIT(TX_WRAP);
-
-		/* First descriptor is header descriptor */
-		if (i == queue->tx_head) {
-			ctrl |= MACB_BF(TX_LSO, lso_ctrl);
-			ctrl |= MACB_BF(TX_TCP_SEQ_SRC, seq_ctrl);
-		} else
-			/* Only set MSS/MFS on payload descriptors
-			 * (second or later descriptor)
-			 */
-			ctrl |= MACB_BF(MSS_MFS, mss_mfs);
-
-		/* Set TX buffer descriptor */
-		macb_set_addr(bp, desc, tx_skb->mapping);
-		/* desc->addr must be visible to hardware before clearing
-		 * 'TX_USED' bit in desc->ctrl.
-		 */
-		wmb();
-		desc->ctrl = ctrl;
-	} while (i != queue->tx_head);
-
-	queue->tx_head = tx_head;
-
-	return count;
-
-dma_error:
-	netdev_err(bp->dev, "TX DMA map failed\n");
-
-	for (i = queue->tx_head; i != tx_head; i++) {
-		tx_skb = macb_tx_skb(queue, i);
-
-		macb_tx_unmap(bp, tx_skb);
-	}
-
-	return 0;
-}
-
-static netdev_features_t macb_features_check(struct sk_buff *skb,
-					     struct net_device *dev,
-					     netdev_features_t features)
-{
-	unsigned int nr_frags, f;
-	unsigned int hdrlen;
-
-	/* Validate LSO compatibility */
-
-	/* there is only one buffer */
-	if (!skb_is_nonlinear(skb))
-		return features;
-
-	/* length of header */
-	hdrlen = skb_transport_offset(skb);
-	if (ip_hdr(skb)->protocol == IPPROTO_TCP)
-		hdrlen += tcp_hdrlen(skb);
-
-	/* For LSO:
-	 * When software supplies two or more payload buffers all payload buffers
-	 * apart from the last must be a multiple of 8 bytes in size.
-	 */
-	if (!IS_ALIGNED(skb_headlen(skb) - hdrlen, MACB_TX_LEN_ALIGN))
-		return features & ~MACB_NETIF_LSO;
-
-	nr_frags = skb_shinfo(skb)->nr_frags;
-	/* No need to check last fragment */
-	nr_frags--;
-	for (f = 0; f < nr_frags; f++) {
-		const skb_frag_t *frag = &skb_shinfo(skb)->frags[f];
-
-		if (!IS_ALIGNED(skb_frag_size(frag), MACB_TX_LEN_ALIGN))
-			return features & ~MACB_NETIF_LSO;
-	}
-	return features;
-}
-
-static inline int macb_clear_csum(struct sk_buff *skb)
-{
-	/* no change for packets without checksum offloading */
-	if (skb->ip_summed != CHECKSUM_PARTIAL)
-		return 0;
-
-	/* make sure we can modify the header */
-	if (unlikely(skb_cow_head(skb, 0)))
-		return -1;
-
-	/* initialize checksum field
-	 * This is required - at least for Zynq, which otherwise calculates
-	 * wrong UDP header checksums for UDP packets with UDP data len <=2
-	 */
-	*(__sum16 *)(skb_checksum_start(skb) + skb->csum_offset) = 0;
-	return 0;
-}
-
-static int macb_start_xmit(struct sk_buff *skb, struct net_device *dev)
-{
-	u16 queue_index = skb_get_queue_mapping(skb);
-	struct macb *bp = netdev_priv(dev);
-	struct macb_queue *queue = &bp->queues[queue_index];
-	unsigned long flags;
-	unsigned int desc_cnt, nr_frags, frag_size, f;
-	unsigned int hdrlen;
-	bool is_lso, is_udp = 0;
-
-	is_lso = (skb_shinfo(skb)->gso_size != 0);
-
-	if (is_lso) {
-		is_udp = !!(ip_hdr(skb)->protocol == IPPROTO_UDP);
-
-		/* length of headers */
-		if (is_udp)
-			/* only queue eth + ip headers separately for UDP */
-			hdrlen = skb_transport_offset(skb);
-		else
-			hdrlen = skb_transport_offset(skb) + tcp_hdrlen(skb);
-		if (skb_headlen(skb) < hdrlen) {
-			netdev_err(bp->dev, "Error - LSO headers fragmented!!!\n");
-			/* if this is required, would need to copy to single buffer */
-			return NETDEV_TX_BUSY;
-		}
-	} else
-		hdrlen = min(skb_headlen(skb), bp->max_tx_length);
-
-#if defined(DEBUG) && defined(VERBOSE_DEBUG)
-	netdev_vdbg(bp->dev,
-		    "start_xmit: queue %hu len %u head %p data %p tail %p end %p\n",
-		    queue_index, skb->len, skb->head, skb->data,
-		    skb_tail_pointer(skb), skb_end_pointer(skb));
-	print_hex_dump(KERN_DEBUG, "data: ", DUMP_PREFIX_OFFSET, 16, 1,
-		       skb->data, 16, true);
-#endif
-
-	/* Count how many TX buffer descriptors are needed to send this
-	 * socket buffer: skb fragments of jumbo frames may need to be
-	 * split into many buffer descriptors.
-	 */
-	if (is_lso && (skb_headlen(skb) > hdrlen))
-		/* extra header descriptor if also payload in first buffer */
-		desc_cnt = DIV_ROUND_UP((skb_headlen(skb) - hdrlen), bp->max_tx_length) + 1;
-	else
-		desc_cnt = DIV_ROUND_UP(skb_headlen(skb), bp->max_tx_length);
-	nr_frags = skb_shinfo(skb)->nr_frags;
-	for (f = 0; f < nr_frags; f++) {
-		frag_size = skb_frag_size(&skb_shinfo(skb)->frags[f]);
-		desc_cnt += DIV_ROUND_UP(frag_size, bp->max_tx_length);
-	}
-
-	spin_lock_irqsave(&bp->lock, flags);
-
-	/* This is a hard error, log it. */
-	if (CIRC_SPACE(queue->tx_head, queue->tx_tail,
-		       bp->tx_ring_size) < desc_cnt) {
-		netif_stop_subqueue(dev, queue_index);
-		spin_unlock_irqrestore(&bp->lock, flags);
-		netdev_dbg(bp->dev, "tx_head = %u, tx_tail = %u\n",
-			   queue->tx_head, queue->tx_tail);
-		return NETDEV_TX_BUSY;
-	}
-
-	if (macb_clear_csum(skb)) {
-		dev_kfree_skb_any(skb);
-		goto unlock;
-	}
-
-	/* Map socket buffer for DMA transfer */
-	if (!macb_tx_map(bp, queue, skb, hdrlen)) {
-		dev_kfree_skb_any(skb);
-		goto unlock;
-	}
-
-	/* Make newly initialized descriptor visible to hardware */
-	wmb();
-
-	skb_tx_timestamp(skb);
-
-	macb_writel(bp, NCR, macb_readl(bp, NCR) | MACB_BIT(TSTART));
-
-	if (CIRC_SPACE(queue->tx_head, queue->tx_tail, bp->tx_ring_size) < 1)
-		netif_stop_subqueue(dev, queue_index);
-
-unlock:
-	spin_unlock_irqrestore(&bp->lock, flags);
-
-	return NETDEV_TX_OK;
-}
-
-static void macb_init_rx_buffer_size(struct macb *bp, size_t size)
-{
-	if (!macb_is_gem(bp)) {
-		bp->rx_buffer_size = MACB_RX_BUFFER_SIZE;
-	} else {
-		bp->rx_buffer_size = size;
-
-		if (bp->rx_buffer_size % RX_BUFFER_MULTIPLE) {
-			netdev_dbg(bp->dev,
-				   "RX buffer must be multiple of %d bytes, expanding\n",
-				   RX_BUFFER_MULTIPLE);
-			bp->rx_buffer_size =
-				roundup(bp->rx_buffer_size, RX_BUFFER_MULTIPLE);
-		}
-	}
-
-	netdev_dbg(bp->dev, "mtu [%u] rx_buffer_size [%zu]\n",
-		   bp->dev->mtu, bp->rx_buffer_size);
-}
-
-static void gem_free_rx_buffers(struct macb *bp)
-{
-	struct sk_buff		*skb;
-	struct macb_dma_desc	*desc;
-	dma_addr_t		addr;
-	int i;
-
-	if (!bp->rx_skbuff)
-		return;
-
-	for (i = 0; i < bp->rx_ring_size; i++) {
-		skb = bp->rx_skbuff[i];
-
-		if (!skb)
-			continue;
-
-		desc = macb_rx_desc(bp, i);
-		addr = macb_get_addr(bp, desc);
-
-		dma_unmap_single(&bp->pdev->dev, addr, bp->rx_buffer_size,
-				 DMA_FROM_DEVICE);
-		dev_kfree_skb_any(skb);
-		skb = NULL;
-	}
-
-	kfree(bp->rx_skbuff);
-	bp->rx_skbuff = NULL;
-}
-
-static void macb_free_rx_buffers(struct macb *bp)
-{
-	if (bp->rx_buffers) {
-		dma_free_coherent(&bp->pdev->dev,
-				  bp->rx_ring_size * bp->rx_buffer_size,
-				  bp->rx_buffers, bp->rx_buffers_dma);
-		bp->rx_buffers = NULL;
-	}
-}
-
-static void macb_free_consistent(struct macb *bp)
-{
-	struct macb_queue *queue;
-	unsigned int q;
-
-	bp->macbgem_ops.mog_free_rx_buffers(bp);
-	if (bp->rx_ring) {
-		dma_free_coherent(&bp->pdev->dev, RX_RING_BYTES(bp),
-				  bp->rx_ring, bp->rx_ring_dma);
-		bp->rx_ring = NULL;
-	}
-
-	for (q = 0, queue = bp->queues; q < bp->num_queues; ++q, ++queue) {
-		kfree(queue->tx_skb);
-		queue->tx_skb = NULL;
-		if (queue->tx_ring) {
-			dma_free_coherent(&bp->pdev->dev, TX_RING_BYTES(bp),
-					  queue->tx_ring, queue->tx_ring_dma);
-			queue->tx_ring = NULL;
-		}
-	}
-}
-
-static int gem_alloc_rx_buffers(struct macb *bp)
-{
-	int size;
-
-	size = bp->rx_ring_size * sizeof(struct sk_buff *);
-	bp->rx_skbuff = kzalloc(size, GFP_KERNEL);
-	if (!bp->rx_skbuff)
-		return -ENOMEM;
-	else
-		netdev_dbg(bp->dev,
-			   "Allocated %d RX struct sk_buff entries at %p\n",
-			   bp->rx_ring_size, bp->rx_skbuff);
-	return 0;
-}
-
-static int macb_alloc_rx_buffers(struct macb *bp)
-{
-	int size;
-
-	size = bp->rx_ring_size * bp->rx_buffer_size;
-	bp->rx_buffers = dma_alloc_coherent(&bp->pdev->dev, size,
-					    &bp->rx_buffers_dma, GFP_KERNEL);
-	if (!bp->rx_buffers)
-		return -ENOMEM;
-
-	netdev_dbg(bp->dev,
-		   "Allocated RX buffers of %d bytes at %08lx (mapped %p)\n",
-		   size, (unsigned long)bp->rx_buffers_dma, bp->rx_buffers);
-	return 0;
-}
-
-static int macb_alloc_consistent(struct macb *bp)
-{
-	struct macb_queue *queue;
-	unsigned int q;
-	int size;
-
-	for (q = 0, queue = bp->queues; q < bp->num_queues; ++q, ++queue) {
-		size = TX_RING_BYTES(bp);
-		queue->tx_ring = dma_alloc_coherent(&bp->pdev->dev, size,
-						    &queue->tx_ring_dma,
-						    GFP_KERNEL);
-		if (!queue->tx_ring)
-			goto out_err;
-		netdev_dbg(bp->dev,
-			   "Allocated TX ring for queue %u of %d bytes at %08lx (mapped %p)\n",
-			   q, size, (unsigned long)queue->tx_ring_dma,
-			   queue->tx_ring);
-
-		size = bp->tx_ring_size * sizeof(struct macb_tx_skb);
-		queue->tx_skb = kmalloc(size, GFP_KERNEL);
-		if (!queue->tx_skb)
-			goto out_err;
-	}
-
-	size = RX_RING_BYTES(bp);
-	bp->rx_ring = dma_alloc_coherent(&bp->pdev->dev, size,
-					 &bp->rx_ring_dma, GFP_KERNEL);
-	if (!bp->rx_ring)
-		goto out_err;
-	netdev_dbg(bp->dev,
-		   "Allocated RX ring of %d bytes at %08lx (mapped %p)\n",
-		   size, (unsigned long)bp->rx_ring_dma, bp->rx_ring);
-
-	if (bp->macbgem_ops.mog_alloc_rx_buffers(bp))
-		goto out_err;
-
-	return 0;
-
-out_err:
-	macb_free_consistent(bp);
-	return -ENOMEM;
-}
-
-static void gem_init_rings(struct macb *bp)
-{
-	struct macb_queue *queue;
-	struct macb_dma_desc *desc = NULL;
-	unsigned int q;
-	int i;
-
-	for (q = 0, queue = bp->queues; q < bp->num_queues; ++q, ++queue) {
-		for (i = 0; i < bp->tx_ring_size; i++) {
-			desc = macb_tx_desc(queue, i);
-			macb_set_addr(bp, desc, 0);
-			desc->ctrl = MACB_BIT(TX_USED);
-		}
-		desc->ctrl |= MACB_BIT(TX_WRAP);
-		queue->tx_head = 0;
-		queue->tx_tail = 0;
-	}
-
-	bp->rx_tail = 0;
-	bp->rx_prepared_head = 0;
-
-	gem_rx_refill(bp);
-}
-
-static void macb_init_rings(struct macb *bp)
-{
-	int i;
-	struct macb_dma_desc *desc = NULL;
-
-	macb_init_rx_ring(bp);
-
-	for (i = 0; i < bp->tx_ring_size; i++) {
-		desc = macb_tx_desc(&bp->queues[0], i);
-		macb_set_addr(bp, desc, 0);
-		desc->ctrl = MACB_BIT(TX_USED);
-	}
-	bp->queues[0].tx_head = 0;
-	bp->queues[0].tx_tail = 0;
-	desc->ctrl |= MACB_BIT(TX_WRAP);
-}
-
-static void macb_reset_hw(struct macb *bp)
-{
-	struct macb_queue *queue;
-	unsigned int q;
-
-	/* Disable RX and TX (XXX: Should we halt the transmission
-	 * more gracefully?)
-	 */
-	macb_writel(bp, NCR, 0);
-
-	/* Clear the stats registers (XXX: Update stats first?) */
-	macb_writel(bp, NCR, MACB_BIT(CLRSTAT));
-
-	/* Clear all status flags */
-	macb_writel(bp, TSR, -1);
-	macb_writel(bp, RSR, -1);
-
-	/* Disable all interrupts */
-	for (q = 0, queue = bp->queues; q < bp->num_queues; ++q, ++queue) {
-		queue_writel(queue, IDR, -1);
-		queue_readl(queue, ISR);
-		if (bp->caps & MACB_CAPS_ISR_CLEAR_ON_WRITE)
-			queue_writel(queue, ISR, -1);
-	}
-}
-
-static u32 gem_mdc_clk_div(struct macb *bp)
-{
-	u32 config;
-	unsigned long pclk_hz = clk_get_rate(bp->pclk);
-
-	if (pclk_hz <= 20000000)
-		config = GEM_BF(CLK, GEM_CLK_DIV8);
-	else if (pclk_hz <= 40000000)
-		config = GEM_BF(CLK, GEM_CLK_DIV16);
-	else if (pclk_hz <= 80000000)
-		config = GEM_BF(CLK, GEM_CLK_DIV32);
-	else if (pclk_hz <= 120000000)
-		config = GEM_BF(CLK, GEM_CLK_DIV48);
-	else if (pclk_hz <= 160000000)
-		config = GEM_BF(CLK, GEM_CLK_DIV64);
-	else
-		config = GEM_BF(CLK, GEM_CLK_DIV96);
-
-	return config;
-}
-
-static u32 macb_mdc_clk_div(struct macb *bp)
-{
-	u32 config;
-	unsigned long pclk_hz;
-
-	if (macb_is_gem(bp))
-		return gem_mdc_clk_div(bp);
-
-	pclk_hz = clk_get_rate(bp->pclk);
-	if (pclk_hz <= 20000000)
-		config = MACB_BF(CLK, MACB_CLK_DIV8);
-	else if (pclk_hz <= 40000000)
-		config = MACB_BF(CLK, MACB_CLK_DIV16);
-	else if (pclk_hz <= 80000000)
-		config = MACB_BF(CLK, MACB_CLK_DIV32);
-	else
-		config = MACB_BF(CLK, MACB_CLK_DIV64);
-
-	return config;
-}
-
-/* Get the DMA bus width field of the network configuration register that we
- * should program.  We find the width from decoding the design configuration
- * register to find the maximum supported data bus width.
- */
-static u32 macb_dbw(struct macb *bp)
-{
-	if (!macb_is_gem(bp))
-		return 0;
-
-	switch (GEM_BFEXT(DBWDEF, gem_readl(bp, DCFG1))) {
-	case 4:
-		return GEM_BF(DBW, GEM_DBW128);
-	case 2:
-		return GEM_BF(DBW, GEM_DBW64);
-	case 1:
-	default:
-		return GEM_BF(DBW, GEM_DBW32);
-	}
-}
-
-/* Configure the receive DMA engine
- * - use the correct receive buffer size
- * - set best burst length for DMA operations
- *   (if not supported by FIFO, it will fallback to default)
- * - set both rx/tx packet buffers to full memory size
- * These are configurable parameters for GEM.
- */
-static void macb_configure_dma(struct macb *bp)
-{
-	u32 dmacfg;
-
-	if (macb_is_gem(bp)) {
-		dmacfg = gem_readl(bp, DMACFG) & ~GEM_BF(RXBS, -1L);
-		dmacfg |= GEM_BF(RXBS, bp->rx_buffer_size / RX_BUFFER_MULTIPLE);
-		if (bp->dma_burst_length)
-			dmacfg = GEM_BFINS(FBLDO, bp->dma_burst_length, dmacfg);
-		dmacfg |= GEM_BIT(TXPBMS) | GEM_BF(RXBMS, -1L);
-		dmacfg &= ~GEM_BIT(ENDIA_PKT);
-
-		if (bp->native_io)
-			dmacfg &= ~GEM_BIT(ENDIA_DESC);
-		else
-			dmacfg |= GEM_BIT(ENDIA_DESC); /* CPU in big endian */
-
-		if (bp->dev->features & NETIF_F_HW_CSUM)
-			dmacfg |= GEM_BIT(TXCOEN);
-		else
-			dmacfg &= ~GEM_BIT(TXCOEN);
-
-#ifdef CONFIG_ARCH_DMA_ADDR_T_64BIT
-		if (bp->hw_dma_cap == HW_DMA_CAP_64B)
-			dmacfg |= GEM_BIT(ADDR64);
-#endif
-		netdev_dbg(bp->dev, "Cadence configure DMA with 0x%08x\n",
-			   dmacfg);
-		gem_writel(bp, DMACFG, dmacfg);
-	}
-}
-
-static void macb_init_hw(struct macb *bp)
-{
-	struct macb_queue *queue;
-	unsigned int q;
-
-	u32 config;
-
-	macb_reset_hw(bp);
-	macb_set_hwaddr(bp);
-
-	config = macb_mdc_clk_div(bp);
-	if (bp->phy_interface == PHY_INTERFACE_MODE_SGMII)
-		config |= GEM_BIT(SGMIIEN) | GEM_BIT(PCSSEL);
-	config |= MACB_BF(RBOF, NET_IP_ALIGN);	/* Make eth data aligned */
-	config |= MACB_BIT(PAE);		/* PAuse Enable */
-	config |= MACB_BIT(DRFCS);		/* Discard Rx FCS */
-	if (bp->caps & MACB_CAPS_JUMBO)
-		config |= MACB_BIT(JFRAME);	/* Enable jumbo frames */
-	else
-		config |= MACB_BIT(BIG);	/* Receive oversized frames */
-	if (bp->dev->flags & IFF_PROMISC)
-		config |= MACB_BIT(CAF);	/* Copy All Frames */
-	else if (macb_is_gem(bp) && bp->dev->features & NETIF_F_RXCSUM)
-		config |= GEM_BIT(RXCOEN);
-	if (!(bp->dev->flags & IFF_BROADCAST))
-		config |= MACB_BIT(NBC);	/* No BroadCast */
-	config |= macb_dbw(bp);
-	macb_writel(bp, NCFGR, config);
-	if ((bp->caps & MACB_CAPS_JUMBO) && bp->jumbo_max_len)
-		gem_writel(bp, JML, bp->jumbo_max_len);
-	bp->speed = SPEED_10;
-	bp->duplex = DUPLEX_HALF;
-	bp->rx_frm_len_mask = MACB_RX_FRMLEN_MASK;
-	if (bp->caps & MACB_CAPS_JUMBO)
-		bp->rx_frm_len_mask = MACB_RX_JFRMLEN_MASK;
-
-	macb_configure_dma(bp);
-
-	/* Initialize TX and RX buffers */
-	macb_writel(bp, RBQP, lower_32_bits(bp->rx_ring_dma));
-#ifdef CONFIG_ARCH_DMA_ADDR_T_64BIT
-	if (bp->hw_dma_cap == HW_DMA_CAP_64B)
-		macb_writel(bp, RBQPH, upper_32_bits(bp->rx_ring_dma));
-#endif
-	for (q = 0, queue = bp->queues; q < bp->num_queues; ++q, ++queue) {
-		queue_writel(queue, TBQP, lower_32_bits(queue->tx_ring_dma));
-#ifdef CONFIG_ARCH_DMA_ADDR_T_64BIT
-		if (bp->hw_dma_cap == HW_DMA_CAP_64B)
-			queue_writel(queue, TBQPH, upper_32_bits(queue->tx_ring_dma));
-#endif
-
-		/* Enable interrupts */
-		queue_writel(queue, IER,
-			     MACB_RX_INT_FLAGS |
-			     MACB_TX_INT_FLAGS |
-			     MACB_BIT(HRESP));
-	}
-
-	/* Enable TX and RX */
-	macb_writel(bp, NCR, MACB_BIT(RE) | MACB_BIT(TE) | MACB_BIT(MPE));
-}
-
-/* The hash address register is 64 bits long and takes up two
- * locations in the memory map.  The least significant bits are stored
- * in EMAC_HSL and the most significant bits in EMAC_HSH.
- *
- * The unicast hash enable and the multicast hash enable bits in the
- * network configuration register enable the reception of hash matched
- * frames. The destination address is reduced to a 6 bit index into
- * the 64 bit hash register using the following hash function.  The
- * hash function is an exclusive or of every sixth bit of the
- * destination address.
- *
- * hi[5] = da[5] ^ da[11] ^ da[17] ^ da[23] ^ da[29] ^ da[35] ^ da[41] ^ da[47]
- * hi[4] = da[4] ^ da[10] ^ da[16] ^ da[22] ^ da[28] ^ da[34] ^ da[40] ^ da[46]
- * hi[3] = da[3] ^ da[09] ^ da[15] ^ da[21] ^ da[27] ^ da[33] ^ da[39] ^ da[45]
- * hi[2] = da[2] ^ da[08] ^ da[14] ^ da[20] ^ da[26] ^ da[32] ^ da[38] ^ da[44]
- * hi[1] = da[1] ^ da[07] ^ da[13] ^ da[19] ^ da[25] ^ da[31] ^ da[37] ^ da[43]
- * hi[0] = da[0] ^ da[06] ^ da[12] ^ da[18] ^ da[24] ^ da[30] ^ da[36] ^ da[42]
- *
- * da[0] represents the least significant bit of the first byte
- * received, that is, the multicast/unicast indicator, and da[47]
- * represents the most significant bit of the last byte received.  If
- * the hash index, hi[n], points to a bit that is set in the hash
- * register then the frame will be matched according to whether the
- * frame is multicast or unicast.  A multicast match will be signalled
- * if the multicast hash enable bit is set, da[0] is 1 and the hash
- * index points to a bit set in the hash register.  A unicast match
- * will be signalled if the unicast hash enable bit is set, da[0] is 0
- * and the hash index points to a bit set in the hash register.  To
- * receive all multicast frames, the hash register should be set with
- * all ones and the multicast hash enable bit should be set in the
- * network configuration register.
- */
-
-static inline int hash_bit_value(int bitnr, __u8 *addr)
-{
-	if (addr[bitnr / 8] & (1 << (bitnr % 8)))
-		return 1;
-	return 0;
-}
-
-/* Return the hash index value for the specified address. */
-static int hash_get_index(__u8 *addr)
-{
-	int i, j, bitval;
-	int hash_index = 0;
-
-	for (j = 0; j < 6; j++) {
-		for (i = 0, bitval = 0; i < 8; i++)
-			bitval ^= hash_bit_value(i * 6 + j, addr);
-
-		hash_index |= (bitval << j);
-	}
-
-	return hash_index;
-}
-
-/* Add multicast addresses to the internal multicast-hash table. */
-static void macb_sethashtable(struct net_device *dev)
-{
-	struct netdev_hw_addr *ha;
-	unsigned long mc_filter[2];
-	unsigned int bitnr;
-	struct macb *bp = netdev_priv(dev);
-
-	mc_filter[0] = 0;
-	mc_filter[1] = 0;
-
-	netdev_for_each_mc_addr(ha, dev) {
-		bitnr = hash_get_index(ha->addr);
-		mc_filter[bitnr >> 5] |= 1 << (bitnr & 31);
-	}
-
-	macb_or_gem_writel(bp, HRB, mc_filter[0]);
-	macb_or_gem_writel(bp, HRT, mc_filter[1]);
-}
-
-/* Enable/Disable promiscuous and multicast modes. */
-static void macb_set_rx_mode(struct net_device *dev)
-{
-	unsigned long cfg;
-	struct macb *bp = netdev_priv(dev);
-
-	cfg = macb_readl(bp, NCFGR);
-
-	if (dev->flags & IFF_PROMISC) {
-		/* Enable promiscuous mode */
-		cfg |= MACB_BIT(CAF);
-
-		/* Disable RX checksum offload */
-		if (macb_is_gem(bp))
-			cfg &= ~GEM_BIT(RXCOEN);
-	} else {
-		/* Disable promiscuous mode */
-		cfg &= ~MACB_BIT(CAF);
-
-		/* Enable RX checksum offload only if requested */
-		if (macb_is_gem(bp) && dev->features & NETIF_F_RXCSUM)
-			cfg |= GEM_BIT(RXCOEN);
-	}
-
-	if (dev->flags & IFF_ALLMULTI) {
-		/* Enable all multicast mode */
-		macb_or_gem_writel(bp, HRB, -1);
-		macb_or_gem_writel(bp, HRT, -1);
-		cfg |= MACB_BIT(NCFGR_MTI);
-	} else if (!netdev_mc_empty(dev)) {
-		/* Enable specific multicasts */
-		macb_sethashtable(dev);
-		cfg |= MACB_BIT(NCFGR_MTI);
-	} else if (dev->flags & (~IFF_ALLMULTI)) {
-		/* Disable all multicast mode */
-		macb_or_gem_writel(bp, HRB, 0);
-		macb_or_gem_writel(bp, HRT, 0);
-		cfg &= ~MACB_BIT(NCFGR_MTI);
-	}
-
-	macb_writel(bp, NCFGR, cfg);
-}
-
-static int macb_open(struct net_device *dev)
-{
-	struct macb *bp = netdev_priv(dev);
-	size_t bufsz = dev->mtu + ETH_HLEN + ETH_FCS_LEN + NET_IP_ALIGN;
-	int err;
-
-	netdev_dbg(bp->dev, "open\n");
-
-	/* carrier starts down */
-	netif_carrier_off(dev);
-
-	/* if the phy is not yet register, retry later*/
-	if (!dev->phydev)
-		return -EAGAIN;
-
-	/* RX buffers initialization */
-	macb_init_rx_buffer_size(bp, bufsz);
-
-	err = macb_alloc_consistent(bp);
-	if (err) {
-		netdev_err(dev, "Unable to allocate DMA memory (error %d)\n",
-			   err);
-		return err;
-	}
-
-	napi_enable(&bp->napi);
-
-	bp->macbgem_ops.mog_init_rings(bp);
-	macb_init_hw(bp);
-
-	/* schedule a link state check */
-	phy_start(dev->phydev);
-
-	netif_tx_start_all_queues(dev);
-
-	if (bp->ptp_info)
-		bp->ptp_info->ptp_init(dev);
-
-	return 0;
-}
-
-static int macb_close(struct net_device *dev)
-{
-	struct macb *bp = netdev_priv(dev);
-	unsigned long flags;
-
-	netif_tx_stop_all_queues(dev);
-	napi_disable(&bp->napi);
-
-	if (dev->phydev)
-		phy_stop(dev->phydev);
-
-	spin_lock_irqsave(&bp->lock, flags);
-	macb_reset_hw(bp);
-	netif_carrier_off(dev);
-	spin_unlock_irqrestore(&bp->lock, flags);
-
-	macb_free_consistent(bp);
-
-	if (bp->ptp_info)
-		bp->ptp_info->ptp_remove(dev);
-
-	return 0;
-}
-
-static int macb_change_mtu(struct net_device *dev, int new_mtu)
-{
-	if (netif_running(dev))
-		return -EBUSY;
-
-	dev->mtu = new_mtu;
-
-	return 0;
-}
-
-static void gem_update_stats(struct macb *bp)
-{
-	unsigned int i;
-	u32 *p = &bp->hw_stats.gem.tx_octets_31_0;
-
-	for (i = 0; i < GEM_STATS_LEN; ++i, ++p) {
-		u32 offset = gem_statistics[i].offset;
-		u64 val = bp->macb_reg_readl(bp, offset);
-
-		bp->ethtool_stats[i] += val;
-		*p += val;
-
-		if (offset == GEM_OCTTXL || offset == GEM_OCTRXL) {
-			/* Add GEM_OCTTXH, GEM_OCTRXH */
-			val = bp->macb_reg_readl(bp, offset + 4);
-			bp->ethtool_stats[i] += ((u64)val) << 32;
-			*(++p) += val;
-		}
-	}
-}
-
-static struct net_device_stats *gem_get_stats(struct macb *bp)
-{
-	struct gem_stats *hwstat = &bp->hw_stats.gem;
-	struct net_device_stats *nstat = &bp->dev->stats;
-
-	gem_update_stats(bp);
-
-	nstat->rx_errors = (hwstat->rx_frame_check_sequence_errors +
-			    hwstat->rx_alignment_errors +
-			    hwstat->rx_resource_errors +
-			    hwstat->rx_overruns +
-			    hwstat->rx_oversize_frames +
-			    hwstat->rx_jabbers +
-			    hwstat->rx_undersized_frames +
-			    hwstat->rx_length_field_frame_errors);
-	nstat->tx_errors = (hwstat->tx_late_collisions +
-			    hwstat->tx_excessive_collisions +
-			    hwstat->tx_underrun +
-			    hwstat->tx_carrier_sense_errors);
-	nstat->multicast = hwstat->rx_multicast_frames;
-	nstat->collisions = (hwstat->tx_single_collision_frames +
-			     hwstat->tx_multiple_collision_frames +
-			     hwstat->tx_excessive_collisions);
-	nstat->rx_length_errors = (hwstat->rx_oversize_frames +
-				   hwstat->rx_jabbers +
-				   hwstat->rx_undersized_frames +
-				   hwstat->rx_length_field_frame_errors);
-	nstat->rx_over_errors = hwstat->rx_resource_errors;
-	nstat->rx_crc_errors = hwstat->rx_frame_check_sequence_errors;
-	nstat->rx_frame_errors = hwstat->rx_alignment_errors;
-	nstat->rx_fifo_errors = hwstat->rx_overruns;
-	nstat->tx_aborted_errors = hwstat->tx_excessive_collisions;
-	nstat->tx_carrier_errors = hwstat->tx_carrier_sense_errors;
-	nstat->tx_fifo_errors = hwstat->tx_underrun;
-
-	return nstat;
-}
-
-static void gem_get_ethtool_stats(struct net_device *dev,
-				  struct ethtool_stats *stats, u64 *data)
-{
-	struct macb *bp;
-
-	bp = netdev_priv(dev);
-	gem_update_stats(bp);
-	memcpy(data, &bp->ethtool_stats, sizeof(u64) * GEM_STATS_LEN);
-}
-
-static int gem_get_sset_count(struct net_device *dev, int sset)
-{
-	switch (sset) {
-	case ETH_SS_STATS:
-		return GEM_STATS_LEN;
-	default:
-		return -EOPNOTSUPP;
-	}
-}
-
-static void gem_get_ethtool_strings(struct net_device *dev, u32 sset, u8 *p)
-{
-	unsigned int i;
-
-	switch (sset) {
-	case ETH_SS_STATS:
-		for (i = 0; i < GEM_STATS_LEN; i++, p += ETH_GSTRING_LEN)
-			memcpy(p, gem_statistics[i].stat_string,
-			       ETH_GSTRING_LEN);
-		break;
-	}
-}
-
-static struct net_device_stats *macb_get_stats(struct net_device *dev)
-{
-	struct macb *bp = netdev_priv(dev);
-	struct net_device_stats *nstat = &bp->dev->stats;
-	struct macb_stats *hwstat = &bp->hw_stats.macb;
-
-	if (macb_is_gem(bp))
-		return gem_get_stats(bp);
-
-	/* read stats from hardware */
-	macb_update_stats(bp);
-
-	/* Convert HW stats into netdevice stats */
-	nstat->rx_errors = (hwstat->rx_fcs_errors +
-			    hwstat->rx_align_errors +
-			    hwstat->rx_resource_errors +
-			    hwstat->rx_overruns +
-			    hwstat->rx_oversize_pkts +
-			    hwstat->rx_jabbers +
-			    hwstat->rx_undersize_pkts +
-			    hwstat->rx_length_mismatch);
-	nstat->tx_errors = (hwstat->tx_late_cols +
-			    hwstat->tx_excessive_cols +
-			    hwstat->tx_underruns +
-			    hwstat->tx_carrier_errors +
-			    hwstat->sqe_test_errors);
-	nstat->collisions = (hwstat->tx_single_cols +
-			     hwstat->tx_multiple_cols +
-			     hwstat->tx_excessive_cols);
-	nstat->rx_length_errors = (hwstat->rx_oversize_pkts +
-				   hwstat->rx_jabbers +
-				   hwstat->rx_undersize_pkts +
-				   hwstat->rx_length_mismatch);
-	nstat->rx_over_errors = hwstat->rx_resource_errors +
-				   hwstat->rx_overruns;
-	nstat->rx_crc_errors = hwstat->rx_fcs_errors;
-	nstat->rx_frame_errors = hwstat->rx_align_errors;
-	nstat->rx_fifo_errors = hwstat->rx_overruns;
-	/* XXX: What does "missed" mean? */
-	nstat->tx_aborted_errors = hwstat->tx_excessive_cols;
-	nstat->tx_carrier_errors = hwstat->tx_carrier_errors;
-	nstat->tx_fifo_errors = hwstat->tx_underruns;
-	/* Don't know about heartbeat or window errors... */
-
-	return nstat;
-}
-
-static int macb_get_regs_len(struct net_device *netdev)
-{
-	return MACB_GREGS_NBR * sizeof(u32);
-}
-
-static void macb_get_regs(struct net_device *dev, struct ethtool_regs *regs,
-			  void *p)
-{
-	struct macb *bp = netdev_priv(dev);
-	unsigned int tail, head;
-	u32 *regs_buff = p;
-
-	regs->version = (macb_readl(bp, MID) & ((1 << MACB_REV_SIZE) - 1))
-			| MACB_GREGS_VERSION;
-
-	tail = macb_tx_ring_wrap(bp, bp->queues[0].tx_tail);
-	head = macb_tx_ring_wrap(bp, bp->queues[0].tx_head);
-
-	regs_buff[0]  = macb_readl(bp, NCR);
-	regs_buff[1]  = macb_or_gem_readl(bp, NCFGR);
-	regs_buff[2]  = macb_readl(bp, NSR);
-	regs_buff[3]  = macb_readl(bp, TSR);
-	regs_buff[4]  = macb_readl(bp, RBQP);
-	regs_buff[5]  = macb_readl(bp, TBQP);
-	regs_buff[6]  = macb_readl(bp, RSR);
-	regs_buff[7]  = macb_readl(bp, IMR);
-
-	regs_buff[8]  = tail;
-	regs_buff[9]  = head;
-	regs_buff[10] = macb_tx_dma(&bp->queues[0], tail);
-	regs_buff[11] = macb_tx_dma(&bp->queues[0], head);
-
-	if (!(bp->caps & MACB_CAPS_USRIO_DISABLED))
-		regs_buff[12] = macb_or_gem_readl(bp, USRIO);
-	if (macb_is_gem(bp))
-		regs_buff[13] = gem_readl(bp, DMACFG);
-}
-
-static void macb_get_wol(struct net_device *netdev, struct ethtool_wolinfo *wol)
-{
-	struct macb *bp = netdev_priv(netdev);
-
-	wol->supported = 0;
-	wol->wolopts = 0;
-
-	if (bp->wol & MACB_WOL_HAS_MAGIC_PACKET) {
-		wol->supported = WAKE_MAGIC;
-
-		if (bp->wol & MACB_WOL_ENABLED)
-			wol->wolopts |= WAKE_MAGIC;
-	}
-}
-
-static int macb_set_wol(struct net_device *netdev, struct ethtool_wolinfo *wol)
-{
-	struct macb *bp = netdev_priv(netdev);
-
-	if (!(bp->wol & MACB_WOL_HAS_MAGIC_PACKET) ||
-	    (wol->wolopts & ~WAKE_MAGIC))
-		return -EOPNOTSUPP;
-
-	if (wol->wolopts & WAKE_MAGIC)
-		bp->wol |= MACB_WOL_ENABLED;
-	else
-		bp->wol &= ~MACB_WOL_ENABLED;
-
-	device_set_wakeup_enable(&bp->pdev->dev, bp->wol & MACB_WOL_ENABLED);
-
-	return 0;
-}
-
-static void macb_get_ringparam(struct net_device *netdev,
-			       struct ethtool_ringparam *ring)
-{
-	struct macb *bp = netdev_priv(netdev);
-
-	ring->rx_max_pending = MAX_RX_RING_SIZE;
-	ring->tx_max_pending = MAX_TX_RING_SIZE;
-
-	ring->rx_pending = bp->rx_ring_size;
-	ring->tx_pending = bp->tx_ring_size;
-}
-
-static int macb_set_ringparam(struct net_device *netdev,
-			      struct ethtool_ringparam *ring)
-{
-	struct macb *bp = netdev_priv(netdev);
-	u32 new_rx_size, new_tx_size;
-	unsigned int reset = 0;
-
-	if ((ring->rx_mini_pending) || (ring->rx_jumbo_pending))
-		return -EINVAL;
-
-	new_rx_size = clamp_t(u32, ring->rx_pending,
-			      MIN_RX_RING_SIZE, MAX_RX_RING_SIZE);
-	new_rx_size = roundup_pow_of_two(new_rx_size);
-
-	new_tx_size = clamp_t(u32, ring->tx_pending,
-			      MIN_TX_RING_SIZE, MAX_TX_RING_SIZE);
-	new_tx_size = roundup_pow_of_two(new_tx_size);
-
-	if ((new_tx_size == bp->tx_ring_size) &&
-	    (new_rx_size == bp->rx_ring_size)) {
-		/* nothing to do */
-		return 0;
-	}
-
-	if (netif_running(bp->dev)) {
-		reset = 1;
-		macb_close(bp->dev);
-	}
-
-	bp->rx_ring_size = new_rx_size;
-	bp->tx_ring_size = new_tx_size;
-
-	if (reset)
-		macb_open(bp->dev);
-
-	return 0;
-}
-
-static int macb_get_ts_info(struct net_device *netdev,
-			    struct ethtool_ts_info *info)
-{
-	struct macb *bp = netdev_priv(netdev);
-
-	if (bp->ptp_info)
-		return bp->ptp_info->get_ts_info(netdev, info);
-
-	return ethtool_op_get_ts_info(netdev, info);
-}
-
-static const struct ethtool_ops macb_ethtool_ops = {
-	.get_regs_len		= macb_get_regs_len,
-	.get_regs		= macb_get_regs,
-	.get_link		= ethtool_op_get_link,
-	.get_ts_info		= ethtool_op_get_ts_info,
-	.get_wol		= macb_get_wol,
-	.set_wol		= macb_set_wol,
-	.get_link_ksettings     = phy_ethtool_get_link_ksettings,
-	.set_link_ksettings     = phy_ethtool_set_link_ksettings,
-	.get_ringparam		= macb_get_ringparam,
-	.set_ringparam		= macb_set_ringparam,
-};
-
-static const struct ethtool_ops gem_ethtool_ops = {
-	.get_regs_len		= macb_get_regs_len,
-	.get_regs		= macb_get_regs,
-	.get_link		= ethtool_op_get_link,
-	.get_ts_info		= macb_get_ts_info,
-	.get_ethtool_stats	= gem_get_ethtool_stats,
-	.get_strings		= gem_get_ethtool_strings,
-	.get_sset_count		= gem_get_sset_count,
-	.get_link_ksettings     = phy_ethtool_get_link_ksettings,
-	.set_link_ksettings     = phy_ethtool_set_link_ksettings,
-	.get_ringparam		= macb_get_ringparam,
-	.set_ringparam		= macb_set_ringparam,
-};
-
-static int macb_ioctl(struct net_device *dev, struct ifreq *rq, int cmd)
-{
-	struct phy_device *phydev = dev->phydev;
-	struct macb *bp = netdev_priv(dev);
-
-	if (!netif_running(dev))
-		return -EINVAL;
-
-	if (!phydev)
-		return -ENODEV;
-
-	if (!bp->ptp_info)
-		return phy_mii_ioctl(phydev, rq, cmd);
-
-	switch (cmd) {
-	case SIOCSHWTSTAMP:
-		return bp->ptp_info->set_hwtst(dev, rq, cmd);
-	case SIOCGHWTSTAMP:
-		return bp->ptp_info->get_hwtst(dev, rq);
-	default:
-		return phy_mii_ioctl(phydev, rq, cmd);
-	}
-}
-
-static int macb_set_features(struct net_device *netdev,
-			     netdev_features_t features)
-{
-	struct macb *bp = netdev_priv(netdev);
-	netdev_features_t changed = features ^ netdev->features;
-
-	/* TX checksum offload */
-	if ((changed & NETIF_F_HW_CSUM) && macb_is_gem(bp)) {
-		u32 dmacfg;
-
-		dmacfg = gem_readl(bp, DMACFG);
-		if (features & NETIF_F_HW_CSUM)
-			dmacfg |= GEM_BIT(TXCOEN);
-		else
-			dmacfg &= ~GEM_BIT(TXCOEN);
-		gem_writel(bp, DMACFG, dmacfg);
-	}
-
-	/* RX checksum offload */
-	if ((changed & NETIF_F_RXCSUM) && macb_is_gem(bp)) {
-		u32 netcfg;
-
-		netcfg = gem_readl(bp, NCFGR);
-		if (features & NETIF_F_RXCSUM &&
-		    !(netdev->flags & IFF_PROMISC))
-			netcfg |= GEM_BIT(RXCOEN);
-		else
-			netcfg &= ~GEM_BIT(RXCOEN);
-		gem_writel(bp, NCFGR, netcfg);
-	}
-
-	return 0;
-}
-
-static const struct net_device_ops macb_netdev_ops = {
-	.ndo_open		= macb_open,
-	.ndo_stop		= macb_close,
-	.ndo_start_xmit		= macb_start_xmit,
-	.ndo_set_rx_mode	= macb_set_rx_mode,
-	.ndo_get_stats		= macb_get_stats,
-	.ndo_do_ioctl		= macb_ioctl,
-	.ndo_validate_addr	= eth_validate_addr,
-	.ndo_change_mtu		= macb_change_mtu,
-	.ndo_set_mac_address	= eth_mac_addr,
-#ifdef CONFIG_NET_POLL_CONTROLLER
-	.ndo_poll_controller	= macb_poll_controller,
-#endif
-	.ndo_set_features	= macb_set_features,
-	.ndo_features_check	= macb_features_check,
-};
-
-/* Configure peripheral capabilities according to device tree
- * and integration options used
- */
-static void macb_configure_caps(struct macb *bp,
-				const struct macb_config *dt_conf)
-{
-	u32 dcfg;
-
-	if (dt_conf)
-		bp->caps = dt_conf->caps;
-
-	if (hw_is_gem(bp->regs, bp->native_io)) {
-		bp->caps |= MACB_CAPS_MACB_IS_GEM;
-
-		dcfg = gem_readl(bp, DCFG1);
-		if (GEM_BFEXT(IRQCOR, dcfg) == 0)
-			bp->caps |= MACB_CAPS_ISR_CLEAR_ON_WRITE;
-		dcfg = gem_readl(bp, DCFG2);
-		if ((dcfg & (GEM_BIT(RX_PKT_BUFF) | GEM_BIT(TX_PKT_BUFF))) == 0)
-			bp->caps |= MACB_CAPS_FIFO_MODE;
-	}
-
-	dev_dbg(&bp->pdev->dev, "Cadence caps 0x%08x\n", bp->caps);
-}
-
-static void macb_probe_queues(void __iomem *mem,
-			      bool native_io,
-			      unsigned int *queue_mask,
-			      unsigned int *num_queues)
-{
-	unsigned int hw_q;
-
-	*queue_mask = 0x1;
-	*num_queues = 1;
-
-	/* is it macb or gem ?
-	 *
-	 * We need to read directly from the hardware here because
-	 * we are early in the probe process and don't have the
-	 * MACB_CAPS_MACB_IS_GEM flag positioned
-	 */
-	if (!hw_is_gem(mem, native_io))
-		return;
-
-	/* bit 0 is never set but queue 0 always exists */
-	*queue_mask = readl_relaxed(mem + GEM_DCFG6) & 0xff;
-
-	*queue_mask |= 0x1;
-
-	for (hw_q = 1; hw_q < MACB_MAX_QUEUES; ++hw_q)
-		if (*queue_mask & (1 << hw_q))
-			(*num_queues)++;
-}
-
-static int macb_clk_init(struct platform_device *pdev, struct clk **pclk,
-			 struct clk **hclk, struct clk **tx_clk,
-			 struct clk **rx_clk)
-{
-	struct macb_platform_data *pdata;
-	int err;
-
-	pdata = dev_get_platdata(&pdev->dev);
-	if (pdata) {
-		*pclk = pdata->pclk;
-		*hclk = pdata->hclk;
-	} else {
-		*pclk = devm_clk_get(&pdev->dev, "pclk");
-		*hclk = devm_clk_get(&pdev->dev, "hclk");
-	}
-
-	if (IS_ERR(*pclk)) {
-		err = PTR_ERR(*pclk);
-		dev_err(&pdev->dev, "failed to get macb_clk (%u)\n", err);
-		return err;
-	}
-
-	if (IS_ERR(*hclk)) {
-		err = PTR_ERR(*hclk);
-		dev_err(&pdev->dev, "failed to get hclk (%u)\n", err);
-		return err;
-	}
-
-	*tx_clk = devm_clk_get(&pdev->dev, "tx_clk");
-	if (IS_ERR(*tx_clk))
-		*tx_clk = NULL;
-
-	*rx_clk = devm_clk_get(&pdev->dev, "rx_clk");
-	if (IS_ERR(*rx_clk))
-		*rx_clk = NULL;
-
-	err = clk_prepare_enable(*pclk);
-	if (err) {
-		dev_err(&pdev->dev, "failed to enable pclk (%u)\n", err);
-		return err;
-	}
-
-	err = clk_prepare_enable(*hclk);
-	if (err) {
-		dev_err(&pdev->dev, "failed to enable hclk (%u)\n", err);
-		goto err_disable_pclk;
-	}
-
-	err = clk_prepare_enable(*tx_clk);
-	if (err) {
-		dev_err(&pdev->dev, "failed to enable tx_clk (%u)\n", err);
-		goto err_disable_hclk;
-	}
-
-	err = clk_prepare_enable(*rx_clk);
-	if (err) {
-		dev_err(&pdev->dev, "failed to enable rx_clk (%u)\n", err);
-		goto err_disable_txclk;
-	}
-
-	return 0;
-
-err_disable_txclk:
-	clk_disable_unprepare(*tx_clk);
-
-err_disable_hclk:
-	clk_disable_unprepare(*hclk);
-
-err_disable_pclk:
-	clk_disable_unprepare(*pclk);
-
-	return err;
-}
-
-static int macb_init(struct platform_device *pdev)
-{
-	struct net_device *dev = platform_get_drvdata(pdev);
-	unsigned int hw_q, q;
-	struct macb *bp = netdev_priv(dev);
-	struct macb_queue *queue;
-	int err;
-	u32 val;
-
-	bp->tx_ring_size = DEFAULT_TX_RING_SIZE;
-	bp->rx_ring_size = DEFAULT_RX_RING_SIZE;
-
-	/* set the queue register mapping once for all: queue0 has a special
-	 * register mapping but we don't want to test the queue index then
-	 * compute the corresponding register offset at run time.
-	 */
-	for (hw_q = 0, q = 0; hw_q < MACB_MAX_QUEUES; ++hw_q) {
-		if (!(bp->queue_mask & (1 << hw_q)))
-			continue;
-
-		queue = &bp->queues[q];
-		queue->bp = bp;
-		if (hw_q) {
-			queue->ISR  = GEM_ISR(hw_q - 1);
-			queue->IER  = GEM_IER(hw_q - 1);
-			queue->IDR  = GEM_IDR(hw_q - 1);
-			queue->IMR  = GEM_IMR(hw_q - 1);
-			queue->TBQP = GEM_TBQP(hw_q - 1);
-#ifdef CONFIG_ARCH_DMA_ADDR_T_64BIT
-			if (bp->hw_dma_cap == HW_DMA_CAP_64B)
-				queue->TBQPH = GEM_TBQPH(hw_q - 1);
-#endif
-		} else {
-			/* queue0 uses legacy registers */
-			queue->ISR  = MACB_ISR;
-			queue->IER  = MACB_IER;
-			queue->IDR  = MACB_IDR;
-			queue->IMR  = MACB_IMR;
-			queue->TBQP = MACB_TBQP;
-#ifdef CONFIG_ARCH_DMA_ADDR_T_64BIT
-			if (bp->hw_dma_cap == HW_DMA_CAP_64B)
-				queue->TBQPH = MACB_TBQPH;
-#endif
-		}
-
-		/* get irq: here we use the linux queue index, not the hardware
-		 * queue index. the queue irq definitions in the device tree
-		 * must remove the optional gaps that could exist in the
-		 * hardware queue mask.
-		 */
-		queue->irq = platform_get_irq(pdev, q);
-		err = devm_request_irq(&pdev->dev, queue->irq, macb_interrupt,
-				       IRQF_SHARED, dev->name, queue);
-		if (err) {
-			dev_err(&pdev->dev,
-				"Unable to request IRQ %d (error %d)\n",
-				queue->irq, err);
-			return err;
-		}
-
-		INIT_WORK(&queue->tx_error_task, macb_tx_error_task);
-		q++;
-	}
-
-	dev->netdev_ops = &macb_netdev_ops;
-	netif_napi_add(dev, &bp->napi, macb_poll, 64);
-
-	/* setup appropriated routines according to adapter type */
-	if (macb_is_gem(bp)) {
-		bp->max_tx_length = GEM_MAX_TX_LEN;
-		bp->macbgem_ops.mog_alloc_rx_buffers = gem_alloc_rx_buffers;
-		bp->macbgem_ops.mog_free_rx_buffers = gem_free_rx_buffers;
-		bp->macbgem_ops.mog_init_rings = gem_init_rings;
-		bp->macbgem_ops.mog_rx = gem_rx;
-		dev->ethtool_ops = &gem_ethtool_ops;
-	} else {
-		bp->max_tx_length = MACB_MAX_TX_LEN;
-		bp->macbgem_ops.mog_alloc_rx_buffers = macb_alloc_rx_buffers;
-		bp->macbgem_ops.mog_free_rx_buffers = macb_free_rx_buffers;
-		bp->macbgem_ops.mog_init_rings = macb_init_rings;
-		bp->macbgem_ops.mog_rx = macb_rx;
-		dev->ethtool_ops = &macb_ethtool_ops;
-	}
-
-	/* Set features */
-	dev->hw_features = NETIF_F_SG;
-
-	/* Check LSO capability */
-	if (GEM_BFEXT(PBUF_LSO, gem_readl(bp, DCFG6)))
-		dev->hw_features |= MACB_NETIF_LSO;
-
-	/* Checksum offload is only available on gem with packet buffer */
-	if (macb_is_gem(bp) && !(bp->caps & MACB_CAPS_FIFO_MODE))
-		dev->hw_features |= NETIF_F_HW_CSUM | NETIF_F_RXCSUM;
-	if (bp->caps & MACB_CAPS_SG_DISABLED)
-		dev->hw_features &= ~NETIF_F_SG;
-	dev->features = dev->hw_features;
-
-	if (!(bp->caps & MACB_CAPS_USRIO_DISABLED)) {
-		val = 0;
-		if (bp->phy_interface == PHY_INTERFACE_MODE_RGMII)
-			val = GEM_BIT(RGMII);
-		else if (bp->phy_interface == PHY_INTERFACE_MODE_RMII &&
-			 (bp->caps & MACB_CAPS_USRIO_DEFAULT_IS_MII_GMII))
-			val = MACB_BIT(RMII);
-		else if (!(bp->caps & MACB_CAPS_USRIO_DEFAULT_IS_MII_GMII))
-			val = MACB_BIT(MII);
-
-		if (bp->caps & MACB_CAPS_USRIO_HAS_CLKEN)
-			val |= MACB_BIT(CLKEN);
-
-		macb_or_gem_writel(bp, USRIO, val);
-	}
-
-	/* Set MII management clock divider */
-	val = macb_mdc_clk_div(bp);
-	val |= macb_dbw(bp);
-	if (bp->phy_interface == PHY_INTERFACE_MODE_SGMII)
-		val |= GEM_BIT(SGMIIEN) | GEM_BIT(PCSSEL);
-	macb_writel(bp, NCFGR, val);
-
-	return 0;
-}
-
-#if defined(CONFIG_OF)
-/* 1518 rounded up */
-#define AT91ETHER_MAX_RBUFF_SZ	0x600
-/* max number of receive buffers */
-#define AT91ETHER_MAX_RX_DESCR	9
-
-/* Initialize and start the Receiver and Transmit subsystems */
-static int at91ether_start(struct net_device *dev)
-{
-	struct macb *lp = netdev_priv(dev);
-	struct macb_dma_desc *desc;
-	dma_addr_t addr;
-	u32 ctl;
-	int i;
-
-	lp->rx_ring = dma_alloc_coherent(&lp->pdev->dev,
-					 (AT91ETHER_MAX_RX_DESCR *
-					  macb_dma_desc_get_size(lp)),
-					 &lp->rx_ring_dma, GFP_KERNEL);
-	if (!lp->rx_ring)
-		return -ENOMEM;
-
-	lp->rx_buffers = dma_alloc_coherent(&lp->pdev->dev,
-					    AT91ETHER_MAX_RX_DESCR *
-					    AT91ETHER_MAX_RBUFF_SZ,
-					    &lp->rx_buffers_dma, GFP_KERNEL);
-	if (!lp->rx_buffers) {
-		dma_free_coherent(&lp->pdev->dev,
-				  AT91ETHER_MAX_RX_DESCR *
-				  macb_dma_desc_get_size(lp),
-				  lp->rx_ring, lp->rx_ring_dma);
-		lp->rx_ring = NULL;
-		return -ENOMEM;
-	}
-
-	addr = lp->rx_buffers_dma;
-	for (i = 0; i < AT91ETHER_MAX_RX_DESCR; i++) {
-		desc = macb_rx_desc(lp, i);
-		macb_set_addr(lp, desc, addr);
-		desc->ctrl = 0;
-		addr += AT91ETHER_MAX_RBUFF_SZ;
-	}
-
-	/* Set the Wrap bit on the last descriptor */
-	desc->addr |= MACB_BIT(RX_WRAP);
-
-	/* Reset buffer index */
-	lp->rx_tail = 0;
-
-	/* Program address of descriptor list in Rx Buffer Queue register */
-	macb_writel(lp, RBQP, lp->rx_ring_dma);
-
-	/* Enable Receive and Transmit */
-	ctl = macb_readl(lp, NCR);
-	macb_writel(lp, NCR, ctl | MACB_BIT(RE) | MACB_BIT(TE));
-
-	return 0;
-}
-
-/* Open the ethernet interface */
-static int at91ether_open(struct net_device *dev)
-{
-	struct macb *lp = netdev_priv(dev);
-	u32 ctl;
-	int ret;
-
-	/* Clear internal statistics */
-	ctl = macb_readl(lp, NCR);
-	macb_writel(lp, NCR, ctl | MACB_BIT(CLRSTAT));
-
-	macb_set_hwaddr(lp);
-
-	ret = at91ether_start(dev);
-	if (ret)
-		return ret;
-
-	/* Enable MAC interrupts */
-	macb_writel(lp, IER, MACB_BIT(RCOMP)	|
-			     MACB_BIT(RXUBR)	|
-			     MACB_BIT(ISR_TUND)	|
-			     MACB_BIT(ISR_RLE)	|
-			     MACB_BIT(TCOMP)	|
-			     MACB_BIT(ISR_ROVR)	|
-			     MACB_BIT(HRESP));
-
-	/* schedule a link state check */
-	phy_start(dev->phydev);
-
-	netif_start_queue(dev);
-
-	return 0;
-}
-
-/* Close the interface */
-static int at91ether_close(struct net_device *dev)
-{
-	struct macb *lp = netdev_priv(dev);
-	u32 ctl;
-
-	/* Disable Receiver and Transmitter */
-	ctl = macb_readl(lp, NCR);
-	macb_writel(lp, NCR, ctl & ~(MACB_BIT(TE) | MACB_BIT(RE)));
-
-	/* Disable MAC interrupts */
-	macb_writel(lp, IDR, MACB_BIT(RCOMP)	|
-			     MACB_BIT(RXUBR)	|
-			     MACB_BIT(ISR_TUND)	|
-			     MACB_BIT(ISR_RLE)	|
-			     MACB_BIT(TCOMP)	|
-			     MACB_BIT(ISR_ROVR) |
-			     MACB_BIT(HRESP));
-
-	netif_stop_queue(dev);
-
-	dma_free_coherent(&lp->pdev->dev,
-			  AT91ETHER_MAX_RX_DESCR *
-			  macb_dma_desc_get_size(lp),
-			  lp->rx_ring, lp->rx_ring_dma);
-	lp->rx_ring = NULL;
-
-	dma_free_coherent(&lp->pdev->dev,
-			  AT91ETHER_MAX_RX_DESCR * AT91ETHER_MAX_RBUFF_SZ,
-			  lp->rx_buffers, lp->rx_buffers_dma);
-	lp->rx_buffers = NULL;
-
-	return 0;
-}
-
-/* Transmit packet */
-static int at91ether_start_xmit(struct sk_buff *skb, struct net_device *dev)
-{
-	struct macb *lp = netdev_priv(dev);
-
-	if (macb_readl(lp, TSR) & MACB_BIT(RM9200_BNQ)) {
-		netif_stop_queue(dev);
-
-		/* Store packet information (to free when Tx completed) */
-		lp->skb = skb;
-		lp->skb_length = skb->len;
-		lp->skb_physaddr = dma_map_single(NULL, skb->data, skb->len,
-							DMA_TO_DEVICE);
-		if (dma_mapping_error(NULL, lp->skb_physaddr)) {
-			dev_kfree_skb_any(skb);
-			dev->stats.tx_dropped++;
-			netdev_err(dev, "%s: DMA mapping error\n", __func__);
-			return NETDEV_TX_OK;
-		}
-
-		/* Set address of the data in the Transmit Address register */
-		macb_writel(lp, TAR, lp->skb_physaddr);
-		/* Set length of the packet in the Transmit Control register */
-		macb_writel(lp, TCR, skb->len);
-
-	} else {
-		netdev_err(dev, "%s called, but device is busy!\n", __func__);
-		return NETDEV_TX_BUSY;
-	}
-
-	return NETDEV_TX_OK;
-}
-
-/* Extract received frame from buffer descriptors and sent to upper layers.
- * (Called from interrupt context)
- */
-static void at91ether_rx(struct net_device *dev)
-{
-	struct macb *lp = netdev_priv(dev);
-	struct macb_dma_desc *desc;
-	unsigned char *p_recv;
-	struct sk_buff *skb;
-	unsigned int pktlen;
-
-	desc = macb_rx_desc(lp, lp->rx_tail);
-	while (desc->addr & MACB_BIT(RX_USED)) {
-		p_recv = lp->rx_buffers + lp->rx_tail * AT91ETHER_MAX_RBUFF_SZ;
-		pktlen = MACB_BF(RX_FRMLEN, desc->ctrl);
-		skb = netdev_alloc_skb(dev, pktlen + 2);
-		if (skb) {
-			skb_reserve(skb, 2);
-			memcpy(skb_put(skb, pktlen), p_recv, pktlen);
-
-			skb->protocol = eth_type_trans(skb, dev);
-			dev->stats.rx_packets++;
-			dev->stats.rx_bytes += pktlen;
-			netif_rx(skb);
-		} else {
-			dev->stats.rx_dropped++;
-		}
-
-		if (desc->ctrl & MACB_BIT(RX_MHASH_MATCH))
-			dev->stats.multicast++;
-
-		/* reset ownership bit */
-		desc->addr &= ~MACB_BIT(RX_USED);
-
-		/* wrap after last buffer */
-		if (lp->rx_tail == AT91ETHER_MAX_RX_DESCR - 1)
-			lp->rx_tail = 0;
-		else
-			lp->rx_tail++;
-
-		desc = macb_rx_desc(lp, lp->rx_tail);
-	}
-}
-
-/* MAC interrupt handler */
-static irqreturn_t at91ether_interrupt(int irq, void *dev_id)
-{
-	struct net_device *dev = dev_id;
-	struct macb *lp = netdev_priv(dev);
-	u32 intstatus, ctl;
-
-	/* MAC Interrupt Status register indicates what interrupts are pending.
-	 * It is automatically cleared once read.
-	 */
-	intstatus = macb_readl(lp, ISR);
-
-	/* Receive complete */
-	if (intstatus & MACB_BIT(RCOMP))
-		at91ether_rx(dev);
-
-	/* Transmit complete */
-	if (intstatus & MACB_BIT(TCOMP)) {
-		/* The TCOM bit is set even if the transmission failed */
-		if (intstatus & (MACB_BIT(ISR_TUND) | MACB_BIT(ISR_RLE)))
-			dev->stats.tx_errors++;
-
-		if (lp->skb) {
-			dev_kfree_skb_irq(lp->skb);
-			lp->skb = NULL;
-			dma_unmap_single(NULL, lp->skb_physaddr,
-					 lp->skb_length, DMA_TO_DEVICE);
-			dev->stats.tx_packets++;
-			dev->stats.tx_bytes += lp->skb_length;
-		}
-		netif_wake_queue(dev);
-	}
-
-	/* Work-around for EMAC Errata section 41.3.1 */
-	if (intstatus & MACB_BIT(RXUBR)) {
-		ctl = macb_readl(lp, NCR);
-		macb_writel(lp, NCR, ctl & ~MACB_BIT(RE));
-		wmb();
-		macb_writel(lp, NCR, ctl | MACB_BIT(RE));
-	}
-
-	if (intstatus & MACB_BIT(ISR_ROVR))
-		netdev_err(dev, "ROVR error\n");
-
-	return IRQ_HANDLED;
-}
-
-#ifdef CONFIG_NET_POLL_CONTROLLER
-static void at91ether_poll_controller(struct net_device *dev)
-{
-	unsigned long flags;
-
-	local_irq_save(flags);
-	at91ether_interrupt(dev->irq, dev);
-	local_irq_restore(flags);
-}
-#endif
-
-static const struct net_device_ops at91ether_netdev_ops = {
-	.ndo_open		= at91ether_open,
-	.ndo_stop		= at91ether_close,
-	.ndo_start_xmit		= at91ether_start_xmit,
-	.ndo_get_stats		= macb_get_stats,
-	.ndo_set_rx_mode	= macb_set_rx_mode,
-	.ndo_set_mac_address	= eth_mac_addr,
-	.ndo_do_ioctl		= macb_ioctl,
-	.ndo_validate_addr	= eth_validate_addr,
-#ifdef CONFIG_NET_POLL_CONTROLLER
-	.ndo_poll_controller	= at91ether_poll_controller,
-#endif
-};
-
-static int at91ether_clk_init(struct platform_device *pdev, struct clk **pclk,
-			      struct clk **hclk, struct clk **tx_clk,
-			      struct clk **rx_clk)
-{
-	int err;
-
-	*hclk = NULL;
-	*tx_clk = NULL;
-	*rx_clk = NULL;
-
-	*pclk = devm_clk_get(&pdev->dev, "ether_clk");
-	if (IS_ERR(*pclk))
-		return PTR_ERR(*pclk);
-
-	err = clk_prepare_enable(*pclk);
-	if (err) {
-		dev_err(&pdev->dev, "failed to enable pclk (%u)\n", err);
-		return err;
-	}
-
-	return 0;
-}
-
-static int at91ether_init(struct platform_device *pdev)
-{
-	struct net_device *dev = platform_get_drvdata(pdev);
-	struct macb *bp = netdev_priv(dev);
-	int err;
-	u32 reg;
-
-	dev->netdev_ops = &at91ether_netdev_ops;
-	dev->ethtool_ops = &macb_ethtool_ops;
-
-	err = devm_request_irq(&pdev->dev, dev->irq, at91ether_interrupt,
-			       0, dev->name, dev);
-	if (err)
-		return err;
-
-	macb_writel(bp, NCR, 0);
-
-	reg = MACB_BF(CLK, MACB_CLK_DIV32) | MACB_BIT(BIG);
-	if (bp->phy_interface == PHY_INTERFACE_MODE_RMII)
-		reg |= MACB_BIT(RM9200_RMII);
-
-	macb_writel(bp, NCFGR, reg);
-
-	return 0;
-}
-
-static const struct macb_config at91sam9260_config = {
-	.caps = MACB_CAPS_USRIO_HAS_CLKEN | MACB_CAPS_USRIO_DEFAULT_IS_MII_GMII,
-	.clk_init = macb_clk_init,
-	.init = macb_init,
-};
-
-static const struct macb_config pc302gem_config = {
-	.caps = MACB_CAPS_SG_DISABLED | MACB_CAPS_GIGABIT_MODE_AVAILABLE,
-	.dma_burst_length = 16,
-	.clk_init = macb_clk_init,
-	.init = macb_init,
-};
-
-static const struct macb_config sama5d2_config = {
-	.caps = MACB_CAPS_USRIO_DEFAULT_IS_MII_GMII,
-	.dma_burst_length = 16,
-	.clk_init = macb_clk_init,
-	.init = macb_init,
-};
-
-static const struct macb_config sama5d3_config = {
-	.caps = MACB_CAPS_SG_DISABLED | MACB_CAPS_GIGABIT_MODE_AVAILABLE
-	      | MACB_CAPS_USRIO_DEFAULT_IS_MII_GMII,
-	.dma_burst_length = 16,
-	.clk_init = macb_clk_init,
-	.init = macb_init,
-};
-
-static const struct macb_config sama5d4_config = {
-	.caps = MACB_CAPS_USRIO_DEFAULT_IS_MII_GMII,
-	.dma_burst_length = 4,
-	.clk_init = macb_clk_init,
-	.init = macb_init,
-};
-
-static const struct macb_config emac_config = {
-	.clk_init = at91ether_clk_init,
-	.init = at91ether_init,
-};
-
-static const struct macb_config np4_config = {
-	.caps = MACB_CAPS_USRIO_DISABLED,
-	.clk_init = macb_clk_init,
-	.init = macb_init,
-};
-
-static const struct macb_config zynqmp_config = {
-	.caps = MACB_CAPS_GIGABIT_MODE_AVAILABLE | MACB_CAPS_JUMBO,
-	.dma_burst_length = 16,
-	.clk_init = macb_clk_init,
-	.init = macb_init,
-	.jumbo_max_len = 10240,
-};
-
-static const struct macb_config zynq_config = {
-	.caps = MACB_CAPS_GIGABIT_MODE_AVAILABLE | MACB_CAPS_NO_GIGABIT_HALF,
-	.dma_burst_length = 16,
-	.clk_init = macb_clk_init,
-	.init = macb_init,
-};
-
-static const struct of_device_id macb_dt_ids[] = {
-	{ .compatible = "cdns,at32ap7000-macb" },
-	{ .compatible = "cdns,at91sam9260-macb", .data = &at91sam9260_config },
-	{ .compatible = "cdns,macb" },
-	{ .compatible = "cdns,np4-macb", .data = &np4_config },
-	{ .compatible = "cdns,pc302-gem", .data = &pc302gem_config },
-	{ .compatible = "cdns,gem", .data = &pc302gem_config },
-	{ .compatible = "atmel,sama5d2-gem", .data = &sama5d2_config },
-	{ .compatible = "atmel,sama5d3-gem", .data = &sama5d3_config },
-	{ .compatible = "atmel,sama5d4-gem", .data = &sama5d4_config },
-	{ .compatible = "cdns,at91rm9200-emac", .data = &emac_config },
-	{ .compatible = "cdns,emac", .data = &emac_config },
-	{ .compatible = "cdns,zynqmp-gem", .data = &zynqmp_config},
-	{ .compatible = "cdns,zynq-gem", .data = &zynq_config },
-	{ /* sentinel */ }
-};
-MODULE_DEVICE_TABLE(of, macb_dt_ids);
-#endif /* CONFIG_OF */
-
-static const struct macb_config default_gem_config = {
-	.caps = MACB_CAPS_GIGABIT_MODE_AVAILABLE | MACB_CAPS_JUMBO,
-	.dma_burst_length = 16,
-	.clk_init = macb_clk_init,
-	.init = macb_init,
-	.jumbo_max_len = 10240,
-};
-
-static int macb_probe(struct platform_device *pdev)
-{
-	const struct macb_config *macb_config = &default_gem_config;
-	int (*clk_init)(struct platform_device *, struct clk **,
-			struct clk **, struct clk **,  struct clk **)
-					      = macb_config->clk_init;
-	int (*init)(struct platform_device *) = macb_config->init;
-	struct device_node *np = pdev->dev.of_node;
-	struct device_node *phy_node;
-	struct clk *pclk, *hclk = NULL, *tx_clk = NULL, *rx_clk = NULL;
-	unsigned int queue_mask, num_queues;
-	struct macb_platform_data *pdata;
-	bool native_io;
-	struct phy_device *phydev;
-	struct net_device *dev;
-	struct resource *regs;
-	void __iomem *mem;
-	const char *mac;
-	struct macb *bp;
-	int err;
-
-	regs = platform_get_resource(pdev, IORESOURCE_MEM, 0);
-	mem = devm_ioremap_resource(&pdev->dev, regs);
-	if (IS_ERR(mem))
-		return PTR_ERR(mem);
-
-	if (np) {
-		const struct of_device_id *match;
-
-		match = of_match_node(macb_dt_ids, np);
-		if (match && match->data) {
-			macb_config = match->data;
-			clk_init = macb_config->clk_init;
-			init = macb_config->init;
-		}
-	}
-
-	err = clk_init(pdev, &pclk, &hclk, &tx_clk, &rx_clk);
-	if (err)
-		return err;
-
-	native_io = hw_is_native_io(mem);
-
-	macb_probe_queues(mem, native_io, &queue_mask, &num_queues);
-	dev = alloc_etherdev_mq(sizeof(*bp), num_queues);
-	if (!dev) {
-		err = -ENOMEM;
-		goto err_disable_clocks;
-	}
-
-	dev->base_addr = regs->start;
-
-	SET_NETDEV_DEV(dev, &pdev->dev);
-
-	bp = netdev_priv(dev);
-	bp->pdev = pdev;
-	bp->dev = dev;
-	bp->regs = mem;
-	bp->native_io = native_io;
-	if (native_io) {
-		bp->macb_reg_readl = hw_readl_native;
-		bp->macb_reg_writel = hw_writel_native;
-	} else {
-		bp->macb_reg_readl = hw_readl;
-		bp->macb_reg_writel = hw_writel;
-	}
-	bp->num_queues = num_queues;
-	bp->queue_mask = queue_mask;
-	if (macb_config)
-		bp->dma_burst_length = macb_config->dma_burst_length;
-	bp->pclk = pclk;
-	bp->hclk = hclk;
-	bp->tx_clk = tx_clk;
-	bp->rx_clk = rx_clk;
-	if (macb_config)
-		bp->jumbo_max_len = macb_config->jumbo_max_len;
-
-	bp->wol = 0;
-	if (of_get_property(np, "magic-packet", NULL))
-		bp->wol |= MACB_WOL_HAS_MAGIC_PACKET;
-	device_init_wakeup(&pdev->dev, bp->wol & MACB_WOL_HAS_MAGIC_PACKET);
-
-#ifdef CONFIG_ARCH_DMA_ADDR_T_64BIT
-	if (GEM_BFEXT(DAW64, gem_readl(bp, DCFG6))) {
-		dma_set_mask(&pdev->dev, DMA_BIT_MASK(44));
-		bp->hw_dma_cap = HW_DMA_CAP_64B;
-	} else
-		bp->hw_dma_cap = HW_DMA_CAP_32B;
-#endif
-
-	spin_lock_init(&bp->lock);
-
-	/* setup capabilities */
-	macb_configure_caps(bp, macb_config);
-
-	platform_set_drvdata(pdev, dev);
-
-	dev->irq = platform_get_irq(pdev, 0);
-	if (dev->irq < 0) {
-		err = dev->irq;
-		goto err_out_free_netdev;
-	}
-
-	/* MTU range: 68 - 1500 or 10240 */
-	dev->min_mtu = GEM_MTU_MIN_SIZE;
-	if (bp->caps & MACB_CAPS_JUMBO)
-		dev->max_mtu = gem_readl(bp, JML) - ETH_HLEN - ETH_FCS_LEN;
-	else
-		dev->max_mtu = ETH_DATA_LEN;
-
-	mac = of_get_mac_address(np);
-	if (mac)
-		ether_addr_copy(bp->dev->dev_addr, mac);
-	else
-		macb_get_hwaddr(bp);
-
-	/* Power up the PHY if there is a GPIO reset */
-	phy_node =  of_get_next_available_child(np, NULL);
-	if (phy_node) {
-		int gpio = of_get_named_gpio(phy_node, "reset-gpios", 0);
-
-		if (gpio_is_valid(gpio)) {
-			bp->reset_gpio = gpio_to_desc(gpio);
-			gpiod_direction_output(bp->reset_gpio, 1);
-		}
-	}
-	of_node_put(phy_node);
-
-	err = of_get_phy_mode(np);
-	if (err < 0) {
-		pdata = dev_get_platdata(&pdev->dev);
-		if (pdata && pdata->is_rmii)
-			bp->phy_interface = PHY_INTERFACE_MODE_RMII;
-		else
-			bp->phy_interface = PHY_INTERFACE_MODE_MII;
-	} else {
-		bp->phy_interface = err;
-	}
-
-	/* IP specific init */
-	err = init(pdev);
-	if (err)
-		goto err_out_free_netdev;
-
-	err = macb_mii_init(bp);
-	if (err)
-		goto err_out_free_netdev;
-
-	phydev = dev->phydev;
-
-	netif_carrier_off(dev);
-
-	err = register_netdev(dev);
-	if (err) {
-		dev_err(&pdev->dev, "Cannot register net device, aborting.\n");
-		goto err_out_unregister_mdio;
-	}
-
-	phy_attached_info(phydev);
-
-	netdev_info(dev, "Cadence %s rev 0x%08x at 0x%08lx irq %d (%pM)\n",
-		    macb_is_gem(bp) ? "GEM" : "MACB", macb_readl(bp, MID),
-		    dev->base_addr, dev->irq, dev->dev_addr);
-
-	return 0;
-
-err_out_unregister_mdio:
-	phy_disconnect(dev->phydev);
-	mdiobus_unregister(bp->mii_bus);
-	mdiobus_free(bp->mii_bus);
-
-	/* Shutdown the PHY if there is a GPIO reset */
-	if (bp->reset_gpio)
-		gpiod_set_value(bp->reset_gpio, 0);
-
-err_out_free_netdev:
-	free_netdev(dev);
-
-err_disable_clocks:
-	clk_disable_unprepare(tx_clk);
-	clk_disable_unprepare(hclk);
-	clk_disable_unprepare(pclk);
-	clk_disable_unprepare(rx_clk);
-
-	return err;
-}
-
-static int macb_remove(struct platform_device *pdev)
-{
-	struct net_device *dev;
-	struct macb *bp;
-
-	dev = platform_get_drvdata(pdev);
-
-	if (dev) {
-		bp = netdev_priv(dev);
-		if (dev->phydev)
-			phy_disconnect(dev->phydev);
-		mdiobus_unregister(bp->mii_bus);
-		dev->phydev = NULL;
-		mdiobus_free(bp->mii_bus);
-
-		/* Shutdown the PHY if there is a GPIO reset */
-		if (bp->reset_gpio)
-			gpiod_set_value(bp->reset_gpio, 0);
-
-		unregister_netdev(dev);
-		clk_disable_unprepare(bp->tx_clk);
-		clk_disable_unprepare(bp->hclk);
-		clk_disable_unprepare(bp->pclk);
-		clk_disable_unprepare(bp->rx_clk);
-		free_netdev(dev);
-	}
-
-	return 0;
-}
-
-static int __maybe_unused macb_suspend(struct device *dev)
-{
-	struct platform_device *pdev = to_platform_device(dev);
-	struct net_device *netdev = platform_get_drvdata(pdev);
-	struct macb *bp = netdev_priv(netdev);
-
-	netif_carrier_off(netdev);
-	netif_device_detach(netdev);
-
-	if (bp->wol & MACB_WOL_ENABLED) {
-		macb_writel(bp, IER, MACB_BIT(WOL));
-		macb_writel(bp, WOL, MACB_BIT(MAG));
-		enable_irq_wake(bp->queues[0].irq);
-	} else {
-		clk_disable_unprepare(bp->tx_clk);
-		clk_disable_unprepare(bp->hclk);
-		clk_disable_unprepare(bp->pclk);
-		clk_disable_unprepare(bp->rx_clk);
-	}
-
-	return 0;
-}
-
-static int __maybe_unused macb_resume(struct device *dev)
-{
-	struct platform_device *pdev = to_platform_device(dev);
-	struct net_device *netdev = platform_get_drvdata(pdev);
-	struct macb *bp = netdev_priv(netdev);
-
-	if (bp->wol & MACB_WOL_ENABLED) {
-		macb_writel(bp, IDR, MACB_BIT(WOL));
-		macb_writel(bp, WOL, 0);
-		disable_irq_wake(bp->queues[0].irq);
-	} else {
-		clk_prepare_enable(bp->pclk);
-		clk_prepare_enable(bp->hclk);
-		clk_prepare_enable(bp->tx_clk);
-		clk_prepare_enable(bp->rx_clk);
-	}
-
-	netif_device_attach(netdev);
-
-	return 0;
-}
-
-static SIMPLE_DEV_PM_OPS(macb_pm_ops, macb_suspend, macb_resume);
-
-static struct platform_driver macb_driver = {
-	.probe		= macb_probe,
-	.remove		= macb_remove,
-	.driver		= {
-		.name		= "macb",
-		.of_match_table	= of_match_ptr(macb_dt_ids),
-		.pm	= &macb_pm_ops,
-	},
-};
-
-module_platform_driver(macb_driver);
-
-MODULE_LICENSE("GPL");
-MODULE_DESCRIPTION("Cadence MACB/GEM Ethernet driver");
-MODULE_AUTHOR("Haavard Skinnemoen (Atmel)");
-MODULE_ALIAS("platform:macb");
diff --git a/drivers/net/ethernet/cadence/macb.h b/drivers/net/ethernet/cadence/macb.h
index ec037b0fa2a4..0830c48973aa 100644
--- a/drivers/net/ethernet/cadence/macb.h
+++ b/drivers/net/ethernet/cadence/macb.h
@@ -1,16 +1,23 @@
+/* SPDX-License-Identifier: GPL-2.0-only */
 /*
  * Atmel MACB Ethernet Controller driver
  *
  * Copyright (C) 2004-2006 Atmel Corporation
- *
- * This program is free software; you can redistribute it and/or modify
- * it under the terms of the GNU General Public License version 2 as
- * published by the Free Software Foundation.
  */
 #ifndef _MACB_H
 #define _MACB_H
 
-#include <linux/phy.h>
+#include <linux/clk.h>
+#include <linux/phylink.h>
+#include <linux/ptp_clock_kernel.h>
+#include <linux/net_tstamp.h>
+#include <linux/interrupt.h>
+#include <linux/phy/phy.h>
+#include <linux/workqueue.h>
+
+#if defined(CONFIG_ARCH_DMA_ADDR_T_64BIT) || defined(CONFIG_MACB_USE_HWSTAMP)
+#define MACB_EXT_DESC
+#endif
 
 #define MACB_GREGS_NBR 16
 #define MACB_GREGS_VERSION 2
@@ -72,10 +79,13 @@
 #define MACB_RBQPH		0x04D4
 
 /* GEM register offsets. */
+#define GEM_NCR			0x0000 /* Network Control */
 #define GEM_NCFGR		0x0004 /* Network Config */
 #define GEM_USRIO		0x000c /* User IO */
 #define GEM_DMACFG		0x0010 /* DMA Configuration */
+#define GEM_PBUFRXCUT		0x0044 /* RX Partial Store and Forward */
 #define GEM_JML			0x0048 /* Jumbo Max Length */
+#define GEM_HS_MAC_CONFIG	0x0050 /* GEM high speed config */
 #define GEM_HRB			0x0080 /* Hash Bottom */
 #define GEM_HRT			0x0084 /* Hash Top */
 #define GEM_SA1B		0x0088 /* Specific1 Bottom */
@@ -86,6 +96,13 @@
 #define GEM_SA3T		0x009C /* Specific3 Top */
 #define GEM_SA4B		0x00A0 /* Specific4 Bottom */
 #define GEM_SA4T		0x00A4 /* Specific4 Top */
+#define GEM_WOL			0x00b8 /* Wake on LAN */
+#define GEM_RXPTPUNI		0x00D4 /* PTP RX Unicast address */
+#define GEM_TXPTPUNI		0x00D8 /* PTP TX Unicast address */
+#define GEM_EFTSH		0x00e8 /* PTP Event Frame Transmitted Seconds Register 47:32 */
+#define GEM_EFRSH		0x00ec /* PTP Event Frame Received Seconds Register 47:32 */
+#define GEM_PEFTSH		0x00f0 /* PTP Peer Event Frame Transmitted Seconds Register 47:32 */
+#define GEM_PEFRSH		0x00f4 /* PTP Peer Event Frame Received Seconds Register 47:32 */
 #define GEM_OTX			0x0100 /* Octets transmitted */
 #define GEM_OCTTXL		0x0100 /* Octets transmitted [31:0] */
 #define GEM_OCTTXH		0x0104 /* Octets transmitted [47:32] */
@@ -147,6 +164,16 @@
 #define GEM_PEFTN		0x01f4 /* PTP Peer Event Frame Tx Ns */
 #define GEM_PEFRSL		0x01f8 /* PTP Peer Event Frame Rx Sec Low */
 #define GEM_PEFRN		0x01fc /* PTP Peer Event Frame Rx Ns */
+#define GEM_PCSCNTRL		0x0200 /* PCS Control */
+#define GEM_PCSSTS		0x0204 /* PCS Status */
+#define GEM_PCSPHYTOPID		0x0208 /* PCS PHY Top ID */
+#define GEM_PCSPHYBOTID		0x020c /* PCS PHY Bottom ID */
+#define GEM_PCSANADV		0x0210 /* PCS AN Advertisement */
+#define GEM_PCSANLPBASE		0x0214 /* PCS AN Link Partner Base */
+#define GEM_PCSANEXP		0x0218 /* PCS AN Expansion */
+#define GEM_PCSANNPTX		0x021c /* PCS AN Next Page TX */
+#define GEM_PCSANNPLP		0x0220 /* PCS AN Next Page LP */
+#define GEM_PCSANEXTSTS		0x023c /* PCS AN Extended Status */
 #define GEM_DCFG1		0x0280 /* Design Config 1 */
 #define GEM_DCFG2		0x0284 /* Design Config 2 */
 #define GEM_DCFG3		0x0288 /* Design Config 3 */
@@ -154,15 +181,58 @@
 #define GEM_DCFG5		0x0290 /* Design Config 5 */
 #define GEM_DCFG6		0x0294 /* Design Config 6 */
 #define GEM_DCFG7		0x0298 /* Design Config 7 */
+#define GEM_DCFG8		0x029C /* Design Config 8 */
+#define GEM_DCFG10		0x02A4 /* Design Config 10 */
+#define GEM_DCFG12		0x02AC /* Design Config 12 */
+#define GEM_ENST_START_TIME_Q0	0x0800 /* ENST Q0 start time */
+#define GEM_ENST_START_TIME_Q1	0x0804 /* ENST Q1 start time */
+#define GEM_ENST_ON_TIME_Q0	0x0820 /* ENST Q0 on time */
+#define GEM_ENST_ON_TIME_Q1	0x0824 /* ENST Q1 on time */
+#define GEM_ENST_OFF_TIME_Q0	0x0840 /* ENST Q0 off time */
+#define GEM_ENST_OFF_TIME_Q1	0x0844 /* ENST Q1 off time */
+#define GEM_ENST_CONTROL	0x0880 /* ENST control register */
+#define GEM_USX_CONTROL		0x0A80 /* High speed PCS control register */
+#define GEM_USX_STATUS		0x0A88 /* High speed PCS status register */
+
+#define GEM_TXBDCTRL	0x04cc /* TX Buffer Descriptor control register */
+#define GEM_RXBDCTRL	0x04d0 /* RX Buffer Descriptor control register */
+
+/* Screener Type 2 match registers */
+#define GEM_SCRT2		0x540
+
+/* EtherType registers */
+#define GEM_ETHT		0x06E0
+
+/* Type 2 compare registers */
+#define GEM_T2CMPW0		0x0700
+#define GEM_T2CMPW1		0x0704
+#define T2CMP_OFST(t2idx)	(t2idx * 2)
+
+/* type 2 compare registers
+ * each location requires 3 compare regs
+ */
+#define GEM_IP4SRC_CMP(idx)		(idx * 3)
+#define GEM_IP4DST_CMP(idx)		(idx * 3 + 1)
+#define GEM_PORT_CMP(idx)		(idx * 3 + 2)
+
+/* Which screening type 2 EtherType register will be used (0 - 7) */
+#define SCRT2_ETHT		0
 
 #define GEM_ISR(hw_q)		(0x0400 + ((hw_q) << 2))
 #define GEM_TBQP(hw_q)		(0x0440 + ((hw_q) << 2))
-#define GEM_TBQPH(hw_q)		(0x04C8)
 #define GEM_RBQP(hw_q)		(0x0480 + ((hw_q) << 2))
+#define GEM_RBQS(hw_q)		(0x04A0 + ((hw_q) << 2))
 #define GEM_IER(hw_q)		(0x0600 + ((hw_q) << 2))
 #define GEM_IDR(hw_q)		(0x0620 + ((hw_q) << 2))
 #define GEM_IMR(hw_q)		(0x0640 + ((hw_q) << 2))
 
+#define GEM_ENST_START_TIME(hw_q)	(0x0800 + ((hw_q) << 2))
+#define GEM_ENST_ON_TIME(hw_q)		(0x0820 + ((hw_q) << 2))
+#define GEM_ENST_OFF_TIME(hw_q)		(0x0840 + ((hw_q) << 2))
+
+/* Bitfields in ENST_CONTROL */
+#define GEM_ENST_DISABLE_QUEUE_OFFSET	16
+
 /* Bitfields in NCR */
 #define MACB_LB_OFFSET		0 /* reserved */
 #define MACB_LB_SIZE		1
@@ -190,7 +260,13 @@
 #define MACB_NCR_TPF_SIZE	1
 #define MACB_TZQ_OFFSET		12 /* Transmit zero quantum pause frame */
 #define MACB_TZQ_SIZE		1
-#define MACB_SRTSM_OFFSET	15
+#define MACB_SRTSM_OFFSET	15 /* Store Receive Timestamp to Memory */
+#define MACB_PTPUNI_OFFSET	20 /* PTP Unicast packet enable */
+#define MACB_PTPUNI_SIZE	1
+#define MACB_OSSMODE_OFFSET	24 /* Enable One Step Synchro Mode */
+#define MACB_OSSMODE_SIZE	1
+#define MACB_MIIONRGMII_OFFSET	28 /* MII Usage on RGMII Interface */
+#define MACB_MIIONRGMII_SIZE	1
 
 /* Bitfields in NCFGR */
 #define MACB_SPD_OFFSET		0 /* Speed */
@@ -232,11 +308,19 @@
 #define MACB_IRXFCS_OFFSET	19
 #define MACB_IRXFCS_SIZE	1
 
+/* GEM specific NCR bitfields. */
+#define GEM_ENABLE_HS_MAC_OFFSET	31
+#define GEM_ENABLE_HS_MAC_SIZE		1
+
 /* GEM specific NCFGR bitfields. */
+#define GEM_FD_OFFSET		1 /* Full duplex */
+#define GEM_FD_SIZE		1
 #define GEM_GBE_OFFSET		10 /* Gigabit mode enable */
 #define GEM_GBE_SIZE		1
 #define GEM_PCSSEL_OFFSET	11
 #define GEM_PCSSEL_SIZE		1
+#define GEM_PAE_OFFSET		13 /* Pause enable */
+#define GEM_PAE_SIZE		1
 #define GEM_CLK_OFFSET		18 /* MDC clock division */
 #define GEM_CLK_SIZE		3
 #define GEM_DBW_OFFSET		21 /* Data bus width */
@@ -269,10 +353,18 @@
 #define GEM_RXBS_SIZE		8
 #define GEM_DDRP_OFFSET		24 /* disc_when_no_ahb */
 #define GEM_DDRP_SIZE		1
+#define GEM_RXEXT_OFFSET	28 /* RX extended Buffer Descriptor mode */
+#define GEM_RXEXT_SIZE		1
+#define GEM_TXEXT_OFFSET	29 /* TX extended Buffer Descriptor mode */
+#define GEM_TXEXT_SIZE		1
 #define GEM_ADDR64_OFFSET	30 /* Address bus width - 64b or 32b */
 #define GEM_ADDR64_SIZE		1
 
 
+/* Bitfields in PBUFRXCUT */
+#define GEM_ENCUTTHRU_OFFSET	31 /* Enable RX partial store and forward */
+#define GEM_ENCUTTHRU_SIZE	1
+
 /* Bitfields in NSR */
 #define MACB_NSR_LINK_OFFSET	0 /* pcs_link_state */
 #define MACB_NSR_LINK_SIZE	1
@@ -322,6 +414,8 @@
 #define MACB_ISR_RLE_SIZE	1
 #define MACB_TXERR_OFFSET	6 /* EN TX frame corrupt from error interrupt */
 #define MACB_TXERR_SIZE		1
+#define MACB_RM9200_TBRE_OFFSET	6 /* EN may send new frame interrupt (RM9200) */
+#define MACB_RM9200_TBRE_SIZE	1
 #define MACB_TCOMP_OFFSET	7 /* Enable transmit complete interrupt */
 #define MACB_TCOMP_SIZE		1
 #define MACB_ISR_LINK_OFFSET	9 /* Enable link change interrupt */
@@ -354,6 +448,8 @@
 #define MACB_PDRSFT_SIZE	1
 #define MACB_SRI_OFFSET		26 /* TSU Seconds Register Increment */
 #define MACB_SRI_SIZE		1
+#define GEM_WOL_OFFSET		28 /* Enable wake-on-lan interrupt */
+#define GEM_WOL_SIZE		1
 
 /* Timer increment fields */
 #define MACB_TI_CNS_OFFSET	0
@@ -413,11 +509,21 @@
 #define MACB_REV_OFFSET				0
 #define MACB_REV_SIZE				16
 
+/* Bitfield in HS_MAC_CONFIG */
+#define GEM_HS_MAC_SPEED_OFFSET			0
+#define GEM_HS_MAC_SPEED_SIZE			3
+
+/* Bitfields in PCSCNTRL */
+#define GEM_PCSAUTONEG_OFFSET			12
+#define GEM_PCSAUTONEG_SIZE			1
+
 /* Bitfields in DCFG1. */
 #define GEM_IRQCOR_OFFSET			23
 #define GEM_IRQCOR_SIZE				1
 #define GEM_DBWDEF_OFFSET			25
 #define GEM_DBWDEF_SIZE				3
+#define GEM_NO_PCS_OFFSET			0
+#define GEM_NO_PCS_SIZE				1
 
 /* Bitfields in DCFG2. */
 #define GEM_RX_PKT_BUFF_OFFSET			20
@@ -425,20 +531,198 @@
 #define GEM_TX_PKT_BUFF_OFFSET			21
 #define GEM_TX_PKT_BUFF_SIZE			1
 
+#define GEM_RX_PBUF_ADDR_OFFSET			22
+#define GEM_RX_PBUF_ADDR_SIZE			4
+
+/* Bitfields in DCFG5. */
+#define GEM_TSU_OFFSET				8
+#define GEM_TSU_SIZE				1
+
 /* Bitfields in DCFG6. */
 #define GEM_PBUF_LSO_OFFSET			27
 #define GEM_PBUF_LSO_SIZE			1
+#define GEM_PBUF_CUTTHRU_OFFSET			25
+#define GEM_PBUF_CUTTHRU_SIZE			1
 #define GEM_DAW64_OFFSET			23
 #define GEM_DAW64_SIZE				1
 
+/* Bitfields in DCFG8. */
+#define GEM_T1SCR_OFFSET			24
+#define GEM_T1SCR_SIZE				8
+#define GEM_T2SCR_OFFSET			16
+#define GEM_T2SCR_SIZE				8
+#define GEM_SCR2ETH_OFFSET			8
+#define GEM_SCR2ETH_SIZE			8
+#define GEM_SCR2CMP_OFFSET			0
+#define GEM_SCR2CMP_SIZE			8
+
+/* Bitfields in DCFG10 */
+#define GEM_TXBD_RDBUFF_OFFSET			12
+#define GEM_TXBD_RDBUFF_SIZE			4
+#define GEM_RXBD_RDBUFF_OFFSET			8
+#define GEM_RXBD_RDBUFF_SIZE			4
+
+/* Bitfields in DCFG12. */
+#define GEM_HIGH_SPEED_OFFSET			26
+#define GEM_HIGH_SPEED_SIZE			1
+
+/* Bitfields in ENST_START_TIME_Qx. */
+#define GEM_START_TIME_SEC_OFFSET		30
+#define GEM_START_TIME_SEC_SIZE			2
+#define GEM_START_TIME_NSEC_OFFSET		0
+#define GEM_START_TIME_NSEC_SIZE		30
+
+/* Bitfields in ENST_ON_TIME_Qx. */
+#define GEM_ON_TIME_OFFSET			0
+#define GEM_ON_TIME_SIZE			17
+
+/* Bitfields in ENST_OFF_TIME_Qx. */
+#define GEM_OFF_TIME_OFFSET			0
+#define GEM_OFF_TIME_SIZE			17
+
+/* Hardware ENST timing registers granularity */
+#define ENST_TIME_GRANULARITY_NS		8
+
+/* Bitfields in USX_CONTROL. */
+#define GEM_USX_CTRL_SPEED_OFFSET		14
+#define GEM_USX_CTRL_SPEED_SIZE			3
+#define GEM_SERDES_RATE_OFFSET			12
+#define GEM_SERDES_RATE_SIZE			2
+#define GEM_RX_SCR_BYPASS_OFFSET		9
+#define GEM_RX_SCR_BYPASS_SIZE			1
+#define GEM_TX_SCR_BYPASS_OFFSET		8
+#define GEM_TX_SCR_BYPASS_SIZE			1
+#define GEM_TX_EN_OFFSET			1
+#define GEM_TX_EN_SIZE				1
+#define GEM_SIGNAL_OK_OFFSET			0
+#define GEM_SIGNAL_OK_SIZE			1
+
+/* Bitfields in USX_STATUS. */
+#define GEM_USX_BLOCK_LOCK_OFFSET		0
+#define GEM_USX_BLOCK_LOCK_SIZE			1
+
 /* Bitfields in TISUBN */
 #define GEM_SUBNSINCR_OFFSET			0
-#define GEM_SUBNSINCR_SIZE			16
+#define GEM_SUBNSINCRL_OFFSET			24
+#define GEM_SUBNSINCRL_SIZE			8
+#define GEM_SUBNSINCRH_OFFSET			0
+#define GEM_SUBNSINCRH_SIZE			16
+#define GEM_SUBNSINCR_SIZE			24
 
 /* Bitfields in TI */
 #define GEM_NSINCR_OFFSET			0
 #define GEM_NSINCR_SIZE				8
 
+/* Bitfields in TSH */
+#define GEM_TSH_OFFSET				0 /* TSU timer value (s). MSB [47:32] of seconds timer count */
+#define GEM_TSH_SIZE				16
+
+/* Bitfields in TSL */
+#define GEM_TSL_OFFSET				0 /* TSU timer value (s). LSB [31:0] of seconds timer count */
+#define GEM_TSL_SIZE				32
+
+/* Bitfields in TN */
+#define GEM_TN_OFFSET				0 /* TSU timer value (ns) */
+#define GEM_TN_SIZE					30
+
+/* Bitfields in TXBDCTRL */
+#define GEM_TXTSMODE_OFFSET			4 /* TX Descriptor Timestamp Insertion mode */
+#define GEM_TXTSMODE_SIZE			2
+
+/* Bitfields in RXBDCTRL */
+#define GEM_RXTSMODE_OFFSET			4 /* RX Descriptor Timestamp Insertion mode */
+#define GEM_RXTSMODE_SIZE			2
+
+/* Bitfields in SCRT2 */
+#define GEM_QUEUE_OFFSET			0 /* Queue Number */
+#define GEM_QUEUE_SIZE				4
+#define GEM_VLANPR_OFFSET			4 /* VLAN Priority */
+#define GEM_VLANPR_SIZE				3
+#define GEM_VLANEN_OFFSET			8 /* VLAN Enable */
+#define GEM_VLANEN_SIZE				1
+#define GEM_ETHT2IDX_OFFSET			9 /* Index to screener type 2 EtherType register */
+#define GEM_ETHT2IDX_SIZE			3
+#define GEM_ETHTEN_OFFSET			12 /* EtherType Enable */
+#define GEM_ETHTEN_SIZE				1
+#define GEM_CMPA_OFFSET				13 /* Compare A - Index to screener type 2 Compare register */
+#define GEM_CMPA_SIZE				5
+#define GEM_CMPAEN_OFFSET			18 /* Compare A Enable */
+#define GEM_CMPAEN_SIZE				1
+#define GEM_CMPB_OFFSET				19 /* Compare B - Index to screener type 2 Compare register */
+#define GEM_CMPB_SIZE				5
+#define GEM_CMPBEN_OFFSET			24 /* Compare B Enable */
+#define GEM_CMPBEN_SIZE				1
+#define GEM_CMPC_OFFSET				25 /* Compare C - Index to screener type 2 Compare register */
+#define GEM_CMPC_SIZE				5
+#define GEM_CMPCEN_OFFSET			30 /* Compare C Enable */
+#define GEM_CMPCEN_SIZE				1
+
+/* Bitfields in ETHT */
+#define GEM_ETHTCMP_OFFSET			0 /* EtherType compare value */
+#define GEM_ETHTCMP_SIZE			16
+
+/* Bitfields in T2CMPW0 */
+#define GEM_T2CMP_OFFSET			16 /* 0xFFFF0000 compare value */
+#define GEM_T2CMP_SIZE				16
+#define GEM_T2MASK_OFFSET			0 /* 0x0000FFFF compare value or mask */
+#define GEM_T2MASK_SIZE				16
+
+/* Bitfields in T2CMPW1 */
+#define GEM_T2DISMSK_OFFSET			9 /* disable mask */
+#define GEM_T2DISMSK_SIZE			1
+#define GEM_T2CMPOFST_OFFSET			7 /* compare offset */
+#define GEM_T2CMPOFST_SIZE			2
+#define GEM_T2OFST_OFFSET			0 /* offset value */
+#define GEM_T2OFST_SIZE				7
+
+/* Bitfields in queue pointer registers */
+#define MACB_QUEUE_DISABLE_OFFSET		0 /* disable queue */
+#define MACB_QUEUE_DISABLE_SIZE			1
+
+/* Offset for screener type 2 compare values (T2CMPOFST).
+ * Note the offset is applied after the specified point,
+ * e.g. GEM_T2COMPOFST_ETYPE denotes the EtherType field, so an offset
+ * of 12 bytes from this would be the source IP address in an IP header
+ */
+#define GEM_T2COMPOFST_SOF		0
+#define GEM_T2COMPOFST_ETYPE	1
+#define GEM_T2COMPOFST_IPHDR	2
+#define GEM_T2COMPOFST_TCPUDP	3
+
+/* offset from EtherType to IP address */
+#define ETYPE_SRCIP_OFFSET			12
+#define ETYPE_DSTIP_OFFSET			16
+
+/* offset from IP header to port */
+#define IPHDR_SRCPORT_OFFSET		0
+#define IPHDR_DSTPORT_OFFSET		2
+
+/* Transmit DMA buffer descriptor Word 1 */
+#define GEM_DMA_TXVALID_OFFSET		23 /* timestamp has been captured in the Buffer Descriptor */
+#define GEM_DMA_TXVALID_SIZE		1
+
+/* Receive DMA buffer descriptor Word 0 */
+#define GEM_DMA_RXVALID_OFFSET		2 /* indicates a valid timestamp in the Buffer Descriptor */
+#define GEM_DMA_RXVALID_SIZE		1
+
+/* DMA buffer descriptor Word 2 (32 bit addressing) or Word 4 (64 bit addressing) */
+#define GEM_DMA_SECL_OFFSET			30 /* Timestamp seconds[1:0]  */
+#define GEM_DMA_SECL_SIZE			2
+#define GEM_DMA_NSEC_OFFSET			0 /* Timestamp nanosecs [29:0] */
+#define GEM_DMA_NSEC_SIZE			30
+
+/* DMA buffer descriptor Word 3 (32 bit addressing) or Word 5 (64 bit addressing) */
+
+/* New hardware supports 12 bit precision of timestamp in DMA buffer descriptor.
+ * Old hardware supports only 6 bit precision but it is enough for PTP.
+ * Less accuracy is used always instead of checking hardware version.
+ */
+#define GEM_DMA_SECH_OFFSET			0 /* Timestamp seconds[5:2] */
+#define GEM_DMA_SECH_SIZE			4
+#define GEM_DMA_SEC_WIDTH			(GEM_DMA_SECH_SIZE + GEM_DMA_SECL_SIZE)
+#define GEM_DMA_SEC_TOP				(1 << GEM_DMA_SEC_WIDTH)
+#define GEM_DMA_SEC_MASK			(GEM_DMA_SEC_TOP - 1)
+
 /* Bitfields in ADJ */
 #define GEM_ADDSUB_OFFSET			31
 #define GEM_ADDSUB_SIZE				1
@@ -455,12 +739,21 @@
 #define GEM_CLK_DIV48				3
 #define GEM_CLK_DIV64				4
 #define GEM_CLK_DIV96				5
+#define GEM_CLK_DIV128				6
+#define GEM_CLK_DIV224				7
 
 /* Constants for MAN register */
-#define MACB_MAN_SOF				1
-#define MACB_MAN_WRITE				1
-#define MACB_MAN_READ				2
-#define MACB_MAN_CODE				2
+#define MACB_MAN_C22_SOF			1
+#define MACB_MAN_C22_WRITE			1
+#define MACB_MAN_C22_READ			2
+#define MACB_MAN_C22_CODE			2
+
+#define MACB_MAN_C45_SOF			0
+#define MACB_MAN_C45_ADDR			0
+#define MACB_MAN_C45_WRITE			1
+#define MACB_MAN_C45_POST_READ_INCR		2
+#define MACB_MAN_C45_READ			3
+#define MACB_MAN_C45_CODE			2
 
 /* Capability mask bits */
 #define MACB_CAPS_ISR_CLEAR_ON_WRITE		0x00000001
@@ -470,6 +763,16 @@
 #define MACB_CAPS_USRIO_DISABLED		0x00000010
 #define MACB_CAPS_JUMBO				0x00000020
 #define MACB_CAPS_GEM_HAS_PTP			0x00000040
+#define MACB_CAPS_BD_RD_PREFETCH		0x00000080
+#define MACB_CAPS_NEEDS_RSTONUBR		0x00000100
+#define MACB_CAPS_MIIONRGMII			0x00000200
+#define MACB_CAPS_NEED_TSUCLK			0x00000400
+#define MACB_CAPS_QUEUE_DISABLE			0x00000800
+#define MACB_CAPS_QBV				0x00001000
+#define MACB_CAPS_PCS				0x01000000
+#define MACB_CAPS_HIGH_SPEED			0x02000000
+#define MACB_CAPS_CLK_HW_CHG			0x04000000
+#define MACB_CAPS_MACB_IS_EMAC			0x08000000
 #define MACB_CAPS_FIFO_MODE			0x10000000
 #define MACB_CAPS_GIGABIT_MODE_AVAILABLE	0x20000000
 #define MACB_CAPS_SG_DISABLED			0x40000000
@@ -513,6 +816,8 @@
 #define gem_writel(port, reg, value)	(port)->macb_reg_writel((port), GEM_##reg, (value))
 #define queue_readl(queue, reg)		(queue)->bp->macb_reg_readl((queue)->bp, (queue)->reg)
 #define queue_writel(queue, reg, value)	(queue)->bp->macb_reg_writel((queue)->bp, (queue)->reg, (value))
+#define gem_readl_n(port, reg, idx)		(port)->macb_reg_readl((port), GEM_##reg + idx * 4)
+#define gem_writel_n(port, reg, idx, value)	(port)->macb_reg_writel((port), GEM_##reg + idx * 4, (value))
 
 /* Conditional GEM/MACB macros.  These perform the operation to the correct
  * register dependent on whether the device is a GEM or a MACB.  For registers
@@ -537,6 +842,8 @@
 		__v; \
 	})
 
+#define MACB_READ_NSR(bp)	macb_readl(bp, NSR)
+
 /* struct macb_dma_desc - Hardware DMA descriptor
  * @addr: DMA address of data buffer
  * @ctrl: Control and status bits
@@ -546,16 +853,21 @@ struct macb_dma_desc {
 	u32	ctrl;
 };
 
-#ifdef CONFIG_ARCH_DMA_ADDR_T_64BIT
-enum macb_hw_dma_cap {
-	HW_DMA_CAP_32B,
-	HW_DMA_CAP_64B,
-};
+#ifdef MACB_EXT_DESC
+#define HW_DMA_CAP_32B		0
+#define HW_DMA_CAP_64B		(1 << 0)
+#define HW_DMA_CAP_PTP		(1 << 1)
+#define HW_DMA_CAP_64B_PTP	(HW_DMA_CAP_64B | HW_DMA_CAP_PTP)
 
 struct macb_dma_desc_64 {
 	u32 addrh;
 	u32 resvd;
 };
+
+struct macb_dma_desc_ptp {
+	u32	ts_1;
+	u32	ts_2;
+};
 #endif
 
 /* DMA descriptor bitfields */
@@ -647,6 +959,9 @@ struct macb_dma_desc_64 {
 /* limit RX checksum offload to TCP and UDP packets */
 #define GEM_RX_CSUM_CHECKED_MASK		2
 
+/* Scaled PPM fraction */
+#define PPM_FRACTION	16
+
 /* struct macb_tx_skb - data about an skb which is being transmitted
  * @skb: skb currently being transmitted, only set for the last buffer
  *       of the frame
@@ -666,75 +981,73 @@ struct macb_tx_skb {
  * device stats by a periodic timer.
  */
 struct macb_stats {
-	u32	rx_pause_frames;
-	u32	tx_ok;
-	u32	tx_single_cols;
-	u32	tx_multiple_cols;
-	u32	rx_ok;
-	u32	rx_fcs_errors;
-	u32	rx_align_errors;
-	u32	tx_deferred;
-	u32	tx_late_cols;
-	u32	tx_excessive_cols;
-	u32	tx_underruns;
-	u32	tx_carrier_errors;
-	u32	rx_resource_errors;
-	u32	rx_overruns;
-	u32	rx_symbol_errors;
-	u32	rx_oversize_pkts;
-	u32	rx_jabbers;
-	u32	rx_undersize_pkts;
-	u32	sqe_test_errors;
-	u32	rx_length_mismatch;
-	u32	tx_pause_frames;
+	u64	rx_pause_frames;
+	u64	tx_ok;
+	u64	tx_single_cols;
+	u64	tx_multiple_cols;
+	u64	rx_ok;
+	u64	rx_fcs_errors;
+	u64	rx_align_errors;
+	u64	tx_deferred;
+	u64	tx_late_cols;
+	u64	tx_excessive_cols;
+	u64	tx_underruns;
+	u64	tx_carrier_errors;
+	u64	rx_resource_errors;
+	u64	rx_overruns;
+	u64	rx_symbol_errors;
+	u64	rx_oversize_pkts;
+	u64	rx_jabbers;
+	u64	rx_undersize_pkts;
+	u64	sqe_test_errors;
+	u64	rx_length_mismatch;
+	u64	tx_pause_frames;
 };
 
 struct gem_stats {
-	u32	tx_octets_31_0;
-	u32	tx_octets_47_32;
-	u32	tx_frames;
-	u32	tx_broadcast_frames;
-	u32	tx_multicast_frames;
-	u32	tx_pause_frames;
-	u32	tx_64_byte_frames;
-	u32	tx_65_127_byte_frames;
-	u32	tx_128_255_byte_frames;
-	u32	tx_256_511_byte_frames;
-	u32	tx_512_1023_byte_frames;
-	u32	tx_1024_1518_byte_frames;
-	u32	tx_greater_than_1518_byte_frames;
-	u32	tx_underrun;
-	u32	tx_single_collision_frames;
-	u32	tx_multiple_collision_frames;
-	u32	tx_excessive_collisions;
-	u32	tx_late_collisions;
-	u32	tx_deferred_frames;
-	u32	tx_carrier_sense_errors;
-	u32	rx_octets_31_0;
-	u32	rx_octets_47_32;
-	u32	rx_frames;
-	u32	rx_broadcast_frames;
-	u32	rx_multicast_frames;
-	u32	rx_pause_frames;
-	u32	rx_64_byte_frames;
-	u32	rx_65_127_byte_frames;
-	u32	rx_128_255_byte_frames;
-	u32	rx_256_511_byte_frames;
-	u32	rx_512_1023_byte_frames;
-	u32	rx_1024_1518_byte_frames;
-	u32	rx_greater_than_1518_byte_frames;
-	u32	rx_undersized_frames;
-	u32	rx_oversize_frames;
-	u32	rx_jabbers;
-	u32	rx_frame_check_sequence_errors;
-	u32	rx_length_field_frame_errors;
-	u32	rx_symbol_errors;
-	u32	rx_alignment_errors;
-	u32	rx_resource_errors;
-	u32	rx_overruns;
-	u32	rx_ip_header_checksum_errors;
-	u32	rx_tcp_checksum_errors;
-	u32	rx_udp_checksum_errors;
+	u64	tx_octets;
+	u64	tx_frames;
+	u64	tx_broadcast_frames;
+	u64	tx_multicast_frames;
+	u64	tx_pause_frames;
+	u64	tx_64_byte_frames;
+	u64	tx_65_127_byte_frames;
+	u64	tx_128_255_byte_frames;
+	u64	tx_256_511_byte_frames;
+	u64	tx_512_1023_byte_frames;
+	u64	tx_1024_1518_byte_frames;
+	u64	tx_greater_than_1518_byte_frames;
+	u64	tx_underrun;
+	u64	tx_single_collision_frames;
+	u64	tx_multiple_collision_frames;
+	u64	tx_excessive_collisions;
+	u64	tx_late_collisions;
+	u64	tx_deferred_frames;
+	u64	tx_carrier_sense_errors;
+	u64	rx_octets;
+	u64	rx_frames;
+	u64	rx_broadcast_frames;
+	u64	rx_multicast_frames;
+	u64	rx_pause_frames;
+	u64	rx_64_byte_frames;
+	u64	rx_65_127_byte_frames;
+	u64	rx_128_255_byte_frames;
+	u64	rx_256_511_byte_frames;
+	u64	rx_512_1023_byte_frames;
+	u64	rx_1024_1518_byte_frames;
+	u64	rx_greater_than_1518_byte_frames;
+	u64	rx_undersized_frames;
+	u64	rx_oversize_frames;
+	u64	rx_jabbers;
+	u64	rx_frame_check_sequence_errors;
+	u64	rx_length_field_frame_errors;
+	u64	rx_symbol_errors;
+	u64	rx_alignment_errors;
+	u64	rx_resource_errors;
+	u64	rx_overruns;
+	u64	rx_ip_header_checksum_errors;
+	u64	rx_tcp_checksum_errors;
+	u64	rx_udp_checksum_errors;
 };
 
 /* Describes the name and offset of an individual statistic register, as
@@ -742,7 +1055,7 @@ struct gem_stats {
  * this register should contribute to.
  */
 struct gem_statistic {
-	char stat_string[ETH_GSTRING_LEN];
+	char stat_string[ETH_GSTRING_LEN] __nonstring;
 	int offset;
 	u32 stat_bits;
 };
@@ -838,13 +1151,43 @@ static const struct gem_statistic gem_statistics[] = {
 
 #define GEM_STATS_LEN ARRAY_SIZE(gem_statistics)
 
+#define QUEUE_STAT_TITLE(title) {	\
+	.stat_string = title,			\
+}
+
+/* per queue statistics, each should be unsigned long type */
+struct queue_stats {
+	union {
+		unsigned long first;
+		unsigned long rx_packets;
+	};
+	unsigned long rx_bytes;
+	unsigned long rx_dropped;
+	unsigned long tx_packets;
+	unsigned long tx_bytes;
+	unsigned long tx_dropped;
+};
+
+static const struct gem_statistic queue_statistics[] = {
+		QUEUE_STAT_TITLE("rx_packets"),
+		QUEUE_STAT_TITLE("rx_bytes"),
+		QUEUE_STAT_TITLE("rx_dropped"),
+		QUEUE_STAT_TITLE("tx_packets"),
+		QUEUE_STAT_TITLE("tx_bytes"),
+		QUEUE_STAT_TITLE("tx_dropped"),
+};
+
+#define QUEUE_STATS_LEN ARRAY_SIZE(queue_statistics)
+
 struct macb;
+struct macb_queue;
 
 struct macb_or_gem_ops {
 	int	(*mog_alloc_rx_buffers)(struct macb *bp);
 	void	(*mog_free_rx_buffers)(struct macb *bp);
 	void	(*mog_init_rings)(struct macb *bp);
-	int	(*mog_rx)(struct macb *bp, int budget);
+	int	(*mog_rx)(struct macb_queue *queue, struct napi_struct *napi,
+			  int budget);
 };
 
 /* MACB-PTP interface: adapt to platform needs. */
@@ -854,11 +1197,25 @@ struct macb_ptp_info {
 	s32 (*get_ptp_max_adj)(void);
 	unsigned int (*get_tsu_rate)(struct macb *bp);
 	int (*get_ts_info)(struct net_device *dev,
-			   struct ethtool_ts_info *info);
+			   struct kernel_ethtool_ts_info *info);
 	int (*get_hwtst)(struct net_device *netdev,
-			 struct ifreq *ifr);
+			 struct kernel_hwtstamp_config *tstamp_config);
 	int (*set_hwtst)(struct net_device *netdev,
-			 struct ifreq *ifr, int cmd);
+			 struct kernel_hwtstamp_config *tstamp_config,
+			 struct netlink_ext_ack *extack);
+};
+
+struct macb_pm_data {
+	u32 scrt2;
+	u32 usrio;
+};
+
+struct macb_usrio_config {
+	u32 mii;
+	u32 rmii;
+	u32 rgmii;
+	u32 refclk;
+	u32 hdfctlen;
 };
 
 struct macb_config {
@@ -866,9 +1223,16 @@ struct macb_config {
 	unsigned int		dma_burst_length;
 	int	(*clk_init)(struct platform_device *pdev, struct clk **pclk,
 			    struct clk **hclk, struct clk **tx_clk,
-			    struct clk **rx_clk);
+			    struct clk **rx_clk, struct clk **tsu_clk);
 	int	(*init)(struct platform_device *pdev);
+	unsigned int		max_tx_length;
 	int	jumbo_max_len;
+	const struct macb_usrio_config *usrio;
+};
+
+struct tsu_incr {
+	u32 sub_ns;
+	u32 ns;
 };
 
 struct macb_queue {
@@ -880,13 +1244,43 @@ struct macb_queue {
 	unsigned int		IDR;
 	unsigned int		IMR;
 	unsigned int		TBQP;
-	unsigned int		TBQPH;
+	unsigned int		RBQS;
+	unsigned int		RBQP;
+
+	/* ENST register offsets for this queue */
+	unsigned int		ENST_START_TIME;
+	unsigned int		ENST_ON_TIME;
+	unsigned int		ENST_OFF_TIME;
 
+	/* Lock to protect tx_head and tx_tail */
+	spinlock_t		tx_ptr_lock;
 	unsigned int		tx_head, tx_tail;
 	struct macb_dma_desc	*tx_ring;
 	struct macb_tx_skb	*tx_skb;
 	dma_addr_t		tx_ring_dma;
 	struct work_struct	tx_error_task;
+	bool			txubr_pending;
+	struct napi_struct	napi_tx;
+
+	dma_addr_t		rx_ring_dma;
+	dma_addr_t		rx_buffers_dma;
+	unsigned int		rx_tail;
+	unsigned int		rx_prepared_head;
+	struct macb_dma_desc	*rx_ring;
+	struct sk_buff		**rx_skbuff;
+	void			*rx_buffers;
+	struct napi_struct	napi_rx;
+	struct queue_stats stats;
+};
+
+struct ethtool_rx_fs_item {
+	struct ethtool_rx_flow_spec fs;
+	struct list_head list;
+};
+
+struct ethtool_rx_fs_list {
+	struct list_head list;
+	unsigned int count;
 };
 
 struct macb {
@@ -897,11 +1291,8 @@ struct macb {
 	u32	(*macb_reg_readl)(struct macb *bp, int offset);
 	void	(*macb_reg_writel)(struct macb *bp, int offset, u32 value);
 
-	unsigned int		rx_tail;
-	unsigned int		rx_prepared_head;
-	struct macb_dma_desc	*rx_ring;
-	struct sk_buff		**rx_skbuff;
-	void			*rx_buffers;
+	struct macb_dma_desc	*rx_ring_tieoff;
+	dma_addr_t		rx_ring_tieoff_dma;
 	size_t			rx_buffer_size;
 
 	unsigned int		rx_ring_size;
@@ -917,48 +1308,118 @@ struct macb {
 	struct clk		*hclk;
 	struct clk		*tx_clk;
 	struct clk		*rx_clk;
+	struct clk		*tsu_clk;
 	struct net_device	*dev;
-	struct napi_struct	napi;
+	/* Protects hw_stats and ethtool_stats */
+	spinlock_t		stats_lock;
 	union {
 		struct macb_stats	macb;
 		struct gem_stats	gem;
 	}			hw_stats;
 
-	dma_addr_t		rx_ring_dma;
-	dma_addr_t		rx_buffers_dma;
-
 	struct macb_or_gem_ops	macbgem_ops;
 
 	struct mii_bus		*mii_bus;
-	int 			link;
-	int 			speed;
-	int 			duplex;
+	struct phylink		*phylink;
+	struct phylink_config	phylink_config;
+	struct phylink_pcs	phylink_usx_pcs;
+	struct phylink_pcs	phylink_sgmii_pcs;
 
 	u32			caps;
 	unsigned int		dma_burst_length;
 
 	phy_interface_t		phy_interface;
-	struct gpio_desc	*reset_gpio;
 
-	/* AT91RM9200 transmit */
-	struct sk_buff *skb;			/* holds skb until xmit interrupt completes */
-	dma_addr_t skb_physaddr;		/* phys addr from pci_map_single */
-	int skb_length;				/* saved skb length for pci_unmap_single */
+	/* AT91RM9200 transmit queue (1 on wire + 1 queued) */
+	struct macb_tx_skb	rm9200_txq[2];
 	unsigned int		max_tx_length;
 
-	u64			ethtool_stats[GEM_STATS_LEN];
+	u64			ethtool_stats[GEM_STATS_LEN + QUEUE_STATS_LEN * MACB_MAX_QUEUES];
 
 	unsigned int		rx_frm_len_mask;
 	unsigned int		jumbo_max_len;
 
 	u32			wol;
+	u32			wolopts;
+
+	/* holds value of rx watermark value for pbuf_rxcutthru register */
+	u32			rx_watermark;
 
 	struct macb_ptp_info	*ptp_info;	/* macb-ptp interface */
-#ifdef CONFIG_ARCH_DMA_ADDR_T_64BIT
-	enum macb_hw_dma_cap hw_dma_cap;
+
+	struct phy		*sgmii_phy;	/* for ZynqMP SGMII mode */
+
+#ifdef MACB_EXT_DESC
+	uint8_t hw_dma_cap;
 #endif
+	spinlock_t tsu_clk_lock; /* gem tsu clock locking */
+	unsigned int tsu_rate;
+	struct ptp_clock *ptp_clock;
+	struct ptp_clock_info ptp_clock_info;
+	struct tsu_incr tsu_incr;
+	struct kernel_hwtstamp_config tstamp_config;
+
+	/* RX queue filer rule set*/
+	struct ethtool_rx_fs_list rx_fs_list;
+	spinlock_t rx_fs_lock;
+	unsigned int max_tuples;
+
+	struct work_struct	hresp_err_bh_work;
+
+	int	rx_bd_rd_prefetch;
+	int	tx_bd_rd_prefetch;
+
+	u32	rx_intr_mask;
+
+	struct macb_pm_data pm_data;
+	const struct macb_usrio_config *usrio;
+};
+
+#ifdef CONFIG_MACB_USE_HWSTAMP
+#define GEM_TSEC_SIZE  (GEM_TSH_SIZE + GEM_TSL_SIZE)
+#define TSU_SEC_MAX_VAL (((u64)1 << GEM_TSEC_SIZE) - 1)
+#define TSU_NSEC_MAX_VAL ((1 << GEM_TN_SIZE) - 1)
+
+enum macb_bd_control {
+	TSTAMP_DISABLED,
+	TSTAMP_FRAME_PTP_EVENT_ONLY,
+	TSTAMP_ALL_PTP_FRAMES,
+	TSTAMP_ALL_FRAMES,
 };
 
+void gem_ptp_init(struct net_device *ndev);
+void gem_ptp_remove(struct net_device *ndev);
+void gem_ptp_txstamp(struct macb *bp, struct sk_buff *skb, struct macb_dma_desc *desc);
+void gem_ptp_rxstamp(struct macb *bp, struct sk_buff *skb, struct macb_dma_desc *desc);
+static inline void gem_ptp_do_txstamp(struct macb *bp, struct sk_buff *skb, struct macb_dma_desc *desc)
+{
+	if (bp->tstamp_config.tx_type == TSTAMP_DISABLED)
+		return;
+
+	gem_ptp_txstamp(bp, skb, desc);
+}
+
+static inline void gem_ptp_do_rxstamp(struct macb *bp, struct sk_buff *skb, struct macb_dma_desc *desc)
+{
+	if (bp->tstamp_config.rx_filter == TSTAMP_DISABLED)
+		return;
+
+	gem_ptp_rxstamp(bp, skb, desc);
+}
+
+int gem_get_hwtst(struct net_device *dev,
+		  struct kernel_hwtstamp_config *tstamp_config);
+int gem_set_hwtst(struct net_device *dev,
+		  struct kernel_hwtstamp_config *tstamp_config,
+		  struct netlink_ext_ack *extack);
+#else
+static inline void gem_ptp_init(struct net_device *ndev) { }
+static inline void gem_ptp_remove(struct net_device *ndev) { }
+
+static inline void gem_ptp_do_txstamp(struct macb *bp, struct sk_buff *skb, struct macb_dma_desc *desc) { }
+static inline void gem_ptp_do_rxstamp(struct macb *bp, struct sk_buff *skb, struct macb_dma_desc *desc) { }
+#endif
+
 static inline bool macb_is_gem(struct macb *bp)
 {
 	return !!(bp->caps & MACB_CAPS_MACB_IS_GEM);
@@ -966,7 +1427,47 @@ static inline bool macb_is_gem(struct macb *bp)
 
 static inline bool gem_has_ptp(struct macb *bp)
 {
-	return !!(bp->caps & MACB_CAPS_GEM_HAS_PTP);
+	return IS_ENABLED(CONFIG_MACB_USE_HWSTAMP) && (bp->caps & MACB_CAPS_GEM_HAS_PTP);
 }
 
+/* ENST Helper functions */
+static inline u64 enst_ns_to_hw_units(size_t ns, u32 speed_mbps)
+{
+	return DIV_ROUND_UP((ns) * (speed_mbps),
+			    (ENST_TIME_GRANULARITY_NS * 1000));
+}
+
+static inline u64 enst_max_hw_interval(u32 speed_mbps)
+{
+	return DIV_ROUND_UP(GENMASK(GEM_ON_TIME_SIZE - 1, 0) *
+			    ENST_TIME_GRANULARITY_NS * 1000, (speed_mbps));
+}
+
+/**
+ * struct macb_platform_data - platform data for MACB Ethernet used for PCI registration
+ * @pclk:		platform clock
+ * @hclk:		AHB clock
+ */
+struct macb_platform_data {
+	struct clk	*pclk;
+	struct clk	*hclk;
+};
+
+/**
+ * struct macb_queue_enst_config - Configuration for Enhanced Scheduled Traffic
+ * @start_time_mask:  Bitmask representing the start time for the queue
+ * @on_time_bytes:    "on" time nsec expressed in bytes
+ * @off_time_bytes:   "off" time nsec expressed in bytes
+ * @queue_id:         Identifier for the queue
+ *
+ * This structure holds the configuration parameters for an ENST queue,
+ * used to control time-based transmission scheduling in the MACB driver.
+ */
+struct macb_queue_enst_config {
+	u32 start_time_mask;
+	u32 on_time_bytes;
+	u32 off_time_bytes;
+	u8 queue_id;
+};
+
 #endif /* _MACB_H */
diff --git a/drivers/net/ethernet/cadence/macb_main.c b/drivers/net/ethernet/cadence/macb_main.c
new file mode 100644
index 000000000000..ca2386b83473
--- /dev/null
+++ b/drivers/net/ethernet/cadence/macb_main.c
@@ -0,0 +1,5948 @@
+// SPDX-License-Identifier: GPL-2.0-only
+/*
+ * Cadence MACB/GEM Ethernet Controller driver
+ *
+ * Copyright (C) 2004-2006 Atmel Corporation
+ */
+
+#define pr_fmt(fmt) KBUILD_MODNAME ": " fmt
+#include <linux/clk.h>
+#include <linux/clk-provider.h>
+#include <linux/crc32.h>
+#include <linux/module.h>
+#include <linux/moduleparam.h>
+#include <linux/kernel.h>
+#include <linux/types.h>
+#include <linux/circ_buf.h>
+#include <linux/slab.h>
+#include <linux/init.h>
+#include <linux/io.h>
+#include <linux/interrupt.h>
+#include <linux/netdevice.h>
+#include <linux/etherdevice.h>
+#include <linux/dma-mapping.h>
+#include <linux/platform_device.h>
+#include <linux/phylink.h>
+#include <linux/of.h>
+#include <linux/of_mdio.h>
+#include <linux/of_net.h>
+#include <linux/ip.h>
+#include <linux/udp.h>
+#include <linux/tcp.h>
+#include <linux/iopoll.h>
+#include <linux/phy/phy.h>
+#include <linux/pm_runtime.h>
+#include <linux/ptp_classify.h>
+#include <linux/reset.h>
+#include <linux/firmware/xlnx-zynqmp.h>
+#include <linux/inetdevice.h>
+#include <net/pkt_sched.h>
+#include "macb.h"
+
+/* This structure is only used for MACB on SiFive FU540 devices */
+struct sifive_fu540_macb_mgmt {
+	void __iomem *reg;
+	unsigned long rate;
+	struct clk_hw hw;
+};
+
+#define MACB_RX_BUFFER_SIZE	128
+#define RX_BUFFER_MULTIPLE	64  /* bytes */
+
+#define DEFAULT_RX_RING_SIZE	512 /* must be power of 2 */
+#define MIN_RX_RING_SIZE	64
+#define MAX_RX_RING_SIZE	8192
+
+#define DEFAULT_TX_RING_SIZE	512 /* must be power of 2 */
+#define MIN_TX_RING_SIZE	64
+#define MAX_TX_RING_SIZE	4096
+
+/* level of occupied TX descriptors under which we wake up TX process */
+#define MACB_TX_WAKEUP_THRESH(bp)	(3 * (bp)->tx_ring_size / 4)
+
+#define MACB_RX_INT_FLAGS	(MACB_BIT(RCOMP) | MACB_BIT(ISR_ROVR))
+#define MACB_TX_ERR_FLAGS	(MACB_BIT(ISR_TUND)			\
+					| MACB_BIT(ISR_RLE)		\
+					| MACB_BIT(TXERR))
+#define MACB_TX_INT_FLAGS	(MACB_TX_ERR_FLAGS | MACB_BIT(TCOMP)	\
+					| MACB_BIT(TXUBR))
+
+/* Max length of transmit frame must be a multiple of 8 bytes */
+#define MACB_TX_LEN_ALIGN	8
+#define MACB_MAX_TX_LEN		((unsigned int)((1 << MACB_TX_FRMLEN_SIZE) - 1) & ~((unsigned int)(MACB_TX_LEN_ALIGN - 1)))
+/* Limit maximum TX length as per Cadence TSO errata. This is to avoid a
+ * false amba_error in TX path from the DMA assuming there is not enough
+ * space in the SRAM (16KB) even when there is.
+ */
+#define GEM_MAX_TX_LEN		(unsigned int)(0x3FC0)
+
+#define GEM_MTU_MIN_SIZE	ETH_MIN_MTU
+#define MACB_NETIF_LSO		NETIF_F_TSO
+
+#define MACB_WOL_ENABLED		BIT(0)
+
+#define HS_SPEED_10000M			4
+#define MACB_SERDES_RATE_10G		1
+
+/* Graceful stop timeouts in us. We should allow up to
+ * 1 frame time (10 Mbits/s, full-duplex, ignoring collisions)
+ */
+#define MACB_HALT_TIMEOUT	14000
+#define MACB_PM_TIMEOUT  100 /* ms */
+
+#define MACB_MDIO_TIMEOUT	1000000 /* in usecs */
+
+/* DMA buffer descriptor might be different size
+ * depends on hardware configuration:
+ *
+ * 1. dma address width 32 bits:
+ *    word 1: 32 bit address of Data Buffer
+ *    word 2: control
+ *
+ * 2. dma address width 64 bits:
+ *    word 1: 32 bit address of Data Buffer
+ *    word 2: control
+ *    word 3: upper 32 bit address of Data Buffer
+ *    word 4: unused
+ *
+ * 3. dma address width 32 bits with hardware timestamping:
+ *    word 1: 32 bit address of Data Buffer
+ *    word 2: control
+ *    word 3: timestamp word 1
+ *    word 4: timestamp word 2
+ *
+ * 4. dma address width 64 bits with hardware timestamping:
+ *    word 1: 32 bit address of Data Buffer
+ *    word 2: control
+ *    word 3: upper 32 bit address of Data Buffer
+ *    word 4: unused
+ *    word 5: timestamp word 1
+ *    word 6: timestamp word 2
+ */
+static unsigned int macb_dma_desc_get_size(struct macb *bp)
+{
+#ifdef MACB_EXT_DESC
+	unsigned int desc_size;
+
+	switch (bp->hw_dma_cap) {
+	case HW_DMA_CAP_64B:
+		desc_size = sizeof(struct macb_dma_desc)
+			+ sizeof(struct macb_dma_desc_64);
+		break;
+	case HW_DMA_CAP_PTP:
+		desc_size = sizeof(struct macb_dma_desc)
+			+ sizeof(struct macb_dma_desc_ptp);
+		break;
+	case HW_DMA_CAP_64B_PTP:
+		desc_size = sizeof(struct macb_dma_desc)
+			+ sizeof(struct macb_dma_desc_64)
+			+ sizeof(struct macb_dma_desc_ptp);
+		break;
+	default:
+		desc_size = sizeof(struct macb_dma_desc);
+	}
+	return desc_size;
+#endif
+	return sizeof(struct macb_dma_desc);
+}
+
+static unsigned int macb_adj_dma_desc_idx(struct macb *bp, unsigned int desc_idx)
+{
+#ifdef MACB_EXT_DESC
+	switch (bp->hw_dma_cap) {
+	case HW_DMA_CAP_64B:
+	case HW_DMA_CAP_PTP:
+		desc_idx <<= 1;
+		break;
+	case HW_DMA_CAP_64B_PTP:
+		desc_idx *= 3;
+		break;
+	default:
+		break;
+	}
+#endif
+	return desc_idx;
+}
+
+#ifdef CONFIG_ARCH_DMA_ADDR_T_64BIT
+static struct macb_dma_desc_64 *macb_64b_desc(struct macb *bp, struct macb_dma_desc *desc)
+{
+	return (struct macb_dma_desc_64 *)((void *)desc
+		+ sizeof(struct macb_dma_desc));
+}
+#endif
+
+/* Ring buffer accessors */
+static unsigned int macb_tx_ring_wrap(struct macb *bp, unsigned int index)
+{
+	return index & (bp->tx_ring_size - 1);
+}
+
+static struct macb_dma_desc *macb_tx_desc(struct macb_queue *queue,
+					  unsigned int index)
+{
+	index = macb_tx_ring_wrap(queue->bp, index);
+	index = macb_adj_dma_desc_idx(queue->bp, index);
+	return &queue->tx_ring[index];
+}
+
+static struct macb_tx_skb *macb_tx_skb(struct macb_queue *queue,
+				       unsigned int index)
+{
+	return &queue->tx_skb[macb_tx_ring_wrap(queue->bp, index)];
+}
+
+static dma_addr_t macb_tx_dma(struct macb_queue *queue, unsigned int index)
+{
+	dma_addr_t offset;
+
+	offset = macb_tx_ring_wrap(queue->bp, index) *
+			macb_dma_desc_get_size(queue->bp);
+
+	return queue->tx_ring_dma + offset;
+}
+
+static unsigned int macb_rx_ring_wrap(struct macb *bp, unsigned int index)
+{
+	return index & (bp->rx_ring_size - 1);
+}
+
+static struct macb_dma_desc *macb_rx_desc(struct macb_queue *queue, unsigned int index)
+{
+	index = macb_rx_ring_wrap(queue->bp, index);
+	index = macb_adj_dma_desc_idx(queue->bp, index);
+	return &queue->rx_ring[index];
+}
+
+static void *macb_rx_buffer(struct macb_queue *queue, unsigned int index)
+{
+	return queue->rx_buffers + queue->bp->rx_buffer_size *
+	       macb_rx_ring_wrap(queue->bp, index);
+}
+
+/* I/O accessors */
+static u32 hw_readl_native(struct macb *bp, int offset)
+{
+	return __raw_readl(bp->regs + offset);
+}
+
+static void hw_writel_native(struct macb *bp, int offset, u32 value)
+{
+	__raw_writel(value, bp->regs + offset);
+}
+
+static u32 hw_readl(struct macb *bp, int offset)
+{
+	return readl_relaxed(bp->regs + offset);
+}
+
+static void hw_writel(struct macb *bp, int offset, u32 value)
+{
+	writel_relaxed(value, bp->regs + offset);
+}
+
+/* Find the CPU endianness by using the loopback bit of NCR register. When the
+ * CPU is in big endian we need to program swapped mode for management
+ * descriptor access.
+ */
+static bool hw_is_native_io(void __iomem *addr)
+{
+	u32 value = MACB_BIT(LLB);
+
+	__raw_writel(value, addr + MACB_NCR);
+	value = __raw_readl(addr + MACB_NCR);
+
+	/* Write 0 back to disable everything */
+	__raw_writel(0, addr + MACB_NCR);
+
+	return value == MACB_BIT(LLB);
+}
+
+static bool hw_is_gem(void __iomem *addr, bool native_io)
+{
+	u32 id;
+
+	if (native_io)
+		id = __raw_readl(addr + MACB_MID);
+	else
+		id = readl_relaxed(addr + MACB_MID);
+
+	return MACB_BFEXT(IDNUM, id) >= 0x2;
+}
+
+static void macb_set_hwaddr(struct macb *bp)
+{
+	u32 bottom;
+	u16 top;
+
+	bottom = get_unaligned_le32(bp->dev->dev_addr);
+	macb_or_gem_writel(bp, SA1B, bottom);
+	top = get_unaligned_le16(bp->dev->dev_addr + 4);
+	macb_or_gem_writel(bp, SA1T, top);
+
+	if (gem_has_ptp(bp)) {
+		gem_writel(bp, RXPTPUNI, bottom);
+		gem_writel(bp, TXPTPUNI, bottom);
+	}
+
+	/* Clear unused address register sets */
+	macb_or_gem_writel(bp, SA2B, 0);
+	macb_or_gem_writel(bp, SA2T, 0);
+	macb_or_gem_writel(bp, SA3B, 0);
+	macb_or_gem_writel(bp, SA3T, 0);
+	macb_or_gem_writel(bp, SA4B, 0);
+	macb_or_gem_writel(bp, SA4T, 0);
+}
+
+static void macb_get_hwaddr(struct macb *bp)
+{
+	u32 bottom;
+	u16 top;
+	u8 addr[6];
+	int i;
+
+	/* Check all 4 address register for valid address */
+	for (i = 0; i < 4; i++) {
+		bottom = macb_or_gem_readl(bp, SA1B + i * 8);
+		top = macb_or_gem_readl(bp, SA1T + i * 8);
+
+		addr[0] = bottom & 0xff;
+		addr[1] = (bottom >> 8) & 0xff;
+		addr[2] = (bottom >> 16) & 0xff;
+		addr[3] = (bottom >> 24) & 0xff;
+		addr[4] = top & 0xff;
+		addr[5] = (top >> 8) & 0xff;
+
+		if (is_valid_ether_addr(addr)) {
+			eth_hw_addr_set(bp->dev, addr);
+			return;
+		}
+	}
+
+	dev_info(&bp->pdev->dev, "invalid hw address, using random\n");
+	eth_hw_addr_random(bp->dev);
+}
+
+static int macb_mdio_wait_for_idle(struct macb *bp)
+{
+	u32 val;
+
+	return readx_poll_timeout(MACB_READ_NSR, bp, val, val & MACB_BIT(IDLE),
+				  1, MACB_MDIO_TIMEOUT);
+}
+
+static int macb_mdio_read_c22(struct mii_bus *bus, int mii_id, int regnum)
+{
+	struct macb *bp = bus->priv;
+	int status;
+
+	status = pm_runtime_resume_and_get(&bp->pdev->dev);
+	if (status < 0)
+		goto mdio_pm_exit;
+
+	status = macb_mdio_wait_for_idle(bp);
+	if (status < 0)
+		goto mdio_read_exit;
+
+	macb_writel(bp, MAN, (MACB_BF(SOF, MACB_MAN_C22_SOF)
+			      | MACB_BF(RW, MACB_MAN_C22_READ)
+			      | MACB_BF(PHYA, mii_id)
+			      | MACB_BF(REGA, regnum)
+			      | MACB_BF(CODE, MACB_MAN_C22_CODE)));
+
+	status = macb_mdio_wait_for_idle(bp);
+	if (status < 0)
+		goto mdio_read_exit;
+
+	status = MACB_BFEXT(DATA, macb_readl(bp, MAN));
+
+mdio_read_exit:
+	pm_runtime_mark_last_busy(&bp->pdev->dev);
+	pm_runtime_put_autosuspend(&bp->pdev->dev);
+mdio_pm_exit:
+	return status;
+}
+
+static int macb_mdio_read_c45(struct mii_bus *bus, int mii_id, int devad,
+			      int regnum)
+{
+	struct macb *bp = bus->priv;
+	int status;
+
+	status = pm_runtime_get_sync(&bp->pdev->dev);
+	if (status < 0) {
+		pm_runtime_put_noidle(&bp->pdev->dev);
+		goto mdio_pm_exit;
+	}
+
+	status = macb_mdio_wait_for_idle(bp);
+	if (status < 0)
+		goto mdio_read_exit;
+
+	macb_writel(bp, MAN, (MACB_BF(SOF, MACB_MAN_C45_SOF)
+			      | MACB_BF(RW, MACB_MAN_C45_ADDR)
+			      | MACB_BF(PHYA, mii_id)
+			      | MACB_BF(REGA, devad & 0x1F)
+			      | MACB_BF(DATA, regnum & 0xFFFF)
+			      | MACB_BF(CODE, MACB_MAN_C45_CODE)));
+
+	status = macb_mdio_wait_for_idle(bp);
+	if (status < 0)
+		goto mdio_read_exit;
+
+	macb_writel(bp, MAN, (MACB_BF(SOF, MACB_MAN_C45_SOF)
+			      | MACB_BF(RW, MACB_MAN_C45_READ)
+			      | MACB_BF(PHYA, mii_id)
+			      | MACB_BF(REGA, devad & 0x1F)
+			      | MACB_BF(CODE, MACB_MAN_C45_CODE)));
+
+	status = macb_mdio_wait_for_idle(bp);
+	if (status < 0)
+		goto mdio_read_exit;
+
+	status = MACB_BFEXT(DATA, macb_readl(bp, MAN));
+
+mdio_read_exit:
+	pm_runtime_mark_last_busy(&bp->pdev->dev);
+	pm_runtime_put_autosuspend(&bp->pdev->dev);
+mdio_pm_exit:
+	return status;
+}
+
+static int macb_mdio_write_c22(struct mii_bus *bus, int mii_id, int regnum,
+			       u16 value)
+{
+	struct macb *bp = bus->priv;
+	int status;
+
+	status = pm_runtime_resume_and_get(&bp->pdev->dev);
+	if (status < 0)
+		goto mdio_pm_exit;
+
+	status = macb_mdio_wait_for_idle(bp);
+	if (status < 0)
+		goto mdio_write_exit;
+
+	macb_writel(bp, MAN, (MACB_BF(SOF, MACB_MAN_C22_SOF)
+			      | MACB_BF(RW, MACB_MAN_C22_WRITE)
+			      | MACB_BF(PHYA, mii_id)
+			      | MACB_BF(REGA, regnum)
+			      | MACB_BF(CODE, MACB_MAN_C22_CODE)
+			      | MACB_BF(DATA, value)));
+
+	status = macb_mdio_wait_for_idle(bp);
+	if (status < 0)
+		goto mdio_write_exit;
+
+mdio_write_exit:
+	pm_runtime_mark_last_busy(&bp->pdev->dev);
+	pm_runtime_put_autosuspend(&bp->pdev->dev);
+mdio_pm_exit:
+	return status;
+}
+
+static int macb_mdio_write_c45(struct mii_bus *bus, int mii_id,
+			       int devad, int regnum,
+			       u16 value)
+{
+	struct macb *bp = bus->priv;
+	int status;
+
+	status = pm_runtime_get_sync(&bp->pdev->dev);
+	if (status < 0) {
+		pm_runtime_put_noidle(&bp->pdev->dev);
+		goto mdio_pm_exit;
+	}
+
+	status = macb_mdio_wait_for_idle(bp);
+	if (status < 0)
+		goto mdio_write_exit;
+
+	macb_writel(bp, MAN, (MACB_BF(SOF, MACB_MAN_C45_SOF)
+			      | MACB_BF(RW, MACB_MAN_C45_ADDR)
+			      | MACB_BF(PHYA, mii_id)
+			      | MACB_BF(REGA, devad & 0x1F)
+			      | MACB_BF(DATA, regnum & 0xFFFF)
+			      | MACB_BF(CODE, MACB_MAN_C45_CODE)));
+
+	status = macb_mdio_wait_for_idle(bp);
+	if (status < 0)
+		goto mdio_write_exit;
+
+	macb_writel(bp, MAN, (MACB_BF(SOF, MACB_MAN_C45_SOF)
+			      | MACB_BF(RW, MACB_MAN_C45_WRITE)
+			      | MACB_BF(PHYA, mii_id)
+			      | MACB_BF(REGA, devad & 0x1F)
+			      | MACB_BF(CODE, MACB_MAN_C45_CODE)
+			      | MACB_BF(DATA, value)));
+
+	status = macb_mdio_wait_for_idle(bp);
+	if (status < 0)
+		goto mdio_write_exit;
+
+mdio_write_exit:
+	pm_runtime_mark_last_busy(&bp->pdev->dev);
+	pm_runtime_put_autosuspend(&bp->pdev->dev);
+mdio_pm_exit:
+	return status;
+}
+
+static void macb_init_buffers(struct macb *bp)
+{
+	struct macb_queue *queue;
+	unsigned int q;
+
+#ifdef CONFIG_ARCH_DMA_ADDR_T_64BIT
+	/* Single register for all queues' high 32 bits. */
+	if (bp->hw_dma_cap & HW_DMA_CAP_64B) {
+		macb_writel(bp, RBQPH,
+			    upper_32_bits(bp->queues[0].rx_ring_dma));
+		macb_writel(bp, TBQPH,
+			    upper_32_bits(bp->queues[0].tx_ring_dma));
+	}
+#endif
+
+	for (q = 0, queue = bp->queues; q < bp->num_queues; ++q, ++queue) {
+		queue_writel(queue, RBQP, lower_32_bits(queue->rx_ring_dma));
+		queue_writel(queue, TBQP, lower_32_bits(queue->tx_ring_dma));
+	}
+}
+
+/**
+ * macb_set_tx_clk() - Set a clock to a new frequency
+ * @bp:		pointer to struct macb
+ * @speed:	New frequency in Hz
+ */
+static void macb_set_tx_clk(struct macb *bp, int speed)
+{
+	long ferr, rate, rate_rounded;
+
+	if (!bp->tx_clk || (bp->caps & MACB_CAPS_CLK_HW_CHG))
+		return;
+
+	/* In case of MII the PHY is the clock master */
+	if (bp->phy_interface == PHY_INTERFACE_MODE_MII)
+		return;
+
+	rate = rgmii_clock(speed);
+	if (rate < 0)
+		return;
+
+	rate_rounded = clk_round_rate(bp->tx_clk, rate);
+	if (rate_rounded < 0)
+		return;
+
+	/* RGMII allows 50 ppm frequency error. Test and warn if this limit
+	 * is not satisfied.
+	 */
+	ferr = abs(rate_rounded - rate);
+	ferr = DIV_ROUND_UP(ferr, rate / 100000);
+	if (ferr > 5)
+		netdev_warn(bp->dev,
+			    "unable to generate target frequency: %ld Hz\n",
+			    rate);
+
+	if (clk_set_rate(bp->tx_clk, rate_rounded))
+		netdev_err(bp->dev, "adjusting tx_clk failed.\n");
+}
+
+static void macb_usx_pcs_link_up(struct phylink_pcs *pcs, unsigned int neg_mode,
+				 phy_interface_t interface, int speed,
+				 int duplex)
+{
+	struct macb *bp = container_of(pcs, struct macb, phylink_usx_pcs);
+	u32 config;
+
+	config = gem_readl(bp, USX_CONTROL);
+	config = GEM_BFINS(SERDES_RATE, MACB_SERDES_RATE_10G, config);
+	config = GEM_BFINS(USX_CTRL_SPEED, HS_SPEED_10000M, config);
+	config &= ~(GEM_BIT(TX_SCR_BYPASS) | GEM_BIT(RX_SCR_BYPASS));
+	config |= GEM_BIT(TX_EN);
+	gem_writel(bp, USX_CONTROL, config);
+}
+
+static void macb_usx_pcs_get_state(struct phylink_pcs *pcs,
+				   unsigned int neg_mode,
+				   struct phylink_link_state *state)
+{
+	struct macb *bp = container_of(pcs, struct macb, phylink_usx_pcs);
+	u32 val;
+
+	state->speed = SPEED_10000;
+	state->duplex = 1;
+	state->an_complete = 1;
+
+	val = gem_readl(bp, USX_STATUS);
+	state->link = !!(val & GEM_BIT(USX_BLOCK_LOCK));
+	val = gem_readl(bp, NCFGR);
+	if (val & GEM_BIT(PAE))
+		state->pause = MLO_PAUSE_RX;
+}
+
+static int macb_usx_pcs_config(struct phylink_pcs *pcs,
+			       unsigned int neg_mode,
+			       phy_interface_t interface,
+			       const unsigned long *advertising,
+			       bool permit_pause_to_mac)
+{
+	struct macb *bp = container_of(pcs, struct macb, phylink_usx_pcs);
+
+	gem_writel(bp, USX_CONTROL, gem_readl(bp, USX_CONTROL) |
+		   GEM_BIT(SIGNAL_OK));
+
+	return 0;
+}
+
+static void macb_pcs_get_state(struct phylink_pcs *pcs, unsigned int neg_mode,
+			       struct phylink_link_state *state)
+{
+	state->link = 0;
+}
+
+static void macb_pcs_an_restart(struct phylink_pcs *pcs)
+{
+	/* Not supported */
+}
+
+static int macb_pcs_config(struct phylink_pcs *pcs,
+			   unsigned int neg_mode,
+			   phy_interface_t interface,
+			   const unsigned long *advertising,
+			   bool permit_pause_to_mac)
+{
+	return 0;
+}
+
+static const struct phylink_pcs_ops macb_phylink_usx_pcs_ops = {
+	.pcs_get_state = macb_usx_pcs_get_state,
+	.pcs_config = macb_usx_pcs_config,
+	.pcs_link_up = macb_usx_pcs_link_up,
+};
+
+static const struct phylink_pcs_ops macb_phylink_pcs_ops = {
+	.pcs_get_state = macb_pcs_get_state,
+	.pcs_an_restart = macb_pcs_an_restart,
+	.pcs_config = macb_pcs_config,
+};
+
+static void macb_mac_config(struct phylink_config *config, unsigned int mode,
+			    const struct phylink_link_state *state)
+{
+	struct net_device *ndev = to_net_dev(config->dev);
+	struct macb *bp = netdev_priv(ndev);
+	unsigned long flags;
+	u32 old_ctrl, ctrl;
+	u32 old_ncr, ncr;
+
+	spin_lock_irqsave(&bp->lock, flags);
+
+	old_ctrl = ctrl = macb_or_gem_readl(bp, NCFGR);
+	old_ncr = ncr = macb_or_gem_readl(bp, NCR);
+
+	if (bp->caps & MACB_CAPS_MACB_IS_EMAC) {
+		if (state->interface == PHY_INTERFACE_MODE_RMII)
+			ctrl |= MACB_BIT(RM9200_RMII);
+	} else if (macb_is_gem(bp)) {
+		ctrl &= ~(GEM_BIT(SGMIIEN) | GEM_BIT(PCSSEL));
+		ncr &= ~GEM_BIT(ENABLE_HS_MAC);
+
+		if (state->interface == PHY_INTERFACE_MODE_SGMII) {
+			ctrl |= GEM_BIT(SGMIIEN) | GEM_BIT(PCSSEL);
+		} else if (state->interface == PHY_INTERFACE_MODE_10GBASER) {
+			ctrl |= GEM_BIT(PCSSEL);
+			ncr |= GEM_BIT(ENABLE_HS_MAC);
+		} else if (bp->caps & MACB_CAPS_MIIONRGMII &&
+			   bp->phy_interface == PHY_INTERFACE_MODE_MII) {
+			ncr |= MACB_BIT(MIIONRGMII);
+		}
+	}
+
+	/* Apply the new configuration, if any */
+	if (old_ctrl ^ ctrl)
+		macb_or_gem_writel(bp, NCFGR, ctrl);
+
+	if (old_ncr ^ ncr)
+		macb_or_gem_writel(bp, NCR, ncr);
+
+	/* Disable AN for SGMII fixed link configuration, enable otherwise.
+	 * Must be written after PCSSEL is set in NCFGR,
+	 * otherwise writes will not take effect.
+	 */
+	if (macb_is_gem(bp) && state->interface == PHY_INTERFACE_MODE_SGMII) {
+		u32 pcsctrl, old_pcsctrl;
+
+		old_pcsctrl = gem_readl(bp, PCSCNTRL);
+		if (mode == MLO_AN_FIXED)
+			pcsctrl = old_pcsctrl & ~GEM_BIT(PCSAUTONEG);
+		else
+			pcsctrl = old_pcsctrl | GEM_BIT(PCSAUTONEG);
+		if (old_pcsctrl != pcsctrl)
+			gem_writel(bp, PCSCNTRL, pcsctrl);
+	}
+
+	spin_unlock_irqrestore(&bp->lock, flags);
+}
+
+static void macb_mac_link_down(struct phylink_config *config, unsigned int mode,
+			       phy_interface_t interface)
+{
+	struct net_device *ndev = to_net_dev(config->dev);
+	struct macb *bp = netdev_priv(ndev);
+	struct macb_queue *queue;
+	unsigned int q;
+	u32 ctrl;
+
+	if (!(bp->caps & MACB_CAPS_MACB_IS_EMAC))
+		for (q = 0, queue = bp->queues; q < bp->num_queues; ++q, ++queue)
+			queue_writel(queue, IDR,
+				     bp->rx_intr_mask | MACB_TX_INT_FLAGS | MACB_BIT(HRESP));
+
+	/* Disable Rx and Tx */
+	ctrl = macb_readl(bp, NCR) & ~(MACB_BIT(RE) | MACB_BIT(TE));
+	macb_writel(bp, NCR, ctrl);
+
+	netif_tx_stop_all_queues(ndev);
+}
+
+static void macb_mac_link_up(struct phylink_config *config,
+			     struct phy_device *phy,
+			     unsigned int mode, phy_interface_t interface,
+			     int speed, int duplex,
+			     bool tx_pause, bool rx_pause)
+{
+	struct net_device *ndev = to_net_dev(config->dev);
+	struct macb *bp = netdev_priv(ndev);
+	struct macb_queue *queue;
+	unsigned long flags;
+	unsigned int q;
+	u32 ctrl;
+
+	spin_lock_irqsave(&bp->lock, flags);
+
+	ctrl = macb_or_gem_readl(bp, NCFGR);
+
+	ctrl &= ~(MACB_BIT(SPD) | MACB_BIT(FD));
+
+	if (speed == SPEED_100)
+		ctrl |= MACB_BIT(SPD);
+
+	if (duplex)
+		ctrl |= MACB_BIT(FD);
+
+	if (!(bp->caps & MACB_CAPS_MACB_IS_EMAC)) {
+		ctrl &= ~MACB_BIT(PAE);
+		if (macb_is_gem(bp)) {
+			ctrl &= ~GEM_BIT(GBE);
+
+			if (speed == SPEED_1000)
+				ctrl |= GEM_BIT(GBE);
+		}
+
+		if (rx_pause)
+			ctrl |= MACB_BIT(PAE);
+
+		/* Initialize rings & buffers as clearing MACB_BIT(TE) in link down
+		 * cleared the pipeline and control registers.
+		 */
+		bp->macbgem_ops.mog_init_rings(bp);
+		macb_init_buffers(bp);
+
+		for (q = 0, queue = bp->queues; q < bp->num_queues; ++q, ++queue)
+			queue_writel(queue, IER,
+				     bp->rx_intr_mask | MACB_TX_INT_FLAGS | MACB_BIT(HRESP));
+	}
+
+	macb_or_gem_writel(bp, NCFGR, ctrl);
+
+	if (bp->phy_interface == PHY_INTERFACE_MODE_10GBASER)
+		gem_writel(bp, HS_MAC_CONFIG, GEM_BFINS(HS_MAC_SPEED, HS_SPEED_10000M,
+							gem_readl(bp, HS_MAC_CONFIG)));
+
+	spin_unlock_irqrestore(&bp->lock, flags);
+
+	if (!(bp->caps & MACB_CAPS_MACB_IS_EMAC))
+		macb_set_tx_clk(bp, speed);
+
+	/* Enable Rx and Tx; Enable PTP unicast */
+	ctrl = macb_readl(bp, NCR);
+	if (gem_has_ptp(bp))
+		ctrl |= MACB_BIT(PTPUNI);
+
+	macb_writel(bp, NCR, ctrl | MACB_BIT(RE) | MACB_BIT(TE));
+
+	netif_tx_wake_all_queues(ndev);
+}
+
+static struct phylink_pcs *macb_mac_select_pcs(struct phylink_config *config,
+					       phy_interface_t interface)
+{
+	struct net_device *ndev = to_net_dev(config->dev);
+	struct macb *bp = netdev_priv(ndev);
+
+	if (interface == PHY_INTERFACE_MODE_10GBASER)
+		return &bp->phylink_usx_pcs;
+	else if (interface == PHY_INTERFACE_MODE_SGMII)
+		return &bp->phylink_sgmii_pcs;
+	else
+		return NULL;
+}
+
+static const struct phylink_mac_ops macb_phylink_ops = {
+	.mac_select_pcs = macb_mac_select_pcs,
+	.mac_config = macb_mac_config,
+	.mac_link_down = macb_mac_link_down,
+	.mac_link_up = macb_mac_link_up,
+};
+
+static bool macb_phy_handle_exists(struct device_node *dn)
+{
+	dn = of_parse_phandle(dn, "phy-handle", 0);
+	of_node_put(dn);
+	return dn != NULL;
+}
+
+static int macb_phylink_connect(struct macb *bp)
+{
+	struct device_node *dn = bp->pdev->dev.of_node;
+	struct net_device *dev = bp->dev;
+	struct phy_device *phydev;
+	int ret;
+
+	if (dn)
+		ret = phylink_of_phy_connect(bp->phylink, dn, 0);
+
+	if (!dn || (ret && !macb_phy_handle_exists(dn))) {
+		phydev = phy_find_first(bp->mii_bus);
+		if (!phydev) {
+			netdev_err(dev, "no PHY found\n");
+			return -ENXIO;
+		}
+
+		/* attach the mac to the phy */
+		ret = phylink_connect_phy(bp->phylink, phydev);
+	}
+
+	if (ret) {
+		netdev_err(dev, "Could not attach PHY (%d)\n", ret);
+		return ret;
+	}
+
+	phylink_start(bp->phylink);
+
+	return 0;
+}
+
+static void macb_get_pcs_fixed_state(struct phylink_config *config,
+				     struct phylink_link_state *state)
+{
+	struct net_device *ndev = to_net_dev(config->dev);
+	struct macb *bp = netdev_priv(ndev);
+
+	state->link = (macb_readl(bp, NSR) & MACB_BIT(NSR_LINK)) != 0;
+}
+
+/* based on au1000_eth. c*/
+static int macb_mii_probe(struct net_device *dev)
+{
+	struct macb *bp = netdev_priv(dev);
+
+	bp->phylink_sgmii_pcs.ops = &macb_phylink_pcs_ops;
+	bp->phylink_usx_pcs.ops = &macb_phylink_usx_pcs_ops;
+
+	bp->phylink_config.dev = &dev->dev;
+	bp->phylink_config.type = PHYLINK_NETDEV;
+	bp->phylink_config.mac_managed_pm = true;
+
+	if (bp->phy_interface == PHY_INTERFACE_MODE_SGMII) {
+		bp->phylink_config.poll_fixed_state = true;
+		bp->phylink_config.get_fixed_state = macb_get_pcs_fixed_state;
+	}
+
+	bp->phylink_config.mac_capabilities = MAC_ASYM_PAUSE |
+		MAC_10 | MAC_100;
+
+	__set_bit(PHY_INTERFACE_MODE_MII,
+		  bp->phylink_config.supported_interfaces);
+	__set_bit(PHY_INTERFACE_MODE_RMII,
+		  bp->phylink_config.supported_interfaces);
+
+	/* Determine what modes are supported */
+	if (macb_is_gem(bp) && (bp->caps & MACB_CAPS_GIGABIT_MODE_AVAILABLE)) {
+		bp->phylink_config.mac_capabilities |= MAC_1000FD;
+		if (!(bp->caps & MACB_CAPS_NO_GIGABIT_HALF))
+			bp->phylink_config.mac_capabilities |= MAC_1000HD;
+
+		__set_bit(PHY_INTERFACE_MODE_GMII,
+			  bp->phylink_config.supported_interfaces);
+		phy_interface_set_rgmii(bp->phylink_config.supported_interfaces);
+
+		if (bp->caps & MACB_CAPS_PCS)
+			__set_bit(PHY_INTERFACE_MODE_SGMII,
+				  bp->phylink_config.supported_interfaces);
+
+		if (bp->caps & MACB_CAPS_HIGH_SPEED) {
+			__set_bit(PHY_INTERFACE_MODE_10GBASER,
+				  bp->phylink_config.supported_interfaces);
+			bp->phylink_config.mac_capabilities |= MAC_10000FD;
+		}
+	}
+
+	bp->phylink = phylink_create(&bp->phylink_config, bp->pdev->dev.fwnode,
+				     bp->phy_interface, &macb_phylink_ops);
+	if (IS_ERR(bp->phylink)) {
+		netdev_err(dev, "Could not create a phylink instance (%ld)\n",
+			   PTR_ERR(bp->phylink));
+		return PTR_ERR(bp->phylink);
+	}
+
+	return 0;
+}
+
+static int macb_mdiobus_register(struct macb *bp, struct device_node *mdio_np)
+{
+	struct device_node *child, *np = bp->pdev->dev.of_node;
+
+	/* If we have a child named mdio, probe it instead of looking for PHYs
+	 * directly under the MAC node
+	 */
+	if (mdio_np)
+		return of_mdiobus_register(bp->mii_bus, mdio_np);
+
+	/* Only create the PHY from the device tree if at least one PHY is
+	 * described. Otherwise scan the entire MDIO bus. We do this to support
+	 * old device tree that did not follow the best practices and did not
+	 * describe their network PHYs.
+	 */
+	for_each_available_child_of_node(np, child)
+		if (of_mdiobus_child_is_phy(child)) {
+			/* The loop increments the child refcount,
+			 * decrement it before returning.
+			 */
+			of_node_put(child);
+
+			return of_mdiobus_register(bp->mii_bus, np);
+		}
+
+	return mdiobus_register(bp->mii_bus);
+}
+
+static int macb_mii_init(struct macb *bp)
+{
+	struct device_node *mdio_np, *np = bp->pdev->dev.of_node;
+	int err = -ENXIO;
+
+	/* With fixed-link, we don't need to register the MDIO bus,
+	 * except if we have a child named "mdio" in the device tree.
+	 * In that case, some devices may be attached to the MACB's MDIO bus.
+	 */
+	mdio_np = of_get_child_by_name(np, "mdio");
+	if (!mdio_np && of_phy_is_fixed_link(np))
+		return macb_mii_probe(bp->dev);
+
+	/* Enable management port */
+	macb_writel(bp, NCR, MACB_BIT(MPE));
+
+	bp->mii_bus = mdiobus_alloc();
+	if (!bp->mii_bus) {
+		err = -ENOMEM;
+		goto err_out;
+	}
+
+	bp->mii_bus->name = "MACB_mii_bus";
+	bp->mii_bus->read = &macb_mdio_read_c22;
+	bp->mii_bus->write = &macb_mdio_write_c22;
+	bp->mii_bus->read_c45 = &macb_mdio_read_c45;
+	bp->mii_bus->write_c45 = &macb_mdio_write_c45;
+	snprintf(bp->mii_bus->id, MII_BUS_ID_SIZE, "%s-%x",
+		 bp->pdev->name, bp->pdev->id);
+	bp->mii_bus->priv = bp;
+	bp->mii_bus->parent = &bp->pdev->dev;
+
+	dev_set_drvdata(&bp->dev->dev, bp->mii_bus);
+
+	err = macb_mdiobus_register(bp, mdio_np);
+	if (err)
+		goto err_out_free_mdiobus;
+
+	err = macb_mii_probe(bp->dev);
+	if (err)
+		goto err_out_unregister_bus;
+
+	return 0;
+
+err_out_unregister_bus:
+	mdiobus_unregister(bp->mii_bus);
+err_out_free_mdiobus:
+	mdiobus_free(bp->mii_bus);
+err_out:
+	of_node_put(mdio_np);
+
+	return err;
+}
+
+static void macb_update_stats(struct macb *bp)
+{
+	u64 *p = &bp->hw_stats.macb.rx_pause_frames;
+	u64 *end = &bp->hw_stats.macb.tx_pause_frames + 1;
+	int offset = MACB_PFR;
+
+	WARN_ON((unsigned long)(end - p - 1) != (MACB_TPF - MACB_PFR) / 4);
+
+	for (; p < end; p++, offset += 4)
+		*p += bp->macb_reg_readl(bp, offset);
+}
+
+static int macb_halt_tx(struct macb *bp)
+{
+	u32 status;
+
+	macb_writel(bp, NCR, macb_readl(bp, NCR) | MACB_BIT(THALT));
+
+	/* Poll TSR until TGO is cleared or timeout. */
+	return read_poll_timeout_atomic(macb_readl, status,
+					!(status & MACB_BIT(TGO)),
+					250, MACB_HALT_TIMEOUT, false,
+					bp, TSR);
+}
+
+static void macb_tx_unmap(struct macb *bp, struct macb_tx_skb *tx_skb, int budget)
+{
+	if (tx_skb->mapping) {
+		if (tx_skb->mapped_as_page)
+			dma_unmap_page(&bp->pdev->dev, tx_skb->mapping,
+				       tx_skb->size, DMA_TO_DEVICE);
+		else
+			dma_unmap_single(&bp->pdev->dev, tx_skb->mapping,
+					 tx_skb->size, DMA_TO_DEVICE);
+		tx_skb->mapping = 0;
+	}
+
+	if (tx_skb->skb) {
+		napi_consume_skb(tx_skb->skb, budget);
+		tx_skb->skb = NULL;
+	}
+}
+
+static void macb_set_addr(struct macb *bp, struct macb_dma_desc *desc, dma_addr_t addr)
+{
+#ifdef CONFIG_ARCH_DMA_ADDR_T_64BIT
+	struct macb_dma_desc_64 *desc_64;
+
+	if (bp->hw_dma_cap & HW_DMA_CAP_64B) {
+		desc_64 = macb_64b_desc(bp, desc);
+		desc_64->addrh = upper_32_bits(addr);
+		/* The low bits of RX address contain the RX_USED bit, clearing
+		 * of which allows packet RX. Make sure the high bits are also
+		 * visible to HW at that point.
+		 */
+		dma_wmb();
+	}
+#endif
+	desc->addr = lower_32_bits(addr);
+}
+
+static dma_addr_t macb_get_addr(struct macb *bp, struct macb_dma_desc *desc)
+{
+	dma_addr_t addr = 0;
+#ifdef CONFIG_ARCH_DMA_ADDR_T_64BIT
+	struct macb_dma_desc_64 *desc_64;
+
+	if (bp->hw_dma_cap & HW_DMA_CAP_64B) {
+		desc_64 = macb_64b_desc(bp, desc);
+		addr = ((u64)(desc_64->addrh) << 32);
+	}
+#endif
+	addr |= MACB_BF(RX_WADDR, MACB_BFEXT(RX_WADDR, desc->addr));
+#ifdef CONFIG_MACB_USE_HWSTAMP
+	if (bp->hw_dma_cap & HW_DMA_CAP_PTP)
+		addr &= ~GEM_BIT(DMA_RXVALID);
+#endif
+	return addr;
+}
+
+static void macb_tx_error_task(struct work_struct *work)
+{
+	struct macb_queue	*queue = container_of(work, struct macb_queue,
+						      tx_error_task);
+	bool			halt_timeout = false;
+	struct macb		*bp = queue->bp;
+	u32			queue_index;
+	u32			packets = 0;
+	u32			bytes = 0;
+	struct macb_tx_skb	*tx_skb;
+	struct macb_dma_desc	*desc;
+	struct sk_buff		*skb;
+	unsigned int		tail;
+	unsigned long		flags;
+
+	queue_index = queue - bp->queues;
+	netdev_vdbg(bp->dev, "macb_tx_error_task: q = %u, t = %u, h = %u\n",
+		    queue_index, queue->tx_tail, queue->tx_head);
+
+	/* Prevent the queue NAPI TX poll from running, as it calls
+	 * macb_tx_complete(), which in turn may call netif_wake_subqueue().
+	 * As explained below, we have to halt the transmission before updating
+	 * TBQP registers so we call netif_tx_stop_all_queues() to notify the
+	 * network engine about the macb/gem being halted.
+	 */
+	napi_disable(&queue->napi_tx);
+	spin_lock_irqsave(&bp->lock, flags);
+
+	/* Make sure nobody is trying to queue up new packets */
+	netif_tx_stop_all_queues(bp->dev);
+
+	/* Stop transmission now
+	 * (in case we have just queued new packets)
+	 * macb/gem must be halted to write TBQP register
+	 */
+	if (macb_halt_tx(bp)) {
+		netdev_err(bp->dev, "BUG: halt tx timed out\n");
+		macb_writel(bp, NCR, macb_readl(bp, NCR) & (~MACB_BIT(TE)));
+		halt_timeout = true;
+	}
+
+	/* Treat frames in TX queue including the ones that caused the error.
+	 * Free transmit buffers in upper layer.
+	 */
+	for (tail = queue->tx_tail; tail != queue->tx_head; tail++) {
+		u32	ctrl;
+
+		desc = macb_tx_desc(queue, tail);
+		ctrl = desc->ctrl;
+		tx_skb = macb_tx_skb(queue, tail);
+		skb = tx_skb->skb;
+
+		if (ctrl & MACB_BIT(TX_USED)) {
+			/* skb is set for the last buffer of the frame */
+			while (!skb) {
+				macb_tx_unmap(bp, tx_skb, 0);
+				tail++;
+				tx_skb = macb_tx_skb(queue, tail);
+				skb = tx_skb->skb;
+			}
+
+			/* ctrl still refers to the first buffer descriptor
+			 * since it's the only one written back by the hardware
+			 */
+			if (!(ctrl & MACB_BIT(TX_BUF_EXHAUSTED))) {
+				netdev_vdbg(bp->dev, "txerr skb %u (data %p) TX complete\n",
+					    macb_tx_ring_wrap(bp, tail),
+					    skb->data);
+				bp->dev->stats.tx_packets++;
+				queue->stats.tx_packets++;
+				packets++;
+				bp->dev->stats.tx_bytes += skb->len;
+				queue->stats.tx_bytes += skb->len;
+				bytes += skb->len;
+			}
+		} else {
+			/* "Buffers exhausted mid-frame" errors may only happen
+			 * if the driver is buggy, so complain loudly about
+			 * those. Statistics are updated by hardware.
+			 */
+			if (ctrl & MACB_BIT(TX_BUF_EXHAUSTED))
+				netdev_err(bp->dev,
+					   "BUG: TX buffers exhausted mid-frame\n");
+
+			desc->ctrl = ctrl | MACB_BIT(TX_USED);
+		}
+
+		macb_tx_unmap(bp, tx_skb, 0);
+	}
+
+	netdev_tx_completed_queue(netdev_get_tx_queue(bp->dev, queue_index),
+				  packets, bytes);
+
+	/* Set end of TX queue */
+	desc = macb_tx_desc(queue, 0);
+	macb_set_addr(bp, desc, 0);
+	desc->ctrl = MACB_BIT(TX_USED);
+
+	/* Make descriptor updates visible to hardware */
+	wmb();
+
+	/* Reinitialize the TX desc queue */
+	queue_writel(queue, TBQP, lower_32_bits(queue->tx_ring_dma));
+	/* Make TX ring reflect state of hardware */
+	queue->tx_head = 0;
+	queue->tx_tail = 0;
+
+	/* Housework before enabling TX IRQ */
+	macb_writel(bp, TSR, macb_readl(bp, TSR));
+	queue_writel(queue, IER, MACB_TX_INT_FLAGS);
+
+	if (halt_timeout)
+		macb_writel(bp, NCR, macb_readl(bp, NCR) | MACB_BIT(TE));
+
+	/* Now we are ready to start transmission again */
+	netif_tx_start_all_queues(bp->dev);
+	macb_writel(bp, NCR, macb_readl(bp, NCR) | MACB_BIT(TSTART));
+
+	spin_unlock_irqrestore(&bp->lock, flags);
+	napi_enable(&queue->napi_tx);
+}
+
+static bool ptp_one_step_sync(struct sk_buff *skb)
+{
+	struct ptp_header *hdr;
+	unsigned int ptp_class;
+	u8 msgtype;
+
+	/* No need to parse packet if PTP TS is not involved */
+	if (likely(!(skb_shinfo(skb)->tx_flags & SKBTX_HW_TSTAMP)))
+		goto not_oss;
+
+	/* Identify and return whether PTP one step sync is being processed */
+	ptp_class = ptp_classify_raw(skb);
+	if (ptp_class == PTP_CLASS_NONE)
+		goto not_oss;
+
+	hdr = ptp_parse_header(skb, ptp_class);
+	if (!hdr)
+		goto not_oss;
+
+	if (hdr->flag_field[0] & PTP_FLAG_TWOSTEP)
+		goto not_oss;
+
+	msgtype = ptp_get_msgtype(hdr, ptp_class);
+	if (msgtype == PTP_MSGTYPE_SYNC)
+		return true;
+
+not_oss:
+	return false;
+}
+
+static int macb_tx_complete(struct macb_queue *queue, int budget)
+{
+	struct macb *bp = queue->bp;
+	u16 queue_index = queue - bp->queues;
+	unsigned long flags;
+	unsigned int tail;
+	unsigned int head;
+	int packets = 0;
+	u32 bytes = 0;
+
+	spin_lock_irqsave(&queue->tx_ptr_lock, flags);
+	head = queue->tx_head;
+	for (tail = queue->tx_tail; tail != head && packets < budget; tail++) {
+		struct macb_tx_skb	*tx_skb;
+		struct sk_buff		*skb;
+		struct macb_dma_desc	*desc;
+		u32			ctrl;
+
+		desc = macb_tx_desc(queue, tail);
+
+		/* Make hw descriptor updates visible to CPU */
+		rmb();
+
+		ctrl = desc->ctrl;
+
+		/* TX_USED bit is only set by hardware on the very first buffer
+		 * descriptor of the transmitted frame.
+		 */
+		if (!(ctrl & MACB_BIT(TX_USED)))
+			break;
+
+		/* Process all buffers of the current transmitted frame */
+		for (;; tail++) {
+			tx_skb = macb_tx_skb(queue, tail);
+			skb = tx_skb->skb;
+
+			/* First, update TX stats if needed */
+			if (skb) {
+				if (unlikely(skb_shinfo(skb)->tx_flags & SKBTX_HW_TSTAMP) &&
+				    !ptp_one_step_sync(skb))
+					gem_ptp_do_txstamp(bp, skb, desc);
+
+				netdev_vdbg(bp->dev, "skb %u (data %p) TX complete\n",
+					    macb_tx_ring_wrap(bp, tail),
+					    skb->data);
+				bp->dev->stats.tx_packets++;
+				queue->stats.tx_packets++;
+				bp->dev->stats.tx_bytes += skb->len;
+				queue->stats.tx_bytes += skb->len;
+				packets++;
+				bytes += skb->len;
+			}
+
+			/* Now we can safely release resources */
+			macb_tx_unmap(bp, tx_skb, budget);
+
+			/* skb is set only for the last buffer of the frame.
+			 * WARNING: at this point skb has been freed by
+			 * macb_tx_unmap().
+			 */
+			if (skb)
+				break;
+		}
+	}
+
+	netdev_tx_completed_queue(netdev_get_tx_queue(bp->dev, queue_index),
+				  packets, bytes);
+
+	queue->tx_tail = tail;
+	if (__netif_subqueue_stopped(bp->dev, queue_index) &&
+	    CIRC_CNT(queue->tx_head, queue->tx_tail,
+		     bp->tx_ring_size) <= MACB_TX_WAKEUP_THRESH(bp))
+		netif_wake_subqueue(bp->dev, queue_index);
+	spin_unlock_irqrestore(&queue->tx_ptr_lock, flags);
+
+	return packets;
+}
+
+static void gem_rx_refill(struct macb_queue *queue)
+{
+	unsigned int		entry;
+	struct sk_buff		*skb;
+	dma_addr_t		paddr;
+	struct macb *bp = queue->bp;
+	struct macb_dma_desc *desc;
+
+	while (CIRC_SPACE(queue->rx_prepared_head, queue->rx_tail,
+			bp->rx_ring_size) > 0) {
+		entry = macb_rx_ring_wrap(bp, queue->rx_prepared_head);
+
+		/* Make hw descriptor updates visible to CPU */
+		rmb();
+
+		desc = macb_rx_desc(queue, entry);
+
+		if (!queue->rx_skbuff[entry]) {
+			/* allocate sk_buff for this free entry in ring */
+			skb = netdev_alloc_skb(bp->dev, bp->rx_buffer_size);
+			if (unlikely(!skb)) {
+				netdev_err(bp->dev,
+					   "Unable to allocate sk_buff\n");
+				break;
+			}
+
+			/* now fill corresponding descriptor entry */
+			paddr = dma_map_single(&bp->pdev->dev, skb->data,
+					       bp->rx_buffer_size,
+					       DMA_FROM_DEVICE);
+			if (dma_mapping_error(&bp->pdev->dev, paddr)) {
+				dev_kfree_skb(skb);
+				break;
+			}
+
+			queue->rx_skbuff[entry] = skb;
+
+			if (entry == bp->rx_ring_size - 1)
+				paddr |= MACB_BIT(RX_WRAP);
+			desc->ctrl = 0;
+			/* Setting addr clears RX_USED and allows reception,
+			 * make sure ctrl is cleared first to avoid a race.
+			 */
+			dma_wmb();
+			macb_set_addr(bp, desc, paddr);
+
+			/* properly align Ethernet header */
+			skb_reserve(skb, NET_IP_ALIGN);
+		} else {
+			desc->ctrl = 0;
+			dma_wmb();
+			desc->addr &= ~MACB_BIT(RX_USED);
+		}
+		queue->rx_prepared_head++;
+	}
+
+	/* Make descriptor updates visible to hardware */
+	wmb();
+
+	netdev_vdbg(bp->dev, "rx ring: queue: %p, prepared head %d, tail %d\n",
+			queue, queue->rx_prepared_head, queue->rx_tail);
+}
+
+/* Mark DMA descriptors from begin up to and not including end as unused */
+static void discard_partial_frame(struct macb_queue *queue, unsigned int begin,
+				  unsigned int end)
+{
+	unsigned int frag;
+
+	for (frag = begin; frag != end; frag++) {
+		struct macb_dma_desc *desc = macb_rx_desc(queue, frag);
+
+		desc->addr &= ~MACB_BIT(RX_USED);
+	}
+
+	/* Make descriptor updates visible to hardware */
+	wmb();
+
+	/* When this happens, the hardware stats registers for
+	 * whatever caused this is updated, so we don't have to record
+	 * anything.
+	 */
+}
+
+static int gem_rx(struct macb_queue *queue, struct napi_struct *napi,
+		  int budget)
+{
+	struct macb *bp = queue->bp;
+	unsigned int		len;
+	unsigned int		entry;
+	struct sk_buff		*skb;
+	struct macb_dma_desc	*desc;
+	int			count = 0;
+
+	while (count < budget) {
+		u32 ctrl;
+		dma_addr_t addr;
+		bool rxused;
+
+		entry = macb_rx_ring_wrap(bp, queue->rx_tail);
+		desc = macb_rx_desc(queue, entry);
+
+		/* Make hw descriptor updates visible to CPU */
+		rmb();
+
+		rxused = (desc->addr & MACB_BIT(RX_USED)) ? true : false;
+		addr = macb_get_addr(bp, desc);
+
+		if (!rxused)
+			break;
+
+		/* Ensure ctrl is at least as up-to-date as rxused */
+		dma_rmb();
+
+		ctrl = desc->ctrl;
+
+		queue->rx_tail++;
+		count++;
+
+		if (!(ctrl & MACB_BIT(RX_SOF) && ctrl & MACB_BIT(RX_EOF))) {
+			netdev_err(bp->dev,
+				   "not whole frame pointed by descriptor\n");
+			bp->dev->stats.rx_dropped++;
+			queue->stats.rx_dropped++;
+			break;
+		}
+		skb = queue->rx_skbuff[entry];
+		if (unlikely(!skb)) {
+			netdev_err(bp->dev,
+				   "inconsistent Rx descriptor chain\n");
+			bp->dev->stats.rx_dropped++;
+			queue->stats.rx_dropped++;
+			break;
+		}
+		/* now everything is ready for receiving packet */
+		queue->rx_skbuff[entry] = NULL;
+		len = ctrl & bp->rx_frm_len_mask;
+
+		netdev_vdbg(bp->dev, "gem_rx %u (len %u)\n", entry, len);
+
+		skb_put(skb, len);
+		dma_unmap_single(&bp->pdev->dev, addr,
+				 bp->rx_buffer_size, DMA_FROM_DEVICE);
+
+		skb->protocol = eth_type_trans(skb, bp->dev);
+		skb_checksum_none_assert(skb);
+		if (bp->dev->features & NETIF_F_RXCSUM &&
+		    !(bp->dev->flags & IFF_PROMISC) &&
+		    GEM_BFEXT(RX_CSUM, ctrl) & GEM_RX_CSUM_CHECKED_MASK)
+			skb->ip_summed = CHECKSUM_UNNECESSARY;
+
+		bp->dev->stats.rx_packets++;
+		queue->stats.rx_packets++;
+		bp->dev->stats.rx_bytes += skb->len;
+		queue->stats.rx_bytes += skb->len;
+
+		gem_ptp_do_rxstamp(bp, skb, desc);
+
+#if defined(DEBUG) && defined(VERBOSE_DEBUG)
+		netdev_vdbg(bp->dev, "received skb of length %u, csum: %08x\n",
+			    skb->len, skb->csum);
+		print_hex_dump(KERN_DEBUG, " mac: ", DUMP_PREFIX_ADDRESS, 16, 1,
+			       skb_mac_header(skb), 16, true);
+		print_hex_dump(KERN_DEBUG, "data: ", DUMP_PREFIX_ADDRESS, 16, 1,
+			       skb->data, 32, true);
+#endif
+
+		napi_gro_receive(napi, skb);
+	}
+
+	gem_rx_refill(queue);
+
+	return count;
+}
+
+static int macb_rx_frame(struct macb_queue *queue, struct napi_struct *napi,
+			 unsigned int first_frag, unsigned int last_frag)
+{
+	unsigned int len;
+	unsigned int frag;
+	unsigned int offset;
+	struct sk_buff *skb;
+	struct macb_dma_desc *desc;
+	struct macb *bp = queue->bp;
+
+	desc = macb_rx_desc(queue, last_frag);
+	len = desc->ctrl & bp->rx_frm_len_mask;
+
+	netdev_vdbg(bp->dev, "macb_rx_frame frags %u - %u (len %u)\n",
+		macb_rx_ring_wrap(bp, first_frag),
+		macb_rx_ring_wrap(bp, last_frag), len);
+
+	/* The ethernet header starts NET_IP_ALIGN bytes into the
+	 * first buffer. Since the header is 14 bytes, this makes the
+	 * payload word-aligned.
+	 *
+	 * Instead of calling skb_reserve(NET_IP_ALIGN), we just copy
+	 * the two padding bytes into the skb so that we avoid hitting
+	 * the slowpath in memcpy(), and pull them off afterwards.
+	 */
+	skb = netdev_alloc_skb(bp->dev, len + NET_IP_ALIGN);
+	if (!skb) {
+		bp->dev->stats.rx_dropped++;
+		for (frag = first_frag; ; frag++) {
+			desc = macb_rx_desc(queue, frag);
+			desc->addr &= ~MACB_BIT(RX_USED);
+			if (frag == last_frag)
+				break;
+		}
+
+		/* Make descriptor updates visible to hardware */
+		wmb();
+
+		return 1;
+	}
+
+	offset = 0;
+	len += NET_IP_ALIGN;
+	skb_checksum_none_assert(skb);
+	skb_put(skb, len);
+
+	for (frag = first_frag; ; frag++) {
+		unsigned int frag_len = bp->rx_buffer_size;
+
+		if (offset + frag_len > len) {
+			if (unlikely(frag != last_frag)) {
+				dev_kfree_skb_any(skb);
+				return -1;
+			}
+			frag_len = len - offset;
+		}
+		skb_copy_to_linear_data_offset(skb, offset,
+					       macb_rx_buffer(queue, frag),
+					       frag_len);
+		offset += bp->rx_buffer_size;
+		desc = macb_rx_desc(queue, frag);
+		desc->addr &= ~MACB_BIT(RX_USED);
+
+		if (frag == last_frag)
+			break;
+	}
+
+	/* Make descriptor updates visible to hardware */
+	wmb();
+
+	__skb_pull(skb, NET_IP_ALIGN);
+	skb->protocol = eth_type_trans(skb, bp->dev);
+
+	bp->dev->stats.rx_packets++;
+	bp->dev->stats.rx_bytes += skb->len;
+	netdev_vdbg(bp->dev, "received skb of length %u, csum: %08x\n",
+		    skb->len, skb->csum);
+	napi_gro_receive(napi, skb);
+
+	return 0;
+}
+
+static inline void macb_init_rx_ring(struct macb_queue *queue)
+{
+	struct macb *bp = queue->bp;
+	dma_addr_t addr;
+	struct macb_dma_desc *desc = NULL;
+	int i;
+
+	addr = queue->rx_buffers_dma;
+	for (i = 0; i < bp->rx_ring_size; i++) {
+		desc = macb_rx_desc(queue, i);
+		macb_set_addr(bp, desc, addr);
+		desc->ctrl = 0;
+		addr += bp->rx_buffer_size;
+	}
+	desc->addr |= MACB_BIT(RX_WRAP);
+	queue->rx_tail = 0;
+}
+
+static int macb_rx(struct macb_queue *queue, struct napi_struct *napi,
+		   int budget)
+{
+	struct macb *bp = queue->bp;
+	bool reset_rx_queue = false;
+	int received = 0;
+	unsigned int tail;
+	int first_frag = -1;
+
+	for (tail = queue->rx_tail; budget > 0; tail++) {
+		struct macb_dma_desc *desc = macb_rx_desc(queue, tail);
+		u32 ctrl;
+
+		/* Make hw descriptor updates visible to CPU */
+		rmb();
+
+		if (!(desc->addr & MACB_BIT(RX_USED)))
+			break;
+
+		/* Ensure ctrl is at least as up-to-date as addr */
+		dma_rmb();
+
+		ctrl = desc->ctrl;
+
+		if (ctrl & MACB_BIT(RX_SOF)) {
+			if (first_frag != -1)
+				discard_partial_frame(queue, first_frag, tail);
+			first_frag = tail;
+		}
+
+		if (ctrl & MACB_BIT(RX_EOF)) {
+			int dropped;
+
+			if (unlikely(first_frag == -1)) {
+				reset_rx_queue = true;
+				continue;
+			}
+
+			dropped = macb_rx_frame(queue, napi, first_frag, tail);
+			first_frag = -1;
+			if (unlikely(dropped < 0)) {
+				reset_rx_queue = true;
+				continue;
+			}
+			if (!dropped) {
+				received++;
+				budget--;
+			}
+		}
+	}
+
+	if (unlikely(reset_rx_queue)) {
+		unsigned long flags;
+		u32 ctrl;
+
+		netdev_err(bp->dev, "RX queue corruption: reset it\n");
+
+		spin_lock_irqsave(&bp->lock, flags);
+
+		ctrl = macb_readl(bp, NCR);
+		macb_writel(bp, NCR, ctrl & ~MACB_BIT(RE));
+
+		macb_init_rx_ring(queue);
+		queue_writel(queue, RBQP, queue->rx_ring_dma);
+
+		macb_writel(bp, NCR, ctrl | MACB_BIT(RE));
+
+		spin_unlock_irqrestore(&bp->lock, flags);
+		return received;
+	}
+
+	if (first_frag != -1)
+		queue->rx_tail = first_frag;
+	else
+		queue->rx_tail = tail;
+
+	return received;
+}
+
+static bool macb_rx_pending(struct macb_queue *queue)
+{
+	struct macb *bp = queue->bp;
+	unsigned int		entry;
+	struct macb_dma_desc	*desc;
+
+	entry = macb_rx_ring_wrap(bp, queue->rx_tail);
+	desc = macb_rx_desc(queue, entry);
+
+	/* Make hw descriptor updates visible to CPU */
+	rmb();
+
+	return (desc->addr & MACB_BIT(RX_USED)) != 0;
+}
+
+static int macb_rx_poll(struct napi_struct *napi, int budget)
+{
+	struct macb_queue *queue = container_of(napi, struct macb_queue, napi_rx);
+	struct macb *bp = queue->bp;
+	int work_done;
+
+	work_done = bp->macbgem_ops.mog_rx(queue, napi, budget);
+
+	netdev_vdbg(bp->dev, "RX poll: queue = %u, work_done = %d, budget = %d\n",
+		    (unsigned int)(queue - bp->queues), work_done, budget);
+
+	if (work_done < budget && napi_complete_done(napi, work_done)) {
+		queue_writel(queue, IER, bp->rx_intr_mask);
+
+		/* Packet completions only seem to propagate to raise
+		 * interrupts when interrupts are enabled at the time, so if
+		 * packets were received while interrupts were disabled,
+		 * they will not cause another interrupt to be generated when
+		 * interrupts are re-enabled.
+		 * Check for this case here to avoid losing a wakeup. This can
+		 * potentially race with the interrupt handler doing the same
+		 * actions if an interrupt is raised just after enabling them,
+		 * but this should be harmless.
+		 */
+		if (macb_rx_pending(queue)) {
+			queue_writel(queue, IDR, bp->rx_intr_mask);
+			if (bp->caps & MACB_CAPS_ISR_CLEAR_ON_WRITE)
+				queue_writel(queue, ISR, MACB_BIT(RCOMP));
+			netdev_vdbg(bp->dev, "poll: packets pending, reschedule\n");
+			napi_schedule(napi);
+		}
+	}
+
+	/* TODO: Handle errors */
+
+	return work_done;
+}
+
+static void macb_tx_restart(struct macb_queue *queue)
+{
+	struct macb *bp = queue->bp;
+	unsigned int head_idx, tbqp;
+	unsigned long flags;
+
+	spin_lock_irqsave(&queue->tx_ptr_lock, flags);
+
+	if (queue->tx_head == queue->tx_tail)
+		goto out_tx_ptr_unlock;
+
+	tbqp = queue_readl(queue, TBQP) / macb_dma_desc_get_size(bp);
+	tbqp = macb_adj_dma_desc_idx(bp, macb_tx_ring_wrap(bp, tbqp));
+	head_idx = macb_adj_dma_desc_idx(bp, macb_tx_ring_wrap(bp, queue->tx_head));
+
+	if (tbqp == head_idx)
+		goto out_tx_ptr_unlock;
+
+	spin_lock(&bp->lock);
+	macb_writel(bp, NCR, macb_readl(bp, NCR) | MACB_BIT(TSTART));
+	spin_unlock(&bp->lock);
+
+out_tx_ptr_unlock:
+	spin_unlock_irqrestore(&queue->tx_ptr_lock, flags);
+}
+
+static bool macb_tx_complete_pending(struct macb_queue *queue)
+{
+	bool retval = false;
+	unsigned long flags;
+
+	spin_lock_irqsave(&queue->tx_ptr_lock, flags);
+	if (queue->tx_head != queue->tx_tail) {
+		/* Make hw descriptor updates visible to CPU */
+		rmb();
+
+		if (macb_tx_desc(queue, queue->tx_tail)->ctrl & MACB_BIT(TX_USED))
+			retval = true;
+	}
+	spin_unlock_irqrestore(&queue->tx_ptr_lock, flags);
+	return retval;
+}
+
+static int macb_tx_poll(struct napi_struct *napi, int budget)
+{
+	struct macb_queue *queue = container_of(napi, struct macb_queue, napi_tx);
+	struct macb *bp = queue->bp;
+	int work_done;
+
+	work_done = macb_tx_complete(queue, budget);
+
+	rmb(); // ensure txubr_pending is up to date
+	if (queue->txubr_pending) {
+		queue->txubr_pending = false;
+		netdev_vdbg(bp->dev, "poll: tx restart\n");
+		macb_tx_restart(queue);
+	}
+
+	netdev_vdbg(bp->dev, "TX poll: queue = %u, work_done = %d, budget = %d\n",
+		    (unsigned int)(queue - bp->queues), work_done, budget);
+
+	if (work_done < budget && napi_complete_done(napi, work_done)) {
+		queue_writel(queue, IER, MACB_BIT(TCOMP));
+
+		/* Packet completions only seem to propagate to raise
+		 * interrupts when interrupts are enabled at the time, so if
+		 * packets were sent while interrupts were disabled,
+		 * they will not cause another interrupt to be generated when
+		 * interrupts are re-enabled.
+		 * Check for this case here to avoid losing a wakeup. This can
+		 * potentially race with the interrupt handler doing the same
+		 * actions if an interrupt is raised just after enabling them,
+		 * but this should be harmless.
+		 */
+		if (macb_tx_complete_pending(queue)) {
+			queue_writel(queue, IDR, MACB_BIT(TCOMP));
+			if (bp->caps & MACB_CAPS_ISR_CLEAR_ON_WRITE)
+				queue_writel(queue, ISR, MACB_BIT(TCOMP));
+			netdev_vdbg(bp->dev, "TX poll: packets pending, reschedule\n");
+			napi_schedule(napi);
+		}
+	}
+
+	return work_done;
+}
+
+static void macb_hresp_error_task(struct work_struct *work)
+{
+	struct macb *bp = from_work(bp, work, hresp_err_bh_work);
+	struct net_device *dev = bp->dev;
+	struct macb_queue *queue;
+	unsigned int q;
+	u32 ctrl;
+
+	for (q = 0, queue = bp->queues; q < bp->num_queues; ++q, ++queue) {
+		queue_writel(queue, IDR, bp->rx_intr_mask |
+					 MACB_TX_INT_FLAGS |
+					 MACB_BIT(HRESP));
+	}
+	ctrl = macb_readl(bp, NCR);
+	ctrl &= ~(MACB_BIT(RE) | MACB_BIT(TE));
+	macb_writel(bp, NCR, ctrl);
+
+	netif_tx_stop_all_queues(dev);
+	netif_carrier_off(dev);
+
+	bp->macbgem_ops.mog_init_rings(bp);
+
+	/* Initialize TX and RX buffers */
+	macb_init_buffers(bp);
+
+	/* Enable interrupts */
+	for (q = 0, queue = bp->queues; q < bp->num_queues; ++q, ++queue)
+		queue_writel(queue, IER,
+			     bp->rx_intr_mask |
+			     MACB_TX_INT_FLAGS |
+			     MACB_BIT(HRESP));
+
+	ctrl |= MACB_BIT(RE) | MACB_BIT(TE);
+	macb_writel(bp, NCR, ctrl);
+
+	netif_carrier_on(dev);
+	netif_tx_start_all_queues(dev);
+}
+
+static irqreturn_t macb_wol_interrupt(int irq, void *dev_id)
+{
+	struct macb_queue *queue = dev_id;
+	struct macb *bp = queue->bp;
+	u32 status;
+
+	status = queue_readl(queue, ISR);
+
+	if (unlikely(!status))
+		return IRQ_NONE;
+
+	spin_lock(&bp->lock);
+
+	if (status & MACB_BIT(WOL)) {
+		queue_writel(queue, IDR, MACB_BIT(WOL));
+		macb_writel(bp, WOL, 0);
+		netdev_vdbg(bp->dev, "MACB WoL: queue = %u, isr = 0x%08lx\n",
+			    (unsigned int)(queue - bp->queues),
+			    (unsigned long)status);
+		if (bp->caps & MACB_CAPS_ISR_CLEAR_ON_WRITE)
+			queue_writel(queue, ISR, MACB_BIT(WOL));
+		pm_wakeup_event(&bp->pdev->dev, 0);
+	}
+
+	spin_unlock(&bp->lock);
+
+	return IRQ_HANDLED;
+}
+
+static irqreturn_t gem_wol_interrupt(int irq, void *dev_id)
+{
+	struct macb_queue *queue = dev_id;
+	struct macb *bp = queue->bp;
+	u32 status;
+
+	status = queue_readl(queue, ISR);
+
+	if (unlikely(!status))
+		return IRQ_NONE;
+
+	spin_lock(&bp->lock);
+
+	if (status & GEM_BIT(WOL)) {
+		queue_writel(queue, IDR, GEM_BIT(WOL));
+		gem_writel(bp, WOL, 0);
+		netdev_vdbg(bp->dev, "GEM WoL: queue = %u, isr = 0x%08lx\n",
+			    (unsigned int)(queue - bp->queues),
+			    (unsigned long)status);
+		if (bp->caps & MACB_CAPS_ISR_CLEAR_ON_WRITE)
+			queue_writel(queue, ISR, GEM_BIT(WOL));
+		pm_wakeup_event(&bp->pdev->dev, 0);
+	}
+
+	spin_unlock(&bp->lock);
+
+	return IRQ_HANDLED;
+}
+
+static irqreturn_t macb_interrupt(int irq, void *dev_id)
+{
+	struct macb_queue *queue = dev_id;
+	struct macb *bp = queue->bp;
+	struct net_device *dev = bp->dev;
+	u32 status, ctrl;
+
+	status = queue_readl(queue, ISR);
+
+	if (unlikely(!status))
+		return IRQ_NONE;
+
+	spin_lock(&bp->lock);
+
+	while (status) {
+		/* close possible race with dev_close */
+		if (unlikely(!netif_running(dev))) {
+			queue_writel(queue, IDR, -1);
+			if (bp->caps & MACB_CAPS_ISR_CLEAR_ON_WRITE)
+				queue_writel(queue, ISR, -1);
+			break;
+		}
+
+		netdev_vdbg(bp->dev, "queue = %u, isr = 0x%08lx\n",
+			    (unsigned int)(queue - bp->queues),
+			    (unsigned long)status);
+
+		if (status & bp->rx_intr_mask) {
+			/* There's no point taking any more interrupts
+			 * until we have processed the buffers. The
+			 * scheduling call may fail if the poll routine
+			 * is already scheduled, so disable interrupts
+			 * now.
+			 */
+			queue_writel(queue, IDR, bp->rx_intr_mask);
+			if (bp->caps & MACB_CAPS_ISR_CLEAR_ON_WRITE)
+				queue_writel(queue, ISR, MACB_BIT(RCOMP));
+
+			if (napi_schedule_prep(&queue->napi_rx)) {
+				netdev_vdbg(bp->dev, "scheduling RX softirq\n");
+				__napi_schedule(&queue->napi_rx);
+			}
+		}
+
+		if (status & (MACB_BIT(TCOMP) |
+			      MACB_BIT(TXUBR))) {
+			queue_writel(queue, IDR, MACB_BIT(TCOMP));
+			if (bp->caps & MACB_CAPS_ISR_CLEAR_ON_WRITE)
+				queue_writel(queue, ISR, MACB_BIT(TCOMP) |
+							 MACB_BIT(TXUBR));
+
+			if (status & MACB_BIT(TXUBR)) {
+				queue->txubr_pending = true;
+				wmb(); // ensure softirq can see update
+			}
+
+			if (napi_schedule_prep(&queue->napi_tx)) {
+				netdev_vdbg(bp->dev, "scheduling TX softirq\n");
+				__napi_schedule(&queue->napi_tx);
+			}
+		}
+
+		if (unlikely(status & (MACB_TX_ERR_FLAGS))) {
+			queue_writel(queue, IDR, MACB_TX_INT_FLAGS);
+			schedule_work(&queue->tx_error_task);
+
+			if (bp->caps & MACB_CAPS_ISR_CLEAR_ON_WRITE)
+				queue_writel(queue, ISR, MACB_TX_ERR_FLAGS);
+
+			break;
+		}
+
+		/* Link change detection isn't possible with RMII, so we'll
+		 * add that if/when we get our hands on a full-blown MII PHY.
+		 */
+
+		/* There is a hardware issue under heavy load where DMA can
+		 * stop, this causes endless "used buffer descriptor read"
+		 * interrupts but it can be cleared by re-enabling RX. See
+		 * the at91rm9200 manual, section 41.3.1 or the Zynq manual
+		 * section 16.7.4 for details. RXUBR is only enabled for
+		 * these two versions.
+		 */
+		if (status & MACB_BIT(RXUBR)) {
+			ctrl = macb_readl(bp, NCR);
+			macb_writel(bp, NCR, ctrl & ~MACB_BIT(RE));
+			wmb();
+			macb_writel(bp, NCR, ctrl | MACB_BIT(RE));
+
+			if (bp->caps & MACB_CAPS_ISR_CLEAR_ON_WRITE)
+				queue_writel(queue, ISR, MACB_BIT(RXUBR));
+		}
+
+		if (status & MACB_BIT(ISR_ROVR)) {
+			/* We missed at least one packet */
+			spin_lock(&bp->stats_lock);
+			if (macb_is_gem(bp))
+				bp->hw_stats.gem.rx_overruns++;
+			else
+				bp->hw_stats.macb.rx_overruns++;
+			spin_unlock(&bp->stats_lock);
+
+			if (bp->caps & MACB_CAPS_ISR_CLEAR_ON_WRITE)
+				queue_writel(queue, ISR, MACB_BIT(ISR_ROVR));
+		}
+
+		if (status & MACB_BIT(HRESP)) {
+			queue_work(system_bh_wq, &bp->hresp_err_bh_work);
+			netdev_err(dev, "DMA bus error: HRESP not OK\n");
+
+			if (bp->caps & MACB_CAPS_ISR_CLEAR_ON_WRITE)
+				queue_writel(queue, ISR, MACB_BIT(HRESP));
+		}
+		status = queue_readl(queue, ISR);
+	}
+
+	spin_unlock(&bp->lock);
+
+	return IRQ_HANDLED;
+}
+
+#ifdef CONFIG_NET_POLL_CONTROLLER
+/* Polling receive - used by netconsole and other diagnostic tools
+ * to allow network i/o with interrupts disabled.
+ */
+static void macb_poll_controller(struct net_device *dev)
+{
+	struct macb *bp = netdev_priv(dev);
+	struct macb_queue *queue;
+	unsigned long flags;
+	unsigned int q;
+
+	local_irq_save(flags);
+	for (q = 0, queue = bp->queues; q < bp->num_queues; ++q, ++queue)
+		macb_interrupt(dev->irq, queue);
+	local_irq_restore(flags);
+}
+#endif
+
+static unsigned int macb_tx_map(struct macb *bp,
+				struct macb_queue *queue,
+				struct sk_buff *skb,
+				unsigned int hdrlen)
+{
+	dma_addr_t mapping;
+	unsigned int len, entry, i, tx_head = queue->tx_head;
+	struct macb_tx_skb *tx_skb = NULL;
+	struct macb_dma_desc *desc;
+	unsigned int offset, size, count = 0;
+	unsigned int f, nr_frags = skb_shinfo(skb)->nr_frags;
+	unsigned int eof = 1, mss_mfs = 0;
+	u32 ctrl, lso_ctrl = 0, seq_ctrl = 0;
+
+	/* LSO */
+	if (skb_shinfo(skb)->gso_size != 0) {
+		if (ip_hdr(skb)->protocol == IPPROTO_UDP)
+			/* UDP - UFO */
+			lso_ctrl = MACB_LSO_UFO_ENABLE;
+		else
+			/* TCP - TSO */
+			lso_ctrl = MACB_LSO_TSO_ENABLE;
+	}
+
+	/* First, map non-paged data */
+	len = skb_headlen(skb);
+
+	/* first buffer length */
+	size = hdrlen;
+
+	offset = 0;
+	while (len) {
+		entry = macb_tx_ring_wrap(bp, tx_head);
+		tx_skb = &queue->tx_skb[entry];
+
+		mapping = dma_map_single(&bp->pdev->dev,
+					 skb->data + offset,
+					 size, DMA_TO_DEVICE);
+		if (dma_mapping_error(&bp->pdev->dev, mapping))
+			goto dma_error;
+
+		/* Save info to properly release resources */
+		tx_skb->skb = NULL;
+		tx_skb->mapping = mapping;
+		tx_skb->size = size;
+		tx_skb->mapped_as_page = false;
+
+		len -= size;
+		offset += size;
+		count++;
+		tx_head++;
+
+		size = min(len, bp->max_tx_length);
+	}
+
+	/* Then, map paged data from fragments */
+	for (f = 0; f < nr_frags; f++) {
+		const skb_frag_t *frag = &skb_shinfo(skb)->frags[f];
+
+		len = skb_frag_size(frag);
+		offset = 0;
+		while (len) {
+			size = min(len, bp->max_tx_length);
+			entry = macb_tx_ring_wrap(bp, tx_head);
+			tx_skb = &queue->tx_skb[entry];
+
+			mapping = skb_frag_dma_map(&bp->pdev->dev, frag,
+						   offset, size, DMA_TO_DEVICE);
+			if (dma_mapping_error(&bp->pdev->dev, mapping))
+				goto dma_error;
+
+			/* Save info to properly release resources */
+			tx_skb->skb = NULL;
+			tx_skb->mapping = mapping;
+			tx_skb->size = size;
+			tx_skb->mapped_as_page = true;
+
+			len -= size;
+			offset += size;
+			count++;
+			tx_head++;
+		}
+	}
+
+	/* Should never happen */
+	if (unlikely(!tx_skb)) {
+		netdev_err(bp->dev, "BUG! empty skb!\n");
+		return 0;
+	}
+
+	/* This is the last buffer of the frame: save socket buffer */
+	tx_skb->skb = skb;
+
+	/* Update TX ring: update buffer descriptors in reverse order
+	 * to avoid race condition
+	 */
+
+	/* Set 'TX_USED' bit in buffer descriptor at tx_head position
+	 * to set the end of TX queue
+	 */
+	i = tx_head;
+	entry = macb_tx_ring_wrap(bp, i);
+	ctrl = MACB_BIT(TX_USED);
+	desc = macb_tx_desc(queue, entry);
+	desc->ctrl = ctrl;
+
+	if (lso_ctrl) {
+		if (lso_ctrl == MACB_LSO_UFO_ENABLE)
+			/* include header and FCS in value given to h/w */
+			mss_mfs = skb_shinfo(skb)->gso_size +
+					skb_transport_offset(skb) +
+					ETH_FCS_LEN;
+		else /* TSO */ {
+			mss_mfs = skb_shinfo(skb)->gso_size;
+			/* TCP Sequence Number Source Select
+			 * can be set only for TSO
+			 */
+			seq_ctrl = 0;
+		}
+	}
+
+	do {
+		i--;
+		entry = macb_tx_ring_wrap(bp, i);
+		tx_skb = &queue->tx_skb[entry];
+		desc = macb_tx_desc(queue, entry);
+
+		ctrl = (u32)tx_skb->size;
+		if (eof) {
+			ctrl |= MACB_BIT(TX_LAST);
+			eof = 0;
+		}
+		if (unlikely(entry == (bp->tx_ring_size - 1)))
+			ctrl |= MACB_BIT(TX_WRAP);
+
+		/* First descriptor is header descriptor */
+		if (i == queue->tx_head) {
+			ctrl |= MACB_BF(TX_LSO, lso_ctrl);
+			ctrl |= MACB_BF(TX_TCP_SEQ_SRC, seq_ctrl);
+			if ((bp->dev->features & NETIF_F_HW_CSUM) &&
+			    skb->ip_summed != CHECKSUM_PARTIAL && !lso_ctrl &&
+			    !ptp_one_step_sync(skb))
+				ctrl |= MACB_BIT(TX_NOCRC);
+		} else
+			/* Only set MSS/MFS on payload descriptors
+			 * (second or later descriptor)
+			 */
+			ctrl |= MACB_BF(MSS_MFS, mss_mfs);
+
+		/* Set TX buffer descriptor */
+		macb_set_addr(bp, desc, tx_skb->mapping);
+		/* desc->addr must be visible to hardware before clearing
+		 * 'TX_USED' bit in desc->ctrl.
+		 */
+		wmb();
+		desc->ctrl = ctrl;
+	} while (i != queue->tx_head);
+
+	queue->tx_head = tx_head;
+
+	return count;
+
+dma_error:
+	netdev_err(bp->dev, "TX DMA map failed\n");
+
+	for (i = queue->tx_head; i != tx_head; i++) {
+		tx_skb = macb_tx_skb(queue, i);
+
+		macb_tx_unmap(bp, tx_skb, 0);
+	}
+
+	return 0;
+}
+
+static netdev_features_t macb_features_check(struct sk_buff *skb,
+					     struct net_device *dev,
+					     netdev_features_t features)
+{
+	unsigned int nr_frags, f;
+	unsigned int hdrlen;
+
+	/* Validate LSO compatibility */
+
+	/* there is only one buffer or protocol is not UDP */
+	if (!skb_is_nonlinear(skb) || (ip_hdr(skb)->protocol != IPPROTO_UDP))
+		return features;
+
+	/* length of header */
+	hdrlen = skb_transport_offset(skb);
+
+	/* For UFO only:
+	 * When software supplies two or more payload buffers all payload buffers
+	 * apart from the last must be a multiple of 8 bytes in size.
+	 */
+	if (!IS_ALIGNED(skb_headlen(skb) - hdrlen, MACB_TX_LEN_ALIGN))
+		return features & ~MACB_NETIF_LSO;
+
+	nr_frags = skb_shinfo(skb)->nr_frags;
+	/* No need to check last fragment */
+	nr_frags--;
+	for (f = 0; f < nr_frags; f++) {
+		const skb_frag_t *frag = &skb_shinfo(skb)->frags[f];
+
+		if (!IS_ALIGNED(skb_frag_size(frag), MACB_TX_LEN_ALIGN))
+			return features & ~MACB_NETIF_LSO;
+	}
+	return features;
+}
+
+static inline int macb_clear_csum(struct sk_buff *skb)
+{
+	/* no change for packets without checksum offloading */
+	if (skb->ip_summed != CHECKSUM_PARTIAL)
+		return 0;
+
+	/* make sure we can modify the header */
+	if (unlikely(skb_cow_head(skb, 0)))
+		return -1;
+
+	/* initialize checksum field
+	 * This is required - at least for Zynq, which otherwise calculates
+	 * wrong UDP header checksums for UDP packets with UDP data len <=2
+	 */
+	*(__sum16 *)(skb_checksum_start(skb) + skb->csum_offset) = 0;
+	return 0;
+}
+
+static int macb_pad_and_fcs(struct sk_buff **skb, struct net_device *ndev)
+{
+	bool cloned = skb_cloned(*skb) || skb_header_cloned(*skb) ||
+		      skb_is_nonlinear(*skb);
+	int padlen = ETH_ZLEN - (*skb)->len;
+	int tailroom = skb_tailroom(*skb);
+	struct sk_buff *nskb;
+	u32 fcs;
+
+	if (!(ndev->features & NETIF_F_HW_CSUM) ||
+	    !((*skb)->ip_summed != CHECKSUM_PARTIAL) ||
+	    skb_shinfo(*skb)->gso_size || ptp_one_step_sync(*skb))
+		return 0;
+
+	if (padlen <= 0) {
+		/* FCS could be appeded to tailroom. */
+		if (tailroom >= ETH_FCS_LEN)
+			goto add_fcs;
+		/* No room for FCS, need to reallocate skb. */
+		else
+			padlen = ETH_FCS_LEN;
+	} else {
+		/* Add room for FCS. */
+		padlen += ETH_FCS_LEN;
+	}
+
+	if (cloned || tailroom < padlen) {
+		nskb = skb_copy_expand(*skb, 0, padlen, GFP_ATOMIC);
+		if (!nskb)
+			return -ENOMEM;
+
+		dev_consume_skb_any(*skb);
+		*skb = nskb;
+	}
+
+	if (padlen > ETH_FCS_LEN)
+		skb_put_zero(*skb, padlen - ETH_FCS_LEN);
+
+add_fcs:
+	/* set FCS to packet */
+	fcs = crc32_le(~0, (*skb)->data, (*skb)->len);
+	fcs = ~fcs;
+
+	skb_put_u8(*skb, fcs		& 0xff);
+	skb_put_u8(*skb, (fcs >> 8)	& 0xff);
+	skb_put_u8(*skb, (fcs >> 16)	& 0xff);
+	skb_put_u8(*skb, (fcs >> 24)	& 0xff);
+
+	return 0;
+}
+
+static netdev_tx_t macb_start_xmit(struct sk_buff *skb, struct net_device *dev)
+{
+	u16 queue_index = skb_get_queue_mapping(skb);
+	struct macb *bp = netdev_priv(dev);
+	struct macb_queue *queue = &bp->queues[queue_index];
+	unsigned int desc_cnt, nr_frags, frag_size, f;
+	unsigned int hdrlen;
+	unsigned long flags;
+	bool is_lso;
+	netdev_tx_t ret = NETDEV_TX_OK;
+
+	if (macb_clear_csum(skb)) {
+		dev_kfree_skb_any(skb);
+		return ret;
+	}
+
+	if (macb_pad_and_fcs(&skb, dev)) {
+		dev_kfree_skb_any(skb);
+		return ret;
+	}
+
+#ifdef CONFIG_MACB_USE_HWSTAMP
+	if ((skb_shinfo(skb)->tx_flags & SKBTX_HW_TSTAMP) &&
+	    (bp->hw_dma_cap & HW_DMA_CAP_PTP))
+		skb_shinfo(skb)->tx_flags |= SKBTX_IN_PROGRESS;
+#endif
+
+	is_lso = (skb_shinfo(skb)->gso_size != 0);
+
+	if (is_lso) {
+		/* length of headers */
+		if (ip_hdr(skb)->protocol == IPPROTO_UDP)
+			/* only queue eth + ip headers separately for UDP */
+			hdrlen = skb_transport_offset(skb);
+		else
+			hdrlen = skb_tcp_all_headers(skb);
+		if (skb_headlen(skb) < hdrlen) {
+			netdev_err(bp->dev, "Error - LSO headers fragmented!!!\n");
+			/* if this is required, would need to copy to single buffer */
+			return NETDEV_TX_BUSY;
+		}
+	} else
+		hdrlen = min(skb_headlen(skb), bp->max_tx_length);
+
+#if defined(DEBUG) && defined(VERBOSE_DEBUG)
+	netdev_vdbg(bp->dev,
+		    "start_xmit: queue %hu len %u head %p data %p tail %p end %p\n",
+		    queue_index, skb->len, skb->head, skb->data,
+		    skb_tail_pointer(skb), skb_end_pointer(skb));
+	print_hex_dump(KERN_DEBUG, "data: ", DUMP_PREFIX_OFFSET, 16, 1,
+		       skb->data, 16, true);
+#endif
+
+	/* Count how many TX buffer descriptors are needed to send this
+	 * socket buffer: skb fragments of jumbo frames may need to be
+	 * split into many buffer descriptors.
+	 */
+	if (is_lso && (skb_headlen(skb) > hdrlen))
+		/* extra header descriptor if also payload in first buffer */
+		desc_cnt = DIV_ROUND_UP((skb_headlen(skb) - hdrlen), bp->max_tx_length) + 1;
+	else
+		desc_cnt = DIV_ROUND_UP(skb_headlen(skb), bp->max_tx_length);
+	nr_frags = skb_shinfo(skb)->nr_frags;
+	for (f = 0; f < nr_frags; f++) {
+		frag_size = skb_frag_size(&skb_shinfo(skb)->frags[f]);
+		desc_cnt += DIV_ROUND_UP(frag_size, bp->max_tx_length);
+	}
+
+	spin_lock_irqsave(&queue->tx_ptr_lock, flags);
+
+	/* This is a hard error, log it. */
+	if (CIRC_SPACE(queue->tx_head, queue->tx_tail,
+		       bp->tx_ring_size) < desc_cnt) {
+		netif_stop_subqueue(dev, queue_index);
+		netdev_dbg(bp->dev, "tx_head = %u, tx_tail = %u\n",
+			   queue->tx_head, queue->tx_tail);
+		ret = NETDEV_TX_BUSY;
+		goto unlock;
+	}
+
+	/* Map socket buffer for DMA transfer */
+	if (!macb_tx_map(bp, queue, skb, hdrlen)) {
+		dev_kfree_skb_any(skb);
+		goto unlock;
+	}
+
+	/* Make newly initialized descriptor visible to hardware */
+	wmb();
+	skb_tx_timestamp(skb);
+	netdev_tx_sent_queue(netdev_get_tx_queue(bp->dev, queue_index),
+			     skb->len);
+
+	spin_lock(&bp->lock);
+	macb_writel(bp, NCR, macb_readl(bp, NCR) | MACB_BIT(TSTART));
+	spin_unlock(&bp->lock);
+
+	if (CIRC_SPACE(queue->tx_head, queue->tx_tail, bp->tx_ring_size) < 1)
+		netif_stop_subqueue(dev, queue_index);
+
+unlock:
+	spin_unlock_irqrestore(&queue->tx_ptr_lock, flags);
+
+	return ret;
+}
+
+static void macb_init_rx_buffer_size(struct macb *bp, size_t size)
+{
+	if (!macb_is_gem(bp)) {
+		bp->rx_buffer_size = MACB_RX_BUFFER_SIZE;
+	} else {
+		bp->rx_buffer_size = size;
+
+		if (bp->rx_buffer_size % RX_BUFFER_MULTIPLE) {
+			netdev_dbg(bp->dev,
+				   "RX buffer must be multiple of %d bytes, expanding\n",
+				   RX_BUFFER_MULTIPLE);
+			bp->rx_buffer_size =
+				roundup(bp->rx_buffer_size, RX_BUFFER_MULTIPLE);
+		}
+	}
+
+	netdev_dbg(bp->dev, "mtu [%u] rx_buffer_size [%zu]\n",
+		   bp->dev->mtu, bp->rx_buffer_size);
+}
+
+static void gem_free_rx_buffers(struct macb *bp)
+{
+	struct sk_buff		*skb;
+	struct macb_dma_desc	*desc;
+	struct macb_queue *queue;
+	dma_addr_t		addr;
+	unsigned int q;
+	int i;
+
+	for (q = 0, queue = bp->queues; q < bp->num_queues; ++q, ++queue) {
+		if (!queue->rx_skbuff)
+			continue;
+
+		for (i = 0; i < bp->rx_ring_size; i++) {
+			skb = queue->rx_skbuff[i];
+
+			if (!skb)
+				continue;
+
+			desc = macb_rx_desc(queue, i);
+			addr = macb_get_addr(bp, desc);
+
+			dma_unmap_single(&bp->pdev->dev, addr, bp->rx_buffer_size,
+					DMA_FROM_DEVICE);
+			dev_kfree_skb_any(skb);
+			skb = NULL;
+		}
+
+		kfree(queue->rx_skbuff);
+		queue->rx_skbuff = NULL;
+	}
+}
+
+static void macb_free_rx_buffers(struct macb *bp)
+{
+	struct macb_queue *queue = &bp->queues[0];
+
+	if (queue->rx_buffers) {
+		dma_free_coherent(&bp->pdev->dev,
+				  bp->rx_ring_size * bp->rx_buffer_size,
+				  queue->rx_buffers, queue->rx_buffers_dma);
+		queue->rx_buffers = NULL;
+	}
+}
+
+static unsigned int macb_tx_ring_size_per_queue(struct macb *bp)
+{
+	return macb_dma_desc_get_size(bp) * bp->tx_ring_size + bp->tx_bd_rd_prefetch;
+}
+
+static unsigned int macb_rx_ring_size_per_queue(struct macb *bp)
+{
+	return macb_dma_desc_get_size(bp) * bp->rx_ring_size + bp->rx_bd_rd_prefetch;
+}
+
+static void macb_free_consistent(struct macb *bp)
+{
+	struct device *dev = &bp->pdev->dev;
+	struct macb_queue *queue;
+	unsigned int q;
+	size_t size;
+
+	if (bp->rx_ring_tieoff) {
+		dma_free_coherent(dev, macb_dma_desc_get_size(bp),
+				  bp->rx_ring_tieoff, bp->rx_ring_tieoff_dma);
+		bp->rx_ring_tieoff = NULL;
+	}
+
+	bp->macbgem_ops.mog_free_rx_buffers(bp);
+
+	size = bp->num_queues * macb_tx_ring_size_per_queue(bp);
+	dma_free_coherent(dev, size, bp->queues[0].tx_ring, bp->queues[0].tx_ring_dma);
+
+	size = bp->num_queues * macb_rx_ring_size_per_queue(bp);
+	dma_free_coherent(dev, size, bp->queues[0].rx_ring, bp->queues[0].rx_ring_dma);
+
+	for (q = 0, queue = bp->queues; q < bp->num_queues; ++q, ++queue) {
+		kfree(queue->tx_skb);
+		queue->tx_skb = NULL;
+		queue->tx_ring = NULL;
+		queue->rx_ring = NULL;
+	}
+}
+
+static int gem_alloc_rx_buffers(struct macb *bp)
+{
+	struct macb_queue *queue;
+	unsigned int q;
+	int size;
+
+	for (q = 0, queue = bp->queues; q < bp->num_queues; ++q, ++queue) {
+		size = bp->rx_ring_size * sizeof(struct sk_buff *);
+		queue->rx_skbuff = kzalloc(size, GFP_KERNEL);
+		if (!queue->rx_skbuff)
+			return -ENOMEM;
+		else
+			netdev_dbg(bp->dev,
+				   "Allocated %d RX struct sk_buff entries at %p\n",
+				   bp->rx_ring_size, queue->rx_skbuff);
+	}
+	return 0;
+}
+
+static int macb_alloc_rx_buffers(struct macb *bp)
+{
+	struct macb_queue *queue = &bp->queues[0];
+	int size;
+
+	size = bp->rx_ring_size * bp->rx_buffer_size;
+	queue->rx_buffers = dma_alloc_coherent(&bp->pdev->dev, size,
+					    &queue->rx_buffers_dma, GFP_KERNEL);
+	if (!queue->rx_buffers)
+		return -ENOMEM;
+
+	netdev_dbg(bp->dev,
+		   "Allocated RX buffers of %d bytes at %08lx (mapped %p)\n",
+		   size, (unsigned long)queue->rx_buffers_dma, queue->rx_buffers);
+	return 0;
+}
+
+static int macb_alloc_consistent(struct macb *bp)
+{
+	struct device *dev = &bp->pdev->dev;
+	dma_addr_t tx_dma, rx_dma;
+	struct macb_queue *queue;
+	unsigned int q;
+	void *tx, *rx;
+	size_t size;
+
+	/*
+	 * Upper 32-bits of Tx/Rx DMA descriptor for each queues much match!
+	 * We cannot enforce this guarantee, the best we can do is do a single
+	 * allocation and hope it will land into alloc_pages() that guarantees
+	 * natural alignment of physical addresses.
+	 */
+
+	size = bp->num_queues * macb_tx_ring_size_per_queue(bp);
+	tx = dma_alloc_coherent(dev, size, &tx_dma, GFP_KERNEL);
+	if (!tx || upper_32_bits(tx_dma) != upper_32_bits(tx_dma + size - 1))
+		goto out_err;
+	netdev_dbg(bp->dev, "Allocated %zu bytes for %u TX rings at %08lx (mapped %p)\n",
+		   size, bp->num_queues, (unsigned long)tx_dma, tx);
+
+	size = bp->num_queues * macb_rx_ring_size_per_queue(bp);
+	rx = dma_alloc_coherent(dev, size, &rx_dma, GFP_KERNEL);
+	if (!rx || upper_32_bits(rx_dma) != upper_32_bits(rx_dma + size - 1))
+		goto out_err;
+	netdev_dbg(bp->dev, "Allocated %zu bytes for %u RX rings at %08lx (mapped %p)\n",
+		   size, bp->num_queues, (unsigned long)rx_dma, rx);
+
+	for (q = 0, queue = bp->queues; q < bp->num_queues; ++q, ++queue) {
+		queue->tx_ring = tx + macb_tx_ring_size_per_queue(bp) * q;
+		queue->tx_ring_dma = tx_dma + macb_tx_ring_size_per_queue(bp) * q;
+
+		queue->rx_ring = rx + macb_rx_ring_size_per_queue(bp) * q;
+		queue->rx_ring_dma = rx_dma + macb_rx_ring_size_per_queue(bp) * q;
+
+		size = bp->tx_ring_size * sizeof(struct macb_tx_skb);
+		queue->tx_skb = kmalloc(size, GFP_KERNEL);
+		if (!queue->tx_skb)
+			goto out_err;
+	}
+	if (bp->macbgem_ops.mog_alloc_rx_buffers(bp))
+		goto out_err;
+
+	/* Required for tie off descriptor for PM cases */
+	if (!(bp->caps & MACB_CAPS_QUEUE_DISABLE)) {
+		bp->rx_ring_tieoff = dma_alloc_coherent(&bp->pdev->dev,
+							macb_dma_desc_get_size(bp),
+							&bp->rx_ring_tieoff_dma,
+							GFP_KERNEL);
+		if (!bp->rx_ring_tieoff)
+			goto out_err;
+	}
+
+	return 0;
+
+out_err:
+	macb_free_consistent(bp);
+	return -ENOMEM;
+}
+
+static void macb_init_tieoff(struct macb *bp)
+{
+	struct macb_dma_desc *desc = bp->rx_ring_tieoff;
+
+	if (bp->caps & MACB_CAPS_QUEUE_DISABLE)
+		return;
+	/* Setup a wrapping descriptor with no free slots
+	 * (WRAP and USED) to tie off/disable unused RX queues.
+	 */
+	macb_set_addr(bp, desc, MACB_BIT(RX_WRAP) | MACB_BIT(RX_USED));
+	desc->ctrl = 0;
+}
+
+static void gem_init_rings(struct macb *bp)
+{
+	struct macb_queue *queue;
+	struct macb_dma_desc *desc = NULL;
+	unsigned int q;
+	int i;
+
+	for (q = 0, queue = bp->queues; q < bp->num_queues; ++q, ++queue) {
+		for (i = 0; i < bp->tx_ring_size; i++) {
+			desc = macb_tx_desc(queue, i);
+			macb_set_addr(bp, desc, 0);
+			desc->ctrl = MACB_BIT(TX_USED);
+		}
+		desc->ctrl |= MACB_BIT(TX_WRAP);
+		queue->tx_head = 0;
+		queue->tx_tail = 0;
+
+		queue->rx_tail = 0;
+		queue->rx_prepared_head = 0;
+
+		gem_rx_refill(queue);
+	}
+
+	macb_init_tieoff(bp);
+}
+
+static void macb_init_rings(struct macb *bp)
+{
+	int i;
+	struct macb_dma_desc *desc = NULL;
+
+	macb_init_rx_ring(&bp->queues[0]);
+
+	for (i = 0; i < bp->tx_ring_size; i++) {
+		desc = macb_tx_desc(&bp->queues[0], i);
+		macb_set_addr(bp, desc, 0);
+		desc->ctrl = MACB_BIT(TX_USED);
+	}
+	bp->queues[0].tx_head = 0;
+	bp->queues[0].tx_tail = 0;
+	desc->ctrl |= MACB_BIT(TX_WRAP);
+
+	macb_init_tieoff(bp);
+}
+
+static void macb_reset_hw(struct macb *bp)
+{
+	struct macb_queue *queue;
+	unsigned int q;
+	u32 ctrl = macb_readl(bp, NCR);
+
+	/* Disable RX and TX (XXX: Should we halt the transmission
+	 * more gracefully?)
+	 */
+	ctrl &= ~(MACB_BIT(RE) | MACB_BIT(TE));
+
+	/* Clear the stats registers (XXX: Update stats first?) */
+	ctrl |= MACB_BIT(CLRSTAT);
+
+	macb_writel(bp, NCR, ctrl);
+
+	/* Clear all status flags */
+	macb_writel(bp, TSR, -1);
+	macb_writel(bp, RSR, -1);
+
+	/* Disable RX partial store and forward and reset watermark value */
+	gem_writel(bp, PBUFRXCUT, 0);
+
+	/* Disable all interrupts */
+	for (q = 0, queue = bp->queues; q < bp->num_queues; ++q, ++queue) {
+		queue_writel(queue, IDR, -1);
+		queue_readl(queue, ISR);
+		if (bp->caps & MACB_CAPS_ISR_CLEAR_ON_WRITE)
+			queue_writel(queue, ISR, -1);
+	}
+}
+
+static u32 gem_mdc_clk_div(struct macb *bp)
+{
+	u32 config;
+	unsigned long pclk_hz = clk_get_rate(bp->pclk);
+
+	if (pclk_hz <= 20000000)
+		config = GEM_BF(CLK, GEM_CLK_DIV8);
+	else if (pclk_hz <= 40000000)
+		config = GEM_BF(CLK, GEM_CLK_DIV16);
+	else if (pclk_hz <= 80000000)
+		config = GEM_BF(CLK, GEM_CLK_DIV32);
+	else if (pclk_hz <= 120000000)
+		config = GEM_BF(CLK, GEM_CLK_DIV48);
+	else if (pclk_hz <= 160000000)
+		config = GEM_BF(CLK, GEM_CLK_DIV64);
+	else if (pclk_hz <= 240000000)
+		config = GEM_BF(CLK, GEM_CLK_DIV96);
+	else if (pclk_hz <= 320000000)
+		config = GEM_BF(CLK, GEM_CLK_DIV128);
+	else
+		config = GEM_BF(CLK, GEM_CLK_DIV224);
+
+	return config;
+}
+
+static u32 macb_mdc_clk_div(struct macb *bp)
+{
+	u32 config;
+	unsigned long pclk_hz;
+
+	if (macb_is_gem(bp))
+		return gem_mdc_clk_div(bp);
+
+	pclk_hz = clk_get_rate(bp->pclk);
+	if (pclk_hz <= 20000000)
+		config = MACB_BF(CLK, MACB_CLK_DIV8);
+	else if (pclk_hz <= 40000000)
+		config = MACB_BF(CLK, MACB_CLK_DIV16);
+	else if (pclk_hz <= 80000000)
+		config = MACB_BF(CLK, MACB_CLK_DIV32);
+	else
+		config = MACB_BF(CLK, MACB_CLK_DIV64);
+
+	return config;
+}
+
+/* Get the DMA bus width field of the network configuration register that we
+ * should program.  We find the width from decoding the design configuration
+ * register to find the maximum supported data bus width.
+ */
+static u32 macb_dbw(struct macb *bp)
+{
+	if (!macb_is_gem(bp))
+		return 0;
+
+	switch (GEM_BFEXT(DBWDEF, gem_readl(bp, DCFG1))) {
+	case 4:
+		return GEM_BF(DBW, GEM_DBW128);
+	case 2:
+		return GEM_BF(DBW, GEM_DBW64);
+	case 1:
+	default:
+		return GEM_BF(DBW, GEM_DBW32);
+	}
+}
+
+/* Configure the receive DMA engine
+ * - use the correct receive buffer size
+ * - set best burst length for DMA operations
+ *   (if not supported by FIFO, it will fallback to default)
+ * - set both rx/tx packet buffers to full memory size
+ * These are configurable parameters for GEM.
+ */
+static void macb_configure_dma(struct macb *bp)
+{
+	struct macb_queue *queue;
+	u32 buffer_size;
+	unsigned int q;
+	u32 dmacfg;
+
+	buffer_size = bp->rx_buffer_size / RX_BUFFER_MULTIPLE;
+	if (macb_is_gem(bp)) {
+		dmacfg = gem_readl(bp, DMACFG) & ~GEM_BF(RXBS, -1L);
+		for (q = 0, queue = bp->queues; q < bp->num_queues; ++q, ++queue) {
+			if (q)
+				queue_writel(queue, RBQS, buffer_size);
+			else
+				dmacfg |= GEM_BF(RXBS, buffer_size);
+		}
+		if (bp->dma_burst_length)
+			dmacfg = GEM_BFINS(FBLDO, bp->dma_burst_length, dmacfg);
+		dmacfg |= GEM_BIT(TXPBMS) | GEM_BF(RXBMS, -1L);
+		dmacfg &= ~GEM_BIT(ENDIA_PKT);
+
+		if (bp->native_io)
+			dmacfg &= ~GEM_BIT(ENDIA_DESC);
+		else
+			dmacfg |= GEM_BIT(ENDIA_DESC); /* CPU in big endian */
+
+		if (bp->dev->features & NETIF_F_HW_CSUM)
+			dmacfg |= GEM_BIT(TXCOEN);
+		else
+			dmacfg &= ~GEM_BIT(TXCOEN);
+
+		dmacfg &= ~GEM_BIT(ADDR64);
+#ifdef CONFIG_ARCH_DMA_ADDR_T_64BIT
+		if (bp->hw_dma_cap & HW_DMA_CAP_64B)
+			dmacfg |= GEM_BIT(ADDR64);
+#endif
+#ifdef CONFIG_MACB_USE_HWSTAMP
+		if (bp->hw_dma_cap & HW_DMA_CAP_PTP)
+			dmacfg |= GEM_BIT(RXEXT) | GEM_BIT(TXEXT);
+#endif
+		netdev_dbg(bp->dev, "Cadence configure DMA with 0x%08x\n",
+			   dmacfg);
+		gem_writel(bp, DMACFG, dmacfg);
+	}
+}
+
+static void macb_init_hw(struct macb *bp)
+{
+	u32 config;
+
+	macb_reset_hw(bp);
+	macb_set_hwaddr(bp);
+
+	config = macb_mdc_clk_div(bp);
+	config |= MACB_BF(RBOF, NET_IP_ALIGN);	/* Make eth data aligned */
+	config |= MACB_BIT(DRFCS);		/* Discard Rx FCS */
+	if (bp->caps & MACB_CAPS_JUMBO)
+		config |= MACB_BIT(JFRAME);	/* Enable jumbo frames */
+	else
+		config |= MACB_BIT(BIG);	/* Receive oversized frames */
+	if (bp->dev->flags & IFF_PROMISC)
+		config |= MACB_BIT(CAF);	/* Copy All Frames */
+	else if (macb_is_gem(bp) && bp->dev->features & NETIF_F_RXCSUM)
+		config |= GEM_BIT(RXCOEN);
+	if (!(bp->dev->flags & IFF_BROADCAST))
+		config |= MACB_BIT(NBC);	/* No BroadCast */
+	config |= macb_dbw(bp);
+	macb_writel(bp, NCFGR, config);
+	if ((bp->caps & MACB_CAPS_JUMBO) && bp->jumbo_max_len)
+		gem_writel(bp, JML, bp->jumbo_max_len);
+	bp->rx_frm_len_mask = MACB_RX_FRMLEN_MASK;
+	if (bp->caps & MACB_CAPS_JUMBO)
+		bp->rx_frm_len_mask = MACB_RX_JFRMLEN_MASK;
+
+	macb_configure_dma(bp);
+
+	/* Enable RX partial store and forward and set watermark */
+	if (bp->rx_watermark)
+		gem_writel(bp, PBUFRXCUT, (bp->rx_watermark | GEM_BIT(ENCUTTHRU)));
+}
+
+/* The hash address register is 64 bits long and takes up two
+ * locations in the memory map.  The least significant bits are stored
+ * in EMAC_HSL and the most significant bits in EMAC_HSH.
+ *
+ * The unicast hash enable and the multicast hash enable bits in the
+ * network configuration register enable the reception of hash matched
+ * frames. The destination address is reduced to a 6 bit index into
+ * the 64 bit hash register using the following hash function.  The
+ * hash function is an exclusive or of every sixth bit of the
+ * destination address.
+ *
+ * hi[5] = da[5] ^ da[11] ^ da[17] ^ da[23] ^ da[29] ^ da[35] ^ da[41] ^ da[47]
+ * hi[4] = da[4] ^ da[10] ^ da[16] ^ da[22] ^ da[28] ^ da[34] ^ da[40] ^ da[46]
+ * hi[3] = da[3] ^ da[09] ^ da[15] ^ da[21] ^ da[27] ^ da[33] ^ da[39] ^ da[45]
+ * hi[2] = da[2] ^ da[08] ^ da[14] ^ da[20] ^ da[26] ^ da[32] ^ da[38] ^ da[44]
+ * hi[1] = da[1] ^ da[07] ^ da[13] ^ da[19] ^ da[25] ^ da[31] ^ da[37] ^ da[43]
+ * hi[0] = da[0] ^ da[06] ^ da[12] ^ da[18] ^ da[24] ^ da[30] ^ da[36] ^ da[42]
+ *
+ * da[0] represents the least significant bit of the first byte
+ * received, that is, the multicast/unicast indicator, and da[47]
+ * represents the most significant bit of the last byte received.  If
+ * the hash index, hi[n], points to a bit that is set in the hash
+ * register then the frame will be matched according to whether the
+ * frame is multicast or unicast.  A multicast match will be signalled
+ * if the multicast hash enable bit is set, da[0] is 1 and the hash
+ * index points to a bit set in the hash register.  A unicast match
+ * will be signalled if the unicast hash enable bit is set, da[0] is 0
+ * and the hash index points to a bit set in the hash register.  To
+ * receive all multicast frames, the hash register should be set with
+ * all ones and the multicast hash enable bit should be set in the
+ * network configuration register.
+ */
+
+static inline int hash_bit_value(int bitnr, __u8 *addr)
+{
+	if (addr[bitnr / 8] & (1 << (bitnr % 8)))
+		return 1;
+	return 0;
+}
+
+/* Return the hash index value for the specified address. */
+static int hash_get_index(__u8 *addr)
+{
+	int i, j, bitval;
+	int hash_index = 0;
+
+	for (j = 0; j < 6; j++) {
+		for (i = 0, bitval = 0; i < 8; i++)
+			bitval ^= hash_bit_value(i * 6 + j, addr);
+
+		hash_index |= (bitval << j);
+	}
+
+	return hash_index;
+}
+
+/* Add multicast addresses to the internal multicast-hash table. */
+static void macb_sethashtable(struct net_device *dev)
+{
+	struct netdev_hw_addr *ha;
+	unsigned long mc_filter[2];
+	unsigned int bitnr;
+	struct macb *bp = netdev_priv(dev);
+
+	mc_filter[0] = 0;
+	mc_filter[1] = 0;
+
+	netdev_for_each_mc_addr(ha, dev) {
+		bitnr = hash_get_index(ha->addr);
+		mc_filter[bitnr >> 5] |= 1 << (bitnr & 31);
+	}
+
+	macb_or_gem_writel(bp, HRB, mc_filter[0]);
+	macb_or_gem_writel(bp, HRT, mc_filter[1]);
+}
+
+/* Enable/Disable promiscuous and multicast modes. */
+static void macb_set_rx_mode(struct net_device *dev)
+{
+	unsigned long cfg;
+	struct macb *bp = netdev_priv(dev);
+
+	cfg = macb_readl(bp, NCFGR);
+
+	if (dev->flags & IFF_PROMISC) {
+		/* Enable promiscuous mode */
+		cfg |= MACB_BIT(CAF);
+
+		/* Disable RX checksum offload */
+		if (macb_is_gem(bp))
+			cfg &= ~GEM_BIT(RXCOEN);
+	} else {
+		/* Disable promiscuous mode */
+		cfg &= ~MACB_BIT(CAF);
+
+		/* Enable RX checksum offload only if requested */
+		if (macb_is_gem(bp) && dev->features & NETIF_F_RXCSUM)
+			cfg |= GEM_BIT(RXCOEN);
+	}
+
+	if (dev->flags & IFF_ALLMULTI) {
+		/* Enable all multicast mode */
+		macb_or_gem_writel(bp, HRB, -1);
+		macb_or_gem_writel(bp, HRT, -1);
+		cfg |= MACB_BIT(NCFGR_MTI);
+	} else if (!netdev_mc_empty(dev)) {
+		/* Enable specific multicasts */
+		macb_sethashtable(dev);
+		cfg |= MACB_BIT(NCFGR_MTI);
+	} else if (dev->flags & (~IFF_ALLMULTI)) {
+		/* Disable all multicast mode */
+		macb_or_gem_writel(bp, HRB, 0);
+		macb_or_gem_writel(bp, HRT, 0);
+		cfg &= ~MACB_BIT(NCFGR_MTI);
+	}
+
+	macb_writel(bp, NCFGR, cfg);
+}
+
+static int macb_open(struct net_device *dev)
+{
+	size_t bufsz = dev->mtu + ETH_HLEN + ETH_FCS_LEN + NET_IP_ALIGN;
+	struct macb *bp = netdev_priv(dev);
+	struct macb_queue *queue;
+	unsigned int q;
+	int err;
+
+	netdev_dbg(bp->dev, "open\n");
+
+	err = pm_runtime_resume_and_get(&bp->pdev->dev);
+	if (err < 0)
+		return err;
+
+	/* RX buffers initialization */
+	macb_init_rx_buffer_size(bp, bufsz);
+
+	err = macb_alloc_consistent(bp);
+	if (err) {
+		netdev_err(dev, "Unable to allocate DMA memory (error %d)\n",
+			   err);
+		goto pm_exit;
+	}
+
+	for (q = 0, queue = bp->queues; q < bp->num_queues; ++q, ++queue) {
+		napi_enable(&queue->napi_rx);
+		napi_enable(&queue->napi_tx);
+	}
+
+	macb_init_hw(bp);
+
+	err = phy_power_on(bp->sgmii_phy);
+	if (err)
+		goto reset_hw;
+
+	err = macb_phylink_connect(bp);
+	if (err)
+		goto phy_off;
+
+	netif_tx_start_all_queues(dev);
+
+	if (bp->ptp_info)
+		bp->ptp_info->ptp_init(dev);
+
+	return 0;
+
+phy_off:
+	phy_power_off(bp->sgmii_phy);
+
+reset_hw:
+	macb_reset_hw(bp);
+	for (q = 0, queue = bp->queues; q < bp->num_queues; ++q, ++queue) {
+		napi_disable(&queue->napi_rx);
+		napi_disable(&queue->napi_tx);
+	}
+	macb_free_consistent(bp);
+pm_exit:
+	pm_runtime_put_sync(&bp->pdev->dev);
+	return err;
+}
+
+static int macb_close(struct net_device *dev)
+{
+	struct macb *bp = netdev_priv(dev);
+	struct macb_queue *queue;
+	unsigned long flags;
+	unsigned int q;
+
+	netif_tx_stop_all_queues(dev);
+
+	for (q = 0, queue = bp->queues; q < bp->num_queues; ++q, ++queue) {
+		napi_disable(&queue->napi_rx);
+		napi_disable(&queue->napi_tx);
+		netdev_tx_reset_queue(netdev_get_tx_queue(dev, q));
+	}
+
+	phylink_stop(bp->phylink);
+	phylink_disconnect_phy(bp->phylink);
+
+	phy_power_off(bp->sgmii_phy);
+
+	spin_lock_irqsave(&bp->lock, flags);
+	macb_reset_hw(bp);
+	netif_carrier_off(dev);
+	spin_unlock_irqrestore(&bp->lock, flags);
+
+	macb_free_consistent(bp);
+
+	if (bp->ptp_info)
+		bp->ptp_info->ptp_remove(dev);
+
+	pm_runtime_put(&bp->pdev->dev);
+
+	return 0;
+}
+
+static int macb_change_mtu(struct net_device *dev, int new_mtu)
+{
+	if (netif_running(dev))
+		return -EBUSY;
+
+	WRITE_ONCE(dev->mtu, new_mtu);
+
+	return 0;
+}
+
+static int macb_set_mac_addr(struct net_device *dev, void *addr)
+{
+	int err;
+
+	err = eth_mac_addr(dev, addr);
+	if (err < 0)
+		return err;
+
+	macb_set_hwaddr(netdev_priv(dev));
+	return 0;
+}
+
+static void gem_update_stats(struct macb *bp)
+{
+	struct macb_queue *queue;
+	unsigned int i, q, idx;
+	unsigned long *stat;
+
+	u64 *p = &bp->hw_stats.gem.tx_octets;
+
+	for (i = 0; i < GEM_STATS_LEN; ++i, ++p) {
+		u32 offset = gem_statistics[i].offset;
+		u64 val = bp->macb_reg_readl(bp, offset);
+
+		bp->ethtool_stats[i] += val;
+		*p += val;
+
+		if (offset == GEM_OCTTXL || offset == GEM_OCTRXL) {
+			/* Add GEM_OCTTXH, GEM_OCTRXH */
+			val = bp->macb_reg_readl(bp, offset + 4);
+			bp->ethtool_stats[i] += ((u64)val) << 32;
+			*p += ((u64)val) << 32;
+		}
+	}
+
+	idx = GEM_STATS_LEN;
+	for (q = 0, queue = bp->queues; q < bp->num_queues; ++q, ++queue)
+		for (i = 0, stat = &queue->stats.first; i < QUEUE_STATS_LEN; ++i, ++stat)
+			bp->ethtool_stats[idx++] = *stat;
+}
+
+static void gem_get_stats(struct macb *bp, struct rtnl_link_stats64 *nstat)
+{
+	struct gem_stats *hwstat = &bp->hw_stats.gem;
+
+	spin_lock_irq(&bp->stats_lock);
+	if (netif_running(bp->dev))
+		gem_update_stats(bp);
+
+	nstat->rx_errors = (hwstat->rx_frame_check_sequence_errors +
+			    hwstat->rx_alignment_errors +
+			    hwstat->rx_resource_errors +
+			    hwstat->rx_overruns +
+			    hwstat->rx_oversize_frames +
+			    hwstat->rx_jabbers +
+			    hwstat->rx_undersized_frames +
+			    hwstat->rx_length_field_frame_errors);
+	nstat->tx_errors = (hwstat->tx_late_collisions +
+			    hwstat->tx_excessive_collisions +
+			    hwstat->tx_underrun +
+			    hwstat->tx_carrier_sense_errors);
+	nstat->multicast = hwstat->rx_multicast_frames;
+	nstat->collisions = (hwstat->tx_single_collision_frames +
+			     hwstat->tx_multiple_collision_frames +
+			     hwstat->tx_excessive_collisions);
+	nstat->rx_length_errors = (hwstat->rx_oversize_frames +
+				   hwstat->rx_jabbers +
+				   hwstat->rx_undersized_frames +
+				   hwstat->rx_length_field_frame_errors);
+	nstat->rx_over_errors = hwstat->rx_resource_errors;
+	nstat->rx_crc_errors = hwstat->rx_frame_check_sequence_errors;
+	nstat->rx_frame_errors = hwstat->rx_alignment_errors;
+	nstat->rx_fifo_errors = hwstat->rx_overruns;
+	nstat->tx_aborted_errors = hwstat->tx_excessive_collisions;
+	nstat->tx_carrier_errors = hwstat->tx_carrier_sense_errors;
+	nstat->tx_fifo_errors = hwstat->tx_underrun;
+	spin_unlock_irq(&bp->stats_lock);
+}
+
+static void gem_get_ethtool_stats(struct net_device *dev,
+				  struct ethtool_stats *stats, u64 *data)
+{
+	struct macb *bp = netdev_priv(dev);
+
+	spin_lock_irq(&bp->stats_lock);
+	gem_update_stats(bp);
+	memcpy(data, &bp->ethtool_stats, sizeof(u64)
+			* (GEM_STATS_LEN + QUEUE_STATS_LEN * MACB_MAX_QUEUES));
+	spin_unlock_irq(&bp->stats_lock);
+}
+
+static int gem_get_sset_count(struct net_device *dev, int sset)
+{
+	struct macb *bp = netdev_priv(dev);
+
+	switch (sset) {
+	case ETH_SS_STATS:
+		return GEM_STATS_LEN + bp->num_queues * QUEUE_STATS_LEN;
+	default:
+		return -EOPNOTSUPP;
+	}
+}
+
+static void gem_get_ethtool_strings(struct net_device *dev, u32 sset, u8 *p)
+{
+	char stat_string[ETH_GSTRING_LEN];
+	struct macb *bp = netdev_priv(dev);
+	struct macb_queue *queue;
+	unsigned int i;
+	unsigned int q;
+
+	switch (sset) {
+	case ETH_SS_STATS:
+		for (i = 0; i < GEM_STATS_LEN; i++, p += ETH_GSTRING_LEN)
+			memcpy(p, gem_statistics[i].stat_string,
+			       ETH_GSTRING_LEN);
+
+		for (q = 0, queue = bp->queues; q < bp->num_queues; ++q, ++queue) {
+			for (i = 0; i < QUEUE_STATS_LEN; i++, p += ETH_GSTRING_LEN) {
+				snprintf(stat_string, ETH_GSTRING_LEN, "q%d_%s",
+						q, queue_statistics[i].stat_string);
+				memcpy(p, stat_string, ETH_GSTRING_LEN);
+			}
+		}
+		break;
+	}
+}
+
+static void macb_get_stats(struct net_device *dev,
+			   struct rtnl_link_stats64 *nstat)
+{
+	struct macb *bp = netdev_priv(dev);
+	struct macb_stats *hwstat = &bp->hw_stats.macb;
+
+	netdev_stats_to_stats64(nstat, &bp->dev->stats);
+	if (macb_is_gem(bp)) {
+		gem_get_stats(bp, nstat);
+		return;
+	}
+
+	/* read stats from hardware */
+	spin_lock_irq(&bp->stats_lock);
+	macb_update_stats(bp);
+
+	/* Convert HW stats into netdevice stats */
+	nstat->rx_errors = (hwstat->rx_fcs_errors +
+			    hwstat->rx_align_errors +
+			    hwstat->rx_resource_errors +
+			    hwstat->rx_overruns +
+			    hwstat->rx_oversize_pkts +
+			    hwstat->rx_jabbers +
+			    hwstat->rx_undersize_pkts +
+			    hwstat->rx_length_mismatch);
+	nstat->tx_errors = (hwstat->tx_late_cols +
+			    hwstat->tx_excessive_cols +
+			    hwstat->tx_underruns +
+			    hwstat->tx_carrier_errors +
+			    hwstat->sqe_test_errors);
+	nstat->collisions = (hwstat->tx_single_cols +
+			     hwstat->tx_multiple_cols +
+			     hwstat->tx_excessive_cols);
+	nstat->rx_length_errors = (hwstat->rx_oversize_pkts +
+				   hwstat->rx_jabbers +
+				   hwstat->rx_undersize_pkts +
+				   hwstat->rx_length_mismatch);
+	nstat->rx_over_errors = hwstat->rx_resource_errors +
+				   hwstat->rx_overruns;
+	nstat->rx_crc_errors = hwstat->rx_fcs_errors;
+	nstat->rx_frame_errors = hwstat->rx_align_errors;
+	nstat->rx_fifo_errors = hwstat->rx_overruns;
+	/* XXX: What does "missed" mean? */
+	nstat->tx_aborted_errors = hwstat->tx_excessive_cols;
+	nstat->tx_carrier_errors = hwstat->tx_carrier_errors;
+	nstat->tx_fifo_errors = hwstat->tx_underruns;
+	/* Don't know about heartbeat or window errors... */
+	spin_unlock_irq(&bp->stats_lock);
+}
+
+static void macb_get_pause_stats(struct net_device *dev,
+				 struct ethtool_pause_stats *pause_stats)
+{
+	struct macb *bp = netdev_priv(dev);
+	struct macb_stats *hwstat = &bp->hw_stats.macb;
+
+	spin_lock_irq(&bp->stats_lock);
+	macb_update_stats(bp);
+	pause_stats->tx_pause_frames = hwstat->tx_pause_frames;
+	pause_stats->rx_pause_frames = hwstat->rx_pause_frames;
+	spin_unlock_irq(&bp->stats_lock);
+}
+
+static void gem_get_pause_stats(struct net_device *dev,
+				struct ethtool_pause_stats *pause_stats)
+{
+	struct macb *bp = netdev_priv(dev);
+	struct gem_stats *hwstat = &bp->hw_stats.gem;
+
+	spin_lock_irq(&bp->stats_lock);
+	gem_update_stats(bp);
+	pause_stats->tx_pause_frames = hwstat->tx_pause_frames;
+	pause_stats->rx_pause_frames = hwstat->rx_pause_frames;
+	spin_unlock_irq(&bp->stats_lock);
+}
+
+static void macb_get_eth_mac_stats(struct net_device *dev,
+				   struct ethtool_eth_mac_stats *mac_stats)
+{
+	struct macb *bp = netdev_priv(dev);
+	struct macb_stats *hwstat = &bp->hw_stats.macb;
+
+	spin_lock_irq(&bp->stats_lock);
+	macb_update_stats(bp);
+	mac_stats->FramesTransmittedOK = hwstat->tx_ok;
+	mac_stats->SingleCollisionFrames = hwstat->tx_single_cols;
+	mac_stats->MultipleCollisionFrames = hwstat->tx_multiple_cols;
+	mac_stats->FramesReceivedOK = hwstat->rx_ok;
+	mac_stats->FrameCheckSequenceErrors = hwstat->rx_fcs_errors;
+	mac_stats->AlignmentErrors = hwstat->rx_align_errors;
+	mac_stats->FramesWithDeferredXmissions = hwstat->tx_deferred;
+	mac_stats->LateCollisions = hwstat->tx_late_cols;
+	mac_stats->FramesAbortedDueToXSColls = hwstat->tx_excessive_cols;
+	mac_stats->FramesLostDueToIntMACXmitError = hwstat->tx_underruns;
+	mac_stats->CarrierSenseErrors = hwstat->tx_carrier_errors;
+	mac_stats->FramesLostDueToIntMACRcvError = hwstat->rx_overruns;
+	mac_stats->InRangeLengthErrors = hwstat->rx_length_mismatch;
+	mac_stats->FrameTooLongErrors = hwstat->rx_oversize_pkts;
+	spin_unlock_irq(&bp->stats_lock);
+}
+
+static void gem_get_eth_mac_stats(struct net_device *dev,
+				  struct ethtool_eth_mac_stats *mac_stats)
+{
+	struct macb *bp = netdev_priv(dev);
+	struct gem_stats *hwstat = &bp->hw_stats.gem;
+
+	spin_lock_irq(&bp->stats_lock);
+	gem_update_stats(bp);
+	mac_stats->FramesTransmittedOK = hwstat->tx_frames;
+	mac_stats->SingleCollisionFrames = hwstat->tx_single_collision_frames;
+	mac_stats->MultipleCollisionFrames =
+		hwstat->tx_multiple_collision_frames;
+	mac_stats->FramesReceivedOK = hwstat->rx_frames;
+	mac_stats->FrameCheckSequenceErrors =
+		hwstat->rx_frame_check_sequence_errors;
+	mac_stats->AlignmentErrors = hwstat->rx_alignment_errors;
+	mac_stats->OctetsTransmittedOK = hwstat->tx_octets;
+	mac_stats->FramesWithDeferredXmissions = hwstat->tx_deferred_frames;
+	mac_stats->LateCollisions = hwstat->tx_late_collisions;
+	mac_stats->FramesAbortedDueToXSColls = hwstat->tx_excessive_collisions;
+	mac_stats->FramesLostDueToIntMACXmitError = hwstat->tx_underrun;
+	mac_stats->CarrierSenseErrors = hwstat->tx_carrier_sense_errors;
+	mac_stats->OctetsReceivedOK = hwstat->rx_octets;
+	mac_stats->MulticastFramesXmittedOK = hwstat->tx_multicast_frames;
+	mac_stats->BroadcastFramesXmittedOK = hwstat->tx_broadcast_frames;
+	mac_stats->MulticastFramesReceivedOK = hwstat->rx_multicast_frames;
+	mac_stats->BroadcastFramesReceivedOK = hwstat->rx_broadcast_frames;
+	mac_stats->InRangeLengthErrors = hwstat->rx_length_field_frame_errors;
+	mac_stats->FrameTooLongErrors = hwstat->rx_oversize_frames;
+	spin_unlock_irq(&bp->stats_lock);
+}
+
+/* TODO: Report SQE test errors when added to phy_stats */
+static void macb_get_eth_phy_stats(struct net_device *dev,
+				   struct ethtool_eth_phy_stats *phy_stats)
+{
+	struct macb *bp = netdev_priv(dev);
+	struct macb_stats *hwstat = &bp->hw_stats.macb;
+
+	spin_lock_irq(&bp->stats_lock);
+	macb_update_stats(bp);
+	phy_stats->SymbolErrorDuringCarrier = hwstat->rx_symbol_errors;
+	spin_unlock_irq(&bp->stats_lock);
+}
+
+static void gem_get_eth_phy_stats(struct net_device *dev,
+				  struct ethtool_eth_phy_stats *phy_stats)
+{
+	struct macb *bp = netdev_priv(dev);
+	struct gem_stats *hwstat = &bp->hw_stats.gem;
+
+	spin_lock_irq(&bp->stats_lock);
+	gem_update_stats(bp);
+	phy_stats->SymbolErrorDuringCarrier = hwstat->rx_symbol_errors;
+	spin_unlock_irq(&bp->stats_lock);
+}
+
+static void macb_get_rmon_stats(struct net_device *dev,
+				struct ethtool_rmon_stats *rmon_stats,
+				const struct ethtool_rmon_hist_range **ranges)
+{
+	struct macb *bp = netdev_priv(dev);
+	struct macb_stats *hwstat = &bp->hw_stats.macb;
+
+	spin_lock_irq(&bp->stats_lock);
+	macb_update_stats(bp);
+	rmon_stats->undersize_pkts = hwstat->rx_undersize_pkts;
+	rmon_stats->oversize_pkts = hwstat->rx_oversize_pkts;
+	rmon_stats->jabbers = hwstat->rx_jabbers;
+	spin_unlock_irq(&bp->stats_lock);
+}
+
+static const struct ethtool_rmon_hist_range gem_rmon_ranges[] = {
+	{   64,    64 },
+	{   65,   127 },
+	{  128,   255 },
+	{  256,   511 },
+	{  512,  1023 },
+	{ 1024,  1518 },
+	{ 1519, 16384 },
+	{ },
+};
+
+static void gem_get_rmon_stats(struct net_device *dev,
+			       struct ethtool_rmon_stats *rmon_stats,
+			       const struct ethtool_rmon_hist_range **ranges)
+{
+	struct macb *bp = netdev_priv(dev);
+	struct gem_stats *hwstat = &bp->hw_stats.gem;
+
+	spin_lock_irq(&bp->stats_lock);
+	gem_update_stats(bp);
+	rmon_stats->undersize_pkts = hwstat->rx_undersized_frames;
+	rmon_stats->oversize_pkts = hwstat->rx_oversize_frames;
+	rmon_stats->jabbers = hwstat->rx_jabbers;
+	rmon_stats->hist[0] = hwstat->rx_64_byte_frames;
+	rmon_stats->hist[1] = hwstat->rx_65_127_byte_frames;
+	rmon_stats->hist[2] = hwstat->rx_128_255_byte_frames;
+	rmon_stats->hist[3] = hwstat->rx_256_511_byte_frames;
+	rmon_stats->hist[4] = hwstat->rx_512_1023_byte_frames;
+	rmon_stats->hist[5] = hwstat->rx_1024_1518_byte_frames;
+	rmon_stats->hist[6] = hwstat->rx_greater_than_1518_byte_frames;
+	rmon_stats->hist_tx[0] = hwstat->tx_64_byte_frames;
+	rmon_stats->hist_tx[1] = hwstat->tx_65_127_byte_frames;
+	rmon_stats->hist_tx[2] = hwstat->tx_128_255_byte_frames;
+	rmon_stats->hist_tx[3] = hwstat->tx_256_511_byte_frames;
+	rmon_stats->hist_tx[4] = hwstat->tx_512_1023_byte_frames;
+	rmon_stats->hist_tx[5] = hwstat->tx_1024_1518_byte_frames;
+	rmon_stats->hist_tx[6] = hwstat->tx_greater_than_1518_byte_frames;
+	spin_unlock_irq(&bp->stats_lock);
+	*ranges = gem_rmon_ranges;
+}
+
+static int macb_get_regs_len(struct net_device *netdev)
+{
+	return MACB_GREGS_NBR * sizeof(u32);
+}
+
+static void macb_get_regs(struct net_device *dev, struct ethtool_regs *regs,
+			  void *p)
+{
+	struct macb *bp = netdev_priv(dev);
+	unsigned int tail, head;
+	u32 *regs_buff = p;
+
+	regs->version = (macb_readl(bp, MID) & ((1 << MACB_REV_SIZE) - 1))
+			| MACB_GREGS_VERSION;
+
+	tail = macb_tx_ring_wrap(bp, bp->queues[0].tx_tail);
+	head = macb_tx_ring_wrap(bp, bp->queues[0].tx_head);
+
+	regs_buff[0]  = macb_readl(bp, NCR);
+	regs_buff[1]  = macb_or_gem_readl(bp, NCFGR);
+	regs_buff[2]  = macb_readl(bp, NSR);
+	regs_buff[3]  = macb_readl(bp, TSR);
+	regs_buff[4]  = macb_readl(bp, RBQP);
+	regs_buff[5]  = macb_readl(bp, TBQP);
+	regs_buff[6]  = macb_readl(bp, RSR);
+	regs_buff[7]  = macb_readl(bp, IMR);
+
+	regs_buff[8]  = tail;
+	regs_buff[9]  = head;
+	regs_buff[10] = macb_tx_dma(&bp->queues[0], tail);
+	regs_buff[11] = macb_tx_dma(&bp->queues[0], head);
+
+	if (!(bp->caps & MACB_CAPS_USRIO_DISABLED))
+		regs_buff[12] = macb_or_gem_readl(bp, USRIO);
+	if (macb_is_gem(bp))
+		regs_buff[13] = gem_readl(bp, DMACFG);
+}
+
+static void macb_get_wol(struct net_device *netdev, struct ethtool_wolinfo *wol)
+{
+	struct macb *bp = netdev_priv(netdev);
+
+	phylink_ethtool_get_wol(bp->phylink, wol);
+	wol->supported |= (WAKE_MAGIC | WAKE_ARP);
+
+	/* Add macb wolopts to phy wolopts */
+	wol->wolopts |= bp->wolopts;
+}
+
+static int macb_set_wol(struct net_device *netdev, struct ethtool_wolinfo *wol)
+{
+	struct macb *bp = netdev_priv(netdev);
+	int ret;
+
+	/* Pass the order to phylink layer */
+	ret = phylink_ethtool_set_wol(bp->phylink, wol);
+	/* Don't manage WoL on MAC, if PHY set_wol() fails */
+	if (ret && ret != -EOPNOTSUPP)
+		return ret;
+
+	bp->wolopts = (wol->wolopts & WAKE_MAGIC) ? WAKE_MAGIC : 0;
+	bp->wolopts |= (wol->wolopts & WAKE_ARP) ? WAKE_ARP : 0;
+	bp->wol = (wol->wolopts) ? MACB_WOL_ENABLED : 0;
+
+	device_set_wakeup_enable(&bp->pdev->dev, bp->wol);
+
+	return 0;
+}
+
+static int macb_get_link_ksettings(struct net_device *netdev,
+				   struct ethtool_link_ksettings *kset)
+{
+	struct macb *bp = netdev_priv(netdev);
+
+	return phylink_ethtool_ksettings_get(bp->phylink, kset);
+}
+
+static int macb_set_link_ksettings(struct net_device *netdev,
+				   const struct ethtool_link_ksettings *kset)
+{
+	struct macb *bp = netdev_priv(netdev);
+
+	return phylink_ethtool_ksettings_set(bp->phylink, kset);
+}
+
+static void macb_get_ringparam(struct net_device *netdev,
+			       struct ethtool_ringparam *ring,
+			       struct kernel_ethtool_ringparam *kernel_ring,
+			       struct netlink_ext_ack *extack)
+{
+	struct macb *bp = netdev_priv(netdev);
+
+	ring->rx_max_pending = MAX_RX_RING_SIZE;
+	ring->tx_max_pending = MAX_TX_RING_SIZE;
+
+	ring->rx_pending = bp->rx_ring_size;
+	ring->tx_pending = bp->tx_ring_size;
+}
+
+static int macb_set_ringparam(struct net_device *netdev,
+			      struct ethtool_ringparam *ring,
+			      struct kernel_ethtool_ringparam *kernel_ring,
+			      struct netlink_ext_ack *extack)
+{
+	struct macb *bp = netdev_priv(netdev);
+	u32 new_rx_size, new_tx_size;
+	unsigned int reset = 0;
+
+	if ((ring->rx_mini_pending) || (ring->rx_jumbo_pending))
+		return -EINVAL;
+
+	new_rx_size = clamp_t(u32, ring->rx_pending,
+			      MIN_RX_RING_SIZE, MAX_RX_RING_SIZE);
+	new_rx_size = roundup_pow_of_two(new_rx_size);
+
+	new_tx_size = clamp_t(u32, ring->tx_pending,
+			      MIN_TX_RING_SIZE, MAX_TX_RING_SIZE);
+	new_tx_size = roundup_pow_of_two(new_tx_size);
+
+	if ((new_tx_size == bp->tx_ring_size) &&
+	    (new_rx_size == bp->rx_ring_size)) {
+		/* nothing to do */
+		return 0;
+	}
+
+	if (netif_running(bp->dev)) {
+		reset = 1;
+		macb_close(bp->dev);
+	}
+
+	bp->rx_ring_size = new_rx_size;
+	bp->tx_ring_size = new_tx_size;
+
+	if (reset)
+		macb_open(bp->dev);
+
+	return 0;
+}
+
+#ifdef CONFIG_MACB_USE_HWSTAMP
+static unsigned int gem_get_tsu_rate(struct macb *bp)
+{
+	struct clk *tsu_clk;
+	unsigned int tsu_rate;
+
+	tsu_clk = devm_clk_get(&bp->pdev->dev, "tsu_clk");
+	if (!IS_ERR(tsu_clk))
+		tsu_rate = clk_get_rate(tsu_clk);
+	/* try pclk instead */
+	else if (!IS_ERR(bp->pclk)) {
+		tsu_clk = bp->pclk;
+		tsu_rate = clk_get_rate(tsu_clk);
+	} else
+		return -ENOTSUPP;
+	return tsu_rate;
+}
+
+static s32 gem_get_ptp_max_adj(void)
+{
+	return 64000000;
+}
+
+static int gem_get_ts_info(struct net_device *dev,
+			   struct kernel_ethtool_ts_info *info)
+{
+	struct macb *bp = netdev_priv(dev);
+
+	if ((bp->hw_dma_cap & HW_DMA_CAP_PTP) == 0) {
+		ethtool_op_get_ts_info(dev, info);
+		return 0;
+	}
+
+	info->so_timestamping =
+		SOF_TIMESTAMPING_TX_SOFTWARE |
+		SOF_TIMESTAMPING_TX_HARDWARE |
+		SOF_TIMESTAMPING_RX_HARDWARE |
+		SOF_TIMESTAMPING_RAW_HARDWARE;
+	info->tx_types =
+		(1 << HWTSTAMP_TX_ONESTEP_SYNC) |
+		(1 << HWTSTAMP_TX_OFF) |
+		(1 << HWTSTAMP_TX_ON);
+	info->rx_filters =
+		(1 << HWTSTAMP_FILTER_NONE) |
+		(1 << HWTSTAMP_FILTER_ALL);
+
+	if (bp->ptp_clock)
+		info->phc_index = ptp_clock_index(bp->ptp_clock);
+
+	return 0;
+}
+
+static struct macb_ptp_info gem_ptp_info = {
+	.ptp_init	 = gem_ptp_init,
+	.ptp_remove	 = gem_ptp_remove,
+	.get_ptp_max_adj = gem_get_ptp_max_adj,
+	.get_tsu_rate	 = gem_get_tsu_rate,
+	.get_ts_info	 = gem_get_ts_info,
+	.get_hwtst	 = gem_get_hwtst,
+	.set_hwtst	 = gem_set_hwtst,
+};
+#endif
+
+static int macb_get_ts_info(struct net_device *netdev,
+			    struct kernel_ethtool_ts_info *info)
+{
+	struct macb *bp = netdev_priv(netdev);
+
+	if (bp->ptp_info)
+		return bp->ptp_info->get_ts_info(netdev, info);
+
+	return ethtool_op_get_ts_info(netdev, info);
+}
+
+static void gem_enable_flow_filters(struct macb *bp, bool enable)
+{
+	struct net_device *netdev = bp->dev;
+	struct ethtool_rx_fs_item *item;
+	u32 t2_scr;
+	int num_t2_scr;
+
+	if (!(netdev->features & NETIF_F_NTUPLE))
+		return;
+
+	num_t2_scr = GEM_BFEXT(T2SCR, gem_readl(bp, DCFG8));
+
+	list_for_each_entry(item, &bp->rx_fs_list.list, list) {
+		struct ethtool_rx_flow_spec *fs = &item->fs;
+		struct ethtool_tcpip4_spec *tp4sp_m;
+
+		if (fs->location >= num_t2_scr)
+			continue;
+
+		t2_scr = gem_readl_n(bp, SCRT2, fs->location);
+
+		/* enable/disable screener regs for the flow entry */
+		t2_scr = GEM_BFINS(ETHTEN, enable, t2_scr);
+
+		/* only enable fields with no masking */
+		tp4sp_m = &(fs->m_u.tcp_ip4_spec);
+
+		if (enable && (tp4sp_m->ip4src == 0xFFFFFFFF))
+			t2_scr = GEM_BFINS(CMPAEN, 1, t2_scr);
+		else
+			t2_scr = GEM_BFINS(CMPAEN, 0, t2_scr);
+
+		if (enable && (tp4sp_m->ip4dst == 0xFFFFFFFF))
+			t2_scr = GEM_BFINS(CMPBEN, 1, t2_scr);
+		else
+			t2_scr = GEM_BFINS(CMPBEN, 0, t2_scr);
+
+		if (enable && ((tp4sp_m->psrc == 0xFFFF) || (tp4sp_m->pdst == 0xFFFF)))
+			t2_scr = GEM_BFINS(CMPCEN, 1, t2_scr);
+		else
+			t2_scr = GEM_BFINS(CMPCEN, 0, t2_scr);
+
+		gem_writel_n(bp, SCRT2, fs->location, t2_scr);
+	}
+}
+
+static void gem_prog_cmp_regs(struct macb *bp, struct ethtool_rx_flow_spec *fs)
+{
+	struct ethtool_tcpip4_spec *tp4sp_v, *tp4sp_m;
+	uint16_t index = fs->location;
+	u32 w0, w1, t2_scr;
+	bool cmp_a = false;
+	bool cmp_b = false;
+	bool cmp_c = false;
+
+	if (!macb_is_gem(bp))
+		return;
+
+	tp4sp_v = &(fs->h_u.tcp_ip4_spec);
+	tp4sp_m = &(fs->m_u.tcp_ip4_spec);
+
+	/* ignore field if any masking set */
+	if (tp4sp_m->ip4src == 0xFFFFFFFF) {
+		/* 1st compare reg - IP source address */
+		w0 = 0;
+		w1 = 0;
+		w0 = tp4sp_v->ip4src;
+		w1 = GEM_BFINS(T2DISMSK, 1, w1); /* 32-bit compare */
+		w1 = GEM_BFINS(T2CMPOFST, GEM_T2COMPOFST_ETYPE, w1);
+		w1 = GEM_BFINS(T2OFST, ETYPE_SRCIP_OFFSET, w1);
+		gem_writel_n(bp, T2CMPW0, T2CMP_OFST(GEM_IP4SRC_CMP(index)), w0);
+		gem_writel_n(bp, T2CMPW1, T2CMP_OFST(GEM_IP4SRC_CMP(index)), w1);
+		cmp_a = true;
+	}
+
+	/* ignore field if any masking set */
+	if (tp4sp_m->ip4dst == 0xFFFFFFFF) {
+		/* 2nd compare reg - IP destination address */
+		w0 = 0;
+		w1 = 0;
+		w0 = tp4sp_v->ip4dst;
+		w1 = GEM_BFINS(T2DISMSK, 1, w1); /* 32-bit compare */
+		w1 = GEM_BFINS(T2CMPOFST, GEM_T2COMPOFST_ETYPE, w1);
+		w1 = GEM_BFINS(T2OFST, ETYPE_DSTIP_OFFSET, w1);
+		gem_writel_n(bp, T2CMPW0, T2CMP_OFST(GEM_IP4DST_CMP(index)), w0);
+		gem_writel_n(bp, T2CMPW1, T2CMP_OFST(GEM_IP4DST_CMP(index)), w1);
+		cmp_b = true;
+	}
+
+	/* ignore both port fields if masking set in both */
+	if ((tp4sp_m->psrc == 0xFFFF) || (tp4sp_m->pdst == 0xFFFF)) {
+		/* 3rd compare reg - source port, destination port */
+		w0 = 0;
+		w1 = 0;
+		w1 = GEM_BFINS(T2CMPOFST, GEM_T2COMPOFST_IPHDR, w1);
+		if (tp4sp_m->psrc == tp4sp_m->pdst) {
+			w0 = GEM_BFINS(T2MASK, tp4sp_v->psrc, w0);
+			w0 = GEM_BFINS(T2CMP, tp4sp_v->pdst, w0);
+			w1 = GEM_BFINS(T2DISMSK, 1, w1); /* 32-bit compare */
+			w1 = GEM_BFINS(T2OFST, IPHDR_SRCPORT_OFFSET, w1);
+		} else {
+			/* only one port definition */
+			w1 = GEM_BFINS(T2DISMSK, 0, w1); /* 16-bit compare */
+			w0 = GEM_BFINS(T2MASK, 0xFFFF, w0);
+			if (tp4sp_m->psrc == 0xFFFF) { /* src port */
+				w0 = GEM_BFINS(T2CMP, tp4sp_v->psrc, w0);
+				w1 = GEM_BFINS(T2OFST, IPHDR_SRCPORT_OFFSET, w1);
+			} else { /* dst port */
+				w0 = GEM_BFINS(T2CMP, tp4sp_v->pdst, w0);
+				w1 = GEM_BFINS(T2OFST, IPHDR_DSTPORT_OFFSET, w1);
+			}
+		}
+		gem_writel_n(bp, T2CMPW0, T2CMP_OFST(GEM_PORT_CMP(index)), w0);
+		gem_writel_n(bp, T2CMPW1, T2CMP_OFST(GEM_PORT_CMP(index)), w1);
+		cmp_c = true;
+	}
+
+	t2_scr = 0;
+	t2_scr = GEM_BFINS(QUEUE, (fs->ring_cookie) & 0xFF, t2_scr);
+	t2_scr = GEM_BFINS(ETHT2IDX, SCRT2_ETHT, t2_scr);
+	if (cmp_a)
+		t2_scr = GEM_BFINS(CMPA, GEM_IP4SRC_CMP(index), t2_scr);
+	if (cmp_b)
+		t2_scr = GEM_BFINS(CMPB, GEM_IP4DST_CMP(index), t2_scr);
+	if (cmp_c)
+		t2_scr = GEM_BFINS(CMPC, GEM_PORT_CMP(index), t2_scr);
+	gem_writel_n(bp, SCRT2, index, t2_scr);
+}
+
+static int gem_add_flow_filter(struct net_device *netdev,
+		struct ethtool_rxnfc *cmd)
+{
+	struct macb *bp = netdev_priv(netdev);
+	struct ethtool_rx_flow_spec *fs = &cmd->fs;
+	struct ethtool_rx_fs_item *item, *newfs;
+	unsigned long flags;
+	int ret = -EINVAL;
+	bool added = false;
+
+	newfs = kmalloc(sizeof(*newfs), GFP_KERNEL);
+	if (newfs == NULL)
+		return -ENOMEM;
+	memcpy(&newfs->fs, fs, sizeof(newfs->fs));
+
+	netdev_dbg(netdev,
+			"Adding flow filter entry,type=%u,queue=%u,loc=%u,src=%08X,dst=%08X,ps=%u,pd=%u\n",
+			fs->flow_type, (int)fs->ring_cookie, fs->location,
+			htonl(fs->h_u.tcp_ip4_spec.ip4src),
+			htonl(fs->h_u.tcp_ip4_spec.ip4dst),
+			be16_to_cpu(fs->h_u.tcp_ip4_spec.psrc),
+			be16_to_cpu(fs->h_u.tcp_ip4_spec.pdst));
+
+	spin_lock_irqsave(&bp->rx_fs_lock, flags);
+
+	/* find correct place to add in list */
+	list_for_each_entry(item, &bp->rx_fs_list.list, list) {
+		if (item->fs.location > newfs->fs.location) {
+			list_add_tail(&newfs->list, &item->list);
+			added = true;
+			break;
+		} else if (item->fs.location == fs->location) {
+			netdev_err(netdev, "Rule not added: location %d not free!\n",
+					fs->location);
+			ret = -EBUSY;
+			goto err;
+		}
+	}
+	if (!added)
+		list_add_tail(&newfs->list, &bp->rx_fs_list.list);
+
+	gem_prog_cmp_regs(bp, fs);
+	bp->rx_fs_list.count++;
+	/* enable filtering if NTUPLE on */
+	gem_enable_flow_filters(bp, 1);
+
+	spin_unlock_irqrestore(&bp->rx_fs_lock, flags);
+	return 0;
+
+err:
+	spin_unlock_irqrestore(&bp->rx_fs_lock, flags);
+	kfree(newfs);
+	return ret;
+}
+
+static int gem_del_flow_filter(struct net_device *netdev,
+		struct ethtool_rxnfc *cmd)
+{
+	struct macb *bp = netdev_priv(netdev);
+	struct ethtool_rx_fs_item *item;
+	struct ethtool_rx_flow_spec *fs;
+	unsigned long flags;
+
+	spin_lock_irqsave(&bp->rx_fs_lock, flags);
+
+	list_for_each_entry(item, &bp->rx_fs_list.list, list) {
+		if (item->fs.location == cmd->fs.location) {
+			/* disable screener regs for the flow entry */
+			fs = &(item->fs);
+			netdev_dbg(netdev,
+					"Deleting flow filter entry,type=%u,queue=%u,loc=%u,src=%08X,dst=%08X,ps=%u,pd=%u\n",
+					fs->flow_type, (int)fs->ring_cookie, fs->location,
+					htonl(fs->h_u.tcp_ip4_spec.ip4src),
+					htonl(fs->h_u.tcp_ip4_spec.ip4dst),
+					be16_to_cpu(fs->h_u.tcp_ip4_spec.psrc),
+					be16_to_cpu(fs->h_u.tcp_ip4_spec.pdst));
+
+			gem_writel_n(bp, SCRT2, fs->location, 0);
+
+			list_del(&item->list);
+			bp->rx_fs_list.count--;
+			spin_unlock_irqrestore(&bp->rx_fs_lock, flags);
+			kfree(item);
+			return 0;
+		}
+	}
+
+	spin_unlock_irqrestore(&bp->rx_fs_lock, flags);
+	return -EINVAL;
+}
+
+static int gem_get_flow_entry(struct net_device *netdev,
+		struct ethtool_rxnfc *cmd)
+{
+	struct macb *bp = netdev_priv(netdev);
+	struct ethtool_rx_fs_item *item;
+
+	list_for_each_entry(item, &bp->rx_fs_list.list, list) {
+		if (item->fs.location == cmd->fs.location) {
+			memcpy(&cmd->fs, &item->fs, sizeof(cmd->fs));
+			return 0;
+		}
+	}
+	return -EINVAL;
+}
+
+static int gem_get_all_flow_entries(struct net_device *netdev,
+		struct ethtool_rxnfc *cmd, u32 *rule_locs)
+{
+	struct macb *bp = netdev_priv(netdev);
+	struct ethtool_rx_fs_item *item;
+	uint32_t cnt = 0;
+
+	list_for_each_entry(item, &bp->rx_fs_list.list, list) {
+		if (cnt == cmd->rule_cnt)
+			return -EMSGSIZE;
+		rule_locs[cnt] = item->fs.location;
+		cnt++;
+	}
+	cmd->data = bp->max_tuples;
+	cmd->rule_cnt = cnt;
+
+	return 0;
+}
+
+static int gem_get_rxnfc(struct net_device *netdev, struct ethtool_rxnfc *cmd,
+		u32 *rule_locs)
+{
+	struct macb *bp = netdev_priv(netdev);
+	int ret = 0;
+
+	switch (cmd->cmd) {
+	case ETHTOOL_GRXRINGS:
+		cmd->data = bp->num_queues;
+		break;
+	case ETHTOOL_GRXCLSRLCNT:
+		cmd->rule_cnt = bp->rx_fs_list.count;
+		break;
+	case ETHTOOL_GRXCLSRULE:
+		ret = gem_get_flow_entry(netdev, cmd);
+		break;
+	case ETHTOOL_GRXCLSRLALL:
+		ret = gem_get_all_flow_entries(netdev, cmd, rule_locs);
+		break;
+	default:
+		netdev_err(netdev,
+			  "Command parameter %d is not supported\n", cmd->cmd);
+		ret = -EOPNOTSUPP;
+	}
+
+	return ret;
+}
+
+static int gem_set_rxnfc(struct net_device *netdev, struct ethtool_rxnfc *cmd)
+{
+	struct macb *bp = netdev_priv(netdev);
+	int ret;
+
+	switch (cmd->cmd) {
+	case ETHTOOL_SRXCLSRLINS:
+		if ((cmd->fs.location >= bp->max_tuples)
+				|| (cmd->fs.ring_cookie >= bp->num_queues)) {
+			ret = -EINVAL;
+			break;
+		}
+		ret = gem_add_flow_filter(netdev, cmd);
+		break;
+	case ETHTOOL_SRXCLSRLDEL:
+		ret = gem_del_flow_filter(netdev, cmd);
+		break;
+	default:
+		netdev_err(netdev,
+			  "Command parameter %d is not supported\n", cmd->cmd);
+		ret = -EOPNOTSUPP;
+	}
+
+	return ret;
+}
+
+static const struct ethtool_ops macb_ethtool_ops = {
+	.get_regs_len		= macb_get_regs_len,
+	.get_regs		= macb_get_regs,
+	.get_link		= ethtool_op_get_link,
+	.get_ts_info		= ethtool_op_get_ts_info,
+	.get_pause_stats	= macb_get_pause_stats,
+	.get_eth_mac_stats	= macb_get_eth_mac_stats,
+	.get_eth_phy_stats	= macb_get_eth_phy_stats,
+	.get_rmon_stats		= macb_get_rmon_stats,
+	.get_wol		= macb_get_wol,
+	.set_wol		= macb_set_wol,
+	.get_link_ksettings     = macb_get_link_ksettings,
+	.set_link_ksettings     = macb_set_link_ksettings,
+	.get_ringparam		= macb_get_ringparam,
+	.set_ringparam		= macb_set_ringparam,
+};
+
+static const struct ethtool_ops gem_ethtool_ops = {
+	.get_regs_len		= macb_get_regs_len,
+	.get_regs		= macb_get_regs,
+	.get_wol		= macb_get_wol,
+	.set_wol		= macb_set_wol,
+	.get_link		= ethtool_op_get_link,
+	.get_ts_info		= macb_get_ts_info,
+	.get_ethtool_stats	= gem_get_ethtool_stats,
+	.get_strings		= gem_get_ethtool_strings,
+	.get_sset_count		= gem_get_sset_count,
+	.get_pause_stats	= gem_get_pause_stats,
+	.get_eth_mac_stats	= gem_get_eth_mac_stats,
+	.get_eth_phy_stats	= gem_get_eth_phy_stats,
+	.get_rmon_stats		= gem_get_rmon_stats,
+	.get_link_ksettings     = macb_get_link_ksettings,
+	.set_link_ksettings     = macb_set_link_ksettings,
+	.get_ringparam		= macb_get_ringparam,
+	.set_ringparam		= macb_set_ringparam,
+	.get_rxnfc			= gem_get_rxnfc,
+	.set_rxnfc			= gem_set_rxnfc,
+};
+
+static int macb_ioctl(struct net_device *dev, struct ifreq *rq, int cmd)
+{
+	struct macb *bp = netdev_priv(dev);
+
+	if (!netif_running(dev))
+		return -EINVAL;
+
+	return phylink_mii_ioctl(bp->phylink, rq, cmd);
+}
+
+static int macb_hwtstamp_get(struct net_device *dev,
+			     struct kernel_hwtstamp_config *cfg)
+{
+	struct macb *bp = netdev_priv(dev);
+
+	if (!netif_running(dev))
+		return -EINVAL;
+
+	if (!bp->ptp_info)
+		return -EOPNOTSUPP;
+
+	return bp->ptp_info->get_hwtst(dev, cfg);
+}
+
+static int macb_hwtstamp_set(struct net_device *dev,
+			     struct kernel_hwtstamp_config *cfg,
+			     struct netlink_ext_ack *extack)
+{
+	struct macb *bp = netdev_priv(dev);
+
+	if (!netif_running(dev))
+		return -EINVAL;
+
+	if (!bp->ptp_info)
+		return -EOPNOTSUPP;
+
+	return bp->ptp_info->set_hwtst(dev, cfg, extack);
+}
+
+static inline void macb_set_txcsum_feature(struct macb *bp,
+					   netdev_features_t features)
+{
+	u32 val;
+
+	if (!macb_is_gem(bp))
+		return;
+
+	val = gem_readl(bp, DMACFG);
+	if (features & NETIF_F_HW_CSUM)
+		val |= GEM_BIT(TXCOEN);
+	else
+		val &= ~GEM_BIT(TXCOEN);
+
+	gem_writel(bp, DMACFG, val);
+}
+
+static inline void macb_set_rxcsum_feature(struct macb *bp,
+					   netdev_features_t features)
+{
+	struct net_device *netdev = bp->dev;
+	u32 val;
+
+	if (!macb_is_gem(bp))
+		return;
+
+	val = gem_readl(bp, NCFGR);
+	if ((features & NETIF_F_RXCSUM) && !(netdev->flags & IFF_PROMISC))
+		val |= GEM_BIT(RXCOEN);
+	else
+		val &= ~GEM_BIT(RXCOEN);
+
+	gem_writel(bp, NCFGR, val);
+}
+
+static inline void macb_set_rxflow_feature(struct macb *bp,
+					   netdev_features_t features)
+{
+	if (!macb_is_gem(bp))
+		return;
+
+	gem_enable_flow_filters(bp, !!(features & NETIF_F_NTUPLE));
+}
+
+static int macb_set_features(struct net_device *netdev,
+			     netdev_features_t features)
+{
+	struct macb *bp = netdev_priv(netdev);
+	netdev_features_t changed = features ^ netdev->features;
+
+	/* TX checksum offload */
+	if (changed & NETIF_F_HW_CSUM)
+		macb_set_txcsum_feature(bp, features);
+
+	/* RX checksum offload */
+	if (changed & NETIF_F_RXCSUM)
+		macb_set_rxcsum_feature(bp, features);
+
+	/* RX Flow Filters */
+	if (changed & NETIF_F_NTUPLE)
+		macb_set_rxflow_feature(bp, features);
+
+	return 0;
+}
+
+static void macb_restore_features(struct macb *bp)
+{
+	struct net_device *netdev = bp->dev;
+	netdev_features_t features = netdev->features;
+	struct ethtool_rx_fs_item *item;
+
+	/* TX checksum offload */
+	macb_set_txcsum_feature(bp, features);
+
+	/* RX checksum offload */
+	macb_set_rxcsum_feature(bp, features);
+
+	/* RX Flow Filters */
+	list_for_each_entry(item, &bp->rx_fs_list.list, list)
+		gem_prog_cmp_regs(bp, &item->fs);
+
+	macb_set_rxflow_feature(bp, features);
+}
+
+static int macb_taprio_setup_replace(struct net_device *ndev,
+				     struct tc_taprio_qopt_offload *conf)
+{
+	u64 total_on_time = 0, start_time_sec = 0, start_time = conf->base_time;
+	u32 configured_queues = 0, speed = 0, start_time_nsec;
+	struct macb_queue_enst_config *enst_queue;
+	struct tc_taprio_sched_entry *entry;
+	struct macb *bp = netdev_priv(ndev);
+	struct ethtool_link_ksettings kset;
+	struct macb_queue *queue;
+	size_t i;
+	int err;
+
+	if (conf->num_entries > bp->num_queues) {
+		netdev_err(ndev, "Too many TAPRIO entries: %zu > %d queues\n",
+			   conf->num_entries, bp->num_queues);
+		return -EINVAL;
+	}
+
+	if (conf->base_time < 0) {
+		netdev_err(ndev, "Invalid base_time: must be 0 or positive, got %lld\n",
+			   conf->base_time);
+		return -ERANGE;
+	}
+
+	/* Get the current link speed */
+	err = phylink_ethtool_ksettings_get(bp->phylink, &kset);
+	if (unlikely(err)) {
+		netdev_err(ndev, "Failed to get link settings: %d\n", err);
+		return err;
+	}
+
+	speed = kset.base.speed;
+	if (unlikely(speed <= 0)) {
+		netdev_err(ndev, "Invalid speed: %d\n", speed);
+		return -EINVAL;
+	}
+
+	enst_queue = kcalloc(conf->num_entries, sizeof(*enst_queue), GFP_KERNEL);
+	if (unlikely(!enst_queue))
+		return -ENOMEM;
+
+	/* Pre-validate all entries before making any hardware changes */
+	for (i = 0; i < conf->num_entries; i++) {
+		entry = &conf->entries[i];
+
+		if (entry->command != TC_TAPRIO_CMD_SET_GATES) {
+			netdev_err(ndev, "Entry %zu: unsupported command %d\n",
+				   i, entry->command);
+			err = -EOPNOTSUPP;
+			goto cleanup;
+		}
+
+		/* Validate gate_mask: must be nonzero, single queue, and within range */
+		if (!is_power_of_2(entry->gate_mask)) {
+			netdev_err(ndev, "Entry %zu: gate_mask 0x%x is not a power of 2 (only one queue per entry allowed)\n",
+				   i, entry->gate_mask);
+			err = -EINVAL;
+			goto cleanup;
+		}
+
+		/* gate_mask must not select queues outside the valid queue_mask */
+		if (entry->gate_mask & ~bp->queue_mask) {
+			netdev_err(ndev, "Entry %zu: gate_mask 0x%x exceeds queue range (max_queues=%d)\n",
+				   i, entry->gate_mask, bp->num_queues);
+			err = -EINVAL;
+			goto cleanup;
+		}
+
+		/* Check for start time limits */
+		start_time_sec = start_time;
+		start_time_nsec = do_div(start_time_sec, NSEC_PER_SEC);
+		if (start_time_sec > GENMASK(GEM_START_TIME_SEC_SIZE - 1, 0)) {
+			netdev_err(ndev, "Entry %zu: Start time %llu s exceeds hardware limit\n",
+				   i, start_time_sec);
+			err = -ERANGE;
+			goto cleanup;
+		}
+
+		/* Check for on time limit */
+		if (entry->interval > enst_max_hw_interval(speed)) {
+			netdev_err(ndev, "Entry %zu: interval %u ns exceeds hardware limit %llu ns\n",
+				   i, entry->interval, enst_max_hw_interval(speed));
+			err = -ERANGE;
+			goto cleanup;
+		}
+
+		/* Check for off time limit*/
+		if ((conf->cycle_time - entry->interval) > enst_max_hw_interval(speed)) {
+			netdev_err(ndev, "Entry %zu: off_time %llu ns exceeds hardware limit %llu ns\n",
+				   i, conf->cycle_time - entry->interval,
+				   enst_max_hw_interval(speed));
+			err = -ERANGE;
+			goto cleanup;
+		}
+
+		enst_queue[i].queue_id = order_base_2(entry->gate_mask);
+		enst_queue[i].start_time_mask =
+			(start_time_sec << GEM_START_TIME_SEC_OFFSET) |
+			start_time_nsec;
+		enst_queue[i].on_time_bytes =
+			enst_ns_to_hw_units(entry->interval, speed);
+		enst_queue[i].off_time_bytes =
+			enst_ns_to_hw_units(conf->cycle_time - entry->interval, speed);
+
+		configured_queues |= entry->gate_mask;
+		total_on_time += entry->interval;
+		start_time += entry->interval;
+	}
+
+	/* Check total interval doesn't exceed cycle time */
+	if (total_on_time > conf->cycle_time) {
+		netdev_err(ndev, "Total ON %llu ns exceeds cycle time %llu ns\n",
+			   total_on_time, conf->cycle_time);
+		err = -EINVAL;
+		goto cleanup;
+	}
+
+	netdev_dbg(ndev, "TAPRIO setup: %zu entries, base_time=%lld ns, cycle_time=%llu ns\n",
+		   conf->num_entries, conf->base_time, conf->cycle_time);
+
+	/* All validations passed - proceed with hardware configuration */
+	scoped_guard(spinlock_irqsave, &bp->lock) {
+		/* Disable ENST queues if running before configuring */
+		gem_writel(bp, ENST_CONTROL,
+			   bp->queue_mask << GEM_ENST_DISABLE_QUEUE_OFFSET);
+
+		for (i = 0; i < conf->num_entries; i++) {
+			queue = &bp->queues[enst_queue[i].queue_id];
+			/* Configure queue timing registers */
+			queue_writel(queue, ENST_START_TIME,
+				     enst_queue[i].start_time_mask);
+			queue_writel(queue, ENST_ON_TIME,
+				     enst_queue[i].on_time_bytes);
+			queue_writel(queue, ENST_OFF_TIME,
+				     enst_queue[i].off_time_bytes);
+		}
+
+		/* Enable ENST for all configured queues in one write */
+		gem_writel(bp, ENST_CONTROL, configured_queues);
+	}
+
+	netdev_info(ndev, "TAPRIO configuration completed successfully: %zu entries, %d queues configured\n",
+		    conf->num_entries, hweight32(configured_queues));
+
+cleanup:
+	kfree(enst_queue);
+	return err;
+}
+
+static void macb_taprio_destroy(struct net_device *ndev)
+{
+	struct macb *bp = netdev_priv(ndev);
+	struct macb_queue *queue;
+	u32 enst_disable_mask;
+	unsigned int q;
+
+	netdev_reset_tc(ndev);
+	enst_disable_mask = bp->queue_mask << GEM_ENST_DISABLE_QUEUE_OFFSET;
+
+	scoped_guard(spinlock_irqsave, &bp->lock) {
+		/* Single disable command for all queues */
+		gem_writel(bp, ENST_CONTROL, enst_disable_mask);
+
+		/* Clear all queue ENST registers in batch */
+		for (q = 0, queue = bp->queues; q < bp->num_queues; ++q, ++queue) {
+			queue_writel(queue, ENST_START_TIME, 0);
+			queue_writel(queue, ENST_ON_TIME, 0);
+			queue_writel(queue, ENST_OFF_TIME, 0);
+		}
+	}
+	netdev_info(ndev, "TAPRIO destroy: All gates disabled\n");
+}
+
+static int macb_setup_taprio(struct net_device *ndev,
+			     struct tc_taprio_qopt_offload *taprio)
+{
+	struct macb *bp = netdev_priv(ndev);
+	int err = 0;
+
+	if (unlikely(!(ndev->hw_features & NETIF_F_HW_TC)))
+		return -EOPNOTSUPP;
+
+	/* Check if Device is in runtime suspend */
+	if (unlikely(pm_runtime_suspended(&bp->pdev->dev))) {
+		netdev_err(ndev, "Device is in runtime suspend\n");
+		return -EOPNOTSUPP;
+	}
+
+	switch (taprio->cmd) {
+	case TAPRIO_CMD_REPLACE:
+		err = macb_taprio_setup_replace(ndev, taprio);
+		break;
+	case TAPRIO_CMD_DESTROY:
+		macb_taprio_destroy(ndev);
+		break;
+	default:
+		err = -EOPNOTSUPP;
+	}
+
+	return err;
+}
+
+static int macb_setup_tc(struct net_device *dev, enum tc_setup_type type,
+			 void *type_data)
+{
+	if (!dev || !type_data)
+		return -EINVAL;
+
+	switch (type) {
+	case TC_SETUP_QDISC_TAPRIO:
+		return macb_setup_taprio(dev, type_data);
+	default:
+		return -EOPNOTSUPP;
+	}
+}
+
+static const struct net_device_ops macb_netdev_ops = {
+	.ndo_open		= macb_open,
+	.ndo_stop		= macb_close,
+	.ndo_start_xmit		= macb_start_xmit,
+	.ndo_set_rx_mode	= macb_set_rx_mode,
+	.ndo_get_stats64	= macb_get_stats,
+	.ndo_eth_ioctl		= macb_ioctl,
+	.ndo_validate_addr	= eth_validate_addr,
+	.ndo_change_mtu		= macb_change_mtu,
+	.ndo_set_mac_address	= macb_set_mac_addr,
+#ifdef CONFIG_NET_POLL_CONTROLLER
+	.ndo_poll_controller	= macb_poll_controller,
+#endif
+	.ndo_set_features	= macb_set_features,
+	.ndo_features_check	= macb_features_check,
+	.ndo_hwtstamp_set	= macb_hwtstamp_set,
+	.ndo_hwtstamp_get	= macb_hwtstamp_get,
+	.ndo_setup_tc		= macb_setup_tc,
+};
+
+/* Configure peripheral capabilities according to device tree
+ * and integration options used
+ */
+static void macb_configure_caps(struct macb *bp,
+				const struct macb_config *dt_conf)
+{
+	struct device_node *np = bp->pdev->dev.of_node;
+	bool refclk_ext;
+	u32 dcfg;
+
+	refclk_ext = of_property_read_bool(np, "cdns,refclk-ext");
+
+	if (dt_conf)
+		bp->caps = dt_conf->caps;
+
+	if (hw_is_gem(bp->regs, bp->native_io)) {
+		bp->caps |= MACB_CAPS_MACB_IS_GEM;
+
+		dcfg = gem_readl(bp, DCFG1);
+		if (GEM_BFEXT(IRQCOR, dcfg) == 0)
+			bp->caps |= MACB_CAPS_ISR_CLEAR_ON_WRITE;
+		if (GEM_BFEXT(NO_PCS, dcfg) == 0)
+			bp->caps |= MACB_CAPS_PCS;
+		dcfg = gem_readl(bp, DCFG12);
+		if (GEM_BFEXT(HIGH_SPEED, dcfg) == 1)
+			bp->caps |= MACB_CAPS_HIGH_SPEED;
+		dcfg = gem_readl(bp, DCFG2);
+		if ((dcfg & (GEM_BIT(RX_PKT_BUFF) | GEM_BIT(TX_PKT_BUFF))) == 0)
+			bp->caps |= MACB_CAPS_FIFO_MODE;
+		if (gem_has_ptp(bp)) {
+			if (!GEM_BFEXT(TSU, gem_readl(bp, DCFG5)))
+				dev_err(&bp->pdev->dev,
+					"GEM doesn't support hardware ptp.\n");
+			else {
+#ifdef CONFIG_MACB_USE_HWSTAMP
+				bp->hw_dma_cap |= HW_DMA_CAP_PTP;
+				bp->ptp_info = &gem_ptp_info;
+#endif
+			}
+		}
+	}
+
+	if (refclk_ext)
+		bp->caps |= MACB_CAPS_USRIO_HAS_CLKEN;
+
+	dev_dbg(&bp->pdev->dev, "Cadence caps 0x%08x\n", bp->caps);
+}
+
+static void macb_probe_queues(void __iomem *mem,
+			      bool native_io,
+			      unsigned int *queue_mask,
+			      unsigned int *num_queues)
+{
+	*queue_mask = 0x1;
+	*num_queues = 1;
+
+	/* is it macb or gem ?
+	 *
+	 * We need to read directly from the hardware here because
+	 * we are early in the probe process and don't have the
+	 * MACB_CAPS_MACB_IS_GEM flag positioned
+	 */
+	if (!hw_is_gem(mem, native_io))
+		return;
+
+	/* bit 0 is never set but queue 0 always exists */
+	*queue_mask |= readl_relaxed(mem + GEM_DCFG6) & 0xff;
+	*num_queues = hweight32(*queue_mask);
+}
+
+static void macb_clks_disable(struct clk *pclk, struct clk *hclk, struct clk *tx_clk,
+			      struct clk *rx_clk, struct clk *tsu_clk)
+{
+	struct clk_bulk_data clks[] = {
+		{ .clk = tsu_clk, },
+		{ .clk = rx_clk, },
+		{ .clk = pclk, },
+		{ .clk = hclk, },
+		{ .clk = tx_clk },
+	};
+
+	clk_bulk_disable_unprepare(ARRAY_SIZE(clks), clks);
+}
+
+static int macb_clk_init(struct platform_device *pdev, struct clk **pclk,
+			 struct clk **hclk, struct clk **tx_clk,
+			 struct clk **rx_clk, struct clk **tsu_clk)
+{
+	struct macb_platform_data *pdata;
+	int err;
+
+	pdata = dev_get_platdata(&pdev->dev);
+	if (pdata) {
+		*pclk = pdata->pclk;
+		*hclk = pdata->hclk;
+	} else {
+		*pclk = devm_clk_get(&pdev->dev, "pclk");
+		*hclk = devm_clk_get(&pdev->dev, "hclk");
+	}
+
+	if (IS_ERR_OR_NULL(*pclk))
+		return dev_err_probe(&pdev->dev,
+				     IS_ERR(*pclk) ? PTR_ERR(*pclk) : -ENODEV,
+				     "failed to get pclk\n");
+
+	if (IS_ERR_OR_NULL(*hclk))
+		return dev_err_probe(&pdev->dev,
+				     IS_ERR(*hclk) ? PTR_ERR(*hclk) : -ENODEV,
+				     "failed to get hclk\n");
+
+	*tx_clk = devm_clk_get_optional(&pdev->dev, "tx_clk");
+	if (IS_ERR(*tx_clk))
+		return PTR_ERR(*tx_clk);
+
+	*rx_clk = devm_clk_get_optional(&pdev->dev, "rx_clk");
+	if (IS_ERR(*rx_clk))
+		return PTR_ERR(*rx_clk);
+
+	*tsu_clk = devm_clk_get_optional(&pdev->dev, "tsu_clk");
+	if (IS_ERR(*tsu_clk))
+		return PTR_ERR(*tsu_clk);
+
+	err = clk_prepare_enable(*pclk);
+	if (err) {
+		dev_err(&pdev->dev, "failed to enable pclk (%d)\n", err);
+		return err;
+	}
+
+	err = clk_prepare_enable(*hclk);
+	if (err) {
+		dev_err(&pdev->dev, "failed to enable hclk (%d)\n", err);
+		goto err_disable_pclk;
+	}
+
+	err = clk_prepare_enable(*tx_clk);
+	if (err) {
+		dev_err(&pdev->dev, "failed to enable tx_clk (%d)\n", err);
+		goto err_disable_hclk;
+	}
+
+	err = clk_prepare_enable(*rx_clk);
+	if (err) {
+		dev_err(&pdev->dev, "failed to enable rx_clk (%d)\n", err);
+		goto err_disable_txclk;
+	}
+
+	err = clk_prepare_enable(*tsu_clk);
+	if (err) {
+		dev_err(&pdev->dev, "failed to enable tsu_clk (%d)\n", err);
+		goto err_disable_rxclk;
+	}
+
+	return 0;
+
+err_disable_rxclk:
+	clk_disable_unprepare(*rx_clk);
+
+err_disable_txclk:
+	clk_disable_unprepare(*tx_clk);
+
+err_disable_hclk:
+	clk_disable_unprepare(*hclk);
+
+err_disable_pclk:
+	clk_disable_unprepare(*pclk);
+
+	return err;
+}
+
+static int macb_init(struct platform_device *pdev)
+{
+	struct net_device *dev = platform_get_drvdata(pdev);
+	unsigned int hw_q, q;
+	struct macb *bp = netdev_priv(dev);
+	struct macb_queue *queue;
+	int err;
+	u32 val, reg;
+
+	bp->tx_ring_size = DEFAULT_TX_RING_SIZE;
+	bp->rx_ring_size = DEFAULT_RX_RING_SIZE;
+
+	/* set the queue register mapping once for all: queue0 has a special
+	 * register mapping but we don't want to test the queue index then
+	 * compute the corresponding register offset at run time.
+	 */
+	for (hw_q = 0, q = 0; hw_q < MACB_MAX_QUEUES; ++hw_q) {
+		if (!(bp->queue_mask & (1 << hw_q)))
+			continue;
+
+		queue = &bp->queues[q];
+		queue->bp = bp;
+		spin_lock_init(&queue->tx_ptr_lock);
+		netif_napi_add(dev, &queue->napi_rx, macb_rx_poll);
+		netif_napi_add(dev, &queue->napi_tx, macb_tx_poll);
+		if (hw_q) {
+			queue->ISR  = GEM_ISR(hw_q - 1);
+			queue->IER  = GEM_IER(hw_q - 1);
+			queue->IDR  = GEM_IDR(hw_q - 1);
+			queue->IMR  = GEM_IMR(hw_q - 1);
+			queue->TBQP = GEM_TBQP(hw_q - 1);
+			queue->RBQP = GEM_RBQP(hw_q - 1);
+			queue->RBQS = GEM_RBQS(hw_q - 1);
+		} else {
+			/* queue0 uses legacy registers */
+			queue->ISR  = MACB_ISR;
+			queue->IER  = MACB_IER;
+			queue->IDR  = MACB_IDR;
+			queue->IMR  = MACB_IMR;
+			queue->TBQP = MACB_TBQP;
+			queue->RBQP = MACB_RBQP;
+		}
+
+		queue->ENST_START_TIME = GEM_ENST_START_TIME(hw_q);
+		queue->ENST_ON_TIME = GEM_ENST_ON_TIME(hw_q);
+		queue->ENST_OFF_TIME = GEM_ENST_OFF_TIME(hw_q);
+
+		/* get irq: here we use the linux queue index, not the hardware
+		 * queue index. the queue irq definitions in the device tree
+		 * must remove the optional gaps that could exist in the
+		 * hardware queue mask.
+		 */
+		queue->irq = platform_get_irq(pdev, q);
+		err = devm_request_irq(&pdev->dev, queue->irq, macb_interrupt,
+				       IRQF_SHARED, dev->name, queue);
+		if (err) {
+			dev_err(&pdev->dev,
+				"Unable to request IRQ %d (error %d)\n",
+				queue->irq, err);
+			return err;
+		}
+
+		INIT_WORK(&queue->tx_error_task, macb_tx_error_task);
+		q++;
+	}
+
+	dev->netdev_ops = &macb_netdev_ops;
+
+	/* setup appropriated routines according to adapter type */
+	if (macb_is_gem(bp)) {
+		bp->macbgem_ops.mog_alloc_rx_buffers = gem_alloc_rx_buffers;
+		bp->macbgem_ops.mog_free_rx_buffers = gem_free_rx_buffers;
+		bp->macbgem_ops.mog_init_rings = gem_init_rings;
+		bp->macbgem_ops.mog_rx = gem_rx;
+		dev->ethtool_ops = &gem_ethtool_ops;
+	} else {
+		bp->macbgem_ops.mog_alloc_rx_buffers = macb_alloc_rx_buffers;
+		bp->macbgem_ops.mog_free_rx_buffers = macb_free_rx_buffers;
+		bp->macbgem_ops.mog_init_rings = macb_init_rings;
+		bp->macbgem_ops.mog_rx = macb_rx;
+		dev->ethtool_ops = &macb_ethtool_ops;
+	}
+
+	netdev_sw_irq_coalesce_default_on(dev);
+
+	dev->priv_flags |= IFF_LIVE_ADDR_CHANGE;
+
+	/* Set features */
+	dev->hw_features = NETIF_F_SG;
+
+	/* Check LSO capability */
+	if (GEM_BFEXT(PBUF_LSO, gem_readl(bp, DCFG6)))
+		dev->hw_features |= MACB_NETIF_LSO;
+
+	/* Checksum offload is only available on gem with packet buffer */
+	if (macb_is_gem(bp) && !(bp->caps & MACB_CAPS_FIFO_MODE))
+		dev->hw_features |= NETIF_F_HW_CSUM | NETIF_F_RXCSUM;
+	if (bp->caps & MACB_CAPS_SG_DISABLED)
+		dev->hw_features &= ~NETIF_F_SG;
+	/* Enable HW_TC if hardware supports QBV */
+	if (bp->caps & MACB_CAPS_QBV)
+		dev->hw_features |= NETIF_F_HW_TC;
+
+	dev->features = dev->hw_features;
+
+	/* Check RX Flow Filters support.
+	 * Max Rx flows set by availability of screeners & compare regs:
+	 * each 4-tuple define requires 1 T2 screener reg + 3 compare regs
+	 */
+	reg = gem_readl(bp, DCFG8);
+	bp->max_tuples = min((GEM_BFEXT(SCR2CMP, reg) / 3),
+			GEM_BFEXT(T2SCR, reg));
+	INIT_LIST_HEAD(&bp->rx_fs_list.list);
+	if (bp->max_tuples > 0) {
+		/* also needs one ethtype match to check IPv4 */
+		if (GEM_BFEXT(SCR2ETH, reg) > 0) {
+			/* program this reg now */
+			reg = 0;
+			reg = GEM_BFINS(ETHTCMP, (uint16_t)ETH_P_IP, reg);
+			gem_writel_n(bp, ETHT, SCRT2_ETHT, reg);
+			/* Filtering is supported in hw but don't enable it in kernel now */
+			dev->hw_features |= NETIF_F_NTUPLE;
+			/* init Rx flow definitions */
+			bp->rx_fs_list.count = 0;
+			spin_lock_init(&bp->rx_fs_lock);
+		} else
+			bp->max_tuples = 0;
+	}
+
+	if (!(bp->caps & MACB_CAPS_USRIO_DISABLED)) {
+		val = 0;
+		if (phy_interface_mode_is_rgmii(bp->phy_interface))
+			val = bp->usrio->rgmii;
+		else if (bp->phy_interface == PHY_INTERFACE_MODE_RMII &&
+			 (bp->caps & MACB_CAPS_USRIO_DEFAULT_IS_MII_GMII))
+			val = bp->usrio->rmii;
+		else if (!(bp->caps & MACB_CAPS_USRIO_DEFAULT_IS_MII_GMII))
+			val = bp->usrio->mii;
+
+		if (bp->caps & MACB_CAPS_USRIO_HAS_CLKEN)
+			val |= bp->usrio->refclk;
+
+		macb_or_gem_writel(bp, USRIO, val);
+	}
+
+	/* Set MII management clock divider */
+	val = macb_mdc_clk_div(bp);
+	val |= macb_dbw(bp);
+	if (bp->phy_interface == PHY_INTERFACE_MODE_SGMII)
+		val |= GEM_BIT(SGMIIEN) | GEM_BIT(PCSSEL);
+	macb_writel(bp, NCFGR, val);
+
+	return 0;
+}
+
+static const struct macb_usrio_config macb_default_usrio = {
+	.mii = MACB_BIT(MII),
+	.rmii = MACB_BIT(RMII),
+	.rgmii = GEM_BIT(RGMII),
+	.refclk = MACB_BIT(CLKEN),
+};
+
+#if defined(CONFIG_OF)
+/* 1518 rounded up */
+#define AT91ETHER_MAX_RBUFF_SZ	0x600
+/* max number of receive buffers */
+#define AT91ETHER_MAX_RX_DESCR	9
+
+static struct sifive_fu540_macb_mgmt *mgmt;
+
+static int at91ether_alloc_coherent(struct macb *lp)
+{
+	struct macb_queue *q = &lp->queues[0];
+
+	q->rx_ring = dma_alloc_coherent(&lp->pdev->dev,
+					 (AT91ETHER_MAX_RX_DESCR *
+					  macb_dma_desc_get_size(lp)),
+					 &q->rx_ring_dma, GFP_KERNEL);
+	if (!q->rx_ring)
+		return -ENOMEM;
+
+	q->rx_buffers = dma_alloc_coherent(&lp->pdev->dev,
+					    AT91ETHER_MAX_RX_DESCR *
+					    AT91ETHER_MAX_RBUFF_SZ,
+					    &q->rx_buffers_dma, GFP_KERNEL);
+	if (!q->rx_buffers) {
+		dma_free_coherent(&lp->pdev->dev,
+				  AT91ETHER_MAX_RX_DESCR *
+				  macb_dma_desc_get_size(lp),
+				  q->rx_ring, q->rx_ring_dma);
+		q->rx_ring = NULL;
+		return -ENOMEM;
+	}
+
+	return 0;
+}
+
+static void at91ether_free_coherent(struct macb *lp)
+{
+	struct macb_queue *q = &lp->queues[0];
+
+	if (q->rx_ring) {
+		dma_free_coherent(&lp->pdev->dev,
+				  AT91ETHER_MAX_RX_DESCR *
+				  macb_dma_desc_get_size(lp),
+				  q->rx_ring, q->rx_ring_dma);
+		q->rx_ring = NULL;
+	}
+
+	if (q->rx_buffers) {
+		dma_free_coherent(&lp->pdev->dev,
+				  AT91ETHER_MAX_RX_DESCR *
+				  AT91ETHER_MAX_RBUFF_SZ,
+				  q->rx_buffers, q->rx_buffers_dma);
+		q->rx_buffers = NULL;
+	}
+}
+
+/* Initialize and start the Receiver and Transmit subsystems */
+static int at91ether_start(struct macb *lp)
+{
+	struct macb_queue *q = &lp->queues[0];
+	struct macb_dma_desc *desc;
+	dma_addr_t addr;
+	u32 ctl;
+	int i, ret;
+
+	ret = at91ether_alloc_coherent(lp);
+	if (ret)
+		return ret;
+
+	addr = q->rx_buffers_dma;
+	for (i = 0; i < AT91ETHER_MAX_RX_DESCR; i++) {
+		desc = macb_rx_desc(q, i);
+		macb_set_addr(lp, desc, addr);
+		desc->ctrl = 0;
+		addr += AT91ETHER_MAX_RBUFF_SZ;
+	}
+
+	/* Set the Wrap bit on the last descriptor */
+	desc->addr |= MACB_BIT(RX_WRAP);
+
+	/* Reset buffer index */
+	q->rx_tail = 0;
+
+	/* Program address of descriptor list in Rx Buffer Queue register */
+	macb_writel(lp, RBQP, q->rx_ring_dma);
+
+	/* Enable Receive and Transmit */
+	ctl = macb_readl(lp, NCR);
+	macb_writel(lp, NCR, ctl | MACB_BIT(RE) | MACB_BIT(TE));
+
+	/* Enable MAC interrupts */
+	macb_writel(lp, IER, MACB_BIT(RCOMP)	|
+			     MACB_BIT(RXUBR)	|
+			     MACB_BIT(ISR_TUND)	|
+			     MACB_BIT(ISR_RLE)	|
+			     MACB_BIT(TCOMP)	|
+			     MACB_BIT(ISR_ROVR)	|
+			     MACB_BIT(HRESP));
+
+	return 0;
+}
+
+static void at91ether_stop(struct macb *lp)
+{
+	u32 ctl;
+
+	/* Disable MAC interrupts */
+	macb_writel(lp, IDR, MACB_BIT(RCOMP)	|
+			     MACB_BIT(RXUBR)	|
+			     MACB_BIT(ISR_TUND)	|
+			     MACB_BIT(ISR_RLE)	|
+			     MACB_BIT(TCOMP)	|
+			     MACB_BIT(ISR_ROVR) |
+			     MACB_BIT(HRESP));
+
+	/* Disable Receiver and Transmitter */
+	ctl = macb_readl(lp, NCR);
+	macb_writel(lp, NCR, ctl & ~(MACB_BIT(TE) | MACB_BIT(RE)));
+
+	/* Free resources. */
+	at91ether_free_coherent(lp);
+}
+
+/* Open the ethernet interface */
+static int at91ether_open(struct net_device *dev)
+{
+	struct macb *lp = netdev_priv(dev);
+	u32 ctl;
+	int ret;
+
+	ret = pm_runtime_resume_and_get(&lp->pdev->dev);
+	if (ret < 0)
+		return ret;
+
+	/* Clear internal statistics */
+	ctl = macb_readl(lp, NCR);
+	macb_writel(lp, NCR, ctl | MACB_BIT(CLRSTAT));
+
+	macb_set_hwaddr(lp);
+
+	ret = at91ether_start(lp);
+	if (ret)
+		goto pm_exit;
+
+	ret = macb_phylink_connect(lp);
+	if (ret)
+		goto stop;
+
+	netif_start_queue(dev);
+
+	return 0;
+
+stop:
+	at91ether_stop(lp);
+pm_exit:
+	pm_runtime_put_sync(&lp->pdev->dev);
+	return ret;
+}
+
+/* Close the interface */
+static int at91ether_close(struct net_device *dev)
+{
+	struct macb *lp = netdev_priv(dev);
+
+	netif_stop_queue(dev);
+
+	phylink_stop(lp->phylink);
+	phylink_disconnect_phy(lp->phylink);
+
+	at91ether_stop(lp);
+
+	return pm_runtime_put(&lp->pdev->dev);
+}
+
+/* Transmit packet */
+static netdev_tx_t at91ether_start_xmit(struct sk_buff *skb,
+					struct net_device *dev)
+{
+	struct macb *lp = netdev_priv(dev);
+
+	if (macb_readl(lp, TSR) & MACB_BIT(RM9200_BNQ)) {
+		int desc = 0;
+
+		netif_stop_queue(dev);
+
+		/* Store packet information (to free when Tx completed) */
+		lp->rm9200_txq[desc].skb = skb;
+		lp->rm9200_txq[desc].size = skb->len;
+		lp->rm9200_txq[desc].mapping = dma_map_single(&lp->pdev->dev, skb->data,
+							      skb->len, DMA_TO_DEVICE);
+		if (dma_mapping_error(&lp->pdev->dev, lp->rm9200_txq[desc].mapping)) {
+			dev_kfree_skb_any(skb);
+			dev->stats.tx_dropped++;
+			netdev_err(dev, "%s: DMA mapping error\n", __func__);
+			return NETDEV_TX_OK;
+		}
+
+		/* Set address of the data in the Transmit Address register */
+		macb_writel(lp, TAR, lp->rm9200_txq[desc].mapping);
+		/* Set length of the packet in the Transmit Control register */
+		macb_writel(lp, TCR, skb->len);
+
+	} else {
+		netdev_err(dev, "%s called, but device is busy!\n", __func__);
+		return NETDEV_TX_BUSY;
+	}
+
+	return NETDEV_TX_OK;
+}
+
+/* Extract received frame from buffer descriptors and sent to upper layers.
+ * (Called from interrupt context)
+ */
+static void at91ether_rx(struct net_device *dev)
+{
+	struct macb *lp = netdev_priv(dev);
+	struct macb_queue *q = &lp->queues[0];
+	struct macb_dma_desc *desc;
+	unsigned char *p_recv;
+	struct sk_buff *skb;
+	unsigned int pktlen;
+
+	desc = macb_rx_desc(q, q->rx_tail);
+	while (desc->addr & MACB_BIT(RX_USED)) {
+		p_recv = q->rx_buffers + q->rx_tail * AT91ETHER_MAX_RBUFF_SZ;
+		pktlen = MACB_BF(RX_FRMLEN, desc->ctrl);
+		skb = netdev_alloc_skb(dev, pktlen + 2);
+		if (skb) {
+			skb_reserve(skb, 2);
+			skb_put_data(skb, p_recv, pktlen);
+
+			skb->protocol = eth_type_trans(skb, dev);
+			dev->stats.rx_packets++;
+			dev->stats.rx_bytes += pktlen;
+			netif_rx(skb);
+		} else {
+			dev->stats.rx_dropped++;
+		}
+
+		if (desc->ctrl & MACB_BIT(RX_MHASH_MATCH))
+			dev->stats.multicast++;
+
+		/* reset ownership bit */
+		desc->addr &= ~MACB_BIT(RX_USED);
+
+		/* wrap after last buffer */
+		if (q->rx_tail == AT91ETHER_MAX_RX_DESCR - 1)
+			q->rx_tail = 0;
+		else
+			q->rx_tail++;
+
+		desc = macb_rx_desc(q, q->rx_tail);
+	}
+}
+
+/* MAC interrupt handler */
+static irqreturn_t at91ether_interrupt(int irq, void *dev_id)
+{
+	struct net_device *dev = dev_id;
+	struct macb *lp = netdev_priv(dev);
+	u32 intstatus, ctl;
+	unsigned int desc;
+
+	/* MAC Interrupt Status register indicates what interrupts are pending.
+	 * It is automatically cleared once read.
+	 */
+	intstatus = macb_readl(lp, ISR);
+
+	/* Receive complete */
+	if (intstatus & MACB_BIT(RCOMP))
+		at91ether_rx(dev);
+
+	/* Transmit complete */
+	if (intstatus & MACB_BIT(TCOMP)) {
+		/* The TCOM bit is set even if the transmission failed */
+		if (intstatus & (MACB_BIT(ISR_TUND) | MACB_BIT(ISR_RLE)))
+			dev->stats.tx_errors++;
+
+		desc = 0;
+		if (lp->rm9200_txq[desc].skb) {
+			dev_consume_skb_irq(lp->rm9200_txq[desc].skb);
+			lp->rm9200_txq[desc].skb = NULL;
+			dma_unmap_single(&lp->pdev->dev, lp->rm9200_txq[desc].mapping,
+					 lp->rm9200_txq[desc].size, DMA_TO_DEVICE);
+			dev->stats.tx_packets++;
+			dev->stats.tx_bytes += lp->rm9200_txq[desc].size;
+		}
+		netif_wake_queue(dev);
+	}
+
+	/* Work-around for EMAC Errata section 41.3.1 */
+	if (intstatus & MACB_BIT(RXUBR)) {
+		ctl = macb_readl(lp, NCR);
+		macb_writel(lp, NCR, ctl & ~MACB_BIT(RE));
+		wmb();
+		macb_writel(lp, NCR, ctl | MACB_BIT(RE));
+	}
+
+	if (intstatus & MACB_BIT(ISR_ROVR))
+		netdev_err(dev, "ROVR error\n");
+
+	return IRQ_HANDLED;
+}
+
+#ifdef CONFIG_NET_POLL_CONTROLLER
+static void at91ether_poll_controller(struct net_device *dev)
+{
+	unsigned long flags;
+
+	local_irq_save(flags);
+	at91ether_interrupt(dev->irq, dev);
+	local_irq_restore(flags);
+}
+#endif
+
+static const struct net_device_ops at91ether_netdev_ops = {
+	.ndo_open		= at91ether_open,
+	.ndo_stop		= at91ether_close,
+	.ndo_start_xmit		= at91ether_start_xmit,
+	.ndo_get_stats64	= macb_get_stats,
+	.ndo_set_rx_mode	= macb_set_rx_mode,
+	.ndo_set_mac_address	= eth_mac_addr,
+	.ndo_eth_ioctl		= macb_ioctl,
+	.ndo_validate_addr	= eth_validate_addr,
+#ifdef CONFIG_NET_POLL_CONTROLLER
+	.ndo_poll_controller	= at91ether_poll_controller,
+#endif
+	.ndo_hwtstamp_set	= macb_hwtstamp_set,
+	.ndo_hwtstamp_get	= macb_hwtstamp_get,
+};
+
+static int at91ether_clk_init(struct platform_device *pdev, struct clk **pclk,
+			      struct clk **hclk, struct clk **tx_clk,
+			      struct clk **rx_clk, struct clk **tsu_clk)
+{
+	int err;
+
+	*hclk = NULL;
+	*tx_clk = NULL;
+	*rx_clk = NULL;
+	*tsu_clk = NULL;
+
+	*pclk = devm_clk_get(&pdev->dev, "ether_clk");
+	if (IS_ERR(*pclk))
+		return PTR_ERR(*pclk);
+
+	err = clk_prepare_enable(*pclk);
+	if (err) {
+		dev_err(&pdev->dev, "failed to enable pclk (%d)\n", err);
+		return err;
+	}
+
+	return 0;
+}
+
+static int at91ether_init(struct platform_device *pdev)
+{
+	struct net_device *dev = platform_get_drvdata(pdev);
+	struct macb *bp = netdev_priv(dev);
+	int err;
+
+	bp->queues[0].bp = bp;
+
+	dev->netdev_ops = &at91ether_netdev_ops;
+	dev->ethtool_ops = &macb_ethtool_ops;
+
+	err = devm_request_irq(&pdev->dev, dev->irq, at91ether_interrupt,
+			       0, dev->name, dev);
+	if (err)
+		return err;
+
+	macb_writel(bp, NCR, 0);
+
+	macb_writel(bp, NCFGR, MACB_BF(CLK, MACB_CLK_DIV32) | MACB_BIT(BIG));
+
+	return 0;
+}
+
+static unsigned long fu540_macb_tx_recalc_rate(struct clk_hw *hw,
+					       unsigned long parent_rate)
+{
+	return mgmt->rate;
+}
+
+static int fu540_macb_tx_determine_rate(struct clk_hw *hw,
+					struct clk_rate_request *req)
+{
+	if (WARN_ON(req->rate < 2500000))
+		req->rate = 2500000;
+	else if (req->rate == 2500000)
+		req->rate = 2500000;
+	else if (WARN_ON(req->rate < 13750000))
+		req->rate = 2500000;
+	else if (WARN_ON(req->rate < 25000000))
+		req->rate = 25000000;
+	else if (req->rate == 25000000)
+		req->rate = 25000000;
+	else if (WARN_ON(req->rate < 75000000))
+		req->rate = 25000000;
+	else if (WARN_ON(req->rate < 125000000))
+		req->rate = 125000000;
+	else if (req->rate == 125000000)
+		req->rate = 125000000;
+	else if (WARN_ON(req->rate > 125000000))
+		req->rate = 125000000;
+	else
+		req->rate = 125000000;
+
+	return 0;
+}
+
+static int fu540_macb_tx_set_rate(struct clk_hw *hw, unsigned long rate,
+				  unsigned long parent_rate)
+{
+	struct clk_rate_request req;
+	int ret;
+
+	clk_hw_init_rate_request(hw, &req, rate);
+	ret = fu540_macb_tx_determine_rate(hw, &req);
+	if (ret != 0)
+		return ret;
+
+	if (req.rate != 125000000)
+		iowrite32(1, mgmt->reg);
+	else
+		iowrite32(0, mgmt->reg);
+	mgmt->rate = rate;
+
+	return 0;
+}
+
+static const struct clk_ops fu540_c000_ops = {
+	.recalc_rate = fu540_macb_tx_recalc_rate,
+	.determine_rate = fu540_macb_tx_determine_rate,
+	.set_rate = fu540_macb_tx_set_rate,
+};
+
+static int fu540_c000_clk_init(struct platform_device *pdev, struct clk **pclk,
+			       struct clk **hclk, struct clk **tx_clk,
+			       struct clk **rx_clk, struct clk **tsu_clk)
+{
+	struct clk_init_data init;
+	int err = 0;
+
+	err = macb_clk_init(pdev, pclk, hclk, tx_clk, rx_clk, tsu_clk);
+	if (err)
+		return err;
+
+	mgmt = devm_kzalloc(&pdev->dev, sizeof(*mgmt), GFP_KERNEL);
+	if (!mgmt) {
+		err = -ENOMEM;
+		goto err_disable_clks;
+	}
+
+	init.name = "sifive-gemgxl-mgmt";
+	init.ops = &fu540_c000_ops;
+	init.flags = 0;
+	init.num_parents = 0;
+
+	mgmt->rate = 0;
+	mgmt->hw.init = &init;
+
+	*tx_clk = devm_clk_register(&pdev->dev, &mgmt->hw);
+	if (IS_ERR(*tx_clk)) {
+		err = PTR_ERR(*tx_clk);
+		goto err_disable_clks;
+	}
+
+	err = clk_prepare_enable(*tx_clk);
+	if (err) {
+		dev_err(&pdev->dev, "failed to enable tx_clk (%u)\n", err);
+		*tx_clk = NULL;
+		goto err_disable_clks;
+	} else {
+		dev_info(&pdev->dev, "Registered clk switch '%s'\n", init.name);
+	}
+
+	return 0;
+
+err_disable_clks:
+	macb_clks_disable(*pclk, *hclk, *tx_clk, *rx_clk, *tsu_clk);
+
+	return err;
+}
+
+static int fu540_c000_init(struct platform_device *pdev)
+{
+	mgmt->reg = devm_platform_ioremap_resource(pdev, 1);
+	if (IS_ERR(mgmt->reg))
+		return PTR_ERR(mgmt->reg);
+
+	return macb_init(pdev);
+}
+
+static int init_reset_optional(struct platform_device *pdev)
+{
+	struct net_device *dev = platform_get_drvdata(pdev);
+	struct macb *bp = netdev_priv(dev);
+	int ret;
+
+	if (bp->phy_interface == PHY_INTERFACE_MODE_SGMII) {
+		/* Ensure PHY device used in SGMII mode is ready */
+		bp->sgmii_phy = devm_phy_optional_get(&pdev->dev, NULL);
+
+		if (IS_ERR(bp->sgmii_phy))
+			return dev_err_probe(&pdev->dev, PTR_ERR(bp->sgmii_phy),
+					     "failed to get SGMII PHY\n");
+
+		ret = phy_init(bp->sgmii_phy);
+		if (ret)
+			return dev_err_probe(&pdev->dev, ret,
+					     "failed to init SGMII PHY\n");
+
+		ret = zynqmp_pm_is_function_supported(PM_IOCTL, IOCTL_SET_GEM_CONFIG);
+		if (!ret) {
+			u32 pm_info[2];
+
+			ret = of_property_read_u32_array(pdev->dev.of_node, "power-domains",
+							 pm_info, ARRAY_SIZE(pm_info));
+			if (ret) {
+				dev_err(&pdev->dev, "Failed to read power management information\n");
+				goto err_out_phy_exit;
+			}
+			ret = zynqmp_pm_set_gem_config(pm_info[1], GEM_CONFIG_FIXED, 0);
+			if (ret)
+				goto err_out_phy_exit;
+
+			ret = zynqmp_pm_set_gem_config(pm_info[1], GEM_CONFIG_SGMII_MODE, 1);
+			if (ret)
+				goto err_out_phy_exit;
+		}
+
+	}
+
+	/* Fully reset controller at hardware level if mapped in device tree */
+	ret = device_reset_optional(&pdev->dev);
+	if (ret) {
+		phy_exit(bp->sgmii_phy);
+		return dev_err_probe(&pdev->dev, ret, "failed to reset controller");
+	}
+
+	ret = macb_init(pdev);
+
+err_out_phy_exit:
+	if (ret)
+		phy_exit(bp->sgmii_phy);
+
+	return ret;
+}
+
+static const struct macb_usrio_config sama7g5_usrio = {
+	.mii = 0,
+	.rmii = 1,
+	.rgmii = 2,
+	.refclk = BIT(2),
+	.hdfctlen = BIT(6),
+};
+
+static const struct macb_config fu540_c000_config = {
+	.caps = MACB_CAPS_GIGABIT_MODE_AVAILABLE | MACB_CAPS_JUMBO |
+		MACB_CAPS_GEM_HAS_PTP,
+	.dma_burst_length = 16,
+	.clk_init = fu540_c000_clk_init,
+	.init = fu540_c000_init,
+	.jumbo_max_len = 10240,
+	.usrio = &macb_default_usrio,
+};
+
+static const struct macb_config at91sam9260_config = {
+	.caps = MACB_CAPS_USRIO_HAS_CLKEN | MACB_CAPS_USRIO_DEFAULT_IS_MII_GMII,
+	.clk_init = macb_clk_init,
+	.init = macb_init,
+	.usrio = &macb_default_usrio,
+};
+
+static const struct macb_config sama5d3macb_config = {
+	.caps = MACB_CAPS_SG_DISABLED |
+		MACB_CAPS_USRIO_HAS_CLKEN | MACB_CAPS_USRIO_DEFAULT_IS_MII_GMII,
+	.clk_init = macb_clk_init,
+	.init = macb_init,
+	.usrio = &macb_default_usrio,
+};
+
+static const struct macb_config pc302gem_config = {
+	.caps = MACB_CAPS_SG_DISABLED | MACB_CAPS_GIGABIT_MODE_AVAILABLE,
+	.dma_burst_length = 16,
+	.clk_init = macb_clk_init,
+	.init = macb_init,
+	.usrio = &macb_default_usrio,
+};
+
+static const struct macb_config sama5d2_config = {
+	.caps = MACB_CAPS_USRIO_DEFAULT_IS_MII_GMII | MACB_CAPS_JUMBO,
+	.dma_burst_length = 16,
+	.clk_init = macb_clk_init,
+	.init = macb_init,
+	.jumbo_max_len = 10240,
+	.usrio = &macb_default_usrio,
+};
+
+static const struct macb_config sama5d29_config = {
+	.caps = MACB_CAPS_USRIO_DEFAULT_IS_MII_GMII | MACB_CAPS_GEM_HAS_PTP,
+	.dma_burst_length = 16,
+	.clk_init = macb_clk_init,
+	.init = macb_init,
+	.usrio = &macb_default_usrio,
+};
+
+static const struct macb_config sama5d3_config = {
+	.caps = MACB_CAPS_SG_DISABLED | MACB_CAPS_GIGABIT_MODE_AVAILABLE |
+		MACB_CAPS_USRIO_DEFAULT_IS_MII_GMII | MACB_CAPS_JUMBO,
+	.dma_burst_length = 16,
+	.clk_init = macb_clk_init,
+	.init = macb_init,
+	.jumbo_max_len = 10240,
+	.usrio = &macb_default_usrio,
+};
+
+static const struct macb_config sama5d4_config = {
+	.caps = MACB_CAPS_USRIO_DEFAULT_IS_MII_GMII,
+	.dma_burst_length = 4,
+	.clk_init = macb_clk_init,
+	.init = macb_init,
+	.usrio = &macb_default_usrio,
+};
+
+static const struct macb_config emac_config = {
+	.caps = MACB_CAPS_NEEDS_RSTONUBR | MACB_CAPS_MACB_IS_EMAC,
+	.clk_init = at91ether_clk_init,
+	.init = at91ether_init,
+	.usrio = &macb_default_usrio,
+};
+
+static const struct macb_config np4_config = {
+	.caps = MACB_CAPS_USRIO_DISABLED,
+	.clk_init = macb_clk_init,
+	.init = macb_init,
+	.usrio = &macb_default_usrio,
+};
+
+static const struct macb_config zynqmp_config = {
+	.caps = MACB_CAPS_GIGABIT_MODE_AVAILABLE |
+		MACB_CAPS_JUMBO |
+		MACB_CAPS_GEM_HAS_PTP | MACB_CAPS_BD_RD_PREFETCH,
+	.dma_burst_length = 16,
+	.clk_init = macb_clk_init,
+	.init = init_reset_optional,
+	.jumbo_max_len = 10240,
+	.usrio = &macb_default_usrio,
+};
+
+static const struct macb_config zynq_config = {
+	.caps = MACB_CAPS_GIGABIT_MODE_AVAILABLE | MACB_CAPS_NO_GIGABIT_HALF |
+		MACB_CAPS_NEEDS_RSTONUBR,
+	.dma_burst_length = 16,
+	.clk_init = macb_clk_init,
+	.init = macb_init,
+	.usrio = &macb_default_usrio,
+};
+
+static const struct macb_config mpfs_config = {
+	.caps = MACB_CAPS_GIGABIT_MODE_AVAILABLE |
+		MACB_CAPS_JUMBO |
+		MACB_CAPS_GEM_HAS_PTP,
+	.dma_burst_length = 16,
+	.clk_init = macb_clk_init,
+	.init = init_reset_optional,
+	.usrio = &macb_default_usrio,
+	.max_tx_length = 4040, /* Cadence Erratum 1686 */
+	.jumbo_max_len = 4040,
+};
+
+static const struct macb_config sama7g5_gem_config = {
+	.caps = MACB_CAPS_GIGABIT_MODE_AVAILABLE | MACB_CAPS_CLK_HW_CHG |
+		MACB_CAPS_USRIO_DEFAULT_IS_MII_GMII |
+		MACB_CAPS_MIIONRGMII | MACB_CAPS_GEM_HAS_PTP,
+	.dma_burst_length = 16,
+	.clk_init = macb_clk_init,
+	.init = macb_init,
+	.usrio = &sama7g5_usrio,
+};
+
+static const struct macb_config sama7g5_emac_config = {
+	.caps = MACB_CAPS_USRIO_DEFAULT_IS_MII_GMII |
+		MACB_CAPS_USRIO_HAS_CLKEN | MACB_CAPS_MIIONRGMII |
+		MACB_CAPS_GEM_HAS_PTP,
+	.dma_burst_length = 16,
+	.clk_init = macb_clk_init,
+	.init = macb_init,
+	.usrio = &sama7g5_usrio,
+};
+
+static const struct macb_config versal_config = {
+	.caps = MACB_CAPS_GIGABIT_MODE_AVAILABLE | MACB_CAPS_JUMBO |
+		MACB_CAPS_GEM_HAS_PTP | MACB_CAPS_BD_RD_PREFETCH |
+		MACB_CAPS_NEED_TSUCLK | MACB_CAPS_QUEUE_DISABLE |
+		MACB_CAPS_QBV,
+	.dma_burst_length = 16,
+	.clk_init = macb_clk_init,
+	.init = init_reset_optional,
+	.jumbo_max_len = 10240,
+	.usrio = &macb_default_usrio,
+};
+
+static const struct macb_config raspberrypi_rp1_config = {
+	.caps = MACB_CAPS_GIGABIT_MODE_AVAILABLE | MACB_CAPS_CLK_HW_CHG |
+		MACB_CAPS_JUMBO |
+		MACB_CAPS_GEM_HAS_PTP,
+	.dma_burst_length = 16,
+	.clk_init = macb_clk_init,
+	.init = macb_init,
+	.usrio = &macb_default_usrio,
+	.jumbo_max_len = 10240,
+};
+
+static const struct of_device_id macb_dt_ids[] = {
+	{ .compatible = "cdns,at91sam9260-macb", .data = &at91sam9260_config },
+	{ .compatible = "cdns,macb" },
+	{ .compatible = "cdns,np4-macb", .data = &np4_config },
+	{ .compatible = "cdns,pc302-gem", .data = &pc302gem_config },
+	{ .compatible = "cdns,gem", .data = &pc302gem_config },
+	{ .compatible = "cdns,sam9x60-macb", .data = &at91sam9260_config },
+	{ .compatible = "atmel,sama5d2-gem", .data = &sama5d2_config },
+	{ .compatible = "atmel,sama5d29-gem", .data = &sama5d29_config },
+	{ .compatible = "atmel,sama5d3-gem", .data = &sama5d3_config },
+	{ .compatible = "atmel,sama5d3-macb", .data = &sama5d3macb_config },
+	{ .compatible = "atmel,sama5d4-gem", .data = &sama5d4_config },
+	{ .compatible = "cdns,at91rm9200-emac", .data = &emac_config },
+	{ .compatible = "cdns,emac", .data = &emac_config },
+	{ .compatible = "cdns,zynqmp-gem", .data = &zynqmp_config}, /* deprecated */
+	{ .compatible = "cdns,zynq-gem", .data = &zynq_config }, /* deprecated */
+	{ .compatible = "sifive,fu540-c000-gem", .data = &fu540_c000_config },
+	{ .compatible = "microchip,mpfs-macb", .data = &mpfs_config },
+	{ .compatible = "microchip,sama7g5-gem", .data = &sama7g5_gem_config },
+	{ .compatible = "microchip,sama7g5-emac", .data = &sama7g5_emac_config },
+	{ .compatible = "raspberrypi,rp1-gem", .data = &raspberrypi_rp1_config },
+	{ .compatible = "xlnx,zynqmp-gem", .data = &zynqmp_config},
+	{ .compatible = "xlnx,zynq-gem", .data = &zynq_config },
+	{ .compatible = "xlnx,versal-gem", .data = &versal_config},
+	{ /* sentinel */ }
+};
+MODULE_DEVICE_TABLE(of, macb_dt_ids);
+#endif /* CONFIG_OF */
+
+static const struct macb_config default_gem_config = {
+	.caps = MACB_CAPS_GIGABIT_MODE_AVAILABLE |
+		MACB_CAPS_JUMBO |
+		MACB_CAPS_GEM_HAS_PTP,
+	.dma_burst_length = 16,
+	.clk_init = macb_clk_init,
+	.init = macb_init,
+	.usrio = &macb_default_usrio,
+	.jumbo_max_len = 10240,
+};
+
+static int macb_probe(struct platform_device *pdev)
+{
+	const struct macb_config *macb_config = &default_gem_config;
+	int (*clk_init)(struct platform_device *, struct clk **,
+			struct clk **, struct clk **,  struct clk **,
+			struct clk **) = macb_config->clk_init;
+	int (*init)(struct platform_device *) = macb_config->init;
+	struct device_node *np = pdev->dev.of_node;
+	struct clk *pclk, *hclk = NULL, *tx_clk = NULL, *rx_clk = NULL;
+	struct clk *tsu_clk = NULL;
+	unsigned int queue_mask, num_queues;
+	bool native_io;
+	phy_interface_t interface;
+	struct net_device *dev;
+	struct resource *regs;
+	u32 wtrmrk_rst_val;
+	void __iomem *mem;
+	struct macb *bp;
+	int err, val;
+
+	mem = devm_platform_get_and_ioremap_resource(pdev, 0, &regs);
+	if (IS_ERR(mem))
+		return PTR_ERR(mem);
+
+	if (np) {
+		const struct of_device_id *match;
+
+		match = of_match_node(macb_dt_ids, np);
+		if (match && match->data) {
+			macb_config = match->data;
+			clk_init = macb_config->clk_init;
+			init = macb_config->init;
+		}
+	}
+
+	err = clk_init(pdev, &pclk, &hclk, &tx_clk, &rx_clk, &tsu_clk);
+	if (err)
+		return err;
+
+	pm_runtime_set_autosuspend_delay(&pdev->dev, MACB_PM_TIMEOUT);
+	pm_runtime_use_autosuspend(&pdev->dev);
+	pm_runtime_get_noresume(&pdev->dev);
+	pm_runtime_set_active(&pdev->dev);
+	pm_runtime_enable(&pdev->dev);
+	native_io = hw_is_native_io(mem);
+
+	macb_probe_queues(mem, native_io, &queue_mask, &num_queues);
+	dev = alloc_etherdev_mq(sizeof(*bp), num_queues);
+	if (!dev) {
+		err = -ENOMEM;
+		goto err_disable_clocks;
+	}
+
+	dev->base_addr = regs->start;
+
+	SET_NETDEV_DEV(dev, &pdev->dev);
+
+	bp = netdev_priv(dev);
+	bp->pdev = pdev;
+	bp->dev = dev;
+	bp->regs = mem;
+	bp->native_io = native_io;
+	if (native_io) {
+		bp->macb_reg_readl = hw_readl_native;
+		bp->macb_reg_writel = hw_writel_native;
+	} else {
+		bp->macb_reg_readl = hw_readl;
+		bp->macb_reg_writel = hw_writel;
+	}
+	bp->num_queues = num_queues;
+	bp->queue_mask = queue_mask;
+	if (macb_config)
+		bp->dma_burst_length = macb_config->dma_burst_length;
+	bp->pclk = pclk;
+	bp->hclk = hclk;
+	bp->tx_clk = tx_clk;
+	bp->rx_clk = rx_clk;
+	bp->tsu_clk = tsu_clk;
+	if (macb_config)
+		bp->jumbo_max_len = macb_config->jumbo_max_len;
+
+	if (!hw_is_gem(bp->regs, bp->native_io))
+		bp->max_tx_length = MACB_MAX_TX_LEN;
+	else if (macb_config->max_tx_length)
+		bp->max_tx_length = macb_config->max_tx_length;
+	else
+		bp->max_tx_length = GEM_MAX_TX_LEN;
+
+	bp->wol = 0;
+	device_set_wakeup_capable(&pdev->dev, 1);
+
+	bp->usrio = macb_config->usrio;
+
+	/* By default we set to partial store and forward mode for zynqmp.
+	 * Disable if not set in devicetree.
+	 */
+	if (GEM_BFEXT(PBUF_CUTTHRU, gem_readl(bp, DCFG6))) {
+		err = of_property_read_u32(bp->pdev->dev.of_node,
+					   "cdns,rx-watermark",
+					   &bp->rx_watermark);
+
+		if (!err) {
+			/* Disable partial store and forward in case of error or
+			 * invalid watermark value
+			 */
+			wtrmrk_rst_val = (1 << (GEM_BFEXT(RX_PBUF_ADDR, gem_readl(bp, DCFG2)))) - 1;
+			if (bp->rx_watermark > wtrmrk_rst_val || !bp->rx_watermark) {
+				dev_info(&bp->pdev->dev, "Invalid watermark value\n");
+				bp->rx_watermark = 0;
+			}
+		}
+	}
+	spin_lock_init(&bp->lock);
+	spin_lock_init(&bp->stats_lock);
+
+	/* setup capabilities */
+	macb_configure_caps(bp, macb_config);
+
+#ifdef CONFIG_ARCH_DMA_ADDR_T_64BIT
+	if (GEM_BFEXT(DAW64, gem_readl(bp, DCFG6))) {
+		err = dma_set_mask_and_coherent(&pdev->dev, DMA_BIT_MASK(44));
+		if (err) {
+			dev_err(&pdev->dev, "failed to set DMA mask\n");
+			goto err_out_free_netdev;
+		}
+		bp->hw_dma_cap |= HW_DMA_CAP_64B;
+	}
+#endif
+	platform_set_drvdata(pdev, dev);
+
+	dev->irq = platform_get_irq(pdev, 0);
+	if (dev->irq < 0) {
+		err = dev->irq;
+		goto err_out_free_netdev;
+	}
+
+	/* MTU range: 68 - 1518 or 10240 */
+	dev->min_mtu = GEM_MTU_MIN_SIZE;
+	if ((bp->caps & MACB_CAPS_JUMBO) && bp->jumbo_max_len)
+		dev->max_mtu = bp->jumbo_max_len - ETH_HLEN - ETH_FCS_LEN;
+	else
+		dev->max_mtu = 1536 - ETH_HLEN - ETH_FCS_LEN;
+
+	if (bp->caps & MACB_CAPS_BD_RD_PREFETCH) {
+		val = GEM_BFEXT(RXBD_RDBUFF, gem_readl(bp, DCFG10));
+		if (val)
+			bp->rx_bd_rd_prefetch = (2 << (val - 1)) *
+						macb_dma_desc_get_size(bp);
+
+		val = GEM_BFEXT(TXBD_RDBUFF, gem_readl(bp, DCFG10));
+		if (val)
+			bp->tx_bd_rd_prefetch = (2 << (val - 1)) *
+						macb_dma_desc_get_size(bp);
+	}
+
+	bp->rx_intr_mask = MACB_RX_INT_FLAGS;
+	if (bp->caps & MACB_CAPS_NEEDS_RSTONUBR)
+		bp->rx_intr_mask |= MACB_BIT(RXUBR);
+
+	err = of_get_ethdev_address(np, bp->dev);
+	if (err == -EPROBE_DEFER)
+		goto err_out_free_netdev;
+	else if (err)
+		macb_get_hwaddr(bp);
+
+	err = of_get_phy_mode(np, &interface);
+	if (err)
+		/* not found in DT, MII by default */
+		bp->phy_interface = PHY_INTERFACE_MODE_MII;
+	else
+		bp->phy_interface = interface;
+
+	/* IP specific init */
+	err = init(pdev);
+	if (err)
+		goto err_out_free_netdev;
+
+	err = macb_mii_init(bp);
+	if (err)
+		goto err_out_phy_exit;
+
+	netif_carrier_off(dev);
+
+	err = register_netdev(dev);
+	if (err) {
+		dev_err(&pdev->dev, "Cannot register net device, aborting.\n");
+		goto err_out_unregister_mdio;
+	}
+
+	INIT_WORK(&bp->hresp_err_bh_work, macb_hresp_error_task);
+
+	netdev_info(dev, "Cadence %s rev 0x%08x at 0x%08lx irq %d (%pM)\n",
+		    macb_is_gem(bp) ? "GEM" : "MACB", macb_readl(bp, MID),
+		    dev->base_addr, dev->irq, dev->dev_addr);
+
+	pm_runtime_mark_last_busy(&bp->pdev->dev);
+	pm_runtime_put_autosuspend(&bp->pdev->dev);
+
+	return 0;
+
+err_out_unregister_mdio:
+	mdiobus_unregister(bp->mii_bus);
+	mdiobus_free(bp->mii_bus);
+
+err_out_phy_exit:
+	phy_exit(bp->sgmii_phy);
+
+err_out_free_netdev:
+	free_netdev(dev);
+
+err_disable_clocks:
+	macb_clks_disable(pclk, hclk, tx_clk, rx_clk, tsu_clk);
+	pm_runtime_disable(&pdev->dev);
+	pm_runtime_set_suspended(&pdev->dev);
+	pm_runtime_dont_use_autosuspend(&pdev->dev);
+
+	return err;
+}
+
+static void macb_remove(struct platform_device *pdev)
+{
+	struct net_device *dev;
+	struct macb *bp;
+
+	dev = platform_get_drvdata(pdev);
+
+	if (dev) {
+		bp = netdev_priv(dev);
+		unregister_netdev(dev);
+		phy_exit(bp->sgmii_phy);
+		mdiobus_unregister(bp->mii_bus);
+		mdiobus_free(bp->mii_bus);
+
+		device_set_wakeup_enable(&bp->pdev->dev, 0);
+		cancel_work_sync(&bp->hresp_err_bh_work);
+		pm_runtime_disable(&pdev->dev);
+		pm_runtime_dont_use_autosuspend(&pdev->dev);
+		pm_runtime_set_suspended(&pdev->dev);
+		phylink_destroy(bp->phylink);
+		free_netdev(dev);
+	}
+}
+
+static int __maybe_unused macb_suspend(struct device *dev)
+{
+	struct net_device *netdev = dev_get_drvdata(dev);
+	struct macb *bp = netdev_priv(netdev);
+	struct in_ifaddr *ifa = NULL;
+	struct macb_queue *queue;
+	struct in_device *idev;
+	unsigned long flags;
+	unsigned int q;
+	int err;
+	u32 tmp;
+
+	if (!device_may_wakeup(&bp->dev->dev))
+		phy_exit(bp->sgmii_phy);
+
+	if (!netif_running(netdev))
+		return 0;
+
+	if (bp->wol & MACB_WOL_ENABLED) {
+		/* Check for IP address in WOL ARP mode */
+		idev = __in_dev_get_rcu(bp->dev);
+		if (idev)
+			ifa = rcu_dereference(idev->ifa_list);
+		if ((bp->wolopts & WAKE_ARP) && !ifa) {
+			netdev_err(netdev, "IP address not assigned as required by WoL walk ARP\n");
+			return -EOPNOTSUPP;
+		}
+		spin_lock_irqsave(&bp->lock, flags);
+
+		/* Disable Tx and Rx engines before  disabling the queues,
+		 * this is mandatory as per the IP spec sheet
+		 */
+		tmp = macb_readl(bp, NCR);
+		macb_writel(bp, NCR, tmp & ~(MACB_BIT(TE) | MACB_BIT(RE)));
+#ifdef CONFIG_ARCH_DMA_ADDR_T_64BIT
+		if (!(bp->caps & MACB_CAPS_QUEUE_DISABLE))
+			macb_writel(bp, RBQPH,
+				    upper_32_bits(bp->rx_ring_tieoff_dma));
+#endif
+		for (q = 0, queue = bp->queues; q < bp->num_queues;
+		     ++q, ++queue) {
+			/* Disable RX queues */
+			if (bp->caps & MACB_CAPS_QUEUE_DISABLE) {
+				queue_writel(queue, RBQP, MACB_BIT(QUEUE_DISABLE));
+			} else {
+				/* Tie off RX queues */
+				queue_writel(queue, RBQP,
+					     lower_32_bits(bp->rx_ring_tieoff_dma));
+			}
+			/* Disable all interrupts */
+			queue_writel(queue, IDR, -1);
+			queue_readl(queue, ISR);
+			if (bp->caps & MACB_CAPS_ISR_CLEAR_ON_WRITE)
+				queue_writel(queue, ISR, -1);
+		}
+		/* Enable Receive engine */
+		macb_writel(bp, NCR, tmp | MACB_BIT(RE));
+		/* Flush all status bits */
+		macb_writel(bp, TSR, -1);
+		macb_writel(bp, RSR, -1);
+
+		tmp = (bp->wolopts & WAKE_MAGIC) ? MACB_BIT(MAG) : 0;
+		if (bp->wolopts & WAKE_ARP) {
+			tmp |= MACB_BIT(ARP);
+			/* write IP address into register */
+			tmp |= MACB_BFEXT(IP, be32_to_cpu(ifa->ifa_local));
+		}
+
+		/* Change interrupt handler and
+		 * Enable WoL IRQ on queue 0
+		 */
+		devm_free_irq(dev, bp->queues[0].irq, bp->queues);
+		if (macb_is_gem(bp)) {
+			err = devm_request_irq(dev, bp->queues[0].irq, gem_wol_interrupt,
+					       IRQF_SHARED, netdev->name, bp->queues);
+			if (err) {
+				dev_err(dev,
+					"Unable to request IRQ %d (error %d)\n",
+					bp->queues[0].irq, err);
+				spin_unlock_irqrestore(&bp->lock, flags);
+				return err;
+			}
+			queue_writel(bp->queues, IER, GEM_BIT(WOL));
+			gem_writel(bp, WOL, tmp);
+		} else {
+			err = devm_request_irq(dev, bp->queues[0].irq, macb_wol_interrupt,
+					       IRQF_SHARED, netdev->name, bp->queues);
+			if (err) {
+				dev_err(dev,
+					"Unable to request IRQ %d (error %d)\n",
+					bp->queues[0].irq, err);
+				spin_unlock_irqrestore(&bp->lock, flags);
+				return err;
+			}
+			queue_writel(bp->queues, IER, MACB_BIT(WOL));
+			macb_writel(bp, WOL, tmp);
+		}
+		spin_unlock_irqrestore(&bp->lock, flags);
+
+		enable_irq_wake(bp->queues[0].irq);
+	}
+
+	netif_device_detach(netdev);
+	for (q = 0, queue = bp->queues; q < bp->num_queues;
+	     ++q, ++queue) {
+		napi_disable(&queue->napi_rx);
+		napi_disable(&queue->napi_tx);
+	}
+
+	if (!(bp->wol & MACB_WOL_ENABLED)) {
+		rtnl_lock();
+		phylink_stop(bp->phylink);
+		rtnl_unlock();
+		spin_lock_irqsave(&bp->lock, flags);
+		macb_reset_hw(bp);
+		spin_unlock_irqrestore(&bp->lock, flags);
+	}
+
+	if (!(bp->caps & MACB_CAPS_USRIO_DISABLED))
+		bp->pm_data.usrio = macb_or_gem_readl(bp, USRIO);
+
+	if (netdev->hw_features & NETIF_F_NTUPLE)
+		bp->pm_data.scrt2 = gem_readl_n(bp, ETHT, SCRT2_ETHT);
+
+	if (bp->ptp_info)
+		bp->ptp_info->ptp_remove(netdev);
+	if (!device_may_wakeup(dev))
+		pm_runtime_force_suspend(dev);
+
+	return 0;
+}
+
+static int __maybe_unused macb_resume(struct device *dev)
+{
+	struct net_device *netdev = dev_get_drvdata(dev);
+	struct macb *bp = netdev_priv(netdev);
+	struct macb_queue *queue;
+	unsigned long flags;
+	unsigned int q;
+	int err;
+
+	if (!device_may_wakeup(&bp->dev->dev))
+		phy_init(bp->sgmii_phy);
+
+	if (!netif_running(netdev))
+		return 0;
+
+	if (!device_may_wakeup(dev))
+		pm_runtime_force_resume(dev);
+
+	if (bp->wol & MACB_WOL_ENABLED) {
+		spin_lock_irqsave(&bp->lock, flags);
+		/* Disable WoL */
+		if (macb_is_gem(bp)) {
+			queue_writel(bp->queues, IDR, GEM_BIT(WOL));
+			gem_writel(bp, WOL, 0);
+		} else {
+			queue_writel(bp->queues, IDR, MACB_BIT(WOL));
+			macb_writel(bp, WOL, 0);
+		}
+		/* Clear ISR on queue 0 */
+		queue_readl(bp->queues, ISR);
+		if (bp->caps & MACB_CAPS_ISR_CLEAR_ON_WRITE)
+			queue_writel(bp->queues, ISR, -1);
+		/* Replace interrupt handler on queue 0 */
+		devm_free_irq(dev, bp->queues[0].irq, bp->queues);
+		err = devm_request_irq(dev, bp->queues[0].irq, macb_interrupt,
+				       IRQF_SHARED, netdev->name, bp->queues);
+		if (err) {
+			dev_err(dev,
+				"Unable to request IRQ %d (error %d)\n",
+				bp->queues[0].irq, err);
+			spin_unlock_irqrestore(&bp->lock, flags);
+			return err;
+		}
+		spin_unlock_irqrestore(&bp->lock, flags);
+
+		disable_irq_wake(bp->queues[0].irq);
+
+		/* Now make sure we disable phy before moving
+		 * to common restore path
+		 */
+		rtnl_lock();
+		phylink_stop(bp->phylink);
+		rtnl_unlock();
+	}
+
+	for (q = 0, queue = bp->queues; q < bp->num_queues;
+	     ++q, ++queue) {
+		napi_enable(&queue->napi_rx);
+		napi_enable(&queue->napi_tx);
+	}
+
+	if (netdev->hw_features & NETIF_F_NTUPLE)
+		gem_writel_n(bp, ETHT, SCRT2_ETHT, bp->pm_data.scrt2);
+
+	if (!(bp->caps & MACB_CAPS_USRIO_DISABLED))
+		macb_or_gem_writel(bp, USRIO, bp->pm_data.usrio);
+
+	macb_writel(bp, NCR, MACB_BIT(MPE));
+	macb_init_hw(bp);
+	macb_set_rx_mode(netdev);
+	macb_restore_features(bp);
+	rtnl_lock();
+
+	phylink_start(bp->phylink);
+	rtnl_unlock();
+
+	netif_device_attach(netdev);
+	if (bp->ptp_info)
+		bp->ptp_info->ptp_init(netdev);
+
+	return 0;
+}
+
+static int __maybe_unused macb_runtime_suspend(struct device *dev)
+{
+	struct net_device *netdev = dev_get_drvdata(dev);
+	struct macb *bp = netdev_priv(netdev);
+
+	if (!(device_may_wakeup(dev)))
+		macb_clks_disable(bp->pclk, bp->hclk, bp->tx_clk, bp->rx_clk, bp->tsu_clk);
+	else if (!(bp->caps & MACB_CAPS_NEED_TSUCLK))
+		macb_clks_disable(NULL, NULL, NULL, NULL, bp->tsu_clk);
+
+	return 0;
+}
+
+static int __maybe_unused macb_runtime_resume(struct device *dev)
+{
+	struct net_device *netdev = dev_get_drvdata(dev);
+	struct macb *bp = netdev_priv(netdev);
+
+	if (!(device_may_wakeup(dev))) {
+		clk_prepare_enable(bp->pclk);
+		clk_prepare_enable(bp->hclk);
+		clk_prepare_enable(bp->tx_clk);
+		clk_prepare_enable(bp->rx_clk);
+		clk_prepare_enable(bp->tsu_clk);
+	} else if (!(bp->caps & MACB_CAPS_NEED_TSUCLK)) {
+		clk_prepare_enable(bp->tsu_clk);
+	}
+
+	return 0;
+}
+
+static void macb_shutdown(struct platform_device *pdev)
+{
+	struct net_device *netdev = platform_get_drvdata(pdev);
+
+	rtnl_lock();
+
+	if (netif_running(netdev))
+		dev_close(netdev);
+
+	netif_device_detach(netdev);
+
+	rtnl_unlock();
+}
+
+static const struct dev_pm_ops macb_pm_ops = {
+	SET_SYSTEM_SLEEP_PM_OPS(macb_suspend, macb_resume)
+	SET_RUNTIME_PM_OPS(macb_runtime_suspend, macb_runtime_resume, NULL)
+};
+
+static struct platform_driver macb_driver = {
+	.probe		= macb_probe,
+	.remove		= macb_remove,
+	.driver		= {
+		.name		= "macb",
+		.of_match_table	= of_match_ptr(macb_dt_ids),
+		.pm	= &macb_pm_ops,
+	},
+	.shutdown	= macb_shutdown,
+};
+
+module_platform_driver(macb_driver);
+
+MODULE_LICENSE("GPL");
+MODULE_DESCRIPTION("Cadence MACB/GEM Ethernet driver");
+MODULE_AUTHOR("Haavard Skinnemoen (Atmel)");
+MODULE_ALIAS("platform:macb");
diff --git a/drivers/net/ethernet/cadence/macb_pci.c b/drivers/net/ethernet/cadence/macb_pci.c
index 9906fda76087..fc4f5aee6ab3 100644
--- a/drivers/net/ethernet/cadence/macb_pci.c
+++ b/drivers/net/ethernet/cadence/macb_pci.c
@@ -1,22 +1,11 @@
-/**
- * Cadence GEM PCI wrapper.
+// SPDX-License-Identifier: GPL-2.0-only
+/*
+ * DOC: Cadence GEM PCI wrapper.
  *
- * Copyright (C) 2016 Cadence Design Systems - http://www.cadence.com
+ * Copyright (C) 2016 Cadence Design Systems - https://www.cadence.com
  *
  * Authors: Rafal Ozieblo <rafalo@cadence.com>
  *	    Bartosz Folta <bfolta@cadence.com>
- *
- * This program is free software: you can redistribute it and/or modify
- * it under the terms of the GNU General Public License version 2  of
- * the License as published by the Free Software Foundation.
- *
- * This program is distributed in the hope that it will be useful,
- * but WITHOUT ANY WARRANTY; without even the implied warranty of
- * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
- * GNU General Public License for more details.
- *
- * You should have received a copy of the GNU General Public License
- * along with this program.  If not, see <http://www.gnu.org/licenses/>.
  */
 
 #include <linux/clk.h>
@@ -24,15 +13,13 @@
 #include <linux/etherdevice.h>
 #include <linux/module.h>
 #include <linux/pci.h>
-#include <linux/platform_data/macb.h>
 #include <linux/platform_device.h>
 #include "macb.h"
 
 #define PCI_DRIVER_NAME "macb_pci"
 #define PLAT_DRIVER_NAME "macb"
 
-#define CDNS_VENDOR_ID 0x17cd
-#define CDNS_DEVICE_ID 0xe007
+#define PCI_DEVICE_ID_CDNS_MACB 0xe007
 
 #define GEM_PCLK_RATE 50000000
 #define GEM_HCLK_RATE 50000000
@@ -123,13 +110,13 @@ static void macb_remove(struct pci_dev *pdev)
 	struct platform_device *plat_dev = pci_get_drvdata(pdev);
 	struct macb_platform_data *plat_data = dev_get_platdata(&plat_dev->dev);
 
-	platform_device_unregister(plat_dev);
 	clk_unregister(plat_data->pclk);
 	clk_unregister(plat_data->hclk);
+	platform_device_unregister(plat_dev);
 }
 
-static struct pci_device_id dev_id_table[] = {
-	{ PCI_DEVICE(CDNS_VENDOR_ID, CDNS_DEVICE_ID), },
+static const struct pci_device_id dev_id_table[] = {
+	{ PCI_VDEVICE(CDNS, PCI_DEVICE_ID_CDNS_MACB) },
 	{ 0, }
 };
 
diff --git a/drivers/net/ethernet/cadence/macb_ptp.c b/drivers/net/ethernet/cadence/macb_ptp.c
new file mode 100644
index 000000000000..a63bf29c4fa8
--- /dev/null
+++ b/drivers/net/ethernet/cadence/macb_ptp.c
@@ -0,0 +1,466 @@
+// SPDX-License-Identifier: GPL-2.0-only
+/*
+ * 1588 PTP support for Cadence GEM device.
+ *
+ * Copyright (C) 2017 Cadence Design Systems - https://www.cadence.com
+ *
+ * Authors: Rafal Ozieblo <rafalo@cadence.com>
+ *          Bartosz Folta <bfolta@cadence.com>
+ */
+#include <linux/kernel.h>
+#include <linux/types.h>
+#include <linux/clk.h>
+#include <linux/device.h>
+#include <linux/etherdevice.h>
+#include <linux/platform_device.h>
+#include <linux/time64.h>
+#include <linux/ptp_classify.h>
+#include <linux/if_ether.h>
+#include <linux/if_vlan.h>
+#include <linux/net_tstamp.h>
+#include <linux/circ_buf.h>
+#include <linux/spinlock.h>
+
+#include "macb.h"
+
+#define  GEM_PTP_TIMER_NAME "gem-ptp-timer"
+
+static struct macb_dma_desc_ptp *macb_ptp_desc(struct macb *bp,
+					       struct macb_dma_desc *desc)
+{
+	if (bp->hw_dma_cap == HW_DMA_CAP_PTP)
+		return (struct macb_dma_desc_ptp *)
+				((u8 *)desc + sizeof(struct macb_dma_desc));
+	if (bp->hw_dma_cap == HW_DMA_CAP_64B_PTP)
+		return (struct macb_dma_desc_ptp *)
+				((u8 *)desc + sizeof(struct macb_dma_desc)
+				+ sizeof(struct macb_dma_desc_64));
+	return NULL;
+}
+
+static int gem_tsu_get_time(struct ptp_clock_info *ptp, struct timespec64 *ts,
+			    struct ptp_system_timestamp *sts)
+{
+	struct macb *bp = container_of(ptp, struct macb, ptp_clock_info);
+	unsigned long flags;
+	long first, second;
+	u32 secl, sech;
+
+	spin_lock_irqsave(&bp->tsu_clk_lock, flags);
+	ptp_read_system_prets(sts);
+	first = gem_readl(bp, TN);
+	ptp_read_system_postts(sts);
+	secl = gem_readl(bp, TSL);
+	sech = gem_readl(bp, TSH);
+	second = gem_readl(bp, TN);
+
+	/* test for nsec rollover */
+	if (first > second) {
+		/* if so, use later read & re-read seconds
+		 * (assume all done within 1s)
+		 */
+		ptp_read_system_prets(sts);
+		ts->tv_nsec = gem_readl(bp, TN);
+		ptp_read_system_postts(sts);
+		secl = gem_readl(bp, TSL);
+		sech = gem_readl(bp, TSH);
+	} else {
+		ts->tv_nsec = first;
+	}
+
+	spin_unlock_irqrestore(&bp->tsu_clk_lock, flags);
+	ts->tv_sec = (((u64)sech << GEM_TSL_SIZE) | secl)
+			& TSU_SEC_MAX_VAL;
+	return 0;
+}
+
+static int gem_tsu_set_time(struct ptp_clock_info *ptp,
+			    const struct timespec64 *ts)
+{
+	struct macb *bp = container_of(ptp, struct macb, ptp_clock_info);
+	unsigned long flags;
+	u32 ns, sech, secl;
+
+	secl = (u32)ts->tv_sec;
+	sech = (ts->tv_sec >> GEM_TSL_SIZE) & ((1 << GEM_TSH_SIZE) - 1);
+	ns = ts->tv_nsec;
+
+	spin_lock_irqsave(&bp->tsu_clk_lock, flags);
+
+	/* TSH doesn't latch the time and no atomicity! */
+	gem_writel(bp, TN, 0); /* clear to avoid overflow */
+	gem_writel(bp, TSH, sech);
+	/* write lower bits 2nd, for synchronized secs update */
+	gem_writel(bp, TSL, secl);
+	gem_writel(bp, TN, ns);
+
+	spin_unlock_irqrestore(&bp->tsu_clk_lock, flags);
+
+	return 0;
+}
+
+static int gem_tsu_incr_set(struct macb *bp, struct tsu_incr *incr_spec)
+{
+	unsigned long flags;
+
+	/* tsu_timer_incr register must be written after
+	 * the tsu_timer_incr_sub_ns register and the write operation
+	 * will cause the value written to the tsu_timer_incr_sub_ns register
+	 * to take effect.
+	 */
+	spin_lock_irqsave(&bp->tsu_clk_lock, flags);
+	/* RegBit[15:0] = Subns[23:8]; RegBit[31:24] = Subns[7:0] */
+	gem_writel(bp, TISUBN, GEM_BF(SUBNSINCRL, incr_spec->sub_ns) |
+		   GEM_BF(SUBNSINCRH, (incr_spec->sub_ns >>
+			  GEM_SUBNSINCRL_SIZE)));
+	gem_writel(bp, TI, GEM_BF(NSINCR, incr_spec->ns));
+	spin_unlock_irqrestore(&bp->tsu_clk_lock, flags);
+
+	return 0;
+}
+
+static int gem_ptp_adjfine(struct ptp_clock_info *ptp, long scaled_ppm)
+{
+	struct macb *bp = container_of(ptp, struct macb, ptp_clock_info);
+	struct tsu_incr incr_spec;
+	bool neg_adj = false;
+	u32 word;
+	u64 adj;
+
+	if (scaled_ppm < 0) {
+		neg_adj = true;
+		scaled_ppm = -scaled_ppm;
+	}
+
+	/* Adjustment is relative to base frequency */
+	incr_spec.sub_ns = bp->tsu_incr.sub_ns;
+	incr_spec.ns = bp->tsu_incr.ns;
+
+	/* scaling: unused(8bit) | ns(8bit) | fractions(16bit) */
+	word = ((u64)incr_spec.ns << GEM_SUBNSINCR_SIZE) + incr_spec.sub_ns;
+	adj = (u64)scaled_ppm * word;
+	/* Divide with rounding, equivalent to floating dividing:
+	 * (temp / USEC_PER_SEC) + 0.5
+	 */
+	adj += (USEC_PER_SEC >> 1);
+	adj >>= PPM_FRACTION; /* remove fractions */
+	adj = div_u64(adj, USEC_PER_SEC);
+	adj = neg_adj ? (word - adj) : (word + adj);
+
+	incr_spec.ns = (adj >> GEM_SUBNSINCR_SIZE)
+			& ((1 << GEM_NSINCR_SIZE) - 1);
+	incr_spec.sub_ns = adj & ((1 << GEM_SUBNSINCR_SIZE) - 1);
+	gem_tsu_incr_set(bp, &incr_spec);
+	return 0;
+}
+
+static int gem_ptp_adjtime(struct ptp_clock_info *ptp, s64 delta)
+{
+	struct macb *bp = container_of(ptp, struct macb, ptp_clock_info);
+	struct timespec64 now, then = ns_to_timespec64(delta);
+	u32 adj, sign = 0;
+
+	if (delta < 0) {
+		sign = 1;
+		delta = -delta;
+	}
+
+	if (delta > TSU_NSEC_MAX_VAL) {
+		gem_tsu_get_time(&bp->ptp_clock_info, &now, NULL);
+		now = timespec64_add(now, then);
+
+		gem_tsu_set_time(&bp->ptp_clock_info,
+				 (const struct timespec64 *)&now);
+	} else {
+		adj = (sign << GEM_ADDSUB_OFFSET) | delta;
+
+		gem_writel(bp, TA, adj);
+	}
+
+	return 0;
+}
+
+static int gem_ptp_enable(struct ptp_clock_info *ptp,
+			  struct ptp_clock_request *rq, int on)
+{
+	return -EOPNOTSUPP;
+}
+
+static const struct ptp_clock_info gem_ptp_caps_template = {
+	.owner		= THIS_MODULE,
+	.name		= GEM_PTP_TIMER_NAME,
+	.max_adj	= 0,
+	.n_alarm	= 0,
+	.n_ext_ts	= 0,
+	.n_per_out	= 0,
+	.n_pins		= 0,
+	.pps		= 1,
+	.adjfine	= gem_ptp_adjfine,
+	.adjtime	= gem_ptp_adjtime,
+	.gettimex64	= gem_tsu_get_time,
+	.settime64	= gem_tsu_set_time,
+	.enable		= gem_ptp_enable,
+};
+
+static void gem_ptp_init_timer(struct macb *bp)
+{
+	u32 rem = 0;
+	u64 adj;
+
+	bp->tsu_incr.ns = div_u64_rem(NSEC_PER_SEC, bp->tsu_rate, &rem);
+	if (rem) {
+		adj = rem;
+		adj <<= GEM_SUBNSINCR_SIZE;
+		bp->tsu_incr.sub_ns = div_u64(adj, bp->tsu_rate);
+	} else {
+		bp->tsu_incr.sub_ns = 0;
+	}
+}
+
+static void gem_ptp_init_tsu(struct macb *bp)
+{
+	struct timespec64 ts;
+
+	/* 1. get current system time */
+	ts = ns_to_timespec64(ktime_to_ns(ktime_get_real()));
+
+	/* 2. set ptp timer */
+	gem_tsu_set_time(&bp->ptp_clock_info, &ts);
+
+	/* 3. set PTP timer increment value to BASE_INCREMENT */
+	gem_tsu_incr_set(bp, &bp->tsu_incr);
+
+	gem_writel(bp, TA, 0);
+}
+
+static void gem_ptp_clear_timer(struct macb *bp)
+{
+	bp->tsu_incr.sub_ns = 0;
+	bp->tsu_incr.ns = 0;
+
+	gem_writel(bp, TISUBN, GEM_BF(SUBNSINCR, 0));
+	gem_writel(bp, TI, GEM_BF(NSINCR, 0));
+	gem_writel(bp, TA, 0);
+}
+
+static int gem_hw_timestamp(struct macb *bp, u32 dma_desc_ts_1,
+			    u32 dma_desc_ts_2, struct timespec64 *ts)
+{
+	struct timespec64 tsu;
+
+	ts->tv_sec = (GEM_BFEXT(DMA_SECH, dma_desc_ts_2) << GEM_DMA_SECL_SIZE) |
+			GEM_BFEXT(DMA_SECL, dma_desc_ts_1);
+	ts->tv_nsec = GEM_BFEXT(DMA_NSEC, dma_desc_ts_1);
+
+	/* TSU overlapping workaround
+	 * The timestamp only contains lower few bits of seconds,
+	 * so add value from 1588 timer
+	 */
+	gem_tsu_get_time(&bp->ptp_clock_info, &tsu, NULL);
+
+	ts->tv_sec |= ((~GEM_DMA_SEC_MASK) & tsu.tv_sec);
+
+	/* If the top bit is set in the timestamp,
+	 * but not in 1588 timer, it has rolled over,
+	 * so subtract max size
+	 */
+	if ((ts->tv_sec & (GEM_DMA_SEC_TOP >> 1)) &&
+	    !(tsu.tv_sec & (GEM_DMA_SEC_TOP >> 1)))
+		ts->tv_sec -= GEM_DMA_SEC_TOP;
+
+	return 0;
+}
+
+void gem_ptp_rxstamp(struct macb *bp, struct sk_buff *skb,
+		     struct macb_dma_desc *desc)
+{
+	struct skb_shared_hwtstamps *shhwtstamps = skb_hwtstamps(skb);
+	struct macb_dma_desc_ptp *desc_ptp;
+	struct timespec64 ts;
+
+	if (GEM_BFEXT(DMA_RXVALID, desc->addr)) {
+		desc_ptp = macb_ptp_desc(bp, desc);
+		/* Unlikely but check */
+		if (!desc_ptp) {
+			dev_warn_ratelimited(&bp->pdev->dev,
+					     "Timestamp not supported in BD\n");
+			return;
+		}
+		gem_hw_timestamp(bp, desc_ptp->ts_1, desc_ptp->ts_2, &ts);
+		memset(shhwtstamps, 0, sizeof(struct skb_shared_hwtstamps));
+		shhwtstamps->hwtstamp = ktime_set(ts.tv_sec, ts.tv_nsec);
+	}
+}
+
+void gem_ptp_txstamp(struct macb *bp, struct sk_buff *skb,
+		     struct macb_dma_desc *desc)
+{
+	struct skb_shared_hwtstamps shhwtstamps;
+	struct macb_dma_desc_ptp *desc_ptp;
+	struct timespec64 ts;
+
+	if (!GEM_BFEXT(DMA_TXVALID, desc->ctrl)) {
+		dev_warn_ratelimited(&bp->pdev->dev,
+				     "Timestamp not set in TX BD as expected\n");
+		return;
+	}
+
+	desc_ptp = macb_ptp_desc(bp, desc);
+	/* Unlikely but check */
+	if (!desc_ptp) {
+		dev_warn_ratelimited(&bp->pdev->dev,
+				     "Timestamp not supported in BD\n");
+		return;
+	}
+
+	/* ensure ts_1/ts_2 is loaded after ctrl (TX_USED check) */
+	dma_rmb();
+	gem_hw_timestamp(bp, desc_ptp->ts_1, desc_ptp->ts_2, &ts);
+
+	memset(&shhwtstamps, 0, sizeof(shhwtstamps));
+	shhwtstamps.hwtstamp = ktime_set(ts.tv_sec, ts.tv_nsec);
+	skb_tstamp_tx(skb, &shhwtstamps);
+}
+
+void gem_ptp_init(struct net_device *dev)
+{
+	struct macb *bp = netdev_priv(dev);
+
+	bp->ptp_clock_info = gem_ptp_caps_template;
+
+	/* nominal frequency and maximum adjustment in ppb */
+	bp->tsu_rate = bp->ptp_info->get_tsu_rate(bp);
+	bp->ptp_clock_info.max_adj = bp->ptp_info->get_ptp_max_adj();
+	gem_ptp_init_timer(bp);
+	bp->ptp_clock = ptp_clock_register(&bp->ptp_clock_info, &dev->dev);
+	if (IS_ERR(bp->ptp_clock)) {
+		pr_err("ptp clock register failed: %ld\n",
+			PTR_ERR(bp->ptp_clock));
+		bp->ptp_clock = NULL;
+		return;
+	} else if (bp->ptp_clock == NULL) {
+		pr_err("ptp clock register failed\n");
+		return;
+	}
+
+	spin_lock_init(&bp->tsu_clk_lock);
+
+	gem_ptp_init_tsu(bp);
+
+	dev_info(&bp->pdev->dev, "%s ptp clock registered.\n",
+		 GEM_PTP_TIMER_NAME);
+}
+
+void gem_ptp_remove(struct net_device *ndev)
+{
+	struct macb *bp = netdev_priv(ndev);
+
+	if (bp->ptp_clock)
+		ptp_clock_unregister(bp->ptp_clock);
+
+	gem_ptp_clear_timer(bp);
+
+	dev_info(&bp->pdev->dev, "%s ptp clock unregistered.\n",
+		 GEM_PTP_TIMER_NAME);
+}
+
+static int gem_ptp_set_ts_mode(struct macb *bp,
+			       enum macb_bd_control tx_bd_control,
+			       enum macb_bd_control rx_bd_control)
+{
+	gem_writel(bp, TXBDCTRL, GEM_BF(TXTSMODE, tx_bd_control));
+	gem_writel(bp, RXBDCTRL, GEM_BF(RXTSMODE, rx_bd_control));
+
+	return 0;
+}
+
+int gem_get_hwtst(struct net_device *dev,
+		  struct kernel_hwtstamp_config *tstamp_config)
+{
+	struct macb *bp = netdev_priv(dev);
+
+	*tstamp_config = bp->tstamp_config;
+	if ((bp->hw_dma_cap & HW_DMA_CAP_PTP) == 0)
+		return -EOPNOTSUPP;
+
+	return 0;
+}
+
+static void gem_ptp_set_one_step_sync(struct macb *bp, u8 enable)
+{
+	u32 reg_val;
+
+	reg_val = macb_readl(bp, NCR);
+
+	if (enable)
+		macb_writel(bp, NCR, reg_val | MACB_BIT(OSSMODE));
+	else
+		macb_writel(bp, NCR, reg_val & ~MACB_BIT(OSSMODE));
+}
+
+int gem_set_hwtst(struct net_device *dev,
+		  struct kernel_hwtstamp_config *tstamp_config,
+		  struct netlink_ext_ack *extack)
+{
+	enum macb_bd_control tx_bd_control = TSTAMP_DISABLED;
+	enum macb_bd_control rx_bd_control = TSTAMP_DISABLED;
+	struct macb *bp = netdev_priv(dev);
+	u32 regval;
+
+	if ((bp->hw_dma_cap & HW_DMA_CAP_PTP) == 0)
+		return -EOPNOTSUPP;
+
+	switch (tstamp_config->tx_type) {
+	case HWTSTAMP_TX_OFF:
+		break;
+	case HWTSTAMP_TX_ONESTEP_SYNC:
+		gem_ptp_set_one_step_sync(bp, 1);
+		tx_bd_control = TSTAMP_ALL_FRAMES;
+		break;
+	case HWTSTAMP_TX_ON:
+		gem_ptp_set_one_step_sync(bp, 0);
+		tx_bd_control = TSTAMP_ALL_FRAMES;
+		break;
+	default:
+		return -ERANGE;
+	}
+
+	switch (tstamp_config->rx_filter) {
+	case HWTSTAMP_FILTER_NONE:
+		break;
+	case HWTSTAMP_FILTER_PTP_V1_L4_SYNC:
+		break;
+	case HWTSTAMP_FILTER_PTP_V1_L4_DELAY_REQ:
+		break;
+	case HWTSTAMP_FILTER_PTP_V2_EVENT:
+	case HWTSTAMP_FILTER_PTP_V2_L2_EVENT:
+	case HWTSTAMP_FILTER_PTP_V2_L4_EVENT:
+	case HWTSTAMP_FILTER_PTP_V2_SYNC:
+	case HWTSTAMP_FILTER_PTP_V2_L2_SYNC:
+	case HWTSTAMP_FILTER_PTP_V2_L4_SYNC:
+	case HWTSTAMP_FILTER_PTP_V2_DELAY_REQ:
+	case HWTSTAMP_FILTER_PTP_V2_L2_DELAY_REQ:
+	case HWTSTAMP_FILTER_PTP_V2_L4_DELAY_REQ:
+		rx_bd_control =  TSTAMP_ALL_PTP_FRAMES;
+		tstamp_config->rx_filter = HWTSTAMP_FILTER_PTP_V2_EVENT;
+		regval = macb_readl(bp, NCR);
+		macb_writel(bp, NCR, (regval | MACB_BIT(SRTSM)));
+		break;
+	case HWTSTAMP_FILTER_PTP_V1_L4_EVENT:
+	case HWTSTAMP_FILTER_ALL:
+		rx_bd_control = TSTAMP_ALL_FRAMES;
+		tstamp_config->rx_filter = HWTSTAMP_FILTER_ALL;
+		break;
+	default:
+		tstamp_config->rx_filter = HWTSTAMP_FILTER_NONE;
+		return -ERANGE;
+	}
+
+	bp->tstamp_config = *tstamp_config;
+
+	if (gem_ptp_set_ts_mode(bp, tx_bd_control, rx_bd_control) != 0)
+		return -ERANGE;
+
+	return 0;
+}
+
diff --git a/drivers/net/ethernet/calxeda/Kconfig b/drivers/net/ethernet/calxeda/Kconfig
index 07d2201530d2..ce42157f13f6 100644
--- a/drivers/net/ethernet/calxeda/Kconfig
+++ b/drivers/net/ethernet/calxeda/Kconfig
@@ -1,6 +1,7 @@
+# SPDX-License-Identifier: GPL-2.0-only
 config NET_CALXEDA_XGMAC
 	tristate "Calxeda 1G/10G XGMAC Ethernet driver"
-	depends on HAS_IOMEM && HAS_DMA
+	depends on HAS_IOMEM
 	depends on ARCH_HIGHBANK || COMPILE_TEST
 	select CRC32
 	help
diff --git a/drivers/net/ethernet/calxeda/Makefile b/drivers/net/ethernet/calxeda/Makefile
index f0ef08067f97..641e5b6b5ac7 100644
--- a/drivers/net/ethernet/calxeda/Makefile
+++ b/drivers/net/ethernet/calxeda/Makefile
@@ -1 +1,2 @@
+# SPDX-License-Identifier: GPL-2.0-only
 obj-$(CONFIG_NET_CALXEDA_XGMAC) += xgmac.o
diff --git a/drivers/net/ethernet/calxeda/xgmac.c b/drivers/net/ethernet/calxeda/xgmac.c
index 2bd7c638b178..331ac6a3dc38 100644
--- a/drivers/net/ethernet/calxeda/xgmac.c
+++ b/drivers/net/ethernet/calxeda/xgmac.c
@@ -1,19 +1,9 @@
+// SPDX-License-Identifier: GPL-2.0-only
 /*
  * Copyright 2010-2011 Calxeda, Inc.
- *
- * This program is free software; you can redistribute it and/or modify it
- * under the terms and conditions of the GNU General Public License,
- * version 2, as published by the Free Software Foundation.
- *
- * This program is distributed in the hope it will be useful, but WITHOUT
- * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
- * FITNESS FOR A PARTICULAR PURPOSE.  See the GNU General Public License for
- * more details.
- *
- * You should have received a copy of the GNU General Public License along with
- * this program.  If not, see <http://www.gnu.org/licenses/>.
  */
 #include <linux/module.h>
+#include <linux/mod_devicetable.h>
 #include <linux/kernel.h>
 #include <linux/circ_buf.h>
 #include <linux/interrupt.h>
@@ -617,7 +607,7 @@ static inline void xgmac_mac_disable(void __iomem *ioaddr)
 	writel(value, ioaddr + XGMAC_CONTROL);
 }
 
-static void xgmac_set_mac_addr(void __iomem *ioaddr, unsigned char *addr,
+static void xgmac_set_mac_addr(void __iomem *ioaddr, const unsigned char *addr,
 			       int num)
 {
 	u32 data;
@@ -721,7 +711,7 @@ static void xgmac_rx_refill(struct xgmac_priv *priv)
 }
 
 /**
- * init_xgmac_dma_desc_rings - init the RX/TX descriptor rings
+ * xgmac_dma_desc_rings_init - init the RX/TX descriptor rings
  * @dev: net device structure
  * Description:  this function initializes the DMA RX/TX descriptors
  * and allocates the socket buffers.
@@ -739,7 +729,7 @@ static int xgmac_dma_desc_rings_init(struct net_device *dev)
 
 	netdev_dbg(priv->dev, "mtu [%d] bfsize [%d]\n", dev->mtu, bfsize);
 
-	priv->rx_skbuff = kzalloc(sizeof(struct sk_buff *) * DMA_RX_RING_SZ,
+	priv->rx_skbuff = kcalloc(DMA_RX_RING_SZ, sizeof(struct sk_buff *),
 				  GFP_KERNEL);
 	if (!priv->rx_skbuff)
 		return -ENOMEM;
@@ -752,7 +742,7 @@ static int xgmac_dma_desc_rings_init(struct net_device *dev)
 	if (!priv->dma_rx)
 		goto err_dma_rx;
 
-	priv->tx_skbuff = kzalloc(sizeof(struct sk_buff *) * DMA_TX_RING_SZ,
+	priv->tx_skbuff = kcalloc(DMA_TX_RING_SZ, sizeof(struct sk_buff *),
 				  GFP_KERNEL);
 	if (!priv->tx_skbuff)
 		goto err_tx_skb;
@@ -869,7 +859,7 @@ static void xgmac_free_dma_desc_rings(struct xgmac_priv *priv)
 }
 
 /**
- * xgmac_tx:
+ * xgmac_tx_complete:
  * @priv: private driver structure
  * Description: it reclaims resources after transmission completes.
  */
@@ -1050,7 +1040,7 @@ static int xgmac_open(struct net_device *dev)
 }
 
 /**
- *  xgmac_release - close entry point of the driver
+ *  xgmac_stop - close entry point of the driver
  *  @dev : device pointer.
  *  Description:
  *  This is the stop entry point of the driver.
@@ -1114,7 +1104,7 @@ static netdev_tx_t xgmac_xmit(struct sk_buff *skb, struct net_device *dev)
 	for (i = 0; i < nfrags; i++) {
 		skb_frag_t *frag = &skb_shinfo(skb)->frags[i];
 
-		len = frag->size;
+		len = skb_frag_size(frag);
 
 		paddr = skb_frag_dma_map(priv->device, frag, 0, len,
 					 DMA_TO_DEVICE);
@@ -1234,7 +1224,7 @@ static int xgmac_rx(struct xgmac_priv *priv, int limit)
  *  @budget : maximum number of packets that the current CPU can receive from
  *	      all interfaces.
  *  Description :
- *   This function implements the the reception process.
+ *   This function implements the reception process.
  *   Also it runs the TX completion thread
  */
 static int xgmac_poll(struct napi_struct *napi, int budget)
@@ -1256,12 +1246,14 @@ static int xgmac_poll(struct napi_struct *napi, int budget)
 /**
  *  xgmac_tx_timeout
  *  @dev : Pointer to net device structure
+ *  @txqueue: index of the hung transmit queue
+ *
  *  Description: this function is called when a packet transmission fails to
  *   complete within a reasonable tmrate. The driver will mark the error in the
  *   netdev structure and arrange for the device to be reset to a sane state
  *   in order to transmit a new packet.
  */
-static void xgmac_tx_timeout(struct net_device *dev)
+static void xgmac_tx_timeout(struct net_device *dev, unsigned int txqueue)
 {
 	struct xgmac_priv *priv = netdev_priv(dev);
 	schedule_work(&priv->tx_timeout_work);
@@ -1366,7 +1358,7 @@ static int xgmac_change_mtu(struct net_device *dev, int new_mtu)
 
 	/* Bring interface down, change mtu and bring interface back up */
 	xgmac_stop(dev);
-	dev->mtu = new_mtu;
+	WRITE_ONCE(dev->mtu, new_mtu);
 	return xgmac_open(dev);
 }
 
@@ -1487,7 +1479,7 @@ static int xgmac_set_mac_address(struct net_device *dev, void *p)
 	if (!is_valid_ether_addr(addr->sa_data))
 		return -EADDRNOTAVAIL;
 
-	memcpy(dev->dev_addr, addr->sa_data, dev->addr_len);
+	eth_hw_addr_set(dev, addr->sa_data);
 
 	xgmac_set_mac_addr(ioaddr, dev->dev_addr, 0);
 
@@ -1701,6 +1693,7 @@ static int xgmac_probe(struct platform_device *pdev)
 	struct resource *res;
 	struct net_device *ndev = NULL;
 	struct xgmac_priv *priv = NULL;
+	u8 addr[ETH_ALEN];
 	u32 uid;
 
 	res = platform_get_resource(pdev, IORESOURCE_MEM, 0);
@@ -1793,12 +1786,13 @@ static int xgmac_probe(struct platform_device *pdev)
 	ndev->max_mtu = XGMAC_MAX_MTU;
 
 	/* Get the MAC address */
-	xgmac_get_mac_addr(priv->base, ndev->dev_addr, 0);
+	xgmac_get_mac_addr(priv->base, addr, 0);
+	eth_hw_addr_set(ndev, addr);
 	if (!is_valid_ether_addr(ndev->dev_addr))
 		netdev_warn(ndev, "MAC address %pM not valid",
 			 ndev->dev_addr);
 
-	netif_napi_add(ndev, &priv->napi, xgmac_poll, 64);
+	netif_napi_add(ndev, &priv->napi, xgmac_poll);
 	ret = register_netdev(ndev);
 	if (ret)
 		goto err_reg;
@@ -1820,13 +1814,13 @@ err_alloc:
 }
 
 /**
- * xgmac_dvr_remove
+ * xgmac_remove
  * @pdev: platform device pointer
  * Description: this function resets the TX/RX processes, disables the MAC RX/TX
  * changes the link status, releases the DMA descriptor rings,
  * unregisters the MDIO bus and unmaps the allocated memory.
  */
-static int xgmac_remove(struct platform_device *pdev)
+static void xgmac_remove(struct platform_device *pdev)
 {
 	struct net_device *ndev = platform_get_drvdata(pdev);
 	struct xgmac_priv *priv = netdev_priv(ndev);
@@ -1846,8 +1840,6 @@ static int xgmac_remove(struct platform_device *pdev)
 	release_mem_region(res->start, resource_size(res));
 
 	free_netdev(ndev);
-
-	return 0;
 }
 
 #ifdef CONFIG_PM_SLEEP
@@ -1865,7 +1857,7 @@ static void xgmac_pmt(void __iomem *ioaddr, unsigned long mode)
 
 static int xgmac_suspend(struct device *dev)
 {
-	struct net_device *ndev = platform_get_drvdata(to_platform_device(dev));
+	struct net_device *ndev = dev_get_drvdata(dev);
 	struct xgmac_priv *priv = netdev_priv(ndev);
 	u32 value;
 
@@ -1891,7 +1883,7 @@ static int xgmac_suspend(struct device *dev)
 
 static int xgmac_resume(struct device *dev)
 {
-	struct net_device *ndev = platform_get_drvdata(to_platform_device(dev));
+	struct net_device *ndev = dev_get_drvdata(dev);
 	struct xgmac_priv *priv = netdev_priv(ndev);
 	void __iomem *ioaddr = priv->base;
 
@@ -1924,10 +1916,10 @@ static struct platform_driver xgmac_driver = {
 	.driver = {
 		.name = "calxedaxgmac",
 		.of_match_table = xgmac_of_match,
+		.pm = &xgmac_pm_ops,
 	},
 	.probe = xgmac_probe,
 	.remove = xgmac_remove,
-	.driver.pm = &xgmac_pm_ops,
 };
 
 module_platform_driver(xgmac_driver);
diff --git a/drivers/net/ethernet/cavium/Kconfig b/drivers/net/ethernet/cavium/Kconfig
index dcbce6cac63e..7dae5aad3689 100644
--- a/drivers/net/ethernet/cavium/Kconfig
+++ b/drivers/net/ethernet/cavium/Kconfig
@@ -1,12 +1,12 @@
+# SPDX-License-Identifier: GPL-2.0-only
 #
 # Cavium ethernet device configuration
 #
 
 config NET_VENDOR_CAVIUM
 	bool "Cavium ethernet drivers"
-	depends on PCI
 	default y
-	---help---
+	help
 	  Select this option if you want enable Cavium network support.
 
 	  If you have a Cavium SoC or network adapter, say Y.
@@ -15,9 +15,9 @@ if NET_VENDOR_CAVIUM
 
 config THUNDER_NIC_PF
 	tristate "Thunder Physical function driver"
-	depends on 64BIT
+	depends on 64BIT && PCI
 	select THUNDER_NIC_BGX
-	---help---
+	help
 	  This driver supports Thunder's NIC physical function.
 	  The NIC provides the controller and DMA engines to
 	  move network traffic to/from the memory. The NIC
@@ -27,36 +27,54 @@ config THUNDER_NIC_PF
 
 config THUNDER_NIC_VF
 	tristate "Thunder Virtual function driver"
-	depends on 64BIT
-	---help---
+	imply CAVIUM_PTP
+	depends on 64BIT && PCI
+	help
 	  This driver supports Thunder's NIC virtual function
 
 config	THUNDER_NIC_BGX
 	tristate "Thunder MAC interface driver (BGX)"
-	depends on 64BIT
+	depends on 64BIT && PCI
 	select PHYLIB
-	select MDIO_THUNDER
+	select MDIO_THUNDER if PCI
 	select THUNDER_NIC_RGX
-	---help---
+	help
 	  This driver supports programming and controlling of MAC
 	  interface from NIC physical function driver.
 
 config	THUNDER_NIC_RGX
 	tristate "Thunder MAC interface driver (RGX)"
-	depends on 64BIT
+	depends on 64BIT && PCI
 	select PHYLIB
-	select MDIO_THUNDER
-	---help---
+	select MDIO_THUNDER if PCI
+	help
 	  This driver supports configuring XCV block of RGX interface
 	  present on CN81XX chip.
 
+config CAVIUM_PTP
+	tristate "Cavium PTP coprocessor as PTP clock"
+	depends on 64BIT && PCI
+	depends on PTP_1588_CLOCK
+	help
+	  This driver adds support for the Precision Time Protocol Clocks and
+	  Timestamping coprocessor (PTP) found on Cavium processors.
+	  PTP provides timestamping mechanism that is suitable for use in IEEE 1588
+	  Precision Time Protocol or other purposes.  Timestamps can be used in
+	  BGX, TNS, GTI, and NIC blocks.
+
+config LIQUIDIO_CORE
+	tristate
+
 config LIQUIDIO
 	tristate "Cavium LiquidIO support"
-	depends on 64BIT
-	imply PTP_1588_CLOCK
+	depends on 64BIT && PCI
+	depends on PCI
+	depends on PTP_1588_CLOCK_OPTIONAL
+	select CRC32
 	select FW_LOADER
-	select LIBCRC32C
-	---help---
+	select LIQUIDIO_CORE
+	select NET_DEVLINK
+	help
 	  This driver supports Cavium LiquidIO Intelligent Server Adapters
 	  based on CN66XX, CN68XX and CN23XX chips.
 
@@ -77,8 +95,9 @@ config OCTEON_MGMT_ETHERNET
 config LIQUIDIO_VF
 	tristate "Cavium LiquidIO VF support"
 	depends on 64BIT && PCI_MSI
-	imply PTP_1588_CLOCK
-	---help---
+	depends on PTP_1588_CLOCK_OPTIONAL
+	select LIQUIDIO_CORE
+	help
 	  This driver supports Cavium LiquidIO Intelligent Server Adapter
 	  based on CN23XX chips.
 
diff --git a/drivers/net/ethernet/cavium/Makefile b/drivers/net/ethernet/cavium/Makefile
index 872da9f7c31a..5d32808210fb 100644
--- a/drivers/net/ethernet/cavium/Makefile
+++ b/drivers/net/ethernet/cavium/Makefile
@@ -1,6 +1,8 @@
+# SPDX-License-Identifier: GPL-2.0-only
 #
 # Makefile for the Cavium ethernet device drivers.
 #
+obj-$(CONFIG_NET_VENDOR_CAVIUM) += common/
 obj-$(CONFIG_NET_VENDOR_CAVIUM) += thunder/
 obj-$(CONFIG_NET_VENDOR_CAVIUM) += liquidio/
 obj-$(CONFIG_NET_VENDOR_CAVIUM) += octeon/
diff --git a/drivers/net/ethernet/cavium/common/Makefile b/drivers/net/ethernet/cavium/common/Makefile
new file mode 100644
index 000000000000..e3f87bd65928
--- /dev/null
+++ b/drivers/net/ethernet/cavium/common/Makefile
@@ -0,0 +1,2 @@
+# SPDX-License-Identifier: GPL-2.0-only
+obj-$(CONFIG_CAVIUM_PTP) += cavium_ptp.o
diff --git a/drivers/net/ethernet/cavium/common/cavium_ptp.c b/drivers/net/ethernet/cavium/common/cavium_ptp.c
new file mode 100644
index 000000000000..61e261657073
--- /dev/null
+++ b/drivers/net/ethernet/cavium/common/cavium_ptp.c
@@ -0,0 +1,347 @@
+// SPDX-License-Identifier: GPL-2.0
+/* cavium_ptp.c - PTP 1588 clock on Cavium hardware
+ * Copyright (c) 2003-2015, 2017 Cavium, Inc.
+ */
+
+#include <linux/device.h>
+#include <linux/module.h>
+#include <linux/timecounter.h>
+#include <linux/pci.h>
+
+#include "cavium_ptp.h"
+
+#define DRV_NAME "cavium_ptp"
+
+#define PCI_DEVICE_ID_CAVIUM_PTP	0xA00C
+#define PCI_SUBSYS_DEVID_88XX_PTP	0xA10C
+#define PCI_SUBSYS_DEVID_81XX_PTP	0XA20C
+#define PCI_SUBSYS_DEVID_83XX_PTP	0xA30C
+#define PCI_DEVICE_ID_CAVIUM_RST	0xA00E
+
+#define PCI_PTP_BAR_NO	0
+#define PCI_RST_BAR_NO	0
+
+#define PTP_CLOCK_CFG		0xF00ULL
+#define  PTP_CLOCK_CFG_PTP_EN	BIT(0)
+#define PTP_CLOCK_LO		0xF08ULL
+#define PTP_CLOCK_HI		0xF10ULL
+#define PTP_CLOCK_COMP		0xF18ULL
+
+#define RST_BOOT	0x1600ULL
+#define CLOCK_BASE_RATE	50000000ULL
+
+static u64 ptp_cavium_clock_get(void)
+{
+	struct pci_dev *pdev;
+	void __iomem *base;
+	u64 ret = CLOCK_BASE_RATE * 16;
+
+	pdev = pci_get_device(PCI_VENDOR_ID_CAVIUM,
+			      PCI_DEVICE_ID_CAVIUM_RST, NULL);
+	if (!pdev)
+		goto error;
+
+	base = pci_ioremap_bar(pdev, PCI_RST_BAR_NO);
+	if (!base)
+		goto error_put_pdev;
+
+	ret = CLOCK_BASE_RATE * ((readq(base + RST_BOOT) >> 33) & 0x3f);
+
+	iounmap(base);
+
+error_put_pdev:
+	pci_dev_put(pdev);
+
+error:
+	return ret;
+}
+
+struct cavium_ptp *cavium_ptp_get(void)
+{
+	struct cavium_ptp *ptp;
+	struct pci_dev *pdev;
+
+	pdev = pci_get_device(PCI_VENDOR_ID_CAVIUM,
+			      PCI_DEVICE_ID_CAVIUM_PTP, NULL);
+	if (!pdev)
+		return ERR_PTR(-ENODEV);
+
+	ptp = pci_get_drvdata(pdev);
+	if (!ptp)
+		ptp = ERR_PTR(-EPROBE_DEFER);
+	if (IS_ERR(ptp))
+		pci_dev_put(pdev);
+
+	return ptp;
+}
+EXPORT_SYMBOL(cavium_ptp_get);
+
+void cavium_ptp_put(struct cavium_ptp *ptp)
+{
+	if (!ptp)
+		return;
+	pci_dev_put(ptp->pdev);
+}
+EXPORT_SYMBOL(cavium_ptp_put);
+
+/**
+ * cavium_ptp_adjfine() - Adjust ptp frequency
+ * @ptp_info: PTP clock info
+ * @scaled_ppm: how much to adjust by, in parts per million, but with a
+ *              16 bit binary fractional field
+ */
+static int cavium_ptp_adjfine(struct ptp_clock_info *ptp_info, long scaled_ppm)
+{
+	struct cavium_ptp *clock =
+		container_of(ptp_info, struct cavium_ptp, ptp_info);
+	unsigned long flags;
+	u64 comp;
+	u64 adj;
+	bool neg_adj = false;
+
+	if (scaled_ppm < 0) {
+		neg_adj = true;
+		scaled_ppm = -scaled_ppm;
+	}
+
+	/* The hardware adds the clock compensation value to the PTP clock
+	 * on every coprocessor clock cycle. Typical convention is that it
+	 * represent number of nanosecond betwen each cycle. In this
+	 * convention compensation value is in 64 bit fixed-point
+	 * representation where upper 32 bits are number of nanoseconds
+	 * and lower is fractions of nanosecond.
+	 * The scaled_ppm represent the ratio in "parts per bilion" by which the
+	 * compensation value should be corrected.
+	 * To calculate new compenstation value we use 64bit fixed point
+	 * arithmetic on following formula
+	 * comp = tbase + tbase * scaled_ppm / (1M * 2^16)
+	 * where tbase is the basic compensation value calculated initialy
+	 * in cavium_ptp_init() -> tbase = 1/Hz. Then we use endian
+	 * independent structure definition to write data to PTP register.
+	 */
+	comp = ((u64)1000000000ull << 32) / clock->clock_rate;
+	adj = comp * scaled_ppm;
+	adj >>= 16;
+	adj = div_u64(adj, 1000000ull);
+	comp = neg_adj ? comp - adj : comp + adj;
+
+	spin_lock_irqsave(&clock->spin_lock, flags);
+	writeq(comp, clock->reg_base + PTP_CLOCK_COMP);
+	spin_unlock_irqrestore(&clock->spin_lock, flags);
+
+	return 0;
+}
+
+/**
+ * cavium_ptp_adjtime() - Adjust ptp time
+ * @ptp_info:   PTP clock info
+ * @delta: how much to adjust by, in nanosecs
+ */
+static int cavium_ptp_adjtime(struct ptp_clock_info *ptp_info, s64 delta)
+{
+	struct cavium_ptp *clock =
+		container_of(ptp_info, struct cavium_ptp, ptp_info);
+	unsigned long flags;
+
+	spin_lock_irqsave(&clock->spin_lock, flags);
+	timecounter_adjtime(&clock->time_counter, delta);
+	spin_unlock_irqrestore(&clock->spin_lock, flags);
+
+	/* Sync, for network driver to get latest value */
+	smp_mb();
+
+	return 0;
+}
+
+/**
+ * cavium_ptp_gettime() - Get hardware clock time with adjustment
+ * @ptp_info: PTP clock info
+ * @ts:  timespec
+ */
+static int cavium_ptp_gettime(struct ptp_clock_info *ptp_info,
+			      struct timespec64 *ts)
+{
+	struct cavium_ptp *clock =
+		container_of(ptp_info, struct cavium_ptp, ptp_info);
+	unsigned long flags;
+	u64 nsec;
+
+	spin_lock_irqsave(&clock->spin_lock, flags);
+	nsec = timecounter_read(&clock->time_counter);
+	spin_unlock_irqrestore(&clock->spin_lock, flags);
+
+	*ts = ns_to_timespec64(nsec);
+
+	return 0;
+}
+
+/**
+ * cavium_ptp_settime() - Set hardware clock time. Reset adjustment
+ * @ptp_info: PTP clock info
+ * @ts:  timespec
+ */
+static int cavium_ptp_settime(struct ptp_clock_info *ptp_info,
+			      const struct timespec64 *ts)
+{
+	struct cavium_ptp *clock =
+		container_of(ptp_info, struct cavium_ptp, ptp_info);
+	unsigned long flags;
+	u64 nsec;
+
+	nsec = timespec64_to_ns(ts);
+
+	spin_lock_irqsave(&clock->spin_lock, flags);
+	timecounter_init(&clock->time_counter, &clock->cycle_counter, nsec);
+	spin_unlock_irqrestore(&clock->spin_lock, flags);
+
+	return 0;
+}
+
+/**
+ * cavium_ptp_enable() - Request to enable or disable an ancillary feature.
+ * @ptp_info: PTP clock info
+ * @rq:  request
+ * @on:  is it on
+ */
+static int cavium_ptp_enable(struct ptp_clock_info *ptp_info,
+			     struct ptp_clock_request *rq, int on)
+{
+	return -EOPNOTSUPP;
+}
+
+static u64 cavium_ptp_cc_read(struct cyclecounter *cc)
+{
+	struct cavium_ptp *clock =
+		container_of(cc, struct cavium_ptp, cycle_counter);
+
+	return readq(clock->reg_base + PTP_CLOCK_HI);
+}
+
+static int cavium_ptp_probe(struct pci_dev *pdev,
+			    const struct pci_device_id *ent)
+{
+	struct device *dev = &pdev->dev;
+	struct cavium_ptp *clock;
+	struct cyclecounter *cc;
+	u64 clock_cfg;
+	u64 clock_comp;
+	int err;
+
+	clock = devm_kzalloc(dev, sizeof(*clock), GFP_KERNEL);
+	if (!clock) {
+		err = -ENOMEM;
+		goto error;
+	}
+
+	clock->pdev = pdev;
+
+	err = pcim_enable_device(pdev);
+	if (err)
+		goto error_free;
+
+	clock->reg_base = pcim_iomap_region(pdev, PCI_PTP_BAR_NO, pci_name(pdev));
+	err = PTR_ERR_OR_ZERO(clock->reg_base);
+	if (err)
+		goto error_free;
+
+	spin_lock_init(&clock->spin_lock);
+
+	cc = &clock->cycle_counter;
+	cc->read = cavium_ptp_cc_read;
+	cc->mask = CYCLECOUNTER_MASK(64);
+	cc->mult = 1;
+	cc->shift = 0;
+
+	timecounter_init(&clock->time_counter, &clock->cycle_counter,
+			 ktime_to_ns(ktime_get_real()));
+
+	clock->clock_rate = ptp_cavium_clock_get();
+
+	clock->ptp_info = (struct ptp_clock_info) {
+		.owner		= THIS_MODULE,
+		.name		= "ThunderX PTP",
+		.max_adj	= 1000000000ull,
+		.n_ext_ts	= 0,
+		.n_pins		= 0,
+		.pps		= 0,
+		.adjfine	= cavium_ptp_adjfine,
+		.adjtime	= cavium_ptp_adjtime,
+		.gettime64	= cavium_ptp_gettime,
+		.settime64	= cavium_ptp_settime,
+		.enable		= cavium_ptp_enable,
+	};
+
+	clock_cfg = readq(clock->reg_base + PTP_CLOCK_CFG);
+	clock_cfg |= PTP_CLOCK_CFG_PTP_EN;
+	writeq(clock_cfg, clock->reg_base + PTP_CLOCK_CFG);
+
+	clock_comp = ((u64)1000000000ull << 32) / clock->clock_rate;
+	writeq(clock_comp, clock->reg_base + PTP_CLOCK_COMP);
+
+	clock->ptp_clock = ptp_clock_register(&clock->ptp_info, dev);
+	if (IS_ERR(clock->ptp_clock)) {
+		err = PTR_ERR(clock->ptp_clock);
+		goto error_stop;
+	}
+
+	pci_set_drvdata(pdev, clock);
+	return 0;
+
+error_stop:
+	clock_cfg = readq(clock->reg_base + PTP_CLOCK_CFG);
+	clock_cfg &= ~PTP_CLOCK_CFG_PTP_EN;
+	writeq(clock_cfg, clock->reg_base + PTP_CLOCK_CFG);
+	pcim_iounmap_region(pdev, PCI_PTP_BAR_NO);
+
+error_free:
+	devm_kfree(dev, clock);
+
+error:
+	/* For `cavium_ptp_get()` we need to differentiate between the case
+	 * when the core has not tried to probe this device and the case when
+	 * the probe failed.  In the later case we pretend that the
+	 * initialization was successful and keep the error in
+	 * `dev->driver_data`.
+	 */
+	pci_set_drvdata(pdev, ERR_PTR(err));
+	return 0;
+}
+
+static void cavium_ptp_remove(struct pci_dev *pdev)
+{
+	struct cavium_ptp *clock = pci_get_drvdata(pdev);
+	u64 clock_cfg;
+
+	if (IS_ERR_OR_NULL(clock))
+		return;
+
+	ptp_clock_unregister(clock->ptp_clock);
+
+	clock_cfg = readq(clock->reg_base + PTP_CLOCK_CFG);
+	clock_cfg &= ~PTP_CLOCK_CFG_PTP_EN;
+	writeq(clock_cfg, clock->reg_base + PTP_CLOCK_CFG);
+}
+
+static const struct pci_device_id cavium_ptp_id_table[] = {
+	{ PCI_DEVICE_SUB(PCI_VENDOR_ID_CAVIUM, PCI_DEVICE_ID_CAVIUM_PTP,
+			PCI_VENDOR_ID_CAVIUM, PCI_SUBSYS_DEVID_88XX_PTP) },
+	{ PCI_DEVICE_SUB(PCI_VENDOR_ID_CAVIUM, PCI_DEVICE_ID_CAVIUM_PTP,
+			PCI_VENDOR_ID_CAVIUM, PCI_SUBSYS_DEVID_81XX_PTP) },
+	{ PCI_DEVICE_SUB(PCI_VENDOR_ID_CAVIUM, PCI_DEVICE_ID_CAVIUM_PTP,
+			PCI_VENDOR_ID_CAVIUM, PCI_SUBSYS_DEVID_83XX_PTP) },
+	{ 0, }
+};
+
+static struct pci_driver cavium_ptp_driver = {
+	.name = DRV_NAME,
+	.id_table = cavium_ptp_id_table,
+	.probe = cavium_ptp_probe,
+	.remove = cavium_ptp_remove,
+};
+
+module_pci_driver(cavium_ptp_driver);
+
+MODULE_DESCRIPTION(DRV_NAME);
+MODULE_AUTHOR("Cavium Networks <support@cavium.com>");
+MODULE_LICENSE("GPL v2");
+MODULE_DEVICE_TABLE(pci, cavium_ptp_id_table);
diff --git a/drivers/net/ethernet/cavium/common/cavium_ptp.h b/drivers/net/ethernet/cavium/common/cavium_ptp.h
new file mode 100644
index 000000000000..1e0ffe8f4152
--- /dev/null
+++ b/drivers/net/ethernet/cavium/common/cavium_ptp.h
@@ -0,0 +1,70 @@
+/* SPDX-License-Identifier: GPL-2.0 */
+/* cavium_ptp.h - PTP 1588 clock on Cavium hardware
+ * Copyright (c) 2003-2015, 2017 Cavium, Inc.
+ */
+
+#ifndef CAVIUM_PTP_H
+#define CAVIUM_PTP_H
+
+#include <linux/ptp_clock_kernel.h>
+#include <linux/timecounter.h>
+
+struct cavium_ptp {
+	struct pci_dev *pdev;
+
+	/* Serialize access to cycle_counter, time_counter and hw_registers */
+	spinlock_t spin_lock;
+	struct cyclecounter cycle_counter;
+	struct timecounter time_counter;
+	void __iomem *reg_base;
+
+	u32 clock_rate;
+
+	struct ptp_clock_info ptp_info;
+	struct ptp_clock *ptp_clock;
+};
+
+#if IS_REACHABLE(CONFIG_CAVIUM_PTP)
+
+struct cavium_ptp *cavium_ptp_get(void);
+void cavium_ptp_put(struct cavium_ptp *ptp);
+
+static inline u64 cavium_ptp_tstamp2time(struct cavium_ptp *ptp, u64 tstamp)
+{
+	unsigned long flags;
+	u64 ret;
+
+	spin_lock_irqsave(&ptp->spin_lock, flags);
+	ret = timecounter_cyc2time(&ptp->time_counter, tstamp);
+	spin_unlock_irqrestore(&ptp->spin_lock, flags);
+
+	return ret;
+}
+
+static inline int cavium_ptp_clock_index(struct cavium_ptp *clock)
+{
+	return ptp_clock_index(clock->ptp_clock);
+}
+
+#else
+
+static inline struct cavium_ptp *cavium_ptp_get(void)
+{
+	return ERR_PTR(-ENODEV);
+}
+
+static inline void cavium_ptp_put(struct cavium_ptp *ptp) {}
+
+static inline u64 cavium_ptp_tstamp2time(struct cavium_ptp *ptp, u64 tstamp)
+{
+	return 0;
+}
+
+static inline int cavium_ptp_clock_index(struct cavium_ptp *clock)
+{
+	return -1;
+}
+
+#endif
+
+#endif
diff --git a/drivers/net/ethernet/cavium/liquidio/Makefile b/drivers/net/ethernet/cavium/liquidio/Makefile
index c4d411d1aa28..4ee80af88e79 100644
--- a/drivers/net/ethernet/cavium/liquidio/Makefile
+++ b/drivers/net/ethernet/cavium/liquidio/Makefile
@@ -1,42 +1,26 @@
+# SPDX-License-Identifier: GPL-2.0
 #
 # Cavium Liquidio ethernet device driver
 #
-obj-$(CONFIG_LIQUIDIO) += liquidio.o
 
-liquidio-$(CONFIG_LIQUIDIO) += lio_ethtool.o \
-			lio_core.o         \
-			request_manager.o  \
-			response_manager.o \
-			octeon_device.o    \
-			cn66xx_device.o    \
-			cn68xx_device.o    \
-			cn23xx_pf_device.o \
-			cn23xx_vf_device.o \
-			octeon_mailbox.o   \
-			octeon_mem_ops.o   \
-			octeon_droq.o      \
-			octeon_nic.o
+obj-$(CONFIG_LIQUIDIO_CORE) += liquidio-core.o
+liquidio-core-y := \
+		lio_ethtool.o		\
+		lio_core.o		\
+		request_manager.o	\
+		response_manager.o	\
+		octeon_device.o		\
+		cn66xx_device.o		\
+		cn68xx_device.o		\
+		cn23xx_pf_device.o	\
+		cn23xx_vf_device.o	\
+		octeon_mailbox.o	\
+		octeon_mem_ops.o	\
+		octeon_droq.o		\
+		octeon_nic.o
 
-liquidio-objs := lio_main.o octeon_console.o $(liquidio-y)
+obj-$(CONFIG_LIQUIDIO) += liquidio.o
+liquidio-y := lio_main.o octeon_console.o lio_vf_rep.o
 
 obj-$(CONFIG_LIQUIDIO_VF) += liquidio_vf.o
-
-ifeq ($(CONFIG_LIQUIDIO)$(CONFIG_LIQUIDIO_VF), yy)
-	liquidio_vf-objs := lio_vf_main.o
-else
-liquidio_vf-$(CONFIG_LIQUIDIO_VF) += lio_ethtool.o \
-			lio_core.o         \
-			request_manager.o  \
-			response_manager.o \
-			octeon_device.o    \
-			cn66xx_device.o    \
-			cn68xx_device.o    \
-			cn23xx_pf_device.o \
-			cn23xx_vf_device.o \
-			octeon_mailbox.o   \
-			octeon_mem_ops.o   \
-			octeon_droq.o      \
-			octeon_nic.o
-
-liquidio_vf-objs := lio_vf_main.o $(liquidio_vf-y)
-endif
+liquidio_vf-y := lio_vf_main.o
diff --git a/drivers/net/ethernet/cavium/liquidio/cn23xx_pf_device.c b/drivers/net/ethernet/cavium/liquidio/cn23xx_pf_device.c
index 962dcbcef8b5..75f22f74774c 100644
--- a/drivers/net/ethernet/cavium/liquidio/cn23xx_pf_device.c
+++ b/drivers/net/ethernet/cavium/liquidio/cn23xx_pf_device.c
@@ -36,175 +36,6 @@
  */
 #define CN23XX_INPUT_JABBER 64600
 
-void cn23xx_dump_pf_initialized_regs(struct octeon_device *oct)
-{
-	int i = 0;
-	u32 regval = 0;
-	struct octeon_cn23xx_pf *cn23xx = (struct octeon_cn23xx_pf *)oct->chip;
-
-	/*In cn23xx_soft_reset*/
-	dev_dbg(&oct->pci_dev->dev, "%s[%llx] : 0x%llx\n",
-		"CN23XX_WIN_WR_MASK_REG", CVM_CAST64(CN23XX_WIN_WR_MASK_REG),
-		CVM_CAST64(octeon_read_csr64(oct, CN23XX_WIN_WR_MASK_REG)));
-	dev_dbg(&oct->pci_dev->dev, "%s[%llx] : 0x%016llx\n",
-		"CN23XX_SLI_SCRATCH1", CVM_CAST64(CN23XX_SLI_SCRATCH1),
-		CVM_CAST64(octeon_read_csr64(oct, CN23XX_SLI_SCRATCH1)));
-	dev_dbg(&oct->pci_dev->dev, "%s[%llx] : 0x%016llx\n",
-		"CN23XX_RST_SOFT_RST", CN23XX_RST_SOFT_RST,
-		lio_pci_readq(oct, CN23XX_RST_SOFT_RST));
-
-	/*In cn23xx_set_dpi_regs*/
-	dev_dbg(&oct->pci_dev->dev, "%s[%llx] : 0x%016llx\n",
-		"CN23XX_DPI_DMA_CONTROL", CN23XX_DPI_DMA_CONTROL,
-		lio_pci_readq(oct, CN23XX_DPI_DMA_CONTROL));
-
-	for (i = 0; i < 6; i++) {
-		dev_dbg(&oct->pci_dev->dev, "%s(%d)[%llx] : 0x%016llx\n",
-			"CN23XX_DPI_DMA_ENG_ENB", i,
-			CN23XX_DPI_DMA_ENG_ENB(i),
-			lio_pci_readq(oct, CN23XX_DPI_DMA_ENG_ENB(i)));
-		dev_dbg(&oct->pci_dev->dev, "%s(%d)[%llx] : 0x%016llx\n",
-			"CN23XX_DPI_DMA_ENG_BUF", i,
-			CN23XX_DPI_DMA_ENG_BUF(i),
-			lio_pci_readq(oct, CN23XX_DPI_DMA_ENG_BUF(i)));
-	}
-
-	dev_dbg(&oct->pci_dev->dev, "%s[%llx] : 0x%016llx\n", "CN23XX_DPI_CTL",
-		CN23XX_DPI_CTL, lio_pci_readq(oct, CN23XX_DPI_CTL));
-
-	/*In cn23xx_setup_pcie_mps and cn23xx_setup_pcie_mrrs */
-	pci_read_config_dword(oct->pci_dev, CN23XX_CONFIG_PCIE_DEVCTL, &regval);
-	dev_dbg(&oct->pci_dev->dev, "%s[%llx] : 0x%016llx\n",
-		"CN23XX_CONFIG_PCIE_DEVCTL",
-		CVM_CAST64(CN23XX_CONFIG_PCIE_DEVCTL), CVM_CAST64(regval));
-
-	dev_dbg(&oct->pci_dev->dev, "%s(%d)[%llx] : 0x%016llx\n",
-		"CN23XX_DPI_SLI_PRTX_CFG", oct->pcie_port,
-		CN23XX_DPI_SLI_PRTX_CFG(oct->pcie_port),
-		lio_pci_readq(oct, CN23XX_DPI_SLI_PRTX_CFG(oct->pcie_port)));
-
-	/*In cn23xx_specific_regs_setup */
-	dev_dbg(&oct->pci_dev->dev, "%s(%d)[%llx] : 0x%016llx\n",
-		"CN23XX_SLI_S2M_PORTX_CTL", oct->pcie_port,
-		CVM_CAST64(CN23XX_SLI_S2M_PORTX_CTL(oct->pcie_port)),
-		CVM_CAST64(octeon_read_csr64(
-			oct, CN23XX_SLI_S2M_PORTX_CTL(oct->pcie_port))));
-
-	dev_dbg(&oct->pci_dev->dev, "%s[%llx] : 0x%016llx\n",
-		"CN23XX_SLI_RING_RST", CVM_CAST64(CN23XX_SLI_PKT_IOQ_RING_RST),
-		(u64)octeon_read_csr64(oct, CN23XX_SLI_PKT_IOQ_RING_RST));
-
-	/*In cn23xx_setup_global_mac_regs*/
-	for (i = 0; i < CN23XX_MAX_MACS; i++) {
-		dev_dbg(&oct->pci_dev->dev, "%s(%d)[%llx] : 0x%016llx\n",
-			"CN23XX_SLI_PKT_MAC_RINFO64", i,
-			CVM_CAST64(CN23XX_SLI_PKT_MAC_RINFO64(i, oct->pf_num)),
-			CVM_CAST64(octeon_read_csr64
-				(oct, CN23XX_SLI_PKT_MAC_RINFO64
-					(i, oct->pf_num))));
-	}
-
-	/*In cn23xx_setup_global_input_regs*/
-	for (i = 0; i < CN23XX_MAX_INPUT_QUEUES; i++) {
-		dev_dbg(&oct->pci_dev->dev, "%s(%d)[%llx] : 0x%016llx\n",
-			"CN23XX_SLI_IQ_PKT_CONTROL64", i,
-			CVM_CAST64(CN23XX_SLI_IQ_PKT_CONTROL64(i)),
-			CVM_CAST64(octeon_read_csr64
-				(oct, CN23XX_SLI_IQ_PKT_CONTROL64(i))));
-	}
-
-	/*In cn23xx_setup_global_output_regs*/
-	dev_dbg(&oct->pci_dev->dev, "%s[%llx] : 0x%016llx\n",
-		"CN23XX_SLI_OQ_WMARK", CVM_CAST64(CN23XX_SLI_OQ_WMARK),
-		CVM_CAST64(octeon_read_csr64(oct, CN23XX_SLI_OQ_WMARK)));
-
-	for (i = 0; i < CN23XX_MAX_OUTPUT_QUEUES; i++) {
-		dev_dbg(&oct->pci_dev->dev, "%s(%d)[%llx] : 0x%016llx\n",
-			"CN23XX_SLI_OQ_PKT_CONTROL", i,
-			CVM_CAST64(CN23XX_SLI_OQ_PKT_CONTROL(i)),
-			CVM_CAST64(octeon_read_csr(
-				oct, CN23XX_SLI_OQ_PKT_CONTROL(i))));
-		dev_dbg(&oct->pci_dev->dev, "%s(%d)[%llx] : 0x%016llx\n",
-			"CN23XX_SLI_OQ_PKT_INT_LEVELS", i,
-			CVM_CAST64(CN23XX_SLI_OQ_PKT_INT_LEVELS(i)),
-			CVM_CAST64(octeon_read_csr64(
-				oct, CN23XX_SLI_OQ_PKT_INT_LEVELS(i))));
-	}
-
-	/*In cn23xx_enable_interrupt and cn23xx_disable_interrupt*/
-	dev_dbg(&oct->pci_dev->dev, "%s[%llx] : 0x%016llx\n",
-		"cn23xx->intr_enb_reg64",
-		CVM_CAST64((long)(cn23xx->intr_enb_reg64)),
-		CVM_CAST64(readq(cn23xx->intr_enb_reg64)));
-
-	dev_dbg(&oct->pci_dev->dev, "%s[%llx] : 0x%016llx\n",
-		"cn23xx->intr_sum_reg64",
-		CVM_CAST64((long)(cn23xx->intr_sum_reg64)),
-		CVM_CAST64(readq(cn23xx->intr_sum_reg64)));
-
-	/*In cn23xx_setup_iq_regs*/
-	for (i = 0; i < CN23XX_MAX_INPUT_QUEUES; i++) {
-		dev_dbg(&oct->pci_dev->dev, "%s(%d)[%llx] : 0x%016llx\n",
-			"CN23XX_SLI_IQ_BASE_ADDR64", i,
-			CVM_CAST64(CN23XX_SLI_IQ_BASE_ADDR64(i)),
-			CVM_CAST64(octeon_read_csr64(
-				oct, CN23XX_SLI_IQ_BASE_ADDR64(i))));
-		dev_dbg(&oct->pci_dev->dev, "%s(%d)[%llx] : 0x%016llx\n",
-			"CN23XX_SLI_IQ_SIZE", i,
-			CVM_CAST64(CN23XX_SLI_IQ_SIZE(i)),
-			CVM_CAST64(octeon_read_csr
-				(oct, CN23XX_SLI_IQ_SIZE(i))));
-		dev_dbg(&oct->pci_dev->dev, "%s(%d)[%llx] : 0x%016llx\n",
-			"CN23XX_SLI_IQ_DOORBELL", i,
-			CVM_CAST64(CN23XX_SLI_IQ_DOORBELL(i)),
-			CVM_CAST64(octeon_read_csr64(
-				oct, CN23XX_SLI_IQ_DOORBELL(i))));
-		dev_dbg(&oct->pci_dev->dev, "%s(%d)[%llx] : 0x%016llx\n",
-			"CN23XX_SLI_IQ_INSTR_COUNT64", i,
-			CVM_CAST64(CN23XX_SLI_IQ_INSTR_COUNT64(i)),
-			CVM_CAST64(octeon_read_csr64(
-				oct, CN23XX_SLI_IQ_INSTR_COUNT64(i))));
-	}
-
-	/*In cn23xx_setup_oq_regs*/
-	for (i = 0; i < CN23XX_MAX_OUTPUT_QUEUES; i++) {
-		dev_dbg(&oct->pci_dev->dev, "%s(%d)[%llx] : 0x%016llx\n",
-			"CN23XX_SLI_OQ_BASE_ADDR64", i,
-			CVM_CAST64(CN23XX_SLI_OQ_BASE_ADDR64(i)),
-			CVM_CAST64(octeon_read_csr64(
-				oct, CN23XX_SLI_OQ_BASE_ADDR64(i))));
-		dev_dbg(&oct->pci_dev->dev, "%s(%d)[%llx] : 0x%016llx\n",
-			"CN23XX_SLI_OQ_SIZE", i,
-			CVM_CAST64(CN23XX_SLI_OQ_SIZE(i)),
-			CVM_CAST64(octeon_read_csr
-				(oct, CN23XX_SLI_OQ_SIZE(i))));
-		dev_dbg(&oct->pci_dev->dev, "%s(%d)[%llx] : 0x%016llx\n",
-			"CN23XX_SLI_OQ_BUFF_INFO_SIZE", i,
-			CVM_CAST64(CN23XX_SLI_OQ_BUFF_INFO_SIZE(i)),
-			CVM_CAST64(octeon_read_csr(
-				oct, CN23XX_SLI_OQ_BUFF_INFO_SIZE(i))));
-		dev_dbg(&oct->pci_dev->dev, "%s(%d)[%llx] : 0x%016llx\n",
-			"CN23XX_SLI_OQ_PKTS_SENT", i,
-			CVM_CAST64(CN23XX_SLI_OQ_PKTS_SENT(i)),
-			CVM_CAST64(octeon_read_csr64(
-				oct, CN23XX_SLI_OQ_PKTS_SENT(i))));
-		dev_dbg(&oct->pci_dev->dev, "%s(%d)[%llx] : 0x%016llx\n",
-			"CN23XX_SLI_OQ_PKTS_CREDIT", i,
-			CVM_CAST64(CN23XX_SLI_OQ_PKTS_CREDIT(i)),
-			CVM_CAST64(octeon_read_csr64(
-				oct, CN23XX_SLI_OQ_PKTS_CREDIT(i))));
-	}
-
-	dev_dbg(&oct->pci_dev->dev, "%s[%llx] : 0x%016llx\n",
-		"CN23XX_SLI_PKT_TIME_INT",
-		CVM_CAST64(CN23XX_SLI_PKT_TIME_INT),
-		CVM_CAST64(octeon_read_csr64(oct, CN23XX_SLI_PKT_TIME_INT)));
-	dev_dbg(&oct->pci_dev->dev, "%s[%llx] : 0x%016llx\n",
-		"CN23XX_SLI_PKT_CNT_INT",
-		CVM_CAST64(CN23XX_SLI_PKT_CNT_INT),
-		CVM_CAST64(octeon_read_csr64(oct, CN23XX_SLI_PKT_CNT_INT)));
-}
-
 static int cn23xx_pf_soft_reset(struct octeon_device *oct)
 {
 	octeon_write_csr64(oct, CN23XX_WIN_WR_MASK_REG, 0xFF);
@@ -218,10 +49,10 @@ static int cn23xx_pf_soft_reset(struct octeon_device *oct)
 	lio_pci_readq(oct, CN23XX_RST_SOFT_RST);
 	lio_pci_writeq(oct, 1, CN23XX_RST_SOFT_RST);
 
-	/* Wait for 100ms as Octeon resets. */
+	/* Wait for 100ms as Octeon resets */
 	mdelay(100);
 
-	if (octeon_read_csr64(oct, CN23XX_SLI_SCRATCH1) == 0x1234ULL) {
+	if (octeon_read_csr64(oct, CN23XX_SLI_SCRATCH1)) {
 		dev_err(&oct->pci_dev->dev, "OCTEON[%d]: Soft reset failed\n",
 			oct->octeon_id);
 		return 1;
@@ -230,7 +61,7 @@ static int cn23xx_pf_soft_reset(struct octeon_device *oct)
 	dev_dbg(&oct->pci_dev->dev, "OCTEON[%d]: Reset completed\n",
 		oct->octeon_id);
 
-	/* restore the  reset value*/
+	/* Restore the reset value */
 	octeon_write_csr64(oct, CN23XX_WIN_WR_MASK_REG, 0xFF);
 
 	return 0;
@@ -290,7 +121,7 @@ u32 cn23xx_pf_get_oq_ticks(struct octeon_device *oct, u32 time_intr_in_us)
 	oqticks_per_us /= 1024;
 
 	/* time_intr is in microseconds. The next 2 steps gives the oq ticks
-	 *  corressponding to time_intr.
+	 * corresponding to time_intr.
 	 */
 	oqticks_per_us *= time_intr_in_us;
 	oqticks_per_us /= 1000;
@@ -305,11 +136,11 @@ static void cn23xx_setup_global_mac_regs(struct octeon_device *oct)
 	u64 reg_val;
 	u64 temp;
 
-	/* programming SRN and TRS for each MAC(0..3)  */
+	/* Programming SRN and TRS for each MAC(0..3) */
 
 	dev_dbg(&oct->pci_dev->dev, "%s:Using pcie port %d\n",
 		__func__, mac_no);
-	/* By default, mapping all 64 IOQs to  a single MACs */
+	/* By default, map all 64 IOQs to a single MAC */
 
 	reg_val =
 	    octeon_read_csr64(oct, CN23XX_SLI_PKT_MAC_RINFO64(mac_no, pf_num));
@@ -333,7 +164,7 @@ static void cn23xx_setup_global_mac_regs(struct octeon_device *oct)
 	temp = oct->sriov_info.max_vfs & 0xff;
 	reg_val |= (temp << CN23XX_PKT_MAC_CTL_RINFO_NVFS_BIT_POS);
 
-	/* write these settings to MAC register */
+	/* Write these settings to MAC register */
 	octeon_write_csr64(oct, CN23XX_SLI_PKT_MAC_RINFO64(mac_no, pf_num),
 			   reg_val);
 
@@ -352,10 +183,10 @@ static int cn23xx_reset_io_queues(struct octeon_device *oct)
 	srn = oct->sriov_info.pf_srn;
 	ern = srn + oct->sriov_info.num_pf_rings;
 
-	/*As per HRM reg description, s/w cant write 0 to ENB. */
-	/*to make the queue off, need to set the RST bit. */
+	/* As per HRM reg description, s/w can't write 0 to ENB. */
+	/* We need to set the RST bit, to turn the queue off. */
 
-	/* Reset the Enable bit for all the 64 IQs.  */
+	/* Reset the enable bit for all the 64 IQs. */
 	for (q_no = srn; q_no < ern; q_no++) {
 		/* set RST bit to 1. This bit applies to both IQ and OQ */
 		d64 = octeon_read_csr64(oct, CN23XX_SLI_IQ_PKT_CONTROL64(q_no));
@@ -363,7 +194,7 @@ static int cn23xx_reset_io_queues(struct octeon_device *oct)
 		octeon_write_csr64(oct, CN23XX_SLI_IQ_PKT_CONTROL64(q_no), d64);
 	}
 
-	/*wait until the RST bit is clear or the RST and quite bits are set*/
+	/* Wait until the RST bit is clear or the RST and quiet bits are set */
 	for (q_no = srn; q_no < ern; q_no++) {
 		u64 reg_val = octeon_read_csr64(oct,
 					CN23XX_SLI_IQ_PKT_CONTROL64(q_no));
@@ -414,15 +245,15 @@ static int cn23xx_pf_setup_global_input_regs(struct octeon_device *oct)
 	if (cn23xx_reset_io_queues(oct))
 		return -1;
 
-	/** Set the MAC_NUM and PVF_NUM in IQ_PKT_CONTROL reg
-	 * for all queues.Only PF can set these bits.
+	/* Set the MAC_NUM and PVF_NUM in IQ_PKT_CONTROL reg
+	 * for all queues. Only PF can set these bits.
 	 * bits 29:30 indicate the MAC num.
 	 * bits 32:47 indicate the PVF num.
 	 */
 	for (q_no = 0; q_no < ern; q_no++) {
-		reg_val = oct->pcie_port << CN23XX_PKT_INPUT_CTL_MAC_NUM_POS;
+		reg_val = (u64)oct->pcie_port << CN23XX_PKT_INPUT_CTL_MAC_NUM_POS;
 
-		/* for VF assigned queues. */
+		/* For VF assigned queues. */
 		if (q_no < oct->sriov_info.pf_srn) {
 			vf_num = q_no / oct->sriov_info.rings_per_vf;
 			vf_num += 1; /* VF1, VF2,........ */
@@ -437,7 +268,7 @@ static int cn23xx_pf_setup_global_input_regs(struct octeon_device *oct)
 				   reg_val);
 	}
 
-	/* Select ES, RO, NS, RDSIZE,DPTR Fomat#0 for
+	/* Select ES, RO, NS, RDSIZE,DPTR Format#0 for
 	 * pf queues
 	 */
 	for (q_no = srn; q_no < ern; q_no++) {
@@ -458,7 +289,7 @@ static int cn23xx_pf_setup_global_input_regs(struct octeon_device *oct)
 		octeon_write_csr64(oct, CN23XX_SLI_IQ_PKT_CONTROL64(q_no),
 				   reg_val);
 
-		/* Set WMARK level for triggering PI_INT */
+		/* Set WMARK level to trigger PI_INT */
 		/* intr_threshold = CN23XX_DEF_IQ_INTR_THRESHOLD & */
 		intr_threshold = CFG_GET_IQ_INTR_PKT(cn23xx->conf) &
 				 CN23XX_PKT_IN_DONE_WMARK_MASK;
@@ -493,9 +324,11 @@ static void cn23xx_pf_setup_global_output_regs(struct octeon_device *oct)
 	for (q_no = srn; q_no < ern; q_no++) {
 		reg_val = octeon_read_csr(oct, CN23XX_SLI_OQ_PKT_CONTROL(q_no));
 
-		/* set IPTR & DPTR */
-		reg_val |=
-		    (CN23XX_PKT_OUTPUT_CTL_IPTR | CN23XX_PKT_OUTPUT_CTL_DPTR);
+		/* clear IPTR */
+		reg_val &= ~CN23XX_PKT_OUTPUT_CTL_IPTR;
+
+		/* set DPTR */
+		reg_val |= CN23XX_PKT_OUTPUT_CTL_DPTR;
 
 		/* reset BMODE */
 		reg_val &= ~(CN23XX_PKT_OUTPUT_CTL_BMODE);
@@ -521,7 +354,7 @@ static void cn23xx_pf_setup_global_output_regs(struct octeon_device *oct)
 		/* set the ES bit */
 		reg_val |= (CN23XX_PKT_OUTPUT_CTL_ES);
 
-		/* write all the selected settings */
+		/* Write all the selected settings */
 		octeon_write_csr(oct, CN23XX_SLI_OQ_PKT_CONTROL(q_no), reg_val);
 
 		/* Enabling these interrupt in oct->fn_list.enable_interrupt()
@@ -540,7 +373,7 @@ static void cn23xx_pf_setup_global_output_regs(struct octeon_device *oct)
 	/** Setting the water mark level for pko back pressure **/
 	writeq(0x40, (u8 *)oct->mmio[0].hw_addr + CN23XX_SLI_OQ_WMARK);
 
-	/** Disabling setting OQs in reset when ring has no dorebells
+	/* Disabling setting OQs in reset when ring has no doorbells
 	 * enabling this will cause of head of line blocking
 	 */
 	/* Do it only for pass1.1. and pass1.2 */
@@ -550,7 +383,7 @@ static void cn23xx_pf_setup_global_output_regs(struct octeon_device *oct)
 				     CN23XX_SLI_GBL_CONTROL) | 0x2,
 		       (u8 *)oct->mmio[0].hw_addr + CN23XX_SLI_GBL_CONTROL);
 
-	/** Enable channel-level backpressure */
+	/** Enable channel-level backpressure **/
 	if (oct->pf_num)
 		writeq(0xffffffffffffffffULL,
 		       (u8 *)oct->mmio[0].hw_addr + CN23XX_SLI_OUT_BP_EN2_W1S);
@@ -563,7 +396,7 @@ static int cn23xx_setup_pf_device_regs(struct octeon_device *oct)
 {
 	cn23xx_enable_error_reporting(oct);
 
-	/* program the MAC(0..3)_RINFO before setting up input/output regs */
+	/* Program the MAC(0..3)_RINFO before setting up input/output regs */
 	cn23xx_setup_global_mac_regs(oct);
 
 	if (cn23xx_pf_setup_global_input_regs(oct))
@@ -577,7 +410,7 @@ static int cn23xx_setup_pf_device_regs(struct octeon_device *oct)
 	octeon_write_csr64(oct, CN23XX_SLI_WINDOW_CTL,
 			   CN23XX_SLI_WINDOW_CTL_DEFAULT);
 
-	/* set SLI_PKT_IN_JABBER to handle large VXLAN packets */
+	/* Set SLI_PKT_IN_JABBER to handle large VXLAN packets */
 	octeon_write_csr64(oct, CN23XX_SLI_PKT_IN_JABBER, CN23XX_INPUT_JABBER);
 	return 0;
 }
@@ -638,7 +471,7 @@ static void cn23xx_setup_oq_regs(struct octeon_device *oct, u32 oq_no)
 	octeon_write_csr(oct, CN23XX_SLI_OQ_SIZE(oq_no), droq->max_count);
 
 	octeon_write_csr(oct, CN23XX_SLI_OQ_BUFF_INFO_SIZE(oq_no),
-			 (droq->buffer_size | (OCT_RH_SIZE << 16)));
+			 droq->buffer_size);
 
 	/* Get the mapped address of the pkt_sent and pkts_credit regs */
 	droq->pkts_sent_reg =
@@ -717,12 +550,10 @@ static int cn23xx_setup_pf_mbox(struct octeon_device *oct)
 	for (i = 0; i < oct->sriov_info.max_vfs; i++) {
 		q_no = i * oct->sriov_info.rings_per_vf;
 
-		mbox = vmalloc(sizeof(*mbox));
+		mbox = vzalloc(sizeof(*mbox));
 		if (!mbox)
 			goto free_mbox;
 
-		memset(mbox, 0, sizeof(struct octeon_mbox));
-
 		spin_lock_init(&mbox->lock);
 
 		mbox->oct_dev = oct;
@@ -743,7 +574,7 @@ static int cn23xx_setup_pf_mbox(struct octeon_device *oct)
 		mbox->mbox_read_reg = (u8 *)oct->mmio[0].hw_addr +
 				      CN23XX_SLI_PKT_PF_VF_MBOX_SIG(q_no, 1);
 
-		/*Mail Box Thread creation*/
+		/* Mail Box Thread creation */
 		INIT_DELAYED_WORK(&mbox->mbox_poll_wk.work,
 				  cn23xx_pf_mbox_thread);
 		mbox->mbox_poll_wk.ctxptr = (void *)mbox;
@@ -795,7 +626,7 @@ static int cn23xx_enable_io_queues(struct octeon_device *oct)
 	ern = srn + oct->num_iqs;
 
 	for (q_no = srn; q_no < ern; q_no++) {
-		/* set the corresponding IQ IS_64B bit */
+		/* Set the corresponding IQ IS_64B bit */
 		if (oct->io_qmask.iq64B & BIT_ULL(q_no - srn)) {
 			reg_val = octeon_read_csr64(
 			    oct, CN23XX_SLI_IQ_PKT_CONTROL64(q_no));
@@ -804,7 +635,7 @@ static int cn23xx_enable_io_queues(struct octeon_device *oct)
 			    oct, CN23XX_SLI_IQ_PKT_CONTROL64(q_no), reg_val);
 		}
 
-		/* set the corresponding IQ ENB bit */
+		/* Set the corresponding IQ ENB bit */
 		if (oct->io_qmask.iq & BIT_ULL(q_no - srn)) {
 			/* IOQs are in reset by default in PEM2 mode,
 			 * clearing reset bit
@@ -850,7 +681,7 @@ static int cn23xx_enable_io_queues(struct octeon_device *oct)
 	}
 	for (q_no = srn; q_no < ern; q_no++) {
 		u32 reg_val;
-		/* set the corresponding OQ ENB bit */
+		/* Set the corresponding OQ ENB bit */
 		if (oct->io_qmask.oq & BIT_ULL(q_no - srn)) {
 			reg_val = octeon_read_csr(
 			    oct, CN23XX_SLI_OQ_PKT_CONTROL(q_no));
@@ -876,7 +707,7 @@ static void cn23xx_disable_io_queues(struct octeon_device *oct)
 	for (q_no = srn; q_no < ern; q_no++) {
 		loop = HZ;
 
-		/* start the Reset for a particular ring */
+		/* Start the Reset for a particular ring */
 		WRITE_ONCE(d64, octeon_read_csr64(
 			   oct, CN23XX_SLI_IQ_PKT_CONTROL64(q_no)));
 		WRITE_ONCE(d64, READ_ONCE(d64) &
@@ -909,7 +740,7 @@ static void cn23xx_disable_io_queues(struct octeon_device *oct)
 		loop = HZ;
 
 		/* Wait until hardware indicates that the particular IQ
-		 * is out of reset.It given that SLI_PKT_RING_RST is
+		 * is out of reset. Given that SLI_PKT_RING_RST is
 		 * common for both IQs and OQs
 		 */
 		WRITE_ONCE(d64, octeon_read_csr64(
@@ -929,7 +760,7 @@ static void cn23xx_disable_io_queues(struct octeon_device *oct)
 			schedule_timeout_uninterruptible(1);
 		}
 
-		/* clear the SLI_PKT(0..63)_CNTS[CNT] reg value */
+		/* Clear the SLI_PKT(0..63)_CNTS[CNT] reg value */
 		WRITE_ONCE(d32, octeon_read_csr(
 					oct, CN23XX_SLI_OQ_PKTS_SENT(q_no)));
 		octeon_write_csr(oct, CN23XX_SLI_OQ_PKTS_SENT(q_no),
@@ -962,7 +793,7 @@ static u64 cn23xx_pf_msix_interrupt_handler(void *dev)
 	if (!pkts_sent || (pkts_sent == 0xFFFFFFFFFFFFFFFFULL))
 		return ret;
 
-	/* Write count reg in sli_pkt_cnts to clear these int.*/
+	/* Write count reg in sli_pkt_cnts to clear these int. */
 	if ((pkts_sent & CN23XX_INTR_PO_INT) ||
 	    (pkts_sent & CN23XX_INTR_PI_INT)) {
 		if (pkts_sent & CN23XX_INTR_PO_INT)
@@ -1077,7 +908,7 @@ static u32 cn23xx_bar1_idx_read(struct octeon_device *oct, u32 idx)
 	    oct, CN23XX_PEM_BAR1_INDEX_REG(oct->pcie_port, idx));
 }
 
-/* always call with lock held */
+/* Always call with lock held */
 static u32 cn23xx_update_read_index(struct octeon_instr_queue *iq)
 {
 	u32 new_idx;
@@ -1088,7 +919,7 @@ static u32 cn23xx_update_read_index(struct octeon_instr_queue *iq)
 	iq->pkt_in_done = pkt_in_done;
 
 	/* Modulo of the new index with the IQ size will give us
-	 * the new index.  The iq->reset_instr_cnt is always zero for
+	 * the new index. The iq->reset_instr_cnt is always zero for
 	 * cn23xx, so no extra adjustments are needed.
 	 */
 	new_idx = (iq->octeon_read_index +
@@ -1103,8 +934,8 @@ static void cn23xx_enable_pf_interrupt(struct octeon_device *oct, u8 intr_flag)
 	struct octeon_cn23xx_pf *cn23xx = (struct octeon_cn23xx_pf *)oct->chip;
 	u64 intr_val = 0;
 
-	/*  Divide the single write to multiple writes based on the flag. */
-	/* Enable Interrupt */
+	/* Divide the single write to multiple writes based on the flag. */
+	/* Enable Interrupts */
 	if (intr_flag == OCTEON_ALL_INTR) {
 		writeq(cn23xx->intr_mask64, cn23xx->intr_enb_reg64);
 	} else if (intr_flag & OCTEON_OUTPUT_INTR) {
@@ -1151,14 +982,50 @@ static void cn23xx_get_pcie_qlmport(struct octeon_device *oct)
 		oct->pcie_port);
 }
 
-static void cn23xx_get_pf_num(struct octeon_device *oct)
+static int cn23xx_get_pf_num(struct octeon_device *oct)
 {
 	u32 fdl_bit = 0;
+	u64 pkt0_in_ctl, d64;
+	int pfnum, mac, trs, ret;
 
-	/** Read Function Dependency Link reg to get the function number */
-	pci_read_config_dword(oct->pci_dev, CN23XX_PCIE_SRIOV_FDL, &fdl_bit);
-	oct->pf_num = ((fdl_bit >> CN23XX_PCIE_SRIOV_FDL_BIT_POS) &
-		       CN23XX_PCIE_SRIOV_FDL_MASK);
+	ret = 0;
+
+	/* Read Function Dependency Link reg to get the function number */
+	if (pci_read_config_dword(oct->pci_dev, CN23XX_PCIE_SRIOV_FDL,
+				  &fdl_bit) == 0) {
+		oct->pf_num = ((fdl_bit >> CN23XX_PCIE_SRIOV_FDL_BIT_POS) &
+			       CN23XX_PCIE_SRIOV_FDL_MASK);
+	} else {
+		ret = -EINVAL;
+
+		/* Under some virtual environments, extended PCI regs are
+		 * inaccessible, in which case the above read will have failed.
+		 * In this case, read the PF number from the
+		 * SLI_PKT0_INPUT_CONTROL reg (written by f/w)
+		 */
+		pkt0_in_ctl =
+			octeon_read_csr64(oct, CN23XX_SLI_IQ_PKT_CONTROL64(0));
+		pfnum = (pkt0_in_ctl >> CN23XX_PKT_INPUT_CTL_PF_NUM_POS) &
+			CN23XX_PKT_INPUT_CTL_PF_NUM_MASK;
+		mac = (octeon_read_csr(oct, CN23XX_SLI_MAC_NUMBER)) & 0xff;
+
+		/* Validate PF num by reading RINFO; f/w writes RINFO.trs == 1 */
+		d64 = octeon_read_csr64(oct,
+					CN23XX_SLI_PKT_MAC_RINFO64(mac, pfnum));
+		trs = (int)(d64 >> CN23XX_PKT_MAC_CTL_RINFO_TRS_BIT_POS) & 0xff;
+		if (trs == 1) {
+			dev_err(&oct->pci_dev->dev,
+				"OCTEON: error reading PCI cfg space pfnum, re-read %u\n",
+				pfnum);
+			oct->pf_num = pfnum;
+			ret = 0;
+		} else {
+			dev_err(&oct->pci_dev->dev,
+				"OCTEON: error reading PCI cfg space pfnum; could not ascertain PF number\n");
+		}
+	}
+
+	return ret;
 }
 
 static void cn23xx_setup_reg_address(struct octeon_device *oct)
@@ -1210,7 +1077,7 @@ static void cn23xx_setup_reg_address(struct octeon_device *oct)
 	    CN23XX_SLI_MAC_PF_INT_ENB64(oct->pcie_port, oct->pf_num);
 }
 
-static int cn23xx_sriov_config(struct octeon_device *oct)
+int cn23xx_sriov_config(struct octeon_device *oct)
 {
 	struct octeon_cn23xx_pf *cn23xx = (struct octeon_cn23xx_pf *)oct->chip;
 	u32 max_rings, total_rings, max_vfs, rings_per_vf;
@@ -1234,8 +1101,8 @@ static int cn23xx_sriov_config(struct octeon_device *oct)
 		break;
 	}
 
-	if (max_rings <= num_present_cpus())
-		num_pf_rings = 1;
+	if (oct->sriov_info.num_pf_rings)
+		num_pf_rings = oct->sriov_info.num_pf_rings;
 	else
 		num_pf_rings = num_present_cpus();
 
@@ -1270,6 +1137,26 @@ static int cn23xx_sriov_config(struct octeon_device *oct)
 
 int setup_cn23xx_octeon_pf_device(struct octeon_device *oct)
 {
+	u32 data32;
+	u64 BAR0, BAR1;
+
+	pci_read_config_dword(oct->pci_dev, PCI_BASE_ADDRESS_0, &data32);
+	BAR0 = (u64)(data32 & ~0xf);
+	pci_read_config_dword(oct->pci_dev, PCI_BASE_ADDRESS_1, &data32);
+	BAR0 |= ((u64)data32 << 32);
+	pci_read_config_dword(oct->pci_dev, PCI_BASE_ADDRESS_2, &data32);
+	BAR1 = (u64)(data32 & ~0xf);
+	pci_read_config_dword(oct->pci_dev, PCI_BASE_ADDRESS_3, &data32);
+	BAR1 |= ((u64)data32 << 32);
+
+	if (!BAR0 || !BAR1) {
+		if (!BAR0)
+			dev_err(&oct->pci_dev->dev, "device BAR0 unassigned\n");
+		if (!BAR1)
+			dev_err(&oct->pci_dev->dev, "device BAR1 unassigned\n");
+		return 1;
+	}
+
 	if (octeon_map_pci_barx(oct, 0, 0))
 		return 1;
 
@@ -1280,7 +1167,8 @@ int setup_cn23xx_octeon_pf_device(struct octeon_device *oct)
 		return 1;
 	}
 
-	cn23xx_get_pf_num(oct);
+	if (cn23xx_get_pf_num(oct) != 0)
+		return 1;
 
 	if (cn23xx_sriov_config(oct)) {
 		octeon_unmap_pci_barx(oct, 0);
@@ -1318,98 +1206,27 @@ int setup_cn23xx_octeon_pf_device(struct octeon_device *oct)
 
 	return 0;
 }
-
-int validate_cn23xx_pf_config_info(struct octeon_device *oct,
-				   struct octeon_config *conf23xx)
-{
-	if (CFG_GET_IQ_MAX_Q(conf23xx) > CN23XX_MAX_INPUT_QUEUES) {
-		dev_err(&oct->pci_dev->dev, "%s: Num IQ (%d) exceeds Max (%d)\n",
-			__func__, CFG_GET_IQ_MAX_Q(conf23xx),
-			CN23XX_MAX_INPUT_QUEUES);
-		return 1;
-	}
-
-	if (CFG_GET_OQ_MAX_Q(conf23xx) > CN23XX_MAX_OUTPUT_QUEUES) {
-		dev_err(&oct->pci_dev->dev, "%s: Num OQ (%d) exceeds Max (%d)\n",
-			__func__, CFG_GET_OQ_MAX_Q(conf23xx),
-			CN23XX_MAX_OUTPUT_QUEUES);
-		return 1;
-	}
-
-	if (CFG_GET_IQ_INSTR_TYPE(conf23xx) != OCTEON_32BYTE_INSTR &&
-	    CFG_GET_IQ_INSTR_TYPE(conf23xx) != OCTEON_64BYTE_INSTR) {
-		dev_err(&oct->pci_dev->dev, "%s: Invalid instr type for IQ\n",
-			__func__);
-		return 1;
-	}
-
-	if (!(CFG_GET_OQ_INFO_PTR(conf23xx)) ||
-	    !(CFG_GET_OQ_REFILL_THRESHOLD(conf23xx))) {
-		dev_err(&oct->pci_dev->dev, "%s: Invalid parameter for OQ\n",
-			__func__);
-		return 1;
-	}
-
-	if (!(CFG_GET_OQ_INTR_TIME(conf23xx))) {
-		dev_err(&oct->pci_dev->dev, "%s: Invalid parameter for OQ\n",
-			__func__);
-		return 1;
-	}
-
-	return 0;
-}
-
-void cn23xx_dump_iq_regs(struct octeon_device *oct)
-{
-	u32 regval, q_no;
-
-	dev_dbg(&oct->pci_dev->dev, "SLI_IQ_DOORBELL_0 [0x%x]: 0x%016llx\n",
-		CN23XX_SLI_IQ_DOORBELL(0),
-		CVM_CAST64(octeon_read_csr64
-			(oct, CN23XX_SLI_IQ_DOORBELL(0))));
-
-	dev_dbg(&oct->pci_dev->dev, "SLI_IQ_BASEADDR_0 [0x%x]: 0x%016llx\n",
-		CN23XX_SLI_IQ_BASE_ADDR64(0),
-		CVM_CAST64(octeon_read_csr64
-			(oct, CN23XX_SLI_IQ_BASE_ADDR64(0))));
-
-	dev_dbg(&oct->pci_dev->dev, "SLI_IQ_FIFO_RSIZE_0 [0x%x]: 0x%016llx\n",
-		CN23XX_SLI_IQ_SIZE(0),
-		CVM_CAST64(octeon_read_csr64(oct, CN23XX_SLI_IQ_SIZE(0))));
-
-	dev_dbg(&oct->pci_dev->dev, "SLI_CTL_STATUS [0x%x]: 0x%016llx\n",
-		CN23XX_SLI_CTL_STATUS,
-		CVM_CAST64(octeon_read_csr64(oct, CN23XX_SLI_CTL_STATUS)));
-
-	for (q_no = 0; q_no < CN23XX_MAX_INPUT_QUEUES; q_no++) {
-		dev_dbg(&oct->pci_dev->dev, "SLI_PKT[%d]_INPUT_CTL [0x%x]: 0x%016llx\n",
-			q_no, CN23XX_SLI_IQ_PKT_CONTROL64(q_no),
-			CVM_CAST64(octeon_read_csr64
-				(oct, CN23XX_SLI_IQ_PKT_CONTROL64(q_no))));
-	}
-
-	pci_read_config_dword(oct->pci_dev, CN23XX_CONFIG_PCIE_DEVCTL, &regval);
-	dev_dbg(&oct->pci_dev->dev, "Config DevCtl [0x%x]: 0x%08x\n",
-		CN23XX_CONFIG_PCIE_DEVCTL, regval);
-
-	dev_dbg(&oct->pci_dev->dev, "SLI_PRT[%d]_CFG [0x%llx]: 0x%016llx\n",
-		oct->pcie_port, CN23XX_DPI_SLI_PRTX_CFG(oct->pcie_port),
-		CVM_CAST64(lio_pci_readq(
-			oct, CN23XX_DPI_SLI_PRTX_CFG(oct->pcie_port))));
-
-	dev_dbg(&oct->pci_dev->dev, "SLI_S2M_PORT[%d]_CTL [0x%x]: 0x%016llx\n",
-		oct->pcie_port, CN23XX_SLI_S2M_PORTX_CTL(oct->pcie_port),
-		CVM_CAST64(octeon_read_csr64(
-			oct, CN23XX_SLI_S2M_PORTX_CTL(oct->pcie_port))));
-}
+EXPORT_SYMBOL_GPL(setup_cn23xx_octeon_pf_device);
 
 int cn23xx_fw_loaded(struct octeon_device *oct)
 {
 	u64 val;
 
-	val = octeon_read_csr64(oct, CN23XX_SLI_SCRATCH1);
-	return (val >> 1) & 1ULL;
+	/* If there's more than one active PF on this NIC, then that
+	 * implies that the NIC firmware is loaded and running. This check
+	 * prevents a rare false negative that might occur if we only relied
+	 * on checking the SCR2_BIT_FW_LOADED flag. The false negative would
+	 * happen if the PF driver sees SCR2_BIT_FW_LOADED as cleared even
+	 * though the firmware was already loaded but still booting and has yet
+	 * to set SCR2_BIT_FW_LOADED.
+	 */
+	if (atomic_read(oct->adapter_refcount) > 1)
+		return 1;
+
+	val = octeon_read_csr64(oct, CN23XX_SLI_SCRATCH2);
+	return (val >> SCR2_BIT_FW_LOADED) & 1ULL;
 }
+EXPORT_SYMBOL_GPL(cn23xx_fw_loaded);
 
 void cn23xx_tell_vf_its_macaddr_changed(struct octeon_device *oct, int vfidx,
 					u8 *mac)
@@ -1425,9 +1242,65 @@ void cn23xx_tell_vf_its_macaddr_changed(struct octeon_device *oct, int vfidx,
 		mbox_cmd.recv_len = 0;
 		mbox_cmd.recv_status = 0;
 		mbox_cmd.fn = NULL;
-		mbox_cmd.fn_arg = 0;
+		mbox_cmd.fn_arg = NULL;
 		ether_addr_copy(mbox_cmd.msg.s.params, mac);
 		mbox_cmd.q_no = vfidx * oct->sriov_info.rings_per_vf;
 		octeon_mbox_write(oct, &mbox_cmd);
 	}
 }
+EXPORT_SYMBOL_GPL(cn23xx_tell_vf_its_macaddr_changed);
+
+static void
+cn23xx_get_vf_stats_callback(struct octeon_device *oct,
+			     struct octeon_mbox_cmd *cmd, void *arg)
+{
+	struct oct_vf_stats_ctx *ctx = arg;
+
+	memcpy(ctx->stats, cmd->data, sizeof(struct oct_vf_stats));
+	atomic_set(&ctx->status, 1);
+}
+
+int cn23xx_get_vf_stats(struct octeon_device *oct, int vfidx,
+			struct oct_vf_stats *stats)
+{
+	u32 timeout = HZ; // 1sec
+	struct octeon_mbox_cmd mbox_cmd;
+	struct oct_vf_stats_ctx ctx;
+	u32 count = 0, ret;
+
+	if (!(oct->sriov_info.vf_drv_loaded_mask & (1ULL << vfidx)))
+		return -1;
+
+	if (sizeof(struct oct_vf_stats) > sizeof(mbox_cmd.data))
+		return -1;
+
+	mbox_cmd.msg.u64 = 0;
+	mbox_cmd.msg.s.type = OCTEON_MBOX_REQUEST;
+	mbox_cmd.msg.s.resp_needed = 1;
+	mbox_cmd.msg.s.cmd = OCTEON_GET_VF_STATS;
+	mbox_cmd.msg.s.len = 1;
+	mbox_cmd.q_no = vfidx * oct->sriov_info.rings_per_vf;
+	mbox_cmd.recv_len = 0;
+	mbox_cmd.recv_status = 0;
+	mbox_cmd.fn = cn23xx_get_vf_stats_callback;
+	ctx.stats = stats;
+	atomic_set(&ctx.status, 0);
+	mbox_cmd.fn_arg = (void *)&ctx;
+	memset(mbox_cmd.data, 0, sizeof(mbox_cmd.data));
+	octeon_mbox_write(oct, &mbox_cmd);
+
+	do {
+		schedule_timeout_uninterruptible(1);
+	} while ((atomic_read(&ctx.status) == 0) && (count++ < timeout));
+
+	ret = atomic_read(&ctx.status);
+	if (ret == 0) {
+		octeon_mbox_cancel(oct, 0);
+		dev_err(&oct->pci_dev->dev, "Unable to get stats from VF-%d, timedout\n",
+			vfidx);
+		return -1;
+	}
+
+	return 0;
+}
+EXPORT_SYMBOL_GPL(cn23xx_get_vf_stats);
diff --git a/drivers/net/ethernet/cavium/liquidio/cn23xx_pf_device.h b/drivers/net/ethernet/cavium/liquidio/cn23xx_pf_device.h
index dee604651ba7..bbe9f3133b07 100644
--- a/drivers/net/ethernet/cavium/liquidio/cn23xx_pf_device.h
+++ b/drivers/net/ethernet/cavium/liquidio/cn23xx_pf_device.h
@@ -24,8 +24,6 @@
 
 #include "cn23xx_pf_regs.h"
 
-#define LIO_CMD_WAIT_TM 100
-
 /* Register address and configuration for a CN23XX devices.
  * If device specific changes need to be made then add a struct to include
  * device specific fields as shown in the commented section
@@ -45,17 +43,26 @@ struct octeon_cn23xx_pf {
 
 #define CN23XX_SLI_DEF_BP			0x40
 
-int setup_cn23xx_octeon_pf_device(struct octeon_device *oct);
+struct oct_vf_stats {
+	u64 rx_packets;
+	u64 tx_packets;
+	u64 rx_bytes;
+	u64 tx_bytes;
+	u64 broadcast;
+	u64 multicast;
+};
 
-int validate_cn23xx_pf_config_info(struct octeon_device *oct,
-				   struct octeon_config *conf23xx);
+int setup_cn23xx_octeon_pf_device(struct octeon_device *oct);
 
 u32 cn23xx_pf_get_oq_ticks(struct octeon_device *oct, u32 time_intr_in_us);
 
-void cn23xx_dump_pf_initialized_regs(struct octeon_device *oct);
+int cn23xx_sriov_config(struct octeon_device *oct);
 
 int cn23xx_fw_loaded(struct octeon_device *oct);
 
 void cn23xx_tell_vf_its_macaddr_changed(struct octeon_device *oct, int vfidx,
 					u8 *mac);
+
+int cn23xx_get_vf_stats(struct octeon_device *oct, int ifidx,
+			struct oct_vf_stats *stats);
 #endif
diff --git a/drivers/net/ethernet/cavium/liquidio/cn23xx_pf_regs.h b/drivers/net/ethernet/cavium/liquidio/cn23xx_pf_regs.h
index e6d4ad99cc38..a0fd32476225 100644
--- a/drivers/net/ethernet/cavium/liquidio/cn23xx_pf_regs.h
+++ b/drivers/net/ethernet/cavium/liquidio/cn23xx_pf_regs.h
@@ -87,8 +87,8 @@
  */
 #define    CN23XX_SLI_PKT_IN_JABBER                0x29170
 /* The input jabber is used to determine the TSO max size.
- * Due to H/W limitation, this need to be reduced to 60000
- * in order to to H/W TSO and avoid the WQE malfarmation
+ * Due to H/W limitation, this needs to be reduced to 60000
+ * in order to use H/W TSO and avoid the WQE malformation
  * PKO_BUG_24989_WQE_LEN
  */
 #define    CN23XX_DEFAULT_INPUT_JABBER             0xEA60 /*60000*/
@@ -521,7 +521,7 @@
 #define    CN23XX_BAR1_INDEX_OFFSET                3
 
 #define    CN23XX_PEM_BAR1_INDEX_REG(port, idx)		\
-		(CN23XX_PEM_BAR1_INDEX_START + ((port) << CN23XX_PEM_OFFSET) + \
+		(CN23XX_PEM_BAR1_INDEX_START + (((u64)port) << CN23XX_PEM_OFFSET) + \
 		 ((idx) << CN23XX_BAR1_INDEX_OFFSET))
 
 /*############################ DPI #########################*/
diff --git a/drivers/net/ethernet/cavium/liquidio/cn23xx_vf_device.c b/drivers/net/ethernet/cavium/liquidio/cn23xx_vf_device.c
index b6117b6a1de2..d2fcb3da484e 100644
--- a/drivers/net/ethernet/cavium/liquidio/cn23xx_vf_device.c
+++ b/drivers/net/ethernet/cavium/liquidio/cn23xx_vf_device.c
@@ -165,9 +165,11 @@ static void cn23xx_vf_setup_global_output_regs(struct octeon_device *oct)
 		reg_val =
 		    octeon_read_csr(oct, CN23XX_VF_SLI_OQ_PKT_CONTROL(q_no));
 
-		/* set IPTR & DPTR */
-		reg_val |=
-		    (CN23XX_PKT_OUTPUT_CTL_IPTR | CN23XX_PKT_OUTPUT_CTL_DPTR);
+		/* clear IPTR */
+		reg_val &= ~CN23XX_PKT_OUTPUT_CTL_IPTR;
+
+		/* set DPTR */
+		reg_val |= CN23XX_PKT_OUTPUT_CTL_DPTR;
 
 		/* reset BMODE */
 		reg_val &= ~(CN23XX_PKT_OUTPUT_CTL_BMODE);
@@ -249,7 +251,7 @@ static void cn23xx_setup_vf_oq_regs(struct octeon_device *oct, u32 oq_no)
 	octeon_write_csr(oct, CN23XX_VF_SLI_OQ_SIZE(oq_no), droq->max_count);
 
 	octeon_write_csr(oct, CN23XX_VF_SLI_OQ_BUFF_INFO_SIZE(oq_no),
-			 (droq->buffer_size | (OCT_RH_SIZE << 16)));
+			 droq->buffer_size);
 
 	/* Get the mapped address of the pkt_sent and pkts_credit regs */
 	droq->pkts_sent_reg =
@@ -277,12 +279,10 @@ static int cn23xx_setup_vf_mbox(struct octeon_device *oct)
 {
 	struct octeon_mbox *mbox = NULL;
 
-	mbox = vmalloc(sizeof(*mbox));
+	mbox = vzalloc(sizeof(*mbox));
 	if (!mbox)
 		return 1;
 
-	memset(mbox, 0, sizeof(struct octeon_mbox));
-
 	spin_lock_init(&mbox->lock);
 
 	mbox->oct_dev = oct;
@@ -380,10 +380,11 @@ void cn23xx_vf_ask_pf_to_do_flr(struct octeon_device *oct)
 	mbox_cmd.recv_len = 0;
 	mbox_cmd.recv_status = 0;
 	mbox_cmd.fn = NULL;
-	mbox_cmd.fn_arg = 0;
+	mbox_cmd.fn_arg = NULL;
 
 	octeon_mbox_write(oct, &mbox_cmd);
 }
+EXPORT_SYMBOL_GPL(cn23xx_vf_ask_pf_to_do_flr);
 
 static void octeon_pfvf_hs_callback(struct octeon_device *oct,
 				    struct octeon_mbox_cmd *cmd,
@@ -428,14 +429,9 @@ int cn23xx_octeon_pfvf_handshake(struct octeon_device *oct)
 	mbox_cmd.q_no = 0;
 	mbox_cmd.recv_len = 0;
 	mbox_cmd.recv_status = 0;
-	mbox_cmd.fn = (octeon_mbox_callback_t)octeon_pfvf_hs_callback;
+	mbox_cmd.fn = octeon_pfvf_hs_callback;
 	mbox_cmd.fn_arg = &status;
 
-	/* Interrupts are not enabled at this point.
-	 * Enable them with default oq ticks
-	 */
-	oct->fn_list.enable_interrupt(oct, OCTEON_ALL_INTR);
-
 	octeon_mbox_write(oct, &mbox_cmd);
 
 	atomic_set(&status, 0);
@@ -444,11 +440,6 @@ int cn23xx_octeon_pfvf_handshake(struct octeon_device *oct)
 		schedule_timeout_uninterruptible(1);
 	} while ((!atomic_read(&status)) && (count++ < 100000));
 
-	/* Disable the interrupt so that the interrupsts will be reenabled
-	 * with the oq ticks received from the PF
-	 */
-	oct->fn_list.disable_interrupt(oct, OCTEON_ALL_INTR);
-
 	ret = atomic_read(&status);
 	if (!ret) {
 		dev_err(&oct->pci_dev->dev, "octeon_pfvf_handshake timeout\n");
@@ -476,6 +467,7 @@ int cn23xx_octeon_pfvf_handshake(struct octeon_device *oct)
 
 	return 0;
 }
+EXPORT_SYMBOL_GPL(cn23xx_octeon_pfvf_handshake);
 
 static void cn23xx_handle_vf_mbox_intr(struct octeon_ioq_vector *ioq_vector)
 {
@@ -624,7 +616,7 @@ static void cn23xx_disable_vf_interrupt(struct octeon_device *oct, u8 intr_flag)
 int cn23xx_setup_octeon_vf_device(struct octeon_device *oct)
 {
 	struct octeon_cn23xx_vf *cn23xx = (struct octeon_cn23xx_vf *)oct->chip;
-	u32 rings_per_vf, ring_flag;
+	u32 rings_per_vf;
 	u64 reg_val;
 
 	if (octeon_map_pci_barx(oct, 0, 0))
@@ -642,8 +634,6 @@ int cn23xx_setup_octeon_vf_device(struct octeon_device *oct)
 
 	rings_per_vf = reg_val & CN23XX_PKT_INPUT_CTL_RPVF_MASK;
 
-	ring_flag = 0;
-
 	cn23xx->conf  = oct_get_config_info(oct, LIO_23XX);
 	if (!cn23xx->conf) {
 		dev_err(&oct->pci_dev->dev, "%s No Config found for CN23XX\n",
@@ -690,33 +680,4 @@ int cn23xx_setup_octeon_vf_device(struct octeon_device *oct)
 
 	return 0;
 }
-
-void cn23xx_dump_vf_iq_regs(struct octeon_device *oct)
-{
-	u32 regval, q_no;
-
-	dev_dbg(&oct->pci_dev->dev, "SLI_IQ_DOORBELL_0 [0x%x]: 0x%016llx\n",
-		CN23XX_VF_SLI_IQ_DOORBELL(0),
-		CVM_CAST64(octeon_read_csr64(
-					oct, CN23XX_VF_SLI_IQ_DOORBELL(0))));
-
-	dev_dbg(&oct->pci_dev->dev, "SLI_IQ_BASEADDR_0 [0x%x]: 0x%016llx\n",
-		CN23XX_VF_SLI_IQ_BASE_ADDR64(0),
-		CVM_CAST64(octeon_read_csr64(
-			oct, CN23XX_VF_SLI_IQ_BASE_ADDR64(0))));
-
-	dev_dbg(&oct->pci_dev->dev, "SLI_IQ_FIFO_RSIZE_0 [0x%x]: 0x%016llx\n",
-		CN23XX_VF_SLI_IQ_SIZE(0),
-		CVM_CAST64(octeon_read_csr64(oct, CN23XX_VF_SLI_IQ_SIZE(0))));
-
-	for (q_no = 0; q_no < oct->sriov_info.rings_per_vf; q_no++) {
-		dev_dbg(&oct->pci_dev->dev, "SLI_PKT[%d]_INPUT_CTL [0x%x]: 0x%016llx\n",
-			q_no, CN23XX_VF_SLI_IQ_PKT_CONTROL64(q_no),
-			CVM_CAST64(octeon_read_csr64(
-				oct, CN23XX_VF_SLI_IQ_PKT_CONTROL64(q_no))));
-	}
-
-	pci_read_config_dword(oct->pci_dev, CN23XX_CONFIG_PCIE_DEVCTL, &regval);
-	dev_dbg(&oct->pci_dev->dev, "Config DevCtl [0x%x]: 0x%08x\n",
-		CN23XX_CONFIG_PCIE_DEVCTL, regval);
-}
+EXPORT_SYMBOL_GPL(cn23xx_setup_octeon_vf_device);
diff --git a/drivers/net/ethernet/cavium/liquidio/cn23xx_vf_device.h b/drivers/net/ethernet/cavium/liquidio/cn23xx_vf_device.h
index 3f98c7334957..40f529d0bc4c 100644
--- a/drivers/net/ethernet/cavium/liquidio/cn23xx_vf_device.h
+++ b/drivers/net/ethernet/cavium/liquidio/cn23xx_vf_device.h
@@ -36,8 +36,6 @@ struct octeon_cn23xx_vf {
 
 #define CN23XX_MAILBOX_MSGPARAM_SIZE		6
 
-#define MAX_VF_IP_OP_PENDING_PKT_COUNT		100
-
 void cn23xx_vf_ask_pf_to_do_flr(struct octeon_device *oct);
 
 int cn23xx_octeon_pfvf_handshake(struct octeon_device *oct);
@@ -45,6 +43,4 @@ int cn23xx_octeon_pfvf_handshake(struct octeon_device *oct);
 int cn23xx_setup_octeon_vf_device(struct octeon_device *oct);
 
 u32 cn23xx_vf_get_oq_ticks(struct octeon_device *oct, u32 time_intr_in_us);
-
-void cn23xx_dump_vf_initialized_regs(struct octeon_device *oct);
 #endif
diff --git a/drivers/net/ethernet/cavium/liquidio/cn23xx_vf_regs.h b/drivers/net/ethernet/cavium/liquidio/cn23xx_vf_regs.h
index d33dd8f4226f..e956109415cd 100644
--- a/drivers/net/ethernet/cavium/liquidio/cn23xx_vf_regs.h
+++ b/drivers/net/ethernet/cavium/liquidio/cn23xx_vf_regs.h
@@ -36,8 +36,8 @@
 #define     CN23XX_CONFIG_PCIE_FLTMSK              0x720
 
 /* The input jabber is used to determine the TSO max size.
- * Due to H/W limitation, this need to be reduced to 60000
- * in order to to H/W TSO and avoid the WQE malfarmation
+ * Due to H/W limitation, this needs to be reduced to 60000
+ * in order to use H/W TSO and avoid the WQE malformation
  * PKO_BUG_24989_WQE_LEN
  */
 #define    CN23XX_DEFAULT_INPUT_JABBER             0xEA60 /*60000*/
diff --git a/drivers/net/ethernet/cavium/liquidio/cn66xx_device.c b/drivers/net/ethernet/cavium/liquidio/cn66xx_device.c
index bdec051107a6..93fccfec288d 100644
--- a/drivers/net/ethernet/cavium/liquidio/cn66xx_device.c
+++ b/drivers/net/ethernet/cavium/liquidio/cn66xx_device.c
@@ -38,13 +38,10 @@ int lio_cn6xxx_soft_reset(struct octeon_device *oct)
 	lio_pci_readq(oct, CN6XXX_CIU_SOFT_RST);
 	lio_pci_writeq(oct, 1, CN6XXX_CIU_SOFT_RST);
 
-	/* make sure that the reset is written before starting timer */
-	mmiowb();
-
 	/* Wait for 10ms as Octeon resets. */
 	mdelay(100);
 
-	if (octeon_read_csr64(oct, CN6XXX_SLI_SCRATCH1) == 0x1234ULL) {
+	if (octeon_read_csr64(oct, CN6XXX_SLI_SCRATCH1)) {
 		dev_err(&oct->pci_dev->dev, "Soft reset failed\n");
 		return 1;
 	}
@@ -209,9 +206,6 @@ void lio_cn6xxx_setup_global_output_regs(struct octeon_device *oct)
 		octeon_write_csr64(oct, CN6XXX_SLI_OQ_WMARK, 0);
 	}
 
-	/* / Select Info Ptr for length & data */
-	octeon_write_csr(oct, CN6XXX_SLI_PKT_IPTR, 0xFFFFFFFF);
-
 	/* / Select Packet count instead of bytes for SLI_PKTi_CNTS[CNT] */
 	octeon_write_csr(oct, CN6XXX_SLI_PKT_OUT_BMODE, 0);
 
@@ -314,7 +308,7 @@ void lio_cn6xxx_setup_oq_regs(struct octeon_device *oct, u32 oq_no)
 	octeon_write_csr(oct, CN6XXX_SLI_OQ_SIZE(oq_no), droq->max_count);
 
 	octeon_write_csr(oct, CN6XXX_SLI_OQ_BUFF_INFO_SIZE(oq_no),
-			 (droq->buffer_size | (OCT_RH_SIZE << 16)));
+			 droq->buffer_size);
 
 	/* Get the mapped address of the pkt_sent and pkts_credit regs */
 	droq->pkts_sent_reg =
@@ -490,9 +484,6 @@ void lio_cn6xxx_disable_interrupt(struct octeon_device *oct,
 
 	/* Disable Interrupts */
 	writeq(0, cn6xxx->intr_enb_reg64);
-
-	/* make sure interrupts are really disabled */
-	mmiowb();
 }
 
 static void lio_cn6xxx_get_pcie_qlmport(struct octeon_device *oct)
@@ -558,10 +549,6 @@ static int lio_cn6xxx_process_droq_intr_regs(struct octeon_device *oct)
 				value &= ~(1 << oq_no);
 				octeon_write_csr(oct, reg, value);
 
-				/* Ensure that the enable register is written.
-				 */
-				mmiowb();
-
 				spin_unlock(&cn6xxx->lock_for_droq_int_enb_reg);
 			}
 		}
@@ -710,6 +697,7 @@ int lio_setup_cn66xx_octeon_device(struct octeon_device *oct)
 
 	return 0;
 }
+EXPORT_SYMBOL_GPL(lio_setup_cn66xx_octeon_device);
 
 int lio_validate_cn6xxx_config_info(struct octeon_device *oct,
 				    struct octeon_config *conf6xxx)
@@ -734,8 +722,7 @@ int lio_validate_cn6xxx_config_info(struct octeon_device *oct,
 			__func__);
 		return 1;
 	}
-	if (!(CFG_GET_OQ_INFO_PTR(conf6xxx)) ||
-	    !(CFG_GET_OQ_REFILL_THRESHOLD(conf6xxx))) {
+	if (!CFG_GET_OQ_REFILL_THRESHOLD(conf6xxx)) {
 		dev_err(&oct->pci_dev->dev, "%s: Invalid parameter for OQ\n",
 			__func__);
 		return 1;
diff --git a/drivers/net/ethernet/cavium/liquidio/cn66xx_device.h b/drivers/net/ethernet/cavium/liquidio/cn66xx_device.h
index 8ed57134ee0c..129c8b84f549 100644
--- a/drivers/net/ethernet/cavium/liquidio/cn66xx_device.h
+++ b/drivers/net/ethernet/cavium/liquidio/cn66xx_device.h
@@ -86,7 +86,6 @@ u32
 lio_cn6xxx_update_read_index(struct octeon_instr_queue *iq);
 void lio_cn6xxx_enable_interrupt(struct octeon_device *oct, u8 unused);
 void lio_cn6xxx_disable_interrupt(struct octeon_device *oct, u8 unused);
-void cn6xxx_get_pcie_qlmport(struct octeon_device *oct);
 void lio_cn6xxx_setup_reg_address(struct octeon_device *oct, void *chip,
 				  struct octeon_reg_list *reg_list);
 u32 lio_cn6xxx_coprocessor_clock(struct octeon_device *oct);
diff --git a/drivers/net/ethernet/cavium/liquidio/cn66xx_regs.h b/drivers/net/ethernet/cavium/liquidio/cn66xx_regs.h
index b248966837b4..7aad40b2aa73 100644
--- a/drivers/net/ethernet/cavium/liquidio/cn66xx_regs.h
+++ b/drivers/net/ethernet/cavium/liquidio/cn66xx_regs.h
@@ -412,7 +412,7 @@
 	   | CN6XXX_INTR_M0UNWI_ERR             \
 	   | CN6XXX_INTR_M1UPB0_ERR             \
 	   | CN6XXX_INTR_M1UPWI_ERR             \
-	   | CN6XXX_INTR_M1UPB0_ERR             \
+	   | CN6XXX_INTR_M1UNB0_ERR             \
 	   | CN6XXX_INTR_M1UNWI_ERR             \
 	   | CN6XXX_INTR_INSTR_DB_OF_ERR        \
 	   | CN6XXX_INTR_SLIST_DB_OF_ERR        \
diff --git a/drivers/net/ethernet/cavium/liquidio/cn68xx_device.c b/drivers/net/ethernet/cavium/liquidio/cn68xx_device.c
index 50b533ff58e6..b5103def3761 100644
--- a/drivers/net/ethernet/cavium/liquidio/cn68xx_device.c
+++ b/drivers/net/ethernet/cavium/liquidio/cn68xx_device.c
@@ -25,6 +25,7 @@
 #include "octeon_main.h"
 #include "cn66xx_regs.h"
 #include "cn66xx_device.h"
+#include "cn68xx_device.h"
 #include "cn68xx_regs.h"
 
 static void lio_cn68xx_set_dpi_regs(struct octeon_device *oct)
@@ -180,3 +181,4 @@ int lio_setup_cn68xx_octeon_device(struct octeon_device *oct)
 
 	return 0;
 }
+EXPORT_SYMBOL_GPL(lio_setup_cn68xx_octeon_device);
diff --git a/drivers/net/ethernet/cavium/liquidio/lio_core.c b/drivers/net/ethernet/cavium/liquidio/lio_core.c
index 796c2cbc11f6..215dac201b4a 100644
--- a/drivers/net/ethernet/cavium/liquidio/lio_core.c
+++ b/drivers/net/ethernet/cavium/liquidio/lio_core.c
@@ -26,9 +26,141 @@
 #include "octeon_main.h"
 #include "octeon_network.h"
 
+MODULE_AUTHOR("Cavium Networks, <support@cavium.com>");
+MODULE_DESCRIPTION("Cavium LiquidIO Intelligent Server Adapter Core");
+MODULE_LICENSE("GPL");
+
 /* OOM task polling interval */
 #define LIO_OOM_POLL_INTERVAL_MS 250
 
+#define OCTNIC_MAX_SG  MAX_SKB_FRAGS
+
+/**
+ * lio_delete_glists - Delete gather lists
+ * @lio: per-network private data
+ */
+void lio_delete_glists(struct lio *lio)
+{
+	struct octnic_gather *g;
+	int i;
+
+	kfree(lio->glist_lock);
+	lio->glist_lock = NULL;
+
+	if (!lio->glist)
+		return;
+
+	for (i = 0; i < lio->oct_dev->num_iqs; i++) {
+		do {
+			g = (struct octnic_gather *)
+			    lio_list_delete_head(&lio->glist[i]);
+			kfree(g);
+		} while (g);
+
+		if (lio->glists_virt_base && lio->glists_virt_base[i] &&
+		    lio->glists_dma_base && lio->glists_dma_base[i]) {
+			lio_dma_free(lio->oct_dev,
+				     lio->glist_entry_size * lio->tx_qsize,
+				     lio->glists_virt_base[i],
+				     lio->glists_dma_base[i]);
+		}
+	}
+
+	kfree(lio->glists_virt_base);
+	lio->glists_virt_base = NULL;
+
+	kfree(lio->glists_dma_base);
+	lio->glists_dma_base = NULL;
+
+	kfree(lio->glist);
+	lio->glist = NULL;
+}
+EXPORT_SYMBOL_GPL(lio_delete_glists);
+
+/**
+ * lio_setup_glists - Setup gather lists
+ * @oct: octeon_device
+ * @lio: per-network private data
+ * @num_iqs: count of iqs to allocate
+ */
+int lio_setup_glists(struct octeon_device *oct, struct lio *lio, int num_iqs)
+{
+	struct octnic_gather *g;
+	int i, j;
+
+	lio->glist_lock =
+	    kcalloc(num_iqs, sizeof(*lio->glist_lock), GFP_KERNEL);
+	if (!lio->glist_lock)
+		return -ENOMEM;
+
+	lio->glist =
+	    kcalloc(num_iqs, sizeof(*lio->glist), GFP_KERNEL);
+	if (!lio->glist) {
+		kfree(lio->glist_lock);
+		lio->glist_lock = NULL;
+		return -ENOMEM;
+	}
+
+	lio->glist_entry_size =
+		ROUNDUP8((ROUNDUP4(OCTNIC_MAX_SG) >> 2) * OCT_SG_ENTRY_SIZE);
+
+	/* allocate memory to store virtual and dma base address of
+	 * per glist consistent memory
+	 */
+	lio->glists_virt_base = kcalloc(num_iqs, sizeof(*lio->glists_virt_base),
+					GFP_KERNEL);
+	lio->glists_dma_base = kcalloc(num_iqs, sizeof(*lio->glists_dma_base),
+				       GFP_KERNEL);
+
+	if (!lio->glists_virt_base || !lio->glists_dma_base) {
+		lio_delete_glists(lio);
+		return -ENOMEM;
+	}
+
+	for (i = 0; i < num_iqs; i++) {
+		int numa_node = dev_to_node(&oct->pci_dev->dev);
+
+		spin_lock_init(&lio->glist_lock[i]);
+
+		INIT_LIST_HEAD(&lio->glist[i]);
+
+		lio->glists_virt_base[i] =
+			lio_dma_alloc(oct,
+				      lio->glist_entry_size * lio->tx_qsize,
+				      &lio->glists_dma_base[i]);
+
+		if (!lio->glists_virt_base[i]) {
+			lio_delete_glists(lio);
+			return -ENOMEM;
+		}
+
+		for (j = 0; j < lio->tx_qsize; j++) {
+			g = kzalloc_node(sizeof(*g), GFP_KERNEL,
+					 numa_node);
+			if (!g)
+				g = kzalloc(sizeof(*g), GFP_KERNEL);
+			if (!g)
+				break;
+
+			g->sg = lio->glists_virt_base[i] +
+				(j * lio->glist_entry_size);
+
+			g->sg_dma_ptr = lio->glists_dma_base[i] +
+					(j * lio->glist_entry_size);
+
+			list_add_tail(&g->list, &lio->glist[i]);
+		}
+
+		if (j != lio->tx_qsize) {
+			lio_delete_glists(lio);
+			return -ENOMEM;
+		}
+	}
+
+	return 0;
+}
+EXPORT_SYMBOL_GPL(lio_setup_glists);
+
 int liquidio_set_feature(struct net_device *netdev, int cmd, u16 param1)
 {
 	struct lio *lio = GET_LIO(netdev);
@@ -42,17 +174,19 @@ int liquidio_set_feature(struct net_device *netdev, int cmd, u16 param1)
 	nctrl.ncmd.s.cmd = cmd;
 	nctrl.ncmd.s.param1 = param1;
 	nctrl.iq_no = lio->linfo.txpciq[0].s.q_no;
-	nctrl.wait_time = 100;
 	nctrl.netpndev = (u64)netdev;
 	nctrl.cb_fn = liquidio_link_ctrl_cmd_completion;
 
 	ret = octnet_send_nic_ctrl_pkt(lio->oct_dev, &nctrl);
-	if (ret < 0) {
+	if (ret) {
 		dev_err(&oct->pci_dev->dev, "Feature change failed in core (ret: 0x%x)\n",
 			ret);
+		if (ret > 0)
+			ret = -EIO;
 	}
 	return ret;
 }
+EXPORT_SYMBOL_GPL(liquidio_set_feature);
 
 void octeon_report_tx_completion_to_bql(void *txq, unsigned int pkts_compl,
 					unsigned int bytes_compl)
@@ -91,7 +225,7 @@ void octeon_update_tx_completion_counters(void *buf, int reqtype,
 	*bytes_compl += skb->len;
 }
 
-void octeon_report_sent_bytes_to_bql(void *buf, int reqtype)
+int octeon_report_sent_bytes_to_bql(void *buf, int reqtype)
 {
 	struct octnet_buf_free_info *finfo;
 	struct sk_buff *skb;
@@ -112,11 +246,13 @@ void octeon_report_sent_bytes_to_bql(void *buf, int reqtype)
 		break;
 
 	default:
-		return;
+		return 0;
 	}
 
 	txq = netdev_get_tx_queue(skb->dev, skb_get_queue_mapping(skb));
 	netdev_tx_sent_queue(txq, skb->len);
+
+	return netif_xmit_stopped(txq);
 }
 
 void liquidio_link_ctrl_cmd_completion(void *nctrl_ptr)
@@ -127,20 +263,13 @@ void liquidio_link_ctrl_cmd_completion(void *nctrl_ptr)
 	struct octeon_device *oct = lio->oct_dev;
 	u8 *mac;
 
-	if (nctrl->completion && nctrl->response_code) {
-		/* Signal whoever is interested that the response code from the
-		 * firmware has arrived.
-		 */
-		WRITE_ONCE(*nctrl->response_code, nctrl->status);
-		complete(nctrl->completion);
-	}
-
-	if (nctrl->status)
+	if (nctrl->sc_status)
 		return;
 
 	switch (nctrl->ncmd.s.cmd) {
 	case OCTNET_CMD_CHANGE_DEVFLAGS:
 	case OCTNET_CMD_SET_MULTI_LIST:
+	case OCTNET_CMD_SET_UC_LIST:
 		break;
 
 	case OCTNET_CMD_CHANGE_MACADDR:
@@ -161,18 +290,6 @@ void liquidio_link_ctrl_cmd_completion(void *nctrl_ptr)
 		}
 		break;
 
-	case OCTNET_CMD_CHANGE_MTU:
-		/* If command is successful, change the MTU. */
-		netif_info(lio, probe, lio->netdev, "MTU Changed from %d to %d\n",
-			   netdev->mtu, nctrl->ncmd.s.param1);
-		dev_info(&oct->pci_dev->dev, "%s MTU Changed from %d to %d\n",
-			 netdev->name, netdev->mtu,
-			 nctrl->ncmd.s.param1);
-		netdev->mtu = nctrl->ncmd.s.param1;
-		queue_delayed_work(lio->link_status_wq.wq,
-				   &lio->link_status_wq.wk.work, 0);
-		break;
-
 	case OCTNET_CMD_GPIO_ACCESS:
 		netif_info(lio, probe, lio->netdev, "LED Flashing visual identification\n");
 
@@ -202,9 +319,13 @@ void liquidio_link_ctrl_cmd_completion(void *nctrl_ptr)
 			 netdev->name);
 		break;
 
-	case OCTNET_CMD_ENABLE_VLAN_FILTER:
-		dev_info(&oct->pci_dev->dev, "%s VLAN filter enabled\n",
-			 netdev->name);
+	case OCTNET_CMD_VLAN_FILTER_CTL:
+		if (nctrl->ncmd.s.param1)
+			dev_info(&oct->pci_dev->dev,
+				 "%s VLAN filter enabled\n", netdev->name);
+		else
+			dev_info(&oct->pci_dev->dev,
+				 "%s VLAN filter disabled\n", netdev->name);
 		break;
 
 	case OCTNET_CMD_ADD_VLAN_FILTER:
@@ -271,11 +392,17 @@ void liquidio_link_ctrl_cmd_completion(void *nctrl_ptr)
 		netif_info(lio, probe, lio->netdev, "Set RX/TX flow control parameters\n");
 		break;
 
+	case OCTNET_CMD_QUEUE_COUNT_CTL:
+		netif_info(lio, probe, lio->netdev, "Queue count updated to %d\n",
+			   nctrl->ncmd.s.param1);
+		break;
+
 	default:
 		dev_err(&oct->pci_dev->dev, "%s Unknown cmd %d\n", __func__,
 			nctrl->ncmd.s.cmd);
 	}
 }
+EXPORT_SYMBOL_GPL(liquidio_link_ctrl_cmd_completion);
 
 void octeon_pf_changed_vf_macaddr(struct octeon_device *oct, u8 *mac)
 {
@@ -292,7 +419,7 @@ void octeon_pf_changed_vf_macaddr(struct octeon_device *oct, u8 *mac)
 
 	if (!ether_addr_equal(netdev->dev_addr, mac)) {
 		macaddr_changed = true;
-		ether_addr_copy(netdev->dev_addr, mac);
+		eth_hw_addr_set(netdev, mac);
 		ether_addr_copy(((u8 *)&lio->linfo.hw_addr) + 2, mac);
 		call_netdevice_notifiers(NETDEV_CHANGEADDR, netdev);
 	}
@@ -308,55 +435,1397 @@ void octeon_pf_changed_vf_macaddr(struct octeon_device *oct, u8 *mac)
 	 */
 }
 
+void octeon_schedule_rxq_oom_work(struct octeon_device *oct,
+				  struct octeon_droq *droq)
+{
+	struct net_device *netdev = oct->props[0].netdev;
+	struct lio *lio = GET_LIO(netdev);
+	struct cavium_wq *wq = &lio->rxq_status_wq[droq->q_no];
+
+	queue_delayed_work(wq->wq, &wq->wk.work,
+			   msecs_to_jiffies(LIO_OOM_POLL_INTERVAL_MS));
+}
+
 static void octnet_poll_check_rxq_oom_status(struct work_struct *work)
 {
 	struct cavium_wk *wk = (struct cavium_wk *)work;
 	struct lio *lio = (struct lio *)wk->ctxptr;
 	struct octeon_device *oct = lio->oct_dev;
+	int q_no = wk->ctxul;
+	struct octeon_droq *droq = oct->droq[q_no];
+
+	if (!ifstate_check(lio, LIO_IFSTATE_RUNNING) || !droq)
+		return;
+
+	if (octeon_retry_droq_refill(droq))
+		octeon_schedule_rxq_oom_work(oct, droq);
+}
+
+int setup_rx_oom_poll_fn(struct net_device *netdev)
+{
+	struct lio *lio = GET_LIO(netdev);
+	struct octeon_device *oct = lio->oct_dev;
+	struct cavium_wq *wq;
+	int q, q_no;
+
+	for (q = 0; q < oct->num_oqs; q++) {
+		q_no = lio->linfo.rxpciq[q].s.q_no;
+		wq = &lio->rxq_status_wq[q_no];
+		wq->wq = alloc_workqueue("rxq-oom-status",
+					 WQ_MEM_RECLAIM | WQ_PERCPU, 0);
+		if (!wq->wq) {
+			dev_err(&oct->pci_dev->dev, "unable to create cavium rxq oom status wq\n");
+			return -ENOMEM;
+		}
+
+		INIT_DELAYED_WORK(&wq->wk.work,
+				  octnet_poll_check_rxq_oom_status);
+		wq->wk.ctxptr = lio;
+		wq->wk.ctxul = q_no;
+	}
+
+	return 0;
+}
+EXPORT_SYMBOL_GPL(setup_rx_oom_poll_fn);
+
+void cleanup_rx_oom_poll_fn(struct net_device *netdev)
+{
+	struct lio *lio = GET_LIO(netdev);
+	struct octeon_device *oct = lio->oct_dev;
+	struct cavium_wq *wq;
+	int q_no;
+
+	for (q_no = 0; q_no < oct->num_oqs; q_no++) {
+		wq = &lio->rxq_status_wq[q_no];
+		if (wq->wq) {
+			cancel_delayed_work_sync(&wq->wk.work);
+			destroy_workqueue(wq->wq);
+			wq->wq = NULL;
+		}
+	}
+}
+EXPORT_SYMBOL_GPL(cleanup_rx_oom_poll_fn);
+
+/* Runs in interrupt context. */
+static void lio_update_txq_status(struct octeon_device *oct, int iq_num)
+{
+	struct octeon_instr_queue *iq = oct->instr_queue[iq_num];
+	struct net_device *netdev;
+	struct lio *lio;
+
+	netdev = oct->props[iq->ifidx].netdev;
+
+	/* This is needed because the first IQ does not have
+	 * a netdev associated with it.
+	 */
+	if (!netdev)
+		return;
+
+	lio = GET_LIO(netdev);
+	if (__netif_subqueue_stopped(netdev, iq->q_index) &&
+	    lio->linfo.link.s.link_up &&
+	    (!octnet_iq_is_full(oct, iq_num))) {
+		netif_wake_subqueue(netdev, iq->q_index);
+		INCR_INSTRQUEUE_PKT_COUNT(lio->oct_dev, iq_num,
+					  tx_restart, 1);
+	}
+}
+
+/**
+ * octeon_setup_droq - Setup output queue
+ * @oct: octeon device
+ * @q_no: which queue
+ * @num_descs: how many descriptors
+ * @desc_size: size of each descriptor
+ * @app_ctx: application context
+ */
+static int octeon_setup_droq(struct octeon_device *oct, int q_no, int num_descs,
+			     int desc_size, void *app_ctx)
+{
+	int ret_val;
+
+	dev_dbg(&oct->pci_dev->dev, "Creating Droq: %d\n", q_no);
+	/* droq creation and local register settings. */
+	ret_val = octeon_create_droq(oct, q_no, num_descs, desc_size, app_ctx);
+	if (ret_val < 0)
+		return ret_val;
+
+	if (ret_val == 1) {
+		dev_dbg(&oct->pci_dev->dev, "Using default droq %d\n", q_no);
+		return 0;
+	}
+
+	/* Enable the droq queues */
+	octeon_set_droq_pkt_op(oct, q_no, 1);
+
+	/* Send Credit for Octeon Output queues. Credits are always
+	 * sent after the output queue is enabled.
+	 */
+	writel(oct->droq[q_no]->max_count, oct->droq[q_no]->pkts_credit_reg);
+
+	return ret_val;
+}
+
+/**
+ * liquidio_push_packet - Routine to push packets arriving on Octeon interface upto network layer.
+ * @octeon_id:octeon device id.
+ * @skbuff:   skbuff struct to be passed to network layer.
+ * @len:      size of total data received.
+ * @rh:       Control header associated with the packet
+ * @param:    additional control data with the packet
+ * @arg:      farg registered in droq_ops
+ */
+static void
+liquidio_push_packet(u32 __maybe_unused octeon_id,
+		     void *skbuff,
+		     u32 len,
+		     union octeon_rh *rh,
+		     void *param,
+		     void *arg)
+{
+	struct net_device *netdev = (struct net_device *)arg;
+	struct octeon_droq *droq =
+	    container_of(param, struct octeon_droq, napi);
+	struct sk_buff *skb = (struct sk_buff *)skbuff;
+	struct skb_shared_hwtstamps *shhwtstamps;
+	struct napi_struct *napi = param;
+	u16 vtag = 0;
+	u32 r_dh_off;
+	u64 ns;
+
+	if (netdev) {
+		struct lio *lio = GET_LIO(netdev);
+		struct octeon_device *oct = lio->oct_dev;
+
+		/* Do not proceed if the interface is not in RUNNING state. */
+		if (!ifstate_check(lio, LIO_IFSTATE_RUNNING)) {
+			recv_buffer_free(skb);
+			droq->stats.rx_dropped++;
+			return;
+		}
+
+		skb->dev = netdev;
+
+		skb_record_rx_queue(skb, droq->q_no);
+		if (likely(len > MIN_SKB_SIZE)) {
+			struct octeon_skb_page_info *pg_info;
+			unsigned char *va;
+
+			pg_info = ((struct octeon_skb_page_info *)(skb->cb));
+			if (pg_info->page) {
+				/* For Paged allocation use the frags */
+				va = page_address(pg_info->page) +
+					pg_info->page_offset;
+				memcpy(skb->data, va, MIN_SKB_SIZE);
+				skb_put(skb, MIN_SKB_SIZE);
+				skb_add_rx_frag(skb, skb_shinfo(skb)->nr_frags,
+						pg_info->page,
+						pg_info->page_offset +
+						MIN_SKB_SIZE,
+						len - MIN_SKB_SIZE,
+						LIO_RXBUFFER_SZ);
+			}
+		} else {
+			struct octeon_skb_page_info *pg_info =
+				((struct octeon_skb_page_info *)(skb->cb));
+			skb_copy_to_linear_data(skb, page_address(pg_info->page)
+						+ pg_info->page_offset, len);
+			skb_put(skb, len);
+			put_page(pg_info->page);
+		}
+
+		r_dh_off = (rh->r_dh.len - 1) * BYTES_PER_DHLEN_UNIT;
+
+		if (oct->ptp_enable) {
+			if (rh->r_dh.has_hwtstamp) {
+				/* timestamp is included from the hardware at
+				 * the beginning of the packet.
+				 */
+				if (ifstate_check
+					(lio,
+					 LIO_IFSTATE_RX_TIMESTAMP_ENABLED)) {
+					/* Nanoseconds are in the first 64-bits
+					 * of the packet.
+					 */
+					memcpy(&ns, (skb->data + r_dh_off),
+					       sizeof(ns));
+					r_dh_off -= BYTES_PER_DHLEN_UNIT;
+					shhwtstamps = skb_hwtstamps(skb);
+					shhwtstamps->hwtstamp =
+						ns_to_ktime(ns +
+							    lio->ptp_adjust);
+				}
+			}
+		}
+
+		if (rh->r_dh.has_hash) {
+			__be32 *hash_be = (__be32 *)(skb->data + r_dh_off);
+			u32 hash = be32_to_cpu(*hash_be);
+
+			skb_set_hash(skb, hash, PKT_HASH_TYPE_L4);
+			r_dh_off -= BYTES_PER_DHLEN_UNIT;
+		}
+
+		skb_pull(skb, rh->r_dh.len * BYTES_PER_DHLEN_UNIT);
+		skb->protocol = eth_type_trans(skb, skb->dev);
+
+		if ((netdev->features & NETIF_F_RXCSUM) &&
+		    (((rh->r_dh.encap_on) &&
+		      (rh->r_dh.csum_verified & CNNIC_TUN_CSUM_VERIFIED)) ||
+		     (!(rh->r_dh.encap_on) &&
+		      ((rh->r_dh.csum_verified & CNNIC_CSUM_VERIFIED) ==
+			CNNIC_CSUM_VERIFIED))))
+			/* checksum has already been verified */
+			skb->ip_summed = CHECKSUM_UNNECESSARY;
+		else
+			skb->ip_summed = CHECKSUM_NONE;
+
+		/* Setting Encapsulation field on basis of status received
+		 * from the firmware
+		 */
+		if (rh->r_dh.encap_on) {
+			skb->encapsulation = 1;
+			skb->csum_level = 1;
+			droq->stats.rx_vxlan++;
+		}
+
+		/* inbound VLAN tag */
+		if ((netdev->features & NETIF_F_HW_VLAN_CTAG_RX) &&
+		    rh->r_dh.vlan) {
+			u16 priority = rh->r_dh.priority;
+			u16 vid = rh->r_dh.vlan;
+
+			vtag = (priority << VLAN_PRIO_SHIFT) | vid;
+			__vlan_hwaccel_put_tag(skb, htons(ETH_P_8021Q), vtag);
+		}
+
+		napi_gro_receive(napi, skb);
+
+		droq->stats.rx_bytes_received += len -
+			rh->r_dh.len * BYTES_PER_DHLEN_UNIT;
+		droq->stats.rx_pkts_received++;
+	} else {
+		recv_buffer_free(skb);
+	}
+}
+
+/**
+ * napi_schedule_wrapper - wrapper for calling napi_schedule
+ * @param: parameters to pass to napi_schedule
+ *
+ * Used when scheduling on different CPUs
+ */
+static void napi_schedule_wrapper(void *param)
+{
+	struct napi_struct *napi = param;
+
+	napi_schedule(napi);
+}
+
+/**
+ * liquidio_napi_drv_callback - callback when receive interrupt occurs and we are in NAPI mode
+ * @arg: pointer to octeon output queue
+ */
+static void liquidio_napi_drv_callback(void *arg)
+{
+	struct octeon_device *oct;
+	struct octeon_droq *droq = arg;
+	int this_cpu = smp_processor_id();
+
+	oct = droq->oct_dev;
+
+	if (OCTEON_CN23XX_PF(oct) || OCTEON_CN23XX_VF(oct) ||
+	    droq->cpu_id == this_cpu) {
+		napi_schedule_irqoff(&droq->napi);
+	} else {
+		INIT_CSD(&droq->csd, napi_schedule_wrapper, &droq->napi);
+		smp_call_function_single_async(droq->cpu_id, &droq->csd);
+	}
+}
+
+/**
+ * liquidio_napi_poll - Entry point for NAPI polling
+ * @napi: NAPI structure
+ * @budget: maximum number of items to process
+ */
+static int liquidio_napi_poll(struct napi_struct *napi, int budget)
+{
+	struct octeon_instr_queue *iq;
+	struct octeon_device *oct;
+	struct octeon_droq *droq;
+	int tx_done = 0, iq_no;
+	int work_done;
+
+	droq = container_of(napi, struct octeon_droq, napi);
+	oct = droq->oct_dev;
+	iq_no = droq->q_no;
+
+	/* Handle Droq descriptors */
+	work_done = octeon_droq_process_poll_pkts(oct, droq, budget);
+
+	/* Flush the instruction queue */
+	iq = oct->instr_queue[iq_no];
+	if (iq) {
+		/* TODO: move this check to inside octeon_flush_iq,
+		 * once check_db_timeout is removed
+		 */
+		if (atomic_read(&iq->instr_pending))
+			/* Process iq buffers with in the budget limits */
+			tx_done = octeon_flush_iq(oct, iq, budget);
+		else
+			tx_done = 1;
+		/* Update iq read-index rather than waiting for next interrupt.
+		 * Return back if tx_done is false.
+		 */
+		/* sub-queue status update */
+		lio_update_txq_status(oct, iq_no);
+	} else {
+		dev_err(&oct->pci_dev->dev, "%s:  iq (%d) num invalid\n",
+			__func__, iq_no);
+	}
+
+#define MAX_REG_CNT  2000000U
+	/* force enable interrupt if reg cnts are high to avoid wraparound */
+	if ((work_done < budget && tx_done) ||
+	    (iq && iq->pkt_in_done >= MAX_REG_CNT) ||
+	    (droq->pkt_count >= MAX_REG_CNT)) {
+		napi_complete_done(napi, work_done);
+
+		octeon_enable_irq(droq->oct_dev, droq->q_no);
+		return 0;
+	}
+
+	return (!tx_done) ? (budget) : (work_done);
+}
+
+/**
+ * liquidio_setup_io_queues - Setup input and output queues
+ * @octeon_dev: octeon device
+ * @ifidx: Interface index
+ * @num_iqs: input io queue count
+ * @num_oqs: output io queue count
+ *
+ * Note: Queues are with respect to the octeon device. Thus
+ * an input queue is for egress packets, and output queues
+ * are for ingress packets.
+ */
+int liquidio_setup_io_queues(struct octeon_device *octeon_dev, int ifidx,
+			     u32 num_iqs, u32 num_oqs)
+{
+	struct octeon_droq_ops droq_ops;
+	struct net_device *netdev;
+	struct octeon_droq *droq;
+	struct napi_struct *napi;
+	int cpu_id_modulus;
+	int num_tx_descs;
+	struct lio *lio;
+	int retval = 0;
+	int q, q_no;
+	int cpu_id;
+
+	netdev = octeon_dev->props[ifidx].netdev;
+
+	lio = GET_LIO(netdev);
+
+	memset(&droq_ops, 0, sizeof(struct octeon_droq_ops));
+
+	droq_ops.fptr = liquidio_push_packet;
+	droq_ops.farg = netdev;
+
+	droq_ops.poll_mode = 1;
+	droq_ops.napi_fn = liquidio_napi_drv_callback;
+	cpu_id = 0;
+	cpu_id_modulus = num_present_cpus();
+
+	/* set up DROQs. */
+	for (q = 0; q < num_oqs; q++) {
+		q_no = lio->linfo.rxpciq[q].s.q_no;
+		dev_dbg(&octeon_dev->pci_dev->dev,
+			"%s index:%d linfo.rxpciq.s.q_no:%d\n",
+			__func__, q, q_no);
+		retval = octeon_setup_droq(
+		    octeon_dev, q_no,
+		    CFG_GET_NUM_RX_DESCS_NIC_IF(octeon_get_conf(octeon_dev),
+						lio->ifidx),
+		    CFG_GET_NUM_RX_BUF_SIZE_NIC_IF(octeon_get_conf(octeon_dev),
+						   lio->ifidx),
+		    NULL);
+		if (retval) {
+			dev_err(&octeon_dev->pci_dev->dev,
+				"%s : Runtime DROQ(RxQ) creation failed.\n",
+				__func__);
+			return 1;
+		}
+
+		droq = octeon_dev->droq[q_no];
+		napi = &droq->napi;
+		dev_dbg(&octeon_dev->pci_dev->dev, "netif_napi_add netdev:%llx oct:%llx\n",
+			(u64)netdev, (u64)octeon_dev);
+		netif_napi_add(netdev, napi, liquidio_napi_poll);
+
+		/* designate a CPU for this droq */
+		droq->cpu_id = cpu_id;
+		cpu_id++;
+		if (cpu_id >= cpu_id_modulus)
+			cpu_id = 0;
+
+		octeon_register_droq_ops(octeon_dev, q_no, &droq_ops);
+	}
+
+	if (OCTEON_CN23XX_PF(octeon_dev) || OCTEON_CN23XX_VF(octeon_dev)) {
+		/* 23XX PF/VF can send/recv control messages (via the first
+		 * PF/VF-owned droq) from the firmware even if the ethX
+		 * interface is down, so that's why poll_mode must be off
+		 * for the first droq.
+		 */
+		octeon_dev->droq[0]->ops.poll_mode = 0;
+	}
+
+	/* set up IQs. */
+	for (q = 0; q < num_iqs; q++) {
+		num_tx_descs = CFG_GET_NUM_TX_DESCS_NIC_IF(
+		    octeon_get_conf(octeon_dev), lio->ifidx);
+		retval = octeon_setup_iq(octeon_dev, ifidx, q,
+					 lio->linfo.txpciq[q], num_tx_descs,
+					 netdev_get_tx_queue(netdev, q));
+		if (retval) {
+			dev_err(&octeon_dev->pci_dev->dev,
+				" %s : Runtime IQ(TxQ) creation failed.\n",
+				__func__);
+			return 1;
+		}
+
+		/* XPS */
+		if (!OCTEON_CN23XX_VF(octeon_dev) && octeon_dev->msix_on &&
+		    octeon_dev->ioq_vector) {
+			struct octeon_ioq_vector    *ioq_vector;
+
+			ioq_vector = &octeon_dev->ioq_vector[q];
+			netif_set_xps_queue(netdev,
+					    &ioq_vector->affinity_mask,
+					    ioq_vector->iq_index);
+		}
+	}
+
+	return 0;
+}
+EXPORT_SYMBOL_GPL(liquidio_setup_io_queues);
+
+static
+int liquidio_schedule_msix_droq_pkt_handler(struct octeon_droq *droq, u64 ret)
+{
+	struct octeon_device *oct = droq->oct_dev;
+	struct octeon_device_priv *oct_priv =
+	    (struct octeon_device_priv *)oct->priv;
+
+	if (droq->ops.poll_mode) {
+		droq->ops.napi_fn(droq);
+	} else {
+		if (ret & MSIX_PO_INT) {
+			if (OCTEON_CN23XX_VF(oct))
+				dev_err(&oct->pci_dev->dev,
+					"should not come here should not get rx when poll mode = 0 for vf\n");
+			tasklet_schedule(&oct_priv->droq_tasklet);
+			return 1;
+		}
+		/* this will be flushed periodically by check iq db */
+		if (ret & MSIX_PI_INT)
+			return 0;
+	}
+
+	return 0;
+}
+
+irqreturn_t
+liquidio_msix_intr_handler(int __maybe_unused irq, void *dev)
+{
+	struct octeon_ioq_vector *ioq_vector = (struct octeon_ioq_vector *)dev;
+	struct octeon_device *oct = ioq_vector->oct_dev;
+	struct octeon_droq *droq = oct->droq[ioq_vector->droq_index];
+	u64 ret;
+
+	ret = oct->fn_list.msix_interrupt_handler(ioq_vector);
+
+	if (ret & MSIX_PO_INT || ret & MSIX_PI_INT)
+		liquidio_schedule_msix_droq_pkt_handler(droq, ret);
+
+	return IRQ_HANDLED;
+}
+
+/**
+ * liquidio_schedule_droq_pkt_handlers - Droq packet processor sceduler
+ * @oct: octeon device
+ */
+static void liquidio_schedule_droq_pkt_handlers(struct octeon_device *oct)
+{
+	struct octeon_device_priv *oct_priv =
+		(struct octeon_device_priv *)oct->priv;
 	struct octeon_droq *droq;
-	int q, q_no = 0;
+	u64 oq_no;
 
-	if (ifstate_check(lio, LIO_IFSTATE_RUNNING)) {
-		for (q = 0; q < lio->linfo.num_rxpciq; q++) {
-			q_no = lio->linfo.rxpciq[q].s.q_no;
-			droq = oct->droq[q_no];
-			if (!droq)
+	if (oct->int_status & OCT_DEV_INTR_PKT_DATA) {
+		for (oq_no = 0; oq_no < MAX_OCTEON_OUTPUT_QUEUES(oct);
+		     oq_no++) {
+			if (!(oct->droq_intr & BIT_ULL(oq_no)))
 				continue;
-			octeon_droq_check_oom(droq);
+
+			droq = oct->droq[oq_no];
+
+			if (droq->ops.poll_mode) {
+				droq->ops.napi_fn(droq);
+				oct_priv->napi_mask |= BIT_ULL(oq_no);
+			} else {
+				tasklet_schedule(&oct_priv->droq_tasklet);
+			}
 		}
 	}
-	queue_delayed_work(lio->rxq_status_wq.wq,
-			   &lio->rxq_status_wq.wk.work,
-			   msecs_to_jiffies(LIO_OOM_POLL_INTERVAL_MS));
 }
 
-int setup_rx_oom_poll_fn(struct net_device *netdev)
+/**
+ * liquidio_legacy_intr_handler - Interrupt handler for octeon
+ * @irq: unused
+ * @dev: octeon device
+ */
+static
+irqreturn_t liquidio_legacy_intr_handler(int __maybe_unused irq, void *dev)
+{
+	struct octeon_device *oct = (struct octeon_device *)dev;
+	irqreturn_t ret;
+
+	/* Disable our interrupts for the duration of ISR */
+	oct->fn_list.disable_interrupt(oct, OCTEON_ALL_INTR);
+
+	ret = oct->fn_list.process_interrupt_regs(oct);
+
+	if (ret == IRQ_HANDLED)
+		liquidio_schedule_droq_pkt_handlers(oct);
+
+	/* Re-enable our interrupts  */
+	if (!(atomic_read(&oct->status) == OCT_DEV_IN_RESET))
+		oct->fn_list.enable_interrupt(oct, OCTEON_ALL_INTR);
+
+	return ret;
+}
+
+/**
+ * octeon_setup_interrupt - Setup interrupt for octeon device
+ * @oct: octeon device
+ * @num_ioqs: number of queues
+ *
+ *  Enable interrupt in Octeon device as given in the PCI interrupt mask.
+ */
+int octeon_setup_interrupt(struct octeon_device *oct, u32 num_ioqs)
+{
+	struct msix_entry *msix_entries;
+	char *queue_irq_names = NULL;
+	int i, num_interrupts = 0;
+	int num_alloc_ioq_vectors;
+	char *aux_irq_name = NULL;
+	int num_ioq_vectors;
+	int irqret, err;
+
+	if (oct->msix_on) {
+		oct->num_msix_irqs = num_ioqs;
+		if (OCTEON_CN23XX_PF(oct)) {
+			num_interrupts = MAX_IOQ_INTERRUPTS_PER_PF + 1;
+
+			/* one non ioq interrupt for handling
+			 * sli_mac_pf_int_sum
+			 */
+			oct->num_msix_irqs += 1;
+		} else if (OCTEON_CN23XX_VF(oct)) {
+			num_interrupts = MAX_IOQ_INTERRUPTS_PER_VF;
+		}
+
+		/* allocate storage for the names assigned to each irq */
+		oct->irq_name_storage =
+			kcalloc(num_interrupts, INTRNAMSIZ, GFP_KERNEL);
+		if (!oct->irq_name_storage) {
+			dev_err(&oct->pci_dev->dev, "Irq name storage alloc failed...\n");
+			return -ENOMEM;
+		}
+
+		queue_irq_names = oct->irq_name_storage;
+
+		if (OCTEON_CN23XX_PF(oct))
+			aux_irq_name = &queue_irq_names
+				[IRQ_NAME_OFF(MAX_IOQ_INTERRUPTS_PER_PF)];
+
+		oct->msix_entries = kcalloc(oct->num_msix_irqs,
+					    sizeof(struct msix_entry),
+					    GFP_KERNEL);
+		if (!oct->msix_entries) {
+			dev_err(&oct->pci_dev->dev, "Memory Alloc failed...\n");
+			kfree(oct->irq_name_storage);
+			oct->irq_name_storage = NULL;
+			return -ENOMEM;
+		}
+
+		msix_entries = (struct msix_entry *)oct->msix_entries;
+
+		/*Assumption is that pf msix vectors start from pf srn to pf to
+		 * trs and not from 0. if not change this code
+		 */
+		if (OCTEON_CN23XX_PF(oct)) {
+			for (i = 0; i < oct->num_msix_irqs - 1; i++)
+				msix_entries[i].entry =
+					oct->sriov_info.pf_srn + i;
+
+			msix_entries[oct->num_msix_irqs - 1].entry =
+				oct->sriov_info.trs;
+		} else if (OCTEON_CN23XX_VF(oct)) {
+			for (i = 0; i < oct->num_msix_irqs; i++)
+				msix_entries[i].entry = i;
+		}
+		num_alloc_ioq_vectors = pci_enable_msix_range(
+						oct->pci_dev, msix_entries,
+						oct->num_msix_irqs,
+						oct->num_msix_irqs);
+		if (num_alloc_ioq_vectors < 0) {
+			dev_err(&oct->pci_dev->dev, "unable to Allocate MSI-X interrupts\n");
+			kfree(oct->msix_entries);
+			oct->msix_entries = NULL;
+			kfree(oct->irq_name_storage);
+			oct->irq_name_storage = NULL;
+			return num_alloc_ioq_vectors;
+		}
+
+		dev_dbg(&oct->pci_dev->dev, "OCTEON: Enough MSI-X interrupts are allocated...\n");
+
+		num_ioq_vectors = oct->num_msix_irqs;
+		/* For PF, there is one non-ioq interrupt handler */
+		if (OCTEON_CN23XX_PF(oct)) {
+			num_ioq_vectors -= 1;
+
+			snprintf(aux_irq_name, INTRNAMSIZ,
+				 "LiquidIO%u-pf%u-aux", oct->octeon_id,
+				 oct->pf_num);
+			irqret = request_irq(
+					msix_entries[num_ioq_vectors].vector,
+					liquidio_legacy_intr_handler, 0,
+					aux_irq_name, oct);
+			if (irqret) {
+				dev_err(&oct->pci_dev->dev,
+					"Request_irq failed for MSIX interrupt Error: %d\n",
+					irqret);
+				pci_disable_msix(oct->pci_dev);
+				kfree(oct->msix_entries);
+				kfree(oct->irq_name_storage);
+				oct->irq_name_storage = NULL;
+				oct->msix_entries = NULL;
+				return irqret;
+			}
+		}
+		for (i = 0 ; i < num_ioq_vectors ; i++) {
+			if (OCTEON_CN23XX_PF(oct))
+				snprintf(&queue_irq_names[IRQ_NAME_OFF(i)],
+					 INTRNAMSIZ, "LiquidIO%u-pf%u-rxtx-%u",
+					 oct->octeon_id, oct->pf_num, i);
+
+			if (OCTEON_CN23XX_VF(oct))
+				snprintf(&queue_irq_names[IRQ_NAME_OFF(i)],
+					 INTRNAMSIZ, "LiquidIO%u-vf%u-rxtx-%u",
+					 oct->octeon_id, oct->vf_num, i);
+
+			irqret = request_irq(msix_entries[i].vector,
+					     liquidio_msix_intr_handler, 0,
+					     &queue_irq_names[IRQ_NAME_OFF(i)],
+					     &oct->ioq_vector[i]);
+
+			if (irqret) {
+				dev_err(&oct->pci_dev->dev,
+					"Request_irq failed for MSIX interrupt Error: %d\n",
+					irqret);
+				/* Freeing the non-ioq irq vector here . */
+				free_irq(msix_entries[num_ioq_vectors].vector,
+					 oct);
+
+				while (i) {
+					i--;
+					/* clearing affinity mask. */
+					irq_set_affinity_hint(
+						      msix_entries[i].vector,
+						      NULL);
+					free_irq(msix_entries[i].vector,
+						 &oct->ioq_vector[i]);
+				}
+				pci_disable_msix(oct->pci_dev);
+				kfree(oct->msix_entries);
+				kfree(oct->irq_name_storage);
+				oct->irq_name_storage = NULL;
+				oct->msix_entries = NULL;
+				return irqret;
+			}
+			oct->ioq_vector[i].vector = msix_entries[i].vector;
+			/* assign the cpu mask for this msix interrupt vector */
+			irq_set_affinity_hint(msix_entries[i].vector,
+					      &oct->ioq_vector[i].affinity_mask
+					      );
+		}
+		dev_dbg(&oct->pci_dev->dev, "OCTEON[%d]: MSI-X enabled\n",
+			oct->octeon_id);
+	} else {
+		err = pci_enable_msi(oct->pci_dev);
+		if (err)
+			dev_warn(&oct->pci_dev->dev, "Reverting to legacy interrupts. Error: %d\n",
+				 err);
+		else
+			oct->flags |= LIO_FLAG_MSI_ENABLED;
+
+		/* allocate storage for the names assigned to the irq */
+		oct->irq_name_storage = kzalloc(INTRNAMSIZ, GFP_KERNEL);
+		if (!oct->irq_name_storage)
+			return -ENOMEM;
+
+		queue_irq_names = oct->irq_name_storage;
+
+		if (OCTEON_CN23XX_PF(oct))
+			snprintf(&queue_irq_names[IRQ_NAME_OFF(0)], INTRNAMSIZ,
+				 "LiquidIO%u-pf%u-rxtx-%u",
+				 oct->octeon_id, oct->pf_num, 0);
+
+		if (OCTEON_CN23XX_VF(oct))
+			snprintf(&queue_irq_names[IRQ_NAME_OFF(0)], INTRNAMSIZ,
+				 "LiquidIO%u-vf%u-rxtx-%u",
+				 oct->octeon_id, oct->vf_num, 0);
+
+		irqret = request_irq(oct->pci_dev->irq,
+				     liquidio_legacy_intr_handler,
+				     IRQF_SHARED,
+				     &queue_irq_names[IRQ_NAME_OFF(0)], oct);
+		if (irqret) {
+			if (oct->flags & LIO_FLAG_MSI_ENABLED)
+				pci_disable_msi(oct->pci_dev);
+			dev_err(&oct->pci_dev->dev, "Request IRQ failed with code: %d\n",
+				irqret);
+			kfree(oct->irq_name_storage);
+			oct->irq_name_storage = NULL;
+			return irqret;
+		}
+	}
+	return 0;
+}
+EXPORT_SYMBOL_GPL(octeon_setup_interrupt);
+
+/**
+ * liquidio_change_mtu - Net device change_mtu
+ * @netdev: network device
+ * @new_mtu: the new max transmit unit size
+ */
+int liquidio_change_mtu(struct net_device *netdev, int new_mtu)
 {
 	struct lio *lio = GET_LIO(netdev);
 	struct octeon_device *oct = lio->oct_dev;
+	struct octeon_soft_command *sc;
+	union octnet_cmd *ncmd;
+	int ret = 0;
 
-	lio->rxq_status_wq.wq = alloc_workqueue("rxq-oom-status",
-						WQ_MEM_RECLAIM, 0);
-	if (!lio->rxq_status_wq.wq) {
-		dev_err(&oct->pci_dev->dev, "unable to create cavium rxq oom status wq\n");
+	sc = (struct octeon_soft_command *)
+		octeon_alloc_soft_command(oct, OCTNET_CMD_SIZE, 16, 0);
+	if (!sc) {
+		netif_info(lio, rx_err, lio->netdev,
+			   "Failed to allocate soft command\n");
 		return -ENOMEM;
 	}
-	INIT_DELAYED_WORK(&lio->rxq_status_wq.wk.work,
-			  octnet_poll_check_rxq_oom_status);
-	lio->rxq_status_wq.wk.ctxptr = lio;
-	queue_delayed_work(lio->rxq_status_wq.wq,
-			   &lio->rxq_status_wq.wk.work,
-			   msecs_to_jiffies(LIO_OOM_POLL_INTERVAL_MS));
+
+	ncmd = (union octnet_cmd *)sc->virtdptr;
+
+	init_completion(&sc->complete);
+	sc->sc_status = OCTEON_REQUEST_PENDING;
+
+	ncmd->u64 = 0;
+	ncmd->s.cmd = OCTNET_CMD_CHANGE_MTU;
+	ncmd->s.param1 = new_mtu;
+
+	octeon_swap_8B_data((u64 *)ncmd, (OCTNET_CMD_SIZE >> 3));
+
+	sc->iq_no = lio->linfo.txpciq[0].s.q_no;
+
+	octeon_prepare_soft_command(oct, sc, OPCODE_NIC,
+				    OPCODE_NIC_CMD, 0, 0, 0);
+
+	ret = octeon_send_soft_command(oct, sc);
+	if (ret == IQ_SEND_FAILED) {
+		netif_info(lio, rx_err, lio->netdev, "Failed to change MTU\n");
+		octeon_free_soft_command(oct, sc);
+		return -EINVAL;
+	}
+	/* Sleep on a wait queue till the cond flag indicates that the
+	 * response arrived or timed-out.
+	 */
+	ret = wait_for_sc_completion_timeout(oct, sc, 0);
+	if (ret)
+		return ret;
+
+	if (sc->sc_status) {
+		WRITE_ONCE(sc->caller_is_done, true);
+		return -EINVAL;
+	}
+
+	WRITE_ONCE(netdev->mtu, new_mtu);
+	lio->mtu = new_mtu;
+
+	WRITE_ONCE(sc->caller_is_done, true);
 	return 0;
 }
+EXPORT_SYMBOL_GPL(liquidio_change_mtu);
 
-void cleanup_rx_oom_poll_fn(struct net_device *netdev)
+int lio_wait_for_clean_oq(struct octeon_device *oct)
 {
-	struct lio *lio = GET_LIO(netdev);
+	int retry = 100, pending_pkts = 0;
+	int idx;
+
+	do {
+		pending_pkts = 0;
+
+		for (idx = 0; idx < MAX_OCTEON_OUTPUT_QUEUES(oct); idx++) {
+			if (!(oct->io_qmask.oq & BIT_ULL(idx)))
+				continue;
+			pending_pkts +=
+				atomic_read(&oct->droq[idx]->pkts_pending);
+		}
+
+		if (pending_pkts > 0)
+			schedule_timeout_uninterruptible(1);
+
+	} while (retry-- && pending_pkts);
+
+	return pending_pkts;
+}
+EXPORT_SYMBOL_GPL(lio_wait_for_clean_oq);
+
+static void
+octnet_nic_stats_callback(struct octeon_device *oct_dev,
+			  u32 status, void *ptr)
+{
+	struct octeon_soft_command *sc = (struct octeon_soft_command *)ptr;
+	struct oct_nic_stats_resp *resp =
+	    (struct oct_nic_stats_resp *)sc->virtrptr;
+	struct nic_rx_stats *rsp_rstats = &resp->stats.fromwire;
+	struct nic_tx_stats *rsp_tstats = &resp->stats.fromhost;
+	struct nic_rx_stats *rstats = &oct_dev->link_stats.fromwire;
+	struct nic_tx_stats *tstats = &oct_dev->link_stats.fromhost;
+
+	if (status != OCTEON_REQUEST_TIMEOUT && !resp->status) {
+		octeon_swap_8B_data((u64 *)&resp->stats,
+				    (sizeof(struct oct_link_stats)) >> 3);
+
+		/* RX link-level stats */
+		rstats->total_rcvd = rsp_rstats->total_rcvd;
+		rstats->bytes_rcvd = rsp_rstats->bytes_rcvd;
+		rstats->total_bcst = rsp_rstats->total_bcst;
+		rstats->total_mcst = rsp_rstats->total_mcst;
+		rstats->runts      = rsp_rstats->runts;
+		rstats->ctl_rcvd   = rsp_rstats->ctl_rcvd;
+		/* Accounts for over/under-run of buffers */
+		rstats->fifo_err  = rsp_rstats->fifo_err;
+		rstats->dmac_drop = rsp_rstats->dmac_drop;
+		rstats->fcs_err   = rsp_rstats->fcs_err;
+		rstats->jabber_err = rsp_rstats->jabber_err;
+		rstats->l2_err    = rsp_rstats->l2_err;
+		rstats->frame_err = rsp_rstats->frame_err;
+		rstats->red_drops = rsp_rstats->red_drops;
+
+		/* RX firmware stats */
+		rstats->fw_total_rcvd = rsp_rstats->fw_total_rcvd;
+		rstats->fw_total_fwd = rsp_rstats->fw_total_fwd;
+		rstats->fw_total_mcast = rsp_rstats->fw_total_mcast;
+		rstats->fw_total_bcast = rsp_rstats->fw_total_bcast;
+		rstats->fw_err_pko = rsp_rstats->fw_err_pko;
+		rstats->fw_err_link = rsp_rstats->fw_err_link;
+		rstats->fw_err_drop = rsp_rstats->fw_err_drop;
+		rstats->fw_rx_vxlan = rsp_rstats->fw_rx_vxlan;
+		rstats->fw_rx_vxlan_err = rsp_rstats->fw_rx_vxlan_err;
+
+		/* Number of packets that are LROed      */
+		rstats->fw_lro_pkts = rsp_rstats->fw_lro_pkts;
+		/* Number of octets that are LROed       */
+		rstats->fw_lro_octs = rsp_rstats->fw_lro_octs;
+		/* Number of LRO packets formed          */
+		rstats->fw_total_lro = rsp_rstats->fw_total_lro;
+		/* Number of times lRO of packet aborted */
+		rstats->fw_lro_aborts = rsp_rstats->fw_lro_aborts;
+		rstats->fw_lro_aborts_port = rsp_rstats->fw_lro_aborts_port;
+		rstats->fw_lro_aborts_seq = rsp_rstats->fw_lro_aborts_seq;
+		rstats->fw_lro_aborts_tsval = rsp_rstats->fw_lro_aborts_tsval;
+		rstats->fw_lro_aborts_timer = rsp_rstats->fw_lro_aborts_timer;
+		/* intrmod: packet forward rate */
+		rstats->fwd_rate = rsp_rstats->fwd_rate;
+
+		/* TX link-level stats */
+		tstats->total_pkts_sent = rsp_tstats->total_pkts_sent;
+		tstats->total_bytes_sent = rsp_tstats->total_bytes_sent;
+		tstats->mcast_pkts_sent = rsp_tstats->mcast_pkts_sent;
+		tstats->bcast_pkts_sent = rsp_tstats->bcast_pkts_sent;
+		tstats->ctl_sent = rsp_tstats->ctl_sent;
+		/* Packets sent after one collision*/
+		tstats->one_collision_sent = rsp_tstats->one_collision_sent;
+		/* Packets sent after multiple collision*/
+		tstats->multi_collision_sent = rsp_tstats->multi_collision_sent;
+		/* Packets not sent due to max collisions */
+		tstats->max_collision_fail = rsp_tstats->max_collision_fail;
+		/* Packets not sent due to max deferrals */
+		tstats->max_deferral_fail = rsp_tstats->max_deferral_fail;
+		/* Accounts for over/under-run of buffers */
+		tstats->fifo_err = rsp_tstats->fifo_err;
+		tstats->runts = rsp_tstats->runts;
+		/* Total number of collisions detected */
+		tstats->total_collisions = rsp_tstats->total_collisions;
+
+		/* firmware stats */
+		tstats->fw_total_sent = rsp_tstats->fw_total_sent;
+		tstats->fw_total_fwd = rsp_tstats->fw_total_fwd;
+		tstats->fw_total_mcast_sent = rsp_tstats->fw_total_mcast_sent;
+		tstats->fw_total_bcast_sent = rsp_tstats->fw_total_bcast_sent;
+		tstats->fw_err_pko = rsp_tstats->fw_err_pko;
+		tstats->fw_err_pki = rsp_tstats->fw_err_pki;
+		tstats->fw_err_link = rsp_tstats->fw_err_link;
+		tstats->fw_err_drop = rsp_tstats->fw_err_drop;
+		tstats->fw_tso = rsp_tstats->fw_tso;
+		tstats->fw_tso_fwd = rsp_tstats->fw_tso_fwd;
+		tstats->fw_err_tso = rsp_tstats->fw_err_tso;
+		tstats->fw_tx_vxlan = rsp_tstats->fw_tx_vxlan;
+
+		resp->status = 1;
+	} else {
+		dev_err(&oct_dev->pci_dev->dev, "sc OPCODE_NIC_PORT_STATS command failed\n");
+		resp->status = -1;
+	}
+}
+
+static int lio_fetch_vf_stats(struct lio *lio)
+{
+	struct octeon_device *oct_dev = lio->oct_dev;
+	struct octeon_soft_command *sc;
+	struct oct_nic_vf_stats_resp *resp;
+
+	int retval;
 
-	if (lio->rxq_status_wq.wq) {
-		cancel_delayed_work_sync(&lio->rxq_status_wq.wk.work);
-		flush_workqueue(lio->rxq_status_wq.wq);
-		destroy_workqueue(lio->rxq_status_wq.wq);
+	/* Alloc soft command */
+	sc = (struct octeon_soft_command *)
+		octeon_alloc_soft_command(oct_dev,
+					  0,
+					  sizeof(struct oct_nic_vf_stats_resp),
+					  0);
+
+	if (!sc) {
+		dev_err(&oct_dev->pci_dev->dev, "Soft command allocation failed\n");
+		retval = -ENOMEM;
+		goto lio_fetch_vf_stats_exit;
+	}
+
+	resp = (struct oct_nic_vf_stats_resp *)sc->virtrptr;
+	memset(resp, 0, sizeof(struct oct_nic_vf_stats_resp));
+
+	init_completion(&sc->complete);
+	sc->sc_status = OCTEON_REQUEST_PENDING;
+
+	sc->iq_no = lio->linfo.txpciq[0].s.q_no;
+
+	octeon_prepare_soft_command(oct_dev, sc, OPCODE_NIC,
+				    OPCODE_NIC_VF_PORT_STATS, 0, 0, 0);
+
+	retval = octeon_send_soft_command(oct_dev, sc);
+	if (retval == IQ_SEND_FAILED) {
+		octeon_free_soft_command(oct_dev, sc);
+		goto lio_fetch_vf_stats_exit;
 	}
+
+	retval =
+		wait_for_sc_completion_timeout(oct_dev, sc,
+					       (2 * LIO_SC_MAX_TMO_MS));
+	if (retval)  {
+		dev_err(&oct_dev->pci_dev->dev,
+			"sc OPCODE_NIC_VF_PORT_STATS command failed\n");
+		goto lio_fetch_vf_stats_exit;
+	}
+
+	if (sc->sc_status != OCTEON_REQUEST_TIMEOUT && !resp->status) {
+		octeon_swap_8B_data((u64 *)&resp->spoofmac_cnt,
+				    (sizeof(u64)) >> 3);
+
+		if (resp->spoofmac_cnt != 0) {
+			dev_warn(&oct_dev->pci_dev->dev,
+				 "%llu Spoofed packets detected\n",
+				 resp->spoofmac_cnt);
+		}
+	}
+	WRITE_ONCE(sc->caller_is_done, 1);
+
+lio_fetch_vf_stats_exit:
+	return retval;
+}
+
+void lio_fetch_stats(struct work_struct *work)
+{
+	struct cavium_wk *wk = (struct cavium_wk *)work;
+	struct lio *lio = wk->ctxptr;
+	struct octeon_device *oct_dev = lio->oct_dev;
+	struct octeon_soft_command *sc;
+	struct oct_nic_stats_resp *resp;
+	unsigned long time_in_jiffies;
+	int retval;
+
+	if (OCTEON_CN23XX_PF(oct_dev)) {
+		/* report spoofchk every 2 seconds */
+		if (!(oct_dev->vfstats_poll % LIO_VFSTATS_POLL) &&
+		    (oct_dev->fw_info.app_cap_flags & LIQUIDIO_SPOOFCHK_CAP) &&
+		    oct_dev->sriov_info.num_vfs_alloced) {
+			lio_fetch_vf_stats(lio);
+		}
+
+		oct_dev->vfstats_poll++;
+	}
+
+	/* Alloc soft command */
+	sc = (struct octeon_soft_command *)
+		octeon_alloc_soft_command(oct_dev,
+					  0,
+					  sizeof(struct oct_nic_stats_resp),
+					  0);
+
+	if (!sc) {
+		dev_err(&oct_dev->pci_dev->dev, "Soft command allocation failed\n");
+		goto lio_fetch_stats_exit;
+	}
+
+	resp = (struct oct_nic_stats_resp *)sc->virtrptr;
+	memset(resp, 0, sizeof(struct oct_nic_stats_resp));
+
+	init_completion(&sc->complete);
+	sc->sc_status = OCTEON_REQUEST_PENDING;
+
+	sc->iq_no = lio->linfo.txpciq[0].s.q_no;
+
+	octeon_prepare_soft_command(oct_dev, sc, OPCODE_NIC,
+				    OPCODE_NIC_PORT_STATS, 0, 0, 0);
+
+	retval = octeon_send_soft_command(oct_dev, sc);
+	if (retval == IQ_SEND_FAILED) {
+		octeon_free_soft_command(oct_dev, sc);
+		goto lio_fetch_stats_exit;
+	}
+
+	retval = wait_for_sc_completion_timeout(oct_dev, sc,
+						(2 * LIO_SC_MAX_TMO_MS));
+	if (retval)  {
+		dev_err(&oct_dev->pci_dev->dev, "sc OPCODE_NIC_PORT_STATS command failed\n");
+		goto lio_fetch_stats_exit;
+	}
+
+	octnet_nic_stats_callback(oct_dev, sc->sc_status, sc);
+	WRITE_ONCE(sc->caller_is_done, true);
+
+lio_fetch_stats_exit:
+	time_in_jiffies = msecs_to_jiffies(LIQUIDIO_NDEV_STATS_POLL_TIME_MS);
+	if (ifstate_check(lio, LIO_IFSTATE_RUNNING))
+		schedule_delayed_work(&lio->stats_wk.work, time_in_jiffies);
+
+	return;
+}
+EXPORT_SYMBOL_GPL(lio_fetch_stats);
+
+int liquidio_set_speed(struct lio *lio, int speed)
+{
+	struct octeon_device *oct = lio->oct_dev;
+	struct oct_nic_seapi_resp *resp;
+	struct octeon_soft_command *sc;
+	union octnet_cmd *ncmd;
+	int retval;
+	u32 var;
+
+	if (oct->speed_setting == speed)
+		return 0;
+
+	if (!OCTEON_CN23XX_PF(oct)) {
+		dev_err(&oct->pci_dev->dev, "%s: SET SPEED only for PF\n",
+			__func__);
+		return -EOPNOTSUPP;
+	}
+
+	sc = octeon_alloc_soft_command(oct, OCTNET_CMD_SIZE,
+				       sizeof(struct oct_nic_seapi_resp),
+				       0);
+	if (!sc)
+		return -ENOMEM;
+
+	ncmd = sc->virtdptr;
+	resp = sc->virtrptr;
+	memset(resp, 0, sizeof(struct oct_nic_seapi_resp));
+
+	init_completion(&sc->complete);
+	sc->sc_status = OCTEON_REQUEST_PENDING;
+
+	ncmd->u64 = 0;
+	ncmd->s.cmd = SEAPI_CMD_SPEED_SET;
+	ncmd->s.param1 = speed;
+
+	octeon_swap_8B_data((u64 *)ncmd, (OCTNET_CMD_SIZE >> 3));
+
+	sc->iq_no = lio->linfo.txpciq[0].s.q_no;
+
+	octeon_prepare_soft_command(oct, sc, OPCODE_NIC,
+				    OPCODE_NIC_UBOOT_CTL, 0, 0, 0);
+
+	retval = octeon_send_soft_command(oct, sc);
+	if (retval == IQ_SEND_FAILED) {
+		dev_info(&oct->pci_dev->dev, "Failed to send soft command\n");
+		octeon_free_soft_command(oct, sc);
+		retval = -EBUSY;
+	} else {
+		/* Wait for response or timeout */
+		retval = wait_for_sc_completion_timeout(oct, sc, 0);
+		if (retval)
+			return retval;
+
+		retval = resp->status;
+
+		if (retval) {
+			dev_err(&oct->pci_dev->dev, "%s failed, retval=%d\n",
+				__func__, retval);
+			WRITE_ONCE(sc->caller_is_done, true);
+
+			return -EIO;
+		}
+
+		var = be32_to_cpu((__force __be32)resp->speed);
+		if (var != speed) {
+			dev_err(&oct->pci_dev->dev,
+				"%s: setting failed speed= %x, expect %x\n",
+				__func__, var, speed);
+		}
+
+		oct->speed_setting = var;
+		WRITE_ONCE(sc->caller_is_done, true);
+	}
+
+	return retval;
+}
+
+int liquidio_get_speed(struct lio *lio)
+{
+	struct octeon_device *oct = lio->oct_dev;
+	struct oct_nic_seapi_resp *resp;
+	struct octeon_soft_command *sc;
+	union octnet_cmd *ncmd;
+	int retval;
+
+	sc = octeon_alloc_soft_command(oct, OCTNET_CMD_SIZE,
+				       sizeof(struct oct_nic_seapi_resp),
+				       0);
+	if (!sc)
+		return -ENOMEM;
+
+	ncmd = sc->virtdptr;
+	resp = sc->virtrptr;
+	memset(resp, 0, sizeof(struct oct_nic_seapi_resp));
+
+	init_completion(&sc->complete);
+	sc->sc_status = OCTEON_REQUEST_PENDING;
+
+	ncmd->u64 = 0;
+	ncmd->s.cmd = SEAPI_CMD_SPEED_GET;
+
+	octeon_swap_8B_data((u64 *)ncmd, (OCTNET_CMD_SIZE >> 3));
+
+	sc->iq_no = lio->linfo.txpciq[0].s.q_no;
+
+	octeon_prepare_soft_command(oct, sc, OPCODE_NIC,
+				    OPCODE_NIC_UBOOT_CTL, 0, 0, 0);
+
+	retval = octeon_send_soft_command(oct, sc);
+	if (retval == IQ_SEND_FAILED) {
+		dev_info(&oct->pci_dev->dev, "Failed to send soft command\n");
+		octeon_free_soft_command(oct, sc);
+		retval = -EIO;
+	} else {
+		retval = wait_for_sc_completion_timeout(oct, sc, 0);
+		if (retval)
+			return retval;
+
+		retval = resp->status;
+		if (retval) {
+			dev_err(&oct->pci_dev->dev,
+				"%s failed retval=%d\n", __func__, retval);
+			retval = -EIO;
+		} else {
+			u32 var;
+
+			var = be32_to_cpu((__force __be32)resp->speed);
+			oct->speed_setting = var;
+			if (var == 0xffff) {
+				/* unable to access boot variables
+				 * get the default value based on the NIC type
+				 */
+				if (oct->subsystem_id ==
+						OCTEON_CN2350_25GB_SUBSYS_ID ||
+				    oct->subsystem_id ==
+						OCTEON_CN2360_25GB_SUBSYS_ID) {
+					oct->no_speed_setting = 1;
+					oct->speed_setting = 25;
+				} else {
+					oct->speed_setting = 10;
+				}
+			}
+
+		}
+		WRITE_ONCE(sc->caller_is_done, true);
+	}
+
+	return retval;
+}
+EXPORT_SYMBOL_GPL(liquidio_get_speed);
+
+int liquidio_set_fec(struct lio *lio, int on_off)
+{
+	struct oct_nic_seapi_resp *resp;
+	struct octeon_soft_command *sc;
+	struct octeon_device *oct;
+	union octnet_cmd *ncmd;
+	int retval;
+	u32 var;
+
+	oct = lio->oct_dev;
+
+	if (oct->props[lio->ifidx].fec == on_off)
+		return 0;
+
+	if (!OCTEON_CN23XX_PF(oct)) {
+		dev_err(&oct->pci_dev->dev, "%s: SET FEC only for PF\n",
+			__func__);
+		return -1;
+	}
+
+	if (oct->speed_boot != 25)  {
+		dev_err(&oct->pci_dev->dev,
+			"Set FEC only when link speed is 25G during insmod\n");
+		return -1;
+	}
+
+	sc = octeon_alloc_soft_command(oct, OCTNET_CMD_SIZE,
+				       sizeof(struct oct_nic_seapi_resp), 0);
+	if (!sc) {
+		dev_err(&oct->pci_dev->dev,
+			"Failed to allocate soft command\n");
+		return -ENOMEM;
+	}
+
+	ncmd = sc->virtdptr;
+	resp = sc->virtrptr;
+	memset(resp, 0, sizeof(struct oct_nic_seapi_resp));
+
+	init_completion(&sc->complete);
+	sc->sc_status = OCTEON_REQUEST_PENDING;
+
+	ncmd->u64 = 0;
+	ncmd->s.cmd = SEAPI_CMD_FEC_SET;
+	ncmd->s.param1 = on_off;
+	/* SEAPI_CMD_FEC_DISABLE(0) or SEAPI_CMD_FEC_RS(1) */
+
+	octeon_swap_8B_data((u64 *)ncmd, (OCTNET_CMD_SIZE >> 3));
+
+	sc->iq_no = lio->linfo.txpciq[0].s.q_no;
+
+	octeon_prepare_soft_command(oct, sc, OPCODE_NIC,
+				    OPCODE_NIC_UBOOT_CTL, 0, 0, 0);
+
+	retval = octeon_send_soft_command(oct, sc);
+	if (retval == IQ_SEND_FAILED) {
+		dev_info(&oct->pci_dev->dev, "Failed to send soft command\n");
+		octeon_free_soft_command(oct, sc);
+		return -EIO;
+	}
+
+	retval = wait_for_sc_completion_timeout(oct, sc, 0);
+	if (retval)
+		return (-EIO);
+
+	var = be32_to_cpu(resp->fec_setting);
+	resp->fec_setting = var;
+	if (var != on_off) {
+		dev_err(&oct->pci_dev->dev,
+			"Setting failed fec= %x, expect %x\n",
+			var, on_off);
+		oct->props[lio->ifidx].fec = var;
+		if (resp->fec_setting == SEAPI_CMD_FEC_SET_RS)
+			oct->props[lio->ifidx].fec = 1;
+		else
+			oct->props[lio->ifidx].fec = 0;
+	}
+
+	WRITE_ONCE(sc->caller_is_done, true);
+
+	if (oct->props[lio->ifidx].fec !=
+	    oct->props[lio->ifidx].fec_boot) {
+		dev_dbg(&oct->pci_dev->dev,
+			"Reload driver to change fec to %s\n",
+			oct->props[lio->ifidx].fec ? "on" : "off");
+	}
+
+	return retval;
+}
+
+int liquidio_get_fec(struct lio *lio)
+{
+	struct oct_nic_seapi_resp *resp;
+	struct octeon_soft_command *sc;
+	struct octeon_device *oct;
+	union octnet_cmd *ncmd;
+	int retval;
+	u32 var;
+
+	oct = lio->oct_dev;
+
+	sc = octeon_alloc_soft_command(oct, OCTNET_CMD_SIZE,
+				       sizeof(struct oct_nic_seapi_resp), 0);
+	if (!sc)
+		return -ENOMEM;
+
+	ncmd = sc->virtdptr;
+	resp = sc->virtrptr;
+	memset(resp, 0, sizeof(struct oct_nic_seapi_resp));
+
+	init_completion(&sc->complete);
+	sc->sc_status = OCTEON_REQUEST_PENDING;
+
+	ncmd->u64 = 0;
+	ncmd->s.cmd = SEAPI_CMD_FEC_GET;
+
+	octeon_swap_8B_data((u64 *)ncmd, (OCTNET_CMD_SIZE >> 3));
+
+	sc->iq_no = lio->linfo.txpciq[0].s.q_no;
+
+	octeon_prepare_soft_command(oct, sc, OPCODE_NIC,
+				    OPCODE_NIC_UBOOT_CTL, 0, 0, 0);
+
+	retval = octeon_send_soft_command(oct, sc);
+	if (retval == IQ_SEND_FAILED) {
+		dev_info(&oct->pci_dev->dev,
+			 "%s: Failed to send soft command\n", __func__);
+		octeon_free_soft_command(oct, sc);
+		return -EIO;
+	}
+
+	retval = wait_for_sc_completion_timeout(oct, sc, 0);
+	if (retval)
+		return retval;
+
+	var = be32_to_cpu(resp->fec_setting);
+	resp->fec_setting = var;
+	if (resp->fec_setting == SEAPI_CMD_FEC_SET_RS)
+		oct->props[lio->ifidx].fec = 1;
+	else
+		oct->props[lio->ifidx].fec = 0;
+
+	WRITE_ONCE(sc->caller_is_done, true);
+
+	if (oct->props[lio->ifidx].fec !=
+	    oct->props[lio->ifidx].fec_boot) {
+		dev_dbg(&oct->pci_dev->dev,
+			"Reload driver to change fec to %s\n",
+			oct->props[lio->ifidx].fec ? "on" : "off");
+	}
+
+	return retval;
 }
+EXPORT_SYMBOL_GPL(liquidio_get_fec);
diff --git a/drivers/net/ethernet/cavium/liquidio/lio_ethtool.c b/drivers/net/ethernet/cavium/liquidio/lio_ethtool.c
index 579dc7336f58..c849e2c871a9 100644
--- a/drivers/net/ethernet/cavium/liquidio/lio_ethtool.c
+++ b/drivers/net/ethernet/cavium/liquidio/lio_ethtool.c
@@ -15,6 +15,7 @@
  * of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE, TITLE, or
  * NONINFRINGEMENT.  See the GNU General Public License for more details.
  ***********************************************************************/
+#include <linux/ethtool.h>
 #include <linux/netdevice.h>
 #include <linux/net_tstamp.h>
 #include <linux/pci.h>
@@ -31,14 +32,7 @@
 #include "cn23xx_pf_device.h"
 #include "cn23xx_vf_device.h"
 
-static int octnet_get_link_stats(struct net_device *netdev);
-
-struct oct_intrmod_context {
-	int octeon_id;
-	wait_queue_head_t wc;
-	int cond;
-	int status;
-};
+static int lio_reset_queues(struct net_device *netdev, uint32_t num_qs);
 
 struct oct_intrmod_resp {
 	u64     rh;
@@ -46,12 +40,6 @@ struct oct_intrmod_resp {
 	u64     status;
 };
 
-struct oct_mdio_cmd_context {
-	int octeon_id;
-	wait_queue_head_t wc;
-	int cond;
-};
-
 struct oct_mdio_cmd_resp {
 	u64 rh;
 	struct oct_mdio_cmd resp;
@@ -95,16 +83,15 @@ static const char oct_stats_strings[][ETH_GSTRING_LEN] = {
 	"tx_packets",
 	"rx_bytes",
 	"tx_bytes",
-	"rx_errors",	/*jabber_err+l2_err+frame_err */
-	"tx_errors",	/*fw_err_pko+fw_err_link+fw_err_drop */
-	"rx_dropped",   /*st->fromwire.total_rcvd - st->fromwire.fw_total_rcvd +
-			 *st->fromwire.dmac_drop + st->fromwire.fw_err_drop
-			 */
+	"rx_errors",
+	"tx_errors",
+	"rx_dropped",
 	"tx_dropped",
 
 	"tx_total_sent",
 	"tx_total_fwd",
 	"tx_err_pko",
+	"tx_err_pki",
 	"tx_err_link",
 	"tx_err_drop",
 
@@ -113,21 +100,26 @@ static const char oct_stats_strings[][ETH_GSTRING_LEN] = {
 	"tx_tso_err",
 	"tx_vxlan",
 
+	"tx_mcast",
+	"tx_bcast",
+
 	"mac_tx_total_pkts",
 	"mac_tx_total_bytes",
 	"mac_tx_mcast_pkts",
 	"mac_tx_bcast_pkts",
-	"mac_tx_ctl_packets",	/*oct->link_stats.fromhost.ctl_sent */
+	"mac_tx_ctl_packets",
 	"mac_tx_total_collisions",
 	"mac_tx_one_collision",
-	"mac_tx_multi_collison",
+	"mac_tx_multi_collision",
 	"mac_tx_max_collision_fail",
-	"mac_tx_max_deferal_fail",
+	"mac_tx_max_deferral_fail",
 	"mac_tx_fifo_err",
 	"mac_tx_runts",
 
 	"rx_total_rcvd",
 	"rx_total_fwd",
+	"rx_mcast",
+	"rx_bcast",
 	"rx_jabber_err",
 	"rx_l2_err",
 	"rx_frame_err",
@@ -168,17 +160,21 @@ static const char oct_vf_stats_strings[][ETH_GSTRING_LEN] = {
 	"tx_packets",
 	"rx_bytes",
 	"tx_bytes",
-	"rx_errors", /* jabber_err + l2_err+frame_err */
-	"tx_errors", /* fw_err_pko + fw_err_link+fw_err_drop */
-	"rx_dropped", /* total_rcvd - fw_total_rcvd + dmac_drop + fw_err_drop */
+	"rx_errors",
+	"tx_errors",
+	"rx_dropped",
 	"tx_dropped",
+	"rx_mcast",
+	"tx_mcast",
+	"rx_bcast",
+	"tx_bcast",
 	"link_state_changes",
 };
 
 /* statistics of host tx queue */
 static const char oct_iq_stats_strings[][ETH_GSTRING_LEN] = {
-	"packets",		/*oct->instr_queue[iq_no]->stats.tx_done*/
-	"bytes",		/*oct->instr_queue[iq_no]->stats.tx_tot_bytes*/
+	"packets",
+	"bytes",
 	"dropped",
 	"iq_busy",
 	"sgentry_sent",
@@ -195,13 +191,9 @@ static const char oct_iq_stats_strings[][ETH_GSTRING_LEN] = {
 
 /* statistics of host rx queue */
 static const char oct_droq_stats_strings[][ETH_GSTRING_LEN] = {
-	"packets",		/*oct->droq[oq_no]->stats.rx_pkts_received */
-	"bytes",		/*oct->droq[oq_no]->stats.rx_bytes_received */
-	"dropped",		/*oct->droq[oq_no]->stats.rx_dropped+
-				 *oct->droq[oq_no]->stats.dropped_nodispatch+
-				 *oct->droq[oq_no]->stats.dropped_toomany+
-				 *oct->droq[oq_no]->stats.dropped_nomem
-				 */
+	"packets",
+	"bytes",
+	"dropped",
 	"dropped_nomem",
 	"dropped_toomany",
 	"fw_dropped",
@@ -226,32 +218,166 @@ static int lio_get_link_ksettings(struct net_device *netdev,
 	struct lio *lio = GET_LIO(netdev);
 	struct octeon_device *oct = lio->oct_dev;
 	struct oct_link_info *linfo;
-	u32 supported = 0, advertising = 0;
 
 	linfo = &lio->linfo;
 
-	if (linfo->link.s.if_mode == INTERFACE_MODE_XAUI ||
-	    linfo->link.s.if_mode == INTERFACE_MODE_RXAUI ||
-	    linfo->link.s.if_mode == INTERFACE_MODE_XLAUI ||
-	    linfo->link.s.if_mode == INTERFACE_MODE_XFI) {
-		ecmd->base.port = PORT_FIBRE;
+	ethtool_link_ksettings_zero_link_mode(ecmd, supported);
+	ethtool_link_ksettings_zero_link_mode(ecmd, advertising);
 
-		if (linfo->link.s.speed == SPEED_10000) {
-			supported = SUPPORTED_10000baseT_Full;
-			advertising = ADVERTISED_10000baseT_Full;
+	switch (linfo->link.s.phy_type) {
+	case LIO_PHY_PORT_TP:
+		ecmd->base.port = PORT_TP;
+		ecmd->base.autoneg = AUTONEG_DISABLE;
+		ethtool_link_ksettings_add_link_mode(ecmd, supported, TP);
+		ethtool_link_ksettings_add_link_mode(ecmd, supported, Pause);
+		ethtool_link_ksettings_add_link_mode(ecmd, supported,
+						     10000baseT_Full);
+
+		ethtool_link_ksettings_add_link_mode(ecmd, advertising, Pause);
+		ethtool_link_ksettings_add_link_mode(ecmd, advertising,
+						     10000baseT_Full);
+
+		break;
+
+	case LIO_PHY_PORT_FIBRE:
+		if (linfo->link.s.if_mode == INTERFACE_MODE_XAUI ||
+		    linfo->link.s.if_mode == INTERFACE_MODE_RXAUI ||
+		    linfo->link.s.if_mode == INTERFACE_MODE_XLAUI ||
+		    linfo->link.s.if_mode == INTERFACE_MODE_XFI) {
+			dev_dbg(&oct->pci_dev->dev, "ecmd->base.transceiver is XCVR_EXTERNAL\n");
+			ecmd->base.transceiver = XCVR_EXTERNAL;
+		} else {
+			dev_err(&oct->pci_dev->dev, "Unknown link interface mode: %d\n",
+				linfo->link.s.if_mode);
 		}
 
-		supported |= SUPPORTED_FIBRE | SUPPORTED_Pause;
-		advertising |= ADVERTISED_Pause;
-		ethtool_convert_legacy_u32_to_link_mode(
-			ecmd->link_modes.supported, supported);
-		ethtool_convert_legacy_u32_to_link_mode(
-			ecmd->link_modes.advertising, advertising);
+		ecmd->base.port = PORT_FIBRE;
 		ecmd->base.autoneg = AUTONEG_DISABLE;
-
-	} else {
-		dev_err(&oct->pci_dev->dev, "Unknown link interface reported %d\n",
-			linfo->link.s.if_mode);
+		ethtool_link_ksettings_add_link_mode(ecmd, supported, FIBRE);
+
+		ethtool_link_ksettings_add_link_mode(ecmd, supported, Pause);
+		ethtool_link_ksettings_add_link_mode(ecmd, advertising, Pause);
+		if (oct->subsystem_id == OCTEON_CN2350_25GB_SUBSYS_ID ||
+		    oct->subsystem_id == OCTEON_CN2360_25GB_SUBSYS_ID) {
+			if (OCTEON_CN23XX_PF(oct)) {
+				ethtool_link_ksettings_add_link_mode
+					(ecmd, supported, 25000baseSR_Full);
+				ethtool_link_ksettings_add_link_mode
+					(ecmd, supported, 25000baseKR_Full);
+				ethtool_link_ksettings_add_link_mode
+					(ecmd, supported, 25000baseCR_Full);
+
+				if (oct->no_speed_setting == 0)  {
+					ethtool_link_ksettings_add_link_mode
+						(ecmd, supported,
+						 10000baseSR_Full);
+					ethtool_link_ksettings_add_link_mode
+						(ecmd, supported,
+						 10000baseKR_Full);
+					ethtool_link_ksettings_add_link_mode
+						(ecmd, supported,
+						 10000baseCR_Full);
+				}
+
+				if (oct->no_speed_setting == 0) {
+					liquidio_get_speed(lio);
+					liquidio_get_fec(lio);
+				} else {
+					oct->speed_setting = 25;
+				}
+
+				if (oct->speed_setting == 10) {
+					ethtool_link_ksettings_add_link_mode
+						(ecmd, advertising,
+						 10000baseSR_Full);
+					ethtool_link_ksettings_add_link_mode
+						(ecmd, advertising,
+						 10000baseKR_Full);
+					ethtool_link_ksettings_add_link_mode
+						(ecmd, advertising,
+						 10000baseCR_Full);
+				}
+				if (oct->speed_setting == 25) {
+					ethtool_link_ksettings_add_link_mode
+						(ecmd, advertising,
+						 25000baseSR_Full);
+					ethtool_link_ksettings_add_link_mode
+						(ecmd, advertising,
+						 25000baseKR_Full);
+					ethtool_link_ksettings_add_link_mode
+						(ecmd, advertising,
+						 25000baseCR_Full);
+				}
+
+				if (oct->no_speed_setting)
+					break;
+
+				ethtool_link_ksettings_add_link_mode
+					(ecmd, supported, FEC_RS);
+				ethtool_link_ksettings_add_link_mode
+					(ecmd, supported, FEC_NONE);
+					/*FEC_OFF*/
+				if (oct->props[lio->ifidx].fec == 1) {
+					/* ETHTOOL_FEC_RS */
+					ethtool_link_ksettings_add_link_mode
+						(ecmd, advertising, FEC_RS);
+				} else {
+					/* ETHTOOL_FEC_OFF */
+					ethtool_link_ksettings_add_link_mode
+						(ecmd, advertising, FEC_NONE);
+				}
+			} else { /* VF */
+				if (linfo->link.s.speed == 10000) {
+					ethtool_link_ksettings_add_link_mode
+						(ecmd, supported,
+						 10000baseSR_Full);
+					ethtool_link_ksettings_add_link_mode
+						(ecmd, supported,
+						 10000baseKR_Full);
+					ethtool_link_ksettings_add_link_mode
+						(ecmd, supported,
+						 10000baseCR_Full);
+
+					ethtool_link_ksettings_add_link_mode
+						(ecmd, advertising,
+						 10000baseSR_Full);
+					ethtool_link_ksettings_add_link_mode
+						(ecmd, advertising,
+						 10000baseKR_Full);
+					ethtool_link_ksettings_add_link_mode
+						(ecmd, advertising,
+						 10000baseCR_Full);
+				}
+
+				if (linfo->link.s.speed == 25000) {
+					ethtool_link_ksettings_add_link_mode
+						(ecmd, supported,
+						 25000baseSR_Full);
+					ethtool_link_ksettings_add_link_mode
+						(ecmd, supported,
+						 25000baseKR_Full);
+					ethtool_link_ksettings_add_link_mode
+						(ecmd, supported,
+						 25000baseCR_Full);
+
+					ethtool_link_ksettings_add_link_mode
+						(ecmd, advertising,
+						 25000baseSR_Full);
+					ethtool_link_ksettings_add_link_mode
+						(ecmd, advertising,
+						 25000baseKR_Full);
+					ethtool_link_ksettings_add_link_mode
+						(ecmd, advertising,
+						 25000baseCR_Full);
+				}
+			}
+		} else {
+			ethtool_link_ksettings_add_link_mode(ecmd, supported,
+							     10000baseT_Full);
+			ethtool_link_ksettings_add_link_mode(ecmd, advertising,
+							     10000baseT_Full);
+		}
+		break;
 	}
 
 	if (linfo->link.s.link_up) {
@@ -265,6 +391,47 @@ static int lio_get_link_ksettings(struct net_device *netdev,
 	return 0;
 }
 
+static int lio_set_link_ksettings(struct net_device *netdev,
+				  const struct ethtool_link_ksettings *ecmd)
+{
+	const int speed = ecmd->base.speed;
+	struct lio *lio = GET_LIO(netdev);
+	struct oct_link_info *linfo;
+	struct octeon_device *oct;
+
+	oct = lio->oct_dev;
+
+	linfo = &lio->linfo;
+
+	if (!(oct->subsystem_id == OCTEON_CN2350_25GB_SUBSYS_ID ||
+	      oct->subsystem_id == OCTEON_CN2360_25GB_SUBSYS_ID))
+		return -EOPNOTSUPP;
+
+	if (oct->no_speed_setting) {
+		dev_err(&oct->pci_dev->dev, "%s: Changing speed is not supported\n",
+			__func__);
+		return -EOPNOTSUPP;
+	}
+
+	if ((ecmd->base.duplex != DUPLEX_UNKNOWN &&
+	     ecmd->base.duplex != linfo->link.s.duplex) ||
+	     ecmd->base.autoneg != AUTONEG_DISABLE ||
+	    (ecmd->base.speed != 10000 && ecmd->base.speed != 25000 &&
+	     ecmd->base.speed != SPEED_UNKNOWN))
+		return -EOPNOTSUPP;
+
+	if ((oct->speed_boot == speed / 1000) &&
+	    oct->speed_boot == oct->speed_setting)
+		return 0;
+
+	liquidio_set_speed(lio, speed / 1000);
+
+	dev_dbg(&oct->pci_dev->dev, "Port speed is set to %dG\n",
+		oct->speed_setting);
+
+	return 0;
+}
+
 static void
 lio_get_drvinfo(struct net_device *netdev, struct ethtool_drvinfo *drvinfo)
 {
@@ -275,11 +442,11 @@ lio_get_drvinfo(struct net_device *netdev, struct ethtool_drvinfo *drvinfo)
 	oct = lio->oct_dev;
 
 	memset(drvinfo, 0, sizeof(struct ethtool_drvinfo));
-	strcpy(drvinfo->driver, "liquidio");
-	strcpy(drvinfo->version, LIQUIDIO_VERSION);
-	strncpy(drvinfo->fw_version, oct->fw_info.liquidio_firmware_version,
-		ETHTOOL_FWVERS_LEN);
-	strncpy(drvinfo->bus_info, pci_name(oct->pci_dev), 32);
+	strscpy(drvinfo->driver, "liquidio", sizeof(drvinfo->driver));
+	strscpy(drvinfo->fw_version, oct->fw_info.liquidio_firmware_version,
+		sizeof(drvinfo->fw_version));
+	strscpy(drvinfo->bus_info, pci_name(oct->pci_dev),
+		sizeof(drvinfo->bus_info));
 }
 
 static void
@@ -292,11 +459,39 @@ lio_get_vf_drvinfo(struct net_device *netdev, struct ethtool_drvinfo *drvinfo)
 	oct = lio->oct_dev;
 
 	memset(drvinfo, 0, sizeof(struct ethtool_drvinfo));
-	strcpy(drvinfo->driver, "liquidio_vf");
-	strcpy(drvinfo->version, LIQUIDIO_VERSION);
-	strncpy(drvinfo->fw_version, oct->fw_info.liquidio_firmware_version,
-		ETHTOOL_FWVERS_LEN);
-	strncpy(drvinfo->bus_info, pci_name(oct->pci_dev), 32);
+	strscpy(drvinfo->driver, "liquidio_vf", sizeof(drvinfo->driver));
+	strscpy(drvinfo->fw_version, oct->fw_info.liquidio_firmware_version,
+		sizeof(drvinfo->fw_version));
+	strscpy(drvinfo->bus_info, pci_name(oct->pci_dev),
+		sizeof(drvinfo->bus_info));
+}
+
+static int
+lio_send_queue_count_update(struct net_device *netdev, uint32_t num_queues)
+{
+	struct lio *lio = GET_LIO(netdev);
+	struct octeon_device *oct = lio->oct_dev;
+	struct octnic_ctrl_pkt nctrl;
+	int ret = 0;
+
+	memset(&nctrl, 0, sizeof(struct octnic_ctrl_pkt));
+
+	nctrl.ncmd.u64 = 0;
+	nctrl.ncmd.s.cmd = OCTNET_CMD_QUEUE_COUNT_CTL;
+	nctrl.ncmd.s.param1 = num_queues;
+	nctrl.ncmd.s.param2 = num_queues;
+	nctrl.iq_no = lio->linfo.txpciq[0].s.q_no;
+	nctrl.netpndev = (u64)netdev;
+	nctrl.cb_fn = liquidio_link_ctrl_cmd_completion;
+
+	ret = octnet_send_nic_ctrl_pkt(lio->oct_dev, &nctrl);
+	if (ret) {
+		dev_err(&oct->pci_dev->dev, "Failed to send Queue reset command (ret: 0x%x)\n",
+			ret);
+		return -1;
+	}
+
+	return 0;
 }
 
 static void
@@ -306,6 +501,7 @@ lio_ethtool_get_channels(struct net_device *dev,
 	struct lio *lio = GET_LIO(dev);
 	struct octeon_device *oct = lio->oct_dev;
 	u32 max_rx = 0, max_tx = 0, tx_count = 0, rx_count = 0;
+	u32 combined_count = 0, max_combined = 0;
 
 	if (OCTEON_CN6XXX(oct)) {
 		struct octeon_config *conf6x = CHIP_CONF(oct, cn6xxx);
@@ -315,27 +511,163 @@ lio_ethtool_get_channels(struct net_device *dev,
 		rx_count = CFG_GET_NUM_RXQS_NIC_IF(conf6x, lio->ifidx);
 		tx_count = CFG_GET_NUM_TXQS_NIC_IF(conf6x, lio->ifidx);
 	} else if (OCTEON_CN23XX_PF(oct)) {
+		if (oct->sriov_info.sriov_enabled) {
+			max_combined = lio->linfo.num_txpciq;
+		} else {
+			struct octeon_config *conf23_pf =
+				CHIP_CONF(oct, cn23xx_pf);
 
-		max_rx = oct->sriov_info.num_pf_rings;
-		max_tx = oct->sriov_info.num_pf_rings;
-		rx_count = lio->linfo.num_rxpciq;
-		tx_count = lio->linfo.num_txpciq;
+			max_combined = CFG_GET_IQ_MAX_Q(conf23_pf);
+		}
+		combined_count = oct->num_iqs;
 	} else if (OCTEON_CN23XX_VF(oct)) {
-		max_tx = oct->sriov_info.rings_per_vf;
-		max_rx = oct->sriov_info.rings_per_vf;
-		rx_count = lio->linfo.num_rxpciq;
-		tx_count = lio->linfo.num_txpciq;
+		u64 reg_val = 0ULL;
+		u64 ctrl = CN23XX_VF_SLI_IQ_PKT_CONTROL64(0);
+
+		reg_val = octeon_read_csr64(oct, ctrl);
+		reg_val = reg_val >> CN23XX_PKT_INPUT_CTL_RPVF_POS;
+		max_combined = reg_val & CN23XX_PKT_INPUT_CTL_RPVF_MASK;
+		combined_count = oct->num_iqs;
 	}
 
 	channel->max_rx = max_rx;
 	channel->max_tx = max_tx;
+	channel->max_combined = max_combined;
 	channel->rx_count = rx_count;
 	channel->tx_count = tx_count;
+	channel->combined_count = combined_count;
+}
+
+static int
+lio_irq_reallocate_irqs(struct octeon_device *oct, uint32_t num_ioqs)
+{
+	struct msix_entry *msix_entries;
+	int num_msix_irqs = 0;
+	int i;
+
+	if (!oct->msix_on)
+		return 0;
+
+	/* Disable the input and output queues now. No more packets will
+	 * arrive from Octeon.
+	 */
+	oct->fn_list.disable_interrupt(oct, OCTEON_ALL_INTR);
+
+	if (oct->msix_on) {
+		if (OCTEON_CN23XX_PF(oct))
+			num_msix_irqs = oct->num_msix_irqs - 1;
+		else if (OCTEON_CN23XX_VF(oct))
+			num_msix_irqs = oct->num_msix_irqs;
+
+		msix_entries = (struct msix_entry *)oct->msix_entries;
+		for (i = 0; i < num_msix_irqs; i++) {
+			if (oct->ioq_vector[i].vector) {
+				/* clear the affinity_cpumask */
+				irq_set_affinity_hint(msix_entries[i].vector,
+						      NULL);
+				free_irq(msix_entries[i].vector,
+					 &oct->ioq_vector[i]);
+				oct->ioq_vector[i].vector = 0;
+			}
+		}
+
+		/* non-iov vector's argument is oct struct */
+		if (OCTEON_CN23XX_PF(oct))
+			free_irq(msix_entries[i].vector, oct);
+
+		pci_disable_msix(oct->pci_dev);
+		kfree(oct->msix_entries);
+		oct->msix_entries = NULL;
+	}
+
+	kfree(oct->irq_name_storage);
+	oct->irq_name_storage = NULL;
+
+	if (octeon_allocate_ioq_vector(oct, num_ioqs)) {
+		dev_err(&oct->pci_dev->dev, "OCTEON: ioq vector allocation failed\n");
+		return -1;
+	}
+
+	if (octeon_setup_interrupt(oct, num_ioqs)) {
+		dev_info(&oct->pci_dev->dev, "Setup interrupt failed\n");
+		return -1;
+	}
+
+	/* Enable Octeon device interrupts */
+	oct->fn_list.enable_interrupt(oct, OCTEON_ALL_INTR);
+
+	return 0;
+}
+
+static int
+lio_ethtool_set_channels(struct net_device *dev,
+			 struct ethtool_channels *channel)
+{
+	u32 combined_count, max_combined;
+	struct lio *lio = GET_LIO(dev);
+	struct octeon_device *oct = lio->oct_dev;
+	int stopped = 0;
+
+	if (strcmp(oct->fw_info.liquidio_firmware_version, "1.6.1") < 0) {
+		dev_err(&oct->pci_dev->dev, "Minimum firmware version required is 1.6.1\n");
+		return -EINVAL;
+	}
+
+	if (!channel->combined_count || channel->other_count ||
+	    channel->rx_count || channel->tx_count)
+		return -EINVAL;
+
+	combined_count = channel->combined_count;
+
+	if (OCTEON_CN23XX_PF(oct)) {
+		if (oct->sriov_info.sriov_enabled) {
+			max_combined = lio->linfo.num_txpciq;
+		} else {
+			struct octeon_config *conf23_pf =
+				CHIP_CONF(oct,
+					  cn23xx_pf);
+
+			max_combined =
+				CFG_GET_IQ_MAX_Q(conf23_pf);
+		}
+	} else if (OCTEON_CN23XX_VF(oct)) {
+		u64 reg_val = 0ULL;
+		u64 ctrl = CN23XX_VF_SLI_IQ_PKT_CONTROL64(0);
+
+		reg_val = octeon_read_csr64(oct, ctrl);
+		reg_val = reg_val >> CN23XX_PKT_INPUT_CTL_RPVF_POS;
+		max_combined = reg_val & CN23XX_PKT_INPUT_CTL_RPVF_MASK;
+	} else {
+		return -EINVAL;
+	}
+
+	if (combined_count > max_combined || combined_count < 1)
+		return -EINVAL;
+
+	if (combined_count == oct->num_iqs)
+		return 0;
+
+	ifstate_set(lio, LIO_IFSTATE_RESETTING);
+
+	if (netif_running(dev)) {
+		dev->netdev_ops->ndo_stop(dev);
+		stopped = 1;
+	}
+
+	if (lio_reset_queues(dev, combined_count))
+		return -EINVAL;
+
+	if (stopped)
+		dev->netdev_ops->ndo_open(dev);
+
+	ifstate_reset(lio, LIO_IFSTATE_RESETTING);
+
+	return 0;
 }
 
 static int lio_get_eeprom_len(struct net_device *netdev)
 {
-	u8 buf[128];
+	u8 buf[192];
 	struct lio *lio = GET_LIO(netdev);
 	struct octeon_device *oct_dev = lio->oct_dev;
 	struct octeon_board_info *board_info;
@@ -384,13 +716,13 @@ static int octnet_gpio_access(struct net_device *netdev, int addr, int val)
 	nctrl.ncmd.s.param1 = addr;
 	nctrl.ncmd.s.param2 = val;
 	nctrl.iq_no = lio->linfo.txpciq[0].s.q_no;
-	nctrl.wait_time = 100;
 	nctrl.netpndev = (u64)netdev;
 	nctrl.cb_fn = liquidio_link_ctrl_cmd_completion;
 
 	ret = octnet_send_nic_ctrl_pkt(lio->oct_dev, &nctrl);
-	if (ret < 0) {
-		dev_err(&oct->pci_dev->dev, "Failed to configure gpio value\n");
+	if (ret) {
+		dev_err(&oct->pci_dev->dev,
+			"Failed to configure gpio value, ret=%d\n", ret);
 		return -EINVAL;
 	}
 
@@ -410,41 +742,19 @@ static int octnet_id_active(struct net_device *netdev, int val)
 	nctrl.ncmd.s.cmd = OCTNET_CMD_ID_ACTIVE;
 	nctrl.ncmd.s.param1 = val;
 	nctrl.iq_no = lio->linfo.txpciq[0].s.q_no;
-	nctrl.wait_time = 100;
 	nctrl.netpndev = (u64)netdev;
 	nctrl.cb_fn = liquidio_link_ctrl_cmd_completion;
 
 	ret = octnet_send_nic_ctrl_pkt(lio->oct_dev, &nctrl);
-	if (ret < 0) {
-		dev_err(&oct->pci_dev->dev, "Failed to configure gpio value\n");
+	if (ret) {
+		dev_err(&oct->pci_dev->dev,
+			"Failed to configure gpio value, ret=%d\n", ret);
 		return -EINVAL;
 	}
 
 	return 0;
 }
 
-/* Callback for when mdio command response arrives
- */
-static void octnet_mdio_resp_callback(struct octeon_device *oct,
-				      u32 status,
-				      void *buf)
-{
-	struct oct_mdio_cmd_context *mdio_cmd_ctx;
-	struct octeon_soft_command *sc = (struct octeon_soft_command *)buf;
-
-	mdio_cmd_ctx = (struct oct_mdio_cmd_context *)sc->ctxptr;
-
-	oct = lio_get_device(mdio_cmd_ctx->octeon_id);
-	if (status) {
-		dev_err(&oct->pci_dev->dev, "MIDO instruction failed. Status: %llx\n",
-			CVM_CAST64(status));
-		WRITE_ONCE(mdio_cmd_ctx->cond, -1);
-	} else {
-		WRITE_ONCE(mdio_cmd_ctx->cond, 1);
-	}
-	wake_up_interruptible(&mdio_cmd_ctx->wc);
-}
-
 /* This routine provides PHY access routines for
  * mdio  clause45 .
  */
@@ -454,25 +764,20 @@ octnet_mdio45_access(struct lio *lio, int op, int loc, int *value)
 	struct octeon_device *oct_dev = lio->oct_dev;
 	struct octeon_soft_command *sc;
 	struct oct_mdio_cmd_resp *mdio_cmd_rsp;
-	struct oct_mdio_cmd_context *mdio_cmd_ctx;
 	struct oct_mdio_cmd *mdio_cmd;
 	int retval = 0;
 
 	sc = (struct octeon_soft_command *)
 		octeon_alloc_soft_command(oct_dev,
 					  sizeof(struct oct_mdio_cmd),
-					  sizeof(struct oct_mdio_cmd_resp),
-					  sizeof(struct oct_mdio_cmd_context));
+					  sizeof(struct oct_mdio_cmd_resp), 0);
 
 	if (!sc)
 		return -ENOMEM;
 
-	mdio_cmd_ctx = (struct oct_mdio_cmd_context *)sc->ctxptr;
 	mdio_cmd_rsp = (struct oct_mdio_cmd_resp *)sc->virtrptr;
 	mdio_cmd = (struct oct_mdio_cmd *)sc->virtdptr;
 
-	WRITE_ONCE(mdio_cmd_ctx->cond, 0);
-	mdio_cmd_ctx->octeon_id = lio_get_device_id(oct_dev);
 	mdio_cmd->op = op;
 	mdio_cmd->mdio_addr = loc;
 	if (op)
@@ -484,42 +789,40 @@ octnet_mdio45_access(struct lio *lio, int op, int loc, int *value)
 	octeon_prepare_soft_command(oct_dev, sc, OPCODE_NIC, OPCODE_NIC_MDIO45,
 				    0, 0, 0);
 
-	sc->wait_time = 1000;
-	sc->callback = octnet_mdio_resp_callback;
-	sc->callback_arg = sc;
-
-	init_waitqueue_head(&mdio_cmd_ctx->wc);
+	init_completion(&sc->complete);
+	sc->sc_status = OCTEON_REQUEST_PENDING;
 
 	retval = octeon_send_soft_command(oct_dev, sc);
-
 	if (retval == IQ_SEND_FAILED) {
 		dev_err(&oct_dev->pci_dev->dev,
 			"octnet_mdio45_access instruction failed status: %x\n",
 			retval);
-		retval = -EBUSY;
+		octeon_free_soft_command(oct_dev, sc);
+		return -EBUSY;
 	} else {
 		/* Sleep on a wait queue till the cond flag indicates that the
 		 * response arrived
 		 */
-		sleep_cond(&mdio_cmd_ctx->wc, &mdio_cmd_ctx->cond);
+		retval = wait_for_sc_completion_timeout(oct_dev, sc, 0);
+		if (retval)
+			return retval;
+
 		retval = mdio_cmd_rsp->status;
 		if (retval) {
-			dev_err(&oct_dev->pci_dev->dev, "octnet mdio45 access failed\n");
-			retval = -EBUSY;
-		} else {
-			octeon_swap_8B_data((u64 *)(&mdio_cmd_rsp->resp),
-					    sizeof(struct oct_mdio_cmd) / 8);
-
-			if (READ_ONCE(mdio_cmd_ctx->cond) == 1) {
-				if (!op)
-					*value = mdio_cmd_rsp->resp.value1;
-			} else {
-				retval = -EINVAL;
-			}
+			dev_err(&oct_dev->pci_dev->dev,
+				"octnet mdio45 access failed: %x\n", retval);
+			WRITE_ONCE(sc->caller_is_done, true);
+			return -EBUSY;
 		}
-	}
 
-	octeon_free_soft_command(oct_dev, sc);
+		octeon_swap_8B_data((u64 *)(&mdio_cmd_rsp->resp),
+				    sizeof(struct oct_mdio_cmd) / 8);
+
+		if (!op)
+			*value = mdio_cmd_rsp->resp.value1;
+
+		WRITE_ONCE(sc->caller_is_done, true);
+	}
 
 	return retval;
 }
@@ -529,7 +832,14 @@ static int lio_set_phys_id(struct net_device *netdev,
 {
 	struct lio *lio = GET_LIO(netdev);
 	struct octeon_device *oct = lio->oct_dev;
+	struct oct_link_info *linfo;
 	int value, ret;
+	u32 cur_ver;
+
+	linfo = &lio->linfo;
+	cur_ver = OCT_FW_VER(oct->fw_info.ver.maj,
+			     oct->fw_info.ver.min,
+			     oct->fw_info.ver.rev);
 
 	switch (state) {
 	case ETHTOOL_ID_ACTIVE:
@@ -568,32 +878,37 @@ static int lio_set_phys_id(struct net_device *netdev,
 				return ret;
 		} else if (oct->chip_id == OCTEON_CN23XX_PF_VID) {
 			octnet_id_active(netdev, LED_IDENTIFICATION_ON);
-
-			/* returns 0 since updates are asynchronous */
-			return 0;
+			if (linfo->link.s.phy_type == LIO_PHY_PORT_TP &&
+			    cur_ver > OCT_FW_VER(1, 7, 2))
+				return 2;
+			else
+				return 0;
 		} else {
 			return -EINVAL;
 		}
 		break;
 
 	case ETHTOOL_ID_ON:
-		if (oct->chip_id == OCTEON_CN66XX) {
+		if (oct->chip_id == OCTEON_CN23XX_PF_VID &&
+		    linfo->link.s.phy_type == LIO_PHY_PORT_TP &&
+		    cur_ver > OCT_FW_VER(1, 7, 2))
+			octnet_id_active(netdev, LED_IDENTIFICATION_ON);
+		else if (oct->chip_id == OCTEON_CN66XX)
 			octnet_gpio_access(netdev, VITESSE_PHY_GPIO_CFG,
 					   VITESSE_PHY_GPIO_HIGH);
-
-		} else if (oct->chip_id == OCTEON_CN68XX) {
-			return -EINVAL;
-		} else {
+		else
 			return -EINVAL;
-		}
+
 		break;
 
 	case ETHTOOL_ID_OFF:
-		if (oct->chip_id == OCTEON_CN66XX)
+		if (oct->chip_id == OCTEON_CN23XX_PF_VID &&
+		    linfo->link.s.phy_type == LIO_PHY_PORT_TP &&
+		    cur_ver > OCT_FW_VER(1, 7, 2))
+			octnet_id_active(netdev, LED_IDENTIFICATION_OFF);
+		else if (oct->chip_id == OCTEON_CN66XX)
 			octnet_gpio_access(netdev, VITESSE_PHY_GPIO_CFG,
 					   VITESSE_PHY_GPIO_LOW);
-		else if (oct->chip_id == OCTEON_CN68XX)
-			return -EINVAL;
 		else
 			return -EINVAL;
 
@@ -634,13 +949,18 @@ static int lio_set_phys_id(struct net_device *netdev,
 
 static void
 lio_ethtool_get_ringparam(struct net_device *netdev,
-			  struct ethtool_ringparam *ering)
+			  struct ethtool_ringparam *ering,
+			  struct kernel_ethtool_ringparam *kernel_ering,
+			  struct netlink_ext_ack *extack)
 {
 	struct lio *lio = GET_LIO(netdev);
 	struct octeon_device *oct = lio->oct_dev;
 	u32 tx_max_pending = 0, rx_max_pending = 0, tx_pending = 0,
 	    rx_pending = 0;
 
+	if (ifstate_check(lio, LIO_IFSTATE_RESETTING))
+		return;
+
 	if (OCTEON_CN6XXX(oct)) {
 		struct octeon_config *conf6x = CHIP_CONF(oct, cn6xxx);
 
@@ -648,33 +968,355 @@ lio_ethtool_get_ringparam(struct net_device *netdev,
 		rx_max_pending = CN6XXX_MAX_OQ_DESCRIPTORS;
 		rx_pending = CFG_GET_NUM_RX_DESCS_NIC_IF(conf6x, lio->ifidx);
 		tx_pending = CFG_GET_NUM_TX_DESCS_NIC_IF(conf6x, lio->ifidx);
-	} else if (OCTEON_CN23XX_PF(oct)) {
-		struct octeon_config *conf23 = CHIP_CONF(oct, cn23xx_pf);
-
+	} else if (OCTEON_CN23XX_PF(oct) || OCTEON_CN23XX_VF(oct)) {
 		tx_max_pending = CN23XX_MAX_IQ_DESCRIPTORS;
 		rx_max_pending = CN23XX_MAX_OQ_DESCRIPTORS;
-		rx_pending = CFG_GET_NUM_RX_DESCS_NIC_IF(conf23, lio->ifidx);
-		tx_pending = CFG_GET_NUM_TX_DESCS_NIC_IF(conf23, lio->ifidx);
-	}
-
-	if (lio->mtu > OCTNET_DEFAULT_FRM_SIZE - OCTNET_FRM_HEADER_SIZE) {
-		ering->rx_pending = 0;
-		ering->rx_max_pending = 0;
-		ering->rx_mini_pending = 0;
-		ering->rx_jumbo_pending = rx_pending;
-		ering->rx_mini_max_pending = 0;
-		ering->rx_jumbo_max_pending = rx_max_pending;
-	} else {
-		ering->rx_pending = rx_pending;
-		ering->rx_max_pending = rx_max_pending;
-		ering->rx_mini_pending = 0;
-		ering->rx_jumbo_pending = 0;
-		ering->rx_mini_max_pending = 0;
-		ering->rx_jumbo_max_pending = 0;
+		rx_pending = oct->droq[0]->max_count;
+		tx_pending = oct->instr_queue[0]->max_count;
 	}
 
 	ering->tx_pending = tx_pending;
 	ering->tx_max_pending = tx_max_pending;
+	ering->rx_pending = rx_pending;
+	ering->rx_max_pending = rx_max_pending;
+	ering->rx_mini_pending = 0;
+	ering->rx_jumbo_pending = 0;
+	ering->rx_mini_max_pending = 0;
+	ering->rx_jumbo_max_pending = 0;
+}
+
+static int lio_23xx_reconfigure_queue_count(struct lio *lio)
+{
+	struct octeon_device *oct = lio->oct_dev;
+	u32 resp_size, data_size;
+	struct liquidio_if_cfg_resp *resp;
+	struct octeon_soft_command *sc;
+	union oct_nic_if_cfg if_cfg;
+	struct lio_version *vdata;
+	u32 ifidx_or_pfnum;
+	int retval;
+	int j;
+
+	resp_size = sizeof(struct liquidio_if_cfg_resp);
+	data_size = sizeof(struct lio_version);
+	sc = (struct octeon_soft_command *)
+		octeon_alloc_soft_command(oct, data_size,
+					  resp_size, 0);
+	if (!sc) {
+		dev_err(&oct->pci_dev->dev, "%s: Failed to allocate soft command\n",
+			__func__);
+		return -1;
+	}
+
+	resp = (struct liquidio_if_cfg_resp *)sc->virtrptr;
+	vdata = (struct lio_version *)sc->virtdptr;
+
+	vdata->major = (__force u16)cpu_to_be16(LIQUIDIO_BASE_MAJOR_VERSION);
+	vdata->minor = (__force u16)cpu_to_be16(LIQUIDIO_BASE_MINOR_VERSION);
+	vdata->micro = (__force u16)cpu_to_be16(LIQUIDIO_BASE_MICRO_VERSION);
+
+	ifidx_or_pfnum = oct->pf_num;
+
+	if_cfg.u64 = 0;
+	if_cfg.s.num_iqueues = oct->sriov_info.num_pf_rings;
+	if_cfg.s.num_oqueues = oct->sriov_info.num_pf_rings;
+	if_cfg.s.base_queue = oct->sriov_info.pf_srn;
+	if_cfg.s.gmx_port_id = oct->pf_num;
+
+	sc->iq_no = 0;
+	octeon_prepare_soft_command(oct, sc, OPCODE_NIC,
+				    OPCODE_NIC_QCOUNT_UPDATE, 0,
+				    if_cfg.u64, 0);
+
+	init_completion(&sc->complete);
+	sc->sc_status = OCTEON_REQUEST_PENDING;
+
+	retval = octeon_send_soft_command(oct, sc);
+	if (retval == IQ_SEND_FAILED) {
+		dev_err(&oct->pci_dev->dev,
+			"Sending iq/oq config failed status: %x\n",
+			retval);
+		octeon_free_soft_command(oct, sc);
+		return -EIO;
+	}
+
+	retval = wait_for_sc_completion_timeout(oct, sc, 0);
+	if (retval)
+		return retval;
+
+	retval = resp->status;
+	if (retval) {
+		dev_err(&oct->pci_dev->dev,
+			"iq/oq config failed: %x\n", retval);
+		WRITE_ONCE(sc->caller_is_done, true);
+		return -1;
+	}
+
+	octeon_swap_8B_data((u64 *)(&resp->cfg_info),
+			    (sizeof(struct liquidio_if_cfg_info)) >> 3);
+
+	lio->ifidx = ifidx_or_pfnum;
+	lio->linfo.num_rxpciq = hweight64(resp->cfg_info.iqmask);
+	lio->linfo.num_txpciq = hweight64(resp->cfg_info.iqmask);
+	for (j = 0; j < lio->linfo.num_rxpciq; j++) {
+		lio->linfo.rxpciq[j].u64 =
+			resp->cfg_info.linfo.rxpciq[j].u64;
+	}
+
+	for (j = 0; j < lio->linfo.num_txpciq; j++) {
+		lio->linfo.txpciq[j].u64 =
+			resp->cfg_info.linfo.txpciq[j].u64;
+	}
+
+	lio->linfo.hw_addr = resp->cfg_info.linfo.hw_addr;
+	lio->linfo.gmxport = resp->cfg_info.linfo.gmxport;
+	lio->linfo.link.u64 = resp->cfg_info.linfo.link.u64;
+	lio->txq = lio->linfo.txpciq[0].s.q_no;
+	lio->rxq = lio->linfo.rxpciq[0].s.q_no;
+
+	dev_info(&oct->pci_dev->dev, "Queue count updated to %d\n",
+		 lio->linfo.num_rxpciq);
+
+	WRITE_ONCE(sc->caller_is_done, true);
+
+	return 0;
+}
+
+static int lio_reset_queues(struct net_device *netdev, uint32_t num_qs)
+{
+	struct lio *lio = GET_LIO(netdev);
+	struct octeon_device *oct = lio->oct_dev;
+	int i, queue_count_update = 0;
+	struct napi_struct *napi, *n;
+	int ret;
+
+	schedule_timeout_uninterruptible(msecs_to_jiffies(100));
+
+	if (wait_for_pending_requests(oct))
+		dev_err(&oct->pci_dev->dev, "There were pending requests\n");
+
+	if (lio_wait_for_instr_fetch(oct))
+		dev_err(&oct->pci_dev->dev, "IQ had pending instructions\n");
+
+	if (octeon_set_io_queues_off(oct)) {
+		dev_err(&oct->pci_dev->dev, "Setting io queues off failed\n");
+		return -1;
+	}
+
+	/* Disable the input and output queues now. No more packets will
+	 * arrive from Octeon.
+	 */
+	oct->fn_list.disable_io_queues(oct);
+	/* Delete NAPI */
+	list_for_each_entry_safe(napi, n, &netdev->napi_list, dev_list)
+		netif_napi_del(napi);
+
+	if (num_qs != oct->num_iqs) {
+		ret = netif_set_real_num_rx_queues(netdev, num_qs);
+		if (ret) {
+			dev_err(&oct->pci_dev->dev,
+				"Setting real number rx failed\n");
+			return ret;
+		}
+
+		ret = netif_set_real_num_tx_queues(netdev, num_qs);
+		if (ret) {
+			dev_err(&oct->pci_dev->dev,
+				"Setting real number tx failed\n");
+			return ret;
+		}
+
+		/* The value of queue_count_update decides whether it is the
+		 * queue count or the descriptor count that is being
+		 * re-configured.
+		 */
+		queue_count_update = 1;
+	}
+
+	/* Re-configuration of queues can happen in two scenarios, SRIOV enabled
+	 * and SRIOV disabled. Few things like recreating queue zero, resetting
+	 * glists and IRQs are required for both. For the latter, some more
+	 * steps like updating sriov_info for the octeon device need to be done.
+	 */
+	if (queue_count_update) {
+		cleanup_rx_oom_poll_fn(netdev);
+
+		lio_delete_glists(lio);
+
+		/* Delete mbox for PF which is SRIOV disabled because sriov_info
+		 * will be now changed.
+		 */
+		if ((OCTEON_CN23XX_PF(oct)) && !oct->sriov_info.sriov_enabled)
+			oct->fn_list.free_mbox(oct);
+	}
+
+	for (i = 0; i < MAX_OCTEON_OUTPUT_QUEUES(oct); i++) {
+		if (!(oct->io_qmask.oq & BIT_ULL(i)))
+			continue;
+		octeon_delete_droq(oct, i);
+	}
+
+	for (i = 0; i < MAX_OCTEON_INSTR_QUEUES(oct); i++) {
+		if (!(oct->io_qmask.iq & BIT_ULL(i)))
+			continue;
+		octeon_delete_instr_queue(oct, i);
+	}
+
+	if (queue_count_update) {
+		/* For PF re-configure sriov related information */
+		if ((OCTEON_CN23XX_PF(oct)) &&
+		    !oct->sriov_info.sriov_enabled) {
+			oct->sriov_info.num_pf_rings = num_qs;
+			if (cn23xx_sriov_config(oct)) {
+				dev_err(&oct->pci_dev->dev,
+					"Queue reset aborted: SRIOV config failed\n");
+				return -1;
+			}
+
+			num_qs = oct->sriov_info.num_pf_rings;
+		}
+	}
+
+	if (oct->fn_list.setup_device_regs(oct)) {
+		dev_err(&oct->pci_dev->dev, "Failed to configure device registers\n");
+		return -1;
+	}
+
+	/* The following are needed in case of queue count re-configuration and
+	 * not for descriptor count re-configuration.
+	 */
+	if (queue_count_update) {
+		if (octeon_setup_instr_queues(oct))
+			return -1;
+
+		if (octeon_setup_output_queues(oct))
+			return -1;
+
+		/* Recreating mbox for PF that is SRIOV disabled */
+		if (OCTEON_CN23XX_PF(oct) && !oct->sriov_info.sriov_enabled) {
+			if (oct->fn_list.setup_mbox(oct)) {
+				dev_err(&oct->pci_dev->dev, "Mailbox setup failed\n");
+				return -1;
+			}
+		}
+
+		/* Deleting and recreating IRQs whether the interface is SRIOV
+		 * enabled or disabled.
+		 */
+		if (lio_irq_reallocate_irqs(oct, num_qs)) {
+			dev_err(&oct->pci_dev->dev, "IRQs could not be allocated\n");
+			return -1;
+		}
+
+		/* Enable the input and output queues for this Octeon device */
+		if (oct->fn_list.enable_io_queues(oct)) {
+			dev_err(&oct->pci_dev->dev, "Failed to enable input/output queues\n");
+			return -1;
+		}
+
+		for (i = 0; i < oct->num_oqs; i++)
+			writel(oct->droq[i]->max_count,
+			       oct->droq[i]->pkts_credit_reg);
+
+		/* Informing firmware about the new queue count. It is required
+		 * for firmware to allocate more number of queues than those at
+		 * load time.
+		 */
+		if (OCTEON_CN23XX_PF(oct) && !oct->sriov_info.sriov_enabled) {
+			if (lio_23xx_reconfigure_queue_count(lio))
+				return -1;
+		}
+	}
+
+	/* Once firmware is aware of the new value, queues can be recreated */
+	if (liquidio_setup_io_queues(oct, 0, num_qs, num_qs)) {
+		dev_err(&oct->pci_dev->dev, "I/O queues creation failed\n");
+		return -1;
+	}
+
+	if (queue_count_update) {
+		if (lio_setup_glists(oct, lio, num_qs)) {
+			dev_err(&oct->pci_dev->dev, "Gather list allocation failed\n");
+			return -1;
+		}
+
+		if (setup_rx_oom_poll_fn(netdev)) {
+			dev_err(&oct->pci_dev->dev, "lio_setup_rx_oom_poll_fn failed\n");
+			return 1;
+		}
+
+		/* Send firmware the information about new number of queues
+		 * if the interface is a VF or a PF that is SRIOV enabled.
+		 */
+		if (oct->sriov_info.sriov_enabled || OCTEON_CN23XX_VF(oct))
+			if (lio_send_queue_count_update(netdev, num_qs))
+				return -1;
+	}
+
+	return 0;
+}
+
+static int
+lio_ethtool_set_ringparam(struct net_device *netdev,
+			  struct ethtool_ringparam *ering,
+			  struct kernel_ethtool_ringparam *kernel_ering,
+			  struct netlink_ext_ack *extack)
+{
+	u32 rx_count, tx_count, rx_count_old, tx_count_old;
+	struct lio *lio = GET_LIO(netdev);
+	struct octeon_device *oct = lio->oct_dev;
+	int stopped = 0;
+
+	if (!OCTEON_CN23XX_PF(oct) && !OCTEON_CN23XX_VF(oct))
+		return -EINVAL;
+
+	if (ering->rx_mini_pending || ering->rx_jumbo_pending)
+		return -EINVAL;
+
+	rx_count = clamp_t(u32, ering->rx_pending, CN23XX_MIN_OQ_DESCRIPTORS,
+			   CN23XX_MAX_OQ_DESCRIPTORS);
+	tx_count = clamp_t(u32, ering->tx_pending, CN23XX_MIN_IQ_DESCRIPTORS,
+			   CN23XX_MAX_IQ_DESCRIPTORS);
+
+	rx_count_old = oct->droq[0]->max_count;
+	tx_count_old = oct->instr_queue[0]->max_count;
+
+	if (rx_count == rx_count_old && tx_count == tx_count_old)
+		return 0;
+
+	ifstate_set(lio, LIO_IFSTATE_RESETTING);
+
+	if (netif_running(netdev)) {
+		netdev->netdev_ops->ndo_stop(netdev);
+		stopped = 1;
+	}
+
+	/* Change RX/TX DESCS  count */
+	if (tx_count != tx_count_old)
+		CFG_SET_NUM_TX_DESCS_NIC_IF(octeon_get_conf(oct), lio->ifidx,
+					    tx_count);
+	if (rx_count != rx_count_old)
+		CFG_SET_NUM_RX_DESCS_NIC_IF(octeon_get_conf(oct), lio->ifidx,
+					    rx_count);
+
+	if (lio_reset_queues(netdev, oct->num_iqs))
+		goto err_lio_reset_queues;
+
+	if (stopped)
+		netdev->netdev_ops->ndo_open(netdev);
+
+	ifstate_reset(lio, LIO_IFSTATE_RESETTING);
+
+	return 0;
+
+err_lio_reset_queues:
+	if (tx_count != tx_count_old)
+		CFG_SET_NUM_TX_DESCS_NIC_IF(octeon_get_conf(oct), lio->ifidx,
+					    tx_count_old);
+	if (rx_count != rx_count_old)
+		CFG_SET_NUM_RX_DESCS_NIC_IF(octeon_get_conf(oct), lio->ifidx,
+					    rx_count_old);
+	return -EINVAL;
 }
 
 static u32 lio_get_msglevel(struct net_device *netdev)
@@ -700,6 +1342,13 @@ static void lio_set_msglevel(struct net_device *netdev, u32 msglvl)
 	lio->msg_enable = msglvl;
 }
 
+static void lio_vf_set_msglevel(struct net_device *netdev, u32 msglvl)
+{
+	struct lio *lio = GET_LIO(netdev);
+
+	lio->msg_enable = msglvl;
+}
+
 static void
 lio_get_pauseparam(struct net_device *netdev, struct ethtool_pauseparam *pause)
 {
@@ -746,7 +1395,6 @@ lio_set_pauseparam(struct net_device *netdev, struct ethtool_pauseparam *pause)
 	nctrl.ncmd.u64 = 0;
 	nctrl.ncmd.s.cmd = OCTNET_CMD_SET_FLOW_CTL;
 	nctrl.iq_no = lio->linfo.txpciq[0].s.q_no;
-	nctrl.wait_time = 100;
 	nctrl.netpndev = (u64)netdev;
 	nctrl.cb_fn = liquidio_link_ctrl_cmd_completion;
 
@@ -767,8 +1415,9 @@ lio_set_pauseparam(struct net_device *netdev, struct ethtool_pauseparam *pause)
 	}
 
 	ret = octnet_send_nic_ctrl_pkt(lio->oct_dev, &nctrl);
-	if (ret < 0) {
-		dev_err(&oct->pci_dev->dev, "Failed to set pause parameter\n");
+	if (ret) {
+		dev_err(&oct->pci_dev->dev,
+			"Failed to set pause parameter, ret=%d\n", ret);
 		return -EINVAL;
 	}
 
@@ -785,30 +1434,48 @@ lio_get_ethtool_stats(struct net_device *netdev,
 {
 	struct lio *lio = GET_LIO(netdev);
 	struct octeon_device *oct_dev = lio->oct_dev;
-	struct net_device_stats *netstats = &netdev->stats;
+	struct rtnl_link_stats64 lstats;
 	int i = 0, j;
 
-	netdev->netdev_ops->ndo_get_stats(netdev);
-	octnet_get_link_stats(netdev);
+	if (ifstate_check(lio, LIO_IFSTATE_RESETTING))
+		return;
 
+	netdev->netdev_ops->ndo_get_stats64(netdev, &lstats);
 	/*sum of oct->droq[oq_no]->stats->rx_pkts_received */
-	data[i++] = CVM_CAST64(netstats->rx_packets);
+	data[i++] = lstats.rx_packets;
 	/*sum of oct->instr_queue[iq_no]->stats.tx_done */
-	data[i++] = CVM_CAST64(netstats->tx_packets);
+	data[i++] = lstats.tx_packets;
 	/*sum of oct->droq[oq_no]->stats->rx_bytes_received */
-	data[i++] = CVM_CAST64(netstats->rx_bytes);
+	data[i++] = lstats.rx_bytes;
 	/*sum of oct->instr_queue[iq_no]->stats.tx_tot_bytes */
-	data[i++] = CVM_CAST64(netstats->tx_bytes);
-	data[i++] = CVM_CAST64(netstats->rx_errors);
-	data[i++] = CVM_CAST64(netstats->tx_errors);
+	data[i++] = lstats.tx_bytes;
+	data[i++] = lstats.rx_errors +
+			oct_dev->link_stats.fromwire.fcs_err +
+			oct_dev->link_stats.fromwire.jabber_err +
+			oct_dev->link_stats.fromwire.l2_err +
+			oct_dev->link_stats.fromwire.frame_err;
+	data[i++] = lstats.tx_errors;
 	/*sum of oct->droq[oq_no]->stats->rx_dropped +
 	 *oct->droq[oq_no]->stats->dropped_nodispatch +
 	 *oct->droq[oq_no]->stats->dropped_toomany +
 	 *oct->droq[oq_no]->stats->dropped_nomem
 	 */
-	data[i++] = CVM_CAST64(netstats->rx_dropped);
+	data[i++] = lstats.rx_dropped +
+			oct_dev->link_stats.fromwire.fifo_err +
+			oct_dev->link_stats.fromwire.dmac_drop +
+			oct_dev->link_stats.fromwire.red_drops +
+			oct_dev->link_stats.fromwire.fw_err_pko +
+			oct_dev->link_stats.fromwire.fw_err_link +
+			oct_dev->link_stats.fromwire.fw_err_drop;
 	/*sum of oct->instr_queue[iq_no]->stats.tx_dropped */
-	data[i++] = CVM_CAST64(netstats->tx_dropped);
+	data[i++] = lstats.tx_dropped +
+			oct_dev->link_stats.fromhost.max_collision_fail +
+			oct_dev->link_stats.fromhost.max_deferral_fail +
+			oct_dev->link_stats.fromhost.total_collisions +
+			oct_dev->link_stats.fromhost.fw_err_pko +
+			oct_dev->link_stats.fromhost.fw_err_link +
+			oct_dev->link_stats.fromhost.fw_err_drop +
+			oct_dev->link_stats.fromhost.fw_err_pki;
 
 	/* firmware tx stats */
 	/*per_core_stats[cvmx_get_core_num()].link_stats[mdata->from_ifidx].
@@ -819,6 +1486,8 @@ lio_get_ethtool_stats(struct net_device *netdev,
 	data[i++] = CVM_CAST64(oct_dev->link_stats.fromhost.fw_total_fwd);
 	/*per_core_stats[j].link_stats[i].fromhost.fw_err_pko */
 	data[i++] = CVM_CAST64(oct_dev->link_stats.fromhost.fw_err_pko);
+	/*per_core_stats[j].link_stats[i].fromhost.fw_err_pki */
+	data[i++] = CVM_CAST64(oct_dev->link_stats.fromhost.fw_err_pki);
 	/*per_core_stats[j].link_stats[i].fromhost.fw_err_link */
 	data[i++] = CVM_CAST64(oct_dev->link_stats.fromhost.fw_err_link);
 	/*per_core_stats[cvmx_get_core_num()].link_stats[idx].fromhost.
@@ -841,6 +1510,10 @@ lio_get_ethtool_stats(struct net_device *netdev,
 	 */
 	data[i++] = CVM_CAST64(oct_dev->link_stats.fromhost.fw_tx_vxlan);
 
+	/* Multicast packets sent by this port */
+	data[i++] = oct_dev->link_stats.fromhost.fw_total_mcast_sent;
+	data[i++] = oct_dev->link_stats.fromhost.fw_total_bcast_sent;
+
 	/* mac tx statistics */
 	/*CVMX_BGXX_CMRX_TX_STAT5 */
 	data[i++] = CVM_CAST64(oct_dev->link_stats.fromhost.total_pkts_sent);
@@ -877,6 +1550,9 @@ lio_get_ethtool_stats(struct net_device *netdev,
 	 *fw_total_fwd
 	 */
 	data[i++] = CVM_CAST64(oct_dev->link_stats.fromwire.fw_total_fwd);
+	/* Multicast packets received on this port */
+	data[i++] = oct_dev->link_stats.fromwire.fw_total_mcast;
+	data[i++] = oct_dev->link_stats.fromwire.fw_total_bcast;
 	/*per_core_stats[core_id].link_stats[ifidx].fromwire.jabber_err */
 	data[i++] = CVM_CAST64(oct_dev->link_stats.fromwire.jabber_err);
 	/*per_core_stats[core_id].link_stats[ifidx].fromwire.l2_err */
@@ -984,11 +1660,11 @@ lio_get_ethtool_stats(struct net_device *netdev,
 		data[i++] =
 			CVM_CAST64(oct_dev->instr_queue[j]->stats.instr_posted);
 		/*# of instructions processed */
-		data[i++] = CVM_CAST64(oct_dev->instr_queue[j]->
-				       stats.instr_processed);
+		data[i++] = CVM_CAST64(
+				oct_dev->instr_queue[j]->stats.instr_processed);
 		/*# of instructions could not be processed */
-		data[i++] = CVM_CAST64(oct_dev->instr_queue[j]->
-				       stats.instr_dropped);
+		data[i++] = CVM_CAST64(
+				oct_dev->instr_queue[j]->stats.instr_dropped);
 		/*bytes sent through the queue */
 		data[i++] =
 			CVM_CAST64(oct_dev->instr_queue[j]->stats.bytes_sent);
@@ -1045,34 +1721,44 @@ static void lio_vf_get_ethtool_stats(struct net_device *netdev,
 				     __attribute__((unused)),
 				     u64 *data)
 {
-	struct net_device_stats *netstats = &netdev->stats;
+	struct rtnl_link_stats64 lstats;
 	struct lio *lio = GET_LIO(netdev);
 	struct octeon_device *oct_dev = lio->oct_dev;
 	int i = 0, j, vj;
 
-	netdev->netdev_ops->ndo_get_stats(netdev);
+	if (ifstate_check(lio, LIO_IFSTATE_RESETTING))
+		return;
+
+	netdev->netdev_ops->ndo_get_stats64(netdev, &lstats);
 	/* sum of oct->droq[oq_no]->stats->rx_pkts_received */
-	data[i++] = CVM_CAST64(netstats->rx_packets);
+	data[i++] = lstats.rx_packets;
 	/* sum of oct->instr_queue[iq_no]->stats.tx_done */
-	data[i++] = CVM_CAST64(netstats->tx_packets);
+	data[i++] = lstats.tx_packets;
 	/* sum of oct->droq[oq_no]->stats->rx_bytes_received */
-	data[i++] = CVM_CAST64(netstats->rx_bytes);
+	data[i++] = lstats.rx_bytes;
 	/* sum of oct->instr_queue[iq_no]->stats.tx_tot_bytes */
-	data[i++] = CVM_CAST64(netstats->tx_bytes);
-	data[i++] = CVM_CAST64(netstats->rx_errors);
-	data[i++] = CVM_CAST64(netstats->tx_errors);
+	data[i++] = lstats.tx_bytes;
+	data[i++] = lstats.rx_errors;
+	data[i++] = lstats.tx_errors;
 	 /* sum of oct->droq[oq_no]->stats->rx_dropped +
 	  * oct->droq[oq_no]->stats->dropped_nodispatch +
 	  * oct->droq[oq_no]->stats->dropped_toomany +
 	  * oct->droq[oq_no]->stats->dropped_nomem
 	  */
-	data[i++] = CVM_CAST64(netstats->rx_dropped);
+	data[i++] = lstats.rx_dropped;
 	/* sum of oct->instr_queue[iq_no]->stats.tx_dropped */
-	data[i++] = CVM_CAST64(netstats->tx_dropped);
+	data[i++] = lstats.tx_dropped +
+		oct_dev->link_stats.fromhost.fw_err_drop;
+
+	data[i++] = oct_dev->link_stats.fromwire.fw_total_mcast;
+	data[i++] = oct_dev->link_stats.fromhost.fw_total_mcast_sent;
+	data[i++] = oct_dev->link_stats.fromwire.fw_total_bcast;
+	data[i++] = oct_dev->link_stats.fromhost.fw_total_bcast_sent;
+
 	/* lio->link_changes */
 	data[i++] = CVM_CAST64(lio->link_changes);
 
-	for (vj = 0; vj < lio->linfo.num_txpciq; vj++) {
+	for (vj = 0; vj < oct_dev->num_iqs; vj++) {
 		j = lio->linfo.txpciq[vj].s.q_no;
 
 		/* packets to network port */
@@ -1114,7 +1800,7 @@ static void lio_vf_get_ethtool_stats(struct net_device *netdev,
 	}
 
 	/* RX */
-	for (vj = 0; vj < lio->linfo.num_rxpciq; vj++) {
+	for (vj = 0; vj < oct_dev->num_oqs; vj++) {
 		j = lio->linfo.rxpciq[vj].s.q_no;
 
 		/* packets send to TCP/IP network stack */
@@ -1311,34 +1997,11 @@ static int lio_vf_get_sset_count(struct net_device *netdev, int sset)
 	}
 }
 
-/* Callback function for intrmod */
-static void octnet_intrmod_callback(struct octeon_device *oct_dev,
-				    u32 status,
-				    void *ptr)
-{
-	struct octeon_soft_command *sc = (struct octeon_soft_command *)ptr;
-	struct oct_intrmod_context *ctx;
-
-	ctx  = (struct oct_intrmod_context *)sc->ctxptr;
-
-	ctx->status = status;
-
-	WRITE_ONCE(ctx->cond, 1);
-
-	/* This barrier is required to be sure that the response has been
-	 * written fully before waking up the handler
-	 */
-	wmb();
-
-	wake_up_interruptible(&ctx->wc);
-}
-
 /*  get interrupt moderation parameters */
 static int octnet_get_intrmod_cfg(struct lio *lio,
 				  struct oct_intrmod_cfg *intr_cfg)
 {
 	struct octeon_soft_command *sc;
-	struct oct_intrmod_context *ctx;
 	struct oct_intrmod_resp *resp;
 	int retval;
 	struct octeon_device *oct_dev = lio->oct_dev;
@@ -1347,8 +2010,7 @@ static int octnet_get_intrmod_cfg(struct lio *lio,
 	sc = (struct octeon_soft_command *)
 		octeon_alloc_soft_command(oct_dev,
 					  0,
-					  sizeof(struct oct_intrmod_resp),
-					  sizeof(struct oct_intrmod_context));
+					  sizeof(struct oct_intrmod_resp), 0);
 
 	if (!sc)
 		return -ENOMEM;
@@ -1356,20 +2018,13 @@ static int octnet_get_intrmod_cfg(struct lio *lio,
 	resp = (struct oct_intrmod_resp *)sc->virtrptr;
 	memset(resp, 0, sizeof(struct oct_intrmod_resp));
 
-	ctx = (struct oct_intrmod_context *)sc->ctxptr;
-	memset(ctx, 0, sizeof(struct oct_intrmod_context));
-	WRITE_ONCE(ctx->cond, 0);
-	ctx->octeon_id = lio_get_device_id(oct_dev);
-	init_waitqueue_head(&ctx->wc);
-
 	sc->iq_no = lio->linfo.txpciq[0].s.q_no;
 
 	octeon_prepare_soft_command(oct_dev, sc, OPCODE_NIC,
 				    OPCODE_NIC_INTRMOD_PARAMS, 0, 0, 0);
 
-	sc->callback = octnet_intrmod_callback;
-	sc->callback_arg = sc;
-	sc->wait_time = 1000;
+	init_completion(&sc->complete);
+	sc->sc_status = OCTEON_REQUEST_PENDING;
 
 	retval = octeon_send_soft_command(oct_dev, sc);
 	if (retval == IQ_SEND_FAILED) {
@@ -1380,32 +2035,23 @@ static int octnet_get_intrmod_cfg(struct lio *lio,
 	/* Sleep on a wait queue till the cond flag indicates that the
 	 * response arrived or timed-out.
 	 */
-	if (sleep_cond(&ctx->wc, &ctx->cond) == -EINTR) {
-		dev_err(&oct_dev->pci_dev->dev, "Wait interrupted\n");
-		goto intrmod_info_wait_intr;
-	}
+	retval = wait_for_sc_completion_timeout(oct_dev, sc, 0);
+	if (retval)
+		return -ENODEV;
 
-	retval = ctx->status || resp->status;
-	if (retval) {
+	if (resp->status) {
 		dev_err(&oct_dev->pci_dev->dev,
 			"Get interrupt moderation parameters failed\n");
-		goto intrmod_info_wait_fail;
+		WRITE_ONCE(sc->caller_is_done, true);
+		return -ENODEV;
 	}
 
 	octeon_swap_8B_data((u64 *)&resp->intrmod,
 			    (sizeof(struct oct_intrmod_cfg)) / 8);
 	memcpy(intr_cfg, &resp->intrmod, sizeof(struct oct_intrmod_cfg));
-	octeon_free_soft_command(oct_dev, sc);
+	WRITE_ONCE(sc->caller_is_done, true);
 
 	return 0;
-
-intrmod_info_wait_fail:
-
-	octeon_free_soft_command(oct_dev, sc);
-
-intrmod_info_wait_intr:
-
-	return -ENODEV;
 }
 
 /*  Configure interrupt moderation parameters */
@@ -1413,7 +2059,6 @@ static int octnet_set_intrmod_cfg(struct lio *lio,
 				  struct oct_intrmod_cfg *intr_cfg)
 {
 	struct octeon_soft_command *sc;
-	struct oct_intrmod_context *ctx;
 	struct oct_intrmod_cfg *cfg;
 	int retval;
 	struct octeon_device *oct_dev = lio->oct_dev;
@@ -1422,18 +2067,11 @@ static int octnet_set_intrmod_cfg(struct lio *lio,
 	sc = (struct octeon_soft_command *)
 		octeon_alloc_soft_command(oct_dev,
 					  sizeof(struct oct_intrmod_cfg),
-					  0,
-					  sizeof(struct oct_intrmod_context));
+					  16, 0);
 
 	if (!sc)
 		return -ENOMEM;
 
-	ctx = (struct oct_intrmod_context *)sc->ctxptr;
-
-	WRITE_ONCE(ctx->cond, 0);
-	ctx->octeon_id = lio_get_device_id(oct_dev);
-	init_waitqueue_head(&ctx->wc);
-
 	cfg = (struct oct_intrmod_cfg *)sc->virtdptr;
 
 	memcpy(cfg, intr_cfg, sizeof(struct oct_intrmod_cfg));
@@ -1444,9 +2082,8 @@ static int octnet_set_intrmod_cfg(struct lio *lio,
 	octeon_prepare_soft_command(oct_dev, sc, OPCODE_NIC,
 				    OPCODE_NIC_INTRMOD_CFG, 0, 0, 0);
 
-	sc->callback = octnet_intrmod_callback;
-	sc->callback_arg = sc;
-	sc->wait_time = 1000;
+	init_completion(&sc->complete);
+	sc->sc_status = OCTEON_REQUEST_PENDING;
 
 	retval = octeon_send_soft_command(oct_dev, sc);
 	if (retval == IQ_SEND_FAILED) {
@@ -1457,184 +2094,30 @@ static int octnet_set_intrmod_cfg(struct lio *lio,
 	/* Sleep on a wait queue till the cond flag indicates that the
 	 * response arrived or timed-out.
 	 */
-	if (sleep_cond(&ctx->wc, &ctx->cond) != -EINTR) {
-		retval = ctx->status;
-		if (retval)
-			dev_err(&oct_dev->pci_dev->dev,
-				"intrmod config failed. Status: %llx\n",
-				CVM_CAST64(retval));
-		else
-			dev_info(&oct_dev->pci_dev->dev,
-				 "Rx-Adaptive Interrupt moderation %s\n",
-				 (intr_cfg->rx_enable) ?
-				 "enabled" : "disabled");
-
-		octeon_free_soft_command(oct_dev, sc);
-
-		return ((retval) ? -ENODEV : 0);
-	}
-
-	dev_err(&oct_dev->pci_dev->dev, "iq/oq config failed\n");
-
-	return -EINTR;
-}
-
-static void
-octnet_nic_stats_callback(struct octeon_device *oct_dev,
-			  u32 status, void *ptr)
-{
-	struct octeon_soft_command *sc = (struct octeon_soft_command *)ptr;
-	struct oct_nic_stats_resp *resp =
-	    (struct oct_nic_stats_resp *)sc->virtrptr;
-	struct oct_nic_stats_ctrl *ctrl =
-	    (struct oct_nic_stats_ctrl *)sc->ctxptr;
-	struct nic_rx_stats *rsp_rstats = &resp->stats.fromwire;
-	struct nic_tx_stats *rsp_tstats = &resp->stats.fromhost;
-
-	struct nic_rx_stats *rstats = &oct_dev->link_stats.fromwire;
-	struct nic_tx_stats *tstats = &oct_dev->link_stats.fromhost;
-
-	if ((status != OCTEON_REQUEST_TIMEOUT) && !resp->status) {
-		octeon_swap_8B_data((u64 *)&resp->stats,
-				    (sizeof(struct oct_link_stats)) >> 3);
-
-		/* RX link-level stats */
-		rstats->total_rcvd = rsp_rstats->total_rcvd;
-		rstats->bytes_rcvd = rsp_rstats->bytes_rcvd;
-		rstats->total_bcst = rsp_rstats->total_bcst;
-		rstats->total_mcst = rsp_rstats->total_mcst;
-		rstats->runts      = rsp_rstats->runts;
-		rstats->ctl_rcvd   = rsp_rstats->ctl_rcvd;
-		/* Accounts for over/under-run of buffers */
-		rstats->fifo_err  = rsp_rstats->fifo_err;
-		rstats->dmac_drop = rsp_rstats->dmac_drop;
-		rstats->fcs_err   = rsp_rstats->fcs_err;
-		rstats->jabber_err = rsp_rstats->jabber_err;
-		rstats->l2_err    = rsp_rstats->l2_err;
-		rstats->frame_err = rsp_rstats->frame_err;
-
-		/* RX firmware stats */
-		rstats->fw_total_rcvd = rsp_rstats->fw_total_rcvd;
-		rstats->fw_total_fwd = rsp_rstats->fw_total_fwd;
-		rstats->fw_err_pko = rsp_rstats->fw_err_pko;
-		rstats->fw_err_link = rsp_rstats->fw_err_link;
-		rstats->fw_err_drop = rsp_rstats->fw_err_drop;
-		rstats->fw_rx_vxlan = rsp_rstats->fw_rx_vxlan;
-		rstats->fw_rx_vxlan_err = rsp_rstats->fw_rx_vxlan_err;
-
-		/* Number of packets that are LROed      */
-		rstats->fw_lro_pkts = rsp_rstats->fw_lro_pkts;
-		/* Number of octets that are LROed       */
-		rstats->fw_lro_octs = rsp_rstats->fw_lro_octs;
-		/* Number of LRO packets formed          */
-		rstats->fw_total_lro = rsp_rstats->fw_total_lro;
-		/* Number of times lRO of packet aborted */
-		rstats->fw_lro_aborts = rsp_rstats->fw_lro_aborts;
-		rstats->fw_lro_aborts_port = rsp_rstats->fw_lro_aborts_port;
-		rstats->fw_lro_aborts_seq = rsp_rstats->fw_lro_aborts_seq;
-		rstats->fw_lro_aborts_tsval = rsp_rstats->fw_lro_aborts_tsval;
-		rstats->fw_lro_aborts_timer = rsp_rstats->fw_lro_aborts_timer;
-		/* intrmod: packet forward rate */
-		rstats->fwd_rate = rsp_rstats->fwd_rate;
-
-		/* TX link-level stats */
-		tstats->total_pkts_sent = rsp_tstats->total_pkts_sent;
-		tstats->total_bytes_sent = rsp_tstats->total_bytes_sent;
-		tstats->mcast_pkts_sent = rsp_tstats->mcast_pkts_sent;
-		tstats->bcast_pkts_sent = rsp_tstats->bcast_pkts_sent;
-		tstats->ctl_sent = rsp_tstats->ctl_sent;
-		/* Packets sent after one collision*/
-		tstats->one_collision_sent = rsp_tstats->one_collision_sent;
-		/* Packets sent after multiple collision*/
-		tstats->multi_collision_sent = rsp_tstats->multi_collision_sent;
-		/* Packets not sent due to max collisions */
-		tstats->max_collision_fail = rsp_tstats->max_collision_fail;
-		/* Packets not sent due to max deferrals */
-		tstats->max_deferral_fail = rsp_tstats->max_deferral_fail;
-		/* Accounts for over/under-run of buffers */
-		tstats->fifo_err = rsp_tstats->fifo_err;
-		tstats->runts = rsp_tstats->runts;
-		/* Total number of collisions detected */
-		tstats->total_collisions = rsp_tstats->total_collisions;
-
-		/* firmware stats */
-		tstats->fw_total_sent = rsp_tstats->fw_total_sent;
-		tstats->fw_total_fwd = rsp_tstats->fw_total_fwd;
-		tstats->fw_err_pko = rsp_tstats->fw_err_pko;
-		tstats->fw_err_link = rsp_tstats->fw_err_link;
-		tstats->fw_err_drop = rsp_tstats->fw_err_drop;
-		tstats->fw_tso = rsp_tstats->fw_tso;
-		tstats->fw_tso_fwd = rsp_tstats->fw_tso_fwd;
-		tstats->fw_err_tso = rsp_tstats->fw_err_tso;
-		tstats->fw_tx_vxlan = rsp_tstats->fw_tx_vxlan;
-
-		resp->status = 1;
-	} else {
-		resp->status = -1;
-	}
-	complete(&ctrl->complete);
-}
-
-/*  Configure interrupt moderation parameters */
-static int octnet_get_link_stats(struct net_device *netdev)
-{
-	struct lio *lio = GET_LIO(netdev);
-	struct octeon_device *oct_dev = lio->oct_dev;
-
-	struct octeon_soft_command *sc;
-	struct oct_nic_stats_ctrl *ctrl;
-	struct oct_nic_stats_resp *resp;
-
-	int retval;
-
-	/* Alloc soft command */
-	sc = (struct octeon_soft_command *)
-		octeon_alloc_soft_command(oct_dev,
-					  0,
-					  sizeof(struct oct_nic_stats_resp),
-					  sizeof(struct octnic_ctrl_pkt));
-
-	if (!sc)
-		return -ENOMEM;
-
-	resp = (struct oct_nic_stats_resp *)sc->virtrptr;
-	memset(resp, 0, sizeof(struct oct_nic_stats_resp));
-
-	ctrl = (struct oct_nic_stats_ctrl *)sc->ctxptr;
-	memset(ctrl, 0, sizeof(struct oct_nic_stats_ctrl));
-	ctrl->netdev = netdev;
-	init_completion(&ctrl->complete);
-
-	sc->iq_no = lio->linfo.txpciq[0].s.q_no;
-
-	octeon_prepare_soft_command(oct_dev, sc, OPCODE_NIC,
-				    OPCODE_NIC_PORT_STATS, 0, 0, 0);
-
-	sc->callback = octnet_nic_stats_callback;
-	sc->callback_arg = sc;
-	sc->wait_time = 500;	/*in milli seconds*/
-
-	retval = octeon_send_soft_command(oct_dev, sc);
-	if (retval == IQ_SEND_FAILED) {
-		octeon_free_soft_command(oct_dev, sc);
-		return -EINVAL;
-	}
-
-	wait_for_completion_timeout(&ctrl->complete, msecs_to_jiffies(1000));
-
-	if (resp->status != 1) {
-		octeon_free_soft_command(oct_dev, sc);
-
-		return -EINVAL;
-	}
-
-	octeon_free_soft_command(oct_dev, sc);
-
-	return 0;
+	retval = wait_for_sc_completion_timeout(oct_dev, sc, 0);
+	if (retval)
+		return retval;
+
+	retval = sc->sc_status;
+	if (retval == 0) {
+		dev_info(&oct_dev->pci_dev->dev,
+			 "Rx-Adaptive Interrupt moderation %s\n",
+			 (intr_cfg->rx_enable) ?
+			 "enabled" : "disabled");
+		WRITE_ONCE(sc->caller_is_done, true);
+		return 0;
+	}
+
+	dev_err(&oct_dev->pci_dev->dev,
+		"intrmod config failed. Status: %x\n", retval);
+	WRITE_ONCE(sc->caller_is_done, true);
+	return -ENODEV;
 }
 
 static int lio_get_intr_coalesce(struct net_device *netdev,
-				 struct ethtool_coalesce *intr_coal)
+				 struct ethtool_coalesce *intr_coal,
+				 struct kernel_ethtool_coalesce *kernel_coal,
+				 struct netlink_ext_ack *extack)
 {
 	struct lio *lio = GET_LIO(netdev);
 	struct octeon_device *oct = lio->oct_dev;
@@ -1801,7 +2284,7 @@ oct_cfg_rx_intrcnt(struct lio *lio,
 			    (octeon_read_csr64(
 				 oct, CN23XX_VF_SLI_OQ_PKT_INT_LEVELS(q_no)) &
 			     (0x3fffff00000000UL)) |
-				rx_max_coalesced_frames);
+				(rx_max_coalesced_frames - 1));
 			/*consider writing to resend bit here*/
 		}
 		intrmod->rx_frames = rx_max_coalesced_frames;
@@ -1938,7 +2421,9 @@ oct_cfg_tx_intrcnt(struct lio *lio,
 }
 
 static int lio_set_intr_coalesce(struct net_device *netdev,
-				 struct ethtool_coalesce *intr_coal)
+				 struct ethtool_coalesce *intr_coal,
+				 struct kernel_ethtool_coalesce *kernel_coal,
+				 struct netlink_ext_ack *extack)
 {
 	struct lio *lio = GET_LIO(netdev);
 	int ret;
@@ -2011,37 +2496,31 @@ ret_intrmod:
 	return ret;
 }
 
+#ifdef PTP_HARDWARE_TIMESTAMPING
 static int lio_get_ts_info(struct net_device *netdev,
-			   struct ethtool_ts_info *info)
+			   struct kernel_ethtool_ts_info *info)
 {
 	struct lio *lio = GET_LIO(netdev);
 
 	info->so_timestamping =
-#ifdef PTP_HARDWARE_TIMESTAMPING
 		SOF_TIMESTAMPING_TX_HARDWARE |
 		SOF_TIMESTAMPING_RX_HARDWARE |
 		SOF_TIMESTAMPING_RAW_HARDWARE |
-		SOF_TIMESTAMPING_TX_SOFTWARE |
-#endif
-		SOF_TIMESTAMPING_RX_SOFTWARE |
-		SOF_TIMESTAMPING_SOFTWARE;
+		SOF_TIMESTAMPING_TX_SOFTWARE;
 
 	if (lio->ptp_clock)
 		info->phc_index = ptp_clock_index(lio->ptp_clock);
-	else
-		info->phc_index = -1;
 
-#ifdef PTP_HARDWARE_TIMESTAMPING
 	info->tx_types = (1 << HWTSTAMP_TX_OFF) | (1 << HWTSTAMP_TX_ON);
 
 	info->rx_filters = (1 << HWTSTAMP_FILTER_NONE) |
 			   (1 << HWTSTAMP_FILTER_PTP_V1_L4_EVENT) |
 			   (1 << HWTSTAMP_FILTER_PTP_V2_L2_EVENT) |
 			   (1 << HWTSTAMP_FILTER_PTP_V2_L4_EVENT);
-#endif
 
 	return 0;
 }
+#endif
 
 /* Return register dump len. */
 static int lio_get_regs_len(struct net_device *dev)
@@ -2575,12 +3054,76 @@ static int lio_set_priv_flags(struct net_device *netdev, u32 flags)
 	return 0;
 }
 
+static int lio_get_fecparam(struct net_device *netdev,
+			    struct ethtool_fecparam *fec)
+{
+	struct lio *lio = GET_LIO(netdev);
+	struct octeon_device *oct = lio->oct_dev;
+
+	fec->active_fec = ETHTOOL_FEC_NONE;
+	fec->fec = ETHTOOL_FEC_NONE;
+
+	if (oct->subsystem_id == OCTEON_CN2350_25GB_SUBSYS_ID ||
+	    oct->subsystem_id == OCTEON_CN2360_25GB_SUBSYS_ID) {
+		if (oct->no_speed_setting == 1)
+			return 0;
+
+		liquidio_get_fec(lio);
+		fec->fec = (ETHTOOL_FEC_RS | ETHTOOL_FEC_OFF);
+		if (oct->props[lio->ifidx].fec == 1)
+			fec->active_fec = ETHTOOL_FEC_RS;
+		else
+			fec->active_fec = ETHTOOL_FEC_OFF;
+	}
+
+	return 0;
+}
+
+static int lio_set_fecparam(struct net_device *netdev,
+			    struct ethtool_fecparam *fec)
+{
+	struct lio *lio = GET_LIO(netdev);
+	struct octeon_device *oct = lio->oct_dev;
+
+	if (oct->subsystem_id == OCTEON_CN2350_25GB_SUBSYS_ID ||
+	    oct->subsystem_id == OCTEON_CN2360_25GB_SUBSYS_ID) {
+		if (oct->no_speed_setting == 1)
+			return -EOPNOTSUPP;
+
+		if (fec->fec & ETHTOOL_FEC_OFF)
+			liquidio_set_fec(lio, 0);
+		else if (fec->fec & ETHTOOL_FEC_RS)
+			liquidio_set_fec(lio, 1);
+		else
+			return -EOPNOTSUPP;
+	} else {
+		return -EOPNOTSUPP;
+	}
+
+	return 0;
+}
+
+#define LIO_ETHTOOL_COALESCE	(ETHTOOL_COALESCE_RX_USECS |		\
+				 ETHTOOL_COALESCE_MAX_FRAMES |		\
+				 ETHTOOL_COALESCE_USE_ADAPTIVE |	\
+				 ETHTOOL_COALESCE_RX_MAX_FRAMES_LOW |	\
+				 ETHTOOL_COALESCE_TX_MAX_FRAMES_LOW |	\
+				 ETHTOOL_COALESCE_RX_MAX_FRAMES_HIGH |	\
+				 ETHTOOL_COALESCE_TX_MAX_FRAMES_HIGH |	\
+				 ETHTOOL_COALESCE_PKT_RATE_RX_USECS)
+
 static const struct ethtool_ops lio_ethtool_ops = {
+	.supported_coalesce_params = LIO_ETHTOOL_COALESCE,
 	.get_link_ksettings	= lio_get_link_ksettings,
+	.set_link_ksettings	= lio_set_link_ksettings,
+	.get_fecparam		= lio_get_fecparam,
+	.set_fecparam		= lio_set_fecparam,
 	.get_link		= ethtool_op_get_link,
 	.get_drvinfo		= lio_get_drvinfo,
 	.get_ringparam		= lio_ethtool_get_ringparam,
+	.set_ringparam		= lio_ethtool_set_ringparam,
 	.get_channels		= lio_ethtool_get_channels,
+	.set_channels		= lio_ethtool_set_channels,
 	.set_phys_id		= lio_set_phys_id,
 	.get_eeprom_len		= lio_get_eeprom_len,
 	.get_eeprom		= lio_get_eeprom,
@@ -2597,27 +3140,34 @@ static const struct ethtool_ops lio_ethtool_ops = {
 	.set_coalesce		= lio_set_intr_coalesce,
 	.get_priv_flags		= lio_get_priv_flags,
 	.set_priv_flags		= lio_set_priv_flags,
+#ifdef PTP_HARDWARE_TIMESTAMPING
 	.get_ts_info		= lio_get_ts_info,
+#endif
 };
 
 static const struct ethtool_ops lio_vf_ethtool_ops = {
+	.supported_coalesce_params = LIO_ETHTOOL_COALESCE,
 	.get_link_ksettings	= lio_get_link_ksettings,
 	.get_link		= ethtool_op_get_link,
 	.get_drvinfo		= lio_get_vf_drvinfo,
 	.get_ringparam		= lio_ethtool_get_ringparam,
+	.set_ringparam          = lio_ethtool_set_ringparam,
 	.get_channels		= lio_ethtool_get_channels,
+	.set_channels		= lio_ethtool_set_channels,
 	.get_strings		= lio_vf_get_strings,
 	.get_ethtool_stats	= lio_vf_get_ethtool_stats,
 	.get_regs_len		= lio_get_regs_len,
 	.get_regs		= lio_get_regs,
 	.get_msglevel		= lio_get_msglevel,
-	.set_msglevel		= lio_set_msglevel,
+	.set_msglevel		= lio_vf_set_msglevel,
 	.get_sset_count		= lio_vf_get_sset_count,
 	.get_coalesce		= lio_get_intr_coalesce,
 	.set_coalesce		= lio_set_intr_coalesce,
 	.get_priv_flags		= lio_get_priv_flags,
 	.set_priv_flags		= lio_set_priv_flags,
+#ifdef PTP_HARDWARE_TIMESTAMPING
 	.get_ts_info		= lio_get_ts_info,
+#endif
 };
 
 void liquidio_set_ethtool_ops(struct net_device *netdev)
@@ -2630,3 +3180,4 @@ void liquidio_set_ethtool_ops(struct net_device *netdev)
 	else
 		netdev->ethtool_ops = &lio_ethtool_ops;
 }
+EXPORT_SYMBOL_GPL(liquidio_set_ethtool_ops);
diff --git a/drivers/net/ethernet/cavium/liquidio/lio_main.c b/drivers/net/ethernet/cavium/liquidio/lio_main.c
index ba012427edd6..8e2fcec26ea1 100644
--- a/drivers/net/ethernet/cavium/liquidio/lio_main.c
+++ b/drivers/net/ethernet/cavium/liquidio/lio_main.c
@@ -34,15 +34,19 @@
 #include "cn68xx_device.h"
 #include "cn23xx_pf_device.h"
 #include "liquidio_image.h"
+#include "lio_vf_rep.h"
 
 MODULE_AUTHOR("Cavium Networks, <support@cavium.com>");
 MODULE_DESCRIPTION("Cavium LiquidIO Intelligent Server Adapter Driver");
 MODULE_LICENSE("GPL");
-MODULE_VERSION(LIQUIDIO_VERSION);
-MODULE_FIRMWARE(LIO_FW_DIR LIO_FW_BASE_NAME LIO_210SV_NAME LIO_FW_NAME_SUFFIX);
-MODULE_FIRMWARE(LIO_FW_DIR LIO_FW_BASE_NAME LIO_210NV_NAME LIO_FW_NAME_SUFFIX);
-MODULE_FIRMWARE(LIO_FW_DIR LIO_FW_BASE_NAME LIO_410NV_NAME LIO_FW_NAME_SUFFIX);
-MODULE_FIRMWARE(LIO_FW_DIR LIO_FW_BASE_NAME LIO_23XX_NAME LIO_FW_NAME_SUFFIX);
+MODULE_FIRMWARE(LIO_FW_DIR LIO_FW_BASE_NAME LIO_210SV_NAME
+		"_" LIO_FW_NAME_TYPE_NIC LIO_FW_NAME_SUFFIX);
+MODULE_FIRMWARE(LIO_FW_DIR LIO_FW_BASE_NAME LIO_210NV_NAME
+		"_" LIO_FW_NAME_TYPE_NIC LIO_FW_NAME_SUFFIX);
+MODULE_FIRMWARE(LIO_FW_DIR LIO_FW_BASE_NAME LIO_410NV_NAME
+		"_" LIO_FW_NAME_TYPE_NIC LIO_FW_NAME_SUFFIX);
+MODULE_FIRMWARE(LIO_FW_DIR LIO_FW_BASE_NAME LIO_23XX_NAME
+		"_" LIO_FW_NAME_TYPE_NIC LIO_FW_NAME_SUFFIX);
 
 static int ddr_timeout = 10000;
 module_param(ddr_timeout, int, 0644);
@@ -55,45 +59,38 @@ static int debug = -1;
 module_param(debug, int, 0644);
 MODULE_PARM_DESC(debug, "NETIF_MSG debug bits");
 
-static char fw_type[LIO_MAX_FW_TYPE_LEN];
-module_param_string(fw_type, fw_type, sizeof(fw_type), 0000);
-MODULE_PARM_DESC(fw_type, "Type of firmware to be loaded. Default \"nic\"");
+static char fw_type[LIO_MAX_FW_TYPE_LEN] = LIO_FW_NAME_TYPE_AUTO;
+module_param_string(fw_type, fw_type, sizeof(fw_type), 0444);
+MODULE_PARM_DESC(fw_type, "Type of firmware to be loaded (default is \"auto\"), which uses firmware in flash, if present, else loads \"nic\".");
 
-static int ptp_enable = 1;
+static u32 console_bitmask;
+module_param(console_bitmask, int, 0644);
+MODULE_PARM_DESC(console_bitmask,
+		 "Bitmask indicating which consoles have debug output redirected to syslog.");
+
+/**
+ * octeon_console_debug_enabled - determines if a given console has debug enabled.
+ * @console: console to check
+ * Return:  1 = enabled. 0 otherwise
+ */
+static int octeon_console_debug_enabled(u32 console)
+{
+	return (console_bitmask >> (console)) & 0x1;
+}
 
 /* Polling interval for determining when NIC application is alive */
 #define LIQUIDIO_STARTER_POLL_INTERVAL_MS 100
 
 /* runtime link query interval */
 #define LIQUIDIO_LINK_QUERY_INTERVAL_MS         1000
+/* update localtime to octeon firmware every 60 seconds.
+ * make firmware to use same time reference, so that it will be easy to
+ * correlate firmware logged events/errors with host events, for debugging.
+ */
+#define LIO_SYNC_OCTEON_TIME_INTERVAL_MS 60000
 
-struct liquidio_if_cfg_context {
-	int octeon_id;
-
-	wait_queue_head_t wc;
-
-	int cond;
-};
-
-struct liquidio_if_cfg_resp {
-	u64 rh;
-	struct liquidio_if_cfg_info cfg_info;
-	u64 status;
-};
-
-struct liquidio_rx_ctl_context {
-	int octeon_id;
-
-	wait_queue_head_t wc;
-
-	int cond;
-};
-
-struct oct_link_status_resp {
-	u64 rh;
-	struct oct_link_info link_info;
-	u64 status;
-};
+/* time to wait for possible in-flight requests in milliseconds */
+#define WAIT_INFLIGHT_REQUEST	msecs_to_jiffies(1000)
 
 struct oct_timestamp_resp {
 	u64 rh;
@@ -118,38 +115,15 @@ union tx_info {
 	} s;
 };
 
-/** Octeon device properties to be used by the NIC module.
+/* Octeon device properties to be used by the NIC module.
  * Each octeon device in the system will be represented
  * by this structure in the NIC module.
  */
 
-#define OCTNIC_MAX_SG  (MAX_SKB_FRAGS)
-
 #define OCTNIC_GSO_MAX_HEADER_SIZE 128
 #define OCTNIC_GSO_MAX_SIZE                                                    \
 	(CN23XX_DEFAULT_INPUT_JABBER - OCTNIC_GSO_MAX_HEADER_SIZE)
 
-/** Structure of a node in list of gather components maintained by
- * NIC driver for each network device.
- */
-struct octnic_gather {
-	/** List manipulation. Next and prev pointers. */
-	struct list_head list;
-
-	/** Size of the gather component at sg in bytes. */
-	int sg_size;
-
-	/** Number of bytes that sg was adjusted to make it 8B-aligned. */
-	int adjust;
-
-	/** Gather component that can accommodate max sized fragment list
-	 *  received from the IP layer.
-	 */
-	struct octeon_sg_entry *sg;
-
-	dma_addr_t sg_dma_ptr;
-};
-
 struct handshake {
 	struct completion init;
 	struct completion started;
@@ -158,16 +132,13 @@ struct handshake {
 	int started_ok;
 };
 
-struct octeon_device_priv {
-	/** Tasklet structures for this device. */
-	struct tasklet_struct droq_tasklet;
-	unsigned long napi_mask;
-};
-
 #ifdef CONFIG_PCI_IOV
 static int liquidio_enable_sriov(struct pci_dev *dev, int num_vfs);
 #endif
 
+static int octeon_dbg_console_print(struct octeon_device *oct, u32 console_num,
+				    char *prefix, char *suffix);
+
 static int octeon_device_init(struct octeon_device *);
 static int liquidio_stop(struct net_device *netdev);
 static void liquidio_remove(struct pci_dev *pdev);
@@ -179,13 +150,13 @@ static int liquidio_set_vf_link_state(struct net_device *netdev, int vfidx,
 static struct handshake handshake[MAX_OCTEON_DEVICES];
 static struct completion first_stage;
 
-static void octeon_droq_bh(unsigned long pdev)
+static void octeon_droq_bh(struct tasklet_struct *t)
 {
 	int q_no;
 	int reschedule = 0;
-	struct octeon_device *oct = (struct octeon_device *)pdev;
-	struct octeon_device_priv *oct_priv =
-		(struct octeon_device_priv *)oct->priv;
+	struct octeon_device_priv *oct_priv = from_tasklet(oct_priv, t,
+							  droq_tasklet);
+	struct octeon_device *oct = oct_priv->dev;
 
 	for (q_no = 0; q_no < MAX_OCTEON_OUTPUT_QUEUES(oct); q_no++) {
 		if (!(oct->io_qmask.oq & BIT_ULL(q_no)))
@@ -214,8 +185,7 @@ static void octeon_droq_bh(unsigned long pdev)
 
 static int lio_wait_for_oq_pkts(struct octeon_device *oct)
 {
-	struct octeon_device_priv *oct_priv =
-		(struct octeon_device_priv *)oct->priv;
+	struct octeon_device_priv *oct_priv = oct->priv;
 	int retry = 100, pkt_cnt = 0, pending_pkts = 0;
 	int i;
 
@@ -240,8 +210,8 @@ static int lio_wait_for_oq_pkts(struct octeon_device *oct)
 }
 
 /**
- * \brief Forces all IO queues off on a given device
- * @param oct Pointer to Octeon device
+ * force_io_queues_off - Forces all IO queues off on a given device
+ * @oct: Pointer to Octeon device
  */
 static void force_io_queues_off(struct octeon_device *oct)
 {
@@ -256,34 +226,8 @@ static void force_io_queues_off(struct octeon_device *oct)
 }
 
 /**
- * \brief wait for all pending requests to complete
- * @param oct Pointer to Octeon device
- *
- * Called during shutdown sequence
- */
-static int wait_for_pending_requests(struct octeon_device *oct)
-{
-	int i, pcount = 0;
-
-	for (i = 0; i < 100; i++) {
-		pcount =
-			atomic_read(&oct->response_list
-				[OCTEON_ORDERED_SC_LIST].pending_req_count);
-		if (pcount)
-			schedule_timeout_uninterruptible(HZ / 10);
-		else
-			break;
-	}
-
-	if (pcount)
-		return 1;
-
-	return 0;
-}
-
-/**
- * \brief Cause device to go quiet so it can be safely removed/reset/etc
- * @param oct Pointer to Octeon device
+ * pcierror_quiesce_device - Cause device to go quiet so it can be safely removed/reset/etc
+ * @oct: Pointer to Octeon device
  */
 static inline void pcierror_quiesce_device(struct octeon_device *oct)
 {
@@ -296,7 +240,7 @@ static inline void pcierror_quiesce_device(struct octeon_device *oct)
 	force_io_queues_off(oct);
 
 	/* To allow for in-flight requests */
-	schedule_timeout_uninterruptible(100);
+	schedule_timeout_uninterruptible(WAIT_INFLIGHT_REQUEST);
 
 	if (wait_for_pending_requests(oct))
 		dev_err(&oct->pci_dev->dev, "There were pending requests\n");
@@ -327,8 +271,8 @@ static inline void pcierror_quiesce_device(struct octeon_device *oct)
 }
 
 /**
- * \brief Cleanup PCI AER uncorrectable error status
- * @param dev Pointer to PCI device
+ * cleanup_aer_uncorrect_error_status - Cleanup PCI AER uncorrectable error status
+ * @dev: Pointer to PCI device
  */
 static void cleanup_aer_uncorrect_error_status(struct pci_dev *dev)
 {
@@ -347,8 +291,8 @@ static void cleanup_aer_uncorrect_error_status(struct pci_dev *dev)
 }
 
 /**
- * \brief Stop all PCI IO to a given device
- * @param dev Pointer to Octeon device
+ * stop_pci_io - Stop all PCI IO to a given device
+ * @oct: Pointer to Octeon device
  */
 static void stop_pci_io(struct octeon_device *oct)
 {
@@ -376,9 +320,9 @@ static void stop_pci_io(struct octeon_device *oct)
 }
 
 /**
- * \brief called when PCI error is detected
- * @param pdev Pointer to PCI device
- * @param state The current pci connection state
+ * liquidio_pcie_error_detected - called when PCI error is detected
+ * @pdev: Pointer to PCI device
+ * @state: The current pci connection state
  *
  * This function is called after a PCI bus error affecting
  * this device has been detected.
@@ -406,11 +350,10 @@ static pci_ers_result_t liquidio_pcie_error_detected(struct pci_dev *pdev,
 }
 
 /**
- * \brief mmio handler
- * @param pdev Pointer to PCI device
+ * liquidio_pcie_mmio_enabled - mmio handler
+ * @pdev: Pointer to PCI device
  */
-static pci_ers_result_t liquidio_pcie_mmio_enabled(
-				struct pci_dev *pdev __attribute__((unused)))
+static pci_ers_result_t liquidio_pcie_mmio_enabled(struct pci_dev __maybe_unused *pdev)
 {
 	/* We should never hit this since we never ask for a reset for a Fatal
 	 * Error. We always return DISCONNECT in io_error above.
@@ -420,14 +363,13 @@ static pci_ers_result_t liquidio_pcie_mmio_enabled(
 }
 
 /**
- * \brief called after the pci bus has been reset.
- * @param pdev Pointer to PCI device
+ * liquidio_pcie_slot_reset - called after the pci bus has been reset.
+ * @pdev: Pointer to PCI device
  *
  * Restart the card from scratch, as if from a cold-boot. Implementation
  * resembles the first-half of the octeon_resume routine.
  */
-static pci_ers_result_t liquidio_pcie_slot_reset(
-				struct pci_dev *pdev __attribute__((unused)))
+static pci_ers_result_t liquidio_pcie_slot_reset(struct pci_dev __maybe_unused *pdev)
 {
 	/* We should never hit this since we never ask for a reset for a Fatal
 	 * Error. We always return DISCONNECT in io_error above.
@@ -437,39 +379,20 @@ static pci_ers_result_t liquidio_pcie_slot_reset(
 }
 
 /**
- * \brief called when traffic can start flowing again.
- * @param pdev Pointer to PCI device
+ * liquidio_pcie_resume - called when traffic can start flowing again.
+ * @pdev: Pointer to PCI device
  *
  * This callback is called when the error recovery driver tells us that
  * its OK to resume normal operation. Implementation resembles the
  * second-half of the octeon_resume routine.
  */
-static void liquidio_pcie_resume(struct pci_dev *pdev __attribute__((unused)))
+static void liquidio_pcie_resume(struct pci_dev __maybe_unused *pdev)
 {
 	/* Nothing to be done here. */
 }
 
-#ifdef CONFIG_PM
-/**
- * \brief called when suspending
- * @param pdev Pointer to PCI device
- * @param state state to suspend to
- */
-static int liquidio_suspend(struct pci_dev *pdev __attribute__((unused)),
-			    pm_message_t state __attribute__((unused)))
-{
-	return 0;
-}
-
-/**
- * \brief called when resuming
- * @param pdev Pointer to PCI device
- */
-static int liquidio_resume(struct pci_dev *pdev __attribute__((unused)))
-{
-	return 0;
-}
-#endif
+#define liquidio_suspend NULL
+#define liquidio_resume NULL
 
 /* For PCI-E Advanced Error Recovery (AER) Interface */
 static const struct pci_error_handlers liquidio_err_handler = {
@@ -495,24 +418,22 @@ static const struct pci_device_id liquidio_pci_tbl[] = {
 };
 MODULE_DEVICE_TABLE(pci, liquidio_pci_tbl);
 
+static SIMPLE_DEV_PM_OPS(liquidio_pm_ops, liquidio_suspend, liquidio_resume);
+
 static struct pci_driver liquidio_pci_driver = {
 	.name		= "LiquidIO",
 	.id_table	= liquidio_pci_tbl,
 	.probe		= liquidio_probe,
 	.remove		= liquidio_remove,
 	.err_handler	= &liquidio_err_handler,    /* For AER */
-
-#ifdef CONFIG_PM
-	.suspend	= liquidio_suspend,
-	.resume		= liquidio_resume,
-#endif
+	.driver.pm	= &liquidio_pm_ops,
 #ifdef CONFIG_PCI_IOV
 	.sriov_configure = liquidio_enable_sriov,
 #endif
 };
 
 /**
- * \brief register PCI driver
+ * liquidio_init_pci - register PCI driver
  */
 static int liquidio_init_pci(void)
 {
@@ -520,7 +441,7 @@ static int liquidio_init_pci(void)
 }
 
 /**
- * \brief unregister PCI driver
+ * liquidio_deinit_pci - unregister PCI driver
  */
 static void liquidio_deinit_pci(void)
 {
@@ -528,302 +449,43 @@ static void liquidio_deinit_pci(void)
 }
 
 /**
- * \brief Stop Tx queues
- * @param netdev network device
- */
-static inline void txqs_stop(struct net_device *netdev)
-{
-	if (netif_is_multiqueue(netdev)) {
-		int i;
-
-		for (i = 0; i < netdev->num_tx_queues; i++)
-			netif_stop_subqueue(netdev, i);
-	} else {
-		netif_stop_queue(netdev);
-	}
-}
-
-/**
- * \brief Start Tx queues
- * @param netdev network device
- */
-static inline void txqs_start(struct net_device *netdev)
-{
-	if (netif_is_multiqueue(netdev)) {
-		int i;
-
-		for (i = 0; i < netdev->num_tx_queues; i++)
-			netif_start_subqueue(netdev, i);
-	} else {
-		netif_start_queue(netdev);
-	}
-}
-
-/**
- * \brief Wake Tx queues
- * @param netdev network device
- */
-static inline void txqs_wake(struct net_device *netdev)
-{
-	struct lio *lio = GET_LIO(netdev);
-
-	if (netif_is_multiqueue(netdev)) {
-		int i;
-
-		for (i = 0; i < netdev->num_tx_queues; i++) {
-			int qno = lio->linfo.txpciq[i %
-				(lio->linfo.num_txpciq)].s.q_no;
-
-			if (__netif_subqueue_stopped(netdev, i)) {
-				INCR_INSTRQUEUE_PKT_COUNT(lio->oct_dev, qno,
-							  tx_restart, 1);
-				netif_wake_subqueue(netdev, i);
-			}
-		}
-	} else {
-		INCR_INSTRQUEUE_PKT_COUNT(lio->oct_dev, lio->txq,
-					  tx_restart, 1);
-		netif_wake_queue(netdev);
-	}
-}
-
-/**
- * \brief Stop Tx queue
- * @param netdev network device
- */
-static void stop_txq(struct net_device *netdev)
-{
-	txqs_stop(netdev);
-}
-
-/**
- * \brief Start Tx queue
- * @param netdev network device
- */
-static void start_txq(struct net_device *netdev)
-{
-	struct lio *lio = GET_LIO(netdev);
-
-	if (lio->linfo.link.s.link_up) {
-		txqs_start(netdev);
-		return;
-	}
-}
-
-/**
- * \brief Wake a queue
- * @param netdev network device
- * @param q which queue to wake
- */
-static inline void wake_q(struct net_device *netdev, int q)
-{
-	if (netif_is_multiqueue(netdev))
-		netif_wake_subqueue(netdev, q);
-	else
-		netif_wake_queue(netdev);
-}
-
-/**
- * \brief Stop a queue
- * @param netdev network device
- * @param q which queue to stop
- */
-static inline void stop_q(struct net_device *netdev, int q)
-{
-	if (netif_is_multiqueue(netdev))
-		netif_stop_subqueue(netdev, q);
-	else
-		netif_stop_queue(netdev);
-}
-
-/**
- * \brief Check Tx queue status, and take appropriate action
- * @param lio per-network private data
- * @returns 0 if full, number of queues woken up otherwise
+ * check_txq_status - Check Tx queue status, and take appropriate action
+ * @lio: per-network private data
+ * Return: 0 if full, number of queues woken up otherwise
  */
 static inline int check_txq_status(struct lio *lio)
 {
+	int numqs = lio->netdev->real_num_tx_queues;
 	int ret_val = 0;
+	int q, iq;
 
-	if (netif_is_multiqueue(lio->netdev)) {
-		int numqs = lio->netdev->num_tx_queues;
-		int q, iq = 0;
-
-		/* check each sub-queue state */
-		for (q = 0; q < numqs; q++) {
-			iq = lio->linfo.txpciq[q %
-				(lio->linfo.num_txpciq)].s.q_no;
-			if (octnet_iq_is_full(lio->oct_dev, iq))
-				continue;
-			if (__netif_subqueue_stopped(lio->netdev, q)) {
-				wake_q(lio->netdev, q);
-				INCR_INSTRQUEUE_PKT_COUNT(lio->oct_dev, iq,
-							  tx_restart, 1);
-				ret_val++;
-			}
-		}
-	} else {
-		if (octnet_iq_is_full(lio->oct_dev, lio->txq))
-			return 0;
-		wake_q(lio->netdev, lio->txq);
-		INCR_INSTRQUEUE_PKT_COUNT(lio->oct_dev, lio->txq,
-					  tx_restart, 1);
-		ret_val = 1;
-	}
-	return ret_val;
-}
-
-/**
- * Remove the node at the head of the list. The list would be empty at
- * the end of this call if there are no more nodes in the list.
- */
-static inline struct list_head *list_delete_head(struct list_head *root)
-{
-	struct list_head *node;
-
-	if ((root->prev == root) && (root->next == root))
-		node = NULL;
-	else
-		node = root->next;
-
-	if (node)
-		list_del(node);
-
-	return node;
-}
-
-/**
- * \brief Delete gather lists
- * @param lio per-network private data
- */
-static void delete_glists(struct lio *lio)
-{
-	struct octnic_gather *g;
-	int i;
-
-	kfree(lio->glist_lock);
-	lio->glist_lock = NULL;
-
-	if (!lio->glist)
-		return;
-
-	for (i = 0; i < lio->linfo.num_txpciq; i++) {
-		do {
-			g = (struct octnic_gather *)
-				list_delete_head(&lio->glist[i]);
-			if (g)
-				kfree(g);
-		} while (g);
-
-		if (lio->glists_virt_base && lio->glists_virt_base[i] &&
-		    lio->glists_dma_base && lio->glists_dma_base[i]) {
-			lio_dma_free(lio->oct_dev,
-				     lio->glist_entry_size * lio->tx_qsize,
-				     lio->glists_virt_base[i],
-				     lio->glists_dma_base[i]);
-		}
-	}
-
-	kfree(lio->glists_virt_base);
-	lio->glists_virt_base = NULL;
-
-	kfree(lio->glists_dma_base);
-	lio->glists_dma_base = NULL;
-
-	kfree(lio->glist);
-	lio->glist = NULL;
-}
-
-/**
- * \brief Setup gather lists
- * @param lio per-network private data
- */
-static int setup_glists(struct octeon_device *oct, struct lio *lio, int num_iqs)
-{
-	int i, j;
-	struct octnic_gather *g;
-
-	lio->glist_lock = kcalloc(num_iqs, sizeof(*lio->glist_lock),
-				  GFP_KERNEL);
-	if (!lio->glist_lock)
-		return -ENOMEM;
-
-	lio->glist = kcalloc(num_iqs, sizeof(*lio->glist),
-			     GFP_KERNEL);
-	if (!lio->glist) {
-		kfree(lio->glist_lock);
-		lio->glist_lock = NULL;
-		return -ENOMEM;
-	}
-
-	lio->glist_entry_size =
-		ROUNDUP8((ROUNDUP4(OCTNIC_MAX_SG) >> 2) * OCT_SG_ENTRY_SIZE);
-
-	/* allocate memory to store virtual and dma base address of
-	 * per glist consistent memory
-	 */
-	lio->glists_virt_base = kcalloc(num_iqs, sizeof(*lio->glists_virt_base),
-					GFP_KERNEL);
-	lio->glists_dma_base = kcalloc(num_iqs, sizeof(*lio->glists_dma_base),
-				       GFP_KERNEL);
-
-	if (!lio->glists_virt_base || !lio->glists_dma_base) {
-		delete_glists(lio);
-		return -ENOMEM;
-	}
-
-	for (i = 0; i < num_iqs; i++) {
-		int numa_node = dev_to_node(&oct->pci_dev->dev);
-
-		spin_lock_init(&lio->glist_lock[i]);
-
-		INIT_LIST_HEAD(&lio->glist[i]);
-
-		lio->glists_virt_base[i] =
-			lio_dma_alloc(oct,
-				      lio->glist_entry_size * lio->tx_qsize,
-				      &lio->glists_dma_base[i]);
-
-		if (!lio->glists_virt_base[i]) {
-			delete_glists(lio);
-			return -ENOMEM;
-		}
-
-		for (j = 0; j < lio->tx_qsize; j++) {
-			g = kzalloc_node(sizeof(*g), GFP_KERNEL,
-					 numa_node);
-			if (!g)
-				g = kzalloc(sizeof(*g), GFP_KERNEL);
-			if (!g)
-				break;
-
-			g->sg = lio->glists_virt_base[i] +
-				(j * lio->glist_entry_size);
-
-			g->sg_dma_ptr = lio->glists_dma_base[i] +
-					(j * lio->glist_entry_size);
-
-			list_add_tail(&g->list, &lio->glist[i]);
-		}
-
-		if (j != lio->tx_qsize) {
-			delete_glists(lio);
-			return -ENOMEM;
+	/* check each sub-queue state */
+	for (q = 0; q < numqs; q++) {
+		iq = lio->linfo.txpciq[q %
+			lio->oct_dev->num_iqs].s.q_no;
+		if (octnet_iq_is_full(lio->oct_dev, iq))
+			continue;
+		if (__netif_subqueue_stopped(lio->netdev, q)) {
+			netif_wake_subqueue(lio->netdev, q);
+			INCR_INSTRQUEUE_PKT_COUNT(lio->oct_dev, iq,
+						  tx_restart, 1);
+			ret_val++;
 		}
 	}
 
-	return 0;
+	return ret_val;
 }
 
 /**
- * \brief Print link information
- * @param netdev network device
+ * print_link_info -  Print link information
+ * @netdev: network device
  */
 static void print_link_info(struct net_device *netdev)
 {
 	struct lio *lio = GET_LIO(netdev);
 
-	if (atomic_read(&lio->ifstate) & LIO_IFSTATE_REGISTERED) {
+	if (!ifstate_check(lio, LIO_IFSTATE_RESETTING) &&
+	    ifstate_check(lio, LIO_IFSTATE_REGISTERED)) {
 		struct oct_link_info *linfo = &lio->linfo;
 
 		if (linfo->link.s.link_up) {
@@ -837,22 +499,26 @@ static void print_link_info(struct net_device *netdev)
 }
 
 /**
- * \brief Routine to notify MTU change
- * @param work work_struct data structure
+ * octnet_link_status_change - Routine to notify MTU change
+ * @work: work_struct data structure
  */
 static void octnet_link_status_change(struct work_struct *work)
 {
 	struct cavium_wk *wk = (struct cavium_wk *)work;
 	struct lio *lio = (struct lio *)wk->ctxptr;
 
+	/* lio->linfo.link.s.mtu always contains max MTU of the lio interface.
+	 * this API is invoked only when new max-MTU of the interface is
+	 * less than current MTU.
+	 */
 	rtnl_lock();
-	call_netdevice_notifiers(NETDEV_CHANGEMTU, lio->netdev);
+	dev_set_mtu(lio->netdev, lio->linfo.link.s.mtu);
 	rtnl_unlock();
 }
 
 /**
- * \brief Sets up the mtu status change work
- * @param netdev network device
+ * setup_link_status_change_wq - Sets up the mtu status change work
+ * @netdev: network device
  */
 static inline int setup_link_status_change_wq(struct net_device *netdev)
 {
@@ -860,7 +526,8 @@ static inline int setup_link_status_change_wq(struct net_device *netdev)
 	struct octeon_device *oct = lio->oct_dev;
 
 	lio->link_status_wq.wq = alloc_workqueue("link-status",
-						 WQ_MEM_RECLAIM, 0);
+						 WQ_MEM_RECLAIM | WQ_PERCPU,
+						 0);
 	if (!lio->link_status_wq.wq) {
 		dev_err(&oct->pci_dev->dev, "unable to create cavium link status wq\n");
 		return -1;
@@ -883,9 +550,9 @@ static inline void cleanup_link_status_change_wq(struct net_device *netdev)
 }
 
 /**
- * \brief Update link status
- * @param netdev network device
- * @param ls link status structure
+ * update_link_status - Update link status
+ * @netdev: network device
+ * @ls: link status structure
  *
  * Called on receipt of a link status response from the core application to
  * update each interface's link status.
@@ -895,7 +562,11 @@ static inline void update_link_status(struct net_device *netdev,
 {
 	struct lio *lio = GET_LIO(netdev);
 	int changed = (lio->linfo.link.u64 != ls->u64);
+	int current_max_mtu = lio->linfo.link.s.mtu;
+	struct octeon_device *oct = lio->oct_dev;
 
+	dev_dbg(&oct->pci_dev->dev, "%s: lio->linfo.link.u64=%llx, ls->u64=%llx\n",
+		__func__, lio->linfo.link.u64, ls->u64);
 	lio->linfo.link.u64 = ls->u64;
 
 	if ((lio->intf_open) && (changed)) {
@@ -903,302 +574,125 @@ static inline void update_link_status(struct net_device *netdev,
 		lio->link_changes++;
 
 		if (lio->linfo.link.s.link_up) {
+			dev_dbg(&oct->pci_dev->dev, "%s: link_up", __func__);
 			netif_carrier_on(netdev);
-			txqs_wake(netdev);
+			wake_txqs(netdev);
 		} else {
+			dev_dbg(&oct->pci_dev->dev, "%s: link_off", __func__);
 			netif_carrier_off(netdev);
-			stop_txq(netdev);
+			stop_txqs(netdev);
+		}
+		if (lio->linfo.link.s.mtu != current_max_mtu) {
+			netif_info(lio, probe, lio->netdev, "Max MTU changed from %d to %d\n",
+				   current_max_mtu, lio->linfo.link.s.mtu);
+			netdev->max_mtu = lio->linfo.link.s.mtu;
+		}
+		if (lio->linfo.link.s.mtu < netdev->mtu) {
+			dev_warn(&oct->pci_dev->dev,
+				 "Current MTU is higher than new max MTU; Reducing the current mtu from %d to %d\n",
+				     netdev->mtu, lio->linfo.link.s.mtu);
+			queue_delayed_work(lio->link_status_wq.wq,
+					   &lio->link_status_wq.wk.work, 0);
 		}
 	}
 }
 
-/* Runs in interrupt context. */
-static void update_txq_status(struct octeon_device *oct, int iq_num)
+/**
+ * lio_sync_octeon_time - send latest localtime to octeon firmware so that
+ * firmware will correct it's time, in case there is a time skew
+ *
+ * @work: work scheduled to send time update to octeon firmware
+ **/
+static void lio_sync_octeon_time(struct work_struct *work)
 {
-	struct net_device *netdev;
-	struct lio *lio;
-	struct octeon_instr_queue *iq = oct->instr_queue[iq_num];
-
-	netdev = oct->props[iq->ifidx].netdev;
+	struct cavium_wk *wk = (struct cavium_wk *)work;
+	struct lio *lio = (struct lio *)wk->ctxptr;
+	struct octeon_device *oct = lio->oct_dev;
+	struct octeon_soft_command *sc;
+	struct timespec64 ts;
+	struct lio_time *lt;
+	int ret;
 
-	/* This is needed because the first IQ does not have
-	 * a netdev associated with it.
-	 */
-	if (!netdev)
+	sc = octeon_alloc_soft_command(oct, sizeof(struct lio_time), 16, 0);
+	if (!sc) {
+		dev_err(&oct->pci_dev->dev,
+			"Failed to sync time to octeon: soft command allocation failed\n");
 		return;
-
-	lio = GET_LIO(netdev);
-	if (netif_is_multiqueue(netdev)) {
-		if (__netif_subqueue_stopped(netdev, iq->q_index) &&
-		    lio->linfo.link.s.link_up &&
-		    (!octnet_iq_is_full(oct, iq_num))) {
-			INCR_INSTRQUEUE_PKT_COUNT(lio->oct_dev, iq_num,
-						  tx_restart, 1);
-			netif_wake_subqueue(netdev, iq->q_index);
-		}
-	} else if (netif_queue_stopped(netdev) &&
-		   lio->linfo.link.s.link_up &&
-		   (!octnet_iq_is_full(oct, lio->txq))) {
-		INCR_INSTRQUEUE_PKT_COUNT(lio->oct_dev,
-					  lio->txq, tx_restart, 1);
-		netif_wake_queue(netdev);
 	}
-}
 
-static
-int liquidio_schedule_msix_droq_pkt_handler(struct octeon_droq *droq, u64 ret)
-{
-	struct octeon_device *oct = droq->oct_dev;
-	struct octeon_device_priv *oct_priv =
-	    (struct octeon_device_priv *)oct->priv;
+	lt = (struct lio_time *)sc->virtdptr;
 
-	if (droq->ops.poll_mode) {
-		droq->ops.napi_fn(droq);
-	} else {
-		if (ret & MSIX_PO_INT) {
-			tasklet_schedule(&oct_priv->droq_tasklet);
-			return 1;
-		}
-		/* this will be flushed periodically by check iq db */
-		if (ret & MSIX_PI_INT)
-			return 0;
-	}
-	return 0;
-}
+	/* Get time of the day */
+	ktime_get_real_ts64(&ts);
+	lt->sec = ts.tv_sec;
+	lt->nsec = ts.tv_nsec;
+	octeon_swap_8B_data((u64 *)lt, (sizeof(struct lio_time)) / 8);
 
-/**
- * \brief Droq packet processor sceduler
- * @param oct octeon device
- */
-static void liquidio_schedule_droq_pkt_handlers(struct octeon_device *oct)
-{
-	struct octeon_device_priv *oct_priv =
-		(struct octeon_device_priv *)oct->priv;
-	u64 oq_no;
-	struct octeon_droq *droq;
-
-	if (oct->int_status & OCT_DEV_INTR_PKT_DATA) {
-		for (oq_no = 0; oq_no < MAX_OCTEON_OUTPUT_QUEUES(oct);
-		     oq_no++) {
-			if (!(oct->droq_intr & BIT_ULL(oq_no)))
-				continue;
+	sc->iq_no = lio->linfo.txpciq[0].s.q_no;
+	octeon_prepare_soft_command(oct, sc, OPCODE_NIC,
+				    OPCODE_NIC_SYNC_OCTEON_TIME, 0, 0, 0);
 
-			droq = oct->droq[oq_no];
+	init_completion(&sc->complete);
+	sc->sc_status = OCTEON_REQUEST_PENDING;
 
-			if (droq->ops.poll_mode) {
-				droq->ops.napi_fn(droq);
-				oct_priv->napi_mask |= (1 << oq_no);
-			} else {
-				tasklet_schedule(&oct_priv->droq_tasklet);
-			}
-		}
+	ret = octeon_send_soft_command(oct, sc);
+	if (ret == IQ_SEND_FAILED) {
+		dev_err(&oct->pci_dev->dev,
+			"Failed to sync time to octeon: failed to send soft command\n");
+		octeon_free_soft_command(oct, sc);
+	} else {
+		WRITE_ONCE(sc->caller_is_done, true);
 	}
-}
-
-static irqreturn_t
-liquidio_msix_intr_handler(int irq __attribute__((unused)), void *dev)
-{
-	u64 ret;
-	struct octeon_ioq_vector *ioq_vector = (struct octeon_ioq_vector *)dev;
-	struct octeon_device *oct = ioq_vector->oct_dev;
-	struct octeon_droq *droq = oct->droq[ioq_vector->droq_index];
-
-	ret = oct->fn_list.msix_interrupt_handler(ioq_vector);
-
-	if ((ret & MSIX_PO_INT) || (ret & MSIX_PI_INT))
-		liquidio_schedule_msix_droq_pkt_handler(droq, ret);
 
-	return IRQ_HANDLED;
+	queue_delayed_work(lio->sync_octeon_time_wq.wq,
+			   &lio->sync_octeon_time_wq.wk.work,
+			   msecs_to_jiffies(LIO_SYNC_OCTEON_TIME_INTERVAL_MS));
 }
 
 /**
- * \brief Interrupt handler for octeon
- * @param irq unused
- * @param dev octeon device
- */
-static
-irqreturn_t liquidio_legacy_intr_handler(int irq __attribute__((unused)),
-					 void *dev)
+ * setup_sync_octeon_time_wq - prepare work to periodically update local time to octeon firmware
+ *
+ * @netdev: network device which should send time update to firmware
+ **/
+static inline int setup_sync_octeon_time_wq(struct net_device *netdev)
 {
-	struct octeon_device *oct = (struct octeon_device *)dev;
-	irqreturn_t ret;
-
-	/* Disable our interrupts for the duration of ISR */
-	oct->fn_list.disable_interrupt(oct, OCTEON_ALL_INTR);
-
-	ret = oct->fn_list.process_interrupt_regs(oct);
-
-	if (ret == IRQ_HANDLED)
-		liquidio_schedule_droq_pkt_handlers(oct);
+	struct lio *lio = GET_LIO(netdev);
+	struct octeon_device *oct = lio->oct_dev;
 
-	/* Re-enable our interrupts  */
-	if (!(atomic_read(&oct->status) == OCT_DEV_IN_RESET))
-		oct->fn_list.enable_interrupt(oct, OCTEON_ALL_INTR);
+	lio->sync_octeon_time_wq.wq =
+		alloc_workqueue("update-octeon-time",
+				WQ_MEM_RECLAIM | WQ_PERCPU, 0);
+	if (!lio->sync_octeon_time_wq.wq) {
+		dev_err(&oct->pci_dev->dev, "Unable to create wq to update octeon time\n");
+		return -1;
+	}
+	INIT_DELAYED_WORK(&lio->sync_octeon_time_wq.wk.work,
+			  lio_sync_octeon_time);
+	lio->sync_octeon_time_wq.wk.ctxptr = lio;
+	queue_delayed_work(lio->sync_octeon_time_wq.wq,
+			   &lio->sync_octeon_time_wq.wk.work,
+			   msecs_to_jiffies(LIO_SYNC_OCTEON_TIME_INTERVAL_MS));
 
-	return ret;
+	return 0;
 }
 
 /**
- * \brief Setup interrupt for octeon device
- * @param oct octeon device
+ * cleanup_sync_octeon_time_wq - destroy wq
  *
- *  Enable interrupt in Octeon device as given in the PCI interrupt mask.
- */
-static int octeon_setup_interrupt(struct octeon_device *oct)
+ * @netdev: network device which should send time update to firmware
+ *
+ * Stop scheduling and destroy the work created to periodically update local
+ * time to octeon firmware.
+ **/
+static inline void cleanup_sync_octeon_time_wq(struct net_device *netdev)
 {
-	int irqret, err;
-	struct msix_entry *msix_entries;
-	int i;
-	int num_ioq_vectors;
-	int num_alloc_ioq_vectors;
-	char *queue_irq_names = NULL;
-	char *aux_irq_name = NULL;
-
-	if (OCTEON_CN23XX_PF(oct) && oct->msix_on) {
-		oct->num_msix_irqs = oct->sriov_info.num_pf_rings;
-		/* one non ioq interrupt for handling sli_mac_pf_int_sum */
-		oct->num_msix_irqs += 1;
-
-		/* allocate storage for the names assigned to each irq */
-		oct->irq_name_storage =
-			kcalloc((MAX_IOQ_INTERRUPTS_PER_PF + 1), INTRNAMSIZ,
-				GFP_KERNEL);
-		if (!oct->irq_name_storage) {
-			dev_err(&oct->pci_dev->dev, "Irq name storage alloc failed...\n");
-			return -ENOMEM;
-		}
-
-		queue_irq_names = oct->irq_name_storage;
-		aux_irq_name = &queue_irq_names
-				[IRQ_NAME_OFF(MAX_IOQ_INTERRUPTS_PER_PF)];
-
-		oct->msix_entries = kcalloc(
-		    oct->num_msix_irqs, sizeof(struct msix_entry), GFP_KERNEL);
-		if (!oct->msix_entries) {
-			dev_err(&oct->pci_dev->dev, "Memory Alloc failed...\n");
-			kfree(oct->irq_name_storage);
-			oct->irq_name_storage = NULL;
-			return -ENOMEM;
-		}
-
-		msix_entries = (struct msix_entry *)oct->msix_entries;
-		/*Assumption is that pf msix vectors start from pf srn to pf to
-		 * trs and not from 0. if not change this code
-		 */
-		for (i = 0; i < oct->num_msix_irqs - 1; i++)
-			msix_entries[i].entry = oct->sriov_info.pf_srn + i;
-		msix_entries[oct->num_msix_irqs - 1].entry =
-		    oct->sriov_info.trs;
-		num_alloc_ioq_vectors = pci_enable_msix_range(
-						oct->pci_dev, msix_entries,
-						oct->num_msix_irqs,
-						oct->num_msix_irqs);
-		if (num_alloc_ioq_vectors < 0) {
-			dev_err(&oct->pci_dev->dev, "unable to Allocate MSI-X interrupts\n");
-			kfree(oct->msix_entries);
-			oct->msix_entries = NULL;
-			kfree(oct->irq_name_storage);
-			oct->irq_name_storage = NULL;
-			return num_alloc_ioq_vectors;
-		}
-		dev_dbg(&oct->pci_dev->dev, "OCTEON: Enough MSI-X interrupts are allocated...\n");
-
-		num_ioq_vectors = oct->num_msix_irqs;
-
-		/** For PF, there is one non-ioq interrupt handler */
-		num_ioq_vectors -= 1;
-
-		snprintf(aux_irq_name, INTRNAMSIZ,
-			 "LiquidIO%u-pf%u-aux", oct->octeon_id, oct->pf_num);
-		irqret = request_irq(msix_entries[num_ioq_vectors].vector,
-				     liquidio_legacy_intr_handler, 0,
-				     aux_irq_name, oct);
-		if (irqret) {
-			dev_err(&oct->pci_dev->dev,
-				"OCTEON: Request_irq failed for MSIX interrupt Error: %d\n",
-				irqret);
-			pci_disable_msix(oct->pci_dev);
-			kfree(oct->msix_entries);
-			oct->msix_entries = NULL;
-			kfree(oct->irq_name_storage);
-			oct->irq_name_storage = NULL;
-			return irqret;
-		}
-
-		for (i = 0; i < num_ioq_vectors; i++) {
-			snprintf(&queue_irq_names[IRQ_NAME_OFF(i)], INTRNAMSIZ,
-				 "LiquidIO%u-pf%u-rxtx-%u",
-				 oct->octeon_id, oct->pf_num, i);
-
-			irqret = request_irq(msix_entries[i].vector,
-					     liquidio_msix_intr_handler, 0,
-					     &queue_irq_names[IRQ_NAME_OFF(i)],
-					     &oct->ioq_vector[i]);
-			if (irqret) {
-				dev_err(&oct->pci_dev->dev,
-					"OCTEON: Request_irq failed for MSIX interrupt Error: %d\n",
-					irqret);
-				/** Freeing the non-ioq irq vector here . */
-				free_irq(msix_entries[num_ioq_vectors].vector,
-					 oct);
-
-				while (i) {
-					i--;
-					/** clearing affinity mask. */
-					irq_set_affinity_hint(
-						msix_entries[i].vector, NULL);
-					free_irq(msix_entries[i].vector,
-						 &oct->ioq_vector[i]);
-				}
-				pci_disable_msix(oct->pci_dev);
-				kfree(oct->msix_entries);
-				oct->msix_entries = NULL;
-				kfree(oct->irq_name_storage);
-				oct->irq_name_storage = NULL;
-				return irqret;
-			}
-			oct->ioq_vector[i].vector = msix_entries[i].vector;
-			/* assign the cpu mask for this msix interrupt vector */
-			irq_set_affinity_hint(
-					msix_entries[i].vector,
-					(&oct->ioq_vector[i].affinity_mask));
-		}
-		dev_dbg(&oct->pci_dev->dev, "OCTEON[%d]: MSI-X enabled\n",
-			oct->octeon_id);
-	} else {
-		err = pci_enable_msi(oct->pci_dev);
-		if (err)
-			dev_warn(&oct->pci_dev->dev, "Reverting to legacy interrupts. Error: %d\n",
-				 err);
-		else
-			oct->flags |= LIO_FLAG_MSI_ENABLED;
-
-		/* allocate storage for the names assigned to the irq */
-		oct->irq_name_storage = kcalloc(1, INTRNAMSIZ, GFP_KERNEL);
-		if (!oct->irq_name_storage)
-			return -ENOMEM;
-
-		queue_irq_names = oct->irq_name_storage;
-
-		snprintf(&queue_irq_names[IRQ_NAME_OFF(0)], INTRNAMSIZ,
-			 "LiquidIO%u-pf%u-rxtx-%u",
-			 oct->octeon_id, oct->pf_num, 0);
+	struct lio *lio = GET_LIO(netdev);
+	struct cavium_wq *time_wq = &lio->sync_octeon_time_wq;
 
-		irqret = request_irq(oct->pci_dev->irq,
-				     liquidio_legacy_intr_handler,
-				     IRQF_SHARED,
-				     &queue_irq_names[IRQ_NAME_OFF(0)], oct);
-		if (irqret) {
-			if (oct->flags & LIO_FLAG_MSI_ENABLED)
-				pci_disable_msi(oct->pci_dev);
-			dev_err(&oct->pci_dev->dev, "Request IRQ failed with code: %d\n",
-				irqret);
-			kfree(oct->irq_name_storage);
-			oct->irq_name_storage = NULL;
-			return irqret;
-		}
+	if (time_wq->wq) {
+		cancel_delayed_work_sync(&time_wq->wk.work);
+		destroy_workqueue(time_wq->wq);
 	}
-	return 0;
 }
 
 static struct octeon_device *get_other_octeon_device(struct octeon_device *oct)
@@ -1291,7 +785,7 @@ static int liquidio_watchdog(void *param)
 				dev_err(&oct->pci_dev->dev,
 					"ERROR: Octeon core %d crashed or got stuck!  See oct-fwdump for details.\n",
 					core);
-					err_msg_was_printed[core] = true;
+				err_msg_was_printed[core] = true;
 			}
 		}
 
@@ -1323,13 +817,12 @@ static int liquidio_watchdog(void *param)
 }
 
 /**
- * \brief PCI probe handler
- * @param pdev PCI device structure
- * @param ent unused
+ * liquidio_probe - PCI probe handler
+ * @pdev: PCI device structure
+ * @ent: unused
  */
 static int
-liquidio_probe(struct pci_dev *pdev,
-	       const struct pci_device_id *ent __attribute__((unused)))
+liquidio_probe(struct pci_dev *pdev, const struct pci_device_id __maybe_unused *ent)
 {
 	struct octeon_device *oct_dev = NULL;
 	struct handshake *hs;
@@ -1344,6 +837,13 @@ liquidio_probe(struct pci_dev *pdev,
 	if (pdev->device == OCTEON_CN23XX_PF_VID)
 		oct_dev->msix_on = LIO_FLAG_MSIX_ENABLED;
 
+	/* Enable PTP for 6XXX Device */
+	if (((pdev->device == OCTEON_CN66XX) ||
+	     (pdev->device == OCTEON_CN68XX)))
+		oct_dev->ptp_enable = true;
+	else
+		oct_dev->ptp_enable = false;
+
 	dev_info(&pdev->dev, "Initializing device %x:%x.\n",
 		 (u32)pdev->vendor, (u32)pdev->device);
 
@@ -1353,6 +853,9 @@ liquidio_probe(struct pci_dev *pdev,
 	/* set linux specific device pointer */
 	oct_dev->pci_dev = (void *)pdev;
 
+	oct_dev->subsystem_id = pdev->subsystem_vendor |
+		(pdev->subsystem_device << 16);
+
 	hs = &handshake[oct_dev->octeon_id];
 	init_completion(&hs->init);
 	init_completion(&hs->started);
@@ -1369,28 +872,21 @@ liquidio_probe(struct pci_dev *pdev,
 	}
 
 	if (OCTEON_CN23XX_PF(oct_dev)) {
-		u64 scratch1;
 		u8 bus, device, function;
 
-		scratch1 = octeon_read_csr64(oct_dev, CN23XX_SLI_SCRATCH1);
-		if (!(scratch1 & 4ULL)) {
-			/* Bit 2 of SLI_SCRATCH_1 is a flag that indicates that
-			 * the lio watchdog kernel thread is running for this
-			 * NIC.  Each NIC gets one watchdog kernel thread.
+		if (atomic_read(oct_dev->adapter_refcount) == 1) {
+			/* Each NIC gets one watchdog kernel thread.  The first
+			 * PF (of each NIC) that gets pci_driver->probe()'d
+			 * creates that thread.
 			 */
-			scratch1 |= 4ULL;
-			octeon_write_csr64(oct_dev, CN23XX_SLI_SCRATCH1,
-					   scratch1);
-
 			bus = pdev->bus->number;
 			device = PCI_SLOT(pdev->devfn);
 			function = PCI_FUNC(pdev->devfn);
-			oct_dev->watchdog_task = kthread_create(
-			    liquidio_watchdog, oct_dev,
-			    "liowd/%02hhx:%02hhx.%hhx", bus, device, function);
-			if (!IS_ERR(oct_dev->watchdog_task)) {
-				wake_up_process(oct_dev->watchdog_task);
-			} else {
+			oct_dev->watchdog_task = kthread_run(liquidio_watchdog,
+							     oct_dev,
+							     "liowd/%02hhx:%02hhx.%hhx",
+							     bus, device, function);
+			if (IS_ERR(oct_dev->watchdog_task)) {
 				oct_dev->watchdog_task = NULL;
 				dev_err(&oct_dev->pci_dev->dev,
 					"failed to create kernel_thread\n");
@@ -1408,23 +904,48 @@ liquidio_probe(struct pci_dev *pdev,
 	return 0;
 }
 
-static bool fw_type_is_none(void)
+static bool fw_type_is_auto(void)
 {
-	return strncmp(fw_type, LIO_FW_NAME_TYPE_NONE,
-		       sizeof(LIO_FW_NAME_TYPE_NONE)) == 0;
+	return strncmp(fw_type, LIO_FW_NAME_TYPE_AUTO,
+		       sizeof(LIO_FW_NAME_TYPE_AUTO)) == 0;
+}
+
+/**
+ * octeon_pci_flr - PCI FLR for each Octeon device.
+ * @oct: octeon device
+ */
+static void octeon_pci_flr(struct octeon_device *oct)
+{
+	int rc;
+
+	pci_save_state(oct->pci_dev);
+
+	pci_cfg_access_lock(oct->pci_dev);
+
+	/* Quiesce the device completely */
+	pci_write_config_word(oct->pci_dev, PCI_COMMAND,
+			      PCI_COMMAND_INTX_DISABLE);
+
+	rc = __pci_reset_function_locked(oct->pci_dev);
+
+	if (rc != 0)
+		dev_err(&oct->pci_dev->dev, "Error %d resetting PCI function %d\n",
+			rc, oct->pf_num);
+
+	pci_cfg_access_unlock(oct->pci_dev);
+
+	pci_restore_state(oct->pci_dev);
 }
 
 /**
- *\brief Destroy resources associated with octeon device
- * @param pdev PCI device structure
- * @param ent unused
+ * octeon_destroy_resources - Destroy resources associated with octeon device
+ * @oct: octeon device
  */
 static void octeon_destroy_resources(struct octeon_device *oct)
 {
 	int i, refcount;
 	struct msix_entry *msix_entries;
-	struct octeon_device_priv *oct_priv =
-		(struct octeon_device_priv *)oct->priv;
+	struct octeon_device_priv *oct_priv = oct->priv;
 
 	struct handshake *hs;
 
@@ -1441,22 +962,21 @@ static void octeon_destroy_resources(struct octeon_device *oct)
 
 		schedule_timeout_uninterruptible(HZ / 10);
 
-		/* fallthrough */
+		fallthrough;
 	case OCT_DEV_HOST_OK:
 
-		/* fallthrough */
 	case OCT_DEV_CONSOLE_INIT_DONE:
 		/* Remove any consoles */
 		octeon_remove_consoles(oct);
 
-		/* fallthrough */
+		fallthrough;
 	case OCT_DEV_IO_QUEUES_DONE:
-		if (wait_for_pending_requests(oct))
-			dev_err(&oct->pci_dev->dev, "There were pending requests\n");
-
 		if (lio_wait_for_instr_fetch(oct))
 			dev_err(&oct->pci_dev->dev, "IQ had pending instructions\n");
 
+		if (wait_for_pending_requests(oct))
+			dev_err(&oct->pci_dev->dev, "There were pending requests\n");
+
 		/* Disable the input and output queues now. No more packets will
 		 * arrive from Octeon, but we should wait for all packet
 		 * processing to finish.
@@ -1466,7 +986,32 @@ static void octeon_destroy_resources(struct octeon_device *oct)
 		if (lio_wait_for_oq_pkts(oct))
 			dev_err(&oct->pci_dev->dev, "OQ had pending packets\n");
 
-	/* fallthrough */
+		/* Force all requests waiting to be fetched by OCTEON to
+		 * complete.
+		 */
+		for (i = 0; i < MAX_OCTEON_INSTR_QUEUES(oct); i++) {
+			struct octeon_instr_queue *iq;
+
+			if (!(oct->io_qmask.iq & BIT_ULL(i)))
+				continue;
+			iq = oct->instr_queue[i];
+
+			if (atomic_read(&iq->instr_pending)) {
+				spin_lock_bh(&iq->lock);
+				iq->fill_cnt = 0;
+				iq->octeon_read_index = iq->host_write_index;
+				iq->stats.instr_processed +=
+					atomic_read(&iq->instr_pending);
+				lio_process_iq_request_list(oct, iq, 0);
+				spin_unlock_bh(&iq->lock);
+			}
+		}
+
+		lio_process_ordered_list(oct, 1);
+		octeon_free_sc_done_list(oct);
+		octeon_free_sc_zombie_list(oct);
+
+		fallthrough;
 	case OCT_DEV_INTR_SET_DONE:
 		/* Disable interrupts  */
 		oct->fn_list.disable_interrupt(oct, OCTEON_ALL_INTR);
@@ -1474,11 +1019,15 @@ static void octeon_destroy_resources(struct octeon_device *oct)
 		if (oct->msix_on) {
 			msix_entries = (struct msix_entry *)oct->msix_entries;
 			for (i = 0; i < oct->num_msix_irqs - 1; i++) {
-				/* clear the affinity_cpumask */
-				irq_set_affinity_hint(msix_entries[i].vector,
-						      NULL);
-				free_irq(msix_entries[i].vector,
-					 &oct->ioq_vector[i]);
+				if (oct->ioq_vector[i].vector) {
+					/* clear the affinity_cpumask */
+					irq_set_affinity_hint(
+							msix_entries[i].vector,
+							NULL);
+					free_irq(msix_entries[i].vector,
+						 &oct->ioq_vector[i]);
+					oct->ioq_vector[i].vector = 0;
+				}
 			}
 			/* non-iov vector's argument is oct struct */
 			free_irq(msix_entries[i].vector, oct);
@@ -1497,17 +1046,17 @@ static void octeon_destroy_resources(struct octeon_device *oct)
 		kfree(oct->irq_name_storage);
 		oct->irq_name_storage = NULL;
 
-	/* fallthrough */
+		fallthrough;
 	case OCT_DEV_MSIX_ALLOC_VECTOR_DONE:
 		if (OCTEON_CN23XX_PF(oct))
 			octeon_free_ioq_vector(oct);
 
-	/* fallthrough */
+		fallthrough;
 	case OCT_DEV_MBOX_SETUP_DONE:
 		if (OCTEON_CN23XX_PF(oct))
 			oct->fn_list.free_mbox(oct);
 
-	/* fallthrough */
+		fallthrough;
 	case OCT_DEV_IN_RESET:
 	case OCT_DEV_DROQ_INIT_DONE:
 		/* Wait for any pending operations */
@@ -1530,11 +1079,11 @@ static void octeon_destroy_resources(struct octeon_device *oct)
 			}
 		}
 
-		/* fallthrough */
+		fallthrough;
 	case OCT_DEV_RESP_LIST_INIT_DONE:
 		octeon_delete_response_list(oct);
 
-		/* fallthrough */
+		fallthrough;
 	case OCT_DEV_INSTR_QUEUE_INIT_DONE:
 		for (i = 0; i < MAX_OCTEON_INSTR_QUEUES(oct); i++) {
 			if (!(oct->io_qmask.iq & BIT_ULL(i)))
@@ -1545,38 +1094,39 @@ static void octeon_destroy_resources(struct octeon_device *oct)
 		if (oct->sriov_info.sriov_enabled)
 			pci_disable_sriov(oct->pci_dev);
 #endif
-		/* fallthrough */
+		fallthrough;
 	case OCT_DEV_SC_BUFF_POOL_INIT_DONE:
 		octeon_free_sc_buffer_pool(oct);
 
-		/* fallthrough */
+		fallthrough;
 	case OCT_DEV_DISPATCH_INIT_DONE:
 		octeon_delete_dispatch_list(oct);
 		cancel_delayed_work_sync(&oct->nic_poll_work.work);
 
-		/* fallthrough */
+		fallthrough;
 	case OCT_DEV_PCI_MAP_DONE:
 		refcount = octeon_deregister_device(oct);
 
-		if (!fw_type_is_none()) {
-			/* Soft reset the octeon device before exiting.
-			 * Implementation note: here, we reset the device
-			 * if it is a CN6XXX OR the last CN23XX device.
-			 */
-			if (OCTEON_CN6XXX(oct) || !refcount)
-				oct->fn_list.soft_reset(oct);
-		}
+		/* Soft reset the octeon device before exiting.
+		 * However, if fw was loaded from card (i.e. autoboot),
+		 * perform an FLR instead.
+		 * Implementation note: only soft-reset the device
+		 * if it is a CN6XXX OR the LAST CN23XX device.
+		 */
+		if (atomic_read(oct->adapter_fw_state) == FW_IS_PRELOADED)
+			octeon_pci_flr(oct);
+		else if (OCTEON_CN6XXX(oct) || !refcount)
+			oct->fn_list.soft_reset(oct);
 
 		octeon_unmap_pci_barx(oct, 0);
 		octeon_unmap_pci_barx(oct, 1);
 
-		/* fallthrough */
+		fallthrough;
 	case OCT_DEV_PCI_ENABLE_DONE:
-		pci_clear_master(oct->pci_dev);
 		/* Disable the device, releasing the PCI INT */
 		pci_disable_device(oct->pci_dev);
 
-		/* fallthrough */
+		fallthrough;
 	case OCT_DEV_BEGIN_STATE:
 		/* Nothing to be done here either */
 		break;
@@ -1586,60 +1136,30 @@ static void octeon_destroy_resources(struct octeon_device *oct)
 }
 
 /**
- * \brief Callback for rx ctrl
- * @param status status of request
- * @param buf pointer to resp structure
- */
-static void rx_ctl_callback(struct octeon_device *oct,
-			    u32 status,
-			    void *buf)
-{
-	struct octeon_soft_command *sc = (struct octeon_soft_command *)buf;
-	struct liquidio_rx_ctl_context *ctx;
-
-	ctx  = (struct liquidio_rx_ctl_context *)sc->ctxptr;
-
-	oct = lio_get_device(ctx->octeon_id);
-	if (status)
-		dev_err(&oct->pci_dev->dev, "rx ctl instruction failed. Status: %llx\n",
-			CVM_CAST64(status));
-	WRITE_ONCE(ctx->cond, 1);
-
-	/* This barrier is required to be sure that the response has been
-	 * written fully before waking up the handler
-	 */
-	wmb();
-
-	wake_up_interruptible(&ctx->wc);
-}
-
-/**
- * \brief Send Rx control command
- * @param lio per-network private data
- * @param start_stop whether to start or stop
+ * send_rx_ctrl_cmd - Send Rx control command
+ * @lio: per-network private data
+ * @start_stop: whether to start or stop
  */
-static void send_rx_ctrl_cmd(struct lio *lio, int start_stop)
+static int send_rx_ctrl_cmd(struct lio *lio, int start_stop)
 {
 	struct octeon_soft_command *sc;
-	struct liquidio_rx_ctl_context *ctx;
 	union octnet_cmd *ncmd;
-	int ctx_size = sizeof(struct liquidio_rx_ctl_context);
 	struct octeon_device *oct = (struct octeon_device *)lio->oct_dev;
 	int retval;
 
 	if (oct->props[lio->ifidx].rx_on == start_stop)
-		return;
+		return 0;
 
 	sc = (struct octeon_soft_command *)
 		octeon_alloc_soft_command(oct, OCTNET_CMD_SIZE,
-					  16, ctx_size);
+					  16, 0);
+	if (!sc) {
+		netif_info(lio, rx_err, lio->netdev,
+			   "Failed to allocate octeon_soft_command struct\n");
+		return -ENOMEM;
+	}
 
 	ncmd = (union octnet_cmd *)sc->virtdptr;
-	ctx  = (struct liquidio_rx_ctl_context *)sc->ctxptr;
-
-	WRITE_ONCE(ctx->cond, 0);
-	ctx->octeon_id = lio_get_device_id(oct);
-	init_waitqueue_head(&ctx->wc);
 
 	ncmd->u64 = 0;
 	ncmd->s.cmd = OCTNET_CMD_RX_CTL;
@@ -1652,29 +1172,32 @@ static void send_rx_ctrl_cmd(struct lio *lio, int start_stop)
 	octeon_prepare_soft_command(oct, sc, OPCODE_NIC,
 				    OPCODE_NIC_CMD, 0, 0, 0);
 
-	sc->callback = rx_ctl_callback;
-	sc->callback_arg = sc;
-	sc->wait_time = 5000;
+	init_completion(&sc->complete);
+	sc->sc_status = OCTEON_REQUEST_PENDING;
 
 	retval = octeon_send_soft_command(oct, sc);
 	if (retval == IQ_SEND_FAILED) {
 		netif_info(lio, rx_err, lio->netdev, "Failed to send RX Control message\n");
+		octeon_free_soft_command(oct, sc);
 	} else {
 		/* Sleep on a wait queue till the cond flag indicates that the
 		 * response arrived or timed-out.
 		 */
-		if (sleep_cond(&ctx->wc, &ctx->cond) == -EINTR)
-			return;
+		retval = wait_for_sc_completion_timeout(oct, sc, 0);
+		if (retval)
+			return retval;
+
 		oct->props[lio->ifidx].rx_on = start_stop;
+		WRITE_ONCE(sc->caller_is_done, true);
 	}
 
-	octeon_free_soft_command(oct, sc);
+	return retval;
 }
 
 /**
- * \brief Destroy NIC device interface
- * @param oct octeon device
- * @param ifidx which interface to destroy
+ * liquidio_destroy_nic_device - Destroy NIC device interface
+ * @oct: octeon device
+ * @ifidx: which interface to destroy
  *
  * Cleanup associated with each interface for an Octeon device  when NIC
  * module is being unloaded or if initialization fails during load.
@@ -1682,8 +1205,9 @@ static void send_rx_ctrl_cmd(struct lio *lio, int start_stop)
 static void liquidio_destroy_nic_device(struct octeon_device *oct, int ifidx)
 {
 	struct net_device *netdev = oct->props[ifidx].netdev;
-	struct lio *lio;
+	struct octeon_device_priv *oct_priv = oct->priv;
 	struct napi_struct *napi, *n;
+	struct lio *lio;
 
 	if (!netdev) {
 		dev_err(&oct->pci_dev->dev, "%s No netdevice ptr for index %d\n",
@@ -1698,15 +1222,6 @@ static void liquidio_destroy_nic_device(struct octeon_device *oct, int ifidx)
 	if (atomic_read(&lio->ifstate) & LIO_IFSTATE_RUNNING)
 		liquidio_stop(netdev);
 
-	if (fw_type_is_none()) {
-		struct octnic_ctrl_pkt nctrl;
-
-		memset(&nctrl, 0, sizeof(struct octnic_ctrl_pkt));
-		nctrl.ncmd.s.cmd = OCTNET_CMD_RESET_PF;
-		nctrl.iq_no = lio->linfo.txpciq[0].s.q_no;
-		octnet_send_nic_ctrl_pkt(oct, &nctrl);
-	}
-
 	if (oct->props[lio->ifidx].napi_enabled == 1) {
 		list_for_each_entry_safe(napi, n, &netdev->napi_list, dev_list)
 			napi_disable(napi);
@@ -1717,14 +1232,21 @@ static void liquidio_destroy_nic_device(struct octeon_device *oct, int ifidx)
 			oct->droq[0]->ops.poll_mode = 0;
 	}
 
+	/* Delete NAPI */
+	list_for_each_entry_safe(napi, n, &netdev->napi_list, dev_list)
+		netif_napi_del(napi);
+
+	tasklet_enable(&oct_priv->droq_tasklet);
+
 	if (atomic_read(&lio->ifstate) & LIO_IFSTATE_REGISTERED)
 		unregister_netdev(netdev);
 
+	cleanup_sync_octeon_time_wq(netdev);
 	cleanup_link_status_change_wq(netdev);
 
 	cleanup_rx_oom_poll_fn(netdev);
 
-	delete_glists(lio);
+	lio_delete_glists(lio);
 
 	free_netdev(netdev);
 
@@ -1734,8 +1256,8 @@ static void liquidio_destroy_nic_device(struct octeon_device *oct, int ifidx)
 }
 
 /**
- * \brief Stop complete NIC functionality
- * @param oct octeon device
+ * liquidio_stop_nic_module - Stop complete NIC functionality
+ * @oct: octeon device
  */
 static int liquidio_stop_nic_module(struct octeon_device *oct)
 {
@@ -1743,6 +1265,14 @@ static int liquidio_stop_nic_module(struct octeon_device *oct)
 	struct lio *lio;
 
 	dev_dbg(&oct->pci_dev->dev, "Stopping network interfaces\n");
+	device_lock(&oct->pci_dev->dev);
+	if (oct->devlink) {
+		devlink_unregister(oct->devlink);
+		devlink_free(oct->devlink);
+		oct->devlink = NULL;
+	}
+	device_unlock(&oct->pci_dev->dev);
+
 	if (!oct->ifcount) {
 		dev_err(&oct->pci_dev->dev, "Init for Octeon was not completed\n");
 		return 1;
@@ -1752,9 +1282,11 @@ static int liquidio_stop_nic_module(struct octeon_device *oct)
 	oct->cmd_resp_state = OCT_DRV_OFFLINE;
 	spin_unlock_bh(&oct->cmd_resp_wqlock);
 
+	lio_vf_rep_destroy(oct);
+
 	for (i = 0; i < oct->ifcount; i++) {
 		lio = GET_LIO(oct->props[i].netdev);
-		for (j = 0; j < lio->linfo.num_rxpciq; j++)
+		for (j = 0; j < oct->num_oqs; j++)
 			octeon_unregister_droq_ops(oct,
 						   lio->linfo.rxpciq[j].s.q_no);
 	}
@@ -1767,8 +1299,8 @@ static int liquidio_stop_nic_module(struct octeon_device *oct)
 }
 
 /**
- * \brief Cleans up resources at unload time
- * @param pdev PCI device structure
+ * liquidio_remove - Cleans up resources at unload time
+ * @pdev: PCI device structure
  */
 static void liquidio_remove(struct pci_dev *pdev)
 {
@@ -1779,6 +1311,10 @@ static void liquidio_remove(struct pci_dev *pdev)
 	if (oct_dev->watchdog_task)
 		kthread_stop(oct_dev->watchdog_task);
 
+	if (!oct_dev->octeon_id &&
+	    oct_dev->fw_info.app_cap_flags & LIQUIDIO_SWITCHDEV_CAP)
+		lio_vf_rep_modexit();
+
 	if (oct_dev->app_mode && (oct_dev->app_mode == CVM_DRV_NIC_APP))
 		liquidio_stop_nic_module(oct_dev);
 
@@ -1796,14 +1332,13 @@ static void liquidio_remove(struct pci_dev *pdev)
 }
 
 /**
- * \brief Identify the Octeon device and to map the BAR address space
- * @param oct octeon device
+ * octeon_chip_specific_setup - Identify the Octeon device and to map the BAR address space
+ * @oct: octeon device
  */
 static int octeon_chip_specific_setup(struct octeon_device *oct)
 {
 	u32 dev_id, rev_id;
 	int ret = 1;
-	char *s;
 
 	pci_read_config_dword(oct->pci_dev, 0, &dev_id);
 	pci_read_config_dword(oct->pci_dev, 8, &rev_id);
@@ -1813,40 +1348,36 @@ static int octeon_chip_specific_setup(struct octeon_device *oct)
 	case OCTEON_CN68XX_PCIID:
 		oct->chip_id = OCTEON_CN68XX;
 		ret = lio_setup_cn68xx_octeon_device(oct);
-		s = "CN68XX";
 		break;
 
 	case OCTEON_CN66XX_PCIID:
 		oct->chip_id = OCTEON_CN66XX;
 		ret = lio_setup_cn66xx_octeon_device(oct);
-		s = "CN66XX";
 		break;
 
 	case OCTEON_CN23XX_PCIID_PF:
 		oct->chip_id = OCTEON_CN23XX_PF_VID;
 		ret = setup_cn23xx_octeon_pf_device(oct);
-		s = "CN23XX";
+		if (ret)
+			break;
+#ifdef CONFIG_PCI_IOV
+		if (!ret)
+			pci_sriov_set_totalvfs(oct->pci_dev,
+					       oct->sriov_info.max_vfs);
+#endif
 		break;
 
 	default:
-		s = "?";
 		dev_err(&oct->pci_dev->dev, "Unknown device found (dev_id: %x)\n",
 			dev_id);
 	}
 
-	if (!ret)
-		dev_info(&oct->pci_dev->dev, "%s PASS%d.%d %s Version: %s\n", s,
-			 OCTEON_MAJOR_REV(oct),
-			 OCTEON_MINOR_REV(oct),
-			 octeon_get_conf(oct)->card_name,
-			 LIQUIDIO_VERSION);
-
 	return ret;
 }
 
 /**
- * \brief PCI initialization for each Octeon device.
- * @param oct octeon device
+ * octeon_pci_os_setup - PCI initialization for each Octeon device.
+ * @oct: octeon device
  */
 static int octeon_pci_os_setup(struct octeon_device *oct)
 {
@@ -1868,46 +1399,9 @@ static int octeon_pci_os_setup(struct octeon_device *oct)
 	return 0;
 }
 
-static inline int skb_iq(struct lio *lio, struct sk_buff *skb)
-{
-	int q = 0;
-
-	if (netif_is_multiqueue(lio->netdev))
-		q = skb->queue_mapping % lio->linfo.num_txpciq;
-
-	return q;
-}
-
-/**
- * \brief Check Tx queue state for a given network buffer
- * @param lio per-network private data
- * @param skb network buffer
- */
-static inline int check_txq_state(struct lio *lio, struct sk_buff *skb)
-{
-	int q = 0, iq = 0;
-
-	if (netif_is_multiqueue(lio->netdev)) {
-		q = skb->queue_mapping;
-		iq = lio->linfo.txpciq[(q % (lio->linfo.num_txpciq))].s.q_no;
-	} else {
-		iq = lio->txq;
-		q = iq;
-	}
-
-	if (octnet_iq_is_full(lio->oct_dev, iq))
-		return 0;
-
-	if (__netif_subqueue_stopped(lio->netdev, q)) {
-		INCR_INSTRQUEUE_PKT_COUNT(lio->oct_dev, iq, tx_restart, 1);
-		wake_q(lio->netdev, q);
-	}
-	return 1;
-}
-
 /**
- * \brief Unmap and free network buffer
- * @param buf buffer
+ * free_netbuf - Unmap and free network buffer
+ * @buf: buffer
  */
 static void free_netbuf(void *buf)
 {
@@ -1922,14 +1416,12 @@ static void free_netbuf(void *buf)
 	dma_unmap_single(&lio->oct_dev->pci_dev->dev, finfo->dptr, skb->len,
 			 DMA_TO_DEVICE);
 
-	check_txq_state(lio, skb);
-
 	tx_buffer_free(skb);
 }
 
 /**
- * \brief Unmap and free gather buffer
- * @param buf buffer
+ * free_netsgbuf - Unmap and free gather buffer
+ * @buf: buffer
  */
 static void free_netsgbuf(void *buf)
 {
@@ -1951,27 +1443,25 @@ static void free_netsgbuf(void *buf)
 
 	i = 1;
 	while (frags--) {
-		struct skb_frag_struct *frag = &skb_shinfo(skb)->frags[i - 1];
+		skb_frag_t *frag = &skb_shinfo(skb)->frags[i - 1];
 
-		pci_unmap_page((lio->oct_dev)->pci_dev,
+		dma_unmap_page(&lio->oct_dev->pci_dev->dev,
 			       g->sg[(i >> 2)].ptr[(i & 3)],
-			       frag->size, DMA_TO_DEVICE);
+			       skb_frag_size(frag), DMA_TO_DEVICE);
 		i++;
 	}
 
-	iq = skb_iq(lio, skb);
+	iq = skb_iq(lio->oct_dev, skb);
 	spin_lock(&lio->glist_lock[iq]);
 	list_add_tail(&g->list, &lio->glist[iq]);
 	spin_unlock(&lio->glist_lock[iq]);
 
-	check_txq_state(lio, skb);     /* mq support: sub-queue state check */
-
 	tx_buffer_free(skb);
 }
 
 /**
- * \brief Unmap and free gather buffer with response
- * @param buf buffer
+ * free_netsgbuf_with_resp - Unmap and free gather buffer with response
+ * @buf: buffer
  */
 static void free_netsgbuf_with_resp(void *buf)
 {
@@ -1996,34 +1486,35 @@ static void free_netsgbuf_with_resp(void *buf)
 
 	i = 1;
 	while (frags--) {
-		struct skb_frag_struct *frag = &skb_shinfo(skb)->frags[i - 1];
+		skb_frag_t *frag = &skb_shinfo(skb)->frags[i - 1];
 
-		pci_unmap_page((lio->oct_dev)->pci_dev,
+		dma_unmap_page(&lio->oct_dev->pci_dev->dev,
 			       g->sg[(i >> 2)].ptr[(i & 3)],
-			       frag->size, DMA_TO_DEVICE);
+			       skb_frag_size(frag), DMA_TO_DEVICE);
 		i++;
 	}
 
-	iq = skb_iq(lio, skb);
+	iq = skb_iq(lio->oct_dev, skb);
 
 	spin_lock(&lio->glist_lock[iq]);
 	list_add_tail(&g->list, &lio->glist[iq]);
 	spin_unlock(&lio->glist_lock[iq]);
 
 	/* Don't free the skb yet */
-
-	check_txq_state(lio, skb);
 }
 
 /**
- * \brief Adjust ptp frequency
- * @param ptp PTP clock info
- * @param ppb how much to adjust by, in parts-per-billion
+ * liquidio_ptp_adjfine - Adjust ptp frequency
+ * @ptp: PTP clock info
+ * @scaled_ppm: how much to adjust by, in scaled parts-per-million
+ *
+ * Scaled parts per million is ppm with a 16-bit binary fractional field.
  */
-static int liquidio_ptp_adjfreq(struct ptp_clock_info *ptp, s32 ppb)
+static int liquidio_ptp_adjfine(struct ptp_clock_info *ptp, long scaled_ppm)
 {
 	struct lio *lio = container_of(ptp, struct lio, ptp_info);
 	struct octeon_device *oct = (struct octeon_device *)lio->oct_dev;
+	s32 ppb = scaled_ppm_to_ppb(scaled_ppm);
 	u64 comp, delta;
 	unsigned long flags;
 	bool neg_adj = false;
@@ -2053,9 +1544,9 @@ static int liquidio_ptp_adjfreq(struct ptp_clock_info *ptp, s32 ppb)
 }
 
 /**
- * \brief Adjust ptp time
- * @param ptp PTP clock info
- * @param delta how much to adjust by, in nanosecs
+ * liquidio_ptp_adjtime - Adjust ptp time
+ * @ptp: PTP clock info
+ * @delta: how much to adjust by, in nanosecs
  */
 static int liquidio_ptp_adjtime(struct ptp_clock_info *ptp, s64 delta)
 {
@@ -2070,9 +1561,9 @@ static int liquidio_ptp_adjtime(struct ptp_clock_info *ptp, s64 delta)
 }
 
 /**
- * \brief Get hardware clock time, including any adjustment
- * @param ptp PTP clock info
- * @param ts timespec
+ * liquidio_ptp_gettime - Get hardware clock time, including any adjustment
+ * @ptp: PTP clock info
+ * @ts: timespec
  */
 static int liquidio_ptp_gettime(struct ptp_clock_info *ptp,
 				struct timespec64 *ts)
@@ -2093,9 +1584,9 @@ static int liquidio_ptp_gettime(struct ptp_clock_info *ptp,
 }
 
 /**
- * \brief Set hardware clock time. Reset adjustment
- * @param ptp PTP clock info
- * @param ts timespec
+ * liquidio_ptp_settime - Set hardware clock time. Reset adjustment
+ * @ptp: PTP clock info
+ * @ts: timespec
  */
 static int liquidio_ptp_settime(struct ptp_clock_info *ptp,
 				const struct timespec64 *ts)
@@ -2105,7 +1596,7 @@ static int liquidio_ptp_settime(struct ptp_clock_info *ptp,
 	struct lio *lio = container_of(ptp, struct lio, ptp_info);
 	struct octeon_device *oct = (struct octeon_device *)lio->oct_dev;
 
-	ns = timespec_to_ns(ts);
+	ns = timespec64_to_ns(ts);
 
 	spin_lock_irqsave(&lio->ptp_lock, flags);
 	lio_pci_writeq(oct, ns, CN6XXX_MIO_PTP_CLOCK_HI);
@@ -2116,22 +1607,22 @@ static int liquidio_ptp_settime(struct ptp_clock_info *ptp,
 }
 
 /**
- * \brief Check if PTP is enabled
- * @param ptp PTP clock info
- * @param rq request
- * @param on is it on
+ * liquidio_ptp_enable - Check if PTP is enabled
+ * @ptp: PTP clock info
+ * @rq: request
+ * @on: is it on
  */
 static int
-liquidio_ptp_enable(struct ptp_clock_info *ptp __attribute__((unused)),
-		    struct ptp_clock_request *rq __attribute__((unused)),
-		    int on __attribute__((unused)))
+liquidio_ptp_enable(struct ptp_clock_info __maybe_unused *ptp,
+		    struct ptp_clock_request __maybe_unused *rq,
+		    int __maybe_unused on)
 {
 	return -EOPNOTSUPP;
 }
 
 /**
- * \brief Open PTP clock source
- * @param netdev network device
+ * oct_ptp_open - Open PTP clock source
+ * @netdev: network device
  */
 static void oct_ptp_open(struct net_device *netdev)
 {
@@ -2147,7 +1638,7 @@ static void oct_ptp_open(struct net_device *netdev)
 	lio->ptp_info.n_ext_ts = 0;
 	lio->ptp_info.n_per_out = 0;
 	lio->ptp_info.pps = 0;
-	lio->ptp_info.adjfreq = liquidio_ptp_adjfreq;
+	lio->ptp_info.adjfine = liquidio_ptp_adjfine;
 	lio->ptp_info.adjtime = liquidio_ptp_adjtime;
 	lio->ptp_info.gettime64 = liquidio_ptp_gettime;
 	lio->ptp_info.settime64 = liquidio_ptp_settime;
@@ -2163,8 +1654,8 @@ static void oct_ptp_open(struct net_device *netdev)
 }
 
 /**
- * \brief Init PTP clock
- * @param oct octeon device
+ * liquidio_ptp_init - Init PTP clock
+ * @oct: octeon device
  */
 static void liquidio_ptp_init(struct octeon_device *oct)
 {
@@ -2180,8 +1671,8 @@ static void liquidio_ptp_init(struct octeon_device *oct)
 }
 
 /**
- * \brief Load firmware to device
- * @param oct octeon device
+ * load_firmware - Load firmware to device
+ * @oct: octeon device
  *
  * Maps device to firmware filename, requests firmware, and downloads it
  */
@@ -2192,15 +1683,12 @@ static int load_firmware(struct octeon_device *oct)
 	char fw_name[LIO_MAX_FW_FILENAME_LEN];
 	char *tmp_fw_type;
 
-	if (fw_type_is_none()) {
-		dev_info(&oct->pci_dev->dev, "Skipping firmware load\n");
-		return ret;
-	}
-
-	if (fw_type[0] == '\0')
+	if (fw_type_is_auto()) {
 		tmp_fw_type = LIO_FW_NAME_TYPE_NIC;
-	else
+		strscpy_pad(fw_type, tmp_fw_type, sizeof(fw_type));
+	} else {
 		tmp_fw_type = fw_type;
+	}
 
 	sprintf(fw_name, "%s%s%s_%s%s", LIO_FW_DIR, LIO_FW_BASE_NAME,
 		octeon_get_conf(oct)->card_name, tmp_fw_type,
@@ -2208,7 +1696,7 @@ static int load_firmware(struct octeon_device *oct)
 
 	ret = request_firmware(&fw, fw_name, &oct->pci_dev->dev);
 	if (ret) {
-		dev_err(&oct->pci_dev->dev, "Request firmware failed. Could not find file %s.\n.",
+		dev_err(&oct->pci_dev->dev, "Request firmware failed. Could not find file %s.\n",
 			fw_name);
 		release_firmware(fw);
 		return ret;
@@ -2222,424 +1710,8 @@ static int load_firmware(struct octeon_device *oct)
 }
 
 /**
- * \brief Setup output queue
- * @param oct octeon device
- * @param q_no which queue
- * @param num_descs how many descriptors
- * @param desc_size size of each descriptor
- * @param app_ctx application context
- */
-static int octeon_setup_droq(struct octeon_device *oct, int q_no, int num_descs,
-			     int desc_size, void *app_ctx)
-{
-	int ret_val = 0;
-
-	dev_dbg(&oct->pci_dev->dev, "Creating Droq: %d\n", q_no);
-	/* droq creation and local register settings. */
-	ret_val = octeon_create_droq(oct, q_no, num_descs, desc_size, app_ctx);
-	if (ret_val < 0)
-		return ret_val;
-
-	if (ret_val == 1) {
-		dev_dbg(&oct->pci_dev->dev, "Using default droq %d\n", q_no);
-		return 0;
-	}
-	/* tasklet creation for the droq */
-
-	/* Enable the droq queues */
-	octeon_set_droq_pkt_op(oct, q_no, 1);
-
-	/* Send Credit for Octeon Output queues. Credits are always
-	 * sent after the output queue is enabled.
-	 */
-	writel(oct->droq[q_no]->max_count,
-	       oct->droq[q_no]->pkts_credit_reg);
-
-	return ret_val;
-}
-
-/**
- * \brief Callback for getting interface configuration
- * @param status status of request
- * @param buf pointer to resp structure
- */
-static void if_cfg_callback(struct octeon_device *oct,
-			    u32 status __attribute__((unused)),
-			    void *buf)
-{
-	struct octeon_soft_command *sc = (struct octeon_soft_command *)buf;
-	struct liquidio_if_cfg_resp *resp;
-	struct liquidio_if_cfg_context *ctx;
-
-	resp = (struct liquidio_if_cfg_resp *)sc->virtrptr;
-	ctx = (struct liquidio_if_cfg_context *)sc->ctxptr;
-
-	oct = lio_get_device(ctx->octeon_id);
-	if (resp->status)
-		dev_err(&oct->pci_dev->dev, "nic if cfg instruction failed. Status: 0x%llx (0x%08x)\n",
-			CVM_CAST64(resp->status), status);
-	WRITE_ONCE(ctx->cond, 1);
-
-	snprintf(oct->fw_info.liquidio_firmware_version, 32, "%s",
-		 resp->cfg_info.liquidio_firmware_version);
-
-	/* This barrier is required to be sure that the response has been
-	 * written fully before waking up the handler
-	 */
-	wmb();
-
-	wake_up_interruptible(&ctx->wc);
-}
-
-/** Routine to push packets arriving on Octeon interface upto network layer.
- * @param oct_id   - octeon device id.
- * @param skbuff   - skbuff struct to be passed to network layer.
- * @param len      - size of total data received.
- * @param rh       - Control header associated with the packet
- * @param param    - additional control data with the packet
- * @param arg	   - farg registered in droq_ops
- */
-static void
-liquidio_push_packet(u32 octeon_id __attribute__((unused)),
-		     void *skbuff,
-		     u32 len,
-		     union octeon_rh *rh,
-		     void *param,
-		     void *arg)
-{
-	struct napi_struct *napi = param;
-	struct sk_buff *skb = (struct sk_buff *)skbuff;
-	struct skb_shared_hwtstamps *shhwtstamps;
-	u64 ns;
-	u16 vtag = 0;
-	u32 r_dh_off;
-	struct net_device *netdev = (struct net_device *)arg;
-	struct octeon_droq *droq = container_of(param, struct octeon_droq,
-						napi);
-	if (netdev) {
-		int packet_was_received;
-		struct lio *lio = GET_LIO(netdev);
-		struct octeon_device *oct = lio->oct_dev;
-
-		/* Do not proceed if the interface is not in RUNNING state. */
-		if (!ifstate_check(lio, LIO_IFSTATE_RUNNING)) {
-			recv_buffer_free(skb);
-			droq->stats.rx_dropped++;
-			return;
-		}
-
-		skb->dev = netdev;
-
-		skb_record_rx_queue(skb, droq->q_no);
-		if (likely(len > MIN_SKB_SIZE)) {
-			struct octeon_skb_page_info *pg_info;
-			unsigned char *va;
-
-			pg_info = ((struct octeon_skb_page_info *)(skb->cb));
-			if (pg_info->page) {
-				/* For Paged allocation use the frags */
-				va = page_address(pg_info->page) +
-					pg_info->page_offset;
-				memcpy(skb->data, va, MIN_SKB_SIZE);
-				skb_put(skb, MIN_SKB_SIZE);
-				skb_add_rx_frag(skb, skb_shinfo(skb)->nr_frags,
-						pg_info->page,
-						pg_info->page_offset +
-						MIN_SKB_SIZE,
-						len - MIN_SKB_SIZE,
-						LIO_RXBUFFER_SZ);
-			}
-		} else {
-			struct octeon_skb_page_info *pg_info =
-				((struct octeon_skb_page_info *)(skb->cb));
-			skb_copy_to_linear_data(skb, page_address(pg_info->page)
-						+ pg_info->page_offset, len);
-			skb_put(skb, len);
-			put_page(pg_info->page);
-		}
-
-		r_dh_off = (rh->r_dh.len - 1) * BYTES_PER_DHLEN_UNIT;
-
-		if (((oct->chip_id == OCTEON_CN66XX) ||
-		     (oct->chip_id == OCTEON_CN68XX)) &&
-		    ptp_enable) {
-			if (rh->r_dh.has_hwtstamp) {
-				/* timestamp is included from the hardware at
-				 * the beginning of the packet.
-				 */
-				if (ifstate_check
-				    (lio, LIO_IFSTATE_RX_TIMESTAMP_ENABLED)) {
-					/* Nanoseconds are in the first 64-bits
-					 * of the packet.
-					 */
-					memcpy(&ns, (skb->data + r_dh_off),
-					       sizeof(ns));
-					r_dh_off -= BYTES_PER_DHLEN_UNIT;
-					shhwtstamps = skb_hwtstamps(skb);
-					shhwtstamps->hwtstamp =
-						ns_to_ktime(ns +
-							    lio->ptp_adjust);
-				}
-			}
-		}
-
-		if (rh->r_dh.has_hash) {
-			__be32 *hash_be = (__be32 *)(skb->data + r_dh_off);
-			u32 hash = be32_to_cpu(*hash_be);
-
-			skb_set_hash(skb, hash, PKT_HASH_TYPE_L4);
-			r_dh_off -= BYTES_PER_DHLEN_UNIT;
-		}
-
-		skb_pull(skb, rh->r_dh.len * BYTES_PER_DHLEN_UNIT);
-
-		skb->protocol = eth_type_trans(skb, skb->dev);
-		if ((netdev->features & NETIF_F_RXCSUM) &&
-		    (((rh->r_dh.encap_on) &&
-		      (rh->r_dh.csum_verified & CNNIC_TUN_CSUM_VERIFIED)) ||
-		     (!(rh->r_dh.encap_on) &&
-		      (rh->r_dh.csum_verified & CNNIC_CSUM_VERIFIED))))
-			/* checksum has already been verified */
-			skb->ip_summed = CHECKSUM_UNNECESSARY;
-		else
-			skb->ip_summed = CHECKSUM_NONE;
-
-		/* Setting Encapsulation field on basis of status received
-		 * from the firmware
-		 */
-		if (rh->r_dh.encap_on) {
-			skb->encapsulation = 1;
-			skb->csum_level = 1;
-			droq->stats.rx_vxlan++;
-		}
-
-		/* inbound VLAN tag */
-		if ((netdev->features & NETIF_F_HW_VLAN_CTAG_RX) &&
-		    (rh->r_dh.vlan != 0)) {
-			u16 vid = rh->r_dh.vlan;
-			u16 priority = rh->r_dh.priority;
-
-			vtag = priority << 13 | vid;
-			__vlan_hwaccel_put_tag(skb, htons(ETH_P_8021Q), vtag);
-		}
-
-		packet_was_received = napi_gro_receive(napi, skb) != GRO_DROP;
-
-		if (packet_was_received) {
-			droq->stats.rx_bytes_received += len;
-			droq->stats.rx_pkts_received++;
-		} else {
-			droq->stats.rx_dropped++;
-			netif_info(lio, rx_err, lio->netdev,
-				   "droq:%d  error rx_dropped:%llu\n",
-				   droq->q_no, droq->stats.rx_dropped);
-		}
-
-	} else {
-		recv_buffer_free(skb);
-	}
-}
-
-/**
- * \brief wrapper for calling napi_schedule
- * @param param parameters to pass to napi_schedule
- *
- * Used when scheduling on different CPUs
- */
-static void napi_schedule_wrapper(void *param)
-{
-	struct napi_struct *napi = param;
-
-	napi_schedule(napi);
-}
-
-/**
- * \brief callback when receive interrupt occurs and we are in NAPI mode
- * @param arg pointer to octeon output queue
- */
-static void liquidio_napi_drv_callback(void *arg)
-{
-	struct octeon_device *oct;
-	struct octeon_droq *droq = arg;
-	int this_cpu = smp_processor_id();
-
-	oct = droq->oct_dev;
-
-	if (OCTEON_CN23XX_PF(oct) || droq->cpu_id == this_cpu) {
-		napi_schedule_irqoff(&droq->napi);
-	} else {
-		struct call_single_data *csd = &droq->csd;
-
-		csd->func = napi_schedule_wrapper;
-		csd->info = &droq->napi;
-		csd->flags = 0;
-
-		smp_call_function_single_async(droq->cpu_id, csd);
-	}
-}
-
-/**
- * \brief Entry point for NAPI polling
- * @param napi NAPI structure
- * @param budget maximum number of items to process
- */
-static int liquidio_napi_poll(struct napi_struct *napi, int budget)
-{
-	struct octeon_droq *droq;
-	int work_done;
-	int tx_done = 0, iq_no;
-	struct octeon_instr_queue *iq;
-	struct octeon_device *oct;
-
-	droq = container_of(napi, struct octeon_droq, napi);
-	oct = droq->oct_dev;
-	iq_no = droq->q_no;
-	/* Handle Droq descriptors */
-	work_done = octeon_process_droq_poll_cmd(oct, droq->q_no,
-						 POLL_EVENT_PROCESS_PKTS,
-						 budget);
-
-	/* Flush the instruction queue */
-	iq = oct->instr_queue[iq_no];
-	if (iq) {
-		if (atomic_read(&iq->instr_pending))
-			/* Process iq buffers with in the budget limits */
-			tx_done = octeon_flush_iq(oct, iq, budget);
-		else
-			tx_done = 1;
-		/* Update iq read-index rather than waiting for next interrupt.
-		 * Return back if tx_done is false.
-		 */
-		update_txq_status(oct, iq_no);
-	} else {
-		dev_err(&oct->pci_dev->dev, "%s:  iq (%d) num invalid\n",
-			__func__, iq_no);
-	}
-
-	/* force enable interrupt if reg cnts are high to avoid wraparound */
-	if ((work_done < budget && tx_done) ||
-	    (iq && iq->pkt_in_done >= MAX_REG_CNT) ||
-	    (droq->pkt_count >= MAX_REG_CNT)) {
-		tx_done = 1;
-		napi_complete_done(napi, work_done);
-		octeon_process_droq_poll_cmd(droq->oct_dev, droq->q_no,
-					     POLL_EVENT_ENABLE_INTR, 0);
-		return 0;
-	}
-
-	return (!tx_done) ? (budget) : (work_done);
-}
-
-/**
- * \brief Setup input and output queues
- * @param octeon_dev octeon device
- * @param ifidx  Interface Index
- *
- * Note: Queues are with respect to the octeon device. Thus
- * an input queue is for egress packets, and output queues
- * are for ingress packets.
- */
-static inline int setup_io_queues(struct octeon_device *octeon_dev,
-				  int ifidx)
-{
-	struct octeon_droq_ops droq_ops;
-	struct net_device *netdev;
-	static int cpu_id;
-	static int cpu_id_modulus;
-	struct octeon_droq *droq;
-	struct napi_struct *napi;
-	int q, q_no, retval = 0;
-	struct lio *lio;
-	int num_tx_descs;
-
-	netdev = octeon_dev->props[ifidx].netdev;
-
-	lio = GET_LIO(netdev);
-
-	memset(&droq_ops, 0, sizeof(struct octeon_droq_ops));
-
-	droq_ops.fptr = liquidio_push_packet;
-	droq_ops.farg = (void *)netdev;
-
-	droq_ops.poll_mode = 1;
-	droq_ops.napi_fn = liquidio_napi_drv_callback;
-	cpu_id = 0;
-	cpu_id_modulus = num_present_cpus();
-
-	/* set up DROQs. */
-	for (q = 0; q < lio->linfo.num_rxpciq; q++) {
-		q_no = lio->linfo.rxpciq[q].s.q_no;
-		dev_dbg(&octeon_dev->pci_dev->dev,
-			"setup_io_queues index:%d linfo.rxpciq.s.q_no:%d\n",
-			q, q_no);
-		retval = octeon_setup_droq(octeon_dev, q_no,
-					   CFG_GET_NUM_RX_DESCS_NIC_IF
-						   (octeon_get_conf(octeon_dev),
-						   lio->ifidx),
-					   CFG_GET_NUM_RX_BUF_SIZE_NIC_IF
-						   (octeon_get_conf(octeon_dev),
-						   lio->ifidx), NULL);
-		if (retval) {
-			dev_err(&octeon_dev->pci_dev->dev,
-				"%s : Runtime DROQ(RxQ) creation failed.\n",
-				__func__);
-			return 1;
-		}
-
-		droq = octeon_dev->droq[q_no];
-		napi = &droq->napi;
-		dev_dbg(&octeon_dev->pci_dev->dev, "netif_napi_add netdev:%llx oct:%llx pf_num:%d\n",
-			(u64)netdev, (u64)octeon_dev, octeon_dev->pf_num);
-		netif_napi_add(netdev, napi, liquidio_napi_poll, 64);
-
-		/* designate a CPU for this droq */
-		droq->cpu_id = cpu_id;
-		cpu_id++;
-		if (cpu_id >= cpu_id_modulus)
-			cpu_id = 0;
-
-		octeon_register_droq_ops(octeon_dev, q_no, &droq_ops);
-	}
-
-	if (OCTEON_CN23XX_PF(octeon_dev)) {
-		/* 23XX PF can receive control messages (via the first PF-owned
-		 * droq) from the firmware even if the ethX interface is down,
-		 * so that's why poll_mode must be off for the first droq.
-		 */
-		octeon_dev->droq[0]->ops.poll_mode = 0;
-	}
-
-	/* set up IQs. */
-	for (q = 0; q < lio->linfo.num_txpciq; q++) {
-		num_tx_descs = CFG_GET_NUM_TX_DESCS_NIC_IF(octeon_get_conf
-							   (octeon_dev),
-							   lio->ifidx);
-		retval = octeon_setup_iq(octeon_dev, ifidx, q,
-					 lio->linfo.txpciq[q], num_tx_descs,
-					 netdev_get_tx_queue(netdev, q));
-		if (retval) {
-			dev_err(&octeon_dev->pci_dev->dev,
-				" %s : Runtime IQ(TxQ) creation failed.\n",
-				__func__);
-			return 1;
-		}
-
-		if (octeon_dev->ioq_vector) {
-			struct octeon_ioq_vector *ioq_vector;
-
-			ioq_vector = &octeon_dev->ioq_vector[q];
-			netif_set_xps_queue(netdev,
-					    &ioq_vector->affinity_mask,
-					    ioq_vector->iq_index);
-		}
-	}
-
-	return 0;
-}
-
-/**
- * \brief Poll routine for checking transmit queue status
- * @param work work_struct data structure
+ * octnet_poll_check_txq_status - Poll routine for checking transmit queue status
+ * @work: work_struct data structure
  */
 static void octnet_poll_check_txq_status(struct work_struct *work)
 {
@@ -2655,8 +1727,8 @@ static void octnet_poll_check_txq_status(struct work_struct *work)
 }
 
 /**
- * \brief Sets up the txq poll check
- * @param netdev network device
+ * setup_tx_poll_fn - Sets up the txq poll check
+ * @netdev: network device
  */
 static inline int setup_tx_poll_fn(struct net_device *netdev)
 {
@@ -2664,7 +1736,7 @@ static inline int setup_tx_poll_fn(struct net_device *netdev)
 	struct octeon_device *oct = lio->oct_dev;
 
 	lio->txq_status_wq.wq = alloc_workqueue("txq-status",
-						WQ_MEM_RECLAIM, 0);
+						WQ_MEM_RECLAIM | WQ_PERCPU, 0);
 	if (!lio->txq_status_wq.wq) {
 		dev_err(&oct->pci_dev->dev, "unable to create cavium txq status wq\n");
 		return -1;
@@ -2688,16 +1760,20 @@ static inline void cleanup_tx_poll_fn(struct net_device *netdev)
 }
 
 /**
- * \brief Net device open for LiquidIO
- * @param netdev network device
+ * liquidio_open - Net device open for LiquidIO
+ * @netdev: network device
  */
 static int liquidio_open(struct net_device *netdev)
 {
 	struct lio *lio = GET_LIO(netdev);
 	struct octeon_device *oct = lio->oct_dev;
+	struct octeon_device_priv *oct_priv = oct->priv;
 	struct napi_struct *napi, *n;
+	int ret = 0;
 
 	if (oct->props[lio->ifidx].napi_enabled == 0) {
+		tasklet_disable(&oct_priv->droq_tasklet);
+
 		list_for_each_entry_safe(napi, n, &netdev->napi_list, dev_list)
 			napi_enable(napi);
 
@@ -2707,58 +1783,92 @@ static int liquidio_open(struct net_device *netdev)
 			oct->droq[0]->ops.poll_mode = 1;
 	}
 
-	if ((oct->chip_id == OCTEON_CN66XX || oct->chip_id == OCTEON_CN68XX) &&
-	    ptp_enable)
+	if (oct->ptp_enable)
 		oct_ptp_open(netdev);
 
 	ifstate_set(lio, LIO_IFSTATE_RUNNING);
 
+	if (!OCTEON_CN23XX_PF(oct) || !oct->msix_on) {
+		ret = setup_tx_poll_fn(netdev);
+		if (ret)
+			goto err_poll;
+	}
+
+	netif_tx_start_all_queues(netdev);
+
 	/* Ready for link status updates */
 	lio->intf_open = 1;
 
 	netif_info(lio, ifup, lio->netdev, "Interface Open, ready for traffic\n");
 
-	if (OCTEON_CN23XX_PF(oct)) {
-		if (!oct->msix_on)
-			if (setup_tx_poll_fn(netdev))
-				return -1;
-	} else {
-		if (setup_tx_poll_fn(netdev))
-			return -1;
-	}
-
-	start_txq(netdev);
-
 	/* tell Octeon to start forwarding packets to host */
-	send_rx_ctrl_cmd(lio, 1);
+	ret = send_rx_ctrl_cmd(lio, 1);
+	if (ret)
+		goto err_rx_ctrl;
+
+	/* start periodical statistics fetch */
+	INIT_DELAYED_WORK(&lio->stats_wk.work, lio_fetch_stats);
+	lio->stats_wk.ctxptr = lio;
+	schedule_delayed_work(&lio->stats_wk.work, msecs_to_jiffies
+					(LIQUIDIO_NDEV_STATS_POLL_TIME_MS));
 
 	dev_info(&oct->pci_dev->dev, "%s interface is opened\n",
 		 netdev->name);
 
 	return 0;
+
+err_rx_ctrl:
+	if (!OCTEON_CN23XX_PF(oct) || !oct->msix_on)
+		cleanup_tx_poll_fn(netdev);
+err_poll:
+	if (lio->ptp_clock) {
+		ptp_clock_unregister(lio->ptp_clock);
+		lio->ptp_clock = NULL;
+	}
+
+	if (oct->props[lio->ifidx].napi_enabled == 1) {
+		list_for_each_entry_safe(napi, n, &netdev->napi_list, dev_list)
+			napi_disable(napi);
+
+		oct->props[lio->ifidx].napi_enabled = 0;
+
+		if (OCTEON_CN23XX_PF(oct))
+			oct->droq[0]->ops.poll_mode = 0;
+	}
+
+	return ret;
 }
 
 /**
- * \brief Net device stop for LiquidIO
- * @param netdev network device
+ * liquidio_stop - Net device stop for LiquidIO
+ * @netdev: network device
  */
 static int liquidio_stop(struct net_device *netdev)
 {
 	struct lio *lio = GET_LIO(netdev);
 	struct octeon_device *oct = lio->oct_dev;
+	struct octeon_device_priv *oct_priv = oct->priv;
+	struct napi_struct *napi, *n;
+	int ret = 0;
 
 	ifstate_reset(lio, LIO_IFSTATE_RUNNING);
 
-	netif_tx_disable(netdev);
+	/* Stop any link updates */
+	lio->intf_open = 0;
+
+	stop_txqs(netdev);
 
 	/* Inform that netif carrier is down */
 	netif_carrier_off(netdev);
-	lio->intf_open = 0;
+	netif_tx_disable(netdev);
+
 	lio->linfo.link.s.link_up = 0;
 	lio->link_changes++;
 
 	/* Tell Octeon that nic interface is down. */
-	send_rx_ctrl_cmd(lio, 0);
+	ret = send_rx_ctrl_cmd(lio, 0);
+	if (ret)
+		return ret;
 
 	if (OCTEON_CN23XX_PF(oct)) {
 		if (!oct->msix_on)
@@ -2767,19 +1877,38 @@ static int liquidio_stop(struct net_device *netdev)
 		cleanup_tx_poll_fn(netdev);
 	}
 
+	cancel_delayed_work_sync(&lio->stats_wk.work);
+
 	if (lio->ptp_clock) {
 		ptp_clock_unregister(lio->ptp_clock);
 		lio->ptp_clock = NULL;
 	}
 
+	/* Wait for any pending Rx descriptors */
+	if (lio_wait_for_clean_oq(oct))
+		netif_info(lio, rx_err, lio->netdev,
+			   "Proceeding with stop interface after partial RX desc processing\n");
+
+	if (oct->props[lio->ifidx].napi_enabled == 1) {
+		list_for_each_entry_safe(napi, n, &netdev->napi_list, dev_list)
+			napi_disable(napi);
+
+		oct->props[lio->ifidx].napi_enabled = 0;
+
+		if (OCTEON_CN23XX_PF(oct))
+			oct->droq[0]->ops.poll_mode = 0;
+
+		tasklet_enable(&oct_priv->droq_tasklet);
+	}
+
 	dev_info(&oct->pci_dev->dev, "%s interface is stopped\n", netdev->name);
 
-	return 0;
+	return ret;
 }
 
 /**
- * \brief Converts a mask based on net device flags
- * @param netdev network device
+ * get_new_flags - Converts a mask based on net device flags
+ * @netdev: network device
  *
  * This routine generates a octnet_ifflags mask from the net device flags
  * received from the OS.
@@ -2811,8 +1940,8 @@ static inline enum octnet_ifflags get_new_flags(struct net_device *netdev)
 }
 
 /**
- * \brief Net device set_multicast_list
- * @param netdev network device
+ * liquidio_set_mcast_list - Net device set_multicast_list
+ * @netdev: network device
  */
 static void liquidio_set_mcast_list(struct net_device *netdev)
 {
@@ -2850,18 +1979,18 @@ static void liquidio_set_mcast_list(struct net_device *netdev)
 	/* Apparently, any activity in this call from the kernel has to
 	 * be atomic. So we won't wait for response.
 	 */
-	nctrl.wait_time = 0;
 
 	ret = octnet_send_nic_ctrl_pkt(lio->oct_dev, &nctrl);
-	if (ret < 0) {
+	if (ret) {
 		dev_err(&oct->pci_dev->dev, "DEVFLAGS change failed in core (ret: 0x%x)\n",
 			ret);
 	}
 }
 
 /**
- * \brief Net device set_mac_address
- * @param netdev network device
+ * liquidio_set_mac - Net device set_mac_address
+ * @netdev: network device
+ * @p: pointer to sockaddr
  */
 static int liquidio_set_mac(struct net_device *netdev, void *p)
 {
@@ -2882,8 +2011,6 @@ static int liquidio_set_mac(struct net_device *netdev, void *p)
 	nctrl.ncmd.s.more = 1;
 	nctrl.iq_no = lio->linfo.txpciq[0].s.q_no;
 	nctrl.netpndev = (u64)netdev;
-	nctrl.cb_fn = liquidio_link_ctrl_cmd_completion;
-	nctrl.wait_time = 100;
 
 	nctrl.udd[0] = 0;
 	/* The MAC Address is presented in network byte order. */
@@ -2894,20 +2021,25 @@ static int liquidio_set_mac(struct net_device *netdev, void *p)
 		dev_err(&oct->pci_dev->dev, "MAC Address change failed\n");
 		return -ENOMEM;
 	}
-	memcpy(netdev->dev_addr, addr->sa_data, netdev->addr_len);
+
+	if (nctrl.sc_status) {
+		dev_err(&oct->pci_dev->dev,
+			"%s: MAC Address change failed. sc return=%x\n",
+			 __func__, nctrl.sc_status);
+		return -EIO;
+	}
+
+	eth_hw_addr_set(netdev, addr->sa_data);
 	memcpy(((u8 *)&lio->linfo.hw_addr) + 2, addr->sa_data, ETH_ALEN);
 
 	return 0;
 }
 
-/**
- * \brief Net device get_stats
- * @param netdev network device
- */
-static struct net_device_stats *liquidio_get_stats(struct net_device *netdev)
+static void
+liquidio_get_stats64(struct net_device *netdev,
+		     struct rtnl_link_stats64 *lstats)
 {
 	struct lio *lio = GET_LIO(netdev);
-	struct net_device_stats *stats = &netdev->stats;
 	struct octeon_device *oct;
 	u64 pkts = 0, drop = 0, bytes = 0;
 	struct oct_droq_stats *oq_stats;
@@ -2916,7 +2048,10 @@ static struct net_device_stats *liquidio_get_stats(struct net_device *netdev)
 
 	oct = lio->oct_dev;
 
-	for (i = 0; i < lio->linfo.num_txpciq; i++) {
+	if (ifstate_check(lio, LIO_IFSTATE_RESETTING))
+		return;
+
+	for (i = 0; i < oct->num_iqs; i++) {
 		iq_no = lio->linfo.txpciq[i].s.q_no;
 		iq_stats = &oct->instr_queue[iq_no]->stats;
 		pkts += iq_stats->tx_done;
@@ -2924,15 +2059,15 @@ static struct net_device_stats *liquidio_get_stats(struct net_device *netdev)
 		bytes += iq_stats->tx_tot_bytes;
 	}
 
-	stats->tx_packets = pkts;
-	stats->tx_bytes = bytes;
-	stats->tx_dropped = drop;
+	lstats->tx_packets = pkts;
+	lstats->tx_bytes = bytes;
+	lstats->tx_dropped = drop;
 
 	pkts = 0;
 	drop = 0;
 	bytes = 0;
 
-	for (i = 0; i < lio->linfo.num_rxpciq; i++) {
+	for (i = 0; i < oct->num_oqs; i++) {
 		oq_no = lio->linfo.rxpciq[i].s.q_no;
 		oq_stats = &oct->droq[oq_no]->stats;
 		pkts += oq_stats->rx_pkts_received;
@@ -2943,50 +2078,39 @@ static struct net_device_stats *liquidio_get_stats(struct net_device *netdev)
 		bytes += oq_stats->rx_bytes_received;
 	}
 
-	stats->rx_bytes = bytes;
-	stats->rx_packets = pkts;
-	stats->rx_dropped = drop;
-
-	return stats;
-}
-
-/**
- * \brief Net device change_mtu
- * @param netdev network device
- */
-static int liquidio_change_mtu(struct net_device *netdev, int new_mtu)
-{
-	struct lio *lio = GET_LIO(netdev);
-	struct octeon_device *oct = lio->oct_dev;
-	struct octnic_ctrl_pkt nctrl;
-	int ret = 0;
-
-	memset(&nctrl, 0, sizeof(struct octnic_ctrl_pkt));
-
-	nctrl.ncmd.u64 = 0;
-	nctrl.ncmd.s.cmd = OCTNET_CMD_CHANGE_MTU;
-	nctrl.ncmd.s.param1 = new_mtu;
-	nctrl.iq_no = lio->linfo.txpciq[0].s.q_no;
-	nctrl.wait_time = 100;
-	nctrl.netpndev = (u64)netdev;
-	nctrl.cb_fn = liquidio_link_ctrl_cmd_completion;
-
-	ret = octnet_send_nic_ctrl_pkt(lio->oct_dev, &nctrl);
-	if (ret < 0) {
-		dev_err(&oct->pci_dev->dev, "Failed to set MTU\n");
-		return -1;
-	}
-
-	lio->mtu = new_mtu;
-
-	return 0;
+	lstats->rx_bytes = bytes;
+	lstats->rx_packets = pkts;
+	lstats->rx_dropped = drop;
+
+	lstats->multicast = oct->link_stats.fromwire.fw_total_mcast;
+	lstats->collisions = oct->link_stats.fromhost.total_collisions;
+
+	/* detailed rx_errors: */
+	lstats->rx_length_errors = oct->link_stats.fromwire.l2_err;
+	/* recved pkt with crc error    */
+	lstats->rx_crc_errors = oct->link_stats.fromwire.fcs_err;
+	/* recv'd frame alignment error */
+	lstats->rx_frame_errors = oct->link_stats.fromwire.frame_err;
+	/* recv'r fifo overrun */
+	lstats->rx_fifo_errors = oct->link_stats.fromwire.fifo_err;
+
+	lstats->rx_errors = lstats->rx_length_errors + lstats->rx_crc_errors +
+		lstats->rx_frame_errors + lstats->rx_fifo_errors;
+
+	/* detailed tx_errors */
+	lstats->tx_aborted_errors = oct->link_stats.fromhost.fw_err_pko;
+	lstats->tx_carrier_errors = oct->link_stats.fromhost.fw_err_link;
+	lstats->tx_fifo_errors = oct->link_stats.fromhost.fifo_err;
+
+	lstats->tx_errors = lstats->tx_aborted_errors +
+		lstats->tx_carrier_errors +
+		lstats->tx_fifo_errors;
 }
 
 /**
- * \brief Handler for SIOCSHWTSTAMP ioctl
- * @param netdev network device
- * @param ifr interface request
- * @param cmd command
+ * hwtstamp_ioctl - Handler for SIOCSHWTSTAMP ioctl
+ * @netdev: network device
+ * @ifr: interface request
  */
 static int hwtstamp_ioctl(struct net_device *netdev, struct ifreq *ifr)
 {
@@ -2996,9 +2120,6 @@ static int hwtstamp_ioctl(struct net_device *netdev, struct ifreq *ifr)
 	if (copy_from_user(&conf, ifr->ifr_data, sizeof(conf)))
 		return -EFAULT;
 
-	if (conf.flags)
-		return -EINVAL;
-
 	switch (conf.tx_type) {
 	case HWTSTAMP_TX_ON:
 	case HWTSTAMP_TX_OFF:
@@ -3041,10 +2162,10 @@ static int hwtstamp_ioctl(struct net_device *netdev, struct ifreq *ifr)
 }
 
 /**
- * \brief ioctl handler
- * @param netdev network device
- * @param ifr interface request
- * @param cmd command
+ * liquidio_ioctl - ioctl handler
+ * @netdev: network device
+ * @ifr: interface request
+ * @cmd: command
  */
 static int liquidio_ioctl(struct net_device *netdev, struct ifreq *ifr, int cmd)
 {
@@ -3052,18 +2173,19 @@ static int liquidio_ioctl(struct net_device *netdev, struct ifreq *ifr, int cmd)
 
 	switch (cmd) {
 	case SIOCSHWTSTAMP:
-		if ((lio->oct_dev->chip_id == OCTEON_CN66XX ||
-		     lio->oct_dev->chip_id == OCTEON_CN68XX) && ptp_enable)
+		if (lio->oct_dev->ptp_enable)
 			return hwtstamp_ioctl(netdev, ifr);
+		fallthrough;
 	default:
 		return -EOPNOTSUPP;
 	}
 }
 
 /**
- * \brief handle a Tx timestamp response
- * @param status response status
- * @param buf pointer to skb
+ * handle_timestamp - handle a Tx timestamp response
+ * @oct: octeon device
+ * @status: response status
+ * @buf: pointer to skb
  */
 static void handle_timestamp(struct octeon_device *oct,
 			     u32 status,
@@ -3104,14 +2226,17 @@ static void handle_timestamp(struct octeon_device *oct,
 	tx_buffer_free(skb);
 }
 
-/* \brief Send a data packet that will be timestamped
- * @param oct octeon device
- * @param ndata pointer to network data
- * @param finfo pointer to private network data
+/**
+ * send_nic_timestamp_pkt - Send a data packet that will be timestamped
+ * @oct: octeon device
+ * @ndata: pointer to network data
+ * @finfo: pointer to private network data
+ * @xmit_more: more is coming
  */
 static inline int send_nic_timestamp_pkt(struct octeon_device *oct,
 					 struct octnic_data_pkt *ndata,
-					 struct octnet_buf_free_info *finfo)
+					 struct octnet_buf_free_info *finfo,
+					 int xmit_more)
 {
 	int retval;
 	struct octeon_soft_command *sc;
@@ -3146,7 +2271,7 @@ static inline int send_nic_timestamp_pkt(struct octeon_device *oct,
 		len = (u32)((struct octeon_instr_ih2 *)
 			    (&sc->cmd.cmd2.ih2))->dlengsz;
 
-	ring_doorbell = 1;
+	ring_doorbell = !xmit_more;
 
 	retval = octeon_send_command(oct, sc->iq_no, ring_doorbell, &sc->cmd,
 				     sc, len, ndata->reqtype);
@@ -3162,13 +2287,15 @@ static inline int send_nic_timestamp_pkt(struct octeon_device *oct,
 	return retval;
 }
 
-/** \brief Transmit networks packets to the Octeon interface
- * @param skbuff   skbuff struct to be passed to network layer.
- * @param netdev    pointer to network device
- * @returns whether the packet was transmitted to the device okay or not
+/**
+ * liquidio_xmit - Transmit networks packets to the Octeon interface
+ * @skb: skbuff struct to be passed to network layer.
+ * @netdev: pointer to network device
+ *
+ * Return: whether the packet was transmitted to the device okay or not
  *             (NETDEV_TX_OK or NETDEV_TX_BUSY)
  */
-static int liquidio_xmit(struct sk_buff *skb, struct net_device *netdev)
+static netdev_tx_t liquidio_xmit(struct sk_buff *skb, struct net_device *netdev)
 {
 	struct lio *lio;
 	struct octnet_buf_free_info *finfo;
@@ -3180,21 +2307,16 @@ static int liquidio_xmit(struct sk_buff *skb, struct net_device *netdev)
 	union tx_info *tx_info;
 	int status = 0;
 	int q_idx = 0, iq_no = 0;
-	int j;
+	int j, xmit_more = 0;
 	u64 dptr = 0;
 	u32 tag = 0;
 
 	lio = GET_LIO(netdev);
 	oct = lio->oct_dev;
 
-	if (netif_is_multiqueue(netdev)) {
-		q_idx = skb->queue_mapping;
-		q_idx = (q_idx % (lio->linfo.num_txpciq));
-		tag = q_idx;
-		iq_no = lio->linfo.txpciq[q_idx].s.q_no;
-	} else {
-		iq_no = lio->txq;
-	}
+	q_idx = skb_iq(oct, skb);
+	tag = q_idx;
+	iq_no = lio->linfo.txpciq[q_idx].s.q_no;
 
 	stats = &oct->instr_queue[iq_no]->stats;
 
@@ -3225,23 +2347,14 @@ static int liquidio_xmit(struct sk_buff *skb, struct net_device *netdev)
 
 	ndata.q_no = iq_no;
 
-	if (netif_is_multiqueue(netdev)) {
-		if (octnet_iq_is_full(oct, ndata.q_no)) {
-			/* defer sending if queue is full */
-			netif_info(lio, tx_err, lio->netdev, "Transmit failed iq:%d full\n",
-				   ndata.q_no);
-			stats->tx_iq_busy++;
-			return NETDEV_TX_BUSY;
-		}
-	} else {
-		if (octnet_iq_is_full(oct, lio->txq)) {
-			/* defer sending if queue is full */
-			stats->tx_iq_busy++;
-			netif_info(lio, tx_err, lio->netdev, "Transmit failed iq:%d full\n",
-				   lio->txq);
-			return NETDEV_TX_BUSY;
-		}
+	if (octnet_iq_is_full(oct, ndata.q_no)) {
+		/* defer sending if queue is full */
+		netif_info(lio, tx_err, lio->netdev, "Transmit failed iq:%d full\n",
+			   ndata.q_no);
+		stats->tx_iq_busy++;
+		return NETDEV_TX_BUSY;
 	}
+
 	/* pr_info(" XMIT - valid Qs: %d, 1st Q no: %d, cpu:  %d, q_no:%d\n",
 	 *	lio->linfo.num_txpciq, lio->txq, cpu, ndata.q_no);
 	 */
@@ -3276,6 +2389,7 @@ static int liquidio_xmit(struct sk_buff *skb, struct net_device *netdev)
 		if (dma_mapping_error(&oct->pci_dev->dev, dptr)) {
 			dev_err(&oct->pci_dev->dev, "%s DMA mapping error 1\n",
 				__func__);
+			stats->tx_dmamap_fail++;
 			return NETDEV_TX_BUSY;
 		}
 
@@ -3288,12 +2402,12 @@ static int liquidio_xmit(struct sk_buff *skb, struct net_device *netdev)
 
 	} else {
 		int i, frags;
-		struct skb_frag_struct *frag;
+		skb_frag_t *frag;
 		struct octnic_gather *g;
 
 		spin_lock(&lio->glist_lock[q_idx]);
 		g = (struct octnic_gather *)
-			list_delete_head(&lio->glist[q_idx]);
+			lio_list_delete_head(&lio->glist[q_idx]);
 		spin_unlock(&lio->glist_lock[q_idx]);
 
 		if (!g) {
@@ -3315,6 +2429,7 @@ static int liquidio_xmit(struct sk_buff *skb, struct net_device *netdev)
 		if (dma_mapping_error(&oct->pci_dev->dev, g->sg[0].ptr[0])) {
 			dev_err(&oct->pci_dev->dev, "%s DMA mapping error 2\n",
 				__func__);
+			stats->tx_dmamap_fail++;
 			return NETDEV_TX_BUSY;
 		}
 		add_sg_size(&g->sg[0], (skb->len - skb->data_len), 0);
@@ -3325,11 +2440,9 @@ static int liquidio_xmit(struct sk_buff *skb, struct net_device *netdev)
 			frag = &skb_shinfo(skb)->frags[i - 1];
 
 			g->sg[(i >> 2)].ptr[(i & 3)] =
-				dma_map_page(&oct->pci_dev->dev,
-					     frag->page.p,
-					     frag->page_offset,
-					     frag->size,
-					     DMA_TO_DEVICE);
+				skb_frag_dma_map(&oct->pci_dev->dev,
+					         frag, 0, skb_frag_size(frag),
+						 DMA_TO_DEVICE);
 
 			if (dma_mapping_error(&oct->pci_dev->dev,
 					      g->sg[i >> 2].ptr[i & 3])) {
@@ -3341,7 +2454,7 @@ static int liquidio_xmit(struct sk_buff *skb, struct net_device *netdev)
 					frag = &skb_shinfo(skb)->frags[j - 1];
 					dma_unmap_page(&oct->pci_dev->dev,
 						       g->sg[j >> 2].ptr[j & 3],
-						       frag->size,
+						       skb_frag_size(frag),
 						       DMA_TO_DEVICE);
 				}
 				dev_err(&oct->pci_dev->dev, "%s DMA mapping error 3\n",
@@ -3349,7 +2462,8 @@ static int liquidio_xmit(struct sk_buff *skb, struct net_device *netdev)
 				return NETDEV_TX_BUSY;
 			}
 
-			add_sg_size(&g->sg[(i >> 2)], frag->size, (i & 3));
+			add_sg_size(&g->sg[(i >> 2)], skb_frag_size(frag),
+				    (i & 3));
 			i++;
 		}
 
@@ -3385,17 +2499,19 @@ static int liquidio_xmit(struct sk_buff *skb, struct net_device *netdev)
 		irh->vlan = skb_vlan_tag_get(skb) & 0xfff;
 	}
 
+	xmit_more = netdev_xmit_more();
+
 	if (unlikely(cmdsetup.s.timestamp))
-		status = send_nic_timestamp_pkt(oct, &ndata, finfo);
+		status = send_nic_timestamp_pkt(oct, &ndata, finfo, xmit_more);
 	else
-		status = octnet_send_nic_data_pkt(oct, &ndata);
+		status = octnet_send_nic_data_pkt(oct, &ndata, xmit_more);
 	if (status == IQ_SEND_FAILED)
 		goto lio_xmit_failed;
 
 	netif_info(lio, tx_queued, lio->netdev, "Transmit queued successfully\n");
 
 	if (status == IQ_SEND_STOP)
-		stop_q(lio->netdev, q_idx);
+		netif_stop_subqueue(netdev, q_idx);
 
 	netif_trans_update(netdev);
 
@@ -3414,14 +2530,19 @@ lio_xmit_failed:
 	if (dptr)
 		dma_unmap_single(&oct->pci_dev->dev, dptr,
 				 ndata.datasize, DMA_TO_DEVICE);
+
+	octeon_ring_doorbell_locked(oct, iq_no);
+
 	tx_buffer_free(skb);
 	return NETDEV_TX_OK;
 }
 
-/** \brief Network device Tx timeout
- * @param netdev    pointer to network device
+/**
+ * liquidio_tx_timeout - Network device Tx timeout
+ * @netdev:    pointer to network device
+ * @txqueue: index of the hung transmit queue
  */
-static void liquidio_tx_timeout(struct net_device *netdev)
+static void liquidio_tx_timeout(struct net_device *netdev, unsigned int txqueue)
 {
 	struct lio *lio;
 
@@ -3431,7 +2552,7 @@ static void liquidio_tx_timeout(struct net_device *netdev)
 		   "Transmit timeout tx_dropped:%ld, waking up queues now!!\n",
 		   netdev->stats.tx_dropped);
 	netif_trans_update(netdev);
-	txqs_wake(netdev);
+	wake_txqs(netdev);
 }
 
 static int liquidio_vlan_rx_add_vid(struct net_device *netdev,
@@ -3449,14 +2570,15 @@ static int liquidio_vlan_rx_add_vid(struct net_device *netdev,
 	nctrl.ncmd.s.cmd = OCTNET_CMD_ADD_VLAN_FILTER;
 	nctrl.ncmd.s.param1 = vid;
 	nctrl.iq_no = lio->linfo.txpciq[0].s.q_no;
-	nctrl.wait_time = 100;
 	nctrl.netpndev = (u64)netdev;
 	nctrl.cb_fn = liquidio_link_ctrl_cmd_completion;
 
 	ret = octnet_send_nic_ctrl_pkt(lio->oct_dev, &nctrl);
-	if (ret < 0) {
+	if (ret) {
 		dev_err(&oct->pci_dev->dev, "Add VLAN filter failed in core (ret: 0x%x)\n",
 			ret);
+		if (ret > 0)
+			ret = -EIO;
 	}
 
 	return ret;
@@ -3477,24 +2599,25 @@ static int liquidio_vlan_rx_kill_vid(struct net_device *netdev,
 	nctrl.ncmd.s.cmd = OCTNET_CMD_DEL_VLAN_FILTER;
 	nctrl.ncmd.s.param1 = vid;
 	nctrl.iq_no = lio->linfo.txpciq[0].s.q_no;
-	nctrl.wait_time = 100;
 	nctrl.netpndev = (u64)netdev;
 	nctrl.cb_fn = liquidio_link_ctrl_cmd_completion;
 
 	ret = octnet_send_nic_ctrl_pkt(lio->oct_dev, &nctrl);
-	if (ret < 0) {
-		dev_err(&oct->pci_dev->dev, "Add VLAN filter failed in core (ret: 0x%x)\n",
+	if (ret) {
+		dev_err(&oct->pci_dev->dev, "Del VLAN filter failed in core (ret: 0x%x)\n",
 			ret);
+		if (ret > 0)
+			ret = -EIO;
 	}
 	return ret;
 }
 
-/** Sending command to enable/disable RX checksum offload
- * @param netdev                pointer to network device
- * @param command               OCTNET_CMD_TNL_RX_CSUM_CTL
- * @param rx_cmd_bit            OCTNET_CMD_RXCSUM_ENABLE/
- *                              OCTNET_CMD_RXCSUM_DISABLE
- * @returns                     SUCCESS or FAILURE
+/**
+ * liquidio_set_rxcsum_command - Sending command to enable/disable RX checksum offload
+ * @netdev:                pointer to network device
+ * @command:               OCTNET_CMD_TNL_RX_CSUM_CTL
+ * @rx_cmd:                OCTNET_CMD_RXCSUM_ENABLE/OCTNET_CMD_RXCSUM_DISABLE
+ * Returns:                SUCCESS or FAILURE
  */
 static int liquidio_set_rxcsum_command(struct net_device *netdev, int command,
 				       u8 rx_cmd)
@@ -3510,26 +2633,28 @@ static int liquidio_set_rxcsum_command(struct net_device *netdev, int command,
 	nctrl.ncmd.s.cmd = command;
 	nctrl.ncmd.s.param1 = rx_cmd;
 	nctrl.iq_no = lio->linfo.txpciq[0].s.q_no;
-	nctrl.wait_time = 100;
 	nctrl.netpndev = (u64)netdev;
 	nctrl.cb_fn = liquidio_link_ctrl_cmd_completion;
 
 	ret = octnet_send_nic_ctrl_pkt(lio->oct_dev, &nctrl);
-	if (ret < 0) {
+	if (ret) {
 		dev_err(&oct->pci_dev->dev,
 			"DEVFLAGS RXCSUM change failed in core(ret:0x%x)\n",
 			ret);
+		if (ret > 0)
+			ret = -EIO;
 	}
 	return ret;
 }
 
-/** Sending command to add/delete VxLAN UDP port to firmware
- * @param netdev                pointer to network device
- * @param command               OCTNET_CMD_VXLAN_PORT_CONFIG
- * @param vxlan_port            VxLAN port to be added or deleted
- * @param vxlan_cmd_bit         OCTNET_CMD_VXLAN_PORT_ADD,
+/**
+ * liquidio_vxlan_port_command - Sending command to add/delete VxLAN UDP port to firmware
+ * @netdev:                pointer to network device
+ * @command:               OCTNET_CMD_VXLAN_PORT_CONFIG
+ * @vxlan_port:            VxLAN port to be added or deleted
+ * @vxlan_cmd_bit:         OCTNET_CMD_VXLAN_PORT_ADD,
  *                              OCTNET_CMD_VXLAN_PORT_DEL
- * @returns                     SUCCESS or FAILURE
+ * Return:                     SUCCESS or FAILURE
  */
 static int liquidio_vxlan_port_command(struct net_device *netdev, int command,
 				       u16 vxlan_port, u8 vxlan_cmd_bit)
@@ -3546,23 +2671,54 @@ static int liquidio_vxlan_port_command(struct net_device *netdev, int command,
 	nctrl.ncmd.s.more = vxlan_cmd_bit;
 	nctrl.ncmd.s.param1 = vxlan_port;
 	nctrl.iq_no = lio->linfo.txpciq[0].s.q_no;
-	nctrl.wait_time = 100;
 	nctrl.netpndev = (u64)netdev;
 	nctrl.cb_fn = liquidio_link_ctrl_cmd_completion;
 
 	ret = octnet_send_nic_ctrl_pkt(lio->oct_dev, &nctrl);
-	if (ret < 0) {
+	if (ret) {
 		dev_err(&oct->pci_dev->dev,
 			"VxLAN port add/delete failed in core (ret:0x%x)\n",
 			ret);
+		if (ret > 0)
+			ret = -EIO;
 	}
 	return ret;
 }
 
-/** \brief Net device fix features
- * @param netdev  pointer to network device
- * @param request features requested
- * @returns updated features list
+static int liquidio_udp_tunnel_set_port(struct net_device *netdev,
+					unsigned int table, unsigned int entry,
+					struct udp_tunnel_info *ti)
+{
+	return liquidio_vxlan_port_command(netdev,
+					   OCTNET_CMD_VXLAN_PORT_CONFIG,
+					   htons(ti->port),
+					   OCTNET_CMD_VXLAN_PORT_ADD);
+}
+
+static int liquidio_udp_tunnel_unset_port(struct net_device *netdev,
+					  unsigned int table,
+					  unsigned int entry,
+					  struct udp_tunnel_info *ti)
+{
+	return liquidio_vxlan_port_command(netdev,
+					   OCTNET_CMD_VXLAN_PORT_CONFIG,
+					   htons(ti->port),
+					   OCTNET_CMD_VXLAN_PORT_DEL);
+}
+
+static const struct udp_tunnel_nic_info liquidio_udp_tunnels = {
+	.set_port	= liquidio_udp_tunnel_set_port,
+	.unset_port	= liquidio_udp_tunnel_unset_port,
+	.tables		= {
+		{ .n_entries = 1024, .tunnel_types = UDP_TUNNEL_TYPE_VXLAN, },
+	},
+};
+
+/**
+ * liquidio_fix_features - Net device fix features
+ * @netdev:  pointer to network device
+ * @request: features requested
+ * Return: updated features list
  */
 static netdev_features_t liquidio_fix_features(struct net_device *netdev,
 					       netdev_features_t request)
@@ -3591,26 +2747,31 @@ static netdev_features_t liquidio_fix_features(struct net_device *netdev,
 	    (lio->dev_capability & NETIF_F_LRO))
 		request &= ~NETIF_F_LRO;
 
+	if ((request & NETIF_F_HW_VLAN_CTAG_FILTER) &&
+	    !(lio->dev_capability & NETIF_F_HW_VLAN_CTAG_FILTER))
+		request &= ~NETIF_F_HW_VLAN_CTAG_FILTER;
+
 	return request;
 }
 
-/** \brief Net device set features
- * @param netdev  pointer to network device
- * @param features features to enable/disable
+/**
+ * liquidio_set_features - Net device set features
+ * @netdev:  pointer to network device
+ * @features: features to enable/disable
  */
 static int liquidio_set_features(struct net_device *netdev,
 				 netdev_features_t features)
 {
 	struct lio *lio = netdev_priv(netdev);
 
-	if (!((netdev->features ^ features) & NETIF_F_LRO))
-		return 0;
-
-	if ((features & NETIF_F_LRO) && (lio->dev_capability & NETIF_F_LRO))
+	if ((features & NETIF_F_LRO) &&
+	    (lio->dev_capability & NETIF_F_LRO) &&
+	    !(netdev->features & NETIF_F_LRO))
 		liquidio_set_feature(netdev, OCTNET_CMD_LRO_ENABLE,
 				     OCTNIC_LROIPV4 | OCTNIC_LROIPV6);
 	else if (!(features & NETIF_F_LRO) &&
-		 (lio->dev_capability & NETIF_F_LRO))
+		 (lio->dev_capability & NETIF_F_LRO) &&
+		 (netdev->features & NETIF_F_LRO))
 		liquidio_set_feature(netdev, OCTNET_CMD_LRO_DISABLE,
 				     OCTNIC_LROIPV4 | OCTNIC_LROIPV6);
 
@@ -3629,31 +2790,18 @@ static int liquidio_set_features(struct net_device *netdev,
 		liquidio_set_rxcsum_command(netdev, OCTNET_CMD_TNL_RX_CSUM_CTL,
 					    OCTNET_CMD_RXCSUM_DISABLE);
 
-	return 0;
-}
-
-static void liquidio_add_vxlan_port(struct net_device *netdev,
-				    struct udp_tunnel_info *ti)
-{
-	if (ti->type != UDP_TUNNEL_TYPE_VXLAN)
-		return;
+	if ((features & NETIF_F_HW_VLAN_CTAG_FILTER) &&
+	    (lio->dev_capability & NETIF_F_HW_VLAN_CTAG_FILTER) &&
+	    !(netdev->features & NETIF_F_HW_VLAN_CTAG_FILTER))
+		liquidio_set_feature(netdev, OCTNET_CMD_VLAN_FILTER_CTL,
+				     OCTNET_CMD_VLAN_FILTER_ENABLE);
+	else if (!(features & NETIF_F_HW_VLAN_CTAG_FILTER) &&
+		 (lio->dev_capability & NETIF_F_HW_VLAN_CTAG_FILTER) &&
+		 (netdev->features & NETIF_F_HW_VLAN_CTAG_FILTER))
+		liquidio_set_feature(netdev, OCTNET_CMD_VLAN_FILTER_CTL,
+				     OCTNET_CMD_VLAN_FILTER_DISABLE);
 
-	liquidio_vxlan_port_command(netdev,
-				    OCTNET_CMD_VXLAN_PORT_CONFIG,
-				    htons(ti->port),
-				    OCTNET_CMD_VXLAN_PORT_ADD);
-}
-
-static void liquidio_del_vxlan_port(struct net_device *netdev,
-				    struct udp_tunnel_info *ti)
-{
-	if (ti->type != UDP_TUNNEL_TYPE_VXLAN)
-		return;
-
-	liquidio_vxlan_port_command(netdev,
-				    OCTNET_CMD_VXLAN_PORT_CONFIG,
-				    htons(ti->port),
-				    OCTNET_CMD_VXLAN_PORT_DEL);
+	return 0;
 }
 
 static int __liquidio_set_vf_mac(struct net_device *netdev, int vfidx,
@@ -3662,6 +2810,7 @@ static int __liquidio_set_vf_mac(struct net_device *netdev, int vfidx,
 	struct lio *lio = GET_LIO(netdev);
 	struct octeon_device *oct = lio->oct_dev;
 	struct octnic_ctrl_pkt nctrl;
+	int ret = 0;
 
 	if (!is_valid_ether_addr(mac))
 		return -EINVAL;
@@ -3675,12 +2824,13 @@ static int __liquidio_set_vf_mac(struct net_device *netdev, int vfidx,
 	nctrl.ncmd.s.cmd = OCTNET_CMD_CHANGE_MACADDR;
 	/* vfidx is 0 based, but vf_num (param1) is 1 based */
 	nctrl.ncmd.s.param1 = vfidx + 1;
-	nctrl.ncmd.s.param2 = (is_admin_assigned ? 1 : 0);
 	nctrl.ncmd.s.more = 1;
 	nctrl.iq_no = lio->linfo.txpciq[0].s.q_no;
 	nctrl.netpndev = (u64)netdev;
-	nctrl.cb_fn = liquidio_link_ctrl_cmd_completion;
-	nctrl.wait_time = LIO_CMD_WAIT_TM;
+	if (is_admin_assigned) {
+		nctrl.ncmd.s.param2 = true;
+		nctrl.cb_fn = liquidio_link_ctrl_cmd_completion;
+	}
 
 	nctrl.udd[0] = 0;
 	/* The MAC Address is presented in network byte order. */
@@ -3688,9 +2838,11 @@ static int __liquidio_set_vf_mac(struct net_device *netdev, int vfidx,
 
 	oct->sriov_info.vf_macaddr[vfidx] = nctrl.udd[0];
 
-	octnet_send_nic_ctrl_pkt(oct, &nctrl);
+	ret = octnet_send_nic_ctrl_pkt(oct, &nctrl);
+	if (ret > 0)
+		ret = -EIO;
 
-	return 0;
+	return ret;
 }
 
 static int liquidio_set_vf_mac(struct net_device *netdev, int vfidx, u8 *mac)
@@ -3709,6 +2861,62 @@ static int liquidio_set_vf_mac(struct net_device *netdev, int vfidx, u8 *mac)
 	return retval;
 }
 
+static int liquidio_set_vf_spoofchk(struct net_device *netdev, int vfidx,
+				    bool enable)
+{
+	struct lio *lio = GET_LIO(netdev);
+	struct octeon_device *oct = lio->oct_dev;
+	struct octnic_ctrl_pkt nctrl;
+	int retval;
+
+	if (!(oct->fw_info.app_cap_flags & LIQUIDIO_SPOOFCHK_CAP)) {
+		netif_info(lio, drv, lio->netdev,
+			   "firmware does not support spoofchk\n");
+		return -EOPNOTSUPP;
+	}
+
+	if (vfidx < 0 || vfidx >= oct->sriov_info.num_vfs_alloced) {
+		netif_info(lio, drv, lio->netdev, "Invalid vfidx %d\n", vfidx);
+		return -EINVAL;
+	}
+
+	if (enable) {
+		if (oct->sriov_info.vf_spoofchk[vfidx])
+			return 0;
+	} else {
+		/* Clear */
+		if (!oct->sriov_info.vf_spoofchk[vfidx])
+			return 0;
+	}
+
+	memset(&nctrl, 0, sizeof(struct octnic_ctrl_pkt));
+	nctrl.ncmd.s.cmdgroup = OCTNET_CMD_GROUP1;
+	nctrl.ncmd.s.cmd = OCTNET_CMD_SET_VF_SPOOFCHK;
+	nctrl.ncmd.s.param1 =
+		vfidx + 1; /* vfidx is 0 based,
+			    * but vf_num (param1) is 1 based
+			    */
+	nctrl.ncmd.s.param2 = enable;
+	nctrl.ncmd.s.more = 0;
+	nctrl.iq_no = lio->linfo.txpciq[0].s.q_no;
+	nctrl.cb_fn = NULL;
+
+	retval = octnet_send_nic_ctrl_pkt(oct, &nctrl);
+
+	if (retval) {
+		netif_info(lio, drv, lio->netdev,
+			   "Failed to set VF %d spoofchk %s\n", vfidx,
+			enable ? "on" : "off");
+		return -1;
+	}
+
+	oct->sriov_info.vf_spoofchk[vfidx] = enable;
+	netif_info(lio, drv, lio->netdev, "VF %u spoofchk is %s\n", vfidx,
+		   enable ? "on" : "off");
+
+	return 0;
+}
+
 static int liquidio_set_vf_vlan(struct net_device *netdev, int vfidx,
 				u16 vlan, u8 qos, __be16 vlan_proto)
 {
@@ -3716,6 +2924,7 @@ static int liquidio_set_vf_vlan(struct net_device *netdev, int vfidx,
 	struct octeon_device *oct = lio->oct_dev;
 	struct octnic_ctrl_pkt nctrl;
 	u16 vlantci;
+	int ret = 0;
 
 	if (vfidx < 0 || vfidx >= oct->sriov_info.num_vfs_alloced)
 		return -EINVAL;
@@ -3746,14 +2955,18 @@ static int liquidio_set_vf_vlan(struct net_device *netdev, int vfidx,
 	    vfidx + 1; /* vfidx is 0 based, but vf_num (param2) is 1 based */
 	nctrl.ncmd.s.more = 0;
 	nctrl.iq_no = lio->linfo.txpciq[0].s.q_no;
-	nctrl.cb_fn = 0;
-	nctrl.wait_time = LIO_CMD_WAIT_TM;
+	nctrl.cb_fn = NULL;
 
-	octnet_send_nic_ctrl_pkt(oct, &nctrl);
+	ret = octnet_send_nic_ctrl_pkt(oct, &nctrl);
+	if (ret) {
+		if (ret > 0)
+			ret = -EIO;
+		return ret;
+	}
 
 	oct->sriov_info.vf_vlantci[vfidx] = vlantci;
 
-	return 0;
+	return ret;
 }
 
 static int liquidio_get_vf_config(struct net_device *netdev, int vfidx,
@@ -3766,12 +2979,111 @@ static int liquidio_get_vf_config(struct net_device *netdev, int vfidx,
 	if (vfidx < 0 || vfidx >= oct->sriov_info.num_vfs_alloced)
 		return -EINVAL;
 
+	memset(ivi, 0, sizeof(struct ifla_vf_info));
+
 	ivi->vf = vfidx;
 	macaddr = 2 + (u8 *)&oct->sriov_info.vf_macaddr[vfidx];
 	ether_addr_copy(&ivi->mac[0], macaddr);
 	ivi->vlan = oct->sriov_info.vf_vlantci[vfidx] & VLAN_VID_MASK;
 	ivi->qos = oct->sriov_info.vf_vlantci[vfidx] >> VLAN_PRIO_SHIFT;
+	if (oct->sriov_info.trusted_vf.active &&
+	    oct->sriov_info.trusted_vf.id == vfidx)
+		ivi->trusted = true;
+	else
+		ivi->trusted = false;
 	ivi->linkstate = oct->sriov_info.vf_linkstate[vfidx];
+	ivi->spoofchk = oct->sriov_info.vf_spoofchk[vfidx];
+	ivi->max_tx_rate = lio->linfo.link.s.speed;
+	ivi->min_tx_rate = 0;
+
+	return 0;
+}
+
+static int liquidio_send_vf_trust_cmd(struct lio *lio, int vfidx, bool trusted)
+{
+	struct octeon_device *oct = lio->oct_dev;
+	struct octeon_soft_command *sc;
+	int retval;
+
+	sc = octeon_alloc_soft_command(oct, 0, 16, 0);
+	if (!sc)
+		return -ENOMEM;
+
+	sc->iq_no = lio->linfo.txpciq[0].s.q_no;
+
+	/* vfidx is 0 based, but vf_num (param1) is 1 based */
+	octeon_prepare_soft_command(oct, sc, OPCODE_NIC,
+				    OPCODE_NIC_SET_TRUSTED_VF, 0, vfidx + 1,
+				    trusted);
+
+	init_completion(&sc->complete);
+	sc->sc_status = OCTEON_REQUEST_PENDING;
+
+	retval = octeon_send_soft_command(oct, sc);
+	if (retval == IQ_SEND_FAILED) {
+		octeon_free_soft_command(oct, sc);
+		retval = -1;
+	} else {
+		/* Wait for response or timeout */
+		retval = wait_for_sc_completion_timeout(oct, sc, 0);
+		if (retval)
+			return (retval);
+
+		WRITE_ONCE(sc->caller_is_done, true);
+	}
+
+	return retval;
+}
+
+static int liquidio_set_vf_trust(struct net_device *netdev, int vfidx,
+				 bool setting)
+{
+	struct lio *lio = GET_LIO(netdev);
+	struct octeon_device *oct = lio->oct_dev;
+
+	if (strcmp(oct->fw_info.liquidio_firmware_version, "1.7.1") < 0) {
+		/* trusted vf is not supported by firmware older than 1.7.1 */
+		return -EOPNOTSUPP;
+	}
+
+	if (vfidx < 0 || vfidx >= oct->sriov_info.num_vfs_alloced) {
+		netif_info(lio, drv, lio->netdev, "Invalid vfidx %d\n", vfidx);
+		return -EINVAL;
+	}
+
+	if (setting) {
+		/* Set */
+
+		if (oct->sriov_info.trusted_vf.active &&
+		    oct->sriov_info.trusted_vf.id == vfidx)
+			return 0;
+
+		if (oct->sriov_info.trusted_vf.active) {
+			netif_info(lio, drv, lio->netdev, "More than one trusted VF is not allowed\n");
+			return -EPERM;
+		}
+	} else {
+		/* Clear */
+
+		if (!oct->sriov_info.trusted_vf.active)
+			return 0;
+	}
+
+	if (!liquidio_send_vf_trust_cmd(lio, vfidx, setting)) {
+		if (setting) {
+			oct->sriov_info.trusted_vf.id = vfidx;
+			oct->sriov_info.trusted_vf.active = true;
+		} else {
+			oct->sriov_info.trusted_vf.active = false;
+		}
+
+		netif_info(lio, drv, lio->netdev, "VF %u is %strusted\n", vfidx,
+			   setting ? "" : "not ");
+	} else {
+		netif_info(lio, drv, lio->netdev, "Failed to set VF trusted\n");
+		return -1;
+	}
+
 	return 0;
 }
 
@@ -3781,6 +3093,7 @@ static int liquidio_set_vf_link_state(struct net_device *netdev, int vfidx,
 	struct lio *lio = GET_LIO(netdev);
 	struct octeon_device *oct = lio->oct_dev;
 	struct octnic_ctrl_pkt nctrl;
+	int ret = 0;
 
 	if (vfidx < 0 || vfidx >= oct->sriov_info.num_vfs_alloced)
 		return -EINVAL;
@@ -3795,21 +3108,118 @@ static int liquidio_set_vf_link_state(struct net_device *netdev, int vfidx,
 	nctrl.ncmd.s.param2 = linkstate;
 	nctrl.ncmd.s.more = 0;
 	nctrl.iq_no = lio->linfo.txpciq[0].s.q_no;
-	nctrl.cb_fn = 0;
-	nctrl.wait_time = LIO_CMD_WAIT_TM;
+	nctrl.cb_fn = NULL;
+
+	ret = octnet_send_nic_ctrl_pkt(oct, &nctrl);
+
+	if (!ret)
+		oct->sriov_info.vf_linkstate[vfidx] = linkstate;
+	else if (ret > 0)
+		ret = -EIO;
 
-	octnet_send_nic_ctrl_pkt(oct, &nctrl);
+	return ret;
+}
+
+static int
+liquidio_eswitch_mode_get(struct devlink *devlink, u16 *mode)
+{
+	struct lio_devlink_priv *priv;
+	struct octeon_device *oct;
+
+	priv = devlink_priv(devlink);
+	oct = priv->oct;
 
-	oct->sriov_info.vf_linkstate[vfidx] = linkstate;
+	*mode = oct->eswitch_mode;
 
 	return 0;
 }
 
+static int
+liquidio_eswitch_mode_set(struct devlink *devlink, u16 mode,
+			  struct netlink_ext_ack *extack)
+{
+	struct lio_devlink_priv *priv;
+	struct octeon_device *oct;
+	int ret = 0;
+
+	priv = devlink_priv(devlink);
+	oct = priv->oct;
+
+	if (!(oct->fw_info.app_cap_flags & LIQUIDIO_SWITCHDEV_CAP))
+		return -EINVAL;
+
+	if (oct->eswitch_mode == mode)
+		return 0;
+
+	switch (mode) {
+	case DEVLINK_ESWITCH_MODE_SWITCHDEV:
+		oct->eswitch_mode = mode;
+		ret = lio_vf_rep_create(oct);
+		break;
+
+	case DEVLINK_ESWITCH_MODE_LEGACY:
+		lio_vf_rep_destroy(oct);
+		oct->eswitch_mode = mode;
+		break;
+
+	default:
+		ret = -EINVAL;
+	}
+
+	return ret;
+}
+
+static const struct devlink_ops liquidio_devlink_ops = {
+	.eswitch_mode_get = liquidio_eswitch_mode_get,
+	.eswitch_mode_set = liquidio_eswitch_mode_set,
+};
+
+static int
+liquidio_get_port_parent_id(struct net_device *dev,
+			    struct netdev_phys_item_id *ppid)
+{
+	struct lio *lio = GET_LIO(dev);
+	struct octeon_device *oct = lio->oct_dev;
+
+	if (oct->eswitch_mode != DEVLINK_ESWITCH_MODE_SWITCHDEV)
+		return -EOPNOTSUPP;
+
+	ppid->id_len = ETH_ALEN;
+	ether_addr_copy(ppid->id, (void *)&lio->linfo.hw_addr + 2);
+
+	return 0;
+}
+
+static int liquidio_get_vf_stats(struct net_device *netdev, int vfidx,
+				 struct ifla_vf_stats *vf_stats)
+{
+	struct lio *lio = GET_LIO(netdev);
+	struct octeon_device *oct = lio->oct_dev;
+	struct oct_vf_stats stats;
+	int ret;
+
+	if (vfidx < 0 || vfidx >= oct->sriov_info.num_vfs_alloced)
+		return -EINVAL;
+
+	memset(&stats, 0, sizeof(struct oct_vf_stats));
+	ret = cn23xx_get_vf_stats(oct, vfidx, &stats);
+	if (!ret) {
+		vf_stats->rx_packets = stats.rx_packets;
+		vf_stats->tx_packets = stats.tx_packets;
+		vf_stats->rx_bytes = stats.rx_bytes;
+		vf_stats->tx_bytes = stats.tx_bytes;
+		vf_stats->broadcast = stats.broadcast;
+		vf_stats->multicast = stats.multicast;
+	}
+
+	return ret;
+}
+
 static const struct net_device_ops lionetdevops = {
 	.ndo_open		= liquidio_open,
 	.ndo_stop		= liquidio_stop,
 	.ndo_start_xmit		= liquidio_xmit,
-	.ndo_get_stats		= liquidio_get_stats,
+	.ndo_get_stats64	= liquidio_get_stats64,
 	.ndo_set_mac_address	= liquidio_set_mac,
 	.ndo_set_rx_mode	= liquidio_set_mcast_list,
 	.ndo_tx_timeout		= liquidio_tx_timeout,
@@ -3817,18 +3227,21 @@ static const struct net_device_ops lionetdevops = {
 	.ndo_vlan_rx_add_vid    = liquidio_vlan_rx_add_vid,
 	.ndo_vlan_rx_kill_vid   = liquidio_vlan_rx_kill_vid,
 	.ndo_change_mtu		= liquidio_change_mtu,
-	.ndo_do_ioctl		= liquidio_ioctl,
+	.ndo_eth_ioctl		= liquidio_ioctl,
 	.ndo_fix_features	= liquidio_fix_features,
 	.ndo_set_features	= liquidio_set_features,
-	.ndo_udp_tunnel_add	= liquidio_add_vxlan_port,
-	.ndo_udp_tunnel_del	= liquidio_del_vxlan_port,
 	.ndo_set_vf_mac		= liquidio_set_vf_mac,
 	.ndo_set_vf_vlan	= liquidio_set_vf_vlan,
 	.ndo_get_vf_config	= liquidio_get_vf_config,
+	.ndo_set_vf_spoofchk	= liquidio_set_vf_spoofchk,
+	.ndo_set_vf_trust	= liquidio_set_vf_trust,
 	.ndo_set_vf_link_state  = liquidio_set_vf_link_state,
+	.ndo_get_vf_stats	= liquidio_get_vf_stats,
+	.ndo_get_port_parent_id	= liquidio_get_port_parent_id,
 };
 
-/** \brief Entry point for the liquidio module
+/**
+ * liquidio_init - Entry point for the liquidio module
  */
 static int __init liquidio_init(void)
 {
@@ -3884,7 +3297,7 @@ static int lio_nic_info(struct octeon_recv_info *recv_info, void *buf)
 	union oct_link_status *ls;
 	int i;
 
-	if (recv_pkt->buffer_size[0] != sizeof(*ls)) {
+	if (recv_pkt->buffer_size[0] != (sizeof(*ls) + OCT_DROQ_INFO_SIZE)) {
 		dev_err(&oct->pci_dev->dev, "Malformed NIC_INFO, len=%d, ifidx=%d\n",
 			recv_pkt->buffer_size[0],
 			recv_pkt->rh.r_nic_info.gmxport);
@@ -3892,7 +3305,8 @@ static int lio_nic_info(struct octeon_recv_info *recv_info, void *buf)
 	}
 
 	gmxport = recv_pkt->rh.r_nic_info.gmxport;
-	ls = (union oct_link_status *)get_rbd(recv_pkt->buffer_ptr[0]);
+	ls = (union oct_link_status *)(get_rbd(recv_pkt->buffer_ptr[0]) +
+		OCT_DROQ_INFO_SIZE);
 
 	octeon_swap_8B_data((u64 *)ls, (sizeof(union oct_link_status)) >> 3);
 	for (i = 0; i < oct->ifcount; i++) {
@@ -3910,8 +3324,8 @@ nic_info_err:
 }
 
 /**
- * \brief Setup network interfaces
- * @param octeon_dev  octeon device
+ * setup_nic_devices - Setup network interfaces
+ * @octeon_dev:  octeon device
  *
  * Called during init time for each device. It assumes the NIC
  * is already up and running.  The link information for each
@@ -3921,18 +3335,22 @@ static int setup_nic_devices(struct octeon_device *octeon_dev)
 {
 	struct lio *lio = NULL;
 	struct net_device *netdev;
-	u8 mac[6], i, j;
+	u8 mac[6], i, j, *fw_ver, *micro_ver;
+	unsigned long micro;
+	u32 cur_ver;
 	struct octeon_soft_command *sc;
-	struct liquidio_if_cfg_context *ctx;
 	struct liquidio_if_cfg_resp *resp;
 	struct octdev_props *props;
 	int retval, num_iqueues, num_oqueues;
+	int max_num_queues = 0;
 	union oct_nic_if_cfg if_cfg;
 	unsigned int base_queue;
 	unsigned int gmx_port_id;
-	u32 resp_size, ctx_size, data_size;
+	u32 resp_size, data_size;
 	u32 ifidx_or_pfnum;
 	struct lio_version *vdata;
+	struct devlink *devlink;
+	struct lio_devlink_priv *lio_devlink;
 
 	/* This is to handle link status changes */
 	octeon_register_dispatch_fn(octeon_dev, OPCODE_NIC,
@@ -3953,13 +3371,11 @@ static int setup_nic_devices(struct octeon_device *octeon_dev)
 
 	for (i = 0; i < octeon_dev->ifcount; i++) {
 		resp_size = sizeof(struct liquidio_if_cfg_resp);
-		ctx_size = sizeof(struct liquidio_if_cfg_context);
 		data_size = sizeof(struct lio_version);
 		sc = (struct octeon_soft_command *)
 			octeon_alloc_soft_command(octeon_dev, data_size,
-						  resp_size, ctx_size);
+						  resp_size, 0);
 		resp = (struct liquidio_if_cfg_resp *)sc->virtrptr;
-		ctx  = (struct liquidio_if_cfg_context *)sc->ctxptr;
 		vdata = (struct lio_version *)sc->virtdptr;
 
 		*((u64 *)vdata) = 0;
@@ -3989,9 +3405,6 @@ static int setup_nic_devices(struct octeon_device *octeon_dev)
 		dev_dbg(&octeon_dev->pci_dev->dev,
 			"requesting config for interface %d, iqs %d, oqs %d\n",
 			ifidx_or_pfnum, num_iqueues, num_oqueues);
-		WRITE_ONCE(ctx->cond, 0);
-		ctx->octeon_id = lio_get_device_id(octeon_dev);
-		init_waitqueue_head(&ctx->wc);
 
 		if_cfg.u64 = 0;
 		if_cfg.s.num_iqueues = num_iqueues;
@@ -4005,9 +3418,8 @@ static int setup_nic_devices(struct octeon_device *octeon_dev)
 					    OPCODE_NIC_IF_CFG, 0,
 					    if_cfg.u64, 0);
 
-		sc->callback = if_cfg_callback;
-		sc->callback_arg = sc;
-		sc->wait_time = 3000;
+		init_completion(&sc->complete);
+		sc->sc_status = OCTEON_REQUEST_PENDING;
 
 		retval = octeon_send_soft_command(octeon_dev, sc);
 		if (retval == IQ_SEND_FAILED) {
@@ -4015,23 +3427,52 @@ static int setup_nic_devices(struct octeon_device *octeon_dev)
 				"iq/oq config failed status: %x\n",
 				retval);
 			/* Soft instr is freed by driver in case of failure. */
-			goto setup_nic_dev_fail;
+			octeon_free_soft_command(octeon_dev, sc);
+			return(-EIO);
 		}
 
 		/* Sleep on a wait queue till the cond flag indicates that the
 		 * response arrived or timed-out.
 		 */
-		if (sleep_cond(&ctx->wc, &ctx->cond) == -EINTR) {
-			dev_err(&octeon_dev->pci_dev->dev, "Wait interrupted\n");
-			goto setup_nic_wait_intr;
-		}
+		retval = wait_for_sc_completion_timeout(octeon_dev, sc, 0);
+		if (retval)
+			return retval;
 
 		retval = resp->status;
 		if (retval) {
 			dev_err(&octeon_dev->pci_dev->dev, "iq/oq config failed\n");
-			goto setup_nic_dev_fail;
+			WRITE_ONCE(sc->caller_is_done, true);
+			goto setup_nic_dev_done;
+		}
+		snprintf(octeon_dev->fw_info.liquidio_firmware_version,
+			 32, "%s",
+			 resp->cfg_info.liquidio_firmware_version);
+
+		/* Verify f/w version (in case of 'auto' loading from flash) */
+		fw_ver = octeon_dev->fw_info.liquidio_firmware_version;
+		if (memcmp(LIQUIDIO_BASE_VERSION,
+			   fw_ver,
+			   strlen(LIQUIDIO_BASE_VERSION))) {
+			dev_err(&octeon_dev->pci_dev->dev,
+				"Unmatched firmware version. Expected %s.x, got %s.\n",
+				LIQUIDIO_BASE_VERSION, fw_ver);
+			WRITE_ONCE(sc->caller_is_done, true);
+			goto setup_nic_dev_done;
+		} else if (atomic_read(octeon_dev->adapter_fw_state) ==
+			   FW_IS_PRELOADED) {
+			dev_info(&octeon_dev->pci_dev->dev,
+				 "Using auto-loaded firmware version %s.\n",
+				 fw_ver);
 		}
 
+		/* extract micro version field; point past '<maj>.<min>.' */
+		micro_ver = fw_ver + strlen(LIQUIDIO_BASE_VERSION) + 1;
+		if (kstrtoul(micro_ver, 10, &micro) != 0)
+			micro = 0;
+		octeon_dev->fw_info.ver.maj = LIQUIDIO_BASE_MAJOR_VERSION;
+		octeon_dev->fw_info.ver.min = LIQUIDIO_BASE_MINOR_VERSION;
+		octeon_dev->fw_info.ver.rev = micro;
+
 		octeon_swap_8B_data((u64 *)(&resp->cfg_info),
 				    (sizeof(struct liquidio_if_cfg_info)) >> 3);
 
@@ -4043,17 +3484,28 @@ static int setup_nic_devices(struct octeon_device *octeon_dev)
 				"Got bad iqueues (%016llx) or oqueues (%016llx) from firmware.\n",
 				resp->cfg_info.iqmask,
 				resp->cfg_info.oqmask);
-			goto setup_nic_dev_fail;
+			WRITE_ONCE(sc->caller_is_done, true);
+			goto setup_nic_dev_done;
+		}
+
+		if (OCTEON_CN6XXX(octeon_dev)) {
+			max_num_queues = CFG_GET_IQ_MAX_Q(CHIP_CONF(octeon_dev,
+								    cn6xxx));
+		} else if (OCTEON_CN23XX_PF(octeon_dev)) {
+			max_num_queues = CFG_GET_IQ_MAX_Q(CHIP_CONF(octeon_dev,
+								    cn23xx_pf));
 		}
+
 		dev_dbg(&octeon_dev->pci_dev->dev,
-			"interface %d, iqmask %016llx, oqmask %016llx, numiqueues %d, numoqueues %d\n",
+			"interface %d, iqmask %016llx, oqmask %016llx, numiqueues %d, numoqueues %d max_num_queues: %d\n",
 			i, resp->cfg_info.iqmask, resp->cfg_info.oqmask,
-			num_iqueues, num_oqueues);
-		netdev = alloc_etherdev_mq(LIO_SIZE, num_iqueues);
+			num_iqueues, num_oqueues, max_num_queues);
+		netdev = alloc_etherdev_mq(LIO_SIZE, max_num_queues);
 
 		if (!netdev) {
 			dev_err(&octeon_dev->pci_dev->dev, "Device allocation failed\n");
-			goto setup_nic_dev_fail;
+			WRITE_ONCE(sc->caller_is_done, true);
+			goto setup_nic_dev_done;
 		}
 
 		SET_NETDEV_DEV(netdev, &octeon_dev->pci_dev->dev);
@@ -4063,6 +3515,22 @@ static int setup_nic_devices(struct octeon_device *octeon_dev)
 		 */
 		netdev->netdev_ops = &lionetdevops;
 
+		retval = netif_set_real_num_rx_queues(netdev, num_oqueues);
+		if (retval) {
+			dev_err(&octeon_dev->pci_dev->dev,
+				"setting real number rx failed\n");
+			WRITE_ONCE(sc->caller_is_done, true);
+			goto setup_nic_dev_free;
+		}
+
+		retval = netif_set_real_num_tx_queues(netdev, num_iqueues);
+		if (retval) {
+			dev_err(&octeon_dev->pci_dev->dev,
+				"setting real number tx failed\n");
+			WRITE_ONCE(sc->caller_is_done, true);
+			goto setup_nic_dev_free;
+		}
+
 		lio = GET_LIO(netdev);
 
 		memset(lio, 0, sizeof(struct lio));
@@ -4087,6 +3555,8 @@ static int setup_nic_devices(struct octeon_device *octeon_dev)
 		lio->linfo.gmxport = resp->cfg_info.linfo.gmxport;
 		lio->linfo.link.u64 = resp->cfg_info.linfo.link.u64;
 
+		WRITE_ONCE(sc->caller_is_done, true);
+
 		lio->msg_enable = netif_msg_init(debug, DEFAULT_MSG_ENABLE);
 
 		if (OCTEON_CN23XX_PF(octeon_dev) ||
@@ -4099,7 +3569,7 @@ static int setup_nic_devices(struct octeon_device *octeon_dev)
 					      | NETIF_F_TSO | NETIF_F_TSO6
 					      | NETIF_F_LRO;
 		}
-		netif_set_gso_max_size(netdev, OCTNIC_GSO_MAX_SIZE);
+		netif_set_tso_max_size(netdev, OCTNIC_GSO_MAX_SIZE);
 
 		/*  Copy of transmit encapsulation capabilities:
 		 *  TSO, TSO6, Checksums for this device
@@ -4115,6 +3585,8 @@ static int setup_nic_devices(struct octeon_device *octeon_dev)
 		netdev->hw_enc_features = (lio->enc_dev_capability &
 					   ~NETIF_F_LRO);
 
+		netdev->udp_tunnel_nic_info = &liquidio_udp_tunnels;
+
 		lio->dev_capability |= NETIF_F_GSO_UDP_TUNNEL;
 
 		netdev->vlan_features = lio->dev_capability;
@@ -4148,13 +3620,12 @@ static int setup_nic_devices(struct octeon_device *octeon_dev)
 		for (j = 0; j < octeon_dev->sriov_info.max_vfs; j++) {
 			u8 vfmac[ETH_ALEN];
 
-			random_ether_addr(&vfmac[0]);
-			if (__liquidio_set_vf_mac(netdev, j,
-						  &vfmac[0], false)) {
+			eth_random_addr(vfmac);
+			if (__liquidio_set_vf_mac(netdev, j, vfmac, false)) {
 				dev_err(&octeon_dev->pci_dev->dev,
 					"Error setting VF%d MAC address\n",
 					j);
-				goto setup_nic_dev_fail;
+				goto setup_nic_dev_free;
 			}
 		}
 
@@ -4165,16 +3636,18 @@ static int setup_nic_devices(struct octeon_device *octeon_dev)
 
 		/* Copy MAC Address to OS network device structure */
 
-		ether_addr_copy(netdev->dev_addr, mac);
+		eth_hw_addr_set(netdev, mac);
 
 		/* By default all interfaces on a single Octeon uses the same
 		 * tx and rx queues
 		 */
 		lio->txq = lio->linfo.txpciq[0].s.q_no;
 		lio->rxq = lio->linfo.rxpciq[0].s.q_no;
-		if (setup_io_queues(octeon_dev, i)) {
+		if (liquidio_setup_io_queues(octeon_dev, i,
+					     lio->linfo.num_txpciq,
+					     lio->linfo.num_rxpciq)) {
 			dev_err(&octeon_dev->pci_dev->dev, "I/O queues creation failed\n");
-			goto setup_nic_dev_fail;
+			goto setup_nic_dev_free;
 		}
 
 		ifstate_set(lio, LIO_IFSTATE_DROQ_OPS);
@@ -4182,10 +3655,10 @@ static int setup_nic_devices(struct octeon_device *octeon_dev)
 		lio->tx_qsize = octeon_get_tx_qsize(octeon_dev, lio->txq);
 		lio->rx_qsize = octeon_get_rx_qsize(octeon_dev, lio->rxq);
 
-		if (setup_glists(octeon_dev, lio, num_iqueues)) {
+		if (lio_setup_glists(octeon_dev, lio, num_iqueues)) {
 			dev_err(&octeon_dev->pci_dev->dev,
 				"Gather list allocation failed\n");
-			goto setup_nic_dev_fail;
+			goto setup_nic_dev_free;
 		}
 
 		/* Register ethtool support */
@@ -4199,22 +3672,28 @@ static int setup_nic_devices(struct octeon_device *octeon_dev)
 			liquidio_set_feature(netdev, OCTNET_CMD_LRO_ENABLE,
 					     OCTNIC_LROIPV4 | OCTNIC_LROIPV6);
 
-		liquidio_set_feature(netdev, OCTNET_CMD_ENABLE_VLAN_FILTER, 0);
+		liquidio_set_feature(netdev, OCTNET_CMD_VLAN_FILTER_CTL,
+				     OCTNET_CMD_VLAN_FILTER_ENABLE);
 
 		if ((debug != -1) && (debug & NETIF_MSG_HW))
 			liquidio_set_feature(netdev,
 					     OCTNET_CMD_VERBOSE_ENABLE, 0);
 
 		if (setup_link_status_change_wq(netdev))
-			goto setup_nic_dev_fail;
+			goto setup_nic_dev_free;
+
+		if ((octeon_dev->fw_info.app_cap_flags &
+		     LIQUIDIO_TIME_SYNC_CAP) &&
+		    setup_sync_octeon_time_wq(netdev))
+			goto setup_nic_dev_free;
 
 		if (setup_rx_oom_poll_fn(netdev))
-			goto setup_nic_dev_fail;
+			goto setup_nic_dev_free;
 
 		/* Register the network device with the OS */
 		if (register_netdev(netdev)) {
 			dev_err(&octeon_dev->pci_dev->dev, "Device registration failed\n");
-			goto setup_nic_dev_fail;
+			goto setup_nic_dev_free;
 		}
 
 		dev_dbg(&octeon_dev->pci_dev->dev,
@@ -4237,22 +3716,73 @@ static int setup_nic_devices(struct octeon_device *octeon_dev)
 		dev_dbg(&octeon_dev->pci_dev->dev,
 			"NIC ifidx:%d Setup successful\n", i);
 
-		octeon_free_soft_command(octeon_dev, sc);
+		if (octeon_dev->subsystem_id ==
+			OCTEON_CN2350_25GB_SUBSYS_ID ||
+		    octeon_dev->subsystem_id ==
+			OCTEON_CN2360_25GB_SUBSYS_ID) {
+			cur_ver = OCT_FW_VER(octeon_dev->fw_info.ver.maj,
+					     octeon_dev->fw_info.ver.min,
+					     octeon_dev->fw_info.ver.rev);
+
+			/* speed control unsupported in f/w older than 1.7.2 */
+			if (cur_ver < OCT_FW_VER(1, 7, 2)) {
+				dev_info(&octeon_dev->pci_dev->dev,
+					 "speed setting not supported by f/w.");
+				octeon_dev->speed_setting = 25;
+				octeon_dev->no_speed_setting = 1;
+			} else {
+				liquidio_get_speed(lio);
+			}
+
+			if (octeon_dev->speed_setting == 0) {
+				octeon_dev->speed_setting = 25;
+				octeon_dev->no_speed_setting = 1;
+			}
+		} else {
+			octeon_dev->no_speed_setting = 1;
+			octeon_dev->speed_setting = 10;
+		}
+		octeon_dev->speed_boot = octeon_dev->speed_setting;
+
+		/* don't read FEC setting if unsupported by f/w (see above) */
+		if (octeon_dev->speed_boot == 25 &&
+		    !octeon_dev->no_speed_setting) {
+			liquidio_get_fec(lio);
+			octeon_dev->props[lio->ifidx].fec_boot =
+				octeon_dev->props[lio->ifidx].fec;
+		}
+	}
+
+	device_lock(&octeon_dev->pci_dev->dev);
+	devlink = devlink_alloc(&liquidio_devlink_ops,
+				sizeof(struct lio_devlink_priv),
+				&octeon_dev->pci_dev->dev);
+	if (!devlink) {
+		device_unlock(&octeon_dev->pci_dev->dev);
+		dev_err(&octeon_dev->pci_dev->dev, "devlink alloc failed\n");
+		goto setup_nic_dev_free;
 	}
 
-	return 0;
+	lio_devlink = devlink_priv(devlink);
+	lio_devlink->oct = octeon_dev;
 
-setup_nic_dev_fail:
+	octeon_dev->devlink = devlink;
+	octeon_dev->eswitch_mode = DEVLINK_ESWITCH_MODE_LEGACY;
+	devlink_register(devlink);
+	device_unlock(&octeon_dev->pci_dev->dev);
 
-	octeon_free_soft_command(octeon_dev, sc);
+	return 0;
 
-setup_nic_wait_intr:
+setup_nic_dev_free:
 
 	while (i--) {
 		dev_err(&octeon_dev->pci_dev->dev,
 			"NIC ifidx:%d Setup failed\n", i);
 		liquidio_destroy_nic_device(octeon_dev, i);
 	}
+
+setup_nic_dev_done:
+
 	return -ENODEV;
 }
 
@@ -4334,6 +3864,7 @@ static int liquidio_enable_sriov(struct pci_dev *dev, int num_vfs)
 	}
 
 	if (!num_vfs) {
+		lio_vf_rep_destroy(oct);
 		ret = lio_pci_sriov_disable(oct);
 	} else if (num_vfs > oct->sriov_info.max_vfs) {
 		dev_err(&oct->pci_dev->dev,
@@ -4345,6 +3876,10 @@ static int liquidio_enable_sriov(struct pci_dev *dev, int num_vfs)
 		ret = octeon_enable_sriov(oct);
 		dev_info(&oct->pci_dev->dev, "oct->pf_num:%d num_vfs:%d\n",
 			 oct->pf_num, num_vfs);
+		ret = lio_vf_rep_create(oct);
+		if (ret)
+			dev_info(&oct->pci_dev->dev,
+				 "vf representor create failed");
 	}
 
 	return ret;
@@ -4352,8 +3887,8 @@ static int liquidio_enable_sriov(struct pci_dev *dev, int num_vfs)
 #endif
 
 /**
- * \brief initialize the NIC
- * @param oct octeon device
+ * liquidio_init_nic_module - initialize the NIC
+ * @oct: octeon device
  *
  * This initialization routine is called once the Octeon device application is
  * up and running
@@ -4382,6 +3917,18 @@ static int liquidio_init_nic_module(struct octeon_device *oct)
 		goto octnet_init_failure;
 	}
 
+	/* Call vf_rep_modinit if the firmware is switchdev capable
+	 * and do it from the first liquidio function probed.
+	 */
+	if (!oct->octeon_id &&
+	    oct->fw_info.app_cap_flags & LIQUIDIO_SWITCHDEV_CAP) {
+		retval = lio_vf_rep_modinit();
+		if (retval) {
+			liquidio_stop_nic_module(oct);
+			goto octnet_init_failure;
+		}
+	}
+
 	liquidio_ptp_init(oct);
 
 	dev_dbg(&oct->pci_dev->dev, "Network interfaces ready\n");
@@ -4396,9 +3943,10 @@ octnet_init_failure:
 }
 
 /**
- * \brief starter callback that invokes the remaining initialization work after
- * the NIC is up and running.
- * @param octptr  work struct work_struct
+ * nic_starter - finish init
+ * @work:  work struct work_struct
+ *
+ * starter callback that invokes the remaining initialization work after the NIC is up and running.
  */
 static void nic_starter(struct work_struct *work)
 {
@@ -4449,7 +3997,7 @@ octeon_recv_vf_drv_notice(struct octeon_recv_info *recv_info, void *buf)
 	u64 *data, vf_num;
 
 	notice = recv_pkt->rh.r.ossp;
-	data = (u64 *)get_rbd(recv_pkt->buffer_ptr[0]);
+	data = (u64 *)(get_rbd(recv_pkt->buffer_ptr[0]) + OCT_DROQ_INFO_SIZE);
 
 	/* the first 64-bit word of data is the vf_num */
 	vf_num = data[0];
@@ -4491,16 +4039,16 @@ octeon_recv_vf_drv_notice(struct octeon_recv_info *recv_info, void *buf)
 }
 
 /**
- * \brief Device initialization for each Octeon device that is probed
- * @param octeon_dev  octeon device
+ * octeon_device_init - Device initialization for each Octeon device that is probed
+ * @octeon_dev:  octeon device
  */
 static int octeon_device_init(struct octeon_device *octeon_dev)
 {
 	int j, ret;
-	int fw_loaded = 0;
 	char bootcmd[] = "\n";
-	struct octeon_device_priv *oct_priv =
-		(struct octeon_device_priv *)octeon_dev->priv;
+	char *dbg_enb = NULL;
+	enum lio_fw_state fw_state;
+	struct octeon_device_priv *oct_priv = octeon_dev->priv;
 	atomic_set(&octeon_dev->status, OCT_DEV_BEGIN_STATE);
 
 	/* Enable access to the octeon device and make its DMA capability
@@ -4530,26 +4078,40 @@ static int octeon_device_init(struct octeon_device *octeon_dev)
 
 	octeon_dev->app_mode = CVM_DRV_INVALID_APP;
 
-	if (OCTEON_CN23XX_PF(octeon_dev)) {
-		if (!cn23xx_fw_loaded(octeon_dev)) {
-			fw_loaded = 0;
-			if (!fw_type_is_none()) {
-				/* Do a soft reset of the Octeon device. */
-				if (octeon_dev->fn_list.soft_reset(octeon_dev))
-					return 1;
-				/* things might have changed */
-				if (!cn23xx_fw_loaded(octeon_dev))
-					fw_loaded = 0;
-				else
-					fw_loaded = 1;
-			}
-		} else {
-			fw_loaded = 1;
-		}
-	} else if (octeon_dev->fn_list.soft_reset(octeon_dev)) {
-		return 1;
+	/* CN23XX supports preloaded firmware if the following is true:
+	 *
+	 * The adapter indicates that firmware is currently running AND
+	 * 'fw_type' is 'auto'.
+	 *
+	 * (default state is NEEDS_TO_BE_LOADED, override it if appropriate).
+	 */
+	if (OCTEON_CN23XX_PF(octeon_dev) &&
+	    cn23xx_fw_loaded(octeon_dev) && fw_type_is_auto()) {
+		atomic_cmpxchg(octeon_dev->adapter_fw_state,
+			       FW_NEEDS_TO_BE_LOADED, FW_IS_PRELOADED);
 	}
 
+	/* If loading firmware, only first device of adapter needs to do so. */
+	fw_state = atomic_cmpxchg(octeon_dev->adapter_fw_state,
+				  FW_NEEDS_TO_BE_LOADED,
+				  FW_IS_BEING_LOADED);
+
+	/* Here, [local variable] 'fw_state' is set to one of:
+	 *
+	 *   FW_IS_PRELOADED:       No firmware is to be loaded (see above)
+	 *   FW_NEEDS_TO_BE_LOADED: The driver's first instance will load
+	 *                          firmware to the adapter.
+	 *   FW_IS_BEING_LOADED:    The driver's second instance will not load
+	 *                          firmware to the adapter.
+	 */
+
+	/* Prior to f/w load, perform a soft reset of the Octeon device;
+	 * if error resetting, return w/error.
+	 */
+	if (fw_state == FW_NEEDS_TO_BE_LOADED)
+		if (octeon_dev->fn_list.soft_reset(octeon_dev))
+			return 1;
+
 	/* Initialize the dispatch mechanism used to push packets arriving on
 	 * Octeon Output queues.
 	 */
@@ -4623,7 +4185,9 @@ static int octeon_device_init(struct octeon_device *octeon_dev)
 		}
 		atomic_set(&octeon_dev->status, OCT_DEV_MBOX_SETUP_DONE);
 
-		if (octeon_allocate_ioq_vector(octeon_dev)) {
+		if (octeon_allocate_ioq_vector
+				(octeon_dev,
+				 octeon_dev->sriov_info.num_pf_rings)) {
 			dev_err(&octeon_dev->pci_dev->dev, "OCTEON: ioq vector allocation failed\n");
 			return 1;
 		}
@@ -4644,12 +4208,12 @@ static int octeon_device_init(struct octeon_device *octeon_dev)
 
 	/* Initialize the tasklet that handles output queue packet processing.*/
 	dev_dbg(&octeon_dev->pci_dev->dev, "Initializing droq tasklet\n");
-	tasklet_init(&oct_priv->droq_tasklet, octeon_droq_bh,
-		     (unsigned long)octeon_dev);
+	tasklet_setup(&oct_priv->droq_tasklet, octeon_droq_bh);
 
 	/* Setup the interrupt handler and record the INT SUM register address
 	 */
-	if (octeon_setup_interrupt(octeon_dev))
+	if (octeon_setup_interrupt(octeon_dev,
+				   octeon_dev->sriov_info.num_pf_rings))
 		return 1;
 
 	/* Enable Octeon device interrupts */
@@ -4657,6 +4221,18 @@ static int octeon_device_init(struct octeon_device *octeon_dev)
 
 	atomic_set(&octeon_dev->status, OCT_DEV_INTR_SET_DONE);
 
+	/* Send Credit for Octeon Output queues. Credits are always sent BEFORE
+	 * the output queue is enabled.
+	 * This ensures that we'll receive the f/w CORE DRV_ACTIVE message in
+	 * case we've configured CN23XX_SLI_GBL_CONTROL[NOPTR_D] = 0.
+	 * Otherwise, it is possible that the DRV_ACTIVE message will be sent
+	 * before any credits have been issued, causing the ring to be reset
+	 * (and the f/w appear to never have started).
+	 */
+	for (j = 0; j < octeon_dev->num_oqs; j++)
+		writel(octeon_dev->droq[j]->max_count,
+		       octeon_dev->droq[j]->pkts_credit_reg);
+
 	/* Enable the input and output queues for this Octeon device */
 	ret = octeon_dev->fn_list.enable_io_queues(octeon_dev);
 	if (ret) {
@@ -4666,7 +4242,7 @@ static int octeon_device_init(struct octeon_device *octeon_dev)
 
 	atomic_set(&octeon_dev->status, OCT_DEV_IO_QUEUES_DONE);
 
-	if ((!OCTEON_CN23XX_PF(octeon_dev)) || !fw_loaded) {
+	if (fw_state == FW_NEEDS_TO_BE_LOADED) {
 		dev_dbg(&octeon_dev->pci_dev->dev, "Waiting for DDR initialization...\n");
 		if (!ddr_timeout) {
 			dev_info(&octeon_dev->pci_dev->dev,
@@ -4705,10 +4281,19 @@ static int octeon_device_init(struct octeon_device *octeon_dev)
 			dev_err(&octeon_dev->pci_dev->dev, "Could not access board consoles\n");
 			return 1;
 		}
-		ret = octeon_add_console(octeon_dev, 0);
+		/* If console debug enabled, specify empty string to use default
+		 * enablement ELSE specify NULL string for 'disabled'.
+		 */
+		dbg_enb = octeon_console_debug_enabled(0) ? "" : NULL;
+		ret = octeon_add_console(octeon_dev, 0, dbg_enb);
 		if (ret) {
 			dev_err(&octeon_dev->pci_dev->dev, "Could not access board console\n");
 			return 1;
+		} else if (octeon_console_debug_enabled(0)) {
+			/* If console was added AND we're logging console output
+			 * then set our console print function.
+			 */
+			octeon_dev->console[0].print = octeon_dbg_console_print;
 		}
 
 		atomic_set(&octeon_dev->status, OCT_DEV_CONSOLE_INIT_DONE);
@@ -4719,32 +4304,48 @@ static int octeon_device_init(struct octeon_device *octeon_dev)
 			dev_err(&octeon_dev->pci_dev->dev, "Could not load firmware to board\n");
 			return 1;
 		}
-		/* set bit 1 of SLI_SCRATCH_1 to indicate that firmware is
-		 * loaded
-		 */
-		if (OCTEON_CN23XX_PF(octeon_dev))
-			octeon_write_csr64(octeon_dev, CN23XX_SLI_SCRATCH1,
-					   2ULL);
+
+		atomic_set(octeon_dev->adapter_fw_state, FW_HAS_BEEN_LOADED);
 	}
 
 	handshake[octeon_dev->octeon_id].init_ok = 1;
 	complete(&handshake[octeon_dev->octeon_id].init);
 
 	atomic_set(&octeon_dev->status, OCT_DEV_HOST_OK);
+	oct_priv->dev = octeon_dev;
 
-	/* Send Credit for Octeon Output queues. Credits are always sent after
-	 * the output queue is enabled.
-	 */
-	for (j = 0; j < octeon_dev->num_oqs; j++)
-		writel(octeon_dev->droq[j]->max_count,
-		       octeon_dev->droq[j]->pkts_credit_reg);
+	return 0;
+}
+
+/**
+ * octeon_dbg_console_print - Debug console print function
+ * @oct:  octeon device
+ * @console_num: console number
+ * @prefix:      first portion of line to display
+ * @suffix:      second portion of line to display
+ *
+ * The OCTEON debug console outputs entire lines (excluding '\n').
+ * Normally, the line will be passed in the 'prefix' parameter.
+ * However, due to buffering, it is possible for a line to be split into two
+ * parts, in which case they will be passed as the 'prefix' parameter and
+ * 'suffix' parameter.
+ */
+static int octeon_dbg_console_print(struct octeon_device *oct, u32 console_num,
+				    char *prefix, char *suffix)
+{
+	if (prefix && suffix)
+		dev_info(&oct->pci_dev->dev, "%u: %s%s\n", console_num, prefix,
+			 suffix);
+	else if (prefix)
+		dev_info(&oct->pci_dev->dev, "%u: %s\n", console_num, prefix);
+	else if (suffix)
+		dev_info(&oct->pci_dev->dev, "%u: %s\n", console_num, suffix);
 
-	/* Packets can start arriving on the output queues from this point. */
 	return 0;
 }
 
 /**
- * \brief Exits the module
+ * liquidio_exit - Exits the module
  */
 static void __exit liquidio_exit(void)
 {
diff --git a/drivers/net/ethernet/cavium/liquidio/lio_vf_main.c b/drivers/net/ethernet/cavium/liquidio/lio_vf_main.c
index 31d737c22648..3230dff5ba05 100644
--- a/drivers/net/ethernet/cavium/liquidio/lio_vf_main.c
+++ b/drivers/net/ethernet/cavium/liquidio/lio_vf_main.c
@@ -32,7 +32,6 @@
 MODULE_AUTHOR("Cavium Networks, <support@cavium.com>");
 MODULE_DESCRIPTION("Cavium LiquidIO Intelligent Server Adapter Virtual Function Driver");
 MODULE_LICENSE("GPL");
-MODULE_VERSION(LIQUIDIO_VERSION);
 
 static int debug = -1;
 module_param(debug, int, 0644);
@@ -40,28 +39,6 @@ MODULE_PARM_DESC(debug, "NETIF_MSG debug bits");
 
 #define DEFAULT_MSG_ENABLE (NETIF_MSG_DRV | NETIF_MSG_PROBE | NETIF_MSG_LINK)
 
-struct liquidio_if_cfg_context {
-	int octeon_id;
-
-	wait_queue_head_t wc;
-
-	int cond;
-};
-
-struct liquidio_if_cfg_resp {
-	u64 rh;
-	struct liquidio_if_cfg_info cfg_info;
-	u64 status;
-};
-
-struct liquidio_rx_ctl_context {
-	int octeon_id;
-
-	wait_queue_head_t wc;
-
-	int cond;
-};
-
 struct oct_timestamp_resp {
 	u64 rh;
 	u64 timestamp;
@@ -83,36 +60,10 @@ union tx_info {
 	} s;
 };
 
-#define OCTNIC_MAX_SG  (MAX_SKB_FRAGS)
-
 #define OCTNIC_GSO_MAX_HEADER_SIZE 128
 #define OCTNIC_GSO_MAX_SIZE \
 		(CN23XX_DEFAULT_INPUT_JABBER - OCTNIC_GSO_MAX_HEADER_SIZE)
 
-struct octnic_gather {
-	/* List manipulation. Next and prev pointers. */
-	struct list_head list;
-
-	/* Size of the gather component at sg in bytes. */
-	int sg_size;
-
-	/* Number of bytes that sg was adjusted to make it 8B-aligned. */
-	int adjust;
-
-	/* Gather component that can accommodate max sized fragment list
-	 * received from the IP layer.
-	 */
-	struct octeon_sg_entry *sg;
-
-	dma_addr_t sg_dma_ptr;
-};
-
-struct octeon_device_priv {
-	/* Tasklet structures for this device. */
-	struct tasklet_struct droq_tasklet;
-	unsigned long napi_mask;
-};
-
 static int
 liquidio_vf_probe(struct pci_dev *pdev, const struct pci_device_id *ent);
 static void liquidio_vf_remove(struct pci_dev *pdev);
@@ -121,9 +72,8 @@ static int liquidio_stop(struct net_device *netdev);
 
 static int lio_wait_for_oq_pkts(struct octeon_device *oct)
 {
-	struct octeon_device_priv *oct_priv =
-	    (struct octeon_device_priv *)oct->priv;
-	int retry = MAX_VF_IP_OP_PENDING_PKT_COUNT;
+	struct octeon_device_priv *oct_priv = oct->priv;
+	int retry = MAX_IO_PENDING_PKT_COUNT;
 	int pkt_cnt = 0, pending_pkts;
 	int i;
 
@@ -148,34 +98,8 @@ static int lio_wait_for_oq_pkts(struct octeon_device *oct)
 }
 
 /**
- * \brief wait for all pending requests to complete
- * @param oct Pointer to Octeon device
- *
- * Called during shutdown sequence
- */
-static int wait_for_pending_requests(struct octeon_device *oct)
-{
-	int i, pcount = 0;
-
-	for (i = 0; i < MAX_VF_IP_OP_PENDING_PKT_COUNT; i++) {
-		pcount = atomic_read(
-		    &oct->response_list[OCTEON_ORDERED_SC_LIST]
-			 .pending_req_count);
-		if (pcount)
-			schedule_timeout_uninterruptible(HZ / 10);
-		else
-			break;
-	}
-
-	if (pcount)
-		return 1;
-
-	return 0;
-}
-
-/**
- * \brief Cause device to go quiet so it can be safely removed/reset/etc
- * @param oct Pointer to Octeon device
+ * pcierror_quiesce_device - Cause device to go quiet so it can be safely removed/reset/etc
+ * @oct: Pointer to Octeon device
  */
 static void pcierror_quiesce_device(struct octeon_device *oct)
 {
@@ -218,8 +142,8 @@ static void pcierror_quiesce_device(struct octeon_device *oct)
 }
 
 /**
- * \brief Cleanup PCI AER uncorrectable error status
- * @param dev Pointer to PCI device
+ * cleanup_aer_uncorrect_error_status - Cleanup PCI AER uncorrectable error status
+ * @dev: Pointer to PCI device
  */
 static void cleanup_aer_uncorrect_error_status(struct pci_dev *dev)
 {
@@ -238,8 +162,8 @@ static void cleanup_aer_uncorrect_error_status(struct pci_dev *dev)
 }
 
 /**
- * \brief Stop all PCI IO to a given device
- * @param dev Pointer to Octeon device
+ * stop_pci_io - Stop all PCI IO to a given device
+ * @oct: Pointer to Octeon device
  */
 static void stop_pci_io(struct octeon_device *oct)
 {
@@ -280,9 +204,9 @@ static void stop_pci_io(struct octeon_device *oct)
 }
 
 /**
- * \brief called when PCI error is detected
- * @param pdev Pointer to PCI device
- * @param state The current pci connection state
+ * liquidio_pcie_error_detected - called when PCI error is detected
+ * @pdev: Pointer to PCI device
+ * @state: The current pci connection state
  *
  * This function is called after a PCI bus error affecting
  * this device has been detected.
@@ -331,250 +255,15 @@ static struct pci_driver liquidio_vf_pci_driver = {
 };
 
 /**
- * \brief Stop Tx queues
- * @param netdev network device
- */
-static void txqs_stop(struct net_device *netdev)
-{
-	if (netif_is_multiqueue(netdev)) {
-		int i;
-
-		for (i = 0; i < netdev->num_tx_queues; i++)
-			netif_stop_subqueue(netdev, i);
-	} else {
-		netif_stop_queue(netdev);
-	}
-}
-
-/**
- * \brief Start Tx queues
- * @param netdev network device
- */
-static void txqs_start(struct net_device *netdev)
-{
-	if (netif_is_multiqueue(netdev)) {
-		int i;
-
-		for (i = 0; i < netdev->num_tx_queues; i++)
-			netif_start_subqueue(netdev, i);
-	} else {
-		netif_start_queue(netdev);
-	}
-}
-
-/**
- * \brief Wake Tx queues
- * @param netdev network device
- */
-static void txqs_wake(struct net_device *netdev)
-{
-	struct lio *lio = GET_LIO(netdev);
-
-	if (netif_is_multiqueue(netdev)) {
-		int i;
-
-		for (i = 0; i < netdev->num_tx_queues; i++) {
-			int qno = lio->linfo.txpciq[i % (lio->linfo.num_txpciq)]
-				      .s.q_no;
-			if (__netif_subqueue_stopped(netdev, i)) {
-				INCR_INSTRQUEUE_PKT_COUNT(lio->oct_dev, qno,
-							  tx_restart, 1);
-				netif_wake_subqueue(netdev, i);
-			}
-		}
-	} else {
-		INCR_INSTRQUEUE_PKT_COUNT(lio->oct_dev, lio->txq,
-					  tx_restart, 1);
-		netif_wake_queue(netdev);
-	}
-}
-
-/**
- * \brief Start Tx queue
- * @param netdev network device
- */
-static void start_txq(struct net_device *netdev)
-{
-	struct lio *lio = GET_LIO(netdev);
-
-	if (lio->linfo.link.s.link_up) {
-		txqs_start(netdev);
-		return;
-	}
-}
-
-/**
- * \brief Wake a queue
- * @param netdev network device
- * @param q which queue to wake
- */
-static void wake_q(struct net_device *netdev, int q)
-{
-	if (netif_is_multiqueue(netdev))
-		netif_wake_subqueue(netdev, q);
-	else
-		netif_wake_queue(netdev);
-}
-
-/**
- * \brief Stop a queue
- * @param netdev network device
- * @param q which queue to stop
- */
-static void stop_q(struct net_device *netdev, int q)
-{
-	if (netif_is_multiqueue(netdev))
-		netif_stop_subqueue(netdev, q);
-	else
-		netif_stop_queue(netdev);
-}
-
-/**
- * Remove the node at the head of the list. The list would be empty at
- * the end of this call if there are no more nodes in the list.
- */
-static struct list_head *list_delete_head(struct list_head *root)
-{
-	struct list_head *node;
-
-	if ((root->prev == root) && (root->next == root))
-		node = NULL;
-	else
-		node = root->next;
-
-	if (node)
-		list_del(node);
-
-	return node;
-}
-
-/**
- * \brief Delete gather lists
- * @param lio per-network private data
- */
-static void delete_glists(struct lio *lio)
-{
-	struct octnic_gather *g;
-	int i;
-
-	kfree(lio->glist_lock);
-	lio->glist_lock = NULL;
-
-	if (!lio->glist)
-		return;
-
-	for (i = 0; i < lio->linfo.num_txpciq; i++) {
-		do {
-			g = (struct octnic_gather *)
-			    list_delete_head(&lio->glist[i]);
-			if (g)
-				kfree(g);
-		} while (g);
-
-		if (lio->glists_virt_base && lio->glists_virt_base[i] &&
-		    lio->glists_dma_base && lio->glists_dma_base[i]) {
-			lio_dma_free(lio->oct_dev,
-				     lio->glist_entry_size * lio->tx_qsize,
-				     lio->glists_virt_base[i],
-				     lio->glists_dma_base[i]);
-		}
-	}
-
-	kfree(lio->glists_virt_base);
-	lio->glists_virt_base = NULL;
-
-	kfree(lio->glists_dma_base);
-	lio->glists_dma_base = NULL;
-
-	kfree(lio->glist);
-	lio->glist = NULL;
-}
-
-/**
- * \brief Setup gather lists
- * @param lio per-network private data
- */
-static int setup_glists(struct lio *lio, int num_iqs)
-{
-	struct octnic_gather *g;
-	int i, j;
-
-	lio->glist_lock =
-	    kzalloc(sizeof(*lio->glist_lock) * num_iqs, GFP_KERNEL);
-	if (!lio->glist_lock)
-		return -ENOMEM;
-
-	lio->glist =
-	    kzalloc(sizeof(*lio->glist) * num_iqs, GFP_KERNEL);
-	if (!lio->glist) {
-		kfree(lio->glist_lock);
-		lio->glist_lock = NULL;
-		return -ENOMEM;
-	}
-
-	lio->glist_entry_size =
-		ROUNDUP8((ROUNDUP4(OCTNIC_MAX_SG) >> 2) * OCT_SG_ENTRY_SIZE);
-
-	/* allocate memory to store virtual and dma base address of
-	 * per glist consistent memory
-	 */
-	lio->glists_virt_base = kcalloc(num_iqs, sizeof(*lio->glists_virt_base),
-					GFP_KERNEL);
-	lio->glists_dma_base = kcalloc(num_iqs, sizeof(*lio->glists_dma_base),
-				       GFP_KERNEL);
-
-	if (!lio->glists_virt_base || !lio->glists_dma_base) {
-		delete_glists(lio);
-		return -ENOMEM;
-	}
-
-	for (i = 0; i < num_iqs; i++) {
-		spin_lock_init(&lio->glist_lock[i]);
-
-		INIT_LIST_HEAD(&lio->glist[i]);
-
-		lio->glists_virt_base[i] =
-			lio_dma_alloc(lio->oct_dev,
-				      lio->glist_entry_size * lio->tx_qsize,
-				      &lio->glists_dma_base[i]);
-
-		if (!lio->glists_virt_base[i]) {
-			delete_glists(lio);
-			return -ENOMEM;
-		}
-
-		for (j = 0; j < lio->tx_qsize; j++) {
-			g = kzalloc(sizeof(*g), GFP_KERNEL);
-			if (!g)
-				break;
-
-			g->sg = lio->glists_virt_base[i] +
-				(j * lio->glist_entry_size);
-
-			g->sg_dma_ptr = lio->glists_dma_base[i] +
-					(j * lio->glist_entry_size);
-
-			list_add_tail(&g->list, &lio->glist[i]);
-		}
-
-		if (j != lio->tx_qsize) {
-			delete_glists(lio);
-			return -ENOMEM;
-		}
-	}
-
-	return 0;
-}
-
-/**
- * \brief Print link information
- * @param netdev network device
+ * print_link_info - Print link information
+ * @netdev: network device
  */
 static void print_link_info(struct net_device *netdev)
 {
 	struct lio *lio = GET_LIO(netdev);
 
-	if (atomic_read(&lio->ifstate) & LIO_IFSTATE_REGISTERED) {
+	if (!ifstate_check(lio, LIO_IFSTATE_RESETTING) &&
+	    ifstate_check(lio, LIO_IFSTATE_REGISTERED)) {
 		struct oct_link_info *linfo = &lio->linfo;
 
 		if (linfo->link.s.link_up) {
@@ -588,22 +277,26 @@ static void print_link_info(struct net_device *netdev)
 }
 
 /**
- * \brief Routine to notify MTU change
- * @param work work_struct data structure
+ * octnet_link_status_change - Routine to notify MTU change
+ * @work: work_struct data structure
  */
 static void octnet_link_status_change(struct work_struct *work)
 {
 	struct cavium_wk *wk = (struct cavium_wk *)work;
 	struct lio *lio = (struct lio *)wk->ctxptr;
 
+	/* lio->linfo.link.s.mtu always contains max MTU of the lio interface.
+	 * this API is invoked only when new max-MTU of the interface is
+	 * less than current MTU.
+	 */
 	rtnl_lock();
-	call_netdevice_notifiers(NETDEV_CHANGEMTU, lio->netdev);
+	dev_set_mtu(lio->netdev, lio->linfo.link.s.mtu);
 	rtnl_unlock();
 }
 
 /**
- * \brief Sets up the mtu status change work
- * @param netdev network device
+ * setup_link_status_change_wq - Sets up the mtu status change work
+ * @netdev: network device
  */
 static int setup_link_status_change_wq(struct net_device *netdev)
 {
@@ -611,7 +304,8 @@ static int setup_link_status_change_wq(struct net_device *netdev)
 	struct octeon_device *oct = lio->oct_dev;
 
 	lio->link_status_wq.wq = alloc_workqueue("link-status",
-						 WQ_MEM_RECLAIM, 0);
+						 WQ_MEM_RECLAIM | WQ_PERCPU,
+						 0);
 	if (!lio->link_status_wq.wq) {
 		dev_err(&oct->pci_dev->dev, "unable to create cavium link status wq\n");
 		return -1;
@@ -634,9 +328,9 @@ static void cleanup_link_status_change_wq(struct net_device *netdev)
 }
 
 /**
- * \brief Update link status
- * @param netdev network device
- * @param ls link status structure
+ * update_link_status - Update link status
+ * @netdev: network device
+ * @ls: link status structure
  *
  * Called on receipt of a link status response from the core application to
  * update each interface's link status.
@@ -645,6 +339,7 @@ static void update_link_status(struct net_device *netdev,
 			       union oct_link_status *ls)
 {
 	struct lio *lio = GET_LIO(netdev);
+	int current_max_mtu = lio->linfo.link.s.mtu;
 	struct octeon_device *oct = lio->oct_dev;
 
 	if ((lio->intf_open) && (lio->linfo.link.u64 != ls->u64)) {
@@ -655,193 +350,37 @@ static void update_link_status(struct net_device *netdev,
 
 		if (lio->linfo.link.s.link_up) {
 			netif_carrier_on(netdev);
-			txqs_wake(netdev);
+			wake_txqs(netdev);
 		} else {
 			netif_carrier_off(netdev);
-			txqs_stop(netdev);
+			stop_txqs(netdev);
+		}
+
+		if (lio->linfo.link.s.mtu != current_max_mtu) {
+			dev_info(&oct->pci_dev->dev,
+				 "Max MTU Changed from %d to %d\n",
+				 current_max_mtu, lio->linfo.link.s.mtu);
+			netdev->max_mtu = lio->linfo.link.s.mtu;
 		}
 
 		if (lio->linfo.link.s.mtu < netdev->mtu) {
 			dev_warn(&oct->pci_dev->dev,
-				 "PF has changed the MTU for gmx port. Reducing the mtu from %d to %d\n",
+				 "Current MTU is higher than new max MTU; Reducing the current mtu from %d to %d\n",
 				 netdev->mtu, lio->linfo.link.s.mtu);
-			lio->mtu = lio->linfo.link.s.mtu;
-			netdev->mtu = lio->linfo.link.s.mtu;
 			queue_delayed_work(lio->link_status_wq.wq,
 					   &lio->link_status_wq.wk.work, 0);
 		}
 	}
 }
 
-static void update_txq_status(struct octeon_device *oct, int iq_num)
-{
-	struct octeon_instr_queue *iq = oct->instr_queue[iq_num];
-	struct net_device *netdev;
-	struct lio *lio;
-
-	netdev = oct->props[iq->ifidx].netdev;
-	lio = GET_LIO(netdev);
-	if (netif_is_multiqueue(netdev)) {
-		if (__netif_subqueue_stopped(netdev, iq->q_index) &&
-		    lio->linfo.link.s.link_up &&
-		    (!octnet_iq_is_full(oct, iq_num))) {
-			netif_wake_subqueue(netdev, iq->q_index);
-			INCR_INSTRQUEUE_PKT_COUNT(lio->oct_dev, iq_num,
-						  tx_restart, 1);
-		}
-	} else if (netif_queue_stopped(netdev) && lio->linfo.link.s.link_up &&
-		   (!octnet_iq_is_full(oct, lio->txq))) {
-		INCR_INSTRQUEUE_PKT_COUNT(lio->oct_dev,
-					  lio->txq, tx_restart, 1);
-		netif_wake_queue(netdev);
-	}
-}
-
-static
-int liquidio_schedule_msix_droq_pkt_handler(struct octeon_droq *droq, u64 ret)
-{
-	struct octeon_device *oct = droq->oct_dev;
-	struct octeon_device_priv *oct_priv =
-	    (struct octeon_device_priv *)oct->priv;
-
-	if (droq->ops.poll_mode) {
-		droq->ops.napi_fn(droq);
-	} else {
-		if (ret & MSIX_PO_INT) {
-			dev_err(&oct->pci_dev->dev,
-				"should not come here should not get rx when poll mode = 0 for vf\n");
-			tasklet_schedule(&oct_priv->droq_tasklet);
-			return 1;
-		}
-		/* this will be flushed periodically by check iq db */
-		if (ret & MSIX_PI_INT)
-			return 0;
-	}
-	return 0;
-}
-
-static irqreturn_t
-liquidio_msix_intr_handler(int irq __attribute__((unused)), void *dev)
-{
-	struct octeon_ioq_vector *ioq_vector = (struct octeon_ioq_vector *)dev;
-	struct octeon_device *oct = ioq_vector->oct_dev;
-	struct octeon_droq *droq = oct->droq[ioq_vector->droq_index];
-	u64 ret;
-
-	ret = oct->fn_list.msix_interrupt_handler(ioq_vector);
-
-	if ((ret & MSIX_PO_INT) || (ret & MSIX_PI_INT))
-		liquidio_schedule_msix_droq_pkt_handler(droq, ret);
-
-	return IRQ_HANDLED;
-}
-
 /**
- * \brief Setup interrupt for octeon device
- * @param oct octeon device
- *
- *  Enable interrupt in Octeon device as given in the PCI interrupt mask.
- */
-static int octeon_setup_interrupt(struct octeon_device *oct)
-{
-	struct msix_entry *msix_entries;
-	char *queue_irq_names = NULL;
-	int num_alloc_ioq_vectors;
-	int num_ioq_vectors;
-	int irqret;
-	int i;
-
-	if (oct->msix_on) {
-		oct->num_msix_irqs = oct->sriov_info.rings_per_vf;
-
-		/* allocate storage for the names assigned to each irq */
-		oct->irq_name_storage =
-			kcalloc(MAX_IOQ_INTERRUPTS_PER_VF, INTRNAMSIZ,
-				GFP_KERNEL);
-		if (!oct->irq_name_storage) {
-			dev_err(&oct->pci_dev->dev, "Irq name storage alloc failed...\n");
-			return -ENOMEM;
-		}
-
-		queue_irq_names = oct->irq_name_storage;
-
-		oct->msix_entries = kcalloc(
-		    oct->num_msix_irqs, sizeof(struct msix_entry), GFP_KERNEL);
-		if (!oct->msix_entries) {
-			dev_err(&oct->pci_dev->dev, "Memory Alloc failed...\n");
-			kfree(oct->irq_name_storage);
-			oct->irq_name_storage = NULL;
-			return -ENOMEM;
-		}
-
-		msix_entries = (struct msix_entry *)oct->msix_entries;
-
-		for (i = 0; i < oct->num_msix_irqs; i++)
-			msix_entries[i].entry = i;
-		num_alloc_ioq_vectors = pci_enable_msix_range(
-						oct->pci_dev, msix_entries,
-						oct->num_msix_irqs,
-						oct->num_msix_irqs);
-		if (num_alloc_ioq_vectors < 0) {
-			dev_err(&oct->pci_dev->dev, "unable to Allocate MSI-X interrupts\n");
-			kfree(oct->msix_entries);
-			oct->msix_entries = NULL;
-			kfree(oct->irq_name_storage);
-			oct->irq_name_storage = NULL;
-			return num_alloc_ioq_vectors;
-		}
-		dev_dbg(&oct->pci_dev->dev, "OCTEON: Enough MSI-X interrupts are allocated...\n");
-
-		num_ioq_vectors = oct->num_msix_irqs;
-
-		for (i = 0; i < num_ioq_vectors; i++) {
-			snprintf(&queue_irq_names[IRQ_NAME_OFF(i)], INTRNAMSIZ,
-				 "LiquidIO%u-vf%u-rxtx-%u",
-				 oct->octeon_id, oct->vf_num, i);
-
-			irqret = request_irq(msix_entries[i].vector,
-					     liquidio_msix_intr_handler, 0,
-					     &queue_irq_names[IRQ_NAME_OFF(i)],
-					     &oct->ioq_vector[i]);
-			if (irqret) {
-				dev_err(&oct->pci_dev->dev,
-					"OCTEON: Request_irq failed for MSIX interrupt Error: %d\n",
-					irqret);
-
-				while (i) {
-					i--;
-					irq_set_affinity_hint(
-					    msix_entries[i].vector, NULL);
-					free_irq(msix_entries[i].vector,
-						 &oct->ioq_vector[i]);
-				}
-				pci_disable_msix(oct->pci_dev);
-				kfree(oct->msix_entries);
-				oct->msix_entries = NULL;
-				kfree(oct->irq_name_storage);
-				oct->irq_name_storage = NULL;
-				return irqret;
-			}
-			oct->ioq_vector[i].vector = msix_entries[i].vector;
-			/* assign the cpu mask for this msix interrupt vector */
-			irq_set_affinity_hint(
-			    msix_entries[i].vector,
-			    (&oct->ioq_vector[i].affinity_mask));
-		}
-		dev_dbg(&oct->pci_dev->dev,
-			"OCTEON[%d]: MSI-X enabled\n", oct->octeon_id);
-	}
-	return 0;
-}
-
-/**
- * \brief PCI probe handler
- * @param pdev PCI device structure
- * @param ent unused
+ * liquidio_vf_probe - PCI probe handler
+ * @pdev: PCI device structure
+ * @ent: unused
  */
 static int
 liquidio_vf_probe(struct pci_dev *pdev,
-		  const struct pci_device_id *ent __attribute__((unused)))
+		  const struct pci_device_id __maybe_unused *ent)
 {
 	struct octeon_device *oct_dev = NULL;
 
@@ -863,6 +402,9 @@ liquidio_vf_probe(struct pci_dev *pdev,
 	/* set linux specific device pointer */
 	oct_dev->pci_dev = pdev;
 
+	oct_dev->subsystem_id = pdev->subsystem_vendor |
+		(pdev->subsystem_device << 16);
+
 	if (octeon_device_init(oct_dev)) {
 		liquidio_vf_remove(pdev);
 		return -ENOMEM;
@@ -874,8 +416,8 @@ liquidio_vf_probe(struct pci_dev *pdev,
 }
 
 /**
- * \brief PCI FLR for each Octeon device.
- * @param oct octeon device
+ * octeon_pci_flr - PCI FLR for each Octeon device.
+ * @oct: octeon device
  */
 static void octeon_pci_flr(struct octeon_device *oct)
 {
@@ -895,12 +437,12 @@ static void octeon_pci_flr(struct octeon_device *oct)
 }
 
 /**
- *\brief Destroy resources associated with octeon device
- * @param pdev PCI device structure
- * @param ent unused
+ * octeon_destroy_resources - Destroy resources associated with octeon device
+ * @oct: octeon device
  */
 static void octeon_destroy_resources(struct octeon_device *oct)
 {
+	struct octeon_device_priv *oct_priv = oct->priv;
 	struct msix_entry *msix_entries;
 	int i;
 
@@ -916,16 +458,15 @@ static void octeon_destroy_resources(struct octeon_device *oct)
 
 		schedule_timeout_uninterruptible(HZ / 10);
 
-		/* fallthrough */
+		fallthrough;
 	case OCT_DEV_HOST_OK:
-		/* fallthrough */
 	case OCT_DEV_IO_QUEUES_DONE:
-		if (wait_for_pending_requests(oct))
-			dev_err(&oct->pci_dev->dev, "There were pending requests\n");
-
 		if (lio_wait_for_instr_fetch(oct))
 			dev_err(&oct->pci_dev->dev, "IQ had pending instructions\n");
 
+		if (wait_for_pending_requests(oct))
+			dev_err(&oct->pci_dev->dev, "There were pending requests\n");
+
 		/* Disable the input and output queues now. No more packets will
 		 * arrive from Octeon, but we should wait for all packet
 		 * processing to finish.
@@ -935,6 +476,32 @@ static void octeon_destroy_resources(struct octeon_device *oct)
 		if (lio_wait_for_oq_pkts(oct))
 			dev_err(&oct->pci_dev->dev, "OQ had pending packets\n");
 
+		/* Force all requests waiting to be fetched by OCTEON to
+		 * complete.
+		 */
+		for (i = 0; i < MAX_OCTEON_INSTR_QUEUES(oct); i++) {
+			struct octeon_instr_queue *iq;
+
+			if (!(oct->io_qmask.iq & BIT_ULL(i)))
+				continue;
+			iq = oct->instr_queue[i];
+
+			if (atomic_read(&iq->instr_pending)) {
+				spin_lock_bh(&iq->lock);
+				iq->fill_cnt = 0;
+				iq->octeon_read_index = iq->host_write_index;
+				iq->stats.instr_processed +=
+					atomic_read(&iq->instr_pending);
+				lio_process_iq_request_list(oct, iq, 0);
+				spin_unlock_bh(&iq->lock);
+			}
+		}
+
+		lio_process_ordered_list(oct, 1);
+		octeon_free_sc_done_list(oct);
+		octeon_free_sc_zombie_list(oct);
+
+		fallthrough;
 	case OCT_DEV_INTR_SET_DONE:
 		/* Disable interrupts  */
 		oct->fn_list.disable_interrupt(oct, OCTEON_ALL_INTR);
@@ -942,10 +509,14 @@ static void octeon_destroy_resources(struct octeon_device *oct)
 		if (oct->msix_on) {
 			msix_entries = (struct msix_entry *)oct->msix_entries;
 			for (i = 0; i < oct->num_msix_irqs; i++) {
-				irq_set_affinity_hint(msix_entries[i].vector,
-						      NULL);
-				free_irq(msix_entries[i].vector,
-					 &oct->ioq_vector[i]);
+				if (oct->ioq_vector[i].vector) {
+					irq_set_affinity_hint(
+							msix_entries[i].vector,
+							NULL);
+					free_irq(msix_entries[i].vector,
+						 &oct->ioq_vector[i]);
+					oct->ioq_vector[i].vector = 0;
+				}
 			}
 			pci_disable_msix(oct->pci_dev);
 			kfree(oct->msix_entries);
@@ -954,20 +525,20 @@ static void octeon_destroy_resources(struct octeon_device *oct)
 			oct->irq_name_storage = NULL;
 		}
 		/* Soft reset the octeon device before exiting */
-		if (oct->pci_dev->reset_fn)
+		if (!pcie_reset_flr(oct->pci_dev, PCI_RESET_PROBE))
 			octeon_pci_flr(oct);
 		else
 			cn23xx_vf_ask_pf_to_do_flr(oct);
 
-		/* fallthrough */
+		fallthrough;
 	case OCT_DEV_MSIX_ALLOC_VECTOR_DONE:
 		octeon_free_ioq_vector(oct);
 
-		/* fallthrough */
+		fallthrough;
 	case OCT_DEV_MBOX_SETUP_DONE:
 		oct->fn_list.free_mbox(oct);
 
-		/* fallthrough */
+		fallthrough;
 	case OCT_DEV_IN_RESET:
 	case OCT_DEV_DROQ_INIT_DONE:
 		mdelay(100);
@@ -977,11 +548,11 @@ static void octeon_destroy_resources(struct octeon_device *oct)
 			octeon_delete_droq(oct, i);
 		}
 
-		/* fallthrough */
+		fallthrough;
 	case OCT_DEV_RESP_LIST_INIT_DONE:
 		octeon_delete_response_list(oct);
 
-		/* fallthrough */
+		fallthrough;
 	case OCT_DEV_INSTR_QUEUE_INIT_DONE:
 		for (i = 0; i < MAX_OCTEON_INSTR_QUEUES(oct); i++) {
 			if (!(oct->io_qmask.iq & BIT_ULL(i)))
@@ -989,87 +560,59 @@ static void octeon_destroy_resources(struct octeon_device *oct)
 			octeon_delete_instr_queue(oct, i);
 		}
 
-		/* fallthrough */
+		fallthrough;
 	case OCT_DEV_SC_BUFF_POOL_INIT_DONE:
 		octeon_free_sc_buffer_pool(oct);
 
-		/* fallthrough */
+		fallthrough;
 	case OCT_DEV_DISPATCH_INIT_DONE:
 		octeon_delete_dispatch_list(oct);
 		cancel_delayed_work_sync(&oct->nic_poll_work.work);
 
-		/* fallthrough */
+		fallthrough;
 	case OCT_DEV_PCI_MAP_DONE:
 		octeon_unmap_pci_barx(oct, 0);
 		octeon_unmap_pci_barx(oct, 1);
 
-		/* fallthrough */
+		fallthrough;
 	case OCT_DEV_PCI_ENABLE_DONE:
-		pci_clear_master(oct->pci_dev);
 		/* Disable the device, releasing the PCI INT */
 		pci_disable_device(oct->pci_dev);
 
-		/* fallthrough */
+		fallthrough;
 	case OCT_DEV_BEGIN_STATE:
 		/* Nothing to be done here either */
 		break;
 	}
-}
-
-/**
- * \brief Callback for rx ctrl
- * @param status status of request
- * @param buf pointer to resp structure
- */
-static void rx_ctl_callback(struct octeon_device *oct,
-			    u32 status, void *buf)
-{
-	struct octeon_soft_command *sc = (struct octeon_soft_command *)buf;
-	struct liquidio_rx_ctl_context *ctx;
-
-	ctx  = (struct liquidio_rx_ctl_context *)sc->ctxptr;
 
-	oct = lio_get_device(ctx->octeon_id);
-	if (status)
-		dev_err(&oct->pci_dev->dev, "rx ctl instruction failed. Status: %llx\n",
-			CVM_CAST64(status));
-	WRITE_ONCE(ctx->cond, 1);
-
-	/* This barrier is required to be sure that the response has been
-	 * written fully before waking up the handler
-	 */
-	wmb();
-
-	wake_up_interruptible(&ctx->wc);
+	tasklet_kill(&oct_priv->droq_tasklet);
 }
 
 /**
- * \brief Send Rx control command
- * @param lio per-network private data
- * @param start_stop whether to start or stop
+ * send_rx_ctrl_cmd - Send Rx control command
+ * @lio: per-network private data
+ * @start_stop: whether to start or stop
  */
-static void send_rx_ctrl_cmd(struct lio *lio, int start_stop)
+static int send_rx_ctrl_cmd(struct lio *lio, int start_stop)
 {
 	struct octeon_device *oct = (struct octeon_device *)lio->oct_dev;
-	int ctx_size = sizeof(struct liquidio_rx_ctl_context);
-	struct liquidio_rx_ctl_context *ctx;
 	struct octeon_soft_command *sc;
 	union octnet_cmd *ncmd;
 	int retval;
 
 	if (oct->props[lio->ifidx].rx_on == start_stop)
-		return;
+		return 0;
 
 	sc = (struct octeon_soft_command *)
 		octeon_alloc_soft_command(oct, OCTNET_CMD_SIZE,
-					  16, ctx_size);
+					  16, 0);
+	if (!sc) {
+		netif_info(lio, rx_err, lio->netdev,
+			   "Failed to allocate octeon_soft_command struct\n");
+		return -ENOMEM;
+	}
 
 	ncmd = (union octnet_cmd *)sc->virtdptr;
-	ctx  = (struct liquidio_rx_ctl_context *)sc->ctxptr;
-
-	WRITE_ONCE(ctx->cond, 0);
-	ctx->octeon_id = lio_get_device_id(oct);
-	init_waitqueue_head(&ctx->wc);
 
 	ncmd->u64 = 0;
 	ncmd->s.cmd = OCTNET_CMD_RX_CTL;
@@ -1082,29 +625,32 @@ static void send_rx_ctrl_cmd(struct lio *lio, int start_stop)
 	octeon_prepare_soft_command(oct, sc, OPCODE_NIC,
 				    OPCODE_NIC_CMD, 0, 0, 0);
 
-	sc->callback = rx_ctl_callback;
-	sc->callback_arg = sc;
-	sc->wait_time = 5000;
+	init_completion(&sc->complete);
+	sc->sc_status = OCTEON_REQUEST_PENDING;
 
 	retval = octeon_send_soft_command(oct, sc);
 	if (retval == IQ_SEND_FAILED) {
 		netif_info(lio, rx_err, lio->netdev, "Failed to send RX Control message\n");
+		octeon_free_soft_command(oct, sc);
 	} else {
 		/* Sleep on a wait queue till the cond flag indicates that the
 		 * response arrived or timed-out.
 		 */
-		if (sleep_cond(&ctx->wc, &ctx->cond) == -EINTR)
-			return;
+		retval = wait_for_sc_completion_timeout(oct, sc, 0);
+		if (retval)
+			return retval;
+
 		oct->props[lio->ifidx].rx_on = start_stop;
+		WRITE_ONCE(sc->caller_is_done, true);
 	}
 
-	octeon_free_soft_command(oct, sc);
+	return retval;
 }
 
 /**
- * \brief Destroy NIC device interface
- * @param oct octeon device
- * @param ifidx which interface to destroy
+ * liquidio_destroy_nic_device - Destroy NIC device interface
+ * @oct: octeon device
+ * @ifidx: which interface to destroy
  *
  * Cleanup associated with each interface for an Octeon device  when NIC
  * module is being unloaded or if initialization fails during load.
@@ -1112,6 +658,7 @@ static void send_rx_ctrl_cmd(struct lio *lio, int start_stop)
 static void liquidio_destroy_nic_device(struct octeon_device *oct, int ifidx)
 {
 	struct net_device *netdev = oct->props[ifidx].netdev;
+	struct octeon_device_priv *oct_priv = oct->priv;
 	struct napi_struct *napi, *n;
 	struct lio *lio;
 
@@ -1137,6 +684,12 @@ static void liquidio_destroy_nic_device(struct octeon_device *oct, int ifidx)
 		oct->droq[0]->ops.poll_mode = 0;
 	}
 
+	/* Delete NAPI */
+	list_for_each_entry_safe(napi, n, &netdev->napi_list, dev_list)
+		netif_napi_del(napi);
+
+	tasklet_enable(&oct_priv->droq_tasklet);
+
 	if (atomic_read(&lio->ifstate) & LIO_IFSTATE_REGISTERED)
 		unregister_netdev(netdev);
 
@@ -1144,7 +697,7 @@ static void liquidio_destroy_nic_device(struct octeon_device *oct, int ifidx)
 
 	cleanup_link_status_change_wq(netdev);
 
-	delete_glists(lio);
+	lio_delete_glists(lio);
 
 	free_netdev(netdev);
 
@@ -1154,8 +707,8 @@ static void liquidio_destroy_nic_device(struct octeon_device *oct, int ifidx)
 }
 
 /**
- * \brief Stop complete NIC functionality
- * @param oct octeon device
+ * liquidio_stop_nic_module - Stop complete NIC functionality
+ * @oct: octeon device
  */
 static int liquidio_stop_nic_module(struct octeon_device *oct)
 {
@@ -1174,7 +727,7 @@ static int liquidio_stop_nic_module(struct octeon_device *oct)
 
 	for (i = 0; i < oct->ifcount; i++) {
 		lio = GET_LIO(oct->props[i].netdev);
-		for (j = 0; j < lio->linfo.num_rxpciq; j++)
+		for (j = 0; j < oct->num_oqs; j++)
 			octeon_unregister_droq_ops(oct,
 						   lio->linfo.rxpciq[j].s.q_no);
 	}
@@ -1187,8 +740,8 @@ static int liquidio_stop_nic_module(struct octeon_device *oct)
 }
 
 /**
- * \brief Cleans up resources at unload time
- * @param pdev PCI device structure
+ * liquidio_vf_remove - Cleans up resources at unload time
+ * @pdev: PCI device structure
  */
 static void liquidio_vf_remove(struct pci_dev *pdev)
 {
@@ -1213,8 +766,8 @@ static void liquidio_vf_remove(struct pci_dev *pdev)
 }
 
 /**
- * \brief PCI initialization for each Octeon device.
- * @param oct octeon device
+ * octeon_pci_os_setup - PCI initialization for each Octeon device.
+ * @oct: octeon device
  */
 static int octeon_pci_os_setup(struct octeon_device *oct)
 {
@@ -1241,47 +794,9 @@ static int octeon_pci_os_setup(struct octeon_device *oct)
 	return 0;
 }
 
-static int skb_iq(struct lio *lio, struct sk_buff *skb)
-{
-	int q = 0;
-
-	if (netif_is_multiqueue(lio->netdev))
-		q = skb->queue_mapping % lio->linfo.num_txpciq;
-
-	return q;
-}
-
 /**
- * \brief Check Tx queue state for a given network buffer
- * @param lio per-network private data
- * @param skb network buffer
- */
-static int check_txq_state(struct lio *lio, struct sk_buff *skb)
-{
-	int q = 0, iq = 0;
-
-	if (netif_is_multiqueue(lio->netdev)) {
-		q = skb->queue_mapping;
-		iq = lio->linfo.txpciq[(q % (lio->linfo.num_txpciq))].s.q_no;
-	} else {
-		iq = lio->txq;
-		q = iq;
-	}
-
-	if (octnet_iq_is_full(lio->oct_dev, iq))
-		return 0;
-
-	if (__netif_subqueue_stopped(lio->netdev, q)) {
-		INCR_INSTRQUEUE_PKT_COUNT(lio->oct_dev, iq, tx_restart, 1);
-		wake_q(lio->netdev, q);
-	}
-
-	return 1;
-}
-
-/**
- * \brief Unmap and free network buffer
- * @param buf buffer
+ * free_netbuf - Unmap and free network buffer
+ * @buf: buffer
  */
 static void free_netbuf(void *buf)
 {
@@ -1296,14 +811,12 @@ static void free_netbuf(void *buf)
 	dma_unmap_single(&lio->oct_dev->pci_dev->dev, finfo->dptr, skb->len,
 			 DMA_TO_DEVICE);
 
-	check_txq_state(lio, skb);
-
 	tx_buffer_free(skb);
 }
 
 /**
- * \brief Unmap and free gather buffer
- * @param buf buffer
+ * free_netsgbuf - Unmap and free gather buffer
+ * @buf: buffer
  */
 static void free_netsgbuf(void *buf)
 {
@@ -1325,28 +838,26 @@ static void free_netsgbuf(void *buf)
 
 	i = 1;
 	while (frags--) {
-		struct skb_frag_struct *frag = &skb_shinfo(skb)->frags[i - 1];
+		skb_frag_t *frag = &skb_shinfo(skb)->frags[i - 1];
 
-		pci_unmap_page((lio->oct_dev)->pci_dev,
+		dma_unmap_page(&lio->oct_dev->pci_dev->dev,
 			       g->sg[(i >> 2)].ptr[(i & 3)],
-			       frag->size, DMA_TO_DEVICE);
+			       skb_frag_size(frag), DMA_TO_DEVICE);
 		i++;
 	}
 
-	iq = skb_iq(lio, skb);
+	iq = skb_iq(lio->oct_dev, skb);
 
 	spin_lock(&lio->glist_lock[iq]);
 	list_add_tail(&g->list, &lio->glist[iq]);
 	spin_unlock(&lio->glist_lock[iq]);
 
-	check_txq_state(lio, skb); /* mq support: sub-queue state check */
-
 	tx_buffer_free(skb);
 }
 
 /**
- * \brief Unmap and free gather buffer with response
- * @param buf buffer
+ * free_netsgbuf_with_resp - Unmap and free gather buffer with response
+ * @buf: buffer
  */
 static void free_netsgbuf_with_resp(void *buf)
 {
@@ -1371,387 +882,38 @@ static void free_netsgbuf_with_resp(void *buf)
 
 	i = 1;
 	while (frags--) {
-		struct skb_frag_struct *frag = &skb_shinfo(skb)->frags[i - 1];
+		skb_frag_t *frag = &skb_shinfo(skb)->frags[i - 1];
 
-		pci_unmap_page((lio->oct_dev)->pci_dev,
+		dma_unmap_page(&lio->oct_dev->pci_dev->dev,
 			       g->sg[(i >> 2)].ptr[(i & 3)],
-			       frag->size, DMA_TO_DEVICE);
+			       skb_frag_size(frag), DMA_TO_DEVICE);
 		i++;
 	}
 
-	iq = skb_iq(lio, skb);
+	iq = skb_iq(lio->oct_dev, skb);
 
 	spin_lock(&lio->glist_lock[iq]);
 	list_add_tail(&g->list, &lio->glist[iq]);
 	spin_unlock(&lio->glist_lock[iq]);
 
 	/* Don't free the skb yet */
-
-	check_txq_state(lio, skb);
 }
 
 /**
- * \brief Setup output queue
- * @param oct octeon device
- * @param q_no which queue
- * @param num_descs how many descriptors
- * @param desc_size size of each descriptor
- * @param app_ctx application context
- */
-static int octeon_setup_droq(struct octeon_device *oct, int q_no, int num_descs,
-			     int desc_size, void *app_ctx)
-{
-	int ret_val;
-
-	dev_dbg(&oct->pci_dev->dev, "Creating Droq: %d\n", q_no);
-	/* droq creation and local register settings. */
-	ret_val = octeon_create_droq(oct, q_no, num_descs, desc_size, app_ctx);
-	if (ret_val < 0)
-		return ret_val;
-
-	if (ret_val == 1) {
-		dev_dbg(&oct->pci_dev->dev, "Using default droq %d\n", q_no);
-		return 0;
-	}
-
-	/* Enable the droq queues */
-	octeon_set_droq_pkt_op(oct, q_no, 1);
-
-	/* Send Credit for Octeon Output queues. Credits are always
-	 * sent after the output queue is enabled.
-	 */
-	writel(oct->droq[q_no]->max_count, oct->droq[q_no]->pkts_credit_reg);
-
-	return ret_val;
-}
-
-/**
- * \brief Callback for getting interface configuration
- * @param status status of request
- * @param buf pointer to resp structure
- */
-static void if_cfg_callback(struct octeon_device *oct,
-			    u32 status __attribute__((unused)), void *buf)
-{
-	struct octeon_soft_command *sc = (struct octeon_soft_command *)buf;
-	struct liquidio_if_cfg_context *ctx;
-	struct liquidio_if_cfg_resp *resp;
-
-	resp = (struct liquidio_if_cfg_resp *)sc->virtrptr;
-	ctx = (struct liquidio_if_cfg_context *)sc->ctxptr;
-
-	oct = lio_get_device(ctx->octeon_id);
-	if (resp->status)
-		dev_err(&oct->pci_dev->dev, "nic if cfg instruction failed. Status: %llx\n",
-			CVM_CAST64(resp->status));
-	WRITE_ONCE(ctx->cond, 1);
-
-	snprintf(oct->fw_info.liquidio_firmware_version, 32, "%s",
-		 resp->cfg_info.liquidio_firmware_version);
-
-	/* This barrier is required to be sure that the response has been
-	 * written fully before waking up the handler
-	 */
-	wmb();
-
-	wake_up_interruptible(&ctx->wc);
-}
-
-/** Routine to push packets arriving on Octeon interface upto network layer.
- * @param oct_id   - octeon device id.
- * @param skbuff   - skbuff struct to be passed to network layer.
- * @param len      - size of total data received.
- * @param rh       - Control header associated with the packet
- * @param param    - additional control data with the packet
- * @param arg      - farg registered in droq_ops
- */
-static void
-liquidio_push_packet(u32 octeon_id __attribute__((unused)),
-		     void *skbuff,
-		     u32 len,
-		     union octeon_rh *rh,
-		     void *param,
-		     void *arg)
-{
-	struct napi_struct *napi = param;
-	struct octeon_droq *droq =
-		container_of(param, struct octeon_droq, napi);
-	struct net_device *netdev = (struct net_device *)arg;
-	struct sk_buff *skb = (struct sk_buff *)skbuff;
-	u16 vtag = 0;
-	u32 r_dh_off;
-
-	if (netdev) {
-		struct lio *lio = GET_LIO(netdev);
-		int packet_was_received;
-
-		/* Do not proceed if the interface is not in RUNNING state. */
-		if (!ifstate_check(lio, LIO_IFSTATE_RUNNING)) {
-			recv_buffer_free(skb);
-			droq->stats.rx_dropped++;
-			return;
-		}
-
-		skb->dev = netdev;
-
-		skb_record_rx_queue(skb, droq->q_no);
-		if (likely(len > MIN_SKB_SIZE)) {
-			struct octeon_skb_page_info *pg_info;
-			unsigned char *va;
-
-			pg_info = ((struct octeon_skb_page_info *)(skb->cb));
-			if (pg_info->page) {
-				/* For Paged allocation use the frags */
-				va = page_address(pg_info->page) +
-					pg_info->page_offset;
-				memcpy(skb->data, va, MIN_SKB_SIZE);
-				skb_put(skb, MIN_SKB_SIZE);
-				skb_add_rx_frag(skb, skb_shinfo(skb)->nr_frags,
-						pg_info->page,
-						pg_info->page_offset +
-						MIN_SKB_SIZE,
-						len - MIN_SKB_SIZE,
-						LIO_RXBUFFER_SZ);
-			}
-		} else {
-			struct octeon_skb_page_info *pg_info =
-				((struct octeon_skb_page_info *)(skb->cb));
-			skb_copy_to_linear_data(skb,
-						page_address(pg_info->page) +
-						pg_info->page_offset, len);
-			skb_put(skb, len);
-			put_page(pg_info->page);
-		}
-
-		r_dh_off = (rh->r_dh.len - 1) * BYTES_PER_DHLEN_UNIT;
-
-		if (rh->r_dh.has_hwtstamp)
-			r_dh_off -= BYTES_PER_DHLEN_UNIT;
-
-		if (rh->r_dh.has_hash) {
-			__be32 *hash_be = (__be32 *)(skb->data + r_dh_off);
-			u32 hash = be32_to_cpu(*hash_be);
-
-			skb_set_hash(skb, hash, PKT_HASH_TYPE_L4);
-			r_dh_off -= BYTES_PER_DHLEN_UNIT;
-		}
-
-		skb_pull(skb, rh->r_dh.len * BYTES_PER_DHLEN_UNIT);
-		skb->protocol = eth_type_trans(skb, skb->dev);
-
-		if ((netdev->features & NETIF_F_RXCSUM) &&
-		    (((rh->r_dh.encap_on) &&
-		      (rh->r_dh.csum_verified & CNNIC_TUN_CSUM_VERIFIED)) ||
-		     (!(rh->r_dh.encap_on) &&
-		      (rh->r_dh.csum_verified & CNNIC_CSUM_VERIFIED))))
-			/* checksum has already been verified */
-			skb->ip_summed = CHECKSUM_UNNECESSARY;
-		else
-			skb->ip_summed = CHECKSUM_NONE;
-
-		/* Setting Encapsulation field on basis of status received
-		 * from the firmware
-		 */
-		if (rh->r_dh.encap_on) {
-			skb->encapsulation = 1;
-			skb->csum_level = 1;
-			droq->stats.rx_vxlan++;
-		}
-
-		/* inbound VLAN tag */
-		if ((netdev->features & NETIF_F_HW_VLAN_CTAG_RX) &&
-		    rh->r_dh.vlan) {
-			u16 priority = rh->r_dh.priority;
-			u16 vid = rh->r_dh.vlan;
-
-			vtag = (priority << VLAN_PRIO_SHIFT) | vid;
-			__vlan_hwaccel_put_tag(skb, htons(ETH_P_8021Q), vtag);
-		}
-
-		packet_was_received = (napi_gro_receive(napi, skb) != GRO_DROP);
-
-		if (packet_was_received) {
-			droq->stats.rx_bytes_received += len;
-			droq->stats.rx_pkts_received++;
-		} else {
-			droq->stats.rx_dropped++;
-			netif_info(lio, rx_err, lio->netdev,
-				   "droq:%d  error rx_dropped:%llu\n",
-				   droq->q_no, droq->stats.rx_dropped);
-		}
-
-	} else {
-		recv_buffer_free(skb);
-	}
-}
-
-/**
- * \brief callback when receive interrupt occurs and we are in NAPI mode
- * @param arg pointer to octeon output queue
- */
-static void liquidio_vf_napi_drv_callback(void *arg)
-{
-	struct octeon_droq *droq = arg;
-
-	napi_schedule_irqoff(&droq->napi);
-}
-
-/**
- * \brief Entry point for NAPI polling
- * @param napi NAPI structure
- * @param budget maximum number of items to process
- */
-static int liquidio_napi_poll(struct napi_struct *napi, int budget)
-{
-	struct octeon_instr_queue *iq;
-	struct octeon_device *oct;
-	struct octeon_droq *droq;
-	int tx_done = 0, iq_no;
-	int work_done;
-
-	droq = container_of(napi, struct octeon_droq, napi);
-	oct = droq->oct_dev;
-	iq_no = droq->q_no;
-
-	/* Handle Droq descriptors */
-	work_done = octeon_process_droq_poll_cmd(oct, droq->q_no,
-						 POLL_EVENT_PROCESS_PKTS,
-						 budget);
-
-	/* Flush the instruction queue */
-	iq = oct->instr_queue[iq_no];
-	if (iq) {
-		if (atomic_read(&iq->instr_pending))
-			/* Process iq buffers with in the budget limits */
-			tx_done = octeon_flush_iq(oct, iq, budget);
-		else
-			tx_done = 1;
-
-		/* Update iq read-index rather than waiting for next interrupt.
-		 * Return back if tx_done is false.
-		 */
-		update_txq_status(oct, iq_no);
-	} else {
-		dev_err(&oct->pci_dev->dev, "%s: iq (%d) num invalid\n",
-			__func__, iq_no);
-	}
-
-	/* force enable interrupt if reg cnts are high to avoid wraparound */
-	if ((work_done < budget && tx_done) ||
-	    (iq && iq->pkt_in_done >= MAX_REG_CNT) ||
-	    (droq->pkt_count >= MAX_REG_CNT)) {
-		tx_done = 1;
-		napi_complete_done(napi, work_done);
-		octeon_process_droq_poll_cmd(droq->oct_dev, droq->q_no,
-					     POLL_EVENT_ENABLE_INTR, 0);
-		return 0;
-	}
-
-	return (!tx_done) ? (budget) : (work_done);
-}
-
-/**
- * \brief Setup input and output queues
- * @param octeon_dev octeon device
- * @param ifidx Interface index
- *
- * Note: Queues are with respect to the octeon device. Thus
- * an input queue is for egress packets, and output queues
- * are for ingress packets.
- */
-static int setup_io_queues(struct octeon_device *octeon_dev, int ifidx)
-{
-	struct octeon_droq_ops droq_ops;
-	struct net_device *netdev;
-	static int cpu_id_modulus;
-	struct octeon_droq *droq;
-	struct napi_struct *napi;
-	static int cpu_id;
-	int num_tx_descs;
-	struct lio *lio;
-	int retval = 0;
-	int q, q_no;
-
-	netdev = octeon_dev->props[ifidx].netdev;
-
-	lio = GET_LIO(netdev);
-
-	memset(&droq_ops, 0, sizeof(struct octeon_droq_ops));
-
-	droq_ops.fptr = liquidio_push_packet;
-	droq_ops.farg = netdev;
-
-	droq_ops.poll_mode = 1;
-	droq_ops.napi_fn = liquidio_vf_napi_drv_callback;
-	cpu_id = 0;
-	cpu_id_modulus = num_present_cpus();
-
-	/* set up DROQs. */
-	for (q = 0; q < lio->linfo.num_rxpciq; q++) {
-		q_no = lio->linfo.rxpciq[q].s.q_no;
-
-		retval = octeon_setup_droq(
-		    octeon_dev, q_no,
-		    CFG_GET_NUM_RX_DESCS_NIC_IF(octeon_get_conf(octeon_dev),
-						lio->ifidx),
-		    CFG_GET_NUM_RX_BUF_SIZE_NIC_IF(octeon_get_conf(octeon_dev),
-						   lio->ifidx),
-		    NULL);
-		if (retval) {
-			dev_err(&octeon_dev->pci_dev->dev,
-				"%s : Runtime DROQ(RxQ) creation failed.\n",
-				__func__);
-			return 1;
-		}
-
-		droq = octeon_dev->droq[q_no];
-		napi = &droq->napi;
-		netif_napi_add(netdev, napi, liquidio_napi_poll, 64);
-
-		/* designate a CPU for this droq */
-		droq->cpu_id = cpu_id;
-		cpu_id++;
-		if (cpu_id >= cpu_id_modulus)
-			cpu_id = 0;
-
-		octeon_register_droq_ops(octeon_dev, q_no, &droq_ops);
-	}
-
-	/* 23XX VF can send/recv control messages (via the first VF-owned
-	 * droq) from the firmware even if the ethX interface is down,
-	 * so that's why poll_mode must be off for the first droq.
-	 */
-	octeon_dev->droq[0]->ops.poll_mode = 0;
-
-	/* set up IQs. */
-	for (q = 0; q < lio->linfo.num_txpciq; q++) {
-		num_tx_descs = CFG_GET_NUM_TX_DESCS_NIC_IF(
-		    octeon_get_conf(octeon_dev), lio->ifidx);
-		retval = octeon_setup_iq(octeon_dev, ifidx, q,
-					 lio->linfo.txpciq[q], num_tx_descs,
-					 netdev_get_tx_queue(netdev, q));
-		if (retval) {
-			dev_err(&octeon_dev->pci_dev->dev,
-				" %s : Runtime IQ(TxQ) creation failed.\n",
-				__func__);
-			return 1;
-		}
-	}
-
-	return 0;
-}
-
-/**
- * \brief Net device open for LiquidIO
- * @param netdev network device
+ * liquidio_open - Net device open for LiquidIO
+ * @netdev: network device
  */
 static int liquidio_open(struct net_device *netdev)
 {
 	struct lio *lio = GET_LIO(netdev);
 	struct octeon_device *oct = lio->oct_dev;
+	struct octeon_device_priv *oct_priv = oct->priv;
 	struct napi_struct *napi, *n;
+	int ret = 0;
 
 	if (!oct->props[lio->ifidx].napi_enabled) {
+		tasklet_disable(&oct_priv->droq_tasklet);
+
 		list_for_each_entry_safe(napi, n, &netdev->napi_list, dev_list)
 			napi_enable(napi);
 
@@ -1766,24 +928,39 @@ static int liquidio_open(struct net_device *netdev)
 	lio->intf_open = 1;
 
 	netif_info(lio, ifup, lio->netdev, "Interface Open, ready for traffic\n");
-	start_txq(netdev);
+	start_txqs(netdev);
+
+	INIT_DELAYED_WORK(&lio->stats_wk.work, lio_fetch_stats);
+	lio->stats_wk.ctxptr = lio;
+	schedule_delayed_work(&lio->stats_wk.work, msecs_to_jiffies
+					(LIQUIDIO_NDEV_STATS_POLL_TIME_MS));
 
 	/* tell Octeon to start forwarding packets to host */
-	send_rx_ctrl_cmd(lio, 1);
+	ret = send_rx_ctrl_cmd(lio, 1);
+	if (ret)
+		return ret;
 
 	dev_info(&oct->pci_dev->dev, "%s interface is opened\n", netdev->name);
 
-	return 0;
+	return ret;
 }
 
 /**
- * \brief Net device stop for LiquidIO
- * @param netdev network device
+ * liquidio_stop - jNet device stop for LiquidIO
+ * @netdev: network device
  */
 static int liquidio_stop(struct net_device *netdev)
 {
 	struct lio *lio = GET_LIO(netdev);
 	struct octeon_device *oct = lio->oct_dev;
+	struct octeon_device_priv *oct_priv = oct->priv;
+	struct napi_struct *napi, *n;
+	int ret = 0;
+
+	/* tell Octeon to stop forwarding packets to host */
+	ret = send_rx_ctrl_cmd(lio, 0);
+	if (ret)
+		return ret;
 
 	netif_info(lio, ifdown, lio->netdev, "Stopping interface!\n");
 	/* Inform that netif carrier is down */
@@ -1793,21 +970,36 @@ static int liquidio_stop(struct net_device *netdev)
 	netif_carrier_off(netdev);
 	lio->link_changes++;
 
-	/* tell Octeon to stop forwarding packets to host */
-	send_rx_ctrl_cmd(lio, 0);
-
 	ifstate_reset(lio, LIO_IFSTATE_RUNNING);
 
-	txqs_stop(netdev);
+	stop_txqs(netdev);
+
+	/* Wait for any pending Rx descriptors */
+	if (lio_wait_for_clean_oq(oct))
+		netif_info(lio, rx_err, lio->netdev,
+			   "Proceeding with stop interface after partial RX desc processing\n");
+
+	if (oct->props[lio->ifidx].napi_enabled == 1) {
+		list_for_each_entry_safe(napi, n, &netdev->napi_list, dev_list)
+			napi_disable(napi);
+
+		oct->props[lio->ifidx].napi_enabled = 0;
+
+		oct->droq[0]->ops.poll_mode = 0;
+
+		tasklet_enable(&oct_priv->droq_tasklet);
+	}
+
+	cancel_delayed_work_sync(&lio->stats_wk.work);
 
 	dev_info(&oct->pci_dev->dev, "%s interface is stopped\n", netdev->name);
 
-	return 0;
+	return ret;
 }
 
 /**
- * \brief Converts a mask based on net device flags
- * @param netdev network device
+ * get_new_flags - Converts a mask based on net device flags
+ * @netdev: network device
  *
  * This routine generates a octnet_ifflags mask from the net device flags
  * received from the OS.
@@ -1875,8 +1067,8 @@ static void liquidio_set_uc_list(struct net_device *netdev)
 }
 
 /**
- * \brief Net device set_multicast_list
- * @param netdev network device
+ * liquidio_set_mcast_list - Net device set_multicast_list
+ * @netdev: network device
  */
 static void liquidio_set_mcast_list(struct net_device *netdev)
 {
@@ -1914,10 +1106,9 @@ static void liquidio_set_mcast_list(struct net_device *netdev)
 	/* Apparently, any activity in this call from the kernel has to
 	 * be atomic. So we won't wait for response.
 	 */
-	nctrl.wait_time = 0;
 
 	ret = octnet_send_nic_ctrl_pkt(lio->oct_dev, &nctrl);
-	if (ret < 0) {
+	if (ret) {
 		dev_err(&oct->pci_dev->dev, "DEVFLAGS change failed in core (ret: 0x%x)\n",
 			ret);
 	}
@@ -1926,8 +1117,9 @@ static void liquidio_set_mcast_list(struct net_device *netdev)
 }
 
 /**
- * \brief Net device set_mac_address
- * @param netdev network device
+ * liquidio_set_mac - Net device set_mac_address
+ * @netdev: network device
+ * @p: opaque pointer to sockaddr
  */
 static int liquidio_set_mac(struct net_device *netdev, void *p)
 {
@@ -1954,8 +1146,6 @@ static int liquidio_set_mac(struct net_device *netdev, void *p)
 	nctrl.ncmd.s.more = 1;
 	nctrl.iq_no = lio->linfo.txpciq[0].s.q_no;
 	nctrl.netpndev = (u64)netdev;
-	nctrl.cb_fn = liquidio_link_ctrl_cmd_completion;
-	nctrl.wait_time = 100;
 
 	nctrl.udd[0] = 0;
 	/* The MAC Address is presented in network byte order. */
@@ -1966,29 +1156,36 @@ static int liquidio_set_mac(struct net_device *netdev, void *p)
 		dev_err(&oct->pci_dev->dev, "MAC Address change failed\n");
 		return -ENOMEM;
 	}
-	memcpy(netdev->dev_addr, addr->sa_data, netdev->addr_len);
+
+	if (nctrl.sc_status ==
+	    FIRMWARE_STATUS_CODE(OCTEON_REQUEST_NO_PERMISSION)) {
+		dev_err(&oct->pci_dev->dev, "MAC Address change failed: no permission\n");
+		return -EPERM;
+	}
+
+	eth_hw_addr_set(netdev, addr->sa_data);
 	ether_addr_copy(((u8 *)&lio->linfo.hw_addr) + 2, addr->sa_data);
 
 	return 0;
 }
 
-/**
- * \brief Net device get_stats
- * @param netdev network device
- */
-static struct net_device_stats *liquidio_get_stats(struct net_device *netdev)
+static void
+liquidio_get_stats64(struct net_device *netdev,
+		     struct rtnl_link_stats64 *lstats)
 {
 	struct lio *lio = GET_LIO(netdev);
-	struct net_device_stats *stats = &netdev->stats;
+	struct octeon_device *oct;
 	u64 pkts = 0, drop = 0, bytes = 0;
 	struct oct_droq_stats *oq_stats;
 	struct oct_iq_stats *iq_stats;
-	struct octeon_device *oct;
 	int i, iq_no, oq_no;
 
 	oct = lio->oct_dev;
 
-	for (i = 0; i < lio->linfo.num_txpciq; i++) {
+	if (ifstate_check(lio, LIO_IFSTATE_RESETTING))
+		return;
+
+	for (i = 0; i < oct->num_iqs; i++) {
 		iq_no = lio->linfo.txpciq[i].s.q_no;
 		iq_stats = &oct->instr_queue[iq_no]->stats;
 		pkts += iq_stats->tx_done;
@@ -1996,15 +1193,15 @@ static struct net_device_stats *liquidio_get_stats(struct net_device *netdev)
 		bytes += iq_stats->tx_tot_bytes;
 	}
 
-	stats->tx_packets = pkts;
-	stats->tx_bytes = bytes;
-	stats->tx_dropped = drop;
+	lstats->tx_packets = pkts;
+	lstats->tx_bytes = bytes;
+	lstats->tx_dropped = drop;
 
 	pkts = 0;
 	drop = 0;
 	bytes = 0;
 
-	for (i = 0; i < lio->linfo.num_rxpciq; i++) {
+	for (i = 0; i < oct->num_oqs; i++) {
 		oq_no = lio->linfo.rxpciq[i].s.q_no;
 		oq_stats = &oct->droq[oq_no]->stats;
 		pkts += oq_stats->rx_pkts_received;
@@ -2015,39 +1212,34 @@ static struct net_device_stats *liquidio_get_stats(struct net_device *netdev)
 		bytes += oq_stats->rx_bytes_received;
 	}
 
-	stats->rx_bytes = bytes;
-	stats->rx_packets = pkts;
-	stats->rx_dropped = drop;
+	lstats->rx_bytes = bytes;
+	lstats->rx_packets = pkts;
+	lstats->rx_dropped = drop;
 
-	return stats;
-}
+	lstats->multicast = oct->link_stats.fromwire.fw_total_mcast;
 
-/**
- * \brief Net device change_mtu
- * @param netdev network device
- */
-static int liquidio_change_mtu(struct net_device *netdev, int new_mtu)
-{
-	struct lio *lio = GET_LIO(netdev);
-	struct octeon_device *oct = lio->oct_dev;
+	/* detailed rx_errors: */
+	lstats->rx_length_errors = oct->link_stats.fromwire.l2_err;
+	/* recved pkt with crc error */
+	lstats->rx_crc_errors = oct->link_stats.fromwire.fcs_err;
+	/* recv'd frame alignment error */
+	lstats->rx_frame_errors = oct->link_stats.fromwire.frame_err;
 
-	lio->mtu = new_mtu;
+	lstats->rx_errors = lstats->rx_length_errors + lstats->rx_crc_errors +
+			    lstats->rx_frame_errors;
 
-	netif_info(lio, probe, lio->netdev, "MTU Changed from %d to %d\n",
-		   netdev->mtu, new_mtu);
-	dev_info(&oct->pci_dev->dev, "%s MTU Changed from %d to %d\n",
-		 netdev->name, netdev->mtu, new_mtu);
+	/* detailed tx_errors */
+	lstats->tx_aborted_errors = oct->link_stats.fromhost.fw_err_pko;
+	lstats->tx_carrier_errors = oct->link_stats.fromhost.fw_err_link;
 
-	netdev->mtu = new_mtu;
-
-	return 0;
+	lstats->tx_errors = lstats->tx_aborted_errors +
+		lstats->tx_carrier_errors;
 }
 
 /**
- * \brief Handler for SIOCSHWTSTAMP ioctl
- * @param netdev network device
- * @param ifr interface request
- * @param cmd command
+ * hwtstamp_ioctl - Handler for SIOCSHWTSTAMP ioctl
+ * @netdev: network device
+ * @ifr: interface request
  */
 static int hwtstamp_ioctl(struct net_device *netdev, struct ifreq *ifr)
 {
@@ -2057,9 +1249,6 @@ static int hwtstamp_ioctl(struct net_device *netdev, struct ifreq *ifr)
 	if (copy_from_user(&conf, ifr->ifr_data, sizeof(conf)))
 		return -EFAULT;
 
-	if (conf.flags)
-		return -EINVAL;
-
 	switch (conf.tx_type) {
 	case HWTSTAMP_TX_ON:
 	case HWTSTAMP_TX_OFF:
@@ -2102,10 +1291,10 @@ static int hwtstamp_ioctl(struct net_device *netdev, struct ifreq *ifr)
 }
 
 /**
- * \brief ioctl handler
- * @param netdev network device
- * @param ifr interface request
- * @param cmd command
+ * liquidio_ioctl - ioctl handler
+ * @netdev: network device
+ * @ifr: interface request
+ * @cmd: command
  */
 static int liquidio_ioctl(struct net_device *netdev, struct ifreq *ifr, int cmd)
 {
@@ -2154,14 +1343,15 @@ static void handle_timestamp(struct octeon_device *oct, u32 status, void *buf)
 	tx_buffer_free(skb);
 }
 
-/* \brief Send a data packet that will be timestamped
- * @param oct octeon device
- * @param ndata pointer to network data
- * @param finfo pointer to private network data
+/* send_nic_timestamp_pkt - Send a data packet that will be timestamped
+ * @oct: octeon device
+ * @ndata: pointer to network data
+ * @finfo: pointer to private network data
  */
 static int send_nic_timestamp_pkt(struct octeon_device *oct,
 				  struct octnic_data_pkt *ndata,
-				  struct octnet_buf_free_info *finfo)
+				  struct octnet_buf_free_info *finfo,
+				  int xmit_more)
 {
 	struct octeon_soft_command *sc;
 	int ring_doorbell;
@@ -2191,7 +1381,7 @@ static int send_nic_timestamp_pkt(struct octeon_device *oct,
 
 	len = (u32)((struct octeon_instr_ih3 *)(&sc->cmd.cmd3.ih3))->dlengsz;
 
-	ring_doorbell = 1;
+	ring_doorbell = !xmit_more;
 
 	retval = octeon_send_command(oct, sc->iq_no, ring_doorbell, &sc->cmd,
 				     sc, len, ndata->reqtype);
@@ -2207,13 +1397,14 @@ static int send_nic_timestamp_pkt(struct octeon_device *oct,
 	return retval;
 }
 
-/** \brief Transmit networks packets to the Octeon interface
- * @param skbuff   skbuff struct to be passed to network layer.
- * @param netdev   pointer to network device
+/**
+ * liquidio_xmit - Transmit networks packets to the Octeon interface
+ * @skb: skbuff struct to be passed to network layer.
+ * @netdev: pointer to network device
  * @returns whether the packet was transmitted to the device okay or not
  *             (NETDEV_TX_OK or NETDEV_TX_BUSY)
  */
-static int liquidio_xmit(struct sk_buff *skb, struct net_device *netdev)
+static netdev_tx_t liquidio_xmit(struct sk_buff *skb, struct net_device *netdev)
 {
 	struct octnet_buf_free_info *finfo;
 	union octnic_cmd_setup cmdsetup;
@@ -2223,6 +1414,7 @@ static int liquidio_xmit(struct sk_buff *skb, struct net_device *netdev)
 	struct octeon_device *oct;
 	int q_idx = 0, iq_no = 0;
 	union tx_info *tx_info;
+	int xmit_more = 0;
 	struct lio *lio;
 	int status = 0;
 	u64 dptr = 0;
@@ -2232,14 +1424,9 @@ static int liquidio_xmit(struct sk_buff *skb, struct net_device *netdev)
 	lio = GET_LIO(netdev);
 	oct = lio->oct_dev;
 
-	if (netif_is_multiqueue(netdev)) {
-		q_idx = skb->queue_mapping;
-		q_idx = (q_idx % (lio->linfo.num_txpciq));
-		tag = q_idx;
-		iq_no = lio->linfo.txpciq[q_idx].s.q_no;
-	} else {
-		iq_no = lio->txq;
-	}
+	q_idx = skb_iq(lio->oct_dev, skb);
+	tag = q_idx;
+	iq_no = lio->linfo.txpciq[q_idx].s.q_no;
 
 	stats = &oct->instr_queue[iq_no]->stats;
 
@@ -2268,22 +1455,12 @@ static int liquidio_xmit(struct sk_buff *skb, struct net_device *netdev)
 
 	ndata.q_no = iq_no;
 
-	if (netif_is_multiqueue(netdev)) {
-		if (octnet_iq_is_full(oct, ndata.q_no)) {
-			/* defer sending if queue is full */
-			netif_info(lio, tx_err, lio->netdev, "Transmit failed iq:%d full\n",
-				   ndata.q_no);
-			stats->tx_iq_busy++;
-			return NETDEV_TX_BUSY;
-		}
-	} else {
-		if (octnet_iq_is_full(oct, lio->txq)) {
-			/* defer sending if queue is full */
-			stats->tx_iq_busy++;
-			netif_info(lio, tx_err, lio->netdev, "Transmit failed iq:%d full\n",
-				   ndata.q_no);
-			return NETDEV_TX_BUSY;
-		}
+	if (octnet_iq_is_full(oct, ndata.q_no)) {
+		/* defer sending if queue is full */
+		netif_info(lio, tx_err, lio->netdev, "Transmit failed iq:%d full\n",
+			   ndata.q_no);
+		stats->tx_iq_busy++;
+		return NETDEV_TX_BUSY;
 	}
 
 	ndata.datasize = skb->len;
@@ -2323,13 +1500,13 @@ static int liquidio_xmit(struct sk_buff *skb, struct net_device *netdev)
 		ndata.reqtype = REQTYPE_NORESP_NET;
 
 	} else {
-		struct skb_frag_struct *frag;
+		skb_frag_t *frag;
 		struct octnic_gather *g;
 		int i, frags;
 
 		spin_lock(&lio->glist_lock[q_idx]);
-		g = (struct octnic_gather *)list_delete_head(
-		    &lio->glist[q_idx]);
+		g = (struct octnic_gather *)
+			lio_list_delete_head(&lio->glist[q_idx]);
 		spin_unlock(&lio->glist_lock[q_idx]);
 
 		if (!g) {
@@ -2361,11 +1538,9 @@ static int liquidio_xmit(struct sk_buff *skb, struct net_device *netdev)
 			frag = &skb_shinfo(skb)->frags[i - 1];
 
 			g->sg[(i >> 2)].ptr[(i & 3)] =
-				dma_map_page(&oct->pci_dev->dev,
-					     frag->page.p,
-					     frag->page_offset,
-					     frag->size,
-					     DMA_TO_DEVICE);
+				skb_frag_dma_map(&oct->pci_dev->dev,
+						 frag, 0, skb_frag_size(frag),
+						 DMA_TO_DEVICE);
 			if (dma_mapping_error(&oct->pci_dev->dev,
 					      g->sg[i >> 2].ptr[i & 3])) {
 				dma_unmap_single(&oct->pci_dev->dev,
@@ -2376,7 +1551,7 @@ static int liquidio_xmit(struct sk_buff *skb, struct net_device *netdev)
 					frag = &skb_shinfo(skb)->frags[j - 1];
 					dma_unmap_page(&oct->pci_dev->dev,
 						       g->sg[j >> 2].ptr[j & 3],
-						       frag->size,
+						       skb_frag_size(frag),
 						       DMA_TO_DEVICE);
 				}
 				dev_err(&oct->pci_dev->dev, "%s DMA mapping error 3\n",
@@ -2384,7 +1559,8 @@ static int liquidio_xmit(struct sk_buff *skb, struct net_device *netdev)
 				return NETDEV_TX_BUSY;
 			}
 
-			add_sg_size(&g->sg[(i >> 2)], frag->size, (i & 3));
+			add_sg_size(&g->sg[(i >> 2)], skb_frag_size(frag),
+				    (i & 3));
 			i++;
 		}
 
@@ -2411,10 +1587,12 @@ static int liquidio_xmit(struct sk_buff *skb, struct net_device *netdev)
 		irh->vlan = skb_vlan_tag_get(skb) & VLAN_VID_MASK;
 	}
 
+	xmit_more = netdev_xmit_more();
+
 	if (unlikely(cmdsetup.s.timestamp))
-		status = send_nic_timestamp_pkt(oct, &ndata, finfo);
+		status = send_nic_timestamp_pkt(oct, &ndata, finfo, xmit_more);
 	else
-		status = octnet_send_nic_data_pkt(oct, &ndata);
+		status = octnet_send_nic_data_pkt(oct, &ndata, xmit_more);
 	if (status == IQ_SEND_FAILED)
 		goto lio_xmit_failed;
 
@@ -2423,7 +1601,7 @@ static int liquidio_xmit(struct sk_buff *skb, struct net_device *netdev)
 	if (status == IQ_SEND_STOP) {
 		dev_err(&oct->pci_dev->dev, "Rcvd IQ_SEND_STOP signal; stopping IQ-%d\n",
 			iq_no);
-		stop_q(lio->netdev, q_idx);
+		netif_stop_subqueue(netdev, q_idx);
 	}
 
 	netif_trans_update(netdev);
@@ -2443,14 +1621,19 @@ lio_xmit_failed:
 	if (dptr)
 		dma_unmap_single(&oct->pci_dev->dev, dptr,
 				 ndata.datasize, DMA_TO_DEVICE);
+
+	octeon_ring_doorbell_locked(oct, iq_no);
+
 	tx_buffer_free(skb);
 	return NETDEV_TX_OK;
 }
 
-/** \brief Network device Tx timeout
- * @param netdev    pointer to network device
+/**
+ * liquidio_tx_timeout - Network device Tx timeout
+ * @netdev: pointer to network device
+ * @txqueue: index of the hung transmit queue
  */
-static void liquidio_tx_timeout(struct net_device *netdev)
+static void liquidio_tx_timeout(struct net_device *netdev, unsigned int txqueue)
 {
 	struct lio *lio;
 
@@ -2460,7 +1643,7 @@ static void liquidio_tx_timeout(struct net_device *netdev)
 		   "Transmit timeout tx_dropped:%ld, waking up queues now!!\n",
 		   netdev->stats.tx_dropped);
 	netif_trans_update(netdev);
-	txqs_wake(netdev);
+	wake_txqs(netdev);
 }
 
 static int
@@ -2470,8 +1653,6 @@ liquidio_vlan_rx_add_vid(struct net_device *netdev,
 	struct lio *lio = GET_LIO(netdev);
 	struct octeon_device *oct = lio->oct_dev;
 	struct octnic_ctrl_pkt nctrl;
-	struct completion compl;
-	u16 response_code;
 	int ret = 0;
 
 	memset(&nctrl, 0, sizeof(struct octnic_ctrl_pkt));
@@ -2480,26 +1661,15 @@ liquidio_vlan_rx_add_vid(struct net_device *netdev,
 	nctrl.ncmd.s.cmd = OCTNET_CMD_ADD_VLAN_FILTER;
 	nctrl.ncmd.s.param1 = vid;
 	nctrl.iq_no = lio->linfo.txpciq[0].s.q_no;
-	nctrl.wait_time = 100;
 	nctrl.netpndev = (u64)netdev;
 	nctrl.cb_fn = liquidio_link_ctrl_cmd_completion;
-	init_completion(&compl);
-	nctrl.completion = &compl;
-	nctrl.response_code = &response_code;
 
 	ret = octnet_send_nic_ctrl_pkt(lio->oct_dev, &nctrl);
-	if (ret < 0) {
+	if (ret) {
 		dev_err(&oct->pci_dev->dev, "Add VLAN filter failed in core (ret: 0x%x)\n",
 			ret);
-		return -EIO;
-	}
-
-	if (!wait_for_completion_timeout(&compl,
-					 msecs_to_jiffies(nctrl.wait_time)))
-		return -EPERM;
-
-	if (READ_ONCE(response_code))
 		return -EPERM;
+	}
 
 	return 0;
 }
@@ -2519,14 +1689,15 @@ liquidio_vlan_rx_kill_vid(struct net_device *netdev,
 	nctrl.ncmd.s.cmd = OCTNET_CMD_DEL_VLAN_FILTER;
 	nctrl.ncmd.s.param1 = vid;
 	nctrl.iq_no = lio->linfo.txpciq[0].s.q_no;
-	nctrl.wait_time = 100;
 	nctrl.netpndev = (u64)netdev;
 	nctrl.cb_fn = liquidio_link_ctrl_cmd_completion;
 
 	ret = octnet_send_nic_ctrl_pkt(lio->oct_dev, &nctrl);
-	if (ret < 0) {
-		dev_err(&oct->pci_dev->dev, "Add VLAN filter failed in core (ret: 0x%x)\n",
+	if (ret) {
+		dev_err(&oct->pci_dev->dev, "Del VLAN filter failed in core (ret: 0x%x)\n",
 			ret);
+		if (ret > 0)
+			ret = -EIO;
 	}
 	return ret;
 }
@@ -2552,14 +1723,15 @@ static int liquidio_set_rxcsum_command(struct net_device *netdev, int command,
 	nctrl.ncmd.s.cmd = command;
 	nctrl.ncmd.s.param1 = rx_cmd;
 	nctrl.iq_no = lio->linfo.txpciq[0].s.q_no;
-	nctrl.wait_time = 100;
 	nctrl.netpndev = (u64)netdev;
 	nctrl.cb_fn = liquidio_link_ctrl_cmd_completion;
 
 	ret = octnet_send_nic_ctrl_pkt(lio->oct_dev, &nctrl);
-	if (ret < 0) {
+	if (ret) {
 		dev_err(&oct->pci_dev->dev, "DEVFLAGS RXCSUM change failed in core (ret:0x%x)\n",
 			ret);
+		if (ret > 0)
+			ret = -EIO;
 	}
 	return ret;
 }
@@ -2587,19 +1759,49 @@ static int liquidio_vxlan_port_command(struct net_device *netdev, int command,
 	nctrl.ncmd.s.more = vxlan_cmd_bit;
 	nctrl.ncmd.s.param1 = vxlan_port;
 	nctrl.iq_no = lio->linfo.txpciq[0].s.q_no;
-	nctrl.wait_time = 100;
 	nctrl.netpndev = (u64)netdev;
 	nctrl.cb_fn = liquidio_link_ctrl_cmd_completion;
 
 	ret = octnet_send_nic_ctrl_pkt(lio->oct_dev, &nctrl);
-	if (ret < 0) {
+	if (ret) {
 		dev_err(&oct->pci_dev->dev,
 			"DEVFLAGS VxLAN port add/delete failed in core (ret : 0x%x)\n",
 			ret);
+		if (ret > 0)
+			ret = -EIO;
 	}
 	return ret;
 }
 
+static int liquidio_udp_tunnel_set_port(struct net_device *netdev,
+					unsigned int table, unsigned int entry,
+					struct udp_tunnel_info *ti)
+{
+	return liquidio_vxlan_port_command(netdev,
+					   OCTNET_CMD_VXLAN_PORT_CONFIG,
+					   htons(ti->port),
+					   OCTNET_CMD_VXLAN_PORT_ADD);
+}
+
+static int liquidio_udp_tunnel_unset_port(struct net_device *netdev,
+					  unsigned int table,
+					  unsigned int entry,
+					  struct udp_tunnel_info *ti)
+{
+	return liquidio_vxlan_port_command(netdev,
+					   OCTNET_CMD_VXLAN_PORT_CONFIG,
+					   htons(ti->port),
+					   OCTNET_CMD_VXLAN_PORT_DEL);
+}
+
+static const struct udp_tunnel_nic_info liquidio_udp_tunnels = {
+	.set_port	= liquidio_udp_tunnel_set_port,
+	.unset_port	= liquidio_udp_tunnel_unset_port,
+	.tables		= {
+		{ .n_entries = 1024, .tunnel_types = UDP_TUNNEL_TYPE_VXLAN, },
+	},
+};
+
 /** \brief Net device fix features
  * @param netdev  pointer to network device
  * @param request features requested
@@ -2668,46 +1870,20 @@ static int liquidio_set_features(struct net_device *netdev,
 	return 0;
 }
 
-static void liquidio_add_vxlan_port(struct net_device *netdev,
-				    struct udp_tunnel_info *ti)
-{
-	if (ti->type != UDP_TUNNEL_TYPE_VXLAN)
-		return;
-
-	liquidio_vxlan_port_command(netdev,
-				    OCTNET_CMD_VXLAN_PORT_CONFIG,
-				    htons(ti->port),
-				    OCTNET_CMD_VXLAN_PORT_ADD);
-}
-
-static void liquidio_del_vxlan_port(struct net_device *netdev,
-				    struct udp_tunnel_info *ti)
-{
-	if (ti->type != UDP_TUNNEL_TYPE_VXLAN)
-		return;
-
-	liquidio_vxlan_port_command(netdev,
-				    OCTNET_CMD_VXLAN_PORT_CONFIG,
-				    htons(ti->port),
-				    OCTNET_CMD_VXLAN_PORT_DEL);
-}
-
 static const struct net_device_ops lionetdevops = {
 	.ndo_open		= liquidio_open,
 	.ndo_stop		= liquidio_stop,
 	.ndo_start_xmit		= liquidio_xmit,
-	.ndo_get_stats		= liquidio_get_stats,
+	.ndo_get_stats64	= liquidio_get_stats64,
 	.ndo_set_mac_address	= liquidio_set_mac,
 	.ndo_set_rx_mode	= liquidio_set_mcast_list,
 	.ndo_tx_timeout		= liquidio_tx_timeout,
 	.ndo_vlan_rx_add_vid    = liquidio_vlan_rx_add_vid,
 	.ndo_vlan_rx_kill_vid   = liquidio_vlan_rx_kill_vid,
 	.ndo_change_mtu		= liquidio_change_mtu,
-	.ndo_do_ioctl		= liquidio_ioctl,
+	.ndo_eth_ioctl		= liquidio_ioctl,
 	.ndo_fix_features	= liquidio_fix_features,
 	.ndo_set_features	= liquidio_set_features,
-	.ndo_udp_tunnel_add     = liquidio_add_vxlan_port,
-	.ndo_udp_tunnel_del     = liquidio_del_vxlan_port,
 };
 
 static int lio_nic_info(struct octeon_recv_info *recv_info, void *buf)
@@ -2718,7 +1894,7 @@ static int lio_nic_info(struct octeon_recv_info *recv_info, void *buf)
 	int gmxport = 0;
 	int i;
 
-	if (recv_pkt->buffer_size[0] != sizeof(*ls)) {
+	if (recv_pkt->buffer_size[0] != (sizeof(*ls) + OCT_DROQ_INFO_SIZE)) {
 		dev_err(&oct->pci_dev->dev, "Malformed NIC_INFO, len=%d, ifidx=%d\n",
 			recv_pkt->buffer_size[0],
 			recv_pkt->rh.r_nic_info.gmxport);
@@ -2726,7 +1902,8 @@ static int lio_nic_info(struct octeon_recv_info *recv_info, void *buf)
 	}
 
 	gmxport = recv_pkt->rh.r_nic_info.gmxport;
-	ls = (union oct_link_status *)get_rbd(recv_pkt->buffer_ptr[0]);
+	ls = (union oct_link_status *)(get_rbd(recv_pkt->buffer_ptr[0]) +
+		OCT_DROQ_INFO_SIZE);
 
 	octeon_swap_8B_data((u64 *)ls, (sizeof(union oct_link_status)) >> 3);
 
@@ -2745,8 +1922,8 @@ nic_info_err:
 }
 
 /**
- * \brief Setup network interfaces
- * @param octeon_dev  octeon device
+ * setup_nic_devices - Setup network interfaces
+ * @octeon_dev:  octeon device
  *
  * Called during init time for each device. It assumes the NIC
  * is already up and running.  The link information for each
@@ -2755,8 +1932,7 @@ nic_info_err:
 static int setup_nic_devices(struct octeon_device *octeon_dev)
 {
 	int retval, num_iqueues, num_oqueues;
-	struct liquidio_if_cfg_context *ctx;
-	u32 resp_size, ctx_size, data_size;
+	u32 resp_size, data_size;
 	struct liquidio_if_cfg_resp *resp;
 	struct octeon_soft_command *sc;
 	union oct_nic_if_cfg if_cfg;
@@ -2787,13 +1963,11 @@ static int setup_nic_devices(struct octeon_device *octeon_dev)
 
 	for (i = 0; i < octeon_dev->ifcount; i++) {
 		resp_size = sizeof(struct liquidio_if_cfg_resp);
-		ctx_size = sizeof(struct liquidio_if_cfg_context);
 		data_size = sizeof(struct lio_version);
 		sc = (struct octeon_soft_command *)
 			octeon_alloc_soft_command(octeon_dev, data_size,
-						  resp_size, ctx_size);
+						  resp_size, 0);
 		resp = (struct liquidio_if_cfg_resp *)sc->virtrptr;
-		ctx  = (struct liquidio_if_cfg_context *)sc->ctxptr;
 		vdata = (struct lio_version *)sc->virtdptr;
 
 		*((u64 *)vdata) = 0;
@@ -2801,10 +1975,6 @@ static int setup_nic_devices(struct octeon_device *octeon_dev)
 		vdata->minor = cpu_to_be16(LIQUIDIO_BASE_MINOR_VERSION);
 		vdata->micro = cpu_to_be16(LIQUIDIO_BASE_MICRO_VERSION);
 
-		WRITE_ONCE(ctx->cond, 0);
-		ctx->octeon_id = lio_get_device_id(octeon_dev);
-		init_waitqueue_head(&ctx->wc);
-
 		if_cfg.u64 = 0;
 
 		if_cfg.s.num_iqueues = octeon_dev->sriov_info.rings_per_vf;
@@ -2817,32 +1987,37 @@ static int setup_nic_devices(struct octeon_device *octeon_dev)
 					    OPCODE_NIC_IF_CFG, 0, if_cfg.u64,
 					    0);
 
-		sc->callback = if_cfg_callback;
-		sc->callback_arg = sc;
-		sc->wait_time = 5000;
+		init_completion(&sc->complete);
+		sc->sc_status = OCTEON_REQUEST_PENDING;
 
 		retval = octeon_send_soft_command(octeon_dev, sc);
 		if (retval == IQ_SEND_FAILED) {
 			dev_err(&octeon_dev->pci_dev->dev,
 				"iq/oq config failed status: %x\n", retval);
 			/* Soft instr is freed by driver in case of failure. */
-			goto setup_nic_dev_fail;
+			octeon_free_soft_command(octeon_dev, sc);
+			return(-EIO);
 		}
 
 		/* Sleep on a wait queue till the cond flag indicates that the
 		 * response arrived or timed-out.
 		 */
-		if (sleep_cond(&ctx->wc, &ctx->cond) == -EINTR) {
-			dev_err(&octeon_dev->pci_dev->dev, "Wait interrupted\n");
-			goto setup_nic_wait_intr;
-		}
+		retval = wait_for_sc_completion_timeout(octeon_dev, sc, 0);
+		if (retval)
+			return retval;
 
 		retval = resp->status;
 		if (retval) {
-			dev_err(&octeon_dev->pci_dev->dev, "iq/oq config failed\n");
-			goto setup_nic_dev_fail;
+			dev_err(&octeon_dev->pci_dev->dev,
+				"iq/oq config failed, retval = %d\n", retval);
+			WRITE_ONCE(sc->caller_is_done, true);
+			return -EIO;
 		}
 
+		snprintf(octeon_dev->fw_info.liquidio_firmware_version,
+			 32, "%s",
+			 resp->cfg_info.liquidio_firmware_version);
+
 		octeon_swap_8B_data((u64 *)(&resp->cfg_info),
 				    (sizeof(struct liquidio_if_cfg_info)) >> 3);
 
@@ -2853,7 +2028,8 @@ static int setup_nic_devices(struct octeon_device *octeon_dev)
 			dev_err(&octeon_dev->pci_dev->dev,
 				"Got bad iqueues (%016llx) or oqueues (%016llx) from firmware.\n",
 				resp->cfg_info.iqmask, resp->cfg_info.oqmask);
-			goto setup_nic_dev_fail;
+			WRITE_ONCE(sc->caller_is_done, true);
+			goto setup_nic_dev_done;
 		}
 		dev_dbg(&octeon_dev->pci_dev->dev,
 			"interface %d, iqmask %016llx, oqmask %016llx, numiqueues %d, numoqueues %d\n",
@@ -2864,7 +2040,8 @@ static int setup_nic_devices(struct octeon_device *octeon_dev)
 
 		if (!netdev) {
 			dev_err(&octeon_dev->pci_dev->dev, "Device allocation failed\n");
-			goto setup_nic_dev_fail;
+			WRITE_ONCE(sc->caller_is_done, true);
+			goto setup_nic_dev_done;
 		}
 
 		SET_NETDEV_DEV(netdev, &octeon_dev->pci_dev->dev);
@@ -2901,6 +2078,8 @@ static int setup_nic_devices(struct octeon_device *octeon_dev)
 		lio->linfo.link.u64 = resp->cfg_info.linfo.link.u64;
 		lio->linfo.macaddr_is_admin_asgnd =
 			resp->cfg_info.linfo.macaddr_is_admin_asgnd;
+		lio->linfo.macaddr_spoofchk =
+			resp->cfg_info.linfo.macaddr_spoofchk;
 
 		lio->msg_enable = netif_msg_init(debug, DEFAULT_MSG_ENABLE);
 
@@ -2910,7 +2089,7 @@ static int setup_nic_devices(struct octeon_device *octeon_dev)
 				      | NETIF_F_TSO | NETIF_F_TSO6
 				      | NETIF_F_GRO
 				      | NETIF_F_LRO;
-		netif_set_gso_max_size(netdev, OCTNIC_GSO_MAX_SIZE);
+		netif_set_tso_max_size(netdev, OCTNIC_GSO_MAX_SIZE);
 
 		/* Copy of transmit encapsulation capabilities:
 		 * TSO, TSO6, Checksums for this device
@@ -2925,6 +2104,8 @@ static int setup_nic_devices(struct octeon_device *octeon_dev)
 
 		netdev->hw_enc_features =
 		    (lio->enc_dev_capability & ~NETIF_F_LRO);
+		netdev->udp_tunnel_nic_info = &liquidio_udp_tunnels;
+
 		netdev->vlan_features = lio->dev_capability;
 		/* Add any unchangeable hw features */
 		lio->dev_capability |= NETIF_F_HW_VLAN_CTAG_FILTER |
@@ -2934,11 +2115,14 @@ static int setup_nic_devices(struct octeon_device *octeon_dev)
 		netdev->features = (lio->dev_capability & ~NETIF_F_LRO);
 
 		netdev->hw_features = lio->dev_capability;
+		netdev->hw_features &= ~NETIF_F_HW_VLAN_CTAG_RX;
 
 		/* MTU range: 68 - 16000 */
 		netdev->min_mtu = LIO_MIN_MTU_SIZE;
 		netdev->max_mtu = LIO_MAX_MTU_SIZE;
 
+		WRITE_ONCE(sc->caller_is_done, true);
+
 		/* Point to the  properties for octeon device to which this
 		 * interface belongs.
 		 */
@@ -2956,11 +2140,13 @@ static int setup_nic_devices(struct octeon_device *octeon_dev)
 			mac[j] = *((u8 *)(((u8 *)&lio->linfo.hw_addr) + 2 + j));
 
 		/* Copy MAC Address to OS network device structure */
-		ether_addr_copy(netdev->dev_addr, mac);
+		eth_hw_addr_set(netdev, mac);
 
-		if (setup_io_queues(octeon_dev, i)) {
+		if (liquidio_setup_io_queues(octeon_dev, i,
+					     lio->linfo.num_txpciq,
+					     lio->linfo.num_rxpciq)) {
 			dev_err(&octeon_dev->pci_dev->dev, "I/O queues creation failed\n");
-			goto setup_nic_dev_fail;
+			goto setup_nic_dev_free;
 		}
 
 		ifstate_set(lio, LIO_IFSTATE_DROQ_OPS);
@@ -2980,10 +2166,10 @@ static int setup_nic_devices(struct octeon_device *octeon_dev)
 		lio->tx_qsize = octeon_get_tx_qsize(octeon_dev, lio->txq);
 		lio->rx_qsize = octeon_get_rx_qsize(octeon_dev, lio->rxq);
 
-		if (setup_glists(lio, num_iqueues)) {
+		if (lio_setup_glists(octeon_dev, lio, num_iqueues)) {
 			dev_err(&octeon_dev->pci_dev->dev,
 				"Gather list allocation failed\n");
-			goto setup_nic_dev_fail;
+			goto setup_nic_dev_free;
 		}
 
 		/* Register ethtool support */
@@ -2997,20 +2183,16 @@ static int setup_nic_devices(struct octeon_device *octeon_dev)
 			liquidio_set_feature(netdev, OCTNET_CMD_LRO_ENABLE,
 					     OCTNIC_LROIPV4 | OCTNIC_LROIPV6);
 
-		if ((debug != -1) && (debug & NETIF_MSG_HW))
-			liquidio_set_feature(netdev, OCTNET_CMD_VERBOSE_ENABLE,
-					     0);
-
 		if (setup_link_status_change_wq(netdev))
-			goto setup_nic_dev_fail;
+			goto setup_nic_dev_free;
 
 		if (setup_rx_oom_poll_fn(netdev))
-			goto setup_nic_dev_fail;
+			goto setup_nic_dev_free;
 
 		/* Register the network device with the OS */
 		if (register_netdev(netdev)) {
 			dev_err(&octeon_dev->pci_dev->dev, "Device registration failed\n");
-			goto setup_nic_dev_fail;
+			goto setup_nic_dev_free;
 		}
 
 		dev_dbg(&octeon_dev->pci_dev->dev,
@@ -3033,28 +2215,27 @@ static int setup_nic_devices(struct octeon_device *octeon_dev)
 		dev_dbg(&octeon_dev->pci_dev->dev,
 			"NIC ifidx:%d Setup successful\n", i);
 
-		octeon_free_soft_command(octeon_dev, sc);
+		octeon_dev->no_speed_setting = 1;
 	}
 
 	return 0;
 
-setup_nic_dev_fail:
-
-	octeon_free_soft_command(octeon_dev, sc);
-
-setup_nic_wait_intr:
+setup_nic_dev_free:
 
 	while (i--) {
 		dev_err(&octeon_dev->pci_dev->dev,
 			"NIC ifidx:%d Setup failed\n", i);
 		liquidio_destroy_nic_device(octeon_dev, i);
 	}
+
+setup_nic_dev_done:
+
 	return -ENODEV;
 }
 
 /**
- * \brief initialize the NIC
- * @param oct octeon device
+ * liquidio_init_nic_module - initialize the NIC
+ * @oct: octeon device
  *
  * This initialization routine is called once the Octeon device application is
  * up and running
@@ -3094,8 +2275,8 @@ octnet_init_failure:
 }
 
 /**
- * \brief Device initialization for each Octeon device that is probed
- * @param octeon_dev  octeon device
+ * octeon_device_init - Device initialization for each Octeon device that is probed
+ * @oct:  octeon device
  */
 static int octeon_device_init(struct octeon_device *oct)
 {
@@ -3175,26 +2356,41 @@ static int octeon_device_init(struct octeon_device *oct)
 	}
 	atomic_set(&oct->status, OCT_DEV_MBOX_SETUP_DONE);
 
-	if (octeon_allocate_ioq_vector(oct)) {
+	if (octeon_allocate_ioq_vector(oct, oct->sriov_info.rings_per_vf)) {
 		dev_err(&oct->pci_dev->dev, "ioq vector allocation failed\n");
 		return 1;
 	}
 	atomic_set(&oct->status, OCT_DEV_MSIX_ALLOC_VECTOR_DONE);
 
-	dev_info(&oct->pci_dev->dev, "OCTEON_CN23XX VF Version: %s, %d ioqs\n",
-		 LIQUIDIO_VERSION, oct->sriov_info.rings_per_vf);
+	dev_info(&oct->pci_dev->dev, "OCTEON_CN23XX VF: %d ioqs\n",
+		 oct->sriov_info.rings_per_vf);
 
 	/* Setup the interrupt handler and record the INT SUM register address*/
-	if (octeon_setup_interrupt(oct))
+	if (octeon_setup_interrupt(oct, oct->sriov_info.rings_per_vf))
 		return 1;
 
+	atomic_set(&oct->status, OCT_DEV_INTR_SET_DONE);
+
+	/* ***************************************************************
+	 * The interrupts need to be enabled for the PF<-->VF handshake.
+	 * They are [re]-enabled after the PF<-->VF handshake so that the
+	 * correct OQ tick value is used (i.e. the value retrieved from
+	 * the PF as part of the handshake).
+	 */
+
+	/* Enable Octeon device interrupts */
+	oct->fn_list.enable_interrupt(oct, OCTEON_ALL_INTR);
+
 	if (cn23xx_octeon_pfvf_handshake(oct))
 		return 1;
 
+	/* Here we [re]-enable the interrupts so that the correct OQ tick value
+	 * is used (i.e. the value that was retrieved during the handshake)
+	 */
+
 	/* Enable Octeon device interrupts */
 	oct->fn_list.enable_interrupt(oct, OCTEON_ALL_INTR);
-
-	atomic_set(&oct->status, OCT_DEV_INTR_SET_DONE);
+	/* *************************************************************** */
 
 	/* Enable the input and output queues for this Octeon device */
 	if (oct->fn_list.enable_io_queues(oct)) {
diff --git a/drivers/net/ethernet/cavium/liquidio/lio_vf_rep.c b/drivers/net/ethernet/cavium/liquidio/lio_vf_rep.c
new file mode 100644
index 000000000000..989b4ddae342
--- /dev/null
+++ b/drivers/net/ethernet/cavium/liquidio/lio_vf_rep.c
@@ -0,0 +1,672 @@
+/**********************************************************************
+ * Author: Cavium, Inc.
+ *
+ * Contact: support@cavium.com
+ *          Please include "LiquidIO" in the subject.
+ *
+ * Copyright (c) 2003-2017 Cavium, Inc.
+ *
+ * This file is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License, Version 2, as
+ * published by the Free Software Foundation.
+ *
+ * This file is distributed in the hope that it will be useful, but
+ * AS-IS and WITHOUT ANY WARRANTY; without even the implied warranty
+ * of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE, TITLE, or
+ * NONINFRINGEMENT.  See the GNU General Public License for more details.
+ ***********************************************************************/
+#include <linux/pci.h>
+#include <linux/if_vlan.h>
+#include "liquidio_common.h"
+#include "octeon_droq.h"
+#include "octeon_iq.h"
+#include "response_manager.h"
+#include "octeon_device.h"
+#include "octeon_nic.h"
+#include "octeon_main.h"
+#include "octeon_network.h"
+#include "lio_vf_rep.h"
+
+static int lio_vf_rep_open(struct net_device *ndev);
+static int lio_vf_rep_stop(struct net_device *ndev);
+static netdev_tx_t lio_vf_rep_pkt_xmit(struct sk_buff *skb,
+				       struct net_device *ndev);
+static void lio_vf_rep_tx_timeout(struct net_device *netdev, unsigned int txqueue);
+static int lio_vf_rep_phys_port_name(struct net_device *dev,
+				     char *buf, size_t len);
+static void lio_vf_rep_get_stats64(struct net_device *dev,
+				   struct rtnl_link_stats64 *stats64);
+static int lio_vf_rep_change_mtu(struct net_device *ndev, int new_mtu);
+static int lio_vf_get_port_parent_id(struct net_device *dev,
+				     struct netdev_phys_item_id *ppid);
+
+static const struct net_device_ops lio_vf_rep_ndev_ops = {
+	.ndo_open = lio_vf_rep_open,
+	.ndo_stop = lio_vf_rep_stop,
+	.ndo_start_xmit = lio_vf_rep_pkt_xmit,
+	.ndo_tx_timeout = lio_vf_rep_tx_timeout,
+	.ndo_get_phys_port_name = lio_vf_rep_phys_port_name,
+	.ndo_get_stats64 = lio_vf_rep_get_stats64,
+	.ndo_change_mtu = lio_vf_rep_change_mtu,
+	.ndo_get_port_parent_id = lio_vf_get_port_parent_id,
+};
+
+static int
+lio_vf_rep_send_soft_command(struct octeon_device *oct,
+			     void *req, int req_size,
+			     void *resp, int resp_size)
+{
+	int tot_resp_size = sizeof(struct lio_vf_rep_resp) + resp_size;
+	struct octeon_soft_command *sc = NULL;
+	struct lio_vf_rep_resp *rep_resp;
+	void *sc_req;
+	int err;
+
+	sc = (struct octeon_soft_command *)
+		octeon_alloc_soft_command(oct, req_size,
+					  tot_resp_size, 0);
+	if (!sc)
+		return -ENOMEM;
+
+	init_completion(&sc->complete);
+	sc->sc_status = OCTEON_REQUEST_PENDING;
+
+	sc_req = (struct lio_vf_rep_req *)sc->virtdptr;
+	memcpy(sc_req, req, req_size);
+
+	rep_resp = (struct lio_vf_rep_resp *)sc->virtrptr;
+	memset(rep_resp, 0, tot_resp_size);
+	WRITE_ONCE(rep_resp->status, 1);
+
+	sc->iq_no = 0;
+	octeon_prepare_soft_command(oct, sc, OPCODE_NIC,
+				    OPCODE_NIC_VF_REP_CMD, 0, 0, 0);
+
+	err = octeon_send_soft_command(oct, sc);
+	if (err == IQ_SEND_FAILED)
+		goto free_buff;
+
+	err = wait_for_sc_completion_timeout(oct, sc, 0);
+	if (err)
+		return err;
+
+	err = READ_ONCE(rep_resp->status) ? -EBUSY : 0;
+	if (err)
+		dev_err(&oct->pci_dev->dev, "VF rep send config failed\n");
+	else if (resp)
+		memcpy(resp, (rep_resp + 1), resp_size);
+
+	WRITE_ONCE(sc->caller_is_done, true);
+	return err;
+
+free_buff:
+	octeon_free_soft_command(oct, sc);
+
+	return err;
+}
+
+static int
+lio_vf_rep_open(struct net_device *ndev)
+{
+	struct lio_vf_rep_desc *vf_rep = netdev_priv(ndev);
+	struct lio_vf_rep_req rep_cfg;
+	struct octeon_device *oct;
+	int ret;
+
+	oct = vf_rep->oct;
+
+	memset(&rep_cfg, 0, sizeof(rep_cfg));
+	rep_cfg.req_type = LIO_VF_REP_REQ_STATE;
+	rep_cfg.ifidx = vf_rep->ifidx;
+	rep_cfg.rep_state.state = LIO_VF_REP_STATE_UP;
+
+	ret = lio_vf_rep_send_soft_command(oct, &rep_cfg,
+					   sizeof(rep_cfg), NULL, 0);
+
+	if (ret) {
+		dev_err(&oct->pci_dev->dev,
+			"VF_REP open failed with err %d\n", ret);
+		return -EIO;
+	}
+
+	atomic_set(&vf_rep->ifstate, (atomic_read(&vf_rep->ifstate) |
+				      LIO_IFSTATE_RUNNING));
+
+	netif_carrier_on(ndev);
+	netif_start_queue(ndev);
+
+	return 0;
+}
+
+static int
+lio_vf_rep_stop(struct net_device *ndev)
+{
+	struct lio_vf_rep_desc *vf_rep = netdev_priv(ndev);
+	struct lio_vf_rep_req rep_cfg;
+	struct octeon_device *oct;
+	int ret;
+
+	oct = vf_rep->oct;
+
+	memset(&rep_cfg, 0, sizeof(rep_cfg));
+	rep_cfg.req_type = LIO_VF_REP_REQ_STATE;
+	rep_cfg.ifidx = vf_rep->ifidx;
+	rep_cfg.rep_state.state = LIO_VF_REP_STATE_DOWN;
+
+	ret = lio_vf_rep_send_soft_command(oct, &rep_cfg,
+					   sizeof(rep_cfg), NULL, 0);
+
+	if (ret) {
+		dev_err(&oct->pci_dev->dev,
+			"VF_REP dev stop failed with err %d\n", ret);
+		return -EIO;
+	}
+
+	atomic_set(&vf_rep->ifstate, (atomic_read(&vf_rep->ifstate) &
+				      ~LIO_IFSTATE_RUNNING));
+
+	netif_tx_disable(ndev);
+	netif_carrier_off(ndev);
+
+	return 0;
+}
+
+static void
+lio_vf_rep_tx_timeout(struct net_device *ndev, unsigned int txqueue)
+{
+	netif_trans_update(ndev);
+
+	netif_wake_queue(ndev);
+}
+
+static void
+lio_vf_rep_get_stats64(struct net_device *dev,
+		       struct rtnl_link_stats64 *stats64)
+{
+	struct lio_vf_rep_desc *vf_rep = netdev_priv(dev);
+
+	/* Swap tx and rx stats as VF rep is a switch port */
+	stats64->tx_packets = vf_rep->stats.rx_packets;
+	stats64->tx_bytes   = vf_rep->stats.rx_bytes;
+	stats64->tx_dropped = vf_rep->stats.rx_dropped;
+
+	stats64->rx_packets = vf_rep->stats.tx_packets;
+	stats64->rx_bytes   = vf_rep->stats.tx_bytes;
+	stats64->rx_dropped = vf_rep->stats.tx_dropped;
+}
+
+static int
+lio_vf_rep_change_mtu(struct net_device *ndev, int new_mtu)
+{
+	struct lio_vf_rep_desc *vf_rep = netdev_priv(ndev);
+	struct lio_vf_rep_req rep_cfg;
+	struct octeon_device *oct;
+	int ret;
+
+	oct = vf_rep->oct;
+
+	memset(&rep_cfg, 0, sizeof(rep_cfg));
+	rep_cfg.req_type = LIO_VF_REP_REQ_MTU;
+	rep_cfg.ifidx = vf_rep->ifidx;
+	rep_cfg.rep_mtu.mtu = cpu_to_be32(new_mtu);
+
+	ret = lio_vf_rep_send_soft_command(oct, &rep_cfg,
+					   sizeof(rep_cfg), NULL, 0);
+	if (ret) {
+		dev_err(&oct->pci_dev->dev,
+			"Change MTU failed with err %d\n", ret);
+		return -EIO;
+	}
+
+	WRITE_ONCE(ndev->mtu, new_mtu);
+
+	return 0;
+}
+
+static int
+lio_vf_rep_phys_port_name(struct net_device *dev,
+			  char *buf, size_t len)
+{
+	struct lio_vf_rep_desc *vf_rep = netdev_priv(dev);
+	struct octeon_device *oct = vf_rep->oct;
+	int ret;
+
+	ret = snprintf(buf, len, "pf%dvf%d", oct->pf_num,
+		       vf_rep->ifidx - oct->pf_num * 64 - 1);
+	if (ret >= len)
+		return -EOPNOTSUPP;
+
+	return 0;
+}
+
+static struct net_device *
+lio_vf_rep_get_ndev(struct octeon_device *oct, int ifidx)
+{
+	int vf_id, max_vfs = CN23XX_MAX_VFS_PER_PF + 1;
+	int vfid_mask = max_vfs - 1;
+
+	if (ifidx <= oct->pf_num * max_vfs ||
+	    ifidx >= oct->pf_num * max_vfs + max_vfs)
+		return NULL;
+
+	/* ifidx 1-63 for PF0 VFs
+	 * ifidx 65-127 for PF1 VFs
+	 */
+	vf_id = (ifidx & vfid_mask) - 1;
+
+	return oct->vf_rep_list.ndev[vf_id];
+}
+
+static void
+lio_vf_rep_copy_packet(struct octeon_device *oct,
+		       struct sk_buff *skb,
+		       int len)
+{
+	if (likely(len > MIN_SKB_SIZE)) {
+		struct octeon_skb_page_info *pg_info;
+		unsigned char *va;
+
+		pg_info = ((struct octeon_skb_page_info *)(skb->cb));
+		if (pg_info->page) {
+			va = page_address(pg_info->page) +
+				pg_info->page_offset;
+			memcpy(skb->data, va, MIN_SKB_SIZE);
+			skb_put(skb, MIN_SKB_SIZE);
+			skb_add_rx_frag(skb, skb_shinfo(skb)->nr_frags,
+					pg_info->page,
+					pg_info->page_offset + MIN_SKB_SIZE,
+					len - MIN_SKB_SIZE,
+					LIO_RXBUFFER_SZ);
+		}
+	} else {
+		struct octeon_skb_page_info *pg_info =
+			((struct octeon_skb_page_info *)(skb->cb));
+
+		skb_copy_to_linear_data(skb, page_address(pg_info->page) +
+					pg_info->page_offset, len);
+		skb_put(skb, len);
+		put_page(pg_info->page);
+	}
+}
+
+static int
+lio_vf_rep_pkt_recv(struct octeon_recv_info *recv_info, void *buf)
+{
+	struct octeon_recv_pkt *recv_pkt = recv_info->recv_pkt;
+	struct lio_vf_rep_desc *vf_rep;
+	struct net_device *vf_ndev;
+	struct octeon_device *oct;
+	union octeon_rh *rh;
+	struct sk_buff *skb;
+	int i, ifidx;
+
+	oct = lio_get_device(recv_pkt->octeon_id);
+	if (!oct)
+		goto free_buffers;
+
+	skb = recv_pkt->buffer_ptr[0];
+	rh = &recv_pkt->rh;
+	ifidx = rh->r.ossp;
+
+	vf_ndev = lio_vf_rep_get_ndev(oct, ifidx);
+	if (!vf_ndev)
+		goto free_buffers;
+
+	vf_rep = netdev_priv(vf_ndev);
+	if (!(atomic_read(&vf_rep->ifstate) & LIO_IFSTATE_RUNNING) ||
+	    recv_pkt->buffer_count > 1)
+		goto free_buffers;
+
+	skb->dev = vf_ndev;
+
+	/* Multiple buffers are not used for vf_rep packets.
+	 * So just buffer_size[0] is valid.
+	 */
+	lio_vf_rep_copy_packet(oct, skb, recv_pkt->buffer_size[0]);
+
+	skb_pull(skb, rh->r_dh.len * BYTES_PER_DHLEN_UNIT);
+	skb->protocol = eth_type_trans(skb, skb->dev);
+	skb->ip_summed = CHECKSUM_NONE;
+
+	netif_rx(skb);
+
+	octeon_free_recv_info(recv_info);
+
+	return 0;
+
+free_buffers:
+	for (i = 0; i < recv_pkt->buffer_count; i++)
+		recv_buffer_free(recv_pkt->buffer_ptr[i]);
+
+	octeon_free_recv_info(recv_info);
+
+	return 0;
+}
+
+static void
+lio_vf_rep_packet_sent_callback(struct octeon_device *oct,
+				u32 status, void *buf)
+{
+	struct octeon_soft_command *sc = (struct octeon_soft_command *)buf;
+	struct sk_buff *skb = sc->ctxptr;
+	struct net_device *ndev = skb->dev;
+	u32 iq_no;
+
+	dma_unmap_single(&oct->pci_dev->dev, sc->dmadptr,
+			 sc->datasize, DMA_TO_DEVICE);
+	dev_kfree_skb_any(skb);
+	iq_no = sc->iq_no;
+	octeon_free_soft_command(oct, sc);
+
+	if (octnet_iq_is_full(oct, iq_no))
+		return;
+
+	if (netif_queue_stopped(ndev))
+		netif_wake_queue(ndev);
+}
+
+static netdev_tx_t
+lio_vf_rep_pkt_xmit(struct sk_buff *skb, struct net_device *ndev)
+{
+	struct lio_vf_rep_desc *vf_rep = netdev_priv(ndev);
+	struct net_device *parent_ndev = vf_rep->parent_ndev;
+	struct octeon_device *oct = vf_rep->oct;
+	struct octeon_instr_pki_ih3 *pki_ih3;
+	struct octeon_soft_command *sc;
+	struct lio *parent_lio;
+	int status;
+
+	parent_lio = GET_LIO(parent_ndev);
+
+	if (!(atomic_read(&vf_rep->ifstate) & LIO_IFSTATE_RUNNING) ||
+	    skb->len <= 0)
+		goto xmit_failed;
+
+	if (octnet_iq_is_full(vf_rep->oct, parent_lio->txq)) {
+		dev_err(&oct->pci_dev->dev, "VF rep: Device IQ full\n");
+		netif_stop_queue(ndev);
+		return NETDEV_TX_BUSY;
+	}
+
+	sc = (struct octeon_soft_command *)
+		octeon_alloc_soft_command(oct, 0, 16, 0);
+	if (!sc) {
+		dev_err(&oct->pci_dev->dev, "VF rep: Soft command alloc failed\n");
+		goto xmit_failed;
+	}
+
+	/* Multiple buffers are not used for vf_rep packets. */
+	if (skb_shinfo(skb)->nr_frags != 0) {
+		dev_err(&oct->pci_dev->dev, "VF rep: nr_frags != 0. Dropping packet\n");
+		octeon_free_soft_command(oct, sc);
+		goto xmit_failed;
+	}
+
+	sc->dmadptr = dma_map_single(&oct->pci_dev->dev,
+				     skb->data, skb->len, DMA_TO_DEVICE);
+	if (dma_mapping_error(&oct->pci_dev->dev, sc->dmadptr)) {
+		dev_err(&oct->pci_dev->dev, "VF rep: DMA mapping failed\n");
+		octeon_free_soft_command(oct, sc);
+		goto xmit_failed;
+	}
+
+	sc->virtdptr = skb->data;
+	sc->datasize = skb->len;
+	sc->ctxptr = skb;
+	sc->iq_no = parent_lio->txq;
+
+	octeon_prepare_soft_command(oct, sc, OPCODE_NIC, OPCODE_NIC_VF_REP_PKT,
+				    vf_rep->ifidx, 0, 0);
+	pki_ih3 = (struct octeon_instr_pki_ih3 *)&sc->cmd.cmd3.pki_ih3;
+	pki_ih3->tagtype = ORDERED_TAG;
+
+	sc->callback = lio_vf_rep_packet_sent_callback;
+	sc->callback_arg = sc;
+
+	status = octeon_send_soft_command(oct, sc);
+	if (status == IQ_SEND_FAILED) {
+		dma_unmap_single(&oct->pci_dev->dev, sc->dmadptr,
+				 sc->datasize, DMA_TO_DEVICE);
+		octeon_free_soft_command(oct, sc);
+		goto xmit_failed;
+	}
+
+	if (status == IQ_SEND_STOP)
+		netif_stop_queue(ndev);
+
+	netif_trans_update(ndev);
+
+	return NETDEV_TX_OK;
+
+xmit_failed:
+	dev_kfree_skb_any(skb);
+
+	return NETDEV_TX_OK;
+}
+
+static int lio_vf_get_port_parent_id(struct net_device *dev,
+				     struct netdev_phys_item_id *ppid)
+{
+	struct lio_vf_rep_desc *vf_rep = netdev_priv(dev);
+	struct net_device *parent_ndev = vf_rep->parent_ndev;
+	struct lio *lio = GET_LIO(parent_ndev);
+
+	ppid->id_len = ETH_ALEN;
+	ether_addr_copy(ppid->id, (void *)&lio->linfo.hw_addr + 2);
+
+	return 0;
+}
+
+static void
+lio_vf_rep_fetch_stats(struct work_struct *work)
+{
+	struct cavium_wk *wk = (struct cavium_wk *)work;
+	struct lio_vf_rep_desc *vf_rep = wk->ctxptr;
+	struct lio_vf_rep_stats stats;
+	struct lio_vf_rep_req rep_cfg;
+	struct octeon_device *oct;
+	int ret;
+
+	oct = vf_rep->oct;
+
+	memset(&rep_cfg, 0, sizeof(rep_cfg));
+	rep_cfg.req_type = LIO_VF_REP_REQ_STATS;
+	rep_cfg.ifidx = vf_rep->ifidx;
+
+	ret = lio_vf_rep_send_soft_command(oct, &rep_cfg, sizeof(rep_cfg),
+					   &stats, sizeof(stats));
+
+	if (!ret) {
+		octeon_swap_8B_data((u64 *)&stats, (sizeof(stats) >> 3));
+		memcpy(&vf_rep->stats, &stats, sizeof(stats));
+	}
+
+	schedule_delayed_work(&vf_rep->stats_wk.work,
+			      msecs_to_jiffies(LIO_VF_REP_STATS_POLL_TIME_MS));
+}
+
+int
+lio_vf_rep_create(struct octeon_device *oct)
+{
+	struct lio_vf_rep_desc *vf_rep;
+	struct net_device *ndev;
+	int i, num_vfs;
+
+	if (oct->eswitch_mode != DEVLINK_ESWITCH_MODE_SWITCHDEV)
+		return 0;
+
+	if (!oct->sriov_info.sriov_enabled)
+		return 0;
+
+	num_vfs = oct->sriov_info.num_vfs_alloced;
+
+	oct->vf_rep_list.num_vfs = 0;
+	for (i = 0; i < num_vfs; i++) {
+		ndev = alloc_etherdev(sizeof(struct lio_vf_rep_desc));
+
+		if (!ndev) {
+			dev_err(&oct->pci_dev->dev,
+				"VF rep device %d creation failed\n", i);
+			goto cleanup;
+		}
+
+		ndev->min_mtu = LIO_MIN_MTU_SIZE;
+		ndev->max_mtu = LIO_MAX_MTU_SIZE;
+		ndev->netdev_ops = &lio_vf_rep_ndev_ops;
+
+		vf_rep = netdev_priv(ndev);
+		memset(vf_rep, 0, sizeof(*vf_rep));
+
+		vf_rep->ndev = ndev;
+		vf_rep->oct = oct;
+		vf_rep->parent_ndev = oct->props[0].netdev;
+		vf_rep->ifidx = (oct->pf_num * 64) + i + 1;
+
+		eth_hw_addr_random(ndev);
+
+		if (register_netdev(ndev)) {
+			dev_err(&oct->pci_dev->dev, "VF rep nerdev registration failed\n");
+
+			free_netdev(ndev);
+			goto cleanup;
+		}
+
+		netif_carrier_off(ndev);
+
+		INIT_DELAYED_WORK(&vf_rep->stats_wk.work,
+				  lio_vf_rep_fetch_stats);
+		vf_rep->stats_wk.ctxptr = (void *)vf_rep;
+		schedule_delayed_work(&vf_rep->stats_wk.work,
+				      msecs_to_jiffies
+				      (LIO_VF_REP_STATS_POLL_TIME_MS));
+		oct->vf_rep_list.num_vfs++;
+		oct->vf_rep_list.ndev[i] = ndev;
+	}
+
+	if (octeon_register_dispatch_fn(oct, OPCODE_NIC,
+					OPCODE_NIC_VF_REP_PKT,
+					lio_vf_rep_pkt_recv, oct)) {
+		dev_err(&oct->pci_dev->dev, "VF rep Dispatch func registration failed\n");
+
+		goto cleanup;
+	}
+
+	return 0;
+
+cleanup:
+	for (i = 0; i < oct->vf_rep_list.num_vfs; i++) {
+		ndev = oct->vf_rep_list.ndev[i];
+		oct->vf_rep_list.ndev[i] = NULL;
+		if (ndev) {
+			vf_rep = netdev_priv(ndev);
+			cancel_delayed_work_sync
+				(&vf_rep->stats_wk.work);
+			unregister_netdev(ndev);
+			free_netdev(ndev);
+		}
+	}
+
+	oct->vf_rep_list.num_vfs = 0;
+
+	return -1;
+}
+
+void
+lio_vf_rep_destroy(struct octeon_device *oct)
+{
+	struct lio_vf_rep_desc *vf_rep;
+	struct net_device *ndev;
+	int i;
+
+	if (oct->eswitch_mode != DEVLINK_ESWITCH_MODE_SWITCHDEV)
+		return;
+
+	if (!oct->sriov_info.sriov_enabled)
+		return;
+
+	for (i = 0; i < oct->vf_rep_list.num_vfs; i++) {
+		ndev = oct->vf_rep_list.ndev[i];
+		oct->vf_rep_list.ndev[i] = NULL;
+		if (ndev) {
+			vf_rep = netdev_priv(ndev);
+			cancel_delayed_work_sync
+				(&vf_rep->stats_wk.work);
+			netif_tx_disable(ndev);
+			netif_carrier_off(ndev);
+
+			unregister_netdev(ndev);
+			free_netdev(ndev);
+		}
+	}
+
+	oct->vf_rep_list.num_vfs = 0;
+}
+
+static int
+lio_vf_rep_netdev_event(struct notifier_block *nb,
+			unsigned long event, void *ptr)
+{
+	struct net_device *ndev = netdev_notifier_info_to_dev(ptr);
+	struct lio_vf_rep_desc *vf_rep;
+	struct lio_vf_rep_req rep_cfg;
+	struct octeon_device *oct;
+	int ret;
+
+	switch (event) {
+	case NETDEV_REGISTER:
+	case NETDEV_CHANGENAME:
+		break;
+
+	default:
+		return NOTIFY_DONE;
+	}
+
+	if (ndev->netdev_ops != &lio_vf_rep_ndev_ops)
+		return NOTIFY_DONE;
+
+	vf_rep = netdev_priv(ndev);
+	oct = vf_rep->oct;
+
+	if (strlen(ndev->name) > LIO_IF_NAME_SIZE) {
+		dev_err(&oct->pci_dev->dev,
+			"Device name change sync failed as the size is > %d\n",
+			LIO_IF_NAME_SIZE);
+		return NOTIFY_DONE;
+	}
+
+	memset(&rep_cfg, 0, sizeof(rep_cfg));
+	rep_cfg.req_type = LIO_VF_REP_REQ_DEVNAME;
+	rep_cfg.ifidx = vf_rep->ifidx;
+	strscpy(rep_cfg.rep_name.name, ndev->name,
+		sizeof(rep_cfg.rep_name.name));
+
+	ret = lio_vf_rep_send_soft_command(oct, &rep_cfg,
+					   sizeof(rep_cfg), NULL, 0);
+	if (ret)
+		dev_err(&oct->pci_dev->dev,
+			"vf_rep netdev name change failed with err %d\n", ret);
+
+	return NOTIFY_DONE;
+}
+
+static struct notifier_block lio_vf_rep_netdev_notifier = {
+	.notifier_call = lio_vf_rep_netdev_event,
+};
+
+int
+lio_vf_rep_modinit(void)
+{
+	if (register_netdevice_notifier(&lio_vf_rep_netdev_notifier)) {
+		pr_err("netdev notifier registration failed\n");
+		return -EFAULT;
+	}
+
+	return 0;
+}
+
+void
+lio_vf_rep_modexit(void)
+{
+	if (unregister_netdevice_notifier(&lio_vf_rep_netdev_notifier))
+		pr_err("netdev notifier unregister failed\n");
+}
diff --git a/drivers/net/ethernet/cavium/liquidio/lio_vf_rep.h b/drivers/net/ethernet/cavium/liquidio/lio_vf_rep.h
new file mode 100644
index 000000000000..bb3cedc63c63
--- /dev/null
+++ b/drivers/net/ethernet/cavium/liquidio/lio_vf_rep.h
@@ -0,0 +1,49 @@
+/**********************************************************************
+ * Author: Cavium, Inc.
+ *
+ * Contact: support@cavium.com
+ *          Please include "LiquidIO" in the subject.
+ *
+ * Copyright (c) 2003-2017 Cavium, Inc.
+ *
+ * This file is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License, Version 2, as
+ * published by the Free Software Foundation.
+ *
+ * This file is distributed in the hope that it will be useful, but
+ * AS-IS and WITHOUT ANY WARRANTY; without even the implied warranty
+ * of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE, TITLE, or
+ * NONINFRINGEMENT.  See the GNU General Public License for more
+ * details.
+ *
+ * This file may also be available under a different license from Cavium.
+ * Contact Cavium, Inc. for more information
+ **********************************************************************/
+
+/*! \file octeon_vf_main.h
+ *  \brief Host Driver: This file defines vf_rep related macros and structures
+ */
+#ifndef __LIO_VF_REP_H__
+#define __LIO_VF_REP_H__
+#define LIO_VF_REP_REQ_TMO_MS 5000
+#define LIO_VF_REP_STATS_POLL_TIME_MS 200
+
+struct lio_vf_rep_desc {
+	struct net_device *parent_ndev;
+	struct net_device *ndev;
+	struct octeon_device *oct;
+	struct lio_vf_rep_stats stats;
+	struct cavium_wk stats_wk;
+	atomic_t ifstate;
+	int ifidx;
+};
+
+struct lio_vf_rep_sc_ctx {
+	struct completion complete;
+};
+
+int lio_vf_rep_create(struct octeon_device *oct);
+void lio_vf_rep_destroy(struct octeon_device *oct);
+int lio_vf_rep_modinit(void);
+void lio_vf_rep_modexit(void);
+#endif
diff --git a/drivers/net/ethernet/cavium/liquidio/liquidio_common.h b/drivers/net/ethernet/cavium/liquidio/liquidio_common.h
index 8ea2323d8d67..4da90757cd3f 100644
--- a/drivers/net/ethernet/cavium/liquidio/liquidio_common.h
+++ b/drivers/net/ethernet/cavium/liquidio/liquidio_common.h
@@ -25,17 +25,11 @@
 
 #include "octeon_config.h"
 
-#define LIQUIDIO_PACKAGE ""
 #define LIQUIDIO_BASE_MAJOR_VERSION 1
-#define LIQUIDIO_BASE_MINOR_VERSION 5
-#define LIQUIDIO_BASE_MICRO_VERSION 1
+#define LIQUIDIO_BASE_MINOR_VERSION 7
+#define LIQUIDIO_BASE_MICRO_VERSION 2
 #define LIQUIDIO_BASE_VERSION   __stringify(LIQUIDIO_BASE_MAJOR_VERSION) "." \
 				__stringify(LIQUIDIO_BASE_MINOR_VERSION)
-#define LIQUIDIO_MICRO_VERSION  "." __stringify(LIQUIDIO_BASE_MICRO_VERSION)
-#define LIQUIDIO_VERSION        LIQUIDIO_PACKAGE \
-				__stringify(LIQUIDIO_BASE_MAJOR_VERSION) "." \
-				__stringify(LIQUIDIO_BASE_MINOR_VERSION) \
-				"." __stringify(LIQUIDIO_BASE_MICRO_VERSION)
 
 struct lio_version {
 	u16  major;
@@ -84,10 +78,17 @@ enum octeon_tag_type {
 #define OPCODE_NIC_IF_CFG              0x09
 #define OPCODE_NIC_VF_DRV_NOTICE       0x0A
 #define OPCODE_NIC_INTRMOD_PARAMS      0x0B
+#define OPCODE_NIC_QCOUNT_UPDATE       0x12
+#define OPCODE_NIC_SET_TRUSTED_VF	0x13
+#define OPCODE_NIC_SYNC_OCTEON_TIME	0x14
 #define VF_DRV_LOADED                  1
 #define VF_DRV_REMOVED                -1
 #define VF_DRV_MACADDR_CHANGED         2
 
+#define OPCODE_NIC_VF_REP_PKT          0x15
+#define OPCODE_NIC_VF_REP_CMD          0x16
+#define OPCODE_NIC_UBOOT_CTL           0x17
+
 #define CORE_DRV_TEST_SCATTER_OP    0xFFF5
 
 /* Application codes advertised by the core driver initialization packet. */
@@ -106,6 +107,15 @@ enum octeon_tag_type {
 #define MAX_IOQ_INTERRUPTS_PER_PF   (64 * 2)
 #define MAX_IOQ_INTERRUPTS_PER_VF   (8 * 2)
 
+#define SCR2_BIT_FW_LOADED	    63
+
+/* App specific capabilities from firmware to pf driver */
+#define LIQUIDIO_TIME_SYNC_CAP 0x1
+#define LIQUIDIO_SWITCHDEV_CAP 0x2
+#define LIQUIDIO_SPOOFCHK_CAP  0x4
+
+/* error status return from firmware */
+#define OCTEON_REQUEST_NO_PERMISSION 0xc
 
 static inline u32 incr_index(u32 index, u32 count, u32 max)
 {
@@ -173,6 +183,8 @@ static inline void add_sg_size(struct octeon_sg_entry *sg_entry,
 
 /*------------------------- End Scatter/Gather ---------------------------*/
 
+#define   OCTNET_FRM_LENGTH_SIZE      8
+
 #define   OCTNET_FRM_PTP_HEADER_SIZE  8
 
 #define   OCTNET_FRM_HEADER_SIZE     22 /* VLAN + Ethernet */
@@ -181,13 +193,13 @@ static inline void add_sg_size(struct octeon_sg_entry *sg_entry,
 
 #define   OCTNET_MAX_FRM_SIZE        (16000 + OCTNET_FRM_HEADER_SIZE)
 
-#define   OCTNET_DEFAULT_FRM_SIZE    (1500 + OCTNET_FRM_HEADER_SIZE)
+#define   OCTNET_DEFAULT_MTU         (1500)
+#define   OCTNET_DEFAULT_FRM_SIZE  (OCTNET_DEFAULT_MTU + OCTNET_FRM_HEADER_SIZE)
 
 /** NIC Commands are sent using this Octeon Input Queue */
 #define   OCTNET_CMD_Q                0
 
 /* NIC Command types */
-#define   OCTNET_CMD_RESET_PF         0x0
 #define   OCTNET_CMD_CHANGE_MTU       0x1
 #define   OCTNET_CMD_CHANGE_MACADDR   0x2
 #define   OCTNET_CMD_CHANGE_DEVFLAGS  0x3
@@ -215,7 +227,7 @@ static inline void add_sg_size(struct octeon_sg_entry *sg_entry,
 #define   OCTNET_CMD_VERBOSE_ENABLE   0x14
 #define   OCTNET_CMD_VERBOSE_DISABLE  0x15
 
-#define   OCTNET_CMD_ENABLE_VLAN_FILTER 0x16
+#define   OCTNET_CMD_VLAN_FILTER_CTL 0x16
 #define   OCTNET_CMD_ADD_VLAN_FILTER  0x17
 #define   OCTNET_CMD_DEL_VLAN_FILTER  0x18
 #define   OCTNET_CMD_VXLAN_PORT_CONFIG 0x19
@@ -224,12 +236,35 @@ static inline void add_sg_size(struct octeon_sg_entry *sg_entry,
 
 #define   OCTNET_CMD_SET_UC_LIST       0x1b
 #define   OCTNET_CMD_SET_VF_LINKSTATE  0x1c
+
+#define   OCTNET_CMD_QUEUE_COUNT_CTL	0x1f
+
+#define   OCTNET_CMD_GROUP1             1
+#define   OCTNET_CMD_SET_VF_SPOOFCHK    0x1
+#define   OCTNET_GROUP1_LAST_CMD        OCTNET_CMD_SET_VF_SPOOFCHK
+
 #define   OCTNET_CMD_VXLAN_PORT_ADD    0x0
 #define   OCTNET_CMD_VXLAN_PORT_DEL    0x1
 #define   OCTNET_CMD_RXCSUM_ENABLE     0x0
 #define   OCTNET_CMD_RXCSUM_DISABLE    0x1
 #define   OCTNET_CMD_TXCSUM_ENABLE     0x0
 #define   OCTNET_CMD_TXCSUM_DISABLE    0x1
+#define   OCTNET_CMD_VLAN_FILTER_ENABLE 0x1
+#define   OCTNET_CMD_VLAN_FILTER_DISABLE 0x0
+
+#define   OCTNET_CMD_FAIL 0x1
+
+#define   SEAPI_CMD_FEC_SET             0x0
+#define   SEAPI_CMD_FEC_SET_DISABLE       0x0
+#define   SEAPI_CMD_FEC_SET_RS            0x1
+#define   SEAPI_CMD_FEC_GET             0x1
+
+#define   SEAPI_CMD_SPEED_SET           0x2
+#define   SEAPI_CMD_SPEED_GET           0x3
+
+#define OPCODE_NIC_VF_PORT_STATS        0x22
+
+#define   LIO_CMD_WAIT_TM 100
 
 /* RX(packets coming from wire) Checksum verification flags */
 /* TCP/UDP csum */
@@ -277,7 +312,8 @@ union octnet_cmd {
 
 		u64 more:6; /* How many udd words follow the command */
 
-		u64 reserved:29;
+		u64 cmdgroup:8;
+		u64 reserved:21;
 
 		u64 param1:16;
 
@@ -289,7 +325,8 @@ union octnet_cmd {
 
 		u64 param1:16;
 
-		u64 reserved:29;
+		u64 reserved:21;
+		u64 cmdgroup:8;
 
 		u64 more:6;
 
@@ -658,9 +695,11 @@ union oct_link_status {
 		u64 if_mode:5;
 		u64 pause:1;
 		u64 flashing:1;
-		u64 reserved:15;
+		u64 phy_type:5;
+		u64 reserved:10;
 #else
-		u64 reserved:15;
+		u64 reserved:10;
+		u64 phy_type:5;
 		u64 flashing:1;
 		u64 pause:1;
 		u64 if_mode:5;
@@ -673,6 +712,12 @@ union oct_link_status {
 	} s;
 };
 
+enum lio_phy_type {
+	LIO_PHY_PORT_TP = 0x0,
+	LIO_PHY_PORT_FIBRE = 0x1,
+	LIO_PHY_PORT_UNKNOWN,
+};
+
 /** The txpciq info passed to host from the firmware */
 
 union oct_txpciq {
@@ -685,9 +730,13 @@ union oct_txpciq {
 		u64 pkind:6;
 		u64 use_qpg:1;
 		u64 qpg:11;
-		u64 reserved:30;
+		u64 reserved0:10;
+		u64 ctrl_qpg:11;
+		u64 reserved:9;
 #else
-		u64 reserved:30;
+		u64 reserved:9;
+		u64 ctrl_qpg:11;
+		u64 reserved0:10;
 		u64 qpg:11;
 		u64 use_qpg:1;
 		u64 pkind:6;
@@ -721,13 +770,17 @@ struct oct_link_info {
 #ifdef __BIG_ENDIAN_BITFIELD
 	u64 gmxport:16;
 	u64 macaddr_is_admin_asgnd:1;
-	u64 rsvd:31;
+	u64 rsvd:13;
+	u64 macaddr_spoofchk:1;
+	u64 rsvd1:17;
 	u64 num_txpciq:8;
 	u64 num_rxpciq:8;
 #else
 	u64 num_rxpciq:8;
 	u64 num_txpciq:8;
-	u64 rsvd:31;
+	u64 rsvd1:17;
+	u64 macaddr_spoofchk:1;
+	u64 rsvd:13;
 	u64 macaddr_is_admin_asgnd:1;
 	u64 gmxport:16;
 #endif
@@ -748,22 +801,32 @@ struct liquidio_if_cfg_info {
 /** Stats for each NIC port in RX direction. */
 struct nic_rx_stats {
 	/* link-level stats */
-	u64 total_rcvd;
-	u64 bytes_rcvd;
-	u64 total_bcst;
-	u64 total_mcst;
-	u64 runts;
-	u64 ctl_rcvd;
-	u64 fifo_err;      /* Accounts for over/under-run of buffers */
-	u64 dmac_drop;
-	u64 fcs_err;
-	u64 jabber_err;
-	u64 l2_err;
-	u64 frame_err;
+	u64 total_rcvd;		/* Received packets */
+	u64 bytes_rcvd;		/* Octets of received packets */
+	u64 total_bcst;		/* Number of non-dropped L2 broadcast packets */
+	u64 total_mcst;		/* Number of non-dropped L2 multicast packets */
+	u64 runts;		/* Packets shorter than allowed */
+	u64 ctl_rcvd;		/* Received PAUSE packets */
+	u64 fifo_err;		/* Packets dropped due to RX FIFO full */
+	u64 dmac_drop;		/* Packets dropped by the DMAC filter */
+	u64 fcs_err;		/* Sum of fragment, overrun, and FCS errors */
+	u64 jabber_err;		/* Packets larger than allowed */
+	u64 l2_err;		/* Sum of DMA, parity, PCAM access, no memory,
+				 * buffer overflow, malformed L2 header or
+				 * length, oversize errors
+				 **/
+	u64 frame_err;		/* Sum of IPv4 and L4 checksum errors */
+	u64 red_drops;		/* Packets dropped by RED due to buffer
+				 * exhaustion
+				 **/
 
 	/* firmware stats */
 	u64 fw_total_rcvd;
 	u64 fw_total_fwd;
+	u64 fw_total_fwd_bytes;
+	u64 fw_total_mcast;
+	u64 fw_total_bcast;
+
 	u64 fw_err_pko;
 	u64 fw_err_link;
 	u64 fw_err_drop;
@@ -774,11 +837,11 @@ struct nic_rx_stats {
 	u64 fw_lro_pkts;   /* Number of packets that are LROed      */
 	u64 fw_lro_octs;   /* Number of octets that are LROed       */
 	u64 fw_total_lro;  /* Number of LRO packets formed          */
-	u64 fw_lro_aborts; /* Number of times lRO of packet aborted */
+	u64 fw_lro_aborts; /* Number of times LRO of packet aborted */
 	u64 fw_lro_aborts_port;
 	u64 fw_lro_aborts_seq;
 	u64 fw_lro_aborts_tsval;
-	u64 fw_lro_aborts_timer;
+	u64 fw_lro_aborts_timer;	/* Timer setting error */
 	/* intrmod: packet forward rate */
 	u64 fwd_rate;
 };
@@ -786,23 +849,42 @@ struct nic_rx_stats {
 /** Stats for each NIC port in RX direction. */
 struct nic_tx_stats {
 	/* link-level stats */
-	u64 total_pkts_sent;
-	u64 total_bytes_sent;
-	u64 mcast_pkts_sent;
-	u64 bcast_pkts_sent;
-	u64 ctl_sent;
-	u64 one_collision_sent;   /* Packets sent after one collision*/
-	u64 multi_collision_sent; /* Packets sent after multiple collision*/
-	u64 max_collision_fail;   /* Packets not sent due to max collisions */
-	u64 max_deferral_fail;   /* Packets not sent due to max deferrals */
-	u64 fifo_err;       /* Accounts for over/under-run of buffers */
-	u64 runts;
-	u64 total_collisions; /* Total number of collisions detected */
+	u64 total_pkts_sent;		/* Total frames sent on the interface */
+	u64 total_bytes_sent;		/* Total octets sent on the interface */
+	u64 mcast_pkts_sent;		/* Packets sent to the multicast DMAC */
+	u64 bcast_pkts_sent;		/* Packets sent to a broadcast DMAC */
+	u64 ctl_sent;			/* Control/PAUSE packets sent */
+	u64 one_collision_sent;		/* Packets sent that experienced a
+					 * single collision before successful
+					 * transmission
+					 **/
+	u64 multi_collision_sent;	/* Packets sent that experienced
+					 * multiple collisions before successful
+					 * transmission
+					 **/
+	u64 max_collision_fail;		/* Packets dropped due to excessive
+					 * collisions
+					 **/
+	u64 max_deferral_fail;		/* Packets not sent due to max
+					 * deferrals
+					 **/
+	u64 fifo_err;			/* Packets sent that experienced a
+					 * transmit underflow and were
+					 * truncated
+					 **/
+	u64 runts;			/* Packets sent with an octet count
+					 * lessthan 64
+					 **/
+	u64 total_collisions;		/* Packets dropped due to excessive
+					 * collisions
+					 **/
 
 	/* firmware stats */
 	u64 fw_total_sent;
 	u64 fw_total_fwd;
 	u64 fw_total_fwd_bytes;
+	u64 fw_total_mcast_sent;
+	u64 fw_total_bcast_sent;
 	u64 fw_err_pko;
 	u64 fw_err_link;
 	u64 fw_err_drop;
@@ -810,6 +892,7 @@ struct nic_tx_stats {
 	u64 fw_tso;		/* number of tso requests */
 	u64 fw_tso_fwd;		/* number of packets segmented in tso */
 	u64 fw_tx_vxlan;
+	u64 fw_err_pki;
 };
 
 struct oct_link_stats {
@@ -841,6 +924,7 @@ static inline int opcode_slow_path(union octeon_rh *rh)
 #define VITESSE_PHY_GPIO_LOW      0x3
 #define LED_IDENTIFICATION_ON     0x1
 #define LED_IDENTIFICATION_OFF    0x0
+#define LIO23XX_COPPERHEAD_LED_GPIO 0x2
 
 struct oct_mdio_cmd {
 	u64 op;
@@ -890,4 +974,66 @@ union oct_nic_if_cfg {
 	} s;
 };
 
+struct lio_trusted_vf {
+	uint64_t active: 1;
+	uint64_t id : 8;
+	uint64_t reserved: 55;
+};
+
+struct lio_time {
+	s64 sec;   /* seconds */
+	s64 nsec;  /* nanoseconds */
+};
+
+struct lio_vf_rep_stats {
+	u64 tx_packets;
+	u64 tx_bytes;
+	u64 tx_dropped;
+
+	u64 rx_packets;
+	u64 rx_bytes;
+	u64 rx_dropped;
+};
+
+enum lio_vf_rep_req_type {
+	LIO_VF_REP_REQ_NONE,
+	LIO_VF_REP_REQ_STATE,
+	LIO_VF_REP_REQ_MTU,
+	LIO_VF_REP_REQ_STATS,
+	LIO_VF_REP_REQ_DEVNAME
+};
+
+enum {
+	LIO_VF_REP_STATE_DOWN,
+	LIO_VF_REP_STATE_UP
+};
+
+#define LIO_IF_NAME_SIZE 16
+struct lio_vf_rep_req {
+	u8 req_type;
+	u8 ifidx;
+	u8 rsvd[6];
+
+	union {
+		struct lio_vf_rep_name {
+			char name[LIO_IF_NAME_SIZE];
+		} rep_name;
+
+		struct lio_vf_rep_mtu {
+			u32 mtu;
+			u32 rsvd;
+		} rep_mtu;
+
+		struct lio_vf_rep_state {
+			u8 state;
+			u8 rsvd[7];
+		} rep_state;
+	};
+};
+
+struct lio_vf_rep_resp {
+	u64 rh;
+	u8  status;
+	u8  rsvd[7];
+};
 #endif
diff --git a/drivers/net/ethernet/cavium/liquidio/liquidio_image.h b/drivers/net/ethernet/cavium/liquidio/liquidio_image.h
index 78a3685f6fe0..5bf5e8791dfb 100644
--- a/drivers/net/ethernet/cavium/liquidio/liquidio_image.h
+++ b/drivers/net/ethernet/cavium/liquidio/liquidio_image.h
@@ -24,6 +24,7 @@
 #define LIO_FW_BASE_NAME        "lio_"
 #define LIO_FW_NAME_SUFFIX      ".bin"
 #define LIO_FW_NAME_TYPE_NIC    "nic"
+#define LIO_FW_NAME_TYPE_AUTO   "auto"
 #define LIO_FW_NAME_TYPE_NONE   "none"
 #define LIO_MAX_FIRMWARE_VERSION_LEN 16
 
diff --git a/drivers/net/ethernet/cavium/liquidio/octeon_config.h b/drivers/net/ethernet/cavium/liquidio/octeon_config.h
index d29ebc531151..24c212001212 100644
--- a/drivers/net/ethernet/cavium/liquidio/octeon_config.h
+++ b/drivers/net/ethernet/cavium/liquidio/octeon_config.h
@@ -37,6 +37,8 @@
 #define   MAX_OCTEON_LINKS	       MAX_OCTEON_NICIF
 #define   MAX_OCTEON_MULTICAST_ADDR    32
 
+#define   MAX_OCTEON_FILL_COUNT        8
+
 /* CN6xxx IQ configuration macros */
 #define   CN6XXX_MAX_INPUT_QUEUES      32
 #define   CN6XXX_MAX_IQ_DESCRIPTORS    2048
@@ -47,7 +49,7 @@
 /* CN6xxx OQ configuration macros */
 #define   CN6XXX_MAX_OUTPUT_QUEUES     32
 #define   CN6XXX_MAX_OQ_DESCRIPTORS    2048
-#define   CN6XXX_OQ_BUF_SIZE           1536
+#define   CN6XXX_OQ_BUF_SIZE           1664
 #define   CN6XXX_OQ_PKTSPER_INTR       ((CN6XXX_MAX_OQ_DESCRIPTORS < 512) ? \
 					(CN6XXX_MAX_OQ_DESCRIPTORS / 4) : 128)
 #define   CN6XXX_OQ_REFIL_THRESHOLD    ((CN6XXX_MAX_OQ_DESCRIPTORS < 512) ? \
@@ -71,14 +73,18 @@
 #define   CN23XX_MAX_RINGS_PER_VF          8
 
 #define   CN23XX_MAX_INPUT_QUEUES	CN23XX_MAX_RINGS_PER_PF
-#define   CN23XX_MAX_IQ_DESCRIPTORS	512
+#define   CN23XX_MAX_IQ_DESCRIPTORS	2048
+#define   CN23XX_DEFAULT_IQ_DESCRIPTORS	512
+#define   CN23XX_MIN_IQ_DESCRIPTORS	128
 #define   CN23XX_DB_MIN                 1
 #define   CN23XX_DB_MAX                 8
 #define   CN23XX_DB_TIMEOUT             1
 
 #define   CN23XX_MAX_OUTPUT_QUEUES	CN23XX_MAX_RINGS_PER_PF
-#define   CN23XX_MAX_OQ_DESCRIPTORS	512
-#define   CN23XX_OQ_BUF_SIZE		1536
+#define   CN23XX_MAX_OQ_DESCRIPTORS	2048
+#define   CN23XX_DEFAULT_OQ_DESCRIPTORS	512
+#define   CN23XX_MIN_OQ_DESCRIPTORS	128
+#define   CN23XX_OQ_BUF_SIZE		1664
 #define   CN23XX_OQ_PKTSPER_INTR	128
 /*#define CAVIUM_ONLY_CN23XX_RX_PERF*/
 #define   CN23XX_OQ_REFIL_THRESHOLD	16
@@ -98,8 +104,6 @@
 #define   OCTEON_32BYTE_INSTR          32
 #define   OCTEON_64BYTE_INSTR          64
 #define   OCTEON_MAX_BASE_IOQ          4
-#define   OCTEON_OQ_BUFPTR_MODE        0
-#define   OCTEON_OQ_INFOPTR_MODE       1
 
 #define   OCTEON_DMA_INTR_PKT          64
 #define   OCTEON_DMA_INTR_TIME         1000
@@ -125,7 +129,6 @@
 #define CFG_SET_IQ_INTR_PKT(cfg, val)            (cfg)->iq.iq_intr_pkt = val
 
 #define CFG_GET_OQ_MAX_Q(cfg)                    ((cfg)->oq.max_oqs)
-#define CFG_GET_OQ_INFO_PTR(cfg)                 ((cfg)->oq.info_ptr)
 #define CFG_GET_OQ_PKTS_PER_INTR(cfg)            ((cfg)->oq.pkts_per_intr)
 #define CFG_GET_OQ_REFILL_THRESHOLD(cfg)         ((cfg)->oq.refill_threshold)
 #define CFG_GET_OQ_INTR_PKT(cfg)                 ((cfg)->oq.oq_intr_pkt)
@@ -166,6 +169,11 @@
 				((cfg)->misc.oct_link_query_interval)
 #define CFG_GET_IS_SLI_BP_ON(cfg)                ((cfg)->misc.enable_sli_oq_bp)
 
+#define CFG_SET_NUM_RX_DESCS_NIC_IF(cfg, idx, value) \
+				((cfg)->nic_if_cfg[idx].num_rx_descs = value)
+#define CFG_SET_NUM_TX_DESCS_NIC_IF(cfg, idx, value) \
+				((cfg)->nic_if_cfg[idx].num_tx_descs = value)
+
 /* Max IOQs per OCTEON Link */
 #define MAX_IOQS_PER_NICIF              64
 
@@ -266,9 +274,6 @@ struct octeon_oq_config {
 	 */
 	u64 refill_threshold:16;
 
-	/** If set, the Output queue uses info-pointer mode. (Default: 1) */
-	u64 info_ptr:32;
-
 	/* Max number of OQs available */
 	u64 max_oqs:8;
 
@@ -276,9 +281,6 @@ struct octeon_oq_config {
 	/* Max number of OQs available */
 	u64 max_oqs:8;
 
-	/** If set, the Output queue uses info-pointer mode. (Default: 1) */
-	u64 info_ptr:32;
-
 	/** The number of buffers that were consumed during packet processing by
 	 *   the driver on this Output queue before the driver attempts to
 	 *   replenish
@@ -436,9 +438,10 @@ struct octeon_config {
 #define  MAX_BAR1_IOREMAP_SIZE  (16 * OCTEON_BAR1_ENTRY_SIZE)
 
 /* Response lists - 1 ordered, 1 unordered-blocking, 1 unordered-nonblocking
+ *                  1 process done list, 1 zombie lists(timeouted sc list)
  * NoResponse Lists are now maintained with each IQ. (Dec' 2007).
  */
-#define MAX_RESPONSE_LISTS           4
+#define MAX_RESPONSE_LISTS           6
 
 /* Opcode hash bits. The opcode is hashed on the lower 6-bits to lookup the
  * dispatch table.
diff --git a/drivers/net/ethernet/cavium/liquidio/octeon_console.c b/drivers/net/ethernet/cavium/liquidio/octeon_console.c
index 53f38d05f7c2..67c3570f875f 100644
--- a/drivers/net/ethernet/cavium/liquidio/octeon_console.c
+++ b/drivers/net/ethernet/cavium/liquidio/octeon_console.c
@@ -15,7 +15,7 @@
  * of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE, TITLE, or
  * NONINFRINGEMENT.  See the GNU General Public License for more details.
  ***********************************************************************/
-/**
+/*
  * @file octeon_console.c
  */
 #include <linux/moduleparam.h>
@@ -37,13 +37,6 @@ static u64 cvmx_bootmem_phy_named_block_find(struct octeon_device *oct,
 					     u32 flags);
 static int octeon_console_read(struct octeon_device *oct, u32 console_num,
 			       char *buffer, u32 buf_size);
-static u32 console_bitmask;
-module_param(console_bitmask, int, 0644);
-MODULE_PARM_DESC(console_bitmask,
-		 "Bitmask indicating which consoles have debug output redirected to syslog.");
-
-#define MIN(a, b) min((a), (b))
-#define CAST_ULL(v) ((u64)(v))
 
 #define BOOTLOADER_PCI_READ_BUFFER_DATA_ADDR    0x0006c008
 #define BOOTLOADER_PCI_READ_BUFFER_LEN_ADDR     0x0006c004
@@ -134,21 +127,11 @@ struct octeon_pci_console_desc {
 	u32 pad;
 	/* must be 64 bit aligned here... */
 	/* Array of addresses of octeon_pci_console structures */
-	u64 console_addr_array[0];
+	u64 console_addr_array[];
 	/* Implicit storage for console_addr_array */
 };
 
-/**
- * \brief determines if a given console has debug enabled.
- * @param console console to check
- * @returns  1 = enabled. 0 otherwise
- */
-static int octeon_console_debug_enabled(u32 console)
-{
-	return (console_bitmask >> (console)) & 0x1;
-}
-
-/**
+/*
  * This function is the implementation of the get macros defined
  * for individual structure members. The argument are generated
  * by the macros inorder to read only the needed memory.
@@ -177,7 +160,7 @@ static inline u64 __cvmx_bootmem_desc_get(struct octeon_device *oct,
 	}
 }
 
-/**
+/*
  * This function retrieves the string name of a named block. It is
  * more complicated than a simple memcpy() since the named block
  * descriptor may not be directly accessible.
@@ -199,7 +182,7 @@ static void CVMX_BOOTMEM_NAMED_GET_NAME(struct octeon_device *oct,
 
 /* See header file for descriptions of functions */
 
-/**
+/*
  * Check the version information on the bootmem descriptor
  *
  * @param exact_match
@@ -222,11 +205,11 @@ static int __cvmx_bootmem_check_version(struct octeon_device *oct,
 	major_version = (u32)__cvmx_bootmem_desc_get(
 			oct, oct->bootmem_desc_addr,
 			offsetof(struct cvmx_bootmem_desc, major_version),
-			FIELD_SIZEOF(struct cvmx_bootmem_desc, major_version));
+			sizeof_field(struct cvmx_bootmem_desc, major_version));
 	minor_version = (u32)__cvmx_bootmem_desc_get(
 			oct, oct->bootmem_desc_addr,
 			offsetof(struct cvmx_bootmem_desc, minor_version),
-			FIELD_SIZEOF(struct cvmx_bootmem_desc, minor_version));
+			sizeof_field(struct cvmx_bootmem_desc, minor_version));
 
 	dev_dbg(&oct->pci_dev->dev, "%s: major_version=%d\n", __func__,
 		major_version);
@@ -234,7 +217,7 @@ static int __cvmx_bootmem_check_version(struct octeon_device *oct,
 	    (exact_match && major_version != exact_match)) {
 		dev_err(&oct->pci_dev->dev, "bootmem ver mismatch %d.%d addr:0x%llx\n",
 			major_version, minor_version,
-			CAST_ULL(oct->bootmem_desc_addr));
+			(long long)oct->bootmem_desc_addr);
 		return -1;
 	} else {
 		return 0;
@@ -254,18 +237,17 @@ static const struct cvmx_bootmem_named_block_desc
 				oct, named_addr,
 				offsetof(struct cvmx_bootmem_named_block_desc,
 					 base_addr),
-				FIELD_SIZEOF(
+				sizeof_field(
 					struct cvmx_bootmem_named_block_desc,
 					base_addr));
 		desc->size = __cvmx_bootmem_desc_get(oct, named_addr,
 				offsetof(struct cvmx_bootmem_named_block_desc,
 					 size),
-				FIELD_SIZEOF(
+				sizeof_field(
 					struct cvmx_bootmem_named_block_desc,
 					size));
 
-		strncpy(desc->name, name, sizeof(desc->name));
-		desc->name[sizeof(desc->name) - 1] = 0;
+		strscpy(desc->name, name, sizeof(desc->name));
 		return &oct->bootmem_named_block_desc;
 	} else {
 		return NULL;
@@ -285,20 +267,20 @@ static u64 cvmx_bootmem_phy_named_block_find(struct octeon_device *oct,
 					oct, oct->bootmem_desc_addr,
 					offsetof(struct cvmx_bootmem_desc,
 						 named_block_array_addr),
-					FIELD_SIZEOF(struct cvmx_bootmem_desc,
+					sizeof_field(struct cvmx_bootmem_desc,
 						     named_block_array_addr));
 		u32 num_blocks = (u32)__cvmx_bootmem_desc_get(
 					oct, oct->bootmem_desc_addr,
 					offsetof(struct cvmx_bootmem_desc,
 						 nb_num_blocks),
-					FIELD_SIZEOF(struct cvmx_bootmem_desc,
+					sizeof_field(struct cvmx_bootmem_desc,
 						     nb_num_blocks));
 
 		u32 name_length = (u32)__cvmx_bootmem_desc_get(
 					oct, oct->bootmem_desc_addr,
 					offsetof(struct cvmx_bootmem_desc,
 						 named_block_name_len),
-					FIELD_SIZEOF(struct cvmx_bootmem_desc,
+					sizeof_field(struct cvmx_bootmem_desc,
 						     named_block_name_len));
 
 		u64 named_addr = named_block_array_addr;
@@ -309,7 +291,7 @@ static u64 cvmx_bootmem_phy_named_block_find(struct octeon_device *oct,
 					 offsetof(
 					struct cvmx_bootmem_named_block_desc,
 					size),
-					 FIELD_SIZEOF(
+					 sizeof_field(
 					struct cvmx_bootmem_named_block_desc,
 					size));
 
@@ -340,7 +322,7 @@ static u64 cvmx_bootmem_phy_named_block_find(struct octeon_device *oct,
 	return result;
 }
 
-/**
+/*
  * Find a named block on the remote Octeon
  *
  * @param name      Name of block to find
@@ -454,20 +436,31 @@ static void output_console_line(struct octeon_device *oct,
 {
 	char *line;
 	s32 i;
+	size_t len;
 
 	line = console_buffer;
 	for (i = 0; i < bytes_read; i++) {
 		/* Output a line at a time, prefixed */
 		if (console_buffer[i] == '\n') {
 			console_buffer[i] = '\0';
-			if (console->leftover[0]) {
-				dev_info(&oct->pci_dev->dev, "%lu: %s%s\n",
-					 console_num, console->leftover,
-					 line);
+			/* We need to output 'line', prefaced by 'leftover'.
+			 * However, it is possible we're being called to
+			 * output 'leftover' by itself (in the case of nothing
+			 * having been read from the console).
+			 *
+			 * To avoid duplication, check for this condition.
+			 */
+			if (console->leftover[0] &&
+			    (line != console->leftover)) {
+				if (console->print)
+					(*console->print)(oct, (u32)console_num,
+							  console->leftover,
+							  line);
 				console->leftover[0] = '\0';
 			} else {
-				dev_info(&oct->pci_dev->dev, "%lu: %s\n",
-					 console_num, line);
+				if (console->print)
+					(*console->print)(oct, (u32)console_num,
+							  line, NULL);
 			}
 			line = &console_buffer[i + 1];
 		}
@@ -476,13 +469,16 @@ static void output_console_line(struct octeon_device *oct,
 	/* Save off any leftovers */
 	if (line != &console_buffer[bytes_read]) {
 		console_buffer[bytes_read] = '\0';
-		strcpy(console->leftover, line);
+		len = strlen(console->leftover);
+		strscpy(&console->leftover[len], line,
+			sizeof(console->leftover) - len + 1);
 	}
 }
 
 static void check_console(struct work_struct *work)
 {
 	s32 bytes_read, tries, total_read;
+	size_t len;
 	struct octeon_console *console;
 	struct cavium_wk *wk = (struct cavium_wk *)work;
 	struct octeon_device *oct = (struct octeon_device *)wk->ctxptr;
@@ -504,7 +500,7 @@ static void check_console(struct work_struct *work)
 			total_read += bytes_read;
 			if (console->waiting)
 				octeon_console_handle_result(oct, console_num);
-			if (octeon_console_debug_enabled(console_num)) {
+			if (console->print) {
 				output_console_line(oct, console, console_num,
 						    console_buffer, bytes_read);
 			}
@@ -519,10 +515,13 @@ static void check_console(struct work_struct *work)
 	/* If nothing is read after polling the console,
 	 * output any leftovers if any
 	 */
-	if (octeon_console_debug_enabled(console_num) &&
-	    (total_read == 0) && (console->leftover[0])) {
-		dev_info(&oct->pci_dev->dev, "%u: %s\n",
-			 console_num, console->leftover);
+	if (console->print && (total_read == 0) &&
+	    (console->leftover[0])) {
+		/* append '\n' as terminator for 'output_console_line' */
+		len = strlen(console->leftover);
+		console->leftover[len] = '\n';
+		output_console_line(oct, console, console_num,
+				    console->leftover, (s32)(len + 1));
 		console->leftover[0] = '\0';
 	}
 
@@ -574,7 +573,84 @@ int octeon_init_consoles(struct octeon_device *oct)
 	return ret;
 }
 
-int octeon_add_console(struct octeon_device *oct, u32 console_num)
+static void octeon_get_uboot_version(struct octeon_device *oct)
+{
+	s32 bytes_read, tries, total_read;
+	struct octeon_console *console;
+	u32 console_num = 0;
+	char *uboot_ver;
+	char *buf;
+	char *p;
+
+#define OCTEON_UBOOT_VER_BUF_SIZE 512
+	buf = kmalloc(OCTEON_UBOOT_VER_BUF_SIZE, GFP_KERNEL);
+	if (!buf)
+		return;
+
+	if (octeon_console_send_cmd(oct, "setenv stdout pci\n", 50)) {
+		kfree(buf);
+		return;
+	}
+
+	if (octeon_console_send_cmd(oct, "version\n", 1)) {
+		kfree(buf);
+		return;
+	}
+
+	console = &oct->console[console_num];
+	tries = 0;
+	total_read = 0;
+
+	do {
+		/* Take console output regardless of whether it will
+		 * be logged
+		 */
+		bytes_read =
+			octeon_console_read(oct,
+					    console_num, buf + total_read,
+					    OCTEON_UBOOT_VER_BUF_SIZE - 1 -
+					    total_read);
+		if (bytes_read > 0) {
+			buf[bytes_read] = '\0';
+
+			total_read += bytes_read;
+			if (console->waiting)
+				octeon_console_handle_result(oct, console_num);
+		} else if (bytes_read < 0) {
+			dev_err(&oct->pci_dev->dev, "Error reading console %u, ret=%d\n",
+				console_num, bytes_read);
+		}
+
+		tries++;
+	} while ((bytes_read > 0) && (tries < 16));
+
+	/* If nothing is read after polling the console,
+	 * output any leftovers if any
+	 */
+	if ((total_read == 0) && (console->leftover[0])) {
+		dev_dbg(&oct->pci_dev->dev, "%u: %s\n",
+			console_num, console->leftover);
+		console->leftover[0] = '\0';
+	}
+
+	buf[OCTEON_UBOOT_VER_BUF_SIZE - 1] = '\0';
+
+	uboot_ver = strstr(buf, "U-Boot");
+	if (uboot_ver) {
+		p = strstr(uboot_ver, "mips");
+		if (p) {
+			p--;
+			*p = '\0';
+			dev_info(&oct->pci_dev->dev, "%s\n", uboot_ver);
+		}
+	}
+
+	kfree(buf);
+	octeon_console_send_cmd(oct, "setenv stdout serial\n", 50);
+}
+
+int octeon_add_console(struct octeon_device *oct, u32 console_num,
+		       char *dbg_enb)
 {
 	int ret = 0;
 	u32 delay;
@@ -610,17 +686,19 @@ int octeon_add_console(struct octeon_device *oct, u32 console_num)
 
 		work = &oct->console_poll_work[console_num].work;
 
+		octeon_get_uboot_version(oct);
+
 		INIT_DELAYED_WORK(work, check_console);
 		oct->console_poll_work[console_num].ctxptr = (void *)oct;
 		oct->console_poll_work[console_num].ctxul = console_num;
 		delay = OCTEON_CONSOLE_POLL_INTERVAL_MS;
 		schedule_delayed_work(work, msecs_to_jiffies(delay));
 
-		if (octeon_console_debug_enabled(console_num)) {
-			ret = octeon_console_send_cmd(oct,
-						      "setenv pci_console_active 1",
-						      2000);
-		}
+		/* an empty string means use default debug console enablement */
+		if (dbg_enb && !dbg_enb[0])
+			dbg_enb = "setenv pci_console_active 1";
+		if (dbg_enb)
+			ret = octeon_console_send_cmd(oct, dbg_enb, 2000);
 
 		console->active = 1;
 	}
@@ -628,7 +706,7 @@ int octeon_add_console(struct octeon_device *oct, u32 console_num)
 	return ret;
 }
 
-/**
+/*
  * Removes all consoles
  *
  * @param oct         octeon device
@@ -704,7 +782,7 @@ static int octeon_console_read(struct octeon_device *oct, u32 console_num,
 	if (bytes_to_read <= 0)
 		return bytes_to_read;
 
-	bytes_to_read = MIN(bytes_to_read, (s32)buf_size);
+	bytes_to_read = min_t(s32, bytes_to_read, buf_size);
 
 	/* Check to see if what we want to read is not contiguous, and limit
 	 * ourselves to the contiguous block
@@ -724,17 +802,18 @@ static int octeon_console_read(struct octeon_device *oct, u32 console_num,
 }
 
 #define FBUF_SIZE	(4 * 1024 * 1024)
-u8 fbuf[FBUF_SIZE];
+#define MAX_BOOTTIME_SIZE    80
 
 int octeon_download_firmware(struct octeon_device *oct, const u8 *data,
 			     size_t size)
 {
-	int ret = 0;
-	u8 *p = fbuf;
+	struct octeon_firmware_file_header *h;
+	char boottime[MAX_BOOTTIME_SIZE];
+	struct timespec64 ts;
 	u32 crc32_result;
 	u64 load_addr;
 	u32 image_len;
-	struct octeon_firmware_file_header *h;
+	int ret = 0;
 	u32 i, rem;
 
 	if (size < sizeof(struct octeon_firmware_file_header)) {
@@ -760,17 +839,11 @@ int octeon_download_firmware(struct octeon_device *oct, const u8 *data,
 		return -EINVAL;
 	}
 
-	if (strncmp(LIQUIDIO_PACKAGE, h->version, strlen(LIQUIDIO_PACKAGE))) {
-		dev_err(&oct->pci_dev->dev, "Unmatched firmware package type. Expected %s, got %s.\n",
-			LIQUIDIO_PACKAGE, h->version);
-		return -EINVAL;
-	}
-
-	if (memcmp(LIQUIDIO_BASE_VERSION, h->version + strlen(LIQUIDIO_PACKAGE),
+	if (memcmp(LIQUIDIO_BASE_VERSION, h->version,
 		   strlen(LIQUIDIO_BASE_VERSION))) {
 		dev_err(&oct->pci_dev->dev, "Unmatched firmware version. Expected %s.x, got %s.\n",
 			LIQUIDIO_BASE_VERSION,
-			h->version + strlen(LIQUIDIO_PACKAGE));
+			h->version);
 		return -EINVAL;
 	}
 
@@ -805,21 +878,42 @@ int octeon_download_firmware(struct octeon_device *oct, const u8 *data,
 			else
 				size = FBUF_SIZE;
 
-			memcpy(p, data, size);
-
 			/* download the image */
-			octeon_pci_write_core_mem(oct, load_addr, p, (u32)size);
+			octeon_pci_write_core_mem(oct, load_addr, data, (u32)size);
 
 			data += size;
 			rem -= (u32)size;
 			load_addr += size;
 		}
 	}
+
+	/* Pass date and time information to NIC at the time of loading
+	 * firmware and periodically update the host time to NIC firmware.
+	 * This is to make NIC firmware use the same time reference as Host,
+	 * so that it is easy to correlate logs from firmware and host for
+	 * debugging.
+	 *
+	 * Octeon always uses UTC time. so timezone information is not sent.
+	 */
+	ktime_get_real_ts64(&ts);
+	ret = snprintf(boottime, MAX_BOOTTIME_SIZE,
+		       " time_sec=%lld time_nsec=%ld",
+		       (s64)ts.tv_sec, ts.tv_nsec);
+	if ((sizeof(h->bootcmd) - strnlen(h->bootcmd, sizeof(h->bootcmd))) <
+		ret) {
+		dev_err(&oct->pci_dev->dev, "Boot command buffer too small\n");
+		return -EINVAL;
+	}
+	strncat(h->bootcmd, boottime,
+		sizeof(h->bootcmd) - strnlen(h->bootcmd, sizeof(h->bootcmd)));
+
 	dev_info(&oct->pci_dev->dev, "Writing boot command: %s\n",
 		 h->bootcmd);
 
 	/* Invoke the bootcmd */
 	ret = octeon_console_send_cmd(oct, h->bootcmd, 50);
+	if (ret)
+		dev_info(&oct->pci_dev->dev, "Boot command send failed\n");
 
-	return 0;
+	return ret;
 }
diff --git a/drivers/net/ethernet/cavium/liquidio/octeon_device.c b/drivers/net/ethernet/cavium/liquidio/octeon_device.c
index 3b7cc9320deb..1753bb87dfbd 100644
--- a/drivers/net/ethernet/cavium/liquidio/octeon_device.c
+++ b/drivers/net/ethernet/cavium/liquidio/octeon_device.c
@@ -51,7 +51,6 @@ static struct octeon_config default_cn66xx_conf = {
 	/** OQ attributes */
 	.oq					= {
 		.max_oqs			= CN6XXX_CFG_IO_QUEUES,
-		.info_ptr			= OCTEON_OQ_INFOPTR_MODE,
 		.refill_threshold		= CN6XXX_OQ_REFIL_THRESHOLD,
 		.oq_intr_pkt			= CN6XXX_OQ_INTR_PKT,
 		.oq_intr_time			= CN6XXX_OQ_INTR_TIME,
@@ -161,7 +160,6 @@ static struct octeon_config default_cn68xx_conf = {
 	/** OQ attributes */
 	.oq					= {
 		.max_oqs			= CN6XXX_CFG_IO_QUEUES,
-		.info_ptr			= OCTEON_OQ_INFOPTR_MODE,
 		.refill_threshold		= CN6XXX_OQ_REFIL_THRESHOLD,
 		.oq_intr_pkt			= CN6XXX_OQ_INTR_PKT,
 		.oq_intr_time			= CN6XXX_OQ_INTR_TIME,
@@ -328,7 +326,6 @@ static struct octeon_config default_cn68xx_210nv_conf = {
 	/** OQ attributes */
 	.oq					= {
 		.max_oqs			= CN6XXX_CFG_IO_QUEUES,
-		.info_ptr			= OCTEON_OQ_INFOPTR_MODE,
 		.refill_threshold		= CN6XXX_OQ_REFIL_THRESHOLD,
 		.oq_intr_pkt			= CN6XXX_OQ_INTR_PKT,
 		.oq_intr_time			= CN6XXX_OQ_INTR_TIME,
@@ -421,7 +418,7 @@ static struct octeon_config default_cn23xx_conf = {
 	/** IQ attributes */
 	.iq = {
 		.max_iqs		= CN23XX_CFG_IO_QUEUES,
-		.pending_list_size	= (CN23XX_MAX_IQ_DESCRIPTORS *
+		.pending_list_size	= (CN23XX_DEFAULT_IQ_DESCRIPTORS *
 					   CN23XX_CFG_IO_QUEUES),
 		.instr_type		= OCTEON_64BYTE_INSTR,
 		.db_min			= CN23XX_DB_MIN,
@@ -432,7 +429,6 @@ static struct octeon_config default_cn23xx_conf = {
 	/** OQ attributes */
 	.oq = {
 		.max_oqs		= CN23XX_CFG_IO_QUEUES,
-		.info_ptr		= OCTEON_OQ_INFOPTR_MODE,
 		.pkts_per_intr	= CN23XX_OQ_PKTSPER_INTR,
 		.refill_threshold	= CN23XX_OQ_REFIL_THRESHOLD,
 		.oq_intr_pkt	= CN23XX_OQ_INTR_PKT,
@@ -440,8 +436,8 @@ static struct octeon_config default_cn23xx_conf = {
 	},
 
 	.num_nic_ports				= DEFAULT_NUM_NIC_PORTS_23XX,
-	.num_def_rx_descs			= CN23XX_MAX_OQ_DESCRIPTORS,
-	.num_def_tx_descs			= CN23XX_MAX_IQ_DESCRIPTORS,
+	.num_def_rx_descs			= CN23XX_DEFAULT_OQ_DESCRIPTORS,
+	.num_def_tx_descs			= CN23XX_DEFAULT_IQ_DESCRIPTORS,
 	.def_rx_buf_size			= CN23XX_OQ_BUF_SIZE,
 
 	/* For ethernet interface 0:  Port cfg Attributes */
@@ -459,10 +455,10 @@ static struct octeon_config default_cn23xx_conf = {
 		.num_rxqs			= DEF_RXQS_PER_INTF,
 
 		/* Num of desc for rx rings */
-		.num_rx_descs			= CN23XX_MAX_OQ_DESCRIPTORS,
+		.num_rx_descs			= CN23XX_DEFAULT_OQ_DESCRIPTORS,
 
 		/* Num of desc for tx rings */
-		.num_tx_descs			= CN23XX_MAX_IQ_DESCRIPTORS,
+		.num_tx_descs			= CN23XX_DEFAULT_IQ_DESCRIPTORS,
 
 		/* SKB size, We need not change buf size even for Jumbo frames.
 		 * Octeon can send jumbo frames in 4 consecutive descriptors,
@@ -488,10 +484,10 @@ static struct octeon_config default_cn23xx_conf = {
 		.num_rxqs			= DEF_RXQS_PER_INTF,
 
 		/* Num of desc for rx rings */
-		.num_rx_descs			= CN23XX_MAX_OQ_DESCRIPTORS,
+		.num_rx_descs			= CN23XX_DEFAULT_OQ_DESCRIPTORS,
 
 		/* Num of desc for tx rings */
-		.num_tx_descs			= CN23XX_MAX_IQ_DESCRIPTORS,
+		.num_tx_descs			= CN23XX_DEFAULT_IQ_DESCRIPTORS,
 
 		/* SKB size, We need not change buf size even for Jumbo frames.
 		 * Octeon can send jumbo frames in 4 consecutive descriptors,
@@ -532,9 +528,10 @@ static struct octeon_config_ptr {
 };
 
 static char oct_dev_state_str[OCT_DEV_STATES + 1][32] = {
-	"BEGIN", "PCI-MAP-DONE", "DISPATCH-INIT-DONE",
+	"BEGIN", "PCI-ENABLE-DONE", "PCI-MAP-DONE", "DISPATCH-INIT-DONE",
 	"IQ-INIT-DONE", "SCBUFF-POOL-INIT-DONE", "RESPLIST-INIT-DONE",
-	"DROQ-INIT-DONE", "IO-QUEUES-INIT-DONE", "CONSOLE-INIT-DONE",
+	"DROQ-INIT-DONE", "MBOX-SETUP-DONE", "MSIX-ALLOC-VECTOR-DONE",
+	"INTR-SET-DONE", "IO-QUEUES-INIT-DONE", "CONSOLE-INIT-DONE",
 	"HOST-READY", "CORE-READY", "RUNNING", "IN-RESET",
 	"INVALID"
 };
@@ -544,10 +541,11 @@ static char oct_dev_app_str[CVM_DRV_APP_COUNT + 1][32] = {
 
 static struct octeon_device *octeon_device[MAX_OCTEON_DEVICES];
 static atomic_t adapter_refcounts[MAX_OCTEON_DEVICES];
+static atomic_t adapter_fw_states[MAX_OCTEON_DEVICES];
 
 static u32 octeon_device_count;
 /* locks device array (i.e. octeon_device[]) */
-static spinlock_t octeon_devices_lock;
+static DEFINE_SPINLOCK(octeon_devices_lock);
 
 static struct octeon_core_setup core_setup[MAX_OCTEON_DEVICES];
 
@@ -565,8 +563,8 @@ void octeon_init_device_list(int conf_type)
 	memset(octeon_device, 0, (sizeof(void *) * MAX_OCTEON_DEVICES));
 	for (i = 0; i <  MAX_OCTEON_DEVICES; i++)
 		oct_set_config_info(i, conf_type);
-	spin_lock_init(&octeon_devices_lock);
 }
+EXPORT_SYMBOL_GPL(octeon_init_device_list);
 
 static void *__retrieve_octeon_config_info(struct octeon_device *oct,
 					   u16 card_type)
@@ -636,6 +634,7 @@ char *lio_get_state_string(atomic_t *state_ptr)
 		return oct_dev_state_str[OCT_DEV_STATE_INVALID];
 	return oct_dev_state_str[istate];
 }
+EXPORT_SYMBOL_GPL(lio_get_state_string);
 
 static char *get_oct_app_string(u32 app_mode)
 {
@@ -664,6 +663,7 @@ void octeon_free_device_mem(struct octeon_device *oct)
 	octeon_device[i] = NULL;
 	octeon_device_count--;
 }
+EXPORT_SYMBOL_GPL(octeon_free_device_mem);
 
 static struct octeon_device *octeon_allocate_device_mem(u32 pci_id,
 							u32 priv_size)
@@ -704,12 +704,10 @@ static struct octeon_device *octeon_allocate_device_mem(u32 pci_id,
 	size = octdevsize + priv_size + configsize +
 		(sizeof(struct octeon_dispatch) * DISPATCH_LIST_SIZE);
 
-	buf = vmalloc(size);
+	buf = vzalloc(size);
 	if (!buf)
 		return NULL;
 
-	memset(buf, 0, size);
-
 	oct = (struct octeon_device *)buf;
 	oct->priv = (void *)(buf + octdevsize);
 	oct->chip = (void *)(buf + octdevsize + priv_size);
@@ -752,6 +750,7 @@ struct octeon_device *octeon_allocate_device(u32 pci_id,
 
 	return oct;
 }
+EXPORT_SYMBOL_GPL(octeon_allocate_device);
 
 /** Register a device's bus location at initialization time.
  *  @param octeon_dev - pointer to the octeon device structure.
@@ -773,6 +772,10 @@ int octeon_register_device(struct octeon_device *oct,
 	oct->adapter_refcount = &adapter_refcounts[oct->octeon_id];
 	atomic_set(oct->adapter_refcount, 0);
 
+	/* Like the reference count, the f/w state is shared 'per-adapter' */
+	oct->adapter_fw_state = &adapter_fw_states[oct->octeon_id];
+	atomic_set(oct->adapter_fw_state, FW_NEEDS_TO_BE_LOADED);
+
 	spin_lock(&octeon_devices_lock);
 	for (idx = (int)oct->octeon_id - 1; idx >= 0; idx--) {
 		if (!octeon_device[idx]) {
@@ -783,11 +786,15 @@ int octeon_register_device(struct octeon_device *oct,
 			atomic_inc(oct->adapter_refcount);
 			return 1; /* here, refcount is guaranteed to be 1 */
 		}
-		/* if another device is at same bus/dev, use its refcounter */
+		/* If another device is at same bus/dev, use its refcounter
+		 * (and f/w state variable).
+		 */
 		if ((octeon_device[idx]->loc.bus == bus) &&
 		    (octeon_device[idx]->loc.dev == dev)) {
 			oct->adapter_refcount =
 				octeon_device[idx]->adapter_refcount;
+			oct->adapter_fw_state =
+				octeon_device[idx]->adapter_fw_state;
 			break;
 		}
 	}
@@ -801,6 +808,7 @@ int octeon_register_device(struct octeon_device *oct,
 
 	return refcount;
 }
+EXPORT_SYMBOL_GPL(octeon_register_device);
 
 /** Deregister a device at de-initialization time.
  *  @param octeon_dev - pointer to the octeon device structure.
@@ -818,26 +826,21 @@ int octeon_deregister_device(struct octeon_device *oct)
 
 	return refcount;
 }
+EXPORT_SYMBOL_GPL(octeon_deregister_device);
 
 int
-octeon_allocate_ioq_vector(struct octeon_device  *oct)
+octeon_allocate_ioq_vector(struct octeon_device *oct, u32 num_ioqs)
 {
-	int i, num_ioqs = 0;
 	struct octeon_ioq_vector *ioq_vector;
 	int cpu_num;
 	int size;
-
-	if (OCTEON_CN23XX_PF(oct))
-		num_ioqs = oct->sriov_info.num_pf_rings;
-	else if (OCTEON_CN23XX_VF(oct))
-		num_ioqs = oct->sriov_info.rings_per_vf;
+	int i;
 
 	size = sizeof(struct octeon_ioq_vector) * num_ioqs;
 
-	oct->ioq_vector = vmalloc(size);
+	oct->ioq_vector = vzalloc(size);
 	if (!oct->ioq_vector)
-		return 1;
-	memset(oct->ioq_vector, 0, size);
+		return -1;
 	for (i = 0; i < num_ioqs; i++) {
 		ioq_vector		= &oct->ioq_vector[i];
 		ioq_vector->oct_dev	= oct;
@@ -853,14 +856,17 @@ octeon_allocate_ioq_vector(struct octeon_device  *oct)
 		else
 			ioq_vector->ioq_num	= i;
 	}
+
 	return 0;
 }
+EXPORT_SYMBOL_GPL(octeon_allocate_ioq_vector);
 
 void
 octeon_free_ioq_vector(struct octeon_device *oct)
 {
 	vfree(oct->ioq_vector);
 }
+EXPORT_SYMBOL_GPL(octeon_free_ioq_vector);
 
 /* this function is only for setting up the first queue */
 int octeon_setup_instr_queues(struct octeon_device *oct)
@@ -880,11 +886,11 @@ int octeon_setup_instr_queues(struct octeon_device *oct)
 
 	oct->num_iqs = 0;
 
-	oct->instr_queue[0] = vmalloc_node(sizeof(*oct->instr_queue[0]),
+	oct->instr_queue[0] = vzalloc_node(sizeof(*oct->instr_queue[0]),
 				numa_node);
 	if (!oct->instr_queue[0])
 		oct->instr_queue[0] =
-			vmalloc(sizeof(struct octeon_instr_queue));
+			vzalloc(sizeof(struct octeon_instr_queue));
 	if (!oct->instr_queue[0])
 		return 1;
 	memset(oct->instr_queue[0], 0, sizeof(struct octeon_instr_queue));
@@ -906,6 +912,7 @@ int octeon_setup_instr_queues(struct octeon_device *oct)
 	oct->num_iqs++;
 	return 0;
 }
+EXPORT_SYMBOL_GPL(octeon_setup_instr_queues);
 
 int octeon_setup_output_queues(struct octeon_device *oct)
 {
@@ -927,9 +934,9 @@ int octeon_setup_output_queues(struct octeon_device *oct)
 		desc_size = CFG_GET_DEF_RX_BUF_SIZE(CHIP_CONF(oct, cn23xx_vf));
 	}
 	oct->num_oqs = 0;
-	oct->droq[0] = vmalloc_node(sizeof(*oct->droq[0]), numa_node);
+	oct->droq[0] = vzalloc_node(sizeof(*oct->droq[0]), numa_node);
 	if (!oct->droq[0])
-		oct->droq[0] = vmalloc(sizeof(*oct->droq[0]));
+		oct->droq[0] = vzalloc(sizeof(*oct->droq[0]));
 	if (!oct->droq[0])
 		return 1;
 
@@ -942,6 +949,7 @@ int octeon_setup_output_queues(struct octeon_device *oct)
 
 	return 0;
 }
+EXPORT_SYMBOL_GPL(octeon_setup_output_queues);
 
 int octeon_set_io_queues_off(struct octeon_device *oct)
 {
@@ -991,6 +999,7 @@ int octeon_set_io_queues_off(struct octeon_device *oct)
 	}
 	return 0;
 }
+EXPORT_SYMBOL_GPL(octeon_set_io_queues_off);
 
 void octeon_set_droq_pkt_op(struct octeon_device *oct,
 			    u32 q_no,
@@ -1029,6 +1038,7 @@ int octeon_init_dispatch_list(struct octeon_device *oct)
 
 	return 0;
 }
+EXPORT_SYMBOL_GPL(octeon_init_dispatch_list);
 
 void octeon_delete_dispatch_list(struct octeon_device *oct)
 {
@@ -1045,8 +1055,7 @@ void octeon_delete_dispatch_list(struct octeon_device *oct)
 		dispatch = &oct->dispatch.dlist[i].list;
 		while (dispatch->next != dispatch) {
 			temp = dispatch->next;
-			list_del(temp);
-			list_add_tail(temp, &freelist);
+			list_move_tail(temp, &freelist);
 		}
 
 		oct->dispatch.dlist[i].opcode = 0;
@@ -1058,9 +1067,10 @@ void octeon_delete_dispatch_list(struct octeon_device *oct)
 
 	list_for_each_safe(temp, tmp2, &freelist) {
 		list_del(temp);
-		vfree(temp);
+		kfree(temp);
 	}
 }
+EXPORT_SYMBOL_GPL(octeon_delete_dispatch_list);
 
 octeon_dispatch_fn_t
 octeon_get_dispatch(struct octeon_device *octeon_dev, u16 opcode,
@@ -1154,13 +1164,10 @@ octeon_register_dispatch_fn(struct octeon_device *oct,
 
 		dev_dbg(&oct->pci_dev->dev,
 			"Adding opcode to dispatch list linked list\n");
-		dispatch = (struct octeon_dispatch *)
-			   vmalloc(sizeof(struct octeon_dispatch));
-		if (!dispatch) {
-			dev_err(&oct->pci_dev->dev,
-				"No memory to add dispatch function\n");
+		dispatch = kmalloc(sizeof(*dispatch), GFP_KERNEL);
+		if (!dispatch)
 			return 1;
-		}
+
 		dispatch->opcode = combined_opcode;
 		dispatch->dispatch_fn = fn;
 		dispatch->arg = fn_arg;
@@ -1174,6 +1181,10 @@ octeon_register_dispatch_fn(struct octeon_device *oct,
 		spin_unlock_bh(&oct->dispatch.lock);
 
 	} else {
+		if (pfn == fn &&
+		    octeon_get_dispatch_arg(oct, opcode, subcode) == fn_arg)
+			return 0;
+
 		dev_err(&oct->pci_dev->dev,
 			"Found previously registered dispatch fn for opcode/subcode: %x/%x\n",
 			opcode, subcode);
@@ -1182,6 +1193,7 @@ octeon_register_dispatch_fn(struct octeon_device *oct,
 
 	return 0;
 }
+EXPORT_SYMBOL_GPL(octeon_register_dispatch_fn);
 
 int octeon_core_drv_init(struct octeon_recv_info *recv_info, void *buf)
 {
@@ -1205,10 +1217,10 @@ int octeon_core_drv_init(struct octeon_recv_info *recv_info, void *buf)
 		goto core_drv_init_err;
 	}
 
-	strncpy(app_name,
+	strscpy(app_name,
 		get_oct_app_string(
 		(u32)recv_pkt->rh.r_core_drv_init.app_mode),
-		sizeof(app_name) - 1);
+		sizeof(app_name));
 	oct->app_mode = (u32)recv_pkt->rh.r_core_drv_init.app_mode;
 	if (recv_pkt->rh.r_core_drv_init.app_mode == CVM_DRV_NIC_APP) {
 		oct->fw_info.max_nic_ports =
@@ -1236,16 +1248,19 @@ int octeon_core_drv_init(struct octeon_recv_info *recv_info, void *buf)
 
 	cs = &core_setup[oct->octeon_id];
 
-	if (recv_pkt->buffer_size[0] != sizeof(*cs)) {
+	if (recv_pkt->buffer_size[0] != (sizeof(*cs) + OCT_DROQ_INFO_SIZE)) {
 		dev_dbg(&oct->pci_dev->dev, "Core setup bytes expected %u found %d\n",
 			(u32)sizeof(*cs),
 			recv_pkt->buffer_size[0]);
 	}
 
-	memcpy(cs, get_rbd(recv_pkt->buffer_ptr[0]), sizeof(*cs));
-	strncpy(oct->boardinfo.name, cs->boardname, OCT_BOARD_NAME);
-	strncpy(oct->boardinfo.serial_number, cs->board_serial_number,
-		OCT_SERIAL_LEN);
+	memcpy(cs, get_rbd(
+	       recv_pkt->buffer_ptr[0]) + OCT_DROQ_INFO_SIZE, sizeof(*cs));
+
+	strscpy(oct->boardinfo.name, cs->boardname,
+		    sizeof(oct->boardinfo.name));
+	strscpy(oct->boardinfo.serial_number, cs->board_serial_number,
+		    sizeof(oct->boardinfo.serial_number));
 
 	octeon_swap_8B_data((u64 *)cs, (sizeof(*cs) >> 3));
 
@@ -1262,6 +1277,7 @@ core_drv_init_err:
 	octeon_free_recv_info(recv_info);
 	return 0;
 }
+EXPORT_SYMBOL_GPL(octeon_core_drv_init);
 
 int octeon_get_tx_qsize(struct octeon_device *oct, u32 q_no)
 
@@ -1272,6 +1288,7 @@ int octeon_get_tx_qsize(struct octeon_device *oct, u32 q_no)
 
 	return -1;
 }
+EXPORT_SYMBOL_GPL(octeon_get_tx_qsize);
 
 int octeon_get_rx_qsize(struct octeon_device *oct, u32 q_no)
 {
@@ -1280,6 +1297,7 @@ int octeon_get_rx_qsize(struct octeon_device *oct, u32 q_no)
 		return oct->droq[q_no]->max_count;
 	return -1;
 }
+EXPORT_SYMBOL_GPL(octeon_get_rx_qsize);
 
 /* Retruns the host firmware handshake OCTEON specific configuration */
 struct octeon_config *octeon_get_conf(struct octeon_device *oct)
@@ -1302,11 +1320,12 @@ struct octeon_config *octeon_get_conf(struct octeon_device *oct)
 	}
 	return default_oct_conf;
 }
+EXPORT_SYMBOL_GPL(octeon_get_conf);
 
 /* scratch register address is same in all the OCT-II and CN70XX models */
 #define CNXX_SLI_SCRATCH1   0x3C0
 
-/** Get the octeon device pointer.
+/* Get the octeon device pointer.
  *  @param octeon_id  - The id for which the octeon device pointer is required.
  *  @return Success: Octeon device pointer.
  *  @return Failure: NULL.
@@ -1318,12 +1337,13 @@ struct octeon_device *lio_get_device(u32 octeon_id)
 	else
 		return octeon_device[octeon_id];
 }
+EXPORT_SYMBOL_GPL(lio_get_device);
 
 u64 lio_pci_readq(struct octeon_device *oct, u64 addr)
 {
 	u64 val64;
 	unsigned long flags;
-	u32 val32, addrhi;
+	u32 addrhi;
 
 	spin_lock_irqsave(&oct->pci_win_lock, flags);
 
@@ -1338,10 +1358,10 @@ u64 lio_pci_readq(struct octeon_device *oct, u64 addr)
 	writel(addrhi, oct->reg_list.pci_win_rd_addr_hi);
 
 	/* Read back to preserve ordering of writes */
-	val32 = readl(oct->reg_list.pci_win_rd_addr_hi);
+	readl(oct->reg_list.pci_win_rd_addr_hi);
 
 	writel(addr & 0xffffffff, oct->reg_list.pci_win_rd_addr_lo);
-	val32 = readl(oct->reg_list.pci_win_rd_addr_lo);
+	readl(oct->reg_list.pci_win_rd_addr_lo);
 
 	val64 = readq(oct->reg_list.pci_win_rd_data);
 
@@ -1349,12 +1369,12 @@ u64 lio_pci_readq(struct octeon_device *oct, u64 addr)
 
 	return val64;
 }
+EXPORT_SYMBOL_GPL(lio_pci_readq);
 
 void lio_pci_writeq(struct octeon_device *oct,
 		    u64 val,
 		    u64 addr)
 {
-	u32 val32;
 	unsigned long flags;
 
 	spin_lock_irqsave(&oct->pci_win_lock, flags);
@@ -1364,12 +1384,13 @@ void lio_pci_writeq(struct octeon_device *oct,
 	/* The write happens when the LSB is written. So write MSB first. */
 	writel(val >> 32, oct->reg_list.pci_win_wr_data_hi);
 	/* Read the MSB to ensure ordering of writes. */
-	val32 = readl(oct->reg_list.pci_win_wr_data_hi);
+	readl(oct->reg_list.pci_win_wr_data_hi);
 
 	writel(val & 0xffffffff, oct->reg_list.pci_win_wr_data_lo);
 
 	spin_unlock_irqrestore(&oct->pci_win_lock, flags);
 }
+EXPORT_SYMBOL_GPL(lio_pci_writeq);
 
 int octeon_mem_access_ok(struct octeon_device *oct)
 {
@@ -1389,6 +1410,7 @@ int octeon_mem_access_ok(struct octeon_device *oct)
 
 	return access_okay ? 0 : 1;
 }
+EXPORT_SYMBOL_GPL(octeon_mem_access_ok);
 
 int octeon_wait_for_ddr_init(struct octeon_device *oct, u32 *timeout)
 {
@@ -1409,22 +1431,7 @@ int octeon_wait_for_ddr_init(struct octeon_device *oct, u32 *timeout)
 
 	return ret;
 }
-
-/** Get the octeon id assigned to the octeon device passed as argument.
- *  This function is exported to other modules.
- *  @param dev - octeon device pointer passed as a void *.
- *  @return octeon device id
- */
-int lio_get_device_id(void *dev)
-{
-	struct octeon_device *octeon_dev = (struct octeon_device *)dev;
-	u32 i;
-
-	for (i = 0; i < MAX_OCTEON_DEVICES; i++)
-		if (octeon_device[i] == octeon_dev)
-			return octeon_dev->octeon_id;
-	return -1;
-}
+EXPORT_SYMBOL_GPL(octeon_wait_for_ddr_init);
 
 void lio_enable_irq(struct octeon_droq *droq, struct octeon_instr_queue *iq)
 {
@@ -1435,20 +1442,16 @@ void lio_enable_irq(struct octeon_droq *droq, struct octeon_instr_queue *iq)
 	/* the whole thing needs to be atomic, ideally */
 	if (droq) {
 		pkts_pend = (u32)atomic_read(&droq->pkts_pending);
-		spin_lock_bh(&droq->lock);
 		writel(droq->pkt_count - pkts_pend, droq->pkts_sent_reg);
 		droq->pkt_count = pkts_pend;
-		/* this write needs to be flushed before we release the lock */
-		mmiowb();
-		spin_unlock_bh(&droq->lock);
 		oct = droq->oct_dev;
 	}
 	if (iq) {
 		spin_lock_bh(&iq->lock);
-		writel(iq->pkt_in_done, iq->inst_cnt_reg);
-		iq->pkt_in_done = 0;
+		writel(iq->pkts_processed, iq->inst_cnt_reg);
+		iq->pkt_in_done -= iq->pkts_processed;
+		iq->pkts_processed = 0;
 		/* this write needs to be flushed before we release the lock */
-		mmiowb();
 		spin_unlock_bh(&iq->lock);
 		oct = iq->oct_dev;
 	}
@@ -1467,3 +1470,4 @@ void lio_enable_irq(struct octeon_droq *droq, struct octeon_instr_queue *iq)
 		}
 	}
 }
+EXPORT_SYMBOL_GPL(lio_enable_irq);
diff --git a/drivers/net/ethernet/cavium/liquidio/octeon_device.h b/drivers/net/ethernet/cavium/liquidio/octeon_device.h
index c90ed48ae8ab..19344b21f8fb 100644
--- a/drivers/net/ethernet/cavium/liquidio/octeon_device.h
+++ b/drivers/net/ethernet/cavium/liquidio/octeon_device.h
@@ -22,6 +22,9 @@
 #ifndef _OCTEON_DEVICE_H_
 #define  _OCTEON_DEVICE_H_
 
+#include <linux/interrupt.h>
+#include <net/devlink.h>
+
 /** PCI VendorId Device Id */
 #define  OCTEON_CN68XX_PCIID          0x91177d
 #define  OCTEON_CN66XX_PCIID          0x92177d
@@ -40,6 +43,13 @@
 #define  OCTEON_CN23XX_REV_1_1        0x01
 #define  OCTEON_CN23XX_REV_2_0        0x80
 
+/**SubsystemId for the chips */
+#define	 OCTEON_CN2350_10GB_SUBSYS_ID_1	0X3177d
+#define	 OCTEON_CN2350_10GB_SUBSYS_ID_2	0X4177d
+#define	 OCTEON_CN2360_10GB_SUBSYS_ID	0X5177d
+#define	 OCTEON_CN2350_25GB_SUBSYS_ID	0X7177d
+#define	 OCTEON_CN2360_25GB_SUBSYS_ID	0X6177d
+
 /** Endian-swap modes supported by Octeon. */
 enum octeon_pci_swap_mode {
 	OCTEON_PCI_PASSTHROUGH = 0,
@@ -48,6 +58,13 @@ enum octeon_pci_swap_mode {
 	OCTEON_PCI_32BIT_LW_SWAP = 3
 };
 
+enum lio_fw_state {
+	FW_IS_PRELOADED = 0,
+	FW_NEEDS_TO_BE_LOADED = 1,
+	FW_IS_BEING_LOADED = 2,
+	FW_HAS_BEEN_LOADED = 3,
+};
+
 enum {
 	OCTEON_CONFIG_TYPE_DEFAULT = 0,
 	NUM_OCTEON_CONFS,
@@ -192,6 +209,8 @@ struct octeon_reg_list {
 };
 
 #define OCTEON_CONSOLE_MAX_READ_BYTES 512
+typedef int (*octeon_console_print_fn)(struct octeon_device *oct,
+				       u32 num, char *pre, char *suf);
 struct octeon_console {
 	u32 active;
 	u32 waiting;
@@ -199,6 +218,7 @@ struct octeon_console {
 	u32 buffer_size;
 	u64 input_base_addr;
 	u64 output_base_addr;
+	octeon_console_print_fn print;
 	char leftover[OCTEON_CONSOLE_MAX_READ_BYTES];
 };
 
@@ -268,8 +288,17 @@ struct oct_fw_info {
 	 */
 	u32 app_mode;
 	char   liquidio_firmware_version[32];
+	/* Fields extracted from legacy string 'liquidio_firmware_version' */
+	struct {
+		u8  maj;
+		u8  min;
+		u8  rev;
+	} ver;
 };
 
+#define OCT_FW_VER(maj, min, rev) \
+	(((u32)(maj) << 16) | ((u32)(min) << 8) | ((u32)(rev)))
+
 /* wrappers around work structs */
 struct cavium_wk {
 	struct delayed_work work;
@@ -287,6 +316,8 @@ struct octdev_props {
 	 * device pointer (used for OS specific calls).
 	 */
 	int    rx_on;
+	int    fec;
+	int    fec_boot;
 	int    napi_enabled;
 	int    gmxport;
 	struct net_device *netdev;
@@ -357,6 +388,8 @@ struct octeon_sriov_info {
 
 	u32	sriov_enabled;
 
+	struct lio_trusted_vf	trusted_vf;
+
 	/*lookup table that maps DPI ring number to VF pci_dev struct pointer*/
 	struct pci_dev *dpiring_to_vfpcidev_lut[MAX_POSSIBLE_VFS];
 
@@ -366,6 +399,8 @@ struct octeon_sriov_info {
 
 	int	vf_linkstate[MAX_POSSIBLE_VFS];
 
+	bool    vf_spoofchk[MAX_POSSIBLE_VFS];
+
 	u64	vf_drv_loaded_mask;
 };
 
@@ -379,6 +414,15 @@ struct octeon_ioq_vector {
 	u32			ioq_num;
 };
 
+struct lio_vf_rep_list {
+	int num_vfs;
+	struct net_device *ndev[CN23XX_MAX_VFS_PER_PF];
+};
+
+struct lio_devlink_priv {
+	struct octeon_device *oct;
+};
+
 /** The Octeon device.
  *  Each Octeon device has this structure to represent all its
  *  components.
@@ -406,6 +450,8 @@ struct octeon_device {
 
 	u16 rev_id;
 
+	u32 subsystem_id;
+
 	u16 pf_num;
 
 	u16 vf_num;
@@ -552,6 +598,22 @@ struct octeon_device {
 	} loc;
 
 	atomic_t *adapter_refcount; /* reference count of adapter */
+
+	atomic_t *adapter_fw_state; /* per-adapter, lio_fw_state */
+
+	bool ptp_enable;
+
+	struct lio_vf_rep_list vf_rep_list;
+	struct devlink *devlink;
+	enum devlink_eswitch_mode eswitch_mode;
+
+	/* for 25G NIC speed change */
+	u8  speed_boot;
+	u8  speed_setting;
+	u8  no_speed_setting;
+
+	u32    vfstats_poll;
+#define LIO_VFSTATS_POLL 10
 };
 
 #define  OCT_DRV_ONLINE 1
@@ -565,6 +627,8 @@ struct octeon_device {
 #define CHIP_CONF(oct, TYPE)             \
 	(((struct octeon_ ## TYPE  *)((oct)->chip))->conf)
 
+#define MAX_IO_PENDING_PKT_COUNT 100
+
 /*------------------ Function Prototypes ----------------------*/
 
 /** Initialize device list memory */
@@ -641,25 +705,6 @@ octeon_get_dispatch(struct octeon_device *octeon_dev, u16 opcode,
  */
 struct octeon_device *lio_get_device(u32 octeon_id);
 
-/** Get the octeon id assigned to the octeon device passed as argument.
- *  This function is exported to other modules.
- *  @param dev - octeon device pointer passed as a void *.
- *  @return octeon device id
- */
-int lio_get_device_id(void *dev);
-
-static inline u16 OCTEON_MAJOR_REV(struct octeon_device *oct)
-{
-	u16 rev = (oct->rev_id & 0xC) >> 2;
-
-	return (rev == 0) ? 1 : rev;
-}
-
-static inline u16 OCTEON_MINOR_REV(struct octeon_device *oct)
-{
-	return oct->rev_id & 0x3;
-}
-
 /** Read windowed register.
  *  @param  oct   -  pointer to the Octeon device.
  *  @param  addr  -  Address of the register to read.
@@ -740,18 +785,17 @@ int octeon_init_consoles(struct octeon_device *oct);
 /**
  * Adds access to a console to the device.
  *
- * @param oct which octeon to add to
- * @param console_num which console
+ * @param oct:          which octeon to add to
+ * @param console_num:  which console
+ * @param dbg_enb:      ptr to debug enablement string, one of:
+ *                    * NULL for no debug output (i.e. disabled)
+ *                    * empty string enables debug output (via default method)
+ *                    * specific string to enable debug console output
+ *
  * @return Zero on success, negative on failure.
  */
-int octeon_add_console(struct octeon_device *oct, u32 console_num);
-
-/** write or read from a console */
-int octeon_console_write(struct octeon_device *oct, u32 console_num,
-			 char *buffer, u32 write_request_size, u32 flags);
-int octeon_console_write_avail(struct octeon_device *oct, u32 console_num);
-
-int octeon_console_read_avail(struct octeon_device *oct, u32 console_num);
+int octeon_add_console(struct octeon_device *oct, u32 console_num,
+		       char *dbg_enb);
 
 /** Removes all attached consoles. */
 void octeon_remove_consoles(struct octeon_device *oct);
@@ -827,7 +871,7 @@ void *oct_get_config_info(struct octeon_device *oct, u16 card_type);
 struct octeon_config *octeon_get_conf(struct octeon_device *oct);
 
 void octeon_free_ioq_vector(struct octeon_device *oct);
-int octeon_allocate_ioq_vector(struct octeon_device  *oct);
+int octeon_allocate_ioq_vector(struct octeon_device  *oct, u32 num_ioqs);
 void lio_enable_irq(struct octeon_droq *droq, struct octeon_instr_queue *iq);
 
 /* LiquidIO driver pivate flags */
diff --git a/drivers/net/ethernet/cavium/liquidio/octeon_droq.c b/drivers/net/ethernet/cavium/liquidio/octeon_droq.c
index 286be5539cef..eef12fdd246d 100644
--- a/drivers/net/ethernet/cavium/liquidio/octeon_droq.c
+++ b/drivers/net/ethernet/cavium/liquidio/octeon_droq.c
@@ -30,11 +30,6 @@
 #include "cn23xx_pf_device.h"
 #include "cn23xx_vf_device.h"
 
-struct niclist {
-	struct list_head list;
-	void *ptr;
-};
-
 struct __dispatch {
 	struct list_head list;
 	struct octeon_recv_info *rinfo;
@@ -52,8 +47,8 @@ struct __dispatch {
  *  @return  Failure: NULL
  *
  */
-static inline void *octeon_get_dispatch_arg(struct octeon_device *octeon_dev,
-					    u16 opcode, u16 subcode)
+void *octeon_get_dispatch_arg(struct octeon_device *octeon_dev,
+			      u16 opcode, u16 subcode)
 {
 	int idx;
 	struct list_head *dispatch;
@@ -107,6 +102,7 @@ u32 octeon_droq_check_hw_for_pkts(struct octeon_droq *droq)
 
 	return last_count;
 }
+EXPORT_SYMBOL_GPL(octeon_droq_check_hw_for_pkts);
 
 static void octeon_droq_compute_max_packet_bufs(struct octeon_droq *droq)
 {
@@ -145,6 +141,8 @@ octeon_droq_destroy_ring_buffers(struct octeon_device *oct,
 
 	for (i = 0; i < droq->max_count; i++) {
 		pg_info = &droq->recv_buf_list[i].pg_info;
+		if (!pg_info)
+			continue;
 
 		if (pg_info->dma)
 			lio_unmap_ring(oct->pci_dev,
@@ -181,10 +179,7 @@ octeon_droq_setup_ring_buffers(struct octeon_device *oct,
 
 		droq->recv_buf_list[i].buffer = buf;
 		droq->recv_buf_list[i].data = get_rbd(buf);
-		droq->info_list[i].length = 0;
-
-		/* map ring buffers into memory */
-		desc_ring[i].info_ptr = lio_map_ring_info(droq, i);
+		desc_ring[i].info_ptr = 0;
 		desc_ring[i].buffer_ptr =
 			lio_map_ring(droq->recv_buf_list[i].buffer);
 	}
@@ -205,17 +200,19 @@ int octeon_delete_droq(struct octeon_device *oct, u32 q_no)
 	octeon_droq_destroy_ring_buffers(oct, droq);
 	vfree(droq->recv_buf_list);
 
-	if (droq->info_base_addr)
-		lio_free_info_buffer(oct, droq);
-
 	if (droq->desc_ring)
 		lio_dma_free(oct, (droq->max_count * OCT_DROQ_DESC_SIZE),
 			     droq->desc_ring, droq->desc_ring_dma);
 
 	memset(droq, 0, OCT_DROQ_SIZE);
+	oct->io_qmask.oq &= ~(1ULL << q_no);
+	vfree(oct->droq[q_no]);
+	oct->droq[q_no] = NULL;
+	oct->num_oqs--;
 
 	return 0;
 }
+EXPORT_SYMBOL_GPL(octeon_delete_droq);
 
 int octeon_init_droq(struct octeon_device *oct,
 		     u32 q_no,
@@ -280,22 +277,10 @@ int octeon_init_droq(struct octeon_device *oct,
 	dev_dbg(&oct->pci_dev->dev, "droq[%d]: num_desc: %d\n", q_no,
 		droq->max_count);
 
-	droq->info_list = lio_alloc_info_buffer(oct, droq);
-	if (!droq->info_list) {
-		dev_err(&oct->pci_dev->dev, "Cannot allocate memory for info list.\n");
-		lio_dma_free(oct, (droq->max_count * OCT_DROQ_DESC_SIZE),
-			     droq->desc_ring, droq->desc_ring_dma);
-		return 1;
-	}
-
-	droq->recv_buf_list = (struct octeon_recv_buffer *)
-			      vmalloc_node(droq->max_count *
-						OCT_DROQ_RECVBUF_SIZE,
-						numa_node);
+	droq->recv_buf_list = vzalloc_node(array_size(droq->max_count, OCT_DROQ_RECVBUF_SIZE),
+					   numa_node);
 	if (!droq->recv_buf_list)
-		droq->recv_buf_list = (struct octeon_recv_buffer *)
-				      vmalloc(droq->max_count *
-						OCT_DROQ_RECVBUF_SIZE);
+		droq->recv_buf_list = vzalloc(array_size(droq->max_count, OCT_DROQ_RECVBUF_SIZE));
 	if (!droq->recv_buf_list) {
 		dev_err(&oct->pci_dev->dev, "Output queue recv buf list alloc failed\n");
 		goto init_droq_fail;
@@ -310,8 +295,6 @@ int octeon_init_droq(struct octeon_device *oct,
 	dev_dbg(&oct->pci_dev->dev, "DROQ INIT: max_empty_descs: %d\n",
 		droq->max_empty_descs);
 
-	spin_lock_init(&droq->lock);
-
 	INIT_LIST_HEAD(&droq->dispatch_list);
 
 	/* For 56xx Pass1, this function won't be called, so no checks. */
@@ -342,8 +325,6 @@ init_droq_fail:
  * Returns:
  *  Success: Pointer to recv_info_t
  *  Failure: NULL.
- * Locks:
- *  The droq->lock is held when this routine is called.
  */
 static inline struct octeon_recv_info *octeon_create_recv_info(
 		struct octeon_device *octeon_dev,
@@ -357,7 +338,7 @@ static inline struct octeon_recv_info *octeon_create_recv_info(
 	u32 i, bytes_left;
 	struct octeon_skb_page_info *pg_info;
 
-	info = &droq->info_list[idx];
+	info = (struct octeon_droq_info *)droq->recv_buf_list[idx].data;
 
 	recv_info = octeon_alloc_recv_info(sizeof(struct __dispatch));
 	if (!recv_info)
@@ -425,8 +406,7 @@ octeon_droq_refill_pullup_descs(struct octeon_droq *droq,
 							      droq->max_count);
 				desc_refilled++;
 				droq->refill_count--;
-			} while (droq->recv_buf_list[droq->refill_idx].
-				 buffer);
+			} while (droq->recv_buf_list[droq->refill_idx].buffer);
 		}
 		refill_index = incr_index(refill_index, 1, droq->max_count);
 	}                       /* while */
@@ -443,8 +423,6 @@ octeon_droq_refill_pullup_descs(struct octeon_droq *droq,
  *  up buffers (that were not dispatched) to form a contiguous ring.
  * Returns:
  *  No of descriptors refilled.
- * Locks:
- *  This routine is called with droq->lock held.
  */
 static u32
 octeon_droq_refill(struct octeon_device *octeon_dev, struct octeon_droq *droq)
@@ -459,8 +437,7 @@ octeon_droq_refill(struct octeon_device *octeon_dev, struct octeon_droq *droq)
 
 	while (droq->refill_count && (desc_refilled < droq->max_count)) {
 		/* If a valid buffer exists (happens if there is no dispatch),
-		 * reuse
-		 * the buffer, else allocate.
+		 * reuse the buffer, else allocate.
 		 */
 		if (!droq->recv_buf_list[droq->refill_idx].buffer) {
 			pg_info =
@@ -490,10 +467,8 @@ octeon_droq_refill(struct octeon_device *octeon_dev, struct octeon_droq *droq)
 		droq->recv_buf_list[droq->refill_idx].data = data;
 
 		desc_ring[droq->refill_idx].buffer_ptr =
-			lio_map_ring(droq->recv_buf_list[droq->
-				     refill_idx].buffer);
-		/* Reset any previous values in the length field. */
-		droq->info_list[droq->refill_idx].length = 0;
+			lio_map_ring(droq->recv_buf_list[
+				     droq->refill_idx].buffer);
 
 		droq->refill_idx = incr_index(droq->refill_idx, 1,
 					      droq->max_count);
@@ -515,38 +490,35 @@ octeon_droq_refill(struct octeon_device *octeon_dev, struct octeon_droq *droq)
 
 /** check if we can allocate packets to get out of oom.
  *  @param  droq - Droq being checked.
- *  @return does not return anything
+ *  @return 1 if fails to refill minimum
  */
-void octeon_droq_check_oom(struct octeon_droq *droq)
+int octeon_retry_droq_refill(struct octeon_droq *droq)
 {
-	int desc_refilled;
 	struct octeon_device *oct = droq->oct_dev;
+	int desc_refilled, reschedule = 1;
+	u32 pkts_credit;
+
+	pkts_credit = readl(droq->pkts_credit_reg);
+	desc_refilled = octeon_droq_refill(oct, droq);
+	if (desc_refilled) {
+		/* Flush the droq descriptor data to memory to be sure
+		 * that when we update the credits the data in memory
+		 * is accurate.
+		 */
+		wmb();
+		writel(desc_refilled, droq->pkts_credit_reg);
 
-	if (readl(droq->pkts_credit_reg) <= CN23XX_SLI_DEF_BP) {
-		spin_lock_bh(&droq->lock);
-		desc_refilled = octeon_droq_refill(oct, droq);
-		if (desc_refilled) {
-			/* Flush the droq descriptor data to memory to be sure
-			 * that when we update the credits the data in memory
-			 * is accurate.
-			 */
-			wmb();
-			writel(desc_refilled, droq->pkts_credit_reg);
-			/* make sure mmio write completes */
-			mmiowb();
-		}
-		spin_unlock_bh(&droq->lock);
+		if (pkts_credit + desc_refilled >= CN23XX_SLI_DEF_BP)
+			reschedule = 0;
 	}
+
+	return reschedule;
 }
 
 static inline u32
 octeon_droq_get_bufcount(u32 buf_size, u32 total_len)
 {
-	u32 buf_cnt = 0;
-
-	while (total_len > (buf_size * buf_cnt))
-		buf_cnt++;
-	return buf_cnt;
+	return DIV_ROUND_UP(total_len, buf_size);
 }
 
 static int
@@ -594,11 +566,12 @@ static inline void octeon_droq_drop_packets(struct octeon_device *oct,
 	struct octeon_droq_info *info;
 
 	for (i = 0; i < cnt; i++) {
-		info = &droq->info_list[droq->read_idx];
+		info = (struct octeon_droq_info *)
+			droq->recv_buf_list[droq->read_idx].data;
 		octeon_swap_8B_data((u64 *)info, 2);
 
 		if (info->length) {
-			info->length -= OCT_RH_SIZE;
+			info->length += OCTNET_FRM_LENGTH_SIZE;
 			droq->stats.bytes_received += info->length;
 			buf_cnt = octeon_droq_get_bufcount(droq->buffer_size,
 							   (u32)info->length);
@@ -618,9 +591,9 @@ octeon_droq_fast_process_packets(struct octeon_device *oct,
 				 struct octeon_droq *droq,
 				 u32 pkts_to_process)
 {
+	u32 pkt, total_len = 0, pkt_count, retval;
 	struct octeon_droq_info *info;
 	union octeon_rh *rh;
-	u32 pkt, total_len = 0, pkt_count;
 
 	pkt_count = pkts_to_process;
 
@@ -630,7 +603,8 @@ octeon_droq_fast_process_packets(struct octeon_device *oct,
 		struct octeon_skb_page_info *pg_info;
 		void *buf;
 
-		info = &droq->info_list[droq->read_idx];
+		info = (struct octeon_droq_info *)
+			droq->recv_buf_list[droq->read_idx].data;
 		octeon_swap_8B_data((u64 *)info, 2);
 
 		if (!info->length) {
@@ -644,9 +618,10 @@ octeon_droq_fast_process_packets(struct octeon_device *oct,
 		}
 
 		/* Len of resp hdr in included in the received data len. */
-		info->length -= OCT_RH_SIZE;
 		rh = &info->rh;
 
+		info->length += OCTNET_FRM_LENGTH_SIZE;
+		rh->r_dh.len += (ROUNDUP8(OCT_DROQ_INFO_SIZE) / sizeof(u64));
 		total_len += (u32)info->length;
 		if (opcode_slow_path(rh)) {
 			u32 buf_cnt;
@@ -690,8 +665,8 @@ octeon_droq_fast_process_packets(struct octeon_device *oct,
 									nicbuf,
 									cpy_len,
 									idx);
-						buf = droq->recv_buf_list[idx].
-							buffer;
+						buf = droq->recv_buf_list[
+							idx].buffer;
 						recv_buffer_fast_free(buf);
 						droq->recv_buf_list[idx].buffer
 							= NULL;
@@ -722,30 +697,41 @@ octeon_droq_fast_process_packets(struct octeon_device *oct,
 		if (droq->refill_count >= droq->refill_threshold) {
 			int desc_refilled = octeon_droq_refill(oct, droq);
 
-			/* Flush the droq descriptor data to memory to be sure
-			 * that when we update the credits the data in memory
-			 * is accurate.
-			 */
-			wmb();
-			writel((desc_refilled), droq->pkts_credit_reg);
-			/* make sure mmio write completes */
-			mmiowb();
+			if (desc_refilled) {
+				/* Flush the droq descriptor data to memory to
+				 * be sure that when we update the credits the
+				 * data in memory is accurate.
+				 */
+				wmb();
+				writel(desc_refilled, droq->pkts_credit_reg);
+			}
 		}
-
 	}                       /* for (each packet)... */
 
 	/* Increment refill_count by the number of buffers processed. */
 	droq->stats.pkts_received += pkt;
 	droq->stats.bytes_received += total_len;
 
+	retval = pkt;
 	if ((droq->ops.drop_on_max) && (pkts_to_process - pkt)) {
 		octeon_droq_drop_packets(oct, droq, (pkts_to_process - pkt));
 
 		droq->stats.dropped_toomany += (pkts_to_process - pkt);
-		return pkts_to_process;
+		retval = pkts_to_process;
 	}
 
-	return pkt;
+	atomic_sub(retval, &droq->pkts_pending);
+
+	if (droq->refill_count >= droq->refill_threshold &&
+	    readl(droq->pkts_credit_reg) < CN23XX_SLI_DEF_BP) {
+		octeon_droq_check_hw_for_pkts(droq);
+
+		/* Make sure there are no pkts_pending */
+		if (!atomic_read(&droq->pkts_pending))
+			octeon_schedule_rxq_oom_work(oct, droq);
+	}
+
+	return retval;
 }
 
 int
@@ -753,29 +739,19 @@ octeon_droq_process_packets(struct octeon_device *oct,
 			    struct octeon_droq *droq,
 			    u32 budget)
 {
-	u32 pkt_count = 0, pkts_processed = 0;
+	u32 pkt_count = 0;
 	struct list_head *tmp, *tmp2;
 
-	/* Grab the droq lock */
-	spin_lock(&droq->lock);
-
 	octeon_droq_check_hw_for_pkts(droq);
 	pkt_count = atomic_read(&droq->pkts_pending);
 
-	if (!pkt_count) {
-		spin_unlock(&droq->lock);
+	if (!pkt_count)
 		return 0;
-	}
 
 	if (pkt_count > budget)
 		pkt_count = budget;
 
-	pkts_processed = octeon_droq_fast_process_packets(oct, droq, pkt_count);
-
-	atomic_sub(pkts_processed, &droq->pkts_pending);
-
-	/* Release the spin lock */
-	spin_unlock(&droq->lock);
+	octeon_droq_fast_process_packets(oct, droq, pkt_count);
 
 	list_for_each_safe(tmp, tmp2, &droq->dispatch_list) {
 		struct __dispatch *rdisp = (struct __dispatch *)tmp;
@@ -794,13 +770,14 @@ octeon_droq_process_packets(struct octeon_device *oct,
 
 	return 0;
 }
+EXPORT_SYMBOL_GPL(octeon_droq_process_packets);
 
-/**
+/*
  * Utility function to poll for packets. check_hw_for_packets must be
  * called before calling this routine.
  */
 
-static int
+int
 octeon_droq_process_poll_pkts(struct octeon_device *oct,
 			      struct octeon_droq *droq, u32 budget)
 {
@@ -811,8 +788,6 @@ octeon_droq_process_poll_pkts(struct octeon_device *oct,
 	if (budget > droq->max_count)
 		budget = droq->max_count;
 
-	spin_lock(&droq->lock);
-
 	while (total_pkts_processed < budget) {
 		octeon_droq_check_hw_for_pkts(droq);
 
@@ -826,13 +801,9 @@ octeon_droq_process_poll_pkts(struct octeon_device *oct,
 			octeon_droq_fast_process_packets(oct, droq,
 							 pkts_available);
 
-		atomic_sub(pkts_processed, &droq->pkts_pending);
-
 		total_pkts_processed += pkts_processed;
 	}
 
-	spin_unlock(&droq->lock);
-
 	list_for_each_safe(tmp, tmp2, &droq->dispatch_list) {
 		struct __dispatch *rdisp = (struct __dispatch *)tmp;
 
@@ -847,79 +818,53 @@ octeon_droq_process_poll_pkts(struct octeon_device *oct,
 	return total_pkts_processed;
 }
 
+/* Enable Pkt Interrupt */
 int
-octeon_process_droq_poll_cmd(struct octeon_device *oct, u32 q_no, int cmd,
-			     u32 arg)
+octeon_enable_irq(struct octeon_device *oct, u32 q_no)
 {
-	struct octeon_droq *droq;
-
-	droq = oct->droq[q_no];
+	switch (oct->chip_id) {
+	case OCTEON_CN66XX:
+	case OCTEON_CN68XX: {
+		struct octeon_cn6xxx *cn6xxx =
+			(struct octeon_cn6xxx *)oct->chip;
+		unsigned long flags;
+		u32 value;
 
-	if (cmd == POLL_EVENT_PROCESS_PKTS)
-		return octeon_droq_process_poll_pkts(oct, droq, arg);
+		spin_lock_irqsave
+			(&cn6xxx->lock_for_droq_int_enb_reg, flags);
+		value = octeon_read_csr(oct, CN6XXX_SLI_PKT_TIME_INT_ENB);
+		value |= (1 << q_no);
+		octeon_write_csr(oct, CN6XXX_SLI_PKT_TIME_INT_ENB, value);
+		value = octeon_read_csr(oct, CN6XXX_SLI_PKT_CNT_INT_ENB);
+		value |= (1 << q_no);
+		octeon_write_csr(oct, CN6XXX_SLI_PKT_CNT_INT_ENB, value);
 
-	if (cmd == POLL_EVENT_PENDING_PKTS) {
-		u32 pkt_cnt = atomic_read(&droq->pkts_pending);
+		/* don't bother flushing the enables */
 
-		return  octeon_droq_process_packets(oct, droq, pkt_cnt);
+		spin_unlock_irqrestore
+			(&cn6xxx->lock_for_droq_int_enb_reg, flags);
 	}
-
-	if (cmd == POLL_EVENT_ENABLE_INTR) {
-		u32 value;
-		unsigned long flags;
-
-		/* Enable Pkt Interrupt */
-		switch (oct->chip_id) {
-		case OCTEON_CN66XX:
-		case OCTEON_CN68XX: {
-			struct octeon_cn6xxx *cn6xxx =
-				(struct octeon_cn6xxx *)oct->chip;
-			spin_lock_irqsave
-				(&cn6xxx->lock_for_droq_int_enb_reg, flags);
-			value =
-				octeon_read_csr(oct,
-						CN6XXX_SLI_PKT_TIME_INT_ENB);
-			value |= (1 << q_no);
-			octeon_write_csr(oct,
-					 CN6XXX_SLI_PKT_TIME_INT_ENB,
-					 value);
-			value =
-				octeon_read_csr(oct,
-						CN6XXX_SLI_PKT_CNT_INT_ENB);
-			value |= (1 << q_no);
-			octeon_write_csr(oct,
-					 CN6XXX_SLI_PKT_CNT_INT_ENB,
-					 value);
-
-			/* don't bother flushing the enables */
-
-			spin_unlock_irqrestore
-				(&cn6xxx->lock_for_droq_int_enb_reg, flags);
-			return 0;
-		}
 		break;
-		case OCTEON_CN23XX_PF_VID: {
-			lio_enable_irq(oct->droq[q_no], oct->instr_queue[q_no]);
-		}
+	case OCTEON_CN23XX_PF_VID:
+		lio_enable_irq(oct->droq[q_no], oct->instr_queue[q_no]);
 		break;
 
-		case OCTEON_CN23XX_VF_VID:
-			lio_enable_irq(oct->droq[q_no], oct->instr_queue[q_no]);
+	case OCTEON_CN23XX_VF_VID:
+		lio_enable_irq(oct->droq[q_no], oct->instr_queue[q_no]);
 		break;
-		}
-		return 0;
+	default:
+		dev_err(&oct->pci_dev->dev, "%s Unknown Chip\n", __func__);
+		return 1;
 	}
 
-	dev_err(&oct->pci_dev->dev, "%s Unknown command: %d\n", __func__, cmd);
-	return -EINVAL;
+	return 0;
 }
 
 int octeon_register_droq_ops(struct octeon_device *oct, u32 q_no,
 			     struct octeon_droq_ops *ops)
 {
-	struct octeon_droq *droq;
-	unsigned long flags;
 	struct octeon_config *oct_cfg = NULL;
+	struct octeon_droq *droq;
 
 	oct_cfg = octeon_get_conf(oct);
 
@@ -939,21 +884,15 @@ int octeon_register_droq_ops(struct octeon_device *oct, u32 q_no,
 	}
 
 	droq = oct->droq[q_no];
-
-	spin_lock_irqsave(&droq->lock, flags);
-
 	memcpy(&droq->ops, ops, sizeof(struct octeon_droq_ops));
 
-	spin_unlock_irqrestore(&droq->lock, flags);
-
 	return 0;
 }
 
 int octeon_unregister_droq_ops(struct octeon_device *oct, u32 q_no)
 {
-	unsigned long flags;
-	struct octeon_droq *droq;
 	struct octeon_config *oct_cfg = NULL;
+	struct octeon_droq *droq;
 
 	oct_cfg = octeon_get_conf(oct);
 
@@ -974,16 +913,13 @@ int octeon_unregister_droq_ops(struct octeon_device *oct, u32 q_no)
 		return 0;
 	}
 
-	spin_lock_irqsave(&droq->lock, flags);
-
 	droq->ops.fptr = NULL;
 	droq->ops.farg = NULL;
 	droq->ops.drop_on_max = 0;
 
-	spin_unlock_irqrestore(&droq->lock, flags);
-
 	return 0;
 }
+EXPORT_SYMBOL_GPL(octeon_unregister_droq_ops);
 
 int octeon_create_droq(struct octeon_device *oct,
 		       u32 q_no, u32 num_descs,
diff --git a/drivers/net/ethernet/cavium/liquidio/octeon_droq.h b/drivers/net/ethernet/cavium/liquidio/octeon_droq.h
index 9781577115e7..232ae72c0e37 100644
--- a/drivers/net/ethernet/cavium/liquidio/octeon_droq.h
+++ b/drivers/net/ethernet/cavium/liquidio/octeon_droq.h
@@ -51,11 +51,11 @@ struct octeon_droq_desc {
  *  about the packet.
  */
 struct octeon_droq_info {
-	/** The Output Receive Header. */
-	union octeon_rh rh;
-
 	/** The Length of the packet. */
 	u64 length;
+
+	/** The Output Receive Header. */
+	union octeon_rh rh;
 };
 
 #define OCT_DROQ_INFO_SIZE   (sizeof(struct octeon_droq_info))
@@ -123,11 +123,6 @@ struct oct_droq_stats {
 
 };
 
-#define POLL_EVENT_INTR_ARRIVED  1
-#define POLL_EVENT_PROCESS_PKTS  2
-#define POLL_EVENT_PENDING_PKTS  3
-#define POLL_EVENT_ENABLE_INTR   4
-
 /* The maximum number of buffers that can be dispatched from the
  * output/dma queue. Set to 64 assuming 1K buffers in DROQ and the fact that
  * max packet size from DROQ is 64K.
@@ -250,9 +245,6 @@ struct octeon_droq_ops {
  *  Octeon DROQ.
  */
 struct octeon_droq {
-	/** A spinlock to protect access to this ring. */
-	spinlock_t lock;
-
 	u32 q_no;
 
 	u32 pkt_count;
@@ -294,9 +286,6 @@ struct octeon_droq {
 	 */
 	u32 max_empty_descs;
 
-	/** The 8B aligned info ptrs begin from this address. */
-	struct octeon_droq_info *info_list;
-
 	/** The receive buffer list. This list has the virtual addresses of the
 	 * buffers.
 	 */
@@ -324,15 +313,6 @@ struct octeon_droq {
 	/** DMA mapped address of the DROQ descriptor ring. */
 	size_t desc_ring_dma;
 
-	/** Info ptr list are allocated at this virtual address. */
-	void *info_base_addr;
-
-	/** DMA mapped address of the info list */
-	dma_addr_t info_list_dma;
-
-	/** Allocated size of info list. */
-	u32 info_alloc_size;
-
 	/** application context */
 	void *app_ctx;
 
@@ -340,7 +320,7 @@ struct octeon_droq {
 
 	u32 cpu_id;
 
-	struct call_single_data csd;
+	call_single_data_t csd;
 };
 
 #define OCT_DROQ_SIZE   (sizeof(struct octeon_droq))
@@ -412,7 +392,8 @@ int octeon_register_dispatch_fn(struct octeon_device *oct,
 				u16 subcode,
 				octeon_dispatch_fn_t fn, void *fn_arg);
 
-void octeon_droq_print_stats(void);
+void *octeon_get_dispatch_arg(struct octeon_device *oct,
+			      u16 opcode, u16 subcode);
 
 u32 octeon_droq_check_hw_for_pkts(struct octeon_droq *droq);
 
@@ -423,9 +404,11 @@ int octeon_droq_process_packets(struct octeon_device *oct,
 				struct octeon_droq *droq,
 				u32 budget);
 
-int octeon_process_droq_poll_cmd(struct octeon_device *oct, u32 q_no,
-				 int cmd, u32 arg);
+int octeon_droq_process_poll_pkts(struct octeon_device *oct,
+				  struct octeon_droq *droq, u32 budget);
+
+int octeon_enable_irq(struct octeon_device *oct, u32 q_no);
 
-void octeon_droq_check_oom(struct octeon_droq *droq);
+int octeon_retry_droq_refill(struct octeon_droq *droq);
 
 #endif	/*__OCTEON_DROQ_H__ */
diff --git a/drivers/net/ethernet/cavium/liquidio/octeon_iq.h b/drivers/net/ethernet/cavium/liquidio/octeon_iq.h
index 5c3c8da976f7..a04f36a0e1a0 100644
--- a/drivers/net/ethernet/cavium/liquidio/octeon_iq.h
+++ b/drivers/net/ethernet/cavium/liquidio/octeon_iq.h
@@ -62,8 +62,8 @@ struct oct_iq_stats {
 	u64 tx_tot_bytes;/**< Total count of bytes sento to network. */
 	u64 tx_gso;  /* count of tso */
 	u64 tx_vxlan; /* tunnel */
-	u64 tx_dmamap_fail;
-	u64 tx_restart;
+	u64 tx_dmamap_fail; /* Number of times dma mapping failed */
+	u64 tx_restart; /* Number of times this queue restarted */
 };
 
 #define OCT_IQ_STATS_SIZE   (sizeof(struct oct_iq_stats))
@@ -82,8 +82,20 @@ struct octeon_instr_queue {
 	/** A spinlock to protect while posting on the ring.  */
 	spinlock_t post_lock;
 
+	/** This flag indicates if the queue can be used for soft commands.
+	 *  If this flag is set, post_lock must be acquired before posting
+	 *  a command to the queue.
+	 *  If this flag is clear, post_lock is invalid for the queue.
+	 *  All control commands (soft commands) will go through only Queue 0
+	 *  (control and data queue). So only queue-0 needs post_lock,
+	 *  other queues are only data queues and does not need post_lock
+	 */
+	bool allow_soft_cmds;
+
 	u32 pkt_in_done;
 
+	u32 pkts_processed;
+
 	/** A spinlock to protect access to the input ring.*/
 	spinlock_t iq_flush_running_lock;
 
@@ -280,13 +292,19 @@ struct octeon_soft_command {
 	u32  ctxsize;
 
 	/** Time out and callback */
-	size_t wait_time;
-	size_t timeout;
+	size_t expiry_time;
 	u32 iq_no;
 	void (*callback)(struct octeon_device *, u32, void *);
 	void *callback_arg;
+
+	int caller_is_done;
+	u32 sc_status;
+	struct completion complete;
 };
 
+/* max timeout (in milli sec) for soft request */
+#define LIO_SC_MAX_TMO_MS       60000
+
 /** Maximum number of buffers to allocate into soft command buffer pool
  */
 #define  MAX_SOFT_COMMAND_BUFFERS	256
@@ -307,6 +325,8 @@ struct octeon_sc_buffer_pool {
 		(((octeon_dev_ptr)->instr_queue[iq_no]->stats.field) += count)
 
 int octeon_setup_sc_buffer_pool(struct octeon_device *oct);
+int octeon_free_sc_done_list(struct octeon_device *oct);
+int octeon_free_sc_zombie_list(struct octeon_device *oct);
 int octeon_free_sc_buffer_pool(struct octeon_device *oct);
 struct octeon_soft_command *
 	octeon_alloc_soft_command(struct octeon_device *oct,
@@ -343,6 +363,9 @@ int octeon_delete_instr_queue(struct octeon_device *octeon_dev, u32 iq_no);
 
 int lio_wait_for_instr_fetch(struct octeon_device *oct);
 
+void
+octeon_ring_doorbell_locked(struct octeon_device *oct, u32 iq_no);
+
 int
 octeon_register_reqtype_free_fn(struct octeon_device *oct, int reqtype,
 				void (*fn)(void *));
diff --git a/drivers/net/ethernet/cavium/liquidio/octeon_mailbox.c b/drivers/net/ethernet/cavium/liquidio/octeon_mailbox.c
index 57af7df74ced..ad685f5d0a13 100644
--- a/drivers/net/ethernet/cavium/liquidio/octeon_mailbox.c
+++ b/drivers/net/ethernet/cavium/liquidio/octeon_mailbox.c
@@ -24,10 +24,11 @@
 #include "octeon_device.h"
 #include "octeon_main.h"
 #include "octeon_mailbox.h"
+#include "cn23xx_pf_device.h"
 
 /**
  * octeon_mbox_read:
- * @oct: Pointer mailbox
+ * @mbox: Pointer mailbox
  *
  * Reads the 8-bytes of data from the mbox register
  * Writes back the acknowldgement inidcating completion of read
@@ -87,7 +88,7 @@ int octeon_mbox_read(struct octeon_mbox *mbox)
 	}
 
 	if (mbox->state & OCTEON_MBOX_STATE_REQUEST_RECEIVING) {
-		if (mbox->mbox_req.recv_len < msg.s.len) {
+		if (mbox->mbox_req.recv_len < mbox->mbox_req.msg.s.len) {
 			ret = 0;
 		} else {
 			mbox->state &= ~OCTEON_MBOX_STATE_REQUEST_RECEIVING;
@@ -96,7 +97,8 @@ int octeon_mbox_read(struct octeon_mbox *mbox)
 		}
 	} else {
 		if (mbox->state & OCTEON_MBOX_STATE_RESPONSE_RECEIVING) {
-			if (mbox->mbox_resp.recv_len < msg.s.len) {
+			if (mbox->mbox_resp.recv_len <
+			    mbox->mbox_resp.msg.s.len) {
 				ret = 0;
 			} else {
 				mbox->state &=
@@ -204,6 +206,26 @@ int octeon_mbox_write(struct octeon_device *oct,
 	return ret;
 }
 
+static void get_vf_stats(struct octeon_device *oct,
+			 struct oct_vf_stats *stats)
+{
+	int i;
+
+	for (i = 0; i < oct->num_iqs; i++) {
+		if (!oct->instr_queue[i])
+			continue;
+		stats->tx_packets += oct->instr_queue[i]->stats.tx_done;
+		stats->tx_bytes += oct->instr_queue[i]->stats.tx_tot_bytes;
+	}
+
+	for (i = 0; i < oct->num_oqs; i++) {
+		if (!oct->droq[i])
+			continue;
+		stats->rx_packets += oct->droq[i]->stats.rx_pkts_received;
+		stats->rx_bytes += oct->droq[i]->stats.rx_bytes_received;
+	}
+}
+
 /**
  * octeon_mbox_process_cmd:
  * @mbox: Pointer mailbox
@@ -238,9 +260,7 @@ static int octeon_mbox_process_cmd(struct octeon_mbox *mbox,
 		dev_info(&oct->pci_dev->dev,
 			 "got a request for FLR from VF that owns DPI ring %u\n",
 			 mbox->q_no);
-		pcie_capability_set_word(
-			oct->sriov_info.dpiring_to_vfpcidev_lut[mbox->q_no],
-			PCI_EXP_DEVCTL, PCI_EXP_DEVCTL_BCR_FLR);
+		pcie_flr(oct->sriov_info.dpiring_to_vfpcidev_lut[mbox->q_no]);
 		break;
 
 	case OCTEON_PF_CHANGED_VF_MACADDR:
@@ -249,6 +269,15 @@ static int octeon_mbox_process_cmd(struct octeon_mbox *mbox,
 						     mbox_cmd->msg.s.params);
 		break;
 
+	case OCTEON_GET_VF_STATS:
+		dev_dbg(&oct->pci_dev->dev, "Got VF stats request. Sending data back\n");
+		mbox_cmd->msg.s.type = OCTEON_MBOX_RESPONSE;
+		mbox_cmd->msg.s.resp_needed = 1;
+		mbox_cmd->msg.s.len = 1 +
+			sizeof(struct oct_vf_stats) / sizeof(u64);
+		get_vf_stats(oct, (struct oct_vf_stats *)mbox_cmd->data);
+		octeon_mbox_write(oct, mbox_cmd);
+		break;
 	default:
 		break;
 	}
@@ -256,7 +285,8 @@ static int octeon_mbox_process_cmd(struct octeon_mbox *mbox,
 }
 
 /**
- *octeon_mbox_process_message:
+ * octeon_mbox_process_message
+ * @mbox: mailbox
  *
  * Process the received mbox message.
  */
@@ -321,3 +351,25 @@ int octeon_mbox_process_message(struct octeon_mbox *mbox)
 
 	return 0;
 }
+
+int octeon_mbox_cancel(struct octeon_device *oct, int q_no)
+{
+	struct octeon_mbox *mbox = oct->mbox[q_no];
+	struct octeon_mbox_cmd *mbox_cmd;
+	unsigned long flags = 0;
+
+	spin_lock_irqsave(&mbox->lock, flags);
+	mbox_cmd = &mbox->mbox_resp;
+
+	if (!(mbox->state & OCTEON_MBOX_STATE_RESPONSE_PENDING)) {
+		spin_unlock_irqrestore(&mbox->lock, flags);
+		return 1;
+	}
+
+	mbox->state = OCTEON_MBOX_STATE_IDLE;
+	memset(mbox_cmd, 0, sizeof(*mbox_cmd));
+	writeq(OCTEON_PFVFSIG, mbox->mbox_read_reg);
+	spin_unlock_irqrestore(&mbox->lock, flags);
+
+	return 0;
+}
diff --git a/drivers/net/ethernet/cavium/liquidio/octeon_mailbox.h b/drivers/net/ethernet/cavium/liquidio/octeon_mailbox.h
index 1def22afeff1..9ac85d22c615 100644
--- a/drivers/net/ethernet/cavium/liquidio/octeon_mailbox.h
+++ b/drivers/net/ethernet/cavium/liquidio/octeon_mailbox.h
@@ -25,6 +25,7 @@
 #define OCTEON_VF_ACTIVE		0x1
 #define OCTEON_VF_FLR_REQUEST		0x2
 #define OCTEON_PF_CHANGED_VF_MACADDR	0x4
+#define OCTEON_GET_VF_STATS		0x8
 
 /*Macro for Read acknowldgement*/
 #define OCTEON_PFVFACK			0xffffffffffffffffULL
@@ -56,7 +57,10 @@ union octeon_mbox_message {
 	} s;
 };
 
-typedef void (*octeon_mbox_callback_t)(void *, void *, void *);
+struct octeon_mbox_cmd;
+
+typedef void (*octeon_mbox_callback_t)(struct octeon_device *,
+				       struct octeon_mbox_cmd *, void *);
 
 struct octeon_mbox_cmd {
 	union octeon_mbox_message msg;
@@ -107,9 +111,15 @@ struct octeon_mbox {
 
 };
 
+struct oct_vf_stats_ctx {
+	atomic_t status;
+	struct oct_vf_stats *stats;
+};
+
 int octeon_mbox_read(struct octeon_mbox *mbox);
 int octeon_mbox_write(struct octeon_device *oct,
 		      struct octeon_mbox_cmd *mbox_cmd);
 int octeon_mbox_process_message(struct octeon_mbox *mbox);
+int octeon_mbox_cancel(struct octeon_device *oct, int q_no);
 
 #endif
diff --git a/drivers/net/ethernet/cavium/liquidio/octeon_main.h b/drivers/net/ethernet/cavium/liquidio/octeon_main.h
index bed9ef17bc26..953edf0c7096 100644
--- a/drivers/net/ethernet/cavium/liquidio/octeon_main.h
+++ b/drivers/net/ethernet/cavium/liquidio/octeon_main.h
@@ -35,6 +35,13 @@
 
 #define DRV_NAME "LiquidIO"
 
+struct octeon_device_priv {
+	/** Tasklet structures for this device. */
+	struct tasklet_struct droq_tasklet;
+	unsigned long napi_mask;
+	struct octeon_device *dev;
+};
+
 /** This structure is used by NIC driver to store information required
  * to free the sk_buff when the packet has been fetched by Octeon.
  * Bytes offset below assume worst-case of a 64-bit system.
@@ -57,13 +64,17 @@ struct octnet_buf_free_info {
 };
 
 /* BQL-related functions */
-void octeon_report_sent_bytes_to_bql(void *buf, int reqtype);
+int octeon_report_sent_bytes_to_bql(void *buf, int reqtype);
 void octeon_update_tx_completion_counters(void *buf, int reqtype,
 					  unsigned int *pkts_compl,
 					  unsigned int *bytes_compl);
 void octeon_report_tx_completion_to_bql(void *txq, unsigned int pkts_compl,
 					unsigned int bytes_compl);
 void octeon_pf_changed_vf_macaddr(struct octeon_device *oct, u8 *mac);
+
+void octeon_schedule_rxq_oom_work(struct octeon_device *oct,
+				  struct octeon_droq *droq);
+
 /** Swap 8B blocks */
 static inline void octeon_swap_8B_data(u64 *data, u32 blocks)
 {
@@ -140,46 +151,70 @@ err_release_region:
 	return 1;
 }
 
+/* input parameter:
+ * sc: pointer to a soft request
+ * timeout: milli sec which an application wants to wait for the
+	    response of the request.
+ *          0: the request will wait until its response gets back
+ *	       from the firmware within LIO_SC_MAX_TMO_MS milli sec.
+ *	       If the response does not return within
+ *	       LIO_SC_MAX_TMO_MS milli sec, lio_process_ordered_list()
+ *	       will move the request to zombie response list.
+ *
+ * return value:
+ * 0: got the response from firmware for the sc request.
+ * errno -EINTR: user abort the command.
+ * errno -ETIME: user spefified timeout value has been expired.
+ * errno -EBUSY: the response of the request does not return in
+ *               resonable time (LIO_SC_MAX_TMO_MS).
+ *               the sc wll be move to zombie response list by
+ *               lio_process_ordered_list()
+ *
+ * A request with non-zero return value, the sc->caller_is_done
+ *  will be marked 1.
+ * When getting a request with zero return value, the requestor
+ *  should mark sc->caller_is_done with 1 after examing the
+ *  response of sc.
+ * lio_process_ordered_list() will free the soft command on behalf
+ * of the soft command requestor.
+ * This is to fix the possible race condition of both timeout process
+ * and lio_process_ordered_list()/callback function to free a
+ * sc strucutre.
+ */
 static inline int
-sleep_cond(wait_queue_head_t *wait_queue, int *condition)
+wait_for_sc_completion_timeout(struct octeon_device *oct_dev,
+			       struct octeon_soft_command *sc,
+			       unsigned long timeout)
 {
 	int errno = 0;
-	wait_queue_t we;
-
-	init_waitqueue_entry(&we, current);
-	add_wait_queue(wait_queue, &we);
-	while (!(READ_ONCE(*condition))) {
-		set_current_state(TASK_INTERRUPTIBLE);
-		if (signal_pending(current)) {
-			errno = -EINTR;
-			goto out;
-		}
-		schedule();
+	long timeout_jiff;
+
+	if (timeout)
+		timeout_jiff = msecs_to_jiffies(timeout);
+	else
+		timeout_jiff = MAX_SCHEDULE_TIMEOUT;
+
+	timeout_jiff =
+		wait_for_completion_interruptible_timeout(&sc->complete,
+							  timeout_jiff);
+	if (timeout_jiff == 0) {
+		dev_err(&oct_dev->pci_dev->dev, "%s: sc is timeout\n",
+			__func__);
+		WRITE_ONCE(sc->caller_is_done, true);
+		errno = -ETIME;
+	} else if (timeout_jiff == -ERESTARTSYS) {
+		dev_err(&oct_dev->pci_dev->dev, "%s: sc is interrupted\n",
+			__func__);
+		WRITE_ONCE(sc->caller_is_done, true);
+		errno = -EINTR;
+	} else  if (sc->sc_status == OCTEON_REQUEST_TIMEOUT) {
+		dev_err(&oct_dev->pci_dev->dev, "%s: sc has fatal timeout\n",
+			__func__);
+		WRITE_ONCE(sc->caller_is_done, true);
+		errno = -EBUSY;
 	}
-out:
-	set_current_state(TASK_RUNNING);
-	remove_wait_queue(wait_queue, &we);
-	return errno;
-}
 
-/* Gives up the CPU for a timeout period.
- * Check that the condition is not true before we go to sleep for a
- * timeout period.
- */
-static inline void
-sleep_timeout_cond(wait_queue_head_t *wait_queue,
-		   int *condition,
-		   int timeout)
-{
-	wait_queue_t we;
-
-	init_waitqueue_entry(&we, current);
-	add_wait_queue(wait_queue, &we);
-	set_current_state(TASK_INTERRUPTIBLE);
-	if (!(*condition))
-		schedule_timeout(timeout);
-	set_current_state(TASK_RUNNING);
-	remove_wait_queue(wait_queue, &we);
+	return errno;
 }
 
 #ifndef ROUNDUP4
diff --git a/drivers/net/ethernet/cavium/liquidio/octeon_mem_ops.c b/drivers/net/ethernet/cavium/liquidio/octeon_mem_ops.c
index 5cd96e7d426c..d70132437af3 100644
--- a/drivers/net/ethernet/cavium/liquidio/octeon_mem_ops.c
+++ b/drivers/net/ethernet/cavium/liquidio/octeon_mem_ops.c
@@ -22,6 +22,7 @@
 #include "octeon_iq.h"
 #include "response_manager.h"
 #include "octeon_device.h"
+#include "octeon_mem_ops.h"
 
 #define MEMOPS_IDX   BAR1_INDEX_DYNAMIC_MAP
 
@@ -163,15 +164,17 @@ octeon_pci_read_core_mem(struct octeon_device *oct,
 {
 	__octeon_pci_rw_core_mem(oct, coreaddr, buf, len, 1);
 }
+EXPORT_SYMBOL_GPL(octeon_pci_read_core_mem);
 
 void
 octeon_pci_write_core_mem(struct octeon_device *oct,
 			  u64 coreaddr,
-			  u8 *buf,
+			  const u8 *buf,
 			  u32 len)
 {
-	__octeon_pci_rw_core_mem(oct, coreaddr, buf, len, 0);
+	__octeon_pci_rw_core_mem(oct, coreaddr, (u8 *)buf, len, 0);
 }
+EXPORT_SYMBOL_GPL(octeon_pci_write_core_mem);
 
 u64 octeon_read_device_mem64(struct octeon_device *oct, u64 coreaddr)
 {
@@ -181,6 +184,7 @@ u64 octeon_read_device_mem64(struct octeon_device *oct, u64 coreaddr)
 
 	return be64_to_cpu(ret);
 }
+EXPORT_SYMBOL_GPL(octeon_read_device_mem64);
 
 u32 octeon_read_device_mem32(struct octeon_device *oct, u64 coreaddr)
 {
@@ -190,6 +194,7 @@ u32 octeon_read_device_mem32(struct octeon_device *oct, u64 coreaddr)
 
 	return be32_to_cpu(ret);
 }
+EXPORT_SYMBOL_GPL(octeon_read_device_mem32);
 
 void octeon_write_device_mem32(struct octeon_device *oct, u64 coreaddr,
 			       u32 val)
@@ -198,3 +203,4 @@ void octeon_write_device_mem32(struct octeon_device *oct, u64 coreaddr,
 
 	__octeon_pci_rw_core_mem(oct, coreaddr, (u8 *)&t, 4, 0);
 }
+EXPORT_SYMBOL_GPL(octeon_write_device_mem32);
diff --git a/drivers/net/ethernet/cavium/liquidio/octeon_mem_ops.h b/drivers/net/ethernet/cavium/liquidio/octeon_mem_ops.h
index bae2fdd89503..47a3ff5f9b1e 100644
--- a/drivers/net/ethernet/cavium/liquidio/octeon_mem_ops.h
+++ b/drivers/net/ethernet/cavium/liquidio/octeon_mem_ops.h
@@ -66,7 +66,7 @@ octeon_pci_read_core_mem(struct octeon_device *oct,
 void
 octeon_pci_write_core_mem(struct octeon_device *oct,
 			  u64 coreaddr,
-			  u8 *buf,
+			  const u8 *buf,
 			  u32 len);
 
 #endif
diff --git a/drivers/net/ethernet/cavium/liquidio/octeon_network.h b/drivers/net/ethernet/cavium/liquidio/octeon_network.h
index bf483932ff25..ebe56bd8849b 100644
--- a/drivers/net/ethernet/cavium/liquidio/octeon_network.h
+++ b/drivers/net/ethernet/cavium/liquidio/octeon_network.h
@@ -33,6 +33,37 @@
 #define   LIO_IFSTATE_REGISTERED           0x02
 #define   LIO_IFSTATE_RUNNING              0x04
 #define   LIO_IFSTATE_RX_TIMESTAMP_ENABLED 0x08
+#define   LIO_IFSTATE_RESETTING		   0x10
+
+struct liquidio_if_cfg_resp {
+	u64 rh;
+	struct liquidio_if_cfg_info cfg_info;
+	u64 status;
+};
+
+#define LIO_IFCFG_WAIT_TIME    3000 /* In milli seconds */
+#define LIQUIDIO_NDEV_STATS_POLL_TIME_MS 200
+
+/* Structure of a node in list of gather components maintained by
+ * NIC driver for each network device.
+ */
+struct octnic_gather {
+	/* List manipulation. Next and prev pointers. */
+	struct list_head list;
+
+	/* Size of the gather component at sg in bytes. */
+	int sg_size;
+
+	/* Number of bytes that sg was adjusted to make it 8B-aligned. */
+	int adjust;
+
+	/* Gather component that can accommodate max sized fragment list
+	 * received from the IP layer.
+	 */
+	struct octeon_sg_entry *sg;
+
+	dma_addr_t sg_dma_ptr;
+};
 
 struct oct_nic_stats_resp {
 	u64     rh;
@@ -40,11 +71,26 @@ struct oct_nic_stats_resp {
 	u64     status;
 };
 
+struct oct_nic_vf_stats_resp {
+	u64     rh;
+	u64	spoofmac_cnt;
+	u64     status;
+};
+
 struct oct_nic_stats_ctrl {
 	struct completion complete;
 	struct net_device *netdev;
 };
 
+struct oct_nic_seapi_resp {
+	u64 rh;
+	union {
+		u32 fec_setting;
+		u32 speed;
+	};
+	u64 status;
+};
+
 /** LiquidIO per-interface network private data */
 struct lio {
 	/** State of the interface. Rx/Tx happens only in the RUNNING state.  */
@@ -130,18 +176,22 @@ struct lio {
 	struct cavium_wq	txq_status_wq;
 
 	/* work queue for  rxq oom status */
-	struct cavium_wq	rxq_status_wq;
+	struct cavium_wq rxq_status_wq[MAX_POSSIBLE_OCTEON_OUTPUT_QUEUES];
 
 	/* work queue for  link status */
 	struct cavium_wq	link_status_wq;
 
+	/* work queue to regularly send local time to octeon firmware */
+	struct cavium_wq	sync_octeon_time_wq;
+
 	int netdev_uc_count;
+	struct cavium_wk stats_wk;
 };
 
 #define LIO_SIZE         (sizeof(struct lio))
 #define GET_LIO(netdev)  ((struct lio *)netdev_priv(netdev))
 
-#define LIO_MAX_CORES                12
+#define LIO_MAX_CORES                16
 
 /**
  * \brief Enable or disable feature
@@ -166,12 +216,40 @@ void cleanup_rx_oom_poll_fn(struct net_device *netdev);
  */
 void liquidio_link_ctrl_cmd_completion(void *nctrl_ptr);
 
+int liquidio_setup_io_queues(struct octeon_device *octeon_dev, int ifidx,
+			     u32 num_iqs, u32 num_oqs);
+
+irqreturn_t liquidio_msix_intr_handler(int irq __attribute__((unused)),
+				       void *dev);
+
+int octeon_setup_interrupt(struct octeon_device *oct, u32 num_ioqs);
+
+void lio_fetch_stats(struct work_struct *work);
+
+int lio_wait_for_clean_oq(struct octeon_device *oct);
 /**
  * \brief Register ethtool operations
  * @param netdev    pointer to network device
  */
 void liquidio_set_ethtool_ops(struct net_device *netdev);
 
+void lio_delete_glists(struct lio *lio);
+
+int lio_setup_glists(struct octeon_device *oct, struct lio *lio, int num_qs);
+
+int liquidio_get_speed(struct lio *lio);
+int liquidio_set_speed(struct lio *lio, int speed);
+int liquidio_get_fec(struct lio *lio);
+int liquidio_set_fec(struct lio *lio, int on_off);
+
+/**
+ * \brief Net device change_mtu
+ * @param netdev network device
+ */
+int liquidio_change_mtu(struct net_device *netdev, int new_mtu);
+#define LIO_CHANGE_MTU_SUCCESS 1
+#define LIO_CHANGE_MTU_FAIL    2
+
 #define SKB_ADJ_MASK  0x3F
 #define SKB_ADJ       (SKB_ADJ_MASK + 1)
 
@@ -186,7 +264,7 @@ static inline void
 	struct sk_buff *skb;
 	struct octeon_skb_page_info *skb_pg_info;
 
-	page = alloc_page(GFP_ATOMIC | __GFP_COLD);
+	page = alloc_page(GFP_ATOMIC);
 	if (unlikely(!page))
 		return NULL;
 
@@ -356,29 +434,6 @@ static inline void tx_buffer_free(void *buffer)
 #define lio_dma_free(oct, size, virt_addr, dma_addr) \
 	dma_free_coherent(&(oct)->pci_dev->dev, size, virt_addr, dma_addr)
 
-static inline void *
-lio_alloc_info_buffer(struct octeon_device *oct,
-		      struct octeon_droq *droq)
-{
-	void *virt_ptr;
-
-	virt_ptr = lio_dma_alloc(oct, (droq->max_count * OCT_DROQ_INFO_SIZE),
-				 &droq->info_list_dma);
-	if (virt_ptr) {
-		droq->info_alloc_size = droq->max_count * OCT_DROQ_INFO_SIZE;
-		droq->info_base_addr = virt_ptr;
-	}
-
-	return virt_ptr;
-}
-
-static inline void lio_free_info_buffer(struct octeon_device *oct,
-					struct octeon_droq *droq)
-{
-	lio_dma_free(oct, droq->info_alloc_size, droq->info_base_addr,
-		     droq->info_list_dma);
-}
-
 static inline
 void *get_rbd(struct sk_buff *skb)
 {
@@ -392,12 +447,6 @@ void *get_rbd(struct sk_buff *skb)
 }
 
 static inline u64
-lio_map_ring_info(struct octeon_droq *droq, u32 i)
-{
-	return droq->info_list_dma + (i * sizeof(struct octeon_droq_info));
-}
-
-static inline u64
 lio_map_ring(void *buf)
 {
 	dma_addr_t dma_addr;
@@ -443,8 +492,8 @@ static inline void octeon_fast_packet_next(struct octeon_droq *droq,
 					   int copy_len,
 					   int idx)
 {
-	memcpy(skb_put(nicbuf, copy_len),
-	       get_rbd(droq->recv_buf_list[idx].buffer), copy_len);
+	skb_put_data(nicbuf, get_rbd(droq->recv_buf_list[idx].buffer),
+		     copy_len);
 }
 
 /**
@@ -477,4 +526,101 @@ static inline void ifstate_reset(struct lio *lio, int state_flag)
 	atomic_set(&lio->ifstate, (atomic_read(&lio->ifstate) & ~(state_flag)));
 }
 
+/**
+ * \brief wait for all pending requests to complete
+ * @param oct Pointer to Octeon device
+ *
+ * Called during shutdown sequence
+ */
+static inline int wait_for_pending_requests(struct octeon_device *oct)
+{
+	int i, pcount = 0;
+
+	for (i = 0; i < MAX_IO_PENDING_PKT_COUNT; i++) {
+		pcount = atomic_read(
+		    &oct->response_list[OCTEON_ORDERED_SC_LIST]
+			 .pending_req_count);
+		if (pcount)
+			schedule_timeout_uninterruptible(HZ / 10);
+		else
+			break;
+	}
+
+	if (pcount)
+		return 1;
+
+	return 0;
+}
+
+/**
+ * \brief Stop Tx queues
+ * @param netdev network device
+ */
+static inline void stop_txqs(struct net_device *netdev)
+{
+	int i;
+
+	for (i = 0; i < netdev->real_num_tx_queues; i++)
+		netif_stop_subqueue(netdev, i);
+}
+
+/**
+ * \brief Wake Tx queues
+ * @param netdev network device
+ */
+static inline void wake_txqs(struct net_device *netdev)
+{
+	struct lio *lio = GET_LIO(netdev);
+	int i, qno;
+
+	for (i = 0; i < netdev->real_num_tx_queues; i++) {
+		qno = lio->linfo.txpciq[i % lio->oct_dev->num_iqs].s.q_no;
+
+		if (__netif_subqueue_stopped(netdev, i)) {
+			INCR_INSTRQUEUE_PKT_COUNT(lio->oct_dev, qno,
+						  tx_restart, 1);
+			netif_wake_subqueue(netdev, i);
+		}
+	}
+}
+
+/**
+ * \brief Start Tx queues
+ * @param netdev network device
+ */
+static inline void start_txqs(struct net_device *netdev)
+{
+	struct lio *lio = GET_LIO(netdev);
+	int i;
+
+	if (lio->linfo.link.s.link_up) {
+		for (i = 0; i < netdev->real_num_tx_queues; i++)
+			netif_start_subqueue(netdev, i);
+	}
+}
+
+static inline int skb_iq(struct octeon_device *oct, struct sk_buff *skb)
+{
+	return skb->queue_mapping % oct->num_iqs;
+}
+
+/**
+ * Remove the node at the head of the list. The list would be empty at
+ * the end of this call if there are no more nodes in the list.
+ */
+static inline struct list_head *lio_list_delete_head(struct list_head *root)
+{
+	struct list_head *node;
+
+	if (list_empty_careful(root))
+		node = NULL;
+	else
+		node = root->next;
+
+	if (node)
+		list_del(node);
+
+	return node;
+}
+
 #endif
diff --git a/drivers/net/ethernet/cavium/liquidio/octeon_nic.c b/drivers/net/ethernet/cavium/liquidio/octeon_nic.c
index b457cf23fce6..dee56ea740e7 100644
--- a/drivers/net/ethernet/cavium/liquidio/octeon_nic.c
+++ b/drivers/net/ethernet/cavium/liquidio/octeon_nic.c
@@ -75,44 +75,23 @@ octeon_alloc_soft_command_resp(struct octeon_device    *oct,
 	else
 		sc->cmd.cmd2.rptr =  sc->dmarptr;
 
-	sc->wait_time = 1000;
-	sc->timeout = jiffies + sc->wait_time;
+	sc->expiry_time = jiffies + msecs_to_jiffies(LIO_SC_MAX_TMO_MS);
 
 	return sc;
 }
+EXPORT_SYMBOL_GPL(octeon_alloc_soft_command_resp);
 
 int octnet_send_nic_data_pkt(struct octeon_device *oct,
-			     struct octnic_data_pkt *ndata)
+			     struct octnic_data_pkt *ndata,
+			     int xmit_more)
 {
-	int ring_doorbell = 1;
+	int ring_doorbell = !xmit_more;
 
 	return octeon_send_command(oct, ndata->q_no, ring_doorbell, &ndata->cmd,
 				   ndata->buf, ndata->datasize,
 				   ndata->reqtype);
 }
-
-static void octnet_link_ctrl_callback(struct octeon_device *oct,
-				      u32 status,
-				      void *sc_ptr)
-{
-	struct octeon_soft_command *sc = (struct octeon_soft_command *)sc_ptr;
-	struct octnic_ctrl_pkt *nctrl;
-
-	nctrl = (struct octnic_ctrl_pkt *)sc->ctxptr;
-
-	/* Call the callback function if status is zero (meaning OK) or status
-	 * contains a firmware status code bigger than zero (meaning the
-	 * firmware is reporting an error).
-	 * If no response was expected, status is OK if the command was posted
-	 * successfully.
-	 */
-	if ((!status || status > FIRMWARE_STATUS_CODE(0)) && nctrl->cb_fn) {
-		nctrl->status = status;
-		nctrl->cb_fn(nctrl);
-	}
-
-	octeon_free_soft_command(oct, sc);
-}
+EXPORT_SYMBOL_GPL(octnet_send_nic_data_pkt);
 
 static inline struct octeon_soft_command
 *octnic_alloc_ctrl_pkt_sc(struct octeon_device *oct,
@@ -126,17 +105,14 @@ static inline struct octeon_soft_command
 	uddsize = (u32)(nctrl->ncmd.s.more * 8);
 
 	datasize = OCTNET_CMD_SIZE + uddsize;
-	rdatasize = (nctrl->wait_time) ? 16 : 0;
+	rdatasize = 16;
 
 	sc = (struct octeon_soft_command *)
-		octeon_alloc_soft_command(oct, datasize, rdatasize,
-					  sizeof(struct octnic_ctrl_pkt));
+		octeon_alloc_soft_command(oct, datasize, rdatasize, 0);
 
 	if (!sc)
 		return NULL;
 
-	memcpy(sc->ctxptr, nctrl, sizeof(struct octnic_ctrl_pkt));
-
 	data = (u8 *)sc->virtdptr;
 
 	memcpy(data, &nctrl->ncmd, OCTNET_CMD_SIZE);
@@ -153,9 +129,8 @@ static inline struct octeon_soft_command
 	octeon_prepare_soft_command(oct, sc, OPCODE_NIC, OPCODE_NIC_CMD,
 				    0, 0, 0);
 
-	sc->callback = octnet_link_ctrl_callback;
-	sc->callback_arg = sc;
-	sc->wait_time = nctrl->wait_time;
+	init_completion(&sc->complete);
+	sc->sc_status = OCTEON_REQUEST_PENDING;
 
 	return sc;
 }
@@ -198,5 +173,29 @@ octnet_send_nic_ctrl_pkt(struct octeon_device *oct,
 	}
 
 	spin_unlock_bh(&oct->cmd_resp_wqlock);
+
+	if (nctrl->ncmd.s.cmdgroup == 0) {
+		switch (nctrl->ncmd.s.cmd) {
+			/* caller holds lock, can not sleep */
+		case OCTNET_CMD_CHANGE_DEVFLAGS:
+		case OCTNET_CMD_SET_MULTI_LIST:
+		case OCTNET_CMD_SET_UC_LIST:
+			WRITE_ONCE(sc->caller_is_done, true);
+			return retval;
+		}
+	}
+
+	retval = wait_for_sc_completion_timeout(oct, sc, 0);
+	if (retval)
+		return (retval);
+
+	nctrl->sc_status = sc->sc_status;
+	retval = nctrl->sc_status;
+	if (nctrl->cb_fn)
+		nctrl->cb_fn(nctrl);
+
+	WRITE_ONCE(sc->caller_is_done, true);
+
 	return retval;
 }
+EXPORT_SYMBOL_GPL(octnet_send_nic_ctrl_pkt);
diff --git a/drivers/net/ethernet/cavium/liquidio/octeon_nic.h b/drivers/net/ethernet/cavium/liquidio/octeon_nic.h
index 6480ef863441..c139fc423764 100644
--- a/drivers/net/ethernet/cavium/liquidio/octeon_nic.h
+++ b/drivers/net/ethernet/cavium/liquidio/octeon_nic.h
@@ -52,20 +52,13 @@ struct octnic_ctrl_pkt {
 	/** Input queue to use to send this command. */
 	u64 iq_no;
 
-	/** Time to wait for Octeon software to respond to this control command.
-	 *  If wait_time is 0, OSI assumes no response is expected.
-	 */
-	size_t wait_time;
-
 	/** The network device that issued the control command. */
 	u64 netpndev;
 
 	/** Callback function called when the command has been fetched */
 	octnic_ctrl_pkt_cb_fn_t cb_fn;
 
-	u32 status;
-	u16 *response_code;
-	struct completion *completion;
+	u32 sc_status;
 };
 
 #define MAX_UDD_SIZE(nctrl) (sizeof((nctrl)->udd))
@@ -275,16 +268,17 @@ octeon_alloc_soft_command_resp(struct octeon_device    *oct,
  * @param oct - octeon device pointer
  * @param ndata - control structure with queueing, and buffer information
  *
- * @returns IQ_FAILED if it failed to add to the input queue. IQ_STOP if it the
+ * @returns IQ_FAILED if it failed to add to the input queue. IQ_STOP if the
  * queue should be stopped, and IQ_SEND_OK if it sent okay.
  */
 int octnet_send_nic_data_pkt(struct octeon_device *oct,
-			     struct octnic_data_pkt *ndata);
+			     struct octnic_data_pkt *ndata,
+			     int xmit_more);
 
 /** Send a NIC control packet to the device
  * @param oct - octeon device pointer
  * @param nctrl - control structure with command, timout, and callback info
- * @returns IQ_FAILED if it failed to add to the input queue. IQ_STOP if it the
+ * @returns IQ_FAILED if it failed to add to the input queue. IQ_STOP if the
  * queue should be stopped, and IQ_SEND_OK if it sent okay.
  */
 int
diff --git a/drivers/net/ethernet/cavium/liquidio/request_manager.c b/drivers/net/ethernet/cavium/liquidio/request_manager.c
index 261f448f9de2..d7cfb20eea00 100644
--- a/drivers/net/ethernet/cavium/liquidio/request_manager.c
+++ b/drivers/net/ethernet/cavium/liquidio/request_manager.c
@@ -40,15 +40,6 @@ static void  __check_db_timeout(struct octeon_device *oct, u64 iq_no);
 
 static void (*reqtype_free_fn[MAX_OCTEON_DEVICES][REQTYPE_LAST + 1]) (void *);
 
-static inline int IQ_INSTR_MODE_64B(struct octeon_device *oct, int iq_no)
-{
-	struct octeon_instr_queue *iq =
-	    (struct octeon_instr_queue *)oct->instr_queue[iq_no];
-	return iq->iqcmd_64B;
-}
-
-#define IQ_INSTR_MODE_32B(oct, iq_no)  (!IQ_INSTR_MODE_64B(oct, iq_no))
-
 /* Define this to return the request status comaptible to old code */
 /*#define OCTEON_USE_OLD_REQ_STATUS*/
 
@@ -77,13 +68,6 @@ int octeon_init_instr_queue(struct octeon_device *oct,
 		return 1;
 	}
 
-	if (num_descs & (num_descs - 1)) {
-		dev_err(&oct->pci_dev->dev,
-			"Number of descriptors for instr queue %d not in power of 2.\n",
-			iq_no);
-		return 1;
-	}
-
 	q_size = (u32)conf->instr_type * num_descs;
 
 	iq = oct->instr_queue[iq_no];
@@ -102,11 +86,10 @@ int octeon_init_instr_queue(struct octeon_device *oct,
 	/* Initialize a list to holds requests that have been posted to Octeon
 	 * but has yet to be fetched by octeon
 	 */
-	iq->request_list = vmalloc_node((sizeof(*iq->request_list) * num_descs),
-					       numa_node);
+	iq->request_list = vzalloc_node(array_size(num_descs, sizeof(*iq->request_list)),
+					numa_node);
 	if (!iq->request_list)
-		iq->request_list = vmalloc(sizeof(*iq->request_list) *
-						  num_descs);
+		iq->request_list = vzalloc(array_size(num_descs, sizeof(*iq->request_list)));
 	if (!iq->request_list) {
 		lio_dma_free(oct, q_size, iq->base_addr, iq->base_addr_dma);
 		dev_err(&oct->pci_dev->dev, "Alloc failed for IQ[%d] nr free list\n",
@@ -114,10 +97,8 @@ int octeon_init_instr_queue(struct octeon_device *oct,
 		return 1;
 	}
 
-	memset(iq->request_list, 0, sizeof(*iq->request_list) * num_descs);
-
-	dev_dbg(&oct->pci_dev->dev, "IQ[%d]: base: %p basedma: %llx count: %d\n",
-		iq_no, iq->base_addr, iq->base_addr_dma, iq->max_count);
+	dev_dbg(&oct->pci_dev->dev, "IQ[%d]: base: %p basedma: %pad count: %d\n",
+		iq_no, iq->base_addr, &iq->base_addr_dma, iq->max_count);
 
 	iq->txpciq.u64 = txpciq.u64;
 	iq->fill_threshold = (u32)conf->db_min;
@@ -129,23 +110,29 @@ int octeon_init_instr_queue(struct octeon_device *oct,
 	iq->do_auto_flush = 1;
 	iq->db_timeout = (u32)conf->db_timeout;
 	atomic_set(&iq->instr_pending, 0);
+	iq->pkts_processed = 0;
 
 	/* Initialize the spinlock for this instruction queue */
 	spin_lock_init(&iq->lock);
-	spin_lock_init(&iq->post_lock);
+	if (iq_no == 0) {
+		iq->allow_soft_cmds = true;
+		spin_lock_init(&iq->post_lock);
+	} else {
+		iq->allow_soft_cmds = false;
+	}
 
 	spin_lock_init(&iq->iq_flush_running_lock);
 
 	oct->io_qmask.iq |= BIT_ULL(iq_no);
 
 	/* Set the 32B/64B mode for each input queue */
-	oct->io_qmask.iq64B |= ((conf->instr_type == 64) << iq_no);
+	oct->io_qmask.iq64B |= ((u64)(conf->instr_type == 64) << iq_no);
 	iq->iqcmd_64B = (conf->instr_type == 64);
 
 	oct->fn_list.setup_iq_regs(oct, iq_no);
 
 	oct->check_db_wq[iq_no].wq = alloc_workqueue("check_iq_db",
-						     WQ_MEM_RECLAIM,
+						     WQ_MEM_RECLAIM | WQ_PERCPU,
 						     0);
 	if (!oct->check_db_wq[iq_no].wq) {
 		vfree(iq->request_list);
@@ -190,10 +177,15 @@ int octeon_delete_instr_queue(struct octeon_device *oct, u32 iq_no)
 		q_size = iq->max_count * desc_size;
 		lio_dma_free(oct, (u32)q_size, iq->base_addr,
 			     iq->base_addr_dma);
+		oct->io_qmask.iq &= ~(1ULL << iq_no);
+		vfree(oct->instr_queue[iq_no]);
+		oct->instr_queue[iq_no] = NULL;
+		oct->num_iqs--;
 		return 0;
 	}
 	return 1;
 }
+EXPORT_SYMBOL_GPL(octeon_delete_instr_queue);
 
 /* Return 0 on success, 1 on failure */
 int octeon_setup_iq(struct octeon_device *oct,
@@ -214,15 +206,13 @@ int octeon_setup_iq(struct octeon_device *oct,
 		return 0;
 	}
 	oct->instr_queue[iq_no] =
-	    vmalloc_node(sizeof(struct octeon_instr_queue), numa_node);
+	    vzalloc_node(sizeof(struct octeon_instr_queue), numa_node);
 	if (!oct->instr_queue[iq_no])
 		oct->instr_queue[iq_no] =
-		    vmalloc(sizeof(struct octeon_instr_queue));
+		    vzalloc(sizeof(struct octeon_instr_queue));
 	if (!oct->instr_queue[iq_no])
 		return 1;
 
-	memset(oct->instr_queue[iq_no], 0,
-	       sizeof(struct octeon_instr_queue));
 
 	oct->instr_queue[iq_no]->q_index = q_index;
 	oct->instr_queue[iq_no]->app_ctx = app_ctx;
@@ -235,8 +225,10 @@ int octeon_setup_iq(struct octeon_device *oct,
 	}
 
 	oct->num_iqs++;
-	if (oct->fn_list.enable_io_queues(oct))
+	if (oct->fn_list.enable_io_queues(oct)) {
+		octeon_delete_instr_queue(oct, iq_no);
 		return 1;
+	}
 
 	return 0;
 }
@@ -252,8 +244,7 @@ int lio_wait_for_instr_fetch(struct octeon_device *oct)
 			if (!(oct->io_qmask.iq & BIT_ULL(i)))
 				continue;
 			pending =
-			    atomic_read(&oct->
-					       instr_queue[i]->instr_pending);
+			    atomic_read(&oct->instr_queue[i]->instr_pending);
 			if (pending)
 				__check_db_timeout(oct, i);
 			instr_cnt += pending;
@@ -268,6 +259,7 @@ int lio_wait_for_instr_fetch(struct octeon_device *oct)
 
 	return instr_cnt;
 }
+EXPORT_SYMBOL_GPL(lio_wait_for_instr_fetch);
 
 static inline void
 ring_doorbell(struct octeon_device *oct, struct octeon_instr_queue *iq)
@@ -275,13 +267,25 @@ ring_doorbell(struct octeon_device *oct, struct octeon_instr_queue *iq)
 	if (atomic_read(&oct->status) == OCT_DEV_RUNNING) {
 		writel(iq->fill_cnt, iq->doorbell_reg);
 		/* make sure doorbell write goes through */
-		mmiowb();
 		iq->fill_cnt = 0;
 		iq->last_db_time = jiffies;
 		return;
 	}
 }
 
+void
+octeon_ring_doorbell_locked(struct octeon_device *oct, u32 iq_no)
+{
+	struct octeon_instr_queue *iq;
+
+	iq = oct->instr_queue[iq_no];
+	spin_lock(&iq->post_lock);
+	if (iq->fill_cnt)
+		ring_doorbell(oct, iq);
+	spin_unlock(&iq->post_lock);
+}
+EXPORT_SYMBOL_GPL(octeon_ring_doorbell_locked);
+
 static inline void __copy_cmd_into_iq(struct octeon_instr_queue *iq,
 				      u8 *cmd)
 {
@@ -344,6 +348,7 @@ octeon_register_reqtype_free_fn(struct octeon_device *oct, int reqtype,
 
 	return 0;
 }
+EXPORT_SYMBOL_GPL(octeon_register_reqtype_free_fn);
 
 static inline void
 __add_to_request_list(struct octeon_instr_queue *iq,
@@ -358,13 +363,13 @@ int
 lio_process_iq_request_list(struct octeon_device *oct,
 			    struct octeon_instr_queue *iq, u32 napi_budget)
 {
+	struct cavium_wq *cwq = &oct->dma_comp_wq;
 	int reqtype;
 	void *buf;
 	u32 old = iq->flush_index;
 	u32 inst_count = 0;
 	unsigned int pkts_compl = 0, bytes_compl = 0;
 	struct octeon_soft_command *sc;
-	struct octeon_instr_irh *irh;
 	unsigned long flags;
 
 	while (old != iq->octeon_read_index) {
@@ -386,40 +391,21 @@ lio_process_iq_request_list(struct octeon_device *oct,
 		case REQTYPE_RESP_NET:
 		case REQTYPE_SOFT_COMMAND:
 			sc = buf;
-
-			if (OCTEON_CN23XX_PF(oct) || OCTEON_CN23XX_VF(oct))
-				irh = (struct octeon_instr_irh *)
-					&sc->cmd.cmd3.irh;
-			else
-				irh = (struct octeon_instr_irh *)
-					&sc->cmd.cmd2.irh;
-			if (irh->rflag) {
-				/* We're expecting a response from Octeon.
-				 * It's up to lio_process_ordered_list() to
-				 * process  sc. Add sc to the ordered soft
-				 * command response list because we expect
-				 * a response from Octeon.
-				 */
-				spin_lock_irqsave
-					(&oct->response_list
-					 [OCTEON_ORDERED_SC_LIST].lock,
-					 flags);
-				atomic_inc(&oct->response_list
-					[OCTEON_ORDERED_SC_LIST].
-					pending_req_count);
-				list_add_tail(&sc->node, &oct->response_list
-					[OCTEON_ORDERED_SC_LIST].head);
-				spin_unlock_irqrestore
-					(&oct->response_list
-					 [OCTEON_ORDERED_SC_LIST].lock,
-					 flags);
-			} else {
-				if (sc->callback) {
-					/* This callback must not sleep */
-					sc->callback(oct, OCTEON_REQUEST_DONE,
-						     sc->callback_arg);
-				}
-			}
+			/* We're expecting a response from Octeon.
+			 * It's up to lio_process_ordered_list() to
+			 * process  sc. Add sc to the ordered soft
+			 * command response list because we expect
+			 * a response from Octeon.
+			 */
+			spin_lock_irqsave(&oct->response_list
+					  [OCTEON_ORDERED_SC_LIST].lock, flags);
+			atomic_inc(&oct->response_list
+				   [OCTEON_ORDERED_SC_LIST].pending_req_count);
+			list_add_tail(&sc->node, &oct->response_list
+				[OCTEON_ORDERED_SC_LIST].head);
+			spin_unlock_irqrestore(&oct->response_list
+					       [OCTEON_ORDERED_SC_LIST].lock,
+					       flags);
 			break;
 		default:
 			dev_err(&oct->pci_dev->dev,
@@ -442,8 +428,13 @@ lio_process_iq_request_list(struct octeon_device *oct,
 						   bytes_compl);
 	iq->flush_index = old;
 
+	if (atomic_read(&oct->response_list
+			[OCTEON_ORDERED_SC_LIST].pending_req_count))
+		queue_work(cwq->wq, &cwq->wk.work.work);
+
 	return inst_count;
 }
+EXPORT_SYMBOL_GPL(lio_process_iq_request_list);
 
 /* Can only be called from process context */
 int
@@ -476,13 +467,12 @@ octeon_flush_iq(struct octeon_device *oct, struct octeon_instr_queue *iq,
 				lio_process_iq_request_list(oct, iq, 0);
 
 		if (inst_processed) {
+			iq->pkts_processed += inst_processed;
 			atomic_sub(inst_processed, &iq->instr_pending);
 			iq->stats.instr_processed += inst_processed;
 		}
 
 		tot_inst_processed += inst_processed;
-		inst_processed = 0;
-
 	} while (tot_inst_processed < napi_budget);
 
 	if (napi_budget && (tot_inst_processed >= napi_budget))
@@ -547,29 +537,33 @@ octeon_send_command(struct octeon_device *oct, u32 iq_no,
 		    u32 force_db, void *cmd, void *buf,
 		    u32 datasize, u32 reqtype)
 {
+	int xmit_stopped;
 	struct iq_post_status st;
 	struct octeon_instr_queue *iq = oct->instr_queue[iq_no];
 
 	/* Get the lock and prevent other tasks and tx interrupt handler from
 	 * running.
 	 */
-	spin_lock_bh(&iq->post_lock);
+	if (iq->allow_soft_cmds)
+		spin_lock_bh(&iq->post_lock);
 
 	st = __post_command2(iq, cmd);
 
 	if (st.status != IQ_SEND_FAILED) {
-		octeon_report_sent_bytes_to_bql(buf, reqtype);
+		xmit_stopped = octeon_report_sent_bytes_to_bql(buf, reqtype);
 		__add_to_request_list(iq, st.index, buf, reqtype);
 		INCR_INSTRQUEUE_PKT_COUNT(oct, iq_no, bytes_sent, datasize);
 		INCR_INSTRQUEUE_PKT_COUNT(oct, iq_no, instr_posted, 1);
 
-		if (force_db)
+		if (iq->fill_cnt >= MAX_OCTEON_FILL_COUNT || force_db ||
+		    xmit_stopped || st.status == IQ_SEND_STOP)
 			ring_doorbell(oct, iq);
 	} else {
 		INCR_INSTRQUEUE_PKT_COUNT(oct, iq_no, instr_dropped, 1);
 	}
 
-	spin_unlock_bh(&iq->post_lock);
+	if (iq->allow_soft_cmds)
+		spin_unlock_bh(&iq->post_lock);
 
 	/* This is only done here to expedite packets being flushed
 	 * for cases where there are no IQ completion interrupts.
@@ -577,6 +571,7 @@ octeon_send_command(struct octeon_device *oct, u32 iq_no,
 
 	return st.status;
 }
+EXPORT_SYMBOL_GPL(octeon_send_command);
 
 void
 octeon_prepare_soft_command(struct octeon_device *oct,
@@ -615,7 +610,8 @@ octeon_prepare_soft_command(struct octeon_device *oct,
 		pki_ih3->tag     = LIO_CONTROL;
 		pki_ih3->tagtype = ATOMIC_TAG;
 		pki_ih3->qpg         =
-			oct->instr_queue[sc->iq_no]->txpciq.s.qpg;
+			oct->instr_queue[sc->iq_no]->txpciq.s.ctrl_qpg;
+
 		pki_ih3->pm          = 0x7;
 		pki_ih3->sl          = 8;
 
@@ -683,15 +679,25 @@ octeon_prepare_soft_command(struct octeon_device *oct,
 		}
 	}
 }
+EXPORT_SYMBOL_GPL(octeon_prepare_soft_command);
 
 int octeon_send_soft_command(struct octeon_device *oct,
 			     struct octeon_soft_command *sc)
 {
+	struct octeon_instr_queue *iq;
 	struct octeon_instr_ih2 *ih2;
 	struct octeon_instr_ih3 *ih3;
 	struct octeon_instr_irh *irh;
 	u32 len;
 
+	iq = oct->instr_queue[sc->iq_no];
+	if (!iq->allow_soft_cmds) {
+		dev_err(&oct->pci_dev->dev, "Soft commands are not allowed on Queue %d\n",
+			sc->iq_no);
+		INCR_INSTRQUEUE_PKT_COUNT(oct, sc->iq_no, instr_dropped, 1);
+		return IQ_SEND_FAILED;
+	}
+
 	if (OCTEON_CN23XX_PF(oct) || OCTEON_CN23XX_VF(oct)) {
 		ih3 =  (struct octeon_instr_ih3 *)&sc->cmd.cmd3.ih3;
 		if (ih3->dlengsz) {
@@ -722,12 +728,12 @@ int octeon_send_soft_command(struct octeon_device *oct,
 		len = (u32)ih2->dlengsz;
 	}
 
-	if (sc->wait_time)
-		sc->timeout = jiffies + sc->wait_time;
+	sc->expiry_time = jiffies + msecs_to_jiffies(LIO_SC_MAX_TMO_MS);
 
 	return (octeon_send_command(oct, sc->iq_no, 1, &sc->cmd, sc,
 				    len, REQTYPE_SOFT_COMMAND));
 }
+EXPORT_SYMBOL_GPL(octeon_send_soft_command);
 
 int octeon_setup_sc_buffer_pool(struct octeon_device *oct)
 {
@@ -757,12 +763,80 @@ int octeon_setup_sc_buffer_pool(struct octeon_device *oct)
 
 	return 0;
 }
+EXPORT_SYMBOL_GPL(octeon_setup_sc_buffer_pool);
+
+int octeon_free_sc_done_list(struct octeon_device *oct)
+{
+	struct octeon_response_list *done_sc_list, *zombie_sc_list;
+	struct octeon_soft_command *sc;
+	struct list_head *tmp, *tmp2;
+	spinlock_t *sc_lists_lock; /* lock for response_list */
+
+	done_sc_list = &oct->response_list[OCTEON_DONE_SC_LIST];
+	zombie_sc_list = &oct->response_list[OCTEON_ZOMBIE_SC_LIST];
+
+	if (!atomic_read(&done_sc_list->pending_req_count))
+		return 0;
+
+	sc_lists_lock = &oct->response_list[OCTEON_ORDERED_SC_LIST].lock;
+
+	spin_lock_bh(sc_lists_lock);
+
+	list_for_each_safe(tmp, tmp2, &done_sc_list->head) {
+		sc = list_entry(tmp, struct octeon_soft_command, node);
+
+		if (READ_ONCE(sc->caller_is_done)) {
+			list_del(&sc->node);
+			atomic_dec(&done_sc_list->pending_req_count);
+
+			if (*sc->status_word == COMPLETION_WORD_INIT) {
+				/* timeout; move sc to zombie list */
+				list_add_tail(&sc->node, &zombie_sc_list->head);
+				atomic_inc(&zombie_sc_list->pending_req_count);
+			} else {
+				octeon_free_soft_command(oct, sc);
+			}
+		}
+	}
+
+	spin_unlock_bh(sc_lists_lock);
+
+	return 0;
+}
+EXPORT_SYMBOL_GPL(octeon_free_sc_done_list);
+
+int octeon_free_sc_zombie_list(struct octeon_device *oct)
+{
+	struct octeon_response_list *zombie_sc_list;
+	struct octeon_soft_command *sc;
+	struct list_head *tmp, *tmp2;
+	spinlock_t *sc_lists_lock; /* lock for response_list */
+
+	zombie_sc_list = &oct->response_list[OCTEON_ZOMBIE_SC_LIST];
+	sc_lists_lock = &oct->response_list[OCTEON_ORDERED_SC_LIST].lock;
+
+	spin_lock_bh(sc_lists_lock);
+
+	list_for_each_safe(tmp, tmp2, &zombie_sc_list->head) {
+		list_del(tmp);
+		atomic_dec(&zombie_sc_list->pending_req_count);
+		sc = list_entry(tmp, struct octeon_soft_command, node);
+		octeon_free_soft_command(oct, sc);
+	}
+
+	spin_unlock_bh(sc_lists_lock);
+
+	return 0;
+}
+EXPORT_SYMBOL_GPL(octeon_free_sc_zombie_list);
 
 int octeon_free_sc_buffer_pool(struct octeon_device *oct)
 {
 	struct list_head *tmp, *tmp2;
 	struct octeon_soft_command *sc;
 
+	octeon_free_sc_zombie_list(oct);
+
 	spin_lock_bh(&oct->sc_buf_pool.lock);
 
 	list_for_each_safe(tmp, tmp2, &oct->sc_buf_pool.head) {
@@ -779,6 +853,7 @@ int octeon_free_sc_buffer_pool(struct octeon_device *oct)
 
 	return 0;
 }
+EXPORT_SYMBOL_GPL(octeon_free_sc_buffer_pool);
 
 struct octeon_soft_command *octeon_alloc_soft_command(struct octeon_device *oct,
 						      u32 datasize,
@@ -791,6 +866,9 @@ struct octeon_soft_command *octeon_alloc_soft_command(struct octeon_device *oct,
 	struct octeon_soft_command *sc = NULL;
 	struct list_head *tmp;
 
+	if (!rdatasize)
+		rdatasize = 16;
+
 	WARN_ON((offset + datasize + rdatasize + ctxsize) >
 	       SOFT_COMMAND_BUFFER_SIZE);
 
@@ -847,6 +925,7 @@ struct octeon_soft_command *octeon_alloc_soft_command(struct octeon_device *oct,
 
 	return sc;
 }
+EXPORT_SYMBOL_GPL(octeon_alloc_soft_command);
 
 void octeon_free_soft_command(struct octeon_device *oct,
 			      struct octeon_soft_command *sc)
@@ -859,3 +938,4 @@ void octeon_free_soft_command(struct octeon_device *oct,
 
 	spin_unlock_bh(&oct->sc_buf_pool.lock);
 }
+EXPORT_SYMBOL_GPL(octeon_free_soft_command);
diff --git a/drivers/net/ethernet/cavium/liquidio/response_manager.c b/drivers/net/ethernet/cavium/liquidio/response_manager.c
index 3d691c69f74d..de1a8335b545 100644
--- a/drivers/net/ethernet/cavium/liquidio/response_manager.c
+++ b/drivers/net/ethernet/cavium/liquidio/response_manager.c
@@ -39,7 +39,8 @@ int octeon_setup_response_list(struct octeon_device *oct)
 	}
 	spin_lock_init(&oct->cmd_resp_wqlock);
 
-	oct->dma_comp_wq.wq = alloc_workqueue("dma-comp", WQ_MEM_RECLAIM, 0);
+	oct->dma_comp_wq.wq = alloc_workqueue("dma-comp",
+					      WQ_MEM_RECLAIM | WQ_PERCPU, 0);
 	if (!oct->dma_comp_wq.wq) {
 		dev_err(&oct->pci_dev->dev, "failed to create wq thread\n");
 		return -ENOMEM;
@@ -49,16 +50,17 @@ int octeon_setup_response_list(struct octeon_device *oct)
 	INIT_DELAYED_WORK(&cwq->wk.work, oct_poll_req_completion);
 	cwq->wk.ctxptr = oct;
 	oct->cmd_resp_state = OCT_DRV_ONLINE;
-	queue_delayed_work(cwq->wq, &cwq->wk.work, msecs_to_jiffies(50));
 
 	return ret;
 }
+EXPORT_SYMBOL_GPL(octeon_setup_response_list);
 
 void octeon_delete_response_list(struct octeon_device *oct)
 {
 	cancel_delayed_work_sync(&oct->dma_comp_wq.wk.work);
 	destroy_workqueue(oct->dma_comp_wq.wq);
 }
+EXPORT_SYMBOL_GPL(octeon_delete_response_list);
 
 int lio_process_ordered_list(struct octeon_device *octeon_dev,
 			     u32 force_quit)
@@ -70,6 +72,8 @@ int lio_process_ordered_list(struct octeon_device *octeon_dev,
 	u32 status;
 	u64 status64;
 
+	octeon_free_sc_done_list(octeon_dev);
+
 	ordered_sc_list = &octeon_dev->response_list[OCTEON_ORDERED_SC_LIST];
 
 	do {
@@ -112,26 +116,88 @@ int lio_process_ordered_list(struct octeon_device *octeon_dev,
 					}
 				}
 			}
-		} else if (force_quit || (sc->timeout &&
-			time_after(jiffies, (unsigned long)sc->timeout))) {
-			dev_err(&octeon_dev->pci_dev->dev, "%s: cmd failed, timeout (%ld, %ld)\n",
-				__func__, (long)jiffies, (long)sc->timeout);
+		} else if (unlikely(force_quit) || (sc->expiry_time &&
+			time_after(jiffies, (unsigned long)sc->expiry_time))) {
+			struct octeon_instr_irh *irh =
+				(struct octeon_instr_irh *)&sc->cmd.cmd3.irh;
+
+			dev_err(&octeon_dev->pci_dev->dev, "%s: ", __func__);
+			dev_err(&octeon_dev->pci_dev->dev,
+				"cmd %x/%x/%llx/%llx failed, ",
+				irh->opcode, irh->subcode,
+				sc->cmd.cmd3.ossp[0], sc->cmd.cmd3.ossp[1]);
+			dev_err(&octeon_dev->pci_dev->dev,
+				"timeout (%ld, %ld)\n",
+				(long)jiffies, (long)sc->expiry_time);
 			status = OCTEON_REQUEST_TIMEOUT;
 		}
 
 		if (status != OCTEON_REQUEST_PENDING) {
+			sc->sc_status = status;
+
 			/* we have received a response or we have timed out */
 			/* remove node from linked list */
 			list_del(&sc->node);
 			atomic_dec(&octeon_dev->response_list
-					  [OCTEON_ORDERED_SC_LIST].
-					  pending_req_count);
-			spin_unlock_bh
-			    (&ordered_sc_list->lock);
+				   [OCTEON_ORDERED_SC_LIST].
+				   pending_req_count);
+
+			if (!sc->callback) {
+				atomic_inc(&octeon_dev->response_list
+					   [OCTEON_DONE_SC_LIST].
+					   pending_req_count);
+				list_add_tail(&sc->node,
+					      &octeon_dev->response_list
+					      [OCTEON_DONE_SC_LIST].head);
+
+				if (unlikely(READ_ONCE(sc->caller_is_done))) {
+					/* caller does not wait for response
+					 * from firmware
+					 */
+					if (status != OCTEON_REQUEST_DONE) {
+						struct octeon_instr_irh *irh;
+
+						irh =
+						    (struct octeon_instr_irh *)
+						    &sc->cmd.cmd3.irh;
+						dev_dbg
+						    (&octeon_dev->pci_dev->dev,
+						    "%s: sc failed: opcode=%x, ",
+						    __func__, irh->opcode);
+						dev_dbg
+						    (&octeon_dev->pci_dev->dev,
+						    "subcode=%x, ossp[0]=%llx, ",
+						    irh->subcode,
+						    sc->cmd.cmd3.ossp[0]);
+						dev_dbg
+						    (&octeon_dev->pci_dev->dev,
+						    "ossp[1]=%llx, status=%d\n",
+						    sc->cmd.cmd3.ossp[1],
+						    status);
+					}
+				} else {
+					complete(&sc->complete);
+				}
+
+				spin_unlock_bh(&ordered_sc_list->lock);
+			} else {
+				/* sc with callback function */
+				if (status == OCTEON_REQUEST_TIMEOUT) {
+					atomic_inc(&octeon_dev->response_list
+						   [OCTEON_ZOMBIE_SC_LIST].
+						   pending_req_count);
+					list_add_tail(&sc->node,
+						      &octeon_dev->response_list
+						      [OCTEON_ZOMBIE_SC_LIST].
+						      head);
+				}
+
+				spin_unlock_bh(&ordered_sc_list->lock);
 
-			if (sc->callback)
 				sc->callback(octeon_dev, status,
 					     sc->callback_arg);
+				/* sc is freed by caller */
+			}
 
 			request_complete++;
 
@@ -156,6 +222,7 @@ int lio_process_ordered_list(struct octeon_device *octeon_dev,
 
 	return 0;
 }
+EXPORT_SYMBOL_GPL(lio_process_ordered_list);
 
 static void oct_poll_req_completion(struct work_struct *work)
 {
@@ -164,5 +231,8 @@ static void oct_poll_req_completion(struct work_struct *work)
 	struct cavium_wq *cwq = &oct->dma_comp_wq;
 
 	lio_process_ordered_list(oct, 0);
-	queue_delayed_work(cwq->wq, &cwq->wk.work, msecs_to_jiffies(50));
+
+	if (atomic_read(&oct->response_list
+			[OCTEON_ORDERED_SC_LIST].pending_req_count))
+		queue_delayed_work(cwq->wq, &cwq->wk.work, msecs_to_jiffies(1));
 }
diff --git a/drivers/net/ethernet/cavium/liquidio/response_manager.h b/drivers/net/ethernet/cavium/liquidio/response_manager.h
index 9169c2815dba..ed4020d26fae 100644
--- a/drivers/net/ethernet/cavium/liquidio/response_manager.h
+++ b/drivers/net/ethernet/cavium/liquidio/response_manager.h
@@ -53,7 +53,9 @@ enum {
 	OCTEON_ORDERED_LIST = 0,
 	OCTEON_UNORDERED_NONBLOCKING_LIST = 1,
 	OCTEON_UNORDERED_BLOCKING_LIST = 2,
-	OCTEON_ORDERED_SC_LIST = 3
+	OCTEON_ORDERED_SC_LIST = 3,
+	OCTEON_DONE_SC_LIST = 4,
+	OCTEON_ZOMBIE_SC_LIST = 5
 };
 
 /** Response Order values for a Octeon Request. */
diff --git a/drivers/net/ethernet/cavium/octeon/Makefile b/drivers/net/ethernet/cavium/octeon/Makefile
index efa41c1d91c5..4f5098f6bc14 100644
--- a/drivers/net/ethernet/cavium/octeon/Makefile
+++ b/drivers/net/ethernet/cavium/octeon/Makefile
@@ -1,3 +1,4 @@
+# SPDX-License-Identifier: GPL-2.0-only
 #
 # Makefile for the Cavium network device drivers.
 #
diff --git a/drivers/net/ethernet/cavium/octeon/octeon_mgmt.c b/drivers/net/ethernet/cavium/octeon/octeon_mgmt.c
index 2887bcaf6af5..393b9951490a 100644
--- a/drivers/net/ethernet/cavium/octeon/octeon_mgmt.c
+++ b/drivers/net/ethernet/cavium/octeon/octeon_mgmt.c
@@ -28,7 +28,6 @@
 #include <asm/octeon/cvmx-agl-defs.h>
 
 #define DRV_NAME "octeon_mgmt"
-#define DRV_VERSION "2.0"
 #define DRV_DESCRIPTION \
 	"Cavium Networks Octeon MII (management) port Network Driver"
 
@@ -235,6 +234,11 @@ static void octeon_mgmt_rx_fill_ring(struct net_device *netdev)
 
 		/* Put it in the ring.  */
 		p->rx_ring[p->rx_next_fill] = re.d64;
+		/* Make sure there is no reorder of filling the ring and ringing
+		 * the bell
+		 */
+		wmb();
+
 		dma_sync_single_for_device(p->dev, p->rx_ring_handle,
 					   ring_size_to_bytes(OCTEON_MGMT_RX_RING_SIZE),
 					   DMA_BIDIRECTIONAL);
@@ -311,9 +315,9 @@ static void octeon_mgmt_clean_tx_buffers(struct octeon_mgmt *p)
 		netif_wake_queue(p->netdev);
 }
 
-static void octeon_mgmt_clean_tx_tasklet(unsigned long arg)
+static void octeon_mgmt_clean_tx_tasklet(struct tasklet_struct *t)
 {
-	struct octeon_mgmt *p = (struct octeon_mgmt *)arg;
+	struct octeon_mgmt *p = from_tasklet(p, t, tx_clean_tasklet);
 	octeon_mgmt_clean_tx_buffers(p);
 	octeon_mgmt_enable_tx_irq(p);
 }
@@ -544,7 +548,7 @@ struct octeon_mgmt_cam_state {
 };
 
 static void octeon_mgmt_cam_state_add(struct octeon_mgmt_cam_state *cs,
-				      unsigned char *addr)
+				      const unsigned char *addr)
 {
 	int i;
 
@@ -643,13 +647,21 @@ static int octeon_mgmt_set_mac_address(struct net_device *netdev, void *addr)
 static int octeon_mgmt_change_mtu(struct net_device *netdev, int new_mtu)
 {
 	struct octeon_mgmt *p = netdev_priv(netdev);
-	int size_without_fcs = new_mtu + OCTEON_MGMT_RX_HEADROOM;
+	int max_packet = new_mtu + ETH_HLEN + ETH_FCS_LEN;
 
-	netdev->mtu = new_mtu;
+	WRITE_ONCE(netdev->mtu, new_mtu);
 
-	cvmx_write_csr(p->agl + AGL_GMX_RX_FRM_MAX, size_without_fcs);
+	/* HW lifts the limit if the frame is VLAN tagged
+	 * (+4 bytes per each tag, up to two tags)
+	 */
+	cvmx_write_csr(p->agl + AGL_GMX_RX_FRM_MAX, max_packet);
+	/* Set the hardware to truncate packets larger than the MTU. The jabber
+	 * register must be set to a multiple of 8 bytes, so round up. JABBER is
+	 * an unconditional limit, so we need to account for two possible VLAN
+	 * tags.
+	 */
 	cvmx_write_csr(p->agl + AGL_GMX_RX_JABBER,
-		       (size_without_fcs + 7) & 0xfff8);
+		       (max_packet + 7 + VLAN_HLEN * 2) & 0xfff8);
 
 	return 0;
 }
@@ -690,9 +702,6 @@ static int octeon_mgmt_ioctl_hwtstamp(struct net_device *netdev,
 	if (copy_from_user(&config, rq->ifr_data, sizeof(config)))
 		return -EFAULT;
 
-	if (config.flags) /* reserved for future extensions */
-		return -EINVAL;
-
 	/* Check the status of hardware for tiemstamps */
 	if (OCTEON_IS_MODEL(OCTEON_CN6XXX)) {
 		/* Get the current state of the PTP clock */
@@ -705,14 +714,15 @@ static int octeon_mgmt_ioctl_hwtstamp(struct net_device *netdev,
 			u64 clock_comp = (NSEC_PER_SEC << 32) /	octeon_get_io_clock_rate();
 			if (!ptp.s.ptp_en)
 				cvmx_write_csr(CVMX_MIO_PTP_CLOCK_COMP, clock_comp);
-			pr_info("PTP Clock: Using sclk reference at %lld Hz\n",
-				(NSEC_PER_SEC << 32) / clock_comp);
+			netdev_info(netdev,
+				    "PTP Clock using sclk reference @ %lldHz\n",
+				    (NSEC_PER_SEC << 32) / clock_comp);
 		} else {
 			/* The clock is already programmed to use a GPIO */
 			u64 clock_comp = cvmx_read_csr(CVMX_MIO_PTP_CLOCK_COMP);
-			pr_info("PTP Clock: Using GPIO %d at %lld Hz\n",
-				ptp.s.ext_clk_in,
-				(NSEC_PER_SEC << 32) / clock_comp);
+			netdev_info(netdev,
+				    "PTP Clock using GPIO%d @ %lld Hz\n",
+				    ptp.s.ext_clk_in, (NSEC_PER_SEC << 32) / clock_comp);
 		}
 
 		/* Enable the clock if it wasn't done already */
@@ -781,9 +791,7 @@ static int octeon_mgmt_ioctl(struct net_device *netdev,
 	case SIOCSHWTSTAMP:
 		return octeon_mgmt_ioctl_hwtstamp(netdev, rq, cmd);
 	default:
-		if (netdev->phydev)
-			return phy_mii_ioctl(netdev->phydev, rq, cmd);
-		return -EINVAL;
+		return phy_do_ioctl(netdev, rq, cmd);
 	}
 }
 
@@ -926,14 +934,11 @@ static void octeon_mgmt_adjust_link(struct net_device *netdev)
 	spin_unlock_irqrestore(&p->lock, flags);
 
 	if (link_changed != 0) {
-		if (link_changed > 0) {
-			pr_info("%s: Link is up - %d/%s\n", netdev->name,
-				phydev->speed,
-				phydev->duplex == DUPLEX_FULL ?
-				"Full" : "Half");
-		} else {
-			pr_info("%s: Link is down\n", netdev->name);
-		}
+		if (link_changed > 0)
+			netdev_info(netdev, "Link is up - %d/%s\n",
+				    phydev->speed, phydev->duplex == DUPLEX_FULL ? "Full" : "Half");
+		else
+			netdev_info(netdev, "Link is down\n");
 	}
 }
 
@@ -953,7 +958,7 @@ static int octeon_mgmt_init_phy(struct net_device *netdev)
 				PHY_INTERFACE_MODE_MII);
 
 	if (!phydev)
-		return -ENODEV;
+		return -EPROBE_DEFER;
 
 	return 0;
 }
@@ -1074,8 +1079,11 @@ static int octeon_mgmt_open(struct net_device *netdev)
 	/* Set the mode of the interface, RGMII/MII. */
 	if (OCTEON_IS_MODEL(OCTEON_CN6XXX) && netdev->phydev) {
 		union cvmx_agl_prtx_ctl agl_prtx_ctl;
-		int rgmii_mode = (netdev->phydev->supported &
-				  (SUPPORTED_1000baseT_Half | SUPPORTED_1000baseT_Full)) != 0;
+		int rgmii_mode =
+			(linkmode_test_bit(ETHTOOL_LINK_MODE_1000baseT_Half_BIT,
+					   netdev->phydev->supported) |
+			 linkmode_test_bit(ETHTOOL_LINK_MODE_1000baseT_Full_BIT,
+					   netdev->phydev->supported)) != 0;
 
 		agl_prtx_ctl.u64 = cvmx_read_csr(p->agl_prt_ctl);
 		agl_prtx_ctl.s.mode = rgmii_mode ? 0 : 1;
@@ -1208,7 +1216,7 @@ static int octeon_mgmt_open(struct net_device *netdev)
 	 */
 	if (netdev->phydev) {
 		netif_carrier_off(netdev);
-		phy_start_aneg(netdev->phydev);
+		phy_start(netdev->phydev);
 	}
 
 	netif_wake_queue(netdev);
@@ -1236,8 +1244,10 @@ static int octeon_mgmt_stop(struct net_device *netdev)
 	napi_disable(&p->napi);
 	netif_stop_queue(netdev);
 
-	if (netdev->phydev)
+	if (netdev->phydev) {
+		phy_stop(netdev->phydev);
 		phy_disconnect(netdev->phydev);
+	}
 
 	netif_carrier_off(netdev);
 
@@ -1262,12 +1272,13 @@ static int octeon_mgmt_stop(struct net_device *netdev)
 	return 0;
 }
 
-static int octeon_mgmt_xmit(struct sk_buff *skb, struct net_device *netdev)
+static netdev_tx_t
+octeon_mgmt_xmit(struct sk_buff *skb, struct net_device *netdev)
 {
 	struct octeon_mgmt *p = netdev_priv(netdev);
 	union mgmt_port_ring_entry re;
 	unsigned long flags;
-	int rv = NETDEV_TX_BUSY;
+	netdev_tx_t rv = NETDEV_TX_BUSY;
 
 	re.d64 = 0;
 	re.s.tstamp = ((skb_shinfo(skb)->tx_flags & SKBTX_HW_TSTAMP) != 0);
@@ -1331,10 +1342,7 @@ static void octeon_mgmt_poll_controller(struct net_device *netdev)
 static void octeon_mgmt_get_drvinfo(struct net_device *netdev,
 				    struct ethtool_drvinfo *info)
 {
-	strlcpy(info->driver, DRV_NAME, sizeof(info->driver));
-	strlcpy(info->version, DRV_VERSION, sizeof(info->version));
-	strlcpy(info->fw_version, "N/A", sizeof(info->fw_version));
-	strlcpy(info->bus_info, "N/A", sizeof(info->bus_info));
+	strscpy(info->driver, DRV_NAME, sizeof(info->driver));
 }
 
 static int octeon_mgmt_nway_reset(struct net_device *dev)
@@ -1362,7 +1370,7 @@ static const struct net_device_ops octeon_mgmt_ops = {
 	.ndo_start_xmit =		octeon_mgmt_xmit,
 	.ndo_set_rx_mode =		octeon_mgmt_set_rx_filtering,
 	.ndo_set_mac_address =		octeon_mgmt_set_mac_address,
-	.ndo_do_ioctl =			octeon_mgmt_ioctl,
+	.ndo_eth_ioctl =			octeon_mgmt_ioctl,
 	.ndo_change_mtu =		octeon_mgmt_change_mtu,
 #ifdef CONFIG_NET_POLL_CONTROLLER
 	.ndo_poll_controller =		octeon_mgmt_poll_controller,
@@ -1374,7 +1382,6 @@ static int octeon_mgmt_probe(struct platform_device *pdev)
 	struct net_device *netdev;
 	struct octeon_mgmt *p;
 	const __be32 *data;
-	const u8 *mac;
 	struct resource *res_mix;
 	struct resource *res_agl;
 	struct resource *res_agl_prt_ctl;
@@ -1389,8 +1396,8 @@ static int octeon_mgmt_probe(struct platform_device *pdev)
 
 	platform_set_drvdata(pdev, netdev);
 	p = netdev_priv(netdev);
-	netif_napi_add(netdev, &p->napi, octeon_mgmt_napi_poll,
-		       OCTEON_MGMT_NAPI_WEIGHT);
+	netif_napi_add_weight(netdev, &p->napi, octeon_mgmt_napi_poll,
+			      OCTEON_MGMT_NAPI_WEIGHT);
 
 	p->netdev = netdev;
 	p->dev = &pdev->dev;
@@ -1480,8 +1487,8 @@ static int octeon_mgmt_probe(struct platform_device *pdev)
 
 	skb_queue_head_init(&p->tx_list);
 	skb_queue_head_init(&p->rx_list);
-	tasklet_init(&p->tx_clean_tasklet,
-		     octeon_mgmt_clean_tx_tasklet, (unsigned long)p);
+	tasklet_setup(&p->tx_clean_tasklet,
+		      octeon_mgmt_clean_tx_tasklet);
 
 	netdev->priv_flags |= IFF_UNICAST_FLT;
 
@@ -1489,13 +1496,10 @@ static int octeon_mgmt_probe(struct platform_device *pdev)
 	netdev->ethtool_ops = &octeon_mgmt_ethtool_ops;
 
 	netdev->min_mtu = 64 - OCTEON_MGMT_RX_HEADROOM;
-	netdev->max_mtu = 16383 - OCTEON_MGMT_RX_HEADROOM;
-
-	mac = of_get_mac_address(pdev->dev.of_node);
+	netdev->max_mtu = 16383 - OCTEON_MGMT_RX_HEADROOM - VLAN_HLEN;
 
-	if (mac)
-		memcpy(netdev->dev_addr, mac, ETH_ALEN);
-	else
+	result = of_get_ethdev_address(pdev->dev.of_node, netdev);
+	if (result)
 		eth_hw_addr_random(netdev);
 
 	p->phy_np = of_parse_phandle(pdev->dev.of_node, "phy-handle", 0);
@@ -1509,7 +1513,6 @@ static int octeon_mgmt_probe(struct platform_device *pdev)
 	if (result)
 		goto err;
 
-	dev_info(&pdev->dev, "Version " DRV_VERSION "\n");
 	return 0;
 
 err:
@@ -1518,7 +1521,7 @@ err:
 	return result;
 }
 
-static int octeon_mgmt_remove(struct platform_device *pdev)
+static void octeon_mgmt_remove(struct platform_device *pdev)
 {
 	struct net_device *netdev = platform_get_drvdata(pdev);
 	struct octeon_mgmt *p = netdev_priv(netdev);
@@ -1526,7 +1529,6 @@ static int octeon_mgmt_remove(struct platform_device *pdev)
 	unregister_netdev(netdev);
 	of_node_put(p->phy_np);
 	free_netdev(netdev);
-	return 0;
 }
 
 static const struct of_device_id octeon_mgmt_match[] = {
@@ -1546,24 +1548,9 @@ static struct platform_driver octeon_mgmt_driver = {
 	.remove		= octeon_mgmt_remove,
 };
 
-extern void octeon_mdiobus_force_mod_depencency(void);
-
-static int __init octeon_mgmt_mod_init(void)
-{
-	/* Force our mdiobus driver module to be loaded first. */
-	octeon_mdiobus_force_mod_depencency();
-	return platform_driver_register(&octeon_mgmt_driver);
-}
-
-static void __exit octeon_mgmt_mod_exit(void)
-{
-	platform_driver_unregister(&octeon_mgmt_driver);
-}
-
-module_init(octeon_mgmt_mod_init);
-module_exit(octeon_mgmt_mod_exit);
+module_platform_driver(octeon_mgmt_driver);
 
+MODULE_SOFTDEP("pre: mdio-cavium");
 MODULE_DESCRIPTION(DRV_DESCRIPTION);
 MODULE_AUTHOR("David Daney");
 MODULE_LICENSE("GPL");
-MODULE_VERSION(DRV_VERSION);
diff --git a/drivers/net/ethernet/cavium/thunder/Makefile b/drivers/net/ethernet/cavium/thunder/Makefile
index 6b4d4add7353..2fc6142d1634 100644
--- a/drivers/net/ethernet/cavium/thunder/Makefile
+++ b/drivers/net/ethernet/cavium/thunder/Makefile
@@ -1,3 +1,4 @@
+# SPDX-License-Identifier: GPL-2.0
 #
 # Makefile for Cavium's Thunder ethernet device
 #
diff --git a/drivers/net/ethernet/cavium/thunder/nic.h b/drivers/net/ethernet/cavium/thunder/nic.h
index 4a02e618e318..090d6b83982a 100644
--- a/drivers/net/ethernet/cavium/thunder/nic.h
+++ b/drivers/net/ethernet/cavium/thunder/nic.h
@@ -1,9 +1,6 @@
+/* SPDX-License-Identifier: GPL-2.0-only */
 /*
  * Copyright (C) 2015 Cavium, Inc.
- *
- * This program is free software; you can redistribute it and/or modify it
- * under the terms of version 2 of the GNU General Public License
- * as published by the Free Software Foundation.
  */
 
 #ifndef NIC_H
@@ -263,6 +260,19 @@ struct nicvf_drv_stats {
 	struct u64_stats_sync   syncp;
 };
 
+struct cavium_ptp;
+
+struct xcast_addr_list {
+	int              count;
+	u64              mc[];
+};
+
+struct nicvf_work {
+	struct work_struct     work;
+	u8                     mode;
+	struct xcast_addr_list *mc;
+};
+
 struct nicvf {
 	struct nicvf		*pnicvf;
 	struct net_device	*netdev;
@@ -311,6 +321,40 @@ struct nicvf {
 	struct nicvf_pfc	pfc;
 	struct tasklet_struct	qs_err_task;
 	struct work_struct	reset_task;
+	struct nicvf_work       rx_mode_work;
+	/* spinlock to protect workqueue arguments from concurrent access */
+	spinlock_t              rx_mode_wq_lock;
+	/* workqueue for handling kernel ndo_set_rx_mode() calls */
+	struct workqueue_struct *nicvf_rx_mode_wq;
+	/* mutex to protect VF's mailbox contents from concurrent access */
+	struct mutex            rx_mode_mtx;
+	struct delayed_work	link_change_work;
+	/* PTP timestamp */
+	struct cavium_ptp	*ptp_clock;
+	/* Inbound timestamping is on */
+	bool			hw_rx_tstamp;
+	/* When the packet that requires timestamping is sent, hardware inserts
+	 * two entries to the completion queue.  First is the regular
+	 * CQE_TYPE_SEND entry that signals that the packet was sent.
+	 * The second is CQE_TYPE_SEND_PTP that contains the actual timestamp
+	 * for that packet.
+	 * `ptp_skb` is initialized in the handler for the CQE_TYPE_SEND
+	 * entry and is used and zeroed in the handler for the CQE_TYPE_SEND_PTP
+	 * entry.
+	 * So `ptp_skb` is used to hold the pointer to the packet between
+	 * the calls to CQE_TYPE_SEND and CQE_TYPE_SEND_PTP handlers.
+	 */
+	struct sk_buff		*ptp_skb;
+	/* `tx_ptp_skbs` is set when the hardware is sending a packet that
+	 * requires timestamping.  Cavium hardware can not process more than one
+	 * such packet at once so this is set each time the driver submits
+	 * a packet that requires timestamping to the send queue and clears
+	 * each time it receives the entry on the completion queue saying
+	 * that such packet was sent.
+	 * So `tx_ptp_skbs` prevents driver from submitting more than one
+	 * packet that requires timestamping to the hardware for transmitting.
+	 */
+	atomic_t		tx_ptp_skbs;
 
 	/* Interrupt coalescing settings */
 	u32			cq_coalesce_usecs;
@@ -371,8 +415,12 @@ struct nicvf {
 #define	NIC_MBOX_MSG_LOOPBACK		0x16	/* Set interface in loopback */
 #define	NIC_MBOX_MSG_RESET_STAT_COUNTER 0x17	/* Reset statistics counters */
 #define	NIC_MBOX_MSG_PFC		0x18	/* Pause frame control */
+#define	NIC_MBOX_MSG_PTP_CFG		0x19	/* HW packet timestamp */
 #define	NIC_MBOX_MSG_CFG_DONE		0xF0	/* VF configuration done */
 #define	NIC_MBOX_MSG_SHUTDOWN		0xF1	/* VF is being shutdown */
+#define	NIC_MBOX_MSG_RESET_XCAST	0xF2    /* Reset DCAM filtering mode */
+#define	NIC_MBOX_MSG_ADD_MCAST		0xF3    /* Add MAC to DCAM filters */
+#define	NIC_MBOX_MSG_SET_XCAST		0xF4    /* Set MCAST/BCAST RX mode */
 
 struct nic_cfg_msg {
 	u8    msg;
@@ -521,6 +569,17 @@ struct pfc {
 	u8    fc_tx;
 };
 
+struct set_ptp {
+	u8    msg;
+	bool  enable;
+};
+
+struct xcast {
+	u8    msg;
+	u8    mode;
+	u64   mac:48;
+};
+
 /* 128 bit shared memory between PF and each VF */
 union nic_mbx {
 	struct { u8 msg; }	msg;
@@ -540,6 +599,8 @@ union nic_mbx {
 	struct set_loopback	lbk;
 	struct reset_stat_cfg	reset_stat;
 	struct pfc		pfc;
+	struct set_ptp		ptp;
+	struct xcast            xcast;
 };
 
 #define NIC_NODE_ID_MASK	0x03
diff --git a/drivers/net/ethernet/cavium/thunder/nic_main.c b/drivers/net/ethernet/cavium/thunder/nic_main.c
index fb770b0182d3..0ec65ec634df 100644
--- a/drivers/net/ethernet/cavium/thunder/nic_main.c
+++ b/drivers/net/ethernet/cavium/thunder/nic_main.c
@@ -1,9 +1,6 @@
+// SPDX-License-Identifier: GPL-2.0-only
 /*
  * Copyright (C) 2015 Cavium, Inc.
- *
- * This program is free software; you can redistribute it and/or modify it
- * under the terms of version 2 of the GNU General Public License
- * as published by the Free Software Foundation.
  */
 
 #include <linux/module.h>
@@ -18,9 +15,11 @@
 #include "q_struct.h"
 #include "thunder_bgx.h"
 
-#define DRV_NAME	"thunder-nic"
+#define DRV_NAME	"nicpf"
 #define DRV_VERSION	"1.0"
 
+#define NIC_VF_PER_MBX_REG      64
+
 struct hw_info {
 	u8		bgx_cnt;
 	u8		chans_per_lmac;
@@ -55,18 +54,12 @@ struct nicpf {
 #define	NIC_GET_BGX_FROM_VF_LMAC_MAP(map)	((map >> 4) & 0xF)
 #define	NIC_GET_LMAC_FROM_VF_LMAC_MAP(map)	(map & 0xF)
 	u8			*vf_lmac_map;
-	struct delayed_work     dwork;
-	struct workqueue_struct *check_link;
-	u8			*link;
-	u8			*duplex;
-	u32			*speed;
 	u16			cpi_base[MAX_NUM_VFS_SUPPORTED];
 	u16			rssi_base[MAX_NUM_VFS_SUPPORTED];
-	bool			mbx_lock[MAX_NUM_VFS_SUPPORTED];
 
 	/* MSI-X */
 	u8			num_vec;
-	bool			irq_allocated[NIC_PF_MSIX_VECTORS];
+	unsigned int		irq_allocated[NIC_PF_MSIX_VECTORS];
 	char			irq_name[NIC_PF_MSIX_VECTORS][20];
 };
 
@@ -361,17 +354,8 @@ static void nic_set_lmac_vf_mapping(struct nicpf *nic)
 	}
 }
 
-static void nic_free_lmacmem(struct nicpf *nic)
-{
-	kfree(nic->vf_lmac_map);
-	kfree(nic->link);
-	kfree(nic->duplex);
-	kfree(nic->speed);
-}
-
-static int nic_get_hw_info(struct nicpf *nic)
+static void nic_get_hw_info(struct nicpf *nic)
 {
-	u8 max_lmac;
 	u16 sdevid;
 	struct hw_info *hw = nic->hw;
 
@@ -419,54 +403,38 @@ static int nic_get_hw_info(struct nicpf *nic)
 		break;
 	}
 	hw->tl4_cnt = MAX_QUEUES_PER_QSET * pci_sriov_get_totalvfs(nic->pdev);
-
-	/* Allocate memory for LMAC tracking elements */
-	max_lmac = hw->bgx_cnt * MAX_LMAC_PER_BGX;
-	nic->vf_lmac_map = kmalloc_array(max_lmac, sizeof(u8), GFP_KERNEL);
-	if (!nic->vf_lmac_map)
-		goto error;
-	nic->link = kmalloc_array(max_lmac, sizeof(u8), GFP_KERNEL);
-	if (!nic->link)
-		goto error;
-	nic->duplex = kmalloc_array(max_lmac, sizeof(u8), GFP_KERNEL);
-	if (!nic->duplex)
-		goto error;
-	nic->speed = kmalloc_array(max_lmac, sizeof(u32), GFP_KERNEL);
-	if (!nic->speed)
-		goto error;
-	return 0;
-
-error:
-	nic_free_lmacmem(nic);
-	return -ENOMEM;
 }
 
 #define BGX0_BLOCK 8
 #define BGX1_BLOCK 9
 
-static int nic_init_hw(struct nicpf *nic)
+static void nic_init_hw(struct nicpf *nic)
 {
-	int i, err;
+	int i;
 	u64 cqm_cfg;
 
-	/* Get HW capability info */
-	err = nic_get_hw_info(nic);
-	if (err)
-		return err;
-
 	/* Enable NIC HW block */
 	nic_reg_write(nic, NIC_PF_CFG, 0x3);
 
 	/* Enable backpressure */
 	nic_reg_write(nic, NIC_PF_BP_CFG, (1ULL << 6) | 0x03);
 
-	/* TNS and TNS bypass modes are present only on 88xx */
+	/* TNS and TNS bypass modes are present only on 88xx
+	 * Also offset of this CSR has changed in 81xx and 83xx.
+	 */
 	if (nic->pdev->subsystem_device == PCI_SUBSYS_DEVID_88XX_NIC_PF) {
 		/* Disable TNS mode on both interfaces */
 		nic_reg_write(nic, NIC_PF_INTF_0_1_SEND_CFG,
-			      (NIC_TNS_BYPASS_MODE << 7) | BGX0_BLOCK);
+			      (NIC_TNS_BYPASS_MODE << 7) |
+			      BGX0_BLOCK | (1ULL << 16));
 		nic_reg_write(nic, NIC_PF_INTF_0_1_SEND_CFG | (1 << 8),
-			      (NIC_TNS_BYPASS_MODE << 7) | BGX1_BLOCK);
+			      (NIC_TNS_BYPASS_MODE << 7) |
+			      BGX1_BLOCK | (1ULL << 16));
+	} else {
+		/* Configure timestamp generation timeout to 10us */
+		for (i = 0; i < nic->hw->bgx_cnt; i++)
+			nic_reg_write(nic, NIC_PF_INTFX_SEND_CFG | (i << 3),
+				      (1ULL << 16));
 	}
 
 	nic_reg_write(nic, NIC_PF_INTF_0_1_BP_CFG,
@@ -498,8 +466,6 @@ static int nic_init_hw(struct nicpf *nic)
 	cqm_cfg = nic_reg_read(nic, NIC_PF_CQM_CFG);
 	if (cqm_cfg < NICPF_CQM_MIN_DROP_LEVEL)
 		nic_reg_write(nic, NIC_PF_CQM_CFG, NICPF_CQM_MIN_DROP_LEVEL);
-
-	return 0;
 }
 
 /* Channel parse index configuration */
@@ -584,9 +550,6 @@ static void nic_config_cpi(struct nicpf *nic, struct cpi_cfg_msg *cfg)
 static void nic_send_rss_size(struct nicpf *nic, int vf)
 {
 	union nic_mbx mbx = {};
-	u64  *msg;
-
-	msg = (u64 *)&mbx;
 
 	mbx.rss_size.msg = NIC_MBOX_MSG_RSS_SIZE;
 	mbx.rss_size.ind_tbl_size = nic->hw->rss_ind_tbl_size;
@@ -608,7 +571,6 @@ static void nic_config_rss(struct nicpf *nic, struct rss_cfg_msg *cfg)
 	rssi_base = nic->rssi_base[cfg->vf_id] + cfg->tbl_offset;
 
 	rssi = rssi_base;
-	qset = cfg->vf_id;
 
 	for (; rssi < (rssi_base + cfg->tbl_len); rssi++) {
 		u8 svf = cfg->ind_tbl[idx] >> 3;
@@ -920,6 +882,73 @@ static void nic_pause_frame(struct nicpf *nic, int vf, struct pfc *cfg)
 	}
 }
 
+/* Enable or disable HW timestamping by BGX for pkts received on a LMAC */
+static void nic_config_timestamp(struct nicpf *nic, int vf, struct set_ptp *ptp)
+{
+	struct pkind_cfg *pkind;
+	u8 lmac, bgx_idx;
+	u64 pkind_val, pkind_idx;
+
+	if (vf >= nic->num_vf_en)
+		return;
+
+	bgx_idx = NIC_GET_BGX_FROM_VF_LMAC_MAP(nic->vf_lmac_map[vf]);
+	lmac = NIC_GET_LMAC_FROM_VF_LMAC_MAP(nic->vf_lmac_map[vf]);
+
+	pkind_idx = lmac + bgx_idx * MAX_LMAC_PER_BGX;
+	pkind_val = nic_reg_read(nic, NIC_PF_PKIND_0_15_CFG | (pkind_idx << 3));
+	pkind = (struct pkind_cfg *)&pkind_val;
+
+	if (ptp->enable && !pkind->hdr_sl) {
+		/* Skiplen to exclude 8byte timestamp while parsing pkt
+		 * If not configured, will result in L2 errors.
+		 */
+		pkind->hdr_sl = 4;
+		/* Adjust max packet length allowed */
+		pkind->maxlen += (pkind->hdr_sl * 2);
+		bgx_config_timestamping(nic->node, bgx_idx, lmac, true);
+		nic_reg_write(nic, NIC_PF_RX_ETYPE_0_7 | (1 << 3),
+			      (ETYPE_ALG_ENDPARSE << 16) | ETH_P_1588);
+	} else if (!ptp->enable && pkind->hdr_sl) {
+		pkind->maxlen -= (pkind->hdr_sl * 2);
+		pkind->hdr_sl = 0;
+		bgx_config_timestamping(nic->node, bgx_idx, lmac, false);
+		nic_reg_write(nic, NIC_PF_RX_ETYPE_0_7 | (1 << 3),
+			      (ETYPE_ALG_SKIP << 16) | ETH_P_8021Q);
+	}
+
+	nic_reg_write(nic, NIC_PF_PKIND_0_15_CFG | (pkind_idx << 3), pkind_val);
+}
+
+/* Get BGX LMAC link status and update corresponding VF
+ * if there is a change, valid only if internal L2 switch
+ * is not present otherwise VF link is always treated as up
+ */
+static void nic_link_status_get(struct nicpf *nic, u8 vf)
+{
+	union nic_mbx mbx = {};
+	struct bgx_link_status link;
+	u8 bgx, lmac;
+
+	mbx.link_status.msg = NIC_MBOX_MSG_BGX_LINK_CHANGE;
+
+	/* Get BGX, LMAC indices for the VF */
+	bgx = NIC_GET_BGX_FROM_VF_LMAC_MAP(nic->vf_lmac_map[vf]);
+	lmac = NIC_GET_LMAC_FROM_VF_LMAC_MAP(nic->vf_lmac_map[vf]);
+
+	/* Get interface link status */
+	bgx_get_lmac_link_state(nic->node, bgx, lmac, &link);
+
+	/* Send a mbox message to VF with current link status */
+	mbx.link_status.link_up = link.link_up;
+	mbx.link_status.duplex = link.duplex;
+	mbx.link_status.speed = link.speed;
+	mbx.link_status.mac_type = link.mac_type;
+
+	/* reply with link status */
+	nic_send_msg_to_vf(nic, vf, &mbx);
+}
+
 /* Interrupt handler to handle mailbox messages from VFs */
 static void nic_handle_mbx_intr(struct nicpf *nic, int vf)
 {
@@ -932,8 +961,6 @@ static void nic_handle_mbx_intr(struct nicpf *nic, int vf)
 	int i;
 	int ret = 0;
 
-	nic->mbx_lock[vf] = true;
-
 	mbx_addr = nic_get_mbx_addr(vf);
 	mbx_data = (u64 *)&mbx;
 
@@ -948,12 +975,7 @@ static void nic_handle_mbx_intr(struct nicpf *nic, int vf)
 	switch (mbx.msg.msg) {
 	case NIC_MBOX_MSG_READY:
 		nic_mbx_send_ready(nic, vf);
-		if (vf < nic->num_vf_en) {
-			nic->link[vf] = 0;
-			nic->duplex[vf] = 0;
-			nic->speed[vf] = 0;
-		}
-		goto unlock;
+		return;
 	case NIC_MBOX_MSG_QS_CFG:
 		reg_addr = NIC_PF_QSET_0_127_CFG |
 			   (mbx.qs.num << NIC_QS_ID_SHIFT);
@@ -1022,7 +1044,7 @@ static void nic_handle_mbx_intr(struct nicpf *nic, int vf)
 		break;
 	case NIC_MBOX_MSG_RSS_SIZE:
 		nic_send_rss_size(nic, vf);
-		goto unlock;
+		return;
 	case NIC_MBOX_MSG_RSS_CFG:
 	case NIC_MBOX_MSG_RSS_CFG_CONT:
 		nic_config_rss(nic, &mbx.rss_cfg);
@@ -1030,7 +1052,7 @@ static void nic_handle_mbx_intr(struct nicpf *nic, int vf)
 	case NIC_MBOX_MSG_CFG_DONE:
 		/* Last message of VF config msg sequence */
 		nic_enable_vf(nic, vf, true);
-		goto unlock;
+		break;
 	case NIC_MBOX_MSG_SHUTDOWN:
 		/* First msg in VF teardown sequence */
 		if (vf >= nic->num_vf_en)
@@ -1040,19 +1062,19 @@ static void nic_handle_mbx_intr(struct nicpf *nic, int vf)
 		break;
 	case NIC_MBOX_MSG_ALLOC_SQS:
 		nic_alloc_sqs(nic, &mbx.sqs_alloc);
-		goto unlock;
+		return;
 	case NIC_MBOX_MSG_NICVF_PTR:
 		nic->nicvf[vf] = mbx.nicvf.nicvf;
 		break;
 	case NIC_MBOX_MSG_PNICVF_PTR:
 		nic_send_pnicvf(nic, vf);
-		goto unlock;
+		return;
 	case NIC_MBOX_MSG_SNICVF_PTR:
 		nic_send_snicvf(nic, &mbx.nicvf);
-		goto unlock;
+		return;
 	case NIC_MBOX_MSG_BGX_STATS:
 		nic_get_bgx_stats(nic, &mbx.bgx_stats);
-		goto unlock;
+		return;
 	case NIC_MBOX_MSG_LOOPBACK:
 		ret = nic_config_loopback(nic, &mbx.lbk);
 		break;
@@ -1061,7 +1083,51 @@ static void nic_handle_mbx_intr(struct nicpf *nic, int vf)
 		break;
 	case NIC_MBOX_MSG_PFC:
 		nic_pause_frame(nic, vf, &mbx.pfc);
-		goto unlock;
+		return;
+	case NIC_MBOX_MSG_PTP_CFG:
+		nic_config_timestamp(nic, vf, &mbx.ptp);
+		break;
+	case NIC_MBOX_MSG_RESET_XCAST:
+		if (vf >= nic->num_vf_en) {
+			ret = -1; /* NACK */
+			break;
+		}
+		bgx = NIC_GET_BGX_FROM_VF_LMAC_MAP(nic->vf_lmac_map[vf]);
+		lmac = NIC_GET_LMAC_FROM_VF_LMAC_MAP(nic->vf_lmac_map[vf]);
+		bgx_reset_xcast_mode(nic->node, bgx, lmac,
+				     vf < NIC_VF_PER_MBX_REG ? vf :
+				     vf - NIC_VF_PER_MBX_REG);
+		break;
+
+	case NIC_MBOX_MSG_ADD_MCAST:
+		if (vf >= nic->num_vf_en) {
+			ret = -1; /* NACK */
+			break;
+		}
+		bgx = NIC_GET_BGX_FROM_VF_LMAC_MAP(nic->vf_lmac_map[vf]);
+		lmac = NIC_GET_LMAC_FROM_VF_LMAC_MAP(nic->vf_lmac_map[vf]);
+		bgx_set_dmac_cam_filter(nic->node, bgx, lmac,
+					mbx.xcast.mac,
+					vf < NIC_VF_PER_MBX_REG ? vf :
+					vf - NIC_VF_PER_MBX_REG);
+		break;
+
+	case NIC_MBOX_MSG_SET_XCAST:
+		if (vf >= nic->num_vf_en) {
+			ret = -1; /* NACK */
+			break;
+		}
+		bgx = NIC_GET_BGX_FROM_VF_LMAC_MAP(nic->vf_lmac_map[vf]);
+		lmac = NIC_GET_LMAC_FROM_VF_LMAC_MAP(nic->vf_lmac_map[vf]);
+		bgx_set_xcast_mode(nic->node, bgx, lmac, mbx.xcast.mode);
+		break;
+	case NIC_MBOX_MSG_BGX_LINK_CHANGE:
+		if (vf >= nic->num_vf_en) {
+			ret = -1; /* NACK */
+			break;
+		}
+		nic_link_status_get(nic, vf);
+		return;
 	default:
 		dev_err(&nic->pdev->dev,
 			"Invalid msg from VF%d, msg 0x%x\n", vf, mbx.msg.msg);
@@ -1075,8 +1141,6 @@ static void nic_handle_mbx_intr(struct nicpf *nic, int vf)
 			mbx.msg.msg, vf);
 		nic_mbx_send_nack(nic, vf);
 	}
-unlock:
-	nic->mbx_lock[vf] = false;
 }
 
 static irqreturn_t nic_mbx_intr_handler(int irq, void *nic_irq)
@@ -1084,21 +1148,22 @@ static irqreturn_t nic_mbx_intr_handler(int irq, void *nic_irq)
 	struct nicpf *nic = (struct nicpf *)nic_irq;
 	int mbx;
 	u64 intr;
-	u8  vf, vf_per_mbx_reg = 64;
+	u8  vf;
 
-	if (irq == pci_irq_vector(nic->pdev, NIC_PF_INTR_ID_MBOX0))
+	if (irq == nic->irq_allocated[NIC_PF_INTR_ID_MBOX0])
 		mbx = 0;
 	else
 		mbx = 1;
 
 	intr = nic_reg_read(nic, NIC_PF_MAILBOX_INT + (mbx << 3));
 	dev_dbg(&nic->pdev->dev, "PF interrupt Mbox%d 0x%llx\n", mbx, intr);
-	for (vf = 0; vf < vf_per_mbx_reg; vf++) {
+	for (vf = 0; vf < NIC_VF_PER_MBX_REG; vf++) {
 		if (intr & (1ULL << vf)) {
 			dev_dbg(&nic->pdev->dev, "Intr from VF %d\n",
-				vf + (mbx * vf_per_mbx_reg));
+				vf + (mbx * NIC_VF_PER_MBX_REG));
 
-			nic_handle_mbx_intr(nic, vf + (mbx * vf_per_mbx_reg));
+			nic_handle_mbx_intr(nic, vf +
+					    (mbx * NIC_VF_PER_MBX_REG));
 			nic_clear_mbx_intr(nic, vf, mbx);
 		}
 	}
@@ -1111,14 +1176,14 @@ static void nic_free_all_interrupts(struct nicpf *nic)
 
 	for (irq = 0; irq < nic->num_vec; irq++) {
 		if (nic->irq_allocated[irq])
-			free_irq(pci_irq_vector(nic->pdev, irq), nic);
-		nic->irq_allocated[irq] = false;
+			free_irq(nic->irq_allocated[irq], nic);
+		nic->irq_allocated[irq] = 0;
 	}
 }
 
 static int nic_register_interrupts(struct nicpf *nic)
 {
-	int i, ret;
+	int i, ret, irq;
 	nic->num_vec = pci_msix_vec_count(nic->pdev);
 
 	/* Enable MSI-X */
@@ -1128,7 +1193,7 @@ static int nic_register_interrupts(struct nicpf *nic)
 		dev_err(&nic->pdev->dev,
 			"Request for #%d msix vectors failed, returned %d\n",
 			   nic->num_vec, ret);
-		return 1;
+		return ret;
 	}
 
 	/* Register mailbox interrupt handler */
@@ -1136,13 +1201,13 @@ static int nic_register_interrupts(struct nicpf *nic)
 		sprintf(nic->irq_name[i],
 			"NICPF Mbox%d", (i - NIC_PF_INTR_ID_MBOX0));
 
-		ret = request_irq(pci_irq_vector(nic->pdev, i),
-				  nic_mbx_intr_handler, 0,
+		irq = pci_irq_vector(nic->pdev, i);
+		ret = request_irq(irq, nic_mbx_intr_handler, 0,
 				  nic->irq_name[i], nic);
 		if (ret)
 			goto fail;
 
-		nic->irq_allocated[i] = true;
+		nic->irq_allocated[i] = irq;
 	}
 
 	/* Enable mailbox interrupt */
@@ -1223,56 +1288,11 @@ static int nic_sriov_init(struct pci_dev *pdev, struct nicpf *nic)
 	return 0;
 }
 
-/* Poll for BGX LMAC link status and update corresponding VF
- * if there is a change, valid only if internal L2 switch
- * is not present otherwise VF link is always treated as up
- */
-static void nic_poll_for_link(struct work_struct *work)
-{
-	union nic_mbx mbx = {};
-	struct nicpf *nic;
-	struct bgx_link_status link;
-	u8 vf, bgx, lmac;
-
-	nic = container_of(work, struct nicpf, dwork.work);
-
-	mbx.link_status.msg = NIC_MBOX_MSG_BGX_LINK_CHANGE;
-
-	for (vf = 0; vf < nic->num_vf_en; vf++) {
-		/* Poll only if VF is UP */
-		if (!nic->vf_enabled[vf])
-			continue;
-
-		/* Get BGX, LMAC indices for the VF */
-		bgx = NIC_GET_BGX_FROM_VF_LMAC_MAP(nic->vf_lmac_map[vf]);
-		lmac = NIC_GET_LMAC_FROM_VF_LMAC_MAP(nic->vf_lmac_map[vf]);
-		/* Get interface link status */
-		bgx_get_lmac_link_state(nic->node, bgx, lmac, &link);
-
-		/* Inform VF only if link status changed */
-		if (nic->link[vf] == link.link_up)
-			continue;
-
-		if (!nic->mbx_lock[vf]) {
-			nic->link[vf] = link.link_up;
-			nic->duplex[vf] = link.duplex;
-			nic->speed[vf] = link.speed;
-
-			/* Send a mbox message to VF with current link status */
-			mbx.link_status.link_up = link.link_up;
-			mbx.link_status.duplex = link.duplex;
-			mbx.link_status.speed = link.speed;
-			mbx.link_status.mac_type = link.mac_type;
-			nic_send_msg_to_vf(nic, vf, &mbx);
-		}
-	}
-	queue_delayed_work(nic->check_link, &nic->dwork, HZ * 2);
-}
-
 static int nic_probe(struct pci_dev *pdev, const struct pci_device_id *ent)
 {
 	struct device *dev = &pdev->dev;
 	struct nicpf *nic;
+	u8     max_lmac;
 	int    err;
 
 	BUILD_BUG_ON(sizeof(union nic_mbx) > 16);
@@ -1282,10 +1302,8 @@ static int nic_probe(struct pci_dev *pdev, const struct pci_device_id *ent)
 		return -ENOMEM;
 
 	nic->hw = devm_kzalloc(dev, sizeof(struct hw_info), GFP_KERNEL);
-	if (!nic->hw) {
-		devm_kfree(dev, nic);
+	if (!nic->hw)
 		return -ENOMEM;
-	}
 
 	pci_set_drvdata(pdev, nic);
 
@@ -1293,9 +1311,8 @@ static int nic_probe(struct pci_dev *pdev, const struct pci_device_id *ent)
 
 	err = pci_enable_device(pdev);
 	if (err) {
-		dev_err(dev, "Failed to enable PCI device\n");
 		pci_set_drvdata(pdev, NULL);
-		return err;
+		return dev_err_probe(dev, err, "Failed to enable PCI device\n");
 	}
 
 	err = pci_request_regions(pdev, DRV_NAME);
@@ -1304,18 +1321,12 @@ static int nic_probe(struct pci_dev *pdev, const struct pci_device_id *ent)
 		goto err_disable_device;
 	}
 
-	err = pci_set_dma_mask(pdev, DMA_BIT_MASK(48));
+	err = dma_set_mask_and_coherent(&pdev->dev, DMA_BIT_MASK(48));
 	if (err) {
 		dev_err(dev, "Unable to get usable DMA configuration\n");
 		goto err_release_regions;
 	}
 
-	err = pci_set_consistent_dma_mask(pdev, DMA_BIT_MASK(48));
-	if (err) {
-		dev_err(dev, "Unable to get 48-bit DMA for consistent allocations\n");
-		goto err_release_regions;
-	}
-
 	/* MAP PF's configuration registers */
 	nic->reg_base = pcim_iomap(pdev, PCI_CFG_REG_BAR_NUM, 0);
 	if (!nic->reg_base) {
@@ -1326,11 +1337,21 @@ static int nic_probe(struct pci_dev *pdev, const struct pci_device_id *ent)
 
 	nic->node = nic_get_node_id(pdev);
 
-	/* Initialize hardware */
-	err = nic_init_hw(nic);
-	if (err)
+	/* Get HW capability info */
+	nic_get_hw_info(nic);
+
+	/* Allocate memory for LMAC tracking elements */
+	err = -ENOMEM;
+	max_lmac = nic->hw->bgx_cnt * MAX_LMAC_PER_BGX;
+
+	nic->vf_lmac_map = devm_kmalloc_array(dev, max_lmac, sizeof(u8),
+					      GFP_KERNEL);
+	if (!nic->vf_lmac_map)
 		goto err_release_regions;
 
+	/* Initialize hardware */
+	nic_init_hw(nic);
+
 	nic_set_lmac_vf_mapping(nic);
 
 	/* Register interrupts */
@@ -1343,30 +1364,13 @@ static int nic_probe(struct pci_dev *pdev, const struct pci_device_id *ent)
 	if (err)
 		goto err_unregister_interrupts;
 
-	/* Register a physical link status poll fn() */
-	nic->check_link = alloc_workqueue("check_link_status",
-					  WQ_UNBOUND | WQ_MEM_RECLAIM, 1);
-	if (!nic->check_link) {
-		err = -ENOMEM;
-		goto err_disable_sriov;
-	}
-
-	INIT_DELAYED_WORK(&nic->dwork, nic_poll_for_link);
-	queue_delayed_work(nic->check_link, &nic->dwork, 0);
-
 	return 0;
 
-err_disable_sriov:
-	if (nic->flags & NIC_SRIOV_ENABLED)
-		pci_disable_sriov(pdev);
 err_unregister_interrupts:
 	nic_unregister_interrupts(nic);
 err_release_regions:
 	pci_release_regions(pdev);
 err_disable_device:
-	nic_free_lmacmem(nic);
-	devm_kfree(dev, nic->hw);
-	devm_kfree(dev, nic);
 	pci_disable_device(pdev);
 	pci_set_drvdata(pdev, NULL);
 	return err;
@@ -1376,22 +1380,15 @@ static void nic_remove(struct pci_dev *pdev)
 {
 	struct nicpf *nic = pci_get_drvdata(pdev);
 
+	if (!nic)
+		return;
+
 	if (nic->flags & NIC_SRIOV_ENABLED)
 		pci_disable_sriov(pdev);
 
-	if (nic->check_link) {
-		/* Destroy work Queue */
-		cancel_delayed_work_sync(&nic->dwork);
-		destroy_workqueue(nic->check_link);
-	}
-
 	nic_unregister_interrupts(nic);
 	pci_release_regions(pdev);
 
-	nic_free_lmacmem(nic);
-	devm_kfree(&pdev->dev, nic->hw);
-	devm_kfree(&pdev->dev, nic);
-
 	pci_disable_device(pdev);
 	pci_set_drvdata(pdev, NULL);
 }
diff --git a/drivers/net/ethernet/cavium/thunder/nic_reg.h b/drivers/net/ethernet/cavium/thunder/nic_reg.h
index 80d46337cf29..b3bd24febe75 100644
--- a/drivers/net/ethernet/cavium/thunder/nic_reg.h
+++ b/drivers/net/ethernet/cavium/thunder/nic_reg.h
@@ -1,9 +1,6 @@
+/* SPDX-License-Identifier: GPL-2.0-only */
 /*
  * Copyright (C) 2015 Cavium, Inc.
- *
- * This program is free software; you can redistribute it and/or modify it
- * under the terms of version 2 of the GNU General Public License
- * as published by the Free Software Foundation.
  */
 
 #ifndef NIC_REG_H
@@ -99,6 +96,7 @@
 #define   NIC_PF_ECC3_DBE_INT_W1S		(0x2708)
 #define   NIC_PF_ECC3_DBE_ENA_W1C		(0x2710)
 #define   NIC_PF_ECC3_DBE_ENA_W1S		(0x2718)
+#define   NIC_PF_INTFX_SEND_CFG			(0x4000)
 #define   NIC_PF_MCAM_0_191_ENA			(0x100000)
 #define   NIC_PF_MCAM_0_191_M_0_5_DATA		(0x110000)
 #define   NIC_PF_MCAM_CTRL			(0x120000)
diff --git a/drivers/net/ethernet/cavium/thunder/nicvf_ethtool.c b/drivers/net/ethernet/cavium/thunder/nicvf_ethtool.c
index b9ece9cbf98b..fc6053414b7d 100644
--- a/drivers/net/ethernet/cavium/thunder/nicvf_ethtool.c
+++ b/drivers/net/ethernet/cavium/thunder/nicvf_ethtool.c
@@ -1,23 +1,22 @@
+// SPDX-License-Identifier: GPL-2.0-only
 /*
  * Copyright (C) 2015 Cavium, Inc.
- *
- * This program is free software; you can redistribute it and/or modify it
- * under the terms of version 2 of the GNU General Public License
- * as published by the Free Software Foundation.
  */
 
 /* ETHTOOL Support for VNIC_VF Device*/
 
+#include <linux/ethtool.h>
 #include <linux/pci.h>
+#include <linux/net_tstamp.h>
 
 #include "nic_reg.h"
 #include "nic.h"
 #include "nicvf_queues.h"
 #include "q_struct.h"
 #include "thunder_bgx.h"
+#include "../common/cavium_ptp.h"
 
-#define DRV_NAME	"thunder-nicvf"
-#define DRV_VERSION     "1.0"
+#define DRV_NAME	"nicvf"
 
 struct nicvf_stat {
 	char name[ETH_GSTRING_LEN];
@@ -192,9 +191,8 @@ static void nicvf_get_drvinfo(struct net_device *netdev,
 {
 	struct nicvf *nic = netdev_priv(netdev);
 
-	strlcpy(info->driver, DRV_NAME, sizeof(info->driver));
-	strlcpy(info->version, DRV_VERSION, sizeof(info->version));
-	strlcpy(info->bus_info, pci_name(nic->pdev), sizeof(info->bus_info));
+	strscpy(info->driver, DRV_NAME, sizeof(info->driver));
+	strscpy(info->bus_info, pci_name(nic->pdev), sizeof(info->bus_info));
 }
 
 static u32 nicvf_get_msglevel(struct net_device *netdev)
@@ -458,7 +456,9 @@ static void nicvf_get_regs(struct net_device *dev,
 }
 
 static int nicvf_get_coalesce(struct net_device *netdev,
-			      struct ethtool_coalesce *cmd)
+			      struct ethtool_coalesce *cmd,
+			      struct kernel_ethtool_coalesce *kernel_coal,
+			      struct netlink_ext_ack *extack)
 {
 	struct nicvf *nic = netdev_priv(netdev);
 
@@ -467,7 +467,9 @@ static int nicvf_get_coalesce(struct net_device *netdev,
 }
 
 static void nicvf_get_ringparam(struct net_device *netdev,
-				struct ethtool_ringparam *ring)
+				struct ethtool_ringparam *ring,
+				struct kernel_ethtool_ringparam *kernel_ring,
+				struct netlink_ext_ack *extack)
 {
 	struct nicvf *nic = netdev_priv(netdev);
 	struct queue_set *qs = nic->qs;
@@ -479,7 +481,9 @@ static void nicvf_get_ringparam(struct net_device *netdev,
 }
 
 static int nicvf_set_ringparam(struct net_device *netdev,
-			       struct ethtool_ringparam *ring)
+			       struct ethtool_ringparam *ring,
+			       struct kernel_ethtool_ringparam *kernel_ring,
+			       struct netlink_ext_ack *extack)
 {
 	struct nicvf *nic = netdev_priv(netdev);
 	struct queue_set *qs = nic->qs;
@@ -512,8 +516,8 @@ static int nicvf_set_ringparam(struct net_device *netdev,
 	return 0;
 }
 
-static int nicvf_get_rss_hash_opts(struct nicvf *nic,
-				   struct ethtool_rxnfc *info)
+static int nicvf_get_rxfh_fields(struct net_device *dev,
+				 struct ethtool_rxfh_fields *info)
 {
 	info->data = 0;
 
@@ -525,6 +529,7 @@ static int nicvf_get_rss_hash_opts(struct nicvf *nic,
 	case SCTP_V4_FLOW:
 	case SCTP_V6_FLOW:
 		info->data |= RXH_L4_B_0_1 | RXH_L4_B_2_3;
+		fallthrough;
 	case IPV4_FLOW:
 	case IPV6_FLOW:
 		info->data |= RXH_IP_SRC | RXH_IP_DST;
@@ -547,25 +552,28 @@ static int nicvf_get_rxnfc(struct net_device *dev,
 		info->data = nic->rx_queues;
 		ret = 0;
 		break;
-	case ETHTOOL_GRXFH:
-		return nicvf_get_rss_hash_opts(nic, info);
 	default:
 		break;
 	}
 	return ret;
 }
 
-static int nicvf_set_rss_hash_opts(struct nicvf *nic,
-				   struct ethtool_rxnfc *info)
+static int nicvf_set_rxfh_fields(struct net_device *dev,
+				 const struct ethtool_rxfh_fields *info,
+				 struct netlink_ext_ack *extack)
 {
-	struct nicvf_rss_info *rss = &nic->rss_info;
-	u64 rss_cfg = nicvf_reg_read(nic, NIC_VNIC_RSS_CFG);
+	struct nicvf *nic = netdev_priv(dev);
+	struct nicvf_rss_info *rss;
+	u64 rss_cfg;
+
+	rss = &nic->rss_info;
+	rss_cfg = nicvf_reg_read(nic, NIC_VNIC_RSS_CFG);
 
 	if (!rss->enable)
 		netdev_err(nic->netdev,
 			   "RSS is disabled, hash cannot be set\n");
 
-	netdev_info(nic->netdev, "Set RSS flow type = %d, data = %lld\n",
+	netdev_info(nic->netdev, "Set RSS flow type = %d, data = %u\n",
 		    info->flow_type, info->data);
 
 	if (!(info->data & RXH_IP_SRC) || !(info->data & RXH_IP_DST))
@@ -623,19 +631,6 @@ static int nicvf_set_rss_hash_opts(struct nicvf *nic,
 	return 0;
 }
 
-static int nicvf_set_rxnfc(struct net_device *dev, struct ethtool_rxnfc *info)
-{
-	struct nicvf *nic = netdev_priv(dev);
-
-	switch (info->cmd) {
-	case ETHTOOL_SRXFH:
-		return nicvf_set_rss_hash_opts(nic, info);
-	default:
-		break;
-	}
-	return -EOPNOTSUPP;
-}
-
 static u32 nicvf_get_rxfh_key_size(struct net_device *netdev)
 {
 	return RSS_HASH_KEY_SIZE * sizeof(u64);
@@ -648,35 +643,36 @@ static u32 nicvf_get_rxfh_indir_size(struct net_device *dev)
 	return nic->rss_info.rss_size;
 }
 
-static int nicvf_get_rxfh(struct net_device *dev, u32 *indir, u8 *hkey,
-			  u8 *hfunc)
+static int nicvf_get_rxfh(struct net_device *dev,
+			  struct ethtool_rxfh_param *rxfh)
 {
 	struct nicvf *nic = netdev_priv(dev);
 	struct nicvf_rss_info *rss = &nic->rss_info;
 	int idx;
 
-	if (indir) {
+	if (rxfh->indir) {
 		for (idx = 0; idx < rss->rss_size; idx++)
-			indir[idx] = rss->ind_tbl[idx];
+			rxfh->indir[idx] = rss->ind_tbl[idx];
 	}
 
-	if (hkey)
-		memcpy(hkey, rss->key, RSS_HASH_KEY_SIZE * sizeof(u64));
+	if (rxfh->key)
+		memcpy(rxfh->key, rss->key, RSS_HASH_KEY_SIZE * sizeof(u64));
 
-	if (hfunc)
-		*hfunc = ETH_RSS_HASH_TOP;
+	rxfh->hfunc = ETH_RSS_HASH_TOP;
 
 	return 0;
 }
 
-static int nicvf_set_rxfh(struct net_device *dev, const u32 *indir,
-			  const u8 *hkey, const u8 hfunc)
+static int nicvf_set_rxfh(struct net_device *dev,
+			  struct ethtool_rxfh_param *rxfh,
+			  struct netlink_ext_ack *extack)
 {
 	struct nicvf *nic = netdev_priv(dev);
 	struct nicvf_rss_info *rss = &nic->rss_info;
 	int idx;
 
-	if (hfunc != ETH_RSS_HASH_NO_CHANGE && hfunc != ETH_RSS_HASH_TOP)
+	if (rxfh->hfunc != ETH_RSS_HASH_NO_CHANGE &&
+	    rxfh->hfunc != ETH_RSS_HASH_TOP)
 		return -EOPNOTSUPP;
 
 	if (!rss->enable) {
@@ -685,13 +681,13 @@ static int nicvf_set_rxfh(struct net_device *dev, const u32 *indir,
 		return -EIO;
 	}
 
-	if (indir) {
+	if (rxfh->indir) {
 		for (idx = 0; idx < rss->rss_size; idx++)
-			rss->ind_tbl[idx] = indir[idx];
+			rss->ind_tbl[idx] = rxfh->indir[idx];
 	}
 
-	if (hkey) {
-		memcpy(rss->key, hkey, RSS_HASH_KEY_SIZE * sizeof(u64));
+	if (rxfh->key) {
+		memcpy(rss->key, rxfh->key, RSS_HASH_KEY_SIZE * sizeof(u64));
 		nicvf_set_rss_key(nic);
 	}
 
@@ -730,12 +726,17 @@ static int nicvf_set_channels(struct net_device *dev,
 	if (channel->tx_count > nic->max_queues)
 		return -EINVAL;
 
-	if (nic->xdp_prog &&
-	    ((channel->tx_count + channel->rx_count) > nic->max_queues)) {
-		netdev_err(nic->netdev,
-			   "XDP mode, RXQs + TXQs > Max %d\n",
-			   nic->max_queues);
-		return -EINVAL;
+	if (channel->tx_count + channel->rx_count > nic->max_queues) {
+		if (nic->xdp_prog) {
+			netdev_err(nic->netdev,
+				   "XDP mode, RXQs + TXQs > Max %d\n",
+				   nic->max_queues);
+			return -EINVAL;
+		}
+
+		xdp_clear_features_flag(nic->netdev);
+	} else if (!pass1_silicon(nic->pdev)) {
+		xdp_set_features_flag(dev, NETDEV_XDP_ACT_BASIC);
 	}
 
 	if (if_up)
@@ -824,6 +825,29 @@ static int nicvf_set_pauseparam(struct net_device *dev,
 	return 0;
 }
 
+static int nicvf_get_ts_info(struct net_device *netdev,
+			     struct kernel_ethtool_ts_info *info)
+{
+	struct nicvf *nic = netdev_priv(netdev);
+
+	if (!nic->ptp_clock)
+		return ethtool_op_get_ts_info(netdev, info);
+
+	info->so_timestamping = SOF_TIMESTAMPING_TX_SOFTWARE |
+				SOF_TIMESTAMPING_TX_HARDWARE |
+				SOF_TIMESTAMPING_RX_HARDWARE |
+				SOF_TIMESTAMPING_RAW_HARDWARE;
+
+	info->phc_index = cavium_ptp_clock_index(nic->ptp_clock);
+
+	info->tx_types = (1 << HWTSTAMP_TX_OFF) | (1 << HWTSTAMP_TX_ON);
+
+	info->rx_filters = (1 << HWTSTAMP_FILTER_NONE) |
+			   (1 << HWTSTAMP_FILTER_ALL);
+
+	return 0;
+}
+
 static const struct ethtool_ops nicvf_ethtool_ops = {
 	.get_link		= nicvf_get_link,
 	.get_drvinfo		= nicvf_get_drvinfo,
@@ -838,16 +862,17 @@ static const struct ethtool_ops nicvf_ethtool_ops = {
 	.get_ringparam		= nicvf_get_ringparam,
 	.set_ringparam		= nicvf_set_ringparam,
 	.get_rxnfc		= nicvf_get_rxnfc,
-	.set_rxnfc		= nicvf_set_rxnfc,
 	.get_rxfh_key_size	= nicvf_get_rxfh_key_size,
 	.get_rxfh_indir_size	= nicvf_get_rxfh_indir_size,
 	.get_rxfh		= nicvf_get_rxfh,
 	.set_rxfh		= nicvf_set_rxfh,
+	.get_rxfh_fields	= nicvf_get_rxfh_fields,
+	.set_rxfh_fields	= nicvf_set_rxfh_fields,
 	.get_channels		= nicvf_get_channels,
 	.set_channels		= nicvf_set_channels,
 	.get_pauseparam         = nicvf_get_pauseparam,
 	.set_pauseparam         = nicvf_set_pauseparam,
-	.get_ts_info		= ethtool_op_get_ts_info,
+	.get_ts_info		= nicvf_get_ts_info,
 	.get_link_ksettings	= nicvf_get_link_ksettings,
 };
 
diff --git a/drivers/net/ethernet/cavium/thunder/nicvf_main.c b/drivers/net/ethernet/cavium/thunder/nicvf_main.c
index d6477af88085..1be2dc40a1a6 100644
--- a/drivers/net/ethernet/cavium/thunder/nicvf_main.c
+++ b/drivers/net/ethernet/cavium/thunder/nicvf_main.c
@@ -1,9 +1,6 @@
+// SPDX-License-Identifier: GPL-2.0-only
 /*
  * Copyright (C) 2015 Cavium, Inc.
- *
- * This program is free software; you can redistribute it and/or modify it
- * under the terms of version 2 of the GNU General Public License
- * as published by the Free Software Foundation.
  */
 
 #include <linux/module.h>
@@ -20,15 +17,25 @@
 #include <linux/bpf.h>
 #include <linux/bpf_trace.h>
 #include <linux/filter.h>
+#include <linux/net_tstamp.h>
+#include <linux/workqueue.h>
 
 #include "nic_reg.h"
 #include "nic.h"
 #include "nicvf_queues.h"
 #include "thunder_bgx.h"
+#include "../common/cavium_ptp.h"
 
-#define DRV_NAME	"thunder-nicvf"
+#define DRV_NAME	"nicvf"
 #define DRV_VERSION	"1.0"
 
+/* NOTE: Packets bigger than 1530 are split across multiple pages and XDP needs
+ * the buffer to be contiguous. Allow XDP to be set up only if we don't exceed
+ * this value, keeping headroom for the 14 byte Ethernet header and two
+ * VLAN tags (for QinQ)
+ */
+#define MAX_XDP_MTU	(1530 - ETH_HLEN - VLAN_HLEN * 2)
+
 /* Supported devices */
 static const struct pci_device_id nicvf_id_table[] = {
 	{ PCI_DEVICE_SUB(PCI_VENDOR_ID_CAVIUM,
@@ -61,7 +68,7 @@ module_param(debug, int, 0644);
 MODULE_PARM_DESC(debug, "Debug message level bitmap");
 
 static int cpi_alg = CPI_ALG_NONE;
-module_param(cpi_alg, int, S_IRUGO);
+module_param(cpi_alg, int, 0444);
 MODULE_PARM_DESC(cpi_alg,
 		 "PFC algorithm (0=none, 1=VLAN, 2=VLAN16, 3=IP Diffserv)");
 
@@ -119,34 +126,39 @@ static void nicvf_write_to_mbx(struct nicvf *nic, union nic_mbx *mbx)
 
 int nicvf_send_msg_to_pf(struct nicvf *nic, union nic_mbx *mbx)
 {
-	int timeout = NIC_MBOX_MSG_TIMEOUT;
-	int sleep = 10;
+	unsigned long timeout;
+	int ret = 0;
+
+	mutex_lock(&nic->rx_mode_mtx);
 
 	nic->pf_acked = false;
 	nic->pf_nacked = false;
 
 	nicvf_write_to_mbx(nic, mbx);
 
+	timeout = jiffies + msecs_to_jiffies(NIC_MBOX_MSG_TIMEOUT);
 	/* Wait for previous message to be acked, timeout 2sec */
 	while (!nic->pf_acked) {
 		if (nic->pf_nacked) {
 			netdev_err(nic->netdev,
 				   "PF NACK to mbox msg 0x%02x from VF%d\n",
 				   (mbx->msg.msg & 0xFF), nic->vf_id);
-			return -EINVAL;
+			ret = -EINVAL;
+			break;
 		}
-		msleep(sleep);
+		usleep_range(8000, 10000);
 		if (nic->pf_acked)
 			break;
-		timeout -= sleep;
-		if (!timeout) {
+		if (time_after(jiffies, timeout)) {
 			netdev_err(nic->netdev,
 				   "PF didn't ACK to mbox msg 0x%02x from VF%d\n",
 				   (mbx->msg.msg & 0xFF), nic->vf_id);
-			return -EBUSY;
+			ret = -EBUSY;
+			break;
 		}
 	}
-	return 0;
+	mutex_unlock(&nic->rx_mode_mtx);
+	return ret;
 }
 
 /* Checks if VF is able to comminicate with PF
@@ -166,6 +178,17 @@ static int nicvf_check_pf_ready(struct nicvf *nic)
 	return 1;
 }
 
+static void nicvf_send_cfg_done(struct nicvf *nic)
+{
+	union nic_mbx mbx = {};
+
+	mbx.msg.msg = NIC_MBOX_MSG_CFG_DONE;
+	if (nicvf_send_msg_to_pf(nic, &mbx)) {
+		netdev_err(nic->netdev,
+			   "PF didn't respond to CFG DONE msg\n");
+	}
+}
+
 static void nicvf_read_bgx_stats(struct nicvf *nic, struct bgx_stats_msg *bgx)
 {
 	if (bgx->rx)
@@ -198,8 +221,7 @@ static void  nicvf_handle_mbx_intr(struct nicvf *nic)
 		nic->tns_mode = mbx.nic_cfg.tns_mode & 0x7F;
 		nic->node = mbx.nic_cfg.node_id;
 		if (!nic->set_mac_pending)
-			ether_addr_copy(nic->netdev->dev_addr,
-					mbx.nic_cfg.mac_addr);
+			eth_hw_addr_set(nic->netdev, mbx.nic_cfg.mac_addr);
 		nic->sqs_mode = mbx.nic_cfg.sqs_mode;
 		nic->loopback_supported = mbx.nic_cfg.loopback_supported;
 		nic->link_up = false;
@@ -222,22 +244,24 @@ static void  nicvf_handle_mbx_intr(struct nicvf *nic)
 		break;
 	case NIC_MBOX_MSG_BGX_LINK_CHANGE:
 		nic->pf_acked = true;
-		nic->link_up = mbx.link_status.link_up;
-		nic->duplex = mbx.link_status.duplex;
-		nic->speed = mbx.link_status.speed;
-		nic->mac_type = mbx.link_status.mac_type;
-		if (nic->link_up) {
-			netdev_info(nic->netdev, "%s: Link is Up %d Mbps %s\n",
-				    nic->netdev->name, nic->speed,
-				    nic->duplex == DUPLEX_FULL ?
-				"Full duplex" : "Half duplex");
-			netif_carrier_on(nic->netdev);
-			netif_tx_start_all_queues(nic->netdev);
-		} else {
-			netdev_info(nic->netdev, "%s: Link is Down\n",
-				    nic->netdev->name);
-			netif_carrier_off(nic->netdev);
-			netif_tx_stop_all_queues(nic->netdev);
+		if (nic->link_up != mbx.link_status.link_up) {
+			nic->link_up = mbx.link_status.link_up;
+			nic->duplex = mbx.link_status.duplex;
+			nic->speed = mbx.link_status.speed;
+			nic->mac_type = mbx.link_status.mac_type;
+			if (nic->link_up) {
+				netdev_info(nic->netdev,
+					    "Link is Up %d Mbps %s duplex\n",
+					    nic->speed,
+					    nic->duplex == DUPLEX_FULL ?
+					    "Full" : "Half");
+				netif_carrier_on(nic->netdev);
+				netif_tx_start_all_queues(nic->netdev);
+			} else {
+				netdev_info(nic->netdev, "Link is Down\n");
+				netif_carrier_off(nic->netdev);
+				netif_tx_stop_all_queues(nic->netdev);
+			}
 		}
 		break;
 	case NIC_MBOX_MSG_ALLOC_SQS:
@@ -503,8 +527,9 @@ static int nicvf_init_resources(struct nicvf *nic)
 
 static inline bool nicvf_xdp_rx(struct nicvf *nic, struct bpf_prog *prog,
 				struct cqe_rx_t *cqe_rx, struct snd_queue *sq,
-				struct sk_buff **skb)
+				struct rcv_queue *rq, struct sk_buff **skb)
 {
+	unsigned char *hard_start, *data;
 	struct xdp_buff xdp;
 	struct page *page;
 	u32 action;
@@ -522,18 +547,18 @@ static inline bool nicvf_xdp_rx(struct nicvf *nic, struct bpf_prog *prog,
 	cpu_addr = (u64)phys_to_virt(cpu_addr);
 	page = virt_to_page((void *)cpu_addr);
 
-	xdp.data_hard_start = page_address(page);
-	xdp.data = (void *)cpu_addr;
-	xdp.data_end = xdp.data + len;
+	xdp_init_buff(&xdp, RCV_FRAG_LEN + XDP_PACKET_HEADROOM,
+		      &rq->xdp_rxq);
+	hard_start = page_address(page);
+	data = (unsigned char *)cpu_addr;
+	xdp_prepare_buff(&xdp, hard_start, data - hard_start, len, false);
 	orig_data = xdp.data;
 
-	rcu_read_lock();
 	action = bpf_prog_run_xdp(prog, &xdp);
-	rcu_read_unlock();
 
+	len = xdp.data_end - xdp.data;
 	/* Check if XDP program has changed headers */
 	if (orig_data != xdp.data) {
-		len = xdp.data_end - xdp.data;
 		offset = orig_data - xdp.data;
 		dma_addr -= offset;
 	}
@@ -565,9 +590,11 @@ static inline bool nicvf_xdp_rx(struct nicvf *nic, struct bpf_prog *prog,
 		nicvf_xdp_sq_append_pkt(nic, sq, (u64)xdp.data, dma_addr, len);
 		return true;
 	default:
-		bpf_warn_invalid_xdp_action(action);
+		bpf_warn_invalid_xdp_action(nic->netdev, prog, action);
+		fallthrough;
 	case XDP_ABORTED:
 		trace_xdp_exception(nic->netdev, prog, action);
+		fallthrough;
 	case XDP_DROP:
 		/* Check if it's a recycled page, if not
 		 * unmap the DMA mapping.
@@ -587,6 +614,44 @@ static inline bool nicvf_xdp_rx(struct nicvf *nic, struct bpf_prog *prog,
 	return false;
 }
 
+static void nicvf_snd_ptp_handler(struct net_device *netdev,
+				  struct cqe_send_t *cqe_tx)
+{
+	struct nicvf *nic = netdev_priv(netdev);
+	struct skb_shared_hwtstamps ts;
+	u64 ns;
+
+	nic = nic->pnicvf;
+
+	/* Sync for 'ptp_skb' */
+	smp_rmb();
+
+	/* New timestamp request can be queued now */
+	atomic_set(&nic->tx_ptp_skbs, 0);
+
+	/* Check for timestamp requested skb */
+	if (!nic->ptp_skb)
+		return;
+
+	/* Check if timestamping is timedout, which is set to 10us */
+	if (cqe_tx->send_status == CQ_TX_ERROP_TSTMP_TIMEOUT ||
+	    cqe_tx->send_status == CQ_TX_ERROP_TSTMP_CONFLICT)
+		goto no_tstamp;
+
+	/* Get the timestamp */
+	memset(&ts, 0, sizeof(ts));
+	ns = cavium_ptp_tstamp2time(nic->ptp_clock, cqe_tx->ptp_timestamp);
+	ts.hwtstamp = ns_to_ktime(ns);
+	skb_tstamp_tx(nic->ptp_skb, &ts);
+
+no_tstamp:
+	/* Free the original skb */
+	dev_kfree_skb_any(nic->ptp_skb);
+	nic->ptp_skb = NULL;
+	/* Sync 'ptp_skb' */
+	smp_wmb();
+}
+
 static void nicvf_snd_pkt_handler(struct net_device *netdev,
 				  struct cqe_send_t *cqe_tx,
 				  int budget, int *subdesc_cnt,
@@ -643,7 +708,12 @@ static void nicvf_snd_pkt_handler(struct net_device *netdev,
 		prefetch(skb);
 		(*tx_pkts)++;
 		*tx_bytes += skb->len;
-		napi_consume_skb(skb, budget);
+		/* If timestamp is requested for this skb, don't free it */
+		if (skb_shinfo(skb)->tx_flags & SKBTX_IN_PROGRESS &&
+		    !nic->pnicvf->ptp_skb)
+			nic->pnicvf->ptp_skb = skb;
+		else
+			napi_consume_skb(skb, budget);
 		sq->skbuff[cqe_tx->sqe_ptr] = (u64)NULL;
 	} else {
 		/* In case of SW TSO on 88xx, only last segment will have
@@ -682,9 +752,25 @@ static inline void nicvf_set_rxhash(struct net_device *netdev,
 	skb_set_hash(skb, hash, hash_type);
 }
 
+static inline void nicvf_set_rxtstamp(struct nicvf *nic, struct sk_buff *skb)
+{
+	u64 ns;
+
+	if (!nic->ptp_clock || !nic->hw_rx_tstamp)
+		return;
+
+	/* The first 8 bytes is the timestamp */
+	ns = cavium_ptp_tstamp2time(nic->ptp_clock,
+				    be64_to_cpu(*(__be64 *)skb->data));
+	skb_hwtstamps(skb)->hwtstamp = ns_to_ktime(ns);
+
+	__skb_pull(skb, 8);
+}
+
 static void nicvf_rcv_pkt_handler(struct net_device *netdev,
 				  struct napi_struct *napi,
-				  struct cqe_rx_t *cqe_rx, struct snd_queue *sq)
+				  struct cqe_rx_t *cqe_rx,
+				  struct snd_queue *sq, struct rcv_queue *rq)
 {
 	struct sk_buff *skb = NULL;
 	struct nicvf *nic = netdev_priv(netdev);
@@ -710,7 +796,7 @@ static void nicvf_rcv_pkt_handler(struct net_device *netdev,
 	/* For XDP, ignore pkts spanning multiple pages */
 	if (nic->xdp_prog && (cqe_rx->rb_cnt == 1)) {
 		/* Packet consumed by XDP */
-		if (nicvf_xdp_rx(snic, nic->xdp_prog, cqe_rx, sq, &skb))
+		if (nicvf_xdp_rx(snic, nic->xdp_prog, cqe_rx, sq, rq, &skb))
 			return;
 	} else {
 		skb = nicvf_get_rcv_skb(snic, cqe_rx,
@@ -721,8 +807,7 @@ static void nicvf_rcv_pkt_handler(struct net_device *netdev,
 		return;
 
 	if (netif_msg_pktdata(nic)) {
-		netdev_info(nic->netdev, "%s: skb 0x%p, len=%d\n", netdev->name,
-			    skb, skb->len);
+		netdev_info(nic->netdev, "skb 0x%p, len=%d\n", skb, skb->len);
 		print_hex_dump(KERN_INFO, "", DUMP_PREFIX_OFFSET, 16, 1,
 			       skb->data, skb->len, true);
 	}
@@ -733,6 +818,7 @@ static void nicvf_rcv_pkt_handler(struct net_device *netdev,
 		return;
 	}
 
+	nicvf_set_rxtstamp(nic, skb);
 	nicvf_set_rxhash(netdev, cqe_rx, skb);
 
 	skb_record_rx_queue(skb, rq_idx);
@@ -768,6 +854,7 @@ static int nicvf_cq_intr_handler(struct net_device *netdev, u8 cq_idx,
 	struct cqe_rx_t *cq_desc;
 	struct netdev_queue *txq;
 	struct snd_queue *sq = &qs->sq[cq_idx];
+	struct rcv_queue *rq = &qs->rq[cq_idx];
 	unsigned int tx_pkts = 0, tx_bytes = 0, txq_idx;
 
 	spin_lock_bh(&cq->lock);
@@ -798,7 +885,7 @@ loop:
 
 		switch (cq_desc->cqe_type) {
 		case CQE_TYPE_RX:
-			nicvf_rcv_pkt_handler(netdev, napi, cq_desc, sq);
+			nicvf_rcv_pkt_handler(netdev, napi, cq_desc, sq, rq);
 			work_done++;
 		break;
 		case CQE_TYPE_SEND:
@@ -807,10 +894,12 @@ loop:
 					      &tx_pkts, &tx_bytes);
 			tx_done++;
 		break;
+		case CQE_TYPE_SEND_PTP:
+			nicvf_snd_ptp_handler(netdev, (void *)cq_desc);
+		break;
 		case CQE_TYPE_INVALID:
 		case CQE_TYPE_RX_SPLIT:
 		case CQE_TYPE_RX_TCP:
-		case CQE_TYPE_SEND_PTP:
 			/* Ignore for now */
 		break;
 		}
@@ -854,10 +943,8 @@ done:
 			netif_tx_wake_queue(txq);
 			nic = nic->pnicvf;
 			this_cpu_inc(nic->drv_stats->txq_wake);
-			if (netif_msg_tx_err(nic))
-				netdev_warn(netdev,
-					    "%s: Transmit queue wakeup SQ%d\n",
-					    netdev->name, txq_idx);
+			netif_warn(nic, tx_err, netdev,
+				   "Transmit queue wakeup SQ%d\n", txq_idx);
 		}
 	}
 
@@ -895,9 +982,9 @@ static int nicvf_poll(struct napi_struct *napi, int budget)
  *
  * As of now only CQ errors are handled
  */
-static void nicvf_handle_qs_err(unsigned long data)
+static void nicvf_handle_qs_err(struct tasklet_struct *t)
 {
-	struct nicvf *nic = (struct nicvf *)data;
+	struct nicvf *nic = from_tasklet(nic, t, qs_err_task);
 	struct queue_set *qs = nic->qs;
 	int qidx;
 	u64 status;
@@ -928,9 +1015,8 @@ static void nicvf_handle_qs_err(unsigned long data)
 
 static void nicvf_dump_intr_status(struct nicvf *nic)
 {
-	if (netif_msg_intr(nic))
-		netdev_info(nic->netdev, "%s: interrupt status 0x%llx\n",
-			    nic->netdev->name, nicvf_reg_read(nic, NIC_VF_INT));
+	netif_info(nic, intr, nic->netdev, "interrupt status 0x%llx\n",
+		   nicvf_reg_read(nic, NIC_VF_INT));
 }
 
 static irqreturn_t nicvf_misc_intr_handler(int irq, void *nicvf_irq)
@@ -1137,7 +1223,7 @@ static int nicvf_register_misc_interrupt(struct nicvf *nic)
 	if (ret < 0) {
 		netdev_err(nic->netdev,
 			   "Req for #%d msix vectors failed\n", nic->num_vec);
-		return 1;
+		return ret;
 	}
 
 	sprintf(nic->irq_name[irq], "%s Mbox", "NICVF");
@@ -1156,7 +1242,7 @@ static int nicvf_register_misc_interrupt(struct nicvf *nic)
 	if (!nicvf_check_pf_ready(nic)) {
 		nicvf_disable_intr(nic, NICVF_INTR_MBOX, 0);
 		nicvf_unregister_interrupts(nic);
-		return 1;
+		return -EIO;
 	}
 
 	return 0;
@@ -1212,10 +1298,8 @@ static netdev_tx_t nicvf_xmit(struct sk_buff *skb, struct net_device *netdev)
 			netif_tx_wake_queue(txq);
 		} else {
 			this_cpu_inc(nic->drv_stats->txq_stop);
-			if (netif_msg_tx_err(nic))
-				netdev_warn(netdev,
-					    "%s: Transmit ring full, stopping SQ%d\n",
-					    netdev->name, qid);
+			netif_warn(nic, tx_err, netdev,
+				   "Transmit ring full, stopping SQ%d\n", qid);
 		}
 		return NETDEV_TX_BUSY;
 	}
@@ -1245,6 +1329,12 @@ int nicvf_stop(struct net_device *netdev)
 	struct nicvf_cq_poll *cq_poll = NULL;
 	union nic_mbx mbx = {};
 
+	/* wait till all queued set_rx_mode tasks completes */
+	if (nic->nicvf_rx_mode_wq) {
+		cancel_delayed_work_sync(&nic->link_change_work);
+		drain_workqueue(nic->nicvf_rx_mode_wq);
+	}
+
 	mbx.msg.msg = NIC_MBOX_MSG_SHUTDOWN;
 	nicvf_send_msg_to_pf(nic, &mbx);
 
@@ -1311,12 +1401,28 @@ int nicvf_stop(struct net_device *netdev)
 
 	nicvf_free_cq_poll(nic);
 
+	/* Free any pending SKB saved to receive timestamp */
+	if (nic->ptp_skb) {
+		dev_kfree_skb_any(nic->ptp_skb);
+		nic->ptp_skb = NULL;
+	}
+
 	/* Clear multiqset info */
 	nic->pnicvf = nic;
 
 	return 0;
 }
 
+static int nicvf_config_hw_rx_tstamp(struct nicvf *nic, bool enable)
+{
+	union nic_mbx mbx = {};
+
+	mbx.ptp.msg = NIC_MBOX_MSG_PTP_CFG;
+	mbx.ptp.enable = enable;
+
+	return nicvf_send_msg_to_pf(nic, &mbx);
+}
+
 static int nicvf_update_hw_max_frs(struct nicvf *nic, int mtu)
 {
 	union nic_mbx mbx = {};
@@ -1328,13 +1434,28 @@ static int nicvf_update_hw_max_frs(struct nicvf *nic, int mtu)
 	return nicvf_send_msg_to_pf(nic, &mbx);
 }
 
+static void nicvf_link_status_check_task(struct work_struct *work_arg)
+{
+	struct nicvf *nic = container_of(work_arg,
+					 struct nicvf,
+					 link_change_work.work);
+	union nic_mbx mbx = {};
+	mbx.msg.msg = NIC_MBOX_MSG_BGX_LINK_CHANGE;
+	nicvf_send_msg_to_pf(nic, &mbx);
+	queue_delayed_work(nic->nicvf_rx_mode_wq,
+			   &nic->link_change_work, 2 * HZ);
+}
+
 int nicvf_open(struct net_device *netdev)
 {
 	int cpu, err, qidx;
 	struct nicvf *nic = netdev_priv(netdev);
 	struct queue_set *qs = nic->qs;
 	struct nicvf_cq_poll *cq_poll = NULL;
-	union nic_mbx mbx = {};
+
+	/* wait till all queued set_rx_mode tasks completes if any */
+	if (nic->nicvf_rx_mode_wq)
+		drain_workqueue(nic->nicvf_rx_mode_wq);
 
 	netif_carrier_off(netdev);
 
@@ -1351,8 +1472,7 @@ int nicvf_open(struct net_device *netdev)
 		}
 		cq_poll->cq_idx = qidx;
 		cq_poll->nicvf = nic;
-		netif_napi_add(netdev, &cq_poll->napi, nicvf_poll,
-			       NAPI_POLL_WEIGHT);
+		netif_napi_add(netdev, &cq_poll->napi, nicvf_poll);
 		napi_enable(&cq_poll->napi);
 		nic->napi[qidx] = cq_poll;
 	}
@@ -1369,12 +1489,10 @@ int nicvf_open(struct net_device *netdev)
 	}
 
 	/* Init tasklet for handling Qset err interrupt */
-	tasklet_init(&nic->qs_err_task, nicvf_handle_qs_err,
-		     (unsigned long)nic);
+	tasklet_setup(&nic->qs_err_task, nicvf_handle_qs_err);
 
 	/* Init RBDR tasklet which will refill RBDR */
-	tasklet_init(&nic->rbdr_task, nicvf_rbdr_task,
-		     (unsigned long)nic);
+	tasklet_setup(&nic->rbdr_task, nicvf_rbdr_task);
 	INIT_DELAYED_WORK(&nic->rbdr_work, nicvf_rbdr_work);
 
 	/* Configure CPI alorithm */
@@ -1386,6 +1504,12 @@ int nicvf_open(struct net_device *netdev)
 	if (nic->sqs_mode)
 		nicvf_get_primary_vf_struct(nic);
 
+	/* Configure PTP timestamp */
+	if (nic->ptp_clock)
+		nicvf_config_hw_rx_tstamp(nic, nic->hw_rx_tstamp);
+	atomic_set(&nic->tx_ptp_skbs, 0);
+	nic->ptp_skb = NULL;
+
 	/* Configure receive side scaling and MTU */
 	if (!nic->sqs_mode) {
 		nicvf_rss_init(nic);
@@ -1424,8 +1548,14 @@ int nicvf_open(struct net_device *netdev)
 		nicvf_enable_intr(nic, NICVF_INTR_RBDR, qidx);
 
 	/* Send VF config done msg to PF */
-	mbx.msg.msg = NIC_MBOX_MSG_CFG_DONE;
-	nicvf_write_to_mbx(nic, &mbx);
+	nicvf_send_cfg_done(nic);
+
+	if (nic->nicvf_rx_mode_wq) {
+		INIT_DELAYED_WORK(&nic->link_change_work,
+				  nicvf_link_status_check_task);
+		queue_delayed_work(nic->nicvf_rx_mode_wq,
+				   &nic->link_change_work, 0);
+	}
 
 	return 0;
 cleanup:
@@ -1448,18 +1578,21 @@ napi_del:
 static int nicvf_change_mtu(struct net_device *netdev, int new_mtu)
 {
 	struct nicvf *nic = netdev_priv(netdev);
-	int orig_mtu = netdev->mtu;
-
-	netdev->mtu = new_mtu;
-
-	if (!netif_running(netdev))
-		return 0;
 
-	if (nicvf_update_hw_max_frs(nic, new_mtu)) {
-		netdev->mtu = orig_mtu;
+	/* For now just support only the usual MTU sized frames,
+	 * plus some headroom for VLAN, QinQ.
+	 */
+	if (nic->xdp_prog && new_mtu > MAX_XDP_MTU) {
+		netdev_warn(netdev, "Jumbo frames not yet supported with XDP, current MTU %d.\n",
+			    netdev->mtu);
 		return -EINVAL;
 	}
 
+	if (netif_running(netdev) && nicvf_update_hw_max_frs(nic, new_mtu))
+		return -EINVAL;
+
+	WRITE_ONCE(netdev->mtu, new_mtu);
+
 	return 0;
 }
 
@@ -1471,7 +1604,7 @@ static int nicvf_set_mac_address(struct net_device *netdev, void *p)
 	if (!is_valid_ether_addr(addr->sa_data))
 		return -EADDRNOTAVAIL;
 
-	memcpy(netdev->dev_addr, addr->sa_data, netdev->addr_len);
+	eth_hw_addr_set(netdev, addr->sa_data);
 
 	if (nic->pdev->msix_enabled) {
 		if (nicvf_hw_set_mac_addr(nic, netdev))
@@ -1596,13 +1729,11 @@ static void nicvf_get_stats64(struct net_device *netdev,
 
 }
 
-static void nicvf_tx_timeout(struct net_device *dev)
+static void nicvf_tx_timeout(struct net_device *dev, unsigned int txqueue)
 {
 	struct nicvf *nic = netdev_priv(dev);
 
-	if (netif_msg_tx_err(nic))
-		netdev_warn(dev, "%s: Transmit timed out, resetting\n",
-			    dev->name);
+	netif_warn(nic, tx_err, dev, "Transmit timed out, resetting\n");
 
 	this_cpu_inc(nic->drv_stats->tx_timeout);
 	schedule_work(&nic->reset_task);
@@ -1698,9 +1829,12 @@ static int nicvf_xdp_setup(struct nicvf *nic, struct bpf_prog *prog)
 	bool if_up = netif_running(nic->netdev);
 	struct bpf_prog *old_prog;
 	bool bpf_attached = false;
+	int ret = 0;
 
-	/* For now just support only the usual MTU sized frames */
-	if (prog && (dev->mtu > 1500)) {
+	/* For now just support only the usual MTU sized frames,
+	 * plus some headroom for VLAN, QinQ.
+	 */
+	if (prog && dev->mtu > MAX_XDP_MTU) {
 		netdev_warn(dev, "Jumbo frames not yet supported with XDP, current MTU %d.\n",
 			    dev->mtu);
 		return -EOPNOTSUPP;
@@ -1730,9 +1864,8 @@ static int nicvf_xdp_setup(struct nicvf *nic, struct bpf_prog *prog)
 
 	if (nic->xdp_prog) {
 		/* Attach BPF program */
-		nic->xdp_prog = bpf_prog_add(nic->xdp_prog, nic->rx_queues - 1);
-		if (!IS_ERR(nic->xdp_prog))
-			bpf_attached = true;
+		bpf_prog_add(nic->xdp_prog, nic->rx_queues - 1);
+		bpf_attached = true;
 	}
 
 	/* Calculate Tx queues needed for XDP and network stack */
@@ -1744,10 +1877,10 @@ static int nicvf_xdp_setup(struct nicvf *nic, struct bpf_prog *prog)
 		netif_trans_update(nic->netdev);
 	}
 
-	return 0;
+	return ret;
 }
 
-static int nicvf_xdp(struct net_device *netdev, struct netdev_xdp *xdp)
+static int nicvf_xdp(struct net_device *netdev, struct netdev_bpf *xdp)
 {
 	struct nicvf *nic = netdev_priv(netdev);
 
@@ -1761,14 +1894,182 @@ static int nicvf_xdp(struct net_device *netdev, struct netdev_xdp *xdp)
 	switch (xdp->command) {
 	case XDP_SETUP_PROG:
 		return nicvf_xdp_setup(nic, xdp->prog);
-	case XDP_QUERY_PROG:
-		xdp->prog_attached = !!nic->xdp_prog;
-		return 0;
 	default:
 		return -EINVAL;
 	}
 }
 
+static int nicvf_config_hwtstamp(struct net_device *netdev, struct ifreq *ifr)
+{
+	struct hwtstamp_config config;
+	struct nicvf *nic = netdev_priv(netdev);
+
+	if (!nic->ptp_clock)
+		return -ENODEV;
+
+	if (copy_from_user(&config, ifr->ifr_data, sizeof(config)))
+		return -EFAULT;
+
+	switch (config.tx_type) {
+	case HWTSTAMP_TX_OFF:
+	case HWTSTAMP_TX_ON:
+		break;
+	default:
+		return -ERANGE;
+	}
+
+	switch (config.rx_filter) {
+	case HWTSTAMP_FILTER_NONE:
+		nic->hw_rx_tstamp = false;
+		break;
+	case HWTSTAMP_FILTER_ALL:
+	case HWTSTAMP_FILTER_SOME:
+	case HWTSTAMP_FILTER_PTP_V1_L4_EVENT:
+	case HWTSTAMP_FILTER_PTP_V1_L4_SYNC:
+	case HWTSTAMP_FILTER_PTP_V1_L4_DELAY_REQ:
+	case HWTSTAMP_FILTER_PTP_V2_L4_EVENT:
+	case HWTSTAMP_FILTER_PTP_V2_L4_SYNC:
+	case HWTSTAMP_FILTER_PTP_V2_L4_DELAY_REQ:
+	case HWTSTAMP_FILTER_PTP_V2_L2_EVENT:
+	case HWTSTAMP_FILTER_PTP_V2_L2_SYNC:
+	case HWTSTAMP_FILTER_PTP_V2_L2_DELAY_REQ:
+	case HWTSTAMP_FILTER_PTP_V2_EVENT:
+	case HWTSTAMP_FILTER_PTP_V2_SYNC:
+	case HWTSTAMP_FILTER_PTP_V2_DELAY_REQ:
+		nic->hw_rx_tstamp = true;
+		config.rx_filter = HWTSTAMP_FILTER_ALL;
+		break;
+	default:
+		return -ERANGE;
+	}
+
+	if (netif_running(netdev))
+		nicvf_config_hw_rx_tstamp(nic, nic->hw_rx_tstamp);
+
+	if (copy_to_user(ifr->ifr_data, &config, sizeof(config)))
+		return -EFAULT;
+
+	return 0;
+}
+
+static int nicvf_ioctl(struct net_device *netdev, struct ifreq *req, int cmd)
+{
+	switch (cmd) {
+	case SIOCSHWTSTAMP:
+		return nicvf_config_hwtstamp(netdev, req);
+	default:
+		return -EOPNOTSUPP;
+	}
+}
+
+static void __nicvf_set_rx_mode_task(u8 mode, struct xcast_addr_list *mc_addrs,
+				     struct nicvf *nic)
+{
+	union nic_mbx mbx = {};
+	int idx;
+
+	/* From the inside of VM code flow we have only 128 bits memory
+	 * available to send message to host's PF, so send all mc addrs
+	 * one by one, starting from flush command in case if kernel
+	 * requests to configure specific MAC filtering
+	 */
+
+	/* flush DMAC filters and reset RX mode */
+	mbx.xcast.msg = NIC_MBOX_MSG_RESET_XCAST;
+	if (nicvf_send_msg_to_pf(nic, &mbx) < 0)
+		goto free_mc;
+
+	if (mode & BGX_XCAST_MCAST_FILTER) {
+		/* once enabling filtering, we need to signal to PF to add
+		 * its' own LMAC to the filter to accept packets for it.
+		 */
+		mbx.xcast.msg = NIC_MBOX_MSG_ADD_MCAST;
+		mbx.xcast.mac = 0;
+		if (nicvf_send_msg_to_pf(nic, &mbx) < 0)
+			goto free_mc;
+	}
+
+	/* check if we have any specific MACs to be added to PF DMAC filter */
+	if (mc_addrs) {
+		/* now go through kernel list of MACs and add them one by one */
+		for (idx = 0; idx < mc_addrs->count; idx++) {
+			mbx.xcast.msg = NIC_MBOX_MSG_ADD_MCAST;
+			mbx.xcast.mac = mc_addrs->mc[idx];
+			if (nicvf_send_msg_to_pf(nic, &mbx) < 0)
+				goto free_mc;
+		}
+	}
+
+	/* and finally set rx mode for PF accordingly */
+	mbx.xcast.msg = NIC_MBOX_MSG_SET_XCAST;
+	mbx.xcast.mode = mode;
+
+	nicvf_send_msg_to_pf(nic, &mbx);
+free_mc:
+	kfree(mc_addrs);
+}
+
+static void nicvf_set_rx_mode_task(struct work_struct *work_arg)
+{
+	struct nicvf_work *vf_work = container_of(work_arg, struct nicvf_work,
+						  work);
+	struct nicvf *nic = container_of(vf_work, struct nicvf, rx_mode_work);
+	u8 mode;
+	struct xcast_addr_list *mc;
+
+	/* Save message data locally to prevent them from
+	 * being overwritten by next ndo_set_rx_mode call().
+	 */
+	spin_lock_bh(&nic->rx_mode_wq_lock);
+	mode = vf_work->mode;
+	mc = vf_work->mc;
+	vf_work->mc = NULL;
+	spin_unlock_bh(&nic->rx_mode_wq_lock);
+
+	__nicvf_set_rx_mode_task(mode, mc, nic);
+}
+
+static void nicvf_set_rx_mode(struct net_device *netdev)
+{
+	struct nicvf *nic = netdev_priv(netdev);
+	struct netdev_hw_addr *ha;
+	struct xcast_addr_list *mc_list = NULL;
+	u8 mode = 0;
+
+	if (netdev->flags & IFF_PROMISC) {
+		mode = BGX_XCAST_BCAST_ACCEPT | BGX_XCAST_MCAST_ACCEPT;
+	} else {
+		if (netdev->flags & IFF_BROADCAST)
+			mode |= BGX_XCAST_BCAST_ACCEPT;
+
+		if (netdev->flags & IFF_ALLMULTI) {
+			mode |= BGX_XCAST_MCAST_ACCEPT;
+		} else if (netdev->flags & IFF_MULTICAST) {
+			mode |= BGX_XCAST_MCAST_FILTER;
+			/* here we need to copy mc addrs */
+			if (netdev_mc_count(netdev)) {
+				mc_list = kmalloc(struct_size(mc_list, mc,
+							      netdev_mc_count(netdev)),
+						  GFP_ATOMIC);
+				if (unlikely(!mc_list))
+					return;
+				mc_list->count = 0;
+				netdev_hw_addr_list_for_each(ha, &netdev->mc) {
+					mc_list->mc[mc_list->count] =
+						ether_addr_to_u64(ha->addr);
+					mc_list->count++;
+				}
+			}
+		}
+	}
+	spin_lock(&nic->rx_mode_wq_lock);
+	kfree(nic->rx_mode_work.mc);
+	nic->rx_mode_work.mc = mc_list;
+	nic->rx_mode_work.mode = mode;
+	queue_work(nic->nicvf_rx_mode_wq, &nic->rx_mode_work.work);
+	spin_unlock(&nic->rx_mode_wq_lock);
+}
+
 static const struct net_device_ops nicvf_netdev_ops = {
 	.ndo_open		= nicvf_open,
 	.ndo_stop		= nicvf_stop,
@@ -1779,7 +2080,9 @@ static const struct net_device_ops nicvf_netdev_ops = {
 	.ndo_tx_timeout         = nicvf_tx_timeout,
 	.ndo_fix_features       = nicvf_fix_features,
 	.ndo_set_features       = nicvf_set_features,
-	.ndo_xdp		= nicvf_xdp,
+	.ndo_bpf		= nicvf_xdp,
+	.ndo_eth_ioctl           = nicvf_ioctl,
+	.ndo_set_rx_mode        = nicvf_set_rx_mode,
 };
 
 static int nicvf_probe(struct pci_dev *pdev, const struct pci_device_id *ent)
@@ -1789,31 +2092,33 @@ static int nicvf_probe(struct pci_dev *pdev, const struct pci_device_id *ent)
 	struct nicvf *nic;
 	int    err, qcount;
 	u16    sdevid;
+	struct cavium_ptp *ptp_clock;
 
-	err = pci_enable_device(pdev);
-	if (err) {
-		dev_err(dev, "Failed to enable PCI device\n");
-		return err;
+	ptp_clock = cavium_ptp_get();
+	if (IS_ERR(ptp_clock)) {
+		if (PTR_ERR(ptp_clock) == -ENODEV)
+			/* In virtualized environment we proceed without ptp */
+			ptp_clock = NULL;
+		else
+			return PTR_ERR(ptp_clock);
 	}
 
+	err = pci_enable_device(pdev);
+	if (err)
+		return dev_err_probe(dev, err, "Failed to enable PCI device\n");
+
 	err = pci_request_regions(pdev, DRV_NAME);
 	if (err) {
 		dev_err(dev, "PCI request regions failed 0x%x\n", err);
 		goto err_disable_device;
 	}
 
-	err = pci_set_dma_mask(pdev, DMA_BIT_MASK(48));
+	err = dma_set_mask_and_coherent(&pdev->dev, DMA_BIT_MASK(48));
 	if (err) {
 		dev_err(dev, "Unable to get usable DMA configuration\n");
 		goto err_release_regions;
 	}
 
-	err = pci_set_consistent_dma_mask(pdev, DMA_BIT_MASK(48));
-	if (err) {
-		dev_err(dev, "unable to get 48-bit DMA for consistent allocations\n");
-		goto err_release_regions;
-	}
-
 	qcount = netif_get_num_default_rss_queues();
 
 	/* Restrict multiqset support only for host bound VFs */
@@ -1838,6 +2143,15 @@ static int nicvf_probe(struct pci_dev *pdev, const struct pci_device_id *ent)
 	nic->pdev = pdev;
 	nic->pnicvf = nic;
 	nic->max_queues = qcount;
+	/* If no of CPUs are too low, there won't be any queues left
+	 * for XDP_TX, hence double it.
+	 */
+	if (!nic->t88)
+		nic->max_queues *= 2;
+	nic->ptp_clock = ptp_clock;
+
+	/* Initialize mutex that serializes usage of VF's mailbox */
+	mutex_init(&nic->rx_mode_mtx);
 
 	/* MAP VF's configuration registers */
 	nic->reg_base = pcim_iomap(pdev, PCI_CFG_REG_BAR_NUM, 0);
@@ -1898,16 +2212,32 @@ static int nicvf_probe(struct pci_dev *pdev, const struct pci_device_id *ent)
 	netdev->netdev_ops = &nicvf_netdev_ops;
 	netdev->watchdog_timeo = NICVF_TX_TIMEOUT;
 
+	if (!pass1_silicon(nic->pdev) &&
+	    nic->rx_queues + nic->tx_queues <= nic->max_queues)
+		netdev->xdp_features = NETDEV_XDP_ACT_BASIC;
+
 	/* MTU range: 64 - 9200 */
 	netdev->min_mtu = NIC_HW_MIN_FRS;
 	netdev->max_mtu = NIC_HW_MAX_FRS;
 
 	INIT_WORK(&nic->reset_task, nicvf_reset_task);
 
+	nic->nicvf_rx_mode_wq = alloc_ordered_workqueue("nicvf_rx_mode_wq_VF%d",
+							WQ_MEM_RECLAIM,
+							nic->vf_id);
+	if (!nic->nicvf_rx_mode_wq) {
+		err = -ENOMEM;
+		dev_err(dev, "Failed to allocate work queue\n");
+		goto err_unregister_interrupts;
+	}
+
+	INIT_WORK(&nic->rx_mode_work.work, nicvf_set_rx_mode_task);
+	spin_lock_init(&nic->rx_mode_wq_lock);
+
 	err = register_netdev(netdev);
 	if (err) {
 		dev_err(dev, "Failed to register netdevice\n");
-		goto err_unregister_interrupts;
+		goto err_destroy_workqueue;
 	}
 
 	nic->msg_enable = debug;
@@ -1916,6 +2246,8 @@ static int nicvf_probe(struct pci_dev *pdev, const struct pci_device_id *ent)
 
 	return 0;
 
+err_destroy_workqueue:
+	destroy_workqueue(nic->nicvf_rx_mode_wq);
 err_unregister_interrupts:
 	nicvf_unregister_interrupts(nic);
 err_free_netdev:
@@ -1947,10 +2279,15 @@ static void nicvf_remove(struct pci_dev *pdev)
 	 */
 	if (pnetdev && (pnetdev->reg_state == NETREG_REGISTERED))
 		unregister_netdev(pnetdev);
+	if (nic->nicvf_rx_mode_wq) {
+		destroy_workqueue(nic->nicvf_rx_mode_wq);
+		nic->nicvf_rx_mode_wq = NULL;
+	}
 	nicvf_unregister_interrupts(nic);
 	pci_set_drvdata(pdev, NULL);
 	if (nic->drv_stats)
 		free_percpu(nic->drv_stats);
+	cavium_ptp_put(nic->ptp_clock);
 	free_netdev(netdev);
 	pci_release_regions(pdev);
 	pci_disable_device(pdev);
@@ -1972,7 +2309,6 @@ static struct pci_driver nicvf_driver = {
 static int __init nicvf_init_module(void)
 {
 	pr_info("%s, ver %s\n", DRV_NAME, DRV_VERSION);
-
 	return pci_register_driver(&nicvf_driver);
 }
 
diff --git a/drivers/net/ethernet/cavium/thunder/nicvf_queues.c b/drivers/net/ethernet/cavium/thunder/nicvf_queues.c
index 2b181762ad49..5211759bfe47 100644
--- a/drivers/net/ethernet/cavium/thunder/nicvf_queues.c
+++ b/drivers/net/ethernet/cavium/thunder/nicvf_queues.c
@@ -1,9 +1,6 @@
+// SPDX-License-Identifier: GPL-2.0-only
 /*
  * Copyright (C) 2015 Cavium, Inc.
- *
- * This program is free software; you can redistribute it and/or modify it
- * under the terms of version 2 of the GNU General Public License
- * as published by the Free Software Foundation.
  */
 
 #include <linux/pci.h>
@@ -13,6 +10,7 @@
 #include <linux/iommu.h>
 #include <net/ip.h>
 #include <net/tso.h>
+#include <uapi/linux/bpf.h>
 
 #include "nic_reg.h"
 #include "nic.h"
@@ -59,7 +57,7 @@ static int nicvf_alloc_q_desc_mem(struct nicvf *nic, struct q_desc_mem *dmem,
 	dmem->q_len = q_len;
 	dmem->size = (desc_size * q_len) + align_bytes;
 	/* Save address, need it while freeing */
-	dmem->unalign_base = dma_zalloc_coherent(&nic->pdev->dev, dmem->size,
+	dmem->unalign_base = dma_alloc_coherent(&nic->pdev->dev, dmem->size,
 						&dmem->dma, GFP_KERNEL);
 	if (!dmem->unalign_base)
 		return -ENOMEM;
@@ -105,20 +103,19 @@ static inline struct pgcache *nicvf_alloc_page(struct nicvf *nic,
 	/* Check if page can be recycled */
 	if (page) {
 		ref_count = page_ref_count(page);
-		/* Check if this page has been used once i.e 'put_page'
-		 * called after packet transmission i.e internal ref_count
-		 * and page's ref_count are equal i.e page can be recycled.
-		 */
-		if (rbdr->is_xdp && (ref_count == pgcache->ref_count))
-			pgcache->ref_count--;
-		else
-			page = NULL;
-
-		/* In non-XDP mode, page's ref_count needs to be '1' for it
-		 * to be recycled.
+		/* This page can be recycled if internal ref_count and page's
+		 * ref_count are equal, indicating that the page has been used
+		 * once for packet transmission. For non-XDP mode, internal
+		 * ref_count is always '1'.
 		 */
-		if (!rbdr->is_xdp && (ref_count != 1))
+		if (rbdr->is_xdp) {
+			if (ref_count == pgcache->ref_count)
+				pgcache->ref_count--;
+			else
+				page = NULL;
+		} else if (ref_count != 1) {
 			page = NULL;
+		}
 	}
 
 	if (!page) {
@@ -292,8 +289,8 @@ static int  nicvf_init_rbdr(struct nicvf *nic, struct rbdr *rbdr,
 		rbdr->is_xdp = true;
 	}
 	rbdr->pgcnt = roundup_pow_of_two(rbdr->pgcnt);
-	rbdr->pgcache = kzalloc(sizeof(*rbdr->pgcache) *
-				rbdr->pgcnt, GFP_KERNEL);
+	rbdr->pgcache = kcalloc(rbdr->pgcnt, sizeof(*rbdr->pgcache),
+				GFP_KERNEL);
 	if (!rbdr->pgcache)
 		return -ENOMEM;
 	rbdr->pgidx = 0;
@@ -365,11 +362,10 @@ static void nicvf_free_rbdr(struct nicvf *nic, struct rbdr *rbdr)
 	while (head < rbdr->pgcnt) {
 		pgcache = &rbdr->pgcache[head];
 		if (pgcache->page && page_ref_count(pgcache->page) != 0) {
-			if (!rbdr->is_xdp) {
-				put_page(pgcache->page);
-				continue;
+			if (rbdr->is_xdp) {
+				page_ref_sub(pgcache->page,
+					     pgcache->ref_count - 1);
 			}
-			page_ref_sub(pgcache->page, pgcache->ref_count - 1);
 			put_page(pgcache->page);
 		}
 		head++;
@@ -465,9 +461,9 @@ void nicvf_rbdr_work(struct work_struct *work)
 }
 
 /* In Softirq context, alloc rcv buffers in atomic mode */
-void nicvf_rbdr_task(unsigned long data)
+void nicvf_rbdr_task(struct tasklet_struct *t)
 {
-	struct nicvf *nic = (struct nicvf *)data;
+	struct nicvf *nic = from_tasklet(nic, t, rbdr_task);
 
 	nicvf_refill_rbdr(nic, GFP_ATOMIC);
 	if (nic->rb_alloc_fail) {
@@ -585,10 +581,12 @@ static void nicvf_free_snd_queue(struct nicvf *nic, struct snd_queue *sq)
 	if (!sq->dmem.base)
 		return;
 
-	if (sq->tso_hdrs)
+	if (sq->tso_hdrs) {
 		dma_free_coherent(&nic->pdev->dev,
 				  sq->dmem.q_len * TSO_HEADER_SIZE,
 				  sq->tso_hdrs, sq->tso_hdrs_phys);
+		sq->tso_hdrs = NULL;
+	}
 
 	/* Free pending skbs in the queue */
 	smp_rmb();
@@ -759,6 +757,7 @@ static void nicvf_rcv_queue_config(struct nicvf *nic, struct queue_set *qs,
 
 	if (!rq->enable) {
 		nicvf_reclaim_rcv_queue(nic, qs, qidx);
+		xdp_rxq_info_unreg(&rq->xdp_rxq);
 		return;
 	}
 
@@ -771,11 +770,14 @@ static void nicvf_rcv_queue_config(struct nicvf *nic, struct queue_set *qs,
 	/* all writes of RBDR data to be loaded into L2 Cache as well*/
 	rq->caching = 1;
 
+	/* Driver have no proper error path for failed XDP RX-queue info reg */
+	WARN_ON(xdp_rxq_info_reg(&rq->xdp_rxq, nic->netdev, qidx, 0) < 0);
+
 	/* Send a mailbox msg to PF to config RQ */
 	mbx.rq.msg = NIC_MBOX_MSG_RQ_CFG;
 	mbx.rq.qs_num = qs->vnic_id;
 	mbx.rq.rq_num = qidx;
-	mbx.rq.cfg = (rq->caching << 26) | (rq->cq_qs << 19) |
+	mbx.rq.cfg = ((u64)rq->caching << 26) | (rq->cq_qs << 19) |
 			  (rq->cq_idx << 16) | (rq->cont_rbdr_qs << 9) |
 			  (rq->cont_qs_rbdr_idx << 8) |
 			  (rq->start_rbdr_qs << 1) | (rq->start_qs_rbdr_idx);
@@ -977,6 +979,9 @@ void nicvf_qset_config(struct nicvf *nic, bool enable)
 		qs_cfg->be = 1;
 #endif
 		qs_cfg->vnic = qs->vnic_id;
+		/* Enable Tx timestamping capability */
+		if (nic->ptp_clock)
+			qs_cfg->send_tstmp_ena = 1;
 	}
 	nicvf_send_msg_to_pf(nic, &mbx);
 }
@@ -1175,13 +1180,12 @@ void nicvf_sq_disable(struct nicvf *nic, int qidx)
 void nicvf_sq_free_used_descs(struct net_device *netdev, struct snd_queue *sq,
 			      int qidx)
 {
-	u64 head, tail;
+	u64 head;
 	struct sk_buff *skb;
 	struct nicvf *nic = netdev_priv(netdev);
 	struct sq_hdr_subdesc *hdr;
 
 	head = nicvf_queue_reg_read(nic, NIC_QSET_SQ_0_7_HEAD, qidx) >> 4;
-	tail = nicvf_queue_reg_read(nic, NIC_QSET_SQ_0_7_TAIL, qidx) >> 4;
 	while (sq->head != head) {
 		hdr = (struct sq_hdr_subdesc *)GET_SQ_DESC(sq, sq->head);
 		if (hdr->subdesc_type != SQ_DESC_TYPE_HEADER) {
@@ -1257,7 +1261,7 @@ int nicvf_xdp_sq_append_pkt(struct nicvf *nic, struct snd_queue *sq,
 static int nicvf_tso_count_subdescs(struct sk_buff *skb)
 {
 	struct skb_shared_info *sh = skb_shinfo(skb);
-	unsigned int sh_len = skb_transport_offset(skb) + tcp_hdrlen(skb);
+	unsigned int sh_len = skb_tcp_all_headers(skb);
 	unsigned int data_len = skb->len - sh_len;
 	unsigned int p_len = sh->gso_size;
 	long f_id = -1;    /* id of the current fragment */
@@ -1355,7 +1359,8 @@ nicvf_sq_add_hdr_subdesc(struct nicvf *nic, struct snd_queue *sq, int qentry,
 
 	/* Offload checksum calculation to HW */
 	if (skb->ip_summed == CHECKSUM_PARTIAL) {
-		hdr->csum_l3 = 1; /* Enable IP csum calculation */
+		if (ip.v4->version == 4)
+			hdr->csum_l3 = 1; /* Enable IP csum calculation */
 		hdr->l3_offset = skb_network_offset(skb);
 		hdr->l4_offset = skb_transport_offset(skb);
 
@@ -1377,12 +1382,33 @@ nicvf_sq_add_hdr_subdesc(struct nicvf *nic, struct snd_queue *sq, int qentry,
 
 	if (nic->hw_tso && skb_shinfo(skb)->gso_size) {
 		hdr->tso = 1;
-		hdr->tso_start = skb_transport_offset(skb) + tcp_hdrlen(skb);
+		hdr->tso_start = skb_tcp_all_headers(skb);
 		hdr->tso_max_paysize = skb_shinfo(skb)->gso_size;
 		/* For non-tunneled pkts, point this to L2 ethertype */
 		hdr->inner_l3_offset = skb_network_offset(skb) - 2;
 		this_cpu_inc(nic->pnicvf->drv_stats->tx_tso);
 	}
+
+	/* Check if hw timestamp is requested */
+	if (!(skb_shinfo(skb)->tx_flags & SKBTX_HW_TSTAMP))
+		return;
+
+	/* Tx timestamping not supported along with TSO, so ignore request */
+	if (skb_shinfo(skb)->gso_size)
+		return;
+
+	/* HW supports only a single outstanding packet to timestamp */
+	if (!atomic_add_unless(&nic->pnicvf->tx_ptp_skbs, 1, 1))
+		return;
+
+	/* Mark the SKB for later reference */
+	skb_shinfo(skb)->tx_flags |= SKBTX_IN_PROGRESS;
+
+	/* Finally enable timestamp generation
+	 * Since 'post_cqe' is also set, two CQEs will be posted
+	 * for this packet i.e CQE_TYPE_SEND and CQE_TYPE_SEND_PTP.
+	 */
+	hdr->tstmp = 1;
 }
 
 /* SQ GATHER subdescriptor
@@ -1444,6 +1470,8 @@ static inline void nicvf_sq_doorbell(struct nicvf *nic, struct sk_buff *skb,
 
 	netdev_tx_sent_queue(txq, skb->len);
 
+	skb_tx_timestamp(skb);
+
 	/* make sure all memory stores are done before ringing doorbell */
 	smp_wmb();
 
@@ -1462,9 +1490,10 @@ static int nicvf_sq_append_tso(struct nicvf *nic, struct snd_queue *sq,
 	int seg_subdescs = 0, desc_cnt = 0;
 	int seg_len, total_len, data_left;
 	int hdr_qentry = qentry;
-	int hdr_len = skb_transport_offset(skb) + tcp_hdrlen(skb);
+	int hdr_len;
+
+	hdr_len = tso_start(skb, &tso);
 
-	tso_start(skb, &tso);
 	total_len = skb->len - hdr_len;
 	while (total_len > 0) {
 		char *hdr;
@@ -1560,15 +1589,13 @@ int nicvf_sq_append_skb(struct nicvf *nic, struct snd_queue *sq,
 		goto doorbell;
 
 	for (i = 0; i < skb_shinfo(skb)->nr_frags; i++) {
-		const struct skb_frag_struct *frag;
-
-		frag = &skb_shinfo(skb)->frags[i];
+		const skb_frag_t *frag = &skb_shinfo(skb)->frags[i];
 
 		qentry = nicvf_get_nxt_sqentry(sq, qentry);
 		size = skb_frag_size(frag);
 		dma_addr = dma_map_page_attrs(&nic->pdev->dev,
 					      skb_frag_page(frag),
-					      frag->page_offset, size,
+					      skb_frag_off(frag), size,
 					      DMA_TO_DEVICE,
 					      DMA_ATTR_SKIP_CPU_SYNC);
 		if (dma_mapping_error(&nic->pdev->dev, dma_addr)) {
@@ -1811,11 +1838,9 @@ void nicvf_update_sq_stats(struct nicvf *nic, int sq_idx)
 /* Check for errors in the receive cmp.queue entry */
 int nicvf_check_cqe_rx_errs(struct nicvf *nic, struct cqe_rx_t *cqe_rx)
 {
-	if (netif_msg_rx_err(nic))
-		netdev_err(nic->netdev,
-			   "%s: RX error CQE err_level 0x%x err_opcode 0x%x\n",
-			   nic->netdev->name,
-			   cqe_rx->err_level, cqe_rx->err_opcode);
+	netif_err(nic, rx_err, nic->netdev,
+		  "RX error CQE err_level 0x%x err_opcode 0x%x\n",
+		  cqe_rx->err_level, cqe_rx->err_opcode);
 
 	switch (cqe_rx->err_opcode) {
 	case CQ_RX_ERROP_RE_PARTIAL:
diff --git a/drivers/net/ethernet/cavium/thunder/nicvf_queues.h b/drivers/net/ethernet/cavium/thunder/nicvf_queues.h
index 57858522c33c..b7531041c56d 100644
--- a/drivers/net/ethernet/cavium/thunder/nicvf_queues.h
+++ b/drivers/net/ethernet/cavium/thunder/nicvf_queues.h
@@ -1,9 +1,6 @@
+/* SPDX-License-Identifier: GPL-2.0-only */
 /*
  * Copyright (C) 2015 Cavium, Inc.
- *
- * This program is free software; you can redistribute it and/or modify it
- * under the terms of version 2 of the GNU General Public License
- * as published by the Free Software Foundation.
  */
 
 #ifndef NICVF_QUEUES_H
@@ -11,6 +8,7 @@
 
 #include <linux/netdevice.h>
 #include <linux/iommu.h>
+#include <net/xdp.h>
 #include "q_struct.h"
 
 #define MAX_QUEUE_SET			128
@@ -102,8 +100,8 @@
  * RED accepts pkt if unused CQE < 2304 & >= 2560
  * DROPs pkts if unused CQE < 2304
  */
-#define RQ_PASS_CQ_LVL         192ULL
-#define RQ_DROP_CQ_LVL         184ULL
+#define RQ_PASS_CQ_LVL         224ULL
+#define RQ_DROP_CQ_LVL         216ULL
 
 /* RED and Backpressure levels of RBDR for pkt reception
  * For RBDR, level is a measure of fullness i.e 0x0 means empty
@@ -208,7 +206,7 @@ struct rx_tx_queue_stats {
 struct q_desc_mem {
 	dma_addr_t	dma;
 	u64		size;
-	u16		q_len;
+	u32		q_len;
 	dma_addr_t	phys_base;
 	void		*base;
 	void		*unalign_base;
@@ -251,6 +249,7 @@ struct rcv_queue {
 	u8		start_qs_rbdr_idx; /* RBDR idx in the above QS */
 	u8		caching;
 	struct		rx_tx_queue_stats stats;
+	struct xdp_rxq_info xdp_rxq;
 } ____cacheline_aligned_in_smp;
 
 struct cmp_queue {
@@ -277,7 +276,6 @@ struct snd_queue {
 	u16		xdp_free_cnt;
 	bool		is_xdp;
 
-#define	TSO_HEADER_SIZE	128
 	/* For TSO segment's header */
 	char		*tso_hdrs;
 	dma_addr_t	tso_hdrs_phys;
@@ -350,7 +348,7 @@ void nicvf_xdp_sq_doorbell(struct nicvf *nic, struct snd_queue *sq, int sq_num);
 
 struct sk_buff *nicvf_get_rcv_skb(struct nicvf *nic,
 				  struct cqe_rx_t *cqe_rx, bool xdp);
-void nicvf_rbdr_task(unsigned long data);
+void nicvf_rbdr_task(struct tasklet_struct *t);
 void nicvf_rbdr_work(struct work_struct *work);
 
 void nicvf_enable_intr(struct nicvf *nic, int int_type, int q_idx);
@@ -361,8 +359,6 @@ int nicvf_is_intr_enabled(struct nicvf *nic, int int_type, int q_idx);
 /* Register access APIs */
 void nicvf_reg_write(struct nicvf *nic, u64 offset, u64 val);
 u64  nicvf_reg_read(struct nicvf *nic, u64 offset);
-void nicvf_qset_reg_write(struct nicvf *nic, u64 offset, u64 val);
-u64 nicvf_qset_reg_read(struct nicvf *nic, u64 offset);
 void nicvf_queue_reg_write(struct nicvf *nic, u64 offset,
 			   u64 qidx, u64 val);
 u64  nicvf_queue_reg_read(struct nicvf *nic,
diff --git a/drivers/net/ethernet/cavium/thunder/q_struct.h b/drivers/net/ethernet/cavium/thunder/q_struct.h
index e47205aa87ea..0df115d42612 100644
--- a/drivers/net/ethernet/cavium/thunder/q_struct.h
+++ b/drivers/net/ethernet/cavium/thunder/q_struct.h
@@ -1,12 +1,9 @@
+/* SPDX-License-Identifier: GPL-2.0-only */
 /*
  * This file contains HW queue descriptor formats, config register
  * structures etc
  *
  * Copyright (C) 2015 Cavium, Inc.
- *
- * This program is free software; you can redistribute it and/or modify it
- * under the terms of version 2 of the GNU General Public License
- * as published by the Free Software Foundation.
  */
 
 #ifndef Q_STRUCT_H
diff --git a/drivers/net/ethernet/cavium/thunder/thunder_bgx.c b/drivers/net/ethernet/cavium/thunder/thunder_bgx.c
index a0ca68ce3fbb..9efb60842ad1 100644
--- a/drivers/net/ethernet/cavium/thunder/thunder_bgx.c
+++ b/drivers/net/ethernet/cavium/thunder/thunder_bgx.c
@@ -1,9 +1,6 @@
+// SPDX-License-Identifier: GPL-2.0-only
 /*
  * Copyright (C) 2015 Cavium, Inc.
- *
- * This program is free software; you can redistribute it and/or modify it
- * under the terms of version 2 of the GNU General Public License
- * as published by the Free Software Foundation.
  */
 
 #include <linux/acpi.h>
@@ -21,12 +18,34 @@
 #include "nic.h"
 #include "thunder_bgx.h"
 
-#define DRV_NAME	"thunder-BGX"
+#define DRV_NAME	"thunder_bgx"
 #define DRV_VERSION	"1.0"
 
+/* RX_DMAC_CTL configuration */
+enum MCAST_MODE {
+		MCAST_MODE_REJECT = 0x0,
+		MCAST_MODE_ACCEPT = 0x1,
+		MCAST_MODE_CAM_FILTER = 0x2,
+		RSVD = 0x3
+};
+
+#define BCAST_ACCEPT      BIT(0)
+#define CAM_ACCEPT        BIT(3)
+#define MCAST_MODE_MASK   0x3
+#define BGX_MCAST_MODE(x) (x << 1)
+
+struct dmac_map {
+	u64                     vf_map;
+	u64                     dmac;
+};
+
 struct lmac {
 	struct bgx		*bgx;
-	int			dmac;
+	/* actual number of DMACs configured */
+	u8			dmacs_cfg;
+	/* overal number of possible DMACs could be configured per LMAC */
+	u8                      dmacs_count;
+	struct dmac_map         *dmacs; /* DMAC:VFs tracking filter array */
 	u8			mac[ETH_ALEN];
 	u8                      lmac_type;
 	u8                      lane_to_sds;
@@ -35,7 +54,7 @@ struct lmac {
 	bool			link_up;
 	int			lmacid; /* ID within BGX */
 	int			lmacid_bd; /* ID on board */
-	struct net_device       netdev;
+	struct net_device       *netdev;
 	struct phy_device       *phydev;
 	unsigned int            last_duplex;
 	unsigned int            last_link;
@@ -223,6 +242,163 @@ void bgx_set_lmac_mac(int node, int bgx_idx, int lmacid, const u8 *mac)
 }
 EXPORT_SYMBOL(bgx_set_lmac_mac);
 
+static void bgx_flush_dmac_cam_filter(struct bgx *bgx, int lmacid)
+{
+	struct lmac *lmac = NULL;
+	u8  idx = 0;
+
+	lmac = &bgx->lmac[lmacid];
+	/* reset CAM filters */
+	for (idx = 0; idx < lmac->dmacs_count; idx++)
+		bgx_reg_write(bgx, 0, BGX_CMR_RX_DMACX_CAM +
+			      ((lmacid * lmac->dmacs_count) + idx) *
+			      sizeof(u64), 0);
+}
+
+static void bgx_lmac_remove_filters(struct lmac *lmac, u8 vf_id)
+{
+	int i = 0;
+
+	if (!lmac)
+		return;
+
+	/* We've got reset filters request from some of attached VF, while the
+	 * others might want to keep their configuration. So in this case lets
+	 * iterate over all of configured filters and decrease number of
+	 * referencies. if some addresses get zero refs remove them from list
+	 */
+	for (i = lmac->dmacs_cfg - 1; i >= 0; i--) {
+		lmac->dmacs[i].vf_map &= ~BIT_ULL(vf_id);
+		if (!lmac->dmacs[i].vf_map) {
+			lmac->dmacs_cfg--;
+			lmac->dmacs[i].dmac = 0;
+			lmac->dmacs[i].vf_map = 0;
+		}
+	}
+}
+
+static int bgx_lmac_save_filter(struct lmac *lmac, u64 dmac, u8 vf_id)
+{
+	u8 i = 0;
+
+	if (!lmac)
+		return -1;
+
+	/* At the same time we could have several VFs 'attached' to some
+	 * particular LMAC, and each VF is represented as network interface
+	 * for kernel. So from user perspective it should be possible to
+	 * manipulate with its' (VF) receive modes. However from PF
+	 * driver perspective we need to keep track of filter configurations
+	 * for different VFs to prevent filter values dupes
+	 */
+	for (i = 0; i < lmac->dmacs_cfg; i++) {
+		if (lmac->dmacs[i].dmac == dmac) {
+			lmac->dmacs[i].vf_map |= BIT_ULL(vf_id);
+			return -1;
+		}
+	}
+
+	if (!(lmac->dmacs_cfg < lmac->dmacs_count))
+		return -1;
+
+	/* keep it for further tracking */
+	lmac->dmacs[lmac->dmacs_cfg].dmac = dmac;
+	lmac->dmacs[lmac->dmacs_cfg].vf_map = BIT_ULL(vf_id);
+	lmac->dmacs_cfg++;
+	return 0;
+}
+
+static int bgx_set_dmac_cam_filter_mac(struct bgx *bgx, int lmacid,
+				       u64 cam_dmac, u8 idx)
+{
+	struct lmac *lmac = NULL;
+	u64 cfg = 0;
+
+	/* skip zero addresses as meaningless */
+	if (!cam_dmac || !bgx)
+		return -1;
+
+	lmac = &bgx->lmac[lmacid];
+
+	/* configure DCAM filtering for designated LMAC */
+	cfg = RX_DMACX_CAM_LMACID(lmacid & LMAC_ID_MASK) |
+		RX_DMACX_CAM_EN | cam_dmac;
+	bgx_reg_write(bgx, 0, BGX_CMR_RX_DMACX_CAM +
+		      ((lmacid * lmac->dmacs_count) + idx) * sizeof(u64), cfg);
+	return 0;
+}
+
+void bgx_set_dmac_cam_filter(int node, int bgx_idx, int lmacid,
+			     u64 cam_dmac, u8 vf_id)
+{
+	struct bgx *bgx = get_bgx(node, bgx_idx);
+	struct lmac *lmac = NULL;
+
+	if (!bgx)
+		return;
+
+	lmac = &bgx->lmac[lmacid];
+
+	if (!cam_dmac)
+		cam_dmac = ether_addr_to_u64(lmac->mac);
+
+	/* since we might have several VFs attached to particular LMAC
+	 * and kernel could call mcast config for each of them with the
+	 * same MAC, check if requested MAC is already in filtering list and
+	 * updare/prepare list of MACs to be applied later to HW filters
+	 */
+	bgx_lmac_save_filter(lmac, cam_dmac, vf_id);
+}
+EXPORT_SYMBOL(bgx_set_dmac_cam_filter);
+
+void bgx_set_xcast_mode(int node, int bgx_idx, int lmacid, u8 mode)
+{
+	struct bgx *bgx = get_bgx(node, bgx_idx);
+	struct lmac *lmac = NULL;
+	u64 cfg = 0;
+	u8 i = 0;
+
+	if (!bgx)
+		return;
+
+	lmac = &bgx->lmac[lmacid];
+
+	cfg = bgx_reg_read(bgx, lmacid, BGX_CMRX_RX_DMAC_CTL);
+	if (mode & BGX_XCAST_BCAST_ACCEPT)
+		cfg |= BCAST_ACCEPT;
+	else
+		cfg &= ~BCAST_ACCEPT;
+
+	/* disable all MCASTs and DMAC filtering */
+	cfg &= ~(CAM_ACCEPT | BGX_MCAST_MODE(MCAST_MODE_MASK));
+
+	/* check requested bits and set filtergin mode appropriately */
+	if (mode & (BGX_XCAST_MCAST_ACCEPT)) {
+		cfg |= (BGX_MCAST_MODE(MCAST_MODE_ACCEPT));
+	} else if (mode & BGX_XCAST_MCAST_FILTER) {
+		cfg |= (BGX_MCAST_MODE(MCAST_MODE_CAM_FILTER) | CAM_ACCEPT);
+		for (i = 0; i < lmac->dmacs_cfg; i++)
+			bgx_set_dmac_cam_filter_mac(bgx, lmacid,
+						    lmac->dmacs[i].dmac, i);
+	}
+	bgx_reg_write(bgx, lmacid, BGX_CMRX_RX_DMAC_CTL, cfg);
+}
+EXPORT_SYMBOL(bgx_set_xcast_mode);
+
+void bgx_reset_xcast_mode(int node, int bgx_idx, int lmacid, u8 vf_id)
+{
+	struct bgx *bgx = get_bgx(node, bgx_idx);
+
+	if (!bgx)
+		return;
+
+	bgx_lmac_remove_filters(&bgx->lmac[lmacid], vf_id);
+	bgx_flush_dmac_cam_filter(bgx, lmacid);
+	bgx_set_xcast_mode(node, bgx_idx, lmacid,
+			   (BGX_XCAST_BCAST_ACCEPT | BGX_XCAST_MCAST_ACCEPT));
+}
+EXPORT_SYMBOL(bgx_reset_xcast_mode);
+
 void bgx_lmac_rx_tx_enable(int node, int bgx_idx, int lmacid, bool enable)
 {
 	struct bgx *bgx = get_bgx(node, bgx_idx);
@@ -234,10 +410,19 @@ void bgx_lmac_rx_tx_enable(int node, int bgx_idx, int lmacid, bool enable)
 	lmac = &bgx->lmac[lmacid];
 
 	cfg = bgx_reg_read(bgx, lmacid, BGX_CMRX_CFG);
-	if (enable)
+	if (enable) {
 		cfg |= CMR_PKT_RX_EN | CMR_PKT_TX_EN;
-	else
+
+		/* enable TX FIFO Underflow interrupt */
+		bgx_reg_modify(bgx, lmacid, BGX_GMP_GMI_TXX_INT_ENA_W1S,
+			       GMI_TXX_INT_UNDFLW);
+	} else {
 		cfg &= ~(CMR_PKT_RX_EN | CMR_PKT_TX_EN);
+
+		/* Disable TX FIFO Underflow interrupt */
+		bgx_reg_modify(bgx, lmacid, BGX_GMP_GMI_TXX_INT_ENA_W1C,
+			       GMI_TXX_INT_UNDFLW);
+	}
 	bgx_reg_write(bgx, lmacid, BGX_CMRX_CFG, cfg);
 
 	if (bgx->is_rgx)
@@ -245,6 +430,35 @@ void bgx_lmac_rx_tx_enable(int node, int bgx_idx, int lmacid, bool enable)
 }
 EXPORT_SYMBOL(bgx_lmac_rx_tx_enable);
 
+/* Enables or disables timestamp insertion by BGX for Rx packets */
+void bgx_config_timestamping(int node, int bgx_idx, int lmacid, bool enable)
+{
+	struct bgx *bgx = get_bgx(node, bgx_idx);
+	struct lmac *lmac;
+	u64 csr_offset, cfg;
+
+	if (!bgx)
+		return;
+
+	lmac = &bgx->lmac[lmacid];
+
+	if (lmac->lmac_type == BGX_MODE_SGMII ||
+	    lmac->lmac_type == BGX_MODE_QSGMII ||
+	    lmac->lmac_type == BGX_MODE_RGMII)
+		csr_offset = BGX_GMP_GMI_RXX_FRM_CTL;
+	else
+		csr_offset = BGX_SMUX_RX_FRM_CTL;
+
+	cfg = bgx_reg_read(bgx, lmacid, csr_offset);
+
+	if (enable)
+		cfg |= BGX_PKT_RX_PTP_EN;
+	else
+		cfg &= ~BGX_PKT_RX_PTP_EN;
+	bgx_reg_write(bgx, lmacid, csr_offset, cfg);
+}
+EXPORT_SYMBOL(bgx_config_timestamping);
+
 void bgx_lmac_get_pfc(int node, int bgx_idx, int lmacid, void *pause)
 {
 	struct pfc *pfc = (struct pfc *)pause;
@@ -292,11 +506,30 @@ static void bgx_sgmii_change_link_state(struct lmac *lmac)
 	u64 cmr_cfg;
 	u64 port_cfg = 0;
 	u64 misc_ctl = 0;
+	bool tx_en, rx_en;
 
 	cmr_cfg = bgx_reg_read(bgx, lmac->lmacid, BGX_CMRX_CFG);
-	cmr_cfg &= ~CMR_EN;
+	tx_en = cmr_cfg & CMR_PKT_TX_EN;
+	rx_en = cmr_cfg & CMR_PKT_RX_EN;
+	cmr_cfg &= ~(CMR_PKT_RX_EN | CMR_PKT_TX_EN);
 	bgx_reg_write(bgx, lmac->lmacid, BGX_CMRX_CFG, cmr_cfg);
 
+	/* Wait for BGX RX to be idle */
+	if (bgx_poll_reg(bgx, lmac->lmacid, BGX_GMP_GMI_PRTX_CFG,
+			 GMI_PORT_CFG_RX_IDLE, false)) {
+		dev_err(&bgx->pdev->dev, "BGX%d LMAC%d GMI RX not idle\n",
+			bgx->bgx_id, lmac->lmacid);
+		return;
+	}
+
+	/* Wait for BGX TX to be idle */
+	if (bgx_poll_reg(bgx, lmac->lmacid, BGX_GMP_GMI_PRTX_CFG,
+			 GMI_PORT_CFG_TX_IDLE, false)) {
+		dev_err(&bgx->pdev->dev, "BGX%d LMAC%d GMI TX not idle\n",
+			bgx->bgx_id, lmac->lmacid);
+		return;
+	}
+
 	port_cfg = bgx_reg_read(bgx, lmac->lmacid, BGX_GMP_GMI_PRTX_CFG);
 	misc_ctl = bgx_reg_read(bgx, lmac->lmacid, BGX_GMP_PCS_MISCX_CTL);
 
@@ -347,10 +580,8 @@ static void bgx_sgmii_change_link_state(struct lmac *lmac)
 	bgx_reg_write(bgx, lmac->lmacid, BGX_GMP_PCS_MISCX_CTL, misc_ctl);
 	bgx_reg_write(bgx, lmac->lmacid, BGX_GMP_GMI_PRTX_CFG, port_cfg);
 
-	port_cfg = bgx_reg_read(bgx, lmac->lmacid, BGX_GMP_GMI_PRTX_CFG);
-
-	/* Re-enable lmac */
-	cmr_cfg |= CMR_EN;
+	/* Restore CMR config settings */
+	cmr_cfg |= (rx_en ? CMR_PKT_RX_EN : 0) | (tx_en ? CMR_PKT_TX_EN : 0);
 	bgx_reg_write(bgx, lmac->lmacid, BGX_CMRX_CFG, cmr_cfg);
 
 	if (bgx->is_rgx && (cmr_cfg & (CMR_PKT_RX_EN | CMR_PKT_TX_EN)))
@@ -359,13 +590,12 @@ static void bgx_sgmii_change_link_state(struct lmac *lmac)
 
 static void bgx_lmac_handler(struct net_device *netdev)
 {
-	struct lmac *lmac = container_of(netdev, struct lmac, netdev);
 	struct phy_device *phydev;
+	struct lmac *lmac, **priv;
 	int link_changed = 0;
 
-	if (!lmac)
-		return;
-
+	priv = netdev_priv(netdev);
+	lmac = *priv;
 	phydev = lmac->phydev;
 
 	if (!phydev->link && lmac->last_link)
@@ -422,18 +652,6 @@ u64 bgx_get_tx_stats(int node, int bgx_idx, int lmac, int idx)
 }
 EXPORT_SYMBOL(bgx_get_tx_stats);
 
-static void bgx_flush_dmac_addrs(struct bgx *bgx, int lmac)
-{
-	u64 offset;
-
-	while (bgx->lmac[lmac].dmac > 0) {
-		offset = ((bgx->lmac[lmac].dmac - 1) * sizeof(u64)) +
-			(lmac * MAX_DMAC_PER_LMAC * sizeof(u64));
-		bgx_reg_write(bgx, 0, BGX_CMR_RX_DMACX_CAM + offset, 0);
-		bgx->lmac[lmac].dmac--;
-	}
-}
-
 /* Configure BGX LMAC in internal loopback mode */
 void bgx_lmac_internal_loopback(int node, int bgx_idx,
 				int lmac_idx, bool enable)
@@ -749,13 +967,13 @@ static void bgx_poll_for_sgmii_link(struct lmac *lmac)
 	lmac->last_duplex = (an_result >> 1) & 0x1;
 	switch (speed) {
 	case 0:
-		lmac->last_speed = 10;
+		lmac->last_speed = SPEED_10;
 		break;
 	case 1:
-		lmac->last_speed = 100;
+		lmac->last_speed = SPEED_100;
 		break;
 	case 2:
-		lmac->last_speed = 1000;
+		lmac->last_speed = SPEED_1000;
 		break;
 	default:
 		lmac->link_up = false;
@@ -797,14 +1015,14 @@ static void bgx_poll_for_link(struct work_struct *work)
 
 	if ((spu_link & SPU_STATUS1_RCV_LNK) &&
 	    !(smu_link & SMU_RX_CTL_STATUS)) {
-		lmac->link_up = 1;
+		lmac->link_up = true;
 		if (lmac->lmac_type == BGX_MODE_XLAUI)
-			lmac->last_speed = 40000;
+			lmac->last_speed = SPEED_40000;
 		else
-			lmac->last_speed = 10000;
-		lmac->last_duplex = 1;
+			lmac->last_speed = SPEED_10000;
+		lmac->last_duplex = DUPLEX_FULL;
 	} else {
-		lmac->link_up = 0;
+		lmac->link_up = false;
 		lmac->last_speed = SPEED_UNKNOWN;
 		lmac->last_duplex = DUPLEX_UNKNOWN;
 	}
@@ -813,7 +1031,7 @@ static void bgx_poll_for_link(struct work_struct *work)
 		if (lmac->link_up) {
 			if (bgx_xaui_check_link(lmac)) {
 				/* Errors, clear link_up state */
-				lmac->link_up = 0;
+				lmac->link_up = false;
 				lmac->last_speed = SPEED_UNKNOWN;
 				lmac->last_duplex = DUPLEX_UNKNOWN;
 			}
@@ -829,7 +1047,7 @@ static int phy_interface_mode(u8 lmac_type)
 	if (lmac_type == BGX_MODE_QSGMII)
 		return PHY_INTERFACE_MODE_QSGMII;
 	if (lmac_type == BGX_MODE_RGMII)
-		return PHY_INTERFACE_MODE_RGMII;
+		return PHY_INTERFACE_MODE_RGMII_RXID;
 
 	return PHY_INTERFACE_MODE_SGMII;
 }
@@ -845,11 +1063,11 @@ static int bgx_lmac_enable(struct bgx *bgx, u8 lmacid)
 	if ((lmac->lmac_type == BGX_MODE_SGMII) ||
 	    (lmac->lmac_type == BGX_MODE_QSGMII) ||
 	    (lmac->lmac_type == BGX_MODE_RGMII)) {
-		lmac->is_sgmii = 1;
+		lmac->is_sgmii = true;
 		if (bgx_lmac_sgmii_init(bgx, lmac))
 			return -1;
 	} else {
-		lmac->is_sgmii = 0;
+		lmac->is_sgmii = false;
 		if (bgx_lmac_xaui_init(bgx, lmac))
 			return -1;
 	}
@@ -866,6 +1084,13 @@ static int bgx_lmac_enable(struct bgx *bgx, u8 lmacid)
 		bgx_reg_write(bgx, lmacid, BGX_SMUX_TX_MIN_PKT, 60 + 4);
 	}
 
+	/* actual number of filters available to exact LMAC */
+	lmac->dmacs_count = (RX_DMAC_COUNT / bgx->lmac_count);
+	lmac->dmacs = kcalloc(lmac->dmacs_count, sizeof(*lmac->dmacs),
+			      GFP_KERNEL);
+	if (!lmac->dmacs)
+		return -ENOMEM;
+
 	/* Enable lmac */
 	bgx_reg_modify(bgx, lmacid, BGX_CMRX_CFG, CMR_EN);
 
@@ -885,26 +1110,25 @@ static int bgx_lmac_enable(struct bgx *bgx, u8 lmacid)
 			} else {
 				/* Default to below link speed and duplex */
 				lmac->link_up = true;
-				lmac->last_speed = 1000;
-				lmac->last_duplex = 1;
+				lmac->last_speed = SPEED_1000;
+				lmac->last_duplex = DUPLEX_FULL;
 				bgx_sgmii_change_link_state(lmac);
 				return 0;
 			}
 		}
 		lmac->phydev->dev_flags = 0;
 
-		if (phy_connect_direct(&lmac->netdev, lmac->phydev,
+		if (phy_connect_direct(lmac->netdev, lmac->phydev,
 				       bgx_lmac_handler,
 				       phy_interface_mode(lmac->lmac_type)))
 			return -ENODEV;
 
-		phy_start_aneg(lmac->phydev);
+		phy_start(lmac->phydev);
 		return 0;
 	}
 
 poll:
-	lmac->check_link = alloc_workqueue("check_link", WQ_UNBOUND |
-					   WQ_MEM_RECLAIM, 1);
+	lmac->check_link = alloc_ordered_workqueue("check_link", WQ_MEM_RECLAIM);
 	if (!lmac->check_link)
 		return -ENOMEM;
 	INIT_DELAYED_WORK(&lmac->dwork, bgx_poll_for_link);
@@ -952,7 +1176,8 @@ static void bgx_lmac_disable(struct bgx *bgx, u8 lmacid)
 	cfg &= ~CMR_EN;
 	bgx_reg_write(bgx, lmacid, BGX_CMRX_CFG, cfg);
 
-	bgx_flush_dmac_addrs(bgx, lmacid);
+	bgx_flush_dmac_cam_filter(bgx, lmacid);
+	kfree(lmac->dmacs);
 
 	if ((lmac->lmac_type != BGX_MODE_XFI) &&
 	    (lmac->lmac_type != BGX_MODE_XLAUI) &&
@@ -996,7 +1221,7 @@ static void bgx_init_hw(struct bgx *bgx)
 
 	/* Disable MAC steering (NCSI traffic) */
 	for (i = 0; i < RX_TRAFFIC_STEER_RULE_COUNT; i++)
-		bgx_reg_write(bgx, 0, BGX_CMR_RX_STREERING + (i * 8), 0x00);
+		bgx_reg_write(bgx, 0, BGX_CMR_RX_STEERING + (i * 8), 0x00);
 }
 
 static u8 bgx_get_lane2sds_cfg(struct bgx *bgx, struct lmac *lmac)
@@ -1008,7 +1233,7 @@ static void bgx_print_qlm_mode(struct bgx *bgx, u8 lmacid)
 {
 	struct device *dev = &bgx->pdev->dev;
 	struct lmac *lmac;
-	char str[20];
+	char str[27];
 
 	if (!bgx->is_dlm && lmacid)
 		return;
@@ -1086,7 +1311,7 @@ static void lmac_set_training(struct bgx *bgx, struct lmac *lmac, int lmacid)
 {
 	if ((lmac->lmac_type != BGX_MODE_10G_KR) &&
 	    (lmac->lmac_type != BGX_MODE_40G_KR)) {
-		lmac->use_training = 0;
+		lmac->use_training = false;
 		return;
 	}
 
@@ -1165,22 +1390,16 @@ static int acpi_get_mac_address(struct device *dev, struct acpi_device *adev,
 	u8 mac[ETH_ALEN];
 	int ret;
 
-	ret = fwnode_property_read_u8_array(acpi_fwnode_handle(adev),
-					    "mac-address", mac, ETH_ALEN);
-	if (ret)
-		goto out;
-
-	if (!is_valid_ether_addr(mac)) {
+	ret = fwnode_get_mac_address(acpi_fwnode_handle(adev), mac);
+	if (ret) {
 		dev_err(dev, "MAC address invalid: %pM\n", mac);
-		ret = -EINVAL;
-		goto out;
+		return -EINVAL;
 	}
 
 	dev_info(dev, "MAC address set to: %pM\n", mac);
 
-	memcpy(dst, mac, ETH_ALEN);
-out:
-	return ret;
+	ether_addr_copy(dst, mac);
+	return 0;
 }
 
 /* Currently only sets the MAC address. */
@@ -1191,12 +1410,13 @@ static acpi_status bgx_acpi_register_phy(acpi_handle handle,
 	struct device *dev = &bgx->pdev->dev;
 	struct acpi_device *adev;
 
-	if (acpi_bus_get_device(handle, &adev))
+	adev = acpi_fetch_acpi_dev(handle);
+	if (!adev)
 		goto out;
 
 	acpi_get_mac_address(dev, adev, bgx->lmac[bgx->acpi_lmac_idx].mac);
 
-	SET_NETDEV_DEV(&bgx->lmac[bgx->acpi_lmac_idx].netdev, dev);
+	SET_NETDEV_DEV(bgx->lmac[bgx->acpi_lmac_idx].netdev, dev);
 
 	bgx->lmac[bgx->acpi_lmac_idx].lmacid = bgx->acpi_lmac_idx;
 	bgx->acpi_lmac_idx++; /* move to next LMAC */
@@ -1209,16 +1429,18 @@ static acpi_status bgx_acpi_match_id(acpi_handle handle, u32 lvl,
 {
 	struct acpi_buffer string = { ACPI_ALLOCATE_BUFFER, NULL };
 	struct bgx *bgx = context;
-	char bgx_sel[5];
+	char bgx_sel[7];
 
-	snprintf(bgx_sel, 5, "BGX%d", bgx->bgx_id);
+	snprintf(bgx_sel, sizeof(bgx_sel), "BGX%d", bgx->bgx_id);
 	if (ACPI_FAILURE(acpi_get_name(handle, ACPI_SINGLE_NAME, &string))) {
 		pr_warn("Invalid link device\n");
 		return AE_OK;
 	}
 
-	if (strncmp(string.pointer, bgx_sel, 4))
+	if (strncmp(string.pointer, bgx_sel, 4)) {
+		kfree(string.pointer);
 		return AE_OK;
+	}
 
 	acpi_walk_namespace(ACPI_TYPE_DEVICE, handle, 1,
 			    bgx_acpi_register_phy, NULL, bgx, NULL);
@@ -1253,7 +1475,6 @@ static int bgx_init_of_phy(struct bgx *bgx)
 	device_for_each_child_node(&bgx->pdev->dev, fwn) {
 		struct phy_device *pd;
 		struct device_node *phy_np;
-		const char *mac;
 
 		/* Should always be an OF node.  But if it is not, we
 		 * cannot handle it, so exit the loop.
@@ -1262,11 +1483,9 @@ static int bgx_init_of_phy(struct bgx *bgx)
 		if (!node)
 			break;
 
-		mac = of_get_mac_address(node);
-		if (mac)
-			ether_addr_copy(bgx->lmac[lmac].mac, mac);
+		of_get_mac_address(node, bgx->lmac[lmac].mac);
 
-		SET_NETDEV_DEV(&bgx->lmac[lmac].netdev, &bgx->pdev->dev);
+		SET_NETDEV_DEV(bgx->lmac[lmac].netdev, &bgx->pdev->dev);
 		bgx->lmac[lmac].lmacid = lmac;
 
 		phy_np = of_parse_phandle(node, "phy-handle", 0);
@@ -1274,13 +1493,17 @@ static int bgx_init_of_phy(struct bgx *bgx)
 		 * this cortina phy, for which there is no driver
 		 * support, ignore it.
 		 */
-		if (phy_np &&
-		    !of_device_is_compatible(phy_np, "cortina,cs4223-slice")) {
-			/* Wait until the phy drivers are available */
-			pd = of_phy_find_device(phy_np);
-			if (!pd)
-				goto defer;
-			bgx->lmac[lmac].phydev = pd;
+		if (phy_np) {
+			if (!of_device_is_compatible(phy_np, "cortina,cs4223-slice")) {
+				/* Wait until the phy drivers are available */
+				pd = of_phy_find_device(phy_np);
+				if (!pd) {
+					of_node_put(phy_np);
+					goto defer;
+				}
+				bgx->lmac[lmac].phydev = pd;
+			}
+			of_node_put(phy_np);
 		}
 
 		lmac++;
@@ -1296,11 +1519,11 @@ defer:
 	 * for phy devices we may have already found.
 	 */
 	while (lmac) {
+		lmac--;
 		if (bgx->lmac[lmac].phydev) {
 			put_device(&bgx->lmac[lmac].phydev->mdio.dev);
 			bgx->lmac[lmac].phydev = NULL;
 		}
-		lmac--;
 	}
 	of_node_put(node);
 	return -EPROBE_DEFER;
@@ -1323,6 +1546,48 @@ static int bgx_init_phy(struct bgx *bgx)
 	return bgx_init_of_phy(bgx);
 }
 
+static irqreturn_t bgx_intr_handler(int irq, void *data)
+{
+	struct bgx *bgx = (struct bgx *)data;
+	u64 status, val;
+	int lmac;
+
+	for (lmac = 0; lmac < bgx->lmac_count; lmac++) {
+		status = bgx_reg_read(bgx, lmac, BGX_GMP_GMI_TXX_INT);
+		if (status & GMI_TXX_INT_UNDFLW) {
+			pci_err(bgx->pdev, "BGX%d lmac%d UNDFLW\n",
+				bgx->bgx_id, lmac);
+			val = bgx_reg_read(bgx, lmac, BGX_CMRX_CFG);
+			val &= ~CMR_EN;
+			bgx_reg_write(bgx, lmac, BGX_CMRX_CFG, val);
+			val |= CMR_EN;
+			bgx_reg_write(bgx, lmac, BGX_CMRX_CFG, val);
+		}
+		/* clear interrupts */
+		bgx_reg_write(bgx, lmac, BGX_GMP_GMI_TXX_INT, status);
+	}
+
+	return IRQ_HANDLED;
+}
+
+static void bgx_register_intr(struct pci_dev *pdev)
+{
+	struct bgx *bgx = pci_get_drvdata(pdev);
+	int ret;
+
+	ret = pci_alloc_irq_vectors(pdev, BGX_LMAC_VEC_OFFSET,
+				    BGX_LMAC_VEC_OFFSET, PCI_IRQ_ALL_TYPES);
+	if (ret < 0) {
+		pci_err(pdev, "Req for #%d msix vectors failed\n",
+			BGX_LMAC_VEC_OFFSET);
+		return;
+	}
+	ret = pci_request_irq(pdev, GMPX_GMI_TX_INT, bgx_intr_handler, NULL,
+			      bgx, "BGX%d", bgx->bgx_id);
+	if (ret)
+		pci_free_irq(pdev, GMPX_GMI_TX_INT, bgx);
+}
+
 static int bgx_probe(struct pci_dev *pdev, const struct pci_device_id *ent)
 {
 	int err;
@@ -1338,17 +1603,16 @@ static int bgx_probe(struct pci_dev *pdev, const struct pci_device_id *ent)
 
 	pci_set_drvdata(pdev, bgx);
 
-	err = pci_enable_device(pdev);
+	err = pcim_enable_device(pdev);
 	if (err) {
-		dev_err(dev, "Failed to enable PCI device\n");
 		pci_set_drvdata(pdev, NULL);
-		return err;
+		return dev_err_probe(dev, err, "Failed to enable PCI device\n");
 	}
 
-	err = pci_request_regions(pdev, DRV_NAME);
+	err = pcim_request_all_regions(pdev, DRV_NAME);
 	if (err) {
 		dev_err(dev, "PCI request regions failed 0x%x\n", err);
-		goto err_disable_device;
+		goto err_zero_drv_data;
 	}
 
 	/* MAP configuration registers */
@@ -1356,7 +1620,7 @@ static int bgx_probe(struct pci_dev *pdev, const struct pci_device_id *ent)
 	if (!bgx->reg_base) {
 		dev_err(dev, "BGX: Cannot map CSR memory space, aborting\n");
 		err = -ENOMEM;
-		goto err_release_regions;
+		goto err_zero_drv_data;
 	}
 
 	set_max_bgx_per_node(pdev);
@@ -1386,12 +1650,31 @@ static int bgx_probe(struct pci_dev *pdev, const struct pci_device_id *ent)
 
 	bgx_get_qlm_mode(bgx);
 
+	for (lmac = 0; lmac < bgx->lmac_count; lmac++) {
+		struct lmac *lmacp, **priv;
+
+		lmacp = &bgx->lmac[lmac];
+		lmacp->netdev = alloc_netdev_dummy(sizeof(struct lmac *));
+
+		if (!lmacp->netdev) {
+			for (int i = 0; i < lmac; i++)
+				free_netdev(bgx->lmac[i].netdev);
+			err = -ENOMEM;
+			goto err_enable;
+		}
+
+		priv = netdev_priv(lmacp->netdev);
+		*priv = lmacp;
+	}
+
 	err = bgx_init_phy(bgx);
 	if (err)
 		goto err_enable;
 
 	bgx_init_hw(bgx);
 
+	bgx_register_intr(pdev);
+
 	/* Enable all LMACs */
 	for (lmac = 0; lmac < bgx->lmac_count; lmac++) {
 		err = bgx_lmac_enable(bgx, lmac);
@@ -1408,10 +1691,8 @@ static int bgx_probe(struct pci_dev *pdev, const struct pci_device_id *ent)
 
 err_enable:
 	bgx_vnic[bgx->bgx_id] = NULL;
-err_release_regions:
-	pci_release_regions(pdev);
-err_disable_device:
-	pci_disable_device(pdev);
+	pci_free_irq(pdev, GMPX_GMI_TX_INT, bgx);
+err_zero_drv_data:
 	pci_set_drvdata(pdev, NULL);
 	return err;
 }
@@ -1422,12 +1703,14 @@ static void bgx_remove(struct pci_dev *pdev)
 	u8 lmac;
 
 	/* Disable all LMACs */
-	for (lmac = 0; lmac < bgx->lmac_count; lmac++)
+	for (lmac = 0; lmac < bgx->lmac_count; lmac++) {
 		bgx_lmac_disable(bgx, lmac);
+		free_netdev(bgx->lmac[lmac].netdev);
+	}
+
+	pci_free_irq(pdev, GMPX_GMI_TX_INT, bgx);
 
 	bgx_vnic[bgx->bgx_id] = NULL;
-	pci_release_regions(pdev);
-	pci_disable_device(pdev);
 	pci_set_drvdata(pdev, NULL);
 }
 
diff --git a/drivers/net/ethernet/cavium/thunder/thunder_bgx.h b/drivers/net/ethernet/cavium/thunder/thunder_bgx.h
index 6b7fe6fdd13b..84f16ababaee 100644
--- a/drivers/net/ethernet/cavium/thunder/thunder_bgx.h
+++ b/drivers/net/ethernet/cavium/thunder/thunder_bgx.h
@@ -1,9 +1,6 @@
+/* SPDX-License-Identifier: GPL-2.0-only */
 /*
  * Copyright (C) 2015 Cavium, Inc.
- *
- * This program is free software; you can redistribute it and/or modify it
- * under the terms of version 2 of the GNU General Public License
- * as published by the Free Software Foundation.
  */
 
 #ifndef THUNDER_BGX_H
@@ -30,6 +27,7 @@
 #define    DEFAULT_PAUSE_TIME			0xFFFF
 
 #define	   BGX_ID_MASK				0x3
+#define	   LMAC_ID_MASK				0x3
 
 #define    MAX_DMAC_PER_LMAC_TNS_BYPASS_MODE	2
 
@@ -57,9 +55,9 @@
 #define BGX_CMRX_RX_FIFO_LEN		0x108
 #define BGX_CMR_RX_DMACX_CAM		0x200
 #define  RX_DMACX_CAM_EN			BIT_ULL(48)
-#define  RX_DMACX_CAM_LMACID(x)			(x << 49)
+#define  RX_DMACX_CAM_LMACID(x)			(((u64)x) << 49)
 #define  RX_DMAC_COUNT				32
-#define BGX_CMR_RX_STREERING		0x300
+#define BGX_CMR_RX_STEERING		0x300
 #define  RX_TRAFFIC_STEER_RULE_COUNT		8
 #define BGX_CMR_CHAN_MSK_AND		0x450
 #define BGX_CMR_BIST_STATUS		0x460
@@ -122,6 +120,8 @@
 #define  SPU_DBG_CTL_AN_NONCE_MCT_DIS		BIT_ULL(29)
 
 #define BGX_SMUX_RX_INT			0x20000
+#define BGX_SMUX_RX_FRM_CTL		0x20020
+#define  BGX_PKT_RX_PTP_EN			BIT_ULL(12)
 #define BGX_SMUX_RX_JABBER		0x20030
 #define BGX_SMUX_RX_CTL			0x20048
 #define  SMU_RX_CTL_STATUS			(3ull << 0)
@@ -170,6 +170,9 @@
 #define  GMI_PORT_CFG_DUPLEX			BIT_ULL(2)
 #define  GMI_PORT_CFG_SLOT_TIME			BIT_ULL(3)
 #define  GMI_PORT_CFG_SPEED_MSB			BIT_ULL(8)
+#define  GMI_PORT_CFG_RX_IDLE			BIT_ULL(12)
+#define  GMI_PORT_CFG_TX_IDLE			BIT_ULL(13)
+#define BGX_GMP_GMI_RXX_FRM_CTL		0x38028
 #define BGX_GMP_GMI_RXX_JABBER		0x38038
 #define BGX_GMP_GMI_TXX_THRESH		0x38210
 #define BGX_GMP_GMI_TXX_APPEND		0x38218
@@ -177,6 +180,15 @@
 #define BGX_GMP_GMI_TXX_BURST		0x38228
 #define BGX_GMP_GMI_TXX_MIN_PKT		0x38240
 #define BGX_GMP_GMI_TXX_SGMII_CTL	0x38300
+#define BGX_GMP_GMI_TXX_INT		0x38500
+#define BGX_GMP_GMI_TXX_INT_W1S		0x38508
+#define BGX_GMP_GMI_TXX_INT_ENA_W1C	0x38510
+#define BGX_GMP_GMI_TXX_INT_ENA_W1S	0x38518
+#define  GMI_TXX_INT_PTP_LOST			BIT_ULL(4)
+#define  GMI_TXX_INT_LATE_COL			BIT_ULL(3)
+#define  GMI_TXX_INT_XSDEF			BIT_ULL(2)
+#define  GMI_TXX_INT_XSCOL			BIT_ULL(1)
+#define  GMI_TXX_INT_UNDFLW			BIT_ULL(0)
 
 #define BGX_MSIX_VEC_0_29_ADDR		0x400000 /* +(0..29) << 4 */
 #define BGX_MSIX_VEC_0_29_CTL		0x400008
@@ -200,20 +212,14 @@
 #define LMAC_INTR_LINK_UP	BIT(0)
 #define LMAC_INTR_LINK_DOWN	BIT(1)
 
-/*  RX_DMAC_CTL configuration*/
-enum MCAST_MODE {
-		MCAST_MODE_REJECT,
-		MCAST_MODE_ACCEPT,
-		MCAST_MODE_CAM_FILTER,
-		RSVD
-};
-
-#define BCAST_ACCEPT	1
-#define CAM_ACCEPT	1
+#define BGX_XCAST_BCAST_ACCEPT  BIT(0)
+#define BGX_XCAST_MCAST_ACCEPT  BIT(1)
+#define BGX_XCAST_MCAST_FILTER  BIT(2)
 
-void octeon_mdiobus_force_mod_depencency(void);
+void bgx_set_dmac_cam_filter(int node, int bgx_idx, int lmacid, u64 mac, u8 vf);
+void bgx_reset_xcast_mode(int node, int bgx_idx, int lmacid, u8 vf);
+void bgx_set_xcast_mode(int node, int bgx_idx, int lmacid, u8 mode);
 void bgx_lmac_rx_tx_enable(int node, int bgx_idx, int lmacid, bool enable);
-void bgx_add_dmac_addr(u64 dmac, int node, int bgx_idx, int lmac);
 unsigned bgx_get_map(int node);
 int bgx_get_lmac_count(int node, int bgx);
 const u8 *bgx_get_lmac_mac(int node, int bgx_idx, int lmacid);
@@ -221,6 +227,7 @@ void bgx_set_lmac_mac(int node, int bgx_idx, int lmacid, const u8 *mac);
 void bgx_get_lmac_link_state(int node, int bgx_idx, int lmacid, void *status);
 void bgx_lmac_internal_loopback(int node, int bgx_idx,
 				int lmac_idx, bool enable);
+void bgx_config_timestamping(int node, int bgx_idx, int lmacid, bool enable);
 void bgx_lmac_get_pfc(int node, int bgx_idx, int lmacid, void *pause);
 void bgx_lmac_set_pfc(int node, int bgx_idx, int lmacid, void *pause);
 
diff --git a/drivers/net/ethernet/cavium/thunder/thunder_xcv.c b/drivers/net/ethernet/cavium/thunder/thunder_xcv.c
index 578c7f8f11bf..3ebb93792831 100644
--- a/drivers/net/ethernet/cavium/thunder/thunder_xcv.c
+++ b/drivers/net/ethernet/cavium/thunder/thunder_xcv.c
@@ -1,9 +1,6 @@
+// SPDX-License-Identifier: GPL-2.0-only
 /*
  * Copyright (C) 2016 Cavium, Inc.
- *
- * This program is free software; you can redistribute it and/or modify it
- * under the terms of version 2 of the GNU General Public License
- * as published by the Free Software Foundation.
  */
 
 #include <linux/acpi.h>
@@ -20,7 +17,7 @@
 #include "nic.h"
 #include "thunder_bgx.h"
 
-#define DRV_NAME	"thunder-xcv"
+#define DRV_NAME	"thunder_xcv"
 #define DRV_VERSION	"1.0"
 
 /* Register offsets */
diff --git a/drivers/net/ethernet/chelsio/Kconfig b/drivers/net/ethernet/chelsio/Kconfig
index 5713e83be08c..c931ec8cac40 100644
--- a/drivers/net/ethernet/chelsio/Kconfig
+++ b/drivers/net/ethernet/chelsio/Kconfig
@@ -1,3 +1,4 @@
+# SPDX-License-Identifier: GPL-2.0-only
 #
 # Chelsio device configuration
 #
@@ -6,7 +7,7 @@ config NET_VENDOR_CHELSIO
 	bool "Chelsio devices"
 	default y
 	depends on PCI
-	---help---
+	help
 	  If you have a network (Ethernet) card belonging to this class, say Y.
 
 	  Note that the answer to this question doesn't directly affect the
@@ -21,10 +22,11 @@ config CHELSIO_T1
 	depends on PCI
 	select CRC32
 	select MDIO
-	---help---
+	help
 	  This driver supports Chelsio gigabit and 10-gigabit
 	  Ethernet cards. More information about adapter features and
-	  performance tuning is in <file:Documentation/networking/cxgb.txt>.
+	  performance tuning is in
+	  <file:Documentation/networking/device_drivers/ethernet/chelsio/cxgb.rst>.
 
 	  For general information about Chelsio and our products, visit
 	  our website at <http://www.chelsio.com>.
@@ -40,7 +42,7 @@ config CHELSIO_T1
 config CHELSIO_T1_1G
 	bool "Chelsio gigabit Ethernet support"
 	depends on CHELSIO_T1
-	---help---
+	help
 	  Enables support for Chelsio's gigabit Ethernet PCI cards.  If you
 	  are using only 10G cards say 'N' here.
 
@@ -49,7 +51,7 @@ config CHELSIO_T3
 	depends on PCI && INET
 	select FW_LOADER
 	select MDIO
-	---help---
+	help
 	  This driver supports Chelsio T3-based gigabit and 10Gb Ethernet
 	  adapters.
 
@@ -66,10 +68,12 @@ config CHELSIO_T3
 
 config CHELSIO_T4
 	tristate "Chelsio Communications T4/T5/T6 Ethernet support"
-	depends on PCI && (IPV6 || IPV6=n)
+	depends on PCI && (IPV6 || IPV6=n) && (TLS || TLS=n)
+	depends on PTP_1588_CLOCK_OPTIONAL
 	select FW_LOADER
 	select MDIO
-	---help---
+	select ZLIB_DEFLATE
+	help
 	  This driver supports Chelsio T4, T5 & T6 based gigabit, 10Gb Ethernet
 	  adapter and T5/T6 based 40Gb and T6 based 25Gb, 50Gb and 100Gb
 	  Ethernet adapters.
@@ -89,7 +93,7 @@ config CHELSIO_T4_DCB
 	bool "Data Center Bridging (DCB) Support for Chelsio T4/T5/T6 cards"
 	default n
 	depends on CHELSIO_T4 && DCB
-	---help---
+	help
 	  Enable DCB support through rtNetlink interface.
 	  Say Y here if you want to enable Data Center Bridging (DCB) support
 	  in the driver.
@@ -100,7 +104,7 @@ config CHELSIO_T4_FCOE
 	bool "Fibre Channel over Ethernet (FCoE) Support for Chelsio T5 cards"
 	default n
 	depends on CHELSIO_T4 && CHELSIO_T4_DCB && FCOE
-	---help---
+	help
 	  Enable FCoE offload features.
 	  Say Y here if you want to enable Fibre Channel over Ethernet (FCoE) support
 	  in the driver.
@@ -110,7 +114,7 @@ config CHELSIO_T4_FCOE
 config CHELSIO_T4VF
 	tristate "Chelsio Communications T4/T5/T6 Virtual Function Ethernet support"
 	depends on PCI
-	---help---
+	help
 	  This driver supports Chelsio T4, T5 & T6 based gigabit, 10Gb Ethernet
 	  adapters and T5/T6 based 40Gb and T6 based 25Gb, 50Gb and 100Gb
 	  Ethernet adapters with PCI-E SR-IOV Virtual Functions.
@@ -128,7 +132,9 @@ config CHELSIO_T4VF
 
 config CHELSIO_LIB
 	tristate
-	---help---
+	help
 	Common library for Chelsio drivers.
 
+source "drivers/net/ethernet/chelsio/inline_crypto/Kconfig"
+
 endif # NET_VENDOR_CHELSIO
diff --git a/drivers/net/ethernet/chelsio/Makefile b/drivers/net/ethernet/chelsio/Makefile
index b6a5eec6ed8e..1a6fd8b2bb7d 100644
--- a/drivers/net/ethernet/chelsio/Makefile
+++ b/drivers/net/ethernet/chelsio/Makefile
@@ -1,3 +1,4 @@
+# SPDX-License-Identifier: GPL-2.0
 #
 # Makefile for the Chelsio network device drivers.
 #
@@ -7,3 +8,4 @@ obj-$(CONFIG_CHELSIO_T3) += cxgb3/
 obj-$(CONFIG_CHELSIO_T4) += cxgb4/
 obj-$(CONFIG_CHELSIO_T4VF) += cxgb4vf/
 obj-$(CONFIG_CHELSIO_LIB) += libcxgb/
+obj-$(CONFIG_CHELSIO_INLINE_CRYPTO) += inline_crypto/
diff --git a/drivers/net/ethernet/chelsio/cxgb/Makefile b/drivers/net/ethernet/chelsio/cxgb/Makefile
index 57a4b262fd3f..8008282a276f 100644
--- a/drivers/net/ethernet/chelsio/cxgb/Makefile
+++ b/drivers/net/ethernet/chelsio/cxgb/Makefile
@@ -1,3 +1,4 @@
+# SPDX-License-Identifier: GPL-2.0-only
 #
 # Chelsio T1 driver
 #
diff --git a/drivers/net/ethernet/chelsio/cxgb/common.h b/drivers/net/ethernet/chelsio/cxgb/common.h
index 94b9482f14a5..304bb282ab03 100644
--- a/drivers/net/ethernet/chelsio/cxgb/common.h
+++ b/drivers/net/ethernet/chelsio/cxgb/common.h
@@ -1,3 +1,4 @@
+/* SPDX-License-Identifier: GPL-2.0-only */
 /*****************************************************************************
  *                                                                           *
  * File: common.h                                                            *
@@ -6,16 +7,6 @@
  * Description:                                                              *
  *  part of the Chelsio 10Gb Ethernet Driver.                                *
  *                                                                           *
- * This program is free software; you can redistribute it and/or modify      *
- * it under the terms of the GNU General Public License, version 2, as       *
- * published by the Free Software Foundation.                                *
- *                                                                           *
- * You should have received a copy of the GNU General Public License along   *
- * with this program; if not, see <http://www.gnu.org/licenses/>.            *
- *                                                                           *
- * THIS SOFTWARE IS PROVIDED ``AS IS'' AND WITHOUT ANY EXPRESS OR IMPLIED    *
- * WARRANTIES, INCLUDING, WITHOUT LIMITATION, THE IMPLIED WARRANTIES OF      *
- * MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE.                     *
  *                                                                           *
  * http://www.chelsio.com                                                    *
  *                                                                           *
@@ -55,7 +46,6 @@
 
 #define DRV_DESCRIPTION "Chelsio 10Gb Ethernet Driver"
 #define DRV_NAME "cxgb"
-#define DRV_VERSION "2.2"
 
 #define CH_DEVICE(devid, ssid, idx) \
 	{ PCI_VENDOR_ID_CHELSIO, devid, PCI_ANY_ID, ssid, 0, 0, idx }
@@ -239,7 +229,6 @@ struct adapter {
 	int msg_enable;
 	u32 mmio_len;
 
-	struct work_struct ext_intr_handler_task;
 	struct adapter_params params;
 
 	/* Terminator modules. */
@@ -258,6 +247,7 @@ struct adapter {
 
 	/* guards async operations */
 	spinlock_t async_lock ____cacheline_aligned;
+	u32 pending_thread_intr;
 	u32 slow_intr_mask;
 	int t1powersave;
 };
@@ -335,20 +325,16 @@ void t1_interrupts_enable(adapter_t *adapter);
 void t1_interrupts_disable(adapter_t *adapter);
 void t1_interrupts_clear(adapter_t *adapter);
 int t1_elmer0_ext_intr_handler(adapter_t *adapter);
-void t1_elmer0_ext_intr(adapter_t *adapter);
-int t1_slow_intr_handler(adapter_t *adapter);
+irqreturn_t t1_slow_intr_handler(adapter_t *adapter);
 
 int t1_link_start(struct cphy *phy, struct cmac *mac, struct link_config *lc);
 const struct board_info *t1_get_board_info(unsigned int board_id);
-const struct board_info *t1_get_board_info_from_ids(unsigned int devid,
-						    unsigned short ssid);
 int t1_seeprom_read(adapter_t *adapter, u32 addr, __le32 *data);
 int t1_get_board_rev(adapter_t *adapter, const struct board_info *bi,
 		     struct adapter_params *p);
 int t1_init_hw_modules(adapter_t *adapter);
 int t1_init_sw_modules(adapter_t *adapter, const struct board_info *bi);
 void t1_free_sw_modules(adapter_t *adapter);
-void t1_fatal_err(adapter_t *adapter);
 void t1_link_changed(adapter_t *adapter, int port_id);
 void t1_link_negotiated(adapter_t *adapter, int port_id, int link_stat,
 			    int speed, int duplex, int pause);
diff --git a/drivers/net/ethernet/chelsio/cxgb/cphy.h b/drivers/net/ethernet/chelsio/cxgb/cphy.h
index bf43da6c6a63..12639b688ddc 100644
--- a/drivers/net/ethernet/chelsio/cxgb/cphy.h
+++ b/drivers/net/ethernet/chelsio/cxgb/cphy.h
@@ -1,3 +1,4 @@
+/* SPDX-License-Identifier: GPL-2.0-only */
 /*****************************************************************************
  *                                                                           *
  * File: cphy.h                                                              *
@@ -6,16 +7,6 @@
  * Description:                                                              *
  *  part of the Chelsio 10Gb Ethernet Driver.                                *
  *                                                                           *
- * This program is free software; you can redistribute it and/or modify      *
- * it under the terms of the GNU General Public License, version 2, as       *
- * published by the Free Software Foundation.                                *
- *                                                                           *
- * You should have received a copy of the GNU General Public License along   *
- * with this program; if not, see <http://www.gnu.org/licenses/>.            *
- *                                                                           *
- * THIS SOFTWARE IS PROVIDED ``AS IS'' AND WITHOUT ANY EXPRESS OR IMPLIED    *
- * WARRANTIES, INCLUDING, WITHOUT LIMITATION, THE IMPLIED WARRANTIES OF      *
- * MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE.                     *
  *                                                                           *
  * http://www.chelsio.com                                                    *
  *                                                                           *
diff --git a/drivers/net/ethernet/chelsio/cxgb/cpl5_cmd.h b/drivers/net/ethernet/chelsio/cxgb/cpl5_cmd.h
index 5249686afe71..a30fb407115d 100644
--- a/drivers/net/ethernet/chelsio/cxgb/cpl5_cmd.h
+++ b/drivers/net/ethernet/chelsio/cxgb/cpl5_cmd.h
@@ -1,3 +1,4 @@
+/* SPDX-License-Identifier: GPL-2.0-only */
 /*****************************************************************************
  *                                                                           *
  * File: cpl5_cmd.h                                                          *
@@ -6,16 +7,6 @@
  * Description:                                                              *
  *  part of the Chelsio 10Gb Ethernet Driver.                                *
  *                                                                           *
- * This program is free software; you can redistribute it and/or modify      *
- * it under the terms of the GNU General Public License, version 2, as       *
- * published by the Free Software Foundation.                                *
- *                                                                           *
- * You should have received a copy of the GNU General Public License along   *
- * with this program; if not, see <http://www.gnu.org/licenses/>.            *
- *                                                                           *
- * THIS SOFTWARE IS PROVIDED ``AS IS'' AND WITHOUT ANY EXPRESS OR IMPLIED    *
- * WARRANTIES, INCLUDING, WITHOUT LIMITATION, THE IMPLIED WARRANTIES OF      *
- * MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE.                     *
  *                                                                           *
  * http://www.chelsio.com                                                    *
  *                                                                           *
@@ -635,4 +626,3 @@ struct cpl_mss_change {
 };
 
 #endif /* _CXGB_CPL5_CMD_H_ */
-
diff --git a/drivers/net/ethernet/chelsio/cxgb/cxgb2.c b/drivers/net/ethernet/chelsio/cxgb/cxgb2.c
index 8623be13bf86..4a0e2d2eb60a 100644
--- a/drivers/net/ethernet/chelsio/cxgb/cxgb2.c
+++ b/drivers/net/ethernet/chelsio/cxgb/cxgb2.c
@@ -109,10 +109,6 @@ static int disable_msi = 0;
 module_param(disable_msi, int, 0);
 MODULE_PARM_DESC(disable_msi, "Disable Message Signaled Interrupt (MSI)");
 
-static const char pci_speed[][4] = {
-	"33", "66", "100", "133"
-};
-
 /*
  * Setup MAC to receive the types of packets we want.
  */
@@ -215,9 +211,10 @@ static int cxgb_up(struct adapter *adapter)
 	t1_interrupts_clear(adapter);
 
 	adapter->params.has_msi = !disable_msi && !pci_enable_msi(adapter->pdev);
-	err = request_irq(adapter->pdev->irq, t1_interrupt,
-			  adapter->params.has_msi ? 0 : IRQF_SHARED,
-			  adapter->name, adapter);
+	err = request_threaded_irq(adapter->pdev->irq, t1_interrupt,
+				   t1_interrupt_thread,
+				   adapter->params.has_msi ? 0 : IRQF_SHARED,
+				   adapter->name, adapter);
 	if (err) {
 		if (adapter->params.has_msi)
 			pci_disable_msi(adapter->pdev);
@@ -354,7 +351,7 @@ static void set_msglevel(struct net_device *dev, u32 val)
 	adapter->msg_enable = val;
 }
 
-static const char stats_strings[][ETH_GSTRING_LEN] = {
+static const char stats_strings[][ETH_GSTRING_LEN] __nonstring_array = {
 	"TxOctetsOK",
 	"TxOctetsBad",
 	"TxUnicastFramesOK",
@@ -432,9 +429,8 @@ static void get_drvinfo(struct net_device *dev, struct ethtool_drvinfo *info)
 {
 	struct adapter *adapter = dev->ml_priv;
 
-	strlcpy(info->driver, DRV_NAME, sizeof(info->driver));
-	strlcpy(info->version, DRV_VERSION, sizeof(info->version));
-	strlcpy(info->bus_info, pci_name(adapter->pdev),
+	strscpy(info->driver, DRV_NAME, sizeof(info->driver));
+	strscpy(info->bus_info, pci_name(adapter->pdev),
 		sizeof(info->bus_info));
 }
 
@@ -714,7 +710,9 @@ static int set_pauseparam(struct net_device *dev,
 	return 0;
 }
 
-static void get_sge_param(struct net_device *dev, struct ethtool_ringparam *e)
+static void get_sge_param(struct net_device *dev, struct ethtool_ringparam *e,
+			  struct kernel_ethtool_ringparam *kernel_e,
+			  struct netlink_ext_ack *extack)
 {
 	struct adapter *adapter = dev->ml_priv;
 	int jumbo_fl = t1_is_T1B(adapter) ? 1 : 0;
@@ -728,7 +726,9 @@ static void get_sge_param(struct net_device *dev, struct ethtool_ringparam *e)
 	e->tx_pending = adapter->params.sge.cmdQ_size[0];
 }
 
-static int set_sge_param(struct net_device *dev, struct ethtool_ringparam *e)
+static int set_sge_param(struct net_device *dev, struct ethtool_ringparam *e,
+			 struct kernel_ethtool_ringparam *kernel_e,
+			 struct netlink_ext_ack *extack)
 {
 	struct adapter *adapter = dev->ml_priv;
 	int jumbo_fl = t1_is_T1B(adapter) ? 1 : 0;
@@ -752,7 +752,9 @@ static int set_sge_param(struct net_device *dev, struct ethtool_ringparam *e)
 	return 0;
 }
 
-static int set_coalesce(struct net_device *dev, struct ethtool_coalesce *c)
+static int set_coalesce(struct net_device *dev, struct ethtool_coalesce *c,
+			struct kernel_ethtool_coalesce *kernel_coal,
+			struct netlink_ext_ack *extack)
 {
 	struct adapter *adapter = dev->ml_priv;
 
@@ -763,7 +765,9 @@ static int set_coalesce(struct net_device *dev, struct ethtool_coalesce *c)
 	return 0;
 }
 
-static int get_coalesce(struct net_device *dev, struct ethtool_coalesce *c)
+static int get_coalesce(struct net_device *dev, struct ethtool_coalesce *c,
+			struct kernel_ethtool_coalesce *kernel_coal,
+			struct netlink_ext_ack *extack)
 {
 	struct adapter *adapter = dev->ml_priv;
 
@@ -798,6 +802,9 @@ static int get_eeprom(struct net_device *dev, struct ethtool_eeprom *e,
 }
 
 static const struct ethtool_ops t1_ethtool_ops = {
+	.supported_coalesce_params = ETHTOOL_COALESCE_RX_USECS |
+				     ETHTOOL_COALESCE_USE_ADAPTIVE_RX |
+				     ETHTOOL_COALESCE_RATE_SAMPLE_INTERVAL,
 	.get_drvinfo       = get_drvinfo,
 	.get_msglevel      = get_msglevel,
 	.set_msglevel      = set_msglevel,
@@ -837,7 +844,7 @@ static int t1_change_mtu(struct net_device *dev, int new_mtu)
 		return -EOPNOTSUPP;
 	if ((ret = mac->ops->set_mtu(mac, new_mtu)))
 		return ret;
-	dev->mtu = new_mtu;
+	WRITE_ONCE(dev->mtu, new_mtu);
 	return 0;
 }
 
@@ -850,7 +857,7 @@ static int t1_set_mac_addr(struct net_device *dev, void *p)
 	if (!mac->ops->macaddress_set)
 		return -EOPNOTSUPP;
 
-	memcpy(dev->dev_addr, addr->sa_data, dev->addr_len);
+	eth_hw_addr_set(dev, addr->sa_data);
 	mac->ops->macaddress_set(mac, dev->dev_addr);
 	return 0;
 }
@@ -918,51 +925,6 @@ static void mac_stats_task(struct work_struct *work)
 	spin_unlock(&adapter->work_lock);
 }
 
-/*
- * Processes elmer0 external interrupts in process context.
- */
-static void ext_intr_task(struct work_struct *work)
-{
-	struct adapter *adapter =
-		container_of(work, struct adapter, ext_intr_handler_task);
-
-	t1_elmer0_ext_intr_handler(adapter);
-
-	/* Now reenable external interrupts */
-	spin_lock_irq(&adapter->async_lock);
-	adapter->slow_intr_mask |= F_PL_INTR_EXT;
-	writel(F_PL_INTR_EXT, adapter->regs + A_PL_CAUSE);
-	writel(adapter->slow_intr_mask | F_PL_INTR_SGE_DATA,
-		   adapter->regs + A_PL_ENABLE);
-	spin_unlock_irq(&adapter->async_lock);
-}
-
-/*
- * Interrupt-context handler for elmer0 external interrupts.
- */
-void t1_elmer0_ext_intr(struct adapter *adapter)
-{
-	/*
-	 * Schedule a task to handle external interrupts as we require
-	 * a process context.  We disable EXT interrupts in the interim
-	 * and let the task reenable them when it's done.
-	 */
-	adapter->slow_intr_mask &= ~F_PL_INTR_EXT;
-	writel(adapter->slow_intr_mask | F_PL_INTR_SGE_DATA,
-		   adapter->regs + A_PL_ENABLE);
-	schedule_work(&adapter->ext_intr_handler_task);
-}
-
-void t1_fatal_err(struct adapter *adapter)
-{
-	if (adapter->flags & FULL_INIT_DONE) {
-		t1_sge_stop(adapter->sge);
-		t1_interrupts_disable(adapter);
-	}
-	pr_alert("%s: encountered fatal error, operation suspended\n",
-		 adapter->name);
-}
-
 static const struct net_device_ops cxgb_netdev_ops = {
 	.ndo_open		= cxgb_open,
 	.ndo_stop		= cxgb_close,
@@ -970,7 +932,7 @@ static const struct net_device_ops cxgb_netdev_ops = {
 	.ndo_get_stats		= t1_get_stats,
 	.ndo_validate_addr	= eth_validate_addr,
 	.ndo_set_rx_mode	= t1_set_rxmode,
-	.ndo_do_ioctl		= t1_ioctl,
+	.ndo_eth_ioctl		= t1_ioctl,
 	.ndo_change_mtu		= t1_change_mtu,
 	.ndo_set_mac_address	= t1_set_mac_addr,
 	.ndo_fix_features	= t1_fix_features,
@@ -982,13 +944,11 @@ static const struct net_device_ops cxgb_netdev_ops = {
 
 static int init_one(struct pci_dev *pdev, const struct pci_device_id *ent)
 {
-	int i, err, pci_using_dac = 0;
 	unsigned long mmio_start, mmio_len;
 	const struct board_info *bi;
 	struct adapter *adapter = NULL;
 	struct port_info *pi;
-
-	pr_info_once("%s - version %s\n", DRV_DESCRIPTION, DRV_VERSION);
+	int i, err;
 
 	err = pci_enable_device(pdev);
 	if (err)
@@ -1001,17 +961,8 @@ static int init_one(struct pci_dev *pdev, const struct pci_device_id *ent)
 		goto out_disable_pdev;
 	}
 
-	if (!pci_set_dma_mask(pdev, DMA_BIT_MASK(64))) {
-		pci_using_dac = 1;
-
-		if (pci_set_consistent_dma_mask(pdev, DMA_BIT_MASK(64))) {
-			pr_err("%s: unable to obtain 64-bit DMA for "
-			       "consistent allocations\n", pci_name(pdev));
-			err = -ENODEV;
-			goto out_disable_pdev;
-		}
-
-	} else if ((err = pci_set_dma_mask(pdev, DMA_BIT_MASK(32))) != 0) {
+	err = dma_set_mask_and_coherent(&pdev->dev, DMA_BIT_MASK(64));
+	if (err) {
 		pr_err("%s: no usable DMA configuration\n", pci_name(pdev));
 		goto out_disable_pdev;
 	}
@@ -1066,8 +1017,6 @@ static int init_one(struct pci_dev *pdev, const struct pci_device_id *ent)
 			spin_lock_init(&adapter->async_lock);
 			spin_lock_init(&adapter->mac_lock);
 
-			INIT_WORK(&adapter->ext_intr_handler_task,
-				  ext_intr_task);
 			INIT_DELAYED_WORK(&adapter->stats_update_task,
 					  mac_stats_task);
 
@@ -1085,10 +1034,9 @@ static int init_one(struct pci_dev *pdev, const struct pci_device_id *ent)
 		netdev->hw_features |= NETIF_F_SG | NETIF_F_IP_CSUM |
 			NETIF_F_RXCSUM;
 		netdev->features |= NETIF_F_SG | NETIF_F_IP_CSUM |
-			NETIF_F_RXCSUM | NETIF_F_LLTX;
+			NETIF_F_RXCSUM | NETIF_F_HIGHDMA;
+		netdev->lltx = true;
 
-		if (pci_using_dac)
-			netdev->features |= NETIF_F_HIGHDMA;
 		if (vlan_tso_capable(adapter)) {
 			netdev->features |=
 				NETIF_F_HW_VLAN_CTAG_TX |
@@ -1106,7 +1054,7 @@ static int init_one(struct pci_dev *pdev, const struct pci_device_id *ent)
 		netdev->hard_header_len += (netdev->hw_features & NETIF_F_TSO) ?
 			sizeof(struct cpl_tx_pkt_lso) : sizeof(struct cpl_tx_pkt);
 
-		netif_napi_add(netdev, &adapter->napi, t1_poll, 64);
+		netif_napi_add(netdev, &adapter->napi, t1_poll);
 
 		netdev->ethtool_ops = &t1_ethtool_ops;
 
@@ -1157,6 +1105,7 @@ static int init_one(struct pci_dev *pdev, const struct pci_device_id *ent)
 	if (!adapter->registered_device_map) {
 		pr_err("%s: could not register any net devices\n",
 		       pci_name(pdev));
+		err = -EINVAL;
 		goto out_release_adapter_res;
 	}
 
diff --git a/drivers/net/ethernet/chelsio/cxgb/elmer0.h b/drivers/net/ethernet/chelsio/cxgb/elmer0.h
index 81526ad36339..0427e894c277 100644
--- a/drivers/net/ethernet/chelsio/cxgb/elmer0.h
+++ b/drivers/net/ethernet/chelsio/cxgb/elmer0.h
@@ -1,3 +1,4 @@
+/* SPDX-License-Identifier: GPL-2.0-only */
 /*****************************************************************************
  *                                                                           *
  * File: elmer0.h                                                            *
@@ -6,16 +7,6 @@
  * Description:                                                              *
  *  part of the Chelsio 10Gb Ethernet Driver.                                *
  *                                                                           *
- * This program is free software; you can redistribute it and/or modify      *
- * it under the terms of the GNU General Public License, version 2, as       *
- * published by the Free Software Foundation.                                *
- *                                                                           *
- * You should have received a copy of the GNU General Public License along   *
- * with this program; if not, see <http://www.gnu.org/licenses/>.            *
- *                                                                           *
- * THIS SOFTWARE IS PROVIDED ``AS IS'' AND WITHOUT ANY EXPRESS OR IMPLIED    *
- * WARRANTIES, INCLUDING, WITHOUT LIMITATION, THE IMPLIED WARRANTIES OF      *
- * MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE.                     *
  *                                                                           *
  * http://www.chelsio.com                                                    *
  *                                                                           *
@@ -154,4 +145,3 @@ enum {
 #define MI1_OP_INDIRECT_READ     3
 
 #endif /* _CXGB_ELMER0_H_ */
-
diff --git a/drivers/net/ethernet/chelsio/cxgb/espi.c b/drivers/net/ethernet/chelsio/cxgb/espi.c
index 3e182eee799e..ef70569435be 100644
--- a/drivers/net/ethernet/chelsio/cxgb/espi.c
+++ b/drivers/net/ethernet/chelsio/cxgb/espi.c
@@ -1,3 +1,4 @@
+// SPDX-License-Identifier: GPL-2.0-only
 /*****************************************************************************
  *                                                                           *
  * File: espi.c                                                              *
@@ -7,16 +8,6 @@
  *  Ethernet SPI functionality.                                              *
  *  part of the Chelsio 10Gb Ethernet Driver.                                *
  *                                                                           *
- * This program is free software; you can redistribute it and/or modify      *
- * it under the terms of the GNU General Public License, version 2, as       *
- * published by the Free Software Foundation.                                *
- *                                                                           *
- * You should have received a copy of the GNU General Public License along   *
- * with this program; if not, see <http://www.gnu.org/licenses/>.            *
- *                                                                           *
- * THIS SOFTWARE IS PROVIDED ``AS IS'' AND WITHOUT ANY EXPRESS OR IMPLIED    *
- * WARRANTIES, INCLUDING, WITHOUT LIMITATION, THE IMPLIED WARRANTIES OF      *
- * MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE.                     *
  *                                                                           *
  * http://www.chelsio.com                                                    *
  *                                                                           *
diff --git a/drivers/net/ethernet/chelsio/cxgb/espi.h b/drivers/net/ethernet/chelsio/cxgb/espi.h
index 162de5259df9..f588e9f3b37a 100644
--- a/drivers/net/ethernet/chelsio/cxgb/espi.h
+++ b/drivers/net/ethernet/chelsio/cxgb/espi.h
@@ -1,3 +1,4 @@
+/* SPDX-License-Identifier: GPL-2.0-only */
 /*****************************************************************************
  *                                                                           *
  * File: espi.h                                                              *
@@ -6,16 +7,6 @@
  * Description:                                                              *
  *  part of the Chelsio 10Gb Ethernet Driver.                                *
  *                                                                           *
- * This program is free software; you can redistribute it and/or modify      *
- * it under the terms of the GNU General Public License, version 2, as       *
- * published by the Free Software Foundation.                                *
- *                                                                           *
- * You should have received a copy of the GNU General Public License along   *
- * with this program; if not, see <http://www.gnu.org/licenses/>.            *
- *                                                                           *
- * THIS SOFTWARE IS PROVIDED ``AS IS'' AND WITHOUT ANY EXPRESS OR IMPLIED    *
- * WARRANTIES, INCLUDING, WITHOUT LIMITATION, THE IMPLIED WARRANTIES OF      *
- * MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE.                     *
  *                                                                           *
  * http://www.chelsio.com                                                    *
  *                                                                           *
diff --git a/drivers/net/ethernet/chelsio/cxgb/fpga_defs.h b/drivers/net/ethernet/chelsio/cxgb/fpga_defs.h
index ccdb2bc9ae98..e9c65d812c4e 100644
--- a/drivers/net/ethernet/chelsio/cxgb/fpga_defs.h
+++ b/drivers/net/ethernet/chelsio/cxgb/fpga_defs.h
@@ -1,3 +1,4 @@
+/* SPDX-License-Identifier: GPL-2.0 */
 /* $Date: 2005/03/07 23:59:05 $ $RCSfile: fpga_defs.h,v $ $Revision: 1.4 $ */
 
 /*
diff --git a/drivers/net/ethernet/chelsio/cxgb/gmac.h b/drivers/net/ethernet/chelsio/cxgb/gmac.h
index dfa77491a910..96077da1ed5e 100644
--- a/drivers/net/ethernet/chelsio/cxgb/gmac.h
+++ b/drivers/net/ethernet/chelsio/cxgb/gmac.h
@@ -1,3 +1,4 @@
+/* SPDX-License-Identifier: GPL-2.0-only */
 /*****************************************************************************
  *                                                                           *
  * File: gmac.h                                                              *
@@ -7,16 +8,6 @@
  *  Generic MAC functionality.                                               *
  *  part of the Chelsio 10Gb Ethernet Driver.                                *
  *                                                                           *
- * This program is free software; you can redistribute it and/or modify      *
- * it under the terms of the GNU General Public License, version 2, as       *
- * published by the Free Software Foundation.                                *
- *                                                                           *
- * You should have received a copy of the GNU General Public License along   *
- * with this program; if not, see <http://www.gnu.org/licenses/>.            *
- *                                                                           *
- * THIS SOFTWARE IS PROVIDED ``AS IS'' AND WITHOUT ANY EXPRESS OR IMPLIED    *
- * WARRANTIES, INCLUDING, WITHOUT LIMITATION, THE IMPLIED WARRANTIES OF      *
- * MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE.                     *
  *                                                                           *
  * http://www.chelsio.com                                                    *
  *                                                                           *
@@ -117,7 +108,7 @@ struct cmac_ops {
 	const struct cmac_statistics *(*statistics_update)(struct cmac *, int);
 
 	int (*macaddress_get)(struct cmac *, u8 mac_addr[6]);
-	int (*macaddress_set)(struct cmac *, u8 mac_addr[6]);
+	int (*macaddress_set)(struct cmac *, const u8 mac_addr[6]);
 };
 
 typedef struct _cmac_instance cmac_instance;
diff --git a/drivers/net/ethernet/chelsio/cxgb/mv88e1xxx.c b/drivers/net/ethernet/chelsio/cxgb/mv88e1xxx.c
index 76ce6e538326..30b003484fc1 100644
--- a/drivers/net/ethernet/chelsio/cxgb/mv88e1xxx.c
+++ b/drivers/net/ethernet/chelsio/cxgb/mv88e1xxx.c
@@ -1,3 +1,4 @@
+// SPDX-License-Identifier: GPL-2.0
 /* $Date: 2005/10/24 23:18:13 $ $RCSfile: mv88e1xxx.c,v $ $Revision: 1.49 $ */
 #include "common.h"
 #include "mv88e1xxx.h"
diff --git a/drivers/net/ethernet/chelsio/cxgb/mv88e1xxx.h b/drivers/net/ethernet/chelsio/cxgb/mv88e1xxx.h
index 967cc4286359..11bf0f7c2f90 100644
--- a/drivers/net/ethernet/chelsio/cxgb/mv88e1xxx.h
+++ b/drivers/net/ethernet/chelsio/cxgb/mv88e1xxx.h
@@ -1,3 +1,4 @@
+/* SPDX-License-Identifier: GPL-2.0 */
 /* $Date: 2005/03/07 23:59:05 $ $RCSfile: mv88e1xxx.h,v $ $Revision: 1.13 $ */
 #ifndef CHELSIO_MV8E1XXX_H
 #define CHELSIO_MV8E1XXX_H
diff --git a/drivers/net/ethernet/chelsio/cxgb/mv88x201x.c b/drivers/net/ethernet/chelsio/cxgb/mv88x201x.c
index 7ddb301bcba0..556c8ad68fa8 100644
--- a/drivers/net/ethernet/chelsio/cxgb/mv88x201x.c
+++ b/drivers/net/ethernet/chelsio/cxgb/mv88x201x.c
@@ -1,3 +1,4 @@
+// SPDX-License-Identifier: GPL-2.0-only
 /*****************************************************************************
  *                                                                           *
  * File: mv88x201x.c                                                         *
@@ -7,16 +8,6 @@
  *  Marvell PHY (mv88x201x) functionality.                                   *
  *  part of the Chelsio 10Gb Ethernet Driver.                                *
  *                                                                           *
- * This program is free software; you can redistribute it and/or modify      *
- * it under the terms of the GNU General Public License, version 2, as       *
- * published by the Free Software Foundation.                                *
- *                                                                           *
- * You should have received a copy of the GNU General Public License along   *
- * with this program; if not, see <http://www.gnu.org/licenses/>.            *
- *                                                                           *
- * THIS SOFTWARE IS PROVIDED ``AS IS'' AND WITHOUT ANY EXPRESS OR IMPLIED    *
- * WARRANTIES, INCLUDING, WITHOUT LIMITATION, THE IMPLIED WARRANTIES OF      *
- * MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE.                     *
  *                                                                           *
  * http://www.chelsio.com                                                    *
  *                                                                           *
diff --git a/drivers/net/ethernet/chelsio/cxgb/my3126.c b/drivers/net/ethernet/chelsio/cxgb/my3126.c
index d546f46c8ef7..60aa45b375b6 100644
--- a/drivers/net/ethernet/chelsio/cxgb/my3126.c
+++ b/drivers/net/ethernet/chelsio/cxgb/my3126.c
@@ -1,3 +1,4 @@
+// SPDX-License-Identifier: GPL-2.0
 /* $Date: 2005/11/12 02:13:49 $ $RCSfile: my3126.c,v $ $Revision: 1.15 $ */
 #include "cphy.h"
 #include "elmer0.h"
@@ -93,7 +94,7 @@ static int my3126_interrupt_handler(struct cphy *cphy)
 	return cphy_cause_link_change;
 }
 
-static void my3216_poll(struct work_struct *work)
+static void my3126_poll(struct work_struct *work)
 {
 	struct cphy *cphy = container_of(work, struct cphy, phy_update.work);
 
@@ -176,7 +177,7 @@ static struct cphy *my3126_phy_create(struct net_device *dev,
 		return NULL;
 
 	cphy_init(cphy, dev, phy_addr, &my3126_ops, mdio_ops);
-	INIT_DELAYED_WORK(&cphy->phy_update, my3216_poll);
+	INIT_DELAYED_WORK(&cphy->phy_update, my3126_poll);
 	cphy->bmsr = 0;
 
 	return cphy;
diff --git a/drivers/net/ethernet/chelsio/cxgb/pm3393.c b/drivers/net/ethernet/chelsio/cxgb/pm3393.c
index c27908e66f5e..f3ada6e7cdc5 100644
--- a/drivers/net/ethernet/chelsio/cxgb/pm3393.c
+++ b/drivers/net/ethernet/chelsio/cxgb/pm3393.c
@@ -1,3 +1,4 @@
+// SPDX-License-Identifier: GPL-2.0-only
 /*****************************************************************************
  *                                                                           *
  * File: pm3393.c                                                            *
@@ -7,16 +8,6 @@
  *  PMC/SIERRA (pm3393) MAC-PHY functionality.                               *
  *  part of the Chelsio 10Gb Ethernet Driver.                                *
  *                                                                           *
- * This program is free software; you can redistribute it and/or modify      *
- * it under the terms of the GNU General Public License, version 2, as       *
- * published by the Free Software Foundation.                                *
- *                                                                           *
- * You should have received a copy of the GNU General Public License along   *
- * with this program; if not, see <http://www.gnu.org/licenses/>.            *
- *                                                                           *
- * THIS SOFTWARE IS PROVIDED ``AS IS'' AND WITHOUT ANY EXPRESS OR IMPLIED    *
- * WARRANTIES, INCLUDING, WITHOUT LIMITATION, THE IMPLIED WARRANTIES OF      *
- * MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE.                     *
  *                                                                           *
  * http://www.chelsio.com                                                    *
  *                                                                           *
@@ -150,7 +141,7 @@ static int pm3393_interrupt_enable(struct cmac *cmac)
 	pmwrite(cmac, SUNI1x10GEXP_REG_GLOBAL_INTERRUPT_ENABLE,
 		0 /*SUNI1x10GEXP_BITMSK_TOP_INTE */ );
 
-	/* TERMINATOR - PL_INTERUPTS_EXT */
+	/* TERMINATOR - PL_INTERRUPTS_EXT */
 	pl_intr = readl(cmac->adapter->regs + A_PL_ENABLE);
 	pl_intr |= F_PL_INTR_EXT;
 	writel(pl_intr, cmac->adapter->regs + A_PL_ENABLE);
@@ -188,7 +179,7 @@ static int pm3393_interrupt_disable(struct cmac *cmac)
 	elmer &= ~ELMER0_GP_BIT1;
 	t1_tpi_write(cmac->adapter, A_ELMER0_INT_ENABLE, elmer);
 
-	/* TERMINATOR - PL_INTERUPTS_EXT */
+	/* TERMINATOR - PL_INTERRUPTS_EXT */
 	/* DO NOT DISABLE TERMINATOR's EXTERNAL INTERRUPTS. ANOTHER CHIP
 	 * COULD WANT THEM ENABLED. We disable PM3393 at the ELMER level.
 	 */
@@ -231,7 +222,7 @@ static int pm3393_interrupt_clear(struct cmac *cmac)
 	elmer |= ELMER0_GP_BIT1;
 	t1_tpi_write(cmac->adapter, A_ELMER0_INT_CAUSE, elmer);
 
-	/* TERMINATOR - PL_INTERUPTS_EXT
+	/* TERMINATOR - PL_INTERRUPTS_EXT
 	 */
 	pl_intr = readl(cmac->adapter->regs + A_PL_CAUSE);
 	pl_intr |= F_PL_INTR_EXT;
@@ -496,7 +487,7 @@ static int pm3393_macaddress_get(struct cmac *cmac, u8 mac_addr[6])
 	return 0;
 }
 
-static int pm3393_macaddress_set(struct cmac *cmac, u8 ma[6])
+static int pm3393_macaddress_set(struct cmac *cmac, const u8 ma[6])
 {
 	u32 val, lo, mid, hi, enabled = cmac->instance->enabled;
 
@@ -765,7 +756,7 @@ static int pm3393_mac_reset(adapter_t * adapter)
 
 		/* ??? If this fails, might be able to software reset the XAUI part
 		 *     and try to recover... thus saving us from doing another HW reset */
-		/* Has the XAUI MABC PLL circuitry stablized? */
+		/* Has the XAUI MABC PLL circuitry stabilized? */
 		is_xaui_mabc_pll_locked =
 		    (val & SUNI1x10GEXP_BITMSK_TOP_SXRA_EXPIRED);
 
diff --git a/drivers/net/ethernet/chelsio/cxgb/regs.h b/drivers/net/ethernet/chelsio/cxgb/regs.h
index 964ce59ee169..f751e680cf7d 100644
--- a/drivers/net/ethernet/chelsio/cxgb/regs.h
+++ b/drivers/net/ethernet/chelsio/cxgb/regs.h
@@ -1,3 +1,4 @@
+/* SPDX-License-Identifier: GPL-2.0-only */
 /*****************************************************************************
  *                                                                           *
  * File: regs.h                                                              *
@@ -6,16 +7,6 @@
  * Description:                                                              *
  *  part of the Chelsio 10Gb Ethernet Driver.                                *
  *                                                                           *
- * This program is free software; you can redistribute it and/or modify      *
- * it under the terms of the GNU General Public License, version 2, as       *
- * published by the Free Software Foundation.                                *
- *                                                                           *
- * You should have received a copy of the GNU General Public License along   *
- * with this program; if not, see <http://www.gnu.org/licenses/>.            *
- *                                                                           *
- * THIS SOFTWARE IS PROVIDED ``AS IS'' AND WITHOUT ANY EXPRESS OR IMPLIED    *
- * WARRANTIES, INCLUDING, WITHOUT LIMITATION, THE IMPLIED WARRANTIES OF      *
- * MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE.                     *
  *                                                                           *
  * http://www.chelsio.com                                                    *
  *                                                                           *
diff --git a/drivers/net/ethernet/chelsio/cxgb/sge.c b/drivers/net/ethernet/chelsio/cxgb/sge.c
index d56142b98534..5f354cf62cdd 100644
--- a/drivers/net/ethernet/chelsio/cxgb/sge.c
+++ b/drivers/net/ethernet/chelsio/cxgb/sge.c
@@ -1,3 +1,4 @@
+// SPDX-License-Identifier: GPL-2.0-only
 /*****************************************************************************
  *                                                                           *
  * File: sge.c                                                               *
@@ -7,16 +8,6 @@
  *  DMA engine.                                                              *
  *  part of the Chelsio 10Gb Ethernet Driver.                                *
  *                                                                           *
- * This program is free software; you can redistribute it and/or modify      *
- * it under the terms of the GNU General Public License, version 2, as       *
- * published by the Free Software Foundation.                                *
- *                                                                           *
- * You should have received a copy of the GNU General Public License along   *
- * with this program; if not, see <http://www.gnu.org/licenses/>.            *
- *                                                                           *
- * THIS SOFTWARE IS PROVIDED ``AS IS'' AND WITHOUT ANY EXPRESS OR IMPLIED    *
- * WARRANTIES, INCLUDING, WITHOUT LIMITATION, THE IMPLIED WARRANTIES OF      *
- * MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE.                     *
  *                                                                           *
  * http://www.chelsio.com                                                    *
  *                                                                           *
@@ -239,8 +230,10 @@ struct sched {
 	unsigned int	num;		/* num skbs in per port queues */
 	struct sched_port p[MAX_NPORTS];
 	struct tasklet_struct sched_tsk;/* tasklet used to run scheduler */
+	struct sge *sge;
 };
-static void restart_sched(unsigned long);
+
+static void restart_sched(struct tasklet_struct *t);
 
 
 /*
@@ -378,7 +371,8 @@ static int tx_sched_init(struct sge *sge)
 		return -ENOMEM;
 
 	pr_debug("tx_sched_init\n");
-	tasklet_init(&s->sched_tsk, restart_sched, (unsigned long) sge);
+	tasklet_setup(&s->sched_tsk, restart_sched);
+	s->sge = sge;
 	sge->tx_sched = s;
 
 	for (i = 0; i < MAX_NPORTS; i++) {
@@ -509,9 +503,8 @@ static void free_freelQ_buffers(struct pci_dev *pdev, struct freelQ *q)
 	while (q->credits--) {
 		struct freelQ_ce *ce = &q->centries[cidx];
 
-		pci_unmap_single(pdev, dma_unmap_addr(ce, dma_addr),
-				 dma_unmap_len(ce, dma_len),
-				 PCI_DMA_FROMDEVICE);
+		dma_unmap_single(&pdev->dev, dma_unmap_addr(ce, dma_addr),
+				 dma_unmap_len(ce, dma_len), DMA_FROM_DEVICE);
 		dev_kfree_skb(ce->skb);
 		ce->skb = NULL;
 		if (++cidx == q->size)
@@ -529,8 +522,8 @@ static void free_rx_resources(struct sge *sge)
 
 	if (sge->respQ.entries) {
 		size = sizeof(struct respQ_e) * sge->respQ.size;
-		pci_free_consistent(pdev, size, sge->respQ.entries,
-				    sge->respQ.dma_addr);
+		dma_free_coherent(&pdev->dev, size, sge->respQ.entries,
+				  sge->respQ.dma_addr);
 	}
 
 	for (i = 0; i < SGE_FREELQ_N; i++) {
@@ -542,8 +535,8 @@ static void free_rx_resources(struct sge *sge)
 		}
 		if (q->entries) {
 			size = sizeof(struct freelQ_e) * q->size;
-			pci_free_consistent(pdev, size, q->entries,
-					    q->dma_addr);
+			dma_free_coherent(&pdev->dev, size, q->entries,
+					  q->dma_addr);
 		}
 	}
 }
@@ -564,7 +557,8 @@ static int alloc_rx_resources(struct sge *sge, struct sge_params *p)
 		q->size = p->freelQ_size[i];
 		q->dma_offset = sge->rx_pkt_pad ? 0 : NET_IP_ALIGN;
 		size = sizeof(struct freelQ_e) * q->size;
-		q->entries = pci_alloc_consistent(pdev, size, &q->dma_addr);
+		q->entries = dma_alloc_coherent(&pdev->dev, size,
+						&q->dma_addr, GFP_KERNEL);
 		if (!q->entries)
 			goto err_no_mem;
 
@@ -585,8 +579,7 @@ static int alloc_rx_resources(struct sge *sge, struct sge_params *p)
 		sizeof(struct cpl_rx_data) +
 		sge->freelQ[!sge->jumbo_fl].dma_offset;
 
-		size = (16 * 1024) -
-		    SKB_DATA_ALIGN(sizeof(struct skb_shared_info));
+	size = (16 * 1024) - SKB_DATA_ALIGN(sizeof(struct skb_shared_info));
 
 	sge->freelQ[sge->jumbo_fl].rx_buffer_size = size;
 
@@ -602,7 +595,8 @@ static int alloc_rx_resources(struct sge *sge, struct sge_params *p)
 	sge->respQ.credits = 0;
 	size = sizeof(struct respQ_e) * sge->respQ.size;
 	sge->respQ.entries =
-		pci_alloc_consistent(pdev, size, &sge->respQ.dma_addr);
+		dma_alloc_coherent(&pdev->dev, size, &sge->respQ.dma_addr,
+				   GFP_KERNEL);
 	if (!sge->respQ.entries)
 		goto err_no_mem;
 	return 0;
@@ -625,9 +619,10 @@ static void free_cmdQ_buffers(struct sge *sge, struct cmdQ *q, unsigned int n)
 	ce = &q->centries[cidx];
 	while (n--) {
 		if (likely(dma_unmap_len(ce, dma_len))) {
-			pci_unmap_single(pdev, dma_unmap_addr(ce, dma_addr),
+			dma_unmap_single(&pdev->dev,
+					 dma_unmap_addr(ce, dma_addr),
 					 dma_unmap_len(ce, dma_len),
-					 PCI_DMA_TODEVICE);
+					 DMA_TO_DEVICE);
 			if (q->sop)
 				q->sop = 0;
 		}
@@ -664,8 +659,8 @@ static void free_tx_resources(struct sge *sge)
 		}
 		if (q->entries) {
 			size = sizeof(struct cmdQ_e) * q->size;
-			pci_free_consistent(pdev, size, q->entries,
-					    q->dma_addr);
+			dma_free_coherent(&pdev->dev, size, q->entries,
+					  q->dma_addr);
 		}
 	}
 }
@@ -690,7 +685,8 @@ static int alloc_tx_resources(struct sge *sge, struct sge_params *p)
 		q->stop_thres = 0;
 		spin_lock_init(&q->lock);
 		size = sizeof(struct cmdQ_e) * q->size;
-		q->entries = pci_alloc_consistent(pdev, size, &q->dma_addr);
+		q->entries = dma_alloc_coherent(&pdev->dev, size,
+						&q->dma_addr, GFP_KERNEL);
 		if (!q->entries)
 			goto err_no_mem;
 
@@ -838,8 +834,8 @@ static void refill_free_list(struct sge *sge, struct freelQ *q)
 			break;
 
 		skb_reserve(skb, q->dma_offset);
-		mapping = pci_map_single(pdev, skb->data, dma_len,
-					 PCI_DMA_FROMDEVICE);
+		mapping = dma_map_single(&pdev->dev, skb->data, dma_len,
+					 DMA_FROM_DEVICE);
 		skb_reserve(skb, sge->rx_pkt_pad);
 
 		ce->skb = skb;
@@ -935,10 +931,11 @@ void t1_sge_intr_clear(struct sge *sge)
 /*
  * SGE 'Error' interrupt handler
  */
-int t1_sge_intr_error_handler(struct sge *sge)
+bool t1_sge_intr_error_handler(struct sge *sge)
 {
 	struct adapter *adapter = sge->adapter;
 	u32 cause = readl(adapter->regs + A_SG_INT_CAUSE);
+	bool wake = false;
 
 	if (adapter->port[0].dev->hw_features & NETIF_F_TSO)
 		cause &= ~F_PACKET_TOO_BIG;
@@ -962,11 +959,14 @@ int t1_sge_intr_error_handler(struct sge *sge)
 		sge->stats.pkt_mismatch++;
 		pr_alert("%s: SGE packet mismatch\n", adapter->name);
 	}
-	if (cause & SGE_INT_FATAL)
-		t1_fatal_err(adapter);
+	if (cause & SGE_INT_FATAL) {
+		t1_interrupts_disable(adapter);
+		adapter->pending_thread_intr |= F_PL_INTR_SGE_ERR;
+		wake = true;
+	}
 
 	writel(cause, adapter->regs + A_SG_INT_CAUSE);
-	return 0;
+	return wake;
 }
 
 const struct sge_intr_counts *t1_sge_get_intr_counts(const struct sge *sge)
@@ -1050,15 +1050,15 @@ static inline struct sk_buff *get_packet(struct adapter *adapter,
 			goto use_orig_buf;
 
 		skb_put(skb, len);
-		pci_dma_sync_single_for_cpu(pdev,
-					    dma_unmap_addr(ce, dma_addr),
-					    dma_unmap_len(ce, dma_len),
-					    PCI_DMA_FROMDEVICE);
+		dma_sync_single_for_cpu(&pdev->dev,
+					dma_unmap_addr(ce, dma_addr),
+					dma_unmap_len(ce, dma_len),
+					DMA_FROM_DEVICE);
 		skb_copy_from_linear_data(ce->skb, skb->data, len);
-		pci_dma_sync_single_for_device(pdev,
-					       dma_unmap_addr(ce, dma_addr),
-					       dma_unmap_len(ce, dma_len),
-					       PCI_DMA_FROMDEVICE);
+		dma_sync_single_for_device(&pdev->dev,
+					   dma_unmap_addr(ce, dma_addr),
+					   dma_unmap_len(ce, dma_len),
+					   DMA_FROM_DEVICE);
 		recycle_fl_buf(fl, fl->cidx);
 		return skb;
 	}
@@ -1069,8 +1069,8 @@ use_orig_buf:
 		return NULL;
 	}
 
-	pci_unmap_single(pdev, dma_unmap_addr(ce, dma_addr),
-			 dma_unmap_len(ce, dma_len), PCI_DMA_FROMDEVICE);
+	dma_unmap_single(&pdev->dev, dma_unmap_addr(ce, dma_addr),
+			 dma_unmap_len(ce, dma_len), DMA_FROM_DEVICE);
 	skb = ce->skb;
 	prefetch(skb->data);
 
@@ -1092,8 +1092,9 @@ static void unexpected_offload(struct adapter *adapter, struct freelQ *fl)
 	struct freelQ_ce *ce = &fl->centries[fl->cidx];
 	struct sk_buff *skb = ce->skb;
 
-	pci_dma_sync_single_for_cpu(adapter->pdev, dma_unmap_addr(ce, dma_addr),
-			    dma_unmap_len(ce, dma_len), PCI_DMA_FROMDEVICE);
+	dma_sync_single_for_cpu(&adapter->pdev->dev,
+				dma_unmap_addr(ce, dma_addr),
+				dma_unmap_len(ce, dma_len), DMA_FROM_DEVICE);
 	pr_err("%s: unexpected offload packet, cmd %u\n",
 	       adapter->name, *skb->data);
 	recycle_fl_buf(fl, fl->cidx);
@@ -1210,8 +1211,8 @@ static inline void write_tx_descs(struct adapter *adapter, struct sk_buff *skb,
 	e = e1 = &q->entries[pidx];
 	ce = &q->centries[pidx];
 
-	mapping = pci_map_single(adapter->pdev, skb->data,
-				 skb_headlen(skb), PCI_DMA_TODEVICE);
+	mapping = dma_map_single(&adapter->pdev->dev, skb->data,
+				 skb_headlen(skb), DMA_TO_DEVICE);
 
 	desc_mapping = mapping;
 	desc_len = skb_headlen(skb);
@@ -1302,9 +1303,10 @@ static inline void reclaim_completed_tx(struct sge *sge, struct cmdQ *q)
  * Called from tasklet. Checks the scheduler for any
  * pending skbs that can be sent.
  */
-static void restart_sched(unsigned long arg)
+static void restart_sched(struct tasklet_struct *t)
 {
-	struct sge *sge = (struct sge *) arg;
+	struct sched *s = from_tasklet(s, t, sched_tsk);
+	struct sge *sge = s->sge;
 	struct adapter *adapter = sge->adapter;
 	struct cmdQ *q = &sge->cmdQ[0];
 	struct sk_buff *skb;
@@ -1348,7 +1350,7 @@ static void restart_sched(unsigned long arg)
  *	@fl: the free list that contains the packet buffer
  *	@len: the packet length
  *
- *	Process an ingress ethernet pakcet and deliver it to the stack.
+ *	Process an ingress ethernet packet and deliver it to the stack.
  */
 static void sge_rx(struct sge *sge, struct freelQ *fl, unsigned int len)
 {
@@ -1612,11 +1614,46 @@ int t1_poll(struct napi_struct *napi, int budget)
 	return work_done;
 }
 
+irqreturn_t t1_interrupt_thread(int irq, void *data)
+{
+	struct adapter *adapter = data;
+	u32 pending_thread_intr;
+
+	spin_lock_irq(&adapter->async_lock);
+	pending_thread_intr = adapter->pending_thread_intr;
+	adapter->pending_thread_intr = 0;
+	spin_unlock_irq(&adapter->async_lock);
+
+	if (!pending_thread_intr)
+		return IRQ_NONE;
+
+	if (pending_thread_intr & F_PL_INTR_EXT)
+		t1_elmer0_ext_intr_handler(adapter);
+
+	/* This error is fatal, interrupts remain off */
+	if (pending_thread_intr & F_PL_INTR_SGE_ERR) {
+		pr_alert("%s: encountered fatal error, operation suspended\n",
+			 adapter->name);
+		t1_sge_stop(adapter->sge);
+		return IRQ_HANDLED;
+	}
+
+	spin_lock_irq(&adapter->async_lock);
+	adapter->slow_intr_mask |= F_PL_INTR_EXT;
+
+	writel(F_PL_INTR_EXT, adapter->regs + A_PL_CAUSE);
+	writel(adapter->slow_intr_mask | F_PL_INTR_SGE_DATA,
+	       adapter->regs + A_PL_ENABLE);
+	spin_unlock_irq(&adapter->async_lock);
+
+	return IRQ_HANDLED;
+}
+
 irqreturn_t t1_interrupt(int irq, void *data)
 {
 	struct adapter *adapter = data;
 	struct sge *sge = adapter->sge;
-	int handled;
+	irqreturn_t handled;
 
 	if (likely(responses_pending(adapter))) {
 		writel(F_PL_INTR_SGE_DATA, adapter->regs + A_PL_CAUSE);
@@ -1638,10 +1675,10 @@ irqreturn_t t1_interrupt(int irq, void *data)
 	handled = t1_slow_intr_handler(adapter);
 	spin_unlock(&adapter->async_lock);
 
-	if (!handled)
+	if (handled == IRQ_NONE)
 		sge->stats.unhandled_irqs++;
 
-	return IRQ_RETVAL(handled != 0);
+	return handled;
 }
 
 /*
@@ -1801,7 +1838,7 @@ netdev_tx_t t1_start_xmit(struct sk_buff *skb, struct net_device *dev)
 		eth_type = skb_network_offset(skb) == ETH_HLEN ?
 			CPL_ETH_II : CPL_ETH_II_VLAN;
 
-		hdr = (struct cpl_tx_pkt_lso *)skb_push(skb, sizeof(*hdr));
+		hdr = skb_push(skb, sizeof(*hdr));
 		hdr->opcode = CPL_TX_PKT_LSO;
 		hdr->ip_csum_dis = hdr->l4_csum_dis = 0;
 		hdr->ip_hdr_words = ip_hdr(skb)->ihl;
@@ -1849,7 +1886,7 @@ netdev_tx_t t1_start_xmit(struct sk_buff *skb, struct net_device *dev)
 			}
 		}
 
-		cpl = (struct cpl_tx_pkt *)__skb_push(skb, sizeof(*cpl));
+		cpl = __skb_push(skb, sizeof(*cpl));
 		cpl->opcode = CPL_TX_PKT;
 		cpl->ip_csum_dis = 1;    /* SW calculates IP csum */
 		cpl->l4_csum_dis = skb->ip_summed == CHECKSUM_PARTIAL ? 0 : 1;
@@ -1882,10 +1919,10 @@ send:
 /*
  * Callback for the Tx buffer reclaim timer.  Runs with softirqs disabled.
  */
-static void sge_tx_reclaim_cb(unsigned long data)
+static void sge_tx_reclaim_cb(struct timer_list *t)
 {
 	int i;
-	struct sge *sge = (struct sge *)data;
+	struct sge *sge = timer_container_of(sge, t, tx_reclaim_timer);
 
 	for (i = 0; i < SGE_CMDQ_N; ++i) {
 		struct cmdQ *q = &sge->cmdQ[i];
@@ -1947,9 +1984,9 @@ void t1_sge_stop(struct sge *sge)
 	readl(sge->adapter->regs + A_SG_CONTROL); /* flush */
 
 	if (is_T2(sge->adapter))
-		del_timer_sync(&sge->espibug_timer);
+		timer_delete_sync(&sge->espibug_timer);
 
-	del_timer_sync(&sge->tx_reclaim_timer);
+	timer_delete_sync(&sge->tx_reclaim_timer);
 	if (sge->tx_sched)
 		tx_sched_stop(sge);
 
@@ -1978,10 +2015,10 @@ void t1_sge_start(struct sge *sge)
 /*
  * Callback for the T2 ESPI 'stuck packet feature' workaorund
  */
-static void espibug_workaround_t204(unsigned long data)
+static void espibug_workaround_t204(struct timer_list *t)
 {
-	struct adapter *adapter = (struct adapter *)data;
-	struct sge *sge = adapter->sge;
+	struct sge *sge = timer_container_of(sge, t, espibug_timer);
+	struct adapter *adapter = sge->adapter;
 	unsigned int nports = adapter->params.nports;
 	u32 seop[MAX_NPORTS];
 
@@ -2021,10 +2058,10 @@ static void espibug_workaround_t204(unsigned long data)
 	mod_timer(&sge->espibug_timer, jiffies + sge->espibug_timeout);
 }
 
-static void espibug_workaround(unsigned long data)
+static void espibug_workaround(struct timer_list *t)
 {
-	struct adapter *adapter = (struct adapter *)data;
-	struct sge *sge = adapter->sge;
+	struct sge *sge = timer_container_of(sge, t, espibug_timer);
+	struct adapter *adapter = sge->adapter;
 
 	if (netif_running(adapter->port[0].dev)) {
 	        struct sk_buff *skb = sge->espibug_skb[0];
@@ -2075,19 +2112,15 @@ struct sge *t1_sge_create(struct adapter *adapter, struct sge_params *p)
 			goto nomem_port;
 	}
 
-	init_timer(&sge->tx_reclaim_timer);
-	sge->tx_reclaim_timer.data = (unsigned long)sge;
-	sge->tx_reclaim_timer.function = sge_tx_reclaim_cb;
+	timer_setup(&sge->tx_reclaim_timer, sge_tx_reclaim_cb, 0);
 
 	if (is_T2(sge->adapter)) {
-		init_timer(&sge->espibug_timer);
+		timer_setup(&sge->espibug_timer,
+			    adapter->params.nports > 1 ? espibug_workaround_t204 : espibug_workaround,
+			    0);
 
-		if (adapter->params.nports > 1) {
+		if (adapter->params.nports > 1)
 			tx_sched_init(sge);
-			sge->espibug_timer.function = espibug_workaround_t204;
-		} else
-			sge->espibug_timer.function = espibug_workaround;
-		sge->espibug_timer.data = (unsigned long)sge->adapter;
 
 		sge->espibug_timeout = 1;
 		/* for T204, every 10ms */
diff --git a/drivers/net/ethernet/chelsio/cxgb/sge.h b/drivers/net/ethernet/chelsio/cxgb/sge.h
index a1ba591b3431..f7e6f64040ea 100644
--- a/drivers/net/ethernet/chelsio/cxgb/sge.h
+++ b/drivers/net/ethernet/chelsio/cxgb/sge.h
@@ -1,3 +1,4 @@
+/* SPDX-License-Identifier: GPL-2.0-only */
 /*****************************************************************************
  *                                                                           *
  * File: sge.h                                                               *
@@ -6,16 +7,6 @@
  * Description:                                                              *
  *  part of the Chelsio 10Gb Ethernet Driver.                                *
  *                                                                           *
- * This program is free software; you can redistribute it and/or modify      *
- * it under the terms of the GNU General Public License, version 2, as       *
- * published by the Free Software Foundation.                                *
- *                                                                           *
- * You should have received a copy of the GNU General Public License along   *
- * with this program; if not, see <http://www.gnu.org/licenses/>.            *
- *                                                                           *
- * THIS SOFTWARE IS PROVIDED ``AS IS'' AND WITHOUT ANY EXPRESS OR IMPLIED    *
- * WARRANTIES, INCLUDING, WITHOUT LIMITATION, THE IMPLIED WARRANTIES OF      *
- * MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE.                     *
  *                                                                           *
  * http://www.chelsio.com                                                    *
  *                                                                           *
@@ -74,6 +65,7 @@ struct sge *t1_sge_create(struct adapter *, struct sge_params *);
 int t1_sge_configure(struct sge *, struct sge_params *);
 int t1_sge_set_coalesce_params(struct sge *, struct sge_params *);
 void t1_sge_destroy(struct sge *);
+irqreturn_t t1_interrupt_thread(int irq, void *data);
 irqreturn_t t1_interrupt(int irq, void *cookie);
 int t1_poll(struct napi_struct *, int);
 
@@ -81,7 +73,7 @@ netdev_tx_t t1_start_xmit(struct sk_buff *skb, struct net_device *dev);
 void t1_vlan_mode(struct adapter *adapter, netdev_features_t features);
 void t1_sge_start(struct sge *);
 void t1_sge_stop(struct sge *);
-int t1_sge_intr_error_handler(struct sge *);
+bool t1_sge_intr_error_handler(struct sge *sge);
 void t1_sge_intr_enable(struct sge *);
 void t1_sge_intr_disable(struct sge *);
 void t1_sge_intr_clear(struct sge *);
diff --git a/drivers/net/ethernet/chelsio/cxgb/subr.c b/drivers/net/ethernet/chelsio/cxgb/subr.c
index ea0f8741d7cf..367a9e4581d5 100644
--- a/drivers/net/ethernet/chelsio/cxgb/subr.c
+++ b/drivers/net/ethernet/chelsio/cxgb/subr.c
@@ -1,3 +1,4 @@
+// SPDX-License-Identifier: GPL-2.0-only
 /*****************************************************************************
  *                                                                           *
  * File: subr.c                                                              *
@@ -7,16 +8,6 @@
  *  Various subroutines (intr,pio,etc.) used by Chelsio 10G Ethernet driver. *
  *  part of the Chelsio 10Gb Ethernet Driver.                                *
  *                                                                           *
- * This program is free software; you can redistribute it and/or modify      *
- * it under the terms of the GNU General Public License, version 2, as       *
- * published by the Free Software Foundation.                                *
- *                                                                           *
- * You should have received a copy of the GNU General Public License along   *
- * with this program; if not, see <http://www.gnu.org/licenses/>.            *
- *                                                                           *
- * THIS SOFTWARE IS PROVIDED ``AS IS'' AND WITHOUT ANY EXPRESS OR IMPLIED    *
- * WARRANTIES, INCLUDING, WITHOUT LIMITATION, THE IMPLIED WARRANTIES OF      *
- * MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE.                     *
  *                                                                           *
  * http://www.chelsio.com                                                    *
  *                                                                           *
@@ -170,7 +161,7 @@ void t1_link_changed(adapter_t *adapter, int port_id)
 	t1_link_negotiated(adapter, port_id, link_ok, speed, duplex, fc);
 }
 
-static int t1_pci_intr_handler(adapter_t *adapter)
+static bool t1_pci_intr_handler(adapter_t *adapter)
 {
 	u32 pcix_cause;
 
@@ -179,9 +170,13 @@ static int t1_pci_intr_handler(adapter_t *adapter)
 	if (pcix_cause) {
 		pci_write_config_dword(adapter->pdev, A_PCICFG_INTR_CAUSE,
 				       pcix_cause);
-		t1_fatal_err(adapter);    /* PCI errors are fatal */
+		/* PCI errors are fatal */
+		t1_interrupts_disable(adapter);
+		adapter->pending_thread_intr |= F_PL_INTR_SGE_ERR;
+		pr_alert("%s: PCI error encountered.\n", adapter->name);
+		return true;
 	}
-	return 0;
+	return false;
 }
 
 #ifdef CONFIG_CHELSIO_T1_1G
@@ -210,13 +205,16 @@ static int fpga_phy_intr_handler(adapter_t *adapter)
 /*
  * Slow path interrupt handler for FPGAs.
  */
-static int fpga_slow_intr(adapter_t *adapter)
+static irqreturn_t fpga_slow_intr(adapter_t *adapter)
 {
 	u32 cause = readl(adapter->regs + A_PL_CAUSE);
+	irqreturn_t ret = IRQ_NONE;
 
 	cause &= ~F_PL_INTR_SGE_DATA;
-	if (cause & F_PL_INTR_SGE_ERR)
-		t1_sge_intr_error_handler(adapter->sge);
+	if (cause & F_PL_INTR_SGE_ERR) {
+		if (t1_sge_intr_error_handler(adapter->sge))
+			ret = IRQ_WAKE_THREAD;
+	}
 
 	if (cause & FPGA_PCIX_INTERRUPT_GMAC)
 		fpga_phy_intr_handler(adapter);
@@ -231,14 +229,19 @@ static int fpga_slow_intr(adapter_t *adapter)
 		/* Clear TP interrupt */
 		writel(tp_cause, adapter->regs + FPGA_TP_ADDR_INTERRUPT_CAUSE);
 	}
-	if (cause & FPGA_PCIX_INTERRUPT_PCIX)
-		t1_pci_intr_handler(adapter);
+	if (cause & FPGA_PCIX_INTERRUPT_PCIX) {
+		if (t1_pci_intr_handler(adapter))
+			ret = IRQ_WAKE_THREAD;
+	}
 
 	/* Clear the interrupts just processed. */
 	if (cause)
 		writel(cause, adapter->regs + A_PL_CAUSE);
 
-	return cause != 0;
+	if (ret != IRQ_NONE)
+		return ret;
+
+	return cause == 0 ? IRQ_NONE : IRQ_HANDLED;
 }
 #endif
 
@@ -842,31 +845,45 @@ void t1_interrupts_clear(adapter_t* adapter)
 /*
  * Slow path interrupt handler for ASICs.
  */
-static int asic_slow_intr(adapter_t *adapter)
+static irqreturn_t asic_slow_intr(adapter_t *adapter)
 {
 	u32 cause = readl(adapter->regs + A_PL_CAUSE);
+	irqreturn_t ret = IRQ_HANDLED;
 
 	cause &= adapter->slow_intr_mask;
 	if (!cause)
-		return 0;
-	if (cause & F_PL_INTR_SGE_ERR)
-		t1_sge_intr_error_handler(adapter->sge);
+		return IRQ_NONE;
+	if (cause & F_PL_INTR_SGE_ERR) {
+		if (t1_sge_intr_error_handler(adapter->sge))
+			ret = IRQ_WAKE_THREAD;
+	}
 	if (cause & F_PL_INTR_TP)
 		t1_tp_intr_handler(adapter->tp);
 	if (cause & F_PL_INTR_ESPI)
 		t1_espi_intr_handler(adapter->espi);
-	if (cause & F_PL_INTR_PCIX)
-		t1_pci_intr_handler(adapter);
-	if (cause & F_PL_INTR_EXT)
-		t1_elmer0_ext_intr(adapter);
+	if (cause & F_PL_INTR_PCIX) {
+		if (t1_pci_intr_handler(adapter))
+			ret = IRQ_WAKE_THREAD;
+	}
+	if (cause & F_PL_INTR_EXT) {
+		/* Wake the threaded interrupt to handle external interrupts as
+		 * we require a process context. We disable EXT interrupts in
+		 * the interim and let the thread reenable them when it's done.
+		 */
+		adapter->pending_thread_intr |= F_PL_INTR_EXT;
+		adapter->slow_intr_mask &= ~F_PL_INTR_EXT;
+		writel(adapter->slow_intr_mask | F_PL_INTR_SGE_DATA,
+		       adapter->regs + A_PL_ENABLE);
+		ret = IRQ_WAKE_THREAD;
+	}
 
 	/* Clear the interrupts just processed. */
 	writel(cause, adapter->regs + A_PL_CAUSE);
 	readl(adapter->regs + A_PL_CAUSE); /* flush writes */
-	return 1;
+	return ret;
 }
 
-int t1_slow_intr_handler(adapter_t *adapter)
+irqreturn_t t1_slow_intr_handler(adapter_t *adapter)
 {
 #ifdef CONFIG_CHELSIO_T1_1G
 	if (!t1_is_asic(adapter))
@@ -1114,7 +1131,7 @@ int t1_init_sw_modules(adapter_t *adapter, const struct board_info *bi)
 			       adapter->port[i].dev->name);
 			goto error;
 		}
-		memcpy(adapter->port[i].dev->dev_addr, hw_addr, ETH_ALEN);
+		eth_hw_addr_set(adapter->port[i].dev, hw_addr);
 		init_link_config(&adapter->port[i].link_config, bi);
 	}
 
diff --git a/drivers/net/ethernet/chelsio/cxgb/suni1x10gexp_regs.h b/drivers/net/ethernet/chelsio/cxgb/suni1x10gexp_regs.h
index 7f79cc7ceb75..4c883170683b 100644
--- a/drivers/net/ethernet/chelsio/cxgb/suni1x10gexp_regs.h
+++ b/drivers/net/ethernet/chelsio/cxgb/suni1x10gexp_regs.h
@@ -1,3 +1,4 @@
+/* SPDX-License-Identifier: GPL-2.0-only */
 /*****************************************************************************
  *                                                                           *
  * File: suni1x10gexp_regs.h                                                 *
@@ -7,16 +8,6 @@
  *  PMC/SIERRA (pm3393) MAC-PHY functionality.                               *
  *  part of the Chelsio 10Gb Ethernet Driver.                                *
  *                                                                           *
- * This program is free software; you can redistribute it and/or modify      *
- * it under the terms of the GNU General Public License, version 2, as       *
- * published by the Free Software Foundation.                                *
- *                                                                           *
- * You should have received a copy of the GNU General Public License along   *
- * with this program; if not, see <http://www.gnu.org/licenses/>.            *
- *                                                                           *
- * THIS SOFTWARE IS PROVIDED ``AS IS'' AND WITHOUT ANY EXPRESS OR IMPLIED    *
- * WARRANTIES, INCLUDING, WITHOUT LIMITATION, THE IMPLIED WARRANTIES OF      *
- * MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE.                     *
  *                                                                           *
  * http://www.chelsio.com                                                    *
  *                                                                           *
@@ -1639,4 +1630,3 @@
 #define SUNI1x10GEXP_BITMSK_PL4IDU_DIP4I       0x0002
 
 #endif /* _CXGB_SUNI1x10GEXP_REGS_H_ */
-
diff --git a/drivers/net/ethernet/chelsio/cxgb/tp.c b/drivers/net/ethernet/chelsio/cxgb/tp.c
index b146acabf982..4337cee0763e 100644
--- a/drivers/net/ethernet/chelsio/cxgb/tp.c
+++ b/drivers/net/ethernet/chelsio/cxgb/tp.c
@@ -1,3 +1,4 @@
+// SPDX-License-Identifier: GPL-2.0
 /* $Date: 2006/02/07 04:21:54 $ $RCSfile: tp.c,v $ $Revision: 1.73 $ */
 #include "common.h"
 #include "regs.h"
diff --git a/drivers/net/ethernet/chelsio/cxgb/tp.h b/drivers/net/ethernet/chelsio/cxgb/tp.h
index dfd8ce25106a..64f93dcc676b 100644
--- a/drivers/net/ethernet/chelsio/cxgb/tp.h
+++ b/drivers/net/ethernet/chelsio/cxgb/tp.h
@@ -1,3 +1,4 @@
+/* SPDX-License-Identifier: GPL-2.0 */
 /* $Date: 2005/03/07 23:59:05 $ $RCSfile: tp.h,v $ $Revision: 1.20 $ */
 #ifndef CHELSIO_TP_H
 #define CHELSIO_TP_H
@@ -64,9 +65,7 @@ void t1_tp_intr_enable(struct petp *tp);
 void t1_tp_intr_clear(struct petp *tp);
 int t1_tp_intr_handler(struct petp *tp);
 
-void t1_tp_get_mib_statistics(adapter_t *adap, struct tp_mib_statistics *tps);
 void t1_tp_set_tcp_checksum_offload(struct petp *tp, int enable);
 void t1_tp_set_ip_checksum_offload(struct petp *tp, int enable);
-int t1_tp_set_coalescing_size(struct petp *tp, unsigned int size);
 int t1_tp_reset(struct petp *tp, struct tp_params *p, unsigned int tp_clk);
 #endif
diff --git a/drivers/net/ethernet/chelsio/cxgb/vsc7326.c b/drivers/net/ethernet/chelsio/cxgb/vsc7326.c
index bdc895bd2a46..2ad3efb550c2 100644
--- a/drivers/net/ethernet/chelsio/cxgb/vsc7326.c
+++ b/drivers/net/ethernet/chelsio/cxgb/vsc7326.c
@@ -1,3 +1,4 @@
+// SPDX-License-Identifier: GPL-2.0
 /* $Date: 2006/04/28 19:20:06 $ $RCSfile: vsc7326.c,v $ $Revision: 1.19 $ */
 
 /* Driver for Vitesse VSC7326 (Schaumburg) MAC */
@@ -378,7 +379,7 @@ static int mac_intr_clear(struct cmac *mac)
 }
 
 /* Expect MAC address to be in network byte order. */
-static int mac_set_address(struct cmac* mac, u8 addr[6])
+static int mac_set_address(struct cmac* mac, const u8 addr[6])
 {
 	u32 val;
 	int port = mac->instance->index;
@@ -590,7 +591,7 @@ static void port_stats_update(struct cmac *mac)
 	} hw_stats[] = {
 
 #define HW_STAT(reg, stat_name) \
-	{ reg, (&((struct cmac_statistics *)NULL)->stat_name) - (u64 *)NULL }
+	{ reg, offsetof(struct cmac_statistics, stat_name) / sizeof(u64) }
 
 		/* Rx stats */
 		HW_STAT(RxUnicast, RxUnicastFramesOK),
diff --git a/drivers/net/ethernet/chelsio/cxgb/vsc7326_reg.h b/drivers/net/ethernet/chelsio/cxgb/vsc7326_reg.h
index 479edbcabe68..04503857c6a9 100644
--- a/drivers/net/ethernet/chelsio/cxgb/vsc7326_reg.h
+++ b/drivers/net/ethernet/chelsio/cxgb/vsc7326_reg.h
@@ -1,3 +1,4 @@
+/* SPDX-License-Identifier: GPL-2.0 */
 /* $Date: 2006/04/28 19:20:17 $ $RCSfile: vsc7326_reg.h,v $ $Revision: 1.5 $ */
 #ifndef _VSC7321_REG_H_
 #define _VSC7321_REG_H_
diff --git a/drivers/net/ethernet/chelsio/cxgb3/Makefile b/drivers/net/ethernet/chelsio/cxgb3/Makefile
index 29aff78c7820..f65f0d93be42 100644
--- a/drivers/net/ethernet/chelsio/cxgb3/Makefile
+++ b/drivers/net/ethernet/chelsio/cxgb3/Makefile
@@ -1,3 +1,4 @@
+# SPDX-License-Identifier: GPL-2.0-only
 #
 # Chelsio T3 driver
 #
diff --git a/drivers/net/ethernet/chelsio/cxgb3/adapter.h b/drivers/net/ethernet/chelsio/cxgb3/adapter.h
index 087ff0ffb597..9d11e55981a0 100644
--- a/drivers/net/ethernet/chelsio/cxgb3/adapter.h
+++ b/drivers/net/ethernet/chelsio/cxgb3/adapter.h
@@ -178,7 +178,7 @@ struct sge_txq {		/* state for an SGE Tx queue */
 	unsigned int token;	/* WR token */
 	dma_addr_t phys_addr;	/* physical address of the ring */
 	struct sk_buff_head sendq;	/* List of backpressured offload packets */
-	struct tasklet_struct qresume_tsk;	/* restarts the queue */
+	struct work_struct qresume_task;	/* restarts the queue */
 	unsigned int cntxt_id;	/* SGE context id for the Tx q */
 	unsigned long stops;	/* # of times q has been stopped */
 	unsigned long restarts;	/* # of queue restarts */
@@ -237,7 +237,7 @@ struct adapter {
 	int msix_nvectors;
 	struct {
 		unsigned short vec;
-		char desc[22];
+		char desc[IFNAMSIZ + 1 + 12];	/* Needs space for "%s-%d" */
 	} msix_info[SGE_QSETS + 1];
 
 	/* T3 modules */
@@ -313,6 +313,7 @@ void t3_os_link_fault(struct adapter *adapter, int port_id, int state);
 void t3_os_link_fault_handler(struct adapter *adapter, int port_id);
 
 void t3_sge_start(struct adapter *adap);
+void t3_sge_stop_dma(struct adapter *adap);
 void t3_sge_stop(struct adapter *adap);
 void t3_start_sge_timers(struct adapter *adap);
 void t3_stop_sge_timers(struct adapter *adap);
diff --git a/drivers/net/ethernet/chelsio/cxgb3/ael1002.c b/drivers/net/ethernet/chelsio/cxgb3/ael1002.c
index dadf11e3dddb..9d591f0ddfc5 100644
--- a/drivers/net/ethernet/chelsio/cxgb3/ael1002.c
+++ b/drivers/net/ethernet/chelsio/cxgb3/ael1002.c
@@ -815,17 +815,12 @@ static const struct cphy_ops ael2020_ops = {
 int t3_ael2020_phy_prep(struct cphy *phy, struct adapter *adapter, int phy_addr,
 			const struct mdio_ops *mdio_ops)
 {
-	int err;
-
 	cphy_init(phy, adapter, phy_addr, &ael2020_ops, mdio_ops,
 		  SUPPORTED_10000baseT_Full | SUPPORTED_AUI | SUPPORTED_FIBRE |
 		  SUPPORTED_IRQ, "10GBASE-R");
 	msleep(125);
 
-	err = set_phy_regs(phy, ael2020_reset_regs);
-	if (err)
-		return err;
-	return 0;
+	return set_phy_regs(phy, ael2020_reset_regs);
 }
 
 /*
diff --git a/drivers/net/ethernet/chelsio/cxgb3/common.h b/drivers/net/ethernet/chelsio/cxgb3/common.h
index 1bd7d89666c4..ecd025dda8d6 100644
--- a/drivers/net/ethernet/chelsio/cxgb3/common.h
+++ b/drivers/net/ethernet/chelsio/cxgb3/common.h
@@ -676,8 +676,6 @@ void t3_link_changed(struct adapter *adapter, int port_id);
 void t3_link_fault(struct adapter *adapter, int port_id);
 int t3_link_start(struct cphy *phy, struct cmac *mac, struct link_config *lc);
 const struct adapter_info *t3_get_adapter_info(unsigned int board_id);
-int t3_seeprom_read(struct adapter *adapter, u32 addr, __le32 *data);
-int t3_seeprom_write(struct adapter *adapter, u32 addr, __le32 data);
 int t3_seeprom_wp(struct adapter *adapter, int enable);
 int t3_get_tp_version(struct adapter *adapter, u32 *vers);
 int t3_check_tpsram_version(struct adapter *adapter);
@@ -710,7 +708,7 @@ int t3_mac_enable(struct cmac *mac, int which);
 int t3_mac_disable(struct cmac *mac, int which);
 int t3_mac_set_mtu(struct cmac *mac, unsigned int mtu);
 int t3_mac_set_rx_mode(struct cmac *mac, struct net_device *dev);
-int t3_mac_set_address(struct cmac *mac, unsigned int idx, u8 addr[6]);
+int t3_mac_set_address(struct cmac *mac, unsigned int idx, const u8 addr[6]);
 int t3_mac_set_num_ucast(struct cmac *mac, int n);
 const struct mac_stats *t3_mac_update_stats(struct cmac *mac);
 int t3_mac_set_speed_duplex_fc(struct cmac *mac, int speed, int duplex, int fc);
@@ -770,4 +768,6 @@ int t3_xaui_direct_phy_prep(struct cphy *phy, struct adapter *adapter,
 			    int phy_addr, const struct mdio_ops *mdio_ops);
 int t3_aq100x_phy_prep(struct cphy *phy, struct adapter *adapter,
 			    int phy_addr, const struct mdio_ops *mdio_ops);
+
+extern struct workqueue_struct *cxgb3_wq;
 #endif				/* __CHELSIO_COMMON_H */
diff --git a/drivers/net/ethernet/chelsio/cxgb3/cxgb3_defs.h b/drivers/net/ethernet/chelsio/cxgb3/cxgb3_defs.h
index f04e81f33795..a08fc762a438 100644
--- a/drivers/net/ethernet/chelsio/cxgb3/cxgb3_defs.h
+++ b/drivers/net/ethernet/chelsio/cxgb3/cxgb3_defs.h
@@ -106,6 +106,4 @@ static inline struct t3c_tid_entry *lookup_atid(const struct tid_info *t,
 	return &e->t3c_tid;
 }
 
-int attach_t3cdev(struct t3cdev *dev);
-void detach_t3cdev(struct t3cdev *dev);
 #endif
diff --git a/drivers/net/ethernet/chelsio/cxgb3/cxgb3_ioctl.h b/drivers/net/ethernet/chelsio/cxgb3/cxgb3_ioctl.h
index b19e4376ba76..401827b82aa1 100644
--- a/drivers/net/ethernet/chelsio/cxgb3/cxgb3_ioctl.h
+++ b/drivers/net/ethernet/chelsio/cxgb3/cxgb3_ioctl.h
@@ -79,7 +79,7 @@ struct ch_mem_range {
 	uint32_t addr;
 	uint32_t len;
 	uint32_t version;
-	uint8_t buf[0];
+	uint8_t buf[];
 };
 
 struct ch_qset_params {
diff --git a/drivers/net/ethernet/chelsio/cxgb3/cxgb3_main.c b/drivers/net/ethernet/chelsio/cxgb3/cxgb3_main.c
index 2ff6bd139c96..f92a3550e480 100644
--- a/drivers/net/ethernet/chelsio/cxgb3/cxgb3_main.c
+++ b/drivers/net/ethernet/chelsio/cxgb3/cxgb3_main.c
@@ -33,7 +33,6 @@
 #define pr_fmt(fmt) KBUILD_MODNAME ": " fmt
 
 #include <linux/module.h>
-#include <linux/moduleparam.h>
 #include <linux/init.h>
 #include <linux/pci.h>
 #include <linux/dma-mapping.h>
@@ -51,6 +50,7 @@
 #include <linux/sched.h>
 #include <linux/slab.h>
 #include <linux/uaccess.h>
+#include <linux/nospec.h>
 
 #include "common.h"
 #include "cxgb3_ioctl.h"
@@ -105,7 +105,6 @@ static const struct pci_device_id cxgb3_pci_tbl[] = {
 MODULE_DESCRIPTION(DRV_DESC);
 MODULE_AUTHOR("Chelsio Communications");
 MODULE_LICENSE("Dual BSD/GPL");
-MODULE_VERSION(DRV_VERSION);
 MODULE_DEVICE_TABLE(pci, cxgb3_pci_tbl);
 
 static int dflt_msg_enable = DFLT_MSG_ENABLE;
@@ -149,7 +148,7 @@ struct workqueue_struct *cxgb3_wq;
 
 /**
  *	link_report - show link status and link speed/duplex
- *	@p: the port whose settings are to be reported
+ *	@dev: the port whose settings are to be reported
  *
  *	Shows the link status, speed, and duplex of a port.
  */
@@ -305,8 +304,8 @@ void t3_os_link_changed(struct adapter *adapter, int port_id, int link_stat,
 
 /**
  *	t3_os_phymod_changed - handle PHY module changes
- *	@phy: the PHY reporting the module change
- *	@mod_type: new module type
+ *	@adap: the adapter associated with the link change
+ *	@port_id: the port index whose limk status has changed
  *
  *	This is the OS-dependent handler for PHY module changes.  It is
  *	invoked when a PHY module is removed or inserted for any OS-specific
@@ -381,19 +380,18 @@ static irqreturn_t t3_async_intr_handler(int irq, void *cookie)
  */
 static void name_msix_vecs(struct adapter *adap)
 {
-	int i, j, msi_idx = 1, n = sizeof(adap->msix_info[0].desc) - 1;
+	int i, j, msi_idx = 1;
 
-	snprintf(adap->msix_info[0].desc, n, "%s", adap->name);
-	adap->msix_info[0].desc[n] = 0;
+	strscpy(adap->msix_info[0].desc, adap->name, sizeof(adap->msix_info[0].desc));
 
 	for_each_port(adap, j) {
 		struct net_device *d = adap->port[j];
 		const struct port_info *pi = netdev_priv(d);
 
 		for (i = 0; i < pi->nqsets; i++, msi_idx++) {
-			snprintf(adap->msix_info[msi_idx].desc, n,
+			snprintf(adap->msix_info[msi_idx].desc,
+				 sizeof(adap->msix_info[0].desc),
 				 "%s-%d", d->name, pi->first_qset + i);
-			adap->msix_info[msi_idx].desc[n] = 0;
 		}
 	}
 }
@@ -471,8 +469,7 @@ static int init_tp_parity(struct adapter *adap)
 		if (!skb)
 			goto alloc_skb_fail;
 
-		req = (struct cpl_smt_write_req *)__skb_put(skb, sizeof(*req));
-		memset(req, 0, sizeof(*req));
+		req = __skb_put_zero(skb, sizeof(*req));
 		req->wr.wr_hi = htonl(V_WR_OP(FW_WROPCODE_FORWARD));
 		OPCODE_TID(req) = htonl(MK_OPCODE_TID(CPL_SMT_WRITE_REQ, i));
 		req->mtu_idx = NMTUS - 1;
@@ -495,8 +492,7 @@ static int init_tp_parity(struct adapter *adap)
 		if (!skb)
 			goto alloc_skb_fail;
 
-		req = (struct cpl_l2t_write_req *)__skb_put(skb, sizeof(*req));
-		memset(req, 0, sizeof(*req));
+		req = __skb_put_zero(skb, sizeof(*req));
 		req->wr.wr_hi = htonl(V_WR_OP(FW_WROPCODE_FORWARD));
 		OPCODE_TID(req) = htonl(MK_OPCODE_TID(CPL_L2T_WRITE_REQ, i));
 		req->params = htonl(V_L2T_W_IDX(i));
@@ -518,8 +514,7 @@ static int init_tp_parity(struct adapter *adap)
 		if (!skb)
 			goto alloc_skb_fail;
 
-		req = (struct cpl_rte_write_req *)__skb_put(skb, sizeof(*req));
-		memset(req, 0, sizeof(*req));
+		req = __skb_put_zero(skb, sizeof(*req));
 		req->wr.wr_hi = htonl(V_WR_OP(FW_WROPCODE_FORWARD));
 		OPCODE_TID(req) = htonl(MK_OPCODE_TID(CPL_RTE_WRITE_REQ, i));
 		req->l2t_idx = htonl(V_L2T_W_IDX(i));
@@ -538,8 +533,7 @@ static int init_tp_parity(struct adapter *adap)
 	if (!skb)
 		goto alloc_skb_fail;
 
-	greq = (struct cpl_set_tcb_field *)__skb_put(skb, sizeof(*greq));
-	memset(greq, 0, sizeof(*greq));
+	greq = __skb_put_zero(skb, sizeof(*greq));
 	greq->wr.wr_hi = htonl(V_WR_OP(FW_WROPCODE_FORWARD));
 	OPCODE_TID(greq) = htonl(MK_OPCODE_TID(CPL_SET_TCB_FIELD, 0));
 	greq->mask = cpu_to_be64(1);
@@ -614,8 +608,7 @@ static void init_napi(struct adapter *adap)
 		struct sge_qset *qs = &adap->sge.qs[i];
 
 		if (qs->adap)
-			netif_napi_add(qs->netdev, &qs->napi, qs->napi.poll,
-				       64);
+			netif_napi_add(qs->netdev, &qs->napi, qs->napi.poll);
 	}
 
 	/*
@@ -780,11 +773,11 @@ static ssize_t store_nservers(struct device *d, struct device_attribute *attr,
 
 #define CXGB3_ATTR_R(name, val_expr) \
 CXGB3_SHOW(name, val_expr) \
-static DEVICE_ATTR(name, S_IRUGO, show_##name, NULL)
+static DEVICE_ATTR(name, 0444, show_##name, NULL)
 
 #define CXGB3_ATTR_RW(name, val_expr, store_method) \
 CXGB3_SHOW(name, val_expr) \
-static DEVICE_ATTR(name, S_IRUGO | S_IWUSR, show_##name, store_method)
+static DEVICE_ATTR(name, 0644, show_##name, store_method)
 
 CXGB3_ATTR_R(cam_size, t3_mc5_size(&adap->mc5));
 CXGB3_ATTR_RW(nfilters, adap->params.mc5.nfilters, store_nfilters);
@@ -797,7 +790,9 @@ static struct attribute *cxgb3_attrs[] = {
 	NULL
 };
 
-static struct attribute_group cxgb3_attr_group = {.attrs = cxgb3_attrs };
+static const struct attribute_group cxgb3_attr_group = {
+	.attrs = cxgb3_attrs,
+};
 
 static ssize_t tm_attr_show(struct device *d,
 			    char *buf, int sched)
@@ -861,7 +856,7 @@ static ssize_t store_##name(struct device *d, struct device_attribute *attr, \
 { \
 	return tm_attr_store(d, buf, len, sched); \
 } \
-static DEVICE_ATTR(name, S_IRUGO | S_IWUSR, show_##name, store_##name)
+static DEVICE_ATTR(name, 0644, show_##name, store_##name)
 
 TM_ATTR(sched0, 0);
 TM_ATTR(sched1, 1);
@@ -884,7 +879,9 @@ static struct attribute *offload_attrs[] = {
 	NULL
 };
 
-static struct attribute_group offload_attr_group = {.attrs = offload_attrs };
+static const struct attribute_group offload_attr_group = {
+	.attrs = offload_attrs,
+};
 
 /*
  * Sends an sk_buff to an offload queue driver
@@ -909,7 +906,7 @@ static int write_smt_entry(struct adapter *adapter, int idx)
 	if (!skb)
 		return -ENOMEM;
 
-	req = (struct cpl_smt_write_req *)__skb_put(skb, sizeof(*req));
+	req = __skb_put(skb, sizeof(*req));
 	req->wr.wr_hi = htonl(V_WR_OP(FW_WROPCODE_FORWARD));
 	OPCODE_TID(req) = htonl(MK_OPCODE_TID(CPL_SMT_WRITE_REQ, idx));
 	req->mtu_idx = NMTUS - 1;	/* should be 0 but there's a T3 bug */
@@ -952,7 +949,7 @@ static int send_pktsched_cmd(struct adapter *adap, int sched, int qidx, int lo,
 	if (!skb)
 		return -ENOMEM;
 
-	req = (struct mngt_pktsched_wr *)skb_put(skb, sizeof(*req));
+	req = skb_put(skb, sizeof(*req));
 	req->wr_hi = htonl(V_WR_OP(FW_WROPCODE_MNGT));
 	req->mngt_opcode = FW_MNGTOPCODE_PKTSCHED_SET;
 	req->sched = sched;
@@ -1201,7 +1198,7 @@ static void cxgb_vlan_mode(struct net_device *dev, netdev_features_t features)
 
 /**
  *	cxgb_up - enable the adapter
- *	@adapter: adapter being enabled
+ *	@adap: adapter being enabled
  *
  *	Called when the first port is enabled, this function performs the
  *	actions necessary to make an adapter operational, such as completing
@@ -1274,14 +1271,14 @@ static int cxgb_up(struct adapter *adap)
 			free_irq(adap->msix_info[0].vec, adap);
 			goto irq_err;
 		}
-	} else if ((err = request_irq(adap->pdev->irq,
-				      t3_intr_handler(adap,
-						      adap->sge.qs[0].rspq.
-						      polling),
-				      (adap->flags & USING_MSI) ?
-				       0 : IRQF_SHARED,
-				      adap->name, adap)))
-		goto irq_err;
+	} else {
+		err = request_irq(adap->pdev->irq,
+				  t3_intr_handler(adap, adap->sge.qs[0].rspq.polling),
+				  (adap->flags & USING_MSI) ? 0 : IRQF_SHARED,
+				  adap->name, adap);
+		if (err)
+			goto irq_err;
+	}
 
 	enable_all_napi(adap);
 	t3_sge_start(adap);
@@ -1303,6 +1300,7 @@ static int cxgb_up(struct adapter *adap)
 		if (ret < 0) {
 			CH_ERR(adap, "failed to bind qsets, err %d\n", ret);
 			t3_intr_disable(adap);
+			quiesce_rx(adap);
 			free_irq_resources(adap);
 			err = ret;
 			goto out;
@@ -1628,9 +1626,8 @@ static void get_drvinfo(struct net_device *dev, struct ethtool_drvinfo *info)
 	t3_get_tp_version(adapter, &tp_vers);
 	spin_unlock(&adapter->stats_lock);
 
-	strlcpy(info->driver, DRV_NAME, sizeof(info->driver));
-	strlcpy(info->version, DRV_VERSION, sizeof(info->version));
-	strlcpy(info->bus_info, pci_name(adapter->pdev),
+	strscpy(info->driver, DRV_NAME, sizeof(info->driver));
+	strscpy(info->bus_info, pci_name(adapter->pdev),
 		sizeof(info->bus_info));
 	if (fw_vers)
 		snprintf(info->fw_version, sizeof(info->fw_version),
@@ -1950,7 +1947,9 @@ static int set_pauseparam(struct net_device *dev,
 	return 0;
 }
 
-static void get_sge_param(struct net_device *dev, struct ethtool_ringparam *e)
+static void get_sge_param(struct net_device *dev, struct ethtool_ringparam *e,
+			  struct kernel_ethtool_ringparam *kernel_e,
+			  struct netlink_ext_ack *extack)
 {
 	struct port_info *pi = netdev_priv(dev);
 	struct adapter *adapter = pi->adapter;
@@ -1966,7 +1965,9 @@ static void get_sge_param(struct net_device *dev, struct ethtool_ringparam *e)
 	e->tx_pending = q->txq_size[0];
 }
 
-static int set_sge_param(struct net_device *dev, struct ethtool_ringparam *e)
+static int set_sge_param(struct net_device *dev, struct ethtool_ringparam *e,
+			 struct kernel_ethtool_ringparam *kernel_e,
+			 struct netlink_ext_ack *extack)
 {
 	struct port_info *pi = netdev_priv(dev);
 	struct adapter *adapter = pi->adapter;
@@ -1998,7 +1999,9 @@ static int set_sge_param(struct net_device *dev, struct ethtool_ringparam *e)
 	return 0;
 }
 
-static int set_coalesce(struct net_device *dev, struct ethtool_coalesce *c)
+static int set_coalesce(struct net_device *dev, struct ethtool_coalesce *c,
+			struct kernel_ethtool_coalesce *kernel_coal,
+			struct netlink_ext_ack *extack)
 {
 	struct port_info *pi = netdev_priv(dev);
 	struct adapter *adapter = pi->adapter;
@@ -2019,7 +2022,9 @@ static int set_coalesce(struct net_device *dev, struct ethtool_coalesce *c)
 	return 0;
 }
 
-static int get_coalesce(struct net_device *dev, struct ethtool_coalesce *c)
+static int get_coalesce(struct net_device *dev, struct ethtool_coalesce *c,
+			struct kernel_ethtool_coalesce *kernel_coal,
+			struct netlink_ext_ack *extack)
 {
 	struct port_info *pi = netdev_priv(dev);
 	struct adapter *adapter = pi->adapter;
@@ -2034,20 +2039,16 @@ static int get_eeprom(struct net_device *dev, struct ethtool_eeprom *e,
 {
 	struct port_info *pi = netdev_priv(dev);
 	struct adapter *adapter = pi->adapter;
-	int i, err = 0;
-
-	u8 *buf = kmalloc(EEPROMSIZE, GFP_KERNEL);
-	if (!buf)
-		return -ENOMEM;
+	int cnt;
 
 	e->magic = EEPROM_MAGIC;
-	for (i = e->offset & ~3; !err && i < e->offset + e->len; i += 4)
-		err = t3_seeprom_read(adapter, i, (__le32 *) & buf[i]);
+	cnt = pci_read_vpd(adapter->pdev, e->offset, e->len, data);
+	if (cnt < 0)
+		return cnt;
 
-	if (!err)
-		memcpy(data, buf + e->offset, e->len);
-	kfree(buf);
-	return err;
+	e->len = cnt;
+
+	return 0;
 }
 
 static int set_eeprom(struct net_device *dev, struct ethtool_eeprom *eeprom,
@@ -2056,7 +2057,6 @@ static int set_eeprom(struct net_device *dev, struct ethtool_eeprom *eeprom,
 	struct port_info *pi = netdev_priv(dev);
 	struct adapter *adapter = pi->adapter;
 	u32 aligned_offset, aligned_len;
-	__le32 *p;
 	u8 *buf;
 	int err;
 
@@ -2070,12 +2070,9 @@ static int set_eeprom(struct net_device *dev, struct ethtool_eeprom *eeprom,
 		buf = kmalloc(aligned_len, GFP_KERNEL);
 		if (!buf)
 			return -ENOMEM;
-		err = t3_seeprom_read(adapter, aligned_offset, (__le32 *) buf);
-		if (!err && aligned_len > 4)
-			err = t3_seeprom_read(adapter,
-					      aligned_offset + aligned_len - 4,
-					      (__le32 *) & buf[aligned_len - 4]);
-		if (err)
+		err = pci_read_vpd(adapter->pdev, aligned_offset, aligned_len,
+				   buf);
+		if (err < 0)
 			goto out;
 		memcpy(buf + (eeprom->offset & 3), data, eeprom->len);
 	} else
@@ -2085,17 +2082,13 @@ static int set_eeprom(struct net_device *dev, struct ethtool_eeprom *eeprom,
 	if (err)
 		goto out;
 
-	for (p = (__le32 *) buf; !err && aligned_len; aligned_len -= 4, p++) {
-		err = t3_seeprom_write(adapter, aligned_offset, *p);
-		aligned_offset += 4;
-	}
-
-	if (!err)
+	err = pci_write_vpd(adapter->pdev, aligned_offset, aligned_len, buf);
+	if (err >= 0)
 		err = t3_seeprom_wp(adapter, 1);
 out:
 	if (buf != data)
 		kfree(buf);
-	return err;
+	return err < 0 ? err : 0;
 }
 
 static void get_wol(struct net_device *dev, struct ethtool_wolinfo *wol)
@@ -2106,6 +2099,7 @@ static void get_wol(struct net_device *dev, struct ethtool_wolinfo *wol)
 }
 
 static const struct ethtool_ops cxgb_ethtool_ops = {
+	.supported_coalesce_params = ETHTOOL_COALESCE_RX_USECS,
 	.get_drvinfo = get_drvinfo,
 	.get_msglevel = get_msglevel,
 	.set_msglevel = set_msglevel,
@@ -2131,18 +2125,23 @@ static const struct ethtool_ops cxgb_ethtool_ops = {
 	.set_link_ksettings = set_link_ksettings,
 };
 
-static int in_range(int val, int lo, int hi)
+static int cxgb_in_range(int val, int lo, int hi)
 {
 	return val < 0 || (val <= hi && val >= lo);
 }
 
-static int cxgb_extension_ioctl(struct net_device *dev, void __user *useraddr)
+static int cxgb_siocdevprivate(struct net_device *dev,
+			       struct ifreq *ifreq,
+			       void __user *useraddr,
+			       int cmd)
 {
 	struct port_info *pi = netdev_priv(dev);
 	struct adapter *adapter = pi->adapter;
-	u32 cmd;
 	int ret;
 
+	if (cmd != SIOCCHIOCTL)
+		return -EOPNOTSUPP;
+
 	if (copy_from_user(&cmd, useraddr, sizeof(cmd)))
 		return -EFAULT;
 
@@ -2158,21 +2157,23 @@ static int cxgb_extension_ioctl(struct net_device *dev, void __user *useraddr)
 			return -EPERM;
 		if (copy_from_user(&t, useraddr, sizeof(t)))
 			return -EFAULT;
+		if (t.cmd != CHELSIO_SET_QSET_PARAMS)
+			return -EINVAL;
 		if (t.qset_idx >= SGE_QSETS)
 			return -EINVAL;
-		if (!in_range(t.intr_lat, 0, M_NEWTIMER) ||
-		    !in_range(t.cong_thres, 0, 255) ||
-		    !in_range(t.txq_size[0], MIN_TXQ_ENTRIES,
+		if (!cxgb_in_range(t.intr_lat, 0, M_NEWTIMER) ||
+		    !cxgb_in_range(t.cong_thres, 0, 255) ||
+		    !cxgb_in_range(t.txq_size[0], MIN_TXQ_ENTRIES,
 			      MAX_TXQ_ENTRIES) ||
-		    !in_range(t.txq_size[1], MIN_TXQ_ENTRIES,
+		    !cxgb_in_range(t.txq_size[1], MIN_TXQ_ENTRIES,
 			      MAX_TXQ_ENTRIES) ||
-		    !in_range(t.txq_size[2], MIN_CTRL_TXQ_ENTRIES,
+		    !cxgb_in_range(t.txq_size[2], MIN_CTRL_TXQ_ENTRIES,
 			      MAX_CTRL_TXQ_ENTRIES) ||
-		    !in_range(t.fl_size[0], MIN_FL_ENTRIES,
+		    !cxgb_in_range(t.fl_size[0], MIN_FL_ENTRIES,
 			      MAX_RX_BUFFERS) ||
-		    !in_range(t.fl_size[1], MIN_FL_ENTRIES,
+		    !cxgb_in_range(t.fl_size[1], MIN_FL_ENTRIES,
 			      MAX_RX_JUMBO_BUFFERS) ||
-		    !in_range(t.rspq_size, MIN_RSPQ_ENTRIES,
+		    !cxgb_in_range(t.rspq_size, MIN_RSPQ_ENTRIES,
 			      MAX_RSPQ_ENTRIES))
 			return -EINVAL;
 
@@ -2257,6 +2258,9 @@ static int cxgb_extension_ioctl(struct net_device *dev, void __user *useraddr)
 		if (copy_from_user(&t, useraddr, sizeof(t)))
 			return -EFAULT;
 
+		if (t.cmd != CHELSIO_GET_QSET_PARAMS)
+			return -EINVAL;
+
 		/* Display qsets for all ports when offload enabled */
 		if (test_bit(OFFLOAD_DEVMAP_BIT, &adapter->open_device_map)) {
 			q1 = 0;
@@ -2268,6 +2272,7 @@ static int cxgb_extension_ioctl(struct net_device *dev, void __user *useraddr)
 
 		if (t.qset_idx >= nqsets)
 			return -EINVAL;
+		t.qset_idx = array_index_nospec(t.qset_idx, nqsets);
 
 		q = &adapter->params.sge.qset[q1 + t.qset_idx];
 		t.rspq_size = q->rspq_size;
@@ -2301,6 +2306,8 @@ static int cxgb_extension_ioctl(struct net_device *dev, void __user *useraddr)
 			return -EBUSY;
 		if (copy_from_user(&edata, useraddr, sizeof(edata)))
 			return -EFAULT;
+		if (edata.cmd != CHELSIO_SET_QSET_NUM)
+			return -EINVAL;
 		if (edata.val < 1 ||
 			(edata.val > 1 && !(adapter->flags & USING_MSIX)))
 			return -EINVAL;
@@ -2341,6 +2348,8 @@ static int cxgb_extension_ioctl(struct net_device *dev, void __user *useraddr)
 			return -EPERM;
 		if (copy_from_user(&t, useraddr, sizeof(t)))
 			return -EFAULT;
+		if (t.cmd != CHELSIO_LOAD_FW)
+			return -EINVAL;
 		/* Check t.len sanity ? */
 		fw_data = memdup_user(useraddr + sizeof(t), t.len);
 		if (IS_ERR(fw_data))
@@ -2364,6 +2373,8 @@ static int cxgb_extension_ioctl(struct net_device *dev, void __user *useraddr)
 			return -EBUSY;
 		if (copy_from_user(&m, useraddr, sizeof(m)))
 			return -EFAULT;
+		if (m.cmd != CHELSIO_SETMTUTAB)
+			return -EINVAL;
 		if (m.nmtus != NMTUS)
 			return -EINVAL;
 		if (m.mtus[0] < 81)	/* accommodate SACK */
@@ -2405,6 +2416,8 @@ static int cxgb_extension_ioctl(struct net_device *dev, void __user *useraddr)
 			return -EBUSY;
 		if (copy_from_user(&m, useraddr, sizeof(m)))
 			return -EFAULT;
+		if (m.cmd != CHELSIO_SET_PM)
+			return -EINVAL;
 		if (!is_power_of_2(m.rx_pg_sz) ||
 			!is_power_of_2(m.tx_pg_sz))
 			return -EINVAL;	/* not power of 2 */
@@ -2434,10 +2447,14 @@ static int cxgb_extension_ioctl(struct net_device *dev, void __user *useraddr)
 
 		if (!is_offload(adapter))
 			return -EOPNOTSUPP;
+		if (!capable(CAP_NET_ADMIN))
+			return -EPERM;
 		if (!(adapter->flags & FULL_INIT_DONE))
 			return -EIO;	/* need the memory controllers */
 		if (copy_from_user(&t, useraddr, sizeof(t)))
 			return -EFAULT;
+		if (t.cmd != CHELSIO_GET_MEM)
+			return -EINVAL;
 		if ((t.addr & 7) || (t.len & 7))
 			return -EINVAL;
 		if (t.mem_id == MEM_CM)
@@ -2490,6 +2507,8 @@ static int cxgb_extension_ioctl(struct net_device *dev, void __user *useraddr)
 			return -EAGAIN;
 		if (copy_from_user(&t, useraddr, sizeof(t)))
 			return -EFAULT;
+		if (t.cmd != CHELSIO_SET_TRACE_FILTER)
+			return -EINVAL;
 
 		tp = (const struct trace_params *)&t.sip;
 		if (t.config_tx)
@@ -2524,11 +2543,9 @@ static int cxgb_ioctl(struct net_device *dev, struct ifreq *req, int cmd)
 		    !(data->phy_id & 0xe0e0))
 			data->phy_id = mdio_phy_id_c45(data->phy_id >> 8,
 						       data->phy_id & 0x1f);
-		/* FALLTHRU */
+		fallthrough;
 	case SIOCGMIIPHY:
 		return mdio_mii_ioctl(&pi->phy.mdio, data, cmd);
-	case SIOCCHIOCTL:
-		return cxgb_extension_ioctl(dev, req->ifr_data);
 	default:
 		return -EOPNOTSUPP;
 	}
@@ -2542,7 +2559,7 @@ static int cxgb_change_mtu(struct net_device *dev, int new_mtu)
 
 	if ((ret = t3_mac_set_mtu(&pi->mac, new_mtu)))
 		return ret;
-	dev->mtu = new_mtu;
+	WRITE_ONCE(dev->mtu, new_mtu);
 	init_port_mtus(adapter);
 	if (adapter->params.rev == 0 && offload_running(adapter))
 		t3_load_mtus(adapter, adapter->params.mtus,
@@ -2560,7 +2577,7 @@ static int cxgb_set_mac_addr(struct net_device *dev, void *p)
 	if (!is_valid_ether_addr(addr->sa_data))
 		return -EADDRNOTAVAIL;
 
-	memcpy(dev->dev_addr, addr->sa_data, dev->addr_len);
+	eth_hw_addr_set(dev, addr->sa_data);
 	t3_mac_set_address(&pi->mac, LAN_MAC_IDX, dev->dev_addr);
 	if (offload_running(adapter))
 		write_smt_entry(adapter, pi->port_id);
@@ -2977,7 +2994,7 @@ void t3_fatal_err(struct adapter *adapter)
 	unsigned int fw_status[4];
 
 	if (adapter->flags & FULL_INIT_DONE) {
-		t3_sge_stop(adapter);
+		t3_sge_stop_dma(adapter);
 		t3_write_reg(adapter, A_XGM_TX_CTRL, 0);
 		t3_write_reg(adapter, A_XGM_RX_CTRL, 0);
 		t3_write_reg(adapter, XGM_REG(A_XGM_TX_CTRL, 1), 0);
@@ -3079,8 +3096,9 @@ static void set_nqsets(struct adapter *adap)
 			nqsets = num_cpus;
 		if (nqsets < 1 || hwports == 4)
 			nqsets = 1;
-	} else
+	} else {
 		nqsets = 1;
+	}
 
 	for_each_port(adap, i) {
 		struct port_info *pi = adap2pinfo(adap, i);
@@ -3161,7 +3179,8 @@ static const struct net_device_ops cxgb_netdev_ops = {
 	.ndo_get_stats		= cxgb_get_stats,
 	.ndo_validate_addr	= eth_validate_addr,
 	.ndo_set_rx_mode	= cxgb_set_rxmode,
-	.ndo_do_ioctl		= cxgb_ioctl,
+	.ndo_eth_ioctl		= cxgb_ioctl,
+	.ndo_siocdevprivate	= cxgb_siocdevprivate,
 	.ndo_change_mtu		= cxgb_change_mtu,
 	.ndo_set_mac_address	= cxgb_set_mac_addr,
 	.ndo_fix_features	= cxgb_fix_features,
@@ -3184,14 +3203,12 @@ static void cxgb3_init_iscsi_mac(struct net_device *dev)
 			NETIF_F_IPV6_CSUM | NETIF_F_HIGHDMA)
 static int init_one(struct pci_dev *pdev, const struct pci_device_id *ent)
 {
-	int i, err, pci_using_dac = 0;
+	int i, err;
 	resource_size_t mmio_start, mmio_len;
 	const struct adapter_info *ai;
 	struct adapter *adapter = NULL;
 	struct port_info *pi;
 
-	pr_info_once("%s - version %s\n", DRV_DESC, DRV_VERSION);
-
 	if (!cxgb3_wq) {
 		cxgb3_wq = create_singlethread_workqueue(DRV_NAME);
 		if (!cxgb3_wq) {
@@ -3213,15 +3230,8 @@ static int init_one(struct pci_dev *pdev, const struct pci_device_id *ent)
 		goto out_disable_device;
 	}
 
-	if (!pci_set_dma_mask(pdev, DMA_BIT_MASK(64))) {
-		pci_using_dac = 1;
-		err = pci_set_consistent_dma_mask(pdev, DMA_BIT_MASK(64));
-		if (err) {
-			dev_err(&pdev->dev, "unable to obtain 64-bit DMA for "
-			       "coherent allocations\n");
-			goto out_release_regions;
-		}
-	} else if ((err = pci_set_dma_mask(pdev, DMA_BIT_MASK(32))) != 0) {
+	err = dma_set_mask_and_coherent(&pdev->dev, DMA_BIT_MASK(64));
+	if (err) {
 		dev_err(&pdev->dev, "no usable DMA configuration\n");
 		goto out_release_regions;
 	}
@@ -3247,11 +3257,11 @@ static int init_one(struct pci_dev *pdev, const struct pci_device_id *ent)
 		goto out_free_adapter;
 	}
 
-	adapter->regs = ioremap_nocache(mmio_start, mmio_len);
+	adapter->regs = ioremap(mmio_start, mmio_len);
 	if (!adapter->regs) {
 		dev_err(&pdev->dev, "cannot map device registers\n");
 		err = -ENOMEM;
-		goto out_free_adapter;
+		goto out_free_adapter_nofail;
 	}
 
 	adapter->pdev = pdev;
@@ -3297,13 +3307,14 @@ static int init_one(struct pci_dev *pdev, const struct pci_device_id *ent)
 		netdev->features |= netdev->hw_features |
 				    NETIF_F_HW_VLAN_CTAG_TX;
 		netdev->vlan_features |= netdev->features & VLAN_FEAT;
-		if (pci_using_dac)
-			netdev->features |= NETIF_F_HIGHDMA;
+
+		netdev->features |= NETIF_F_HIGHDMA;
 
 		netdev->netdev_ops = &cxgb_netdev_ops;
 		netdev->ethtool_ops = &cxgb_ethtool_ops;
 		netdev->min_mtu = 81;
 		netdev->max_mtu = ETH_MAX_MTU;
+		netdev->dev_port = pi->port_id;
 	}
 
 	pci_set_drvdata(pdev, adapter);
@@ -3337,6 +3348,7 @@ static int init_one(struct pci_dev *pdev, const struct pci_device_id *ent)
 	}
 	if (!adapter->registered_device_map) {
 		dev_err(&pdev->dev, "could not register any net devices\n");
+		err = -ENODEV;
 		goto out_free_dev;
 	}
 
@@ -3361,16 +3373,26 @@ static int init_one(struct pci_dev *pdev, const struct pci_device_id *ent)
 
 	err = sysfs_create_group(&adapter->port[0]->dev.kobj,
 				 &cxgb3_attr_group);
+	if (err) {
+		dev_err(&pdev->dev, "cannot create sysfs group\n");
+		goto out_close_led;
+	}
 
 	print_port_info(adapter, ai);
 	return 0;
 
+out_close_led:
+	t3_set_reg_field(adapter, A_T3DBG_GPIO_EN, F_GPIO0_OUT_VAL, 0);
+
 out_free_dev:
 	iounmap(adapter->regs);
 	for (i = ai->nports0 + ai->nports1 - 1; i >= 0; --i)
 		if (adapter->port[i])
 			free_netdev(adapter->port[i]);
 
+out_free_adapter_nofail:
+	kfree_skb(adapter->nofail_skb);
+
 out_free_adapter:
 	kfree(adapter);
 
@@ -3413,8 +3435,7 @@ static void remove_one(struct pci_dev *pdev)
 				free_netdev(adapter->port[i]);
 
 		iounmap(adapter->regs);
-		if (adapter->nofail_skb)
-			kfree_skb(adapter->nofail_skb);
+		kfree_skb(adapter->nofail_skb);
 		kfree(adapter);
 		pci_release_regions(pdev);
 		pci_disable_device(pdev);
diff --git a/drivers/net/ethernet/chelsio/cxgb3/cxgb3_offload.c b/drivers/net/ethernet/chelsio/cxgb3/cxgb3_offload.c
index fa81445e334c..5a9f6925e1fa 100644
--- a/drivers/net/ethernet/chelsio/cxgb3/cxgb3_offload.c
+++ b/drivers/net/ethernet/chelsio/cxgb3/cxgb3_offload.c
@@ -243,7 +243,7 @@ static int cxgb_ulp_iscsi_ctl(struct adapter *adapter, unsigned int req,
 
 		/*
 		 * on rx, the iscsi pdu has to be < rx page size and the
-		 * the max rx data length programmed in TP
+		 * max rx data length programmed in TP
 		 */
 		val = min(adapter->params.tp.rx_pg_size,
 			  ((t3_read_reg(adapter, A_TP_PARA_REG2)) >>
@@ -515,23 +515,6 @@ void *cxgb3_free_atid(struct t3cdev *tdev, int atid)
 
 EXPORT_SYMBOL(cxgb3_free_atid);
 
-/*
- * Free a server TID and return it to the free pool.
- */
-void cxgb3_free_stid(struct t3cdev *tdev, int stid)
-{
-	struct tid_info *t = &(T3C_DATA(tdev))->tid_maps;
-	union listen_entry *p = stid2entry(t, stid);
-
-	spin_lock_bh(&t->stid_lock);
-	p->next = t->sfree;
-	t->sfree = p;
-	t->stids_in_use--;
-	spin_unlock_bh(&t->stid_lock);
-}
-
-EXPORT_SYMBOL(cxgb3_free_stid);
-
 void cxgb3_insert_tid(struct t3cdev *tdev, struct cxgb3_client *client,
 		      void *ctx, unsigned int tid)
 {
@@ -552,7 +535,7 @@ static inline void mk_tid_release(struct sk_buff *skb, unsigned int tid)
 	struct cpl_tid_release *req;
 
 	skb->priority = CPL_PRIORITY_SETUP;
-	req = (struct cpl_tid_release *)__skb_put(skb, sizeof(*req));
+	req = __skb_put(skb, sizeof(*req));
 	req->wr.wr_hi = htonl(V_WR_OP(FW_WROPCODE_FORWARD));
 	OPCODE_TID(req) = htonl(MK_OPCODE_TID(CPL_TID_RELEASE, tid));
 }
@@ -671,28 +654,6 @@ int cxgb3_alloc_atid(struct t3cdev *tdev, struct cxgb3_client *client,
 
 EXPORT_SYMBOL(cxgb3_alloc_atid);
 
-int cxgb3_alloc_stid(struct t3cdev *tdev, struct cxgb3_client *client,
-		     void *ctx)
-{
-	int stid = -1;
-	struct tid_info *t = &(T3C_DATA(tdev))->tid_maps;
-
-	spin_lock_bh(&t->stid_lock);
-	if (t->sfree) {
-		union listen_entry *p = t->sfree;
-
-		stid = (p - t->stid_tab) + t->stid_base;
-		t->sfree = p->next;
-		p->t3c_tid.ctx = ctx;
-		p->t3c_tid.client = client;
-		t->stids_in_use++;
-	}
-	spin_unlock_bh(&t->stid_lock);
-	return stid;
-}
-
-EXPORT_SYMBOL(cxgb3_alloc_stid);
-
 /* Get the t3cdev associated with a net_device */
 struct t3cdev *dev2t3cdev(struct net_device *dev)
 {
@@ -1096,7 +1057,7 @@ static void set_l2t_ix(struct t3cdev *tdev, u32 tid, struct l2t_entry *e)
 		return;
 	}
 	skb->priority = CPL_PRIORITY_CONTROL;
-	req = (struct cpl_set_tcb_field *)skb_put(skb, sizeof(*req));
+	req = skb_put(skb, sizeof(*req));
 	req->wr.wr_hi = htonl(V_WR_OP(FW_WROPCODE_FORWARD));
 	OPCODE_TID(req) = htonl(MK_OPCODE_TID(CPL_SET_TCB_FIELD, tid));
 	req->reply = 0;
@@ -1302,8 +1263,7 @@ void cxgb3_offload_deactivate(struct adapter *adapter)
 	rcu_read_unlock();
 	RCU_INIT_POINTER(tdev->l2opt, NULL);
 	call_rcu(&d->rcu_head, clean_l2_data);
-	if (t->nofail_skb)
-		kfree_skb(t->nofail_skb);
+	kfree_skb(t->nofail_skb);
 	kfree(t);
 }
 
diff --git a/drivers/net/ethernet/chelsio/cxgb3/cxgb3_offload.h b/drivers/net/ethernet/chelsio/cxgb3/cxgb3_offload.h
index 929c298115ca..7419824f9926 100644
--- a/drivers/net/ethernet/chelsio/cxgb3/cxgb3_offload.h
+++ b/drivers/net/ethernet/chelsio/cxgb3/cxgb3_offload.h
@@ -95,10 +95,7 @@ struct cxgb3_client {
  */
 int cxgb3_alloc_atid(struct t3cdev *dev, struct cxgb3_client *client,
 		     void *ctx);
-int cxgb3_alloc_stid(struct t3cdev *dev, struct cxgb3_client *client,
-		     void *ctx);
 void *cxgb3_free_atid(struct t3cdev *dev, int atid);
-void cxgb3_free_stid(struct t3cdev *dev, int stid);
 void cxgb3_insert_tid(struct t3cdev *dev, struct cxgb3_client *client,
 		      void *ctx, unsigned int tid);
 void cxgb3_queue_tid_release(struct t3cdev *dev, unsigned int tid);
diff --git a/drivers/net/ethernet/chelsio/cxgb3/l2t.c b/drivers/net/ethernet/chelsio/cxgb3/l2t.c
index 26264125865f..5d5f3380ecca 100644
--- a/drivers/net/ethernet/chelsio/cxgb3/l2t.c
+++ b/drivers/net/ethernet/chelsio/cxgb3/l2t.c
@@ -96,7 +96,7 @@ static int setup_l2e_send_pending(struct t3cdev *dev, struct sk_buff *skb,
 			return -ENOMEM;
 	}
 
-	req = (struct cpl_l2t_write_req *)__skb_put(skb, sizeof(*req));
+	req = __skb_put(skb, sizeof(*req));
 	req->wr.wr_hi = htonl(V_WR_OP(FW_WROPCODE_FORWARD));
 	OPCODE_TID(req) = htonl(MK_OPCODE_TID(CPL_L2T_WRITE_REQ, e->idx));
 	req->params = htonl(V_L2T_W_IDX(e->idx) | V_L2T_W_IFF(e->smt_idx) |
@@ -136,6 +136,7 @@ again:
 		if (e->state == L2T_STATE_STALE)
 			e->state = L2T_STATE_VALID;
 		spin_unlock_bh(&e->lock);
+		fallthrough;
 	case L2T_STATE_VALID:	/* fast-path, send the packet on */
 		return cxgb3_ofld_send(dev, skb);
 	case L2T_STATE_RESOLVING:
@@ -175,43 +176,6 @@ again:
 
 EXPORT_SYMBOL(t3_l2t_send_slow);
 
-void t3_l2t_send_event(struct t3cdev *dev, struct l2t_entry *e)
-{
-again:
-	switch (e->state) {
-	case L2T_STATE_STALE:	/* entry is stale, kick off revalidation */
-		neigh_event_send(e->neigh, NULL);
-		spin_lock_bh(&e->lock);
-		if (e->state == L2T_STATE_STALE) {
-			e->state = L2T_STATE_VALID;
-		}
-		spin_unlock_bh(&e->lock);
-		return;
-	case L2T_STATE_VALID:	/* fast-path, send the packet on */
-		return;
-	case L2T_STATE_RESOLVING:
-		spin_lock_bh(&e->lock);
-		if (e->state != L2T_STATE_RESOLVING) {
-			/* ARP already completed */
-			spin_unlock_bh(&e->lock);
-			goto again;
-		}
-		spin_unlock_bh(&e->lock);
-
-		/*
-		 * Only the first packet added to the arpq should kick off
-		 * resolution.  However, because the alloc_skb below can fail,
-		 * we allow each packet added to the arpq to retry resolution
-		 * as a way of recovering from transient memory exhaustion.
-		 * A better way would be to use a work request to retry L2T
-		 * entries when there's no memory.
-		 */
-		neigh_event_send(e->neigh, NULL);
-	}
-}
-
-EXPORT_SYMBOL(t3_l2t_send_event);
-
 /*
  * Allocate a free L2T entry.  Must be called with l2t_data.lock held.
  */
@@ -442,9 +406,9 @@ found:
 struct l2t_data *t3_init_l2t(unsigned int l2t_capacity)
 {
 	struct l2t_data *d;
-	int i, size = sizeof(*d) + l2t_capacity * sizeof(struct l2t_entry);
+	int i;
 
-	d = kvzalloc(size, GFP_KERNEL);
+	d = kvzalloc(struct_size(d, l2tab, l2t_capacity), GFP_KERNEL);
 	if (!d)
 		return NULL;
 
diff --git a/drivers/net/ethernet/chelsio/cxgb3/l2t.h b/drivers/net/ethernet/chelsio/cxgb3/l2t.h
index c2fd323c4078..33558f177497 100644
--- a/drivers/net/ethernet/chelsio/cxgb3/l2t.h
+++ b/drivers/net/ethernet/chelsio/cxgb3/l2t.h
@@ -75,8 +75,8 @@ struct l2t_data {
 	struct l2t_entry *rover;	/* starting point for next allocation */
 	atomic_t nfree;		/* number of free entries */
 	rwlock_t lock;
-	struct l2t_entry l2tab[0];
 	struct rcu_head rcu_head;	/* to handle rcu cleanup */
+	struct l2t_entry l2tab[] __counted_by(nentries);
 };
 
 typedef void (*arp_failure_handler_func)(struct t3cdev * dev,
@@ -113,7 +113,6 @@ struct l2t_entry *t3_l2t_get(struct t3cdev *cdev, struct dst_entry *dst,
 			     struct net_device *dev, const void *daddr);
 int t3_l2t_send_slow(struct t3cdev *dev, struct sk_buff *skb,
 		     struct l2t_entry *e);
-void t3_l2t_send_event(struct t3cdev *dev, struct l2t_entry *e);
 struct l2t_data *t3_init_l2t(unsigned int l2t_capacity);
 
 int cxgb3_ofld_send(struct t3cdev *dev, struct sk_buff *skb);
diff --git a/drivers/net/ethernet/chelsio/cxgb3/regs.h b/drivers/net/ethernet/chelsio/cxgb3/regs.h
index 81029b872bdd..174eb45100a4 100644
--- a/drivers/net/ethernet/chelsio/cxgb3/regs.h
+++ b/drivers/net/ethernet/chelsio/cxgb3/regs.h
@@ -1,3 +1,4 @@
+/* SPDX-License-Identifier: GPL-2.0 */
 #define A_SG_CONTROL 0x0
 
 #define S_CONGMODE    29
diff --git a/drivers/net/ethernet/chelsio/cxgb3/sge.c b/drivers/net/ethernet/chelsio/cxgb3/sge.c
index 1b9d154f1149..b59735d0e065 100644
--- a/drivers/net/ethernet/chelsio/cxgb3/sge.c
+++ b/drivers/net/ethernet/chelsio/cxgb3/sge.c
@@ -126,8 +126,10 @@ struct rsp_desc {		/* response queue descriptor */
 	struct rss_header rss_hdr;
 	__be32 flags;
 	__be32 len_cq;
-	u8 imm_data[47];
-	u8 intr_gen;
+	struct_group(immediate,
+		u8 imm_data[47];
+		u8 intr_gen;
+	);
 };
 
 /*
@@ -164,11 +166,6 @@ static u8 flit_desc_map[] = {
 #endif
 };
 
-static inline struct sge_qset *fl_to_qset(const struct sge_fl *q, int qidx)
-{
-	return container_of(q, struct sge_qset, fl[qidx]);
-}
-
 static inline struct sge_qset *rspq_to_qset(const struct sge_rspq *q)
 {
 	return container_of(q, struct sge_qset, rspq);
@@ -244,8 +241,8 @@ static inline void unmap_skb(struct sk_buff *skb, struct sge_txq *q,
 	frag_idx = d->fragidx;
 
 	if (frag_idx == 0 && skb_headlen(skb)) {
-		pci_unmap_single(pdev, be64_to_cpu(sgp->addr[0]),
-				 skb_headlen(skb), PCI_DMA_TODEVICE);
+		dma_unmap_single(&pdev->dev, be64_to_cpu(sgp->addr[0]),
+				 skb_headlen(skb), DMA_TO_DEVICE);
 		j = 1;
 	}
 
@@ -253,9 +250,9 @@ static inline void unmap_skb(struct sk_buff *skb, struct sge_txq *q,
 	nfrags = skb_shinfo(skb)->nr_frags;
 
 	while (frag_idx < nfrags && curflit < WR_FLITS) {
-		pci_unmap_page(pdev, be64_to_cpu(sgp->addr[j]),
+		dma_unmap_page(&pdev->dev, be64_to_cpu(sgp->addr[j]),
 			       skb_frag_size(&skb_shinfo(skb)->frags[frag_idx]),
-			       PCI_DMA_TODEVICE);
+			       DMA_TO_DEVICE);
 		j ^= 1;
 		if (j == 0) {
 			sgp++;
@@ -355,15 +352,14 @@ static void clear_rx_desc(struct pci_dev *pdev, const struct sge_fl *q,
 	if (q->use_pages && d->pg_chunk.page) {
 		(*d->pg_chunk.p_cnt)--;
 		if (!*d->pg_chunk.p_cnt)
-			pci_unmap_page(pdev,
-				       d->pg_chunk.mapping,
-				       q->alloc_size, PCI_DMA_FROMDEVICE);
+			dma_unmap_page(&pdev->dev, d->pg_chunk.mapping,
+				       q->alloc_size, DMA_FROM_DEVICE);
 
 		put_page(d->pg_chunk.page);
 		d->pg_chunk.page = NULL;
 	} else {
-		pci_unmap_single(pdev, dma_unmap_addr(d, dma_addr),
-				 q->buf_size, PCI_DMA_FROMDEVICE);
+		dma_unmap_single(&pdev->dev, dma_unmap_addr(d, dma_addr),
+				 q->buf_size, DMA_FROM_DEVICE);
 		kfree_skb(d->skb);
 		d->skb = NULL;
 	}
@@ -372,7 +368,7 @@ static void clear_rx_desc(struct pci_dev *pdev, const struct sge_fl *q,
 /**
  *	free_rx_bufs - free the Rx buffers on an SGE free list
  *	@pdev: the PCI device associated with the adapter
- *	@rxq: the SGE free list to clean up
+ *	@q: the SGE free list to clean up
  *
  *	Release the buffers on an SGE free-buffer Rx queue.  HW fetching from
  *	this queue should be stopped before calling this function.
@@ -414,8 +410,8 @@ static inline int add_one_rx_buf(void *va, unsigned int len,
 {
 	dma_addr_t mapping;
 
-	mapping = pci_map_single(pdev, va, len, PCI_DMA_FROMDEVICE);
-	if (unlikely(pci_dma_mapping_error(pdev, mapping)))
+	mapping = dma_map_single(&pdev->dev, va, len, DMA_FROM_DEVICE);
+	if (unlikely(dma_mapping_error(&pdev->dev, mapping)))
 		return -ENOMEM;
 
 	dma_unmap_addr_set(sd, dma_addr, mapping);
@@ -453,8 +449,13 @@ static int alloc_pg_chunk(struct adapter *adapter, struct sge_fl *q,
 		q->pg_chunk.p_cnt = q->pg_chunk.va + (PAGE_SIZE << order) -
 				    SGE_PG_RSVD;
 		q->pg_chunk.offset = 0;
-		mapping = pci_map_page(adapter->pdev, q->pg_chunk.page,
-				       0, q->alloc_size, PCI_DMA_FROMDEVICE);
+		mapping = dma_map_page(&adapter->pdev->dev, q->pg_chunk.page,
+				       0, q->alloc_size, DMA_FROM_DEVICE);
+		if (unlikely(dma_mapping_error(&adapter->pdev->dev, mapping))) {
+			__free_pages(q->pg_chunk.page, order);
+			q->pg_chunk.page = NULL;
+			return -EIO;
+		}
 		q->pg_chunk.mapping = mapping;
 	}
 	sd->pg_chunk = q->pg_chunk;
@@ -488,7 +489,7 @@ static inline void ring_fl_db(struct adapter *adap, struct sge_fl *q)
 
 /**
  *	refill_fl - refill an SGE free-buffer list
- *	@adapter: the adapter
+ *	@adap: the adapter
  *	@q: the free-list to refill
  *	@n: the number of new buffers to allocate
  *	@gfp: the gfp flags for allocating new buffers
@@ -517,9 +518,9 @@ nomem:				q->alloc_failed++;
 			dma_unmap_addr_set(sd, dma_addr, mapping);
 
 			add_one_rx_chunk(mapping, d, q->gen);
-			pci_dma_sync_single_for_device(adap->pdev, mapping,
-						q->buf_size - SGE_PG_RSVD,
-						PCI_DMA_FROMDEVICE);
+			dma_sync_single_for_device(&adap->pdev->dev, mapping,
+						   q->buf_size - SGE_PG_RSVD,
+						   DMA_FROM_DEVICE);
 		} else {
 			void *buf_start;
 
@@ -563,7 +564,7 @@ static inline void __refill_fl(struct adapter *adap, struct sge_fl *fl)
 
 /**
  *	recycle_rx_buf - recycle a receive buffer
- *	@adapter: the adapter
+ *	@adap: the adapter
  *	@q: the SGE free list
  *	@idx: index of buffer to recycle
  *
@@ -628,7 +629,6 @@ static void *alloc_ring(struct pci_dev *pdev, size_t nelem, size_t elem_size,
 		}
 		*(void **)metadata = s;
 	}
-	memset(p, 0, len);
 	return p;
 }
 
@@ -661,7 +661,7 @@ static void t3_reset_qset(struct sge_qset *q)
 
 
 /**
- *	free_qset - free the resources of an SGE queue set
+ *	t3_free_qset - free the resources of an SGE queue set
  *	@adapter: the adapter owning the queue set
  *	@q: the queue set
  *
@@ -789,13 +789,13 @@ static struct sk_buff *get_packet(struct adapter *adap, struct sge_fl *fl,
 		skb = alloc_skb(len, GFP_ATOMIC);
 		if (likely(skb != NULL)) {
 			__skb_put(skb, len);
-			pci_dma_sync_single_for_cpu(adap->pdev,
-					    dma_unmap_addr(sd, dma_addr), len,
-					    PCI_DMA_FROMDEVICE);
+			dma_sync_single_for_cpu(&adap->pdev->dev,
+						dma_unmap_addr(sd, dma_addr),
+						len, DMA_FROM_DEVICE);
 			memcpy(skb->data, sd->skb->data, len);
-			pci_dma_sync_single_for_device(adap->pdev,
-					    dma_unmap_addr(sd, dma_addr), len,
-					    PCI_DMA_FROMDEVICE);
+			dma_sync_single_for_device(&adap->pdev->dev,
+						   dma_unmap_addr(sd, dma_addr),
+						   len, DMA_FROM_DEVICE);
 		} else if (!drop_thres)
 			goto use_orig_buf;
 recycle:
@@ -809,8 +809,8 @@ recycle:
 		goto recycle;
 
 use_orig_buf:
-	pci_unmap_single(adap->pdev, dma_unmap_addr(sd, dma_addr),
-			 fl->buf_size, PCI_DMA_FROMDEVICE);
+	dma_unmap_single(&adap->pdev->dev, dma_unmap_addr(sd, dma_addr),
+			 fl->buf_size, DMA_FROM_DEVICE);
 	skb = sd->skb;
 	skb_put(skb, len);
 	__refill_fl(adap, fl);
@@ -821,6 +821,7 @@ use_orig_buf:
  *	get_packet_pg - return the next ingress packet buffer from a free list
  *	@adap: the adapter that received the packet
  *	@fl: the SGE free list holding the packet
+ *	@q: the queue
  *	@len: the packet length including any SGE padding
  *	@drop_thres: # of remaining buffers before we start dropping packets
  *
@@ -849,12 +850,11 @@ static struct sk_buff *get_packet_pg(struct adapter *adap, struct sge_fl *fl,
 		newskb = alloc_skb(len, GFP_ATOMIC);
 		if (likely(newskb != NULL)) {
 			__skb_put(newskb, len);
-			pci_dma_sync_single_for_cpu(adap->pdev, dma_addr, len,
-					    PCI_DMA_FROMDEVICE);
+			dma_sync_single_for_cpu(&adap->pdev->dev, dma_addr,
+						len, DMA_FROM_DEVICE);
 			memcpy(newskb->data, sd->pg_chunk.va, len);
-			pci_dma_sync_single_for_device(adap->pdev, dma_addr,
-						       len,
-						       PCI_DMA_FROMDEVICE);
+			dma_sync_single_for_device(&adap->pdev->dev, dma_addr,
+						   len, DMA_FROM_DEVICE);
 		} else if (!drop_thres)
 			return NULL;
 recycle:
@@ -878,14 +878,12 @@ recycle:
 		goto recycle;
 	}
 
-	pci_dma_sync_single_for_cpu(adap->pdev, dma_addr, len,
-				    PCI_DMA_FROMDEVICE);
+	dma_sync_single_for_cpu(&adap->pdev->dev, dma_addr, len,
+				DMA_FROM_DEVICE);
 	(*sd->pg_chunk.p_cnt)--;
 	if (!*sd->pg_chunk.p_cnt && sd->pg_chunk.page != fl->pg_chunk.page)
-		pci_unmap_page(adap->pdev,
-			       sd->pg_chunk.mapping,
-			       fl->alloc_size,
-			       PCI_DMA_FROMDEVICE);
+		dma_unmap_page(&adap->pdev->dev, sd->pg_chunk.mapping,
+			       fl->alloc_size, DMA_FROM_DEVICE);
 	if (!skb) {
 		__skb_put(newskb, SGE_RX_PULL_LEN);
 		memcpy(newskb->data, sd->pg_chunk.va, SGE_RX_PULL_LEN);
@@ -924,7 +922,8 @@ static inline struct sk_buff *get_imm_packet(const struct rsp_desc *resp)
 
 	if (skb) {
 		__skb_put(skb, IMMED_PKT_SIZE);
-		skb_copy_to_linear_data(skb, resp->imm_data, IMMED_PKT_SIZE);
+		BUILD_BUG_ON(IMMED_PKT_SIZE != sizeof(resp->immediate));
+		skb_copy_to_linear_data(skb, &resp->immediate, IMMED_PKT_SIZE);
 	}
 	return skb;
 }
@@ -949,40 +948,79 @@ static inline unsigned int calc_tx_descs(const struct sk_buff *skb)
 	return flits_to_desc(flits);
 }
 
+/*	map_skb - map a packet main body and its page fragments
+ *	@pdev: the PCI device
+ *	@skb: the packet
+ *	@addr: placeholder to save the mapped addresses
+ *
+ *	map the main body of an sk_buff and its page fragments, if any.
+ */
+static int map_skb(struct pci_dev *pdev, const struct sk_buff *skb,
+		   dma_addr_t *addr)
+{
+	const skb_frag_t *fp, *end;
+	const struct skb_shared_info *si;
+
+	if (skb_headlen(skb)) {
+		*addr = dma_map_single(&pdev->dev, skb->data,
+				       skb_headlen(skb), DMA_TO_DEVICE);
+		if (dma_mapping_error(&pdev->dev, *addr))
+			goto out_err;
+		addr++;
+	}
+
+	si = skb_shinfo(skb);
+	end = &si->frags[si->nr_frags];
+
+	for (fp = si->frags; fp < end; fp++) {
+		*addr = skb_frag_dma_map(&pdev->dev, fp, 0, skb_frag_size(fp),
+					 DMA_TO_DEVICE);
+		if (dma_mapping_error(&pdev->dev, *addr))
+			goto unwind;
+		addr++;
+	}
+	return 0;
+
+unwind:
+	while (fp-- > si->frags)
+		dma_unmap_page(&pdev->dev, *--addr, skb_frag_size(fp),
+			       DMA_TO_DEVICE);
+
+	dma_unmap_single(&pdev->dev, addr[-1], skb_headlen(skb),
+			 DMA_TO_DEVICE);
+out_err:
+	return -ENOMEM;
+}
+
 /**
- *	make_sgl - populate a scatter/gather list for a packet
+ *	write_sgl - populate a scatter/gather list for a packet
  *	@skb: the packet
  *	@sgp: the SGL to populate
  *	@start: start address of skb main body data to include in the SGL
  *	@len: length of skb main body data to include in the SGL
- *	@pdev: the PCI device
+ *	@addr: the list of the mapped addresses
  *
- *	Generates a scatter/gather list for the buffers that make up a packet
+ *	Copies the scatter/gather list for the buffers that make up a packet
  *	and returns the SGL size in 8-byte words.  The caller must size the SGL
  *	appropriately.
  */
-static inline unsigned int make_sgl(const struct sk_buff *skb,
-				    struct sg_ent *sgp, unsigned char *start,
-				    unsigned int len, struct pci_dev *pdev)
+static inline unsigned int write_sgl(const struct sk_buff *skb,
+				     struct sg_ent *sgp, unsigned char *start,
+				     unsigned int len, const dma_addr_t *addr)
 {
-	dma_addr_t mapping;
-	unsigned int i, j = 0, nfrags;
+	unsigned int i, j = 0, k = 0, nfrags;
 
 	if (len) {
-		mapping = pci_map_single(pdev, start, len, PCI_DMA_TODEVICE);
 		sgp->len[0] = cpu_to_be32(len);
-		sgp->addr[0] = cpu_to_be64(mapping);
-		j = 1;
+		sgp->addr[j++] = cpu_to_be64(addr[k++]);
 	}
 
 	nfrags = skb_shinfo(skb)->nr_frags;
 	for (i = 0; i < nfrags; i++) {
 		const skb_frag_t *frag = &skb_shinfo(skb)->frags[i];
 
-		mapping = skb_frag_dma_map(&pdev->dev, frag, 0, skb_frag_size(frag),
-					   DMA_TO_DEVICE);
 		sgp->len[j] = cpu_to_be32(skb_frag_size(frag));
-		sgp->addr[j] = cpu_to_be64(mapping);
+		sgp->addr[j] = cpu_to_be64(addr[k++]);
 		j ^= 1;
 		if (j == 0)
 			++sgp;
@@ -1131,6 +1169,7 @@ static void write_wr_hdr_sgl(unsigned int ndesc, struct sk_buff *skb,
  *	@q: the Tx queue
  *	@ndesc: number of descriptors the packet will occupy
  *	@compl: the value of the COMPL bit to use
+ *	@addr: address
  *
  *	Generate a TX_PKT work request to send the supplied packet.
  */
@@ -1138,7 +1177,7 @@ static void write_tx_pkt_wr(struct adapter *adap, struct sk_buff *skb,
 			    const struct port_info *pi,
 			    unsigned int pidx, unsigned int gen,
 			    struct sge_txq *q, unsigned int ndesc,
-			    unsigned int compl)
+			    unsigned int compl, const dma_addr_t *addr)
 {
 	unsigned int flits, sgl_flits, cntrl, tso_info;
 	struct sg_ent *sgp, sgl[MAX_SKB_FRAGS / 2 + 1];
@@ -1196,7 +1235,7 @@ static void write_tx_pkt_wr(struct adapter *adap, struct sk_buff *skb,
 	}
 
 	sgp = ndesc == 1 ? (struct sg_ent *)&d->flit[flits] : sgl;
-	sgl_flits = make_sgl(skb, sgp, skb->data, skb_headlen(skb), adap->pdev);
+	sgl_flits = write_sgl(skb, sgp, skb->data, skb_headlen(skb), addr);
 
 	write_wr_hdr_sgl(ndesc, skb, d, pidx, q, sgl, flits, sgl_flits, gen,
 			 htonl(V_WR_OP(FW_WROPCODE_TUNNEL_TX_PKT) | compl),
@@ -1212,7 +1251,7 @@ static inline void t3_stop_tx_queue(struct netdev_queue *txq,
 }
 
 /**
- *	eth_xmit - add a packet to the Ethernet Tx queue
+ *	t3_eth_xmit - add a packet to the Ethernet Tx queue
  *	@skb: the packet
  *	@dev: the egress net device
  *
@@ -1227,6 +1266,7 @@ netdev_tx_t t3_eth_xmit(struct sk_buff *skb, struct net_device *dev)
 	struct netdev_queue *txq;
 	struct sge_qset *qs;
 	struct sge_txq *q;
+	dma_addr_t addr[MAX_SKB_FRAGS + 1];
 
 	/*
 	 * The chip min packet length is 9 octets but play safe and reject
@@ -1255,6 +1295,14 @@ netdev_tx_t t3_eth_xmit(struct sk_buff *skb, struct net_device *dev)
 		return NETDEV_TX_BUSY;
 	}
 
+	/* Check if ethernet packet can't be sent as immediate data */
+	if (skb->len > (WR_LEN - sizeof(struct cpl_tx_pkt))) {
+		if (unlikely(map_skb(adap->pdev, skb, addr) < 0)) {
+			dev_kfree_skb(skb);
+			return NETDEV_TX_OK;
+		}
+	}
+
 	q->in_use += ndesc;
 	if (unlikely(credits - ndesc < q->stop_thres)) {
 		t3_stop_tx_queue(txq, qs, q);
@@ -1312,7 +1360,7 @@ netdev_tx_t t3_eth_xmit(struct sk_buff *skb, struct net_device *dev)
 	if (likely(!skb_shared(skb)))
 		skb_orphan(skb);
 
-	write_tx_pkt_wr(adap, skb, pi, pidx, gen, q, ndesc, compl);
+	write_tx_pkt_wr(adap, skb, pi, pidx, gen, q, ndesc, compl, addr);
 	check_ring_tx_db(adap, q);
 	return NETDEV_TX_OK;
 }
@@ -1465,14 +1513,15 @@ static int ctrl_xmit(struct adapter *adap, struct sge_txq *q,
 
 /**
  *	restart_ctrlq - restart a suspended control queue
- *	@qs: the queue set cotaining the control queue
+ *	@w: pointer to the work associated with this handler
  *
  *	Resumes transmission on a suspended Tx control queue.
  */
-static void restart_ctrlq(unsigned long data)
+static void restart_ctrlq(struct work_struct *w)
 {
 	struct sk_buff *skb;
-	struct sge_qset *qs = (struct sge_qset *)data;
+	struct sge_qset *qs = container_of(w, struct sge_qset,
+					   txq[TXQ_CTRL].qresume_task);
 	struct sge_txq *q = &qs->txq[TXQ_CTRL];
 
 	spin_lock(&q->lock);
@@ -1538,13 +1587,14 @@ static void deferred_unmap_destructor(struct sk_buff *skb)
 	p = dui->addr;
 
 	if (skb_tail_pointer(skb) - skb_transport_header(skb))
-		pci_unmap_single(dui->pdev, *p++, skb_tail_pointer(skb) -
-				 skb_transport_header(skb), PCI_DMA_TODEVICE);
+		dma_unmap_single(&dui->pdev->dev, *p++,
+				 skb_tail_pointer(skb) - skb_transport_header(skb),
+				 DMA_TO_DEVICE);
 
 	si = skb_shinfo(skb);
 	for (i = 0; i < si->nr_frags; i++)
-		pci_unmap_page(dui->pdev, *p++, skb_frag_size(&si->frags[i]),
-			       PCI_DMA_TODEVICE);
+		dma_unmap_page(&dui->pdev->dev, *p++,
+			       skb_frag_size(&si->frags[i]), DMA_TO_DEVICE);
 }
 
 static void setup_deferred_unmapping(struct sk_buff *skb, struct pci_dev *pdev,
@@ -1571,13 +1621,15 @@ static void setup_deferred_unmapping(struct sk_buff *skb, struct pci_dev *pdev,
  *	@pidx: index of the first Tx descriptor to write
  *	@gen: the generation value to use
  *	@ndesc: number of descriptors the packet will occupy
+ *	@addr: the address
  *
  *	Write an offload work request to send the supplied packet.  The packet
  *	data already carry the work request with most fields populated.
  */
 static void write_ofld_wr(struct adapter *adap, struct sk_buff *skb,
 			  struct sge_txq *q, unsigned int pidx,
-			  unsigned int gen, unsigned int ndesc)
+			  unsigned int gen, unsigned int ndesc,
+			  const dma_addr_t *addr)
 {
 	unsigned int sgl_flits, flits;
 	struct work_request_hdr *from;
@@ -1598,10 +1650,9 @@ static void write_ofld_wr(struct adapter *adap, struct sk_buff *skb,
 
 	flits = skb_transport_offset(skb) / 8;
 	sgp = ndesc == 1 ? (struct sg_ent *)&d->flit[flits] : sgl;
-	sgl_flits = make_sgl(skb, sgp, skb_transport_header(skb),
-			     skb_tail_pointer(skb) -
-			     skb_transport_header(skb),
-			     adap->pdev);
+	sgl_flits = write_sgl(skb, sgp, skb_transport_header(skb),
+			      skb_tail_pointer(skb) - skb_transport_header(skb),
+			      addr);
 	if (need_skb_unmap()) {
 		setup_deferred_unmapping(skb, adap->pdev, sgp, sgl_flits);
 		skb->destructor = deferred_unmap_destructor;
@@ -1659,6 +1710,12 @@ again:	reclaim_completed_tx(adap, q, TX_RECLAIM_CHUNK);
 		goto again;
 	}
 
+	if (!immediate(skb) &&
+	    map_skb(adap->pdev, skb, (dma_addr_t *)skb->head)) {
+		spin_unlock(&q->lock);
+		return NET_XMIT_SUCCESS;
+	}
+
 	gen = q->gen;
 	q->in_use += ndesc;
 	pidx = q->pidx;
@@ -1669,24 +1726,26 @@ again:	reclaim_completed_tx(adap, q, TX_RECLAIM_CHUNK);
 	}
 	spin_unlock(&q->lock);
 
-	write_ofld_wr(adap, skb, q, pidx, gen, ndesc);
+	write_ofld_wr(adap, skb, q, pidx, gen, ndesc, (dma_addr_t *)skb->head);
 	check_ring_tx_db(adap, q);
 	return NET_XMIT_SUCCESS;
 }
 
 /**
  *	restart_offloadq - restart a suspended offload queue
- *	@qs: the queue set cotaining the offload queue
+ *	@w: pointer to the work associated with this handler
  *
  *	Resumes transmission on a suspended Tx offload queue.
  */
-static void restart_offloadq(unsigned long data)
+static void restart_offloadq(struct work_struct *w)
 {
 	struct sk_buff *skb;
-	struct sge_qset *qs = (struct sge_qset *)data;
+	struct sge_qset *qs = container_of(w, struct sge_qset,
+					   txq[TXQ_OFLD].qresume_task);
 	struct sge_txq *q = &qs->txq[TXQ_OFLD];
 	const struct port_info *pi = netdev_priv(qs->netdev);
 	struct adapter *adap = pi->adapter;
+	unsigned int written = 0;
 
 	spin_lock(&q->lock);
 again:	reclaim_completed_tx(adap, q, TX_RECLAIM_CHUNK);
@@ -1706,10 +1765,15 @@ again:	reclaim_completed_tx(adap, q, TX_RECLAIM_CHUNK);
 			break;
 		}
 
+		if (!immediate(skb) &&
+		    map_skb(adap->pdev, skb, (dma_addr_t *)skb->head))
+			break;
+
 		gen = q->gen;
 		q->in_use += ndesc;
 		pidx = q->pidx;
 		q->pidx += ndesc;
+		written += ndesc;
 		if (q->pidx >= q->size) {
 			q->pidx -= q->size;
 			q->gen ^= 1;
@@ -1717,7 +1781,8 @@ again:	reclaim_completed_tx(adap, q, TX_RECLAIM_CHUNK);
 		__skb_unlink(skb, &q->sendq);
 		spin_unlock(&q->lock);
 
-		write_ofld_wr(adap, skb, q, pidx, gen, ndesc);
+		write_ofld_wr(adap, skb, q, pidx, gen, ndesc,
+			      (dma_addr_t *)skb->head);
 		spin_lock(&q->lock);
 	}
 	spin_unlock(&q->lock);
@@ -1727,8 +1792,9 @@ again:	reclaim_completed_tx(adap, q, TX_RECLAIM_CHUNK);
 	set_bit(TXQ_LAST_PKT_DB, &q->flags);
 #endif
 	wmb();
-	t3_write_reg(adap, A_SG_KDOORBELL,
-		     F_SELEGRCNTX | V_EGRCNTX(q->cntxt_id));
+	if (likely(written))
+		t3_write_reg(adap, A_SG_KDOORBELL,
+			     F_SELEGRCNTX | V_EGRCNTX(q->cntxt_id));
 }
 
 /**
@@ -1818,7 +1884,7 @@ static inline void deliver_partial_bundle(struct t3cdev *tdev,
 
 /**
  *	ofld_poll - NAPI handler for offload packets in interrupt mode
- *	@dev: the network device doing the polling
+ *	@napi: the network device doing the polling
  *	@budget: polling budget
  *
  *	The NAPI handler for offload packets when a response queue is serviced
@@ -1885,7 +1951,7 @@ static int ofld_poll(struct napi_struct *napi, int budget)
  *	@rx_gather: a gather list of packets if we are building a bundle
  *	@gather_idx: index of the next available slot in the bundle
  *
- *	Process an ingress offload pakcet and add it to the offload ingress
+ *	Process an ingress offload packet and add it to the offload ingress
  *	queue. 	Returns the index of the next available slot in the bundle.
  */
 static inline int rx_offload(struct t3cdev *tdev, struct sge_rspq *rq,
@@ -1930,19 +1996,23 @@ static void restart_tx(struct sge_qset *qs)
 	    should_restart_tx(&qs->txq[TXQ_OFLD]) &&
 	    test_and_clear_bit(TXQ_OFLD, &qs->txq_stopped)) {
 		qs->txq[TXQ_OFLD].restarts++;
-		tasklet_schedule(&qs->txq[TXQ_OFLD].qresume_tsk);
+
+		/* The work can be quite lengthy so we use driver's own queue */
+		queue_work(cxgb3_wq, &qs->txq[TXQ_OFLD].qresume_task);
 	}
 	if (test_bit(TXQ_CTRL, &qs->txq_stopped) &&
 	    should_restart_tx(&qs->txq[TXQ_CTRL]) &&
 	    test_and_clear_bit(TXQ_CTRL, &qs->txq_stopped)) {
 		qs->txq[TXQ_CTRL].restarts++;
-		tasklet_schedule(&qs->txq[TXQ_CTRL].qresume_tsk);
+
+		/* The work can be quite lengthy so we use driver's own queue */
+		queue_work(cxgb3_wq, &qs->txq[TXQ_CTRL].qresume_task);
 	}
 }
 
 /**
  *	cxgb3_arp_process - process an ARP request probing a private IP address
- *	@adapter: the adapter
+ *	@pi: the port info
  *	@skb: the skbuff containing the ARP request
  *
  *	Check if the ARP request is probing the private IP address
@@ -2004,9 +2074,10 @@ static void cxgb3_process_iscsi_prov_pack(struct port_info *pi,
  *	@adap: the adapter
  *	@rq: the response queue that received the packet
  *	@skb: the packet
- *	@pad: amount of padding at the start of the buffer
+ *	@pad: padding
+ *	@lro: large receive offload
  *
- *	Process an ingress ethernet pakcet and deliver it to the stack.
+ *	Process an ingress ethernet packet and deliver it to the stack.
  *	The padding is 2 if the packet was delivered in an Rx buffer and 0
  *	if it was immediate data in a response.
  */
@@ -2067,7 +2138,7 @@ static void lro_add_page(struct adapter *adap, struct sge_qset *qs,
 	struct port_info *pi = netdev_priv(qs->netdev);
 	struct sk_buff *skb = NULL;
 	struct cpl_rx_pkt *cpl;
-	struct skb_frag_struct *rx_frag;
+	skb_frag_t *rx_frag;
 	int nr_frags;
 	int offset = 0;
 
@@ -2078,17 +2149,14 @@ static void lro_add_page(struct adapter *adap, struct sge_qset *qs,
 
 	fl->credits--;
 
-	pci_dma_sync_single_for_cpu(adap->pdev,
-				    dma_unmap_addr(sd, dma_addr),
-				    fl->buf_size - SGE_PG_RSVD,
-				    PCI_DMA_FROMDEVICE);
+	dma_sync_single_for_cpu(&adap->pdev->dev,
+				dma_unmap_addr(sd, dma_addr),
+				fl->buf_size - SGE_PG_RSVD, DMA_FROM_DEVICE);
 
 	(*sd->pg_chunk.p_cnt)--;
 	if (!*sd->pg_chunk.p_cnt && sd->pg_chunk.page != fl->pg_chunk.page)
-		pci_unmap_page(adap->pdev,
-			       sd->pg_chunk.mapping,
-			       fl->alloc_size,
-			       PCI_DMA_FROMDEVICE);
+		dma_unmap_page(&adap->pdev->dev, sd->pg_chunk.mapping,
+			       fl->alloc_size, DMA_FROM_DEVICE);
 
 	if (!skb) {
 		put_page(sd->pg_chunk.page);
@@ -2116,9 +2184,8 @@ static void lro_add_page(struct adapter *adap, struct sge_qset *qs,
 	len -= offset;
 
 	rx_frag += nr_frags;
-	__skb_frag_set_page(rx_frag, sd->pg_chunk.page);
-	rx_frag->page_offset = sd->pg_chunk.offset + offset;
-	skb_frag_size_set(rx_frag, len);
+	skb_frag_fill_page_desc(rx_frag, sd->pg_chunk.page,
+				sd->pg_chunk.offset + offset, len);
 
 	skb->len += len;
 	skb->data_len += len;
@@ -2174,7 +2241,7 @@ static inline void handle_rsp_cntrl_info(struct sge_qset *qs, u32 flags)
 
 /**
  *	check_ring_db - check if we need to ring any doorbells
- *	@adapter: the adapter
+ *	@adap: the adapter
  *	@qs: the queue set whose Tx queues are to be examined
  *	@sleeping: indicates which Tx queue sent GTS
  *
@@ -2282,7 +2349,7 @@ static int process_responses(struct adapter *adap, struct sge_qset *qs,
 			if (!skb)
 				goto no_mem;
 
-			memcpy(__skb_put(skb, AN_PKT_SIZE), r, AN_PKT_SIZE);
+			__skb_put_data(skb, r, AN_PKT_SIZE);
 			skb->data[0] = CPL_ASYNC_NOTIF;
 			rss_hi = htonl(CPL_ASYNC_NOTIF << 24);
 			q->async_notif++;
@@ -2307,16 +2374,13 @@ no_mem:
 			if (fl->use_pages) {
 				void *addr = fl->sdesc[fl->cidx].pg_chunk.va;
 
-				prefetch(addr);
-#if L1_CACHE_BYTES < 128
-				prefetch(addr + L1_CACHE_BYTES);
-#endif
+				net_prefetch(addr);
 				__refill_fl(adap, fl);
 				if (lro > 0) {
 					lro_add_page(adap, qs, fl,
 						     G_RSPD_LEN(len),
 						     flags & F_RSPD_EOP);
-					 goto next_fl;
+					goto next_fl;
 				}
 
 				skb = get_packet_pg(adap, fl, q,
@@ -2437,14 +2501,6 @@ static int napi_rx_handler(struct napi_struct *napi, int budget)
 	return work_done;
 }
 
-/*
- * Returns true if the device is already scheduled for polling.
- */
-static inline int napi_is_scheduled(struct napi_struct *napi)
-{
-	return test_bit(NAPI_STATE_SCHED, &napi->state);
-}
-
 /**
  *	process_pure_responses - process pure responses from a response queue
  *	@adap: the adapter
@@ -2610,12 +2666,7 @@ static int rspq_check_napi(struct sge_qset *qs)
 {
 	struct sge_rspq *q = &qs->rspq;
 
-	if (!napi_is_scheduled(&qs->napi) &&
-	    is_new_response(&q->desc[q->cidx], q)) {
-		napi_schedule(&qs->napi);
-		return 1;
-	}
-	return 0;
+	return is_new_response(&q->desc[q->cidx], q) && napi_schedule(&qs->napi);
 }
 
 /*
@@ -2837,7 +2888,7 @@ void t3_sge_err_intr_handler(struct adapter *adapter)
 
 /**
  *	sge_timer_tx - perform periodic maintenance of an SGE qset
- *	@data: the SGE queue set to maintain
+ *	@t: a timer list containing the SGE queue set to maintain
  *
  *	Runs periodically from a timer to perform maintenance of an SGE queue
  *	set.  It performs two tasks:
@@ -2853,9 +2904,9 @@ void t3_sge_err_intr_handler(struct adapter *adapter)
  *	bother cleaning them up here.
  *
  */
-static void sge_timer_tx(unsigned long data)
+static void sge_timer_tx(struct timer_list *t)
 {
-	struct sge_qset *qs = (struct sge_qset *)data;
+	struct sge_qset *qs = timer_container_of(qs, t, tx_reclaim_timer);
 	struct port_info *pi = netdev_priv(qs->netdev);
 	struct adapter *adap = pi->adapter;
 	unsigned int tbd[SGE_TXQ_PER_SET] = {0, 0};
@@ -2881,7 +2932,7 @@ static void sge_timer_tx(unsigned long data)
 
 /**
  *	sge_timer_rx - perform periodic maintenance of an SGE qset
- *	@data: the SGE queue set to maintain
+ *	@t: the timer list containing the SGE queue set to maintain
  *
  *	a) Replenishes Rx queues that have run out due to memory shortage.
  *	Normally new Rx buffers are added when existing ones are consumed but
@@ -2893,10 +2944,10 @@ static void sge_timer_tx(unsigned long data)
  *	starved.
  *
  */
-static void sge_timer_rx(unsigned long data)
+static void sge_timer_rx(struct timer_list *t)
 {
 	spinlock_t *lock;
-	struct sge_qset *qs = (struct sge_qset *)data;
+	struct sge_qset *qs = timer_container_of(qs, t, rx_reclaim_timer);
 	struct port_info *pi = netdev_priv(qs->netdev);
 	struct adapter *adap = pi->adapter;
 	u32 status;
@@ -2959,7 +3010,7 @@ void t3_update_qset_coalesce(struct sge_qset *qs, const struct qset_params *p)
  *	@irq_vec_idx: the IRQ vector index for response queue interrupts
  *	@p: configuration parameters for this queue set
  *	@ntxq: number of Tx queues for the queue set
- *	@netdev: net device associated with this queue set
+ *	@dev: net device associated with this queue set
  *	@netdevq: net device TX queue associated with this queue set
  *
  *	Allocate resources and initialize an SGE queue set.  A queue set
@@ -2976,8 +3027,8 @@ int t3_sge_alloc_qset(struct adapter *adapter, unsigned int id, int nports,
 	struct sge_qset *q = &adapter->sge.qs[id];
 
 	init_qset_cntxt(q, id);
-	setup_timer(&q->tx_reclaim_timer, sge_timer_tx, (unsigned long)q);
-	setup_timer(&q->rx_reclaim_timer, sge_timer_rx, (unsigned long)q);
+	timer_setup(&q->tx_reclaim_timer, sge_timer_tx, 0);
+	timer_setup(&q->rx_reclaim_timer, sge_timer_rx, 0);
 
 	q->fl[0].desc = alloc_ring(adapter->pdev, p->fl_size,
 				   sizeof(struct rx_desc),
@@ -3019,10 +3070,8 @@ int t3_sge_alloc_qset(struct adapter *adapter, unsigned int id, int nports,
 		skb_queue_head_init(&q->txq[i].sendq);
 	}
 
-	tasklet_init(&q->txq[TXQ_OFLD].qresume_tsk, restart_offloadq,
-		     (unsigned long)q);
-	tasklet_init(&q->txq[TXQ_CTRL].qresume_tsk, restart_ctrlq,
-		     (unsigned long)q);
+	INIT_WORK(&q->txq[TXQ_OFLD].qresume_task, restart_offloadq);
+	INIT_WORK(&q->txq[TXQ_CTRL].qresume_task, restart_ctrlq);
 
 	q->fl[0].gen = q->fl[1].gen = 1;
 	q->fl[0].size = p->fl_size;
@@ -3111,6 +3160,7 @@ int t3_sge_alloc_qset(struct adapter *adapter, unsigned int id, int nports,
 			  GFP_KERNEL | __GFP_COMP);
 	if (!avail) {
 		CH_ALERT(adapter, "free list queue 0 initialization failed\n");
+		ret = -ENOMEM;
 		goto err;
 	}
 	if (avail < q->fl[0].size)
@@ -3149,11 +3199,13 @@ void t3_start_sge_timers(struct adapter *adap)
 	for (i = 0; i < SGE_QSETS; ++i) {
 		struct sge_qset *q = &adap->sge.qs[i];
 
-	if (q->tx_reclaim_timer.function)
-		mod_timer(&q->tx_reclaim_timer, jiffies + TX_RECLAIM_PERIOD);
+		if (q->tx_reclaim_timer.function)
+			mod_timer(&q->tx_reclaim_timer,
+				  jiffies + TX_RECLAIM_PERIOD);
 
-	if (q->rx_reclaim_timer.function)
-		mod_timer(&q->rx_reclaim_timer, jiffies + RX_RECLAIM_PERIOD);
+		if (q->rx_reclaim_timer.function)
+			mod_timer(&q->rx_reclaim_timer,
+				  jiffies + RX_RECLAIM_PERIOD);
 	}
 }
 
@@ -3171,9 +3223,9 @@ void t3_stop_sge_timers(struct adapter *adap)
 		struct sge_qset *q = &adap->sge.qs[i];
 
 		if (q->tx_reclaim_timer.function)
-			del_timer_sync(&q->tx_reclaim_timer);
+			timer_delete_sync(&q->tx_reclaim_timer);
 		if (q->rx_reclaim_timer.function)
-			del_timer_sync(&q->rx_reclaim_timer);
+			timer_delete_sync(&q->rx_reclaim_timer);
 	}
 }
 
@@ -3204,30 +3256,43 @@ void t3_sge_start(struct adapter *adap)
 }
 
 /**
- *	t3_sge_stop - disable SGE operation
+ *	t3_sge_stop_dma - Disable SGE DMA engine operation
  *	@adap: the adapter
  *
- *	Disables the DMA engine.  This can be called in emeregencies (e.g.,
- *	from error interrupts) or from normal process context.  In the latter
- *	case it also disables any pending queue restart tasklets.  Note that
- *	if it is called in interrupt context it cannot disable the restart
- *	tasklets as it cannot wait, however the tasklets will have no effect
- *	since the doorbells are disabled and the driver will call this again
- *	later from process context, at which time the tasklets will be stopped
- *	if they are still running.
+ *	Can be invoked from interrupt context e.g.  error handler.
+ *
+ *	Note that this function cannot disable the restart of works as
+ *	it cannot wait if called from interrupt context, however the
+ *	works will have no effect since the doorbells are disabled. The
+ *	driver will call tg3_sge_stop() later from process context, at
+ *	which time the works will be stopped if they are still running.
  */
-void t3_sge_stop(struct adapter *adap)
+void t3_sge_stop_dma(struct adapter *adap)
 {
 	t3_set_reg_field(adap, A_SG_CONTROL, F_GLOBALENABLE, 0);
-	if (!in_interrupt()) {
-		int i;
+}
+
+/**
+ *	t3_sge_stop - disable SGE operation completly
+ *	@adap: the adapter
+ *
+ *	Called from process context. Disables the DMA engine and any
+ *	pending queue restart works.
+ */
+void t3_sge_stop(struct adapter *adap)
+{
+	int i;
 
-		for (i = 0; i < SGE_QSETS; ++i) {
-			struct sge_qset *qs = &adap->sge.qs[i];
+	t3_sge_stop_dma(adap);
 
-			tasklet_kill(&qs->txq[TXQ_OFLD].qresume_tsk);
-			tasklet_kill(&qs->txq[TXQ_CTRL].qresume_tsk);
-		}
+	/* workqueues aren't initialized otherwise */
+	if (!(adap->flags & FULL_INIT_DONE))
+		return;
+	for (i = 0; i < SGE_QSETS; ++i) {
+		struct sge_qset *qs = &adap->sge.qs[i];
+
+		cancel_work_sync(&qs->txq[TXQ_OFLD].qresume_task);
+		cancel_work_sync(&qs->txq[TXQ_CTRL].qresume_task);
 	}
 }
 
@@ -3294,7 +3359,7 @@ void t3_sge_prep(struct adapter *adap, struct sge_params *p)
 		q->coalesce_usecs = 5;
 		q->rspq_size = 1024;
 		q->fl_size = 1024;
- 		q->jumbo_size = 512;
+		q->jumbo_size = 512;
 		q->txq_size[TXQ_ETH] = 1024;
 		q->txq_size[TXQ_OFLD] = 1024;
 		q->txq_size[TXQ_CTRL] = 256;
diff --git a/drivers/net/ethernet/chelsio/cxgb3/sge_defs.h b/drivers/net/ethernet/chelsio/cxgb3/sge_defs.h
index 29b6c800b238..c31ce8dc95fb 100644
--- a/drivers/net/ethernet/chelsio/cxgb3/sge_defs.h
+++ b/drivers/net/ethernet/chelsio/cxgb3/sge_defs.h
@@ -1,3 +1,4 @@
+/* SPDX-License-Identifier: GPL-2.0 */
 /*
  * This file is automatically generated --- any changes will be lost.
  */
diff --git a/drivers/net/ethernet/chelsio/cxgb3/t3_cpl.h b/drivers/net/ethernet/chelsio/cxgb3/t3_cpl.h
index 852c399a8b0a..68bb5f39f3f1 100644
--- a/drivers/net/ethernet/chelsio/cxgb3/t3_cpl.h
+++ b/drivers/net/ethernet/chelsio/cxgb3/t3_cpl.h
@@ -1448,7 +1448,7 @@ struct cpl_rdma_terminate {
 #endif
 	__be32 msn;
 	__be32 mo;
-	__u8 data[0];
+	__u8 data[];
 };
 
 /* cpl_rdma_terminate.tid_len fields */
diff --git a/drivers/net/ethernet/chelsio/cxgb3/t3_hw.c b/drivers/net/ethernet/chelsio/cxgb3/t3_hw.c
index a89721fad633..a06003bfa04b 100644
--- a/drivers/net/ethernet/chelsio/cxgb3/t3_hw.c
+++ b/drivers/net/ethernet/chelsio/cxgb3/t3_hw.c
@@ -29,6 +29,7 @@
  * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
  * SOFTWARE.
  */
+#include <linux/etherdevice.h>
 #include "common.h"
 #include "regs.h"
 #include "sge_defs.h"
@@ -595,81 +596,10 @@ struct t3_vpd {
 	u32 pad;		/* for multiple-of-4 sizing and alignment */
 };
 
-#define EEPROM_MAX_POLL   40
 #define EEPROM_STAT_ADDR  0x4000
 #define VPD_BASE          0xc00
 
 /**
- *	t3_seeprom_read - read a VPD EEPROM location
- *	@adapter: adapter to read
- *	@addr: EEPROM address
- *	@data: where to store the read data
- *
- *	Read a 32-bit word from a location in VPD EEPROM using the card's PCI
- *	VPD ROM capability.  A zero is written to the flag bit when the
- *	address is written to the control register.  The hardware device will
- *	set the flag to 1 when 4 bytes have been read into the data register.
- */
-int t3_seeprom_read(struct adapter *adapter, u32 addr, __le32 *data)
-{
-	u16 val;
-	int attempts = EEPROM_MAX_POLL;
-	u32 v;
-	unsigned int base = adapter->params.pci.vpd_cap_addr;
-
-	if ((addr >= EEPROMSIZE && addr != EEPROM_STAT_ADDR) || (addr & 3))
-		return -EINVAL;
-
-	pci_write_config_word(adapter->pdev, base + PCI_VPD_ADDR, addr);
-	do {
-		udelay(10);
-		pci_read_config_word(adapter->pdev, base + PCI_VPD_ADDR, &val);
-	} while (!(val & PCI_VPD_ADDR_F) && --attempts);
-
-	if (!(val & PCI_VPD_ADDR_F)) {
-		CH_ERR(adapter, "reading EEPROM address 0x%x failed\n", addr);
-		return -EIO;
-	}
-	pci_read_config_dword(adapter->pdev, base + PCI_VPD_DATA, &v);
-	*data = cpu_to_le32(v);
-	return 0;
-}
-
-/**
- *	t3_seeprom_write - write a VPD EEPROM location
- *	@adapter: adapter to write
- *	@addr: EEPROM address
- *	@data: value to write
- *
- *	Write a 32-bit word to a location in VPD EEPROM using the card's PCI
- *	VPD ROM capability.
- */
-int t3_seeprom_write(struct adapter *adapter, u32 addr, __le32 data)
-{
-	u16 val;
-	int attempts = EEPROM_MAX_POLL;
-	unsigned int base = adapter->params.pci.vpd_cap_addr;
-
-	if ((addr >= EEPROMSIZE && addr != EEPROM_STAT_ADDR) || (addr & 3))
-		return -EINVAL;
-
-	pci_write_config_dword(adapter->pdev, base + PCI_VPD_DATA,
-			       le32_to_cpu(data));
-	pci_write_config_word(adapter->pdev,base + PCI_VPD_ADDR,
-			      addr | PCI_VPD_ADDR_F);
-	do {
-		msleep(1);
-		pci_read_config_word(adapter->pdev, base + PCI_VPD_ADDR, &val);
-	} while ((val & PCI_VPD_ADDR_F) && --attempts);
-
-	if (val & PCI_VPD_ADDR_F) {
-		CH_ERR(adapter, "write to EEPROM address 0x%x failed\n", addr);
-		return -EIO;
-	}
-	return 0;
-}
-
-/**
  *	t3_seeprom_wp - enable/disable EEPROM write protection
  *	@adapter: the adapter
  *	@enable: 1 to enable write protection, 0 to disable it
@@ -678,21 +608,28 @@ int t3_seeprom_write(struct adapter *adapter, u32 addr, __le32 data)
  */
 int t3_seeprom_wp(struct adapter *adapter, int enable)
 {
-	return t3_seeprom_write(adapter, EEPROM_STAT_ADDR, enable ? 0xc : 0);
+	u32 data = enable ? 0xc : 0;
+	int ret;
+
+	/* EEPROM_STAT_ADDR is outside VPD area, use pci_write_vpd_any() */
+	ret = pci_write_vpd_any(adapter->pdev, EEPROM_STAT_ADDR, sizeof(u32),
+				&data);
+
+	return ret < 0 ? ret : 0;
 }
 
-static int vpdstrtouint(char *s, int len, unsigned int base, unsigned int *val)
+static int vpdstrtouint(char *s, u8 len, unsigned int base, unsigned int *val)
 {
-	char tok[len + 1];
+	char tok[256];
 
 	memcpy(tok, s, len);
 	tok[len] = 0;
 	return kstrtouint(strim(tok), base, val);
 }
 
-static int vpdstrtou16(char *s, int len, unsigned int base, u16 *val)
+static int vpdstrtou16(char *s, u8 len, unsigned int base, u16 *val)
 {
-	char tok[len + 1];
+	char tok[256];
 
 	memcpy(tok, s, len);
 	tok[len] = 0;
@@ -708,24 +645,22 @@ static int vpdstrtou16(char *s, int len, unsigned int base, u16 *val)
  */
 static int get_vpd_params(struct adapter *adapter, struct vpd_params *p)
 {
-	int i, addr, ret;
 	struct t3_vpd vpd;
+	u8 base_val = 0;
+	int addr, ret;
 
 	/*
 	 * Card information is normally at VPD_BASE but some early cards had
 	 * it at 0.
 	 */
-	ret = t3_seeprom_read(adapter, VPD_BASE, (__le32 *)&vpd);
-	if (ret)
+	ret = pci_read_vpd(adapter->pdev, VPD_BASE, 1, &base_val);
+	if (ret < 0)
 		return ret;
-	addr = vpd.id_tag == 0x82 ? VPD_BASE : 0;
+	addr = base_val == PCI_VPD_LRDT_ID_STRING ? VPD_BASE : 0;
 
-	for (i = 0; i < sizeof(vpd); i += 4) {
-		ret = t3_seeprom_read(adapter, addr + i,
-				      (__le32 *)((u8 *)&vpd + i));
-		if (ret)
-			return ret;
-	}
+	ret = pci_read_vpd(adapter->pdev, addr, sizeof(vpd), &vpd);
+	if (ret < 0)
+		return ret;
 
 	ret = vpdstrtouint(vpd.cclk_data, vpd.cclk_len, 10, &p->cclk);
 	if (ret)
@@ -1082,7 +1017,7 @@ int t3_check_fw_version(struct adapter *adapter)
 		CH_WARN(adapter, "found newer FW version(%u.%u), "
 		        "driver compiled for version %u.%u\n", major, minor,
 			FW_VERSION_MAJOR, FW_VERSION_MINOR);
-			return 0;
+		return 0;
 	}
 	return -EINVAL;
 }
@@ -2195,7 +2130,7 @@ static int t3_sge_write_context(struct adapter *adapter, unsigned int id,
 
 /**
  *	clear_sge_ctxt - completely clear an SGE context
- *	@adapter: the adapter
+ *	@adap: the adapter
  *	@id: the context id
  *	@type: the context type
  *
@@ -2484,6 +2419,7 @@ int t3_sge_disable_cqcntxt(struct adapter *adapter, unsigned int id)
  *	@adapter: the adapter
  *	@id: the context id
  *	@op: the operation to perform
+ *	@credits: credit value to write
  *
  *	Perform the selected operation on an SGE completion queue context.
  *	The caller is responsible for ensuring only one context operation
@@ -2885,7 +2821,7 @@ static void init_cong_ctrl(unsigned short *a, unsigned short *b)
  *	t3_load_mtus - write the MTU and congestion control HW tables
  *	@adap: the adapter
  *	@mtus: the unrestricted values for the MTU table
- *	@alphs: the values for the congestion control alpha parameter
+ *	@alpha: the values for the congestion control alpha parameter
  *	@beta: the values for the congestion control beta parameter
  *	@mtu_cap: the maximum permitted effective MTU
  *
@@ -2966,7 +2902,7 @@ static void ulp_config(struct adapter *adap, const struct tp_params *p)
 
 /**
  *	t3_set_proto_sram - set the contents of the protocol sram
- *	@adapter: the adapter
+ *	@adap: the adapter
  *	@data: the protocol image
  *
  *	Write the contents of the protocol SRAM.
@@ -3483,7 +3419,7 @@ static void get_pci_mode(struct adapter *adapter, struct pci_params *p)
 /**
  *	init_link_config - initialize a link's SW state
  *	@lc: structure holding the link state
- *	@ai: information about the current card
+ *	@caps: information about the current card
  *
  *	Initializes the SW state maintained for each link, including the link's
  *	capabilities and default speed/duplex/flow-control/autonegotiation
@@ -3619,7 +3555,7 @@ int t3_reset_adapter(struct adapter *adapter)
 
 static int init_parity(struct adapter *adap)
 {
-		int i, err, addr;
+	int i, err, addr;
 
 	if (t3_read_reg(adap, A_SG_CONTEXT_CMD) & F_CONTEXT_CMD_BUSY)
 		return -EBUSY;
@@ -3677,6 +3613,8 @@ int t3_prep_adapter(struct adapter *adapter, const struct adapter_info *ai,
 	    MAC_STATS_ACCUM_SECS : (MAC_STATS_ACCUM_SECS * 10);
 	adapter->params.pci.vpd_cap_addr =
 	    pci_find_capability(adapter->pdev, PCI_CAP_ID_VPD);
+	if (!adapter->params.pci.vpd_cap_addr)
+		return -ENODEV;
 	ret = get_vpd_params(adapter, &adapter->params.vpd);
 	if (ret < 0)
 		return ret;
@@ -3757,8 +3695,7 @@ int t3_prep_adapter(struct adapter *adapter, const struct adapter_info *ai,
 		memcpy(hw_addr, adapter->params.vpd.eth_base, 5);
 		hw_addr[5] = adapter->params.vpd.eth_base[5] + i;
 
-		memcpy(adapter->port[i]->dev_addr, hw_addr,
-		       ETH_ALEN);
+		eth_hw_addr_set(adapter->port[i], hw_addr);
 		init_link_config(&p->link_config, p->phy.caps);
 		p->phy.ops->power_down(&p->phy, 1);
 
@@ -3806,6 +3743,6 @@ int t3_replay_prep_adapter(struct adapter *adapter)
 		p->phy.ops->power_down(&p->phy, 1);
 	}
 
-return 0;
+	return 0;
 }
 
diff --git a/drivers/net/ethernet/chelsio/cxgb3/t3cdev.h b/drivers/net/ethernet/chelsio/cxgb3/t3cdev.h
index 705713b56636..3c3e6cf6aca6 100644
--- a/drivers/net/ethernet/chelsio/cxgb3/t3cdev.h
+++ b/drivers/net/ethernet/chelsio/cxgb3/t3cdev.h
@@ -60,7 +60,7 @@ struct t3cdev {
 	int (*ctl)(struct t3cdev *dev, unsigned int req, void *data);
 	void (*neigh_update)(struct t3cdev *dev, struct neighbour *neigh);
 	void *priv;		/* driver private data */
-	void *l2opt;		/* optional layer 2 data */
+	void __rcu *l2opt;	/* optional layer 2 data */
 	void *l3opt;		/* optional layer 3 data */
 	void *l4opt;		/* optional layer 4 data */
 	void *ulp;		/* ulp stuff */
diff --git a/drivers/net/ethernet/chelsio/cxgb3/version.h b/drivers/net/ethernet/chelsio/cxgb3/version.h
index 165bfb91487a..b4b2547efc86 100644
--- a/drivers/net/ethernet/chelsio/cxgb3/version.h
+++ b/drivers/net/ethernet/chelsio/cxgb3/version.h
@@ -34,8 +34,6 @@
 #define __CHELSIO_VERSION_H
 #define DRV_DESC "Chelsio T3 Network Driver"
 #define DRV_NAME "cxgb3"
-/* Driver version */
-#define DRV_VERSION "1.1.5-ko"
 
 /* Firmware version */
 #define FW_VERSION_MAJOR 7
diff --git a/drivers/net/ethernet/chelsio/cxgb3/xgmac.c b/drivers/net/ethernet/chelsio/cxgb3/xgmac.c
index 3af19a550372..1bdc6cad1e49 100644
--- a/drivers/net/ethernet/chelsio/cxgb3/xgmac.c
+++ b/drivers/net/ethernet/chelsio/cxgb3/xgmac.c
@@ -240,7 +240,7 @@ static void set_addr_filter(struct cmac *mac, int idx, const u8 * addr)
 }
 
 /* Set one of the station's unicast MAC addresses. */
-int t3_mac_set_address(struct cmac *mac, unsigned int idx, u8 addr[6])
+int t3_mac_set_address(struct cmac *mac, unsigned int idx, const u8 addr[6])
 {
 	if (idx >= mac->nucast)
 		return -EINVAL;
diff --git a/drivers/net/ethernet/chelsio/cxgb4/Makefile b/drivers/net/ethernet/chelsio/cxgb4/Makefile
index c6b71f656992..a4b4d475abf8 100644
--- a/drivers/net/ethernet/chelsio/cxgb4/Makefile
+++ b/drivers/net/ethernet/chelsio/cxgb4/Makefile
@@ -1,10 +1,16 @@
+# SPDX-License-Identifier: GPL-2.0
 #
 # Chelsio T4 driver
 #
 
 obj-$(CONFIG_CHELSIO_T4) += cxgb4.o
 
-cxgb4-objs := cxgb4_main.o l2t.o t4_hw.o sge.o clip_tbl.o cxgb4_ethtool.o cxgb4_uld.o sched.o cxgb4_filter.o cxgb4_tc_u32.o
+cxgb4-objs := cxgb4_main.o l2t.o smt.o t4_hw.o sge.o clip_tbl.o cxgb4_ethtool.o \
+	      cxgb4_uld.o srq.o sched.o cxgb4_filter.o cxgb4_tc_u32.o \
+	      cxgb4_ptp.o cxgb4_tc_flower.o cxgb4_cudbg.o cxgb4_mps.o \
+	      cudbg_common.o cudbg_lib.o cudbg_zlib.o cxgb4_tc_mqprio.o \
+	      cxgb4_tc_matchall.o
 cxgb4-$(CONFIG_CHELSIO_T4_DCB) +=  cxgb4_dcb.o
 cxgb4-$(CONFIG_CHELSIO_T4_FCOE) +=  cxgb4_fcoe.o
 cxgb4-$(CONFIG_DEBUG_FS) += cxgb4_debugfs.o
+cxgb4-$(CONFIG_THERMAL) += cxgb4_thermal.o
diff --git a/drivers/net/ethernet/chelsio/cxgb4/clip_tbl.c b/drivers/net/ethernet/chelsio/cxgb4/clip_tbl.c
index 3103ef9b561d..5060d3998889 100644
--- a/drivers/net/ethernet/chelsio/cxgb4/clip_tbl.c
+++ b/drivers/net/ethernet/chelsio/cxgb4/clip_tbl.c
@@ -96,7 +96,8 @@ int cxgb4_clip_get(const struct net_device *dev, const u32 *lip, u8 v6)
 		if (!ret) {
 			ce = cte;
 			read_unlock_bh(&ctbl->lock);
-			goto found;
+			refcount_inc(&ce->refcnt);
+			return 0;
 		}
 	}
 	read_unlock_bh(&ctbl->lock);
@@ -105,10 +106,9 @@ int cxgb4_clip_get(const struct net_device *dev, const u32 *lip, u8 v6)
 	if (!list_empty(&ctbl->ce_free_head)) {
 		ce = list_first_entry(&ctbl->ce_free_head,
 				      struct clip_entry, list);
-		list_del(&ce->list);
-		INIT_LIST_HEAD(&ce->list);
+		list_del_init(&ce->list);
 		spin_lock_init(&ce->lock);
-		atomic_set(&ce->refcnt, 0);
+		refcount_set(&ce->refcnt, 0);
 		atomic_dec(&ctbl->nfree);
 		list_add_tail(&ce->list, &ctbl->hash_list[hash]);
 		if (v6) {
@@ -120,7 +120,7 @@ int cxgb4_clip_get(const struct net_device *dev, const u32 *lip, u8 v6)
 				write_unlock_bh(&ctbl->lock);
 				dev_err(adap->pdev_dev,
 					"CLIP FW cmd failed with error %d, "
-					"Connections using %pI6c wont be "
+					"Connections using %pI6c won't be "
 					"offloaded",
 					ret, ce->addr6.sin6_addr.s6_addr);
 				return ret;
@@ -133,14 +133,12 @@ int cxgb4_clip_get(const struct net_device *dev, const u32 *lip, u8 v6)
 	} else {
 		write_unlock_bh(&ctbl->lock);
 		dev_info(adap->pdev_dev, "CLIP table overflow, "
-			 "Connections using %pI6c wont be offloaded",
+			 "Connections using %pI6c won't be offloaded",
 			 (void *)lip);
 		return -ENOMEM;
 	}
 	write_unlock_bh(&ctbl->lock);
-found:
-	atomic_inc(&ce->refcnt);
-
+	refcount_set(&ce->refcnt, 1);
 	return 0;
 }
 EXPORT_SYMBOL(cxgb4_clip_get);
@@ -179,9 +177,8 @@ void cxgb4_clip_release(const struct net_device *dev, const u32 *lip, u8 v6)
 found:
 	write_lock_bh(&ctbl->lock);
 	spin_lock_bh(&ce->lock);
-	if (atomic_dec_and_test(&ce->refcnt)) {
-		list_del(&ce->list);
-		INIT_LIST_HEAD(&ce->list);
+	if (refcount_dec_and_test(&ce->refcnt)) {
+		list_del_init(&ce->list);
 		list_add_tail(&ce->list, &ctbl->ce_free_head);
 		atomic_inc(&ctbl->nfree);
 		if (v6)
@@ -266,7 +263,7 @@ int clip_tbl_show(struct seq_file *seq, void *v)
 			ip[0] = '\0';
 			sprintf(ip, "%pISc", &ce->addr);
 			seq_printf(seq, "%-25s   %u\n", ip,
-				   atomic_read(&ce->refcnt));
+				   refcount_read(&ce->refcnt));
 		}
 	}
 	seq_printf(seq, "Free clip entries : %d\n", atomic_read(&ctbl->nfree));
@@ -290,8 +287,7 @@ struct clip_tbl *t4_init_clip_tbl(unsigned int clipt_start,
 	if (clipt_size < CLIPT_MIN_HASH_BUCKETS)
 		return NULL;
 
-	ctbl = kvzalloc(sizeof(*ctbl) +
-			    clipt_size*sizeof(struct list_head), GFP_KERNEL);
+	ctbl = kvzalloc(struct_size(ctbl, hash_list, clipt_size), GFP_KERNEL);
 	if (!ctbl)
 		return NULL;
 
@@ -305,7 +301,7 @@ struct clip_tbl *t4_init_clip_tbl(unsigned int clipt_start,
 	for (i = 0; i < ctbl->clipt_size; ++i)
 		INIT_LIST_HEAD(&ctbl->hash_list[i]);
 
-	cl_list = kvzalloc(clipt_size*sizeof(struct clip_entry), GFP_KERNEL);
+	cl_list = kvcalloc(clipt_size, sizeof(struct clip_entry), GFP_KERNEL);
 	if (!cl_list) {
 		kvfree(ctbl);
 		return NULL;
@@ -325,8 +321,7 @@ void t4_cleanup_clip_tbl(struct adapter *adap)
 	struct clip_tbl *ctbl = adap->clipt;
 
 	if (ctbl) {
-		if (ctbl->cl_list)
-			kvfree(ctbl->cl_list);
+		kvfree(ctbl->cl_list);
 		kvfree(ctbl);
 	}
 }
diff --git a/drivers/net/ethernet/chelsio/cxgb4/clip_tbl.h b/drivers/net/ethernet/chelsio/cxgb4/clip_tbl.h
index 35eb43c6bcbb..847c7fc2bbd9 100644
--- a/drivers/net/ethernet/chelsio/cxgb4/clip_tbl.h
+++ b/drivers/net/ethernet/chelsio/cxgb4/clip_tbl.h
@@ -10,9 +10,11 @@
  *  release for licensing terms and conditions.
  */
 
+#include <linux/refcount.h>
+
 struct clip_entry {
 	spinlock_t lock;	/* Hold while modifying clip reference */
-	atomic_t refcnt;
+	refcount_t refcnt;
 	struct list_head list;
 	union {
 		struct sockaddr_in addr;
@@ -27,7 +29,7 @@ struct clip_tbl {
 	atomic_t nfree;
 	struct list_head ce_free_head;
 	void *cl_list;
-	struct list_head hash_list[0];
+	struct list_head hash_list[] __counted_by(clipt_size);
 };
 
 enum {
diff --git a/drivers/net/ethernet/chelsio/cxgb4/cudbg_common.c b/drivers/net/ethernet/chelsio/cxgb4/cudbg_common.c
new file mode 100644
index 000000000000..175e1a675de5
--- /dev/null
+++ b/drivers/net/ethernet/chelsio/cxgb4/cudbg_common.c
@@ -0,0 +1,56 @@
+// SPDX-License-Identifier: GPL-2.0-only
+/*
+ *  Copyright (C) 2017 Chelsio Communications.  All rights reserved.
+ */
+
+#include "cxgb4.h"
+#include "cudbg_if.h"
+#include "cudbg_lib_common.h"
+
+int cudbg_get_buff(struct cudbg_init *pdbg_init,
+		   struct cudbg_buffer *pdbg_buff, u32 size,
+		   struct cudbg_buffer *pin_buff)
+{
+	u32 offset;
+
+	offset = pdbg_buff->offset;
+	if (offset + size > pdbg_buff->size)
+		return CUDBG_STATUS_NO_MEM;
+
+	if (pdbg_init->compress_type != CUDBG_COMPRESSION_NONE) {
+		if (size > pdbg_init->compress_buff_size)
+			return CUDBG_STATUS_NO_MEM;
+
+		pin_buff->data = (char *)pdbg_init->compress_buff;
+		pin_buff->offset = 0;
+		pin_buff->size = size;
+		return 0;
+	}
+
+	pin_buff->data = (char *)pdbg_buff->data + offset;
+	pin_buff->offset = offset;
+	pin_buff->size = size;
+	return 0;
+}
+
+void cudbg_put_buff(struct cudbg_init *pdbg_init,
+		    struct cudbg_buffer *pin_buff)
+{
+	/* Clear compression buffer for re-use */
+	if (pdbg_init->compress_type != CUDBG_COMPRESSION_NONE)
+		memset(pdbg_init->compress_buff, 0,
+		       pdbg_init->compress_buff_size);
+
+	pin_buff->data = NULL;
+	pin_buff->offset = 0;
+	pin_buff->size = 0;
+}
+
+void cudbg_update_buff(struct cudbg_buffer *pin_buff,
+		       struct cudbg_buffer *pout_buff)
+{
+	/* We already write to buffer provided by ethool, so just
+	 * increment offset to next free space.
+	 */
+	pout_buff->offset += pin_buff->size;
+}
diff --git a/drivers/net/ethernet/chelsio/cxgb4/cudbg_entity.h b/drivers/net/ethernet/chelsio/cxgb4/cudbg_entity.h
new file mode 100644
index 000000000000..d5218e74284c
--- /dev/null
+++ b/drivers/net/ethernet/chelsio/cxgb4/cudbg_entity.h
@@ -0,0 +1,352 @@
+/* SPDX-License-Identifier: GPL-2.0-only */
+/*
+ *  Copyright (C) 2017 Chelsio Communications.  All rights reserved.
+ */
+
+#ifndef __CUDBG_ENTITY_H__
+#define __CUDBG_ENTITY_H__
+
+#define EDC0_FLAG 0
+#define EDC1_FLAG 1
+#define MC_FLAG 2
+#define MC0_FLAG 3
+#define MC1_FLAG 4
+#define HMA_FLAG 5
+
+#define CUDBG_ENTITY_SIGNATURE 0xCCEDB001
+
+struct cudbg_mbox_log {
+	struct mbox_cmd entry;
+	u32 hi[MBOX_LEN / 8];
+	u32 lo[MBOX_LEN / 8];
+};
+
+struct cudbg_cim_qcfg {
+	u8 chip;
+	u16 base[CIM_NUM_IBQ + CIM_NUM_OBQ_T5];
+	u16 size[CIM_NUM_IBQ + CIM_NUM_OBQ_T5];
+	u16 thres[CIM_NUM_IBQ];
+	u32 obq_wr[2 * CIM_NUM_OBQ_T5];
+	u32 stat[4 * (CIM_NUM_IBQ + CIM_NUM_OBQ_T5)];
+};
+
+struct cudbg_rss_vf_conf {
+	u32 rss_vf_vfl;
+	u32 rss_vf_vfh;
+};
+
+struct cudbg_pm_stats {
+	u32 tx_cnt[T6_PM_NSTATS];
+	u32 rx_cnt[T6_PM_NSTATS];
+	u64 tx_cyc[T6_PM_NSTATS];
+	u64 rx_cyc[T6_PM_NSTATS];
+};
+
+struct cudbg_hw_sched {
+	u32 kbps[NTX_SCHED];
+	u32 ipg[NTX_SCHED];
+	u32 pace_tab[NTX_SCHED];
+	u32 mode;
+	u32 map;
+};
+
+#define SGE_QBASE_DATA_REG_NUM 4
+
+struct sge_qbase_reg_field {
+	u32 reg_addr;
+	u32 reg_data[SGE_QBASE_DATA_REG_NUM];
+	/* Max supported PFs */
+	u32 pf_data_value[PCIE_FW_MASTER_M + 1][SGE_QBASE_DATA_REG_NUM];
+	/* Max supported VFs */
+	u32 vf_data_value[T6_VF_M + 1][SGE_QBASE_DATA_REG_NUM];
+	u32 vfcount; /* Actual number of max vfs in current configuration */
+};
+
+struct ireg_field {
+	u32 ireg_addr;
+	u32 ireg_data;
+	u32 ireg_local_offset;
+	u32 ireg_offset_range;
+};
+
+struct ireg_buf {
+	struct ireg_field tp_pio;
+	u32 outbuf[32];
+};
+
+struct cudbg_ulprx_la {
+	u32 data[ULPRX_LA_SIZE * 8];
+	u32 size;
+};
+
+struct cudbg_tp_la {
+	u32 size;
+	u32 mode;
+	u8 data[];
+};
+
+static const char * const cudbg_region[] = {
+	"DBQ contexts:", "IMSG contexts:", "FLM cache:", "TCBs:",
+	"Pstructs:", "Timers:", "Rx FL:", "Tx FL:", "Pstruct FL:",
+	"Tx payload:", "Rx payload:", "LE hash:", "iSCSI region:",
+	"TDDP region:", "TPT region:", "STAG region:", "RQ region:",
+	"RQUDP region:", "PBL region:", "TXPBL region:",
+	"DBVFIFO region:", "ULPRX state:", "ULPTX state:",
+	"On-chip queues:"
+};
+
+/* Memory region info relative to current memory (i.e. wrt 0). */
+struct cudbg_region_info {
+	bool exist; /* Does region exists in current memory? */
+	u32 start;  /* Start wrt 0 */
+	u32 end;    /* End wrt 0 */
+};
+
+struct cudbg_mem_desc {
+	u32 base;
+	u32 limit;
+	u32 idx;
+};
+
+#define CUDBG_MEMINFO_REV 1
+
+struct cudbg_meminfo {
+	struct cudbg_mem_desc avail[4];
+	struct cudbg_mem_desc mem[ARRAY_SIZE(cudbg_region) + 3];
+	u32 avail_c;
+	u32 mem_c;
+	u32 up_ram_lo;
+	u32 up_ram_hi;
+	u32 up_extmem2_lo;
+	u32 up_extmem2_hi;
+	u32 rx_pages_data[3];
+	u32 tx_pages_data[4];
+	u32 p_structs;
+	u32 reserved[12];
+	u32 port_used[4];
+	u32 port_alloc[4];
+	u32 loopback_used[NCHAN];
+	u32 loopback_alloc[NCHAN];
+	u32 p_structs_free_cnt;
+	u32 free_rx_cnt;
+	u32 free_tx_cnt;
+};
+
+struct cudbg_cim_pif_la {
+	int size;
+	u8 data[];
+};
+
+struct cudbg_clk_info {
+	u64 retransmit_min;
+	u64 retransmit_max;
+	u64 persist_timer_min;
+	u64 persist_timer_max;
+	u64 keepalive_idle_timer;
+	u64 keepalive_interval;
+	u64 initial_srtt;
+	u64 finwait2_timer;
+	u32 dack_timer;
+	u32 res;
+	u32 cclk_ps;
+	u32 tre;
+	u32 dack_re;
+};
+
+struct cudbg_tid_info_region {
+	u32 ntids;
+	u32 nstids;
+	u32 stid_base;
+	u32 hash_base;
+
+	u32 natids;
+	u32 nftids;
+	u32 ftid_base;
+	u32 aftid_base;
+	u32 aftid_end;
+
+	u32 sftid_base;
+	u32 nsftids;
+
+	u32 uotid_base;
+	u32 nuotids;
+
+	u32 sb;
+	u32 flags;
+	u32 le_db_conf;
+	u32 ip_users;
+	u32 ipv6_users;
+
+	u32 hpftid_base;
+	u32 nhpftids;
+};
+
+#define CUDBG_TID_INFO_REV 1
+
+struct cudbg_tid_info_region_rev1 {
+	struct cudbg_ver_hdr ver_hdr;
+	struct cudbg_tid_info_region tid;
+	u32 tid_start;
+	u32 reserved[16];
+};
+
+#define CUDBG_LOWMEM_MAX_CTXT_QIDS 256
+#define CUDBG_MAX_FL_QIDS 1024
+
+struct cudbg_ch_cntxt {
+	u32 cntxt_type;
+	u32 cntxt_id;
+	u32 data[SGE_CTXT_SIZE / 4];
+};
+
+#define CUDBG_MAX_RPLC_SIZE 128
+
+struct cudbg_mps_tcam {
+	u64 mask;
+	u32 rplc[8];
+	u32 idx;
+	u32 cls_lo;
+	u32 cls_hi;
+	u32 rplc_size;
+	u32 vniy;
+	u32 vnix;
+	u32 dip_hit;
+	u32 vlan_vld;
+	u32 repli;
+	u16 ivlan;
+	u8 addr[ETH_ALEN];
+	u8 lookup_type;
+	u8 port_num;
+	u8 reserved[2];
+};
+
+#define CUDBG_VPD_VER_ADDR 0x18c7
+#define CUDBG_VPD_VER_LEN 2
+
+struct cudbg_vpd_data {
+	u8 sn[SERNUM_LEN + 1];
+	u8 bn[PN_LEN + 1];
+	u8 na[MACADDR_LEN + 1];
+	u8 mn[ID_LEN + 1];
+	u16 fw_major;
+	u16 fw_minor;
+	u16 fw_micro;
+	u16 fw_build;
+	u32 scfg_vers;
+	u32 vpd_vers;
+};
+
+#define CUDBG_MAX_TCAM_TID 0x800
+#define CUDBG_T6_CLIP 1536
+#define CUDBG_MAX_TID_COMP_EN 6144
+#define CUDBG_MAX_TID_COMP_DIS 3072
+
+enum cudbg_le_entry_types {
+	LE_ET_UNKNOWN = 0,
+	LE_ET_TCAM_CON = 1,
+	LE_ET_TCAM_SERVER = 2,
+	LE_ET_TCAM_FILTER = 3,
+	LE_ET_TCAM_CLIP = 4,
+	LE_ET_TCAM_ROUTING = 5,
+	LE_ET_HASH_CON = 6,
+	LE_ET_INVALID_TID = 8,
+};
+
+struct cudbg_tcam {
+	u32 filter_start;
+	u32 server_start;
+	u32 clip_start;
+	u32 routing_start;
+	u32 tid_hash_base;
+	u32 max_tid;
+};
+
+struct cudbg_tid_data {
+	u32 tid;
+	u32 dbig_cmd;
+	u32 dbig_conf;
+	u32 dbig_rsp_stat;
+	u32 data[NUM_LE_DB_DBGI_RSP_DATA_INSTANCES];
+};
+
+#define CUDBG_NUM_ULPTX 11
+#define CUDBG_NUM_ULPTX_READ 512
+#define CUDBG_NUM_ULPTX_ASIC 6
+#define CUDBG_NUM_ULPTX_ASIC_READ 128
+
+#define CUDBG_ULPTX_LA_REV 1
+
+struct cudbg_ulptx_la {
+	u32 rdptr[CUDBG_NUM_ULPTX];
+	u32 wrptr[CUDBG_NUM_ULPTX];
+	u32 rddata[CUDBG_NUM_ULPTX];
+	u32 rd_data[CUDBG_NUM_ULPTX][CUDBG_NUM_ULPTX_READ];
+	u32 rdptr_asic[CUDBG_NUM_ULPTX_ASIC_READ];
+	u32 rddata_asic[CUDBG_NUM_ULPTX_ASIC_READ][CUDBG_NUM_ULPTX_ASIC];
+};
+
+#define CUDBG_CHAC_PBT_ADDR 0x2800
+#define CUDBG_CHAC_PBT_LRF  0x3000
+#define CUDBG_CHAC_PBT_DATA 0x3800
+#define CUDBG_PBT_DYNAMIC_ENTRIES 8
+#define CUDBG_PBT_STATIC_ENTRIES 16
+#define CUDBG_LRF_ENTRIES 8
+#define CUDBG_PBT_DATA_ENTRIES 512
+
+struct cudbg_pbt_tables {
+	u32 pbt_dynamic[CUDBG_PBT_DYNAMIC_ENTRIES];
+	u32 pbt_static[CUDBG_PBT_STATIC_ENTRIES];
+	u32 lrf_table[CUDBG_LRF_ENTRIES];
+	u32 pbt_data[CUDBG_PBT_DATA_ENTRIES];
+};
+
+enum cudbg_qdesc_qtype {
+	CUDBG_QTYPE_UNKNOWN = 0,
+	CUDBG_QTYPE_NIC_TXQ,
+	CUDBG_QTYPE_NIC_RXQ,
+	CUDBG_QTYPE_NIC_FLQ,
+	CUDBG_QTYPE_CTRLQ,
+	CUDBG_QTYPE_FWEVTQ,
+	CUDBG_QTYPE_INTRQ,
+	CUDBG_QTYPE_PTP_TXQ,
+	CUDBG_QTYPE_OFLD_TXQ,
+	CUDBG_QTYPE_RDMA_RXQ,
+	CUDBG_QTYPE_RDMA_FLQ,
+	CUDBG_QTYPE_RDMA_CIQ,
+	CUDBG_QTYPE_ISCSI_RXQ,
+	CUDBG_QTYPE_ISCSI_FLQ,
+	CUDBG_QTYPE_ISCSIT_RXQ,
+	CUDBG_QTYPE_ISCSIT_FLQ,
+	CUDBG_QTYPE_CRYPTO_TXQ,
+	CUDBG_QTYPE_CRYPTO_RXQ,
+	CUDBG_QTYPE_CRYPTO_FLQ,
+	CUDBG_QTYPE_TLS_RXQ,
+	CUDBG_QTYPE_TLS_FLQ,
+	CUDBG_QTYPE_ETHOFLD_TXQ,
+	CUDBG_QTYPE_ETHOFLD_RXQ,
+	CUDBG_QTYPE_ETHOFLD_FLQ,
+	CUDBG_QTYPE_MAX,
+};
+
+#define CUDBG_QDESC_REV 1
+
+struct cudbg_qdesc_entry {
+	u32 data_size;
+	u32 qtype;
+	u32 qid;
+	u32 desc_size;
+	u32 num_desc;
+	u8 data[]; /* Must be last */
+};
+
+struct cudbg_qdesc_info {
+	u32 qdesc_entry_size;
+	u32 num_queues;
+	u8 data[]; /* Must be last */
+};
+
+#define IREG_NUM_ELEM 4
+
+#define CUDBG_NUM_PCIE_CONFIG_REGS 0x61
+
+#endif /* __CUDBG_ENTITY_H__ */
diff --git a/drivers/net/ethernet/chelsio/cxgb4/cudbg_if.h b/drivers/net/ethernet/chelsio/cxgb4/cudbg_if.h
new file mode 100644
index 000000000000..c84719e3ca08
--- /dev/null
+++ b/drivers/net/ethernet/chelsio/cxgb4/cudbg_if.h
@@ -0,0 +1,91 @@
+/* SPDX-License-Identifier: GPL-2.0-only */
+/*
+ *  Copyright (C) 2017 Chelsio Communications.  All rights reserved.
+ */
+
+#ifndef __CUDBG_IF_H__
+#define __CUDBG_IF_H__
+
+/* Error codes */
+#define CUDBG_STATUS_NO_MEM -19
+#define CUDBG_STATUS_ENTITY_NOT_FOUND -24
+#define CUDBG_STATUS_NOT_IMPLEMENTED -28
+#define CUDBG_SYSTEM_ERROR -29
+#define CUDBG_STATUS_CCLK_NOT_DEFINED -32
+#define CUDBG_STATUS_PARTIAL_DATA -41
+
+#define CUDBG_MAJOR_VERSION 1
+#define CUDBG_MINOR_VERSION 14
+
+enum cudbg_dbg_entity_type {
+	CUDBG_REG_DUMP = 1,
+	CUDBG_DEV_LOG = 2,
+	CUDBG_CIM_LA = 3,
+	CUDBG_CIM_MA_LA = 4,
+	CUDBG_CIM_QCFG = 5,
+	CUDBG_CIM_IBQ_TP0 = 6,
+	CUDBG_CIM_IBQ_TP1 = 7,
+	CUDBG_CIM_IBQ_ULP = 8,
+	CUDBG_CIM_IBQ_SGE0 = 9,
+	CUDBG_CIM_IBQ_SGE1 = 10,
+	CUDBG_CIM_IBQ_NCSI = 11,
+	CUDBG_CIM_OBQ_ULP0 = 12,
+	CUDBG_CIM_OBQ_ULP1 = 13,
+	CUDBG_CIM_OBQ_ULP2 = 14,
+	CUDBG_CIM_OBQ_ULP3 = 15,
+	CUDBG_CIM_OBQ_SGE = 16,
+	CUDBG_CIM_OBQ_NCSI = 17,
+	CUDBG_EDC0 = 18,
+	CUDBG_EDC1 = 19,
+	CUDBG_MC0 = 20,
+	CUDBG_MC1 = 21,
+	CUDBG_RSS = 22,
+	CUDBG_RSS_VF_CONF = 25,
+	CUDBG_PATH_MTU = 27,
+	CUDBG_PM_STATS = 30,
+	CUDBG_HW_SCHED = 31,
+	CUDBG_TP_INDIRECT = 36,
+	CUDBG_SGE_INDIRECT = 37,
+	CUDBG_ULPRX_LA = 41,
+	CUDBG_TP_LA = 43,
+	CUDBG_MEMINFO = 44,
+	CUDBG_CIM_PIF_LA = 45,
+	CUDBG_CLK = 46,
+	CUDBG_CIM_OBQ_RXQ0 = 47,
+	CUDBG_CIM_OBQ_RXQ1 = 48,
+	CUDBG_PCIE_INDIRECT = 50,
+	CUDBG_PM_INDIRECT = 51,
+	CUDBG_TID_INFO = 54,
+	CUDBG_PCIE_CONFIG = 55,
+	CUDBG_DUMP_CONTEXT = 56,
+	CUDBG_MPS_TCAM = 57,
+	CUDBG_VPD_DATA = 58,
+	CUDBG_LE_TCAM = 59,
+	CUDBG_CCTRL = 60,
+	CUDBG_MA_INDIRECT = 61,
+	CUDBG_ULPTX_LA = 62,
+	CUDBG_UP_CIM_INDIRECT = 64,
+	CUDBG_PBT_TABLE = 65,
+	CUDBG_MBOX_LOG = 66,
+	CUDBG_HMA_INDIRECT = 67,
+	CUDBG_HMA = 68,
+	CUDBG_QDESC = 70,
+	CUDBG_FLASH = 71,
+	CUDBG_MAX_ENTITY = 72,
+};
+
+struct cudbg_init {
+	struct adapter *adap; /* Pointer to adapter structure */
+	void *outbuf; /* Output buffer */
+	u32 outbuf_size;  /* Output buffer size */
+	u8 compress_type; /* Type of compression to use */
+	void *compress_buff; /* Compression buffer */
+	u32 compress_buff_size; /* Compression buffer size */
+	void *workspace; /* Workspace for zlib */
+};
+
+static inline unsigned int cudbg_mbytes_to_bytes(unsigned int size)
+{
+	return size * 1024 * 1024;
+}
+#endif /* __CUDBG_IF_H__ */
diff --git a/drivers/net/ethernet/chelsio/cxgb4/cudbg_lib.c b/drivers/net/ethernet/chelsio/cxgb4/cudbg_lib.c
new file mode 100644
index 000000000000..557c591a6ce3
--- /dev/null
+++ b/drivers/net/ethernet/chelsio/cxgb4/cudbg_lib.c
@@ -0,0 +1,3609 @@
+// SPDX-License-Identifier: GPL-2.0-only
+/*
+ *  Copyright (C) 2017 Chelsio Communications.  All rights reserved.
+ */
+
+#include <linux/sort.h>
+#include <linux/string.h>
+
+#include "t4_regs.h"
+#include "cxgb4.h"
+#include "cxgb4_cudbg.h"
+#include "cudbg_if.h"
+#include "cudbg_lib_common.h"
+#include "cudbg_entity.h"
+#include "cudbg_lib.h"
+#include "cudbg_zlib.h"
+#include "cxgb4_tc_mqprio.h"
+
+static const u32 t6_tp_pio_array[][IREG_NUM_ELEM] = {
+	{0x7e40, 0x7e44, 0x020, 28}, /* t6_tp_pio_regs_20_to_3b */
+	{0x7e40, 0x7e44, 0x040, 10}, /* t6_tp_pio_regs_40_to_49 */
+	{0x7e40, 0x7e44, 0x050, 10}, /* t6_tp_pio_regs_50_to_59 */
+	{0x7e40, 0x7e44, 0x060, 14}, /* t6_tp_pio_regs_60_to_6d */
+	{0x7e40, 0x7e44, 0x06F, 1}, /* t6_tp_pio_regs_6f */
+	{0x7e40, 0x7e44, 0x070, 6}, /* t6_tp_pio_regs_70_to_75 */
+	{0x7e40, 0x7e44, 0x130, 18}, /* t6_tp_pio_regs_130_to_141 */
+	{0x7e40, 0x7e44, 0x145, 19}, /* t6_tp_pio_regs_145_to_157 */
+	{0x7e40, 0x7e44, 0x160, 1}, /* t6_tp_pio_regs_160 */
+	{0x7e40, 0x7e44, 0x230, 25}, /* t6_tp_pio_regs_230_to_248 */
+	{0x7e40, 0x7e44, 0x24a, 3}, /* t6_tp_pio_regs_24c */
+	{0x7e40, 0x7e44, 0x8C0, 1} /* t6_tp_pio_regs_8c0 */
+};
+
+static const u32 t5_tp_pio_array[][IREG_NUM_ELEM] = {
+	{0x7e40, 0x7e44, 0x020, 28}, /* t5_tp_pio_regs_20_to_3b */
+	{0x7e40, 0x7e44, 0x040, 19}, /* t5_tp_pio_regs_40_to_52 */
+	{0x7e40, 0x7e44, 0x054, 2}, /* t5_tp_pio_regs_54_to_55 */
+	{0x7e40, 0x7e44, 0x060, 13}, /* t5_tp_pio_regs_60_to_6c */
+	{0x7e40, 0x7e44, 0x06F, 1}, /* t5_tp_pio_regs_6f */
+	{0x7e40, 0x7e44, 0x120, 4}, /* t5_tp_pio_regs_120_to_123 */
+	{0x7e40, 0x7e44, 0x12b, 2}, /* t5_tp_pio_regs_12b_to_12c */
+	{0x7e40, 0x7e44, 0x12f, 21}, /* t5_tp_pio_regs_12f_to_143 */
+	{0x7e40, 0x7e44, 0x145, 19}, /* t5_tp_pio_regs_145_to_157 */
+	{0x7e40, 0x7e44, 0x230, 25}, /* t5_tp_pio_regs_230_to_248 */
+	{0x7e40, 0x7e44, 0x8C0, 1} /* t5_tp_pio_regs_8c0 */
+};
+
+static const u32 t6_tp_tm_pio_array[][IREG_NUM_ELEM] = {
+	{0x7e18, 0x7e1c, 0x0, 12}
+};
+
+static const u32 t5_tp_tm_pio_array[][IREG_NUM_ELEM] = {
+	{0x7e18, 0x7e1c, 0x0, 12}
+};
+
+static const u32 t6_tp_mib_index_array[6][IREG_NUM_ELEM] = {
+	{0x7e50, 0x7e54, 0x0, 13},
+	{0x7e50, 0x7e54, 0x10, 6},
+	{0x7e50, 0x7e54, 0x18, 21},
+	{0x7e50, 0x7e54, 0x30, 32},
+	{0x7e50, 0x7e54, 0x50, 22},
+	{0x7e50, 0x7e54, 0x68, 12}
+};
+
+static const u32 t5_tp_mib_index_array[9][IREG_NUM_ELEM] = {
+	{0x7e50, 0x7e54, 0x0, 13},
+	{0x7e50, 0x7e54, 0x10, 6},
+	{0x7e50, 0x7e54, 0x18, 8},
+	{0x7e50, 0x7e54, 0x20, 13},
+	{0x7e50, 0x7e54, 0x30, 16},
+	{0x7e50, 0x7e54, 0x40, 16},
+	{0x7e50, 0x7e54, 0x50, 16},
+	{0x7e50, 0x7e54, 0x60, 6},
+	{0x7e50, 0x7e54, 0x68, 4}
+};
+
+static const u32 t5_sge_dbg_index_array[2][IREG_NUM_ELEM] = {
+	{0x10cc, 0x10d0, 0x0, 16},
+	{0x10cc, 0x10d4, 0x0, 16},
+};
+
+static const u32 t6_sge_qbase_index_array[] = {
+	/* 1 addr reg SGE_QBASE_INDEX and 4 data reg SGE_QBASE_MAP[0-3] */
+	0x1250, 0x1240, 0x1244, 0x1248, 0x124c,
+};
+
+static const u32 t5_pcie_pdbg_array[][IREG_NUM_ELEM] = {
+	{0x5a04, 0x5a0c, 0x00, 0x20}, /* t5_pcie_pdbg_regs_00_to_20 */
+	{0x5a04, 0x5a0c, 0x21, 0x20}, /* t5_pcie_pdbg_regs_21_to_40 */
+	{0x5a04, 0x5a0c, 0x41, 0x10}, /* t5_pcie_pdbg_regs_41_to_50 */
+};
+
+static const u32 t5_pcie_cdbg_array[][IREG_NUM_ELEM] = {
+	{0x5a10, 0x5a18, 0x00, 0x20}, /* t5_pcie_cdbg_regs_00_to_20 */
+	{0x5a10, 0x5a18, 0x21, 0x18}, /* t5_pcie_cdbg_regs_21_to_37 */
+};
+
+static const u32 t5_pm_rx_array[][IREG_NUM_ELEM] = {
+	{0x8FD0, 0x8FD4, 0x10000, 0x20}, /* t5_pm_rx_regs_10000_to_10020 */
+	{0x8FD0, 0x8FD4, 0x10021, 0x0D}, /* t5_pm_rx_regs_10021_to_1002c */
+};
+
+static const u32 t5_pm_tx_array[][IREG_NUM_ELEM] = {
+	{0x8FF0, 0x8FF4, 0x10000, 0x20}, /* t5_pm_tx_regs_10000_to_10020 */
+	{0x8FF0, 0x8FF4, 0x10021, 0x1D}, /* t5_pm_tx_regs_10021_to_1003c */
+};
+
+static const u32 t5_pcie_config_array[][2] = {
+	{0x0, 0x34},
+	{0x3c, 0x40},
+	{0x50, 0x64},
+	{0x70, 0x80},
+	{0x94, 0xa0},
+	{0xb0, 0xb8},
+	{0xd0, 0xd4},
+	{0x100, 0x128},
+	{0x140, 0x148},
+	{0x150, 0x164},
+	{0x170, 0x178},
+	{0x180, 0x194},
+	{0x1a0, 0x1b8},
+	{0x1c0, 0x208},
+};
+
+static const u32 t6_ma_ireg_array[][IREG_NUM_ELEM] = {
+	{0x78f8, 0x78fc, 0xa000, 23}, /* t6_ma_regs_a000_to_a016 */
+	{0x78f8, 0x78fc, 0xa400, 30}, /* t6_ma_regs_a400_to_a41e */
+	{0x78f8, 0x78fc, 0xa800, 20} /* t6_ma_regs_a800_to_a813 */
+};
+
+static const u32 t6_ma_ireg_array2[][IREG_NUM_ELEM] = {
+	{0x78f8, 0x78fc, 0xe400, 17}, /* t6_ma_regs_e400_to_e600 */
+	{0x78f8, 0x78fc, 0xe640, 13} /* t6_ma_regs_e640_to_e7c0 */
+};
+
+static const u32 t6_up_cim_reg_array[][IREG_NUM_ELEM + 1] = {
+	{0x7b50, 0x7b54, 0x2000, 0x20, 0}, /* up_cim_2000_to_207c */
+	{0x7b50, 0x7b54, 0x2080, 0x1d, 0}, /* up_cim_2080_to_20fc */
+	{0x7b50, 0x7b54, 0x00, 0x20, 0}, /* up_cim_00_to_7c */
+	{0x7b50, 0x7b54, 0x80, 0x20, 0}, /* up_cim_80_to_fc */
+	{0x7b50, 0x7b54, 0x100, 0x11, 0}, /* up_cim_100_to_14c */
+	{0x7b50, 0x7b54, 0x200, 0x10, 0}, /* up_cim_200_to_23c */
+	{0x7b50, 0x7b54, 0x240, 0x2, 0}, /* up_cim_240_to_244 */
+	{0x7b50, 0x7b54, 0x250, 0x2, 0}, /* up_cim_250_to_254 */
+	{0x7b50, 0x7b54, 0x260, 0x2, 0}, /* up_cim_260_to_264 */
+	{0x7b50, 0x7b54, 0x270, 0x2, 0}, /* up_cim_270_to_274 */
+	{0x7b50, 0x7b54, 0x280, 0x20, 0}, /* up_cim_280_to_2fc */
+	{0x7b50, 0x7b54, 0x300, 0x20, 0}, /* up_cim_300_to_37c */
+	{0x7b50, 0x7b54, 0x380, 0x14, 0}, /* up_cim_380_to_3cc */
+	{0x7b50, 0x7b54, 0x4900, 0x4, 0x4}, /* up_cim_4900_to_4c60 */
+	{0x7b50, 0x7b54, 0x4904, 0x4, 0x4}, /* up_cim_4904_to_4c64 */
+	{0x7b50, 0x7b54, 0x4908, 0x4, 0x4}, /* up_cim_4908_to_4c68 */
+	{0x7b50, 0x7b54, 0x4910, 0x4, 0x4}, /* up_cim_4910_to_4c70 */
+	{0x7b50, 0x7b54, 0x4914, 0x4, 0x4}, /* up_cim_4914_to_4c74 */
+	{0x7b50, 0x7b54, 0x4920, 0x10, 0x10}, /* up_cim_4920_to_4a10 */
+	{0x7b50, 0x7b54, 0x4924, 0x10, 0x10}, /* up_cim_4924_to_4a14 */
+	{0x7b50, 0x7b54, 0x4928, 0x10, 0x10}, /* up_cim_4928_to_4a18 */
+	{0x7b50, 0x7b54, 0x492c, 0x10, 0x10}, /* up_cim_492c_to_4a1c */
+};
+
+static const u32 t5_up_cim_reg_array[][IREG_NUM_ELEM + 1] = {
+	{0x7b50, 0x7b54, 0x2000, 0x20, 0}, /* up_cim_2000_to_207c */
+	{0x7b50, 0x7b54, 0x2080, 0x19, 0}, /* up_cim_2080_to_20ec */
+	{0x7b50, 0x7b54, 0x00, 0x20, 0}, /* up_cim_00_to_7c */
+	{0x7b50, 0x7b54, 0x80, 0x20, 0}, /* up_cim_80_to_fc */
+	{0x7b50, 0x7b54, 0x100, 0x11, 0}, /* up_cim_100_to_14c */
+	{0x7b50, 0x7b54, 0x200, 0x10, 0}, /* up_cim_200_to_23c */
+	{0x7b50, 0x7b54, 0x240, 0x2, 0}, /* up_cim_240_to_244 */
+	{0x7b50, 0x7b54, 0x250, 0x2, 0}, /* up_cim_250_to_254 */
+	{0x7b50, 0x7b54, 0x260, 0x2, 0}, /* up_cim_260_to_264 */
+	{0x7b50, 0x7b54, 0x270, 0x2, 0}, /* up_cim_270_to_274 */
+	{0x7b50, 0x7b54, 0x280, 0x20, 0}, /* up_cim_280_to_2fc */
+	{0x7b50, 0x7b54, 0x300, 0x20, 0}, /* up_cim_300_to_37c */
+	{0x7b50, 0x7b54, 0x380, 0x14, 0}, /* up_cim_380_to_3cc */
+};
+
+static const u32 t6_hma_ireg_array[][IREG_NUM_ELEM] = {
+	{0x51320, 0x51324, 0xa000, 32} /* t6_hma_regs_a000_to_a01f */
+};
+
+u32 cudbg_get_entity_length(struct adapter *adap, u32 entity)
+{
+	struct cudbg_tcam tcam_region = { 0 };
+	u32 value, n = 0, len = 0;
+
+	switch (entity) {
+	case CUDBG_REG_DUMP:
+		switch (CHELSIO_CHIP_VERSION(adap->params.chip)) {
+		case CHELSIO_T4:
+			len = T4_REGMAP_SIZE;
+			break;
+		case CHELSIO_T5:
+		case CHELSIO_T6:
+			len = T5_REGMAP_SIZE;
+			break;
+		default:
+			break;
+		}
+		break;
+	case CUDBG_DEV_LOG:
+		len = adap->params.devlog.size;
+		break;
+	case CUDBG_CIM_LA:
+		if (is_t6(adap->params.chip)) {
+			len = adap->params.cim_la_size / 10 + 1;
+			len *= 10 * sizeof(u32);
+		} else {
+			len = adap->params.cim_la_size / 8;
+			len *= 8 * sizeof(u32);
+		}
+		len += sizeof(u32); /* for reading CIM LA configuration */
+		break;
+	case CUDBG_CIM_MA_LA:
+		len = 2 * CIM_MALA_SIZE * 5 * sizeof(u32);
+		break;
+	case CUDBG_CIM_QCFG:
+		len = sizeof(struct cudbg_cim_qcfg);
+		break;
+	case CUDBG_CIM_IBQ_TP0:
+	case CUDBG_CIM_IBQ_TP1:
+	case CUDBG_CIM_IBQ_ULP:
+	case CUDBG_CIM_IBQ_SGE0:
+	case CUDBG_CIM_IBQ_SGE1:
+	case CUDBG_CIM_IBQ_NCSI:
+		len = CIM_IBQ_SIZE * 4 * sizeof(u32);
+		break;
+	case CUDBG_CIM_OBQ_ULP0:
+		len = cudbg_cim_obq_size(adap, 0);
+		break;
+	case CUDBG_CIM_OBQ_ULP1:
+		len = cudbg_cim_obq_size(adap, 1);
+		break;
+	case CUDBG_CIM_OBQ_ULP2:
+		len = cudbg_cim_obq_size(adap, 2);
+		break;
+	case CUDBG_CIM_OBQ_ULP3:
+		len = cudbg_cim_obq_size(adap, 3);
+		break;
+	case CUDBG_CIM_OBQ_SGE:
+		len = cudbg_cim_obq_size(adap, 4);
+		break;
+	case CUDBG_CIM_OBQ_NCSI:
+		len = cudbg_cim_obq_size(adap, 5);
+		break;
+	case CUDBG_CIM_OBQ_RXQ0:
+		len = cudbg_cim_obq_size(adap, 6);
+		break;
+	case CUDBG_CIM_OBQ_RXQ1:
+		len = cudbg_cim_obq_size(adap, 7);
+		break;
+	case CUDBG_EDC0:
+		value = t4_read_reg(adap, MA_TARGET_MEM_ENABLE_A);
+		if (value & EDRAM0_ENABLE_F) {
+			value = t4_read_reg(adap, MA_EDRAM0_BAR_A);
+			len = EDRAM0_SIZE_G(value);
+		}
+		len = cudbg_mbytes_to_bytes(len);
+		break;
+	case CUDBG_EDC1:
+		value = t4_read_reg(adap, MA_TARGET_MEM_ENABLE_A);
+		if (value & EDRAM1_ENABLE_F) {
+			value = t4_read_reg(adap, MA_EDRAM1_BAR_A);
+			len = EDRAM1_SIZE_G(value);
+		}
+		len = cudbg_mbytes_to_bytes(len);
+		break;
+	case CUDBG_MC0:
+		value = t4_read_reg(adap, MA_TARGET_MEM_ENABLE_A);
+		if (value & EXT_MEM0_ENABLE_F) {
+			value = t4_read_reg(adap, MA_EXT_MEMORY0_BAR_A);
+			len = EXT_MEM0_SIZE_G(value);
+		}
+		len = cudbg_mbytes_to_bytes(len);
+		break;
+	case CUDBG_MC1:
+		value = t4_read_reg(adap, MA_TARGET_MEM_ENABLE_A);
+		if (value & EXT_MEM1_ENABLE_F) {
+			value = t4_read_reg(adap, MA_EXT_MEMORY1_BAR_A);
+			len = EXT_MEM1_SIZE_G(value);
+		}
+		len = cudbg_mbytes_to_bytes(len);
+		break;
+	case CUDBG_RSS:
+		len = t4_chip_rss_size(adap) * sizeof(u16);
+		break;
+	case CUDBG_RSS_VF_CONF:
+		len = adap->params.arch.vfcount *
+		      sizeof(struct cudbg_rss_vf_conf);
+		break;
+	case CUDBG_PATH_MTU:
+		len = NMTUS * sizeof(u16);
+		break;
+	case CUDBG_PM_STATS:
+		len = sizeof(struct cudbg_pm_stats);
+		break;
+	case CUDBG_HW_SCHED:
+		len = sizeof(struct cudbg_hw_sched);
+		break;
+	case CUDBG_TP_INDIRECT:
+		switch (CHELSIO_CHIP_VERSION(adap->params.chip)) {
+		case CHELSIO_T5:
+			n = sizeof(t5_tp_pio_array) +
+			    sizeof(t5_tp_tm_pio_array) +
+			    sizeof(t5_tp_mib_index_array);
+			break;
+		case CHELSIO_T6:
+			n = sizeof(t6_tp_pio_array) +
+			    sizeof(t6_tp_tm_pio_array) +
+			    sizeof(t6_tp_mib_index_array);
+			break;
+		default:
+			break;
+		}
+		n = n / (IREG_NUM_ELEM * sizeof(u32));
+		len = sizeof(struct ireg_buf) * n;
+		break;
+	case CUDBG_SGE_INDIRECT:
+		len = sizeof(struct ireg_buf) * 2 +
+		      sizeof(struct sge_qbase_reg_field);
+		break;
+	case CUDBG_ULPRX_LA:
+		len = sizeof(struct cudbg_ulprx_la);
+		break;
+	case CUDBG_TP_LA:
+		len = sizeof(struct cudbg_tp_la) + TPLA_SIZE * sizeof(u64);
+		break;
+	case CUDBG_MEMINFO:
+		len = sizeof(struct cudbg_ver_hdr) +
+		      sizeof(struct cudbg_meminfo);
+		break;
+	case CUDBG_CIM_PIF_LA:
+		len = sizeof(struct cudbg_cim_pif_la);
+		len += 2 * CIM_PIFLA_SIZE * 6 * sizeof(u32);
+		break;
+	case CUDBG_CLK:
+		len = sizeof(struct cudbg_clk_info);
+		break;
+	case CUDBG_PCIE_INDIRECT:
+		n = sizeof(t5_pcie_pdbg_array) / (IREG_NUM_ELEM * sizeof(u32));
+		len = sizeof(struct ireg_buf) * n * 2;
+		break;
+	case CUDBG_PM_INDIRECT:
+		n = sizeof(t5_pm_rx_array) / (IREG_NUM_ELEM * sizeof(u32));
+		len = sizeof(struct ireg_buf) * n * 2;
+		break;
+	case CUDBG_TID_INFO:
+		len = sizeof(struct cudbg_tid_info_region_rev1);
+		break;
+	case CUDBG_PCIE_CONFIG:
+		len = sizeof(u32) * CUDBG_NUM_PCIE_CONFIG_REGS;
+		break;
+	case CUDBG_DUMP_CONTEXT:
+		len = cudbg_dump_context_size(adap);
+		break;
+	case CUDBG_MPS_TCAM:
+		len = sizeof(struct cudbg_mps_tcam) *
+		      adap->params.arch.mps_tcam_size;
+		break;
+	case CUDBG_VPD_DATA:
+		len = sizeof(struct cudbg_vpd_data);
+		break;
+	case CUDBG_LE_TCAM:
+		cudbg_fill_le_tcam_info(adap, &tcam_region);
+		len = sizeof(struct cudbg_tcam) +
+		      sizeof(struct cudbg_tid_data) * tcam_region.max_tid;
+		break;
+	case CUDBG_CCTRL:
+		len = sizeof(u16) * NMTUS * NCCTRL_WIN;
+		break;
+	case CUDBG_MA_INDIRECT:
+		if (CHELSIO_CHIP_VERSION(adap->params.chip) > CHELSIO_T5) {
+			n = sizeof(t6_ma_ireg_array) /
+			    (IREG_NUM_ELEM * sizeof(u32));
+			len = sizeof(struct ireg_buf) * n * 2;
+		}
+		break;
+	case CUDBG_ULPTX_LA:
+		len = sizeof(struct cudbg_ver_hdr) +
+		      sizeof(struct cudbg_ulptx_la);
+		break;
+	case CUDBG_UP_CIM_INDIRECT:
+		n = 0;
+		if (is_t5(adap->params.chip))
+			n = sizeof(t5_up_cim_reg_array) /
+			    ((IREG_NUM_ELEM + 1) * sizeof(u32));
+		else if (is_t6(adap->params.chip))
+			n = sizeof(t6_up_cim_reg_array) /
+			    ((IREG_NUM_ELEM + 1) * sizeof(u32));
+		len = sizeof(struct ireg_buf) * n;
+		break;
+	case CUDBG_PBT_TABLE:
+		len = sizeof(struct cudbg_pbt_tables);
+		break;
+	case CUDBG_MBOX_LOG:
+		len = sizeof(struct cudbg_mbox_log) * adap->mbox_log->size;
+		break;
+	case CUDBG_HMA_INDIRECT:
+		if (CHELSIO_CHIP_VERSION(adap->params.chip) > CHELSIO_T5) {
+			n = sizeof(t6_hma_ireg_array) /
+			    (IREG_NUM_ELEM * sizeof(u32));
+			len = sizeof(struct ireg_buf) * n;
+		}
+		break;
+	case CUDBG_HMA:
+		value = t4_read_reg(adap, MA_TARGET_MEM_ENABLE_A);
+		if (value & HMA_MUX_F) {
+			/* In T6, there's no MC1.  So, HMA shares MC1
+			 * address space.
+			 */
+			value = t4_read_reg(adap, MA_EXT_MEMORY1_BAR_A);
+			len = EXT_MEM1_SIZE_G(value);
+		}
+		len = cudbg_mbytes_to_bytes(len);
+		break;
+	case CUDBG_QDESC:
+		cudbg_fill_qdesc_num_and_size(adap, NULL, &len);
+		break;
+	default:
+		break;
+	}
+
+	return len;
+}
+
+static int cudbg_do_compression(struct cudbg_init *pdbg_init,
+				struct cudbg_buffer *pin_buff,
+				struct cudbg_buffer *dbg_buff)
+{
+	struct cudbg_buffer temp_in_buff = { 0 };
+	int bytes_left, bytes_read, bytes;
+	u32 offset = dbg_buff->offset;
+	int rc;
+
+	temp_in_buff.offset = pin_buff->offset;
+	temp_in_buff.data = pin_buff->data;
+	temp_in_buff.size = pin_buff->size;
+
+	bytes_left = pin_buff->size;
+	bytes_read = 0;
+	while (bytes_left > 0) {
+		/* Do compression in smaller chunks */
+		bytes = min_t(unsigned long, bytes_left,
+			      (unsigned long)CUDBG_CHUNK_SIZE);
+		temp_in_buff.data = (char *)pin_buff->data + bytes_read;
+		temp_in_buff.size = bytes;
+		rc = cudbg_compress_buff(pdbg_init, &temp_in_buff, dbg_buff);
+		if (rc)
+			return rc;
+		bytes_left -= bytes;
+		bytes_read += bytes;
+	}
+
+	pin_buff->size = dbg_buff->offset - offset;
+	return 0;
+}
+
+static int cudbg_write_and_release_buff(struct cudbg_init *pdbg_init,
+					struct cudbg_buffer *pin_buff,
+					struct cudbg_buffer *dbg_buff)
+{
+	int rc = 0;
+
+	if (pdbg_init->compress_type == CUDBG_COMPRESSION_NONE) {
+		cudbg_update_buff(pin_buff, dbg_buff);
+	} else {
+		rc = cudbg_do_compression(pdbg_init, pin_buff, dbg_buff);
+		if (rc)
+			goto out;
+	}
+
+out:
+	cudbg_put_buff(pdbg_init, pin_buff);
+	return rc;
+}
+
+static int is_fw_attached(struct cudbg_init *pdbg_init)
+{
+	struct adapter *padap = pdbg_init->adap;
+
+	if (!(padap->flags & CXGB4_FW_OK) || padap->use_bd)
+		return 0;
+
+	return 1;
+}
+
+/* This function will add additional padding bytes into debug_buffer to make it
+ * 4 byte aligned.
+ */
+void cudbg_align_debug_buffer(struct cudbg_buffer *dbg_buff,
+			      struct cudbg_entity_hdr *entity_hdr)
+{
+	u8 zero_buf[4] = {0};
+	u8 padding, remain;
+
+	remain = (dbg_buff->offset - entity_hdr->start_offset) % 4;
+	padding = 4 - remain;
+	if (remain) {
+		memcpy(((u8 *)dbg_buff->data) + dbg_buff->offset, &zero_buf,
+		       padding);
+		dbg_buff->offset += padding;
+		entity_hdr->num_pad = padding;
+	}
+	entity_hdr->size = dbg_buff->offset - entity_hdr->start_offset;
+}
+
+struct cudbg_entity_hdr *cudbg_get_entity_hdr(void *outbuf, int i)
+{
+	struct cudbg_hdr *cudbg_hdr = (struct cudbg_hdr *)outbuf;
+
+	return (struct cudbg_entity_hdr *)
+	       ((char *)outbuf + cudbg_hdr->hdr_len +
+		(sizeof(struct cudbg_entity_hdr) * (i - 1)));
+}
+
+static int cudbg_read_vpd_reg(struct adapter *padap, u32 addr, u32 len,
+			      void *dest)
+{
+	int vaddr, rc;
+
+	vaddr = t4_eeprom_ptov(addr, padap->pf, EEPROMPFSIZE);
+	if (vaddr < 0)
+		return vaddr;
+
+	rc = pci_read_vpd(padap->pdev, vaddr, len, dest);
+	if (rc < 0)
+		return rc;
+
+	return 0;
+}
+
+static int cudbg_mem_desc_cmp(const void *a, const void *b)
+{
+	return ((const struct cudbg_mem_desc *)a)->base -
+	       ((const struct cudbg_mem_desc *)b)->base;
+}
+
+int cudbg_fill_meminfo(struct adapter *padap,
+		       struct cudbg_meminfo *meminfo_buff)
+{
+	struct cudbg_mem_desc *md;
+	u32 lo, hi, used, alloc;
+	int n, i;
+
+	memset(meminfo_buff->avail, 0,
+	       ARRAY_SIZE(meminfo_buff->avail) *
+	       sizeof(struct cudbg_mem_desc));
+	memset(meminfo_buff->mem, 0,
+	       (ARRAY_SIZE(cudbg_region) + 3) * sizeof(struct cudbg_mem_desc));
+	md  = meminfo_buff->mem;
+
+	for (i = 0; i < ARRAY_SIZE(meminfo_buff->mem); i++) {
+		meminfo_buff->mem[i].limit = 0;
+		meminfo_buff->mem[i].idx = i;
+	}
+
+	/* Find and sort the populated memory ranges */
+	i = 0;
+	lo = t4_read_reg(padap, MA_TARGET_MEM_ENABLE_A);
+	if (lo & EDRAM0_ENABLE_F) {
+		hi = t4_read_reg(padap, MA_EDRAM0_BAR_A);
+		meminfo_buff->avail[i].base =
+			cudbg_mbytes_to_bytes(EDRAM0_BASE_G(hi));
+		meminfo_buff->avail[i].limit =
+			meminfo_buff->avail[i].base +
+			cudbg_mbytes_to_bytes(EDRAM0_SIZE_G(hi));
+		meminfo_buff->avail[i].idx = 0;
+		i++;
+	}
+
+	if (lo & EDRAM1_ENABLE_F) {
+		hi =  t4_read_reg(padap, MA_EDRAM1_BAR_A);
+		meminfo_buff->avail[i].base =
+			cudbg_mbytes_to_bytes(EDRAM1_BASE_G(hi));
+		meminfo_buff->avail[i].limit =
+			meminfo_buff->avail[i].base +
+			cudbg_mbytes_to_bytes(EDRAM1_SIZE_G(hi));
+		meminfo_buff->avail[i].idx = 1;
+		i++;
+	}
+
+	if (is_t5(padap->params.chip)) {
+		if (lo & EXT_MEM0_ENABLE_F) {
+			hi = t4_read_reg(padap, MA_EXT_MEMORY0_BAR_A);
+			meminfo_buff->avail[i].base =
+				cudbg_mbytes_to_bytes(EXT_MEM_BASE_G(hi));
+			meminfo_buff->avail[i].limit =
+				meminfo_buff->avail[i].base +
+				cudbg_mbytes_to_bytes(EXT_MEM_SIZE_G(hi));
+			meminfo_buff->avail[i].idx = 3;
+			i++;
+		}
+
+		if (lo & EXT_MEM1_ENABLE_F) {
+			hi = t4_read_reg(padap, MA_EXT_MEMORY1_BAR_A);
+			meminfo_buff->avail[i].base =
+				cudbg_mbytes_to_bytes(EXT_MEM1_BASE_G(hi));
+			meminfo_buff->avail[i].limit =
+				meminfo_buff->avail[i].base +
+				cudbg_mbytes_to_bytes(EXT_MEM1_SIZE_G(hi));
+			meminfo_buff->avail[i].idx = 4;
+			i++;
+		}
+	} else {
+		if (lo & EXT_MEM_ENABLE_F) {
+			hi = t4_read_reg(padap, MA_EXT_MEMORY_BAR_A);
+			meminfo_buff->avail[i].base =
+				cudbg_mbytes_to_bytes(EXT_MEM_BASE_G(hi));
+			meminfo_buff->avail[i].limit =
+				meminfo_buff->avail[i].base +
+				cudbg_mbytes_to_bytes(EXT_MEM_SIZE_G(hi));
+			meminfo_buff->avail[i].idx = 2;
+			i++;
+		}
+
+		if (lo & HMA_MUX_F) {
+			hi = t4_read_reg(padap, MA_EXT_MEMORY1_BAR_A);
+			meminfo_buff->avail[i].base =
+				cudbg_mbytes_to_bytes(EXT_MEM1_BASE_G(hi));
+			meminfo_buff->avail[i].limit =
+				meminfo_buff->avail[i].base +
+				cudbg_mbytes_to_bytes(EXT_MEM1_SIZE_G(hi));
+			meminfo_buff->avail[i].idx = 5;
+			i++;
+		}
+	}
+
+	if (!i) /* no memory available */
+		return CUDBG_STATUS_ENTITY_NOT_FOUND;
+
+	meminfo_buff->avail_c = i;
+	sort(meminfo_buff->avail, i, sizeof(struct cudbg_mem_desc),
+	     cudbg_mem_desc_cmp, NULL);
+	(md++)->base = t4_read_reg(padap, SGE_DBQ_CTXT_BADDR_A);
+	(md++)->base = t4_read_reg(padap, SGE_IMSG_CTXT_BADDR_A);
+	(md++)->base = t4_read_reg(padap, SGE_FLM_CACHE_BADDR_A);
+	(md++)->base = t4_read_reg(padap, TP_CMM_TCB_BASE_A);
+	(md++)->base = t4_read_reg(padap, TP_CMM_MM_BASE_A);
+	(md++)->base = t4_read_reg(padap, TP_CMM_TIMER_BASE_A);
+	(md++)->base = t4_read_reg(padap, TP_CMM_MM_RX_FLST_BASE_A);
+	(md++)->base = t4_read_reg(padap, TP_CMM_MM_TX_FLST_BASE_A);
+	(md++)->base = t4_read_reg(padap, TP_CMM_MM_PS_FLST_BASE_A);
+
+	/* the next few have explicit upper bounds */
+	md->base = t4_read_reg(padap, TP_PMM_TX_BASE_A);
+	md->limit = md->base - 1 +
+		    t4_read_reg(padap, TP_PMM_TX_PAGE_SIZE_A) *
+		    PMTXMAXPAGE_G(t4_read_reg(padap, TP_PMM_TX_MAX_PAGE_A));
+	md++;
+
+	md->base = t4_read_reg(padap, TP_PMM_RX_BASE_A);
+	md->limit = md->base - 1 +
+		    t4_read_reg(padap, TP_PMM_RX_PAGE_SIZE_A) *
+		    PMRXMAXPAGE_G(t4_read_reg(padap, TP_PMM_RX_MAX_PAGE_A));
+	md++;
+
+	if (t4_read_reg(padap, LE_DB_CONFIG_A) & HASHEN_F) {
+		if (CHELSIO_CHIP_VERSION(padap->params.chip) <= CHELSIO_T5) {
+			hi = t4_read_reg(padap, LE_DB_TID_HASHBASE_A) / 4;
+			md->base = t4_read_reg(padap, LE_DB_HASH_TID_BASE_A);
+		} else {
+			hi = t4_read_reg(padap, LE_DB_HASH_TID_BASE_A);
+			md->base = t4_read_reg(padap,
+					       LE_DB_HASH_TBL_BASE_ADDR_A);
+		}
+		md->limit = 0;
+	} else {
+		md->base = 0;
+		md->idx = ARRAY_SIZE(cudbg_region);  /* hide it */
+	}
+	md++;
+
+#define ulp_region(reg) do { \
+	md->base = t4_read_reg(padap, ULP_ ## reg ## _LLIMIT_A);\
+	(md++)->limit = t4_read_reg(padap, ULP_ ## reg ## _ULIMIT_A);\
+} while (0)
+
+	ulp_region(RX_ISCSI);
+	ulp_region(RX_TDDP);
+	ulp_region(TX_TPT);
+	ulp_region(RX_STAG);
+	ulp_region(RX_RQ);
+	ulp_region(RX_RQUDP);
+	ulp_region(RX_PBL);
+	ulp_region(TX_PBL);
+#undef ulp_region
+	md->base = 0;
+	md->idx = ARRAY_SIZE(cudbg_region);
+	if (!is_t4(padap->params.chip)) {
+		u32 fifo_size = t4_read_reg(padap, SGE_DBVFIFO_SIZE_A);
+		u32 sge_ctrl = t4_read_reg(padap, SGE_CONTROL2_A);
+		u32 size = 0;
+
+		if (is_t5(padap->params.chip)) {
+			if (sge_ctrl & VFIFO_ENABLE_F)
+				size = DBVFIFO_SIZE_G(fifo_size);
+		} else {
+			size = T6_DBVFIFO_SIZE_G(fifo_size);
+		}
+
+		if (size) {
+			md->base = BASEADDR_G(t4_read_reg(padap,
+							  SGE_DBVFIFO_BADDR_A));
+			md->limit = md->base + (size << 2) - 1;
+		}
+	}
+
+	md++;
+
+	md->base = t4_read_reg(padap, ULP_RX_CTX_BASE_A);
+	md->limit = 0;
+	md++;
+	md->base = t4_read_reg(padap, ULP_TX_ERR_TABLE_BASE_A);
+	md->limit = 0;
+	md++;
+
+	md->base = padap->vres.ocq.start;
+	if (padap->vres.ocq.size)
+		md->limit = md->base + padap->vres.ocq.size - 1;
+	else
+		md->idx = ARRAY_SIZE(cudbg_region);  /* hide it */
+	md++;
+
+	/* add any address-space holes, there can be up to 3 */
+	for (n = 0; n < i - 1; n++)
+		if (meminfo_buff->avail[n].limit <
+		    meminfo_buff->avail[n + 1].base)
+			(md++)->base = meminfo_buff->avail[n].limit;
+
+	if (meminfo_buff->avail[n].limit)
+		(md++)->base = meminfo_buff->avail[n].limit;
+
+	n = md - meminfo_buff->mem;
+	meminfo_buff->mem_c = n;
+
+	sort(meminfo_buff->mem, n, sizeof(struct cudbg_mem_desc),
+	     cudbg_mem_desc_cmp, NULL);
+
+	lo = t4_read_reg(padap, CIM_SDRAM_BASE_ADDR_A);
+	hi = t4_read_reg(padap, CIM_SDRAM_ADDR_SIZE_A) + lo - 1;
+	meminfo_buff->up_ram_lo = lo;
+	meminfo_buff->up_ram_hi = hi;
+
+	lo = t4_read_reg(padap, CIM_EXTMEM2_BASE_ADDR_A);
+	hi = t4_read_reg(padap, CIM_EXTMEM2_ADDR_SIZE_A) + lo - 1;
+	meminfo_buff->up_extmem2_lo = lo;
+	meminfo_buff->up_extmem2_hi = hi;
+
+	lo = t4_read_reg(padap, TP_PMM_RX_MAX_PAGE_A);
+	for (i = 0, meminfo_buff->free_rx_cnt = 0; i < 2; i++)
+		meminfo_buff->free_rx_cnt +=
+			FREERXPAGECOUNT_G(t4_read_reg(padap,
+						      TP_FLM_FREE_RX_CNT_A));
+
+	meminfo_buff->rx_pages_data[0] =  PMRXMAXPAGE_G(lo);
+	meminfo_buff->rx_pages_data[1] =
+		t4_read_reg(padap, TP_PMM_RX_PAGE_SIZE_A) >> 10;
+	meminfo_buff->rx_pages_data[2] = (lo & PMRXNUMCHN_F) ? 2 : 1;
+
+	lo = t4_read_reg(padap, TP_PMM_TX_MAX_PAGE_A);
+	hi = t4_read_reg(padap, TP_PMM_TX_PAGE_SIZE_A);
+	for (i = 0, meminfo_buff->free_tx_cnt = 0; i < 4; i++)
+		meminfo_buff->free_tx_cnt +=
+			FREETXPAGECOUNT_G(t4_read_reg(padap,
+						      TP_FLM_FREE_TX_CNT_A));
+
+	meminfo_buff->tx_pages_data[0] = PMTXMAXPAGE_G(lo);
+	meminfo_buff->tx_pages_data[1] =
+		hi >= (1 << 20) ? (hi >> 20) : (hi >> 10);
+	meminfo_buff->tx_pages_data[2] =
+		hi >= (1 << 20) ? 'M' : 'K';
+	meminfo_buff->tx_pages_data[3] = 1 << PMTXNUMCHN_G(lo);
+
+	meminfo_buff->p_structs = t4_read_reg(padap, TP_CMM_MM_MAX_PSTRUCT_A);
+	meminfo_buff->p_structs_free_cnt =
+		FREEPSTRUCTCOUNT_G(t4_read_reg(padap, TP_FLM_FREE_PS_CNT_A));
+
+	for (i = 0; i < 4; i++) {
+		if (CHELSIO_CHIP_VERSION(padap->params.chip) > CHELSIO_T5)
+			lo = t4_read_reg(padap,
+					 MPS_RX_MAC_BG_PG_CNT0_A + i * 4);
+		else
+			lo = t4_read_reg(padap, MPS_RX_PG_RSV0_A + i * 4);
+		if (is_t5(padap->params.chip)) {
+			used = T5_USED_G(lo);
+			alloc = T5_ALLOC_G(lo);
+		} else {
+			used = USED_G(lo);
+			alloc = ALLOC_G(lo);
+		}
+		meminfo_buff->port_used[i] = used;
+		meminfo_buff->port_alloc[i] = alloc;
+	}
+
+	for (i = 0; i < padap->params.arch.nchan; i++) {
+		if (CHELSIO_CHIP_VERSION(padap->params.chip) > CHELSIO_T5)
+			lo = t4_read_reg(padap,
+					 MPS_RX_LPBK_BG_PG_CNT0_A + i * 4);
+		else
+			lo = t4_read_reg(padap, MPS_RX_PG_RSV4_A + i * 4);
+		if (is_t5(padap->params.chip)) {
+			used = T5_USED_G(lo);
+			alloc = T5_ALLOC_G(lo);
+		} else {
+			used = USED_G(lo);
+			alloc = ALLOC_G(lo);
+		}
+		meminfo_buff->loopback_used[i] = used;
+		meminfo_buff->loopback_alloc[i] = alloc;
+	}
+
+	return 0;
+}
+
+int cudbg_collect_reg_dump(struct cudbg_init *pdbg_init,
+			   struct cudbg_buffer *dbg_buff,
+			   struct cudbg_error *cudbg_err)
+{
+	struct adapter *padap = pdbg_init->adap;
+	struct cudbg_buffer temp_buff = { 0 };
+	u32 buf_size = 0;
+	int rc = 0;
+
+	if (is_t4(padap->params.chip))
+		buf_size = T4_REGMAP_SIZE;
+	else if (is_t5(padap->params.chip) || is_t6(padap->params.chip))
+		buf_size = T5_REGMAP_SIZE;
+
+	rc = cudbg_get_buff(pdbg_init, dbg_buff, buf_size, &temp_buff);
+	if (rc)
+		return rc;
+	t4_get_regs(padap, (void *)temp_buff.data, temp_buff.size);
+	return cudbg_write_and_release_buff(pdbg_init, &temp_buff, dbg_buff);
+}
+
+int cudbg_collect_fw_devlog(struct cudbg_init *pdbg_init,
+			    struct cudbg_buffer *dbg_buff,
+			    struct cudbg_error *cudbg_err)
+{
+	struct adapter *padap = pdbg_init->adap;
+	struct cudbg_buffer temp_buff = { 0 };
+	struct devlog_params *dparams;
+	int rc = 0;
+
+	rc = t4_init_devlog_params(padap);
+	if (rc < 0) {
+		cudbg_err->sys_err = rc;
+		return rc;
+	}
+
+	dparams = &padap->params.devlog;
+	rc = cudbg_get_buff(pdbg_init, dbg_buff, dparams->size, &temp_buff);
+	if (rc)
+		return rc;
+
+	/* Collect FW devlog */
+	if (dparams->start != 0) {
+		spin_lock(&padap->win0_lock);
+		rc = t4_memory_rw(padap, padap->params.drv_memwin,
+				  dparams->memtype, dparams->start,
+				  dparams->size,
+				  (__be32 *)(char *)temp_buff.data,
+				  1);
+		spin_unlock(&padap->win0_lock);
+		if (rc) {
+			cudbg_err->sys_err = rc;
+			cudbg_put_buff(pdbg_init, &temp_buff);
+			return rc;
+		}
+	}
+	return cudbg_write_and_release_buff(pdbg_init, &temp_buff, dbg_buff);
+}
+
+int cudbg_collect_cim_la(struct cudbg_init *pdbg_init,
+			 struct cudbg_buffer *dbg_buff,
+			 struct cudbg_error *cudbg_err)
+{
+	struct adapter *padap = pdbg_init->adap;
+	struct cudbg_buffer temp_buff = { 0 };
+	int size, rc;
+	u32 cfg = 0;
+
+	if (is_t6(padap->params.chip)) {
+		size = padap->params.cim_la_size / 10 + 1;
+		size *= 10 * sizeof(u32);
+	} else {
+		size = padap->params.cim_la_size / 8;
+		size *= 8 * sizeof(u32);
+	}
+
+	size += sizeof(cfg);
+	rc = cudbg_get_buff(pdbg_init, dbg_buff, size, &temp_buff);
+	if (rc)
+		return rc;
+
+	rc = t4_cim_read(padap, UP_UP_DBG_LA_CFG_A, 1, &cfg);
+	if (rc) {
+		cudbg_err->sys_err = rc;
+		cudbg_put_buff(pdbg_init, &temp_buff);
+		return rc;
+	}
+
+	memcpy((char *)temp_buff.data, &cfg, sizeof(cfg));
+	rc = t4_cim_read_la(padap,
+			    (u32 *)((char *)temp_buff.data + sizeof(cfg)),
+			    NULL);
+	if (rc < 0) {
+		cudbg_err->sys_err = rc;
+		cudbg_put_buff(pdbg_init, &temp_buff);
+		return rc;
+	}
+	return cudbg_write_and_release_buff(pdbg_init, &temp_buff, dbg_buff);
+}
+
+int cudbg_collect_cim_ma_la(struct cudbg_init *pdbg_init,
+			    struct cudbg_buffer *dbg_buff,
+			    struct cudbg_error *cudbg_err)
+{
+	struct adapter *padap = pdbg_init->adap;
+	struct cudbg_buffer temp_buff = { 0 };
+	int size, rc;
+
+	size = 2 * CIM_MALA_SIZE * 5 * sizeof(u32);
+	rc = cudbg_get_buff(pdbg_init, dbg_buff, size, &temp_buff);
+	if (rc)
+		return rc;
+
+	t4_cim_read_ma_la(padap,
+			  (u32 *)temp_buff.data,
+			  (u32 *)((char *)temp_buff.data +
+				  5 * CIM_MALA_SIZE));
+	return cudbg_write_and_release_buff(pdbg_init, &temp_buff, dbg_buff);
+}
+
+int cudbg_collect_cim_qcfg(struct cudbg_init *pdbg_init,
+			   struct cudbg_buffer *dbg_buff,
+			   struct cudbg_error *cudbg_err)
+{
+	struct adapter *padap = pdbg_init->adap;
+	struct cudbg_buffer temp_buff = { 0 };
+	struct cudbg_cim_qcfg *cim_qcfg_data;
+	int rc;
+
+	rc = cudbg_get_buff(pdbg_init, dbg_buff, sizeof(struct cudbg_cim_qcfg),
+			    &temp_buff);
+	if (rc)
+		return rc;
+
+	cim_qcfg_data = (struct cudbg_cim_qcfg *)temp_buff.data;
+	cim_qcfg_data->chip = padap->params.chip;
+	rc = t4_cim_read(padap, UP_IBQ_0_RDADDR_A,
+			 ARRAY_SIZE(cim_qcfg_data->stat), cim_qcfg_data->stat);
+	if (rc) {
+		cudbg_err->sys_err = rc;
+		cudbg_put_buff(pdbg_init, &temp_buff);
+		return rc;
+	}
+
+	rc = t4_cim_read(padap, UP_OBQ_0_REALADDR_A,
+			 ARRAY_SIZE(cim_qcfg_data->obq_wr),
+			 cim_qcfg_data->obq_wr);
+	if (rc) {
+		cudbg_err->sys_err = rc;
+		cudbg_put_buff(pdbg_init, &temp_buff);
+		return rc;
+	}
+
+	t4_read_cimq_cfg(padap, cim_qcfg_data->base, cim_qcfg_data->size,
+			 cim_qcfg_data->thres);
+	return cudbg_write_and_release_buff(pdbg_init, &temp_buff, dbg_buff);
+}
+
+static int cudbg_read_cim_ibq(struct cudbg_init *pdbg_init,
+			      struct cudbg_buffer *dbg_buff,
+			      struct cudbg_error *cudbg_err, int qid)
+{
+	struct adapter *padap = pdbg_init->adap;
+	struct cudbg_buffer temp_buff = { 0 };
+	int no_of_read_words, rc = 0;
+	u32 qsize;
+
+	/* collect CIM IBQ */
+	qsize = CIM_IBQ_SIZE * 4 * sizeof(u32);
+	rc = cudbg_get_buff(pdbg_init, dbg_buff, qsize, &temp_buff);
+	if (rc)
+		return rc;
+
+	/* t4_read_cim_ibq will return no. of read words or error */
+	no_of_read_words = t4_read_cim_ibq(padap, qid,
+					   (u32 *)temp_buff.data, qsize);
+	/* no_of_read_words is less than or equal to 0 means error */
+	if (no_of_read_words <= 0) {
+		if (!no_of_read_words)
+			rc = CUDBG_SYSTEM_ERROR;
+		else
+			rc = no_of_read_words;
+		cudbg_err->sys_err = rc;
+		cudbg_put_buff(pdbg_init, &temp_buff);
+		return rc;
+	}
+	return cudbg_write_and_release_buff(pdbg_init, &temp_buff, dbg_buff);
+}
+
+int cudbg_collect_cim_ibq_tp0(struct cudbg_init *pdbg_init,
+			      struct cudbg_buffer *dbg_buff,
+			      struct cudbg_error *cudbg_err)
+{
+	return cudbg_read_cim_ibq(pdbg_init, dbg_buff, cudbg_err, 0);
+}
+
+int cudbg_collect_cim_ibq_tp1(struct cudbg_init *pdbg_init,
+			      struct cudbg_buffer *dbg_buff,
+			      struct cudbg_error *cudbg_err)
+{
+	return cudbg_read_cim_ibq(pdbg_init, dbg_buff, cudbg_err, 1);
+}
+
+int cudbg_collect_cim_ibq_ulp(struct cudbg_init *pdbg_init,
+			      struct cudbg_buffer *dbg_buff,
+			      struct cudbg_error *cudbg_err)
+{
+	return cudbg_read_cim_ibq(pdbg_init, dbg_buff, cudbg_err, 2);
+}
+
+int cudbg_collect_cim_ibq_sge0(struct cudbg_init *pdbg_init,
+			       struct cudbg_buffer *dbg_buff,
+			       struct cudbg_error *cudbg_err)
+{
+	return cudbg_read_cim_ibq(pdbg_init, dbg_buff, cudbg_err, 3);
+}
+
+int cudbg_collect_cim_ibq_sge1(struct cudbg_init *pdbg_init,
+			       struct cudbg_buffer *dbg_buff,
+			       struct cudbg_error *cudbg_err)
+{
+	return cudbg_read_cim_ibq(pdbg_init, dbg_buff, cudbg_err, 4);
+}
+
+int cudbg_collect_cim_ibq_ncsi(struct cudbg_init *pdbg_init,
+			       struct cudbg_buffer *dbg_buff,
+			       struct cudbg_error *cudbg_err)
+{
+	return cudbg_read_cim_ibq(pdbg_init, dbg_buff, cudbg_err, 5);
+}
+
+u32 cudbg_cim_obq_size(struct adapter *padap, int qid)
+{
+	u32 value;
+
+	t4_write_reg(padap, CIM_QUEUE_CONFIG_REF_A, OBQSELECT_F |
+		     QUENUMSELECT_V(qid));
+	value = t4_read_reg(padap, CIM_QUEUE_CONFIG_CTRL_A);
+	value = CIMQSIZE_G(value) * 64; /* size in number of words */
+	return value * sizeof(u32);
+}
+
+static int cudbg_read_cim_obq(struct cudbg_init *pdbg_init,
+			      struct cudbg_buffer *dbg_buff,
+			      struct cudbg_error *cudbg_err, int qid)
+{
+	struct adapter *padap = pdbg_init->adap;
+	struct cudbg_buffer temp_buff = { 0 };
+	int no_of_read_words, rc = 0;
+	u32 qsize;
+
+	/* collect CIM OBQ */
+	qsize =  cudbg_cim_obq_size(padap, qid);
+	rc = cudbg_get_buff(pdbg_init, dbg_buff, qsize, &temp_buff);
+	if (rc)
+		return rc;
+
+	/* t4_read_cim_obq will return no. of read words or error */
+	no_of_read_words = t4_read_cim_obq(padap, qid,
+					   (u32 *)temp_buff.data, qsize);
+	/* no_of_read_words is less than or equal to 0 means error */
+	if (no_of_read_words <= 0) {
+		if (!no_of_read_words)
+			rc = CUDBG_SYSTEM_ERROR;
+		else
+			rc = no_of_read_words;
+		cudbg_err->sys_err = rc;
+		cudbg_put_buff(pdbg_init, &temp_buff);
+		return rc;
+	}
+	return cudbg_write_and_release_buff(pdbg_init, &temp_buff, dbg_buff);
+}
+
+int cudbg_collect_cim_obq_ulp0(struct cudbg_init *pdbg_init,
+			       struct cudbg_buffer *dbg_buff,
+			       struct cudbg_error *cudbg_err)
+{
+	return cudbg_read_cim_obq(pdbg_init, dbg_buff, cudbg_err, 0);
+}
+
+int cudbg_collect_cim_obq_ulp1(struct cudbg_init *pdbg_init,
+			       struct cudbg_buffer *dbg_buff,
+			       struct cudbg_error *cudbg_err)
+{
+	return cudbg_read_cim_obq(pdbg_init, dbg_buff, cudbg_err, 1);
+}
+
+int cudbg_collect_cim_obq_ulp2(struct cudbg_init *pdbg_init,
+			       struct cudbg_buffer *dbg_buff,
+			       struct cudbg_error *cudbg_err)
+{
+	return cudbg_read_cim_obq(pdbg_init, dbg_buff, cudbg_err, 2);
+}
+
+int cudbg_collect_cim_obq_ulp3(struct cudbg_init *pdbg_init,
+			       struct cudbg_buffer *dbg_buff,
+			       struct cudbg_error *cudbg_err)
+{
+	return cudbg_read_cim_obq(pdbg_init, dbg_buff, cudbg_err, 3);
+}
+
+int cudbg_collect_cim_obq_sge(struct cudbg_init *pdbg_init,
+			      struct cudbg_buffer *dbg_buff,
+			      struct cudbg_error *cudbg_err)
+{
+	return cudbg_read_cim_obq(pdbg_init, dbg_buff, cudbg_err, 4);
+}
+
+int cudbg_collect_cim_obq_ncsi(struct cudbg_init *pdbg_init,
+			       struct cudbg_buffer *dbg_buff,
+			       struct cudbg_error *cudbg_err)
+{
+	return cudbg_read_cim_obq(pdbg_init, dbg_buff, cudbg_err, 5);
+}
+
+int cudbg_collect_obq_sge_rx_q0(struct cudbg_init *pdbg_init,
+				struct cudbg_buffer *dbg_buff,
+				struct cudbg_error *cudbg_err)
+{
+	return cudbg_read_cim_obq(pdbg_init, dbg_buff, cudbg_err, 6);
+}
+
+int cudbg_collect_obq_sge_rx_q1(struct cudbg_init *pdbg_init,
+				struct cudbg_buffer *dbg_buff,
+				struct cudbg_error *cudbg_err)
+{
+	return cudbg_read_cim_obq(pdbg_init, dbg_buff, cudbg_err, 7);
+}
+
+static int cudbg_meminfo_get_mem_index(struct adapter *padap,
+				       struct cudbg_meminfo *mem_info,
+				       u8 mem_type, u8 *idx)
+{
+	u8 i, flag;
+
+	switch (mem_type) {
+	case MEM_EDC0:
+		flag = EDC0_FLAG;
+		break;
+	case MEM_EDC1:
+		flag = EDC1_FLAG;
+		break;
+	case MEM_MC0:
+		/* Some T5 cards have both MC0 and MC1. */
+		flag = is_t5(padap->params.chip) ? MC0_FLAG : MC_FLAG;
+		break;
+	case MEM_MC1:
+		flag = MC1_FLAG;
+		break;
+	case MEM_HMA:
+		flag = HMA_FLAG;
+		break;
+	default:
+		return CUDBG_STATUS_ENTITY_NOT_FOUND;
+	}
+
+	for (i = 0; i < mem_info->avail_c; i++) {
+		if (mem_info->avail[i].idx == flag) {
+			*idx = i;
+			return 0;
+		}
+	}
+
+	return CUDBG_STATUS_ENTITY_NOT_FOUND;
+}
+
+/* Fetch the @region_name's start and end from @meminfo. */
+static int cudbg_get_mem_region(struct adapter *padap,
+				struct cudbg_meminfo *meminfo,
+				u8 mem_type, const char *region_name,
+				struct cudbg_mem_desc *mem_desc)
+{
+	u8 mc, found = 0;
+	u32 idx = 0;
+	int rc, i;
+
+	rc = cudbg_meminfo_get_mem_index(padap, meminfo, mem_type, &mc);
+	if (rc)
+		return rc;
+
+	i = match_string(cudbg_region, ARRAY_SIZE(cudbg_region), region_name);
+	if (i < 0)
+		return -EINVAL;
+
+	idx = i;
+	for (i = 0; i < meminfo->mem_c; i++) {
+		if (meminfo->mem[i].idx >= ARRAY_SIZE(cudbg_region))
+			continue; /* Skip holes */
+
+		if (!(meminfo->mem[i].limit))
+			meminfo->mem[i].limit =
+				i < meminfo->mem_c - 1 ?
+				meminfo->mem[i + 1].base - 1 : ~0;
+
+		if (meminfo->mem[i].idx == idx) {
+			/* Check if the region exists in @mem_type memory */
+			if (meminfo->mem[i].base < meminfo->avail[mc].base &&
+			    meminfo->mem[i].limit < meminfo->avail[mc].base)
+				return -EINVAL;
+
+			if (meminfo->mem[i].base > meminfo->avail[mc].limit)
+				return -EINVAL;
+
+			memcpy(mem_desc, &meminfo->mem[i],
+			       sizeof(struct cudbg_mem_desc));
+			found = 1;
+			break;
+		}
+	}
+	if (!found)
+		return -EINVAL;
+
+	return 0;
+}
+
+/* Fetch and update the start and end of the requested memory region w.r.t 0
+ * in the corresponding EDC/MC/HMA.
+ */
+static int cudbg_get_mem_relative(struct adapter *padap,
+				  struct cudbg_meminfo *meminfo,
+				  u8 mem_type, u32 *out_base, u32 *out_end)
+{
+	u8 mc_idx;
+	int rc;
+
+	rc = cudbg_meminfo_get_mem_index(padap, meminfo, mem_type, &mc_idx);
+	if (rc)
+		return rc;
+
+	if (*out_base < meminfo->avail[mc_idx].base)
+		*out_base = 0;
+	else
+		*out_base -= meminfo->avail[mc_idx].base;
+
+	if (*out_end > meminfo->avail[mc_idx].limit)
+		*out_end = meminfo->avail[mc_idx].limit;
+	else
+		*out_end -= meminfo->avail[mc_idx].base;
+
+	return 0;
+}
+
+/* Get TX and RX Payload region */
+static int cudbg_get_payload_range(struct adapter *padap, u8 mem_type,
+				   const char *region_name,
+				   struct cudbg_region_info *payload)
+{
+	struct cudbg_mem_desc mem_desc = { 0 };
+	struct cudbg_meminfo meminfo;
+	int rc;
+
+	rc = cudbg_fill_meminfo(padap, &meminfo);
+	if (rc)
+		return rc;
+
+	rc = cudbg_get_mem_region(padap, &meminfo, mem_type, region_name,
+				  &mem_desc);
+	if (rc) {
+		payload->exist = false;
+		return 0;
+	}
+
+	payload->exist = true;
+	payload->start = mem_desc.base;
+	payload->end = mem_desc.limit;
+
+	return cudbg_get_mem_relative(padap, &meminfo, mem_type,
+				      &payload->start, &payload->end);
+}
+
+static int cudbg_memory_read(struct cudbg_init *pdbg_init, int win,
+			     int mtype, u32 addr, u32 len, void *hbuf)
+{
+	u32 win_pf, memoffset, mem_aperture, mem_base;
+	struct adapter *adap = pdbg_init->adap;
+	u32 pos, offset, resid;
+	u32 *res_buf;
+	u64 *buf;
+	int ret;
+
+	/* Argument sanity checks ...
+	 */
+	if (addr & 0x3 || (uintptr_t)hbuf & 0x3)
+		return -EINVAL;
+
+	buf = (u64 *)hbuf;
+
+	/* Try to do 64-bit reads.  Residual will be handled later. */
+	resid = len & 0x7;
+	len -= resid;
+
+	ret = t4_memory_rw_init(adap, win, mtype, &memoffset, &mem_base,
+				&mem_aperture);
+	if (ret)
+		return ret;
+
+	addr = addr + memoffset;
+	win_pf = is_t4(adap->params.chip) ? 0 : PFNUM_V(adap->pf);
+
+	pos = addr & ~(mem_aperture - 1);
+	offset = addr - pos;
+
+	/* Set up initial PCI-E Memory Window to cover the start of our
+	 * transfer.
+	 */
+	t4_memory_update_win(adap, win, pos | win_pf);
+
+	/* Transfer data from the adapter */
+	while (len > 0) {
+		*buf++ = le64_to_cpu((__force __le64)
+				     t4_read_reg64(adap, mem_base + offset));
+		offset += sizeof(u64);
+		len -= sizeof(u64);
+
+		/* If we've reached the end of our current window aperture,
+		 * move the PCI-E Memory Window on to the next.
+		 */
+		if (offset == mem_aperture) {
+			pos += mem_aperture;
+			offset = 0;
+			t4_memory_update_win(adap, win, pos | win_pf);
+		}
+	}
+
+	res_buf = (u32 *)buf;
+	/* Read residual in 32-bit multiples */
+	while (resid > sizeof(u32)) {
+		*res_buf++ = le32_to_cpu((__force __le32)
+					 t4_read_reg(adap, mem_base + offset));
+		offset += sizeof(u32);
+		resid -= sizeof(u32);
+
+		/* If we've reached the end of our current window aperture,
+		 * move the PCI-E Memory Window on to the next.
+		 */
+		if (offset == mem_aperture) {
+			pos += mem_aperture;
+			offset = 0;
+			t4_memory_update_win(adap, win, pos | win_pf);
+		}
+	}
+
+	/* Transfer residual < 32-bits */
+	if (resid)
+		t4_memory_rw_residual(adap, resid, mem_base + offset,
+				      (u8 *)res_buf, T4_MEMORY_READ);
+
+	return 0;
+}
+
+#define CUDBG_YIELD_ITERATION 256
+
+static int cudbg_read_fw_mem(struct cudbg_init *pdbg_init,
+			     struct cudbg_buffer *dbg_buff, u8 mem_type,
+			     unsigned long tot_len,
+			     struct cudbg_error *cudbg_err)
+{
+	static const char * const region_name[] = { "Tx payload:",
+						    "Rx payload:" };
+	unsigned long bytes, bytes_left, bytes_read = 0;
+	struct adapter *padap = pdbg_init->adap;
+	struct cudbg_buffer temp_buff = { 0 };
+	struct cudbg_region_info payload[2];
+	u32 yield_count = 0;
+	int rc = 0;
+	u8 i;
+
+	/* Get TX/RX Payload region range if they exist */
+	memset(payload, 0, sizeof(payload));
+	for (i = 0; i < ARRAY_SIZE(region_name); i++) {
+		rc = cudbg_get_payload_range(padap, mem_type, region_name[i],
+					     &payload[i]);
+		if (rc)
+			return rc;
+
+		if (payload[i].exist) {
+			/* Align start and end to avoid wrap around */
+			payload[i].start = roundup(payload[i].start,
+						   CUDBG_CHUNK_SIZE);
+			payload[i].end = rounddown(payload[i].end,
+						   CUDBG_CHUNK_SIZE);
+		}
+	}
+
+	bytes_left = tot_len;
+	while (bytes_left > 0) {
+		/* As MC size is huge and read through PIO access, this
+		 * loop will hold cpu for a longer time. OS may think that
+		 * the process is hanged and will generate CPU stall traces.
+		 * So yield the cpu regularly.
+		 */
+		yield_count++;
+		if (!(yield_count % CUDBG_YIELD_ITERATION))
+			schedule();
+
+		bytes = min_t(unsigned long, bytes_left,
+			      (unsigned long)CUDBG_CHUNK_SIZE);
+		rc = cudbg_get_buff(pdbg_init, dbg_buff, bytes, &temp_buff);
+		if (rc)
+			return rc;
+
+		for (i = 0; i < ARRAY_SIZE(payload); i++)
+			if (payload[i].exist &&
+			    bytes_read >= payload[i].start &&
+			    bytes_read + bytes <= payload[i].end)
+				/* TX and RX Payload regions can't overlap */
+				goto skip_read;
+
+		spin_lock(&padap->win0_lock);
+		rc = cudbg_memory_read(pdbg_init, MEMWIN_NIC, mem_type,
+				       bytes_read, bytes, temp_buff.data);
+		spin_unlock(&padap->win0_lock);
+		if (rc) {
+			cudbg_err->sys_err = rc;
+			cudbg_put_buff(pdbg_init, &temp_buff);
+			return rc;
+		}
+
+skip_read:
+		bytes_left -= bytes;
+		bytes_read += bytes;
+		rc = cudbg_write_and_release_buff(pdbg_init, &temp_buff,
+						  dbg_buff);
+		if (rc) {
+			cudbg_put_buff(pdbg_init, &temp_buff);
+			return rc;
+		}
+	}
+	return rc;
+}
+
+static void cudbg_t4_fwcache(struct cudbg_init *pdbg_init,
+			     struct cudbg_error *cudbg_err)
+{
+	struct adapter *padap = pdbg_init->adap;
+	int rc;
+
+	if (is_fw_attached(pdbg_init)) {
+		/* Flush uP dcache before reading edcX/mcX  */
+		rc = t4_fwcache(padap, FW_PARAM_DEV_FWCACHE_FLUSH);
+		if (rc)
+			cudbg_err->sys_warn = rc;
+	}
+}
+
+static int cudbg_mem_region_size(struct cudbg_init *pdbg_init,
+				 struct cudbg_error *cudbg_err,
+				 u8 mem_type, unsigned long *region_size)
+{
+	struct adapter *padap = pdbg_init->adap;
+	struct cudbg_meminfo mem_info;
+	u8 mc_idx;
+	int rc;
+
+	memset(&mem_info, 0, sizeof(struct cudbg_meminfo));
+	rc = cudbg_fill_meminfo(padap, &mem_info);
+	if (rc) {
+		cudbg_err->sys_err = rc;
+		return rc;
+	}
+
+	cudbg_t4_fwcache(pdbg_init, cudbg_err);
+	rc = cudbg_meminfo_get_mem_index(padap, &mem_info, mem_type, &mc_idx);
+	if (rc) {
+		cudbg_err->sys_err = rc;
+		return rc;
+	}
+
+	if (region_size)
+		*region_size = mem_info.avail[mc_idx].limit -
+			       mem_info.avail[mc_idx].base;
+
+	return 0;
+}
+
+static int cudbg_collect_mem_region(struct cudbg_init *pdbg_init,
+				    struct cudbg_buffer *dbg_buff,
+				    struct cudbg_error *cudbg_err,
+				    u8 mem_type)
+{
+	unsigned long size = 0;
+	int rc;
+
+	rc = cudbg_mem_region_size(pdbg_init, cudbg_err, mem_type, &size);
+	if (rc)
+		return rc;
+
+	return cudbg_read_fw_mem(pdbg_init, dbg_buff, mem_type, size,
+				 cudbg_err);
+}
+
+int cudbg_collect_edc0_meminfo(struct cudbg_init *pdbg_init,
+			       struct cudbg_buffer *dbg_buff,
+			       struct cudbg_error *cudbg_err)
+{
+	return cudbg_collect_mem_region(pdbg_init, dbg_buff, cudbg_err,
+					MEM_EDC0);
+}
+
+int cudbg_collect_edc1_meminfo(struct cudbg_init *pdbg_init,
+			       struct cudbg_buffer *dbg_buff,
+			       struct cudbg_error *cudbg_err)
+{
+	return cudbg_collect_mem_region(pdbg_init, dbg_buff, cudbg_err,
+					MEM_EDC1);
+}
+
+int cudbg_collect_mc0_meminfo(struct cudbg_init *pdbg_init,
+			      struct cudbg_buffer *dbg_buff,
+			      struct cudbg_error *cudbg_err)
+{
+	return cudbg_collect_mem_region(pdbg_init, dbg_buff, cudbg_err,
+					MEM_MC0);
+}
+
+int cudbg_collect_mc1_meminfo(struct cudbg_init *pdbg_init,
+			      struct cudbg_buffer *dbg_buff,
+			      struct cudbg_error *cudbg_err)
+{
+	return cudbg_collect_mem_region(pdbg_init, dbg_buff, cudbg_err,
+					MEM_MC1);
+}
+
+int cudbg_collect_hma_meminfo(struct cudbg_init *pdbg_init,
+			      struct cudbg_buffer *dbg_buff,
+			      struct cudbg_error *cudbg_err)
+{
+	return cudbg_collect_mem_region(pdbg_init, dbg_buff, cudbg_err,
+					MEM_HMA);
+}
+
+int cudbg_collect_rss(struct cudbg_init *pdbg_init,
+		      struct cudbg_buffer *dbg_buff,
+		      struct cudbg_error *cudbg_err)
+{
+	struct adapter *padap = pdbg_init->adap;
+	struct cudbg_buffer temp_buff = { 0 };
+	int rc, nentries;
+
+	nentries = t4_chip_rss_size(padap);
+	rc = cudbg_get_buff(pdbg_init, dbg_buff, nentries * sizeof(u16),
+			    &temp_buff);
+	if (rc)
+		return rc;
+
+	rc = t4_read_rss(padap, (u16 *)temp_buff.data);
+	if (rc) {
+		cudbg_err->sys_err = rc;
+		cudbg_put_buff(pdbg_init, &temp_buff);
+		return rc;
+	}
+	return cudbg_write_and_release_buff(pdbg_init, &temp_buff, dbg_buff);
+}
+
+int cudbg_collect_rss_vf_config(struct cudbg_init *pdbg_init,
+				struct cudbg_buffer *dbg_buff,
+				struct cudbg_error *cudbg_err)
+{
+	struct adapter *padap = pdbg_init->adap;
+	struct cudbg_buffer temp_buff = { 0 };
+	struct cudbg_rss_vf_conf *vfconf;
+	int vf, rc, vf_count;
+
+	vf_count = padap->params.arch.vfcount;
+	rc = cudbg_get_buff(pdbg_init, dbg_buff,
+			    vf_count * sizeof(struct cudbg_rss_vf_conf),
+			    &temp_buff);
+	if (rc)
+		return rc;
+
+	vfconf = (struct cudbg_rss_vf_conf *)temp_buff.data;
+	for (vf = 0; vf < vf_count; vf++)
+		t4_read_rss_vf_config(padap, vf, &vfconf[vf].rss_vf_vfl,
+				      &vfconf[vf].rss_vf_vfh, true);
+	return cudbg_write_and_release_buff(pdbg_init, &temp_buff, dbg_buff);
+}
+
+int cudbg_collect_path_mtu(struct cudbg_init *pdbg_init,
+			   struct cudbg_buffer *dbg_buff,
+			   struct cudbg_error *cudbg_err)
+{
+	struct adapter *padap = pdbg_init->adap;
+	struct cudbg_buffer temp_buff = { 0 };
+	int rc;
+
+	rc = cudbg_get_buff(pdbg_init, dbg_buff, NMTUS * sizeof(u16),
+			    &temp_buff);
+	if (rc)
+		return rc;
+
+	t4_read_mtu_tbl(padap, (u16 *)temp_buff.data, NULL);
+	return cudbg_write_and_release_buff(pdbg_init, &temp_buff, dbg_buff);
+}
+
+int cudbg_collect_pm_stats(struct cudbg_init *pdbg_init,
+			   struct cudbg_buffer *dbg_buff,
+			   struct cudbg_error *cudbg_err)
+{
+	struct adapter *padap = pdbg_init->adap;
+	struct cudbg_buffer temp_buff = { 0 };
+	struct cudbg_pm_stats *pm_stats_buff;
+	int rc;
+
+	rc = cudbg_get_buff(pdbg_init, dbg_buff, sizeof(struct cudbg_pm_stats),
+			    &temp_buff);
+	if (rc)
+		return rc;
+
+	pm_stats_buff = (struct cudbg_pm_stats *)temp_buff.data;
+	t4_pmtx_get_stats(padap, pm_stats_buff->tx_cnt, pm_stats_buff->tx_cyc);
+	t4_pmrx_get_stats(padap, pm_stats_buff->rx_cnt, pm_stats_buff->rx_cyc);
+	return cudbg_write_and_release_buff(pdbg_init, &temp_buff, dbg_buff);
+}
+
+int cudbg_collect_hw_sched(struct cudbg_init *pdbg_init,
+			   struct cudbg_buffer *dbg_buff,
+			   struct cudbg_error *cudbg_err)
+{
+	struct adapter *padap = pdbg_init->adap;
+	struct cudbg_buffer temp_buff = { 0 };
+	struct cudbg_hw_sched *hw_sched_buff;
+	int i, rc = 0;
+
+	if (!padap->params.vpd.cclk)
+		return CUDBG_STATUS_CCLK_NOT_DEFINED;
+
+	rc = cudbg_get_buff(pdbg_init, dbg_buff, sizeof(struct cudbg_hw_sched),
+			    &temp_buff);
+
+	if (rc)
+		return rc;
+
+	hw_sched_buff = (struct cudbg_hw_sched *)temp_buff.data;
+	hw_sched_buff->map = t4_read_reg(padap, TP_TX_MOD_QUEUE_REQ_MAP_A);
+	hw_sched_buff->mode = TIMERMODE_G(t4_read_reg(padap, TP_MOD_CONFIG_A));
+	t4_read_pace_tbl(padap, hw_sched_buff->pace_tab);
+	for (i = 0; i < NTX_SCHED; ++i)
+		t4_get_tx_sched(padap, i, &hw_sched_buff->kbps[i],
+				&hw_sched_buff->ipg[i], true);
+	return cudbg_write_and_release_buff(pdbg_init, &temp_buff, dbg_buff);
+}
+
+int cudbg_collect_tp_indirect(struct cudbg_init *pdbg_init,
+			      struct cudbg_buffer *dbg_buff,
+			      struct cudbg_error *cudbg_err)
+{
+	struct adapter *padap = pdbg_init->adap;
+	struct cudbg_buffer temp_buff = { 0 };
+	struct ireg_buf *ch_tp_pio;
+	int i, rc, n = 0;
+	u32 size;
+
+	if (is_t5(padap->params.chip))
+		n = sizeof(t5_tp_pio_array) +
+		    sizeof(t5_tp_tm_pio_array) +
+		    sizeof(t5_tp_mib_index_array);
+	else
+		n = sizeof(t6_tp_pio_array) +
+		    sizeof(t6_tp_tm_pio_array) +
+		    sizeof(t6_tp_mib_index_array);
+
+	n = n / (IREG_NUM_ELEM * sizeof(u32));
+	size = sizeof(struct ireg_buf) * n;
+	rc = cudbg_get_buff(pdbg_init, dbg_buff, size, &temp_buff);
+	if (rc)
+		return rc;
+
+	ch_tp_pio = (struct ireg_buf *)temp_buff.data;
+
+	/* TP_PIO */
+	if (is_t5(padap->params.chip))
+		n = sizeof(t5_tp_pio_array) / (IREG_NUM_ELEM * sizeof(u32));
+	else if (is_t6(padap->params.chip))
+		n = sizeof(t6_tp_pio_array) / (IREG_NUM_ELEM * sizeof(u32));
+
+	for (i = 0; i < n; i++) {
+		struct ireg_field *tp_pio = &ch_tp_pio->tp_pio;
+		u32 *buff = ch_tp_pio->outbuf;
+
+		if (is_t5(padap->params.chip)) {
+			tp_pio->ireg_addr = t5_tp_pio_array[i][0];
+			tp_pio->ireg_data = t5_tp_pio_array[i][1];
+			tp_pio->ireg_local_offset = t5_tp_pio_array[i][2];
+			tp_pio->ireg_offset_range = t5_tp_pio_array[i][3];
+		} else if (is_t6(padap->params.chip)) {
+			tp_pio->ireg_addr = t6_tp_pio_array[i][0];
+			tp_pio->ireg_data = t6_tp_pio_array[i][1];
+			tp_pio->ireg_local_offset = t6_tp_pio_array[i][2];
+			tp_pio->ireg_offset_range = t6_tp_pio_array[i][3];
+		}
+		t4_tp_pio_read(padap, buff, tp_pio->ireg_offset_range,
+			       tp_pio->ireg_local_offset, true);
+		ch_tp_pio++;
+	}
+
+	/* TP_TM_PIO */
+	if (is_t5(padap->params.chip))
+		n = sizeof(t5_tp_tm_pio_array) / (IREG_NUM_ELEM * sizeof(u32));
+	else if (is_t6(padap->params.chip))
+		n = sizeof(t6_tp_tm_pio_array) / (IREG_NUM_ELEM * sizeof(u32));
+
+	for (i = 0; i < n; i++) {
+		struct ireg_field *tp_pio = &ch_tp_pio->tp_pio;
+		u32 *buff = ch_tp_pio->outbuf;
+
+		if (is_t5(padap->params.chip)) {
+			tp_pio->ireg_addr = t5_tp_tm_pio_array[i][0];
+			tp_pio->ireg_data = t5_tp_tm_pio_array[i][1];
+			tp_pio->ireg_local_offset = t5_tp_tm_pio_array[i][2];
+			tp_pio->ireg_offset_range = t5_tp_tm_pio_array[i][3];
+		} else if (is_t6(padap->params.chip)) {
+			tp_pio->ireg_addr = t6_tp_tm_pio_array[i][0];
+			tp_pio->ireg_data = t6_tp_tm_pio_array[i][1];
+			tp_pio->ireg_local_offset = t6_tp_tm_pio_array[i][2];
+			tp_pio->ireg_offset_range = t6_tp_tm_pio_array[i][3];
+		}
+		t4_tp_tm_pio_read(padap, buff, tp_pio->ireg_offset_range,
+				  tp_pio->ireg_local_offset, true);
+		ch_tp_pio++;
+	}
+
+	/* TP_MIB_INDEX */
+	if (is_t5(padap->params.chip))
+		n = sizeof(t5_tp_mib_index_array) /
+		    (IREG_NUM_ELEM * sizeof(u32));
+	else if (is_t6(padap->params.chip))
+		n = sizeof(t6_tp_mib_index_array) /
+		    (IREG_NUM_ELEM * sizeof(u32));
+
+	for (i = 0; i < n ; i++) {
+		struct ireg_field *tp_pio = &ch_tp_pio->tp_pio;
+		u32 *buff = ch_tp_pio->outbuf;
+
+		if (is_t5(padap->params.chip)) {
+			tp_pio->ireg_addr = t5_tp_mib_index_array[i][0];
+			tp_pio->ireg_data = t5_tp_mib_index_array[i][1];
+			tp_pio->ireg_local_offset =
+				t5_tp_mib_index_array[i][2];
+			tp_pio->ireg_offset_range =
+				t5_tp_mib_index_array[i][3];
+		} else if (is_t6(padap->params.chip)) {
+			tp_pio->ireg_addr = t6_tp_mib_index_array[i][0];
+			tp_pio->ireg_data = t6_tp_mib_index_array[i][1];
+			tp_pio->ireg_local_offset =
+				t6_tp_mib_index_array[i][2];
+			tp_pio->ireg_offset_range =
+				t6_tp_mib_index_array[i][3];
+		}
+		t4_tp_mib_read(padap, buff, tp_pio->ireg_offset_range,
+			       tp_pio->ireg_local_offset, true);
+		ch_tp_pio++;
+	}
+	return cudbg_write_and_release_buff(pdbg_init, &temp_buff, dbg_buff);
+}
+
+static void cudbg_read_sge_qbase_indirect_reg(struct adapter *padap,
+					      struct sge_qbase_reg_field *qbase,
+					      u32 func, bool is_pf)
+{
+	u32 *buff, i;
+
+	if (is_pf) {
+		buff = qbase->pf_data_value[func];
+	} else {
+		buff = qbase->vf_data_value[func];
+		/* In SGE_QBASE_INDEX,
+		 * Entries 0->7 are PF0->7, Entries 8->263 are VFID0->256.
+		 */
+		func += 8;
+	}
+
+	t4_write_reg(padap, qbase->reg_addr, func);
+	for (i = 0; i < SGE_QBASE_DATA_REG_NUM; i++, buff++)
+		*buff = t4_read_reg(padap, qbase->reg_data[i]);
+}
+
+int cudbg_collect_sge_indirect(struct cudbg_init *pdbg_init,
+			       struct cudbg_buffer *dbg_buff,
+			       struct cudbg_error *cudbg_err)
+{
+	struct adapter *padap = pdbg_init->adap;
+	struct cudbg_buffer temp_buff = { 0 };
+	struct sge_qbase_reg_field *sge_qbase;
+	struct ireg_buf *ch_sge_dbg;
+	u8 padap_running = 0;
+	int i, rc;
+	u32 size;
+
+	/* Accessing SGE_QBASE_MAP[0-3] and SGE_QBASE_INDEX regs can
+	 * lead to SGE missing doorbells under heavy traffic. So, only
+	 * collect them when adapter is idle.
+	 */
+	for_each_port(padap, i) {
+		padap_running = netif_running(padap->port[i]);
+		if (padap_running)
+			break;
+	}
+
+	size = sizeof(*ch_sge_dbg) * 2;
+	if (!padap_running)
+		size += sizeof(*sge_qbase);
+
+	rc = cudbg_get_buff(pdbg_init, dbg_buff, size, &temp_buff);
+	if (rc)
+		return rc;
+
+	ch_sge_dbg = (struct ireg_buf *)temp_buff.data;
+	for (i = 0; i < 2; i++) {
+		struct ireg_field *sge_pio = &ch_sge_dbg->tp_pio;
+		u32 *buff = ch_sge_dbg->outbuf;
+
+		sge_pio->ireg_addr = t5_sge_dbg_index_array[i][0];
+		sge_pio->ireg_data = t5_sge_dbg_index_array[i][1];
+		sge_pio->ireg_local_offset = t5_sge_dbg_index_array[i][2];
+		sge_pio->ireg_offset_range = t5_sge_dbg_index_array[i][3];
+		t4_read_indirect(padap,
+				 sge_pio->ireg_addr,
+				 sge_pio->ireg_data,
+				 buff,
+				 sge_pio->ireg_offset_range,
+				 sge_pio->ireg_local_offset);
+		ch_sge_dbg++;
+	}
+
+	if (CHELSIO_CHIP_VERSION(padap->params.chip) > CHELSIO_T5 &&
+	    !padap_running) {
+		sge_qbase = (struct sge_qbase_reg_field *)ch_sge_dbg;
+		/* 1 addr reg SGE_QBASE_INDEX and 4 data reg
+		 * SGE_QBASE_MAP[0-3]
+		 */
+		sge_qbase->reg_addr = t6_sge_qbase_index_array[0];
+		for (i = 0; i < SGE_QBASE_DATA_REG_NUM; i++)
+			sge_qbase->reg_data[i] =
+				t6_sge_qbase_index_array[i + 1];
+
+		for (i = 0; i <= PCIE_FW_MASTER_M; i++)
+			cudbg_read_sge_qbase_indirect_reg(padap, sge_qbase,
+							  i, true);
+
+		for (i = 0; i < padap->params.arch.vfcount; i++)
+			cudbg_read_sge_qbase_indirect_reg(padap, sge_qbase,
+							  i, false);
+
+		sge_qbase->vfcount = padap->params.arch.vfcount;
+	}
+
+	return cudbg_write_and_release_buff(pdbg_init, &temp_buff, dbg_buff);
+}
+
+int cudbg_collect_ulprx_la(struct cudbg_init *pdbg_init,
+			   struct cudbg_buffer *dbg_buff,
+			   struct cudbg_error *cudbg_err)
+{
+	struct adapter *padap = pdbg_init->adap;
+	struct cudbg_buffer temp_buff = { 0 };
+	struct cudbg_ulprx_la *ulprx_la_buff;
+	int rc;
+
+	rc = cudbg_get_buff(pdbg_init, dbg_buff, sizeof(struct cudbg_ulprx_la),
+			    &temp_buff);
+	if (rc)
+		return rc;
+
+	ulprx_la_buff = (struct cudbg_ulprx_la *)temp_buff.data;
+	t4_ulprx_read_la(padap, (u32 *)ulprx_la_buff->data);
+	ulprx_la_buff->size = ULPRX_LA_SIZE;
+	return cudbg_write_and_release_buff(pdbg_init, &temp_buff, dbg_buff);
+}
+
+int cudbg_collect_tp_la(struct cudbg_init *pdbg_init,
+			struct cudbg_buffer *dbg_buff,
+			struct cudbg_error *cudbg_err)
+{
+	struct adapter *padap = pdbg_init->adap;
+	struct cudbg_buffer temp_buff = { 0 };
+	struct cudbg_tp_la *tp_la_buff;
+	int size, rc;
+
+	size = sizeof(struct cudbg_tp_la) + TPLA_SIZE *  sizeof(u64);
+	rc = cudbg_get_buff(pdbg_init, dbg_buff, size, &temp_buff);
+	if (rc)
+		return rc;
+
+	tp_la_buff = (struct cudbg_tp_la *)temp_buff.data;
+	tp_la_buff->mode = DBGLAMODE_G(t4_read_reg(padap, TP_DBG_LA_CONFIG_A));
+	t4_tp_read_la(padap, (u64 *)tp_la_buff->data, NULL);
+	return cudbg_write_and_release_buff(pdbg_init, &temp_buff, dbg_buff);
+}
+
+int cudbg_collect_meminfo(struct cudbg_init *pdbg_init,
+			  struct cudbg_buffer *dbg_buff,
+			  struct cudbg_error *cudbg_err)
+{
+	struct adapter *padap = pdbg_init->adap;
+	struct cudbg_buffer temp_buff = { 0 };
+	struct cudbg_meminfo *meminfo_buff;
+	struct cudbg_ver_hdr *ver_hdr;
+	int rc;
+
+	rc = cudbg_get_buff(pdbg_init, dbg_buff,
+			    sizeof(struct cudbg_ver_hdr) +
+			    sizeof(struct cudbg_meminfo),
+			    &temp_buff);
+	if (rc)
+		return rc;
+
+	ver_hdr = (struct cudbg_ver_hdr *)temp_buff.data;
+	ver_hdr->signature = CUDBG_ENTITY_SIGNATURE;
+	ver_hdr->revision = CUDBG_MEMINFO_REV;
+	ver_hdr->size = sizeof(struct cudbg_meminfo);
+
+	meminfo_buff = (struct cudbg_meminfo *)(temp_buff.data +
+						sizeof(*ver_hdr));
+	rc = cudbg_fill_meminfo(padap, meminfo_buff);
+	if (rc) {
+		cudbg_err->sys_err = rc;
+		cudbg_put_buff(pdbg_init, &temp_buff);
+		return rc;
+	}
+
+	return cudbg_write_and_release_buff(pdbg_init, &temp_buff, dbg_buff);
+}
+
+int cudbg_collect_cim_pif_la(struct cudbg_init *pdbg_init,
+			     struct cudbg_buffer *dbg_buff,
+			     struct cudbg_error *cudbg_err)
+{
+	struct cudbg_cim_pif_la *cim_pif_la_buff;
+	struct adapter *padap = pdbg_init->adap;
+	struct cudbg_buffer temp_buff = { 0 };
+	int size, rc;
+
+	size = sizeof(struct cudbg_cim_pif_la) +
+	       2 * CIM_PIFLA_SIZE * 6 * sizeof(u32);
+	rc = cudbg_get_buff(pdbg_init, dbg_buff, size, &temp_buff);
+	if (rc)
+		return rc;
+
+	cim_pif_la_buff = (struct cudbg_cim_pif_la *)temp_buff.data;
+	cim_pif_la_buff->size = CIM_PIFLA_SIZE;
+	t4_cim_read_pif_la(padap, (u32 *)cim_pif_la_buff->data,
+			   (u32 *)cim_pif_la_buff->data + 6 * CIM_PIFLA_SIZE,
+			   NULL, NULL);
+	return cudbg_write_and_release_buff(pdbg_init, &temp_buff, dbg_buff);
+}
+
+int cudbg_collect_clk_info(struct cudbg_init *pdbg_init,
+			   struct cudbg_buffer *dbg_buff,
+			   struct cudbg_error *cudbg_err)
+{
+	struct adapter *padap = pdbg_init->adap;
+	struct cudbg_buffer temp_buff = { 0 };
+	struct cudbg_clk_info *clk_info_buff;
+	u64 tp_tick_us;
+	int rc;
+
+	if (!padap->params.vpd.cclk)
+		return CUDBG_STATUS_CCLK_NOT_DEFINED;
+
+	rc = cudbg_get_buff(pdbg_init, dbg_buff, sizeof(struct cudbg_clk_info),
+			    &temp_buff);
+	if (rc)
+		return rc;
+
+	clk_info_buff = (struct cudbg_clk_info *)temp_buff.data;
+	clk_info_buff->cclk_ps = 1000000000 / padap->params.vpd.cclk; /* psec */
+	clk_info_buff->res = t4_read_reg(padap, TP_TIMER_RESOLUTION_A);
+	clk_info_buff->tre = TIMERRESOLUTION_G(clk_info_buff->res);
+	clk_info_buff->dack_re = DELAYEDACKRESOLUTION_G(clk_info_buff->res);
+	tp_tick_us = (clk_info_buff->cclk_ps << clk_info_buff->tre) / 1000000;
+
+	clk_info_buff->dack_timer =
+		(clk_info_buff->cclk_ps << clk_info_buff->dack_re) / 1000000 *
+		t4_read_reg(padap, TP_DACK_TIMER_A);
+	clk_info_buff->retransmit_min =
+		tp_tick_us * t4_read_reg(padap, TP_RXT_MIN_A);
+	clk_info_buff->retransmit_max =
+		tp_tick_us * t4_read_reg(padap, TP_RXT_MAX_A);
+	clk_info_buff->persist_timer_min =
+		tp_tick_us * t4_read_reg(padap, TP_PERS_MIN_A);
+	clk_info_buff->persist_timer_max =
+		tp_tick_us * t4_read_reg(padap, TP_PERS_MAX_A);
+	clk_info_buff->keepalive_idle_timer =
+		tp_tick_us * t4_read_reg(padap, TP_KEEP_IDLE_A);
+	clk_info_buff->keepalive_interval =
+		tp_tick_us * t4_read_reg(padap, TP_KEEP_INTVL_A);
+	clk_info_buff->initial_srtt =
+		tp_tick_us * INITSRTT_G(t4_read_reg(padap, TP_INIT_SRTT_A));
+	clk_info_buff->finwait2_timer =
+		tp_tick_us * t4_read_reg(padap, TP_FINWAIT2_TIMER_A);
+
+	return cudbg_write_and_release_buff(pdbg_init, &temp_buff, dbg_buff);
+}
+
+int cudbg_collect_pcie_indirect(struct cudbg_init *pdbg_init,
+				struct cudbg_buffer *dbg_buff,
+				struct cudbg_error *cudbg_err)
+{
+	struct adapter *padap = pdbg_init->adap;
+	struct cudbg_buffer temp_buff = { 0 };
+	struct ireg_buf *ch_pcie;
+	int i, rc, n;
+	u32 size;
+
+	n = sizeof(t5_pcie_pdbg_array) / (IREG_NUM_ELEM * sizeof(u32));
+	size = sizeof(struct ireg_buf) * n * 2;
+	rc = cudbg_get_buff(pdbg_init, dbg_buff, size, &temp_buff);
+	if (rc)
+		return rc;
+
+	ch_pcie = (struct ireg_buf *)temp_buff.data;
+	/* PCIE_PDBG */
+	for (i = 0; i < n; i++) {
+		struct ireg_field *pcie_pio = &ch_pcie->tp_pio;
+		u32 *buff = ch_pcie->outbuf;
+
+		pcie_pio->ireg_addr = t5_pcie_pdbg_array[i][0];
+		pcie_pio->ireg_data = t5_pcie_pdbg_array[i][1];
+		pcie_pio->ireg_local_offset = t5_pcie_pdbg_array[i][2];
+		pcie_pio->ireg_offset_range = t5_pcie_pdbg_array[i][3];
+		t4_read_indirect(padap,
+				 pcie_pio->ireg_addr,
+				 pcie_pio->ireg_data,
+				 buff,
+				 pcie_pio->ireg_offset_range,
+				 pcie_pio->ireg_local_offset);
+		ch_pcie++;
+	}
+
+	/* PCIE_CDBG */
+	n = sizeof(t5_pcie_cdbg_array) / (IREG_NUM_ELEM * sizeof(u32));
+	for (i = 0; i < n; i++) {
+		struct ireg_field *pcie_pio = &ch_pcie->tp_pio;
+		u32 *buff = ch_pcie->outbuf;
+
+		pcie_pio->ireg_addr = t5_pcie_cdbg_array[i][0];
+		pcie_pio->ireg_data = t5_pcie_cdbg_array[i][1];
+		pcie_pio->ireg_local_offset = t5_pcie_cdbg_array[i][2];
+		pcie_pio->ireg_offset_range = t5_pcie_cdbg_array[i][3];
+		t4_read_indirect(padap,
+				 pcie_pio->ireg_addr,
+				 pcie_pio->ireg_data,
+				 buff,
+				 pcie_pio->ireg_offset_range,
+				 pcie_pio->ireg_local_offset);
+		ch_pcie++;
+	}
+	return cudbg_write_and_release_buff(pdbg_init, &temp_buff, dbg_buff);
+}
+
+int cudbg_collect_pm_indirect(struct cudbg_init *pdbg_init,
+			      struct cudbg_buffer *dbg_buff,
+			      struct cudbg_error *cudbg_err)
+{
+	struct adapter *padap = pdbg_init->adap;
+	struct cudbg_buffer temp_buff = { 0 };
+	struct ireg_buf *ch_pm;
+	int i, rc, n;
+	u32 size;
+
+	n = sizeof(t5_pm_rx_array) / (IREG_NUM_ELEM * sizeof(u32));
+	size = sizeof(struct ireg_buf) * n * 2;
+	rc = cudbg_get_buff(pdbg_init, dbg_buff, size, &temp_buff);
+	if (rc)
+		return rc;
+
+	ch_pm = (struct ireg_buf *)temp_buff.data;
+	/* PM_RX */
+	for (i = 0; i < n; i++) {
+		struct ireg_field *pm_pio = &ch_pm->tp_pio;
+		u32 *buff = ch_pm->outbuf;
+
+		pm_pio->ireg_addr = t5_pm_rx_array[i][0];
+		pm_pio->ireg_data = t5_pm_rx_array[i][1];
+		pm_pio->ireg_local_offset = t5_pm_rx_array[i][2];
+		pm_pio->ireg_offset_range = t5_pm_rx_array[i][3];
+		t4_read_indirect(padap,
+				 pm_pio->ireg_addr,
+				 pm_pio->ireg_data,
+				 buff,
+				 pm_pio->ireg_offset_range,
+				 pm_pio->ireg_local_offset);
+		ch_pm++;
+	}
+
+	/* PM_TX */
+	n = sizeof(t5_pm_tx_array) / (IREG_NUM_ELEM * sizeof(u32));
+	for (i = 0; i < n; i++) {
+		struct ireg_field *pm_pio = &ch_pm->tp_pio;
+		u32 *buff = ch_pm->outbuf;
+
+		pm_pio->ireg_addr = t5_pm_tx_array[i][0];
+		pm_pio->ireg_data = t5_pm_tx_array[i][1];
+		pm_pio->ireg_local_offset = t5_pm_tx_array[i][2];
+		pm_pio->ireg_offset_range = t5_pm_tx_array[i][3];
+		t4_read_indirect(padap,
+				 pm_pio->ireg_addr,
+				 pm_pio->ireg_data,
+				 buff,
+				 pm_pio->ireg_offset_range,
+				 pm_pio->ireg_local_offset);
+		ch_pm++;
+	}
+	return cudbg_write_and_release_buff(pdbg_init, &temp_buff, dbg_buff);
+}
+
+int cudbg_collect_tid(struct cudbg_init *pdbg_init,
+		      struct cudbg_buffer *dbg_buff,
+		      struct cudbg_error *cudbg_err)
+{
+	struct adapter *padap = pdbg_init->adap;
+	struct cudbg_tid_info_region_rev1 *tid1;
+	struct cudbg_buffer temp_buff = { 0 };
+	struct cudbg_tid_info_region *tid;
+	u32 para[2], val[2];
+	int rc;
+
+	rc = cudbg_get_buff(pdbg_init, dbg_buff,
+			    sizeof(struct cudbg_tid_info_region_rev1),
+			    &temp_buff);
+	if (rc)
+		return rc;
+
+	tid1 = (struct cudbg_tid_info_region_rev1 *)temp_buff.data;
+	tid = &tid1->tid;
+	tid1->ver_hdr.signature = CUDBG_ENTITY_SIGNATURE;
+	tid1->ver_hdr.revision = CUDBG_TID_INFO_REV;
+	tid1->ver_hdr.size = sizeof(struct cudbg_tid_info_region_rev1) -
+			     sizeof(struct cudbg_ver_hdr);
+
+	/* If firmware is not attached/alive, use backdoor register
+	 * access to collect dump.
+	 */
+	if (!is_fw_attached(pdbg_init))
+		goto fill_tid;
+
+#define FW_PARAM_PFVF_A(param) \
+	(FW_PARAMS_MNEM_V(FW_PARAMS_MNEM_PFVF) | \
+	 FW_PARAMS_PARAM_X_V(FW_PARAMS_PARAM_PFVF_##param) | \
+	 FW_PARAMS_PARAM_Y_V(0) | \
+	 FW_PARAMS_PARAM_Z_V(0))
+
+	para[0] = FW_PARAM_PFVF_A(ETHOFLD_START);
+	para[1] = FW_PARAM_PFVF_A(ETHOFLD_END);
+	rc = t4_query_params(padap, padap->mbox, padap->pf, 0, 2, para, val);
+	if (rc <  0) {
+		cudbg_err->sys_err = rc;
+		cudbg_put_buff(pdbg_init, &temp_buff);
+		return rc;
+	}
+	tid->uotid_base = val[0];
+	tid->nuotids = val[1] - val[0] + 1;
+
+	if (is_t5(padap->params.chip)) {
+		tid->sb = t4_read_reg(padap, LE_DB_SERVER_INDEX_A) / 4;
+	} else if (is_t6(padap->params.chip)) {
+		tid1->tid_start =
+			t4_read_reg(padap, LE_DB_ACTIVE_TABLE_START_INDEX_A);
+		tid->sb = t4_read_reg(padap, LE_DB_SRVR_START_INDEX_A);
+
+		para[0] = FW_PARAM_PFVF_A(HPFILTER_START);
+		para[1] = FW_PARAM_PFVF_A(HPFILTER_END);
+		rc = t4_query_params(padap, padap->mbox, padap->pf, 0, 2,
+				     para, val);
+		if (rc < 0) {
+			cudbg_err->sys_err = rc;
+			cudbg_put_buff(pdbg_init, &temp_buff);
+			return rc;
+		}
+		tid->hpftid_base = val[0];
+		tid->nhpftids = val[1] - val[0] + 1;
+	}
+
+#undef FW_PARAM_PFVF_A
+
+fill_tid:
+	tid->ntids = padap->tids.ntids;
+	tid->nstids = padap->tids.nstids;
+	tid->stid_base = padap->tids.stid_base;
+	tid->hash_base = padap->tids.hash_base;
+
+	tid->natids = padap->tids.natids;
+	tid->nftids = padap->tids.nftids;
+	tid->ftid_base = padap->tids.ftid_base;
+	tid->aftid_base = padap->tids.aftid_base;
+	tid->aftid_end = padap->tids.aftid_end;
+
+	tid->sftid_base = padap->tids.sftid_base;
+	tid->nsftids = padap->tids.nsftids;
+
+	tid->flags = padap->flags;
+	tid->le_db_conf = t4_read_reg(padap, LE_DB_CONFIG_A);
+	tid->ip_users = t4_read_reg(padap, LE_DB_ACT_CNT_IPV4_A);
+	tid->ipv6_users = t4_read_reg(padap, LE_DB_ACT_CNT_IPV6_A);
+
+	return cudbg_write_and_release_buff(pdbg_init, &temp_buff, dbg_buff);
+}
+
+int cudbg_collect_pcie_config(struct cudbg_init *pdbg_init,
+			      struct cudbg_buffer *dbg_buff,
+			      struct cudbg_error *cudbg_err)
+{
+	struct adapter *padap = pdbg_init->adap;
+	struct cudbg_buffer temp_buff = { 0 };
+	u32 size, *value, j;
+	int i, rc, n;
+
+	size = sizeof(u32) * CUDBG_NUM_PCIE_CONFIG_REGS;
+	n = sizeof(t5_pcie_config_array) / (2 * sizeof(u32));
+	rc = cudbg_get_buff(pdbg_init, dbg_buff, size, &temp_buff);
+	if (rc)
+		return rc;
+
+	value = (u32 *)temp_buff.data;
+	for (i = 0; i < n; i++) {
+		for (j = t5_pcie_config_array[i][0];
+		     j <= t5_pcie_config_array[i][1]; j += 4) {
+			t4_hw_pci_read_cfg4(padap, j, value);
+			value++;
+		}
+	}
+	return cudbg_write_and_release_buff(pdbg_init, &temp_buff, dbg_buff);
+}
+
+static int cudbg_sge_ctxt_check_valid(u32 *buf, int type)
+{
+	int index, bit, bit_pos = 0;
+
+	switch (type) {
+	case CTXT_EGRESS:
+		bit_pos = 176;
+		break;
+	case CTXT_INGRESS:
+		bit_pos = 141;
+		break;
+	case CTXT_FLM:
+		bit_pos = 89;
+		break;
+	}
+	index = bit_pos / 32;
+	bit =  bit_pos % 32;
+	return buf[index] & (1U << bit);
+}
+
+static int cudbg_get_ctxt_region_info(struct adapter *padap,
+				      struct cudbg_region_info *ctx_info,
+				      u8 *mem_type)
+{
+	struct cudbg_mem_desc mem_desc;
+	struct cudbg_meminfo meminfo;
+	u32 i, j, value, found;
+	u8 flq;
+	int rc;
+
+	rc = cudbg_fill_meminfo(padap, &meminfo);
+	if (rc)
+		return rc;
+
+	/* Get EGRESS and INGRESS context region size */
+	for (i = CTXT_EGRESS; i <= CTXT_INGRESS; i++) {
+		found = 0;
+		memset(&mem_desc, 0, sizeof(struct cudbg_mem_desc));
+		for (j = 0; j < ARRAY_SIZE(meminfo.avail); j++) {
+			rc = cudbg_get_mem_region(padap, &meminfo, j,
+						  cudbg_region[i],
+						  &mem_desc);
+			if (!rc) {
+				found = 1;
+				rc = cudbg_get_mem_relative(padap, &meminfo, j,
+							    &mem_desc.base,
+							    &mem_desc.limit);
+				if (rc) {
+					ctx_info[i].exist = false;
+					break;
+				}
+				ctx_info[i].exist = true;
+				ctx_info[i].start = mem_desc.base;
+				ctx_info[i].end = mem_desc.limit;
+				mem_type[i] = j;
+				break;
+			}
+		}
+		if (!found)
+			ctx_info[i].exist = false;
+	}
+
+	/* Get FLM and CNM max qid. */
+	value = t4_read_reg(padap, SGE_FLM_CFG_A);
+
+	/* Get number of data freelist queues */
+	flq = HDRSTARTFLQ_G(value);
+	ctx_info[CTXT_FLM].exist = true;
+	ctx_info[CTXT_FLM].end = (CUDBG_MAX_FL_QIDS >> flq) * SGE_CTXT_SIZE;
+
+	/* The number of CONM contexts are same as number of freelist
+	 * queues.
+	 */
+	ctx_info[CTXT_CNM].exist = true;
+	ctx_info[CTXT_CNM].end = ctx_info[CTXT_FLM].end;
+
+	return 0;
+}
+
+int cudbg_dump_context_size(struct adapter *padap)
+{
+	struct cudbg_region_info region_info[CTXT_CNM + 1] = { {0} };
+	u8 mem_type[CTXT_INGRESS + 1] = { 0 };
+	u32 i, size = 0;
+	int rc;
+
+	/* Get max valid qid for each type of queue */
+	rc = cudbg_get_ctxt_region_info(padap, region_info, mem_type);
+	if (rc)
+		return rc;
+
+	for (i = 0; i < CTXT_CNM; i++) {
+		if (!region_info[i].exist) {
+			if (i == CTXT_EGRESS || i == CTXT_INGRESS)
+				size += CUDBG_LOWMEM_MAX_CTXT_QIDS *
+					SGE_CTXT_SIZE;
+			continue;
+		}
+
+		size += (region_info[i].end - region_info[i].start + 1) /
+			SGE_CTXT_SIZE;
+	}
+	return size * sizeof(struct cudbg_ch_cntxt);
+}
+
+static void cudbg_read_sge_ctxt(struct cudbg_init *pdbg_init, u32 cid,
+				enum ctxt_type ctype, u32 *data)
+{
+	struct adapter *padap = pdbg_init->adap;
+	int rc = -1;
+
+	/* Under heavy traffic, the SGE Queue contexts registers will be
+	 * frequently accessed by firmware.
+	 *
+	 * To avoid conflicts with firmware, always ask firmware to fetch
+	 * the SGE Queue contexts via mailbox. On failure, fallback to
+	 * accessing hardware registers directly.
+	 */
+	if (is_fw_attached(pdbg_init))
+		rc = t4_sge_ctxt_rd(padap, padap->mbox, cid, ctype, data);
+	if (rc)
+		t4_sge_ctxt_rd_bd(padap, cid, ctype, data);
+}
+
+static void cudbg_get_sge_ctxt_fw(struct cudbg_init *pdbg_init, u32 max_qid,
+				  u8 ctxt_type,
+				  struct cudbg_ch_cntxt **out_buff)
+{
+	struct cudbg_ch_cntxt *buff = *out_buff;
+	int rc;
+	u32 j;
+
+	for (j = 0; j < max_qid; j++) {
+		cudbg_read_sge_ctxt(pdbg_init, j, ctxt_type, buff->data);
+		rc = cudbg_sge_ctxt_check_valid(buff->data, ctxt_type);
+		if (!rc)
+			continue;
+
+		buff->cntxt_type = ctxt_type;
+		buff->cntxt_id = j;
+		buff++;
+		if (ctxt_type == CTXT_FLM) {
+			cudbg_read_sge_ctxt(pdbg_init, j, CTXT_CNM, buff->data);
+			buff->cntxt_type = CTXT_CNM;
+			buff->cntxt_id = j;
+			buff++;
+		}
+	}
+
+	*out_buff = buff;
+}
+
+int cudbg_collect_dump_context(struct cudbg_init *pdbg_init,
+			       struct cudbg_buffer *dbg_buff,
+			       struct cudbg_error *cudbg_err)
+{
+	struct cudbg_region_info region_info[CTXT_CNM + 1] = { {0} };
+	struct adapter *padap = pdbg_init->adap;
+	u32 j, size, max_ctx_size, max_ctx_qid;
+	u8 mem_type[CTXT_INGRESS + 1] = { 0 };
+	struct cudbg_buffer temp_buff = { 0 };
+	struct cudbg_ch_cntxt *buff;
+	u8 *ctx_buf;
+	u8 i, k;
+	int rc;
+
+	/* Get max valid qid for each type of queue */
+	rc = cudbg_get_ctxt_region_info(padap, region_info, mem_type);
+	if (rc)
+		return rc;
+
+	rc = cudbg_dump_context_size(padap);
+	if (rc <= 0)
+		return CUDBG_STATUS_ENTITY_NOT_FOUND;
+
+	size = rc;
+	rc = cudbg_get_buff(pdbg_init, dbg_buff, size, &temp_buff);
+	if (rc)
+		return rc;
+
+	/* Get buffer with enough space to read the biggest context
+	 * region in memory.
+	 */
+	max_ctx_size = max(region_info[CTXT_EGRESS].end -
+			   region_info[CTXT_EGRESS].start + 1,
+			   region_info[CTXT_INGRESS].end -
+			   region_info[CTXT_INGRESS].start + 1);
+
+	ctx_buf = kvzalloc(max_ctx_size, GFP_KERNEL);
+	if (!ctx_buf) {
+		cudbg_put_buff(pdbg_init, &temp_buff);
+		return -ENOMEM;
+	}
+
+	buff = (struct cudbg_ch_cntxt *)temp_buff.data;
+
+	/* Collect EGRESS and INGRESS context data.
+	 * In case of failures, fallback to collecting via FW or
+	 * backdoor access.
+	 */
+	for (i = CTXT_EGRESS; i <= CTXT_INGRESS; i++) {
+		if (!region_info[i].exist) {
+			max_ctx_qid = CUDBG_LOWMEM_MAX_CTXT_QIDS;
+			cudbg_get_sge_ctxt_fw(pdbg_init, max_ctx_qid, i,
+					      &buff);
+			continue;
+		}
+
+		max_ctx_size = region_info[i].end - region_info[i].start + 1;
+		max_ctx_qid = max_ctx_size / SGE_CTXT_SIZE;
+
+		/* If firmware is not attached/alive, use backdoor register
+		 * access to collect dump.
+		 */
+		if (is_fw_attached(pdbg_init)) {
+			t4_sge_ctxt_flush(padap, padap->mbox, i);
+
+			rc = t4_memory_rw(padap, MEMWIN_NIC, mem_type[i],
+					  region_info[i].start, max_ctx_size,
+					  (__be32 *)ctx_buf, 1);
+		}
+
+		if (rc || !is_fw_attached(pdbg_init)) {
+			max_ctx_qid = CUDBG_LOWMEM_MAX_CTXT_QIDS;
+			cudbg_get_sge_ctxt_fw(pdbg_init, max_ctx_qid, i,
+					      &buff);
+			continue;
+		}
+
+		for (j = 0; j < max_ctx_qid; j++) {
+			__be64 *dst_off;
+			u64 *src_off;
+
+			src_off = (u64 *)(ctx_buf + j * SGE_CTXT_SIZE);
+			dst_off = (__be64 *)buff->data;
+
+			/* The data is stored in 64-bit cpu order.  Convert it
+			 * to big endian before parsing.
+			 */
+			for (k = 0; k < SGE_CTXT_SIZE / sizeof(u64); k++)
+				dst_off[k] = cpu_to_be64(src_off[k]);
+
+			rc = cudbg_sge_ctxt_check_valid(buff->data, i);
+			if (!rc)
+				continue;
+
+			buff->cntxt_type = i;
+			buff->cntxt_id = j;
+			buff++;
+		}
+	}
+
+	kvfree(ctx_buf);
+
+	/* Collect FREELIST and CONGESTION MANAGER contexts */
+	max_ctx_size = region_info[CTXT_FLM].end -
+		       region_info[CTXT_FLM].start + 1;
+	max_ctx_qid = max_ctx_size / SGE_CTXT_SIZE;
+	/* Since FLM and CONM are 1-to-1 mapped, the below function
+	 * will fetch both FLM and CONM contexts.
+	 */
+	cudbg_get_sge_ctxt_fw(pdbg_init, max_ctx_qid, CTXT_FLM, &buff);
+
+	return cudbg_write_and_release_buff(pdbg_init, &temp_buff, dbg_buff);
+}
+
+static inline void cudbg_tcamxy2valmask(u64 x, u64 y, u8 *addr, u64 *mask)
+{
+	*mask = x | y;
+	y = (__force u64)cpu_to_be64(y);
+	memcpy(addr, (char *)&y + 2, ETH_ALEN);
+}
+
+static void cudbg_mps_rpl_backdoor(struct adapter *padap,
+				   struct fw_ldst_mps_rplc *mps_rplc)
+{
+	if (is_t5(padap->params.chip)) {
+		mps_rplc->rplc255_224 = htonl(t4_read_reg(padap,
+							  MPS_VF_RPLCT_MAP3_A));
+		mps_rplc->rplc223_192 = htonl(t4_read_reg(padap,
+							  MPS_VF_RPLCT_MAP2_A));
+		mps_rplc->rplc191_160 = htonl(t4_read_reg(padap,
+							  MPS_VF_RPLCT_MAP1_A));
+		mps_rplc->rplc159_128 = htonl(t4_read_reg(padap,
+							  MPS_VF_RPLCT_MAP0_A));
+	} else {
+		mps_rplc->rplc255_224 = htonl(t4_read_reg(padap,
+							  MPS_VF_RPLCT_MAP7_A));
+		mps_rplc->rplc223_192 = htonl(t4_read_reg(padap,
+							  MPS_VF_RPLCT_MAP6_A));
+		mps_rplc->rplc191_160 = htonl(t4_read_reg(padap,
+							  MPS_VF_RPLCT_MAP5_A));
+		mps_rplc->rplc159_128 = htonl(t4_read_reg(padap,
+							  MPS_VF_RPLCT_MAP4_A));
+	}
+	mps_rplc->rplc127_96 = htonl(t4_read_reg(padap, MPS_VF_RPLCT_MAP3_A));
+	mps_rplc->rplc95_64 = htonl(t4_read_reg(padap, MPS_VF_RPLCT_MAP2_A));
+	mps_rplc->rplc63_32 = htonl(t4_read_reg(padap, MPS_VF_RPLCT_MAP1_A));
+	mps_rplc->rplc31_0 = htonl(t4_read_reg(padap, MPS_VF_RPLCT_MAP0_A));
+}
+
+static int cudbg_collect_tcam_index(struct cudbg_init *pdbg_init,
+				    struct cudbg_mps_tcam *tcam, u32 idx)
+{
+	struct adapter *padap = pdbg_init->adap;
+	u64 tcamy, tcamx, val;
+	u32 ctl, data2;
+	int rc = 0;
+
+	if (CHELSIO_CHIP_VERSION(padap->params.chip) >= CHELSIO_T6) {
+		/* CtlReqID   - 1: use Host Driver Requester ID
+		 * CtlCmdType - 0: Read, 1: Write
+		 * CtlTcamSel - 0: TCAM0, 1: TCAM1
+		 * CtlXYBitSel- 0: Y bit, 1: X bit
+		 */
+
+		/* Read tcamy */
+		ctl = CTLREQID_V(1) | CTLCMDTYPE_V(0) | CTLXYBITSEL_V(0);
+		if (idx < 256)
+			ctl |= CTLTCAMINDEX_V(idx) | CTLTCAMSEL_V(0);
+		else
+			ctl |= CTLTCAMINDEX_V(idx - 256) | CTLTCAMSEL_V(1);
+
+		t4_write_reg(padap, MPS_CLS_TCAM_DATA2_CTL_A, ctl);
+		val = t4_read_reg(padap, MPS_CLS_TCAM_RDATA1_REQ_ID1_A);
+		tcamy = DMACH_G(val) << 32;
+		tcamy |= t4_read_reg(padap, MPS_CLS_TCAM_RDATA0_REQ_ID1_A);
+		data2 = t4_read_reg(padap, MPS_CLS_TCAM_RDATA2_REQ_ID1_A);
+		tcam->lookup_type = DATALKPTYPE_G(data2);
+
+		/* 0 - Outer header, 1 - Inner header
+		 * [71:48] bit locations are overloaded for
+		 * outer vs. inner lookup types.
+		 */
+		if (tcam->lookup_type && tcam->lookup_type != DATALKPTYPE_M) {
+			/* Inner header VNI */
+			tcam->vniy = (data2 & DATAVIDH2_F) | DATAVIDH1_G(data2);
+			tcam->vniy = (tcam->vniy << 16) | VIDL_G(val);
+			tcam->dip_hit = data2 & DATADIPHIT_F;
+		} else {
+			tcam->vlan_vld = data2 & DATAVIDH2_F;
+			tcam->ivlan = VIDL_G(val);
+		}
+
+		tcam->port_num = DATAPORTNUM_G(data2);
+
+		/* Read tcamx. Change the control param */
+		ctl |= CTLXYBITSEL_V(1);
+		t4_write_reg(padap, MPS_CLS_TCAM_DATA2_CTL_A, ctl);
+		val = t4_read_reg(padap, MPS_CLS_TCAM_RDATA1_REQ_ID1_A);
+		tcamx = DMACH_G(val) << 32;
+		tcamx |= t4_read_reg(padap, MPS_CLS_TCAM_RDATA0_REQ_ID1_A);
+		data2 = t4_read_reg(padap, MPS_CLS_TCAM_RDATA2_REQ_ID1_A);
+		if (tcam->lookup_type && tcam->lookup_type != DATALKPTYPE_M) {
+			/* Inner header VNI mask */
+			tcam->vnix = (data2 & DATAVIDH2_F) | DATAVIDH1_G(data2);
+			tcam->vnix = (tcam->vnix << 16) | VIDL_G(val);
+		}
+	} else {
+		tcamy = t4_read_reg64(padap, MPS_CLS_TCAM_Y_L(idx));
+		tcamx = t4_read_reg64(padap, MPS_CLS_TCAM_X_L(idx));
+	}
+
+	/* If no entry, return */
+	if (tcamx & tcamy)
+		return rc;
+
+	tcam->cls_lo = t4_read_reg(padap, MPS_CLS_SRAM_L(idx));
+	tcam->cls_hi = t4_read_reg(padap, MPS_CLS_SRAM_H(idx));
+
+	if (is_t5(padap->params.chip))
+		tcam->repli = (tcam->cls_lo & REPLICATE_F);
+	else if (is_t6(padap->params.chip))
+		tcam->repli = (tcam->cls_lo & T6_REPLICATE_F);
+
+	if (tcam->repli) {
+		struct fw_ldst_cmd ldst_cmd;
+		struct fw_ldst_mps_rplc mps_rplc;
+
+		memset(&ldst_cmd, 0, sizeof(ldst_cmd));
+		ldst_cmd.op_to_addrspace =
+			htonl(FW_CMD_OP_V(FW_LDST_CMD) |
+			      FW_CMD_REQUEST_F | FW_CMD_READ_F |
+			      FW_LDST_CMD_ADDRSPACE_V(FW_LDST_ADDRSPC_MPS));
+		ldst_cmd.cycles_to_len16 = htonl(FW_LEN16(ldst_cmd));
+		ldst_cmd.u.mps.rplc.fid_idx =
+			htons(FW_LDST_CMD_FID_V(FW_LDST_MPS_RPLC) |
+			      FW_LDST_CMD_IDX_V(idx));
+
+		/* If firmware is not attached/alive, use backdoor register
+		 * access to collect dump.
+		 */
+		if (is_fw_attached(pdbg_init))
+			rc = t4_wr_mbox(padap, padap->mbox, &ldst_cmd,
+					sizeof(ldst_cmd), &ldst_cmd);
+
+		if (rc || !is_fw_attached(pdbg_init)) {
+			cudbg_mps_rpl_backdoor(padap, &mps_rplc);
+			/* Ignore error since we collected directly from
+			 * reading registers.
+			 */
+			rc = 0;
+		} else {
+			mps_rplc = ldst_cmd.u.mps.rplc;
+		}
+
+		tcam->rplc[0] = ntohl(mps_rplc.rplc31_0);
+		tcam->rplc[1] = ntohl(mps_rplc.rplc63_32);
+		tcam->rplc[2] = ntohl(mps_rplc.rplc95_64);
+		tcam->rplc[3] = ntohl(mps_rplc.rplc127_96);
+		if (padap->params.arch.mps_rplc_size > CUDBG_MAX_RPLC_SIZE) {
+			tcam->rplc[4] = ntohl(mps_rplc.rplc159_128);
+			tcam->rplc[5] = ntohl(mps_rplc.rplc191_160);
+			tcam->rplc[6] = ntohl(mps_rplc.rplc223_192);
+			tcam->rplc[7] = ntohl(mps_rplc.rplc255_224);
+		}
+	}
+	cudbg_tcamxy2valmask(tcamx, tcamy, tcam->addr, &tcam->mask);
+	tcam->idx = idx;
+	tcam->rplc_size = padap->params.arch.mps_rplc_size;
+	return rc;
+}
+
+int cudbg_collect_mps_tcam(struct cudbg_init *pdbg_init,
+			   struct cudbg_buffer *dbg_buff,
+			   struct cudbg_error *cudbg_err)
+{
+	struct adapter *padap = pdbg_init->adap;
+	struct cudbg_buffer temp_buff = { 0 };
+	u32 size = 0, i, n, total_size = 0;
+	struct cudbg_mps_tcam *tcam;
+	int rc;
+
+	n = padap->params.arch.mps_tcam_size;
+	size = sizeof(struct cudbg_mps_tcam) * n;
+	rc = cudbg_get_buff(pdbg_init, dbg_buff, size, &temp_buff);
+	if (rc)
+		return rc;
+
+	tcam = (struct cudbg_mps_tcam *)temp_buff.data;
+	for (i = 0; i < n; i++) {
+		rc = cudbg_collect_tcam_index(pdbg_init, tcam, i);
+		if (rc) {
+			cudbg_err->sys_err = rc;
+			cudbg_put_buff(pdbg_init, &temp_buff);
+			return rc;
+		}
+		total_size += sizeof(struct cudbg_mps_tcam);
+		tcam++;
+	}
+
+	if (!total_size) {
+		rc = CUDBG_SYSTEM_ERROR;
+		cudbg_err->sys_err = rc;
+		cudbg_put_buff(pdbg_init, &temp_buff);
+		return rc;
+	}
+	return cudbg_write_and_release_buff(pdbg_init, &temp_buff, dbg_buff);
+}
+
+int cudbg_collect_vpd_data(struct cudbg_init *pdbg_init,
+			   struct cudbg_buffer *dbg_buff,
+			   struct cudbg_error *cudbg_err)
+{
+	struct adapter *padap = pdbg_init->adap;
+	struct cudbg_buffer temp_buff = { 0 };
+	char vpd_str[CUDBG_VPD_VER_LEN + 1];
+	struct cudbg_vpd_data *vpd_data;
+	struct vpd_params vpd = { 0 };
+	u32 vpd_vers, fw_vers;
+	int rc;
+
+	rc = t4_get_raw_vpd_params(padap, &vpd);
+	if (rc)
+		return rc;
+
+	rc = t4_get_fw_version(padap, &fw_vers);
+	if (rc)
+		return rc;
+
+	rc = cudbg_read_vpd_reg(padap, CUDBG_VPD_VER_ADDR, CUDBG_VPD_VER_LEN,
+				vpd_str);
+	if (rc)
+		return rc;
+
+	vpd_str[CUDBG_VPD_VER_LEN] = '\0';
+	rc = kstrtouint(vpd_str, 0, &vpd_vers);
+	if (rc)
+		return rc;
+
+	rc = cudbg_get_buff(pdbg_init, dbg_buff, sizeof(struct cudbg_vpd_data),
+			    &temp_buff);
+	if (rc)
+		return rc;
+
+	vpd_data = (struct cudbg_vpd_data *)temp_buff.data;
+	memcpy(vpd_data->sn, vpd.sn, SERNUM_LEN + 1);
+	memcpy(vpd_data->bn, vpd.pn, PN_LEN + 1);
+	memcpy(vpd_data->na, vpd.na, MACADDR_LEN + 1);
+	memcpy(vpd_data->mn, vpd.id, ID_LEN + 1);
+	vpd_data->scfg_vers = t4_read_reg(padap, PCIE_STATIC_SPARE2_A);
+	vpd_data->vpd_vers = vpd_vers;
+	vpd_data->fw_major = FW_HDR_FW_VER_MAJOR_G(fw_vers);
+	vpd_data->fw_minor = FW_HDR_FW_VER_MINOR_G(fw_vers);
+	vpd_data->fw_micro = FW_HDR_FW_VER_MICRO_G(fw_vers);
+	vpd_data->fw_build = FW_HDR_FW_VER_BUILD_G(fw_vers);
+	return cudbg_write_and_release_buff(pdbg_init, &temp_buff, dbg_buff);
+}
+
+static int cudbg_read_tid(struct cudbg_init *pdbg_init, u32 tid,
+			  struct cudbg_tid_data *tid_data)
+{
+	struct adapter *padap = pdbg_init->adap;
+	int i, cmd_retry = 8;
+	u32 val;
+
+	/* Fill REQ_DATA regs with 0's */
+	for (i = 0; i < NUM_LE_DB_DBGI_REQ_DATA_INSTANCES; i++)
+		t4_write_reg(padap, LE_DB_DBGI_REQ_DATA_A + (i << 2), 0);
+
+	/* Write DBIG command */
+	val = DBGICMD_V(4) | DBGITID_V(tid);
+	t4_write_reg(padap, LE_DB_DBGI_REQ_TCAM_CMD_A, val);
+	tid_data->dbig_cmd = val;
+
+	val = DBGICMDSTRT_F | DBGICMDMODE_V(1); /* LE mode */
+	t4_write_reg(padap, LE_DB_DBGI_CONFIG_A, val);
+	tid_data->dbig_conf = val;
+
+	/* Poll the DBGICMDBUSY bit */
+	val = 1;
+	while (val) {
+		val = t4_read_reg(padap, LE_DB_DBGI_CONFIG_A);
+		val = val & DBGICMDBUSY_F;
+		cmd_retry--;
+		if (!cmd_retry)
+			return CUDBG_SYSTEM_ERROR;
+	}
+
+	/* Check RESP status */
+	val = t4_read_reg(padap, LE_DB_DBGI_RSP_STATUS_A);
+	tid_data->dbig_rsp_stat = val;
+	if (!(val & 1))
+		return CUDBG_SYSTEM_ERROR;
+
+	/* Read RESP data */
+	for (i = 0; i < NUM_LE_DB_DBGI_RSP_DATA_INSTANCES; i++)
+		tid_data->data[i] = t4_read_reg(padap,
+						LE_DB_DBGI_RSP_DATA_A +
+						(i << 2));
+	tid_data->tid = tid;
+	return 0;
+}
+
+static int cudbg_get_le_type(u32 tid, struct cudbg_tcam tcam_region)
+{
+	int type = LE_ET_UNKNOWN;
+
+	if (tid < tcam_region.server_start)
+		type = LE_ET_TCAM_CON;
+	else if (tid < tcam_region.filter_start)
+		type = LE_ET_TCAM_SERVER;
+	else if (tid < tcam_region.clip_start)
+		type = LE_ET_TCAM_FILTER;
+	else if (tid < tcam_region.routing_start)
+		type = LE_ET_TCAM_CLIP;
+	else if (tid < tcam_region.tid_hash_base)
+		type = LE_ET_TCAM_ROUTING;
+	else if (tid < tcam_region.max_tid)
+		type = LE_ET_HASH_CON;
+	else
+		type = LE_ET_INVALID_TID;
+
+	return type;
+}
+
+static int cudbg_is_ipv6_entry(struct cudbg_tid_data *tid_data,
+			       struct cudbg_tcam tcam_region)
+{
+	int ipv6 = 0;
+	int le_type;
+
+	le_type = cudbg_get_le_type(tid_data->tid, tcam_region);
+	if (tid_data->tid & 1)
+		return 0;
+
+	if (le_type == LE_ET_HASH_CON) {
+		ipv6 = tid_data->data[16] & 0x8000;
+	} else if (le_type == LE_ET_TCAM_CON) {
+		ipv6 = tid_data->data[16] & 0x8000;
+		if (ipv6)
+			ipv6 = tid_data->data[9] == 0x00C00000;
+	} else {
+		ipv6 = 0;
+	}
+	return ipv6;
+}
+
+void cudbg_fill_le_tcam_info(struct adapter *padap,
+			     struct cudbg_tcam *tcam_region)
+{
+	u32 value;
+
+	/* Get the LE regions */
+	value = t4_read_reg(padap, LE_DB_TID_HASHBASE_A); /* hash base index */
+	tcam_region->tid_hash_base = value;
+
+	/* Get routing table index */
+	value = t4_read_reg(padap, LE_DB_ROUTING_TABLE_INDEX_A);
+	tcam_region->routing_start = value;
+
+	/* Get clip table index. For T6 there is separate CLIP TCAM */
+	if (is_t6(padap->params.chip))
+		value = t4_read_reg(padap, LE_DB_CLCAM_TID_BASE_A);
+	else
+		value = t4_read_reg(padap, LE_DB_CLIP_TABLE_INDEX_A);
+	tcam_region->clip_start = value;
+
+	/* Get filter table index */
+	value = t4_read_reg(padap, LE_DB_FILTER_TABLE_INDEX_A);
+	tcam_region->filter_start = value;
+
+	/* Get server table index */
+	value = t4_read_reg(padap, LE_DB_SERVER_INDEX_A);
+	tcam_region->server_start = value;
+
+	/* Check whether hash is enabled and calculate the max tids */
+	value = t4_read_reg(padap, LE_DB_CONFIG_A);
+	if ((value >> HASHEN_S) & 1) {
+		value = t4_read_reg(padap, LE_DB_HASH_CONFIG_A);
+		if (CHELSIO_CHIP_VERSION(padap->params.chip) > CHELSIO_T5) {
+			tcam_region->max_tid = (value & 0xFFFFF) +
+					       tcam_region->tid_hash_base;
+		} else {
+			value = HASHTIDSIZE_G(value);
+			value = 1 << value;
+			tcam_region->max_tid = value +
+					       tcam_region->tid_hash_base;
+		}
+	} else { /* hash not enabled */
+		if (is_t6(padap->params.chip))
+			tcam_region->max_tid = (value & ASLIPCOMPEN_F) ?
+					       CUDBG_MAX_TID_COMP_EN :
+					       CUDBG_MAX_TID_COMP_DIS;
+		else
+			tcam_region->max_tid = CUDBG_MAX_TCAM_TID;
+	}
+
+	if (is_t6(padap->params.chip))
+		tcam_region->max_tid += CUDBG_T6_CLIP;
+}
+
+int cudbg_collect_le_tcam(struct cudbg_init *pdbg_init,
+			  struct cudbg_buffer *dbg_buff,
+			  struct cudbg_error *cudbg_err)
+{
+	struct adapter *padap = pdbg_init->adap;
+	struct cudbg_buffer temp_buff = { 0 };
+	struct cudbg_tcam tcam_region = { 0 };
+	struct cudbg_tid_data *tid_data;
+	u32 bytes = 0;
+	int rc, size;
+	u32 i;
+
+	cudbg_fill_le_tcam_info(padap, &tcam_region);
+
+	size = sizeof(struct cudbg_tid_data) * tcam_region.max_tid;
+	size += sizeof(struct cudbg_tcam);
+	rc = cudbg_get_buff(pdbg_init, dbg_buff, size, &temp_buff);
+	if (rc)
+		return rc;
+
+	memcpy(temp_buff.data, &tcam_region, sizeof(struct cudbg_tcam));
+	bytes = sizeof(struct cudbg_tcam);
+	tid_data = (struct cudbg_tid_data *)(temp_buff.data + bytes);
+	/* read all tid */
+	for (i = 0; i < tcam_region.max_tid; ) {
+		rc = cudbg_read_tid(pdbg_init, i, tid_data);
+		if (rc) {
+			cudbg_err->sys_warn = CUDBG_STATUS_PARTIAL_DATA;
+			/* Update tcam header and exit */
+			tcam_region.max_tid = i;
+			memcpy(temp_buff.data, &tcam_region,
+			       sizeof(struct cudbg_tcam));
+			goto out;
+		}
+
+		if (cudbg_is_ipv6_entry(tid_data, tcam_region)) {
+			/* T6 CLIP TCAM: ipv6 takes 4 entries */
+			if (is_t6(padap->params.chip) &&
+			    i >= tcam_region.clip_start &&
+			    i < tcam_region.clip_start + CUDBG_T6_CLIP)
+				i += 4;
+			else /* Main TCAM: ipv6 takes two tids */
+				i += 2;
+		} else {
+			i++;
+		}
+
+		tid_data++;
+		bytes += sizeof(struct cudbg_tid_data);
+	}
+
+out:
+	return cudbg_write_and_release_buff(pdbg_init, &temp_buff, dbg_buff);
+}
+
+int cudbg_collect_cctrl(struct cudbg_init *pdbg_init,
+			struct cudbg_buffer *dbg_buff,
+			struct cudbg_error *cudbg_err)
+{
+	struct adapter *padap = pdbg_init->adap;
+	struct cudbg_buffer temp_buff = { 0 };
+	u32 size;
+	int rc;
+
+	size = sizeof(u16) * NMTUS * NCCTRL_WIN;
+	rc = cudbg_get_buff(pdbg_init, dbg_buff, size, &temp_buff);
+	if (rc)
+		return rc;
+
+	t4_read_cong_tbl(padap, (void *)temp_buff.data);
+	return cudbg_write_and_release_buff(pdbg_init, &temp_buff, dbg_buff);
+}
+
+int cudbg_collect_ma_indirect(struct cudbg_init *pdbg_init,
+			      struct cudbg_buffer *dbg_buff,
+			      struct cudbg_error *cudbg_err)
+{
+	struct adapter *padap = pdbg_init->adap;
+	struct cudbg_buffer temp_buff = { 0 };
+	struct ireg_buf *ma_indr;
+	int i, rc, n;
+	u32 size, j;
+
+	if (CHELSIO_CHIP_VERSION(padap->params.chip) < CHELSIO_T6)
+		return CUDBG_STATUS_ENTITY_NOT_FOUND;
+
+	n = sizeof(t6_ma_ireg_array) / (IREG_NUM_ELEM * sizeof(u32));
+	size = sizeof(struct ireg_buf) * n * 2;
+	rc = cudbg_get_buff(pdbg_init, dbg_buff, size, &temp_buff);
+	if (rc)
+		return rc;
+
+	ma_indr = (struct ireg_buf *)temp_buff.data;
+	for (i = 0; i < n; i++) {
+		struct ireg_field *ma_fli = &ma_indr->tp_pio;
+		u32 *buff = ma_indr->outbuf;
+
+		ma_fli->ireg_addr = t6_ma_ireg_array[i][0];
+		ma_fli->ireg_data = t6_ma_ireg_array[i][1];
+		ma_fli->ireg_local_offset = t6_ma_ireg_array[i][2];
+		ma_fli->ireg_offset_range = t6_ma_ireg_array[i][3];
+		t4_read_indirect(padap, ma_fli->ireg_addr, ma_fli->ireg_data,
+				 buff, ma_fli->ireg_offset_range,
+				 ma_fli->ireg_local_offset);
+		ma_indr++;
+	}
+
+	n = sizeof(t6_ma_ireg_array2) / (IREG_NUM_ELEM * sizeof(u32));
+	for (i = 0; i < n; i++) {
+		struct ireg_field *ma_fli = &ma_indr->tp_pio;
+		u32 *buff = ma_indr->outbuf;
+
+		ma_fli->ireg_addr = t6_ma_ireg_array2[i][0];
+		ma_fli->ireg_data = t6_ma_ireg_array2[i][1];
+		ma_fli->ireg_local_offset = t6_ma_ireg_array2[i][2];
+		for (j = 0; j < t6_ma_ireg_array2[i][3]; j++) {
+			t4_read_indirect(padap, ma_fli->ireg_addr,
+					 ma_fli->ireg_data, buff, 1,
+					 ma_fli->ireg_local_offset);
+			buff++;
+			ma_fli->ireg_local_offset += 0x20;
+		}
+		ma_indr++;
+	}
+	return cudbg_write_and_release_buff(pdbg_init, &temp_buff, dbg_buff);
+}
+
+int cudbg_collect_ulptx_la(struct cudbg_init *pdbg_init,
+			   struct cudbg_buffer *dbg_buff,
+			   struct cudbg_error *cudbg_err)
+{
+	struct adapter *padap = pdbg_init->adap;
+	struct cudbg_buffer temp_buff = { 0 };
+	struct cudbg_ulptx_la *ulptx_la_buff;
+	struct cudbg_ver_hdr *ver_hdr;
+	u32 i, j;
+	int rc;
+
+	rc = cudbg_get_buff(pdbg_init, dbg_buff,
+			    sizeof(struct cudbg_ver_hdr) +
+			    sizeof(struct cudbg_ulptx_la),
+			    &temp_buff);
+	if (rc)
+		return rc;
+
+	ver_hdr = (struct cudbg_ver_hdr *)temp_buff.data;
+	ver_hdr->signature = CUDBG_ENTITY_SIGNATURE;
+	ver_hdr->revision = CUDBG_ULPTX_LA_REV;
+	ver_hdr->size = sizeof(struct cudbg_ulptx_la);
+
+	ulptx_la_buff = (struct cudbg_ulptx_la *)(temp_buff.data +
+						  sizeof(*ver_hdr));
+	for (i = 0; i < CUDBG_NUM_ULPTX; i++) {
+		ulptx_la_buff->rdptr[i] = t4_read_reg(padap,
+						      ULP_TX_LA_RDPTR_0_A +
+						      0x10 * i);
+		ulptx_la_buff->wrptr[i] = t4_read_reg(padap,
+						      ULP_TX_LA_WRPTR_0_A +
+						      0x10 * i);
+		ulptx_la_buff->rddata[i] = t4_read_reg(padap,
+						       ULP_TX_LA_RDDATA_0_A +
+						       0x10 * i);
+		for (j = 0; j < CUDBG_NUM_ULPTX_READ; j++)
+			ulptx_la_buff->rd_data[i][j] =
+				t4_read_reg(padap,
+					    ULP_TX_LA_RDDATA_0_A + 0x10 * i);
+	}
+
+	for (i = 0; i < CUDBG_NUM_ULPTX_ASIC_READ; i++) {
+		t4_write_reg(padap, ULP_TX_ASIC_DEBUG_CTRL_A, 0x1);
+		ulptx_la_buff->rdptr_asic[i] =
+				t4_read_reg(padap, ULP_TX_ASIC_DEBUG_CTRL_A);
+		ulptx_la_buff->rddata_asic[i][0] =
+				t4_read_reg(padap, ULP_TX_ASIC_DEBUG_0_A);
+		ulptx_la_buff->rddata_asic[i][1] =
+				t4_read_reg(padap, ULP_TX_ASIC_DEBUG_1_A);
+		ulptx_la_buff->rddata_asic[i][2] =
+				t4_read_reg(padap, ULP_TX_ASIC_DEBUG_2_A);
+		ulptx_la_buff->rddata_asic[i][3] =
+				t4_read_reg(padap, ULP_TX_ASIC_DEBUG_3_A);
+		ulptx_la_buff->rddata_asic[i][4] =
+				t4_read_reg(padap, ULP_TX_ASIC_DEBUG_4_A);
+		ulptx_la_buff->rddata_asic[i][5] =
+				t4_read_reg(padap, PM_RX_BASE_ADDR);
+	}
+
+	return cudbg_write_and_release_buff(pdbg_init, &temp_buff, dbg_buff);
+}
+
+int cudbg_collect_up_cim_indirect(struct cudbg_init *pdbg_init,
+				  struct cudbg_buffer *dbg_buff,
+				  struct cudbg_error *cudbg_err)
+{
+	struct adapter *padap = pdbg_init->adap;
+	struct cudbg_buffer temp_buff = { 0 };
+	u32 local_offset, local_range;
+	struct ireg_buf *up_cim;
+	u32 size, j, iter;
+	u32 instance = 0;
+	int i, rc, n;
+
+	if (is_t5(padap->params.chip))
+		n = sizeof(t5_up_cim_reg_array) /
+		    ((IREG_NUM_ELEM + 1) * sizeof(u32));
+	else if (is_t6(padap->params.chip))
+		n = sizeof(t6_up_cim_reg_array) /
+		    ((IREG_NUM_ELEM + 1) * sizeof(u32));
+	else
+		return CUDBG_STATUS_NOT_IMPLEMENTED;
+
+	size = sizeof(struct ireg_buf) * n;
+	rc = cudbg_get_buff(pdbg_init, dbg_buff, size, &temp_buff);
+	if (rc)
+		return rc;
+
+	up_cim = (struct ireg_buf *)temp_buff.data;
+	for (i = 0; i < n; i++) {
+		struct ireg_field *up_cim_reg = &up_cim->tp_pio;
+		u32 *buff = up_cim->outbuf;
+
+		if (is_t5(padap->params.chip)) {
+			up_cim_reg->ireg_addr = t5_up_cim_reg_array[i][0];
+			up_cim_reg->ireg_data = t5_up_cim_reg_array[i][1];
+			up_cim_reg->ireg_local_offset =
+						t5_up_cim_reg_array[i][2];
+			up_cim_reg->ireg_offset_range =
+						t5_up_cim_reg_array[i][3];
+			instance = t5_up_cim_reg_array[i][4];
+		} else if (is_t6(padap->params.chip)) {
+			up_cim_reg->ireg_addr = t6_up_cim_reg_array[i][0];
+			up_cim_reg->ireg_data = t6_up_cim_reg_array[i][1];
+			up_cim_reg->ireg_local_offset =
+						t6_up_cim_reg_array[i][2];
+			up_cim_reg->ireg_offset_range =
+						t6_up_cim_reg_array[i][3];
+			instance = t6_up_cim_reg_array[i][4];
+		}
+
+		switch (instance) {
+		case NUM_CIM_CTL_TSCH_CHANNEL_INSTANCES:
+			iter = up_cim_reg->ireg_offset_range;
+			local_offset = 0x120;
+			local_range = 1;
+			break;
+		case NUM_CIM_CTL_TSCH_CHANNEL_TSCH_CLASS_INSTANCES:
+			iter = up_cim_reg->ireg_offset_range;
+			local_offset = 0x10;
+			local_range = 1;
+			break;
+		default:
+			iter = 1;
+			local_offset = 0;
+			local_range = up_cim_reg->ireg_offset_range;
+			break;
+		}
+
+		for (j = 0; j < iter; j++, buff++) {
+			rc = t4_cim_read(padap,
+					 up_cim_reg->ireg_local_offset +
+					 (j * local_offset), local_range, buff);
+			if (rc) {
+				cudbg_put_buff(pdbg_init, &temp_buff);
+				return rc;
+			}
+		}
+		up_cim++;
+	}
+	return cudbg_write_and_release_buff(pdbg_init, &temp_buff, dbg_buff);
+}
+
+int cudbg_collect_pbt_tables(struct cudbg_init *pdbg_init,
+			     struct cudbg_buffer *dbg_buff,
+			     struct cudbg_error *cudbg_err)
+{
+	struct adapter *padap = pdbg_init->adap;
+	struct cudbg_buffer temp_buff = { 0 };
+	struct cudbg_pbt_tables *pbt;
+	int i, rc;
+	u32 addr;
+
+	rc = cudbg_get_buff(pdbg_init, dbg_buff,
+			    sizeof(struct cudbg_pbt_tables),
+			    &temp_buff);
+	if (rc)
+		return rc;
+
+	pbt = (struct cudbg_pbt_tables *)temp_buff.data;
+	/* PBT dynamic entries */
+	addr = CUDBG_CHAC_PBT_ADDR;
+	for (i = 0; i < CUDBG_PBT_DYNAMIC_ENTRIES; i++) {
+		rc = t4_cim_read(padap, addr + (i * 4), 1,
+				 &pbt->pbt_dynamic[i]);
+		if (rc) {
+			cudbg_err->sys_err = rc;
+			cudbg_put_buff(pdbg_init, &temp_buff);
+			return rc;
+		}
+	}
+
+	/* PBT static entries */
+	/* static entries start when bit 6 is set */
+	addr = CUDBG_CHAC_PBT_ADDR + (1 << 6);
+	for (i = 0; i < CUDBG_PBT_STATIC_ENTRIES; i++) {
+		rc = t4_cim_read(padap, addr + (i * 4), 1,
+				 &pbt->pbt_static[i]);
+		if (rc) {
+			cudbg_err->sys_err = rc;
+			cudbg_put_buff(pdbg_init, &temp_buff);
+			return rc;
+		}
+	}
+
+	/* LRF entries */
+	addr = CUDBG_CHAC_PBT_LRF;
+	for (i = 0; i < CUDBG_LRF_ENTRIES; i++) {
+		rc = t4_cim_read(padap, addr + (i * 4), 1,
+				 &pbt->lrf_table[i]);
+		if (rc) {
+			cudbg_err->sys_err = rc;
+			cudbg_put_buff(pdbg_init, &temp_buff);
+			return rc;
+		}
+	}
+
+	/* PBT data entries */
+	addr = CUDBG_CHAC_PBT_DATA;
+	for (i = 0; i < CUDBG_PBT_DATA_ENTRIES; i++) {
+		rc = t4_cim_read(padap, addr + (i * 4), 1,
+				 &pbt->pbt_data[i]);
+		if (rc) {
+			cudbg_err->sys_err = rc;
+			cudbg_put_buff(pdbg_init, &temp_buff);
+			return rc;
+		}
+	}
+	return cudbg_write_and_release_buff(pdbg_init, &temp_buff, dbg_buff);
+}
+
+int cudbg_collect_mbox_log(struct cudbg_init *pdbg_init,
+			   struct cudbg_buffer *dbg_buff,
+			   struct cudbg_error *cudbg_err)
+{
+	struct adapter *padap = pdbg_init->adap;
+	struct cudbg_mbox_log *mboxlog = NULL;
+	struct cudbg_buffer temp_buff = { 0 };
+	struct mbox_cmd_log *log = NULL;
+	struct mbox_cmd *entry;
+	unsigned int entry_idx;
+	u16 mbox_cmds;
+	int i, k, rc;
+	u64 flit;
+	u32 size;
+
+	log = padap->mbox_log;
+	mbox_cmds = padap->mbox_log->size;
+	size = sizeof(struct cudbg_mbox_log) * mbox_cmds;
+	rc = cudbg_get_buff(pdbg_init, dbg_buff, size, &temp_buff);
+	if (rc)
+		return rc;
+
+	mboxlog = (struct cudbg_mbox_log *)temp_buff.data;
+	for (k = 0; k < mbox_cmds; k++) {
+		entry_idx = log->cursor + k;
+		if (entry_idx >= log->size)
+			entry_idx -= log->size;
+
+		entry = mbox_cmd_log_entry(log, entry_idx);
+		/* skip over unused entries */
+		if (entry->timestamp == 0)
+			continue;
+
+		memcpy(&mboxlog->entry, entry, sizeof(struct mbox_cmd));
+		for (i = 0; i < MBOX_LEN / 8; i++) {
+			flit = entry->cmd[i];
+			mboxlog->hi[i] = (u32)(flit >> 32);
+			mboxlog->lo[i] = (u32)flit;
+		}
+		mboxlog++;
+	}
+	return cudbg_write_and_release_buff(pdbg_init, &temp_buff, dbg_buff);
+}
+
+int cudbg_collect_hma_indirect(struct cudbg_init *pdbg_init,
+			       struct cudbg_buffer *dbg_buff,
+			       struct cudbg_error *cudbg_err)
+{
+	struct adapter *padap = pdbg_init->adap;
+	struct cudbg_buffer temp_buff = { 0 };
+	struct ireg_buf *hma_indr;
+	int i, rc, n;
+	u32 size;
+
+	if (CHELSIO_CHIP_VERSION(padap->params.chip) < CHELSIO_T6)
+		return CUDBG_STATUS_ENTITY_NOT_FOUND;
+
+	n = sizeof(t6_hma_ireg_array) / (IREG_NUM_ELEM * sizeof(u32));
+	size = sizeof(struct ireg_buf) * n;
+	rc = cudbg_get_buff(pdbg_init, dbg_buff, size, &temp_buff);
+	if (rc)
+		return rc;
+
+	hma_indr = (struct ireg_buf *)temp_buff.data;
+	for (i = 0; i < n; i++) {
+		struct ireg_field *hma_fli = &hma_indr->tp_pio;
+		u32 *buff = hma_indr->outbuf;
+
+		hma_fli->ireg_addr = t6_hma_ireg_array[i][0];
+		hma_fli->ireg_data = t6_hma_ireg_array[i][1];
+		hma_fli->ireg_local_offset = t6_hma_ireg_array[i][2];
+		hma_fli->ireg_offset_range = t6_hma_ireg_array[i][3];
+		t4_read_indirect(padap, hma_fli->ireg_addr, hma_fli->ireg_data,
+				 buff, hma_fli->ireg_offset_range,
+				 hma_fli->ireg_local_offset);
+		hma_indr++;
+	}
+	return cudbg_write_and_release_buff(pdbg_init, &temp_buff, dbg_buff);
+}
+
+void cudbg_fill_qdesc_num_and_size(const struct adapter *padap,
+				   u32 *num, u32 *size)
+{
+	u32 tot_entries = 0, tot_size = 0;
+
+	/* NIC TXQ, RXQ, FLQ, and CTRLQ */
+	tot_entries += MAX_ETH_QSETS * 3;
+	tot_entries += MAX_CTRL_QUEUES;
+
+	tot_size += MAX_ETH_QSETS * MAX_TXQ_ENTRIES * MAX_TXQ_DESC_SIZE;
+	tot_size += MAX_ETH_QSETS * MAX_RSPQ_ENTRIES * MAX_RXQ_DESC_SIZE;
+	tot_size += MAX_ETH_QSETS * MAX_RX_BUFFERS * MAX_FL_DESC_SIZE;
+	tot_size += MAX_CTRL_QUEUES * MAX_CTRL_TXQ_ENTRIES *
+		    MAX_CTRL_TXQ_DESC_SIZE;
+
+	/* FW_EVTQ and INTRQ */
+	tot_entries += INGQ_EXTRAS;
+	tot_size += INGQ_EXTRAS * MAX_RSPQ_ENTRIES * MAX_RXQ_DESC_SIZE;
+
+	/* PTP_TXQ */
+	tot_entries += 1;
+	tot_size += MAX_TXQ_ENTRIES * MAX_TXQ_DESC_SIZE;
+
+	/* ULD TXQ, RXQ, and FLQ */
+	tot_entries += CXGB4_TX_MAX * MAX_OFLD_QSETS;
+	tot_entries += CXGB4_ULD_MAX * MAX_ULD_QSETS * 2;
+
+	tot_size += CXGB4_TX_MAX * MAX_OFLD_QSETS * MAX_TXQ_ENTRIES *
+		    MAX_TXQ_DESC_SIZE;
+	tot_size += CXGB4_ULD_MAX * MAX_ULD_QSETS * MAX_RSPQ_ENTRIES *
+		    MAX_RXQ_DESC_SIZE;
+	tot_size += CXGB4_ULD_MAX * MAX_ULD_QSETS * MAX_RX_BUFFERS *
+		    MAX_FL_DESC_SIZE;
+
+	/* ULD CIQ */
+	tot_entries += CXGB4_ULD_MAX * MAX_ULD_QSETS;
+	tot_size += CXGB4_ULD_MAX * MAX_ULD_QSETS * SGE_MAX_IQ_SIZE *
+		    MAX_RXQ_DESC_SIZE;
+
+	/* ETHOFLD TXQ, RXQ, and FLQ */
+	tot_entries += MAX_OFLD_QSETS * 3;
+	tot_size += MAX_OFLD_QSETS * MAX_TXQ_ENTRIES * MAX_TXQ_DESC_SIZE;
+
+	tot_size += sizeof(struct cudbg_ver_hdr) +
+		    sizeof(struct cudbg_qdesc_info) +
+		    sizeof(struct cudbg_qdesc_entry) * tot_entries;
+
+	if (num)
+		*num = tot_entries;
+
+	if (size)
+		*size = tot_size;
+}
+
+int cudbg_collect_qdesc(struct cudbg_init *pdbg_init,
+			struct cudbg_buffer *dbg_buff,
+			struct cudbg_error *cudbg_err)
+{
+	u32 num_queues = 0, tot_entries = 0, size = 0;
+	struct adapter *padap = pdbg_init->adap;
+	struct cudbg_buffer temp_buff = { 0 };
+	struct cudbg_qdesc_entry *qdesc_entry;
+	struct cudbg_qdesc_info *qdesc_info;
+	struct cudbg_ver_hdr *ver_hdr;
+	struct sge *s = &padap->sge;
+	u32 i, j, cur_off, tot_len;
+	u8 *data;
+	int rc;
+
+	cudbg_fill_qdesc_num_and_size(padap, &tot_entries, &size);
+	size = min_t(u32, size, CUDBG_DUMP_BUFF_SIZE);
+	tot_len = size;
+	data = kvzalloc(size, GFP_KERNEL);
+	if (!data)
+		return -ENOMEM;
+
+	ver_hdr = (struct cudbg_ver_hdr *)data;
+	ver_hdr->signature = CUDBG_ENTITY_SIGNATURE;
+	ver_hdr->revision = CUDBG_QDESC_REV;
+	ver_hdr->size = sizeof(struct cudbg_qdesc_info);
+	size -= sizeof(*ver_hdr);
+
+	qdesc_info = (struct cudbg_qdesc_info *)(data +
+						 sizeof(*ver_hdr));
+	size -= sizeof(*qdesc_info);
+	qdesc_entry = (struct cudbg_qdesc_entry *)qdesc_info->data;
+
+#define QDESC_GET(q, desc, type, label) do { \
+	if (size <= 0) { \
+		goto label; \
+	} \
+	if (desc) { \
+		cudbg_fill_qdesc_##q(q, type, qdesc_entry); \
+		size -= sizeof(*qdesc_entry) + qdesc_entry->data_size; \
+		num_queues++; \
+		qdesc_entry = cudbg_next_qdesc(qdesc_entry); \
+	} \
+} while (0)
+
+#define QDESC_GET_TXQ(q, type, label) do { \
+	struct sge_txq *txq = (struct sge_txq *)q; \
+	QDESC_GET(txq, txq->desc, type, label); \
+} while (0)
+
+#define QDESC_GET_RXQ(q, type, label) do { \
+	struct sge_rspq *rxq = (struct sge_rspq *)q; \
+	QDESC_GET(rxq, rxq->desc, type, label); \
+} while (0)
+
+#define QDESC_GET_FLQ(q, type, label) do { \
+	struct sge_fl *flq = (struct sge_fl *)q; \
+	QDESC_GET(flq, flq->desc, type, label); \
+} while (0)
+
+	/* NIC TXQ */
+	for (i = 0; i < s->ethqsets; i++)
+		QDESC_GET_TXQ(&s->ethtxq[i].q, CUDBG_QTYPE_NIC_TXQ, out);
+
+	/* NIC RXQ */
+	for (i = 0; i < s->ethqsets; i++)
+		QDESC_GET_RXQ(&s->ethrxq[i].rspq, CUDBG_QTYPE_NIC_RXQ, out);
+
+	/* NIC FLQ */
+	for (i = 0; i < s->ethqsets; i++)
+		QDESC_GET_FLQ(&s->ethrxq[i].fl, CUDBG_QTYPE_NIC_FLQ, out);
+
+	/* NIC CTRLQ */
+	for (i = 0; i < padap->params.nports; i++)
+		QDESC_GET_TXQ(&s->ctrlq[i].q, CUDBG_QTYPE_CTRLQ, out);
+
+	/* FW_EVTQ */
+	QDESC_GET_RXQ(&s->fw_evtq, CUDBG_QTYPE_FWEVTQ, out);
+
+	/* INTRQ */
+	QDESC_GET_RXQ(&s->intrq, CUDBG_QTYPE_INTRQ, out);
+
+	/* PTP_TXQ */
+	QDESC_GET_TXQ(&s->ptptxq.q, CUDBG_QTYPE_PTP_TXQ, out);
+
+	/* ULD Queues */
+	mutex_lock(&uld_mutex);
+
+	if (s->uld_txq_info) {
+		struct sge_uld_txq_info *utxq;
+
+		/* ULD TXQ */
+		for (j = 0; j < CXGB4_TX_MAX; j++) {
+			if (!s->uld_txq_info[j])
+				continue;
+
+			utxq = s->uld_txq_info[j];
+			for (i = 0; i < utxq->ntxq; i++)
+				QDESC_GET_TXQ(&utxq->uldtxq[i].q,
+					      cudbg_uld_txq_to_qtype(j),
+					      out_unlock_uld);
+		}
+	}
+
+	if (s->uld_rxq_info) {
+		struct sge_uld_rxq_info *urxq;
+		u32 base;
+
+		/* ULD RXQ */
+		for (j = 0; j < CXGB4_ULD_MAX; j++) {
+			if (!s->uld_rxq_info[j])
+				continue;
+
+			urxq = s->uld_rxq_info[j];
+			for (i = 0; i < urxq->nrxq; i++)
+				QDESC_GET_RXQ(&urxq->uldrxq[i].rspq,
+					      cudbg_uld_rxq_to_qtype(j),
+					      out_unlock_uld);
+		}
+
+		/* ULD FLQ */
+		for (j = 0; j < CXGB4_ULD_MAX; j++) {
+			if (!s->uld_rxq_info[j])
+				continue;
+
+			urxq = s->uld_rxq_info[j];
+			for (i = 0; i < urxq->nrxq; i++)
+				QDESC_GET_FLQ(&urxq->uldrxq[i].fl,
+					      cudbg_uld_flq_to_qtype(j),
+					      out_unlock_uld);
+		}
+
+		/* ULD CIQ */
+		for (j = 0; j < CXGB4_ULD_MAX; j++) {
+			if (!s->uld_rxq_info[j])
+				continue;
+
+			urxq = s->uld_rxq_info[j];
+			base = urxq->nrxq;
+			for (i = 0; i < urxq->nciq; i++)
+				QDESC_GET_RXQ(&urxq->uldrxq[base + i].rspq,
+					      cudbg_uld_ciq_to_qtype(j),
+					      out_unlock_uld);
+		}
+	}
+	mutex_unlock(&uld_mutex);
+
+	if (!padap->tc_mqprio)
+		goto out;
+
+	mutex_lock(&padap->tc_mqprio->mqprio_mutex);
+	/* ETHOFLD TXQ */
+	if (s->eohw_txq)
+		for (i = 0; i < s->eoqsets; i++)
+			QDESC_GET_TXQ(&s->eohw_txq[i].q,
+				      CUDBG_QTYPE_ETHOFLD_TXQ, out_unlock_mqprio);
+
+	/* ETHOFLD RXQ and FLQ */
+	if (s->eohw_rxq) {
+		for (i = 0; i < s->eoqsets; i++)
+			QDESC_GET_RXQ(&s->eohw_rxq[i].rspq,
+				      CUDBG_QTYPE_ETHOFLD_RXQ, out_unlock_mqprio);
+
+		for (i = 0; i < s->eoqsets; i++)
+			QDESC_GET_FLQ(&s->eohw_rxq[i].fl,
+				      CUDBG_QTYPE_ETHOFLD_FLQ, out_unlock_mqprio);
+	}
+
+out_unlock_mqprio:
+	mutex_unlock(&padap->tc_mqprio->mqprio_mutex);
+
+out:
+	qdesc_info->qdesc_entry_size = sizeof(*qdesc_entry);
+	qdesc_info->num_queues = num_queues;
+	cur_off = 0;
+	while (tot_len) {
+		u32 chunk_size = min_t(u32, tot_len, CUDBG_CHUNK_SIZE);
+
+		rc = cudbg_get_buff(pdbg_init, dbg_buff, chunk_size,
+				    &temp_buff);
+		if (rc) {
+			cudbg_err->sys_warn = CUDBG_STATUS_PARTIAL_DATA;
+			goto out_free;
+		}
+
+		memcpy(temp_buff.data, data + cur_off, chunk_size);
+		tot_len -= chunk_size;
+		cur_off += chunk_size;
+		rc = cudbg_write_and_release_buff(pdbg_init, &temp_buff,
+						  dbg_buff);
+		if (rc) {
+			cudbg_put_buff(pdbg_init, &temp_buff);
+			cudbg_err->sys_warn = CUDBG_STATUS_PARTIAL_DATA;
+			goto out_free;
+		}
+	}
+
+out_free:
+	kvfree(data);
+
+#undef QDESC_GET_FLQ
+#undef QDESC_GET_RXQ
+#undef QDESC_GET_TXQ
+#undef QDESC_GET
+
+	return rc;
+
+out_unlock_uld:
+	mutex_unlock(&uld_mutex);
+	goto out;
+}
+
+int cudbg_collect_flash(struct cudbg_init *pdbg_init,
+			struct cudbg_buffer *dbg_buff,
+			struct cudbg_error *cudbg_err)
+{
+	struct adapter *padap = pdbg_init->adap;
+	u32 count = padap->params.sf_size, n;
+	struct cudbg_buffer temp_buff = {0};
+	u32 addr, i;
+	int rc;
+
+	addr = FLASH_EXP_ROM_START;
+
+	for (i = 0; i < count; i += SF_PAGE_SIZE) {
+		n = min_t(u32, count - i, SF_PAGE_SIZE);
+
+		rc = cudbg_get_buff(pdbg_init, dbg_buff, n, &temp_buff);
+		if (rc) {
+			cudbg_err->sys_warn = CUDBG_STATUS_PARTIAL_DATA;
+			goto out;
+		}
+		rc = t4_read_flash(padap, addr, n, (u32 *)temp_buff.data, 0);
+		if (rc)
+			goto out;
+
+		addr += (n * 4);
+		rc = cudbg_write_and_release_buff(pdbg_init, &temp_buff,
+						  dbg_buff);
+		if (rc) {
+			cudbg_err->sys_warn = CUDBG_STATUS_PARTIAL_DATA;
+			goto out;
+		}
+	}
+
+out:
+	return rc;
+}
diff --git a/drivers/net/ethernet/chelsio/cxgb4/cudbg_lib.h b/drivers/net/ethernet/chelsio/cxgb4/cudbg_lib.h
new file mode 100644
index 000000000000..d6d6cd298930
--- /dev/null
+++ b/drivers/net/ethernet/chelsio/cxgb4/cudbg_lib.h
@@ -0,0 +1,283 @@
+/* SPDX-License-Identifier: GPL-2.0-only */
+/*
+ *  Copyright (C) 2017 Chelsio Communications.  All rights reserved.
+ */
+
+#ifndef __CUDBG_LIB_H__
+#define __CUDBG_LIB_H__
+
+int cudbg_collect_reg_dump(struct cudbg_init *pdbg_init,
+			   struct cudbg_buffer *dbg_buff,
+			   struct cudbg_error *cudbg_err);
+int cudbg_collect_fw_devlog(struct cudbg_init *pdbg_init,
+			    struct cudbg_buffer *dbg_buff,
+			    struct cudbg_error *cudbg_err);
+int cudbg_collect_cim_la(struct cudbg_init *pdbg_init,
+			 struct cudbg_buffer *dbg_buff,
+			 struct cudbg_error *cudbg_err);
+int cudbg_collect_cim_ma_la(struct cudbg_init *pdbg_init,
+			    struct cudbg_buffer *dbg_buff,
+			    struct cudbg_error *cudbg_err);
+int cudbg_collect_cim_qcfg(struct cudbg_init *pdbg_init,
+			   struct cudbg_buffer *dbg_buff,
+			   struct cudbg_error *cudbg_err);
+int cudbg_collect_cim_ibq_tp0(struct cudbg_init *pdbg_init,
+			      struct cudbg_buffer *dbg_buff,
+			      struct cudbg_error *cudbg_err);
+int cudbg_collect_cim_ibq_tp1(struct cudbg_init *pdbg_init,
+			      struct cudbg_buffer *dbg_buff,
+			      struct cudbg_error *cudbg_err);
+int cudbg_collect_cim_ibq_ulp(struct cudbg_init *pdbg_init,
+			      struct cudbg_buffer *dbg_buff,
+			      struct cudbg_error *cudbg_err);
+int cudbg_collect_cim_ibq_sge0(struct cudbg_init *pdbg_init,
+			       struct cudbg_buffer *dbg_buff,
+			       struct cudbg_error *cudbg_err);
+int cudbg_collect_cim_ibq_sge1(struct cudbg_init *pdbg_init,
+			       struct cudbg_buffer *dbg_buff,
+			       struct cudbg_error *cudbg_err);
+int cudbg_collect_cim_ibq_ncsi(struct cudbg_init *pdbg_init,
+			       struct cudbg_buffer *dbg_buff,
+			       struct cudbg_error *cudbg_err);
+int cudbg_collect_cim_obq_ulp0(struct cudbg_init *pdbg_init,
+			       struct cudbg_buffer *dbg_buff,
+			       struct cudbg_error *cudbg_err);
+int cudbg_collect_cim_obq_ulp1(struct cudbg_init *pdbg_init,
+			       struct cudbg_buffer *dbg_buff,
+			       struct cudbg_error *cudbg_err);
+int cudbg_collect_cim_obq_ulp2(struct cudbg_init *pdbg_init,
+			       struct cudbg_buffer *dbg_buff,
+			       struct cudbg_error *cudbg_err);
+int cudbg_collect_cim_obq_ulp3(struct cudbg_init *pdbg_init,
+			       struct cudbg_buffer *dbg_buff,
+			       struct cudbg_error *cudbg_err);
+int cudbg_collect_cim_obq_sge(struct cudbg_init *pdbg_init,
+			      struct cudbg_buffer *dbg_buff,
+			      struct cudbg_error *cudbg_err);
+int cudbg_collect_cim_obq_ncsi(struct cudbg_init *pdbg_init,
+			       struct cudbg_buffer *dbg_buff,
+			       struct cudbg_error *cudbg_err);
+int cudbg_collect_edc0_meminfo(struct cudbg_init *pdbg_init,
+			       struct cudbg_buffer *dbg_buff,
+			       struct cudbg_error *cudbg_err);
+int cudbg_collect_edc1_meminfo(struct cudbg_init *pdbg_init,
+			       struct cudbg_buffer *dbg_buff,
+			       struct cudbg_error *cudbg_err);
+int cudbg_collect_mc0_meminfo(struct cudbg_init *pdbg_init,
+			      struct cudbg_buffer *dbg_buff,
+			      struct cudbg_error *cudbg_err);
+int cudbg_collect_mc1_meminfo(struct cudbg_init *pdbg_init,
+			      struct cudbg_buffer *dbg_buff,
+			      struct cudbg_error *cudbg_err);
+int cudbg_collect_rss(struct cudbg_init *pdbg_init,
+		      struct cudbg_buffer *dbg_buff,
+		      struct cudbg_error *cudbg_err);
+int cudbg_collect_rss_vf_config(struct cudbg_init *pdbg_init,
+				struct cudbg_buffer *dbg_buff,
+				struct cudbg_error *cudbg_err);
+int cudbg_collect_tp_indirect(struct cudbg_init *pdbg_init,
+			      struct cudbg_buffer *dbg_buff,
+			      struct cudbg_error *cudbg_err);
+int cudbg_collect_path_mtu(struct cudbg_init *pdbg_init,
+			   struct cudbg_buffer *dbg_buff,
+			   struct cudbg_error *cudbg_err);
+int cudbg_collect_pm_stats(struct cudbg_init *pdbg_init,
+			   struct cudbg_buffer *dbg_buff,
+			   struct cudbg_error *cudbg_err);
+int cudbg_collect_hw_sched(struct cudbg_init *pdbg_init,
+			   struct cudbg_buffer *dbg_buff,
+			   struct cudbg_error *cudbg_err);
+int cudbg_collect_sge_indirect(struct cudbg_init *pdbg_init,
+			       struct cudbg_buffer *dbg_buff,
+			       struct cudbg_error *cudbg_err);
+int cudbg_collect_ulprx_la(struct cudbg_init *pdbg_init,
+			   struct cudbg_buffer *dbg_buff,
+			   struct cudbg_error *cudbg_err);
+int cudbg_collect_tp_la(struct cudbg_init *pdbg_init,
+			struct cudbg_buffer *dbg_buff,
+			struct cudbg_error *cudbg_err);
+int cudbg_collect_meminfo(struct cudbg_init *pdbg_init,
+			  struct cudbg_buffer *dbg_buff,
+			  struct cudbg_error *cudbg_err);
+int cudbg_collect_cim_pif_la(struct cudbg_init *pdbg_init,
+			     struct cudbg_buffer *dbg_buff,
+			     struct cudbg_error *cudbg_err);
+int cudbg_collect_clk_info(struct cudbg_init *pdbg_init,
+			   struct cudbg_buffer *dbg_buff,
+			   struct cudbg_error *cudbg_err);
+int cudbg_collect_obq_sge_rx_q0(struct cudbg_init *pdbg_init,
+				struct cudbg_buffer *dbg_buff,
+				struct cudbg_error *cudbg_err);
+int cudbg_collect_obq_sge_rx_q1(struct cudbg_init *pdbg_init,
+				struct cudbg_buffer *dbg_buff,
+				struct cudbg_error *cudbg_err);
+int cudbg_collect_pcie_indirect(struct cudbg_init *pdbg_init,
+				struct cudbg_buffer *dbg_buff,
+				struct cudbg_error *cudbg_err);
+int cudbg_collect_pm_indirect(struct cudbg_init *pdbg_init,
+			      struct cudbg_buffer *dbg_buff,
+			      struct cudbg_error *cudbg_err);
+int cudbg_collect_tid(struct cudbg_init *pdbg_init,
+		      struct cudbg_buffer *dbg_buff,
+		      struct cudbg_error *cudbg_err);
+int cudbg_collect_pcie_config(struct cudbg_init *pdbg_init,
+			      struct cudbg_buffer *dbg_buff,
+			      struct cudbg_error *cudbg_err);
+int cudbg_collect_dump_context(struct cudbg_init *pdbg_init,
+			       struct cudbg_buffer *dbg_buff,
+			       struct cudbg_error *cudbg_err);
+int cudbg_collect_mps_tcam(struct cudbg_init *pdbg_init,
+			   struct cudbg_buffer *dbg_buff,
+			   struct cudbg_error *cudbg_err);
+int cudbg_collect_vpd_data(struct cudbg_init *pdbg_init,
+			   struct cudbg_buffer *dbg_buff,
+			   struct cudbg_error *cudbg_err);
+int cudbg_collect_le_tcam(struct cudbg_init *pdbg_init,
+			  struct cudbg_buffer *dbg_buff,
+			  struct cudbg_error *cudbg_err);
+int cudbg_collect_cctrl(struct cudbg_init *pdbg_init,
+			struct cudbg_buffer *dbg_buff,
+			struct cudbg_error *cudbg_err);
+int cudbg_collect_ma_indirect(struct cudbg_init *pdbg_init,
+			      struct cudbg_buffer *dbg_buff,
+			      struct cudbg_error *cudbg_err);
+int cudbg_collect_ulptx_la(struct cudbg_init *pdbg_init,
+			   struct cudbg_buffer *dbg_buff,
+			   struct cudbg_error *cudbg_err);
+int cudbg_collect_up_cim_indirect(struct cudbg_init *pdbg_init,
+				  struct cudbg_buffer *dbg_buff,
+				  struct cudbg_error *cudbg_err);
+int cudbg_collect_pbt_tables(struct cudbg_init *pdbg_init,
+			     struct cudbg_buffer *dbg_buff,
+			     struct cudbg_error *cudbg_err);
+int cudbg_collect_mbox_log(struct cudbg_init *pdbg_init,
+			   struct cudbg_buffer *dbg_buff,
+			   struct cudbg_error *cudbg_err);
+int cudbg_collect_hma_indirect(struct cudbg_init *pdbg_init,
+			       struct cudbg_buffer *dbg_buff,
+			       struct cudbg_error *cudbg_err);
+int cudbg_collect_hma_meminfo(struct cudbg_init *pdbg_init,
+			      struct cudbg_buffer *dbg_buff,
+			      struct cudbg_error *cudbg_err);
+int cudbg_collect_qdesc(struct cudbg_init *pdbg_init,
+			struct cudbg_buffer *dbg_buff,
+			struct cudbg_error *cudbg_err);
+int cudbg_collect_flash(struct cudbg_init *pdbg_init,
+			struct cudbg_buffer *dbg_buff,
+			struct cudbg_error *cudbg_err);
+
+u32 cudbg_get_entity_length(struct adapter *adap, u32 entity);
+struct cudbg_entity_hdr *cudbg_get_entity_hdr(void *outbuf, int i);
+void cudbg_align_debug_buffer(struct cudbg_buffer *dbg_buff,
+			      struct cudbg_entity_hdr *entity_hdr);
+u32 cudbg_cim_obq_size(struct adapter *padap, int qid);
+int cudbg_dump_context_size(struct adapter *padap);
+
+int cudbg_fill_meminfo(struct adapter *padap,
+		       struct cudbg_meminfo *meminfo_buff);
+void cudbg_fill_le_tcam_info(struct adapter *padap,
+			     struct cudbg_tcam *tcam_region);
+void cudbg_fill_qdesc_num_and_size(const struct adapter *padap,
+				   u32 *num, u32 *size);
+
+static inline u32 cudbg_uld_txq_to_qtype(u32 uld)
+{
+	switch (uld) {
+	case CXGB4_TX_OFLD:
+		return CUDBG_QTYPE_OFLD_TXQ;
+	case CXGB4_TX_CRYPTO:
+		return CUDBG_QTYPE_CRYPTO_TXQ;
+	}
+
+	return CUDBG_QTYPE_UNKNOWN;
+}
+
+static inline u32 cudbg_uld_rxq_to_qtype(u32 uld)
+{
+	switch (uld) {
+	case CXGB4_ULD_RDMA:
+		return CUDBG_QTYPE_RDMA_RXQ;
+	case CXGB4_ULD_ISCSI:
+		return CUDBG_QTYPE_ISCSI_RXQ;
+	case CXGB4_ULD_ISCSIT:
+		return CUDBG_QTYPE_ISCSIT_RXQ;
+	case CXGB4_ULD_CRYPTO:
+		return CUDBG_QTYPE_CRYPTO_RXQ;
+	case CXGB4_ULD_TLS:
+		return CUDBG_QTYPE_TLS_RXQ;
+	}
+
+	return CUDBG_QTYPE_UNKNOWN;
+}
+
+static inline u32 cudbg_uld_flq_to_qtype(u32 uld)
+{
+	switch (uld) {
+	case CXGB4_ULD_RDMA:
+		return CUDBG_QTYPE_RDMA_FLQ;
+	case CXGB4_ULD_ISCSI:
+		return CUDBG_QTYPE_ISCSI_FLQ;
+	case CXGB4_ULD_ISCSIT:
+		return CUDBG_QTYPE_ISCSIT_FLQ;
+	case CXGB4_ULD_CRYPTO:
+		return CUDBG_QTYPE_CRYPTO_FLQ;
+	case CXGB4_ULD_TLS:
+		return CUDBG_QTYPE_TLS_FLQ;
+	}
+
+	return CUDBG_QTYPE_UNKNOWN;
+}
+
+static inline u32 cudbg_uld_ciq_to_qtype(u32 uld)
+{
+	switch (uld) {
+	case CXGB4_ULD_RDMA:
+		return CUDBG_QTYPE_RDMA_CIQ;
+	}
+
+	return CUDBG_QTYPE_UNKNOWN;
+}
+
+static inline void cudbg_fill_qdesc_txq(const struct sge_txq *txq,
+					enum cudbg_qdesc_qtype type,
+					struct cudbg_qdesc_entry *entry)
+{
+	entry->qtype = type;
+	entry->qid = txq->cntxt_id;
+	entry->desc_size = sizeof(struct tx_desc);
+	entry->num_desc = txq->size;
+	entry->data_size = txq->size * sizeof(struct tx_desc);
+	memcpy(entry->data, txq->desc, entry->data_size);
+}
+
+static inline void cudbg_fill_qdesc_rxq(const struct sge_rspq *rxq,
+					enum cudbg_qdesc_qtype type,
+					struct cudbg_qdesc_entry *entry)
+{
+	entry->qtype = type;
+	entry->qid = rxq->cntxt_id;
+	entry->desc_size = rxq->iqe_len;
+	entry->num_desc = rxq->size;
+	entry->data_size = rxq->size * rxq->iqe_len;
+	memcpy(entry->data, rxq->desc, entry->data_size);
+}
+
+static inline void cudbg_fill_qdesc_flq(const struct sge_fl *flq,
+					enum cudbg_qdesc_qtype type,
+					struct cudbg_qdesc_entry *entry)
+{
+	entry->qtype = type;
+	entry->qid = flq->cntxt_id;
+	entry->desc_size = sizeof(__be64);
+	entry->num_desc = flq->size;
+	entry->data_size = flq->size * sizeof(__be64);
+	memcpy(entry->data, flq->desc, entry->data_size);
+}
+
+static inline
+struct cudbg_qdesc_entry *cudbg_next_qdesc(struct cudbg_qdesc_entry *e)
+{
+	return (struct cudbg_qdesc_entry *)
+	       ((u8 *)e + sizeof(*e) + e->data_size);
+}
+#endif /* __CUDBG_LIB_H__ */
diff --git a/drivers/net/ethernet/chelsio/cxgb4/cudbg_lib_common.h b/drivers/net/ethernet/chelsio/cxgb4/cudbg_lib_common.h
new file mode 100644
index 000000000000..9fac777b0b24
--- /dev/null
+++ b/drivers/net/ethernet/chelsio/cxgb4/cudbg_lib_common.h
@@ -0,0 +1,77 @@
+/* SPDX-License-Identifier: GPL-2.0-only */
+/*
+ *  Copyright (C) 2017 Chelsio Communications.  All rights reserved.
+ */
+
+#ifndef __CUDBG_LIB_COMMON_H__
+#define __CUDBG_LIB_COMMON_H__
+
+#define CUDBG_SIGNATURE 67856866 /* CUDB in ascii */
+
+enum cudbg_dump_type {
+	CUDBG_DUMP_TYPE_MINI = 1,
+};
+
+enum cudbg_compression_type {
+	CUDBG_COMPRESSION_NONE = 1,
+	CUDBG_COMPRESSION_ZLIB,
+};
+
+struct cudbg_hdr {
+	u32 signature;
+	u32 hdr_len;
+	u16 major_ver;
+	u16 minor_ver;
+	u32 data_len;
+	u32 hdr_flags;
+	u16 max_entities;
+	u8 chip_ver;
+	u8 dump_type:3;
+	u8 reserved1:1;
+	u8 compress_type:4;
+	u32 reserved[8];
+};
+
+struct cudbg_entity_hdr {
+	u32 entity_type;
+	u32 start_offset;
+	u32 size;
+	int hdr_flags;
+	u32 sys_warn;
+	u32 sys_err;
+	u8 num_pad;
+	u8 flag;             /* bit 0 is used to indicate ext data */
+	u8 reserved1[2];
+	u32 next_ext_offset; /* pointer to next extended entity meta data */
+	u32 reserved[5];
+};
+
+struct cudbg_ver_hdr {
+	u32 signature;
+	u16 revision;
+	u16 size;
+};
+
+struct cudbg_buffer {
+	u32 size;
+	u32 offset;
+	char *data;
+};
+
+struct cudbg_error {
+	int sys_err;
+	int sys_warn;
+	int app_err;
+};
+
+#define CDUMP_MAX_COMP_BUF_SIZE ((64 * 1024) - 1)
+#define CUDBG_CHUNK_SIZE ((CDUMP_MAX_COMP_BUF_SIZE / 1024) * 1024)
+
+int cudbg_get_buff(struct cudbg_init *pdbg_init,
+		   struct cudbg_buffer *pdbg_buff, u32 size,
+		   struct cudbg_buffer *pin_buff);
+void cudbg_put_buff(struct cudbg_init *pdbg_init,
+		    struct cudbg_buffer *pin_buff);
+void cudbg_update_buff(struct cudbg_buffer *pin_buff,
+		       struct cudbg_buffer *pout_buff);
+#endif /* __CUDBG_LIB_COMMON_H__ */
diff --git a/drivers/net/ethernet/chelsio/cxgb4/cudbg_zlib.c b/drivers/net/ethernet/chelsio/cxgb4/cudbg_zlib.c
new file mode 100644
index 000000000000..aad55fb3585f
--- /dev/null
+++ b/drivers/net/ethernet/chelsio/cxgb4/cudbg_zlib.c
@@ -0,0 +1,70 @@
+// SPDX-License-Identifier: GPL-2.0-only
+/*
+ *  Copyright (C) 2018 Chelsio Communications.  All rights reserved.
+ */
+
+#include <linux/zlib.h>
+
+#include "cxgb4.h"
+#include "cudbg_if.h"
+#include "cudbg_lib_common.h"
+#include "cudbg_zlib.h"
+
+static int cudbg_get_compress_hdr(struct cudbg_buffer *pdbg_buff,
+				  struct cudbg_buffer *pin_buff)
+{
+	if (pdbg_buff->offset + sizeof(struct cudbg_compress_hdr) >
+	    pdbg_buff->size)
+		return CUDBG_STATUS_NO_MEM;
+
+	pin_buff->data = (char *)pdbg_buff->data + pdbg_buff->offset;
+	pin_buff->offset = 0;
+	pin_buff->size = sizeof(struct cudbg_compress_hdr);
+	pdbg_buff->offset += sizeof(struct cudbg_compress_hdr);
+	return 0;
+}
+
+int cudbg_compress_buff(struct cudbg_init *pdbg_init,
+			struct cudbg_buffer *pin_buff,
+			struct cudbg_buffer *pout_buff)
+{
+	struct cudbg_buffer temp_buff = { 0 };
+	struct z_stream_s compress_stream;
+	struct cudbg_compress_hdr *c_hdr;
+	int rc;
+
+	/* Write compression header to output buffer before compression */
+	rc = cudbg_get_compress_hdr(pout_buff, &temp_buff);
+	if (rc)
+		return rc;
+
+	c_hdr = (struct cudbg_compress_hdr *)temp_buff.data;
+	c_hdr->compress_id = CUDBG_ZLIB_COMPRESS_ID;
+
+	memset(&compress_stream, 0, sizeof(struct z_stream_s));
+	compress_stream.workspace = pdbg_init->workspace;
+	rc = zlib_deflateInit2(&compress_stream, Z_DEFAULT_COMPRESSION,
+			       Z_DEFLATED, CUDBG_ZLIB_WIN_BITS,
+			       CUDBG_ZLIB_MEM_LVL, Z_DEFAULT_STRATEGY);
+	if (rc != Z_OK)
+		return CUDBG_SYSTEM_ERROR;
+
+	compress_stream.next_in = pin_buff->data;
+	compress_stream.avail_in = pin_buff->size;
+	compress_stream.next_out = pout_buff->data + pout_buff->offset;
+	compress_stream.avail_out = pout_buff->size - pout_buff->offset;
+
+	rc = zlib_deflate(&compress_stream, Z_FINISH);
+	if (rc != Z_STREAM_END)
+		return CUDBG_SYSTEM_ERROR;
+
+	rc = zlib_deflateEnd(&compress_stream);
+	if (rc != Z_OK)
+		return CUDBG_SYSTEM_ERROR;
+
+	c_hdr->compress_size = compress_stream.total_out;
+	c_hdr->decompress_size = pin_buff->size;
+	pout_buff->offset += compress_stream.total_out;
+
+	return 0;
+}
diff --git a/drivers/net/ethernet/chelsio/cxgb4/cudbg_zlib.h b/drivers/net/ethernet/chelsio/cxgb4/cudbg_zlib.h
new file mode 100644
index 000000000000..f6d83289ff1e
--- /dev/null
+++ b/drivers/net/ethernet/chelsio/cxgb4/cudbg_zlib.h
@@ -0,0 +1,31 @@
+/* SPDX-License-Identifier: GPL-2.0-only */
+/*
+ *  Copyright (C) 2018 Chelsio Communications.  All rights reserved.
+ */
+
+#ifndef __CUDBG_ZLIB_H__
+#define __CUDBG_ZLIB_H__
+
+#include <linux/zlib.h>
+
+#define CUDBG_ZLIB_COMPRESS_ID 17
+#define CUDBG_ZLIB_WIN_BITS 12
+#define CUDBG_ZLIB_MEM_LVL 4
+
+struct cudbg_compress_hdr {
+	u32 compress_id;
+	u64 decompress_size;
+	u64 compress_size;
+	u64 rsvd[32];
+};
+
+static inline int cudbg_get_workspace_size(void)
+{
+	return zlib_deflate_workspacesize(CUDBG_ZLIB_WIN_BITS,
+					  CUDBG_ZLIB_MEM_LVL);
+}
+
+int cudbg_compress_buff(struct cudbg_init *pdbg_init,
+			struct cudbg_buffer *pin_buff,
+			struct cudbg_buffer *pout_buff);
+#endif /* __CUDBG_ZLIB_H__ */
diff --git a/drivers/net/ethernet/chelsio/cxgb4/cxgb4.h b/drivers/net/ethernet/chelsio/cxgb4/cxgb4.h
index 1cf3e2f89fc1..0d85198fb03d 100644
--- a/drivers/net/ethernet/chelsio/cxgb4/cxgb4.h
+++ b/drivers/net/ethernet/chelsio/cxgb4/cxgb4.h
@@ -39,6 +39,7 @@
 
 #include <linux/bitops.h>
 #include <linux/cache.h>
+#include <linux/ethtool.h>
 #include <linux/interrupt.h>
 #include <linux/list.h>
 #include <linux/netdevice.h>
@@ -46,20 +47,43 @@
 #include <linux/spinlock.h>
 #include <linux/timer.h>
 #include <linux/vmalloc.h>
+#include <linux/rhashtable.h>
 #include <linux/etherdevice.h>
 #include <linux/net_tstamp.h>
+#include <linux/ptp_clock_kernel.h>
+#include <linux/ptp_classify.h>
+#include <linux/crash_dump.h>
+#include <linux/thermal.h>
 #include <asm/io.h>
 #include "t4_chip_type.h"
 #include "cxgb4_uld.h"
+#include "t4fw_api.h"
 
 #define CH_WARN(adap, fmt, ...) dev_warn(adap->pdev_dev, fmt, ## __VA_ARGS__)
 extern struct list_head adapter_list;
+extern struct list_head uld_list;
 extern struct mutex uld_mutex;
 
+/* Suspend an Ethernet Tx queue with fewer available descriptors than this.
+ * This is the same as calc_tx_descs() for a TSO packet with
+ * nr_frags == MAX_SKB_FRAGS.
+ */
+#define ETHTXQ_STOP_THRES \
+	(1 + DIV_ROUND_UP((3 * MAX_SKB_FRAGS) / 2 + (MAX_SKB_FRAGS & 1), 8))
+
+#define FW_PARAM_DEV(param) \
+	(FW_PARAMS_MNEM_V(FW_PARAMS_MNEM_DEV) | \
+	 FW_PARAMS_PARAM_X_V(FW_PARAMS_PARAM_DEV_##param))
+
+#define FW_PARAM_PFVF(param) \
+	(FW_PARAMS_MNEM_V(FW_PARAMS_MNEM_PFVF) | \
+	 FW_PARAMS_PARAM_X_V(FW_PARAMS_PARAM_PFVF_##param) |  \
+	 FW_PARAMS_PARAM_Y_V(0) | \
+	 FW_PARAMS_PARAM_Z_V(0))
+
 enum {
 	MAX_NPORTS	= 4,     /* max # of ports */
 	SERNUM_LEN	= 24,    /* Serial # length */
-	EC_LEN		= 16,    /* E/C length */
 	ID_LEN		= 16,    /* ID length */
 	PN_LEN		= 16,    /* Part Number length */
 	MACADDR_LEN	= 12,    /* MAC Address length */
@@ -75,7 +99,8 @@ enum {
 	MEM_EDC1,
 	MEM_MC,
 	MEM_MC0 = MEM_MC,
-	MEM_MC1
+	MEM_MC1,
+	MEM_HMA,
 };
 
 enum {
@@ -102,18 +127,81 @@ enum dev_state {
 	DEV_STATE_ERR
 };
 
-enum {
+enum cc_pause {
 	PAUSE_RX      = 1 << 0,
 	PAUSE_TX      = 1 << 1,
 	PAUSE_AUTONEG = 1 << 2
 };
 
-enum {
+enum cc_fec {
 	FEC_AUTO      = 1 << 0,	 /* IEEE 802.3 "automatic" */
 	FEC_RS        = 1 << 1,  /* Reed-Solomon */
 	FEC_BASER_RS  = 1 << 2   /* BaseR/Reed-Solomon */
 };
 
+enum {
+	CXGB4_ETHTOOL_FLASH_FW = 1,
+	CXGB4_ETHTOOL_FLASH_PHY = 2,
+	CXGB4_ETHTOOL_FLASH_BOOT = 3,
+	CXGB4_ETHTOOL_FLASH_BOOTCFG = 4
+};
+
+enum cxgb4_netdev_tls_ops {
+	CXGB4_TLSDEV_OPS  = 1,
+	CXGB4_XFRMDEV_OPS
+};
+
+struct cxgb4_bootcfg_data {
+	__le16 signature;
+	__u8 reserved[2];
+};
+
+struct cxgb4_pcir_data {
+	__le32 signature;	/* Signature. The string "PCIR" */
+	__le16 vendor_id;	/* Vendor Identification */
+	__le16 device_id;	/* Device Identification */
+	__u8 vital_product[2];	/* Pointer to Vital Product Data */
+	__u8 length[2];		/* PCIR Data Structure Length */
+	__u8 revision;		/* PCIR Data Structure Revision */
+	__u8 class_code[3];	/* Class Code */
+	__u8 image_length[2];	/* Image Length. Multiple of 512B */
+	__u8 code_revision[2];	/* Revision Level of Code/Data */
+	__u8 code_type;
+	__u8 indicator;
+	__u8 reserved[2];
+};
+
+/* BIOS boot headers */
+struct cxgb4_pci_exp_rom_header {
+	__le16 signature;	/* ROM Signature. Should be 0xaa55 */
+	__u8 reserved[22];	/* Reserved per processor Architecture data */
+	__le16 pcir_offset;	/* Offset to PCI Data Structure */
+};
+
+/* Legacy PCI Expansion ROM Header */
+struct legacy_pci_rom_hdr {
+	__u8 signature[2];	/* ROM Signature. Should be 0xaa55 */
+	__u8 size512;		/* Current Image Size in units of 512 bytes */
+	__u8 initentry_point[4];
+	__u8 cksum;		/* Checksum computed on the entire Image */
+	__u8 reserved[16];	/* Reserved */
+	__le16 pcir_offset;	/* Offset to PCI Data Struture */
+};
+
+#define CXGB4_HDR_CODE1 0x00
+#define CXGB4_HDR_CODE2 0x03
+#define CXGB4_HDR_INDI 0x80
+
+/* BOOT constants */
+enum {
+	BOOT_CFG_SIG = 0x4243,
+	BOOT_SIZE_INC = 512,
+	BOOT_SIGNATURE = 0xaa55,
+	BOOT_MIN_SIZE = sizeof(struct cxgb4_pci_exp_rom_header),
+	BOOT_MAX_SIZE = 1024 * BOOT_SIZE_INC,
+	PCIR_SIGNATURE = 0x52494350
+};
+
 struct port_stats {
 	u64 tx_octets;            /* total # of octets in good frames */
 	u64 tx_frames;            /* all good frames */
@@ -267,6 +355,7 @@ struct tp_params {
 	unsigned short tx_modq[NCHAN];	/* channel to modulation queue map */
 
 	u32 vlan_pri_map;               /* cached TP_VLAN_PRI_MAP */
+	u32 filter_mask;
 	u32 ingress_config;             /* cached TP_INGRESS_CONFIG */
 
 	/* cached TP_OUT_CONFIG compressed error vector
@@ -285,21 +374,43 @@ struct tp_params {
 	 * places we store their offsets here, or a -1 if the field isn't
 	 * present.
 	 */
-	int vlan_shift;
-	int vnic_shift;
+	int fcoe_shift;
 	int port_shift;
+	int vnic_shift;
+	int vlan_shift;
+	int tos_shift;
 	int protocol_shift;
+	int ethertype_shift;
+	int macmatch_shift;
+	int matchtype_shift;
+	int frag_shift;
+
+	u64 hash_filter_mask;
 };
 
 struct vpd_params {
 	unsigned int cclk;
-	u8 ec[EC_LEN + 1];
 	u8 sn[SERNUM_LEN + 1];
 	u8 id[ID_LEN + 1];
 	u8 pn[PN_LEN + 1];
 	u8 na[MACADDR_LEN + 1];
 };
 
+/* Maximum resources provisioned for a PCI PF.
+ */
+struct pf_resources {
+	unsigned int nvi;		/* N virtual interfaces */
+	unsigned int neq;		/* N egress Qs */
+	unsigned int nethctrl;		/* N egress ETH or CTRL Qs */
+	unsigned int niqflint;		/* N ingress Qs/w free list(s) & intr */
+	unsigned int niq;		/* N ingress Qs */
+	unsigned int tc;		/* PCI-E traffic class */
+	unsigned int pmask;		/* port access rights mask */
+	unsigned int nexactf;		/* N exact MPS filters */
+	unsigned int r_caps;		/* read capabilities */
+	unsigned int wx_caps;		/* write/execute capabilities */
+};
+
 struct pci_params {
 	unsigned char speed;
 	unsigned char width;
@@ -326,6 +437,7 @@ struct adapter_params {
 	struct sge_params sge;
 	struct tp_params  tp;
 	struct vpd_params vpd;
+	struct pf_resources pfres;
 	struct pci_params pci;
 	struct devlog_params devlog;
 	enum pcie_memwin drv_memwin;
@@ -334,12 +446,13 @@ struct adapter_params {
 
 	unsigned int sf_size;             /* serial flash size in bytes */
 	unsigned int sf_nsec;             /* # of flash sectors */
-	unsigned int sf_fw_start;         /* start of FW image in flash */
 
-	unsigned int fw_vers;
-	unsigned int bs_vers;		/* bootstrap version */
-	unsigned int tp_vers;
-	unsigned int er_vers;		/* expansion ROM version */
+	unsigned int fw_vers;		  /* firmware version */
+	unsigned int bs_vers;		  /* bootstrap version */
+	unsigned int tp_vers;		  /* TP microcode version */
+	unsigned int er_vers;		  /* expansion ROM version */
+	unsigned int scfg_vers;		  /* Serial Configuration version */
+	unsigned int vpd_vers;		  /* VPD Version */
 	u8 api_vers[7];
 
 	unsigned short mtus[NMTUS];
@@ -352,8 +465,10 @@ struct adapter_params {
 	struct arch_specific_params arch;  /* chip specific params */
 	unsigned char offload;
 	unsigned char crypto;		/* HW capability for crypto */
+	unsigned char ethofld;		/* QoS support */
 
 	unsigned char bypass;
+	unsigned char hash_filter;
 
 	unsigned int ofldq_wr_cred;
 	bool ulptx_memwrite_dsgl;          /* use of T5 DSGL allowed */
@@ -362,6 +477,16 @@ struct adapter_params {
 	unsigned int max_ordird_qp;       /* Max read depth per RDMA QP */
 	unsigned int max_ird_adapter;     /* Max read depth per adapter */
 	bool fr_nsmr_tpte_wr_support;	  /* FW support for FR_NSMR_TPTE_WR */
+	u8 fw_caps_support;		/* 32-bit Port Capabilities */
+	bool filter2_wr_support;	/* FW support for FILTER2_WR */
+	unsigned int viid_smt_extn_support:1; /* FW returns vin and smt index */
+
+	/* MPS Buffer Group Map[per Port].  Bit i is set if buffer group i is
+	 * used by the Port
+	 */
+	u8 mps_bg_map[MAX_NPORTS];	/* MPS Buffer Group Map */
+	bool write_w_imm_support;       /* FW supports WRITE_WITH_IMMEDIATE */
+	bool write_cmpl_support;        /* FW supports WRITE_CMPL */
 };
 
 /* State needed to monitor the forward progress of SGE Ingress DMA activities
@@ -403,8 +528,6 @@ static inline struct mbox_cmd *mbox_cmd_log_entry(struct mbox_cmd_log *log,
 	return &((struct mbox_cmd *)&(log)[1])[entry_idx];
 }
 
-#include "t4fw_api.h"
-
 #define FW_VERSION(chip) ( \
 		FW_HDR_FW_VER_MAJOR_G(chip##FW_VERSION_MAJOR) | \
 		FW_HDR_FW_VER_MINOR_G(chip##FW_VERSION_MINOR) | \
@@ -412,6 +535,12 @@ static inline struct mbox_cmd *mbox_cmd_log_entry(struct mbox_cmd_log *log,
 		FW_HDR_FW_VER_BUILD_G(chip##FW_VERSION_BUILD))
 #define FW_INTFVER(chip, intf) (FW_HDR_INTFVER_##intf)
 
+struct cxgb4_ethtool_lb_test {
+	struct completion completion;
+	int result;
+	int loopback;
+};
+
 struct fw_info {
 	u8 chip;
 	char *fs_name;
@@ -430,20 +559,45 @@ struct trace_params {
 	unsigned char port;
 };
 
+struct cxgb4_fw_data {
+	__be32 signature;
+	__u8 reserved[4];
+};
+
+/* Firmware Port Capabilities types. */
+
+typedef u16 fw_port_cap16_t;	/* 16-bit Port Capabilities integral value */
+typedef u32 fw_port_cap32_t;	/* 32-bit Port Capabilities integral value */
+
+enum fw_caps {
+	FW_CAPS_UNKNOWN	= 0,	/* 0'ed out initial state */
+	FW_CAPS16	= 1,	/* old Firmware: 16-bit Port Capabilities */
+	FW_CAPS32	= 2,	/* new Firmware: 32-bit Port Capabilities */
+};
+
 struct link_config {
-	unsigned short supported;        /* link capabilities */
-	unsigned short advertising;      /* advertised capabilities */
-	unsigned short lp_advertising;   /* peer advertised capabilities */
-	unsigned int   requested_speed;  /* speed user has requested */
-	unsigned int   speed;            /* actual link speed */
-	unsigned char  requested_fc;     /* flow control user has requested */
-	unsigned char  fc;               /* actual link flow control */
-	unsigned char  auto_fec;	 /* Forward Error Correction: */
-	unsigned char  requested_fec;	 /* "automatic" (IEEE 802.3), */
-	unsigned char  fec;		 /* requested, and actual in use */
+	fw_port_cap32_t pcaps;           /* link capabilities */
+	fw_port_cap32_t def_acaps;       /* default advertised capabilities */
+	fw_port_cap32_t acaps;           /* advertised capabilities */
+	fw_port_cap32_t lpacaps;         /* peer advertised capabilities */
+
+	fw_port_cap32_t speed_caps;      /* speed(s) user has requested */
+	unsigned int   speed;            /* actual link speed (Mb/s) */
+
+	enum cc_pause  requested_fc;     /* flow control user has requested */
+	enum cc_pause  fc;               /* actual link flow control */
+	enum cc_pause  advertised_fc;    /* actual advertised flow control */
+
+	enum cc_fec    requested_fec;	 /* Forward Error Correction: */
+	enum cc_fec    fec;		 /* requested and actual in use */
+
 	unsigned char  autoneg;          /* autonegotiating? */
+
 	unsigned char  link_ok;          /* link up? */
 	unsigned char  link_down_rc;     /* link down reason */
+
+	bool new_module;		 /* ->OS Transceiver Module inserted */
+	bool redo_l1cfg;		 /* ->CC redo current "sticky" L1 CFG */
 };
 
 #define FW_LEN16(fw_struct) FW_CMD_LEN16_V(sizeof(fw_struct) / 16)
@@ -466,11 +620,27 @@ enum {
 };
 
 enum {
+	MAX_TXQ_DESC_SIZE      = 64,
+	MAX_RXQ_DESC_SIZE      = 128,
+	MAX_FL_DESC_SIZE       = 8,
+	MAX_CTRL_TXQ_DESC_SIZE = 64,
+};
+
+enum {
 	INGQ_EXTRAS = 2,        /* firmware event queue and */
 				/*   forwarded interrupts */
 	MAX_INGQ = MAX_ETH_QSETS + INGQ_EXTRAS,
 };
 
+enum {
+	PRIV_FLAG_PORT_TX_VM_BIT,
+};
+
+#define PRIV_FLAG_PORT_TX_VM		BIT(PRIV_FLAG_PORT_TX_VM_BIT)
+
+#define PRIV_FLAGS_ADAP			0
+#define PRIV_FLAGS_PORT			PRIV_FLAG_PORT_TX_VM
+
 struct adapter;
 struct sge_rspq;
 
@@ -483,7 +653,7 @@ struct sge_rspq;
 struct port_info {
 	struct adapter *adapter;
 	u16    viid;
-	s16    xact_addr_filt;        /* index of exact MAC address filter */
+	int    xact_addr_filt;        /* index of exact MAC address filter */
 	u16    rss_size;              /* size of VI's RSS table slice */
 	s8     mdio_addr;
 	enum fw_port_type port_type;
@@ -505,28 +675,59 @@ struct port_info {
 #endif /* CONFIG_CHELSIO_T4_FCOE */
 	bool rxtstamp;  /* Enable TS */
 	struct hwtstamp_config tstamp_config;
+	bool ptp_enable;
 	struct sched_table *sched_tbl;
+	u32 eth_flags;
+
+	/* viid and smt fields either returned by fw
+	 * or decoded by parsing viid by driver.
+	 */
+	u8 vin;
+	u8 vivld;
+	u8 smt_idx;
+	u8 rx_cchan;
+
+	bool tc_block_shared;
+
+	/* Mirror VI information */
+	u16 viid_mirror;
+	u16 nmirrorqsets;
+	u32 vi_mirror_count;
+	struct mutex vi_mirror_mutex; /* Sync access to Mirror VI info */
+	struct cxgb4_ethtool_lb_test ethtool_lb;
 };
 
 struct dentry;
 struct work_struct;
 
 enum {                                 /* adapter flags */
-	FULL_INIT_DONE     = (1 << 0),
-	DEV_ENABLED        = (1 << 1),
-	USING_MSI          = (1 << 2),
-	USING_MSIX         = (1 << 3),
-	FW_OK              = (1 << 4),
-	RSS_TNLALLLOOKUP   = (1 << 5),
-	USING_SOFT_PARAMS  = (1 << 6),
-	MASTER_PF          = (1 << 7),
-	FW_OFLD_CONN       = (1 << 9),
+	CXGB4_FULL_INIT_DONE		= (1 << 0),
+	CXGB4_DEV_ENABLED		= (1 << 1),
+	CXGB4_USING_MSI			= (1 << 2),
+	CXGB4_USING_MSIX		= (1 << 3),
+	CXGB4_FW_OK			= (1 << 4),
+	CXGB4_RSS_TNLALLLOOKUP		= (1 << 5),
+	CXGB4_USING_SOFT_PARAMS		= (1 << 6),
+	CXGB4_MASTER_PF			= (1 << 7),
+	CXGB4_FW_OFLD_CONN		= (1 << 9),
+	CXGB4_ROOT_NO_RELAXED_ORDERING	= (1 << 10),
+	CXGB4_SHUTTING_DOWN		= (1 << 11),
+	CXGB4_SGE_DBQ_TIMER		= (1 << 12),
 };
 
 enum {
 	ULP_CRYPTO_LOOKASIDE = 1 << 0,
+	ULP_CRYPTO_IPSEC_INLINE = 1 << 1,
+	ULP_CRYPTO_KTLS_INLINE  = 1 << 3,
 };
 
+#define CXGB4_MIRROR_RXQ_DEFAULT_DESC_NUM 1024
+#define CXGB4_MIRROR_RXQ_DEFAULT_DESC_SIZE 64
+#define CXGB4_MIRROR_RXQ_DEFAULT_INTR_USEC 5
+#define CXGB4_MIRROR_RXQ_DEFAULT_PKT_CNT 8
+
+#define CXGB4_MIRROR_FLQ_DEFAULT_DESC_NUM 72
+
 struct rx_sw_desc;
 
 struct sge_fl {                     /* SGE free-buffer queue state */
@@ -604,12 +805,14 @@ struct sge_eth_stats {              /* Ethernet queue statistics */
 	unsigned long rx_cso;       /* # of Rx checksum offloads */
 	unsigned long vlan_ex;      /* # of Rx VLAN extractions */
 	unsigned long rx_drops;     /* # of packets dropped due to no mem */
+	unsigned long bad_rx_pkts;  /* # of packets with err_vec!=0 */
 };
 
 struct sge_eth_rxq {                /* SW Ethernet Rx queue */
 	struct sge_rspq rspq;
 	struct sge_fl fl;
 	struct sge_eth_stats stats;
+	struct msix_info *msix;
 } ____cacheline_aligned_in_smp;
 
 struct sge_ofld_stats {             /* offload queue statistics */
@@ -623,13 +826,19 @@ struct sge_ofld_rxq {               /* SW offload Rx queue */
 	struct sge_rspq rspq;
 	struct sge_fl fl;
 	struct sge_ofld_stats stats;
+	struct msix_info *msix;
 } ____cacheline_aligned_in_smp;
 
 struct tx_desc {
 	__be64 flit[8];
 };
 
-struct tx_sw_desc;
+struct ulptx_sgl;
+
+struct tx_sw_desc {
+	struct sk_buff *skb; /* SKB to free after getting completion */
+	dma_addr_t addr[MAX_SKB_FRAGS + 1]; /* DMA mapped addresses */
+};
 
 struct sge_txq {
 	unsigned int  in_use;       /* # of in-use Tx descriptors */
@@ -658,7 +867,10 @@ struct sge_eth_txq {                /* state for an SGE Ethernet Tx queue */
 #ifdef CONFIG_CHELSIO_T4_DCB
 	u8 dcb_prio;		    /* DCB Priority bound to queue */
 #endif
+	u8 dbqt;                    /* SGE Doorbell Queue Timer in use */
+	unsigned int dbqtimerix;    /* SGE Doorbell Queue Timer Index */
 	unsigned long tso;          /* # of TSO requests */
+	unsigned long uso;          /* # of USO requests */
 	unsigned long tx_cso;       /* # of Tx checksum offloads */
 	unsigned long vlan_ins;     /* # of Tx VLAN insertions */
 	unsigned long mapping_err;  /* # of I/O MMU packet mapping errors */
@@ -685,7 +897,6 @@ struct sge_ctrl_txq {               /* state for an SGE control Tx queue */
 struct sge_uld_rxq_info {
 	char name[IFNAMSIZ];	/* name of ULD driver */
 	struct sge_ofld_rxq *uldrxq; /* Rxq's for ULD */
-	u16 *msix_tbl;		/* msix_tbl for uld */
 	u16 *rspq_id;		/* response queue id's of rxq */
 	u16 nrxq;		/* # of ingress uld queues */
 	u16 nciq;		/* # of completion queues */
@@ -698,8 +909,61 @@ struct sge_uld_txq_info {
 	u16 ntxq;		/* # of egress uld queues */
 };
 
+/* struct to maintain ULD list to reallocate ULD resources on hotplug */
+struct cxgb4_uld_list {
+	struct cxgb4_uld_info uld_info;
+	struct list_head list_node;
+	enum cxgb4_uld uld_type;
+};
+
+enum sge_eosw_state {
+	CXGB4_EO_STATE_CLOSED = 0, /* Not ready to accept traffic */
+	CXGB4_EO_STATE_FLOWC_OPEN_SEND, /* Send FLOWC open request */
+	CXGB4_EO_STATE_FLOWC_OPEN_REPLY, /* Waiting for FLOWC open reply */
+	CXGB4_EO_STATE_ACTIVE, /* Ready to accept traffic */
+	CXGB4_EO_STATE_FLOWC_CLOSE_SEND, /* Send FLOWC close request */
+	CXGB4_EO_STATE_FLOWC_CLOSE_REPLY, /* Waiting for FLOWC close reply */
+};
+
+struct sge_eosw_txq {
+	spinlock_t lock; /* Per queue lock to synchronize completions */
+	enum sge_eosw_state state; /* Current ETHOFLD State */
+	struct tx_sw_desc *desc; /* Descriptor ring to hold packets */
+	u32 ndesc; /* Number of descriptors */
+	u32 pidx; /* Current Producer Index */
+	u32 last_pidx; /* Last successfully transmitted Producer Index */
+	u32 cidx; /* Current Consumer Index */
+	u32 last_cidx; /* Last successfully reclaimed Consumer Index */
+	u32 flowc_idx; /* Descriptor containing a FLOWC request */
+	u32 inuse; /* Number of packets held in ring */
+
+	u32 cred; /* Current available credits */
+	u32 ncompl; /* # of completions posted */
+	u32 last_compl; /* # of credits consumed since last completion req */
+
+	u32 eotid; /* Index into EOTID table in software */
+	u32 hwtid; /* Hardware EOTID index */
+
+	u32 hwqid; /* Underlying hardware queue index */
+	struct net_device *netdev; /* Pointer to netdevice */
+	struct tasklet_struct qresume_tsk; /* Restarts the queue */
+	struct completion completion; /* completion for FLOWC rendezvous */
+};
+
+struct sge_eohw_txq {
+	spinlock_t lock; /* Per queue lock */
+	struct sge_txq q; /* HW Txq */
+	struct adapter *adap; /* Backpointer to adapter */
+	unsigned long tso; /* # of TSO requests */
+	unsigned long uso; /* # of USO requests */
+	unsigned long tx_cso; /* # of Tx checksum offloads */
+	unsigned long vlan_ins; /* # of Tx VLAN insertions */
+	unsigned long mapping_err; /* # of I/O MMU packet mapping errors */
+};
+
 struct sge {
 	struct sge_eth_txq ethtxq[MAX_ETH_QSETS];
+	struct sge_eth_txq ptptxq;
 	struct sge_ctrl_txq ctrlq[MAX_CTRL_QUEUES];
 
 	struct sge_eth_rxq ethrxq[MAX_ETH_QSETS];
@@ -710,13 +974,23 @@ struct sge {
 	struct sge_rspq intrq ____cacheline_aligned_in_smp;
 	spinlock_t intrq_lock;
 
+	struct sge_eohw_txq *eohw_txq;
+	struct sge_ofld_rxq *eohw_rxq;
+
+	struct sge_eth_rxq *mirror_rxq[NCHAN];
+
 	u16 max_ethqsets;           /* # of available Ethernet queue sets */
 	u16 ethqsets;               /* # of active Ethernet queue sets */
 	u16 ethtxq_rover;           /* Tx queue to clean up next */
 	u16 ofldqsets;              /* # of active ofld queue sets */
 	u16 nqs_per_uld;	    /* # of Rx queues per ULD */
+	u16 eoqsets;                /* # of ETHOFLD queues */
+	u16 mirrorqsets;            /* # of Mirror queues */
+
 	u16 timer_val[SGE_NTIMERS];
 	u8 counter_val[SGE_NCOUNTERS];
+	u16 dbqtimer_tick;
+	u16 dbqtimer_val[SGE_NDBQTIMERS];
 	u32 fl_pg_order;            /* large page allocation size */
 	u32 stat_len;               /* length of status page at ring end */
 	u32 pktshift;               /* padding between CPL & packet data */
@@ -735,6 +1009,9 @@ struct sge {
 	unsigned long *blocked_fl;
 	struct timer_list rx_timer; /* refills starving FLs */
 	struct timer_list tx_timer; /* checks Tx queues */
+
+	int fwevtq_msix_idx; /* Index to firmware event queue MSI-X info */
+	int nd_msix_idx; /* Index to non-data interrupts MSI-X info */
 };
 
 #define for_each_ethrxq(sge, i) for (i = 0; i < (sge)->ethqsets; i++)
@@ -761,29 +1038,69 @@ struct doorbell_stats {
 struct hash_mac_addr {
 	struct list_head list;
 	u8 addr[ETH_ALEN];
+	unsigned int iface_mac;
 };
 
-struct uld_msix_bmap {
+struct msix_bmap {
 	unsigned long *msix_bmap;
 	unsigned int mapsize;
 	spinlock_t lock; /* lock for acquiring bitmap */
 };
 
-struct uld_msix_info {
+struct msix_info {
 	unsigned short vec;
 	char desc[IFNAMSIZ + 10];
 	unsigned int idx;
+	cpumask_var_t aff_mask;
 };
 
 struct vf_info {
 	unsigned char vf_mac_addr[ETH_ALEN];
+	unsigned int tx_rate;
 	bool pf_set_mac;
+	u16 vlan;
+	int link_state;
+};
+
+enum {
+	HMA_DMA_MAPPED_FLAG = 1
+};
+
+struct hma_data {
+	unsigned char flags;
+	struct sg_table *sgt;
+	dma_addr_t *phy_addr;	/* physical address of the page */
 };
 
 struct mbox_list {
 	struct list_head list;
 };
 
+#if IS_ENABLED(CONFIG_THERMAL)
+struct ch_thermal {
+	struct thermal_zone_device *tzdev;
+};
+#endif
+
+struct mps_entries_ref {
+	struct list_head list;
+	u8 addr[ETH_ALEN];
+	u8 mask[ETH_ALEN];
+	u16 idx;
+	refcount_t refcnt;
+};
+
+struct cxgb4_ethtool_filter_info {
+	u32 *loc_array; /* Array holding the actual TIDs set to filters */
+	unsigned long *bmap; /* Bitmap for managing filters in use */
+	u32 in_use; /* # of filters in use */
+};
+
+struct cxgb4_ethtool_filter {
+	u32 nentries; /* Adapter wide number of supported filters */
+	struct cxgb4_ethtool_filter_info *port; /* Per port entry */
+};
+
 struct adapter {
 	void __iomem *regs;
 	void __iomem *bar2;
@@ -796,20 +1113,19 @@ struct adapter {
 	unsigned int flags;
 	unsigned int adap_idx;
 	enum chip_type chip;
+	u32 eth_flags;
 
 	int msg_enable;
+	__be16 vxlan_port;
+	__be16 geneve_port;
 
 	struct adapter_params params;
 	struct cxgb4_virt_res vres;
 	unsigned int swintr;
 
-	struct {
-		unsigned short vec;
-		char desc[IFNAMSIZ + 10];
-	} msix_info[MAX_INGQ + 1];
-	struct uld_msix_info *msix_info_ulds; /* msix info for uld's */
-	struct uld_msix_bmap msix_bmap_ulds; /* msix bitmap for all uld */
-	int msi_idx;
+	/* MSI-X Info for NIC and OFLD queues */
+	struct msix_info *msix_info;
+	struct msix_bmap msix_bmap;
 
 	struct doorbell_stats db_stats;
 	struct sge sge;
@@ -827,6 +1143,9 @@ struct adapter {
 	unsigned int clipt_start;
 	unsigned int clipt_end;
 	struct clip_tbl *clipt;
+	unsigned int rawf_start;
+	unsigned int rawf_cnt;
+	struct smt_data *smt;
 	struct cxgb4_uld_info *uld;
 	void *uld_handle[CXGB4_ULD_MAX];
 	unsigned int num_uld;
@@ -834,6 +1153,8 @@ struct adapter {
 	struct list_head list_node;
 	struct list_head rcu_node;
 	struct list_head mac_hlist; /* list of MAC addresses in MPS Hash */
+	struct list_head mps_ref;
+	spinlock_t mps_ref_lock; /* lock for syncing mps ref/def activities */
 
 	void *iscsi_ppm;
 
@@ -844,6 +1165,7 @@ struct adapter {
 	struct work_struct tid_release_task;
 	struct work_struct db_full_task;
 	struct work_struct db_drop_task;
+	struct work_struct fatal_err_notify_task;
 	bool tid_release_task_busy;
 
 	/* lock for mailbox cmd list */
@@ -863,30 +1185,75 @@ struct adapter {
 			 * used for all 4 filters.
 			 */
 
+	struct ptp_clock *ptp_clock;
+	struct ptp_clock_info ptp_clock_info;
+	struct sk_buff *ptp_tx_skb;
+	/* ptp lock */
+	spinlock_t ptp_lock;
 	spinlock_t stats_lock;
 	spinlock_t win0_lock ____cacheline_aligned_in_smp;
 
 	/* TC u32 offload */
 	struct cxgb4_tc_u32_table *tc_u32;
+	struct chcr_ktls chcr_ktls;
+	struct chcr_stats_debug chcr_stats;
+#if IS_ENABLED(CONFIG_CHELSIO_TLS_DEVICE)
+	struct ch_ktls_stats_debug ch_ktls_stats;
+#endif
+#if IS_ENABLED(CONFIG_CHELSIO_IPSEC_INLINE)
+	struct ch_ipsec_stats_debug ch_ipsec_stats;
+#endif
+
+	/* TC flower offload */
+	bool tc_flower_initialized;
+	struct rhashtable flower_tbl;
+	struct rhashtable_params flower_ht_params;
+	struct timer_list flower_stats_timer;
+	struct work_struct flower_stats_work;
+
+	/* HMA */
+	struct hma_data hma;
+
+	struct srq_data *srq;
+
+	/* Dump buffer for collecting logs in kdump kernel */
+	struct vmcoredd_data vmcoredd;
+#if IS_ENABLED(CONFIG_THERMAL)
+	struct ch_thermal ch_thermal;
+#endif
+
+	/* TC MQPRIO offload */
+	struct cxgb4_tc_mqprio *tc_mqprio;
+
+	/* TC MATCHALL classifier offload */
+	struct cxgb4_tc_matchall *tc_matchall;
+
+	/* Ethtool n-tuple */
+	struct cxgb4_ethtool_filter *ethtool_filters;
+
+	/* Ethtool Dump */
+	/* Must be last - ends in a flex-array member. */
+	struct ethtool_dump eth_dump;
 };
 
 /* Support for "sched-class" command to allow a TX Scheduling Class to be
  * programmed with various parameters.
  */
 struct ch_sched_params {
-	s8   type;                     /* packet or flow */
+	u8   type;                     /* packet or flow */
 	union {
 		struct {
-			s8   level;    /* scheduler hierarchy level */
-			s8   mode;     /* per-class or per-flow */
-			s8   rateunit; /* bit or packet rate */
-			s8   ratemode; /* %port relative or kbps absolute */
-			s8   channel;  /* scheduler channel [0..N] */
-			s8   class;    /* scheduler class [0..N] */
-			s32  minrate;  /* minimum rate */
-			s32  maxrate;  /* maximum rate */
-			s16  weight;   /* percent weight */
-			s16  pktsize;  /* average packet size */
+			u8   level;    /* scheduler hierarchy level */
+			u8   mode;     /* per-class or per-flow */
+			u8   rateunit; /* bit or packet rate */
+			u8   ratemode; /* %port relative or kbps absolute */
+			u8   channel;  /* scheduler channel [0..N] */
+			u8   class;    /* scheduler class [0..N] */
+			u32  minrate;  /* minimum rate */
+			u32  maxrate;  /* maximum rate */
+			u16  weight;   /* percent weight */
+			u16  pktsize;  /* average packet size */
+			u16  burstsize;  /* burst buffer size */
 		} params;
 	} u;
 };
@@ -897,10 +1264,12 @@ enum {
 
 enum {
 	SCHED_CLASS_LEVEL_CL_RL = 0,    /* class rate limiter */
+	SCHED_CLASS_LEVEL_CH_RL = 2,    /* channel rate limiter */
 };
 
 enum {
 	SCHED_CLASS_MODE_CLASS = 0,     /* per-class scheduling */
+	SCHED_CLASS_MODE_FLOW,          /* per-flow scheduling */
 };
 
 enum {
@@ -919,6 +1288,14 @@ struct ch_sched_queue {
 	s8   class;    /* class index */
 };
 
+/* Support for "sched_flowc" command to allow one or more FLOWC
+ * to be bound to a TX Scheduling Class.
+ */
+struct ch_sched_flowc {
+	s32 tid;   /* TID to bind */
+	s8  class; /* class index */
+};
+
 /* Defined bit width of user definable filter tuples
  */
 #define ETHTYPE_BITWIDTH 16
@@ -933,12 +1310,13 @@ struct ch_sched_queue {
 #define VF_BITWIDTH 8
 #define IVLAN_BITWIDTH 16
 #define OVLAN_BITWIDTH 16
+#define ENCAP_VNI_BITWIDTH 24
 
 /* Filter matching rules.  These consist of a set of ingress packet field
  * (value, mask) tuples.  The associated ingress packet field matches the
  * tuple when ((field & mask) == value).  (Thus a wildcard "don't care" field
  * rule can be constructed by specifying a tuple of (0, 0).)  A filter rule
- * matches an ingress packet when all of the individual individual field
+ * matches an ingress packet when all of the individual field
  * matching rules are true.
  *
  * Partial field masks are always valid, however, while it may be easy to
@@ -963,6 +1341,7 @@ struct ch_filter_tuple {
 	uint32_t ivlan_vld:1;                   /* inner VLAN valid */
 	uint32_t ovlan_vld:1;                   /* outer VLAN valid */
 	uint32_t pfvf_vld:1;                    /* PF/VF valid */
+	uint32_t encap_vld:1;			/* Encapsulation valid */
 	uint32_t macidx:MACIDX_BITWIDTH;        /* exact match MAC index */
 	uint32_t fcoe:FCOE_BITWIDTH;            /* FCoE packet */
 	uint32_t iport:IPORT_BITWIDTH;          /* ingress port */
@@ -973,6 +1352,7 @@ struct ch_filter_tuple {
 	uint32_t vf:VF_BITWIDTH;                /* PCI-E VF ID */
 	uint32_t ivlan:IVLAN_BITWIDTH;          /* inner VLAN */
 	uint32_t ovlan:OVLAN_BITWIDTH;          /* outer VLAN */
+	uint32_t vni:ENCAP_VNI_BITWIDTH;	/* VNI of tunnel */
 
 	/* Uncompressed header matching field rules.  These are always
 	 * available for field rules.
@@ -995,6 +1375,7 @@ struct ch_filter_specification {
 	 * matching that doesn't exist as a (value, mask) tuple.
 	 */
 	uint32_t type:1;        /* 0 => IPv4, 1 => IPv6 */
+	u32 hash:1;		/* 0 => wild-card, 1 => exact-match */
 
 	/* Packet dispatch information.  Ingress packets which match the
 	 * filter rules will be dropped, passed to the host or switched back
@@ -1019,10 +1400,22 @@ struct ch_filter_specification {
 	uint32_t newdmac:1;     /* rewrite destination MAC address */
 	uint32_t newsmac:1;     /* rewrite source MAC address */
 	uint32_t newvlan:2;     /* rewrite VLAN Tag */
+	uint32_t nat_mode:3;    /* specify NAT operation mode */
 	uint8_t dmac[ETH_ALEN]; /* new destination MAC address */
 	uint8_t smac[ETH_ALEN]; /* new source MAC address */
 	uint16_t vlan;          /* VLAN Tag to insert */
 
+	u8 nat_lip[16];		/* local IP to use after NAT'ing */
+	u8 nat_fip[16];		/* foreign IP to use after NAT'ing */
+	u16 nat_lport;		/* local port to use after NAT'ing */
+	u16 nat_fport;		/* foreign port to use after NAT'ing */
+
+	u32 tc_prio;		/* TC's filter priority index */
+	u64 tc_cookie;		/* Unique cookie identifying TC rules */
+
+	/* reservation for future additions */
+	u8 rsvd[12];
+
 	/* Filter rule value/mask pairs.
 	 */
 	struct ch_filter_tuple val;
@@ -1042,6 +1435,19 @@ enum {
 	VLAN_REWRITE
 };
 
+enum {
+	NAT_MODE_NONE = 0,	/* No NAT performed */
+	NAT_MODE_DIP,		/* NAT on Dst IP */
+	NAT_MODE_DIP_DP,	/* NAT on Dst IP, Dst Port */
+	NAT_MODE_DIP_DP_SIP,	/* NAT on Dst IP, Dst Port and Src IP */
+	NAT_MODE_DIP_DP_SP,	/* NAT on Dst IP, Dst Port and Src Port */
+	NAT_MODE_SIP_SP,	/* NAT on Src IP and Src Port */
+	NAT_MODE_DIP_SIP_SP,	/* NAT on Dst IP, Src IP and Src Port */
+	NAT_MODE_ALL		/* NAT on entire 4-tuple */
+};
+
+#define CXGB4_FILTER_TYPE_MAX 2
+
 /* Host shadow copy of ingress filter entry.  This is in host native format
  * and doesn't match the ordering or bit order, etc. of the hardware of the
  * firmware command.  The use of bit-field structure elements is purely to
@@ -1054,9 +1460,9 @@ struct filter_entry {
 	u32 locked:1;           /* filter is administratively locked */
 
 	u32 pending:1;          /* filter action is pending firmware reply */
-	u32 smtidx:8;           /* Source MAC Table index for smac */
 	struct filter_ctx *ctx; /* Caller's completion hook */
 	struct l2t_entry *l2t;  /* Layer Two Table entry for dmac */
+	struct smt_entry *smt;  /* Source Mac Table entry for smac */
 	struct net_device *dev; /* Associated net device */
 	u32 tid;                /* This will store the actual tid */
 
@@ -1073,6 +1479,11 @@ static inline int is_offload(const struct adapter *adap)
 	return adap->params.offload;
 }
 
+static inline int is_hashfilter(const struct adapter *adap)
+{
+	return adap->params.hash_filter;
+}
+
 static inline int is_pci_uld(const struct adapter *adap)
 {
 	return adap->params.crypto;
@@ -1083,6 +1494,11 @@ static inline int is_uld(const struct adapter *adap)
 	return (adap->params.offload || adap->params.crypto);
 }
 
+static inline int is_ethofld(const struct adapter *adap)
+{
+	return adap->params.ethofld;
+}
+
 static inline u32 t4_read_reg(struct adapter *adap, u32 reg_addr)
 {
 	return readl(adap->regs + reg_addr);
@@ -1128,7 +1544,7 @@ static inline void t4_write_reg64(struct adapter *adap, u32 reg_addr, u64 val)
 static inline void t4_set_hw_addr(struct adapter *adapter, int port_idx,
 				  u8 hw_addr[])
 {
-	ether_addr_copy(adapter->port[port_idx]->dev_addr, hw_addr);
+	eth_hw_addr_set(adapter->port[port_idx], hw_addr);
 	ether_addr_copy(adapter->port[port_idx]->perm_addr, hw_addr);
 }
 
@@ -1186,17 +1602,16 @@ static inline unsigned int qtimer_val(const struct adapter *adap,
 	return idx < SGE_NTIMERS ? adap->sge.timer_val[idx] : 0;
 }
 
-/* driver version & name used for ethtool_drvinfo */
+/* driver name used for ethtool_drvinfo */
 extern char cxgb4_driver_name[];
-extern const char cxgb4_driver_version[];
 
-void t4_os_portmod_changed(const struct adapter *adap, int port_id);
+void t4_os_portmod_changed(struct adapter *adap, int port_id);
 void t4_os_link_changed(struct adapter *adap, int port_id, int link_stat);
 
 void t4_free_sge_resources(struct adapter *adap);
-void t4_free_ofld_rxqs(struct adapter *adap, int n, struct sge_ofld_rxq *q);
 irq_handler_t t4_intr_handler(struct adapter *adap);
-netdev_tx_t t4_eth_xmit(struct sk_buff *skb, struct net_device *dev);
+netdev_tx_t t4_start_xmit(struct sk_buff *skb, struct net_device *dev);
+int cxgb4_selftest_lb_pkt(struct net_device *netdev);
 int t4_ethrx_handler(struct sge_rspq *q, const __be64 *rsp,
 		     const struct pkt_gl *gl);
 int t4_mgmt_tx(struct adapter *adap, struct sk_buff *skb);
@@ -1207,7 +1622,7 @@ int t4_sge_alloc_rxq(struct adapter *adap, struct sge_rspq *iq, bool fwevtq,
 		     rspq_flush_handler_t flush_handler, int cong);
 int t4_sge_alloc_eth_txq(struct adapter *adap, struct sge_eth_txq *txq,
 			 struct net_device *dev, struct netdev_queue *netdevq,
-			 unsigned int iqid);
+			 unsigned int iqid, u8 dbqt);
 int t4_sge_alloc_ctrl_txq(struct adapter *adap, struct sge_ctrl_txq *txq,
 			  struct net_device *dev, unsigned int iqid,
 			  unsigned int cmplqid);
@@ -1216,12 +1631,18 @@ int t4_sge_mod_ctrl_txq(struct adapter *adap, unsigned int eqid,
 int t4_sge_alloc_uld_txq(struct adapter *adap, struct sge_uld_txq *txq,
 			 struct net_device *dev, unsigned int iqid,
 			 unsigned int uld_type);
+int t4_sge_alloc_ethofld_txq(struct adapter *adap, struct sge_eohw_txq *txq,
+			     struct net_device *dev, u32 iqid);
+void t4_sge_free_ethofld_txq(struct adapter *adap, struct sge_eohw_txq *txq);
 irqreturn_t t4_sge_intr_msix(int irq, void *cookie);
 int t4_sge_init(struct adapter *adap);
 void t4_sge_start(struct adapter *adap);
 void t4_sge_stop(struct adapter *adap);
+int t4_sge_eth_txq_egress_update(struct adapter *adap, struct sge_eth_txq *q,
+				 int maxreclaim);
 void cxgb4_set_ethtool_ops(struct net_device *netdev);
 int cxgb4_write_rss(const struct port_info *pi, const u16 *queues);
+enum cpl_tx_tnl_lso_type cxgb_encap_offload_supported(struct sk_buff *skb);
 extern int dbfifo_int_thresh;
 
 #define for_each_port(adapter, iter) \
@@ -1276,6 +1697,12 @@ static inline unsigned int core_ticks_to_us(const struct adapter *adapter,
 		adapter->params.vpd.cclk);
 }
 
+static inline unsigned int dack_ticks_to_usec(const struct adapter *adap,
+					      unsigned int ticks)
+{
+	return (ticks << adap->params.tp.dack_re) / core_ticks_per_usec(adap);
+}
+
 void t4_set_reg_field(struct adapter *adap, unsigned int addr, u32 mask,
 		      u32 val);
 
@@ -1334,6 +1761,21 @@ static inline void init_rspq(struct adapter *adap, struct sge_rspq *q,
 	q->size = size;
 }
 
+/**
+ *     t4_is_inserted_mod_type - is a plugged in Firmware Module Type
+ *     @fw_mod_type: the Firmware Mofule Type
+ *
+ *     Return whether the Firmware Module Type represents a real Transceiver
+ *     Module/Cable Module Type which has been inserted.
+ */
+static inline bool t4_is_inserted_mod_type(unsigned int fw_mod_type)
+{
+	return (fw_mod_type != FW_PORT_MOD_TYPE_NONE &&
+		fw_mod_type != FW_PORT_MOD_TYPE_NOTSUPPORTED &&
+		fw_mod_type != FW_PORT_MOD_TYPE_UNKNOWN &&
+		fw_mod_type != FW_PORT_MOD_TYPE_ERROR);
+}
+
 void t4_write_indirect(struct adapter *adap, unsigned int addr_reg,
 		       unsigned int data_reg, const u32 *vals,
 		       unsigned int nregs, unsigned int start_idx);
@@ -1349,14 +1791,38 @@ void t4_intr_disable(struct adapter *adapter);
 int t4_slow_intr_handler(struct adapter *adapter);
 
 int t4_wait_dev_ready(void __iomem *regs);
-int t4_link_l1cfg(struct adapter *adap, unsigned int mbox, unsigned int port,
-		  struct link_config *lc);
+
+fw_port_cap32_t t4_link_acaps(struct adapter *adapter, unsigned int port,
+			      struct link_config *lc);
+int t4_link_l1cfg_core(struct adapter *adap, unsigned int mbox,
+		       unsigned int port, struct link_config *lc,
+		       u8 sleep_ok, int timeout);
+
+static inline int t4_link_l1cfg(struct adapter *adapter, unsigned int mbox,
+				unsigned int port, struct link_config *lc)
+{
+	return t4_link_l1cfg_core(adapter, mbox, port, lc,
+				  true, FW_CMD_MAX_TIMEOUT);
+}
+
+static inline int t4_link_l1cfg_ns(struct adapter *adapter, unsigned int mbox,
+				   unsigned int port, struct link_config *lc)
+{
+	return t4_link_l1cfg_core(adapter, mbox, port, lc,
+				  false, FW_CMD_MAX_TIMEOUT);
+}
+
 int t4_restart_aneg(struct adapter *adap, unsigned int mbox, unsigned int port);
 
 u32 t4_read_pcie_cfg4(struct adapter *adap, int reg);
 u32 t4_get_util_window(struct adapter *adap);
 void t4_setup_memwin(struct adapter *adap, u32 memwin_base, u32 window);
 
+int t4_memory_rw_init(struct adapter *adap, int win, int mtype, u32 *mem_off,
+		      u32 *mem_base, u32 *mem_aperture);
+void t4_memory_update_win(struct adapter *adap, int win, u32 addr);
+void t4_memory_rw_residual(struct adapter *adap, u32 off, u32 addr, u8 *buf,
+			   int dir);
 #define T4_MEMORY_WRITE	0
 #define T4_MEMORY_READ	1
 int t4_memory_rw(struct adapter *adap, int win, int mtype, u32 addr, u32 len,
@@ -1370,14 +1836,15 @@ static inline int t4_memory_write(struct adapter *adap, int mtype, u32 addr,
 unsigned int t4_get_regs_len(struct adapter *adapter);
 void t4_get_regs(struct adapter *adap, void *buf, size_t buf_size);
 
+int t4_eeprom_ptov(unsigned int phys_addr, unsigned int fn, unsigned int sz);
 int t4_seeprom_wp(struct adapter *adapter, bool enable);
 int t4_get_raw_vpd_params(struct adapter *adapter, struct vpd_params *p);
 int t4_get_vpd_params(struct adapter *adapter, struct vpd_params *p);
+int t4_get_pfres(struct adapter *adapter);
 int t4_read_flash(struct adapter *adapter, unsigned int addr,
 		  unsigned int nwords, u32 *data, int byte_oriented);
 int t4_load_fw(struct adapter *adapter, const u8 *fw_data, unsigned int size);
-int t4_load_phy_fw(struct adapter *adap,
-		   int win, spinlock_t *lock,
+int t4_load_phy_fw(struct adapter *adap, int win,
 		   int (*phy_fw_version)(const u8 *, size_t),
 		   const u8 *phy_fw_data, size_t phy_fw_size);
 int t4_phy_fw_ver(struct adapter *adap, int *phy_fw_ver);
@@ -1387,10 +1854,15 @@ int t4_fw_upgrade(struct adapter *adap, unsigned int mbox,
 int t4_fl_pkt_align(struct adapter *adap);
 unsigned int t4_flash_cfg_addr(struct adapter *adapter);
 int t4_check_fw_version(struct adapter *adap);
+int t4_load_cfg(struct adapter *adapter, const u8 *cfg_data, unsigned int size);
 int t4_get_fw_version(struct adapter *adapter, u32 *vers);
 int t4_get_bs_version(struct adapter *adapter, u32 *vers);
 int t4_get_tp_version(struct adapter *adapter, u32 *vers);
 int t4_get_exprom_version(struct adapter *adapter, u32 *vers);
+int t4_get_scfg_version(struct adapter *adapter, u32 *vers);
+int t4_get_vpd_version(struct adapter *adapter, u32 *vers);
+int t4_get_version_info(struct adapter *adapter);
+void t4_dump_version_info(struct adapter *adapter);
 int t4_prep_fw(struct adapter *adap, struct fw_info *fw_info,
 	       const u8 *fw_data, unsigned int fw_size,
 	       struct fw_hdr *card_fw, enum dev_state state, int *reset);
@@ -1410,13 +1882,16 @@ unsigned int qtimer_val(const struct adapter *adap,
 
 int t4_init_devlog_params(struct adapter *adapter);
 int t4_init_sge_params(struct adapter *adapter);
-int t4_init_tp_params(struct adapter *adap);
+int t4_init_tp_params(struct adapter *adap, bool sleep_ok);
 int t4_filter_field_shift(const struct adapter *adap, int filter_sel);
 int t4_init_rss_mode(struct adapter *adap, int mbox);
 int t4_init_portinfo(struct port_info *pi, int mbox,
 		     int port, int pf, int vf, u8 mac[]);
 int t4_port_init(struct adapter *adap, int mbox, int pf, int vf);
+int t4_init_port_mirror(struct port_info *pi, u8 mbox, u8 port, u8 pf, u8 vf,
+			u16 *mirror_viid);
 void t4_fatal_err(struct adapter *adapter);
+unsigned int t4_chip_rss_size(struct adapter *adapter);
 int t4_config_rss_range(struct adapter *adapter, int mbox, unsigned int viid,
 			int start, int n, const u16 *rspq, unsigned int nrspq);
 int t4_config_glbl_rss(struct adapter *adapter, int mbox, unsigned int mode,
@@ -1424,16 +1899,18 @@ int t4_config_glbl_rss(struct adapter *adapter, int mbox, unsigned int mode,
 int t4_config_vi_rss(struct adapter *adapter, int mbox, unsigned int viid,
 		     unsigned int flags, unsigned int defq);
 int t4_read_rss(struct adapter *adapter, u16 *entries);
-void t4_read_rss_key(struct adapter *adapter, u32 *key);
-void t4_write_rss_key(struct adapter *adap, const u32 *key, int idx);
+void t4_read_rss_key(struct adapter *adapter, u32 *key, bool sleep_ok);
+void t4_write_rss_key(struct adapter *adap, const u32 *key, int idx,
+		      bool sleep_ok);
 void t4_read_rss_pf_config(struct adapter *adapter, unsigned int index,
-			   u32 *valp);
+			   u32 *valp, bool sleep_ok);
 void t4_read_rss_vf_config(struct adapter *adapter, unsigned int index,
-			   u32 *vfl, u32 *vfh);
-u32 t4_read_rss_pf_map(struct adapter *adapter);
-u32 t4_read_rss_pf_mask(struct adapter *adapter);
+			   u32 *vfl, u32 *vfh, bool sleep_ok);
+u32 t4_read_rss_pf_map(struct adapter *adapter, bool sleep_ok);
+u32 t4_read_rss_pf_mask(struct adapter *adapter, bool sleep_ok);
 
-unsigned int t4_get_mps_bg_map(struct adapter *adapter, int idx);
+unsigned int t4_get_mps_bg_map(struct adapter *adapter, int pidx);
+unsigned int t4_get_tp_ch_map(struct adapter *adapter, int pidx);
 void t4_pmtx_get_stats(struct adapter *adap, u32 cnt[], u64 cycles[]);
 void t4_pmrx_get_stats(struct adapter *adap, u32 cnt[], u64 cycles[]);
 int t4_read_cim_ibq(struct adapter *adap, unsigned int qid, u32 *data,
@@ -1461,14 +1938,18 @@ void t4_read_cong_tbl(struct adapter *adap, u16 incr[NMTUS][NCCTRL_WIN]);
 void t4_tp_wr_bits_indirect(struct adapter *adap, unsigned int addr,
 			    unsigned int mask, unsigned int val);
 void t4_tp_read_la(struct adapter *adap, u64 *la_buf, unsigned int *wrptr);
-void t4_tp_get_err_stats(struct adapter *adap, struct tp_err_stats *st);
-void t4_tp_get_cpl_stats(struct adapter *adap, struct tp_cpl_stats *st);
-void t4_tp_get_rdma_stats(struct adapter *adap, struct tp_rdma_stats *st);
-void t4_get_usm_stats(struct adapter *adap, struct tp_usm_stats *st);
+void t4_tp_get_err_stats(struct adapter *adap, struct tp_err_stats *st,
+			 bool sleep_ok);
+void t4_tp_get_cpl_stats(struct adapter *adap, struct tp_cpl_stats *st,
+			 bool sleep_ok);
+void t4_tp_get_rdma_stats(struct adapter *adap, struct tp_rdma_stats *st,
+			  bool sleep_ok);
+void t4_get_usm_stats(struct adapter *adap, struct tp_usm_stats *st,
+		      bool sleep_ok);
 void t4_tp_get_tcp_stats(struct adapter *adap, struct tp_tcp_stats *v4,
-			 struct tp_tcp_stats *v6);
+			 struct tp_tcp_stats *v6, bool sleep_ok);
 void t4_get_fcoe_stats(struct adapter *adap, unsigned int idx,
-		       struct tp_fcoe_stats *st);
+		       struct tp_fcoe_stats *st, bool sleep_ok);
 void t4_load_mtus(struct adapter *adap, const unsigned short *mtus,
 		  const unsigned short *alpha, const unsigned short *beta);
 
@@ -1477,11 +1958,6 @@ void t4_ulprx_read_la(struct adapter *adap, u32 *la_buf);
 void t4_get_chan_txrate(struct adapter *adap, u64 *nic_rate, u64 *ofld_rate);
 void t4_mk_filtdelwr(unsigned int ftid, struct fw_filter_wr *wr, int qid);
 
-void t4_wol_magic_enable(struct adapter *adap, unsigned int port,
-			 const u8 *addr);
-int t4_wol_pat_enable(struct adapter *adap, unsigned int port, unsigned int map,
-		      u64 mask0, u64 mask1, unsigned int crc, bool enable);
-
 int t4_fw_hello(struct adapter *adap, unsigned int mbox, unsigned int evt_mbox,
 		enum dev_master master, enum dev_state *state);
 int t4_fw_bye(struct adapter *adap, unsigned int mbox);
@@ -1493,9 +1969,12 @@ int t4_fw_initialize(struct adapter *adap, unsigned int mbox);
 int t4_query_params(struct adapter *adap, unsigned int mbox, unsigned int pf,
 		    unsigned int vf, unsigned int nparams, const u32 *params,
 		    u32 *val);
+int t4_query_params_ns(struct adapter *adap, unsigned int mbox, unsigned int pf,
+		       unsigned int vf, unsigned int nparams, const u32 *params,
+		       u32 *val);
 int t4_query_params_rw(struct adapter *adap, unsigned int mbox, unsigned int pf,
 		       unsigned int vf, unsigned int nparams, const u32 *params,
-		       u32 *val, int rw);
+		       u32 *val, int rw, bool sleep_ok);
 int t4_set_params_timeout(struct adapter *adap, unsigned int mbox,
 			  unsigned int pf, unsigned int vf,
 			  unsigned int nparams, const u32 *params,
@@ -1510,13 +1989,25 @@ int t4_cfg_pfvf(struct adapter *adap, unsigned int mbox, unsigned int pf,
 		unsigned int nexact, unsigned int rcaps, unsigned int wxcaps);
 int t4_alloc_vi(struct adapter *adap, unsigned int mbox, unsigned int port,
 		unsigned int pf, unsigned int vf, unsigned int nmac, u8 *mac,
-		unsigned int *rss_size);
+		unsigned int *rss_size, u8 *vivld, u8 *vin);
 int t4_free_vi(struct adapter *adap, unsigned int mbox,
 	       unsigned int pf, unsigned int vf,
 	       unsigned int viid);
 int t4_set_rxmode(struct adapter *adap, unsigned int mbox, unsigned int viid,
-		int mtu, int promisc, int all_multi, int bcast, int vlanex,
-		bool sleep_ok);
+		  unsigned int viid_mirror, int mtu, int promisc, int all_multi,
+		  int bcast, int vlanex, bool sleep_ok);
+int t4_free_raw_mac_filt(struct adapter *adap, unsigned int viid,
+			 const u8 *addr, const u8 *mask, unsigned int idx,
+			 u8 lookup_type, u8 port_id, bool sleep_ok);
+int t4_free_encap_mac_filt(struct adapter *adap, unsigned int viid, int idx,
+			   bool sleep_ok);
+int t4_alloc_encap_mac_filt(struct adapter *adap, unsigned int viid,
+			    const u8 *addr, const u8 *mask, unsigned int vni,
+			    unsigned int vni_mask, u8 dip_hit, u8 lookup_type,
+			    bool sleep_ok);
+int t4_alloc_raw_mac_filt(struct adapter *adap, unsigned int viid,
+			  const u8 *addr, const u8 *mask, unsigned int idx,
+			  u8 lookup_type, u8 port_id, bool sleep_ok);
 int t4_alloc_mac_filt(struct adapter *adap, unsigned int mbox,
 		      unsigned int viid, bool free, unsigned int naddr,
 		      const u8 **addr, u16 *idx, u64 *hash, bool sleep_ok);
@@ -1524,11 +2015,14 @@ int t4_free_mac_filt(struct adapter *adap, unsigned int mbox,
 		     unsigned int viid, unsigned int naddr,
 		     const u8 **addr, bool sleep_ok);
 int t4_change_mac(struct adapter *adap, unsigned int mbox, unsigned int viid,
-		  int idx, const u8 *addr, bool persist, bool add_smt);
+		  int idx, const u8 *addr, bool persist, u8 *smt_idx);
 int t4_set_addr_hash(struct adapter *adap, unsigned int mbox, unsigned int viid,
 		     bool ucast, u64 vec, bool sleep_ok);
 int t4_enable_vi_params(struct adapter *adap, unsigned int mbox,
 			unsigned int viid, bool rx_en, bool tx_en, bool dcb_en);
+int t4_enable_pi_params(struct adapter *adap, unsigned int mbox,
+			struct port_info *pi,
+			bool rx_en, bool tx_en, bool dcb_en);
 int t4_enable_vi(struct adapter *adap, unsigned int mbox, unsigned int viid,
 		 bool rx_en, bool tx_en);
 int t4_identify_port(struct adapter *adap, unsigned int mbox, unsigned int viid,
@@ -1549,9 +2043,13 @@ int t4_ctrl_eq_free(struct adapter *adap, unsigned int mbox, unsigned int pf,
 		    unsigned int vf, unsigned int eqid);
 int t4_ofld_eq_free(struct adapter *adap, unsigned int mbox, unsigned int pf,
 		    unsigned int vf, unsigned int eqid);
-int t4_sge_ctxt_flush(struct adapter *adap, unsigned int mbox);
+int t4_sge_ctxt_flush(struct adapter *adap, unsigned int mbox, int ctxt_type);
+int t4_read_sge_dbqtimers(struct adapter *adap, unsigned int ndbqtimers,
+			  u16 *dbqtimers);
 void t4_handle_get_port_info(struct port_info *pi, const __be64 *rpl);
 int t4_update_port_info(struct port_info *pi);
+int t4_get_link_params(struct port_info *pi, unsigned int *link_okp,
+		       unsigned int *speedp, unsigned int *mtup);
 int t4_handle_fw_rpl(struct adapter *adap, const __be64 *rpl);
 void t4_db_full(struct adapter *adapter);
 void t4_db_dropped(struct adapter *adapter);
@@ -1561,9 +2059,17 @@ void t4_get_trace_filter(struct adapter *adapter, struct trace_params *tp,
 			 int filter_index, int *enabled);
 int t4_fwaddrspace_write(struct adapter *adap, unsigned int mbox,
 			 u32 addr, u32 val);
-int t4_sched_params(struct adapter *adapter, int type, int level, int mode,
-		    int rateunit, int ratemode, int channel, int class,
-		    int minrate, int maxrate, int weight, int pktsize);
+void t4_read_pace_tbl(struct adapter *adap, unsigned int pace_vals[NTX_SCHED]);
+void t4_get_tx_sched(struct adapter *adap, unsigned int sched,
+		     unsigned int *kbps, unsigned int *ipg, bool sleep_ok);
+int t4_sge_ctxt_rd(struct adapter *adap, unsigned int mbox, unsigned int cid,
+		   enum ctxt_type ctype, u32 *data);
+int t4_sge_ctxt_rd_bd(struct adapter *adap, unsigned int cid,
+		      enum ctxt_type ctype, u32 *data);
+int t4_sched_params(struct adapter *adapter, u8 type, u8 level, u8 mode,
+		    u8 rateunit, u8 ratemode, u8 channel, u8 class,
+		    u32 minrate, u32 maxrate, u16 weight, u16 pktsize,
+		    u16 burstsize);
 void t4_sge_decode_idma_state(struct adapter *adapter, int state);
 void t4_idma_monitor_init(struct adapter *adapter,
 			  struct sge_idma_monitor_state *idma);
@@ -1571,13 +2077,80 @@ void t4_idma_monitor(struct adapter *adapter,
 		     struct sge_idma_monitor_state *idma,
 		     int hz, int ticks);
 int t4_set_vf_mac_acl(struct adapter *adapter, unsigned int vf,
-		      unsigned int naddr, u8 *addr);
+		      u8 start, unsigned int naddr, u8 *addr);
+void t4_tp_pio_read(struct adapter *adap, u32 *buff, u32 nregs,
+		    u32 start_index, bool sleep_ok);
+void t4_tp_tm_pio_read(struct adapter *adap, u32 *buff, u32 nregs,
+		       u32 start_index, bool sleep_ok);
+void t4_tp_mib_read(struct adapter *adap, u32 *buff, u32 nregs,
+		    u32 start_index, bool sleep_ok);
+
 void t4_uld_mem_free(struct adapter *adap);
 int t4_uld_mem_alloc(struct adapter *adap);
 void t4_uld_clean_up(struct adapter *adap);
 void t4_register_netevent_notifier(void);
+int t4_i2c_rd(struct adapter *adap, unsigned int mbox, int port,
+	      unsigned int devid, unsigned int offset,
+	      unsigned int len, u8 *buf);
+int t4_load_boot(struct adapter *adap, u8 *boot_data,
+		 unsigned int boot_addr, unsigned int size);
+int t4_load_bootcfg(struct adapter *adap,
+		    const u8 *cfg_data, unsigned int size);
 void free_rspq_fl(struct adapter *adap, struct sge_rspq *rq, struct sge_fl *fl);
 void free_tx_desc(struct adapter *adap, struct sge_txq *q,
 		  unsigned int n, bool unmap);
+void cxgb4_eosw_txq_free_desc(struct adapter *adap, struct sge_eosw_txq *txq,
+			      u32 ndesc);
+int cxgb4_ethofld_send_flowc(struct net_device *dev, u32 eotid, u32 tc);
+void cxgb4_ethofld_restart(struct tasklet_struct *t);
+int cxgb4_ethofld_rx_handler(struct sge_rspq *q, const __be64 *rsp,
+			     const struct pkt_gl *si);
 void free_txq(struct adapter *adap, struct sge_txq *q);
+void cxgb4_reclaim_completed_tx(struct adapter *adap,
+				struct sge_txq *q, bool unmap);
+int cxgb4_map_skb(struct device *dev, const struct sk_buff *skb,
+		  dma_addr_t *addr);
+void cxgb4_inline_tx_skb(const struct sk_buff *skb, const struct sge_txq *q,
+			 void *pos);
+void cxgb4_write_sgl(const struct sk_buff *skb, struct sge_txq *q,
+		     struct ulptx_sgl *sgl, u64 *end, unsigned int start,
+		     const dma_addr_t *addr);
+void cxgb4_write_partial_sgl(const struct sk_buff *skb, struct sge_txq *q,
+			     struct ulptx_sgl *sgl, u64 *end,
+			     const dma_addr_t *addr, u32 start, u32 send_len);
+void cxgb4_ring_tx_db(struct adapter *adap, struct sge_txq *q, int n);
+int t4_set_vlan_acl(struct adapter *adap, unsigned int mbox, unsigned int vf,
+		    u16 vlan);
+int cxgb4_dcb_enabled(const struct net_device *dev);
+
+int cxgb4_thermal_init(struct adapter *adap);
+int cxgb4_thermal_remove(struct adapter *adap);
+int cxgb4_set_msix_aff(struct adapter *adap, unsigned short vec,
+		       cpumask_var_t *aff_mask, int idx);
+void cxgb4_clear_msix_aff(unsigned short vec, cpumask_var_t aff_mask);
+
+int cxgb4_change_mac(struct port_info *pi, unsigned int viid,
+		     int *tcam_idx, const u8 *addr,
+		     bool persistent, u8 *smt_idx);
+
+int cxgb4_alloc_mac_filt(struct adapter *adap, unsigned int viid,
+			 bool free, unsigned int naddr,
+			 const u8 **addr, u16 *idx,
+			 u64 *hash, bool sleep_ok);
+int cxgb4_free_mac_filt(struct adapter *adap, unsigned int viid,
+			unsigned int naddr, const u8 **addr, bool sleep_ok);
+int cxgb4_init_mps_ref_entries(struct adapter *adap);
+void cxgb4_free_mps_ref_entries(struct adapter *adap);
+int cxgb4_update_mac_filt(struct port_info *pi, unsigned int viid,
+			  int *tcam_idx, const u8 *addr,
+			  bool persistent, u8 *smt_idx);
+int cxgb4_get_msix_idx_from_bmap(struct adapter *adap);
+void cxgb4_free_msix_idx_in_bmap(struct adapter *adap, u32 msix_idx);
+void cxgb4_enable_rx(struct adapter *adap, struct sge_rspq *q);
+void cxgb4_quiesce_rx(struct sge_rspq *q);
+int cxgb4_port_mirror_alloc(struct net_device *dev);
+void cxgb4_port_mirror_free(struct net_device *dev);
+#if IS_ENABLED(CONFIG_CHELSIO_TLS_DEVICE)
+int cxgb4_set_ktls_feature(struct adapter *adap, bool enable);
+#endif
 #endif /* __CXGB4_H__ */
diff --git a/drivers/net/ethernet/chelsio/cxgb4/cxgb4_cudbg.c b/drivers/net/ethernet/chelsio/cxgb4/cxgb4_cudbg.c
new file mode 100644
index 000000000000..dd66b244466d
--- /dev/null
+++ b/drivers/net/ethernet/chelsio/cxgb4/cxgb4_cudbg.c
@@ -0,0 +1,279 @@
+// SPDX-License-Identifier: GPL-2.0-only
+/*
+ *  Copyright (C) 2017 Chelsio Communications.  All rights reserved.
+ */
+
+#include "t4_regs.h"
+#include "cxgb4.h"
+#include "cxgb4_cudbg.h"
+#include "cudbg_zlib.h"
+
+static const struct cxgb4_collect_entity cxgb4_collect_mem_dump[] = {
+	{ CUDBG_EDC0, cudbg_collect_edc0_meminfo },
+	{ CUDBG_EDC1, cudbg_collect_edc1_meminfo },
+	{ CUDBG_MC0, cudbg_collect_mc0_meminfo },
+	{ CUDBG_MC1, cudbg_collect_mc1_meminfo },
+	{ CUDBG_HMA, cudbg_collect_hma_meminfo },
+};
+
+static const struct cxgb4_collect_entity cxgb4_collect_hw_dump[] = {
+	{ CUDBG_MBOX_LOG, cudbg_collect_mbox_log },
+	{ CUDBG_QDESC, cudbg_collect_qdesc },
+	{ CUDBG_DEV_LOG, cudbg_collect_fw_devlog },
+	{ CUDBG_REG_DUMP, cudbg_collect_reg_dump },
+	{ CUDBG_CIM_LA, cudbg_collect_cim_la },
+	{ CUDBG_CIM_MA_LA, cudbg_collect_cim_ma_la },
+	{ CUDBG_CIM_QCFG, cudbg_collect_cim_qcfg },
+	{ CUDBG_CIM_IBQ_TP0, cudbg_collect_cim_ibq_tp0 },
+	{ CUDBG_CIM_IBQ_TP1, cudbg_collect_cim_ibq_tp1 },
+	{ CUDBG_CIM_IBQ_ULP, cudbg_collect_cim_ibq_ulp },
+	{ CUDBG_CIM_IBQ_SGE0, cudbg_collect_cim_ibq_sge0 },
+	{ CUDBG_CIM_IBQ_SGE1, cudbg_collect_cim_ibq_sge1 },
+	{ CUDBG_CIM_IBQ_NCSI, cudbg_collect_cim_ibq_ncsi },
+	{ CUDBG_CIM_OBQ_ULP0, cudbg_collect_cim_obq_ulp0 },
+	{ CUDBG_CIM_OBQ_ULP1, cudbg_collect_cim_obq_ulp1 },
+	{ CUDBG_CIM_OBQ_ULP2, cudbg_collect_cim_obq_ulp2 },
+	{ CUDBG_CIM_OBQ_ULP3, cudbg_collect_cim_obq_ulp3 },
+	{ CUDBG_CIM_OBQ_SGE, cudbg_collect_cim_obq_sge },
+	{ CUDBG_CIM_OBQ_NCSI, cudbg_collect_cim_obq_ncsi },
+	{ CUDBG_RSS, cudbg_collect_rss },
+	{ CUDBG_RSS_VF_CONF, cudbg_collect_rss_vf_config },
+	{ CUDBG_PATH_MTU, cudbg_collect_path_mtu },
+	{ CUDBG_PM_STATS, cudbg_collect_pm_stats },
+	{ CUDBG_HW_SCHED, cudbg_collect_hw_sched },
+	{ CUDBG_TP_INDIRECT, cudbg_collect_tp_indirect },
+	{ CUDBG_SGE_INDIRECT, cudbg_collect_sge_indirect },
+	{ CUDBG_ULPRX_LA, cudbg_collect_ulprx_la },
+	{ CUDBG_TP_LA, cudbg_collect_tp_la },
+	{ CUDBG_MEMINFO, cudbg_collect_meminfo },
+	{ CUDBG_CIM_PIF_LA, cudbg_collect_cim_pif_la },
+	{ CUDBG_CLK, cudbg_collect_clk_info },
+	{ CUDBG_CIM_OBQ_RXQ0, cudbg_collect_obq_sge_rx_q0 },
+	{ CUDBG_CIM_OBQ_RXQ1, cudbg_collect_obq_sge_rx_q1 },
+	{ CUDBG_PCIE_INDIRECT, cudbg_collect_pcie_indirect },
+	{ CUDBG_PM_INDIRECT, cudbg_collect_pm_indirect },
+	{ CUDBG_TID_INFO, cudbg_collect_tid },
+	{ CUDBG_PCIE_CONFIG, cudbg_collect_pcie_config },
+	{ CUDBG_DUMP_CONTEXT, cudbg_collect_dump_context },
+	{ CUDBG_MPS_TCAM, cudbg_collect_mps_tcam },
+	{ CUDBG_VPD_DATA, cudbg_collect_vpd_data },
+	{ CUDBG_LE_TCAM, cudbg_collect_le_tcam },
+	{ CUDBG_CCTRL, cudbg_collect_cctrl },
+	{ CUDBG_MA_INDIRECT, cudbg_collect_ma_indirect },
+	{ CUDBG_ULPTX_LA, cudbg_collect_ulptx_la },
+	{ CUDBG_UP_CIM_INDIRECT, cudbg_collect_up_cim_indirect },
+	{ CUDBG_PBT_TABLE, cudbg_collect_pbt_tables },
+	{ CUDBG_HMA_INDIRECT, cudbg_collect_hma_indirect },
+};
+
+static const struct cxgb4_collect_entity cxgb4_collect_flash_dump[] = {
+	{ CUDBG_FLASH, cudbg_collect_flash },
+};
+
+u32 cxgb4_get_dump_length(struct adapter *adap, u32 flag)
+{
+	u32 i, entity;
+	u32 len = 0;
+	u32 wsize;
+
+	if (flag & CXGB4_ETH_DUMP_HW) {
+		for (i = 0; i < ARRAY_SIZE(cxgb4_collect_hw_dump); i++) {
+			entity = cxgb4_collect_hw_dump[i].entity;
+			len += cudbg_get_entity_length(adap, entity);
+		}
+	}
+
+	if (flag & CXGB4_ETH_DUMP_MEM) {
+		for (i = 0; i < ARRAY_SIZE(cxgb4_collect_mem_dump); i++) {
+			entity = cxgb4_collect_mem_dump[i].entity;
+			len += cudbg_get_entity_length(adap, entity);
+		}
+	}
+
+	if (flag & CXGB4_ETH_DUMP_FLASH)
+		len += adap->params.sf_size;
+
+	/* If compression is enabled, a smaller destination buffer is enough */
+	wsize = cudbg_get_workspace_size();
+	if (wsize && len > CUDBG_DUMP_BUFF_SIZE)
+		len = CUDBG_DUMP_BUFF_SIZE;
+
+	return len;
+}
+
+static void cxgb4_cudbg_collect_entity(struct cudbg_init *pdbg_init,
+				       struct cudbg_buffer *dbg_buff,
+				       const struct cxgb4_collect_entity *e_arr,
+				       u32 arr_size, void *buf, u32 *tot_size)
+{
+	struct cudbg_error cudbg_err = { 0 };
+	struct cudbg_entity_hdr *entity_hdr;
+	u32 i, total_size = 0;
+	int ret;
+
+	for (i = 0; i < arr_size; i++) {
+		const struct cxgb4_collect_entity *e = &e_arr[i];
+
+		entity_hdr = cudbg_get_entity_hdr(buf, e->entity);
+		entity_hdr->entity_type = e->entity;
+		entity_hdr->start_offset = dbg_buff->offset;
+		memset(&cudbg_err, 0, sizeof(struct cudbg_error));
+		ret = e->collect_cb(pdbg_init, dbg_buff, &cudbg_err);
+		if (ret) {
+			entity_hdr->size = 0;
+			dbg_buff->offset = entity_hdr->start_offset;
+		} else {
+			cudbg_align_debug_buffer(dbg_buff, entity_hdr);
+		}
+
+		/* Log error and continue with next entity */
+		if (cudbg_err.sys_err)
+			ret = CUDBG_SYSTEM_ERROR;
+
+		entity_hdr->hdr_flags = ret;
+		entity_hdr->sys_err = cudbg_err.sys_err;
+		entity_hdr->sys_warn = cudbg_err.sys_warn;
+		total_size += entity_hdr->size;
+	}
+
+	*tot_size += total_size;
+}
+
+static int cudbg_alloc_compress_buff(struct cudbg_init *pdbg_init)
+{
+	u32 workspace_size;
+
+	workspace_size = cudbg_get_workspace_size();
+	pdbg_init->compress_buff = vzalloc(CUDBG_COMPRESS_BUFF_SIZE +
+					   workspace_size);
+	if (!pdbg_init->compress_buff)
+		return -ENOMEM;
+
+	pdbg_init->compress_buff_size = CUDBG_COMPRESS_BUFF_SIZE;
+	pdbg_init->workspace = (u8 *)pdbg_init->compress_buff +
+			       CUDBG_COMPRESS_BUFF_SIZE - workspace_size;
+	return 0;
+}
+
+static void cudbg_free_compress_buff(struct cudbg_init *pdbg_init)
+{
+	vfree(pdbg_init->compress_buff);
+}
+
+int cxgb4_cudbg_collect(struct adapter *adap, void *buf, u32 *buf_size,
+			u32 flag)
+{
+	struct cudbg_buffer dbg_buff = { 0 };
+	u32 size, min_size, total_size = 0;
+	struct cudbg_init cudbg_init;
+	struct cudbg_hdr *cudbg_hdr;
+	int rc;
+
+	size = *buf_size;
+
+	memset(&cudbg_init, 0, sizeof(struct cudbg_init));
+	cudbg_init.adap = adap;
+	cudbg_init.outbuf = buf;
+	cudbg_init.outbuf_size = size;
+
+	dbg_buff.data = buf;
+	dbg_buff.size = size;
+	dbg_buff.offset = 0;
+
+	cudbg_hdr = (struct cudbg_hdr *)buf;
+	cudbg_hdr->signature = CUDBG_SIGNATURE;
+	cudbg_hdr->hdr_len = sizeof(struct cudbg_hdr);
+	cudbg_hdr->major_ver = CUDBG_MAJOR_VERSION;
+	cudbg_hdr->minor_ver = CUDBG_MINOR_VERSION;
+	cudbg_hdr->max_entities = CUDBG_MAX_ENTITY;
+	cudbg_hdr->chip_ver = adap->params.chip;
+	cudbg_hdr->dump_type = CUDBG_DUMP_TYPE_MINI;
+
+	min_size = sizeof(struct cudbg_hdr) +
+		   sizeof(struct cudbg_entity_hdr) *
+		   cudbg_hdr->max_entities;
+	if (size < min_size)
+		return -ENOMEM;
+
+	rc = cudbg_get_workspace_size();
+	if (rc) {
+		/* Zlib available.  So, use zlib deflate */
+		cudbg_init.compress_type = CUDBG_COMPRESSION_ZLIB;
+		rc = cudbg_alloc_compress_buff(&cudbg_init);
+		if (rc) {
+			/* Ignore error and continue without compression. */
+			dev_warn(adap->pdev_dev,
+				 "Fail allocating compression buffer ret: %d.  Continuing without compression.\n",
+				 rc);
+			cudbg_init.compress_type = CUDBG_COMPRESSION_NONE;
+			rc = 0;
+		}
+	} else {
+		cudbg_init.compress_type = CUDBG_COMPRESSION_NONE;
+	}
+
+	cudbg_hdr->compress_type = cudbg_init.compress_type;
+	dbg_buff.offset += min_size;
+	total_size = dbg_buff.offset;
+
+	if (flag & CXGB4_ETH_DUMP_HW)
+		cxgb4_cudbg_collect_entity(&cudbg_init, &dbg_buff,
+					   cxgb4_collect_hw_dump,
+					   ARRAY_SIZE(cxgb4_collect_hw_dump),
+					   buf,
+					   &total_size);
+
+	if (flag & CXGB4_ETH_DUMP_MEM)
+		cxgb4_cudbg_collect_entity(&cudbg_init, &dbg_buff,
+					   cxgb4_collect_mem_dump,
+					   ARRAY_SIZE(cxgb4_collect_mem_dump),
+					   buf,
+					   &total_size);
+
+	if (flag & CXGB4_ETH_DUMP_FLASH)
+		cxgb4_cudbg_collect_entity(&cudbg_init, &dbg_buff,
+					   cxgb4_collect_flash_dump,
+					   ARRAY_SIZE(cxgb4_collect_flash_dump),
+					   buf,
+					   &total_size);
+
+	cudbg_free_compress_buff(&cudbg_init);
+	cudbg_hdr->data_len = total_size;
+	if (cudbg_init.compress_type != CUDBG_COMPRESSION_NONE)
+		*buf_size = size;
+	else
+		*buf_size = total_size;
+	return 0;
+}
+
+void cxgb4_init_ethtool_dump(struct adapter *adapter)
+{
+	adapter->eth_dump.flag = CXGB4_ETH_DUMP_NONE;
+	adapter->eth_dump.version = adapter->params.fw_vers;
+	adapter->eth_dump.len = 0;
+}
+
+static int cxgb4_cudbg_vmcoredd_collect(struct vmcoredd_data *data, void *buf)
+{
+	struct adapter *adap = container_of(data, struct adapter, vmcoredd);
+	u32 len = data->size;
+
+	return cxgb4_cudbg_collect(adap, buf, &len, CXGB4_ETH_DUMP_ALL);
+}
+
+int cxgb4_cudbg_vmcore_add_dump(struct adapter *adap)
+{
+	struct vmcoredd_data *data = &adap->vmcoredd;
+	u32 len;
+
+	len = sizeof(struct cudbg_hdr) +
+	      sizeof(struct cudbg_entity_hdr) * CUDBG_MAX_ENTITY;
+	len += CUDBG_DUMP_BUFF_SIZE;
+
+	data->size = len;
+	snprintf(data->dump_name, sizeof(data->dump_name), "%s_%s",
+		 cxgb4_driver_name, adap->name);
+	data->vmcoredd_callback = cxgb4_cudbg_vmcoredd_collect;
+
+	return vmcore_add_device_dump(data);
+}
diff --git a/drivers/net/ethernet/chelsio/cxgb4/cxgb4_cudbg.h b/drivers/net/ethernet/chelsio/cxgb4/cxgb4_cudbg.h
new file mode 100644
index 000000000000..c04a49b6378d
--- /dev/null
+++ b/drivers/net/ethernet/chelsio/cxgb4/cxgb4_cudbg.h
@@ -0,0 +1,40 @@
+/* SPDX-License-Identifier: GPL-2.0-only */
+/*
+ *  Copyright (C) 2017 Chelsio Communications.  All rights reserved.
+ */
+
+#ifndef __CXGB4_CUDBG_H__
+#define __CXGB4_CUDBG_H__
+
+#include "cudbg_if.h"
+#include "cudbg_lib_common.h"
+#include "cudbg_entity.h"
+#include "cudbg_lib.h"
+
+#define CUDBG_DUMP_BUFF_SIZE (32 * 1024 * 1024) /* 32 MB */
+#define CUDBG_COMPRESS_BUFF_SIZE (4 * 1024 * 1024) /* 4 MB */
+
+typedef int (*cudbg_collect_callback_t)(struct cudbg_init *pdbg_init,
+					struct cudbg_buffer *dbg_buff,
+					struct cudbg_error *cudbg_err);
+
+struct cxgb4_collect_entity {
+	enum cudbg_dbg_entity_type entity;
+	cudbg_collect_callback_t collect_cb;
+};
+
+enum CXGB4_ETHTOOL_DUMP_FLAGS {
+	CXGB4_ETH_DUMP_NONE = ETH_FW_DUMP_DISABLE,
+	CXGB4_ETH_DUMP_MEM = (1 << 0), /* On-Chip Memory Dumps */
+	CXGB4_ETH_DUMP_HW = (1 << 1), /* various FW and HW dumps */
+	CXGB4_ETH_DUMP_FLASH = (1 << 2), /* Dump flash memory */
+};
+
+#define CXGB4_ETH_DUMP_ALL (CXGB4_ETH_DUMP_MEM | CXGB4_ETH_DUMP_HW)
+
+u32 cxgb4_get_dump_length(struct adapter *adap, u32 flag);
+int cxgb4_cudbg_collect(struct adapter *adap, void *buf, u32 *buf_size,
+			u32 flag);
+void cxgb4_init_ethtool_dump(struct adapter *adapter);
+int cxgb4_cudbg_vmcore_add_dump(struct adapter *adap);
+#endif /* __CXGB4_CUDBG_H__ */
diff --git a/drivers/net/ethernet/chelsio/cxgb4/cxgb4_dcb.c b/drivers/net/ethernet/chelsio/cxgb4/cxgb4_dcb.c
index 6ee2ed30626b..7d5204834ee2 100644
--- a/drivers/net/ethernet/chelsio/cxgb4/cxgb4_dcb.c
+++ b/drivers/net/ethernet/chelsio/cxgb4/cxgb4_dcb.c
@@ -1,28 +1,16 @@
+// SPDX-License-Identifier: GPL-2.0-only
 /*
  *  Copyright (C) 2013-2014 Chelsio Communications.  All rights reserved.
  *
  *  Written by Anish Bhatt (anish@chelsio.com)
  *	       Casey Leedom (leedom@chelsio.com)
- *
- *  This program is free software; you can redistribute it and/or modify it
- *  under the terms and conditions of the GNU General Public License,
- *  version 2, as published by the Free Software Foundation.
- *
- *  This program is distributed in the hope it will be useful, but WITHOUT
- * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
- *  FITNESS FOR A PARTICULAR PURPOSE.  See the GNU General Public License for
- *  more details.
- *
- *  The full GNU General Public License is included in this distribution in
- *  the file called "COPYING".
- *
  */
 
 #include "cxgb4.h"
 
 /* DCBx version control
  */
-static const char * const dcb_ver_array[] = {
+const char * const dcb_ver_array[] = {
 	"Unknown",
 	"DCBx-CIN",
 	"DCBx-CEE 1.01",
@@ -40,8 +28,7 @@ static inline bool cxgb4_dcb_state_synced(enum cxgb4_dcb_state state)
 		return false;
 }
 
-/* Initialize a port's Data Center Bridging state.  Typically used after a
- * Link Down event.
+/* Initialize a port's Data Center Bridging state.
  */
 void cxgb4_dcb_state_init(struct net_device *dev)
 {
@@ -98,7 +85,7 @@ static void cxgb4_dcb_cleanup_apps(struct net_device *dev)
 
 		if (err) {
 			dev_err(adap->pdev_dev,
-				"Failed DCB Clear %s Application Priority: sel=%d, prot=%d, , err=%d\n",
+				"Failed DCB Clear %s Application Priority: sel=%d, prot=%d, err=%d\n",
 				dcb_ver_array[dcb->dcb_version], app.selector,
 				app.protocol, -err);
 			break;
@@ -106,6 +93,33 @@ static void cxgb4_dcb_cleanup_apps(struct net_device *dev)
 	}
 }
 
+/* Reset a port's Data Center Bridging state.  Typically used after a
+ * Link Down event.
+ */
+void cxgb4_dcb_reset(struct net_device *dev)
+{
+	cxgb4_dcb_cleanup_apps(dev);
+	cxgb4_dcb_state_init(dev);
+}
+
+/* update the dcb port support, if version is IEEE then set it to
+ * FW_PORT_DCB_VER_IEEE and if DCB_CAP_DCBX_VER_CEE is already set then
+ * clear that. and if it is set to CEE then set dcb supported to
+ * DCB_CAP_DCBX_VER_CEE & if DCB_CAP_DCBX_VER_IEEE is set, clear it
+ */
+static inline void cxgb4_dcb_update_support(struct port_dcb_info *dcb)
+{
+	if (dcb->dcb_version == FW_PORT_DCB_VER_IEEE) {
+		if (dcb->supported & DCB_CAP_DCBX_VER_CEE)
+			dcb->supported &= ~DCB_CAP_DCBX_VER_CEE;
+		dcb->supported |= DCB_CAP_DCBX_VER_IEEE;
+	} else if (dcb->dcb_version == FW_PORT_DCB_VER_CEE1D01) {
+		if (dcb->supported & DCB_CAP_DCBX_VER_IEEE)
+			dcb->supported &= ~DCB_CAP_DCBX_VER_IEEE;
+		dcb->supported |= DCB_CAP_DCBX_VER_CEE;
+	}
+}
+
 /* Finite State machine for Data Center Bridging.
  */
 void cxgb4_dcb_state_fsm(struct net_device *dev,
@@ -157,6 +171,15 @@ void cxgb4_dcb_state_fsm(struct net_device *dev,
 	}
 
 	case CXGB4_DCB_STATE_FW_INCOMPLETE: {
+		if (transition_to != CXGB4_DCB_INPUT_FW_DISABLED) {
+			/* during this CXGB4_DCB_STATE_FW_INCOMPLETE state,
+			 * check if the dcb version is changed (there can be
+			 * mismatch in default config & the negotiated switch
+			 * configuration at FW, so update the dcb support
+			 * accordingly.
+			 */
+			cxgb4_dcb_update_support(dcb);
+		}
 		switch (transition_to) {
 		case CXGB4_DCB_INPUT_FW_ENABLED: {
 			/* we're alreaady in firmware DCB mode */
@@ -194,8 +217,7 @@ void cxgb4_dcb_state_fsm(struct net_device *dev,
 			 * state.  We need to reset back to a ground state
 			 * of incomplete.
 			 */
-			cxgb4_dcb_cleanup_apps(dev);
-			cxgb4_dcb_state_init(dev);
+			cxgb4_dcb_reset(dev);
 			dcb->state = CXGB4_DCB_STATE_FW_INCOMPLETE;
 			dcb->supported = CXGB4_DCBX_FW_SUPPORT;
 			linkwatch_fire_event(dev);
@@ -266,8 +288,8 @@ void cxgb4_dcb_handle_fw_update(struct adapter *adap,
 		enum cxgb4_dcb_state_input input =
 			((pcmd->u.dcb.control.all_syncd_pkd &
 			  FW_PORT_CMD_ALL_SYNCD_F)
-			 ? CXGB4_DCB_STATE_FW_ALLSYNCED
-			 : CXGB4_DCB_STATE_FW_INCOMPLETE);
+			 ? CXGB4_DCB_INPUT_FW_ALLSYNCED
+			 : CXGB4_DCB_INPUT_FW_INCOMPLETE);
 
 		if (dcb->dcb_version != FW_PORT_DCB_VER_UNKNOWN) {
 			dcb_running_version = FW_PORT_CMD_DCB_VERSION_G(
diff --git a/drivers/net/ethernet/chelsio/cxgb4/cxgb4_dcb.h b/drivers/net/ethernet/chelsio/cxgb4/cxgb4_dcb.h
index ccf24d3dc982..c80a93347a8c 100644
--- a/drivers/net/ethernet/chelsio/cxgb4/cxgb4_dcb.h
+++ b/drivers/net/ethernet/chelsio/cxgb4/cxgb4_dcb.h
@@ -1,20 +1,8 @@
+/* SPDX-License-Identifier: GPL-2.0-only */
 /*
  *  Copyright (C) 2013-2014 Chelsio Communications.  All rights reserved.
  *
  *  Written by Anish Bhatt (anish@chelsio.com)
- *
- *  This program is free software; you can redistribute it and/or modify it
- *  under the terms and conditions of the GNU General Public License,
- *  version 2, as published by the Free Software Foundation.
- *
- *  This program is distributed in the hope it will be useful, but WITHOUT
- *  ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
- *  FITNESS FOR A PARTICULAR PURPOSE.  See the GNU General Public License for
- *  more details.
- *
- *  The full GNU General Public License is included in this distribution in
- *  the file called "COPYING".
- *
  */
 
 #ifndef __CXGB4_DCB_H
@@ -67,7 +55,7 @@
 	do { \
 		if ((__dcb)->dcb_version == FW_PORT_DCB_VER_IEEE) \
 			cxgb4_dcb_state_fsm((__dev), \
-					    CXGB4_DCB_STATE_FW_ALLSYNCED); \
+					    CXGB4_DCB_INPUT_FW_ALLSYNCED); \
 	} while (0)
 
 /* States we can be in for a port's Data Center Bridging.
@@ -131,9 +119,9 @@ struct port_dcb_info {
 
 void cxgb4_dcb_state_init(struct net_device *);
 void cxgb4_dcb_version_init(struct net_device *);
+void cxgb4_dcb_reset(struct net_device *dev);
 void cxgb4_dcb_state_fsm(struct net_device *, enum cxgb4_dcb_state_input);
 void cxgb4_dcb_handle_fw_update(struct adapter *, const struct fw_port_cmd *);
-void cxgb4_dcb_set_caps(struct adapter *, const struct fw_port_cmd *);
 extern const struct dcbnl_rtnl_ops cxgb4_dcb_ops;
 
 static inline __u8 bitswap_1(unsigned char val)
@@ -147,6 +135,9 @@ static inline __u8 bitswap_1(unsigned char val)
 	       ((val & 0x02) << 5) |
 	       ((val & 0x01) << 7);
 }
+
+extern const char * const dcb_ver_array[];
+
 #define CXGB4_DCB_ENABLED true
 
 #else /* !CONFIG_CHELSIO_T4_DCB */
diff --git a/drivers/net/ethernet/chelsio/cxgb4/cxgb4_debugfs.c b/drivers/net/ethernet/chelsio/cxgb4/cxgb4_debugfs.c
index 1fa34b009891..14e0d989c3ba 100644
--- a/drivers/net/ethernet/chelsio/cxgb4/cxgb4_debugfs.c
+++ b/drivers/net/ethernet/chelsio/cxgb4/cxgb4_debugfs.c
@@ -45,6 +45,11 @@
 #include "cxgb4_debugfs.h"
 #include "clip_tbl.h"
 #include "l2t.h"
+#include "cudbg_if.h"
+#include "cudbg_lib_common.h"
+#include "cudbg_entity.h"
+#include "cudbg_lib.h"
+#include "cxgb4_tc_mqprio.h"
 
 /* generic seq_file support for showing a table of size rows x width. */
 static void *seq_tab_get_idx(struct seq_tab *tb, loff_t pos)
@@ -66,8 +71,7 @@ static void *seq_tab_start(struct seq_file *seq, loff_t *pos)
 static void *seq_tab_next(struct seq_file *seq, void *v, loff_t *pos)
 {
 	v = seq_tab_get_idx(seq->private, *pos + 1);
-	if (v)
-		++*pos;
+	++(*pos);
 	return v;
 }
 
@@ -374,19 +378,7 @@ static int cim_qcfg_show(struct seq_file *seq, void *v)
 			   QUEREMFLITS_G(p[2]) * 16);
 	return 0;
 }
-
-static int cim_qcfg_open(struct inode *inode, struct file *file)
-{
-	return single_open(file, cim_qcfg_show, inode->i_private);
-}
-
-static const struct file_operations cim_qcfg_fops = {
-	.owner   = THIS_MODULE,
-	.open    = cim_qcfg_open,
-	.read    = seq_read,
-	.llseek  = seq_lseek,
-	.release = single_release,
-};
+DEFINE_SHOW_ATTRIBUTE(cim_qcfg);
 
 static int cimq_show(struct seq_file *seq, void *v, int idx)
 {
@@ -856,8 +848,7 @@ static int tx_rate_show(struct seq_file *seq, void *v)
 	}
 	return 0;
 }
-
-DEFINE_SIMPLE_DEBUGFS_FILE(tx_rate);
+DEFINE_SHOW_ATTRIBUTE(tx_rate);
 
 static int cctrl_tbl_show(struct seq_file *seq, void *v)
 {
@@ -869,7 +860,7 @@ static int cctrl_tbl_show(struct seq_file *seq, void *v)
 	u16 (*incr)[NCCTRL_WIN];
 	struct adapter *adap = seq->private;
 
-	incr = kmalloc(sizeof(*incr) * NMTUS, GFP_KERNEL);
+	incr = kmalloc_array(NMTUS, sizeof(*incr), GFP_KERNEL);
 	if (!incr)
 		return -ENOMEM;
 
@@ -889,8 +880,7 @@ static int cctrl_tbl_show(struct seq_file *seq, void *v)
 	kfree(incr);
 	return 0;
 }
-
-DEFINE_SIMPLE_DEBUGFS_FILE(cctrl_tbl);
+DEFINE_SHOW_ATTRIBUTE(cctrl_tbl);
 
 /* Format a value in a unit that differs from the value's native unit by the
  * given factor.
@@ -951,8 +941,7 @@ static int clk_show(struct seq_file *seq, void *v)
 
 	return 0;
 }
-
-DEFINE_SIMPLE_DEBUGFS_FILE(clk);
+DEFINE_SHOW_ATTRIBUTE(clk);
 
 /* Firmware Device Log dump. */
 static const char * const devlog_level_strings[] = {
@@ -996,7 +985,7 @@ static const char * const devlog_facility_strings[] = {
 struct devlog_info {
 	unsigned int nentries;		/* number of entries in log[] */
 	unsigned int first;		/* first [temporal] entry in log[] */
-	struct fw_devlog_e log[0];	/* Firmware Device Log */
+	struct fw_devlog_e log[];	/* Firmware Device Log */
 };
 
 /* Dump a Firmaware Device Log entry.
@@ -1739,7 +1728,7 @@ static int mps_tcam_show(struct seq_file *seq, void *v)
 			 */
 			if (lookup_type && (lookup_type != DATALKPTYPE_M)) {
 				/* Inner header VNI */
-				vniy = ((data2 & DATAVIDH2_F) << 23) |
+				vniy = (data2 & DATAVIDH2_F) |
 				       (DATAVIDH1_G(data2) << 16) | VIDL_G(val);
 				dip_hit = data2 & DATADIPHIT_F;
 			} else {
@@ -1749,6 +1738,7 @@ static int mps_tcam_show(struct seq_file *seq, void *v)
 			port_num = DATAPORTNUM_G(data2);
 
 			/* Read tcamx. Change the control param */
+			vnix = 0;
 			ctl |= CTLXYBITSEL_V(1);
 			t4_write_reg(adap, MPS_CLS_TCAM_DATA2_CTL_A, ctl);
 			val = t4_read_reg(adap, MPS_CLS_TCAM_DATA1_A);
@@ -1757,7 +1747,7 @@ static int mps_tcam_show(struct seq_file *seq, void *v)
 			data2 = t4_read_reg(adap, MPS_CLS_TCAM_DATA2_CTL_A);
 			if (lookup_type && (lookup_type != DATALKPTYPE_M)) {
 				/* Inner header VNI mask */
-				vnix = ((data2 & DATAVIDH2_F) << 23) |
+				vnix = (data2 & DATAVIDH2_F) |
 				       (DATAVIDH1_G(data2) << 16) | VIDL_G(val);
 			}
 		} else {
@@ -1823,14 +1813,11 @@ static int mps_tcam_show(struct seq_file *seq, void *v)
 			/* Inner header lookup */
 			if (lookup_type && (lookup_type != DATALKPTYPE_M)) {
 				seq_printf(seq,
-					   "%3u %02x:%02x:%02x:%02x:%02x:%02x "
-					   "%012llx %06x %06x    -    -   %3c"
-					   "      'I'  %4x   "
-					   "%3c   %#x%4u%4d", idx, addr[0],
-					   addr[1], addr[2], addr[3],
-					   addr[4], addr[5],
+					   "%3u %pM %012llx %06x %06x    -    -   %3c      'I'  %4x   %3c   %#x%4u%4d",
+					   idx, addr,
 					   (unsigned long long)mask,
-					   vniy, vnix, dip_hit ? 'Y' : 'N',
+					   vniy, (vnix | vniy),
+					   dip_hit ? 'Y' : 'N',
 					   port_num,
 					   (cls_lo & T6_SRAM_VLD_F) ? 'Y' : 'N',
 					   PORTMAP_G(cls_hi),
@@ -1839,10 +1826,8 @@ static int mps_tcam_show(struct seq_file *seq, void *v)
 					   T6_VF_G(cls_lo) : -1);
 			} else {
 				seq_printf(seq,
-					   "%3u %02x:%02x:%02x:%02x:%02x:%02x "
-					   "%012llx    -       -   ",
-					   idx, addr[0], addr[1], addr[2],
-					   addr[3], addr[4], addr[5],
+					   "%3u %pM %012llx    -       -   ",
+					   idx, addr,
 					   (unsigned long long)mask);
 
 				if (vlan_vld)
@@ -1860,10 +1845,8 @@ static int mps_tcam_show(struct seq_file *seq, void *v)
 					   T6_VF_G(cls_lo) : -1);
 			}
 		} else
-			seq_printf(seq, "%3u %02x:%02x:%02x:%02x:%02x:%02x "
-				   "%012llx%3c   %#x%4u%4d",
-				   idx, addr[0], addr[1], addr[2], addr[3],
-				   addr[4], addr[5], (unsigned long long)mask,
+			seq_printf(seq, "%3u %pM %012llx%3c   %#x%4u%4d",
+				   idx, addr, (unsigned long long)mask,
 				   (cls_lo & SRAM_VLD_F) ? 'Y' : 'N',
 				   PORTMAP_G(cls_hi),
 				   PF_G(cls_lo),
@@ -1984,22 +1967,10 @@ static int sensors_show(struct seq_file *seq, void *v)
 
 	return 0;
 }
-
-DEFINE_SIMPLE_DEBUGFS_FILE(sensors);
+DEFINE_SHOW_ATTRIBUTE(sensors);
 
 #if IS_ENABLED(CONFIG_IPV6)
-static int clip_tbl_open(struct inode *inode, struct file *file)
-{
-	return single_open(file, clip_tbl_show, inode->i_private);
-}
-
-static const struct file_operations clip_tbl_debugfs_fops = {
-	.owner   = THIS_MODULE,
-	.open    = clip_tbl_open,
-	.read    = seq_read,
-	.llseek  = seq_lseek,
-	.release = single_release
-};
+DEFINE_SHOW_ATTRIBUTE(clip_tbl);
 #endif
 
 /*RSS Table.
@@ -2017,11 +1988,12 @@ static int rss_show(struct seq_file *seq, void *v, int idx)
 
 static int rss_open(struct inode *inode, struct file *file)
 {
-	int ret;
-	struct seq_tab *p;
 	struct adapter *adap = inode->i_private;
+	int ret, nentries;
+	struct seq_tab *p;
 
-	p = seq_open_tab(file, RSS_NENTRIES / 8, 8 * sizeof(u16), 0, rss_show);
+	nentries = t4_chip_rss_size(adap);
+	p = seq_open_tab(file, nentries / 8, 8 * sizeof(u16), 0, rss_show);
 	if (!p)
 		return -ENOMEM;
 
@@ -2201,8 +2173,7 @@ static int rss_config_show(struct seq_file *seq, void *v)
 
 	return 0;
 }
-
-DEFINE_SIMPLE_DEBUGFS_FILE(rss_config);
+DEFINE_SHOW_ATTRIBUTE(rss_config);
 
 /* RSS Secret Key.
  */
@@ -2211,7 +2182,7 @@ static int rss_key_show(struct seq_file *seq, void *v)
 {
 	u32 key[10];
 
-	t4_read_rss_key(seq->private, key);
+	t4_read_rss_key(seq->private, key, true);
 	seq_printf(seq, "%08x%08x%08x%08x%08x%08x%08x%08x%08x%08x\n",
 		   key[9], key[8], key[7], key[6], key[5], key[4], key[3],
 		   key[2], key[1], key[0]);
@@ -2248,7 +2219,7 @@ static ssize_t rss_key_write(struct file *file, const char __user *buf,
 		}
 	}
 
-	t4_write_rss_key(adap, key, -1);
+	t4_write_rss_key(adap, key, -1, true);
 	return count;
 }
 
@@ -2325,12 +2296,13 @@ static int rss_pf_config_open(struct inode *inode, struct file *file)
 		return -ENOMEM;
 
 	pfconf = (struct rss_pf_conf *)p->data;
-	rss_pf_map = t4_read_rss_pf_map(adapter);
-	rss_pf_mask = t4_read_rss_pf_mask(adapter);
+	rss_pf_map = t4_read_rss_pf_map(adapter, true);
+	rss_pf_mask = t4_read_rss_pf_mask(adapter, true);
 	for (pf = 0; pf < 8; pf++) {
 		pfconf[pf].rss_pf_map = rss_pf_map;
 		pfconf[pf].rss_pf_mask = rss_pf_mask;
-		t4_read_rss_pf_config(adapter, pf, &pfconf[pf].rss_pf_config);
+		t4_read_rss_pf_config(adapter, pf, &pfconf[pf].rss_pf_config,
+				      true);
 	}
 	return 0;
 }
@@ -2393,7 +2365,7 @@ static int rss_vf_config_open(struct inode *inode, struct file *file)
 	vfconf = (struct rss_vf_conf *)p->data;
 	for (vf = 0; vf < vfcount; vf++) {
 		t4_read_rss_vf_config(adapter, vf, &vfconf[vf].rss_vf_vfl,
-				      &vfconf[vf].rss_vf_vfh);
+				      &vfconf[vf].rss_vf_vfh, true);
 	}
 	return 0;
 }
@@ -2406,6 +2378,221 @@ static const struct file_operations rss_vf_config_debugfs_fops = {
 	.release = seq_release_private
 };
 
+#ifdef CONFIG_CHELSIO_T4_DCB
+
+/* Data Center Briging information for each port.
+ */
+static int dcb_info_show(struct seq_file *seq, void *v)
+{
+	struct adapter *adap = seq->private;
+
+	if (v == SEQ_START_TOKEN) {
+		seq_puts(seq, "Data Center Bridging Information\n");
+	} else {
+		int port = (uintptr_t)v - 2;
+		struct net_device *dev = adap->port[port];
+		struct port_info *pi = netdev2pinfo(dev);
+		struct port_dcb_info *dcb = &pi->dcb;
+
+		seq_puts(seq, "\n");
+		seq_printf(seq, "Port: %d (DCB negotiated: %s)\n",
+			   port,
+			   cxgb4_dcb_enabled(dev) ? "yes" : "no");
+
+		if (cxgb4_dcb_enabled(dev))
+			seq_printf(seq, "[ DCBx Version %s ]\n",
+				   dcb_ver_array[dcb->dcb_version]);
+
+		if (dcb->msgs) {
+			int i;
+
+			seq_puts(seq, "\n  Index\t\t\t  :\t");
+			for (i = 0; i < 8; i++)
+				seq_printf(seq, " %3d", i);
+			seq_puts(seq, "\n\n");
+		}
+
+		if (dcb->msgs & CXGB4_DCB_FW_PGID) {
+			int prio, pgid;
+
+			seq_puts(seq, "  Priority Group IDs\t  :\t");
+			for (prio = 0; prio < 8; prio++) {
+				pgid = (dcb->pgid >> 4 * (7 - prio)) & 0xf;
+				seq_printf(seq, " %3d", pgid);
+			}
+			seq_puts(seq, "\n");
+		}
+
+		if (dcb->msgs & CXGB4_DCB_FW_PGRATE) {
+			int pg;
+
+			seq_puts(seq, "  Priority Group BW(%)\t  :\t");
+			for (pg = 0; pg < 8; pg++)
+				seq_printf(seq, " %3d", dcb->pgrate[pg]);
+			seq_puts(seq, "\n");
+
+			if (dcb->dcb_version == FW_PORT_DCB_VER_IEEE) {
+				seq_puts(seq, "  TSA Algorithm\t\t  :\t");
+				for (pg = 0; pg < 8; pg++)
+					seq_printf(seq, " %3d", dcb->tsa[pg]);
+				seq_puts(seq, "\n");
+			}
+
+			seq_printf(seq, "  Max PG Traffic Classes  [%3d  ]\n",
+				   dcb->pg_num_tcs_supported);
+
+			seq_puts(seq, "\n");
+		}
+
+		if (dcb->msgs & CXGB4_DCB_FW_PRIORATE) {
+			int prio;
+
+			seq_puts(seq, "  Priority Rate\t:\t");
+			for (prio = 0; prio < 8; prio++)
+				seq_printf(seq, " %3d", dcb->priorate[prio]);
+			seq_puts(seq, "\n");
+		}
+
+		if (dcb->msgs & CXGB4_DCB_FW_PFC) {
+			int prio;
+
+			seq_puts(seq, "  Priority Flow Control   :\t");
+			for (prio = 0; prio < 8; prio++) {
+				int pfcen = (dcb->pfcen >> 1 * (7 - prio))
+					    & 0x1;
+				seq_printf(seq, " %3d", pfcen);
+			}
+			seq_puts(seq, "\n");
+
+			seq_printf(seq, "  Max PFC Traffic Classes [%3d  ]\n",
+				   dcb->pfc_num_tcs_supported);
+
+			seq_puts(seq, "\n");
+		}
+
+		if (dcb->msgs & CXGB4_DCB_FW_APP_ID) {
+			int app, napps;
+
+			seq_puts(seq, "  Application Information:\n");
+			seq_puts(seq, "  App    Priority    Selection         Protocol\n");
+			seq_puts(seq, "  Index  Map         Field             ID\n");
+			for (app = 0, napps = 0;
+			     app < CXGB4_MAX_DCBX_APP_SUPPORTED; app++) {
+				struct app_priority *ap;
+				static const char * const sel_names[] = {
+					"Ethertype",
+					"Socket TCP",
+					"Socket UDP",
+					"Socket All",
+				};
+				const char *sel_name;
+
+				ap = &dcb->app_priority[app];
+				/* skip empty slots */
+				if (ap->protocolid == 0)
+					continue;
+				napps++;
+
+				if (ap->sel_field < ARRAY_SIZE(sel_names))
+					sel_name = sel_names[ap->sel_field];
+				else
+					sel_name = "UNKNOWN";
+
+				seq_printf(seq, "  %3d    %#04x        %-10s (%d)    %#06x (%d)\n",
+					   app,
+					   ap->user_prio_map,
+					   sel_name, ap->sel_field,
+					   ap->protocolid, ap->protocolid);
+			}
+			if (napps == 0)
+				seq_puts(seq, "    --- None ---\n");
+		}
+	}
+	return 0;
+}
+
+static inline void *dcb_info_get_idx(struct adapter *adap, loff_t pos)
+{
+	return (pos <= adap->params.nports
+		? (void *)((uintptr_t)pos + 1)
+		: NULL);
+}
+
+static void *dcb_info_start(struct seq_file *seq, loff_t *pos)
+{
+	struct adapter *adap = seq->private;
+
+	return (*pos
+		? dcb_info_get_idx(adap, *pos)
+		: SEQ_START_TOKEN);
+}
+
+static void dcb_info_stop(struct seq_file *seq, void *v)
+{
+}
+
+static void *dcb_info_next(struct seq_file *seq, void *v, loff_t *pos)
+{
+	struct adapter *adap = seq->private;
+
+	(*pos)++;
+	return dcb_info_get_idx(adap, *pos);
+}
+
+static const struct seq_operations dcb_info_seq_ops = {
+	.start = dcb_info_start,
+	.next  = dcb_info_next,
+	.stop  = dcb_info_stop,
+	.show  = dcb_info_show
+};
+
+static int dcb_info_open(struct inode *inode, struct file *file)
+{
+	int res = seq_open(file, &dcb_info_seq_ops);
+
+	if (!res) {
+		struct seq_file *seq = file->private_data;
+
+		seq->private = inode->i_private;
+	}
+	return res;
+}
+
+static const struct file_operations dcb_info_debugfs_fops = {
+	.owner   = THIS_MODULE,
+	.open    = dcb_info_open,
+	.read    = seq_read,
+	.llseek  = seq_lseek,
+	.release = seq_release,
+};
+#endif /* CONFIG_CHELSIO_T4_DCB */
+
+static int resources_show(struct seq_file *seq, void *v)
+{
+	struct adapter *adapter = seq->private;
+	struct pf_resources *pfres = &adapter->params.pfres;
+
+	#define S(desc, fmt, var) \
+		seq_printf(seq, "%-60s " fmt "\n", \
+			   desc " (" #var "):", pfres->var)
+
+	S("Virtual Interfaces", "%d", nvi);
+	S("Egress Queues", "%d", neq);
+	S("Ethernet Control", "%d", nethctrl);
+	S("Ingress Queues/w Free Lists/Interrupts", "%d", niqflint);
+	S("Ingress Queues", "%d", niq);
+	S("Traffic Class", "%d", tc);
+	S("Port Access Rights Mask", "%#x", pmask);
+	S("MAC Address Filters", "%d", nexactf);
+	S("Firmware Command Read Capabilities", "%#x", r_caps);
+	S("Firmware Command Write/Execute Capabilities", "%#x", wx_caps);
+
+	#undef S
+
+	return 0;
+}
+DEFINE_SHOW_ATTRIBUTE(resources);
+
 /**
  * ethqset2pinfo - return port_info of an Ethernet Queue Set
  * @adap: the adapter
@@ -2424,20 +2611,57 @@ static inline struct port_info *ethqset2pinfo(struct adapter *adap, int qset)
 	}
 
 	/* should never happen! */
-	BUG_ON(1);
+	BUG();
 	return NULL;
 }
 
+static int sge_qinfo_uld_txq_entries(const struct adapter *adap, int uld)
+{
+	const struct sge_uld_txq_info *utxq_info = adap->sge.uld_txq_info[uld];
+
+	if (!utxq_info)
+		return 0;
+
+	return DIV_ROUND_UP(utxq_info->ntxq, 4);
+}
+
+static int sge_qinfo_uld_rspq_entries(const struct adapter *adap, int uld,
+				      bool ciq)
+{
+	const struct sge_uld_rxq_info *urxq_info = adap->sge.uld_rxq_info[uld];
+
+	if (!urxq_info)
+		return 0;
+
+	return ciq ? DIV_ROUND_UP(urxq_info->nciq, 4) :
+		     DIV_ROUND_UP(urxq_info->nrxq, 4);
+}
+
+static int sge_qinfo_uld_rxq_entries(const struct adapter *adap, int uld)
+{
+	return sge_qinfo_uld_rspq_entries(adap, uld, false);
+}
+
+static int sge_qinfo_uld_ciq_entries(const struct adapter *adap, int uld)
+{
+	return sge_qinfo_uld_rspq_entries(adap, uld, true);
+}
+
 static int sge_qinfo_show(struct seq_file *seq, void *v)
 {
+	int eth_entries, ctrl_entries, eohw_entries = 0, eosw_entries = 0;
+	int uld_rxq_entries[CXGB4_ULD_MAX] = { 0 };
+	int uld_ciq_entries[CXGB4_ULD_MAX] = { 0 };
+	int uld_txq_entries[CXGB4_TX_MAX] = { 0 };
+	const struct sge_uld_txq_info *utxq_info;
+	const struct sge_uld_rxq_info *urxq_info;
+	struct cxgb4_tc_port_mqprio *port_mqprio;
 	struct adapter *adap = seq->private;
-	int eth_entries = DIV_ROUND_UP(adap->sge.ethqsets, 4);
-	int ofld_entries = DIV_ROUND_UP(adap->sge.ofldqsets, 4);
-	int ctrl_entries = DIV_ROUND_UP(MAX_CTRL_QUEUES, 4);
-	int i, r = (uintptr_t)v - 1;
-	int ofld_idx = r - eth_entries;
-	int ctrl_idx =  ofld_idx - ofld_entries;
-	int fq_idx =  ctrl_idx - ctrl_entries;
+	int i, j, n, r = (uintptr_t)v - 1;
+	struct sge *s = &adap->sge;
+
+	eth_entries = DIV_ROUND_UP(adap->sge.ethqsets, 4);
+	ctrl_entries = DIV_ROUND_UP(MAX_CTRL_QUEUES, 4);
 
 	if (r)
 		seq_putc(seq, '\n');
@@ -2447,7 +2671,7 @@ do { \
 	seq_printf(seq, "%-12s", s); \
 	for (i = 0; i < n; ++i) \
 		seq_printf(seq, " %16" fmt_spec, v); \
-		seq_putc(seq, '\n'); \
+	seq_putc(seq, '\n'); \
 } while (0)
 #define S(s, v) S3("s", s, v)
 #define T3(fmt_spec, s, v) S3(fmt_spec, s, tx[i].v)
@@ -2459,9 +2683,10 @@ do { \
 
 	if (r < eth_entries) {
 		int base_qset = r * 4;
-		const struct sge_eth_rxq *rx = &adap->sge.ethrxq[base_qset];
-		const struct sge_eth_txq *tx = &adap->sge.ethtxq[base_qset];
-		int n = min(4, adap->sge.ethqsets - 4 * r);
+		const struct sge_eth_rxq *rx = &s->ethrxq[base_qset];
+		const struct sge_eth_txq *tx = &s->ethtxq[base_qset];
+
+		n = min(4, s->ethqsets - 4 * r);
 
 		S("QType:", "Ethernet");
 		S("Interface:",
@@ -2486,8 +2711,66 @@ do { \
 		R("RspQ CIDX:", rspq.cidx);
 		R("RspQ Gen:", rspq.gen);
 		S3("u", "Intr delay:", qtimer_val(adap, &rx[i].rspq));
-		S3("u", "Intr pktcnt:",
-		   adap->sge.counter_val[rx[i].rspq.pktcnt_idx]);
+		S3("u", "Intr pktcnt:", s->counter_val[rx[i].rspq.pktcnt_idx]);
+		R("FL ID:", fl.cntxt_id);
+		R("FL size:", fl.size - 8);
+		R("FL pend:", fl.pend_cred);
+		R("FL avail:", fl.avail);
+		R("FL PIDX:", fl.pidx);
+		R("FL CIDX:", fl.cidx);
+		RL("RxPackets:", stats.pkts);
+		RL("RxCSO:", stats.rx_cso);
+		RL("VLANxtract:", stats.vlan_ex);
+		RL("LROmerged:", stats.lro_merged);
+		RL("LROpackets:", stats.lro_pkts);
+		RL("RxDrops:", stats.rx_drops);
+		RL("RxBadPkts:", stats.bad_rx_pkts);
+		TL("TSO:", tso);
+		TL("USO:", uso);
+		TL("TxCSO:", tx_cso);
+		TL("VLANins:", vlan_ins);
+		TL("TxQFull:", q.stops);
+		TL("TxQRestarts:", q.restarts);
+		TL("TxMapErr:", mapping_err);
+		RL("FLAllocErr:", fl.alloc_failed);
+		RL("FLLrgAlcErr:", fl.large_alloc_failed);
+		RL("FLMapErr:", fl.mapping_err);
+		RL("FLLow:", fl.low);
+		RL("FLStarving:", fl.starving);
+
+		goto out;
+	}
+
+	r -= eth_entries;
+	for_each_port(adap, j) {
+		struct port_info *pi = adap2pinfo(adap, j);
+		const struct sge_eth_rxq *rx;
+
+		mutex_lock(&pi->vi_mirror_mutex);
+		if (!pi->vi_mirror_count) {
+			mutex_unlock(&pi->vi_mirror_mutex);
+			continue;
+		}
+
+		if (r >= DIV_ROUND_UP(pi->nmirrorqsets, 4)) {
+			r -= DIV_ROUND_UP(pi->nmirrorqsets, 4);
+			mutex_unlock(&pi->vi_mirror_mutex);
+			continue;
+		}
+
+		rx = &s->mirror_rxq[j][r * 4];
+		n = min(4, pi->nmirrorqsets - 4 * r);
+
+		S("QType:", "Mirror-Rxq");
+		S("Interface:",
+		  rx[i].rspq.netdev ? rx[i].rspq.netdev->name : "N/A");
+		R("RspQ ID:", rspq.abs_id);
+		R("RspQ size:", rspq.size);
+		R("RspQE size:", rspq.iqe_len);
+		R("RspQ CIDX:", rspq.cidx);
+		R("RspQ Gen:", rspq.gen);
+		S3("u", "Intr delay:", qtimer_val(adap, &rx[i].rspq));
+		S3("u", "Intr pktcnt:", s->counter_val[rx[i].rspq.pktcnt_idx]);
 		R("FL ID:", fl.cntxt_id);
 		R("FL size:", fl.size - 8);
 		R("FL pend:", fl.pend_cred);
@@ -2500,7 +2783,61 @@ do { \
 		RL("LROmerged:", stats.lro_merged);
 		RL("LROpackets:", stats.lro_pkts);
 		RL("RxDrops:", stats.rx_drops);
+		RL("RxBadPkts:", stats.bad_rx_pkts);
+		RL("FLAllocErr:", fl.alloc_failed);
+		RL("FLLrgAlcErr:", fl.large_alloc_failed);
+		RL("FLMapErr:", fl.mapping_err);
+		RL("FLLow:", fl.low);
+		RL("FLStarving:", fl.starving);
+
+		mutex_unlock(&pi->vi_mirror_mutex);
+		goto out;
+	}
+
+	if (!adap->tc_mqprio)
+		goto skip_mqprio;
+
+	mutex_lock(&adap->tc_mqprio->mqprio_mutex);
+	if (!refcount_read(&adap->tc_mqprio->refcnt)) {
+		mutex_unlock(&adap->tc_mqprio->mqprio_mutex);
+		goto skip_mqprio;
+	}
+
+	eohw_entries = DIV_ROUND_UP(adap->sge.eoqsets, 4);
+	if (r < eohw_entries) {
+		int base_qset = r * 4;
+		const struct sge_ofld_rxq *rx = &s->eohw_rxq[base_qset];
+		const struct sge_eohw_txq *tx = &s->eohw_txq[base_qset];
+
+		n = min(4, s->eoqsets - 4 * r);
+
+		S("QType:", "ETHOFLD");
+		S("Interface:",
+		  rx[i].rspq.netdev ? rx[i].rspq.netdev->name : "N/A");
+		T("TxQ ID:", q.cntxt_id);
+		T("TxQ size:", q.size);
+		T("TxQ inuse:", q.in_use);
+		T("TxQ CIDX:", q.cidx);
+		T("TxQ PIDX:", q.pidx);
+		R("RspQ ID:", rspq.abs_id);
+		R("RspQ size:", rspq.size);
+		R("RspQE size:", rspq.iqe_len);
+		R("RspQ CIDX:", rspq.cidx);
+		R("RspQ Gen:", rspq.gen);
+		S3("u", "Intr delay:", qtimer_val(adap, &rx[i].rspq));
+		S3("u", "Intr pktcnt:", s->counter_val[rx[i].rspq.pktcnt_idx]);
+		R("FL ID:", fl.cntxt_id);
+		S3("u", "FL size:", rx->fl.size ? rx->fl.size - 8 : 0);
+		R("FL pend:", fl.pend_cred);
+		R("FL avail:", fl.avail);
+		R("FL PIDX:", fl.pidx);
+		R("FL CIDX:", fl.cidx);
+		RL("RxPackets:", stats.pkts);
+		RL("RxImm:", stats.imm);
+		RL("RxAN", stats.an);
+		RL("RxNoMem", stats.nomem);
 		TL("TSO:", tso);
+		TL("USO:", uso);
 		TL("TxCSO:", tx_cso);
 		TL("VLANins:", vlan_ins);
 		TL("TxQFull:", q.stops);
@@ -2512,9 +2849,260 @@ do { \
 		RL("FLLow:", fl.low);
 		RL("FLStarving:", fl.starving);
 
-	} else if (ctrl_idx < ctrl_entries) {
-		const struct sge_ctrl_txq *tx = &adap->sge.ctrlq[ctrl_idx * 4];
-		int n = min(4, adap->params.nports - 4 * ctrl_idx);
+		mutex_unlock(&adap->tc_mqprio->mqprio_mutex);
+		goto out;
+	}
+
+	r -= eohw_entries;
+	for (j = 0; j < adap->params.nports; j++) {
+		int entries;
+		u8 tc;
+
+		port_mqprio = &adap->tc_mqprio->port_mqprio[j];
+		entries = 0;
+		for (tc = 0; tc < port_mqprio->mqprio.qopt.num_tc; tc++)
+			entries += port_mqprio->mqprio.qopt.count[tc];
+
+		if (!entries)
+			continue;
+
+		eosw_entries = DIV_ROUND_UP(entries, 4);
+		if (r < eosw_entries) {
+			const struct sge_eosw_txq *tx;
+
+			n = min(4, entries - 4 * r);
+			tx = &port_mqprio->eosw_txq[4 * r];
+
+			S("QType:", "EOSW-TXQ");
+			S("Interface:",
+			  adap->port[j] ? adap->port[j]->name : "N/A");
+			T("EOTID:", hwtid);
+			T("HWQID:", hwqid);
+			T("State:", state);
+			T("Size:", ndesc);
+			T("In-Use:", inuse);
+			T("Credits:", cred);
+			T("Compl:", ncompl);
+			T("Last-Compl:", last_compl);
+			T("PIDX:", pidx);
+			T("Last-PIDX:", last_pidx);
+			T("CIDX:", cidx);
+			T("Last-CIDX:", last_cidx);
+			T("FLOWC-IDX:", flowc_idx);
+
+			mutex_unlock(&adap->tc_mqprio->mqprio_mutex);
+			goto out;
+		}
+
+		r -= eosw_entries;
+	}
+	mutex_unlock(&adap->tc_mqprio->mqprio_mutex);
+
+skip_mqprio:
+	if (!is_uld(adap))
+		goto skip_uld;
+
+	mutex_lock(&uld_mutex);
+	if (s->uld_txq_info)
+		for (i = 0; i < ARRAY_SIZE(uld_txq_entries); i++)
+			uld_txq_entries[i] = sge_qinfo_uld_txq_entries(adap, i);
+
+	if (s->uld_rxq_info) {
+		for (i = 0; i < ARRAY_SIZE(uld_rxq_entries); i++) {
+			uld_rxq_entries[i] = sge_qinfo_uld_rxq_entries(adap, i);
+			uld_ciq_entries[i] = sge_qinfo_uld_ciq_entries(adap, i);
+		}
+	}
+
+	if (r < uld_txq_entries[CXGB4_TX_OFLD]) {
+		const struct sge_uld_txq *tx;
+
+		utxq_info = s->uld_txq_info[CXGB4_TX_OFLD];
+		tx = &utxq_info->uldtxq[r * 4];
+		n = min(4, utxq_info->ntxq - 4 * r);
+
+		S("QType:", "OFLD-TXQ");
+		T("TxQ ID:", q.cntxt_id);
+		T("TxQ size:", q.size);
+		T("TxQ inuse:", q.in_use);
+		T("TxQ CIDX:", q.cidx);
+		T("TxQ PIDX:", q.pidx);
+
+		goto unlock;
+	}
+
+	r -= uld_txq_entries[CXGB4_TX_OFLD];
+	if (r < uld_rxq_entries[CXGB4_ULD_RDMA]) {
+		const struct sge_ofld_rxq *rx;
+
+		urxq_info = s->uld_rxq_info[CXGB4_ULD_RDMA];
+		rx = &urxq_info->uldrxq[r * 4];
+		n = min(4, urxq_info->nrxq - 4 * r);
+
+		S("QType:", "RDMA-CPL");
+		S("Interface:",
+		  rx[i].rspq.netdev ? rx[i].rspq.netdev->name : "N/A");
+		R("RspQ ID:", rspq.abs_id);
+		R("RspQ size:", rspq.size);
+		R("RspQE size:", rspq.iqe_len);
+		R("RspQ CIDX:", rspq.cidx);
+		R("RspQ Gen:", rspq.gen);
+		S3("u", "Intr delay:", qtimer_val(adap, &rx[i].rspq));
+		S3("u", "Intr pktcnt:",	s->counter_val[rx[i].rspq.pktcnt_idx]);
+		R("FL ID:", fl.cntxt_id);
+		R("FL size:", fl.size - 8);
+		R("FL pend:", fl.pend_cred);
+		R("FL avail:", fl.avail);
+		R("FL PIDX:", fl.pidx);
+		R("FL CIDX:", fl.cidx);
+
+		goto unlock;
+	}
+
+	r -= uld_rxq_entries[CXGB4_ULD_RDMA];
+	if (r < uld_ciq_entries[CXGB4_ULD_RDMA]) {
+		const struct sge_ofld_rxq *rx;
+		int ciq_idx = 0;
+
+		urxq_info = s->uld_rxq_info[CXGB4_ULD_RDMA];
+		ciq_idx = urxq_info->nrxq + (r * 4);
+		rx = &urxq_info->uldrxq[ciq_idx];
+		n = min(4, urxq_info->nciq - 4 * r);
+
+		S("QType:", "RDMA-CIQ");
+		S("Interface:",
+		  rx[i].rspq.netdev ? rx[i].rspq.netdev->name : "N/A");
+		R("RspQ ID:", rspq.abs_id);
+		R("RspQ size:", rspq.size);
+		R("RspQE size:", rspq.iqe_len);
+		R("RspQ CIDX:", rspq.cidx);
+		R("RspQ Gen:", rspq.gen);
+		S3("u", "Intr delay:", qtimer_val(adap, &rx[i].rspq));
+		S3("u", "Intr pktcnt:",	s->counter_val[rx[i].rspq.pktcnt_idx]);
+
+		goto unlock;
+	}
+
+	r -= uld_ciq_entries[CXGB4_ULD_RDMA];
+	if (r < uld_rxq_entries[CXGB4_ULD_ISCSI]) {
+		const struct sge_ofld_rxq *rx;
+
+		urxq_info = s->uld_rxq_info[CXGB4_ULD_ISCSI];
+		rx = &urxq_info->uldrxq[r * 4];
+		n = min(4, urxq_info->nrxq - 4 * r);
+
+		S("QType:", "iSCSI");
+		R("RspQ ID:", rspq.abs_id);
+		R("RspQ size:", rspq.size);
+		R("RspQE size:", rspq.iqe_len);
+		R("RspQ CIDX:", rspq.cidx);
+		R("RspQ Gen:", rspq.gen);
+		S3("u", "Intr delay:", qtimer_val(adap, &rx[i].rspq));
+		S3("u", "Intr pktcnt:",	s->counter_val[rx[i].rspq.pktcnt_idx]);
+		R("FL ID:", fl.cntxt_id);
+		R("FL size:", fl.size - 8);
+		R("FL pend:", fl.pend_cred);
+		R("FL avail:", fl.avail);
+		R("FL PIDX:", fl.pidx);
+		R("FL CIDX:", fl.cidx);
+
+		goto unlock;
+	}
+
+	r -= uld_rxq_entries[CXGB4_ULD_ISCSI];
+	if (r < uld_rxq_entries[CXGB4_ULD_ISCSIT]) {
+		const struct sge_ofld_rxq *rx;
+
+		urxq_info = s->uld_rxq_info[CXGB4_ULD_ISCSIT];
+		rx = &urxq_info->uldrxq[r * 4];
+		n = min(4, urxq_info->nrxq - 4 * r);
+
+		S("QType:", "iSCSIT");
+		R("RspQ ID:", rspq.abs_id);
+		R("RspQ size:", rspq.size);
+		R("RspQE size:", rspq.iqe_len);
+		R("RspQ CIDX:", rspq.cidx);
+		R("RspQ Gen:", rspq.gen);
+		S3("u", "Intr delay:", qtimer_val(adap, &rx[i].rspq));
+		S3("u", "Intr pktcnt:",	s->counter_val[rx[i].rspq.pktcnt_idx]);
+		R("FL ID:", fl.cntxt_id);
+		R("FL size:", fl.size - 8);
+		R("FL pend:", fl.pend_cred);
+		R("FL avail:", fl.avail);
+		R("FL PIDX:", fl.pidx);
+		R("FL CIDX:", fl.cidx);
+
+		goto unlock;
+	}
+
+	r -= uld_rxq_entries[CXGB4_ULD_ISCSIT];
+	if (r < uld_rxq_entries[CXGB4_ULD_TLS]) {
+		const struct sge_ofld_rxq *rx;
+
+		urxq_info = s->uld_rxq_info[CXGB4_ULD_TLS];
+		rx = &urxq_info->uldrxq[r * 4];
+		n = min(4, urxq_info->nrxq - 4 * r);
+
+		S("QType:", "TLS");
+		R("RspQ ID:", rspq.abs_id);
+		R("RspQ size:", rspq.size);
+		R("RspQE size:", rspq.iqe_len);
+		R("RspQ CIDX:", rspq.cidx);
+		R("RspQ Gen:", rspq.gen);
+		S3("u", "Intr delay:", qtimer_val(adap, &rx[i].rspq));
+		S3("u", "Intr pktcnt:",	s->counter_val[rx[i].rspq.pktcnt_idx]);
+		R("FL ID:", fl.cntxt_id);
+		R("FL size:", fl.size - 8);
+		R("FL pend:", fl.pend_cred);
+		R("FL avail:", fl.avail);
+		R("FL PIDX:", fl.pidx);
+		R("FL CIDX:", fl.cidx);
+
+		goto unlock;
+	}
+
+	r -= uld_rxq_entries[CXGB4_ULD_TLS];
+	if (r < uld_txq_entries[CXGB4_TX_CRYPTO]) {
+		const struct sge_ofld_rxq *rx;
+		const struct sge_uld_txq *tx;
+
+		utxq_info = s->uld_txq_info[CXGB4_TX_CRYPTO];
+		urxq_info = s->uld_rxq_info[CXGB4_ULD_CRYPTO];
+		tx = &utxq_info->uldtxq[r * 4];
+		rx = &urxq_info->uldrxq[r * 4];
+		n = min(4, utxq_info->ntxq - 4 * r);
+
+		S("QType:", "Crypto");
+		T("TxQ ID:", q.cntxt_id);
+		T("TxQ size:", q.size);
+		T("TxQ inuse:", q.in_use);
+		T("TxQ CIDX:", q.cidx);
+		T("TxQ PIDX:", q.pidx);
+		R("RspQ ID:", rspq.abs_id);
+		R("RspQ size:", rspq.size);
+		R("RspQE size:", rspq.iqe_len);
+		R("RspQ CIDX:", rspq.cidx);
+		R("RspQ Gen:", rspq.gen);
+		S3("u", "Intr delay:", qtimer_val(adap, &rx[i].rspq));
+		S3("u", "Intr pktcnt:",	s->counter_val[rx[i].rspq.pktcnt_idx]);
+		R("FL ID:", fl.cntxt_id);
+		R("FL size:", fl.size - 8);
+		R("FL pend:", fl.pend_cred);
+		R("FL avail:", fl.avail);
+		R("FL PIDX:", fl.pidx);
+		R("FL CIDX:", fl.cidx);
+
+		goto unlock;
+	}
+
+	r -= uld_txq_entries[CXGB4_TX_CRYPTO];
+	mutex_unlock(&uld_mutex);
+
+skip_uld:
+	if (r < ctrl_entries) {
+		const struct sge_ctrl_txq *tx = &s->ctrlq[r * 4];
+
+		n = min(4, adap->params.nports - 4 * r);
 
 		S("QType:", "Control");
 		T("TxQ ID:", q.cntxt_id);
@@ -2524,8 +3112,13 @@ do { \
 		T("TxQ PIDX:", q.pidx);
 		TL("TxQFull:", q.stops);
 		TL("TxQRestarts:", q.restarts);
-	} else if (fq_idx == 0) {
-		const struct sge_rspq *evtq = &adap->sge.fw_evtq;
+
+		goto out;
+	}
+
+	r -= ctrl_entries;
+	if (r < 1) {
+		const struct sge_rspq *evtq = &s->fw_evtq;
 
 		seq_printf(seq, "%-12s %16s\n", "QType:", "FW event queue");
 		seq_printf(seq, "%-12s %16u\n", "RspQ ID:", evtq->abs_id);
@@ -2536,8 +3129,11 @@ do { \
 		seq_printf(seq, "%-12s %16u\n", "Intr delay:",
 			   qtimer_val(adap, evtq));
 		seq_printf(seq, "%-12s %16u\n", "Intr pktcnt:",
-			   adap->sge.counter_val[evtq->pktcnt_idx]);
+			   s->counter_val[evtq->pktcnt_idx]);
+
+		goto out;
 	}
+
 #undef R
 #undef RL
 #undef T
@@ -2546,13 +3142,65 @@ do { \
 #undef R3
 #undef T3
 #undef S3
+out:
+	return 0;
+
+unlock:
+	mutex_unlock(&uld_mutex);
 	return 0;
 }
 
-static int sge_queue_entries(const struct adapter *adap)
+static int sge_queue_entries(struct adapter *adap)
 {
-	return DIV_ROUND_UP(adap->sge.ethqsets, 4) +
-	       DIV_ROUND_UP(adap->sge.ofldqsets, 4) +
+	int i, tot_uld_entries = 0, eohw_entries = 0, eosw_entries = 0;
+	int mirror_rxq_entries = 0;
+
+	if (adap->tc_mqprio) {
+		struct cxgb4_tc_port_mqprio *port_mqprio;
+		u8 tc;
+
+		mutex_lock(&adap->tc_mqprio->mqprio_mutex);
+		if (adap->sge.eohw_txq)
+			eohw_entries = DIV_ROUND_UP(adap->sge.eoqsets, 4);
+
+		for (i = 0; i < adap->params.nports; i++) {
+			u32 entries = 0;
+
+			port_mqprio = &adap->tc_mqprio->port_mqprio[i];
+			for (tc = 0; tc < port_mqprio->mqprio.qopt.num_tc; tc++)
+				entries += port_mqprio->mqprio.qopt.count[tc];
+
+			if (entries)
+				eosw_entries += DIV_ROUND_UP(entries, 4);
+		}
+		mutex_unlock(&adap->tc_mqprio->mqprio_mutex);
+	}
+
+	for_each_port(adap, i) {
+		struct port_info *pi = adap2pinfo(adap, i);
+
+		mutex_lock(&pi->vi_mirror_mutex);
+		if (pi->vi_mirror_count)
+			mirror_rxq_entries += DIV_ROUND_UP(pi->nmirrorqsets, 4);
+		mutex_unlock(&pi->vi_mirror_mutex);
+	}
+
+	if (!is_uld(adap))
+		goto lld_only;
+
+	mutex_lock(&uld_mutex);
+	for (i = 0; i < CXGB4_TX_MAX; i++)
+		tot_uld_entries += sge_qinfo_uld_txq_entries(adap, i);
+
+	for (i = 0; i < CXGB4_ULD_MAX; i++) {
+		tot_uld_entries += sge_qinfo_uld_rxq_entries(adap, i);
+		tot_uld_entries += sge_qinfo_uld_ciq_entries(adap, i);
+	}
+	mutex_unlock(&uld_mutex);
+
+lld_only:
+	return DIV_ROUND_UP(adap->sge.ethqsets, 4) + mirror_rxq_entries +
+	       eohw_entries + eosw_entries + tot_uld_entries +
 	       DIV_ROUND_UP(MAX_CTRL_QUEUES, 4) + 1;
 }
 
@@ -2609,7 +3257,7 @@ int mem_open(struct inode *inode, struct file *file)
 
 	file->private_data = inode->i_private;
 
-	mem = (uintptr_t)file->private_data & 0x3;
+	mem = (uintptr_t)file->private_data & 0x7;
 	adap = file->private_data - mem;
 
 	(void)t4_fwcache(adap, FW_PARAM_DEV_FWCACHE_FLUSH);
@@ -2622,7 +3270,7 @@ static ssize_t mem_read(struct file *file, char __user *buf, size_t count,
 {
 	loff_t pos = *ppos;
 	loff_t avail = file_inode(file)->i_size;
-	unsigned int mem = (uintptr_t)file->private_data & 3;
+	unsigned int mem = (uintptr_t)file->private_data & 0x7;
 	struct adapter *adap = file->private_data - mem;
 	__be32 *data;
 	int ret;
@@ -2664,11 +3312,15 @@ static const struct file_operations mem_debugfs_fops = {
 static int tid_info_show(struct seq_file *seq, void *v)
 {
 	struct adapter *adap = seq->private;
-	const struct tid_info *t = &adap->tids;
-	enum chip_type chip = CHELSIO_CHIP_VERSION(adap->params.chip);
+	const struct tid_info *t;
+	enum chip_type chip;
 
+	t = &adap->tids;
+	chip = CHELSIO_CHIP_VERSION(adap->params.chip);
 	if (t4_read_reg(adap, LE_DB_CONFIG_A) & HASHEN_F) {
 		unsigned int sb;
+		seq_printf(seq, "Connections in use: %u\n",
+			   atomic_read(&t->conns_in_use));
 
 		if (chip <= CHELSIO_T5)
 			sb = t4_read_reg(adap, LE_DB_SERVER_INDEX_A) / 4;
@@ -2676,40 +3328,47 @@ static int tid_info_show(struct seq_file *seq, void *v)
 			sb = t4_read_reg(adap, LE_DB_SRVR_START_INDEX_A);
 
 		if (sb) {
-			seq_printf(seq, "TID range: 0..%u/%u..%u", sb - 1,
-				   adap->tids.hash_base,
-				   t->ntids - 1);
+			seq_printf(seq, "TID range: %u..%u/%u..%u", t->tid_base,
+				   sb - 1, adap->tids.hash_base,
+				   t->tid_base + t->ntids - 1);
 			seq_printf(seq, ", in use: %u/%u\n",
 				   atomic_read(&t->tids_in_use),
 				   atomic_read(&t->hash_tids_in_use));
-		} else if (adap->flags & FW_OFLD_CONN) {
+		} else if (adap->flags & CXGB4_FW_OFLD_CONN) {
 			seq_printf(seq, "TID range: %u..%u/%u..%u",
 				   t->aftid_base,
 				   t->aftid_end,
 				   adap->tids.hash_base,
-				   t->ntids - 1);
+				   t->tid_base + t->ntids - 1);
 			seq_printf(seq, ", in use: %u/%u\n",
 				   atomic_read(&t->tids_in_use),
 				   atomic_read(&t->hash_tids_in_use));
 		} else {
 			seq_printf(seq, "TID range: %u..%u",
 				   adap->tids.hash_base,
-				   t->ntids - 1);
+				   t->tid_base + t->ntids - 1);
 			seq_printf(seq, ", in use: %u\n",
 				   atomic_read(&t->hash_tids_in_use));
 		}
 	} else if (t->ntids) {
-		seq_printf(seq, "TID range: 0..%u", t->ntids - 1);
+		seq_printf(seq, "Connections in use: %u\n",
+			   atomic_read(&t->conns_in_use));
+
+		seq_printf(seq, "TID range: %u..%u", t->tid_base,
+			   t->tid_base + t->ntids - 1);
 		seq_printf(seq, ", in use: %u\n",
 			   atomic_read(&t->tids_in_use));
 	}
 
 	if (t->nstids)
-		seq_printf(seq, "STID range: %u..%u, in use: %u\n",
+		seq_printf(seq, "STID range: %u..%u, in use-IPv4/IPv6: %u/%u\n",
 			   (!t->stid_base &&
 			   (chip <= CHELSIO_T5)) ?
 			   t->stid_base + 1 : t->stid_base,
-			   t->stid_base + t->nstids - 1, t->stids_in_use);
+			   t->stid_base + t->nstids - 1,
+			   t->stids_in_use - t->v6_stids_in_use,
+			   t->v6_stids_in_use);
+
 	if (t->natids)
 		seq_printf(seq, "ATID range: 0..%u, in use: %u\n",
 			   t->natids - 1, t->atids_in_use);
@@ -2719,19 +3378,25 @@ static int tid_info_show(struct seq_file *seq, void *v)
 		seq_printf(seq, "SFTID range: %u..%u in use: %u\n",
 			   t->sftid_base, t->sftid_base + t->nsftids - 2,
 			   t->sftids_in_use);
+	if (t->nhpftids)
+		seq_printf(seq, "HPFTID range: %u..%u\n", t->hpftid_base,
+			   t->hpftid_base + t->nhpftids - 1);
+	if (t->neotids)
+		seq_printf(seq, "EOTID range: %u..%u, in use: %u\n",
+			   t->eotid_base, t->eotid_base + t->neotids - 1,
+			   atomic_read(&t->eotids_in_use));
 	if (t->ntids)
 		seq_printf(seq, "HW TID usage: %u IP users, %u IPv6 users\n",
 			   t4_read_reg(adap, LE_DB_ACT_CNT_IPV4_A),
 			   t4_read_reg(adap, LE_DB_ACT_CNT_IPV6_A));
 	return 0;
 }
-
-DEFINE_SIMPLE_DEBUGFS_FILE(tid_info);
+DEFINE_SHOW_ATTRIBUTE(tid_info);
 
 static void add_debugfs_mem(struct adapter *adap, const char *name,
 			    unsigned int idx, unsigned int size_mb)
 {
-	debugfs_create_file_size(name, S_IRUSR, adap->debugfs_root,
+	debugfs_create_file_size(name, 0400, adap->debugfs_root,
 				 (void *)adap + idx, &mem_debugfs_fops,
 				 size_mb << 20);
 }
@@ -2753,7 +3418,7 @@ static ssize_t blocked_fl_read(struct file *filp, char __user *ubuf,
 		       adap->sge.egr_sz, adap->sge.blocked_fl);
 	len += sprintf(buf + len, "\n");
 	size = simple_read_from_buffer(ubuf, count, ppos, buf, len);
-	kvfree(buf);
+	kfree(buf);
 	return size;
 }
 
@@ -2764,16 +3429,18 @@ static ssize_t blocked_fl_write(struct file *filp, const char __user *ubuf,
 	unsigned long *t;
 	struct adapter *adap = filp->private_data;
 
-	t = kcalloc(BITS_TO_LONGS(adap->sge.egr_sz), sizeof(long), GFP_KERNEL);
+	t = bitmap_zalloc(adap->sge.egr_sz, GFP_KERNEL);
 	if (!t)
 		return -ENOMEM;
 
 	err = bitmap_parse_user(ubuf, count, t, adap->sge.egr_sz);
-	if (err)
+	if (err) {
+		bitmap_free(t);
 		return err;
+	}
 
 	bitmap_copy(adap->sge.blocked_fl, t, adap->sge.egr_sz);
-	kvfree(t);
+	bitmap_free(t);
 	return count;
 }
 
@@ -2785,18 +3452,6 @@ static const struct file_operations blocked_fl_fops = {
 	.llseek  = generic_file_llseek,
 };
 
-struct mem_desc {
-	unsigned int base;
-	unsigned int limit;
-	unsigned int idx;
-};
-
-static int mem_desc_cmp(const void *a, const void *b)
-{
-	return ((const struct mem_desc *)a)->base -
-	       ((const struct mem_desc *)b)->base;
-}
-
 static void mem_region_show(struct seq_file *seq, const char *name,
 			    unsigned int from, unsigned int to)
 {
@@ -2810,265 +3465,293 @@ static void mem_region_show(struct seq_file *seq, const char *name,
 static int meminfo_show(struct seq_file *seq, void *v)
 {
 	static const char * const memory[] = { "EDC0:", "EDC1:", "MC:",
-					"MC0:", "MC1:"};
-	static const char * const region[] = {
-		"DBQ contexts:", "IMSG contexts:", "FLM cache:", "TCBs:",
-		"Pstructs:", "Timers:", "Rx FL:", "Tx FL:", "Pstruct FL:",
-		"Tx payload:", "Rx payload:", "LE hash:", "iSCSI region:",
-		"TDDP region:", "TPT region:", "STAG region:", "RQ region:",
-		"RQUDP region:", "PBL region:", "TXPBL region:",
-		"DBVFIFO region:", "ULPRX state:", "ULPTX state:",
-		"On-chip queues:"
-	};
-
-	int i, n;
-	u32 lo, hi, used, alloc;
-	struct mem_desc avail[4];
-	struct mem_desc mem[ARRAY_SIZE(region) + 3];      /* up to 3 holes */
-	struct mem_desc *md = mem;
+					       "MC0:", "MC1:", "HMA:"};
 	struct adapter *adap = seq->private;
+	struct cudbg_meminfo meminfo;
+	int i, rc;
 
-	for (i = 0; i < ARRAY_SIZE(mem); i++) {
-		mem[i].limit = 0;
-		mem[i].idx = i;
-	}
+	memset(&meminfo, 0, sizeof(struct cudbg_meminfo));
+	rc = cudbg_fill_meminfo(adap, &meminfo);
+	if (rc)
+		return -ENXIO;
 
-	/* Find and sort the populated memory ranges */
-	i = 0;
-	lo = t4_read_reg(adap, MA_TARGET_MEM_ENABLE_A);
-	if (lo & EDRAM0_ENABLE_F) {
-		hi = t4_read_reg(adap, MA_EDRAM0_BAR_A);
-		avail[i].base = EDRAM0_BASE_G(hi) << 20;
-		avail[i].limit = avail[i].base + (EDRAM0_SIZE_G(hi) << 20);
-		avail[i].idx = 0;
-		i++;
+	for (i = 0; i < meminfo.avail_c; i++)
+		mem_region_show(seq, memory[meminfo.avail[i].idx],
+				meminfo.avail[i].base,
+				meminfo.avail[i].limit - 1);
+
+	seq_putc(seq, '\n');
+	for (i = 0; i < meminfo.mem_c; i++) {
+		if (meminfo.mem[i].idx >= ARRAY_SIZE(cudbg_region))
+			continue;                        /* skip holes */
+		if (!meminfo.mem[i].limit)
+			meminfo.mem[i].limit =
+				i < meminfo.mem_c - 1 ?
+				meminfo.mem[i + 1].base - 1 : ~0;
+		mem_region_show(seq, cudbg_region[meminfo.mem[i].idx],
+				meminfo.mem[i].base, meminfo.mem[i].limit);
 	}
-	if (lo & EDRAM1_ENABLE_F) {
-		hi = t4_read_reg(adap, MA_EDRAM1_BAR_A);
-		avail[i].base = EDRAM1_BASE_G(hi) << 20;
-		avail[i].limit = avail[i].base + (EDRAM1_SIZE_G(hi) << 20);
-		avail[i].idx = 1;
+
+	seq_putc(seq, '\n');
+	mem_region_show(seq, "uP RAM:", meminfo.up_ram_lo, meminfo.up_ram_hi);
+	mem_region_show(seq, "uP Extmem2:", meminfo.up_extmem2_lo,
+			meminfo.up_extmem2_hi);
+
+	seq_printf(seq, "\n%u Rx pages (%u free) of size %uKiB for %u channels\n",
+		   meminfo.rx_pages_data[0], meminfo.free_rx_cnt,
+		   meminfo.rx_pages_data[1], meminfo.rx_pages_data[2]);
+
+	seq_printf(seq, "%u Tx pages (%u free) of size %u%ciB for %u channels\n",
+		   meminfo.tx_pages_data[0], meminfo.free_tx_cnt,
+		   meminfo.tx_pages_data[1], meminfo.tx_pages_data[2],
+		   meminfo.tx_pages_data[3]);
+
+	seq_printf(seq, "%u p-structs (%u free)\n\n",
+		   meminfo.p_structs, meminfo.p_structs_free_cnt);
+
+	for (i = 0; i < 4; i++)
+		/* For T6 these are MAC buffer groups */
+		seq_printf(seq, "Port %d using %u pages out of %u allocated\n",
+			   i, meminfo.port_used[i], meminfo.port_alloc[i]);
+
+	for (i = 0; i < adap->params.arch.nchan; i++)
+		/* For T6 these are MAC buffer groups */
+		seq_printf(seq,
+			   "Loopback %d using %u pages out of %u allocated\n",
+			   i, meminfo.loopback_used[i],
+			   meminfo.loopback_alloc[i]);
+
+	return 0;
+}
+DEFINE_SHOW_ATTRIBUTE(meminfo);
+
+static int chcr_stats_show(struct seq_file *seq, void *v)
+{
+#if IS_ENABLED(CONFIG_CHELSIO_TLS_DEVICE)
+	struct ch_ktls_port_stats_debug *ktls_port;
+	int i = 0;
+#endif
+	struct adapter *adap = seq->private;
+
+	seq_puts(seq, "Chelsio Crypto Accelerator Stats \n");
+	seq_printf(seq, "Cipher Ops: %10u \n",
+		   atomic_read(&adap->chcr_stats.cipher_rqst));
+	seq_printf(seq, "Digest Ops: %10u \n",
+		   atomic_read(&adap->chcr_stats.digest_rqst));
+	seq_printf(seq, "Aead Ops: %10u \n",
+		   atomic_read(&adap->chcr_stats.aead_rqst));
+	seq_printf(seq, "Completion: %10u \n",
+		   atomic_read(&adap->chcr_stats.complete));
+	seq_printf(seq, "Error: %10u \n",
+		   atomic_read(&adap->chcr_stats.error));
+	seq_printf(seq, "Fallback: %10u \n",
+		   atomic_read(&adap->chcr_stats.fallback));
+	seq_printf(seq, "TLS PDU Tx: %10u\n",
+		   atomic_read(&adap->chcr_stats.tls_pdu_tx));
+	seq_printf(seq, "TLS PDU Rx: %10u\n",
+		   atomic_read(&adap->chcr_stats.tls_pdu_rx));
+	seq_printf(seq, "TLS Keys (DDR) Count: %10u\n",
+		   atomic_read(&adap->chcr_stats.tls_key));
+#if IS_ENABLED(CONFIG_CHELSIO_IPSEC_INLINE)
+	seq_puts(seq, "\nChelsio Inline IPsec Crypto Accelerator Stats\n");
+	seq_printf(seq, "IPSec PDU: %10u\n",
+		   atomic_read(&adap->ch_ipsec_stats.ipsec_cnt));
+#endif
+#if IS_ENABLED(CONFIG_CHELSIO_TLS_DEVICE)
+	seq_puts(seq, "\nChelsio KTLS Crypto Accelerator Stats\n");
+	seq_printf(seq, "Tx TLS offload refcount:          %20u\n",
+		   refcount_read(&adap->chcr_ktls.ktls_refcount));
+	seq_printf(seq, "Tx records send:                  %20llu\n",
+		   atomic64_read(&adap->ch_ktls_stats.ktls_tx_send_records));
+	seq_printf(seq, "Tx partial start of records:      %20llu\n",
+		   atomic64_read(&adap->ch_ktls_stats.ktls_tx_start_pkts));
+	seq_printf(seq, "Tx partial middle of records:     %20llu\n",
+		   atomic64_read(&adap->ch_ktls_stats.ktls_tx_middle_pkts));
+	seq_printf(seq, "Tx partial end of record:         %20llu\n",
+		   atomic64_read(&adap->ch_ktls_stats.ktls_tx_end_pkts));
+	seq_printf(seq, "Tx complete records:              %20llu\n",
+		   atomic64_read(&adap->ch_ktls_stats.ktls_tx_complete_pkts));
+	seq_printf(seq, "TX trim pkts :                    %20llu\n",
+		   atomic64_read(&adap->ch_ktls_stats.ktls_tx_trimmed_pkts));
+	seq_printf(seq, "TX sw fallback :                  %20llu\n",
+		   atomic64_read(&adap->ch_ktls_stats.ktls_tx_fallback));
+	while (i < MAX_NPORTS) {
+		ktls_port = &adap->ch_ktls_stats.ktls_port[i];
+		seq_printf(seq, "Port %d\n", i);
+		seq_printf(seq, "Tx connection created:            %20llu\n",
+			   atomic64_read(&ktls_port->ktls_tx_connection_open));
+		seq_printf(seq, "Tx connection failed:             %20llu\n",
+			   atomic64_read(&ktls_port->ktls_tx_connection_fail));
+		seq_printf(seq, "Tx connection closed:             %20llu\n",
+			   atomic64_read(&ktls_port->ktls_tx_connection_close));
 		i++;
 	}
+#endif
+	return 0;
+}
+DEFINE_SHOW_ATTRIBUTE(chcr_stats);
 
-	if (is_t5(adap->params.chip)) {
-		if (lo & EXT_MEM0_ENABLE_F) {
-			hi = t4_read_reg(adap, MA_EXT_MEMORY0_BAR_A);
-			avail[i].base = EXT_MEM0_BASE_G(hi) << 20;
-			avail[i].limit =
-				avail[i].base + (EXT_MEM0_SIZE_G(hi) << 20);
-			avail[i].idx = 3;
-			i++;
-		}
-		if (lo & EXT_MEM1_ENABLE_F) {
-			hi = t4_read_reg(adap, MA_EXT_MEMORY1_BAR_A);
-			avail[i].base = EXT_MEM1_BASE_G(hi) << 20;
-			avail[i].limit =
-				avail[i].base + (EXT_MEM1_SIZE_G(hi) << 20);
-			avail[i].idx = 4;
-			i++;
-		}
-	} else {
-		if (lo & EXT_MEM_ENABLE_F) {
-			hi = t4_read_reg(adap, MA_EXT_MEMORY_BAR_A);
-			avail[i].base = EXT_MEM_BASE_G(hi) << 20;
-			avail[i].limit =
-				avail[i].base + (EXT_MEM_SIZE_G(hi) << 20);
-			avail[i].idx = 2;
-			i++;
-		}
-	}
-	if (!i)                                    /* no memory available */
-		return 0;
-	sort(avail, i, sizeof(struct mem_desc), mem_desc_cmp, NULL);
-
-	(md++)->base = t4_read_reg(adap, SGE_DBQ_CTXT_BADDR_A);
-	(md++)->base = t4_read_reg(adap, SGE_IMSG_CTXT_BADDR_A);
-	(md++)->base = t4_read_reg(adap, SGE_FLM_CACHE_BADDR_A);
-	(md++)->base = t4_read_reg(adap, TP_CMM_TCB_BASE_A);
-	(md++)->base = t4_read_reg(adap, TP_CMM_MM_BASE_A);
-	(md++)->base = t4_read_reg(adap, TP_CMM_TIMER_BASE_A);
-	(md++)->base = t4_read_reg(adap, TP_CMM_MM_RX_FLST_BASE_A);
-	(md++)->base = t4_read_reg(adap, TP_CMM_MM_TX_FLST_BASE_A);
-	(md++)->base = t4_read_reg(adap, TP_CMM_MM_PS_FLST_BASE_A);
-
-	/* the next few have explicit upper bounds */
-	md->base = t4_read_reg(adap, TP_PMM_TX_BASE_A);
-	md->limit = md->base - 1 +
-		    t4_read_reg(adap, TP_PMM_TX_PAGE_SIZE_A) *
-		    PMTXMAXPAGE_G(t4_read_reg(adap, TP_PMM_TX_MAX_PAGE_A));
-	md++;
-
-	md->base = t4_read_reg(adap, TP_PMM_RX_BASE_A);
-	md->limit = md->base - 1 +
-		    t4_read_reg(adap, TP_PMM_RX_PAGE_SIZE_A) *
-		    PMRXMAXPAGE_G(t4_read_reg(adap, TP_PMM_RX_MAX_PAGE_A));
-	md++;
+#define PRINT_ADAP_STATS(string, value) \
+	seq_printf(seq, "%-25s %-20llu\n", (string), \
+		   (unsigned long long)(value))
 
-	if (t4_read_reg(adap, LE_DB_CONFIG_A) & HASHEN_F) {
-		if (CHELSIO_CHIP_VERSION(adap->params.chip) <= CHELSIO_T5) {
-			hi = t4_read_reg(adap, LE_DB_TID_HASHBASE_A) / 4;
-			md->base = t4_read_reg(adap, LE_DB_HASH_TID_BASE_A);
-		 } else {
-			hi = t4_read_reg(adap, LE_DB_HASH_TID_BASE_A);
-			md->base = t4_read_reg(adap,
-					       LE_DB_HASH_TBL_BASE_ADDR_A);
-		}
-		md->limit = 0;
-	} else {
-		md->base = 0;
-		md->idx = ARRAY_SIZE(region);  /* hide it */
-	}
-	md++;
+#define PRINT_CH_STATS(string, value) \
+do { \
+	seq_printf(seq, "%-25s ", (string)); \
+	for (i = 0; i < adap->params.arch.nchan; i++) \
+		seq_printf(seq, "%-20llu ", \
+			   (unsigned long long)stats.value[i]); \
+	seq_printf(seq, "\n"); \
+} while (0)
 
-#define ulp_region(reg) do { \
-	md->base = t4_read_reg(adap, ULP_ ## reg ## _LLIMIT_A);\
-	(md++)->limit = t4_read_reg(adap, ULP_ ## reg ## _ULIMIT_A); \
+#define PRINT_CH_STATS2(string, value) \
+do { \
+	seq_printf(seq, "%-25s ", (string)); \
+	for (i = 0; i < adap->params.arch.nchan; i++) \
+		seq_printf(seq, "%-20llu ", \
+			   (unsigned long long)stats[i].value); \
+	seq_printf(seq, "\n"); \
 } while (0)
 
-	ulp_region(RX_ISCSI);
-	ulp_region(RX_TDDP);
-	ulp_region(TX_TPT);
-	ulp_region(RX_STAG);
-	ulp_region(RX_RQ);
-	ulp_region(RX_RQUDP);
-	ulp_region(RX_PBL);
-	ulp_region(TX_PBL);
-#undef ulp_region
-	md->base = 0;
-	md->idx = ARRAY_SIZE(region);
-	if (!is_t4(adap->params.chip)) {
-		u32 size = 0;
-		u32 sge_ctrl = t4_read_reg(adap, SGE_CONTROL2_A);
-		u32 fifo_size = t4_read_reg(adap, SGE_DBVFIFO_SIZE_A);
-
-		if (is_t5(adap->params.chip)) {
-			if (sge_ctrl & VFIFO_ENABLE_F)
-				size = DBVFIFO_SIZE_G(fifo_size);
-		} else {
-			size = T6_DBVFIFO_SIZE_G(fifo_size);
-		}
+static void show_tcp_stats(struct seq_file *seq)
+{
+	struct adapter *adap = seq->private;
+	struct tp_tcp_stats v4, v6;
 
-		if (size) {
-			md->base = BASEADDR_G(t4_read_reg(adap,
-					SGE_DBVFIFO_BADDR_A));
-			md->limit = md->base + (size << 2) - 1;
-		}
-	}
+	spin_lock(&adap->stats_lock);
+	t4_tp_get_tcp_stats(adap, &v4, &v6, false);
+	spin_unlock(&adap->stats_lock);
+
+	PRINT_ADAP_STATS("tcp_ipv4_out_rsts:", v4.tcp_out_rsts);
+	PRINT_ADAP_STATS("tcp_ipv4_in_segs:", v4.tcp_in_segs);
+	PRINT_ADAP_STATS("tcp_ipv4_out_segs:", v4.tcp_out_segs);
+	PRINT_ADAP_STATS("tcp_ipv4_retrans_segs:", v4.tcp_retrans_segs);
+	PRINT_ADAP_STATS("tcp_ipv6_out_rsts:", v6.tcp_out_rsts);
+	PRINT_ADAP_STATS("tcp_ipv6_in_segs:", v6.tcp_in_segs);
+	PRINT_ADAP_STATS("tcp_ipv6_out_segs:", v6.tcp_out_segs);
+	PRINT_ADAP_STATS("tcp_ipv6_retrans_segs:", v6.tcp_retrans_segs);
+}
+
+static void show_ddp_stats(struct seq_file *seq)
+{
+	struct adapter *adap = seq->private;
+	struct tp_usm_stats stats;
 
-	md++;
+	spin_lock(&adap->stats_lock);
+	t4_get_usm_stats(adap, &stats, false);
+	spin_unlock(&adap->stats_lock);
 
-	md->base = t4_read_reg(adap, ULP_RX_CTX_BASE_A);
-	md->limit = 0;
-	md++;
-	md->base = t4_read_reg(adap, ULP_TX_ERR_TABLE_BASE_A);
-	md->limit = 0;
-	md++;
+	PRINT_ADAP_STATS("usm_ddp_frames:", stats.frames);
+	PRINT_ADAP_STATS("usm_ddp_octets:", stats.octets);
+	PRINT_ADAP_STATS("usm_ddp_drops:", stats.drops);
+}
 
-	md->base = adap->vres.ocq.start;
-	if (adap->vres.ocq.size)
-		md->limit = md->base + adap->vres.ocq.size - 1;
-	else
-		md->idx = ARRAY_SIZE(region);  /* hide it */
-	md++;
+static void show_rdma_stats(struct seq_file *seq)
+{
+	struct adapter *adap = seq->private;
+	struct tp_rdma_stats stats;
 
-	/* add any address-space holes, there can be up to 3 */
-	for (n = 0; n < i - 1; n++)
-		if (avail[n].limit < avail[n + 1].base)
-			(md++)->base = avail[n].limit;
-	if (avail[n].limit)
-		(md++)->base = avail[n].limit;
+	spin_lock(&adap->stats_lock);
+	t4_tp_get_rdma_stats(adap, &stats, false);
+	spin_unlock(&adap->stats_lock);
 
-	n = md - mem;
-	sort(mem, n, sizeof(struct mem_desc), mem_desc_cmp, NULL);
+	PRINT_ADAP_STATS("rdma_no_rqe_mod_defer:", stats.rqe_dfr_mod);
+	PRINT_ADAP_STATS("rdma_no_rqe_pkt_defer:", stats.rqe_dfr_pkt);
+}
 
-	for (lo = 0; lo < i; lo++)
-		mem_region_show(seq, memory[avail[lo].idx], avail[lo].base,
-				avail[lo].limit - 1);
+static void show_tp_err_adapter_stats(struct seq_file *seq)
+{
+	struct adapter *adap = seq->private;
+	struct tp_err_stats stats;
 
-	seq_putc(seq, '\n');
-	for (i = 0; i < n; i++) {
-		if (mem[i].idx >= ARRAY_SIZE(region))
-			continue;                        /* skip holes */
-		if (!mem[i].limit)
-			mem[i].limit = i < n - 1 ? mem[i + 1].base - 1 : ~0;
-		mem_region_show(seq, region[mem[i].idx], mem[i].base,
-				mem[i].limit);
-	}
+	spin_lock(&adap->stats_lock);
+	t4_tp_get_err_stats(adap, &stats, false);
+	spin_unlock(&adap->stats_lock);
 
-	seq_putc(seq, '\n');
-	lo = t4_read_reg(adap, CIM_SDRAM_BASE_ADDR_A);
-	hi = t4_read_reg(adap, CIM_SDRAM_ADDR_SIZE_A) + lo - 1;
-	mem_region_show(seq, "uP RAM:", lo, hi);
-
-	lo = t4_read_reg(adap, CIM_EXTMEM2_BASE_ADDR_A);
-	hi = t4_read_reg(adap, CIM_EXTMEM2_ADDR_SIZE_A) + lo - 1;
-	mem_region_show(seq, "uP Extmem2:", lo, hi);
-
-	lo = t4_read_reg(adap, TP_PMM_RX_MAX_PAGE_A);
-	seq_printf(seq, "\n%u Rx pages of size %uKiB for %u channels\n",
-		   PMRXMAXPAGE_G(lo),
-		   t4_read_reg(adap, TP_PMM_RX_PAGE_SIZE_A) >> 10,
-		   (lo & PMRXNUMCHN_F) ? 2 : 1);
-
-	lo = t4_read_reg(adap, TP_PMM_TX_MAX_PAGE_A);
-	hi = t4_read_reg(adap, TP_PMM_TX_PAGE_SIZE_A);
-	seq_printf(seq, "%u Tx pages of size %u%ciB for %u channels\n",
-		   PMTXMAXPAGE_G(lo),
-		   hi >= (1 << 20) ? (hi >> 20) : (hi >> 10),
-		   hi >= (1 << 20) ? 'M' : 'K', 1 << PMTXNUMCHN_G(lo));
-	seq_printf(seq, "%u p-structs\n\n",
-		   t4_read_reg(adap, TP_CMM_MM_MAX_PSTRUCT_A));
-
-	for (i = 0; i < 4; i++) {
-		if (CHELSIO_CHIP_VERSION(adap->params.chip) > CHELSIO_T5)
-			lo = t4_read_reg(adap, MPS_RX_MAC_BG_PG_CNT0_A + i * 4);
-		else
-			lo = t4_read_reg(adap, MPS_RX_PG_RSV0_A + i * 4);
-		if (is_t5(adap->params.chip)) {
-			used = T5_USED_G(lo);
-			alloc = T5_ALLOC_G(lo);
-		} else {
-			used = USED_G(lo);
-			alloc = ALLOC_G(lo);
-		}
-		/* For T6 these are MAC buffer groups */
-		seq_printf(seq, "Port %d using %u pages out of %u allocated\n",
-			   i, used, alloc);
-	}
-	for (i = 0; i < adap->params.arch.nchan; i++) {
-		if (CHELSIO_CHIP_VERSION(adap->params.chip) > CHELSIO_T5)
-			lo = t4_read_reg(adap,
-					 MPS_RX_LPBK_BG_PG_CNT0_A + i * 4);
-		else
-			lo = t4_read_reg(adap, MPS_RX_PG_RSV4_A + i * 4);
-		if (is_t5(adap->params.chip)) {
-			used = T5_USED_G(lo);
-			alloc = T5_ALLOC_G(lo);
-		} else {
-			used = USED_G(lo);
-			alloc = ALLOC_G(lo);
-		}
-		/* For T6 these are MAC buffer groups */
-		seq_printf(seq,
-			   "Loopback %d using %u pages out of %u allocated\n",
-			   i, used, alloc);
-	}
-	return 0;
+	PRINT_ADAP_STATS("tp_err_ofld_no_neigh:", stats.ofld_no_neigh);
+	PRINT_ADAP_STATS("tp_err_ofld_cong_defer:", stats.ofld_cong_defer);
 }
 
-static int meminfo_open(struct inode *inode, struct file *file)
+static void show_cpl_stats(struct seq_file *seq)
 {
-	return single_open(file, meminfo_show, inode->i_private);
+	struct adapter *adap = seq->private;
+	struct tp_cpl_stats stats;
+	u8 i;
+
+	spin_lock(&adap->stats_lock);
+	t4_tp_get_cpl_stats(adap, &stats, false);
+	spin_unlock(&adap->stats_lock);
+
+	PRINT_CH_STATS("tp_cpl_requests:", req);
+	PRINT_CH_STATS("tp_cpl_responses:", rsp);
 }
 
-static const struct file_operations meminfo_fops = {
-	.owner   = THIS_MODULE,
-	.open    = meminfo_open,
-	.read    = seq_read,
-	.llseek  = seq_lseek,
-	.release = single_release,
-};
+static void show_tp_err_channel_stats(struct seq_file *seq)
+{
+	struct adapter *adap = seq->private;
+	struct tp_err_stats stats;
+	u8 i;
+
+	spin_lock(&adap->stats_lock);
+	t4_tp_get_err_stats(adap, &stats, false);
+	spin_unlock(&adap->stats_lock);
+
+	PRINT_CH_STATS("tp_mac_in_errs:", mac_in_errs);
+	PRINT_CH_STATS("tp_hdr_in_errs:", hdr_in_errs);
+	PRINT_CH_STATS("tp_tcp_in_errs:", tcp_in_errs);
+	PRINT_CH_STATS("tp_tcp6_in_errs:", tcp6_in_errs);
+	PRINT_CH_STATS("tp_tnl_cong_drops:", tnl_cong_drops);
+	PRINT_CH_STATS("tp_tnl_tx_drops:", tnl_tx_drops);
+	PRINT_CH_STATS("tp_ofld_vlan_drops:", ofld_vlan_drops);
+	PRINT_CH_STATS("tp_ofld_chan_drops:", ofld_chan_drops);
+}
+
+static void show_fcoe_stats(struct seq_file *seq)
+{
+	struct adapter *adap = seq->private;
+	struct tp_fcoe_stats stats[NCHAN];
+	u8 i;
+
+	spin_lock(&adap->stats_lock);
+	for (i = 0; i < adap->params.arch.nchan; i++)
+		t4_get_fcoe_stats(adap, i, &stats[i], false);
+	spin_unlock(&adap->stats_lock);
+
+	PRINT_CH_STATS2("fcoe_octets_ddp", octets_ddp);
+	PRINT_CH_STATS2("fcoe_frames_ddp", frames_ddp);
+	PRINT_CH_STATS2("fcoe_frames_drop", frames_drop);
+}
+
+#undef PRINT_CH_STATS2
+#undef PRINT_CH_STATS
+#undef PRINT_ADAP_STATS
+
+static int tp_stats_show(struct seq_file *seq, void *v)
+{
+	struct adapter *adap = seq->private;
+
+	seq_puts(seq, "\n--------Adapter Stats--------\n");
+	show_tcp_stats(seq);
+	show_ddp_stats(seq);
+	show_rdma_stats(seq);
+	show_tp_err_adapter_stats(seq);
+
+	seq_puts(seq, "\n-------- Channel Stats --------\n");
+	if (adap->params.arch.nchan == NCHAN)
+		seq_printf(seq, "%-25s %-20s %-20s %-20s %-20s\n",
+			   " ", "channel 0", "channel 1",
+			   "channel 2", "channel 3");
+	else
+		seq_printf(seq, "%-25s %-20s %-20s\n",
+			   " ", "channel 0", "channel 1");
+	show_cpl_stats(seq);
+	show_tp_err_channel_stats(seq);
+	show_fcoe_stats(seq);
+
+	return 0;
+}
+DEFINE_SHOW_ATTRIBUTE(tp_stats);
+
 /* Add an array of Debug FS files.
  */
 void add_debugfs_files(struct adapter *adap,
@@ -3089,67 +3772,72 @@ int t4_setup_debugfs(struct adapter *adap)
 {
 	int i;
 	u32 size = 0;
-	struct dentry *de;
 
 	static struct t4_debugfs_entry t4_debugfs_files[] = {
-		{ "cim_la", &cim_la_fops, S_IRUSR, 0 },
-		{ "cim_pif_la", &cim_pif_la_fops, S_IRUSR, 0 },
-		{ "cim_ma_la", &cim_ma_la_fops, S_IRUSR, 0 },
-		{ "cim_qcfg", &cim_qcfg_fops, S_IRUSR, 0 },
-		{ "clk", &clk_debugfs_fops, S_IRUSR, 0 },
-		{ "devlog", &devlog_fops, S_IRUSR, 0 },
-		{ "mboxlog", &mboxlog_fops, S_IRUSR, 0 },
-		{ "mbox0", &mbox_debugfs_fops, S_IRUSR | S_IWUSR, 0 },
-		{ "mbox1", &mbox_debugfs_fops, S_IRUSR | S_IWUSR, 1 },
-		{ "mbox2", &mbox_debugfs_fops, S_IRUSR | S_IWUSR, 2 },
-		{ "mbox3", &mbox_debugfs_fops, S_IRUSR | S_IWUSR, 3 },
-		{ "mbox4", &mbox_debugfs_fops, S_IRUSR | S_IWUSR, 4 },
-		{ "mbox5", &mbox_debugfs_fops, S_IRUSR | S_IWUSR, 5 },
-		{ "mbox6", &mbox_debugfs_fops, S_IRUSR | S_IWUSR, 6 },
-		{ "mbox7", &mbox_debugfs_fops, S_IRUSR | S_IWUSR, 7 },
-		{ "trace0", &mps_trc_debugfs_fops, S_IRUSR | S_IWUSR, 0 },
-		{ "trace1", &mps_trc_debugfs_fops, S_IRUSR | S_IWUSR, 1 },
-		{ "trace2", &mps_trc_debugfs_fops, S_IRUSR | S_IWUSR, 2 },
-		{ "trace3", &mps_trc_debugfs_fops, S_IRUSR | S_IWUSR, 3 },
-		{ "l2t", &t4_l2t_fops, S_IRUSR, 0},
-		{ "mps_tcam", &mps_tcam_debugfs_fops, S_IRUSR, 0 },
-		{ "rss", &rss_debugfs_fops, S_IRUSR, 0 },
-		{ "rss_config", &rss_config_debugfs_fops, S_IRUSR, 0 },
-		{ "rss_key", &rss_key_debugfs_fops, S_IRUSR, 0 },
-		{ "rss_pf_config", &rss_pf_config_debugfs_fops, S_IRUSR, 0 },
-		{ "rss_vf_config", &rss_vf_config_debugfs_fops, S_IRUSR, 0 },
-		{ "sge_qinfo", &sge_qinfo_debugfs_fops, S_IRUSR, 0 },
-		{ "ibq_tp0",  &cim_ibq_fops, S_IRUSR, 0 },
-		{ "ibq_tp1",  &cim_ibq_fops, S_IRUSR, 1 },
-		{ "ibq_ulp",  &cim_ibq_fops, S_IRUSR, 2 },
-		{ "ibq_sge0", &cim_ibq_fops, S_IRUSR, 3 },
-		{ "ibq_sge1", &cim_ibq_fops, S_IRUSR, 4 },
-		{ "ibq_ncsi", &cim_ibq_fops, S_IRUSR, 5 },
-		{ "obq_ulp0", &cim_obq_fops, S_IRUSR, 0 },
-		{ "obq_ulp1", &cim_obq_fops, S_IRUSR, 1 },
-		{ "obq_ulp2", &cim_obq_fops, S_IRUSR, 2 },
-		{ "obq_ulp3", &cim_obq_fops, S_IRUSR, 3 },
-		{ "obq_sge",  &cim_obq_fops, S_IRUSR, 4 },
-		{ "obq_ncsi", &cim_obq_fops, S_IRUSR, 5 },
-		{ "tp_la", &tp_la_fops, S_IRUSR, 0 },
-		{ "ulprx_la", &ulprx_la_fops, S_IRUSR, 0 },
-		{ "sensors", &sensors_debugfs_fops, S_IRUSR, 0 },
-		{ "pm_stats", &pm_stats_debugfs_fops, S_IRUSR, 0 },
-		{ "tx_rate", &tx_rate_debugfs_fops, S_IRUSR, 0 },
-		{ "cctrl", &cctrl_tbl_debugfs_fops, S_IRUSR, 0 },
+		{ "cim_la", &cim_la_fops, 0400, 0 },
+		{ "cim_pif_la", &cim_pif_la_fops, 0400, 0 },
+		{ "cim_ma_la", &cim_ma_la_fops, 0400, 0 },
+		{ "cim_qcfg", &cim_qcfg_fops, 0400, 0 },
+		{ "clk", &clk_fops, 0400, 0 },
+		{ "devlog", &devlog_fops, 0400, 0 },
+		{ "mboxlog", &mboxlog_fops, 0400, 0 },
+		{ "mbox0", &mbox_debugfs_fops, 0600, 0 },
+		{ "mbox1", &mbox_debugfs_fops, 0600, 1 },
+		{ "mbox2", &mbox_debugfs_fops, 0600, 2 },
+		{ "mbox3", &mbox_debugfs_fops, 0600, 3 },
+		{ "mbox4", &mbox_debugfs_fops, 0600, 4 },
+		{ "mbox5", &mbox_debugfs_fops, 0600, 5 },
+		{ "mbox6", &mbox_debugfs_fops, 0600, 6 },
+		{ "mbox7", &mbox_debugfs_fops, 0600, 7 },
+		{ "trace0", &mps_trc_debugfs_fops, 0600, 0 },
+		{ "trace1", &mps_trc_debugfs_fops, 0600, 1 },
+		{ "trace2", &mps_trc_debugfs_fops, 0600, 2 },
+		{ "trace3", &mps_trc_debugfs_fops, 0600, 3 },
+		{ "l2t", &t4_l2t_fops, 0400, 0},
+		{ "mps_tcam", &mps_tcam_debugfs_fops, 0400, 0 },
+		{ "rss", &rss_debugfs_fops, 0400, 0 },
+		{ "rss_config", &rss_config_fops, 0400, 0 },
+		{ "rss_key", &rss_key_debugfs_fops, 0400, 0 },
+		{ "rss_pf_config", &rss_pf_config_debugfs_fops, 0400, 0 },
+		{ "rss_vf_config", &rss_vf_config_debugfs_fops, 0400, 0 },
+		{ "resources", &resources_fops, 0400, 0 },
+#ifdef CONFIG_CHELSIO_T4_DCB
+		{ "dcb_info", &dcb_info_debugfs_fops, 0400, 0 },
+#endif
+		{ "sge_qinfo", &sge_qinfo_debugfs_fops, 0400, 0 },
+		{ "ibq_tp0",  &cim_ibq_fops, 0400, 0 },
+		{ "ibq_tp1",  &cim_ibq_fops, 0400, 1 },
+		{ "ibq_ulp",  &cim_ibq_fops, 0400, 2 },
+		{ "ibq_sge0", &cim_ibq_fops, 0400, 3 },
+		{ "ibq_sge1", &cim_ibq_fops, 0400, 4 },
+		{ "ibq_ncsi", &cim_ibq_fops, 0400, 5 },
+		{ "obq_ulp0", &cim_obq_fops, 0400, 0 },
+		{ "obq_ulp1", &cim_obq_fops, 0400, 1 },
+		{ "obq_ulp2", &cim_obq_fops, 0400, 2 },
+		{ "obq_ulp3", &cim_obq_fops, 0400, 3 },
+		{ "obq_sge",  &cim_obq_fops, 0400, 4 },
+		{ "obq_ncsi", &cim_obq_fops, 0400, 5 },
+		{ "tp_la", &tp_la_fops, 0400, 0 },
+		{ "ulprx_la", &ulprx_la_fops, 0400, 0 },
+		{ "sensors", &sensors_fops, 0400, 0 },
+		{ "pm_stats", &pm_stats_debugfs_fops, 0400, 0 },
+		{ "tx_rate", &tx_rate_fops, 0400, 0 },
+		{ "cctrl", &cctrl_tbl_fops, 0400, 0 },
 #if IS_ENABLED(CONFIG_IPV6)
-		{ "clip_tbl", &clip_tbl_debugfs_fops, S_IRUSR, 0 },
+		{ "clip_tbl", &clip_tbl_fops, 0400, 0 },
 #endif
-		{ "tids", &tid_info_debugfs_fops, S_IRUSR, 0},
-		{ "blocked_fl", &blocked_fl_fops, S_IRUSR | S_IWUSR, 0 },
-		{ "meminfo", &meminfo_fops, S_IRUSR, 0 },
+		{ "tids", &tid_info_fops, 0400, 0},
+		{ "blocked_fl", &blocked_fl_fops, 0600, 0 },
+		{ "meminfo", &meminfo_fops, 0400, 0 },
+		{ "crypto", &chcr_stats_fops, 0400, 0 },
+		{ "tp_stats", &tp_stats_fops, 0400, 0 },
 	};
 
 	/* Debug FS nodes common to all T5 and later adapters.
 	 */
 	static struct t4_debugfs_entry t5_debugfs_files[] = {
-		{ "obq_sge_rx_q0", &cim_obq_fops, S_IRUSR, 6 },
-		{ "obq_sge_rx_q1", &cim_obq_fops, S_IRUSR, 7 },
+		{ "obq_sge_rx_q0", &cim_obq_fops, 0400, 6 },
+		{ "obq_sge_rx_q1", &cim_obq_fops, 0400, 7 },
 	};
 
 	add_debugfs_files(adap,
@@ -3186,13 +3874,19 @@ int t4_setup_debugfs(struct adapter *adap)
 			add_debugfs_mem(adap, "mc", MEM_MC,
 					EXT_MEM_SIZE_G(size));
 		}
+
+		if (i & HMA_MUX_F) {
+			size = t4_read_reg(adap, MA_EXT_MEMORY1_BAR_A);
+			add_debugfs_mem(adap, "hma", MEM_HMA,
+					EXT_MEM1_SIZE_G(size));
+		}
 	}
 
-	de = debugfs_create_file_size("flash", S_IRUSR, adap->debugfs_root, adap,
-				      &flash_debugfs_fops, adap->params.sf_size);
-	debugfs_create_bool("use_backdoor", S_IWUSR | S_IRUSR,
+	debugfs_create_file_size("flash", 0400, adap->debugfs_root, adap,
+				 &flash_debugfs_fops, adap->params.sf_size);
+	debugfs_create_bool("use_backdoor", 0600,
 			    adap->debugfs_root, &adap->use_bd);
-	debugfs_create_bool("trace_rss", S_IWUSR | S_IRUSR,
+	debugfs_create_bool("trace_rss", 0600,
 			    adap->debugfs_root, &adap->trace_rss);
 
 	return 0;
diff --git a/drivers/net/ethernet/chelsio/cxgb4/cxgb4_debugfs.h b/drivers/net/ethernet/chelsio/cxgb4/cxgb4_debugfs.h
index 23f43a0f8950..1471cf0deb58 100644
--- a/drivers/net/ethernet/chelsio/cxgb4/cxgb4_debugfs.h
+++ b/drivers/net/ethernet/chelsio/cxgb4/cxgb4_debugfs.h
@@ -37,19 +37,6 @@
 
 #include <linux/export.h>
 
-#define DEFINE_SIMPLE_DEBUGFS_FILE(name) \
-static int name##_open(struct inode *inode, struct file *file) \
-{ \
-	return single_open(file, name##_show, inode->i_private); \
-} \
-static const struct file_operations name##_debugfs_fops = { \
-	.owner   = THIS_MODULE, \
-	.open    = name##_open, \
-	.read    = seq_read, \
-	.llseek  = seq_lseek, \
-	.release = single_release \
-}
-
 struct t4_debugfs_entry {
 	const char *name;
 	const struct file_operations *ops;
@@ -62,7 +49,7 @@ struct seq_tab {
 	unsigned int rows;        /* # of entries */
 	unsigned char width;      /* size in bytes of each entry */
 	unsigned char skip_first; /* whether the first line is a header */
-	char data[0];             /* the table data */
+	char data[];             /* the table data */
 };
 
 static inline unsigned int hex2val(char c)
diff --git a/drivers/net/ethernet/chelsio/cxgb4/cxgb4_ethtool.c b/drivers/net/ethernet/chelsio/cxgb4/cxgb4_ethtool.c
index e9bab72253bb..23326235d4ab 100644
--- a/drivers/net/ethernet/chelsio/cxgb4/cxgb4_ethtool.c
+++ b/drivers/net/ethernet/chelsio/cxgb4/cxgb4_ethtool.c
@@ -1,18 +1,6 @@
+// SPDX-License-Identifier: GPL-2.0-only
 /*
  *  Copyright (C) 2013-2015 Chelsio Communications.  All rights reserved.
- *
- *  This program is free software; you can redistribute it and/or modify it
- *  under the terms and conditions of the GNU General Public License,
- *  version 2, as published by the Free Software Foundation.
- *
- *  This program is distributed in the hope it will be useful, but WITHOUT
- *  ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
- *  FITNESS FOR A PARTICULAR PURPOSE.  See the GNU General Public License for
- *  more details.
- *
- *  The full GNU General Public License is included in this distribution in
- *  the file called "COPYING".
- *
  */
 
 #include <linux/firmware.h>
@@ -21,6 +9,9 @@
 #include "cxgb4.h"
 #include "t4_regs.h"
 #include "t4fw_api.h"
+#include "cxgb4_cudbg.h"
+#include "cxgb4_filter.h"
+#include "cxgb4_tc_flower.h"
 
 #define EEPROM_MAGIC 0x38E2F10C
 
@@ -34,6 +25,23 @@ static void set_msglevel(struct net_device *dev, u32 val)
 	netdev2adap(dev)->msg_enable = val;
 }
 
+enum cxgb4_ethtool_tests {
+	CXGB4_ETHTOOL_LB_TEST,
+	CXGB4_ETHTOOL_MAX_TEST,
+};
+
+static const char cxgb4_selftest_strings[CXGB4_ETHTOOL_MAX_TEST][ETH_GSTRING_LEN] = {
+	"Loop back test (offline)",
+};
+
+static const char * const flash_region_strings[] = {
+	"All",
+	"Firmware",
+	"PHY Firmware",
+	"Boot",
+	"Boot CFG",
+};
+
 static const char stats_strings[][ETH_GSTRING_LEN] = {
 	"tx_octets_ok           ",
 	"tx_frames_ok           ",
@@ -102,54 +110,32 @@ static const char stats_strings[][ETH_GSTRING_LEN] = {
 	"rx_bg3_frames_trunc    ",
 
 	"tso                    ",
+	"uso                    ",
 	"tx_csum_offload        ",
 	"rx_csum_good           ",
 	"vlan_extractions       ",
 	"vlan_insertions        ",
 	"gro_packets            ",
 	"gro_merged             ",
+#if  IS_ENABLED(CONFIG_CHELSIO_TLS_DEVICE)
+	"tx_tls_encrypted_packets",
+	"tx_tls_encrypted_bytes  ",
+	"tx_tls_ctx              ",
+	"tx_tls_ooo              ",
+	"tx_tls_skip_no_sync_data",
+	"tx_tls_drop_no_sync_data",
+	"tx_tls_drop_bypass_req  ",
+#endif
 };
 
 static char adapter_stats_strings[][ETH_GSTRING_LEN] = {
 	"db_drop                ",
 	"db_full                ",
 	"db_empty               ",
-	"tcp_ipv4_out_rsts      ",
-	"tcp_ipv4_in_segs       ",
-	"tcp_ipv4_out_segs      ",
-	"tcp_ipv4_retrans_segs  ",
-	"tcp_ipv6_out_rsts      ",
-	"tcp_ipv6_in_segs       ",
-	"tcp_ipv6_out_segs      ",
-	"tcp_ipv6_retrans_segs  ",
-	"usm_ddp_frames         ",
-	"usm_ddp_octets         ",
-	"usm_ddp_drops          ",
-	"rdma_no_rqe_mod_defer  ",
-	"rdma_no_rqe_pkt_defer  ",
-	"tp_err_ofld_no_neigh   ",
-	"tp_err_ofld_cong_defer ",
 	"write_coal_success     ",
 	"write_coal_fail        ",
 };
 
-static char channel_stats_strings[][ETH_GSTRING_LEN] = {
-	"--------Channel--------- ",
-	"tp_cpl_requests        ",
-	"tp_cpl_responses       ",
-	"tp_mac_in_errs         ",
-	"tp_hdr_in_errs         ",
-	"tp_tcp_in_errs         ",
-	"tp_tcp6_in_errs        ",
-	"tp_tnl_cong_drops      ",
-	"tp_tnl_tx_drops        ",
-	"tp_ofld_vlan_drops     ",
-	"tp_ofld_chan_drops     ",
-	"fcoe_octets_ddp        ",
-	"fcoe_frames_ddp        ",
-	"fcoe_frames_drop       ",
-};
-
 static char loopback_stats_strings[][ETH_GSTRING_LEN] = {
 	"-------Loopback----------- ",
 	"octets_ok              ",
@@ -176,14 +162,21 @@ static char loopback_stats_strings[][ETH_GSTRING_LEN] = {
 	"bg3_frames_trunc       ",
 };
 
+static const char cxgb4_priv_flags_strings[][ETH_GSTRING_LEN] = {
+	[PRIV_FLAG_PORT_TX_VM_BIT] = "port_tx_vm_wr",
+};
+
 static int get_sset_count(struct net_device *dev, int sset)
 {
 	switch (sset) {
 	case ETH_SS_STATS:
 		return ARRAY_SIZE(stats_strings) +
 		       ARRAY_SIZE(adapter_stats_strings) +
-		       ARRAY_SIZE(channel_stats_strings) +
 		       ARRAY_SIZE(loopback_stats_strings);
+	case ETH_SS_PRIV_FLAGS:
+		return ARRAY_SIZE(cxgb4_priv_flags_strings);
+	case ETH_SS_TEST:
+		return ARRAY_SIZE(cxgb4_selftest_strings);
 	default:
 		return -EOPNOTSUPP;
 	}
@@ -206,16 +199,12 @@ static void get_drvinfo(struct net_device *dev, struct ethtool_drvinfo *info)
 	struct adapter *adapter = netdev2adap(dev);
 	u32 exprom_vers;
 
-	strlcpy(info->driver, cxgb4_driver_name, sizeof(info->driver));
-	strlcpy(info->version, cxgb4_driver_version,
-		sizeof(info->version));
-	strlcpy(info->bus_info, pci_name(adapter->pdev),
+	strscpy(info->driver, cxgb4_driver_name, sizeof(info->driver));
+	strscpy(info->bus_info, pci_name(adapter->pdev),
 		sizeof(info->bus_info));
 	info->regdump_len = get_regs_len(dev);
 
-	if (!adapter->params.fw_vers)
-		strcpy(info->fw_version, "N/A");
-	else
+	if (adapter->params.fw_vers)
 		snprintf(info->fw_version, sizeof(info->fw_version),
 			 "%u.%u.%u.%u, TP %u.%u.%u.%u",
 			 FW_HDR_FW_VER_MAJOR_G(adapter->params.fw_vers),
@@ -234,6 +223,7 @@ static void get_drvinfo(struct net_device *dev, struct ethtool_drvinfo *info)
 			 FW_HDR_FW_VER_MINOR_G(exprom_vers),
 			 FW_HDR_FW_VER_MICRO_G(exprom_vers),
 			 FW_HDR_FW_VER_BUILD_G(exprom_vers));
+	info->n_priv_flags = ARRAY_SIZE(cxgb4_priv_flags_strings);
 }
 
 static void get_strings(struct net_device *dev, u32 stringset, u8 *data)
@@ -244,11 +234,14 @@ static void get_strings(struct net_device *dev, u32 stringset, u8 *data)
 		memcpy(data, adapter_stats_strings,
 		       sizeof(adapter_stats_strings));
 		data += sizeof(adapter_stats_strings);
-		memcpy(data, channel_stats_strings,
-		       sizeof(channel_stats_strings));
-		data += sizeof(channel_stats_strings);
 		memcpy(data, loopback_stats_strings,
 		       sizeof(loopback_stats_strings));
+	} else if (stringset == ETH_SS_PRIV_FLAGS) {
+		memcpy(data, cxgb4_priv_flags_strings,
+		       sizeof(cxgb4_priv_flags_strings));
+	} else if (stringset == ETH_SS_TEST) {
+		memcpy(data, cxgb4_selftest_strings,
+		       sizeof(cxgb4_selftest_strings));
 	}
 }
 
@@ -257,64 +250,48 @@ static void get_strings(struct net_device *dev, u32 stringset, u8 *data)
  */
 struct queue_port_stats {
 	u64 tso;
+	u64 uso;
 	u64 tx_csum;
 	u64 rx_csum;
 	u64 vlan_ex;
 	u64 vlan_ins;
 	u64 gro_pkts;
 	u64 gro_merged;
+#if IS_ENABLED(CONFIG_CHELSIO_TLS_DEVICE)
+	u64 tx_tls_encrypted_packets;
+	u64 tx_tls_encrypted_bytes;
+	u64 tx_tls_ctx;
+	u64 tx_tls_ooo;
+	u64 tx_tls_skip_no_sync_data;
+	u64 tx_tls_drop_no_sync_data;
+	u64 tx_tls_drop_bypass_req;
+#endif
 };
 
 struct adapter_stats {
 	u64 db_drop;
 	u64 db_full;
 	u64 db_empty;
-	u64 tcp_v4_out_rsts;
-	u64 tcp_v4_in_segs;
-	u64 tcp_v4_out_segs;
-	u64 tcp_v4_retrans_segs;
-	u64 tcp_v6_out_rsts;
-	u64 tcp_v6_in_segs;
-	u64 tcp_v6_out_segs;
-	u64 tcp_v6_retrans_segs;
-	u64 frames;
-	u64 octets;
-	u64 drops;
-	u64 rqe_dfr_mod;
-	u64 rqe_dfr_pkt;
-	u64 ofld_no_neigh;
-	u64 ofld_cong_defer;
 	u64 wc_success;
 	u64 wc_fail;
 };
 
-struct channel_stats {
-	u64 cpl_req;
-	u64 cpl_rsp;
-	u64 mac_in_errs;
-	u64 hdr_in_errs;
-	u64 tcp_in_errs;
-	u64 tcp6_in_errs;
-	u64 tnl_cong_drops;
-	u64 tnl_tx_drops;
-	u64 ofld_vlan_drops;
-	u64 ofld_chan_drops;
-	u64 octets_ddp;
-	u64 frames_ddp;
-	u64 frames_drop;
-};
-
 static void collect_sge_port_stats(const struct adapter *adap,
 				   const struct port_info *p,
 				   struct queue_port_stats *s)
 {
-	int i;
 	const struct sge_eth_txq *tx = &adap->sge.ethtxq[p->first_qset];
 	const struct sge_eth_rxq *rx = &adap->sge.ethrxq[p->first_qset];
+#if IS_ENABLED(CONFIG_CHELSIO_TLS_DEVICE)
+	const struct ch_ktls_port_stats_debug *ktls_stats;
+#endif
+	struct sge_eohw_txq *eohw_tx;
+	unsigned int i;
 
 	memset(s, 0, sizeof(*s));
 	for (i = 0; i < p->nqsets; i++, rx++, tx++) {
 		s->tso += tx->tso;
+		s->uso += tx->uso;
 		s->tx_csum += tx->tx_cso;
 		s->rx_csum += rx->stats.rx_cso;
 		s->vlan_ex += rx->stats.vlan_ex;
@@ -322,49 +299,43 @@ static void collect_sge_port_stats(const struct adapter *adap,
 		s->gro_pkts += rx->stats.lro_pkts;
 		s->gro_merged += rx->stats.lro_merged;
 	}
+
+	if (adap->sge.eohw_txq) {
+		eohw_tx = &adap->sge.eohw_txq[p->first_qset];
+		for (i = 0; i < p->nqsets; i++, eohw_tx++) {
+			s->tso += eohw_tx->tso;
+			s->uso += eohw_tx->uso;
+			s->tx_csum += eohw_tx->tx_cso;
+			s->vlan_ins += eohw_tx->vlan_ins;
+		}
+	}
+#if IS_ENABLED(CONFIG_CHELSIO_TLS_DEVICE)
+	ktls_stats = &adap->ch_ktls_stats.ktls_port[p->port_id];
+	s->tx_tls_encrypted_packets =
+		atomic64_read(&ktls_stats->ktls_tx_encrypted_packets);
+	s->tx_tls_encrypted_bytes =
+		atomic64_read(&ktls_stats->ktls_tx_encrypted_bytes);
+	s->tx_tls_ctx = atomic64_read(&ktls_stats->ktls_tx_ctx);
+	s->tx_tls_ooo = atomic64_read(&ktls_stats->ktls_tx_ooo);
+	s->tx_tls_skip_no_sync_data =
+		atomic64_read(&ktls_stats->ktls_tx_skip_no_sync_data);
+	s->tx_tls_drop_no_sync_data =
+		atomic64_read(&ktls_stats->ktls_tx_drop_no_sync_data);
+	s->tx_tls_drop_bypass_req =
+		atomic64_read(&ktls_stats->ktls_tx_drop_bypass_req);
+#endif
 }
 
 static void collect_adapter_stats(struct adapter *adap, struct adapter_stats *s)
 {
-	struct tp_tcp_stats v4, v6;
-	struct tp_rdma_stats rdma_stats;
-	struct tp_err_stats err_stats;
-	struct tp_usm_stats usm_stats;
 	u64 val1, val2;
 
 	memset(s, 0, sizeof(*s));
 
-	spin_lock(&adap->stats_lock);
-	t4_tp_get_tcp_stats(adap, &v4, &v6);
-	t4_tp_get_rdma_stats(adap, &rdma_stats);
-	t4_get_usm_stats(adap, &usm_stats);
-	t4_tp_get_err_stats(adap, &err_stats);
-	spin_unlock(&adap->stats_lock);
-
 	s->db_drop = adap->db_stats.db_drop;
 	s->db_full = adap->db_stats.db_full;
 	s->db_empty = adap->db_stats.db_empty;
 
-	s->tcp_v4_out_rsts = v4.tcp_out_rsts;
-	s->tcp_v4_in_segs = v4.tcp_in_segs;
-	s->tcp_v4_out_segs = v4.tcp_out_segs;
-	s->tcp_v4_retrans_segs = v4.tcp_retrans_segs;
-	s->tcp_v6_out_rsts = v6.tcp_out_rsts;
-	s->tcp_v6_in_segs = v6.tcp_in_segs;
-	s->tcp_v6_out_segs = v6.tcp_out_segs;
-	s->tcp_v6_retrans_segs = v6.tcp_retrans_segs;
-
-	if (is_offload(adap)) {
-		s->frames = usm_stats.frames;
-		s->octets = usm_stats.octets;
-		s->drops = usm_stats.drops;
-		s->rqe_dfr_mod = rdma_stats.rqe_dfr_mod;
-		s->rqe_dfr_pkt = rdma_stats.rqe_dfr_pkt;
-	}
-
-	s->ofld_no_neigh = err_stats.ofld_no_neigh;
-	s->ofld_cong_defer = err_stats.ofld_cong_defer;
-
 	if (!is_t4(adap->params.chip)) {
 		int v;
 
@@ -378,36 +349,6 @@ static void collect_adapter_stats(struct adapter *adap, struct adapter_stats *s)
 	}
 }
 
-static void collect_channel_stats(struct adapter *adap, struct channel_stats *s,
-				  u8 i)
-{
-	struct tp_cpl_stats cpl_stats;
-	struct tp_err_stats err_stats;
-	struct tp_fcoe_stats fcoe_stats;
-
-	memset(s, 0, sizeof(*s));
-
-	spin_lock(&adap->stats_lock);
-	t4_tp_get_cpl_stats(adap, &cpl_stats);
-	t4_tp_get_err_stats(adap, &err_stats);
-	t4_get_fcoe_stats(adap, i, &fcoe_stats);
-	spin_unlock(&adap->stats_lock);
-
-	s->cpl_req = cpl_stats.req[i];
-	s->cpl_rsp = cpl_stats.rsp[i];
-	s->mac_in_errs = err_stats.mac_in_errs[i];
-	s->hdr_in_errs = err_stats.hdr_in_errs[i];
-	s->tcp_in_errs = err_stats.tcp_in_errs[i];
-	s->tcp6_in_errs = err_stats.tcp6_in_errs[i];
-	s->tnl_cong_drops = err_stats.tnl_cong_drops[i];
-	s->tnl_tx_drops = err_stats.tnl_tx_drops[i];
-	s->ofld_vlan_drops = err_stats.ofld_vlan_drops[i];
-	s->ofld_chan_drops = err_stats.ofld_chan_drops[i];
-	s->octets_ddp = fcoe_stats.octets_ddp;
-	s->frames_ddp = fcoe_stats.frames_ddp;
-	s->frames_drop = fcoe_stats.frames_drop;
-}
-
 static void get_stats(struct net_device *dev, struct ethtool_stats *stats,
 		      u64 *data)
 {
@@ -428,11 +369,6 @@ static void get_stats(struct net_device *dev, struct ethtool_stats *stats,
 	data += sizeof(struct adapter_stats) / sizeof(u64);
 
 	*data++ = (u64)pi->port_id;
-	collect_channel_stats(adapter, (struct channel_stats *)data,
-			      pi->port_id);
-	data += sizeof(struct channel_stats) / sizeof(u64);
-
-	*data++ = (u64)pi->port_id;
 	memset(&s, 0, sizeof(s));
 	t4_get_lb_stats(adapter, pi->port_id, &s);
 
@@ -516,7 +452,8 @@ static int from_fw_port_mod_type(enum fw_port_type port_type,
 		else
 			return PORT_OTHER;
 	} else if (port_type == FW_PORT_TYPE_KR4_100G ||
-		   port_type == FW_PORT_TYPE_KR_SFP28) {
+		   port_type == FW_PORT_TYPE_KR_SFP28 ||
+		   port_type == FW_PORT_TYPE_KR_XLAUI) {
 		return PORT_NONE;
 	}
 
@@ -533,17 +470,23 @@ static int from_fw_port_mod_type(enum fw_port_type port_type,
 static unsigned int speed_to_fw_caps(int speed)
 {
 	if (speed == 100)
-		return FW_PORT_CAP_SPEED_100M;
+		return FW_PORT_CAP32_SPEED_100M;
 	if (speed == 1000)
-		return FW_PORT_CAP_SPEED_1G;
+		return FW_PORT_CAP32_SPEED_1G;
 	if (speed == 10000)
-		return FW_PORT_CAP_SPEED_10G;
+		return FW_PORT_CAP32_SPEED_10G;
 	if (speed == 25000)
-		return FW_PORT_CAP_SPEED_25G;
+		return FW_PORT_CAP32_SPEED_25G;
 	if (speed == 40000)
-		return FW_PORT_CAP_SPEED_40G;
+		return FW_PORT_CAP32_SPEED_40G;
+	if (speed == 50000)
+		return FW_PORT_CAP32_SPEED_50G;
 	if (speed == 100000)
-		return FW_PORT_CAP_SPEED_100G;
+		return FW_PORT_CAP32_SPEED_100G;
+	if (speed == 200000)
+		return FW_PORT_CAP32_SPEED_200G;
+	if (speed == 400000)
+		return FW_PORT_CAP32_SPEED_400G;
 	return 0;
 }
 
@@ -557,15 +500,18 @@ static unsigned int speed_to_fw_caps(int speed)
  *	Link Mode Mask.
  */
 static void fw_caps_to_lmm(enum fw_port_type port_type,
-			   unsigned int fw_caps,
+			   fw_port_cap32_t fw_caps,
 			   unsigned long *link_mode_mask)
 {
-	#define SET_LMM(__lmm_name) __set_bit(ETHTOOL_LINK_MODE_ ## __lmm_name \
-					## _BIT, link_mode_mask)
+	#define SET_LMM(__lmm_name) \
+		do { \
+			__set_bit(ETHTOOL_LINK_MODE_ ## __lmm_name ## _BIT, \
+				  link_mode_mask); \
+		} while (0)
 
 	#define FW_CAPS_TO_LMM(__fw_name, __lmm_name) \
 		do { \
-			if (fw_caps & FW_PORT_CAP_ ## __fw_name) \
+			if (fw_caps & FW_PORT_CAP32_ ## __fw_name) \
 				SET_LMM(__lmm_name); \
 		} while (0)
 
@@ -588,22 +534,22 @@ static void fw_caps_to_lmm(enum fw_port_type port_type,
 
 	case FW_PORT_TYPE_KR:
 		SET_LMM(Backplane);
-		SET_LMM(10000baseKR_Full);
+		FW_CAPS_TO_LMM(SPEED_10G, 10000baseKR_Full);
 		break;
 
 	case FW_PORT_TYPE_BP_AP:
 		SET_LMM(Backplane);
-		SET_LMM(10000baseR_FEC);
-		SET_LMM(10000baseKR_Full);
-		SET_LMM(1000baseKX_Full);
+		FW_CAPS_TO_LMM(SPEED_1G, 1000baseKX_Full);
+		FW_CAPS_TO_LMM(SPEED_10G, 10000baseR_FEC);
+		FW_CAPS_TO_LMM(SPEED_10G, 10000baseKR_Full);
 		break;
 
 	case FW_PORT_TYPE_BP4_AP:
 		SET_LMM(Backplane);
-		SET_LMM(10000baseR_FEC);
-		SET_LMM(10000baseKR_Full);
-		SET_LMM(1000baseKX_Full);
-		SET_LMM(10000baseKX4_Full);
+		FW_CAPS_TO_LMM(SPEED_1G, 1000baseKX_Full);
+		FW_CAPS_TO_LMM(SPEED_10G, 10000baseR_FEC);
+		FW_CAPS_TO_LMM(SPEED_10G, 10000baseKR_Full);
+		FW_CAPS_TO_LMM(SPEED_10G, 10000baseKX4_Full);
 		break;
 
 	case FW_PORT_TYPE_FIBER_XFI:
@@ -619,7 +565,9 @@ static void fw_caps_to_lmm(enum fw_port_type port_type,
 	case FW_PORT_TYPE_BP40_BA:
 	case FW_PORT_TYPE_QSFP:
 		SET_LMM(FIBRE);
-		SET_LMM(40000baseSR4_Full);
+		FW_CAPS_TO_LMM(SPEED_1G, 1000baseT_Full);
+		FW_CAPS_TO_LMM(SPEED_10G, 10000baseT_Full);
+		FW_CAPS_TO_LMM(SPEED_40G, 40000baseSR4_Full);
 		break;
 
 	case FW_PORT_TYPE_CR_QSFP:
@@ -637,21 +585,40 @@ static void fw_caps_to_lmm(enum fw_port_type port_type,
 		FW_CAPS_TO_LMM(SPEED_25G, 25000baseKR_Full);
 		break;
 
+	case FW_PORT_TYPE_KR_XLAUI:
+		SET_LMM(Backplane);
+		FW_CAPS_TO_LMM(SPEED_1G, 1000baseKX_Full);
+		FW_CAPS_TO_LMM(SPEED_10G, 10000baseKR_Full);
+		FW_CAPS_TO_LMM(SPEED_40G, 40000baseKR4_Full);
+		break;
+
 	case FW_PORT_TYPE_CR2_QSFP:
 		SET_LMM(FIBRE);
-		SET_LMM(50000baseSR2_Full);
+		FW_CAPS_TO_LMM(SPEED_50G, 50000baseSR2_Full);
 		break;
 
 	case FW_PORT_TYPE_KR4_100G:
 	case FW_PORT_TYPE_CR4_QSFP:
 		SET_LMM(FIBRE);
-		SET_LMM(100000baseCR4_Full);
+		FW_CAPS_TO_LMM(SPEED_1G,  1000baseT_Full);
+		FW_CAPS_TO_LMM(SPEED_10G, 10000baseKR_Full);
+		FW_CAPS_TO_LMM(SPEED_40G, 40000baseSR4_Full);
+		FW_CAPS_TO_LMM(SPEED_25G, 25000baseCR_Full);
+		FW_CAPS_TO_LMM(SPEED_50G, 50000baseCR2_Full);
+		FW_CAPS_TO_LMM(SPEED_100G, 100000baseCR4_Full);
 		break;
 
 	default:
 		break;
 	}
 
+	if (fw_caps & FW_PORT_CAP32_FEC_V(FW_PORT_CAP32_FEC_M)) {
+		FW_CAPS_TO_LMM(FEC_RS, FEC_RS);
+		FW_CAPS_TO_LMM(FEC_BASER_RS, FEC_BASER);
+	} else {
+		SET_LMM(FEC_NONE);
+	}
+
 	FW_CAPS_TO_LMM(ANEG, Autoneg);
 	FW_CAPS_TO_LMM(802_3_PAUSE, Pause);
 	FW_CAPS_TO_LMM(802_3_ASM_DIR, Asym_Pause);
@@ -663,7 +630,6 @@ static void fw_caps_to_lmm(enum fw_port_type port_type,
 /**
  *	lmm_to_fw_caps - translate ethtool Link Mode Mask to Firmware
  *	capabilities
- *
  *	@link_mode_mask: ethtool Link Mode Mask
  *
  *	Translate ethtool Link Mode Mask into a Firmware Port capabilities
@@ -677,7 +643,7 @@ static unsigned int lmm_to_fw_caps(const unsigned long *link_mode_mask)
 		do { \
 			if (test_bit(ETHTOOL_LINK_MODE_ ## __lmm_name ## _BIT, \
 				     link_mode_mask)) \
-				fw_caps |= FW_PORT_CAP_ ## __fw_name; \
+				fw_caps |= FW_PORT_CAP32_ ## __fw_name; \
 		} while (0)
 
 	LMM_TO_FW_CAPS(100baseT_Full, SPEED_100M);
@@ -685,6 +651,7 @@ static unsigned int lmm_to_fw_caps(const unsigned long *link_mode_mask)
 	LMM_TO_FW_CAPS(10000baseT_Full, SPEED_10G);
 	LMM_TO_FW_CAPS(40000baseSR4_Full, SPEED_40G);
 	LMM_TO_FW_CAPS(25000baseCR_Full, SPEED_25G);
+	LMM_TO_FW_CAPS(50000baseCR2_Full, SPEED_50G);
 	LMM_TO_FW_CAPS(100000baseCR4_Full, SPEED_100G);
 
 	#undef LMM_TO_FW_CAPS
@@ -698,10 +665,6 @@ static int get_link_ksettings(struct net_device *dev,
 	struct port_info *pi = netdev_priv(dev);
 	struct ethtool_link_settings *base = &link_ksettings->base;
 
-	ethtool_link_ksettings_zero_link_mode(link_ksettings, supported);
-	ethtool_link_ksettings_zero_link_mode(link_ksettings, advertising);
-	ethtool_link_ksettings_zero_link_mode(link_ksettings, lp_advertising);
-
 	/* For the nonce, the Firmware doesn't send up Port State changes
 	 * when the Virtual Interface attached to the Port is down.  So
 	 * if it's down, let's grab any changes.
@@ -709,6 +672,10 @@ static int get_link_ksettings(struct net_device *dev,
 	if (!netif_running(dev))
 		(void)t4_update_port_info(pi);
 
+	ethtool_link_ksettings_zero_link_mode(link_ksettings, supported);
+	ethtool_link_ksettings_zero_link_mode(link_ksettings, advertising);
+	ethtool_link_ksettings_zero_link_mode(link_ksettings, lp_advertising);
+
 	base->port = from_fw_port_mod_type(pi->port_type, pi->mod_type);
 
 	if (pi->mdio_addr >= 0) {
@@ -721,23 +688,23 @@ static int get_link_ksettings(struct net_device *dev,
 		base->mdio_support = 0;
 	}
 
-	fw_caps_to_lmm(pi->port_type, pi->link_cfg.supported,
+	fw_caps_to_lmm(pi->port_type, pi->link_cfg.pcaps,
 		       link_ksettings->link_modes.supported);
-	fw_caps_to_lmm(pi->port_type, pi->link_cfg.advertising,
+	fw_caps_to_lmm(pi->port_type,
+		       t4_link_acaps(pi->adapter,
+				     pi->lport,
+				     &pi->link_cfg),
 		       link_ksettings->link_modes.advertising);
-	fw_caps_to_lmm(pi->port_type, pi->link_cfg.lp_advertising,
+	fw_caps_to_lmm(pi->port_type, pi->link_cfg.lpacaps,
 		       link_ksettings->link_modes.lp_advertising);
 
-	if (netif_carrier_ok(dev)) {
-		base->speed = pi->link_cfg.speed;
-		base->duplex = DUPLEX_FULL;
-	} else {
-		base->speed = SPEED_UNKNOWN;
-		base->duplex = DUPLEX_UNKNOWN;
-	}
+	base->speed = (netif_carrier_ok(dev)
+		       ? pi->link_cfg.speed
+		       : SPEED_UNKNOWN);
+	base->duplex = DUPLEX_FULL;
 
 	base->autoneg = pi->link_cfg.autoneg;
-	if (pi->link_cfg.supported & FW_PORT_CAP_ANEG)
+	if (pi->link_cfg.pcaps & FW_PORT_CAP32_ANEG)
 		ethtool_link_ksettings_add_link_mode(link_ksettings,
 						     supported, Autoneg);
 	if (pi->link_cfg.autoneg)
@@ -748,8 +715,7 @@ static int get_link_ksettings(struct net_device *dev,
 }
 
 static int set_link_ksettings(struct net_device *dev,
-			      const struct ethtool_link_ksettings
-						*link_ksettings)
+			    const struct ethtool_link_ksettings *link_ksettings)
 {
 	struct port_info *pi = netdev_priv(dev);
 	struct link_config *lc = &pi->link_cfg;
@@ -762,32 +728,24 @@ static int set_link_ksettings(struct net_device *dev,
 	if (base->duplex != DUPLEX_FULL)
 		return -EINVAL;
 
-	if (!(lc->supported & FW_PORT_CAP_ANEG)) {
-		/* PHY offers a single speed.  See if that's what's
-		 * being requested.
-		 */
-		if (base->autoneg == AUTONEG_DISABLE &&
-		    (lc->supported & speed_to_fw_caps(base->speed)))
-			return 0;
-		return -EINVAL;
-	}
-
 	old_lc = *lc;
-	if (base->autoneg == AUTONEG_DISABLE) {
+	if (!(lc->pcaps & FW_PORT_CAP32_ANEG) ||
+	    base->autoneg == AUTONEG_DISABLE) {
 		fw_caps = speed_to_fw_caps(base->speed);
 
-		if (!(lc->supported & fw_caps))
+		/* Speed must be supported by Physical Port Capabilities. */
+		if (!(lc->pcaps & fw_caps))
 			return -EINVAL;
-		lc->requested_speed = fw_caps;
-		lc->advertising = 0;
+
+		lc->speed_caps = fw_caps;
+		lc->acaps = fw_caps;
 	} else {
 		fw_caps =
 			lmm_to_fw_caps(link_ksettings->link_modes.advertising);
-
-		if (!(lc->supported & fw_caps))
+		if (!(lc->pcaps & fw_caps))
 			return -EINVAL;
-		lc->requested_speed = 0;
-		lc->advertising = fw_caps | FW_PORT_CAP_ANEG;
+		lc->speed_caps = 0;
+		lc->acaps = fw_caps | FW_PORT_CAP32_ANEG;
 	}
 	lc->autoneg = base->autoneg;
 
@@ -801,14 +759,112 @@ static int set_link_ksettings(struct net_device *dev,
 	return ret;
 }
 
+/* Translate the Firmware FEC value into the ethtool value. */
+static inline unsigned int fwcap_to_eth_fec(unsigned int fw_fec)
+{
+	unsigned int eth_fec = 0;
+
+	if (fw_fec & FW_PORT_CAP32_FEC_RS)
+		eth_fec |= ETHTOOL_FEC_RS;
+	if (fw_fec & FW_PORT_CAP32_FEC_BASER_RS)
+		eth_fec |= ETHTOOL_FEC_BASER;
+
+	/* if nothing is set, then FEC is off */
+	if (!eth_fec)
+		eth_fec = ETHTOOL_FEC_OFF;
+
+	return eth_fec;
+}
+
+/* Translate Common Code FEC value into ethtool value. */
+static inline unsigned int cc_to_eth_fec(unsigned int cc_fec)
+{
+	unsigned int eth_fec = 0;
+
+	if (cc_fec & FEC_AUTO)
+		eth_fec |= ETHTOOL_FEC_AUTO;
+	if (cc_fec & FEC_RS)
+		eth_fec |= ETHTOOL_FEC_RS;
+	if (cc_fec & FEC_BASER_RS)
+		eth_fec |= ETHTOOL_FEC_BASER;
+
+	/* if nothing is set, then FEC is off */
+	if (!eth_fec)
+		eth_fec = ETHTOOL_FEC_OFF;
+
+	return eth_fec;
+}
+
+/* Translate ethtool FEC value into Common Code value. */
+static inline unsigned int eth_to_cc_fec(unsigned int eth_fec)
+{
+	unsigned int cc_fec = 0;
+
+	if (eth_fec & ETHTOOL_FEC_OFF)
+		return cc_fec;
+
+	if (eth_fec & ETHTOOL_FEC_AUTO)
+		cc_fec |= FEC_AUTO;
+	if (eth_fec & ETHTOOL_FEC_RS)
+		cc_fec |= FEC_RS;
+	if (eth_fec & ETHTOOL_FEC_BASER)
+		cc_fec |= FEC_BASER_RS;
+
+	return cc_fec;
+}
+
+static int get_fecparam(struct net_device *dev, struct ethtool_fecparam *fec)
+{
+	const struct port_info *pi = netdev_priv(dev);
+	const struct link_config *lc = &pi->link_cfg;
+
+	/* Translate the Firmware FEC Support into the ethtool value.  We
+	 * always support IEEE 802.3 "automatic" selection of Link FEC type if
+	 * any FEC is supported.
+	 */
+	fec->fec = fwcap_to_eth_fec(lc->pcaps);
+	if (fec->fec != ETHTOOL_FEC_OFF)
+		fec->fec |= ETHTOOL_FEC_AUTO;
+
+	/* Translate the current internal FEC parameters into the
+	 * ethtool values.
+	 */
+	fec->active_fec = cc_to_eth_fec(lc->fec);
+
+	return 0;
+}
+
+static int set_fecparam(struct net_device *dev, struct ethtool_fecparam *fec)
+{
+	struct port_info *pi = netdev_priv(dev);
+	struct link_config *lc = &pi->link_cfg;
+	struct link_config old_lc;
+	int ret;
+
+	/* Save old Link Configuration in case the L1 Configure below
+	 * fails.
+	 */
+	old_lc = *lc;
+
+	/* Try to perform the L1 Configure and return the result of that
+	 * effort.  If it fails, revert the attempted change.
+	 */
+	lc->requested_fec = eth_to_cc_fec(fec->fec);
+	ret = t4_link_l1cfg(pi->adapter, pi->adapter->mbox,
+			    pi->tx_chan, lc);
+	if (ret)
+		*lc = old_lc;
+	return ret;
+}
+
 static void get_pauseparam(struct net_device *dev,
 			   struct ethtool_pauseparam *epause)
 {
 	struct port_info *p = netdev_priv(dev);
 
 	epause->autoneg = (p->link_cfg.requested_fc & PAUSE_AUTONEG) != 0;
-	epause->rx_pause = (p->link_cfg.fc & PAUSE_RX) != 0;
-	epause->tx_pause = (p->link_cfg.fc & PAUSE_TX) != 0;
+	epause->rx_pause = (p->link_cfg.advertised_fc & PAUSE_RX) != 0;
+	epause->tx_pause = (p->link_cfg.advertised_fc & PAUSE_TX) != 0;
 }
 
 static int set_pauseparam(struct net_device *dev,
@@ -819,7 +875,7 @@ static int set_pauseparam(struct net_device *dev,
 
 	if (epause->autoneg == AUTONEG_DISABLE)
 		lc->requested_fc = 0;
-	else if (lc->supported & FW_PORT_CAP_ANEG)
+	else if (lc->pcaps & FW_PORT_CAP32_ANEG)
 		lc->requested_fc = PAUSE_AUTONEG;
 	else
 		return -EINVAL;
@@ -834,7 +890,9 @@ static int set_pauseparam(struct net_device *dev,
 	return 0;
 }
 
-static void get_sge_param(struct net_device *dev, struct ethtool_ringparam *e)
+static void get_sge_param(struct net_device *dev, struct ethtool_ringparam *e,
+			  struct kernel_ethtool_ringparam *kernel_e,
+			  struct netlink_ext_ack *extack)
 {
 	const struct port_info *pi = netdev_priv(dev);
 	const struct sge *s = &pi->adapter->sge;
@@ -850,7 +908,9 @@ static void get_sge_param(struct net_device *dev, struct ethtool_ringparam *e)
 	e->tx_pending = s->ethtxq[pi->first_qset].q.size;
 }
 
-static int set_sge_param(struct net_device *dev, struct ethtool_ringparam *e)
+static int set_sge_param(struct net_device *dev, struct ethtool_ringparam *e,
+			 struct kernel_ethtool_ringparam *kernel_e,
+			 struct netlink_ext_ack *extack)
 {
 	int i;
 	const struct port_info *pi = netdev_priv(dev);
@@ -864,7 +924,7 @@ static int set_sge_param(struct net_device *dev, struct ethtool_ringparam *e)
 	    e->rx_pending < MIN_FL_ENTRIES || e->tx_pending < MIN_TXQ_ENTRIES)
 		return -EINVAL;
 
-	if (adapter->flags & FULL_INIT_DONE)
+	if (adapter->flags & CXGB4_FULL_INIT_DONE)
 		return -EBUSY;
 
 	for (i = 0; i < pi->nqsets; ++i) {
@@ -921,14 +981,197 @@ static int get_adaptive_rx_setting(struct net_device *dev)
 	return q->rspq.adaptive_rx;
 }
 
-static int set_coalesce(struct net_device *dev, struct ethtool_coalesce *c)
+/* Return the current global Adapter SGE Doorbell Queue Timer Tick for all
+ * Ethernet TX Queues.
+ */
+static int get_dbqtimer_tick(struct net_device *dev)
+{
+	struct port_info *pi = netdev_priv(dev);
+	struct adapter *adap = pi->adapter;
+
+	if (!(adap->flags & CXGB4_SGE_DBQ_TIMER))
+		return 0;
+
+	return adap->sge.dbqtimer_tick;
+}
+
+/* Return the SGE Doorbell Queue Timer Value for the Ethernet TX Queues
+ * associated with a Network Device.
+ */
+static int get_dbqtimer(struct net_device *dev)
+{
+	struct port_info *pi = netdev_priv(dev);
+	struct adapter *adap = pi->adapter;
+	struct sge_eth_txq *txq;
+
+	txq = &adap->sge.ethtxq[pi->first_qset];
+
+	if (!(adap->flags & CXGB4_SGE_DBQ_TIMER))
+		return 0;
+
+	/* all of the TX Queues use the same Timer Index */
+	return adap->sge.dbqtimer_val[txq->dbqtimerix];
+}
+
+/* Set the global Adapter SGE Doorbell Queue Timer Tick for all Ethernet TX
+ * Queues.  This is the fundamental "Tick" that sets the scale of values which
+ * can be used.  Individual Ethernet TX Queues index into a relatively small
+ * array of Tick Multipliers.  Changing the base Tick will thus change all of
+ * the resulting Timer Values associated with those multipliers for all
+ * Ethernet TX Queues.
+ */
+static int set_dbqtimer_tick(struct net_device *dev, int usecs)
+{
+	struct port_info *pi = netdev_priv(dev);
+	struct adapter *adap = pi->adapter;
+	struct sge *s = &adap->sge;
+	u32 param, val;
+	int ret;
+
+	if (!(adap->flags & CXGB4_SGE_DBQ_TIMER))
+		return 0;
+
+	/* return early if it's the same Timer Tick we're already using */
+	if (s->dbqtimer_tick == usecs)
+		return 0;
+
+	/* attempt to set the new Timer Tick value */
+	param = (FW_PARAMS_MNEM_V(FW_PARAMS_MNEM_DEV) |
+		 FW_PARAMS_PARAM_X_V(FW_PARAMS_PARAM_DEV_DBQ_TIMERTICK));
+	val = usecs;
+	ret = t4_set_params(adap, adap->mbox, adap->pf, 0, 1, &param, &val);
+	if (ret)
+		return ret;
+	s->dbqtimer_tick = usecs;
+
+	/* if successful, reread resulting dependent Timer values */
+	ret = t4_read_sge_dbqtimers(adap, ARRAY_SIZE(s->dbqtimer_val),
+				    s->dbqtimer_val);
+	return ret;
+}
+
+/* Set the SGE Doorbell Queue Timer Value for the Ethernet TX Queues
+ * associated with a Network Device.  There is a relatively small array of
+ * possible Timer Values so we need to pick the closest value available.
+ */
+static int set_dbqtimer(struct net_device *dev, int usecs)
+{
+	int qix, timerix, min_timerix, delta, min_delta;
+	struct port_info *pi = netdev_priv(dev);
+	struct adapter *adap = pi->adapter;
+	struct sge *s = &adap->sge;
+	struct sge_eth_txq *txq;
+	u32 param, val;
+	int ret;
+
+	if (!(adap->flags & CXGB4_SGE_DBQ_TIMER))
+		return 0;
+
+	/* Find the SGE Doorbell Timer Value that's closest to the requested
+	 * value.
+	 */
+	min_delta = INT_MAX;
+	min_timerix = 0;
+	for (timerix = 0; timerix < ARRAY_SIZE(s->dbqtimer_val); timerix++) {
+		delta = s->dbqtimer_val[timerix] - usecs;
+		if (delta < 0)
+			delta = -delta;
+		if (delta < min_delta) {
+			min_delta = delta;
+			min_timerix = timerix;
+		}
+	}
+
+	/* Return early if it's the same Timer Index we're already using.
+	 * We use the same Timer Index for all of the TX Queues for an
+	 * interface so it's only necessary to check the first one.
+	 */
+	txq = &s->ethtxq[pi->first_qset];
+	if (txq->dbqtimerix == min_timerix)
+		return 0;
+
+	for (qix = 0; qix < pi->nqsets; qix++, txq++) {
+		if (adap->flags & CXGB4_FULL_INIT_DONE) {
+			param =
+			 (FW_PARAMS_MNEM_V(FW_PARAMS_MNEM_DMAQ) |
+			  FW_PARAMS_PARAM_X_V(FW_PARAMS_PARAM_DMAQ_EQ_TIMERIX) |
+			  FW_PARAMS_PARAM_YZ_V(txq->q.cntxt_id));
+			val = min_timerix;
+			ret = t4_set_params(adap, adap->mbox, adap->pf, 0,
+					    1, &param, &val);
+			if (ret)
+				return ret;
+		}
+		txq->dbqtimerix = min_timerix;
+	}
+	return 0;
+}
+
+/* Set the global Adapter SGE Doorbell Queue Timer Tick for all Ethernet TX
+ * Queues and the Timer Value for the Ethernet TX Queues associated with a
+ * Network Device.  Since changing the global Tick changes all of the
+ * available Timer Values, we need to do this first before selecting the
+ * resulting closest Timer Value.  Moreover, since the Tick is global,
+ * changing it affects the Timer Values for all Network Devices on the
+ * adapter.  So, before changing the Tick, we grab all of the current Timer
+ * Values for other Network Devices on this Adapter and then attempt to select
+ * new Timer Values which are close to the old values ...
+ */
+static int set_dbqtimer_tickval(struct net_device *dev,
+				int tick_usecs, int timer_usecs)
+{
+	struct port_info *pi = netdev_priv(dev);
+	struct adapter *adap = pi->adapter;
+	int timer[MAX_NPORTS];
+	unsigned int port;
+	int ret;
+
+	/* Grab the other adapter Network Interface current timers and fill in
+	 * the new one for this Network Interface.
+	 */
+	for_each_port(adap, port)
+		if (port == pi->port_id)
+			timer[port] = timer_usecs;
+		else
+			timer[port] = get_dbqtimer(adap->port[port]);
+
+	/* Change the global Tick first ... */
+	ret = set_dbqtimer_tick(dev, tick_usecs);
+	if (ret)
+		return ret;
+
+	/* ... and then set all of the Network Interface Timer Values ... */
+	for_each_port(adap, port) {
+		ret = set_dbqtimer(adap->port[port], timer[port]);
+		if (ret)
+			return ret;
+	}
+
+	return 0;
+}
+
+static int set_coalesce(struct net_device *dev,
+			struct ethtool_coalesce *coalesce,
+			struct kernel_ethtool_coalesce *kernel_coal,
+			struct netlink_ext_ack *extack)
 {
-	set_adaptive_rx_setting(dev, c->use_adaptive_rx_coalesce);
-	return set_rx_intr_params(dev, c->rx_coalesce_usecs,
-				  c->rx_max_coalesced_frames);
+	int ret;
+
+	set_adaptive_rx_setting(dev, coalesce->use_adaptive_rx_coalesce);
+
+	ret = set_rx_intr_params(dev, coalesce->rx_coalesce_usecs,
+				 coalesce->rx_max_coalesced_frames);
+	if (ret)
+		return ret;
+
+	return set_dbqtimer_tickval(dev,
+				    coalesce->tx_coalesce_usecs_irq,
+				    coalesce->tx_coalesce_usecs);
 }
 
-static int get_coalesce(struct net_device *dev, struct ethtool_coalesce *c)
+static int get_coalesce(struct net_device *dev, struct ethtool_coalesce *c,
+			struct kernel_ethtool_coalesce *kernel_coal,
+			struct netlink_ext_ack *extack)
 {
 	const struct port_info *pi = netdev_priv(dev);
 	const struct adapter *adap = pi->adapter;
@@ -938,43 +1181,16 @@ static int get_coalesce(struct net_device *dev, struct ethtool_coalesce *c)
 	c->rx_max_coalesced_frames = (rq->intr_params & QINTR_CNT_EN_F) ?
 		adap->sge.counter_val[rq->pktcnt_idx] : 0;
 	c->use_adaptive_rx_coalesce = get_adaptive_rx_setting(dev);
+	c->tx_coalesce_usecs_irq = get_dbqtimer_tick(dev);
+	c->tx_coalesce_usecs = get_dbqtimer(dev);
 	return 0;
 }
 
-/**
- *	eeprom_ptov - translate a physical EEPROM address to virtual
- *	@phys_addr: the physical EEPROM address
- *	@fn: the PCI function number
- *	@sz: size of function-specific area
- *
- *	Translate a physical EEPROM address to virtual.  The first 1K is
- *	accessed through virtual addresses starting at 31K, the rest is
- *	accessed through virtual addresses starting at 0.
- *
- *	The mapping is as follows:
- *	[0..1K) -> [31K..32K)
- *	[1K..1K+A) -> [31K-A..31K)
- *	[1K+A..ES) -> [0..ES-A-1K)
- *
- *	where A = @fn * @sz, and ES = EEPROM size.
- */
-static int eeprom_ptov(unsigned int phys_addr, unsigned int fn, unsigned int sz)
-{
-	fn *= sz;
-	if (phys_addr < 1024)
-		return phys_addr + (31 << 10);
-	if (phys_addr < 1024 + fn)
-		return 31744 - fn + phys_addr - 1024;
-	if (phys_addr < EEPROMSIZE)
-		return phys_addr - 1024 - fn;
-	return -EINVAL;
-}
-
 /* The next two routines implement eeprom read/write from physical addresses.
  */
 static int eeprom_rd_phys(struct adapter *adap, unsigned int phys_addr, u32 *v)
 {
-	int vaddr = eeprom_ptov(phys_addr, adap->pf, EEPROMPFSIZE);
+	int vaddr = t4_eeprom_ptov(phys_addr, adap->pf, EEPROMPFSIZE);
 
 	if (vaddr >= 0)
 		vaddr = pci_read_vpd(adap->pdev, vaddr, sizeof(u32), v);
@@ -983,7 +1199,7 @@ static int eeprom_rd_phys(struct adapter *adap, unsigned int phys_addr, u32 *v)
 
 static int eeprom_wr_phys(struct adapter *adap, unsigned int phys_addr, u32 v)
 {
-	int vaddr = eeprom_ptov(phys_addr, adap->pf, EEPROMPFSIZE);
+	int vaddr = t4_eeprom_ptov(phys_addr, adap->pf, EEPROMPFSIZE);
 
 	if (vaddr >= 0)
 		vaddr = pci_write_vpd(adap->pdev, vaddr, sizeof(u32), &v);
@@ -1069,15 +1285,225 @@ out:
 	return err;
 }
 
-static int set_flash(struct net_device *netdev, struct ethtool_flash *ef)
+static int cxgb4_ethtool_flash_bootcfg(struct net_device *netdev,
+				       const u8 *data, u32 size)
 {
+	struct adapter *adap = netdev2adap(netdev);
 	int ret;
-	const struct firmware *fw;
+
+	ret = t4_load_bootcfg(adap, data, size);
+	if (ret)
+		dev_err(adap->pdev_dev, "Failed to load boot cfg image\n");
+
+	return ret;
+}
+
+static int cxgb4_ethtool_flash_boot(struct net_device *netdev,
+				    const u8 *bdata, u32 size)
+{
+	struct adapter *adap = netdev2adap(netdev);
+	unsigned int offset;
+	u8 *data;
+	int ret;
+
+	data = kmemdup(bdata, size, GFP_KERNEL);
+	if (!data)
+		return -ENOMEM;
+
+	offset = OFFSET_G(t4_read_reg(adap, PF_REG(0, PCIE_PF_EXPROM_OFST_A)));
+
+	ret = t4_load_boot(adap, data, offset, size);
+	if (ret)
+		dev_err(adap->pdev_dev, "Failed to load boot image\n");
+
+	kfree(data);
+	return ret;
+}
+
+#define CXGB4_PHY_SIG 0x130000ea
+
+static int cxgb4_validate_phy_image(const u8 *data, u32 *size)
+{
+	struct cxgb4_fw_data *header;
+
+	header = (struct cxgb4_fw_data *)data;
+	if (be32_to_cpu(header->signature) != CXGB4_PHY_SIG)
+		return -EINVAL;
+
+	return 0;
+}
+
+static int cxgb4_ethtool_flash_phy(struct net_device *netdev,
+				   const u8 *data, u32 size)
+{
+	struct adapter *adap = netdev2adap(netdev);
+	int ret;
+
+	ret = cxgb4_validate_phy_image(data, NULL);
+	if (ret) {
+		dev_err(adap->pdev_dev, "PHY signature mismatch\n");
+		return ret;
+	}
+
+	/* We have to RESET the chip/firmware because we need the
+	 * chip in uninitialized state for loading new PHY image.
+	 * Otherwise, the running firmware will only store the PHY
+	 * image in local RAM which will be lost after next reset.
+	 */
+	ret = t4_fw_reset(adap, adap->mbox, PIORSTMODE_F | PIORST_F);
+	if (ret < 0) {
+		dev_err(adap->pdev_dev,
+			"Set FW to RESET for flashing PHY FW failed. ret: %d\n",
+			ret);
+		return ret;
+	}
+
+	ret = t4_load_phy_fw(adap, MEMWIN_NIC, NULL, data, size);
+	if (ret < 0) {
+		dev_err(adap->pdev_dev, "Failed to load PHY FW. ret: %d\n",
+			ret);
+		return ret;
+	}
+
+	return 0;
+}
+
+static int cxgb4_ethtool_flash_fw(struct net_device *netdev,
+				  const u8 *data, u32 size)
+{
 	struct adapter *adap = netdev2adap(netdev);
 	unsigned int mbox = PCIE_FW_MASTER_M + 1;
-	u32 pcie_fw;
+	int ret;
+
+	/* If the adapter has been fully initialized then we'll go ahead and
+	 * try to get the firmware's cooperation in upgrading to the new
+	 * firmware image otherwise we'll try to do the entire job from the
+	 * host ... and we always "force" the operation in this path.
+	 */
+	if (adap->flags & CXGB4_FULL_INIT_DONE)
+		mbox = adap->mbox;
+
+	ret = t4_fw_upgrade(adap, mbox, data, size, 1);
+	if (ret)
+		dev_err(adap->pdev_dev,
+			"Failed to flash firmware\n");
+
+	return ret;
+}
+
+static int cxgb4_ethtool_flash_region(struct net_device *netdev,
+				      const u8 *data, u32 size, u32 region)
+{
+	struct adapter *adap = netdev2adap(netdev);
+	int ret;
+
+	switch (region) {
+	case CXGB4_ETHTOOL_FLASH_FW:
+		ret = cxgb4_ethtool_flash_fw(netdev, data, size);
+		break;
+	case CXGB4_ETHTOOL_FLASH_PHY:
+		ret = cxgb4_ethtool_flash_phy(netdev, data, size);
+		break;
+	case CXGB4_ETHTOOL_FLASH_BOOT:
+		ret = cxgb4_ethtool_flash_boot(netdev, data, size);
+		break;
+	case CXGB4_ETHTOOL_FLASH_BOOTCFG:
+		ret = cxgb4_ethtool_flash_bootcfg(netdev, data, size);
+		break;
+	default:
+		ret = -EOPNOTSUPP;
+		break;
+	}
+
+	if (!ret)
+		dev_info(adap->pdev_dev,
+			 "loading %s successful, reload cxgb4 driver\n",
+			 flash_region_strings[region]);
+	return ret;
+}
+
+#define CXGB4_FW_SIG 0x4368656c
+#define CXGB4_FW_SIG_OFFSET 0x160
+
+static int cxgb4_validate_fw_image(const u8 *data, u32 *size)
+{
+	struct cxgb4_fw_data *header;
+
+	header = (struct cxgb4_fw_data *)&data[CXGB4_FW_SIG_OFFSET];
+	if (be32_to_cpu(header->signature) != CXGB4_FW_SIG)
+		return -EINVAL;
+
+	if (size)
+		*size = be16_to_cpu(((struct fw_hdr *)data)->len512) * 512;
+
+	return 0;
+}
+
+static int cxgb4_validate_bootcfg_image(const u8 *data, u32 *size)
+{
+	struct cxgb4_bootcfg_data *header;
+
+	header = (struct cxgb4_bootcfg_data *)data;
+	if (le16_to_cpu(header->signature) != BOOT_CFG_SIG)
+		return -EINVAL;
+
+	return 0;
+}
+
+static int cxgb4_validate_boot_image(const u8 *data, u32 *size)
+{
+	struct cxgb4_pci_exp_rom_header *exp_header;
+	struct cxgb4_pcir_data *pcir_header;
+	struct legacy_pci_rom_hdr *header;
+	const u8 *cur_header = data;
+	u16 pcir_offset;
+
+	exp_header = (struct cxgb4_pci_exp_rom_header *)data;
+
+	if (le16_to_cpu(exp_header->signature) != BOOT_SIGNATURE)
+		return -EINVAL;
+
+	if (size) {
+		do {
+			header = (struct legacy_pci_rom_hdr *)cur_header;
+			pcir_offset = le16_to_cpu(header->pcir_offset);
+			pcir_header = (struct cxgb4_pcir_data *)(cur_header +
+				      pcir_offset);
+
+			*size += header->size512 * 512;
+			cur_header += header->size512 * 512;
+		} while (!(pcir_header->indicator & CXGB4_HDR_INDI));
+	}
+
+	return 0;
+}
+
+static int cxgb4_ethtool_get_flash_region(const u8 *data, u32 *size)
+{
+	if (!cxgb4_validate_fw_image(data, size))
+		return CXGB4_ETHTOOL_FLASH_FW;
+	if (!cxgb4_validate_boot_image(data, size))
+		return CXGB4_ETHTOOL_FLASH_BOOT;
+	if (!cxgb4_validate_phy_image(data, size))
+		return CXGB4_ETHTOOL_FLASH_PHY;
+	if (!cxgb4_validate_bootcfg_image(data, size))
+		return CXGB4_ETHTOOL_FLASH_BOOTCFG;
+
+	return -EOPNOTSUPP;
+}
+
+static int set_flash(struct net_device *netdev, struct ethtool_flash *ef)
+{
+	struct adapter *adap = netdev2adap(netdev);
+	const struct firmware *fw;
 	unsigned int master;
 	u8 master_vld = 0;
+	const u8 *fw_data;
+	size_t fw_size;
+	u32 size = 0;
+	u32 pcie_fw;
+	int region;
+	int ret;
 
 	pcie_fw = t4_read_reg(adap, PCIE_FW_A);
 	master = PCIE_FW_MASTER_G(pcie_fw);
@@ -1095,32 +1521,57 @@ static int set_flash(struct net_device *netdev, struct ethtool_flash *ef)
 	if (ret < 0)
 		return ret;
 
-	/* If the adapter has been fully initialized then we'll go ahead and
-	 * try to get the firmware's cooperation in upgrading to the new
-	 * firmware image otherwise we'll try to do the entire job from the
-	 * host ... and we always "force" the operation in this path.
-	 */
-	if (adap->flags & FULL_INIT_DONE)
-		mbox = adap->mbox;
+	fw_data = fw->data;
+	fw_size = fw->size;
+	if (ef->region == ETHTOOL_FLASH_ALL_REGIONS) {
+		while (fw_size > 0) {
+			size = 0;
+			region = cxgb4_ethtool_get_flash_region(fw_data, &size);
+			if (region < 0 || !size) {
+				ret = region;
+				goto out_free_fw;
+			}
+
+			ret = cxgb4_ethtool_flash_region(netdev, fw_data, size,
+							 region);
+			if (ret)
+				goto out_free_fw;
+
+			fw_data += size;
+			fw_size -= size;
+		}
+	} else {
+		ret = cxgb4_ethtool_flash_region(netdev, fw_data, fw_size,
+						 ef->region);
+	}
 
-	ret = t4_fw_upgrade(adap, mbox, fw->data, fw->size, 1);
+out_free_fw:
 	release_firmware(fw);
-	if (!ret)
-		dev_info(adap->pdev_dev,
-			 "loaded firmware %s, reload cxgb4 driver\n", ef->data);
 	return ret;
 }
 
-static int get_ts_info(struct net_device *dev, struct ethtool_ts_info *ts_info)
+static int get_ts_info(struct net_device *dev, struct kernel_ethtool_ts_info *ts_info)
 {
+	struct port_info *pi = netdev_priv(dev);
+	struct  adapter *adapter = pi->adapter;
+
 	ts_info->so_timestamping = SOF_TIMESTAMPING_TX_SOFTWARE |
-				   SOF_TIMESTAMPING_RX_SOFTWARE |
-				   SOF_TIMESTAMPING_SOFTWARE;
+				   SOF_TIMESTAMPING_RX_HARDWARE |
+				   SOF_TIMESTAMPING_TX_HARDWARE |
+				   SOF_TIMESTAMPING_RAW_HARDWARE;
+
+	ts_info->tx_types = (1 << HWTSTAMP_TX_OFF) |
+			    (1 << HWTSTAMP_TX_ON);
 
-	ts_info->so_timestamping |= SOF_TIMESTAMPING_RX_HARDWARE |
-				    SOF_TIMESTAMPING_RAW_HARDWARE;
+	ts_info->rx_filters = (1 << HWTSTAMP_FILTER_NONE) |
+			      (1 << HWTSTAMP_FILTER_PTP_V2_L4_EVENT) |
+			      (1 << HWTSTAMP_FILTER_PTP_V1_L4_SYNC) |
+			      (1 << HWTSTAMP_FILTER_PTP_V1_L4_DELAY_REQ) |
+			      (1 << HWTSTAMP_FILTER_PTP_V2_L4_SYNC) |
+			      (1 << HWTSTAMP_FILTER_PTP_V2_L4_DELAY_REQ);
 
-	ts_info->phc_index = -1;
+	if (adapter->ptp_clock)
+		ts_info->phc_index = ptp_clock_index(adapter->ptp_clock);
 
 	return 0;
 }
@@ -1132,22 +1583,23 @@ static u32 get_rss_table_size(struct net_device *dev)
 	return pi->rss_size;
 }
 
-static int get_rss_table(struct net_device *dev, u32 *p, u8 *key, u8 *hfunc)
+static int get_rss_table(struct net_device *dev,
+			 struct ethtool_rxfh_param *rxfh)
 {
 	const struct port_info *pi = netdev_priv(dev);
 	unsigned int n = pi->rss_size;
 
-	if (hfunc)
-		*hfunc = ETH_RSS_HASH_TOP;
-	if (!p)
+	rxfh->hfunc = ETH_RSS_HASH_TOP;
+	if (!rxfh->indir)
 		return 0;
 	while (n--)
-		p[n] = pi->rss[n];
+		rxfh->indir[n] = pi->rss[n];
 	return 0;
 }
 
-static int set_rss_table(struct net_device *dev, const u32 *p, const u8 *key,
-			 const u8 hfunc)
+static int set_rss_table(struct net_device *dev,
+			 struct ethtool_rxfh_param *rxfh,
+			 struct netlink_ext_ack *extack)
 {
 	unsigned int i;
 	struct port_info *pi = netdev_priv(dev);
@@ -1155,16 +1607,17 @@ static int set_rss_table(struct net_device *dev, const u32 *p, const u8 *key,
 	/* We require at least one supported parameter to be changed and no
 	 * change in any of the unsupported parameters
 	 */
-	if (key ||
-	    (hfunc != ETH_RSS_HASH_NO_CHANGE && hfunc != ETH_RSS_HASH_TOP))
+	if (rxfh->key ||
+	    (rxfh->hfunc != ETH_RSS_HASH_NO_CHANGE &&
+	     rxfh->hfunc != ETH_RSS_HASH_TOP))
 		return -EOPNOTSUPP;
-	if (!p)
+	if (!rxfh->indir)
 		return 0;
 
 	/* Interface must be brought up atleast once */
-	if (pi->adapter->flags & FULL_INIT_DONE) {
+	if (pi->adapter->flags & CXGB4_FULL_INIT_DONE) {
 		for (i = 0; i < pi->rss_size; i++)
-			pi->rss[i] = p[i];
+			pi->rss[i] = rxfh->indir[i];
 
 		return cxgb4_write_rss(pi, pi->rss);
 	}
@@ -1172,72 +1625,559 @@ static int set_rss_table(struct net_device *dev, const u32 *p, const u8 *key,
 	return -EPERM;
 }
 
+static struct filter_entry *cxgb4_get_filter_entry(struct adapter *adap,
+						   u32 ftid)
+{
+	struct tid_info *t = &adap->tids;
+
+	if (ftid >= t->hpftid_base && ftid < t->hpftid_base + t->nhpftids)
+		return &t->hpftid_tab[ftid - t->hpftid_base];
+
+	if (ftid >= t->ftid_base && ftid < t->ftid_base + t->nftids)
+		return &t->ftid_tab[ftid - t->ftid_base];
+
+	return lookup_tid(t, ftid);
+}
+
+static void cxgb4_fill_filter_rule(struct ethtool_rx_flow_spec *fs,
+				   struct ch_filter_specification *dfs)
+{
+	switch (dfs->val.proto) {
+	case IPPROTO_TCP:
+		if (dfs->type)
+			fs->flow_type = TCP_V6_FLOW;
+		else
+			fs->flow_type = TCP_V4_FLOW;
+		break;
+	case IPPROTO_UDP:
+		if (dfs->type)
+			fs->flow_type = UDP_V6_FLOW;
+		else
+			fs->flow_type = UDP_V4_FLOW;
+		break;
+	}
+
+	if (dfs->type) {
+		fs->h_u.tcp_ip6_spec.psrc = cpu_to_be16(dfs->val.fport);
+		fs->m_u.tcp_ip6_spec.psrc = cpu_to_be16(dfs->mask.fport);
+		fs->h_u.tcp_ip6_spec.pdst = cpu_to_be16(dfs->val.lport);
+		fs->m_u.tcp_ip6_spec.pdst = cpu_to_be16(dfs->mask.lport);
+		memcpy(&fs->h_u.tcp_ip6_spec.ip6src, &dfs->val.fip[0],
+		       sizeof(fs->h_u.tcp_ip6_spec.ip6src));
+		memcpy(&fs->m_u.tcp_ip6_spec.ip6src, &dfs->mask.fip[0],
+		       sizeof(fs->m_u.tcp_ip6_spec.ip6src));
+		memcpy(&fs->h_u.tcp_ip6_spec.ip6dst, &dfs->val.lip[0],
+		       sizeof(fs->h_u.tcp_ip6_spec.ip6dst));
+		memcpy(&fs->m_u.tcp_ip6_spec.ip6dst, &dfs->mask.lip[0],
+		       sizeof(fs->m_u.tcp_ip6_spec.ip6dst));
+		fs->h_u.tcp_ip6_spec.tclass = dfs->val.tos;
+		fs->m_u.tcp_ip6_spec.tclass = dfs->mask.tos;
+	} else {
+		fs->h_u.tcp_ip4_spec.psrc = cpu_to_be16(dfs->val.fport);
+		fs->m_u.tcp_ip4_spec.psrc = cpu_to_be16(dfs->mask.fport);
+		fs->h_u.tcp_ip4_spec.pdst = cpu_to_be16(dfs->val.lport);
+		fs->m_u.tcp_ip4_spec.pdst = cpu_to_be16(dfs->mask.lport);
+		memcpy(&fs->h_u.tcp_ip4_spec.ip4src, &dfs->val.fip[0],
+		       sizeof(fs->h_u.tcp_ip4_spec.ip4src));
+		memcpy(&fs->m_u.tcp_ip4_spec.ip4src, &dfs->mask.fip[0],
+		       sizeof(fs->m_u.tcp_ip4_spec.ip4src));
+		memcpy(&fs->h_u.tcp_ip4_spec.ip4dst, &dfs->val.lip[0],
+		       sizeof(fs->h_u.tcp_ip4_spec.ip4dst));
+		memcpy(&fs->m_u.tcp_ip4_spec.ip4dst, &dfs->mask.lip[0],
+		       sizeof(fs->m_u.tcp_ip4_spec.ip4dst));
+		fs->h_u.tcp_ip4_spec.tos = dfs->val.tos;
+		fs->m_u.tcp_ip4_spec.tos = dfs->mask.tos;
+	}
+	fs->h_ext.vlan_tci = cpu_to_be16(dfs->val.ivlan);
+	fs->m_ext.vlan_tci = cpu_to_be16(dfs->mask.ivlan);
+	fs->flow_type |= FLOW_EXT;
+
+	if (dfs->action == FILTER_DROP)
+		fs->ring_cookie = RX_CLS_FLOW_DISC;
+	else
+		fs->ring_cookie = dfs->iq;
+}
+
+static int cxgb4_ntuple_get_filter(struct net_device *dev,
+				   struct ethtool_rxnfc *cmd,
+				   unsigned int loc)
+{
+	const struct port_info *pi = netdev_priv(dev);
+	struct adapter *adap = netdev2adap(dev);
+	struct filter_entry *f;
+	int ftid;
+
+	if (!(adap->flags & CXGB4_FULL_INIT_DONE))
+		return -EAGAIN;
+
+	/* Check for maximum filter range */
+	if (!adap->ethtool_filters)
+		return -EOPNOTSUPP;
+
+	if (loc >= adap->ethtool_filters->nentries)
+		return -ERANGE;
+
+	if (!test_bit(loc, adap->ethtool_filters->port[pi->port_id].bmap))
+		return -ENOENT;
+
+	ftid = adap->ethtool_filters->port[pi->port_id].loc_array[loc];
+
+	/* Fetch filter_entry */
+	f = cxgb4_get_filter_entry(adap, ftid);
+
+	cxgb4_fill_filter_rule(&cmd->fs, &f->fs);
+
+	return 0;
+}
+
+static int cxgb4_get_rxfh_fields(struct net_device *dev,
+				 struct ethtool_rxfh_fields *info)
+{
+	const struct port_info *pi = netdev_priv(dev);
+	unsigned int v = pi->rss_mode;
+
+	info->data = 0;
+	switch (info->flow_type) {
+	case TCP_V4_FLOW:
+		if (v & FW_RSS_VI_CONFIG_CMD_IP4FOURTUPEN_F)
+			info->data = RXH_IP_SRC | RXH_IP_DST |
+				RXH_L4_B_0_1 | RXH_L4_B_2_3;
+		else if (v & FW_RSS_VI_CONFIG_CMD_IP4TWOTUPEN_F)
+			info->data = RXH_IP_SRC | RXH_IP_DST;
+		break;
+	case UDP_V4_FLOW:
+		if ((v & FW_RSS_VI_CONFIG_CMD_IP4FOURTUPEN_F) &&
+		    (v & FW_RSS_VI_CONFIG_CMD_UDPEN_F))
+			info->data = RXH_IP_SRC | RXH_IP_DST |
+				RXH_L4_B_0_1 | RXH_L4_B_2_3;
+		else if (v & FW_RSS_VI_CONFIG_CMD_IP4TWOTUPEN_F)
+			info->data = RXH_IP_SRC | RXH_IP_DST;
+		break;
+	case SCTP_V4_FLOW:
+	case AH_ESP_V4_FLOW:
+	case IPV4_FLOW:
+		if (v & FW_RSS_VI_CONFIG_CMD_IP4TWOTUPEN_F)
+			info->data = RXH_IP_SRC | RXH_IP_DST;
+		break;
+	case TCP_V6_FLOW:
+		if (v & FW_RSS_VI_CONFIG_CMD_IP6FOURTUPEN_F)
+			info->data = RXH_IP_SRC | RXH_IP_DST |
+				RXH_L4_B_0_1 | RXH_L4_B_2_3;
+		else if (v & FW_RSS_VI_CONFIG_CMD_IP6TWOTUPEN_F)
+			info->data = RXH_IP_SRC | RXH_IP_DST;
+		break;
+	case UDP_V6_FLOW:
+		if ((v & FW_RSS_VI_CONFIG_CMD_IP6FOURTUPEN_F) &&
+		    (v & FW_RSS_VI_CONFIG_CMD_UDPEN_F))
+			info->data = RXH_IP_SRC | RXH_IP_DST |
+				RXH_L4_B_0_1 | RXH_L4_B_2_3;
+		else if (v & FW_RSS_VI_CONFIG_CMD_IP6TWOTUPEN_F)
+			info->data = RXH_IP_SRC | RXH_IP_DST;
+		break;
+	case SCTP_V6_FLOW:
+	case AH_ESP_V6_FLOW:
+	case IPV6_FLOW:
+		if (v & FW_RSS_VI_CONFIG_CMD_IP6TWOTUPEN_F)
+			info->data = RXH_IP_SRC | RXH_IP_DST;
+		break;
+	}
+	return 0;
+}
+
 static int get_rxnfc(struct net_device *dev, struct ethtool_rxnfc *info,
 		     u32 *rules)
 {
 	const struct port_info *pi = netdev_priv(dev);
+	struct adapter *adap = netdev2adap(dev);
+	unsigned int count = 0, index = 0;
+	int ret = 0;
 
 	switch (info->cmd) {
-	case ETHTOOL_GRXFH: {
-		unsigned int v = pi->rss_mode;
-
-		info->data = 0;
-		switch (info->flow_type) {
-		case TCP_V4_FLOW:
-			if (v & FW_RSS_VI_CONFIG_CMD_IP4FOURTUPEN_F)
-				info->data = RXH_IP_SRC | RXH_IP_DST |
-					     RXH_L4_B_0_1 | RXH_L4_B_2_3;
-			else if (v & FW_RSS_VI_CONFIG_CMD_IP4TWOTUPEN_F)
-				info->data = RXH_IP_SRC | RXH_IP_DST;
-			break;
-		case UDP_V4_FLOW:
-			if ((v & FW_RSS_VI_CONFIG_CMD_IP4FOURTUPEN_F) &&
-			    (v & FW_RSS_VI_CONFIG_CMD_UDPEN_F))
-				info->data = RXH_IP_SRC | RXH_IP_DST |
-					     RXH_L4_B_0_1 | RXH_L4_B_2_3;
-			else if (v & FW_RSS_VI_CONFIG_CMD_IP4TWOTUPEN_F)
-				info->data = RXH_IP_SRC | RXH_IP_DST;
-			break;
-		case SCTP_V4_FLOW:
-		case AH_ESP_V4_FLOW:
-		case IPV4_FLOW:
-			if (v & FW_RSS_VI_CONFIG_CMD_IP4TWOTUPEN_F)
-				info->data = RXH_IP_SRC | RXH_IP_DST;
-			break;
-		case TCP_V6_FLOW:
-			if (v & FW_RSS_VI_CONFIG_CMD_IP6FOURTUPEN_F)
-				info->data = RXH_IP_SRC | RXH_IP_DST |
-					     RXH_L4_B_0_1 | RXH_L4_B_2_3;
-			else if (v & FW_RSS_VI_CONFIG_CMD_IP6TWOTUPEN_F)
-				info->data = RXH_IP_SRC | RXH_IP_DST;
-			break;
-		case UDP_V6_FLOW:
-			if ((v & FW_RSS_VI_CONFIG_CMD_IP6FOURTUPEN_F) &&
-			    (v & FW_RSS_VI_CONFIG_CMD_UDPEN_F))
-				info->data = RXH_IP_SRC | RXH_IP_DST |
-					     RXH_L4_B_0_1 | RXH_L4_B_2_3;
-			else if (v & FW_RSS_VI_CONFIG_CMD_IP6TWOTUPEN_F)
-				info->data = RXH_IP_SRC | RXH_IP_DST;
-			break;
-		case SCTP_V6_FLOW:
-		case AH_ESP_V6_FLOW:
-		case IPV6_FLOW:
-			if (v & FW_RSS_VI_CONFIG_CMD_IP6TWOTUPEN_F)
-				info->data = RXH_IP_SRC | RXH_IP_DST;
-			break;
-		}
-		return 0;
-	}
 	case ETHTOOL_GRXRINGS:
 		info->data = pi->nqsets;
 		return 0;
+	case ETHTOOL_GRXCLSRLCNT:
+		info->rule_cnt =
+		       adap->ethtool_filters->port[pi->port_id].in_use;
+		return 0;
+	case ETHTOOL_GRXCLSRULE:
+		return cxgb4_ntuple_get_filter(dev, info, info->fs.location);
+	case ETHTOOL_GRXCLSRLALL:
+		info->data = adap->ethtool_filters->nentries;
+		while (count < info->rule_cnt) {
+			ret = cxgb4_ntuple_get_filter(dev, info, index);
+			if (!ret)
+				rules[count++] = index;
+			index++;
+		}
+		return 0;
 	}
+
 	return -EOPNOTSUPP;
 }
 
+static int cxgb4_ntuple_del_filter(struct net_device *dev,
+				   struct ethtool_rxnfc *cmd)
+{
+	struct cxgb4_ethtool_filter_info *filter_info;
+	struct adapter *adapter = netdev2adap(dev);
+	struct port_info *pi = netdev_priv(dev);
+	struct filter_entry *f;
+	u32 filter_id;
+	int ret;
+
+	if (!(adapter->flags & CXGB4_FULL_INIT_DONE))
+		return -EAGAIN;  /* can still change nfilters */
+
+	if (!adapter->ethtool_filters)
+		return -EOPNOTSUPP;
+
+	if (cmd->fs.location >= adapter->ethtool_filters->nentries) {
+		dev_err(adapter->pdev_dev,
+			"Location must be < %u",
+			adapter->ethtool_filters->nentries);
+		return -ERANGE;
+	}
+
+	filter_info = &adapter->ethtool_filters->port[pi->port_id];
+
+	if (!test_bit(cmd->fs.location, filter_info->bmap))
+		return -ENOENT;
+
+	filter_id = filter_info->loc_array[cmd->fs.location];
+	f = cxgb4_get_filter_entry(adapter, filter_id);
+
+	if (f->fs.prio)
+		filter_id -= adapter->tids.hpftid_base;
+	else if (!f->fs.hash)
+		filter_id -= (adapter->tids.ftid_base - adapter->tids.nhpftids);
+
+	ret = cxgb4_flow_rule_destroy(dev, f->fs.tc_prio, &f->fs, filter_id);
+	if (ret)
+		goto err;
+
+	clear_bit(cmd->fs.location, filter_info->bmap);
+	filter_info->in_use--;
+
+err:
+	return ret;
+}
+
+/* Add Ethtool n-tuple filters. */
+static int cxgb4_ntuple_set_filter(struct net_device *netdev,
+				   struct ethtool_rxnfc *cmd)
+{
+	struct ethtool_rx_flow_spec_input input = {};
+	struct cxgb4_ethtool_filter_info *filter_info;
+	struct adapter *adapter = netdev2adap(netdev);
+	struct port_info *pi = netdev_priv(netdev);
+	struct ch_filter_specification fs;
+	struct ethtool_rx_flow_rule *flow;
+	u32 tid;
+	int ret;
+
+	if (!(adapter->flags & CXGB4_FULL_INIT_DONE))
+		return -EAGAIN;  /* can still change nfilters */
+
+	if (!adapter->ethtool_filters)
+		return -EOPNOTSUPP;
+
+	if (cmd->fs.location >= adapter->ethtool_filters->nentries) {
+		dev_err(adapter->pdev_dev,
+			"Location must be < %u",
+			adapter->ethtool_filters->nentries);
+		return -ERANGE;
+	}
+
+	if (test_bit(cmd->fs.location,
+		     adapter->ethtool_filters->port[pi->port_id].bmap))
+		return -EEXIST;
+
+	memset(&fs, 0, sizeof(fs));
+
+	input.fs = &cmd->fs;
+	flow = ethtool_rx_flow_rule_create(&input);
+	if (IS_ERR(flow)) {
+		ret = PTR_ERR(flow);
+		goto exit;
+	}
+
+	fs.hitcnts = 1;
+
+	ret = cxgb4_flow_rule_replace(netdev, flow->rule, cmd->fs.location,
+				      NULL, &fs, &tid);
+	if (ret)
+		goto free;
+
+	filter_info = &adapter->ethtool_filters->port[pi->port_id];
+
+	if (fs.prio)
+		tid += adapter->tids.hpftid_base;
+	else if (!fs.hash)
+		tid += (adapter->tids.ftid_base - adapter->tids.nhpftids);
+
+	filter_info->loc_array[cmd->fs.location] = tid;
+	set_bit(cmd->fs.location, filter_info->bmap);
+	filter_info->in_use++;
+
+free:
+	ethtool_rx_flow_rule_destroy(flow);
+exit:
+	return ret;
+}
+
+static int set_rxnfc(struct net_device *dev, struct ethtool_rxnfc *cmd)
+{
+	int ret = -EOPNOTSUPP;
+
+	switch (cmd->cmd) {
+	case ETHTOOL_SRXCLSRLINS:
+		ret = cxgb4_ntuple_set_filter(dev, cmd);
+		break;
+	case ETHTOOL_SRXCLSRLDEL:
+		ret = cxgb4_ntuple_del_filter(dev, cmd);
+		break;
+	default:
+		break;
+	}
+
+	return ret;
+}
+
+static int set_dump(struct net_device *dev, struct ethtool_dump *eth_dump)
+{
+	struct adapter *adapter = netdev2adap(dev);
+	u32 len = 0;
+
+	len = sizeof(struct cudbg_hdr) +
+	      sizeof(struct cudbg_entity_hdr) * CUDBG_MAX_ENTITY;
+	len += cxgb4_get_dump_length(adapter, eth_dump->flag);
+
+	adapter->eth_dump.flag = eth_dump->flag;
+	adapter->eth_dump.len = len;
+	return 0;
+}
+
+static int get_dump_flag(struct net_device *dev, struct ethtool_dump *eth_dump)
+{
+	struct adapter *adapter = netdev2adap(dev);
+
+	eth_dump->flag = adapter->eth_dump.flag;
+	eth_dump->len = adapter->eth_dump.len;
+	eth_dump->version = adapter->eth_dump.version;
+	return 0;
+}
+
+static int get_dump_data(struct net_device *dev, struct ethtool_dump *eth_dump,
+			 void *buf)
+{
+	struct adapter *adapter = netdev2adap(dev);
+	u32 len = 0;
+	int ret = 0;
+
+	if (adapter->eth_dump.flag == CXGB4_ETH_DUMP_NONE)
+		return -ENOENT;
+
+	len = sizeof(struct cudbg_hdr) +
+	      sizeof(struct cudbg_entity_hdr) * CUDBG_MAX_ENTITY;
+	len += cxgb4_get_dump_length(adapter, adapter->eth_dump.flag);
+	if (eth_dump->len < len)
+		return -ENOMEM;
+
+	ret = cxgb4_cudbg_collect(adapter, buf, &len, adapter->eth_dump.flag);
+	if (ret)
+		return ret;
+
+	eth_dump->flag = adapter->eth_dump.flag;
+	eth_dump->len = len;
+	eth_dump->version = adapter->eth_dump.version;
+	return 0;
+}
+
+static bool cxgb4_fw_mod_type_info_available(unsigned int fw_mod_type)
+{
+	/* Read port module EEPROM as long as it is plugged-in and
+	 * safe to read.
+	 */
+	return (fw_mod_type != FW_PORT_MOD_TYPE_NONE &&
+		fw_mod_type != FW_PORT_MOD_TYPE_ERROR);
+}
+
+static int cxgb4_get_module_info(struct net_device *dev,
+				 struct ethtool_modinfo *modinfo)
+{
+	struct port_info *pi = netdev_priv(dev);
+	u8 sff8472_comp, sff_diag_type, sff_rev;
+	struct adapter *adapter = pi->adapter;
+	int ret;
+
+	if (!cxgb4_fw_mod_type_info_available(pi->mod_type))
+		return -EINVAL;
+
+	switch (pi->port_type) {
+	case FW_PORT_TYPE_SFP:
+	case FW_PORT_TYPE_QSA:
+	case FW_PORT_TYPE_SFP28:
+		ret = t4_i2c_rd(adapter, adapter->mbox, pi->tx_chan,
+				I2C_DEV_ADDR_A0, SFF_8472_COMP_ADDR,
+				SFF_8472_COMP_LEN, &sff8472_comp);
+		if (ret)
+			return ret;
+		ret = t4_i2c_rd(adapter, adapter->mbox, pi->tx_chan,
+				I2C_DEV_ADDR_A0, SFP_DIAG_TYPE_ADDR,
+				SFP_DIAG_TYPE_LEN, &sff_diag_type);
+		if (ret)
+			return ret;
+
+		if (!sff8472_comp || (sff_diag_type & SFP_DIAG_ADDRMODE)) {
+			modinfo->type = ETH_MODULE_SFF_8079;
+			modinfo->eeprom_len = ETH_MODULE_SFF_8079_LEN;
+		} else {
+			modinfo->type = ETH_MODULE_SFF_8472;
+			if (sff_diag_type & SFP_DIAG_IMPLEMENTED)
+				modinfo->eeprom_len = ETH_MODULE_SFF_8472_LEN;
+			else
+				modinfo->eeprom_len = ETH_MODULE_SFF_8472_LEN / 2;
+		}
+		break;
+
+	case FW_PORT_TYPE_QSFP:
+	case FW_PORT_TYPE_QSFP_10G:
+	case FW_PORT_TYPE_CR_QSFP:
+	case FW_PORT_TYPE_CR2_QSFP:
+	case FW_PORT_TYPE_CR4_QSFP:
+		ret = t4_i2c_rd(adapter, adapter->mbox, pi->tx_chan,
+				I2C_DEV_ADDR_A0, SFF_REV_ADDR,
+				SFF_REV_LEN, &sff_rev);
+		/* For QSFP type ports, revision value >= 3
+		 * means the SFP is 8636 compliant.
+		 */
+		if (ret)
+			return ret;
+		if (sff_rev >= 0x3) {
+			modinfo->type = ETH_MODULE_SFF_8636;
+			modinfo->eeprom_len = ETH_MODULE_SFF_8636_LEN;
+		} else {
+			modinfo->type = ETH_MODULE_SFF_8436;
+			modinfo->eeprom_len = ETH_MODULE_SFF_8436_LEN;
+		}
+		break;
+
+	default:
+		return -EINVAL;
+	}
+
+	return 0;
+}
+
+static int cxgb4_get_module_eeprom(struct net_device *dev,
+				   struct ethtool_eeprom *eprom, u8 *data)
+{
+	int ret = 0, offset = eprom->offset, len = eprom->len;
+	struct port_info *pi = netdev_priv(dev);
+	struct adapter *adapter = pi->adapter;
+
+	memset(data, 0, eprom->len);
+	if (offset + len <= I2C_PAGE_SIZE)
+		return t4_i2c_rd(adapter, adapter->mbox, pi->tx_chan,
+				 I2C_DEV_ADDR_A0, offset, len, data);
+
+	/* offset + len spans 0xa0 and 0xa1 pages */
+	if (offset <= I2C_PAGE_SIZE) {
+		/* read 0xa0 page */
+		len = I2C_PAGE_SIZE - offset;
+		ret =  t4_i2c_rd(adapter, adapter->mbox, pi->tx_chan,
+				 I2C_DEV_ADDR_A0, offset, len, data);
+		if (ret)
+			return ret;
+		offset = I2C_PAGE_SIZE;
+		/* Remaining bytes to be read from second page =
+		 * Total length - bytes read from first page
+		 */
+		len = eprom->len - len;
+	}
+	/* Read additional optical diagnostics from page 0xa2 if supported */
+	return t4_i2c_rd(adapter, adapter->mbox, pi->tx_chan, I2C_DEV_ADDR_A2,
+			 offset, len, &data[eprom->len - len]);
+}
+
+static u32 cxgb4_get_priv_flags(struct net_device *netdev)
+{
+	struct port_info *pi = netdev_priv(netdev);
+	struct adapter *adapter = pi->adapter;
+
+	return (adapter->eth_flags | pi->eth_flags);
+}
+
+/**
+ *	set_flags - set/unset specified flags if passed in new_flags
+ *	@cur_flags: pointer to current flags
+ *	@new_flags: new incoming flags
+ *	@flags: set of flags to set/unset
+ */
+static inline void set_flags(u32 *cur_flags, u32 new_flags, u32 flags)
+{
+	*cur_flags = (*cur_flags & ~flags) | (new_flags & flags);
+}
+
+static int cxgb4_set_priv_flags(struct net_device *netdev, u32 flags)
+{
+	struct port_info *pi = netdev_priv(netdev);
+	struct adapter *adapter = pi->adapter;
+
+	set_flags(&adapter->eth_flags, flags, PRIV_FLAGS_ADAP);
+	set_flags(&pi->eth_flags, flags, PRIV_FLAGS_PORT);
+
+	return 0;
+}
+
+static void cxgb4_lb_test(struct net_device *netdev, u64 *lb_status)
+{
+	int dev_state = netif_running(netdev);
+
+	if (dev_state) {
+		netif_tx_stop_all_queues(netdev);
+		netif_carrier_off(netdev);
+	}
+
+	*lb_status = cxgb4_selftest_lb_pkt(netdev);
+
+	if (dev_state) {
+		netif_tx_start_all_queues(netdev);
+		netif_carrier_on(netdev);
+	}
+}
+
+static void cxgb4_self_test(struct net_device *netdev,
+			    struct ethtool_test *eth_test, u64 *data)
+{
+	struct port_info *pi = netdev_priv(netdev);
+	struct adapter *adap = pi->adapter;
+
+	memset(data, 0, sizeof(u64) * CXGB4_ETHTOOL_MAX_TEST);
+
+	if (!(adap->flags & CXGB4_FULL_INIT_DONE) ||
+	    !(adap->flags & CXGB4_FW_OK)) {
+		eth_test->flags |= ETH_TEST_FL_FAILED;
+		return;
+	}
+
+	if (eth_test->flags & ETH_TEST_FL_OFFLINE)
+		cxgb4_lb_test(netdev, &data[CXGB4_ETHTOOL_LB_TEST]);
+
+	if (data[CXGB4_ETHTOOL_LB_TEST])
+		eth_test->flags |= ETH_TEST_FL_FAILED;
+}
+
 static const struct ethtool_ops cxgb_ethtool_ops = {
+	.supported_coalesce_params = ETHTOOL_COALESCE_USECS |
+				     ETHTOOL_COALESCE_RX_MAX_FRAMES |
+				     ETHTOOL_COALESCE_TX_USECS_IRQ |
+				     ETHTOOL_COALESCE_USE_ADAPTIVE_RX,
 	.get_link_ksettings = get_link_ksettings,
 	.set_link_ksettings = set_link_ksettings,
+	.get_fecparam      = get_fecparam,
+	.set_fecparam      = set_fecparam,
 	.get_drvinfo       = get_drvinfo,
 	.get_msglevel      = get_msglevel,
 	.set_msglevel      = set_msglevel,
@@ -1259,13 +2199,103 @@ static const struct ethtool_ops cxgb_ethtool_ops = {
 	.get_regs_len      = get_regs_len,
 	.get_regs          = get_regs,
 	.get_rxnfc         = get_rxnfc,
+	.set_rxnfc         = set_rxnfc,
 	.get_rxfh_indir_size = get_rss_table_size,
 	.get_rxfh	   = get_rss_table,
 	.set_rxfh	   = set_rss_table,
+	.get_rxfh_fields   = cxgb4_get_rxfh_fields,
+	.self_test	   = cxgb4_self_test,
 	.flash_device      = set_flash,
-	.get_ts_info       = get_ts_info
+	.get_ts_info       = get_ts_info,
+	.set_dump          = set_dump,
+	.get_dump_flag     = get_dump_flag,
+	.get_dump_data     = get_dump_data,
+	.get_module_info   = cxgb4_get_module_info,
+	.get_module_eeprom = cxgb4_get_module_eeprom,
+	.get_priv_flags    = cxgb4_get_priv_flags,
+	.set_priv_flags    = cxgb4_set_priv_flags,
 };
 
+void cxgb4_cleanup_ethtool_filters(struct adapter *adap)
+{
+	struct cxgb4_ethtool_filter_info *eth_filter_info;
+	u8 i;
+
+	if (!adap->ethtool_filters)
+		return;
+
+	eth_filter_info = adap->ethtool_filters->port;
+
+	if (eth_filter_info) {
+		for (i = 0; i < adap->params.nports; i++) {
+			kvfree(eth_filter_info[i].loc_array);
+			bitmap_free(eth_filter_info[i].bmap);
+		}
+		kfree(eth_filter_info);
+	}
+
+	kfree(adap->ethtool_filters);
+}
+
+int cxgb4_init_ethtool_filters(struct adapter *adap)
+{
+	struct cxgb4_ethtool_filter_info *eth_filter_info;
+	struct cxgb4_ethtool_filter *eth_filter;
+	struct tid_info *tids = &adap->tids;
+	u32 nentries, i;
+	int ret;
+
+	eth_filter = kzalloc(sizeof(*eth_filter), GFP_KERNEL);
+	if (!eth_filter)
+		return -ENOMEM;
+
+	eth_filter_info = kcalloc(adap->params.nports,
+				  sizeof(*eth_filter_info),
+				  GFP_KERNEL);
+	if (!eth_filter_info) {
+		ret = -ENOMEM;
+		goto free_eth_filter;
+	}
+
+	eth_filter->port = eth_filter_info;
+
+	nentries = tids->nhpftids + tids->nftids;
+	if (is_hashfilter(adap))
+		nentries += tids->nhash +
+			    (adap->tids.stid_base - adap->tids.tid_base);
+	eth_filter->nentries = nentries;
+
+	for (i = 0; i < adap->params.nports; i++) {
+		eth_filter->port[i].loc_array = kvzalloc(nentries, GFP_KERNEL);
+		if (!eth_filter->port[i].loc_array) {
+			ret = -ENOMEM;
+			goto free_eth_finfo;
+		}
+
+		eth_filter->port[i].bmap = bitmap_zalloc(nentries, GFP_KERNEL);
+		if (!eth_filter->port[i].bmap) {
+			ret = -ENOMEM;
+			kvfree(eth_filter->port[i].loc_array);
+			goto free_eth_finfo;
+		}
+	}
+
+	adap->ethtool_filters = eth_filter;
+	return 0;
+
+free_eth_finfo:
+	while (i-- > 0) {
+		bitmap_free(eth_filter->port[i].bmap);
+		kvfree(eth_filter->port[i].loc_array);
+	}
+	kfree(eth_filter_info);
+
+free_eth_filter:
+	kfree(eth_filter);
+
+	return ret;
+}
+
 void cxgb4_set_ethtool_ops(struct net_device *netdev)
 {
 	netdev->ethtool_ops = &cxgb_ethtool_ops;
diff --git a/drivers/net/ethernet/chelsio/cxgb4/cxgb4_fcoe.c b/drivers/net/ethernet/chelsio/cxgb4/cxgb4_fcoe.c
index 6c8a62eefe51..f6f745f5c022 100644
--- a/drivers/net/ethernet/chelsio/cxgb4/cxgb4_fcoe.c
+++ b/drivers/net/ethernet/chelsio/cxgb4/cxgb4_fcoe.c
@@ -74,15 +74,14 @@ int cxgb_fcoe_enable(struct net_device *netdev)
 	if (is_t4(adap->params.chip))
 		return -EINVAL;
 
-	if (!(adap->flags & FULL_INIT_DONE))
+	if (!(adap->flags & CXGB4_FULL_INIT_DONE))
 		return -EINVAL;
 
 	dev_info(adap->pdev_dev, "Enabling FCoE offload features\n");
 
 	netdev->features |= NETIF_F_FCOE_CRC;
 	netdev->vlan_features |= NETIF_F_FCOE_CRC;
-	netdev->features |= NETIF_F_FCOE_MTU;
-	netdev->vlan_features |= NETIF_F_FCOE_MTU;
+	netdev->fcoe_mtu = true;
 
 	netdev_features_change(netdev);
 
@@ -112,8 +111,7 @@ int cxgb_fcoe_disable(struct net_device *netdev)
 
 	netdev->features &= ~NETIF_F_FCOE_CRC;
 	netdev->vlan_features &= ~NETIF_F_FCOE_CRC;
-	netdev->features &= ~NETIF_F_FCOE_MTU;
-	netdev->vlan_features &= ~NETIF_F_FCOE_MTU;
+	netdev->fcoe_mtu = false;
 
 	netdev_features_change(netdev);
 
diff --git a/drivers/net/ethernet/chelsio/cxgb4/cxgb4_filter.c b/drivers/net/ethernet/chelsio/cxgb4/cxgb4_filter.c
index 10736738ff30..dd9e68465e69 100644
--- a/drivers/net/ethernet/chelsio/cxgb4/cxgb4_filter.c
+++ b/drivers/net/ethernet/chelsio/cxgb4/cxgb4_filter.c
@@ -31,10 +31,15 @@
  * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
  * SOFTWARE.
  */
+#include <net/ipv6.h>
 
 #include "cxgb4.h"
 #include "t4_regs.h"
+#include "t4_tcb.h"
+#include "t4_values.h"
+#include "clip_tbl.h"
 #include "l2t.h"
+#include "smt.h"
 #include "t4fw_api.h"
 #include "cxgb4_filter.h"
 
@@ -48,6 +53,197 @@ static inline bool unsupported(u32 conf, u32 conf_mask, u32 val, u32 mask)
 	return !(conf & conf_mask) && is_field_set(val, mask);
 }
 
+static int set_tcb_field(struct adapter *adap, struct filter_entry *f,
+			 unsigned int ftid,  u16 word, u64 mask, u64 val,
+			 int no_reply)
+{
+	struct cpl_set_tcb_field *req;
+	struct sk_buff *skb;
+
+	skb = alloc_skb(sizeof(struct cpl_set_tcb_field), GFP_ATOMIC);
+	if (!skb)
+		return -ENOMEM;
+
+	req = (struct cpl_set_tcb_field *)__skb_put_zero(skb, sizeof(*req));
+	INIT_TP_WR_CPL(req, CPL_SET_TCB_FIELD, ftid);
+	req->reply_ctrl = htons(REPLY_CHAN_V(0) |
+				QUEUENO_V(adap->sge.fw_evtq.abs_id) |
+				NO_REPLY_V(no_reply));
+	req->word_cookie = htons(TCB_WORD_V(word) | TCB_COOKIE_V(ftid));
+	req->mask = cpu_to_be64(mask);
+	req->val = cpu_to_be64(val);
+	set_wr_txq(skb, CPL_PRIORITY_CONTROL, f->fs.val.iport & 0x3);
+	t4_ofld_send(adap, skb);
+	return 0;
+}
+
+/* Set one of the t_flags bits in the TCB.
+ */
+static int set_tcb_tflag(struct adapter *adap, struct filter_entry *f,
+			 unsigned int ftid, unsigned int bit_pos,
+			 unsigned int val, int no_reply)
+{
+	return set_tcb_field(adap, f, ftid,  TCB_T_FLAGS_W, 1ULL << bit_pos,
+			     (unsigned long long)val << bit_pos, no_reply);
+}
+
+static void mk_abort_req_ulp(struct cpl_abort_req *abort_req, unsigned int tid)
+{
+	struct ulp_txpkt *txpkt = (struct ulp_txpkt *)abort_req;
+	struct ulptx_idata *sc = (struct ulptx_idata *)(txpkt + 1);
+
+	txpkt->cmd_dest = htonl(ULPTX_CMD_V(ULP_TX_PKT) | ULP_TXPKT_DEST_V(0));
+	txpkt->len = htonl(DIV_ROUND_UP(sizeof(*abort_req), 16));
+	sc->cmd_more = htonl(ULPTX_CMD_V(ULP_TX_SC_IMM));
+	sc->len = htonl(sizeof(*abort_req) - sizeof(struct work_request_hdr));
+	OPCODE_TID(abort_req) = htonl(MK_OPCODE_TID(CPL_ABORT_REQ, tid));
+	abort_req->rsvd0 = htonl(0);
+	abort_req->rsvd1 = 0;
+	abort_req->cmd = CPL_ABORT_NO_RST;
+}
+
+static void mk_abort_rpl_ulp(struct cpl_abort_rpl *abort_rpl, unsigned int tid)
+{
+	struct ulp_txpkt *txpkt = (struct ulp_txpkt *)abort_rpl;
+	struct ulptx_idata *sc = (struct ulptx_idata *)(txpkt + 1);
+
+	txpkt->cmd_dest = htonl(ULPTX_CMD_V(ULP_TX_PKT) | ULP_TXPKT_DEST_V(0));
+	txpkt->len = htonl(DIV_ROUND_UP(sizeof(*abort_rpl), 16));
+	sc->cmd_more = htonl(ULPTX_CMD_V(ULP_TX_SC_IMM));
+	sc->len = htonl(sizeof(*abort_rpl) - sizeof(struct work_request_hdr));
+	OPCODE_TID(abort_rpl) = htonl(MK_OPCODE_TID(CPL_ABORT_RPL, tid));
+	abort_rpl->rsvd0 = htonl(0);
+	abort_rpl->rsvd1 = 0;
+	abort_rpl->cmd = CPL_ABORT_NO_RST;
+}
+
+static void mk_set_tcb_ulp(struct filter_entry *f,
+			   struct cpl_set_tcb_field *req,
+			   unsigned int word, u64 mask, u64 val,
+			   u8 cookie, int no_reply)
+{
+	struct ulp_txpkt *txpkt = (struct ulp_txpkt *)req;
+	struct ulptx_idata *sc = (struct ulptx_idata *)(txpkt + 1);
+
+	txpkt->cmd_dest = htonl(ULPTX_CMD_V(ULP_TX_PKT) | ULP_TXPKT_DEST_V(0));
+	txpkt->len = htonl(DIV_ROUND_UP(sizeof(*req), 16));
+	sc->cmd_more = htonl(ULPTX_CMD_V(ULP_TX_SC_IMM));
+	sc->len = htonl(sizeof(*req) - sizeof(struct work_request_hdr));
+	OPCODE_TID(req) = htonl(MK_OPCODE_TID(CPL_SET_TCB_FIELD, f->tid));
+	req->reply_ctrl = htons(NO_REPLY_V(no_reply) | REPLY_CHAN_V(0) |
+				QUEUENO_V(0));
+	req->word_cookie = htons(TCB_WORD_V(word) | TCB_COOKIE_V(cookie));
+	req->mask = cpu_to_be64(mask);
+	req->val = cpu_to_be64(val);
+	sc = (struct ulptx_idata *)(req + 1);
+	sc->cmd_more = htonl(ULPTX_CMD_V(ULP_TX_SC_NOOP));
+	sc->len = htonl(0);
+}
+
+static int configure_filter_smac(struct adapter *adap, struct filter_entry *f)
+{
+	int err;
+
+	/* do a set-tcb for smac-sel and CWR bit.. */
+	err = set_tcb_field(adap, f, f->tid, TCB_SMAC_SEL_W,
+			    TCB_SMAC_SEL_V(TCB_SMAC_SEL_M),
+			    TCB_SMAC_SEL_V(f->smt->idx), 1);
+	if (err)
+		goto smac_err;
+
+	err = set_tcb_tflag(adap, f, f->tid, TF_CCTRL_CWR_S, 1, 1);
+	if (!err)
+		return 0;
+
+smac_err:
+	dev_err(adap->pdev_dev, "filter %u smac config failed with error %u\n",
+		f->tid, err);
+	return err;
+}
+
+static void set_nat_params(struct adapter *adap, struct filter_entry *f,
+			   unsigned int tid, bool dip, bool sip, bool dp,
+			   bool sp)
+{
+	u8 *nat_lp = (u8 *)&f->fs.nat_lport;
+	u8 *nat_fp = (u8 *)&f->fs.nat_fport;
+
+	if (dip) {
+		if (f->fs.type) {
+			set_tcb_field(adap, f, tid, TCB_SND_UNA_RAW_W,
+				      WORD_MASK, f->fs.nat_lip[15] |
+				      f->fs.nat_lip[14] << 8 |
+				      f->fs.nat_lip[13] << 16 |
+				      (u64)f->fs.nat_lip[12] << 24, 1);
+
+			set_tcb_field(adap, f, tid, TCB_SND_UNA_RAW_W + 1,
+				      WORD_MASK, f->fs.nat_lip[11] |
+				      f->fs.nat_lip[10] << 8 |
+				      f->fs.nat_lip[9] << 16 |
+				      (u64)f->fs.nat_lip[8] << 24, 1);
+
+			set_tcb_field(adap, f, tid, TCB_SND_UNA_RAW_W + 2,
+				      WORD_MASK, f->fs.nat_lip[7] |
+				      f->fs.nat_lip[6] << 8 |
+				      f->fs.nat_lip[5] << 16 |
+				      (u64)f->fs.nat_lip[4] << 24, 1);
+
+			set_tcb_field(adap, f, tid, TCB_SND_UNA_RAW_W + 3,
+				      WORD_MASK, f->fs.nat_lip[3] |
+				      f->fs.nat_lip[2] << 8 |
+				      f->fs.nat_lip[1] << 16 |
+				      (u64)f->fs.nat_lip[0] << 24, 1);
+		} else {
+			set_tcb_field(adap, f, tid, TCB_RX_FRAG3_LEN_RAW_W,
+				      WORD_MASK, f->fs.nat_lip[3] |
+				      f->fs.nat_lip[2] << 8 |
+				      f->fs.nat_lip[1] << 16 |
+				      (u64)f->fs.nat_lip[0] << 24, 1);
+		}
+	}
+
+	if (sip) {
+		if (f->fs.type) {
+			set_tcb_field(adap, f, tid, TCB_RX_FRAG2_PTR_RAW_W,
+				      WORD_MASK, f->fs.nat_fip[15] |
+				      f->fs.nat_fip[14] << 8 |
+				      f->fs.nat_fip[13] << 16 |
+				      (u64)f->fs.nat_fip[12] << 24, 1);
+
+			set_tcb_field(adap, f, tid, TCB_RX_FRAG2_PTR_RAW_W + 1,
+				      WORD_MASK, f->fs.nat_fip[11] |
+				      f->fs.nat_fip[10] << 8 |
+				      f->fs.nat_fip[9] << 16 |
+				      (u64)f->fs.nat_fip[8] << 24, 1);
+
+			set_tcb_field(adap, f, tid, TCB_RX_FRAG2_PTR_RAW_W + 2,
+				      WORD_MASK, f->fs.nat_fip[7] |
+				      f->fs.nat_fip[6] << 8 |
+				      f->fs.nat_fip[5] << 16 |
+				      (u64)f->fs.nat_fip[4] << 24, 1);
+
+			set_tcb_field(adap, f, tid, TCB_RX_FRAG2_PTR_RAW_W + 3,
+				      WORD_MASK, f->fs.nat_fip[3] |
+				      f->fs.nat_fip[2] << 8 |
+				      f->fs.nat_fip[1] << 16 |
+				      (u64)f->fs.nat_fip[0] << 24, 1);
+
+		} else {
+			set_tcb_field(adap, f, tid,
+				      TCB_RX_FRAG3_START_IDX_OFFSET_RAW_W,
+				      WORD_MASK, f->fs.nat_fip[3] |
+				      f->fs.nat_fip[2] << 8 |
+				      f->fs.nat_fip[1] << 16 |
+				      (u64)f->fs.nat_fip[0] << 24, 1);
+		}
+	}
+
+	set_tcb_field(adap, f, tid, TCB_PDU_HDR_LEN_W, WORD_MASK,
+		      (dp ? (nat_lp[1] | nat_lp[0] << 8) : 0) |
+		      (sp ? (nat_fp[1] << 16 | (u64)nat_fp[0] << 24) : 0),
+		      1);
+}
+
 /* Validate filter spec against configuration done on the card. */
 static int validate_filter(struct net_device *dev,
 			   struct ch_filter_specification *fs)
@@ -56,8 +252,9 @@ static int validate_filter(struct net_device *dev,
 	u32 fconf, iconf;
 
 	/* Check for unconfigured fields being used. */
-	fconf = adapter->params.tp.vlan_pri_map;
 	iconf = adapter->params.tp.ingress_config;
+	fconf = fs->hash ? adapter->params.tp.filter_mask :
+			   adapter->params.tp.vlan_pri_map;
 
 	if (unsupported(fconf, FCOE_F, fs->val.fcoe, fs->mask.fcoe) ||
 	    unsupported(fconf, PORT_F, fs->val.iport, fs->mask.iport) ||
@@ -73,6 +270,8 @@ static int validate_filter(struct net_device *dev,
 			fs->mask.pfvf_vld) ||
 	    unsupported(fconf, VNIC_ID_F, fs->val.ovlan_vld,
 			fs->mask.ovlan_vld) ||
+	    unsupported(fconf, VNIC_ID_F, fs->val.encap_vld,
+			fs->mask.encap_vld) ||
 	    unsupported(fconf, VLAN_F, fs->val.ivlan_vld, fs->mask.ivlan_vld))
 		return -EOPNOTSUPP;
 
@@ -83,8 +282,12 @@ static int validate_filter(struct net_device *dev,
 	 * carries that overlap, we need to translate any PF/VF
 	 * specification into that internal format below.
 	 */
-	if (is_field_set(fs->val.pfvf_vld, fs->mask.pfvf_vld) &&
-	    is_field_set(fs->val.ovlan_vld, fs->mask.ovlan_vld))
+	if ((is_field_set(fs->val.pfvf_vld, fs->mask.pfvf_vld) &&
+	     is_field_set(fs->val.ovlan_vld, fs->mask.ovlan_vld)) ||
+	    (is_field_set(fs->val.pfvf_vld, fs->mask.pfvf_vld) &&
+	     is_field_set(fs->val.encap_vld, fs->mask.encap_vld)) ||
+	    (is_field_set(fs->val.ovlan_vld, fs->mask.ovlan_vld) &&
+	     is_field_set(fs->val.encap_vld, fs->mask.encap_vld)))
 		return -EOPNOTSUPP;
 	if (unsupported(iconf, VNIC_F, fs->val.pfvf_vld, fs->mask.pfvf_vld) ||
 	    (is_field_set(fs->val.ovlan_vld, fs->mask.ovlan_vld) &&
@@ -114,6 +317,9 @@ static int validate_filter(struct net_device *dev,
 	     fs->newvlan == VLAN_REWRITE))
 		return -EOPNOTSUPP;
 
+	if (fs->val.encap_vld &&
+	    CHELSIO_CHIP_VERSION(adapter->params.chip) < CHELSIO_T6)
+		return -EOPNOTSUPP;
 	return 0;
 }
 
@@ -148,7 +354,340 @@ static int get_filter_steerq(struct net_device *dev,
 	return iq;
 }
 
-static int cxgb4_set_ftid(struct tid_info *t, int fidx, int family)
+static int get_filter_count(struct adapter *adapter, unsigned int fidx,
+			    u64 *pkts, u64 *bytes, bool hash)
+{
+	unsigned int tcb_base, tcbaddr;
+	unsigned int word_offset;
+	struct filter_entry *f;
+	__be64 be64_byte_count;
+	int ret;
+
+	tcb_base = t4_read_reg(adapter, TP_CMM_TCB_BASE_A);
+	if (is_hashfilter(adapter) && hash) {
+		if (tid_out_of_range(&adapter->tids, fidx))
+			return -E2BIG;
+		f = adapter->tids.tid_tab[fidx - adapter->tids.tid_base];
+		if (!f)
+			return -EINVAL;
+	} else {
+		if ((fidx != (adapter->tids.nftids + adapter->tids.nsftids +
+			      adapter->tids.nhpftids - 1)) &&
+		    fidx >= (adapter->tids.nftids + adapter->tids.nhpftids))
+			return -E2BIG;
+
+		if (fidx < adapter->tids.nhpftids)
+			f = &adapter->tids.hpftid_tab[fidx];
+		else
+			f = &adapter->tids.ftid_tab[fidx -
+						    adapter->tids.nhpftids];
+		if (!f->valid)
+			return -EINVAL;
+	}
+	tcbaddr = tcb_base + f->tid * TCB_SIZE;
+
+	spin_lock(&adapter->win0_lock);
+	if (is_t4(adapter->params.chip)) {
+		__be64 be64_count;
+
+		/* T4 doesn't maintain byte counts in hw */
+		*bytes = 0;
+
+		/* Get pkts */
+		word_offset = 4;
+		ret = t4_memory_rw(adapter, MEMWIN_NIC, MEM_EDC0,
+				   tcbaddr + (word_offset * sizeof(__be32)),
+				   sizeof(be64_count),
+				   (__be32 *)&be64_count,
+				   T4_MEMORY_READ);
+		if (ret < 0)
+			goto out;
+		*pkts = be64_to_cpu(be64_count);
+	} else {
+		__be32 be32_count;
+
+		/* Get bytes */
+		word_offset = 4;
+		ret = t4_memory_rw(adapter, MEMWIN_NIC, MEM_EDC0,
+				   tcbaddr + (word_offset * sizeof(__be32)),
+				   sizeof(be64_byte_count),
+				   &be64_byte_count,
+				   T4_MEMORY_READ);
+		if (ret < 0)
+			goto out;
+		*bytes = be64_to_cpu(be64_byte_count);
+
+		/* Get pkts */
+		word_offset = 6;
+		ret = t4_memory_rw(adapter, MEMWIN_NIC, MEM_EDC0,
+				   tcbaddr + (word_offset * sizeof(__be32)),
+				   sizeof(be32_count),
+				   &be32_count,
+				   T4_MEMORY_READ);
+		if (ret < 0)
+			goto out;
+		*pkts = (u64)be32_to_cpu(be32_count);
+	}
+
+out:
+	spin_unlock(&adapter->win0_lock);
+	return ret;
+}
+
+int cxgb4_get_filter_counters(struct net_device *dev, unsigned int fidx,
+			      u64 *hitcnt, u64 *bytecnt, bool hash)
+{
+	struct adapter *adapter = netdev2adap(dev);
+
+	return get_filter_count(adapter, fidx, hitcnt, bytecnt, hash);
+}
+
+static bool cxgb4_filter_prio_in_range(struct tid_info *t, u32 idx, u8 nslots,
+				       u32 prio)
+{
+	struct filter_entry *prev_tab, *next_tab, *prev_fe, *next_fe;
+	u32 prev_ftid, next_ftid;
+
+	/* Only insert the rule if both of the following conditions
+	 * are met:
+	 * 1. The immediate previous rule has priority <= @prio.
+	 * 2. The immediate next rule has priority >= @prio.
+	 */
+
+	/* High Priority (HPFILTER) region always has higher priority
+	 * than normal FILTER region. So, all rules in HPFILTER region
+	 * must have prio value <= rules in normal FILTER region.
+	 */
+	if (idx < t->nhpftids) {
+		/* Don't insert if there's a rule already present at @idx
+		 * in HPFILTER region.
+		 */
+		if (test_bit(idx, t->hpftid_bmap))
+			return false;
+
+		next_tab = t->hpftid_tab;
+		next_ftid = find_next_bit(t->hpftid_bmap, t->nhpftids, idx);
+		if (next_ftid >= t->nhpftids) {
+			/* No next entry found in HPFILTER region.
+			 * See if there's any next entry in normal
+			 * FILTER region.
+			 */
+			next_ftid = find_first_bit(t->ftid_bmap, t->nftids);
+			if (next_ftid >= t->nftids)
+				next_ftid = idx;
+			else
+				next_tab = t->ftid_tab;
+		}
+
+		/* Search for the closest previous filter entry in HPFILTER
+		 * region. No need to search in normal FILTER region because
+		 * there can never be any entry in normal FILTER region whose
+		 * prio value is < last entry in HPFILTER region.
+		 */
+		prev_ftid = find_last_bit(t->hpftid_bmap, idx);
+		if (prev_ftid >= idx)
+			prev_ftid = idx;
+
+		prev_tab = t->hpftid_tab;
+	} else {
+		idx -= t->nhpftids;
+
+		/* Don't insert if there's a rule already present at @idx
+		 * in normal FILTER region.
+		 */
+		if (test_bit(idx, t->ftid_bmap))
+			return false;
+
+		prev_tab = t->ftid_tab;
+		prev_ftid = find_last_bit(t->ftid_bmap, idx);
+		if (prev_ftid >= idx) {
+			/* No previous entry found in normal FILTER
+			 * region. See if there's any previous entry
+			 * in HPFILTER region.
+			 */
+			prev_ftid = find_last_bit(t->hpftid_bmap, t->nhpftids);
+			if (prev_ftid >= t->nhpftids)
+				prev_ftid = idx;
+			else
+				prev_tab = t->hpftid_tab;
+		}
+
+		/* Search for the closest next filter entry in normal
+		 * FILTER region. No need to search in HPFILTER region
+		 * because there can never be any entry in HPFILTER
+		 * region whose prio value is > first entry in normal
+		 * FILTER region.
+		 */
+		next_ftid = find_next_bit(t->ftid_bmap, t->nftids, idx);
+		if (next_ftid >= t->nftids)
+			next_ftid = idx;
+
+		next_tab = t->ftid_tab;
+	}
+
+	next_fe = &next_tab[next_ftid];
+
+	/* See if the filter entry belongs to an IPv6 rule, which
+	 * occupy 4 slots on T5 and 2 slots on T6. Adjust the
+	 * reference to the previously inserted filter entry
+	 * accordingly.
+	 */
+	prev_fe = &prev_tab[prev_ftid & ~(nslots - 1)];
+	if (!prev_fe->fs.type)
+		prev_fe = &prev_tab[prev_ftid];
+
+	if ((prev_fe->valid && prev_fe->fs.tc_prio > prio) ||
+	    (next_fe->valid && next_fe->fs.tc_prio < prio))
+		return false;
+
+	return true;
+}
+
+int cxgb4_get_free_ftid(struct net_device *dev, u8 family, bool hash_en,
+			u32 tc_prio)
+{
+	struct adapter *adap = netdev2adap(dev);
+	struct tid_info *t = &adap->tids;
+	u32 bmap_ftid, max_ftid;
+	struct filter_entry *f;
+	unsigned long *bmap;
+	bool found = false;
+	u8 i, cnt, n;
+	int ftid = 0;
+
+	/* IPv4 occupy 1 slot. IPv6 occupy 2 slots on T6 and 4 slots
+	 * on T5.
+	 */
+	n = 1;
+	if (family == PF_INET6) {
+		n++;
+		if (CHELSIO_CHIP_VERSION(adap->params.chip) < CHELSIO_T6)
+			n += 2;
+	}
+
+	/* There are 3 filter regions available in hardware in
+	 * following order of priority:
+	 *
+	 * 1. High Priority (HPFILTER) region (Highest Priority).
+	 * 2. HASH region.
+	 * 3. Normal FILTER region (Lowest Priority).
+	 *
+	 * Entries in HPFILTER and normal FILTER region have index
+	 * 0 as the highest priority and the rules will be scanned
+	 * in ascending order until either a rule hits or end of
+	 * the region is reached.
+	 *
+	 * All HASH region entries have same priority. The set of
+	 * fields to match in headers are pre-determined. The same
+	 * set of header match fields must be compulsorily specified
+	 * in all the rules wanting to get inserted in HASH region.
+	 * Hence, HASH region is an exact-match region. A HASH is
+	 * generated for a rule based on the values in the
+	 * pre-determined set of header match fields. The generated
+	 * HASH serves as an index into the HASH region. There can
+	 * never be 2 rules having the same HASH. Hardware will
+	 * compute a HASH for every incoming packet based on the
+	 * values in the pre-determined set of header match fields
+	 * and uses it as an index to check if there's a rule
+	 * inserted in the HASH region at the specified index. If
+	 * there's a rule inserted, then it's considered as a filter
+	 * hit. Otherwise, it's a filter miss and normal FILTER region
+	 * is scanned afterwards.
+	 */
+
+	spin_lock_bh(&t->ftid_lock);
+
+	ftid = (tc_prio <= t->nhpftids) ? 0 : t->nhpftids;
+	max_ftid = t->nftids + t->nhpftids;
+	while (ftid < max_ftid) {
+		if (ftid < t->nhpftids) {
+			/* If the new rule wants to get inserted into
+			 * HPFILTER region, but its prio is greater
+			 * than the rule with the highest prio in HASH
+			 * region, or if there's not enough slots
+			 * available in HPFILTER region, then skip
+			 * trying to insert this rule into HPFILTER
+			 * region and directly go to the next region.
+			 */
+			if ((t->tc_hash_tids_max_prio &&
+			     tc_prio > t->tc_hash_tids_max_prio) ||
+			     (ftid + n) > t->nhpftids) {
+				ftid = t->nhpftids;
+				continue;
+			}
+
+			bmap = t->hpftid_bmap;
+			bmap_ftid = ftid;
+		} else if (hash_en) {
+			/* Ensure priority is >= last rule in HPFILTER
+			 * region.
+			 */
+			ftid = find_last_bit(t->hpftid_bmap, t->nhpftids);
+			if (ftid < t->nhpftids) {
+				f = &t->hpftid_tab[ftid];
+				if (f->valid && tc_prio < f->fs.tc_prio)
+					break;
+			}
+
+			/* Ensure priority is <= first rule in normal
+			 * FILTER region.
+			 */
+			ftid = find_first_bit(t->ftid_bmap, t->nftids);
+			if (ftid < t->nftids) {
+				f = &t->ftid_tab[ftid];
+				if (f->valid && tc_prio > f->fs.tc_prio)
+					break;
+			}
+
+			found = true;
+			ftid = t->nhpftids;
+			goto out_unlock;
+		} else {
+			/* If the new rule wants to get inserted into
+			 * normal FILTER region, but its prio is less
+			 * than the rule with the highest prio in HASH
+			 * region, then reject the rule.
+			 */
+			if (t->tc_hash_tids_max_prio &&
+			    tc_prio < t->tc_hash_tids_max_prio)
+				break;
+
+			if (ftid + n > max_ftid)
+				break;
+
+			bmap = t->ftid_bmap;
+			bmap_ftid = ftid - t->nhpftids;
+		}
+
+		cnt = 0;
+		for (i = 0; i < n; i++) {
+			if (test_bit(bmap_ftid + i, bmap))
+				break;
+			cnt++;
+		}
+
+		if (cnt == n) {
+			/* Ensure the new rule's prio doesn't conflict
+			 * with existing rules.
+			 */
+			if (cxgb4_filter_prio_in_range(t, ftid, n,
+						       tc_prio)) {
+				ftid &= ~(n - 1);
+				found = true;
+				break;
+			}
+		}
+
+		ftid += n;
+	}
+
+out_unlock:
+	spin_unlock_bh(&t->ftid_lock);
+	return found ? ftid : -ENOMEM;
+}
+
+static int cxgb4_set_ftid(struct tid_info *t, int fidx, int family,
+			  unsigned int chip_ver)
 {
 	spin_lock_bh(&t->ftid_lock);
 
@@ -157,40 +696,84 @@ static int cxgb4_set_ftid(struct tid_info *t, int fidx, int family)
 		return -EBUSY;
 	}
 
-	if (family == PF_INET)
+	if (family == PF_INET) {
 		__set_bit(fidx, t->ftid_bmap);
-	else
-		bitmap_allocate_region(t->ftid_bmap, fidx, 2);
+	} else {
+		if (chip_ver < CHELSIO_T6)
+			bitmap_allocate_region(t->ftid_bmap, fidx, 2);
+		else
+			bitmap_allocate_region(t->ftid_bmap, fidx, 1);
+	}
 
 	spin_unlock_bh(&t->ftid_lock);
 	return 0;
 }
 
-static void cxgb4_clear_ftid(struct tid_info *t, int fidx, int family)
+static int cxgb4_set_hpftid(struct tid_info *t, int fidx, int family)
 {
 	spin_lock_bh(&t->ftid_lock);
+
+	if (test_bit(fidx, t->hpftid_bmap)) {
+		spin_unlock_bh(&t->ftid_lock);
+		return -EBUSY;
+	}
+
 	if (family == PF_INET)
+		__set_bit(fidx, t->hpftid_bmap);
+	else
+		bitmap_allocate_region(t->hpftid_bmap, fidx, 1);
+
+	spin_unlock_bh(&t->ftid_lock);
+	return 0;
+}
+
+static void cxgb4_clear_ftid(struct tid_info *t, int fidx, int family,
+			     unsigned int chip_ver)
+{
+	spin_lock_bh(&t->ftid_lock);
+	if (family == PF_INET) {
 		__clear_bit(fidx, t->ftid_bmap);
+	} else {
+		if (chip_ver < CHELSIO_T6)
+			bitmap_release_region(t->ftid_bmap, fidx, 2);
+		else
+			bitmap_release_region(t->ftid_bmap, fidx, 1);
+	}
+	spin_unlock_bh(&t->ftid_lock);
+}
+
+static void cxgb4_clear_hpftid(struct tid_info *t, int fidx, int family)
+{
+	spin_lock_bh(&t->ftid_lock);
+
+	if (family == PF_INET)
+		__clear_bit(fidx, t->hpftid_bmap);
 	else
-		bitmap_release_region(t->ftid_bmap, fidx, 2);
+		bitmap_release_region(t->hpftid_bmap, fidx, 1);
+
 	spin_unlock_bh(&t->ftid_lock);
 }
 
 /* Delete the filter at a specified index. */
 static int del_filter_wr(struct adapter *adapter, int fidx)
 {
-	struct filter_entry *f = &adapter->tids.ftid_tab[fidx];
 	struct fw_filter_wr *fwr;
+	struct filter_entry *f;
 	struct sk_buff *skb;
 	unsigned int len;
 
+	if (fidx < adapter->tids.nhpftids)
+		f = &adapter->tids.hpftid_tab[fidx];
+	else
+		f = &adapter->tids.ftid_tab[fidx - adapter->tids.nhpftids];
+
 	len = sizeof(*fwr);
 
 	skb = alloc_skb(len, GFP_KERNEL);
 	if (!skb)
 		return -ENOMEM;
 
-	fwr = (struct fw_filter_wr *)__skb_put(skb, len);
+	fwr = __skb_put(skb, len);
 	t4_mk_filtdelwr(f->tid, fwr, adapter->sge.fw_evtq.abs_id);
 
 	/* Mark the filter as "pending" and ship off the Filter Work Request.
@@ -209,10 +792,15 @@ static int del_filter_wr(struct adapter *adapter, int fidx)
  */
 int set_filter_wr(struct adapter *adapter, int fidx)
 {
-	struct filter_entry *f = &adapter->tids.ftid_tab[fidx];
-	struct fw_filter_wr *fwr;
+	struct fw_filter2_wr *fwr;
+	struct filter_entry *f;
 	struct sk_buff *skb;
 
+	if (fidx < adapter->tids.nhpftids)
+		f = &adapter->tids.hpftid_tab[fidx];
+	else
+		f = &adapter->tids.ftid_tab[fidx - adapter->tids.nhpftids];
+
 	skb = alloc_skb(sizeof(*fwr), GFP_KERNEL);
 	if (!skb)
 		return -ENOMEM;
@@ -231,8 +819,22 @@ int set_filter_wr(struct adapter *adapter, int fidx)
 		}
 	}
 
-	fwr = (struct fw_filter_wr *)__skb_put(skb, sizeof(*fwr));
-	memset(fwr, 0, sizeof(*fwr));
+	/* If the new filter requires loopback Source MAC rewriting then
+	 * we need to allocate a SMT entry for the filter.
+	 */
+	if (f->fs.newsmac) {
+		f->smt = cxgb4_smt_alloc_switching(f->dev, f->fs.smac);
+		if (!f->smt) {
+			if (f->l2t) {
+				cxgb4_l2t_release(f->l2t);
+				f->l2t = NULL;
+			}
+			kfree_skb(skb);
+			return -ENOMEM;
+		}
+	}
+
+	fwr = __skb_put_zero(skb, sizeof(*fwr));
 
 	/* It would be nice to put most of the following in t4_hw.c but most
 	 * of the work is translating the cxgbtool ch_filter_specification
@@ -242,7 +844,10 @@ int set_filter_wr(struct adapter *adapter, int fidx)
 	 * filter specification structure but for now it's easiest to simply
 	 * put this fairly direct code in line ...
 	 */
-	fwr->op_pkd = htonl(FW_WR_OP_V(FW_FILTER_WR));
+	if (adapter->params.filter2_wr_support)
+		fwr->op_pkd = htonl(FW_WR_OP_V(FW_FILTER2_WR));
+	else
+		fwr->op_pkd = htonl(FW_WR_OP_V(FW_FILTER_WR));
 	fwr->len16_pkd = htonl(FW_WR_LEN16_V(sizeof(*fwr) / 16));
 	fwr->tid_to_iq =
 		htonl(FW_FILTER_WR_TID_V(f->tid) |
@@ -275,7 +880,8 @@ int set_filter_wr(struct adapter *adapter, int fidx)
 		 FW_FILTER_WR_OVLAN_VLD_V(f->fs.val.ovlan_vld) |
 		 FW_FILTER_WR_IVLAN_VLDM_V(f->fs.mask.ivlan_vld) |
 		 FW_FILTER_WR_OVLAN_VLDM_V(f->fs.mask.ovlan_vld));
-	fwr->smac_sel = 0;
+	if (f->fs.newsmac)
+		fwr->smac_sel = f->smt->idx;
 	fwr->rx_chan_rx_rpl_iq =
 		htons(FW_FILTER_WR_RX_CHAN_V(0) |
 		      FW_FILTER_WR_RX_RPL_IQ_V(adapter->sge.fw_evtq.abs_id));
@@ -304,8 +910,21 @@ int set_filter_wr(struct adapter *adapter, int fidx)
 	fwr->lpm = htons(f->fs.mask.lport);
 	fwr->fp = htons(f->fs.val.fport);
 	fwr->fpm = htons(f->fs.mask.fport);
-	if (f->fs.newsmac)
-		memcpy(fwr->sma, f->fs.smac, sizeof(fwr->sma));
+
+	if (adapter->params.filter2_wr_support) {
+		u8 *nat_lp = (u8 *)&f->fs.nat_lport;
+		u8 *nat_fp = (u8 *)&f->fs.nat_fport;
+
+		fwr->natmode_to_ulp_type =
+			FW_FILTER2_WR_ULP_TYPE_V(f->fs.nat_mode ?
+						 ULP_MODE_TCPDDP :
+						 ULP_MODE_NONE) |
+			FW_FILTER2_WR_NATMODE_V(f->fs.nat_mode);
+		memcpy(fwr->newlip, f->fs.nat_lip, sizeof(fwr->newlip));
+		memcpy(fwr->newfip, f->fs.nat_fip, sizeof(fwr->newfip));
+		fwr->newlport = htons(nat_lp[1] | nat_lp[0] << 8);
+		fwr->newfport = htons(nat_fp[1] | nat_fp[0] << 8);
+	}
 
 	/* Mark the filter as "pending" and ship off the Filter Work Request.
 	 * When we get the Work Request Reply we'll clear the pending status.
@@ -336,10 +955,14 @@ int delete_filter(struct adapter *adapter, unsigned int fidx)
 	struct filter_entry *f;
 	int ret;
 
-	if (fidx >= adapter->tids.nftids + adapter->tids.nsftids)
+	if (fidx >= adapter->tids.nftids + adapter->tids.nsftids +
+		    adapter->tids.nhpftids)
 		return -EINVAL;
 
-	f = &adapter->tids.ftid_tab[fidx];
+	if (fidx < adapter->tids.nhpftids)
+		f = &adapter->tids.hpftid_tab[fidx];
+	else
+		f = &adapter->tids.ftid_tab[fidx - adapter->tids.nhpftids];
 	ret = writable_filter(f);
 	if (ret)
 		return ret;
@@ -354,15 +977,25 @@ int delete_filter(struct adapter *adapter, unsigned int fidx)
  */
 void clear_filter(struct adapter *adap, struct filter_entry *f)
 {
-	/* If the new or old filter have loopback rewriteing rules then we'll
-	 * need to free any existing Layer Two Table (L2T) entries of the old
-	 * filter rule.  The firmware will handle freeing up any Source MAC
-	 * Table (SMT) entries used for rewriting Source MAC Addresses in
-	 * loopback rules.
+	struct port_info *pi = netdev_priv(f->dev);
+
+	/* If the new or old filter have loopback rewriting rules then we'll
+	 * need to free any existing L2T, SMT, CLIP entries of filter
+	 * rule.
 	 */
 	if (f->l2t)
 		cxgb4_l2t_release(f->l2t);
 
+	if (f->smt)
+		cxgb4_smt_release(f->smt);
+
+	if (f->fs.val.encap_vld && f->fs.val.ovlan_vld)
+		t4_free_encap_mac_filt(adap, pi->viid,
+				       f->fs.val.ovlan & 0x1ff, 0);
+
+	if ((f->fs.hash || is_t6(adap->params.chip)) && f->fs.type)
+		cxgb4_clip_release(f->dev, (const u32 *)&f->fs.val.lip, 1);
+
 	/* The zeroing of the filter rule below clears the filter valid,
 	 * pending, locked flags, l2t pointer, etc. so it's all we need for
 	 * this operation.
@@ -372,16 +1005,50 @@ void clear_filter(struct adapter *adap, struct filter_entry *f)
 
 void clear_all_filters(struct adapter *adapter)
 {
+	struct net_device *dev = adapter->port[0];
 	unsigned int i;
 
+	if (adapter->tids.hpftid_tab) {
+		struct filter_entry *f = &adapter->tids.hpftid_tab[0];
+
+		for (i = 0; i < adapter->tids.nhpftids; i++, f++)
+			if (f->valid || f->pending)
+				cxgb4_del_filter(dev, i, &f->fs);
+	}
+
 	if (adapter->tids.ftid_tab) {
 		struct filter_entry *f = &adapter->tids.ftid_tab[0];
 		unsigned int max_ftid = adapter->tids.nftids +
-					adapter->tids.nsftids;
+					adapter->tids.nsftids +
+					adapter->tids.nhpftids;
 
-		for (i = 0; i < max_ftid; i++, f++)
+		/* Clear all TCAM filters */
+		for (i = adapter->tids.nhpftids; i < max_ftid; i++, f++)
 			if (f->valid || f->pending)
-				clear_filter(adapter, f);
+				cxgb4_del_filter(dev, i, &f->fs);
+	}
+
+	/* Clear all hash filters */
+	if (is_hashfilter(adapter) && adapter->tids.tid_tab) {
+		struct filter_entry *f;
+		unsigned int sb;
+
+		for (i = adapter->tids.hash_base;
+		     i <= adapter->tids.ntids; i++) {
+			f = (struct filter_entry *)
+				adapter->tids.tid_tab[i];
+
+			if (f && (f->valid || f->pending))
+				cxgb4_del_filter(dev, f->tid, &f->fs);
+		}
+
+		sb = adapter->tids.stid_base;
+		for (i = 0; i < sb; i++) {
+			f = (struct filter_entry *)adapter->tids.tid_tab[i];
+
+			if (f && (f->valid || f->pending))
+				cxgb4_del_filter(dev, f->tid, &f->fs);
+		}
 	}
 }
 
@@ -412,6 +1079,12 @@ static void fill_default_mask(struct ch_filter_specification *fs)
 		fs->mask.tos |= ~0;
 	if (fs->val.proto && !fs->mask.proto)
 		fs->mask.proto |= ~0;
+	if (fs->val.pfvf_vld && !fs->mask.pfvf_vld)
+		fs->mask.pfvf_vld |= ~0;
+	if (fs->val.pf && !fs->mask.pf)
+		fs->mask.pf |= ~0;
+	if (fs->val.vf && !fs->mask.vf)
+		fs->mask.vf |= ~0;
 
 	for (i = 0; i < ARRAY_SIZE(fs->val.lip); i++) {
 		lip |= fs->val.lip[i];
@@ -432,23 +1105,447 @@ static void fill_default_mask(struct ch_filter_specification *fs)
 		fs->mask.fport = ~0;
 }
 
+static bool is_addr_all_mask(u8 *ipmask, int family)
+{
+	if (family == AF_INET) {
+		struct in_addr *addr;
+
+		addr = (struct in_addr *)ipmask;
+		if (addr->s_addr == htonl(0xffffffff))
+			return true;
+	} else if (family == AF_INET6) {
+		struct in6_addr *addr6;
+
+		addr6 = (struct in6_addr *)ipmask;
+		if (addr6->s6_addr32[0] == htonl(0xffffffff) &&
+		    addr6->s6_addr32[1] == htonl(0xffffffff) &&
+		    addr6->s6_addr32[2] == htonl(0xffffffff) &&
+		    addr6->s6_addr32[3] == htonl(0xffffffff))
+			return true;
+	}
+	return false;
+}
+
+static bool is_inaddr_any(u8 *ip, int family)
+{
+	int addr_type;
+
+	if (family == AF_INET) {
+		struct in_addr *addr;
+
+		addr = (struct in_addr *)ip;
+		if (addr->s_addr == htonl(INADDR_ANY))
+			return true;
+	} else if (family == AF_INET6) {
+		struct in6_addr *addr6;
+
+		addr6 = (struct in6_addr *)ip;
+		addr_type = ipv6_addr_type((const struct in6_addr *)
+					   &addr6);
+		if (addr_type == IPV6_ADDR_ANY)
+			return true;
+	}
+	return false;
+}
+
+bool is_filter_exact_match(struct adapter *adap,
+			   struct ch_filter_specification *fs)
+{
+	struct tp_params *tp = &adap->params.tp;
+	u64 hash_filter_mask = tp->hash_filter_mask;
+	u64 ntuple_mask = 0;
+
+	if (!is_hashfilter(adap))
+		return false;
+
+	if ((atomic_read(&adap->tids.hash_tids_in_use) +
+	     atomic_read(&adap->tids.tids_in_use)) >=
+	    (adap->tids.nhash + (adap->tids.stid_base - adap->tids.tid_base)))
+		return false;
+
+	 /* Keep tunnel VNI match disabled for hash-filters for now */
+	if (fs->mask.encap_vld)
+		return false;
+
+	if (fs->type) {
+		if (is_inaddr_any(fs->val.fip, AF_INET6) ||
+		    !is_addr_all_mask(fs->mask.fip, AF_INET6))
+			return false;
+
+		if (is_inaddr_any(fs->val.lip, AF_INET6) ||
+		    !is_addr_all_mask(fs->mask.lip, AF_INET6))
+			return false;
+	} else {
+		if (is_inaddr_any(fs->val.fip, AF_INET) ||
+		    !is_addr_all_mask(fs->mask.fip, AF_INET))
+			return false;
+
+		if (is_inaddr_any(fs->val.lip, AF_INET) ||
+		    !is_addr_all_mask(fs->mask.lip, AF_INET))
+			return false;
+	}
+
+	if (!fs->val.lport || fs->mask.lport != 0xffff)
+		return false;
+
+	if (!fs->val.fport || fs->mask.fport != 0xffff)
+		return false;
+
+	/* calculate tuple mask and compare with mask configured in hw */
+	if (tp->fcoe_shift >= 0)
+		ntuple_mask |= (u64)fs->mask.fcoe << tp->fcoe_shift;
+
+	if (tp->port_shift >= 0)
+		ntuple_mask |= (u64)fs->mask.iport << tp->port_shift;
+
+	if (tp->vnic_shift >= 0) {
+		if ((adap->params.tp.ingress_config & VNIC_F))
+			ntuple_mask |= (u64)fs->mask.pfvf_vld << tp->vnic_shift;
+		else
+			ntuple_mask |= (u64)fs->mask.ovlan_vld <<
+				tp->vnic_shift;
+	}
+
+	if (tp->vlan_shift >= 0)
+		ntuple_mask |= (u64)fs->mask.ivlan << tp->vlan_shift;
+
+	if (tp->tos_shift >= 0)
+		ntuple_mask |= (u64)fs->mask.tos << tp->tos_shift;
+
+	if (tp->protocol_shift >= 0)
+		ntuple_mask |= (u64)fs->mask.proto << tp->protocol_shift;
+
+	if (tp->ethertype_shift >= 0)
+		ntuple_mask |= (u64)fs->mask.ethtype << tp->ethertype_shift;
+
+	if (tp->macmatch_shift >= 0)
+		ntuple_mask |= (u64)fs->mask.macidx << tp->macmatch_shift;
+
+	if (tp->matchtype_shift >= 0)
+		ntuple_mask |= (u64)fs->mask.matchtype << tp->matchtype_shift;
+
+	if (tp->frag_shift >= 0)
+		ntuple_mask |= (u64)fs->mask.frag << tp->frag_shift;
+
+	if (ntuple_mask != hash_filter_mask)
+		return false;
+
+	return true;
+}
+
+static u64 hash_filter_ntuple(struct ch_filter_specification *fs,
+			      struct net_device *dev)
+{
+	struct adapter *adap = netdev2adap(dev);
+	struct tp_params *tp = &adap->params.tp;
+	u64 ntuple = 0;
+
+	/* Initialize each of the fields which we care about which are present
+	 * in the Compressed Filter Tuple.
+	 */
+	if (tp->vlan_shift >= 0 && fs->mask.ivlan)
+		ntuple |= (u64)(FT_VLAN_VLD_F |
+				fs->val.ivlan) << tp->vlan_shift;
+
+	if (tp->port_shift >= 0 && fs->mask.iport)
+		ntuple |= (u64)fs->val.iport << tp->port_shift;
+
+	if (tp->protocol_shift >= 0) {
+		if (!fs->val.proto)
+			ntuple |= (u64)IPPROTO_TCP << tp->protocol_shift;
+		else
+			ntuple |= (u64)fs->val.proto << tp->protocol_shift;
+	}
+
+	if (tp->tos_shift >= 0 && fs->mask.tos)
+		ntuple |= (u64)(fs->val.tos) << tp->tos_shift;
+
+	if (tp->vnic_shift >= 0) {
+		if ((adap->params.tp.ingress_config & USE_ENC_IDX_F) &&
+		    fs->mask.encap_vld)
+			ntuple |= (u64)((fs->val.encap_vld << 16) |
+					(fs->val.ovlan)) << tp->vnic_shift;
+		else if ((adap->params.tp.ingress_config & VNIC_F) &&
+			 fs->mask.pfvf_vld)
+			ntuple |= (u64)((fs->val.pfvf_vld << 16) |
+					(fs->val.pf << 13) |
+					(fs->val.vf)) << tp->vnic_shift;
+		else
+			ntuple |= (u64)((fs->val.ovlan_vld << 16) |
+					(fs->val.ovlan)) << tp->vnic_shift;
+	}
+
+	if (tp->macmatch_shift >= 0 && fs->mask.macidx)
+		ntuple |= (u64)(fs->val.macidx) << tp->macmatch_shift;
+
+	if (tp->ethertype_shift >= 0 && fs->mask.ethtype)
+		ntuple |= (u64)(fs->val.ethtype) << tp->ethertype_shift;
+
+	if (tp->matchtype_shift >= 0 && fs->mask.matchtype)
+		ntuple |= (u64)(fs->val.matchtype) << tp->matchtype_shift;
+
+	if (tp->frag_shift >= 0 && fs->mask.frag)
+		ntuple |= (u64)(fs->val.frag) << tp->frag_shift;
+
+	if (tp->fcoe_shift >= 0 && fs->mask.fcoe)
+		ntuple |= (u64)(fs->val.fcoe) << tp->fcoe_shift;
+	return ntuple;
+}
+
+static void mk_act_open_req6(struct filter_entry *f, struct sk_buff *skb,
+			     unsigned int qid_filterid, struct adapter *adap)
+{
+	struct cpl_t6_act_open_req6 *t6req = NULL;
+	struct cpl_act_open_req6 *req = NULL;
+
+	t6req = (struct cpl_t6_act_open_req6 *)__skb_put(skb, sizeof(*t6req));
+	INIT_TP_WR(t6req, 0);
+	req = (struct cpl_act_open_req6 *)t6req;
+	OPCODE_TID(req) = htonl(MK_OPCODE_TID(CPL_ACT_OPEN_REQ6, qid_filterid));
+	req->local_port = cpu_to_be16(f->fs.val.lport);
+	req->peer_port = cpu_to_be16(f->fs.val.fport);
+	req->local_ip_hi = *(__be64 *)(&f->fs.val.lip);
+	req->local_ip_lo = *(((__be64 *)&f->fs.val.lip) + 1);
+	req->peer_ip_hi = *(__be64 *)(&f->fs.val.fip);
+	req->peer_ip_lo = *(((__be64 *)&f->fs.val.fip) + 1);
+	req->opt0 = cpu_to_be64(NAGLE_V(f->fs.newvlan == VLAN_REMOVE ||
+					f->fs.newvlan == VLAN_REWRITE) |
+				DELACK_V(f->fs.hitcnts) |
+				L2T_IDX_V(f->l2t ? f->l2t->idx : 0) |
+				SMAC_SEL_V((cxgb4_port_viid(f->dev) &
+					    0x7F) << 1) |
+				TX_CHAN_V(f->fs.eport) |
+				NO_CONG_V(f->fs.rpttid) |
+				ULP_MODE_V(f->fs.nat_mode ?
+					   ULP_MODE_TCPDDP : ULP_MODE_NONE) |
+				TCAM_BYPASS_F | NON_OFFLOAD_F);
+	t6req->params = cpu_to_be64(FILTER_TUPLE_V(hash_filter_ntuple(&f->fs,
+								      f->dev)));
+	t6req->opt2 = htonl(RSS_QUEUE_VALID_F |
+			    RSS_QUEUE_V(f->fs.iq) |
+			    TX_QUEUE_V(f->fs.nat_mode) |
+			    T5_OPT_2_VALID_F |
+			    RX_CHANNEL_V(cxgb4_port_e2cchan(f->dev)) |
+			    PACE_V((f->fs.maskhash) |
+				   ((f->fs.dirsteerhash) << 1)));
+}
+
+static void mk_act_open_req(struct filter_entry *f, struct sk_buff *skb,
+			    unsigned int qid_filterid, struct adapter *adap)
+{
+	struct cpl_t6_act_open_req *t6req = NULL;
+	struct cpl_act_open_req *req = NULL;
+
+	t6req = (struct cpl_t6_act_open_req *)__skb_put(skb, sizeof(*t6req));
+	INIT_TP_WR(t6req, 0);
+	req = (struct cpl_act_open_req *)t6req;
+	OPCODE_TID(req) = htonl(MK_OPCODE_TID(CPL_ACT_OPEN_REQ, qid_filterid));
+	req->local_port = cpu_to_be16(f->fs.val.lport);
+	req->peer_port = cpu_to_be16(f->fs.val.fport);
+	memcpy(&req->local_ip, f->fs.val.lip, 4);
+	memcpy(&req->peer_ip, f->fs.val.fip, 4);
+	req->opt0 = cpu_to_be64(NAGLE_V(f->fs.newvlan == VLAN_REMOVE ||
+					f->fs.newvlan == VLAN_REWRITE) |
+				DELACK_V(f->fs.hitcnts) |
+				L2T_IDX_V(f->l2t ? f->l2t->idx : 0) |
+				SMAC_SEL_V((cxgb4_port_viid(f->dev) &
+					    0x7F) << 1) |
+				TX_CHAN_V(f->fs.eport) |
+				NO_CONG_V(f->fs.rpttid) |
+				ULP_MODE_V(f->fs.nat_mode ?
+					   ULP_MODE_TCPDDP : ULP_MODE_NONE) |
+				TCAM_BYPASS_F | NON_OFFLOAD_F);
+
+	t6req->params = cpu_to_be64(FILTER_TUPLE_V(hash_filter_ntuple(&f->fs,
+								      f->dev)));
+	t6req->opt2 = htonl(RSS_QUEUE_VALID_F |
+			    RSS_QUEUE_V(f->fs.iq) |
+			    TX_QUEUE_V(f->fs.nat_mode) |
+			    T5_OPT_2_VALID_F |
+			    RX_CHANNEL_V(cxgb4_port_e2cchan(f->dev)) |
+			    PACE_V((f->fs.maskhash) |
+				   ((f->fs.dirsteerhash) << 1)));
+}
+
+static int cxgb4_set_hash_filter(struct net_device *dev,
+				 struct ch_filter_specification *fs,
+				 struct filter_ctx *ctx)
+{
+	struct adapter *adapter = netdev2adap(dev);
+	struct port_info *pi = netdev_priv(dev);
+	struct tid_info *t = &adapter->tids;
+	struct filter_entry *f;
+	struct sk_buff *skb;
+	int iq, atid, size;
+	int ret = 0;
+	u32 iconf;
+
+	fill_default_mask(fs);
+	ret = validate_filter(dev, fs);
+	if (ret)
+		return ret;
+
+	iq = get_filter_steerq(dev, fs);
+	if (iq < 0)
+		return iq;
+
+	f = kzalloc(sizeof(*f), GFP_KERNEL);
+	if (!f)
+		return -ENOMEM;
+
+	f->fs = *fs;
+	f->ctx = ctx;
+	f->dev = dev;
+	f->fs.iq = iq;
+
+	/* If the new filter requires loopback Destination MAC and/or VLAN
+	 * rewriting then we need to allocate a Layer 2 Table (L2T) entry for
+	 * the filter.
+	 */
+	if (f->fs.newdmac || f->fs.newvlan) {
+		/* allocate L2T entry for new filter */
+		f->l2t = t4_l2t_alloc_switching(adapter, f->fs.vlan,
+						f->fs.eport, f->fs.dmac);
+		if (!f->l2t) {
+			ret = -ENOMEM;
+			goto out_err;
+		}
+	}
+
+	/* If the new filter requires loopback Source MAC rewriting then
+	 * we need to allocate a SMT entry for the filter.
+	 */
+	if (f->fs.newsmac) {
+		f->smt = cxgb4_smt_alloc_switching(f->dev, f->fs.smac);
+		if (!f->smt) {
+			if (f->l2t) {
+				cxgb4_l2t_release(f->l2t);
+				f->l2t = NULL;
+			}
+			ret = -ENOMEM;
+			goto free_l2t;
+		}
+	}
+
+	atid = cxgb4_alloc_atid(t, f);
+	if (atid < 0) {
+		ret = atid;
+		goto free_smt;
+	}
+
+	iconf = adapter->params.tp.ingress_config;
+	if (iconf & VNIC_F) {
+		f->fs.val.ovlan = (fs->val.pf << 13) | fs->val.vf;
+		f->fs.mask.ovlan = (fs->mask.pf << 13) | fs->mask.vf;
+		f->fs.val.ovlan_vld = fs->val.pfvf_vld;
+		f->fs.mask.ovlan_vld = fs->mask.pfvf_vld;
+	} else if (iconf & USE_ENC_IDX_F) {
+		if (f->fs.val.encap_vld) {
+			struct port_info *pi = netdev_priv(f->dev);
+			static const u8 match_all_mac[] = { 0, 0, 0, 0, 0, 0 };
+
+			/* allocate MPS TCAM entry */
+			ret = t4_alloc_encap_mac_filt(adapter, pi->viid,
+						      match_all_mac,
+						      match_all_mac,
+						      f->fs.val.vni,
+						      f->fs.mask.vni,
+						      0, 1, 1);
+			if (ret < 0)
+				goto free_atid;
+
+			f->fs.val.ovlan = ret;
+			f->fs.mask.ovlan = 0xffff;
+			f->fs.val.ovlan_vld = 1;
+			f->fs.mask.ovlan_vld = 1;
+		}
+	}
+
+	size = sizeof(struct cpl_t6_act_open_req);
+	if (f->fs.type) {
+		ret = cxgb4_clip_get(f->dev, (const u32 *)&f->fs.val.lip, 1);
+		if (ret)
+			goto free_mps;
+
+		skb = alloc_skb(size, GFP_KERNEL);
+		if (!skb) {
+			ret = -ENOMEM;
+			goto free_clip;
+		}
+
+		mk_act_open_req6(f, skb,
+				 ((adapter->sge.fw_evtq.abs_id << 14) | atid),
+				 adapter);
+	} else {
+		skb = alloc_skb(size, GFP_KERNEL);
+		if (!skb) {
+			ret = -ENOMEM;
+			goto free_mps;
+		}
+
+		mk_act_open_req(f, skb,
+				((adapter->sge.fw_evtq.abs_id << 14) | atid),
+				adapter);
+	}
+
+	f->pending = 1;
+	set_wr_txq(skb, CPL_PRIORITY_SETUP, f->fs.val.iport & 0x3);
+	t4_ofld_send(adapter, skb);
+	return 0;
+
+free_clip:
+	cxgb4_clip_release(f->dev, (const u32 *)&f->fs.val.lip, 1);
+
+free_mps:
+	if (f->fs.val.encap_vld && f->fs.val.ovlan_vld)
+		t4_free_encap_mac_filt(adapter, pi->viid, f->fs.val.ovlan, 1);
+
+free_atid:
+	cxgb4_free_atid(t, atid);
+
+free_smt:
+	if (f->smt) {
+		cxgb4_smt_release(f->smt);
+		f->smt = NULL;
+	}
+
+free_l2t:
+	if (f->l2t) {
+		cxgb4_l2t_release(f->l2t);
+		f->l2t = NULL;
+	}
+
+out_err:
+	kfree(f);
+	return ret;
+}
+
 /* Check a Chelsio Filter Request for validity, convert it into our internal
  * format and send it to the hardware.  Return 0 on success, an error number
  * otherwise.  We attach any provided filter operation context to the internal
  * filter specification in order to facilitate signaling completion of the
  * operation.
  */
-int __cxgb4_set_filter(struct net_device *dev, int filter_id,
+int __cxgb4_set_filter(struct net_device *dev, int ftid,
 		       struct ch_filter_specification *fs,
 		       struct filter_ctx *ctx)
 {
 	struct adapter *adapter = netdev2adap(dev);
-	unsigned int max_fidx, fidx;
-	struct filter_entry *f;
+	unsigned int max_fidx, fidx, chip_ver;
+	int iq, ret, filter_id = ftid;
+	struct filter_entry *f, *tab;
 	u32 iconf;
-	int iq, ret;
 
-	max_fidx = adapter->tids.nftids;
+	chip_ver = CHELSIO_CHIP_VERSION(adapter->params.chip);
+	if (fs->hash) {
+		if (is_hashfilter(adapter))
+			return cxgb4_set_hash_filter(dev, fs, ctx);
+		netdev_err(dev, "%s: Exact-match filters only supported with Hash Filter configuration\n",
+			   __func__);
+		return -EINVAL;
+	}
+
+	max_fidx = adapter->tids.nftids + adapter->tids.nhpftids;
 	if (filter_id != (max_fidx + adapter->tids.nsftids - 1) &&
 	    filter_id >= max_fidx)
 		return -E2BIG;
@@ -463,6 +1560,13 @@ int __cxgb4_set_filter(struct net_device *dev, int filter_id,
 	if (iq < 0)
 		return iq;
 
+	if (fs->prio) {
+		tab = &adapter->tids.hpftid_tab[0];
+	} else {
+		tab = &adapter->tids.ftid_tab[0];
+		filter_id = ftid - adapter->tids.nhpftids;
+	}
+
 	/* IPv6 filters occupy four slots and must be aligned on
 	 * four-slot boundaries.  IPv4 filters only occupy a single
 	 * slot and have no alignment requirements but writing a new
@@ -471,15 +1575,20 @@ int __cxgb4_set_filter(struct net_device *dev, int filter_id,
 	 * insertion.
 	 */
 	if (fs->type == 0) { /* IPv4 */
-		/* If our IPv4 filter isn't being written to a
-		 * multiple of four filter index and there's an IPv6
-		 * filter at the multiple of 4 base slot, then we
-		 * prevent insertion.
+		/* For T6, If our IPv4 filter isn't being written to a
+		 * multiple of two filter index and there's an IPv6
+		 * filter at the multiple of 2 base slot, then we need
+		 * to delete that IPv6 filter ...
+		 * For adapters below T6, IPv6 filter occupies 4 entries.
+		 * Hence we need to delete the filter in multiple of 4 slot.
 		 */
-		fidx = filter_id & ~0x3;
-		if (fidx != filter_id &&
-		    adapter->tids.ftid_tab[fidx].fs.type) {
-			f = &adapter->tids.ftid_tab[fidx];
+		if (chip_ver < CHELSIO_T6)
+			fidx = filter_id & ~0x3;
+		else
+			fidx = filter_id & ~0x1;
+
+		if (fidx != filter_id && tab[fidx].fs.type) {
+			f = &tab[fidx];
 			if (f->valid) {
 				dev_err(adapter->pdev_dev,
 					"Invalid location. IPv6 requires 4 slots and is occupying slots %u to %u\n",
@@ -488,55 +1597,80 @@ int __cxgb4_set_filter(struct net_device *dev, int filter_id,
 			}
 		}
 	} else { /* IPv6 */
-		/* Ensure that the IPv6 filter is aligned on a
-		 * multiple of 4 boundary.
-		 */
-		if (filter_id & 0x3) {
-			dev_err(adapter->pdev_dev,
-				"Invalid location. IPv6 must be aligned on a 4-slot boundary\n");
-			return -EINVAL;
-		}
-
-		/* Check all except the base overlapping IPv4 filter slots. */
-		for (fidx = filter_id + 1; fidx < filter_id + 4; fidx++) {
-			f = &adapter->tids.ftid_tab[fidx];
-			if (f->valid) {
+		if (chip_ver < CHELSIO_T6) {
+			/* Ensure that the IPv6 filter is aligned on a
+			 * multiple of 4 boundary.
+			 */
+			if (filter_id & 0x3) {
 				dev_err(adapter->pdev_dev,
-					"Invalid location.  IPv6 requires 4 slots and an IPv4 filter exists at %u\n",
-					fidx);
+					"Invalid location. IPv6 must be aligned on a 4-slot boundary\n");
 				return -EINVAL;
 			}
+
+			/* Check all except the base overlapping IPv4 filter
+			 * slots.
+			 */
+			for (fidx = filter_id + 1; fidx < filter_id + 4;
+			     fidx++) {
+				f = &tab[fidx];
+				if (f->valid) {
+					dev_err(adapter->pdev_dev,
+						"Invalid location.  IPv6 requires 4 slots and an IPv4 filter exists at %u\n",
+						fidx);
+					return -EBUSY;
+				}
+			}
+		} else {
+			/* For T6, CLIP being enabled, IPv6 filter would occupy
+			 * 2 entries.
+			 */
+			if (filter_id & 0x1)
+				return -EINVAL;
+			/* Check overlapping IPv4 filter slot */
+			fidx = filter_id + 1;
+			f = &tab[fidx];
+			if (f->valid) {
+				pr_err("%s: IPv6 filter requires 2 indices. IPv4 filter already present at %d. Please remove IPv4 filter first.\n",
+				       __func__, fidx);
+				return -EBUSY;
+			}
 		}
 	}
 
 	/* Check to make sure that provided filter index is not
 	 * already in use by someone else
 	 */
-	f = &adapter->tids.ftid_tab[filter_id];
+	f = &tab[filter_id];
 	if (f->valid)
 		return -EBUSY;
 
-	fidx = filter_id + adapter->tids.ftid_base;
-	ret = cxgb4_set_ftid(&adapter->tids, filter_id,
-			     fs->type ? PF_INET6 : PF_INET);
+	if (fs->prio) {
+		fidx = filter_id + adapter->tids.hpftid_base;
+		ret = cxgb4_set_hpftid(&adapter->tids, filter_id,
+				       fs->type ? PF_INET6 : PF_INET);
+	} else {
+		fidx = filter_id + adapter->tids.ftid_base;
+		ret = cxgb4_set_ftid(&adapter->tids, filter_id,
+				     fs->type ? PF_INET6 : PF_INET,
+				     chip_ver);
+	}
+
 	if (ret)
 		return ret;
 
-	/* Check to make sure the filter requested is writable ... */
+	/* Check t  make sure the filter requested is writable ... */
 	ret = writable_filter(f);
-	if (ret) {
-		/* Clear the bits we have set above */
-		cxgb4_clear_ftid(&adapter->tids, filter_id,
-				 fs->type ? PF_INET6 : PF_INET);
-		return ret;
+	if (ret)
+		goto free_tid;
+
+	if (is_t6(adapter->params.chip) && fs->type &&
+	    ipv6_addr_type((const struct in6_addr *)fs->val.lip) !=
+	    IPV6_ADDR_ANY) {
+		ret = cxgb4_clip_get(dev, (const u32 *)&fs->val.lip, 1);
+		if (ret)
+			goto free_tid;
 	}
 
-	/* Clear out any old resources being used by the filter before
-	 * we start constructing the new filter.
-	 */
-	if (f->valid)
-		clear_filter(adapter, f);
-
 	/* Convert the filter specification into our internal format.
 	 * We copy the PF/VF specification into the Outer VLAN field
 	 * here so the rest of the code -- including the interface to
@@ -552,6 +1686,26 @@ int __cxgb4_set_filter(struct net_device *dev, int filter_id,
 		f->fs.mask.ovlan = (fs->mask.pf << 13) | fs->mask.vf;
 		f->fs.val.ovlan_vld = fs->val.pfvf_vld;
 		f->fs.mask.ovlan_vld = fs->mask.pfvf_vld;
+	} else if (iconf & USE_ENC_IDX_F) {
+		if (f->fs.val.encap_vld) {
+			struct port_info *pi = netdev_priv(f->dev);
+			static const u8 match_all_mac[] = { 0, 0, 0, 0, 0, 0 };
+
+			/* allocate MPS TCAM entry */
+			ret = t4_alloc_encap_mac_filt(adapter, pi->viid,
+						      match_all_mac,
+						      match_all_mac,
+						      f->fs.val.vni,
+						      f->fs.mask.vni,
+						      0, 1, 1);
+			if (ret < 0)
+				goto free_tid;
+
+			f->fs.val.ovlan = ret;
+			f->fs.mask.ovlan = 0x1ff;
+			f->fs.val.ovlan_vld = 1;
+			f->fs.mask.ovlan_vld = 1;
+		}
 	}
 
 	/* Attempt to set the filter.  If we don't succeed, we clear
@@ -559,43 +1713,134 @@ int __cxgb4_set_filter(struct net_device *dev, int filter_id,
 	 */
 	f->ctx = ctx;
 	f->tid = fidx; /* Save the actual tid */
-	ret = set_filter_wr(adapter, filter_id);
-	if (ret) {
+	ret = set_filter_wr(adapter, ftid);
+	if (ret)
+		goto free_tid;
+
+	return ret;
+
+free_tid:
+	if (f->fs.prio)
+		cxgb4_clear_hpftid(&adapter->tids, filter_id,
+				   fs->type ? PF_INET6 : PF_INET);
+	else
 		cxgb4_clear_ftid(&adapter->tids, filter_id,
-				 fs->type ? PF_INET6 : PF_INET);
-		clear_filter(adapter, f);
-	}
+				 fs->type ? PF_INET6 : PF_INET,
+				 chip_ver);
 
+	clear_filter(adapter, f);
 	return ret;
 }
 
+static int cxgb4_del_hash_filter(struct net_device *dev, int filter_id,
+				 struct filter_ctx *ctx)
+{
+	struct adapter *adapter = netdev2adap(dev);
+	struct tid_info *t = &adapter->tids;
+	struct cpl_abort_req *abort_req;
+	struct cpl_abort_rpl *abort_rpl;
+	struct cpl_set_tcb_field *req;
+	struct ulptx_idata *aligner;
+	struct work_request_hdr *wr;
+	struct filter_entry *f;
+	struct sk_buff *skb;
+	unsigned int wrlen;
+	int ret;
+
+	netdev_dbg(dev, "%s: filter_id = %d ; nftids = %d\n",
+		   __func__, filter_id, adapter->tids.nftids);
+
+	if (tid_out_of_range(t, filter_id))
+		return -E2BIG;
+
+	f = lookup_tid(t, filter_id);
+	if (!f) {
+		netdev_err(dev, "%s: no filter entry for filter_id = %d",
+			   __func__, filter_id);
+		return -EINVAL;
+	}
+
+	ret = writable_filter(f);
+	if (ret)
+		return ret;
+
+	if (!f->valid)
+		return -EINVAL;
+
+	f->ctx = ctx;
+	f->pending = 1;
+	wrlen = roundup(sizeof(*wr) + (sizeof(*req) + sizeof(*aligner))
+			+ sizeof(*abort_req) + sizeof(*abort_rpl), 16);
+	skb = alloc_skb(wrlen, GFP_KERNEL);
+	if (!skb) {
+		netdev_err(dev, "%s: could not allocate skb ..\n", __func__);
+		return -ENOMEM;
+	}
+	set_wr_txq(skb, CPL_PRIORITY_CONTROL, f->fs.val.iport & 0x3);
+	req = (struct cpl_set_tcb_field *)__skb_put(skb, wrlen);
+	INIT_ULPTX_WR(req, wrlen, 0, 0);
+	wr = (struct work_request_hdr *)req;
+	wr++;
+	req = (struct cpl_set_tcb_field *)wr;
+	mk_set_tcb_ulp(f, req, TCB_RSS_INFO_W, TCB_RSS_INFO_V(TCB_RSS_INFO_M),
+		       TCB_RSS_INFO_V(adapter->sge.fw_evtq.abs_id), 0, 1);
+	aligner = (struct ulptx_idata *)(req + 1);
+	abort_req = (struct cpl_abort_req *)(aligner + 1);
+	mk_abort_req_ulp(abort_req, f->tid);
+	abort_rpl = (struct cpl_abort_rpl *)(abort_req + 1);
+	mk_abort_rpl_ulp(abort_rpl, f->tid);
+	t4_ofld_send(adapter, skb);
+	return 0;
+}
+
 /* Check a delete filter request for validity and send it to the hardware.
  * Return 0 on success, an error number otherwise.  We attach any provided
  * filter operation context to the internal filter specification in order to
  * facilitate signaling completion of the operation.
  */
 int __cxgb4_del_filter(struct net_device *dev, int filter_id,
+		       struct ch_filter_specification *fs,
 		       struct filter_ctx *ctx)
 {
 	struct adapter *adapter = netdev2adap(dev);
+	unsigned int max_fidx, chip_ver;
 	struct filter_entry *f;
-	unsigned int max_fidx;
 	int ret;
 
-	max_fidx = adapter->tids.nftids;
+	chip_ver = CHELSIO_CHIP_VERSION(adapter->params.chip);
+	if (fs && fs->hash) {
+		if (is_hashfilter(adapter))
+			return cxgb4_del_hash_filter(dev, filter_id, ctx);
+		netdev_err(dev, "%s: Exact-match filters only supported with Hash Filter configuration\n",
+			   __func__);
+		return -EINVAL;
+	}
+
+	max_fidx = adapter->tids.nftids + adapter->tids.nhpftids;
 	if (filter_id != (max_fidx + adapter->tids.nsftids - 1) &&
 	    filter_id >= max_fidx)
 		return -E2BIG;
 
-	f = &adapter->tids.ftid_tab[filter_id];
+	if (filter_id < adapter->tids.nhpftids)
+		f = &adapter->tids.hpftid_tab[filter_id];
+	else
+		f = &adapter->tids.ftid_tab[filter_id - adapter->tids.nhpftids];
+
 	ret = writable_filter(f);
 	if (ret)
 		return ret;
 
 	if (f->valid) {
 		f->ctx = ctx;
-		cxgb4_clear_ftid(&adapter->tids, filter_id,
-				 f->fs.type ? PF_INET6 : PF_INET);
+		if (f->fs.prio)
+			cxgb4_clear_hpftid(&adapter->tids,
+					   f->tid - adapter->tids.hpftid_base,
+					   f->fs.type ? PF_INET6 : PF_INET);
+		else
+			cxgb4_clear_ftid(&adapter->tids,
+					 f->tid - adapter->tids.ftid_base,
+					 f->fs.type ? PF_INET6 : PF_INET,
+					 chip_ver);
 		return del_filter_wr(adapter, filter_id);
 	}
 
@@ -632,14 +1877,18 @@ out:
 	return ret;
 }
 
-int cxgb4_del_filter(struct net_device *dev, int filter_id)
+int cxgb4_del_filter(struct net_device *dev, int filter_id,
+		     struct ch_filter_specification *fs)
 {
 	struct filter_ctx ctx;
 	int ret;
 
+	if (netdev2adap(dev)->flags & CXGB4_SHUTTING_DOWN)
+		return 0;
+
 	init_completion(&ctx.completion);
 
-	ret = __cxgb4_del_filter(dev, filter_id, &ctx);
+	ret = __cxgb4_del_filter(dev, filter_id, fs, &ctx);
 	if (ret)
 		goto out;
 
@@ -653,6 +1902,175 @@ out:
 	return ret;
 }
 
+static int configure_filter_tcb(struct adapter *adap, unsigned int tid,
+				struct filter_entry *f)
+{
+	if (f->fs.hitcnts) {
+		set_tcb_field(adap, f, tid, TCB_TIMESTAMP_W,
+			      TCB_TIMESTAMP_V(TCB_TIMESTAMP_M),
+			      TCB_TIMESTAMP_V(0ULL),
+			      1);
+		set_tcb_field(adap, f, tid, TCB_RTT_TS_RECENT_AGE_W,
+			      TCB_RTT_TS_RECENT_AGE_V(TCB_RTT_TS_RECENT_AGE_M),
+			      TCB_RTT_TS_RECENT_AGE_V(0ULL),
+			      1);
+	}
+
+	if (f->fs.newdmac)
+		set_tcb_tflag(adap, f, tid, TF_CCTRL_ECE_S, 1,
+			      1);
+
+	if (f->fs.newvlan == VLAN_INSERT ||
+	    f->fs.newvlan == VLAN_REWRITE)
+		set_tcb_tflag(adap, f, tid, TF_CCTRL_RFR_S, 1,
+			      1);
+	if (f->fs.newsmac)
+		configure_filter_smac(adap, f);
+
+	if (f->fs.nat_mode) {
+		switch (f->fs.nat_mode) {
+		case NAT_MODE_DIP:
+			set_nat_params(adap, f, tid, true, false, false, false);
+			break;
+
+		case NAT_MODE_DIP_DP:
+			set_nat_params(adap, f, tid, true, false, true, false);
+			break;
+
+		case NAT_MODE_DIP_DP_SIP:
+			set_nat_params(adap, f, tid, true, true, true, false);
+			break;
+		case NAT_MODE_DIP_DP_SP:
+			set_nat_params(adap, f, tid, true, false, true, true);
+			break;
+
+		case NAT_MODE_SIP_SP:
+			set_nat_params(adap, f, tid, false, true, false, true);
+			break;
+
+		case NAT_MODE_DIP_SIP_SP:
+			set_nat_params(adap, f, tid, true, true, false, true);
+			break;
+
+		case NAT_MODE_ALL:
+			set_nat_params(adap, f, tid, true, true, true, true);
+			break;
+
+		default:
+			pr_err("%s: Invalid NAT mode: %d\n",
+			       __func__, f->fs.nat_mode);
+			return -EINVAL;
+		}
+	}
+	return 0;
+}
+
+void hash_del_filter_rpl(struct adapter *adap,
+			 const struct cpl_abort_rpl_rss *rpl)
+{
+	unsigned int status = rpl->status;
+	struct tid_info *t = &adap->tids;
+	unsigned int tid = GET_TID(rpl);
+	struct filter_ctx *ctx = NULL;
+	struct filter_entry *f;
+
+	dev_dbg(adap->pdev_dev, "%s: status = %u; tid = %u\n",
+		__func__, status, tid);
+
+	f = lookup_tid(t, tid);
+	if (!f) {
+		dev_err(adap->pdev_dev, "%s:could not find filter entry",
+			__func__);
+		return;
+	}
+	ctx = f->ctx;
+	f->ctx = NULL;
+	clear_filter(adap, f);
+	cxgb4_remove_tid(t, 0, tid, 0);
+	kfree(f);
+	if (ctx) {
+		ctx->result = 0;
+		complete(&ctx->completion);
+	}
+}
+
+void hash_filter_rpl(struct adapter *adap, const struct cpl_act_open_rpl *rpl)
+{
+	unsigned int ftid = TID_TID_G(AOPEN_ATID_G(ntohl(rpl->atid_status)));
+	unsigned int status  = AOPEN_STATUS_G(ntohl(rpl->atid_status));
+	struct tid_info *t = &adap->tids;
+	unsigned int tid = GET_TID(rpl);
+	struct filter_ctx *ctx = NULL;
+	struct filter_entry *f;
+
+	dev_dbg(adap->pdev_dev, "%s: tid = %u; atid = %u; status = %u\n",
+		__func__, tid, ftid, status);
+
+	f = lookup_atid(t, ftid);
+	if (!f) {
+		dev_err(adap->pdev_dev, "%s:could not find filter entry",
+			__func__);
+		return;
+	}
+	ctx = f->ctx;
+	f->ctx = NULL;
+
+	switch (status) {
+	case CPL_ERR_NONE:
+		f->tid = tid;
+		f->pending = 0;
+		f->valid = 1;
+		cxgb4_insert_tid(t, f, f->tid, 0);
+		cxgb4_free_atid(t, ftid);
+		if (ctx) {
+			ctx->tid = f->tid;
+			ctx->result = 0;
+		}
+		if (configure_filter_tcb(adap, tid, f)) {
+			clear_filter(adap, f);
+			cxgb4_remove_tid(t, 0, tid, 0);
+			kfree(f);
+			if (ctx) {
+				ctx->result = -EINVAL;
+				complete(&ctx->completion);
+			}
+			return;
+		}
+		switch (f->fs.action) {
+		case FILTER_PASS:
+			if (f->fs.dirsteer)
+				set_tcb_tflag(adap, f, tid,
+					      TF_DIRECT_STEER_S, 1, 1);
+			break;
+		case FILTER_DROP:
+			set_tcb_tflag(adap, f, tid, TF_DROP_S, 1, 1);
+			break;
+		case FILTER_SWITCH:
+			set_tcb_tflag(adap, f, tid, TF_LPBK_S, 1, 1);
+			break;
+		}
+
+		break;
+
+	default:
+		if (status != CPL_ERR_TCAM_FULL)
+			dev_err(adap->pdev_dev, "%s: filter creation PROBLEM; status = %u\n",
+				__func__, status);
+
+		if (ctx) {
+			if (status == CPL_ERR_TCAM_FULL)
+				ctx->result = -ENOSPC;
+			else
+				ctx->result = -EINVAL;
+		}
+		clear_filter(adap, f);
+		cxgb4_free_atid(t, ftid);
+		kfree(f);
+	}
+	if (ctx)
+		complete(&ctx->completion);
+}
+
 /* Handle a filter write/deletion reply. */
 void filter_rpl(struct adapter *adap, const struct cpl_set_tcb_rpl *rpl)
 {
@@ -664,11 +2082,18 @@ void filter_rpl(struct adapter *adap, const struct cpl_set_tcb_rpl *rpl)
 	max_fidx = adap->tids.nftids + adap->tids.nsftids;
 	/* Get the corresponding filter entry for this tid */
 	if (adap->tids.ftid_tab) {
-		/* Check this in normal filter region */
-		idx = tid - adap->tids.ftid_base;
-		if (idx >= max_fidx)
-			return;
-		f = &adap->tids.ftid_tab[idx];
+		idx = tid - adap->tids.hpftid_base;
+		if (idx < adap->tids.nhpftids) {
+			f = &adap->tids.hpftid_tab[idx];
+		} else {
+			/* Check this in normal filter region */
+			idx = tid - adap->tids.ftid_base;
+			if (idx >= max_fidx)
+				return;
+			f = &adap->tids.ftid_tab[idx];
+			idx += adap->tids.nhpftids;
+		}
+
 		if (f->tid != tid)
 			return;
 	}
@@ -691,15 +2116,8 @@ void filter_rpl(struct adapter *adap, const struct cpl_set_tcb_rpl *rpl)
 			clear_filter(adap, f);
 			if (ctx)
 				ctx->result = 0;
-		} else if (ret == FW_FILTER_WR_SMT_TBL_FULL) {
-			dev_err(adap->pdev_dev, "filter %u setup failed due to full SMT\n",
-				idx);
-			clear_filter(adap, f);
-			if (ctx)
-				ctx->result = -ENOMEM;
 		} else if (ret == FW_FILTER_WR_FLT_ADDED) {
-			f->smtidx = (be64_to_cpu(rpl->oldval) >> 24) & 0xff;
-			f->pending = 0;  /* asynchronous setup completed */
+			f->pending = 0;  /* async setup completed */
 			f->valid = 1;
 			if (ctx) {
 				ctx->result = 0;
@@ -719,3 +2137,39 @@ void filter_rpl(struct adapter *adap, const struct cpl_set_tcb_rpl *rpl)
 			complete(&ctx->completion);
 	}
 }
+
+void init_hash_filter(struct adapter *adap)
+{
+	u32 reg;
+
+	/* On T6, verify the necessary register configs and warn the user in
+	 * case of improper config
+	 */
+	if (is_t6(adap->params.chip)) {
+		if (is_offload(adap)) {
+			if (!(t4_read_reg(adap, TP_GLOBAL_CONFIG_A)
+			   & ACTIVEFILTERCOUNTS_F)) {
+				dev_err(adap->pdev_dev, "Invalid hash filter + ofld config\n");
+				return;
+			}
+		} else {
+			reg = t4_read_reg(adap, LE_DB_RSP_CODE_0_A);
+			if (TCAM_ACTV_HIT_G(reg) != 4) {
+				dev_err(adap->pdev_dev, "Invalid hash filter config\n");
+				return;
+			}
+
+			reg = t4_read_reg(adap, LE_DB_RSP_CODE_1_A);
+			if (HASH_ACTV_HIT_G(reg) != 4) {
+				dev_err(adap->pdev_dev, "Invalid hash filter config\n");
+				return;
+			}
+		}
+
+	} else {
+		dev_err(adap->pdev_dev, "Hash filter supported only on T6\n");
+		return;
+	}
+
+	adap->params.hash_filter = 1;
+}
diff --git a/drivers/net/ethernet/chelsio/cxgb4/cxgb4_filter.h b/drivers/net/ethernet/chelsio/cxgb4/cxgb4_filter.h
index 23742cb1c69f..807a8dafec45 100644
--- a/drivers/net/ethernet/chelsio/cxgb4/cxgb4_filter.h
+++ b/drivers/net/ethernet/chelsio/cxgb4/cxgb4_filter.h
@@ -37,7 +37,12 @@
 
 #include "t4_msg.h"
 
+#define WORD_MASK	0xffffffff
+
 void filter_rpl(struct adapter *adap, const struct cpl_set_tcb_rpl *rpl);
+void hash_filter_rpl(struct adapter *adap, const struct cpl_act_open_rpl *rpl);
+void hash_del_filter_rpl(struct adapter *adap,
+			 const struct cpl_abort_rpl_rss *rpl);
 void clear_filter(struct adapter *adap, struct filter_entry *f);
 
 int set_filter_wr(struct adapter *adapter, int fidx);
@@ -45,4 +50,9 @@ int delete_filter(struct adapter *adapter, unsigned int fidx);
 
 int writable_filter(struct filter_entry *f);
 void clear_all_filters(struct adapter *adapter);
+void init_hash_filter(struct adapter *adap);
+bool is_filter_exact_match(struct adapter *adap,
+			   struct ch_filter_specification *fs);
+void cxgb4_cleanup_ethtool_filters(struct adapter *adap);
+int cxgb4_init_ethtool_filters(struct adapter *adap);
 #endif /* __CXGB4_FILTER_H */
diff --git a/drivers/net/ethernet/chelsio/cxgb4/cxgb4_main.c b/drivers/net/ethernet/chelsio/cxgb4/cxgb4_main.c
index 8c69046be025..392723ef14e5 100644
--- a/drivers/net/ethernet/chelsio/cxgb4/cxgb4_main.c
+++ b/drivers/net/ethernet/chelsio/cxgb4/cxgb4_main.c
@@ -51,7 +51,6 @@
 #include <linux/mutex.h>
 #include <linux/netdevice.h>
 #include <linux/pci.h>
-#include <linux/aer.h>
 #include <linux/rtnetlink.h>
 #include <linux/sched.h>
 #include <linux/seq_file.h>
@@ -62,9 +61,13 @@
 #include <net/netevent.h>
 #include <net/addrconf.h>
 #include <net/bonding.h>
-#include <net/addrconf.h>
 #include <linux/uaccess.h>
 #include <linux/crash_dump.h>
+#include <net/udp_tunnel.h>
+#include <net/xfrm.h>
+#if IS_ENABLED(CONFIG_CHELSIO_TLS_DEVICE)
+#include <net/tls.h>
+#endif
 
 #include "cxgb4.h"
 #include "cxgb4_filter.h"
@@ -74,19 +77,21 @@
 #include "t4fw_api.h"
 #include "t4fw_version.h"
 #include "cxgb4_dcb.h"
+#include "srq.h"
 #include "cxgb4_debugfs.h"
 #include "clip_tbl.h"
 #include "l2t.h"
+#include "smt.h"
 #include "sched.h"
 #include "cxgb4_tc_u32.h"
+#include "cxgb4_tc_flower.h"
+#include "cxgb4_tc_mqprio.h"
+#include "cxgb4_tc_matchall.h"
+#include "cxgb4_ptp.h"
+#include "cxgb4_cudbg.h"
 
 char cxgb4_driver_name[] = KBUILD_MODNAME;
 
-#ifdef DRV_VERSION
-#undef DRV_VERSION
-#endif
-#define DRV_VERSION "2.0.0-ko"
-const char cxgb4_driver_version[] = DRV_VERSION;
 #define DRV_DESC "Chelsio T4/T5/T6 Network Driver"
 
 #define DFLT_MSG_ENABLE (NETIF_MSG_DRV | NETIF_MSG_PROBE | NETIF_MSG_LINK | \
@@ -97,7 +102,9 @@ const char cxgb4_driver_version[] = DRV_VERSION;
  */
 #define CH_PCI_DEVICE_ID_TABLE_DEFINE_BEGIN \
 	static const struct pci_device_id cxgb4_pci_tbl[] = {
-#define CH_PCI_DEVICE_ID_FUNCTION 0x4
+#define CXGB4_UNIFIED_PF 0x4
+
+#define CH_PCI_DEVICE_ID_FUNCTION CXGB4_UNIFIED_PF
 
 /* Include PCI Device IDs for both PF4 and PF0-3 so our PCI probe() routine is
  * called for both.
@@ -105,7 +112,7 @@ const char cxgb4_driver_version[] = DRV_VERSION;
 #define CH_PCI_DEVICE_ID_FUNCTION2 0x0
 
 #define CH_PCI_ID_TABLE_ENTRY(devid) \
-		{PCI_VDEVICE(CHELSIO, (devid)), 4}
+		{PCI_VDEVICE(CHELSIO, (devid)), CXGB4_UNIFIED_PF}
 
 #define CH_PCI_DEVICE_ID_TABLE_DEFINE_END \
 		{ 0, } \
@@ -127,7 +134,6 @@ const char cxgb4_driver_version[] = DRV_VERSION;
 MODULE_DESCRIPTION(DRV_DESC);
 MODULE_AUTHOR("Chelsio Communications");
 MODULE_LICENSE("Dual BSD/GPL");
-MODULE_VERSION(DRV_VERSION);
 MODULE_DEVICE_TABLE(pci, cxgb4_pci_tbl);
 MODULE_FIRMWARE(FW4_FNAME);
 MODULE_FIRMWARE(FW5_FNAME);
@@ -176,6 +182,9 @@ static struct dentry *cxgb4_debugfs_root;
 
 LIST_HEAD(adapter_list);
 DEFINE_MUTEX(uld_mutex);
+LIST_HEAD(uld_list);
+
+static int cfg_queues(struct adapter *adap);
 
 static void link_report(struct net_device *dev)
 {
@@ -203,6 +212,9 @@ static void link_report(struct net_device *dev)
 		case 40000:
 			s = "40Gbps";
 			break;
+		case 50000:
+			s = "50Gbps";
+			break;
 		case 100000:
 			s = "100Gbps";
 			break;
@@ -252,11 +264,11 @@ static void dcb_tx_queue_prio_enable(struct net_device *dev, int enable)
 				"Can't %s DCB Priority on port %d, TX Queue %d: err=%d\n",
 				enable ? "set" : "unset", pi->port_id, i, -err);
 		else
-			txq->dcb_prio = value;
+			txq->dcb_prio = enable ? value : 0;
 	}
 }
 
-static int cxgb4_dcb_enabled(const struct net_device *dev)
+int cxgb4_dcb_enabled(const struct net_device *dev)
 {
 	struct port_info *pi = netdev_priv(dev);
 
@@ -279,7 +291,7 @@ void t4_os_link_changed(struct adapter *adapter, int port_id, int link_stat)
 		else {
 #ifdef CONFIG_CHELSIO_T4_DCB
 			if (cxgb4_dcb_enabled(dev)) {
-				cxgb4_dcb_state_init(dev);
+				cxgb4_dcb_reset(dev);
 				dcb_tx_queue_prio_enable(dev, false);
 			}
 #endif /* CONFIG_CHELSIO_T4_DCB */
@@ -290,14 +302,14 @@ void t4_os_link_changed(struct adapter *adapter, int port_id, int link_stat)
 	}
 }
 
-void t4_os_portmod_changed(const struct adapter *adap, int port_id)
+void t4_os_portmod_changed(struct adapter *adap, int port_id)
 {
 	static const char *mod_str[] = {
 		NULL, "LR", "SR", "ER", "passive DA", "active DA", "LRM"
 	};
 
-	const struct net_device *dev = adap->port[port_id];
-	const struct port_info *pi = netdev_priv(dev);
+	struct net_device *dev = adap->port[port_id];
+	struct port_info *pi = netdev_priv(dev);
 
 	if (pi->mod_type == FW_PORT_MOD_TYPE_NONE)
 		netdev_info(dev, "port module unplugged\n");
@@ -314,6 +326,11 @@ void t4_os_portmod_changed(const struct adapter *adap, int port_id)
 	else
 		netdev_info(dev, "%s: unknown module type %d inserted\n",
 			    dev->name, pi->mod_type);
+
+	/* If the interface is running, then we'll need any "sticky" Link
+	 * Parameters redone with a new Transceiver Module.
+	 */
+	pi->link_cfg.redo_l1cfg = netif_running(dev);
 }
 
 int dbfifo_int_thresh = 10; /* 10 == 640 entry threshold */
@@ -351,13 +368,19 @@ static int cxgb4_mac_sync(struct net_device *netdev, const u8 *mac_addr)
 	int ret;
 	u64 mhash = 0;
 	u64 uhash = 0;
+	/* idx stores the index of allocated filters,
+	 * its size should be modified based on the number of
+	 * MAC addresses that we allocate filters for
+	 */
+
+	u16 idx[1] = {};
 	bool free = false;
 	bool ucast = is_unicast_ether_addr(mac_addr);
 	const u8 *maclist[1] = {mac_addr};
 	struct hash_mac_addr *new_entry;
 
-	ret = t4_alloc_mac_filt(adap, adap->mbox, pi->viid, free, 1, maclist,
-				NULL, ucast ? &uhash : &mhash, false);
+	ret = cxgb4_alloc_mac_filt(adap, pi->viid, free, 1, maclist,
+				   idx, ucast ? &uhash : &mhash, false);
 	if (ret < 0)
 		goto out;
 	/* if hash != 0, then add the addr to hash addr list
@@ -395,7 +418,7 @@ static int cxgb4_mac_unsync(struct net_device *netdev, const u8 *mac_addr)
 		}
 	}
 
-	ret = t4_free_mac_filt(adap, adap->mbox, pi->viid, 1, maclist, false);
+	ret = cxgb4_free_mac_filt(adap, pi->viid, 1, maclist, false);
 	return ret < 0 ? -EINVAL : 0;
 }
 
@@ -411,13 +434,67 @@ static int set_rxmode(struct net_device *dev, int mtu, bool sleep_ok)
 	__dev_uc_sync(dev, cxgb4_mac_sync, cxgb4_mac_unsync);
 	__dev_mc_sync(dev, cxgb4_mac_sync, cxgb4_mac_unsync);
 
-	return t4_set_rxmode(adapter, adapter->mbox, pi->viid, mtu,
-			     (dev->flags & IFF_PROMISC) ? 1 : 0,
+	return t4_set_rxmode(adapter, adapter->mbox, pi->viid, pi->viid_mirror,
+			     mtu, (dev->flags & IFF_PROMISC) ? 1 : 0,
 			     (dev->flags & IFF_ALLMULTI) ? 1 : 0, 1, -1,
 			     sleep_ok);
 }
 
 /**
+ *	cxgb4_change_mac - Update match filter for a MAC address.
+ *	@pi: the port_info
+ *	@viid: the VI id
+ *	@tcam_idx: TCAM index of existing filter for old value of MAC address,
+ *		   or -1
+ *	@addr: the new MAC address value
+ *	@persist: whether a new MAC allocation should be persistent
+ *	@smt_idx: the destination to store the new SMT index.
+ *
+ *	Modifies an MPS filter and sets it to the new MAC address if
+ *	@tcam_idx >= 0, or adds the MAC address to a new filter if
+ *	@tcam_idx < 0. In the latter case the address is added persistently
+ *	if @persist is %true.
+ *	Addresses are programmed to hash region, if tcam runs out of entries.
+ *
+ */
+int cxgb4_change_mac(struct port_info *pi, unsigned int viid,
+		     int *tcam_idx, const u8 *addr, bool persist,
+		     u8 *smt_idx)
+{
+	struct adapter *adapter = pi->adapter;
+	struct hash_mac_addr *entry, *new_entry;
+	int ret;
+
+	ret = t4_change_mac(adapter, adapter->mbox, viid,
+			    *tcam_idx, addr, persist, smt_idx);
+	/* We ran out of TCAM entries. try programming hash region. */
+	if (ret == -ENOMEM) {
+		/* If the MAC address to be updated is in the hash addr
+		 * list, update it from the list
+		 */
+		list_for_each_entry(entry, &adapter->mac_hlist, list) {
+			if (entry->iface_mac) {
+				ether_addr_copy(entry->addr, addr);
+				goto set_hash;
+			}
+		}
+		new_entry = kzalloc(sizeof(*new_entry), GFP_KERNEL);
+		if (!new_entry)
+			return -ENOMEM;
+		ether_addr_copy(new_entry->addr, addr);
+		new_entry->iface_mac = true;
+		list_add_tail(&new_entry->list, &adapter->mac_hlist);
+set_hash:
+		ret = cxgb4_set_addr_hash(pi);
+	} else if (ret >= 0) {
+		*tcam_idx = ret;
+		ret = 0;
+	}
+
+	return ret;
+}
+
+/*
  *	link_start - enable a port
  *	@dev: the port to enable
  *
@@ -425,31 +502,26 @@ static int set_rxmode(struct net_device *dev, int mtu, bool sleep_ok)
  */
 static int link_start(struct net_device *dev)
 {
-	int ret;
 	struct port_info *pi = netdev_priv(dev);
-	unsigned int mb = pi->adapter->pf;
+	unsigned int mb = pi->adapter->mbox;
+	int ret;
 
 	/*
 	 * We do not set address filters and promiscuity here, the stack does
 	 * that step explicitly.
 	 */
-	ret = t4_set_rxmode(pi->adapter, mb, pi->viid, dev->mtu, -1, -1, -1,
+	ret = t4_set_rxmode(pi->adapter, mb, pi->viid, pi->viid_mirror,
+			    dev->mtu, -1, -1, -1,
 			    !!(dev->features & NETIF_F_HW_VLAN_CTAG_RX), true);
-	if (ret == 0) {
-		ret = t4_change_mac(pi->adapter, mb, pi->viid,
-				    pi->xact_addr_filt, dev->dev_addr, true,
-				    true);
-		if (ret >= 0) {
-			pi->xact_addr_filt = ret;
-			ret = 0;
-		}
-	}
+	if (ret == 0)
+		ret = cxgb4_update_mac_filt(pi, pi->viid, &pi->xact_addr_filt,
+					    dev->dev_addr, true, &pi->smt_idx);
 	if (ret == 0)
 		ret = t4_link_l1cfg(pi->adapter, mb, pi->tx_chan,
 				    &pi->link_cfg);
 	if (ret == 0) {
 		local_bh_disable();
-		ret = t4_enable_vi_params(pi->adapter, mb, pi->viid, true,
+		ret = t4_enable_pi_params(pi->adapter, mb, pi, true,
 					  true, CXGB4_DCB_ENABLED);
 		local_bh_enable();
 	}
@@ -512,7 +584,7 @@ static int fwevtq_handler(struct sge_rspq *q, const __be64 *rsp,
 			struct sge_eth_txq *eq;
 
 			eq = container_of(txq, struct sge_eth_txq, q);
-			netif_tx_wake_queue(eq->txq);
+			t4_sge_eth_txq_egress_update(q->adap, eq, -1);
 		} else {
 			struct sge_uld_txq *oq;
 
@@ -529,15 +601,21 @@ static int fwevtq_handler(struct sge_rspq *q, const __be64 *rsp,
 			FW_PORT_CMD_ACTION_G(ntohl(pcmd->action_to_len16));
 
 		if (cmd == FW_PORT_CMD &&
-		    action == FW_PORT_ACTION_GET_PORT_INFO) {
+		    (action == FW_PORT_ACTION_GET_PORT_INFO ||
+		     action == FW_PORT_ACTION_GET_PORT_INFO32)) {
 			int port = FW_PORT_CMD_PORTID_G(
 					be32_to_cpu(pcmd->op_to_portid));
-			struct net_device *dev =
-				q->adap->port[q->adap->chan_map[port]];
-			int state_input = ((pcmd->u.info.dcbxdis_pkd &
-					    FW_PORT_CMD_DCBXDIS_F)
-					   ? CXGB4_DCB_INPUT_FW_DISABLED
-					   : CXGB4_DCB_INPUT_FW_ENABLED);
+			struct net_device *dev;
+			int dcbxdis, state_input;
+
+			dev = q->adap->port[q->adap->chan_map[port]];
+			dcbxdis = (action == FW_PORT_ACTION_GET_PORT_INFO
+			  ? !!(pcmd->u.info.dcbxdis_pkd & FW_PORT_CMD_DCBXDIS_F)
+			  : !!(be32_to_cpu(pcmd->u.info32.lstatus32_to_cbllen32)
+			       & FW_PORT_CMD_DCBXDIS32_F));
+			state_input = (dcbxdis
+				       ? CXGB4_DCB_INPUT_FW_DISABLED
+				       : CXGB4_DCB_INPUT_FW_ENABLED);
 
 			cxgb4_dcb_state_fsm(dev, state_input);
 		}
@@ -553,10 +631,26 @@ static int fwevtq_handler(struct sge_rspq *q, const __be64 *rsp,
 		const struct cpl_l2t_write_rpl *p = (void *)rsp;
 
 		do_l2t_write_rpl(q->adap, p);
+	} else if (opcode == CPL_SMT_WRITE_RPL) {
+		const struct cpl_smt_write_rpl *p = (void *)rsp;
+
+		do_smt_write_rpl(q->adap, p);
 	} else if (opcode == CPL_SET_TCB_RPL) {
 		const struct cpl_set_tcb_rpl *p = (void *)rsp;
 
 		filter_rpl(q->adap, p);
+	} else if (opcode == CPL_ACT_OPEN_RPL) {
+		const struct cpl_act_open_rpl *p = (void *)rsp;
+
+		hash_filter_rpl(q->adap, p);
+	} else if (opcode == CPL_ABORT_RPL_RSS) {
+		const struct cpl_abort_rpl_rss *p = (void *)rsp;
+
+		hash_del_filter_rpl(q->adap, p);
+	} else if (opcode == CPL_SRQ_TABLE_RPL) {
+		const struct cpl_srq_table_rpl *p = (void *)rsp;
+
+		do_srq_table_rpl(q->adap, p);
 	} else
 		dev_err(q->adap->pdev_dev,
 			"unexpected CPL %#x on FW event queue\n", opcode);
@@ -566,12 +660,12 @@ out:
 
 static void disable_msi(struct adapter *adapter)
 {
-	if (adapter->flags & USING_MSIX) {
+	if (adapter->flags & CXGB4_USING_MSIX) {
 		pci_disable_msix(adapter->pdev);
-		adapter->flags &= ~USING_MSIX;
-	} else if (adapter->flags & USING_MSI) {
+		adapter->flags &= ~CXGB4_USING_MSIX;
+	} else if (adapter->flags & CXGB4_USING_MSI) {
 		pci_disable_msi(adapter->pdev);
-		adapter->flags &= ~USING_MSI;
+		adapter->flags &= ~CXGB4_USING_MSI;
 	}
 }
 
@@ -587,74 +681,170 @@ static irqreturn_t t4_nondata_intr(int irq, void *cookie)
 		adap->swintr = 1;
 		t4_write_reg(adap, MYPF_REG(PL_PF_INT_CAUSE_A), v);
 	}
-	if (adap->flags & MASTER_PF)
+	if (adap->flags & CXGB4_MASTER_PF)
 		t4_slow_intr_handler(adap);
 	return IRQ_HANDLED;
 }
 
-/*
- * Name the MSI-X interrupts.
- */
-static void name_msix_vecs(struct adapter *adap)
+int cxgb4_set_msix_aff(struct adapter *adap, unsigned short vec,
+		       cpumask_var_t *aff_mask, int idx)
 {
-	int i, j, msi_idx = 2, n = sizeof(adap->msix_info[0].desc);
+	int rv;
 
-	/* non-data interrupts */
-	snprintf(adap->msix_info[0].desc, n, "%s", adap->port[0]->name);
+	if (!zalloc_cpumask_var(aff_mask, GFP_KERNEL)) {
+		dev_err(adap->pdev_dev, "alloc_cpumask_var failed\n");
+		return -ENOMEM;
+	}
 
-	/* FW events */
-	snprintf(adap->msix_info[1].desc, n, "%s-FWeventq",
-		 adap->port[0]->name);
+	cpumask_set_cpu(cpumask_local_spread(idx, dev_to_node(adap->pdev_dev)),
+			*aff_mask);
 
-	/* Ethernet queues */
-	for_each_port(adap, j) {
-		struct net_device *d = adap->port[j];
-		const struct port_info *pi = netdev_priv(d);
+	rv = irq_set_affinity_hint(vec, *aff_mask);
+	if (rv)
+		dev_warn(adap->pdev_dev,
+			 "irq_set_affinity_hint %u failed %d\n",
+			 vec, rv);
 
-		for (i = 0; i < pi->nqsets; i++, msi_idx++)
-			snprintf(adap->msix_info[msi_idx].desc, n, "%s-Rx%d",
-				 d->name, i);
-	}
+	return 0;
+}
+
+void cxgb4_clear_msix_aff(unsigned short vec, cpumask_var_t aff_mask)
+{
+	irq_set_affinity_hint(vec, NULL);
+	free_cpumask_var(aff_mask);
 }
 
 static int request_msix_queue_irqs(struct adapter *adap)
 {
 	struct sge *s = &adap->sge;
+	struct msix_info *minfo;
 	int err, ethqidx;
-	int msi_index = 2;
 
-	err = request_irq(adap->msix_info[1].vec, t4_sge_intr_msix, 0,
-			  adap->msix_info[1].desc, &s->fw_evtq);
+	if (s->fwevtq_msix_idx < 0)
+		return -ENOMEM;
+
+	err = request_irq(adap->msix_info[s->fwevtq_msix_idx].vec,
+			  t4_sge_intr_msix, 0,
+			  adap->msix_info[s->fwevtq_msix_idx].desc,
+			  &s->fw_evtq);
 	if (err)
 		return err;
 
 	for_each_ethrxq(s, ethqidx) {
-		err = request_irq(adap->msix_info[msi_index].vec,
+		minfo = s->ethrxq[ethqidx].msix;
+		err = request_irq(minfo->vec,
 				  t4_sge_intr_msix, 0,
-				  adap->msix_info[msi_index].desc,
+				  minfo->desc,
 				  &s->ethrxq[ethqidx].rspq);
 		if (err)
 			goto unwind;
-		msi_index++;
+
+		cxgb4_set_msix_aff(adap, minfo->vec,
+				   &minfo->aff_mask, ethqidx);
 	}
 	return 0;
 
 unwind:
-	while (--ethqidx >= 0)
-		free_irq(adap->msix_info[--msi_index].vec,
-			 &s->ethrxq[ethqidx].rspq);
-	free_irq(adap->msix_info[1].vec, &s->fw_evtq);
+	while (--ethqidx >= 0) {
+		minfo = s->ethrxq[ethqidx].msix;
+		cxgb4_clear_msix_aff(minfo->vec, minfo->aff_mask);
+		free_irq(minfo->vec, &s->ethrxq[ethqidx].rspq);
+	}
+	free_irq(adap->msix_info[s->fwevtq_msix_idx].vec, &s->fw_evtq);
 	return err;
 }
 
 static void free_msix_queue_irqs(struct adapter *adap)
 {
-	int i, msi_index = 2;
 	struct sge *s = &adap->sge;
+	struct msix_info *minfo;
+	int i;
+
+	free_irq(adap->msix_info[s->fwevtq_msix_idx].vec, &s->fw_evtq);
+	for_each_ethrxq(s, i) {
+		minfo = s->ethrxq[i].msix;
+		cxgb4_clear_msix_aff(minfo->vec, minfo->aff_mask);
+		free_irq(minfo->vec, &s->ethrxq[i].rspq);
+	}
+}
+
+static int setup_ppod_edram(struct adapter *adap)
+{
+	unsigned int param, val;
+	int ret;
+
+	/* Driver sends FW_PARAMS_PARAM_DEV_PPOD_EDRAM read command to check
+	 * if firmware supports ppod edram feature or not. If firmware
+	 * returns 1, then driver can enable this feature by sending
+	 * FW_PARAMS_PARAM_DEV_PPOD_EDRAM write command with value 1 to
+	 * enable ppod edram feature.
+	 */
+	param = (FW_PARAMS_MNEM_V(FW_PARAMS_MNEM_DEV) |
+		FW_PARAMS_PARAM_X_V(FW_PARAMS_PARAM_DEV_PPOD_EDRAM));
+
+	ret = t4_query_params(adap, adap->mbox, adap->pf, 0, 1, &param, &val);
+	if (ret < 0) {
+		dev_warn(adap->pdev_dev,
+			 "querying PPOD_EDRAM support failed: %d\n",
+			 ret);
+		return -1;
+	}
+
+	if (val != 1)
+		return -1;
+
+	ret = t4_set_params(adap, adap->mbox, adap->pf, 0, 1, &param, &val);
+	if (ret < 0) {
+		dev_err(adap->pdev_dev,
+			"setting PPOD_EDRAM failed: %d\n", ret);
+		return -1;
+	}
+	return 0;
+}
+
+static void adap_config_hpfilter(struct adapter *adapter)
+{
+	u32 param, val = 0;
+	int ret;
+
+	/* Enable HP filter region. Older fw will fail this request and
+	 * it is fine.
+	 */
+	param = FW_PARAM_DEV(HPFILTER_REGION_SUPPORT);
+	ret = t4_set_params(adapter, adapter->mbox, adapter->pf, 0,
+			    1, &param, &val);
+
+	/* An error means FW doesn't know about HP filter support,
+	 * it's not a problem, don't return an error.
+	 */
+	if (ret < 0)
+		dev_err(adapter->pdev_dev,
+			"HP filter region isn't supported by FW\n");
+}
+
+static int cxgb4_config_rss(const struct port_info *pi, u16 *rss,
+			    u16 rss_size, u16 viid)
+{
+	struct adapter *adap = pi->adapter;
+	int ret;
+
+	ret = t4_config_rss_range(adap, adap->mbox, viid, 0, rss_size, rss,
+				  rss_size);
+	if (ret)
+		return ret;
 
-	free_irq(adap->msix_info[1].vec, &s->fw_evtq);
-	for_each_ethrxq(s, i)
-		free_irq(adap->msix_info[msi_index++].vec, &s->ethrxq[i].rspq);
+	/* If Tunnel All Lookup isn't specified in the global RSS
+	 * Configuration, then we need to specify a default Ingress
+	 * Queue for any ingress packets which aren't hashed.  We'll
+	 * use our first ingress queue ...
+	 */
+	return t4_config_vi_rss(adap, adap->mbox, viid,
+				FW_RSS_VI_CONFIG_CMD_IP6FOURTUPEN_F |
+				FW_RSS_VI_CONFIG_CMD_IP6TWOTUPEN_F |
+				FW_RSS_VI_CONFIG_CMD_IP4FOURTUPEN_F |
+				FW_RSS_VI_CONFIG_CMD_IP4TWOTUPEN_F |
+				FW_RSS_VI_CONFIG_CMD_UDPEN_F,
+				rss[0]);
 }
 
 /**
@@ -668,13 +858,13 @@ static void free_msix_queue_irqs(struct adapter *adap)
  */
 int cxgb4_write_rss(const struct port_info *pi, const u16 *queues)
 {
-	u16 *rss;
-	int i, err;
 	struct adapter *adapter = pi->adapter;
 	const struct sge_eth_rxq *rxq;
+	int i, err;
+	u16 *rss;
 
 	rxq = &adapter->sge.ethrxq[pi->first_qset];
-	rss = kmalloc(pi->rss_size * sizeof(u16), GFP_KERNEL);
+	rss = kmalloc_array(pi->rss_size, sizeof(u16), GFP_KERNEL);
 	if (!rss)
 		return -ENOMEM;
 
@@ -682,21 +872,7 @@ int cxgb4_write_rss(const struct port_info *pi, const u16 *queues)
 	for (i = 0; i < pi->rss_size; i++, queues++)
 		rss[i] = rxq[*queues].rspq.abs_id;
 
-	err = t4_config_rss_range(adapter, adapter->pf, pi->viid, 0,
-				  pi->rss_size, rss, pi->rss_size);
-	/* If Tunnel All Lookup isn't specified in the global RSS
-	 * Configuration, then we need to specify a default Ingress
-	 * Queue for any ingress packets which aren't hashed.  We'll
-	 * use our first ingress queue ...
-	 */
-	if (!err)
-		err = t4_config_vi_rss(adapter, adapter->mbox, pi->viid,
-				       FW_RSS_VI_CONFIG_CMD_IP6FOURTUPEN_F |
-				       FW_RSS_VI_CONFIG_CMD_IP6TWOTUPEN_F |
-				       FW_RSS_VI_CONFIG_CMD_IP4FOURTUPEN_F |
-				       FW_RSS_VI_CONFIG_CMD_IP4TWOTUPEN_F |
-				       FW_RSS_VI_CONFIG_CMD_UDPEN_F,
-				       rss[0]);
+	err = cxgb4_config_rss(pi, rss, pi->rss_size, pi->viid);
 	kfree(rss);
 	return err;
 }
@@ -734,6 +910,12 @@ static unsigned int rxq_to_chan(const struct sge *p, unsigned int qid)
 	return netdev2pinfo(p->ingr_map[qid]->netdev)->tx_chan;
 }
 
+void cxgb4_quiesce_rx(struct sge_rspq *q)
+{
+	if (q->handler)
+		napi_disable(&q->napi);
+}
+
 /*
  * Wait until all NAPI handlers are descheduled.
  */
@@ -744,19 +926,24 @@ static void quiesce_rx(struct adapter *adap)
 	for (i = 0; i < adap->sge.ingr_sz; i++) {
 		struct sge_rspq *q = adap->sge.ingr_map[i];
 
-		if (q && q->handler)
-			napi_disable(&q->napi);
+		if (!q)
+			continue;
+
+		cxgb4_quiesce_rx(q);
 	}
 }
 
 /* Disable interrupt and napi handler */
 static void disable_interrupts(struct adapter *adap)
 {
-	if (adap->flags & FULL_INIT_DONE) {
+	struct sge *s = &adap->sge;
+
+	if (adap->flags & CXGB4_FULL_INIT_DONE) {
 		t4_intr_disable(adap);
-		if (adap->flags & USING_MSIX) {
+		if (adap->flags & CXGB4_USING_MSIX) {
 			free_msix_queue_irqs(adap);
-			free_irq(adap->msix_info[0].vec, adap);
+			free_irq(adap->msix_info[s->nd_msix_idx].vec,
+				 adap);
 		} else {
 			free_irq(adap->pdev->irq, adap);
 		}
@@ -764,6 +951,17 @@ static void disable_interrupts(struct adapter *adap)
 	}
 }
 
+void cxgb4_enable_rx(struct adapter *adap, struct sge_rspq *q)
+{
+	if (q->handler)
+		napi_enable(&q->napi);
+
+	/* 0-increment GTS to start the timer and enable interrupts */
+	t4_write_reg(adap, MYPF_REG(SGE_PF_GTS_A),
+		     SEINTARM_V(q->intr_params) |
+		     INGRESSQID_V(q->cntxt_id));
+}
+
 /*
  * Enable NAPI scheduling and interrupt generation for all Rx queues.
  */
@@ -776,39 +974,63 @@ static void enable_rx(struct adapter *adap)
 
 		if (!q)
 			continue;
-		if (q->handler)
-			napi_enable(&q->napi);
 
-		/* 0-increment GTS to start the timer and enable interrupts */
-		t4_write_reg(adap, MYPF_REG(SGE_PF_GTS_A),
-			     SEINTARM_V(q->intr_params) |
-			     INGRESSQID_V(q->cntxt_id));
+		cxgb4_enable_rx(adap, q);
 	}
 }
 
+static int setup_non_data_intr(struct adapter *adap)
+{
+	int msix;
+
+	adap->sge.nd_msix_idx = -1;
+	if (!(adap->flags & CXGB4_USING_MSIX))
+		return 0;
+
+	/* Request MSI-X vector for non-data interrupt */
+	msix = cxgb4_get_msix_idx_from_bmap(adap);
+	if (msix < 0)
+		return -ENOMEM;
+
+	snprintf(adap->msix_info[msix].desc,
+		 sizeof(adap->msix_info[msix].desc),
+		 "%s", adap->port[0]->name);
+
+	adap->sge.nd_msix_idx = msix;
+	return 0;
+}
 
 static int setup_fw_sge_queues(struct adapter *adap)
 {
 	struct sge *s = &adap->sge;
-	int err = 0;
+	int msix, err = 0;
 
 	bitmap_zero(s->starving_fl, s->egr_sz);
 	bitmap_zero(s->txq_maperr, s->egr_sz);
 
-	if (adap->flags & USING_MSIX)
-		adap->msi_idx = 1;         /* vector 0 is for non-queue interrupts */
-	else {
+	if (adap->flags & CXGB4_USING_MSIX) {
+		s->fwevtq_msix_idx = -1;
+		msix = cxgb4_get_msix_idx_from_bmap(adap);
+		if (msix < 0)
+			return -ENOMEM;
+
+		snprintf(adap->msix_info[msix].desc,
+			 sizeof(adap->msix_info[msix].desc),
+			 "%s-FWeventq", adap->port[0]->name);
+	} else {
 		err = t4_sge_alloc_rxq(adap, &s->intrq, false, adap->port[0], 0,
 				       NULL, NULL, NULL, -1);
 		if (err)
 			return err;
-		adap->msi_idx = -((int)s->intrq.abs_id + 1);
+		msix = -((int)s->intrq.abs_id + 1);
 	}
 
 	err = t4_sge_alloc_rxq(adap, &s->fw_evtq, true, adap->port[0],
-			       adap->msi_idx, NULL, fwevtq_handler, NULL, -1);
-	if (err)
-		t4_free_sge_resources(adap);
+			       msix, NULL, fwevtq_handler, NULL, -1);
+	if (err && msix >= 0)
+		cxgb4_free_msix_idx_in_bmap(adap, msix);
+
+	s->fwevtq_msix_idx = msix;
 	return err;
 }
 
@@ -822,10 +1044,16 @@ static int setup_fw_sge_queues(struct adapter *adap)
  */
 static int setup_sge_queues(struct adapter *adap)
 {
-	int err, i, j;
+	struct sge_uld_rxq_info *rxq_info = NULL;
 	struct sge *s = &adap->sge;
-	struct sge_uld_rxq_info *rxq_info = s->uld_rxq_info[CXGB4_ULD_RDMA];
 	unsigned int cmplqid = 0;
+	int err, i, j, msix = 0;
+
+	if (is_uld(adap))
+		rxq_info = s->uld_rxq_info[CXGB4_ULD_RDMA];
+
+	if (!(adap->flags & CXGB4_USING_MSIX))
+		msix = -((int)s->intrq.abs_id + 1);
 
 	for_each_port(adap, i) {
 		struct net_device *dev = adap->port[i];
@@ -834,23 +1062,37 @@ static int setup_sge_queues(struct adapter *adap)
 		struct sge_eth_txq *t = &s->ethtxq[pi->first_qset];
 
 		for (j = 0; j < pi->nqsets; j++, q++) {
-			if (adap->msi_idx > 0)
-				adap->msi_idx++;
+			if (msix >= 0) {
+				msix = cxgb4_get_msix_idx_from_bmap(adap);
+				if (msix < 0) {
+					err = msix;
+					goto freeout;
+				}
+
+				snprintf(adap->msix_info[msix].desc,
+					 sizeof(adap->msix_info[msix].desc),
+					 "%s-Rx%d", dev->name, j);
+				q->msix = &adap->msix_info[msix];
+			}
+
 			err = t4_sge_alloc_rxq(adap, &q->rspq, false, dev,
-					       adap->msi_idx, &q->fl,
+					       msix, &q->fl,
 					       t4_ethrx_handler,
 					       NULL,
-					       t4_get_mps_bg_map(adap,
-								 pi->tx_chan));
+					       t4_get_tp_ch_map(adap,
+								pi->tx_chan));
 			if (err)
 				goto freeout;
 			q->rspq.idx = j;
 			memset(&q->stats, 0, sizeof(q->stats));
 		}
-		for (j = 0; j < pi->nqsets; j++, t++) {
+
+		q = &s->ethrxq[pi->first_qset];
+		for (j = 0; j < pi->nqsets; j++, t++, q++) {
 			err = t4_sge_alloc_eth_txq(adap, t, dev,
 					netdev_get_tx_queue(dev, j),
-					s->fw_evtq.cntxt_id);
+					q->rspq.cntxt_id,
+					!!(adap->flags & CXGB4_SGE_DBQ_TIMER));
 			if (err)
 				goto freeout;
 		}
@@ -869,6 +1111,14 @@ static int setup_sge_queues(struct adapter *adap)
 			goto freeout;
 	}
 
+	if (!is_t4(adap->params.chip)) {
+		err = t4_sge_alloc_eth_txq(adap, &s->ptptxq, adap->port[0],
+					   netdev_get_tx_queue(adap->port[0], 0)
+					   , s->fw_evtq.cntxt_id, false);
+		if (err)
+			goto freeout;
+	}
+
 	t4_write_reg(adap, is_t4(adap->params.chip) ?
 				MPS_TRC_RSS_CONTROL_A :
 				MPS_T5_TRC_RSS_CONTROL_A,
@@ -876,12 +1126,13 @@ static int setup_sge_queues(struct adapter *adap)
 		     QUEUENUMBER_V(s->ethrxq[0].rspq.abs_id));
 	return 0;
 freeout:
+	dev_err(adap->pdev_dev, "Can't allocate queues, err=%d\n", -err);
 	t4_free_sge_resources(adap);
 	return err;
 }
 
 static u16 cxgb_select_queue(struct net_device *dev, struct sk_buff *skb,
-			     void *accel_priv, select_queue_fallback_t fallback)
+			     struct net_device *sb_dev)
 {
 	int txq;
 
@@ -912,6 +1163,25 @@ static u16 cxgb_select_queue(struct net_device *dev, struct sk_buff *skb,
 	}
 #endif /* CONFIG_CHELSIO_T4_DCB */
 
+	if (dev->num_tc) {
+		struct port_info *pi = netdev2pinfo(dev);
+		u8 ver, proto;
+
+		ver = ip_hdr(skb)->version;
+		proto = (ver == 6) ? ipv6_hdr(skb)->nexthdr :
+				     ip_hdr(skb)->protocol;
+
+		/* Send unsupported traffic pattern to normal NIC queues. */
+		txq = netdev_pick_tx(dev, skb, sb_dev);
+		if (xfrm_offload(skb) || is_ptp_enabled(skb, dev) ||
+		    skb->encapsulation ||
+		    tls_is_skb_tx_device_offloaded(skb) ||
+		    (proto != IPPROTO_TCP && proto != IPPROTO_UDP))
+			txq = txq % pi->nqsets;
+
+		return txq;
+	}
+
 	if (select_queue) {
 		txq = (skb_rx_queue_recorded(skb)
 			? skb_get_rx_queue(skb)
@@ -923,7 +1193,7 @@ static u16 cxgb_select_queue(struct net_device *dev, struct sk_buff *skb,
 		return txq;
 	}
 
-	return fallback(dev, skb) % dev->real_num_tx_queues;
+	return netdev_pick_tx(dev, skb, NULL) % dev->real_num_tx_queues;
 }
 
 static int closest_timer(const struct sge *s, int time)
@@ -1001,15 +1271,15 @@ int cxgb4_set_rspq_intr_params(struct sge_rspq *q,
 
 static int cxgb_set_features(struct net_device *dev, netdev_features_t features)
 {
-	const struct port_info *pi = netdev_priv(dev);
 	netdev_features_t changed = dev->features ^ features;
+	const struct port_info *pi = netdev_priv(dev);
 	int err;
 
 	if (!(changed & NETIF_F_HW_VLAN_CTAG_RX))
 		return 0;
 
-	err = t4_set_rxmode(pi->adapter, pi->adapter->pf, pi->viid, -1,
-			    -1, -1, -1,
+	err = t4_set_rxmode(pi->adapter, pi->adapter->mbox, pi->viid,
+			    pi->viid_mirror, -1, -1, -1, -1,
 			    !!(features & NETIF_F_HW_VLAN_CTAG_RX), true);
 	if (unlikely(err))
 		dev->features = features ^ NETIF_F_HW_VLAN_CTAG_RX;
@@ -1027,6 +1297,292 @@ static int setup_debugfs(struct adapter *adap)
 	return 0;
 }
 
+static void cxgb4_port_mirror_free_rxq(struct adapter *adap,
+				       struct sge_eth_rxq *mirror_rxq)
+{
+	if ((adap->flags & CXGB4_FULL_INIT_DONE) &&
+	    !(adap->flags & CXGB4_SHUTTING_DOWN))
+		cxgb4_quiesce_rx(&mirror_rxq->rspq);
+
+	if (adap->flags & CXGB4_USING_MSIX) {
+		cxgb4_clear_msix_aff(mirror_rxq->msix->vec,
+				     mirror_rxq->msix->aff_mask);
+		free_irq(mirror_rxq->msix->vec, &mirror_rxq->rspq);
+		cxgb4_free_msix_idx_in_bmap(adap, mirror_rxq->msix->idx);
+	}
+
+	free_rspq_fl(adap, &mirror_rxq->rspq, &mirror_rxq->fl);
+}
+
+static int cxgb4_port_mirror_alloc_queues(struct net_device *dev)
+{
+	struct port_info *pi = netdev2pinfo(dev);
+	struct adapter *adap = netdev2adap(dev);
+	struct sge_eth_rxq *mirror_rxq;
+	struct sge *s = &adap->sge;
+	int ret = 0, msix = 0;
+	u16 i, rxqid;
+	u16 *rss;
+
+	if (!pi->vi_mirror_count)
+		return 0;
+
+	if (s->mirror_rxq[pi->port_id])
+		return 0;
+
+	mirror_rxq = kcalloc(pi->nmirrorqsets, sizeof(*mirror_rxq), GFP_KERNEL);
+	if (!mirror_rxq)
+		return -ENOMEM;
+
+	s->mirror_rxq[pi->port_id] = mirror_rxq;
+
+	if (!(adap->flags & CXGB4_USING_MSIX))
+		msix = -((int)adap->sge.intrq.abs_id + 1);
+
+	for (i = 0, rxqid = 0; i < pi->nmirrorqsets; i++, rxqid++) {
+		mirror_rxq = &s->mirror_rxq[pi->port_id][i];
+
+		/* Allocate Mirror Rxqs */
+		if (msix >= 0) {
+			msix = cxgb4_get_msix_idx_from_bmap(adap);
+			if (msix < 0) {
+				ret = msix;
+				goto out_free_queues;
+			}
+
+			mirror_rxq->msix = &adap->msix_info[msix];
+			snprintf(mirror_rxq->msix->desc,
+				 sizeof(mirror_rxq->msix->desc),
+				 "%s-mirrorrxq%d", dev->name, i);
+		}
+
+		init_rspq(adap, &mirror_rxq->rspq,
+			  CXGB4_MIRROR_RXQ_DEFAULT_INTR_USEC,
+			  CXGB4_MIRROR_RXQ_DEFAULT_PKT_CNT,
+			  CXGB4_MIRROR_RXQ_DEFAULT_DESC_NUM,
+			  CXGB4_MIRROR_RXQ_DEFAULT_DESC_SIZE);
+
+		mirror_rxq->fl.size = CXGB4_MIRROR_FLQ_DEFAULT_DESC_NUM;
+
+		ret = t4_sge_alloc_rxq(adap, &mirror_rxq->rspq, false,
+				       dev, msix, &mirror_rxq->fl,
+				       t4_ethrx_handler, NULL, 0);
+		if (ret)
+			goto out_free_msix_idx;
+
+		/* Setup MSI-X vectors for Mirror Rxqs */
+		if (adap->flags & CXGB4_USING_MSIX) {
+			ret = request_irq(mirror_rxq->msix->vec,
+					  t4_sge_intr_msix, 0,
+					  mirror_rxq->msix->desc,
+					  &mirror_rxq->rspq);
+			if (ret)
+				goto out_free_rxq;
+
+			cxgb4_set_msix_aff(adap, mirror_rxq->msix->vec,
+					   &mirror_rxq->msix->aff_mask, i);
+		}
+
+		/* Start NAPI for Mirror Rxqs */
+		cxgb4_enable_rx(adap, &mirror_rxq->rspq);
+	}
+
+	/* Setup RSS for Mirror Rxqs */
+	rss = kcalloc(pi->rss_size, sizeof(u16), GFP_KERNEL);
+	if (!rss) {
+		ret = -ENOMEM;
+		goto out_free_queues;
+	}
+
+	mirror_rxq = &s->mirror_rxq[pi->port_id][0];
+	for (i = 0; i < pi->rss_size; i++)
+		rss[i] = mirror_rxq[i % pi->nmirrorqsets].rspq.abs_id;
+
+	ret = cxgb4_config_rss(pi, rss, pi->rss_size, pi->viid_mirror);
+	kfree(rss);
+	if (ret)
+		goto out_free_queues;
+
+	return 0;
+
+out_free_rxq:
+	free_rspq_fl(adap, &mirror_rxq->rspq, &mirror_rxq->fl);
+
+out_free_msix_idx:
+	cxgb4_free_msix_idx_in_bmap(adap, mirror_rxq->msix->idx);
+
+out_free_queues:
+	while (rxqid-- > 0)
+		cxgb4_port_mirror_free_rxq(adap,
+					   &s->mirror_rxq[pi->port_id][rxqid]);
+
+	kfree(s->mirror_rxq[pi->port_id]);
+	s->mirror_rxq[pi->port_id] = NULL;
+	return ret;
+}
+
+static void cxgb4_port_mirror_free_queues(struct net_device *dev)
+{
+	struct port_info *pi = netdev2pinfo(dev);
+	struct adapter *adap = netdev2adap(dev);
+	struct sge *s = &adap->sge;
+	u16 i;
+
+	if (!pi->vi_mirror_count)
+		return;
+
+	if (!s->mirror_rxq[pi->port_id])
+		return;
+
+	for (i = 0; i < pi->nmirrorqsets; i++)
+		cxgb4_port_mirror_free_rxq(adap,
+					   &s->mirror_rxq[pi->port_id][i]);
+
+	kfree(s->mirror_rxq[pi->port_id]);
+	s->mirror_rxq[pi->port_id] = NULL;
+}
+
+static int cxgb4_port_mirror_start(struct net_device *dev)
+{
+	struct port_info *pi = netdev2pinfo(dev);
+	struct adapter *adap = netdev2adap(dev);
+	int ret, idx = -1;
+
+	if (!pi->vi_mirror_count)
+		return 0;
+
+	/* Mirror VIs can be created dynamically after stack had
+	 * already setup Rx modes like MTU, promisc, allmulti, etc.
+	 * on main VI. So, parse what the stack had setup on the
+	 * main VI and update the same on the mirror VI.
+	 */
+	ret = t4_set_rxmode(adap, adap->mbox, pi->viid, pi->viid_mirror,
+			    dev->mtu, (dev->flags & IFF_PROMISC) ? 1 : 0,
+			    (dev->flags & IFF_ALLMULTI) ? 1 : 0, 1,
+			    !!(dev->features & NETIF_F_HW_VLAN_CTAG_RX), true);
+	if (ret) {
+		dev_err(adap->pdev_dev,
+			"Failed start up Rx mode for Mirror VI 0x%x, ret: %d\n",
+			pi->viid_mirror, ret);
+		return ret;
+	}
+
+	/* Enable replication bit for the device's MAC address
+	 * in MPS TCAM, so that the packets for the main VI are
+	 * replicated to mirror VI.
+	 */
+	ret = cxgb4_update_mac_filt(pi, pi->viid_mirror, &idx,
+				    dev->dev_addr, true, NULL);
+	if (ret) {
+		dev_err(adap->pdev_dev,
+			"Failed updating MAC filter for Mirror VI 0x%x, ret: %d\n",
+			pi->viid_mirror, ret);
+		return ret;
+	}
+
+	/* Enabling a Virtual Interface can result in an interrupt
+	 * during the processing of the VI Enable command and, in some
+	 * paths, result in an attempt to issue another command in the
+	 * interrupt context. Thus, we disable interrupts during the
+	 * course of the VI Enable command ...
+	 */
+	local_bh_disable();
+	ret = t4_enable_vi_params(adap, adap->mbox, pi->viid_mirror, true, true,
+				  false);
+	local_bh_enable();
+	if (ret)
+		dev_err(adap->pdev_dev,
+			"Failed starting Mirror VI 0x%x, ret: %d\n",
+			pi->viid_mirror, ret);
+
+	return ret;
+}
+
+static void cxgb4_port_mirror_stop(struct net_device *dev)
+{
+	struct port_info *pi = netdev2pinfo(dev);
+	struct adapter *adap = netdev2adap(dev);
+
+	if (!pi->vi_mirror_count)
+		return;
+
+	t4_enable_vi_params(adap, adap->mbox, pi->viid_mirror, false, false,
+			    false);
+}
+
+int cxgb4_port_mirror_alloc(struct net_device *dev)
+{
+	struct port_info *pi = netdev2pinfo(dev);
+	struct adapter *adap = netdev2adap(dev);
+	int ret = 0;
+
+	if (!pi->nmirrorqsets)
+		return -EOPNOTSUPP;
+
+	mutex_lock(&pi->vi_mirror_mutex);
+	if (pi->viid_mirror) {
+		pi->vi_mirror_count++;
+		goto out_unlock;
+	}
+
+	ret = t4_init_port_mirror(pi, adap->mbox, pi->port_id, adap->pf, 0,
+				  &pi->viid_mirror);
+	if (ret)
+		goto out_unlock;
+
+	pi->vi_mirror_count = 1;
+
+	if (adap->flags & CXGB4_FULL_INIT_DONE) {
+		ret = cxgb4_port_mirror_alloc_queues(dev);
+		if (ret)
+			goto out_free_vi;
+
+		ret = cxgb4_port_mirror_start(dev);
+		if (ret)
+			goto out_free_queues;
+	}
+
+	mutex_unlock(&pi->vi_mirror_mutex);
+	return 0;
+
+out_free_queues:
+	cxgb4_port_mirror_free_queues(dev);
+
+out_free_vi:
+	pi->vi_mirror_count = 0;
+	t4_free_vi(adap, adap->mbox, adap->pf, 0, pi->viid_mirror);
+	pi->viid_mirror = 0;
+
+out_unlock:
+	mutex_unlock(&pi->vi_mirror_mutex);
+	return ret;
+}
+
+void cxgb4_port_mirror_free(struct net_device *dev)
+{
+	struct port_info *pi = netdev2pinfo(dev);
+	struct adapter *adap = netdev2adap(dev);
+
+	mutex_lock(&pi->vi_mirror_mutex);
+	if (!pi->viid_mirror)
+		goto out_unlock;
+
+	if (pi->vi_mirror_count > 1) {
+		pi->vi_mirror_count--;
+		goto out_unlock;
+	}
+
+	cxgb4_port_mirror_stop(dev);
+	cxgb4_port_mirror_free_queues(dev);
+
+	pi->vi_mirror_count = 0;
+	t4_free_vi(adap, adap->mbox, adap->pf, 0, pi->viid_mirror);
+	pi->viid_mirror = 0;
+
+out_unlock:
+	mutex_unlock(&pi->vi_mirror_mutex);
+}
+
 /*
  * upper-layer driver support
  */
@@ -1093,10 +1649,12 @@ int cxgb4_alloc_stid(struct tid_info *t, int family, void *data)
 		 * This is equivalent to 4 TIDs. With CLIP enabled it
 		 * needs 2 TIDs.
 		 */
-		if (family == PF_INET)
-			t->stids_in_use++;
-		else
+		if (family == PF_INET6) {
 			t->stids_in_use += 2;
+			t->v6_stids_in_use += 2;
+		} else {
+			t->stids_in_use++;
+		}
 	}
 	spin_unlock_bh(&t->stid_lock);
 	return stid;
@@ -1150,13 +1708,16 @@ void cxgb4_free_stid(struct tid_info *t, unsigned int stid, int family)
 		bitmap_release_region(t->stid_bmap, stid, 1);
 	t->stid_tab[stid].data = NULL;
 	if (stid < t->nstids) {
-		if (family == PF_INET)
-			t->stids_in_use--;
-		else
+		if (family == PF_INET6) {
 			t->stids_in_use -= 2;
+			t->v6_stids_in_use -= 2;
+		} else {
+			t->stids_in_use--;
+		}
 	} else {
 		t->sftids_in_use--;
 	}
+
 	spin_unlock_bh(&t->stid_lock);
 }
 EXPORT_SYMBOL(cxgb4_free_stid);
@@ -1170,7 +1731,7 @@ static void mk_tid_release(struct sk_buff *skb, unsigned int chan,
 	struct cpl_tid_release *req;
 
 	set_wr_txq(skb, CPL_PRIORITY_SETUP, chan);
-	req = (struct cpl_tid_release *)__skb_put(skb, sizeof(*req));
+	req = __skb_put(skb, sizeof(*req));
 	INIT_TP_WR(req, tid);
 	OPCODE_TID(req) = htonl(MK_OPCODE_TID(CPL_TID_RELEASE, tid));
 }
@@ -1182,8 +1743,8 @@ static void mk_tid_release(struct sk_buff *skb, unsigned int chan,
 static void cxgb4_queue_tid_release(struct tid_info *t, unsigned int chan,
 				    unsigned int tid)
 {
-	void **p = &t->tid_tab[tid];
 	struct adapter *adap = container_of(t, struct adapter, tids);
+	void **p = &t->tid_tab[tid - t->tid_base];
 
 	spin_lock_bh(&adap->tid_release_lock);
 	*p = adap->tid_release_head;
@@ -1232,19 +1793,31 @@ static void process_tid_release_list(struct work_struct *work)
  * Release a TID and inform HW.  If we are unable to allocate the release
  * message we defer to a work queue.
  */
-void cxgb4_remove_tid(struct tid_info *t, unsigned int chan, unsigned int tid)
+void cxgb4_remove_tid(struct tid_info *t, unsigned int chan, unsigned int tid,
+		      unsigned short family)
 {
-	struct sk_buff *skb;
 	struct adapter *adap = container_of(t, struct adapter, tids);
+	struct sk_buff *skb;
 
-	WARN_ON(tid >= t->ntids);
+	if (tid_out_of_range(&adap->tids, tid)) {
+		dev_err(adap->pdev_dev, "tid %d out of range\n", tid);
+		return;
+	}
 
-	if (t->tid_tab[tid]) {
-		t->tid_tab[tid] = NULL;
-		if (t->hash_base && (tid >= t->hash_base))
-			atomic_dec(&t->hash_tids_in_use);
-		else
-			atomic_dec(&t->tids_in_use);
+	if (t->tid_tab[tid - adap->tids.tid_base]) {
+		t->tid_tab[tid - adap->tids.tid_base] = NULL;
+		atomic_dec(&t->conns_in_use);
+		if (t->hash_base && (tid >= t->hash_base)) {
+			if (family == AF_INET6)
+				atomic_sub(2, &t->hash_tids_in_use);
+			else
+				atomic_dec(&t->hash_tids_in_use);
+		} else {
+			if (family == AF_INET6)
+				atomic_sub(2, &t->tids_in_use);
+			else
+				atomic_dec(&t->tids_in_use);
+		}
 	}
 
 	skb = alloc_skb(sizeof(struct cpl_tid_release), GFP_ATOMIC);
@@ -1264,19 +1837,27 @@ static int tid_init(struct tid_info *t)
 	struct adapter *adap = container_of(t, struct adapter, tids);
 	unsigned int max_ftids = t->nftids + t->nsftids;
 	unsigned int natids = t->natids;
+	unsigned int hpftid_bmap_size;
+	unsigned int eotid_bmap_size;
 	unsigned int stid_bmap_size;
 	unsigned int ftid_bmap_size;
 	size_t size;
 
 	stid_bmap_size = BITS_TO_LONGS(t->nstids + t->nsftids);
 	ftid_bmap_size = BITS_TO_LONGS(t->nftids);
+	hpftid_bmap_size = BITS_TO_LONGS(t->nhpftids);
+	eotid_bmap_size = BITS_TO_LONGS(t->neotids);
 	size = t->ntids * sizeof(*t->tid_tab) +
 	       natids * sizeof(*t->atid_tab) +
 	       t->nstids * sizeof(*t->stid_tab) +
 	       t->nsftids * sizeof(*t->stid_tab) +
 	       stid_bmap_size * sizeof(long) +
+	       t->nhpftids * sizeof(*t->hpftid_tab) +
+	       hpftid_bmap_size * sizeof(long) +
 	       max_ftids * sizeof(*t->ftid_tab) +
-	       ftid_bmap_size * sizeof(long);
+	       ftid_bmap_size * sizeof(long) +
+	       t->neotids * sizeof(*t->eotid_tab) +
+	       eotid_bmap_size * sizeof(long);
 
 	t->tid_tab = kvzalloc(size, GFP_KERNEL);
 	if (!t->tid_tab)
@@ -1285,18 +1866,25 @@ static int tid_init(struct tid_info *t)
 	t->atid_tab = (union aopen_entry *)&t->tid_tab[t->ntids];
 	t->stid_tab = (struct serv_entry *)&t->atid_tab[natids];
 	t->stid_bmap = (unsigned long *)&t->stid_tab[t->nstids + t->nsftids];
-	t->ftid_tab = (struct filter_entry *)&t->stid_bmap[stid_bmap_size];
+	t->hpftid_tab = (struct filter_entry *)&t->stid_bmap[stid_bmap_size];
+	t->hpftid_bmap = (unsigned long *)&t->hpftid_tab[t->nhpftids];
+	t->ftid_tab = (struct filter_entry *)&t->hpftid_bmap[hpftid_bmap_size];
 	t->ftid_bmap = (unsigned long *)&t->ftid_tab[max_ftids];
+	t->eotid_tab = (struct eotid_entry *)&t->ftid_bmap[ftid_bmap_size];
+	t->eotid_bmap = (unsigned long *)&t->eotid_tab[t->neotids];
 	spin_lock_init(&t->stid_lock);
 	spin_lock_init(&t->atid_lock);
 	spin_lock_init(&t->ftid_lock);
 
 	t->stids_in_use = 0;
+	t->v6_stids_in_use = 0;
 	t->sftids_in_use = 0;
 	t->afree = NULL;
 	t->atids_in_use = 0;
 	atomic_set(&t->tids_in_use, 0);
+	atomic_set(&t->conns_in_use, 0);
 	atomic_set(&t->hash_tids_in_use, 0);
+	atomic_set(&t->eotids_in_use, 0);
 
 	/* Setup the free list for atid_tab and clear the stid bitmap. */
 	if (natids) {
@@ -1311,8 +1899,13 @@ static int tid_init(struct tid_info *t)
 		if (!t->stid_base &&
 		    CHELSIO_CHIP_VERSION(adap->params.chip) <= CHELSIO_T5)
 			__set_bit(0, t->stid_bmap);
+
+		if (t->neotids)
+			bitmap_zero(t->eotid_bmap, t->neotids);
 	}
 
+	if (t->nhpftids)
+		bitmap_zero(t->hpftid_bmap, t->nhpftids);
 	bitmap_zero(t->ftid_bmap, t->nftids);
 	return 0;
 }
@@ -1323,6 +1916,7 @@ static int tid_init(struct tid_info *t)
  *	@stid: the server TID
  *	@sip: local IP address to bind server to
  *	@sport: the server's TCP port
+ *	@vlan: the VLAN header information
  *	@queue: queue to direct messages from this server to
  *
  *	Create an IP server for the given port and address.
@@ -1343,7 +1937,7 @@ int cxgb4_create_server(const struct net_device *dev, unsigned int stid,
 		return -ENOMEM;
 
 	adap = netdev2adap(dev);
-	req = (struct cpl_pass_open_req *)__skb_put(skb, sizeof(*req));
+	req = __skb_put(skb, sizeof(*req));
 	INIT_TP_WR(req, 0);
 	OPCODE_TID(req) = htonl(MK_OPCODE_TID(CPL_PASS_OPEN_REQ, stid));
 	req->local_port = sport;
@@ -1384,7 +1978,7 @@ int cxgb4_create_server6(const struct net_device *dev, unsigned int stid,
 		return -ENOMEM;
 
 	adap = netdev2adap(dev);
-	req = (struct cpl_pass_open_req6 *)__skb_put(skb, sizeof(*req));
+	req = __skb_put(skb, sizeof(*req));
 	INIT_TP_WR(req, 0);
 	OPCODE_TID(req) = htonl(MK_OPCODE_TID(CPL_PASS_OPEN_REQ6, stid));
 	req->local_port = sport;
@@ -1416,7 +2010,7 @@ int cxgb4_remove_server(const struct net_device *dev, unsigned int stid,
 	if (!skb)
 		return -ENOMEM;
 
-	req = (struct cpl_close_listsvr_req *)__skb_put(skb, sizeof(*req));
+	req = __skb_put(skb, sizeof(*req));
 	INIT_TP_WR(req, 0);
 	OPCODE_TID(req) = htonl(MK_OPCODE_TID(CPL_CLOSE_LISTSRV_REQ, stid));
 	req->reply_ctrl = htons(NO_REPLY_V(0) | (ipv6 ? LISTSVR_IPV6_V(1) :
@@ -1521,28 +2115,6 @@ unsigned int cxgb4_best_aligned_mtu(const unsigned short *mtus,
 EXPORT_SYMBOL(cxgb4_best_aligned_mtu);
 
 /**
- *	cxgb4_tp_smt_idx - Get the Source Mac Table index for this VI
- *	@chip: chip type
- *	@viid: VI id of the given port
- *
- *	Return the SMT index for this VI.
- */
-unsigned int cxgb4_tp_smt_idx(enum chip_type chip, unsigned int viid)
-{
-	/* In T4/T5, SMT contains 256 SMAC entries organized in
-	 * 128 rows of 2 entries each.
-	 * In T6, SMT contains 256 SMAC entries in 256 rows.
-	 * TODO: The below code needs to be updated when we add support
-	 * for 256 VFs.
-	 */
-	if (CHELSIO_CHIP_VERSION(chip) <= CHELSIO_T5)
-		return ((viid & 0x7f) << 1);
-	else
-		return (viid & 0x7f);
-}
-EXPORT_SYMBOL(cxgb4_tp_smt_idx);
-
-/**
  *	cxgb4_port_chan - get the HW channel of a port
  *	@dev: the net device for the port
  *
@@ -1554,6 +2126,18 @@ unsigned int cxgb4_port_chan(const struct net_device *dev)
 }
 EXPORT_SYMBOL(cxgb4_port_chan);
 
+/**
+ *      cxgb4_port_e2cchan - get the HW c-channel of a port
+ *      @dev: the net device for the port
+ *
+ *      Return the HW RX c-channel of the given port.
+ */
+unsigned int cxgb4_port_e2cchan(const struct net_device *dev)
+{
+	return netdev2pinfo(dev)->rx_cchan;
+}
+EXPORT_SYMBOL(cxgb4_port_e2cchan);
+
 unsigned int cxgb4_dbfifo_count(const struct net_device *dev, int lpfifo)
 {
 	struct adapter *adap = netdev2adap(dev);
@@ -1602,28 +2186,16 @@ void cxgb4_get_tcp_stats(struct pci_dev *pdev, struct tp_tcp_stats *v4,
 	struct adapter *adap = pci_get_drvdata(pdev);
 
 	spin_lock(&adap->stats_lock);
-	t4_tp_get_tcp_stats(adap, v4, v6);
+	t4_tp_get_tcp_stats(adap, v4, v6, false);
 	spin_unlock(&adap->stats_lock);
 }
 EXPORT_SYMBOL(cxgb4_get_tcp_stats);
 
-void cxgb4_iscsi_init(struct net_device *dev, unsigned int tag_mask,
-		      const unsigned int *pgsz_order)
-{
-	struct adapter *adap = netdev2adap(dev);
-
-	t4_write_reg(adap, ULP_RX_ISCSI_TAGMASK_A, tag_mask);
-	t4_write_reg(adap, ULP_RX_ISCSI_PSZ_A, HPZ0_V(pgsz_order[0]) |
-		     HPZ1_V(pgsz_order[1]) | HPZ2_V(pgsz_order[2]) |
-		     HPZ3_V(pgsz_order[3]));
-}
-EXPORT_SYMBOL(cxgb4_iscsi_init);
-
 int cxgb4_flush_eq_cache(struct net_device *dev)
 {
 	struct adapter *adap = netdev2adap(dev);
 
-	return t4_sge_ctxt_flush(adap, adap->mbox);
+	return t4_sge_ctxt_flush(adap, adap->mbox, CTXT_EGRESS);
 }
 EXPORT_SYMBOL(cxgb4_flush_eq_cache);
 
@@ -1680,10 +2252,11 @@ EXPORT_SYMBOL(cxgb4_sync_txq_pidx);
 
 int cxgb4_read_tpte(struct net_device *dev, u32 stag, __be32 *tpte)
 {
-	struct adapter *adap;
-	u32 offset, memtype, memaddr;
 	u32 edc0_size, edc1_size, mc0_size, mc1_size, size;
 	u32 edc0_end, edc1_end, mc0_end, mc1_end;
+	u32 offset, memtype, memaddr;
+	struct adapter *adap;
+	u32 hma_size = 0;
 	int ret;
 
 	adap = netdev2adap(dev);
@@ -1703,6 +2276,10 @@ int cxgb4_read_tpte(struct net_device *dev, u32 stag, __be32 *tpte)
 	size = t4_read_reg(adap, MA_EXT_MEMORY0_BAR_A);
 	mc0_size = EXT_MEM0_SIZE_G(size) << 20;
 
+	if (t4_read_reg(adap, MA_TARGET_MEM_ENABLE_A) & HMA_MUX_F) {
+		size = t4_read_reg(adap, MA_EXT_MEMORY1_BAR_A);
+		hma_size = EXT_MEM1_SIZE_G(size) << 20;
+	}
 	edc0_end = edc0_size;
 	edc1_end = edc0_end + edc1_size;
 	mc0_end = edc1_end + mc0_size;
@@ -1714,7 +2291,10 @@ int cxgb4_read_tpte(struct net_device *dev, u32 stag, __be32 *tpte)
 		memtype = MEM_EDC1;
 		memaddr = offset - edc0_end;
 	} else {
-		if (offset < mc0_end) {
+		if (hma_size && (offset < (edc1_end + hma_size))) {
+			memtype = MEM_HMA;
+			memaddr = offset - edc1_end;
+		} else if (offset < mc0_end) {
 			memtype = MEM_MC0;
 			memaddr = offset - edc1_end;
 		} else if (is_t5(adap->params.chip)) {
@@ -2053,14 +2633,17 @@ static void detach_ulds(struct adapter *adap)
 {
 	unsigned int i;
 
+	if (!is_uld(adap))
+		return;
+
 	mutex_lock(&uld_mutex);
 	list_del(&adap->list_node);
+
 	for (i = 0; i < CXGB4_ULD_MAX; i++)
-		if (adap->uld && adap->uld[i].handle) {
+		if (adap->uld && adap->uld[i].handle)
 			adap->uld[i].state_change(adap->uld[i].handle,
 					     CXGB4_STATE_DETACH);
-			adap->uld[i].handle = NULL;
-		}
+
 	if (netevent_registered && list_empty(&adapter_list)) {
 		unregister_netevent_notifier(&cxgb4_netevent_nb);
 		netevent_registered = false;
@@ -2169,50 +2752,62 @@ static void update_clip(const struct adapter *adap)
  */
 static int cxgb_up(struct adapter *adap)
 {
+	struct sge *s = &adap->sge;
 	int err;
 
+	mutex_lock(&uld_mutex);
 	err = setup_sge_queues(adap);
 	if (err)
-		goto out;
+		goto rel_lock;
 	err = setup_rss(adap);
 	if (err)
 		goto freeq;
 
-	if (adap->flags & USING_MSIX) {
-		name_msix_vecs(adap);
-		err = request_irq(adap->msix_info[0].vec, t4_nondata_intr, 0,
-				  adap->msix_info[0].desc, adap);
+	if (adap->flags & CXGB4_USING_MSIX) {
+		if (s->nd_msix_idx < 0) {
+			err = -ENOMEM;
+			goto irq_err;
+		}
+
+		err = request_irq(adap->msix_info[s->nd_msix_idx].vec,
+				  t4_nondata_intr, 0,
+				  adap->msix_info[s->nd_msix_idx].desc, adap);
 		if (err)
 			goto irq_err;
+
 		err = request_msix_queue_irqs(adap);
-		if (err) {
-			free_irq(adap->msix_info[0].vec, adap);
-			goto irq_err;
-		}
+		if (err)
+			goto irq_err_free_nd_msix;
 	} else {
 		err = request_irq(adap->pdev->irq, t4_intr_handler(adap),
-				  (adap->flags & USING_MSI) ? 0 : IRQF_SHARED,
+				  (adap->flags & CXGB4_USING_MSI) ? 0
+								  : IRQF_SHARED,
 				  adap->port[0]->name, adap);
 		if (err)
 			goto irq_err;
 	}
+
 	enable_rx(adap);
 	t4_sge_start(adap);
 	t4_intr_enable(adap);
-	adap->flags |= FULL_INIT_DONE;
+	adap->flags |= CXGB4_FULL_INIT_DONE;
+	mutex_unlock(&uld_mutex);
+
 	notify_ulds(adap, CXGB4_STATE_UP);
 #if IS_ENABLED(CONFIG_IPV6)
 	update_clip(adap);
 #endif
-	/* Initialize hash mac addr list*/
-	INIT_LIST_HEAD(&adap->mac_hlist);
- out:
 	return err;
- irq_err:
+
+irq_err_free_nd_msix:
+	free_irq(adap->msix_info[s->nd_msix_idx].vec, adap);
+irq_err:
 	dev_err(adap->pdev_dev, "request_irq failed, err %d\n", err);
- freeq:
+freeq:
 	t4_free_sge_resources(adap);
-	goto out;
+rel_lock:
+	mutex_unlock(&uld_mutex);
+	return err;
 }
 
 static void cxgb_down(struct adapter *adapter)
@@ -2225,7 +2820,8 @@ static void cxgb_down(struct adapter *adapter)
 
 	t4_sge_stop(adapter);
 	t4_free_sge_resources(adapter);
-	adapter->flags &= ~FULL_INIT_DONE;
+
+	adapter->flags &= ~CXGB4_FULL_INIT_DONE;
 }
 
 /*
@@ -2233,13 +2829,13 @@ static void cxgb_down(struct adapter *adapter)
  */
 static int cxgb_open(struct net_device *dev)
 {
-	int err;
 	struct port_info *pi = netdev_priv(dev);
 	struct adapter *adapter = pi->adapter;
+	int err;
 
 	netif_carrier_off(dev);
 
-	if (!(adapter->flags & FULL_INIT_DONE)) {
+	if (!(adapter->flags & CXGB4_FULL_INIT_DONE)) {
 		err = cxgb_up(adapter);
 		if (err < 0)
 			return err;
@@ -2253,8 +2849,29 @@ static int cxgb_open(struct net_device *dev)
 		return err;
 
 	err = link_start(dev);
-	if (!err)
-		netif_tx_start_all_queues(dev);
+	if (err)
+		return err;
+
+	if (pi->nmirrorqsets) {
+		mutex_lock(&pi->vi_mirror_mutex);
+		err = cxgb4_port_mirror_alloc_queues(dev);
+		if (err)
+			goto out_unlock;
+
+		err = cxgb4_port_mirror_start(dev);
+		if (err)
+			goto out_free_queues;
+		mutex_unlock(&pi->vi_mirror_mutex);
+	}
+
+	netif_tx_start_all_queues(dev);
+	return 0;
+
+out_free_queues:
+	cxgb4_port_mirror_free_queues(dev);
+
+out_unlock:
+	mutex_unlock(&pi->vi_mirror_mutex);
 	return err;
 }
 
@@ -2262,10 +2879,27 @@ static int cxgb_close(struct net_device *dev)
 {
 	struct port_info *pi = netdev_priv(dev);
 	struct adapter *adapter = pi->adapter;
+	int ret;
 
 	netif_tx_stop_all_queues(dev);
 	netif_carrier_off(dev);
-	return t4_enable_vi(adapter, adapter->pf, pi->viid, false, false);
+	ret = t4_enable_pi_params(adapter, adapter->pf, pi,
+				  false, false, false);
+#ifdef CONFIG_CHELSIO_T4_DCB
+	cxgb4_dcb_reset(dev);
+	dcb_tx_queue_prio_enable(dev, false);
+#endif
+	if (ret)
+		return ret;
+
+	if (pi->nmirrorqsets) {
+		mutex_lock(&pi->vi_mirror_mutex);
+		cxgb4_port_mirror_stop(dev);
+		cxgb4_port_mirror_free_queues(dev);
+		mutex_unlock(&pi->vi_mirror_mutex);
+	}
+
+	return 0;
 }
 
 int cxgb4_create_server_filter(const struct net_device *dev, unsigned int stid,
@@ -2299,7 +2933,7 @@ int cxgb4_create_server_filter(const struct net_device *dev, unsigned int stid,
 
 	/* Clear out filter specifications */
 	memset(&f->fs, 0, sizeof(struct ch_filter_specification));
-	f->fs.val.lport = cpu_to_be16(sport);
+	f->fs.val.lport = be16_to_cpu(sport);
 	f->fs.mask.lport  = ~0;
 	val = (u8 *)&sip;
 	if ((val[0] | val[1] | val[2] | val[3]) != 0) {
@@ -2413,6 +3047,7 @@ static int cxgb_ioctl(struct net_device *dev, struct ifreq *req, int cmd)
 	unsigned int mbox;
 	int ret = 0, prtad, devad;
 	struct port_info *pi = netdev_priv(dev);
+	struct adapter *adapter = pi->adapter;
 	struct mii_ioctl_data *data = (struct mii_ioctl_data *)&req->ifr_data;
 
 	switch (cmd) {
@@ -2450,18 +3085,69 @@ static int cxgb_ioctl(struct net_device *dev, struct ifreq *req, int cmd)
 				   sizeof(pi->tstamp_config)))
 			return -EFAULT;
 
-		switch (pi->tstamp_config.rx_filter) {
-		case HWTSTAMP_FILTER_NONE:
+		if (!is_t4(adapter->params.chip)) {
+			switch (pi->tstamp_config.tx_type) {
+			case HWTSTAMP_TX_OFF:
+			case HWTSTAMP_TX_ON:
+				break;
+			default:
+				return -ERANGE;
+			}
+
+			switch (pi->tstamp_config.rx_filter) {
+			case HWTSTAMP_FILTER_NONE:
+				pi->rxtstamp = false;
+				break;
+			case HWTSTAMP_FILTER_PTP_V1_L4_EVENT:
+			case HWTSTAMP_FILTER_PTP_V2_L4_EVENT:
+				cxgb4_ptprx_timestamping(pi, pi->port_id,
+							 PTP_TS_L4);
+				break;
+			case HWTSTAMP_FILTER_PTP_V2_EVENT:
+				cxgb4_ptprx_timestamping(pi, pi->port_id,
+							 PTP_TS_L2_L4);
+				break;
+			case HWTSTAMP_FILTER_ALL:
+			case HWTSTAMP_FILTER_PTP_V1_L4_SYNC:
+			case HWTSTAMP_FILTER_PTP_V1_L4_DELAY_REQ:
+			case HWTSTAMP_FILTER_PTP_V2_L4_SYNC:
+			case HWTSTAMP_FILTER_PTP_V2_L4_DELAY_REQ:
+				pi->rxtstamp = true;
+				break;
+			default:
+				pi->tstamp_config.rx_filter =
+					HWTSTAMP_FILTER_NONE;
+				return -ERANGE;
+			}
+
+			if ((pi->tstamp_config.tx_type == HWTSTAMP_TX_OFF) &&
+			    (pi->tstamp_config.rx_filter ==
+				HWTSTAMP_FILTER_NONE)) {
+				if (cxgb4_ptp_txtype(adapter, pi->port_id) >= 0)
+					pi->ptp_enable = false;
+			}
+
+			if (pi->tstamp_config.rx_filter !=
+				HWTSTAMP_FILTER_NONE) {
+				if (cxgb4_ptp_redirect_rx_packet(adapter,
+								 pi) >= 0)
+					pi->ptp_enable = true;
+			}
+		} else {
+			/* For T4 Adapters */
+			switch (pi->tstamp_config.rx_filter) {
+			case HWTSTAMP_FILTER_NONE:
 			pi->rxtstamp = false;
 			break;
-		case HWTSTAMP_FILTER_ALL:
+			case HWTSTAMP_FILTER_ALL:
 			pi->rxtstamp = true;
 			break;
-		default:
-			pi->tstamp_config.rx_filter = HWTSTAMP_FILTER_NONE;
+			default:
+			pi->tstamp_config.rx_filter =
+			HWTSTAMP_FILTER_NONE;
 			return -ERANGE;
+			}
 		}
-
 		return copy_to_user(req->ifr_data, &pi->tstamp_config,
 				    sizeof(pi->tstamp_config)) ?
 			-EFAULT : 0;
@@ -2479,18 +3165,18 @@ static void cxgb_set_rxmode(struct net_device *dev)
 
 static int cxgb_change_mtu(struct net_device *dev, int new_mtu)
 {
-	int ret;
 	struct port_info *pi = netdev_priv(dev);
+	int ret;
 
-	ret = t4_set_rxmode(pi->adapter, pi->adapter->pf, pi->viid, new_mtu, -1,
-			    -1, -1, -1, true);
+	ret = t4_set_rxmode(pi->adapter, pi->adapter->mbox, pi->viid,
+			    pi->viid_mirror, new_mtu, -1, -1, -1, -1, true);
 	if (!ret)
-		dev->mtu = new_mtu;
+		WRITE_ONCE(dev->mtu, new_mtu);
 	return ret;
 }
 
 #ifdef CONFIG_PCI_IOV
-static int dummy_open(struct net_device *dev)
+static int cxgb4_mgmt_open(struct net_device *dev)
 {
 	/* Turn carrier off since we don't have to transmit anything on this
 	 * interface.
@@ -2500,39 +3186,42 @@ static int dummy_open(struct net_device *dev)
 }
 
 /* Fill MAC address that will be assigned by the FW */
-static void fill_vf_station_mac_addr(struct adapter *adap)
+static void cxgb4_mgmt_fill_vf_station_mac_addr(struct adapter *adap)
 {
-	unsigned int i;
 	u8 hw_addr[ETH_ALEN], macaddr[ETH_ALEN];
+	unsigned int i, vf, nvfs;
+	u16 a, b;
 	int err;
 	u8 *na;
-	u16 a, b;
 
 	err = t4_get_raw_vpd_params(adap, &adap->params.vpd);
-	if (!err) {
-		na = adap->params.vpd.na;
-		for (i = 0; i < ETH_ALEN; i++)
-			hw_addr[i] = (hex2val(na[2 * i + 0]) * 16 +
-				      hex2val(na[2 * i + 1]));
-		a = (hw_addr[0] << 8) | hw_addr[1];
-		b = (hw_addr[1] << 8) | hw_addr[2];
-		a ^= b;
-		a |= 0x0200;    /* locally assigned Ethernet MAC address */
-		a &= ~0x0100;   /* not a multicast Ethernet MAC address */
-		macaddr[0] = a >> 8;
-		macaddr[1] = a & 0xff;
-
-		for (i = 2; i < 5; i++)
-			macaddr[i] = hw_addr[i + 1];
-
-		for (i = 0; i < adap->num_vfs; i++) {
-			macaddr[5] = adap->pf * 16 + i;
-			ether_addr_copy(adap->vfinfo[i].vf_mac_addr, macaddr);
-		}
+	if (err)
+		return;
+
+	na = adap->params.vpd.na;
+	for (i = 0; i < ETH_ALEN; i++)
+		hw_addr[i] = (hex2val(na[2 * i + 0]) * 16 +
+			      hex2val(na[2 * i + 1]));
+
+	a = (hw_addr[0] << 8) | hw_addr[1];
+	b = (hw_addr[1] << 8) | hw_addr[2];
+	a ^= b;
+	a |= 0x0200;    /* locally assigned Ethernet MAC address */
+	a &= ~0x0100;   /* not a multicast Ethernet MAC address */
+	macaddr[0] = a >> 8;
+	macaddr[1] = a & 0xff;
+
+	for (i = 2; i < 5; i++)
+		macaddr[i] = hw_addr[i + 1];
+
+	for (vf = 0, nvfs = pci_sriov_get_totalvfs(adap->pdev);
+		vf < nvfs; vf++) {
+		macaddr[5] = adap->pf * nvfs + vf;
+		ether_addr_copy(adap->vfinfo[vf].vf_mac_addr, macaddr);
 	}
 }
 
-static int cxgb_set_vf_mac(struct net_device *dev, int vf, u8 *mac)
+static int cxgb4_mgmt_set_vf_mac(struct net_device *dev, int vf, u8 *mac)
 {
 	struct port_info *pi = netdev_priv(dev);
 	struct adapter *adap = pi->adapter;
@@ -2548,27 +3237,34 @@ static int cxgb_set_vf_mac(struct net_device *dev, int vf, u8 *mac)
 
 	dev_info(pi->adapter->pdev_dev,
 		 "Setting MAC %pM on VF %d\n", mac, vf);
-	ret = t4_set_vf_mac_acl(adap, vf + 1, 1, mac);
+	ret = t4_set_vf_mac_acl(adap, vf + 1, pi->lport, 1, mac);
 	if (!ret)
 		ether_addr_copy(adap->vfinfo[vf].vf_mac_addr, mac);
 	return ret;
 }
 
-static int cxgb_get_vf_config(struct net_device *dev,
-			      int vf, struct ifla_vf_info *ivi)
+static int cxgb4_mgmt_get_vf_config(struct net_device *dev,
+				    int vf, struct ifla_vf_info *ivi)
 {
 	struct port_info *pi = netdev_priv(dev);
 	struct adapter *adap = pi->adapter;
+	struct vf_info *vfinfo;
 
 	if (vf >= adap->num_vfs)
 		return -EINVAL;
+	vfinfo = &adap->vfinfo[vf];
+
 	ivi->vf = vf;
-	ether_addr_copy(ivi->mac, adap->vfinfo[vf].vf_mac_addr);
+	ivi->max_tx_rate = vfinfo->tx_rate;
+	ivi->min_tx_rate = 0;
+	ether_addr_copy(ivi->mac, vfinfo->vf_mac_addr);
+	ivi->vlan = vfinfo->vlan;
+	ivi->linkstate = vfinfo->link_state;
 	return 0;
 }
 
-static int cxgb_get_phys_port_id(struct net_device *dev,
-				 struct netdev_phys_item_id *ppid)
+static int cxgb4_mgmt_get_phys_port_id(struct net_device *dev,
+				       struct netdev_phys_item_id *ppid)
 {
 	struct port_info *pi = netdev_priv(dev);
 	unsigned int phy_port_id;
@@ -2579,7 +3275,174 @@ static int cxgb_get_phys_port_id(struct net_device *dev,
 	return 0;
 }
 
-#endif
+static int cxgb4_mgmt_set_vf_rate(struct net_device *dev, int vf,
+				  int min_tx_rate, int max_tx_rate)
+{
+	struct port_info *pi = netdev_priv(dev);
+	struct adapter *adap = pi->adapter;
+	unsigned int link_ok, speed, mtu;
+	u32 fw_pfvf, fw_class;
+	int class_id = vf;
+	int ret;
+	u16 pktsize;
+
+	if (vf >= adap->num_vfs)
+		return -EINVAL;
+
+	if (min_tx_rate) {
+		dev_err(adap->pdev_dev,
+			"Min tx rate (%d) (> 0) for VF %d is Invalid.\n",
+			min_tx_rate, vf);
+		return -EINVAL;
+	}
+
+	if (max_tx_rate == 0) {
+		/* unbind VF to any Traffic Class */
+		fw_pfvf =
+		    (FW_PARAMS_MNEM_V(FW_PARAMS_MNEM_PFVF) |
+		     FW_PARAMS_PARAM_X_V(FW_PARAMS_PARAM_PFVF_SCHEDCLASS_ETH));
+		fw_class = 0xffffffff;
+		ret = t4_set_params(adap, adap->mbox, adap->pf, vf + 1, 1,
+				    &fw_pfvf, &fw_class);
+		if (ret) {
+			dev_err(adap->pdev_dev,
+				"Err %d in unbinding PF %d VF %d from TX Rate Limiting\n",
+				ret, adap->pf, vf);
+			return -EINVAL;
+		}
+		dev_info(adap->pdev_dev,
+			 "PF %d VF %d is unbound from TX Rate Limiting\n",
+			 adap->pf, vf);
+		adap->vfinfo[vf].tx_rate = 0;
+		return 0;
+	}
+
+	ret = t4_get_link_params(pi, &link_ok, &speed, &mtu);
+	if (ret != FW_SUCCESS) {
+		dev_err(adap->pdev_dev,
+			"Failed to get link information for VF %d\n", vf);
+		return -EINVAL;
+	}
+
+	if (!link_ok) {
+		dev_err(adap->pdev_dev, "Link down for VF %d\n", vf);
+		return -EINVAL;
+	}
+
+	if (max_tx_rate > speed) {
+		dev_err(adap->pdev_dev,
+			"Max tx rate %d for VF %d can't be > link-speed %u",
+			max_tx_rate, vf, speed);
+		return -EINVAL;
+	}
+
+	pktsize = mtu;
+	/* subtract ethhdr size and 4 bytes crc since, f/w appends it */
+	pktsize = pktsize - sizeof(struct ethhdr) - 4;
+	/* subtract ipv4 hdr size, tcp hdr size to get typical IPv4 MSS size */
+	pktsize = pktsize - sizeof(struct iphdr) - sizeof(struct tcphdr);
+	/* configure Traffic Class for rate-limiting */
+	ret = t4_sched_params(adap, SCHED_CLASS_TYPE_PACKET,
+			      SCHED_CLASS_LEVEL_CL_RL,
+			      SCHED_CLASS_MODE_CLASS,
+			      SCHED_CLASS_RATEUNIT_BITS,
+			      SCHED_CLASS_RATEMODE_ABS,
+			      pi->tx_chan, class_id, 0,
+			      max_tx_rate * 1000, 0, pktsize, 0);
+	if (ret) {
+		dev_err(adap->pdev_dev, "Err %d for Traffic Class config\n",
+			ret);
+		return -EINVAL;
+	}
+	dev_info(adap->pdev_dev,
+		 "Class %d with MSS %u configured with rate %u\n",
+		 class_id, pktsize, max_tx_rate);
+
+	/* bind VF to configured Traffic Class */
+	fw_pfvf = (FW_PARAMS_MNEM_V(FW_PARAMS_MNEM_PFVF) |
+		   FW_PARAMS_PARAM_X_V(FW_PARAMS_PARAM_PFVF_SCHEDCLASS_ETH));
+	fw_class = class_id;
+	ret = t4_set_params(adap, adap->mbox, adap->pf, vf + 1, 1, &fw_pfvf,
+			    &fw_class);
+	if (ret) {
+		dev_err(adap->pdev_dev,
+			"Err %d in binding PF %d VF %d to Traffic Class %d\n",
+			ret, adap->pf, vf, class_id);
+		return -EINVAL;
+	}
+	dev_info(adap->pdev_dev, "PF %d VF %d is bound to Class %d\n",
+		 adap->pf, vf, class_id);
+	adap->vfinfo[vf].tx_rate = max_tx_rate;
+	return 0;
+}
+
+static int cxgb4_mgmt_set_vf_vlan(struct net_device *dev, int vf,
+				  u16 vlan, u8 qos, __be16 vlan_proto)
+{
+	struct port_info *pi = netdev_priv(dev);
+	struct adapter *adap = pi->adapter;
+	int ret;
+
+	if (vf >= adap->num_vfs || vlan > 4095 || qos > 7)
+		return -EINVAL;
+
+	if (vlan_proto != htons(ETH_P_8021Q) || qos != 0)
+		return -EPROTONOSUPPORT;
+
+	ret = t4_set_vlan_acl(adap, adap->mbox, vf + 1, vlan);
+	if (!ret) {
+		adap->vfinfo[vf].vlan = vlan;
+		return 0;
+	}
+
+	dev_err(adap->pdev_dev, "Err %d %s VLAN ACL for PF/VF %d/%d\n",
+		ret, (vlan ? "setting" : "clearing"), adap->pf, vf);
+	return ret;
+}
+
+static int cxgb4_mgmt_set_vf_link_state(struct net_device *dev, int vf,
+					int link)
+{
+	struct port_info *pi = netdev_priv(dev);
+	struct adapter *adap = pi->adapter;
+	u32 param, val;
+	int ret = 0;
+
+	if (vf >= adap->num_vfs)
+		return -EINVAL;
+
+	switch (link) {
+	case IFLA_VF_LINK_STATE_AUTO:
+		val = FW_VF_LINK_STATE_AUTO;
+		break;
+
+	case IFLA_VF_LINK_STATE_ENABLE:
+		val = FW_VF_LINK_STATE_ENABLE;
+		break;
+
+	case IFLA_VF_LINK_STATE_DISABLE:
+		val = FW_VF_LINK_STATE_DISABLE;
+		break;
+
+	default:
+		return -EINVAL;
+	}
+
+	param = (FW_PARAMS_MNEM_V(FW_PARAMS_MNEM_PFVF) |
+		 FW_PARAMS_PARAM_X_V(FW_PARAMS_PARAM_PFVF_LINK_STATE));
+	ret = t4_set_params(adap, adap->mbox, adap->pf, vf + 1, 1,
+			    &param, &val);
+	if (ret) {
+		dev_err(adap->pdev_dev,
+			"Error %d in setting PF %d VF %d link state\n",
+			ret, adap->pf, vf);
+		return -EINVAL;
+	}
+
+	adap->vfinfo[vf].link_state = link;
+	return ret;
+}
+#endif /* CONFIG_PCI_IOV */
 
 static int cxgb_set_mac_addr(struct net_device *dev, void *p)
 {
@@ -2590,13 +3453,12 @@ static int cxgb_set_mac_addr(struct net_device *dev, void *p)
 	if (!is_valid_ether_addr(addr->sa_data))
 		return -EADDRNOTAVAIL;
 
-	ret = t4_change_mac(pi->adapter, pi->adapter->pf, pi->viid,
-			    pi->xact_addr_filt, addr->sa_data, true, true);
+	ret = cxgb4_update_mac_filt(pi, pi->viid, &pi->xact_addr_filt,
+				    addr->sa_data, true, &pi->smt_idx);
 	if (ret < 0)
 		return ret;
 
-	memcpy(dev->dev_addr, addr->sa_data, dev->addr_len);
-	pi->xact_addr_filt = ret;
+	eth_hw_addr_set(dev, addr->sa_data);
 	return 0;
 }
 
@@ -2606,7 +3468,7 @@ static void cxgb_netpoll(struct net_device *dev)
 	struct port_info *pi = netdev_priv(dev);
 	struct adapter *adap = pi->adapter;
 
-	if (adap->flags & USING_MSIX) {
+	if (adap->flags & CXGB4_USING_MSIX) {
 		int i;
 		struct sge_eth_rxq *rx = &adap->sge.ethrxq[pi->first_qset];
 
@@ -2621,9 +3483,9 @@ static int cxgb_set_tx_maxrate(struct net_device *dev, int index, u32 rate)
 {
 	struct port_info *pi = netdev_priv(dev);
 	struct adapter *adap = pi->adapter;
+	struct ch_sched_queue qe = { 0 };
+	struct ch_sched_params p = { 0 };
 	struct sched_class *e;
-	struct ch_sched_params p;
-	struct ch_sched_queue qe;
 	u32 req_rate;
 	int err = 0;
 
@@ -2633,21 +3495,30 @@ static int cxgb_set_tx_maxrate(struct net_device *dev, int index, u32 rate)
 	if (index < 0 || index > pi->nqsets - 1)
 		return -EINVAL;
 
-	if (!(adap->flags & FULL_INIT_DONE)) {
+	if (!(adap->flags & CXGB4_FULL_INIT_DONE)) {
 		dev_err(adap->pdev_dev,
 			"Failed to rate limit on queue %d. Link Down?\n",
 			index);
 		return -EINVAL;
 	}
 
+	qe.queue = index;
+	e = cxgb4_sched_queue_lookup(dev, &qe);
+	if (e && e->info.u.params.level != SCHED_CLASS_LEVEL_CL_RL) {
+		dev_err(adap->pdev_dev,
+			"Queue %u already bound to class %u of type: %u\n",
+			index, e->idx, e->info.u.params.level);
+		return -EBUSY;
+	}
+
 	/* Convert from Mbps to Kbps */
-	req_rate = rate << 10;
+	req_rate = rate * 1000;
 
-	/* Max rate is 10 Gbps */
-	if (req_rate >= SCHED_MAX_RATE_KBPS) {
+	/* Max rate is 100 Gbps */
+	if (req_rate > SCHED_MAX_RATE_KBPS) {
 		dev_err(adap->pdev_dev,
-			"Invalid rate %u Mbps, Max rate is %u Gbps\n",
-			rate, SCHED_MAX_RATE_KBPS);
+			"Invalid rate %u Mbps, Max rate is %u Mbps\n",
+			rate, SCHED_MAX_RATE_KBPS / 1000);
 		return -ERANGE;
 	}
 
@@ -2669,7 +3540,6 @@ static int cxgb_set_tx_maxrate(struct net_device *dev, int index, u32 rate)
 		return 0;
 
 	/* Fetch any available unused or matching scheduling class */
-	memset(&p, 0, sizeof(p));
 	p.type = SCHED_CLASS_TYPE_PACKET;
 	p.u.params.level    = SCHED_CLASS_LEVEL_CL_RL;
 	p.u.params.mode     = SCHED_CLASS_MODE_CLASS;
@@ -2698,35 +3568,280 @@ static int cxgb_set_tx_maxrate(struct net_device *dev, int index, u32 rate)
 	return err;
 }
 
-static int cxgb_setup_tc(struct net_device *dev, u32 handle, __be16 proto,
-			 struct tc_to_netdev *tc)
+static int cxgb_setup_tc_flower(struct net_device *dev,
+				struct flow_cls_offload *cls_flower)
+{
+	switch (cls_flower->command) {
+	case FLOW_CLS_REPLACE:
+		return cxgb4_tc_flower_replace(dev, cls_flower);
+	case FLOW_CLS_DESTROY:
+		return cxgb4_tc_flower_destroy(dev, cls_flower);
+	case FLOW_CLS_STATS:
+		return cxgb4_tc_flower_stats(dev, cls_flower);
+	default:
+		return -EOPNOTSUPP;
+	}
+}
+
+static int cxgb_setup_tc_cls_u32(struct net_device *dev,
+				 struct tc_cls_u32_offload *cls_u32)
+{
+	switch (cls_u32->command) {
+	case TC_CLSU32_NEW_KNODE:
+	case TC_CLSU32_REPLACE_KNODE:
+		return cxgb4_config_knode(dev, cls_u32);
+	case TC_CLSU32_DELETE_KNODE:
+		return cxgb4_delete_knode(dev, cls_u32);
+	default:
+		return -EOPNOTSUPP;
+	}
+}
+
+static int cxgb_setup_tc_matchall(struct net_device *dev,
+				  struct tc_cls_matchall_offload *cls_matchall,
+				  bool ingress)
 {
+	struct adapter *adap = netdev2adap(dev);
+
+	if (!adap->tc_matchall)
+		return -ENOMEM;
+
+	switch (cls_matchall->command) {
+	case TC_CLSMATCHALL_REPLACE:
+		return cxgb4_tc_matchall_replace(dev, cls_matchall, ingress);
+	case TC_CLSMATCHALL_DESTROY:
+		return cxgb4_tc_matchall_destroy(dev, cls_matchall, ingress);
+	case TC_CLSMATCHALL_STATS:
+		if (ingress)
+			return cxgb4_tc_matchall_stats(dev, cls_matchall);
+		break;
+	default:
+		break;
+	}
+
+	return -EOPNOTSUPP;
+}
+
+static int cxgb_setup_tc_block_ingress_cb(enum tc_setup_type type,
+					  void *type_data, void *cb_priv)
+{
+	struct net_device *dev = cb_priv;
 	struct port_info *pi = netdev2pinfo(dev);
 	struct adapter *adap = netdev2adap(dev);
 
-	if (!(adap->flags & FULL_INIT_DONE)) {
+	if (!(adap->flags & CXGB4_FULL_INIT_DONE)) {
 		dev_err(adap->pdev_dev,
 			"Failed to setup tc on port %d. Link Down?\n",
 			pi->port_id);
 		return -EINVAL;
 	}
 
-	if (TC_H_MAJ(handle) == TC_H_MAJ(TC_H_INGRESS) &&
-	    tc->type == TC_SETUP_CLSU32) {
-		switch (tc->cls_u32->command) {
-		case TC_CLSU32_NEW_KNODE:
-		case TC_CLSU32_REPLACE_KNODE:
-			return cxgb4_config_knode(dev, proto, tc->cls_u32);
-		case TC_CLSU32_DELETE_KNODE:
-			return cxgb4_delete_knode(dev, proto, tc->cls_u32);
-		default:
-			return -EOPNOTSUPP;
-		}
+	if (!tc_cls_can_offload_and_chain0(dev, type_data))
+		return -EOPNOTSUPP;
+
+	switch (type) {
+	case TC_SETUP_CLSU32:
+		return cxgb_setup_tc_cls_u32(dev, type_data);
+	case TC_SETUP_CLSFLOWER:
+		return cxgb_setup_tc_flower(dev, type_data);
+	case TC_SETUP_CLSMATCHALL:
+		return cxgb_setup_tc_matchall(dev, type_data, true);
+	default:
+		return -EOPNOTSUPP;
+	}
+}
+
+static int cxgb_setup_tc_block_egress_cb(enum tc_setup_type type,
+					 void *type_data, void *cb_priv)
+{
+	struct net_device *dev = cb_priv;
+	struct port_info *pi = netdev2pinfo(dev);
+	struct adapter *adap = netdev2adap(dev);
+
+	if (!(adap->flags & CXGB4_FULL_INIT_DONE)) {
+		dev_err(adap->pdev_dev,
+			"Failed to setup tc on port %d. Link Down?\n",
+			pi->port_id);
+		return -EINVAL;
+	}
+
+	if (!tc_cls_can_offload_and_chain0(dev, type_data))
+		return -EOPNOTSUPP;
+
+	switch (type) {
+	case TC_SETUP_CLSMATCHALL:
+		return cxgb_setup_tc_matchall(dev, type_data, false);
+	default:
+		break;
 	}
 
 	return -EOPNOTSUPP;
 }
 
+static int cxgb_setup_tc_mqprio(struct net_device *dev,
+				struct tc_mqprio_qopt_offload *mqprio)
+{
+	struct adapter *adap = netdev2adap(dev);
+
+	if (!is_ethofld(adap) || !adap->tc_mqprio)
+		return -ENOMEM;
+
+	return cxgb4_setup_tc_mqprio(dev, mqprio);
+}
+
+static LIST_HEAD(cxgb_block_cb_list);
+
+static int cxgb_setup_tc_block(struct net_device *dev,
+			       struct flow_block_offload *f)
+{
+	struct port_info *pi = netdev_priv(dev);
+	flow_setup_cb_t *cb;
+	bool ingress_only;
+
+	pi->tc_block_shared = f->block_shared;
+	if (f->binder_type == FLOW_BLOCK_BINDER_TYPE_CLSACT_EGRESS) {
+		cb = cxgb_setup_tc_block_egress_cb;
+		ingress_only = false;
+	} else {
+		cb = cxgb_setup_tc_block_ingress_cb;
+		ingress_only = true;
+	}
+
+	return flow_block_cb_setup_simple(f, &cxgb_block_cb_list,
+					  cb, pi, dev, ingress_only);
+}
+
+static int cxgb_setup_tc(struct net_device *dev, enum tc_setup_type type,
+			 void *type_data)
+{
+	switch (type) {
+	case TC_SETUP_QDISC_MQPRIO:
+		return cxgb_setup_tc_mqprio(dev, type_data);
+	case TC_SETUP_BLOCK:
+		return cxgb_setup_tc_block(dev, type_data);
+	default:
+		return -EOPNOTSUPP;
+	}
+}
+
+static int cxgb_udp_tunnel_unset_port(struct net_device *netdev,
+				      unsigned int table, unsigned int entry,
+				      struct udp_tunnel_info *ti)
+{
+	struct port_info *pi = netdev_priv(netdev);
+	struct adapter *adapter = pi->adapter;
+	u8 match_all_mac[] = { 0, 0, 0, 0, 0, 0 };
+	int ret = 0, i;
+
+	switch (ti->type) {
+	case UDP_TUNNEL_TYPE_VXLAN:
+		adapter->vxlan_port = 0;
+		t4_write_reg(adapter, MPS_RX_VXLAN_TYPE_A, 0);
+		break;
+	case UDP_TUNNEL_TYPE_GENEVE:
+		adapter->geneve_port = 0;
+		t4_write_reg(adapter, MPS_RX_GENEVE_TYPE_A, 0);
+		break;
+	default:
+		return -EINVAL;
+	}
+
+	/* Matchall mac entries can be deleted only after all tunnel ports
+	 * are brought down or removed.
+	 */
+	if (!adapter->rawf_cnt)
+		return 0;
+	for_each_port(adapter, i) {
+		pi = adap2pinfo(adapter, i);
+		ret = t4_free_raw_mac_filt(adapter, pi->viid,
+					   match_all_mac, match_all_mac,
+					   adapter->rawf_start + pi->port_id,
+					   1, pi->port_id, false);
+		if (ret < 0) {
+			netdev_info(netdev, "Failed to free mac filter entry, for port %d\n",
+				    i);
+			return ret;
+		}
+	}
+
+	return 0;
+}
+
+static int cxgb_udp_tunnel_set_port(struct net_device *netdev,
+				    unsigned int table, unsigned int entry,
+				    struct udp_tunnel_info *ti)
+{
+	struct port_info *pi = netdev_priv(netdev);
+	struct adapter *adapter = pi->adapter;
+	u8 match_all_mac[] = { 0, 0, 0, 0, 0, 0 };
+	int i, ret;
+
+	switch (ti->type) {
+	case UDP_TUNNEL_TYPE_VXLAN:
+		adapter->vxlan_port = ti->port;
+		t4_write_reg(adapter, MPS_RX_VXLAN_TYPE_A,
+			     VXLAN_V(be16_to_cpu(ti->port)) | VXLAN_EN_F);
+		break;
+	case UDP_TUNNEL_TYPE_GENEVE:
+		adapter->geneve_port = ti->port;
+		t4_write_reg(adapter, MPS_RX_GENEVE_TYPE_A,
+			     GENEVE_V(be16_to_cpu(ti->port)) | GENEVE_EN_F);
+		break;
+	default:
+		return -EINVAL;
+	}
+
+	/* Create a 'match all' mac filter entry for inner mac,
+	 * if raw mac interface is supported. Once the linux kernel provides
+	 * driver entry points for adding/deleting the inner mac addresses,
+	 * we will remove this 'match all' entry and fallback to adding
+	 * exact match filters.
+	 */
+	for_each_port(adapter, i) {
+		pi = adap2pinfo(adapter, i);
+
+		ret = t4_alloc_raw_mac_filt(adapter, pi->viid,
+					    match_all_mac,
+					    match_all_mac,
+					    adapter->rawf_start + pi->port_id,
+					    1, pi->port_id, false);
+		if (ret < 0) {
+			netdev_info(netdev, "Failed to allocate a mac filter entry, not adding port %d\n",
+				    be16_to_cpu(ti->port));
+			return ret;
+		}
+	}
+
+	return 0;
+}
+
+static const struct udp_tunnel_nic_info cxgb_udp_tunnels = {
+	.set_port	= cxgb_udp_tunnel_set_port,
+	.unset_port	= cxgb_udp_tunnel_unset_port,
+	.tables		= {
+		{ .n_entries = 1, .tunnel_types = UDP_TUNNEL_TYPE_VXLAN,  },
+		{ .n_entries = 1, .tunnel_types = UDP_TUNNEL_TYPE_GENEVE, },
+	},
+};
+
+static netdev_features_t cxgb_features_check(struct sk_buff *skb,
+					     struct net_device *dev,
+					     netdev_features_t features)
+{
+	struct port_info *pi = netdev_priv(dev);
+	struct adapter *adapter = pi->adapter;
+
+	if (CHELSIO_CHIP_VERSION(adapter->params.chip) < CHELSIO_T6)
+		return features;
+
+	/* Check if hw supports offload for this packet */
+	if (!skb->encapsulation || cxgb_encap_offload_supported(skb))
+		return features;
+
+	/* Offload is not supported for this encapsulated packet */
+	return features & ~(NETIF_F_CSUM_MASK | NETIF_F_GSO_MASK);
+}
+
 static netdev_features_t cxgb_fix_features(struct net_device *dev,
 					   netdev_features_t features)
 {
@@ -2740,14 +3855,14 @@ static netdev_features_t cxgb_fix_features(struct net_device *dev,
 static const struct net_device_ops cxgb4_netdev_ops = {
 	.ndo_open             = cxgb_open,
 	.ndo_stop             = cxgb_close,
-	.ndo_start_xmit       = t4_eth_xmit,
+	.ndo_start_xmit       = t4_start_xmit,
 	.ndo_select_queue     =	cxgb_select_queue,
 	.ndo_get_stats64      = cxgb_get_stats,
 	.ndo_set_rx_mode      = cxgb_set_rxmode,
 	.ndo_set_mac_address  = cxgb_set_mac_addr,
 	.ndo_set_features     = cxgb_set_features,
 	.ndo_validate_addr    = eth_validate_addr,
-	.ndo_do_ioctl         = cxgb_ioctl,
+	.ndo_eth_ioctl         = cxgb_ioctl,
 	.ndo_change_mtu       = cxgb_change_mtu,
 #ifdef CONFIG_NET_POLL_CONTROLLER
 	.ndo_poll_controller  = cxgb_netpoll,
@@ -2758,37 +3873,51 @@ static const struct net_device_ops cxgb4_netdev_ops = {
 #endif /* CONFIG_CHELSIO_T4_FCOE */
 	.ndo_set_tx_maxrate   = cxgb_set_tx_maxrate,
 	.ndo_setup_tc         = cxgb_setup_tc,
+	.ndo_features_check   = cxgb_features_check,
 	.ndo_fix_features     = cxgb_fix_features,
 };
 
 #ifdef CONFIG_PCI_IOV
 static const struct net_device_ops cxgb4_mgmt_netdev_ops = {
-	.ndo_open             = dummy_open,
-	.ndo_set_vf_mac       = cxgb_set_vf_mac,
-	.ndo_get_vf_config    = cxgb_get_vf_config,
-	.ndo_get_phys_port_id = cxgb_get_phys_port_id,
+	.ndo_open               = cxgb4_mgmt_open,
+	.ndo_set_vf_mac         = cxgb4_mgmt_set_vf_mac,
+	.ndo_get_vf_config      = cxgb4_mgmt_get_vf_config,
+	.ndo_set_vf_rate        = cxgb4_mgmt_set_vf_rate,
+	.ndo_get_phys_port_id   = cxgb4_mgmt_get_phys_port_id,
+	.ndo_set_vf_vlan        = cxgb4_mgmt_set_vf_vlan,
+	.ndo_set_vf_link_state	= cxgb4_mgmt_set_vf_link_state,
 };
-#endif
 
-static void get_drvinfo(struct net_device *dev, struct ethtool_drvinfo *info)
+static void cxgb4_mgmt_get_drvinfo(struct net_device *dev,
+				   struct ethtool_drvinfo *info)
 {
 	struct adapter *adapter = netdev2adap(dev);
 
-	strlcpy(info->driver, cxgb4_driver_name, sizeof(info->driver));
-	strlcpy(info->version, cxgb4_driver_version,
-		sizeof(info->version));
-	strlcpy(info->bus_info, pci_name(adapter->pdev),
+	strscpy(info->driver, cxgb4_driver_name, sizeof(info->driver));
+	strscpy(info->bus_info, pci_name(adapter->pdev),
 		sizeof(info->bus_info));
 }
 
 static const struct ethtool_ops cxgb4_mgmt_ethtool_ops = {
-	.get_drvinfo       = get_drvinfo,
+	.get_drvinfo       = cxgb4_mgmt_get_drvinfo,
 };
+#endif
+
+static void notify_fatal_err(struct work_struct *work)
+{
+	struct adapter *adap;
+
+	adap = container_of(work, struct adapter, fatal_err_notify_task);
+	notify_ulds(adap, CXGB4_STATE_FATAL_ERROR);
+}
 
 void t4_fatal_err(struct adapter *adap)
 {
 	int port;
 
+	if (pci_channel_offline(adap->pdev))
+		return;
+
 	/* Disable the SGE since ULDs are going to free resources that
 	 * could be exposed to the adapter.  RDMA MWs for example...
 	 */
@@ -2806,6 +3935,7 @@ void t4_fatal_err(struct adapter *adap)
 		netif_carrier_off(dev);
 	}
 	dev_alert(adap->pdev_dev, "encountered fatal error, adapter stopped\n");
+	queue_work(adap->workq, &adap->fatal_err_notify_task);
 }
 
 static void setup_memwin(struct adapter *adap)
@@ -2836,11 +3966,221 @@ static void setup_memwin_rdma(struct adapter *adap)
 	}
 }
 
+/* HMA Definitions */
+
+/* The maximum number of address that can be send in a single FW cmd */
+#define HMA_MAX_ADDR_IN_CMD	5
+
+#define HMA_PAGE_SIZE		PAGE_SIZE
+
+#define HMA_MAX_NO_FW_ADDRESS	(16 << 10)  /* FW supports 16K addresses */
+
+#define HMA_PAGE_ORDER					\
+	((HMA_PAGE_SIZE < HMA_MAX_NO_FW_ADDRESS) ?	\
+	ilog2(HMA_MAX_NO_FW_ADDRESS / HMA_PAGE_SIZE) : 0)
+
+/* The minimum and maximum possible HMA sizes that can be specified in the FW
+ * configuration(in units of MB).
+ */
+#define HMA_MIN_TOTAL_SIZE	1
+#define HMA_MAX_TOTAL_SIZE				\
+	(((HMA_PAGE_SIZE << HMA_PAGE_ORDER) *		\
+	  HMA_MAX_NO_FW_ADDRESS) >> 20)
+
+static void adap_free_hma_mem(struct adapter *adapter)
+{
+	struct scatterlist *iter;
+	struct page *page;
+	int i;
+
+	if (!adapter->hma.sgt)
+		return;
+
+	if (adapter->hma.flags & HMA_DMA_MAPPED_FLAG) {
+		dma_unmap_sg(adapter->pdev_dev, adapter->hma.sgt->sgl,
+			     adapter->hma.sgt->nents, DMA_BIDIRECTIONAL);
+		adapter->hma.flags &= ~HMA_DMA_MAPPED_FLAG;
+	}
+
+	for_each_sg(adapter->hma.sgt->sgl, iter,
+		    adapter->hma.sgt->orig_nents, i) {
+		page = sg_page(iter);
+		if (page)
+			__free_pages(page, HMA_PAGE_ORDER);
+	}
+
+	kfree(adapter->hma.phy_addr);
+	sg_free_table(adapter->hma.sgt);
+	kfree(adapter->hma.sgt);
+	adapter->hma.sgt = NULL;
+}
+
+static int adap_config_hma(struct adapter *adapter)
+{
+	struct scatterlist *sgl, *iter;
+	struct sg_table *sgt;
+	struct page *newpage;
+	unsigned int i, j, k;
+	u32 param, hma_size;
+	unsigned int ncmds;
+	size_t page_size;
+	u32 page_order;
+	int node, ret;
+
+	/* HMA is supported only for T6+ cards.
+	 * Avoid initializing HMA in kdump kernels.
+	 */
+	if (is_kdump_kernel() ||
+	    CHELSIO_CHIP_VERSION(adapter->params.chip) < CHELSIO_T6)
+		return 0;
+
+	/* Get the HMA region size required by fw */
+	param = (FW_PARAMS_MNEM_V(FW_PARAMS_MNEM_DEV) |
+		 FW_PARAMS_PARAM_X_V(FW_PARAMS_PARAM_DEV_HMA_SIZE));
+	ret = t4_query_params(adapter, adapter->mbox, adapter->pf, 0,
+			      1, &param, &hma_size);
+	/* An error means card has its own memory or HMA is not supported by
+	 * the firmware. Return without any errors.
+	 */
+	if (ret || !hma_size)
+		return 0;
+
+	if (hma_size < HMA_MIN_TOTAL_SIZE ||
+	    hma_size > HMA_MAX_TOTAL_SIZE) {
+		dev_err(adapter->pdev_dev,
+			"HMA size %uMB beyond bounds(%u-%lu)MB\n",
+			hma_size, HMA_MIN_TOTAL_SIZE, HMA_MAX_TOTAL_SIZE);
+		return -EINVAL;
+	}
+
+	page_size = HMA_PAGE_SIZE;
+	page_order = HMA_PAGE_ORDER;
+	adapter->hma.sgt = kzalloc(sizeof(*adapter->hma.sgt), GFP_KERNEL);
+	if (unlikely(!adapter->hma.sgt)) {
+		dev_err(adapter->pdev_dev, "HMA SG table allocation failed\n");
+		return -ENOMEM;
+	}
+	sgt = adapter->hma.sgt;
+	/* FW returned value will be in MB's
+	 */
+	sgt->orig_nents = (hma_size << 20) / (page_size << page_order);
+	if (sg_alloc_table(sgt, sgt->orig_nents, GFP_KERNEL)) {
+		dev_err(adapter->pdev_dev, "HMA SGL allocation failed\n");
+		kfree(adapter->hma.sgt);
+		adapter->hma.sgt = NULL;
+		return -ENOMEM;
+	}
+
+	sgl = adapter->hma.sgt->sgl;
+	node = dev_to_node(adapter->pdev_dev);
+	for_each_sg(sgl, iter, sgt->orig_nents, i) {
+		newpage = alloc_pages_node(node, __GFP_NOWARN | GFP_KERNEL |
+					   __GFP_ZERO, page_order);
+		if (!newpage) {
+			dev_err(adapter->pdev_dev,
+				"Not enough memory for HMA page allocation\n");
+			ret = -ENOMEM;
+			goto free_hma;
+		}
+		sg_set_page(iter, newpage, page_size << page_order, 0);
+	}
+
+	sgt->nents = dma_map_sg(adapter->pdev_dev, sgl, sgt->orig_nents,
+				DMA_BIDIRECTIONAL);
+	if (!sgt->nents) {
+		dev_err(adapter->pdev_dev,
+			"Not enough memory for HMA DMA mapping");
+		ret = -ENOMEM;
+		goto free_hma;
+	}
+	adapter->hma.flags |= HMA_DMA_MAPPED_FLAG;
+
+	adapter->hma.phy_addr = kcalloc(sgt->nents, sizeof(dma_addr_t),
+					GFP_KERNEL);
+	if (unlikely(!adapter->hma.phy_addr))
+		goto free_hma;
+
+	for_each_sg(sgl, iter, sgt->nents, i) {
+		newpage = sg_page(iter);
+		adapter->hma.phy_addr[i] = sg_dma_address(iter);
+	}
+
+	ncmds = DIV_ROUND_UP(sgt->nents, HMA_MAX_ADDR_IN_CMD);
+	/* Pass on the addresses to firmware */
+	for (i = 0, k = 0; i < ncmds; i++, k += HMA_MAX_ADDR_IN_CMD) {
+		struct fw_hma_cmd hma_cmd;
+		u8 naddr = HMA_MAX_ADDR_IN_CMD;
+		u8 soc = 0, eoc = 0;
+		u8 hma_mode = 1; /* Presently we support only Page table mode */
+
+		soc = (i == 0) ? 1 : 0;
+		eoc = (i == ncmds - 1) ? 1 : 0;
+
+		/* For last cmd, set naddr corresponding to remaining
+		 * addresses
+		 */
+		if (i == ncmds - 1) {
+			naddr = sgt->nents % HMA_MAX_ADDR_IN_CMD;
+			naddr = naddr ? naddr : HMA_MAX_ADDR_IN_CMD;
+		}
+		memset(&hma_cmd, 0, sizeof(hma_cmd));
+		hma_cmd.op_pkd = htonl(FW_CMD_OP_V(FW_HMA_CMD) |
+				       FW_CMD_REQUEST_F | FW_CMD_WRITE_F);
+		hma_cmd.retval_len16 = htonl(FW_LEN16(hma_cmd));
+
+		hma_cmd.mode_to_pcie_params =
+			htonl(FW_HMA_CMD_MODE_V(hma_mode) |
+			      FW_HMA_CMD_SOC_V(soc) | FW_HMA_CMD_EOC_V(eoc));
+
+		/* HMA cmd size specified in MB's */
+		hma_cmd.naddr_size =
+			htonl(FW_HMA_CMD_SIZE_V(hma_size) |
+			      FW_HMA_CMD_NADDR_V(naddr));
+
+		/* Total Page size specified in units of 4K */
+		hma_cmd.addr_size_pkd =
+			htonl(FW_HMA_CMD_ADDR_SIZE_V
+				((page_size << page_order) >> 12));
+
+		/* Fill the 5 addresses */
+		for (j = 0; j < naddr; j++) {
+			hma_cmd.phy_address[j] =
+				cpu_to_be64(adapter->hma.phy_addr[j + k]);
+		}
+		ret = t4_wr_mbox(adapter, adapter->mbox, &hma_cmd,
+				 sizeof(hma_cmd), &hma_cmd);
+		if (ret) {
+			dev_err(adapter->pdev_dev,
+				"HMA FW command failed with err %d\n", ret);
+			goto free_hma;
+		}
+	}
+
+	if (!ret)
+		dev_info(adapter->pdev_dev,
+			 "Reserved %uMB host memory for HMA\n", hma_size);
+	return ret;
+
+free_hma:
+	adap_free_hma_mem(adapter);
+	return ret;
+}
+
 static int adap_init1(struct adapter *adap, struct fw_caps_config_cmd *c)
 {
 	u32 v;
 	int ret;
 
+	/* Now that we've successfully configured and initialized the adapter
+	 * can ask the Firmware what resources it has provisioned for us.
+	 */
+	ret = t4_get_pfres(adap);
+	if (ret) {
+		dev_err(adap->pdev_dev,
+			"Unable to retrieve resource provisioning information\n");
+		return ret;
+	}
+
 	/* get device capabilities */
 	memset(c, 0, sizeof(*c));
 	c->op_to_write = htonl(FW_CMD_OP_V(FW_CAPS_CONFIG_CMD) |
@@ -3077,8 +4417,7 @@ static int adap_init0_phy(struct adapter *adap)
 
 	/* Load PHY Firmware onto adapter.
 	 */
-	ret = t4_load_phy_fw(adap, MEMWIN_NIC, &adap->win0_lock,
-			     phy_info->phy_fw_version,
+	ret = t4_load_phy_fw(adap, MEMWIN_NIC, phy_info->phy_fw_version,
 			     (u8 *)phyf->data, phyf->size);
 	if (ret < 0)
 		dev_err(adap->pdev_dev, "PHY Firmware transfer error %d\n",
@@ -3104,14 +4443,14 @@ static int adap_init0_phy(struct adapter *adap)
  */
 static int adap_init0_config(struct adapter *adapter, int reset)
 {
+	char *fw_config_file, fw_config_file_path[256];
+	u32 finiver, finicsum, cfcsum, param, val;
 	struct fw_caps_config_cmd caps_cmd;
-	const struct firmware *cf;
 	unsigned long mtype = 0, maddr = 0;
-	u32 finiver, finicsum, cfcsum;
-	int ret;
-	int config_issued = 0;
-	char *fw_config_file, fw_config_file_path[256];
+	const struct firmware *cf;
 	char *config_name = NULL;
+	int config_issued = 0;
+	int ret;
 
 	/*
 	 * Reset device if necessary.
@@ -3219,6 +4558,24 @@ static int adap_init0_config(struct adapter *adapter, int reset)
 			goto bye;
 	}
 
+	val = 0;
+
+	/* Ofld + Hash filter is supported. Older fw will fail this request and
+	 * it is fine.
+	 */
+	param = (FW_PARAMS_MNEM_V(FW_PARAMS_MNEM_DEV) |
+		 FW_PARAMS_PARAM_X_V(FW_PARAMS_PARAM_DEV_HASHFILTER_WITH_OFLD));
+	ret = t4_set_params(adapter, adapter->mbox, adapter->pf, 0,
+			    1, &param, &val);
+
+	/* FW doesn't know about Hash filter + ofld support,
+	 * it's not a problem, don't return an error.
+	 */
+	if (ret < 0) {
+		dev_warn(adapter->pdev_dev,
+			 "Hash filter with ofld is not supported by FW\n");
+	}
+
 	/*
 	 * Issue a Capability Configuration command to the firmware to get it
 	 * to parse the Configuration File.  We don't use t4_fw_config_file()
@@ -3289,6 +4646,20 @@ static int adap_init0_config(struct adapter *adapter, int reset)
 	if (ret < 0)
 		goto bye;
 
+	/* We will proceed even if HMA init fails. */
+	ret = adap_config_hma(adapter);
+	if (ret)
+		dev_err(adapter->pdev_dev,
+			"HMA configuration failed with error %d\n", ret);
+
+	if (is_t6(adapter->params.chip)) {
+		adap_config_hpfilter(adapter);
+		ret = setup_ppod_edram(adapter);
+		if (!ret)
+			dev_info(adapter->pdev_dev, "Successfully enabled "
+				 "ppod edram feature\n");
+	}
+
 	/*
 	 * And finally tell the firmware to initialize itself using the
 	 * parameters from the Configuration File.
@@ -3378,14 +4749,14 @@ static struct fw_info *find_fw_info(int chip)
 /*
  * Phase 0 of initialization: contact FW, obtain config, perform basic init.
  */
-static int adap_init0(struct adapter *adap)
+static int adap_init0(struct adapter *adap, int vpd_skip)
 {
-	int ret;
-	u32 v, port_vec;
-	enum dev_state state;
-	u32 params[7], val[7];
 	struct fw_caps_config_cmd caps_cmd;
+	u32 params[7], val[7];
+	enum dev_state state;
+	u32 v, port_vec;
 	int reset = 1;
+	int ret;
 
 	/* Grab Firmware Device Log parameters as early as possible so we have
 	 * access to it for debugging, etc.
@@ -3403,7 +4774,7 @@ static int adap_init0(struct adapter *adap)
 		return ret;
 	}
 	if (ret == adap->mbox)
-		adap->flags |= MASTER_PF;
+		adap->flags |= CXGB4_MASTER_PF;
 
 	/*
 	 * If we're the Master PF Driver and the device is uninitialized,
@@ -3412,16 +4783,13 @@ static int adap_init0(struct adapter *adap)
 	 * later reporting and B. to warn if the currently loaded firmware
 	 * is excessively mismatched relative to the driver.)
 	 */
-	t4_get_fw_version(adap, &adap->params.fw_vers);
-	t4_get_bs_version(adap, &adap->params.bs_vers);
-	t4_get_tp_version(adap, &adap->params.tp_vers);
-	t4_get_exprom_version(adap, &adap->params.er_vers);
 
+	t4_get_version_info(adap);
 	ret = t4_check_fw_version(adap);
 	/* If firmware is too old (not supported by driver) force an update. */
 	if (ret)
 		state = DEV_STATE_UNINIT;
-	if ((adap->flags & MASTER_PF) && state != DEV_STATE_INIT) {
+	if ((adap->flags & CXGB4_MASTER_PF) && state != DEV_STATE_INIT) {
 		struct fw_info *fw_info;
 		struct fw_hdr *card_fw;
 		const struct firmware *fw;
@@ -3443,8 +4811,12 @@ static int adap_init0(struct adapter *adap)
 		 * card
 		 */
 		card_fw = kvzalloc(sizeof(*card_fw), GFP_KERNEL);
+		if (!card_fw) {
+			ret = -ENOMEM;
+			goto bye;
+		}
 
-		/* Get FW from from /lib/firmware/ */
+		/* Get FW from /lib/firmware/ */
 		ret = request_firmware(&fw, fw_info->fw_mod_name,
 				       adap->pdev_dev);
 		if (ret < 0) {
@@ -3468,39 +4840,18 @@ static int adap_init0(struct adapter *adap)
 			goto bye;
 	}
 
-	/*
-	 * Grab VPD parameters.  This should be done after we establish a
-	 * connection to the firmware since some of the VPD parameters
-	 * (notably the Core Clock frequency) are retrieved via requests to
-	 * the firmware.  On the other hand, we need these fairly early on
-	 * so we do this right after getting ahold of the firmware.
-	 */
-	ret = t4_get_vpd_params(adap, &adap->params.vpd);
-	if (ret < 0)
-		goto bye;
-
-	/*
-	 * Find out what ports are available to us.  Note that we need to do
-	 * this before calling adap_init0_no_config() since it needs nports
-	 * and portvec ...
-	 */
-	v =
-	    FW_PARAMS_MNEM_V(FW_PARAMS_MNEM_DEV) |
-	    FW_PARAMS_PARAM_X_V(FW_PARAMS_PARAM_DEV_PORTVEC);
-	ret = t4_query_params(adap, adap->mbox, adap->pf, 0, 1, &v, &port_vec);
-	if (ret < 0)
-		goto bye;
-
-	adap->params.nports = hweight32(port_vec);
-	adap->params.portvec = port_vec;
-
 	/* If the firmware is initialized already, emit a simply note to that
 	 * effect. Otherwise, it's time to try initializing the adapter.
 	 */
 	if (state == DEV_STATE_INIT) {
+		ret = adap_config_hma(adap);
+		if (ret)
+			dev_err(adap->pdev_dev,
+				"HMA configuration failed with error %d\n",
+				ret);
 		dev_info(adap->pdev_dev, "Coming up as %s: "\
 			 "Adapter already initialized\n",
-			 adap->flags & MASTER_PF ? "MASTER" : "SLAVE");
+			 adap->flags & CXGB4_MASTER_PF ? "MASTER" : "SLAVE");
 	} else {
 		dev_info(adap->pdev_dev, "Coming up as MASTER: "\
 			 "Initializing adapter\n");
@@ -3539,6 +4890,47 @@ static int adap_init0(struct adapter *adap)
 		}
 	}
 
+	/* Now that we've successfully configured and initialized the adapter
+	 * (or found it already initialized), we can ask the Firmware what
+	 * resources it has provisioned for us.
+	 */
+	ret = t4_get_pfres(adap);
+	if (ret) {
+		dev_err(adap->pdev_dev,
+			"Unable to retrieve resource provisioning information\n");
+		goto bye;
+	}
+
+	/* Grab VPD parameters.  This should be done after we establish a
+	 * connection to the firmware since some of the VPD parameters
+	 * (notably the Core Clock frequency) are retrieved via requests to
+	 * the firmware.  On the other hand, we need these fairly early on
+	 * so we do this right after getting ahold of the firmware.
+	 *
+	 * We need to do this after initializing the adapter because someone
+	 * could have FLASHed a new VPD which won't be read by the firmware
+	 * until we do the RESET ...
+	 */
+	if (!vpd_skip) {
+		ret = t4_get_vpd_params(adap, &adap->params.vpd);
+		if (ret < 0)
+			goto bye;
+	}
+
+	/* Find out what ports are available to us.  Note that we need to do
+	 * this before calling adap_init0_no_config() since it needs nports
+	 * and portvec ...
+	 */
+	v =
+	    FW_PARAMS_MNEM_V(FW_PARAMS_MNEM_DEV) |
+	    FW_PARAMS_PARAM_X_V(FW_PARAMS_PARAM_DEV_PORTVEC);
+	ret = t4_query_params(adap, adap->mbox, adap->pf, 0, 1, &v, &port_vec);
+	if (ret < 0)
+		goto bye;
+
+	adap->params.nports = hweight32(port_vec);
+	adap->params.portvec = port_vec;
+
 	/* Give the SGE code a chance to pull in anything that it needs ...
 	 * Note that this must be called after we retrieve our VPD parameters
 	 * in order to know how to convert core ticks to seconds, etc.
@@ -3547,22 +4939,30 @@ static int adap_init0(struct adapter *adap)
 	if (ret < 0)
 		goto bye;
 
+	/* Grab the SGE Doorbell Queue Timer values.  If successful, that
+	 * indicates that the Firmware and Hardware support this.
+	 */
+	params[0] = (FW_PARAMS_MNEM_V(FW_PARAMS_MNEM_DEV) |
+		    FW_PARAMS_PARAM_X_V(FW_PARAMS_PARAM_DEV_DBQ_TIMERTICK));
+	ret = t4_query_params(adap, adap->mbox, adap->pf, 0,
+			      1, params, val);
+
+	if (!ret) {
+		adap->sge.dbqtimer_tick = val[0];
+		ret = t4_read_sge_dbqtimers(adap,
+					    ARRAY_SIZE(adap->sge.dbqtimer_val),
+					    adap->sge.dbqtimer_val);
+	}
+
+	if (!ret)
+		adap->flags |= CXGB4_SGE_DBQ_TIMER;
+
 	if (is_bypass_device(adap->pdev->device))
 		adap->params.bypass = 1;
 
 	/*
 	 * Grab some of our basic fundamental operating parameters.
 	 */
-#define FW_PARAM_DEV(param) \
-	(FW_PARAMS_MNEM_V(FW_PARAMS_MNEM_DEV) | \
-	FW_PARAMS_PARAM_X_V(FW_PARAMS_PARAM_DEV_##param))
-
-#define FW_PARAM_PFVF(param) \
-	FW_PARAMS_MNEM_V(FW_PARAMS_MNEM_PFVF) | \
-	FW_PARAMS_PARAM_X_V(FW_PARAMS_PARAM_PFVF_##param)|  \
-	FW_PARAMS_PARAM_Y_V(0) | \
-	FW_PARAMS_PARAM_Z_V(0)
-
 	params[0] = FW_PARAM_PFVF(EQ_START);
 	params[1] = FW_PARAM_PFVF(L2T_START);
 	params[2] = FW_PARAM_PFVF(L2T_END);
@@ -3579,6 +4979,33 @@ static int adap_init0(struct adapter *adap)
 	adap->tids.nftids = val[4] - val[3] + 1;
 	adap->sge.ingr_start = val[5];
 
+	if (CHELSIO_CHIP_VERSION(adap->params.chip) > CHELSIO_T5) {
+		params[0] = FW_PARAM_PFVF(HPFILTER_START);
+		params[1] = FW_PARAM_PFVF(HPFILTER_END);
+		ret = t4_query_params(adap, adap->mbox, adap->pf, 0, 2,
+				      params, val);
+		if (ret < 0)
+			goto bye;
+
+		adap->tids.hpftid_base = val[0];
+		adap->tids.nhpftids = val[1] - val[0] + 1;
+
+		/* Read the raw mps entries. In T6, the last 2 tcam entries
+		 * are reserved for raw mac addresses (rawf = 2, one per port).
+		 */
+		params[0] = FW_PARAM_PFVF(RAWF_START);
+		params[1] = FW_PARAM_PFVF(RAWF_END);
+		ret = t4_query_params(adap, adap->mbox, adap->pf, 0, 2,
+				      params, val);
+		if (ret == 0) {
+			adap->rawf_start = val[0];
+			adap->rawf_cnt = val[1] - val[0] + 1;
+		}
+
+		adap->tids.tid_base =
+			t4_read_reg(adap, LE_DB_ACTIVE_TABLE_START_INDEX_A);
+	}
+
 	/* qids (ingress/egress) returned from firmware can be anywhere
 	 * in the range from EQ(IQFLINT)_START to EQ(IQFLINT)_END.
 	 * Hence driver needs to allocate memory for this range to
@@ -3610,23 +5037,20 @@ static int adap_init0(struct adapter *adap)
 	/* Allocate the memory for the vaious egress queue bitmaps
 	 * ie starving_fl, txq_maperr and blocked_fl.
 	 */
-	adap->sge.starving_fl =	kcalloc(BITS_TO_LONGS(adap->sge.egr_sz),
-					sizeof(long), GFP_KERNEL);
+	adap->sge.starving_fl = bitmap_zalloc(adap->sge.egr_sz, GFP_KERNEL);
 	if (!adap->sge.starving_fl) {
 		ret = -ENOMEM;
 		goto bye;
 	}
 
-	adap->sge.txq_maperr = kcalloc(BITS_TO_LONGS(adap->sge.egr_sz),
-				       sizeof(long), GFP_KERNEL);
+	adap->sge.txq_maperr = bitmap_zalloc(adap->sge.egr_sz, GFP_KERNEL);
 	if (!adap->sge.txq_maperr) {
 		ret = -ENOMEM;
 		goto bye;
 	}
 
 #ifdef CONFIG_DEBUG_FS
-	adap->sge.blocked_fl = kcalloc(BITS_TO_LONGS(adap->sge.egr_sz),
-				       sizeof(long), GFP_KERNEL);
+	adap->sge.blocked_fl = bitmap_zalloc(adap->sge.egr_sz, GFP_KERNEL);
 	if (!adap->sge.blocked_fl) {
 		ret = -ENOMEM;
 		goto bye;
@@ -3641,11 +5065,18 @@ static int adap_init0(struct adapter *adap)
 	adap->clipt_start = val[0];
 	adap->clipt_end = val[1];
 
-	/* We don't yet have a PARAMs calls to retrieve the number of Traffic
-	 * Classes supported by the hardware/firmware so we hard code it here
-	 * for now.
-	 */
-	adap->params.nsched_cls = is_t4(adap->params.chip) ? 15 : 16;
+	/* Get the supported number of traffic classes */
+	params[0] = FW_PARAM_DEV(NUM_TM_CLASS);
+	ret = t4_query_params(adap, adap->mbox, adap->pf, 0, 1, params, val);
+	if (ret < 0) {
+		/* We couldn't retrieve the number of Traffic Classes
+		 * supported by the hardware/firmware. So we hard
+		 * code it here.
+		 */
+		adap->params.nsched_cls = is_t4(adap->params.chip) ? 15 : 16;
+	} else {
+		adap->params.nsched_cls = val[0];
+	}
 
 	/* query params related to active filter region */
 	params[0] = FW_PARAM_PFVF(ACTIVE_FILTER_START);
@@ -3655,7 +5086,7 @@ static int adap_init0(struct adapter *adap)
 	 * offload connection through firmware work request
 	 */
 	if ((val[0] != val[1]) && (ret >= 0)) {
-		adap->flags |= FW_OFLD_CONN;
+		adap->flags |= CXGB4_FW_OFLD_CONN;
 		adap->tids.aftid_base = val[0];
 		adap->tids.aftid_end = val[1];
 	}
@@ -3690,6 +5121,25 @@ static int adap_init0(struct adapter *adap)
 			      1, params, val);
 	adap->params.fr_nsmr_tpte_wr_support = (ret == 0 && val[0] != 0);
 
+	/* See if FW supports FW_FILTER2 work request */
+	if (is_t4(adap->params.chip)) {
+		adap->params.filter2_wr_support = false;
+	} else {
+		params[0] = FW_PARAM_DEV(FILTER2_WR);
+		ret = t4_query_params(adap, adap->mbox, adap->pf, 0,
+				      1, params, val);
+		adap->params.filter2_wr_support = (ret == 0 && val[0] != 0);
+	}
+
+	/* Check if FW supports returning vin and smt index.
+	 * If this is not supported, driver will interpret
+	 * these values from viid.
+	 */
+	params[0] = FW_PARAM_DEV(OPAQUE_VIID_SMT_EXTN);
+	ret = t4_query_params(adap, adap->mbox, adap->pf, 0,
+			      1, params, val);
+	adap->params.viid_smt_extn_support = (ret == 0 && val[0] != 0);
+
 	/*
 	 * Get device capabilities so we can determine what resources we need
 	 * to manage.
@@ -3703,7 +5153,16 @@ static int adap_init0(struct adapter *adap)
 	if (ret < 0)
 		goto bye;
 
-	if (caps_cmd.ofldcaps) {
+	/* hash filter has some mandatory register settings to be tested and for
+	 * that it needs to test whether offload is enabled or not, hence
+	 * checking and setting it here.
+	 */
+	if (caps_cmd.ofldcaps)
+		adap->params.offload = 1;
+
+	if (caps_cmd.ofldcaps ||
+	    (caps_cmd.niccaps & htons(FW_CAPS_CONFIG_NIC_HASHFILTER)) ||
+	    (caps_cmd.niccaps & htons(FW_CAPS_CONFIG_NIC_ETHOFLD))) {
 		/* query offload-related parameters */
 		params[0] = FW_PARAM_DEV(NTID);
 		params[1] = FW_PARAM_PFVF(SERVER_START);
@@ -3728,7 +5187,7 @@ static int adap_init0(struct adapter *adap)
 		 * 2. Server filter: This are special filters which are used
 		 * to redirect SYN packets to offload queue.
 		 */
-		if (adap->flags & FW_OFLD_CONN && !is_bypass(adap)) {
+		if (adap->flags & CXGB4_FW_OFLD_CONN && !is_bypass(adap)) {
 			adap->tids.sftid_base = adap->tids.ftid_base +
 					DIV_ROUND_UP(adap->tids.nftids, 3);
 			adap->tids.nsftids = adap->tids.nftids -
@@ -3740,8 +5199,24 @@ static int adap_init0(struct adapter *adap)
 		adap->vres.ddp.size = val[4] - val[3] + 1;
 		adap->params.ofldq_wr_cred = val[5];
 
-		adap->params.offload = 1;
-		adap->num_ofld_uld += 1;
+		if (caps_cmd.niccaps & htons(FW_CAPS_CONFIG_NIC_HASHFILTER)) {
+			init_hash_filter(adap);
+		} else {
+			adap->num_ofld_uld += 1;
+		}
+
+		if (caps_cmd.niccaps & htons(FW_CAPS_CONFIG_NIC_ETHOFLD)) {
+			params[0] = FW_PARAM_PFVF(ETHOFLD_START);
+			params[1] = FW_PARAM_PFVF(ETHOFLD_END);
+			ret = t4_query_params(adap, adap->mbox, adap->pf, 0, 2,
+					      params, val);
+			if (!ret) {
+				adap->tids.eotid_base = val[0];
+				adap->tids.neotids = min_t(u32, MAX_ATIDS,
+							   val[1] - val[0] + 1);
+				adap->params.ethofld = 1;
+			}
+		}
 	}
 	if (caps_cmd.rdmacaps) {
 		params[0] = FW_PARAM_PFVF(STAG_START);
@@ -3761,6 +5236,20 @@ static int adap_init0(struct adapter *adap)
 		adap->vres.pbl.start = val[4];
 		adap->vres.pbl.size = val[5] - val[4] + 1;
 
+		params[0] = FW_PARAM_PFVF(SRQ_START);
+		params[1] = FW_PARAM_PFVF(SRQ_END);
+		ret = t4_query_params(adap, adap->mbox, adap->pf, 0, 2,
+				      params, val);
+		if (!ret) {
+			adap->vres.srq.start = val[0];
+			adap->vres.srq.size = val[1] - val[0] + 1;
+		}
+		if (adap->vres.srq.size) {
+			adap->srq = t4_init_srq(adap->vres.srq.size);
+			if (!adap->srq)
+				dev_warn(&adap->pdev->dev, "could not allocate SRQ, continuing\n");
+		}
+
 		params[0] = FW_PARAM_PFVF(SQRQ_START);
 		params[1] = FW_PARAM_PFVF(SQRQ_END);
 		params[2] = FW_PARAM_PFVF(CQ_START);
@@ -3794,6 +5283,18 @@ static int adap_init0(struct adapter *adap)
 			 "max_ordird_qp %d max_ird_adapter %d\n",
 			 adap->params.max_ordird_qp,
 			 adap->params.max_ird_adapter);
+
+		/* Enable write_with_immediate if FW supports it */
+		params[0] = FW_PARAM_DEV(RDMA_WRITE_WITH_IMM);
+		ret = t4_query_params(adap, adap->mbox, adap->pf, 0, 1, params,
+				      val);
+		adap->params.write_w_imm_support = (ret == 0 && val[0] != 0);
+
+		/* Enable write_cmpl if FW supports it */
+		params[0] = FW_PARAM_DEV(RI_WRITE_CMPL_WR);
+		ret = t4_query_params(adap, adap->mbox, adap->pf, 0, 1, params,
+				      val);
+		adap->params.write_cmpl_support = (ret == 0 && val[0] != 0);
 		adap->num_ofld_uld += 2;
 	}
 	if (caps_cmd.iscsicaps) {
@@ -3805,25 +5306,53 @@ static int adap_init0(struct adapter *adap)
 			goto bye;
 		adap->vres.iscsi.start = val[0];
 		adap->vres.iscsi.size = val[1] - val[0] + 1;
+		if (is_t6(adap->params.chip)) {
+			params[0] = FW_PARAM_PFVF(PPOD_EDRAM_START);
+			params[1] = FW_PARAM_PFVF(PPOD_EDRAM_END);
+			ret = t4_query_params(adap, adap->mbox, adap->pf, 0, 2,
+					      params, val);
+			if (!ret) {
+				adap->vres.ppod_edram.start = val[0];
+				adap->vres.ppod_edram.size =
+					val[1] - val[0] + 1;
+
+				dev_info(adap->pdev_dev,
+					 "ppod edram start 0x%x end 0x%x size 0x%x\n",
+					 val[0], val[1],
+					 adap->vres.ppod_edram.size);
+			}
+		}
 		/* LIO target and cxgb4i initiaitor */
 		adap->num_ofld_uld += 2;
 	}
 	if (caps_cmd.cryptocaps) {
-		/* Should query params here...TODO */
-		params[0] = FW_PARAM_PFVF(NCRYPTO_LOOKASIDE);
-		ret = t4_query_params(adap, adap->mbox, adap->pf, 0, 2,
-				      params, val);
-		if (ret < 0) {
-			if (ret != -EINVAL)
+		if (ntohs(caps_cmd.cryptocaps) &
+		    FW_CAPS_CONFIG_CRYPTO_LOOKASIDE) {
+			params[0] = FW_PARAM_PFVF(NCRYPTO_LOOKASIDE);
+			ret = t4_query_params(adap, adap->mbox, adap->pf, 0,
+					      2, params, val);
+			if (ret < 0) {
+				if (ret != -EINVAL)
+					goto bye;
+			} else {
+				adap->vres.ncrypto_fc = val[0];
+			}
+			adap->num_ofld_uld += 1;
+		}
+		if (ntohs(caps_cmd.cryptocaps) &
+		    FW_CAPS_CONFIG_TLS_INLINE) {
+			params[0] = FW_PARAM_PFVF(TLS_START);
+			params[1] = FW_PARAM_PFVF(TLS_END);
+			ret = t4_query_params(adap, adap->mbox, adap->pf, 0,
+					      2, params, val);
+			if (ret < 0)
 				goto bye;
-		} else {
-			adap->vres.ncrypto_fc = val[0];
+			adap->vres.key.start = val[0];
+			adap->vres.key.size = val[1] - val[0] + 1;
+			adap->num_uld += 1;
 		}
-		adap->params.crypto |= ULP_CRYPTO_LOOKASIDE;
-		adap->num_uld += 1;
+		adap->params.crypto = ntohs(caps_cmd.cryptocaps);
 	}
-#undef FW_PARAM_PFVF
-#undef FW_PARAM_DEV
 
 	/* The MTU/MSS Table is initialized by now, so load their values.  If
 	 * we're initializing the adapter, then we'll make any modifications
@@ -3861,8 +5390,8 @@ static int adap_init0(struct adapter *adap)
 			     adap->params.b_wnd);
 	}
 	t4_init_sge_params(adap);
-	adap->flags |= FW_OK;
-	t4_init_tp_params(adap);
+	adap->flags |= CXGB4_FW_OK;
+	t4_init_tp_params(adap, true);
 	return 0;
 
 	/*
@@ -3871,12 +5400,13 @@ static int adap_init0(struct adapter *adap)
 	 * happened to HW/FW, stop issuing commands.
 	 */
 bye:
+	adap_free_hma_mem(adap);
 	kfree(adap->sge.egr_map);
 	kfree(adap->sge.ingr_map);
-	kfree(adap->sge.starving_fl);
-	kfree(adap->sge.txq_maperr);
+	bitmap_free(adap->sge.starving_fl);
+	bitmap_free(adap->sge.txq_maperr);
 #ifdef CONFIG_DEBUG_FS
-	kfree(adap->sge.blocked_fl);
+	bitmap_free(adap->sge.blocked_fl);
 #endif
 	if (ret != -ETIMEDOUT && ret != -EIO)
 		t4_fw_bye(adap, adap->mbox);
@@ -3895,23 +5425,24 @@ static pci_ers_result_t eeh_err_detected(struct pci_dev *pdev,
 		goto out;
 
 	rtnl_lock();
-	adap->flags &= ~FW_OK;
+	adap->flags &= ~CXGB4_FW_OK;
 	notify_ulds(adap, CXGB4_STATE_START_RECOVERY);
 	spin_lock(&adap->stats_lock);
 	for_each_port(adap, i) {
 		struct net_device *dev = adap->port[i];
-
-		netif_device_detach(dev);
-		netif_carrier_off(dev);
+		if (dev) {
+			netif_device_detach(dev);
+			netif_carrier_off(dev);
+		}
 	}
 	spin_unlock(&adap->stats_lock);
 	disable_interrupts(adap);
-	if (adap->flags & FULL_INIT_DONE)
+	if (adap->flags & CXGB4_FULL_INIT_DONE)
 		cxgb_down(adap);
 	rtnl_unlock();
-	if ((adap->flags & DEV_ENABLED)) {
+	if ((adap->flags & CXGB4_DEV_ENABLED)) {
 		pci_disable_device(pdev);
-		adap->flags &= ~DEV_ENABLED;
+		adap->flags &= ~CXGB4_DEV_ENABLED;
 	}
 out:	return state == pci_channel_io_perm_failure ?
 		PCI_ERS_RESULT_DISCONNECT : PCI_ERS_RESULT_NEED_RESET;
@@ -3929,37 +5460,48 @@ static pci_ers_result_t eeh_slot_reset(struct pci_dev *pdev)
 		return PCI_ERS_RESULT_RECOVERED;
 	}
 
-	if (!(adap->flags & DEV_ENABLED)) {
+	if (!(adap->flags & CXGB4_DEV_ENABLED)) {
 		if (pci_enable_device(pdev)) {
 			dev_err(&pdev->dev, "Cannot reenable PCI "
 					    "device after reset\n");
 			return PCI_ERS_RESULT_DISCONNECT;
 		}
-		adap->flags |= DEV_ENABLED;
+		adap->flags |= CXGB4_DEV_ENABLED;
 	}
 
 	pci_set_master(pdev);
 	pci_restore_state(pdev);
 	pci_save_state(pdev);
-	pci_cleanup_aer_uncorrect_error_status(pdev);
 
 	if (t4_wait_dev_ready(adap->regs) < 0)
 		return PCI_ERS_RESULT_DISCONNECT;
 	if (t4_fw_hello(adap, adap->mbox, adap->pf, MASTER_MUST, NULL) < 0)
 		return PCI_ERS_RESULT_DISCONNECT;
-	adap->flags |= FW_OK;
+	adap->flags |= CXGB4_FW_OK;
 	if (adap_init1(adap, &c))
 		return PCI_ERS_RESULT_DISCONNECT;
 
 	for_each_port(adap, i) {
-		struct port_info *p = adap2pinfo(adap, i);
+		struct port_info *pi = adap2pinfo(adap, i);
+		u8 vivld = 0, vin = 0;
 
-		ret = t4_alloc_vi(adap, adap->mbox, p->tx_chan, adap->pf, 0, 1,
-				  NULL, NULL);
+		ret = t4_alloc_vi(adap, adap->mbox, pi->tx_chan, adap->pf, 0, 1,
+				  NULL, NULL, &vivld, &vin);
 		if (ret < 0)
 			return PCI_ERS_RESULT_DISCONNECT;
-		p->viid = ret;
-		p->xact_addr_filt = -1;
+		pi->viid = ret;
+		pi->xact_addr_filt = -1;
+		/* If fw supports returning the VIN as part of FW_VI_CMD,
+		 * save the returned values.
+		 */
+		if (adap->params.viid_smt_extn_support) {
+			pi->vivld = vivld;
+			pi->vin = vin;
+		} else {
+			/* Retrieve the values from VIID */
+			pi->vivld = FW_VIID_VIVLD_G(pi->viid);
+			pi->vin = FW_VIID_VIN_G(pi->viid);
+		}
 	}
 
 	t4_load_mtus(adap, adap->params.mtus, adap->params.a_wnd,
@@ -3981,20 +5523,104 @@ static void eeh_resume(struct pci_dev *pdev)
 	rtnl_lock();
 	for_each_port(adap, i) {
 		struct net_device *dev = adap->port[i];
-
-		if (netif_running(dev)) {
-			link_start(dev);
-			cxgb_set_rxmode(dev);
+		if (dev) {
+			if (netif_running(dev)) {
+				link_start(dev);
+				cxgb_set_rxmode(dev);
+			}
+			netif_device_attach(dev);
 		}
-		netif_device_attach(dev);
 	}
 	rtnl_unlock();
 }
 
+static void eeh_reset_prepare(struct pci_dev *pdev)
+{
+	struct adapter *adapter = pci_get_drvdata(pdev);
+	int i;
+
+	if (adapter->pf != 4)
+		return;
+
+	adapter->flags &= ~CXGB4_FW_OK;
+
+	notify_ulds(adapter, CXGB4_STATE_DOWN);
+
+	for_each_port(adapter, i)
+		if (adapter->port[i]->reg_state == NETREG_REGISTERED)
+			cxgb_close(adapter->port[i]);
+
+	disable_interrupts(adapter);
+	cxgb4_free_mps_ref_entries(adapter);
+
+	adap_free_hma_mem(adapter);
+
+	if (adapter->flags & CXGB4_FULL_INIT_DONE)
+		cxgb_down(adapter);
+}
+
+static void eeh_reset_done(struct pci_dev *pdev)
+{
+	struct adapter *adapter = pci_get_drvdata(pdev);
+	int err, i;
+
+	if (adapter->pf != 4)
+		return;
+
+	err = t4_wait_dev_ready(adapter->regs);
+	if (err < 0) {
+		dev_err(adapter->pdev_dev,
+			"Device not ready, err %d", err);
+		return;
+	}
+
+	setup_memwin(adapter);
+
+	err = adap_init0(adapter, 1);
+	if (err) {
+		dev_err(adapter->pdev_dev,
+			"Adapter init failed, err %d", err);
+		return;
+	}
+
+	setup_memwin_rdma(adapter);
+
+	if (adapter->flags & CXGB4_FW_OK) {
+		err = t4_port_init(adapter, adapter->pf, adapter->pf, 0);
+		if (err) {
+			dev_err(adapter->pdev_dev,
+				"Port init failed, err %d", err);
+			return;
+		}
+	}
+
+	err = cfg_queues(adapter);
+	if (err) {
+		dev_err(adapter->pdev_dev,
+			"Config queues failed, err %d", err);
+		return;
+	}
+
+	cxgb4_init_mps_ref_entries(adapter);
+
+	err = setup_fw_sge_queues(adapter);
+	if (err) {
+		dev_err(adapter->pdev_dev,
+			"FW sge queue allocation failed, err %d", err);
+		return;
+	}
+
+	for_each_port(adapter, i)
+		if (adapter->port[i]->reg_state == NETREG_REGISTERED)
+			cxgb_open(adapter->port[i]);
+}
+
 static const struct pci_error_handlers cxgb4_eeh = {
 	.error_detected = eeh_err_detected,
 	.slot_reset     = eeh_slot_reset,
 	.resume         = eeh_resume,
+	.reset_prepare  = eeh_reset_prepare,
+	.reset_done     = eeh_reset_done,
 };
 
 /* Return true if the Link Configuration supports "High Speeds" (those greater
@@ -4004,90 +5630,158 @@ static inline bool is_x_10g_port(const struct link_config *lc)
 {
 	unsigned int speeds, high_speeds;
 
-	speeds = FW_PORT_CAP_SPEED_V(FW_PORT_CAP_SPEED_G(lc->supported));
-	high_speeds = speeds & ~(FW_PORT_CAP_SPEED_100M | FW_PORT_CAP_SPEED_1G);
+	speeds = FW_PORT_CAP32_SPEED_V(FW_PORT_CAP32_SPEED_G(lc->pcaps));
+	high_speeds = speeds &
+			~(FW_PORT_CAP32_SPEED_100M | FW_PORT_CAP32_SPEED_1G);
 
 	return high_speeds != 0;
 }
 
-/*
- * Perform default configuration of DMA queues depending on the number and type
+/* Perform default configuration of DMA queues depending on the number and type
  * of ports we found and the number of available CPUs.  Most settings can be
  * modified by the admin prior to actual use.
  */
-static void cfg_queues(struct adapter *adap)
+static int cfg_queues(struct adapter *adap)
 {
+	u32 avail_qsets, avail_eth_qsets, avail_uld_qsets;
+	u32 ncpus = num_online_cpus();
+	u32 niqflint, neq, num_ulds;
 	struct sge *s = &adap->sge;
-	int i = 0, n10g = 0, qidx = 0;
-#ifndef CONFIG_CHELSIO_T4_DCB
-	int q10g = 0;
-#endif
+	u32 i, n10g = 0, qidx = 0;
+	u32 q10g = 0, q1g;
 
-	/* Reduce memory usage in kdump environment, disable all offload.
-	 */
+	/* Reduce memory usage in kdump environment, disable all offload. */
 	if (is_kdump_kernel() || (is_uld(adap) && t4_uld_mem_alloc(adap))) {
 		adap->params.offload = 0;
 		adap->params.crypto = 0;
+		adap->params.ethofld = 0;
 	}
 
-	n10g += is_x_10g_port(&adap2pinfo(adap, i)->link_cfg);
+	/* Calculate the number of Ethernet Queue Sets available based on
+	 * resources provisioned for us.  We always have an Asynchronous
+	 * Firmware Event Ingress Queue.  If we're operating in MSI or Legacy
+	 * IRQ Pin Interrupt mode, then we'll also have a Forwarded Interrupt
+	 * Ingress Queue.  Meanwhile, we need two Egress Queues for each
+	 * Queue Set: one for the Free List and one for the Ethernet TX Queue.
+	 *
+	 * Note that we should also take into account all of the various
+	 * Offload Queues.  But, in any situation where we're operating in
+	 * a Resource Constrained Provisioning environment, doing any Offload
+	 * at all is problematic ...
+	 */
+	niqflint = adap->params.pfres.niqflint - 1;
+	if (!(adap->flags & CXGB4_USING_MSIX))
+		niqflint--;
+	neq = adap->params.pfres.neq / 2;
+	avail_qsets = min(niqflint, neq);
+
+	if (avail_qsets < adap->params.nports) {
+		dev_err(adap->pdev_dev, "avail_eth_qsets=%d < nports=%d\n",
+			avail_qsets, adap->params.nports);
+		return -ENOMEM;
+	}
+
+	/* Count the number of 10Gb/s or better ports */
+	for_each_port(adap, i)
+		n10g += is_x_10g_port(&adap2pinfo(adap, i)->link_cfg);
+
+	avail_eth_qsets = min_t(u32, avail_qsets, MAX_ETH_QSETS);
+
+	/* We default to 1 queue per non-10G port and up to # of cores queues
+	 * per 10G port.
+	 */
+	if (n10g)
+		q10g = (avail_eth_qsets - (adap->params.nports - n10g)) / n10g;
+
 #ifdef CONFIG_CHELSIO_T4_DCB
 	/* For Data Center Bridging support we need to be able to support up
 	 * to 8 Traffic Priorities; each of which will be assigned to its
 	 * own TX Queue in order to prevent Head-Of-Line Blocking.
 	 */
-	if (adap->params.nports * 8 > MAX_ETH_QSETS) {
-		dev_err(adap->pdev_dev, "MAX_ETH_QSETS=%d < %d!\n",
-			MAX_ETH_QSETS, adap->params.nports * 8);
-		BUG_ON(1);
+	q1g = 8;
+	if (adap->params.nports * 8 > avail_eth_qsets) {
+		dev_err(adap->pdev_dev, "DCB avail_eth_qsets=%d < %d!\n",
+			avail_eth_qsets, adap->params.nports * 8);
+		return -ENOMEM;
 	}
 
-	for_each_port(adap, i) {
-		struct port_info *pi = adap2pinfo(adap, i);
+	if (adap->params.nports * ncpus < avail_eth_qsets)
+		q10g = max(8U, ncpus);
+	else
+		q10g = max(8U, q10g);
 
-		pi->first_qset = qidx;
-		pi->nqsets = is_kdump_kernel() ? 1 : 8;
-		qidx += pi->nqsets;
-	}
-#else /* !CONFIG_CHELSIO_T4_DCB */
-	/*
-	 * We default to 1 queue per non-10G port and up to # of cores queues
-	 * per 10G port.
-	 */
-	if (n10g)
-		q10g = (MAX_ETH_QSETS - (adap->params.nports - n10g)) / n10g;
-	if (q10g > netif_get_num_default_rss_queues())
-		q10g = netif_get_num_default_rss_queues();
+	while ((q10g * n10g) >
+	       (avail_eth_qsets - (adap->params.nports - n10g) * q1g))
+		q10g--;
 
-	if (is_kdump_kernel())
+#else /* !CONFIG_CHELSIO_T4_DCB */
+	q1g = 1;
+	q10g = min(q10g, ncpus);
+#endif /* !CONFIG_CHELSIO_T4_DCB */
+	if (is_kdump_kernel()) {
 		q10g = 1;
+		q1g = 1;
+	}
 
 	for_each_port(adap, i) {
 		struct port_info *pi = adap2pinfo(adap, i);
 
 		pi->first_qset = qidx;
-		pi->nqsets = is_x_10g_port(&pi->link_cfg) ? q10g : 1;
+		pi->nqsets = is_x_10g_port(&pi->link_cfg) ? q10g : q1g;
 		qidx += pi->nqsets;
 	}
-#endif /* !CONFIG_CHELSIO_T4_DCB */
 
 	s->ethqsets = qidx;
 	s->max_ethqsets = qidx;   /* MSI-X may lower it later */
+	avail_qsets -= qidx;
 
 	if (is_uld(adap)) {
-		/*
-		 * For offload we use 1 queue/channel if all ports are up to 1G,
+		/* For offload we use 1 queue/channel if all ports are up to 1G,
 		 * otherwise we divide all available queues amongst the channels
 		 * capped by the number of available cores.
 		 */
-		if (n10g) {
-			i = min_t(int, MAX_OFLD_QSETS, num_online_cpus());
-			s->ofldqsets = roundup(i, adap->params.nports);
-		} else {
+		num_ulds = adap->num_uld + adap->num_ofld_uld;
+		i = min_t(u32, MAX_OFLD_QSETS, ncpus);
+		avail_uld_qsets = roundup(i, adap->params.nports);
+		if (avail_qsets < num_ulds * adap->params.nports) {
+			adap->params.offload = 0;
+			adap->params.crypto = 0;
+			s->ofldqsets = 0;
+		} else if (avail_qsets < num_ulds * avail_uld_qsets || !n10g) {
 			s->ofldqsets = adap->params.nports;
+		} else {
+			s->ofldqsets = avail_uld_qsets;
 		}
+
+		avail_qsets -= num_ulds * s->ofldqsets;
 	}
 
+	/* ETHOFLD Queues used for QoS offload should follow same
+	 * allocation scheme as normal Ethernet Queues.
+	 */
+	if (is_ethofld(adap)) {
+		if (avail_qsets < s->max_ethqsets) {
+			adap->params.ethofld = 0;
+			s->eoqsets = 0;
+		} else {
+			s->eoqsets = s->max_ethqsets;
+		}
+		avail_qsets -= s->eoqsets;
+	}
+
+	/* Mirror queues must follow same scheme as normal Ethernet
+	 * Queues, when there are enough queues available. Otherwise,
+	 * allocate at least 1 queue per port. If even 1 queue is not
+	 * available, then disable mirror queues support.
+	 */
+	if (avail_qsets >= s->max_ethqsets)
+		s->mirrorqsets = s->max_ethqsets;
+	else if (avail_qsets >= adap->params.nports)
+		s->mirrorqsets = adap->params.nports;
+	else
+		s->mirrorqsets = 0;
+	avail_qsets -= s->mirrorqsets;
+
 	for (i = 0; i < ARRAY_SIZE(s->ethrxq); i++) {
 		struct sge_eth_rxq *r = &s->ethrxq[i];
 
@@ -4101,8 +5795,13 @@ static void cfg_queues(struct adapter *adap)
 	for (i = 0; i < ARRAY_SIZE(s->ctrlq); i++)
 		s->ctrlq[i].q.size = 512;
 
+	if (!is_t4(adap->params.chip))
+		s->ptptxq.q.size = 8;
+
 	init_rspq(adap, &s->fw_evtq, 0, 1, 1024, 64);
 	init_rspq(adap, &s->intrq, 0, 1, 512, 64);
+
+	return 0;
 }
 
 /*
@@ -4133,42 +5832,61 @@ static void reduce_ethqs(struct adapter *adap, int n)
 	}
 }
 
-static int get_msix_info(struct adapter *adap)
+static int alloc_msix_info(struct adapter *adap, u32 num_vec)
 {
-	struct uld_msix_info *msix_info;
-	unsigned int max_ingq = 0;
-
-	if (is_offload(adap))
-		max_ingq += MAX_OFLD_QSETS * adap->num_ofld_uld;
-	if (is_pci_uld(adap))
-		max_ingq += MAX_OFLD_QSETS * adap->num_uld;
-
-	if (!max_ingq)
-		goto out;
+	struct msix_info *msix_info;
 
-	msix_info = kcalloc(max_ingq, sizeof(*msix_info), GFP_KERNEL);
+	msix_info = kcalloc(num_vec, sizeof(*msix_info), GFP_KERNEL);
 	if (!msix_info)
 		return -ENOMEM;
 
-	adap->msix_bmap_ulds.msix_bmap = kcalloc(BITS_TO_LONGS(max_ingq),
-						 sizeof(long), GFP_KERNEL);
-	if (!adap->msix_bmap_ulds.msix_bmap) {
+	adap->msix_bmap.msix_bmap = bitmap_zalloc(num_vec, GFP_KERNEL);
+	if (!adap->msix_bmap.msix_bmap) {
 		kfree(msix_info);
 		return -ENOMEM;
 	}
-	spin_lock_init(&adap->msix_bmap_ulds.lock);
-	adap->msix_info_ulds = msix_info;
-out:
+
+	spin_lock_init(&adap->msix_bmap.lock);
+	adap->msix_bmap.mapsize = num_vec;
+
+	adap->msix_info = msix_info;
 	return 0;
 }
 
 static void free_msix_info(struct adapter *adap)
 {
-	if (!(adap->num_uld && adap->num_ofld_uld))
-		return;
+	bitmap_free(adap->msix_bmap.msix_bmap);
+	kfree(adap->msix_info);
+}
+
+int cxgb4_get_msix_idx_from_bmap(struct adapter *adap)
+{
+	struct msix_bmap *bmap = &adap->msix_bmap;
+	unsigned int msix_idx;
+	unsigned long flags;
+
+	spin_lock_irqsave(&bmap->lock, flags);
+	msix_idx = find_first_zero_bit(bmap->msix_bmap, bmap->mapsize);
+	if (msix_idx < bmap->mapsize) {
+		__set_bit(msix_idx, bmap->msix_bmap);
+	} else {
+		spin_unlock_irqrestore(&bmap->lock, flags);
+		return -ENOSPC;
+	}
+
+	spin_unlock_irqrestore(&bmap->lock, flags);
+	return msix_idx;
+}
+
+void cxgb4_free_msix_idx_in_bmap(struct adapter *adap,
+				 unsigned int msix_idx)
+{
+	struct msix_bmap *bmap = &adap->msix_bmap;
+	unsigned long flags;
 
-	kfree(adap->msix_info_ulds);
-	kfree(adap->msix_bmap_ulds.msix_bmap);
+	spin_lock_irqsave(&bmap->lock, flags);
+	__clear_bit(msix_idx, bmap->msix_bmap);
+	spin_unlock_irqrestore(&bmap->lock, flags);
 }
 
 /* 2 MSI-X vectors needed for the FW queue and non-data interrupts */
@@ -4176,88 +5894,194 @@ static void free_msix_info(struct adapter *adap)
 
 static int enable_msix(struct adapter *adap)
 {
-	int ofld_need = 0, uld_need = 0;
-	int i, j, want, need, allocated;
+	u32 eth_need, uld_need = 0, ethofld_need = 0, mirror_need = 0;
+	u32 ethqsets = 0, ofldqsets = 0, eoqsets = 0, mirrorqsets = 0;
+	u8 num_uld = 0, nchan = adap->params.nports;
+	u32 i, want, need, num_vec;
 	struct sge *s = &adap->sge;
-	unsigned int nchan = adap->params.nports;
 	struct msix_entry *entries;
-	int max_ingq = MAX_INGQ;
-
-	if (is_pci_uld(adap))
-		max_ingq += (MAX_OFLD_QSETS * adap->num_uld);
-	if (is_offload(adap))
-		max_ingq += (MAX_OFLD_QSETS * adap->num_ofld_uld);
-	entries = kmalloc(sizeof(*entries) * (max_ingq + 1),
-			  GFP_KERNEL);
-	if (!entries)
-		return -ENOMEM;
-
-	/* map for msix */
-	if (get_msix_info(adap)) {
-		adap->params.offload = 0;
-		adap->params.crypto = 0;
-	}
-
-	for (i = 0; i < max_ingq + 1; ++i)
-		entries[i].entry = i;
+	struct port_info *pi;
+	int allocated, ret;
 
-	want = s->max_ethqsets + EXTRA_VECS;
-	if (is_offload(adap)) {
-		want += adap->num_ofld_uld * s->ofldqsets;
-		ofld_need = adap->num_ofld_uld * nchan;
-	}
-	if (is_pci_uld(adap)) {
-		want += adap->num_uld * s->ofldqsets;
-		uld_need = adap->num_uld * nchan;
-	}
+	want = s->max_ethqsets;
 #ifdef CONFIG_CHELSIO_T4_DCB
 	/* For Data Center Bridging we need 8 Ethernet TX Priority Queues for
 	 * each port.
 	 */
-	need = 8 * adap->params.nports + EXTRA_VECS + ofld_need + uld_need;
+	need = 8 * nchan;
 #else
-	need = adap->params.nports + EXTRA_VECS + ofld_need + uld_need;
+	need = nchan;
 #endif
+	eth_need = need;
+	if (is_uld(adap)) {
+		num_uld = adap->num_ofld_uld + adap->num_uld;
+		want += num_uld * s->ofldqsets;
+		uld_need = num_uld * nchan;
+		need += uld_need;
+	}
+
+	if (is_ethofld(adap)) {
+		want += s->eoqsets;
+		ethofld_need = eth_need;
+		need += ethofld_need;
+	}
+
+	if (s->mirrorqsets) {
+		want += s->mirrorqsets;
+		mirror_need = nchan;
+		need += mirror_need;
+	}
+
+	want += EXTRA_VECS;
+	need += EXTRA_VECS;
+
+	entries = kmalloc_array(want, sizeof(*entries), GFP_KERNEL);
+	if (!entries)
+		return -ENOMEM;
+
+	for (i = 0; i < want; i++)
+		entries[i].entry = i;
+
 	allocated = pci_enable_msix_range(adap->pdev, entries, need, want);
 	if (allocated < 0) {
-		dev_info(adap->pdev_dev, "not enough MSI-X vectors left,"
-			 " not using MSI-X\n");
-		kfree(entries);
-		return allocated;
+		/* Disable offload and attempt to get vectors for NIC
+		 * only mode.
+		 */
+		want = s->max_ethqsets + EXTRA_VECS;
+		need = eth_need + EXTRA_VECS;
+		allocated = pci_enable_msix_range(adap->pdev, entries,
+						  need, want);
+		if (allocated < 0) {
+			dev_info(adap->pdev_dev,
+				 "Disabling MSI-X due to insufficient MSI-X vectors\n");
+			ret = allocated;
+			goto out_free;
+		}
+
+		dev_info(adap->pdev_dev,
+			 "Disabling offload due to insufficient MSI-X vectors\n");
+		adap->params.offload = 0;
+		adap->params.crypto = 0;
+		adap->params.ethofld = 0;
+		s->ofldqsets = 0;
+		s->eoqsets = 0;
+		s->mirrorqsets = 0;
+		uld_need = 0;
+		ethofld_need = 0;
+		mirror_need = 0;
+	}
+
+	num_vec = allocated;
+	if (num_vec < want) {
+		/* Distribute available vectors to the various queue groups.
+		 * Every group gets its minimum requirement and NIC gets top
+		 * priority for leftovers.
+		 */
+		ethqsets = eth_need;
+		if (is_uld(adap))
+			ofldqsets = nchan;
+		if (is_ethofld(adap))
+			eoqsets = ethofld_need;
+		if (s->mirrorqsets)
+			mirrorqsets = mirror_need;
+
+		num_vec -= need;
+		while (num_vec) {
+			if (num_vec < eth_need + ethofld_need ||
+			    ethqsets > s->max_ethqsets)
+				break;
+
+			for_each_port(adap, i) {
+				pi = adap2pinfo(adap, i);
+				if (pi->nqsets < 2)
+					continue;
+
+				ethqsets++;
+				num_vec--;
+				if (ethofld_need) {
+					eoqsets++;
+					num_vec--;
+				}
+			}
+		}
+
+		if (is_uld(adap)) {
+			while (num_vec) {
+				if (num_vec < uld_need ||
+				    ofldqsets > s->ofldqsets)
+					break;
+
+				ofldqsets++;
+				num_vec -= uld_need;
+			}
+		}
+
+		if (s->mirrorqsets) {
+			while (num_vec) {
+				if (num_vec < mirror_need ||
+				    mirrorqsets > s->mirrorqsets)
+					break;
+
+				mirrorqsets++;
+				num_vec -= mirror_need;
+			}
+		}
+	} else {
+		ethqsets = s->max_ethqsets;
+		if (is_uld(adap))
+			ofldqsets = s->ofldqsets;
+		if (is_ethofld(adap))
+			eoqsets = s->eoqsets;
+		if (s->mirrorqsets)
+			mirrorqsets = s->mirrorqsets;
 	}
 
-	/* Distribute available vectors to the various queue groups.
-	 * Every group gets its minimum requirement and NIC gets top
-	 * priority for leftovers.
-	 */
-	i = allocated - EXTRA_VECS - ofld_need - uld_need;
-	if (i < s->max_ethqsets) {
-		s->max_ethqsets = i;
-		if (i < s->ethqsets)
-			reduce_ethqs(adap, i);
+	if (ethqsets < s->max_ethqsets) {
+		s->max_ethqsets = ethqsets;
+		reduce_ethqs(adap, ethqsets);
 	}
+
 	if (is_uld(adap)) {
-		if (allocated < want)
-			s->nqs_per_uld = nchan;
-		else
-			s->nqs_per_uld = s->ofldqsets;
+		s->ofldqsets = ofldqsets;
+		s->nqs_per_uld = s->ofldqsets;
 	}
 
-	for (i = 0; i < (s->max_ethqsets + EXTRA_VECS); ++i)
-		adap->msix_info[i].vec = entries[i].vector;
-	if (is_uld(adap)) {
-		for (j = 0 ; i < allocated; ++i, j++) {
-			adap->msix_info_ulds[j].vec = entries[i].vector;
-			adap->msix_info_ulds[j].idx = i;
+	if (is_ethofld(adap))
+		s->eoqsets = eoqsets;
+
+	if (s->mirrorqsets) {
+		s->mirrorqsets = mirrorqsets;
+		for_each_port(adap, i) {
+			pi = adap2pinfo(adap, i);
+			pi->nmirrorqsets = s->mirrorqsets / nchan;
+			mutex_init(&pi->vi_mirror_mutex);
 		}
-		adap->msix_bmap_ulds.mapsize = j;
 	}
-	dev_info(adap->pdev_dev, "%d MSI-X vectors allocated, "
-		 "nic %d per uld %d\n",
-		 allocated, s->max_ethqsets, s->nqs_per_uld);
+
+	/* map for msix */
+	ret = alloc_msix_info(adap, allocated);
+	if (ret)
+		goto out_disable_msix;
+
+	for (i = 0; i < allocated; i++) {
+		adap->msix_info[i].vec = entries[i].vector;
+		adap->msix_info[i].idx = i;
+	}
+
+	dev_info(adap->pdev_dev,
+		 "%d MSI-X vectors allocated, nic %d eoqsets %d per uld %d mirrorqsets %d\n",
+		 allocated, s->max_ethqsets, s->eoqsets, s->nqs_per_uld,
+		 s->mirrorqsets);
 
 	kfree(entries);
 	return 0;
+
+out_disable_msix:
+	pci_disable_msix(adap->pdev);
+
+out_free:
+	kfree(entries);
+	return ret;
 }
 
 #undef EXTRA_VECS
@@ -4281,138 +6105,17 @@ static int init_rss(struct adapter *adap)
 	return 0;
 }
 
-static int cxgb4_get_pcie_dev_link_caps(struct adapter *adap,
-					enum pci_bus_speed *speed,
-					enum pcie_link_width *width)
-{
-	u32 lnkcap1, lnkcap2;
-	int err1, err2;
-
-#define  PCIE_MLW_CAP_SHIFT 4   /* start of MLW mask in link capabilities */
-
-	*speed = PCI_SPEED_UNKNOWN;
-	*width = PCIE_LNK_WIDTH_UNKNOWN;
-
-	err1 = pcie_capability_read_dword(adap->pdev, PCI_EXP_LNKCAP,
-					  &lnkcap1);
-	err2 = pcie_capability_read_dword(adap->pdev, PCI_EXP_LNKCAP2,
-					  &lnkcap2);
-	if (!err2 && lnkcap2) { /* PCIe r3.0-compliant */
-		if (lnkcap2 & PCI_EXP_LNKCAP2_SLS_8_0GB)
-			*speed = PCIE_SPEED_8_0GT;
-		else if (lnkcap2 & PCI_EXP_LNKCAP2_SLS_5_0GB)
-			*speed = PCIE_SPEED_5_0GT;
-		else if (lnkcap2 & PCI_EXP_LNKCAP2_SLS_2_5GB)
-			*speed = PCIE_SPEED_2_5GT;
-	}
-	if (!err1) {
-		*width = (lnkcap1 & PCI_EXP_LNKCAP_MLW) >> PCIE_MLW_CAP_SHIFT;
-		if (!lnkcap2) { /* pre-r3.0 */
-			if (lnkcap1 & PCI_EXP_LNKCAP_SLS_5_0GB)
-				*speed = PCIE_SPEED_5_0GT;
-			else if (lnkcap1 & PCI_EXP_LNKCAP_SLS_2_5GB)
-				*speed = PCIE_SPEED_2_5GT;
-		}
-	}
-
-	if (*speed == PCI_SPEED_UNKNOWN || *width == PCIE_LNK_WIDTH_UNKNOWN)
-		return err1 ? err1 : err2 ? err2 : -EINVAL;
-	return 0;
-}
-
-static void cxgb4_check_pcie_caps(struct adapter *adap)
-{
-	enum pcie_link_width width, width_cap;
-	enum pci_bus_speed speed, speed_cap;
-
-#define PCIE_SPEED_STR(speed) \
-	(speed == PCIE_SPEED_8_0GT ? "8.0GT/s" : \
-	 speed == PCIE_SPEED_5_0GT ? "5.0GT/s" : \
-	 speed == PCIE_SPEED_2_5GT ? "2.5GT/s" : \
-	 "Unknown")
-
-	if (cxgb4_get_pcie_dev_link_caps(adap, &speed_cap, &width_cap)) {
-		dev_warn(adap->pdev_dev,
-			 "Unable to determine PCIe device BW capabilities\n");
-		return;
-	}
-
-	if (pcie_get_minimum_link(adap->pdev, &speed, &width) ||
-	    speed == PCI_SPEED_UNKNOWN || width == PCIE_LNK_WIDTH_UNKNOWN) {
-		dev_warn(adap->pdev_dev,
-			 "Unable to determine PCI Express bandwidth.\n");
-		return;
-	}
-
-	dev_info(adap->pdev_dev, "PCIe link speed is %s, device supports %s\n",
-		 PCIE_SPEED_STR(speed), PCIE_SPEED_STR(speed_cap));
-	dev_info(adap->pdev_dev, "PCIe link width is x%d, device supports x%d\n",
-		 width, width_cap);
-	if (speed < speed_cap || width < width_cap)
-		dev_info(adap->pdev_dev,
-			 "A slot with more lanes and/or higher speed is "
-			 "suggested for optimal performance.\n");
-}
-
 /* Dump basic information about the adapter */
 static void print_adapter_info(struct adapter *adapter)
 {
-	/* Device information */
-	dev_info(adapter->pdev_dev, "Chelsio %s rev %d\n",
-		 adapter->params.vpd.id,
-		 CHELSIO_CHIP_RELEASE(adapter->params.chip));
-	dev_info(adapter->pdev_dev, "S/N: %s, P/N: %s\n",
-		 adapter->params.vpd.sn, adapter->params.vpd.pn);
-
-	/* Firmware Version */
-	if (!adapter->params.fw_vers)
-		dev_warn(adapter->pdev_dev, "No firmware loaded\n");
-	else
-		dev_info(adapter->pdev_dev, "Firmware version: %u.%u.%u.%u\n",
-			 FW_HDR_FW_VER_MAJOR_G(adapter->params.fw_vers),
-			 FW_HDR_FW_VER_MINOR_G(adapter->params.fw_vers),
-			 FW_HDR_FW_VER_MICRO_G(adapter->params.fw_vers),
-			 FW_HDR_FW_VER_BUILD_G(adapter->params.fw_vers));
-
-	/* Bootstrap Firmware Version. (Some adapters don't have Bootstrap
-	 * Firmware, so dev_info() is more appropriate here.)
-	 */
-	if (!adapter->params.bs_vers)
-		dev_info(adapter->pdev_dev, "No bootstrap loaded\n");
-	else
-		dev_info(adapter->pdev_dev, "Bootstrap version: %u.%u.%u.%u\n",
-			 FW_HDR_FW_VER_MAJOR_G(adapter->params.bs_vers),
-			 FW_HDR_FW_VER_MINOR_G(adapter->params.bs_vers),
-			 FW_HDR_FW_VER_MICRO_G(adapter->params.bs_vers),
-			 FW_HDR_FW_VER_BUILD_G(adapter->params.bs_vers));
-
-	/* TP Microcode Version */
-	if (!adapter->params.tp_vers)
-		dev_warn(adapter->pdev_dev, "No TP Microcode loaded\n");
-	else
-		dev_info(adapter->pdev_dev,
-			 "TP Microcode version: %u.%u.%u.%u\n",
-			 FW_HDR_FW_VER_MAJOR_G(adapter->params.tp_vers),
-			 FW_HDR_FW_VER_MINOR_G(adapter->params.tp_vers),
-			 FW_HDR_FW_VER_MICRO_G(adapter->params.tp_vers),
-			 FW_HDR_FW_VER_BUILD_G(adapter->params.tp_vers));
-
-	/* Expansion ROM version */
-	if (!adapter->params.er_vers)
-		dev_info(adapter->pdev_dev, "No Expansion ROM loaded\n");
-	else
-		dev_info(adapter->pdev_dev,
-			 "Expansion ROM version: %u.%u.%u.%u\n",
-			 FW_HDR_FW_VER_MAJOR_G(adapter->params.er_vers),
-			 FW_HDR_FW_VER_MINOR_G(adapter->params.er_vers),
-			 FW_HDR_FW_VER_MICRO_G(adapter->params.er_vers),
-			 FW_HDR_FW_VER_BUILD_G(adapter->params.er_vers));
+	/* Hardware/Firmware/etc. Version/Revision IDs */
+	t4_dump_version_info(adapter);
 
 	/* Software/Hardware configuration */
 	dev_info(adapter->pdev_dev, "Configuration: %sNIC %s, %s capable\n",
 		 is_offload(adapter) ? "R" : "",
-		 ((adapter->flags & USING_MSIX) ? "MSI-X" :
-		  (adapter->flags & USING_MSI) ? "MSI" : ""),
+		 ((adapter->flags & CXGB4_USING_MSIX) ? "MSI-X" :
+		  (adapter->flags & CXGB4_USING_MSI) ? "MSI" : ""),
 		 is_offload(adapter) ? "Offload" : "non-Offload");
 }
 
@@ -4420,40 +6123,32 @@ static void print_port_info(const struct net_device *dev)
 {
 	char buf[80];
 	char *bufp = buf;
-	const char *spd = "";
 	const struct port_info *pi = netdev_priv(dev);
 	const struct adapter *adap = pi->adapter;
 
-	if (adap->params.pci.speed == PCI_EXP_LNKSTA_CLS_2_5GB)
-		spd = " 2.5 GT/s";
-	else if (adap->params.pci.speed == PCI_EXP_LNKSTA_CLS_5_0GB)
-		spd = " 5 GT/s";
-	else if (adap->params.pci.speed == PCI_EXP_LNKSTA_CLS_8_0GB)
-		spd = " 8 GT/s";
-
-	if (pi->link_cfg.supported & FW_PORT_CAP_SPEED_100M)
+	if (pi->link_cfg.pcaps & FW_PORT_CAP32_SPEED_100M)
 		bufp += sprintf(bufp, "100M/");
-	if (pi->link_cfg.supported & FW_PORT_CAP_SPEED_1G)
+	if (pi->link_cfg.pcaps & FW_PORT_CAP32_SPEED_1G)
 		bufp += sprintf(bufp, "1G/");
-	if (pi->link_cfg.supported & FW_PORT_CAP_SPEED_10G)
+	if (pi->link_cfg.pcaps & FW_PORT_CAP32_SPEED_10G)
 		bufp += sprintf(bufp, "10G/");
-	if (pi->link_cfg.supported & FW_PORT_CAP_SPEED_25G)
+	if (pi->link_cfg.pcaps & FW_PORT_CAP32_SPEED_25G)
 		bufp += sprintf(bufp, "25G/");
-	if (pi->link_cfg.supported & FW_PORT_CAP_SPEED_40G)
+	if (pi->link_cfg.pcaps & FW_PORT_CAP32_SPEED_40G)
 		bufp += sprintf(bufp, "40G/");
-	if (pi->link_cfg.supported & FW_PORT_CAP_SPEED_100G)
+	if (pi->link_cfg.pcaps & FW_PORT_CAP32_SPEED_50G)
+		bufp += sprintf(bufp, "50G/");
+	if (pi->link_cfg.pcaps & FW_PORT_CAP32_SPEED_100G)
 		bufp += sprintf(bufp, "100G/");
+	if (pi->link_cfg.pcaps & FW_PORT_CAP32_SPEED_200G)
+		bufp += sprintf(bufp, "200G/");
+	if (pi->link_cfg.pcaps & FW_PORT_CAP32_SPEED_400G)
+		bufp += sprintf(bufp, "400G/");
 	if (bufp != buf)
 		--bufp;
 	sprintf(bufp, "BASE-%s", t4_get_port_type_description(pi->port_type));
 
-	netdev_info(dev, "%s: Chelsio %s (%s) %s\n",
-		    dev->name, adap->params.vpd.id, adap->name, buf);
-}
-
-static void enable_pcie_relaxed_ordering(struct pci_dev *dev)
-{
-	pcie_capability_set_word(dev, PCI_EXP_DEVCTL, PCI_EXP_DEVCTL_RELAX_EN);
+	netdev_info(dev, "Chelsio %s %s\n", adap->params.vpd.id, buf);
 }
 
 /*
@@ -4467,16 +6162,22 @@ static void free_some_resources(struct adapter *adapter)
 {
 	unsigned int i;
 
+	kvfree(adapter->smt);
 	kvfree(adapter->l2t);
+	kvfree(adapter->srq);
 	t4_cleanup_sched(adapter);
 	kvfree(adapter->tids.tid_tab);
+	cxgb4_cleanup_tc_matchall(adapter);
+	cxgb4_cleanup_tc_mqprio(adapter);
+	cxgb4_cleanup_tc_flower(adapter);
 	cxgb4_cleanup_tc_u32(adapter);
+	cxgb4_cleanup_ethtool_filters(adapter);
 	kfree(adapter->sge.egr_map);
 	kfree(adapter->sge.ingr_map);
-	kfree(adapter->sge.starving_fl);
-	kfree(adapter->sge.txq_maperr);
+	bitmap_free(adapter->sge.starving_fl);
+	bitmap_free(adapter->sge.txq_maperr);
 #ifdef CONFIG_DEBUG_FS
-	kfree(adapter->sge.blocked_fl);
+	bitmap_free(adapter->sge.blocked_fl);
 #endif
 	disable_msi(adapter);
 
@@ -4490,23 +6191,21 @@ static void free_some_resources(struct adapter *adapter)
 			kfree(adap2pinfo(adapter, i)->rss);
 			free_netdev(adapter->port[i]);
 		}
-	if (adapter->flags & FW_OK)
+	if (adapter->flags & CXGB4_FW_OK)
 		t4_fw_bye(adapter, adapter->pf);
 }
 
-#define TSO_FLAGS (NETIF_F_TSO | NETIF_F_TSO6 | NETIF_F_TSO_ECN)
+#define TSO_FLAGS (NETIF_F_TSO | NETIF_F_TSO6 | NETIF_F_TSO_ECN | \
+		   NETIF_F_GSO_UDP_L4)
 #define VLAN_FEAT (NETIF_F_SG | NETIF_F_IP_CSUM | TSO_FLAGS | \
-		   NETIF_F_IPV6_CSUM | NETIF_F_HIGHDMA)
+		   NETIF_F_GRO | NETIF_F_IPV6_CSUM | NETIF_F_HIGHDMA)
 #define SEGMENT_SIZE 128
 
-static int get_chip_type(struct pci_dev *pdev, u32 pl_rev)
+static int t4_get_chip_type(struct adapter *adap, int ver)
 {
-	u16 device_id;
-
-	/* Retrieve adapter's device ID */
-	pci_read_config_word(pdev, PCI_DEVICE_ID, &device_id);
+	u32 pl_rev = REV_G(t4_read_reg(adap, PL_REV_A));
 
-	switch (device_id >> 12) {
+	switch (ver) {
 	case CHELSIO_T4:
 		return CHELSIO_CHIP_CODE(CHELSIO_T4, pl_rev);
 	case CHELSIO_T5:
@@ -4514,14 +6213,13 @@ static int get_chip_type(struct pci_dev *pdev, u32 pl_rev)
 	case CHELSIO_T6:
 		return CHELSIO_CHIP_CODE(CHELSIO_T6, pl_rev);
 	default:
-		dev_err(&pdev->dev, "Device %d is not supported\n",
-			device_id);
+		break;
 	}
 	return -EINVAL;
 }
 
 #ifdef CONFIG_PCI_IOV
-static void dummy_setup(struct net_device *dev)
+static void cxgb4_mgmt_setup(struct net_device *dev)
 {
 	dev->type = ARPHRD_NONE;
 	dev->mtu = 0;
@@ -4534,38 +6232,6 @@ static void dummy_setup(struct net_device *dev)
 	/* Initialize the device structure. */
 	dev->netdev_ops = &cxgb4_mgmt_netdev_ops;
 	dev->ethtool_ops = &cxgb4_mgmt_ethtool_ops;
-	dev->destructor = free_netdev;
-}
-
-static int config_mgmt_dev(struct pci_dev *pdev)
-{
-	struct adapter *adap = pci_get_drvdata(pdev);
-	struct net_device *netdev;
-	struct port_info *pi;
-	char name[IFNAMSIZ];
-	int err;
-
-	snprintf(name, IFNAMSIZ, "mgmtpf%d%d", adap->adap_idx, adap->pf);
-	netdev = alloc_netdev(sizeof(struct port_info), name, NET_NAME_UNKNOWN,
-			      dummy_setup);
-	if (!netdev)
-		return -ENOMEM;
-
-	pi = netdev_priv(netdev);
-	pi->adapter = adap;
-	pi->port_id = adap->pf % adap->params.nports;
-	SET_NETDEV_DEV(netdev, &pdev->dev);
-
-	adap->port[0] = netdev;
-
-	err = register_netdev(adap->port[0]);
-	if (err) {
-		pr_info("Unable to register VF mgmt netdev %s\n", name);
-		free_netdev(adap->port[0]);
-		adap->port[0] = NULL;
-		return err;
-	}
-	return 0;
 }
 
 static int cxgb4_iov_configure(struct pci_dev *pdev, int num_vfs)
@@ -4576,12 +6242,9 @@ static int cxgb4_iov_configure(struct pci_dev *pdev, int num_vfs)
 	u32 pcie_fw;
 
 	pcie_fw = readl(adap->regs + PCIE_FW_A);
-	/* Check if cxgb4 is the MASTER and fw is initialized */
-	if (!(pcie_fw & PCIE_FW_INIT_F) ||
-	    !(pcie_fw & PCIE_FW_MASTER_VLD_F) ||
-	    PCIE_FW_MASTER_G(pcie_fw) != 4) {
-		dev_warn(&pdev->dev,
-			 "cxgb4 driver needs to be MASTER to support SRIOV\n");
+	/* Check if fw is initialized */
+	if (!(pcie_fw & PCIE_FW_INIT_F)) {
+		dev_warn(&pdev->dev, "Device not initialized\n");
 		return -EOPNOTSUPP;
 	}
 
@@ -4591,63 +6254,335 @@ static int cxgb4_iov_configure(struct pci_dev *pdev, int num_vfs)
 	if (current_vfs && pci_vfs_assigned(pdev)) {
 		dev_err(&pdev->dev,
 			"Cannot modify SR-IOV while VFs are assigned\n");
-		num_vfs = current_vfs;
-		return num_vfs;
+		return current_vfs;
 	}
-
-	/* Disable SRIOV when zero is passed.
-	 * One needs to disable SRIOV before modifying it, else
-	 * stack throws the below warning:
-	 * " 'n' VFs already enabled. Disable before enabling 'm' VFs."
+	/* Note that the upper-level code ensures that we're never called with
+	 * a non-zero "num_vfs" when we already have VFs instantiated.  But
+	 * it never hurts to code defensively.
 	 */
+	if (num_vfs != 0 && current_vfs != 0)
+		return -EBUSY;
+
+	/* Nothing to do for no change. */
+	if (num_vfs == current_vfs)
+		return num_vfs;
+
+	/* Disable SRIOV when zero is passed. */
 	if (!num_vfs) {
 		pci_disable_sriov(pdev);
-		if (adap->port[0]) {
-			unregister_netdev(adap->port[0]);
-			adap->port[0] = NULL;
-		}
+		/* free VF Management Interface */
+		unregister_netdev(adap->port[0]);
+		free_netdev(adap->port[0]);
+		adap->port[0] = NULL;
+
 		/* free VF resources */
+		adap->num_vfs = 0;
 		kfree(adap->vfinfo);
 		adap->vfinfo = NULL;
-		adap->num_vfs = 0;
-		return num_vfs;
+		return 0;
 	}
 
-	if (num_vfs != current_vfs) {
-		err = pci_enable_sriov(pdev, num_vfs);
+	if (!current_vfs) {
+		struct fw_pfvf_cmd port_cmd, port_rpl;
+		struct net_device *netdev;
+		unsigned int pmask, port;
+		struct pci_dev *pbridge;
+		struct port_info *pi;
+		char name[IFNAMSIZ];
+		u32 devcap2;
+		u16 flags;
+
+		/* If we want to instantiate Virtual Functions, then our
+		 * parent bridge's PCI-E needs to support Alternative Routing
+		 * ID (ARI) because our VFs will show up at function offset 8
+		 * and above.
+		 */
+		pbridge = pdev->bus->self;
+		pcie_capability_read_word(pbridge, PCI_EXP_FLAGS, &flags);
+		pcie_capability_read_dword(pbridge, PCI_EXP_DEVCAP2, &devcap2);
+
+		if ((flags & PCI_EXP_FLAGS_VERS) < 2 ||
+		    !(devcap2 & PCI_EXP_DEVCAP2_ARI)) {
+			/* Our parent bridge does not support ARI so issue a
+			 * warning and skip instantiating the VFs.  They
+			 * won't be reachable.
+			 */
+			dev_warn(&pdev->dev, "Parent bridge %02x:%02x.%x doesn't support ARI; can't instantiate Virtual Functions\n",
+				 pbridge->bus->number, PCI_SLOT(pbridge->devfn),
+				 PCI_FUNC(pbridge->devfn));
+			return -ENOTSUPP;
+		}
+		memset(&port_cmd, 0, sizeof(port_cmd));
+		port_cmd.op_to_vfn = cpu_to_be32(FW_CMD_OP_V(FW_PFVF_CMD) |
+						 FW_CMD_REQUEST_F |
+						 FW_CMD_READ_F |
+						 FW_PFVF_CMD_PFN_V(adap->pf) |
+						 FW_PFVF_CMD_VFN_V(0));
+		port_cmd.retval_len16 = cpu_to_be32(FW_LEN16(port_cmd));
+		err = t4_wr_mbox(adap, adap->mbox, &port_cmd, sizeof(port_cmd),
+				 &port_rpl);
 		if (err)
 			return err;
+		pmask = FW_PFVF_CMD_PMASK_G(be32_to_cpu(port_rpl.type_to_neq));
+		port = ffs(pmask) - 1;
+		/* Allocate VF Management Interface. */
+		snprintf(name, IFNAMSIZ, "mgmtpf%d,%d", adap->adap_idx,
+			 adap->pf);
+		netdev = alloc_netdev(sizeof(struct port_info),
+				      name, NET_NAME_UNKNOWN, cxgb4_mgmt_setup);
+		if (!netdev)
+			return -ENOMEM;
 
-		adap->num_vfs = num_vfs;
-		err = config_mgmt_dev(pdev);
-		if (err)
+		pi = netdev_priv(netdev);
+		pi->adapter = adap;
+		pi->lport = port;
+		pi->tx_chan = port;
+		SET_NETDEV_DEV(netdev, &pdev->dev);
+
+		adap->port[0] = netdev;
+		pi->port_id = 0;
+
+		err = register_netdev(adap->port[0]);
+		if (err) {
+			pr_info("Unable to register VF mgmt netdev %s\n", name);
+			free_netdev(adap->port[0]);
+			adap->port[0] = NULL;
 			return err;
+		}
+		/* Allocate and set up VF Information. */
+		adap->vfinfo = kcalloc(pci_sriov_get_totalvfs(pdev),
+				       sizeof(struct vf_info), GFP_KERNEL);
+		if (!adap->vfinfo) {
+			unregister_netdev(adap->port[0]);
+			free_netdev(adap->port[0]);
+			adap->port[0] = NULL;
+			return -ENOMEM;
+		}
+		cxgb4_mgmt_fill_vf_station_mac_addr(adap);
+	}
+	/* Instantiate the requested number of VFs. */
+	err = pci_enable_sriov(pdev, num_vfs);
+	if (err) {
+		pr_info("Unable to instantiate %d VFs\n", num_vfs);
+		if (!current_vfs) {
+			unregister_netdev(adap->port[0]);
+			free_netdev(adap->port[0]);
+			adap->port[0] = NULL;
+			kfree(adap->vfinfo);
+			adap->vfinfo = NULL;
+		}
+		return err;
 	}
 
-	adap->vfinfo = kcalloc(adap->num_vfs,
-			       sizeof(struct vf_info), GFP_KERNEL);
-	if (adap->vfinfo)
-		fill_vf_station_mac_addr(adap);
+	adap->num_vfs = num_vfs;
 	return num_vfs;
 }
-#endif
+#endif /* CONFIG_PCI_IOV */
+
+#if IS_ENABLED(CONFIG_CHELSIO_TLS_DEVICE) || IS_ENABLED(CONFIG_CHELSIO_IPSEC_INLINE)
+
+static int chcr_offload_state(struct adapter *adap,
+			      enum cxgb4_netdev_tls_ops op_val)
+{
+	switch (op_val) {
+#if IS_ENABLED(CONFIG_CHELSIO_TLS_DEVICE)
+	case CXGB4_TLSDEV_OPS:
+		if (!adap->uld[CXGB4_ULD_KTLS].handle) {
+			dev_dbg(adap->pdev_dev, "ch_ktls driver is not loaded\n");
+			return -EOPNOTSUPP;
+		}
+		if (!adap->uld[CXGB4_ULD_KTLS].tlsdev_ops) {
+			dev_dbg(adap->pdev_dev,
+				"ch_ktls driver has no registered tlsdev_ops\n");
+			return -EOPNOTSUPP;
+		}
+		break;
+#endif /* CONFIG_CHELSIO_TLS_DEVICE */
+#if IS_ENABLED(CONFIG_CHELSIO_IPSEC_INLINE)
+	case CXGB4_XFRMDEV_OPS:
+		if (!adap->uld[CXGB4_ULD_IPSEC].handle) {
+			dev_dbg(adap->pdev_dev, "chipsec driver is not loaded\n");
+			return -EOPNOTSUPP;
+		}
+		if (!adap->uld[CXGB4_ULD_IPSEC].xfrmdev_ops) {
+			dev_dbg(adap->pdev_dev,
+				"chipsec driver has no registered xfrmdev_ops\n");
+			return -EOPNOTSUPP;
+		}
+		break;
+#endif /* CONFIG_CHELSIO_IPSEC_INLINE */
+	default:
+		dev_dbg(adap->pdev_dev,
+			"driver has no support for offload %d\n", op_val);
+		return -EOPNOTSUPP;
+	}
+
+	return 0;
+}
+
+#endif /* CONFIG_CHELSIO_TLS_DEVICE || CONFIG_CHELSIO_IPSEC_INLINE */
+
+#if IS_ENABLED(CONFIG_CHELSIO_TLS_DEVICE)
+
+static int cxgb4_ktls_dev_add(struct net_device *netdev, struct sock *sk,
+			      enum tls_offload_ctx_dir direction,
+			      struct tls_crypto_info *crypto_info,
+			      u32 tcp_sn)
+{
+	struct adapter *adap = netdev2adap(netdev);
+	int ret;
+
+	mutex_lock(&uld_mutex);
+	ret = chcr_offload_state(adap, CXGB4_TLSDEV_OPS);
+	if (ret)
+		goto out_unlock;
+
+	ret = cxgb4_set_ktls_feature(adap, FW_PARAMS_PARAM_DEV_KTLS_HW_ENABLE);
+	if (ret)
+		goto out_unlock;
+
+	ret = adap->uld[CXGB4_ULD_KTLS].tlsdev_ops->tls_dev_add(netdev, sk,
+								direction,
+								crypto_info,
+								tcp_sn);
+	/* if there is a failure, clear the refcount */
+	if (ret)
+		cxgb4_set_ktls_feature(adap,
+				       FW_PARAMS_PARAM_DEV_KTLS_HW_DISABLE);
+out_unlock:
+	mutex_unlock(&uld_mutex);
+	return ret;
+}
+
+static void cxgb4_ktls_dev_del(struct net_device *netdev,
+			       struct tls_context *tls_ctx,
+			       enum tls_offload_ctx_dir direction)
+{
+	struct adapter *adap = netdev2adap(netdev);
+
+	mutex_lock(&uld_mutex);
+	if (chcr_offload_state(adap, CXGB4_TLSDEV_OPS))
+		goto out_unlock;
+
+	adap->uld[CXGB4_ULD_KTLS].tlsdev_ops->tls_dev_del(netdev, tls_ctx,
+							  direction);
+
+out_unlock:
+	cxgb4_set_ktls_feature(adap, FW_PARAMS_PARAM_DEV_KTLS_HW_DISABLE);
+	mutex_unlock(&uld_mutex);
+}
+
+static const struct tlsdev_ops cxgb4_ktls_ops = {
+	.tls_dev_add = cxgb4_ktls_dev_add,
+	.tls_dev_del = cxgb4_ktls_dev_del,
+};
+#endif /* CONFIG_CHELSIO_TLS_DEVICE */
+
+#if IS_ENABLED(CONFIG_CHELSIO_IPSEC_INLINE)
+
+static int cxgb4_xfrm_add_state(struct net_device *dev,
+				struct xfrm_state *x,
+				struct netlink_ext_ack *extack)
+{
+	struct adapter *adap = netdev2adap(dev);
+	int ret;
+
+	if (!mutex_trylock(&uld_mutex)) {
+		NL_SET_ERR_MSG_MOD(extack, "crypto uld critical resource is under use");
+		return -EBUSY;
+	}
+	ret = chcr_offload_state(adap, CXGB4_XFRMDEV_OPS);
+	if (ret)
+		goto out_unlock;
+
+	ret = adap->uld[CXGB4_ULD_IPSEC].xfrmdev_ops->xdo_dev_state_add(dev, x,
+									extack);
+
+out_unlock:
+	mutex_unlock(&uld_mutex);
+
+	return ret;
+}
+
+static void cxgb4_xfrm_del_state(struct net_device *dev, struct xfrm_state *x)
+{
+	struct adapter *adap = netdev2adap(dev);
+
+	if (!mutex_trylock(&uld_mutex)) {
+		dev_dbg(adap->pdev_dev,
+			"crypto uld critical resource is under use\n");
+		return;
+	}
+	if (chcr_offload_state(adap, CXGB4_XFRMDEV_OPS))
+		goto out_unlock;
+
+	adap->uld[CXGB4_ULD_IPSEC].xfrmdev_ops->xdo_dev_state_delete(dev, x);
+
+out_unlock:
+	mutex_unlock(&uld_mutex);
+}
+
+static void cxgb4_xfrm_free_state(struct net_device *dev, struct xfrm_state *x)
+{
+	struct adapter *adap = netdev2adap(dev);
+
+	if (!mutex_trylock(&uld_mutex)) {
+		dev_dbg(adap->pdev_dev,
+			"crypto uld critical resource is under use\n");
+		return;
+	}
+	if (chcr_offload_state(adap, CXGB4_XFRMDEV_OPS))
+		goto out_unlock;
+
+	adap->uld[CXGB4_ULD_IPSEC].xfrmdev_ops->xdo_dev_state_free(dev, x);
+
+out_unlock:
+	mutex_unlock(&uld_mutex);
+}
+
+static void cxgb4_advance_esn_state(struct xfrm_state *x)
+{
+	struct adapter *adap = netdev2adap(x->xso.dev);
+
+	if (x->xso.dir != XFRM_DEV_OFFLOAD_IN)
+		return;
+
+	if (!mutex_trylock(&uld_mutex)) {
+		dev_dbg(adap->pdev_dev,
+			"crypto uld critical resource is under use\n");
+		return;
+	}
+	if (chcr_offload_state(adap, CXGB4_XFRMDEV_OPS))
+		goto out_unlock;
+
+	adap->uld[CXGB4_ULD_IPSEC].xfrmdev_ops->xdo_dev_state_advance_esn(x);
+
+out_unlock:
+	mutex_unlock(&uld_mutex);
+}
+
+static const struct xfrmdev_ops cxgb4_xfrmdev_ops = {
+	.xdo_dev_state_add      = cxgb4_xfrm_add_state,
+	.xdo_dev_state_delete   = cxgb4_xfrm_del_state,
+	.xdo_dev_state_free     = cxgb4_xfrm_free_state,
+	.xdo_dev_state_advance_esn = cxgb4_advance_esn_state,
+};
+
+#endif /* CONFIG_CHELSIO_IPSEC_INLINE */
 
 static int init_one(struct pci_dev *pdev, const struct pci_device_id *ent)
 {
-	int func, i, err, s_qpp, qpp, num_seg;
-	struct port_info *pi;
-	bool highdma = false;
-	struct adapter *adapter = NULL;
 	struct net_device *netdev;
-	void __iomem *regs;
-	u32 whoami, pl_rev;
-	enum chip_type chip;
+	struct adapter *adapter;
 	static int adap_idx = 1;
-#ifdef CONFIG_PCI_IOV
-	u32 v, port_vec;
-#endif
-
-	printk_once(KERN_INFO "%s - version %s\n", DRV_DESC, DRV_VERSION);
+	int s_qpp, qpp, num_seg;
+	struct port_info *pi;
+	enum chip_type chip;
+	void __iomem *regs;
+	int func, chip_ver;
+	u16 device_id;
+	int i, err;
+	u32 whoami;
 
 	err = pci_request_regions(pdev, KBUILD_MODNAME);
 	if (err) {
@@ -4669,96 +6604,116 @@ static int init_one(struct pci_dev *pdev, const struct pci_device_id *ent)
 		goto out_disable_device;
 	}
 
+	adapter = kzalloc(sizeof(*adapter), GFP_KERNEL);
+	if (!adapter) {
+		err = -ENOMEM;
+		goto out_unmap_bar0;
+	}
+
+	adapter->regs = regs;
 	err = t4_wait_dev_ready(regs);
 	if (err < 0)
-		goto out_unmap_bar0;
+		goto out_free_adapter;
 
 	/* We control everything through one PF */
-	whoami = readl(regs + PL_WHOAMI_A);
-	pl_rev = REV_G(readl(regs + PL_REV_A));
-	chip = get_chip_type(pdev, pl_rev);
-	func = CHELSIO_CHIP_VERSION(chip) <= CHELSIO_T5 ?
-		SOURCEPF_G(whoami) : T6_SOURCEPF_G(whoami);
+	whoami = t4_read_reg(adapter, PL_WHOAMI_A);
+	pci_read_config_word(pdev, PCI_DEVICE_ID, &device_id);
+	chip = t4_get_chip_type(adapter, CHELSIO_PCI_ID_VER(device_id));
+	if ((int)chip < 0) {
+		dev_err(&pdev->dev, "Device %d is not supported\n", device_id);
+		err = chip;
+		goto out_free_adapter;
+	}
+	chip_ver = CHELSIO_CHIP_VERSION(chip);
+	func = chip_ver <= CHELSIO_T5 ?
+	       SOURCEPF_G(whoami) : T6_SOURCEPF_G(whoami);
+
+	adapter->pdev = pdev;
+	adapter->pdev_dev = &pdev->dev;
+	adapter->name = pci_name(pdev);
+	adapter->mbox = func;
+	adapter->pf = func;
+	adapter->params.chip = chip;
+	adapter->adap_idx = adap_idx;
+	adapter->msg_enable = DFLT_MSG_ENABLE;
+	adapter->mbox_log = kzalloc(sizeof(*adapter->mbox_log) +
+				    (sizeof(struct mbox_cmd) *
+				     T4_OS_LOG_MBOX_CMDS),
+				    GFP_KERNEL);
+	if (!adapter->mbox_log) {
+		err = -ENOMEM;
+		goto out_free_adapter;
+	}
+	spin_lock_init(&adapter->mbox_lock);
+	INIT_LIST_HEAD(&adapter->mlist.list);
+	adapter->mbox_log->size = T4_OS_LOG_MBOX_CMDS;
+	pci_set_drvdata(pdev, adapter);
+
 	if (func != ent->driver_data) {
-#ifndef CONFIG_PCI_IOV
-		iounmap(regs);
-#endif
 		pci_disable_device(pdev);
 		pci_save_state(pdev);        /* to restore SR-IOV later */
-		goto sriov;
+		return 0;
 	}
 
-	if (!pci_set_dma_mask(pdev, DMA_BIT_MASK(64))) {
-		highdma = true;
-		err = pci_set_consistent_dma_mask(pdev, DMA_BIT_MASK(64));
-		if (err) {
-			dev_err(&pdev->dev, "unable to obtain 64-bit DMA for "
-				"coherent allocations\n");
-			goto out_unmap_bar0;
-		}
-	} else {
-		err = pci_set_dma_mask(pdev, DMA_BIT_MASK(32));
-		if (err) {
-			dev_err(&pdev->dev, "no usable DMA configuration\n");
-			goto out_unmap_bar0;
-		}
+	err = dma_set_mask_and_coherent(&pdev->dev, DMA_BIT_MASK(64));
+	if (err) {
+		dev_err(&pdev->dev, "no usable DMA configuration\n");
+		goto out_free_adapter;
 	}
 
-	pci_enable_pcie_error_reporting(pdev);
-	enable_pcie_relaxed_ordering(pdev);
 	pci_set_master(pdev);
 	pci_save_state(pdev);
-
-	adapter = kzalloc(sizeof(*adapter), GFP_KERNEL);
-	if (!adapter) {
-		err = -ENOMEM;
-		goto out_unmap_bar0;
-	}
 	adap_idx++;
-
 	adapter->workq = create_singlethread_workqueue("cxgb4");
 	if (!adapter->workq) {
 		err = -ENOMEM;
 		goto out_free_adapter;
 	}
 
-	adapter->mbox_log = kzalloc(sizeof(*adapter->mbox_log) +
-				    (sizeof(struct mbox_cmd) *
-				     T4_OS_LOG_MBOX_CMDS),
-				    GFP_KERNEL);
-	if (!adapter->mbox_log) {
-		err = -ENOMEM;
-		goto out_free_adapter;
-	}
-	adapter->mbox_log->size = T4_OS_LOG_MBOX_CMDS;
-
 	/* PCI device has been enabled */
-	adapter->flags |= DEV_ENABLED;
-
-	adapter->regs = regs;
-	adapter->pdev = pdev;
-	adapter->pdev_dev = &pdev->dev;
-	adapter->name = pci_name(pdev);
-	adapter->mbox = func;
-	adapter->pf = func;
-	adapter->msg_enable = DFLT_MSG_ENABLE;
+	adapter->flags |= CXGB4_DEV_ENABLED;
 	memset(adapter->chan_map, 0xff, sizeof(adapter->chan_map));
 
+	/* If possible, we use PCIe Relaxed Ordering Attribute to deliver
+	 * Ingress Packet Data to Free List Buffers in order to allow for
+	 * chipset performance optimizations between the Root Complex and
+	 * Memory Controllers.  (Messages to the associated Ingress Queue
+	 * notifying new Packet Placement in the Free Lists Buffers will be
+	 * send without the Relaxed Ordering Attribute thus guaranteeing that
+	 * all preceding PCIe Transaction Layer Packets will be processed
+	 * first.)  But some Root Complexes have various issues with Upstream
+	 * Transaction Layer Packets with the Relaxed Ordering Attribute set.
+	 * The PCIe devices which under the Root Complexes will be cleared the
+	 * Relaxed Ordering bit in the configuration space, So we check our
+	 * PCIe configuration space to see if it's flagged with advice against
+	 * using Relaxed Ordering.
+	 */
+	if (!pcie_relaxed_ordering_enabled(pdev))
+		adapter->flags |= CXGB4_ROOT_NO_RELAXED_ORDERING;
+
 	spin_lock_init(&adapter->stats_lock);
 	spin_lock_init(&adapter->tid_release_lock);
 	spin_lock_init(&adapter->win0_lock);
-	spin_lock_init(&adapter->mbox_lock);
-
-	INIT_LIST_HEAD(&adapter->mlist.list);
 
 	INIT_WORK(&adapter->tid_release_task, process_tid_release_list);
 	INIT_WORK(&adapter->db_full_task, process_db_full);
 	INIT_WORK(&adapter->db_drop_task, process_db_drop);
+	INIT_WORK(&adapter->fatal_err_notify_task, notify_fatal_err);
 
 	err = t4_prep_adapter(adapter);
 	if (err)
 		goto out_free_adapter;
 
+	if (is_kdump_kernel()) {
+		/* Collect hardware state and append to /proc/vmcore */
+		err = cxgb4_cudbg_vmcore_add_dump(adapter);
+		if (err) {
+			dev_warn(adapter->pdev_dev,
+				 "Fail collecting vmcore device dump, err: %d. Continuing\n",
+				 err);
+			err = 0;
+		}
+	}
 
 	if (!is_t4(adapter->params.chip)) {
 		s_qpp = (QUEUESPERPAGEPF0_S +
@@ -4789,23 +6744,30 @@ static int init_one(struct pci_dev *pdev, const struct pci_device_id *ent)
 	}
 
 	setup_memwin(adapter);
-	err = adap_init0(adapter);
-#ifdef CONFIG_DEBUG_FS
-	bitmap_zero(adapter->sge.blocked_fl, adapter->sge.egr_sz);
-#endif
-	setup_memwin_rdma(adapter);
+	err = adap_init0(adapter, 0);
 	if (err)
 		goto out_unmap_bar;
 
+	setup_memwin_rdma(adapter);
+
 	/* configure SGE_STAT_CFG_A to read WC stats */
 	if (!is_t4(adapter->params.chip))
 		t4_write_reg(adapter, SGE_STAT_CFG_A, STATSOURCE_T5_V(7) |
 			     (is_t5(adapter->params.chip) ? STATMODE_V(0) :
 			      T6_STATMODE_V(0)));
 
+	/* Initialize hash mac addr list */
+	INIT_LIST_HEAD(&adapter->mac_hlist);
+
 	for_each_port(adapter, i) {
+		/* For supporting MQPRIO Offload, need some extra
+		 * queues for each ETHOFLD TIDs. Keep it equal to
+		 * MAX_ATIDs for now. Once we connect to firmware
+		 * later and query the EOTID params, we'll come to
+		 * know the actual # of EOTIDs supported.
+		 */
 		netdev = alloc_etherdev_mq(sizeof(struct port_info),
-					   MAX_ETH_QSETS);
+					   MAX_ETH_QSETS + MAX_ATIDS);
 		if (!netdev) {
 			err = -ENOMEM;
 			goto out_free_dev;
@@ -4822,31 +6784,64 @@ static int init_one(struct pci_dev *pdev, const struct pci_device_id *ent)
 
 		netdev->hw_features = NETIF_F_SG | TSO_FLAGS |
 			NETIF_F_IP_CSUM | NETIF_F_IPV6_CSUM |
-			NETIF_F_RXCSUM | NETIF_F_RXHASH |
+			NETIF_F_RXCSUM | NETIF_F_RXHASH | NETIF_F_GRO |
 			NETIF_F_HW_VLAN_CTAG_TX | NETIF_F_HW_VLAN_CTAG_RX |
-			NETIF_F_HW_TC;
-		if (highdma)
-			netdev->hw_features |= NETIF_F_HIGHDMA;
+			NETIF_F_HW_TC | NETIF_F_NTUPLE | NETIF_F_HIGHDMA;
+
+		if (chip_ver > CHELSIO_T5) {
+			netdev->hw_enc_features |= NETIF_F_IP_CSUM |
+						   NETIF_F_IPV6_CSUM |
+						   NETIF_F_RXCSUM |
+						   NETIF_F_GSO_UDP_TUNNEL |
+						   NETIF_F_GSO_UDP_TUNNEL_CSUM |
+						   NETIF_F_TSO | NETIF_F_TSO6;
+
+			netdev->hw_features |= NETIF_F_GSO_UDP_TUNNEL |
+					       NETIF_F_GSO_UDP_TUNNEL_CSUM |
+					       NETIF_F_HW_TLS_RECORD;
+
+			if (adapter->rawf_cnt)
+				netdev->udp_tunnel_nic_info = &cxgb_udp_tunnels;
+		}
+
 		netdev->features |= netdev->hw_features;
 		netdev->vlan_features = netdev->features & VLAN_FEAT;
+#if IS_ENABLED(CONFIG_CHELSIO_TLS_DEVICE)
+		if (pi->adapter->params.crypto & FW_CAPS_CONFIG_TLS_HW) {
+			netdev->hw_features |= NETIF_F_HW_TLS_TX;
+			netdev->tlsdev_ops = &cxgb4_ktls_ops;
+			/* initialize the refcount */
+			refcount_set(&pi->adapter->chcr_ktls.ktls_refcount, 0);
+		}
+#endif /* CONFIG_CHELSIO_TLS_DEVICE */
+#if IS_ENABLED(CONFIG_CHELSIO_IPSEC_INLINE)
+		if (pi->adapter->params.crypto & FW_CAPS_CONFIG_IPSEC_INLINE) {
+			netdev->hw_enc_features |= NETIF_F_HW_ESP;
+			netdev->features |= NETIF_F_HW_ESP;
+			netdev->xfrmdev_ops = &cxgb4_xfrmdev_ops;
+		}
+#endif /* CONFIG_CHELSIO_IPSEC_INLINE */
 
 		netdev->priv_flags |= IFF_UNICAST_FLT;
 
 		/* MTU range: 81 - 9600 */
-		netdev->min_mtu = 81;
+		netdev->min_mtu = 81;              /* accommodate SACK */
 		netdev->max_mtu = MAX_MTU;
 
 		netdev->netdev_ops = &cxgb4_netdev_ops;
 #ifdef CONFIG_CHELSIO_T4_DCB
 		netdev->dcbnl_ops = &cxgb4_dcb_ops;
 		cxgb4_dcb_state_init(netdev);
+		cxgb4_dcb_version_init(netdev);
 #endif
 		cxgb4_set_ethtool_ops(netdev);
 	}
 
+	cxgb4_init_ethtool_dump(adapter);
+
 	pci_set_drvdata(pdev, adapter);
 
-	if (adapter->flags & FW_OK) {
+	if (adapter->flags & CXGB4_FW_OK) {
 		err = t4_port_init(adapter, func, func, 0);
 		if (err)
 			goto out_free_dev;
@@ -4868,10 +6863,21 @@ static int init_one(struct pci_dev *pdev, const struct pci_device_id *ent)
 		}
 	}
 
+	if (!(adapter->flags & CXGB4_FW_OK))
+		goto fw_attach_fail;
+
 	/* Configure queues and allocate tables now, they can be needed as
 	 * soon as the first register_netdev completes.
 	 */
-	cfg_queues(adapter);
+	err = cfg_queues(adapter);
+	if (err)
+		goto out_free_dev;
+
+	adapter->smt = t4_init_smt();
+	if (!adapter->smt) {
+		/* We tolerate a lack of SMT, giving up some functionality */
+		dev_warn(&pdev->dev, "could not allocate SMT, continuing\n");
+	}
 
 	adapter->l2t = t4_init_l2t(adapter->l2t_start, adapter->l2t_end);
 	if (!adapter->l2t) {
@@ -4881,7 +6887,7 @@ static int init_one(struct pci_dev *pdev, const struct pci_device_id *ent)
 	}
 
 #if IS_ENABLED(CONFIG_IPV6)
-	if ((CHELSIO_CHIP_VERSION(adapter->params.chip) <= CHELSIO_T5) &&
+	if (chip_ver <= CHELSIO_T5 &&
 	    (!(t4_read_reg(adapter, LE_DB_CONFIG_A) & ASLIPCOMPEN_F))) {
 		/* CLIP functionality is not present in hardware,
 		 * hence disable all offload features
@@ -4912,6 +6918,24 @@ static int init_one(struct pci_dev *pdev, const struct pci_device_id *ent)
 				 i);
 	}
 
+	if (is_offload(adapter) || is_hashfilter(adapter)) {
+		if (t4_read_reg(adapter, LE_DB_CONFIG_A) & HASHEN_F) {
+			u32 v;
+
+			v = t4_read_reg(adapter, LE_DB_HASH_CONFIG_A);
+			if (chip_ver <= CHELSIO_T5) {
+				adapter->tids.nhash = 1 << HASHTIDSIZE_G(v);
+				v = t4_read_reg(adapter, LE_DB_TID_HASHBASE_A);
+				adapter->tids.hash_base = v / 4;
+			} else {
+				adapter->tids.nhash = HASHTBLSIZE_G(v) << 3;
+				v = t4_read_reg(adapter,
+						T6_LE_DB_HASH_TID_BASE_A);
+				adapter->tids.hash_base = v;
+			}
+		}
+	}
+
 	if (tid_init(&adapter->tids) < 0) {
 		dev_warn(&pdev->dev, "could not allocate TID table, "
 			 "continuing\n");
@@ -4921,40 +6945,56 @@ static int init_one(struct pci_dev *pdev, const struct pci_device_id *ent)
 		if (!adapter->tc_u32)
 			dev_warn(&pdev->dev,
 				 "could not offload tc u32, continuing\n");
-	}
 
-	if (is_offload(adapter)) {
-		if (t4_read_reg(adapter, LE_DB_CONFIG_A) & HASHEN_F) {
-			u32 hash_base, hash_reg;
+		if (cxgb4_init_tc_flower(adapter))
+			dev_warn(&pdev->dev,
+				 "could not offload tc flower, continuing\n");
 
-			if (chip <= CHELSIO_T5) {
-				hash_reg = LE_DB_TID_HASHBASE_A;
-				hash_base = t4_read_reg(adapter, hash_reg);
-				adapter->tids.hash_base = hash_base / 4;
-			} else {
-				hash_reg = T6_LE_DB_HASH_TID_BASE_A;
-				hash_base = t4_read_reg(adapter, hash_reg);
-				adapter->tids.hash_base = hash_base;
-			}
-		}
+		if (cxgb4_init_tc_mqprio(adapter))
+			dev_warn(&pdev->dev,
+				 "could not offload tc mqprio, continuing\n");
+
+		if (cxgb4_init_tc_matchall(adapter))
+			dev_warn(&pdev->dev,
+				 "could not offload tc matchall, continuing\n");
+		if (cxgb4_init_ethtool_filters(adapter))
+			dev_warn(&pdev->dev,
+				 "could not initialize ethtool filters, continuing\n");
 	}
 
 	/* See what interrupts we'll be using */
 	if (msi > 1 && enable_msix(adapter) == 0)
-		adapter->flags |= USING_MSIX;
+		adapter->flags |= CXGB4_USING_MSIX;
 	else if (msi > 0 && pci_enable_msi(pdev) == 0) {
-		adapter->flags |= USING_MSI;
+		adapter->flags |= CXGB4_USING_MSI;
 		if (msi > 1)
 			free_msix_info(adapter);
 	}
 
 	/* check for PCI Express bandwidth capabiltites */
-	cxgb4_check_pcie_caps(adapter);
+	pcie_print_link_status(pdev);
+
+	cxgb4_init_mps_ref_entries(adapter);
 
 	err = init_rss(adapter);
 	if (err)
 		goto out_free_dev;
 
+	err = setup_non_data_intr(adapter);
+	if (err) {
+		dev_err(adapter->pdev_dev,
+			"Non Data interrupt allocation failed, err: %d\n", err);
+		goto out_free_dev;
+	}
+
+	err = setup_fw_sge_queues(adapter);
+	if (err) {
+		dev_err(adapter->pdev_dev,
+			"FW sge queue allocation failed, err %d", err);
+		goto out_free_dev;
+	}
+
+fw_attach_fail:
 	/*
 	 * The card is now ready to go.  If any errors occur during device
 	 * registration we do not fail the whole card but rather proceed only
@@ -4993,69 +7033,23 @@ static int init_one(struct pci_dev *pdev, const struct pci_device_id *ent)
 	/* PCIe EEH recovery on powerpc platforms needs fundamental reset */
 	pdev->needs_freset = 1;
 
-	if (is_uld(adapter)) {
-		mutex_lock(&uld_mutex);
-		list_add_tail(&adapter->list_node, &adapter_list);
-		mutex_unlock(&uld_mutex);
-	}
-
-	print_adapter_info(adapter);
-	setup_fw_sge_queues(adapter);
-	return 0;
-
-sriov:
-#ifdef CONFIG_PCI_IOV
-	adapter = kzalloc(sizeof(*adapter), GFP_KERNEL);
-	if (!adapter) {
-		err = -ENOMEM;
-		goto free_pci_region;
-	}
+	if (is_uld(adapter))
+		cxgb4_uld_enable(adapter);
 
-	adapter->pdev = pdev;
-	adapter->pdev_dev = &pdev->dev;
-	adapter->name = pci_name(pdev);
-	adapter->mbox = func;
-	adapter->pf = func;
-	adapter->regs = regs;
-	adapter->adap_idx = adap_idx;
-	adapter->mbox_log = kzalloc(sizeof(*adapter->mbox_log) +
-				    (sizeof(struct mbox_cmd) *
-				     T4_OS_LOG_MBOX_CMDS),
-				    GFP_KERNEL);
-	if (!adapter->mbox_log) {
-		err = -ENOMEM;
-		goto free_adapter;
-	}
-	spin_lock_init(&adapter->mbox_lock);
-	INIT_LIST_HEAD(&adapter->mlist.list);
-
-	v = FW_PARAMS_MNEM_V(FW_PARAMS_MNEM_DEV) |
-	    FW_PARAMS_PARAM_X_V(FW_PARAMS_PARAM_DEV_PORTVEC);
-	err = t4_query_params(adapter, adapter->mbox, adapter->pf, 0, 1,
-			      &v, &port_vec);
-	if (err < 0) {
-		dev_err(adapter->pdev_dev, "Could not fetch port params\n");
-		goto free_adapter;
-	}
+	if (!is_t4(adapter->params.chip))
+		cxgb4_ptp_init(adapter);
 
-	adapter->params.nports = hweight32(port_vec);
-	pci_set_drvdata(pdev, adapter);
-	return 0;
+	if (IS_REACHABLE(CONFIG_THERMAL) &&
+	    !is_t4(adapter->params.chip) && (adapter->flags & CXGB4_FW_OK))
+		cxgb4_thermal_init(adapter);
 
- free_adapter:
-	kfree(adapter);
- free_pci_region:
-	iounmap(regs);
-	pci_disable_sriov(pdev);
-	pci_release_regions(pdev);
-	return err;
-#else
+	print_adapter_info(adapter);
 	return 0;
-#endif
 
  out_free_dev:
+	t4_free_sge_resources(adapter);
 	free_some_resources(adapter);
-	if (adapter->flags & USING_MSIX)
+	if (adapter->flags & CXGB4_USING_MSIX)
 		free_msix_info(adapter);
 	if (adapter->num_uld || adapter->num_ofld_uld)
 		t4_uld_mem_free(adapter);
@@ -5071,7 +7065,6 @@ sriov:
  out_unmap_bar0:
 	iounmap(regs);
  out_disable_device:
-	pci_disable_pcie_error_reporting(pdev);
 	pci_disable_device(pdev);
  out_release_regions:
 	pci_release_regions(pdev);
@@ -5081,12 +7074,20 @@ sriov:
 static void remove_one(struct pci_dev *pdev)
 {
 	struct adapter *adapter = pci_get_drvdata(pdev);
+	struct hash_mac_addr *entry, *tmp;
 
 	if (!adapter) {
 		pci_release_regions(pdev);
 		return;
 	}
 
+	/* If we allocated filters, free up state associated with any
+	 * valid filters ...
+	 */
+	clear_all_filters(adapter);
+
+	adapter->flags |= CXGB4_SHUTTING_DOWN;
+
 	if (adapter->pf == 4) {
 		int i;
 
@@ -5095,57 +7096,61 @@ static void remove_one(struct pci_dev *pdev)
 		 */
 		destroy_workqueue(adapter->workq);
 
-		if (is_uld(adapter))
-			detach_ulds(adapter);
-
-		disable_interrupts(adapter);
+		detach_ulds(adapter);
 
 		for_each_port(adapter, i)
 			if (adapter->port[i]->reg_state == NETREG_REGISTERED)
 				unregister_netdev(adapter->port[i]);
 
+		t4_uld_clean_up(adapter);
+
+		adap_free_hma_mem(adapter);
+
+		disable_interrupts(adapter);
+
+		cxgb4_free_mps_ref_entries(adapter);
+
 		debugfs_remove_recursive(adapter->debugfs_root);
 
-		/* If we allocated filters, free up state associated with any
-		 * valid filters ...
-		 */
-		clear_all_filters(adapter);
+		if (!is_t4(adapter->params.chip))
+			cxgb4_ptp_stop(adapter);
+		if (IS_REACHABLE(CONFIG_THERMAL))
+			cxgb4_thermal_remove(adapter);
 
-		if (adapter->flags & FULL_INIT_DONE)
+		if (adapter->flags & CXGB4_FULL_INIT_DONE)
 			cxgb_down(adapter);
 
-		if (adapter->flags & USING_MSIX)
+		if (adapter->flags & CXGB4_USING_MSIX)
 			free_msix_info(adapter);
 		if (adapter->num_uld || adapter->num_ofld_uld)
 			t4_uld_mem_free(adapter);
 		free_some_resources(adapter);
+		list_for_each_entry_safe(entry, tmp, &adapter->mac_hlist,
+					 list) {
+			list_del(&entry->list);
+			kfree(entry);
+		}
+
 #if IS_ENABLED(CONFIG_IPV6)
 		t4_cleanup_clip_tbl(adapter);
 #endif
-		iounmap(adapter->regs);
 		if (!is_t4(adapter->params.chip))
 			iounmap(adapter->bar2);
-		pci_disable_pcie_error_reporting(pdev);
-		if ((adapter->flags & DEV_ENABLED)) {
-			pci_disable_device(pdev);
-			adapter->flags &= ~DEV_ENABLED;
-		}
-		pci_release_regions(pdev);
-		kfree(adapter->mbox_log);
-		synchronize_rcu();
-		kfree(adapter);
 	}
 #ifdef CONFIG_PCI_IOV
 	else {
-		if (adapter->port[0])
-			unregister_netdev(adapter->port[0]);
-		iounmap(adapter->regs);
-		kfree(adapter->vfinfo);
-		kfree(adapter);
-		pci_disable_sriov(pdev);
-		pci_release_regions(pdev);
+		cxgb4_iov_configure(adapter->pdev, 0);
 	}
 #endif
+	iounmap(adapter->regs);
+	if ((adapter->flags & CXGB4_DEV_ENABLED)) {
+		pci_disable_device(pdev);
+		adapter->flags &= ~CXGB4_DEV_ENABLED;
+	}
+	pci_release_regions(pdev);
+	kfree(adapter->mbox_log);
+	synchronize_rcu();
+	kfree(adapter);
 }
 
 /* "Shutdown" quiesces the device, stopping Ingress Packet and Interrupt
@@ -5166,6 +7171,8 @@ static void shutdown_one(struct pci_dev *pdev)
 		return;
 	}
 
+	adapter->flags |= CXGB4_SHUTTING_DOWN;
+
 	if (adapter->pf == 4) {
 		int i;
 
@@ -5173,25 +7180,22 @@ static void shutdown_one(struct pci_dev *pdev)
 			if (adapter->port[i]->reg_state == NETREG_REGISTERED)
 				cxgb_close(adapter->port[i]);
 
-		t4_uld_clean_up(adapter);
+		rtnl_lock();
+		cxgb4_mqprio_stop_offload(adapter);
+		rtnl_unlock();
+
+		if (is_uld(adapter)) {
+			detach_ulds(adapter);
+			t4_uld_clean_up(adapter);
+		}
+
 		disable_interrupts(adapter);
 		disable_msi(adapter);
 
 		t4_sge_stop(adapter);
-		if (adapter->flags & FW_OK)
+		if (adapter->flags & CXGB4_FW_OK)
 			t4_fw_bye(adapter, adapter->mbox);
 	}
-#ifdef CONFIG_PCI_IOV
-	else {
-		if (adapter->port[0])
-			unregister_netdev(adapter->port[0]);
-		iounmap(adapter->regs);
-		kfree(adapter->vfinfo);
-		kfree(adapter);
-		pci_disable_sriov(pdev);
-		pci_release_regions(pdev);
-	}
-#endif
 }
 
 static struct pci_driver cxgb4_driver = {
@@ -5210,22 +7214,28 @@ static int __init cxgb4_init_module(void)
 {
 	int ret;
 
-	/* Debugfs support is optional, just warn if this fails */
 	cxgb4_debugfs_root = debugfs_create_dir(KBUILD_MODNAME, NULL);
-	if (!cxgb4_debugfs_root)
-		pr_warn("could not create debugfs entry, continuing\n");
 
 	ret = pci_register_driver(&cxgb4_driver);
 	if (ret < 0)
-		debugfs_remove(cxgb4_debugfs_root);
+		goto err_pci;
 
 #if IS_ENABLED(CONFIG_IPV6)
 	if (!inet6addr_registered) {
-		register_inet6addr_notifier(&cxgb4_inet6addr_notifier);
-		inet6addr_registered = true;
+		ret = register_inet6addr_notifier(&cxgb4_inet6addr_notifier);
+		if (ret)
+			pci_unregister_driver(&cxgb4_driver);
+		else
+			inet6addr_registered = true;
 	}
 #endif
 
+	if (ret == 0)
+		return ret;
+
+err_pci:
+	debugfs_remove(cxgb4_debugfs_root);
+
 	return ret;
 }
 
diff --git a/drivers/net/ethernet/chelsio/cxgb4/cxgb4_mps.c b/drivers/net/ethernet/chelsio/cxgb4/cxgb4_mps.c
new file mode 100644
index 000000000000..60f4d5b5eb3a
--- /dev/null
+++ b/drivers/net/ethernet/chelsio/cxgb4/cxgb4_mps.c
@@ -0,0 +1,143 @@
+// SPDX-License-Identifier: GPL-2.0
+/* Copyright (c) 2019 Chelsio Communications, Inc. All rights reserved. */
+
+#include "cxgb4.h"
+
+static int cxgb4_mps_ref_dec_by_mac(struct adapter *adap,
+				    const u8 *addr, const u8 *mask)
+{
+	u8 bitmask[] = { 0xff, 0xff, 0xff, 0xff, 0xff, 0xff };
+	struct mps_entries_ref *mps_entry, *tmp;
+	int ret = -EINVAL;
+
+	spin_lock_bh(&adap->mps_ref_lock);
+	list_for_each_entry_safe(mps_entry, tmp, &adap->mps_ref, list) {
+		if (ether_addr_equal(mps_entry->addr, addr) &&
+		    ether_addr_equal(mps_entry->mask, mask ? mask : bitmask)) {
+			if (!refcount_dec_and_test(&mps_entry->refcnt)) {
+				spin_unlock_bh(&adap->mps_ref_lock);
+				return -EBUSY;
+			}
+			list_del(&mps_entry->list);
+			kfree(mps_entry);
+			ret = 0;
+			break;
+		}
+	}
+	spin_unlock_bh(&adap->mps_ref_lock);
+	return ret;
+}
+
+static int cxgb4_mps_ref_inc(struct adapter *adap, const u8 *mac_addr,
+			     u16 idx, const u8 *mask)
+{
+	u8 bitmask[] = { 0xff, 0xff, 0xff, 0xff, 0xff, 0xff };
+	struct mps_entries_ref *mps_entry;
+	int ret = 0;
+
+	spin_lock_bh(&adap->mps_ref_lock);
+	list_for_each_entry(mps_entry, &adap->mps_ref, list) {
+		if (mps_entry->idx == idx) {
+			refcount_inc(&mps_entry->refcnt);
+			goto unlock;
+		}
+	}
+	mps_entry = kzalloc(sizeof(*mps_entry), GFP_ATOMIC);
+	if (!mps_entry) {
+		ret = -ENOMEM;
+		goto unlock;
+	}
+	ether_addr_copy(mps_entry->mask, mask ? mask : bitmask);
+	ether_addr_copy(mps_entry->addr, mac_addr);
+	mps_entry->idx = idx;
+	refcount_set(&mps_entry->refcnt, 1);
+	list_add_tail(&mps_entry->list, &adap->mps_ref);
+unlock:
+	spin_unlock_bh(&adap->mps_ref_lock);
+	return ret;
+}
+
+int cxgb4_free_mac_filt(struct adapter *adap, unsigned int viid,
+			unsigned int naddr, const u8 **addr, bool sleep_ok)
+{
+	int ret, i;
+
+	for (i = 0; i < naddr; i++) {
+		if (!cxgb4_mps_ref_dec_by_mac(adap, addr[i], NULL)) {
+			ret = t4_free_mac_filt(adap, adap->mbox, viid,
+					       1, &addr[i], sleep_ok);
+			if (ret < 0)
+				return ret;
+		}
+	}
+
+	/* return number of filters freed */
+	return naddr;
+}
+
+int cxgb4_alloc_mac_filt(struct adapter *adap, unsigned int viid,
+			 bool free, unsigned int naddr, const u8 **addr,
+			 u16 *idx, u64 *hash, bool sleep_ok)
+{
+	int ret, i;
+
+	ret = t4_alloc_mac_filt(adap, adap->mbox, viid, free,
+				naddr, addr, idx, hash, sleep_ok);
+	if (ret < 0)
+		return ret;
+
+	for (i = 0; i < naddr; i++) {
+		if (idx[i] != 0xffff) {
+			if (cxgb4_mps_ref_inc(adap, addr[i], idx[i], NULL)) {
+				ret = -ENOMEM;
+				goto error;
+			}
+		}
+	}
+
+	goto out;
+error:
+	cxgb4_free_mac_filt(adap, viid, naddr, addr, sleep_ok);
+
+out:
+	/* Returns a negative error number or the number of filters allocated */
+	return ret;
+}
+
+int cxgb4_update_mac_filt(struct port_info *pi, unsigned int viid,
+			  int *tcam_idx, const u8 *addr,
+			  bool persistent, u8 *smt_idx)
+{
+	int ret;
+
+	ret = cxgb4_change_mac(pi, viid, tcam_idx,
+			       addr, persistent, smt_idx);
+	if (ret < 0)
+		return ret;
+
+	cxgb4_mps_ref_inc(pi->adapter, addr, *tcam_idx, NULL);
+	return ret;
+}
+
+int cxgb4_init_mps_ref_entries(struct adapter *adap)
+{
+	spin_lock_init(&adap->mps_ref_lock);
+	INIT_LIST_HEAD(&adap->mps_ref);
+
+	return 0;
+}
+
+void cxgb4_free_mps_ref_entries(struct adapter *adap)
+{
+	struct mps_entries_ref *mps_entry, *tmp;
+
+	if (list_empty(&adap->mps_ref))
+		return;
+
+	spin_lock(&adap->mps_ref_lock);
+	list_for_each_entry_safe(mps_entry, tmp, &adap->mps_ref, list) {
+		list_del(&mps_entry->list);
+		kfree(mps_entry);
+	}
+	spin_unlock(&adap->mps_ref_lock);
+}
diff --git a/drivers/net/ethernet/chelsio/cxgb4/cxgb4_ptp.c b/drivers/net/ethernet/chelsio/cxgb4/cxgb4_ptp.c
new file mode 100644
index 000000000000..cbd06d9b95d4
--- /dev/null
+++ b/drivers/net/ethernet/chelsio/cxgb4/cxgb4_ptp.c
@@ -0,0 +1,468 @@
+/*
+ * cxgb4_ptp.c:Chelsio PTP support for T5/T6
+ *
+ * Copyright (c) 2003-2017 Chelsio Communications, Inc. All rights reserved.
+ *
+ * This software is available to you under a choice of one of two
+ * licenses.  You may choose to be licensed under the terms of the GNU
+ * General Public License (GPL) Version 2, available from the file
+ * COPYING in the main directory of this source tree, or the
+ * OpenIB.org BSD license below:
+ *
+ *     Redistribution and use in source and binary forms, with or
+ *     without modification, are permitted provided that the following
+ *     conditions are met:
+ *
+ *      - Redistributions of source code must retain the above
+ *        copyright notice, this list of conditions and the following
+ *        disclaimer.
+ *
+ *      - Redistributions in binary form must reproduce the above
+ *        copyright notice, this list of conditions and the following
+ *        disclaimer in the documentation and/or other materials
+ *        provided with the distribution.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND,
+ * EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF
+ * MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND
+ * NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS
+ * BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN
+ * ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN
+ * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
+ * SOFTWARE.
+ *
+ * Written by: Atul Gupta (atul.gupta@chelsio.com)
+ */
+
+#include <linux/module.h>
+#include <linux/net_tstamp.h>
+#include <linux/skbuff.h>
+#include <linux/netdevice.h>
+#include <linux/pps_kernel.h>
+#include <linux/ptp_clock_kernel.h>
+#include <linux/ptp_classify.h>
+#include <linux/udp.h>
+
+#include "cxgb4.h"
+#include "t4_hw.h"
+#include "t4_regs.h"
+#include "t4_msg.h"
+#include "t4fw_api.h"
+#include "cxgb4_ptp.h"
+
+/**
+ * cxgb4_ptp_is_ptp_tx - determine whether TX packet is PTP or not
+ * @skb: skb of outgoing ptp request
+ *
+ */
+bool cxgb4_ptp_is_ptp_tx(struct sk_buff *skb)
+{
+	struct udphdr *uh;
+
+	uh = udp_hdr(skb);
+	return skb->len >= PTP_MIN_LENGTH &&
+		skb->len <= PTP_IN_TRANSMIT_PACKET_MAXNUM &&
+		likely(skb->protocol == htons(ETH_P_IP)) &&
+		ip_hdr(skb)->protocol == IPPROTO_UDP &&
+		uh->dest == htons(PTP_EVENT_PORT);
+}
+
+bool is_ptp_enabled(struct sk_buff *skb, struct net_device *dev)
+{
+	struct port_info *pi;
+
+	pi = netdev_priv(dev);
+	return (pi->ptp_enable && cxgb4_xmit_with_hwtstamp(skb) &&
+		cxgb4_ptp_is_ptp_tx(skb));
+}
+
+/**
+ * cxgb4_ptp_is_ptp_rx - determine whether RX packet is PTP or not
+ * @skb: skb of incoming ptp request
+ *
+ */
+bool cxgb4_ptp_is_ptp_rx(struct sk_buff *skb)
+{
+	struct udphdr *uh = (struct udphdr *)(skb->data + ETH_HLEN +
+					      IPV4_HLEN(skb->data));
+
+	return  uh->dest == htons(PTP_EVENT_PORT) &&
+		uh->source == htons(PTP_EVENT_PORT);
+}
+
+/**
+ * cxgb4_ptp_read_hwstamp - read timestamp for TX event PTP message
+ * @adapter: board private structure
+ * @pi: port private structure
+ *
+ */
+void cxgb4_ptp_read_hwstamp(struct adapter *adapter, struct port_info *pi)
+{
+	struct skb_shared_hwtstamps *skb_ts = NULL;
+	u64 tx_ts;
+
+	skb_ts = skb_hwtstamps(adapter->ptp_tx_skb);
+
+	tx_ts = t4_read_reg(adapter,
+			    T5_PORT_REG(pi->port_id, MAC_PORT_TX_TS_VAL_LO));
+
+	tx_ts |= (u64)t4_read_reg(adapter,
+				  T5_PORT_REG(pi->port_id,
+					      MAC_PORT_TX_TS_VAL_HI)) << 32;
+	skb_ts->hwtstamp = ns_to_ktime(tx_ts);
+	skb_tstamp_tx(adapter->ptp_tx_skb, skb_ts);
+	dev_kfree_skb_any(adapter->ptp_tx_skb);
+	spin_lock(&adapter->ptp_lock);
+	adapter->ptp_tx_skb = NULL;
+	spin_unlock(&adapter->ptp_lock);
+}
+
+/**
+ * cxgb4_ptprx_timestamping - Enable Timestamp for RX PTP event message
+ * @pi: port private structure
+ * @port: pot number
+ * @mode: RX mode
+ *
+ */
+int cxgb4_ptprx_timestamping(struct port_info *pi, u8 port, u16 mode)
+{
+	struct adapter *adapter = pi->adapter;
+	struct fw_ptp_cmd c;
+	int err;
+
+	memset(&c, 0, sizeof(c));
+	c.op_to_portid = cpu_to_be32(FW_CMD_OP_V(FW_PTP_CMD) |
+				     FW_CMD_REQUEST_F |
+				     FW_CMD_WRITE_F |
+				     FW_PTP_CMD_PORTID_V(port));
+	c.retval_len16 = cpu_to_be32(FW_CMD_LEN16_V(sizeof(c) / 16));
+	c.u.init.sc = FW_PTP_SC_RXTIME_STAMP;
+	c.u.init.mode = cpu_to_be16(mode);
+
+	err = t4_wr_mbox(adapter, adapter->mbox, &c, sizeof(c), NULL);
+	if (err < 0)
+		dev_err(adapter->pdev_dev,
+			"PTP: %s error %d\n", __func__, -err);
+	return err;
+}
+
+int cxgb4_ptp_txtype(struct adapter *adapter, u8 port)
+{
+	struct fw_ptp_cmd c;
+	int err;
+
+	memset(&c, 0, sizeof(c));
+	c.op_to_portid = cpu_to_be32(FW_CMD_OP_V(FW_PTP_CMD) |
+				     FW_CMD_REQUEST_F |
+				     FW_CMD_WRITE_F |
+				     FW_PTP_CMD_PORTID_V(port));
+	c.retval_len16 = cpu_to_be32(FW_CMD_LEN16_V(sizeof(c) / 16));
+	c.u.init.sc = FW_PTP_SC_TX_TYPE;
+	c.u.init.mode = cpu_to_be16(PTP_TS_NONE);
+
+	err = t4_wr_mbox(adapter, adapter->mbox, &c, sizeof(c), NULL);
+	if (err < 0)
+		dev_err(adapter->pdev_dev,
+			"PTP: %s error %d\n", __func__, -err);
+
+	return err;
+}
+
+int cxgb4_ptp_redirect_rx_packet(struct adapter *adapter, struct port_info *pi)
+{
+	struct sge *s = &adapter->sge;
+	struct sge_eth_rxq *receive_q =  &s->ethrxq[pi->first_qset];
+	struct fw_ptp_cmd c;
+	int err;
+
+	memset(&c, 0, sizeof(c));
+	c.op_to_portid = cpu_to_be32(FW_CMD_OP_V(FW_PTP_CMD) |
+				     FW_CMD_REQUEST_F |
+				     FW_CMD_WRITE_F |
+				     FW_PTP_CMD_PORTID_V(pi->port_id));
+
+	c.retval_len16 = cpu_to_be32(FW_CMD_LEN16_V(sizeof(c) / 16));
+	c.u.init.sc = FW_PTP_SC_RDRX_TYPE;
+	c.u.init.txchan = pi->tx_chan;
+	c.u.init.absid = cpu_to_be16(receive_q->rspq.abs_id);
+
+	err = t4_wr_mbox(adapter, adapter->mbox, &c, sizeof(c), NULL);
+	if (err < 0)
+		dev_err(adapter->pdev_dev,
+			"PTP: %s error %d\n", __func__, -err);
+	return err;
+}
+
+/**
+ * cxgb4_ptp_adjfine - Adjust frequency of PHC cycle counter
+ * @ptp: ptp clock structure
+ * @scaled_ppm: Desired frequency in scaled parts per billion
+ *
+ * Adjust the frequency of the PHC cycle counter by the indicated amount from
+ * the base frequency.
+ *
+ * Scaled parts per million is ppm with a 16-bit binary fractional field.
+ */
+static int cxgb4_ptp_adjfine(struct ptp_clock_info *ptp, long scaled_ppm)
+{
+	struct adapter *adapter = (struct adapter *)container_of(ptp,
+				   struct adapter, ptp_clock_info);
+	s32 ppb = scaled_ppm_to_ppb(scaled_ppm);
+	struct fw_ptp_cmd c;
+	int err;
+
+	memset(&c, 0, sizeof(c));
+	c.op_to_portid = cpu_to_be32(FW_CMD_OP_V(FW_PTP_CMD) |
+				     FW_CMD_REQUEST_F |
+				     FW_CMD_WRITE_F |
+				     FW_PTP_CMD_PORTID_V(0));
+	c.retval_len16 = cpu_to_be32(FW_CMD_LEN16_V(sizeof(c) / 16));
+	c.u.ts.sc = FW_PTP_SC_ADJ_FREQ;
+	c.u.ts.sign = (ppb < 0) ? 1 : 0;
+	if (ppb < 0)
+		ppb = -ppb;
+	c.u.ts.ppb = cpu_to_be32(ppb);
+
+	err = t4_wr_mbox(adapter, adapter->mbox, &c, sizeof(c), NULL);
+	if (err < 0)
+		dev_err(adapter->pdev_dev,
+			"PTP: %s error %d\n", __func__, -err);
+
+	return err;
+}
+
+/**
+ * cxgb4_ptp_fineadjtime - Shift the time of the hardware clock
+ * @adapter: board private structure
+ * @delta: Desired change in nanoseconds
+ *
+ * Adjust the timer by resetting the timecounter structure.
+ */
+static int  cxgb4_ptp_fineadjtime(struct adapter *adapter, s64 delta)
+{
+	struct fw_ptp_cmd c;
+	int err;
+
+	memset(&c, 0, sizeof(c));
+	c.op_to_portid = cpu_to_be32(FW_CMD_OP_V(FW_PTP_CMD) |
+			     FW_CMD_REQUEST_F |
+			     FW_CMD_WRITE_F |
+			     FW_PTP_CMD_PORTID_V(0));
+	c.retval_len16 = cpu_to_be32(FW_CMD_LEN16_V(sizeof(c) / 16));
+	c.u.ts.sc = FW_PTP_SC_ADJ_FTIME;
+	c.u.ts.sign = (delta < 0) ? 1 : 0;
+	if (delta < 0)
+		delta = -delta;
+	c.u.ts.tm = cpu_to_be64(delta);
+
+	err = t4_wr_mbox(adapter, adapter->mbox, &c, sizeof(c), NULL);
+	if (err < 0)
+		dev_err(adapter->pdev_dev,
+			"PTP: %s error %d\n", __func__, -err);
+	return err;
+}
+
+/**
+ * cxgb4_ptp_adjtime - Shift the time of the hardware clock
+ * @ptp: ptp clock structure
+ * @delta: Desired change in nanoseconds
+ *
+ * Adjust the timer by resetting the timecounter structure.
+ */
+static int cxgb4_ptp_adjtime(struct ptp_clock_info *ptp, s64 delta)
+{
+	struct adapter *adapter =
+		(struct adapter *)container_of(ptp, struct adapter,
+					       ptp_clock_info);
+	struct fw_ptp_cmd c;
+	s64 sign = 1;
+	int err;
+
+	if (delta < 0)
+		sign = -1;
+
+	if (delta * sign > PTP_CLOCK_MAX_ADJTIME) {
+		memset(&c, 0, sizeof(c));
+		c.op_to_portid = cpu_to_be32(FW_CMD_OP_V(FW_PTP_CMD) |
+					     FW_CMD_REQUEST_F |
+					     FW_CMD_WRITE_F |
+					     FW_PTP_CMD_PORTID_V(0));
+		c.retval_len16 = cpu_to_be32(FW_CMD_LEN16_V(sizeof(c) / 16));
+		c.u.ts.sc = FW_PTP_SC_ADJ_TIME;
+		c.u.ts.sign = (delta < 0) ? 1 : 0;
+		if (delta < 0)
+			delta = -delta;
+		c.u.ts.tm = cpu_to_be64(delta);
+
+		err = t4_wr_mbox(adapter, adapter->mbox, &c, sizeof(c), NULL);
+		if (err < 0)
+			dev_err(adapter->pdev_dev,
+				"PTP: %s error %d\n", __func__, -err);
+	} else {
+		err = cxgb4_ptp_fineadjtime(adapter, delta);
+	}
+
+	return err;
+}
+
+/**
+ * cxgb4_ptp_gettime - Reads the current time from the hardware clock
+ * @ptp: ptp clock structure
+ * @ts: timespec structure to hold the current time value
+ *
+ * Read the timecounter and return the correct value in ns after converting
+ * it into a struct timespec.
+ */
+static int cxgb4_ptp_gettime(struct ptp_clock_info *ptp, struct timespec64 *ts)
+{
+	struct adapter *adapter = container_of(ptp, struct adapter,
+					       ptp_clock_info);
+	u64 ns;
+
+	ns = t4_read_reg(adapter, T5_PORT_REG(0, MAC_PORT_PTP_SUM_LO_A));
+	ns |= (u64)t4_read_reg(adapter,
+			       T5_PORT_REG(0, MAC_PORT_PTP_SUM_HI_A)) << 32;
+
+	/* convert to timespec*/
+	*ts = ns_to_timespec64(ns);
+	return 0;
+}
+
+/**
+ *  cxgb4_ptp_settime - Set the current time on the hardware clock
+ *  @ptp: ptp clock structure
+ *  @ts: timespec containing the new time for the cycle counter
+ *
+ *  Reset value to new base value instead of the kernel
+ *  wall timer value.
+ */
+static int cxgb4_ptp_settime(struct ptp_clock_info *ptp,
+			     const struct timespec64 *ts)
+{
+	struct adapter *adapter = (struct adapter *)container_of(ptp,
+				   struct adapter, ptp_clock_info);
+	struct fw_ptp_cmd c;
+	u64 ns;
+	int err;
+
+	memset(&c, 0, sizeof(c));
+	c.op_to_portid = cpu_to_be32(FW_CMD_OP_V(FW_PTP_CMD) |
+				     FW_CMD_REQUEST_F |
+				     FW_CMD_WRITE_F |
+				     FW_PTP_CMD_PORTID_V(0));
+	c.retval_len16 = cpu_to_be32(FW_CMD_LEN16_V(sizeof(c) / 16));
+	c.u.ts.sc = FW_PTP_SC_SET_TIME;
+
+	ns = timespec64_to_ns(ts);
+	c.u.ts.tm = cpu_to_be64(ns);
+
+	err =  t4_wr_mbox(adapter, adapter->mbox, &c, sizeof(c), NULL);
+	if (err < 0)
+		dev_err(adapter->pdev_dev,
+			"PTP: %s error %d\n", __func__, -err);
+
+	return err;
+}
+
+static void cxgb4_init_ptp_timer(struct adapter *adapter)
+{
+	struct fw_ptp_cmd c;
+	int err;
+
+	memset(&c, 0, sizeof(c));
+	c.op_to_portid = cpu_to_be32(FW_CMD_OP_V(FW_PTP_CMD) |
+				     FW_CMD_REQUEST_F |
+				     FW_CMD_WRITE_F |
+				     FW_PTP_CMD_PORTID_V(0));
+	c.retval_len16 = cpu_to_be32(FW_CMD_LEN16_V(sizeof(c) / 16));
+	c.u.scmd.sc = FW_PTP_SC_INIT_TIMER;
+
+	err = t4_wr_mbox(adapter, adapter->mbox, &c, sizeof(c), NULL);
+	if (err < 0)
+		dev_err(adapter->pdev_dev,
+			"PTP: %s error %d\n", __func__, -err);
+}
+
+/**
+ * cxgb4_ptp_enable - enable or disable an ancillary feature
+ * @ptp: ptp clock structure
+ * @request: Desired resource to enable or disable
+ * @on: Caller passes one to enable or zero to disable
+ *
+ * Enable (or disable) ancillary features of the PHC subsystem.
+ * Currently, no ancillary features are supported.
+ */
+static int cxgb4_ptp_enable(struct ptp_clock_info __always_unused *ptp,
+			    struct ptp_clock_request __always_unused *request,
+			    int __always_unused on)
+{
+	return -ENOTSUPP;
+}
+
+static const struct ptp_clock_info cxgb4_ptp_clock_info = {
+	.owner          = THIS_MODULE,
+	.name           = "cxgb4_clock",
+	.max_adj        = MAX_PTP_FREQ_ADJ,
+	.n_alarm        = 0,
+	.n_ext_ts       = 0,
+	.n_per_out      = 0,
+	.pps            = 0,
+	.adjfine        = cxgb4_ptp_adjfine,
+	.adjtime        = cxgb4_ptp_adjtime,
+	.gettime64      = cxgb4_ptp_gettime,
+	.settime64      = cxgb4_ptp_settime,
+	.enable         = cxgb4_ptp_enable,
+};
+
+/**
+ * cxgb4_ptp_init - initialize PTP for devices which support it
+ * @adapter: board private structure
+ *
+ * This function performs the required steps for enabling PTP support.
+ */
+void cxgb4_ptp_init(struct adapter *adapter)
+{
+	struct timespec64 now;
+	 /* no need to create a clock device if we already have one */
+	if (!IS_ERR_OR_NULL(adapter->ptp_clock))
+		return;
+
+	adapter->ptp_tx_skb = NULL;
+	adapter->ptp_clock_info = cxgb4_ptp_clock_info;
+	spin_lock_init(&adapter->ptp_lock);
+
+	adapter->ptp_clock = ptp_clock_register(&adapter->ptp_clock_info,
+						&adapter->pdev->dev);
+	if (IS_ERR_OR_NULL(adapter->ptp_clock)) {
+		adapter->ptp_clock = NULL;
+		dev_err(adapter->pdev_dev,
+			"PTP %s Clock registration has failed\n", __func__);
+		return;
+	}
+
+	now = ktime_to_timespec64(ktime_get_real());
+	cxgb4_init_ptp_timer(adapter);
+	if (cxgb4_ptp_settime(&adapter->ptp_clock_info, &now) < 0) {
+		ptp_clock_unregister(adapter->ptp_clock);
+		adapter->ptp_clock = NULL;
+	}
+}
+
+/**
+ * cxgb4_ptp_stop - disable PTP device and stop the overflow check
+ * @adapter: board private structure
+ *
+ * Stop the PTP support.
+ */
+void cxgb4_ptp_stop(struct adapter *adapter)
+{
+	if (adapter->ptp_tx_skb) {
+		dev_kfree_skb_any(adapter->ptp_tx_skb);
+		adapter->ptp_tx_skb = NULL;
+	}
+
+	if (adapter->ptp_clock) {
+		ptp_clock_unregister(adapter->ptp_clock);
+		adapter->ptp_clock = NULL;
+	}
+}
diff --git a/drivers/net/ethernet/chelsio/cxgb4/cxgb4_ptp.h b/drivers/net/ethernet/chelsio/cxgb4/cxgb4_ptp.h
new file mode 100644
index 000000000000..cccfae84bb84
--- /dev/null
+++ b/drivers/net/ethernet/chelsio/cxgb4/cxgb4_ptp.h
@@ -0,0 +1,74 @@
+/*
+ * This file is part of the Chelsio T4 Ethernet driver for Linux.
+ *
+ * Copyright (c) 2003-2017 Chelsio Communications, Inc. All rights reserved.
+ *
+ * This software is available to you under a choice of one of two
+ * licenses.  You may choose to be licensed under the terms of the GNU
+ * General Public License (GPL) Version 2, available from the file
+ * COPYING in the main directory of this source tree, or the
+ * OpenIB.org BSD license below:
+ *
+ *     Redistribution and use in source and binary forms, with or
+ *     without modification, are permitted provided that the following
+ *     conditions are met:
+ *
+ *      - Redistributions of source code must retain the above
+ *        copyright notice, this list of conditions and the following
+ *        disclaimer.
+ *
+ *      - Redistributions in binary form must reproduce the above
+ *        copyright notice, this list of conditions and the following
+ *        disclaimer in the documentation and/or other materials
+ *        provided with the distribution.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND,
+ * EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF
+ * MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND
+ * NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS
+ * BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN
+ * ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN
+ * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
+ * SOFTWARE.
+ */
+
+#ifndef __CXGB4_PTP_H__
+#define __CXGB4_PTP_H__
+
+/* Maximum parts-per-billion adjustment that is acceptable */
+#define MAX_PTP_FREQ_ADJ		1000000
+#define PTP_CLOCK_MAX_ADJTIME		10000000 /* 10 ms */
+
+#define PTP_MIN_LENGTH			63
+#define PTP_IN_TRANSMIT_PACKET_MAXNUM	240
+#define PTP_EVENT_PORT			319
+
+enum ptp_rx_filter_mode {
+	PTP_TS_NONE = 0,
+	PTP_TS_L2,
+	PTP_TS_L4,
+	PTP_TS_L2_L4
+};
+
+struct port_info;
+
+static inline bool cxgb4_xmit_with_hwtstamp(struct sk_buff *skb)
+{
+	return skb_shinfo(skb)->tx_flags & SKBTX_HW_TSTAMP;
+}
+
+static inline void cxgb4_xmit_hwtstamp_pending(struct sk_buff *skb)
+{
+	skb_shinfo(skb)->tx_flags |= SKBTX_IN_PROGRESS;
+}
+
+void cxgb4_ptp_init(struct adapter *adap);
+void cxgb4_ptp_stop(struct adapter *adap);
+bool cxgb4_ptp_is_ptp_tx(struct sk_buff *skb);
+bool cxgb4_ptp_is_ptp_rx(struct sk_buff *skb);
+int cxgb4_ptprx_timestamping(struct port_info *pi, u8 port, u16 mode);
+int cxgb4_ptp_redirect_rx_packet(struct adapter *adap, struct port_info *pi);
+int cxgb4_ptp_txtype(struct adapter *adap, u8 port_id);
+void cxgb4_ptp_read_hwstamp(struct adapter *adap, struct port_info *pi);
+bool is_ptp_enabled(struct sk_buff *skb, struct net_device *dev);
+#endif /* __CXGB4_PTP_H__ */
diff --git a/drivers/net/ethernet/chelsio/cxgb4/cxgb4_tc_flower.c b/drivers/net/ethernet/chelsio/cxgb4/cxgb4_tc_flower.c
new file mode 100644
index 000000000000..0765d000eaef
--- /dev/null
+++ b/drivers/net/ethernet/chelsio/cxgb4/cxgb4_tc_flower.c
@@ -0,0 +1,1141 @@
+/*
+ * This file is part of the Chelsio T4/T5/T6 Ethernet driver for Linux.
+ *
+ * Copyright (c) 2017 Chelsio Communications, Inc. All rights reserved.
+ *
+ * This software is available to you under a choice of one of two
+ * licenses.  You may choose to be licensed under the terms of the GNU
+ * General Public License (GPL) Version 2, available from the file
+ * COPYING in the main directory of this source tree, or the
+ * OpenIB.org BSD license below:
+ *
+ *     Redistribution and use in source and binary forms, with or
+ *     without modification, are permitted provided that the following
+ *     conditions are met:
+ *
+ *      - Redistributions of source code must retain the above
+ *        copyright notice, this list of conditions and the following
+ *        disclaimer.
+ *
+ *      - Redistributions in binary form must reproduce the above
+ *        copyright notice, this list of conditions and the following
+ *        disclaimer in the documentation and/or other materials
+ *        provided with the distribution.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND,
+ * EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF
+ * MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND
+ * NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS
+ * BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN
+ * ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN
+ * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
+ * SOFTWARE.
+ */
+
+#include <net/tc_act/tc_mirred.h>
+#include <net/tc_act/tc_pedit.h>
+#include <net/tc_act/tc_gact.h>
+#include <net/tc_act/tc_vlan.h>
+
+#include "cxgb4.h"
+#include "cxgb4_filter.h"
+#include "cxgb4_tc_flower.h"
+
+#define STATS_CHECK_PERIOD (HZ / 2)
+
+static struct ch_tc_pedit_fields pedits[] = {
+	PEDIT_FIELDS(ETH_, DMAC_31_0, 4, dmac, 0),
+	PEDIT_FIELDS(ETH_, DMAC_47_32, 2, dmac, 4),
+	PEDIT_FIELDS(ETH_, SMAC_15_0, 2, smac, 0),
+	PEDIT_FIELDS(ETH_, SMAC_47_16, 4, smac, 2),
+	PEDIT_FIELDS(IP4_, SRC, 4, nat_fip, 0),
+	PEDIT_FIELDS(IP4_, DST, 4, nat_lip, 0),
+	PEDIT_FIELDS(IP6_, SRC_31_0, 4, nat_fip, 0),
+	PEDIT_FIELDS(IP6_, SRC_63_32, 4, nat_fip, 4),
+	PEDIT_FIELDS(IP6_, SRC_95_64, 4, nat_fip, 8),
+	PEDIT_FIELDS(IP6_, SRC_127_96, 4, nat_fip, 12),
+	PEDIT_FIELDS(IP6_, DST_31_0, 4, nat_lip, 0),
+	PEDIT_FIELDS(IP6_, DST_63_32, 4, nat_lip, 4),
+	PEDIT_FIELDS(IP6_, DST_95_64, 4, nat_lip, 8),
+	PEDIT_FIELDS(IP6_, DST_127_96, 4, nat_lip, 12),
+};
+
+static const struct cxgb4_natmode_config cxgb4_natmode_config_array[] = {
+	/* Default supported NAT modes */
+	{
+		.chip = CHELSIO_T5,
+		.flags = CXGB4_ACTION_NATMODE_NONE,
+		.natmode = NAT_MODE_NONE,
+	},
+	{
+		.chip = CHELSIO_T5,
+		.flags = CXGB4_ACTION_NATMODE_DIP,
+		.natmode = NAT_MODE_DIP,
+	},
+	{
+		.chip = CHELSIO_T5,
+		.flags = CXGB4_ACTION_NATMODE_DIP | CXGB4_ACTION_NATMODE_DPORT,
+		.natmode = NAT_MODE_DIP_DP,
+	},
+	{
+		.chip = CHELSIO_T5,
+		.flags = CXGB4_ACTION_NATMODE_DIP | CXGB4_ACTION_NATMODE_DPORT |
+			 CXGB4_ACTION_NATMODE_SIP,
+		.natmode = NAT_MODE_DIP_DP_SIP,
+	},
+	{
+		.chip = CHELSIO_T5,
+		.flags = CXGB4_ACTION_NATMODE_DIP | CXGB4_ACTION_NATMODE_DPORT |
+			 CXGB4_ACTION_NATMODE_SPORT,
+		.natmode = NAT_MODE_DIP_DP_SP,
+	},
+	{
+		.chip = CHELSIO_T5,
+		.flags = CXGB4_ACTION_NATMODE_SIP | CXGB4_ACTION_NATMODE_SPORT,
+		.natmode = NAT_MODE_SIP_SP,
+	},
+	{
+		.chip = CHELSIO_T5,
+		.flags = CXGB4_ACTION_NATMODE_DIP | CXGB4_ACTION_NATMODE_SIP |
+			 CXGB4_ACTION_NATMODE_SPORT,
+		.natmode = NAT_MODE_DIP_SIP_SP,
+	},
+	{
+		.chip = CHELSIO_T5,
+		.flags = CXGB4_ACTION_NATMODE_DIP | CXGB4_ACTION_NATMODE_SIP |
+			 CXGB4_ACTION_NATMODE_DPORT |
+			 CXGB4_ACTION_NATMODE_SPORT,
+		.natmode = NAT_MODE_ALL,
+	},
+	/* T6+ can ignore L4 ports when they're disabled. */
+	{
+		.chip = CHELSIO_T6,
+		.flags = CXGB4_ACTION_NATMODE_SIP,
+		.natmode = NAT_MODE_SIP_SP,
+	},
+	{
+		.chip = CHELSIO_T6,
+		.flags = CXGB4_ACTION_NATMODE_DIP | CXGB4_ACTION_NATMODE_SPORT,
+		.natmode = NAT_MODE_DIP_DP_SP,
+	},
+	{
+		.chip = CHELSIO_T6,
+		.flags = CXGB4_ACTION_NATMODE_DIP | CXGB4_ACTION_NATMODE_SIP,
+		.natmode = NAT_MODE_ALL,
+	},
+};
+
+static void cxgb4_action_natmode_tweak(struct ch_filter_specification *fs,
+				       u8 natmode_flags)
+{
+	u8 i = 0;
+
+	/* Translate the enabled NAT 4-tuple fields to one of the
+	 * hardware supported NAT mode configurations. This ensures
+	 * that we pick a valid combination, where the disabled fields
+	 * do not get overwritten to 0.
+	 */
+	for (i = 0; i < ARRAY_SIZE(cxgb4_natmode_config_array); i++) {
+		if (cxgb4_natmode_config_array[i].flags == natmode_flags) {
+			fs->nat_mode = cxgb4_natmode_config_array[i].natmode;
+			return;
+		}
+	}
+}
+
+static struct ch_tc_flower_entry *allocate_flower_entry(void)
+{
+	struct ch_tc_flower_entry *new = kzalloc(sizeof(*new), GFP_KERNEL);
+	if (new)
+		spin_lock_init(&new->lock);
+	return new;
+}
+
+/* Must be called with either RTNL or rcu_read_lock */
+static struct ch_tc_flower_entry *ch_flower_lookup(struct adapter *adap,
+						   unsigned long flower_cookie)
+{
+	return rhashtable_lookup_fast(&adap->flower_tbl, &flower_cookie,
+				      adap->flower_ht_params);
+}
+
+static void cxgb4_process_flow_match(struct net_device *dev,
+				     struct flow_rule *rule,
+				     struct ch_filter_specification *fs)
+{
+	u16 addr_type = 0;
+
+	if (flow_rule_match_key(rule, FLOW_DISSECTOR_KEY_CONTROL)) {
+		struct flow_match_control match;
+
+		flow_rule_match_control(rule, &match);
+		addr_type = match.key->addr_type;
+	} else if (flow_rule_match_key(rule, FLOW_DISSECTOR_KEY_IPV4_ADDRS)) {
+		addr_type = FLOW_DISSECTOR_KEY_IPV4_ADDRS;
+	} else if (flow_rule_match_key(rule, FLOW_DISSECTOR_KEY_IPV6_ADDRS)) {
+		addr_type = FLOW_DISSECTOR_KEY_IPV6_ADDRS;
+	}
+
+	if (flow_rule_match_key(rule, FLOW_DISSECTOR_KEY_BASIC)) {
+		struct flow_match_basic match;
+		u16 ethtype_key, ethtype_mask;
+
+		flow_rule_match_basic(rule, &match);
+		ethtype_key = ntohs(match.key->n_proto);
+		ethtype_mask = ntohs(match.mask->n_proto);
+
+		if (ethtype_key == ETH_P_ALL) {
+			ethtype_key = 0;
+			ethtype_mask = 0;
+		}
+
+		if (ethtype_key == ETH_P_IPV6)
+			fs->type = 1;
+
+		fs->val.ethtype = ethtype_key;
+		fs->mask.ethtype = ethtype_mask;
+		fs->val.proto = match.key->ip_proto;
+		fs->mask.proto = match.mask->ip_proto;
+	}
+
+	if (addr_type == FLOW_DISSECTOR_KEY_IPV4_ADDRS) {
+		struct flow_match_ipv4_addrs match;
+
+		flow_rule_match_ipv4_addrs(rule, &match);
+		fs->type = 0;
+		memcpy(&fs->val.lip[0], &match.key->dst, sizeof(match.key->dst));
+		memcpy(&fs->val.fip[0], &match.key->src, sizeof(match.key->src));
+		memcpy(&fs->mask.lip[0], &match.mask->dst, sizeof(match.mask->dst));
+		memcpy(&fs->mask.fip[0], &match.mask->src, sizeof(match.mask->src));
+
+		/* also initialize nat_lip/fip to same values */
+		memcpy(&fs->nat_lip[0], &match.key->dst, sizeof(match.key->dst));
+		memcpy(&fs->nat_fip[0], &match.key->src, sizeof(match.key->src));
+	}
+
+	if (addr_type == FLOW_DISSECTOR_KEY_IPV6_ADDRS) {
+		struct flow_match_ipv6_addrs match;
+
+		flow_rule_match_ipv6_addrs(rule, &match);
+		fs->type = 1;
+		memcpy(&fs->val.lip[0], match.key->dst.s6_addr,
+		       sizeof(match.key->dst));
+		memcpy(&fs->val.fip[0], match.key->src.s6_addr,
+		       sizeof(match.key->src));
+		memcpy(&fs->mask.lip[0], match.mask->dst.s6_addr,
+		       sizeof(match.mask->dst));
+		memcpy(&fs->mask.fip[0], match.mask->src.s6_addr,
+		       sizeof(match.mask->src));
+
+		/* also initialize nat_lip/fip to same values */
+		memcpy(&fs->nat_lip[0], match.key->dst.s6_addr,
+		       sizeof(match.key->dst));
+		memcpy(&fs->nat_fip[0], match.key->src.s6_addr,
+		       sizeof(match.key->src));
+	}
+
+	if (flow_rule_match_key(rule, FLOW_DISSECTOR_KEY_PORTS)) {
+		struct flow_match_ports match;
+
+		flow_rule_match_ports(rule, &match);
+		fs->val.lport = be16_to_cpu(match.key->dst);
+		fs->mask.lport = be16_to_cpu(match.mask->dst);
+		fs->val.fport = be16_to_cpu(match.key->src);
+		fs->mask.fport = be16_to_cpu(match.mask->src);
+
+		/* also initialize nat_lport/fport to same values */
+		fs->nat_lport = fs->val.lport;
+		fs->nat_fport = fs->val.fport;
+	}
+
+	if (flow_rule_match_key(rule, FLOW_DISSECTOR_KEY_IP)) {
+		struct flow_match_ip match;
+
+		flow_rule_match_ip(rule, &match);
+		fs->val.tos = match.key->tos;
+		fs->mask.tos = match.mask->tos;
+	}
+
+	if (flow_rule_match_key(rule, FLOW_DISSECTOR_KEY_ENC_KEYID)) {
+		struct flow_match_enc_keyid match;
+
+		flow_rule_match_enc_keyid(rule, &match);
+		fs->val.vni = be32_to_cpu(match.key->keyid);
+		fs->mask.vni = be32_to_cpu(match.mask->keyid);
+		if (fs->mask.vni) {
+			fs->val.encap_vld = 1;
+			fs->mask.encap_vld = 1;
+		}
+	}
+
+	if (flow_rule_match_key(rule, FLOW_DISSECTOR_KEY_VLAN)) {
+		struct flow_match_vlan match;
+		u16 vlan_tci, vlan_tci_mask;
+
+		flow_rule_match_vlan(rule, &match);
+		vlan_tci = match.key->vlan_id | (match.key->vlan_priority <<
+					       VLAN_PRIO_SHIFT);
+		vlan_tci_mask = match.mask->vlan_id | (match.mask->vlan_priority <<
+						     VLAN_PRIO_SHIFT);
+		fs->val.ivlan = vlan_tci;
+		fs->mask.ivlan = vlan_tci_mask;
+
+		fs->val.ivlan_vld = 1;
+		fs->mask.ivlan_vld = 1;
+
+		/* Chelsio adapters use ivlan_vld bit to match vlan packets
+		 * as 802.1Q. Also, when vlan tag is present in packets,
+		 * ethtype match is used then to match on ethtype of inner
+		 * header ie. the header following the vlan header.
+		 * So, set the ivlan_vld based on ethtype info supplied by
+		 * TC for vlan packets if its 802.1Q. And then reset the
+		 * ethtype value else, hw will try to match the supplied
+		 * ethtype value with ethtype of inner header.
+		 */
+		if (fs->val.ethtype == ETH_P_8021Q) {
+			fs->val.ethtype = 0;
+			fs->mask.ethtype = 0;
+		}
+	}
+
+	/* Match only packets coming from the ingress port where this
+	 * filter will be created.
+	 */
+	fs->val.iport = netdev2pinfo(dev)->port_id;
+	fs->mask.iport = ~0;
+}
+
+static int cxgb4_validate_flow_match(struct netlink_ext_ack *extack,
+				     struct flow_rule *rule)
+{
+	struct flow_dissector *dissector = rule->match.dissector;
+	u16 ethtype_mask = 0;
+	u16 ethtype_key = 0;
+
+	if (dissector->used_keys &
+	    ~(BIT_ULL(FLOW_DISSECTOR_KEY_CONTROL) |
+	      BIT_ULL(FLOW_DISSECTOR_KEY_BASIC) |
+	      BIT_ULL(FLOW_DISSECTOR_KEY_IPV4_ADDRS) |
+	      BIT_ULL(FLOW_DISSECTOR_KEY_IPV6_ADDRS) |
+	      BIT_ULL(FLOW_DISSECTOR_KEY_PORTS) |
+	      BIT_ULL(FLOW_DISSECTOR_KEY_ENC_KEYID) |
+	      BIT_ULL(FLOW_DISSECTOR_KEY_VLAN) |
+	      BIT_ULL(FLOW_DISSECTOR_KEY_IP))) {
+		NL_SET_ERR_MSG_FMT_MOD(extack,
+				       "Unsupported key used: 0x%llx",
+				       dissector->used_keys);
+		return -EOPNOTSUPP;
+	}
+
+	if (flow_rule_match_has_control_flags(rule, extack))
+		return -EOPNOTSUPP;
+
+	if (flow_rule_match_key(rule, FLOW_DISSECTOR_KEY_BASIC)) {
+		struct flow_match_basic match;
+
+		flow_rule_match_basic(rule, &match);
+		ethtype_key = ntohs(match.key->n_proto);
+		ethtype_mask = ntohs(match.mask->n_proto);
+	}
+
+	if (flow_rule_match_key(rule, FLOW_DISSECTOR_KEY_IP)) {
+		u16 eth_ip_type = ethtype_key & ethtype_mask;
+		struct flow_match_ip match;
+
+		if (eth_ip_type != ETH_P_IP && eth_ip_type != ETH_P_IPV6) {
+			NL_SET_ERR_MSG_MOD(extack,
+					   "IP Key supported only with IPv4/v6");
+			return -EINVAL;
+		}
+
+		flow_rule_match_ip(rule, &match);
+		if (match.mask->ttl) {
+			NL_SET_ERR_MSG_MOD(extack,
+					   "ttl match unsupported for offload");
+			return -EOPNOTSUPP;
+		}
+	}
+
+	return 0;
+}
+
+static void offload_pedit(struct ch_filter_specification *fs, u32 val, u32 mask,
+			  u8 field)
+{
+	u32 set_val = val & ~mask;
+	u32 offset = 0;
+	u8 size = 1;
+	int i;
+
+	for (i = 0; i < ARRAY_SIZE(pedits); i++) {
+		if (pedits[i].field == field) {
+			offset = pedits[i].offset;
+			size = pedits[i].size;
+			break;
+		}
+	}
+	memcpy((u8 *)fs + offset, &set_val, size);
+}
+
+static void process_pedit_field(struct ch_filter_specification *fs, u32 val,
+				u32 mask, u32 offset, u8 htype,
+				u8 *natmode_flags)
+{
+	switch (htype) {
+	case FLOW_ACT_MANGLE_HDR_TYPE_ETH:
+		switch (offset) {
+		case PEDIT_ETH_DMAC_31_0:
+			fs->newdmac = 1;
+			offload_pedit(fs, val, mask, ETH_DMAC_31_0);
+			break;
+		case PEDIT_ETH_DMAC_47_32_SMAC_15_0:
+			if (~mask & PEDIT_ETH_DMAC_MASK)
+				offload_pedit(fs, val, mask, ETH_DMAC_47_32);
+			else
+				offload_pedit(fs, val >> 16, mask >> 16,
+					      ETH_SMAC_15_0);
+			break;
+		case PEDIT_ETH_SMAC_47_16:
+			fs->newsmac = 1;
+			offload_pedit(fs, val, mask, ETH_SMAC_47_16);
+		}
+		break;
+	case FLOW_ACT_MANGLE_HDR_TYPE_IP4:
+		switch (offset) {
+		case PEDIT_IP4_SRC:
+			offload_pedit(fs, val, mask, IP4_SRC);
+			*natmode_flags |= CXGB4_ACTION_NATMODE_SIP;
+			break;
+		case PEDIT_IP4_DST:
+			offload_pedit(fs, val, mask, IP4_DST);
+			*natmode_flags |= CXGB4_ACTION_NATMODE_DIP;
+		}
+		break;
+	case FLOW_ACT_MANGLE_HDR_TYPE_IP6:
+		switch (offset) {
+		case PEDIT_IP6_SRC_31_0:
+			offload_pedit(fs, val, mask, IP6_SRC_31_0);
+			*natmode_flags |= CXGB4_ACTION_NATMODE_SIP;
+			break;
+		case PEDIT_IP6_SRC_63_32:
+			offload_pedit(fs, val, mask, IP6_SRC_63_32);
+			*natmode_flags |=  CXGB4_ACTION_NATMODE_SIP;
+			break;
+		case PEDIT_IP6_SRC_95_64:
+			offload_pedit(fs, val, mask, IP6_SRC_95_64);
+			*natmode_flags |= CXGB4_ACTION_NATMODE_SIP;
+			break;
+		case PEDIT_IP6_SRC_127_96:
+			offload_pedit(fs, val, mask, IP6_SRC_127_96);
+			*natmode_flags |= CXGB4_ACTION_NATMODE_SIP;
+			break;
+		case PEDIT_IP6_DST_31_0:
+			offload_pedit(fs, val, mask, IP6_DST_31_0);
+			*natmode_flags |= CXGB4_ACTION_NATMODE_DIP;
+			break;
+		case PEDIT_IP6_DST_63_32:
+			offload_pedit(fs, val, mask, IP6_DST_63_32);
+			*natmode_flags |= CXGB4_ACTION_NATMODE_DIP;
+			break;
+		case PEDIT_IP6_DST_95_64:
+			offload_pedit(fs, val, mask, IP6_DST_95_64);
+			*natmode_flags |= CXGB4_ACTION_NATMODE_DIP;
+			break;
+		case PEDIT_IP6_DST_127_96:
+			offload_pedit(fs, val, mask, IP6_DST_127_96);
+			*natmode_flags |= CXGB4_ACTION_NATMODE_DIP;
+		}
+		break;
+	case FLOW_ACT_MANGLE_HDR_TYPE_TCP:
+		switch (offset) {
+		case PEDIT_TCP_SPORT_DPORT:
+			if (~mask & PEDIT_TCP_UDP_SPORT_MASK) {
+				fs->nat_fport = val;
+				*natmode_flags |= CXGB4_ACTION_NATMODE_SPORT;
+			} else {
+				fs->nat_lport = val >> 16;
+				*natmode_flags |= CXGB4_ACTION_NATMODE_DPORT;
+			}
+		}
+		break;
+	case FLOW_ACT_MANGLE_HDR_TYPE_UDP:
+		switch (offset) {
+		case PEDIT_UDP_SPORT_DPORT:
+			if (~mask & PEDIT_TCP_UDP_SPORT_MASK) {
+				fs->nat_fport = val;
+				*natmode_flags |= CXGB4_ACTION_NATMODE_SPORT;
+			} else {
+				fs->nat_lport = val >> 16;
+				*natmode_flags |= CXGB4_ACTION_NATMODE_DPORT;
+			}
+		}
+		break;
+	}
+}
+
+static int cxgb4_action_natmode_validate(struct adapter *adap, u8 natmode_flags,
+					 struct netlink_ext_ack *extack)
+{
+	u8 i = 0;
+
+	/* Extract the NAT mode to enable based on what 4-tuple fields
+	 * are enabled to be overwritten. This ensures that the
+	 * disabled fields don't get overwritten to 0.
+	 */
+	for (i = 0; i < ARRAY_SIZE(cxgb4_natmode_config_array); i++) {
+		const struct cxgb4_natmode_config *c;
+
+		c = &cxgb4_natmode_config_array[i];
+		if (CHELSIO_CHIP_VERSION(adap->params.chip) >= c->chip &&
+		    natmode_flags == c->flags)
+			return 0;
+	}
+	NL_SET_ERR_MSG_MOD(extack, "Unsupported NAT mode 4-tuple combination");
+	return -EOPNOTSUPP;
+}
+
+void cxgb4_process_flow_actions(struct net_device *in,
+				struct flow_action *actions,
+				struct ch_filter_specification *fs)
+{
+	struct flow_action_entry *act;
+	u8 natmode_flags = 0;
+	int i;
+
+	flow_action_for_each(i, act, actions) {
+		switch (act->id) {
+		case FLOW_ACTION_ACCEPT:
+			fs->action = FILTER_PASS;
+			break;
+		case FLOW_ACTION_DROP:
+			fs->action = FILTER_DROP;
+			break;
+		case FLOW_ACTION_MIRRED:
+		case FLOW_ACTION_REDIRECT: {
+			struct net_device *out = act->dev;
+			struct port_info *pi = netdev_priv(out);
+
+			fs->action = FILTER_SWITCH;
+			fs->eport = pi->port_id;
+			}
+			break;
+		case FLOW_ACTION_VLAN_POP:
+		case FLOW_ACTION_VLAN_PUSH:
+		case FLOW_ACTION_VLAN_MANGLE: {
+			u8 prio = act->vlan.prio;
+			u16 vid = act->vlan.vid;
+			u16 vlan_tci = (prio << VLAN_PRIO_SHIFT) | vid;
+			switch (act->id) {
+			case FLOW_ACTION_VLAN_POP:
+				fs->newvlan |= VLAN_REMOVE;
+				break;
+			case FLOW_ACTION_VLAN_PUSH:
+				fs->newvlan |= VLAN_INSERT;
+				fs->vlan = vlan_tci;
+				break;
+			case FLOW_ACTION_VLAN_MANGLE:
+				fs->newvlan |= VLAN_REWRITE;
+				fs->vlan = vlan_tci;
+				break;
+			default:
+				break;
+			}
+			}
+			break;
+		case FLOW_ACTION_MANGLE: {
+			u32 mask, val, offset;
+			u8 htype;
+
+			htype = act->mangle.htype;
+			mask = act->mangle.mask;
+			val = act->mangle.val;
+			offset = act->mangle.offset;
+
+			process_pedit_field(fs, val, mask, offset, htype,
+					    &natmode_flags);
+			}
+			break;
+		case FLOW_ACTION_QUEUE:
+			fs->action = FILTER_PASS;
+			fs->dirsteer = 1;
+			fs->iq = act->queue.index;
+			break;
+		default:
+			break;
+		}
+	}
+	if (natmode_flags)
+		cxgb4_action_natmode_tweak(fs, natmode_flags);
+
+}
+
+static bool valid_l4_mask(u32 mask)
+{
+	u16 hi, lo;
+
+	/* Either the upper 16-bits (SPORT) OR the lower
+	 * 16-bits (DPORT) can be set, but NOT BOTH.
+	 */
+	hi = (mask >> 16) & 0xFFFF;
+	lo = mask & 0xFFFF;
+
+	return hi && lo ? false : true;
+}
+
+static bool valid_pedit_action(struct netlink_ext_ack *extack,
+			       const struct flow_action_entry *act,
+			       u8 *natmode_flags)
+{
+	u32 mask, offset;
+	u8 htype;
+
+	htype = act->mangle.htype;
+	mask = act->mangle.mask;
+	offset = act->mangle.offset;
+
+	switch (htype) {
+	case FLOW_ACT_MANGLE_HDR_TYPE_ETH:
+		switch (offset) {
+		case PEDIT_ETH_DMAC_31_0:
+		case PEDIT_ETH_DMAC_47_32_SMAC_15_0:
+		case PEDIT_ETH_SMAC_47_16:
+			break;
+		default:
+			NL_SET_ERR_MSG_MOD(extack, "Unsupported pedit field");
+			return false;
+		}
+		break;
+	case FLOW_ACT_MANGLE_HDR_TYPE_IP4:
+		switch (offset) {
+		case PEDIT_IP4_SRC:
+			*natmode_flags |= CXGB4_ACTION_NATMODE_SIP;
+			break;
+		case PEDIT_IP4_DST:
+			*natmode_flags |= CXGB4_ACTION_NATMODE_DIP;
+			break;
+		default:
+			NL_SET_ERR_MSG_MOD(extack, "Unsupported pedit field");
+			return false;
+		}
+		break;
+	case FLOW_ACT_MANGLE_HDR_TYPE_IP6:
+		switch (offset) {
+		case PEDIT_IP6_SRC_31_0:
+		case PEDIT_IP6_SRC_63_32:
+		case PEDIT_IP6_SRC_95_64:
+		case PEDIT_IP6_SRC_127_96:
+			*natmode_flags |= CXGB4_ACTION_NATMODE_SIP;
+			break;
+		case PEDIT_IP6_DST_31_0:
+		case PEDIT_IP6_DST_63_32:
+		case PEDIT_IP6_DST_95_64:
+		case PEDIT_IP6_DST_127_96:
+			*natmode_flags |= CXGB4_ACTION_NATMODE_DIP;
+			break;
+		default:
+			NL_SET_ERR_MSG_MOD(extack, "Unsupported pedit field");
+			return false;
+		}
+		break;
+	case FLOW_ACT_MANGLE_HDR_TYPE_TCP:
+		switch (offset) {
+		case PEDIT_TCP_SPORT_DPORT:
+			if (!valid_l4_mask(~mask)) {
+				NL_SET_ERR_MSG_MOD(extack,
+						   "Unsupported mask for TCP L4 ports");
+				return false;
+			}
+			if (~mask & PEDIT_TCP_UDP_SPORT_MASK)
+				*natmode_flags |= CXGB4_ACTION_NATMODE_SPORT;
+			else
+				*natmode_flags |= CXGB4_ACTION_NATMODE_DPORT;
+			break;
+		default:
+			NL_SET_ERR_MSG_MOD(extack, "Unsupported pedit field");
+			return false;
+		}
+		break;
+	case FLOW_ACT_MANGLE_HDR_TYPE_UDP:
+		switch (offset) {
+		case PEDIT_UDP_SPORT_DPORT:
+			if (!valid_l4_mask(~mask)) {
+				NL_SET_ERR_MSG_MOD(extack,
+						   "Unsupported mask for UDP L4 ports");
+				return false;
+			}
+			if (~mask & PEDIT_TCP_UDP_SPORT_MASK)
+				*natmode_flags |= CXGB4_ACTION_NATMODE_SPORT;
+			else
+				*natmode_flags |= CXGB4_ACTION_NATMODE_DPORT;
+			break;
+		default:
+			NL_SET_ERR_MSG_MOD(extack, "Unsupported pedit field");
+			return false;
+		}
+		break;
+	default:
+		NL_SET_ERR_MSG_MOD(extack, "Unsupported pedit type");
+		return false;
+	}
+	return true;
+}
+
+int cxgb4_validate_flow_actions(struct net_device *dev,
+				struct flow_action *actions,
+				struct netlink_ext_ack *extack,
+				u8 matchall_filter)
+{
+	struct adapter *adap = netdev2adap(dev);
+	struct flow_action_entry *act;
+	bool act_redir = false;
+	bool act_pedit = false;
+	bool act_vlan = false;
+	u8 natmode_flags = 0;
+	int i;
+
+	if (!flow_action_basic_hw_stats_check(actions, extack))
+		return -EOPNOTSUPP;
+
+	flow_action_for_each(i, act, actions) {
+		switch (act->id) {
+		case FLOW_ACTION_ACCEPT:
+		case FLOW_ACTION_DROP:
+			/* Do nothing */
+			break;
+		case FLOW_ACTION_MIRRED:
+		case FLOW_ACTION_REDIRECT: {
+			struct net_device *n_dev, *target_dev;
+			bool found = false;
+			unsigned int i;
+
+			if (act->id == FLOW_ACTION_MIRRED &&
+			    !matchall_filter) {
+				NL_SET_ERR_MSG_MOD(extack,
+						   "Egress mirror action is only supported for tc-matchall");
+				return -EOPNOTSUPP;
+			}
+
+			target_dev = act->dev;
+			for_each_port(adap, i) {
+				n_dev = adap->port[i];
+				if (target_dev == n_dev) {
+					found = true;
+					break;
+				}
+			}
+
+			/* If interface doesn't belong to our hw, then
+			 * the provided output port is not valid
+			 */
+			if (!found) {
+				NL_SET_ERR_MSG_MOD(extack, "Out port invalid");
+				return -EINVAL;
+			}
+			act_redir = true;
+			}
+			break;
+		case FLOW_ACTION_VLAN_POP:
+		case FLOW_ACTION_VLAN_PUSH:
+		case FLOW_ACTION_VLAN_MANGLE: {
+			u16 proto = be16_to_cpu(act->vlan.proto);
+
+			switch (act->id) {
+			case FLOW_ACTION_VLAN_POP:
+				break;
+			case FLOW_ACTION_VLAN_PUSH:
+			case FLOW_ACTION_VLAN_MANGLE:
+				if (proto != ETH_P_8021Q) {
+					NL_SET_ERR_MSG_MOD(extack,
+							   "Unsupported vlan proto");
+					return -EOPNOTSUPP;
+				}
+				break;
+			default:
+				NL_SET_ERR_MSG_MOD(extack,
+						   "Unsupported vlan action");
+				return -EOPNOTSUPP;
+			}
+			act_vlan = true;
+			}
+			break;
+		case FLOW_ACTION_MANGLE: {
+			bool pedit_valid = valid_pedit_action(extack, act,
+							      &natmode_flags);
+
+			if (!pedit_valid)
+				return -EOPNOTSUPP;
+			act_pedit = true;
+			}
+			break;
+		case FLOW_ACTION_QUEUE:
+			/* Do nothing. cxgb4_set_filter will validate */
+			break;
+		default:
+			NL_SET_ERR_MSG_MOD(extack, "Unsupported action");
+			return -EOPNOTSUPP;
+		}
+	}
+
+	if ((act_pedit || act_vlan) && !act_redir) {
+		NL_SET_ERR_MSG_MOD(extack,
+				   "pedit/vlan rewrite invalid without egress redirect");
+		return -EINVAL;
+	}
+
+	if (act_pedit) {
+		int ret;
+
+		ret = cxgb4_action_natmode_validate(adap, natmode_flags,
+						    extack);
+		if (ret)
+			return ret;
+	}
+
+	return 0;
+}
+
+static void cxgb4_tc_flower_hash_prio_add(struct adapter *adap, u32 tc_prio)
+{
+	spin_lock_bh(&adap->tids.ftid_lock);
+	if (adap->tids.tc_hash_tids_max_prio < tc_prio)
+		adap->tids.tc_hash_tids_max_prio = tc_prio;
+	spin_unlock_bh(&adap->tids.ftid_lock);
+}
+
+static void cxgb4_tc_flower_hash_prio_del(struct adapter *adap, u32 tc_prio)
+{
+	struct tid_info *t = &adap->tids;
+	struct ch_tc_flower_entry *fe;
+	struct rhashtable_iter iter;
+	u32 found = 0;
+
+	spin_lock_bh(&t->ftid_lock);
+	/* Bail if the current rule is not the one with the max
+	 * prio.
+	 */
+	if (t->tc_hash_tids_max_prio != tc_prio)
+		goto out_unlock;
+
+	/* Search for the next rule having the same or next lower
+	 * max prio.
+	 */
+	rhashtable_walk_enter(&adap->flower_tbl, &iter);
+	do {
+		rhashtable_walk_start(&iter);
+
+		fe = rhashtable_walk_next(&iter);
+		while (!IS_ERR_OR_NULL(fe)) {
+			if (fe->fs.hash &&
+			    fe->fs.tc_prio <= t->tc_hash_tids_max_prio) {
+				t->tc_hash_tids_max_prio = fe->fs.tc_prio;
+				found++;
+
+				/* Bail if we found another rule
+				 * having the same prio as the
+				 * current max one.
+				 */
+				if (fe->fs.tc_prio == tc_prio)
+					break;
+			}
+
+			fe = rhashtable_walk_next(&iter);
+		}
+
+		rhashtable_walk_stop(&iter);
+	} while (fe == ERR_PTR(-EAGAIN));
+	rhashtable_walk_exit(&iter);
+
+	if (!found)
+		t->tc_hash_tids_max_prio = 0;
+
+out_unlock:
+	spin_unlock_bh(&t->ftid_lock);
+}
+
+int cxgb4_flow_rule_replace(struct net_device *dev, struct flow_rule *rule,
+			    u32 tc_prio, struct netlink_ext_ack *extack,
+			    struct ch_filter_specification *fs, u32 *tid)
+{
+	struct adapter *adap = netdev2adap(dev);
+	struct filter_ctx ctx;
+	u8 inet_family;
+	int fidx, ret;
+
+	if (cxgb4_validate_flow_actions(dev, &rule->action, extack, 0))
+		return -EOPNOTSUPP;
+
+	if (cxgb4_validate_flow_match(extack, rule))
+		return -EOPNOTSUPP;
+
+	cxgb4_process_flow_match(dev, rule, fs);
+	cxgb4_process_flow_actions(dev, &rule->action, fs);
+
+	fs->hash = is_filter_exact_match(adap, fs);
+	inet_family = fs->type ? PF_INET6 : PF_INET;
+
+	/* Get a free filter entry TID, where we can insert this new
+	 * rule. Only insert rule if its prio doesn't conflict with
+	 * existing rules.
+	 */
+	fidx = cxgb4_get_free_ftid(dev, inet_family, fs->hash,
+				   tc_prio);
+	if (fidx < 0) {
+		NL_SET_ERR_MSG_MOD(extack,
+				   "No free LETCAM index available");
+		return -ENOMEM;
+	}
+
+	if (fidx < adap->tids.nhpftids) {
+		fs->prio = 1;
+		fs->hash = 0;
+	}
+
+	/* If the rule can be inserted into HASH region, then ignore
+	 * the index to normal FILTER region.
+	 */
+	if (fs->hash)
+		fidx = 0;
+
+	fs->tc_prio = tc_prio;
+
+	init_completion(&ctx.completion);
+	ret = __cxgb4_set_filter(dev, fidx, fs, &ctx);
+	if (ret) {
+		NL_SET_ERR_MSG_FMT_MOD(extack, "filter creation err %d", ret);
+		return ret;
+	}
+
+	/* Wait for reply */
+	ret = wait_for_completion_timeout(&ctx.completion, 10 * HZ);
+	if (!ret)
+		return -ETIMEDOUT;
+
+	/* Check if hw returned error for filter creation */
+	if (ctx.result)
+		return ctx.result;
+
+	*tid = ctx.tid;
+
+	if (fs->hash)
+		cxgb4_tc_flower_hash_prio_add(adap, tc_prio);
+
+	return 0;
+}
+
+int cxgb4_tc_flower_replace(struct net_device *dev,
+			    struct flow_cls_offload *cls)
+{
+	struct flow_rule *rule = flow_cls_offload_flow_rule(cls);
+	struct netlink_ext_ack *extack = cls->common.extack;
+	struct adapter *adap = netdev2adap(dev);
+	struct ch_tc_flower_entry *ch_flower;
+	struct ch_filter_specification *fs;
+	int ret;
+
+	ch_flower = allocate_flower_entry();
+	if (!ch_flower) {
+		netdev_err(dev, "%s: ch_flower alloc failed.\n", __func__);
+		return -ENOMEM;
+	}
+
+	fs = &ch_flower->fs;
+	fs->hitcnts = 1;
+	fs->tc_cookie = cls->cookie;
+
+	ret = cxgb4_flow_rule_replace(dev, rule, cls->common.prio, extack, fs,
+				      &ch_flower->filter_id);
+	if (ret)
+		goto free_entry;
+
+	ch_flower->tc_flower_cookie = cls->cookie;
+	ret = rhashtable_insert_fast(&adap->flower_tbl, &ch_flower->node,
+				     adap->flower_ht_params);
+	if (ret)
+		goto del_filter;
+
+	return 0;
+
+del_filter:
+	if (fs->hash)
+		cxgb4_tc_flower_hash_prio_del(adap, cls->common.prio);
+
+	cxgb4_del_filter(dev, ch_flower->filter_id, &ch_flower->fs);
+
+free_entry:
+	kfree(ch_flower);
+	return ret;
+}
+
+int cxgb4_flow_rule_destroy(struct net_device *dev, u32 tc_prio,
+			    struct ch_filter_specification *fs, int tid)
+{
+	struct adapter *adap = netdev2adap(dev);
+	u8 hash;
+	int ret;
+
+	hash = fs->hash;
+
+	ret = cxgb4_del_filter(dev, tid, fs);
+	if (ret)
+		return ret;
+
+	if (hash)
+		cxgb4_tc_flower_hash_prio_del(adap, tc_prio);
+
+	return ret;
+}
+
+int cxgb4_tc_flower_destroy(struct net_device *dev,
+			    struct flow_cls_offload *cls)
+{
+	struct adapter *adap = netdev2adap(dev);
+	struct ch_tc_flower_entry *ch_flower;
+	int ret;
+
+	ch_flower = ch_flower_lookup(adap, cls->cookie);
+	if (!ch_flower)
+		return -ENOENT;
+
+	rhashtable_remove_fast(&adap->flower_tbl, &ch_flower->node,
+			       adap->flower_ht_params);
+
+	ret = cxgb4_flow_rule_destroy(dev, ch_flower->fs.tc_prio,
+				      &ch_flower->fs, ch_flower->filter_id);
+	if (ret)
+		netdev_err(dev, "Flow rule destroy failed for tid: %u, ret: %d",
+			   ch_flower->filter_id, ret);
+
+	kfree_rcu(ch_flower, rcu);
+	return ret;
+}
+
+static void ch_flower_stats_handler(struct work_struct *work)
+{
+	struct adapter *adap = container_of(work, struct adapter,
+					    flower_stats_work);
+	struct ch_tc_flower_entry *flower_entry;
+	struct ch_tc_flower_stats *ofld_stats;
+	struct rhashtable_iter iter;
+	u64 packets;
+	u64 bytes;
+	int ret;
+
+	rhashtable_walk_enter(&adap->flower_tbl, &iter);
+	do {
+		rhashtable_walk_start(&iter);
+
+		while ((flower_entry = rhashtable_walk_next(&iter)) &&
+		       !IS_ERR(flower_entry)) {
+			ret = cxgb4_get_filter_counters(adap->port[0],
+							flower_entry->filter_id,
+							&packets, &bytes,
+							flower_entry->fs.hash);
+			if (!ret) {
+				spin_lock(&flower_entry->lock);
+				ofld_stats = &flower_entry->stats;
+
+				if (ofld_stats->prev_packet_count != packets) {
+					ofld_stats->prev_packet_count = packets;
+					ofld_stats->last_used = jiffies;
+				}
+				spin_unlock(&flower_entry->lock);
+			}
+		}
+
+		rhashtable_walk_stop(&iter);
+
+	} while (flower_entry == ERR_PTR(-EAGAIN));
+	rhashtable_walk_exit(&iter);
+	mod_timer(&adap->flower_stats_timer, jiffies + STATS_CHECK_PERIOD);
+}
+
+static void ch_flower_stats_cb(struct timer_list *t)
+{
+	struct adapter *adap = timer_container_of(adap, t, flower_stats_timer);
+
+	schedule_work(&adap->flower_stats_work);
+}
+
+int cxgb4_tc_flower_stats(struct net_device *dev,
+			  struct flow_cls_offload *cls)
+{
+	struct adapter *adap = netdev2adap(dev);
+	struct ch_tc_flower_stats *ofld_stats;
+	struct ch_tc_flower_entry *ch_flower;
+	u64 packets;
+	u64 bytes;
+	int ret;
+
+	ch_flower = ch_flower_lookup(adap, cls->cookie);
+	if (!ch_flower) {
+		ret = -ENOENT;
+		goto err;
+	}
+
+	ret = cxgb4_get_filter_counters(dev, ch_flower->filter_id,
+					&packets, &bytes,
+					ch_flower->fs.hash);
+	if (ret < 0)
+		goto err;
+
+	spin_lock_bh(&ch_flower->lock);
+	ofld_stats = &ch_flower->stats;
+	if (ofld_stats->packet_count != packets) {
+		if (ofld_stats->prev_packet_count != packets)
+			ofld_stats->last_used = jiffies;
+		flow_stats_update(&cls->stats, bytes - ofld_stats->byte_count,
+				  packets - ofld_stats->packet_count, 0,
+				  ofld_stats->last_used,
+				  FLOW_ACTION_HW_STATS_IMMEDIATE);
+
+		ofld_stats->packet_count = packets;
+		ofld_stats->byte_count = bytes;
+		ofld_stats->prev_packet_count = packets;
+	}
+	spin_unlock_bh(&ch_flower->lock);
+	return 0;
+
+err:
+	return ret;
+}
+
+static const struct rhashtable_params cxgb4_tc_flower_ht_params = {
+	.nelem_hint = 384,
+	.head_offset = offsetof(struct ch_tc_flower_entry, node),
+	.key_offset = offsetof(struct ch_tc_flower_entry, tc_flower_cookie),
+	.key_len = sizeof(((struct ch_tc_flower_entry *)0)->tc_flower_cookie),
+	.max_size = 524288,
+	.min_size = 512,
+	.automatic_shrinking = true
+};
+
+int cxgb4_init_tc_flower(struct adapter *adap)
+{
+	int ret;
+
+	if (adap->tc_flower_initialized)
+		return -EEXIST;
+
+	adap->flower_ht_params = cxgb4_tc_flower_ht_params;
+	ret = rhashtable_init(&adap->flower_tbl, &adap->flower_ht_params);
+	if (ret)
+		return ret;
+
+	INIT_WORK(&adap->flower_stats_work, ch_flower_stats_handler);
+	timer_setup(&adap->flower_stats_timer, ch_flower_stats_cb, 0);
+	mod_timer(&adap->flower_stats_timer, jiffies + STATS_CHECK_PERIOD);
+	adap->tc_flower_initialized = true;
+	return 0;
+}
+
+void cxgb4_cleanup_tc_flower(struct adapter *adap)
+{
+	if (!adap->tc_flower_initialized)
+		return;
+
+	if (adap->flower_stats_timer.function)
+		timer_shutdown_sync(&adap->flower_stats_timer);
+	cancel_work_sync(&adap->flower_stats_work);
+	rhashtable_destroy(&adap->flower_tbl);
+	adap->tc_flower_initialized = false;
+}
diff --git a/drivers/net/ethernet/chelsio/cxgb4/cxgb4_tc_flower.h b/drivers/net/ethernet/chelsio/cxgb4/cxgb4_tc_flower.h
new file mode 100644
index 000000000000..3a2fa00c8cde
--- /dev/null
+++ b/drivers/net/ethernet/chelsio/cxgb4/cxgb4_tc_flower.h
@@ -0,0 +1,148 @@
+/*
+ * This file is part of the Chelsio T4/T5/T6 Ethernet driver for Linux.
+ *
+ * Copyright (c) 2017 Chelsio Communications, Inc. All rights reserved.
+ *
+ * This software is available to you under a choice of one of two
+ * licenses.  You may choose to be licensed under the terms of the GNU
+ * General Public License (GPL) Version 2, available from the file
+ * COPYING in the main directory of this source tree, or the
+ * OpenIB.org BSD license below:
+ *
+ *     Redistribution and use in source and binary forms, with or
+ *     without modification, are permitted provided that the following
+ *     conditions are met:
+ *
+ *      - Redistributions of source code must retain the above
+ *        copyright notice, this list of conditions and the following
+ *        disclaimer.
+ *
+ *      - Redistributions in binary form must reproduce the above
+ *        copyright notice, this list of conditions and the following
+ *        disclaimer in the documentation and/or other materials
+ *        provided with the distribution.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND,
+ * EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF
+ * MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND
+ * NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS
+ * BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN
+ * ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN
+ * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
+ * SOFTWARE.
+ */
+
+#ifndef __CXGB4_TC_FLOWER_H
+#define __CXGB4_TC_FLOWER_H
+
+#include <net/pkt_cls.h>
+
+struct ch_tc_flower_stats {
+	u64 prev_packet_count;
+	u64 packet_count;
+	u64 byte_count;
+	u64 last_used;
+};
+
+struct ch_tc_flower_entry {
+	struct ch_filter_specification fs;
+	struct ch_tc_flower_stats stats;
+	unsigned long tc_flower_cookie;
+	struct rhash_head node;
+	struct rcu_head rcu;
+	spinlock_t lock; /* lock for stats */
+	u32 filter_id;
+};
+
+enum {
+	ETH_DMAC_31_0,	/* dmac bits 0.. 31 */
+	ETH_DMAC_47_32,	/* dmac bits 32..47 */
+	ETH_SMAC_15_0,	/* smac bits 0.. 15 */
+	ETH_SMAC_47_16,	/* smac bits 16..47 */
+
+	IP4_SRC,	/* 32-bit IPv4 src  */
+	IP4_DST,	/* 32-bit IPv4 dst  */
+
+	IP6_SRC_31_0,	/* src bits 0..  31 */
+	IP6_SRC_63_32,	/* src bits 63.. 32 */
+	IP6_SRC_95_64,	/* src bits 95.. 64 */
+	IP6_SRC_127_96,	/* src bits 127..96 */
+
+	IP6_DST_31_0,	/* dst bits 0..  31 */
+	IP6_DST_63_32,	/* dst bits 63.. 32 */
+	IP6_DST_95_64,	/* dst bits 95.. 64 */
+	IP6_DST_127_96,	/* dst bits 127..96 */
+
+	TCP_SPORT,	/* 16-bit TCP sport */
+	TCP_DPORT,	/* 16-bit TCP dport */
+
+	UDP_SPORT,	/* 16-bit UDP sport */
+	UDP_DPORT,	/* 16-bit UDP dport */
+};
+
+struct ch_tc_pedit_fields {
+	u8 field;
+	u8 size;
+	u32 offset;
+};
+
+#define PEDIT_FIELDS(type, field, size, fs_field, offset) \
+	{ type## field, size, \
+		offsetof(struct ch_filter_specification, fs_field) + (offset) }
+
+#define PEDIT_ETH_DMAC_MASK		0xffff
+#define PEDIT_TCP_UDP_SPORT_MASK	0xffff
+#define PEDIT_ETH_DMAC_31_0		0x0
+#define PEDIT_ETH_DMAC_47_32_SMAC_15_0	0x4
+#define PEDIT_ETH_SMAC_47_16		0x8
+#define PEDIT_IP4_SRC			0xC
+#define PEDIT_IP4_DST			0x10
+#define PEDIT_IP6_SRC_31_0		0x8
+#define PEDIT_IP6_SRC_63_32		0xC
+#define PEDIT_IP6_SRC_95_64		0x10
+#define PEDIT_IP6_SRC_127_96		0x14
+#define PEDIT_IP6_DST_31_0		0x18
+#define PEDIT_IP6_DST_63_32		0x1C
+#define PEDIT_IP6_DST_95_64		0x20
+#define PEDIT_IP6_DST_127_96		0x24
+#define PEDIT_TCP_SPORT_DPORT		0x0
+#define PEDIT_UDP_SPORT_DPORT		0x0
+
+enum cxgb4_action_natmode_flags {
+	CXGB4_ACTION_NATMODE_NONE = 0,
+	CXGB4_ACTION_NATMODE_DIP = (1 << 0),
+	CXGB4_ACTION_NATMODE_SIP = (1 << 1),
+	CXGB4_ACTION_NATMODE_DPORT = (1 << 2),
+	CXGB4_ACTION_NATMODE_SPORT = (1 << 3),
+};
+
+/* TC PEDIT action to NATMODE translation entry */
+struct cxgb4_natmode_config {
+	enum chip_type chip;
+	u8 flags;
+	u8 natmode;
+};
+
+void cxgb4_process_flow_actions(struct net_device *in,
+				struct flow_action *actions,
+				struct ch_filter_specification *fs);
+int cxgb4_validate_flow_actions(struct net_device *dev,
+				struct flow_action *actions,
+				struct netlink_ext_ack *extack,
+				u8 matchall_filter);
+
+int cxgb4_tc_flower_replace(struct net_device *dev,
+			    struct flow_cls_offload *cls);
+int cxgb4_tc_flower_destroy(struct net_device *dev,
+			    struct flow_cls_offload *cls);
+int cxgb4_tc_flower_stats(struct net_device *dev,
+			  struct flow_cls_offload *cls);
+int cxgb4_flow_rule_replace(struct net_device *dev, struct flow_rule *rule,
+			    u32 tc_prio, struct netlink_ext_ack *extack,
+			    struct ch_filter_specification *fs, u32 *tid);
+int cxgb4_flow_rule_destroy(struct net_device *dev, u32 tc_prio,
+			    struct ch_filter_specification *fs, int tid);
+
+int cxgb4_init_tc_flower(struct adapter *adap);
+void cxgb4_cleanup_tc_flower(struct adapter *adap);
+#endif /* __CXGB4_TC_FLOWER_H */
diff --git a/drivers/net/ethernet/chelsio/cxgb4/cxgb4_tc_matchall.c b/drivers/net/ethernet/chelsio/cxgb4/cxgb4_tc_matchall.c
new file mode 100644
index 000000000000..1672d3afe5be
--- /dev/null
+++ b/drivers/net/ethernet/chelsio/cxgb4/cxgb4_tc_matchall.c
@@ -0,0 +1,571 @@
+// SPDX-License-Identifier: GPL-2.0-only
+/* Copyright (C) 2019 Chelsio Communications.  All rights reserved. */
+
+#include "cxgb4.h"
+#include "cxgb4_tc_matchall.h"
+#include "sched.h"
+#include "cxgb4_uld.h"
+#include "cxgb4_filter.h"
+#include "cxgb4_tc_flower.h"
+
+static int cxgb4_policer_validate(const struct flow_action *action,
+				  const struct flow_action_entry *act,
+				  struct netlink_ext_ack *extack)
+{
+	if (act->police.exceed.act_id != FLOW_ACTION_DROP) {
+		NL_SET_ERR_MSG_MOD(extack,
+				   "Offload not supported when exceed action is not drop");
+		return -EOPNOTSUPP;
+	}
+
+	if (act->police.notexceed.act_id != FLOW_ACTION_PIPE &&
+	    act->police.notexceed.act_id != FLOW_ACTION_ACCEPT) {
+		NL_SET_ERR_MSG_MOD(extack,
+				   "Offload not supported when conform action is not pipe or ok");
+		return -EOPNOTSUPP;
+	}
+
+	if (act->police.notexceed.act_id == FLOW_ACTION_ACCEPT &&
+	    !flow_action_is_last_entry(action, act)) {
+		NL_SET_ERR_MSG_MOD(extack,
+				   "Offload not supported when conform action is ok, but action is not last");
+		return -EOPNOTSUPP;
+	}
+
+	if (act->police.peakrate_bytes_ps ||
+	    act->police.avrate || act->police.overhead) {
+		NL_SET_ERR_MSG_MOD(extack,
+				   "Offload not supported when peakrate/avrate/overhead is configured");
+		return -EOPNOTSUPP;
+	}
+
+	if (act->police.rate_pkt_ps) {
+		NL_SET_ERR_MSG_MOD(extack,
+				   "QoS offload not support packets per second");
+		return -EOPNOTSUPP;
+	}
+
+	return 0;
+}
+
+static int cxgb4_matchall_egress_validate(struct net_device *dev,
+					  struct tc_cls_matchall_offload *cls)
+{
+	struct netlink_ext_ack *extack = cls->common.extack;
+	struct flow_action *actions = &cls->rule->action;
+	struct port_info *pi = netdev2pinfo(dev);
+	struct flow_action_entry *entry;
+	struct ch_sched_queue qe;
+	struct sched_class *e;
+	u64 max_link_rate;
+	u32 i, speed;
+	int ret;
+
+	if (!flow_action_has_entries(actions)) {
+		NL_SET_ERR_MSG_MOD(extack,
+				   "Egress MATCHALL offload needs at least 1 policing action");
+		return -EINVAL;
+	} else if (!flow_offload_has_one_action(actions)) {
+		NL_SET_ERR_MSG_MOD(extack,
+				   "Egress MATCHALL offload only supports 1 policing action");
+		return -EINVAL;
+	} else if (pi->tc_block_shared) {
+		NL_SET_ERR_MSG_MOD(extack,
+				   "Egress MATCHALL offload not supported with shared blocks");
+		return -EINVAL;
+	}
+
+	ret = t4_get_link_params(pi, NULL, &speed, NULL);
+	if (ret) {
+		NL_SET_ERR_MSG_MOD(extack,
+				   "Failed to get max speed supported by the link");
+		return -EINVAL;
+	}
+
+	/* Convert from Mbps to bps */
+	max_link_rate = (u64)speed * 1000 * 1000;
+
+	flow_action_for_each(i, entry, actions) {
+		switch (entry->id) {
+		case FLOW_ACTION_POLICE:
+			ret = cxgb4_policer_validate(actions, entry, extack);
+			if (ret)
+				return ret;
+
+			/* Convert bytes per second to bits per second */
+			if (entry->police.rate_bytes_ps * 8 > max_link_rate) {
+				NL_SET_ERR_MSG_MOD(extack,
+						   "Specified policing max rate is larger than underlying link speed");
+				return -ERANGE;
+			}
+			break;
+		default:
+			NL_SET_ERR_MSG_MOD(extack,
+					   "Only policing action supported with Egress MATCHALL offload");
+			return -EOPNOTSUPP;
+		}
+	}
+
+	for (i = 0; i < pi->nqsets; i++) {
+		memset(&qe, 0, sizeof(qe));
+		qe.queue = i;
+
+		e = cxgb4_sched_queue_lookup(dev, &qe);
+		if (e && e->info.u.params.level != SCHED_CLASS_LEVEL_CH_RL) {
+			NL_SET_ERR_MSG_MOD(extack,
+					   "Some queues are already bound to different class");
+			return -EBUSY;
+		}
+	}
+
+	return 0;
+}
+
+static int cxgb4_matchall_tc_bind_queues(struct net_device *dev, u32 tc)
+{
+	struct port_info *pi = netdev2pinfo(dev);
+	struct ch_sched_queue qe;
+	int ret;
+	u32 i;
+
+	for (i = 0; i < pi->nqsets; i++) {
+		qe.queue = i;
+		qe.class = tc;
+		ret = cxgb4_sched_class_bind(dev, &qe, SCHED_QUEUE);
+		if (ret)
+			goto out_free;
+	}
+
+	return 0;
+
+out_free:
+	while (i--) {
+		qe.queue = i;
+		qe.class = SCHED_CLS_NONE;
+		cxgb4_sched_class_unbind(dev, &qe, SCHED_QUEUE);
+	}
+
+	return ret;
+}
+
+static void cxgb4_matchall_tc_unbind_queues(struct net_device *dev)
+{
+	struct port_info *pi = netdev2pinfo(dev);
+	struct ch_sched_queue qe;
+	u32 i;
+
+	for (i = 0; i < pi->nqsets; i++) {
+		qe.queue = i;
+		qe.class = SCHED_CLS_NONE;
+		cxgb4_sched_class_unbind(dev, &qe, SCHED_QUEUE);
+	}
+}
+
+static int cxgb4_matchall_alloc_tc(struct net_device *dev,
+				   struct tc_cls_matchall_offload *cls)
+{
+	struct ch_sched_params p = {
+		.type = SCHED_CLASS_TYPE_PACKET,
+		.u.params.level = SCHED_CLASS_LEVEL_CH_RL,
+		.u.params.mode = SCHED_CLASS_MODE_CLASS,
+		.u.params.rateunit = SCHED_CLASS_RATEUNIT_BITS,
+		.u.params.ratemode = SCHED_CLASS_RATEMODE_ABS,
+		.u.params.class = SCHED_CLS_NONE,
+		.u.params.minrate = 0,
+		.u.params.weight = 0,
+		.u.params.pktsize = dev->mtu,
+	};
+	struct netlink_ext_ack *extack = cls->common.extack;
+	struct cxgb4_tc_port_matchall *tc_port_matchall;
+	struct port_info *pi = netdev2pinfo(dev);
+	struct adapter *adap = netdev2adap(dev);
+	struct flow_action_entry *entry;
+	struct sched_class *e;
+	int ret;
+	u32 i;
+
+	tc_port_matchall = &adap->tc_matchall->port_matchall[pi->port_id];
+
+	flow_action_for_each(i, entry, &cls->rule->action)
+		if (entry->id == FLOW_ACTION_POLICE)
+			break;
+
+	ret = cxgb4_policer_validate(&cls->rule->action, entry, extack);
+	if (ret)
+		return ret;
+
+	/* Convert from bytes per second to Kbps */
+	p.u.params.maxrate = div_u64(entry->police.rate_bytes_ps * 8, 1000);
+	p.u.params.channel = pi->tx_chan;
+	e = cxgb4_sched_class_alloc(dev, &p);
+	if (!e) {
+		NL_SET_ERR_MSG_MOD(extack,
+				   "No free traffic class available for policing action");
+		return -ENOMEM;
+	}
+
+	ret = cxgb4_matchall_tc_bind_queues(dev, e->idx);
+	if (ret) {
+		NL_SET_ERR_MSG_MOD(extack,
+				   "Could not bind queues to traffic class");
+		goto out_free;
+	}
+
+	tc_port_matchall->egress.hwtc = e->idx;
+	tc_port_matchall->egress.cookie = cls->cookie;
+	tc_port_matchall->egress.state = CXGB4_MATCHALL_STATE_ENABLED;
+	return 0;
+
+out_free:
+	cxgb4_sched_class_free(dev, e->idx);
+	return ret;
+}
+
+static void cxgb4_matchall_free_tc(struct net_device *dev)
+{
+	struct cxgb4_tc_port_matchall *tc_port_matchall;
+	struct port_info *pi = netdev2pinfo(dev);
+	struct adapter *adap = netdev2adap(dev);
+
+	tc_port_matchall = &adap->tc_matchall->port_matchall[pi->port_id];
+	cxgb4_matchall_tc_unbind_queues(dev);
+	cxgb4_sched_class_free(dev, tc_port_matchall->egress.hwtc);
+
+	tc_port_matchall->egress.hwtc = SCHED_CLS_NONE;
+	tc_port_matchall->egress.cookie = 0;
+	tc_port_matchall->egress.state = CXGB4_MATCHALL_STATE_DISABLED;
+}
+
+static int cxgb4_matchall_mirror_alloc(struct net_device *dev,
+				       struct tc_cls_matchall_offload *cls)
+{
+	struct netlink_ext_ack *extack = cls->common.extack;
+	struct cxgb4_tc_port_matchall *tc_port_matchall;
+	struct port_info *pi = netdev2pinfo(dev);
+	struct adapter *adap = netdev2adap(dev);
+	struct flow_action_entry *act;
+	int ret;
+	u32 i;
+
+	tc_port_matchall = &adap->tc_matchall->port_matchall[pi->port_id];
+	flow_action_for_each(i, act, &cls->rule->action) {
+		if (act->id == FLOW_ACTION_MIRRED) {
+			ret = cxgb4_port_mirror_alloc(dev);
+			if (ret) {
+				NL_SET_ERR_MSG_MOD(extack,
+						   "Couldn't allocate mirror");
+				return ret;
+			}
+
+			tc_port_matchall->ingress.viid_mirror = pi->viid_mirror;
+			break;
+		}
+	}
+
+	return 0;
+}
+
+static void cxgb4_matchall_mirror_free(struct net_device *dev)
+{
+	struct cxgb4_tc_port_matchall *tc_port_matchall;
+	struct port_info *pi = netdev2pinfo(dev);
+	struct adapter *adap = netdev2adap(dev);
+
+	tc_port_matchall = &adap->tc_matchall->port_matchall[pi->port_id];
+	if (!tc_port_matchall->ingress.viid_mirror)
+		return;
+
+	cxgb4_port_mirror_free(dev);
+	tc_port_matchall->ingress.viid_mirror = 0;
+}
+
+static int cxgb4_matchall_del_filter(struct net_device *dev, u8 filter_type)
+{
+	struct cxgb4_tc_port_matchall *tc_port_matchall;
+	struct port_info *pi = netdev2pinfo(dev);
+	struct adapter *adap = netdev2adap(dev);
+	int ret;
+
+	tc_port_matchall = &adap->tc_matchall->port_matchall[pi->port_id];
+	ret = cxgb4_del_filter(dev, tc_port_matchall->ingress.tid[filter_type],
+			       &tc_port_matchall->ingress.fs[filter_type]);
+	if (ret)
+		return ret;
+
+	tc_port_matchall->ingress.tid[filter_type] = 0;
+	return 0;
+}
+
+static int cxgb4_matchall_add_filter(struct net_device *dev,
+				     struct tc_cls_matchall_offload *cls,
+				     u8 filter_type)
+{
+	struct netlink_ext_ack *extack = cls->common.extack;
+	struct cxgb4_tc_port_matchall *tc_port_matchall;
+	struct port_info *pi = netdev2pinfo(dev);
+	struct adapter *adap = netdev2adap(dev);
+	struct ch_filter_specification *fs;
+	int ret, fidx;
+
+	/* Get a free filter entry TID, where we can insert this new
+	 * rule. Only insert rule if its prio doesn't conflict with
+	 * existing rules.
+	 */
+	fidx = cxgb4_get_free_ftid(dev, filter_type ? PF_INET6 : PF_INET,
+				   false, cls->common.prio);
+	if (fidx < 0) {
+		NL_SET_ERR_MSG_MOD(extack,
+				   "No free LETCAM index available");
+		return -ENOMEM;
+	}
+
+	tc_port_matchall = &adap->tc_matchall->port_matchall[pi->port_id];
+	fs = &tc_port_matchall->ingress.fs[filter_type];
+	memset(fs, 0, sizeof(*fs));
+
+	if (fidx < adap->tids.nhpftids)
+		fs->prio = 1;
+	fs->tc_prio = cls->common.prio;
+	fs->tc_cookie = cls->cookie;
+	fs->type = filter_type;
+	fs->hitcnts = 1;
+
+	fs->val.pfvf_vld = 1;
+	fs->val.pf = adap->pf;
+	fs->val.vf = pi->vin;
+
+	cxgb4_process_flow_actions(dev, &cls->rule->action, fs);
+
+	ret = cxgb4_set_filter(dev, fidx, fs);
+	if (ret)
+		return ret;
+
+	tc_port_matchall->ingress.tid[filter_type] = fidx;
+	return 0;
+}
+
+static int cxgb4_matchall_alloc_filter(struct net_device *dev,
+				       struct tc_cls_matchall_offload *cls)
+{
+	struct cxgb4_tc_port_matchall *tc_port_matchall;
+	struct port_info *pi = netdev2pinfo(dev);
+	struct adapter *adap = netdev2adap(dev);
+	int ret, i;
+
+	tc_port_matchall = &adap->tc_matchall->port_matchall[pi->port_id];
+
+	ret = cxgb4_matchall_mirror_alloc(dev, cls);
+	if (ret)
+		return ret;
+
+	for (i = 0; i < CXGB4_FILTER_TYPE_MAX; i++) {
+		ret = cxgb4_matchall_add_filter(dev, cls, i);
+		if (ret)
+			goto out_free;
+	}
+
+	tc_port_matchall->ingress.state = CXGB4_MATCHALL_STATE_ENABLED;
+	return 0;
+
+out_free:
+	while (i-- > 0)
+		cxgb4_matchall_del_filter(dev, i);
+
+	cxgb4_matchall_mirror_free(dev);
+	return ret;
+}
+
+static int cxgb4_matchall_free_filter(struct net_device *dev)
+{
+	struct cxgb4_tc_port_matchall *tc_port_matchall;
+	struct port_info *pi = netdev2pinfo(dev);
+	struct adapter *adap = netdev2adap(dev);
+	int ret;
+	u8 i;
+
+	tc_port_matchall = &adap->tc_matchall->port_matchall[pi->port_id];
+
+	for (i = 0; i < CXGB4_FILTER_TYPE_MAX; i++) {
+		ret = cxgb4_matchall_del_filter(dev, i);
+		if (ret)
+			return ret;
+	}
+
+	cxgb4_matchall_mirror_free(dev);
+
+	tc_port_matchall->ingress.packets = 0;
+	tc_port_matchall->ingress.bytes = 0;
+	tc_port_matchall->ingress.last_used = 0;
+	tc_port_matchall->ingress.state = CXGB4_MATCHALL_STATE_DISABLED;
+	return 0;
+}
+
+int cxgb4_tc_matchall_replace(struct net_device *dev,
+			      struct tc_cls_matchall_offload *cls_matchall,
+			      bool ingress)
+{
+	struct netlink_ext_ack *extack = cls_matchall->common.extack;
+	struct cxgb4_tc_port_matchall *tc_port_matchall;
+	struct port_info *pi = netdev2pinfo(dev);
+	struct adapter *adap = netdev2adap(dev);
+	int ret;
+
+	tc_port_matchall = &adap->tc_matchall->port_matchall[pi->port_id];
+	if (ingress) {
+		if (tc_port_matchall->ingress.state ==
+		    CXGB4_MATCHALL_STATE_ENABLED) {
+			NL_SET_ERR_MSG_MOD(extack,
+					   "Only 1 Ingress MATCHALL can be offloaded");
+			return -ENOMEM;
+		}
+
+		ret = cxgb4_validate_flow_actions(dev,
+						  &cls_matchall->rule->action,
+						  extack, 1);
+		if (ret)
+			return ret;
+
+		return cxgb4_matchall_alloc_filter(dev, cls_matchall);
+	}
+
+	if (tc_port_matchall->egress.state == CXGB4_MATCHALL_STATE_ENABLED) {
+		NL_SET_ERR_MSG_MOD(extack,
+				   "Only 1 Egress MATCHALL can be offloaded");
+		return -ENOMEM;
+	}
+
+	ret = cxgb4_matchall_egress_validate(dev, cls_matchall);
+	if (ret)
+		return ret;
+
+	return cxgb4_matchall_alloc_tc(dev, cls_matchall);
+}
+
+int cxgb4_tc_matchall_destroy(struct net_device *dev,
+			      struct tc_cls_matchall_offload *cls_matchall,
+			      bool ingress)
+{
+	struct cxgb4_tc_port_matchall *tc_port_matchall;
+	struct port_info *pi = netdev2pinfo(dev);
+	struct adapter *adap = netdev2adap(dev);
+
+	tc_port_matchall = &adap->tc_matchall->port_matchall[pi->port_id];
+	if (ingress) {
+		/* All the filter types of this matchall rule save the
+		 * same cookie. So, checking for the first one is
+		 * enough.
+		 */
+		if (cls_matchall->cookie !=
+		    tc_port_matchall->ingress.fs[0].tc_cookie)
+			return -ENOENT;
+
+		return cxgb4_matchall_free_filter(dev);
+	}
+
+	if (cls_matchall->cookie != tc_port_matchall->egress.cookie)
+		return -ENOENT;
+
+	cxgb4_matchall_free_tc(dev);
+	return 0;
+}
+
+int cxgb4_tc_matchall_stats(struct net_device *dev,
+			    struct tc_cls_matchall_offload *cls_matchall)
+{
+	u64 tmp_packets, tmp_bytes, packets = 0, bytes = 0;
+	struct cxgb4_tc_port_matchall *tc_port_matchall;
+	struct cxgb4_matchall_ingress_entry *ingress;
+	struct port_info *pi = netdev2pinfo(dev);
+	struct adapter *adap = netdev2adap(dev);
+	int ret;
+	u8 i;
+
+	tc_port_matchall = &adap->tc_matchall->port_matchall[pi->port_id];
+	if (tc_port_matchall->ingress.state == CXGB4_MATCHALL_STATE_DISABLED)
+		return -ENOENT;
+
+	ingress = &tc_port_matchall->ingress;
+	for (i = 0; i < CXGB4_FILTER_TYPE_MAX; i++) {
+		ret = cxgb4_get_filter_counters(dev, ingress->tid[i],
+						&tmp_packets, &tmp_bytes,
+						ingress->fs[i].hash);
+		if (ret)
+			return ret;
+
+		packets += tmp_packets;
+		bytes += tmp_bytes;
+	}
+
+	if (tc_port_matchall->ingress.packets != packets) {
+		flow_stats_update(&cls_matchall->stats,
+				  bytes - tc_port_matchall->ingress.bytes,
+				  packets - tc_port_matchall->ingress.packets,
+				  0, tc_port_matchall->ingress.last_used,
+				  FLOW_ACTION_HW_STATS_IMMEDIATE);
+
+		tc_port_matchall->ingress.packets = packets;
+		tc_port_matchall->ingress.bytes = bytes;
+		tc_port_matchall->ingress.last_used = jiffies;
+	}
+
+	return 0;
+}
+
+static void cxgb4_matchall_disable_offload(struct net_device *dev)
+{
+	struct cxgb4_tc_port_matchall *tc_port_matchall;
+	struct port_info *pi = netdev2pinfo(dev);
+	struct adapter *adap = netdev2adap(dev);
+
+	tc_port_matchall = &adap->tc_matchall->port_matchall[pi->port_id];
+	if (tc_port_matchall->egress.state == CXGB4_MATCHALL_STATE_ENABLED)
+		cxgb4_matchall_free_tc(dev);
+
+	if (tc_port_matchall->ingress.state == CXGB4_MATCHALL_STATE_ENABLED)
+		cxgb4_matchall_free_filter(dev);
+}
+
+int cxgb4_init_tc_matchall(struct adapter *adap)
+{
+	struct cxgb4_tc_port_matchall *tc_port_matchall;
+	struct cxgb4_tc_matchall *tc_matchall;
+	int ret;
+
+	tc_matchall = kzalloc(sizeof(*tc_matchall), GFP_KERNEL);
+	if (!tc_matchall)
+		return -ENOMEM;
+
+	tc_port_matchall = kcalloc(adap->params.nports,
+				   sizeof(*tc_port_matchall),
+				   GFP_KERNEL);
+	if (!tc_port_matchall) {
+		ret = -ENOMEM;
+		goto out_free_matchall;
+	}
+
+	tc_matchall->port_matchall = tc_port_matchall;
+	adap->tc_matchall = tc_matchall;
+	return 0;
+
+out_free_matchall:
+	kfree(tc_matchall);
+	return ret;
+}
+
+void cxgb4_cleanup_tc_matchall(struct adapter *adap)
+{
+	u8 i;
+
+	if (adap->tc_matchall) {
+		if (adap->tc_matchall->port_matchall) {
+			for (i = 0; i < adap->params.nports; i++) {
+				struct net_device *dev = adap->port[i];
+
+				if (dev)
+					cxgb4_matchall_disable_offload(dev);
+			}
+			kfree(adap->tc_matchall->port_matchall);
+		}
+		kfree(adap->tc_matchall);
+	}
+}
diff --git a/drivers/net/ethernet/chelsio/cxgb4/cxgb4_tc_matchall.h b/drivers/net/ethernet/chelsio/cxgb4/cxgb4_tc_matchall.h
new file mode 100644
index 000000000000..fe7ec423a4c9
--- /dev/null
+++ b/drivers/net/ethernet/chelsio/cxgb4/cxgb4_tc_matchall.h
@@ -0,0 +1,51 @@
+/* SPDX-License-Identifier: GPL-2.0-only */
+/* Copyright (C) 2019 Chelsio Communications.  All rights reserved. */
+
+#ifndef __CXGB4_TC_MATCHALL_H__
+#define __CXGB4_TC_MATCHALL_H__
+
+#include <net/pkt_cls.h>
+
+enum cxgb4_matchall_state {
+	CXGB4_MATCHALL_STATE_DISABLED = 0,
+	CXGB4_MATCHALL_STATE_ENABLED,
+};
+
+struct cxgb4_matchall_egress_entry {
+	enum cxgb4_matchall_state state; /* Current MATCHALL offload state */
+	u8 hwtc; /* Traffic class bound to port */
+	u64 cookie; /* Used to identify the MATCHALL rule offloaded */
+};
+
+struct cxgb4_matchall_ingress_entry {
+	enum cxgb4_matchall_state state; /* Current MATCHALL offload state */
+	u32 tid[CXGB4_FILTER_TYPE_MAX]; /* Index to hardware filter entries */
+	/* Filter entries */
+	struct ch_filter_specification fs[CXGB4_FILTER_TYPE_MAX];
+	u16 viid_mirror; /* Identifier for allocated Mirror VI */
+	u64 bytes; /* # of bytes hitting the filter */
+	u64 packets; /* # of packets hitting the filter */
+	u64 last_used; /* Last updated jiffies time */
+};
+
+struct cxgb4_tc_port_matchall {
+	struct cxgb4_matchall_egress_entry egress; /* Egress offload info */
+	struct cxgb4_matchall_ingress_entry ingress; /* Ingress offload info */
+};
+
+struct cxgb4_tc_matchall {
+	struct cxgb4_tc_port_matchall *port_matchall; /* Per port entry */
+};
+
+int cxgb4_tc_matchall_replace(struct net_device *dev,
+			      struct tc_cls_matchall_offload *cls_matchall,
+			      bool ingress);
+int cxgb4_tc_matchall_destroy(struct net_device *dev,
+			      struct tc_cls_matchall_offload *cls_matchall,
+			      bool ingress);
+int cxgb4_tc_matchall_stats(struct net_device *dev,
+			    struct tc_cls_matchall_offload *cls_matchall);
+
+int cxgb4_init_tc_matchall(struct adapter *adap);
+void cxgb4_cleanup_tc_matchall(struct adapter *adap);
+#endif /* __CXGB4_TC_MATCHALL_H__ */
diff --git a/drivers/net/ethernet/chelsio/cxgb4/cxgb4_tc_mqprio.c b/drivers/net/ethernet/chelsio/cxgb4/cxgb4_tc_mqprio.c
new file mode 100644
index 000000000000..338b04f339b3
--- /dev/null
+++ b/drivers/net/ethernet/chelsio/cxgb4/cxgb4_tc_mqprio.c
@@ -0,0 +1,722 @@
+// SPDX-License-Identifier: GPL-2.0-only
+/* Copyright (C) 2019 Chelsio Communications.  All rights reserved. */
+
+#include "cxgb4.h"
+#include "cxgb4_tc_mqprio.h"
+#include "sched.h"
+
+static int cxgb4_mqprio_validate(struct net_device *dev,
+				 struct tc_mqprio_qopt_offload *mqprio)
+{
+	u64 min_rate = 0, max_rate = 0, max_link_rate;
+	struct port_info *pi = netdev2pinfo(dev);
+	struct adapter *adap = netdev2adap(dev);
+	u32 speed, qcount = 0, qoffset = 0;
+	u32 start_a, start_b, end_a, end_b;
+	int ret;
+	u8 i, j;
+
+	if (!mqprio->qopt.num_tc)
+		return 0;
+
+	if (mqprio->qopt.hw != TC_MQPRIO_HW_OFFLOAD_TCS) {
+		netdev_err(dev, "Only full TC hardware offload is supported\n");
+		return -EINVAL;
+	} else if (mqprio->mode != TC_MQPRIO_MODE_CHANNEL) {
+		netdev_err(dev, "Only channel mode offload is supported\n");
+		return -EINVAL;
+	} else if (mqprio->shaper != TC_MQPRIO_SHAPER_BW_RATE) {
+		netdev_err(dev,	"Only bandwidth rate shaper supported\n");
+		return -EINVAL;
+	} else if (mqprio->qopt.num_tc > adap->params.nsched_cls) {
+		netdev_err(dev,
+			   "Only %u traffic classes supported by hardware\n",
+			   adap->params.nsched_cls);
+		return -ERANGE;
+	}
+
+	ret = t4_get_link_params(pi, NULL, &speed, NULL);
+	if (ret) {
+		netdev_err(dev, "Failed to get link speed, ret: %d\n", ret);
+		return -EINVAL;
+	}
+
+	/* Convert from Mbps to bps */
+	max_link_rate = (u64)speed * 1000 * 1000;
+
+	for (i = 0; i < mqprio->qopt.num_tc; i++) {
+		qoffset = max_t(u16, mqprio->qopt.offset[i], qoffset);
+		qcount += mqprio->qopt.count[i];
+
+		start_a = mqprio->qopt.offset[i];
+		end_a = start_a + mqprio->qopt.count[i] - 1;
+		for (j = i + 1; j < mqprio->qopt.num_tc; j++) {
+			start_b = mqprio->qopt.offset[j];
+			end_b = start_b + mqprio->qopt.count[j] - 1;
+
+			/* If queue count is 0, then the traffic
+			 * belonging to this class will not use
+			 * ETHOFLD queues. So, no need to validate
+			 * further.
+			 */
+			if (!mqprio->qopt.count[i])
+				break;
+
+			if (!mqprio->qopt.count[j])
+				continue;
+
+			if (max_t(u32, start_a, start_b) <=
+			    min_t(u32, end_a, end_b)) {
+				netdev_err(dev,
+					   "Queues can't overlap across tc\n");
+				return -EINVAL;
+			}
+		}
+
+		/* Convert byte per second to bits per second */
+		min_rate += (mqprio->min_rate[i] * 8);
+		max_rate += (mqprio->max_rate[i] * 8);
+	}
+
+	if (qoffset >= adap->tids.neotids || qcount > adap->tids.neotids)
+		return -ENOMEM;
+
+	if (min_rate > max_link_rate || max_rate > max_link_rate) {
+		netdev_err(dev,
+			   "Total Min/Max (%llu/%llu) Rate > supported (%llu)\n",
+			   min_rate, max_rate, max_link_rate);
+		return -EINVAL;
+	}
+
+	return 0;
+}
+
+static int cxgb4_init_eosw_txq(struct net_device *dev,
+			       struct sge_eosw_txq *eosw_txq,
+			       u32 eotid, u32 hwqid)
+{
+	struct adapter *adap = netdev2adap(dev);
+	struct tx_sw_desc *ring;
+
+	memset(eosw_txq, 0, sizeof(*eosw_txq));
+
+	ring = kcalloc(CXGB4_EOSW_TXQ_DEFAULT_DESC_NUM,
+		       sizeof(*ring), GFP_KERNEL);
+	if (!ring)
+		return -ENOMEM;
+
+	eosw_txq->desc = ring;
+	eosw_txq->ndesc = CXGB4_EOSW_TXQ_DEFAULT_DESC_NUM;
+	spin_lock_init(&eosw_txq->lock);
+	eosw_txq->state = CXGB4_EO_STATE_CLOSED;
+	eosw_txq->eotid = eotid;
+	eosw_txq->hwtid = adap->tids.eotid_base + eosw_txq->eotid;
+	eosw_txq->cred = adap->params.ofldq_wr_cred;
+	eosw_txq->hwqid = hwqid;
+	eosw_txq->netdev = dev;
+	tasklet_setup(&eosw_txq->qresume_tsk, cxgb4_ethofld_restart);
+	return 0;
+}
+
+static void cxgb4_clean_eosw_txq(struct net_device *dev,
+				 struct sge_eosw_txq *eosw_txq)
+{
+	struct adapter *adap = netdev2adap(dev);
+
+	cxgb4_eosw_txq_free_desc(adap, eosw_txq, eosw_txq->ndesc);
+	eosw_txq->pidx = 0;
+	eosw_txq->last_pidx = 0;
+	eosw_txq->cidx = 0;
+	eosw_txq->last_cidx = 0;
+	eosw_txq->flowc_idx = 0;
+	eosw_txq->inuse = 0;
+	eosw_txq->cred = adap->params.ofldq_wr_cred;
+	eosw_txq->ncompl = 0;
+	eosw_txq->last_compl = 0;
+	eosw_txq->state = CXGB4_EO_STATE_CLOSED;
+}
+
+static void cxgb4_free_eosw_txq(struct net_device *dev,
+				struct sge_eosw_txq *eosw_txq)
+{
+	spin_lock_bh(&eosw_txq->lock);
+	cxgb4_clean_eosw_txq(dev, eosw_txq);
+	kfree(eosw_txq->desc);
+	spin_unlock_bh(&eosw_txq->lock);
+	tasklet_kill(&eosw_txq->qresume_tsk);
+}
+
+static int cxgb4_mqprio_alloc_hw_resources(struct net_device *dev)
+{
+	struct port_info *pi = netdev2pinfo(dev);
+	struct adapter *adap = netdev2adap(dev);
+	struct sge_ofld_rxq *eorxq;
+	struct sge_eohw_txq *eotxq;
+	int ret, msix = 0;
+	u32 i;
+
+	/* Allocate ETHOFLD hardware queue structures if not done already */
+	if (!refcount_read(&adap->tc_mqprio->refcnt)) {
+		adap->sge.eohw_rxq = kcalloc(adap->sge.eoqsets,
+					     sizeof(struct sge_ofld_rxq),
+					     GFP_KERNEL);
+		if (!adap->sge.eohw_rxq)
+			return -ENOMEM;
+
+		adap->sge.eohw_txq = kcalloc(adap->sge.eoqsets,
+					     sizeof(struct sge_eohw_txq),
+					     GFP_KERNEL);
+		if (!adap->sge.eohw_txq) {
+			kfree(adap->sge.eohw_rxq);
+			return -ENOMEM;
+		}
+
+		refcount_set(&adap->tc_mqprio->refcnt, 1);
+	} else {
+		refcount_inc(&adap->tc_mqprio->refcnt);
+	}
+
+	if (!(adap->flags & CXGB4_USING_MSIX))
+		msix = -((int)adap->sge.intrq.abs_id + 1);
+
+	for (i = 0; i < pi->nqsets; i++) {
+		eorxq = &adap->sge.eohw_rxq[pi->first_qset + i];
+		eotxq = &adap->sge.eohw_txq[pi->first_qset + i];
+
+		/* Allocate Rxqs for receiving ETHOFLD Tx completions */
+		if (msix >= 0) {
+			msix = cxgb4_get_msix_idx_from_bmap(adap);
+			if (msix < 0) {
+				ret = msix;
+				goto out_free_queues;
+			}
+
+			eorxq->msix = &adap->msix_info[msix];
+			snprintf(eorxq->msix->desc,
+				 sizeof(eorxq->msix->desc),
+				 "%s-eorxq%d", dev->name, i);
+		}
+
+		init_rspq(adap, &eorxq->rspq,
+			  CXGB4_EOHW_RXQ_DEFAULT_INTR_USEC,
+			  CXGB4_EOHW_RXQ_DEFAULT_PKT_CNT,
+			  CXGB4_EOHW_RXQ_DEFAULT_DESC_NUM,
+			  CXGB4_EOHW_RXQ_DEFAULT_DESC_SIZE);
+
+		eorxq->fl.size = CXGB4_EOHW_FLQ_DEFAULT_DESC_NUM;
+
+		ret = t4_sge_alloc_rxq(adap, &eorxq->rspq, false,
+				       dev, msix, &eorxq->fl,
+				       cxgb4_ethofld_rx_handler,
+				       NULL, 0);
+		if (ret)
+			goto out_free_queues;
+
+		/* Allocate ETHOFLD hardware Txqs */
+		eotxq->q.size = CXGB4_EOHW_TXQ_DEFAULT_DESC_NUM;
+		ret = t4_sge_alloc_ethofld_txq(adap, eotxq, dev,
+					       eorxq->rspq.cntxt_id);
+		if (ret)
+			goto out_free_queues;
+
+		/* Allocate IRQs, set IRQ affinity, and start Rx */
+		if (adap->flags & CXGB4_USING_MSIX) {
+			ret = request_irq(eorxq->msix->vec, t4_sge_intr_msix, 0,
+					  eorxq->msix->desc, &eorxq->rspq);
+			if (ret)
+				goto out_free_msix;
+
+			cxgb4_set_msix_aff(adap, eorxq->msix->vec,
+					   &eorxq->msix->aff_mask, i);
+		}
+
+		if (adap->flags & CXGB4_FULL_INIT_DONE)
+			cxgb4_enable_rx(adap, &eorxq->rspq);
+	}
+
+	return 0;
+
+out_free_msix:
+	while (i-- > 0) {
+		eorxq = &adap->sge.eohw_rxq[pi->first_qset + i];
+
+		if (adap->flags & CXGB4_FULL_INIT_DONE)
+			cxgb4_quiesce_rx(&eorxq->rspq);
+
+		if (adap->flags & CXGB4_USING_MSIX) {
+			cxgb4_clear_msix_aff(eorxq->msix->vec,
+					     eorxq->msix->aff_mask);
+			free_irq(eorxq->msix->vec, &eorxq->rspq);
+		}
+	}
+
+out_free_queues:
+	for (i = 0; i < pi->nqsets; i++) {
+		eorxq = &adap->sge.eohw_rxq[pi->first_qset + i];
+		eotxq = &adap->sge.eohw_txq[pi->first_qset + i];
+
+		if (eorxq->rspq.desc)
+			free_rspq_fl(adap, &eorxq->rspq, &eorxq->fl);
+		if (eorxq->msix)
+			cxgb4_free_msix_idx_in_bmap(adap, eorxq->msix->idx);
+		t4_sge_free_ethofld_txq(adap, eotxq);
+	}
+
+	if (refcount_dec_and_test(&adap->tc_mqprio->refcnt)) {
+		kfree(adap->sge.eohw_txq);
+		kfree(adap->sge.eohw_rxq);
+	}
+	return ret;
+}
+
+static void cxgb4_mqprio_free_hw_resources(struct net_device *dev)
+{
+	struct port_info *pi = netdev2pinfo(dev);
+	struct adapter *adap = netdev2adap(dev);
+	struct sge_ofld_rxq *eorxq;
+	struct sge_eohw_txq *eotxq;
+	u32 i;
+
+	/* Return if no ETHOFLD structures have been allocated yet */
+	if (!refcount_read(&adap->tc_mqprio->refcnt))
+		return;
+
+	/* Return if no hardware queues have been allocated */
+	if (!adap->sge.eohw_rxq[pi->first_qset].rspq.desc)
+		return;
+
+	for (i = 0; i < pi->nqsets; i++) {
+		eorxq = &adap->sge.eohw_rxq[pi->first_qset + i];
+		eotxq = &adap->sge.eohw_txq[pi->first_qset + i];
+
+		/* Device removal path will already disable NAPI
+		 * before unregistering netdevice. So, only disable
+		 * NAPI if we're not in device removal path
+		 */
+		if (!(adap->flags & CXGB4_SHUTTING_DOWN))
+			cxgb4_quiesce_rx(&eorxq->rspq);
+
+		if (adap->flags & CXGB4_USING_MSIX) {
+			cxgb4_clear_msix_aff(eorxq->msix->vec,
+					     eorxq->msix->aff_mask);
+			free_irq(eorxq->msix->vec, &eorxq->rspq);
+			cxgb4_free_msix_idx_in_bmap(adap, eorxq->msix->idx);
+		}
+
+		free_rspq_fl(adap, &eorxq->rspq, &eorxq->fl);
+		t4_sge_free_ethofld_txq(adap, eotxq);
+	}
+
+	/* Free up ETHOFLD structures if there are no users */
+	if (refcount_dec_and_test(&adap->tc_mqprio->refcnt)) {
+		kfree(adap->sge.eohw_txq);
+		kfree(adap->sge.eohw_rxq);
+	}
+}
+
+static int cxgb4_mqprio_alloc_tc(struct net_device *dev,
+				 struct tc_mqprio_qopt_offload *mqprio)
+{
+	struct ch_sched_params p = {
+		.type = SCHED_CLASS_TYPE_PACKET,
+		.u.params.level = SCHED_CLASS_LEVEL_CL_RL,
+		.u.params.mode = SCHED_CLASS_MODE_FLOW,
+		.u.params.rateunit = SCHED_CLASS_RATEUNIT_BITS,
+		.u.params.ratemode = SCHED_CLASS_RATEMODE_ABS,
+		.u.params.class = SCHED_CLS_NONE,
+		.u.params.weight = 0,
+		.u.params.pktsize = dev->mtu,
+	};
+	struct cxgb4_tc_port_mqprio *tc_port_mqprio;
+	struct port_info *pi = netdev2pinfo(dev);
+	struct adapter *adap = netdev2adap(dev);
+	struct sched_class *e;
+	int ret;
+	u8 i;
+
+	tc_port_mqprio = &adap->tc_mqprio->port_mqprio[pi->port_id];
+	p.u.params.channel = pi->tx_chan;
+	for (i = 0; i < mqprio->qopt.num_tc; i++) {
+		/* Convert from bytes per second to Kbps */
+		p.u.params.minrate = div_u64(mqprio->min_rate[i] * 8, 1000);
+		p.u.params.maxrate = div_u64(mqprio->max_rate[i] * 8, 1000);
+
+		/* Request larger burst buffer for smaller MTU, so
+		 * that hardware can work on more data per burst
+		 * cycle.
+		 */
+		if (dev->mtu <= ETH_DATA_LEN)
+			p.u.params.burstsize = 8 * dev->mtu;
+
+		e = cxgb4_sched_class_alloc(dev, &p);
+		if (!e) {
+			ret = -ENOMEM;
+			goto out_err;
+		}
+
+		tc_port_mqprio->tc_hwtc_map[i] = e->idx;
+	}
+
+	return 0;
+
+out_err:
+	while (i--)
+		cxgb4_sched_class_free(dev, tc_port_mqprio->tc_hwtc_map[i]);
+
+	return ret;
+}
+
+static void cxgb4_mqprio_free_tc(struct net_device *dev)
+{
+	struct cxgb4_tc_port_mqprio *tc_port_mqprio;
+	struct port_info *pi = netdev2pinfo(dev);
+	struct adapter *adap = netdev2adap(dev);
+	u8 i;
+
+	tc_port_mqprio = &adap->tc_mqprio->port_mqprio[pi->port_id];
+	for (i = 0; i < tc_port_mqprio->mqprio.qopt.num_tc; i++)
+		cxgb4_sched_class_free(dev, tc_port_mqprio->tc_hwtc_map[i]);
+}
+
+static int cxgb4_mqprio_class_bind(struct net_device *dev,
+				   struct sge_eosw_txq *eosw_txq,
+				   u8 tc)
+{
+	struct ch_sched_flowc fe;
+	int ret;
+
+	init_completion(&eosw_txq->completion);
+
+	fe.tid = eosw_txq->eotid;
+	fe.class = tc;
+
+	ret = cxgb4_sched_class_bind(dev, &fe, SCHED_FLOWC);
+	if (ret)
+		return ret;
+
+	ret = wait_for_completion_timeout(&eosw_txq->completion,
+					  CXGB4_FLOWC_WAIT_TIMEOUT);
+	if (!ret)
+		return -ETIMEDOUT;
+
+	return 0;
+}
+
+static void cxgb4_mqprio_class_unbind(struct net_device *dev,
+				      struct sge_eosw_txq *eosw_txq,
+				      u8 tc)
+{
+	struct adapter *adap = netdev2adap(dev);
+	struct ch_sched_flowc fe;
+
+	/* If we're shutting down, interrupts are disabled and no completions
+	 * come back. So, skip waiting for completions in this scenario.
+	 */
+	if (!(adap->flags & CXGB4_SHUTTING_DOWN))
+		init_completion(&eosw_txq->completion);
+
+	fe.tid = eosw_txq->eotid;
+	fe.class = tc;
+	cxgb4_sched_class_unbind(dev, &fe, SCHED_FLOWC);
+
+	if (!(adap->flags & CXGB4_SHUTTING_DOWN))
+		wait_for_completion_timeout(&eosw_txq->completion,
+					    CXGB4_FLOWC_WAIT_TIMEOUT);
+}
+
+static int cxgb4_mqprio_enable_offload(struct net_device *dev,
+				       struct tc_mqprio_qopt_offload *mqprio)
+{
+	struct cxgb4_tc_port_mqprio *tc_port_mqprio;
+	u32 qoffset, qcount, tot_qcount, qid, hwqid;
+	struct port_info *pi = netdev2pinfo(dev);
+	struct adapter *adap = netdev2adap(dev);
+	struct sge_eosw_txq *eosw_txq;
+	int eotid, ret;
+	u16 i, j;
+	u8 hwtc;
+
+	ret = cxgb4_mqprio_alloc_hw_resources(dev);
+	if (ret)
+		return -ENOMEM;
+
+	tc_port_mqprio = &adap->tc_mqprio->port_mqprio[pi->port_id];
+	for (i = 0; i < mqprio->qopt.num_tc; i++) {
+		qoffset = mqprio->qopt.offset[i];
+		qcount = mqprio->qopt.count[i];
+		for (j = 0; j < qcount; j++) {
+			eotid = cxgb4_get_free_eotid(&adap->tids);
+			if (eotid < 0) {
+				ret = -ENOMEM;
+				goto out_free_eotids;
+			}
+
+			qid = qoffset + j;
+			hwqid = pi->first_qset + (eotid % pi->nqsets);
+			eosw_txq = &tc_port_mqprio->eosw_txq[qid];
+			ret = cxgb4_init_eosw_txq(dev, eosw_txq,
+						  eotid, hwqid);
+			if (ret)
+				goto out_free_eotids;
+
+			cxgb4_alloc_eotid(&adap->tids, eotid, eosw_txq);
+
+			hwtc = tc_port_mqprio->tc_hwtc_map[i];
+			ret = cxgb4_mqprio_class_bind(dev, eosw_txq, hwtc);
+			if (ret)
+				goto out_free_eotids;
+		}
+	}
+
+	memcpy(&tc_port_mqprio->mqprio, mqprio,
+	       sizeof(struct tc_mqprio_qopt_offload));
+
+	/* Inform the stack about the configured tc params.
+	 *
+	 * Set the correct queue map. If no queue count has been
+	 * specified, then send the traffic through default NIC
+	 * queues; instead of ETHOFLD queues.
+	 */
+	ret = netdev_set_num_tc(dev, mqprio->qopt.num_tc);
+	if (ret)
+		goto out_free_eotids;
+
+	tot_qcount = pi->nqsets;
+	for (i = 0; i < mqprio->qopt.num_tc; i++) {
+		qcount = mqprio->qopt.count[i];
+		if (qcount) {
+			qoffset = mqprio->qopt.offset[i] + pi->nqsets;
+		} else {
+			qcount = pi->nqsets;
+			qoffset = 0;
+		}
+
+		ret = netdev_set_tc_queue(dev, i, qcount, qoffset);
+		if (ret)
+			goto out_reset_tc;
+
+		tot_qcount += mqprio->qopt.count[i];
+	}
+
+	ret = netif_set_real_num_tx_queues(dev, tot_qcount);
+	if (ret)
+		goto out_reset_tc;
+
+	tc_port_mqprio->state = CXGB4_MQPRIO_STATE_ACTIVE;
+	return 0;
+
+out_reset_tc:
+	netdev_reset_tc(dev);
+	i = mqprio->qopt.num_tc;
+
+out_free_eotids:
+	while (i-- > 0) {
+		qoffset = mqprio->qopt.offset[i];
+		qcount = mqprio->qopt.count[i];
+		for (j = 0; j < qcount; j++) {
+			eosw_txq = &tc_port_mqprio->eosw_txq[qoffset + j];
+
+			hwtc = tc_port_mqprio->tc_hwtc_map[i];
+			cxgb4_mqprio_class_unbind(dev, eosw_txq, hwtc);
+
+			cxgb4_free_eotid(&adap->tids, eosw_txq->eotid);
+			cxgb4_free_eosw_txq(dev, eosw_txq);
+		}
+	}
+
+	cxgb4_mqprio_free_hw_resources(dev);
+	return ret;
+}
+
+static void cxgb4_mqprio_disable_offload(struct net_device *dev)
+{
+	struct cxgb4_tc_port_mqprio *tc_port_mqprio;
+	struct port_info *pi = netdev2pinfo(dev);
+	struct adapter *adap = netdev2adap(dev);
+	struct sge_eosw_txq *eosw_txq;
+	u32 qoffset, qcount;
+	u16 i, j;
+	u8 hwtc;
+
+	tc_port_mqprio = &adap->tc_mqprio->port_mqprio[pi->port_id];
+	if (tc_port_mqprio->state != CXGB4_MQPRIO_STATE_ACTIVE)
+		return;
+
+	netdev_reset_tc(dev);
+	netif_set_real_num_tx_queues(dev, pi->nqsets);
+
+	for (i = 0; i < tc_port_mqprio->mqprio.qopt.num_tc; i++) {
+		qoffset = tc_port_mqprio->mqprio.qopt.offset[i];
+		qcount = tc_port_mqprio->mqprio.qopt.count[i];
+		for (j = 0; j < qcount; j++) {
+			eosw_txq = &tc_port_mqprio->eosw_txq[qoffset + j];
+
+			hwtc = tc_port_mqprio->tc_hwtc_map[i];
+			cxgb4_mqprio_class_unbind(dev, eosw_txq, hwtc);
+
+			cxgb4_free_eotid(&adap->tids, eosw_txq->eotid);
+			cxgb4_free_eosw_txq(dev, eosw_txq);
+		}
+	}
+
+	cxgb4_mqprio_free_hw_resources(dev);
+
+	/* Free up the traffic classes */
+	cxgb4_mqprio_free_tc(dev);
+
+	memset(&tc_port_mqprio->mqprio, 0,
+	       sizeof(struct tc_mqprio_qopt_offload));
+
+	tc_port_mqprio->state = CXGB4_MQPRIO_STATE_DISABLED;
+}
+
+int cxgb4_setup_tc_mqprio(struct net_device *dev,
+			  struct tc_mqprio_qopt_offload *mqprio)
+{
+	struct adapter *adap = netdev2adap(dev);
+	bool needs_bring_up = false;
+	int ret;
+
+	ret = cxgb4_mqprio_validate(dev, mqprio);
+	if (ret)
+		return ret;
+
+	mutex_lock(&adap->tc_mqprio->mqprio_mutex);
+
+	/* To configure tc params, the current allocated EOTIDs must
+	 * be freed up. However, they can't be freed up if there's
+	 * traffic running on the interface. So, ensure interface is
+	 * down before configuring tc params.
+	 */
+	if (netif_running(dev)) {
+		netif_tx_stop_all_queues(dev);
+		netif_carrier_off(dev);
+		needs_bring_up = true;
+	}
+
+	cxgb4_mqprio_disable_offload(dev);
+
+	/* If requested for clear, then just return since resources are
+	 * already freed up by now.
+	 */
+	if (!mqprio->qopt.num_tc)
+		goto out;
+
+	/* Allocate free available traffic classes and configure
+	 * their rate parameters.
+	 */
+	ret = cxgb4_mqprio_alloc_tc(dev, mqprio);
+	if (ret)
+		goto out;
+
+	ret = cxgb4_mqprio_enable_offload(dev, mqprio);
+	if (ret) {
+		cxgb4_mqprio_free_tc(dev);
+		goto out;
+	}
+
+out:
+	if (needs_bring_up) {
+		netif_tx_start_all_queues(dev);
+		netif_carrier_on(dev);
+	}
+
+	mutex_unlock(&adap->tc_mqprio->mqprio_mutex);
+	return ret;
+}
+
+void cxgb4_mqprio_stop_offload(struct adapter *adap)
+{
+	struct cxgb4_tc_port_mqprio *tc_port_mqprio;
+	struct net_device *dev;
+	u8 i;
+
+	if (!adap->tc_mqprio || !adap->tc_mqprio->port_mqprio)
+		return;
+
+	mutex_lock(&adap->tc_mqprio->mqprio_mutex);
+	for_each_port(adap, i) {
+		dev = adap->port[i];
+		if (!dev)
+			continue;
+
+		tc_port_mqprio = &adap->tc_mqprio->port_mqprio[i];
+		if (!tc_port_mqprio->mqprio.qopt.num_tc)
+			continue;
+
+		cxgb4_mqprio_disable_offload(dev);
+	}
+	mutex_unlock(&adap->tc_mqprio->mqprio_mutex);
+}
+
+int cxgb4_init_tc_mqprio(struct adapter *adap)
+{
+	struct cxgb4_tc_port_mqprio *tc_port_mqprio, *port_mqprio;
+	struct cxgb4_tc_mqprio *tc_mqprio;
+	struct sge_eosw_txq *eosw_txq;
+	int ret = 0;
+	u8 i;
+
+	tc_mqprio = kzalloc(sizeof(*tc_mqprio), GFP_KERNEL);
+	if (!tc_mqprio)
+		return -ENOMEM;
+
+	tc_port_mqprio = kcalloc(adap->params.nports, sizeof(*tc_port_mqprio),
+				 GFP_KERNEL);
+	if (!tc_port_mqprio) {
+		ret = -ENOMEM;
+		goto out_free_mqprio;
+	}
+
+	mutex_init(&tc_mqprio->mqprio_mutex);
+
+	tc_mqprio->port_mqprio = tc_port_mqprio;
+	for (i = 0; i < adap->params.nports; i++) {
+		port_mqprio = &tc_mqprio->port_mqprio[i];
+		eosw_txq = kcalloc(adap->tids.neotids, sizeof(*eosw_txq),
+				   GFP_KERNEL);
+		if (!eosw_txq) {
+			ret = -ENOMEM;
+			goto out_free_ports;
+		}
+		port_mqprio->eosw_txq = eosw_txq;
+	}
+
+	adap->tc_mqprio = tc_mqprio;
+	refcount_set(&adap->tc_mqprio->refcnt, 0);
+	return 0;
+
+out_free_ports:
+	for (i = 0; i < adap->params.nports; i++) {
+		port_mqprio = &tc_mqprio->port_mqprio[i];
+		kfree(port_mqprio->eosw_txq);
+	}
+	kfree(tc_port_mqprio);
+
+out_free_mqprio:
+	kfree(tc_mqprio);
+	return ret;
+}
+
+void cxgb4_cleanup_tc_mqprio(struct adapter *adap)
+{
+	struct cxgb4_tc_port_mqprio *port_mqprio;
+	u8 i;
+
+	if (adap->tc_mqprio) {
+		mutex_lock(&adap->tc_mqprio->mqprio_mutex);
+		if (adap->tc_mqprio->port_mqprio) {
+			for (i = 0; i < adap->params.nports; i++) {
+				struct net_device *dev = adap->port[i];
+
+				if (dev)
+					cxgb4_mqprio_disable_offload(dev);
+				port_mqprio = &adap->tc_mqprio->port_mqprio[i];
+				kfree(port_mqprio->eosw_txq);
+			}
+			kfree(adap->tc_mqprio->port_mqprio);
+		}
+		mutex_unlock(&adap->tc_mqprio->mqprio_mutex);
+		kfree(adap->tc_mqprio);
+	}
+}
diff --git a/drivers/net/ethernet/chelsio/cxgb4/cxgb4_tc_mqprio.h b/drivers/net/ethernet/chelsio/cxgb4/cxgb4_tc_mqprio.h
new file mode 100644
index 000000000000..d4a862a9fd7d
--- /dev/null
+++ b/drivers/net/ethernet/chelsio/cxgb4/cxgb4_tc_mqprio.h
@@ -0,0 +1,45 @@
+/* SPDX-License-Identifier: GPL-2.0-only */
+/* Copyright (C) 2019 Chelsio Communications.  All rights reserved. */
+
+#ifndef __CXGB4_TC_MQPRIO_H__
+#define __CXGB4_TC_MQPRIO_H__
+
+#include <net/pkt_sched.h>
+
+#define CXGB4_EOSW_TXQ_DEFAULT_DESC_NUM 128
+
+#define CXGB4_EOHW_TXQ_DEFAULT_DESC_NUM 1024
+
+#define CXGB4_EOHW_RXQ_DEFAULT_DESC_NUM 1024
+#define CXGB4_EOHW_RXQ_DEFAULT_DESC_SIZE 64
+#define CXGB4_EOHW_RXQ_DEFAULT_INTR_USEC 5
+#define CXGB4_EOHW_RXQ_DEFAULT_PKT_CNT 8
+
+#define CXGB4_EOHW_FLQ_DEFAULT_DESC_NUM 72
+
+#define CXGB4_FLOWC_WAIT_TIMEOUT (5 * HZ)
+
+enum cxgb4_mqprio_state {
+	CXGB4_MQPRIO_STATE_DISABLED = 0,
+	CXGB4_MQPRIO_STATE_ACTIVE,
+};
+
+struct cxgb4_tc_port_mqprio {
+	enum cxgb4_mqprio_state state; /* Current MQPRIO offload state */
+	struct tc_mqprio_qopt_offload mqprio; /* MQPRIO offload params */
+	struct sge_eosw_txq *eosw_txq; /* Netdev SW Tx queue array */
+	u8 tc_hwtc_map[TC_QOPT_MAX_QUEUE]; /* MQPRIO tc to hardware tc map */
+};
+
+struct cxgb4_tc_mqprio {
+	refcount_t refcnt; /* Refcount for adapter-wide resources */
+	struct mutex mqprio_mutex; /* Lock for accessing MQPRIO info */
+	struct cxgb4_tc_port_mqprio *port_mqprio; /* Per port MQPRIO info */
+};
+
+int cxgb4_setup_tc_mqprio(struct net_device *dev,
+			  struct tc_mqprio_qopt_offload *mqprio);
+void cxgb4_mqprio_stop_offload(struct adapter *adap);
+int cxgb4_init_tc_mqprio(struct adapter *adap);
+void cxgb4_cleanup_tc_mqprio(struct adapter *adap);
+#endif /* __CXGB4_TC_MQPRIO_H__ */
diff --git a/drivers/net/ethernet/chelsio/cxgb4/cxgb4_tc_u32.c b/drivers/net/ethernet/chelsio/cxgb4/cxgb4_tc_u32.c
index ef06ce8247ab..8524246fd67e 100644
--- a/drivers/net/ethernet/chelsio/cxgb4/cxgb4_tc_u32.c
+++ b/drivers/net/ethernet/chelsio/cxgb4/cxgb4_tc_u32.c
@@ -36,6 +36,7 @@
 #include <net/tc_act/tc_mirred.h>
 
 #include "cxgb4.h"
+#include "cxgb4_filter.h"
 #include "cxgb4_tc_u32_parse.h"
 #include "cxgb4_tc_u32.h"
 
@@ -47,7 +48,7 @@ static int fill_match_fields(struct adapter *adap,
 			     bool next_header)
 {
 	unsigned int i, j;
-	u32 val, mask;
+	__be32 val, mask;
 	int off, err;
 	bool found;
 
@@ -93,14 +94,13 @@ static int fill_action_fields(struct adapter *adap,
 	unsigned int num_actions = 0;
 	const struct tc_action *a;
 	struct tcf_exts *exts;
-	LIST_HEAD(actions);
+	int i;
 
 	exts = cls->knode.exts;
-	if (tc_no_actions(exts))
+	if (!tcf_exts_has_actions(exts))
 		return -EINVAL;
 
-	tcf_exts_to_list(exts, &actions);
-	list_for_each_entry(a, &actions, list) {
+	tcf_exts_for_each_action(i, a, exts) {
 		/* Don't allow more than one action per rule. */
 		if (num_actions)
 			return -EINVAL;
@@ -114,14 +114,14 @@ static int fill_action_fields(struct adapter *adap,
 
 		/* Re-direct to specified port in hardware. */
 		if (is_tcf_mirred_egress_redirect(a)) {
-			struct net_device *n_dev;
-			unsigned int i, index;
+			struct net_device *n_dev, *target_dev;
 			bool found = false;
+			unsigned int i;
 
-			index = tcf_mirred_ifindex(a);
+			target_dev = tcf_mirred_dev(a);
 			for_each_port(adap, i) {
 				n_dev = adap->port[i];
-				if (index == n_dev->ifindex) {
+				if (target_dev == n_dev) {
 					fs->action = FILTER_SWITCH;
 					fs->eport = i;
 					found = true;
@@ -146,17 +146,19 @@ static int fill_action_fields(struct adapter *adap,
 	return 0;
 }
 
-int cxgb4_config_knode(struct net_device *dev, __be16 protocol,
-		       struct tc_cls_u32_offload *cls)
+int cxgb4_config_knode(struct net_device *dev, struct tc_cls_u32_offload *cls)
 {
 	const struct cxgb4_match_field *start, *link_start = NULL;
+	struct netlink_ext_ack *extack = cls->common.extack;
 	struct adapter *adapter = netdev2adap(dev);
+	__be16 protocol = cls->common.protocol;
 	struct ch_filter_specification fs;
 	struct cxgb4_tc_u32_table *t;
 	struct cxgb4_link *link;
-	unsigned int filter_id;
 	u32 uhtid, link_uhtid;
 	bool is_ipv6 = false;
+	u8 inet_family;
+	int filter_id;
 	int ret;
 
 	if (!can_tc_u32_offload(dev))
@@ -165,14 +167,18 @@ int cxgb4_config_knode(struct net_device *dev, __be16 protocol,
 	if (protocol != htons(ETH_P_IP) && protocol != htons(ETH_P_IPV6))
 		return -EOPNOTSUPP;
 
-	/* Fetch the location to insert the filter. */
-	filter_id = cls->knode.handle & 0xFFFFF;
+	inet_family = (protocol == htons(ETH_P_IPV6)) ? PF_INET6 : PF_INET;
 
-	if (filter_id > adapter->tids.nftids) {
-		dev_err(adapter->pdev_dev,
-			"Location %d out of range for insertion. Max: %d\n",
-			filter_id, adapter->tids.nftids);
-		return -ERANGE;
+	/* Get a free filter entry TID, where we can insert this new
+	 * rule. Only insert rule if its prio doesn't conflict with
+	 * existing rules.
+	 */
+	filter_id = cxgb4_get_free_ftid(dev, inet_family, false,
+					TC_U32_NODE(cls->knode.handle));
+	if (filter_id < 0) {
+		NL_SET_ERR_MSG_MOD(extack,
+				   "No free LETCAM index available");
+		return -ENOMEM;
 	}
 
 	t = adapter->tc_u32;
@@ -180,7 +186,7 @@ int cxgb4_config_knode(struct net_device *dev, __be16 protocol,
 	link_uhtid = TC_U32_USERHTID(cls->knode.link_handle);
 
 	/* Ensure that uhtid is either root u32 (i.e. 0x800)
-	 * or a a valid linked bucket.
+	 * or a valid linked bucket.
 	 */
 	if (uhtid != 0x800 && uhtid >= t->size)
 		return -EINVAL;
@@ -191,6 +197,11 @@ int cxgb4_config_knode(struct net_device *dev, __be16 protocol,
 
 	memset(&fs, 0, sizeof(fs));
 
+	if (filter_id < adapter->tids.nhpftids)
+		fs.prio = 1;
+	fs.tc_prio = cls->common.prio;
+	fs.tc_cookie = cls->knode.handle;
+
 	if (protocol == htons(ETH_P_IPV6)) {
 		start = cxgb4_ipv6_fields;
 		is_ipv6 = true;
@@ -217,7 +228,7 @@ int cxgb4_config_knode(struct net_device *dev, __be16 protocol,
 		const struct cxgb4_next_header *next;
 		bool found = false;
 		unsigned int i, j;
-		u32 val, mask;
+		__be32 val, mask;
 		int off;
 
 		if (t->table[link_uhtid - 1].link_handle) {
@@ -231,10 +242,10 @@ int cxgb4_config_knode(struct net_device *dev, __be16 protocol,
 
 		/* Try to find matches that allow jumps to next header. */
 		for (i = 0; next[i].jump; i++) {
-			if (next[i].offoff != cls->knode.sel->offoff ||
-			    next[i].shift != cls->knode.sel->offshift ||
-			    next[i].mask != cls->knode.sel->offmask ||
-			    next[i].offset != cls->knode.sel->off)
+			if (next[i].sel.offoff != cls->knode.sel->offoff ||
+			    next[i].sel.offshift != cls->knode.sel->offshift ||
+			    next[i].sel.offmask != cls->knode.sel->offmask ||
+			    next[i].sel.off != cls->knode.sel->off)
 				continue;
 
 			/* Found a possible candidate.  Find a key that
@@ -246,9 +257,9 @@ int cxgb4_config_knode(struct net_device *dev, __be16 protocol,
 				val = cls->knode.sel->keys[j].val;
 				mask = cls->knode.sel->keys[j].mask;
 
-				if (next[i].match_off == off &&
-				    next[i].match_val == val &&
-				    next[i].match_mask == mask) {
+				if (next[i].key.off == off &&
+				    next[i].key.val == val &&
+				    next[i].key.mask == mask) {
 					found = true;
 					break;
 				}
@@ -338,35 +349,80 @@ out:
 	return ret;
 }
 
-int cxgb4_delete_knode(struct net_device *dev, __be16 protocol,
-		       struct tc_cls_u32_offload *cls)
+int cxgb4_delete_knode(struct net_device *dev, struct tc_cls_u32_offload *cls)
 {
 	struct adapter *adapter = netdev2adap(dev);
 	unsigned int filter_id, max_tids, i, j;
 	struct cxgb4_link *link = NULL;
 	struct cxgb4_tc_u32_table *t;
+	struct filter_entry *f;
+	bool found = false;
 	u32 handle, uhtid;
+	u8 nslots;
 	int ret;
 
 	if (!can_tc_u32_offload(dev))
 		return -EOPNOTSUPP;
 
 	/* Fetch the location to delete the filter. */
-	filter_id = cls->knode.handle & 0xFFFFF;
+	max_tids = adapter->tids.nhpftids + adapter->tids.nftids;
+
+	spin_lock_bh(&adapter->tids.ftid_lock);
+	filter_id = 0;
+	while (filter_id < max_tids) {
+		if (filter_id < adapter->tids.nhpftids) {
+			i = filter_id;
+			f = &adapter->tids.hpftid_tab[i];
+			if (f->valid && f->fs.tc_cookie == cls->knode.handle) {
+				found = true;
+				break;
+			}
 
-	if (filter_id > adapter->tids.nftids) {
-		dev_err(adapter->pdev_dev,
-			"Location %d out of range for deletion. Max: %d\n",
-			filter_id, adapter->tids.nftids);
-		return -ERANGE;
+			i = find_next_bit(adapter->tids.hpftid_bmap,
+					  adapter->tids.nhpftids, i + 1);
+			if (i >= adapter->tids.nhpftids) {
+				filter_id = adapter->tids.nhpftids;
+				continue;
+			}
+
+			filter_id = i;
+		} else {
+			i = filter_id - adapter->tids.nhpftids;
+			f = &adapter->tids.ftid_tab[i];
+			if (f->valid && f->fs.tc_cookie == cls->knode.handle) {
+				found = true;
+				break;
+			}
+
+			i = find_next_bit(adapter->tids.ftid_bmap,
+					  adapter->tids.nftids, i + 1);
+			if (i >= adapter->tids.nftids)
+				break;
+
+			filter_id = i + adapter->tids.nhpftids;
+		}
+
+		nslots = 0;
+		if (f->fs.type) {
+			nslots++;
+			if (CHELSIO_CHIP_VERSION(adapter->params.chip) <
+			    CHELSIO_T6)
+				nslots += 2;
+		}
+
+		filter_id += nslots;
 	}
+	spin_unlock_bh(&adapter->tids.ftid_lock);
+
+	if (!found)
+		return -ERANGE;
 
 	t = adapter->tc_u32;
 	handle = cls->knode.handle;
 	uhtid = TC_U32_USERHTID(cls->knode.handle);
 
 	/* Ensure that uhtid is either root u32 (i.e. 0x800)
-	 * or a a valid linked bucket.
+	 * or a valid linked bucket.
 	 */
 	if (uhtid != 0x800 && uhtid >= t->size)
 		return -EINVAL;
@@ -381,7 +437,7 @@ int cxgb4_delete_knode(struct net_device *dev, __be16 protocol,
 			return -EINVAL;
 	}
 
-	ret = cxgb4_del_filter(dev, filter_id);
+	ret = cxgb4_del_filter(dev, filter_id, NULL);
 	if (ret)
 		goto out;
 
@@ -391,7 +447,6 @@ int cxgb4_delete_knode(struct net_device *dev, __be16 protocol,
 	/* If a link is being deleted, then delete all filters
 	 * associated with the link.
 	 */
-	max_tids = adapter->tids.nftids;
 	for (i = 0; i < t->size; i++) {
 		link = &t->table[i];
 
@@ -400,7 +455,7 @@ int cxgb4_delete_knode(struct net_device *dev, __be16 protocol,
 				if (!test_bit(j, link->tid_map))
 					continue;
 
-				ret = __cxgb4_del_filter(dev, j, NULL);
+				ret = __cxgb4_del_filter(dev, j, NULL, NULL);
 				if (ret)
 					goto out;
 
@@ -439,15 +494,14 @@ void cxgb4_cleanup_tc_u32(struct adapter *adap)
 
 struct cxgb4_tc_u32_table *cxgb4_init_tc_u32(struct adapter *adap)
 {
-	unsigned int max_tids = adap->tids.nftids;
+	unsigned int max_tids = adap->tids.nftids + adap->tids.nhpftids;
 	struct cxgb4_tc_u32_table *t;
 	unsigned int i;
 
 	if (!max_tids)
 		return NULL;
 
-	t = kvzalloc(sizeof(*t) +
-			 (max_tids * sizeof(struct cxgb4_link)), GFP_KERNEL);
+	t = kvzalloc(struct_size(t, table, max_tids), GFP_KERNEL);
 	if (!t)
 		return NULL;
 
@@ -458,7 +512,8 @@ struct cxgb4_tc_u32_table *cxgb4_init_tc_u32(struct adapter *adap)
 		unsigned int bmap_size;
 
 		bmap_size = BITS_TO_LONGS(max_tids);
-		link->tid_map = kvzalloc(sizeof(unsigned long) * bmap_size, GFP_KERNEL);
+		link->tid_map = kvcalloc(bmap_size, sizeof(unsigned long),
+					 GFP_KERNEL);
 		if (!link->tid_map)
 			goto out_no_mem;
 		bitmap_zero(link->tid_map, max_tids);
@@ -469,13 +524,9 @@ struct cxgb4_tc_u32_table *cxgb4_init_tc_u32(struct adapter *adap)
 out_no_mem:
 	for (i = 0; i < t->size; i++) {
 		struct cxgb4_link *link = &t->table[i];
-
-		if (link->tid_map)
-			kvfree(link->tid_map);
+		kvfree(link->tid_map);
 	}
-
-	if (t)
-		kvfree(t);
+	kvfree(t);
 
 	return NULL;
 }
diff --git a/drivers/net/ethernet/chelsio/cxgb4/cxgb4_tc_u32.h b/drivers/net/ethernet/chelsio/cxgb4/cxgb4_tc_u32.h
index 021261a41c13..70a07b7cca56 100644
--- a/drivers/net/ethernet/chelsio/cxgb4/cxgb4_tc_u32.h
+++ b/drivers/net/ethernet/chelsio/cxgb4/cxgb4_tc_u32.h
@@ -44,10 +44,8 @@ static inline bool can_tc_u32_offload(struct net_device *dev)
 	return (dev->features & NETIF_F_HW_TC) && adap->tc_u32 ? true : false;
 }
 
-int cxgb4_config_knode(struct net_device *dev, __be16 protocol,
-		       struct tc_cls_u32_offload *cls);
-int cxgb4_delete_knode(struct net_device *dev, __be16 protocol,
-		       struct tc_cls_u32_offload *cls);
+int cxgb4_config_knode(struct net_device *dev, struct tc_cls_u32_offload *cls);
+int cxgb4_delete_knode(struct net_device *dev, struct tc_cls_u32_offload *cls);
 
 void cxgb4_cleanup_tc_u32(struct adapter *adapter);
 struct cxgb4_tc_u32_table *cxgb4_init_tc_u32(struct adapter *adap);
diff --git a/drivers/net/ethernet/chelsio/cxgb4/cxgb4_tc_u32_parse.h b/drivers/net/ethernet/chelsio/cxgb4/cxgb4_tc_u32_parse.h
index a4b99edcc339..64663112cad8 100644
--- a/drivers/net/ethernet/chelsio/cxgb4/cxgb4_tc_u32_parse.h
+++ b/drivers/net/ethernet/chelsio/cxgb4/cxgb4_tc_u32_parse.h
@@ -38,12 +38,12 @@
 struct cxgb4_match_field {
 	int off; /* Offset from the beginning of the header to match */
 	/* Fill the value/mask pair in the spec if matched */
-	int (*val)(struct ch_filter_specification *f, u32 val, u32 mask);
+	int (*val)(struct ch_filter_specification *f, __be32 val, __be32 mask);
 };
 
 /* IPv4 match fields */
 static inline int cxgb4_fill_ipv4_tos(struct ch_filter_specification *f,
-				      u32 val, u32 mask)
+				      __be32 val, __be32 mask)
 {
 	f->val.tos  = (ntohl(val)  >> 16) & 0x000000FF;
 	f->mask.tos = (ntohl(mask) >> 16) & 0x000000FF;
@@ -52,7 +52,7 @@ static inline int cxgb4_fill_ipv4_tos(struct ch_filter_specification *f,
 }
 
 static inline int cxgb4_fill_ipv4_frag(struct ch_filter_specification *f,
-				       u32 val, u32 mask)
+				       __be32 val, __be32 mask)
 {
 	u32 mask_val;
 	u8 frag_val;
@@ -74,7 +74,7 @@ static inline int cxgb4_fill_ipv4_frag(struct ch_filter_specification *f,
 }
 
 static inline int cxgb4_fill_ipv4_proto(struct ch_filter_specification *f,
-					u32 val, u32 mask)
+					__be32 val, __be32 mask)
 {
 	f->val.proto  = (ntohl(val)  >> 16) & 0x000000FF;
 	f->mask.proto = (ntohl(mask) >> 16) & 0x000000FF;
@@ -83,7 +83,7 @@ static inline int cxgb4_fill_ipv4_proto(struct ch_filter_specification *f,
 }
 
 static inline int cxgb4_fill_ipv4_src_ip(struct ch_filter_specification *f,
-					 u32 val, u32 mask)
+					 __be32 val, __be32 mask)
 {
 	memcpy(&f->val.fip[0],  &val,  sizeof(u32));
 	memcpy(&f->mask.fip[0], &mask, sizeof(u32));
@@ -92,7 +92,7 @@ static inline int cxgb4_fill_ipv4_src_ip(struct ch_filter_specification *f,
 }
 
 static inline int cxgb4_fill_ipv4_dst_ip(struct ch_filter_specification *f,
-					 u32 val, u32 mask)
+					 __be32 val, __be32 mask)
 {
 	memcpy(&f->val.lip[0],  &val,  sizeof(u32));
 	memcpy(&f->mask.lip[0], &mask, sizeof(u32));
@@ -111,7 +111,7 @@ static const struct cxgb4_match_field cxgb4_ipv4_fields[] = {
 
 /* IPv6 match fields */
 static inline int cxgb4_fill_ipv6_tos(struct ch_filter_specification *f,
-				      u32 val, u32 mask)
+				      __be32 val, __be32 mask)
 {
 	f->val.tos  = (ntohl(val)  >> 20) & 0x000000FF;
 	f->mask.tos = (ntohl(mask) >> 20) & 0x000000FF;
@@ -120,7 +120,7 @@ static inline int cxgb4_fill_ipv6_tos(struct ch_filter_specification *f,
 }
 
 static inline int cxgb4_fill_ipv6_proto(struct ch_filter_specification *f,
-					u32 val, u32 mask)
+					__be32 val, __be32 mask)
 {
 	f->val.proto  = (ntohl(val)  >> 8) & 0x000000FF;
 	f->mask.proto = (ntohl(mask) >> 8) & 0x000000FF;
@@ -129,7 +129,7 @@ static inline int cxgb4_fill_ipv6_proto(struct ch_filter_specification *f,
 }
 
 static inline int cxgb4_fill_ipv6_src_ip0(struct ch_filter_specification *f,
-					  u32 val, u32 mask)
+					  __be32 val, __be32 mask)
 {
 	memcpy(&f->val.fip[0],  &val,  sizeof(u32));
 	memcpy(&f->mask.fip[0], &mask, sizeof(u32));
@@ -138,7 +138,7 @@ static inline int cxgb4_fill_ipv6_src_ip0(struct ch_filter_specification *f,
 }
 
 static inline int cxgb4_fill_ipv6_src_ip1(struct ch_filter_specification *f,
-					  u32 val, u32 mask)
+					  __be32 val, __be32 mask)
 {
 	memcpy(&f->val.fip[4],  &val,  sizeof(u32));
 	memcpy(&f->mask.fip[4], &mask, sizeof(u32));
@@ -147,7 +147,7 @@ static inline int cxgb4_fill_ipv6_src_ip1(struct ch_filter_specification *f,
 }
 
 static inline int cxgb4_fill_ipv6_src_ip2(struct ch_filter_specification *f,
-					  u32 val, u32 mask)
+					  __be32 val, __be32 mask)
 {
 	memcpy(&f->val.fip[8],  &val,  sizeof(u32));
 	memcpy(&f->mask.fip[8], &mask, sizeof(u32));
@@ -156,7 +156,7 @@ static inline int cxgb4_fill_ipv6_src_ip2(struct ch_filter_specification *f,
 }
 
 static inline int cxgb4_fill_ipv6_src_ip3(struct ch_filter_specification *f,
-					  u32 val, u32 mask)
+					  __be32 val, __be32 mask)
 {
 	memcpy(&f->val.fip[12],  &val,  sizeof(u32));
 	memcpy(&f->mask.fip[12], &mask, sizeof(u32));
@@ -165,7 +165,7 @@ static inline int cxgb4_fill_ipv6_src_ip3(struct ch_filter_specification *f,
 }
 
 static inline int cxgb4_fill_ipv6_dst_ip0(struct ch_filter_specification *f,
-					  u32 val, u32 mask)
+					  __be32 val, __be32 mask)
 {
 	memcpy(&f->val.lip[0],  &val,  sizeof(u32));
 	memcpy(&f->mask.lip[0], &mask, sizeof(u32));
@@ -174,7 +174,7 @@ static inline int cxgb4_fill_ipv6_dst_ip0(struct ch_filter_specification *f,
 }
 
 static inline int cxgb4_fill_ipv6_dst_ip1(struct ch_filter_specification *f,
-					  u32 val, u32 mask)
+					  __be32 val, __be32 mask)
 {
 	memcpy(&f->val.lip[4],  &val,  sizeof(u32));
 	memcpy(&f->mask.lip[4], &mask, sizeof(u32));
@@ -183,7 +183,7 @@ static inline int cxgb4_fill_ipv6_dst_ip1(struct ch_filter_specification *f,
 }
 
 static inline int cxgb4_fill_ipv6_dst_ip2(struct ch_filter_specification *f,
-					  u32 val, u32 mask)
+					  __be32 val, __be32 mask)
 {
 	memcpy(&f->val.lip[8],  &val,  sizeof(u32));
 	memcpy(&f->mask.lip[8], &mask, sizeof(u32));
@@ -192,7 +192,7 @@ static inline int cxgb4_fill_ipv6_dst_ip2(struct ch_filter_specification *f,
 }
 
 static inline int cxgb4_fill_ipv6_dst_ip3(struct ch_filter_specification *f,
-					  u32 val, u32 mask)
+					  __be32 val, __be32 mask)
 {
 	memcpy(&f->val.lip[12],  &val,  sizeof(u32));
 	memcpy(&f->mask.lip[12], &mask, sizeof(u32));
@@ -216,7 +216,7 @@ static const struct cxgb4_match_field cxgb4_ipv6_fields[] = {
 
 /* TCP/UDP match */
 static inline int cxgb4_fill_l4_ports(struct ch_filter_specification *f,
-				      u32 val, u32 mask)
+				      __be32 val, __be32 mask)
 {
 	f->val.fport  = ntohl(val)  >> 16;
 	f->mask.fport = ntohl(mask) >> 16;
@@ -237,19 +237,13 @@ static const struct cxgb4_match_field cxgb4_udp_fields[] = {
 };
 
 struct cxgb4_next_header {
-	unsigned int offset; /* Offset to next header */
-	/* offset, shift, and mask added to offset above
+	/* Offset, shift, and mask added to beginning of the header
 	 * to get to next header.  Useful when using a header
 	 * field's value to jump to next header such as IHL field
 	 * in IPv4 header.
 	 */
-	unsigned int offoff;
-	u32 shift;
-	u32 mask;
-	/* match criteria to make this jump */
-	unsigned int match_off;
-	u32 match_val;
-	u32 match_mask;
+	struct tc_u32_sel_hdr sel;
+	struct tc_u32_key key;
 	/* location of jump to make */
 	const struct cxgb4_match_field *jump;
 };
@@ -258,26 +252,74 @@ struct cxgb4_next_header {
  * IPv4 header.
  */
 static const struct cxgb4_next_header cxgb4_ipv4_jumps[] = {
-	{ .offset = 0, .offoff = 0, .shift = 6, .mask = 0xF,
-	  .match_off = 8, .match_val = 0x600, .match_mask = 0xFF00,
-	  .jump = cxgb4_tcp_fields },
-	{ .offset = 0, .offoff = 0, .shift = 6, .mask = 0xF,
-	  .match_off = 8, .match_val = 0x1100, .match_mask = 0xFF00,
-	  .jump = cxgb4_udp_fields },
-	{ .jump = NULL }
+	{
+		/* TCP Jump */
+		.sel = {
+			.off = 0,
+			.offoff = 0,
+			.offshift = 6,
+			.offmask = cpu_to_be16(0x0f00),
+		},
+		.key = {
+			.off = 8,
+			.val = cpu_to_be32(0x00060000),
+			.mask = cpu_to_be32(0x00ff0000),
+		},
+		.jump = cxgb4_tcp_fields,
+	},
+	{
+		/* UDP Jump */
+		.sel = {
+			.off = 0,
+			.offoff = 0,
+			.offshift = 6,
+			.offmask = cpu_to_be16(0x0f00),
+		},
+		.key = {
+			.off = 8,
+			.val = cpu_to_be32(0x00110000),
+			.mask = cpu_to_be32(0x00ff0000),
+		},
+		.jump = cxgb4_udp_fields,
+	},
+	{ .jump = NULL },
 };
 
 /* Accept a rule with a jump directly past the 40 Bytes of IPv6 fixed header
  * to get to transport layer header.
  */
 static const struct cxgb4_next_header cxgb4_ipv6_jumps[] = {
-	{ .offset = 0x28, .offoff = 0, .shift = 0, .mask = 0,
-	  .match_off = 4, .match_val = 0x60000, .match_mask = 0xFF0000,
-	  .jump = cxgb4_tcp_fields },
-	{ .offset = 0x28, .offoff = 0, .shift = 0, .mask = 0,
-	  .match_off = 4, .match_val = 0x110000, .match_mask = 0xFF0000,
-	  .jump = cxgb4_udp_fields },
-	{ .jump = NULL }
+	{
+		/* TCP Jump */
+		.sel = {
+			.off = 40,
+			.offoff = 0,
+			.offshift = 0,
+			.offmask = 0,
+		},
+		.key = {
+			.off = 4,
+			.val = cpu_to_be32(0x00000600),
+			.mask = cpu_to_be32(0x0000ff00),
+		},
+		.jump = cxgb4_tcp_fields,
+	},
+	{
+		/* UDP Jump */
+		.sel = {
+			.off = 40,
+			.offoff = 0,
+			.offshift = 0,
+			.offmask = 0,
+		},
+		.key = {
+			.off = 4,
+			.val = cpu_to_be32(0x00001100),
+			.mask = cpu_to_be32(0x0000ff00),
+		},
+		.jump = cxgb4_udp_fields,
+	},
+	{ .jump = NULL },
 };
 
 struct cxgb4_link {
@@ -289,6 +331,6 @@ struct cxgb4_link {
 
 struct cxgb4_tc_u32_table {
 	unsigned int size;          /* number of entries in table */
-	struct cxgb4_link table[0]; /* Jump table */
+	struct cxgb4_link table[] __counted_by(size); /* Jump table */
 };
 #endif /* __CXGB4_TC_U32_PARSE_H */
diff --git a/drivers/net/ethernet/chelsio/cxgb4/cxgb4_thermal.c b/drivers/net/ethernet/chelsio/cxgb4/cxgb4_thermal.c
new file mode 100644
index 000000000000..7bab8da8f6e6
--- /dev/null
+++ b/drivers/net/ethernet/chelsio/cxgb4/cxgb4_thermal.c
@@ -0,0 +1,90 @@
+// SPDX-License-Identifier: GPL-2.0-only
+/*
+ *  Copyright (C) 2017 Chelsio Communications.  All rights reserved.
+ *
+ *  Written by: Ganesh Goudar (ganeshgr@chelsio.com)
+ */
+
+#include "cxgb4.h"
+
+#define CXGB4_NUM_TRIPS 1
+
+static int cxgb4_thermal_get_temp(struct thermal_zone_device *tzdev,
+				  int *temp)
+{
+	struct adapter *adap = thermal_zone_device_priv(tzdev);
+	u32 param, val;
+	int ret;
+
+	param = (FW_PARAMS_MNEM_V(FW_PARAMS_MNEM_DEV) |
+		 FW_PARAMS_PARAM_X_V(FW_PARAMS_PARAM_DEV_DIAG) |
+		 FW_PARAMS_PARAM_Y_V(FW_PARAM_DEV_DIAG_TMP));
+
+	ret = t4_query_params(adap, adap->mbox, adap->pf, 0, 1,
+			      &param, &val);
+	if (ret < 0 || val == 0)
+		return -1;
+
+	*temp = val * 1000;
+	return 0;
+}
+
+static const struct thermal_zone_device_ops cxgb4_thermal_ops = {
+	.get_temp = cxgb4_thermal_get_temp,
+};
+
+static struct thermal_trip trip = { .type = THERMAL_TRIP_CRITICAL } ;
+
+int cxgb4_thermal_init(struct adapter *adap)
+{
+	struct ch_thermal *ch_thermal = &adap->ch_thermal;
+	char ch_tz_name[THERMAL_NAME_LENGTH];
+	int num_trip = CXGB4_NUM_TRIPS;
+	u32 param, val;
+	int ret;
+
+	/* on older firmwares we may not get the trip temperature,
+	 * set the num of trips to 0.
+	 */
+	param = (FW_PARAMS_MNEM_V(FW_PARAMS_MNEM_DEV) |
+		 FW_PARAMS_PARAM_X_V(FW_PARAMS_PARAM_DEV_DIAG) |
+		 FW_PARAMS_PARAM_Y_V(FW_PARAM_DEV_DIAG_MAXTMPTHRESH));
+
+	ret = t4_query_params(adap, adap->mbox, adap->pf, 0, 1,
+			      &param, &val);
+	if (ret < 0) {
+		num_trip = 0; /* could not get trip temperature */
+	} else {
+		trip.temperature = val * 1000;
+	}
+
+	snprintf(ch_tz_name, sizeof(ch_tz_name), "cxgb4_%s", adap->name);
+	ch_thermal->tzdev = thermal_zone_device_register_with_trips(ch_tz_name, &trip, num_trip,
+								    adap,
+								    &cxgb4_thermal_ops,
+								    NULL, 0, 0);
+	if (IS_ERR(ch_thermal->tzdev)) {
+		ret = PTR_ERR(ch_thermal->tzdev);
+		dev_err(adap->pdev_dev, "Failed to register thermal zone\n");
+		ch_thermal->tzdev = NULL;
+		return ret;
+	}
+
+	ret = thermal_zone_device_enable(ch_thermal->tzdev);
+	if (ret) {
+		dev_err(adap->pdev_dev, "Failed to enable thermal zone\n");
+		thermal_zone_device_unregister(adap->ch_thermal.tzdev);
+		return ret;
+	}
+
+	return 0;
+}
+
+int cxgb4_thermal_remove(struct adapter *adap)
+{
+	if (adap->ch_thermal.tzdev) {
+		thermal_zone_device_unregister(adap->ch_thermal.tzdev);
+		adap->ch_thermal.tzdev = NULL;
+	}
+	return 0;
+}
diff --git a/drivers/net/ethernet/chelsio/cxgb4/cxgb4_uld.c b/drivers/net/ethernet/chelsio/cxgb4/cxgb4_uld.c
index d0868c2320da..5c13bcb4550d 100644
--- a/drivers/net/ethernet/chelsio/cxgb4/cxgb4_uld.c
+++ b/drivers/net/ethernet/chelsio/cxgb4/cxgb4_uld.c
@@ -53,35 +53,6 @@
 
 #define for_each_uldrxq(m, i) for (i = 0; i < ((m)->nrxq + (m)->nciq); i++)
 
-static int get_msix_idx_from_bmap(struct adapter *adap)
-{
-	struct uld_msix_bmap *bmap = &adap->msix_bmap_ulds;
-	unsigned long flags;
-	unsigned int msix_idx;
-
-	spin_lock_irqsave(&bmap->lock, flags);
-	msix_idx = find_first_zero_bit(bmap->msix_bmap, bmap->mapsize);
-	if (msix_idx < bmap->mapsize) {
-		__set_bit(msix_idx, bmap->msix_bmap);
-	} else {
-		spin_unlock_irqrestore(&bmap->lock, flags);
-		return -ENOSPC;
-	}
-
-	spin_unlock_irqrestore(&bmap->lock, flags);
-	return msix_idx;
-}
-
-static void free_msix_idx_in_bmap(struct adapter *adap, unsigned int msix_idx)
-{
-	struct uld_msix_bmap *bmap = &adap->msix_bmap_ulds;
-	unsigned long flags;
-
-	spin_lock_irqsave(&bmap->lock, flags);
-	 __clear_bit(msix_idx, bmap->msix_bmap);
-	spin_unlock_irqrestore(&bmap->lock, flags);
-}
-
 /* Flush the aggregated lro sessions */
 static void uldrx_flush_handler(struct sge_rspq *q)
 {
@@ -137,17 +108,16 @@ static int uldrx_handler(struct sge_rspq *q, const __be64 *rsp,
 static int alloc_uld_rxqs(struct adapter *adap,
 			  struct sge_uld_rxq_info *rxq_info, bool lro)
 {
-	struct sge *s = &adap->sge;
 	unsigned int nq = rxq_info->nrxq + rxq_info->nciq;
 	struct sge_ofld_rxq *q = rxq_info->uldrxq;
 	unsigned short *ids = rxq_info->rspq_id;
-	unsigned int bmap_idx = 0;
-	unsigned int per_chan;
 	int i, err, msi_idx, que_idx = 0;
+	struct sge *s = &adap->sge;
+	unsigned int per_chan;
 
 	per_chan = rxq_info->nrxq / adap->params.nports;
 
-	if (adap->flags & USING_MSIX)
+	if (adap->flags & CXGB4_USING_MSIX)
 		msi_idx = 1;
 	else
 		msi_idx = -((int)s->intrq.abs_id + 1);
@@ -160,8 +130,18 @@ static int alloc_uld_rxqs(struct adapter *adap,
 		}
 
 		if (msi_idx >= 0) {
-			bmap_idx = get_msix_idx_from_bmap(adap);
-			msi_idx = adap->msix_info_ulds[bmap_idx].idx;
+			msi_idx = cxgb4_get_msix_idx_from_bmap(adap);
+			if (msi_idx < 0) {
+				err = -ENOSPC;
+				goto freeout;
+			}
+
+			snprintf(adap->msix_info[msi_idx].desc,
+				 sizeof(adap->msix_info[msi_idx].desc),
+				 "%s-%s%d",
+				 adap->port[0]->name, rxq_info->name, i);
+
+			q->msix = &adap->msix_info[msi_idx];
 		}
 		err = t4_sge_alloc_rxq(adap, &q->rspq, false,
 				       adap->port[que_idx++ / per_chan],
@@ -172,8 +152,7 @@ static int alloc_uld_rxqs(struct adapter *adap,
 				       0);
 		if (err)
 			goto freeout;
-		if (msi_idx >= 0)
-			rxq_info->msix_tbl[i] = bmap_idx;
+
 		memset(&q->stats, 0, sizeof(q->stats));
 		if (ids)
 			ids[i] = q->rspq.abs_id;
@@ -185,6 +164,8 @@ freeout:
 		if (q->rspq.desc)
 			free_rspq_fl(adap, &q->rspq,
 				     q->fl.size ? &q->fl : NULL);
+		if (q->msix)
+			cxgb4_free_msix_idx_in_bmap(adap, q->msix->idx);
 	}
 	return err;
 }
@@ -193,21 +174,14 @@ static int
 setup_sge_queues_uld(struct adapter *adap, unsigned int uld_type, bool lro)
 {
 	struct sge_uld_rxq_info *rxq_info = adap->sge.uld_rxq_info[uld_type];
-	int i, ret = 0;
-
-	if (adap->flags & USING_MSIX) {
-		rxq_info->msix_tbl = kcalloc((rxq_info->nrxq + rxq_info->nciq),
-					     sizeof(unsigned short),
-					     GFP_KERNEL);
-		if (!rxq_info->msix_tbl)
-			return -ENOMEM;
-	}
+	int i, ret;
 
-	ret = !(!alloc_uld_rxqs(adap, rxq_info, lro));
+	ret = alloc_uld_rxqs(adap, rxq_info, lro);
+	if (ret)
+		return ret;
 
 	/* Tell uP to route control queue completions to rdma rspq */
-	if (adap->flags & FULL_INIT_DONE &&
-	    !ret && uld_type == CXGB4_ULD_RDMA) {
+	if (adap->flags & CXGB4_FULL_INIT_DONE && uld_type == CXGB4_ULD_RDMA) {
 		struct sge *s = &adap->sge;
 		unsigned int cmplqid;
 		u32 param, cmdop;
@@ -239,7 +213,7 @@ static void free_sge_queues_uld(struct adapter *adap, unsigned int uld_type)
 {
 	struct sge_uld_rxq_info *rxq_info = adap->sge.uld_rxq_info[uld_type];
 
-	if (adap->flags & FULL_INIT_DONE && uld_type == CXGB4_ULD_RDMA) {
+	if (adap->flags & CXGB4_FULL_INIT_DONE && uld_type == CXGB4_ULD_RDMA) {
 		struct sge *s = &adap->sge;
 		u32 param, cmdop, cmplqid = 0;
 		int i;
@@ -258,8 +232,6 @@ static void free_sge_queues_uld(struct adapter *adap, unsigned int uld_type)
 		t4_free_uld_rxqs(adap, rxq_info->nciq,
 				 rxq_info->uldrxq + rxq_info->nrxq);
 	t4_free_uld_rxqs(adap, rxq_info->nrxq, rxq_info->uldrxq);
-	if (adap->flags & USING_MSIX)
-		kfree(rxq_info->msix_tbl);
 }
 
 static int cfg_queues_uld(struct adapter *adap, unsigned int uld_type,
@@ -273,7 +245,7 @@ static int cfg_queues_uld(struct adapter *adap, unsigned int uld_type,
 	if (!rxq_info)
 		return -ENOMEM;
 
-	if (adap->flags & USING_MSIX && uld_info->nrxq > s->nqs_per_uld) {
+	if (adap->flags & CXGB4_USING_MSIX && uld_info->nrxq > s->nqs_per_uld) {
 		i = s->nqs_per_uld;
 		rxq_info->nrxq = roundup(i, adap->params.nports);
 	} else {
@@ -284,7 +256,7 @@ static int cfg_queues_uld(struct adapter *adap, unsigned int uld_type,
 	if (!uld_info->ciq) {
 		rxq_info->nciq = 0;
 	} else  {
-		if (adap->flags & USING_MSIX)
+		if (adap->flags & CXGB4_USING_MSIX)
 			rxq_info->nciq = min_t(int, s->nqs_per_uld,
 					       num_online_cpus());
 		else
@@ -342,6 +314,7 @@ static void free_queues_uld(struct adapter *adap, unsigned int uld_type)
 {
 	struct sge_uld_rxq_info *rxq_info = adap->sge.uld_rxq_info[uld_type];
 
+	adap->sge.uld_rxq_info[uld_type] = NULL;
 	kfree(rxq_info->rspq_id);
 	kfree(rxq_info->uldrxq);
 	kfree(rxq_info);
@@ -351,25 +324,30 @@ static int
 request_msix_queue_irqs_uld(struct adapter *adap, unsigned int uld_type)
 {
 	struct sge_uld_rxq_info *rxq_info = adap->sge.uld_rxq_info[uld_type];
+	struct msix_info *minfo;
+	unsigned int idx;
 	int err = 0;
-	unsigned int idx, bmap_idx;
 
 	for_each_uldrxq(rxq_info, idx) {
-		bmap_idx = rxq_info->msix_tbl[idx];
-		err = request_irq(adap->msix_info_ulds[bmap_idx].vec,
+		minfo = rxq_info->uldrxq[idx].msix;
+		err = request_irq(minfo->vec,
 				  t4_sge_intr_msix, 0,
-				  adap->msix_info_ulds[bmap_idx].desc,
+				  minfo->desc,
 				  &rxq_info->uldrxq[idx].rspq);
 		if (err)
 			goto unwind;
+
+		cxgb4_set_msix_aff(adap, minfo->vec,
+				   &minfo->aff_mask, idx);
 	}
 	return 0;
+
 unwind:
 	while (idx-- > 0) {
-		bmap_idx = rxq_info->msix_tbl[idx];
-		free_msix_idx_in_bmap(adap, bmap_idx);
-		free_irq(adap->msix_info_ulds[bmap_idx].vec,
-			 &rxq_info->uldrxq[idx].rspq);
+		minfo = rxq_info->uldrxq[idx].msix;
+		cxgb4_clear_msix_aff(minfo->vec, minfo->aff_mask);
+		cxgb4_free_msix_idx_in_bmap(adap, minfo->idx);
+		free_irq(minfo->vec, &rxq_info->uldrxq[idx].rspq);
 	}
 	return err;
 }
@@ -378,67 +356,45 @@ static void
 free_msix_queue_irqs_uld(struct adapter *adap, unsigned int uld_type)
 {
 	struct sge_uld_rxq_info *rxq_info = adap->sge.uld_rxq_info[uld_type];
-	unsigned int idx, bmap_idx;
+	struct msix_info *minfo;
+	unsigned int idx;
 
 	for_each_uldrxq(rxq_info, idx) {
-		bmap_idx = rxq_info->msix_tbl[idx];
-
-		free_msix_idx_in_bmap(adap, bmap_idx);
-		free_irq(adap->msix_info_ulds[bmap_idx].vec,
-			 &rxq_info->uldrxq[idx].rspq);
+		minfo = rxq_info->uldrxq[idx].msix;
+		cxgb4_clear_msix_aff(minfo->vec, minfo->aff_mask);
+		cxgb4_free_msix_idx_in_bmap(adap, minfo->idx);
+		free_irq(minfo->vec, &rxq_info->uldrxq[idx].rspq);
 	}
 }
 
-static void name_msix_vecs_uld(struct adapter *adap, unsigned int uld_type)
+static void enable_rx_uld(struct adapter *adap, unsigned int uld_type)
 {
 	struct sge_uld_rxq_info *rxq_info = adap->sge.uld_rxq_info[uld_type];
-	int n = sizeof(adap->msix_info_ulds[0].desc);
-	unsigned int idx, bmap_idx;
+	int idx;
 
 	for_each_uldrxq(rxq_info, idx) {
-		bmap_idx = rxq_info->msix_tbl[idx];
-
-		snprintf(adap->msix_info_ulds[bmap_idx].desc, n, "%s-%s%d",
-			 adap->port[0]->name, rxq_info->name, idx);
-	}
-}
-
-static void enable_rx(struct adapter *adap, struct sge_rspq *q)
-{
-	if (!q)
-		return;
-
-	if (q->handler)
-		napi_enable(&q->napi);
+		struct sge_rspq *q = &rxq_info->uldrxq[idx].rspq;
 
-	/* 0-increment GTS to start the timer and enable interrupts */
-	t4_write_reg(adap, MYPF_REG(SGE_PF_GTS_A),
-		     SEINTARM_V(q->intr_params) |
-		     INGRESSQID_V(q->cntxt_id));
-}
+		if (!q)
+			continue;
 
-static void quiesce_rx(struct adapter *adap, struct sge_rspq *q)
-{
-	if (q && q->handler)
-		napi_disable(&q->napi);
+		cxgb4_enable_rx(adap, q);
+	}
 }
 
-static void enable_rx_uld(struct adapter *adap, unsigned int uld_type)
+static void quiesce_rx_uld(struct adapter *adap, unsigned int uld_type)
 {
 	struct sge_uld_rxq_info *rxq_info = adap->sge.uld_rxq_info[uld_type];
 	int idx;
 
-	for_each_uldrxq(rxq_info, idx)
-		enable_rx(adap, &rxq_info->uldrxq[idx].rspq);
-}
+	for_each_uldrxq(rxq_info, idx) {
+		struct sge_rspq *q = &rxq_info->uldrxq[idx].rspq;
 
-static void quiesce_rx_uld(struct adapter *adap, unsigned int uld_type)
-{
-	struct sge_uld_rxq_info *rxq_info = adap->sge.uld_rxq_info[uld_type];
-	int idx;
+		if (!q)
+			continue;
 
-	for_each_uldrxq(rxq_info, idx)
-		quiesce_rx(adap, &rxq_info->uldrxq[idx].rspq);
+		cxgb4_quiesce_rx(q);
+	}
 }
 
 static void
@@ -450,7 +406,7 @@ free_sge_txq_uld(struct adapter *adap, struct sge_uld_txq_info *txq_info)
 	for (i = 0; i < nq; i++) {
 		struct sge_uld_txq *txq = &txq_info->uldtxq[i];
 
-		if (txq && txq->q.desc) {
+		if (txq->q.desc) {
 			tasklet_kill(&txq->qresume_tsk);
 			t4_ofld_eq_free(adap, adap->mbox, adap->pf, 0,
 					txq->q.cntxt_id);
@@ -519,10 +475,20 @@ setup_sge_txq_uld(struct adapter *adap, unsigned int uld_type,
 	txq_info = kzalloc(sizeof(*txq_info), GFP_KERNEL);
 	if (!txq_info)
 		return -ENOMEM;
+	if (uld_type == CXGB4_ULD_CRYPTO) {
+		i = min_t(int, adap->vres.ncrypto_fc,
+			  num_online_cpus());
+		txq_info->ntxq = rounddown(i, adap->params.nports);
+		if (txq_info->ntxq <= 0) {
+			dev_warn(adap->pdev_dev, "Crypto Tx Queues can't be zero\n");
+			kfree(txq_info);
+			return -EINVAL;
+		}
 
-	i = min_t(int, uld_info->ntxq, num_online_cpus());
-	txq_info->ntxq = roundup(i, adap->params.nports);
-
+	} else {
+		i = min_t(int, uld_info->ntxq, num_online_cpus());
+		txq_info->ntxq = roundup(i, adap->params.nports);
+	}
 	txq_info->uldtxq = kcalloc(txq_info->ntxq, sizeof(struct sge_uld_txq),
 				   GFP_KERNEL);
 	if (!txq_info->uldtxq) {
@@ -545,11 +511,14 @@ static void uld_queue_init(struct adapter *adap, unsigned int uld_type,
 			   struct cxgb4_lld_info *lli)
 {
 	struct sge_uld_rxq_info *rxq_info = adap->sge.uld_rxq_info[uld_type];
+	int tx_uld_type = TX_ULD(uld_type);
+	struct sge_uld_txq_info *txq_info = adap->sge.uld_txq_info[tx_uld_type];
 
 	lli->rxq_ids = rxq_info->rspq_id;
 	lli->nrxq = rxq_info->nrxq;
 	lli->ciq_ids = rxq_info->rspq_id + rxq_info->nrxq;
 	lli->nciq = rxq_info->nciq;
+	lli->ntxq = txq_info->ntxq;
 }
 
 int t4_uld_mem_alloc(struct adapter *adap)
@@ -560,13 +529,13 @@ int t4_uld_mem_alloc(struct adapter *adap)
 	if (!adap->uld)
 		return -ENOMEM;
 
-	s->uld_rxq_info = kzalloc(CXGB4_ULD_MAX *
+	s->uld_rxq_info = kcalloc(CXGB4_ULD_MAX,
 				  sizeof(struct sge_uld_rxq_info *),
 				  GFP_KERNEL);
 	if (!s->uld_rxq_info)
 		goto err_uld;
 
-	s->uld_txq_info = kzalloc(CXGB4_TX_MAX *
+	s->uld_txq_info = kcalloc(CXGB4_TX_MAX,
 				  sizeof(struct sge_uld_txq_info *),
 				  GFP_KERNEL);
 	if (!s->uld_txq_info)
@@ -589,22 +558,40 @@ void t4_uld_mem_free(struct adapter *adap)
 	kfree(adap->uld);
 }
 
+/* This function should be called with uld_mutex taken. */
+static void cxgb4_shutdown_uld_adapter(struct adapter *adap, enum cxgb4_uld type)
+{
+	if (adap->uld[type].handle) {
+		adap->uld[type].handle = NULL;
+		adap->uld[type].add = NULL;
+		release_sge_txq_uld(adap, type);
+
+		if (adap->flags & CXGB4_FULL_INIT_DONE)
+			quiesce_rx_uld(adap, type);
+
+		if (adap->flags & CXGB4_USING_MSIX)
+			free_msix_queue_irqs_uld(adap, type);
+
+		free_sge_queues_uld(adap, type);
+		free_queues_uld(adap, type);
+	}
+}
+
 void t4_uld_clean_up(struct adapter *adap)
 {
 	unsigned int i;
 
-	if (!adap->uld)
+	if (!is_uld(adap))
 		return;
+
+	mutex_lock(&uld_mutex);
 	for (i = 0; i < CXGB4_ULD_MAX; i++) {
 		if (!adap->uld[i].handle)
 			continue;
-		if (adap->flags & FULL_INIT_DONE)
-			quiesce_rx_uld(adap, i);
-		if (adap->flags & USING_MSIX)
-			free_msix_queue_irqs_uld(adap, i);
-		free_sge_queues_uld(adap, i);
-		free_queues_uld(adap, i);
+
+		cxgb4_shutdown_uld_adapter(adap, i);
 	}
+	mutex_unlock(&uld_mutex);
 }
 
 static void uld_init(struct adapter *adap, struct cxgb4_lld_info *lld)
@@ -618,10 +605,10 @@ static void uld_init(struct adapter *adap, struct cxgb4_lld_info *lld)
 	lld->ports = adap->port;
 	lld->vr = &adap->vres;
 	lld->mtus = adap->params.mtus;
-	lld->ntxq = adap->sge.ofldqsets;
 	lld->nchan = adap->params.nports;
 	lld->nports = adap->params.nports;
 	lld->wr_cred = adap->params.ofldq_wr_cred;
+	lld->crypto = adap->params.crypto;
 	lld->iscsi_iolen = MAXRXDATA_G(t4_read_reg(adap, TP_PARA_REG2_A));
 	lld->iscsi_tagmask = t4_read_reg(adap, ULP_RX_ISCSI_TAGMASK_A);
 	lld->iscsi_pgsz_order = t4_read_reg(adap, ULP_RX_ISCSI_PSZ_A);
@@ -631,6 +618,7 @@ static void uld_init(struct adapter *adap, struct cxgb4_lld_info *lld)
 	lld->cclk_ps = 1000000000 / adap->params.vpd.cclk;
 	lld->udb_density = 1 << adap->params.sge.eq_qpp;
 	lld->ucq_density = 1 << adap->params.sge.iq_qpp;
+	lld->sge_host_page_size = 1 << (adap->params.sge.hps + 10);
 	lld->filt_mode = adap->params.tp.vlan_pri_map;
 	/* MODQ_REQ_MAP sets queues 0-3 to chan 0-3 */
 	for (i = 0; i < NCHAN; i++)
@@ -642,18 +630,21 @@ static void uld_init(struct adapter *adap, struct cxgb4_lld_info *lld)
 	lld->sge_ingpadboundary = adap->sge.fl_align;
 	lld->sge_egrstatuspagesize = adap->sge.stat_len;
 	lld->sge_pktshift = adap->sge.pktshift;
-	lld->enable_fw_ofld_conn = adap->flags & FW_OFLD_CONN;
+	lld->ulp_crypto = adap->params.crypto;
+	lld->enable_fw_ofld_conn = adap->flags & CXGB4_FW_OFLD_CONN;
 	lld->max_ordird_qp = adap->params.max_ordird_qp;
 	lld->max_ird_adapter = adap->params.max_ird_adapter;
 	lld->ulptx_memwrite_dsgl = adap->params.ulptx_memwrite_dsgl;
 	lld->nodeid = dev_to_node(adap->pdev_dev);
 	lld->fr_nsmr_tpte_wr_support = adap->params.fr_nsmr_tpte_wr_support;
+	lld->write_w_imm_support = adap->params.write_w_imm_support;
+	lld->write_cmpl_support = adap->params.write_cmpl_support;
 }
 
-static void uld_attach(struct adapter *adap, unsigned int uld)
+static int uld_attach(struct adapter *adap, unsigned int uld)
 {
-	void *handle;
 	struct cxgb4_lld_info lli;
+	void *handle;
 
 	uld_init(adap, &lli);
 	uld_queue_init(adap, uld, &lli);
@@ -663,102 +654,170 @@ static void uld_attach(struct adapter *adap, unsigned int uld)
 		dev_warn(adap->pdev_dev,
 			 "could not attach to the %s driver, error %ld\n",
 			 adap->uld[uld].name, PTR_ERR(handle));
-		return;
+		return PTR_ERR(handle);
 	}
 
 	adap->uld[uld].handle = handle;
 	t4_register_netevent_notifier();
 
-	if (adap->flags & FULL_INIT_DONE)
+	if (adap->flags & CXGB4_FULL_INIT_DONE)
 		adap->uld[uld].state_change(handle, CXGB4_STATE_UP);
+
+	return 0;
 }
 
-/**
- *	cxgb4_register_uld - register an upper-layer driver
- *	@type: the ULD type
- *	@p: the ULD methods
- *
- *	Registers an upper-layer driver with this driver and notifies the ULD
- *	about any presently available devices that support its type.  Returns
- *	%-EBUSY if a ULD of the same type is already registered.
- */
-int cxgb4_register_uld(enum cxgb4_uld type,
-		       const struct cxgb4_uld_info *p)
+#if IS_ENABLED(CONFIG_CHELSIO_TLS_DEVICE)
+static bool cxgb4_uld_in_use(struct adapter *adap)
 {
-	int ret = 0;
-	unsigned int adap_idx = 0;
-	struct adapter *adap;
+	const struct tid_info *t = &adap->tids;
 
-	if (type >= CXGB4_ULD_MAX)
-		return -EINVAL;
+	return (atomic_read(&t->conns_in_use) || t->stids_in_use);
+}
 
-	mutex_lock(&uld_mutex);
-	list_for_each_entry(adap, &adapter_list, list_node) {
-		if ((type == CXGB4_ULD_CRYPTO && !is_pci_uld(adap)) ||
-		    (type != CXGB4_ULD_CRYPTO && !is_offload(adap)))
-			continue;
-		if (type == CXGB4_ULD_ISCSIT && is_t4(adap->params.chip))
-			continue;
-		ret = cfg_queues_uld(adap, type, p);
-		if (ret)
-			goto out;
-		ret = setup_sge_queues_uld(adap, type, p->lro);
-		if (ret)
-			goto free_queues;
-		if (adap->flags & USING_MSIX) {
-			name_msix_vecs_uld(adap, type);
-			ret = request_msix_queue_irqs_uld(adap, type);
+/* cxgb4_set_ktls_feature: request FW to enable/disable ktls settings.
+ * @adap: adapter info
+ * @enable: 1 to enable / 0 to disable ktls settings.
+ */
+int cxgb4_set_ktls_feature(struct adapter *adap, bool enable)
+{
+	int ret = 0;
+	u32 params =
+		FW_PARAMS_MNEM_V(FW_PARAMS_MNEM_DEV) |
+		FW_PARAMS_PARAM_X_V(FW_PARAMS_PARAM_DEV_KTLS_HW) |
+		FW_PARAMS_PARAM_Y_V(enable) |
+		FW_PARAMS_PARAM_Z_V(FW_PARAMS_PARAM_DEV_KTLS_HW_USER_ENABLE);
+
+	if (enable) {
+		if (!refcount_read(&adap->chcr_ktls.ktls_refcount)) {
+			/* At this moment if ULD connection are up means, other
+			 * ULD is/are already active, return failure.
+			 */
+			if (cxgb4_uld_in_use(adap)) {
+				dev_dbg(adap->pdev_dev,
+					"ULD connections (tid/stid) active. Can't enable kTLS\n");
+				return -EINVAL;
+			}
+			ret = t4_set_params(adap, adap->mbox, adap->pf,
+					    0, 1, &params, &params);
 			if (ret)
-				goto free_rxq;
+				return ret;
+			refcount_set(&adap->chcr_ktls.ktls_refcount, 1);
+			pr_debug("kTLS has been enabled. Restrictions placed on ULD support\n");
+		} else {
+			/* ktls settings already up, just increment refcount. */
+			refcount_inc(&adap->chcr_ktls.ktls_refcount);
 		}
-		if (adap->flags & FULL_INIT_DONE)
-			enable_rx_uld(adap, type);
-		if (adap->uld[type].add) {
-			ret = -EBUSY;
-			goto free_irq;
+	} else {
+		/* return failure if refcount is already 0. */
+		if (!refcount_read(&adap->chcr_ktls.ktls_refcount))
+			return -EINVAL;
+		/* decrement refcount and test, if 0, disable ktls feature,
+		 * else return command success.
+		 */
+		if (refcount_dec_and_test(&adap->chcr_ktls.ktls_refcount)) {
+			ret = t4_set_params(adap, adap->mbox, adap->pf,
+					    0, 1, &params, &params);
+			if (ret)
+				return ret;
+			pr_debug("kTLS is disabled. Restrictions on ULD support removed\n");
 		}
-		ret = setup_sge_txq_uld(adap, type, p);
-		if (ret)
-			goto free_irq;
-		adap->uld[type] = *p;
-		uld_attach(adap, type);
-		adap_idx++;
 	}
-	mutex_unlock(&uld_mutex);
-	return 0;
 
+	return ret;
+}
+#endif
+
+static void cxgb4_uld_alloc_resources(struct adapter *adap,
+				      enum cxgb4_uld type,
+				      const struct cxgb4_uld_info *p)
+{
+	int ret = 0;
+
+	if ((type == CXGB4_ULD_CRYPTO && !is_pci_uld(adap)) ||
+	    (type != CXGB4_ULD_CRYPTO && !is_offload(adap)))
+		return;
+	if (type == CXGB4_ULD_ISCSIT && is_t4(adap->params.chip))
+		return;
+	ret = cfg_queues_uld(adap, type, p);
+	if (ret)
+		goto out;
+	ret = setup_sge_queues_uld(adap, type, p->lro);
+	if (ret)
+		goto free_queues;
+	if (adap->flags & CXGB4_USING_MSIX) {
+		ret = request_msix_queue_irqs_uld(adap, type);
+		if (ret)
+			goto free_rxq;
+	}
+	if (adap->flags & CXGB4_FULL_INIT_DONE)
+		enable_rx_uld(adap, type);
+	if (adap->uld[type].add)
+		goto free_irq;
+	ret = setup_sge_txq_uld(adap, type, p);
+	if (ret)
+		goto free_irq;
+	adap->uld[type] = *p;
+	ret = uld_attach(adap, type);
+	if (ret)
+		goto free_txq;
+	return;
+free_txq:
+	release_sge_txq_uld(adap, type);
 free_irq:
-	if (adap->flags & FULL_INIT_DONE)
+	if (adap->flags & CXGB4_FULL_INIT_DONE)
 		quiesce_rx_uld(adap, type);
-	if (adap->flags & USING_MSIX)
+	if (adap->flags & CXGB4_USING_MSIX)
 		free_msix_queue_irqs_uld(adap, type);
 free_rxq:
 	free_sge_queues_uld(adap, type);
 free_queues:
 	free_queues_uld(adap, type);
 out:
+	dev_warn(adap->pdev_dev,
+		 "ULD registration failed for uld type %d\n", type);
+}
 
-	list_for_each_entry(adap, &adapter_list, list_node) {
-		if ((type == CXGB4_ULD_CRYPTO && !is_pci_uld(adap)) ||
-		    (type != CXGB4_ULD_CRYPTO && !is_offload(adap)))
-			continue;
-		if (type == CXGB4_ULD_ISCSIT && is_t4(adap->params.chip))
-			continue;
-		if (!adap_idx)
-			break;
-		adap->uld[type].handle = NULL;
-		adap->uld[type].add = NULL;
-		release_sge_txq_uld(adap, type);
-		if (adap->flags & FULL_INIT_DONE)
-			quiesce_rx_uld(adap, type);
-		if (adap->flags & USING_MSIX)
-			free_msix_queue_irqs_uld(adap, type);
-		free_sge_queues_uld(adap, type);
-		free_queues_uld(adap, type);
-		adap_idx--;
-	}
+void cxgb4_uld_enable(struct adapter *adap)
+{
+	struct cxgb4_uld_list *uld_entry;
+
+	mutex_lock(&uld_mutex);
+	list_add_tail(&adap->list_node, &adapter_list);
+	list_for_each_entry(uld_entry, &uld_list, list_node)
+		cxgb4_uld_alloc_resources(adap, uld_entry->uld_type,
+					  &uld_entry->uld_info);
 	mutex_unlock(&uld_mutex);
-	return ret;
+}
+
+/* cxgb4_register_uld - register an upper-layer driver
+ * @type: the ULD type
+ * @p: the ULD methods
+ *
+ * Registers an upper-layer driver with this driver and notifies the ULD
+ * about any presently available devices that support its type.
+ */
+void cxgb4_register_uld(enum cxgb4_uld type,
+			const struct cxgb4_uld_info *p)
+{
+	struct cxgb4_uld_list *uld_entry;
+	struct adapter *adap;
+
+	if (type >= CXGB4_ULD_MAX)
+		return;
+
+	uld_entry = kzalloc(sizeof(*uld_entry), GFP_KERNEL);
+	if (!uld_entry)
+		return;
+
+	memcpy(&uld_entry->uld_info, p, sizeof(struct cxgb4_uld_info));
+	mutex_lock(&uld_mutex);
+	list_for_each_entry(adap, &adapter_list, list_node)
+		cxgb4_uld_alloc_resources(adap, type, p);
+
+	uld_entry->uld_type = type;
+	list_add_tail(&uld_entry->list_node, &uld_list);
+	mutex_unlock(&uld_mutex);
+	return;
 }
 EXPORT_SYMBOL(cxgb4_register_uld);
 
@@ -770,6 +829,7 @@ EXPORT_SYMBOL(cxgb4_register_uld);
  */
 int cxgb4_unregister_uld(enum cxgb4_uld type)
 {
+	struct cxgb4_uld_list *uld_entry, *tmp;
 	struct adapter *adap;
 
 	if (type >= CXGB4_ULD_MAX)
@@ -782,15 +842,15 @@ int cxgb4_unregister_uld(enum cxgb4_uld type)
 			continue;
 		if (type == CXGB4_ULD_ISCSIT && is_t4(adap->params.chip))
 			continue;
-		adap->uld[type].handle = NULL;
-		adap->uld[type].add = NULL;
-		release_sge_txq_uld(adap, type);
-		if (adap->flags & FULL_INIT_DONE)
-			quiesce_rx_uld(adap, type);
-		if (adap->flags & USING_MSIX)
-			free_msix_queue_irqs_uld(adap, type);
-		free_sge_queues_uld(adap, type);
-		free_queues_uld(adap, type);
+
+		cxgb4_shutdown_uld_adapter(adap, type);
+	}
+
+	list_for_each_entry_safe(uld_entry, tmp, &uld_list, list_node) {
+		if (uld_entry->uld_type == type) {
+			list_del(&uld_entry->list_node);
+			kfree(uld_entry);
+		}
 	}
 	mutex_unlock(&uld_mutex);
 
diff --git a/drivers/net/ethernet/chelsio/cxgb4/cxgb4_uld.h b/drivers/net/ethernet/chelsio/cxgb4/cxgb4_uld.h
index 6e74040af49a..d7713038386c 100644
--- a/drivers/net/ethernet/chelsio/cxgb4/cxgb4_uld.h
+++ b/drivers/net/ethernet/chelsio/cxgb4/cxgb4_uld.h
@@ -40,9 +40,14 @@
 #include <linux/skbuff.h>
 #include <linux/inetdevice.h>
 #include <linux/atomic.h>
+#include <net/tls.h>
 #include "cxgb4.h"
 
 #define MAX_ULD_QSETS 16
+#define MAX_ULD_NPORTS 4
+
+/* ulp_mem_io + ulptx_idata + payload + padding */
+#define MAX_IMM_ULPTX_WR_LEN (32 + 8 + 256 + 8)
 
 /* CPL message priority levels */
 enum {
@@ -89,24 +94,36 @@ union aopen_entry {
 	union aopen_entry *next;
 };
 
+struct eotid_entry {
+	void *data;
+};
+
 /*
  * Holds the size, base address, free list start, etc of the TID, server TID,
  * and active-open TID tables.  The tables themselves are allocated dynamically.
  */
 struct tid_info {
 	void **tid_tab;
+	unsigned int tid_base;
 	unsigned int ntids;
 
 	struct serv_entry *stid_tab;
 	unsigned long *stid_bmap;
 	unsigned int nstids;
 	unsigned int stid_base;
+
+	unsigned int nhash;
 	unsigned int hash_base;
 
 	union aopen_entry *atid_tab;
 	unsigned int natids;
 	unsigned int atid_base;
 
+	struct filter_entry *hpftid_tab;
+	unsigned long *hpftid_bmap;
+	unsigned int nhpftids;
+	unsigned int hpftid_base;
+
 	struct filter_entry *ftid_tab;
 	unsigned long *ftid_bmap;
 	unsigned int nftids;
@@ -123,21 +140,40 @@ struct tid_info {
 
 	spinlock_t stid_lock;
 	unsigned int stids_in_use;
+	unsigned int v6_stids_in_use;
 	unsigned int sftids_in_use;
 
+	/* ETHOFLD range */
+	struct eotid_entry *eotid_tab;
+	unsigned long *eotid_bmap;
+	unsigned int eotid_base;
+	unsigned int neotids;
+
 	/* TIDs in the TCAM */
 	atomic_t tids_in_use;
 	/* TIDs in the HASH */
 	atomic_t hash_tids_in_use;
+	atomic_t conns_in_use;
+	/* ETHOFLD TIDs used for rate limiting */
+	atomic_t eotids_in_use;
+
 	/* lock for setting/clearing filter bitmap */
 	spinlock_t ftid_lock;
+
+	unsigned int tc_hash_tids_max_prio;
 };
 
 static inline void *lookup_tid(const struct tid_info *t, unsigned int tid)
 {
+	tid -= t->tid_base;
 	return tid < t->ntids ? t->tid_tab[tid] : NULL;
 }
 
+static inline bool tid_out_of_range(const struct tid_info *t, unsigned int tid)
+{
+	return ((tid - t->tid_base) >= t->ntids);
+}
+
 static inline void *lookup_atid(const struct tid_info *t, unsigned int atid)
 {
 	return atid < t->natids ? t->atid_tab[atid].data : NULL;
@@ -157,13 +193,52 @@ static inline void *lookup_stid(const struct tid_info *t, unsigned int stid)
 }
 
 static inline void cxgb4_insert_tid(struct tid_info *t, void *data,
-				    unsigned int tid)
+				    unsigned int tid, unsigned short family)
+{
+	t->tid_tab[tid - t->tid_base] = data;
+	if (t->hash_base && (tid >= t->hash_base)) {
+		if (family == AF_INET6)
+			atomic_add(2, &t->hash_tids_in_use);
+		else
+			atomic_inc(&t->hash_tids_in_use);
+	} else {
+		if (family == AF_INET6)
+			atomic_add(2, &t->tids_in_use);
+		else
+			atomic_inc(&t->tids_in_use);
+	}
+	atomic_inc(&t->conns_in_use);
+}
+
+static inline struct eotid_entry *cxgb4_lookup_eotid(struct tid_info *t,
+						     u32 eotid)
+{
+	return eotid < t->neotids ? &t->eotid_tab[eotid] : NULL;
+}
+
+static inline int cxgb4_get_free_eotid(struct tid_info *t)
 {
-	t->tid_tab[tid] = data;
-	if (t->hash_base && (tid >= t->hash_base))
-		atomic_inc(&t->hash_tids_in_use);
-	else
-		atomic_inc(&t->tids_in_use);
+	int eotid;
+
+	eotid = find_first_zero_bit(t->eotid_bmap, t->neotids);
+	if (eotid >= t->neotids)
+		eotid = -1;
+
+	return eotid;
+}
+
+static inline void cxgb4_alloc_eotid(struct tid_info *t, u32 eotid, void *data)
+{
+	set_bit(eotid, t->eotid_bmap);
+	t->eotid_tab[eotid].data = data;
+	atomic_inc(&t->eotids_in_use);
+}
+
+static inline void cxgb4_free_eotid(struct tid_info *t, u32 eotid)
+{
+	clear_bit(eotid, t->eotid_bmap);
+	t->eotid_tab[eotid].data = NULL;
+	atomic_dec(&t->eotids_in_use);
 }
 
 int cxgb4_alloc_atid(struct tid_info *t, void *data);
@@ -171,8 +246,8 @@ int cxgb4_alloc_stid(struct tid_info *t, int family, void *data);
 int cxgb4_alloc_sftid(struct tid_info *t, int family, void *data);
 void cxgb4_free_atid(struct tid_info *t, unsigned int atid);
 void cxgb4_free_stid(struct tid_info *t, unsigned int stid, int family);
-void cxgb4_remove_tid(struct tid_info *t, unsigned int qid, unsigned int tid);
-
+void cxgb4_remove_tid(struct tid_info *t, unsigned int qid, unsigned int tid,
+		      unsigned short family);
 struct in6_addr;
 
 int cxgb4_create_server(const struct net_device *dev, unsigned int stid,
@@ -200,16 +275,26 @@ struct filter_ctx {
 	u32 tid;			/* to store tid */
 };
 
+struct chcr_ktls {
+	refcount_t ktls_refcount;
+};
+
 struct ch_filter_specification;
 
+int cxgb4_get_free_ftid(struct net_device *dev, u8 family, bool hash_en,
+			u32 tc_prio);
 int __cxgb4_set_filter(struct net_device *dev, int filter_id,
 		       struct ch_filter_specification *fs,
 		       struct filter_ctx *ctx);
 int __cxgb4_del_filter(struct net_device *dev, int filter_id,
+		       struct ch_filter_specification *fs,
 		       struct filter_ctx *ctx);
 int cxgb4_set_filter(struct net_device *dev, int filter_id,
 		     struct ch_filter_specification *fs);
-int cxgb4_del_filter(struct net_device *dev, int filter_id);
+int cxgb4_del_filter(struct net_device *dev, int filter_id,
+		     struct ch_filter_specification *fs);
+int cxgb4_get_filter_counters(struct net_device *dev, unsigned int fidx,
+			      u64 *hitcnt, u64 *bytecnt, bool hash);
 
 static inline void set_wr_txq(struct sk_buff *skb, int prio, int queue)
 {
@@ -222,6 +307,9 @@ enum cxgb4_uld {
 	CXGB4_ULD_ISCSI,
 	CXGB4_ULD_ISCSIT,
 	CXGB4_ULD_CRYPTO,
+	CXGB4_ULD_IPSEC,
+	CXGB4_ULD_TLS,
+	CXGB4_ULD_KTLS,
 	CXGB4_ULD_MAX
 };
 
@@ -242,7 +330,8 @@ enum cxgb4_state {
 	CXGB4_STATE_UP,
 	CXGB4_STATE_START_RECOVERY,
 	CXGB4_STATE_DOWN,
-	CXGB4_STATE_DETACH
+	CXGB4_STATE_DETACH,
+	CXGB4_STATE_FATAL_ERROR
 };
 
 enum cxgb4_control {
@@ -251,6 +340,7 @@ enum cxgb4_control {
 	CXGB4_CONTROL_DB_DROP,
 };
 
+struct adapter;
 struct pci_dev;
 struct l2t_data;
 struct net_device;
@@ -268,12 +358,60 @@ struct cxgb4_virt_res {                      /* virtualized HW resources */
 	struct cxgb4_range iscsi;
 	struct cxgb4_range stag;
 	struct cxgb4_range rq;
+	struct cxgb4_range srq;
 	struct cxgb4_range pbl;
 	struct cxgb4_range qp;
 	struct cxgb4_range cq;
 	struct cxgb4_range ocq;
+	struct cxgb4_range key;
 	unsigned int ncrypto_fc;
+	struct cxgb4_range ppod_edram;
+};
+
+#if IS_ENABLED(CONFIG_CHELSIO_TLS_DEVICE)
+struct ch_ktls_port_stats_debug {
+	atomic64_t ktls_tx_connection_open;
+	atomic64_t ktls_tx_connection_fail;
+	atomic64_t ktls_tx_connection_close;
+	atomic64_t ktls_tx_encrypted_packets;
+	atomic64_t ktls_tx_encrypted_bytes;
+	atomic64_t ktls_tx_ctx;
+	atomic64_t ktls_tx_ooo;
+	atomic64_t ktls_tx_skip_no_sync_data;
+	atomic64_t ktls_tx_drop_no_sync_data;
+	atomic64_t ktls_tx_drop_bypass_req;
+};
+
+struct ch_ktls_stats_debug {
+	struct ch_ktls_port_stats_debug ktls_port[MAX_ULD_NPORTS];
+	atomic64_t ktls_tx_send_records;
+	atomic64_t ktls_tx_end_pkts;
+	atomic64_t ktls_tx_start_pkts;
+	atomic64_t ktls_tx_middle_pkts;
+	atomic64_t ktls_tx_retransmit_pkts;
+	atomic64_t ktls_tx_complete_pkts;
+	atomic64_t ktls_tx_trimmed_pkts;
+	atomic64_t ktls_tx_fallback;
+};
+#endif
+
+struct chcr_stats_debug {
+	atomic_t cipher_rqst;
+	atomic_t digest_rqst;
+	atomic_t aead_rqst;
+	atomic_t complete;
+	atomic_t error;
+	atomic_t fallback;
+	atomic_t tls_pdu_tx;
+	atomic_t tls_pdu_rx;
+	atomic_t tls_key;
+};
+
+#if IS_ENABLED(CONFIG_CHELSIO_IPSEC_INLINE)
+struct ch_ipsec_stats_debug {
+	atomic_t ipsec_cnt;
 };
+#endif
 
 #define OCQ_WIN_OFFSET(pdev, vres) \
 	(pci_resource_len((pdev), 2) - roundup_pow_of_two((vres)->ocq.size))
@@ -298,11 +436,13 @@ struct cxgb4_lld_info {
 	unsigned char wr_cred;               /* WR 16-byte credits */
 	unsigned char adapter_type;          /* type of adapter */
 	unsigned char fw_api_ver;            /* FW API version */
+	unsigned char crypto;                /* crypto support */
 	unsigned int fw_vers;                /* FW version */
 	unsigned int iscsi_iolen;            /* iSCSI max I/O length */
 	unsigned int cclk_ps;                /* Core clock period in psec */
 	unsigned short udb_density;          /* # of user DB/page */
 	unsigned short ucq_density;          /* # of user CQs/page */
+	unsigned int sge_host_page_size;     /* SGE host page size */
 	unsigned short filt_mode;            /* filter optional components */
 	unsigned short tx_modq[NCHAN];       /* maps each tx channel to a */
 					     /* scheduler queue */
@@ -322,13 +462,16 @@ struct cxgb4_lld_info {
 	unsigned int iscsi_tagmask;	     /* iscsi ddp tag mask */
 	unsigned int iscsi_pgsz_order;	     /* iscsi ddp page size orders */
 	unsigned int iscsi_llimit;	     /* chip's iscsi region llimit */
+	unsigned int ulp_crypto;             /* crypto lookaside support */
 	void **iscsi_ppm;		     /* iscsi page pod manager */
 	int nodeid;			     /* device numa node id */
 	bool fr_nsmr_tpte_wr_support;	     /* FW supports FR_NSMR_TPTE_WR */
+	bool write_w_imm_support;         /* FW supports WRITE_WITH_IMMEDIATE */
+	bool write_cmpl_support;             /* FW supports WRITE_CMPL WR */
 };
 
 struct cxgb4_uld_info {
-	const char *name;
+	char name[IFNAMSIZ];
 	void *handle;
 	unsigned int nrxq;
 	unsigned int rxq_size;
@@ -345,16 +488,26 @@ struct cxgb4_uld_info {
 			      struct t4_lro_mgr *lro_mgr,
 			      struct napi_struct *napi);
 	void (*lro_flush)(struct t4_lro_mgr *);
+	int (*tx_handler)(struct sk_buff *skb, struct net_device *dev);
+#if IS_ENABLED(CONFIG_CHELSIO_TLS_DEVICE)
+	const struct tlsdev_ops *tlsdev_ops;
+#endif
+#if IS_ENABLED(CONFIG_XFRM_OFFLOAD)
+	const struct xfrmdev_ops *xfrmdev_ops;
+#endif
 };
 
-int cxgb4_register_uld(enum cxgb4_uld type, const struct cxgb4_uld_info *p);
+void cxgb4_uld_enable(struct adapter *adap);
+void cxgb4_register_uld(enum cxgb4_uld type, const struct cxgb4_uld_info *p);
 int cxgb4_unregister_uld(enum cxgb4_uld type);
 int cxgb4_ofld_send(struct net_device *dev, struct sk_buff *skb);
+int cxgb4_immdata_send(struct net_device *dev, unsigned int idx,
+		       const void *src, unsigned int len);
 int cxgb4_crypto_send(struct net_device *dev, struct sk_buff *skb);
 unsigned int cxgb4_dbfifo_count(const struct net_device *dev, int lpfifo);
 unsigned int cxgb4_port_chan(const struct net_device *dev);
+unsigned int cxgb4_port_e2cchan(const struct net_device *dev);
 unsigned int cxgb4_port_viid(const struct net_device *dev);
-unsigned int cxgb4_tp_smt_idx(enum chip_type chip, unsigned int viid);
 unsigned int cxgb4_port_idx(const struct net_device *dev);
 unsigned int cxgb4_best_mtu(const unsigned short *mtus, unsigned short mtu,
 			    unsigned int *idx);
@@ -365,8 +518,6 @@ unsigned int cxgb4_best_aligned_mtu(const unsigned short *mtus,
 				    unsigned int *mtu_idxp);
 void cxgb4_get_tcp_stats(struct pci_dev *pdev, struct tp_tcp_stats *v4,
 			 struct tp_tcp_stats *v6);
-void cxgb4_iscsi_init(struct net_device *dev, unsigned int tag_mask,
-		      const unsigned int *pgsz_order);
 struct sk_buff *cxgb4_pktgl_to_skb(const struct pkt_gl *gl,
 				   unsigned int skb_len, unsigned int pull_len);
 int cxgb4_sync_txq_pidx(struct net_device *dev, u16 qid, u16 pidx, u16 size);
diff --git a/drivers/net/ethernet/chelsio/cxgb4/l2t.c b/drivers/net/ethernet/chelsio/cxgb4/l2t.c
index 6f3692db29af..c02b4e9c06b2 100644
--- a/drivers/net/ethernet/chelsio/cxgb4/l2t.c
+++ b/drivers/net/ethernet/chelsio/cxgb4/l2t.c
@@ -59,7 +59,7 @@ struct l2t_data {
 	rwlock_t lock;
 	atomic_t nfree;             /* number of free entries */
 	struct l2t_entry *rover;    /* starting point for next allocation */
-	struct l2t_entry l2tab[0];  /* MUST BE LAST */
+	struct l2t_entry l2tab[] __counted_by(l2t_size);  /* MUST BE LAST */
 };
 
 static inline unsigned int vlan_prio(const struct l2t_entry *e)
@@ -146,7 +146,7 @@ static int write_l2e(struct adapter *adap, struct l2t_entry *e, int sync)
 	if (!skb)
 		return -ENOMEM;
 
-	req = (struct cpl_l2t_write_req *)__skb_put(skb, sizeof(*req));
+	req = __skb_put(skb, sizeof(*req));
 	INIT_TP_WR(req, 0);
 
 	OPCODE_TID(req) = htonl(MK_OPCODE_TID(CPL_L2T_WRITE_REQ,
@@ -231,6 +231,7 @@ again:
 		if (e->state == L2T_STATE_STALE)
 			e->state = L2T_STATE_VALID;
 		spin_unlock_bh(&e->lock);
+		fallthrough;
 	case L2T_STATE_VALID:     /* fast-path, send the packet on */
 		return t4_ofld_send(adap, skb);
 	case L2T_STATE_RESOLVING:
@@ -350,15 +351,13 @@ exists:
 static void _t4_l2e_free(struct l2t_entry *e)
 {
 	struct l2t_data *d;
-	struct sk_buff *skb;
 
 	if (atomic_read(&e->refcnt) == 0) {  /* hasn't been recycled */
 		if (e->neigh) {
 			neigh_release(e->neigh);
 			e->neigh = NULL;
 		}
-		while ((skb = __skb_dequeue(&e->arpq)) != NULL)
-			kfree_skb(skb);
+		__skb_queue_purge(&e->arpq);
 	}
 
 	d = container_of(e, struct l2t_data, l2tab[e->idx]);
@@ -369,7 +368,6 @@ static void _t4_l2e_free(struct l2t_entry *e)
 static void t4_l2e_free(struct l2t_entry *e)
 {
 	struct l2t_data *d;
-	struct sk_buff *skb;
 
 	spin_lock_bh(&e->lock);
 	if (atomic_read(&e->refcnt) == 0) {  /* hasn't been recycled */
@@ -377,8 +375,7 @@ static void t4_l2e_free(struct l2t_entry *e)
 			neigh_release(e->neigh);
 			e->neigh = NULL;
 		}
-		while ((skb = __skb_dequeue(&e->arpq)) != NULL)
-			kfree_skb(skb);
+		__skb_queue_purge(&e->arpq);
 	}
 	spin_unlock_bh(&e->lock);
 
@@ -422,7 +419,7 @@ struct l2t_entry *cxgb4_l2t_get(struct l2t_data *d, struct neighbour *neigh,
 	u8 lport;
 	u16 vlan;
 	struct l2t_entry *e;
-	int addr_len = neigh->tbl->key_len;
+	unsigned int addr_len = neigh->tbl->key_len;
 	u32 *addr = (u32 *)neigh->primary_key;
 	int ifidx = neigh->dev->ifindex;
 	int hash = addr_hash(d, addr, addr_len, ifidx);
@@ -432,10 +429,12 @@ struct l2t_entry *cxgb4_l2t_get(struct l2t_data *d, struct neighbour *neigh,
 	else
 		lport = netdev2pinfo(physdev)->lport;
 
-	if (is_vlan_dev(neigh->dev))
+	if (is_vlan_dev(neigh->dev)) {
 		vlan = vlan_dev_vlan_id(neigh->dev);
-	else
+		vlan |= vlan_dev_get_egress_qos_mask(neigh->dev, priority);
+	} else {
 		vlan = VLAN_NONE;
+	}
 
 	write_lock_bh(&d->lock);
 	for (e = d->l2tab[hash].first; e; e = e->next)
@@ -491,15 +490,12 @@ u64 cxgb4_select_ntuple(struct net_device *dev,
 	if (tp->protocol_shift >= 0)
 		ntuple |= (u64)IPPROTO_TCP << tp->protocol_shift;
 
-	if (tp->vnic_shift >= 0) {
-		u32 viid = cxgb4_port_viid(dev);
-		u32 vf = FW_VIID_VIN_G(viid);
-		u32 pf = FW_VIID_PFN_G(viid);
-		u32 vld = FW_VIID_VIVLD_G(viid);
+	if (tp->vnic_shift >= 0 && (tp->ingress_config & VNIC_F)) {
+		struct port_info *pi = (struct port_info *)netdev_priv(dev);
 
-		ntuple |= (u64)(FT_VNID_ID_VF_V(vf) |
-				FT_VNID_ID_PF_V(pf) |
-				FT_VNID_ID_VLD_V(vld)) << tp->vnic_shift;
+		ntuple |= (u64)(FT_VNID_ID_VF_V(pi->vin) |
+				FT_VNID_ID_PF_V(adap->pf) |
+				FT_VNID_ID_VLD_V(pi->vivld)) << tp->vnic_shift;
 	}
 
 	return ntuple;
@@ -507,40 +503,19 @@ u64 cxgb4_select_ntuple(struct net_device *dev,
 EXPORT_SYMBOL(cxgb4_select_ntuple);
 
 /*
- * Called when address resolution fails for an L2T entry to handle packets
- * on the arpq head.  If a packet specifies a failure handler it is invoked,
- * otherwise the packet is sent to the device.
- */
-static void handle_failed_resolution(struct adapter *adap, struct l2t_entry *e)
-{
-	struct sk_buff *skb;
-
-	while ((skb = __skb_dequeue(&e->arpq)) != NULL) {
-		const struct l2t_skb_cb *cb = L2T_SKB_CB(skb);
-
-		spin_unlock(&e->lock);
-		if (cb->arp_err_handler)
-			cb->arp_err_handler(cb->handle, skb);
-		else
-			t4_ofld_send(adap, skb);
-		spin_lock(&e->lock);
-	}
-}
-
-/*
  * Called when the host's neighbor layer makes a change to some entry that is
  * loaded into the HW L2 table.
  */
 void t4_l2t_update(struct adapter *adap, struct neighbour *neigh)
 {
-	struct l2t_entry *e;
+	unsigned int addr_len = neigh->tbl->key_len;
+	u32 *addr = (u32 *) neigh->primary_key;
+	int hash, ifidx = neigh->dev->ifindex;
 	struct sk_buff_head *arpq = NULL;
 	struct l2t_data *d = adap->l2t;
-	int addr_len = neigh->tbl->key_len;
-	u32 *addr = (u32 *) neigh->primary_key;
-	int ifidx = neigh->dev->ifindex;
-	int hash = addr_hash(d, addr, addr_len, ifidx);
+	struct l2t_entry *e;
 
+	hash = addr_hash(d, addr, addr_len, ifidx);
 	read_lock_bh(&d->lock);
 	for (e = d->l2tab[hash].first; e; e = e->next)
 		if (!addreq(e, addr) && e->ifindex == ifidx) {
@@ -573,8 +548,25 @@ void t4_l2t_update(struct adapter *adap, struct neighbour *neigh)
 			write_l2e(adap, e, 0);
 	}
 
-	if (arpq)
-		handle_failed_resolution(adap, e);
+	if (arpq) {
+		struct sk_buff *skb;
+
+		/* Called when address resolution fails for an L2T
+		 * entry to handle packets on the arpq head. If a
+		 * packet specifies a failure handler it is invoked,
+		 * otherwise the packet is sent to the device.
+		 */
+		while ((skb = __skb_dequeue(&e->arpq)) != NULL) {
+			const struct l2t_skb_cb *cb = L2T_SKB_CB(skb);
+
+			spin_unlock(&e->lock);
+			if (cb->arp_err_handler)
+				cb->arp_err_handler(cb->handle, skb);
+			else
+				t4_ofld_send(adap, skb);
+			spin_lock(&e->lock);
+		}
+	}
 	spin_unlock_bh(&e->lock);
 }
 
@@ -616,24 +608,6 @@ struct l2t_entry *t4_l2t_alloc_switching(struct adapter *adap, u16 vlan,
 	return e;
 }
 
-/**
- * @dev: net_device pointer
- * @vlan: VLAN Id
- * @port: Associated port
- * @dmac: Destination MAC address to add to L2T
- * Returns pointer to the allocated l2t entry
- *
- * Allocates an L2T entry for use by switching rule of a filter
- */
-struct l2t_entry *cxgb4_l2t_alloc_switching(struct net_device *dev, u16 vlan,
-					    u8 port, u8 *dmac)
-{
-	struct adapter *adap = netdev2adap(dev);
-
-	return t4_l2t_alloc_switching(adap, vlan, port, dmac);
-}
-EXPORT_SYMBOL(cxgb4_l2t_alloc_switching);
-
 struct l2t_data *t4_init_l2t(unsigned int l2t_start, unsigned int l2t_end)
 {
 	unsigned int l2t_size;
@@ -646,7 +620,7 @@ struct l2t_data *t4_init_l2t(unsigned int l2t_start, unsigned int l2t_end)
 	if (l2t_size < L2T_MIN_HASH_BUCKETS)
 		return NULL;
 
-	d = kvzalloc(sizeof(*d) + l2t_size * sizeof(struct l2t_entry), GFP_KERNEL);
+	d = kvzalloc(struct_size(d, l2tab, l2t_size), GFP_KERNEL);
 	if (!d)
 		return NULL;
 
@@ -682,8 +656,7 @@ static void *l2t_seq_start(struct seq_file *seq, loff_t *pos)
 static void *l2t_seq_next(struct seq_file *seq, void *v, loff_t *pos)
 {
 	v = l2t_get_idx(seq, *pos);
-	if (v)
-		++*pos;
+	++(*pos);
 	return v;
 }
 
@@ -705,6 +678,17 @@ static char l2e_state(const struct l2t_entry *e)
 	}
 }
 
+bool cxgb4_check_l2t_valid(struct l2t_entry *e)
+{
+	bool valid;
+
+	spin_lock(&e->lock);
+	valid = (e->state == L2T_STATE_VALID);
+	spin_unlock(&e->lock);
+	return valid;
+}
+EXPORT_SYMBOL(cxgb4_check_l2t_valid);
+
 static int l2t_seq_show(struct seq_file *seq, void *v)
 {
 	if (v == SEQ_START_TOKEN)
diff --git a/drivers/net/ethernet/chelsio/cxgb4/l2t.h b/drivers/net/ethernet/chelsio/cxgb4/l2t.h
index 79665bd8f881..8aad7e9dee6d 100644
--- a/drivers/net/ethernet/chelsio/cxgb4/l2t.h
+++ b/drivers/net/ethernet/chelsio/cxgb4/l2t.h
@@ -115,13 +115,12 @@ struct l2t_entry *cxgb4_l2t_get(struct l2t_data *d, struct neighbour *neigh,
 				unsigned int priority);
 u64 cxgb4_select_ntuple(struct net_device *dev,
 			const struct l2t_entry *l2t);
-struct l2t_entry *cxgb4_l2t_alloc_switching(struct net_device *dev, u16 vlan,
-					    u8 port, u8 *dmac);
 void t4_l2t_update(struct adapter *adap, struct neighbour *neigh);
 struct l2t_entry *t4_l2t_alloc_switching(struct adapter *adap, u16 vlan,
 					 u8 port, u8 *dmac);
 struct l2t_data *t4_init_l2t(unsigned int l2t_start, unsigned int l2t_end);
 void do_l2t_write_rpl(struct adapter *p, const struct cpl_l2t_write_rpl *rpl);
+bool cxgb4_check_l2t_valid(struct l2t_entry *e);
 
 extern const struct file_operations t4_l2t_fops;
 #endif  /* __CXGB4_L2T_H */
diff --git a/drivers/net/ethernet/chelsio/cxgb4/sched.c b/drivers/net/ethernet/chelsio/cxgb4/sched.c
index 02acff741f11..a1b14468d1ff 100644
--- a/drivers/net/ethernet/chelsio/cxgb4/sched.c
+++ b/drivers/net/ethernet/chelsio/cxgb4/sched.c
@@ -38,7 +38,6 @@
 #include "cxgb4.h"
 #include "sched.h"
 
-/* Spinlock must be held by caller */
 static int t4_sched_class_fw_cmd(struct port_info *pi,
 				 struct ch_sched_params *p,
 				 enum sched_fw_ops op)
@@ -51,13 +50,15 @@ static int t4_sched_class_fw_cmd(struct port_info *pi,
 	e = &s->tab[p->u.params.class];
 	switch (op) {
 	case SCHED_FW_OP_ADD:
+	case SCHED_FW_OP_DEL:
 		err = t4_sched_params(adap, p->type,
 				      p->u.params.level, p->u.params.mode,
 				      p->u.params.rateunit,
 				      p->u.params.ratemode,
 				      p->u.params.channel, e->idx,
 				      p->u.params.minrate, p->u.params.maxrate,
-				      p->u.params.weight, p->u.params.pktsize);
+				      p->u.params.weight, p->u.params.pktsize,
+				      p->u.params.burstsize);
 		break;
 	default:
 		err = -ENOTSUPP;
@@ -67,7 +68,6 @@ static int t4_sched_class_fw_cmd(struct port_info *pi,
 	return err;
 }
 
-/* Spinlock must be held by caller */
 static int t4_sched_bind_unbind_op(struct port_info *pi, void *arg,
 				   enum sched_bind_type type, bool bind)
 {
@@ -94,45 +94,69 @@ static int t4_sched_bind_unbind_op(struct port_info *pi, void *arg,
 
 		pf = adap->pf;
 		vf = 0;
+
+		err = t4_set_params(adap, adap->mbox, pf, vf, 1,
+				    &fw_param, &fw_class);
+		break;
+	}
+	case SCHED_FLOWC: {
+		struct sched_flowc_entry *fe;
+
+		fe = (struct sched_flowc_entry *)arg;
+
+		fw_class = bind ? fe->param.class : FW_SCHED_CLS_NONE;
+		err = cxgb4_ethofld_send_flowc(adap->port[pi->port_id],
+					       fe->param.tid, fw_class);
 		break;
 	}
 	default:
 		err = -ENOTSUPP;
-		goto out;
+		break;
 	}
 
-	err = t4_set_params(adap, adap->mbox, pf, vf, 1, &fw_param, &fw_class);
-
-out:
 	return err;
 }
 
-static struct sched_class *t4_sched_queue_lookup(struct port_info *pi,
-						 const unsigned int qid,
-						 int *index)
+static void *t4_sched_entry_lookup(struct port_info *pi,
+				   enum sched_bind_type type,
+				   const u32 val)
 {
 	struct sched_table *s = pi->sched_tbl;
 	struct sched_class *e, *end;
-	struct sched_class *found = NULL;
-	int i;
+	void *found = NULL;
 
-	/* Look for a class with matching bound queue parameters */
+	/* Look for an entry with matching @val */
 	end = &s->tab[s->sched_size];
 	for (e = &s->tab[0]; e != end; ++e) {
-		struct sched_queue_entry *qe;
-
-		i = 0;
-		if (e->state == SCHED_STATE_UNUSED)
+		if (e->state == SCHED_STATE_UNUSED ||
+		    e->bind_type != type)
 			continue;
 
-		list_for_each_entry(qe, &e->queue_list, list) {
-			if (qe->cntxt_id == qid) {
-				found = e;
-				if (index)
-					*index = i;
-				break;
+		switch (type) {
+		case SCHED_QUEUE: {
+			struct sched_queue_entry *qe;
+
+			list_for_each_entry(qe, &e->entry_list, list) {
+				if (qe->cntxt_id == val) {
+					found = qe;
+					break;
+				}
 			}
-			i++;
+			break;
+		}
+		case SCHED_FLOWC: {
+			struct sched_flowc_entry *fe;
+
+			list_for_each_entry(fe, &e->entry_list, list) {
+				if (fe->param.tid == val) {
+					found = fe;
+					break;
+				}
+			}
+			break;
+		}
+		default:
+			return NULL;
 		}
 
 		if (found)
@@ -142,59 +166,59 @@ static struct sched_class *t4_sched_queue_lookup(struct port_info *pi,
 	return found;
 }
 
-static int t4_sched_queue_unbind(struct port_info *pi, struct ch_sched_queue *p)
+struct sched_class *cxgb4_sched_queue_lookup(struct net_device *dev,
+					     struct ch_sched_queue *p)
 {
+	struct port_info *pi = netdev2pinfo(dev);
+	struct sched_queue_entry *qe = NULL;
 	struct adapter *adap = pi->adapter;
-	struct sched_class *e;
+	struct sge_eth_txq *txq;
+
+	if (p->queue < 0 || p->queue >= pi->nqsets)
+		return NULL;
+
+	txq = &adap->sge.ethtxq[pi->first_qset + p->queue];
+	qe = t4_sched_entry_lookup(pi, SCHED_QUEUE, txq->q.cntxt_id);
+	return qe ? &pi->sched_tbl->tab[qe->param.class] : NULL;
+}
+
+static int t4_sched_queue_unbind(struct port_info *pi, struct ch_sched_queue *p)
+{
 	struct sched_queue_entry *qe = NULL;
+	struct adapter *adap = pi->adapter;
 	struct sge_eth_txq *txq;
-	unsigned int qid;
-	int index = -1;
+	struct sched_class *e;
 	int err = 0;
 
 	if (p->queue < 0 || p->queue >= pi->nqsets)
 		return -ERANGE;
 
 	txq = &adap->sge.ethtxq[pi->first_qset + p->queue];
-	qid = txq->q.cntxt_id;
-
-	/* Find the existing class that the queue is bound to */
-	e = t4_sched_queue_lookup(pi, qid, &index);
-	if (e && index >= 0) {
-		int i = 0;
 
-		spin_lock(&e->lock);
-		list_for_each_entry(qe, &e->queue_list, list) {
-			if (i == index)
-				break;
-			i++;
-		}
+	/* Find the existing entry that the queue is bound to */
+	qe = t4_sched_entry_lookup(pi, SCHED_QUEUE, txq->q.cntxt_id);
+	if (qe) {
 		err = t4_sched_bind_unbind_op(pi, (void *)qe, SCHED_QUEUE,
 					      false);
-		if (err) {
-			spin_unlock(&e->lock);
-			goto out;
-		}
+		if (err)
+			return err;
 
+		e = &pi->sched_tbl->tab[qe->param.class];
 		list_del(&qe->list);
 		kvfree(qe);
-		if (atomic_dec_and_test(&e->refcnt)) {
-			e->state = SCHED_STATE_UNUSED;
-			memset(&e->info, 0, sizeof(e->info));
-		}
-		spin_unlock(&e->lock);
+		if (atomic_dec_and_test(&e->refcnt))
+			cxgb4_sched_class_free(adap->port[pi->port_id], e->idx);
 	}
-out:
 	return err;
 }
 
 static int t4_sched_queue_bind(struct port_info *pi, struct ch_sched_queue *p)
 {
-	struct adapter *adap = pi->adapter;
 	struct sched_table *s = pi->sched_tbl;
-	struct sched_class *e;
 	struct sched_queue_entry *qe = NULL;
+	struct adapter *adap = pi->adapter;
 	struct sge_eth_txq *txq;
+	struct sched_class *e;
 	unsigned int qid;
 	int err = 0;
 
@@ -210,29 +234,90 @@ static int t4_sched_queue_bind(struct port_info *pi, struct ch_sched_queue *p)
 
 	/* Unbind queue from any existing class */
 	err = t4_sched_queue_unbind(pi, p);
-	if (err) {
-		kvfree(qe);
-		goto out;
-	}
+	if (err)
+		goto out_err;
 
 	/* Bind queue to specified class */
-	memset(qe, 0, sizeof(*qe));
 	qe->cntxt_id = qid;
 	memcpy(&qe->param, p, sizeof(qe->param));
 
 	e = &s->tab[qe->param.class];
-	spin_lock(&e->lock);
 	err = t4_sched_bind_unbind_op(pi, (void *)qe, SCHED_QUEUE, true);
-	if (err) {
-		kvfree(qe);
-		spin_unlock(&e->lock);
-		goto out;
+	if (err)
+		goto out_err;
+
+	list_add_tail(&qe->list, &e->entry_list);
+	e->bind_type = SCHED_QUEUE;
+	atomic_inc(&e->refcnt);
+	return err;
+
+out_err:
+	kvfree(qe);
+	return err;
+}
+
+static int t4_sched_flowc_unbind(struct port_info *pi, struct ch_sched_flowc *p)
+{
+	struct sched_flowc_entry *fe = NULL;
+	struct adapter *adap = pi->adapter;
+	struct sched_class *e;
+	int err = 0;
+
+	if (p->tid < 0 || p->tid >= adap->tids.neotids)
+		return -ERANGE;
+
+	/* Find the existing entry that the flowc is bound to */
+	fe = t4_sched_entry_lookup(pi, SCHED_FLOWC, p->tid);
+	if (fe) {
+		err = t4_sched_bind_unbind_op(pi, (void *)fe, SCHED_FLOWC,
+					      false);
+		if (err)
+			return err;
+
+		e = &pi->sched_tbl->tab[fe->param.class];
+		list_del(&fe->list);
+		kvfree(fe);
+		if (atomic_dec_and_test(&e->refcnt))
+			cxgb4_sched_class_free(adap->port[pi->port_id], e->idx);
 	}
+	return err;
+}
 
-	list_add_tail(&qe->list, &e->queue_list);
+static int t4_sched_flowc_bind(struct port_info *pi, struct ch_sched_flowc *p)
+{
+	struct sched_table *s = pi->sched_tbl;
+	struct sched_flowc_entry *fe = NULL;
+	struct adapter *adap = pi->adapter;
+	struct sched_class *e;
+	int err = 0;
+
+	if (p->tid < 0 || p->tid >= adap->tids.neotids)
+		return -ERANGE;
+
+	fe = kvzalloc(sizeof(*fe), GFP_KERNEL);
+	if (!fe)
+		return -ENOMEM;
+
+	/* Unbind flowc from any existing class */
+	err = t4_sched_flowc_unbind(pi, p);
+	if (err)
+		goto out_err;
+
+	/* Bind flowc to specified class */
+	memcpy(&fe->param, p, sizeof(fe->param));
+
+	e = &s->tab[fe->param.class];
+	err = t4_sched_bind_unbind_op(pi, (void *)fe, SCHED_FLOWC, true);
+	if (err)
+		goto out_err;
+
+	list_add_tail(&fe->list, &e->entry_list);
+	e->bind_type = SCHED_FLOWC;
 	atomic_inc(&e->refcnt);
-	spin_unlock(&e->lock);
-out:
+	return err;
+
+out_err:
+	kvfree(fe);
 	return err;
 }
 
@@ -247,10 +332,17 @@ static void t4_sched_class_unbind_all(struct port_info *pi,
 	case SCHED_QUEUE: {
 		struct sched_queue_entry *qe;
 
-		list_for_each_entry(qe, &e->queue_list, list)
+		list_for_each_entry(qe, &e->entry_list, list)
 			t4_sched_queue_unbind(pi, &qe->param);
 		break;
 	}
+	case SCHED_FLOWC: {
+		struct sched_flowc_entry *fe;
+
+		list_for_each_entry(fe, &e->entry_list, list)
+			t4_sched_flowc_unbind(pi, &fe->param);
+		break;
+	}
 	default:
 		break;
 	}
@@ -274,6 +366,15 @@ static int t4_sched_class_bind_unbind_op(struct port_info *pi, void *arg,
 			err = t4_sched_queue_unbind(pi, qe);
 		break;
 	}
+	case SCHED_FLOWC: {
+		struct ch_sched_flowc *fe = (struct ch_sched_flowc *)arg;
+
+		if (bind)
+			err = t4_sched_flowc_bind(pi, fe);
+		else
+			err = t4_sched_flowc_unbind(pi, fe);
+		break;
+	}
 	default:
 		err = -ENOTSUPP;
 		break;
@@ -296,8 +397,6 @@ int cxgb4_sched_class_bind(struct net_device *dev, void *arg,
 			   enum sched_bind_type type)
 {
 	struct port_info *pi = netdev2pinfo(dev);
-	struct sched_table *s;
-	int err = 0;
 	u8 class_id;
 
 	if (!can_sched(dev))
@@ -313,6 +412,12 @@ int cxgb4_sched_class_bind(struct net_device *dev, void *arg,
 		class_id = qe->class;
 		break;
 	}
+	case SCHED_FLOWC: {
+		struct ch_sched_flowc *fe = (struct ch_sched_flowc *)arg;
+
+		class_id = fe->class;
+		break;
+	}
 	default:
 		return -ENOTSUPP;
 	}
@@ -323,12 +428,8 @@ int cxgb4_sched_class_bind(struct net_device *dev, void *arg,
 	if (class_id == SCHED_CLS_NONE)
 		return -ENOTSUPP;
 
-	s = pi->sched_tbl;
-	write_lock(&s->rw_lock);
-	err = t4_sched_class_bind_unbind_op(pi, arg, type, true);
-	write_unlock(&s->rw_lock);
+	return t4_sched_class_bind_unbind_op(pi, arg, type, true);
 
-	return err;
 }
 
 /**
@@ -343,8 +444,6 @@ int cxgb4_sched_class_unbind(struct net_device *dev, void *arg,
 			     enum sched_bind_type type)
 {
 	struct port_info *pi = netdev2pinfo(dev);
-	struct sched_table *s;
-	int err = 0;
 	u8 class_id;
 
 	if (!can_sched(dev))
@@ -360,6 +459,12 @@ int cxgb4_sched_class_unbind(struct net_device *dev, void *arg,
 		class_id = qe->class;
 		break;
 	}
+	case SCHED_FLOWC: {
+		struct ch_sched_flowc *fe = (struct ch_sched_flowc *)arg;
+
+		class_id = fe->class;
+		break;
+	}
 	default:
 		return -ENOTSUPP;
 	}
@@ -367,12 +472,7 @@ int cxgb4_sched_class_unbind(struct net_device *dev, void *arg,
 	if (!valid_class_id(dev, class_id))
 		return -EINVAL;
 
-	s = pi->sched_tbl;
-	write_lock(&s->rw_lock);
-	err = t4_sched_class_bind_unbind_op(pi, arg, type, false);
-	write_unlock(&s->rw_lock);
-
-	return err;
+	return t4_sched_class_bind_unbind_op(pi, arg, type, false);
 }
 
 /* If @p is NULL, fetch any available unused class */
@@ -380,8 +480,8 @@ static struct sched_class *t4_sched_class_lookup(struct port_info *pi,
 						const struct ch_sched_params *p)
 {
 	struct sched_table *s = pi->sched_tbl;
-	struct sched_class *e, *end;
 	struct sched_class *found = NULL;
+	struct sched_class *e, *end;
 
 	if (!p) {
 		/* Get any available unused class */
@@ -425,8 +525,7 @@ static struct sched_class *t4_sched_class_lookup(struct port_info *pi,
 static struct sched_class *t4_sched_class_alloc(struct port_info *pi,
 						struct ch_sched_params *p)
 {
-	struct sched_table *s = pi->sched_tbl;
-	struct sched_class *e;
+	struct sched_class *e = NULL;
 	u8 class_id;
 	int err;
 
@@ -441,38 +540,32 @@ static struct sched_class *t4_sched_class_alloc(struct port_info *pi,
 	if (class_id != SCHED_CLS_NONE)
 		return NULL;
 
-	write_lock(&s->rw_lock);
-	/* See if there's an exisiting class with same
-	 * requested sched params
+	/* See if there's an exisiting class with same requested sched
+	 * params. Classes can only be shared among FLOWC types. For
+	 * other types, always request a new class.
 	 */
-	e = t4_sched_class_lookup(pi, p);
+	if (p->u.params.mode == SCHED_CLASS_MODE_FLOW)
+		e = t4_sched_class_lookup(pi, p);
+
 	if (!e) {
 		struct ch_sched_params np;
 
 		/* Fetch any available unused class */
 		e = t4_sched_class_lookup(pi, NULL);
 		if (!e)
-			goto out;
+			return NULL;
 
 		memcpy(&np, p, sizeof(np));
 		np.u.params.class = e->idx;
-
-		spin_lock(&e->lock);
 		/* New class */
 		err = t4_sched_class_fw_cmd(pi, &np, SCHED_FW_OP_ADD);
-		if (err) {
-			spin_unlock(&e->lock);
-			e = NULL;
-			goto out;
-		}
+		if (err)
+			return NULL;
 		memcpy(&e->info, &np, sizeof(e->info));
 		atomic_set(&e->refcnt, 0);
 		e->state = SCHED_STATE_ACTIVE;
-		spin_unlock(&e->lock);
 	}
 
-out:
-	write_unlock(&s->rw_lock);
 	return e;
 }
 
@@ -502,9 +595,57 @@ struct sched_class *cxgb4_sched_class_alloc(struct net_device *dev,
 	return t4_sched_class_alloc(pi, p);
 }
 
-static void t4_sched_class_free(struct port_info *pi, struct sched_class *e)
+/**
+ * cxgb4_sched_class_free - free a scheduling class
+ * @dev: net_device pointer
+ * @classid: scheduling class id to free
+ *
+ * Frees a scheduling class if there are no users.
+ */
+void cxgb4_sched_class_free(struct net_device *dev, u8 classid)
+{
+	struct port_info *pi = netdev2pinfo(dev);
+	struct sched_table *s = pi->sched_tbl;
+	struct ch_sched_params p;
+	struct sched_class *e;
+	u32 speed;
+	int ret;
+
+	e = &s->tab[classid];
+	if (!atomic_read(&e->refcnt) && e->state != SCHED_STATE_UNUSED) {
+		/* Port based rate limiting needs explicit reset back
+		 * to max rate. But, we'll do explicit reset for all
+		 * types, instead of just port based type, to be on
+		 * the safer side.
+		 */
+		memcpy(&p, &e->info, sizeof(p));
+		/* Always reset mode to 0. Otherwise, FLOWC mode will
+		 * still be enabled even after resetting the traffic
+		 * class.
+		 */
+		p.u.params.mode = 0;
+		p.u.params.minrate = 0;
+		p.u.params.pktsize = 0;
+
+		ret = t4_get_link_params(pi, NULL, &speed, NULL);
+		if (!ret)
+			p.u.params.maxrate = speed * 1000; /* Mbps to Kbps */
+		else
+			p.u.params.maxrate = SCHED_MAX_RATE_KBPS;
+
+		t4_sched_class_fw_cmd(pi, &p, SCHED_FW_OP_DEL);
+
+		e->state = SCHED_STATE_UNUSED;
+		memset(&e->info, 0, sizeof(e->info));
+	}
+}
+
+static void t4_sched_class_free(struct net_device *dev, struct sched_class *e)
 {
-	t4_sched_class_unbind_all(pi, e, SCHED_QUEUE);
+	struct port_info *pi = netdev2pinfo(dev);
+
+	t4_sched_class_unbind_all(pi, e, e->bind_type);
+	cxgb4_sched_class_free(dev, e->idx);
 }
 
 struct sched_table *t4_init_sched(unsigned int sched_size)
@@ -512,19 +653,17 @@ struct sched_table *t4_init_sched(unsigned int sched_size)
 	struct sched_table *s;
 	unsigned int i;
 
-	s = kvzalloc(sizeof(*s) + sched_size * sizeof(struct sched_class), GFP_KERNEL);
+	s = kvzalloc(struct_size(s, tab, sched_size), GFP_KERNEL);
 	if (!s)
 		return NULL;
 
 	s->sched_size = sched_size;
-	rwlock_init(&s->rw_lock);
 
 	for (i = 0; i < s->sched_size; i++) {
 		memset(&s->tab[i], 0, sizeof(struct sched_class));
 		s->tab[i].idx = i;
 		s->tab[i].state = SCHED_STATE_UNUSED;
-		INIT_LIST_HEAD(&s->tab[i].queue_list);
-		spin_lock_init(&s->tab[i].lock);
+		INIT_LIST_HEAD(&s->tab[i].entry_list);
 		atomic_set(&s->tab[i].refcnt, 0);
 	}
 	return s;
@@ -533,20 +672,21 @@ struct sched_table *t4_init_sched(unsigned int sched_size)
 void t4_cleanup_sched(struct adapter *adap)
 {
 	struct sched_table *s;
-	unsigned int i;
+	unsigned int j, i;
 
-	for_each_port(adap, i) {
-		struct port_info *pi = netdev2pinfo(adap->port[i]);
+	for_each_port(adap, j) {
+		struct port_info *pi = netdev2pinfo(adap->port[j]);
 
 		s = pi->sched_tbl;
+		if (!s)
+			continue;
+
 		for (i = 0; i < s->sched_size; i++) {
 			struct sched_class *e;
 
-			write_lock(&s->rw_lock);
 			e = &s->tab[i];
 			if (e->state == SCHED_STATE_ACTIVE)
-				t4_sched_class_free(pi, e);
-			write_unlock(&s->rw_lock);
+				t4_sched_class_free(adap->port[j], e);
 		}
 		kvfree(s);
 	}
diff --git a/drivers/net/ethernet/chelsio/cxgb4/sched.h b/drivers/net/ethernet/chelsio/cxgb4/sched.h
index 77b2b3fd9021..6b3c778815f0 100644
--- a/drivers/net/ethernet/chelsio/cxgb4/sched.h
+++ b/drivers/net/ethernet/chelsio/cxgb4/sched.h
@@ -42,8 +42,8 @@
 
 #define FW_SCHED_CLS_NONE 0xffffffff
 
-/* Max rate that can be set to a scheduling class is 10 Gbps */
-#define SCHED_MAX_RATE_KBPS 10000000U
+/* Max rate that can be set to a scheduling class is 100 Gbps */
+#define SCHED_MAX_RATE_KBPS 100000000U
 
 enum {
 	SCHED_STATE_ACTIVE,
@@ -52,10 +52,12 @@ enum {
 
 enum sched_fw_ops {
 	SCHED_FW_OP_ADD,
+	SCHED_FW_OP_DEL,
 };
 
 enum sched_bind_type {
 	SCHED_QUEUE,
+	SCHED_FLOWC,
 };
 
 struct sched_queue_entry {
@@ -64,19 +66,23 @@ struct sched_queue_entry {
 	struct ch_sched_queue param;
 };
 
+struct sched_flowc_entry {
+	struct list_head list;
+	struct ch_sched_flowc param;
+};
+
 struct sched_class {
 	u8 state;
 	u8 idx;
 	struct ch_sched_params info;
-	struct list_head queue_list;
-	spinlock_t lock; /* Per class lock */
+	enum sched_bind_type bind_type;
+	struct list_head entry_list;
 	atomic_t refcnt;
 };
 
 struct sched_table {      /* per port scheduling table */
 	u8 sched_size;
-	rwlock_t rw_lock; /* Table lock */
-	struct sched_class tab[0];
+	struct sched_class tab[] __counted_by(sched_size);
 };
 
 static inline bool can_sched(struct net_device *dev)
@@ -97,6 +103,8 @@ static inline bool valid_class_id(struct net_device *dev, u8 class_id)
 	return true;
 }
 
+struct sched_class *cxgb4_sched_queue_lookup(struct net_device *dev,
+					     struct ch_sched_queue *p);
 int cxgb4_sched_class_bind(struct net_device *dev, void *arg,
 			   enum sched_bind_type type);
 int cxgb4_sched_class_unbind(struct net_device *dev, void *arg,
@@ -104,6 +112,7 @@ int cxgb4_sched_class_unbind(struct net_device *dev, void *arg,
 
 struct sched_class *cxgb4_sched_class_alloc(struct net_device *dev,
 					    struct ch_sched_params *p);
+void cxgb4_sched_class_free(struct net_device *dev, u8 classid);
 
 struct sched_table *t4_init_sched(unsigned int size);
 void t4_cleanup_sched(struct adapter *adap);
diff --git a/drivers/net/ethernet/chelsio/cxgb4/sge.c b/drivers/net/ethernet/chelsio/cxgb4/sge.c
index f05f0d400324..9fccb8ea9bcd 100644
--- a/drivers/net/ethernet/chelsio/cxgb4/sge.c
+++ b/drivers/net/ethernet/chelsio/cxgb4/sge.c
@@ -41,6 +41,7 @@
 #include <linux/jiffies.h>
 #include <linux/prefetch.h>
 #include <linux/export.h>
+#include <net/xfrm.h>
 #include <net/ipv6.h>
 #include <net/tcp.h>
 #include <net/busy_poll.h>
@@ -52,6 +53,10 @@
 #include "t4_values.h"
 #include "t4_msg.h"
 #include "t4fw_api.h"
+#include "cxgb4_ptp.h"
+#include "cxgb4_uld.h"
+#include "cxgb4_tc_mqprio.h"
+#include "sched.h"
 
 /*
  * Rx buffer size.  We use largish buffers if possible but settle for single
@@ -77,9 +82,10 @@
  * Max number of Tx descriptors we clean up at a time.  Should be modest as
  * freeing skbs isn't cheap and it happens while holding locks.  We just need
  * to free packets faster than they arrive, we eventually catch up and keep
- * the amortized cost reasonable.  Must be >= 2 * TXQ_STOP_THRES.
+ * the amortized cost reasonable.  Must be >= 2 * TXQ_STOP_THRES.  It should
+ * also match the CIDX Flush Threshold.
  */
-#define MAX_TX_RECLAIM 16
+#define MAX_TX_RECLAIM 32
 
 /*
  * Max number of Rx buffers we replenish at a time.  Again keep this modest,
@@ -109,14 +115,6 @@
 #define NOMEM_TMR_IDX (SGE_NTIMERS - 1)
 
 /*
- * Suspend an Ethernet Tx queue with fewer available descriptors than this.
- * This is the same as calc_tx_descs() for a TSO packet with
- * nr_frags == MAX_SKB_FRAGS.
- */
-#define ETHTXQ_STOP_THRES \
-	(1 + DIV_ROUND_UP((3 * MAX_SKB_FRAGS) / 2 + (MAX_SKB_FRAGS & 1), 8))
-
-/*
  * Suspension threshold for non-Ethernet Tx queues.  We require enough room
  * for a full sized WR.
  */
@@ -133,11 +131,6 @@
  */
 #define MAX_CTRL_WR_LEN SGE_MAX_WR_LEN
 
-struct tx_sw_desc {                /* SW state per Tx descriptor */
-	struct sk_buff *skb;
-	struct ulptx_sgl *sgl;
-};
-
 struct rx_sw_desc {                /* SW state per Rx descriptor */
 	struct page *page;
 	dma_addr_t dma_addr;
@@ -170,7 +163,7 @@ static inline unsigned int fl_mtu_bufsize(struct adapter *adapter,
  * for DMA, but this is of course never sent to the hardware and is only used
  * to prevent double unmappings.  All of the above requires that the Free List
  * Buffers which we allocate have the bottom 5 bits free (0) -- i.e. are
- * 32-byte or or a power of 2 greater in alignment.  Since the SGE's minimal
+ * 32-byte or a power of 2 greater in alignment.  Since the SGE's minimal
  * Free List Buffer alignment is 32 bytes, this works out for us ...
  */
 enum {
@@ -247,8 +240,8 @@ static inline bool fl_starving(const struct adapter *adapter,
 	return fl->avail - fl->pend_cred <= s->fl_starve_thres;
 }
 
-static int map_skb(struct device *dev, const struct sk_buff *skb,
-		   dma_addr_t *addr)
+int cxgb4_map_skb(struct device *dev, const struct sk_buff *skb,
+		  dma_addr_t *addr)
 {
 	const skb_frag_t *fp, *end;
 	const struct skb_shared_info *si;
@@ -276,8 +269,8 @@ unwind:
 out_err:
 	return -ENOMEM;
 }
+EXPORT_SYMBOL(cxgb4_map_skb);
 
-#ifdef CONFIG_NEED_DMA_MAP_STATE
 static void unmap_skb(struct device *dev, const struct sk_buff *skb,
 		      const dma_addr_t *addr)
 {
@@ -292,6 +285,7 @@ static void unmap_skb(struct device *dev, const struct sk_buff *skb,
 		dma_unmap_page(dev, *addr++, skb_frag_size(fp), DMA_TO_DEVICE);
 }
 
+#ifdef CONFIG_NEED_DMA_MAP_STATE
 /**
  *	deferred_unmap_destructor - unmap a packet when it is freed
  *	@skb: the packet
@@ -306,68 +300,9 @@ static void deferred_unmap_destructor(struct sk_buff *skb)
 }
 #endif
 
-static void unmap_sgl(struct device *dev, const struct sk_buff *skb,
-		      const struct ulptx_sgl *sgl, const struct sge_txq *q)
-{
-	const struct ulptx_sge_pair *p;
-	unsigned int nfrags = skb_shinfo(skb)->nr_frags;
-
-	if (likely(skb_headlen(skb)))
-		dma_unmap_single(dev, be64_to_cpu(sgl->addr0), ntohl(sgl->len0),
-				 DMA_TO_DEVICE);
-	else {
-		dma_unmap_page(dev, be64_to_cpu(sgl->addr0), ntohl(sgl->len0),
-			       DMA_TO_DEVICE);
-		nfrags--;
-	}
-
-	/*
-	 * the complexity below is because of the possibility of a wrap-around
-	 * in the middle of an SGL
-	 */
-	for (p = sgl->sge; nfrags >= 2; nfrags -= 2) {
-		if (likely((u8 *)(p + 1) <= (u8 *)q->stat)) {
-unmap:			dma_unmap_page(dev, be64_to_cpu(p->addr[0]),
-				       ntohl(p->len[0]), DMA_TO_DEVICE);
-			dma_unmap_page(dev, be64_to_cpu(p->addr[1]),
-				       ntohl(p->len[1]), DMA_TO_DEVICE);
-			p++;
-		} else if ((u8 *)p == (u8 *)q->stat) {
-			p = (const struct ulptx_sge_pair *)q->desc;
-			goto unmap;
-		} else if ((u8 *)p + 8 == (u8 *)q->stat) {
-			const __be64 *addr = (const __be64 *)q->desc;
-
-			dma_unmap_page(dev, be64_to_cpu(addr[0]),
-				       ntohl(p->len[0]), DMA_TO_DEVICE);
-			dma_unmap_page(dev, be64_to_cpu(addr[1]),
-				       ntohl(p->len[1]), DMA_TO_DEVICE);
-			p = (const struct ulptx_sge_pair *)&addr[2];
-		} else {
-			const __be64 *addr = (const __be64 *)q->desc;
-
-			dma_unmap_page(dev, be64_to_cpu(p->addr[0]),
-				       ntohl(p->len[0]), DMA_TO_DEVICE);
-			dma_unmap_page(dev, be64_to_cpu(addr[0]),
-				       ntohl(p->len[1]), DMA_TO_DEVICE);
-			p = (const struct ulptx_sge_pair *)&addr[1];
-		}
-	}
-	if (nfrags) {
-		__be64 addr;
-
-		if ((u8 *)p == (u8 *)q->stat)
-			p = (const struct ulptx_sge_pair *)q->desc;
-		addr = (u8 *)p + 16 <= (u8 *)q->stat ? p->addr[0] :
-						       *(const __be64 *)q->desc;
-		dma_unmap_page(dev, be64_to_cpu(addr), ntohl(p->len[0]),
-			       DMA_TO_DEVICE);
-	}
-}
-
 /**
  *	free_tx_desc - reclaims Tx descriptors and their buffers
- *	@adapter: the adapter
+ *	@adap: the adapter
  *	@q: the Tx queue to reclaim descriptors from
  *	@n: the number of descriptors to reclaim
  *	@unmap: whether the buffers should be unmapped for DMA
@@ -378,15 +313,16 @@ unmap:			dma_unmap_page(dev, be64_to_cpu(p->addr[0]),
 void free_tx_desc(struct adapter *adap, struct sge_txq *q,
 		  unsigned int n, bool unmap)
 {
-	struct tx_sw_desc *d;
 	unsigned int cidx = q->cidx;
-	struct device *dev = adap->pdev_dev;
+	struct tx_sw_desc *d;
 
 	d = &q->sdesc[cidx];
 	while (n--) {
 		if (d->skb) {                       /* an SGL is present */
-			if (unmap)
-				unmap_sgl(dev, d->skb, d->sgl, q);
+			if (unmap && d->addr[0]) {
+				unmap_skb(adap->pdev_dev, d->skb, d->addr);
+				memset(d->addr, 0, sizeof(d->addr));
+			}
 			dev_consume_skb_any(d->skb);
 			d->skb = NULL;
 		}
@@ -404,38 +340,60 @@ void free_tx_desc(struct adapter *adap, struct sge_txq *q,
  */
 static inline int reclaimable(const struct sge_txq *q)
 {
-	int hw_cidx = ntohs(ACCESS_ONCE(q->stat->cidx));
+	int hw_cidx = ntohs(READ_ONCE(q->stat->cidx));
 	hw_cidx -= q->cidx;
 	return hw_cidx < 0 ? hw_cidx + q->size : hw_cidx;
 }
 
 /**
- *	reclaim_completed_tx - reclaims completed Tx descriptors
+ *	reclaim_completed_tx - reclaims completed TX Descriptors
  *	@adap: the adapter
  *	@q: the Tx queue to reclaim completed descriptors from
+ *	@maxreclaim: the maximum number of TX Descriptors to reclaim or -1
  *	@unmap: whether the buffers should be unmapped for DMA
  *
- *	Reclaims Tx descriptors that the SGE has indicated it has processed,
- *	and frees the associated buffers if possible.  Called with the Tx
- *	queue locked.
+ *	Reclaims Tx Descriptors that the SGE has indicated it has processed,
+ *	and frees the associated buffers if possible.  If @max == -1, then
+ *	we'll use a defaiult maximum.  Called with the TX Queue locked.
  */
-static inline void reclaim_completed_tx(struct adapter *adap, struct sge_txq *q,
-					bool unmap)
+static inline int reclaim_completed_tx(struct adapter *adap, struct sge_txq *q,
+				       int maxreclaim, bool unmap)
 {
-	int avail = reclaimable(q);
+	int reclaim = reclaimable(q);
 
-	if (avail) {
+	if (reclaim) {
 		/*
 		 * Limit the amount of clean up work we do at a time to keep
 		 * the Tx lock hold time O(1).
 		 */
-		if (avail > MAX_TX_RECLAIM)
-			avail = MAX_TX_RECLAIM;
+		if (maxreclaim < 0)
+			maxreclaim = MAX_TX_RECLAIM;
+		if (reclaim > maxreclaim)
+			reclaim = maxreclaim;
 
-		free_tx_desc(adap, q, avail, unmap);
-		q->in_use -= avail;
+		free_tx_desc(adap, q, reclaim, unmap);
+		q->in_use -= reclaim;
 	}
+
+	return reclaim;
+}
+
+/**
+ *	cxgb4_reclaim_completed_tx - reclaims completed Tx descriptors
+ *	@adap: the adapter
+ *	@q: the Tx queue to reclaim completed descriptors from
+ *	@unmap: whether the buffers should be unmapped for DMA
+ *
+ *	Reclaims Tx descriptors that the SGE has indicated it has processed,
+ *	and frees the associated buffers if possible.  Called with the Tx
+ *	queue locked.
+ */
+void cxgb4_reclaim_completed_tx(struct adapter *adap, struct sge_txq *q,
+				bool unmap)
+{
+	(void)reclaim_completed_tx(adap, q, -1, unmap);
 }
+EXPORT_SYMBOL(cxgb4_reclaim_completed_tx);
 
 static inline int get_buf_size(struct adapter *adapter,
 			       const struct rx_sw_desc *d)
@@ -462,7 +420,7 @@ static inline int get_buf_size(struct adapter *adapter,
 		break;
 
 	default:
-		BUG_ON(1);
+		BUG();
 	}
 
 	return buf_size;
@@ -485,7 +443,7 @@ static void free_rx_bufs(struct adapter *adap, struct sge_fl *q, int n)
 		if (is_buf_mapped(d))
 			dma_unmap_page(adap->pdev_dev, get_buf_addr(d),
 				       get_buf_size(adap, d),
-				       PCI_DMA_FROMDEVICE);
+				       DMA_FROM_DEVICE);
 		put_page(d->page);
 		d->page = NULL;
 		if (++q->cidx == q->size)
@@ -511,7 +469,7 @@ static void unmap_rx_buf(struct adapter *adap, struct sge_fl *q)
 
 	if (is_buf_mapped(d))
 		dma_unmap_page(adap->pdev_dev, get_buf_addr(d),
-			       get_buf_size(adap, d), PCI_DMA_FROMDEVICE);
+			       get_buf_size(adap, d), DMA_FROM_DEVICE);
 	d->page = NULL;
 	if (++q->cidx == q->size)
 		q->cidx = 0;
@@ -608,7 +566,7 @@ static unsigned int refill_fl(struct adapter *adap, struct sge_fl *q, int n,
 
 		mapping = dma_map_page(adap->pdev_dev, pg, 0,
 				       PAGE_SIZE << s->fl_pg_order,
-				       PCI_DMA_FROMDEVICE);
+				       DMA_FROM_DEVICE);
 		if (unlikely(dma_mapping_error(adap->pdev_dev, mapping))) {
 			__free_pages(pg, s->fl_pg_order);
 			q->mapping_err++;
@@ -638,7 +596,7 @@ alloc_small_pages:
 		}
 
 		mapping = dma_map_page(adap->pdev_dev, pg, 0, PAGE_SIZE,
-				       PCI_DMA_FROMDEVICE);
+				       DMA_FROM_DEVICE);
 		if (unlikely(dma_mapping_error(adap->pdev_dev, mapping))) {
 			put_page(pg);
 			q->mapping_err++;
@@ -707,7 +665,7 @@ static void *alloc_ring(struct device *dev, size_t nelem, size_t elem_size,
 	if (!p)
 		return NULL;
 	if (sw_size) {
-		s = kzalloc_node(nelem * sw_size, GFP_KERNEL, node);
+		s = kcalloc_node(sw_size, nelem, GFP_KERNEL, node);
 
 		if (!s) {
 			dma_free_coherent(dev, len, p, *phys);
@@ -716,7 +674,6 @@ static void *alloc_ring(struct device *dev, size_t nelem, size_t elem_size,
 	}
 	if (metadata)
 		*(void **)metadata = s;
-	memset(p, 0, len);
 	return p;
 }
 
@@ -765,16 +722,26 @@ static inline unsigned int flits_to_desc(unsigned int n)
 /**
  *	is_eth_imm - can an Ethernet packet be sent as immediate data?
  *	@skb: the packet
+ *	@chip_ver: chip version
  *
  *	Returns whether an Ethernet packet is small enough to fit as
  *	immediate data. Return value corresponds to headroom required.
  */
-static inline int is_eth_imm(const struct sk_buff *skb)
+static inline int is_eth_imm(const struct sk_buff *skb, unsigned int chip_ver)
 {
-	int hdrlen = skb_shinfo(skb)->gso_size ?
-			sizeof(struct cpl_tx_pkt_lso_core) : 0;
+	int hdrlen = 0;
 
-	hdrlen += sizeof(struct cpl_tx_pkt);
+	if (skb->encapsulation && skb_shinfo(skb)->gso_size &&
+	    chip_ver > CHELSIO_T5) {
+		hdrlen = sizeof(struct cpl_tx_tnl_lso);
+		hdrlen += sizeof(struct cpl_tx_pkt_core);
+	} else if (skb_shinfo(skb)->gso_type & SKB_GSO_UDP_L4) {
+		return 0;
+	} else {
+		hdrlen = skb_shinfo(skb)->gso_size ?
+			 sizeof(struct cpl_tx_pkt_lso_core) : 0;
+		hdrlen += sizeof(struct cpl_tx_pkt);
+	}
 	if (skb->len <= MAX_IMM_TX_PKT_LEN - hdrlen)
 		return hdrlen;
 	return 0;
@@ -783,14 +750,16 @@ static inline int is_eth_imm(const struct sk_buff *skb)
 /**
  *	calc_tx_flits - calculate the number of flits for a packet Tx WR
  *	@skb: the packet
+ *	@chip_ver: chip version
  *
  *	Returns the number of flits needed for a Tx WR for the given Ethernet
  *	packet, including the needed WR and CPL headers.
  */
-static inline unsigned int calc_tx_flits(const struct sk_buff *skb)
+static inline unsigned int calc_tx_flits(const struct sk_buff *skb,
+					 unsigned int chip_ver)
 {
 	unsigned int flits;
-	int hdrlen = is_eth_imm(skb);
+	int hdrlen = is_eth_imm(skb, chip_ver);
 
 	/* If the skb is small enough, we can pump it out as a work request
 	 * with only immediate data.  In that case we just have to have the
@@ -809,30 +778,33 @@ static inline unsigned int calc_tx_flits(const struct sk_buff *skb)
 	 * with an embedded TX Packet Write CPL message.
 	 */
 	flits = sgl_len(skb_shinfo(skb)->nr_frags + 1);
-	if (skb_shinfo(skb)->gso_size)
-		flits += (sizeof(struct fw_eth_tx_pkt_wr) +
-			  sizeof(struct cpl_tx_pkt_lso_core) +
-			  sizeof(struct cpl_tx_pkt_core)) / sizeof(__be64);
-	else
+	if (skb_shinfo(skb)->gso_size) {
+		if (skb->encapsulation && chip_ver > CHELSIO_T5) {
+			hdrlen = sizeof(struct fw_eth_tx_pkt_wr) +
+				 sizeof(struct cpl_tx_tnl_lso);
+		} else if (skb_shinfo(skb)->gso_type & SKB_GSO_UDP_L4) {
+			u32 pkt_hdrlen;
+
+			pkt_hdrlen = eth_get_headlen(skb->dev, skb->data,
+						     skb_headlen(skb));
+			hdrlen = sizeof(struct fw_eth_tx_eo_wr) +
+				 round_up(pkt_hdrlen, 16);
+		} else {
+			hdrlen = sizeof(struct fw_eth_tx_pkt_wr) +
+				 sizeof(struct cpl_tx_pkt_lso_core);
+		}
+
+		hdrlen += sizeof(struct cpl_tx_pkt_core);
+		flits += (hdrlen / sizeof(__be64));
+	} else {
 		flits += (sizeof(struct fw_eth_tx_pkt_wr) +
 			  sizeof(struct cpl_tx_pkt_core)) / sizeof(__be64);
+	}
 	return flits;
 }
 
 /**
- *	calc_tx_descs - calculate the number of Tx descriptors for a packet
- *	@skb: the packet
- *
- *	Returns the number of Tx descriptors needed for the given Ethernet
- *	packet, including the needed WR and CPL headers.
- */
-static inline unsigned int calc_tx_descs(const struct sk_buff *skb)
-{
-	return flits_to_desc(calc_tx_flits(skb));
-}
-
-/**
- *	write_sgl - populate a scatter/gather list for a packet
+ *	cxgb4_write_sgl - populate a scatter/gather list for a packet
  *	@skb: the packet
  *	@q: the Tx queue we are writing into
  *	@sgl: starting location for writing the SGL
@@ -848,9 +820,9 @@ static inline unsigned int calc_tx_descs(const struct sk_buff *skb)
  *	right after the end of the SGL but does not account for any potential
  *	wrap around, i.e., @end > @sgl.
  */
-static void write_sgl(const struct sk_buff *skb, struct sge_txq *q,
-		      struct ulptx_sgl *sgl, u64 *end, unsigned int start,
-		      const dma_addr_t *addr)
+void cxgb4_write_sgl(const struct sk_buff *skb, struct sge_txq *q,
+		     struct ulptx_sgl *sgl, u64 *end, unsigned int start,
+		     const dma_addr_t *addr)
 {
 	unsigned int i, len;
 	struct ulptx_sge_pair *to;
@@ -902,6 +874,115 @@ static void write_sgl(const struct sk_buff *skb, struct sge_txq *q,
 	if ((uintptr_t)end & 8)           /* 0-pad to multiple of 16 */
 		*end = 0;
 }
+EXPORT_SYMBOL(cxgb4_write_sgl);
+
+/*	cxgb4_write_partial_sgl - populate SGL for partial packet
+ *	@skb: the packet
+ *	@q: the Tx queue we are writing into
+ *	@sgl: starting location for writing the SGL
+ *	@end: points right after the end of the SGL
+ *	@addr: the list of bus addresses for the SGL elements
+ *	@start: start offset in the SKB where partial data starts
+ *	@len: length of data from @start to send out
+ *
+ *	This API will handle sending out partial data of a skb if required.
+ *	Unlike cxgb4_write_sgl, @start can be any offset into the skb data,
+ *	and @len will decide how much data after @start offset to send out.
+ */
+void cxgb4_write_partial_sgl(const struct sk_buff *skb, struct sge_txq *q,
+			     struct ulptx_sgl *sgl, u64 *end,
+			     const dma_addr_t *addr, u32 start, u32 len)
+{
+	struct ulptx_sge_pair buf[MAX_SKB_FRAGS / 2 + 1] = {0}, *to;
+	u32 frag_size, skb_linear_data_len = skb_headlen(skb);
+	struct skb_shared_info *si = skb_shinfo(skb);
+	u8 i = 0, frag_idx = 0, nfrags = 0;
+	skb_frag_t *frag;
+
+	/* Fill the first SGL either from linear data or from partial
+	 * frag based on @start.
+	 */
+	if (unlikely(start < skb_linear_data_len)) {
+		frag_size = min(len, skb_linear_data_len - start);
+		sgl->len0 = htonl(frag_size);
+		sgl->addr0 = cpu_to_be64(addr[0] + start);
+		len -= frag_size;
+		nfrags++;
+	} else {
+		start -= skb_linear_data_len;
+		frag = &si->frags[frag_idx];
+		frag_size = skb_frag_size(frag);
+		/* find the first frag */
+		while (start >= frag_size) {
+			start -= frag_size;
+			frag_idx++;
+			frag = &si->frags[frag_idx];
+			frag_size = skb_frag_size(frag);
+		}
+
+		frag_size = min(len, skb_frag_size(frag) - start);
+		sgl->len0 = cpu_to_be32(frag_size);
+		sgl->addr0 = cpu_to_be64(addr[frag_idx + 1] + start);
+		len -= frag_size;
+		nfrags++;
+		frag_idx++;
+	}
+
+	/* If the entire partial data fit in one SGL, then send it out
+	 * now.
+	 */
+	if (!len)
+		goto done;
+
+	/* Most of the complexity below deals with the possibility we hit the
+	 * end of the queue in the middle of writing the SGL.  For this case
+	 * only we create the SGL in a temporary buffer and then copy it.
+	 */
+	to = (u8 *)end > (u8 *)q->stat ? buf : sgl->sge;
+
+	/* If the skb couldn't fit in first SGL completely, fill the
+	 * rest of the frags in subsequent SGLs. Note that each SGL
+	 * pair can store 2 frags.
+	 */
+	while (len) {
+		frag_size = min(len, skb_frag_size(&si->frags[frag_idx]));
+		to->len[i & 1] = cpu_to_be32(frag_size);
+		to->addr[i & 1] = cpu_to_be64(addr[frag_idx + 1]);
+		if (i && (i & 1))
+			to++;
+		nfrags++;
+		frag_idx++;
+		i++;
+		len -= frag_size;
+	}
+
+	/* If we ended in an odd boundary, then set the second SGL's
+	 * length in the pair to 0.
+	 */
+	if (i & 1)
+		to->len[1] = cpu_to_be32(0);
+
+	/* Copy from temporary buffer to Tx ring, in case we hit the
+	 * end of the queue in the middle of writing the SGL.
+	 */
+	if (unlikely((u8 *)end > (u8 *)q->stat)) {
+		u32 part0 = (u8 *)q->stat - (u8 *)sgl->sge, part1;
+
+		if (likely(part0))
+			memcpy(sgl->sge, buf, part0);
+		part1 = (u8 *)end - (u8 *)q->stat;
+		memcpy(q->desc, (u8 *)buf + part0, part1);
+		end = (void *)q->desc + part1;
+	}
+
+	/* 0-pad to multiple of 16 */
+	if ((uintptr_t)end & 8)
+		*end = 0;
+done:
+	sgl->cmd_nsge = htonl(ULPTX_CMD_V(ULP_TX_SC_DSGL) |
+			ULPTX_NSGE_V(nfrags));
+}
+EXPORT_SYMBOL(cxgb4_write_partial_sgl);
 
 /* This function copies 64 byte coalesced work request to
  * memory mapped BAR2 space. For coalesced WR SGE fetches
@@ -920,14 +1001,14 @@ static void cxgb_pio_copy(u64 __iomem *dst, u64 *src)
 }
 
 /**
- *	ring_tx_db - check and potentially ring a Tx queue's doorbell
+ *	cxgb4_ring_tx_db - check and potentially ring a Tx queue's doorbell
  *	@adap: the adapter
  *	@q: the Tx queue
  *	@n: number of new descriptors to give to HW
  *
  *	Ring the doorbel for a Tx queue.
  */
-static inline void ring_tx_db(struct adapter *adap, struct sge_txq *q, int n)
+inline void cxgb4_ring_tx_db(struct adapter *adap, struct sge_txq *q, int n)
 {
 	/* Make sure that all writes to the TX Descriptors are committed
 	 * before we tell the hardware about them.
@@ -994,9 +1075,10 @@ static inline void ring_tx_db(struct adapter *adap, struct sge_txq *q, int n)
 		wmb();
 	}
 }
+EXPORT_SYMBOL(cxgb4_ring_tx_db);
 
 /**
- *	inline_tx_skb - inline a packet's data into Tx descriptors
+ *	cxgb4_inline_tx_skb - inline a packet's data into Tx descriptors
  *	@skb: the packet
  *	@q: the Tx queue where the packet will be inlined
  *	@pos: starting position in the Tx queue where to inline the packet
@@ -1006,11 +1088,11 @@ static inline void ring_tx_db(struct adapter *adap, struct sge_txq *q, int n)
  *	Most of the complexity of this operation is dealing with wrap arounds
  *	in the middle of the packet we want to inline.
  */
-static void inline_tx_skb(const struct sk_buff *skb, const struct sge_txq *q,
-			  void *pos)
+void cxgb4_inline_tx_skb(const struct sk_buff *skb,
+			 const struct sge_txq *q, void *pos)
 {
-	u64 *p;
 	int left = (void *)q->stat - pos;
+	u64 *p;
 
 	if (likely(skb->len <= left)) {
 		if (likely(!skb->data_len))
@@ -1029,6 +1111,7 @@ static void inline_tx_skb(const struct sk_buff *skb, const struct sge_txq *q,
 	if ((uintptr_t)p & 8)
 		*p = 0;
 }
+EXPORT_SYMBOL(cxgb4_inline_tx_skb);
 
 static void *inline_tx_skb_header(const struct sk_buff *skb,
 				  const struct sge_txq *q,  void *pos,
@@ -1061,12 +1144,27 @@ static void *inline_tx_skb_header(const struct sk_buff *skb,
 static u64 hwcsum(enum chip_type chip, const struct sk_buff *skb)
 {
 	int csum_type;
-	const struct iphdr *iph = ip_hdr(skb);
+	bool inner_hdr_csum = false;
+	u16 proto, ver;
+
+	if (skb->encapsulation &&
+	    (CHELSIO_CHIP_VERSION(chip) > CHELSIO_T5))
+		inner_hdr_csum = true;
+
+	if (inner_hdr_csum) {
+		ver = inner_ip_hdr(skb)->version;
+		proto = (ver == 4) ? inner_ip_hdr(skb)->protocol :
+			inner_ipv6_hdr(skb)->nexthdr;
+	} else {
+		ver = ip_hdr(skb)->version;
+		proto = (ver == 4) ? ip_hdr(skb)->protocol :
+			ipv6_hdr(skb)->nexthdr;
+	}
 
-	if (iph->version == 4) {
-		if (iph->protocol == IPPROTO_TCP)
+	if (ver == 4) {
+		if (proto == IPPROTO_TCP)
 			csum_type = TX_CSUM_TCPIP;
-		else if (iph->protocol == IPPROTO_UDP)
+		else if (proto == IPPROTO_UDP)
 			csum_type = TX_CSUM_UDPIP;
 		else {
 nocsum:			/*
@@ -1079,19 +1177,29 @@ nocsum:			/*
 		/*
 		 * this doesn't work with extension headers
 		 */
-		const struct ipv6hdr *ip6h = (const struct ipv6hdr *)iph;
-
-		if (ip6h->nexthdr == IPPROTO_TCP)
+		if (proto == IPPROTO_TCP)
 			csum_type = TX_CSUM_TCPIP6;
-		else if (ip6h->nexthdr == IPPROTO_UDP)
+		else if (proto == IPPROTO_UDP)
 			csum_type = TX_CSUM_UDPIP6;
 		else
 			goto nocsum;
 	}
 
 	if (likely(csum_type >= TX_CSUM_TCPIP)) {
-		u64 hdr_len = TXPKT_IPHDR_LEN_V(skb_network_header_len(skb));
-		int eth_hdr_len = skb_network_offset(skb) - ETH_HLEN;
+		int eth_hdr_len, l4_len;
+		u64 hdr_len;
+
+		if (inner_hdr_csum) {
+			/* This allows checksum offload for all encapsulated
+			 * packets like GRE etc..
+			 */
+			l4_len = skb_inner_network_header_len(skb);
+			eth_hdr_len = skb_inner_network_offset(skb) - ETH_HLEN;
+		} else {
+			l4_len = skb_network_header_len(skb);
+			eth_hdr_len = skb_network_offset(skb) - ETH_HLEN;
+		}
+		hdr_len = TXPKT_IPHDR_LEN_V(l4_len);
 
 		if (CHELSIO_CHIP_VERSION(chip) <= CHELSIO_T5)
 			hdr_len |= TXPKT_ETHHDR_LEN_V(eth_hdr_len);
@@ -1153,63 +1261,290 @@ cxgb_fcoe_offload(struct sk_buff *skb, struct adapter *adap,
 }
 #endif /* CONFIG_CHELSIO_T4_FCOE */
 
+/* Returns tunnel type if hardware supports offloading of the same.
+ * It is called only for T5 and onwards.
+ */
+enum cpl_tx_tnl_lso_type cxgb_encap_offload_supported(struct sk_buff *skb)
+{
+	u8 l4_hdr = 0;
+	enum cpl_tx_tnl_lso_type tnl_type = TX_TNL_TYPE_OPAQUE;
+	struct port_info *pi = netdev_priv(skb->dev);
+	struct adapter *adapter = pi->adapter;
+
+	if (skb->inner_protocol_type != ENCAP_TYPE_ETHER ||
+	    skb->inner_protocol != htons(ETH_P_TEB))
+		return tnl_type;
+
+	switch (vlan_get_protocol(skb)) {
+	case htons(ETH_P_IP):
+		l4_hdr = ip_hdr(skb)->protocol;
+		break;
+	case htons(ETH_P_IPV6):
+		l4_hdr = ipv6_hdr(skb)->nexthdr;
+		break;
+	default:
+		return tnl_type;
+	}
+
+	switch (l4_hdr) {
+	case IPPROTO_UDP:
+		if (adapter->vxlan_port == udp_hdr(skb)->dest)
+			tnl_type = TX_TNL_TYPE_VXLAN;
+		else if (adapter->geneve_port == udp_hdr(skb)->dest)
+			tnl_type = TX_TNL_TYPE_GENEVE;
+		break;
+	default:
+		return tnl_type;
+	}
+
+	return tnl_type;
+}
+
+static inline void t6_fill_tnl_lso(struct sk_buff *skb,
+				   struct cpl_tx_tnl_lso *tnl_lso,
+				   enum cpl_tx_tnl_lso_type tnl_type)
+{
+	u32 val;
+	int in_eth_xtra_len;
+	int l3hdr_len = skb_network_header_len(skb);
+	int eth_xtra_len = skb_network_offset(skb) - ETH_HLEN;
+	const struct skb_shared_info *ssi = skb_shinfo(skb);
+	bool v6 = (ip_hdr(skb)->version == 6);
+
+	val = CPL_TX_TNL_LSO_OPCODE_V(CPL_TX_TNL_LSO) |
+	      CPL_TX_TNL_LSO_FIRST_F |
+	      CPL_TX_TNL_LSO_LAST_F |
+	      (v6 ? CPL_TX_TNL_LSO_IPV6OUT_F : 0) |
+	      CPL_TX_TNL_LSO_ETHHDRLENOUT_V(eth_xtra_len / 4) |
+	      CPL_TX_TNL_LSO_IPHDRLENOUT_V(l3hdr_len / 4) |
+	      (v6 ? 0 : CPL_TX_TNL_LSO_IPHDRCHKOUT_F) |
+	      CPL_TX_TNL_LSO_IPLENSETOUT_F |
+	      (v6 ? 0 : CPL_TX_TNL_LSO_IPIDINCOUT_F);
+	tnl_lso->op_to_IpIdSplitOut = htonl(val);
+
+	tnl_lso->IpIdOffsetOut = 0;
+
+	/* Get the tunnel header length */
+	val = skb_inner_mac_header(skb) - skb_mac_header(skb);
+	in_eth_xtra_len = skb_inner_network_header(skb) -
+			  skb_inner_mac_header(skb) - ETH_HLEN;
+
+	switch (tnl_type) {
+	case TX_TNL_TYPE_VXLAN:
+	case TX_TNL_TYPE_GENEVE:
+		tnl_lso->UdpLenSetOut_to_TnlHdrLen =
+			htons(CPL_TX_TNL_LSO_UDPCHKCLROUT_F |
+			CPL_TX_TNL_LSO_UDPLENSETOUT_F);
+		break;
+	default:
+		tnl_lso->UdpLenSetOut_to_TnlHdrLen = 0;
+		break;
+	}
+
+	tnl_lso->UdpLenSetOut_to_TnlHdrLen |=
+		 htons(CPL_TX_TNL_LSO_TNLHDRLEN_V(val) |
+		       CPL_TX_TNL_LSO_TNLTYPE_V(tnl_type));
+
+	tnl_lso->r1 = 0;
+
+	val = CPL_TX_TNL_LSO_ETHHDRLEN_V(in_eth_xtra_len / 4) |
+	      CPL_TX_TNL_LSO_IPV6_V(inner_ip_hdr(skb)->version == 6) |
+	      CPL_TX_TNL_LSO_IPHDRLEN_V(skb_inner_network_header_len(skb) / 4) |
+	      CPL_TX_TNL_LSO_TCPHDRLEN_V(inner_tcp_hdrlen(skb) / 4);
+	tnl_lso->Flow_to_TcpHdrLen = htonl(val);
+
+	tnl_lso->IpIdOffset = htons(0);
+
+	tnl_lso->IpIdSplit_to_Mss = htons(CPL_TX_TNL_LSO_MSS_V(ssi->gso_size));
+	tnl_lso->TCPSeqOffset = htonl(0);
+	tnl_lso->EthLenOffset_Size = htonl(CPL_TX_TNL_LSO_SIZE_V(skb->len));
+}
+
+static inline void *write_tso_wr(struct adapter *adap, struct sk_buff *skb,
+				 struct cpl_tx_pkt_lso_core *lso)
+{
+	int eth_xtra_len = skb_network_offset(skb) - ETH_HLEN;
+	int l3hdr_len = skb_network_header_len(skb);
+	const struct skb_shared_info *ssi;
+	bool ipv6 = false;
+
+	ssi = skb_shinfo(skb);
+	if (ssi->gso_type & SKB_GSO_TCPV6)
+		ipv6 = true;
+
+	lso->lso_ctrl = htonl(LSO_OPCODE_V(CPL_TX_PKT_LSO) |
+			      LSO_FIRST_SLICE_F | LSO_LAST_SLICE_F |
+			      LSO_IPV6_V(ipv6) |
+			      LSO_ETHHDR_LEN_V(eth_xtra_len / 4) |
+			      LSO_IPHDR_LEN_V(l3hdr_len / 4) |
+			      LSO_TCPHDR_LEN_V(tcp_hdr(skb)->doff));
+	lso->ipid_ofst = htons(0);
+	lso->mss = htons(ssi->gso_size);
+	lso->seqno_offset = htonl(0);
+	if (is_t4(adap->params.chip))
+		lso->len = htonl(skb->len);
+	else
+		lso->len = htonl(LSO_T5_XFER_SIZE_V(skb->len));
+
+	return (void *)(lso + 1);
+}
+
 /**
- *	t4_eth_xmit - add a packet to an Ethernet Tx queue
- *	@skb: the packet
- *	@dev: the egress net device
+ *	t4_sge_eth_txq_egress_update - handle Ethernet TX Queue update
+ *	@adap: the adapter
+ *	@eq: the Ethernet TX Queue
+ *	@maxreclaim: the maximum number of TX Descriptors to reclaim or -1
  *
- *	Add a packet to an SGE Ethernet Tx queue.  Runs with softirqs disabled.
+ *	We're typically called here to update the state of an Ethernet TX
+ *	Queue with respect to the hardware's progress in consuming the TX
+ *	Work Requests that we've put on that Egress Queue.  This happens
+ *	when we get Egress Queue Update messages and also prophylactically
+ *	in regular timer-based Ethernet TX Queue maintenance.
  */
-netdev_tx_t t4_eth_xmit(struct sk_buff *skb, struct net_device *dev)
+int t4_sge_eth_txq_egress_update(struct adapter *adap, struct sge_eth_txq *eq,
+				 int maxreclaim)
 {
-	u32 wr_mid, ctrl0;
-	u64 cntrl, *end;
-	int qidx, credits;
-	unsigned int flits, ndesc;
-	struct adapter *adap;
-	struct sge_eth_txq *q;
-	const struct port_info *pi;
-	struct fw_eth_tx_pkt_wr *wr;
-	struct cpl_tx_pkt_core *cpl;
-	const struct skb_shared_info *ssi;
-	dma_addr_t addr[MAX_SKB_FRAGS + 1];
-	bool immediate = false;
-	int len, max_pkt_len;
-#ifdef CONFIG_CHELSIO_T4_FCOE
-	int err;
-#endif /* CONFIG_CHELSIO_T4_FCOE */
+	unsigned int reclaimed, hw_cidx;
+	struct sge_txq *q = &eq->q;
+	int hw_in_use;
 
-	/*
-	 * The chip min packet length is 10 octets but play safe and reject
-	 * anything shorter than an Ethernet header.
+	if (!q->in_use || !__netif_tx_trylock(eq->txq))
+		return 0;
+
+	/* Reclaim pending completed TX Descriptors. */
+	reclaimed = reclaim_completed_tx(adap, &eq->q, maxreclaim, true);
+
+	hw_cidx = ntohs(READ_ONCE(q->stat->cidx));
+	hw_in_use = q->pidx - hw_cidx;
+	if (hw_in_use < 0)
+		hw_in_use += q->size;
+
+	/* If the TX Queue is currently stopped and there's now more than half
+	 * the queue available, restart it.  Otherwise bail out since the rest
+	 * of what we want do here is with the possibility of shipping any
+	 * currently buffered Coalesced TX Work Request.
 	 */
-	if (unlikely(skb->len < ETH_HLEN)) {
-out_free:	dev_kfree_skb_any(skb);
-		return NETDEV_TX_OK;
+	if (netif_tx_queue_stopped(eq->txq) && hw_in_use < (q->size / 2)) {
+		netif_tx_wake_queue(eq->txq);
+		eq->q.restarts++;
 	}
 
+	__netif_tx_unlock(eq->txq);
+	return reclaimed;
+}
+
+static inline int cxgb4_validate_skb(struct sk_buff *skb,
+				     struct net_device *dev,
+				     u32 min_pkt_len)
+{
+	u32 max_pkt_len;
+
+	/* The chip min packet length is 10 octets but some firmware
+	 * commands have a minimum packet length requirement. So, play
+	 * safe and reject anything shorter than @min_pkt_len.
+	 */
+	if (unlikely(skb->len < min_pkt_len))
+		return -EINVAL;
+
 	/* Discard the packet if the length is greater than mtu */
 	max_pkt_len = ETH_HLEN + dev->mtu;
+
 	if (skb_vlan_tagged(skb))
 		max_pkt_len += VLAN_HLEN;
+
 	if (!skb_shinfo(skb)->gso_size && (unlikely(skb->len > max_pkt_len)))
+		return -EINVAL;
+
+	return 0;
+}
+
+static void *write_eo_udp_wr(struct sk_buff *skb, struct fw_eth_tx_eo_wr *wr,
+			     u32 hdr_len)
+{
+	wr->u.udpseg.type = FW_ETH_TX_EO_TYPE_UDPSEG;
+	wr->u.udpseg.ethlen = skb_network_offset(skb);
+	wr->u.udpseg.iplen = cpu_to_be16(skb_network_header_len(skb));
+	wr->u.udpseg.udplen = sizeof(struct udphdr);
+	wr->u.udpseg.rtplen = 0;
+	wr->u.udpseg.r4 = 0;
+	if (skb_shinfo(skb)->gso_size)
+		wr->u.udpseg.mss = cpu_to_be16(skb_shinfo(skb)->gso_size);
+	else
+		wr->u.udpseg.mss = cpu_to_be16(skb->len - hdr_len);
+	wr->u.udpseg.schedpktsize = wr->u.udpseg.mss;
+	wr->u.udpseg.plen = cpu_to_be32(skb->len - hdr_len);
+
+	return (void *)(wr + 1);
+}
+
+/**
+ *	cxgb4_eth_xmit - add a packet to an Ethernet Tx queue
+ *	@skb: the packet
+ *	@dev: the egress net device
+ *
+ *	Add a packet to an SGE Ethernet Tx queue.  Runs with softirqs disabled.
+ */
+static netdev_tx_t cxgb4_eth_xmit(struct sk_buff *skb, struct net_device *dev)
+{
+	enum cpl_tx_tnl_lso_type tnl_type = TX_TNL_TYPE_OPAQUE;
+	bool ptp_enabled = is_ptp_enabled(skb, dev);
+	unsigned int last_desc, flits, ndesc;
+	u32 wr_mid, ctrl0, op, sgl_off = 0;
+	const struct skb_shared_info *ssi;
+	int len, qidx, credits, ret, left;
+	struct tx_sw_desc *sgl_sdesc;
+	struct fw_eth_tx_eo_wr *eowr;
+	struct fw_eth_tx_pkt_wr *wr;
+	struct cpl_tx_pkt_core *cpl;
+	const struct port_info *pi;
+	bool immediate = false;
+	u64 cntrl, *end, *sgl;
+	struct sge_eth_txq *q;
+	unsigned int chip_ver;
+	struct adapter *adap;
+
+	ret = cxgb4_validate_skb(skb, dev, ETH_HLEN);
+	if (ret)
 		goto out_free;
 
 	pi = netdev_priv(dev);
 	adap = pi->adapter;
+	ssi = skb_shinfo(skb);
+#if IS_ENABLED(CONFIG_CHELSIO_IPSEC_INLINE)
+	if (xfrm_offload(skb) && !ssi->gso_size)
+		return adap->uld[CXGB4_ULD_IPSEC].tx_handler(skb, dev);
+#endif /* CHELSIO_IPSEC_INLINE */
+
+#if IS_ENABLED(CONFIG_CHELSIO_TLS_DEVICE)
+	if (tls_is_skb_tx_device_offloaded(skb) &&
+	    (skb->len - skb_tcp_all_headers(skb)))
+		return adap->uld[CXGB4_ULD_KTLS].tx_handler(skb, dev);
+#endif /* CHELSIO_TLS_DEVICE */
+
 	qidx = skb_get_queue_mapping(skb);
-	q = &adap->sge.ethtxq[qidx + pi->first_qset];
+	if (ptp_enabled) {
+		if (!(adap->ptp_tx_skb)) {
+			skb_shinfo(skb)->tx_flags |= SKBTX_IN_PROGRESS;
+			adap->ptp_tx_skb = skb_get(skb);
+		} else {
+			goto out_free;
+		}
+		q = &adap->sge.ptptxq;
+	} else {
+		q = &adap->sge.ethtxq[qidx + pi->first_qset];
+	}
 
-	reclaim_completed_tx(adap, &q->q, true);
+	reclaim_completed_tx(adap, &q->q, -1, true);
 	cntrl = TXPKT_L4CSUM_DIS_F | TXPKT_IPCSUM_DIS_F;
 
 #ifdef CONFIG_CHELSIO_T4_FCOE
-	err = cxgb_fcoe_offload(skb, adap, pi, &cntrl);
-	if (unlikely(err == -ENOTSUPP))
+	ret = cxgb_fcoe_offload(skb, adap, pi, &cntrl);
+	if (unlikely(ret == -EOPNOTSUPP))
 		goto out_free;
 #endif /* CONFIG_CHELSIO_T4_FCOE */
 
-	flits = calc_tx_flits(skb);
+	chip_ver = CHELSIO_CHIP_VERSION(adap->params.chip);
+	flits = calc_tx_flits(skb, chip_ver);
 	ndesc = flits_to_desc(flits);
 	credits = txq_avail(&q->q) - ndesc;
 
@@ -1221,67 +1556,113 @@ out_free:	dev_kfree_skb_any(skb);
 		return NETDEV_TX_BUSY;
 	}
 
-	if (is_eth_imm(skb))
+	if (is_eth_imm(skb, chip_ver))
 		immediate = true;
 
+	if (skb->encapsulation && chip_ver > CHELSIO_T5)
+		tnl_type = cxgb_encap_offload_supported(skb);
+
+	last_desc = q->q.pidx + ndesc - 1;
+	if (last_desc >= q->q.size)
+		last_desc -= q->q.size;
+	sgl_sdesc = &q->q.sdesc[last_desc];
+
 	if (!immediate &&
-	    unlikely(map_skb(adap->pdev_dev, skb, addr) < 0)) {
+	    unlikely(cxgb4_map_skb(adap->pdev_dev, skb, sgl_sdesc->addr) < 0)) {
+		memset(sgl_sdesc->addr, 0, sizeof(sgl_sdesc->addr));
 		q->mapping_err++;
 		goto out_free;
 	}
 
 	wr_mid = FW_WR_LEN16_V(DIV_ROUND_UP(flits, 2));
 	if (unlikely(credits < ETHTXQ_STOP_THRES)) {
+		/* After we're done injecting the Work Request for this
+		 * packet, we'll be below our "stop threshold" so stop the TX
+		 * Queue now and schedule a request for an SGE Egress Queue
+		 * Update message. The queue will get started later on when
+		 * the firmware processes this Work Request and sends us an
+		 * Egress Queue Status Update message indicating that space
+		 * has opened up.
+		 */
 		eth_txq_stop(q);
-		wr_mid |= FW_WR_EQUEQ_F | FW_WR_EQUIQ_F;
+		if (chip_ver > CHELSIO_T5)
+			wr_mid |= FW_WR_EQUEQ_F | FW_WR_EQUIQ_F;
 	}
 
 	wr = (void *)&q->q.desc[q->q.pidx];
+	eowr = (void *)&q->q.desc[q->q.pidx];
 	wr->equiq_to_len16 = htonl(wr_mid);
 	wr->r3 = cpu_to_be64(0);
-	end = (u64 *)wr + flits;
+	if (skb_shinfo(skb)->gso_type & SKB_GSO_UDP_L4)
+		end = (u64 *)eowr + flits;
+	else
+		end = (u64 *)wr + flits;
 
 	len = immediate ? skb->len : 0;
-	ssi = skb_shinfo(skb);
-	if (ssi->gso_size) {
-		struct cpl_tx_pkt_lso *lso = (void *)wr;
-		bool v6 = (ssi->gso_type & SKB_GSO_TCPV6) != 0;
-		int l3hdr_len = skb_network_header_len(skb);
-		int eth_xtra_len = skb_network_offset(skb) - ETH_HLEN;
+	len += sizeof(*cpl);
+	if (ssi->gso_size && !(ssi->gso_type & SKB_GSO_UDP_L4)) {
+		struct cpl_tx_pkt_lso_core *lso = (void *)(wr + 1);
+		struct cpl_tx_tnl_lso *tnl_lso = (void *)(wr + 1);
 
-		len += sizeof(*lso);
-		wr->op_immdlen = htonl(FW_WR_OP_V(FW_ETH_TX_PKT_WR) |
-				       FW_WR_IMMDLEN_V(len));
-		lso->c.lso_ctrl = htonl(LSO_OPCODE_V(CPL_TX_PKT_LSO) |
-					LSO_FIRST_SLICE_F | LSO_LAST_SLICE_F |
-					LSO_IPV6_V(v6) |
-					LSO_ETHHDR_LEN_V(eth_xtra_len / 4) |
-					LSO_IPHDR_LEN_V(l3hdr_len / 4) |
-					LSO_TCPHDR_LEN_V(tcp_hdr(skb)->doff));
-		lso->c.ipid_ofst = htons(0);
-		lso->c.mss = htons(ssi->gso_size);
-		lso->c.seqno_offset = htonl(0);
-		if (is_t4(adap->params.chip))
-			lso->c.len = htonl(skb->len);
+		if (tnl_type)
+			len += sizeof(*tnl_lso);
 		else
-			lso->c.len = htonl(LSO_T5_XFER_SIZE_V(skb->len));
-		cpl = (void *)(lso + 1);
+			len += sizeof(*lso);
 
-		if (CHELSIO_CHIP_VERSION(adap->params.chip) <= CHELSIO_T5)
-			cntrl =	TXPKT_ETHHDR_LEN_V(eth_xtra_len);
-		else
-			cntrl = T6_TXPKT_ETHHDR_LEN_V(eth_xtra_len);
+		wr->op_immdlen = htonl(FW_WR_OP_V(FW_ETH_TX_PKT_WR) |
+				       FW_WR_IMMDLEN_V(len));
+		if (tnl_type) {
+			struct iphdr *iph = ip_hdr(skb);
 
-		cntrl |= TXPKT_CSUM_TYPE_V(v6 ?
-					   TX_CSUM_TCPIP6 : TX_CSUM_TCPIP) |
-			 TXPKT_IPHDR_LEN_V(l3hdr_len);
+			t6_fill_tnl_lso(skb, tnl_lso, tnl_type);
+			cpl = (void *)(tnl_lso + 1);
+			/* Driver is expected to compute partial checksum that
+			 * does not include the IP Total Length.
+			 */
+			if (iph->version == 4) {
+				iph->check = 0;
+				iph->tot_len = 0;
+				iph->check = ~ip_fast_csum((u8 *)iph, iph->ihl);
+			}
+			if (skb->ip_summed == CHECKSUM_PARTIAL)
+				cntrl = hwcsum(adap->params.chip, skb);
+		} else {
+			cpl = write_tso_wr(adap, skb, lso);
+			cntrl = hwcsum(adap->params.chip, skb);
+		}
+		sgl = (u64 *)(cpl + 1); /* sgl start here */
 		q->tso++;
 		q->tx_cso += ssi->gso_segs;
+	} else if (ssi->gso_size) {
+		u64 *start;
+		u32 hdrlen;
+
+		hdrlen = eth_get_headlen(dev, skb->data, skb_headlen(skb));
+		len += hdrlen;
+		wr->op_immdlen = cpu_to_be32(FW_WR_OP_V(FW_ETH_TX_EO_WR) |
+					     FW_ETH_TX_EO_WR_IMMDLEN_V(len));
+		cpl = write_eo_udp_wr(skb, eowr, hdrlen);
+		cntrl = hwcsum(adap->params.chip, skb);
+
+		start = (u64 *)(cpl + 1);
+		sgl = (u64 *)inline_tx_skb_header(skb, &q->q, (void *)start,
+						  hdrlen);
+		if (unlikely(start > sgl)) {
+			left = (u8 *)end - (u8 *)q->q.stat;
+			end = (void *)q->q.desc + left;
+		}
+		sgl_off = hdrlen;
+		q->uso++;
+		q->tx_cso += ssi->gso_segs;
 	} else {
-		len += sizeof(*cpl);
-		wr->op_immdlen = htonl(FW_WR_OP_V(FW_ETH_TX_PKT_WR) |
+		if (ptp_enabled)
+			op = FW_PTP_TX_PKT_WR;
+		else
+			op = FW_ETH_TX_PKT_WR;
+		wr->op_immdlen = htonl(FW_WR_OP_V(op) |
 				       FW_WR_IMMDLEN_V(len));
 		cpl = (void *)(wr + 1);
+		sgl = (u64 *)(cpl + 1);
 		if (skb->ip_summed == CHECKSUM_PARTIAL) {
 			cntrl = hwcsum(adap->params.chip, skb) |
 				TXPKT_IPCSUM_DIS_F;
@@ -1289,6 +1670,16 @@ out_free:	dev_kfree_skb_any(skb);
 		}
 	}
 
+	if (unlikely((u8 *)sgl >= (u8 *)q->q.stat)) {
+		/* If current position is already at the end of the
+		 * txq, reset the current to point to start of the queue
+		 * and update the end ptr as well.
+		 */
+		left = (u8 *)end - (u8 *)q->q.stat;
+		end = (void *)q->q.desc + left;
+		sgl = (void *)q->q.desc;
+	}
+
 	if (skb_vlan_tag_present(skb)) {
 		q->vlan_ins++;
 		cntrl |= TXPKT_VLAN_VLD_F | TXPKT_VLAN_V(skb_vlan_tag_get(skb));
@@ -1301,6 +1692,8 @@ out_free:	dev_kfree_skb_any(skb);
 
 	ctrl0 = TXPKT_OPCODE_V(CPL_TX_PKT_XT) | TXPKT_INTF_V(pi->tx_chan) |
 		TXPKT_PF_V(adap->pf);
+	if (ptp_enabled)
+		ctrl0 |= TXPKT_TSTAMP_F;
 #ifdef CONFIG_CHELSIO_T4_DCB
 	if (is_t4(adap->params.chip))
 		ctrl0 |= TXPKT_OVLAN_IDX_V(q->dcb_prio);
@@ -1312,40 +1705,396 @@ out_free:	dev_kfree_skb_any(skb);
 	cpl->len = htons(skb->len);
 	cpl->ctrl1 = cpu_to_be64(cntrl);
 
+	skb_tx_timestamp(skb);
+
 	if (immediate) {
-		inline_tx_skb(skb, &q->q, cpl + 1);
+		cxgb4_inline_tx_skb(skb, &q->q, sgl);
 		dev_consume_skb_any(skb);
 	} else {
-		int last_desc;
-
-		write_sgl(skb, &q->q, (struct ulptx_sgl *)(cpl + 1), end, 0,
-			  addr);
+		cxgb4_write_sgl(skb, &q->q, (void *)sgl, end, sgl_off,
+				sgl_sdesc->addr);
 		skb_orphan(skb);
-
-		last_desc = q->q.pidx + ndesc - 1;
-		if (last_desc >= q->q.size)
-			last_desc -= q->q.size;
-		q->q.sdesc[last_desc].skb = skb;
-		q->q.sdesc[last_desc].sgl = (struct ulptx_sgl *)(cpl + 1);
+		sgl_sdesc->skb = skb;
 	}
 
 	txq_advance(&q->q, ndesc);
 
-	ring_tx_db(adap, &q->q, ndesc);
+	cxgb4_ring_tx_db(adap, &q->q, ndesc);
+	return NETDEV_TX_OK;
+
+out_free:
+	dev_kfree_skb_any(skb);
 	return NETDEV_TX_OK;
 }
 
+/* Constants ... */
+enum {
+	/* Egress Queue sizes, producer and consumer indices are all in units
+	 * of Egress Context Units bytes.  Note that as far as the hardware is
+	 * concerned, the free list is an Egress Queue (the host produces free
+	 * buffers which the hardware consumes) and free list entries are
+	 * 64-bit PCI DMA addresses.
+	 */
+	EQ_UNIT = SGE_EQ_IDXSIZE,
+	FL_PER_EQ_UNIT = EQ_UNIT / sizeof(__be64),
+	TXD_PER_EQ_UNIT = EQ_UNIT / sizeof(__be64),
+
+	T4VF_ETHTXQ_MAX_HDR = (sizeof(struct fw_eth_tx_pkt_vm_wr) +
+			       sizeof(struct cpl_tx_pkt_lso_core) +
+			       sizeof(struct cpl_tx_pkt_core)) / sizeof(__be64),
+};
+
+/**
+ *	t4vf_is_eth_imm - can an Ethernet packet be sent as immediate data?
+ *	@skb: the packet
+ *
+ *	Returns whether an Ethernet packet is small enough to fit completely as
+ *	immediate data.
+ */
+static inline int t4vf_is_eth_imm(const struct sk_buff *skb)
+{
+	/* The VF Driver uses the FW_ETH_TX_PKT_VM_WR firmware Work Request
+	 * which does not accommodate immediate data.  We could dike out all
+	 * of the support code for immediate data but that would tie our hands
+	 * too much if we ever want to enhace the firmware.  It would also
+	 * create more differences between the PF and VF Drivers.
+	 */
+	return false;
+}
+
 /**
- *	reclaim_completed_tx_imm - reclaim completed control-queue Tx descs
- *	@q: the SGE control Tx queue
+ *	t4vf_calc_tx_flits - calculate the number of flits for a packet TX WR
+ *	@skb: the packet
  *
- *	This is a variant of reclaim_completed_tx() that is used for Tx queues
- *	that send only immediate data (presently just the control queues) and
- *	thus do not have any sk_buffs to release.
+ *	Returns the number of flits needed for a TX Work Request for the
+ *	given Ethernet packet, including the needed WR and CPL headers.
+ */
+static inline unsigned int t4vf_calc_tx_flits(const struct sk_buff *skb)
+{
+	unsigned int flits;
+
+	/* If the skb is small enough, we can pump it out as a work request
+	 * with only immediate data.  In that case we just have to have the
+	 * TX Packet header plus the skb data in the Work Request.
+	 */
+	if (t4vf_is_eth_imm(skb))
+		return DIV_ROUND_UP(skb->len + sizeof(struct cpl_tx_pkt),
+				    sizeof(__be64));
+
+	/* Otherwise, we're going to have to construct a Scatter gather list
+	 * of the skb body and fragments.  We also include the flits necessary
+	 * for the TX Packet Work Request and CPL.  We always have a firmware
+	 * Write Header (incorporated as part of the cpl_tx_pkt_lso and
+	 * cpl_tx_pkt structures), followed by either a TX Packet Write CPL
+	 * message or, if we're doing a Large Send Offload, an LSO CPL message
+	 * with an embedded TX Packet Write CPL message.
+	 */
+	flits = sgl_len(skb_shinfo(skb)->nr_frags + 1);
+	if (skb_shinfo(skb)->gso_size)
+		flits += (sizeof(struct fw_eth_tx_pkt_vm_wr) +
+			  sizeof(struct cpl_tx_pkt_lso_core) +
+			  sizeof(struct cpl_tx_pkt_core)) / sizeof(__be64);
+	else
+		flits += (sizeof(struct fw_eth_tx_pkt_vm_wr) +
+			  sizeof(struct cpl_tx_pkt_core)) / sizeof(__be64);
+	return flits;
+}
+
+/**
+ *	cxgb4_vf_eth_xmit - add a packet to an Ethernet TX queue
+ *	@skb: the packet
+ *	@dev: the egress net device
+ *
+ *	Add a packet to an SGE Ethernet TX queue.  Runs with softirqs disabled.
+ */
+static netdev_tx_t cxgb4_vf_eth_xmit(struct sk_buff *skb,
+				     struct net_device *dev)
+{
+	unsigned int last_desc, flits, ndesc;
+	const struct skb_shared_info *ssi;
+	struct fw_eth_tx_pkt_vm_wr *wr;
+	struct tx_sw_desc *sgl_sdesc;
+	struct cpl_tx_pkt_core *cpl;
+	const struct port_info *pi;
+	struct sge_eth_txq *txq;
+	struct adapter *adapter;
+	int qidx, credits, ret;
+	size_t fw_hdr_copy_len;
+	unsigned int chip_ver;
+	u64 cntrl, *end;
+	u32 wr_mid;
+
+	/* The chip minimum packet length is 10 octets but the firmware
+	 * command that we are using requires that we copy the Ethernet header
+	 * (including the VLAN tag) into the header so we reject anything
+	 * smaller than that ...
+	 */
+	BUILD_BUG_ON(sizeof(wr->firmware) !=
+		     (sizeof(wr->ethmacdst) + sizeof(wr->ethmacsrc) +
+		      sizeof(wr->ethtype) + sizeof(wr->vlantci)));
+	fw_hdr_copy_len = sizeof(wr->firmware);
+	ret = cxgb4_validate_skb(skb, dev, fw_hdr_copy_len);
+	if (ret)
+		goto out_free;
+
+	/* Figure out which TX Queue we're going to use. */
+	pi = netdev_priv(dev);
+	adapter = pi->adapter;
+	qidx = skb_get_queue_mapping(skb);
+	WARN_ON(qidx >= pi->nqsets);
+	txq = &adapter->sge.ethtxq[pi->first_qset + qidx];
+
+	/* Take this opportunity to reclaim any TX Descriptors whose DMA
+	 * transfers have completed.
+	 */
+	reclaim_completed_tx(adapter, &txq->q, -1, true);
+
+	/* Calculate the number of flits and TX Descriptors we're going to
+	 * need along with how many TX Descriptors will be left over after
+	 * we inject our Work Request.
+	 */
+	flits = t4vf_calc_tx_flits(skb);
+	ndesc = flits_to_desc(flits);
+	credits = txq_avail(&txq->q) - ndesc;
+
+	if (unlikely(credits < 0)) {
+		/* Not enough room for this packet's Work Request.  Stop the
+		 * TX Queue and return a "busy" condition.  The queue will get
+		 * started later on when the firmware informs us that space
+		 * has opened up.
+		 */
+		eth_txq_stop(txq);
+		dev_err(adapter->pdev_dev,
+			"%s: TX ring %u full while queue awake!\n",
+			dev->name, qidx);
+		return NETDEV_TX_BUSY;
+	}
+
+	last_desc = txq->q.pidx + ndesc - 1;
+	if (last_desc >= txq->q.size)
+		last_desc -= txq->q.size;
+	sgl_sdesc = &txq->q.sdesc[last_desc];
+
+	if (!t4vf_is_eth_imm(skb) &&
+	    unlikely(cxgb4_map_skb(adapter->pdev_dev, skb,
+				   sgl_sdesc->addr) < 0)) {
+		/* We need to map the skb into PCI DMA space (because it can't
+		 * be in-lined directly into the Work Request) and the mapping
+		 * operation failed.  Record the error and drop the packet.
+		 */
+		memset(sgl_sdesc->addr, 0, sizeof(sgl_sdesc->addr));
+		txq->mapping_err++;
+		goto out_free;
+	}
+
+	chip_ver = CHELSIO_CHIP_VERSION(adapter->params.chip);
+	wr_mid = FW_WR_LEN16_V(DIV_ROUND_UP(flits, 2));
+	if (unlikely(credits < ETHTXQ_STOP_THRES)) {
+		/* After we're done injecting the Work Request for this
+		 * packet, we'll be below our "stop threshold" so stop the TX
+		 * Queue now and schedule a request for an SGE Egress Queue
+		 * Update message.  The queue will get started later on when
+		 * the firmware processes this Work Request and sends us an
+		 * Egress Queue Status Update message indicating that space
+		 * has opened up.
+		 */
+		eth_txq_stop(txq);
+		if (chip_ver > CHELSIO_T5)
+			wr_mid |= FW_WR_EQUEQ_F | FW_WR_EQUIQ_F;
+	}
+
+	/* Start filling in our Work Request.  Note that we do _not_ handle
+	 * the WR Header wrapping around the TX Descriptor Ring.  If our
+	 * maximum header size ever exceeds one TX Descriptor, we'll need to
+	 * do something else here.
+	 */
+	WARN_ON(DIV_ROUND_UP(T4VF_ETHTXQ_MAX_HDR, TXD_PER_EQ_UNIT) > 1);
+	wr = (void *)&txq->q.desc[txq->q.pidx];
+	wr->equiq_to_len16 = cpu_to_be32(wr_mid);
+	wr->r3[0] = cpu_to_be32(0);
+	wr->r3[1] = cpu_to_be32(0);
+	skb_copy_from_linear_data(skb, &wr->firmware, fw_hdr_copy_len);
+	end = (u64 *)wr + flits;
+
+	/* If this is a Large Send Offload packet we'll put in an LSO CPL
+	 * message with an encapsulated TX Packet CPL message.  Otherwise we
+	 * just use a TX Packet CPL message.
+	 */
+	ssi = skb_shinfo(skb);
+	if (ssi->gso_size) {
+		struct cpl_tx_pkt_lso_core *lso = (void *)(wr + 1);
+		bool v6 = (ssi->gso_type & SKB_GSO_TCPV6) != 0;
+		int l3hdr_len = skb_network_header_len(skb);
+		int eth_xtra_len = skb_network_offset(skb) - ETH_HLEN;
+
+		wr->op_immdlen =
+			cpu_to_be32(FW_WR_OP_V(FW_ETH_TX_PKT_VM_WR) |
+				    FW_WR_IMMDLEN_V(sizeof(*lso) +
+						    sizeof(*cpl)));
+		 /* Fill in the LSO CPL message. */
+		lso->lso_ctrl =
+			cpu_to_be32(LSO_OPCODE_V(CPL_TX_PKT_LSO) |
+				    LSO_FIRST_SLICE_F |
+				    LSO_LAST_SLICE_F |
+				    LSO_IPV6_V(v6) |
+				    LSO_ETHHDR_LEN_V(eth_xtra_len / 4) |
+				    LSO_IPHDR_LEN_V(l3hdr_len / 4) |
+				    LSO_TCPHDR_LEN_V(tcp_hdr(skb)->doff));
+		lso->ipid_ofst = cpu_to_be16(0);
+		lso->mss = cpu_to_be16(ssi->gso_size);
+		lso->seqno_offset = cpu_to_be32(0);
+		if (is_t4(adapter->params.chip))
+			lso->len = cpu_to_be32(skb->len);
+		else
+			lso->len = cpu_to_be32(LSO_T5_XFER_SIZE_V(skb->len));
+
+		/* Set up TX Packet CPL pointer, control word and perform
+		 * accounting.
+		 */
+		cpl = (void *)(lso + 1);
+
+		if (chip_ver <= CHELSIO_T5)
+			cntrl = TXPKT_ETHHDR_LEN_V(eth_xtra_len);
+		else
+			cntrl = T6_TXPKT_ETHHDR_LEN_V(eth_xtra_len);
+
+		cntrl |= TXPKT_CSUM_TYPE_V(v6 ?
+					   TX_CSUM_TCPIP6 : TX_CSUM_TCPIP) |
+			 TXPKT_IPHDR_LEN_V(l3hdr_len);
+		txq->tso++;
+		txq->tx_cso += ssi->gso_segs;
+	} else {
+		int len;
+
+		len = (t4vf_is_eth_imm(skb)
+		       ? skb->len + sizeof(*cpl)
+		       : sizeof(*cpl));
+		wr->op_immdlen =
+			cpu_to_be32(FW_WR_OP_V(FW_ETH_TX_PKT_VM_WR) |
+				    FW_WR_IMMDLEN_V(len));
+
+		/* Set up TX Packet CPL pointer, control word and perform
+		 * accounting.
+		 */
+		cpl = (void *)(wr + 1);
+		if (skb->ip_summed == CHECKSUM_PARTIAL) {
+			cntrl = hwcsum(adapter->params.chip, skb) |
+				TXPKT_IPCSUM_DIS_F;
+			txq->tx_cso++;
+		} else {
+			cntrl = TXPKT_L4CSUM_DIS_F | TXPKT_IPCSUM_DIS_F;
+		}
+	}
+
+	/* If there's a VLAN tag present, add that to the list of things to
+	 * do in this Work Request.
+	 */
+	if (skb_vlan_tag_present(skb)) {
+		txq->vlan_ins++;
+		cntrl |= TXPKT_VLAN_VLD_F | TXPKT_VLAN_V(skb_vlan_tag_get(skb));
+	}
+
+	 /* Fill in the TX Packet CPL message header. */
+	cpl->ctrl0 = cpu_to_be32(TXPKT_OPCODE_V(CPL_TX_PKT_XT) |
+				 TXPKT_INTF_V(pi->port_id) |
+				 TXPKT_PF_V(0));
+	cpl->pack = cpu_to_be16(0);
+	cpl->len = cpu_to_be16(skb->len);
+	cpl->ctrl1 = cpu_to_be64(cntrl);
+
+	/* Fill in the body of the TX Packet CPL message with either in-lined
+	 * data or a Scatter/Gather List.
+	 */
+	if (t4vf_is_eth_imm(skb)) {
+		/* In-line the packet's data and free the skb since we don't
+		 * need it any longer.
+		 */
+		cxgb4_inline_tx_skb(skb, &txq->q, cpl + 1);
+		dev_consume_skb_any(skb);
+	} else {
+		/* Write the skb's Scatter/Gather list into the TX Packet CPL
+		 * message and retain a pointer to the skb so we can free it
+		 * later when its DMA completes.  (We store the skb pointer
+		 * in the Software Descriptor corresponding to the last TX
+		 * Descriptor used by the Work Request.)
+		 *
+		 * The retained skb will be freed when the corresponding TX
+		 * Descriptors are reclaimed after their DMAs complete.
+		 * However, this could take quite a while since, in general,
+		 * the hardware is set up to be lazy about sending DMA
+		 * completion notifications to us and we mostly perform TX
+		 * reclaims in the transmit routine.
+		 *
+		 * This is good for performamce but means that we rely on new
+		 * TX packets arriving to run the destructors of completed
+		 * packets, which open up space in their sockets' send queues.
+		 * Sometimes we do not get such new packets causing TX to
+		 * stall.  A single UDP transmitter is a good example of this
+		 * situation.  We have a clean up timer that periodically
+		 * reclaims completed packets but it doesn't run often enough
+		 * (nor do we want it to) to prevent lengthy stalls.  A
+		 * solution to this problem is to run the destructor early,
+		 * after the packet is queued but before it's DMAd.  A con is
+		 * that we lie to socket memory accounting, but the amount of
+		 * extra memory is reasonable (limited by the number of TX
+		 * descriptors), the packets do actually get freed quickly by
+		 * new packets almost always, and for protocols like TCP that
+		 * wait for acks to really free up the data the extra memory
+		 * is even less.  On the positive side we run the destructors
+		 * on the sending CPU rather than on a potentially different
+		 * completing CPU, usually a good thing.
+		 *
+		 * Run the destructor before telling the DMA engine about the
+		 * packet to make sure it doesn't complete and get freed
+		 * prematurely.
+		 */
+		struct ulptx_sgl *sgl = (struct ulptx_sgl *)(cpl + 1);
+		struct sge_txq *tq = &txq->q;
+
+		/* If the Work Request header was an exact multiple of our TX
+		 * Descriptor length, then it's possible that the starting SGL
+		 * pointer lines up exactly with the end of our TX Descriptor
+		 * ring.  If that's the case, wrap around to the beginning
+		 * here ...
+		 */
+		if (unlikely((void *)sgl == (void *)tq->stat)) {
+			sgl = (void *)tq->desc;
+			end = (void *)((void *)tq->desc +
+				       ((void *)end - (void *)tq->stat));
+		}
+
+		cxgb4_write_sgl(skb, tq, sgl, end, 0, sgl_sdesc->addr);
+		skb_orphan(skb);
+		sgl_sdesc->skb = skb;
+	}
+
+	/* Advance our internal TX Queue state, tell the hardware about
+	 * the new TX descriptors and return success.
+	 */
+	txq_advance(&txq->q, ndesc);
+
+	cxgb4_ring_tx_db(adapter, &txq->q, ndesc);
+	return NETDEV_TX_OK;
+
+out_free:
+	/* An error of some sort happened.  Free the TX skb and tell the
+	 * OS that we've "dealt" with the packet ...
+	 */
+	dev_kfree_skb_any(skb);
+	return NETDEV_TX_OK;
+}
+
+/**
+ * reclaim_completed_tx_imm - reclaim completed control-queue Tx descs
+ * @q: the SGE control Tx queue
+ *
+ * This is a variant of cxgb4_reclaim_completed_tx() that is used
+ * for Tx queues that send only immediate data (presently just
+ * the control queues) and	thus do not have any sk_buffs to release.
  */
 static inline void reclaim_completed_tx_imm(struct sge_txq *q)
 {
-	int hw_cidx = ntohs(ACCESS_ONCE(q->stat->cidx));
+	int hw_cidx = ntohs(READ_ONCE(q->stat->cidx));
 	int reclaim = hw_cidx - q->cidx;
 
 	if (reclaim < 0)
@@ -1355,6 +2104,518 @@ static inline void reclaim_completed_tx_imm(struct sge_txq *q)
 	q->cidx = hw_cidx;
 }
 
+static inline void eosw_txq_advance_index(u32 *idx, u32 n, u32 max)
+{
+	u32 val = *idx + n;
+
+	if (val >= max)
+		val -= max;
+
+	*idx = val;
+}
+
+void cxgb4_eosw_txq_free_desc(struct adapter *adap,
+			      struct sge_eosw_txq *eosw_txq, u32 ndesc)
+{
+	struct tx_sw_desc *d;
+
+	d = &eosw_txq->desc[eosw_txq->last_cidx];
+	while (ndesc--) {
+		if (d->skb) {
+			if (d->addr[0]) {
+				unmap_skb(adap->pdev_dev, d->skb, d->addr);
+				memset(d->addr, 0, sizeof(d->addr));
+			}
+			dev_consume_skb_any(d->skb);
+			d->skb = NULL;
+		}
+		eosw_txq_advance_index(&eosw_txq->last_cidx, 1,
+				       eosw_txq->ndesc);
+		d = &eosw_txq->desc[eosw_txq->last_cidx];
+	}
+}
+
+static inline void eosw_txq_advance(struct sge_eosw_txq *eosw_txq, u32 n)
+{
+	eosw_txq_advance_index(&eosw_txq->pidx, n, eosw_txq->ndesc);
+	eosw_txq->inuse += n;
+}
+
+static inline int eosw_txq_enqueue(struct sge_eosw_txq *eosw_txq,
+				   struct sk_buff *skb)
+{
+	if (eosw_txq->inuse == eosw_txq->ndesc)
+		return -ENOMEM;
+
+	eosw_txq->desc[eosw_txq->pidx].skb = skb;
+	return 0;
+}
+
+static inline struct sk_buff *eosw_txq_peek(struct sge_eosw_txq *eosw_txq)
+{
+	return eosw_txq->desc[eosw_txq->last_pidx].skb;
+}
+
+static inline u8 ethofld_calc_tx_flits(struct adapter *adap,
+				       struct sk_buff *skb, u32 hdr_len)
+{
+	u8 flits, nsgl = 0;
+	u32 wrlen;
+
+	wrlen = sizeof(struct fw_eth_tx_eo_wr) + sizeof(struct cpl_tx_pkt_core);
+	if (skb_shinfo(skb)->gso_size &&
+	    !(skb_shinfo(skb)->gso_type & SKB_GSO_UDP_L4))
+		wrlen += sizeof(struct cpl_tx_pkt_lso_core);
+
+	wrlen += roundup(hdr_len, 16);
+
+	/* Packet headers + WR + CPLs */
+	flits = DIV_ROUND_UP(wrlen, 8);
+
+	if (skb_shinfo(skb)->nr_frags > 0) {
+		if (skb_headlen(skb) - hdr_len)
+			nsgl = sgl_len(skb_shinfo(skb)->nr_frags + 1);
+		else
+			nsgl = sgl_len(skb_shinfo(skb)->nr_frags);
+	} else if (skb->len - hdr_len) {
+		nsgl = sgl_len(1);
+	}
+
+	return flits + nsgl;
+}
+
+static void *write_eo_wr(struct adapter *adap, struct sge_eosw_txq *eosw_txq,
+			 struct sk_buff *skb, struct fw_eth_tx_eo_wr *wr,
+			 u32 hdr_len, u32 wrlen)
+{
+	const struct skb_shared_info *ssi = skb_shinfo(skb);
+	struct cpl_tx_pkt_core *cpl;
+	u32 immd_len, wrlen16;
+	bool compl = false;
+	u8 ver, proto;
+
+	ver = ip_hdr(skb)->version;
+	proto = (ver == 6) ? ipv6_hdr(skb)->nexthdr : ip_hdr(skb)->protocol;
+
+	wrlen16 = DIV_ROUND_UP(wrlen, 16);
+	immd_len = sizeof(struct cpl_tx_pkt_core);
+	if (skb_shinfo(skb)->gso_size &&
+	    !(skb_shinfo(skb)->gso_type & SKB_GSO_UDP_L4))
+		immd_len += sizeof(struct cpl_tx_pkt_lso_core);
+	immd_len += hdr_len;
+
+	if (!eosw_txq->ncompl ||
+	    (eosw_txq->last_compl + wrlen16) >=
+	    (adap->params.ofldq_wr_cred / 2)) {
+		compl = true;
+		eosw_txq->ncompl++;
+		eosw_txq->last_compl = 0;
+	}
+
+	wr->op_immdlen = cpu_to_be32(FW_WR_OP_V(FW_ETH_TX_EO_WR) |
+				     FW_ETH_TX_EO_WR_IMMDLEN_V(immd_len) |
+				     FW_WR_COMPL_V(compl));
+	wr->equiq_to_len16 = cpu_to_be32(FW_WR_LEN16_V(wrlen16) |
+					 FW_WR_FLOWID_V(eosw_txq->hwtid));
+	wr->r3 = 0;
+	if (proto == IPPROTO_UDP) {
+		cpl = write_eo_udp_wr(skb, wr, hdr_len);
+	} else {
+		wr->u.tcpseg.type = FW_ETH_TX_EO_TYPE_TCPSEG;
+		wr->u.tcpseg.ethlen = skb_network_offset(skb);
+		wr->u.tcpseg.iplen = cpu_to_be16(skb_network_header_len(skb));
+		wr->u.tcpseg.tcplen = tcp_hdrlen(skb);
+		wr->u.tcpseg.tsclk_tsoff = 0;
+		wr->u.tcpseg.r4 = 0;
+		wr->u.tcpseg.r5 = 0;
+		wr->u.tcpseg.plen = cpu_to_be32(skb->len - hdr_len);
+
+		if (ssi->gso_size) {
+			struct cpl_tx_pkt_lso_core *lso = (void *)(wr + 1);
+
+			wr->u.tcpseg.mss = cpu_to_be16(ssi->gso_size);
+			cpl = write_tso_wr(adap, skb, lso);
+		} else {
+			wr->u.tcpseg.mss = cpu_to_be16(0xffff);
+			cpl = (void *)(wr + 1);
+		}
+	}
+
+	eosw_txq->cred -= wrlen16;
+	eosw_txq->last_compl += wrlen16;
+	return cpl;
+}
+
+static int ethofld_hard_xmit(struct net_device *dev,
+			     struct sge_eosw_txq *eosw_txq)
+{
+	struct port_info *pi = netdev2pinfo(dev);
+	struct adapter *adap = netdev2adap(dev);
+	u32 wrlen, wrlen16, hdr_len, data_len;
+	enum sge_eosw_state next_state;
+	u64 cntrl, *start, *end, *sgl;
+	struct sge_eohw_txq *eohw_txq;
+	struct cpl_tx_pkt_core *cpl;
+	struct fw_eth_tx_eo_wr *wr;
+	bool skip_eotx_wr = false;
+	struct tx_sw_desc *d;
+	struct sk_buff *skb;
+	int left, ret = 0;
+	u8 flits, ndesc;
+
+	eohw_txq = &adap->sge.eohw_txq[eosw_txq->hwqid];
+	spin_lock(&eohw_txq->lock);
+	reclaim_completed_tx_imm(&eohw_txq->q);
+
+	d = &eosw_txq->desc[eosw_txq->last_pidx];
+	skb = d->skb;
+
+	wr = (struct fw_eth_tx_eo_wr *)&eohw_txq->q.desc[eohw_txq->q.pidx];
+	if (unlikely(eosw_txq->state != CXGB4_EO_STATE_ACTIVE &&
+		     eosw_txq->last_pidx == eosw_txq->flowc_idx)) {
+		hdr_len = skb->len;
+		data_len = 0;
+		flits = DIV_ROUND_UP(hdr_len, 8);
+		if (eosw_txq->state == CXGB4_EO_STATE_FLOWC_OPEN_SEND)
+			next_state = CXGB4_EO_STATE_FLOWC_OPEN_REPLY;
+		else
+			next_state = CXGB4_EO_STATE_FLOWC_CLOSE_REPLY;
+		skip_eotx_wr = true;
+	} else {
+		hdr_len = eth_get_headlen(dev, skb->data, skb_headlen(skb));
+		data_len = skb->len - hdr_len;
+		flits = ethofld_calc_tx_flits(adap, skb, hdr_len);
+	}
+	ndesc = flits_to_desc(flits);
+	wrlen = flits * 8;
+	wrlen16 = DIV_ROUND_UP(wrlen, 16);
+
+	left = txq_avail(&eohw_txq->q) - ndesc;
+
+	/* If there are no descriptors left in hardware queues or no
+	 * CPL credits left in software queues, then wait for them
+	 * to come back and retry again. Note that we always request
+	 * for credits update via interrupt for every half credits
+	 * consumed. So, the interrupt will eventually restore the
+	 * credits and invoke the Tx path again.
+	 */
+	if (unlikely(left < 0 || wrlen16 > eosw_txq->cred)) {
+		ret = -ENOMEM;
+		goto out_unlock;
+	}
+
+	if (unlikely(skip_eotx_wr)) {
+		start = (u64 *)wr;
+		eosw_txq->state = next_state;
+		eosw_txq->cred -= wrlen16;
+		eosw_txq->ncompl++;
+		eosw_txq->last_compl = 0;
+		goto write_wr_headers;
+	}
+
+	cpl = write_eo_wr(adap, eosw_txq, skb, wr, hdr_len, wrlen);
+	cntrl = hwcsum(adap->params.chip, skb);
+	if (skb_vlan_tag_present(skb))
+		cntrl |= TXPKT_VLAN_VLD_F | TXPKT_VLAN_V(skb_vlan_tag_get(skb));
+
+	cpl->ctrl0 = cpu_to_be32(TXPKT_OPCODE_V(CPL_TX_PKT_XT) |
+				 TXPKT_INTF_V(pi->tx_chan) |
+				 TXPKT_PF_V(adap->pf));
+	cpl->pack = 0;
+	cpl->len = cpu_to_be16(skb->len);
+	cpl->ctrl1 = cpu_to_be64(cntrl);
+
+	start = (u64 *)(cpl + 1);
+
+write_wr_headers:
+	sgl = (u64 *)inline_tx_skb_header(skb, &eohw_txq->q, (void *)start,
+					  hdr_len);
+	if (data_len) {
+		ret = cxgb4_map_skb(adap->pdev_dev, skb, d->addr);
+		if (unlikely(ret)) {
+			memset(d->addr, 0, sizeof(d->addr));
+			eohw_txq->mapping_err++;
+			goto out_unlock;
+		}
+
+		end = (u64 *)wr + flits;
+		if (unlikely(start > sgl)) {
+			left = (u8 *)end - (u8 *)eohw_txq->q.stat;
+			end = (void *)eohw_txq->q.desc + left;
+		}
+
+		if (unlikely((u8 *)sgl >= (u8 *)eohw_txq->q.stat)) {
+			/* If current position is already at the end of the
+			 * txq, reset the current to point to start of the queue
+			 * and update the end ptr as well.
+			 */
+			left = (u8 *)end - (u8 *)eohw_txq->q.stat;
+
+			end = (void *)eohw_txq->q.desc + left;
+			sgl = (void *)eohw_txq->q.desc;
+		}
+
+		cxgb4_write_sgl(skb, &eohw_txq->q, (void *)sgl, end, hdr_len,
+				d->addr);
+	}
+
+	if (skb_shinfo(skb)->gso_size) {
+		if (skb_shinfo(skb)->gso_type & SKB_GSO_UDP_L4)
+			eohw_txq->uso++;
+		else
+			eohw_txq->tso++;
+		eohw_txq->tx_cso += skb_shinfo(skb)->gso_segs;
+	} else if (skb->ip_summed == CHECKSUM_PARTIAL) {
+		eohw_txq->tx_cso++;
+	}
+
+	if (skb_vlan_tag_present(skb))
+		eohw_txq->vlan_ins++;
+
+	txq_advance(&eohw_txq->q, ndesc);
+	skb_tx_timestamp(skb);
+	cxgb4_ring_tx_db(adap, &eohw_txq->q, ndesc);
+	eosw_txq_advance_index(&eosw_txq->last_pidx, 1, eosw_txq->ndesc);
+
+out_unlock:
+	spin_unlock(&eohw_txq->lock);
+	return ret;
+}
+
+static void ethofld_xmit(struct net_device *dev, struct sge_eosw_txq *eosw_txq)
+{
+	struct sk_buff *skb;
+	int pktcount, ret;
+
+	switch (eosw_txq->state) {
+	case CXGB4_EO_STATE_ACTIVE:
+	case CXGB4_EO_STATE_FLOWC_OPEN_SEND:
+	case CXGB4_EO_STATE_FLOWC_CLOSE_SEND:
+		pktcount = eosw_txq->pidx - eosw_txq->last_pidx;
+		if (pktcount < 0)
+			pktcount += eosw_txq->ndesc;
+		break;
+	case CXGB4_EO_STATE_FLOWC_OPEN_REPLY:
+	case CXGB4_EO_STATE_FLOWC_CLOSE_REPLY:
+	case CXGB4_EO_STATE_CLOSED:
+	default:
+		return;
+	}
+
+	while (pktcount--) {
+		skb = eosw_txq_peek(eosw_txq);
+		if (!skb) {
+			eosw_txq_advance_index(&eosw_txq->last_pidx, 1,
+					       eosw_txq->ndesc);
+			continue;
+		}
+
+		ret = ethofld_hard_xmit(dev, eosw_txq);
+		if (ret)
+			break;
+	}
+}
+
+static netdev_tx_t cxgb4_ethofld_xmit(struct sk_buff *skb,
+				      struct net_device *dev)
+{
+	struct cxgb4_tc_port_mqprio *tc_port_mqprio;
+	struct port_info *pi = netdev2pinfo(dev);
+	struct adapter *adap = netdev2adap(dev);
+	struct sge_eosw_txq *eosw_txq;
+	u32 qid;
+	int ret;
+
+	ret = cxgb4_validate_skb(skb, dev, ETH_HLEN);
+	if (ret)
+		goto out_free;
+
+	tc_port_mqprio = &adap->tc_mqprio->port_mqprio[pi->port_id];
+	qid = skb_get_queue_mapping(skb) - pi->nqsets;
+	eosw_txq = &tc_port_mqprio->eosw_txq[qid];
+	spin_lock_bh(&eosw_txq->lock);
+	if (eosw_txq->state != CXGB4_EO_STATE_ACTIVE)
+		goto out_unlock;
+
+	ret = eosw_txq_enqueue(eosw_txq, skb);
+	if (ret)
+		goto out_unlock;
+
+	/* SKB is queued for processing until credits are available.
+	 * So, call the destructor now and we'll free the skb later
+	 * after it has been successfully transmitted.
+	 */
+	skb_orphan(skb);
+
+	eosw_txq_advance(eosw_txq, 1);
+	ethofld_xmit(dev, eosw_txq);
+	spin_unlock_bh(&eosw_txq->lock);
+	return NETDEV_TX_OK;
+
+out_unlock:
+	spin_unlock_bh(&eosw_txq->lock);
+out_free:
+	dev_kfree_skb_any(skb);
+	return NETDEV_TX_OK;
+}
+
+netdev_tx_t t4_start_xmit(struct sk_buff *skb, struct net_device *dev)
+{
+	struct port_info *pi = netdev_priv(dev);
+	u16 qid = skb_get_queue_mapping(skb);
+
+	if (unlikely(pi->eth_flags & PRIV_FLAG_PORT_TX_VM))
+		return cxgb4_vf_eth_xmit(skb, dev);
+
+	if (unlikely(qid >= pi->nqsets))
+		return cxgb4_ethofld_xmit(skb, dev);
+
+	if (is_ptp_enabled(skb, dev)) {
+		struct adapter *adap = netdev2adap(dev);
+		netdev_tx_t ret;
+
+		spin_lock(&adap->ptp_lock);
+		ret = cxgb4_eth_xmit(skb, dev);
+		spin_unlock(&adap->ptp_lock);
+		return ret;
+	}
+
+	return cxgb4_eth_xmit(skb, dev);
+}
+
+static void eosw_txq_flush_pending_skbs(struct sge_eosw_txq *eosw_txq)
+{
+	int pktcount = eosw_txq->pidx - eosw_txq->last_pidx;
+	int pidx = eosw_txq->pidx;
+	struct sk_buff *skb;
+
+	if (!pktcount)
+		return;
+
+	if (pktcount < 0)
+		pktcount += eosw_txq->ndesc;
+
+	while (pktcount--) {
+		pidx--;
+		if (pidx < 0)
+			pidx += eosw_txq->ndesc;
+
+		skb = eosw_txq->desc[pidx].skb;
+		if (skb) {
+			dev_consume_skb_any(skb);
+			eosw_txq->desc[pidx].skb = NULL;
+			eosw_txq->inuse--;
+		}
+	}
+
+	eosw_txq->pidx = eosw_txq->last_pidx + 1;
+}
+
+/**
+ * cxgb4_ethofld_send_flowc - Send ETHOFLD flowc request to bind eotid to tc.
+ * @dev: netdevice
+ * @eotid: ETHOFLD tid to bind/unbind
+ * @tc: traffic class. If set to FW_SCHED_CLS_NONE, then unbinds the @eotid
+ *
+ * Send a FLOWC work request to bind an ETHOFLD TID to a traffic class.
+ * If @tc is set to FW_SCHED_CLS_NONE, then the @eotid is unbound from
+ * a traffic class.
+ */
+int cxgb4_ethofld_send_flowc(struct net_device *dev, u32 eotid, u32 tc)
+{
+	struct port_info *pi = netdev2pinfo(dev);
+	struct adapter *adap = netdev2adap(dev);
+	enum sge_eosw_state next_state;
+	struct sge_eosw_txq *eosw_txq;
+	u32 len, len16, nparams = 6;
+	struct fw_flowc_wr *flowc;
+	struct eotid_entry *entry;
+	struct sge_ofld_rxq *rxq;
+	struct sk_buff *skb;
+	int ret = 0;
+
+	len = struct_size(flowc, mnemval, nparams);
+	len16 = DIV_ROUND_UP(len, 16);
+
+	entry = cxgb4_lookup_eotid(&adap->tids, eotid);
+	if (!entry)
+		return -ENOMEM;
+
+	eosw_txq = (struct sge_eosw_txq *)entry->data;
+	if (!eosw_txq)
+		return -ENOMEM;
+
+	if (!(adap->flags & CXGB4_FW_OK)) {
+		/* Don't stall caller when access to FW is lost */
+		complete(&eosw_txq->completion);
+		return -EIO;
+	}
+
+	skb = alloc_skb(len, GFP_KERNEL);
+	if (!skb)
+		return -ENOMEM;
+
+	spin_lock_bh(&eosw_txq->lock);
+	if (tc != FW_SCHED_CLS_NONE) {
+		if (eosw_txq->state != CXGB4_EO_STATE_CLOSED)
+			goto out_free_skb;
+
+		next_state = CXGB4_EO_STATE_FLOWC_OPEN_SEND;
+	} else {
+		if (eosw_txq->state != CXGB4_EO_STATE_ACTIVE)
+			goto out_free_skb;
+
+		next_state = CXGB4_EO_STATE_FLOWC_CLOSE_SEND;
+	}
+
+	flowc = __skb_put(skb, len);
+	memset(flowc, 0, len);
+
+	rxq = &adap->sge.eohw_rxq[eosw_txq->hwqid];
+	flowc->flowid_len16 = cpu_to_be32(FW_WR_LEN16_V(len16) |
+					  FW_WR_FLOWID_V(eosw_txq->hwtid));
+	flowc->op_to_nparams = cpu_to_be32(FW_WR_OP_V(FW_FLOWC_WR) |
+					   FW_FLOWC_WR_NPARAMS_V(nparams) |
+					   FW_WR_COMPL_V(1));
+	flowc->mnemval[0].mnemonic = FW_FLOWC_MNEM_PFNVFN;
+	flowc->mnemval[0].val = cpu_to_be32(FW_PFVF_CMD_PFN_V(adap->pf));
+	flowc->mnemval[1].mnemonic = FW_FLOWC_MNEM_CH;
+	flowc->mnemval[1].val = cpu_to_be32(pi->tx_chan);
+	flowc->mnemval[2].mnemonic = FW_FLOWC_MNEM_PORT;
+	flowc->mnemval[2].val = cpu_to_be32(pi->tx_chan);
+	flowc->mnemval[3].mnemonic = FW_FLOWC_MNEM_IQID;
+	flowc->mnemval[3].val = cpu_to_be32(rxq->rspq.abs_id);
+	flowc->mnemval[4].mnemonic = FW_FLOWC_MNEM_SCHEDCLASS;
+	flowc->mnemval[4].val = cpu_to_be32(tc);
+	flowc->mnemval[5].mnemonic = FW_FLOWC_MNEM_EOSTATE;
+	flowc->mnemval[5].val = cpu_to_be32(tc == FW_SCHED_CLS_NONE ?
+					    FW_FLOWC_MNEM_EOSTATE_CLOSING :
+					    FW_FLOWC_MNEM_EOSTATE_ESTABLISHED);
+
+	/* Free up any pending skbs to ensure there's room for
+	 * termination FLOWC.
+	 */
+	if (tc == FW_SCHED_CLS_NONE)
+		eosw_txq_flush_pending_skbs(eosw_txq);
+
+	ret = eosw_txq_enqueue(eosw_txq, skb);
+	if (ret)
+		goto out_free_skb;
+
+	eosw_txq->state = next_state;
+	eosw_txq->flowc_idx = eosw_txq->pidx;
+	eosw_txq_advance(eosw_txq, 1);
+	ethofld_xmit(dev, eosw_txq);
+
+	spin_unlock_bh(&eosw_txq->lock);
+	return 0;
+
+out_free_skb:
+	dev_consume_skb_any(skb);
+	spin_unlock_bh(&eosw_txq->lock);
+	return ret;
+}
+
 /**
  *	is_imm - check whether a packet can be sent as immediate data
  *	@skb: the packet
@@ -1386,6 +2647,84 @@ static void ctrlq_check_stop(struct sge_ctrl_txq *q, struct fw_wr_hdr *wr)
 	}
 }
 
+#define CXGB4_SELFTEST_LB_STR "CHELSIO_SELFTEST"
+
+int cxgb4_selftest_lb_pkt(struct net_device *netdev)
+{
+	struct port_info *pi = netdev_priv(netdev);
+	struct adapter *adap = pi->adapter;
+	struct cxgb4_ethtool_lb_test *lb;
+	int ret, i = 0, pkt_len, credits;
+	struct fw_eth_tx_pkt_wr *wr;
+	struct cpl_tx_pkt_core *cpl;
+	u32 ctrl0, ndesc, flits;
+	struct sge_eth_txq *q;
+	u8 *sgl;
+
+	pkt_len = ETH_HLEN + sizeof(CXGB4_SELFTEST_LB_STR);
+
+	flits = DIV_ROUND_UP(pkt_len + sizeof(*cpl) + sizeof(*wr),
+			     sizeof(__be64));
+	ndesc = flits_to_desc(flits);
+
+	lb = &pi->ethtool_lb;
+	lb->loopback = 1;
+
+	q = &adap->sge.ethtxq[pi->first_qset];
+	__netif_tx_lock_bh(q->txq);
+
+	reclaim_completed_tx(adap, &q->q, -1, true);
+	credits = txq_avail(&q->q) - ndesc;
+	if (unlikely(credits < 0)) {
+		__netif_tx_unlock_bh(q->txq);
+		return -ENOMEM;
+	}
+
+	wr = (void *)&q->q.desc[q->q.pidx];
+	memset(wr, 0, sizeof(struct tx_desc));
+
+	wr->op_immdlen = htonl(FW_WR_OP_V(FW_ETH_TX_PKT_WR) |
+			       FW_WR_IMMDLEN_V(pkt_len +
+			       sizeof(*cpl)));
+	wr->equiq_to_len16 = htonl(FW_WR_LEN16_V(DIV_ROUND_UP(flits, 2)));
+	wr->r3 = cpu_to_be64(0);
+
+	cpl = (void *)(wr + 1);
+	sgl = (u8 *)(cpl + 1);
+
+	ctrl0 = TXPKT_OPCODE_V(CPL_TX_PKT_XT) | TXPKT_PF_V(adap->pf) |
+		TXPKT_INTF_V(pi->tx_chan + 4);
+
+	cpl->ctrl0 = htonl(ctrl0);
+	cpl->pack = htons(0);
+	cpl->len = htons(pkt_len);
+	cpl->ctrl1 = cpu_to_be64(TXPKT_L4CSUM_DIS_F | TXPKT_IPCSUM_DIS_F);
+
+	eth_broadcast_addr(sgl);
+	i += ETH_ALEN;
+	ether_addr_copy(&sgl[i], netdev->dev_addr);
+	i += ETH_ALEN;
+
+	snprintf(&sgl[i], sizeof(CXGB4_SELFTEST_LB_STR), "%s",
+		 CXGB4_SELFTEST_LB_STR);
+
+	init_completion(&lb->completion);
+	txq_advance(&q->q, ndesc);
+	cxgb4_ring_tx_db(adap, &q->q, ndesc);
+	__netif_tx_unlock_bh(q->txq);
+
+	/* wait for the pkt to return */
+	ret = wait_for_completion_timeout(&lb->completion, 10 * HZ);
+	if (!ret)
+		ret = -ETIMEDOUT;
+	else
+		ret = lb->result;
+
+	lb->loopback = 0;
+
+	return ret;
+}
+
 /**
  *	ctrl_xmit - send a packet through an SGE control Tx queue
  *	@q: the control queue
@@ -1416,13 +2755,13 @@ static int ctrl_xmit(struct sge_ctrl_txq *q, struct sk_buff *skb)
 	}
 
 	wr = (struct fw_wr_hdr *)&q->q.desc[q->q.pidx];
-	inline_tx_skb(skb, &q->q, wr);
+	cxgb4_inline_tx_skb(skb, &q->q, wr);
 
 	txq_advance(&q->q, ndesc);
 	if (unlikely(txq_avail(&q->q) < TXQ_STOP_THRES))
 		ctrlq_check_stop(q, wr);
 
-	ring_tx_db(q->adap, &q->q, ndesc);
+	cxgb4_ring_tx_db(q->adap, &q->q, ndesc);
 	spin_unlock(&q->sendq.lock);
 
 	kfree_skb(skb);
@@ -1431,15 +2770,15 @@ static int ctrl_xmit(struct sge_ctrl_txq *q, struct sk_buff *skb)
 
 /**
  *	restart_ctrlq - restart a suspended control queue
- *	@data: the control queue to restart
+ *	@t: pointer to the tasklet associated with this handler
  *
  *	Resumes transmission on a suspended Tx control queue.
  */
-static void restart_ctrlq(unsigned long data)
+static void restart_ctrlq(struct tasklet_struct *t)
 {
 	struct sk_buff *skb;
 	unsigned int written = 0;
-	struct sge_ctrl_txq *q = (struct sge_ctrl_txq *)data;
+	struct sge_ctrl_txq *q = from_tasklet(q, t, qresume_tsk);
 
 	spin_lock(&q->sendq.lock);
 	reclaim_completed_tx_imm(&q->q);
@@ -1457,7 +2796,7 @@ static void restart_ctrlq(unsigned long data)
 		txq_advance(&q->q, ndesc);
 		spin_unlock(&q->sendq.lock);
 
-		inline_tx_skb(skb, &q->q, wr);
+		cxgb4_inline_tx_skb(skb, &q->q, wr);
 		kfree_skb(skb);
 
 		if (unlikely(txq_avail(&q->q) < TXQ_STOP_THRES)) {
@@ -1470,14 +2809,15 @@ static void restart_ctrlq(unsigned long data)
 			}
 		}
 		if (written > 16) {
-			ring_tx_db(q->adap, &q->q, written);
+			cxgb4_ring_tx_db(q->adap, &q->q, written);
 			written = 0;
 		}
 		spin_lock(&q->sendq.lock);
 	}
 	q->full = 0;
-ringdb: if (written)
-		ring_tx_db(q->adap, &q->q, written);
+ringdb:
+	if (written)
+		cxgb4_ring_tx_db(q->adap, &q->q, written);
 	spin_unlock(&q->sendq.lock);
 }
 
@@ -1503,11 +2843,22 @@ int t4_mgmt_tx(struct adapter *adap, struct sk_buff *skb)
  *	@skb: the packet
  *
  *	Returns true if a packet can be sent as an offload WR with immediate
- *	data.  We currently use the same limit as for Ethernet packets.
+ *	data.
+ *	FW_OFLD_TX_DATA_WR limits the payload to 255 bytes due to 8-bit field.
+ *      However, FW_ULPTX_WR commands have a 256 byte immediate only
+ *      payload limit.
  */
 static inline int is_ofld_imm(const struct sk_buff *skb)
 {
-	return skb->len <= MAX_IMM_TX_PKT_LEN;
+	struct work_request_hdr *req = (struct work_request_hdr *)skb->data;
+	unsigned long opcode = FW_WR_OP_G(ntohl(req->wr_hi));
+
+	if (unlikely(opcode == FW_ULPTX_WR))
+		return skb->len <= MAX_IMM_ULPTX_WR_LEN;
+	else if (opcode == FW_CRYPTO_LOOKASIDE_WR)
+		return skb->len <= SGE_MAX_WR_LEN;
+	else
+		return skb->len <= MAX_IMM_OFLD_TX_DATA_WR_LEN;
 }
 
 /**
@@ -1534,7 +2885,6 @@ static inline unsigned int calc_tx_flits_ofld(const struct sk_buff *skb)
 
 /**
  *	txq_stop_maperr - stop a Tx queue due to I/O MMU exhaustion
- *	@adap: the adapter
  *	@q: the queue to stop
  *
  *	Mark a Tx queue stopped due to I/O MMU exhaustion and resulting
@@ -1552,15 +2902,13 @@ static void txq_stop_maperr(struct sge_uld_txq *q)
 /**
  *	ofldtxq_stop - stop an offload Tx queue that has become full
  *	@q: the queue to stop
- *	@skb: the packet causing the queue to become full
+ *	@wr: the Work Request causing the queue to become full
  *
  *	Stops an offload Tx queue that has become full and modifies the packet
  *	being written to request a wakeup.
  */
-static void ofldtxq_stop(struct sge_uld_txq *q, struct sk_buff *skb)
+static void ofldtxq_stop(struct sge_uld_txq *q, struct fw_wr_hdr *wr)
 {
-	struct fw_wr_hdr *wr = (struct fw_wr_hdr *)skb->data;
-
 	wr->lo |= htonl(FW_WR_EQUEQ_F | FW_WR_EQUIQ_F);
 	q->q.stops++;
 	q->full = 1;
@@ -1585,6 +2933,7 @@ static void ofldtxq_stop(struct sge_uld_txq *q, struct sk_buff *skb)
  *	is ever running at a time ...
  */
 static void service_ofldq(struct sge_uld_txq *q)
+	__must_hold(&q->sendq.lock)
 {
 	u64 *pos, *before, *end;
 	int credits;
@@ -1614,20 +2963,20 @@ static void service_ofldq(struct sge_uld_txq *q)
 		 */
 		spin_unlock(&q->sendq.lock);
 
-		reclaim_completed_tx(q->adap, &q->q, false);
+		cxgb4_reclaim_completed_tx(q->adap, &q->q, false);
 
 		flits = skb->priority;                /* previously saved */
 		ndesc = flits_to_desc(flits);
 		credits = txq_avail(&q->q) - ndesc;
 		BUG_ON(credits < 0);
 		if (unlikely(credits < TXQ_STOP_THRES))
-			ofldtxq_stop(q, skb);
+			ofldtxq_stop(q, (struct fw_wr_hdr *)skb->data);
 
 		pos = (u64 *)&q->q.desc[q->q.pidx];
 		if (is_ofld_imm(skb))
-			inline_tx_skb(skb, &q->q, pos);
-		else if (map_skb(q->adap->pdev_dev, skb,
-				 (dma_addr_t *)skb->head)) {
+			cxgb4_inline_tx_skb(skb, &q->q, pos);
+		else if (cxgb4_map_skb(q->adap->pdev_dev, skb,
+				       (dma_addr_t *)skb->head)) {
 			txq_stop_maperr(q);
 			spin_lock(&q->sendq.lock);
 			break;
@@ -1658,9 +3007,9 @@ static void service_ofldq(struct sge_uld_txq *q)
 				pos = (void *)txq->desc;
 			}
 
-			write_sgl(skb, &q->q, (void *)pos,
-				  end, hdr_len,
-				  (dma_addr_t *)skb->head);
+			cxgb4_write_sgl(skb, &q->q, (void *)pos,
+					end, hdr_len,
+					(dma_addr_t *)skb->head);
 #ifdef CONFIG_NEED_DMA_MAP_STATE
 			skb->dev = q->adap->port[0];
 			skb->destructor = deferred_unmap_destructor;
@@ -1674,7 +3023,7 @@ static void service_ofldq(struct sge_uld_txq *q)
 		txq_advance(&q->q, ndesc);
 		written += ndesc;
 		if (unlikely(written > 32)) {
-			ring_tx_db(q->adap, &q->q, written);
+			cxgb4_ring_tx_db(q->adap, &q->q, written);
 			written = 0;
 		}
 
@@ -1689,7 +3038,7 @@ static void service_ofldq(struct sge_uld_txq *q)
 			kfree_skb(skb);
 	}
 	if (likely(written))
-		ring_tx_db(q->adap, &q->q, written);
+		cxgb4_ring_tx_db(q->adap, &q->q, written);
 
 	/*Indicate that no thread is processing the Pending Send Queue
 	 * currently.
@@ -1727,13 +3076,13 @@ static int ofld_xmit(struct sge_uld_txq *q, struct sk_buff *skb)
 
 /**
  *	restart_ofldq - restart a suspended offload queue
- *	@data: the offload queue to restart
+ *	@t: pointer to the tasklet associated with this handler
  *
  *	Resumes transmission on a suspended Tx offload queue.
  */
-static void restart_ofldq(unsigned long data)
+static void restart_ofldq(struct tasklet_struct *t)
 {
-	struct sge_uld_txq *q = (struct sge_uld_txq *)data;
+	struct sge_uld_txq *q = from_tasklet(q, t, qresume_tsk);
 
 	spin_lock(&q->sendq.lock);
 	q->full = 0;            /* the queue actually is completely empty now */
@@ -1782,6 +3131,7 @@ static inline int uld_send(struct adapter *adap, struct sk_buff *skb,
 	txq_info = adap->sge.uld_txq_info[tx_uld_type];
 	if (unlikely(!txq_info)) {
 		WARN_ON(true);
+		kfree_skb(skb);
 		return NET_XMIT_DROP;
 	}
 
@@ -1822,6 +3172,103 @@ int cxgb4_ofld_send(struct net_device *dev, struct sk_buff *skb)
 }
 EXPORT_SYMBOL(cxgb4_ofld_send);
 
+static void *inline_tx_header(const void *src,
+			      const struct sge_txq *q,
+			      void *pos, int length)
+{
+	int left = (void *)q->stat - pos;
+	u64 *p;
+
+	if (likely(length <= left)) {
+		memcpy(pos, src, length);
+		pos += length;
+	} else {
+		memcpy(pos, src, left);
+		memcpy(q->desc, src + left, length - left);
+		pos = (void *)q->desc + (length - left);
+	}
+	/* 0-pad to multiple of 16 */
+	p = PTR_ALIGN(pos, 8);
+	if ((uintptr_t)p & 8) {
+		*p = 0;
+		return p + 1;
+	}
+	return p;
+}
+
+/**
+ *      ofld_xmit_direct - copy a WR into offload queue
+ *      @q: the Tx offload queue
+ *      @src: location of WR
+ *      @len: WR length
+ *
+ *      Copy an immediate WR into an uncontended SGE offload queue.
+ */
+static int ofld_xmit_direct(struct sge_uld_txq *q, const void *src,
+			    unsigned int len)
+{
+	unsigned int ndesc;
+	int credits;
+	u64 *pos;
+
+	/* Use the lower limit as the cut-off */
+	if (len > MAX_IMM_OFLD_TX_DATA_WR_LEN) {
+		WARN_ON(1);
+		return NET_XMIT_DROP;
+	}
+
+	/* Don't return NET_XMIT_CN here as the current
+	 * implementation doesn't queue the request
+	 * using an skb when the following conditions not met
+	 */
+	if (!spin_trylock(&q->sendq.lock))
+		return NET_XMIT_DROP;
+
+	if (q->full || !skb_queue_empty(&q->sendq) ||
+	    q->service_ofldq_running) {
+		spin_unlock(&q->sendq.lock);
+		return NET_XMIT_DROP;
+	}
+	ndesc = flits_to_desc(DIV_ROUND_UP(len, 8));
+	credits = txq_avail(&q->q) - ndesc;
+	pos = (u64 *)&q->q.desc[q->q.pidx];
+
+	/* ofldtxq_stop modifies WR header in-situ */
+	inline_tx_header(src, &q->q, pos, len);
+	if (unlikely(credits < TXQ_STOP_THRES))
+		ofldtxq_stop(q, (struct fw_wr_hdr *)pos);
+	txq_advance(&q->q, ndesc);
+	cxgb4_ring_tx_db(q->adap, &q->q, ndesc);
+
+	spin_unlock(&q->sendq.lock);
+	return NET_XMIT_SUCCESS;
+}
+
+int cxgb4_immdata_send(struct net_device *dev, unsigned int idx,
+		       const void *src, unsigned int len)
+{
+	struct sge_uld_txq_info *txq_info;
+	struct sge_uld_txq *txq;
+	struct adapter *adap;
+	int ret;
+
+	adap = netdev2adap(dev);
+
+	local_bh_disable();
+	txq_info = adap->sge.uld_txq_info[CXGB4_TX_OFLD];
+	if (unlikely(!txq_info)) {
+		WARN_ON(true);
+		local_bh_enable();
+		return NET_XMIT_DROP;
+	}
+	txq = &txq_info->uldtxq[idx];
+
+	ret = ofld_xmit_direct(txq, src, len);
+	local_bh_enable();
+	return net_xmit_eval(ret);
+}
+EXPORT_SYMBOL(cxgb4_immdata_send);
+
 /**
  *	t4_crypto_send - send crypto packet
  *	@adap: the adapter
@@ -1981,7 +3428,7 @@ static void cxgb4_sgetim_to_hwtstamp(struct adapter *adap,
 }
 
 static void do_gro(struct sge_eth_rxq *rxq, const struct pkt_gl *gl,
-		   const struct cpl_rx_pkt *pkt)
+		   const struct cpl_rx_pkt *pkt, unsigned long tnl_hdr_len)
 {
 	struct adapter *adapter = rxq->rspq.adap;
 	struct sge *s = &adapter->sge;
@@ -1997,6 +3444,8 @@ static void do_gro(struct sge_eth_rxq *rxq, const struct pkt_gl *gl,
 	}
 
 	copy_frags(skb, gl, s->pktshift);
+	if (tnl_hdr_len)
+		skb->csum_level = 1;
 	skb->len = gl->tot_len - s->pktshift;
 	skb->data_len = skb->len;
 	skb->truesize += skb->data_len;
@@ -2023,6 +3472,185 @@ static void do_gro(struct sge_eth_rxq *rxq, const struct pkt_gl *gl,
 	rxq->stats.rx_cso++;
 }
 
+enum {
+	RX_NON_PTP_PKT = 0,
+	RX_PTP_PKT_SUC = 1,
+	RX_PTP_PKT_ERR = 2
+};
+
+/**
+ *     t4_systim_to_hwstamp - read hardware time stamp
+ *     @adapter: the adapter
+ *     @skb: the packet
+ *
+ *     Read Time Stamp from MPS packet and insert in skb which
+ *     is forwarded to PTP application
+ */
+static noinline int t4_systim_to_hwstamp(struct adapter *adapter,
+					 struct sk_buff *skb)
+{
+	struct skb_shared_hwtstamps *hwtstamps;
+	struct cpl_rx_mps_pkt *cpl = NULL;
+	unsigned char *data;
+	int offset;
+
+	cpl = (struct cpl_rx_mps_pkt *)skb->data;
+	if (!(CPL_RX_MPS_PKT_TYPE_G(ntohl(cpl->op_to_r1_hi)) &
+	     X_CPL_RX_MPS_PKT_TYPE_PTP))
+		return RX_PTP_PKT_ERR;
+
+	data = skb->data + sizeof(*cpl);
+	skb_pull(skb, 2 * sizeof(u64) + sizeof(struct cpl_rx_mps_pkt));
+	offset = ETH_HLEN + IPV4_HLEN(skb->data) + UDP_HLEN;
+	if (skb->len < offset + OFF_PTP_SEQUENCE_ID + sizeof(short))
+		return RX_PTP_PKT_ERR;
+
+	hwtstamps = skb_hwtstamps(skb);
+	memset(hwtstamps, 0, sizeof(*hwtstamps));
+	hwtstamps->hwtstamp = ns_to_ktime(get_unaligned_be64(data));
+
+	return RX_PTP_PKT_SUC;
+}
+
+/**
+ *     t4_rx_hststamp - Recv PTP Event Message
+ *     @adapter: the adapter
+ *     @rsp: the response queue descriptor holding the RX_PKT message
+ *     @rxq: the response queue holding the RX_PKT message
+ *     @skb: the packet
+ *
+ *     PTP enabled and MPS packet, read HW timestamp
+ */
+static int t4_rx_hststamp(struct adapter *adapter, const __be64 *rsp,
+			  struct sge_eth_rxq *rxq, struct sk_buff *skb)
+{
+	int ret;
+
+	if (unlikely((*(u8 *)rsp == CPL_RX_MPS_PKT) &&
+		     !is_t4(adapter->params.chip))) {
+		ret = t4_systim_to_hwstamp(adapter, skb);
+		if (ret == RX_PTP_PKT_ERR) {
+			kfree_skb(skb);
+			rxq->stats.rx_drops++;
+		}
+		return ret;
+	}
+	return RX_NON_PTP_PKT;
+}
+
+/**
+ *      t4_tx_hststamp - Loopback PTP Transmit Event Message
+ *      @adapter: the adapter
+ *      @skb: the packet
+ *      @dev: the ingress net device
+ *
+ *      Read hardware timestamp for the loopback PTP Tx event message
+ */
+static int t4_tx_hststamp(struct adapter *adapter, struct sk_buff *skb,
+			  struct net_device *dev)
+{
+	struct port_info *pi = netdev_priv(dev);
+
+	if (!is_t4(adapter->params.chip) && adapter->ptp_tx_skb) {
+		cxgb4_ptp_read_hwstamp(adapter, pi);
+		kfree_skb(skb);
+		return 0;
+	}
+	return 1;
+}
+
+/**
+ *	t4_tx_completion_handler - handle CPL_SGE_EGR_UPDATE messages
+ *	@rspq: Ethernet RX Response Queue associated with Ethernet TX Queue
+ *	@rsp: Response Entry pointer into Response Queue
+ *	@gl: Gather List pointer
+ *
+ *	For adapters which support the SGE Doorbell Queue Timer facility,
+ *	we configure the Ethernet TX Queues to send CIDX Updates to the
+ *	Associated Ethernet RX Response Queue with CPL_SGE_EGR_UPDATE
+ *	messages.  This adds a small load to PCIe Link RX bandwidth and,
+ *	potentially, higher CPU Interrupt load, but allows us to respond
+ *	much more quickly to the CIDX Updates.  This is important for
+ *	Upper Layer Software which isn't willing to have a large amount
+ *	of TX Data outstanding before receiving DMA Completions.
+ */
+static void t4_tx_completion_handler(struct sge_rspq *rspq,
+				     const __be64 *rsp,
+				     const struct pkt_gl *gl)
+{
+	u8 opcode = ((const struct rss_header *)rsp)->opcode;
+	struct port_info *pi = netdev_priv(rspq->netdev);
+	struct adapter *adapter = rspq->adap;
+	struct sge *s = &adapter->sge;
+	struct sge_eth_txq *txq;
+
+	/* skip RSS header */
+	rsp++;
+
+	/* FW can send EGR_UPDATEs encapsulated in a CPL_FW4_MSG.
+	 */
+	if (unlikely(opcode == CPL_FW4_MSG &&
+		     ((const struct cpl_fw4_msg *)rsp)->type ==
+							FW_TYPE_RSSCPL)) {
+		rsp++;
+		opcode = ((const struct rss_header *)rsp)->opcode;
+		rsp++;
+	}
+
+	if (unlikely(opcode != CPL_SGE_EGR_UPDATE)) {
+		pr_info("%s: unexpected FW4/CPL %#x on Rx queue\n",
+			__func__, opcode);
+		return;
+	}
+
+	txq = &s->ethtxq[pi->first_qset + rspq->idx];
+
+	/* We've got the Hardware Consumer Index Update in the Egress Update
+	 * message. These Egress Update messages will be our sole CIDX Updates
+	 * we get since we don't want to chew up PCIe bandwidth for both Ingress
+	 * Messages and Status Page writes.  However, The code which manages
+	 * reclaiming successfully DMA'ed TX Work Requests uses the CIDX value
+	 * stored in the Status Page at the end of the TX Queue.  It's easiest
+	 * to simply copy the CIDX Update value from the Egress Update message
+	 * to the Status Page.  Also note that no Endian issues need to be
+	 * considered here since both are Big Endian and we're just copying
+	 * bytes consistently ...
+	 */
+	if (CHELSIO_CHIP_VERSION(adapter->params.chip) <= CHELSIO_T5) {
+		struct cpl_sge_egr_update *egr;
+
+		egr = (struct cpl_sge_egr_update *)rsp;
+		WRITE_ONCE(txq->q.stat->cidx, egr->cidx);
+	}
+
+	t4_sge_eth_txq_egress_update(adapter, txq, -1);
+}
+
+static int cxgb4_validate_lb_pkt(struct port_info *pi, const struct pkt_gl *si)
+{
+	struct adapter *adap = pi->adapter;
+	struct cxgb4_ethtool_lb_test *lb;
+	struct sge *s = &adap->sge;
+	struct net_device *netdev;
+	u8 *data;
+	int i;
+
+	netdev = adap->port[pi->port_id];
+	lb = &pi->ethtool_lb;
+	data = si->va + s->pktshift;
+
+	i = ETH_ALEN;
+	if (!ether_addr_equal(data + i, netdev->dev_addr))
+		return -1;
+
+	i += ETH_ALEN;
+	if (strcmp(&data[i], CXGB4_SELFTEST_LB_STR))
+		lb->result = -EIO;
+
+	complete(&lb->completion);
+	return 0;
+}
+
 /**
  *	t4_ethrx_handler - process an ingress ethernet packet
  *	@q: the response queue that received the packet
@@ -2038,27 +3666,52 @@ int t4_ethrx_handler(struct sge_rspq *q, const __be64 *rsp,
 	struct sk_buff *skb;
 	const struct cpl_rx_pkt *pkt;
 	struct sge_eth_rxq *rxq = container_of(q, struct sge_eth_rxq, rspq);
+	struct adapter *adapter = q->adap;
 	struct sge *s = &q->adap->sge;
 	int cpl_trace_pkt = is_t4(q->adap->params.chip) ?
 			    CPL_TRACE_PKT : CPL_TRACE_PKT_T5;
-	u16 err_vec;
+	u16 err_vec, tnl_hdr_len = 0;
 	struct port_info *pi;
+	int ret = 0;
+
+	pi = netdev_priv(q->netdev);
+	/* If we're looking at TX Queue CIDX Update, handle that separately
+	 * and return.
+	 */
+	if (unlikely((*(u8 *)rsp == CPL_FW4_MSG) ||
+		     (*(u8 *)rsp == CPL_SGE_EGR_UPDATE))) {
+		t4_tx_completion_handler(q, rsp, si);
+		return 0;
+	}
 
 	if (unlikely(*(u8 *)rsp == cpl_trace_pkt))
 		return handle_trace_pkt(q->adap, si);
 
 	pkt = (const struct cpl_rx_pkt *)rsp;
 	/* Compressed error vector is enabled for T6 only */
-	if (q->adap->params.tp.rx_pkt_encap)
+	if (q->adap->params.tp.rx_pkt_encap) {
 		err_vec = T6_COMPR_RXERR_VEC_G(be16_to_cpu(pkt->err_vec));
-	else
+		tnl_hdr_len = T6_RX_TNLHDR_LEN_G(ntohs(pkt->err_vec));
+	} else {
 		err_vec = be16_to_cpu(pkt->err_vec);
+	}
 
 	csum_ok = pkt->csum_calc && !err_vec &&
 		  (q->netdev->features & NETIF_F_RXCSUM);
-	if ((pkt->l2info & htonl(RXF_TCP_F)) &&
+
+	if (err_vec)
+		rxq->stats.bad_rx_pkts++;
+
+	if (unlikely(pi->ethtool_lb.loopback && pkt->iff >= NCHAN)) {
+		ret = cxgb4_validate_lb_pkt(pi, si);
+		if (!ret)
+			return 0;
+	}
+
+	if (((pkt->l2info & htonl(RXF_TCP_F)) ||
+	     tnl_hdr_len) &&
 	    (q->netdev->features & NETIF_F_GRO) && csum_ok && !pkt->ip_frag) {
-		do_gro(rxq, si, pkt);
+		do_gro(rxq, si, pkt, tnl_hdr_len);
 		return 0;
 	}
 
@@ -2069,7 +3722,23 @@ int t4_ethrx_handler(struct sge_rspq *q, const __be64 *rsp,
 		return 0;
 	}
 
-	__skb_pull(skb, s->pktshift);      /* remove ethernet header padding */
+	/* Handle PTP Event Rx packet */
+	if (unlikely(pi->ptp_enable)) {
+		ret = t4_rx_hststamp(adapter, rsp, rxq, skb);
+		if (ret == RX_PTP_PKT_ERR)
+			return 0;
+	}
+	if (likely(!ret))
+		__skb_pull(skb, s->pktshift); /* remove ethernet header pad */
+
+	/* Handle the PTP Event Tx Loopback packet */
+	if (unlikely(pi->ptp_enable && !ret &&
+		     (pkt->l2info & htonl(RXF_UDP_F)) &&
+		     cxgb4_ptp_is_ptp_rx(skb))) {
+		if (!t4_tx_hststamp(adapter, skb, q->netdev))
+			return 0;
+	}
+
 	skb->protocol = eth_type_trans(skb, q->netdev);
 	skb_record_rx_queue(skb, q->idx);
 	if (skb->dev->features & NETIF_F_RXHASH)
@@ -2078,7 +3747,6 @@ int t4_ethrx_handler(struct sge_rspq *q, const __be64 *rsp,
 
 	rxq->stats.pkts++;
 
-	pi = netdev_priv(skb->dev);
 	if (pi->rxtstamp)
 		cxgb4_sgetim_to_hwtstamp(q->adap, skb_hwtstamps(skb),
 					 si->sgetstamp);
@@ -2089,7 +3757,13 @@ int t4_ethrx_handler(struct sge_rspq *q, const __be64 *rsp,
 		} else if (pkt->l2info & htonl(RXF_IP_F)) {
 			__sum16 c = (__force __sum16)pkt->csum;
 			skb->csum = csum_unfold(c);
-			skb->ip_summed = CHECKSUM_COMPLETE;
+
+			if (tnl_hdr_len) {
+				skb->ip_summed = CHECKSUM_UNNECESSARY;
+				skb->csum_level = 1;
+			} else {
+				skb->ip_summed = CHECKSUM_COMPLETE;
+			}
 			rxq->stats.rx_cso++;
 		}
 	} else {
@@ -2347,6 +4021,113 @@ static int napi_rx_handler(struct napi_struct *napi, int budget)
 	return work_done;
 }
 
+void cxgb4_ethofld_restart(struct tasklet_struct *t)
+{
+	struct sge_eosw_txq *eosw_txq = from_tasklet(eosw_txq, t,
+						     qresume_tsk);
+	int pktcount;
+
+	spin_lock(&eosw_txq->lock);
+	pktcount = eosw_txq->cidx - eosw_txq->last_cidx;
+	if (pktcount < 0)
+		pktcount += eosw_txq->ndesc;
+
+	if (pktcount) {
+		cxgb4_eosw_txq_free_desc(netdev2adap(eosw_txq->netdev),
+					 eosw_txq, pktcount);
+		eosw_txq->inuse -= pktcount;
+	}
+
+	/* There may be some packets waiting for completions. So,
+	 * attempt to send these packets now.
+	 */
+	ethofld_xmit(eosw_txq->netdev, eosw_txq);
+	spin_unlock(&eosw_txq->lock);
+}
+
+/* cxgb4_ethofld_rx_handler - Process ETHOFLD Tx completions
+ * @q: the response queue that received the packet
+ * @rsp: the response queue descriptor holding the CPL message
+ * @si: the gather list of packet fragments
+ *
+ * Process a ETHOFLD Tx completion. Increment the cidx here, but
+ * free up the descriptors in a tasklet later.
+ */
+int cxgb4_ethofld_rx_handler(struct sge_rspq *q, const __be64 *rsp,
+			     const struct pkt_gl *si)
+{
+	u8 opcode = ((const struct rss_header *)rsp)->opcode;
+
+	/* skip RSS header */
+	rsp++;
+
+	if (opcode == CPL_FW4_ACK) {
+		const struct cpl_fw4_ack *cpl;
+		struct sge_eosw_txq *eosw_txq;
+		struct eotid_entry *entry;
+		struct sk_buff *skb;
+		u32 hdr_len, eotid;
+		u8 flits, wrlen16;
+		int credits;
+
+		cpl = (const struct cpl_fw4_ack *)rsp;
+		eotid = CPL_FW4_ACK_FLOWID_G(ntohl(OPCODE_TID(cpl))) -
+			q->adap->tids.eotid_base;
+		entry = cxgb4_lookup_eotid(&q->adap->tids, eotid);
+		if (!entry)
+			goto out_done;
+
+		eosw_txq = (struct sge_eosw_txq *)entry->data;
+		if (!eosw_txq)
+			goto out_done;
+
+		spin_lock(&eosw_txq->lock);
+		credits = cpl->credits;
+		while (credits > 0) {
+			skb = eosw_txq->desc[eosw_txq->cidx].skb;
+			if (!skb)
+				break;
+
+			if (unlikely((eosw_txq->state ==
+				      CXGB4_EO_STATE_FLOWC_OPEN_REPLY ||
+				      eosw_txq->state ==
+				      CXGB4_EO_STATE_FLOWC_CLOSE_REPLY) &&
+				     eosw_txq->cidx == eosw_txq->flowc_idx)) {
+				flits = DIV_ROUND_UP(skb->len, 8);
+				if (eosw_txq->state ==
+				    CXGB4_EO_STATE_FLOWC_OPEN_REPLY)
+					eosw_txq->state = CXGB4_EO_STATE_ACTIVE;
+				else
+					eosw_txq->state = CXGB4_EO_STATE_CLOSED;
+				complete(&eosw_txq->completion);
+			} else {
+				hdr_len = eth_get_headlen(eosw_txq->netdev,
+							  skb->data,
+							  skb_headlen(skb));
+				flits = ethofld_calc_tx_flits(q->adap, skb,
+							      hdr_len);
+			}
+			eosw_txq_advance_index(&eosw_txq->cidx, 1,
+					       eosw_txq->ndesc);
+			wrlen16 = DIV_ROUND_UP(flits * 8, 16);
+			credits -= wrlen16;
+		}
+
+		eosw_txq->cred += cpl->credits;
+		eosw_txq->ncompl--;
+
+		spin_unlock(&eosw_txq->lock);
+
+		/* Schedule a tasklet to reclaim SKBs and restart ETHOFLD Tx,
+		 * if there were packets waiting for completion.
+		 */
+		tasklet_schedule(&eosw_txq->qresume_tsk);
+	}
+
+out_done:
+	return 0;
+}
+
 /*
  * The MSI-X interrupt handler for an SGE response queue.
  */
@@ -2411,7 +4192,7 @@ static irqreturn_t t4_intr_msi(int irq, void *cookie)
 {
 	struct adapter *adap = cookie;
 
-	if (adap->flags & MASTER_PF)
+	if (adap->flags & CXGB4_MASTER_PF)
 		t4_slow_intr_handler(adap);
 	process_intrq(adap);
 	return IRQ_HANDLED;
@@ -2427,7 +4208,7 @@ static irqreturn_t t4_intr_intx(int irq, void *cookie)
 	struct adapter *adap = cookie;
 
 	t4_write_reg(adap, MYPF_REG(PCIE_PF_CLI_A), 0);
-	if (((adap->flags & MASTER_PF) && t4_slow_intr_handler(adap)) |
+	if (((adap->flags & CXGB4_MASTER_PF) && t4_slow_intr_handler(adap)) |
 	    process_intrq(adap))
 		return IRQ_HANDLED;
 	return IRQ_NONE;             /* probably shared interrupt */
@@ -2442,18 +4223,18 @@ static irqreturn_t t4_intr_intx(int irq, void *cookie)
  */
 irq_handler_t t4_intr_handler(struct adapter *adap)
 {
-	if (adap->flags & USING_MSIX)
+	if (adap->flags & CXGB4_USING_MSIX)
 		return t4_sge_intr_msix;
-	if (adap->flags & USING_MSI)
+	if (adap->flags & CXGB4_USING_MSI)
 		return t4_intr_msi;
 	return t4_intr_intx;
 }
 
-static void sge_rx_timer_cb(unsigned long data)
+static void sge_rx_timer_cb(struct timer_list *t)
 {
 	unsigned long m;
 	unsigned int i;
-	struct adapter *adap = (struct adapter *)data;
+	struct adapter *adap = timer_container_of(adap, t, sge.rx_timer);
 	struct sge *s = &adap->sge;
 
 	for (i = 0; i < BITS_TO_LONGS(s->egr_sz); i++)
@@ -2467,7 +4248,7 @@ static void sge_rx_timer_cb(unsigned long data)
 
 			if (fl_starving(adap, fl)) {
 				rxq = container_of(fl, struct sge_eth_rxq, fl);
-				if (napi_reschedule(&rxq->rspq.napi))
+				if (napi_schedule(&rxq->rspq.napi))
 					fl->starving++;
 				else
 					set_bit(id, s->starving_fl);
@@ -2477,7 +4258,7 @@ static void sge_rx_timer_cb(unsigned long data)
 	 * global Master PF activities like checking for chip ingress stalls,
 	 * etc.
 	 */
-	if (!(adap->flags & MASTER_PF))
+	if (!(adap->flags & CXGB4_MASTER_PF))
 		goto done;
 
 	t4_idma_monitor(adap, &s->idma_monitor, HZ, RX_QCHECK_PERIOD);
@@ -2486,12 +4267,12 @@ done:
 	mod_timer(&s->rx_timer, jiffies + RX_QCHECK_PERIOD);
 }
 
-static void sge_tx_timer_cb(unsigned long data)
+static void sge_tx_timer_cb(struct timer_list *t)
 {
-	unsigned long m;
-	unsigned int i, budget;
-	struct adapter *adap = (struct adapter *)data;
+	struct adapter *adap = timer_container_of(adap, t, sge.tx_timer);
 	struct sge *s = &adap->sge;
+	unsigned long m, period;
+	unsigned int i, budget;
 
 	for (i = 0; i < BITS_TO_LONGS(s->egr_sz); i++)
 		for (m = s->txq_maperr[i]; m; m &= m - 1) {
@@ -2502,32 +4283,46 @@ static void sge_tx_timer_cb(unsigned long data)
 			tasklet_schedule(&txq->qresume_tsk);
 		}
 
-	budget = MAX_TIMER_TX_RECLAIM;
-	i = s->ethtxq_rover;
-	do {
-		struct sge_eth_txq *q = &s->ethtxq[i];
-
-		if (q->q.in_use &&
-		    time_after_eq(jiffies, q->txq->trans_start + HZ / 100) &&
-		    __netif_tx_trylock(q->txq)) {
-			int avail = reclaimable(&q->q);
+	if (!is_t4(adap->params.chip)) {
+		struct sge_eth_txq *q = &s->ptptxq;
+		int avail;
 
-			if (avail) {
-				if (avail > budget)
-					avail = budget;
+		spin_lock(&adap->ptp_lock);
+		avail = reclaimable(&q->q);
 
-				free_tx_desc(adap, &q->q, avail, true);
-				q->q.in_use -= avail;
-				budget -= avail;
-			}
-			__netif_tx_unlock(q->txq);
+		if (avail) {
+			free_tx_desc(adap, &q->q, avail, false);
+			q->q.in_use -= avail;
 		}
+		spin_unlock(&adap->ptp_lock);
+	}
+
+	budget = MAX_TIMER_TX_RECLAIM;
+	i = s->ethtxq_rover;
+	do {
+		budget -= t4_sge_eth_txq_egress_update(adap, &s->ethtxq[i],
+						       budget);
+		if (!budget)
+			break;
 
 		if (++i >= s->ethqsets)
 			i = 0;
-	} while (budget && i != s->ethtxq_rover);
+	} while (i != s->ethtxq_rover);
 	s->ethtxq_rover = i;
-	mod_timer(&s->tx_timer, jiffies + (budget ? TX_QCHECK_PERIOD : 2));
+
+	if (budget == 0) {
+		/* If we found too many reclaimable packets schedule a timer
+		 * in the near future to continue where we left off.
+		 */
+		period = 2;
+	} else {
+		/* We reclaimed all reclaimable TX Descriptors, so reschedule
+		 * at the normal period.
+		 */
+		period = TX_QCHECK_PERIOD;
+	}
+
+	mod_timer(&s->tx_timer, jiffies + period);
 }
 
 /**
@@ -2571,6 +4366,7 @@ int t4_sge_alloc_rxq(struct adapter *adap, struct sge_rspq *iq, bool fwevtq,
 	struct fw_iq_cmd c;
 	struct sge *s = &adap->sge;
 	struct port_info *pi = netdev_priv(dev);
+	int relaxed = !(adap->flags & CXGB4_ROOT_NO_RELAXED_ORDERING);
 
 	/* Size needs to be multiple of 16, including status entry. */
 	iq->size = roundup(iq->size, 16);
@@ -2600,10 +4396,13 @@ int t4_sge_alloc_rxq(struct adapter *adap, struct sge_rspq *iq, bool fwevtq,
 	c.iqsize = htons(iq->size);
 	c.iqaddr = cpu_to_be64(iq->phys_addr);
 	if (cong >= 0)
-		c.iqns_to_fl0congen = htonl(FW_IQ_CMD_IQFLINTCONGEN_F);
+		c.iqns_to_fl0congen = htonl(FW_IQ_CMD_IQFLINTCONGEN_F |
+				FW_IQ_CMD_IQTYPE_V(cong ? FW_IQ_IQTYPE_NIC
+							:  FW_IQ_IQTYPE_OFLD));
 
 	if (fl) {
-		enum chip_type chip = CHELSIO_CHIP_VERSION(adap->params.chip);
+		unsigned int chip_ver =
+			CHELSIO_CHIP_VERSION(adap->params.chip);
 
 		/* Allocate the ring for the hardware free list (with space
 		 * for its status page) along with the associated software
@@ -2624,8 +4423,8 @@ int t4_sge_alloc_rxq(struct adapter *adap, struct sge_rspq *iq, bool fwevtq,
 
 		flsz = fl->size / 8 + s->stat_len / sizeof(struct tx_desc);
 		c.iqns_to_fl0congen |= htonl(FW_IQ_CMD_FL0PACKEN_F |
-					     FW_IQ_CMD_FL0FETCHRO_F |
-					     FW_IQ_CMD_FL0DATARO_F |
+					     FW_IQ_CMD_FL0FETCHRO_V(relaxed) |
+					     FW_IQ_CMD_FL0DATARO_V(relaxed) |
 					     FW_IQ_CMD_FL0PADEN_F);
 		if (cong >= 0)
 			c.iqns_to_fl0congen |=
@@ -2641,10 +4440,10 @@ int t4_sge_alloc_rxq(struct adapter *adap, struct sge_rspq *iq, bool fwevtq,
 		 * the smaller 64-byte value there).
 		 */
 		c.fl0dcaen_to_fl0cidxfthresh =
-			htons(FW_IQ_CMD_FL0FBMIN_V(chip <= CHELSIO_T5 ?
+			htons(FW_IQ_CMD_FL0FBMIN_V(chip_ver <= CHELSIO_T5 ?
 						   FETCHBURSTMIN_128B_X :
-						   FETCHBURSTMIN_64B_X) |
-			      FW_IQ_CMD_FL0FBMAX_V((chip <= CHELSIO_T5) ?
+						   FETCHBURSTMIN_64B_T6_X) |
+			      FW_IQ_CMD_FL0FBMAX_V((chip_ver <= CHELSIO_T5) ?
 						   FETCHBURSTMAX_512B_X :
 						   FETCHBURSTMAX_256B_X));
 		c.fl0size = htons(flsz);
@@ -2655,7 +4454,7 @@ int t4_sge_alloc_rxq(struct adapter *adap, struct sge_rspq *iq, bool fwevtq,
 	if (ret)
 		goto err;
 
-	netif_napi_add(dev, &iq->napi, napi_rx_handler, 64);
+	netif_napi_add(dev, &iq->napi, napi_rx_handler);
 	iq->cur_desc = iq->desc;
 	iq->cidx = 0;
 	iq->gen = 1;
@@ -2766,14 +4565,24 @@ static void init_txq(struct adapter *adap, struct sge_txq *q, unsigned int id)
 	adap->sge.egr_map[id - adap->sge.egr_start] = q;
 }
 
+/**
+ *	t4_sge_alloc_eth_txq - allocate an Ethernet TX Queue
+ *	@adap: the adapter
+ *	@txq: the SGE Ethernet TX Queue to initialize
+ *	@dev: the Linux Network Device
+ *	@netdevq: the corresponding Linux TX Queue
+ *	@iqid: the Ingress Queue to which to deliver CIDX Update messages
+ *	@dbqt: whether this TX Queue will use the SGE Doorbell Queue Timers
+ */
 int t4_sge_alloc_eth_txq(struct adapter *adap, struct sge_eth_txq *txq,
 			 struct net_device *dev, struct netdev_queue *netdevq,
-			 unsigned int iqid)
+			 unsigned int iqid, u8 dbqt)
 {
-	int ret, nentries;
-	struct fw_eq_eth_cmd c;
-	struct sge *s = &adap->sge;
+	unsigned int chip_ver = CHELSIO_CHIP_VERSION(adap->params.chip);
 	struct port_info *pi = netdev_priv(dev);
+	struct sge *s = &adap->sge;
+	struct fw_eq_eth_cmd c;
+	int ret, nentries;
 
 	/* Add status entries */
 	nentries = txq->q.size + s->stat_len / sizeof(struct tx_desc);
@@ -2792,19 +4601,48 @@ int t4_sge_alloc_eth_txq(struct adapter *adap, struct sge_eth_txq *txq,
 			    FW_EQ_ETH_CMD_VFN_V(0));
 	c.alloc_to_len16 = htonl(FW_EQ_ETH_CMD_ALLOC_F |
 				 FW_EQ_ETH_CMD_EQSTART_F | FW_LEN16(c));
-	c.viid_pkd = htonl(FW_EQ_ETH_CMD_AUTOEQUEQE_F |
-			   FW_EQ_ETH_CMD_VIID_V(pi->viid));
+
+	/* For TX Ethernet Queues using the SGE Doorbell Queue Timer
+	 * mechanism, we use Ingress Queue messages for Hardware Consumer
+	 * Index Updates on the TX Queue.  Otherwise we have the Hardware
+	 * write the CIDX Updates into the Status Page at the end of the
+	 * TX Queue.
+	 */
+	c.autoequiqe_to_viid = htonl(((chip_ver <= CHELSIO_T5) ?
+				      FW_EQ_ETH_CMD_AUTOEQUIQE_F :
+				      FW_EQ_ETH_CMD_AUTOEQUEQE_F) |
+				     FW_EQ_ETH_CMD_VIID_V(pi->viid));
+
 	c.fetchszm_to_iqid =
-		htonl(FW_EQ_ETH_CMD_HOSTFCMODE_V(HOSTFCMODE_STATUS_PAGE_X) |
+		htonl(FW_EQ_ETH_CMD_HOSTFCMODE_V((chip_ver <= CHELSIO_T5) ?
+						 HOSTFCMODE_INGRESS_QUEUE_X :
+						 HOSTFCMODE_STATUS_PAGE_X) |
 		      FW_EQ_ETH_CMD_PCIECHN_V(pi->tx_chan) |
 		      FW_EQ_ETH_CMD_FETCHRO_F | FW_EQ_ETH_CMD_IQID_V(iqid));
+
+	/* Note that the CIDX Flush Threshold should match MAX_TX_RECLAIM. */
 	c.dcaen_to_eqsize =
-		htonl(FW_EQ_ETH_CMD_FBMIN_V(FETCHBURSTMIN_64B_X) |
+		htonl(FW_EQ_ETH_CMD_FBMIN_V(chip_ver <= CHELSIO_T5
+					    ? FETCHBURSTMIN_64B_X
+					    : FETCHBURSTMIN_64B_T6_X) |
 		      FW_EQ_ETH_CMD_FBMAX_V(FETCHBURSTMAX_512B_X) |
 		      FW_EQ_ETH_CMD_CIDXFTHRESH_V(CIDXFLUSHTHRESH_32_X) |
+		      FW_EQ_ETH_CMD_CIDXFTHRESHO_V(chip_ver == CHELSIO_T5) |
 		      FW_EQ_ETH_CMD_EQSIZE_V(nentries));
+
 	c.eqaddr = cpu_to_be64(txq->q.phys_addr);
 
+	/* If we're using the SGE Doorbell Queue Timer mechanism, pass in the
+	 * currently configured Timer Index.  THis can be changed later via an
+	 * ethtool -C tx-usecs {Timer Val} command.  Note that the SGE
+	 * Doorbell Queue mode is currently automatically enabled in the
+	 * Firmware by setting either AUTOEQUEQE or AUTOEQUIQE ...
+	 */
+	if (dbqt)
+		c.timeren_timerix =
+			cpu_to_be32(FW_EQ_ETH_CMD_TIMEREN_F |
+				    FW_EQ_ETH_CMD_TIMERIX_V(txq->dbqtimerix));
+
 	ret = t4_wr_mbox(adap, adap->mbox, &c, sizeof(c), &c);
 	if (ret) {
 		kfree(txq->q.sdesc);
@@ -2819,8 +4657,13 @@ int t4_sge_alloc_eth_txq(struct adapter *adap, struct sge_eth_txq *txq,
 	txq->q.q_type = CXGB4_TXQ_ETH;
 	init_txq(adap, &txq->q, FW_EQ_ETH_CMD_EQID_G(ntohl(c.eqid_pkd)));
 	txq->txq = netdevq;
-	txq->tso = txq->tx_cso = txq->vlan_ins = 0;
+	txq->tso = 0;
+	txq->uso = 0;
+	txq->tx_cso = 0;
+	txq->vlan_ins = 0;
 	txq->mapping_err = 0;
+	txq->dbqt = dbqt;
+
 	return 0;
 }
 
@@ -2828,10 +4671,11 @@ int t4_sge_alloc_ctrl_txq(struct adapter *adap, struct sge_ctrl_txq *txq,
 			  struct net_device *dev, unsigned int iqid,
 			  unsigned int cmplqid)
 {
-	int ret, nentries;
-	struct fw_eq_ctrl_cmd c;
-	struct sge *s = &adap->sge;
+	unsigned int chip_ver = CHELSIO_CHIP_VERSION(adap->params.chip);
 	struct port_info *pi = netdev_priv(dev);
+	struct sge *s = &adap->sge;
+	struct fw_eq_ctrl_cmd c;
+	int ret, nentries;
 
 	/* Add status entries */
 	nentries = txq->q.size + s->stat_len / sizeof(struct tx_desc);
@@ -2855,7 +4699,9 @@ int t4_sge_alloc_ctrl_txq(struct adapter *adap, struct sge_ctrl_txq *txq,
 		      FW_EQ_CTRL_CMD_PCIECHN_V(pi->tx_chan) |
 		      FW_EQ_CTRL_CMD_FETCHRO_F | FW_EQ_CTRL_CMD_IQID_V(iqid));
 	c.dcaen_to_eqsize =
-		htonl(FW_EQ_CTRL_CMD_FBMIN_V(FETCHBURSTMIN_64B_X) |
+		htonl(FW_EQ_CTRL_CMD_FBMIN_V(chip_ver <= CHELSIO_T5
+					     ? FETCHBURSTMIN_64B_X
+					     : FETCHBURSTMIN_64B_T6_X) |
 		      FW_EQ_CTRL_CMD_FBMAX_V(FETCHBURSTMAX_512B_X) |
 		      FW_EQ_CTRL_CMD_CIDXFTHRESH_V(CIDXFLUSHTHRESH_32_X) |
 		      FW_EQ_CTRL_CMD_EQSIZE_V(nentries));
@@ -2874,7 +4720,7 @@ int t4_sge_alloc_ctrl_txq(struct adapter *adap, struct sge_ctrl_txq *txq,
 	init_txq(adap, &txq->q, FW_EQ_CTRL_CMD_EQID_G(ntohl(c.cmpliqid_eqid)));
 	txq->adap = adap;
 	skb_queue_head_init(&txq->sendq);
-	tasklet_init(&txq->qresume_tsk, restart_ctrlq, (unsigned long)txq);
+	tasklet_setup(&txq->qresume_tsk, restart_ctrlq);
 	txq->full = 0;
 	return 0;
 }
@@ -2891,29 +4737,30 @@ int t4_sge_mod_ctrl_txq(struct adapter *adap, unsigned int eqid,
 	return t4_set_params(adap, adap->mbox, adap->pf, 0, 1, &param, &val);
 }
 
-int t4_sge_alloc_uld_txq(struct adapter *adap, struct sge_uld_txq *txq,
-			 struct net_device *dev, unsigned int iqid,
-			 unsigned int uld_type)
+static int t4_sge_alloc_ofld_txq(struct adapter *adap, struct sge_txq *q,
+				 struct net_device *dev, u32 cmd, u32 iqid)
 {
-	int ret, nentries;
-	struct fw_eq_ofld_cmd c;
-	struct sge *s = &adap->sge;
+	unsigned int chip_ver = CHELSIO_CHIP_VERSION(adap->params.chip);
 	struct port_info *pi = netdev_priv(dev);
-	int cmd = FW_EQ_OFLD_CMD;
+	struct sge *s = &adap->sge;
+	struct fw_eq_ofld_cmd c;
+	u32 fb_min, nentries;
+	int ret;
 
 	/* Add status entries */
-	nentries = txq->q.size + s->stat_len / sizeof(struct tx_desc);
-
-	txq->q.desc = alloc_ring(adap->pdev_dev, txq->q.size,
-			sizeof(struct tx_desc), sizeof(struct tx_sw_desc),
-			&txq->q.phys_addr, &txq->q.sdesc, s->stat_len,
-			NUMA_NO_NODE);
-	if (!txq->q.desc)
+	nentries = q->size + s->stat_len / sizeof(struct tx_desc);
+	q->desc = alloc_ring(adap->pdev_dev, q->size, sizeof(struct tx_desc),
+			     sizeof(struct tx_sw_desc), &q->phys_addr,
+			     &q->sdesc, s->stat_len, NUMA_NO_NODE);
+	if (!q->desc)
 		return -ENOMEM;
 
+	if (chip_ver <= CHELSIO_T5)
+		fb_min = FETCHBURSTMIN_64B_X;
+	else
+		fb_min = FETCHBURSTMIN_64B_T6_X;
+
 	memset(&c, 0, sizeof(c));
-	if (unlikely(uld_type == CXGB4_TX_CRYPTO))
-		cmd = FW_EQ_CTRL_CMD;
 	c.op_to_vfn = htonl(FW_CMD_OP_V(cmd) | FW_CMD_REQUEST_F |
 			    FW_CMD_WRITE_F | FW_CMD_EXEC_F |
 			    FW_EQ_OFLD_CMD_PFN_V(adap->pf) |
@@ -2925,33 +4772,70 @@ int t4_sge_alloc_uld_txq(struct adapter *adap, struct sge_uld_txq *txq,
 		      FW_EQ_OFLD_CMD_PCIECHN_V(pi->tx_chan) |
 		      FW_EQ_OFLD_CMD_FETCHRO_F | FW_EQ_OFLD_CMD_IQID_V(iqid));
 	c.dcaen_to_eqsize =
-		htonl(FW_EQ_OFLD_CMD_FBMIN_V(FETCHBURSTMIN_64B_X) |
+		htonl(FW_EQ_OFLD_CMD_FBMIN_V(fb_min) |
 		      FW_EQ_OFLD_CMD_FBMAX_V(FETCHBURSTMAX_512B_X) |
 		      FW_EQ_OFLD_CMD_CIDXFTHRESH_V(CIDXFLUSHTHRESH_32_X) |
 		      FW_EQ_OFLD_CMD_EQSIZE_V(nentries));
-	c.eqaddr = cpu_to_be64(txq->q.phys_addr);
+	c.eqaddr = cpu_to_be64(q->phys_addr);
 
 	ret = t4_wr_mbox(adap, adap->mbox, &c, sizeof(c), &c);
 	if (ret) {
-		kfree(txq->q.sdesc);
-		txq->q.sdesc = NULL;
+		kfree(q->sdesc);
+		q->sdesc = NULL;
 		dma_free_coherent(adap->pdev_dev,
 				  nentries * sizeof(struct tx_desc),
-				  txq->q.desc, txq->q.phys_addr);
-		txq->q.desc = NULL;
+				  q->desc, q->phys_addr);
+		q->desc = NULL;
 		return ret;
 	}
 
+	init_txq(adap, q, FW_EQ_OFLD_CMD_EQID_G(ntohl(c.eqid_pkd)));
+	return 0;
+}
+
+int t4_sge_alloc_uld_txq(struct adapter *adap, struct sge_uld_txq *txq,
+			 struct net_device *dev, unsigned int iqid,
+			 unsigned int uld_type)
+{
+	u32 cmd = FW_EQ_OFLD_CMD;
+	int ret;
+
+	if (unlikely(uld_type == CXGB4_TX_CRYPTO))
+		cmd = FW_EQ_CTRL_CMD;
+
+	ret = t4_sge_alloc_ofld_txq(adap, &txq->q, dev, cmd, iqid);
+	if (ret)
+		return ret;
+
 	txq->q.q_type = CXGB4_TXQ_ULD;
-	init_txq(adap, &txq->q, FW_EQ_OFLD_CMD_EQID_G(ntohl(c.eqid_pkd)));
 	txq->adap = adap;
 	skb_queue_head_init(&txq->sendq);
-	tasklet_init(&txq->qresume_tsk, restart_ofldq, (unsigned long)txq);
+	tasklet_setup(&txq->qresume_tsk, restart_ofldq);
 	txq->full = 0;
 	txq->mapping_err = 0;
 	return 0;
 }
 
+int t4_sge_alloc_ethofld_txq(struct adapter *adap, struct sge_eohw_txq *txq,
+			     struct net_device *dev, u32 iqid)
+{
+	int ret;
+
+	ret = t4_sge_alloc_ofld_txq(adap, &txq->q, dev, FW_EQ_OFLD_CMD, iqid);
+	if (ret)
+		return ret;
+
+	txq->q.q_type = CXGB4_TXQ_ULD;
+	spin_lock_init(&txq->lock);
+	txq->adap = adap;
+	txq->tso = 0;
+	txq->uso = 0;
+	txq->tx_cso = 0;
+	txq->vlan_ins = 0;
+	txq->mapping_err = 0;
+	return 0;
+}
+
 void free_txq(struct adapter *adap, struct sge_txq *q)
 {
 	struct sge *s = &adap->sge;
@@ -2991,20 +4875,15 @@ void free_rspq_fl(struct adapter *adap, struct sge_rspq *rq,
 	}
 }
 
-/**
- *      t4_free_ofld_rxqs - free a block of consecutive Rx queues
- *      @adap: the adapter
- *      @n: number of queues
- *      @q: pointer to first queue
- *
- *      Release the resources of a consecutive block of offload Rx queues.
- */
-void t4_free_ofld_rxqs(struct adapter *adap, int n, struct sge_ofld_rxq *q)
+void t4_sge_free_ethofld_txq(struct adapter *adap, struct sge_eohw_txq *txq)
 {
-	for ( ; n; n--, q++)
-		if (q->rspq.desc)
-			free_rspq_fl(adap, &q->rspq,
-				     q->fl.size ? &q->fl : NULL);
+	if (txq->q.desc) {
+		t4_ofld_eq_free(adap, adap->mbox, adap->pf, 0,
+				txq->q.cntxt_id);
+		free_tx_desc(adap, &txq->q, txq->q.in_use, false);
+		kfree(txq->q.sdesc);
+		free_txq(adap, &txq->q);
+	}
 }
 
 /**
@@ -3036,6 +4915,10 @@ void t4_free_sge_resources(struct adapter *adap)
 		if (eq->rspq.desc)
 			free_rspq_fl(adap, &eq->rspq,
 				     eq->fl.size ? &eq->fl : NULL);
+		if (eq->msix) {
+			cxgb4_free_msix_idx_in_bmap(adap, eq->msix->idx);
+			eq->msix = NULL;
+		}
 
 		etq = &adap->sge.ethtxq[i];
 		if (etq->q.desc) {
@@ -3062,12 +4945,32 @@ void t4_free_sge_resources(struct adapter *adap)
 		}
 	}
 
-	if (adap->sge.fw_evtq.desc)
+	if (adap->sge.fw_evtq.desc) {
 		free_rspq_fl(adap, &adap->sge.fw_evtq, NULL);
+		if (adap->sge.fwevtq_msix_idx >= 0)
+			cxgb4_free_msix_idx_in_bmap(adap,
+						    adap->sge.fwevtq_msix_idx);
+	}
+
+	if (adap->sge.nd_msix_idx >= 0)
+		cxgb4_free_msix_idx_in_bmap(adap, adap->sge.nd_msix_idx);
 
 	if (adap->sge.intrq.desc)
 		free_rspq_fl(adap, &adap->sge.intrq, NULL);
 
+	if (!is_t4(adap->params.chip)) {
+		etq = &adap->sge.ptptxq;
+		if (etq->q.desc) {
+			t4_eth_eq_free(adap, adap->mbox, adap->pf, 0,
+				       etq->q.cntxt_id);
+			spin_lock_bh(&adap->ptp_lock);
+			free_tx_desc(adap, &etq->q, etq->q.in_use, true);
+			spin_unlock_bh(&adap->ptp_lock);
+			kfree(etq->q.sdesc);
+			free_txq(adap, &etq->q);
+		}
+	}
+
 	/* clear the reverse egress queue map */
 	memset(adap->sge.egr_map, 0,
 	       adap->sge.egr_sz * sizeof(*adap->sge.egr_map));
@@ -3093,13 +4996,10 @@ void t4_sge_stop(struct adapter *adap)
 	int i;
 	struct sge *s = &adap->sge;
 
-	if (in_interrupt())  /* actions below require waiting */
-		return;
-
 	if (s->rx_timer.function)
-		del_timer_sync(&s->rx_timer);
+		timer_delete_sync(&s->rx_timer);
 	if (s->tx_timer.function)
-		del_timer_sync(&s->tx_timer);
+		timer_delete_sync(&s->tx_timer);
 
 	if (is_offload(adap)) {
 		struct sge_uld_txq_info *txq_info;
@@ -3296,8 +5196,8 @@ int t4_sge_init(struct adapter *adap)
 	/* Set up timers used for recuring callbacks to process RX and TX
 	 * administrative tasks.
 	 */
-	setup_timer(&s->rx_timer, sge_rx_timer_cb, (unsigned long)adap);
-	setup_timer(&s->tx_timer, sge_tx_timer_cb, (unsigned long)adap);
+	timer_setup(&s->rx_timer, sge_rx_timer_cb, 0);
+	timer_setup(&s->tx_timer, sge_tx_timer_cb, 0);
 
 	spin_lock_init(&s->intrq_lock);
 
diff --git a/drivers/net/ethernet/chelsio/cxgb4/smt.c b/drivers/net/ethernet/chelsio/cxgb4/smt.c
new file mode 100644
index 000000000000..e617e4aabbcc
--- /dev/null
+++ b/drivers/net/ethernet/chelsio/cxgb4/smt.c
@@ -0,0 +1,248 @@
+/*
+ * This file is part of the Chelsio T4/T5/T6 Ethernet driver for Linux.
+ *
+ * Copyright (c) 2017 Chelsio Communications, Inc. All rights reserved.
+ *
+ * This software is available to you under a choice of one of two
+ * licenses.  You may choose to be licensed under the terms of the GNU
+ * General Public License (GPL) Version 2, available from the file
+ * COPYING in the main directory of this source tree, or the
+ * OpenIB.org BSD license below:
+ *
+ *     Redistribution and use in source and binary forms, with or
+ *     without modification, are permitted provided that the following
+ *     conditions are met:
+ *
+ *      - Redistributions of source code must retain the above
+ *        copyright notice, this list of conditions and the following
+ *        disclaimer.
+ *
+ *      - Redistributions in binary form must reproduce the above
+ *        copyright notice, this list of conditions and the following
+ *        disclaimer in the documentation and/or other materials
+ *        provided with the distribution.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND,
+ * EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF
+ * MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND
+ * NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS
+ * BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN
+ * ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN
+ * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
+ * SOFTWARE.
+ */
+
+#include "cxgb4.h"
+#include "smt.h"
+#include "t4_msg.h"
+#include "t4fw_api.h"
+#include "t4_regs.h"
+#include "t4_values.h"
+
+struct smt_data *t4_init_smt(void)
+{
+	unsigned int smt_size;
+	struct smt_data *s;
+	int i;
+
+	smt_size = SMT_SIZE;
+
+	s = kvzalloc(struct_size(s, smtab, smt_size), GFP_KERNEL);
+	if (!s)
+		return NULL;
+	s->smt_size = smt_size;
+	rwlock_init(&s->lock);
+	for (i = 0; i < s->smt_size; ++i) {
+		s->smtab[i].idx = i;
+		s->smtab[i].state = SMT_STATE_UNUSED;
+		eth_zero_addr(s->smtab[i].src_mac);
+		spin_lock_init(&s->smtab[i].lock);
+		s->smtab[i].refcnt = 0;
+	}
+	return s;
+}
+
+static struct smt_entry *find_or_alloc_smte(struct smt_data *s, u8 *smac)
+{
+	struct smt_entry *first_free = NULL;
+	struct smt_entry *e, *end;
+
+	for (e = &s->smtab[0], end = &s->smtab[s->smt_size]; e != end; ++e) {
+		if (e->refcnt == 0) {
+			if (!first_free)
+				first_free = e;
+		} else {
+			if (e->state == SMT_STATE_SWITCHING) {
+				/* This entry is actually in use. See if we can
+				 * re-use it ?
+				 */
+				if (memcmp(e->src_mac, smac, ETH_ALEN) == 0)
+					goto found_reuse;
+			}
+		}
+	}
+
+	if (first_free) {
+		e = first_free;
+		goto found;
+	}
+	return NULL;
+
+found:
+	e->state = SMT_STATE_UNUSED;
+
+found_reuse:
+	return e;
+}
+
+static void t4_smte_free(struct smt_entry *e)
+{
+	if (e->refcnt == 0) {  /* hasn't been recycled */
+		e->state = SMT_STATE_UNUSED;
+	}
+}
+
+/**
+ * cxgb4_smt_release - Release SMT entry
+ * @e: smt entry to release
+ *
+ * Releases ref count and frees up an smt entry from SMT table
+ */
+void cxgb4_smt_release(struct smt_entry *e)
+{
+	spin_lock_bh(&e->lock);
+	if ((--e->refcnt) == 0)
+		t4_smte_free(e);
+	spin_unlock_bh(&e->lock);
+}
+EXPORT_SYMBOL(cxgb4_smt_release);
+
+void do_smt_write_rpl(struct adapter *adap, const struct cpl_smt_write_rpl *rpl)
+{
+	unsigned int smtidx = TID_TID_G(GET_TID(rpl));
+	struct smt_data *s = adap->smt;
+
+	if (unlikely(rpl->status != CPL_ERR_NONE)) {
+		struct smt_entry *e = &s->smtab[smtidx];
+
+		dev_err(adap->pdev_dev,
+			"Unexpected SMT_WRITE_RPL status %u for entry %u\n",
+			rpl->status, smtidx);
+		spin_lock(&e->lock);
+		e->state = SMT_STATE_ERROR;
+		spin_unlock(&e->lock);
+		return;
+	}
+}
+
+static int write_smt_entry(struct adapter *adapter, struct smt_entry *e)
+{
+	struct cpl_t6_smt_write_req *t6req;
+	struct smt_data *s = adapter->smt;
+	struct cpl_smt_write_req *req;
+	struct sk_buff *skb;
+	int size;
+	u8 row;
+
+	if (CHELSIO_CHIP_VERSION(adapter->params.chip) <= CHELSIO_T5) {
+		size = sizeof(*req);
+		skb = alloc_skb(size, GFP_ATOMIC);
+		if (!skb)
+			return -ENOMEM;
+		/* Source MAC Table (SMT) contains 256 SMAC entries
+		 * organized in 128 rows of 2 entries each.
+		 */
+		req = (struct cpl_smt_write_req *)__skb_put(skb, size);
+		INIT_TP_WR(req, 0);
+
+		/* Each row contains an SMAC pair.
+		 * LSB selects the SMAC entry within a row
+		 */
+		row = (e->idx >> 1);
+		if (e->idx & 1) {
+			req->pfvf1 = 0x0;
+			memcpy(req->src_mac1, e->src_mac, ETH_ALEN);
+
+			/* fill pfvf0/src_mac0 with entry
+			 * at prev index from smt-tab.
+			 */
+			req->pfvf0 = 0x0;
+			memcpy(req->src_mac0, s->smtab[e->idx - 1].src_mac,
+			       ETH_ALEN);
+		} else {
+			req->pfvf0 = 0x0;
+			memcpy(req->src_mac0, e->src_mac, ETH_ALEN);
+
+			/* fill pfvf1/src_mac1 with entry
+			 * at next index from smt-tab
+			 */
+			req->pfvf1 = 0x0;
+			memcpy(req->src_mac1, s->smtab[e->idx + 1].src_mac,
+			       ETH_ALEN);
+		}
+	} else {
+		size = sizeof(*t6req);
+		skb = alloc_skb(size, GFP_ATOMIC);
+		if (!skb)
+			return -ENOMEM;
+		/* Source MAC Table (SMT) contains 256 SMAC entries */
+		t6req = (struct cpl_t6_smt_write_req *)__skb_put(skb, size);
+		INIT_TP_WR(t6req, 0);
+		req = (struct cpl_smt_write_req *)t6req;
+
+		/* fill pfvf0/src_mac0 from smt-tab */
+		req->pfvf0 = 0x0;
+		memcpy(req->src_mac0, s->smtab[e->idx].src_mac, ETH_ALEN);
+		row = e->idx;
+	}
+
+	OPCODE_TID(req) =
+		htonl(MK_OPCODE_TID(CPL_SMT_WRITE_REQ, e->idx |
+				    TID_QID_V(adapter->sge.fw_evtq.abs_id)));
+	req->params = htonl(SMTW_NORPL_V(0) |
+			    SMTW_IDX_V(row) |
+			    SMTW_OVLAN_IDX_V(0));
+	t4_mgmt_tx(adapter, skb);
+	return 0;
+}
+
+static struct smt_entry *t4_smt_alloc_switching(struct adapter *adap, u16 pfvf,
+						u8 *smac)
+{
+	struct smt_data *s = adap->smt;
+	struct smt_entry *e;
+
+	write_lock_bh(&s->lock);
+	e = find_or_alloc_smte(s, smac);
+	if (e) {
+		spin_lock(&e->lock);
+		if (!e->refcnt) {
+			e->refcnt = 1;
+			e->state = SMT_STATE_SWITCHING;
+			e->pfvf = pfvf;
+			memcpy(e->src_mac, smac, ETH_ALEN);
+			write_smt_entry(adap, e);
+		} else {
+			++e->refcnt;
+		}
+		spin_unlock(&e->lock);
+	}
+	write_unlock_bh(&s->lock);
+	return e;
+}
+
+/**
+ * cxgb4_smt_alloc_switching - Allocates an SMT entry for switch filters.
+ * @dev: net_device pointer
+ * @smac: MAC address to add to SMT
+ * Returns pointer to the SMT entry created
+ *
+ * Allocates an SMT entry to be used by switching rule of a filter.
+ */
+struct smt_entry *cxgb4_smt_alloc_switching(struct net_device *dev, u8 *smac)
+{
+	struct adapter *adap = netdev2adap(dev);
+
+	return t4_smt_alloc_switching(adap, 0x0, smac);
+}
+EXPORT_SYMBOL(cxgb4_smt_alloc_switching);
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/vxlan.h b/drivers/net/ethernet/chelsio/cxgb4/smt.h
index 5def12c048e3..109c1dff563a 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/vxlan.h
+++ b/drivers/net/ethernet/chelsio/cxgb4/smt.h
@@ -1,5 +1,7 @@
 /*
- * Copyright (c) 2016, Mellanox Technologies, Ltd.  All rights reserved.
+ * This file is part of the Chelsio T4/T5/T6 Ethernet driver for Linux.
+ *
+ * Copyright (c) 2017 Chelsio Communications, Inc. All rights reserved.
  *
  * This software is available to you under a choice of one of two
  * licenses.  You may choose to be licensed under the terms of the GNU
@@ -29,34 +31,46 @@
  * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
  * SOFTWARE.
  */
-#ifndef __MLX5_VXLAN_H__
-#define __MLX5_VXLAN_H__
 
-#include <linux/mlx5/driver.h>
-#include "en.h"
+#ifndef __CXGB4_SMT_H
+#define __CXGB4_SMT_H
 
-struct mlx5e_vxlan {
-	u16 udp_port;
-};
+#include <linux/spinlock.h>
+#include <linux/if_ether.h>
+#include <linux/atomic.h>
 
-struct mlx5e_vxlan_work {
-	struct work_struct	work;
-	struct mlx5e_priv	*priv;
-	sa_family_t		sa_family;
-	u16			port;
+struct adapter;
+struct cpl_smt_write_rpl;
+
+/* SMT related handling. Heavily adapted based on l2t ops in l2t.h/l2t.c
+ */
+enum {
+	SMT_STATE_SWITCHING,
+	SMT_STATE_UNUSED,
+	SMT_STATE_ERROR
 };
 
-static inline bool mlx5e_vxlan_allowed(struct mlx5_core_dev *mdev)
-{
-	return (MLX5_CAP_ETH(mdev, tunnel_stateless_vxlan) &&
-		mlx5_core_is_pf(mdev));
-}
+enum {
+	SMT_SIZE = 256
+};
 
-void mlx5e_vxlan_init(struct mlx5e_priv *priv);
-void mlx5e_vxlan_cleanup(struct mlx5e_priv *priv);
+struct smt_entry {
+	u16 state;
+	u16 idx;
+	u16 pfvf;
+	u8 src_mac[ETH_ALEN];
+	int refcnt;
+	spinlock_t lock;	/* protect smt entry add,removal */
+};
 
-void mlx5e_vxlan_queue_work(struct mlx5e_priv *priv, sa_family_t sa_family,
-			    u16 port, int add);
-struct mlx5e_vxlan *mlx5e_vxlan_lookup_port(struct mlx5e_priv *priv, u16 port);
+struct smt_data {
+	unsigned int smt_size;
+	rwlock_t lock;
+	struct smt_entry smtab[] __counted_by(smt_size);
+};
 
-#endif /* __MLX5_VXLAN_H__ */
+struct smt_data *t4_init_smt(void);
+struct smt_entry *cxgb4_smt_alloc_switching(struct net_device *dev, u8 *smac);
+void cxgb4_smt_release(struct smt_entry *e);
+void do_smt_write_rpl(struct adapter *p, const struct cpl_smt_write_rpl *rpl);
+#endif /* __CXGB4_SMT_H */
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/mad.c b/drivers/net/ethernet/chelsio/cxgb4/srq.c
index 3a3b0005fd2b..a77d6ac1ee8c 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/mad.c
+++ b/drivers/net/ethernet/chelsio/cxgb4/srq.c
@@ -1,5 +1,7 @@
 /*
- * Copyright (c) 2013-2015, Mellanox Technologies. All rights reserved.
+ * This file is part of the Chelsio T6 Ethernet driver for Linux.
+ *
+ * Copyright (c) 2017-2018 Chelsio Communications, Inc. All rights reserved.
  *
  * This software is available to you under a choice of one of two
  * licenses.  You may choose to be licensed under the terms of the GNU
@@ -30,46 +32,48 @@
  * SOFTWARE.
  */
 
-#include <linux/kernel.h>
-#include <linux/module.h>
-#include <linux/mlx5/driver.h>
-#include <linux/mlx5/cmd.h>
-#include "mlx5_core.h"
+#include "cxgb4.h"
+#include "t4_msg.h"
+#include "srq.h"
 
-int mlx5_core_mad_ifc(struct mlx5_core_dev *dev, const void *inb, void *outb,
-		      u16 opmod, u8 port)
+struct srq_data *t4_init_srq(int srq_size)
 {
-	int outlen = MLX5_ST_SZ_BYTES(mad_ifc_out);
-	int inlen = MLX5_ST_SZ_BYTES(mad_ifc_in);
-	int err = -ENOMEM;
-	void *data;
-	void *resp;
-	u32 *out;
-	u32 *in;
+	struct srq_data *s;
 
-	in = kzalloc(inlen, GFP_KERNEL);
-	out = kzalloc(outlen, GFP_KERNEL);
-	if (!in || !out)
-		goto out;
+	s = kvzalloc(sizeof(*s), GFP_KERNEL);
+	if (!s)
+		return NULL;
+
+	s->srq_size = srq_size;
+	init_completion(&s->comp);
+	mutex_init(&s->lock);
 
-	MLX5_SET(mad_ifc_in, in, opcode, MLX5_CMD_OP_MAD_IFC);
-	MLX5_SET(mad_ifc_in, in, op_mod, opmod);
-	MLX5_SET(mad_ifc_in, in, port, port);
+	return s;
+}
 
-	data = MLX5_ADDR_OF(mad_ifc_in, in, mad);
-	memcpy(data, inb, MLX5_FLD_SZ_BYTES(mad_ifc_in, mad));
+void do_srq_table_rpl(struct adapter *adap,
+		      const struct cpl_srq_table_rpl *rpl)
+{
+	unsigned int idx = TID_TID_G(GET_TID(rpl));
+	struct srq_data *s = adap->srq;
+	struct srq_entry *e;
 
-	err = mlx5_cmd_exec(dev, in, inlen, out, outlen);
-	if (err)
+	if (unlikely(rpl->status != CPL_CONTAINS_READ_RPL)) {
+		dev_err(adap->pdev_dev,
+			"Unexpected SRQ_TABLE_RPL status %u for entry %u\n",
+				rpl->status, idx);
 		goto out;
+	}
 
-	resp = MLX5_ADDR_OF(mad_ifc_out, out, response_mad_packet);
-	memcpy(outb, resp,
-	       MLX5_FLD_SZ_BYTES(mad_ifc_out, response_mad_packet));
-
+	/* Store the read entry */
+	e = s->entryp;
+	e->valid = 1;
+	e->idx = idx;
+	e->pdid = SRQT_PDID_G(be64_to_cpu(rpl->rsvd_pdid));
+	e->qlen = SRQT_QLEN_G(be32_to_cpu(rpl->qlen_qbase));
+	e->qbase = SRQT_QBASE_G(be32_to_cpu(rpl->qlen_qbase));
+	e->cur_msn = be16_to_cpu(rpl->cur_msn);
+	e->max_msn = be16_to_cpu(rpl->max_msn);
 out:
-	kfree(out);
-	kfree(in);
-	return err;
+	complete(&s->comp);
 }
-EXPORT_SYMBOL_GPL(mlx5_core_mad_ifc);
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/ipoib.h b/drivers/net/ethernet/chelsio/cxgb4/srq.h
index 213191a78464..d9f04bd5ffa3 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/ipoib.h
+++ b/drivers/net/ethernet/chelsio/cxgb4/srq.h
@@ -1,5 +1,7 @@
 /*
- * Copyright (c) 2017, Mellanox Technologies. All rights reserved.
+ * This file is part of the Chelsio T6 Ethernet driver for Linux.
+ *
+ * Copyright (c) 2017-2018 Chelsio Communications, Inc. All rights reserved.
  *
  * This software is available to you under a choice of one of two
  * licenses.  You may choose to be licensed under the terms of the GNU
@@ -30,27 +32,32 @@
  * SOFTWARE.
  */
 
-#ifndef __MLX5E_IPOB_H__
-#define __MLX5E_IPOB_H__
+#ifndef __CXGB4_SRQ_H
+#define __CXGB4_SRQ_H
 
-#include <linux/mlx5/fs.h>
-#include "en.h"
+struct adapter;
+struct cpl_srq_table_rpl;
 
-#define MLX5I_MAX_NUM_TC 1
+#define SRQ_WAIT_TO	(HZ * 5)
 
-/* ipoib rdma netdev's private data structure */
-struct mlx5i_priv {
-	struct rdma_netdev rn; /* keep this first */
-	struct mlx5_core_qp qp;
-	u32    qkey;
-	char  *mlx5e_priv[0];
+struct srq_entry {
+	u8 valid;
+	u8 idx;
+	u8 qlen;
+	u16 pdid;
+	u16 cur_msn;
+	u16 max_msn;
+	u32 qbase;
 };
 
-/* Extract mlx5e_priv from IPoIB netdev */
-#define mlx5i_epriv(netdev) ((void *)(((struct mlx5i_priv *)netdev_priv(netdev))->mlx5e_priv))
-
-netdev_tx_t mlx5i_sq_xmit(struct mlx5e_txqsq *sq, struct sk_buff *skb,
-			  struct mlx5_av *av, u32 dqpn, u32 dqkey);
-void mlx5i_handle_rx_cqe(struct mlx5e_rq *rq, struct mlx5_cqe64 *cqe);
+struct srq_data {
+	unsigned int srq_size;
+	struct srq_entry *entryp;
+	struct completion comp;
+	struct mutex lock; /* generic mutex for srq data */
+};
 
-#endif /* __MLX5E_IPOB_H__ */
+struct srq_data *t4_init_srq(int srq_size);
+void do_srq_table_rpl(struct adapter *adap,
+		      const struct cpl_srq_table_rpl *rpl);
+#endif  /* __CXGB4_SRQ_H */
diff --git a/drivers/net/ethernet/chelsio/cxgb4/t4_chip_type.h b/drivers/net/ethernet/chelsio/cxgb4/t4_chip_type.h
index 54b718111e3f..721c77577ec5 100644
--- a/drivers/net/ethernet/chelsio/cxgb4/t4_chip_type.h
+++ b/drivers/net/ethernet/chelsio/cxgb4/t4_chip_type.h
@@ -34,6 +34,8 @@
 #ifndef __T4_CHIP_TYPE_H__
 #define __T4_CHIP_TYPE_H__
 
+#define CHELSIO_PCI_ID_VER(__DeviceID)  ((__DeviceID) >> 12)
+
 #define CHELSIO_T4		0x4
 #define CHELSIO_T5		0x5
 #define CHELSIO_T6		0x6
diff --git a/drivers/net/ethernet/chelsio/cxgb4/t4_hw.c b/drivers/net/ethernet/chelsio/cxgb4/t4_hw.c
index 9160c882fbfc..171750fad44f 100644
--- a/drivers/net/ethernet/chelsio/cxgb4/t4_hw.c
+++ b/drivers/net/ethernet/chelsio/cxgb4/t4_hw.c
@@ -195,9 +195,11 @@ static void t4_report_fw_error(struct adapter *adap)
 	u32 pcie_fw;
 
 	pcie_fw = t4_read_reg(adap, PCIE_FW_A);
-	if (pcie_fw & PCIE_FW_ERR_F)
+	if (pcie_fw & PCIE_FW_ERR_F) {
 		dev_err(adap->pdev_dev, "Firmware reports adapter error: %s\n",
 			reason[PCIE_FW_EVAL_G(pcie_fw)]);
+		adap->flags &= ~CXGB4_FW_OK;
+	}
 }
 
 /*
@@ -317,9 +319,9 @@ int t4_wr_mbox_meat_timeout(struct adapter *adap, int mbox, const void *cmd,
 	 * wait [for a while] till we're at the front [or bail out with an
 	 * EBUSY] ...
 	 */
-	spin_lock(&adap->mbox_lock);
+	spin_lock_bh(&adap->mbox_lock);
 	list_add_tail(&entry.list, &adap->mlist.list);
-	spin_unlock(&adap->mbox_lock);
+	spin_unlock_bh(&adap->mbox_lock);
 
 	delay_idx = 0;
 	ms = delay[0];
@@ -327,14 +329,14 @@ int t4_wr_mbox_meat_timeout(struct adapter *adap, int mbox, const void *cmd,
 	for (i = 0; ; i += ms) {
 		/* If we've waited too long, return a busy indication.  This
 		 * really ought to be based on our initial position in the
-		 * mailbox access list but this is a start.  We very rearely
+		 * mailbox access list but this is a start.  We very rarely
 		 * contend on access to the mailbox ...
 		 */
 		pcie_fw = t4_read_reg(adap, PCIE_FW_A);
 		if (i > FW_CMD_MAX_TIMEOUT || (pcie_fw & PCIE_FW_ERR_F)) {
-			spin_lock(&adap->mbox_lock);
+			spin_lock_bh(&adap->mbox_lock);
 			list_del(&entry.list);
-			spin_unlock(&adap->mbox_lock);
+			spin_unlock_bh(&adap->mbox_lock);
 			ret = (pcie_fw & PCIE_FW_ERR_F) ? -ENXIO : -EBUSY;
 			t4_record_mbox(adap, cmd, size, access, ret);
 			return ret;
@@ -365,16 +367,16 @@ int t4_wr_mbox_meat_timeout(struct adapter *adap, int mbox, const void *cmd,
 	for (i = 0; v == MBOX_OWNER_NONE && i < 3; i++)
 		v = MBOWNER_G(t4_read_reg(adap, ctl_reg));
 	if (v != MBOX_OWNER_DRV) {
-		spin_lock(&adap->mbox_lock);
+		spin_lock_bh(&adap->mbox_lock);
 		list_del(&entry.list);
-		spin_unlock(&adap->mbox_lock);
+		spin_unlock_bh(&adap->mbox_lock);
 		ret = (v == MBOX_OWNER_FW) ? -EBUSY : -ETIMEDOUT;
-		t4_record_mbox(adap, cmd, MBOX_LEN, access, ret);
+		t4_record_mbox(adap, cmd, size, access, ret);
 		return ret;
 	}
 
 	/* Copy in the new mailbox command and send it on its way ... */
-	t4_record_mbox(adap, cmd, MBOX_LEN, access, 0);
+	t4_record_mbox(adap, cmd, size, access, 0);
 	for (i = 0; i < size; i += 8)
 		t4_write_reg64(adap, data_reg + i, be64_to_cpu(*p++));
 
@@ -418,21 +420,21 @@ int t4_wr_mbox_meat_timeout(struct adapter *adap, int mbox, const void *cmd,
 			execute = i + ms;
 			t4_record_mbox(adap, cmd_rpl,
 				       MBOX_LEN, access, execute);
-			spin_lock(&adap->mbox_lock);
+			spin_lock_bh(&adap->mbox_lock);
 			list_del(&entry.list);
-			spin_unlock(&adap->mbox_lock);
+			spin_unlock_bh(&adap->mbox_lock);
 			return -FW_CMD_RETVAL_G((int)res);
 		}
 	}
 
 	ret = (pcie_fw & PCIE_FW_ERR_F) ? -ENXIO : -ETIMEDOUT;
-	t4_record_mbox(adap, cmd, MBOX_LEN, access, ret);
+	t4_record_mbox(adap, cmd, size, access, ret);
 	dev_err(adap->pdev_dev, "command %#x in mailbox %d timed out\n",
 		*(const u8 *)cmd, mbox);
 	t4_report_fw_error(adap);
-	spin_lock(&adap->mbox_lock);
+	spin_lock_bh(&adap->mbox_lock);
 	list_del(&entry.list);
-	spin_unlock(&adap->mbox_lock);
+	spin_unlock_bh(&adap->mbox_lock);
 	t4_fatal_err(adap);
 	return ret;
 }
@@ -482,6 +484,117 @@ static int t4_edc_err_read(struct adapter *adap, int idx)
 }
 
 /**
+ * t4_memory_rw_init - Get memory window relative offset, base, and size.
+ * @adap: the adapter
+ * @win: PCI-E Memory Window to use
+ * @mtype: memory type: MEM_EDC0, MEM_EDC1, MEM_HMA or MEM_MC
+ * @mem_off: memory relative offset with respect to @mtype.
+ * @mem_base: configured memory base address.
+ * @mem_aperture: configured memory window aperture.
+ *
+ * Get the configured memory window's relative offset, base, and size.
+ */
+int t4_memory_rw_init(struct adapter *adap, int win, int mtype, u32 *mem_off,
+		      u32 *mem_base, u32 *mem_aperture)
+{
+	u32 edc_size, mc_size, mem_reg;
+
+	/* Offset into the region of memory which is being accessed
+	 * MEM_EDC0 = 0
+	 * MEM_EDC1 = 1
+	 * MEM_MC   = 2 -- MEM_MC for chips with only 1 memory controller
+	 * MEM_MC1  = 3 -- for chips with 2 memory controllers (e.g. T5)
+	 * MEM_HMA  = 4
+	 */
+	edc_size  = EDRAM0_SIZE_G(t4_read_reg(adap, MA_EDRAM0_BAR_A));
+	if (mtype == MEM_HMA) {
+		*mem_off = 2 * (edc_size * 1024 * 1024);
+	} else if (mtype != MEM_MC1) {
+		*mem_off = (mtype * (edc_size * 1024 * 1024));
+	} else {
+		mc_size = EXT_MEM0_SIZE_G(t4_read_reg(adap,
+						      MA_EXT_MEMORY0_BAR_A));
+		*mem_off = (MEM_MC0 * edc_size + mc_size) * 1024 * 1024;
+	}
+
+	/* Each PCI-E Memory Window is programmed with a window size -- or
+	 * "aperture" -- which controls the granularity of its mapping onto
+	 * adapter memory.  We need to grab that aperture in order to know
+	 * how to use the specified window.  The window is also programmed
+	 * with the base address of the Memory Window in BAR0's address
+	 * space.  For T4 this is an absolute PCI-E Bus Address.  For T5
+	 * the address is relative to BAR0.
+	 */
+	mem_reg = t4_read_reg(adap,
+			      PCIE_MEM_ACCESS_REG(PCIE_MEM_ACCESS_BASE_WIN_A,
+						  win));
+	/* a dead adapter will return 0xffffffff for PIO reads */
+	if (mem_reg == 0xffffffff)
+		return -ENXIO;
+
+	*mem_aperture = 1 << (WINDOW_G(mem_reg) + WINDOW_SHIFT_X);
+	*mem_base = PCIEOFST_G(mem_reg) << PCIEOFST_SHIFT_X;
+	if (is_t4(adap->params.chip))
+		*mem_base -= adap->t4_bar0;
+
+	return 0;
+}
+
+/**
+ * t4_memory_update_win - Move memory window to specified address.
+ * @adap: the adapter
+ * @win: PCI-E Memory Window to use
+ * @addr: location to move.
+ *
+ * Move memory window to specified address.
+ */
+void t4_memory_update_win(struct adapter *adap, int win, u32 addr)
+{
+	t4_write_reg(adap,
+		     PCIE_MEM_ACCESS_REG(PCIE_MEM_ACCESS_OFFSET_A, win),
+		     addr);
+	/* Read it back to ensure that changes propagate before we
+	 * attempt to use the new value.
+	 */
+	t4_read_reg(adap,
+		    PCIE_MEM_ACCESS_REG(PCIE_MEM_ACCESS_OFFSET_A, win));
+}
+
+/**
+ * t4_memory_rw_residual - Read/Write residual data.
+ * @adap: the adapter
+ * @off: relative offset within residual to start read/write.
+ * @addr: address within indicated memory type.
+ * @buf: host memory buffer
+ * @dir: direction of transfer T4_MEMORY_READ (1) or T4_MEMORY_WRITE (0)
+ *
+ * Read/Write residual data less than 32-bits.
+ */
+void t4_memory_rw_residual(struct adapter *adap, u32 off, u32 addr, u8 *buf,
+			   int dir)
+{
+	union {
+		u32 word;
+		char byte[4];
+	} last;
+	unsigned char *bp;
+	int i;
+
+	if (dir == T4_MEMORY_READ) {
+		last.word = le32_to_cpu((__force __le32)
+					t4_read_reg(adap, addr));
+		for (bp = (unsigned char *)buf, i = off; i < 4; i++)
+			bp[i] = last.byte[i];
+	} else {
+		last.word = *buf;
+		for (i = off; i < 4; i++)
+			last.byte[i] = 0;
+		t4_write_reg(adap, addr,
+			     (__force u32)cpu_to_le32(last.word));
+	}
+}
+
+/**
  *	t4_memory_rw - read/write EDC 0, EDC 1 or MC via PCIE memory window
  *	@adap: the adapter
  *	@win: PCI-E Memory Window to use
@@ -493,7 +606,7 @@ static int t4_edc_err_read(struct adapter *adap, int idx)
  *
  *	Reads/writes an [almost] arbitrary memory region in the firmware: the
  *	firmware memory address and host buffer must be aligned on 32-bit
- *	boudaries; the length may be arbitrary.  The memory is transferred as
+ *	boundaries; the length may be arbitrary.  The memory is transferred as
  *	a raw byte sequence from/to the firmware's memory.  If this memory
  *	contains data structures which contain multi-byte integers, it's the
  *	caller's responsibility to perform appropriate byte order conversions.
@@ -502,8 +615,9 @@ int t4_memory_rw(struct adapter *adap, int win, int mtype, u32 addr,
 		 u32 len, void *hbuf, int dir)
 {
 	u32 pos, offset, resid, memoffset;
-	u32 edc_size, mc_size, win_pf, mem_reg, mem_aperture, mem_base;
+	u32 win_pf, mem_aperture, mem_base;
 	u32 *buf;
+	int ret;
 
 	/* Argument sanity checks ...
 	 */
@@ -519,56 +633,26 @@ int t4_memory_rw(struct adapter *adap, int win, int mtype, u32 addr,
 	resid = len & 0x3;
 	len -= resid;
 
-	/* Offset into the region of memory which is being accessed
-	 * MEM_EDC0 = 0
-	 * MEM_EDC1 = 1
-	 * MEM_MC   = 2 -- MEM_MC for chips with only 1 memory controller
-	 * MEM_MC1  = 3 -- for chips with 2 memory controllers (e.g. T5)
-	 */
-	edc_size  = EDRAM0_SIZE_G(t4_read_reg(adap, MA_EDRAM0_BAR_A));
-	if (mtype != MEM_MC1)
-		memoffset = (mtype * (edc_size * 1024 * 1024));
-	else {
-		mc_size = EXT_MEM0_SIZE_G(t4_read_reg(adap,
-						      MA_EXT_MEMORY0_BAR_A));
-		memoffset = (MEM_MC0 * edc_size + mc_size) * 1024 * 1024;
-	}
+	ret = t4_memory_rw_init(adap, win, mtype, &memoffset, &mem_base,
+				&mem_aperture);
+	if (ret)
+		return ret;
 
 	/* Determine the PCIE_MEM_ACCESS_OFFSET */
 	addr = addr + memoffset;
 
-	/* Each PCI-E Memory Window is programmed with a window size -- or
-	 * "aperture" -- which controls the granularity of its mapping onto
-	 * adapter memory.  We need to grab that aperture in order to know
-	 * how to use the specified window.  The window is also programmed
-	 * with the base address of the Memory Window in BAR0's address
-	 * space.  For T4 this is an absolute PCI-E Bus Address.  For T5
-	 * the address is relative to BAR0.
-	 */
-	mem_reg = t4_read_reg(adap,
-			      PCIE_MEM_ACCESS_REG(PCIE_MEM_ACCESS_BASE_WIN_A,
-						  win));
-	mem_aperture = 1 << (WINDOW_G(mem_reg) + WINDOW_SHIFT_X);
-	mem_base = PCIEOFST_G(mem_reg) << PCIEOFST_SHIFT_X;
-	if (is_t4(adap->params.chip))
-		mem_base -= adap->t4_bar0;
 	win_pf = is_t4(adap->params.chip) ? 0 : PFNUM_V(adap->pf);
 
 	/* Calculate our initial PCI-E Memory Window Position and Offset into
 	 * that Window.
 	 */
-	pos = addr & ~(mem_aperture-1);
+	pos = addr & ~(mem_aperture - 1);
 	offset = addr - pos;
 
 	/* Set up initial PCI-E Memory Window to cover the start of our
-	 * transfer.  (Read it back to ensure that changes propagate before we
-	 * attempt to use the new value.)
+	 * transfer.
 	 */
-	t4_write_reg(adap,
-		     PCIE_MEM_ACCESS_REG(PCIE_MEM_ACCESS_OFFSET_A, win),
-		     pos | win_pf);
-	t4_read_reg(adap,
-		    PCIE_MEM_ACCESS_REG(PCIE_MEM_ACCESS_OFFSET_A, win));
+	t4_memory_update_win(adap, win, pos | win_pf);
 
 	/* Transfer data to/from the adapter as long as there's an integral
 	 * number of 32-bit transfers to complete.
@@ -623,12 +707,7 @@ int t4_memory_rw(struct adapter *adap, int win, int mtype, u32 addr,
 		if (offset == mem_aperture) {
 			pos += mem_aperture;
 			offset = 0;
-			t4_write_reg(adap,
-				PCIE_MEM_ACCESS_REG(PCIE_MEM_ACCESS_OFFSET_A,
-						    win), pos | win_pf);
-			t4_read_reg(adap,
-				PCIE_MEM_ACCESS_REG(PCIE_MEM_ACCESS_OFFSET_A,
-						    win));
+			t4_memory_update_win(adap, win, pos | win_pf);
 		}
 	}
 
@@ -637,28 +716,9 @@ int t4_memory_rw(struct adapter *adap, int win, int mtype, u32 addr,
 	 * residual amount.  The PCI-E Memory Window has already been moved
 	 * above (if necessary) to cover this final transfer.
 	 */
-	if (resid) {
-		union {
-			u32 word;
-			char byte[4];
-		} last;
-		unsigned char *bp;
-		int i;
-
-		if (dir == T4_MEMORY_READ) {
-			last.word = le32_to_cpu(
-					(__force __le32)t4_read_reg(adap,
-						mem_base + offset));
-			for (bp = (unsigned char *)buf, i = resid; i < 4; i++)
-				bp[i] = last.byte[i];
-		} else {
-			last.word = *buf;
-			for (i = resid; i < 4; i++)
-				last.byte[i] = 0;
-			t4_write_reg(adap, mem_base + offset,
-				     (__force u32)cpu_to_le32(last.word));
-		}
-	}
+	if (resid)
+		t4_memory_rw_residual(adap, resid, mem_base + offset,
+				      (u8 *)buf, dir);
 
 	return 0;
 }
@@ -913,7 +973,8 @@ void t4_get_regs(struct adapter *adap, void *buf, size_t buf_size)
 		0xd010, 0xd03c,
 		0xdfc0, 0xdfe0,
 		0xe000, 0xea7c,
-		0xf000, 0x11190,
+		0xf000, 0x11110,
+		0x11118, 0x11190,
 		0x19040, 0x1906c,
 		0x19078, 0x19080,
 		0x1908c, 0x190e4,
@@ -1318,8 +1379,7 @@ void t4_get_regs(struct adapter *adap, void *buf, size_t buf_size)
 		0x9608, 0x9638,
 		0x9640, 0x96f4,
 		0x9800, 0x9808,
-		0x9820, 0x983c,
-		0x9850, 0x9864,
+		0x9810, 0x9864,
 		0x9c00, 0x9c6c,
 		0x9c80, 0x9cec,
 		0x9d00, 0x9d6c,
@@ -1328,7 +1388,7 @@ void t4_get_regs(struct adapter *adap, void *buf, size_t buf_size)
 		0x9e80, 0x9eec,
 		0x9f00, 0x9f6c,
 		0x9f80, 0xa020,
-		0xd004, 0xd004,
+		0xd000, 0xd004,
 		0xd010, 0xd03c,
 		0xdfc0, 0xdfe0,
 		0xe000, 0x1106c,
@@ -1369,10 +1429,8 @@ void t4_get_regs(struct adapter *adap, void *buf, size_t buf_size)
 		0x1a0b0, 0x1a0e4,
 		0x1a0ec, 0x1a0f8,
 		0x1a100, 0x1a108,
-		0x1a114, 0x1a120,
-		0x1a128, 0x1a130,
-		0x1a138, 0x1a138,
-		0x1a190, 0x1a1c4,
+		0x1a114, 0x1a130,
+		0x1a138, 0x1a1c4,
 		0x1a1fc, 0x1a1fc,
 		0x1e008, 0x1e00c,
 		0x1e040, 0x1e044,
@@ -1439,8 +1497,6 @@ void t4_get_regs(struct adapter *adap, void *buf, size_t buf_size)
 		0x1ff00, 0x1ff84,
 		0x1ffc0, 0x1ffc8,
 		0x30000, 0x30030,
-		0x30038, 0x30038,
-		0x30040, 0x30040,
 		0x30100, 0x30144,
 		0x30190, 0x301a0,
 		0x301a8, 0x301b8,
@@ -1551,8 +1607,6 @@ void t4_get_regs(struct adapter *adap, void *buf, size_t buf_size)
 		0x33c3c, 0x33c50,
 		0x33cf0, 0x33cfc,
 		0x34000, 0x34030,
-		0x34038, 0x34038,
-		0x34040, 0x34040,
 		0x34100, 0x34144,
 		0x34190, 0x341a0,
 		0x341a8, 0x341b8,
@@ -1663,8 +1717,6 @@ void t4_get_regs(struct adapter *adap, void *buf, size_t buf_size)
 		0x37c3c, 0x37c50,
 		0x37cf0, 0x37cfc,
 		0x38000, 0x38030,
-		0x38038, 0x38038,
-		0x38040, 0x38040,
 		0x38100, 0x38144,
 		0x38190, 0x381a0,
 		0x381a8, 0x381b8,
@@ -1775,8 +1827,6 @@ void t4_get_regs(struct adapter *adap, void *buf, size_t buf_size)
 		0x3bc3c, 0x3bc50,
 		0x3bcf0, 0x3bcfc,
 		0x3c000, 0x3c030,
-		0x3c038, 0x3c038,
-		0x3c040, 0x3c040,
 		0x3c100, 0x3c144,
 		0x3c190, 0x3c1a0,
 		0x3c1a8, 0x3c1b8,
@@ -2040,12 +2090,9 @@ void t4_get_regs(struct adapter *adap, void *buf, size_t buf_size)
 		0x1190, 0x1194,
 		0x11a0, 0x11a4,
 		0x11b0, 0x11b4,
-		0x11fc, 0x1258,
-		0x1280, 0x12d4,
-		0x12d9, 0x12d9,
-		0x12de, 0x12de,
-		0x12e3, 0x12e3,
-		0x12e8, 0x133c,
+		0x11fc, 0x123c,
+		0x1254, 0x1274,
+		0x1280, 0x133c,
 		0x1800, 0x18fc,
 		0x3000, 0x302c,
 		0x3060, 0x30b0,
@@ -2076,6 +2123,9 @@ void t4_get_regs(struct adapter *adap, void *buf, size_t buf_size)
 		0x5ea0, 0x5eb0,
 		0x5ec0, 0x5ec0,
 		0x5ec8, 0x5ed0,
+		0x5ee0, 0x5ee0,
+		0x5ef0, 0x5ef0,
+		0x5f00, 0x5f00,
 		0x6000, 0x6020,
 		0x6028, 0x6040,
 		0x6058, 0x609c,
@@ -2110,8 +2160,7 @@ void t4_get_regs(struct adapter *adap, void *buf, size_t buf_size)
 		0x9640, 0x9704,
 		0x9710, 0x971c,
 		0x9800, 0x9808,
-		0x9820, 0x983c,
-		0x9850, 0x9864,
+		0x9810, 0x9864,
 		0x9c00, 0x9c6c,
 		0x9c80, 0x9cec,
 		0x9d00, 0x9d6c,
@@ -2120,7 +2169,7 @@ void t4_get_regs(struct adapter *adap, void *buf, size_t buf_size)
 		0x9e80, 0x9eec,
 		0x9f00, 0x9f6c,
 		0x9f80, 0xa020,
-		0xd004, 0xd03c,
+		0xd000, 0xd03c,
 		0xd100, 0xd118,
 		0xd200, 0xd214,
 		0xd220, 0xd234,
@@ -2133,6 +2182,8 @@ void t4_get_regs(struct adapter *adap, void *buf, size_t buf_size)
 		0xd300, 0xd31c,
 		0xdfc0, 0xdfe0,
 		0xe000, 0xf008,
+		0xf010, 0xf018,
+		0xf020, 0xf028,
 		0x11000, 0x11014,
 		0x11048, 0x1106c,
 		0x11074, 0x11088,
@@ -2186,10 +2237,8 @@ void t4_get_regs(struct adapter *adap, void *buf, size_t buf_size)
 		0x1a0b0, 0x1a0e4,
 		0x1a0ec, 0x1a0f8,
 		0x1a100, 0x1a108,
-		0x1a114, 0x1a120,
-		0x1a128, 0x1a130,
-		0x1a138, 0x1a138,
-		0x1a190, 0x1a1c4,
+		0x1a114, 0x1a130,
+		0x1a138, 0x1a1c4,
 		0x1a1fc, 0x1a1fc,
 		0x1e008, 0x1e00c,
 		0x1e040, 0x1e044,
@@ -2256,13 +2305,6 @@ void t4_get_regs(struct adapter *adap, void *buf, size_t buf_size)
 		0x1ff00, 0x1ff84,
 		0x1ffc0, 0x1ffc8,
 		0x30000, 0x30030,
-		0x30038, 0x30038,
-		0x30040, 0x30040,
-		0x30048, 0x30048,
-		0x30050, 0x30050,
-		0x3005c, 0x30060,
-		0x30068, 0x30068,
-		0x30070, 0x30070,
 		0x30100, 0x30168,
 		0x30190, 0x301a0,
 		0x301a8, 0x301b8,
@@ -2325,13 +2367,12 @@ void t4_get_regs(struct adapter *adap, void *buf, size_t buf_size)
 		0x326a8, 0x326a8,
 		0x326ec, 0x326ec,
 		0x32a00, 0x32abc,
-		0x32b00, 0x32b38,
+		0x32b00, 0x32b18,
+		0x32b20, 0x32b38,
 		0x32b40, 0x32b58,
 		0x32b60, 0x32b78,
 		0x32c00, 0x32c00,
 		0x32c08, 0x32c3c,
-		0x32e00, 0x32e2c,
-		0x32f00, 0x32f2c,
 		0x33000, 0x3302c,
 		0x33034, 0x33050,
 		0x33058, 0x33058,
@@ -2396,13 +2437,6 @@ void t4_get_regs(struct adapter *adap, void *buf, size_t buf_size)
 		0x33c38, 0x33c50,
 		0x33cf0, 0x33cfc,
 		0x34000, 0x34030,
-		0x34038, 0x34038,
-		0x34040, 0x34040,
-		0x34048, 0x34048,
-		0x34050, 0x34050,
-		0x3405c, 0x34060,
-		0x34068, 0x34068,
-		0x34070, 0x34070,
 		0x34100, 0x34168,
 		0x34190, 0x341a0,
 		0x341a8, 0x341b8,
@@ -2465,13 +2499,12 @@ void t4_get_regs(struct adapter *adap, void *buf, size_t buf_size)
 		0x366a8, 0x366a8,
 		0x366ec, 0x366ec,
 		0x36a00, 0x36abc,
-		0x36b00, 0x36b38,
+		0x36b00, 0x36b18,
+		0x36b20, 0x36b38,
 		0x36b40, 0x36b58,
 		0x36b60, 0x36b78,
 		0x36c00, 0x36c00,
 		0x36c08, 0x36c3c,
-		0x36e00, 0x36e2c,
-		0x36f00, 0x36f2c,
 		0x37000, 0x3702c,
 		0x37034, 0x37050,
 		0x37058, 0x37058,
@@ -2545,8 +2578,7 @@ void t4_get_regs(struct adapter *adap, void *buf, size_t buf_size)
 		0x40280, 0x40280,
 		0x40304, 0x40304,
 		0x40330, 0x4033c,
-		0x41304, 0x413b8,
-		0x413c0, 0x413c8,
+		0x41304, 0x413c8,
 		0x413d0, 0x413dc,
 		0x413f0, 0x413f0,
 		0x41400, 0x4140c,
@@ -2655,11 +2687,38 @@ void t4_get_regs(struct adapter *adap, void *buf, size_t buf_size)
 }
 
 #define EEPROM_STAT_ADDR   0x7bfc
-#define VPD_SIZE           0x800
 #define VPD_BASE           0x400
 #define VPD_BASE_OLD       0
 #define VPD_LEN            1024
-#define CHELSIO_VPD_UNIQUE_ID 0x82
+
+/**
+ * t4_eeprom_ptov - translate a physical EEPROM address to virtual
+ * @phys_addr: the physical EEPROM address
+ * @fn: the PCI function number
+ * @sz: size of function-specific area
+ *
+ * Translate a physical EEPROM address to virtual.  The first 1K is
+ * accessed through virtual addresses starting at 31K, the rest is
+ * accessed through virtual addresses starting at 0.
+ *
+ * The mapping is as follows:
+ * [0..1K) -> [31K..32K)
+ * [1K..1K+A) -> [31K-A..31K)
+ * [1K+A..ES) -> [0..ES-A-1K)
+ *
+ * where A = @fn * @sz, and ES = EEPROM size.
+ */
+int t4_eeprom_ptov(unsigned int phys_addr, unsigned int fn, unsigned int sz)
+{
+	fn *= sz;
+	if (phys_addr < 1024)
+		return phys_addr + (31 << 10);
+	if (phys_addr < 1024 + fn)
+		return 31744 - fn + phys_addr - 1024;
+	if (phys_addr < EEPROMSIZE)
+		return phys_addr - 1024 - fn;
+	return -EINVAL;
+}
 
 /**
  *	t4_seeprom_wp - enable/disable EEPROM write protection
@@ -2684,111 +2743,73 @@ int t4_seeprom_wp(struct adapter *adapter, bool enable)
  */
 int t4_get_raw_vpd_params(struct adapter *adapter, struct vpd_params *p)
 {
-	int i, ret = 0, addr;
-	int ec, sn, pn, na;
-	u8 *vpd, csum;
-	unsigned int vpdr_len, kw_offset, id_len;
+	unsigned int id_len, pn_len, sn_len, na_len;
+	int id, sn, pn, na, addr, ret = 0;
+	u8 *vpd, base_val = 0;
 
 	vpd = vmalloc(VPD_LEN);
 	if (!vpd)
 		return -ENOMEM;
 
-	/* We have two VPD data structures stored in the adapter VPD area.
-	 * By default, Linux calculates the size of the VPD area by traversing
-	 * the first VPD area at offset 0x0, so we need to tell the OS what
-	 * our real VPD size is.
-	 */
-	ret = pci_set_vpd_size(adapter->pdev, VPD_SIZE);
-	if (ret < 0)
-		goto out;
-
 	/* Card information normally starts at VPD_BASE but early cards had
 	 * it at 0.
 	 */
-	ret = pci_read_vpd(adapter->pdev, VPD_BASE, sizeof(u32), vpd);
+	ret = pci_read_vpd(adapter->pdev, VPD_BASE, 1, &base_val);
 	if (ret < 0)
 		goto out;
 
-	/* The VPD shall have a unique identifier specified by the PCI SIG.
-	 * For chelsio adapters, the identifier is 0x82. The first byte of a VPD
-	 * shall be CHELSIO_VPD_UNIQUE_ID (0x82). The VPD programming software
-	 * is expected to automatically put this entry at the
-	 * beginning of the VPD.
-	 */
-	addr = *vpd == CHELSIO_VPD_UNIQUE_ID ? VPD_BASE : VPD_BASE_OLD;
+	addr = base_val == PCI_VPD_LRDT_ID_STRING ? VPD_BASE : VPD_BASE_OLD;
 
 	ret = pci_read_vpd(adapter->pdev, addr, VPD_LEN, vpd);
 	if (ret < 0)
 		goto out;
 
-	if (vpd[0] != PCI_VPD_LRDT_ID_STRING) {
-		dev_err(adapter->pdev_dev, "missing VPD ID string\n");
-		ret = -EINVAL;
+	ret = pci_vpd_find_id_string(vpd, VPD_LEN, &id_len);
+	if (ret < 0)
 		goto out;
-	}
-
-	id_len = pci_vpd_lrdt_size(vpd);
-	if (id_len > ID_LEN)
-		id_len = ID_LEN;
+	id = ret;
 
-	i = pci_vpd_find_tag(vpd, 0, VPD_LEN, PCI_VPD_LRDT_RO_DATA);
-	if (i < 0) {
-		dev_err(adapter->pdev_dev, "missing VPD-R section\n");
+	ret = pci_vpd_check_csum(vpd, VPD_LEN);
+	if (ret) {
+		dev_err(adapter->pdev_dev, "VPD checksum incorrect or missing\n");
 		ret = -EINVAL;
 		goto out;
 	}
 
-	vpdr_len = pci_vpd_lrdt_size(&vpd[i]);
-	kw_offset = i + PCI_VPD_LRDT_TAG_SIZE;
-	if (vpdr_len + kw_offset > VPD_LEN) {
-		dev_err(adapter->pdev_dev, "bad VPD-R length %u\n", vpdr_len);
-		ret = -EINVAL;
+	ret = pci_vpd_find_ro_info_keyword(vpd, VPD_LEN,
+					   PCI_VPD_RO_KEYWORD_SERIALNO, &sn_len);
+	if (ret < 0)
 		goto out;
-	}
+	sn = ret;
 
-#define FIND_VPD_KW(var, name) do { \
-	var = pci_vpd_find_info_keyword(vpd, kw_offset, vpdr_len, name); \
-	if (var < 0) { \
-		dev_err(adapter->pdev_dev, "missing VPD keyword " name "\n"); \
-		ret = -EINVAL; \
-		goto out; \
-	} \
-	var += PCI_VPD_INFO_FLD_HDR_SIZE; \
-} while (0)
-
-	FIND_VPD_KW(i, "RV");
-	for (csum = 0; i >= 0; i--)
-		csum += vpd[i];
-
-	if (csum) {
-		dev_err(adapter->pdev_dev,
-			"corrupted VPD EEPROM, actual csum %u\n", csum);
-		ret = -EINVAL;
+	ret = pci_vpd_find_ro_info_keyword(vpd, VPD_LEN,
+					   PCI_VPD_RO_KEYWORD_PARTNO, &pn_len);
+	if (ret < 0)
 		goto out;
-	}
+	pn = ret;
 
-	FIND_VPD_KW(ec, "EC");
-	FIND_VPD_KW(sn, "SN");
-	FIND_VPD_KW(pn, "PN");
-	FIND_VPD_KW(na, "NA");
-#undef FIND_VPD_KW
+	ret = pci_vpd_find_ro_info_keyword(vpd, VPD_LEN, "NA", &na_len);
+	if (ret < 0)
+		goto out;
+	na = ret;
 
-	memcpy(p->id, vpd + PCI_VPD_LRDT_TAG_SIZE, id_len);
+	memcpy(p->id, vpd + id, min_t(unsigned int, id_len, ID_LEN));
 	strim(p->id);
-	memcpy(p->ec, vpd + ec, EC_LEN);
-	strim(p->ec);
-	i = pci_vpd_info_field_size(vpd + sn - PCI_VPD_INFO_FLD_HDR_SIZE);
-	memcpy(p->sn, vpd + sn, min(i, SERNUM_LEN));
+	memcpy(p->sn, vpd + sn, min_t(unsigned int, sn_len, SERNUM_LEN));
 	strim(p->sn);
-	i = pci_vpd_info_field_size(vpd + pn - PCI_VPD_INFO_FLD_HDR_SIZE);
-	memcpy(p->pn, vpd + pn, min(i, PN_LEN));
+	memcpy(p->pn, vpd + pn, min_t(unsigned int, pn_len, PN_LEN));
 	strim(p->pn);
-	memcpy(p->na, vpd + na, min(i, MACADDR_LEN));
-	strim((char *)p->na);
+	memcpy(p->na, vpd + na, min_t(unsigned int, na_len, MACADDR_LEN));
+	strim(p->na);
 
 out:
 	vfree(vpd);
-	return ret < 0 ? ret : 0;
+	if (ret < 0) {
+		dev_err(adapter->pdev_dev, "error reading VPD\n");
+		return ret;
+	}
+
+	return 0;
 }
 
 /**
@@ -2826,6 +2847,57 @@ int t4_get_vpd_params(struct adapter *adapter, struct vpd_params *p)
 	return 0;
 }
 
+/**
+ *	t4_get_pfres - retrieve VF resource limits
+ *	@adapter: the adapter
+ *
+ *	Retrieves configured resource limits and capabilities for a physical
+ *	function.  The results are stored in @adapter->pfres.
+ */
+int t4_get_pfres(struct adapter *adapter)
+{
+	struct pf_resources *pfres = &adapter->params.pfres;
+	struct fw_pfvf_cmd cmd, rpl;
+	int v;
+	u32 word;
+
+	/* Execute PFVF Read command to get VF resource limits; bail out early
+	 * with error on command failure.
+	 */
+	memset(&cmd, 0, sizeof(cmd));
+	cmd.op_to_vfn = cpu_to_be32(FW_CMD_OP_V(FW_PFVF_CMD) |
+				    FW_CMD_REQUEST_F |
+				    FW_CMD_READ_F |
+				    FW_PFVF_CMD_PFN_V(adapter->pf) |
+				    FW_PFVF_CMD_VFN_V(0));
+	cmd.retval_len16 = cpu_to_be32(FW_LEN16(cmd));
+	v = t4_wr_mbox(adapter, adapter->mbox, &cmd, sizeof(cmd), &rpl);
+	if (v != FW_SUCCESS)
+		return v;
+
+	/* Extract PF resource limits and return success.
+	 */
+	word = be32_to_cpu(rpl.niqflint_niq);
+	pfres->niqflint = FW_PFVF_CMD_NIQFLINT_G(word);
+	pfres->niq = FW_PFVF_CMD_NIQ_G(word);
+
+	word = be32_to_cpu(rpl.type_to_neq);
+	pfres->neq = FW_PFVF_CMD_NEQ_G(word);
+	pfres->pmask = FW_PFVF_CMD_PMASK_G(word);
+
+	word = be32_to_cpu(rpl.tc_to_nexactf);
+	pfres->tc = FW_PFVF_CMD_TC_G(word);
+	pfres->nvi = FW_PFVF_CMD_NVI_G(word);
+	pfres->nexactf = FW_PFVF_CMD_NEXACTF_G(word);
+
+	word = be32_to_cpu(rpl.r_caps_to_nethctrl);
+	pfres->r_caps = FW_PFVF_CMD_R_CAPS_G(word);
+	pfres->wx_caps = FW_PFVF_CMD_WX_CAPS_G(word);
+	pfres->nethctrl = FW_PFVF_CMD_NETHCTRL_G(word);
+
+	return 0;
+}
+
 /* serial flash and firmware constants */
 enum {
 	SF_ATTEMPTS = 10,             /* max retries for SF operations */
@@ -2838,8 +2910,6 @@ enum {
 	SF_RD_DATA_FAST = 0xb,        /* read flash */
 	SF_RD_ID        = 0x9f,       /* read ID */
 	SF_ERASE_SECTOR = 0xd8,       /* erase sector */
-
-	FW_MAX_SIZE = 16 * SF_SEC_SIZE,
 };
 
 /**
@@ -2967,16 +3037,19 @@ int t4_read_flash(struct adapter *adapter, unsigned int addr,
  *	@addr: the start address to write
  *	@n: length of data to write in bytes
  *	@data: the data to write
+ *	@byte_oriented: whether to store data as bytes or as words
  *
  *	Writes up to a page of data (256 bytes) to the serial flash starting
  *	at the given address.  All the data must be written to the same page.
+ *	If @byte_oriented is set the write data is stored as byte stream
+ *	(i.e. matches what on disk), otherwise in big-endian.
  */
 static int t4_write_flash(struct adapter *adapter, unsigned int addr,
-			  unsigned int n, const u8 *data)
+			  unsigned int n, const u8 *data, bool byte_oriented)
 {
-	int ret;
-	u32 buf[64];
 	unsigned int i, c, left, val, offset = addr & 0xff;
+	u32 buf[64];
+	int ret;
 
 	if (addr >= adapter->params.sf_size || offset + n > SF_PAGE_SIZE)
 		return -EINVAL;
@@ -2987,10 +3060,14 @@ static int t4_write_flash(struct adapter *adapter, unsigned int addr,
 	    (ret = sf1_write(adapter, 4, 1, 1, val)) != 0)
 		goto unlock;
 
-	for (left = n; left; left -= c) {
+	for (left = n; left; left -= c, data += c) {
 		c = min(left, 4U);
-		for (val = 0, i = 0; i < c; ++i)
-			val = (val << 8) + *data++;
+		for (val = 0, i = 0; i < c; ++i) {
+			if (byte_oriented)
+				val = (val << 8) + data[i];
+			else
+				val = (val << 8) + data[c - i - 1];
+		}
 
 		ret = sf1_write(adapter, c, c != left, 1, val);
 		if (ret)
@@ -3003,7 +3080,8 @@ static int t4_write_flash(struct adapter *adapter, unsigned int addr,
 	t4_write_reg(adapter, SF_OP_A, 0);    /* unlock SF */
 
 	/* Read the page to verify the write succeeded */
-	ret = t4_read_flash(adapter, addr & ~0xff, ARRAY_SIZE(buf), buf, 1);
+	ret = t4_read_flash(adapter, addr & ~0xff, ARRAY_SIZE(buf), buf,
+			    byte_oriented);
 	if (ret)
 		return ret;
 
@@ -3064,7 +3142,7 @@ int t4_get_tp_version(struct adapter *adapter, u32 *vers)
 
 /**
  *	t4_get_exprom_version - return the Expansion ROM version (if any)
- *	@adapter: the adapter
+ *	@adap: the adapter
  *	@vers: where to place the version
  *
  *	Reads the Expansion ROM header from FLASH and returns the version
@@ -3100,6 +3178,179 @@ int t4_get_exprom_version(struct adapter *adap, u32 *vers)
 }
 
 /**
+ *      t4_get_vpd_version - return the VPD version
+ *      @adapter: the adapter
+ *      @vers: where to place the version
+ *
+ *      Reads the VPD via the Firmware interface (thus this can only be called
+ *      once we're ready to issue Firmware commands).  The format of the
+ *      VPD version is adapter specific.  Returns 0 on success, an error on
+ *      failure.
+ *
+ *      Note that early versions of the Firmware didn't include the ability
+ *      to retrieve the VPD version, so we zero-out the return-value parameter
+ *      in that case to avoid leaving it with garbage in it.
+ *
+ *      Also note that the Firmware will return its cached copy of the VPD
+ *      Revision ID, not the actual Revision ID as written in the Serial
+ *      EEPROM.  This is only an issue if a new VPD has been written and the
+ *      Firmware/Chip haven't yet gone through a RESET sequence.  So it's best
+ *      to defer calling this routine till after a FW_RESET_CMD has been issued
+ *      if the Host Driver will be performing a full adapter initialization.
+ */
+int t4_get_vpd_version(struct adapter *adapter, u32 *vers)
+{
+	u32 vpdrev_param;
+	int ret;
+
+	vpdrev_param = (FW_PARAMS_MNEM_V(FW_PARAMS_MNEM_DEV) |
+			FW_PARAMS_PARAM_X_V(FW_PARAMS_PARAM_DEV_VPDREV));
+	ret = t4_query_params(adapter, adapter->mbox, adapter->pf, 0,
+			      1, &vpdrev_param, vers);
+	if (ret)
+		*vers = 0;
+	return ret;
+}
+
+/**
+ *      t4_get_scfg_version - return the Serial Configuration version
+ *      @adapter: the adapter
+ *      @vers: where to place the version
+ *
+ *      Reads the Serial Configuration Version via the Firmware interface
+ *      (thus this can only be called once we're ready to issue Firmware
+ *      commands).  The format of the Serial Configuration version is
+ *      adapter specific.  Returns 0 on success, an error on failure.
+ *
+ *      Note that early versions of the Firmware didn't include the ability
+ *      to retrieve the Serial Configuration version, so we zero-out the
+ *      return-value parameter in that case to avoid leaving it with
+ *      garbage in it.
+ *
+ *      Also note that the Firmware will return its cached copy of the Serial
+ *      Initialization Revision ID, not the actual Revision ID as written in
+ *      the Serial EEPROM.  This is only an issue if a new VPD has been written
+ *      and the Firmware/Chip haven't yet gone through a RESET sequence.  So
+ *      it's best to defer calling this routine till after a FW_RESET_CMD has
+ *      been issued if the Host Driver will be performing a full adapter
+ *      initialization.
+ */
+int t4_get_scfg_version(struct adapter *adapter, u32 *vers)
+{
+	u32 scfgrev_param;
+	int ret;
+
+	scfgrev_param = (FW_PARAMS_MNEM_V(FW_PARAMS_MNEM_DEV) |
+			 FW_PARAMS_PARAM_X_V(FW_PARAMS_PARAM_DEV_SCFGREV));
+	ret = t4_query_params(adapter, adapter->mbox, adapter->pf, 0,
+			      1, &scfgrev_param, vers);
+	if (ret)
+		*vers = 0;
+	return ret;
+}
+
+/**
+ *      t4_get_version_info - extract various chip/firmware version information
+ *      @adapter: the adapter
+ *
+ *      Reads various chip/firmware version numbers and stores them into the
+ *      adapter Adapter Parameters structure.  If any of the efforts fails
+ *      the first failure will be returned, but all of the version numbers
+ *      will be read.
+ */
+int t4_get_version_info(struct adapter *adapter)
+{
+	int ret = 0;
+
+	#define FIRST_RET(__getvinfo) \
+	do { \
+		int __ret = __getvinfo; \
+		if (__ret && !ret) \
+			ret = __ret; \
+	} while (0)
+
+	FIRST_RET(t4_get_fw_version(adapter, &adapter->params.fw_vers));
+	FIRST_RET(t4_get_bs_version(adapter, &adapter->params.bs_vers));
+	FIRST_RET(t4_get_tp_version(adapter, &adapter->params.tp_vers));
+	FIRST_RET(t4_get_exprom_version(adapter, &adapter->params.er_vers));
+	FIRST_RET(t4_get_scfg_version(adapter, &adapter->params.scfg_vers));
+	FIRST_RET(t4_get_vpd_version(adapter, &adapter->params.vpd_vers));
+
+	#undef FIRST_RET
+	return ret;
+}
+
+/**
+ *      t4_dump_version_info - dump all of the adapter configuration IDs
+ *      @adapter: the adapter
+ *
+ *      Dumps all of the various bits of adapter configuration version/revision
+ *      IDs information.  This is typically called at some point after
+ *      t4_get_version_info() has been called.
+ */
+void t4_dump_version_info(struct adapter *adapter)
+{
+	/* Device information */
+	dev_info(adapter->pdev_dev, "Chelsio %s rev %d\n",
+		 adapter->params.vpd.id,
+		 CHELSIO_CHIP_RELEASE(adapter->params.chip));
+	dev_info(adapter->pdev_dev, "S/N: %s, P/N: %s\n",
+		 adapter->params.vpd.sn, adapter->params.vpd.pn);
+
+	/* Firmware Version */
+	if (!adapter->params.fw_vers)
+		dev_warn(adapter->pdev_dev, "No firmware loaded\n");
+	else
+		dev_info(adapter->pdev_dev, "Firmware version: %u.%u.%u.%u\n",
+			 FW_HDR_FW_VER_MAJOR_G(adapter->params.fw_vers),
+			 FW_HDR_FW_VER_MINOR_G(adapter->params.fw_vers),
+			 FW_HDR_FW_VER_MICRO_G(adapter->params.fw_vers),
+			 FW_HDR_FW_VER_BUILD_G(adapter->params.fw_vers));
+
+	/* Bootstrap Firmware Version. (Some adapters don't have Bootstrap
+	 * Firmware, so dev_info() is more appropriate here.)
+	 */
+	if (!adapter->params.bs_vers)
+		dev_info(adapter->pdev_dev, "No bootstrap loaded\n");
+	else
+		dev_info(adapter->pdev_dev, "Bootstrap version: %u.%u.%u.%u\n",
+			 FW_HDR_FW_VER_MAJOR_G(adapter->params.bs_vers),
+			 FW_HDR_FW_VER_MINOR_G(adapter->params.bs_vers),
+			 FW_HDR_FW_VER_MICRO_G(adapter->params.bs_vers),
+			 FW_HDR_FW_VER_BUILD_G(adapter->params.bs_vers));
+
+	/* TP Microcode Version */
+	if (!adapter->params.tp_vers)
+		dev_warn(adapter->pdev_dev, "No TP Microcode loaded\n");
+	else
+		dev_info(adapter->pdev_dev,
+			 "TP Microcode version: %u.%u.%u.%u\n",
+			 FW_HDR_FW_VER_MAJOR_G(adapter->params.tp_vers),
+			 FW_HDR_FW_VER_MINOR_G(adapter->params.tp_vers),
+			 FW_HDR_FW_VER_MICRO_G(adapter->params.tp_vers),
+			 FW_HDR_FW_VER_BUILD_G(adapter->params.tp_vers));
+
+	/* Expansion ROM version */
+	if (!adapter->params.er_vers)
+		dev_info(adapter->pdev_dev, "No Expansion ROM loaded\n");
+	else
+		dev_info(adapter->pdev_dev,
+			 "Expansion ROM version: %u.%u.%u.%u\n",
+			 FW_HDR_FW_VER_MAJOR_G(adapter->params.er_vers),
+			 FW_HDR_FW_VER_MINOR_G(adapter->params.er_vers),
+			 FW_HDR_FW_VER_MICRO_G(adapter->params.er_vers),
+			 FW_HDR_FW_VER_BUILD_G(adapter->params.er_vers));
+
+	/* Serial Configuration version */
+	dev_info(adapter->pdev_dev, "Serial Configuration version: %#x\n",
+		 adapter->params.scfg_vers);
+
+	/* VPD Version */
+	dev_info(adapter->pdev_dev, "VPD version: %#x\n",
+		 adapter->params.vpd_vers);
+}
+
+/**
  *	t4_check_fw_version - check if the FW is supported with this driver
  *	@adap: the adapter
  *
@@ -3221,7 +3472,7 @@ int t4_prep_fw(struct adapter *adap, struct fw_info *fw_info,
 	drv_fw = &fw_info->fw_hdr;
 
 	/* Read the header of the firmware on the card */
-	ret = -t4_read_flash(adap, FLASH_FW_START,
+	ret = t4_read_flash(adap, FLASH_FW_START,
 			    sizeof(*card_fw) / sizeof(uint32_t),
 			    (uint32_t *)card_fw, 1);
 	if (ret == 0) {
@@ -3250,8 +3501,8 @@ int t4_prep_fw(struct adapter *adap, struct fw_info *fw_info,
 		   should_install_fs_fw(adap, card_fw_usable,
 					be32_to_cpu(fs_fw->fw_ver),
 					be32_to_cpu(card_fw->fw_ver))) {
-		ret = -t4_fw_upgrade(adap, adap->mbox, fw_data,
-				     fw_size, 0);
+		ret = t4_fw_upgrade(adap, adap->mbox, fw_data,
+				    fw_size, 0);
 		if (ret != 0) {
 			dev_err(adap->pdev_dev,
 				"failed to install firmware: %d\n", ret);
@@ -3282,7 +3533,7 @@ int t4_prep_fw(struct adapter *adap, struct fw_info *fw_info,
 			FW_HDR_FW_VER_MICRO_G(c), FW_HDR_FW_VER_BUILD_G(c),
 			FW_HDR_FW_VER_MAJOR_G(k), FW_HDR_FW_VER_MINOR_G(k),
 			FW_HDR_FW_VER_MICRO_G(k), FW_HDR_FW_VER_BUILD_G(k));
-		ret = EINVAL;
+		ret = -EINVAL;
 		goto bye;
 	}
 
@@ -3379,8 +3630,9 @@ int t4_load_fw(struct adapter *adap, const u8 *fw_data, unsigned int size)
 	const __be32 *p = (const __be32 *)fw_data;
 	const struct fw_hdr *hdr = (const struct fw_hdr *)fw_data;
 	unsigned int sf_sec_size = adap->params.sf_size / adap->params.sf_nsec;
-	unsigned int fw_img_start = adap->params.sf_fw_start;
-	unsigned int fw_start_sec = fw_img_start / sf_sec_size;
+	unsigned int fw_start_sec = FLASH_FW_START_SEC;
+	unsigned int fw_size = FLASH_FW_MAX_SIZE;
+	unsigned int fw_start = FLASH_FW_START;
 
 	if (!size) {
 		dev_err(adap->pdev_dev, "FW image has no data\n");
@@ -3396,9 +3648,9 @@ int t4_load_fw(struct adapter *adap, const u8 *fw_data, unsigned int size)
 			"FW image size differs from size in FW header\n");
 		return -EINVAL;
 	}
-	if (size > FW_MAX_SIZE) {
+	if (size > fw_size) {
 		dev_err(adap->pdev_dev, "FW image too large, max is %u bytes\n",
-			FW_MAX_SIZE);
+			fw_size);
 		return -EFBIG;
 	}
 	if (!t4_fw_matches_chip(adap, hdr))
@@ -3425,22 +3677,22 @@ int t4_load_fw(struct adapter *adap, const u8 *fw_data, unsigned int size)
 	 */
 	memcpy(first_page, fw_data, SF_PAGE_SIZE);
 	((struct fw_hdr *)first_page)->fw_ver = cpu_to_be32(0xffffffff);
-	ret = t4_write_flash(adap, fw_img_start, SF_PAGE_SIZE, first_page);
+	ret = t4_write_flash(adap, fw_start, SF_PAGE_SIZE, first_page, true);
 	if (ret)
 		goto out;
 
-	addr = fw_img_start;
+	addr = fw_start;
 	for (size -= SF_PAGE_SIZE; size; size -= SF_PAGE_SIZE) {
 		addr += SF_PAGE_SIZE;
 		fw_data += SF_PAGE_SIZE;
-		ret = t4_write_flash(adap, addr, SF_PAGE_SIZE, fw_data);
+		ret = t4_write_flash(adap, addr, SF_PAGE_SIZE, fw_data, true);
 		if (ret)
 			goto out;
 	}
 
-	ret = t4_write_flash(adap,
-			     fw_img_start + offsetof(struct fw_hdr, fw_ver),
-			     sizeof(hdr->fw_ver), (const u8 *)&hdr->fw_ver);
+	ret = t4_write_flash(adap, fw_start + offsetof(struct fw_hdr, fw_ver),
+			     sizeof(hdr->fw_ver), (const u8 *)&hdr->fw_ver,
+			     true);
 out:
 	if (ret)
 		dev_err(adap->pdev_dev, "firmware download failed, error %d\n",
@@ -3469,7 +3721,7 @@ int t4_phy_fw_ver(struct adapter *adap, int *phy_fw_ver)
 		 FW_PARAMS_PARAM_Z_V(FW_PARAMS_PARAM_DEV_PHYFW_VERSION));
 	ret = t4_query_params(adap, adap->mbox, adap->pf, 0, 1,
 			      &param, &val);
-	if (ret < 0)
+	if (ret)
 		return ret;
 	*phy_fw_ver = val;
 	return 0;
@@ -3479,7 +3731,6 @@ int t4_phy_fw_ver(struct adapter *adap, int *phy_fw_ver)
  *	t4_load_phy_fw - download port PHY firmware
  *	@adap: the adapter
  *	@win: the PCI-E Memory Window index to use for t4_memory_rw()
- *	@win_lock: the lock to use to guard the memory copy
  *	@phy_fw_version: function to check PHY firmware versions
  *	@phy_fw_data: the PHY firmware image to write
  *	@phy_fw_size: image size
@@ -3488,34 +3739,31 @@ int t4_phy_fw_ver(struct adapter *adap, int *phy_fw_ver)
  *	@phy_fw_version is supplied, then it will be used to determine if
  *	it's necessary to perform the transfer by comparing the version
  *	of any existing adapter PHY firmware with that of the passed in
- *	PHY firmware image.  If @win_lock is non-NULL then it will be used
- *	around the call to t4_memory_rw() which transfers the PHY firmware
- *	to the adapter.
+ *	PHY firmware image.
  *
  *	A negative error number will be returned if an error occurs.  If
  *	version number support is available and there's no need to upgrade
  *	the firmware, 0 will be returned.  If firmware is successfully
- *	transferred to the adapter, 1 will be retured.
+ *	transferred to the adapter, 1 will be returned.
  *
  *	NOTE: some adapters only have local RAM to store the PHY firmware.  As
  *	a result, a RESET of the adapter would cause that RAM to lose its
  *	contents.  Thus, loading PHY firmware on such adapters must happen
  *	after any FW_RESET_CMDs ...
  */
-int t4_load_phy_fw(struct adapter *adap,
-		   int win, spinlock_t *win_lock,
+int t4_load_phy_fw(struct adapter *adap, int win,
 		   int (*phy_fw_version)(const u8 *, size_t),
 		   const u8 *phy_fw_data, size_t phy_fw_size)
 {
+	int cur_phy_fw_ver = 0, new_phy_fw_vers = 0;
 	unsigned long mtype = 0, maddr = 0;
 	u32 param, val;
-	int cur_phy_fw_ver = 0, new_phy_fw_vers = 0;
 	int ret;
 
 	/* If we have version number support, then check to see if the adapter
 	 * already has up-to-date PHY firmware loaded.
 	 */
-	 if (phy_fw_version) {
+	if (phy_fw_version) {
 		new_phy_fw_vers = phy_fw_version(phy_fw_data, phy_fw_size);
 		ret = t4_phy_fw_ver(adap, &cur_phy_fw_ver);
 		if (ret < 0)
@@ -3529,7 +3777,7 @@ int t4_load_phy_fw(struct adapter *adap,
 	}
 
 	/* Ask the firmware where it wants us to copy the PHY firmware image.
-	 * The size of the file requires a special version of the READ coommand
+	 * The size of the file requires a special version of the READ command
 	 * which will pass the file size via the values field in PARAMS_CMD and
 	 * retrieve the return value from firmware and place it in the same
 	 * buffer values
@@ -3540,7 +3788,7 @@ int t4_load_phy_fw(struct adapter *adap,
 		 FW_PARAMS_PARAM_Z_V(FW_PARAMS_PARAM_DEV_PHYFW_DOWNLOAD));
 	val = phy_fw_size;
 	ret = t4_query_params_rw(adap, adap->mbox, adap->pf, 0, 1,
-				 &param, &val, 1);
+				 &param, &val, 1, true);
 	if (ret < 0)
 		return ret;
 	mtype = val >> 8;
@@ -3549,13 +3797,11 @@ int t4_load_phy_fw(struct adapter *adap,
 	/* Copy the supplied PHY Firmware image to the adapter memory location
 	 * allocated by the adapter firmware.
 	 */
-	if (win_lock)
-		spin_lock_bh(win_lock);
+	spin_lock_bh(&adap->win0_lock);
 	ret = t4_memory_rw(adap, win, mtype, maddr,
 			   phy_fw_size, (__be32 *)phy_fw_data,
 			   T4_MEMORY_WRITE);
-	if (win_lock)
-		spin_unlock_bh(win_lock);
+	spin_unlock_bh(&adap->win0_lock);
 	if (ret)
 		return ret;
 
@@ -3570,6 +3816,8 @@ int t4_load_phy_fw(struct adapter *adap,
 		 FW_PARAMS_PARAM_Z_V(FW_PARAMS_PARAM_DEV_PHYFW_DOWNLOAD));
 	ret = t4_set_params_timeout(adap, adap->mbox, adap->pf, 0, 1,
 				    &param, &val, 30000);
+	if (ret)
+		return ret;
 
 	/* If we have version number support, then check to see that the new
 	 * firmware got loaded properly.
@@ -3610,7 +3858,7 @@ int t4_fwcache(struct adapter *adap, enum fw_params_param_dev_fwcache op)
 	c.param[0].mnem =
 		cpu_to_be32(FW_PARAMS_MNEM_V(FW_PARAMS_MNEM_DEV) |
 			    FW_PARAMS_PARAM_X_V(FW_PARAMS_PARAM_DEV_FWCACHE));
-	c.param[0].val = (__force __be32)op;
+	c.param[0].val = cpu_to_be32(op);
 
 	return t4_wr_mbox(adap, adap->mbox, &c, sizeof(c), NULL);
 }
@@ -3685,16 +3933,246 @@ void t4_ulprx_read_la(struct adapter *adap, u32 *la_buf)
 	}
 }
 
-#define ADVERT_MASK (FW_PORT_CAP_SPEED_100M | FW_PORT_CAP_SPEED_1G |\
-		     FW_PORT_CAP_SPEED_10G | FW_PORT_CAP_SPEED_25G | \
-		     FW_PORT_CAP_SPEED_40G | FW_PORT_CAP_SPEED_100G | \
-		     FW_PORT_CAP_ANEG)
+/* The ADVERT_MASK is used to mask out all of the Advertised Firmware Port
+ * Capabilities which we control with separate controls -- see, for instance,
+ * Pause Frames and Forward Error Correction.  In order to determine what the
+ * full set of Advertised Port Capabilities are, the base Advertised Port
+ * Capabilities (masked by ADVERT_MASK) must be combined with the Advertised
+ * Port Capabilities associated with those other controls.  See
+ * t4_link_acaps() for how this is done.
+ */
+#define ADVERT_MASK (FW_PORT_CAP32_SPEED_V(FW_PORT_CAP32_SPEED_M) | \
+		     FW_PORT_CAP32_ANEG)
+
+/**
+ *	fwcaps16_to_caps32 - convert 16-bit Port Capabilities to 32-bits
+ *	@caps16: a 16-bit Port Capabilities value
+ *
+ *	Returns the equivalent 32-bit Port Capabilities value.
+ */
+static fw_port_cap32_t fwcaps16_to_caps32(fw_port_cap16_t caps16)
+{
+	fw_port_cap32_t caps32 = 0;
+
+	#define CAP16_TO_CAP32(__cap) \
+		do { \
+			if (caps16 & FW_PORT_CAP_##__cap) \
+				caps32 |= FW_PORT_CAP32_##__cap; \
+		} while (0)
+
+	CAP16_TO_CAP32(SPEED_100M);
+	CAP16_TO_CAP32(SPEED_1G);
+	CAP16_TO_CAP32(SPEED_25G);
+	CAP16_TO_CAP32(SPEED_10G);
+	CAP16_TO_CAP32(SPEED_40G);
+	CAP16_TO_CAP32(SPEED_100G);
+	CAP16_TO_CAP32(FC_RX);
+	CAP16_TO_CAP32(FC_TX);
+	CAP16_TO_CAP32(ANEG);
+	CAP16_TO_CAP32(FORCE_PAUSE);
+	CAP16_TO_CAP32(MDIAUTO);
+	CAP16_TO_CAP32(MDISTRAIGHT);
+	CAP16_TO_CAP32(FEC_RS);
+	CAP16_TO_CAP32(FEC_BASER_RS);
+	CAP16_TO_CAP32(802_3_PAUSE);
+	CAP16_TO_CAP32(802_3_ASM_DIR);
+
+	#undef CAP16_TO_CAP32
+
+	return caps32;
+}
 
 /**
- *	t4_link_l1cfg - apply link configuration to MAC/PHY
- *	@phy: the PHY to setup
- *	@mac: the MAC to setup
- *	@lc: the requested link configuration
+ *	fwcaps32_to_caps16 - convert 32-bit Port Capabilities to 16-bits
+ *	@caps32: a 32-bit Port Capabilities value
+ *
+ *	Returns the equivalent 16-bit Port Capabilities value.  Note that
+ *	not all 32-bit Port Capabilities can be represented in the 16-bit
+ *	Port Capabilities and some fields/values may not make it.
+ */
+static fw_port_cap16_t fwcaps32_to_caps16(fw_port_cap32_t caps32)
+{
+	fw_port_cap16_t caps16 = 0;
+
+	#define CAP32_TO_CAP16(__cap) \
+		do { \
+			if (caps32 & FW_PORT_CAP32_##__cap) \
+				caps16 |= FW_PORT_CAP_##__cap; \
+		} while (0)
+
+	CAP32_TO_CAP16(SPEED_100M);
+	CAP32_TO_CAP16(SPEED_1G);
+	CAP32_TO_CAP16(SPEED_10G);
+	CAP32_TO_CAP16(SPEED_25G);
+	CAP32_TO_CAP16(SPEED_40G);
+	CAP32_TO_CAP16(SPEED_100G);
+	CAP32_TO_CAP16(FC_RX);
+	CAP32_TO_CAP16(FC_TX);
+	CAP32_TO_CAP16(802_3_PAUSE);
+	CAP32_TO_CAP16(802_3_ASM_DIR);
+	CAP32_TO_CAP16(ANEG);
+	CAP32_TO_CAP16(FORCE_PAUSE);
+	CAP32_TO_CAP16(MDIAUTO);
+	CAP32_TO_CAP16(MDISTRAIGHT);
+	CAP32_TO_CAP16(FEC_RS);
+	CAP32_TO_CAP16(FEC_BASER_RS);
+
+	#undef CAP32_TO_CAP16
+
+	return caps16;
+}
+
+/* Translate Firmware Port Capabilities Pause specification to Common Code */
+static inline enum cc_pause fwcap_to_cc_pause(fw_port_cap32_t fw_pause)
+{
+	enum cc_pause cc_pause = 0;
+
+	if (fw_pause & FW_PORT_CAP32_FC_RX)
+		cc_pause |= PAUSE_RX;
+	if (fw_pause & FW_PORT_CAP32_FC_TX)
+		cc_pause |= PAUSE_TX;
+
+	return cc_pause;
+}
+
+/* Translate Common Code Pause specification into Firmware Port Capabilities */
+static inline fw_port_cap32_t cc_to_fwcap_pause(enum cc_pause cc_pause)
+{
+	/* Translate orthogonal RX/TX Pause Controls for L1 Configure
+	 * commands, etc.
+	 */
+	fw_port_cap32_t fw_pause = 0;
+
+	if (cc_pause & PAUSE_RX)
+		fw_pause |= FW_PORT_CAP32_FC_RX;
+	if (cc_pause & PAUSE_TX)
+		fw_pause |= FW_PORT_CAP32_FC_TX;
+	if (!(cc_pause & PAUSE_AUTONEG))
+		fw_pause |= FW_PORT_CAP32_FORCE_PAUSE;
+
+	/* Translate orthogonal Pause controls into IEEE 802.3 Pause,
+	 * Asymmetrical Pause for use in reporting to upper layer OS code, etc.
+	 * Note that these bits are ignored in L1 Configure commands.
+	 */
+	if (cc_pause & PAUSE_RX) {
+		if (cc_pause & PAUSE_TX)
+			fw_pause |= FW_PORT_CAP32_802_3_PAUSE;
+		else
+			fw_pause |= FW_PORT_CAP32_802_3_ASM_DIR |
+				    FW_PORT_CAP32_802_3_PAUSE;
+	} else if (cc_pause & PAUSE_TX) {
+		fw_pause |= FW_PORT_CAP32_802_3_ASM_DIR;
+	}
+
+	return fw_pause;
+}
+
+/* Translate Firmware Forward Error Correction specification to Common Code */
+static inline enum cc_fec fwcap_to_cc_fec(fw_port_cap32_t fw_fec)
+{
+	enum cc_fec cc_fec = 0;
+
+	if (fw_fec & FW_PORT_CAP32_FEC_RS)
+		cc_fec |= FEC_RS;
+	if (fw_fec & FW_PORT_CAP32_FEC_BASER_RS)
+		cc_fec |= FEC_BASER_RS;
+
+	return cc_fec;
+}
+
+/* Translate Common Code Forward Error Correction specification to Firmware */
+static inline fw_port_cap32_t cc_to_fwcap_fec(enum cc_fec cc_fec)
+{
+	fw_port_cap32_t fw_fec = 0;
+
+	if (cc_fec & FEC_RS)
+		fw_fec |= FW_PORT_CAP32_FEC_RS;
+	if (cc_fec & FEC_BASER_RS)
+		fw_fec |= FW_PORT_CAP32_FEC_BASER_RS;
+
+	return fw_fec;
+}
+
+/**
+ *	t4_link_acaps - compute Link Advertised Port Capabilities
+ *	@adapter: the adapter
+ *	@port: the Port ID
+ *	@lc: the Port's Link Configuration
+ *
+ *	Synthesize the Advertised Port Capabilities we'll be using based on
+ *	the base Advertised Port Capabilities (which have been filtered by
+ *	ADVERT_MASK) plus the individual controls for things like Pause
+ *	Frames, Forward Error Correction, MDI, etc.
+ */
+fw_port_cap32_t t4_link_acaps(struct adapter *adapter, unsigned int port,
+			      struct link_config *lc)
+{
+	fw_port_cap32_t fw_fc, fw_fec, acaps;
+	unsigned int fw_mdi;
+	char cc_fec;
+
+	fw_mdi = (FW_PORT_CAP32_MDI_V(FW_PORT_CAP32_MDI_AUTO) & lc->pcaps);
+
+	/* Convert driver coding of Pause Frame Flow Control settings into the
+	 * Firmware's API.
+	 */
+	fw_fc = cc_to_fwcap_pause(lc->requested_fc);
+
+	/* Convert Common Code Forward Error Control settings into the
+	 * Firmware's API.  If the current Requested FEC has "Automatic"
+	 * (IEEE 802.3) specified, then we use whatever the Firmware
+	 * sent us as part of its IEEE 802.3-based interpretation of
+	 * the Transceiver Module EPROM FEC parameters.  Otherwise we
+	 * use whatever is in the current Requested FEC settings.
+	 */
+	if (lc->requested_fec & FEC_AUTO)
+		cc_fec = fwcap_to_cc_fec(lc->def_acaps);
+	else
+		cc_fec = lc->requested_fec;
+	fw_fec = cc_to_fwcap_fec(cc_fec);
+
+	/* Figure out what our Requested Port Capabilities are going to be.
+	 * Note parallel structure in t4_handle_get_port_info() and
+	 * init_link_config().
+	 */
+	if (!(lc->pcaps & FW_PORT_CAP32_ANEG)) {
+		acaps = lc->acaps | fw_fc | fw_fec;
+		lc->fc = lc->requested_fc & ~PAUSE_AUTONEG;
+		lc->fec = cc_fec;
+	} else if (lc->autoneg == AUTONEG_DISABLE) {
+		acaps = lc->speed_caps | fw_fc | fw_fec | fw_mdi;
+		lc->fc = lc->requested_fc & ~PAUSE_AUTONEG;
+		lc->fec = cc_fec;
+	} else {
+		acaps = lc->acaps | fw_fc | fw_fec | fw_mdi;
+	}
+
+	/* Some Requested Port Capabilities are trivially wrong if they exceed
+	 * the Physical Port Capabilities.  We can check that here and provide
+	 * moderately useful feedback in the system log.
+	 *
+	 * Note that older Firmware doesn't have FW_PORT_CAP32_FORCE_PAUSE, so
+	 * we need to exclude this from this check in order to maintain
+	 * compatibility ...
+	 */
+	if ((acaps & ~lc->pcaps) & ~FW_PORT_CAP32_FORCE_PAUSE) {
+		dev_err(adapter->pdev_dev, "Requested Port Capabilities %#x exceed Physical Port Capabilities %#x\n",
+			acaps, lc->pcaps);
+		return -EINVAL;
+	}
+
+	return acaps;
+}
+
+/**
+ *	t4_link_l1cfg_core - apply link configuration to MAC/PHY
+ *	@adapter: the adapter
+ *	@mbox: the Firmware Mailbox to use
+ *	@port: the Port ID
+ *	@lc: the Port's Link Configuration
+ *	@sleep_ok: if true we may sleep while awaiting command completion
+ *	@timeout: time to wait for command to finish before timing out
+ *		(negative implies @sleep_ok=false)
  *
  *	Set up a port's MAC and PHY according to a desired link configuration.
  *	- If the PHY can auto-negotiate first decide what to advertise, then
@@ -3703,47 +4181,54 @@ void t4_ulprx_read_la(struct adapter *adap, u32 *la_buf)
  *	- If auto-negotiation is off set the MAC to the proper speed/duplex/FC,
  *	  otherwise do it later based on the outcome of auto-negotiation.
  */
-int t4_link_l1cfg(struct adapter *adap, unsigned int mbox, unsigned int port,
-		  struct link_config *lc)
+int t4_link_l1cfg_core(struct adapter *adapter, unsigned int mbox,
+		       unsigned int port, struct link_config *lc,
+		       u8 sleep_ok, int timeout)
 {
-	struct fw_port_cmd c;
-	unsigned int mdi = FW_PORT_CAP_MDI_V(FW_PORT_CAP_MDI_AUTO);
-	unsigned int fc = 0, fec = 0, fw_fec = 0;
-
-	lc->link_ok = 0;
-	if (lc->requested_fc & PAUSE_RX)
-		fc |= FW_PORT_CAP_FC_RX;
-	if (lc->requested_fc & PAUSE_TX)
-		fc |= FW_PORT_CAP_FC_TX;
-
-	fec = lc->requested_fec & FEC_AUTO ? lc->auto_fec : lc->requested_fec;
+	unsigned int fw_caps = adapter->params.fw_caps_support;
+	struct fw_port_cmd cmd;
+	fw_port_cap32_t rcap;
+	int ret;
 
-	if (fec & FEC_RS)
-		fw_fec |= FW_PORT_CAP_FEC_RS;
-	if (fec & FEC_BASER_RS)
-		fw_fec |= FW_PORT_CAP_FEC_BASER_RS;
+	if (!(lc->pcaps & FW_PORT_CAP32_ANEG) &&
+	    lc->autoneg == AUTONEG_ENABLE) {
+		return -EINVAL;
+	}
 
-	memset(&c, 0, sizeof(c));
-	c.op_to_portid = cpu_to_be32(FW_CMD_OP_V(FW_PORT_CMD) |
-				     FW_CMD_REQUEST_F | FW_CMD_EXEC_F |
-				     FW_PORT_CMD_PORTID_V(port));
-	c.action_to_len16 =
-		cpu_to_be32(FW_PORT_CMD_ACTION_V(FW_PORT_ACTION_L1_CFG) |
-			    FW_LEN16(c));
+	/* Compute our Requested Port Capabilities and send that on to the
+	 * Firmware.
+	 */
+	rcap = t4_link_acaps(adapter, port, lc);
+	memset(&cmd, 0, sizeof(cmd));
+	cmd.op_to_portid = cpu_to_be32(FW_CMD_OP_V(FW_PORT_CMD) |
+				       FW_CMD_REQUEST_F | FW_CMD_EXEC_F |
+				       FW_PORT_CMD_PORTID_V(port));
+	cmd.action_to_len16 =
+		cpu_to_be32(FW_PORT_CMD_ACTION_V(fw_caps == FW_CAPS16
+						 ? FW_PORT_ACTION_L1_CFG
+						 : FW_PORT_ACTION_L1_CFG32) |
+						 FW_LEN16(cmd));
+	if (fw_caps == FW_CAPS16)
+		cmd.u.l1cfg.rcap = cpu_to_be32(fwcaps32_to_caps16(rcap));
+	else
+		cmd.u.l1cfg32.rcap32 = cpu_to_be32(rcap);
 
-	if (!(lc->supported & FW_PORT_CAP_ANEG)) {
-		c.u.l1cfg.rcap = cpu_to_be32((lc->supported & ADVERT_MASK) |
-					     fc | fw_fec);
-		lc->fc = lc->requested_fc & (PAUSE_RX | PAUSE_TX);
-	} else if (lc->autoneg == AUTONEG_DISABLE) {
-		c.u.l1cfg.rcap = cpu_to_be32(lc->requested_speed | fc |
-					     fw_fec | mdi);
-		lc->fc = lc->requested_fc & (PAUSE_RX | PAUSE_TX);
-	} else
-		c.u.l1cfg.rcap = cpu_to_be32(lc->advertising | fc |
-					     fw_fec | mdi);
+	ret = t4_wr_mbox_meat_timeout(adapter, mbox, &cmd, sizeof(cmd), NULL,
+				      sleep_ok, timeout);
 
-	return t4_wr_mbox(adap, mbox, &c, sizeof(c), NULL);
+	/* Unfortunately, even if the Requested Port Capabilities "fit" within
+	 * the Physical Port Capabilities, some combinations of features may
+	 * still not be legal.  For example, 40Gb/s and Reed-Solomon Forward
+	 * Error Correction.  So if the Firmware rejects the L1 Configure
+	 * request, flag that here.
+	 */
+	if (ret) {
+		dev_err(adapter->pdev_dev,
+			"Requested Port Capabilities %#x rejected, error %d\n",
+			rcap, -ret);
+		return ret;
+	}
+	return 0;
 }
 
 /**
@@ -3756,6 +4241,7 @@ int t4_link_l1cfg(struct adapter *adap, unsigned int mbox, unsigned int port,
  */
 int t4_restart_aneg(struct adapter *adap, unsigned int mbox, unsigned int port)
 {
+	unsigned int fw_caps = adap->params.fw_caps_support;
 	struct fw_port_cmd c;
 
 	memset(&c, 0, sizeof(c));
@@ -3763,9 +4249,14 @@ int t4_restart_aneg(struct adapter *adap, unsigned int mbox, unsigned int port)
 				     FW_CMD_REQUEST_F | FW_CMD_EXEC_F |
 				     FW_PORT_CMD_PORTID_V(port));
 	c.action_to_len16 =
-		cpu_to_be32(FW_PORT_CMD_ACTION_V(FW_PORT_ACTION_L1_CFG) |
+		cpu_to_be32(FW_PORT_CMD_ACTION_V(fw_caps == FW_CAPS16
+						 ? FW_PORT_ACTION_L1_CFG
+						 : FW_PORT_ACTION_L1_CFG32) |
 			    FW_LEN16(c));
-	c.u.l1cfg.rcap = cpu_to_be32(FW_PORT_CAP_ANEG);
+	if (fw_caps == FW_CAPS16)
+		c.u.l1cfg.rcap = cpu_to_be32(FW_PORT_CAP_ANEG);
+	else
+		c.u.l1cfg32.rcap32 = cpu_to_be32(FW_PORT_CAP32_ANEG);
 	return t4_wr_mbox(adap, mbox, &c, sizeof(c), NULL);
 }
 
@@ -3958,7 +4449,7 @@ static void tp_intr_handler(struct adapter *adapter)
  */
 static void sge_intr_handler(struct adapter *adapter)
 {
-	u64 v;
+	u32 v = 0, perr;
 	u32 err;
 
 	static const struct intr_info sge_intr_info[] = {
@@ -3993,13 +4484,29 @@ static void sge_intr_handler(struct adapter *adapter)
 		{ 0 }
 	};
 
-	v = (u64)t4_read_reg(adapter, SGE_INT_CAUSE1_A) |
-		((u64)t4_read_reg(adapter, SGE_INT_CAUSE2_A) << 32);
-	if (v) {
-		dev_alert(adapter->pdev_dev, "SGE parity error (%#llx)\n",
-				(unsigned long long)v);
-		t4_write_reg(adapter, SGE_INT_CAUSE1_A, v);
-		t4_write_reg(adapter, SGE_INT_CAUSE2_A, v >> 32);
+	perr = t4_read_reg(adapter, SGE_INT_CAUSE1_A);
+	if (perr) {
+		v |= perr;
+		dev_alert(adapter->pdev_dev, "SGE Cause1 Parity Error %#x\n",
+			  perr);
+	}
+
+	perr = t4_read_reg(adapter, SGE_INT_CAUSE2_A);
+	if (perr) {
+		v |= perr;
+		dev_alert(adapter->pdev_dev, "SGE Cause2 Parity Error %#x\n",
+			  perr);
+	}
+
+	if (CHELSIO_CHIP_VERSION(adapter->params.chip) >= CHELSIO_T5) {
+		perr = t4_read_reg(adapter, SGE_INT_CAUSE5_A);
+		/* Parity error (CRC) for err_T_RxCRC is trivial, ignore it */
+		perr &= ~ERR_T_RXCRC_F;
+		if (perr) {
+			v |= perr;
+			dev_alert(adapter->pdev_dev,
+				  "SGE Cause5 Parity Error %#x\n", perr);
+		}
 	}
 
 	v |= t4_handle_intr_status(adapter, SGE_INT_CAUSE3_A, sge_intr_info);
@@ -4040,6 +4547,7 @@ static void cim_intr_handler(struct adapter *adapter)
 		{ MBHOSTPARERR_F, "CIM mailbox host parity error", -1, 1 },
 		{ TIEQINPARERRINT_F, "CIM TIEQ outgoing parity error", -1, 1 },
 		{ TIEQOUTPARERRINT_F, "CIM TIEQ incoming parity error", -1, 1 },
+		{ TIMER0INT_F, "CIM TIMER0 interrupt", -1, 1 },
 		{ 0 }
 	};
 	static const struct intr_info cim_upintr_info[] = {
@@ -4074,11 +4582,27 @@ static void cim_intr_handler(struct adapter *adapter)
 		{ 0 }
 	};
 
+	u32 val, fw_err;
 	int fat;
 
-	if (t4_read_reg(adapter, PCIE_FW_A) & PCIE_FW_ERR_F)
+	fw_err = t4_read_reg(adapter, PCIE_FW_A);
+	if (fw_err & PCIE_FW_ERR_F)
 		t4_report_fw_error(adapter);
 
+	/* When the Firmware detects an internal error which normally
+	 * wouldn't raise a Host Interrupt, it forces a CIM Timer0 interrupt
+	 * in order to make sure the Host sees the Firmware Crash.  So
+	 * if we have a Timer0 interrupt and don't see a Firmware Crash,
+	 * ignore the Timer0 interrupt.
+	 */
+
+	val = t4_read_reg(adapter, CIM_HOST_INT_CAUSE_A);
+	if (val & TIMER0INT_F)
+		if (!(fw_err & PCIE_FW_ERR_F) ||
+		    (PCIE_FW_EVAL_G(fw_err) != PCIE_FW_EVAL_CRASH))
+			t4_write_reg(adapter, CIM_HOST_INT_CAUSE_A,
+				     TIMER0INT_F);
+
 	fat = t4_handle_intr_status(adapter, CIM_HOST_INT_CAUSE_A,
 				    cim_intr_info) +
 	      t4_handle_intr_status(adapter, CIM_HOST_UPACC_INT_CAUSE_A,
@@ -4204,9 +4728,11 @@ static void le_intr_handler(struct adapter *adap)
 	static struct intr_info t6_le_intr_info[] = {
 		{ T6_LIPMISS_F, "LE LIP miss", -1, 0 },
 		{ T6_LIP0_F, "LE 0 LIP error", -1, 0 },
+		{ CMDTIDERR_F, "LE cmd tid error", -1, 1 },
 		{ TCAMINTPERR_F, "LE parity error", -1, 1 },
 		{ T6_UNKNOWNCMD_F, "LE unknown command", -1, 1 },
 		{ SSRAMINTPERR_F, "LE request queue parity error", -1, 1 },
+		{ HASHTBLMEMCRCERR_F, "LE hash table mem crc error", -1, 0 },
 		{ 0 }
 	};
 
@@ -4237,6 +4763,18 @@ static void mps_intr_handler(struct adapter *adapter)
 		{ FRMERR_F, "MPS Tx framing error", -1, 1 },
 		{ 0 }
 	};
+	static const struct intr_info t6_mps_tx_intr_info[] = {
+		{ TPFIFO_V(TPFIFO_M), "MPS Tx TP FIFO parity error", -1, 1 },
+		{ NCSIFIFO_F, "MPS Tx NC-SI FIFO parity error", -1, 1 },
+		{ TXDATAFIFO_V(TXDATAFIFO_M), "MPS Tx data FIFO parity error",
+		  -1, 1 },
+		{ TXDESCFIFO_V(TXDESCFIFO_M), "MPS Tx desc FIFO parity error",
+		  -1, 1 },
+		/* MPS Tx Bubble is normal for T6 */
+		{ SECNTERR_F, "MPS Tx SOP/EOP error", -1, 1 },
+		{ FRMERR_F, "MPS Tx framing error", -1, 1 },
+		{ 0 }
+	};
 	static const struct intr_info mps_trc_intr_info[] = {
 		{ FILTMEM_V(FILTMEM_M), "MPS TRC filter parity error", -1, 1 },
 		{ PKTFIFO_V(PKTFIFO_M), "MPS TRC packet FIFO parity error",
@@ -4268,7 +4806,9 @@ static void mps_intr_handler(struct adapter *adapter)
 	fat = t4_handle_intr_status(adapter, MPS_RX_PERR_INT_CAUSE_A,
 				    mps_rx_intr_info) +
 	      t4_handle_intr_status(adapter, MPS_TX_INT_CAUSE_A,
-				    mps_tx_intr_info) +
+				    is_t6(adapter->params.chip)
+				    ? t6_mps_tx_intr_info
+				    : mps_tx_intr_info) +
 	      t4_handle_intr_status(adapter, MPS_TRC_INT_CAUSE_A,
 				    mps_trc_intr_info) +
 	      t4_handle_intr_status(adapter, MPS_STAT_PERR_INT_CAUSE_SRAM_A,
@@ -4445,7 +4985,7 @@ static void pl_intr_handler(struct adapter *adap)
 #define PF_INTR_MASK (PFSW_F)
 #define GLBL_INTR_MASK (CIM_F | MPS_F | PL_F | PCIE_F | MC_F | EDC0_F | \
 		EDC1_F | LE_F | TP_F | MA_F | PM_TX_F | PM_RX_F | ULP_RX_F | \
-		CPL_SWITCH_F | SGE_F | ULP_TX_F)
+		CPL_SWITCH_F | SGE_F | ULP_TX_F | SF_F)
 
 /**
  *	t4_slow_intr_handler - control path interrupt handler
@@ -4457,7 +4997,13 @@ static void pl_intr_handler(struct adapter *adap)
  */
 int t4_slow_intr_handler(struct adapter *adapter)
 {
-	u32 cause = t4_read_reg(adapter, PL_INT_CAUSE_A);
+	/* There are rare cases where a PL_INT_CAUSE bit may end up getting
+	 * set when the corresponding PL_INT_ENABLE bit isn't set.  It's
+	 * easiest just to mask that case here.
+	 */
+	u32 raw_cause = t4_read_reg(adapter, PL_INT_CAUSE_A);
+	u32 enable = t4_read_reg(adapter, PL_INT_ENABLE_A);
+	u32 cause = raw_cause & enable;
 
 	if (!(cause & GLBL_INTR_MASK))
 		return 0;
@@ -4509,7 +5055,7 @@ int t4_slow_intr_handler(struct adapter *adapter)
 		ulptx_intr_handler(adapter);
 
 	/* Clear the interrupts just processed for which we are the master. */
-	t4_write_reg(adapter, PL_INT_CAUSE_A, cause & GLBL_INTR_MASK);
+	t4_write_reg(adapter, PL_INT_CAUSE_A, raw_cause & GLBL_INTR_MASK);
 	(void)t4_read_reg(adapter, PL_INT_CAUSE_A); /* flush */
 	return 1;
 }
@@ -4557,14 +5103,27 @@ void t4_intr_enable(struct adapter *adapter)
  */
 void t4_intr_disable(struct adapter *adapter)
 {
-	u32 whoami = t4_read_reg(adapter, PL_WHOAMI_A);
-	u32 pf = CHELSIO_CHIP_VERSION(adapter->params.chip) <= CHELSIO_T5 ?
+	u32 whoami, pf;
+
+	if (pci_channel_offline(adapter->pdev))
+		return;
+
+	whoami = t4_read_reg(adapter, PL_WHOAMI_A);
+	pf = CHELSIO_CHIP_VERSION(adapter->params.chip) <= CHELSIO_T5 ?
 			SOURCEPF_G(whoami) : T6_SOURCEPF_G(whoami);
 
 	t4_write_reg(adapter, MYPF_REG(PL_PF_INT_ENABLE_A), 0);
 	t4_set_reg_field(adapter, PL_INT_MAP0_A, 1 << pf, 0);
 }
 
+unsigned int t4_chip_rss_size(struct adapter *adap)
+{
+	if (CHELSIO_CHIP_VERSION(adap->params.chip) <= CHELSIO_T5)
+		return RSS_NENTRIES;
+	else
+		return T6_RSS_NENTRIES;
+}
+
 /**
  *	t4_config_rss_range - configure a portion of the RSS mapping table
  *	@adapter: the adapter
@@ -4703,10 +5262,11 @@ static int rd_rss_row(struct adapter *adap, int row, u32 *val)
  */
 int t4_read_rss(struct adapter *adapter, u16 *map)
 {
+	int i, ret, nentries;
 	u32 val;
-	int i, ret;
 
-	for (i = 0; i < RSS_NENTRIES / 2; ++i) {
+	nentries = t4_chip_rss_size(adapter);
+	for (i = 0; i < nentries / 2; ++i) {
 		ret = rd_rss_row(adapter, i, &val);
 		if (ret)
 			return ret;
@@ -4718,27 +5278,30 @@ int t4_read_rss(struct adapter *adapter, u16 *map)
 
 static unsigned int t4_use_ldst(struct adapter *adap)
 {
-	return (adap->flags & FW_OK) || !adap->use_bd;
+	return (adap->flags & CXGB4_FW_OK) && !adap->use_bd;
 }
 
 /**
- *	t4_fw_tp_pio_rw - Access TP PIO through LDST
- *	@adap: the adapter
- *	@vals: where the indirect register values are stored/written
- *	@nregs: how many indirect registers to read/write
- *	@start_idx: index of first indirect register to read/write
- *	@rw: Read (1) or Write (0)
+ * t4_tp_fw_ldst_rw - Access TP indirect register through LDST
+ * @adap: the adapter
+ * @cmd: TP fw ldst address space type
+ * @vals: where the indirect register values are stored/written
+ * @nregs: how many indirect registers to read/write
+ * @start_index: index of first indirect register to read/write
+ * @rw: Read (1) or Write (0)
+ * @sleep_ok: if true we may sleep while awaiting command completion
  *
- *	Access TP PIO registers through LDST
+ * Access TP indirect registers through LDST
  */
-static void t4_fw_tp_pio_rw(struct adapter *adap, u32 *vals, unsigned int nregs,
-			    unsigned int start_index, unsigned int rw)
+static int t4_tp_fw_ldst_rw(struct adapter *adap, int cmd, u32 *vals,
+			    unsigned int nregs, unsigned int start_index,
+			    unsigned int rw, bool sleep_ok)
 {
-	int ret, i;
-	int cmd = FW_LDST_ADDRSPC_TP_PIO;
+	int ret = 0;
+	unsigned int i;
 	struct fw_ldst_cmd c;
 
-	for (i = 0 ; i < nregs; i++) {
+	for (i = 0; i < nregs; i++) {
 		memset(&c, 0, sizeof(c));
 		c.op_to_addrspace = cpu_to_be32(FW_CMD_OP_V(FW_LDST_CMD) |
 						FW_CMD_REQUEST_F |
@@ -4749,26 +5312,147 @@ static void t4_fw_tp_pio_rw(struct adapter *adap, u32 *vals, unsigned int nregs,
 
 		c.u.addrval.addr = cpu_to_be32(start_index + i);
 		c.u.addrval.val  = rw ? 0 : cpu_to_be32(vals[i]);
-		ret = t4_wr_mbox(adap, adap->mbox, &c, sizeof(c), &c);
-		if (!ret && rw)
+		ret = t4_wr_mbox_meat(adap, adap->mbox, &c, sizeof(c), &c,
+				      sleep_ok);
+		if (ret)
+			return ret;
+
+		if (rw)
 			vals[i] = be32_to_cpu(c.u.addrval.val);
 	}
+	return 0;
+}
+
+/**
+ * t4_tp_indirect_rw - Read/Write TP indirect register through LDST or backdoor
+ * @adap: the adapter
+ * @reg_addr: Address Register
+ * @reg_data: Data register
+ * @buff: where the indirect register values are stored/written
+ * @nregs: how many indirect registers to read/write
+ * @start_index: index of first indirect register to read/write
+ * @rw: READ(1) or WRITE(0)
+ * @sleep_ok: if true we may sleep while awaiting command completion
+ *
+ * Read/Write TP indirect registers through LDST if possible.
+ * Else, use backdoor access
+ **/
+static void t4_tp_indirect_rw(struct adapter *adap, u32 reg_addr, u32 reg_data,
+			      u32 *buff, u32 nregs, u32 start_index, int rw,
+			      bool sleep_ok)
+{
+	int rc = -EINVAL;
+	int cmd;
+
+	switch (reg_addr) {
+	case TP_PIO_ADDR_A:
+		cmd = FW_LDST_ADDRSPC_TP_PIO;
+		break;
+	case TP_TM_PIO_ADDR_A:
+		cmd = FW_LDST_ADDRSPC_TP_TM_PIO;
+		break;
+	case TP_MIB_INDEX_A:
+		cmd = FW_LDST_ADDRSPC_TP_MIB;
+		break;
+	default:
+		goto indirect_access;
+	}
+
+	if (t4_use_ldst(adap))
+		rc = t4_tp_fw_ldst_rw(adap, cmd, buff, nregs, start_index, rw,
+				      sleep_ok);
+
+indirect_access:
+
+	if (rc) {
+		if (rw)
+			t4_read_indirect(adap, reg_addr, reg_data, buff, nregs,
+					 start_index);
+		else
+			t4_write_indirect(adap, reg_addr, reg_data, buff, nregs,
+					  start_index);
+	}
+}
+
+/**
+ * t4_tp_pio_read - Read TP PIO registers
+ * @adap: the adapter
+ * @buff: where the indirect register values are written
+ * @nregs: how many indirect registers to read
+ * @start_index: index of first indirect register to read
+ * @sleep_ok: if true we may sleep while awaiting command completion
+ *
+ * Read TP PIO Registers
+ **/
+void t4_tp_pio_read(struct adapter *adap, u32 *buff, u32 nregs,
+		    u32 start_index, bool sleep_ok)
+{
+	t4_tp_indirect_rw(adap, TP_PIO_ADDR_A, TP_PIO_DATA_A, buff, nregs,
+			  start_index, 1, sleep_ok);
+}
+
+/**
+ * t4_tp_pio_write - Write TP PIO registers
+ * @adap: the adapter
+ * @buff: where the indirect register values are stored
+ * @nregs: how many indirect registers to write
+ * @start_index: index of first indirect register to write
+ * @sleep_ok: if true we may sleep while awaiting command completion
+ *
+ * Write TP PIO Registers
+ **/
+static void t4_tp_pio_write(struct adapter *adap, u32 *buff, u32 nregs,
+			    u32 start_index, bool sleep_ok)
+{
+	t4_tp_indirect_rw(adap, TP_PIO_ADDR_A, TP_PIO_DATA_A, buff, nregs,
+			  start_index, 0, sleep_ok);
+}
+
+/**
+ * t4_tp_tm_pio_read - Read TP TM PIO registers
+ * @adap: the adapter
+ * @buff: where the indirect register values are written
+ * @nregs: how many indirect registers to read
+ * @start_index: index of first indirect register to read
+ * @sleep_ok: if true we may sleep while awaiting command completion
+ *
+ * Read TP TM PIO Registers
+ **/
+void t4_tp_tm_pio_read(struct adapter *adap, u32 *buff, u32 nregs,
+		       u32 start_index, bool sleep_ok)
+{
+	t4_tp_indirect_rw(adap, TP_TM_PIO_ADDR_A, TP_TM_PIO_DATA_A, buff,
+			  nregs, start_index, 1, sleep_ok);
+}
+
+/**
+ * t4_tp_mib_read - Read TP MIB registers
+ * @adap: the adapter
+ * @buff: where the indirect register values are written
+ * @nregs: how many indirect registers to read
+ * @start_index: index of first indirect register to read
+ * @sleep_ok: if true we may sleep while awaiting command completion
+ *
+ * Read TP MIB Registers
+ **/
+void t4_tp_mib_read(struct adapter *adap, u32 *buff, u32 nregs, u32 start_index,
+		    bool sleep_ok)
+{
+	t4_tp_indirect_rw(adap, TP_MIB_INDEX_A, TP_MIB_DATA_A, buff, nregs,
+			  start_index, 1, sleep_ok);
 }
 
 /**
  *	t4_read_rss_key - read the global RSS key
  *	@adap: the adapter
  *	@key: 10-entry array holding the 320-bit RSS key
+ *      @sleep_ok: if true we may sleep while awaiting command completion
  *
  *	Reads the global 320-bit RSS key.
  */
-void t4_read_rss_key(struct adapter *adap, u32 *key)
+void t4_read_rss_key(struct adapter *adap, u32 *key, bool sleep_ok)
 {
-	if (t4_use_ldst(adap))
-		t4_fw_tp_pio_rw(adap, key, 10, TP_RSS_SECRET_KEY0_A, 1);
-	else
-		t4_read_indirect(adap, TP_PIO_ADDR_A, TP_PIO_DATA_A, key, 10,
-				 TP_RSS_SECRET_KEY0_A);
+	t4_tp_pio_read(adap, key, 10, TP_RSS_SECRET_KEY0_A, sleep_ok);
 }
 
 /**
@@ -4776,12 +5460,14 @@ void t4_read_rss_key(struct adapter *adap, u32 *key)
  *	@adap: the adapter
  *	@key: 10-entry array holding the 320-bit RSS key
  *	@idx: which RSS key to write
+ *      @sleep_ok: if true we may sleep while awaiting command completion
  *
  *	Writes one of the RSS keys with the given 320-bit value.  If @idx is
  *	0..15 the corresponding entry in the RSS key table is written,
  *	otherwise the global RSS key is written.
  */
-void t4_write_rss_key(struct adapter *adap, const u32 *key, int idx)
+void t4_write_rss_key(struct adapter *adap, const u32 *key, int idx,
+		      bool sleep_ok)
 {
 	u8 rss_key_addr_cnt = 16;
 	u32 vrt = t4_read_reg(adap, TP_RSS_CONFIG_VRT_A);
@@ -4794,11 +5480,7 @@ void t4_write_rss_key(struct adapter *adap, const u32 *key, int idx)
 	    (vrt & KEYEXTEND_F) && (KEYMODE_G(vrt) == 3))
 		rss_key_addr_cnt = 32;
 
-	if (t4_use_ldst(adap))
-		t4_fw_tp_pio_rw(adap, (void *)key, 10, TP_RSS_SECRET_KEY0_A, 0);
-	else
-		t4_write_indirect(adap, TP_PIO_ADDR_A, TP_PIO_DATA_A, key, 10,
-				  TP_RSS_SECRET_KEY0_A);
+	t4_tp_pio_write(adap, (void *)key, 10, TP_RSS_SECRET_KEY0_A, sleep_ok);
 
 	if (idx >= 0 && idx < rss_key_addr_cnt) {
 		if (rss_key_addr_cnt > 16)
@@ -4816,19 +5498,15 @@ void t4_write_rss_key(struct adapter *adap, const u32 *key, int idx)
  *	@adapter: the adapter
  *	@index: the entry in the PF RSS table to read
  *	@valp: where to store the returned value
+ *      @sleep_ok: if true we may sleep while awaiting command completion
  *
  *	Reads the PF RSS Configuration Table at the specified index and returns
  *	the value found there.
  */
 void t4_read_rss_pf_config(struct adapter *adapter, unsigned int index,
-			   u32 *valp)
+			   u32 *valp, bool sleep_ok)
 {
-	if (t4_use_ldst(adapter))
-		t4_fw_tp_pio_rw(adapter, valp, 1,
-				TP_RSS_PF0_CONFIG_A + index, 1);
-	else
-		t4_read_indirect(adapter, TP_PIO_ADDR_A, TP_PIO_DATA_A,
-				 valp, 1, TP_RSS_PF0_CONFIG_A + index);
+	t4_tp_pio_read(adapter, valp, 1, TP_RSS_PF0_CONFIG_A + index, sleep_ok);
 }
 
 /**
@@ -4837,12 +5515,13 @@ void t4_read_rss_pf_config(struct adapter *adapter, unsigned int index,
  *	@index: the entry in the VF RSS table to read
  *	@vfl: where to store the returned VFL
  *	@vfh: where to store the returned VFH
+ *      @sleep_ok: if true we may sleep while awaiting command completion
  *
  *	Reads the VF RSS Configuration Table at the specified index and returns
  *	the (VFL, VFH) values found there.
  */
 void t4_read_rss_vf_config(struct adapter *adapter, unsigned int index,
-			   u32 *vfl, u32 *vfh)
+			   u32 *vfl, u32 *vfh, bool sleep_ok)
 {
 	u32 vrt, mask, data;
 
@@ -4863,50 +5542,37 @@ void t4_read_rss_vf_config(struct adapter *adapter, unsigned int index,
 
 	/* Grab the VFL/VFH values ...
 	 */
-	if (t4_use_ldst(adapter)) {
-		t4_fw_tp_pio_rw(adapter, vfl, 1, TP_RSS_VFL_CONFIG_A, 1);
-		t4_fw_tp_pio_rw(adapter, vfh, 1, TP_RSS_VFH_CONFIG_A, 1);
-	} else {
-		t4_read_indirect(adapter, TP_PIO_ADDR_A, TP_PIO_DATA_A,
-				 vfl, 1, TP_RSS_VFL_CONFIG_A);
-		t4_read_indirect(adapter, TP_PIO_ADDR_A, TP_PIO_DATA_A,
-				 vfh, 1, TP_RSS_VFH_CONFIG_A);
-	}
+	t4_tp_pio_read(adapter, vfl, 1, TP_RSS_VFL_CONFIG_A, sleep_ok);
+	t4_tp_pio_read(adapter, vfh, 1, TP_RSS_VFH_CONFIG_A, sleep_ok);
 }
 
 /**
  *	t4_read_rss_pf_map - read PF RSS Map
  *	@adapter: the adapter
+ *      @sleep_ok: if true we may sleep while awaiting command completion
  *
  *	Reads the PF RSS Map register and returns its value.
  */
-u32 t4_read_rss_pf_map(struct adapter *adapter)
+u32 t4_read_rss_pf_map(struct adapter *adapter, bool sleep_ok)
 {
 	u32 pfmap;
 
-	if (t4_use_ldst(adapter))
-		t4_fw_tp_pio_rw(adapter, &pfmap, 1, TP_RSS_PF_MAP_A, 1);
-	else
-		t4_read_indirect(adapter, TP_PIO_ADDR_A, TP_PIO_DATA_A,
-				 &pfmap, 1, TP_RSS_PF_MAP_A);
+	t4_tp_pio_read(adapter, &pfmap, 1, TP_RSS_PF_MAP_A, sleep_ok);
 	return pfmap;
 }
 
 /**
  *	t4_read_rss_pf_mask - read PF RSS Mask
  *	@adapter: the adapter
+ *      @sleep_ok: if true we may sleep while awaiting command completion
  *
  *	Reads the PF RSS Mask register and returns its value.
  */
-u32 t4_read_rss_pf_mask(struct adapter *adapter)
+u32 t4_read_rss_pf_mask(struct adapter *adapter, bool sleep_ok)
 {
 	u32 pfmask;
 
-	if (t4_use_ldst(adapter))
-		t4_fw_tp_pio_rw(adapter, &pfmask, 1, TP_RSS_PF_MSK_A, 1);
-	else
-		t4_read_indirect(adapter, TP_PIO_ADDR_A, TP_PIO_DATA_A,
-				 &pfmask, 1, TP_RSS_PF_MSK_A);
+	t4_tp_pio_read(adapter, &pfmask, 1, TP_RSS_PF_MSK_A, sleep_ok);
 	return pfmask;
 }
 
@@ -4915,12 +5581,13 @@ u32 t4_read_rss_pf_mask(struct adapter *adapter)
  *	@adap: the adapter
  *	@v4: holds the TCP/IP counter values
  *	@v6: holds the TCP/IPv6 counter values
+ *      @sleep_ok: if true we may sleep while awaiting command completion
  *
  *	Returns the values of TP's TCP/IP and TCP/IPv6 MIB counters.
  *	Either @v4 or @v6 may be %NULL to skip the corresponding stats.
  */
 void t4_tp_get_tcp_stats(struct adapter *adap, struct tp_tcp_stats *v4,
-			 struct tp_tcp_stats *v6)
+			 struct tp_tcp_stats *v6, bool sleep_ok)
 {
 	u32 val[TP_MIB_TCP_RXT_SEG_LO_A - TP_MIB_TCP_OUT_RST_A + 1];
 
@@ -4929,16 +5596,16 @@ void t4_tp_get_tcp_stats(struct adapter *adap, struct tp_tcp_stats *v4,
 #define STAT64(x)   (((u64)STAT(x##_HI) << 32) | STAT(x##_LO))
 
 	if (v4) {
-		t4_read_indirect(adap, TP_MIB_INDEX_A, TP_MIB_DATA_A, val,
-				 ARRAY_SIZE(val), TP_MIB_TCP_OUT_RST_A);
+		t4_tp_mib_read(adap, val, ARRAY_SIZE(val),
+			       TP_MIB_TCP_OUT_RST_A, sleep_ok);
 		v4->tcp_out_rsts = STAT(OUT_RST);
 		v4->tcp_in_segs  = STAT64(IN_SEG);
 		v4->tcp_out_segs = STAT64(OUT_SEG);
 		v4->tcp_retrans_segs = STAT64(RXT_SEG);
 	}
 	if (v6) {
-		t4_read_indirect(adap, TP_MIB_INDEX_A, TP_MIB_DATA_A, val,
-				 ARRAY_SIZE(val), TP_MIB_TCP_V6OUT_RST_A);
+		t4_tp_mib_read(adap, val, ARRAY_SIZE(val),
+			       TP_MIB_TCP_V6OUT_RST_A, sleep_ok);
 		v6->tcp_out_rsts = STAT(OUT_RST);
 		v6->tcp_in_segs  = STAT64(IN_SEG);
 		v6->tcp_out_segs = STAT64(OUT_SEG);
@@ -4953,63 +5620,66 @@ void t4_tp_get_tcp_stats(struct adapter *adap, struct tp_tcp_stats *v4,
  *	t4_tp_get_err_stats - read TP's error MIB counters
  *	@adap: the adapter
  *	@st: holds the counter values
+ *      @sleep_ok: if true we may sleep while awaiting command completion
  *
  *	Returns the values of TP's error counters.
  */
-void t4_tp_get_err_stats(struct adapter *adap, struct tp_err_stats *st)
+void t4_tp_get_err_stats(struct adapter *adap, struct tp_err_stats *st,
+			 bool sleep_ok)
 {
 	int nchan = adap->params.arch.nchan;
 
-	t4_read_indirect(adap, TP_MIB_INDEX_A, TP_MIB_DATA_A,
-			 st->mac_in_errs, nchan, TP_MIB_MAC_IN_ERR_0_A);
-	t4_read_indirect(adap, TP_MIB_INDEX_A, TP_MIB_DATA_A,
-			 st->hdr_in_errs, nchan, TP_MIB_HDR_IN_ERR_0_A);
-	t4_read_indirect(adap, TP_MIB_INDEX_A, TP_MIB_DATA_A,
-			 st->tcp_in_errs, nchan, TP_MIB_TCP_IN_ERR_0_A);
-	t4_read_indirect(adap, TP_MIB_INDEX_A, TP_MIB_DATA_A,
-			 st->tnl_cong_drops, nchan, TP_MIB_TNL_CNG_DROP_0_A);
-	t4_read_indirect(adap, TP_MIB_INDEX_A, TP_MIB_DATA_A,
-			 st->ofld_chan_drops, nchan, TP_MIB_OFD_CHN_DROP_0_A);
-	t4_read_indirect(adap, TP_MIB_INDEX_A, TP_MIB_DATA_A,
-			 st->tnl_tx_drops, nchan, TP_MIB_TNL_DROP_0_A);
-	t4_read_indirect(adap, TP_MIB_INDEX_A, TP_MIB_DATA_A,
-			 st->ofld_vlan_drops, nchan, TP_MIB_OFD_VLN_DROP_0_A);
-	t4_read_indirect(adap, TP_MIB_INDEX_A, TP_MIB_DATA_A,
-			 st->tcp6_in_errs, nchan, TP_MIB_TCP_V6IN_ERR_0_A);
-
-	t4_read_indirect(adap, TP_MIB_INDEX_A, TP_MIB_DATA_A,
-			 &st->ofld_no_neigh, 2, TP_MIB_OFD_ARP_DROP_A);
+	t4_tp_mib_read(adap, st->mac_in_errs, nchan, TP_MIB_MAC_IN_ERR_0_A,
+		       sleep_ok);
+	t4_tp_mib_read(adap, st->hdr_in_errs, nchan, TP_MIB_HDR_IN_ERR_0_A,
+		       sleep_ok);
+	t4_tp_mib_read(adap, st->tcp_in_errs, nchan, TP_MIB_TCP_IN_ERR_0_A,
+		       sleep_ok);
+	t4_tp_mib_read(adap, st->tnl_cong_drops, nchan,
+		       TP_MIB_TNL_CNG_DROP_0_A, sleep_ok);
+	t4_tp_mib_read(adap, st->ofld_chan_drops, nchan,
+		       TP_MIB_OFD_CHN_DROP_0_A, sleep_ok);
+	t4_tp_mib_read(adap, st->tnl_tx_drops, nchan, TP_MIB_TNL_DROP_0_A,
+		       sleep_ok);
+	t4_tp_mib_read(adap, st->ofld_vlan_drops, nchan,
+		       TP_MIB_OFD_VLN_DROP_0_A, sleep_ok);
+	t4_tp_mib_read(adap, st->tcp6_in_errs, nchan,
+		       TP_MIB_TCP_V6IN_ERR_0_A, sleep_ok);
+	t4_tp_mib_read(adap, &st->ofld_no_neigh, 2, TP_MIB_OFD_ARP_DROP_A,
+		       sleep_ok);
 }
 
 /**
  *	t4_tp_get_cpl_stats - read TP's CPL MIB counters
  *	@adap: the adapter
  *	@st: holds the counter values
+ *      @sleep_ok: if true we may sleep while awaiting command completion
  *
  *	Returns the values of TP's CPL counters.
  */
-void t4_tp_get_cpl_stats(struct adapter *adap, struct tp_cpl_stats *st)
+void t4_tp_get_cpl_stats(struct adapter *adap, struct tp_cpl_stats *st,
+			 bool sleep_ok)
 {
 	int nchan = adap->params.arch.nchan;
 
-	t4_read_indirect(adap, TP_MIB_INDEX_A, TP_MIB_DATA_A, st->req,
-			 nchan, TP_MIB_CPL_IN_REQ_0_A);
-	t4_read_indirect(adap, TP_MIB_INDEX_A, TP_MIB_DATA_A, st->rsp,
-			 nchan, TP_MIB_CPL_OUT_RSP_0_A);
+	t4_tp_mib_read(adap, st->req, nchan, TP_MIB_CPL_IN_REQ_0_A, sleep_ok);
 
+	t4_tp_mib_read(adap, st->rsp, nchan, TP_MIB_CPL_OUT_RSP_0_A, sleep_ok);
 }
 
 /**
  *	t4_tp_get_rdma_stats - read TP's RDMA MIB counters
  *	@adap: the adapter
  *	@st: holds the counter values
+ *      @sleep_ok: if true we may sleep while awaiting command completion
  *
  *	Returns the values of TP's RDMA counters.
  */
-void t4_tp_get_rdma_stats(struct adapter *adap, struct tp_rdma_stats *st)
+void t4_tp_get_rdma_stats(struct adapter *adap, struct tp_rdma_stats *st,
+			  bool sleep_ok)
 {
-	t4_read_indirect(adap, TP_MIB_INDEX_A, TP_MIB_DATA_A, &st->rqe_dfr_pkt,
-			 2, TP_MIB_RQE_DFR_PKT_A);
+	t4_tp_mib_read(adap, &st->rqe_dfr_pkt, 2, TP_MIB_RQE_DFR_PKT_A,
+		       sleep_ok);
 }
 
 /**
@@ -5017,20 +5687,24 @@ void t4_tp_get_rdma_stats(struct adapter *adap, struct tp_rdma_stats *st)
  *	@adap: the adapter
  *	@idx: the port index
  *	@st: holds the counter values
+ *      @sleep_ok: if true we may sleep while awaiting command completion
  *
  *	Returns the values of TP's FCoE counters for the selected port.
  */
 void t4_get_fcoe_stats(struct adapter *adap, unsigned int idx,
-		       struct tp_fcoe_stats *st)
+		       struct tp_fcoe_stats *st, bool sleep_ok)
 {
 	u32 val[2];
 
-	t4_read_indirect(adap, TP_MIB_INDEX_A, TP_MIB_DATA_A, &st->frames_ddp,
-			 1, TP_MIB_FCOE_DDP_0_A + idx);
-	t4_read_indirect(adap, TP_MIB_INDEX_A, TP_MIB_DATA_A, &st->frames_drop,
-			 1, TP_MIB_FCOE_DROP_0_A + idx);
-	t4_read_indirect(adap, TP_MIB_INDEX_A, TP_MIB_DATA_A, val,
-			 2, TP_MIB_FCOE_BYTE_0_HI_A + 2 * idx);
+	t4_tp_mib_read(adap, &st->frames_ddp, 1, TP_MIB_FCOE_DDP_0_A + idx,
+		       sleep_ok);
+
+	t4_tp_mib_read(adap, &st->frames_drop, 1,
+		       TP_MIB_FCOE_DROP_0_A + idx, sleep_ok);
+
+	t4_tp_mib_read(adap, val, 2, TP_MIB_FCOE_BYTE_0_HI_A + 2 * idx,
+		       sleep_ok);
+
 	st->octets_ddp = ((u64)val[0] << 32) | val[1];
 }
 
@@ -5038,15 +5712,16 @@ void t4_get_fcoe_stats(struct adapter *adap, unsigned int idx,
  *	t4_get_usm_stats - read TP's non-TCP DDP MIB counters
  *	@adap: the adapter
  *	@st: holds the counter values
+ *      @sleep_ok: if true we may sleep while awaiting command completion
  *
  *	Returns the values of TP's counters for non-TCP directly-placed packets.
  */
-void t4_get_usm_stats(struct adapter *adap, struct tp_usm_stats *st)
+void t4_get_usm_stats(struct adapter *adap, struct tp_usm_stats *st,
+		      bool sleep_ok)
 {
 	u32 val[4];
 
-	t4_read_indirect(adap, TP_MIB_INDEX_A, TP_MIB_DATA_A, val, 4,
-			 TP_MIB_USM_PKTS_A);
+	t4_tp_mib_read(adap, val, 4, TP_MIB_USM_PKTS_A, sleep_ok);
 	st->frames = val[0];
 	st->drops = val[1];
 	st->octets = ((u64)val[2] << 32) | val[3];
@@ -5266,7 +5941,6 @@ int t4_set_trace_filter(struct adapter *adap, const struct trace_params *tp,
 {
 	int i, ofst = idx * 4;
 	u32 data_reg, mask_reg, cfg;
-	u32 multitrc = TRCMULTIFILTER_F;
 
 	if (!enable) {
 		t4_write_reg(adap, MPS_TRC_FILTER_MATCH_CTL_A_A + ofst, 0);
@@ -5286,7 +5960,6 @@ int t4_set_trace_filter(struct adapter *adap, const struct trace_params *tp,
 		 * maximum packet capture size of 9600 bytes is recommended.
 		 * Also in this mode, only trace0 can be enabled and running.
 		 */
-		multitrc = 0;
 		if (tp->snap_len > 9600 || idx)
 			return -EINVAL;
 	}
@@ -5418,30 +6091,187 @@ void t4_pmrx_get_stats(struct adapter *adap, u32 cnt[], u64 cycles[])
 }
 
 /**
+ *	compute_mps_bg_map - compute the MPS Buffer Group Map for a Port
+ *	@adapter: the adapter
+ *	@pidx: the port index
+ *
+ *	Computes and returns a bitmap indicating which MPS buffer groups are
+ *	associated with the given Port.  Bit i is set if buffer group i is
+ *	used by the Port.
+ */
+static inline unsigned int compute_mps_bg_map(struct adapter *adapter,
+					      int pidx)
+{
+	unsigned int chip_version, nports;
+
+	chip_version = CHELSIO_CHIP_VERSION(adapter->params.chip);
+	nports = 1 << NUMPORTS_G(t4_read_reg(adapter, MPS_CMN_CTL_A));
+
+	switch (chip_version) {
+	case CHELSIO_T4:
+	case CHELSIO_T5:
+		switch (nports) {
+		case 1: return 0xf;
+		case 2: return 3 << (2 * pidx);
+		case 4: return 1 << pidx;
+		}
+		break;
+
+	case CHELSIO_T6:
+		switch (nports) {
+		case 2: return 1 << (2 * pidx);
+		}
+		break;
+	}
+
+	dev_err(adapter->pdev_dev, "Need MPS Buffer Group Map for Chip %0x, Nports %d\n",
+		chip_version, nports);
+
+	return 0;
+}
+
+/**
  *	t4_get_mps_bg_map - return the buffer groups associated with a port
- *	@adap: the adapter
- *	@idx: the port index
+ *	@adapter: the adapter
+ *	@pidx: the port index
  *
  *	Returns a bitmap indicating which MPS buffer groups are associated
- *	with the given port.  Bit i is set if buffer group i is used by the
- *	port.
+ *	with the given Port.  Bit i is set if buffer group i is used by the
+ *	Port.
  */
-unsigned int t4_get_mps_bg_map(struct adapter *adap, int idx)
+unsigned int t4_get_mps_bg_map(struct adapter *adapter, int pidx)
 {
-	u32 n = NUMPORTS_G(t4_read_reg(adap, MPS_CMN_CTL_A));
+	u8 *mps_bg_map;
+	unsigned int nports;
+
+	nports = 1 << NUMPORTS_G(t4_read_reg(adapter, MPS_CMN_CTL_A));
+	if (pidx >= nports) {
+		CH_WARN(adapter, "MPS Port Index %d >= Nports %d\n",
+			pidx, nports);
+		return 0;
+	}
+
+	/* If we've already retrieved/computed this, just return the result.
+	 */
+	mps_bg_map = adapter->params.mps_bg_map;
+	if (mps_bg_map[pidx])
+		return mps_bg_map[pidx];
+
+	/* Newer Firmware can tell us what the MPS Buffer Group Map is.
+	 * If we're talking to such Firmware, let it tell us.  If the new
+	 * API isn't supported, revert back to old hardcoded way.  The value
+	 * obtained from Firmware is encoded in below format:
+	 *
+	 * val = (( MPSBGMAP[Port 3] << 24 ) |
+	 *        ( MPSBGMAP[Port 2] << 16 ) |
+	 *        ( MPSBGMAP[Port 1] <<  8 ) |
+	 *        ( MPSBGMAP[Port 0] <<  0 ))
+	 */
+	if (adapter->flags & CXGB4_FW_OK) {
+		u32 param, val;
+		int ret;
+
+		param = (FW_PARAMS_MNEM_V(FW_PARAMS_MNEM_DEV) |
+			 FW_PARAMS_PARAM_X_V(FW_PARAMS_PARAM_DEV_MPSBGMAP));
+		ret = t4_query_params_ns(adapter, adapter->mbox, adapter->pf,
+					 0, 1, &param, &val);
+		if (!ret) {
+			int p;
+
+			/* Store the BG Map for all of the Ports in order to
+			 * avoid more calls to the Firmware in the future.
+			 */
+			for (p = 0; p < MAX_NPORTS; p++, val >>= 8)
+				mps_bg_map[p] = val & 0xff;
+
+			return mps_bg_map[pidx];
+		}
+	}
 
-	if (n == 0)
-		return idx == 0 ? 0xf : 0;
-	/* In T6 (which is a 2 port card),
-	 * port 0 is mapped to channel 0 and port 1 is mapped to channel 1.
-	 * For 2 port T4/T5 adapter,
-	 * port 0 is mapped to channel 0 and 1,
-	 * port 1 is mapped to channel 2 and 3.
+	/* Either we're not talking to the Firmware or we're dealing with
+	 * older Firmware which doesn't support the new API to get the MPS
+	 * Buffer Group Map.  Fall back to computing it ourselves.
 	 */
-	if ((n == 1) &&
-	    (CHELSIO_CHIP_VERSION(adap->params.chip) <= CHELSIO_T5))
-		return idx < 2 ? (3 << (2 * idx)) : 0;
-	return 1 << idx;
+	mps_bg_map[pidx] = compute_mps_bg_map(adapter, pidx);
+	return mps_bg_map[pidx];
+}
+
+/**
+ *      t4_get_tp_e2c_map - return the E2C channel map associated with a port
+ *      @adapter: the adapter
+ *      @pidx: the port index
+ */
+static unsigned int t4_get_tp_e2c_map(struct adapter *adapter, int pidx)
+{
+	unsigned int nports;
+	u32 param, val = 0;
+	int ret;
+
+	nports = 1 << NUMPORTS_G(t4_read_reg(adapter, MPS_CMN_CTL_A));
+	if (pidx >= nports) {
+		CH_WARN(adapter, "TP E2C Channel Port Index %d >= Nports %d\n",
+			pidx, nports);
+		return 0;
+	}
+
+	/* FW version >= 1.16.44.0 can determine E2C channel map using
+	 * FW_PARAMS_PARAM_DEV_TPCHMAP API.
+	 */
+	param = (FW_PARAMS_MNEM_V(FW_PARAMS_MNEM_DEV) |
+		 FW_PARAMS_PARAM_X_V(FW_PARAMS_PARAM_DEV_TPCHMAP));
+	ret = t4_query_params_ns(adapter, adapter->mbox, adapter->pf,
+				 0, 1, &param, &val);
+	if (!ret)
+		return (val >> (8 * pidx)) & 0xff;
+
+	return 0;
+}
+
+/**
+ *	t4_get_tp_ch_map - return TP ingress channels associated with a port
+ *	@adap: the adapter
+ *	@pidx: the port index
+ *
+ *	Returns a bitmap indicating which TP Ingress Channels are associated
+ *	with a given Port.  Bit i is set if TP Ingress Channel i is used by
+ *	the Port.
+ */
+unsigned int t4_get_tp_ch_map(struct adapter *adap, int pidx)
+{
+	unsigned int chip_version = CHELSIO_CHIP_VERSION(adap->params.chip);
+	unsigned int nports = 1 << NUMPORTS_G(t4_read_reg(adap, MPS_CMN_CTL_A));
+
+	if (pidx >= nports) {
+		dev_warn(adap->pdev_dev, "TP Port Index %d >= Nports %d\n",
+			 pidx, nports);
+		return 0;
+	}
+
+	switch (chip_version) {
+	case CHELSIO_T4:
+	case CHELSIO_T5:
+		/* Note that this happens to be the same values as the MPS
+		 * Buffer Group Map for these Chips.  But we replicate the code
+		 * here because they're really separate concepts.
+		 */
+		switch (nports) {
+		case 1: return 0xf;
+		case 2: return 3 << (2 * pidx);
+		case 4: return 1 << pidx;
+		}
+		break;
+
+	case CHELSIO_T6:
+		switch (nports) {
+		case 1:
+		case 2: return 1 << pidx;
+		}
+		break;
+	}
+
+	dev_err(adap->pdev_dev, "Need TP Channel Map for Chip %0x, Nports %d\n",
+		chip_version, nports);
+	return 0;
 }
 
 /**
@@ -5473,6 +6303,7 @@ const char *t4_get_port_type_description(enum fw_port_type port_type)
 		"CR2_QSFP",
 		"SFP28",
 		"KR_SFP28",
+		"KR_XLAUI"
 	};
 
 	if (port_type < ARRAY_SIZE(port_type_description))
@@ -5546,10 +6377,8 @@ void t4_get_port_stats(struct adapter *adap, int idx, struct port_stats *p)
 	p->tx_ppp7             = GET_STAT(TX_PORT_PPP7);
 
 	if (CHELSIO_CHIP_VERSION(adap->params.chip) >= CHELSIO_T5) {
-		if (stat_ctl & COUNTPAUSESTATTX_F) {
-			p->tx_frames -= p->tx_pause;
-			p->tx_octets -= p->tx_pause * 64;
-		}
+		if (stat_ctl & COUNTPAUSESTATTX_F)
+			p->tx_frames_64 -= p->tx_pause;
 		if (stat_ctl & COUNTPAUSEMCTX_F)
 			p->tx_mcast_frames -= p->tx_pause;
 	}
@@ -5582,10 +6411,8 @@ void t4_get_port_stats(struct adapter *adap, int idx, struct port_stats *p)
 	p->rx_ppp7             = GET_STAT(RX_PORT_PPP7);
 
 	if (CHELSIO_CHIP_VERSION(adap->params.chip) >= CHELSIO_T5) {
-		if (stat_ctl & COUNTPAUSESTATRX_F) {
-			p->rx_frames -= p->rx_pause;
-			p->rx_octets -= p->rx_pause * 64;
-		}
+		if (stat_ctl & COUNTPAUSESTATRX_F)
+			p->rx_frames_64 -= p->rx_pause;
 		if (stat_ctl & COUNTPAUSEMCRX_F)
 			p->rx_mcast_frames -= p->rx_pause;
 	}
@@ -5739,7 +6566,7 @@ int t4_mdio_rd(struct adapter *adap, unsigned int mbox, unsigned int phy_addr,
  *	@phy_addr: the PHY address
  *	@mmd: the PHY MMD to access (0 for clause 22 PHYs)
  *	@reg: the register to write
- *	@valp: value to write
+ *	@val: value to write
  *
  *	Issues a FW command through the given mailbox to write a PHY register.
  */
@@ -5765,7 +6592,7 @@ int t4_mdio_wr(struct adapter *adap, unsigned int mbox, unsigned int phy_addr,
 
 /**
  *	t4_sge_decode_idma_state - decode the idma state
- *	@adap: the adapter
+ *	@adapter: the adapter
  *	@state: the state idma is stuck in
  */
 void t4_sge_decode_idma_state(struct adapter *adapter, int state)
@@ -5932,18 +6759,21 @@ void t4_sge_decode_idma_state(struct adapter *adapter, int state)
  *      t4_sge_ctxt_flush - flush the SGE context cache
  *      @adap: the adapter
  *      @mbox: mailbox to use for the FW command
+ *      @ctxt_type: Egress or Ingress
  *
  *      Issues a FW command through the given mailbox to flush the
  *      SGE context cache.
  */
-int t4_sge_ctxt_flush(struct adapter *adap, unsigned int mbox)
+int t4_sge_ctxt_flush(struct adapter *adap, unsigned int mbox, int ctxt_type)
 {
 	int ret;
 	u32 ldst_addrspace;
 	struct fw_ldst_cmd c;
 
 	memset(&c, 0, sizeof(c));
-	ldst_addrspace = FW_LDST_CMD_ADDRSPACE_V(FW_LDST_ADDRSPC_SGE_EGRC);
+	ldst_addrspace = FW_LDST_CMD_ADDRSPACE_V(ctxt_type == CTXT_EGRESS ?
+						 FW_LDST_ADDRSPC_SGE_EGRC :
+						 FW_LDST_ADDRSPC_SGE_INGC);
 	c.op_to_addrspace = cpu_to_be32(FW_CMD_OP_V(FW_LDST_CMD) |
 					FW_CMD_REQUEST_F | FW_CMD_READ_F |
 					ldst_addrspace);
@@ -5955,6 +6785,47 @@ int t4_sge_ctxt_flush(struct adapter *adap, unsigned int mbox)
 }
 
 /**
+ *	t4_read_sge_dbqtimers - read SGE Doorbell Queue Timer values
+ *	@adap: the adapter
+ *	@ndbqtimers: size of the provided SGE Doorbell Queue Timer table
+ *	@dbqtimers: SGE Doorbell Queue Timer table
+ *
+ *	Reads the SGE Doorbell Queue Timer values into the provided table.
+ *	Returns 0 on success (Firmware and Hardware support this feature),
+ *	an error on failure.
+ */
+int t4_read_sge_dbqtimers(struct adapter *adap, unsigned int ndbqtimers,
+			  u16 *dbqtimers)
+{
+	int ret, dbqtimerix;
+
+	ret = 0;
+	dbqtimerix = 0;
+	while (dbqtimerix < ndbqtimers) {
+		int nparams, param;
+		u32 params[7], vals[7];
+
+		nparams = ndbqtimers - dbqtimerix;
+		if (nparams > ARRAY_SIZE(params))
+			nparams = ARRAY_SIZE(params);
+
+		for (param = 0; param < nparams; param++)
+			params[param] =
+			  (FW_PARAMS_MNEM_V(FW_PARAMS_MNEM_DEV) |
+			   FW_PARAMS_PARAM_X_V(FW_PARAMS_PARAM_DEV_DBQ_TIMER) |
+			   FW_PARAMS_PARAM_Y_V(dbqtimerix + param));
+		ret = t4_query_params(adap, adap->mbox, adap->pf, 0,
+				      nparams, params, vals);
+		if (ret)
+			break;
+
+		for (param = 0; param < nparams; param++)
+			dbqtimers[dbqtimerix++] = vals[param];
+	}
+	return ret;
+}
+
+/**
  *      t4_fw_hello - establish communication with FW
  *      @adap: the adapter
  *      @mbox: mailbox to use for the FW command
@@ -6042,8 +6913,8 @@ retry:
 			waiting -= 50;
 
 			/*
-			 * If neither Error nor Initialialized are indicated
-			 * by the firmware keep waiting till we exaust our
+			 * If neither Error nor Initialized are indicated
+			 * by the firmware keep waiting till we exhaust our
 			 * timeout ... and then retry if we haven't exhausted
 			 * our retries ...
 			 */
@@ -6101,7 +6972,7 @@ int t4_fw_bye(struct adapter *adap, unsigned int mbox)
 }
 
 /**
- *	t4_init_cmd - ask FW to initialize the device
+ *	t4_early_init - ask FW to initialize the device
  *	@adap: the adapter
  *	@mbox: mailbox to use for the FW command
  *
@@ -6198,6 +7069,7 @@ static int t4_fw_halt(struct adapter *adap, unsigned int mbox, int force)
 /**
  *	t4_fw_restart - restart the firmware by taking the uP out of RESET
  *	@adap: the adapter
+ *	@mbox: mailbox to use for the FW command
  *	@reset: if we want to do a RESET to restart things
  *
  *	Restart firmware previously halted by t4_fw_halt().  On successful
@@ -6288,10 +7160,10 @@ int t4_fw_upgrade(struct adapter *adap, unsigned int mbox,
 	if (!t4_fw_matches_chip(adap, fw_hdr))
 		return -EINVAL;
 
-	/* Disable FW_OK flag so that mbox commands with FW_OK flag set
-	 * wont be sent when we are flashing FW.
+	/* Disable CXGB4_FW_OK flag so that mbox commands with CXGB4_FW_OK flag
+	 * set wont be sent when we are flashing FW.
 	 */
-	adap->flags &= ~FW_OK;
+	adap->flags &= ~CXGB4_FW_OK;
 
 	ret = t4_fw_halt(adap, mbox, force);
 	if (ret < 0 && !force)
@@ -6302,6 +7174,17 @@ int t4_fw_upgrade(struct adapter *adap, unsigned int mbox,
 		goto out;
 
 	/*
+	 * If there was a Firmware Configuration File stored in FLASH,
+	 * there's a good chance that it won't be compatible with the new
+	 * Firmware.  In order to prevent difficult to diagnose adapter
+	 * initialization issues, we clear out the Firmware Configuration File
+	 * portion of the FLASH .  The user will need to re-FLASH a new
+	 * Firmware Configuration File which is compatible with the new
+	 * Firmware if that's desired.
+	 */
+	(void)t4_load_cfg(adap, NULL, 0);
+
+	/*
 	 * Older versions of the firmware don't understand the new
 	 * PCIE_FW.HALT flag and so won't know to perform a RESET when they
 	 * restart.  So for newly loaded older firmware we'll have to do the
@@ -6319,7 +7202,7 @@ int t4_fw_upgrade(struct adapter *adap, unsigned int mbox,
 	 */
 	(void)t4_init_devlog_params(adap);
 out:
-	adap->flags |= FW_OK;
+	adap->flags |= CXGB4_FW_OK;
 	return ret;
 }
 
@@ -6344,7 +7227,7 @@ int t4_fl_pkt_align(struct adapter *adap)
 	 * separately.  The actual Ingress Packet Data alignment boundary
 	 * within Packed Buffer Mode is the maximum of these two
 	 * specifications.  (Note that it makes no real practical sense to
-	 * have the Pading Boudary be larger than the Packing Boundary but you
+	 * have the Padding Boundary be larger than the Packing Boundary but you
 	 * could set the chip up that way and, in fact, legacy T4 code would
 	 * end doing this because it would initialize the Padding Boundary and
 	 * leave the Packing Boundary initialized to 0 (16 bytes).)
@@ -6415,7 +7298,6 @@ int t4_fixup_host_params(struct adapter *adap, unsigned int page_size,
 	} else {
 		unsigned int pack_align;
 		unsigned int ingpad, ingpack;
-		unsigned int pcie_cap;
 
 		/* T5 introduced the separation of the Free List Padding and
 		 * Packing Boundaries.  Thus, we can select a smaller Padding
@@ -6440,8 +7322,7 @@ int t4_fixup_host_params(struct adapter *adap, unsigned int page_size,
 		 * multiple of the Maximum Payload Size.
 		 */
 		pack_align = fl_align;
-		pcie_cap = pci_find_capability(adap->pdev, PCI_CAP_ID_EXP);
-		if (pcie_cap) {
+		if (pci_is_pcie(adap->pdev)) {
 			unsigned int mps, mps_log;
 			u16 devctl;
 
@@ -6449,9 +7330,8 @@ int t4_fixup_host_params(struct adapter *adap, unsigned int page_size,
 			 * [bits 7:5] encodes sizes as powers of 2 starting at
 			 * 128 bytes.
 			 */
-			pci_read_config_word(adap->pdev,
-					     pcie_cap + PCI_EXP_DEVCTL,
-					     &devctl);
+			pcie_capability_read_word(adap->pdev, PCI_EXP_DEVCTL,
+						  &devctl);
 			mps_log = ((devctl & PCI_EXP_DEVCTL_PAYLOAD) >> 5) + 7;
 			mps = 1 << mps_log;
 			if (mps > pack_align)
@@ -6556,13 +7436,14 @@ int t4_fw_initialize(struct adapter *adap, unsigned int mbox)
  *	@params: the parameter names
  *	@val: the parameter values
  *	@rw: Write and read flag
+ *	@sleep_ok: if true, we may sleep awaiting mbox cmd completion
  *
  *	Reads the value of FW or device parameters.  Up to 7 parameters can be
  *	queried at once.
  */
 int t4_query_params_rw(struct adapter *adap, unsigned int mbox, unsigned int pf,
 		       unsigned int vf, unsigned int nparams, const u32 *params,
-		       u32 *val, int rw)
+		       u32 *val, int rw, bool sleep_ok)
 {
 	int i, ret;
 	struct fw_params_cmd c;
@@ -6585,7 +7466,7 @@ int t4_query_params_rw(struct adapter *adap, unsigned int mbox, unsigned int pf,
 		p++;
 	}
 
-	ret = t4_wr_mbox(adap, mbox, &c, sizeof(c), &c);
+	ret = t4_wr_mbox_meat(adap, mbox, &c, sizeof(c), &c, sleep_ok);
 	if (ret == 0)
 		for (i = 0, p = &c.param[0].val; i < nparams; i++, p += 2)
 			*val++ = be32_to_cpu(*p);
@@ -6596,7 +7477,16 @@ int t4_query_params(struct adapter *adap, unsigned int mbox, unsigned int pf,
 		    unsigned int vf, unsigned int nparams, const u32 *params,
 		    u32 *val)
 {
-	return t4_query_params_rw(adap, mbox, pf, vf, nparams, params, val, 0);
+	return t4_query_params_rw(adap, mbox, pf, vf, nparams, params, val, 0,
+				  true);
+}
+
+int t4_query_params_ns(struct adapter *adap, unsigned int mbox, unsigned int pf,
+		       unsigned int vf, unsigned int nparams, const u32 *params,
+		       u32 *val)
+{
+	return t4_query_params_rw(adap, mbox, pf, vf, nparams, params, val, 0,
+				  false);
 }
 
 /**
@@ -6718,6 +7608,8 @@ int t4_cfg_pfvf(struct adapter *adap, unsigned int mbox, unsigned int pf,
  *	@nmac: number of MAC addresses needed (1 to 5)
  *	@mac: the MAC addresses of the VI
  *	@rss_size: size of RSS table slice associated with this VI
+ *	@vivld: the destination to store the VI Valid value.
+ *	@vin: the destination to store the VIN value.
  *
  *	Allocates a virtual interface for the given physical port.  If @mac is
  *	not %NULL it contains the MAC addresses of the VI as assigned by FW.
@@ -6727,7 +7619,7 @@ int t4_cfg_pfvf(struct adapter *adap, unsigned int mbox, unsigned int pf,
  */
 int t4_alloc_vi(struct adapter *adap, unsigned int mbox, unsigned int port,
 		unsigned int pf, unsigned int vf, unsigned int nmac, u8 *mac,
-		unsigned int *rss_size)
+		unsigned int *rss_size, u8 *vivld, u8 *vin)
 {
 	int ret;
 	struct fw_vi_cmd c;
@@ -6749,16 +7641,26 @@ int t4_alloc_vi(struct adapter *adap, unsigned int mbox, unsigned int port,
 		switch (nmac) {
 		case 5:
 			memcpy(mac + 24, c.nmac3, sizeof(c.nmac3));
+			fallthrough;
 		case 4:
 			memcpy(mac + 18, c.nmac2, sizeof(c.nmac2));
+			fallthrough;
 		case 3:
 			memcpy(mac + 12, c.nmac1, sizeof(c.nmac1));
+			fallthrough;
 		case 2:
 			memcpy(mac + 6,  c.nmac0, sizeof(c.nmac0));
 		}
 	}
 	if (rss_size)
 		*rss_size = FW_VI_CMD_RSSSIZE_G(be16_to_cpu(c.rsssize_pkd));
+
+	if (vivld)
+		*vivld = FW_VI_CMD_VFVLD_G(be32_to_cpu(c.alloc_to_len16));
+
+	if (vin)
+		*vin = FW_VI_CMD_VIN_G(be32_to_cpu(c.alloc_to_len16));
+
 	return FW_VI_CMD_VIID_G(be16_to_cpu(c.type_viid));
 }
 
@@ -6794,6 +7696,7 @@ int t4_free_vi(struct adapter *adap, unsigned int mbox, unsigned int pf,
  *	@adap: the adapter
  *	@mbox: mailbox to use for the FW command
  *	@viid: the VI id
+ *	@viid_mirror: the mirror VI id
  *	@mtu: the new MTU or -1
  *	@promisc: 1 to enable promiscuous mode, 0 to disable it, -1 no change
  *	@all_multi: 1 to enable all-multi mode, 0 to disable it, -1 no change
@@ -6804,10 +7707,11 @@ int t4_free_vi(struct adapter *adap, unsigned int mbox, unsigned int pf,
  *	Sets Rx properties of a virtual interface.
  */
 int t4_set_rxmode(struct adapter *adap, unsigned int mbox, unsigned int viid,
-		  int mtu, int promisc, int all_multi, int bcast, int vlanex,
-		  bool sleep_ok)
+		  unsigned int viid_mirror, int mtu, int promisc, int all_multi,
+		  int bcast, int vlanex, bool sleep_ok)
 {
-	struct fw_vi_rxmode_cmd c;
+	struct fw_vi_rxmode_cmd c, c_mirror;
+	int ret;
 
 	/* convert to FW values */
 	if (mtu < 0)
@@ -6832,7 +7736,215 @@ int t4_set_rxmode(struct adapter *adap, unsigned int mbox, unsigned int viid,
 			    FW_VI_RXMODE_CMD_ALLMULTIEN_V(all_multi) |
 			    FW_VI_RXMODE_CMD_BROADCASTEN_V(bcast) |
 			    FW_VI_RXMODE_CMD_VLANEXEN_V(vlanex));
-	return t4_wr_mbox_meat(adap, mbox, &c, sizeof(c), NULL, sleep_ok);
+
+	if (viid_mirror) {
+		memcpy(&c_mirror, &c, sizeof(c_mirror));
+		c_mirror.op_to_viid =
+			cpu_to_be32(FW_CMD_OP_V(FW_VI_RXMODE_CMD) |
+				    FW_CMD_REQUEST_F | FW_CMD_WRITE_F |
+				    FW_VI_RXMODE_CMD_VIID_V(viid_mirror));
+	}
+
+	ret = t4_wr_mbox_meat(adap, mbox, &c, sizeof(c), NULL, sleep_ok);
+	if (ret)
+		return ret;
+
+	if (viid_mirror)
+		ret = t4_wr_mbox_meat(adap, mbox, &c_mirror, sizeof(c_mirror),
+				      NULL, sleep_ok);
+
+	return ret;
+}
+
+/**
+ *      t4_free_encap_mac_filt - frees MPS entry at given index
+ *      @adap: the adapter
+ *      @viid: the VI id
+ *      @idx: index of MPS entry to be freed
+ *      @sleep_ok: call is allowed to sleep
+ *
+ *      Frees the MPS entry at supplied index
+ *
+ *      Returns a negative error number or zero on success
+ */
+int t4_free_encap_mac_filt(struct adapter *adap, unsigned int viid,
+			   int idx, bool sleep_ok)
+{
+	struct fw_vi_mac_exact *p;
+	struct fw_vi_mac_cmd c;
+	int ret = 0;
+	u32 exact;
+
+	memset(&c, 0, sizeof(c));
+	c.op_to_viid = cpu_to_be32(FW_CMD_OP_V(FW_VI_MAC_CMD) |
+				   FW_CMD_REQUEST_F | FW_CMD_WRITE_F |
+				   FW_CMD_EXEC_V(0) |
+				   FW_VI_MAC_CMD_VIID_V(viid));
+	exact = FW_VI_MAC_CMD_ENTRY_TYPE_V(FW_VI_MAC_TYPE_EXACTMAC);
+	c.freemacs_to_len16 = cpu_to_be32(FW_VI_MAC_CMD_FREEMACS_V(0) |
+					  exact |
+					  FW_CMD_LEN16_V(1));
+	p = c.u.exact;
+	p->valid_to_idx = cpu_to_be16(FW_VI_MAC_CMD_VALID_F |
+				      FW_VI_MAC_CMD_IDX_V(idx));
+	eth_zero_addr(p->macaddr);
+	ret = t4_wr_mbox_meat(adap, adap->mbox, &c, sizeof(c), &c, sleep_ok);
+	return ret;
+}
+
+/**
+ *	t4_free_raw_mac_filt - Frees a raw mac entry in mps tcam
+ *	@adap: the adapter
+ *	@viid: the VI id
+ *	@addr: the MAC address
+ *	@mask: the mask
+ *	@idx: index of the entry in mps tcam
+ *	@lookup_type: MAC address for inner (1) or outer (0) header
+ *	@port_id: the port index
+ *	@sleep_ok: call is allowed to sleep
+ *
+ *	Removes the mac entry at the specified index using raw mac interface.
+ *
+ *	Returns a negative error number on failure.
+ */
+int t4_free_raw_mac_filt(struct adapter *adap, unsigned int viid,
+			 const u8 *addr, const u8 *mask, unsigned int idx,
+			 u8 lookup_type, u8 port_id, bool sleep_ok)
+{
+	struct fw_vi_mac_cmd c;
+	struct fw_vi_mac_raw *p = &c.u.raw;
+	u32 val;
+
+	memset(&c, 0, sizeof(c));
+	c.op_to_viid = cpu_to_be32(FW_CMD_OP_V(FW_VI_MAC_CMD) |
+				   FW_CMD_REQUEST_F | FW_CMD_WRITE_F |
+				   FW_CMD_EXEC_V(0) |
+				   FW_VI_MAC_CMD_VIID_V(viid));
+	val = FW_CMD_LEN16_V(1) |
+	      FW_VI_MAC_CMD_ENTRY_TYPE_V(FW_VI_MAC_TYPE_RAW);
+	c.freemacs_to_len16 = cpu_to_be32(FW_VI_MAC_CMD_FREEMACS_V(0) |
+					  FW_CMD_LEN16_V(val));
+
+	p->raw_idx_pkd = cpu_to_be32(FW_VI_MAC_CMD_RAW_IDX_V(idx) |
+				     FW_VI_MAC_ID_BASED_FREE);
+
+	/* Lookup Type. Outer header: 0, Inner header: 1 */
+	p->data0_pkd = cpu_to_be32(DATALKPTYPE_V(lookup_type) |
+				   DATAPORTNUM_V(port_id));
+	/* Lookup mask and port mask */
+	p->data0m_pkd = cpu_to_be64(DATALKPTYPE_V(DATALKPTYPE_M) |
+				    DATAPORTNUM_V(DATAPORTNUM_M));
+
+	/* Copy the address and the mask */
+	memcpy((u8 *)&p->data1[0] + 2, addr, ETH_ALEN);
+	memcpy((u8 *)&p->data1m[0] + 2, mask, ETH_ALEN);
+
+	return t4_wr_mbox_meat(adap, adap->mbox, &c, sizeof(c), &c, sleep_ok);
+}
+
+/**
+ *      t4_alloc_encap_mac_filt - Adds a mac entry in mps tcam with VNI support
+ *      @adap: the adapter
+ *      @viid: the VI id
+ *      @addr: the MAC address
+ *      @mask: the mask
+ *      @vni: the VNI id for the tunnel protocol
+ *      @vni_mask: mask for the VNI id
+ *      @dip_hit: to enable DIP match for the MPS entry
+ *      @lookup_type: MAC address for inner (1) or outer (0) header
+ *      @sleep_ok: call is allowed to sleep
+ *
+ *      Allocates an MPS entry with specified MAC address and VNI value.
+ *
+ *      Returns a negative error number or the allocated index for this mac.
+ */
+int t4_alloc_encap_mac_filt(struct adapter *adap, unsigned int viid,
+			    const u8 *addr, const u8 *mask, unsigned int vni,
+			    unsigned int vni_mask, u8 dip_hit, u8 lookup_type,
+			    bool sleep_ok)
+{
+	struct fw_vi_mac_cmd c;
+	struct fw_vi_mac_vni *p = c.u.exact_vni;
+	int ret = 0;
+	u32 val;
+
+	memset(&c, 0, sizeof(c));
+	c.op_to_viid = cpu_to_be32(FW_CMD_OP_V(FW_VI_MAC_CMD) |
+				   FW_CMD_REQUEST_F | FW_CMD_WRITE_F |
+				   FW_VI_MAC_CMD_VIID_V(viid));
+	val = FW_CMD_LEN16_V(1) |
+	      FW_VI_MAC_CMD_ENTRY_TYPE_V(FW_VI_MAC_TYPE_EXACTMAC_VNI);
+	c.freemacs_to_len16 = cpu_to_be32(val);
+	p->valid_to_idx = cpu_to_be16(FW_VI_MAC_CMD_VALID_F |
+				      FW_VI_MAC_CMD_IDX_V(FW_VI_MAC_ADD_MAC));
+	memcpy(p->macaddr, addr, sizeof(p->macaddr));
+	memcpy(p->macaddr_mask, mask, sizeof(p->macaddr_mask));
+
+	p->lookup_type_to_vni =
+		cpu_to_be32(FW_VI_MAC_CMD_VNI_V(vni) |
+			    FW_VI_MAC_CMD_DIP_HIT_V(dip_hit) |
+			    FW_VI_MAC_CMD_LOOKUP_TYPE_V(lookup_type));
+	p->vni_mask_pkd = cpu_to_be32(FW_VI_MAC_CMD_VNI_MASK_V(vni_mask));
+	ret = t4_wr_mbox_meat(adap, adap->mbox, &c, sizeof(c), &c, sleep_ok);
+	if (ret == 0)
+		ret = FW_VI_MAC_CMD_IDX_G(be16_to_cpu(p->valid_to_idx));
+	return ret;
+}
+
+/**
+ *	t4_alloc_raw_mac_filt - Adds a mac entry in mps tcam
+ *	@adap: the adapter
+ *	@viid: the VI id
+ *	@addr: the MAC address
+ *	@mask: the mask
+ *	@idx: index at which to add this entry
+ *	@lookup_type: MAC address for inner (1) or outer (0) header
+ *	@port_id: the port index
+ *	@sleep_ok: call is allowed to sleep
+ *
+ *	Adds the mac entry at the specified index using raw mac interface.
+ *
+ *	Returns a negative error number or the allocated index for this mac.
+ */
+int t4_alloc_raw_mac_filt(struct adapter *adap, unsigned int viid,
+			  const u8 *addr, const u8 *mask, unsigned int idx,
+			  u8 lookup_type, u8 port_id, bool sleep_ok)
+{
+	int ret = 0;
+	struct fw_vi_mac_cmd c;
+	struct fw_vi_mac_raw *p = &c.u.raw;
+	u32 val;
+
+	memset(&c, 0, sizeof(c));
+	c.op_to_viid = cpu_to_be32(FW_CMD_OP_V(FW_VI_MAC_CMD) |
+				   FW_CMD_REQUEST_F | FW_CMD_WRITE_F |
+				   FW_VI_MAC_CMD_VIID_V(viid));
+	val = FW_CMD_LEN16_V(1) |
+	      FW_VI_MAC_CMD_ENTRY_TYPE_V(FW_VI_MAC_TYPE_RAW);
+	c.freemacs_to_len16 = cpu_to_be32(val);
+
+	/* Specify that this is an inner mac address */
+	p->raw_idx_pkd = cpu_to_be32(FW_VI_MAC_CMD_RAW_IDX_V(idx));
+
+	/* Lookup Type. Outer header: 0, Inner header: 1 */
+	p->data0_pkd = cpu_to_be32(DATALKPTYPE_V(lookup_type) |
+				   DATAPORTNUM_V(port_id));
+	/* Lookup mask and port mask */
+	p->data0m_pkd = cpu_to_be64(DATALKPTYPE_V(DATALKPTYPE_M) |
+				    DATAPORTNUM_V(DATAPORTNUM_M));
+
+	/* Copy the address and the mask */
+	memcpy((u8 *)&p->data1[0] + 2, addr, ETH_ALEN);
+	memcpy((u8 *)&p->data1m[0] + 2, mask, ETH_ALEN);
+
+	ret = t4_wr_mbox_meat(adap, adap->mbox, &c, sizeof(c), &c, sleep_ok);
+	if (ret == 0) {
+		ret = FW_VI_MAC_CMD_RAW_IDX_G(be32_to_cpu(p->raw_idx_pkd));
+		if (ret != idx)
+			ret = -ENOMEM;
+	}
+
+	return ret;
 }
 
 /**
@@ -7012,7 +8124,7 @@ int t4_free_mac_filt(struct adapter *adap, unsigned int mbox,
  *	@idx: index of existing filter for old value of MAC address, or -1
  *	@addr: the new MAC address value
  *	@persist: whether a new MAC allocation should be persistent
- *	@add_smt: if true also add the address to the HW SMT
+ *	@smt_idx: the destination to store the new SMT index.
  *
  *	Modifies an exact-match filter and sets it to the new MAC address.
  *	Note that in general it is not possible to modify the value of a given
@@ -7024,7 +8136,7 @@ int t4_free_mac_filt(struct adapter *adap, unsigned int mbox,
  *	MAC value.
  */
 int t4_change_mac(struct adapter *adap, unsigned int mbox, unsigned int viid,
-		  int idx, const u8 *addr, bool persist, bool add_smt)
+		  int idx, const u8 *addr, bool persist, u8 *smt_idx)
 {
 	int ret, mode;
 	struct fw_vi_mac_cmd c;
@@ -7033,7 +8145,7 @@ int t4_change_mac(struct adapter *adap, unsigned int mbox, unsigned int viid,
 
 	if (idx < 0)                             /* new allocation */
 		idx = persist ? FW_VI_MAC_ADD_PERSIST_MAC : FW_VI_MAC_ADD_MAC;
-	mode = add_smt ? FW_VI_MAC_SMT_AND_MPSTCAM : FW_VI_MAC_MPS_TCAM_ENTRY;
+	mode = smt_idx ? FW_VI_MAC_SMT_AND_MPSTCAM : FW_VI_MAC_MPS_TCAM_ENTRY;
 
 	memset(&c, 0, sizeof(c));
 	c.op_to_viid = cpu_to_be32(FW_CMD_OP_V(FW_VI_MAC_CMD) |
@@ -7050,6 +8162,23 @@ int t4_change_mac(struct adapter *adap, unsigned int mbox, unsigned int viid,
 		ret = FW_VI_MAC_CMD_IDX_G(be16_to_cpu(p->valid_to_idx));
 		if (ret >= max_mac_addr)
 			ret = -ENOMEM;
+		if (smt_idx) {
+			if (adap->params.viid_smt_extn_support) {
+				*smt_idx = FW_VI_MAC_CMD_SMTID_G
+						    (be32_to_cpu(c.op_to_viid));
+			} else {
+				/* In T4/T5, SMT contains 256 SMAC entries
+				 * organized in 128 rows of 2 entries each.
+				 * In T6, SMT contains 256 SMAC entries in
+				 * 256 rows.
+				 */
+				if (CHELSIO_CHIP_VERSION(adap->params.chip) <=
+								     CHELSIO_T5)
+					*smt_idx = (viid & FW_VIID_VIN_M) << 1;
+				else
+					*smt_idx = (viid & FW_VIID_VIN_M);
+			}
+		}
 	}
 	return ret;
 }
@@ -7126,6 +8255,34 @@ int t4_enable_vi(struct adapter *adap, unsigned int mbox, unsigned int viid,
 }
 
 /**
+ *	t4_enable_pi_params - enable/disable a Port's Virtual Interface
+ *      @adap: the adapter
+ *      @mbox: mailbox to use for the FW command
+ *      @pi: the Port Information structure
+ *      @rx_en: 1=enable Rx, 0=disable Rx
+ *      @tx_en: 1=enable Tx, 0=disable Tx
+ *      @dcb_en: 1=enable delivery of Data Center Bridging messages.
+ *
+ *      Enables/disables a Port's Virtual Interface.  Note that setting DCB
+ *	Enable only makes sense when enabling a Virtual Interface ...
+ *	If the Virtual Interface enable/disable operation is successful,
+ *	we notify the OS-specific code of a potential Link Status change
+ *	via the OS Contract API t4_os_link_changed().
+ */
+int t4_enable_pi_params(struct adapter *adap, unsigned int mbox,
+			struct port_info *pi,
+			bool rx_en, bool tx_en, bool dcb_en)
+{
+	int ret = t4_enable_vi_params(adap, mbox, pi->viid,
+				      rx_en, tx_en, dcb_en);
+	if (ret)
+		return ret;
+	t4_os_link_changed(adap, pi->port_id,
+			   rx_en && tx_en && pi->link_cfg.link_ok);
+	return 0;
+}
+
+/**
  *	t4_identify_port - identify a VI's port by blinking its LED
  *	@adap: the adapter
  *	@mbox: mailbox to use for the FW command
@@ -7289,7 +8446,6 @@ int t4_ofld_eq_free(struct adapter *adap, unsigned int mbox, unsigned int pf,
 
 /**
  *	t4_link_down_rc_str - return a string for a Link Down Reason Code
- *	@adap: the adapter
  *	@link_down_rc: Link Down Reason Code
  *
  *	Returns a string representation of the Link Down Reason Code.
@@ -7313,6 +8469,96 @@ static const char *t4_link_down_rc_str(unsigned char link_down_rc)
 	return reason[link_down_rc];
 }
 
+/* Return the highest speed set in the port capabilities, in Mb/s. */
+static unsigned int fwcap_to_speed(fw_port_cap32_t caps)
+{
+	#define TEST_SPEED_RETURN(__caps_speed, __speed) \
+		do { \
+			if (caps & FW_PORT_CAP32_SPEED_##__caps_speed) \
+				return __speed; \
+		} while (0)
+
+	TEST_SPEED_RETURN(400G, 400000);
+	TEST_SPEED_RETURN(200G, 200000);
+	TEST_SPEED_RETURN(100G, 100000);
+	TEST_SPEED_RETURN(50G,   50000);
+	TEST_SPEED_RETURN(40G,   40000);
+	TEST_SPEED_RETURN(25G,   25000);
+	TEST_SPEED_RETURN(10G,   10000);
+	TEST_SPEED_RETURN(1G,     1000);
+	TEST_SPEED_RETURN(100M,    100);
+
+	#undef TEST_SPEED_RETURN
+
+	return 0;
+}
+
+/**
+ *	fwcap_to_fwspeed - return highest speed in Port Capabilities
+ *	@acaps: advertised Port Capabilities
+ *
+ *	Get the highest speed for the port from the advertised Port
+ *	Capabilities.  It will be either the highest speed from the list of
+ *	speeds or whatever user has set using ethtool.
+ */
+static fw_port_cap32_t fwcap_to_fwspeed(fw_port_cap32_t acaps)
+{
+	#define TEST_SPEED_RETURN(__caps_speed) \
+		do { \
+			if (acaps & FW_PORT_CAP32_SPEED_##__caps_speed) \
+				return FW_PORT_CAP32_SPEED_##__caps_speed; \
+		} while (0)
+
+	TEST_SPEED_RETURN(400G);
+	TEST_SPEED_RETURN(200G);
+	TEST_SPEED_RETURN(100G);
+	TEST_SPEED_RETURN(50G);
+	TEST_SPEED_RETURN(40G);
+	TEST_SPEED_RETURN(25G);
+	TEST_SPEED_RETURN(10G);
+	TEST_SPEED_RETURN(1G);
+	TEST_SPEED_RETURN(100M);
+
+	#undef TEST_SPEED_RETURN
+
+	return 0;
+}
+
+/**
+ *	lstatus_to_fwcap - translate old lstatus to 32-bit Port Capabilities
+ *	@lstatus: old FW_PORT_ACTION_GET_PORT_INFO lstatus value
+ *
+ *	Translates old FW_PORT_ACTION_GET_PORT_INFO lstatus field into new
+ *	32-bit Port Capabilities value.
+ */
+static fw_port_cap32_t lstatus_to_fwcap(u32 lstatus)
+{
+	fw_port_cap32_t linkattr = 0;
+
+	/* Unfortunately the format of the Link Status in the old
+	 * 16-bit Port Information message isn't the same as the
+	 * 16-bit Port Capabilities bitfield used everywhere else ...
+	 */
+	if (lstatus & FW_PORT_CMD_RXPAUSE_F)
+		linkattr |= FW_PORT_CAP32_FC_RX;
+	if (lstatus & FW_PORT_CMD_TXPAUSE_F)
+		linkattr |= FW_PORT_CAP32_FC_TX;
+	if (lstatus & FW_PORT_CMD_LSPEED_V(FW_PORT_CAP_SPEED_100M))
+		linkattr |= FW_PORT_CAP32_SPEED_100M;
+	if (lstatus & FW_PORT_CMD_LSPEED_V(FW_PORT_CAP_SPEED_1G))
+		linkattr |= FW_PORT_CAP32_SPEED_1G;
+	if (lstatus & FW_PORT_CMD_LSPEED_V(FW_PORT_CAP_SPEED_10G))
+		linkattr |= FW_PORT_CAP32_SPEED_10G;
+	if (lstatus & FW_PORT_CMD_LSPEED_V(FW_PORT_CAP_SPEED_25G))
+		linkattr |= FW_PORT_CAP32_SPEED_25G;
+	if (lstatus & FW_PORT_CMD_LSPEED_V(FW_PORT_CAP_SPEED_40G))
+		linkattr |= FW_PORT_CAP32_SPEED_40G;
+	if (lstatus & FW_PORT_CMD_LSPEED_V(FW_PORT_CAP_SPEED_100G))
+		linkattr |= FW_PORT_CAP32_SPEED_100G;
+
+	return linkattr;
+}
+
 /**
  *	t4_handle_get_port_info - process a FW reply message
  *	@pi: the port info
@@ -7322,57 +8568,176 @@ static const char *t4_link_down_rc_str(unsigned char link_down_rc)
  */
 void t4_handle_get_port_info(struct port_info *pi, const __be64 *rpl)
 {
-	const struct fw_port_cmd *p = (const void *)rpl;
-	struct adapter *adap = pi->adapter;
-
-	/* link/module state change message */
-	int speed = 0, fc = 0;
-	struct link_config *lc;
-	u32 stat = be32_to_cpu(p->u.info.lstatus_to_modtype);
-	int link_ok = (stat & FW_PORT_CMD_LSTATUS_F) != 0;
-	u32 mod = FW_PORT_CMD_MODTYPE_G(stat);
-
-	if (stat & FW_PORT_CMD_RXPAUSE_F)
-		fc |= PAUSE_RX;
-	if (stat & FW_PORT_CMD_TXPAUSE_F)
-		fc |= PAUSE_TX;
-	if (stat & FW_PORT_CMD_LSPEED_V(FW_PORT_CAP_SPEED_100M))
-		speed = 100;
-	else if (stat & FW_PORT_CMD_LSPEED_V(FW_PORT_CAP_SPEED_1G))
-		speed = 1000;
-	else if (stat & FW_PORT_CMD_LSPEED_V(FW_PORT_CAP_SPEED_10G))
-		speed = 10000;
-	else if (stat & FW_PORT_CMD_LSPEED_V(FW_PORT_CAP_SPEED_25G))
-		speed = 25000;
-	else if (stat & FW_PORT_CMD_LSPEED_V(FW_PORT_CAP_SPEED_40G))
-		speed = 40000;
-	else if (stat & FW_PORT_CMD_LSPEED_V(FW_PORT_CAP_SPEED_100G))
-		speed = 100000;
-
-	lc = &pi->link_cfg;
-
-	if (mod != pi->mod_type) {
-		pi->mod_type = mod;
-		t4_os_portmod_changed(adap, pi->port_id);
+	const struct fw_port_cmd *cmd = (const void *)rpl;
+	fw_port_cap32_t pcaps, acaps, lpacaps, linkattr;
+	struct link_config *lc = &pi->link_cfg;
+	struct adapter *adapter = pi->adapter;
+	unsigned int speed, fc, fec, adv_fc;
+	enum fw_port_module_type mod_type;
+	int action, link_ok, linkdnrc;
+	enum fw_port_type port_type;
+
+	/* Extract the various fields from the Port Information message.
+	 */
+	action = FW_PORT_CMD_ACTION_G(be32_to_cpu(cmd->action_to_len16));
+	switch (action) {
+	case FW_PORT_ACTION_GET_PORT_INFO: {
+		u32 lstatus = be32_to_cpu(cmd->u.info.lstatus_to_modtype);
+
+		link_ok = (lstatus & FW_PORT_CMD_LSTATUS_F) != 0;
+		linkdnrc = FW_PORT_CMD_LINKDNRC_G(lstatus);
+		port_type = FW_PORT_CMD_PTYPE_G(lstatus);
+		mod_type = FW_PORT_CMD_MODTYPE_G(lstatus);
+		pcaps = fwcaps16_to_caps32(be16_to_cpu(cmd->u.info.pcap));
+		acaps = fwcaps16_to_caps32(be16_to_cpu(cmd->u.info.acap));
+		lpacaps = fwcaps16_to_caps32(be16_to_cpu(cmd->u.info.lpacap));
+		linkattr = lstatus_to_fwcap(lstatus);
+		break;
+	}
+
+	case FW_PORT_ACTION_GET_PORT_INFO32: {
+		u32 lstatus32;
+
+		lstatus32 = be32_to_cpu(cmd->u.info32.lstatus32_to_cbllen32);
+		link_ok = (lstatus32 & FW_PORT_CMD_LSTATUS32_F) != 0;
+		linkdnrc = FW_PORT_CMD_LINKDNRC32_G(lstatus32);
+		port_type = FW_PORT_CMD_PORTTYPE32_G(lstatus32);
+		mod_type = FW_PORT_CMD_MODTYPE32_G(lstatus32);
+		pcaps = be32_to_cpu(cmd->u.info32.pcaps32);
+		acaps = be32_to_cpu(cmd->u.info32.acaps32);
+		lpacaps = be32_to_cpu(cmd->u.info32.lpacaps32);
+		linkattr = be32_to_cpu(cmd->u.info32.linkattr32);
+		break;
 	}
+
+	default:
+		dev_err(adapter->pdev_dev, "Handle Port Information: Bad Command/Action %#x\n",
+			be32_to_cpu(cmd->action_to_len16));
+		return;
+	}
+
+	fec = fwcap_to_cc_fec(acaps);
+	adv_fc = fwcap_to_cc_pause(acaps);
+	fc = fwcap_to_cc_pause(linkattr);
+	speed = fwcap_to_speed(linkattr);
+
+	/* Reset state for communicating new Transceiver Module status and
+	 * whether the OS-dependent layer wants us to redo the current
+	 * "sticky" L1 Configure Link Parameters.
+	 */
+	lc->new_module = false;
+	lc->redo_l1cfg = false;
+
+	if (mod_type != pi->mod_type) {
+		/* With the newer SFP28 and QSFP28 Transceiver Module Types,
+		 * various fundamental Port Capabilities which used to be
+		 * immutable can now change radically.  We can now have
+		 * Speeds, Auto-Negotiation, Forward Error Correction, etc.
+		 * all change based on what Transceiver Module is inserted.
+		 * So we need to record the Physical "Port" Capabilities on
+		 * every Transceiver Module change.
+		 */
+		lc->pcaps = pcaps;
+
+		/* When a new Transceiver Module is inserted, the Firmware
+		 * will examine its i2c EPROM to determine its type and
+		 * general operating parameters including things like Forward
+		 * Error Control, etc.  Various IEEE 802.3 standards dictate
+		 * how to interpret these i2c values to determine default
+		 * "sutomatic" settings.  We record these for future use when
+		 * the user explicitly requests these standards-based values.
+		 */
+		lc->def_acaps = acaps;
+
+		/* Some versions of the early T6 Firmware "cheated" when
+		 * handling different Transceiver Modules by changing the
+		 * underlaying Port Type reported to the Host Drivers.  As
+		 * such we need to capture whatever Port Type the Firmware
+		 * sends us and record it in case it's different from what we
+		 * were told earlier.  Unfortunately, since Firmware is
+		 * forever, we'll need to keep this code here forever, but in
+		 * later T6 Firmware it should just be an assignment of the
+		 * same value already recorded.
+		 */
+		pi->port_type = port_type;
+
+		/* Record new Module Type information.
+		 */
+		pi->mod_type = mod_type;
+
+		/* Let the OS-dependent layer know if we have a new
+		 * Transceiver Module inserted.
+		 */
+		lc->new_module = t4_is_inserted_mod_type(mod_type);
+
+		t4_os_portmod_changed(adapter, pi->port_id);
+	}
+
 	if (link_ok != lc->link_ok || speed != lc->speed ||
-	    fc != lc->fc) {	/* something changed */
+	    fc != lc->fc || adv_fc != lc->advertised_fc ||
+	    fec != lc->fec) {
+		/* something changed */
 		if (!link_ok && lc->link_ok) {
-			unsigned char rc = FW_PORT_CMD_LINKDNRC_G(stat);
-
-			lc->link_down_rc = rc;
-			dev_warn(adap->pdev_dev,
-				 "Port %d link down, reason: %s\n",
-				 pi->port_id, t4_link_down_rc_str(rc));
+			lc->link_down_rc = linkdnrc;
+			dev_warn_ratelimited(adapter->pdev_dev,
+					     "Port %d link down, reason: %s\n",
+					     pi->tx_chan,
+					     t4_link_down_rc_str(linkdnrc));
 		}
 		lc->link_ok = link_ok;
 		lc->speed = speed;
+		lc->advertised_fc = adv_fc;
 		lc->fc = fc;
-		lc->supported = be16_to_cpu(p->u.info.pcap);
-		lc->lp_advertising = be16_to_cpu(p->u.info.lpacap);
+		lc->fec = fec;
+
+		lc->lpacaps = lpacaps;
+		lc->acaps = acaps & ADVERT_MASK;
 
-		t4_os_link_changed(adap, pi->port_id, link_ok);
+		/* If we're not physically capable of Auto-Negotiation, note
+		 * this as Auto-Negotiation disabled.  Otherwise, we track
+		 * what Auto-Negotiation settings we have.  Note parallel
+		 * structure in t4_link_l1cfg_core() and init_link_config().
+		 */
+		if (!(lc->acaps & FW_PORT_CAP32_ANEG)) {
+			lc->autoneg = AUTONEG_DISABLE;
+		} else if (lc->acaps & FW_PORT_CAP32_ANEG) {
+			lc->autoneg = AUTONEG_ENABLE;
+		} else {
+			/* When Autoneg is disabled, user needs to set
+			 * single speed.
+			 * Similar to cxgb4_ethtool.c: set_link_ksettings
+			 */
+			lc->acaps = 0;
+			lc->speed_caps = fwcap_to_fwspeed(acaps);
+			lc->autoneg = AUTONEG_DISABLE;
+		}
+
+		t4_os_link_changed(adapter, pi->port_id, link_ok);
+	}
+
+	/* If we have a new Transceiver Module and the OS-dependent code has
+	 * told us that it wants us to redo whatever "sticky" L1 Configuration
+	 * Link Parameters are set, do that now.
+	 */
+	if (lc->new_module && lc->redo_l1cfg) {
+		struct link_config old_lc;
+		int ret;
+
+		/* Save the current L1 Configuration and restore it if an
+		 * error occurs.  We probably should fix the l1_cfg*()
+		 * routines not to change the link_config when an error
+		 * occurs ...
+		 */
+		old_lc = *lc;
+		ret = t4_link_l1cfg_ns(adapter, adapter->mbox, pi->lport, lc);
+		if (ret) {
+			*lc = old_lc;
+			dev_warn(adapter->pdev_dev,
+				 "Attempt to update new Transceiver Module settings failed\n");
+		}
 	}
+	lc->new_module = false;
+	lc->redo_l1cfg = false;
 }
 
 /**
@@ -7385,15 +8750,18 @@ void t4_handle_get_port_info(struct port_info *pi, const __be64 *rpl)
  */
 int t4_update_port_info(struct port_info *pi)
 {
+	unsigned int fw_caps = pi->adapter->params.fw_caps_support;
 	struct fw_port_cmd port_cmd;
 	int ret;
 
 	memset(&port_cmd, 0, sizeof(port_cmd));
 	port_cmd.op_to_portid = cpu_to_be32(FW_CMD_OP_V(FW_PORT_CMD) |
 					    FW_CMD_REQUEST_F | FW_CMD_READ_F |
-					    FW_PORT_CMD_PORTID_V(pi->port_id));
+					    FW_PORT_CMD_PORTID_V(pi->tx_chan));
 	port_cmd.action_to_len16 = cpu_to_be32(
-		FW_PORT_CMD_ACTION_V(FW_PORT_ACTION_GET_PORT_INFO) |
+		FW_PORT_CMD_ACTION_V(fw_caps == FW_CAPS16
+				     ? FW_PORT_ACTION_GET_PORT_INFO
+				     : FW_PORT_ACTION_GET_PORT_INFO32) |
 		FW_LEN16(port_cmd));
 	ret = t4_wr_mbox(pi->adapter, pi->adapter->mbox,
 			 &port_cmd, sizeof(port_cmd), &port_cmd);
@@ -7405,6 +8773,67 @@ int t4_update_port_info(struct port_info *pi)
 }
 
 /**
+ *	t4_get_link_params - retrieve basic link parameters for given port
+ *	@pi: the port
+ *	@link_okp: value return pointer for link up/down
+ *	@speedp: value return pointer for speed (Mb/s)
+ *	@mtup: value return pointer for mtu
+ *
+ *	Retrieves basic link parameters for a port: link up/down, speed (Mb/s),
+ *	and MTU for a specified port.  A negative error is returned on
+ *	failure; 0 on success.
+ */
+int t4_get_link_params(struct port_info *pi, unsigned int *link_okp,
+		       unsigned int *speedp, unsigned int *mtup)
+{
+	unsigned int fw_caps = pi->adapter->params.fw_caps_support;
+	unsigned int action, link_ok, mtu;
+	struct fw_port_cmd port_cmd;
+	fw_port_cap32_t linkattr;
+	int ret;
+
+	memset(&port_cmd, 0, sizeof(port_cmd));
+	port_cmd.op_to_portid = cpu_to_be32(FW_CMD_OP_V(FW_PORT_CMD) |
+					    FW_CMD_REQUEST_F | FW_CMD_READ_F |
+					    FW_PORT_CMD_PORTID_V(pi->tx_chan));
+	action = (fw_caps == FW_CAPS16
+		  ? FW_PORT_ACTION_GET_PORT_INFO
+		  : FW_PORT_ACTION_GET_PORT_INFO32);
+	port_cmd.action_to_len16 = cpu_to_be32(
+		FW_PORT_CMD_ACTION_V(action) |
+		FW_LEN16(port_cmd));
+	ret = t4_wr_mbox(pi->adapter, pi->adapter->mbox,
+			 &port_cmd, sizeof(port_cmd), &port_cmd);
+	if (ret)
+		return ret;
+
+	if (action == FW_PORT_ACTION_GET_PORT_INFO) {
+		u32 lstatus = be32_to_cpu(port_cmd.u.info.lstatus_to_modtype);
+
+		link_ok = !!(lstatus & FW_PORT_CMD_LSTATUS_F);
+		linkattr = lstatus_to_fwcap(lstatus);
+		mtu = be16_to_cpu(port_cmd.u.info.mtu);
+	} else {
+		u32 lstatus32 =
+			   be32_to_cpu(port_cmd.u.info32.lstatus32_to_cbllen32);
+
+		link_ok = !!(lstatus32 & FW_PORT_CMD_LSTATUS32_F);
+		linkattr = be32_to_cpu(port_cmd.u.info32.linkattr32);
+		mtu = FW_PORT_CMD_MTU32_G(
+			be32_to_cpu(port_cmd.u.info32.auxlinfo32_mtu32));
+	}
+
+	if (link_okp)
+		*link_okp = link_ok;
+	if (speedp)
+		*speedp = fwcap_to_speed(linkattr);
+	if (mtup)
+		*mtup = mtu;
+
+	return 0;
+}
+
+/**
  *      t4_handle_fw_rpl - process a FW reply message
  *      @adap: the adapter
  *      @rpl: start of the FW message
@@ -7424,7 +8853,9 @@ int t4_handle_fw_rpl(struct adapter *adap, const __be64 *rpl)
 	unsigned int action =
 		FW_PORT_CMD_ACTION_G(be32_to_cpu(p->action_to_len16));
 
-	if (opcode == FW_PORT_CMD && action == FW_PORT_ACTION_GET_PORT_INFO) {
+	if (opcode == FW_PORT_CMD &&
+	    (action == FW_PORT_ACTION_GET_PORT_INFO ||
+	     action == FW_PORT_ACTION_GET_PORT_INFO32)) {
 		int i;
 		int chan = FW_PORT_CMD_PORTID_G(be32_to_cpu(p->op_to_portid));
 		struct port_info *pi = NULL;
@@ -7437,7 +8868,8 @@ int t4_handle_fw_rpl(struct adapter *adap, const __be64 *rpl)
 
 		t4_handle_get_port_info(pi, rpl);
 	} else {
-		dev_warn(adap->pdev_dev, "Unknown firmware reply %d\n", opcode);
+		dev_warn(adap->pdev_dev, "Unknown firmware reply %d\n",
+			 opcode);
 		return -EINVAL;
 	}
 	return 0;
@@ -7456,39 +8888,44 @@ static void get_pci_mode(struct adapter *adapter, struct pci_params *p)
 
 /**
  *	init_link_config - initialize a link's SW state
- *	@lc: structure holding the link state
- *	@caps: link capabilities
+ *	@lc: pointer to structure holding the link state
+ *	@pcaps: link Port Capabilities
+ *	@acaps: link current Advertised Port Capabilities
  *
  *	Initializes the SW state maintained for each link, including the link's
  *	capabilities and default speed/flow-control/autonegotiation settings.
  */
-static void init_link_config(struct link_config *lc, unsigned int pcaps,
-			     unsigned int acaps)
+static void init_link_config(struct link_config *lc, fw_port_cap32_t pcaps,
+			     fw_port_cap32_t acaps)
 {
-	lc->supported = pcaps;
-	lc->lp_advertising = 0;
-	lc->requested_speed = 0;
+	lc->pcaps = pcaps;
+	lc->def_acaps = acaps;
+	lc->lpacaps = 0;
+	lc->speed_caps = 0;
 	lc->speed = 0;
 	lc->requested_fc = lc->fc = PAUSE_RX | PAUSE_TX;
-	lc->auto_fec = 0;
 
 	/* For Forward Error Control, we default to whatever the Firmware
 	 * tells us the Link is currently advertising.
 	 */
-	if (acaps & FW_PORT_CAP_FEC_RS)
-		lc->auto_fec |= FEC_RS;
-	if (acaps & FW_PORT_CAP_FEC_BASER_RS)
-		lc->auto_fec |= FEC_BASER_RS;
 	lc->requested_fec = FEC_AUTO;
-	lc->fec = lc->auto_fec;
-
-	if (lc->supported & FW_PORT_CAP_ANEG) {
-		lc->advertising = lc->supported & ADVERT_MASK;
+	lc->fec = fwcap_to_cc_fec(lc->def_acaps);
+
+	/* If the Port is capable of Auto-Negtotiation, initialize it as
+	 * "enabled" and copy over all of the Physical Port Capabilities
+	 * to the Advertised Port Capabilities.  Otherwise mark it as
+	 * Auto-Negotiate disabled and select the highest supported speed
+	 * for the link.  Note parallel structure in t4_link_l1cfg_core()
+	 * and t4_handle_get_port_info().
+	 */
+	if (lc->pcaps & FW_PORT_CAP32_ANEG) {
+		lc->acaps = lc->pcaps & ADVERT_MASK;
 		lc->autoneg = AUTONEG_ENABLE;
 		lc->requested_fc |= PAUSE_AUTONEG;
 	} else {
-		lc->advertising = 0;
+		lc->acaps = 0;
 		lc->autoneg = AUTONEG_DISABLE;
+		lc->speed_caps = fwcap_to_fwspeed(acaps);
 	}
 }
 
@@ -7512,7 +8949,7 @@ struct flash_desc {
 	u32 size_mb;
 };
 
-static int get_flash_params(struct adapter *adap)
+static int t4_get_flash_params(struct adapter *adap)
 {
 	/* Table for non-Numonix supported flash parts.  Numonix parts are left
 	 * to the preexisting code.  All flash parts have 64KB sectors.
@@ -7521,63 +8958,153 @@ static int get_flash_params(struct adapter *adap)
 		{ 0x150201, 4 << 20 },       /* Spansion 4MB S25FL032P */
 	};
 
+	unsigned int part, manufacturer;
+	unsigned int density, size = 0;
+	u32 flashid = 0;
 	int ret;
-	u32 info;
+
+	/* Issue a Read ID Command to the Flash part.  We decode supported
+	 * Flash parts and their sizes from this.  There's a newer Query
+	 * Command which can retrieve detailed geometry information but many
+	 * Flash parts don't support it.
+	 */
 
 	ret = sf1_write(adap, 1, 1, 0, SF_RD_ID);
 	if (!ret)
-		ret = sf1_read(adap, 3, 0, 1, &info);
+		ret = sf1_read(adap, 3, 0, 1, &flashid);
 	t4_write_reg(adap, SF_OP_A, 0);                    /* unlock SF */
 	if (ret)
 		return ret;
 
-	for (ret = 0; ret < ARRAY_SIZE(supported_flash); ++ret)
-		if (supported_flash[ret].vendor_and_model_id == info) {
-			adap->params.sf_size = supported_flash[ret].size_mb;
+	/* Check to see if it's one of our non-standard supported Flash parts.
+	 */
+	for (part = 0; part < ARRAY_SIZE(supported_flash); part++)
+		if (supported_flash[part].vendor_and_model_id == flashid) {
+			adap->params.sf_size = supported_flash[part].size_mb;
 			adap->params.sf_nsec =
 				adap->params.sf_size / SF_SEC_SIZE;
-			return 0;
+			goto found;
 		}
 
-	if ((info & 0xff) != 0x20)             /* not a Numonix flash */
-		return -EINVAL;
-	info >>= 16;                           /* log2 of size */
-	if (info >= 0x14 && info < 0x18)
-		adap->params.sf_nsec = 1 << (info - 16);
-	else if (info == 0x18)
-		adap->params.sf_nsec = 64;
-	else
-		return -EINVAL;
-	adap->params.sf_size = 1 << info;
-	adap->params.sf_fw_start =
-		t4_read_reg(adap, CIM_BOOT_CFG_A) & BOOTADDR_M;
+	/* Decode Flash part size.  The code below looks repetitive with
+	 * common encodings, but that's not guaranteed in the JEDEC
+	 * specification for the Read JEDEC ID command.  The only thing that
+	 * we're guaranteed by the JEDEC specification is where the
+	 * Manufacturer ID is in the returned result.  After that each
+	 * Manufacturer ~could~ encode things completely differently.
+	 * Note, all Flash parts must have 64KB sectors.
+	 */
+	manufacturer = flashid & 0xff;
+	switch (manufacturer) {
+	case 0x20: { /* Micron/Numonix */
+		/* This Density -> Size decoding table is taken from Micron
+		 * Data Sheets.
+		 */
+		density = (flashid >> 16) & 0xff;
+		switch (density) {
+		case 0x14: /* 1MB */
+			size = 1 << 20;
+			break;
+		case 0x15: /* 2MB */
+			size = 1 << 21;
+			break;
+		case 0x16: /* 4MB */
+			size = 1 << 22;
+			break;
+		case 0x17: /* 8MB */
+			size = 1 << 23;
+			break;
+		case 0x18: /* 16MB */
+			size = 1 << 24;
+			break;
+		case 0x19: /* 32MB */
+			size = 1 << 25;
+			break;
+		case 0x20: /* 64MB */
+			size = 1 << 26;
+			break;
+		case 0x21: /* 128MB */
+			size = 1 << 27;
+			break;
+		case 0x22: /* 256MB */
+			size = 1 << 28;
+			break;
+		}
+		break;
+	}
+	case 0x9d: { /* ISSI -- Integrated Silicon Solution, Inc. */
+		/* This Density -> Size decoding table is taken from ISSI
+		 * Data Sheets.
+		 */
+		density = (flashid >> 16) & 0xff;
+		switch (density) {
+		case 0x16: /* 32 MB */
+			size = 1 << 25;
+			break;
+		case 0x17: /* 64MB */
+			size = 1 << 26;
+			break;
+		}
+		break;
+	}
+	case 0xc2: { /* Macronix */
+		/* This Density -> Size decoding table is taken from Macronix
+		 * Data Sheets.
+		 */
+		density = (flashid >> 16) & 0xff;
+		switch (density) {
+		case 0x17: /* 8MB */
+			size = 1 << 23;
+			break;
+		case 0x18: /* 16MB */
+			size = 1 << 24;
+			break;
+		}
+		break;
+	}
+	case 0xef: { /* Winbond */
+		/* This Density -> Size decoding table is taken from Winbond
+		 * Data Sheets.
+		 */
+		density = (flashid >> 16) & 0xff;
+		switch (density) {
+		case 0x17: /* 8MB */
+			size = 1 << 23;
+			break;
+		case 0x18: /* 16MB */
+			size = 1 << 24;
+			break;
+		}
+		break;
+	}
+	}
 
-	if (adap->params.sf_size < FLASH_MIN_SIZE)
-		dev_warn(adap->pdev_dev, "WARNING!!! FLASH size %#x < %#x!!!\n",
-			 adap->params.sf_size, FLASH_MIN_SIZE);
-	return 0;
-}
+	/* If we didn't recognize the FLASH part, that's no real issue: the
+	 * Hardware/Software contract says that Hardware will _*ALWAYS*_
+	 * use a FLASH part which is at least 4MB in size and has 64KB
+	 * sectors.  The unrecognized FLASH part is likely to be much larger
+	 * than 4MB, but that's all we really need.
+	 */
+	if (size == 0) {
+		dev_warn(adap->pdev_dev, "Unknown Flash Part, ID = %#x, assuming 4MB\n",
+			 flashid);
+		size = 1 << 22;
+	}
 
-static void set_pcie_completion_timeout(struct adapter *adapter, u8 range)
-{
-	u16 val;
-	u32 pcie_cap;
+	/* Store decoded Flash size and fall through into vetting code. */
+	adap->params.sf_size = size;
+	adap->params.sf_nsec = size / SF_SEC_SIZE;
 
-	pcie_cap = pci_find_capability(adapter->pdev, PCI_CAP_ID_EXP);
-	if (pcie_cap) {
-		pci_read_config_word(adapter->pdev,
-				     pcie_cap + PCI_EXP_DEVCTL2, &val);
-		val &= ~PCI_EXP_DEVCTL2_COMP_TIMEOUT;
-		val |= range;
-		pci_write_config_word(adapter->pdev,
-				      pcie_cap + PCI_EXP_DEVCTL2, val);
-	}
+found:
+	if (adap->params.sf_size < FLASH_MIN_SIZE)
+		dev_warn(adap->pdev_dev, "WARNING: Flash Part ID %#x, size %#x < %#x\n",
+			 flashid, adap->params.sf_size, FLASH_MIN_SIZE);
+	return 0;
 }
 
 /**
  *	t4_prep_adapter - prepare SW and HW for operation
  *	@adapter: the adapter
- *	@reset: if true perform a HW reset
  *
  *	Initialize adapter SW state for the various HW modules, set initial
  *	values for some adapter tunables, take PHYs out of reset, and
@@ -7592,7 +9119,7 @@ int t4_prep_adapter(struct adapter *adapter)
 	get_pci_mode(adapter, &adapter->params.pci);
 	pl_rev = REV_G(t4_read_reg(adapter, PL_REV_A));
 
-	ret = get_flash_params(adapter);
+	ret = t4_get_flash_params(adapter);
 	if (ret < 0) {
 		dev_err(adapter->pdev_dev, "error %d identifying flash\n", ret);
 		return ret;
@@ -7659,8 +9186,9 @@ int t4_prep_adapter(struct adapter *adapter)
 	adapter->params.portvec = 1;
 	adapter->params.vpd.cclk = 50000;
 
-	/* Set pci completion timeout value to 4 seconds. */
-	set_pcie_completion_timeout(adapter, 0xd);
+	/* Set PCIe completion timeout to 4 seconds. */
+	pcie_capability_clear_and_set_word(adapter->pdev, PCI_EXP_DEVCTL2,
+					   PCI_EXP_DEVCTL2_COMP_TIMEOUT, 0xd);
 	return 0;
 }
 
@@ -7683,10 +9211,9 @@ int t4_shutdown_adapter(struct adapter *adapter)
 	t4_intr_disable(adapter);
 	t4_write_reg(adapter, DBG_GPIO_EN_A, 0);
 	for_each_port(adapter, port) {
-		u32 a_port_cfg = PORT_REG(port,
-					  is_t4(adapter->params.chip)
-					  ? XGMAC_PORT_CFG_A
-					  : MAC_PORT_CFG_A);
+		u32 a_port_cfg = is_t4(adapter->params.chip) ?
+				       PORT_REG(port, XGMAC_PORT_CFG_A) :
+				       T5_PORT_REG(port, MAC_PORT_CFG_A);
 
 		t4_write_reg(adapter, a_port_cfg,
 			     t4_read_reg(adapter, a_port_cfg)
@@ -7802,7 +9329,7 @@ int t4_init_devlog_params(struct adapter *adap)
 	struct fw_devlog_cmd devlog_cmd;
 	int ret;
 
-	/* If we're dealing with newer firmware, the Device Log Paramerters
+	/* If we're dealing with newer firmware, the Device Log Parameters
 	 * are stored in a designated register which allows us to access the
 	 * Device Log even if we can't talk to the firmware.
 	 */
@@ -7821,7 +9348,7 @@ int t4_init_devlog_params(struct adapter *adap)
 		return 0;
 	}
 
-	/* Otherwise, ask the firmware for it's Device Log Parameters.
+	/* Otherwise, ask the firmware for its Device Log Parameters.
 	 */
 	memset(&devlog_cmd, 0, sizeof(devlog_cmd));
 	devlog_cmd.op_to_write = cpu_to_be32(FW_CMD_OP_V(FW_DEVLOG_CMD) |
@@ -7875,13 +9402,15 @@ int t4_init_sge_params(struct adapter *adapter)
 /**
  *      t4_init_tp_params - initialize adap->params.tp
  *      @adap: the adapter
+ *      @sleep_ok: if true we may sleep while awaiting command completion
  *
  *      Initialize various fields of the adapter's TP Parameters structure.
  */
-int t4_init_tp_params(struct adapter *adap)
+int t4_init_tp_params(struct adapter *adap, bool sleep_ok)
 {
-	int chan;
-	u32 v;
+	u32 param, val, v;
+	int chan, ret;
+
 
 	v = t4_read_reg(adap, TP_TIMER_RESOLUTION_A);
 	adap->params.tp.tre = TIMERRESOLUTION_G(v);
@@ -7891,22 +9420,50 @@ int t4_init_tp_params(struct adapter *adap)
 	for (chan = 0; chan < NCHAN; chan++)
 		adap->params.tp.tx_modq[chan] = chan;
 
-	/* Cache the adapter's Compressed Filter Mode and global Incress
+	/* Cache the adapter's Compressed Filter Mode/Mask and global Ingress
 	 * Configuration.
 	 */
-	if (t4_use_ldst(adap)) {
-		t4_fw_tp_pio_rw(adap, &adap->params.tp.vlan_pri_map, 1,
-				TP_VLAN_PRI_MAP_A, 1);
-		t4_fw_tp_pio_rw(adap, &adap->params.tp.ingress_config, 1,
-				TP_INGRESS_CONFIG_A, 1);
+	param = (FW_PARAMS_MNEM_V(FW_PARAMS_MNEM_DEV) |
+		 FW_PARAMS_PARAM_X_V(FW_PARAMS_PARAM_DEV_FILTER) |
+		 FW_PARAMS_PARAM_Y_V(FW_PARAM_DEV_FILTER_MODE_MASK));
+
+	/* Read current value */
+	ret = t4_query_params(adap, adap->mbox, adap->pf, 0, 1,
+			      &param, &val);
+	if (ret == 0) {
+		dev_info(adap->pdev_dev,
+			 "Current filter mode/mask 0x%x:0x%x\n",
+			 FW_PARAMS_PARAM_FILTER_MODE_G(val),
+			 FW_PARAMS_PARAM_FILTER_MASK_G(val));
+		adap->params.tp.vlan_pri_map =
+			FW_PARAMS_PARAM_FILTER_MODE_G(val);
+		adap->params.tp.filter_mask =
+			FW_PARAMS_PARAM_FILTER_MASK_G(val);
 	} else {
-		t4_read_indirect(adap, TP_PIO_ADDR_A, TP_PIO_DATA_A,
-				 &adap->params.tp.vlan_pri_map, 1,
-				 TP_VLAN_PRI_MAP_A);
-		t4_read_indirect(adap, TP_PIO_ADDR_A, TP_PIO_DATA_A,
-				 &adap->params.tp.ingress_config, 1,
-				 TP_INGRESS_CONFIG_A);
+		dev_info(adap->pdev_dev,
+			 "Failed to read filter mode/mask via fw api, using indirect-reg-read\n");
+
+		/* Incase of older-fw (which doesn't expose the api
+		 * FW_PARAM_DEV_FILTER_MODE_MASK) and newer-driver (which uses
+		 * the fw api) combination, fall-back to older method of reading
+		 * the filter mode from indirect-register
+		 */
+		t4_tp_pio_read(adap, &adap->params.tp.vlan_pri_map, 1,
+			       TP_VLAN_PRI_MAP_A, sleep_ok);
+
+		/* With the older-fw and newer-driver combination we might run
+		 * into an issue when user wants to use hash filter region but
+		 * the filter_mask is zero, in this case filter_mask validation
+		 * is tough. To avoid that we set the filter_mask same as filter
+		 * mode, which will behave exactly as the older way of ignoring
+		 * the filter mask validation.
+		 */
+		adap->params.tp.filter_mask = adap->params.tp.vlan_pri_map;
 	}
+
+	t4_tp_pio_read(adap, &adap->params.tp.ingress_config, 1,
+		       TP_INGRESS_CONFIG_A, sleep_ok);
+
 	/* For T6, cache the adapter's compressed error vector
 	 * and passing outer header info for encapsulated packets.
 	 */
@@ -7919,11 +9476,21 @@ int t4_init_tp_params(struct adapter *adap)
 	 * shift positions of several elements of the Compressed Filter Tuple
 	 * for this adapter which we need frequently ...
 	 */
-	adap->params.tp.vlan_shift = t4_filter_field_shift(adap, VLAN_F);
-	adap->params.tp.vnic_shift = t4_filter_field_shift(adap, VNIC_ID_F);
+	adap->params.tp.fcoe_shift = t4_filter_field_shift(adap, FCOE_F);
 	adap->params.tp.port_shift = t4_filter_field_shift(adap, PORT_F);
+	adap->params.tp.vnic_shift = t4_filter_field_shift(adap, VNIC_ID_F);
+	adap->params.tp.vlan_shift = t4_filter_field_shift(adap, VLAN_F);
+	adap->params.tp.tos_shift = t4_filter_field_shift(adap, TOS_F);
 	adap->params.tp.protocol_shift = t4_filter_field_shift(adap,
 							       PROTOCOL_F);
+	adap->params.tp.ethertype_shift = t4_filter_field_shift(adap,
+								ETHERTYPE_F);
+	adap->params.tp.macmatch_shift = t4_filter_field_shift(adap,
+							       MACMATCH_F);
+	adap->params.tp.matchtype_shift = t4_filter_field_shift(adap,
+								MPSHITTYPE_F);
+	adap->params.tp.frag_shift = t4_filter_field_shift(adap,
+							   FRAGMENTATION_F);
 
 	/* If TP_INGRESS_CONFIG.VNID == 0, then TP_VLAN_PRI_MAP.VNIC_ID
 	 * represents the presence of an Outer VLAN instead of a VNIC ID.
@@ -7931,6 +9498,10 @@ int t4_init_tp_params(struct adapter *adap)
 	if ((adap->params.tp.ingress_config & VNIC_F) == 0)
 		adap->params.tp.vnic_shift = -1;
 
+	v = t4_read_reg(adap, LE_3_DB_HASH_MASK_GEN_IPV4_T6_A);
+	adap->params.tp.hash_filter_mask = v;
+	v = t4_read_reg(adap, LE_4_DB_HASH_MASK_GEN_IPV4_T6_A);
+	adap->params.tp.hash_filter_mask |= ((u64)v << 32);
 	return 0;
 }
 
@@ -8013,7 +9584,7 @@ int t4_init_rss_mode(struct adapter *adap, int mbox)
 }
 
 /**
- *	t4_init_portinfo - allocate a virtual interface amd initialize port_info
+ *	t4_init_portinfo - allocate a virtual interface and initialize port_info
  *	@pi: the port_info
  *	@mbox: mailbox to use for the FW command
  *	@port: physical port associated with the VI
@@ -8029,22 +9600,70 @@ int t4_init_rss_mode(struct adapter *adap, int mbox)
 int t4_init_portinfo(struct port_info *pi, int mbox,
 		     int port, int pf, int vf, u8 mac[])
 {
-	int ret;
-	struct fw_port_cmd c;
+	struct adapter *adapter = pi->adapter;
+	unsigned int fw_caps = adapter->params.fw_caps_support;
+	struct fw_port_cmd cmd;
 	unsigned int rss_size;
+	enum fw_port_type port_type;
+	int mdio_addr;
+	fw_port_cap32_t pcaps, acaps;
+	u8 vivld = 0, vin = 0;
+	int ret;
 
-	memset(&c, 0, sizeof(c));
-	c.op_to_portid = cpu_to_be32(FW_CMD_OP_V(FW_PORT_CMD) |
-				     FW_CMD_REQUEST_F | FW_CMD_READ_F |
-				     FW_PORT_CMD_PORTID_V(port));
-	c.action_to_len16 = cpu_to_be32(
-		FW_PORT_CMD_ACTION_V(FW_PORT_ACTION_GET_PORT_INFO) |
-		FW_LEN16(c));
-	ret = t4_wr_mbox(pi->adapter, mbox, &c, sizeof(c), &c);
+	/* If we haven't yet determined whether we're talking to Firmware
+	 * which knows the new 32-bit Port Capabilities, it's time to find
+	 * out now.  This will also tell new Firmware to send us Port Status
+	 * Updates using the new 32-bit Port Capabilities version of the
+	 * Port Information message.
+	 */
+	if (fw_caps == FW_CAPS_UNKNOWN) {
+		u32 param, val;
+
+		param = (FW_PARAMS_MNEM_V(FW_PARAMS_MNEM_PFVF) |
+			 FW_PARAMS_PARAM_X_V(FW_PARAMS_PARAM_PFVF_PORT_CAPS32));
+		val = 1;
+		ret = t4_set_params(adapter, mbox, pf, vf, 1, &param, &val);
+		fw_caps = (ret == 0 ? FW_CAPS32 : FW_CAPS16);
+		adapter->params.fw_caps_support = fw_caps;
+	}
+
+	memset(&cmd, 0, sizeof(cmd));
+	cmd.op_to_portid = cpu_to_be32(FW_CMD_OP_V(FW_PORT_CMD) |
+				       FW_CMD_REQUEST_F | FW_CMD_READ_F |
+				       FW_PORT_CMD_PORTID_V(port));
+	cmd.action_to_len16 = cpu_to_be32(
+		FW_PORT_CMD_ACTION_V(fw_caps == FW_CAPS16
+				     ? FW_PORT_ACTION_GET_PORT_INFO
+				     : FW_PORT_ACTION_GET_PORT_INFO32) |
+		FW_LEN16(cmd));
+	ret = t4_wr_mbox(pi->adapter, mbox, &cmd, sizeof(cmd), &cmd);
 	if (ret)
 		return ret;
 
-	ret = t4_alloc_vi(pi->adapter, mbox, port, pf, vf, 1, mac, &rss_size);
+	/* Extract the various fields from the Port Information message.
+	 */
+	if (fw_caps == FW_CAPS16) {
+		u32 lstatus = be32_to_cpu(cmd.u.info.lstatus_to_modtype);
+
+		port_type = FW_PORT_CMD_PTYPE_G(lstatus);
+		mdio_addr = ((lstatus & FW_PORT_CMD_MDIOCAP_F)
+			     ? FW_PORT_CMD_MDIOADDR_G(lstatus)
+			     : -1);
+		pcaps = fwcaps16_to_caps32(be16_to_cpu(cmd.u.info.pcap));
+		acaps = fwcaps16_to_caps32(be16_to_cpu(cmd.u.info.acap));
+	} else {
+		u32 lstatus32 = be32_to_cpu(cmd.u.info32.lstatus32_to_cbllen32);
+
+		port_type = FW_PORT_CMD_PORTTYPE32_G(lstatus32);
+		mdio_addr = ((lstatus32 & FW_PORT_CMD_MDIOCAP32_F)
+			     ? FW_PORT_CMD_MDIOADDR32_G(lstatus32)
+			     : -1);
+		pcaps = be32_to_cpu(cmd.u.info32.pcaps32);
+		acaps = be32_to_cpu(cmd.u.info32.acaps32);
+	}
+
+	ret = t4_alloc_vi(pi->adapter, mbox, port, pf, vf, 1, mac, &rss_size,
+			  &vivld, &vin);
 	if (ret < 0)
 		return ret;
 
@@ -8052,15 +9671,25 @@ int t4_init_portinfo(struct port_info *pi, int mbox,
 	pi->tx_chan = port;
 	pi->lport = port;
 	pi->rss_size = rss_size;
+	pi->rx_cchan = t4_get_tp_e2c_map(pi->adapter, port);
+
+	/* If fw supports returning the VIN as part of FW_VI_CMD,
+	 * save the returned values.
+	 */
+	if (adapter->params.viid_smt_extn_support) {
+		pi->vivld = vivld;
+		pi->vin = vin;
+	} else {
+		/* Retrieve the values from VIID */
+		pi->vivld = FW_VIID_VIVLD_G(pi->viid);
+		pi->vin =  FW_VIID_VIN_G(pi->viid);
+	}
 
-	ret = be32_to_cpu(c.u.info.lstatus_to_modtype);
-	pi->mdio_addr = (ret & FW_PORT_CMD_MDIOCAP_F) ?
-		FW_PORT_CMD_MDIOADDR_G(ret) : -1;
-	pi->port_type = FW_PORT_CMD_PTYPE_G(ret);
+	pi->port_type = port_type;
+	pi->mdio_addr = mdio_addr;
 	pi->mod_type = FW_PORT_MOD_TYPE_NA;
 
-	init_link_config(&pi->link_cfg, be16_to_cpu(c.u.info.pcap),
-			 be16_to_cpu(c.u.info.acap));
+	init_link_config(&pi->link_cfg, pcaps, acaps);
 	return 0;
 }
 
@@ -8079,12 +9708,28 @@ int t4_port_init(struct adapter *adap, int mbox, int pf, int vf)
 		if (ret)
 			return ret;
 
-		memcpy(adap->port[i]->dev_addr, addr, ETH_ALEN);
+		eth_hw_addr_set(adap->port[i], addr);
 		j++;
 	}
 	return 0;
 }
 
+int t4_init_port_mirror(struct port_info *pi, u8 mbox, u8 port, u8 pf, u8 vf,
+			u16 *mirror_viid)
+{
+	int ret;
+
+	ret = t4_alloc_vi(pi->adapter, mbox, port, pf, vf, 1, NULL, NULL,
+			  NULL, NULL);
+	if (ret < 0)
+		return ret;
+
+	if (mirror_viid)
+		*mirror_viid = ret;
+
+	return 0;
+}
+
 /**
  *	t4_read_cimq_cfg - read CIM queue configuration
  *	@adap: the adapter
@@ -8312,7 +9957,16 @@ int t4_cim_read_la(struct adapter *adap, u32 *la_buf, unsigned int *wrptr)
 		ret = t4_cim_read(adap, UP_UP_DBG_LA_DATA_A, 1, &la_buf[i]);
 		if (ret)
 			break;
-		idx = (idx + 1) & UPDBGLARDPTR_M;
+
+		/* Bits 0-3 of UpDbgLaRdPtr can be between 0000 to 1001 to
+		 * identify the 32-bit portion of the full 312-bit data
+		 */
+		if (is_t6(adap->params.chip) && (idx & 0xf) >= 9)
+			idx = (idx & 0xff0) + 0x10;
+		else
+			idx++;
+		/* address can't exceed 0xfff */
+		idx &= UPDBGLARDPTR_M;
 	}
 restart:
 	if (cfg & UPDBGLAEN_F) {
@@ -8499,14 +10153,74 @@ void t4_idma_monitor(struct adapter *adapter,
 }
 
 /**
- *	t4_set_vf_mac - Set MAC address for the specified VF
+ *	t4_load_cfg - download config file
+ *	@adap: the adapter
+ *	@cfg_data: the cfg text file to write
+ *	@size: text file size
+ *
+ *	Write the supplied config text file to the card's serial flash.
+ */
+int t4_load_cfg(struct adapter *adap, const u8 *cfg_data, unsigned int size)
+{
+	int ret, i, n, cfg_addr;
+	unsigned int addr;
+	unsigned int flash_cfg_start_sec;
+	unsigned int sf_sec_size = adap->params.sf_size / adap->params.sf_nsec;
+
+	cfg_addr = t4_flash_cfg_addr(adap);
+	if (cfg_addr < 0)
+		return cfg_addr;
+
+	addr = cfg_addr;
+	flash_cfg_start_sec = addr / SF_SEC_SIZE;
+
+	if (size > FLASH_CFG_MAX_SIZE) {
+		dev_err(adap->pdev_dev, "cfg file too large, max is %u bytes\n",
+			FLASH_CFG_MAX_SIZE);
+		return -EFBIG;
+	}
+
+	i = DIV_ROUND_UP(FLASH_CFG_MAX_SIZE,	/* # of sectors spanned */
+			 sf_sec_size);
+	ret = t4_flash_erase_sectors(adap, flash_cfg_start_sec,
+				     flash_cfg_start_sec + i - 1);
+	/* If size == 0 then we're simply erasing the FLASH sectors associated
+	 * with the on-adapter Firmware Configuration File.
+	 */
+	if (ret || size == 0)
+		goto out;
+
+	/* this will write to the flash up to SF_PAGE_SIZE at a time */
+	for (i = 0; i < size; i += SF_PAGE_SIZE) {
+		if ((size - i) <  SF_PAGE_SIZE)
+			n = size - i;
+		else
+			n = SF_PAGE_SIZE;
+		ret = t4_write_flash(adap, addr, n, cfg_data, true);
+		if (ret)
+			goto out;
+
+		addr += SF_PAGE_SIZE;
+		cfg_data += SF_PAGE_SIZE;
+	}
+
+out:
+	if (ret)
+		dev_err(adap->pdev_dev, "config file %s failed %d\n",
+			(size == 0 ? "clear" : "download"), ret);
+	return ret;
+}
+
+/**
+ *	t4_set_vf_mac_acl - Set MAC address for the specified VF
  *	@adapter: The adapter
  *	@vf: one of the VFs instantiated by the specified PF
+ *	@start: The start port id associated with specified VF
  *	@naddr: the number of MAC addresses
  *	@addr: the MAC address(es) to be set to the specified VF
  */
 int t4_set_vf_mac_acl(struct adapter *adapter, unsigned int vf,
-		      unsigned int naddr, u8 *addr)
+		      u8 start, unsigned int naddr, u8 *addr)
 {
 	struct fw_acl_mac_cmd cmd;
 
@@ -8521,7 +10235,7 @@ int t4_set_vf_mac_acl(struct adapter *adapter, unsigned int vf,
 	cmd.en_to_len16 = cpu_to_be32((unsigned int)FW_LEN16(cmd));
 	cmd.nmac = naddr;
 
-	switch (adapter->pf) {
+	switch (start) {
 	case 3:
 		memcpy(cmd.macaddr3, addr, sizeof(cmd.macaddr3));
 		break;
@@ -8539,9 +10253,129 @@ int t4_set_vf_mac_acl(struct adapter *adapter, unsigned int vf,
 	return t4_wr_mbox(adapter, adapter->mbox, &cmd, sizeof(cmd), &cmd);
 }
 
-int t4_sched_params(struct adapter *adapter, int type, int level, int mode,
-		    int rateunit, int ratemode, int channel, int class,
-		    int minrate, int maxrate, int weight, int pktsize)
+/**
+ * t4_read_pace_tbl - read the pace table
+ * @adap: the adapter
+ * @pace_vals: holds the returned values
+ *
+ * Returns the values of TP's pace table in microseconds.
+ */
+void t4_read_pace_tbl(struct adapter *adap, unsigned int pace_vals[NTX_SCHED])
+{
+	unsigned int i, v;
+
+	for (i = 0; i < NTX_SCHED; i++) {
+		t4_write_reg(adap, TP_PACE_TABLE_A, 0xffff0000 + i);
+		v = t4_read_reg(adap, TP_PACE_TABLE_A);
+		pace_vals[i] = dack_ticks_to_usec(adap, v);
+	}
+}
+
+/**
+ * t4_get_tx_sched - get the configuration of a Tx HW traffic scheduler
+ * @adap: the adapter
+ * @sched: the scheduler index
+ * @kbps: the byte rate in Kbps
+ * @ipg: the interpacket delay in tenths of nanoseconds
+ * @sleep_ok: if true we may sleep while awaiting command completion
+ *
+ * Return the current configuration of a HW Tx scheduler.
+ */
+void t4_get_tx_sched(struct adapter *adap, unsigned int sched,
+		     unsigned int *kbps, unsigned int *ipg, bool sleep_ok)
+{
+	unsigned int v, addr, bpt, cpt;
+
+	if (kbps) {
+		addr = TP_TX_MOD_Q1_Q0_RATE_LIMIT_A - sched / 2;
+		t4_tp_tm_pio_read(adap, &v, 1, addr, sleep_ok);
+		if (sched & 1)
+			v >>= 16;
+		bpt = (v >> 8) & 0xff;
+		cpt = v & 0xff;
+		if (!cpt) {
+			*kbps = 0;	/* scheduler disabled */
+		} else {
+			v = (adap->params.vpd.cclk * 1000) / cpt; /* ticks/s */
+			*kbps = (v * bpt) / 125;
+		}
+	}
+	if (ipg) {
+		addr = TP_TX_MOD_Q1_Q0_TIMER_SEPARATOR_A - sched / 2;
+		t4_tp_tm_pio_read(adap, &v, 1, addr, sleep_ok);
+		if (sched & 1)
+			v >>= 16;
+		v &= 0xffff;
+		*ipg = (10000 * v) / core_ticks_per_usec(adap);
+	}
+}
+
+/* t4_sge_ctxt_rd - read an SGE context through FW
+ * @adap: the adapter
+ * @mbox: mailbox to use for the FW command
+ * @cid: the context id
+ * @ctype: the context type
+ * @data: where to store the context data
+ *
+ * Issues a FW command through the given mailbox to read an SGE context.
+ */
+int t4_sge_ctxt_rd(struct adapter *adap, unsigned int mbox, unsigned int cid,
+		   enum ctxt_type ctype, u32 *data)
+{
+	struct fw_ldst_cmd c;
+	int ret;
+
+	if (ctype == CTXT_FLM)
+		ret = FW_LDST_ADDRSPC_SGE_FLMC;
+	else
+		ret = FW_LDST_ADDRSPC_SGE_CONMC;
+
+	memset(&c, 0, sizeof(c));
+	c.op_to_addrspace = cpu_to_be32(FW_CMD_OP_V(FW_LDST_CMD) |
+					FW_CMD_REQUEST_F | FW_CMD_READ_F |
+					FW_LDST_CMD_ADDRSPACE_V(ret));
+	c.cycles_to_len16 = cpu_to_be32(FW_LEN16(c));
+	c.u.idctxt.physid = cpu_to_be32(cid);
+
+	ret = t4_wr_mbox(adap, mbox, &c, sizeof(c), &c);
+	if (ret == 0) {
+		data[0] = be32_to_cpu(c.u.idctxt.ctxt_data0);
+		data[1] = be32_to_cpu(c.u.idctxt.ctxt_data1);
+		data[2] = be32_to_cpu(c.u.idctxt.ctxt_data2);
+		data[3] = be32_to_cpu(c.u.idctxt.ctxt_data3);
+		data[4] = be32_to_cpu(c.u.idctxt.ctxt_data4);
+		data[5] = be32_to_cpu(c.u.idctxt.ctxt_data5);
+	}
+	return ret;
+}
+
+/**
+ * t4_sge_ctxt_rd_bd - read an SGE context bypassing FW
+ * @adap: the adapter
+ * @cid: the context id
+ * @ctype: the context type
+ * @data: where to store the context data
+ *
+ * Reads an SGE context directly, bypassing FW.  This is only for
+ * debugging when FW is unavailable.
+ */
+int t4_sge_ctxt_rd_bd(struct adapter *adap, unsigned int cid,
+		      enum ctxt_type ctype, u32 *data)
+{
+	int i, ret;
+
+	t4_write_reg(adap, SGE_CTXT_CMD_A, CTXTQID_V(cid) | CTXTTYPE_V(ctype));
+	ret = t4_wait_op_done(adap, SGE_CTXT_CMD_A, BUSY_F, 0, 3, 1);
+	if (!ret)
+		for (i = SGE_CTXT_DATA0_A; i <= SGE_CTXT_DATA5_A; i += 4)
+			*data++ = t4_read_reg(adap, i);
+	return ret;
+}
+
+int t4_sched_params(struct adapter *adapter, u8 type, u8 level, u8 mode,
+		    u8 rateunit, u8 ratemode, u8 channel, u8 class,
+		    u32 minrate, u32 maxrate, u16 weight, u16 pktsize,
+		    u16 burstsize)
 {
 	struct fw_sched_cmd cmd;
 
@@ -8563,7 +10397,378 @@ int t4_sched_params(struct adapter *adapter, int type, int level, int mode,
 	cmd.u.params.max = cpu_to_be32(maxrate);
 	cmd.u.params.weight = cpu_to_be16(weight);
 	cmd.u.params.pktsize = cpu_to_be16(pktsize);
+	cmd.u.params.burstsize = cpu_to_be16(burstsize);
 
 	return t4_wr_mbox_meat(adapter, adapter->mbox, &cmd, sizeof(cmd),
 			       NULL, 1);
 }
+
+/**
+ *	t4_i2c_rd - read I2C data from adapter
+ *	@adap: the adapter
+ *	@mbox: mailbox to use for the FW command
+ *	@port: Port number if per-port device; <0 if not
+ *	@devid: per-port device ID or absolute device ID
+ *	@offset: byte offset into device I2C space
+ *	@len: byte length of I2C space data
+ *	@buf: buffer in which to return I2C data
+ *
+ *	Reads the I2C data from the indicated device and location.
+ */
+int t4_i2c_rd(struct adapter *adap, unsigned int mbox, int port,
+	      unsigned int devid, unsigned int offset,
+	      unsigned int len, u8 *buf)
+{
+	struct fw_ldst_cmd ldst_cmd, ldst_rpl;
+	unsigned int i2c_max = sizeof(ldst_cmd.u.i2c.data);
+	int ret = 0;
+
+	if (len > I2C_PAGE_SIZE)
+		return -EINVAL;
+
+	/* Dont allow reads that spans multiple pages */
+	if (offset < I2C_PAGE_SIZE && offset + len > I2C_PAGE_SIZE)
+		return -EINVAL;
+
+	memset(&ldst_cmd, 0, sizeof(ldst_cmd));
+	ldst_cmd.op_to_addrspace =
+		cpu_to_be32(FW_CMD_OP_V(FW_LDST_CMD) |
+			    FW_CMD_REQUEST_F |
+			    FW_CMD_READ_F |
+			    FW_LDST_CMD_ADDRSPACE_V(FW_LDST_ADDRSPC_I2C));
+	ldst_cmd.cycles_to_len16 = cpu_to_be32(FW_LEN16(ldst_cmd));
+	ldst_cmd.u.i2c.pid = (port < 0 ? 0xff : port);
+	ldst_cmd.u.i2c.did = devid;
+
+	while (len > 0) {
+		unsigned int i2c_len = (len < i2c_max) ? len : i2c_max;
+
+		ldst_cmd.u.i2c.boffset = offset;
+		ldst_cmd.u.i2c.blen = i2c_len;
+
+		ret = t4_wr_mbox(adap, mbox, &ldst_cmd, sizeof(ldst_cmd),
+				 &ldst_rpl);
+		if (ret)
+			break;
+
+		memcpy(buf, ldst_rpl.u.i2c.data, i2c_len);
+		offset += i2c_len;
+		buf += i2c_len;
+		len -= i2c_len;
+	}
+
+	return ret;
+}
+
+/**
+ *      t4_set_vlan_acl - Set a VLAN id for the specified VF
+ *      @adap: the adapter
+ *      @mbox: mailbox to use for the FW command
+ *      @vf: one of the VFs instantiated by the specified PF
+ *      @vlan: The vlanid to be set
+ */
+int t4_set_vlan_acl(struct adapter *adap, unsigned int mbox, unsigned int vf,
+		    u16 vlan)
+{
+	struct fw_acl_vlan_cmd vlan_cmd;
+	unsigned int enable;
+
+	enable = (vlan ? FW_ACL_VLAN_CMD_EN_F : 0);
+	memset(&vlan_cmd, 0, sizeof(vlan_cmd));
+	vlan_cmd.op_to_vfn = cpu_to_be32(FW_CMD_OP_V(FW_ACL_VLAN_CMD) |
+					 FW_CMD_REQUEST_F |
+					 FW_CMD_WRITE_F |
+					 FW_CMD_EXEC_F |
+					 FW_ACL_VLAN_CMD_PFN_V(adap->pf) |
+					 FW_ACL_VLAN_CMD_VFN_V(vf));
+	vlan_cmd.en_to_len16 = cpu_to_be32(enable | FW_LEN16(vlan_cmd));
+	/* Drop all packets that donot match vlan id */
+	vlan_cmd.dropnovlan_fm = (enable
+				  ? (FW_ACL_VLAN_CMD_DROPNOVLAN_F |
+				     FW_ACL_VLAN_CMD_FM_F) : 0);
+	if (enable != 0) {
+		vlan_cmd.nvlan = 1;
+		vlan_cmd.vlanid[0] = cpu_to_be16(vlan);
+	}
+
+	return t4_wr_mbox(adap, adap->mbox, &vlan_cmd, sizeof(vlan_cmd), NULL);
+}
+
+/**
+ *	modify_device_id - Modifies the device ID of the Boot BIOS image
+ *	@device_id: the device ID to write.
+ *	@boot_data: the boot image to modify.
+ *
+ *	Write the supplied device ID to the boot BIOS image.
+ */
+static void modify_device_id(int device_id, u8 *boot_data)
+{
+	struct cxgb4_pcir_data *pcir_header;
+	struct legacy_pci_rom_hdr *header;
+	u8 *cur_header = boot_data;
+	u16 pcir_offset;
+
+	 /* Loop through all chained images and change the device ID's */
+	do {
+		header = (struct legacy_pci_rom_hdr *)cur_header;
+		pcir_offset = le16_to_cpu(header->pcir_offset);
+		pcir_header = (struct cxgb4_pcir_data *)(cur_header +
+			      pcir_offset);
+
+		/**
+		 * Only modify the Device ID if code type is Legacy or HP.
+		 * 0x00: Okay to modify
+		 * 0x01: FCODE. Do not modify
+		 * 0x03: Okay to modify
+		 * 0x04-0xFF: Do not modify
+		 */
+		if (pcir_header->code_type == CXGB4_HDR_CODE1) {
+			u8 csum = 0;
+			int i;
+
+			/**
+			 * Modify Device ID to match current adatper
+			 */
+			pcir_header->device_id = cpu_to_le16(device_id);
+
+			/**
+			 * Set checksum temporarily to 0.
+			 * We will recalculate it later.
+			 */
+			header->cksum = 0x0;
+
+			/**
+			 * Calculate and update checksum
+			 */
+			for (i = 0; i < (header->size512 * 512); i++)
+				csum += cur_header[i];
+
+			/**
+			 * Invert summed value to create the checksum
+			 * Writing new checksum value directly to the boot data
+			 */
+			cur_header[7] = -csum;
+
+		} else if (pcir_header->code_type == CXGB4_HDR_CODE2) {
+			/**
+			 * Modify Device ID to match current adatper
+			 */
+			pcir_header->device_id = cpu_to_le16(device_id);
+		}
+
+		/**
+		 * Move header pointer up to the next image in the ROM.
+		 */
+		cur_header += header->size512 * 512;
+	} while (!(pcir_header->indicator & CXGB4_HDR_INDI));
+}
+
+/**
+ *	t4_load_boot - download boot flash
+ *	@adap: the adapter
+ *	@boot_data: the boot image to write
+ *	@boot_addr: offset in flash to write boot_data
+ *	@size: image size
+ *
+ *	Write the supplied boot image to the card's serial flash.
+ *	The boot image has the following sections: a 28-byte header and the
+ *	boot image.
+ */
+int t4_load_boot(struct adapter *adap, u8 *boot_data,
+		 unsigned int boot_addr, unsigned int size)
+{
+	unsigned int sf_sec_size = adap->params.sf_size / adap->params.sf_nsec;
+	unsigned int boot_sector = (boot_addr * 1024);
+	struct cxgb4_pci_exp_rom_header *header;
+	struct cxgb4_pcir_data *pcir_header;
+	int pcir_offset;
+	unsigned int i;
+	u16 device_id;
+	int ret, addr;
+
+	/**
+	 * Make sure the boot image does not encroach on the firmware region
+	 */
+	if ((boot_sector + size) >> 16 > FLASH_FW_START_SEC) {
+		dev_err(adap->pdev_dev, "boot image encroaching on firmware region\n");
+		return -EFBIG;
+	}
+
+	/* Get boot header */
+	header = (struct cxgb4_pci_exp_rom_header *)boot_data;
+	pcir_offset = le16_to_cpu(header->pcir_offset);
+	/* PCIR Data Structure */
+	pcir_header = (struct cxgb4_pcir_data *)&boot_data[pcir_offset];
+
+	/**
+	 * Perform some primitive sanity testing to avoid accidentally
+	 * writing garbage over the boot sectors.  We ought to check for
+	 * more but it's not worth it for now ...
+	 */
+	if (size < BOOT_MIN_SIZE || size > BOOT_MAX_SIZE) {
+		dev_err(adap->pdev_dev, "boot image too small/large\n");
+		return -EFBIG;
+	}
+
+	if (le16_to_cpu(header->signature) != BOOT_SIGNATURE) {
+		dev_err(adap->pdev_dev, "Boot image missing signature\n");
+		return -EINVAL;
+	}
+
+	/* Check PCI header signature */
+	if (le32_to_cpu(pcir_header->signature) != PCIR_SIGNATURE) {
+		dev_err(adap->pdev_dev, "PCI header missing signature\n");
+		return -EINVAL;
+	}
+
+	/* Check Vendor ID matches Chelsio ID*/
+	if (le16_to_cpu(pcir_header->vendor_id) != PCI_VENDOR_ID_CHELSIO) {
+		dev_err(adap->pdev_dev, "Vendor ID missing signature\n");
+		return -EINVAL;
+	}
+
+	/**
+	 * The boot sector is comprised of the Expansion-ROM boot, iSCSI boot,
+	 * and Boot configuration data sections. These 3 boot sections span
+	 * sectors 0 to 7 in flash and live right before the FW image location.
+	 */
+	i = DIV_ROUND_UP(size ? size : FLASH_FW_START,  sf_sec_size);
+	ret = t4_flash_erase_sectors(adap, boot_sector >> 16,
+				     (boot_sector >> 16) + i - 1);
+
+	/**
+	 * If size == 0 then we're simply erasing the FLASH sectors associated
+	 * with the on-adapter option ROM file
+	 */
+	if (ret || size == 0)
+		goto out;
+	/* Retrieve adapter's device ID */
+	pci_read_config_word(adap->pdev, PCI_DEVICE_ID, &device_id);
+       /* Want to deal with PF 0 so I strip off PF 4 indicator */
+	device_id = device_id & 0xf0ff;
+
+	 /* Check PCIE Device ID */
+	if (le16_to_cpu(pcir_header->device_id) != device_id) {
+		/**
+		 * Change the device ID in the Boot BIOS image to match
+		 * the Device ID of the current adapter.
+		 */
+		modify_device_id(device_id, boot_data);
+	}
+
+	/**
+	 * Skip over the first SF_PAGE_SIZE worth of data and write it after
+	 * we finish copying the rest of the boot image. This will ensure
+	 * that the BIOS boot header will only be written if the boot image
+	 * was written in full.
+	 */
+	addr = boot_sector;
+	for (size -= SF_PAGE_SIZE; size; size -= SF_PAGE_SIZE) {
+		addr += SF_PAGE_SIZE;
+		boot_data += SF_PAGE_SIZE;
+		ret = t4_write_flash(adap, addr, SF_PAGE_SIZE, boot_data,
+				     false);
+		if (ret)
+			goto out;
+	}
+
+	ret = t4_write_flash(adap, boot_sector, SF_PAGE_SIZE,
+			     (const u8 *)header, false);
+
+out:
+	if (ret)
+		dev_err(adap->pdev_dev, "boot image load failed, error %d\n",
+			ret);
+	return ret;
+}
+
+/**
+ *	t4_flash_bootcfg_addr - return the address of the flash
+ *	optionrom configuration
+ *	@adapter: the adapter
+ *
+ *	Return the address within the flash where the OptionROM Configuration
+ *	is stored, or an error if the device FLASH is too small to contain
+ *	a OptionROM Configuration.
+ */
+static int t4_flash_bootcfg_addr(struct adapter *adapter)
+{
+	/**
+	 * If the device FLASH isn't large enough to hold a Firmware
+	 * Configuration File, return an error.
+	 */
+	if (adapter->params.sf_size <
+	    FLASH_BOOTCFG_START + FLASH_BOOTCFG_MAX_SIZE)
+		return -ENOSPC;
+
+	return FLASH_BOOTCFG_START;
+}
+
+int t4_load_bootcfg(struct adapter *adap, const u8 *cfg_data, unsigned int size)
+{
+	unsigned int sf_sec_size = adap->params.sf_size / adap->params.sf_nsec;
+	struct cxgb4_bootcfg_data *header;
+	unsigned int flash_cfg_start_sec;
+	unsigned int addr, npad;
+	int ret, i, n, cfg_addr;
+
+	cfg_addr = t4_flash_bootcfg_addr(adap);
+	if (cfg_addr < 0)
+		return cfg_addr;
+
+	addr = cfg_addr;
+	flash_cfg_start_sec = addr / SF_SEC_SIZE;
+
+	if (size > FLASH_BOOTCFG_MAX_SIZE) {
+		dev_err(adap->pdev_dev, "bootcfg file too large, max is %u bytes\n",
+			FLASH_BOOTCFG_MAX_SIZE);
+		return -EFBIG;
+	}
+
+	header = (struct cxgb4_bootcfg_data *)cfg_data;
+	if (le16_to_cpu(header->signature) != BOOT_CFG_SIG) {
+		dev_err(adap->pdev_dev, "Wrong bootcfg signature\n");
+		ret = -EINVAL;
+		goto out;
+	}
+
+	i = DIV_ROUND_UP(FLASH_BOOTCFG_MAX_SIZE,
+			 sf_sec_size);
+	ret = t4_flash_erase_sectors(adap, flash_cfg_start_sec,
+				     flash_cfg_start_sec + i - 1);
+
+	/**
+	 * If size == 0 then we're simply erasing the FLASH sectors associated
+	 * with the on-adapter OptionROM Configuration File.
+	 */
+	if (ret || size == 0)
+		goto out;
+
+	/* this will write to the flash up to SF_PAGE_SIZE at a time */
+	for (i = 0; i < size; i += SF_PAGE_SIZE) {
+		n = min_t(u32, size - i, SF_PAGE_SIZE);
+
+		ret = t4_write_flash(adap, addr, n, cfg_data, false);
+		if (ret)
+			goto out;
+
+		addr += SF_PAGE_SIZE;
+		cfg_data += SF_PAGE_SIZE;
+	}
+
+	npad = ((size + 4 - 1) & ~3) - size;
+	for (i = 0; i < npad; i++) {
+		u8 data = 0;
+
+		ret = t4_write_flash(adap, cfg_addr + size + i, 1, &data,
+				     false);
+		if (ret)
+			goto out;
+	}
+
+out:
+	if (ret)
+		dev_err(adap->pdev_dev, "boot config data %s failed %d\n",
+			(size == 0 ? "clear" : "download"), ret);
+	return ret;
+}
diff --git a/drivers/net/ethernet/chelsio/cxgb4/t4_hw.h b/drivers/net/ethernet/chelsio/cxgb4/t4_hw.h
index 7f59ca458431..63bc956d2037 100644
--- a/drivers/net/ethernet/chelsio/cxgb4/t4_hw.h
+++ b/drivers/net/ethernet/chelsio/cxgb4/t4_hw.h
@@ -38,20 +38,22 @@
 #include <linux/types.h>
 
 enum {
-	NCHAN          = 4,     /* # of HW channels */
-	MAX_MTU        = 9600,  /* max MAC MTU, excluding header + FCS */
-	EEPROMSIZE     = 17408, /* Serial EEPROM physical size */
-	EEPROMVSIZE    = 32768, /* Serial EEPROM virtual address space size */
-	EEPROMPFSIZE   = 1024,  /* EEPROM writable area size for PFn, n>0 */
-	RSS_NENTRIES   = 2048,  /* # of entries in RSS mapping table */
-	TCB_SIZE       = 128,   /* TCB size */
-	NMTUS          = 16,    /* size of MTU table */
-	NCCTRL_WIN     = 32,    /* # of congestion control windows */
-	PM_NSTATS      = 5,     /* # of PM stats */
-	T6_PM_NSTATS   = 7,     /* # of PM stats in T6 */
-	MBOX_LEN       = 64,    /* mailbox size in bytes */
-	TRACE_LEN      = 112,   /* length of trace data and mask */
-	FILTER_OPT_LEN = 36,    /* filter tuple width for optional components */
+	NCHAN           = 4,    /* # of HW channels */
+	MAX_MTU         = 9600, /* max MAC MTU, excluding header + FCS */
+	EEPROMSIZE      = 17408,/* Serial EEPROM physical size */
+	EEPROMVSIZE     = 32768,/* Serial EEPROM virtual address space size */
+	EEPROMPFSIZE    = 1024, /* EEPROM writable area size for PFn, n>0 */
+	RSS_NENTRIES    = 2048, /* # of entries in RSS mapping table */
+	T6_RSS_NENTRIES = 4096, /* # of entries in RSS mapping table */
+	TCB_SIZE        = 128,  /* TCB size */
+	NMTUS           = 16,   /* size of MTU table */
+	NCCTRL_WIN      = 32,   /* # of congestion control windows */
+	NTX_SCHED       = 8,    /* # of HW Tx scheduling queues */
+	PM_NSTATS       = 5,    /* # of PM stats */
+	T6_PM_NSTATS    = 7,    /* # of PM stats in T6 */
+	MBOX_LEN        = 64,   /* mailbox size in bytes */
+	TRACE_LEN       = 112,  /* length of trace data and mask */
+	FILTER_OPT_LEN  = 36,   /* filter tuple width for optional components */
 };
 
 enum {
@@ -67,6 +69,14 @@ enum {
 	ULPRX_LA_SIZE  = 512,   /* # of 256-bit words in ULP_RX LA */
 };
 
+/* SGE context types */
+enum ctxt_type {
+	CTXT_EGRESS,
+	CTXT_INGRESS,
+	CTXT_FLM,
+	CTXT_CNM,
+};
+
 enum {
 	SF_PAGE_SIZE = 256,           /* serial flash page size */
 	SF_SEC_SIZE = 64 * 1024,      /* serial flash sector size */
@@ -78,8 +88,10 @@ enum { MBOX_OWNER_NONE, MBOX_OWNER_FW, MBOX_OWNER_DRV };    /* mailbox owners */
 
 enum {
 	SGE_MAX_WR_LEN = 512,     /* max WR size in bytes */
+	SGE_CTXT_SIZE = 24,       /* size of SGE context */
 	SGE_NTIMERS = 6,          /* # of interrupt holdoff timer values */
 	SGE_NCOUNTERS = 4,        /* # of interrupt packet counter values */
+	SGE_NDBQTIMERS = 8,       /* # of Doorbell Queue Timer values */
 	SGE_MAX_IQ_SIZE = 65520,
 
 	SGE_TIMER_RSTRT_CNTR = 6, /* restart RX packet threshold counter */
@@ -276,4 +288,16 @@ enum {
 #define SGE_TIMESTAMP_V(x) ((__u64)(x) << SGE_TIMESTAMP_S)
 #define SGE_TIMESTAMP_G(x) (((__u64)(x) >> SGE_TIMESTAMP_S) & SGE_TIMESTAMP_M)
 
+#define I2C_DEV_ADDR_A0		0xa0
+#define I2C_DEV_ADDR_A2		0xa2
+#define I2C_PAGE_SIZE		0x100
+#define SFP_DIAG_TYPE_ADDR	0x5c
+#define SFP_DIAG_TYPE_LEN	0x1
+#define SFP_DIAG_ADDRMODE	BIT(2)
+#define SFP_DIAG_IMPLEMENTED	BIT(6)
+#define SFF_8472_COMP_ADDR	0x5e
+#define SFF_8472_COMP_LEN	0x1
+#define SFF_REV_ADDR		0x1
+#define SFF_REV_LEN		0x1
+
 #endif /* __T4_HW_H */
diff --git a/drivers/net/ethernet/chelsio/cxgb4/t4_msg.h b/drivers/net/ethernet/chelsio/cxgb4/t4_msg.h
index 8098c93cd16e..fed5f93bf620 100644
--- a/drivers/net/ethernet/chelsio/cxgb4/t4_msg.h
+++ b/drivers/net/ethernet/chelsio/cxgb4/t4_msg.h
@@ -47,19 +47,24 @@ enum {
 	CPL_CLOSE_LISTSRV_REQ = 0x9,
 	CPL_ABORT_REQ         = 0xA,
 	CPL_ABORT_RPL         = 0xB,
+	CPL_TX_DATA           = 0xC,
 	CPL_RX_DATA_ACK       = 0xD,
 	CPL_TX_PKT            = 0xE,
 	CPL_L2T_WRITE_REQ     = 0x12,
+	CPL_SMT_WRITE_REQ     = 0x14,
 	CPL_TID_RELEASE       = 0x1A,
+	CPL_SRQ_TABLE_REQ     = 0x1C,
 	CPL_TX_DATA_ISO	      = 0x1F,
 
 	CPL_CLOSE_LISTSRV_RPL = 0x20,
+	CPL_GET_TCB_RPL       = 0x22,
 	CPL_L2T_WRITE_RPL     = 0x23,
 	CPL_PASS_OPEN_RPL     = 0x24,
 	CPL_ACT_OPEN_RPL      = 0x25,
 	CPL_PEER_CLOSE        = 0x26,
 	CPL_ABORT_REQ_RSS     = 0x2B,
 	CPL_ABORT_RPL_RSS     = 0x2D,
+	CPL_SMT_WRITE_RPL     = 0x2E,
 
 	CPL_RX_PHYS_ADDR      = 0x30,
 	CPL_CLOSE_CON_RPL     = 0x32,
@@ -79,31 +84,37 @@ enum {
 	CPL_RX_ISCSI_CMP      = 0x45,
 	CPL_TRACE_PKT_T5      = 0x48,
 	CPL_RX_ISCSI_DDP      = 0x49,
+	CPL_RX_TLS_CMP        = 0x4E,
 
 	CPL_RDMA_READ_REQ     = 0x60,
 
 	CPL_PASS_OPEN_REQ6    = 0x81,
 	CPL_ACT_OPEN_REQ6     = 0x83,
 
-	CPL_TX_TLS_PDU     =    0x88,
+	CPL_TX_TLS_PDU        = 0x88,
+	CPL_TX_TLS_SFO        = 0x89,
 	CPL_TX_SEC_PDU        = 0x8A,
 	CPL_TX_TLS_ACK        = 0x8B,
 
 	CPL_RDMA_TERMINATE    = 0xA2,
 	CPL_RDMA_WRITE        = 0xA4,
 	CPL_SGE_EGR_UPDATE    = 0xA5,
+	CPL_RX_MPS_PKT        = 0xAF,
 
 	CPL_TRACE_PKT         = 0xB0,
+	CPL_TLS_DATA          = 0xB1,
 	CPL_ISCSI_DATA	      = 0xB2,
 
 	CPL_FW4_MSG           = 0xC0,
 	CPL_FW4_PLD           = 0xC1,
 	CPL_FW4_ACK           = 0xC3,
+	CPL_SRQ_TABLE_RPL     = 0xCC,
 
 	CPL_RX_PHYS_DSGL      = 0xD0,
 
 	CPL_FW6_MSG           = 0xE0,
 	CPL_FW6_PLD           = 0xE1,
+	CPL_TX_TNL_LSO        = 0xEC,
 	CPL_TX_PKT_LSO        = 0xED,
 	CPL_TX_PKT_XT         = 0xEE,
 
@@ -132,6 +143,8 @@ enum CPL_error {
 	CPL_ERR_KEEPALV_NEG_ADVICE = 37,
 	CPL_ERR_ABORT_FAILED       = 42,
 	CPL_ERR_IWARP_FLM          = 50,
+	CPL_CONTAINS_READ_RPL      = 60,
+	CPL_CONTAINS_WRITE_RPL     = 61,
 };
 
 enum {
@@ -147,6 +160,7 @@ enum {
 	ULP_MODE_RDMA          = 4,
 	ULP_MODE_TCPDDP	       = 5,
 	ULP_MODE_FCOE          = 6,
+	ULP_MODE_TLS           = 8,
 };
 
 enum {
@@ -194,6 +208,7 @@ union opcode_tid {
 /* partitioning of TID fields that also carry a queue id */
 #define TID_TID_S    0
 #define TID_TID_M    0x3fff
+#define TID_TID_V(x) ((x) << TID_TID_S)
 #define TID_TID_G(x) (((x) >> TID_TID_S) & TID_TID_M)
 
 #define TID_QID_S    14
@@ -283,6 +298,7 @@ struct work_request_hdr {
 
 #define RX_CHANNEL_S    26
 #define RX_CHANNEL_V(x) ((x) << RX_CHANNEL_S)
+#define RX_CHANNEL_F	RX_CHANNEL_V(1U)
 
 #define WND_SCALE_EN_S    28
 #define WND_SCALE_EN_V(x) ((x) << WND_SCALE_EN_S)
@@ -312,6 +328,10 @@ struct cpl_pass_open_req {
 #define DELACK_V(x) ((x) << DELACK_S)
 #define DELACK_F    DELACK_V(1U)
 
+#define NON_OFFLOAD_S		7
+#define NON_OFFLOAD_V(x)	((x) << NON_OFFLOAD_S)
+#define NON_OFFLOAD_F		NON_OFFLOAD_V(1U)
+
 #define DSCP_S    22
 #define DSCP_M    0x3F
 #define DSCP_V(x) ((x) << DSCP_S)
@@ -670,6 +690,13 @@ struct cpl_get_tcb {
 #define NO_REPLY_V(x) ((x) << NO_REPLY_S)
 #define NO_REPLY_F    NO_REPLY_V(1U)
 
+struct cpl_get_tcb_rpl {
+	union opcode_tid ot;
+	__u8 cookie;
+	__u8 status;
+	__be16 len;
+};
+
 struct cpl_set_tcb_field {
 	WR_HDR;
 	union opcode_tid ot;
@@ -679,9 +706,17 @@ struct cpl_set_tcb_field {
 	__be64 val;
 };
 
+struct cpl_set_tcb_field_core {
+	union opcode_tid ot;
+	__be16 reply_ctrl;
+	__be16 word_cookie;
+	__be64 mask;
+	__be64 val;
+};
+
 /* cpl_set_tcb_field.word_cookie fields */
-#define TCB_WORD_S    0
-#define TCB_WORD(x)   ((x) << TCB_WORD_S)
+#define TCB_WORD_S	0
+#define TCB_WORD_V(x)	((x) << TCB_WORD_S)
 
 #define TCB_COOKIE_S    5
 #define TCB_COOKIE_M    0x7
@@ -734,6 +769,21 @@ struct cpl_abort_req_rss {
 	u8 status;
 };
 
+struct cpl_abort_req_rss6 {
+	union opcode_tid ot;
+	__be32 srqidx_status;
+};
+
+#define ABORT_RSS_STATUS_S    0
+#define ABORT_RSS_STATUS_M    0xff
+#define ABORT_RSS_STATUS_V(x) ((x) << ABORT_RSS_STATUS_S)
+#define ABORT_RSS_STATUS_G(x) (((x) >> ABORT_RSS_STATUS_S) & ABORT_RSS_STATUS_M)
+
+#define ABORT_RSS_SRQIDX_S    8
+#define ABORT_RSS_SRQIDX_M    0xffffff
+#define ABORT_RSS_SRQIDX_V(x) ((x) << ABORT_RSS_SRQIDX_S)
+#define ABORT_RSS_SRQIDX_G(x) (((x) >> ABORT_RSS_SRQIDX_S) & ABORT_RSS_SRQIDX_M)
+
 struct cpl_abort_req {
 	WR_HDR;
 	union opcode_tid ot;
@@ -749,6 +799,11 @@ struct cpl_abort_rpl_rss {
 	u8 status;
 };
 
+struct cpl_abort_rpl_rss6 {
+	union opcode_tid ot;
+	__be32 srqidx_status;
+};
+
 struct cpl_abort_rpl {
 	WR_HDR;
 	union opcode_tid ot;
@@ -807,6 +862,10 @@ struct cpl_tx_pkt {
 #define TXPKT_INS_OVLAN_V(x) ((x) << TXPKT_INS_OVLAN_S)
 #define TXPKT_INS_OVLAN_F    TXPKT_INS_OVLAN_V(1U)
 
+#define TXPKT_TSTAMP_S    23
+#define TXPKT_TSTAMP_V(x) ((x) << TXPKT_TSTAMP_S)
+#define TXPKT_TSTAMP_F    TXPKT_TSTAMP_V(1ULL)
+
 #define TXPKT_OPCODE_S    24
 #define TXPKT_OPCODE_V(x) ((x) << TXPKT_OPCODE_S)
 
@@ -1190,6 +1249,11 @@ struct cpl_rx_pkt {
 #define T6_COMPR_RXERR_SUM_V(x) ((x) << T6_COMPR_RXERR_SUM_S)
 #define T6_COMPR_RXERR_SUM_F    T6_COMPR_RXERR_SUM_V(1U)
 
+#define T6_RX_TNLHDR_LEN_S    8
+#define T6_RX_TNLHDR_LEN_M    0xFF
+#define T6_RX_TNLHDR_LEN_V(x) ((x) << T6_RX_TNLHDR_LEN_S)
+#define T6_RX_TNLHDR_LEN_G(x) (((x) >> T6_RX_TNLHDR_LEN_S) & T6_RX_TNLHDR_LEN_M)
+
 struct cpl_trace_pkt {
 	u8 opcode;
 	u8 intf;
@@ -1261,6 +1325,44 @@ struct cpl_l2t_write_rpl {
 	u8 rsvd[3];
 };
 
+struct cpl_smt_write_req {
+	WR_HDR;
+	union opcode_tid ot;
+	__be32 params;
+	__be16 pfvf1;
+	u8 src_mac1[6];
+	__be16 pfvf0;
+	u8 src_mac0[6];
+};
+
+struct cpl_t6_smt_write_req {
+	WR_HDR;
+	union opcode_tid ot;
+	__be32 params;
+	__be64 tag;
+	__be16 pfvf0;
+	u8 src_mac0[6];
+	__be32 local_ip;
+	__be32 rsvd;
+};
+
+struct cpl_smt_write_rpl {
+	union opcode_tid ot;
+	u8 status;
+	u8 rsvd[3];
+};
+
+/* cpl_smt_{read,write}_req.params fields */
+#define SMTW_OVLAN_IDX_S	16
+#define SMTW_OVLAN_IDX_V(x)	((x) << SMTW_OVLAN_IDX_S)
+
+#define SMTW_IDX_S	20
+#define SMTW_IDX_V(x)	((x) << SMTW_IDX_S)
+
+#define SMTW_NORPL_S	31
+#define SMTW_NORPL_V(x)	((x) << SMTW_NORPL_S)
+#define SMTW_NORPL_F	SMTW_NORPL_V(1U)
+
 struct cpl_rdma_terminate {
 	union opcode_tid ot;
 	__be16 rsvd;
@@ -1328,6 +1430,11 @@ enum {
 	CPL_FW4_ACK_FLAGS_FLOWC		= 0x4,	/* fw_flowc_wr complete */
 };
 
+#define CPL_FW4_ACK_FLOWID_S    0
+#define CPL_FW4_ACK_FLOWID_M    0xffffff
+#define CPL_FW4_ACK_FLOWID_G(x) \
+	(((x) >> CPL_FW4_ACK_FLOWID_S) & CPL_FW4_ACK_FLOWID_M)
+
 struct cpl_fw6_msg {
 	u8 opcode;
 	u8 type;
@@ -1364,10 +1471,40 @@ struct cpl_tx_data {
 #define TX_FORCE_S	13
 #define TX_FORCE_V(x)	((x) << TX_FORCE_S)
 
+#define TX_DATA_MSS_S    16
+#define TX_DATA_MSS_M    0xFFFF
+#define TX_DATA_MSS_V(x) ((x) << TX_DATA_MSS_S)
+#define TX_DATA_MSS_G(x) (((x) >> TX_DATA_MSS_S) & TX_DATA_MSS_M)
+
+#define TX_LENGTH_S    0
+#define TX_LENGTH_M    0xFFFF
+#define TX_LENGTH_V(x) ((x) << TX_LENGTH_S)
+#define TX_LENGTH_G(x) (((x) >> TX_LENGTH_S) & TX_LENGTH_M)
+
 #define T6_TX_FORCE_S		20
 #define T6_TX_FORCE_V(x)	((x) << T6_TX_FORCE_S)
 #define T6_TX_FORCE_F		T6_TX_FORCE_V(1U)
 
+#define TX_URG_S    16
+#define TX_URG_V(x) ((x) << TX_URG_S)
+
+#define TX_SHOVE_S    14
+#define TX_SHOVE_V(x) ((x) << TX_SHOVE_S)
+#define TX_SHOVE_F    TX_SHOVE_V(1U)
+
+#define TX_BYPASS_S    21
+#define TX_BYPASS_V(x) ((x) << TX_BYPASS_S)
+#define TX_BYPASS_F    TX_BYPASS_V(1U)
+
+#define TX_PUSH_S    22
+#define TX_PUSH_V(x) ((x) << TX_PUSH_S)
+#define TX_PUSH_F    TX_PUSH_V(1U)
+
+#define TX_ULP_MODE_S    10
+#define TX_ULP_MODE_M    0x7
+#define TX_ULP_MODE_V(x) ((x) << TX_ULP_MODE_S)
+#define TX_ULP_MODE_G(x) (((x) >> TX_ULP_MODE_S) & TX_ULP_MODE_M)
+
 enum {
 	ULP_TX_MEM_READ = 2,
 	ULP_TX_MEM_WRITE = 3,
@@ -1378,12 +1515,21 @@ enum {
 	ULP_TX_SC_NOOP = 0x80,
 	ULP_TX_SC_IMM  = 0x81,
 	ULP_TX_SC_DSGL = 0x82,
-	ULP_TX_SC_ISGL = 0x83
+	ULP_TX_SC_ISGL = 0x83,
+	ULP_TX_SC_MEMRD = 0x86
 };
 
 #define ULPTX_CMD_S    24
 #define ULPTX_CMD_V(x) ((x) << ULPTX_CMD_S)
 
+#define ULPTX_LEN16_S    0
+#define ULPTX_LEN16_M    0xFF
+#define ULPTX_LEN16_V(x) ((x) << ULPTX_LEN16_S)
+
+#define ULP_TX_SC_MORE_S 23
+#define ULP_TX_SC_MORE_V(x) ((x) << ULP_TX_SC_MORE_S)
+#define ULP_TX_SC_MORE_F  ULP_TX_SC_MORE_V(1U)
+
 struct ulptx_sge_pair {
 	__be32 len[2];
 	__be64 addr[2];
@@ -1393,7 +1539,7 @@ struct ulptx_sgl {
 	__be32 cmd_nsge;
 	__be32 len0;
 	__be64 addr0;
-	struct ulptx_sge_pair sge[0];
+	struct ulptx_sge_pair sge[];
 };
 
 struct ulptx_idata {
@@ -1429,6 +1575,169 @@ struct ulp_txpkt {
 #define ULP_TXPKT_RO_V(x) ((x) << ULP_TXPKT_RO_S)
 #define ULP_TXPKT_RO_F ULP_TXPKT_RO_V(1U)
 
+enum cpl_tx_tnl_lso_type {
+	TX_TNL_TYPE_OPAQUE,
+	TX_TNL_TYPE_NVGRE,
+	TX_TNL_TYPE_VXLAN,
+	TX_TNL_TYPE_GENEVE,
+};
+
+struct cpl_tx_tnl_lso {
+	__be32 op_to_IpIdSplitOut;
+	__be16 IpIdOffsetOut;
+	__be16 UdpLenSetOut_to_TnlHdrLen;
+	__be64 r1;
+	__be32 Flow_to_TcpHdrLen;
+	__be16 IpIdOffset;
+	__be16 IpIdSplit_to_Mss;
+	__be32 TCPSeqOffset;
+	__be32 EthLenOffset_Size;
+	/* encapsulated CPL (TX_PKT_XT) follows here */
+};
+
+#define CPL_TX_TNL_LSO_OPCODE_S		24
+#define CPL_TX_TNL_LSO_OPCODE_M		0xff
+#define CPL_TX_TNL_LSO_OPCODE_V(x)      ((x) << CPL_TX_TNL_LSO_OPCODE_S)
+#define CPL_TX_TNL_LSO_OPCODE_G(x)      \
+	(((x) >> CPL_TX_TNL_LSO_OPCODE_S) & CPL_TX_TNL_LSO_OPCODE_M)
+
+#define CPL_TX_TNL_LSO_FIRST_S		23
+#define CPL_TX_TNL_LSO_FIRST_M		0x1
+#define CPL_TX_TNL_LSO_FIRST_V(x)	((x) << CPL_TX_TNL_LSO_FIRST_S)
+#define CPL_TX_TNL_LSO_FIRST_G(x)	\
+	(((x) >> CPL_TX_TNL_LSO_FIRST_S) & CPL_TX_TNL_LSO_FIRST_M)
+#define CPL_TX_TNL_LSO_FIRST_F		CPL_TX_TNL_LSO_FIRST_V(1U)
+
+#define CPL_TX_TNL_LSO_LAST_S		22
+#define CPL_TX_TNL_LSO_LAST_M		0x1
+#define CPL_TX_TNL_LSO_LAST_V(x)	((x) << CPL_TX_TNL_LSO_LAST_S)
+#define CPL_TX_TNL_LSO_LAST_G(x)	\
+	(((x) >> CPL_TX_TNL_LSO_LAST_S) & CPL_TX_TNL_LSO_LAST_M)
+#define CPL_TX_TNL_LSO_LAST_F		CPL_TX_TNL_LSO_LAST_V(1U)
+
+#define CPL_TX_TNL_LSO_ETHHDRLENXOUT_S	21
+#define CPL_TX_TNL_LSO_ETHHDRLENXOUT_M	0x1
+#define CPL_TX_TNL_LSO_ETHHDRLENXOUT_V(x) \
+	((x) << CPL_TX_TNL_LSO_ETHHDRLENXOUT_S)
+#define CPL_TX_TNL_LSO_ETHHDRLENXOUT_G(x) \
+	(((x) >> CPL_TX_TNL_LSO_ETHHDRLENXOUT_S) & \
+	 CPL_TX_TNL_LSO_ETHHDRLENXOUT_M)
+#define CPL_TX_TNL_LSO_ETHHDRLENXOUT_F CPL_TX_TNL_LSO_ETHHDRLENXOUT_V(1U)
+
+#define CPL_TX_TNL_LSO_IPV6OUT_S	20
+#define CPL_TX_TNL_LSO_IPV6OUT_M	0x1
+#define CPL_TX_TNL_LSO_IPV6OUT_V(x)	((x) << CPL_TX_TNL_LSO_IPV6OUT_S)
+#define CPL_TX_TNL_LSO_IPV6OUT_G(x)	\
+	(((x) >> CPL_TX_TNL_LSO_IPV6OUT_S) & CPL_TX_TNL_LSO_IPV6OUT_M)
+#define CPL_TX_TNL_LSO_IPV6OUT_F        CPL_TX_TNL_LSO_IPV6OUT_V(1U)
+
+#define CPL_TX_TNL_LSO_ETHHDRLEN_S	16
+#define CPL_TX_TNL_LSO_ETHHDRLEN_M	0xf
+#define CPL_TX_TNL_LSO_ETHHDRLEN_V(x)	((x) << CPL_TX_TNL_LSO_ETHHDRLEN_S)
+#define CPL_TX_TNL_LSO_ETHHDRLEN_G(x)	\
+	(((x) >> CPL_TX_TNL_LSO_ETHHDRLEN_S) & CPL_TX_TNL_LSO_ETHHDRLEN_M)
+
+#define CPL_TX_TNL_LSO_IPHDRLEN_S	4
+#define CPL_TX_TNL_LSO_IPHDRLEN_M	0xfff
+#define CPL_TX_TNL_LSO_IPHDRLEN_V(x)	((x) << CPL_TX_TNL_LSO_IPHDRLEN_S)
+#define CPL_TX_TNL_LSO_IPHDRLEN_G(x)    \
+	(((x) >> CPL_TX_TNL_LSO_IPHDRLEN_S) & CPL_TX_TNL_LSO_IPHDRLEN_M)
+
+#define CPL_TX_TNL_LSO_TCPHDRLEN_S	0
+#define CPL_TX_TNL_LSO_TCPHDRLEN_M	0xf
+#define CPL_TX_TNL_LSO_TCPHDRLEN_V(x)	((x) << CPL_TX_TNL_LSO_TCPHDRLEN_S)
+#define CPL_TX_TNL_LSO_TCPHDRLEN_G(x)   \
+	(((x) >> CPL_TX_TNL_LSO_TCPHDRLEN_S) & CPL_TX_TNL_LSO_TCPHDRLEN_M)
+
+#define CPL_TX_TNL_LSO_MSS_S            0
+#define CPL_TX_TNL_LSO_MSS_M            0x3fff
+#define CPL_TX_TNL_LSO_MSS_V(x)         ((x) << CPL_TX_TNL_LSO_MSS_S)
+#define CPL_TX_TNL_LSO_MSS_G(x)         \
+	(((x) >> CPL_TX_TNL_LSO_MSS_S) & CPL_TX_TNL_LSO_MSS_M)
+
+#define CPL_TX_TNL_LSO_SIZE_S		0
+#define CPL_TX_TNL_LSO_SIZE_M		0xfffffff
+#define CPL_TX_TNL_LSO_SIZE_V(x)	((x) << CPL_TX_TNL_LSO_SIZE_S)
+#define CPL_TX_TNL_LSO_SIZE_G(x)	\
+	(((x) >> CPL_TX_TNL_LSO_SIZE_S) & CPL_TX_TNL_LSO_SIZE_M)
+
+#define CPL_TX_TNL_LSO_ETHHDRLENOUT_S   16
+#define CPL_TX_TNL_LSO_ETHHDRLENOUT_M   0xf
+#define CPL_TX_TNL_LSO_ETHHDRLENOUT_V(x) \
+	((x) << CPL_TX_TNL_LSO_ETHHDRLENOUT_S)
+#define CPL_TX_TNL_LSO_ETHHDRLENOUT_G(x) \
+	(((x) >> CPL_TX_TNL_LSO_ETHHDRLENOUT_S) & CPL_TX_TNL_LSO_ETHHDRLENOUT_M)
+
+#define CPL_TX_TNL_LSO_IPHDRLENOUT_S    4
+#define CPL_TX_TNL_LSO_IPHDRLENOUT_M    0xfff
+#define CPL_TX_TNL_LSO_IPHDRLENOUT_V(x) ((x) << CPL_TX_TNL_LSO_IPHDRLENOUT_S)
+#define CPL_TX_TNL_LSO_IPHDRLENOUT_G(x) \
+	(((x) >> CPL_TX_TNL_LSO_IPHDRLENOUT_S) & CPL_TX_TNL_LSO_IPHDRLENOUT_M)
+
+#define CPL_TX_TNL_LSO_IPHDRCHKOUT_S    3
+#define CPL_TX_TNL_LSO_IPHDRCHKOUT_M    0x1
+#define CPL_TX_TNL_LSO_IPHDRCHKOUT_V(x) ((x) << CPL_TX_TNL_LSO_IPHDRCHKOUT_S)
+#define CPL_TX_TNL_LSO_IPHDRCHKOUT_G(x) \
+	(((x) >> CPL_TX_TNL_LSO_IPHDRCHKOUT_S) & CPL_TX_TNL_LSO_IPHDRCHKOUT_M)
+#define CPL_TX_TNL_LSO_IPHDRCHKOUT_F    CPL_TX_TNL_LSO_IPHDRCHKOUT_V(1U)
+
+#define CPL_TX_TNL_LSO_IPLENSETOUT_S	2
+#define CPL_TX_TNL_LSO_IPLENSETOUT_M	0x1
+#define CPL_TX_TNL_LSO_IPLENSETOUT_V(x) ((x) << CPL_TX_TNL_LSO_IPLENSETOUT_S)
+#define CPL_TX_TNL_LSO_IPLENSETOUT_G(x) \
+	(((x) >> CPL_TX_TNL_LSO_IPLENSETOUT_S) & CPL_TX_TNL_LSO_IPLENSETOUT_M)
+#define CPL_TX_TNL_LSO_IPLENSETOUT_F	CPL_TX_TNL_LSO_IPLENSETOUT_V(1U)
+
+#define CPL_TX_TNL_LSO_IPIDINCOUT_S	1
+#define CPL_TX_TNL_LSO_IPIDINCOUT_M	0x1
+#define CPL_TX_TNL_LSO_IPIDINCOUT_V(x)  ((x) << CPL_TX_TNL_LSO_IPIDINCOUT_S)
+#define CPL_TX_TNL_LSO_IPIDINCOUT_G(x)  \
+	(((x) >> CPL_TX_TNL_LSO_IPIDINCOUT_S) & CPL_TX_TNL_LSO_IPIDINCOUT_M)
+#define CPL_TX_TNL_LSO_IPIDINCOUT_F     CPL_TX_TNL_LSO_IPIDINCOUT_V(1U)
+
+#define CPL_TX_TNL_LSO_UDPCHKCLROUT_S   14
+#define CPL_TX_TNL_LSO_UDPCHKCLROUT_M   0x1
+#define CPL_TX_TNL_LSO_UDPCHKCLROUT_V(x) \
+	((x) << CPL_TX_TNL_LSO_UDPCHKCLROUT_S)
+#define CPL_TX_TNL_LSO_UDPCHKCLROUT_G(x) \
+	(((x) >> CPL_TX_TNL_LSO_UDPCHKCLROUT_S) & \
+	 CPL_TX_TNL_LSO_UDPCHKCLROUT_M)
+#define CPL_TX_TNL_LSO_UDPCHKCLROUT_F   CPL_TX_TNL_LSO_UDPCHKCLROUT_V(1U)
+
+#define CPL_TX_TNL_LSO_UDPLENSETOUT_S   15
+#define CPL_TX_TNL_LSO_UDPLENSETOUT_M   0x1
+#define CPL_TX_TNL_LSO_UDPLENSETOUT_V(x) \
+	((x) << CPL_TX_TNL_LSO_UDPLENSETOUT_S)
+#define CPL_TX_TNL_LSO_UDPLENSETOUT_G(x) \
+	(((x) >> CPL_TX_TNL_LSO_UDPLENSETOUT_S) & \
+	 CPL_TX_TNL_LSO_UDPLENSETOUT_M)
+#define CPL_TX_TNL_LSO_UDPLENSETOUT_F   CPL_TX_TNL_LSO_UDPLENSETOUT_V(1U)
+
+#define CPL_TX_TNL_LSO_TNLTYPE_S	12
+#define CPL_TX_TNL_LSO_TNLTYPE_M	0x3
+#define CPL_TX_TNL_LSO_TNLTYPE_V(x)	((x) << CPL_TX_TNL_LSO_TNLTYPE_S)
+#define CPL_TX_TNL_LSO_TNLTYPE_G(x)	\
+	(((x) >> CPL_TX_TNL_LSO_TNLTYPE_S) & CPL_TX_TNL_LSO_TNLTYPE_M)
+
+#define S_CPL_TX_TNL_LSO_ETHHDRLEN	16
+#define M_CPL_TX_TNL_LSO_ETHHDRLEN	0xf
+#define V_CPL_TX_TNL_LSO_ETHHDRLEN(x)	((x) << S_CPL_TX_TNL_LSO_ETHHDRLEN)
+#define G_CPL_TX_TNL_LSO_ETHHDRLEN(x)	\
+	(((x) >> S_CPL_TX_TNL_LSO_ETHHDRLEN) & M_CPL_TX_TNL_LSO_ETHHDRLEN)
+
+#define CPL_TX_TNL_LSO_TNLHDRLEN_S      0
+#define CPL_TX_TNL_LSO_TNLHDRLEN_M      0xfff
+#define CPL_TX_TNL_LSO_TNLHDRLEN_V(x)	((x) << CPL_TX_TNL_LSO_TNLHDRLEN_S)
+#define CPL_TX_TNL_LSO_TNLHDRLEN_G(x)   \
+	(((x) >> CPL_TX_TNL_LSO_TNLHDRLEN_S) & CPL_TX_TNL_LSO_TNLHDRLEN_M)
+
+#define CPL_TX_TNL_LSO_IPV6_S		20
+#define CPL_TX_TNL_LSO_IPV6_M		0x1
+#define CPL_TX_TNL_LSO_IPV6_V(x)	((x) << CPL_TX_TNL_LSO_IPV6_S)
+#define CPL_TX_TNL_LSO_IPV6_G(x)	\
+	(((x) >> CPL_TX_TNL_LSO_IPV6_S) & CPL_TX_TNL_LSO_IPV6_M)
+#define CPL_TX_TNL_LSO_IPV6_F		CPL_TX_TNL_LSO_IPV6_V(1U)
+
 #define ULP_TX_SC_MORE_S 23
 #define ULP_TX_SC_MORE_V(x) ((x) << ULP_TX_SC_MORE_S)
 #define ULP_TX_SC_MORE_F  ULP_TX_SC_MORE_V(1U)
@@ -1875,4 +2184,169 @@ struct cpl_rx_phys_dsgl {
 	(((x) >> CPL_RX_PHYS_DSGL_NOOFSGENTR_S) & \
 	 CPL_RX_PHYS_DSGL_NOOFSGENTR_M)
 
+struct cpl_rx_mps_pkt {
+	__be32 op_to_r1_hi;
+	__be32 r1_lo_length;
+};
+
+#define CPL_RX_MPS_PKT_OP_S     24
+#define CPL_RX_MPS_PKT_OP_M     0xff
+#define CPL_RX_MPS_PKT_OP_V(x)  ((x) << CPL_RX_MPS_PKT_OP_S)
+#define CPL_RX_MPS_PKT_OP_G(x)  \
+	(((x) >> CPL_RX_MPS_PKT_OP_S) & CPL_RX_MPS_PKT_OP_M)
+
+#define CPL_RX_MPS_PKT_TYPE_S           20
+#define CPL_RX_MPS_PKT_TYPE_M           0xf
+#define CPL_RX_MPS_PKT_TYPE_V(x)        ((x) << CPL_RX_MPS_PKT_TYPE_S)
+#define CPL_RX_MPS_PKT_TYPE_G(x)        \
+	(((x) >> CPL_RX_MPS_PKT_TYPE_S) & CPL_RX_MPS_PKT_TYPE_M)
+
+enum {
+	X_CPL_RX_MPS_PKT_TYPE_PAUSE = 1 << 0,
+	X_CPL_RX_MPS_PKT_TYPE_PPP   = 1 << 1,
+	X_CPL_RX_MPS_PKT_TYPE_QFC   = 1 << 2,
+	X_CPL_RX_MPS_PKT_TYPE_PTP   = 1 << 3
+};
+
+struct cpl_srq_table_req {
+	WR_HDR;
+	union opcode_tid ot;
+	__u8 status;
+	__u8 rsvd[2];
+	__u8 idx;
+	__be64 rsvd_pdid;
+	__be32 qlen_qbase;
+	__be16 cur_msn;
+	__be16 max_msn;
+};
+
+struct cpl_srq_table_rpl {
+	union opcode_tid ot;
+	__u8 status;
+	__u8 rsvd[2];
+	__u8 idx;
+	__be64 rsvd_pdid;
+	__be32 qlen_qbase;
+	__be16 cur_msn;
+	__be16 max_msn;
+};
+
+/* cpl_srq_table_{req,rpl}.params fields */
+#define SRQT_QLEN_S   28
+#define SRQT_QLEN_M   0xF
+#define SRQT_QLEN_V(x) ((x) << SRQT_QLEN_S)
+#define SRQT_QLEN_G(x) (((x) >> SRQT_QLEN_S) & SRQT_QLEN_M)
+
+#define SRQT_QBASE_S    0
+#define SRQT_QBASE_M   0x3FFFFFF
+#define SRQT_QBASE_V(x) ((x) << SRQT_QBASE_S)
+#define SRQT_QBASE_G(x) (((x) >> SRQT_QBASE_S) & SRQT_QBASE_M)
+
+#define SRQT_PDID_S    0
+#define SRQT_PDID_M   0xFF
+#define SRQT_PDID_V(x) ((x) << SRQT_PDID_S)
+#define SRQT_PDID_G(x) (((x) >> SRQT_PDID_S) & SRQT_PDID_M)
+
+#define SRQT_IDX_S    0
+#define SRQT_IDX_M    0xF
+#define SRQT_IDX_V(x) ((x) << SRQT_IDX_S)
+#define SRQT_IDX_G(x) (((x) >> SRQT_IDX_S) & SRQT_IDX_M)
+
+struct cpl_tx_tls_sfo {
+	__be32 op_to_seg_len;
+	__be32 pld_len;
+	__be32 type_protover;
+	__be32 r1_lo;
+	__be32 seqno_numivs;
+	__be32 ivgen_hdrlen;
+	__be64 scmd1;
+};
+
+/* cpl_tx_tls_sfo macros */
+#define CPL_TX_TLS_SFO_OPCODE_S         24
+#define CPL_TX_TLS_SFO_OPCODE_V(x)      ((x) << CPL_TX_TLS_SFO_OPCODE_S)
+
+#define CPL_TX_TLS_SFO_DATA_TYPE_S      20
+#define CPL_TX_TLS_SFO_DATA_TYPE_V(x)   ((x) << CPL_TX_TLS_SFO_DATA_TYPE_S)
+
+#define CPL_TX_TLS_SFO_CPL_LEN_S        16
+#define CPL_TX_TLS_SFO_CPL_LEN_V(x)     ((x) << CPL_TX_TLS_SFO_CPL_LEN_S)
+
+#define CPL_TX_TLS_SFO_SEG_LEN_S        0
+#define CPL_TX_TLS_SFO_SEG_LEN_M        0xffff
+#define CPL_TX_TLS_SFO_SEG_LEN_V(x)     ((x) << CPL_TX_TLS_SFO_SEG_LEN_S)
+#define CPL_TX_TLS_SFO_SEG_LEN_G(x)     \
+	(((x) >> CPL_TX_TLS_SFO_SEG_LEN_S) & CPL_TX_TLS_SFO_SEG_LEN_M)
+
+#define CPL_TX_TLS_SFO_TYPE_S           24
+#define CPL_TX_TLS_SFO_TYPE_M           0xff
+#define CPL_TX_TLS_SFO_TYPE_V(x)        ((x) << CPL_TX_TLS_SFO_TYPE_S)
+#define CPL_TX_TLS_SFO_TYPE_G(x)        \
+	(((x) >> CPL_TX_TLS_SFO_TYPE_S) & CPL_TX_TLS_SFO_TYPE_M)
+
+#define CPL_TX_TLS_SFO_PROTOVER_S       8
+#define CPL_TX_TLS_SFO_PROTOVER_M       0xffff
+#define CPL_TX_TLS_SFO_PROTOVER_V(x)    ((x) << CPL_TX_TLS_SFO_PROTOVER_S)
+#define CPL_TX_TLS_SFO_PROTOVER_G(x)    \
+	(((x) >> CPL_TX_TLS_SFO_PROTOVER_S) & CPL_TX_TLS_SFO_PROTOVER_M)
+
+struct cpl_tls_data {
+	struct rss_header rsshdr;
+	union opcode_tid ot;
+	__be32 length_pkd;
+	__be32 seq;
+	__be32 r1;
+};
+
+#define CPL_TLS_DATA_OPCODE_S           24
+#define CPL_TLS_DATA_OPCODE_M           0xff
+#define CPL_TLS_DATA_OPCODE_V(x)        ((x) << CPL_TLS_DATA_OPCODE_S)
+#define CPL_TLS_DATA_OPCODE_G(x)        \
+	(((x) >> CPL_TLS_DATA_OPCODE_S) & CPL_TLS_DATA_OPCODE_M)
+
+#define CPL_TLS_DATA_TID_S              0
+#define CPL_TLS_DATA_TID_M              0xffffff
+#define CPL_TLS_DATA_TID_V(x)           ((x) << CPL_TLS_DATA_TID_S)
+#define CPL_TLS_DATA_TID_G(x)           \
+	(((x) >> CPL_TLS_DATA_TID_S) & CPL_TLS_DATA_TID_M)
+
+#define CPL_TLS_DATA_LENGTH_S           0
+#define CPL_TLS_DATA_LENGTH_M           0xffff
+#define CPL_TLS_DATA_LENGTH_V(x)        ((x) << CPL_TLS_DATA_LENGTH_S)
+#define CPL_TLS_DATA_LENGTH_G(x)        \
+	(((x) >> CPL_TLS_DATA_LENGTH_S) & CPL_TLS_DATA_LENGTH_M)
+
+struct cpl_rx_tls_cmp {
+	struct rss_header rsshdr;
+	union opcode_tid ot;
+	__be32 pdulength_length;
+	__be32 seq;
+	__be32 ddp_report;
+	__be32 r;
+	__be32 ddp_valid;
+};
+
+#define CPL_RX_TLS_CMP_OPCODE_S         24
+#define CPL_RX_TLS_CMP_OPCODE_M         0xff
+#define CPL_RX_TLS_CMP_OPCODE_V(x)      ((x) << CPL_RX_TLS_CMP_OPCODE_S)
+#define CPL_RX_TLS_CMP_OPCODE_G(x)      \
+	(((x) >> CPL_RX_TLS_CMP_OPCODE_S) & CPL_RX_TLS_CMP_OPCODE_M)
+
+#define CPL_RX_TLS_CMP_TID_S            0
+#define CPL_RX_TLS_CMP_TID_M            0xffffff
+#define CPL_RX_TLS_CMP_TID_V(x)         ((x) << CPL_RX_TLS_CMP_TID_S)
+#define CPL_RX_TLS_CMP_TID_G(x)         \
+	(((x) >> CPL_RX_TLS_CMP_TID_S) & CPL_RX_TLS_CMP_TID_M)
+
+#define CPL_RX_TLS_CMP_PDULENGTH_S      16
+#define CPL_RX_TLS_CMP_PDULENGTH_M      0xffff
+#define CPL_RX_TLS_CMP_PDULENGTH_V(x)   ((x) << CPL_RX_TLS_CMP_PDULENGTH_S)
+#define CPL_RX_TLS_CMP_PDULENGTH_G(x)   \
+	(((x) >> CPL_RX_TLS_CMP_PDULENGTH_S) & CPL_RX_TLS_CMP_PDULENGTH_M)
+
+#define CPL_RX_TLS_CMP_LENGTH_S         0
+#define CPL_RX_TLS_CMP_LENGTH_M         0xffff
+#define CPL_RX_TLS_CMP_LENGTH_V(x)      ((x) << CPL_RX_TLS_CMP_LENGTH_S)
+#define CPL_RX_TLS_CMP_LENGTH_G(x)      \
+	(((x) >> CPL_RX_TLS_CMP_LENGTH_S) & CPL_RX_TLS_CMP_LENGTH_M)
 #endif  /* __T4_MSG_H */
diff --git a/drivers/net/ethernet/chelsio/cxgb4/t4_pci_id_tbl.h b/drivers/net/ethernet/chelsio/cxgb4/t4_pci_id_tbl.h
index be7041f6cf71..0b1b5f9c67d4 100644
--- a/drivers/net/ethernet/chelsio/cxgb4/t4_pci_id_tbl.h
+++ b/drivers/net/ethernet/chelsio/cxgb4/t4_pci_id_tbl.h
@@ -145,6 +145,9 @@ CH_PCI_DEVICE_ID_TABLE_DEFINE_BEGIN
 	CH_PCI_ID_TABLE_FENTRY(0x5016),	/* T580-OCP-SO */
 	CH_PCI_ID_TABLE_FENTRY(0x5017),	/* T520-OCP-SO */
 	CH_PCI_ID_TABLE_FENTRY(0x5018),	/* T540-BT */
+	CH_PCI_ID_TABLE_FENTRY(0x5019),	/* T540-LP-BT */
+	CH_PCI_ID_TABLE_FENTRY(0x501a),	/* T540-SO-BT */
+	CH_PCI_ID_TABLE_FENTRY(0x501b),	/* T540-SO-CR */
 	CH_PCI_ID_TABLE_FENTRY(0x5080),	/* Custom T540-cr */
 	CH_PCI_ID_TABLE_FENTRY(0x5081),	/* Custom T540-LL-cr */
 	CH_PCI_ID_TABLE_FENTRY(0x5082),	/* Custom T504-cr */
@@ -174,6 +177,20 @@ CH_PCI_DEVICE_ID_TABLE_DEFINE_BEGIN
 	CH_PCI_ID_TABLE_FENTRY(0x50a0), /* Custom T540-CR */
 	CH_PCI_ID_TABLE_FENTRY(0x50a1), /* Custom T540-CR */
 	CH_PCI_ID_TABLE_FENTRY(0x50a2), /* Custom T540-KR4 */
+	CH_PCI_ID_TABLE_FENTRY(0x50a3), /* Custom T580-KR4 */
+	CH_PCI_ID_TABLE_FENTRY(0x50a4), /* Custom 2x T540-CR */
+	CH_PCI_ID_TABLE_FENTRY(0x50a5), /* Custom T522-BT */
+	CH_PCI_ID_TABLE_FENTRY(0x50a6), /* Custom T522-BT-SO */
+	CH_PCI_ID_TABLE_FENTRY(0x50a7), /* Custom T580-CR */
+	CH_PCI_ID_TABLE_FENTRY(0x50a8), /* Custom T580-KR */
+	CH_PCI_ID_TABLE_FENTRY(0x50a9), /* Custom T580-KR */
+	CH_PCI_ID_TABLE_FENTRY(0x50aa), /* Custom T580-CR */
+	CH_PCI_ID_TABLE_FENTRY(0x50ab), /* Custom T520-CR */
+	CH_PCI_ID_TABLE_FENTRY(0x50ac), /* Custom T540-BT */
+	CH_PCI_ID_TABLE_FENTRY(0x50ad), /* Custom T520-CR */
+	CH_PCI_ID_TABLE_FENTRY(0x50ae), /* Custom T540-XL-SO */
+	CH_PCI_ID_TABLE_FENTRY(0x50af), /* Custom T580-KR-SO */
+	CH_PCI_ID_TABLE_FENTRY(0x50b0), /* Custom T520-CR-LOM */
 
 	/* T6 adapters:
 	 */
@@ -192,6 +209,17 @@ CH_PCI_DEVICE_ID_TABLE_DEFINE_BEGIN
 	CH_PCI_ID_TABLE_FENTRY(0x6015),
 	CH_PCI_ID_TABLE_FENTRY(0x6080),
 	CH_PCI_ID_TABLE_FENTRY(0x6081),
+	CH_PCI_ID_TABLE_FENTRY(0x6082), /* Custom T6225-CR SFP28 */
+	CH_PCI_ID_TABLE_FENTRY(0x6083), /* Custom T62100-CR QSFP28 */
+	CH_PCI_ID_TABLE_FENTRY(0x6084), /* Custom T64100-CR QSFP28 */
+	CH_PCI_ID_TABLE_FENTRY(0x6085), /* Custom T6240-SO */
+	CH_PCI_ID_TABLE_FENTRY(0x6086), /* Custom T6225-SO-CR */
+	CH_PCI_ID_TABLE_FENTRY(0x6087), /* Custom T6225-CR */
+	CH_PCI_ID_TABLE_FENTRY(0x6088), /* Custom T62100-CR */
+	CH_PCI_ID_TABLE_FENTRY(0x6089), /* Custom T62100-KR */
+	CH_PCI_ID_TABLE_FENTRY(0x608a), /* Custom T62100-CR */
+	CH_PCI_ID_TABLE_FENTRY(0x608b), /* Custom T6225-CR */
+	CH_PCI_ID_TABLE_FENTRY(0x6092), /* Custom T62100-CR-LOM */
 CH_PCI_DEVICE_ID_TABLE_DEFINE_END;
 
 #endif /* __T4_PCI_ID_TBL_H__ */
diff --git a/drivers/net/ethernet/chelsio/cxgb4/t4_regs.h b/drivers/net/ethernet/chelsio/cxgb4/t4_regs.h
index 3348d33c36fa..695916ba0405 100644
--- a/drivers/net/ethernet/chelsio/cxgb4/t4_regs.h
+++ b/drivers/net/ethernet/chelsio/cxgb4/t4_regs.h
@@ -45,6 +45,9 @@
 #define PF_BASE(idx) (PF0_BASE + (idx) * PF_STRIDE)
 #define PF_REG(idx, reg) (PF_BASE(idx) + (reg))
 
+#define NUM_CIM_CTL_TSCH_CHANNEL_INSTANCES 4
+#define NUM_CIM_CTL_TSCH_CHANNEL_TSCH_CLASS_INSTANCES 16
+
 #define MYPORT_BASE 0x1c000
 #define MYPORT_REG(reg_addr) (MYPORT_BASE + (reg_addr))
 
@@ -65,6 +68,9 @@
 
 #define PCIE_FW_REG(reg_addr, idx) ((reg_addr) + (idx) * 4)
 
+#define NUM_LE_DB_DBGI_REQ_DATA_INSTANCES 17
+#define NUM_LE_DB_DBGI_RSP_DATA_INSTANCES 17
+
 #define SGE_PF_KDOORBELL_A 0x0
 
 #define QID_S    15
@@ -150,6 +156,23 @@
 #define T6_DBVFIFO_SIZE_M    0x1fffU
 #define T6_DBVFIFO_SIZE_G(x) (((x) >> T6_DBVFIFO_SIZE_S) & T6_DBVFIFO_SIZE_M)
 
+#define SGE_CTXT_CMD_A 0x11fc
+
+#define BUSY_S    31
+#define BUSY_V(x) ((x) << BUSY_S)
+#define BUSY_F    BUSY_V(1U)
+
+#define CTXTTYPE_S    24
+#define CTXTTYPE_M    0x3U
+#define CTXTTYPE_V(x) ((x) << CTXTTYPE_S)
+
+#define CTXTQID_S    0
+#define CTXTQID_M    0x1ffffU
+#define CTXTQID_V(x) ((x) << CTXTQID_S)
+
+#define SGE_CTXT_DATA0_A 0x1200
+#define SGE_CTXT_DATA5_A 0x1214
+
 #define GLOBALENABLE_S    0
 #define GLOBALENABLE_V(x) ((x) << GLOBALENABLE_S)
 #define GLOBALENABLE_F    GLOBALENABLE_V(1U)
@@ -319,6 +342,16 @@
 
 #define SGE_IMSG_CTXT_BADDR_A 0x1088
 #define SGE_FLM_CACHE_BADDR_A 0x108c
+#define SGE_FLM_CFG_A 0x1090
+
+#define NOHDR_S    18
+#define NOHDR_V(x) ((x) << NOHDR_S)
+#define NOHDR_F    NOHDR_V(1U)
+
+#define HDRSTARTFLQ_S    11
+#define HDRSTARTFLQ_M    0x7U
+#define HDRSTARTFLQ_G(x) (((x) >> HDRSTARTFLQ_S) & HDRSTARTFLQ_M)
+
 #define SGE_INGRESS_RX_THRESHOLD_A 0x10a0
 
 #define THRESHOLD_0_S    24
@@ -454,6 +487,12 @@
 #define ERROR_QID_M    0x1ffffU
 #define ERROR_QID_G(x) (((x) >> ERROR_QID_S) & ERROR_QID_M)
 
+#define SGE_INT_CAUSE5_A        0x110c
+
+#define ERR_T_RXCRC_S    31
+#define ERR_T_RXCRC_V(x) ((x) << ERR_T_RXCRC_S)
+#define ERR_T_RXCRC_F    ERR_T_RXCRC_V(1U)
+
 #define HP_INT_THRESH_S    28
 #define HP_INT_THRESH_M    0xfU
 #define HP_INT_THRESH_V(x) ((x) << HP_INT_THRESH_S)
@@ -524,6 +563,12 @@
 #define AIVEC_V(x) ((x) << AIVEC_S)
 
 #define PCIE_PF_CLI_A	0x44
+
+#define PCIE_PF_EXPROM_OFST_A 0x4c
+#define OFFSET_S    10
+#define OFFSET_M    0x3fffU
+#define OFFSET_G(x) (((x) >> OFFSET_S) & OFFSET_M)
+
 #define PCIE_INT_CAUSE_A	0x3004
 
 #define UNXSPLCPLERR_S    29
@@ -839,6 +884,12 @@
 #define TDUE_V(x) ((x) << TDUE_S)
 #define TDUE_F    TDUE_V(1U)
 
+/* SPARE2 register contains 32-bit value at offset 0x6 in Serial INIT
+ * Configuration flashed on EEPROM. This value corresponds to 32-bit
+ * Serial Configuration Version information.
+ */
+#define PCIE_STATIC_SPARE2_A	0x5bfc
+
 /* registers for module MC */
 #define MC_INT_CAUSE_A		0x7518
 #define MC_P_INT_CAUSE_A	0x41318
@@ -931,6 +982,10 @@
 
 #define MA_EXT_MEMORY1_BAR_A 0x7808
 
+#define HMA_MUX_S    5
+#define HMA_MUX_V(x) ((x) << HMA_MUX_S)
+#define HMA_MUX_F    HMA_MUX_V(1U)
+
 #define EXT_MEM1_BASE_S    16
 #define EXT_MEM1_BASE_M    0xfffU
 #define EXT_MEM1_BASE_G(x) (((x) >> EXT_MEM1_BASE_S) & EXT_MEM1_BASE_M)
@@ -1077,6 +1132,10 @@
 #define TIEQINPARERRINT_V(x) ((x) << TIEQINPARERRINT_S)
 #define TIEQINPARERRINT_F    TIEQINPARERRINT_V(1U)
 
+#define TIMER0INT_S    2
+#define TIMER0INT_V(x) ((x) << TIMER0INT_S)
+#define TIMER0INT_F    TIMER0INT_V(1U)
+
 #define PREFDROPINT_S    1
 #define PREFDROPINT_V(x) ((x) << PREFDROPINT_S)
 #define PREFDROPINT_F    PREFDROPINT_V(1U)
@@ -1293,6 +1352,10 @@
 #define TP_OUT_CONFIG_A		0x7d04
 #define TP_GLOBAL_CONFIG_A	0x7d08
 
+#define ACTIVEFILTERCOUNTS_S    22
+#define ACTIVEFILTERCOUNTS_V(x) ((x) << ACTIVEFILTERCOUNTS_S)
+#define ACTIVEFILTERCOUNTS_F    ACTIVEFILTERCOUNTS_V(1U)
+
 #define TP_CMM_TCB_BASE_A 0x7d10
 #define TP_CMM_MM_BASE_A 0x7d14
 #define TP_CMM_TIMER_BASE_A 0x7d18
@@ -1411,6 +1474,7 @@
 #define ROWINDEX_V(x) ((x) << ROWINDEX_S)
 
 #define TP_CCTRL_TABLE_A	0x7ddc
+#define TP_PACE_TABLE_A 0x7dd8
 #define TP_MTU_TABLE_A		0x7de4
 
 #define MTUINDEX_S    24
@@ -1443,12 +1507,42 @@
 #define LKPTBLQUEUE0_M    0x3ffU
 #define LKPTBLQUEUE0_G(x) (((x) >> LKPTBLQUEUE0_S) & LKPTBLQUEUE0_M)
 
+#define TP_TM_PIO_ADDR_A 0x7e18
+#define TP_TM_PIO_DATA_A 0x7e1c
+#define TP_MOD_CONFIG_A 0x7e24
+
+#define TIMERMODE_S    8
+#define TIMERMODE_M    0xffU
+#define TIMERMODE_G(x) (((x) >> TIMERMODE_S) & TIMERMODE_M)
+
+#define TP_TX_MOD_Q1_Q0_TIMER_SEPARATOR_A 0x3
+#define TP_TX_MOD_Q1_Q0_RATE_LIMIT_A 0x8
+
 #define TP_PIO_ADDR_A	0x7e40
 #define TP_PIO_DATA_A	0x7e44
 #define TP_MIB_INDEX_A	0x7e50
 #define TP_MIB_DATA_A	0x7e54
 #define TP_INT_CAUSE_A	0x7e74
 
+#define TP_FLM_FREE_PS_CNT_A 0x7e80
+#define TP_FLM_FREE_RX_CNT_A 0x7e84
+
+#define FREEPSTRUCTCOUNT_S    0
+#define FREEPSTRUCTCOUNT_M    0x1fffffU
+#define FREEPSTRUCTCOUNT_G(x) (((x) >> FREEPSTRUCTCOUNT_S) & FREEPSTRUCTCOUNT_M)
+
+#define FREERXPAGECOUNT_S    0
+#define FREERXPAGECOUNT_M    0x1fffffU
+#define FREERXPAGECOUNT_V(x) ((x) << FREERXPAGECOUNT_S)
+#define FREERXPAGECOUNT_G(x) (((x) >> FREERXPAGECOUNT_S) & FREERXPAGECOUNT_M)
+
+#define TP_FLM_FREE_TX_CNT_A 0x7e88
+
+#define FREETXPAGECOUNT_S    0
+#define FREETXPAGECOUNT_M    0x1fffffU
+#define FREETXPAGECOUNT_V(x) ((x) << FREETXPAGECOUNT_S)
+#define FREETXPAGECOUNT_G(x) (((x) >> FREETXPAGECOUNT_S) & FREETXPAGECOUNT_M)
+
 #define FLMTXFLSTEMPTY_S    30
 #define FLMTXFLSTEMPTY_V(x) ((x) << FLMTXFLSTEMPTY_S)
 #define FLMTXFLSTEMPTY_F    FLMTXFLSTEMPTY_V(1U)
@@ -1545,6 +1639,10 @@
 #define VNIC_V(x) ((x) << VNIC_S)
 #define VNIC_F    VNIC_V(1U)
 
+#define USE_ENC_IDX_S		13
+#define USE_ENC_IDX_V(x)	((x) << USE_ENC_IDX_S)
+#define USE_ENC_IDX_F		USE_ENC_IDX_V(1U)
+
 #define CSUM_HAS_PSEUDO_HDR_S    10
 #define CSUM_HAS_PSEUDO_HDR_V(x) ((x) << CSUM_HAS_PSEUDO_HDR_S)
 #define CSUM_HAS_PSEUDO_HDR_F    CSUM_HAS_PSEUDO_HDR_V(1U)
@@ -1623,6 +1721,20 @@
 #define IESPI_PAR_ERROR_V(x) ((x) << IESPI_PAR_ERROR_S)
 #define IESPI_PAR_ERROR_F    IESPI_PAR_ERROR_V(1U)
 
+#define ULP_TX_LA_RDPTR_0_A 0x8ec0
+#define ULP_TX_LA_RDDATA_0_A 0x8ec4
+#define ULP_TX_LA_WRPTR_0_A 0x8ec8
+#define ULP_TX_ASIC_DEBUG_CTRL_A 0x8f70
+
+#define ULP_TX_ASIC_DEBUG_0_A 0x8f74
+#define ULP_TX_ASIC_DEBUG_1_A 0x8f78
+#define ULP_TX_ASIC_DEBUG_2_A 0x8f7c
+#define ULP_TX_ASIC_DEBUG_3_A 0x8f80
+#define ULP_TX_ASIC_DEBUG_4_A 0x8f84
+
+/* registers for module PM_RX */
+#define PM_RX_BASE_ADDR 0x8fc0
+
 #define PMRX_E_PCMD_PAR_ERROR_S    0
 #define PMRX_E_PCMD_PAR_ERROR_V(x) ((x) << PMRX_E_PCMD_PAR_ERROR_S)
 #define PMRX_E_PCMD_PAR_ERROR_F    PMRX_E_PCMD_PAR_ERROR_V(1U)
@@ -1795,6 +1907,8 @@
 #define MPS_PORT_STAT_RX_PORT_LESS_64B_H 0x614
 #define MAC_PORT_MAGIC_MACID_LO 0x824
 #define MAC_PORT_MAGIC_MACID_HI 0x828
+#define MAC_PORT_TX_TS_VAL_LO   0x928
+#define MAC_PORT_TX_TS_VAL_HI   0x92c
 
 #define MAC_PORT_EPIO_DATA0_A 0x8c0
 #define MAC_PORT_EPIO_DATA1_A 0x8c4
@@ -1804,6 +1918,9 @@
 
 #define MAC_PORT_CFG2_A 0x818
 
+#define MAC_PORT_PTP_SUM_LO_A 0x990
+#define MAC_PORT_PTP_SUM_HI_A 0x994
+
 #define MPS_CMN_CTL_A	0x9000
 
 #define COUNTPAUSEMCRX_S    5
@@ -2251,6 +2368,35 @@
 #define CHNENABLE_V(x) ((x) << CHNENABLE_S)
 #define CHNENABLE_F    CHNENABLE_V(1U)
 
+#define LE_DB_DBGI_CONFIG_A 0x19cf0
+
+#define DBGICMDBUSY_S    3
+#define DBGICMDBUSY_V(x) ((x) << DBGICMDBUSY_S)
+#define DBGICMDBUSY_F    DBGICMDBUSY_V(1U)
+
+#define DBGICMDSTRT_S    2
+#define DBGICMDSTRT_V(x) ((x) << DBGICMDSTRT_S)
+#define DBGICMDSTRT_F    DBGICMDSTRT_V(1U)
+
+#define DBGICMDMODE_S    0
+#define DBGICMDMODE_M    0x3U
+#define DBGICMDMODE_V(x) ((x) << DBGICMDMODE_S)
+
+#define LE_DB_DBGI_REQ_TCAM_CMD_A 0x19cf4
+
+#define DBGICMD_S    20
+#define DBGICMD_M    0xfU
+#define DBGICMD_V(x) ((x) << DBGICMD_S)
+
+#define DBGITID_S    0
+#define DBGITID_M    0xfffffU
+#define DBGITID_V(x) ((x) << DBGITID_S)
+
+#define LE_DB_DBGI_REQ_DATA_A 0x19d00
+#define LE_DB_DBGI_RSP_STATUS_A 0x19d94
+
+#define LE_DB_DBGI_RSP_DATA_A 0x19da0
+
 #define PRTENABLE_S    29
 #define PRTENABLE_V(x) ((x) << PRTENABLE_S)
 #define PRTENABLE_F    PRTENABLE_V(1U)
@@ -2423,10 +2569,44 @@
 #define MPS_RX_MAC_BG_PG_CNT0_A 0x11208
 #define MPS_RX_LPBK_BG_PG_CNT0_A 0x11218
 
+#define MPS_RX_VXLAN_TYPE_A 0x11234
+
+#define VXLAN_EN_S    16
+#define VXLAN_EN_V(x) ((x) << VXLAN_EN_S)
+#define VXLAN_EN_F    VXLAN_EN_V(1U)
+
+#define VXLAN_S    0
+#define VXLAN_M    0xffffU
+#define VXLAN_V(x) ((x) << VXLAN_S)
+#define VXLAN_G(x) (((x) >> VXLAN_S) & VXLAN_M)
+
+#define MPS_RX_GENEVE_TYPE_A 0x11238
+
+#define GENEVE_EN_S    16
+#define GENEVE_EN_V(x) ((x) << GENEVE_EN_S)
+#define GENEVE_EN_F    GENEVE_EN_V(1U)
+
+#define GENEVE_S    0
+#define GENEVE_M    0xffffU
+#define GENEVE_V(x) ((x) << GENEVE_S)
+#define GENEVE_G(x) (((x) >> GENEVE_S) & GENEVE_M)
+
 #define MPS_CLS_TCAM_Y_L_A 0xf000
 #define MPS_CLS_TCAM_DATA0_A 0xf000
 #define MPS_CLS_TCAM_DATA1_A 0xf004
 
+#define CTLREQID_S    30
+#define CTLREQID_V(x) ((x) << CTLREQID_S)
+
+#define MPS_VF_RPLCT_MAP0_A 0x1111c
+#define MPS_VF_RPLCT_MAP1_A 0x11120
+#define MPS_VF_RPLCT_MAP2_A 0x11124
+#define MPS_VF_RPLCT_MAP3_A 0x11128
+#define MPS_VF_RPLCT_MAP4_A 0x11300
+#define MPS_VF_RPLCT_MAP5_A 0x11304
+#define MPS_VF_RPLCT_MAP6_A 0x11308
+#define MPS_VF_RPLCT_MAP7_A 0x1130c
+
 #define VIDL_S    16
 #define VIDL_M    0xffffU
 #define VIDL_G(x) (((x) >> VIDL_S) & VIDL_M)
@@ -2437,8 +2617,14 @@
 
 #define DATAPORTNUM_S    12
 #define DATAPORTNUM_M    0xfU
+#define DATAPORTNUM_V(x) ((x) << DATAPORTNUM_S)
 #define DATAPORTNUM_G(x) (((x) >> DATAPORTNUM_S) & DATAPORTNUM_M)
 
+#define DATALKPTYPE_S    10
+#define DATALKPTYPE_M    0x3U
+#define DATALKPTYPE_V(x) ((x) << DATALKPTYPE_S)
+#define DATALKPTYPE_G(x) (((x) >> DATALKPTYPE_S) & DATALKPTYPE_M)
+
 #define DATADIPHIT_S    8
 #define DATADIPHIT_V(x) ((x) << DATADIPHIT_S)
 #define DATADIPHIT_F    DATADIPHIT_V(1U)
@@ -2451,6 +2637,10 @@
 #define DATAVIDH1_M    0x7fU
 #define DATAVIDH1_G(x) (((x) >> DATAVIDH1_S) & DATAVIDH1_M)
 
+#define MPS_CLS_TCAM_RDATA0_REQ_ID1_A 0xf020
+#define MPS_CLS_TCAM_RDATA1_REQ_ID1_A 0xf024
+#define MPS_CLS_TCAM_RDATA2_REQ_ID1_A 0xf028
+
 #define USED_S    16
 #define USED_M    0x7ffU
 #define USED_G(x) (((x) >> USED_S) & USED_M)
@@ -2643,6 +2833,8 @@
 #define ULP_RX_LA_RDPTR_A 0x19240
 #define ULP_RX_LA_RDDATA_A 0x19244
 #define ULP_RX_LA_WRPTR_A 0x19248
+#define ULP_RX_TLS_KEY_LLIMIT_A 0x192ac
+#define ULP_RX_TLS_KEY_ULIMIT_A 0x192b0
 
 #define HPZ3_S    24
 #define HPZ3_V(x) ((x) << HPZ3_S)
@@ -2831,6 +3023,14 @@
 #define REV_V(x) ((x) << REV_S)
 #define REV_G(x) (((x) >> REV_S) & REV_M)
 
+#define HASHTBLMEMCRCERR_S    27
+#define HASHTBLMEMCRCERR_V(x) ((x) << HASHTBLMEMCRCERR_S)
+#define HASHTBLMEMCRCERR_F    HASHTBLMEMCRCERR_V(1U)
+
+#define CMDTIDERR_S    22
+#define CMDTIDERR_V(x) ((x) << CMDTIDERR_S)
+#define CMDTIDERR_F    CMDTIDERR_V(1U)
+
 #define T6_UNKNOWNCMD_S    3
 #define T6_UNKNOWNCMD_V(x) ((x) << T6_UNKNOWNCMD_S)
 #define T6_UNKNOWNCMD_F    T6_UNKNOWNCMD_V(1U)
@@ -2844,13 +3044,28 @@
 #define T6_LIPMISS_F    T6_LIPMISS_V(1U)
 
 #define LE_DB_CONFIG_A 0x19c04
+#define LE_DB_ROUTING_TABLE_INDEX_A 0x19c10
+#define LE_DB_ACTIVE_TABLE_START_INDEX_A 0x19c10
+#define LE_DB_FILTER_TABLE_INDEX_A 0x19c14
 #define LE_DB_SERVER_INDEX_A 0x19c18
 #define LE_DB_SRVR_START_INDEX_A 0x19c18
+#define LE_DB_CLIP_TABLE_INDEX_A 0x19c1c
 #define LE_DB_ACT_CNT_IPV4_A 0x19c20
 #define LE_DB_ACT_CNT_IPV6_A 0x19c24
+#define LE_DB_HASH_CONFIG_A 0x19c28
+
+#define HASHTIDSIZE_S    16
+#define HASHTIDSIZE_M    0x3fU
+#define HASHTIDSIZE_G(x) (((x) >> HASHTIDSIZE_S) & HASHTIDSIZE_M)
+
+#define HASHTBLSIZE_S    3
+#define HASHTBLSIZE_M    0x1ffffU
+#define HASHTBLSIZE_G(x) (((x) >> HASHTBLSIZE_S) & HASHTBLSIZE_M)
+
 #define LE_DB_HASH_TID_BASE_A 0x19c30
 #define LE_DB_HASH_TBL_BASE_ADDR_A 0x19c30
 #define LE_DB_INT_CAUSE_A 0x19c3c
+#define LE_DB_CLCAM_TID_BASE_A 0x19df4
 #define LE_DB_TID_HASHBASE_A 0x19df8
 #define T6_LE_DB_HASH_TID_BASE_A 0x19df8
 
@@ -2894,6 +3109,23 @@
 #define SSRAMINTPERR_V(x) ((x) << SSRAMINTPERR_S)
 #define SSRAMINTPERR_F    SSRAMINTPERR_V(1U)
 
+#define LE_DB_RSP_CODE_0_A	0x19c74
+
+#define TCAM_ACTV_HIT_S		0
+#define TCAM_ACTV_HIT_M		0x1fU
+#define TCAM_ACTV_HIT_V(x)	((x) << TCAM_ACTV_HIT_S)
+#define TCAM_ACTV_HIT_G(x)	(((x) >> TCAM_ACTV_HIT_S) & TCAM_ACTV_HIT_M)
+
+#define LE_DB_RSP_CODE_1_A     0x19c78
+
+#define HASH_ACTV_HIT_S		25
+#define HASH_ACTV_HIT_M		0x1fU
+#define HASH_ACTV_HIT_V(x)	((x) << HASH_ACTV_HIT_S)
+#define HASH_ACTV_HIT_G(x)	(((x) >> HASH_ACTV_HIT_S) & HASH_ACTV_HIT_M)
+
+#define LE_3_DB_HASH_MASK_GEN_IPV4_T6_A	0x19eac
+#define LE_4_DB_HASH_MASK_GEN_IPV4_T6_A	0x19eb0
+
 #define NCSI_INT_CAUSE_A 0x1a0d8
 
 #define CIM_DM_PRTY_ERR_S    8
diff --git a/drivers/net/ethernet/chelsio/cxgb4/t4_tcb.h b/drivers/net/ethernet/chelsio/cxgb4/t4_tcb.h
new file mode 100644
index 000000000000..22a0220123ad
--- /dev/null
+++ b/drivers/net/ethernet/chelsio/cxgb4/t4_tcb.h
@@ -0,0 +1,144 @@
+/*
+ * This file is part of the Chelsio T4/T5/T6 Ethernet driver for Linux.
+ *
+ * Copyright (c) 2017 Chelsio Communications, Inc. All rights reserved.
+ *
+ * This software is available to you under a choice of one of two
+ * licenses.  You may choose to be licensed under the terms of the GNU
+ * General Public License (GPL) Version 2, available from the file
+ * COPYING in the main directory of this source tree, or the
+ * OpenIB.org BSD license below:
+ *
+ *     Redistribution and use in source and binary forms, with or
+ *     without modification, are permitted provided that the following
+ *     conditions are met:
+ *
+ *      - Redistributions of source code must retain the above
+ *        copyright notice, this list of conditions and the following
+ *        disclaimer.
+ *
+ *      - Redistributions in binary form must reproduce the above
+ *        copyright notice, this list of conditions and the following
+ *        disclaimer in the documentation and/or other materials
+ *        provided with the distribution.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND,
+ * EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF
+ * MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND
+ * NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS
+ * BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN
+ * ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN
+ * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
+ * SOFTWARE.
+ */
+
+#ifndef __T4_TCB_H
+#define __T4_TCB_H
+
+#define TCB_L2T_IX_W		0
+#define TCB_L2T_IX_S		12
+#define TCB_L2T_IX_M		0xfffULL
+#define TCB_L2T_IX_V(x)		((x) << TCB_L2T_IX_S)
+
+#define TCB_T_FLAGS_W           1
+#define TCB_T_FLAGS_S           0
+#define TCB_T_FLAGS_M           0xffffffffffffffffULL
+#define TCB_T_FLAGS_V(x)        ((__u64)(x) << TCB_T_FLAGS_S)
+
+#define TCB_FIELD_COOKIE_TFLAG	1
+
+#define TCB_SMAC_SEL_W		0
+#define TCB_SMAC_SEL_S		24
+#define TCB_SMAC_SEL_M		0xffULL
+#define TCB_SMAC_SEL_V(x)	((x) << TCB_SMAC_SEL_S)
+
+#define TCB_T_FLAGS_W		1
+#define TCB_T_FLAGS_S		0
+#define TCB_T_FLAGS_M		0xffffffffffffffffULL
+#define TCB_T_FLAGS_V(x)	((__u64)(x) << TCB_T_FLAGS_S)
+
+#define TF_DROP_S		22
+#define TF_DIRECT_STEER_S	23
+#define TF_LPBK_S		59
+
+#define TF_CCTRL_ECE_S		60
+#define TF_CCTRL_CWR_S		61
+#define TF_CCTRL_RFR_S		62
+
+#define TCB_RSS_INFO_W		3
+#define TCB_RSS_INFO_S		0
+#define TCB_RSS_INFO_M		0x3ffULL
+#define TCB_RSS_INFO_V(x)	((x) << TCB_RSS_INFO_S)
+
+#define TCB_T_STATE_W		3
+#define TCB_T_STATE_S		16
+#define TCB_T_STATE_M		0xfULL
+#define TCB_T_STATE_V(x)	((x) << TCB_T_STATE_S)
+
+#define TCB_TIMESTAMP_W		5
+#define TCB_TIMESTAMP_S		0
+#define TCB_TIMESTAMP_M		0xffffffffULL
+#define TCB_TIMESTAMP_V(x)	((x) << TCB_TIMESTAMP_S)
+
+#define TCB_RTT_TS_RECENT_AGE_W		6
+#define TCB_RTT_TS_RECENT_AGE_S		0
+#define TCB_RTT_TS_RECENT_AGE_M		0xffffffffULL
+#define TCB_RTT_TS_RECENT_AGE_V(x)	((x) << TCB_RTT_TS_RECENT_AGE_S)
+
+#define TCB_T_RTSEQ_RECENT_W    7
+#define TCB_T_RTSEQ_RECENT_S    0
+#define TCB_T_RTSEQ_RECENT_M    0xffffffffULL
+#define TCB_T_RTSEQ_RECENT_V(x) ((x) << TCB_T_RTSEQ_RECENT_S)
+
+#define TCB_TX_MAX_W		9
+#define TCB_TX_MAX_S		0
+#define TCB_TX_MAX_M		0xffffffffULL
+#define TCB_TX_MAX_V(x)		((x) << TCB_TX_MAX_S)
+
+#define TCB_SND_UNA_RAW_W	10
+#define TCB_SND_UNA_RAW_S	0
+#define TCB_SND_UNA_RAW_M	0xfffffffULL
+#define TCB_SND_UNA_RAW_V(x)	((x) << TCB_SND_UNA_RAW_S)
+
+#define TCB_SND_NXT_RAW_W	10
+#define TCB_SND_NXT_RAW_S	28
+#define TCB_SND_NXT_RAW_M	0xfffffffULL
+#define TCB_SND_NXT_RAW_V(x)	((x) << TCB_SND_NXT_RAW_S)
+
+#define TCB_SND_MAX_RAW_W	11
+#define TCB_SND_MAX_RAW_S	24
+#define TCB_SND_MAX_RAW_M	0xfffffffULL
+#define TCB_SND_MAX_RAW_V(x)	((x) << TCB_SND_MAX_RAW_S)
+
+#define TCB_RCV_NXT_W		16
+#define TCB_RCV_NXT_S		10
+#define TCB_RCV_NXT_M		0xffffffffULL
+#define TCB_RCV_NXT_V(x)	((x) << TCB_RCV_NXT_S)
+
+#define TCB_RCV_WND_W		17
+#define TCB_RCV_WND_S		10
+#define TCB_RCV_WND_M		0xffffffULL
+#define TCB_RCV_WND_V(x)	((x) << TCB_RCV_WND_S)
+
+#define TCB_RX_FRAG2_PTR_RAW_W	27
+#define TCB_RX_FRAG3_LEN_RAW_W	29
+#define TCB_RX_FRAG3_START_IDX_OFFSET_RAW_W	30
+#define TCB_PDU_HDR_LEN_W	31
+
+#define TCB_RQ_START_W		30
+#define TCB_RQ_START_S		0
+#define TCB_RQ_START_M		0x3ffffffULL
+#define TCB_RQ_START_V(x)	((x) << TCB_RQ_START_S)
+
+#define TF_RX_PDU_OUT_S		49
+#define TF_RX_PDU_OUT_V(x)	((__u64)(x) << TF_RX_PDU_OUT_S)
+
+#define TF_CORE_BYPASS_S	63
+#define TF_CORE_BYPASS_V(x)	((__u64)(x) << TF_CORE_BYPASS_S)
+#define TF_CORE_BYPASS_F	TF_CORE_BYPASS_V(1)
+
+#define TF_NON_OFFLOAD_S	1
+#define TF_NON_OFFLOAD_V(x)	((x) << TF_NON_OFFLOAD_S)
+#define TF_NON_OFFLOAD_F	TF_NON_OFFLOAD_V(1)
+
+#endif /* __T4_TCB_H */
diff --git a/drivers/net/ethernet/chelsio/cxgb4/t4_values.h b/drivers/net/ethernet/chelsio/cxgb4/t4_values.h
index f6558cbfc54e..eb1aa82149db 100644
--- a/drivers/net/ethernet/chelsio/cxgb4/t4_values.h
+++ b/drivers/net/ethernet/chelsio/cxgb4/t4_values.h
@@ -71,12 +71,18 @@
 #define FETCHBURSTMIN_64B_X		2
 #define FETCHBURSTMIN_128B_X		3
 
+/* T6 and later use a single-bit encoding for FetchBurstMin */
+#define FETCHBURSTMIN_64B_T6_X		0
+#define FETCHBURSTMIN_128B_T6_X		1
+
 #define FETCHBURSTMAX_256B_X		2
 #define FETCHBURSTMAX_512B_X		3
 
+#define HOSTFCMODE_INGRESS_QUEUE_X	1
 #define HOSTFCMODE_STATUS_PAGE_X	2
 
 #define CIDXFLUSHTHRESH_32_X		5
+#define CIDXFLUSHTHRESH_128_X		7
 
 #define UPDATEDELIVERY_INTERRUPT_X	1
 
diff --git a/drivers/net/ethernet/chelsio/cxgb4/t4fw_api.h b/drivers/net/ethernet/chelsio/cxgb4/t4fw_api.h
index c65c33c03bcb..2419459a0b85 100644
--- a/drivers/net/ethernet/chelsio/cxgb4/t4fw_api.h
+++ b/drivers/net/ethernet/chelsio/cxgb4/t4fw_api.h
@@ -87,6 +87,7 @@ enum fw_wr_opcodes {
 	FW_ULPTX_WR                    = 0x04,
 	FW_TP_WR                       = 0x05,
 	FW_ETH_TX_PKT_WR               = 0x08,
+	FW_ETH_TX_EO_WR                = 0x1c,
 	FW_OFLD_CONNECTION_WR          = 0x2f,
 	FW_FLOWC_WR                    = 0x0a,
 	FW_OFLD_TX_DATA_WR             = 0x0b,
@@ -101,10 +102,14 @@ enum fw_wr_opcodes {
 	FW_RI_BIND_MW_WR               = 0x18,
 	FW_RI_FR_NSMR_WR               = 0x19,
 	FW_RI_FR_NSMR_TPTE_WR	       = 0x20,
+	FW_RI_RDMA_WRITE_CMPL_WR       = 0x21,
 	FW_RI_INV_LSTAG_WR             = 0x1a,
 	FW_ISCSI_TX_DATA_WR	       = 0x45,
+	FW_PTP_TX_PKT_WR               = 0x46,
+	FW_TLSTX_DATA_WR	       = 0x68,
 	FW_CRYPTO_LOOKASIDE_WR         = 0X6d,
-	FW_LASTC2E_WR                  = 0x70
+	FW_LASTC2E_WR                  = 0x70,
+	FW_FILTER2_WR		       = 0x77
 };
 
 struct fw_wr_hdr {
@@ -200,6 +205,51 @@ struct fw_filter_wr {
 	__u8   sma[6];
 };
 
+struct fw_filter2_wr {
+	__be32 op_pkd;
+	__be32 len16_pkd;
+	__be64 r3;
+	__be32 tid_to_iq;
+	__be32 del_filter_to_l2tix;
+	__be16 ethtype;
+	__be16 ethtypem;
+	__u8   frag_to_ovlan_vldm;
+	__u8   smac_sel;
+	__be16 rx_chan_rx_rpl_iq;
+	__be32 maci_to_matchtypem;
+	__u8   ptcl;
+	__u8   ptclm;
+	__u8   ttyp;
+	__u8   ttypm;
+	__be16 ivlan;
+	__be16 ivlanm;
+	__be16 ovlan;
+	__be16 ovlanm;
+	__u8   lip[16];
+	__u8   lipm[16];
+	__u8   fip[16];
+	__u8   fipm[16];
+	__be16 lp;
+	__be16 lpm;
+	__be16 fp;
+	__be16 fpm;
+	__be16 r7;
+	__u8   sma[6];
+	__be16 r8;
+	__u8   filter_type_swapmac;
+	__u8   natmode_to_ulp_type;
+	__be16 newlport;
+	__be16 newfport;
+	__u8   newlip[16];
+	__u8   newfip[16];
+	__be32 natseqcheck;
+	__be32 r9;
+	__be64 r10;
+	__be64 r11;
+	__be64 r12;
+	__be64 r13;
+};
+
 #define FW_FILTER_WR_TID_S      12
 #define FW_FILTER_WR_TID_M      0xfffff
 #define FW_FILTER_WR_TID_V(x)   ((x) << FW_FILTER_WR_TID_S)
@@ -384,6 +434,32 @@ struct fw_filter_wr {
 #define FW_FILTER_WR_RX_RPL_IQ_G(x)     \
 	(((x) >> FW_FILTER_WR_RX_RPL_IQ_S) & FW_FILTER_WR_RX_RPL_IQ_M)
 
+#define FW_FILTER2_WR_FILTER_TYPE_S	1
+#define FW_FILTER2_WR_FILTER_TYPE_M	0x1
+#define FW_FILTER2_WR_FILTER_TYPE_V(x)	((x) << FW_FILTER2_WR_FILTER_TYPE_S)
+#define FW_FILTER2_WR_FILTER_TYPE_G(x)  \
+	(((x) >> FW_FILTER2_WR_FILTER_TYPE_S) & FW_FILTER2_WR_FILTER_TYPE_M)
+#define FW_FILTER2_WR_FILTER_TYPE_F	FW_FILTER2_WR_FILTER_TYPE_V(1U)
+
+#define FW_FILTER2_WR_NATMODE_S		5
+#define FW_FILTER2_WR_NATMODE_M		0x7
+#define FW_FILTER2_WR_NATMODE_V(x)	((x) << FW_FILTER2_WR_NATMODE_S)
+#define FW_FILTER2_WR_NATMODE_G(x)      \
+	(((x) >> FW_FILTER2_WR_NATMODE_S) & FW_FILTER2_WR_NATMODE_M)
+
+#define FW_FILTER2_WR_NATFLAGCHECK_S	4
+#define FW_FILTER2_WR_NATFLAGCHECK_M	0x1
+#define FW_FILTER2_WR_NATFLAGCHECK_V(x)	((x) << FW_FILTER2_WR_NATFLAGCHECK_S)
+#define FW_FILTER2_WR_NATFLAGCHECK_G(x) \
+	(((x) >> FW_FILTER2_WR_NATFLAGCHECK_S) & FW_FILTER2_WR_NATFLAGCHECK_M)
+#define FW_FILTER2_WR_NATFLAGCHECK_F	FW_FILTER2_WR_NATFLAGCHECK_V(1U)
+
+#define FW_FILTER2_WR_ULP_TYPE_S	0
+#define FW_FILTER2_WR_ULP_TYPE_M	0xf
+#define FW_FILTER2_WR_ULP_TYPE_V(x)	((x) << FW_FILTER2_WR_ULP_TYPE_S)
+#define FW_FILTER2_WR_ULP_TYPE_G(x)     \
+	(((x) >> FW_FILTER2_WR_ULP_TYPE_S) & FW_FILTER2_WR_ULP_TYPE_M)
+
 #define FW_FILTER_WR_MACI_S     23
 #define FW_FILTER_WR_MACI_M     0x1ff
 #define FW_FILTER_WR_MACI_V(x)  ((x) << FW_FILTER_WR_MACI_S)
@@ -440,6 +516,13 @@ struct fw_ulptx_wr {
 	u64 cookie;
 };
 
+#define FW_ULPTX_WR_DATA_S      28
+#define FW_ULPTX_WR_DATA_M      0x1
+#define FW_ULPTX_WR_DATA_V(x)   ((x) << FW_ULPTX_WR_DATA_S)
+#define FW_ULPTX_WR_DATA_G(x)   \
+	(((x) >> FW_ULPTX_WR_DATA_S) & FW_ULPTX_WR_DATA_M)
+#define FW_ULPTX_WR_DATA_F      FW_ULPTX_WR_DATA_V(1U)
+
 struct fw_tp_wr {
 	__be32 op_to_immdlen;
 	__be32 flowid_len16;
@@ -452,6 +535,47 @@ struct fw_eth_tx_pkt_wr {
 	__be64 r3;
 };
 
+enum fw_eth_tx_eo_type {
+	FW_ETH_TX_EO_TYPE_UDPSEG = 0,
+	FW_ETH_TX_EO_TYPE_TCPSEG,
+};
+
+struct fw_eth_tx_eo_wr {
+	__be32 op_immdlen;
+	__be32 equiq_to_len16;
+	__be64 r3;
+	union fw_eth_tx_eo {
+		struct fw_eth_tx_eo_udpseg {
+			__u8   type;
+			__u8   ethlen;
+			__be16 iplen;
+			__u8   udplen;
+			__u8   rtplen;
+			__be16 r4;
+			__be16 mss;
+			__be16 schedpktsize;
+			__be32 plen;
+		} udpseg;
+		struct fw_eth_tx_eo_tcpseg {
+			__u8   type;
+			__u8   ethlen;
+			__be16 iplen;
+			__u8   tcplen;
+			__u8   tsclk_tsoff;
+			__be16 r4;
+			__be16 mss;
+			__be16 r5;
+			__be32 plen;
+		} tcpseg;
+	} u;
+};
+
+#define FW_ETH_TX_EO_WR_IMMDLEN_S	0
+#define FW_ETH_TX_EO_WR_IMMDLEN_M	0x1ff
+#define FW_ETH_TX_EO_WR_IMMDLEN_V(x)	((x) << FW_ETH_TX_EO_WR_IMMDLEN_S)
+#define FW_ETH_TX_EO_WR_IMMDLEN_G(x)	\
+	(((x) >> FW_ETH_TX_EO_WR_IMMDLEN_S) & FW_ETH_TX_EO_WR_IMMDLEN_M)
+
 struct fw_ofld_connection_wr {
 	__be32 op_compl;
 	__be32 len16_pkd;
@@ -554,6 +678,36 @@ struct fw_ofld_connection_wr {
 #define FW_OFLD_CONNECTION_WR_CPLPASSACCEPTRPL_F       \
 	FW_OFLD_CONNECTION_WR_CPLPASSACCEPTRPL_V(1U)
 
+enum fw_flowc_mnem_tcpstate {
+	FW_FLOWC_MNEM_TCPSTATE_CLOSED   = 0, /* illegal */
+	FW_FLOWC_MNEM_TCPSTATE_LISTEN   = 1, /* illegal */
+	FW_FLOWC_MNEM_TCPSTATE_SYNSENT  = 2, /* illegal */
+	FW_FLOWC_MNEM_TCPSTATE_SYNRECEIVED = 3, /* illegal */
+	FW_FLOWC_MNEM_TCPSTATE_ESTABLISHED = 4, /* default */
+	FW_FLOWC_MNEM_TCPSTATE_CLOSEWAIT = 5, /* got peer close already */
+	FW_FLOWC_MNEM_TCPSTATE_FINWAIT1 = 6, /* haven't gotten ACK for FIN and
+					      * will resend FIN - equiv ESTAB
+					      */
+	FW_FLOWC_MNEM_TCPSTATE_CLOSING  = 7, /* haven't gotten ACK for FIN and
+					      * will resend FIN but have
+					      * received FIN
+					      */
+	FW_FLOWC_MNEM_TCPSTATE_LASTACK  = 8, /* haven't gotten ACK for FIN and
+					      * will resend FIN but have
+					      * received FIN
+					      */
+	FW_FLOWC_MNEM_TCPSTATE_FINWAIT2 = 9, /* sent FIN and got FIN + ACK,
+					      * waiting for FIN
+					      */
+	FW_FLOWC_MNEM_TCPSTATE_TIMEWAIT = 10, /* not expected */
+};
+
+enum fw_flowc_mnem_eostate {
+	FW_FLOWC_MNEM_EOSTATE_ESTABLISHED = 1, /* default */
+	/* graceful close, after sending outstanding payload */
+	FW_FLOWC_MNEM_EOSTATE_CLOSING = 2,
+};
+
 enum fw_flowc_mnem {
 	FW_FLOWC_MNEM_PFNVFN,		/* PFN [15:8] VFN [7:0] */
 	FW_FLOWC_MNEM_CH,
@@ -570,6 +724,8 @@ enum fw_flowc_mnem {
 	FW_FLOWC_MNEM_DCBPRIO,
 	FW_FLOWC_MNEM_SND_SCALE,
 	FW_FLOWC_MNEM_RCV_SCALE,
+	FW_FLOWC_MNEM_ULD_MODE,
+	FW_FLOWC_MNEM_MAX,
 };
 
 struct fw_flowc_mnemval {
@@ -581,7 +737,7 @@ struct fw_flowc_mnemval {
 struct fw_flowc_wr {
 	__be32 op_to_nparams;
 	__be32 flowid_len16;
-	struct fw_flowc_mnemval mnemval[0];
+	struct fw_flowc_mnemval mnemval[];
 };
 
 #define FW_FLOWC_WR_NPARAMS_S           0
@@ -594,6 +750,14 @@ struct fw_ofld_tx_data_wr {
 	__be32 tunnel_to_proxy;
 };
 
+#define FW_OFLD_TX_DATA_WR_ALIGNPLD_S   30
+#define FW_OFLD_TX_DATA_WR_ALIGNPLD_V(x) ((x) << FW_OFLD_TX_DATA_WR_ALIGNPLD_S)
+#define FW_OFLD_TX_DATA_WR_ALIGNPLD_F   FW_OFLD_TX_DATA_WR_ALIGNPLD_V(1U)
+
+#define FW_OFLD_TX_DATA_WR_SHOVE_S      29
+#define FW_OFLD_TX_DATA_WR_SHOVE_V(x)   ((x) << FW_OFLD_TX_DATA_WR_SHOVE_S)
+#define FW_OFLD_TX_DATA_WR_SHOVE_F      FW_OFLD_TX_DATA_WR_SHOVE_V(1U)
+
 #define FW_OFLD_TX_DATA_WR_TUNNEL_S     19
 #define FW_OFLD_TX_DATA_WR_TUNNEL_V(x)  ((x) << FW_OFLD_TX_DATA_WR_TUNNEL_S)
 
@@ -610,10 +774,6 @@ struct fw_ofld_tx_data_wr {
 #define FW_OFLD_TX_DATA_WR_MORE_S       15
 #define FW_OFLD_TX_DATA_WR_MORE_V(x)    ((x) << FW_OFLD_TX_DATA_WR_MORE_S)
 
-#define FW_OFLD_TX_DATA_WR_SHOVE_S      14
-#define FW_OFLD_TX_DATA_WR_SHOVE_V(x)   ((x) << FW_OFLD_TX_DATA_WR_SHOVE_S)
-#define FW_OFLD_TX_DATA_WR_SHOVE_F      FW_OFLD_TX_DATA_WR_SHOVE_V(1U)
-
 #define FW_OFLD_TX_DATA_WR_ULPMODE_S    10
 #define FW_OFLD_TX_DATA_WR_ULPMODE_V(x) ((x) << FW_OFLD_TX_DATA_WR_ULPMODE_S)
 
@@ -634,10 +794,12 @@ struct fw_eth_tx_pkt_vm_wr {
 	__be32 op_immdlen;
 	__be32 equiq_to_len16;
 	__be32 r3[2];
-	u8 ethmacdst[6];
-	u8 ethmacsrc[6];
-	__be16 ethtype;
-	__be16 vlantci;
+	struct_group(firmware,
+		u8 ethmacdst[ETH_ALEN];
+		u8 ethmacsrc[ETH_ALEN];
+		__be16 ethtype;
+		__be16 vlantci;
+	);
 };
 
 #define FW_CMD_MAX_TIMEOUT 10000
@@ -685,6 +847,8 @@ enum fw_cmd_opcodes {
 	FW_SCHED_CMD                   = 0x24,
 	FW_DEVLOG_CMD                  = 0x25,
 	FW_CLIP_CMD                    = 0x28,
+	FW_PTP_CMD                     = 0x3e,
+	FW_HMA_CMD                     = 0x3f,
 	FW_LASTC2E_CMD                 = 0x40,
 	FW_ERROR_CMD                   = 0x80,
 	FW_DEBUG_CMD                   = 0x81,
@@ -754,6 +918,7 @@ enum fw_ldst_addrspc {
 	FW_LDST_ADDRSPC_MPS       = 0x0020,
 	FW_LDST_ADDRSPC_FUNC      = 0x0028,
 	FW_LDST_ADDRSPC_FUNC_PCIE = 0x0029,
+	FW_LDST_ADDRSPC_I2C       = 0x0038,
 };
 
 enum fw_ldst_mps_fid {
@@ -1018,6 +1183,8 @@ enum fw_caps_config_switch {
 enum fw_caps_config_nic {
 	FW_CAPS_CONFIG_NIC		= 0x00000001,
 	FW_CAPS_CONFIG_NIC_VM		= 0x00000002,
+	FW_CAPS_CONFIG_NIC_HASHFILTER	= 0x00000020,
+	FW_CAPS_CONFIG_NIC_ETHOFLD	= 0x00000040,
 };
 
 enum fw_caps_config_ofld {
@@ -1036,6 +1203,13 @@ enum fw_caps_config_iscsi {
 	FW_CAPS_CONFIG_ISCSI_TARGET_CNXOFLD = 0x00000008,
 };
 
+enum fw_caps_config_crypto {
+	FW_CAPS_CONFIG_CRYPTO_LOOKASIDE = 0x00000001,
+	FW_CAPS_CONFIG_TLS_INLINE = 0x00000002,
+	FW_CAPS_CONFIG_IPSEC_INLINE = 0x00000004,
+	FW_CAPS_CONFIG_TLS_HW = 0x00000008,
+};
+
 enum fw_caps_config_fcoe {
 	FW_CAPS_CONFIG_FCOE_INITIATOR	= 0x00000001,
 	FW_CAPS_CONFIG_FCOE_TARGET	= 0x00000002,
@@ -1049,6 +1223,7 @@ enum fw_memtype_cf {
 	FW_MEMTYPE_CF_FLASH		= 0x4,
 	FW_MEMTYPE_CF_INTERNAL		= 0x5,
 	FW_MEMTYPE_CF_EXTMEM1           = 0x6,
+	FW_MEMTYPE_CF_HMA		= 0x7,
 };
 
 struct fw_caps_config_cmd {
@@ -1098,6 +1273,23 @@ enum fw_params_mnem {
 /*
  * device parameters
  */
+
+#define FW_PARAMS_PARAM_FILTER_MODE_S 16
+#define FW_PARAMS_PARAM_FILTER_MODE_M 0xffff
+#define FW_PARAMS_PARAM_FILTER_MODE_V(x)          \
+	((x) << FW_PARAMS_PARAM_FILTER_MODE_S)
+#define FW_PARAMS_PARAM_FILTER_MODE_G(x)          \
+	(((x) >> FW_PARAMS_PARAM_FILTER_MODE_S) & \
+	FW_PARAMS_PARAM_FILTER_MODE_M)
+
+#define FW_PARAMS_PARAM_FILTER_MASK_S 0
+#define FW_PARAMS_PARAM_FILTER_MASK_M 0xffff
+#define FW_PARAMS_PARAM_FILTER_MASK_V(x)          \
+	((x) << FW_PARAMS_PARAM_FILTER_MASK_S)
+#define FW_PARAMS_PARAM_FILTER_MASK_G(x)          \
+	(((x) >> FW_PARAMS_PARAM_FILTER_MASK_S) & \
+	FW_PARAMS_PARAM_FILTER_MASK_M)
+
 enum fw_params_param_dev {
 	FW_PARAMS_PARAM_DEV_CCLK	= 0x00, /* chip core clock in khz */
 	FW_PARAMS_PARAM_DEV_PORTVEC	= 0x01, /* the port vector */
@@ -1122,7 +1314,24 @@ enum fw_params_param_dev {
 	FW_PARAMS_PARAM_DEV_MAXIRD_ADAPTER = 0x14, /* max supported adap IRD */
 	FW_PARAMS_PARAM_DEV_ULPTX_MEMWRITE_DSGL = 0x17,
 	FW_PARAMS_PARAM_DEV_FWCACHE = 0x18,
+	FW_PARAMS_PARAM_DEV_SCFGREV = 0x1A,
+	FW_PARAMS_PARAM_DEV_VPDREV = 0x1B,
 	FW_PARAMS_PARAM_DEV_RI_FR_NSMR_TPTE_WR	= 0x1C,
+	FW_PARAMS_PARAM_DEV_FILTER2_WR  = 0x1D,
+	FW_PARAMS_PARAM_DEV_MPSBGMAP	= 0x1E,
+	FW_PARAMS_PARAM_DEV_TPCHMAP     = 0x1F,
+	FW_PARAMS_PARAM_DEV_HMA_SIZE	= 0x20,
+	FW_PARAMS_PARAM_DEV_RDMA_WRITE_WITH_IMM = 0x21,
+	FW_PARAMS_PARAM_DEV_PPOD_EDRAM  = 0x23,
+	FW_PARAMS_PARAM_DEV_RI_WRITE_CMPL_WR    = 0x24,
+	FW_PARAMS_PARAM_DEV_HPFILTER_REGION_SUPPORT = 0x26,
+	FW_PARAMS_PARAM_DEV_OPAQUE_VIID_SMT_EXTN = 0x27,
+	FW_PARAMS_PARAM_DEV_HASHFILTER_WITH_OFLD = 0x28,
+	FW_PARAMS_PARAM_DEV_DBQ_TIMER	= 0x29,
+	FW_PARAMS_PARAM_DEV_DBQ_TIMERTICK = 0x2A,
+	FW_PARAMS_PARAM_DEV_NUM_TM_CLASS = 0x2B,
+	FW_PARAMS_PARAM_DEV_FILTER = 0x2E,
+	FW_PARAMS_PARAM_DEV_KTLS_HW = 0x31,
 };
 
 /*
@@ -1154,6 +1363,8 @@ enum fw_params_param_pfvf {
 	FW_PARAMS_PARAM_PFVF_SQRQ_END	= 0x16,
 	FW_PARAMS_PARAM_PFVF_CQ_START	= 0x17,
 	FW_PARAMS_PARAM_PFVF_CQ_END	= 0x18,
+	FW_PARAMS_PARAM_PFVF_SRQ_START  = 0x19,
+	FW_PARAMS_PARAM_PFVF_SRQ_END    = 0x1A,
 	FW_PARAMS_PARAM_PFVF_SCHEDCLASS_ETH = 0x20,
 	FW_PARAMS_PARAM_PFVF_VIID       = 0x24,
 	FW_PARAMS_PARAM_PFVF_CPMASK     = 0x25,
@@ -1166,9 +1377,27 @@ enum fw_params_param_pfvf {
 	FW_PARAMS_PARAM_PFVF_EQ_END	= 0x2C,
 	FW_PARAMS_PARAM_PFVF_ACTIVE_FILTER_START = 0x2D,
 	FW_PARAMS_PARAM_PFVF_ACTIVE_FILTER_END = 0x2E,
+	FW_PARAMS_PARAM_PFVF_ETHOFLD_START = 0x2F,
 	FW_PARAMS_PARAM_PFVF_ETHOFLD_END = 0x30,
 	FW_PARAMS_PARAM_PFVF_CPLFW4MSG_ENCAP = 0x31,
-	FW_PARAMS_PARAM_PFVF_NCRYPTO_LOOKASIDE = 0x32
+	FW_PARAMS_PARAM_PFVF_HPFILTER_START = 0x32,
+	FW_PARAMS_PARAM_PFVF_HPFILTER_END = 0x33,
+	FW_PARAMS_PARAM_PFVF_TLS_START = 0x34,
+	FW_PARAMS_PARAM_PFVF_TLS_END = 0x35,
+	FW_PARAMS_PARAM_PFVF_RAWF_START = 0x36,
+	FW_PARAMS_PARAM_PFVF_RAWF_END = 0x37,
+	FW_PARAMS_PARAM_PFVF_NCRYPTO_LOOKASIDE = 0x39,
+	FW_PARAMS_PARAM_PFVF_PORT_CAPS32 = 0x3A,
+	FW_PARAMS_PARAM_PFVF_PPOD_EDRAM_START = 0x3B,
+	FW_PARAMS_PARAM_PFVF_PPOD_EDRAM_END = 0x3C,
+	FW_PARAMS_PARAM_PFVF_LINK_STATE = 0x40,
+};
+
+/* Virtual link state as seen by the specified VF */
+enum vf_link_states {
+	FW_VF_LINK_STATE_AUTO		= 0x00,
+	FW_VF_LINK_STATE_ENABLE		= 0x01,
+	FW_VF_LINK_STATE_DISABLE	= 0x02,
 };
 
 /*
@@ -1181,9 +1410,16 @@ enum fw_params_param_dmaq {
 	FW_PARAMS_PARAM_DMAQ_EQ_CMPLIQID_CTRL = 0x11,
 	FW_PARAMS_PARAM_DMAQ_EQ_SCHEDCLASS_ETH = 0x12,
 	FW_PARAMS_PARAM_DMAQ_EQ_DCBPRIO_ETH = 0x13,
+	FW_PARAMS_PARAM_DMAQ_EQ_TIMERIX	= 0x15,
 	FW_PARAMS_PARAM_DMAQ_CONM_CTXT = 0x20,
 };
 
+enum fw_params_param_dev_ktls_hw {
+	FW_PARAMS_PARAM_DEV_KTLS_HW_DISABLE      = 0x00,
+	FW_PARAMS_PARAM_DEV_KTLS_HW_ENABLE       = 0x01,
+	FW_PARAMS_PARAM_DEV_KTLS_HW_USER_ENABLE  = 0x01,
+};
+
 enum fw_params_param_dev_phyfw {
 	FW_PARAMS_PARAM_DEV_PHYFW_DOWNLOAD = 0x00,
 	FW_PARAMS_PARAM_DEV_PHYFW_VERSION = 0x01,
@@ -1192,6 +1428,12 @@ enum fw_params_param_dev_phyfw {
 enum fw_params_param_dev_diag {
 	FW_PARAM_DEV_DIAG_TMP		= 0x00,
 	FW_PARAM_DEV_DIAG_VDD		= 0x01,
+	FW_PARAM_DEV_DIAG_MAXTMPTHRESH	= 0x02,
+};
+
+enum fw_params_param_dev_filter {
+	FW_PARAM_DEV_FILTER_VNIC_MODE   = 0x00,
+	FW_PARAM_DEV_FILTER_MODE_MASK   = 0x01,
 };
 
 enum fw_params_param_dev_fwcache {
@@ -1332,6 +1574,12 @@ enum fw_iq_type {
 	FW_IQ_TYPE_NO_FL_INT_CAP
 };
 
+enum fw_iq_iqtype {
+	FW_IQ_IQTYPE_OTHER,
+	FW_IQ_IQTYPE_NIC,
+	FW_IQ_IQTYPE_OFLD,
+};
+
 struct fw_iq_cmd {
 	__be32 op_to_vfn;
 	__be32 alloc_to_len16;
@@ -1446,6 +1694,12 @@ struct fw_iq_cmd {
 #define FW_IQ_CMD_IQFLINTISCSIC_S	26
 #define FW_IQ_CMD_IQFLINTISCSIC_V(x)	((x) << FW_IQ_CMD_IQFLINTISCSIC_S)
 
+#define FW_IQ_CMD_IQTYPE_S		24
+#define FW_IQ_CMD_IQTYPE_M		0x3
+#define FW_IQ_CMD_IQTYPE_V(x)		((x) << FW_IQ_CMD_IQTYPE_S)
+#define FW_IQ_CMD_IQTYPE_G(x)		\
+	(((x) >> FW_IQ_CMD_IQTYPE_S) & FW_IQ_CMD_IQTYPE_M)
+
 #define FW_IQ_CMD_FL0CNGCHMAP_S		20
 #define FW_IQ_CMD_FL0CNGCHMAP_V(x)	((x) << FW_IQ_CMD_FL0CNGCHMAP_S)
 
@@ -1597,8 +1851,8 @@ struct fw_eq_eth_cmd {
 	__be32 fetchszm_to_iqid;
 	__be32 dcaen_to_eqsize;
 	__be64 eqaddr;
-	__be32 viid_pkd;
-	__be32 r8_lo;
+	__be32 autoequiqe_to_viid;
+	__be32 timeren_timerix;
 	__be64 r9;
 };
 
@@ -1693,6 +1947,10 @@ struct fw_eq_eth_cmd {
 #define FW_EQ_ETH_CMD_EQSIZE_S		0
 #define FW_EQ_ETH_CMD_EQSIZE_V(x)	((x) << FW_EQ_ETH_CMD_EQSIZE_S)
 
+#define FW_EQ_ETH_CMD_AUTOEQUIQE_S	31
+#define FW_EQ_ETH_CMD_AUTOEQUIQE_V(x)	((x) << FW_EQ_ETH_CMD_AUTOEQUIQE_S)
+#define FW_EQ_ETH_CMD_AUTOEQUIQE_F	FW_EQ_ETH_CMD_AUTOEQUIQE_V(1U)
+
 #define FW_EQ_ETH_CMD_AUTOEQUEQE_S	30
 #define FW_EQ_ETH_CMD_AUTOEQUEQE_V(x)	((x) << FW_EQ_ETH_CMD_AUTOEQUEQE_S)
 #define FW_EQ_ETH_CMD_AUTOEQUEQE_F	FW_EQ_ETH_CMD_AUTOEQUEQE_V(1U)
@@ -1700,6 +1958,19 @@ struct fw_eq_eth_cmd {
 #define FW_EQ_ETH_CMD_VIID_S	16
 #define FW_EQ_ETH_CMD_VIID_V(x)	((x) << FW_EQ_ETH_CMD_VIID_S)
 
+#define FW_EQ_ETH_CMD_TIMEREN_S		3
+#define FW_EQ_ETH_CMD_TIMEREN_M		0x1
+#define FW_EQ_ETH_CMD_TIMEREN_V(x)	((x) << FW_EQ_ETH_CMD_TIMEREN_S)
+#define FW_EQ_ETH_CMD_TIMEREN_G(x)	\
+    (((x) >> FW_EQ_ETH_CMD_TIMEREN_S) & FW_EQ_ETH_CMD_TIMEREN_M)
+#define FW_EQ_ETH_CMD_TIMEREN_F	FW_EQ_ETH_CMD_TIMEREN_V(1U)
+
+#define FW_EQ_ETH_CMD_TIMERIX_S		0
+#define FW_EQ_ETH_CMD_TIMERIX_M		0x7
+#define FW_EQ_ETH_CMD_TIMERIX_V(x)	((x) << FW_EQ_ETH_CMD_TIMERIX_S)
+#define FW_EQ_ETH_CMD_TIMERIX_G(x)	\
+    (((x) >> FW_EQ_ETH_CMD_TIMERIX_S) & FW_EQ_ETH_CMD_TIMERIX_M)
+
 struct fw_eq_ctrl_cmd {
 	__be32 op_to_vfn;
 	__be32 alloc_to_len16;
@@ -1956,6 +2227,19 @@ struct fw_vi_cmd {
 #define FW_VI_CMD_FREE_V(x)	((x) << FW_VI_CMD_FREE_S)
 #define FW_VI_CMD_FREE_F	FW_VI_CMD_FREE_V(1U)
 
+#define FW_VI_CMD_VFVLD_S	24
+#define FW_VI_CMD_VFVLD_M	0x1
+#define FW_VI_CMD_VFVLD_V(x)	((x) << FW_VI_CMD_VFVLD_S)
+#define FW_VI_CMD_VFVLD_G(x)	\
+	(((x) >> FW_VI_CMD_VFVLD_S) & FW_VI_CMD_VFVLD_M)
+#define FW_VI_CMD_VFVLD_F	FW_VI_CMD_VFVLD_V(1U)
+
+#define FW_VI_CMD_VIN_S		16
+#define FW_VI_CMD_VIN_M		0xff
+#define FW_VI_CMD_VIN_V(x)	((x) << FW_VI_CMD_VIN_S)
+#define FW_VI_CMD_VIN_G(x)	\
+	(((x) >> FW_VI_CMD_VIN_S) & FW_VI_CMD_VIN_M)
+
 #define FW_VI_CMD_VIID_S	0
 #define FW_VI_CMD_VIID_M	0xfff
 #define FW_VI_CMD_VIID_V(x)	((x) << FW_VI_CMD_VIID_S)
@@ -1976,6 +2260,7 @@ struct fw_vi_cmd {
 #define FW_VI_MAC_ADD_MAC		0x3FF
 #define FW_VI_MAC_ADD_PERSIST_MAC	0x3FE
 #define FW_VI_MAC_MAC_BASED_FREE	0x3FD
+#define FW_VI_MAC_ID_BASED_FREE		0x3FC
 #define FW_CLS_TCAM_NUM_ENTRIES		336
 
 enum fw_vi_mac_smac {
@@ -1992,6 +2277,13 @@ enum fw_vi_mac_result {
 	FW_VI_MAC_R_F_ACL_CHECK
 };
 
+enum fw_vi_mac_entry_types {
+	FW_VI_MAC_TYPE_EXACTMAC,
+	FW_VI_MAC_TYPE_HASHVEC,
+	FW_VI_MAC_TYPE_RAW,
+	FW_VI_MAC_TYPE_EXACTMAC_VNI,
+};
+
 struct fw_vi_mac_cmd {
 	__be32 op_to_viid;
 	__be32 freemacs_to_len16;
@@ -2003,15 +2295,42 @@ struct fw_vi_mac_cmd {
 		struct fw_vi_mac_hash {
 			__be64 hashvec;
 		} hash;
+		struct fw_vi_mac_raw {
+			__be32 raw_idx_pkd;
+			__be32 data0_pkd;
+			__be32 data1[2];
+			__be64 data0m_pkd;
+			__be32 data1m[2];
+		} raw;
+		struct fw_vi_mac_vni {
+			__be16 valid_to_idx;
+			__u8 macaddr[6];
+			__be16 r7;
+			__u8 macaddr_mask[6];
+			__be32 lookup_type_to_vni;
+			__be32 vni_mask_pkd;
+		} exact_vni[2];
 	} u;
 };
 
+#define FW_VI_MAC_CMD_SMTID_S		12
+#define FW_VI_MAC_CMD_SMTID_M		0xff
+#define FW_VI_MAC_CMD_SMTID_V(x)	((x) << FW_VI_MAC_CMD_SMTID_S)
+#define FW_VI_MAC_CMD_SMTID_G(x)	\
+	(((x) >> FW_VI_MAC_CMD_SMTID_S) & FW_VI_MAC_CMD_SMTID_M)
+
 #define FW_VI_MAC_CMD_VIID_S	0
 #define FW_VI_MAC_CMD_VIID_V(x)	((x) << FW_VI_MAC_CMD_VIID_S)
 
 #define FW_VI_MAC_CMD_FREEMACS_S	31
 #define FW_VI_MAC_CMD_FREEMACS_V(x)	((x) << FW_VI_MAC_CMD_FREEMACS_S)
 
+#define FW_VI_MAC_CMD_ENTRY_TYPE_S      23
+#define FW_VI_MAC_CMD_ENTRY_TYPE_M      0x7
+#define FW_VI_MAC_CMD_ENTRY_TYPE_V(x)   ((x) << FW_VI_MAC_CMD_ENTRY_TYPE_S)
+#define FW_VI_MAC_CMD_ENTRY_TYPE_G(x)	\
+	(((x) >> FW_VI_MAC_CMD_ENTRY_TYPE_S) & FW_VI_MAC_CMD_ENTRY_TYPE_M)
+
 #define FW_VI_MAC_CMD_HASHVECEN_S	23
 #define FW_VI_MAC_CMD_HASHVECEN_V(x)	((x) << FW_VI_MAC_CMD_HASHVECEN_S)
 #define FW_VI_MAC_CMD_HASHVECEN_F	FW_VI_MAC_CMD_HASHVECEN_V(1U)
@@ -2038,6 +2357,38 @@ struct fw_vi_mac_cmd {
 #define FW_VI_MAC_CMD_IDX_G(x)	\
 	(((x) >> FW_VI_MAC_CMD_IDX_S) & FW_VI_MAC_CMD_IDX_M)
 
+#define FW_VI_MAC_CMD_RAW_IDX_S         16
+#define FW_VI_MAC_CMD_RAW_IDX_M         0xffff
+#define FW_VI_MAC_CMD_RAW_IDX_V(x)      ((x) << FW_VI_MAC_CMD_RAW_IDX_S)
+#define FW_VI_MAC_CMD_RAW_IDX_G(x)      \
+	(((x) >> FW_VI_MAC_CMD_RAW_IDX_S) & FW_VI_MAC_CMD_RAW_IDX_M)
+
+#define FW_VI_MAC_CMD_LOOKUP_TYPE_S	31
+#define FW_VI_MAC_CMD_LOOKUP_TYPE_M	0x1
+#define FW_VI_MAC_CMD_LOOKUP_TYPE_V(x)	((x) << FW_VI_MAC_CMD_LOOKUP_TYPE_S)
+#define FW_VI_MAC_CMD_LOOKUP_TYPE_G(x)	\
+	(((x) >> FW_VI_MAC_CMD_LOOKUP_TYPE_S) & FW_VI_MAC_CMD_LOOKUP_TYPE_M)
+#define FW_VI_MAC_CMD_LOOKUP_TYPE_F	FW_VI_MAC_CMD_LOOKUP_TYPE_V(1U)
+
+#define FW_VI_MAC_CMD_DIP_HIT_S		30
+#define FW_VI_MAC_CMD_DIP_HIT_M		0x1
+#define FW_VI_MAC_CMD_DIP_HIT_V(x)	((x) << FW_VI_MAC_CMD_DIP_HIT_S)
+#define FW_VI_MAC_CMD_DIP_HIT_G(x)	\
+	(((x) >> FW_VI_MAC_CMD_DIP_HIT_S) & FW_VI_MAC_CMD_DIP_HIT_M)
+#define FW_VI_MAC_CMD_DIP_HIT_F		FW_VI_MAC_CMD_DIP_HIT_V(1U)
+
+#define FW_VI_MAC_CMD_VNI_S		0
+#define FW_VI_MAC_CMD_VNI_M		0xffffff
+#define FW_VI_MAC_CMD_VNI_V(x)		((x) << FW_VI_MAC_CMD_VNI_S)
+#define FW_VI_MAC_CMD_VNI_G(x)		\
+	(((x) >> FW_VI_MAC_CMD_VNI_S) & FW_VI_MAC_CMD_VNI_M)
+
+#define FW_VI_MAC_CMD_VNI_MASK_S	0
+#define FW_VI_MAC_CMD_VNI_MASK_M	0xffffff
+#define FW_VI_MAC_CMD_VNI_MASK_V(x)	((x) << FW_VI_MAC_CMD_VNI_MASK_S)
+#define FW_VI_MAC_CMD_VNI_MASK_G(x)	\
+	(((x) >> FW_VI_MAC_CMD_VNI_MASK_S) & FW_VI_MAC_CMD_VNI_MASK_M)
+
 #define FW_RXMODE_MTU_NO_CHG	65535
 
 struct fw_vi_rxmode_cmd {
@@ -2242,15 +2593,25 @@ struct fw_acl_vlan_cmd {
 #define FW_ACL_VLAN_CMD_VFN_S		0
 #define FW_ACL_VLAN_CMD_VFN_V(x)	((x) << FW_ACL_VLAN_CMD_VFN_S)
 
-#define FW_ACL_VLAN_CMD_EN_S	31
-#define FW_ACL_VLAN_CMD_EN_V(x)	((x) << FW_ACL_VLAN_CMD_EN_S)
+#define FW_ACL_VLAN_CMD_EN_S		31
+#define FW_ACL_VLAN_CMD_EN_M		0x1
+#define FW_ACL_VLAN_CMD_EN_V(x)		((x) << FW_ACL_VLAN_CMD_EN_S)
+#define FW_ACL_VLAN_CMD_EN_G(x)         \
+	(((x) >> S_FW_ACL_VLAN_CMD_EN_S) & FW_ACL_VLAN_CMD_EN_M)
+#define FW_ACL_VLAN_CMD_EN_F            FW_ACL_VLAN_CMD_EN_V(1U)
 
 #define FW_ACL_VLAN_CMD_DROPNOVLAN_S	7
 #define FW_ACL_VLAN_CMD_DROPNOVLAN_V(x)	((x) << FW_ACL_VLAN_CMD_DROPNOVLAN_S)
+#define FW_ACL_VLAN_CMD_DROPNOVLAN_F    FW_ACL_VLAN_CMD_DROPNOVLAN_V(1U)
 
-#define FW_ACL_VLAN_CMD_FM_S	6
-#define FW_ACL_VLAN_CMD_FM_V(x)	((x) << FW_ACL_VLAN_CMD_FM_S)
+#define FW_ACL_VLAN_CMD_FM_S		6
+#define FW_ACL_VLAN_CMD_FM_M		0x1
+#define FW_ACL_VLAN_CMD_FM_V(x)         ((x) << FW_ACL_VLAN_CMD_FM_S)
+#define FW_ACL_VLAN_CMD_FM_G(x)         \
+	(((x) >> FW_ACL_VLAN_CMD_FM_S) & FW_ACL_VLAN_CMD_FM_M)
+#define FW_ACL_VLAN_CMD_FM_F            FW_ACL_VLAN_CMD_FM_V(1U)
 
+/* old 16-bit port capabilities bitmap (fw_port_cap16_t) */
 enum fw_port_cap {
 	FW_PORT_CAP_SPEED_100M		= 0x0001,
 	FW_PORT_CAP_SPEED_1G		= 0x0002,
@@ -2261,11 +2622,11 @@ enum fw_port_cap {
 	FW_PORT_CAP_FC_RX		= 0x0040,
 	FW_PORT_CAP_FC_TX		= 0x0080,
 	FW_PORT_CAP_ANEG		= 0x0100,
-	FW_PORT_CAP_MDIX		= 0x0200,
-	FW_PORT_CAP_MDIAUTO		= 0x0400,
+	FW_PORT_CAP_MDIAUTO		= 0x0200,
+	FW_PORT_CAP_MDISTRAIGHT		= 0x0400,
 	FW_PORT_CAP_FEC_RS		= 0x0800,
 	FW_PORT_CAP_FEC_BASER_RS	= 0x1000,
-	FW_PORT_CAP_FEC_RESERVED	= 0x2000,
+	FW_PORT_CAP_FORCE_PAUSE		= 0x2000,
 	FW_PORT_CAP_802_3_PAUSE		= 0x4000,
 	FW_PORT_CAP_802_3_ASM_DIR	= 0x8000,
 };
@@ -2286,6 +2647,85 @@ enum fw_port_mdi {
 #define FW_PORT_CAP_MDI_S 9
 #define FW_PORT_CAP_MDI_V(x) ((x) << FW_PORT_CAP_MDI_S)
 
+/* new 32-bit port capabilities bitmap (fw_port_cap32_t) */
+#define	FW_PORT_CAP32_SPEED_100M	0x00000001UL
+#define	FW_PORT_CAP32_SPEED_1G		0x00000002UL
+#define	FW_PORT_CAP32_SPEED_10G		0x00000004UL
+#define	FW_PORT_CAP32_SPEED_25G		0x00000008UL
+#define	FW_PORT_CAP32_SPEED_40G		0x00000010UL
+#define	FW_PORT_CAP32_SPEED_50G		0x00000020UL
+#define	FW_PORT_CAP32_SPEED_100G	0x00000040UL
+#define	FW_PORT_CAP32_SPEED_200G	0x00000080UL
+#define	FW_PORT_CAP32_SPEED_400G	0x00000100UL
+#define	FW_PORT_CAP32_SPEED_RESERVED1	0x00000200UL
+#define	FW_PORT_CAP32_SPEED_RESERVED2	0x00000400UL
+#define	FW_PORT_CAP32_SPEED_RESERVED3	0x00000800UL
+#define	FW_PORT_CAP32_RESERVED1		0x0000f000UL
+#define	FW_PORT_CAP32_FC_RX		0x00010000UL
+#define	FW_PORT_CAP32_FC_TX		0x00020000UL
+#define	FW_PORT_CAP32_802_3_PAUSE	0x00040000UL
+#define	FW_PORT_CAP32_802_3_ASM_DIR	0x00080000UL
+#define	FW_PORT_CAP32_ANEG		0x00100000UL
+#define	FW_PORT_CAP32_MDIAUTO		0x00200000UL
+#define	FW_PORT_CAP32_MDISTRAIGHT	0x00400000UL
+#define	FW_PORT_CAP32_FEC_RS		0x00800000UL
+#define	FW_PORT_CAP32_FEC_BASER_RS	0x01000000UL
+#define	FW_PORT_CAP32_FEC_RESERVED1	0x02000000UL
+#define	FW_PORT_CAP32_FEC_RESERVED2	0x04000000UL
+#define	FW_PORT_CAP32_FEC_RESERVED3	0x08000000UL
+#define FW_PORT_CAP32_FORCE_PAUSE	0x10000000UL
+#define FW_PORT_CAP32_RESERVED2		0xe0000000UL
+
+#define FW_PORT_CAP32_SPEED_S	0
+#define FW_PORT_CAP32_SPEED_M	0xfff
+#define FW_PORT_CAP32_SPEED_V(x)	((x) << FW_PORT_CAP32_SPEED_S)
+#define FW_PORT_CAP32_SPEED_G(x) \
+	(((x) >> FW_PORT_CAP32_SPEED_S) & FW_PORT_CAP32_SPEED_M)
+
+#define FW_PORT_CAP32_FC_S	16
+#define FW_PORT_CAP32_FC_M	0x3
+#define FW_PORT_CAP32_FC_V(x)	((x) << FW_PORT_CAP32_FC_S)
+#define FW_PORT_CAP32_FC_G(x) \
+	(((x) >> FW_PORT_CAP32_FC_S) & FW_PORT_CAP32_FC_M)
+
+#define FW_PORT_CAP32_802_3_S	18
+#define FW_PORT_CAP32_802_3_M	0x3
+#define FW_PORT_CAP32_802_3_V(x)	((x) << FW_PORT_CAP32_802_3_S)
+#define FW_PORT_CAP32_802_3_G(x) \
+	(((x) >> FW_PORT_CAP32_802_3_S) & FW_PORT_CAP32_802_3_M)
+
+#define FW_PORT_CAP32_ANEG_S	20
+#define FW_PORT_CAP32_ANEG_M	0x1
+#define FW_PORT_CAP32_ANEG_V(x)	((x) << FW_PORT_CAP32_ANEG_S)
+#define FW_PORT_CAP32_ANEG_G(x) \
+	(((x) >> FW_PORT_CAP32_ANEG_S) & FW_PORT_CAP32_ANEG_M)
+
+enum fw_port_mdi32 {
+	FW_PORT_CAP32_MDI_UNCHANGED,
+	FW_PORT_CAP32_MDI_AUTO,
+	FW_PORT_CAP32_MDI_F_STRAIGHT,
+	FW_PORT_CAP32_MDI_F_CROSSOVER
+};
+
+#define FW_PORT_CAP32_MDI_S 21
+#define FW_PORT_CAP32_MDI_M 3
+#define FW_PORT_CAP32_MDI_V(x) ((x) << FW_PORT_CAP32_MDI_S)
+#define FW_PORT_CAP32_MDI_G(x) \
+	(((x) >> FW_PORT_CAP32_MDI_S) & FW_PORT_CAP32_MDI_M)
+
+#define FW_PORT_CAP32_FEC_S	23
+#define FW_PORT_CAP32_FEC_M	0x1f
+#define FW_PORT_CAP32_FEC_V(x)	((x) << FW_PORT_CAP32_FEC_S)
+#define FW_PORT_CAP32_FEC_G(x) \
+	(((x) >> FW_PORT_CAP32_FEC_S) & FW_PORT_CAP32_FEC_M)
+
+/* macros to isolate various 32-bit Port Capabilities sub-fields */
+#define CAP32_SPEED(__cap32) \
+	(FW_PORT_CAP32_SPEED_V(FW_PORT_CAP32_SPEED_M) & __cap32)
+
+#define CAP32_FEC(__cap32) \
+	(FW_PORT_CAP32_FEC_V(FW_PORT_CAP32_FEC_M) & __cap32)
+
 enum fw_port_action {
 	FW_PORT_ACTION_L1_CFG		= 0x0001,
 	FW_PORT_ACTION_L2_CFG		= 0x0002,
@@ -2295,6 +2735,8 @@ enum fw_port_action {
 	FW_PORT_ACTION_DCB_READ_TRANS	= 0x0006,
 	FW_PORT_ACTION_DCB_READ_RECV	= 0x0007,
 	FW_PORT_ACTION_DCB_READ_DET	= 0x0008,
+	FW_PORT_ACTION_L1_CFG32		= 0x0009,
+	FW_PORT_ACTION_GET_PORT_INFO32	= 0x000a,
 	FW_PORT_ACTION_LOW_PWR_TO_NORMAL = 0x0010,
 	FW_PORT_ACTION_L1_LOW_PWR_EN	= 0x0011,
 	FW_PORT_ACTION_L2_WOL_MODE_EN	= 0x0012,
@@ -2442,6 +2884,18 @@ struct fw_port_cmd {
 				__be64 r12;
 			} control;
 		} dcb;
+		struct fw_port_l1cfg32 {
+			__be32 rcap32;
+			__be32 r;
+		} l1cfg32;
+		struct fw_port_info32 {
+			__be32 lstatus32_to_cbllen32;
+			__be32 auxlinfo32_mtu32;
+			__be32 linkattr32;
+			__be32 pcaps32;
+			__be32 acaps32;
+			__be32 lpacaps32;
+		} info32;
 	} u;
 };
 
@@ -2550,6 +3004,85 @@ struct fw_port_cmd {
 #define FW_PORT_CMD_DCB_VERSION_G(x)	\
 	(((x) >> FW_PORT_CMD_DCB_VERSION_S) & FW_PORT_CMD_DCB_VERSION_M)
 
+#define FW_PORT_CMD_LSTATUS32_S		31
+#define FW_PORT_CMD_LSTATUS32_M		0x1
+#define FW_PORT_CMD_LSTATUS32_V(x)	((x) << FW_PORT_CMD_LSTATUS32_S)
+#define FW_PORT_CMD_LSTATUS32_G(x)	\
+	(((x) >> FW_PORT_CMD_LSTATUS32_S) & FW_PORT_CMD_LSTATUS32_M)
+#define FW_PORT_CMD_LSTATUS32_F	FW_PORT_CMD_LSTATUS32_V(1U)
+
+#define FW_PORT_CMD_LINKDNRC32_S	28
+#define FW_PORT_CMD_LINKDNRC32_M	0x7
+#define FW_PORT_CMD_LINKDNRC32_V(x)	((x) << FW_PORT_CMD_LINKDNRC32_S)
+#define FW_PORT_CMD_LINKDNRC32_G(x)	\
+	(((x) >> FW_PORT_CMD_LINKDNRC32_S) & FW_PORT_CMD_LINKDNRC32_M)
+
+#define FW_PORT_CMD_DCBXDIS32_S		27
+#define FW_PORT_CMD_DCBXDIS32_M		0x1
+#define FW_PORT_CMD_DCBXDIS32_V(x)	((x) << FW_PORT_CMD_DCBXDIS32_S)
+#define FW_PORT_CMD_DCBXDIS32_G(x)	\
+	(((x) >> FW_PORT_CMD_DCBXDIS32_S) & FW_PORT_CMD_DCBXDIS32_M)
+#define FW_PORT_CMD_DCBXDIS32_F	FW_PORT_CMD_DCBXDIS32_V(1U)
+
+#define FW_PORT_CMD_MDIOCAP32_S		26
+#define FW_PORT_CMD_MDIOCAP32_M		0x1
+#define FW_PORT_CMD_MDIOCAP32_V(x)	((x) << FW_PORT_CMD_MDIOCAP32_S)
+#define FW_PORT_CMD_MDIOCAP32_G(x)	\
+	(((x) >> FW_PORT_CMD_MDIOCAP32_S) & FW_PORT_CMD_MDIOCAP32_M)
+#define FW_PORT_CMD_MDIOCAP32_F	FW_PORT_CMD_MDIOCAP32_V(1U)
+
+#define FW_PORT_CMD_MDIOADDR32_S	21
+#define FW_PORT_CMD_MDIOADDR32_M	0x1f
+#define FW_PORT_CMD_MDIOADDR32_V(x)	((x) << FW_PORT_CMD_MDIOADDR32_S)
+#define FW_PORT_CMD_MDIOADDR32_G(x)	\
+	(((x) >> FW_PORT_CMD_MDIOADDR32_S) & FW_PORT_CMD_MDIOADDR32_M)
+
+#define FW_PORT_CMD_PORTTYPE32_S	13
+#define FW_PORT_CMD_PORTTYPE32_M	0xff
+#define FW_PORT_CMD_PORTTYPE32_V(x)	((x) << FW_PORT_CMD_PORTTYPE32_S)
+#define FW_PORT_CMD_PORTTYPE32_G(x)	\
+	(((x) >> FW_PORT_CMD_PORTTYPE32_S) & FW_PORT_CMD_PORTTYPE32_M)
+
+#define FW_PORT_CMD_MODTYPE32_S		8
+#define FW_PORT_CMD_MODTYPE32_M		0x1f
+#define FW_PORT_CMD_MODTYPE32_V(x)	((x) << FW_PORT_CMD_MODTYPE32_S)
+#define FW_PORT_CMD_MODTYPE32_G(x)	\
+	(((x) >> FW_PORT_CMD_MODTYPE32_S) & FW_PORT_CMD_MODTYPE32_M)
+
+#define FW_PORT_CMD_CBLLEN32_S		0
+#define FW_PORT_CMD_CBLLEN32_M		0xff
+#define FW_PORT_CMD_CBLLEN32_V(x)	((x) << FW_PORT_CMD_CBLLEN32_S)
+#define FW_PORT_CMD_CBLLEN32_G(x)	\
+	(((x) >> FW_PORT_CMD_CBLLEN32_S) & FW_PORT_CMD_CBLLEN32_M)
+
+#define FW_PORT_CMD_AUXLINFO32_S	24
+#define FW_PORT_CMD_AUXLINFO32_M	0xff
+#define FW_PORT_CMD_AUXLINFO32_V(x)	((x) << FW_PORT_CMD_AUXLINFO32_S)
+#define FW_PORT_CMD_AUXLINFO32_G(x)	\
+	(((x) >> FW_PORT_CMD_AUXLINFO32_S) & FW_PORT_CMD_AUXLINFO32_M)
+
+#define FW_PORT_AUXLINFO32_KX4_S	2
+#define FW_PORT_AUXLINFO32_KX4_M	0x1
+#define FW_PORT_AUXLINFO32_KX4_V(x) \
+	((x) << FW_PORT_AUXLINFO32_KX4_S)
+#define FW_PORT_AUXLINFO32_KX4_G(x) \
+	(((x) >> FW_PORT_AUXLINFO32_KX4_S) & FW_PORT_AUXLINFO32_KX4_M)
+#define FW_PORT_AUXLINFO32_KX4_F	FW_PORT_AUXLINFO32_KX4_V(1U)
+
+#define FW_PORT_AUXLINFO32_KR_S	1
+#define FW_PORT_AUXLINFO32_KR_M	0x1
+#define FW_PORT_AUXLINFO32_KR_V(x) \
+	((x) << FW_PORT_AUXLINFO32_KR_S)
+#define FW_PORT_AUXLINFO32_KR_G(x) \
+	(((x) >> FW_PORT_AUXLINFO32_KR_S) & FW_PORT_AUXLINFO32_KR_M)
+#define FW_PORT_AUXLINFO32_KR_F	FW_PORT_AUXLINFO32_KR_V(1U)
+
+#define FW_PORT_CMD_MTU32_S	0
+#define FW_PORT_CMD_MTU32_M	0xffff
+#define FW_PORT_CMD_MTU32_V(x)	((x) << FW_PORT_CMD_MTU32_S)
+#define FW_PORT_CMD_MTU32_G(x)	\
+	(((x) >> FW_PORT_CMD_MTU32_S) & FW_PORT_CMD_MTU32_M)
+
 enum fw_port_type {
 	FW_PORT_TYPE_FIBER_XFI,
 	FW_PORT_TYPE_FIBER_XAUI,
@@ -2573,6 +3106,7 @@ enum fw_port_type {
 	FW_PORT_TYPE_CR2_QSFP,
 	FW_PORT_TYPE_SFP28,
 	FW_PORT_TYPE_KR_SFP28,
+	FW_PORT_TYPE_KR_XLAUI,
 
 	FW_PORT_TYPE_NONE = FW_PORT_CMD_PTYPE_M
 };
@@ -2801,6 +3335,54 @@ struct fw_port_lb_stats_cmd {
 	} u;
 };
 
+enum fw_ptp_subop {
+	/* none */
+	FW_PTP_SC_INIT_TIMER            = 0x00,
+	FW_PTP_SC_TX_TYPE               = 0x01,
+	/* init */
+	FW_PTP_SC_RXTIME_STAMP          = 0x08,
+	FW_PTP_SC_RDRX_TYPE             = 0x09,
+	/* ts */
+	FW_PTP_SC_ADJ_FREQ              = 0x10,
+	FW_PTP_SC_ADJ_TIME              = 0x11,
+	FW_PTP_SC_ADJ_FTIME             = 0x12,
+	FW_PTP_SC_WALL_CLOCK            = 0x13,
+	FW_PTP_SC_GET_TIME              = 0x14,
+	FW_PTP_SC_SET_TIME              = 0x15,
+};
+
+struct fw_ptp_cmd {
+	__be32 op_to_portid;
+	__be32 retval_len16;
+	union fw_ptp {
+		struct fw_ptp_sc {
+			__u8   sc;
+			__u8   r3[7];
+		} scmd;
+		struct fw_ptp_init {
+			__u8   sc;
+			__u8   txchan;
+			__be16 absid;
+			__be16 mode;
+			__be16 r3;
+		} init;
+		struct fw_ptp_ts {
+			__u8   sc;
+			__u8   sign;
+			__be16 r3;
+			__be32 ppb;
+			__be64 tm;
+		} ts;
+	} u;
+	__be64 r3;
+};
+
+#define FW_PTP_CMD_PORTID_S             0
+#define FW_PTP_CMD_PORTID_M             0xf
+#define FW_PTP_CMD_PORTID_V(x)          ((x) << FW_PTP_CMD_PORTID_S)
+#define FW_PTP_CMD_PORTID_G(x)          \
+	(((x) >> FW_PTP_CMD_PORTID_S) & FW_PTP_CMD_PORTID_M)
+
 struct fw_rss_ind_tbl_cmd {
 	__be32 op_to_viid;
 	__be32 retval_len16;
@@ -3088,6 +3670,63 @@ struct fw_debug_cmd {
 #define FW_DEBUG_CMD_TYPE_G(x)	\
 	(((x) >> FW_DEBUG_CMD_TYPE_S) & FW_DEBUG_CMD_TYPE_M)
 
+struct fw_hma_cmd {
+	__be32 op_pkd;
+	__be32 retval_len16;
+	__be32 mode_to_pcie_params;
+	__be32 naddr_size;
+	__be32 addr_size_pkd;
+	__be32 r6;
+	__be64 phy_address[5];
+};
+
+#define FW_HMA_CMD_MODE_S	31
+#define FW_HMA_CMD_MODE_M	0x1
+#define FW_HMA_CMD_MODE_V(x)	((x) << FW_HMA_CMD_MODE_S)
+#define FW_HMA_CMD_MODE_G(x)	\
+	(((x) >> FW_HMA_CMD_MODE_S) & FW_HMA_CMD_MODE_M)
+#define FW_HMA_CMD_MODE_F	FW_HMA_CMD_MODE_V(1U)
+
+#define FW_HMA_CMD_SOC_S	30
+#define FW_HMA_CMD_SOC_M	0x1
+#define FW_HMA_CMD_SOC_V(x)	((x) << FW_HMA_CMD_SOC_S)
+#define FW_HMA_CMD_SOC_G(x)	(((x) >> FW_HMA_CMD_SOC_S) & FW_HMA_CMD_SOC_M)
+#define FW_HMA_CMD_SOC_F	FW_HMA_CMD_SOC_V(1U)
+
+#define FW_HMA_CMD_EOC_S	29
+#define FW_HMA_CMD_EOC_M	0x1
+#define FW_HMA_CMD_EOC_V(x)	((x) << FW_HMA_CMD_EOC_S)
+#define FW_HMA_CMD_EOC_G(x)	(((x) >> FW_HMA_CMD_EOC_S) & FW_HMA_CMD_EOC_M)
+#define FW_HMA_CMD_EOC_F	FW_HMA_CMD_EOC_V(1U)
+
+#define FW_HMA_CMD_PCIE_PARAMS_S	0
+#define FW_HMA_CMD_PCIE_PARAMS_M	0x7ffffff
+#define FW_HMA_CMD_PCIE_PARAMS_V(x)	((x) << FW_HMA_CMD_PCIE_PARAMS_S)
+#define FW_HMA_CMD_PCIE_PARAMS_G(x)	\
+	(((x) >> FW_HMA_CMD_PCIE_PARAMS_S) & FW_HMA_CMD_PCIE_PARAMS_M)
+
+#define FW_HMA_CMD_NADDR_S	12
+#define FW_HMA_CMD_NADDR_M	0x3f
+#define FW_HMA_CMD_NADDR_V(x)	((x) << FW_HMA_CMD_NADDR_S)
+#define FW_HMA_CMD_NADDR_G(x)	\
+	(((x) >> FW_HMA_CMD_NADDR_S) & FW_HMA_CMD_NADDR_M)
+
+#define FW_HMA_CMD_SIZE_S	0
+#define FW_HMA_CMD_SIZE_M	0xfff
+#define FW_HMA_CMD_SIZE_V(x)	((x) << FW_HMA_CMD_SIZE_S)
+#define FW_HMA_CMD_SIZE_G(x)	\
+	(((x) >> FW_HMA_CMD_SIZE_S) & FW_HMA_CMD_SIZE_M)
+
+#define FW_HMA_CMD_ADDR_SIZE_S		11
+#define FW_HMA_CMD_ADDR_SIZE_M		0x1fffff
+#define FW_HMA_CMD_ADDR_SIZE_V(x)	((x) << FW_HMA_CMD_ADDR_SIZE_S)
+#define FW_HMA_CMD_ADDR_SIZE_G(x)	\
+	(((x) >> FW_HMA_CMD_ADDR_SIZE_S) & FW_HMA_CMD_ADDR_SIZE_M)
+
+enum pcie_fw_eval {
+	PCIE_FW_EVAL_CRASH = 0,
+};
+
 #define PCIE_FW_ERR_S		31
 #define PCIE_FW_ERR_V(x)	((x) << PCIE_FW_ERR_S)
 #define PCIE_FW_ERR_F		PCIE_FW_ERR_V(1U)
@@ -3427,4 +4066,122 @@ struct fw_crypto_lookaside_wr {
 	(((x) >> FW_CRYPTO_LOOKASIDE_WR_HASH_SIZE_S) & \
 	 FW_CRYPTO_LOOKASIDE_WR_HASH_SIZE_M)
 
+struct fw_tlstx_data_wr {
+	__be32 op_to_immdlen;
+	__be32 flowid_len16;
+	__be32 plen;
+	__be32 lsodisable_to_flags;
+	__be32 r5;
+	__be32 ctxloc_to_exp;
+	__be16 mfs;
+	__be16 adjustedplen_pkd;
+	__be16 expinplenmax_pkd;
+	u8   pdusinplenmax_pkd;
+	u8   r10;
+};
+
+#define FW_TLSTX_DATA_WR_OPCODE_S       24
+#define FW_TLSTX_DATA_WR_OPCODE_M       0xff
+#define FW_TLSTX_DATA_WR_OPCODE_V(x)    ((x) << FW_TLSTX_DATA_WR_OPCODE_S)
+#define FW_TLSTX_DATA_WR_OPCODE_G(x)    \
+	(((x) >> FW_TLSTX_DATA_WR_OPCODE_S) & FW_TLSTX_DATA_WR_OPCODE_M)
+
+#define FW_TLSTX_DATA_WR_COMPL_S        21
+#define FW_TLSTX_DATA_WR_COMPL_M        0x1
+#define FW_TLSTX_DATA_WR_COMPL_V(x)     ((x) << FW_TLSTX_DATA_WR_COMPL_S)
+#define FW_TLSTX_DATA_WR_COMPL_G(x)     \
+	(((x) >> FW_TLSTX_DATA_WR_COMPL_S) & FW_TLSTX_DATA_WR_COMPL_M)
+#define FW_TLSTX_DATA_WR_COMPL_F        FW_TLSTX_DATA_WR_COMPL_V(1U)
+
+#define FW_TLSTX_DATA_WR_IMMDLEN_S      0
+#define FW_TLSTX_DATA_WR_IMMDLEN_M      0xff
+#define FW_TLSTX_DATA_WR_IMMDLEN_V(x)   ((x) << FW_TLSTX_DATA_WR_IMMDLEN_S)
+#define FW_TLSTX_DATA_WR_IMMDLEN_G(x)   \
+	(((x) >> FW_TLSTX_DATA_WR_IMMDLEN_S) & FW_TLSTX_DATA_WR_IMMDLEN_M)
+
+#define FW_TLSTX_DATA_WR_FLOWID_S       8
+#define FW_TLSTX_DATA_WR_FLOWID_M       0xfffff
+#define FW_TLSTX_DATA_WR_FLOWID_V(x)    ((x) << FW_TLSTX_DATA_WR_FLOWID_S)
+#define FW_TLSTX_DATA_WR_FLOWID_G(x)    \
+	(((x) >> FW_TLSTX_DATA_WR_FLOWID_S) & FW_TLSTX_DATA_WR_FLOWID_M)
+
+#define FW_TLSTX_DATA_WR_LEN16_S        0
+#define FW_TLSTX_DATA_WR_LEN16_M        0xff
+#define FW_TLSTX_DATA_WR_LEN16_V(x)     ((x) << FW_TLSTX_DATA_WR_LEN16_S)
+#define FW_TLSTX_DATA_WR_LEN16_G(x)     \
+	(((x) >> FW_TLSTX_DATA_WR_LEN16_S) & FW_TLSTX_DATA_WR_LEN16_M)
+
+#define FW_TLSTX_DATA_WR_LSODISABLE_S   31
+#define FW_TLSTX_DATA_WR_LSODISABLE_M   0x1
+#define FW_TLSTX_DATA_WR_LSODISABLE_V(x) \
+	((x) << FW_TLSTX_DATA_WR_LSODISABLE_S)
+#define FW_TLSTX_DATA_WR_LSODISABLE_G(x) \
+	(((x) >> FW_TLSTX_DATA_WR_LSODISABLE_S) & FW_TLSTX_DATA_WR_LSODISABLE_M)
+#define FW_TLSTX_DATA_WR_LSODISABLE_F   FW_TLSTX_DATA_WR_LSODISABLE_V(1U)
+
+#define FW_TLSTX_DATA_WR_ALIGNPLD_S     30
+#define FW_TLSTX_DATA_WR_ALIGNPLD_M     0x1
+#define FW_TLSTX_DATA_WR_ALIGNPLD_V(x)  ((x) << FW_TLSTX_DATA_WR_ALIGNPLD_S)
+#define FW_TLSTX_DATA_WR_ALIGNPLD_G(x)  \
+	(((x) >> FW_TLSTX_DATA_WR_ALIGNPLD_S) & FW_TLSTX_DATA_WR_ALIGNPLD_M)
+#define FW_TLSTX_DATA_WR_ALIGNPLD_F     FW_TLSTX_DATA_WR_ALIGNPLD_V(1U)
+
+#define FW_TLSTX_DATA_WR_ALIGNPLDSHOVE_S 29
+#define FW_TLSTX_DATA_WR_ALIGNPLDSHOVE_M 0x1
+#define FW_TLSTX_DATA_WR_ALIGNPLDSHOVE_V(x) \
+	((x) << FW_TLSTX_DATA_WR_ALIGNPLDSHOVE_S)
+#define FW_TLSTX_DATA_WR_ALIGNPLDSHOVE_G(x) \
+	(((x) >> FW_TLSTX_DATA_WR_ALIGNPLDSHOVE_S) & \
+	FW_TLSTX_DATA_WR_ALIGNPLDSHOVE_M)
+#define FW_TLSTX_DATA_WR_ALIGNPLDSHOVE_F FW_TLSTX_DATA_WR_ALIGNPLDSHOVE_V(1U)
+
+#define FW_TLSTX_DATA_WR_FLAGS_S        0
+#define FW_TLSTX_DATA_WR_FLAGS_M        0xfffffff
+#define FW_TLSTX_DATA_WR_FLAGS_V(x)     ((x) << FW_TLSTX_DATA_WR_FLAGS_S)
+#define FW_TLSTX_DATA_WR_FLAGS_G(x)     \
+	(((x) >> FW_TLSTX_DATA_WR_FLAGS_S) & FW_TLSTX_DATA_WR_FLAGS_M)
+
+#define FW_TLSTX_DATA_WR_CTXLOC_S       30
+#define FW_TLSTX_DATA_WR_CTXLOC_M       0x3
+#define FW_TLSTX_DATA_WR_CTXLOC_V(x)    ((x) << FW_TLSTX_DATA_WR_CTXLOC_S)
+#define FW_TLSTX_DATA_WR_CTXLOC_G(x)    \
+	(((x) >> FW_TLSTX_DATA_WR_CTXLOC_S) & FW_TLSTX_DATA_WR_CTXLOC_M)
+
+#define FW_TLSTX_DATA_WR_IVDSGL_S       29
+#define FW_TLSTX_DATA_WR_IVDSGL_M       0x1
+#define FW_TLSTX_DATA_WR_IVDSGL_V(x)    ((x) << FW_TLSTX_DATA_WR_IVDSGL_S)
+#define FW_TLSTX_DATA_WR_IVDSGL_G(x)    \
+	(((x) >> FW_TLSTX_DATA_WR_IVDSGL_S) & FW_TLSTX_DATA_WR_IVDSGL_M)
+#define FW_TLSTX_DATA_WR_IVDSGL_F       FW_TLSTX_DATA_WR_IVDSGL_V(1U)
+
+#define FW_TLSTX_DATA_WR_KEYSIZE_S      24
+#define FW_TLSTX_DATA_WR_KEYSIZE_M      0x1f
+#define FW_TLSTX_DATA_WR_KEYSIZE_V(x)   ((x) << FW_TLSTX_DATA_WR_KEYSIZE_S)
+#define FW_TLSTX_DATA_WR_KEYSIZE_G(x)   \
+	(((x) >> FW_TLSTX_DATA_WR_KEYSIZE_S) & FW_TLSTX_DATA_WR_KEYSIZE_M)
+
+#define FW_TLSTX_DATA_WR_NUMIVS_S       14
+#define FW_TLSTX_DATA_WR_NUMIVS_M       0xff
+#define FW_TLSTX_DATA_WR_NUMIVS_V(x)    ((x) << FW_TLSTX_DATA_WR_NUMIVS_S)
+#define FW_TLSTX_DATA_WR_NUMIVS_G(x)    \
+	(((x) >> FW_TLSTX_DATA_WR_NUMIVS_S) & FW_TLSTX_DATA_WR_NUMIVS_M)
+
+#define FW_TLSTX_DATA_WR_EXP_S          0
+#define FW_TLSTX_DATA_WR_EXP_M          0x3fff
+#define FW_TLSTX_DATA_WR_EXP_V(x)       ((x) << FW_TLSTX_DATA_WR_EXP_S)
+#define FW_TLSTX_DATA_WR_EXP_G(x)       \
+	(((x) >> FW_TLSTX_DATA_WR_EXP_S) & FW_TLSTX_DATA_WR_EXP_M)
+
+#define FW_TLSTX_DATA_WR_ADJUSTEDPLEN_S 1
+#define FW_TLSTX_DATA_WR_ADJUSTEDPLEN_V(x) \
+	((x) << FW_TLSTX_DATA_WR_ADJUSTEDPLEN_S)
+
+#define FW_TLSTX_DATA_WR_EXPINPLENMAX_S 4
+#define FW_TLSTX_DATA_WR_EXPINPLENMAX_V(x) \
+	((x) << FW_TLSTX_DATA_WR_EXPINPLENMAX_S)
+
+#define FW_TLSTX_DATA_WR_PDUSINPLENMAX_S 2
+#define FW_TLSTX_DATA_WR_PDUSINPLENMAX_V(x) \
+	((x) << FW_TLSTX_DATA_WR_PDUSINPLENMAX_S)
+
 #endif /* _T4FW_INTERFACE_H_ */
diff --git a/drivers/net/ethernet/chelsio/cxgb4/t4fw_version.h b/drivers/net/ethernet/chelsio/cxgb4/t4fw_version.h
index 3549d3876278..a02b1dff403e 100644
--- a/drivers/net/ethernet/chelsio/cxgb4/t4fw_version.h
+++ b/drivers/net/ethernet/chelsio/cxgb4/t4fw_version.h
@@ -36,8 +36,8 @@
 #define __T4FW_VERSION_H__
 
 #define T4FW_VERSION_MAJOR 0x01
-#define T4FW_VERSION_MINOR 0x10
-#define T4FW_VERSION_MICRO 0x2B
+#define T4FW_VERSION_MINOR 0x17
+#define T4FW_VERSION_MICRO 0x03
 #define T4FW_VERSION_BUILD 0x00
 
 #define T4FW_MIN_VERSION_MAJOR 0x01
@@ -45,8 +45,8 @@
 #define T4FW_MIN_VERSION_MICRO 0x00
 
 #define T5FW_VERSION_MAJOR 0x01
-#define T5FW_VERSION_MINOR 0x10
-#define T5FW_VERSION_MICRO 0x2B
+#define T5FW_VERSION_MINOR 0x17
+#define T5FW_VERSION_MICRO 0x03
 #define T5FW_VERSION_BUILD 0x00
 
 #define T5FW_MIN_VERSION_MAJOR 0x00
@@ -54,8 +54,8 @@
 #define T5FW_MIN_VERSION_MICRO 0x00
 
 #define T6FW_VERSION_MAJOR 0x01
-#define T6FW_VERSION_MINOR 0x10
-#define T6FW_VERSION_MICRO 0x2B
+#define T6FW_VERSION_MINOR 0x17
+#define T6FW_VERSION_MICRO 0x03
 #define T6FW_VERSION_BUILD 0x00
 
 #define T6FW_MIN_VERSION_MAJOR 0x00
diff --git a/drivers/net/ethernet/chelsio/cxgb4vf/Makefile b/drivers/net/ethernet/chelsio/cxgb4vf/Makefile
index d72ee26cb4c7..f527ab13a008 100644
--- a/drivers/net/ethernet/chelsio/cxgb4vf/Makefile
+++ b/drivers/net/ethernet/chelsio/cxgb4vf/Makefile
@@ -1,3 +1,4 @@
+# SPDX-License-Identifier: GPL-2.0-only
 #
 # Chelsio T4 SR-IOV Virtual Function Driver
 #
diff --git a/drivers/net/ethernet/chelsio/cxgb4vf/adapter.h b/drivers/net/ethernet/chelsio/cxgb4vf/adapter.h
index 109bc630408b..03cb1410d6fc 100644
--- a/drivers/net/ethernet/chelsio/cxgb4vf/adapter.h
+++ b/drivers/net/ethernet/chelsio/cxgb4vf/adapter.h
@@ -40,6 +40,7 @@
 #ifndef __CXGB4VF_ADAPTER_H__
 #define __CXGB4VF_ADAPTER_H__
 
+#include <linux/etherdevice.h>
 #include <linux/interrupt.h>
 #include <linux/pci.h>
 #include <linux/spinlock.h>
@@ -92,8 +93,9 @@ struct sge_rspq;
  */
 struct port_info {
 	struct adapter *adapter;	/* our adapter */
+	u32 vlan_id;			/* vlan id for VST */
 	u16 viid;			/* virtual interface ID */
-	s16 xact_addr_filt;		/* index of our MAC address filter */
+	int xact_addr_filt;		/* index of our MAC address filter */
 	u16 rss_size;			/* size of VI's RSS table slice */
 	u8 pidx;			/* index into adapter port[] */
 	s8 mdio_addr;
@@ -351,6 +353,7 @@ struct sge {
 struct hash_mac_addr {
 	struct list_head list;
 	u8 addr[ETH_ALEN];
+	unsigned int iface_mac;
 };
 
 struct mbox_list {
@@ -404,10 +407,12 @@ struct adapter {
 };
 
 enum { /* adapter flags */
-	FULL_INIT_DONE     = (1UL << 0),
-	USING_MSI          = (1UL << 1),
-	USING_MSIX         = (1UL << 2),
-	QUEUES_BOUND       = (1UL << 3),
+	CXGB4VF_FULL_INIT_DONE			= (1UL << 0),
+	CXGB4VF_USING_MSI			= (1UL << 1),
+	CXGB4VF_USING_MSIX			= (1UL << 2),
+	CXGB4VF_QUEUES_BOUND			= (1UL << 3),
+	CXGB4VF_ROOT_NO_RELAXED_ORDERING	= (1UL << 4),
+	CXGB4VF_FW_OK				= (1UL << 5),
 };
 
 /*
@@ -503,7 +508,7 @@ static inline const char *port_name(struct adapter *adapter, int pidx)
 static inline void t4_os_set_hw_addr(struct adapter *adapter, int pidx,
 				     u8 hw_addr[])
 {
-	memcpy(adapter->port[pidx]->dev_addr, hw_addr, ETH_ALEN);
+	eth_hw_addr_set(adapter->port[pidx], hw_addr);
 }
 
 /**
@@ -558,7 +563,7 @@ int t4vf_sge_alloc_eth_txq(struct adapter *, struct sge_eth_txq *,
 			   unsigned int);
 void t4vf_free_sge_resources(struct adapter *);
 
-int t4vf_eth_xmit(struct sk_buff *, struct net_device *);
+netdev_tx_t t4vf_eth_xmit(struct sk_buff *, struct net_device *);
 int t4vf_ethrx_handler(struct sge_rspq *, const __be64 *,
 		       const struct pkt_gl *);
 
diff --git a/drivers/net/ethernet/chelsio/cxgb4vf/cxgb4vf_main.c b/drivers/net/ethernet/chelsio/cxgb4vf/cxgb4vf_main.c
index ac7a150c54e9..2fbe0f059a0b 100644
--- a/drivers/net/ethernet/chelsio/cxgb4vf/cxgb4vf_main.c
+++ b/drivers/net/ethernet/chelsio/cxgb4vf/cxgb4vf_main.c
@@ -55,7 +55,6 @@
 /*
  * Generic information about the driver.
  */
-#define DRV_VERSION "2.0.0-ko"
 #define DRV_DESC "Chelsio T4/T5/T6 Virtual Function (VF) Network Driver"
 
 /*
@@ -182,7 +181,7 @@ void t4vf_os_link_changed(struct adapter *adapter, int pidx, int link_ok)
 			break;
 		}
 
-		switch (pi->link_cfg.fc) {
+		switch ((int)pi->link_cfg.fc) {
 		case PAUSE_RX:
 			fc = "RX";
 			break;
@@ -191,7 +190,7 @@ void t4vf_os_link_changed(struct adapter *adapter, int pidx, int link_ok)
 			fc = "TX";
 			break;
 
-		case PAUSE_RX|PAUSE_TX:
+		case PAUSE_RX | PAUSE_TX:
 			fc = "RX/TX";
 			break;
 
@@ -239,6 +238,72 @@ void t4vf_os_portmod_changed(struct adapter *adapter, int pidx)
 			 "inserted\n", dev->name, pi->mod_type);
 }
 
+static int cxgb4vf_set_addr_hash(struct port_info *pi)
+{
+	struct adapter *adapter = pi->adapter;
+	u64 vec = 0;
+	bool ucast = false;
+	struct hash_mac_addr *entry;
+
+	/* Calculate the hash vector for the updated list and program it */
+	list_for_each_entry(entry, &adapter->mac_hlist, list) {
+		ucast |= is_unicast_ether_addr(entry->addr);
+		vec |= (1ULL << hash_mac_addr(entry->addr));
+	}
+	return t4vf_set_addr_hash(adapter, pi->viid, ucast, vec, false);
+}
+
+/**
+ *	cxgb4vf_change_mac - Update match filter for a MAC address.
+ *	@pi: the port_info
+ *	@viid: the VI id
+ *	@tcam_idx: TCAM index of existing filter for old value of MAC address,
+ *		   or -1
+ *	@addr: the new MAC address value
+ *	@persistent: whether a new MAC allocation should be persistent
+ *
+ *	Modifies an MPS filter and sets it to the new MAC address if
+ *	@tcam_idx >= 0, or adds the MAC address to a new filter if
+ *	@tcam_idx < 0. In the latter case the address is added persistently
+ *	if @persist is %true.
+ *	Addresses are programmed to hash region, if tcam runs out of entries.
+ *
+ */
+static int cxgb4vf_change_mac(struct port_info *pi, unsigned int viid,
+			      int *tcam_idx, const u8 *addr, bool persistent)
+{
+	struct hash_mac_addr *new_entry, *entry;
+	struct adapter *adapter = pi->adapter;
+	int ret;
+
+	ret = t4vf_change_mac(adapter, viid, *tcam_idx, addr, persistent);
+	/* We ran out of TCAM entries. try programming hash region. */
+	if (ret == -ENOMEM) {
+		/* If the MAC address to be updated is in the hash addr
+		 * list, update it from the list
+		 */
+		list_for_each_entry(entry, &adapter->mac_hlist, list) {
+			if (entry->iface_mac) {
+				ether_addr_copy(entry->addr, addr);
+				goto set_hash;
+			}
+		}
+		new_entry = kzalloc(sizeof(*new_entry), GFP_KERNEL);
+		if (!new_entry)
+			return -ENOMEM;
+		ether_addr_copy(new_entry->addr, addr);
+		new_entry->iface_mac = true;
+		list_add_tail(&new_entry->list, &adapter->mac_hlist);
+set_hash:
+		ret = cxgb4vf_set_addr_hash(pi);
+	} else if (ret >= 0) {
+		*tcam_idx = ret;
+		ret = 0;
+	}
+
+	return ret;
+}
+
 /*
  * Net device operations.
  * ======================
@@ -262,14 +327,10 @@ static int link_start(struct net_device *dev)
 	 */
 	ret = t4vf_set_rxmode(pi->adapter, pi->viid, dev->mtu, -1, -1, -1, 1,
 			      true);
-	if (ret == 0) {
-		ret = t4vf_change_mac(pi->adapter, pi->viid,
-				      pi->xact_addr_filt, dev->dev_addr, true);
-		if (ret >= 0) {
-			pi->xact_addr_filt = ret;
-			ret = 0;
-		}
-	}
+	if (ret == 0)
+		ret = cxgb4vf_change_mac(pi, pi->viid,
+					 &pi->xact_addr_filt,
+					 dev->dev_addr, true);
 
 	/*
 	 * We don't need to actually "start the link" itself since the
@@ -277,7 +338,8 @@ static int link_start(struct net_device *dev)
 	 * is enabled on a port.
 	 */
 	if (ret == 0)
-		ret = t4vf_enable_vi(pi->adapter, pi->viid, true, true);
+		ret = t4vf_enable_pi(pi->adapter, pi, true, true);
+
 	return ret;
 }
 
@@ -398,7 +460,7 @@ static void enable_rx(struct adapter *adapter)
 	 * The interrupt queue doesn't use NAPI so we do the 0-increment of
 	 * its Going To Sleep register here to get it started.
 	 */
-	if (adapter->flags & USING_MSI)
+	if (adapter->flags & CXGB4VF_USING_MSI)
 		t4_write_reg(adapter, T4VF_SGE_BASE_ADDR + SGE_VF_GTS,
 			     CIDXINC_V(0) |
 			     SEINTARM_V(s->intrq.intr_params) |
@@ -454,8 +516,8 @@ static int fwevtq_handler(struct sge_rspq *rspq, const __be64 *rsp,
 			break;
 		}
 		cpl = (void *)p;
-		/*FALLTHROUGH*/
 	}
+		fallthrough;
 
 	case CPL_SGE_EGR_UPDATE: {
 		/*
@@ -542,7 +604,7 @@ static int setup_sge_queues(struct adapter *adapter)
 	 * the intrq's queue ID as the interrupt forwarding queue for the
 	 * subsequent calls ...
 	 */
-	if (adapter->flags & USING_MSI) {
+	if (adapter->flags & CXGB4VF_USING_MSI) {
 		err = t4vf_sge_alloc_rxq(adapter, &s->intrq, false,
 					 adapter->port[0], 0, NULL, NULL);
 		if (err)
@@ -702,7 +764,7 @@ static int adapter_up(struct adapter *adapter)
 	 * adapter setup.  Once we've done this, many of our adapter
 	 * parameters can no longer be changed ...
 	 */
-	if ((adapter->flags & FULL_INIT_DONE) == 0) {
+	if ((adapter->flags & CXGB4VF_FULL_INIT_DONE) == 0) {
 		err = setup_sge_queues(adapter);
 		if (err)
 			return err;
@@ -712,16 +774,18 @@ static int adapter_up(struct adapter *adapter)
 			return err;
 		}
 
-		if (adapter->flags & USING_MSIX)
+		if (adapter->flags & CXGB4VF_USING_MSIX)
 			name_msix_vecs(adapter);
-		adapter->flags |= FULL_INIT_DONE;
+
+		adapter->flags |= CXGB4VF_FULL_INIT_DONE;
 	}
 
 	/*
 	 * Acquire our interrupt resources.  We only support MSI-X and MSI.
 	 */
-	BUG_ON((adapter->flags & (USING_MSIX|USING_MSI)) == 0);
-	if (adapter->flags & USING_MSIX)
+	BUG_ON((adapter->flags &
+	       (CXGB4VF_USING_MSIX | CXGB4VF_USING_MSI)) == 0);
+	if (adapter->flags & CXGB4VF_USING_MSIX)
 		err = request_msix_queue_irqs(adapter);
 	else
 		err = request_irq(adapter->pdev->irq,
@@ -739,8 +803,6 @@ static int adapter_up(struct adapter *adapter)
 	enable_rx(adapter);
 	t4vf_sge_start(adapter);
 
-	/* Initialize hash mac addr list*/
-	INIT_LIST_HEAD(&adapter->mac_hlist);
 	return 0;
 }
 
@@ -754,7 +816,7 @@ static void adapter_down(struct adapter *adapter)
 	/*
 	 * Free interrupt resources.
 	 */
-	if (adapter->flags & USING_MSIX)
+	if (adapter->flags & CXGB4VF_USING_MSIX)
 		free_msix_queue_irqs(adapter);
 	else
 		free_irq(adapter->pdev->irq, adapter);
@@ -775,6 +837,13 @@ static int cxgb4vf_open(struct net_device *dev)
 	struct adapter *adapter = pi->adapter;
 
 	/*
+	 * If we don't have a connection to the firmware there's nothing we
+	 * can do.
+	 */
+	if (!(adapter->flags & CXGB4VF_FW_OK))
+		return -ENXIO;
+
+	/*
 	 * If this is the first interface that we're opening on the "adapter",
 	 * bring the "adapter" up now.
 	 */
@@ -784,6 +853,13 @@ static int cxgb4vf_open(struct net_device *dev)
 			return err;
 	}
 
+	/* It's possible that the basic port information could have
+	 * changed since we first read it.
+	 */
+	err = t4vf_update_port_info(pi);
+	if (err < 0)
+		goto err_unwind;
+
 	/*
 	 * Note that this interface is up and start everything up ...
 	 */
@@ -791,6 +867,8 @@ static int cxgb4vf_open(struct net_device *dev)
 	if (err)
 		goto err_unwind;
 
+	pi->vlan_id = t4vf_get_vf_vlan_acl(adapter);
+
 	netif_tx_start_all_queues(dev);
 	set_bit(pi->port_id, &adapter->open_device_map);
 	return 0;
@@ -812,8 +890,7 @@ static int cxgb4vf_stop(struct net_device *dev)
 
 	netif_tx_stop_all_queues(dev);
 	netif_carrier_off(dev);
-	t4vf_enable_vi(adapter, pi->viid, false, false);
-	pi->link_cfg.link_ok = 0;
+	t4vf_enable_pi(adapter, pi, false, false);
 
 	clear_bit(pi->port_id, &adapter->open_device_map);
 	if (adapter->open_device_map == 0)
@@ -855,21 +932,6 @@ static struct net_device_stats *cxgb4vf_get_stats(struct net_device *dev)
 	return ns;
 }
 
-static inline int cxgb4vf_set_addr_hash(struct port_info *pi)
-{
-	struct adapter *adapter = pi->adapter;
-	u64 vec = 0;
-	bool ucast = false;
-	struct hash_mac_addr *entry;
-
-	/* Calculate the hash vector for the updated list and program it */
-	list_for_each_entry(entry, &adapter->mac_hlist, list) {
-		ucast |= is_unicast_ether_addr(entry->addr);
-		vec |= (1ULL << hash_mac_addr(entry->addr));
-	}
-	return t4vf_set_addr_hash(adapter, pi->viid, ucast, vec, false);
-}
-
 static int cxgb4vf_mac_sync(struct net_device *netdev, const u8 *mac_addr)
 {
 	struct port_info *pi = netdev_priv(netdev);
@@ -1107,7 +1169,7 @@ static int cxgb4vf_change_mtu(struct net_device *dev, int new_mtu)
 	ret = t4vf_set_rxmode(pi->adapter, pi->viid, new_mtu,
 			      -1, -1, -1, -1, true);
 	if (!ret)
-		dev->mtu = new_mtu;
+		WRITE_ONCE(dev->mtu, new_mtu);
 	return ret;
 }
 
@@ -1151,13 +1213,12 @@ static int cxgb4vf_set_mac_addr(struct net_device *dev, void *_addr)
 	if (!is_valid_ether_addr(addr->sa_data))
 		return -EADDRNOTAVAIL;
 
-	ret = t4vf_change_mac(pi->adapter, pi->viid, pi->xact_addr_filt,
-			      addr->sa_data, true);
+	ret = cxgb4vf_change_mac(pi, pi->viid, &pi->xact_addr_filt,
+				 addr->sa_data, true);
 	if (ret < 0)
 		return ret;
 
-	memcpy(dev->dev_addr, addr->sa_data, dev->addr_len);
-	pi->xact_addr_filt = ret;
+	eth_hw_addr_set(dev, addr->sa_data);
 	return 0;
 }
 
@@ -1171,7 +1232,7 @@ static void cxgb4vf_poll_controller(struct net_device *dev)
 	struct port_info *pi = netdev_priv(dev);
 	struct adapter *adapter = pi->adapter;
 
-	if (adapter->flags & USING_MSIX) {
+	if (adapter->flags & CXGB4VF_USING_MSIX) {
 		struct sge_eth_rxq *rxq;
 		int nqsets;
 
@@ -1213,7 +1274,11 @@ static int from_fw_port_mod_type(enum fw_port_type port_type,
 	} else if (port_type == FW_PORT_TYPE_SFP ||
 		   port_type == FW_PORT_TYPE_QSFP_10G ||
 		   port_type == FW_PORT_TYPE_QSA ||
-		   port_type == FW_PORT_TYPE_QSFP) {
+		   port_type == FW_PORT_TYPE_QSFP ||
+		   port_type == FW_PORT_TYPE_CR4_QSFP ||
+		   port_type == FW_PORT_TYPE_CR_QSFP ||
+		   port_type == FW_PORT_TYPE_CR2_QSFP ||
+		   port_type == FW_PORT_TYPE_SFP28) {
 		if (mod_type == FW_PORT_MOD_TYPE_LR ||
 		    mod_type == FW_PORT_MOD_TYPE_SR ||
 		    mod_type == FW_PORT_MOD_TYPE_ER ||
@@ -1224,6 +1289,10 @@ static int from_fw_port_mod_type(enum fw_port_type port_type,
 			return PORT_DA;
 		else
 			return PORT_OTHER;
+	} else if (port_type == FW_PORT_TYPE_KR4_100G ||
+		   port_type == FW_PORT_TYPE_KR_SFP28 ||
+		   port_type == FW_PORT_TYPE_KR_XLAUI) {
+		return PORT_NONE;
 	}
 
 	return PORT_OTHER;
@@ -1242,12 +1311,13 @@ static void fw_caps_to_lmm(enum fw_port_type port_type,
 			   unsigned int fw_caps,
 			   unsigned long *link_mode_mask)
 {
-	#define SET_LMM(__lmm_name) __set_bit(ETHTOOL_LINK_MODE_ ## __lmm_name\
-			 ## _BIT, link_mode_mask)
+	#define SET_LMM(__lmm_name) \
+		__set_bit(ETHTOOL_LINK_MODE_ ## __lmm_name ## _BIT, \
+			  link_mode_mask)
 
 	#define FW_CAPS_TO_LMM(__fw_name, __lmm_name) \
 		do { \
-			if (fw_caps & FW_PORT_CAP_ ## __fw_name) \
+			if (fw_caps & FW_PORT_CAP32_ ## __fw_name) \
 				SET_LMM(__lmm_name); \
 		} while (0)
 
@@ -1270,22 +1340,22 @@ static void fw_caps_to_lmm(enum fw_port_type port_type,
 
 	case FW_PORT_TYPE_KR:
 		SET_LMM(Backplane);
-		SET_LMM(10000baseKR_Full);
+		FW_CAPS_TO_LMM(SPEED_10G, 10000baseKR_Full);
 		break;
 
 	case FW_PORT_TYPE_BP_AP:
 		SET_LMM(Backplane);
-		SET_LMM(10000baseR_FEC);
-		SET_LMM(10000baseKR_Full);
-		SET_LMM(1000baseKX_Full);
+		FW_CAPS_TO_LMM(SPEED_1G, 1000baseKX_Full);
+		FW_CAPS_TO_LMM(SPEED_10G, 10000baseR_FEC);
+		FW_CAPS_TO_LMM(SPEED_10G, 10000baseKR_Full);
 		break;
 
 	case FW_PORT_TYPE_BP4_AP:
 		SET_LMM(Backplane);
-		SET_LMM(10000baseR_FEC);
-		SET_LMM(10000baseKR_Full);
-		SET_LMM(1000baseKX_Full);
-		SET_LMM(10000baseKX4_Full);
+		FW_CAPS_TO_LMM(SPEED_1G, 1000baseKX_Full);
+		FW_CAPS_TO_LMM(SPEED_10G, 10000baseR_FEC);
+		FW_CAPS_TO_LMM(SPEED_10G, 10000baseKR_Full);
+		FW_CAPS_TO_LMM(SPEED_10G, 10000baseKX4_Full);
 		break;
 
 	case FW_PORT_TYPE_FIBER_XFI:
@@ -1301,25 +1371,60 @@ static void fw_caps_to_lmm(enum fw_port_type port_type,
 	case FW_PORT_TYPE_BP40_BA:
 	case FW_PORT_TYPE_QSFP:
 		SET_LMM(FIBRE);
-		SET_LMM(40000baseSR4_Full);
+		FW_CAPS_TO_LMM(SPEED_1G, 1000baseT_Full);
+		FW_CAPS_TO_LMM(SPEED_10G, 10000baseT_Full);
+		FW_CAPS_TO_LMM(SPEED_40G, 40000baseSR4_Full);
 		break;
 
 	case FW_PORT_TYPE_CR_QSFP:
 	case FW_PORT_TYPE_SFP28:
 		SET_LMM(FIBRE);
-		SET_LMM(25000baseCR_Full);
+		FW_CAPS_TO_LMM(SPEED_1G, 1000baseT_Full);
+		FW_CAPS_TO_LMM(SPEED_10G, 10000baseT_Full);
+		FW_CAPS_TO_LMM(SPEED_25G, 25000baseCR_Full);
+		break;
+
+	case FW_PORT_TYPE_KR_SFP28:
+		SET_LMM(Backplane);
+		FW_CAPS_TO_LMM(SPEED_1G, 1000baseT_Full);
+		FW_CAPS_TO_LMM(SPEED_10G, 10000baseKR_Full);
+		FW_CAPS_TO_LMM(SPEED_25G, 25000baseKR_Full);
+		break;
+
+	case FW_PORT_TYPE_KR_XLAUI:
+		SET_LMM(Backplane);
+		FW_CAPS_TO_LMM(SPEED_1G, 1000baseKX_Full);
+		FW_CAPS_TO_LMM(SPEED_10G, 10000baseKR_Full);
+		FW_CAPS_TO_LMM(SPEED_40G, 40000baseKR4_Full);
+		break;
+
+	case FW_PORT_TYPE_CR2_QSFP:
+		SET_LMM(FIBRE);
+		FW_CAPS_TO_LMM(SPEED_50G, 50000baseSR2_Full);
 		break;
 
 	case FW_PORT_TYPE_KR4_100G:
 	case FW_PORT_TYPE_CR4_QSFP:
 		SET_LMM(FIBRE);
-		SET_LMM(100000baseCR4_Full);
+		FW_CAPS_TO_LMM(SPEED_1G,  1000baseT_Full);
+		FW_CAPS_TO_LMM(SPEED_10G, 10000baseKR_Full);
+		FW_CAPS_TO_LMM(SPEED_40G, 40000baseSR4_Full);
+		FW_CAPS_TO_LMM(SPEED_25G, 25000baseCR_Full);
+		FW_CAPS_TO_LMM(SPEED_50G, 50000baseCR2_Full);
+		FW_CAPS_TO_LMM(SPEED_100G, 100000baseCR4_Full);
 		break;
 
 	default:
 		break;
 	}
 
+	if (fw_caps & FW_PORT_CAP32_FEC_V(FW_PORT_CAP32_FEC_M)) {
+		FW_CAPS_TO_LMM(FEC_RS, FEC_RS);
+		FW_CAPS_TO_LMM(FEC_BASER_RS, FEC_BASER);
+	} else {
+		SET_LMM(FEC_NONE);
+	}
+
 	FW_CAPS_TO_LMM(ANEG, Autoneg);
 	FW_CAPS_TO_LMM(802_3_PAUSE, Pause);
 	FW_CAPS_TO_LMM(802_3_ASM_DIR, Asym_Pause);
@@ -1329,12 +1434,18 @@ static void fw_caps_to_lmm(enum fw_port_type port_type,
 }
 
 static int cxgb4vf_get_link_ksettings(struct net_device *dev,
-				      struct ethtool_link_ksettings
-							*link_ksettings)
+				  struct ethtool_link_ksettings *link_ksettings)
 {
-	const struct port_info *pi = netdev_priv(dev);
+	struct port_info *pi = netdev_priv(dev);
 	struct ethtool_link_settings *base = &link_ksettings->base;
 
+	/* For the nonce, the Firmware doesn't send up Port State changes
+	 * when the Virtual Interface attached to the Port is down.  So
+	 * if it's down, let's grab any changes.
+	 */
+	if (!netif_running(dev))
+		(void)t4vf_update_port_info(pi);
+
 	ethtool_link_ksettings_zero_link_mode(link_ksettings, supported);
 	ethtool_link_ksettings_zero_link_mode(link_ksettings, advertising);
 	ethtool_link_ksettings_zero_link_mode(link_ksettings, lp_advertising);
@@ -1351,11 +1462,11 @@ static int cxgb4vf_get_link_ksettings(struct net_device *dev,
 		base->mdio_support = 0;
 	}
 
-	fw_caps_to_lmm(pi->port_type, pi->link_cfg.supported,
+	fw_caps_to_lmm(pi->port_type, pi->link_cfg.pcaps,
 		       link_ksettings->link_modes.supported);
-	fw_caps_to_lmm(pi->port_type, pi->link_cfg.advertising,
+	fw_caps_to_lmm(pi->port_type, pi->link_cfg.acaps,
 		       link_ksettings->link_modes.advertising);
-	fw_caps_to_lmm(pi->port_type, pi->link_cfg.lp_advertising,
+	fw_caps_to_lmm(pi->port_type, pi->link_cfg.lpacaps,
 		       link_ksettings->link_modes.lp_advertising);
 
 	if (netif_carrier_ok(dev)) {
@@ -1367,7 +1478,7 @@ static int cxgb4vf_get_link_ksettings(struct net_device *dev,
 	}
 
 	base->autoneg = pi->link_cfg.autoneg;
-	if (pi->link_cfg.supported & FW_PORT_CAP_ANEG)
+	if (pi->link_cfg.pcaps & FW_PORT_CAP32_ANEG)
 		ethtool_link_ksettings_add_link_mode(link_ksettings,
 						     supported, Autoneg);
 	if (pi->link_cfg.autoneg)
@@ -1377,6 +1488,63 @@ static int cxgb4vf_get_link_ksettings(struct net_device *dev,
 	return 0;
 }
 
+/* Translate the Firmware FEC value into the ethtool value. */
+static inline unsigned int fwcap_to_eth_fec(unsigned int fw_fec)
+{
+	unsigned int eth_fec = 0;
+
+	if (fw_fec & FW_PORT_CAP32_FEC_RS)
+		eth_fec |= ETHTOOL_FEC_RS;
+	if (fw_fec & FW_PORT_CAP32_FEC_BASER_RS)
+		eth_fec |= ETHTOOL_FEC_BASER;
+
+	/* if nothing is set, then FEC is off */
+	if (!eth_fec)
+		eth_fec = ETHTOOL_FEC_OFF;
+
+	return eth_fec;
+}
+
+/* Translate Common Code FEC value into ethtool value. */
+static inline unsigned int cc_to_eth_fec(unsigned int cc_fec)
+{
+	unsigned int eth_fec = 0;
+
+	if (cc_fec & FEC_AUTO)
+		eth_fec |= ETHTOOL_FEC_AUTO;
+	if (cc_fec & FEC_RS)
+		eth_fec |= ETHTOOL_FEC_RS;
+	if (cc_fec & FEC_BASER_RS)
+		eth_fec |= ETHTOOL_FEC_BASER;
+
+	/* if nothing is set, then FEC is off */
+	if (!eth_fec)
+		eth_fec = ETHTOOL_FEC_OFF;
+
+	return eth_fec;
+}
+
+static int cxgb4vf_get_fecparam(struct net_device *dev,
+				struct ethtool_fecparam *fec)
+{
+	const struct port_info *pi = netdev_priv(dev);
+	const struct link_config *lc = &pi->link_cfg;
+
+	/* Translate the Firmware FEC Support into the ethtool value.  We
+	 * always support IEEE 802.3 "automatic" selection of Link FEC type if
+	 * any FEC is supported.
+	 */
+	fec->fec = fwcap_to_eth_fec(lc->pcaps);
+	if (fec->fec != ETHTOOL_FEC_OFF)
+		fec->fec |= ETHTOOL_FEC_AUTO;
+
+	/* Translate the current internal FEC parameters into the
+	 * ethtool values.
+	 */
+	fec->active_fec = cc_to_eth_fec(lc->fec);
+	return 0;
+}
+
 /*
  * Return our driver information.
  */
@@ -1385,9 +1553,8 @@ static void cxgb4vf_get_drvinfo(struct net_device *dev,
 {
 	struct adapter *adapter = netdev2adap(dev);
 
-	strlcpy(drvinfo->driver, KBUILD_MODNAME, sizeof(drvinfo->driver));
-	strlcpy(drvinfo->version, DRV_VERSION, sizeof(drvinfo->version));
-	strlcpy(drvinfo->bus_info, pci_name(to_pci_dev(dev->dev.parent)),
+	strscpy(drvinfo->driver, KBUILD_MODNAME, sizeof(drvinfo->driver));
+	strscpy(drvinfo->bus_info, pci_name(to_pci_dev(dev->dev.parent)),
 		sizeof(drvinfo->bus_info));
 	snprintf(drvinfo->fw_version, sizeof(drvinfo->fw_version),
 		 "%u.%u.%u.%u, TP %u.%u.%u.%u",
@@ -1424,7 +1591,9 @@ static void cxgb4vf_set_msglevel(struct net_device *dev, u32 msglevel)
  * first Queue Set.
  */
 static void cxgb4vf_get_ringparam(struct net_device *dev,
-				  struct ethtool_ringparam *rp)
+				  struct ethtool_ringparam *rp,
+				  struct kernel_ethtool_ringparam *kernel_rp,
+				  struct netlink_ext_ack *extack)
 {
 	const struct port_info *pi = netdev_priv(dev);
 	const struct sge *s = &pi->adapter->sge;
@@ -1447,7 +1616,9 @@ static void cxgb4vf_get_ringparam(struct net_device *dev,
  * device -- after vetting them of course!
  */
 static int cxgb4vf_set_ringparam(struct net_device *dev,
-				 struct ethtool_ringparam *rp)
+				 struct ethtool_ringparam *rp,
+				 struct kernel_ethtool_ringparam *kernel_rp,
+				 struct netlink_ext_ack *extack)
 {
 	const struct port_info *pi = netdev_priv(dev);
 	struct adapter *adapter = pi->adapter;
@@ -1463,7 +1634,7 @@ static int cxgb4vf_set_ringparam(struct net_device *dev,
 	    rp->tx_pending < MIN_TXQ_ENTRIES)
 		return -EINVAL;
 
-	if (adapter->flags & FULL_INIT_DONE)
+	if (adapter->flags & CXGB4VF_FULL_INIT_DONE)
 		return -EBUSY;
 
 	for (qs = pi->first_qset; qs < pi->first_qset + pi->nqsets; qs++) {
@@ -1480,7 +1651,9 @@ static int cxgb4vf_set_ringparam(struct net_device *dev,
  * interrupt holdoff timer to be read on all of the device's Queue Sets.
  */
 static int cxgb4vf_get_coalesce(struct net_device *dev,
-				struct ethtool_coalesce *coalesce)
+				struct ethtool_coalesce *coalesce,
+				struct kernel_ethtool_coalesce *kernel_coal,
+				struct netlink_ext_ack *extack)
 {
 	const struct port_info *pi = netdev_priv(dev);
 	const struct adapter *adapter = pi->adapter;
@@ -1500,7 +1673,9 @@ static int cxgb4vf_get_coalesce(struct net_device *dev,
  * the interrupt holdoff timer on any of the device's Queue Sets.
  */
 static int cxgb4vf_set_coalesce(struct net_device *dev,
-				struct ethtool_coalesce *coalesce)
+				struct ethtool_coalesce *coalesce,
+				struct kernel_ethtool_coalesce *kernel_coal,
+				struct netlink_ext_ack *extack)
 {
 	const struct port_info *pi = netdev_priv(dev);
 	struct adapter *adapter = pi->adapter;
@@ -1520,8 +1695,8 @@ static void cxgb4vf_get_pauseparam(struct net_device *dev,
 	struct port_info *pi = netdev_priv(dev);
 
 	pauseparam->autoneg = (pi->link_cfg.requested_fc & PAUSE_AUTONEG) != 0;
-	pauseparam->rx_pause = (pi->link_cfg.fc & PAUSE_RX) != 0;
-	pauseparam->tx_pause = (pi->link_cfg.fc & PAUSE_TX) != 0;
+	pauseparam->rx_pause = (pi->link_cfg.advertised_fc & PAUSE_RX) != 0;
+	pauseparam->tx_pause = (pi->link_cfg.advertised_fc & PAUSE_TX) != 0;
 }
 
 /*
@@ -1747,9 +1922,14 @@ static void cxgb4vf_get_wol(struct net_device *dev,
  * TCP Segmentation Offload flags which we support.
  */
 #define TSO_FLAGS (NETIF_F_TSO | NETIF_F_TSO6 | NETIF_F_TSO_ECN)
+#define VLAN_FEAT (NETIF_F_SG | NETIF_F_IP_CSUM | TSO_FLAGS | \
+		   NETIF_F_GRO | NETIF_F_IPV6_CSUM | NETIF_F_HIGHDMA)
 
 static const struct ethtool_ops cxgb4vf_ethtool_ops = {
+	.supported_coalesce_params = ETHTOOL_COALESCE_RX_USECS |
+				     ETHTOOL_COALESCE_RX_MAX_FRAMES,
 	.get_link_ksettings	= cxgb4vf_get_link_ksettings,
+	.get_fecparam		= cxgb4vf_get_fecparam,
 	.get_drvinfo		= cxgb4vf_get_drvinfo,
 	.get_msglevel		= cxgb4vf_get_msglevel,
 	.set_msglevel		= cxgb4vf_set_msglevel,
@@ -1845,33 +2025,14 @@ static void mboxlog_stop(struct seq_file *seq, void *v)
 {
 }
 
-static const struct seq_operations mboxlog_seq_ops = {
+static const struct seq_operations mboxlog_sops = {
 	.start = mboxlog_start,
 	.next  = mboxlog_next,
 	.stop  = mboxlog_stop,
 	.show  = mboxlog_show
 };
 
-static int mboxlog_open(struct inode *inode, struct file *file)
-{
-	int res = seq_open(file, &mboxlog_seq_ops);
-
-	if (!res) {
-		struct seq_file *seq = file->private_data;
-
-		seq->private = inode->i_private;
-	}
-	return res;
-}
-
-static const struct file_operations mboxlog_fops = {
-	.owner   = THIS_MODULE,
-	.open    = mboxlog_open,
-	.read    = seq_read,
-	.llseek  = seq_lseek,
-	.release = seq_release,
-};
-
+DEFINE_SEQ_ATTRIBUTE(mboxlog);
 /*
  * Show SGE Queue Set information.  We display QPL Queues Sets per line.
  */
@@ -1977,7 +2138,7 @@ static int sge_qinfo_show(struct seq_file *seq, void *v)
 static int sge_queue_entries(const struct adapter *adapter)
 {
 	return DIV_ROUND_UP(adapter->sge.ethqsets, QPL) + 1 +
-		((adapter->flags & USING_MSI) != 0);
+		((adapter->flags & CXGB4VF_USING_MSI) != 0);
 }
 
 static void *sge_queue_start(struct seq_file *seq, loff_t *pos)
@@ -1999,31 +2160,14 @@ static void *sge_queue_next(struct seq_file *seq, void *v, loff_t *pos)
 	return *pos < entries ? (void *)((uintptr_t)*pos + 1) : NULL;
 }
 
-static const struct seq_operations sge_qinfo_seq_ops = {
+static const struct seq_operations sge_qinfo_sops = {
 	.start = sge_queue_start,
 	.next  = sge_queue_next,
 	.stop  = sge_queue_stop,
 	.show  = sge_qinfo_show
 };
 
-static int sge_qinfo_open(struct inode *inode, struct file *file)
-{
-	int res = seq_open(file, &sge_qinfo_seq_ops);
-
-	if (!res) {
-		struct seq_file *seq = file->private_data;
-		seq->private = inode->i_private;
-	}
-	return res;
-}
-
-static const struct file_operations sge_qinfo_debugfs_fops = {
-	.owner   = THIS_MODULE,
-	.open    = sge_qinfo_open,
-	.read    = seq_read,
-	.llseek  = seq_lseek,
-	.release = seq_release,
-};
+DEFINE_SEQ_ATTRIBUTE(sge_qinfo);
 
 /*
  * Show SGE Queue Set statistics.  We display QPL Queues Sets per line.
@@ -2123,7 +2267,7 @@ static int sge_qstats_show(struct seq_file *seq, void *v)
 static int sge_qstats_entries(const struct adapter *adapter)
 {
 	return DIV_ROUND_UP(adapter->sge.ethqsets, QPL) + 1 +
-		((adapter->flags & USING_MSI) != 0);
+		((adapter->flags & CXGB4VF_USING_MSI) != 0);
 }
 
 static void *sge_qstats_start(struct seq_file *seq, loff_t *pos)
@@ -2145,31 +2289,14 @@ static void *sge_qstats_next(struct seq_file *seq, void *v, loff_t *pos)
 	return *pos < entries ? (void *)((uintptr_t)*pos + 1) : NULL;
 }
 
-static const struct seq_operations sge_qstats_seq_ops = {
+static const struct seq_operations sge_qstats_sops = {
 	.start = sge_qstats_start,
 	.next  = sge_qstats_next,
 	.stop  = sge_qstats_stop,
 	.show  = sge_qstats_show
 };
 
-static int sge_qstats_open(struct inode *inode, struct file *file)
-{
-	int res = seq_open(file, &sge_qstats_seq_ops);
-
-	if (res == 0) {
-		struct seq_file *seq = file->private_data;
-		seq->private = inode->i_private;
-	}
-	return res;
-}
-
-static const struct file_operations sge_qstats_proc_fops = {
-	.owner   = THIS_MODULE,
-	.open    = sge_qstats_open,
-	.read    = seq_read,
-	.llseek  = seq_lseek,
-	.release = seq_release,
-};
+DEFINE_SEQ_ATTRIBUTE(sge_qstats);
 
 /*
  * Show PCI-E SR-IOV Virtual Function Resource Limits.
@@ -2198,19 +2325,7 @@ static int resources_show(struct seq_file *seq, void *v)
 
 	return 0;
 }
-
-static int resources_open(struct inode *inode, struct file *file)
-{
-	return single_open(file, resources_show, inode->i_private);
-}
-
-static const struct file_operations resources_proc_fops = {
-	.owner   = THIS_MODULE,
-	.open    = resources_open,
-	.read    = seq_read,
-	.llseek  = seq_lseek,
-	.release = single_release,
-};
+DEFINE_SHOW_ATTRIBUTE(resources);
 
 /*
  * Show Virtual Interfaces.
@@ -2255,31 +2370,14 @@ static void interfaces_stop(struct seq_file *seq, void *v)
 {
 }
 
-static const struct seq_operations interfaces_seq_ops = {
+static const struct seq_operations interfaces_sops = {
 	.start = interfaces_start,
 	.next  = interfaces_next,
 	.stop  = interfaces_stop,
 	.show  = interfaces_show
 };
 
-static int interfaces_open(struct inode *inode, struct file *file)
-{
-	int res = seq_open(file, &interfaces_seq_ops);
-
-	if (res == 0) {
-		struct seq_file *seq = file->private_data;
-		seq->private = inode->i_private;
-	}
-	return res;
-}
-
-static const struct file_operations interfaces_proc_fops = {
-	.owner   = THIS_MODULE,
-	.open    = interfaces_open,
-	.read    = seq_read,
-	.llseek  = seq_lseek,
-	.release = seq_release,
-};
+DEFINE_SEQ_ATTRIBUTE(interfaces);
 
 /*
  * /sys/kernel/debugfs/cxgb4vf/ files list.
@@ -2291,11 +2389,11 @@ struct cxgb4vf_debugfs_entry {
 };
 
 static struct cxgb4vf_debugfs_entry debugfs_files[] = {
-	{ "mboxlog",    S_IRUGO, &mboxlog_fops },
-	{ "sge_qinfo",  S_IRUGO, &sge_qinfo_debugfs_fops },
-	{ "sge_qstats", S_IRUGO, &sge_qstats_proc_fops },
-	{ "resources",  S_IRUGO, &resources_proc_fops },
-	{ "interfaces", S_IRUGO, &interfaces_proc_fops },
+	{ "mboxlog",    0444, &mboxlog_fops },
+	{ "sge_qinfo",  0444, &sge_qinfo_fops },
+	{ "sge_qstats", 0444, &sge_qstats_fops },
+	{ "resources",  0444, &resources_fops },
+	{ "interfaces", 0444, &interfaces_fops },
 };
 
 /*
@@ -2317,11 +2415,10 @@ static int setup_debugfs(struct adapter *adapter)
 	 * Debugfs support is best effort.
 	 */
 	for (i = 0; i < ARRAY_SIZE(debugfs_files); i++)
-		(void)debugfs_create_file(debugfs_files[i].name,
-				  debugfs_files[i].mode,
-				  adapter->debugfs_root,
-				  (void *)adapter,
-				  debugfs_files[i].fops);
+		debugfs_create_file(debugfs_files[i].name,
+				    debugfs_files[i].mode,
+				    adapter->debugfs_root, adapter,
+				    debugfs_files[i].fops);
 
 	return 0;
 }
@@ -2544,6 +2641,7 @@ static int adap_init0(struct adapter *adapter)
 	 */
 	size_nports_qsets(adapter);
 
+	adapter->flags |= CXGB4VF_FW_OK;
 	return 0;
 }
 
@@ -2578,7 +2676,8 @@ static void cfg_queues(struct adapter *adapter)
 	 * support.  In particular, this means that we need to know what kind
 	 * of interrupts we'll be using ...
 	 */
-	BUG_ON((adapter->flags & (USING_MSIX|USING_MSI)) == 0);
+	BUG_ON((adapter->flags &
+	       (CXGB4VF_USING_MSIX | CXGB4VF_USING_MSI)) == 0);
 
 	/*
 	 * Count the number of 10GbE Virtual Interfaces that we have.
@@ -2746,7 +2845,7 @@ static const struct net_device_ops cxgb4vf_netdev_ops	= {
 	.ndo_set_rx_mode	= cxgb4vf_set_rxmode,
 	.ndo_set_mac_address	= cxgb4vf_set_mac_addr,
 	.ndo_validate_addr	= eth_validate_addr,
-	.ndo_do_ioctl		= cxgb4vf_do_ioctl,
+	.ndo_eth_ioctl		= cxgb4vf_do_ioctl,
 	.ndo_change_mtu		= cxgb4vf_change_mtu,
 	.ndo_fix_features	= cxgb4vf_fix_features,
 	.ndo_set_features	= cxgb4vf_set_features,
@@ -2755,6 +2854,39 @@ static const struct net_device_ops cxgb4vf_netdev_ops	= {
 #endif
 };
 
+/**
+ *	cxgb4vf_get_port_mask - Get port mask for the VF based on mac
+ *				address stored on the adapter
+ *	@adapter: The adapter
+ *
+ *	Find the port mask for the VF based on the index of mac
+ *	address stored in the adapter. If no mac address is stored on
+ *	the adapter for the VF, use the port mask received from the
+ *	firmware.
+ */
+static unsigned int cxgb4vf_get_port_mask(struct adapter *adapter)
+{
+	unsigned int naddr = 1, pidx = 0;
+	unsigned int pmask, rmask = 0;
+	u8 mac[ETH_ALEN];
+	int err;
+
+	pmask = adapter->params.vfres.pmask;
+	while (pmask) {
+		if (pmask & 1) {
+			err = t4vf_get_vf_mac_acl(adapter, pidx, &naddr, mac);
+			if (!err && !is_zero_ether_addr(mac))
+				rmask |= (1 << pidx);
+		}
+		pmask >>= 1;
+		pidx++;
+	}
+	if (!rmask)
+		rmask = adapter->params.vfres.pmask;
+
+	return rmask;
+}
+
 /*
  * "Probe" a device: initialize a device and construct all kernel and driver
  * state needed to manage the device.  This routine is called "init_one" in
@@ -2763,28 +2895,18 @@ static const struct net_device_ops cxgb4vf_netdev_ops	= {
 static int cxgb4vf_pci_probe(struct pci_dev *pdev,
 			     const struct pci_device_id *ent)
 {
-	int pci_using_dac;
-	int err, pidx;
-	unsigned int pmask;
 	struct adapter *adapter;
-	struct port_info *pi;
 	struct net_device *netdev;
-	unsigned int pf;
-
-	/*
-	 * Print our driver banner the first time we're called to initialize a
-	 * device.
-	 */
-	pr_info_once("%s - version %s\n", DRV_DESC, DRV_VERSION);
+	struct port_info *pi;
+	unsigned int pmask;
+	int err, pidx;
 
 	/*
 	 * Initialize generic PCI device state.
 	 */
 	err = pci_enable_device(pdev);
-	if (err) {
-		dev_err(&pdev->dev, "cannot enable PCI device\n");
-		return err;
-	}
+	if (err)
+		return dev_err_probe(&pdev->dev, err, "cannot enable PCI device\n");
 
 	/*
 	 * Reserve PCI resources for the device.  If we can't get them some
@@ -2797,25 +2919,12 @@ static int cxgb4vf_pci_probe(struct pci_dev *pdev,
 	}
 
 	/*
-	 * Set up our DMA mask: try for 64-bit address masking first and
-	 * fall back to 32-bit if we can't get 64 bits ...
+	 * Set up our DMA mask
 	 */
-	err = pci_set_dma_mask(pdev, DMA_BIT_MASK(64));
-	if (err == 0) {
-		err = pci_set_consistent_dma_mask(pdev, DMA_BIT_MASK(64));
-		if (err) {
-			dev_err(&pdev->dev, "unable to obtain 64-bit DMA for"
-				" coherent allocations\n");
-			goto err_release_regions;
-		}
-		pci_using_dac = 1;
-	} else {
-		err = pci_set_dma_mask(pdev, DMA_BIT_MASK(32));
-		if (err != 0) {
-			dev_err(&pdev->dev, "no usable DMA configuration\n");
-			goto err_release_regions;
-		}
-		pci_using_dac = 0;
+	err = dma_set_mask_and_coherent(&pdev->dev, DMA_BIT_MASK(64));
+	if (err) {
+		dev_err(&pdev->dev, "no usable DMA configuration\n");
+		goto err_release_regions;
 	}
 
 	/*
@@ -2888,15 +2997,37 @@ static int cxgb4vf_pci_probe(struct pci_dev *pdev,
 	 */
 	adapter->name = pci_name(pdev);
 	adapter->msg_enable = DFLT_MSG_ENABLE;
+
+	/* If possible, we use PCIe Relaxed Ordering Attribute to deliver
+	 * Ingress Packet Data to Free List Buffers in order to allow for
+	 * chipset performance optimizations between the Root Complex and
+	 * Memory Controllers.  (Messages to the associated Ingress Queue
+	 * notifying new Packet Placement in the Free Lists Buffers will be
+	 * send without the Relaxed Ordering Attribute thus guaranteeing that
+	 * all preceding PCIe Transaction Layer Packets will be processed
+	 * first.)  But some Root Complexes have various issues with Upstream
+	 * Transaction Layer Packets with the Relaxed Ordering Attribute set.
+	 * The PCIe devices which under the Root Complexes will be cleared the
+	 * Relaxed Ordering bit in the configuration space, So we check our
+	 * PCIe configuration space to see if it's flagged with advice against
+	 * using Relaxed Ordering.
+	 */
+	if (!pcie_relaxed_ordering_enabled(pdev))
+		adapter->flags |= CXGB4VF_ROOT_NO_RELAXED_ORDERING;
+
 	err = adap_init0(adapter);
 	if (err)
-		goto err_unmap_bar;
+		dev_err(&pdev->dev,
+			"Adapter initialization failed, error %d. Continuing in debug mode\n",
+			err);
+
+	/* Initialize hash mac addr list */
+	INIT_LIST_HEAD(&adapter->mac_hlist);
 
 	/*
 	 * Allocate our "adapter ports" and stitch everything together.
 	 */
-	pmask = adapter->params.vfres.pmask;
-	pf = t4vf_get_pf_from_vf(adapter);
+	pmask = cxgb4vf_get_port_mask(adapter);
 	for_each_port(adapter, pidx) {
 		int port_id, viid;
 		u8 mac[ETH_ALEN];
@@ -2912,13 +3043,6 @@ static int cxgb4vf_pci_probe(struct pci_dev *pdev,
 			break;
 		port_id = ffs(pmask) - 1;
 		pmask &= ~(1 << port_id);
-		viid = t4vf_alloc_vi(adapter, port_id);
-		if (viid < 0) {
-			dev_err(&pdev->dev, "cannot allocate VI for port %d:"
-				" err=%d\n", port_id, viid);
-			err = viid;
-			goto err_free_dev;
-		}
 
 		/*
 		 * Allocate our network device and stitch things together.
@@ -2926,7 +3050,6 @@ static int cxgb4vf_pci_probe(struct pci_dev *pdev,
 		netdev = alloc_etherdev_mq(sizeof(struct port_info),
 					   MAX_PORT_QSETS);
 		if (netdev == NULL) {
-			t4vf_free_vi(adapter, viid);
 			err = -ENOMEM;
 			goto err_free_dev;
 		}
@@ -2936,26 +3059,19 @@ static int cxgb4vf_pci_probe(struct pci_dev *pdev,
 		pi->adapter = adapter;
 		pi->pidx = pidx;
 		pi->port_id = port_id;
-		pi->viid = viid;
 
 		/*
 		 * Initialize the starting state of our "port" and register
 		 * it.
 		 */
 		pi->xact_addr_filt = -1;
-		netif_carrier_off(netdev);
 		netdev->irq = pdev->irq;
 
-		netdev->hw_features = NETIF_F_SG | TSO_FLAGS |
-			NETIF_F_IP_CSUM | NETIF_F_IPV6_CSUM |
-			NETIF_F_HW_VLAN_CTAG_RX | NETIF_F_RXCSUM;
-		netdev->vlan_features = NETIF_F_SG | TSO_FLAGS |
-			NETIF_F_IP_CSUM | NETIF_F_IPV6_CSUM |
-			NETIF_F_HIGHDMA;
-		netdev->features = netdev->hw_features |
-				   NETIF_F_HW_VLAN_CTAG_TX;
-		if (pci_using_dac)
-			netdev->features |= NETIF_F_HIGHDMA;
+		netdev->hw_features = NETIF_F_SG | TSO_FLAGS | NETIF_F_GRO |
+			NETIF_F_IP_CSUM | NETIF_F_IPV6_CSUM | NETIF_F_RXCSUM |
+			NETIF_F_HW_VLAN_CTAG_TX | NETIF_F_HW_VLAN_CTAG_RX;
+		netdev->features = netdev->hw_features | NETIF_F_HIGHDMA;
+		netdev->vlan_features = netdev->features & VLAN_FEAT;
 
 		netdev->priv_flags |= IFF_UNICAST_FLT;
 		netdev->min_mtu = 81;
@@ -2966,6 +3082,23 @@ static int cxgb4vf_pci_probe(struct pci_dev *pdev,
 		netdev->dev_port = pi->port_id;
 
 		/*
+		 * If we haven't been able to contact the firmware, there's
+		 * nothing else we can do for this "port" ...
+		 */
+		if (!(adapter->flags & CXGB4VF_FW_OK))
+			continue;
+
+		viid = t4vf_alloc_vi(adapter, port_id);
+		if (viid < 0) {
+			dev_err(&pdev->dev,
+				"cannot allocate VI for port %d: err=%d\n",
+				port_id, viid);
+			err = viid;
+			goto err_free_dev;
+		}
+		pi->viid = viid;
+
+		/*
 		 * Initialize the hardware/software state for the port.
 		 */
 		err = t4vf_port_init(adapter, pidx);
@@ -2975,7 +3108,7 @@ static int cxgb4vf_pci_probe(struct pci_dev *pdev,
 			goto err_free_dev;
 		}
 
-		err = t4vf_get_vf_mac_acl(adapter, pf, &naddr, mac);
+		err = t4vf_get_vf_mac_acl(adapter, port_id, &naddr, mac);
 		if (err) {
 			dev_err(&pdev->dev,
 				"unable to determine MAC ACL address, "
@@ -3002,7 +3135,7 @@ static int cxgb4vf_pci_probe(struct pci_dev *pdev,
 	 * get MSI interrupts we bail with the error.
 	 */
 	if (msi == MSI_MSIX && enable_msix(adapter) == 0)
-		adapter->flags |= USING_MSIX;
+		adapter->flags |= CXGB4VF_USING_MSIX;
 	else {
 		if (msi == MSI_MSIX) {
 			dev_info(adapter->pdev_dev,
@@ -3022,7 +3155,7 @@ static int cxgb4vf_pci_probe(struct pci_dev *pdev,
 				" err=%d\n", err);
 			goto err_free_dev;
 		}
-		adapter->flags |= USING_MSI;
+		adapter->flags |= CXGB4VF_USING_MSI;
 	}
 
 	/* Now that we know how many "ports" we have and what interrupt
@@ -3052,10 +3185,12 @@ static int cxgb4vf_pci_probe(struct pci_dev *pdev,
 			continue;
 		}
 
+		netif_carrier_off(netdev);
 		set_bit(pidx, &adapter->registered_device_map);
 	}
 	if (adapter->registered_device_map == 0) {
 		dev_err(&pdev->dev, "could not register any net devices\n");
+		err = -EINVAL;
 		goto err_disable_interrupts;
 	}
 
@@ -3066,11 +3201,7 @@ static int cxgb4vf_pci_probe(struct pci_dev *pdev,
 		adapter->debugfs_root =
 			debugfs_create_dir(pci_name(pdev),
 					   cxgb4vf_debugfs_root);
-		if (IS_ERR_OR_NULL(adapter->debugfs_root))
-			dev_warn(&pdev->dev, "could not create debugfs"
-				 " directory");
-		else
-			setup_debugfs(adapter);
+		setup_debugfs(adapter);
 	}
 
 	/*
@@ -3080,8 +3211,8 @@ static int cxgb4vf_pci_probe(struct pci_dev *pdev,
 	for_each_port(adapter, pidx) {
 		dev_info(adapter->pdev_dev, "%s: Chelsio VF NIC PCIe %s\n",
 			 adapter->port[pidx]->name,
-			 (adapter->flags & USING_MSIX) ? "MSI-X" :
-			 (adapter->flags & USING_MSI)  ? "MSI" : "");
+			 (adapter->flags & CXGB4VF_USING_MSIX) ? "MSI-X" :
+			 (adapter->flags & CXGB4VF_USING_MSI)  ? "MSI" : "");
 	}
 
 	/*
@@ -3094,12 +3225,12 @@ static int cxgb4vf_pci_probe(struct pci_dev *pdev,
 	 * so far and return the error.
 	 */
 err_disable_interrupts:
-	if (adapter->flags & USING_MSIX) {
+	if (adapter->flags & CXGB4VF_USING_MSIX) {
 		pci_disable_msix(adapter->pdev);
-		adapter->flags &= ~USING_MSIX;
-	} else if (adapter->flags & USING_MSI) {
+		adapter->flags &= ~CXGB4VF_USING_MSIX;
+	} else if (adapter->flags & CXGB4VF_USING_MSI) {
 		pci_disable_msi(adapter->pdev);
-		adapter->flags &= ~USING_MSI;
+		adapter->flags &= ~CXGB4VF_USING_MSI;
 	}
 
 err_free_dev:
@@ -3108,13 +3239,13 @@ err_free_dev:
 		if (netdev == NULL)
 			continue;
 		pi = netdev_priv(netdev);
-		t4vf_free_vi(adapter, pi->viid);
+		if (pi->viid)
+			t4vf_free_vi(adapter, pi->viid);
 		if (test_bit(pidx, &adapter->registered_device_map))
 			unregister_netdev(netdev);
 		free_netdev(netdev);
 	}
 
-err_unmap_bar:
 	if (!is_t4(adapter->params.chip))
 		iounmap(adapter->bar2);
 
@@ -3127,7 +3258,6 @@ err_free_adapter:
 
 err_release_regions:
 	pci_release_regions(pdev);
-	pci_clear_master(pdev);
 
 err_disable_device:
 	pci_disable_device(pdev);
@@ -3143,6 +3273,7 @@ err_disable_device:
 static void cxgb4vf_pci_remove(struct pci_dev *pdev)
 {
 	struct adapter *adapter = pci_get_drvdata(pdev);
+	struct hash_mac_addr *entry, *tmp;
 
 	/*
 	 * Tear down driver state associated with device.
@@ -3158,12 +3289,12 @@ static void cxgb4vf_pci_remove(struct pci_dev *pdev)
 			if (test_bit(pidx, &adapter->registered_device_map))
 				unregister_netdev(adapter->port[pidx]);
 		t4vf_sge_stop(adapter);
-		if (adapter->flags & USING_MSIX) {
+		if (adapter->flags & CXGB4VF_USING_MSIX) {
 			pci_disable_msix(adapter->pdev);
-			adapter->flags &= ~USING_MSIX;
-		} else if (adapter->flags & USING_MSI) {
+			adapter->flags &= ~CXGB4VF_USING_MSIX;
+		} else if (adapter->flags & CXGB4VF_USING_MSI) {
 			pci_disable_msi(adapter->pdev);
-			adapter->flags &= ~USING_MSI;
+			adapter->flags &= ~CXGB4VF_USING_MSI;
 		}
 
 		/*
@@ -3186,13 +3317,19 @@ static void cxgb4vf_pci_remove(struct pci_dev *pdev)
 				continue;
 
 			pi = netdev_priv(netdev);
-			t4vf_free_vi(adapter, pi->viid);
+			if (pi->viid)
+				t4vf_free_vi(adapter, pi->viid);
 			free_netdev(netdev);
 		}
 		iounmap(adapter->regs);
 		if (!is_t4(adapter->params.chip))
 			iounmap(adapter->bar2);
 		kfree(adapter->mbox_log);
+		list_for_each_entry_safe(entry, tmp, &adapter->mac_hlist,
+					 list) {
+			list_del(&entry->list);
+			kfree(entry);
+		}
 		kfree(adapter);
 	}
 
@@ -3200,7 +3337,6 @@ static void cxgb4vf_pci_remove(struct pci_dev *pdev)
 	 * Disable the device and release its PCI resources.
 	 */
 	pci_disable_device(pdev);
-	pci_clear_master(pdev);
 	pci_release_regions(pdev);
 }
 
@@ -3229,12 +3365,12 @@ static void cxgb4vf_pci_shutdown(struct pci_dev *pdev)
 	 * Interrupts allowing various internal pathways to drain.
 	 */
 	t4vf_sge_stop(adapter);
-	if (adapter->flags & USING_MSIX) {
+	if (adapter->flags & CXGB4VF_USING_MSIX) {
 		pci_disable_msix(adapter->pdev);
-		adapter->flags &= ~USING_MSIX;
-	} else if (adapter->flags & USING_MSI) {
+		adapter->flags &= ~CXGB4VF_USING_MSIX;
+	} else if (adapter->flags & CXGB4VF_USING_MSI) {
 		pci_disable_msi(adapter->pdev);
-		adapter->flags &= ~USING_MSI;
+		adapter->flags &= ~CXGB4VF_USING_MSI;
 	}
 
 	/*
@@ -3261,7 +3397,6 @@ static void cxgb4vf_pci_shutdown(struct pci_dev *pdev)
 MODULE_DESCRIPTION(DRV_DESC);
 MODULE_AUTHOR("Chelsio Communications");
 MODULE_LICENSE("Dual BSD/GPL");
-MODULE_VERSION(DRV_VERSION);
 MODULE_DEVICE_TABLE(pci, cxgb4vf_pci_tbl);
 
 static struct pci_driver cxgb4vf_driver = {
@@ -3288,13 +3423,11 @@ static int __init cxgb4vf_module_init(void)
 		return -EINVAL;
 	}
 
-	/* Debugfs support is optional, just warn if this fails */
+	/* Debugfs support is optional, debugfs will warn if this fails */
 	cxgb4vf_debugfs_root = debugfs_create_dir(KBUILD_MODNAME, NULL);
-	if (IS_ERR_OR_NULL(cxgb4vf_debugfs_root))
-		pr_warn("could not create debugfs entry, continuing\n");
 
 	ret = pci_register_driver(&cxgb4vf_driver);
-	if (ret < 0 && !IS_ERR_OR_NULL(cxgb4vf_debugfs_root))
+	if (ret < 0)
 		debugfs_remove(cxgb4vf_debugfs_root);
 	return ret;
 }
diff --git a/drivers/net/ethernet/chelsio/cxgb4vf/sge.c b/drivers/net/ethernet/chelsio/cxgb4vf/sge.c
index e37dde2ba97f..31fab2415743 100644
--- a/drivers/net/ethernet/chelsio/cxgb4vf/sge.c
+++ b/drivers/net/ethernet/chelsio/cxgb4vf/sge.c
@@ -478,7 +478,7 @@ static void free_rx_bufs(struct adapter *adapter, struct sge_fl *fl, int n)
 		if (is_buf_mapped(sdesc))
 			dma_unmap_page(adapter->pdev_dev, get_buf_addr(sdesc),
 				       get_buf_size(adapter, sdesc),
-				       PCI_DMA_FROMDEVICE);
+				       DMA_FROM_DEVICE);
 		put_page(sdesc->page);
 		sdesc->page = NULL;
 		if (++fl->cidx == fl->size)
@@ -507,7 +507,7 @@ static void unmap_rx_buf(struct adapter *adapter, struct sge_fl *fl)
 	if (is_buf_mapped(sdesc))
 		dma_unmap_page(adapter->pdev_dev, get_buf_addr(sdesc),
 			       get_buf_size(adapter, sdesc),
-			       PCI_DMA_FROMDEVICE);
+			       DMA_FROM_DEVICE);
 	sdesc->page = NULL;
 	if (++fl->cidx == fl->size)
 		fl->cidx = 0;
@@ -644,7 +644,7 @@ static unsigned int refill_fl(struct adapter *adapter, struct sge_fl *fl,
 
 		dma_addr = dma_map_page(adapter->pdev_dev, page, 0,
 					PAGE_SIZE << s->fl_pg_order,
-					PCI_DMA_FROMDEVICE);
+					DMA_FROM_DEVICE);
 		if (unlikely(dma_mapping_error(adapter->pdev_dev, dma_addr))) {
 			/*
 			 * We've run out of DMA mapping space.  Free up the
@@ -682,7 +682,7 @@ alloc_small_pages:
 		poison_buf(page, PAGE_SIZE);
 
 		dma_addr = dma_map_page(adapter->pdev_dev, page, 0, PAGE_SIZE,
-				       PCI_DMA_FROMDEVICE);
+				       DMA_FROM_DEVICE);
 		if (unlikely(dma_mapping_error(adapter->pdev_dev, dma_addr))) {
 			put_page(page);
 			break;
@@ -776,11 +776,6 @@ static void *alloc_ring(struct device *dev, size_t nelem, size_t hwsize,
 		*(void **)swringp = swring;
 	}
 
-	/*
-	 * Zero out the hardware ring and return its address as our function
-	 * value.
-	 */
-	memset(hwring, 0, hwlen);
 	return hwring;
 }
 
@@ -959,7 +954,7 @@ static void write_sgl(const struct sk_buff *skb, struct sge_txq *tq,
 }
 
 /**
- *	check_ring_tx_db - check and potentially ring a TX queue's doorbell
+ *	ring_tx_db - check and potentially ring a TX queue's doorbell
  *	@adapter: the adapter
  *	@tq: the TX queue
  *	@n: number of new descriptors to give to HW
@@ -1159,7 +1154,7 @@ static inline void txq_advance(struct sge_txq *tq, unsigned int n)
  *
  *	Add a packet to an SGE Ethernet TX queue.  Runs with softirqs disabled.
  */
-int t4vf_eth_xmit(struct sk_buff *skb, struct net_device *dev)
+netdev_tx_t t4vf_eth_xmit(struct sk_buff *skb, struct net_device *dev)
 {
 	u32 wr_mid;
 	u64 cntrl, *end;
@@ -1172,10 +1167,7 @@ int t4vf_eth_xmit(struct sk_buff *skb, struct net_device *dev)
 	struct cpl_tx_pkt_core *cpl;
 	const struct skb_shared_info *ssi;
 	dma_addr_t addr[MAX_SKB_FRAGS + 1];
-	const size_t fw_hdr_copy_len = (sizeof(wr->ethmacdst) +
-					sizeof(wr->ethmacsrc) +
-					sizeof(wr->ethtype) +
-					sizeof(wr->vlantci));
+	const size_t fw_hdr_copy_len = sizeof(wr->firmware);
 
 	/*
 	 * The chip minimum packet length is 10 octets but the firmware
@@ -1202,6 +1194,10 @@ int t4vf_eth_xmit(struct sk_buff *skb, struct net_device *dev)
 	BUG_ON(qidx >= pi->nqsets);
 	txq = &adapter->sge.ethtxq[pi->first_qset + qidx];
 
+	if (pi->vlan_id && !skb_vlan_tag_present(skb))
+		__vlan_hwaccel_put_tag(skb, cpu_to_be16(ETH_P_8021Q),
+				       pi->vlan_id);
+
 	/*
 	 * Take this opportunity to reclaim any TX Descriptors whose DMA
 	 * transfers have completed.
@@ -1268,7 +1264,7 @@ int t4vf_eth_xmit(struct sk_buff *skb, struct net_device *dev)
 	wr->equiq_to_len16 = cpu_to_be32(wr_mid);
 	wr->r3[0] = cpu_to_be32(0);
 	wr->r3[1] = cpu_to_be32(0);
-	skb_copy_from_linear_data(skb, (void *)wr->ethmacdst, fw_hdr_copy_len);
+	skb_copy_from_linear_data(skb, &wr->firmware, fw_hdr_copy_len);
 	end = (u64 *)wr + flits;
 
 	/*
@@ -1570,6 +1566,7 @@ static void do_gro(struct sge_eth_rxq *rxq, const struct pkt_gl *gl,
 {
 	struct adapter *adapter = rxq->rspq.adapter;
 	struct sge *s = &adapter->sge;
+	struct port_info *pi;
 	int ret;
 	struct sk_buff *skb;
 
@@ -1586,8 +1583,9 @@ static void do_gro(struct sge_eth_rxq *rxq, const struct pkt_gl *gl,
 	skb->truesize += skb->data_len;
 	skb->ip_summed = CHECKSUM_UNNECESSARY;
 	skb_record_rx_queue(skb, rxq->rspq.idx);
+	pi = netdev_priv(skb->dev);
 
-	if (pkt->vlan_ex) {
+	if (pkt->vlan_ex && !pi->vlan_id) {
 		__vlan_hwaccel_put_tag(skb, cpu_to_be16(ETH_P_8021Q),
 					be16_to_cpu(pkt->vlan));
 		rxq->stats.vlan_ex++;
@@ -1620,6 +1618,7 @@ int t4vf_ethrx_handler(struct sge_rspq *rspq, const __be64 *rsp,
 	struct sge_eth_rxq *rxq = container_of(rspq, struct sge_eth_rxq, rspq);
 	struct adapter *adapter = rspq->adapter;
 	struct sge *s = &adapter->sge;
+	struct port_info *pi;
 
 	/*
 	 * If this is a good TCP packet and we have Generic Receive Offload
@@ -1644,6 +1643,7 @@ int t4vf_ethrx_handler(struct sge_rspq *rspq, const __be64 *rsp,
 	__skb_pull(skb, s->pktshift);
 	skb->protocol = eth_type_trans(skb, rspq->netdev);
 	skb_record_rx_queue(skb, rspq->idx);
+	pi = netdev_priv(skb->dev);
 	rxq->stats.pkts++;
 
 	if (csum_ok && !pkt->err_vec &&
@@ -1660,9 +1660,10 @@ int t4vf_ethrx_handler(struct sge_rspq *rspq, const __be64 *rsp,
 	} else
 		skb_checksum_none_assert(skb);
 
-	if (pkt->vlan_ex) {
+	if (pkt->vlan_ex && !pi->vlan_id) {
 		rxq->stats.vlan_ex++;
-		__vlan_hwaccel_put_tag(skb, htons(ETH_P_8021Q), be16_to_cpu(pkt->vlan));
+		__vlan_hwaccel_put_tag(skb, htons(ETH_P_8021Q),
+				       be16_to_cpu(pkt->vlan));
 	}
 
 	netif_receive_skb(skb);
@@ -1688,7 +1689,7 @@ static inline bool is_new_response(const struct rsp_ctrl *rc,
  *	restore_rx_bufs - put back a packet's RX buffers
  *	@gl: the packet gather list
  *	@fl: the SGE Free List
- *	@nfrags: how many fragments in @si
+ *	@frags: how many fragments in @si
  *
  *	Called when we find out that the current packet, @si, can't be
  *	processed right away for some reason.  This is a very rare event and
@@ -2040,8 +2041,9 @@ static irqreturn_t t4vf_intr_msi(int irq, void *cookie)
  */
 irq_handler_t t4vf_intr_handler(struct adapter *adapter)
 {
-	BUG_ON((adapter->flags & (USING_MSIX|USING_MSI)) == 0);
-	if (adapter->flags & USING_MSIX)
+	BUG_ON((adapter->flags &
+	       (CXGB4VF_USING_MSIX | CXGB4VF_USING_MSI)) == 0);
+	if (adapter->flags & CXGB4VF_USING_MSIX)
 		return t4vf_sge_intr_msix;
 	else
 		return t4vf_intr_msi;
@@ -2049,7 +2051,7 @@ irq_handler_t t4vf_intr_handler(struct adapter *adapter)
 
 /**
  *	sge_rx_timer_cb - perform periodic maintenance of SGE RX queues
- *	@data: the adapter
+ *	@t: Rx timer
  *
  *	Runs periodically from a timer to perform maintenance of SGE RX queues.
  *
@@ -2058,9 +2060,9 @@ irq_handler_t t4vf_intr_handler(struct adapter *adapter)
  *	when out of memory a queue can become empty.  We schedule NAPI to do
  *	the actual refill.
  */
-static void sge_rx_timer_cb(unsigned long data)
+static void sge_rx_timer_cb(struct timer_list *t)
 {
-	struct adapter *adapter = (struct adapter *)data;
+	struct adapter *adapter = timer_container_of(adapter, t, sge.rx_timer);
 	struct sge *s = &adapter->sge;
 	unsigned int i;
 
@@ -2092,7 +2094,7 @@ static void sge_rx_timer_cb(unsigned long data)
 				struct sge_eth_rxq *rxq;
 
 				rxq = container_of(fl, struct sge_eth_rxq, fl);
-				if (napi_reschedule(&rxq->rspq.napi))
+				if (napi_schedule(&rxq->rspq.napi))
 					fl->starving++;
 				else
 					set_bit(id, s->starving_fl);
@@ -2108,7 +2110,7 @@ static void sge_rx_timer_cb(unsigned long data)
 
 /**
  *	sge_tx_timer_cb - perform periodic maintenance of SGE Tx queues
- *	@data: the adapter
+ *	@t: Tx timer
  *
  *	Runs periodically from a timer to perform maintenance of SGE TX queues.
  *
@@ -2117,9 +2119,9 @@ static void sge_rx_timer_cb(unsigned long data)
  *	when no new packets are being submitted.  This is essential for pktgen,
  *	at least.
  */
-static void sge_tx_timer_cb(unsigned long data)
+static void sge_tx_timer_cb(struct timer_list *t)
 {
-	struct adapter *adapter = (struct adapter *)data;
+	struct adapter *adapter = timer_container_of(adapter, t, sge.tx_timer);
 	struct sge *s = &adapter->sge;
 	unsigned int i, budget;
 
@@ -2189,7 +2191,7 @@ static void __iomem *bar2_address(struct adapter *adapter,
 /**
  *	t4vf_sge_alloc_rxq - allocate an SGE RX Queue
  *	@adapter: the adapter
- *	@rspq: pointer to to the new rxq's Response Queue to be filled in
+ *	@rspq: pointer to the new rxq's Response Queue to be filled in
  *	@iqasynch: if 0, a normal rspq; if 1, an asynchronous event queue
  *	@dev: the network device associated with the new rspq
  *	@intr_dest: MSI-X vector index (overriden in MSI mode)
@@ -2205,6 +2207,7 @@ int t4vf_sge_alloc_rxq(struct adapter *adapter, struct sge_rspq *rspq,
 	struct port_info *pi = netdev_priv(dev);
 	struct fw_iq_cmd cmd, rpl;
 	int ret, iqandst, flsz = 0;
+	int relaxed = !(adapter->flags & CXGB4VF_ROOT_NO_RELAXED_ORDERING);
 
 	/*
 	 * If we're using MSI interrupts and we're not initializing the
@@ -2213,7 +2216,8 @@ int t4vf_sge_alloc_rxq(struct adapter *adapter, struct sge_rspq *rspq,
 	 * the Forwarded Interrupt Queue must be set up before any other
 	 * ingress queue ...
 	 */
-	if ((adapter->flags & USING_MSI) && rspq != &adapter->sge.intrq) {
+	if ((adapter->flags & CXGB4VF_USING_MSI) &&
+	    rspq != &adapter->sge.intrq) {
 		iqandst = SGE_INTRDST_IQ;
 		intr_dest = adapter->sge.intrq.abs_id;
 	} else
@@ -2263,7 +2267,7 @@ int t4vf_sge_alloc_rxq(struct adapter *adapter, struct sge_rspq *rspq,
 	cmd.iqaddr = cpu_to_be64(rspq->phys_addr);
 
 	if (fl) {
-		enum chip_type chip =
+		unsigned int chip_ver =
 			CHELSIO_CHIP_VERSION(adapter->params.chip);
 		/*
 		 * Allocate the ring for the hardware free list (with space
@@ -2300,6 +2304,8 @@ int t4vf_sge_alloc_rxq(struct adapter *adapter, struct sge_rspq *rspq,
 			cpu_to_be32(
 				FW_IQ_CMD_FL0HOSTFCMODE_V(SGE_HOSTFCMODE_NONE) |
 				FW_IQ_CMD_FL0PACKEN_F |
+				FW_IQ_CMD_FL0FETCHRO_V(relaxed) |
+				FW_IQ_CMD_FL0DATARO_V(relaxed) |
 				FW_IQ_CMD_FL0PADEN_F);
 
 		/* In T6, for egress queue type FL there is internal overhead
@@ -2312,10 +2318,10 @@ int t4vf_sge_alloc_rxq(struct adapter *adapter, struct sge_rspq *rspq,
 		 */
 		cmd.fl0dcaen_to_fl0cidxfthresh =
 			cpu_to_be16(
-				FW_IQ_CMD_FL0FBMIN_V(chip <= CHELSIO_T5 ?
-						     FETCHBURSTMIN_128B_X :
-						     FETCHBURSTMIN_64B_X) |
-				FW_IQ_CMD_FL0FBMAX_V((chip <= CHELSIO_T5) ?
+				FW_IQ_CMD_FL0FBMIN_V(chip_ver <= CHELSIO_T5
+						     ? FETCHBURSTMIN_128B_X
+						     : FETCHBURSTMIN_64B_T6_X) |
+				FW_IQ_CMD_FL0FBMAX_V((chip_ver <= CHELSIO_T5) ?
 						     FETCHBURSTMAX_512B_X :
 						     FETCHBURSTMAX_256B_X));
 		cmd.fl0size = cpu_to_be16(flsz);
@@ -2330,7 +2336,7 @@ int t4vf_sge_alloc_rxq(struct adapter *adapter, struct sge_rspq *rspq,
 	if (ret)
 		goto err;
 
-	netif_napi_add(dev, &rspq->napi, napi_rx_handler, 64);
+	netif_napi_add(dev, &rspq->napi, napi_rx_handler);
 	rspq->cur_desc = rspq->desc;
 	rspq->cidx = 0;
 	rspq->gen = 1;
@@ -2396,6 +2402,7 @@ err:
  *	t4vf_sge_alloc_eth_txq - allocate an SGE Ethernet TX Queue
  *	@adapter: the adapter
  *	@txq: pointer to the new txq to be filled in
+ *	@dev: the network device
  *	@devq: the network TX queue associated with the new txq
  *	@iqid: the relative ingress queue ID to which events relating to
  *		the new txq should be directed
@@ -2404,10 +2411,11 @@ int t4vf_sge_alloc_eth_txq(struct adapter *adapter, struct sge_eth_txq *txq,
 			   struct net_device *dev, struct netdev_queue *devq,
 			   unsigned int iqid)
 {
+	unsigned int chip_ver = CHELSIO_CHIP_VERSION(adapter->params.chip);
+	struct port_info *pi = netdev_priv(dev);
+	struct fw_eq_eth_cmd cmd, rpl;
 	struct sge *s = &adapter->sge;
 	int ret, nentries;
-	struct fw_eq_eth_cmd cmd, rpl;
-	struct port_info *pi = netdev_priv(dev);
 
 	/*
 	 * Calculate the size of the hardware TX Queue (including the Status
@@ -2441,17 +2449,19 @@ int t4vf_sge_alloc_eth_txq(struct adapter *adapter, struct sge_eth_txq *txq,
 	cmd.alloc_to_len16 = cpu_to_be32(FW_EQ_ETH_CMD_ALLOC_F |
 					 FW_EQ_ETH_CMD_EQSTART_F |
 					 FW_LEN16(cmd));
-	cmd.viid_pkd = cpu_to_be32(FW_EQ_ETH_CMD_AUTOEQUEQE_F |
-				   FW_EQ_ETH_CMD_VIID_V(pi->viid));
+	cmd.autoequiqe_to_viid = cpu_to_be32(FW_EQ_ETH_CMD_AUTOEQUEQE_F |
+					     FW_EQ_ETH_CMD_VIID_V(pi->viid));
 	cmd.fetchszm_to_iqid =
 		cpu_to_be32(FW_EQ_ETH_CMD_HOSTFCMODE_V(SGE_HOSTFCMODE_STPG) |
 			    FW_EQ_ETH_CMD_PCIECHN_V(pi->port_id) |
 			    FW_EQ_ETH_CMD_IQID_V(iqid));
 	cmd.dcaen_to_eqsize =
-		cpu_to_be32(FW_EQ_ETH_CMD_FBMIN_V(SGE_FETCHBURSTMIN_64B) |
-			    FW_EQ_ETH_CMD_FBMAX_V(SGE_FETCHBURSTMAX_512B) |
+		cpu_to_be32(FW_EQ_ETH_CMD_FBMIN_V(chip_ver <= CHELSIO_T5
+						  ? FETCHBURSTMIN_64B_X
+						  : FETCHBURSTMIN_64B_T6_X) |
+			    FW_EQ_ETH_CMD_FBMAX_V(FETCHBURSTMAX_512B_X) |
 			    FW_EQ_ETH_CMD_CIDXFTHRESH_V(
-						SGE_CIDXFLUSHTHRESH_32) |
+						CIDXFLUSHTHRESH_32_X) |
 			    FW_EQ_ETH_CMD_EQSIZE_V(nentries));
 	cmd.eqaddr = cpu_to_be64(txq->q.phys_addr);
 
@@ -2599,9 +2609,9 @@ void t4vf_sge_stop(struct adapter *adapter)
 	struct sge *s = &adapter->sge;
 
 	if (s->rx_timer.function)
-		del_timer_sync(&s->rx_timer);
+		timer_delete_sync(&s->rx_timer);
 	if (s->tx_timer.function)
-		del_timer_sync(&s->tx_timer);
+		timer_delete_sync(&s->tx_timer);
 }
 
 /**
@@ -2616,8 +2626,8 @@ void t4vf_sge_stop(struct adapter *adapter)
 int t4vf_sge_init(struct adapter *adapter)
 {
 	struct sge_params *sge_params = &adapter->params.sge;
-	u32 fl0 = sge_params->sge_fl_buffer_size[0];
-	u32 fl1 = sge_params->sge_fl_buffer_size[1];
+	u32 fl_small_pg = sge_params->sge_fl_buffer_size[0];
+	u32 fl_large_pg = sge_params->sge_fl_buffer_size[1];
 	struct sge *s = &adapter->sge;
 
 	/*
@@ -2625,9 +2635,20 @@ int t4vf_sge_init(struct adapter *adapter)
 	 * the Physical Function Driver.  Ideally we should be able to deal
 	 * with _any_ configuration.  Practice is different ...
 	 */
-	if (fl0 != PAGE_SIZE || (fl1 != 0 && fl1 <= fl0)) {
+
+	/* We only bother using the Large Page logic if the Large Page Buffer
+	 * is larger than our Page Size Buffer.
+	 */
+	if (fl_large_pg <= fl_small_pg)
+		fl_large_pg = 0;
+
+	/* The Page Size Buffer must be exactly equal to our Page Size and the
+	 * Large Page Size Buffer should be 0 (per above) or a power of 2.
+	 */
+	if (fl_small_pg != PAGE_SIZE ||
+	    (fl_large_pg & (fl_large_pg - 1)) != 0) {
 		dev_err(adapter->pdev_dev, "bad SGE FL buffer sizes [%d, %d]\n",
-			fl0, fl1);
+			fl_small_pg, fl_large_pg);
 		return -EINVAL;
 	}
 	if ((sge_params->sge_control & RXPKTCPLMODE_F) !=
@@ -2639,8 +2660,8 @@ int t4vf_sge_init(struct adapter *adapter)
 	/*
 	 * Now translate the adapter parameters into our internal forms.
 	 */
-	if (fl1)
-		s->fl_pg_order = ilog2(fl1) - PAGE_SHIFT;
+	if (fl_large_pg)
+		s->fl_pg_order = ilog2(fl_large_pg) - PAGE_SHIFT;
 	s->stat_len = ((sge_params->sge_control & EGRSTATUSPAGESIZE_F)
 			? 128 : 64);
 	s->pktshift = PKTSHIFT_G(sge_params->sge_control);
@@ -2673,8 +2694,8 @@ int t4vf_sge_init(struct adapter *adapter)
 	/*
 	 * Set up tasklet timers.
 	 */
-	setup_timer(&s->rx_timer, sge_rx_timer_cb, (unsigned long)adapter);
-	setup_timer(&s->tx_timer, sge_tx_timer_cb, (unsigned long)adapter);
+	timer_setup(&s->rx_timer, sge_rx_timer_cb, 0);
+	timer_setup(&s->tx_timer, sge_tx_timer_cb, 0);
 
 	/*
 	 * Initialize Forwarded Interrupt Queue lock.
diff --git a/drivers/net/ethernet/chelsio/cxgb4vf/t4vf_common.h b/drivers/net/ethernet/chelsio/cxgb4vf/t4vf_common.h
index b3903fe411aa..03777145efec 100644
--- a/drivers/net/ethernet/chelsio/cxgb4vf/t4vf_common.h
+++ b/drivers/net/ethernet/chelsio/cxgb4vf/t4vf_common.h
@@ -104,24 +104,63 @@ struct t4vf_port_stats {
 /*
  * Per-"port" (Virtual Interface) link configuration ...
  */
-struct link_config {
-	unsigned int   supported;        /* link capabilities */
-	unsigned int   advertising;      /* advertised capabilities */
-	unsigned short lp_advertising;   /* peer advertised capabilities */
-	unsigned int   requested_speed;  /* speed user has requested */
-	unsigned int   speed;            /* actual link speed */
-	unsigned char  requested_fc;     /* flow control user has requested */
-	unsigned char  fc;               /* actual link flow control */
-	unsigned char  autoneg;          /* autonegotiating? */
-	unsigned char  link_ok;          /* link up? */
+typedef u16 fw_port_cap16_t;    /* 16-bit Port Capabilities integral value */
+typedef u32 fw_port_cap32_t;    /* 32-bit Port Capabilities integral value */
+
+enum fw_caps {
+	FW_CAPS_UNKNOWN	= 0,	/* 0'ed out initial state */
+	FW_CAPS16	= 1,	/* old Firmware: 16-bit Port Capabilities */
+	FW_CAPS32	= 2,	/* new Firmware: 32-bit Port Capabilities */
 };
 
-enum {
-	PAUSE_RX      = 1 << 0,
-	PAUSE_TX      = 1 << 1,
-	PAUSE_AUTONEG = 1 << 2
+enum cc_pause {
+	PAUSE_RX	= 1 << 0,
+	PAUSE_TX	= 1 << 1,
+	PAUSE_AUTONEG	= 1 << 2
+};
+
+enum cc_fec {
+	FEC_AUTO	= 1 << 0,	/* IEEE 802.3 "automatic" */
+	FEC_RS		= 1 << 1,	/* Reed-Solomon */
+	FEC_BASER_RS	= 1 << 2,	/* BaseR/Reed-Solomon */
+};
+
+struct link_config {
+	fw_port_cap32_t pcaps;		/* link capabilities */
+	fw_port_cap32_t	acaps;		/* advertised capabilities */
+	fw_port_cap32_t	lpacaps;	/* peer advertised capabilities */
+
+	fw_port_cap32_t	speed_caps;	/* speed(s) user has requested */
+	u32		speed;		/* actual link speed */
+
+	enum cc_pause	requested_fc;	/* flow control user has requested */
+	enum cc_pause	fc;		/* actual link flow control */
+	enum cc_pause   advertised_fc;  /* actual advertised flow control */
+
+	enum cc_fec	auto_fec;	/* Forward Error Correction: */
+	enum cc_fec	requested_fec;	/*   "automatic" (IEEE 802.3), */
+	enum cc_fec	fec;		/*   requested, and actual in use */
+
+	unsigned char	autoneg;	/* autonegotiating? */
+
+	unsigned char	link_ok;	/* link up? */
+	unsigned char	link_down_rc;	/* link down reason */
 };
 
+/* Return true if the Link Configuration supports "High Speeds" (those greater
+ * than 1Gb/s).
+ */
+static inline bool is_x_10g_port(const struct link_config *lc)
+{
+	fw_port_cap32_t speeds, high_speeds;
+
+	speeds = FW_PORT_CAP32_SPEED_V(FW_PORT_CAP32_SPEED_G(lc->pcaps));
+	high_speeds =
+		speeds & ~(FW_PORT_CAP32_SPEED_100M | FW_PORT_CAP32_SPEED_1G);
+
+	return high_speeds != 0;
+}
+
 /*
  * General device parameters ...
  */
@@ -227,6 +266,7 @@ struct adapter_params {
 	struct arch_specific_params arch; /* chip specific params */
 	enum chip_type chip;		/* chip code */
 	u8 nports;			/* # of Ethernet "ports" */
+	u8 fw_caps_support;		/* 32-bit Port Capabilities */
 };
 
 /* Firmware Mailbox Command/Reply log.  All values are in Host-Endian format.
@@ -266,24 +306,6 @@ static inline struct mbox_cmd *mbox_cmd_log_entry(struct mbox_cmd_log *log,
 #define for_each_port(adapter, iter) \
 	for (iter = 0; iter < (adapter)->params.nports; iter++)
 
-static inline bool is_10g_port(const struct link_config *lc)
-{
-	return (lc->supported & FW_PORT_CAP_SPEED_10G) != 0;
-}
-
-/* Return true if the Link Configuration supports "High Speeds" (those greater
- * than 1Gb/s).
- */
-static inline bool is_x_10g_port(const struct link_config *lc)
-{
-	unsigned int speeds, high_speeds;
-
-	speeds = FW_PORT_CAP_SPEED_V(FW_PORT_CAP_SPEED_G(lc->supported));
-	high_speeds = speeds & ~(FW_PORT_CAP_SPEED_100M | FW_PORT_CAP_SPEED_1G);
-
-	return high_speeds != 0;
-}
-
 static inline unsigned int core_ticks_per_usec(const struct adapter *adapter)
 {
 	return adapter->params.vpd.cclk / 1000;
@@ -370,7 +392,10 @@ int t4vf_config_rss_range(struct adapter *, unsigned int, int, int,
 
 int t4vf_alloc_vi(struct adapter *, int);
 int t4vf_free_vi(struct adapter *, int);
-int t4vf_enable_vi(struct adapter *, unsigned int, bool, bool);
+int t4vf_enable_vi(struct adapter *adapter, unsigned int viid, bool rx_en,
+		   bool tx_en);
+int t4vf_enable_pi(struct adapter *adapter, struct port_info *pi, bool rx_en,
+		   bool tx_en);
 int t4vf_identify_port(struct adapter *, unsigned int, unsigned int);
 
 int t4vf_set_rxmode(struct adapter *, unsigned int, int, int, int, int, int,
@@ -387,9 +412,11 @@ int t4vf_iq_free(struct adapter *, unsigned int, unsigned int, unsigned int,
 		 unsigned int);
 int t4vf_eth_eq_free(struct adapter *, unsigned int);
 
+int t4vf_update_port_info(struct port_info *pi);
 int t4vf_handle_fw_rpl(struct adapter *, const __be64 *);
 int t4vf_prep_adapter(struct adapter *);
-int t4vf_get_vf_mac_acl(struct adapter *adapter, unsigned int pf,
+int t4vf_get_vf_mac_acl(struct adapter *adapter, unsigned int port,
 			unsigned int *naddr, u8 *addr);
+int t4vf_get_vf_vlan_acl(struct adapter *adapter);
 
 #endif /* __T4VF_COMMON_H__ */
diff --git a/drivers/net/ethernet/chelsio/cxgb4vf/t4vf_hw.c b/drivers/net/ethernet/chelsio/cxgb4vf/t4vf_hw.c
index e98248f00fef..56fcc531af2e 100644
--- a/drivers/net/ethernet/chelsio/cxgb4vf/t4vf_hw.c
+++ b/drivers/net/ethernet/chelsio/cxgb4vf/t4vf_hw.c
@@ -33,6 +33,7 @@
  * SOFTWARE.
  */
 
+#include <linux/ethtool.h>
 #include <linux/pci.h>
 
 #include "t4vf_common.h"
@@ -313,33 +314,177 @@ int t4vf_wr_mbox_core(struct adapter *adapter, const void *cmd, int size,
 	return ret;
 }
 
-#define ADVERT_MASK (FW_PORT_CAP_SPEED_100M | FW_PORT_CAP_SPEED_1G |\
-		     FW_PORT_CAP_SPEED_10G | FW_PORT_CAP_SPEED_25G | \
-		     FW_PORT_CAP_SPEED_40G | FW_PORT_CAP_SPEED_100G | \
-		     FW_PORT_CAP_ANEG)
+/* In the Physical Function Driver Common Code, the ADVERT_MASK is used to
+ * mask out bits in the Advertised Port Capabilities which are managed via
+ * separate controls, like Pause Frames and Forward Error Correction.  In the
+ * Virtual Function Common Code, since we never perform L1 Configuration on
+ * the Link, the only things we really need to filter out are things which
+ * we decode and report separately like Speed.
+ */
+#define ADVERT_MASK (FW_PORT_CAP32_SPEED_V(FW_PORT_CAP32_SPEED_M) | \
+		     FW_PORT_CAP32_802_3_PAUSE | \
+		     FW_PORT_CAP32_802_3_ASM_DIR | \
+		     FW_PORT_CAP32_FEC_V(FW_PORT_CAP32_FEC_M) | \
+		     FW_PORT_CAP32_ANEG)
 
 /**
+ *	fwcaps16_to_caps32 - convert 16-bit Port Capabilities to 32-bits
+ *	@caps16: a 16-bit Port Capabilities value
+ *
+ *	Returns the equivalent 32-bit Port Capabilities value.
+ */
+static fw_port_cap32_t fwcaps16_to_caps32(fw_port_cap16_t caps16)
+{
+	fw_port_cap32_t caps32 = 0;
+
+	#define CAP16_TO_CAP32(__cap) \
+		do { \
+			if (caps16 & FW_PORT_CAP_##__cap) \
+				caps32 |= FW_PORT_CAP32_##__cap; \
+		} while (0)
+
+	CAP16_TO_CAP32(SPEED_100M);
+	CAP16_TO_CAP32(SPEED_1G);
+	CAP16_TO_CAP32(SPEED_25G);
+	CAP16_TO_CAP32(SPEED_10G);
+	CAP16_TO_CAP32(SPEED_40G);
+	CAP16_TO_CAP32(SPEED_100G);
+	CAP16_TO_CAP32(FC_RX);
+	CAP16_TO_CAP32(FC_TX);
+	CAP16_TO_CAP32(ANEG);
+	CAP16_TO_CAP32(MDIAUTO);
+	CAP16_TO_CAP32(MDISTRAIGHT);
+	CAP16_TO_CAP32(FEC_RS);
+	CAP16_TO_CAP32(FEC_BASER_RS);
+	CAP16_TO_CAP32(802_3_PAUSE);
+	CAP16_TO_CAP32(802_3_ASM_DIR);
+
+	#undef CAP16_TO_CAP32
+
+	return caps32;
+}
+
+/* Translate Firmware Pause specification to Common Code */
+static inline enum cc_pause fwcap_to_cc_pause(fw_port_cap32_t fw_pause)
+{
+	enum cc_pause cc_pause = 0;
+
+	if (fw_pause & FW_PORT_CAP32_FC_RX)
+		cc_pause |= PAUSE_RX;
+	if (fw_pause & FW_PORT_CAP32_FC_TX)
+		cc_pause |= PAUSE_TX;
+
+	return cc_pause;
+}
+
+/* Translate Firmware Forward Error Correction specification to Common Code */
+static inline enum cc_fec fwcap_to_cc_fec(fw_port_cap32_t fw_fec)
+{
+	enum cc_fec cc_fec = 0;
+
+	if (fw_fec & FW_PORT_CAP32_FEC_RS)
+		cc_fec |= FEC_RS;
+	if (fw_fec & FW_PORT_CAP32_FEC_BASER_RS)
+		cc_fec |= FEC_BASER_RS;
+
+	return cc_fec;
+}
+
+/* Return the highest speed set in the port capabilities, in Mb/s. */
+static unsigned int fwcap_to_speed(fw_port_cap32_t caps)
+{
+	#define TEST_SPEED_RETURN(__caps_speed, __speed) \
+		do { \
+			if (caps & FW_PORT_CAP32_SPEED_##__caps_speed) \
+				return __speed; \
+		} while (0)
+
+	TEST_SPEED_RETURN(400G, 400000);
+	TEST_SPEED_RETURN(200G, 200000);
+	TEST_SPEED_RETURN(100G, 100000);
+	TEST_SPEED_RETURN(50G,   50000);
+	TEST_SPEED_RETURN(40G,   40000);
+	TEST_SPEED_RETURN(25G,   25000);
+	TEST_SPEED_RETURN(10G,   10000);
+	TEST_SPEED_RETURN(1G,     1000);
+	TEST_SPEED_RETURN(100M,    100);
+
+	#undef TEST_SPEED_RETURN
+
+	return 0;
+}
+
+/**
+ *      fwcap_to_fwspeed - return highest speed in Port Capabilities
+ *      @acaps: advertised Port Capabilities
+ *
+ *      Get the highest speed for the port from the advertised Port
+ *      Capabilities.  It will be either the highest speed from the list of
+ *      speeds or whatever user has set using ethtool.
+ */
+static fw_port_cap32_t fwcap_to_fwspeed(fw_port_cap32_t acaps)
+{
+	#define TEST_SPEED_RETURN(__caps_speed) \
+		do { \
+			if (acaps & FW_PORT_CAP32_SPEED_##__caps_speed) \
+				return FW_PORT_CAP32_SPEED_##__caps_speed; \
+		} while (0)
+
+	TEST_SPEED_RETURN(400G);
+	TEST_SPEED_RETURN(200G);
+	TEST_SPEED_RETURN(100G);
+	TEST_SPEED_RETURN(50G);
+	TEST_SPEED_RETURN(40G);
+	TEST_SPEED_RETURN(25G);
+	TEST_SPEED_RETURN(10G);
+	TEST_SPEED_RETURN(1G);
+	TEST_SPEED_RETURN(100M);
+
+	#undef TEST_SPEED_RETURN
+	return 0;
+}
+
+/*
  *	init_link_config - initialize a link's SW state
  *	@lc: structure holding the link state
- *	@caps: link capabilities
+ *	@pcaps: link Port Capabilities
+ *	@acaps: link current Advertised Port Capabilities
  *
  *	Initializes the SW state maintained for each link, including the link's
  *	capabilities and default speed/flow-control/autonegotiation settings.
  */
-static void init_link_config(struct link_config *lc, unsigned int caps)
+static void init_link_config(struct link_config *lc,
+			     fw_port_cap32_t pcaps,
+			     fw_port_cap32_t acaps)
 {
-	lc->supported = caps;
-	lc->lp_advertising = 0;
-	lc->requested_speed = 0;
+	lc->pcaps = pcaps;
+	lc->lpacaps = 0;
+	lc->speed_caps = 0;
 	lc->speed = 0;
 	lc->requested_fc = lc->fc = PAUSE_RX | PAUSE_TX;
-	if (lc->supported & FW_PORT_CAP_ANEG) {
-		lc->advertising = lc->supported & ADVERT_MASK;
+
+	/* For Forward Error Control, we default to whatever the Firmware
+	 * tells us the Link is currently advertising.
+	 */
+	lc->auto_fec = fwcap_to_cc_fec(acaps);
+	lc->requested_fec = FEC_AUTO;
+	lc->fec = lc->auto_fec;
+
+	/* If the Port is capable of Auto-Negtotiation, initialize it as
+	 * "enabled" and copy over all of the Physical Port Capabilities
+	 * to the Advertised Port Capabilities.  Otherwise mark it as
+	 * Auto-Negotiate disabled and select the highest supported speed
+	 * for the link.  Note parallel structure in t4_link_l1cfg_core()
+	 * and t4_handle_get_port_info().
+	 */
+	if (lc->pcaps & FW_PORT_CAP32_ANEG) {
+		lc->acaps = acaps & ADVERT_MASK;
 		lc->autoneg = AUTONEG_ENABLE;
 		lc->requested_fc |= PAUSE_AUTONEG;
 	} else {
-		lc->advertising = 0;
+		lc->acaps = 0;
 		lc->autoneg = AUTONEG_DISABLE;
+		lc->speed_caps = fwcap_to_fwspeed(acaps);
 	}
 }
 
@@ -351,9 +496,30 @@ static void init_link_config(struct link_config *lc, unsigned int caps)
 int t4vf_port_init(struct adapter *adapter, int pidx)
 {
 	struct port_info *pi = adap2pinfo(adapter, pidx);
+	unsigned int fw_caps = adapter->params.fw_caps_support;
 	struct fw_vi_cmd vi_cmd, vi_rpl;
 	struct fw_port_cmd port_cmd, port_rpl;
-	int v;
+	enum fw_port_type port_type;
+	int mdio_addr;
+	fw_port_cap32_t pcaps, acaps;
+	int ret;
+
+	/* If we haven't yet determined whether we're talking to Firmware
+	 * which knows the new 32-bit Port Capabilities, it's time to find
+	 * out now.  This will also tell new Firmware to send us Port Status
+	 * Updates using the new 32-bit Port Capabilities version of the
+	 * Port Information message.
+	 */
+	if (fw_caps == FW_CAPS_UNKNOWN) {
+		u32 param, val;
+
+		param = (FW_PARAMS_MNEM_V(FW_PARAMS_MNEM_PFVF) |
+			 FW_PARAMS_PARAM_X_V(FW_PARAMS_PARAM_PFVF_PORT_CAPS32));
+		val = 1;
+		ret = t4vf_set_params(adapter, 1, &param, &val);
+		fw_caps = (ret == 0 ? FW_CAPS32 : FW_CAPS16);
+		adapter->params.fw_caps_support = fw_caps;
+	}
 
 	/*
 	 * Execute a VI Read command to get our Virtual Interface information
@@ -365,9 +531,9 @@ int t4vf_port_init(struct adapter *adapter, int pidx)
 				       FW_CMD_READ_F);
 	vi_cmd.alloc_to_len16 = cpu_to_be32(FW_LEN16(vi_cmd));
 	vi_cmd.type_viid = cpu_to_be16(FW_VI_CMD_VIID_V(pi->viid));
-	v = t4vf_wr_mbox(adapter, &vi_cmd, sizeof(vi_cmd), &vi_rpl);
-	if (v)
-		return v;
+	ret = t4vf_wr_mbox(adapter, &vi_cmd, sizeof(vi_cmd), &vi_rpl);
+	if (ret != FW_SUCCESS)
+		return ret;
 
 	BUG_ON(pi->port_id != FW_VI_CMD_PORTID_G(vi_rpl.portid_pkd));
 	pi->rss_size = FW_VI_CMD_RSSSIZE_G(be16_to_cpu(vi_rpl.rsssize_pkd));
@@ -385,21 +551,42 @@ int t4vf_port_init(struct adapter *adapter, int pidx)
 					    FW_CMD_REQUEST_F |
 					    FW_CMD_READ_F |
 					    FW_PORT_CMD_PORTID_V(pi->port_id));
-	port_cmd.action_to_len16 =
-		cpu_to_be32(FW_PORT_CMD_ACTION_V(FW_PORT_ACTION_GET_PORT_INFO) |
-			    FW_LEN16(port_cmd));
-	v = t4vf_wr_mbox(adapter, &port_cmd, sizeof(port_cmd), &port_rpl);
-	if (v)
-		return v;
+	port_cmd.action_to_len16 = cpu_to_be32(
+		FW_PORT_CMD_ACTION_V(fw_caps == FW_CAPS16
+				     ? FW_PORT_ACTION_GET_PORT_INFO
+				     : FW_PORT_ACTION_GET_PORT_INFO32) |
+		FW_LEN16(port_cmd));
+	ret = t4vf_wr_mbox(adapter, &port_cmd, sizeof(port_cmd), &port_rpl);
+	if (ret != FW_SUCCESS)
+		return ret;
 
-	v = be32_to_cpu(port_rpl.u.info.lstatus_to_modtype);
-	pi->mdio_addr = (v & FW_PORT_CMD_MDIOCAP_F) ?
-			FW_PORT_CMD_MDIOADDR_G(v) : -1;
-	pi->port_type = FW_PORT_CMD_PTYPE_G(v);
-	pi->mod_type = FW_PORT_MOD_TYPE_NA;
+	/* Extract the various fields from the Port Information message. */
+	if (fw_caps == FW_CAPS16) {
+		u32 lstatus = be32_to_cpu(port_rpl.u.info.lstatus_to_modtype);
+
+		port_type = FW_PORT_CMD_PTYPE_G(lstatus);
+		mdio_addr = ((lstatus & FW_PORT_CMD_MDIOCAP_F)
+			     ? FW_PORT_CMD_MDIOADDR_G(lstatus)
+			     : -1);
+		pcaps = fwcaps16_to_caps32(be16_to_cpu(port_rpl.u.info.pcap));
+		acaps = fwcaps16_to_caps32(be16_to_cpu(port_rpl.u.info.acap));
+	} else {
+		u32 lstatus32 =
+			   be32_to_cpu(port_rpl.u.info32.lstatus32_to_cbllen32);
+
+		port_type = FW_PORT_CMD_PORTTYPE32_G(lstatus32);
+		mdio_addr = ((lstatus32 & FW_PORT_CMD_MDIOCAP32_F)
+			     ? FW_PORT_CMD_MDIOADDR32_G(lstatus32)
+			     : -1);
+		pcaps = be32_to_cpu(port_rpl.u.info32.pcaps32);
+		acaps = be32_to_cpu(port_rpl.u.info32.acaps32);
+	}
 
-	init_link_config(&pi->link_cfg, be16_to_cpu(port_rpl.u.info.pcap));
+	pi->port_type = port_type;
+	pi->mdio_addr = mdio_addr;
+	pi->mod_type = FW_PORT_MOD_TYPE_NA;
 
+	init_link_config(&pi->link_cfg, pcaps, acaps);
 	return 0;
 }
 
@@ -519,7 +706,7 @@ int t4vf_fl_pkt_align(struct adapter *adapter)
 	 * separately.  The actual Ingress Packet Data alignment boundary
 	 * within Packed Buffer Mode is the maximum of these two
 	 * specifications.  (Note that it makes no real practical sense to
-	 * have the Pading Boudary be larger than the Packing Boundary but you
+	 * have the Padding Boundary be larger than the Packing Boundary but you
 	 * could set the chip up that way and, in fact, legacy T4 code would
 	 * end doing this because it would initialize the Padding Boundary and
 	 * leave the Packing Boundary initialized to 0 (16 bytes).)
@@ -690,7 +877,7 @@ int t4vf_get_sge_params(struct adapter *adapter)
 
 	/* T4 uses a single control field to specify both the PCIe Padding and
 	 * Packing Boundary.  T5 introduced the ability to specify these
-	 * separately with the Padding Boundary in SGE_CONTROL and and Packing
+	 * separately with the Padding Boundary in SGE_CONTROL and Packing
 	 * Boundary in SGE_CONTROL2.  So for T5 and later we need to grab
 	 * SGE_CONTROL in order to determine how ingress packet data will be
 	 * laid out in Packed Buffer Mode.  Unfortunately, older versions of
@@ -1222,6 +1409,30 @@ int t4vf_enable_vi(struct adapter *adapter, unsigned int viid,
 }
 
 /**
+ *	t4vf_enable_pi - enable/disable a Port's virtual interface
+ *	@adapter: the adapter
+ *	@pi: the Port Information structure
+ *	@rx_en: 1=enable Rx, 0=disable Rx
+ *	@tx_en: 1=enable Tx, 0=disable Tx
+ *
+ *	Enables/disables a Port's virtual interface.  If the Virtual
+ *	Interface enable/disable operation is successful, we notify the
+ *	OS-specific code of a potential Link Status change via the OS Contract
+ *	API t4vf_os_link_changed().
+ */
+int t4vf_enable_pi(struct adapter *adapter, struct port_info *pi,
+		   bool rx_en, bool tx_en)
+{
+	int ret = t4vf_enable_vi(adapter, pi->viid, rx_en, tx_en);
+
+	if (ret)
+		return ret;
+	t4vf_os_link_changed(adapter, pi->pidx,
+			     rx_en && tx_en && pi->link_cfg.link_ok);
+	return 0;
+}
+
+/**
  *	t4vf_identify_port - identify a VI's port by blinking its LED
  *	@adapter: the adapter
  *	@viid: the Virtual Interface ID
@@ -1255,6 +1466,7 @@ int t4vf_identify_port(struct adapter *adapter, unsigned int viid,
  *	@bcast: 1 to enable broadcast Rx, 0 to disable it, -1 no change
  *	@vlanex: 1 to enable hardware VLAN Tag extraction, 0 to disable it,
  *		-1 no change
+ *	@sleep_ok: call is allowed to sleep
  *
  *	Sets Rx properties of a virtual interface.
  */
@@ -1667,6 +1879,215 @@ int t4vf_eth_eq_free(struct adapter *adapter, unsigned int eqid)
 }
 
 /**
+ *	t4vf_link_down_rc_str - return a string for a Link Down Reason Code
+ *	@link_down_rc: Link Down Reason Code
+ *
+ *	Returns a string representation of the Link Down Reason Code.
+ */
+static const char *t4vf_link_down_rc_str(unsigned char link_down_rc)
+{
+	static const char * const reason[] = {
+		"Link Down",
+		"Remote Fault",
+		"Auto-negotiation Failure",
+		"Reserved",
+		"Insufficient Airflow",
+		"Unable To Determine Reason",
+		"No RX Signal Detected",
+		"Reserved",
+	};
+
+	if (link_down_rc >= ARRAY_SIZE(reason))
+		return "Bad Reason Code";
+
+	return reason[link_down_rc];
+}
+
+/**
+ *	t4vf_handle_get_port_info - process a FW reply message
+ *	@pi: the port info
+ *	@cmd: start of the FW message
+ *
+ *	Processes a GET_PORT_INFO FW reply message.
+ */
+static void t4vf_handle_get_port_info(struct port_info *pi,
+				      const struct fw_port_cmd *cmd)
+{
+	fw_port_cap32_t pcaps, acaps, lpacaps, linkattr;
+	struct link_config *lc = &pi->link_cfg;
+	struct adapter *adapter = pi->adapter;
+	unsigned int speed, fc, fec, adv_fc;
+	enum fw_port_module_type mod_type;
+	int action, link_ok, linkdnrc;
+	enum fw_port_type port_type;
+
+	/* Extract the various fields from the Port Information message. */
+	action = FW_PORT_CMD_ACTION_G(be32_to_cpu(cmd->action_to_len16));
+	switch (action) {
+	case FW_PORT_ACTION_GET_PORT_INFO: {
+		u32 lstatus = be32_to_cpu(cmd->u.info.lstatus_to_modtype);
+
+		link_ok = (lstatus & FW_PORT_CMD_LSTATUS_F) != 0;
+		linkdnrc = FW_PORT_CMD_LINKDNRC_G(lstatus);
+		port_type = FW_PORT_CMD_PTYPE_G(lstatus);
+		mod_type = FW_PORT_CMD_MODTYPE_G(lstatus);
+		pcaps = fwcaps16_to_caps32(be16_to_cpu(cmd->u.info.pcap));
+		acaps = fwcaps16_to_caps32(be16_to_cpu(cmd->u.info.acap));
+		lpacaps = fwcaps16_to_caps32(be16_to_cpu(cmd->u.info.lpacap));
+
+		/* Unfortunately the format of the Link Status in the old
+		 * 16-bit Port Information message isn't the same as the
+		 * 16-bit Port Capabilities bitfield used everywhere else ...
+		 */
+		linkattr = 0;
+		if (lstatus & FW_PORT_CMD_RXPAUSE_F)
+			linkattr |= FW_PORT_CAP32_FC_RX;
+		if (lstatus & FW_PORT_CMD_TXPAUSE_F)
+			linkattr |= FW_PORT_CAP32_FC_TX;
+		if (lstatus & FW_PORT_CMD_LSPEED_V(FW_PORT_CAP_SPEED_100M))
+			linkattr |= FW_PORT_CAP32_SPEED_100M;
+		if (lstatus & FW_PORT_CMD_LSPEED_V(FW_PORT_CAP_SPEED_1G))
+			linkattr |= FW_PORT_CAP32_SPEED_1G;
+		if (lstatus & FW_PORT_CMD_LSPEED_V(FW_PORT_CAP_SPEED_10G))
+			linkattr |= FW_PORT_CAP32_SPEED_10G;
+		if (lstatus & FW_PORT_CMD_LSPEED_V(FW_PORT_CAP_SPEED_25G))
+			linkattr |= FW_PORT_CAP32_SPEED_25G;
+		if (lstatus & FW_PORT_CMD_LSPEED_V(FW_PORT_CAP_SPEED_40G))
+			linkattr |= FW_PORT_CAP32_SPEED_40G;
+		if (lstatus & FW_PORT_CMD_LSPEED_V(FW_PORT_CAP_SPEED_100G))
+			linkattr |= FW_PORT_CAP32_SPEED_100G;
+
+		break;
+	}
+
+	case FW_PORT_ACTION_GET_PORT_INFO32: {
+		u32 lstatus32;
+
+		lstatus32 = be32_to_cpu(cmd->u.info32.lstatus32_to_cbllen32);
+		link_ok = (lstatus32 & FW_PORT_CMD_LSTATUS32_F) != 0;
+		linkdnrc = FW_PORT_CMD_LINKDNRC32_G(lstatus32);
+		port_type = FW_PORT_CMD_PORTTYPE32_G(lstatus32);
+		mod_type = FW_PORT_CMD_MODTYPE32_G(lstatus32);
+		pcaps = be32_to_cpu(cmd->u.info32.pcaps32);
+		acaps = be32_to_cpu(cmd->u.info32.acaps32);
+		lpacaps = be32_to_cpu(cmd->u.info32.lpacaps32);
+		linkattr = be32_to_cpu(cmd->u.info32.linkattr32);
+		break;
+	}
+
+	default:
+		dev_err(adapter->pdev_dev, "Handle Port Information: Bad Command/Action %#x\n",
+			be32_to_cpu(cmd->action_to_len16));
+		return;
+	}
+
+	fec = fwcap_to_cc_fec(acaps);
+	adv_fc = fwcap_to_cc_pause(acaps);
+	fc = fwcap_to_cc_pause(linkattr);
+	speed = fwcap_to_speed(linkattr);
+
+	if (mod_type != pi->mod_type) {
+		/* When a new Transceiver Module is inserted, the Firmware
+		 * will examine any Forward Error Correction parameters
+		 * present in the Transceiver Module i2c EPROM and determine
+		 * the supported and recommended FEC settings from those
+		 * based on IEEE 802.3 standards.  We always record the
+		 * IEEE 802.3 recommended "automatic" settings.
+		 */
+		lc->auto_fec = fec;
+
+		/* Some versions of the early T6 Firmware "cheated" when
+		 * handling different Transceiver Modules by changing the
+		 * underlaying Port Type reported to the Host Drivers.  As
+		 * such we need to capture whatever Port Type the Firmware
+		 * sends us and record it in case it's different from what we
+		 * were told earlier.  Unfortunately, since Firmware is
+		 * forever, we'll need to keep this code here forever, but in
+		 * later T6 Firmware it should just be an assignment of the
+		 * same value already recorded.
+		 */
+		pi->port_type = port_type;
+
+		pi->mod_type = mod_type;
+		t4vf_os_portmod_changed(adapter, pi->pidx);
+	}
+
+	if (link_ok != lc->link_ok || speed != lc->speed ||
+	    fc != lc->fc || adv_fc != lc->advertised_fc ||
+	    fec != lc->fec) {
+		/* something changed */
+		if (!link_ok && lc->link_ok) {
+			lc->link_down_rc = linkdnrc;
+			dev_warn_ratelimited(adapter->pdev_dev,
+					     "Port %d link down, reason: %s\n",
+					     pi->port_id,
+					     t4vf_link_down_rc_str(linkdnrc));
+		}
+		lc->link_ok = link_ok;
+		lc->speed = speed;
+		lc->advertised_fc = adv_fc;
+		lc->fc = fc;
+		lc->fec = fec;
+
+		lc->pcaps = pcaps;
+		lc->lpacaps = lpacaps;
+		lc->acaps = acaps & ADVERT_MASK;
+
+		/* If we're not physically capable of Auto-Negotiation, note
+		 * this as Auto-Negotiation disabled.  Otherwise, we track
+		 * what Auto-Negotiation settings we have.  Note parallel
+		 * structure in init_link_config().
+		 */
+		if (!(lc->pcaps & FW_PORT_CAP32_ANEG)) {
+			lc->autoneg = AUTONEG_DISABLE;
+		} else if (lc->acaps & FW_PORT_CAP32_ANEG) {
+			lc->autoneg = AUTONEG_ENABLE;
+		} else {
+			/* When Autoneg is disabled, user needs to set
+			 * single speed.
+			 * Similar to cxgb4_ethtool.c: set_link_ksettings
+			 */
+			lc->acaps = 0;
+			lc->speed_caps = fwcap_to_speed(acaps);
+			lc->autoneg = AUTONEG_DISABLE;
+		}
+
+		t4vf_os_link_changed(adapter, pi->pidx, link_ok);
+	}
+}
+
+/**
+ *	t4vf_update_port_info - retrieve and update port information if changed
+ *	@pi: the port_info
+ *
+ *	We issue a Get Port Information Command to the Firmware and, if
+ *	successful, we check to see if anything is different from what we
+ *	last recorded and update things accordingly.
+ */
+int t4vf_update_port_info(struct port_info *pi)
+{
+	unsigned int fw_caps = pi->adapter->params.fw_caps_support;
+	struct fw_port_cmd port_cmd;
+	int ret;
+
+	memset(&port_cmd, 0, sizeof(port_cmd));
+	port_cmd.op_to_portid = cpu_to_be32(FW_CMD_OP_V(FW_PORT_CMD) |
+					    FW_CMD_REQUEST_F | FW_CMD_READ_F |
+					    FW_PORT_CMD_PORTID_V(pi->port_id));
+	port_cmd.action_to_len16 = cpu_to_be32(
+		FW_PORT_CMD_ACTION_V(fw_caps == FW_CAPS16
+				     ? FW_PORT_ACTION_GET_PORT_INFO
+				     : FW_PORT_ACTION_GET_PORT_INFO32) |
+		FW_LEN16(port_cmd));
+	ret = t4vf_wr_mbox(pi->adapter, &port_cmd, sizeof(port_cmd),
+			   &port_cmd);
+	if (ret)
+		return ret;
+	t4vf_handle_get_port_info(pi, &port_cmd);
+	return 0;
+}
+
+/**
  *	t4vf_handle_fw_rpl - process a firmware reply message
  *	@adapter: the adapter
  *	@rpl: start of the firmware message
@@ -1685,15 +2106,12 @@ int t4vf_handle_fw_rpl(struct adapter *adapter, const __be64 *rpl)
 		 */
 		const struct fw_port_cmd *port_cmd =
 			(const struct fw_port_cmd *)rpl;
-		u32 stat, mod;
-		int action, port_id, link_ok, speed, fc, pidx;
-
-		/*
-		 * Extract various fields from port status change message.
-		 */
-		action = FW_PORT_CMD_ACTION_G(
+		int action = FW_PORT_CMD_ACTION_G(
 			be32_to_cpu(port_cmd->action_to_len16));
-		if (action != FW_PORT_ACTION_GET_PORT_INFO) {
+		int port_id, pidx;
+
+		if (action != FW_PORT_ACTION_GET_PORT_INFO &&
+		    action != FW_PORT_ACTION_GET_PORT_INFO32) {
 			dev_err(adapter->pdev_dev,
 				"Unknown firmware PORT reply action %x\n",
 				action);
@@ -1702,61 +2120,12 @@ int t4vf_handle_fw_rpl(struct adapter *adapter, const __be64 *rpl)
 
 		port_id = FW_PORT_CMD_PORTID_G(
 			be32_to_cpu(port_cmd->op_to_portid));
-
-		stat = be32_to_cpu(port_cmd->u.info.lstatus_to_modtype);
-		link_ok = (stat & FW_PORT_CMD_LSTATUS_F) != 0;
-		speed = 0;
-		fc = 0;
-		if (stat & FW_PORT_CMD_RXPAUSE_F)
-			fc |= PAUSE_RX;
-		if (stat & FW_PORT_CMD_TXPAUSE_F)
-			fc |= PAUSE_TX;
-		if (stat & FW_PORT_CMD_LSPEED_V(FW_PORT_CAP_SPEED_100M))
-			speed = 100;
-		else if (stat & FW_PORT_CMD_LSPEED_V(FW_PORT_CAP_SPEED_1G))
-			speed = 1000;
-		else if (stat & FW_PORT_CMD_LSPEED_V(FW_PORT_CAP_SPEED_10G))
-			speed = 10000;
-		else if (stat & FW_PORT_CMD_LSPEED_V(FW_PORT_CAP_SPEED_25G))
-			speed = 25000;
-		else if (stat & FW_PORT_CMD_LSPEED_V(FW_PORT_CAP_SPEED_40G))
-			speed = 40000;
-		else if (stat & FW_PORT_CMD_LSPEED_V(FW_PORT_CAP_SPEED_100G))
-			speed = 100000;
-
-		/*
-		 * Scan all of our "ports" (Virtual Interfaces) looking for
-		 * those bound to the physical port which has changed.  If
-		 * our recorded state doesn't match the current state,
-		 * signal that change to the OS code.
-		 */
 		for_each_port(adapter, pidx) {
 			struct port_info *pi = adap2pinfo(adapter, pidx);
-			struct link_config *lc;
 
 			if (pi->port_id != port_id)
 				continue;
-
-			lc = &pi->link_cfg;
-
-			mod = FW_PORT_CMD_MODTYPE_G(stat);
-			if (mod != pi->mod_type) {
-				pi->mod_type = mod;
-				t4vf_os_portmod_changed(adapter, pidx);
-			}
-
-			if (link_ok != lc->link_ok || speed != lc->speed ||
-			    fc != lc->fc) {
-				/* something changed */
-				lc->link_ok = link_ok;
-				lc->speed = speed;
-				lc->fc = fc;
-				lc->supported =
-					be16_to_cpu(port_cmd->u.info.pcap);
-				lc->lp_advertising =
-					be16_to_cpu(port_cmd->u.info.lpacap);
-				t4vf_os_link_changed(adapter, pidx, link_ok);
-			}
+			t4vf_handle_get_port_info(pi, port_cmd);
 		}
 		break;
 	}
@@ -1768,8 +2137,6 @@ int t4vf_handle_fw_rpl(struct adapter *adapter, const __be64 *rpl)
 	return 0;
 }
 
-/**
- */
 int t4vf_prep_adapter(struct adapter *adapter)
 {
 	int err;
@@ -1821,14 +2188,14 @@ int t4vf_prep_adapter(struct adapter *adapter)
  *	t4vf_get_vf_mac_acl - Get the MAC address to be set to
  *			      the VI of this VF.
  *	@adapter: The adapter
- *	@pf: The pf associated with vf
+ *	@port: The port associated with vf
  *	@naddr: the number of ACL MAC addresses returned in addr
  *	@addr: Placeholder for MAC addresses
  *
  *	Find the MAC address to be set to the VF's VI. The requested MAC address
  *	is from the host OS via callback in the PF driver.
  */
-int t4vf_get_vf_mac_acl(struct adapter *adapter, unsigned int pf,
+int t4vf_get_vf_mac_acl(struct adapter *adapter, unsigned int port,
 			unsigned int *naddr, u8 *addr)
 {
 	struct fw_acl_mac_cmd cmd;
@@ -1846,7 +2213,7 @@ int t4vf_get_vf_mac_acl(struct adapter *adapter, unsigned int pf,
 	if (cmd.nmac < *naddr)
 		*naddr = cmd.nmac;
 
-	switch (pf) {
+	switch (port) {
 	case 3:
 		memcpy(addr, cmd.macaddr3, sizeof(cmd.macaddr3));
 		break;
@@ -1863,3 +2230,31 @@ int t4vf_get_vf_mac_acl(struct adapter *adapter, unsigned int pf,
 
 	return ret;
 }
+
+/**
+ *	t4vf_get_vf_vlan_acl - Get the VLAN ID to be set to
+ *                             the VI of this VF.
+ *	@adapter: The adapter
+ *
+ *	Find the VLAN ID to be set to the VF's VI. The requested VLAN ID
+ *	is from the host OS via callback in the PF driver.
+ */
+int t4vf_get_vf_vlan_acl(struct adapter *adapter)
+{
+	struct fw_acl_vlan_cmd cmd;
+	int vlan = 0;
+	int ret = 0;
+
+	cmd.op_to_vfn = htonl(FW_CMD_OP_V(FW_ACL_VLAN_CMD) |
+			      FW_CMD_REQUEST_F | FW_CMD_READ_F);
+
+	/* Note: Do not enable the ACL */
+	cmd.en_to_len16 = cpu_to_be32((unsigned int)FW_LEN16(cmd));
+
+	ret = t4vf_wr_mbox(adapter, &cmd, sizeof(cmd), &cmd);
+
+	if (!ret)
+		vlan = be16_to_cpu(cmd.vlanid[0]);
+
+	return vlan;
+}
diff --git a/drivers/net/ethernet/chelsio/inline_crypto/Kconfig b/drivers/net/ethernet/chelsio/inline_crypto/Kconfig
new file mode 100644
index 000000000000..521955e1f894
--- /dev/null
+++ b/drivers/net/ethernet/chelsio/inline_crypto/Kconfig
@@ -0,0 +1,54 @@
+# SPDX-License-Identifier: GPL-2.0-only
+#
+# Chelsio inline crypto configuration
+#
+
+config CHELSIO_INLINE_CRYPTO
+	bool "Chelsio Inline Crypto support"
+	depends on CHELSIO_T4
+	default y
+	help
+	  Enable support for inline crypto.
+	  Allows enable/disable from list of inline crypto drivers.
+
+if CHELSIO_INLINE_CRYPTO
+
+config CRYPTO_DEV_CHELSIO_TLS
+	tristate "Chelsio Crypto Inline TLS Driver"
+	depends on CHELSIO_T4
+	depends on TLS
+	depends on TLS_TOE
+	help
+	  Support Chelsio Inline TLS with Chelsio crypto accelerator.
+	  Enable inline TLS support for Tx and Rx.
+
+	  To compile this driver as a module, choose M here: the module
+	  will be called chtls.
+
+config CHELSIO_IPSEC_INLINE
+       tristate "Chelsio IPSec XFRM Tx crypto offload"
+       depends on CHELSIO_T4
+       depends on XFRM_OFFLOAD
+       depends on INET_ESP_OFFLOAD || INET6_ESP_OFFLOAD
+       help
+        Support Chelsio Inline IPsec with Chelsio crypto accelerator.
+        Enable inline IPsec support for Tx.
+
+        To compile this driver as a module, choose M here: the module
+        will be called ch_ipsec.
+
+config CHELSIO_TLS_DEVICE
+        tristate "Chelsio Inline KTLS Offload"
+        depends on CHELSIO_T4
+        depends on TLS
+        depends on TLS_DEVICE
+	select CRYPTO_LIB_AES
+        help
+          This flag enables support for kernel tls offload over Chelsio T6
+          crypto accelerator. CONFIG_CHELSIO_TLS_DEVICE flag can be enabled
+          only if CONFIG_TLS and CONFIG_TLS_DEVICE flags are enabled.
+
+	  To compile this driver as a module, choose M here: the module
+	  will be called ch_ktls.
+
+endif # CHELSIO_INLINE_CRYPTO
diff --git a/drivers/net/ethernet/chelsio/inline_crypto/Makefile b/drivers/net/ethernet/chelsio/inline_crypto/Makefile
new file mode 100644
index 000000000000..27e6d7e2f1eb
--- /dev/null
+++ b/drivers/net/ethernet/chelsio/inline_crypto/Makefile
@@ -0,0 +1,4 @@
+# SPDX-License-Identifier: GPL-2.0-only
+obj-$(CONFIG_CRYPTO_DEV_CHELSIO_TLS) += chtls/
+obj-$(CONFIG_CHELSIO_IPSEC_INLINE) += ch_ipsec/
+obj-$(CONFIG_CHELSIO_TLS_DEVICE) += ch_ktls/
diff --git a/drivers/net/ethernet/chelsio/inline_crypto/ch_ipsec/Makefile b/drivers/net/ethernet/chelsio/inline_crypto/ch_ipsec/Makefile
new file mode 100644
index 000000000000..efdcaaebc455
--- /dev/null
+++ b/drivers/net/ethernet/chelsio/inline_crypto/ch_ipsec/Makefile
@@ -0,0 +1,8 @@
+# SPDX-License-Identifier: GPL-2.0-only
+ccflags-y := -I $(srctree)/drivers/net/ethernet/chelsio/cxgb4 \
+             -I $(srctree)/drivers/crypto/chelsio
+
+obj-$(CONFIG_CHELSIO_IPSEC_INLINE) += ch_ipsec.o
+ch_ipsec-objs := chcr_ipsec.o
+
+
diff --git a/drivers/net/ethernet/chelsio/inline_crypto/ch_ipsec/chcr_ipsec.c b/drivers/net/ethernet/chelsio/inline_crypto/ch_ipsec/chcr_ipsec.c
new file mode 100644
index 000000000000..ecd9a0bd5e18
--- /dev/null
+++ b/drivers/net/ethernet/chelsio/inline_crypto/ch_ipsec/chcr_ipsec.c
@@ -0,0 +1,817 @@
+/*
+ * This file is part of the Chelsio T6 Crypto driver for Linux.
+ *
+ * Copyright (c) 2003-2017 Chelsio Communications, Inc. All rights reserved.
+ *
+ * This software is available to you under a choice of one of two
+ * licenses.  You may choose to be licensed under the terms of the GNU
+ * General Public License (GPL) Version 2, available from the file
+ * COPYING in the main directory of this source tree, or the
+ * OpenIB.org BSD license below:
+ *
+ *     Redistribution and use in source and binary forms, with or
+ *     without modification, are permitted provided that the following
+ *     conditions are met:
+ *
+ *      - Redistributions of source code must retain the above
+ *        copyright notice, this list of conditions and the following
+ *        disclaimer.
+ *
+ *      - Redistributions in binary form must reproduce the above
+ *        copyright notice, this list of conditions and the following
+ *        disclaimer in the documentation and/or other materials
+ *        provided with the distribution.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND,
+ * EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF
+ * MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND
+ * NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS
+ * BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN
+ * ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN
+ * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
+ * SOFTWARE.
+ *
+ * Written and Maintained by:
+ *	Atul Gupta (atul.gupta@chelsio.com)
+ */
+
+#define pr_fmt(fmt) "ch_ipsec: " fmt
+
+#include <linux/kernel.h>
+#include <linux/module.h>
+#include <linux/skbuff.h>
+#include <linux/rtnetlink.h>
+#include <linux/highmem.h>
+#include <linux/if_vlan.h>
+#include <linux/ip.h>
+#include <linux/netdevice.h>
+#include <net/esp.h>
+#include <net/xfrm.h>
+#include <crypto/aes.h>
+#include <crypto/hash.h>
+#include <crypto/sha1.h>
+#include <crypto/sha2.h>
+#include <crypto/authenc.h>
+#include <crypto/internal/aead.h>
+#include <crypto/null.h>
+#include <crypto/internal/skcipher.h>
+#include <crypto/aead.h>
+#include <crypto/scatterwalk.h>
+#include <crypto/internal/hash.h>
+
+#include "chcr_ipsec.h"
+
+/*
+ * Max Tx descriptor space we allow for an Ethernet packet to be inlined
+ * into a WR.
+ */
+#define MAX_IMM_TX_PKT_LEN 256
+#define GCM_ESP_IV_SIZE     8
+
+static LIST_HEAD(uld_ctx_list);
+static DEFINE_MUTEX(dev_mutex);
+
+static int ch_ipsec_uld_state_change(void *handle, enum cxgb4_state new_state);
+static int ch_ipsec_xmit(struct sk_buff *skb, struct net_device *dev);
+static void *ch_ipsec_uld_add(const struct cxgb4_lld_info *infop);
+static void ch_ipsec_advance_esn_state(struct xfrm_state *x);
+static void ch_ipsec_xfrm_free_state(struct net_device *dev,
+				     struct xfrm_state *x);
+static void ch_ipsec_xfrm_del_state(struct net_device *dev,
+				    struct xfrm_state *x);
+static int ch_ipsec_xfrm_add_state(struct net_device *dev,
+				   struct xfrm_state *x,
+				   struct netlink_ext_ack *extack);
+
+static const struct xfrmdev_ops ch_ipsec_xfrmdev_ops = {
+	.xdo_dev_state_add      = ch_ipsec_xfrm_add_state,
+	.xdo_dev_state_delete   = ch_ipsec_xfrm_del_state,
+	.xdo_dev_state_free     = ch_ipsec_xfrm_free_state,
+	.xdo_dev_state_advance_esn = ch_ipsec_advance_esn_state,
+};
+
+static struct cxgb4_uld_info ch_ipsec_uld_info = {
+	.name = CHIPSEC_DRV_MODULE_NAME,
+	.add = ch_ipsec_uld_add,
+	.state_change = ch_ipsec_uld_state_change,
+	.tx_handler = ch_ipsec_xmit,
+	.xfrmdev_ops = &ch_ipsec_xfrmdev_ops,
+};
+
+static void *ch_ipsec_uld_add(const struct cxgb4_lld_info *infop)
+{
+	struct ipsec_uld_ctx *u_ctx;
+
+	pr_info_once("%s - version %s\n", CHIPSEC_DRV_DESC,
+		     CHIPSEC_DRV_VERSION);
+	u_ctx = kzalloc(sizeof(*u_ctx), GFP_KERNEL);
+	if (!u_ctx) {
+		u_ctx = ERR_PTR(-ENOMEM);
+		goto out;
+	}
+	u_ctx->lldi = *infop;
+out:
+	return u_ctx;
+}
+
+static int ch_ipsec_uld_state_change(void *handle, enum cxgb4_state new_state)
+{
+	struct ipsec_uld_ctx *u_ctx = handle;
+
+	pr_debug("new_state %u\n", new_state);
+	switch (new_state) {
+	case CXGB4_STATE_UP:
+		pr_info("%s: Up\n", pci_name(u_ctx->lldi.pdev));
+		mutex_lock(&dev_mutex);
+		list_add_tail(&u_ctx->entry, &uld_ctx_list);
+		mutex_unlock(&dev_mutex);
+		break;
+	case CXGB4_STATE_START_RECOVERY:
+	case CXGB4_STATE_DOWN:
+	case CXGB4_STATE_DETACH:
+		pr_info("%s: Down\n", pci_name(u_ctx->lldi.pdev));
+		list_del(&u_ctx->entry);
+		break;
+	default:
+		break;
+	}
+
+	return 0;
+}
+
+static int ch_ipsec_setauthsize(struct xfrm_state *x,
+				struct ipsec_sa_entry *sa_entry)
+{
+	int hmac_ctrl;
+	int authsize = x->aead->alg_icv_len / 8;
+
+	sa_entry->authsize = authsize;
+
+	switch (authsize) {
+	case ICV_8:
+		hmac_ctrl = CHCR_SCMD_HMAC_CTRL_DIV2;
+		break;
+	case ICV_12:
+		hmac_ctrl = CHCR_SCMD_HMAC_CTRL_IPSEC_96BIT;
+		break;
+	case ICV_16:
+		hmac_ctrl = CHCR_SCMD_HMAC_CTRL_NO_TRUNC;
+		break;
+	default:
+		return -EINVAL;
+	}
+	return hmac_ctrl;
+}
+
+static int ch_ipsec_setkey(struct xfrm_state *x,
+			   struct ipsec_sa_entry *sa_entry)
+{
+	int keylen = (x->aead->alg_key_len + 7) / 8;
+	unsigned char *key = x->aead->alg_key;
+	int ck_size, key_ctx_size = 0;
+	unsigned char ghash_h[AEAD_H_SIZE];
+	struct crypto_aes_ctx aes;
+	int ret = 0;
+
+	if (keylen > 3) {
+		keylen -= 4;  /* nonce/salt is present in the last 4 bytes */
+		memcpy(sa_entry->salt, key + keylen, 4);
+	}
+
+	if (keylen == AES_KEYSIZE_128) {
+		ck_size = CHCR_KEYCTX_CIPHER_KEY_SIZE_128;
+	} else if (keylen == AES_KEYSIZE_192) {
+		ck_size = CHCR_KEYCTX_CIPHER_KEY_SIZE_192;
+	} else if (keylen == AES_KEYSIZE_256) {
+		ck_size = CHCR_KEYCTX_CIPHER_KEY_SIZE_256;
+	} else {
+		pr_err("GCM: Invalid key length %d\n", keylen);
+		ret = -EINVAL;
+		goto out;
+	}
+
+	memcpy(sa_entry->key, key, keylen);
+	sa_entry->enckey_len = keylen;
+	key_ctx_size = sizeof(struct _key_ctx) +
+			      ((DIV_ROUND_UP(keylen, 16)) << 4) +
+			      AEAD_H_SIZE;
+
+	sa_entry->key_ctx_hdr = FILL_KEY_CTX_HDR(ck_size,
+						 CHCR_KEYCTX_MAC_KEY_SIZE_128,
+						 0, 0,
+						 key_ctx_size >> 4);
+
+	/* Calculate the H = CIPH(K, 0 repeated 16 times).
+	 * It will go in key context
+	 */
+	ret = aes_expandkey(&aes, key, keylen);
+	if (ret) {
+		sa_entry->enckey_len = 0;
+		goto out;
+	}
+	memset(ghash_h, 0, AEAD_H_SIZE);
+	aes_encrypt(&aes, ghash_h, ghash_h);
+	memzero_explicit(&aes, sizeof(aes));
+
+	memcpy(sa_entry->key + (DIV_ROUND_UP(sa_entry->enckey_len, 16) *
+	       16), ghash_h, AEAD_H_SIZE);
+	sa_entry->kctx_len = ((DIV_ROUND_UP(sa_entry->enckey_len, 16)) << 4) +
+			      AEAD_H_SIZE;
+out:
+	return ret;
+}
+
+/*
+ * ch_ipsec_xfrm_add_state
+ * returns 0 on success, negative error if failed to send message to FPGA
+ * positive error if FPGA returned a bad response
+ */
+static int ch_ipsec_xfrm_add_state(struct net_device *dev,
+				   struct xfrm_state *x,
+				   struct netlink_ext_ack *extack)
+{
+	struct ipsec_sa_entry *sa_entry;
+	int res = 0;
+
+	if (x->props.aalgo != SADB_AALG_NONE) {
+		NL_SET_ERR_MSG_MOD(extack, "Cannot offload authenticated xfrm states");
+		return -EINVAL;
+	}
+	if (x->props.calgo != SADB_X_CALG_NONE) {
+		NL_SET_ERR_MSG_MOD(extack, "Cannot offload compressed xfrm states");
+		return -EINVAL;
+	}
+	if (x->props.family != AF_INET &&
+	    x->props.family != AF_INET6) {
+		NL_SET_ERR_MSG_MOD(extack, "Only IPv4/6 xfrm state offloaded");
+		return -EINVAL;
+	}
+	if (x->props.mode != XFRM_MODE_TRANSPORT &&
+	    x->props.mode != XFRM_MODE_TUNNEL) {
+		NL_SET_ERR_MSG_MOD(extack, "Only transport and tunnel xfrm offload");
+		return -EINVAL;
+	}
+	if (x->id.proto != IPPROTO_ESP) {
+		NL_SET_ERR_MSG_MOD(extack, "Only ESP xfrm state offloaded");
+		return -EINVAL;
+	}
+	if (x->encap) {
+		NL_SET_ERR_MSG_MOD(extack, "Encapsulated xfrm state not offloaded");
+		return -EINVAL;
+	}
+	if (!x->aead) {
+		NL_SET_ERR_MSG_MOD(extack, "Cannot offload xfrm states without aead");
+		return -EINVAL;
+	}
+	if (x->aead->alg_icv_len != 128 &&
+	    x->aead->alg_icv_len != 96) {
+		NL_SET_ERR_MSG_MOD(extack, "Cannot offload xfrm states with AEAD ICV length other than 96b & 128b");
+		return -EINVAL;
+	}
+	if ((x->aead->alg_key_len != 128 + 32) &&
+	    (x->aead->alg_key_len != 256 + 32)) {
+		NL_SET_ERR_MSG_MOD(extack, "cannot offload xfrm states with AEAD key length other than 128/256 bit");
+		return -EINVAL;
+	}
+	if (x->tfcpad) {
+		NL_SET_ERR_MSG_MOD(extack, "Cannot offload xfrm states with tfc padding");
+		return -EINVAL;
+	}
+	if (!x->geniv) {
+		NL_SET_ERR_MSG_MOD(extack, "Cannot offload xfrm states without geniv");
+		return -EINVAL;
+	}
+	if (strcmp(x->geniv, "seqiv")) {
+		NL_SET_ERR_MSG_MOD(extack, "Cannot offload xfrm states with geniv other than seqiv");
+		return -EINVAL;
+	}
+	if (x->xso.type != XFRM_DEV_OFFLOAD_CRYPTO) {
+		NL_SET_ERR_MSG_MOD(extack, "Unsupported xfrm offload");
+		return -EINVAL;
+	}
+
+	sa_entry = kzalloc(sizeof(*sa_entry), GFP_KERNEL);
+	if (!sa_entry) {
+		res = -ENOMEM;
+		goto out;
+	}
+
+	sa_entry->hmac_ctrl = ch_ipsec_setauthsize(x, sa_entry);
+	if (x->props.flags & XFRM_STATE_ESN)
+		sa_entry->esn = 1;
+	ch_ipsec_setkey(x, sa_entry);
+	x->xso.offload_handle = (unsigned long)sa_entry;
+	try_module_get(THIS_MODULE);
+out:
+	return res;
+}
+
+static void ch_ipsec_xfrm_del_state(struct net_device *dev,
+				    struct xfrm_state *x)
+{
+	/* do nothing */
+	if (!x->xso.offload_handle)
+		return;
+}
+
+static void ch_ipsec_xfrm_free_state(struct net_device *dev,
+				     struct xfrm_state *x)
+{
+	struct ipsec_sa_entry *sa_entry;
+
+	if (!x->xso.offload_handle)
+		return;
+
+	sa_entry = (struct ipsec_sa_entry *)x->xso.offload_handle;
+	kfree(sa_entry);
+	module_put(THIS_MODULE);
+}
+
+static void ch_ipsec_advance_esn_state(struct xfrm_state *x)
+{
+	/* do nothing */
+	if (!x->xso.offload_handle)
+		return;
+}
+
+static int is_eth_imm(const struct sk_buff *skb,
+		      struct ipsec_sa_entry *sa_entry)
+{
+	unsigned int kctx_len;
+	int hdrlen;
+
+	kctx_len = sa_entry->kctx_len;
+	hdrlen = sizeof(struct fw_ulptx_wr) +
+		 sizeof(struct chcr_ipsec_req) + kctx_len;
+
+	hdrlen += sizeof(struct cpl_tx_pkt);
+	if (sa_entry->esn)
+		hdrlen += (DIV_ROUND_UP(sizeof(struct chcr_ipsec_aadiv), 16)
+			   << 4);
+	if (skb->len <= MAX_IMM_TX_PKT_LEN - hdrlen)
+		return hdrlen;
+	return 0;
+}
+
+static unsigned int calc_tx_sec_flits(const struct sk_buff *skb,
+				      struct ipsec_sa_entry *sa_entry,
+				      bool *immediate)
+{
+	unsigned int kctx_len;
+	unsigned int flits;
+	int aadivlen;
+	int hdrlen;
+
+	kctx_len = sa_entry->kctx_len;
+	hdrlen = is_eth_imm(skb, sa_entry);
+	aadivlen = sa_entry->esn ? DIV_ROUND_UP(sizeof(struct chcr_ipsec_aadiv),
+						16) : 0;
+	aadivlen <<= 4;
+
+	/* If the skb is small enough, we can pump it out as a work request
+	 * with only immediate data.  In that case we just have to have the
+	 * TX Packet header plus the skb data in the Work Request.
+	 */
+
+	if (hdrlen) {
+		*immediate = true;
+		return DIV_ROUND_UP(skb->len + hdrlen, sizeof(__be64));
+	}
+
+	flits = sgl_len(skb_shinfo(skb)->nr_frags + 1);
+
+	/* Otherwise, we're going to have to construct a Scatter gather list
+	 * of the skb body and fragments.  We also include the flits necessary
+	 * for the TX Packet Work Request and CPL.  We always have a firmware
+	 * Write Header (incorporated as part of the cpl_tx_pkt_lso and
+	 * cpl_tx_pkt structures), followed by either a TX Packet Write CPL
+	 * message or, if we're doing a Large Send Offload, an LSO CPL message
+	 * with an embedded TX Packet Write CPL message.
+	 */
+	flits += (sizeof(struct fw_ulptx_wr) +
+		  sizeof(struct chcr_ipsec_req) +
+		  kctx_len +
+		  sizeof(struct cpl_tx_pkt_core) +
+		  aadivlen) / sizeof(__be64);
+	return flits;
+}
+
+static void *copy_esn_pktxt(struct sk_buff *skb,
+			    struct net_device *dev,
+			    void *pos,
+			    struct ipsec_sa_entry *sa_entry)
+{
+	struct chcr_ipsec_aadiv *aadiv;
+	struct ulptx_idata *sc_imm;
+	struct ip_esp_hdr *esphdr;
+	struct xfrm_offload *xo;
+	struct sge_eth_txq *q;
+	struct adapter *adap;
+	struct port_info *pi;
+	__be64 seqno;
+	u32 qidx;
+	u32 seqlo;
+	u8 *iv;
+	int eoq;
+	int len;
+
+	pi = netdev_priv(dev);
+	adap = pi->adapter;
+	qidx = skb->queue_mapping;
+	q = &adap->sge.ethtxq[qidx + pi->first_qset];
+
+	/* end of queue, reset pos to start of queue */
+	eoq = (void *)q->q.stat - pos;
+	if (!eoq)
+		pos = q->q.desc;
+
+	len = DIV_ROUND_UP(sizeof(struct chcr_ipsec_aadiv), 16) << 4;
+	memset(pos, 0, len);
+	aadiv = (struct chcr_ipsec_aadiv *)pos;
+	esphdr = (struct ip_esp_hdr *)skb_transport_header(skb);
+	iv = skb_transport_header(skb) + sizeof(struct ip_esp_hdr);
+	xo = xfrm_offload(skb);
+
+	aadiv->spi = (esphdr->spi);
+	seqlo = ntohl(esphdr->seq_no);
+	seqno = cpu_to_be64(seqlo + ((u64)xo->seq.hi << 32));
+	memcpy(aadiv->seq_no, &seqno, 8);
+	iv = skb_transport_header(skb) + sizeof(struct ip_esp_hdr);
+	memcpy(aadiv->iv, iv, 8);
+
+	if (is_eth_imm(skb, sa_entry) && !skb_is_nonlinear(skb)) {
+		sc_imm = (struct ulptx_idata *)(pos +
+			  (DIV_ROUND_UP(sizeof(struct chcr_ipsec_aadiv),
+					sizeof(__be64)) << 3));
+		sc_imm->cmd_more = FILL_CMD_MORE(0);
+		sc_imm->len = cpu_to_be32(skb->len);
+	}
+	pos += len;
+	return pos;
+}
+
+static void *copy_cpltx_pktxt(struct sk_buff *skb,
+			      struct net_device *dev,
+			      void *pos,
+			      struct ipsec_sa_entry *sa_entry)
+{
+	struct cpl_tx_pkt_core *cpl;
+	struct sge_eth_txq *q;
+	struct adapter *adap;
+	struct port_info *pi;
+	u32 ctrl0, qidx;
+	u64 cntrl = 0;
+	int left;
+
+	pi = netdev_priv(dev);
+	adap = pi->adapter;
+	qidx = skb->queue_mapping;
+	q = &adap->sge.ethtxq[qidx + pi->first_qset];
+
+	left = (void *)q->q.stat - pos;
+	if (!left)
+		pos = q->q.desc;
+
+	cpl = (struct cpl_tx_pkt_core *)pos;
+
+	cntrl = TXPKT_L4CSUM_DIS_F | TXPKT_IPCSUM_DIS_F;
+	ctrl0 = TXPKT_OPCODE_V(CPL_TX_PKT_XT) | TXPKT_INTF_V(pi->tx_chan) |
+			       TXPKT_PF_V(adap->pf);
+	if (skb_vlan_tag_present(skb)) {
+		q->vlan_ins++;
+		cntrl |= TXPKT_VLAN_VLD_F | TXPKT_VLAN_V(skb_vlan_tag_get(skb));
+	}
+
+	cpl->ctrl0 = htonl(ctrl0);
+	cpl->pack = htons(0);
+	cpl->len = htons(skb->len);
+	cpl->ctrl1 = cpu_to_be64(cntrl);
+
+	pos += sizeof(struct cpl_tx_pkt_core);
+	/* Copy ESN info for HW */
+	if (sa_entry->esn)
+		pos = copy_esn_pktxt(skb, dev, pos, sa_entry);
+	return pos;
+}
+
+static void *copy_key_cpltx_pktxt(struct sk_buff *skb,
+				  struct net_device *dev,
+				  void *pos,
+				  struct ipsec_sa_entry *sa_entry)
+{
+	struct _key_ctx *key_ctx;
+	int left, eoq, key_len;
+	struct sge_eth_txq *q;
+	struct adapter *adap;
+	struct port_info *pi;
+	unsigned int qidx;
+
+	pi = netdev_priv(dev);
+	adap = pi->adapter;
+	qidx = skb->queue_mapping;
+	q = &adap->sge.ethtxq[qidx + pi->first_qset];
+	key_len = sa_entry->kctx_len;
+
+	/* end of queue, reset pos to start of queue */
+	eoq = (void *)q->q.stat - pos;
+	left = eoq;
+	if (!eoq) {
+		pos = q->q.desc;
+		left = 64 * q->q.size;
+	}
+
+	/* Copy the Key context header */
+	key_ctx = (struct _key_ctx *)pos;
+	key_ctx->ctx_hdr = sa_entry->key_ctx_hdr;
+	memcpy(key_ctx->salt, sa_entry->salt, MAX_SALT);
+	pos += sizeof(struct _key_ctx);
+	left -= sizeof(struct _key_ctx);
+
+	if (likely(key_len <= left)) {
+		memcpy(key_ctx->key, sa_entry->key, key_len);
+		pos += key_len;
+	} else {
+		memcpy(pos, sa_entry->key, left);
+		memcpy(q->q.desc, sa_entry->key + left,
+		       key_len - left);
+		pos = (u8 *)q->q.desc + (key_len - left);
+	}
+	/* Copy CPL TX PKT XT */
+	pos = copy_cpltx_pktxt(skb, dev, pos, sa_entry);
+
+	return pos;
+}
+
+static void *ch_ipsec_crypto_wreq(struct sk_buff *skb,
+				  struct net_device *dev,
+				  void *pos,
+				  int credits,
+				  struct ipsec_sa_entry *sa_entry)
+{
+	struct port_info *pi = netdev_priv(dev);
+	struct adapter *adap = pi->adapter;
+	unsigned int ivsize = GCM_ESP_IV_SIZE;
+	struct chcr_ipsec_wr *wr;
+	bool immediate = false;
+	u16 immdatalen = 0;
+	unsigned int flits;
+	u32 ivinoffset;
+	u32 aadstart;
+	u32 aadstop;
+	u32 ciphstart;
+	u16 sc_more = 0;
+	u32 ivdrop = 0;
+	u32 esnlen = 0;
+	u32 wr_mid;
+	u16 ndesc;
+	int qidx = skb_get_queue_mapping(skb);
+	struct sge_eth_txq *q = &adap->sge.ethtxq[qidx + pi->first_qset];
+	unsigned int kctx_len = sa_entry->kctx_len;
+	int qid = q->q.cntxt_id;
+
+	atomic_inc(&adap->ch_ipsec_stats.ipsec_cnt);
+
+	flits = calc_tx_sec_flits(skb, sa_entry, &immediate);
+	ndesc = DIV_ROUND_UP(flits, 2);
+	if (sa_entry->esn)
+		ivdrop = 1;
+
+	if (immediate)
+		immdatalen = skb->len;
+
+	if (sa_entry->esn) {
+		esnlen = sizeof(struct chcr_ipsec_aadiv);
+		if (!skb_is_nonlinear(skb))
+			sc_more  = 1;
+	}
+
+	/* WR Header */
+	wr = (struct chcr_ipsec_wr *)pos;
+	wr->wreq.op_to_compl = htonl(FW_WR_OP_V(FW_ULPTX_WR));
+	wr_mid = FW_CRYPTO_LOOKASIDE_WR_LEN16_V(ndesc);
+
+	if (unlikely(credits < ETHTXQ_STOP_THRES)) {
+		netif_tx_stop_queue(q->txq);
+		q->q.stops++;
+		if (!q->dbqt)
+			wr_mid |= FW_WR_EQUEQ_F | FW_WR_EQUIQ_F;
+	}
+	wr_mid |= FW_ULPTX_WR_DATA_F;
+	wr->wreq.flowid_len16 = htonl(wr_mid);
+
+	/* ULPTX */
+	wr->req.ulptx.cmd_dest = FILL_ULPTX_CMD_DEST(pi->port_id, qid);
+	wr->req.ulptx.len = htonl(ndesc - 1);
+
+	/* Sub-command */
+	wr->req.sc_imm.cmd_more = FILL_CMD_MORE(!immdatalen || sc_more);
+	wr->req.sc_imm.len = cpu_to_be32(sizeof(struct cpl_tx_sec_pdu) +
+					 sizeof(wr->req.key_ctx) +
+					 kctx_len +
+					 sizeof(struct cpl_tx_pkt_core) +
+					 esnlen +
+					 (esnlen ? 0 : immdatalen));
+
+	/* CPL_SEC_PDU */
+	ivinoffset = sa_entry->esn ? (ESN_IV_INSERT_OFFSET + 1) :
+				     (skb_transport_offset(skb) +
+				      sizeof(struct ip_esp_hdr) + 1);
+	wr->req.sec_cpl.op_ivinsrtofst = htonl(
+				CPL_TX_SEC_PDU_OPCODE_V(CPL_TX_SEC_PDU) |
+				CPL_TX_SEC_PDU_CPLLEN_V(2) |
+				CPL_TX_SEC_PDU_PLACEHOLDER_V(1) |
+				CPL_TX_SEC_PDU_IVINSRTOFST_V(
+							     ivinoffset));
+
+	wr->req.sec_cpl.pldlen = htonl(skb->len + esnlen);
+	aadstart = sa_entry->esn ? 1 : (skb_transport_offset(skb) + 1);
+	aadstop = sa_entry->esn ? ESN_IV_INSERT_OFFSET :
+				  (skb_transport_offset(skb) +
+				   sizeof(struct ip_esp_hdr));
+	ciphstart = skb_transport_offset(skb) + sizeof(struct ip_esp_hdr) +
+		    GCM_ESP_IV_SIZE + 1;
+	ciphstart += sa_entry->esn ?  esnlen : 0;
+
+	wr->req.sec_cpl.aadstart_cipherstop_hi = FILL_SEC_CPL_CIPHERSTOP_HI(
+							aadstart,
+							aadstop,
+							ciphstart, 0);
+
+	wr->req.sec_cpl.cipherstop_lo_authinsert =
+		FILL_SEC_CPL_AUTHINSERT(0, ciphstart,
+					sa_entry->authsize,
+					 sa_entry->authsize);
+	wr->req.sec_cpl.seqno_numivs =
+		FILL_SEC_CPL_SCMD0_SEQNO(CHCR_ENCRYPT_OP, 1,
+					 CHCR_SCMD_CIPHER_MODE_AES_GCM,
+					 CHCR_SCMD_AUTH_MODE_GHASH,
+					 sa_entry->hmac_ctrl,
+					 ivsize >> 1);
+	wr->req.sec_cpl.ivgen_hdrlen =  FILL_SEC_CPL_IVGEN_HDRLEN(0, 0, 1,
+								  0, ivdrop, 0);
+
+	pos += sizeof(struct fw_ulptx_wr) +
+	       sizeof(struct ulp_txpkt) +
+	       sizeof(struct ulptx_idata) +
+	       sizeof(struct cpl_tx_sec_pdu);
+
+	pos = copy_key_cpltx_pktxt(skb, dev, pos, sa_entry);
+
+	return pos;
+}
+
+/**
+ *      flits_to_desc - returns the num of Tx descriptors for the given flits
+ *      @n: the number of flits
+ *
+ *      Returns the number of Tx descriptors needed for the supplied number
+ *      of flits.
+ */
+static unsigned int flits_to_desc(unsigned int n)
+{
+	WARN_ON(n > SGE_MAX_WR_LEN / 8);
+	return DIV_ROUND_UP(n, 8);
+}
+
+static unsigned int txq_avail(const struct sge_txq *q)
+{
+	return q->size - 1 - q->in_use;
+}
+
+static void eth_txq_stop(struct sge_eth_txq *q)
+{
+	netif_tx_stop_queue(q->txq);
+	q->q.stops++;
+}
+
+static void txq_advance(struct sge_txq *q, unsigned int n)
+{
+	q->in_use += n;
+	q->pidx += n;
+	if (q->pidx >= q->size)
+		q->pidx -= q->size;
+}
+
+/*
+ *      ch_ipsec_xmit called from ULD Tx handler
+ */
+int ch_ipsec_xmit(struct sk_buff *skb, struct net_device *dev)
+{
+	struct xfrm_state *x = xfrm_input_state(skb);
+	unsigned int last_desc, ndesc, flits = 0;
+	struct ipsec_sa_entry *sa_entry;
+	u64 *pos, *end, *before, *sgl;
+	struct tx_sw_desc *sgl_sdesc;
+	int qidx, left, credits;
+	bool immediate = false;
+	struct sge_eth_txq *q;
+	struct adapter *adap;
+	struct port_info *pi;
+	struct sec_path *sp;
+
+	if (!x->xso.offload_handle)
+		return NETDEV_TX_BUSY;
+
+	sa_entry = (struct ipsec_sa_entry *)x->xso.offload_handle;
+
+	sp = skb_sec_path(skb);
+	if (sp->len != 1) {
+out_free:       dev_kfree_skb_any(skb);
+		return NETDEV_TX_OK;
+	}
+
+	pi = netdev_priv(dev);
+	adap = pi->adapter;
+	qidx = skb->queue_mapping;
+	q = &adap->sge.ethtxq[qidx + pi->first_qset];
+
+	cxgb4_reclaim_completed_tx(adap, &q->q, true);
+
+	flits = calc_tx_sec_flits(skb, sa_entry, &immediate);
+	ndesc = flits_to_desc(flits);
+	credits = txq_avail(&q->q) - ndesc;
+
+	if (unlikely(credits < 0)) {
+		eth_txq_stop(q);
+		dev_err(adap->pdev_dev,
+			"%s: Tx ring %u full while queue awake! cred:%d %d %d flits:%d\n",
+			dev->name, qidx, credits, ndesc, txq_avail(&q->q),
+			flits);
+		return NETDEV_TX_BUSY;
+	}
+
+	last_desc = q->q.pidx + ndesc - 1;
+	if (last_desc >= q->q.size)
+		last_desc -= q->q.size;
+	sgl_sdesc = &q->q.sdesc[last_desc];
+
+	if (!immediate &&
+	    unlikely(cxgb4_map_skb(adap->pdev_dev, skb, sgl_sdesc->addr) < 0)) {
+		memset(sgl_sdesc->addr, 0, sizeof(sgl_sdesc->addr));
+		q->mapping_err++;
+		goto out_free;
+	}
+
+	pos = (u64 *)&q->q.desc[q->q.pidx];
+	before = (u64 *)pos;
+	end = (u64 *)pos + flits;
+	/* Setup IPSec CPL */
+	pos = (void *)ch_ipsec_crypto_wreq(skb, dev, (void *)pos,
+					   credits, sa_entry);
+	if (before > (u64 *)pos) {
+		left = (u8 *)end - (u8 *)q->q.stat;
+		end = (void *)q->q.desc + left;
+	}
+	if (pos == (u64 *)q->q.stat) {
+		left = (u8 *)end - (u8 *)q->q.stat;
+		end = (void *)q->q.desc + left;
+		pos = (void *)q->q.desc;
+	}
+
+	sgl = (void *)pos;
+	if (immediate) {
+		cxgb4_inline_tx_skb(skb, &q->q, sgl);
+		dev_consume_skb_any(skb);
+	} else {
+		cxgb4_write_sgl(skb, &q->q, (void *)sgl, end,
+				0, sgl_sdesc->addr);
+		skb_orphan(skb);
+		sgl_sdesc->skb = skb;
+	}
+	txq_advance(&q->q, ndesc);
+
+	cxgb4_ring_tx_db(adap, &q->q, ndesc);
+	return NETDEV_TX_OK;
+}
+
+static int __init ch_ipsec_init(void)
+{
+	cxgb4_register_uld(CXGB4_ULD_IPSEC, &ch_ipsec_uld_info);
+
+	return 0;
+}
+
+static void __exit ch_ipsec_exit(void)
+{
+	struct ipsec_uld_ctx *u_ctx, *tmp;
+	struct adapter *adap;
+
+	mutex_lock(&dev_mutex);
+	list_for_each_entry_safe(u_ctx, tmp, &uld_ctx_list, entry) {
+		adap = pci_get_drvdata(u_ctx->lldi.pdev);
+		atomic_set(&adap->ch_ipsec_stats.ipsec_cnt, 0);
+		list_del(&u_ctx->entry);
+		kfree(u_ctx);
+	}
+	mutex_unlock(&dev_mutex);
+	cxgb4_unregister_uld(CXGB4_ULD_IPSEC);
+}
+
+module_init(ch_ipsec_init);
+module_exit(ch_ipsec_exit);
+
+MODULE_DESCRIPTION("Crypto IPSEC for Chelsio Terminator cards.");
+MODULE_LICENSE("GPL");
+MODULE_AUTHOR("Chelsio Communications");
+MODULE_VERSION(CHIPSEC_DRV_VERSION);
+
diff --git a/drivers/net/ethernet/chelsio/inline_crypto/ch_ipsec/chcr_ipsec.h b/drivers/net/ethernet/chelsio/inline_crypto/ch_ipsec/chcr_ipsec.h
new file mode 100644
index 000000000000..0d42e7d15714
--- /dev/null
+++ b/drivers/net/ethernet/chelsio/inline_crypto/ch_ipsec/chcr_ipsec.h
@@ -0,0 +1,57 @@
+/* SPDX-License-Identifier: GPL-2.0-only */
+/* Copyright (c) 2018 Chelsio Communications, Inc. */
+
+#ifndef __CHCR_IPSEC_H__
+#define __CHCR_IPSEC_H__
+
+#include "t4_hw.h"
+#include "cxgb4.h"
+#include "t4_msg.h"
+#include "cxgb4_uld.h"
+
+#include "chcr_core.h"
+#include "chcr_algo.h"
+#include "chcr_crypto.h"
+
+#define CHIPSEC_DRV_MODULE_NAME "ch_ipsec"
+#define CHIPSEC_DRV_VERSION "1.0.0.0-ko"
+#define CHIPSEC_DRV_DESC "Chelsio T6 Crypto Ipsec offload Driver"
+
+struct ipsec_uld_ctx {
+	struct list_head entry;
+	struct cxgb4_lld_info lldi;
+};
+
+struct chcr_ipsec_req {
+	struct ulp_txpkt ulptx;
+	struct ulptx_idata sc_imm;
+	struct cpl_tx_sec_pdu sec_cpl;
+	struct _key_ctx key_ctx;
+};
+
+struct chcr_ipsec_wr {
+	struct fw_ulptx_wr wreq;
+	struct chcr_ipsec_req req;
+};
+
+#define ESN_IV_INSERT_OFFSET 12
+struct chcr_ipsec_aadiv {
+	__be32 spi;
+	u8 seq_no[8];
+	u8 iv[8];
+};
+
+struct ipsec_sa_entry {
+	int hmac_ctrl;
+	u16 esn;
+	u16 resv;
+	unsigned int enckey_len;
+	unsigned int kctx_len;
+	unsigned int authsize;
+	__be32 key_ctx_hdr;
+	char salt[MAX_SALT];
+	char key[2 * AES_MAX_KEY_SIZE];
+};
+
+#endif /* __CHCR_IPSEC_H__ */
+
diff --git a/drivers/net/ethernet/chelsio/inline_crypto/ch_ktls/Makefile b/drivers/net/ethernet/chelsio/inline_crypto/ch_ktls/Makefile
new file mode 100644
index 000000000000..5e7d161c3199
--- /dev/null
+++ b/drivers/net/ethernet/chelsio/inline_crypto/ch_ktls/Makefile
@@ -0,0 +1,5 @@
+# SPDX-License-Identifier: GPL-2.0-only
+ccflags-y := -I $(srctree)/drivers/net/ethernet/chelsio/cxgb4
+
+obj-$(CONFIG_CHELSIO_TLS_DEVICE) += ch_ktls.o
+ch_ktls-objs := chcr_ktls.o
diff --git a/drivers/net/ethernet/chelsio/inline_crypto/ch_ktls/chcr_common.h b/drivers/net/ethernet/chelsio/inline_crypto/ch_ktls/chcr_common.h
new file mode 100644
index 000000000000..38319f4c3121
--- /dev/null
+++ b/drivers/net/ethernet/chelsio/inline_crypto/ch_ktls/chcr_common.h
@@ -0,0 +1,111 @@
+/* SPDX-License-Identifier: GPL-2.0-only */
+/* Copyright (C) 2020 Chelsio Communications.  All rights reserved. */
+
+#ifndef __CHCR_COMMON_H__
+#define __CHCR_COMMON_H__
+
+#include "cxgb4.h"
+
+#define CHCR_MAX_SALT                      4
+#define CHCR_KEYCTX_MAC_KEY_SIZE_128       0
+#define CHCR_KEYCTX_CIPHER_KEY_SIZE_128    0
+#define CHCR_SCMD_CIPHER_MODE_AES_GCM      2
+#define CHCR_SCMD_CIPHER_MODE_AES_CTR      3
+#define CHCR_CPL_TX_SEC_PDU_LEN_64BIT      2
+#define CHCR_SCMD_SEQ_NO_CTRL_64BIT        3
+#define CHCR_SCMD_PROTO_VERSION_TLS        0
+#define CHCR_SCMD_PROTO_VERSION_GENERIC    4
+#define CHCR_SCMD_AUTH_MODE_GHASH          4
+#define AES_BLOCK_LEN                      16
+
+struct ktls_key_ctx {
+	__be32 ctx_hdr;
+	u8 salt[CHCR_MAX_SALT];
+	__be64 iv_to_auth;
+	unsigned char key[TLS_CIPHER_AES_GCM_128_KEY_SIZE +
+			  TLS_CIPHER_AES_GCM_256_TAG_SIZE];
+};
+
+/* Crypto key context */
+#define KEY_CONTEXT_CTX_LEN_S           24
+#define KEY_CONTEXT_CTX_LEN_V(x)        ((x) << KEY_CONTEXT_CTX_LEN_S)
+
+#define KEY_CONTEXT_SALT_PRESENT_S      10
+#define KEY_CONTEXT_SALT_PRESENT_V(x)   ((x) << KEY_CONTEXT_SALT_PRESENT_S)
+#define KEY_CONTEXT_SALT_PRESENT_F      KEY_CONTEXT_SALT_PRESENT_V(1U)
+
+#define KEY_CONTEXT_VALID_S     0
+#define KEY_CONTEXT_VALID_V(x)  ((x) << KEY_CONTEXT_VALID_S)
+#define KEY_CONTEXT_VALID_F     KEY_CONTEXT_VALID_V(1U)
+
+#define KEY_CONTEXT_CK_SIZE_S           6
+#define KEY_CONTEXT_CK_SIZE_V(x)        ((x) << KEY_CONTEXT_CK_SIZE_S)
+
+#define KEY_CONTEXT_MK_SIZE_S           2
+#define KEY_CONTEXT_MK_SIZE_V(x)        ((x) << KEY_CONTEXT_MK_SIZE_S)
+
+#define KEY_CONTEXT_OPAD_PRESENT_S      11
+#define KEY_CONTEXT_OPAD_PRESENT_V(x)   ((x) << KEY_CONTEXT_OPAD_PRESENT_S)
+#define KEY_CONTEXT_OPAD_PRESENT_F      KEY_CONTEXT_OPAD_PRESENT_V(1U)
+
+#define FILL_KEY_CTX_HDR(ck_size, mk_size, ctx_len) \
+		htonl(KEY_CONTEXT_MK_SIZE_V(mk_size) | \
+		      KEY_CONTEXT_CK_SIZE_V(ck_size) | \
+		      KEY_CONTEXT_VALID_F | \
+		      KEY_CONTEXT_SALT_PRESENT_F | \
+		      KEY_CONTEXT_CTX_LEN_V((ctx_len)))
+
+static inline void *chcr_copy_to_txd(const void *src, const struct sge_txq *q,
+				     void *pos, int length)
+{
+	int left = (void *)q->stat - pos;
+	u64 *p;
+
+	if (likely(length <= left)) {
+		memcpy(pos, src, length);
+		pos += length;
+	} else {
+		memcpy(pos, src, left);
+		memcpy(q->desc, src + left, length - left);
+		pos = (void *)q->desc + (length - left);
+	}
+	/* 0-pad to multiple of 16 */
+	p = PTR_ALIGN(pos, 8);
+	if ((uintptr_t)p & 8) {
+		*p = 0;
+		return p + 1;
+	}
+	return p;
+}
+
+static inline unsigned int chcr_txq_avail(const struct sge_txq *q)
+{
+	return q->size - 1 - q->in_use;
+}
+
+static inline void chcr_txq_advance(struct sge_txq *q, unsigned int n)
+{
+	q->in_use += n;
+	q->pidx += n;
+	if (q->pidx >= q->size)
+		q->pidx -= q->size;
+}
+
+static inline void chcr_eth_txq_stop(struct sge_eth_txq *q)
+{
+	netif_tx_stop_queue(q->txq);
+	q->q.stops++;
+}
+
+static inline unsigned int chcr_sgl_len(unsigned int n)
+{
+	n--;
+	return (3 * n) / 2 + (n & 1) + 2;
+}
+
+static inline unsigned int chcr_flits_to_desc(unsigned int n)
+{
+	WARN_ON(n > SGE_MAX_WR_LEN / 8);
+	return DIV_ROUND_UP(n, 8);
+}
+#endif /* __CHCR_COMMON_H__ */
diff --git a/drivers/net/ethernet/chelsio/inline_crypto/ch_ktls/chcr_ktls.c b/drivers/net/ethernet/chelsio/inline_crypto/ch_ktls/chcr_ktls.c
new file mode 100644
index 000000000000..4e2096e49684
--- /dev/null
+++ b/drivers/net/ethernet/chelsio/inline_crypto/ch_ktls/chcr_ktls.c
@@ -0,0 +1,2271 @@
+// SPDX-License-Identifier: GPL-2.0-only
+/* Copyright (C) 2020 Chelsio Communications.  All rights reserved. */
+
+#define pr_fmt(fmt) KBUILD_MODNAME ": " fmt
+
+#include <linux/skbuff.h>
+#include <linux/module.h>
+#include <linux/highmem.h>
+#include <linux/ip.h>
+#include <net/ipv6.h>
+#include <linux/netdevice.h>
+#include <crypto/aes.h>
+#include <linux/skbuff_ref.h>
+#include "chcr_ktls.h"
+
+static LIST_HEAD(uld_ctx_list);
+static DEFINE_MUTEX(dev_mutex);
+
+/* chcr_get_nfrags_to_send: get the remaining nfrags after start offset
+ * @skb: skb
+ * @start: start offset.
+ * @len: how much data to send after @start
+ */
+static int chcr_get_nfrags_to_send(struct sk_buff *skb, u32 start, u32 len)
+{
+	struct skb_shared_info *si = skb_shinfo(skb);
+	u32 frag_size, skb_linear_data_len = skb_headlen(skb);
+	u8 nfrags = 0, frag_idx = 0;
+	skb_frag_t *frag;
+
+	/* if its a linear skb then return 1 */
+	if (!skb_is_nonlinear(skb))
+		return 1;
+
+	if (unlikely(start < skb_linear_data_len)) {
+		frag_size = min(len, skb_linear_data_len - start);
+	} else {
+		start -= skb_linear_data_len;
+
+		frag = &si->frags[frag_idx];
+		frag_size = skb_frag_size(frag);
+		while (start >= frag_size) {
+			start -= frag_size;
+			frag_idx++;
+			frag = &si->frags[frag_idx];
+			frag_size = skb_frag_size(frag);
+		}
+		frag_size = min(len, skb_frag_size(frag) - start);
+	}
+	len -= frag_size;
+	nfrags++;
+
+	while (len) {
+		frag_size = min(len, skb_frag_size(&si->frags[frag_idx]));
+		len -= frag_size;
+		nfrags++;
+		frag_idx++;
+	}
+	return nfrags;
+}
+
+static int chcr_init_tcb_fields(struct chcr_ktls_info *tx_info);
+static void clear_conn_resources(struct chcr_ktls_info *tx_info);
+/*
+ * chcr_ktls_save_keys: calculate and save crypto keys.
+ * @tx_info - driver specific tls info.
+ * @crypto_info - tls crypto information.
+ * @direction - TX/RX direction.
+ * return - SUCCESS/FAILURE.
+ */
+static int chcr_ktls_save_keys(struct chcr_ktls_info *tx_info,
+			       struct tls_crypto_info *crypto_info,
+			       enum tls_offload_ctx_dir direction)
+{
+	int ck_size, key_ctx_size, mac_key_size, keylen, ghash_size, ret;
+	unsigned char ghash_h[TLS_CIPHER_AES_GCM_256_TAG_SIZE];
+	struct tls12_crypto_info_aes_gcm_128 *info_128_gcm;
+	struct ktls_key_ctx *kctx = &tx_info->key_ctx;
+	struct crypto_aes_ctx aes_ctx;
+	unsigned char *key, *salt;
+
+	switch (crypto_info->cipher_type) {
+	case TLS_CIPHER_AES_GCM_128:
+		info_128_gcm =
+			(struct tls12_crypto_info_aes_gcm_128 *)crypto_info;
+		keylen = TLS_CIPHER_AES_GCM_128_KEY_SIZE;
+		ck_size = CHCR_KEYCTX_CIPHER_KEY_SIZE_128;
+		tx_info->salt_size = TLS_CIPHER_AES_GCM_128_SALT_SIZE;
+		mac_key_size = CHCR_KEYCTX_MAC_KEY_SIZE_128;
+		tx_info->iv_size = TLS_CIPHER_AES_GCM_128_IV_SIZE;
+		tx_info->iv = be64_to_cpu(*(__be64 *)info_128_gcm->iv);
+
+		ghash_size = TLS_CIPHER_AES_GCM_128_TAG_SIZE;
+		key = info_128_gcm->key;
+		salt = info_128_gcm->salt;
+		tx_info->record_no = *(u64 *)info_128_gcm->rec_seq;
+
+		/* The SCMD fields used when encrypting a full TLS
+		 * record. Its a one time calculation till the
+		 * connection exists.
+		 */
+		tx_info->scmd0_seqno_numivs =
+			SCMD_SEQ_NO_CTRL_V(CHCR_SCMD_SEQ_NO_CTRL_64BIT) |
+			SCMD_CIPH_AUTH_SEQ_CTRL_F |
+			SCMD_PROTO_VERSION_V(CHCR_SCMD_PROTO_VERSION_TLS) |
+			SCMD_CIPH_MODE_V(CHCR_SCMD_CIPHER_MODE_AES_GCM) |
+			SCMD_AUTH_MODE_V(CHCR_SCMD_AUTH_MODE_GHASH) |
+			SCMD_IV_SIZE_V(TLS_CIPHER_AES_GCM_128_IV_SIZE >> 1) |
+			SCMD_NUM_IVS_V(1);
+
+		/* keys will be sent inline. */
+		tx_info->scmd0_ivgen_hdrlen = SCMD_KEY_CTX_INLINE_F;
+
+		/* The SCMD fields used when encrypting a partial TLS
+		 * record (no trailer and possibly a truncated payload).
+		 */
+		tx_info->scmd0_short_seqno_numivs =
+			SCMD_CIPH_AUTH_SEQ_CTRL_F |
+			SCMD_PROTO_VERSION_V(CHCR_SCMD_PROTO_VERSION_GENERIC) |
+			SCMD_CIPH_MODE_V(CHCR_SCMD_CIPHER_MODE_AES_CTR) |
+			SCMD_IV_SIZE_V(AES_BLOCK_LEN >> 1);
+
+		tx_info->scmd0_short_ivgen_hdrlen =
+			tx_info->scmd0_ivgen_hdrlen | SCMD_AADIVDROP_F;
+
+		break;
+
+	default:
+		pr_err("GCM: cipher type 0x%x not supported\n",
+		       crypto_info->cipher_type);
+		ret = -EINVAL;
+		goto out;
+	}
+
+	key_ctx_size = CHCR_KTLS_KEY_CTX_LEN +
+		       roundup(keylen, 16) + ghash_size;
+	/* Calculate the H = CIPH(K, 0 repeated 16 times).
+	 * It will go in key context
+	 */
+
+	ret = aes_expandkey(&aes_ctx, key, keylen);
+	if (ret)
+		goto out;
+
+	memset(ghash_h, 0, ghash_size);
+	aes_encrypt(&aes_ctx, ghash_h, ghash_h);
+	memzero_explicit(&aes_ctx, sizeof(aes_ctx));
+
+	/* fill the Key context */
+	if (direction == TLS_OFFLOAD_CTX_DIR_TX) {
+		kctx->ctx_hdr = FILL_KEY_CTX_HDR(ck_size,
+						 mac_key_size,
+						 key_ctx_size >> 4);
+	} else {
+		ret = -EINVAL;
+		goto out;
+	}
+
+	memcpy(kctx->salt, salt, tx_info->salt_size);
+	memcpy(kctx->key, key, keylen);
+	memcpy(kctx->key + keylen, ghash_h, ghash_size);
+	tx_info->key_ctx_len = key_ctx_size;
+
+out:
+	return ret;
+}
+
+/*
+ * chcr_ktls_act_open_req: creates TCB entry for ipv4 connection.
+ * @sk - tcp socket.
+ * @tx_info - driver specific tls info.
+ * @atid - connection active tid.
+ * return - send success/failure.
+ */
+static int chcr_ktls_act_open_req(struct sock *sk,
+				  struct chcr_ktls_info *tx_info,
+				  int atid)
+{
+	struct inet_sock *inet = inet_sk(sk);
+	struct cpl_t6_act_open_req *cpl6;
+	struct cpl_act_open_req *cpl;
+	struct sk_buff *skb;
+	unsigned int len;
+	int qid_atid;
+	u64 options;
+
+	len = sizeof(*cpl6);
+	skb = alloc_skb(len, GFP_KERNEL);
+	if (unlikely(!skb))
+		return -ENOMEM;
+	/* mark it a control pkt */
+	set_wr_txq(skb, CPL_PRIORITY_CONTROL, tx_info->port_id);
+
+	cpl6 = __skb_put_zero(skb, len);
+	cpl = (struct cpl_act_open_req *)cpl6;
+	INIT_TP_WR(cpl6, 0);
+	qid_atid = TID_QID_V(tx_info->rx_qid) |
+		   TID_TID_V(atid);
+	OPCODE_TID(cpl) = htonl(MK_OPCODE_TID(CPL_ACT_OPEN_REQ, qid_atid));
+	cpl->local_port = inet->inet_sport;
+	cpl->peer_port = inet->inet_dport;
+	cpl->local_ip = inet->inet_rcv_saddr;
+	cpl->peer_ip = inet->inet_daddr;
+
+	/* fill first 64 bit option field. */
+	options = TCAM_BYPASS_F | ULP_MODE_V(ULP_MODE_NONE) | NON_OFFLOAD_F |
+		  SMAC_SEL_V(tx_info->smt_idx) | TX_CHAN_V(tx_info->tx_chan);
+	cpl->opt0 = cpu_to_be64(options);
+
+	/* next 64 bit option field. */
+	options =
+		TX_QUEUE_V(tx_info->adap->params.tp.tx_modq[tx_info->tx_chan]);
+	cpl->opt2 = htonl(options);
+
+	return cxgb4_l2t_send(tx_info->netdev, skb, tx_info->l2te);
+}
+
+#if IS_ENABLED(CONFIG_IPV6)
+/*
+ * chcr_ktls_act_open_req6: creates TCB entry for ipv6 connection.
+ * @sk - tcp socket.
+ * @tx_info - driver specific tls info.
+ * @atid - connection active tid.
+ * return - send success/failure.
+ */
+static int chcr_ktls_act_open_req6(struct sock *sk,
+				   struct chcr_ktls_info *tx_info,
+				   int atid)
+{
+	struct inet_sock *inet = inet_sk(sk);
+	struct cpl_t6_act_open_req6 *cpl6;
+	struct cpl_act_open_req6 *cpl;
+	struct sk_buff *skb;
+	unsigned int len;
+	int qid_atid;
+	u64 options;
+
+	len = sizeof(*cpl6);
+	skb = alloc_skb(len, GFP_KERNEL);
+	if (unlikely(!skb))
+		return -ENOMEM;
+	/* mark it a control pkt */
+	set_wr_txq(skb, CPL_PRIORITY_CONTROL, tx_info->port_id);
+
+	cpl6 = __skb_put_zero(skb, len);
+	cpl = (struct cpl_act_open_req6 *)cpl6;
+	INIT_TP_WR(cpl6, 0);
+	qid_atid = TID_QID_V(tx_info->rx_qid) | TID_TID_V(atid);
+	OPCODE_TID(cpl) = htonl(MK_OPCODE_TID(CPL_ACT_OPEN_REQ6, qid_atid));
+	cpl->local_port = inet->inet_sport;
+	cpl->peer_port = inet->inet_dport;
+	cpl->local_ip_hi = *(__be64 *)&sk->sk_v6_rcv_saddr.in6_u.u6_addr8[0];
+	cpl->local_ip_lo = *(__be64 *)&sk->sk_v6_rcv_saddr.in6_u.u6_addr8[8];
+	cpl->peer_ip_hi = *(__be64 *)&sk->sk_v6_daddr.in6_u.u6_addr8[0];
+	cpl->peer_ip_lo = *(__be64 *)&sk->sk_v6_daddr.in6_u.u6_addr8[8];
+
+	/* first 64 bit option field. */
+	options = TCAM_BYPASS_F | ULP_MODE_V(ULP_MODE_NONE) | NON_OFFLOAD_F |
+		  SMAC_SEL_V(tx_info->smt_idx) | TX_CHAN_V(tx_info->tx_chan);
+	cpl->opt0 = cpu_to_be64(options);
+	/* next 64 bit option field. */
+	options =
+		TX_QUEUE_V(tx_info->adap->params.tp.tx_modq[tx_info->tx_chan]);
+	cpl->opt2 = htonl(options);
+
+	return cxgb4_l2t_send(tx_info->netdev, skb, tx_info->l2te);
+}
+#endif /* #if IS_ENABLED(CONFIG_IPV6) */
+
+/*
+ * chcr_setup_connection:  create a TCB entry so that TP will form tcp packets.
+ * @sk - tcp socket.
+ * @tx_info - driver specific tls info.
+ * return: NET_TX_OK/NET_XMIT_DROP
+ */
+static int chcr_setup_connection(struct sock *sk,
+				 struct chcr_ktls_info *tx_info)
+{
+	struct tid_info *t = &tx_info->adap->tids;
+	int atid, ret = 0;
+
+	atid = cxgb4_alloc_atid(t, tx_info);
+	if (atid == -1)
+		return -EINVAL;
+
+	tx_info->atid = atid;
+
+	if (tx_info->ip_family == AF_INET) {
+		ret = chcr_ktls_act_open_req(sk, tx_info, atid);
+#if IS_ENABLED(CONFIG_IPV6)
+	} else {
+		ret = cxgb4_clip_get(tx_info->netdev, (const u32 *)
+				     &sk->sk_v6_rcv_saddr,
+				     1);
+		if (ret)
+			return ret;
+		ret = chcr_ktls_act_open_req6(sk, tx_info, atid);
+#endif
+	}
+
+	/* if return type is NET_XMIT_CN, msg will be sent but delayed, mark ret
+	 * success, if any other return type clear atid and return that failure.
+	 */
+	if (ret) {
+		if (ret == NET_XMIT_CN) {
+			ret = 0;
+		} else {
+#if IS_ENABLED(CONFIG_IPV6)
+			/* clear clip entry */
+			if (tx_info->ip_family == AF_INET6)
+				cxgb4_clip_release(tx_info->netdev,
+						   (const u32 *)
+						   &sk->sk_v6_rcv_saddr,
+						   1);
+#endif
+			cxgb4_free_atid(t, atid);
+		}
+	}
+
+	return ret;
+}
+
+/*
+ * chcr_set_tcb_field: update tcb fields.
+ * @tx_info - driver specific tls info.
+ * @word - TCB word.
+ * @mask - TCB word related mask.
+ * @val - TCB word related value.
+ * @no_reply - set 1 if not looking for TP response.
+ */
+static int chcr_set_tcb_field(struct chcr_ktls_info *tx_info, u16 word,
+			      u64 mask, u64 val, int no_reply)
+{
+	struct cpl_set_tcb_field *req;
+	struct sk_buff *skb;
+
+	skb = alloc_skb(sizeof(struct cpl_set_tcb_field), GFP_ATOMIC);
+	if (!skb)
+		return -ENOMEM;
+
+	req = (struct cpl_set_tcb_field *)__skb_put_zero(skb, sizeof(*req));
+	INIT_TP_WR_CPL(req, CPL_SET_TCB_FIELD, tx_info->tid);
+	req->reply_ctrl = htons(QUEUENO_V(tx_info->rx_qid) |
+				NO_REPLY_V(no_reply));
+	req->word_cookie = htons(TCB_WORD_V(word));
+	req->mask = cpu_to_be64(mask);
+	req->val = cpu_to_be64(val);
+
+	set_wr_txq(skb, CPL_PRIORITY_CONTROL, tx_info->port_id);
+	return cxgb4_ofld_send(tx_info->netdev, skb);
+}
+
+/*
+ * chcr_ktls_dev_del:  call back for tls_dev_del.
+ * Remove the tid and l2t entry and close the connection.
+ * it per connection basis.
+ * @netdev - net device.
+ * @tls_cts - tls context.
+ * @direction - TX/RX crypto direction
+ */
+static void chcr_ktls_dev_del(struct net_device *netdev,
+			      struct tls_context *tls_ctx,
+			      enum tls_offload_ctx_dir direction)
+{
+	struct chcr_ktls_info *tx_info = chcr_get_ktls_tx_info(tls_ctx);
+	struct ch_ktls_port_stats_debug *port_stats;
+	struct chcr_ktls_uld_ctx *u_ctx;
+
+	if (!tx_info)
+		return;
+
+	u_ctx = tx_info->adap->uld[CXGB4_ULD_KTLS].handle;
+	if (u_ctx && u_ctx->detach)
+		return;
+	/* clear l2t entry */
+	if (tx_info->l2te)
+		cxgb4_l2t_release(tx_info->l2te);
+
+#if IS_ENABLED(CONFIG_IPV6)
+	/* clear clip entry */
+	if (tx_info->ip_family == AF_INET6)
+		cxgb4_clip_release(netdev, (const u32 *)
+				   &tx_info->sk->sk_v6_rcv_saddr,
+				   1);
+#endif
+
+	/* clear tid */
+	if (tx_info->tid != -1) {
+		cxgb4_remove_tid(&tx_info->adap->tids, tx_info->tx_chan,
+				 tx_info->tid, tx_info->ip_family);
+
+		xa_erase(&u_ctx->tid_list, tx_info->tid);
+	}
+
+	port_stats = &tx_info->adap->ch_ktls_stats.ktls_port[tx_info->port_id];
+	atomic64_inc(&port_stats->ktls_tx_connection_close);
+	kvfree(tx_info);
+	chcr_set_ktls_tx_info(tls_ctx, NULL);
+	/* release module refcount */
+	module_put(THIS_MODULE);
+}
+
+/*
+ * chcr_ktls_dev_add:  call back for tls_dev_add.
+ * Create a tcb entry for TP. Also add l2t entry for the connection. And
+ * generate keys & save those keys locally.
+ * @netdev - net device.
+ * @tls_cts - tls context.
+ * @direction - TX/RX crypto direction
+ * return: SUCCESS/FAILURE.
+ */
+static int chcr_ktls_dev_add(struct net_device *netdev, struct sock *sk,
+			     enum tls_offload_ctx_dir direction,
+			     struct tls_crypto_info *crypto_info,
+			     u32 start_offload_tcp_sn)
+{
+	struct tls_context *tls_ctx = tls_get_ctx(sk);
+	struct ch_ktls_port_stats_debug *port_stats;
+	struct chcr_ktls_uld_ctx *u_ctx;
+	struct chcr_ktls_info *tx_info;
+	struct dst_entry *dst;
+	struct adapter *adap;
+	struct port_info *pi;
+	struct neighbour *n;
+	u8 daaddr[16];
+	int ret = -1;
+
+	pi = netdev_priv(netdev);
+	adap = pi->adapter;
+	port_stats = &adap->ch_ktls_stats.ktls_port[pi->port_id];
+	atomic64_inc(&port_stats->ktls_tx_connection_open);
+	u_ctx = adap->uld[CXGB4_ULD_KTLS].handle;
+
+	if (direction == TLS_OFFLOAD_CTX_DIR_RX) {
+		pr_err("not expecting for RX direction\n");
+		goto out;
+	}
+
+	if (chcr_get_ktls_tx_info(tls_ctx))
+		goto out;
+
+	if (u_ctx && u_ctx->detach)
+		goto out;
+
+	tx_info = kvzalloc(sizeof(*tx_info), GFP_KERNEL);
+	if (!tx_info)
+		goto out;
+
+	tx_info->sk = sk;
+	spin_lock_init(&tx_info->lock);
+	/* initialize tid and atid to -1, 0 is a also a valid id. */
+	tx_info->tid = -1;
+	tx_info->atid = -1;
+
+	tx_info->adap = adap;
+	tx_info->netdev = netdev;
+	tx_info->first_qset = pi->first_qset;
+	tx_info->tx_chan = pi->tx_chan;
+	tx_info->smt_idx = pi->smt_idx;
+	tx_info->port_id = pi->port_id;
+	tx_info->prev_ack = 0;
+	tx_info->prev_win = 0;
+
+	tx_info->rx_qid = chcr_get_first_rx_qid(adap);
+	if (unlikely(tx_info->rx_qid < 0))
+		goto free_tx_info;
+
+	tx_info->prev_seq = start_offload_tcp_sn;
+	tx_info->tcp_start_seq_number = start_offload_tcp_sn;
+
+	/* save crypto keys */
+	ret = chcr_ktls_save_keys(tx_info, crypto_info, direction);
+	if (ret < 0)
+		goto free_tx_info;
+
+	/* get peer ip */
+	if (sk->sk_family == AF_INET) {
+		memcpy(daaddr, &sk->sk_daddr, 4);
+		tx_info->ip_family = AF_INET;
+#if IS_ENABLED(CONFIG_IPV6)
+	} else {
+		if (!ipv6_only_sock(sk) &&
+		    ipv6_addr_type(&sk->sk_v6_daddr) == IPV6_ADDR_MAPPED) {
+			memcpy(daaddr, &sk->sk_daddr, 4);
+			tx_info->ip_family = AF_INET;
+		} else {
+			memcpy(daaddr, sk->sk_v6_daddr.in6_u.u6_addr8, 16);
+			tx_info->ip_family = AF_INET6;
+		}
+#endif
+	}
+
+	/* get the l2t index */
+	dst = sk_dst_get(sk);
+	if (!dst) {
+		pr_err("DST entry not found\n");
+		goto free_tx_info;
+	}
+	n = dst_neigh_lookup(dst, daaddr);
+	if (!n || !n->dev) {
+		pr_err("neighbour not found\n");
+		dst_release(dst);
+		goto free_tx_info;
+	}
+	tx_info->l2te  = cxgb4_l2t_get(adap->l2t, n, n->dev, 0);
+
+	neigh_release(n);
+	dst_release(dst);
+
+	if (!tx_info->l2te) {
+		pr_err("l2t entry not found\n");
+		goto free_tx_info;
+	}
+
+	/* Driver shouldn't be removed until any single connection exists */
+	if (!try_module_get(THIS_MODULE))
+		goto free_l2t;
+
+	init_completion(&tx_info->completion);
+	/* create a filter and call cxgb4_l2t_send to send the packet out, which
+	 * will take care of updating l2t entry in hw if not already done.
+	 */
+	tx_info->open_state = CH_KTLS_OPEN_PENDING;
+
+	if (chcr_setup_connection(sk, tx_info))
+		goto put_module;
+
+	/* Wait for reply */
+	wait_for_completion_timeout(&tx_info->completion, 30 * HZ);
+	spin_lock_bh(&tx_info->lock);
+	if (tx_info->open_state) {
+		/* need to wait for hw response, can't free tx_info yet. */
+		if (tx_info->open_state == CH_KTLS_OPEN_PENDING)
+			tx_info->pending_close = true;
+		else
+			spin_unlock_bh(&tx_info->lock);
+		/* if in pending close, free the lock after the cleanup */
+		goto put_module;
+	}
+	spin_unlock_bh(&tx_info->lock);
+
+	/* initialize tcb */
+	reinit_completion(&tx_info->completion);
+	/* mark it pending for hw response */
+	tx_info->open_state = CH_KTLS_OPEN_PENDING;
+
+	if (chcr_init_tcb_fields(tx_info))
+		goto free_tid;
+
+	/* Wait for reply */
+	wait_for_completion_timeout(&tx_info->completion, 30 * HZ);
+	spin_lock_bh(&tx_info->lock);
+	if (tx_info->open_state) {
+		/* need to wait for hw response, can't free tx_info yet. */
+		tx_info->pending_close = true;
+		/* free the lock after cleanup */
+		goto free_tid;
+	}
+	spin_unlock_bh(&tx_info->lock);
+
+	if (!cxgb4_check_l2t_valid(tx_info->l2te))
+		goto free_tid;
+
+	atomic64_inc(&port_stats->ktls_tx_ctx);
+	chcr_set_ktls_tx_info(tls_ctx, tx_info);
+
+	return 0;
+
+free_tid:
+#if IS_ENABLED(CONFIG_IPV6)
+	/* clear clip entry */
+	if (tx_info->ip_family == AF_INET6)
+		cxgb4_clip_release(netdev, (const u32 *)
+				   &sk->sk_v6_rcv_saddr,
+				   1);
+#endif
+	cxgb4_remove_tid(&tx_info->adap->tids, tx_info->tx_chan,
+			 tx_info->tid, tx_info->ip_family);
+
+	xa_erase(&u_ctx->tid_list, tx_info->tid);
+
+put_module:
+	/* release module refcount */
+	module_put(THIS_MODULE);
+free_l2t:
+	cxgb4_l2t_release(tx_info->l2te);
+free_tx_info:
+	if (tx_info->pending_close)
+		spin_unlock_bh(&tx_info->lock);
+	else
+		kvfree(tx_info);
+out:
+	atomic64_inc(&port_stats->ktls_tx_connection_fail);
+	return -1;
+}
+
+/*
+ * chcr_init_tcb_fields:  Initialize tcb fields to handle TCP seq number
+ *			  handling.
+ * @tx_info - driver specific tls info.
+ * return: NET_TX_OK/NET_XMIT_DROP
+ */
+static int chcr_init_tcb_fields(struct chcr_ktls_info *tx_info)
+{
+	int  ret = 0;
+
+	/* set tcb in offload and bypass */
+	ret =
+	chcr_set_tcb_field(tx_info, TCB_T_FLAGS_W,
+			   TCB_T_FLAGS_V(TF_CORE_BYPASS_F | TF_NON_OFFLOAD_F),
+			   TCB_T_FLAGS_V(TF_CORE_BYPASS_F), 1);
+	if (ret)
+		return ret;
+	/* reset snd_una and snd_next fields in tcb */
+	ret = chcr_set_tcb_field(tx_info, TCB_SND_UNA_RAW_W,
+				 TCB_SND_NXT_RAW_V(TCB_SND_NXT_RAW_M) |
+				 TCB_SND_UNA_RAW_V(TCB_SND_UNA_RAW_M),
+				 0, 1);
+	if (ret)
+		return ret;
+
+	/* reset send max */
+	ret = chcr_set_tcb_field(tx_info, TCB_SND_MAX_RAW_W,
+				 TCB_SND_MAX_RAW_V(TCB_SND_MAX_RAW_M),
+				 0, 1);
+	if (ret)
+		return ret;
+
+	/* update l2t index and request for tp reply to confirm tcb is
+	 * initialised to handle tx traffic.
+	 */
+	ret = chcr_set_tcb_field(tx_info, TCB_L2T_IX_W,
+				 TCB_L2T_IX_V(TCB_L2T_IX_M),
+				 TCB_L2T_IX_V(tx_info->l2te->idx), 0);
+	return ret;
+}
+
+/*
+ * chcr_ktls_cpl_act_open_rpl: connection reply received from TP.
+ */
+static int chcr_ktls_cpl_act_open_rpl(struct adapter *adap,
+				      unsigned char *input)
+{
+	const struct cpl_act_open_rpl *p = (void *)input;
+	struct chcr_ktls_info *tx_info = NULL;
+	struct tls_offload_context_tx *tx_ctx;
+	struct chcr_ktls_uld_ctx *u_ctx;
+	unsigned int atid, tid, status;
+	struct tls_context *tls_ctx;
+	struct tid_info *t;
+	int ret = 0;
+
+	tid = GET_TID(p);
+	status = AOPEN_STATUS_G(ntohl(p->atid_status));
+	atid = TID_TID_G(AOPEN_ATID_G(ntohl(p->atid_status)));
+
+	t = &adap->tids;
+	tx_info = lookup_atid(t, atid);
+
+	if (!tx_info || tx_info->atid != atid) {
+		pr_err("%s: incorrect tx_info or atid\n", __func__);
+		return -1;
+	}
+
+	cxgb4_free_atid(t, atid);
+	tx_info->atid = -1;
+
+	spin_lock(&tx_info->lock);
+	/* HW response is very close, finish pending cleanup */
+	if (tx_info->pending_close) {
+		spin_unlock(&tx_info->lock);
+		if (!status) {
+			cxgb4_remove_tid(&tx_info->adap->tids, tx_info->tx_chan,
+					 tid, tx_info->ip_family);
+		}
+		kvfree(tx_info);
+		return 0;
+	}
+
+	if (!status) {
+		tx_info->tid = tid;
+		cxgb4_insert_tid(t, tx_info, tx_info->tid, tx_info->ip_family);
+		/* Adding tid */
+		tls_ctx = tls_get_ctx(tx_info->sk);
+		tx_ctx = tls_offload_ctx_tx(tls_ctx);
+		u_ctx = adap->uld[CXGB4_ULD_KTLS].handle;
+		if (u_ctx) {
+			ret = xa_insert_bh(&u_ctx->tid_list, tid, tx_ctx,
+					   GFP_NOWAIT);
+			if (ret < 0) {
+				pr_err("%s: Failed to allocate tid XA entry = %d\n",
+				       __func__, tx_info->tid);
+				tx_info->open_state = CH_KTLS_OPEN_FAILURE;
+				goto out;
+			}
+		}
+		tx_info->open_state = CH_KTLS_OPEN_SUCCESS;
+	} else {
+		tx_info->open_state = CH_KTLS_OPEN_FAILURE;
+	}
+out:
+	spin_unlock(&tx_info->lock);
+
+	complete(&tx_info->completion);
+	return ret;
+}
+
+/*
+ * chcr_ktls_cpl_set_tcb_rpl: TCB reply received from TP.
+ */
+static int chcr_ktls_cpl_set_tcb_rpl(struct adapter *adap, unsigned char *input)
+{
+	const struct cpl_set_tcb_rpl *p = (void *)input;
+	struct chcr_ktls_info *tx_info = NULL;
+	struct tid_info *t;
+	u32 tid;
+
+	tid = GET_TID(p);
+
+	t = &adap->tids;
+	tx_info = lookup_tid(t, tid);
+
+	if (!tx_info || tx_info->tid != tid) {
+		pr_err("%s: incorrect tx_info or tid\n", __func__);
+		return -1;
+	}
+
+	spin_lock(&tx_info->lock);
+	if (tx_info->pending_close) {
+		spin_unlock(&tx_info->lock);
+		kvfree(tx_info);
+		return 0;
+	}
+	tx_info->open_state = CH_KTLS_OPEN_SUCCESS;
+	spin_unlock(&tx_info->lock);
+
+	complete(&tx_info->completion);
+	return 0;
+}
+
+static void *__chcr_write_cpl_set_tcb_ulp(struct chcr_ktls_info *tx_info,
+					u32 tid, void *pos, u16 word,
+					struct sge_eth_txq *q, u64 mask,
+					u64 val, u32 reply)
+{
+	struct cpl_set_tcb_field_core *cpl;
+	struct ulptx_idata *idata;
+	struct ulp_txpkt *txpkt;
+
+	/* ULP_TXPKT */
+	txpkt = pos;
+	txpkt->cmd_dest = htonl(ULPTX_CMD_V(ULP_TX_PKT) |
+				ULP_TXPKT_CHANNELID_V(tx_info->port_id) |
+				ULP_TXPKT_FID_V(q->q.cntxt_id) |
+				ULP_TXPKT_RO_F);
+	txpkt->len = htonl(DIV_ROUND_UP(CHCR_SET_TCB_FIELD_LEN, 16));
+
+	/* ULPTX_IDATA sub-command */
+	idata = (struct ulptx_idata *)(txpkt + 1);
+	idata->cmd_more = htonl(ULPTX_CMD_V(ULP_TX_SC_IMM));
+	idata->len = htonl(sizeof(*cpl));
+	pos = idata + 1;
+
+	cpl = pos;
+	/* CPL_SET_TCB_FIELD */
+	OPCODE_TID(cpl) = htonl(MK_OPCODE_TID(CPL_SET_TCB_FIELD, tid));
+	cpl->reply_ctrl = htons(QUEUENO_V(tx_info->rx_qid) |
+			NO_REPLY_V(!reply));
+	cpl->word_cookie = htons(TCB_WORD_V(word));
+	cpl->mask = cpu_to_be64(mask);
+	cpl->val = cpu_to_be64(val);
+
+	/* ULPTX_NOOP */
+	idata = (struct ulptx_idata *)(cpl + 1);
+	idata->cmd_more = htonl(ULPTX_CMD_V(ULP_TX_SC_NOOP));
+	idata->len = htonl(0);
+	pos = idata + 1;
+
+	return pos;
+}
+
+
+/*
+ * chcr_write_cpl_set_tcb_ulp: update tcb values.
+ * TCB is responsible to create tcp headers, so all the related values
+ * should be correctly updated.
+ * @tx_info - driver specific tls info.
+ * @q - tx queue on which packet is going out.
+ * @tid - TCB identifier.
+ * @pos - current index where should we start writing.
+ * @word - TCB word.
+ * @mask - TCB word related mask.
+ * @val - TCB word related value.
+ * @reply - set 1 if looking for TP response.
+ * return - next position to write.
+ */
+static void *chcr_write_cpl_set_tcb_ulp(struct chcr_ktls_info *tx_info,
+					struct sge_eth_txq *q, u32 tid,
+					void *pos, u16 word, u64 mask,
+					u64 val, u32 reply)
+{
+	int left = (void *)q->q.stat - pos;
+
+	if (unlikely(left < CHCR_SET_TCB_FIELD_LEN)) {
+		if (!left) {
+			pos = q->q.desc;
+		} else {
+			u8 buf[48] = {0};
+
+			__chcr_write_cpl_set_tcb_ulp(tx_info, tid, buf, word, q,
+						     mask, val, reply);
+
+			return chcr_copy_to_txd(buf, &q->q, pos,
+						CHCR_SET_TCB_FIELD_LEN);
+		}
+	}
+
+	pos = __chcr_write_cpl_set_tcb_ulp(tx_info, tid, pos, word, q,
+					   mask, val, reply);
+
+	/* check again if we are at the end of the queue */
+	if (left == CHCR_SET_TCB_FIELD_LEN)
+		pos = q->q.desc;
+
+	return pos;
+}
+
+/*
+ * chcr_ktls_xmit_tcb_cpls: update tcb entry so that TP will create the header
+ * with updated values like tcp seq, ack, window etc.
+ * @tx_info - driver specific tls info.
+ * @q - TX queue.
+ * @tcp_seq
+ * @tcp_ack
+ * @tcp_win
+ * return: NETDEV_TX_BUSY/NET_TX_OK.
+ */
+static int chcr_ktls_xmit_tcb_cpls(struct chcr_ktls_info *tx_info,
+				   struct sge_eth_txq *q, u64 tcp_seq,
+				   u64 tcp_ack, u64 tcp_win, bool offset)
+{
+	bool first_wr = ((tx_info->prev_ack == 0) && (tx_info->prev_win == 0));
+	struct ch_ktls_port_stats_debug *port_stats;
+	u32 len, cpl = 0, ndesc, wr_len, wr_mid = 0;
+	struct fw_ulptx_wr *wr;
+	int credits;
+	void *pos;
+
+	wr_len = sizeof(*wr);
+	/* there can be max 4 cpls, check if we have enough credits */
+	len = wr_len + 4 * roundup(CHCR_SET_TCB_FIELD_LEN, 16);
+	ndesc = DIV_ROUND_UP(len, 64);
+
+	credits = chcr_txq_avail(&q->q) - ndesc;
+	if (unlikely(credits < 0)) {
+		chcr_eth_txq_stop(q);
+		return NETDEV_TX_BUSY;
+	}
+
+	if (unlikely(credits < ETHTXQ_STOP_THRES)) {
+		chcr_eth_txq_stop(q);
+		wr_mid |= FW_WR_EQUEQ_F | FW_WR_EQUIQ_F;
+	}
+
+	pos = &q->q.desc[q->q.pidx];
+	/* make space for WR, we'll fill it later when we know all the cpls
+	 * being sent out and have complete length.
+	 */
+	wr = pos;
+	pos += wr_len;
+	/* update tx_max if its a re-transmit or the first wr */
+	if (first_wr || tcp_seq != tx_info->prev_seq) {
+		pos = chcr_write_cpl_set_tcb_ulp(tx_info, q, tx_info->tid, pos,
+						 TCB_TX_MAX_W,
+						 TCB_TX_MAX_V(TCB_TX_MAX_M),
+						 TCB_TX_MAX_V(tcp_seq), 0);
+		cpl++;
+	}
+	/* reset snd una if it's a re-transmit pkt */
+	if (tcp_seq != tx_info->prev_seq || offset) {
+		/* reset snd_una */
+		port_stats =
+			&tx_info->adap->ch_ktls_stats.ktls_port[tx_info->port_id];
+		pos = chcr_write_cpl_set_tcb_ulp(tx_info, q, tx_info->tid, pos,
+						 TCB_SND_UNA_RAW_W,
+						 TCB_SND_UNA_RAW_V
+						 (TCB_SND_UNA_RAW_M),
+						 TCB_SND_UNA_RAW_V(0), 0);
+		if (tcp_seq != tx_info->prev_seq)
+			atomic64_inc(&port_stats->ktls_tx_ooo);
+		cpl++;
+	}
+	/* update ack */
+	if (first_wr || tx_info->prev_ack != tcp_ack) {
+		pos = chcr_write_cpl_set_tcb_ulp(tx_info, q, tx_info->tid, pos,
+						 TCB_RCV_NXT_W,
+						 TCB_RCV_NXT_V(TCB_RCV_NXT_M),
+						 TCB_RCV_NXT_V(tcp_ack), 0);
+		tx_info->prev_ack = tcp_ack;
+		cpl++;
+	}
+	/* update receive window */
+	if (first_wr || tx_info->prev_win != tcp_win) {
+		chcr_write_cpl_set_tcb_ulp(tx_info, q, tx_info->tid, pos,
+					   TCB_RCV_WND_W,
+					   TCB_RCV_WND_V(TCB_RCV_WND_M),
+					   TCB_RCV_WND_V(tcp_win), 0);
+		tx_info->prev_win = tcp_win;
+		cpl++;
+	}
+
+	if (cpl) {
+		/* get the actual length */
+		len = wr_len + cpl * roundup(CHCR_SET_TCB_FIELD_LEN, 16);
+		/* ULPTX wr */
+		wr->op_to_compl = htonl(FW_WR_OP_V(FW_ULPTX_WR));
+		wr->cookie = 0;
+		/* fill len in wr field */
+		wr->flowid_len16 = htonl(wr_mid |
+					 FW_WR_LEN16_V(DIV_ROUND_UP(len, 16)));
+
+		ndesc = DIV_ROUND_UP(len, 64);
+		chcr_txq_advance(&q->q, ndesc);
+		cxgb4_ring_tx_db(tx_info->adap, &q->q, ndesc);
+	}
+	return 0;
+}
+
+/*
+ * chcr_ktls_get_tx_flits
+ * returns number of flits to be sent out, it includes key context length, WR
+ * size and skb fragments.
+ */
+static unsigned int
+chcr_ktls_get_tx_flits(u32 nr_frags, unsigned int key_ctx_len)
+{
+	return chcr_sgl_len(nr_frags) +
+	       DIV_ROUND_UP(key_ctx_len + CHCR_KTLS_WR_SIZE, 8);
+}
+
+/*
+ * chcr_ktls_check_tcp_options: To check if there is any TCP option available
+ * other than timestamp.
+ * @skb - skb contains partial record..
+ * return: 1 / 0
+ */
+static int
+chcr_ktls_check_tcp_options(struct tcphdr *tcp)
+{
+	int cnt, opt, optlen;
+	u_char *cp;
+
+	cp = (u_char *)(tcp + 1);
+	cnt = (tcp->doff << 2) - sizeof(struct tcphdr);
+	for (; cnt > 0; cnt -= optlen, cp += optlen) {
+		opt = cp[0];
+		if (opt == TCPOPT_EOL)
+			break;
+		if (opt == TCPOPT_NOP) {
+			optlen = 1;
+		} else {
+			if (cnt < 2)
+				break;
+			optlen = cp[1];
+			if (optlen < 2 || optlen > cnt)
+				break;
+		}
+		switch (opt) {
+		case TCPOPT_NOP:
+			break;
+		default:
+			return 1;
+		}
+	}
+	return 0;
+}
+
+/*
+ * chcr_ktls_write_tcp_options : TP can't send out all the options, we need to
+ * send out separately.
+ * @tx_info - driver specific tls info.
+ * @skb - skb contains partial record..
+ * @q - TX queue.
+ * @tx_chan - channel number.
+ * return: NETDEV_TX_OK/NETDEV_TX_BUSY.
+ */
+static int
+chcr_ktls_write_tcp_options(struct chcr_ktls_info *tx_info, struct sk_buff *skb,
+			    struct sge_eth_txq *q, uint32_t tx_chan)
+{
+	struct fw_eth_tx_pkt_wr *wr;
+	struct cpl_tx_pkt_core *cpl;
+	u32 ctrl, iplen, maclen;
+	struct ipv6hdr *ip6;
+	unsigned int ndesc;
+	struct tcphdr *tcp;
+	int len16, pktlen;
+	struct iphdr *ip;
+	u32 wr_mid = 0;
+	int credits;
+	u8 buf[150];
+	u64 cntrl1;
+	void *pos;
+
+	iplen = skb_network_header_len(skb);
+	maclen = skb_mac_header_len(skb);
+
+	/* packet length = eth hdr len + ip hdr len + tcp hdr len
+	 * (including options).
+	 */
+	pktlen = skb_tcp_all_headers(skb);
+
+	ctrl = sizeof(*cpl) + pktlen;
+	len16 = DIV_ROUND_UP(sizeof(*wr) + ctrl, 16);
+	/* check how many descriptors needed */
+	ndesc = DIV_ROUND_UP(len16, 4);
+
+	credits = chcr_txq_avail(&q->q) - ndesc;
+	if (unlikely(credits < 0)) {
+		chcr_eth_txq_stop(q);
+		return NETDEV_TX_BUSY;
+	}
+
+	if (unlikely(credits < ETHTXQ_STOP_THRES)) {
+		chcr_eth_txq_stop(q);
+		wr_mid |= FW_WR_EQUEQ_F | FW_WR_EQUIQ_F;
+	}
+
+	pos = &q->q.desc[q->q.pidx];
+	wr = pos;
+
+	/* Firmware work request header */
+	wr->op_immdlen = htonl(FW_WR_OP_V(FW_ETH_TX_PKT_WR) |
+			       FW_WR_IMMDLEN_V(ctrl));
+
+	wr->equiq_to_len16 = htonl(wr_mid | FW_WR_LEN16_V(len16));
+	wr->r3 = 0;
+
+	cpl = (void *)(wr + 1);
+
+	/* CPL header */
+	cpl->ctrl0 = htonl(TXPKT_OPCODE_V(CPL_TX_PKT) | TXPKT_INTF_V(tx_chan) |
+			   TXPKT_PF_V(tx_info->adap->pf));
+	cpl->pack = 0;
+	cpl->len = htons(pktlen);
+
+	memcpy(buf, skb->data, pktlen);
+	if (!IS_ENABLED(CONFIG_IPV6) || tx_info->ip_family == AF_INET) {
+		/* we need to correct ip header len */
+		ip = (struct iphdr *)(buf + maclen);
+		ip->tot_len = htons(pktlen - maclen);
+		cntrl1 = TXPKT_CSUM_TYPE_V(TX_CSUM_TCPIP);
+	} else {
+		ip6 = (struct ipv6hdr *)(buf + maclen);
+		ip6->payload_len = htons(pktlen - maclen - iplen);
+		cntrl1 = TXPKT_CSUM_TYPE_V(TX_CSUM_TCPIP6);
+	}
+
+	cntrl1 |= T6_TXPKT_ETHHDR_LEN_V(maclen - ETH_HLEN) |
+		  TXPKT_IPHDR_LEN_V(iplen);
+	/* checksum offload */
+	cpl->ctrl1 = cpu_to_be64(cntrl1);
+
+	pos = cpl + 1;
+
+	/* now take care of the tcp header, if fin is not set then clear push
+	 * bit as well, and if fin is set, it will be sent at the last so we
+	 * need to update the tcp sequence number as per the last packet.
+	 */
+	tcp = (struct tcphdr *)(buf + maclen + iplen);
+
+	if (!tcp->fin)
+		tcp->psh = 0;
+	else
+		tcp->seq = htonl(tx_info->prev_seq);
+
+	chcr_copy_to_txd(buf, &q->q, pos, pktlen);
+
+	chcr_txq_advance(&q->q, ndesc);
+	cxgb4_ring_tx_db(tx_info->adap, &q->q, ndesc);
+	return 0;
+}
+
+/*
+ * chcr_ktls_xmit_wr_complete: This sends out the complete record. If an skb
+ * received has partial end part of the record, send out the complete record, so
+ * that crypto block will be able to generate TAG/HASH.
+ * @skb - segment which has complete or partial end part.
+ * @tx_info - driver specific tls info.
+ * @q - TX queue.
+ * @tcp_seq
+ * @tcp_push - tcp push bit.
+ * @mss - segment size.
+ * return: NETDEV_TX_BUSY/NET_TX_OK.
+ */
+static int chcr_ktls_xmit_wr_complete(struct sk_buff *skb,
+				      struct chcr_ktls_info *tx_info,
+				      struct sge_eth_txq *q, u32 tcp_seq,
+				      bool is_last_wr, u32 data_len,
+				      u32 skb_offset, u32 nfrags,
+				      bool tcp_push, u32 mss)
+{
+	u32 len16, wr_mid = 0, flits = 0, ndesc, cipher_start;
+	struct adapter *adap = tx_info->adap;
+	int credits, left, last_desc;
+	struct tx_sw_desc *sgl_sdesc;
+	struct cpl_tx_data *tx_data;
+	struct cpl_tx_sec_pdu *cpl;
+	struct ulptx_idata *idata;
+	struct ulp_txpkt *ulptx;
+	struct fw_ulptx_wr *wr;
+	void *pos;
+	u64 *end;
+
+	/* get the number of flits required */
+	flits = chcr_ktls_get_tx_flits(nfrags, tx_info->key_ctx_len);
+	/* number of descriptors */
+	ndesc = chcr_flits_to_desc(flits);
+	/* check if enough credits available */
+	credits = chcr_txq_avail(&q->q) - ndesc;
+	if (unlikely(credits < 0)) {
+		chcr_eth_txq_stop(q);
+		return NETDEV_TX_BUSY;
+	}
+
+	if (unlikely(credits < ETHTXQ_STOP_THRES)) {
+		/* Credits are below the threshold values, stop the queue after
+		 * injecting the Work Request for this packet.
+		 */
+		chcr_eth_txq_stop(q);
+		wr_mid |= FW_WR_EQUEQ_F | FW_WR_EQUIQ_F;
+	}
+
+	last_desc = q->q.pidx + ndesc - 1;
+	if (last_desc >= q->q.size)
+		last_desc -= q->q.size;
+	sgl_sdesc = &q->q.sdesc[last_desc];
+
+	if (unlikely(cxgb4_map_skb(adap->pdev_dev, skb, sgl_sdesc->addr) < 0)) {
+		memset(sgl_sdesc->addr, 0, sizeof(sgl_sdesc->addr));
+		q->mapping_err++;
+		return NETDEV_TX_BUSY;
+	}
+
+	if (!is_last_wr)
+		skb_get(skb);
+
+	pos = &q->q.desc[q->q.pidx];
+	end = (u64 *)pos + flits;
+	/* FW_ULPTX_WR */
+	wr = pos;
+	/* WR will need len16 */
+	len16 = DIV_ROUND_UP(flits, 2);
+	wr->op_to_compl = htonl(FW_WR_OP_V(FW_ULPTX_WR));
+	wr->flowid_len16 = htonl(wr_mid | FW_WR_LEN16_V(len16));
+	wr->cookie = 0;
+	pos += sizeof(*wr);
+	/* ULP_TXPKT */
+	ulptx = pos;
+	ulptx->cmd_dest = htonl(ULPTX_CMD_V(ULP_TX_PKT) |
+				ULP_TXPKT_CHANNELID_V(tx_info->port_id) |
+				ULP_TXPKT_FID_V(q->q.cntxt_id) |
+				ULP_TXPKT_RO_F);
+	ulptx->len = htonl(len16 - 1);
+	/* ULPTX_IDATA sub-command */
+	idata = (struct ulptx_idata *)(ulptx + 1);
+	idata->cmd_more = htonl(ULPTX_CMD_V(ULP_TX_SC_IMM) | ULP_TX_SC_MORE_F);
+	/* idata length will include cpl_tx_sec_pdu + key context size +
+	 * cpl_tx_data header.
+	 */
+	idata->len = htonl(sizeof(*cpl) + tx_info->key_ctx_len +
+			   sizeof(*tx_data));
+	/* SEC CPL */
+	cpl = (struct cpl_tx_sec_pdu *)(idata + 1);
+	cpl->op_ivinsrtofst =
+		htonl(CPL_TX_SEC_PDU_OPCODE_V(CPL_TX_SEC_PDU) |
+		      CPL_TX_SEC_PDU_CPLLEN_V(CHCR_CPL_TX_SEC_PDU_LEN_64BIT) |
+		      CPL_TX_SEC_PDU_PLACEHOLDER_V(1) |
+		      CPL_TX_SEC_PDU_IVINSRTOFST_V(TLS_HEADER_SIZE + 1));
+	cpl->pldlen = htonl(data_len);
+
+	/* encryption should start after tls header size + iv size */
+	cipher_start = TLS_HEADER_SIZE + tx_info->iv_size + 1;
+
+	cpl->aadstart_cipherstop_hi =
+		htonl(CPL_TX_SEC_PDU_AADSTART_V(1) |
+		      CPL_TX_SEC_PDU_AADSTOP_V(TLS_HEADER_SIZE) |
+		      CPL_TX_SEC_PDU_CIPHERSTART_V(cipher_start));
+
+	/* authentication will also start after tls header + iv size */
+	cpl->cipherstop_lo_authinsert =
+	htonl(CPL_TX_SEC_PDU_AUTHSTART_V(cipher_start) |
+	      CPL_TX_SEC_PDU_AUTHSTOP_V(TLS_CIPHER_AES_GCM_128_TAG_SIZE) |
+	      CPL_TX_SEC_PDU_AUTHINSERT_V(TLS_CIPHER_AES_GCM_128_TAG_SIZE));
+
+	/* These two flits are actually a CPL_TLS_TX_SCMD_FMT. */
+	cpl->seqno_numivs = htonl(tx_info->scmd0_seqno_numivs);
+	cpl->ivgen_hdrlen = htonl(tx_info->scmd0_ivgen_hdrlen);
+	cpl->scmd1 = cpu_to_be64(tx_info->record_no);
+
+	pos = cpl + 1;
+	/* check if space left to fill the keys */
+	left = (void *)q->q.stat - pos;
+	if (!left) {
+		left = (void *)end - (void *)q->q.stat;
+		pos = q->q.desc;
+		end = pos + left;
+	}
+
+	pos = chcr_copy_to_txd(&tx_info->key_ctx, &q->q, pos,
+			       tx_info->key_ctx_len);
+	left = (void *)q->q.stat - pos;
+
+	if (!left) {
+		left = (void *)end - (void *)q->q.stat;
+		pos = q->q.desc;
+		end = pos + left;
+	}
+	/* CPL_TX_DATA */
+	tx_data = (void *)pos;
+	OPCODE_TID(tx_data) = htonl(MK_OPCODE_TID(CPL_TX_DATA, tx_info->tid));
+	tx_data->len = htonl(TX_DATA_MSS_V(mss) | TX_LENGTH_V(data_len));
+
+	tx_data->rsvd = htonl(tcp_seq);
+
+	tx_data->flags = htonl(TX_BYPASS_F);
+	if (tcp_push)
+		tx_data->flags |= htonl(TX_PUSH_F | TX_SHOVE_F);
+
+	/* check left again, it might go beyond queue limit */
+	pos = tx_data + 1;
+	left = (void *)q->q.stat - pos;
+
+	/* check the position again */
+	if (!left) {
+		left = (void *)end - (void *)q->q.stat;
+		pos = q->q.desc;
+		end = pos + left;
+	}
+
+	/* send the complete packet except the header */
+	cxgb4_write_partial_sgl(skb, &q->q, pos, end, sgl_sdesc->addr,
+				skb_offset, data_len);
+	sgl_sdesc->skb = skb;
+
+	chcr_txq_advance(&q->q, ndesc);
+	cxgb4_ring_tx_db(adap, &q->q, ndesc);
+	atomic64_inc(&adap->ch_ktls_stats.ktls_tx_send_records);
+
+	return 0;
+}
+
+/*
+ * chcr_ktls_xmit_wr_short: This is to send out partial records. If its
+ * a middle part of a record, fetch the prior data to make it 16 byte aligned
+ * and then only send it out.
+ *
+ * @skb - skb contains partial record..
+ * @tx_info - driver specific tls info.
+ * @q - TX queue.
+ * @tcp_seq
+ * @tcp_push - tcp push bit.
+ * @mss - segment size.
+ * @tls_rec_offset - offset from start of the tls record.
+ * @perior_data - data before the current segment, required to make this record
+ *		  16 byte aligned.
+ * @prior_data_len - prior_data length (less than 16)
+ * return: NETDEV_TX_BUSY/NET_TX_OK.
+ */
+static int chcr_ktls_xmit_wr_short(struct sk_buff *skb,
+				   struct chcr_ktls_info *tx_info,
+				   struct sge_eth_txq *q,
+				   u32 tcp_seq, bool tcp_push, u32 mss,
+				   u32 tls_rec_offset, u8 *prior_data,
+				   u32 prior_data_len, u32 data_len,
+				   u32 skb_offset)
+{
+	u32 len16, wr_mid = 0, cipher_start, nfrags;
+	struct adapter *adap = tx_info->adap;
+	unsigned int flits = 0, ndesc;
+	int credits, left, last_desc;
+	struct tx_sw_desc *sgl_sdesc;
+	struct cpl_tx_data *tx_data;
+	struct cpl_tx_sec_pdu *cpl;
+	struct ulptx_idata *idata;
+	struct ulp_txpkt *ulptx;
+	struct fw_ulptx_wr *wr;
+	__be64 iv_record;
+	void *pos;
+	u64 *end;
+
+	nfrags = chcr_get_nfrags_to_send(skb, skb_offset, data_len);
+	/* get the number of flits required, it's a partial record so 2 flits
+	 * (AES_BLOCK_SIZE) will be added.
+	 */
+	flits = chcr_ktls_get_tx_flits(nfrags, tx_info->key_ctx_len) + 2;
+	/* get the correct 8 byte IV of this record */
+	iv_record = cpu_to_be64(tx_info->iv + tx_info->record_no);
+	/* If it's a middle record and not 16 byte aligned to run AES CTR, need
+	 * to make it 16 byte aligned. So atleadt 2 extra flits of immediate
+	 * data will be added.
+	 */
+	if (prior_data_len)
+		flits += 2;
+	/* number of descriptors */
+	ndesc = chcr_flits_to_desc(flits);
+	/* check if enough credits available */
+	credits = chcr_txq_avail(&q->q) - ndesc;
+	if (unlikely(credits < 0)) {
+		chcr_eth_txq_stop(q);
+		return NETDEV_TX_BUSY;
+	}
+
+	if (unlikely(credits < ETHTXQ_STOP_THRES)) {
+		chcr_eth_txq_stop(q);
+		wr_mid |= FW_WR_EQUEQ_F | FW_WR_EQUIQ_F;
+	}
+
+	last_desc = q->q.pidx + ndesc - 1;
+	if (last_desc >= q->q.size)
+		last_desc -= q->q.size;
+	sgl_sdesc = &q->q.sdesc[last_desc];
+
+	if (unlikely(cxgb4_map_skb(adap->pdev_dev, skb, sgl_sdesc->addr) < 0)) {
+		memset(sgl_sdesc->addr, 0, sizeof(sgl_sdesc->addr));
+		q->mapping_err++;
+		return NETDEV_TX_BUSY;
+	}
+
+	pos = &q->q.desc[q->q.pidx];
+	end = (u64 *)pos + flits;
+	/* FW_ULPTX_WR */
+	wr = pos;
+	/* WR will need len16 */
+	len16 = DIV_ROUND_UP(flits, 2);
+	wr->op_to_compl = htonl(FW_WR_OP_V(FW_ULPTX_WR));
+	wr->flowid_len16 = htonl(wr_mid | FW_WR_LEN16_V(len16));
+	wr->cookie = 0;
+	pos += sizeof(*wr);
+	/* ULP_TXPKT */
+	ulptx = pos;
+	ulptx->cmd_dest = htonl(ULPTX_CMD_V(ULP_TX_PKT) |
+				ULP_TXPKT_CHANNELID_V(tx_info->port_id) |
+				ULP_TXPKT_FID_V(q->q.cntxt_id) |
+				ULP_TXPKT_RO_F);
+	ulptx->len = htonl(len16 - 1);
+	/* ULPTX_IDATA sub-command */
+	idata = (struct ulptx_idata *)(ulptx + 1);
+	idata->cmd_more = htonl(ULPTX_CMD_V(ULP_TX_SC_IMM) | ULP_TX_SC_MORE_F);
+	/* idata length will include cpl_tx_sec_pdu + key context size +
+	 * cpl_tx_data header.
+	 */
+	idata->len = htonl(sizeof(*cpl) + tx_info->key_ctx_len +
+			   sizeof(*tx_data) + AES_BLOCK_LEN + prior_data_len);
+	/* SEC CPL */
+	cpl = (struct cpl_tx_sec_pdu *)(idata + 1);
+	/* cipher start will have tls header + iv size extra if its a header
+	 * part of tls record. else only 16 byte IV will be added.
+	 */
+	cipher_start =
+		AES_BLOCK_LEN + 1 +
+		(!tls_rec_offset ? TLS_HEADER_SIZE + tx_info->iv_size : 0);
+
+	cpl->op_ivinsrtofst =
+		htonl(CPL_TX_SEC_PDU_OPCODE_V(CPL_TX_SEC_PDU) |
+		      CPL_TX_SEC_PDU_CPLLEN_V(CHCR_CPL_TX_SEC_PDU_LEN_64BIT) |
+		      CPL_TX_SEC_PDU_IVINSRTOFST_V(1));
+	cpl->pldlen = htonl(data_len + AES_BLOCK_LEN + prior_data_len);
+	cpl->aadstart_cipherstop_hi =
+		htonl(CPL_TX_SEC_PDU_CIPHERSTART_V(cipher_start));
+	cpl->cipherstop_lo_authinsert = 0;
+	/* These two flits are actually a CPL_TLS_TX_SCMD_FMT. */
+	cpl->seqno_numivs = htonl(tx_info->scmd0_short_seqno_numivs);
+	cpl->ivgen_hdrlen = htonl(tx_info->scmd0_short_ivgen_hdrlen);
+	cpl->scmd1 = 0;
+
+	pos = cpl + 1;
+	/* check if space left to fill the keys */
+	left = (void *)q->q.stat - pos;
+	if (!left) {
+		left = (void *)end - (void *)q->q.stat;
+		pos = q->q.desc;
+		end = pos + left;
+	}
+
+	pos = chcr_copy_to_txd(&tx_info->key_ctx, &q->q, pos,
+			       tx_info->key_ctx_len);
+	left = (void *)q->q.stat - pos;
+
+	if (!left) {
+		left = (void *)end - (void *)q->q.stat;
+		pos = q->q.desc;
+		end = pos + left;
+	}
+	/* CPL_TX_DATA */
+	tx_data = (void *)pos;
+	OPCODE_TID(tx_data) = htonl(MK_OPCODE_TID(CPL_TX_DATA, tx_info->tid));
+	tx_data->len = htonl(TX_DATA_MSS_V(mss) |
+			     TX_LENGTH_V(data_len + prior_data_len));
+	tx_data->rsvd = htonl(tcp_seq);
+	tx_data->flags = htonl(TX_BYPASS_F);
+	if (tcp_push)
+		tx_data->flags |= htonl(TX_PUSH_F | TX_SHOVE_F);
+
+	/* check left again, it might go beyond queue limit */
+	pos = tx_data + 1;
+	left = (void *)q->q.stat - pos;
+
+	/* check the position again */
+	if (!left) {
+		left = (void *)end - (void *)q->q.stat;
+		pos = q->q.desc;
+		end = pos + left;
+	}
+	/* copy the 16 byte IV for AES-CTR, which includes 4 bytes of salt, 8
+	 * bytes of actual IV and 4 bytes of 16 byte-sequence.
+	 */
+	memcpy(pos, tx_info->key_ctx.salt, tx_info->salt_size);
+	memcpy(pos + tx_info->salt_size, &iv_record, tx_info->iv_size);
+	*(__be32 *)(pos + tx_info->salt_size + tx_info->iv_size) =
+		htonl(2 + (tls_rec_offset ? ((tls_rec_offset -
+		(TLS_HEADER_SIZE + tx_info->iv_size)) / AES_BLOCK_LEN) : 0));
+
+	pos += 16;
+	/* Prior_data_len will always be less than 16 bytes, fill the
+	 * prio_data_len after AES_CTRL_BLOCK and clear the remaining length
+	 * to 0.
+	 */
+	if (prior_data_len)
+		pos = chcr_copy_to_txd(prior_data, &q->q, pos, 16);
+	/* send the complete packet except the header */
+	cxgb4_write_partial_sgl(skb, &q->q, pos, end, sgl_sdesc->addr,
+				skb_offset, data_len);
+	sgl_sdesc->skb = skb;
+
+	chcr_txq_advance(&q->q, ndesc);
+	cxgb4_ring_tx_db(adap, &q->q, ndesc);
+
+	return 0;
+}
+
+/*
+ * chcr_ktls_tx_plaintxt: This handler will take care of the records which has
+ * only plain text (only tls header and iv)
+ * @tx_info - driver specific tls info.
+ * @skb - skb contains partial record..
+ * @tcp_seq
+ * @mss - segment size.
+ * @tcp_push - tcp push bit.
+ * @q - TX queue.
+ * @port_id : port number
+ * @perior_data - data before the current segment, required to make this record
+ *		 16 byte aligned.
+ * @prior_data_len - prior_data length (less than 16)
+ * return: NETDEV_TX_BUSY/NET_TX_OK.
+ */
+static int chcr_ktls_tx_plaintxt(struct chcr_ktls_info *tx_info,
+				 struct sk_buff *skb, u32 tcp_seq, u32 mss,
+				 bool tcp_push, struct sge_eth_txq *q,
+				 u32 port_id, u8 *prior_data,
+				 u32 data_len, u32 skb_offset,
+				 u32 prior_data_len)
+{
+	int credits, left, len16, last_desc;
+	unsigned int flits = 0, ndesc;
+	struct tx_sw_desc *sgl_sdesc;
+	struct cpl_tx_data *tx_data;
+	struct ulptx_idata *idata;
+	struct ulp_txpkt *ulptx;
+	struct fw_ulptx_wr *wr;
+	u32 wr_mid = 0, nfrags;
+	void *pos;
+	u64 *end;
+
+	flits = DIV_ROUND_UP(CHCR_PLAIN_TX_DATA_LEN, 8);
+	nfrags = chcr_get_nfrags_to_send(skb, skb_offset, data_len);
+	flits += chcr_sgl_len(nfrags);
+	if (prior_data_len)
+		flits += 2;
+
+	/* WR will need len16 */
+	len16 = DIV_ROUND_UP(flits, 2);
+	/* check how many descriptors needed */
+	ndesc = DIV_ROUND_UP(flits, 8);
+
+	credits = chcr_txq_avail(&q->q) - ndesc;
+	if (unlikely(credits < 0)) {
+		chcr_eth_txq_stop(q);
+		return NETDEV_TX_BUSY;
+	}
+
+	if (unlikely(credits < ETHTXQ_STOP_THRES)) {
+		chcr_eth_txq_stop(q);
+		wr_mid |= FW_WR_EQUEQ_F | FW_WR_EQUIQ_F;
+	}
+
+	last_desc = q->q.pidx + ndesc - 1;
+	if (last_desc >= q->q.size)
+		last_desc -= q->q.size;
+	sgl_sdesc = &q->q.sdesc[last_desc];
+
+	if (unlikely(cxgb4_map_skb(tx_info->adap->pdev_dev, skb,
+				   sgl_sdesc->addr) < 0)) {
+		memset(sgl_sdesc->addr, 0, sizeof(sgl_sdesc->addr));
+		q->mapping_err++;
+		return NETDEV_TX_BUSY;
+	}
+
+	pos = &q->q.desc[q->q.pidx];
+	end = (u64 *)pos + flits;
+	/* FW_ULPTX_WR */
+	wr = pos;
+	wr->op_to_compl = htonl(FW_WR_OP_V(FW_ULPTX_WR));
+	wr->flowid_len16 = htonl(wr_mid | FW_WR_LEN16_V(len16));
+	wr->cookie = 0;
+	/* ULP_TXPKT */
+	ulptx = (struct ulp_txpkt *)(wr + 1);
+	ulptx->cmd_dest = htonl(ULPTX_CMD_V(ULP_TX_PKT) |
+			ULP_TXPKT_DATAMODIFY_V(0) |
+			ULP_TXPKT_CHANNELID_V(tx_info->port_id) |
+			ULP_TXPKT_DEST_V(0) |
+			ULP_TXPKT_FID_V(q->q.cntxt_id) | ULP_TXPKT_RO_V(1));
+	ulptx->len = htonl(len16 - 1);
+	/* ULPTX_IDATA sub-command */
+	idata = (struct ulptx_idata *)(ulptx + 1);
+	idata->cmd_more = htonl(ULPTX_CMD_V(ULP_TX_SC_IMM) | ULP_TX_SC_MORE_F);
+	idata->len = htonl(sizeof(*tx_data) + prior_data_len);
+	/* CPL_TX_DATA */
+	tx_data = (struct cpl_tx_data *)(idata + 1);
+	OPCODE_TID(tx_data) = htonl(MK_OPCODE_TID(CPL_TX_DATA, tx_info->tid));
+	tx_data->len = htonl(TX_DATA_MSS_V(mss) |
+			     TX_LENGTH_V(data_len + prior_data_len));
+	/* set tcp seq number */
+	tx_data->rsvd = htonl(tcp_seq);
+	tx_data->flags = htonl(TX_BYPASS_F);
+	if (tcp_push)
+		tx_data->flags |= htonl(TX_PUSH_F | TX_SHOVE_F);
+
+	pos = tx_data + 1;
+	/* apart from prior_data_len, we should set remaining part of 16 bytes
+	 * to be zero.
+	 */
+	if (prior_data_len)
+		pos = chcr_copy_to_txd(prior_data, &q->q, pos, 16);
+
+	/* check left again, it might go beyond queue limit */
+	left = (void *)q->q.stat - pos;
+
+	/* check the position again */
+	if (!left) {
+		left = (void *)end - (void *)q->q.stat;
+		pos = q->q.desc;
+		end = pos + left;
+	}
+	/* send the complete packet including the header */
+	cxgb4_write_partial_sgl(skb, &q->q, pos, end, sgl_sdesc->addr,
+				skb_offset, data_len);
+	sgl_sdesc->skb = skb;
+
+	chcr_txq_advance(&q->q, ndesc);
+	cxgb4_ring_tx_db(tx_info->adap, &q->q, ndesc);
+	return 0;
+}
+
+static int chcr_ktls_tunnel_pkt(struct chcr_ktls_info *tx_info,
+				struct sk_buff *skb,
+				struct sge_eth_txq *q)
+{
+	u32 ctrl, iplen, maclen, wr_mid = 0, len16;
+	struct tx_sw_desc *sgl_sdesc;
+	struct fw_eth_tx_pkt_wr *wr;
+	struct cpl_tx_pkt_core *cpl;
+	unsigned int flits, ndesc;
+	int credits, last_desc;
+	u64 cntrl1, *end;
+	void *pos;
+
+	ctrl = sizeof(*cpl);
+	flits = DIV_ROUND_UP(sizeof(*wr) + ctrl, 8);
+
+	flits += chcr_sgl_len(skb_shinfo(skb)->nr_frags + 1);
+	len16 = DIV_ROUND_UP(flits, 2);
+	/* check how many descriptors needed */
+	ndesc = DIV_ROUND_UP(flits, 8);
+
+	credits = chcr_txq_avail(&q->q) - ndesc;
+	if (unlikely(credits < 0)) {
+		chcr_eth_txq_stop(q);
+		return -ENOMEM;
+	}
+
+	if (unlikely(credits < ETHTXQ_STOP_THRES)) {
+		chcr_eth_txq_stop(q);
+		wr_mid |= FW_WR_EQUEQ_F | FW_WR_EQUIQ_F;
+	}
+
+	last_desc = q->q.pidx + ndesc - 1;
+	if (last_desc >= q->q.size)
+		last_desc -= q->q.size;
+	sgl_sdesc = &q->q.sdesc[last_desc];
+
+	if (unlikely(cxgb4_map_skb(tx_info->adap->pdev_dev, skb,
+				   sgl_sdesc->addr) < 0)) {
+		memset(sgl_sdesc->addr, 0, sizeof(sgl_sdesc->addr));
+		q->mapping_err++;
+		return -ENOMEM;
+	}
+
+	iplen = skb_network_header_len(skb);
+	maclen = skb_mac_header_len(skb);
+
+	pos = &q->q.desc[q->q.pidx];
+	end = (u64 *)pos + flits;
+	wr = pos;
+
+	/* Firmware work request header */
+	wr->op_immdlen = htonl(FW_WR_OP_V(FW_ETH_TX_PKT_WR) |
+			       FW_WR_IMMDLEN_V(ctrl));
+
+	wr->equiq_to_len16 = htonl(wr_mid | FW_WR_LEN16_V(len16));
+	wr->r3 = 0;
+
+	cpl = (void *)(wr + 1);
+
+	/* CPL header */
+	cpl->ctrl0 = htonl(TXPKT_OPCODE_V(CPL_TX_PKT) |
+			   TXPKT_INTF_V(tx_info->tx_chan) |
+			   TXPKT_PF_V(tx_info->adap->pf));
+	cpl->pack = 0;
+	cntrl1 = TXPKT_CSUM_TYPE_V(tx_info->ip_family == AF_INET ?
+				   TX_CSUM_TCPIP : TX_CSUM_TCPIP6);
+	cntrl1 |= T6_TXPKT_ETHHDR_LEN_V(maclen - ETH_HLEN) |
+		  TXPKT_IPHDR_LEN_V(iplen);
+	/* checksum offload */
+	cpl->ctrl1 = cpu_to_be64(cntrl1);
+	cpl->len = htons(skb->len);
+
+	pos = cpl + 1;
+
+	cxgb4_write_sgl(skb, &q->q, pos, end, 0, sgl_sdesc->addr);
+	sgl_sdesc->skb = skb;
+	chcr_txq_advance(&q->q, ndesc);
+	skb_tx_timestamp(skb);
+	cxgb4_ring_tx_db(tx_info->adap, &q->q, ndesc);
+	return 0;
+}
+
+/*
+ * chcr_ktls_copy_record_in_skb
+ * @nskb - new skb where the frags to be added.
+ * @skb - old skb, to copy socket and destructor details.
+ * @record - specific record which has complete 16k record in frags.
+ */
+static void chcr_ktls_copy_record_in_skb(struct sk_buff *nskb,
+					 struct sk_buff *skb,
+					 struct tls_record_info *record)
+{
+	int i = 0;
+
+	for (i = 0; i < record->num_frags; i++) {
+		skb_shinfo(nskb)->frags[i] = record->frags[i];
+		/* increase the frag ref count */
+		__skb_frag_ref(&skb_shinfo(nskb)->frags[i]);
+	}
+
+	skb_shinfo(nskb)->nr_frags = record->num_frags;
+	nskb->data_len = record->len;
+	nskb->len += record->len;
+	nskb->truesize += record->len;
+	nskb->sk = skb->sk;
+	nskb->destructor = skb->destructor;
+	refcount_add(nskb->truesize, &nskb->sk->sk_wmem_alloc);
+}
+
+/*
+ * chcr_end_part_handler: This handler will handle the record which
+ * is complete or if record's end part is received. T6 adapter has a issue that
+ * it can't send out TAG with partial record so if its an end part then we have
+ * to send TAG as well and for which we need to fetch the complete record and
+ * send it to crypto module.
+ * @tx_info - driver specific tls info.
+ * @skb - skb contains partial record.
+ * @record - complete record of 16K size.
+ * @tcp_seq
+ * @mss - segment size in which TP needs to chop a packet.
+ * @tcp_push_no_fin - tcp push if fin is not set.
+ * @q - TX queue.
+ * @tls_end_offset - offset from end of the record.
+ * @last wr : check if this is the last part of the skb going out.
+ * return: NETDEV_TX_OK/NETDEV_TX_BUSY.
+ */
+static int chcr_end_part_handler(struct chcr_ktls_info *tx_info,
+				 struct sk_buff *skb,
+				 struct tls_record_info *record,
+				 u32 tcp_seq, int mss, bool tcp_push_no_fin,
+				 struct sge_eth_txq *q, u32 skb_offset,
+				 u32 tls_end_offset, bool last_wr)
+{
+	bool free_skb_if_tx_fails = false;
+	struct sk_buff *nskb = NULL;
+
+	/* check if it is a complete record */
+	if (tls_end_offset == record->len) {
+		nskb = skb;
+		atomic64_inc(&tx_info->adap->ch_ktls_stats.ktls_tx_complete_pkts);
+	} else {
+		nskb = alloc_skb(0, GFP_ATOMIC);
+		if (!nskb) {
+			dev_kfree_skb_any(skb);
+			return NETDEV_TX_BUSY;
+		}
+
+		/* copy complete record in skb */
+		chcr_ktls_copy_record_in_skb(nskb, skb, record);
+		/* packet is being sent from the beginning, update the tcp_seq
+		 * accordingly.
+		 */
+		tcp_seq = tls_record_start_seq(record);
+		/* reset skb offset */
+		skb_offset = 0;
+
+		if (last_wr)
+			dev_kfree_skb_any(skb);
+		else
+			free_skb_if_tx_fails = true;
+
+		last_wr = true;
+
+		atomic64_inc(&tx_info->adap->ch_ktls_stats.ktls_tx_end_pkts);
+	}
+
+	if (chcr_ktls_xmit_wr_complete(nskb, tx_info, q, tcp_seq,
+				       last_wr, record->len, skb_offset,
+				       record->num_frags,
+				       (last_wr && tcp_push_no_fin),
+				       mss)) {
+		if (free_skb_if_tx_fails)
+			dev_kfree_skb_any(skb);
+		goto out;
+	}
+	tx_info->prev_seq = record->end_seq;
+	return 0;
+out:
+	dev_kfree_skb_any(nskb);
+	return NETDEV_TX_BUSY;
+}
+
+/*
+ * chcr_short_record_handler: This handler will take care of the records which
+ * doesn't have end part (1st part or the middle part(/s) of a record). In such
+ * cases, AES CTR will be used in place of AES GCM to send out partial packet.
+ * This partial record might be the first part of the record, or the middle
+ * part. In case of middle record we should fetch the prior data to make it 16
+ * byte aligned. If it has a partial tls header or iv then get to the start of
+ * tls header. And if it has partial TAG, then remove the complete TAG and send
+ * only the payload.
+ * There is one more possibility that it gets a partial header, send that
+ * portion as a plaintext.
+ * @tx_info - driver specific tls info.
+ * @skb - skb contains partial record..
+ * @record - complete record of 16K size.
+ * @tcp_seq
+ * @mss - segment size in which TP needs to chop a packet.
+ * @tcp_push_no_fin - tcp push if fin is not set.
+ * @q - TX queue.
+ * @tls_end_offset - offset from end of the record.
+ * return: NETDEV_TX_OK/NETDEV_TX_BUSY.
+ */
+static int chcr_short_record_handler(struct chcr_ktls_info *tx_info,
+				     struct sk_buff *skb,
+				     struct tls_record_info *record,
+				     u32 tcp_seq, int mss, bool tcp_push_no_fin,
+				     u32 data_len, u32 skb_offset,
+				     struct sge_eth_txq *q, u32 tls_end_offset)
+{
+	u32 tls_rec_offset = tcp_seq - tls_record_start_seq(record);
+	u8 prior_data[16] = {0};
+	u32 prior_data_len = 0;
+
+	/* check if the skb is ending in middle of tag/HASH, its a big
+	 * trouble, send the packet before the HASH.
+	 */
+	int remaining_record = tls_end_offset - data_len;
+
+	if (remaining_record > 0 &&
+	    remaining_record < TLS_CIPHER_AES_GCM_128_TAG_SIZE) {
+		int trimmed_len = 0;
+
+		if (tls_end_offset > TLS_CIPHER_AES_GCM_128_TAG_SIZE)
+			trimmed_len = data_len -
+				      (TLS_CIPHER_AES_GCM_128_TAG_SIZE -
+				       remaining_record);
+		if (!trimmed_len)
+			return FALLBACK;
+
+		WARN_ON(trimmed_len > data_len);
+
+		data_len = trimmed_len;
+		atomic64_inc(&tx_info->adap->ch_ktls_stats.ktls_tx_trimmed_pkts);
+	}
+
+	/* check if it is only the header part. */
+	if (tls_rec_offset + data_len <= (TLS_HEADER_SIZE + tx_info->iv_size)) {
+		if (chcr_ktls_tx_plaintxt(tx_info, skb, tcp_seq, mss,
+					  tcp_push_no_fin, q,
+					  tx_info->port_id, prior_data,
+					  data_len, skb_offset, prior_data_len))
+			goto out;
+
+		tx_info->prev_seq = tcp_seq + data_len;
+		return 0;
+	}
+
+	/* check if the middle record's start point is 16 byte aligned. CTR
+	 * needs 16 byte aligned start point to start encryption.
+	 */
+	if (tls_rec_offset) {
+		/* there is an offset from start, means its a middle record */
+		int remaining = 0;
+
+		if (tls_rec_offset < (TLS_HEADER_SIZE + tx_info->iv_size)) {
+			prior_data_len = tls_rec_offset;
+			tls_rec_offset = 0;
+			remaining = 0;
+		} else {
+			prior_data_len =
+				(tls_rec_offset -
+				(TLS_HEADER_SIZE + tx_info->iv_size))
+				% AES_BLOCK_LEN;
+			remaining = tls_rec_offset - prior_data_len;
+		}
+
+		/* if prior_data_len is not zero, means we need to fetch prior
+		 * data to make this record 16 byte aligned, or we need to reach
+		 * to start offset.
+		 */
+		if (prior_data_len) {
+			int i = 0;
+			skb_frag_t *f;
+			int frag_size = 0, frag_delta = 0;
+
+			while (remaining > 0) {
+				frag_size = skb_frag_size(&record->frags[i]);
+				if (remaining < frag_size)
+					break;
+
+				remaining -= frag_size;
+				i++;
+			}
+			f = &record->frags[i];
+			frag_delta = skb_frag_size(f) - remaining;
+
+			if (frag_delta >= prior_data_len) {
+				memcpy_from_page(prior_data, skb_frag_page(f),
+						 skb_frag_off(f) + remaining,
+						 prior_data_len);
+			} else {
+				memcpy_from_page(prior_data, skb_frag_page(f),
+						 skb_frag_off(f) + remaining,
+						 frag_delta);
+
+				/* get the next page */
+				f = &record->frags[i + 1];
+
+				memcpy_from_page(prior_data + frag_delta,
+						 skb_frag_page(f),
+						 skb_frag_off(f),
+						 prior_data_len - frag_delta);
+			}
+			/* reset tcp_seq as per the prior_data_required len */
+			tcp_seq -= prior_data_len;
+		}
+		atomic64_inc(&tx_info->adap->ch_ktls_stats.ktls_tx_middle_pkts);
+	} else {
+		atomic64_inc(&tx_info->adap->ch_ktls_stats.ktls_tx_start_pkts);
+	}
+
+	if (chcr_ktls_xmit_wr_short(skb, tx_info, q, tcp_seq, tcp_push_no_fin,
+				    mss, tls_rec_offset, prior_data,
+				    prior_data_len, data_len, skb_offset)) {
+		goto out;
+	}
+
+	tx_info->prev_seq = tcp_seq + data_len + prior_data_len;
+	return 0;
+out:
+	dev_kfree_skb_any(skb);
+	return NETDEV_TX_BUSY;
+}
+
+static int chcr_ktls_sw_fallback(struct sk_buff *skb,
+				 struct chcr_ktls_info *tx_info,
+				 struct sge_eth_txq *q)
+{
+	u32 data_len, skb_offset;
+	struct sk_buff *nskb;
+	struct tcphdr *th;
+
+	nskb = tls_encrypt_skb(skb);
+
+	if (!nskb)
+		return 0;
+
+	th = tcp_hdr(nskb);
+	skb_offset = skb_tcp_all_headers(nskb);
+	data_len = nskb->len - skb_offset;
+
+	if (chcr_ktls_tunnel_pkt(tx_info, nskb, q))
+		goto out;
+
+	tx_info->prev_seq = ntohl(th->seq) + data_len;
+	atomic64_inc(&tx_info->adap->ch_ktls_stats.ktls_tx_fallback);
+	return 0;
+out:
+	dev_kfree_skb_any(nskb);
+	return 0;
+}
+/* nic tls TX handler */
+static int chcr_ktls_xmit(struct sk_buff *skb, struct net_device *dev)
+{
+	u32 tls_end_offset, tcp_seq, skb_data_len, skb_offset;
+	struct ch_ktls_port_stats_debug *port_stats;
+	struct tls_offload_context_tx *tx_ctx;
+	struct ch_ktls_stats_debug *stats;
+	struct tcphdr *th = tcp_hdr(skb);
+	int data_len, qidx, ret = 0, mss;
+	struct tls_record_info *record;
+	struct chcr_ktls_info *tx_info;
+	struct net_device *tls_netdev;
+	struct tls_context *tls_ctx;
+	struct sge_eth_txq *q;
+	struct adapter *adap;
+	unsigned long flags;
+
+	tcp_seq = ntohl(th->seq);
+	skb_offset = skb_tcp_all_headers(skb);
+	skb_data_len = skb->len - skb_offset;
+	data_len = skb_data_len;
+
+	mss = skb_is_gso(skb) ? skb_shinfo(skb)->gso_size : data_len;
+
+	tls_ctx = tls_get_ctx(skb->sk);
+	tx_ctx = tls_offload_ctx_tx(tls_ctx);
+	tls_netdev = rcu_dereference_bh(tls_ctx->netdev);
+	/* Don't quit on NULL: if tls_device_down is running in parallel,
+	 * netdev might become NULL, even if tls_is_skb_tx_device_offloaded was
+	 * true. Rather continue processing this packet.
+	 */
+	if (unlikely(tls_netdev && tls_netdev != dev))
+		goto out;
+
+	tx_info = chcr_get_ktls_tx_info(tls_ctx);
+
+	if (unlikely(!tx_info))
+		goto out;
+
+	adap = tx_info->adap;
+	stats = &adap->ch_ktls_stats;
+	port_stats = &stats->ktls_port[tx_info->port_id];
+
+	qidx = skb->queue_mapping;
+	q = &adap->sge.ethtxq[qidx + tx_info->first_qset];
+	cxgb4_reclaim_completed_tx(adap, &q->q, true);
+	/* if tcp options are set but finish is not send the options first */
+	if (!th->fin && chcr_ktls_check_tcp_options(th)) {
+		ret = chcr_ktls_write_tcp_options(tx_info, skb, q,
+						  tx_info->tx_chan);
+		if (ret)
+			return NETDEV_TX_BUSY;
+	}
+
+	/* TCP segments can be in received either complete or partial.
+	 * chcr_end_part_handler will handle cases if complete record or end
+	 * part of the record is received. In case of partial end part of record,
+	 * we will send the complete record again.
+	 */
+
+	spin_lock_irqsave(&tx_ctx->lock, flags);
+
+	do {
+
+		cxgb4_reclaim_completed_tx(adap, &q->q, true);
+		/* fetch the tls record */
+		record = tls_get_record(tx_ctx, tcp_seq,
+					&tx_info->record_no);
+		/* By the time packet reached to us, ACK is received, and record
+		 * won't be found in that case, handle it gracefully.
+		 */
+		if (unlikely(!record)) {
+			spin_unlock_irqrestore(&tx_ctx->lock, flags);
+			atomic64_inc(&port_stats->ktls_tx_drop_no_sync_data);
+			goto out;
+		}
+
+		tls_end_offset = record->end_seq - tcp_seq;
+
+		pr_debug("seq 0x%x, end_seq 0x%x prev_seq 0x%x, datalen 0x%x\n",
+			 tcp_seq, record->end_seq, tx_info->prev_seq, data_len);
+		/* update tcb for the skb */
+		if (skb_data_len == data_len) {
+			u32 tx_max = tcp_seq;
+
+			if (!tls_record_is_start_marker(record) &&
+			    tls_end_offset < TLS_CIPHER_AES_GCM_128_TAG_SIZE)
+				tx_max = record->end_seq -
+					TLS_CIPHER_AES_GCM_128_TAG_SIZE;
+
+			ret = chcr_ktls_xmit_tcb_cpls(tx_info, q, tx_max,
+						      ntohl(th->ack_seq),
+						      ntohs(th->window),
+						      tls_end_offset !=
+						      record->len);
+			if (ret) {
+				spin_unlock_irqrestore(&tx_ctx->lock,
+						       flags);
+				goto out;
+			}
+
+			if (th->fin)
+				skb_get(skb);
+		}
+
+		if (unlikely(tls_record_is_start_marker(record))) {
+			atomic64_inc(&port_stats->ktls_tx_skip_no_sync_data);
+			/* If tls_end_offset < data_len, means there is some
+			 * data after start marker, which needs encryption, send
+			 * plaintext first and take skb refcount. else send out
+			 * complete pkt as plaintext.
+			 */
+			if (tls_end_offset < data_len)
+				skb_get(skb);
+			else
+				tls_end_offset = data_len;
+
+			ret = chcr_ktls_tx_plaintxt(tx_info, skb, tcp_seq, mss,
+						    (!th->fin && th->psh), q,
+						    tx_info->port_id, NULL,
+						    tls_end_offset, skb_offset,
+						    0);
+
+			if (ret) {
+				/* free the refcount taken earlier */
+				if (tls_end_offset < data_len)
+					dev_kfree_skb_any(skb);
+				spin_unlock_irqrestore(&tx_ctx->lock, flags);
+				goto out;
+			}
+
+			data_len -= tls_end_offset;
+			tcp_seq = record->end_seq;
+			skb_offset += tls_end_offset;
+			continue;
+		}
+
+		/* if a tls record is finishing in this SKB */
+		if (tls_end_offset <= data_len) {
+			ret = chcr_end_part_handler(tx_info, skb, record,
+						    tcp_seq, mss,
+						    (!th->fin && th->psh), q,
+						    skb_offset,
+						    tls_end_offset,
+						    skb_offset +
+						    tls_end_offset == skb->len);
+
+			data_len -= tls_end_offset;
+			/* tcp_seq increment is required to handle next record.
+			 */
+			tcp_seq += tls_end_offset;
+			skb_offset += tls_end_offset;
+		} else {
+			ret = chcr_short_record_handler(tx_info, skb,
+							record, tcp_seq, mss,
+							(!th->fin && th->psh),
+							data_len, skb_offset,
+							q, tls_end_offset);
+			data_len = 0;
+		}
+
+		/* if any failure, come out from the loop. */
+		if (ret) {
+			spin_unlock_irqrestore(&tx_ctx->lock, flags);
+			if (th->fin)
+				dev_kfree_skb_any(skb);
+
+			if (ret == FALLBACK)
+				return chcr_ktls_sw_fallback(skb, tx_info, q);
+
+			return NETDEV_TX_OK;
+		}
+
+		/* length should never be less than 0 */
+		WARN_ON(data_len < 0);
+
+	} while (data_len > 0);
+
+	spin_unlock_irqrestore(&tx_ctx->lock, flags);
+	atomic64_inc(&port_stats->ktls_tx_encrypted_packets);
+	atomic64_add(skb_data_len, &port_stats->ktls_tx_encrypted_bytes);
+
+	/* tcp finish is set, send a separate tcp msg including all the options
+	 * as well.
+	 */
+	if (th->fin) {
+		chcr_ktls_write_tcp_options(tx_info, skb, q, tx_info->tx_chan);
+		dev_kfree_skb_any(skb);
+	}
+
+	return NETDEV_TX_OK;
+out:
+	dev_kfree_skb_any(skb);
+	return NETDEV_TX_OK;
+}
+
+static void *chcr_ktls_uld_add(const struct cxgb4_lld_info *lldi)
+{
+	struct chcr_ktls_uld_ctx *u_ctx;
+
+	pr_info_once("%s - version %s\n", CHCR_KTLS_DRV_DESC,
+		     CHCR_KTLS_DRV_VERSION);
+	u_ctx = kzalloc(sizeof(*u_ctx), GFP_KERNEL);
+	if (!u_ctx) {
+		u_ctx = ERR_PTR(-ENOMEM);
+		goto out;
+	}
+	u_ctx->lldi = *lldi;
+	u_ctx->detach = false;
+	xa_init_flags(&u_ctx->tid_list, XA_FLAGS_LOCK_BH);
+out:
+	return u_ctx;
+}
+
+static const struct tlsdev_ops chcr_ktls_ops = {
+	.tls_dev_add = chcr_ktls_dev_add,
+	.tls_dev_del = chcr_ktls_dev_del,
+};
+
+static chcr_handler_func work_handlers[NUM_CPL_CMDS] = {
+	[CPL_ACT_OPEN_RPL] = chcr_ktls_cpl_act_open_rpl,
+	[CPL_SET_TCB_RPL] = chcr_ktls_cpl_set_tcb_rpl,
+};
+
+static int chcr_ktls_uld_rx_handler(void *handle, const __be64 *rsp,
+				    const struct pkt_gl *pgl)
+{
+	const struct cpl_act_open_rpl *rpl = (struct cpl_act_open_rpl *)rsp;
+	struct chcr_ktls_uld_ctx *u_ctx = handle;
+	u8 opcode = rpl->ot.opcode;
+	struct adapter *adap;
+
+	adap = pci_get_drvdata(u_ctx->lldi.pdev);
+
+	if (!work_handlers[opcode]) {
+		pr_err("Unsupported opcode %d received\n", opcode);
+		return 0;
+	}
+
+	work_handlers[opcode](adap, (unsigned char *)&rsp[1]);
+	return 0;
+}
+
+static void clear_conn_resources(struct chcr_ktls_info *tx_info)
+{
+	/* clear l2t entry */
+	if (tx_info->l2te)
+		cxgb4_l2t_release(tx_info->l2te);
+
+#if IS_ENABLED(CONFIG_IPV6)
+	/* clear clip entry */
+	if (tx_info->ip_family == AF_INET6)
+		cxgb4_clip_release(tx_info->netdev, (const u32 *)
+				   &tx_info->sk->sk_v6_rcv_saddr,
+				   1);
+#endif
+
+	/* clear tid */
+	if (tx_info->tid != -1)
+		cxgb4_remove_tid(&tx_info->adap->tids, tx_info->tx_chan,
+				 tx_info->tid, tx_info->ip_family);
+}
+
+static void ch_ktls_reset_all_conn(struct chcr_ktls_uld_ctx *u_ctx)
+{
+	struct ch_ktls_port_stats_debug *port_stats;
+	struct tls_offload_context_tx *tx_ctx;
+	struct chcr_ktls_info *tx_info;
+	unsigned long index;
+
+	xa_for_each(&u_ctx->tid_list, index, tx_ctx) {
+		tx_info = __chcr_get_ktls_tx_info(tx_ctx);
+		clear_conn_resources(tx_info);
+		port_stats = &tx_info->adap->ch_ktls_stats.ktls_port[tx_info->port_id];
+		atomic64_inc(&port_stats->ktls_tx_connection_close);
+		kvfree(tx_info);
+		memset(tx_ctx->driver_state, 0, TLS_DRIVER_STATE_SIZE_TX);
+		/* release module refcount */
+		module_put(THIS_MODULE);
+	}
+}
+
+static int chcr_ktls_uld_state_change(void *handle, enum cxgb4_state new_state)
+{
+	struct chcr_ktls_uld_ctx *u_ctx = handle;
+
+	switch (new_state) {
+	case CXGB4_STATE_UP:
+		pr_info("%s: Up\n", pci_name(u_ctx->lldi.pdev));
+		mutex_lock(&dev_mutex);
+		list_add_tail(&u_ctx->entry, &uld_ctx_list);
+		mutex_unlock(&dev_mutex);
+		break;
+	case CXGB4_STATE_START_RECOVERY:
+	case CXGB4_STATE_DOWN:
+	case CXGB4_STATE_DETACH:
+		pr_info("%s: Down\n", pci_name(u_ctx->lldi.pdev));
+		mutex_lock(&dev_mutex);
+		u_ctx->detach = true;
+		list_del(&u_ctx->entry);
+		ch_ktls_reset_all_conn(u_ctx);
+		xa_destroy(&u_ctx->tid_list);
+		mutex_unlock(&dev_mutex);
+		break;
+	default:
+		break;
+	}
+
+	return 0;
+}
+
+static struct cxgb4_uld_info chcr_ktls_uld_info = {
+	.name = CHCR_KTLS_DRV_MODULE_NAME,
+	.nrxq = 1,
+	.rxq_size = 1024,
+	.add = chcr_ktls_uld_add,
+	.tx_handler = chcr_ktls_xmit,
+	.rx_handler = chcr_ktls_uld_rx_handler,
+	.state_change = chcr_ktls_uld_state_change,
+	.tlsdev_ops = &chcr_ktls_ops,
+};
+
+static int __init chcr_ktls_init(void)
+{
+	cxgb4_register_uld(CXGB4_ULD_KTLS, &chcr_ktls_uld_info);
+	return 0;
+}
+
+static void __exit chcr_ktls_exit(void)
+{
+	struct chcr_ktls_uld_ctx *u_ctx, *tmp;
+	struct adapter *adap;
+
+	pr_info("driver unloaded\n");
+
+	mutex_lock(&dev_mutex);
+	list_for_each_entry_safe(u_ctx, tmp, &uld_ctx_list, entry) {
+		adap = pci_get_drvdata(u_ctx->lldi.pdev);
+		memset(&adap->ch_ktls_stats, 0, sizeof(adap->ch_ktls_stats));
+		list_del(&u_ctx->entry);
+		xa_destroy(&u_ctx->tid_list);
+		kfree(u_ctx);
+	}
+	mutex_unlock(&dev_mutex);
+	cxgb4_unregister_uld(CXGB4_ULD_KTLS);
+}
+
+module_init(chcr_ktls_init);
+module_exit(chcr_ktls_exit);
+
+MODULE_DESCRIPTION("Chelsio NIC TLS ULD driver");
+MODULE_LICENSE("GPL");
+MODULE_AUTHOR("Chelsio Communications");
+MODULE_VERSION(CHCR_KTLS_DRV_VERSION);
diff --git a/drivers/net/ethernet/chelsio/inline_crypto/ch_ktls/chcr_ktls.h b/drivers/net/ethernet/chelsio/inline_crypto/ch_ktls/chcr_ktls.h
new file mode 100644
index 000000000000..dbbba92bf540
--- /dev/null
+++ b/drivers/net/ethernet/chelsio/inline_crypto/ch_ktls/chcr_ktls.h
@@ -0,0 +1,121 @@
+/* SPDX-License-Identifier: GPL-2.0-only */
+/* Copyright (C) 2020 Chelsio Communications.  All rights reserved. */
+
+#ifndef __CHCR_KTLS_H__
+#define __CHCR_KTLS_H__
+
+#include "cxgb4.h"
+#include "t4_msg.h"
+#include "t4_tcb.h"
+#include "l2t.h"
+#include "chcr_common.h"
+#include "cxgb4_uld.h"
+#include "clip_tbl.h"
+
+#define CHCR_KTLS_DRV_MODULE_NAME "ch_ktls"
+#define CHCR_KTLS_DRV_VERSION "1.0.0.0-ko"
+#define CHCR_KTLS_DRV_DESC "Chelsio NIC TLS ULD Driver"
+
+#define CHCR_TCB_STATE_CLOSED	0
+#define CHCR_KTLS_KEY_CTX_LEN	16
+#define CHCR_SET_TCB_FIELD_LEN	sizeof(struct cpl_set_tcb_field)
+#define CHCR_PLAIN_TX_DATA_LEN	(sizeof(struct fw_ulptx_wr) +\
+				 sizeof(struct ulp_txpkt) +\
+				 sizeof(struct ulptx_idata) +\
+				 sizeof(struct cpl_tx_data))
+
+#define CHCR_KTLS_WR_SIZE	(CHCR_PLAIN_TX_DATA_LEN +\
+				 sizeof(struct cpl_tx_sec_pdu))
+#define FALLBACK		35
+
+enum ch_ktls_open_state {
+	CH_KTLS_OPEN_SUCCESS = 0,
+	CH_KTLS_OPEN_PENDING = 1,
+	CH_KTLS_OPEN_FAILURE = 2,
+};
+
+struct chcr_ktls_info {
+	struct sock *sk;
+	spinlock_t lock; /* lock for pending_close */
+	struct ktls_key_ctx key_ctx;
+	struct adapter *adap;
+	struct l2t_entry *l2te;
+	struct net_device *netdev;
+	struct completion completion;
+	u64 iv;
+	u64 record_no;
+	int tid;
+	int atid;
+	int rx_qid;
+	u32 iv_size;
+	u32 prev_seq;
+	u32 prev_ack;
+	u32 salt_size;
+	u32 key_ctx_len;
+	u32 scmd0_seqno_numivs;
+	u32 scmd0_ivgen_hdrlen;
+	u32 tcp_start_seq_number;
+	u32 scmd0_short_seqno_numivs;
+	u32 scmd0_short_ivgen_hdrlen;
+	u16 prev_win;
+	u8 tx_chan;
+	u8 smt_idx;
+	u8 port_id;
+	u8 ip_family;
+	u8 first_qset;
+	enum ch_ktls_open_state open_state;
+	bool pending_close;
+};
+
+struct chcr_ktls_ctx_tx {
+	struct chcr_ktls_info *chcr_info;
+};
+
+struct chcr_ktls_uld_ctx {
+	struct list_head entry;
+	struct cxgb4_lld_info lldi;
+	struct xarray tid_list;
+	bool detach;
+};
+
+static inline struct chcr_ktls_info *
+__chcr_get_ktls_tx_info(struct tls_offload_context_tx *octx)
+{
+	struct chcr_ktls_ctx_tx *priv_ctx;
+
+	BUILD_BUG_ON(sizeof(struct chcr_ktls_ctx_tx) > TLS_DRIVER_STATE_SIZE_TX);
+	priv_ctx = (struct chcr_ktls_ctx_tx *)octx->driver_state;
+	return priv_ctx->chcr_info;
+}
+
+static inline struct chcr_ktls_info *
+chcr_get_ktls_tx_info(struct tls_context *tls_ctx)
+{
+	struct chcr_ktls_ctx_tx *priv_ctx;
+
+	BUILD_BUG_ON(sizeof(struct chcr_ktls_ctx_tx) > TLS_DRIVER_STATE_SIZE_TX);
+	priv_ctx = (struct chcr_ktls_ctx_tx *)__tls_driver_ctx(tls_ctx, TLS_OFFLOAD_CTX_DIR_TX);
+	return priv_ctx->chcr_info;
+}
+
+static inline void
+chcr_set_ktls_tx_info(struct tls_context *tls_ctx, struct chcr_ktls_info *chcr_info)
+{
+	struct chcr_ktls_ctx_tx *priv_ctx;
+
+	priv_ctx = __tls_driver_ctx(tls_ctx, TLS_OFFLOAD_CTX_DIR_TX);
+	priv_ctx->chcr_info = chcr_info;
+}
+
+static inline int chcr_get_first_rx_qid(struct adapter *adap)
+{
+	/* u_ctx is saved in adap, fetch it */
+	struct chcr_ktls_uld_ctx *u_ctx = adap->uld[CXGB4_ULD_KTLS].handle;
+
+	if (!u_ctx)
+		return -1;
+	return u_ctx->lldi.rxq_ids[0];
+}
+
+typedef int (*chcr_handler_func)(struct adapter *adap, unsigned char *input);
+#endif /* __CHCR_KTLS_H__ */
diff --git a/drivers/net/ethernet/chelsio/inline_crypto/chtls/Makefile b/drivers/net/ethernet/chelsio/inline_crypto/chtls/Makefile
new file mode 100644
index 000000000000..bc11495acdb3
--- /dev/null
+++ b/drivers/net/ethernet/chelsio/inline_crypto/chtls/Makefile
@@ -0,0 +1,6 @@
+# SPDX-License-Identifier: GPL-2.0-only
+ccflags-y := -I $(srctree)/drivers/net/ethernet/chelsio/cxgb4 \
+	     -I $(srctree)/drivers/crypto/chelsio
+
+obj-$(CONFIG_CRYPTO_DEV_CHELSIO_TLS) += chtls.o
+chtls-objs := chtls_main.o chtls_cm.o chtls_io.o chtls_hw.o
diff --git a/drivers/net/ethernet/chelsio/inline_crypto/chtls/chtls.h b/drivers/net/ethernet/chelsio/inline_crypto/chtls/chtls.h
new file mode 100644
index 000000000000..21e0dfeff158
--- /dev/null
+++ b/drivers/net/ethernet/chelsio/inline_crypto/chtls/chtls.h
@@ -0,0 +1,584 @@
+/* SPDX-License-Identifier: GPL-2.0-only */
+/*
+ * Copyright (c) 2018 Chelsio Communications, Inc.
+ */
+
+#ifndef __CHTLS_H__
+#define __CHTLS_H__
+
+#include <crypto/aes.h>
+#include <crypto/hash.h>
+#include <crypto/sha1.h>
+#include <crypto/sha2.h>
+#include <crypto/authenc.h>
+#include <crypto/ctr.h>
+#include <crypto/gf128mul.h>
+#include <crypto/internal/aead.h>
+#include <crypto/null.h>
+#include <crypto/internal/skcipher.h>
+#include <crypto/aead.h>
+#include <crypto/scatterwalk.h>
+#include <crypto/internal/hash.h>
+#include <linux/tls.h>
+#include <net/tls.h>
+#include <net/tls_prot.h>
+#include <net/tls_toe.h>
+
+#include "t4fw_api.h"
+#include "t4_msg.h"
+#include "cxgb4.h"
+#include "cxgb4_uld.h"
+#include "l2t.h"
+#include "chcr_algo.h"
+#include "chcr_core.h"
+#include "chcr_crypto.h"
+
+#define CHTLS_DRV_VERSION "1.0.0.0-ko"
+
+#define TLS_KEYCTX_RXFLIT_CNT_S 24
+#define TLS_KEYCTX_RXFLIT_CNT_V(x) ((x) << TLS_KEYCTX_RXFLIT_CNT_S)
+
+#define TLS_KEYCTX_RXPROT_VER_S 20
+#define TLS_KEYCTX_RXPROT_VER_M 0xf
+#define TLS_KEYCTX_RXPROT_VER_V(x) ((x) << TLS_KEYCTX_RXPROT_VER_S)
+
+#define TLS_KEYCTX_RXCIPH_MODE_S 16
+#define TLS_KEYCTX_RXCIPH_MODE_M 0xf
+#define TLS_KEYCTX_RXCIPH_MODE_V(x) ((x) << TLS_KEYCTX_RXCIPH_MODE_S)
+
+#define TLS_KEYCTX_RXAUTH_MODE_S 12
+#define TLS_KEYCTX_RXAUTH_MODE_M 0xf
+#define TLS_KEYCTX_RXAUTH_MODE_V(x) ((x) << TLS_KEYCTX_RXAUTH_MODE_S)
+
+#define TLS_KEYCTX_RXCIAU_CTRL_S 11
+#define TLS_KEYCTX_RXCIAU_CTRL_V(x) ((x) << TLS_KEYCTX_RXCIAU_CTRL_S)
+
+#define TLS_KEYCTX_RX_SEQCTR_S 9
+#define TLS_KEYCTX_RX_SEQCTR_M 0x3
+#define TLS_KEYCTX_RX_SEQCTR_V(x) ((x) << TLS_KEYCTX_RX_SEQCTR_S)
+
+#define TLS_KEYCTX_RX_VALID_S 8
+#define TLS_KEYCTX_RX_VALID_V(x) ((x) << TLS_KEYCTX_RX_VALID_S)
+
+#define TLS_KEYCTX_RXCK_SIZE_S 3
+#define TLS_KEYCTX_RXCK_SIZE_M 0x7
+#define TLS_KEYCTX_RXCK_SIZE_V(x) ((x) << TLS_KEYCTX_RXCK_SIZE_S)
+
+#define TLS_KEYCTX_RXMK_SIZE_S 0
+#define TLS_KEYCTX_RXMK_SIZE_M 0x7
+#define TLS_KEYCTX_RXMK_SIZE_V(x) ((x) << TLS_KEYCTX_RXMK_SIZE_S)
+
+#define KEYCTX_TX_WR_IV_S  55
+#define KEYCTX_TX_WR_IV_M  0x1ffULL
+#define KEYCTX_TX_WR_IV_V(x) ((x) << KEYCTX_TX_WR_IV_S)
+#define KEYCTX_TX_WR_IV_G(x) \
+	(((x) >> KEYCTX_TX_WR_IV_S) & KEYCTX_TX_WR_IV_M)
+
+#define KEYCTX_TX_WR_AAD_S 47
+#define KEYCTX_TX_WR_AAD_M 0xffULL
+#define KEYCTX_TX_WR_AAD_V(x) ((x) << KEYCTX_TX_WR_AAD_S)
+#define KEYCTX_TX_WR_AAD_G(x) (((x) >> KEYCTX_TX_WR_AAD_S) & \
+				KEYCTX_TX_WR_AAD_M)
+
+#define KEYCTX_TX_WR_AADST_S 39
+#define KEYCTX_TX_WR_AADST_M 0xffULL
+#define KEYCTX_TX_WR_AADST_V(x) ((x) << KEYCTX_TX_WR_AADST_S)
+#define KEYCTX_TX_WR_AADST_G(x) \
+	(((x) >> KEYCTX_TX_WR_AADST_S) & KEYCTX_TX_WR_AADST_M)
+
+#define KEYCTX_TX_WR_CIPHER_S 30
+#define KEYCTX_TX_WR_CIPHER_M 0x1ffULL
+#define KEYCTX_TX_WR_CIPHER_V(x) ((x) << KEYCTX_TX_WR_CIPHER_S)
+#define KEYCTX_TX_WR_CIPHER_G(x) \
+	(((x) >> KEYCTX_TX_WR_CIPHER_S) & KEYCTX_TX_WR_CIPHER_M)
+
+#define KEYCTX_TX_WR_CIPHERST_S 23
+#define KEYCTX_TX_WR_CIPHERST_M 0x7f
+#define KEYCTX_TX_WR_CIPHERST_V(x) ((x) << KEYCTX_TX_WR_CIPHERST_S)
+#define KEYCTX_TX_WR_CIPHERST_G(x) \
+	(((x) >> KEYCTX_TX_WR_CIPHERST_S) & KEYCTX_TX_WR_CIPHERST_M)
+
+#define KEYCTX_TX_WR_AUTH_S 14
+#define KEYCTX_TX_WR_AUTH_M 0x1ff
+#define KEYCTX_TX_WR_AUTH_V(x) ((x) << KEYCTX_TX_WR_AUTH_S)
+#define KEYCTX_TX_WR_AUTH_G(x) \
+	(((x) >> KEYCTX_TX_WR_AUTH_S) & KEYCTX_TX_WR_AUTH_M)
+
+#define KEYCTX_TX_WR_AUTHST_S 7
+#define KEYCTX_TX_WR_AUTHST_M 0x7f
+#define KEYCTX_TX_WR_AUTHST_V(x) ((x) << KEYCTX_TX_WR_AUTHST_S)
+#define KEYCTX_TX_WR_AUTHST_G(x) \
+	(((x) >> KEYCTX_TX_WR_AUTHST_S) & KEYCTX_TX_WR_AUTHST_M)
+
+#define KEYCTX_TX_WR_AUTHIN_S 0
+#define KEYCTX_TX_WR_AUTHIN_M 0x7f
+#define KEYCTX_TX_WR_AUTHIN_V(x) ((x) << KEYCTX_TX_WR_AUTHIN_S)
+#define KEYCTX_TX_WR_AUTHIN_G(x) \
+	(((x) >> KEYCTX_TX_WR_AUTHIN_S) & KEYCTX_TX_WR_AUTHIN_M)
+
+struct sge_opaque_hdr {
+	void *dev;
+	dma_addr_t addr[MAX_SKB_FRAGS + 1];
+};
+
+#define MAX_IVS_PAGE			256
+#define TLS_KEY_CONTEXT_SZ		64
+#define CIPHER_BLOCK_SIZE		16
+#define GCM_TAG_SIZE			16
+#define KEY_ON_MEM_SZ			16
+#define AEAD_EXPLICIT_DATA_SIZE		8
+#define TLS_HEADER_LENGTH		5
+#define SCMD_CIPH_MODE_AES_GCM		2
+/* Any MFS size should work and come from openssl */
+#define TLS_MFS				16384
+
+#define RSS_HDR sizeof(struct rss_header)
+#define TLS_WR_CPL_LEN \
+	(sizeof(struct fw_tlstx_data_wr) + sizeof(struct cpl_tx_tls_sfo))
+
+enum {
+	CHTLS_KEY_CONTEXT_DSGL,
+	CHTLS_KEY_CONTEXT_IMM,
+	CHTLS_KEY_CONTEXT_DDR,
+};
+
+enum {
+	CHTLS_LISTEN_START,
+	CHTLS_LISTEN_STOP,
+};
+
+/* Flags for return value of CPL message handlers */
+enum {
+	CPL_RET_BUF_DONE =    1,   /* buffer processing done */
+	CPL_RET_BAD_MSG =     2,   /* bad CPL message */
+	CPL_RET_UNKNOWN_TID = 4    /* unexpected unknown TID */
+};
+
+#define LISTEN_INFO_HASH_SIZE 32
+#define RSPQ_HASH_BITS 5
+struct listen_info {
+	struct listen_info *next;  /* Link to next entry */
+	struct sock *sk;           /* The listening socket */
+	unsigned int stid;         /* The server TID */
+};
+
+enum {
+	T4_LISTEN_START_PENDING,
+	T4_LISTEN_STARTED
+};
+
+enum csk_flags {
+	CSK_CALLBACKS_CHKD,	/* socket callbacks have been sanitized */
+	CSK_ABORT_REQ_RCVD,	/* received one ABORT_REQ_RSS message */
+	CSK_TX_MORE_DATA,	/* sending ULP data; don't set SHOVE bit */
+	CSK_TX_WAIT_IDLE,	/* suspend Tx until in-flight data is ACKed */
+	CSK_ABORT_SHUTDOWN,	/* shouldn't send more abort requests */
+	CSK_ABORT_RPL_PENDING,	/* expecting an abort reply */
+	CSK_CLOSE_CON_REQUESTED,/* we've sent a close_conn_req */
+	CSK_TX_DATA_SENT,	/* sent a TX_DATA WR on this connection */
+	CSK_TX_FAILOVER,	/* Tx traffic failing over */
+	CSK_UPDATE_RCV_WND,	/* Need to update rcv window */
+	CSK_RST_ABORTED,	/* outgoing RST was aborted */
+	CSK_TLS_HANDSHK,	/* TLS Handshake */
+	CSK_CONN_INLINE,	/* Connection on HW */
+};
+
+enum chtls_cdev_state {
+	CHTLS_CDEV_STATE_UP = 1
+};
+
+struct listen_ctx {
+	struct sock *lsk;
+	struct chtls_dev *cdev;
+	struct sk_buff_head synq;
+	u32 state;
+};
+
+struct key_map {
+	unsigned long *addr;
+	unsigned int start;
+	unsigned int available;
+	unsigned int size;
+	spinlock_t lock; /* lock for key id request from map */
+} __packed;
+
+struct tls_scmd {
+	u32 seqno_numivs;
+	u32 ivgen_hdrlen;
+};
+
+struct chtls_dev {
+	struct tls_toe_device tlsdev;
+	struct list_head list;
+	struct cxgb4_lld_info *lldi;
+	struct pci_dev *pdev;
+	struct listen_info *listen_hash_tab[LISTEN_INFO_HASH_SIZE];
+	spinlock_t listen_lock; /* lock for listen list */
+	struct net_device **ports;
+	struct tid_info *tids;
+	unsigned int pfvf;
+	const unsigned short *mtus;
+
+	struct idr hwtid_idr;
+	struct idr stid_idr;
+
+	spinlock_t idr_lock ____cacheline_aligned_in_smp;
+
+	struct net_device *egr_dev[NCHAN * 2];
+	struct sk_buff *rspq_skb_cache[1 << RSPQ_HASH_BITS];
+	struct sk_buff *askb;
+
+	struct sk_buff_head deferq;
+	struct work_struct deferq_task;
+
+	struct list_head list_node;
+	struct list_head rcu_node;
+	struct list_head na_node;
+	unsigned int send_page_order;
+	int max_host_sndbuf;
+	u32 round_robin_cnt;
+	struct key_map kmap;
+	unsigned int cdev_state;
+};
+
+struct chtls_listen {
+	struct chtls_dev *cdev;
+	struct sock *sk;
+};
+
+struct chtls_hws {
+	struct sk_buff_head sk_recv_queue;
+	u8 txqid;
+	u8 ofld;
+	u16 type;
+	u16 rstate;
+	u16 keyrpl;
+	u16 pldlen;
+	u16 rcvpld;
+	u16 compute;
+	u16 expansion;
+	u16 keylen;
+	u16 pdus;
+	u16 adjustlen;
+	u16 ivsize;
+	u16 txleft;
+	u32 mfs;
+	s32 txkey;
+	s32 rxkey;
+	u32 fcplenmax;
+	u32 copied_seq;
+	u64 tx_seq_no;
+	struct tls_scmd scmd;
+	union {
+		struct tls12_crypto_info_aes_gcm_128 aes_gcm_128;
+		struct tls12_crypto_info_aes_gcm_256 aes_gcm_256;
+	} crypto_info;
+};
+
+struct chtls_sock {
+	struct sock *sk;
+	struct chtls_dev *cdev;
+	struct l2t_entry *l2t_entry;    /* pointer to the L2T entry */
+	struct net_device *egress_dev;  /* TX_CHAN for act open retry */
+
+	struct sk_buff_head txq;
+	struct sk_buff *wr_skb_head;
+	struct sk_buff *wr_skb_tail;
+	struct sk_buff *ctrl_skb_cache;
+	struct sk_buff *txdata_skb_cache; /* abort path messages */
+	struct kref kref;
+	unsigned long flags;
+	u32 opt2;
+	u32 wr_credits;
+	u32 wr_unacked;
+	u32 wr_max_credits;
+	u32 wr_nondata;
+	u32 hwtid;               /* TCP Control Block ID */
+	u32 txq_idx;
+	u32 rss_qid;
+	u32 tid;
+	u32 idr;
+	u32 mss;
+	u32 ulp_mode;
+	u32 tx_chan;
+	u32 rx_chan;
+	u32 sndbuf;
+	u32 txplen_max;
+	u32 mtu_idx;           /* MTU table index */
+	u32 smac_idx;
+	u8 port_id;
+	u8 tos;
+	u16 resv2;
+	u32 delack_mode;
+	u32 delack_seq;
+	u32 snd_win;
+	u32 rcv_win;
+
+	void *passive_reap_next;        /* placeholder for passive */
+	struct chtls_hws tlshws;
+	struct synq {
+		struct sk_buff *next;
+		struct sk_buff *prev;
+	} synq;
+	struct listen_ctx *listen_ctx;
+};
+
+struct tls_hdr {
+	u8  type;
+	u16 version;
+	u16 length;
+} __packed;
+
+struct tlsrx_cmp_hdr {
+	u8  type;
+	u16 version;
+	u16 length;
+
+	u64 tls_seq;
+	u16 reserved1;
+	u8  res_to_mac_error;
+} __packed;
+
+/* res_to_mac_error fields */
+#define TLSRX_HDR_PKT_INT_ERROR_S   4
+#define TLSRX_HDR_PKT_INT_ERROR_M   0x1
+#define TLSRX_HDR_PKT_INT_ERROR_V(x) \
+	((x) << TLSRX_HDR_PKT_INT_ERROR_S)
+#define TLSRX_HDR_PKT_INT_ERROR_G(x) \
+	(((x) >> TLSRX_HDR_PKT_INT_ERROR_S) & TLSRX_HDR_PKT_INT_ERROR_M)
+#define TLSRX_HDR_PKT_INT_ERROR_F   TLSRX_HDR_PKT_INT_ERROR_V(1U)
+
+#define TLSRX_HDR_PKT_SPP_ERROR_S        3
+#define TLSRX_HDR_PKT_SPP_ERROR_M        0x1
+#define TLSRX_HDR_PKT_SPP_ERROR_V(x)     ((x) << TLSRX_HDR_PKT_SPP_ERROR)
+#define TLSRX_HDR_PKT_SPP_ERROR_G(x)     \
+	(((x) >> TLSRX_HDR_PKT_SPP_ERROR_S) & TLSRX_HDR_PKT_SPP_ERROR_M)
+#define TLSRX_HDR_PKT_SPP_ERROR_F        TLSRX_HDR_PKT_SPP_ERROR_V(1U)
+
+#define TLSRX_HDR_PKT_CCDX_ERROR_S       2
+#define TLSRX_HDR_PKT_CCDX_ERROR_M       0x1
+#define TLSRX_HDR_PKT_CCDX_ERROR_V(x)    ((x) << TLSRX_HDR_PKT_CCDX_ERROR_S)
+#define TLSRX_HDR_PKT_CCDX_ERROR_G(x)    \
+	(((x) >> TLSRX_HDR_PKT_CCDX_ERROR_S) & TLSRX_HDR_PKT_CCDX_ERROR_M)
+#define TLSRX_HDR_PKT_CCDX_ERROR_F       TLSRX_HDR_PKT_CCDX_ERROR_V(1U)
+
+#define TLSRX_HDR_PKT_PAD_ERROR_S        1
+#define TLSRX_HDR_PKT_PAD_ERROR_M        0x1
+#define TLSRX_HDR_PKT_PAD_ERROR_V(x)     ((x) << TLSRX_HDR_PKT_PAD_ERROR_S)
+#define TLSRX_HDR_PKT_PAD_ERROR_G(x)     \
+	(((x) >> TLSRX_HDR_PKT_PAD_ERROR_S) & TLSRX_HDR_PKT_PAD_ERROR_M)
+#define TLSRX_HDR_PKT_PAD_ERROR_F        TLSRX_HDR_PKT_PAD_ERROR_V(1U)
+
+#define TLSRX_HDR_PKT_MAC_ERROR_S        0
+#define TLSRX_HDR_PKT_MAC_ERROR_M        0x1
+#define TLSRX_HDR_PKT_MAC_ERROR_V(x)     ((x) << TLSRX_HDR_PKT_MAC_ERROR)
+#define TLSRX_HDR_PKT_MAC_ERROR_G(x)     \
+	(((x) >> S_TLSRX_HDR_PKT_MAC_ERROR_S) & TLSRX_HDR_PKT_MAC_ERROR_M)
+#define TLSRX_HDR_PKT_MAC_ERROR_F        TLSRX_HDR_PKT_MAC_ERROR_V(1U)
+
+#define TLSRX_HDR_PKT_ERROR_M           0x1F
+#define CONTENT_TYPE_ERROR		0x7F
+
+struct ulp_mem_rw {
+	__be32 cmd;
+	__be32 len16;             /* command length */
+	__be32 dlen;              /* data length in 32-byte units */
+	__be32 lock_addr;
+};
+
+struct tls_key_wr {
+	__be32 op_to_compl;
+	__be32 flowid_len16;
+	__be32 ftid;
+	u8   reneg_to_write_rx;
+	u8   protocol;
+	__be16 mfs;
+};
+
+struct tls_key_req {
+	struct tls_key_wr wr;
+	struct ulp_mem_rw req;
+	struct ulptx_idata sc_imm;
+};
+
+/*
+ * This lives in skb->cb and is used to chain WRs in a linked list.
+ */
+struct wr_skb_cb {
+	struct l2t_skb_cb l2t;          /* reserve space for l2t CB */
+	struct sk_buff *next_wr;        /* next write request */
+};
+
+/* Per-skb backlog handler.  Run when a socket's backlog is processed. */
+struct blog_skb_cb {
+	void (*backlog_rcv)(struct sock *sk, struct sk_buff *skb);
+	struct chtls_dev *cdev;
+};
+
+/*
+ * Similar to tcp_skb_cb but with ULP elements added to support TLS,
+ * etc.
+ */
+struct ulp_skb_cb {
+	struct wr_skb_cb wr;		/* reserve space for write request */
+	u16 flags;			/* TCP-like flags */
+	u8 psh;
+	u8 ulp_mode;			/* ULP mode/submode of sk_buff */
+	u32 seq;			/* TCP sequence number */
+	union { /* ULP-specific fields */
+		struct {
+			u8  type;
+			u8  ofld;
+			u8  iv;
+		} tls;
+	} ulp;
+};
+
+#define ULP_SKB_CB(skb) ((struct ulp_skb_cb *)&((skb)->cb[0]))
+#define BLOG_SKB_CB(skb) ((struct blog_skb_cb *)(skb)->cb)
+
+/*
+ * Flags for ulp_skb_cb.flags.
+ */
+enum {
+	ULPCB_FLAG_NEED_HDR  = 1 << 0,	/* packet needs a TX_DATA_WR header */
+	ULPCB_FLAG_NO_APPEND = 1 << 1,	/* don't grow this skb */
+	ULPCB_FLAG_BARRIER   = 1 << 2,	/* set TX_WAIT_IDLE after sending */
+	ULPCB_FLAG_HOLD      = 1 << 3,	/* skb not ready for Tx yet */
+	ULPCB_FLAG_COMPL     = 1 << 4,	/* request WR completion */
+	ULPCB_FLAG_URG       = 1 << 5,	/* urgent data */
+	ULPCB_FLAG_TLS_HDR   = 1 << 6,  /* payload with tls hdr */
+	ULPCB_FLAG_NO_HDR    = 1 << 7,  /* not a ofld wr */
+};
+
+/* The ULP mode/submode of an skbuff */
+#define skb_ulp_mode(skb)  (ULP_SKB_CB(skb)->ulp_mode)
+#define TCP_PAGE(sk)   (sk->sk_frag.page)
+#define TCP_OFF(sk)    (sk->sk_frag.offset)
+
+static inline struct chtls_dev *to_chtls_dev(struct tls_toe_device *tlsdev)
+{
+	return container_of(tlsdev, struct chtls_dev, tlsdev);
+}
+
+static inline void csk_set_flag(struct chtls_sock *csk,
+				enum csk_flags flag)
+{
+	__set_bit(flag, &csk->flags);
+}
+
+static inline void csk_reset_flag(struct chtls_sock *csk,
+				  enum csk_flags flag)
+{
+	__clear_bit(flag, &csk->flags);
+}
+
+static inline bool csk_conn_inline(const struct chtls_sock *csk)
+{
+	return test_bit(CSK_CONN_INLINE, &csk->flags);
+}
+
+static inline int csk_flag(const struct sock *sk, enum csk_flags flag)
+{
+	struct chtls_sock *csk = rcu_dereference_sk_user_data(sk);
+
+	if (!csk_conn_inline(csk))
+		return 0;
+	return test_bit(flag, &csk->flags);
+}
+
+static inline int csk_flag_nochk(const struct chtls_sock *csk,
+				 enum csk_flags flag)
+{
+	return test_bit(flag, &csk->flags);
+}
+
+static inline void *cplhdr(struct sk_buff *skb)
+{
+	return skb->data;
+}
+
+static inline int is_neg_adv(unsigned int status)
+{
+	return status == CPL_ERR_RTX_NEG_ADVICE ||
+	       status == CPL_ERR_KEEPALV_NEG_ADVICE ||
+	       status == CPL_ERR_PERSIST_NEG_ADVICE;
+}
+
+static inline void process_cpl_msg(void (*fn)(struct sock *, struct sk_buff *),
+				   struct sock *sk,
+				   struct sk_buff *skb)
+{
+	skb_reset_mac_header(skb);
+	skb_reset_network_header(skb);
+	skb_reset_transport_header(skb);
+
+	bh_lock_sock(sk);
+	if (unlikely(sock_owned_by_user(sk))) {
+		BLOG_SKB_CB(skb)->backlog_rcv = fn;
+		__sk_add_backlog(sk, skb);
+	} else {
+		fn(sk, skb);
+	}
+	bh_unlock_sock(sk);
+}
+
+static inline void chtls_sock_free(struct kref *ref)
+{
+	struct chtls_sock *csk = container_of(ref, struct chtls_sock,
+					      kref);
+	kfree(csk);
+}
+
+static inline void __chtls_sock_put(const char *fn, struct chtls_sock *csk)
+{
+	kref_put(&csk->kref, chtls_sock_free);
+}
+
+static inline void __chtls_sock_get(const char *fn,
+				    struct chtls_sock *csk)
+{
+	kref_get(&csk->kref);
+}
+
+static inline void send_or_defer(struct sock *sk, struct tcp_sock *tp,
+				 struct sk_buff *skb, int through_l2t)
+{
+	struct chtls_sock *csk = rcu_dereference_sk_user_data(sk);
+
+	if (through_l2t) {
+		/* send through L2T */
+		cxgb4_l2t_send(csk->egress_dev, skb, csk->l2t_entry);
+	} else {
+		/* send directly */
+		cxgb4_ofld_send(csk->egress_dev, skb);
+	}
+}
+
+typedef int (*chtls_handler_func)(struct chtls_dev *, struct sk_buff *);
+extern chtls_handler_func chtls_handlers[NUM_CPL_CMDS];
+void chtls_install_cpl_ops(struct sock *sk);
+int chtls_init_kmap(struct chtls_dev *cdev, struct cxgb4_lld_info *lldi);
+void chtls_listen_stop(struct chtls_dev *cdev, struct sock *sk);
+int chtls_listen_start(struct chtls_dev *cdev, struct sock *sk);
+void chtls_close(struct sock *sk, long timeout);
+int chtls_disconnect(struct sock *sk, int flags);
+void chtls_shutdown(struct sock *sk, int how);
+void chtls_destroy_sock(struct sock *sk);
+int chtls_sendmsg(struct sock *sk, struct msghdr *msg, size_t size);
+int chtls_recvmsg(struct sock *sk, struct msghdr *msg,
+		  size_t len, int flags, int *addr_len);
+void chtls_splice_eof(struct socket *sock);
+int send_tx_flowc_wr(struct sock *sk, int compl,
+		     u32 snd_nxt, u32 rcv_nxt);
+void chtls_tcp_push(struct sock *sk, int flags);
+int chtls_push_frames(struct chtls_sock *csk, int comp);
+void chtls_set_tcb_field_rpl_skb(struct sock *sk, u16 word,
+				 u64 mask, u64 val, u8 cookie,
+				 int through_l2t);
+int chtls_setkey(struct chtls_sock *csk, u32 keylen, u32 mode, int cipher_type);
+void chtls_set_quiesce_ctrl(struct sock *sk, int val);
+void skb_entail(struct sock *sk, struct sk_buff *skb, int flags);
+unsigned int keyid_to_addr(int start_addr, int keyid);
+void free_tls_keyid(struct sock *sk);
+#endif
diff --git a/drivers/net/ethernet/chelsio/inline_crypto/chtls/chtls_cm.c b/drivers/net/ethernet/chelsio/inline_crypto/chtls/chtls_cm.c
new file mode 100644
index 000000000000..4ee970f3bad6
--- /dev/null
+++ b/drivers/net/ethernet/chelsio/inline_crypto/chtls/chtls_cm.c
@@ -0,0 +1,2336 @@
+// SPDX-License-Identifier: GPL-2.0-only
+/*
+ * Copyright (c) 2018 Chelsio Communications, Inc.
+ *
+ * Written by: Atul Gupta (atul.gupta@chelsio.com)
+ */
+
+#include <linux/module.h>
+#include <linux/list.h>
+#include <linux/workqueue.h>
+#include <linux/skbuff.h>
+#include <linux/timer.h>
+#include <linux/notifier.h>
+#include <linux/inetdevice.h>
+#include <linux/ip.h>
+#include <linux/tcp.h>
+#include <linux/sched/signal.h>
+#include <linux/kallsyms.h>
+#include <linux/kprobes.h>
+#include <linux/if_vlan.h>
+#include <linux/ipv6.h>
+#include <net/ipv6.h>
+#include <net/transp_v6.h>
+#include <net/ip6_route.h>
+#include <net/inet_common.h>
+#include <net/tcp.h>
+#include <net/dst.h>
+#include <net/tls.h>
+#include <net/addrconf.h>
+#include <net/secure_seq.h>
+
+#include "chtls.h"
+#include "chtls_cm.h"
+#include "clip_tbl.h"
+#include "t4_tcb.h"
+
+/*
+ * State transitions and actions for close.  Note that if we are in SYN_SENT
+ * we remain in that state as we cannot control a connection while it's in
+ * SYN_SENT; such connections are allowed to establish and are then aborted.
+ */
+static unsigned char new_state[16] = {
+	/* current state:     new state:      action: */
+	/* (Invalid)       */ TCP_CLOSE,
+	/* TCP_ESTABLISHED */ TCP_FIN_WAIT1 | TCP_ACTION_FIN,
+	/* TCP_SYN_SENT    */ TCP_SYN_SENT,
+	/* TCP_SYN_RECV    */ TCP_FIN_WAIT1 | TCP_ACTION_FIN,
+	/* TCP_FIN_WAIT1   */ TCP_FIN_WAIT1,
+	/* TCP_FIN_WAIT2   */ TCP_FIN_WAIT2,
+	/* TCP_TIME_WAIT   */ TCP_CLOSE,
+	/* TCP_CLOSE       */ TCP_CLOSE,
+	/* TCP_CLOSE_WAIT  */ TCP_LAST_ACK | TCP_ACTION_FIN,
+	/* TCP_LAST_ACK    */ TCP_LAST_ACK,
+	/* TCP_LISTEN      */ TCP_CLOSE,
+	/* TCP_CLOSING     */ TCP_CLOSING,
+};
+
+static struct chtls_sock *chtls_sock_create(struct chtls_dev *cdev)
+{
+	struct chtls_sock *csk = kzalloc(sizeof(*csk), GFP_ATOMIC);
+
+	if (!csk)
+		return NULL;
+
+	csk->txdata_skb_cache = alloc_skb(TXDATA_SKB_LEN, GFP_ATOMIC);
+	if (!csk->txdata_skb_cache) {
+		kfree(csk);
+		return NULL;
+	}
+
+	kref_init(&csk->kref);
+	csk->cdev = cdev;
+	skb_queue_head_init(&csk->txq);
+	csk->wr_skb_head = NULL;
+	csk->wr_skb_tail = NULL;
+	csk->mss = MAX_MSS;
+	csk->tlshws.ofld = 1;
+	csk->tlshws.txkey = -1;
+	csk->tlshws.rxkey = -1;
+	csk->tlshws.mfs = TLS_MFS;
+	skb_queue_head_init(&csk->tlshws.sk_recv_queue);
+	return csk;
+}
+
+static void chtls_sock_release(struct kref *ref)
+{
+	struct chtls_sock *csk =
+		container_of(ref, struct chtls_sock, kref);
+
+	kfree(csk);
+}
+
+static struct net_device *chtls_find_netdev(struct chtls_dev *cdev,
+					    struct sock *sk)
+{
+	struct adapter *adap = pci_get_drvdata(cdev->pdev);
+	struct net_device *ndev = cdev->ports[0];
+#if IS_ENABLED(CONFIG_IPV6)
+	struct net_device *temp;
+	int addr_type;
+#endif
+	int i;
+
+	switch (sk->sk_family) {
+	case PF_INET:
+		if (likely(!inet_sk(sk)->inet_rcv_saddr))
+			return ndev;
+		ndev = __ip_dev_find(&init_net, inet_sk(sk)->inet_rcv_saddr, false);
+		break;
+#if IS_ENABLED(CONFIG_IPV6)
+	case PF_INET6:
+		addr_type = ipv6_addr_type(&sk->sk_v6_rcv_saddr);
+		if (likely(addr_type == IPV6_ADDR_ANY))
+			return ndev;
+
+		for_each_netdev_rcu(&init_net, temp) {
+			if (ipv6_chk_addr(&init_net, (struct in6_addr *)
+					  &sk->sk_v6_rcv_saddr, temp, 1)) {
+				ndev = temp;
+				break;
+			}
+		}
+	break;
+#endif
+	default:
+		return NULL;
+	}
+
+	if (!ndev)
+		return NULL;
+
+	if (is_vlan_dev(ndev))
+		ndev = vlan_dev_real_dev(ndev);
+
+	for_each_port(adap, i)
+		if (cdev->ports[i] == ndev)
+			return ndev;
+	return NULL;
+}
+
+static void assign_rxopt(struct sock *sk, unsigned int opt)
+{
+	const struct chtls_dev *cdev;
+	struct chtls_sock *csk;
+	struct tcp_sock *tp;
+
+	csk = rcu_dereference_sk_user_data(sk);
+	tp = tcp_sk(sk);
+
+	cdev = csk->cdev;
+	tp->tcp_header_len           = sizeof(struct tcphdr);
+	tp->rx_opt.mss_clamp         = cdev->mtus[TCPOPT_MSS_G(opt)] - 40;
+	tp->mss_cache                = tp->rx_opt.mss_clamp;
+	tp->rx_opt.tstamp_ok         = TCPOPT_TSTAMP_G(opt);
+	tp->rx_opt.snd_wscale        = TCPOPT_SACK_G(opt);
+	tp->rx_opt.wscale_ok         = TCPOPT_WSCALE_OK_G(opt);
+	SND_WSCALE(tp)               = TCPOPT_SND_WSCALE_G(opt);
+	if (!tp->rx_opt.wscale_ok)
+		tp->rx_opt.rcv_wscale = 0;
+	if (tp->rx_opt.tstamp_ok) {
+		tp->tcp_header_len += TCPOLEN_TSTAMP_ALIGNED;
+		tp->rx_opt.mss_clamp -= TCPOLEN_TSTAMP_ALIGNED;
+	} else if (csk->opt2 & TSTAMPS_EN_F) {
+		csk->opt2 &= ~TSTAMPS_EN_F;
+		csk->mtu_idx = TCPOPT_MSS_G(opt);
+	}
+}
+
+static void chtls_purge_receive_queue(struct sock *sk)
+{
+	struct sk_buff *skb;
+
+	while ((skb = __skb_dequeue(&sk->sk_receive_queue)) != NULL) {
+		skb_dstref_steal(skb);
+		kfree_skb(skb);
+	}
+}
+
+static void chtls_purge_write_queue(struct sock *sk)
+{
+	struct chtls_sock *csk = rcu_dereference_sk_user_data(sk);
+	struct sk_buff *skb;
+
+	while ((skb = __skb_dequeue(&csk->txq))) {
+		sk->sk_wmem_queued -= skb->truesize;
+		__kfree_skb(skb);
+	}
+}
+
+static void chtls_purge_recv_queue(struct sock *sk)
+{
+	struct chtls_sock *csk = rcu_dereference_sk_user_data(sk);
+	struct chtls_hws *tlsk = &csk->tlshws;
+	struct sk_buff *skb;
+
+	while ((skb = __skb_dequeue(&tlsk->sk_recv_queue)) != NULL) {
+		skb_dstref_steal(skb);
+		kfree_skb(skb);
+	}
+}
+
+static void abort_arp_failure(void *handle, struct sk_buff *skb)
+{
+	struct cpl_abort_req *req = cplhdr(skb);
+	struct chtls_dev *cdev;
+
+	cdev = (struct chtls_dev *)handle;
+	req->cmd = CPL_ABORT_NO_RST;
+	cxgb4_ofld_send(cdev->lldi->ports[0], skb);
+}
+
+static struct sk_buff *alloc_ctrl_skb(struct sk_buff *skb, int len)
+{
+	if (likely(skb && !skb_shared(skb) && !skb_cloned(skb))) {
+		__skb_trim(skb, 0);
+		refcount_inc(&skb->users);
+	} else {
+		skb = alloc_skb(len, GFP_KERNEL | __GFP_NOFAIL);
+	}
+	return skb;
+}
+
+static void chtls_send_abort(struct sock *sk, int mode, struct sk_buff *skb)
+{
+	struct cpl_abort_req *req;
+	struct chtls_sock *csk;
+	struct tcp_sock *tp;
+
+	csk = rcu_dereference_sk_user_data(sk);
+	tp = tcp_sk(sk);
+
+	if (!skb)
+		skb = alloc_ctrl_skb(csk->txdata_skb_cache, sizeof(*req));
+
+	req = (struct cpl_abort_req *)skb_put(skb, sizeof(*req));
+	INIT_TP_WR_CPL(req, CPL_ABORT_REQ, csk->tid);
+	skb_set_queue_mapping(skb, (csk->txq_idx << 1) | CPL_PRIORITY_DATA);
+	req->rsvd0 = htonl(tp->snd_nxt);
+	req->rsvd1 = !csk_flag_nochk(csk, CSK_TX_DATA_SENT);
+	req->cmd = mode;
+	t4_set_arp_err_handler(skb, csk->cdev, abort_arp_failure);
+	send_or_defer(sk, tp, skb, mode == CPL_ABORT_SEND_RST);
+}
+
+static void chtls_send_reset(struct sock *sk, int mode, struct sk_buff *skb)
+{
+	struct chtls_sock *csk = rcu_dereference_sk_user_data(sk);
+
+	if (unlikely(csk_flag_nochk(csk, CSK_ABORT_SHUTDOWN) ||
+		     !csk->cdev)) {
+		if (sk->sk_state == TCP_SYN_RECV)
+			csk_set_flag(csk, CSK_RST_ABORTED);
+		goto out;
+	}
+
+	if (!csk_flag_nochk(csk, CSK_TX_DATA_SENT)) {
+		struct tcp_sock *tp = tcp_sk(sk);
+
+		if (send_tx_flowc_wr(sk, 0, tp->snd_nxt, tp->rcv_nxt) < 0)
+			WARN_ONCE(1, "send tx flowc error");
+		csk_set_flag(csk, CSK_TX_DATA_SENT);
+	}
+
+	csk_set_flag(csk, CSK_ABORT_RPL_PENDING);
+	chtls_purge_write_queue(sk);
+
+	csk_set_flag(csk, CSK_ABORT_SHUTDOWN);
+	if (sk->sk_state != TCP_SYN_RECV)
+		chtls_send_abort(sk, mode, skb);
+	else
+		chtls_set_tcb_field_rpl_skb(sk, TCB_T_FLAGS_W,
+					    TCB_T_FLAGS_V(TCB_T_FLAGS_M), 0,
+					    TCB_FIELD_COOKIE_TFLAG, 1);
+
+	return;
+out:
+	kfree_skb(skb);
+}
+
+static void release_tcp_port(struct sock *sk)
+{
+	if (inet_csk(sk)->icsk_bind_hash)
+		inet_put_port(sk);
+}
+
+static void tcp_uncork(struct sock *sk)
+{
+	struct tcp_sock *tp = tcp_sk(sk);
+
+	if (tp->nonagle & TCP_NAGLE_CORK) {
+		tp->nonagle &= ~TCP_NAGLE_CORK;
+		chtls_tcp_push(sk, 0);
+	}
+}
+
+static void chtls_close_conn(struct sock *sk)
+{
+	struct cpl_close_con_req *req;
+	struct chtls_sock *csk;
+	struct sk_buff *skb;
+	unsigned int tid;
+	unsigned int len;
+
+	len = roundup(sizeof(struct cpl_close_con_req), 16);
+	csk = rcu_dereference_sk_user_data(sk);
+	tid = csk->tid;
+
+	skb = alloc_skb(len, GFP_KERNEL | __GFP_NOFAIL);
+	req = (struct cpl_close_con_req *)__skb_put(skb, len);
+	memset(req, 0, len);
+	req->wr.wr_hi = htonl(FW_WR_OP_V(FW_TP_WR) |
+			      FW_WR_IMMDLEN_V(sizeof(*req) -
+					      sizeof(req->wr)));
+	req->wr.wr_mid = htonl(FW_WR_LEN16_V(DIV_ROUND_UP(sizeof(*req), 16)) |
+			       FW_WR_FLOWID_V(tid));
+
+	OPCODE_TID(req) = htonl(MK_OPCODE_TID(CPL_CLOSE_CON_REQ, tid));
+
+	tcp_uncork(sk);
+	skb_entail(sk, skb, ULPCB_FLAG_NO_HDR | ULPCB_FLAG_NO_APPEND);
+	if (sk->sk_state != TCP_SYN_SENT)
+		chtls_push_frames(csk, 1);
+}
+
+/*
+ * Perform a state transition during close and return the actions indicated
+ * for the transition.  Do not make this function inline, the main reason
+ * it exists at all is to avoid multiple inlining of tcp_set_state.
+ */
+static int make_close_transition(struct sock *sk)
+{
+	int next = (int)new_state[sk->sk_state];
+
+	tcp_set_state(sk, next & TCP_STATE_MASK);
+	return next & TCP_ACTION_FIN;
+}
+
+void chtls_close(struct sock *sk, long timeout)
+{
+	int data_lost, prev_state;
+	struct chtls_sock *csk;
+
+	csk = rcu_dereference_sk_user_data(sk);
+
+	lock_sock(sk);
+	sk->sk_shutdown |= SHUTDOWN_MASK;
+
+	data_lost = skb_queue_len(&sk->sk_receive_queue);
+	data_lost |= skb_queue_len(&csk->tlshws.sk_recv_queue);
+	chtls_purge_recv_queue(sk);
+	chtls_purge_receive_queue(sk);
+
+	if (sk->sk_state == TCP_CLOSE) {
+		goto wait;
+	} else if (data_lost || sk->sk_state == TCP_SYN_SENT) {
+		chtls_send_reset(sk, CPL_ABORT_SEND_RST, NULL);
+		release_tcp_port(sk);
+		goto unlock;
+	} else if (sock_flag(sk, SOCK_LINGER) && !sk->sk_lingertime) {
+		sk->sk_prot->disconnect(sk, 0);
+	} else if (make_close_transition(sk)) {
+		chtls_close_conn(sk);
+	}
+wait:
+	if (timeout)
+		sk_stream_wait_close(sk, timeout);
+
+unlock:
+	prev_state = sk->sk_state;
+	sock_hold(sk);
+	sock_orphan(sk);
+
+	release_sock(sk);
+
+	local_bh_disable();
+	bh_lock_sock(sk);
+
+	if (prev_state != TCP_CLOSE && sk->sk_state == TCP_CLOSE)
+		goto out;
+
+	if (sk->sk_state == TCP_FIN_WAIT2 && tcp_sk(sk)->linger2 < 0 &&
+	    !csk_flag(sk, CSK_ABORT_SHUTDOWN)) {
+		struct sk_buff *skb;
+
+		skb = alloc_skb(sizeof(struct cpl_abort_req), GFP_ATOMIC);
+		if (skb)
+			chtls_send_reset(sk, CPL_ABORT_SEND_RST, skb);
+	}
+
+	if (sk->sk_state == TCP_CLOSE)
+		inet_csk_destroy_sock(sk);
+
+out:
+	bh_unlock_sock(sk);
+	local_bh_enable();
+	sock_put(sk);
+}
+
+/*
+ * Wait until a socket enters on of the given states.
+ */
+static int wait_for_states(struct sock *sk, unsigned int states)
+{
+	DECLARE_WAITQUEUE(wait, current);
+	struct socket_wq _sk_wq;
+	long current_timeo;
+	int err = 0;
+
+	current_timeo = 200;
+
+	/*
+	 * We want this to work even when there's no associated struct socket.
+	 * In that case we provide a temporary wait_queue_head_t.
+	 */
+	if (!sk->sk_wq) {
+		init_waitqueue_head(&_sk_wq.wait);
+		_sk_wq.fasync_list = NULL;
+		init_rcu_head_on_stack(&_sk_wq.rcu);
+		RCU_INIT_POINTER(sk->sk_wq, &_sk_wq);
+	}
+
+	add_wait_queue(sk_sleep(sk), &wait);
+	while (!sk_in_state(sk, states)) {
+		if (!current_timeo) {
+			err = -EBUSY;
+			break;
+		}
+		if (signal_pending(current)) {
+			err = sock_intr_errno(current_timeo);
+			break;
+		}
+		set_current_state(TASK_UNINTERRUPTIBLE);
+		release_sock(sk);
+		if (!sk_in_state(sk, states))
+			current_timeo = schedule_timeout(current_timeo);
+		__set_current_state(TASK_RUNNING);
+		lock_sock(sk);
+	}
+	remove_wait_queue(sk_sleep(sk), &wait);
+
+	if (rcu_dereference(sk->sk_wq) == &_sk_wq)
+		sk->sk_wq = NULL;
+	return err;
+}
+
+int chtls_disconnect(struct sock *sk, int flags)
+{
+	struct tcp_sock *tp;
+	int err;
+
+	tp = tcp_sk(sk);
+	chtls_purge_recv_queue(sk);
+	chtls_purge_receive_queue(sk);
+	chtls_purge_write_queue(sk);
+
+	if (sk->sk_state != TCP_CLOSE) {
+		sk->sk_err = ECONNRESET;
+		chtls_send_reset(sk, CPL_ABORT_SEND_RST, NULL);
+		err = wait_for_states(sk, TCPF_CLOSE);
+		if (err)
+			return err;
+	}
+	chtls_purge_recv_queue(sk);
+	chtls_purge_receive_queue(sk);
+	tp->max_window = 0xFFFF << (tp->rx_opt.snd_wscale);
+	return tcp_disconnect(sk, flags);
+}
+
+#define SHUTDOWN_ELIGIBLE_STATE (TCPF_ESTABLISHED | \
+				 TCPF_SYN_RECV | TCPF_CLOSE_WAIT)
+void chtls_shutdown(struct sock *sk, int how)
+{
+	if ((how & SEND_SHUTDOWN) &&
+	    sk_in_state(sk, SHUTDOWN_ELIGIBLE_STATE) &&
+	    make_close_transition(sk))
+		chtls_close_conn(sk);
+}
+
+void chtls_destroy_sock(struct sock *sk)
+{
+	struct chtls_sock *csk;
+
+	csk = rcu_dereference_sk_user_data(sk);
+	chtls_purge_recv_queue(sk);
+	csk->ulp_mode = ULP_MODE_NONE;
+	chtls_purge_write_queue(sk);
+	free_tls_keyid(sk);
+	kref_put(&csk->kref, chtls_sock_release);
+	if (sk->sk_family == AF_INET)
+		sk->sk_prot = &tcp_prot;
+#if IS_ENABLED(CONFIG_IPV6)
+	else
+		sk->sk_prot = &tcpv6_prot;
+#endif
+	sk->sk_prot->destroy(sk);
+}
+
+static void reset_listen_child(struct sock *child)
+{
+	struct chtls_sock *csk = rcu_dereference_sk_user_data(child);
+	struct sk_buff *skb;
+
+	skb = alloc_ctrl_skb(csk->txdata_skb_cache,
+			     sizeof(struct cpl_abort_req));
+
+	chtls_send_reset(child, CPL_ABORT_SEND_RST, skb);
+	sock_orphan(child);
+	tcp_orphan_count_inc();
+	if (child->sk_state == TCP_CLOSE)
+		inet_csk_destroy_sock(child);
+}
+
+static void chtls_disconnect_acceptq(struct sock *listen_sk)
+{
+	struct request_sock **pprev;
+
+	pprev = ACCEPT_QUEUE(listen_sk);
+	while (*pprev) {
+		struct request_sock *req = *pprev;
+
+		if (req->rsk_ops == &chtls_rsk_ops ||
+		    req->rsk_ops == &chtls_rsk_opsv6) {
+			struct sock *child = req->sk;
+
+			*pprev = req->dl_next;
+			sk_acceptq_removed(listen_sk);
+			reqsk_put(req);
+			sock_hold(child);
+			local_bh_disable();
+			bh_lock_sock(child);
+			release_tcp_port(child);
+			reset_listen_child(child);
+			bh_unlock_sock(child);
+			local_bh_enable();
+			sock_put(child);
+		} else {
+			pprev = &req->dl_next;
+		}
+	}
+}
+
+static int listen_hashfn(const struct sock *sk)
+{
+	return ((unsigned long)sk >> 10) & (LISTEN_INFO_HASH_SIZE - 1);
+}
+
+static struct listen_info *listen_hash_add(struct chtls_dev *cdev,
+					   struct sock *sk,
+					   unsigned int stid)
+{
+	struct listen_info *p = kmalloc(sizeof(*p), GFP_KERNEL);
+
+	if (p) {
+		int key = listen_hashfn(sk);
+
+		p->sk = sk;
+		p->stid = stid;
+		spin_lock(&cdev->listen_lock);
+		p->next = cdev->listen_hash_tab[key];
+		cdev->listen_hash_tab[key] = p;
+		spin_unlock(&cdev->listen_lock);
+	}
+	return p;
+}
+
+static int listen_hash_find(struct chtls_dev *cdev,
+			    struct sock *sk)
+{
+	struct listen_info *p;
+	int stid = -1;
+	int key;
+
+	key = listen_hashfn(sk);
+
+	spin_lock(&cdev->listen_lock);
+	for (p = cdev->listen_hash_tab[key]; p; p = p->next)
+		if (p->sk == sk) {
+			stid = p->stid;
+			break;
+		}
+	spin_unlock(&cdev->listen_lock);
+	return stid;
+}
+
+static int listen_hash_del(struct chtls_dev *cdev,
+			   struct sock *sk)
+{
+	struct listen_info *p, **prev;
+	int stid = -1;
+	int key;
+
+	key = listen_hashfn(sk);
+	prev = &cdev->listen_hash_tab[key];
+
+	spin_lock(&cdev->listen_lock);
+	for (p = *prev; p; prev = &p->next, p = p->next)
+		if (p->sk == sk) {
+			stid = p->stid;
+			*prev = p->next;
+			kfree(p);
+			break;
+		}
+	spin_unlock(&cdev->listen_lock);
+	return stid;
+}
+
+static void cleanup_syn_rcv_conn(struct sock *child, struct sock *parent)
+{
+	struct request_sock *req;
+	struct chtls_sock *csk;
+
+	csk = rcu_dereference_sk_user_data(child);
+	req = csk->passive_reap_next;
+
+	reqsk_queue_removed(&inet_csk(parent)->icsk_accept_queue, req);
+	__skb_unlink((struct sk_buff *)&csk->synq, &csk->listen_ctx->synq);
+	chtls_reqsk_free(req);
+	csk->passive_reap_next = NULL;
+}
+
+static void chtls_reset_synq(struct listen_ctx *listen_ctx)
+{
+	struct sock *listen_sk = listen_ctx->lsk;
+
+	while (!skb_queue_empty(&listen_ctx->synq)) {
+		struct chtls_sock *csk =
+			container_of((struct synq *)skb_peek
+				(&listen_ctx->synq), struct chtls_sock, synq);
+		struct sock *child = csk->sk;
+
+		cleanup_syn_rcv_conn(child, listen_sk);
+		sock_hold(child);
+		local_bh_disable();
+		bh_lock_sock(child);
+		release_tcp_port(child);
+		reset_listen_child(child);
+		bh_unlock_sock(child);
+		local_bh_enable();
+		sock_put(child);
+	}
+}
+
+int chtls_listen_start(struct chtls_dev *cdev, struct sock *sk)
+{
+	struct net_device *ndev;
+#if IS_ENABLED(CONFIG_IPV6)
+	bool clip_valid = false;
+#endif
+	struct listen_ctx *ctx;
+	struct adapter *adap;
+	struct port_info *pi;
+	int ret = 0;
+	int stid;
+
+	rcu_read_lock();
+	ndev = chtls_find_netdev(cdev, sk);
+	rcu_read_unlock();
+	if (!ndev)
+		return -EBADF;
+
+	pi = netdev_priv(ndev);
+	adap = pi->adapter;
+	if (!(adap->flags & CXGB4_FULL_INIT_DONE))
+		return -EBADF;
+
+	if (listen_hash_find(cdev, sk) >= 0)   /* already have it */
+		return -EADDRINUSE;
+
+	ctx = kmalloc(sizeof(*ctx), GFP_KERNEL);
+	if (!ctx)
+		return -ENOMEM;
+
+	__module_get(THIS_MODULE);
+	ctx->lsk = sk;
+	ctx->cdev = cdev;
+	ctx->state = T4_LISTEN_START_PENDING;
+	skb_queue_head_init(&ctx->synq);
+
+	stid = cxgb4_alloc_stid(cdev->tids, sk->sk_family, ctx);
+	if (stid < 0)
+		goto free_ctx;
+
+	sock_hold(sk);
+	if (!listen_hash_add(cdev, sk, stid))
+		goto free_stid;
+
+	if (sk->sk_family == PF_INET) {
+		ret = cxgb4_create_server(ndev, stid,
+					  inet_sk(sk)->inet_rcv_saddr,
+					  inet_sk(sk)->inet_sport, 0,
+					  cdev->lldi->rxq_ids[0]);
+#if IS_ENABLED(CONFIG_IPV6)
+	} else {
+		int addr_type;
+
+		addr_type = ipv6_addr_type(&sk->sk_v6_rcv_saddr);
+		if (addr_type != IPV6_ADDR_ANY) {
+			ret = cxgb4_clip_get(ndev, (const u32 *)
+					     &sk->sk_v6_rcv_saddr, 1);
+			if (ret)
+				goto del_hash;
+			clip_valid = true;
+		}
+		ret = cxgb4_create_server6(ndev, stid,
+					   &sk->sk_v6_rcv_saddr,
+					   inet_sk(sk)->inet_sport,
+					   cdev->lldi->rxq_ids[0]);
+#endif
+	}
+	if (ret > 0)
+		ret = net_xmit_errno(ret);
+	if (ret)
+		goto del_hash;
+	return 0;
+del_hash:
+#if IS_ENABLED(CONFIG_IPV6)
+	if (clip_valid)
+		cxgb4_clip_release(ndev, (const u32 *)&sk->sk_v6_rcv_saddr, 1);
+#endif
+	listen_hash_del(cdev, sk);
+free_stid:
+	cxgb4_free_stid(cdev->tids, stid, sk->sk_family);
+	sock_put(sk);
+free_ctx:
+	kfree(ctx);
+	module_put(THIS_MODULE);
+	return -EBADF;
+}
+
+void chtls_listen_stop(struct chtls_dev *cdev, struct sock *sk)
+{
+	struct listen_ctx *listen_ctx;
+	int stid;
+
+	stid = listen_hash_del(cdev, sk);
+	if (stid < 0)
+		return;
+
+	listen_ctx = (struct listen_ctx *)lookup_stid(cdev->tids, stid);
+	chtls_reset_synq(listen_ctx);
+
+	cxgb4_remove_server(cdev->lldi->ports[0], stid,
+			    cdev->lldi->rxq_ids[0], sk->sk_family == PF_INET6);
+
+#if IS_ENABLED(CONFIG_IPV6)
+	if (sk->sk_family == PF_INET6) {
+		struct net_device *ndev = chtls_find_netdev(cdev, sk);
+		int addr_type = 0;
+
+		addr_type = ipv6_addr_type((const struct in6_addr *)
+					  &sk->sk_v6_rcv_saddr);
+		if (addr_type != IPV6_ADDR_ANY)
+			cxgb4_clip_release(ndev, (const u32 *)
+					   &sk->sk_v6_rcv_saddr, 1);
+	}
+#endif
+	chtls_disconnect_acceptq(sk);
+}
+
+static int chtls_pass_open_rpl(struct chtls_dev *cdev, struct sk_buff *skb)
+{
+	struct cpl_pass_open_rpl *rpl = cplhdr(skb) + RSS_HDR;
+	unsigned int stid = GET_TID(rpl);
+	struct listen_ctx *listen_ctx;
+
+	listen_ctx = (struct listen_ctx *)lookup_stid(cdev->tids, stid);
+	if (!listen_ctx)
+		return CPL_RET_BUF_DONE;
+
+	if (listen_ctx->state == T4_LISTEN_START_PENDING) {
+		listen_ctx->state = T4_LISTEN_STARTED;
+		return CPL_RET_BUF_DONE;
+	}
+
+	if (rpl->status != CPL_ERR_NONE) {
+		pr_info("Unexpected PASS_OPEN_RPL status %u for STID %u\n",
+			rpl->status, stid);
+	} else {
+		cxgb4_free_stid(cdev->tids, stid, listen_ctx->lsk->sk_family);
+		sock_put(listen_ctx->lsk);
+		kfree(listen_ctx);
+		module_put(THIS_MODULE);
+	}
+	return CPL_RET_BUF_DONE;
+}
+
+static int chtls_close_listsrv_rpl(struct chtls_dev *cdev, struct sk_buff *skb)
+{
+	struct cpl_close_listsvr_rpl *rpl = cplhdr(skb) + RSS_HDR;
+	struct listen_ctx *listen_ctx;
+	unsigned int stid;
+	void *data;
+
+	stid = GET_TID(rpl);
+	data = lookup_stid(cdev->tids, stid);
+	listen_ctx = (struct listen_ctx *)data;
+
+	if (rpl->status != CPL_ERR_NONE) {
+		pr_info("Unexpected CLOSE_LISTSRV_RPL status %u for STID %u\n",
+			rpl->status, stid);
+	} else {
+		cxgb4_free_stid(cdev->tids, stid, listen_ctx->lsk->sk_family);
+		sock_put(listen_ctx->lsk);
+		kfree(listen_ctx);
+		module_put(THIS_MODULE);
+	}
+	return CPL_RET_BUF_DONE;
+}
+
+static void chtls_purge_wr_queue(struct sock *sk)
+{
+	struct sk_buff *skb;
+
+	while ((skb = dequeue_wr(sk)) != NULL)
+		kfree_skb(skb);
+}
+
+static void chtls_release_resources(struct sock *sk)
+{
+	struct chtls_sock *csk = rcu_dereference_sk_user_data(sk);
+	struct chtls_dev *cdev = csk->cdev;
+	unsigned int tid = csk->tid;
+	struct tid_info *tids;
+
+	if (!cdev)
+		return;
+
+	tids = cdev->tids;
+	kfree_skb(csk->txdata_skb_cache);
+	csk->txdata_skb_cache = NULL;
+
+	if (csk->wr_credits != csk->wr_max_credits) {
+		chtls_purge_wr_queue(sk);
+		chtls_reset_wr_list(csk);
+	}
+
+	if (csk->l2t_entry) {
+		cxgb4_l2t_release(csk->l2t_entry);
+		csk->l2t_entry = NULL;
+	}
+
+	if (sk->sk_state != TCP_SYN_SENT) {
+		cxgb4_remove_tid(tids, csk->port_id, tid, sk->sk_family);
+		sock_put(sk);
+	}
+}
+
+static void chtls_conn_done(struct sock *sk)
+{
+	if (sock_flag(sk, SOCK_DEAD))
+		chtls_purge_receive_queue(sk);
+	sk_wakeup_sleepers(sk, 0);
+	tcp_done(sk);
+}
+
+static void do_abort_syn_rcv(struct sock *child, struct sock *parent)
+{
+	/*
+	 * If the server is still open we clean up the child connection,
+	 * otherwise the server already did the clean up as it was purging
+	 * its SYN queue and the skb was just sitting in its backlog.
+	 */
+	if (likely(parent->sk_state == TCP_LISTEN)) {
+		cleanup_syn_rcv_conn(child, parent);
+		/* Without the below call to sock_orphan,
+		 * we leak the socket resource with syn_flood test
+		 * as inet_csk_destroy_sock will not be called
+		 * in tcp_done since SOCK_DEAD flag is not set.
+		 * Kernel handles this differently where new socket is
+		 * created only after 3 way handshake is done.
+		 */
+		sock_orphan(child);
+		tcp_orphan_count_inc();
+		chtls_release_resources(child);
+		chtls_conn_done(child);
+	} else {
+		if (csk_flag(child, CSK_RST_ABORTED)) {
+			chtls_release_resources(child);
+			chtls_conn_done(child);
+		}
+	}
+}
+
+static void pass_open_abort(struct sock *child, struct sock *parent,
+			    struct sk_buff *skb)
+{
+	do_abort_syn_rcv(child, parent);
+	kfree_skb(skb);
+}
+
+static void bl_pass_open_abort(struct sock *lsk, struct sk_buff *skb)
+{
+	pass_open_abort(skb->sk, lsk, skb);
+}
+
+static void chtls_pass_open_arp_failure(struct sock *sk,
+					struct sk_buff *skb)
+{
+	const struct request_sock *oreq;
+	struct chtls_sock *csk;
+	struct chtls_dev *cdev;
+	struct sock *parent;
+	void *data;
+
+	csk = rcu_dereference_sk_user_data(sk);
+	cdev = csk->cdev;
+
+	/*
+	 * If the connection is being aborted due to the parent listening
+	 * socket going away there's nothing to do, the ABORT_REQ will close
+	 * the connection.
+	 */
+	if (csk_flag(sk, CSK_ABORT_RPL_PENDING)) {
+		kfree_skb(skb);
+		return;
+	}
+
+	oreq = csk->passive_reap_next;
+	data = lookup_stid(cdev->tids, oreq->ts_recent);
+	parent = ((struct listen_ctx *)data)->lsk;
+
+	bh_lock_sock(parent);
+	if (!sock_owned_by_user(parent)) {
+		pass_open_abort(sk, parent, skb);
+	} else {
+		BLOG_SKB_CB(skb)->backlog_rcv = bl_pass_open_abort;
+		__sk_add_backlog(parent, skb);
+	}
+	bh_unlock_sock(parent);
+}
+
+static void chtls_accept_rpl_arp_failure(void *handle,
+					 struct sk_buff *skb)
+{
+	struct sock *sk = (struct sock *)handle;
+
+	sock_hold(sk);
+	process_cpl_msg(chtls_pass_open_arp_failure, sk, skb);
+	sock_put(sk);
+}
+
+static unsigned int chtls_select_mss(const struct chtls_sock *csk,
+				     unsigned int pmtu,
+				     struct cpl_pass_accept_req *req)
+{
+	struct chtls_dev *cdev;
+	struct dst_entry *dst;
+	unsigned int tcpoptsz;
+	unsigned int iphdrsz;
+	unsigned int mtu_idx;
+	struct tcp_sock *tp;
+	unsigned int mss;
+	struct sock *sk;
+	u16 user_mss;
+
+	mss = ntohs(req->tcpopt.mss);
+	sk = csk->sk;
+	dst = __sk_dst_get(sk);
+	cdev = csk->cdev;
+	tp = tcp_sk(sk);
+	tcpoptsz = 0;
+
+#if IS_ENABLED(CONFIG_IPV6)
+	if (sk->sk_family == AF_INET6)
+		iphdrsz = sizeof(struct ipv6hdr) + sizeof(struct tcphdr);
+	else
+#endif
+		iphdrsz = sizeof(struct iphdr) + sizeof(struct tcphdr);
+	if (req->tcpopt.tstamp)
+		tcpoptsz += round_up(TCPOLEN_TIMESTAMP, 4);
+
+	tp->advmss = dst_metric_advmss(dst);
+	user_mss = USER_MSS(tp);
+	if (user_mss && tp->advmss > user_mss)
+		tp->advmss = user_mss;
+	if (tp->advmss > pmtu - iphdrsz)
+		tp->advmss = pmtu - iphdrsz;
+	if (mss && tp->advmss > mss)
+		tp->advmss = mss;
+
+	tp->advmss = cxgb4_best_aligned_mtu(cdev->lldi->mtus,
+					    iphdrsz + tcpoptsz,
+					    tp->advmss - tcpoptsz,
+					    8, &mtu_idx);
+	tp->advmss -= iphdrsz;
+
+	inet_csk(sk)->icsk_pmtu_cookie = pmtu;
+	return mtu_idx;
+}
+
+static unsigned int select_rcv_wscale(int space, int wscale_ok, int win_clamp)
+{
+	int wscale = 0;
+
+	if (space > MAX_RCV_WND)
+		space = MAX_RCV_WND;
+	if (win_clamp && win_clamp < space)
+		space = win_clamp;
+
+	if (wscale_ok) {
+		while (wscale < 14 && (65535 << wscale) < space)
+			wscale++;
+	}
+	return wscale;
+}
+
+static void chtls_pass_accept_rpl(struct sk_buff *skb,
+				  struct cpl_pass_accept_req *req,
+				  unsigned int tid)
+
+{
+	struct cpl_t5_pass_accept_rpl *rpl5;
+	struct cxgb4_lld_info *lldi;
+	const struct tcphdr *tcph;
+	const struct tcp_sock *tp;
+	struct chtls_sock *csk;
+	unsigned int len;
+	struct sock *sk;
+	u32 opt2, hlen;
+	u64 opt0;
+
+	sk = skb->sk;
+	tp = tcp_sk(sk);
+	csk = sk->sk_user_data;
+	csk->tid = tid;
+	lldi = csk->cdev->lldi;
+	len = roundup(sizeof(*rpl5), 16);
+
+	rpl5 = __skb_put_zero(skb, len);
+	INIT_TP_WR(rpl5, tid);
+
+	OPCODE_TID(rpl5) = cpu_to_be32(MK_OPCODE_TID(CPL_PASS_ACCEPT_RPL,
+						     csk->tid));
+	csk->mtu_idx = chtls_select_mss(csk, dst_mtu(__sk_dst_get(sk)),
+					req);
+	opt0 = TCAM_BYPASS_F |
+	       WND_SCALE_V(RCV_WSCALE(tp)) |
+	       MSS_IDX_V(csk->mtu_idx) |
+	       L2T_IDX_V(csk->l2t_entry->idx) |
+	       NAGLE_V(!(tp->nonagle & TCP_NAGLE_OFF)) |
+	       TX_CHAN_V(csk->tx_chan) |
+	       SMAC_SEL_V(csk->smac_idx) |
+	       DSCP_V(csk->tos >> 2) |
+	       ULP_MODE_V(ULP_MODE_TLS) |
+	       RCV_BUFSIZ_V(min(tp->rcv_wnd >> 10, RCV_BUFSIZ_M));
+
+	opt2 = RX_CHANNEL_V(0) |
+		RSS_QUEUE_VALID_F | RSS_QUEUE_V(csk->rss_qid);
+
+	if (!is_t5(lldi->adapter_type))
+		opt2 |= RX_FC_DISABLE_F;
+	if (req->tcpopt.tstamp)
+		opt2 |= TSTAMPS_EN_F;
+	if (req->tcpopt.sack)
+		opt2 |= SACK_EN_F;
+	hlen = ntohl(req->hdr_len);
+
+	tcph = (struct tcphdr *)((u8 *)(req + 1) +
+			T6_ETH_HDR_LEN_G(hlen) + T6_IP_HDR_LEN_G(hlen));
+	if (tcph->ece && tcph->cwr)
+		opt2 |= CCTRL_ECN_V(1);
+	opt2 |= CONG_CNTRL_V(CONG_ALG_NEWRENO);
+	opt2 |= T5_ISS_F;
+	opt2 |= T5_OPT_2_VALID_F;
+	opt2 |= WND_SCALE_EN_V(WSCALE_OK(tp));
+	rpl5->opt0 = cpu_to_be64(opt0);
+	rpl5->opt2 = cpu_to_be32(opt2);
+	rpl5->iss = cpu_to_be32((get_random_u32() & ~7UL) - 1);
+	set_wr_txq(skb, CPL_PRIORITY_SETUP, csk->port_id);
+	t4_set_arp_err_handler(skb, sk, chtls_accept_rpl_arp_failure);
+	cxgb4_l2t_send(csk->egress_dev, skb, csk->l2t_entry);
+}
+
+static void inet_inherit_port(struct sock *lsk, struct sock *newsk)
+{
+	local_bh_disable();
+	__inet_inherit_port(lsk, newsk);
+	local_bh_enable();
+}
+
+static int chtls_backlog_rcv(struct sock *sk, struct sk_buff *skb)
+{
+	if (skb->protocol) {
+		kfree_skb(skb);
+		return 0;
+	}
+	BLOG_SKB_CB(skb)->backlog_rcv(sk, skb);
+	return 0;
+}
+
+static void chtls_set_tcp_window(struct chtls_sock *csk)
+{
+	struct net_device *ndev = csk->egress_dev;
+	struct port_info *pi = netdev_priv(ndev);
+	unsigned int linkspeed;
+	u8 scale;
+
+	linkspeed = pi->link_cfg.speed;
+	scale = linkspeed / SPEED_10000;
+#define CHTLS_10G_RCVWIN (256 * 1024)
+	csk->rcv_win = CHTLS_10G_RCVWIN;
+	if (scale)
+		csk->rcv_win *= scale;
+#define CHTLS_10G_SNDWIN (256 * 1024)
+	csk->snd_win = CHTLS_10G_SNDWIN;
+	if (scale)
+		csk->snd_win *= scale;
+}
+
+static struct sock *chtls_recv_sock(struct sock *lsk,
+				    struct request_sock *oreq,
+				    void *network_hdr,
+				    const struct cpl_pass_accept_req *req,
+				    struct chtls_dev *cdev)
+{
+	struct adapter *adap = pci_get_drvdata(cdev->pdev);
+	struct neighbour *n = NULL;
+	struct inet_sock *newinet;
+	const struct iphdr *iph;
+	struct tls_context *ctx;
+	struct net_device *ndev;
+	struct chtls_sock *csk;
+	struct dst_entry *dst;
+	struct tcp_sock *tp;
+	struct sock *newsk;
+	bool found = false;
+	u16 port_id;
+	int rxq_idx;
+	int step, i;
+
+	iph = (const struct iphdr *)network_hdr;
+	newsk = tcp_create_openreq_child(lsk, oreq, cdev->askb);
+	if (!newsk)
+		goto free_oreq;
+
+	if (lsk->sk_family == AF_INET) {
+		dst = inet_csk_route_child_sock(lsk, newsk, oreq);
+		if (!dst)
+			goto free_sk;
+
+		n = dst_neigh_lookup(dst, &iph->saddr);
+#if IS_ENABLED(CONFIG_IPV6)
+	} else {
+		const struct ipv6hdr *ip6h;
+		struct flowi6 fl6;
+
+		ip6h = (const struct ipv6hdr *)network_hdr;
+		memset(&fl6, 0, sizeof(fl6));
+		fl6.flowi6_proto = IPPROTO_TCP;
+		fl6.saddr = ip6h->daddr;
+		fl6.daddr = ip6h->saddr;
+		fl6.fl6_dport = inet_rsk(oreq)->ir_rmt_port;
+		fl6.fl6_sport = htons(inet_rsk(oreq)->ir_num);
+		security_req_classify_flow(oreq, flowi6_to_flowi_common(&fl6));
+		dst = ip6_dst_lookup_flow(sock_net(lsk), lsk, &fl6, NULL);
+		if (IS_ERR(dst))
+			goto free_sk;
+		n = dst_neigh_lookup(dst, &ip6h->saddr);
+#endif
+	}
+	if (!n || !n->dev)
+		goto free_dst;
+
+	ndev = n->dev;
+	if (is_vlan_dev(ndev))
+		ndev = vlan_dev_real_dev(ndev);
+
+	for_each_port(adap, i)
+		if (cdev->ports[i] == ndev)
+			found = true;
+
+	if (!found)
+		goto free_dst;
+
+	port_id = cxgb4_port_idx(ndev);
+
+	csk = chtls_sock_create(cdev);
+	if (!csk)
+		goto free_dst;
+
+	csk->l2t_entry = cxgb4_l2t_get(cdev->lldi->l2t, n, ndev, 0);
+	if (!csk->l2t_entry)
+		goto free_csk;
+
+	newsk->sk_user_data = csk;
+	newsk->sk_backlog_rcv = chtls_backlog_rcv;
+
+	tp = tcp_sk(newsk);
+	newinet = inet_sk(newsk);
+
+	if (iph->version == 0x4) {
+		newinet->inet_daddr = iph->saddr;
+		newinet->inet_rcv_saddr = iph->daddr;
+		newinet->inet_saddr = iph->daddr;
+#if IS_ENABLED(CONFIG_IPV6)
+	} else {
+		struct tcp6_sock *newtcp6sk = (struct tcp6_sock *)newsk;
+		struct inet_request_sock *treq = inet_rsk(oreq);
+		struct ipv6_pinfo *newnp = inet6_sk(newsk);
+		struct ipv6_pinfo *np = inet6_sk(lsk);
+
+		inet_sk(newsk)->pinet6 = &newtcp6sk->inet6;
+		memcpy(newnp, np, sizeof(struct ipv6_pinfo));
+		newsk->sk_v6_daddr = treq->ir_v6_rmt_addr;
+		newsk->sk_v6_rcv_saddr = treq->ir_v6_loc_addr;
+		inet6_sk(newsk)->saddr = treq->ir_v6_loc_addr;
+		newnp->ipv6_fl_list = NULL;
+		newnp->pktoptions = NULL;
+		newsk->sk_bound_dev_if = treq->ir_iif;
+		newinet->inet_opt = NULL;
+		newinet->inet_daddr = LOOPBACK4_IPV6;
+		newinet->inet_saddr = LOOPBACK4_IPV6;
+#endif
+	}
+
+	oreq->ts_recent = PASS_OPEN_TID_G(ntohl(req->tos_stid));
+	sk_setup_caps(newsk, dst);
+	ctx = tls_get_ctx(lsk);
+	newsk->sk_destruct = ctx->sk_destruct;
+	newsk->sk_prot_creator = lsk->sk_prot_creator;
+	csk->sk = newsk;
+	csk->passive_reap_next = oreq;
+	csk->tx_chan = cxgb4_port_chan(ndev);
+	csk->port_id = port_id;
+	csk->egress_dev = ndev;
+	csk->tos = PASS_OPEN_TOS_G(ntohl(req->tos_stid));
+	chtls_set_tcp_window(csk);
+	tp->rcv_wnd = csk->rcv_win;
+	csk->sndbuf = csk->snd_win;
+	csk->ulp_mode = ULP_MODE_TLS;
+	step = cdev->lldi->nrxq / cdev->lldi->nchan;
+	rxq_idx = port_id * step;
+	rxq_idx += cdev->round_robin_cnt++ % step;
+	csk->rss_qid = cdev->lldi->rxq_ids[rxq_idx];
+	csk->txq_idx = (rxq_idx < cdev->lldi->ntxq) ? rxq_idx :
+			port_id * step;
+	csk->sndbuf = newsk->sk_sndbuf;
+	csk->smac_idx = ((struct port_info *)netdev_priv(ndev))->smt_idx;
+	RCV_WSCALE(tp) = select_rcv_wscale(tcp_full_space(newsk),
+					   READ_ONCE(sock_net(newsk)->
+						     ipv4.sysctl_tcp_window_scaling),
+					   tp->window_clamp);
+	neigh_release(n);
+	inet_inherit_port(lsk, newsk);
+	csk_set_flag(csk, CSK_CONN_INLINE);
+	bh_unlock_sock(newsk); /* tcp_create_openreq_child ->sk_clone_lock */
+
+	return newsk;
+free_csk:
+	chtls_sock_release(&csk->kref);
+free_dst:
+	if (n)
+		neigh_release(n);
+	dst_release(dst);
+free_sk:
+	inet_csk_prepare_forced_close(newsk);
+	tcp_done(newsk);
+free_oreq:
+	chtls_reqsk_free(oreq);
+	return NULL;
+}
+
+/*
+ * Populate a TID_RELEASE WR.  The skb must be already propely sized.
+ */
+static  void mk_tid_release(struct sk_buff *skb,
+			    unsigned int chan, unsigned int tid)
+{
+	struct cpl_tid_release *req;
+	unsigned int len;
+
+	len = roundup(sizeof(struct cpl_tid_release), 16);
+	req = (struct cpl_tid_release *)__skb_put(skb, len);
+	memset(req, 0, len);
+	set_wr_txq(skb, CPL_PRIORITY_SETUP, chan);
+	INIT_TP_WR_CPL(req, CPL_TID_RELEASE, tid);
+}
+
+static int chtls_get_module(struct sock *sk)
+{
+	struct inet_connection_sock *icsk = inet_csk(sk);
+
+	if (!try_module_get(icsk->icsk_ulp_ops->owner))
+		return -1;
+
+	return 0;
+}
+
+static void chtls_pass_accept_request(struct sock *sk,
+				      struct sk_buff *skb)
+{
+	struct cpl_t5_pass_accept_rpl *rpl;
+	struct cpl_pass_accept_req *req;
+	struct listen_ctx *listen_ctx;
+	struct vlan_ethhdr *vlan_eh;
+	struct request_sock *oreq;
+	struct sk_buff *reply_skb;
+	struct chtls_sock *csk;
+	struct chtls_dev *cdev;
+	struct ipv6hdr *ip6h;
+	struct tcphdr *tcph;
+	struct sock *newsk;
+	struct ethhdr *eh;
+	struct iphdr *iph;
+	void *network_hdr;
+	unsigned int stid;
+	unsigned int len;
+	unsigned int tid;
+	bool th_ecn, ect;
+	__u8 ip_dsfield; /* IPv4 tos or IPv6 dsfield */
+	u16 eth_hdr_len;
+	bool ecn_ok;
+
+	req = cplhdr(skb) + RSS_HDR;
+	tid = GET_TID(req);
+	cdev = BLOG_SKB_CB(skb)->cdev;
+	newsk = lookup_tid(cdev->tids, tid);
+	stid = PASS_OPEN_TID_G(ntohl(req->tos_stid));
+	if (newsk) {
+		pr_info("tid (%d) already in use\n", tid);
+		return;
+	}
+
+	len = roundup(sizeof(*rpl), 16);
+	reply_skb = alloc_skb(len, GFP_ATOMIC);
+	if (!reply_skb) {
+		cxgb4_remove_tid(cdev->tids, 0, tid, sk->sk_family);
+		kfree_skb(skb);
+		return;
+	}
+
+	if (sk->sk_state != TCP_LISTEN)
+		goto reject;
+
+	if (inet_csk_reqsk_queue_is_full(sk))
+		goto reject;
+
+	if (sk_acceptq_is_full(sk))
+		goto reject;
+
+
+	eth_hdr_len = T6_ETH_HDR_LEN_G(ntohl(req->hdr_len));
+	if (eth_hdr_len == ETH_HLEN) {
+		eh = (struct ethhdr *)(req + 1);
+		iph = (struct iphdr *)(eh + 1);
+		ip6h = (struct ipv6hdr *)(eh + 1);
+		network_hdr = (void *)(eh + 1);
+	} else {
+		vlan_eh = (struct vlan_ethhdr *)(req + 1);
+		iph = (struct iphdr *)(vlan_eh + 1);
+		ip6h = (struct ipv6hdr *)(vlan_eh + 1);
+		network_hdr = (void *)(vlan_eh + 1);
+	}
+
+	if (iph->version == 0x4) {
+		tcph = (struct tcphdr *)(iph + 1);
+		skb_set_network_header(skb, (void *)iph - (void *)req);
+		oreq = inet_reqsk_alloc(&chtls_rsk_ops, sk, true);
+	} else {
+		tcph = (struct tcphdr *)(ip6h + 1);
+		skb_set_network_header(skb, (void *)ip6h - (void *)req);
+		oreq = inet_reqsk_alloc(&chtls_rsk_opsv6, sk, false);
+	}
+
+	if (!oreq)
+		goto reject;
+
+	oreq->rsk_rcv_wnd = 0;
+	oreq->rsk_window_clamp = 0;
+	oreq->syncookie = 0;
+	oreq->mss = 0;
+	oreq->ts_recent = 0;
+
+	tcp_rsk(oreq)->tfo_listener = false;
+	tcp_rsk(oreq)->rcv_isn = ntohl(tcph->seq);
+	chtls_set_req_port(oreq, tcph->source, tcph->dest);
+	if (iph->version == 0x4) {
+		chtls_set_req_addr(oreq, iph->daddr, iph->saddr);
+		ip_dsfield = ipv4_get_dsfield(iph);
+#if IS_ENABLED(CONFIG_IPV6)
+	} else {
+		inet_rsk(oreq)->ir_v6_rmt_addr = ipv6_hdr(skb)->saddr;
+		inet_rsk(oreq)->ir_v6_loc_addr = ipv6_hdr(skb)->daddr;
+		ip_dsfield = ipv6_get_dsfield(ipv6_hdr(skb));
+#endif
+	}
+	if (req->tcpopt.wsf <= 14 &&
+	    READ_ONCE(sock_net(sk)->ipv4.sysctl_tcp_window_scaling)) {
+		inet_rsk(oreq)->wscale_ok = 1;
+		inet_rsk(oreq)->snd_wscale = req->tcpopt.wsf;
+	}
+	inet_rsk(oreq)->ir_iif = sk->sk_bound_dev_if;
+	th_ecn = tcph->ece && tcph->cwr;
+	if (th_ecn) {
+		ect = !INET_ECN_is_not_ect(ip_dsfield);
+		ecn_ok = READ_ONCE(sock_net(sk)->ipv4.sysctl_tcp_ecn);
+		if ((!ect && ecn_ok) || tcp_ca_needs_ecn(sk))
+			inet_rsk(oreq)->ecn_ok = 1;
+	}
+
+	newsk = chtls_recv_sock(sk, oreq, network_hdr, req, cdev);
+	if (!newsk)
+		goto reject;
+
+	if (chtls_get_module(newsk))
+		goto reject;
+	inet_csk_reqsk_queue_added(sk);
+	reply_skb->sk = newsk;
+	chtls_install_cpl_ops(newsk);
+	cxgb4_insert_tid(cdev->tids, newsk, tid, newsk->sk_family);
+	csk = rcu_dereference_sk_user_data(newsk);
+	listen_ctx = (struct listen_ctx *)lookup_stid(cdev->tids, stid);
+	csk->listen_ctx = listen_ctx;
+	__skb_queue_tail(&listen_ctx->synq, (struct sk_buff *)&csk->synq);
+	chtls_pass_accept_rpl(reply_skb, req, tid);
+	kfree_skb(skb);
+	return;
+
+reject:
+	mk_tid_release(reply_skb, 0, tid);
+	cxgb4_ofld_send(cdev->lldi->ports[0], reply_skb);
+	kfree_skb(skb);
+}
+
+/*
+ * Handle a CPL_PASS_ACCEPT_REQ message.
+ */
+static int chtls_pass_accept_req(struct chtls_dev *cdev, struct sk_buff *skb)
+{
+	struct cpl_pass_accept_req *req = cplhdr(skb) + RSS_HDR;
+	struct listen_ctx *ctx;
+	unsigned int stid;
+	unsigned int tid;
+	struct sock *lsk;
+	void *data;
+
+	stid = PASS_OPEN_TID_G(ntohl(req->tos_stid));
+	tid = GET_TID(req);
+
+	data = lookup_stid(cdev->tids, stid);
+	if (!data)
+		return 1;
+
+	ctx = (struct listen_ctx *)data;
+	lsk = ctx->lsk;
+
+	if (unlikely(tid_out_of_range(cdev->tids, tid))) {
+		pr_info("passive open TID %u too large\n", tid);
+		return 1;
+	}
+
+	BLOG_SKB_CB(skb)->cdev = cdev;
+	process_cpl_msg(chtls_pass_accept_request, lsk, skb);
+	return 0;
+}
+
+/*
+ * Completes some final bits of initialization for just established connections
+ * and changes their state to TCP_ESTABLISHED.
+ *
+ * snd_isn here is the ISN after the SYN, i.e., the true ISN + 1.
+ */
+static void make_established(struct sock *sk, u32 snd_isn, unsigned int opt)
+{
+	struct tcp_sock *tp = tcp_sk(sk);
+
+	tp->pushed_seq = snd_isn;
+	tp->write_seq = snd_isn;
+	tp->snd_nxt = snd_isn;
+	tp->snd_una = snd_isn;
+	atomic_set(&inet_sk(sk)->inet_id, get_random_u16());
+	assign_rxopt(sk, opt);
+
+	if (tp->rcv_wnd > (RCV_BUFSIZ_M << 10))
+		tp->rcv_wup -= tp->rcv_wnd - (RCV_BUFSIZ_M << 10);
+
+	smp_mb();
+	tcp_set_state(sk, TCP_ESTABLISHED);
+}
+
+static void chtls_abort_conn(struct sock *sk, struct sk_buff *skb)
+{
+	struct sk_buff *abort_skb;
+
+	abort_skb = alloc_skb(sizeof(struct cpl_abort_req), GFP_ATOMIC);
+	if (abort_skb)
+		chtls_send_reset(sk, CPL_ABORT_SEND_RST, abort_skb);
+}
+
+static struct sock *reap_list;
+static DEFINE_SPINLOCK(reap_list_lock);
+
+/*
+ * Process the reap list.
+ */
+DECLARE_TASK_FUNC(process_reap_list, task_param)
+{
+	spin_lock_bh(&reap_list_lock);
+	while (reap_list) {
+		struct sock *sk = reap_list;
+		struct chtls_sock *csk = rcu_dereference_sk_user_data(sk);
+
+		reap_list = csk->passive_reap_next;
+		csk->passive_reap_next = NULL;
+		spin_unlock(&reap_list_lock);
+		sock_hold(sk);
+
+		bh_lock_sock(sk);
+		chtls_abort_conn(sk, NULL);
+		sock_orphan(sk);
+		if (sk->sk_state == TCP_CLOSE)
+			inet_csk_destroy_sock(sk);
+		bh_unlock_sock(sk);
+		sock_put(sk);
+		spin_lock(&reap_list_lock);
+	}
+	spin_unlock_bh(&reap_list_lock);
+}
+
+static DECLARE_WORK(reap_task, process_reap_list);
+
+static void add_to_reap_list(struct sock *sk)
+{
+	struct chtls_sock *csk = sk->sk_user_data;
+
+	local_bh_disable();
+	release_tcp_port(sk); /* release the port immediately */
+
+	spin_lock(&reap_list_lock);
+	csk->passive_reap_next = reap_list;
+	reap_list = sk;
+	if (!csk->passive_reap_next)
+		schedule_work(&reap_task);
+	spin_unlock(&reap_list_lock);
+	local_bh_enable();
+}
+
+static void add_pass_open_to_parent(struct sock *child, struct sock *lsk,
+				    struct chtls_dev *cdev)
+{
+	struct request_sock *oreq;
+	struct chtls_sock *csk;
+
+	if (lsk->sk_state != TCP_LISTEN)
+		return;
+
+	csk = child->sk_user_data;
+	oreq = csk->passive_reap_next;
+	csk->passive_reap_next = NULL;
+
+	reqsk_queue_removed(&inet_csk(lsk)->icsk_accept_queue, oreq);
+	__skb_unlink((struct sk_buff *)&csk->synq, &csk->listen_ctx->synq);
+
+	if (sk_acceptq_is_full(lsk)) {
+		chtls_reqsk_free(oreq);
+		add_to_reap_list(child);
+	} else {
+		refcount_set(&oreq->rsk_refcnt, 1);
+		inet_csk_reqsk_queue_add(lsk, oreq, child);
+		lsk->sk_data_ready(lsk);
+	}
+}
+
+static void bl_add_pass_open_to_parent(struct sock *lsk, struct sk_buff *skb)
+{
+	struct sock *child = skb->sk;
+
+	skb->sk = NULL;
+	add_pass_open_to_parent(child, lsk, BLOG_SKB_CB(skb)->cdev);
+	kfree_skb(skb);
+}
+
+static int chtls_pass_establish(struct chtls_dev *cdev, struct sk_buff *skb)
+{
+	struct cpl_pass_establish *req = cplhdr(skb) + RSS_HDR;
+	struct chtls_sock *csk;
+	struct sock *lsk, *sk;
+	unsigned int hwtid;
+
+	hwtid = GET_TID(req);
+	sk = lookup_tid(cdev->tids, hwtid);
+	if (!sk)
+		return (CPL_RET_UNKNOWN_TID | CPL_RET_BUF_DONE);
+
+	bh_lock_sock(sk);
+	if (unlikely(sock_owned_by_user(sk))) {
+		kfree_skb(skb);
+	} else {
+		unsigned int stid;
+		void *data;
+
+		csk = sk->sk_user_data;
+		csk->wr_max_credits = 64;
+		csk->wr_credits = 64;
+		csk->wr_unacked = 0;
+		make_established(sk, ntohl(req->snd_isn), ntohs(req->tcp_opt));
+		stid = PASS_OPEN_TID_G(ntohl(req->tos_stid));
+		sk->sk_state_change(sk);
+		if (unlikely(sk->sk_socket))
+			sk_wake_async(sk, 0, POLL_OUT);
+
+		data = lookup_stid(cdev->tids, stid);
+		if (!data) {
+			/* listening server close */
+			kfree_skb(skb);
+			goto unlock;
+		}
+		lsk = ((struct listen_ctx *)data)->lsk;
+
+		bh_lock_sock(lsk);
+		if (unlikely(skb_queue_empty(&csk->listen_ctx->synq))) {
+			/* removed from synq */
+			bh_unlock_sock(lsk);
+			kfree_skb(skb);
+			goto unlock;
+		}
+
+		if (likely(!sock_owned_by_user(lsk))) {
+			kfree_skb(skb);
+			add_pass_open_to_parent(sk, lsk, cdev);
+		} else {
+			skb->sk = sk;
+			BLOG_SKB_CB(skb)->cdev = cdev;
+			BLOG_SKB_CB(skb)->backlog_rcv =
+				bl_add_pass_open_to_parent;
+			__sk_add_backlog(lsk, skb);
+		}
+		bh_unlock_sock(lsk);
+	}
+unlock:
+	bh_unlock_sock(sk);
+	return 0;
+}
+
+/*
+ * Handle receipt of an urgent pointer.
+ */
+static void handle_urg_ptr(struct sock *sk, u32 urg_seq)
+{
+	struct tcp_sock *tp = tcp_sk(sk);
+
+	urg_seq--;
+	if (tp->urg_data && !after(urg_seq, tp->urg_seq))
+		return;	/* duplicate pointer */
+
+	sk_send_sigurg(sk);
+	if (tp->urg_seq == tp->copied_seq && tp->urg_data &&
+	    !sock_flag(sk, SOCK_URGINLINE) &&
+	    tp->copied_seq != tp->rcv_nxt) {
+		struct sk_buff *skb = skb_peek(&sk->sk_receive_queue);
+
+		tp->copied_seq++;
+		if (skb && tp->copied_seq - ULP_SKB_CB(skb)->seq >= skb->len)
+			chtls_free_skb(sk, skb);
+	}
+
+	tp->urg_data = TCP_URG_NOTYET;
+	tp->urg_seq = urg_seq;
+}
+
+static void check_sk_callbacks(struct chtls_sock *csk)
+{
+	struct sock *sk = csk->sk;
+
+	if (unlikely(sk->sk_user_data &&
+		     !csk_flag_nochk(csk, CSK_CALLBACKS_CHKD)))
+		csk_set_flag(csk, CSK_CALLBACKS_CHKD);
+}
+
+/*
+ * Handles Rx data that arrives in a state where the socket isn't accepting
+ * new data.
+ */
+static void handle_excess_rx(struct sock *sk, struct sk_buff *skb)
+{
+	if (!csk_flag(sk, CSK_ABORT_SHUTDOWN))
+		chtls_abort_conn(sk, skb);
+
+	kfree_skb(skb);
+}
+
+static void chtls_recv_data(struct sock *sk, struct sk_buff *skb)
+{
+	struct cpl_rx_data *hdr = cplhdr(skb) + RSS_HDR;
+	struct chtls_sock *csk;
+	struct tcp_sock *tp;
+
+	csk = rcu_dereference_sk_user_data(sk);
+	tp = tcp_sk(sk);
+
+	if (unlikely(sk->sk_shutdown & RCV_SHUTDOWN)) {
+		handle_excess_rx(sk, skb);
+		return;
+	}
+
+	ULP_SKB_CB(skb)->seq = ntohl(hdr->seq);
+	ULP_SKB_CB(skb)->psh = hdr->psh;
+	skb_ulp_mode(skb) = ULP_MODE_NONE;
+
+	skb_reset_transport_header(skb);
+	__skb_pull(skb, sizeof(*hdr) + RSS_HDR);
+	if (!skb->data_len)
+		__skb_trim(skb, ntohs(hdr->len));
+
+	if (unlikely(hdr->urg))
+		handle_urg_ptr(sk, tp->rcv_nxt + ntohs(hdr->urg));
+	if (unlikely(tp->urg_data == TCP_URG_NOTYET &&
+		     tp->urg_seq - tp->rcv_nxt < skb->len))
+		tp->urg_data = TCP_URG_VALID |
+			       skb->data[tp->urg_seq - tp->rcv_nxt];
+
+	if (unlikely(hdr->dack_mode != csk->delack_mode)) {
+		csk->delack_mode = hdr->dack_mode;
+		csk->delack_seq = tp->rcv_nxt;
+	}
+
+	tcp_hdr(skb)->fin = 0;
+	tp->rcv_nxt += skb->len;
+
+	__skb_queue_tail(&sk->sk_receive_queue, skb);
+
+	if (!sock_flag(sk, SOCK_DEAD)) {
+		check_sk_callbacks(csk);
+		sk->sk_data_ready(sk);
+	}
+}
+
+static int chtls_rx_data(struct chtls_dev *cdev, struct sk_buff *skb)
+{
+	struct cpl_rx_data *req = cplhdr(skb) + RSS_HDR;
+	unsigned int hwtid = GET_TID(req);
+	struct sock *sk;
+
+	sk = lookup_tid(cdev->tids, hwtid);
+	if (unlikely(!sk)) {
+		pr_err("can't find conn. for hwtid %u.\n", hwtid);
+		return -EINVAL;
+	}
+	skb_dstref_steal(skb);
+	process_cpl_msg(chtls_recv_data, sk, skb);
+	return 0;
+}
+
+static void chtls_recv_pdu(struct sock *sk, struct sk_buff *skb)
+{
+	struct cpl_tls_data *hdr = cplhdr(skb);
+	struct chtls_sock *csk;
+	struct chtls_hws *tlsk;
+	struct tcp_sock *tp;
+
+	csk = rcu_dereference_sk_user_data(sk);
+	tlsk = &csk->tlshws;
+	tp = tcp_sk(sk);
+
+	if (unlikely(sk->sk_shutdown & RCV_SHUTDOWN)) {
+		handle_excess_rx(sk, skb);
+		return;
+	}
+
+	ULP_SKB_CB(skb)->seq = ntohl(hdr->seq);
+	ULP_SKB_CB(skb)->flags = 0;
+	skb_ulp_mode(skb) = ULP_MODE_TLS;
+
+	skb_reset_transport_header(skb);
+	__skb_pull(skb, sizeof(*hdr));
+	if (!skb->data_len)
+		__skb_trim(skb,
+			   CPL_TLS_DATA_LENGTH_G(ntohl(hdr->length_pkd)));
+
+	if (unlikely(tp->urg_data == TCP_URG_NOTYET && tp->urg_seq -
+		     tp->rcv_nxt < skb->len))
+		tp->urg_data = TCP_URG_VALID |
+			       skb->data[tp->urg_seq - tp->rcv_nxt];
+
+	tcp_hdr(skb)->fin = 0;
+	tlsk->pldlen = CPL_TLS_DATA_LENGTH_G(ntohl(hdr->length_pkd));
+	__skb_queue_tail(&tlsk->sk_recv_queue, skb);
+}
+
+static int chtls_rx_pdu(struct chtls_dev *cdev, struct sk_buff *skb)
+{
+	struct cpl_tls_data *req = cplhdr(skb);
+	unsigned int hwtid = GET_TID(req);
+	struct sock *sk;
+
+	sk = lookup_tid(cdev->tids, hwtid);
+	if (unlikely(!sk)) {
+		pr_err("can't find conn. for hwtid %u.\n", hwtid);
+		return -EINVAL;
+	}
+	skb_dstref_steal(skb);
+	process_cpl_msg(chtls_recv_pdu, sk, skb);
+	return 0;
+}
+
+static void chtls_set_hdrlen(struct sk_buff *skb, unsigned int nlen)
+{
+	struct tlsrx_cmp_hdr *tls_cmp_hdr = cplhdr(skb);
+
+	skb->hdr_len = ntohs((__force __be16)tls_cmp_hdr->length);
+	tls_cmp_hdr->length = ntohs((__force __be16)nlen);
+}
+
+static void chtls_rx_hdr(struct sock *sk, struct sk_buff *skb)
+{
+	struct tlsrx_cmp_hdr *tls_hdr_pkt;
+	struct cpl_rx_tls_cmp *cmp_cpl;
+	struct sk_buff *skb_rec;
+	struct chtls_sock *csk;
+	struct chtls_hws *tlsk;
+	struct tcp_sock *tp;
+
+	cmp_cpl = cplhdr(skb);
+	csk = rcu_dereference_sk_user_data(sk);
+	tlsk = &csk->tlshws;
+	tp = tcp_sk(sk);
+
+	ULP_SKB_CB(skb)->seq = ntohl(cmp_cpl->seq);
+	ULP_SKB_CB(skb)->flags = 0;
+
+	skb_reset_transport_header(skb);
+	__skb_pull(skb, sizeof(*cmp_cpl));
+	tls_hdr_pkt = (struct tlsrx_cmp_hdr *)skb->data;
+	if (tls_hdr_pkt->res_to_mac_error & TLSRX_HDR_PKT_ERROR_M)
+		tls_hdr_pkt->type = CONTENT_TYPE_ERROR;
+	if (!skb->data_len)
+		__skb_trim(skb, TLS_HEADER_LENGTH);
+
+	tp->rcv_nxt +=
+		CPL_RX_TLS_CMP_PDULENGTH_G(ntohl(cmp_cpl->pdulength_length));
+
+	ULP_SKB_CB(skb)->flags |= ULPCB_FLAG_TLS_HDR;
+	skb_rec = __skb_dequeue(&tlsk->sk_recv_queue);
+	if (!skb_rec) {
+		__skb_queue_tail(&sk->sk_receive_queue, skb);
+	} else {
+		chtls_set_hdrlen(skb, tlsk->pldlen);
+		tlsk->pldlen = 0;
+		__skb_queue_tail(&sk->sk_receive_queue, skb);
+		__skb_queue_tail(&sk->sk_receive_queue, skb_rec);
+	}
+
+	if (!sock_flag(sk, SOCK_DEAD)) {
+		check_sk_callbacks(csk);
+		sk->sk_data_ready(sk);
+	}
+}
+
+static int chtls_rx_cmp(struct chtls_dev *cdev, struct sk_buff *skb)
+{
+	struct cpl_rx_tls_cmp *req = cplhdr(skb);
+	unsigned int hwtid = GET_TID(req);
+	struct sock *sk;
+
+	sk = lookup_tid(cdev->tids, hwtid);
+	if (unlikely(!sk)) {
+		pr_err("can't find conn. for hwtid %u.\n", hwtid);
+		return -EINVAL;
+	}
+	skb_dstref_steal(skb);
+	process_cpl_msg(chtls_rx_hdr, sk, skb);
+
+	return 0;
+}
+
+static void chtls_timewait(struct sock *sk)
+{
+	struct tcp_sock *tp = tcp_sk(sk);
+
+	tp->rcv_nxt++;
+	tp->rx_opt.ts_recent_stamp = ktime_get_seconds();
+	tp->srtt_us = 0;
+	tcp_time_wait(sk, TCP_TIME_WAIT, 0);
+}
+
+static void chtls_peer_close(struct sock *sk, struct sk_buff *skb)
+{
+	struct chtls_sock *csk = rcu_dereference_sk_user_data(sk);
+
+	if (csk_flag_nochk(csk, CSK_ABORT_RPL_PENDING))
+		goto out;
+
+	sk->sk_shutdown |= RCV_SHUTDOWN;
+	sock_set_flag(sk, SOCK_DONE);
+
+	switch (sk->sk_state) {
+	case TCP_SYN_RECV:
+	case TCP_ESTABLISHED:
+		tcp_set_state(sk, TCP_CLOSE_WAIT);
+		break;
+	case TCP_FIN_WAIT1:
+		tcp_set_state(sk, TCP_CLOSING);
+		break;
+	case TCP_FIN_WAIT2:
+		chtls_release_resources(sk);
+		if (csk_flag_nochk(csk, CSK_ABORT_RPL_PENDING))
+			chtls_conn_done(sk);
+		else
+			chtls_timewait(sk);
+		break;
+	default:
+		pr_info("cpl_peer_close in bad state %d\n", sk->sk_state);
+	}
+
+	if (!sock_flag(sk, SOCK_DEAD)) {
+		sk->sk_state_change(sk);
+		/* Do not send POLL_HUP for half duplex close. */
+
+		if ((sk->sk_shutdown & SEND_SHUTDOWN) ||
+		    sk->sk_state == TCP_CLOSE)
+			sk_wake_async(sk, SOCK_WAKE_WAITD, POLL_HUP);
+		else
+			sk_wake_async(sk, SOCK_WAKE_WAITD, POLL_IN);
+	}
+out:
+	kfree_skb(skb);
+}
+
+static void chtls_close_con_rpl(struct sock *sk, struct sk_buff *skb)
+{
+	struct cpl_close_con_rpl *rpl = cplhdr(skb) + RSS_HDR;
+	struct chtls_sock *csk;
+	struct tcp_sock *tp;
+
+	csk = rcu_dereference_sk_user_data(sk);
+
+	if (csk_flag_nochk(csk, CSK_ABORT_RPL_PENDING))
+		goto out;
+
+	tp = tcp_sk(sk);
+
+	tp->snd_una = ntohl(rpl->snd_nxt) - 1;  /* exclude FIN */
+
+	switch (sk->sk_state) {
+	case TCP_CLOSING:
+		chtls_release_resources(sk);
+		if (csk_flag_nochk(csk, CSK_ABORT_RPL_PENDING))
+			chtls_conn_done(sk);
+		else
+			chtls_timewait(sk);
+		break;
+	case TCP_LAST_ACK:
+		chtls_release_resources(sk);
+		chtls_conn_done(sk);
+		break;
+	case TCP_FIN_WAIT1:
+		tcp_set_state(sk, TCP_FIN_WAIT2);
+		sk->sk_shutdown |= SEND_SHUTDOWN;
+
+		if (!sock_flag(sk, SOCK_DEAD))
+			sk->sk_state_change(sk);
+		else if (tcp_sk(sk)->linger2 < 0 &&
+			 !csk_flag_nochk(csk, CSK_ABORT_SHUTDOWN))
+			chtls_abort_conn(sk, skb);
+		else if (csk_flag_nochk(csk, CSK_TX_DATA_SENT))
+			chtls_set_quiesce_ctrl(sk, 0);
+		break;
+	default:
+		pr_info("close_con_rpl in bad state %d\n", sk->sk_state);
+	}
+out:
+	kfree_skb(skb);
+}
+
+static struct sk_buff *get_cpl_skb(struct sk_buff *skb,
+				   size_t len, gfp_t gfp)
+{
+	if (likely(!skb_is_nonlinear(skb) && !skb_cloned(skb))) {
+		WARN_ONCE(skb->len < len, "skb alloc error");
+		__skb_trim(skb, len);
+		skb_get(skb);
+	} else {
+		skb = alloc_skb(len, gfp);
+		if (skb)
+			__skb_put(skb, len);
+	}
+	return skb;
+}
+
+static void set_abort_rpl_wr(struct sk_buff *skb, unsigned int tid,
+			     int cmd)
+{
+	struct cpl_abort_rpl *rpl = cplhdr(skb);
+
+	INIT_TP_WR_CPL(rpl, CPL_ABORT_RPL, tid);
+	rpl->cmd = cmd;
+}
+
+static void send_defer_abort_rpl(struct chtls_dev *cdev, struct sk_buff *skb)
+{
+	struct cpl_abort_req_rss *req = cplhdr(skb);
+	struct sk_buff *reply_skb;
+
+	reply_skb = alloc_skb(sizeof(struct cpl_abort_rpl),
+			      GFP_KERNEL | __GFP_NOFAIL);
+	__skb_put(reply_skb, sizeof(struct cpl_abort_rpl));
+	set_abort_rpl_wr(reply_skb, GET_TID(req),
+			 (req->status & CPL_ABORT_NO_RST));
+	set_wr_txq(reply_skb, CPL_PRIORITY_DATA, req->status >> 1);
+	cxgb4_ofld_send(cdev->lldi->ports[0], reply_skb);
+	kfree_skb(skb);
+}
+
+/*
+ * Add an skb to the deferred skb queue for processing from process context.
+ */
+static void t4_defer_reply(struct sk_buff *skb, struct chtls_dev *cdev,
+			   defer_handler_t handler)
+{
+	DEFERRED_SKB_CB(skb)->handler = handler;
+	spin_lock_bh(&cdev->deferq.lock);
+	__skb_queue_tail(&cdev->deferq, skb);
+	if (skb_queue_len(&cdev->deferq) == 1)
+		schedule_work(&cdev->deferq_task);
+	spin_unlock_bh(&cdev->deferq.lock);
+}
+
+static void chtls_send_abort_rpl(struct sock *sk, struct sk_buff *skb,
+				 struct chtls_dev *cdev,
+				 int status, int queue)
+{
+	struct cpl_abort_req_rss *req = cplhdr(skb) + RSS_HDR;
+	struct sk_buff *reply_skb;
+	struct chtls_sock *csk;
+	unsigned int tid;
+
+	csk = rcu_dereference_sk_user_data(sk);
+	tid = GET_TID(req);
+
+	reply_skb = get_cpl_skb(skb, sizeof(struct cpl_abort_rpl), gfp_any());
+	if (!reply_skb) {
+		req->status = (queue << 1) | status;
+		t4_defer_reply(skb, cdev, send_defer_abort_rpl);
+		return;
+	}
+
+	set_abort_rpl_wr(reply_skb, tid, status);
+	kfree_skb(skb);
+	set_wr_txq(reply_skb, CPL_PRIORITY_DATA, queue);
+	if (csk_conn_inline(csk)) {
+		struct l2t_entry *e = csk->l2t_entry;
+
+		if (e && sk->sk_state != TCP_SYN_RECV) {
+			cxgb4_l2t_send(csk->egress_dev, reply_skb, e);
+			return;
+		}
+	}
+	cxgb4_ofld_send(cdev->lldi->ports[0], reply_skb);
+}
+
+/*
+ * This is run from a listener's backlog to abort a child connection in
+ * SYN_RCV state (i.e., one on the listener's SYN queue).
+ */
+static void bl_abort_syn_rcv(struct sock *lsk, struct sk_buff *skb)
+{
+	struct chtls_sock *csk;
+	struct sock *child;
+	int queue;
+
+	child = skb->sk;
+	csk = rcu_dereference_sk_user_data(child);
+	queue = csk->txq_idx;
+
+	skb->sk	= NULL;
+	chtls_send_abort_rpl(child, skb, BLOG_SKB_CB(skb)->cdev,
+			     CPL_ABORT_NO_RST, queue);
+	do_abort_syn_rcv(child, lsk);
+}
+
+static int abort_syn_rcv(struct sock *sk, struct sk_buff *skb)
+{
+	const struct request_sock *oreq;
+	struct listen_ctx *listen_ctx;
+	struct chtls_sock *csk;
+	struct chtls_dev *cdev;
+	struct sock *psk;
+	void *ctx;
+
+	csk = sk->sk_user_data;
+	oreq = csk->passive_reap_next;
+	cdev = csk->cdev;
+
+	if (!oreq)
+		return -1;
+
+	ctx = lookup_stid(cdev->tids, oreq->ts_recent);
+	if (!ctx)
+		return -1;
+
+	listen_ctx = (struct listen_ctx *)ctx;
+	psk = listen_ctx->lsk;
+
+	bh_lock_sock(psk);
+	if (!sock_owned_by_user(psk)) {
+		int queue = csk->txq_idx;
+
+		chtls_send_abort_rpl(sk, skb, cdev, CPL_ABORT_NO_RST, queue);
+		do_abort_syn_rcv(sk, psk);
+	} else {
+		skb->sk = sk;
+		BLOG_SKB_CB(skb)->backlog_rcv = bl_abort_syn_rcv;
+		__sk_add_backlog(psk, skb);
+	}
+	bh_unlock_sock(psk);
+	return 0;
+}
+
+static void chtls_abort_req_rss(struct sock *sk, struct sk_buff *skb)
+{
+	const struct cpl_abort_req_rss *req = cplhdr(skb) + RSS_HDR;
+	struct chtls_sock *csk = sk->sk_user_data;
+	int rst_status = CPL_ABORT_NO_RST;
+	int queue = csk->txq_idx;
+
+	if (is_neg_adv(req->status)) {
+		kfree_skb(skb);
+		return;
+	}
+
+	csk_reset_flag(csk, CSK_ABORT_REQ_RCVD);
+
+	if (!csk_flag_nochk(csk, CSK_ABORT_SHUTDOWN) &&
+	    !csk_flag_nochk(csk, CSK_TX_DATA_SENT)) {
+		struct tcp_sock *tp = tcp_sk(sk);
+
+		if (send_tx_flowc_wr(sk, 0, tp->snd_nxt, tp->rcv_nxt) < 0)
+			WARN_ONCE(1, "send_tx_flowc error");
+		csk_set_flag(csk, CSK_TX_DATA_SENT);
+	}
+
+	csk_set_flag(csk, CSK_ABORT_SHUTDOWN);
+
+	if (!csk_flag_nochk(csk, CSK_ABORT_RPL_PENDING)) {
+		sk->sk_err = ETIMEDOUT;
+
+		if (!sock_flag(sk, SOCK_DEAD))
+			sk_error_report(sk);
+
+		if (sk->sk_state == TCP_SYN_RECV && !abort_syn_rcv(sk, skb))
+			return;
+
+	}
+
+	chtls_send_abort_rpl(sk, skb, BLOG_SKB_CB(skb)->cdev,
+			     rst_status, queue);
+	chtls_release_resources(sk);
+	chtls_conn_done(sk);
+}
+
+static void chtls_abort_rpl_rss(struct sock *sk, struct sk_buff *skb)
+{
+	struct cpl_abort_rpl_rss *rpl = cplhdr(skb) + RSS_HDR;
+	struct chtls_sock *csk;
+	struct chtls_dev *cdev;
+
+	csk = rcu_dereference_sk_user_data(sk);
+	cdev = csk->cdev;
+
+	if (csk_flag_nochk(csk, CSK_ABORT_RPL_PENDING)) {
+		csk_reset_flag(csk, CSK_ABORT_RPL_PENDING);
+		if (!csk_flag_nochk(csk, CSK_ABORT_REQ_RCVD)) {
+			if (sk->sk_state == TCP_SYN_SENT) {
+				cxgb4_remove_tid(cdev->tids,
+						 csk->port_id,
+						 GET_TID(rpl),
+						 sk->sk_family);
+				sock_put(sk);
+			}
+			chtls_release_resources(sk);
+			chtls_conn_done(sk);
+		}
+	}
+	kfree_skb(skb);
+}
+
+static int chtls_conn_cpl(struct chtls_dev *cdev, struct sk_buff *skb)
+{
+	struct cpl_peer_close *req = cplhdr(skb) + RSS_HDR;
+	void (*fn)(struct sock *sk, struct sk_buff *skb);
+	unsigned int hwtid = GET_TID(req);
+	struct chtls_sock *csk;
+	struct sock *sk;
+	u8 opcode;
+
+	opcode = ((const struct rss_header *)cplhdr(skb))->opcode;
+
+	sk = lookup_tid(cdev->tids, hwtid);
+	if (!sk)
+		goto rel_skb;
+
+	csk = sk->sk_user_data;
+
+	switch (opcode) {
+	case CPL_PEER_CLOSE:
+		fn = chtls_peer_close;
+		break;
+	case CPL_CLOSE_CON_RPL:
+		fn = chtls_close_con_rpl;
+		break;
+	case CPL_ABORT_REQ_RSS:
+		/*
+		 * Save the offload device in the skb, we may process this
+		 * message after the socket has closed.
+		 */
+		BLOG_SKB_CB(skb)->cdev = csk->cdev;
+		fn = chtls_abort_req_rss;
+		break;
+	case CPL_ABORT_RPL_RSS:
+		fn = chtls_abort_rpl_rss;
+		break;
+	default:
+		goto rel_skb;
+	}
+
+	process_cpl_msg(fn, sk, skb);
+	return 0;
+
+rel_skb:
+	kfree_skb(skb);
+	return 0;
+}
+
+static void chtls_rx_ack(struct sock *sk, struct sk_buff *skb)
+{
+	struct cpl_fw4_ack *hdr = cplhdr(skb) + RSS_HDR;
+	struct chtls_sock *csk = sk->sk_user_data;
+	struct tcp_sock *tp = tcp_sk(sk);
+	u32 credits = hdr->credits;
+	u32 snd_una;
+
+	snd_una = ntohl(hdr->snd_una);
+	csk->wr_credits += credits;
+
+	if (csk->wr_unacked > csk->wr_max_credits - csk->wr_credits)
+		csk->wr_unacked = csk->wr_max_credits - csk->wr_credits;
+
+	while (credits) {
+		struct sk_buff *pskb = csk->wr_skb_head;
+		u32 csum;
+
+		if (unlikely(!pskb)) {
+			if (csk->wr_nondata)
+				csk->wr_nondata -= credits;
+			break;
+		}
+		csum = (__force u32)pskb->csum;
+		if (unlikely(credits < csum)) {
+			pskb->csum = (__force __wsum)(csum - credits);
+			break;
+		}
+		dequeue_wr(sk);
+		credits -= csum;
+		kfree_skb(pskb);
+	}
+	if (hdr->seq_vld & CPL_FW4_ACK_FLAGS_SEQVAL) {
+		if (unlikely(before(snd_una, tp->snd_una))) {
+			kfree_skb(skb);
+			return;
+		}
+
+		if (tp->snd_una != snd_una) {
+			tp->snd_una = snd_una;
+			tp->rcv_tstamp = tcp_jiffies32;
+			if (tp->snd_una == tp->snd_nxt &&
+			    !csk_flag_nochk(csk, CSK_TX_FAILOVER))
+				csk_reset_flag(csk, CSK_TX_WAIT_IDLE);
+		}
+	}
+
+	if (hdr->seq_vld & CPL_FW4_ACK_FLAGS_CH) {
+		unsigned int fclen16 = roundup(failover_flowc_wr_len, 16);
+
+		csk->wr_credits -= fclen16;
+		csk_reset_flag(csk, CSK_TX_WAIT_IDLE);
+		csk_reset_flag(csk, CSK_TX_FAILOVER);
+	}
+	if (skb_queue_len(&csk->txq) && chtls_push_frames(csk, 0))
+		sk->sk_write_space(sk);
+
+	kfree_skb(skb);
+}
+
+static int chtls_wr_ack(struct chtls_dev *cdev, struct sk_buff *skb)
+{
+	struct cpl_fw4_ack *rpl = cplhdr(skb) + RSS_HDR;
+	unsigned int hwtid = GET_TID(rpl);
+	struct sock *sk;
+
+	sk = lookup_tid(cdev->tids, hwtid);
+	if (unlikely(!sk)) {
+		pr_err("can't find conn. for hwtid %u.\n", hwtid);
+		return -EINVAL;
+	}
+	process_cpl_msg(chtls_rx_ack, sk, skb);
+
+	return 0;
+}
+
+static int chtls_set_tcb_rpl(struct chtls_dev *cdev, struct sk_buff *skb)
+{
+	struct cpl_set_tcb_rpl *rpl = cplhdr(skb) + RSS_HDR;
+	unsigned int hwtid = GET_TID(rpl);
+	struct sock *sk;
+
+	sk = lookup_tid(cdev->tids, hwtid);
+
+	/* return EINVAL if socket doesn't exist */
+	if (!sk)
+		return -EINVAL;
+
+	/* Reusing the skb as size of cpl_set_tcb_field structure
+	 * is greater than cpl_abort_req
+	 */
+	if (TCB_COOKIE_G(rpl->cookie) == TCB_FIELD_COOKIE_TFLAG)
+		chtls_send_abort(sk, CPL_ABORT_SEND_RST, NULL);
+
+	kfree_skb(skb);
+	return 0;
+}
+
+chtls_handler_func chtls_handlers[NUM_CPL_CMDS] = {
+	[CPL_PASS_OPEN_RPL]     = chtls_pass_open_rpl,
+	[CPL_CLOSE_LISTSRV_RPL] = chtls_close_listsrv_rpl,
+	[CPL_PASS_ACCEPT_REQ]   = chtls_pass_accept_req,
+	[CPL_PASS_ESTABLISH]    = chtls_pass_establish,
+	[CPL_RX_DATA]           = chtls_rx_data,
+	[CPL_TLS_DATA]          = chtls_rx_pdu,
+	[CPL_RX_TLS_CMP]        = chtls_rx_cmp,
+	[CPL_PEER_CLOSE]        = chtls_conn_cpl,
+	[CPL_CLOSE_CON_RPL]     = chtls_conn_cpl,
+	[CPL_ABORT_REQ_RSS]     = chtls_conn_cpl,
+	[CPL_ABORT_RPL_RSS]     = chtls_conn_cpl,
+	[CPL_FW4_ACK]		= chtls_wr_ack,
+	[CPL_SET_TCB_RPL]	= chtls_set_tcb_rpl,
+};
diff --git a/drivers/net/ethernet/chelsio/inline_crypto/chtls/chtls_cm.h b/drivers/net/ethernet/chelsio/inline_crypto/chtls/chtls_cm.h
new file mode 100644
index 000000000000..29ceff5a5fcb
--- /dev/null
+++ b/drivers/net/ethernet/chelsio/inline_crypto/chtls/chtls_cm.h
@@ -0,0 +1,218 @@
+/* SPDX-License-Identifier: GPL-2.0-only */
+/*
+ * Copyright (c) 2018 Chelsio Communications, Inc.
+ */
+
+#ifndef __CHTLS_CM_H__
+#define __CHTLS_CM_H__
+
+/*
+ * TCB settings
+ */
+/* 3:0 */
+#define TCB_ULP_TYPE_W    0
+#define TCB_ULP_TYPE_S    0
+#define TCB_ULP_TYPE_M    0xfULL
+#define TCB_ULP_TYPE_V(x) ((x) << TCB_ULP_TYPE_S)
+
+/* 11:4 */
+#define TCB_ULP_RAW_W    0
+#define TCB_ULP_RAW_S    4
+#define TCB_ULP_RAW_M    0xffULL
+#define TCB_ULP_RAW_V(x) ((x) << TCB_ULP_RAW_S)
+
+#define TF_TLS_KEY_SIZE_S    7
+#define TF_TLS_KEY_SIZE_V(x) ((x) << TF_TLS_KEY_SIZE_S)
+
+#define TF_TLS_CONTROL_S     2
+#define TF_TLS_CONTROL_V(x) ((x) << TF_TLS_CONTROL_S)
+
+#define TF_TLS_ACTIVE_S      1
+#define TF_TLS_ACTIVE_V(x) ((x) << TF_TLS_ACTIVE_S)
+
+#define TF_TLS_ENABLE_S      0
+#define TF_TLS_ENABLE_V(x) ((x) << TF_TLS_ENABLE_S)
+
+#define TF_RX_QUIESCE_S    15
+#define TF_RX_QUIESCE_V(x) ((x) << TF_RX_QUIESCE_S)
+
+/*
+ * Max receive window supported by HW in bytes.  Only a small part of it can
+ * be set through option0, the rest needs to be set through RX_DATA_ACK.
+ */
+#define MAX_RCV_WND ((1U << 27) - 1)
+#define MAX_MSS     65536
+
+/*
+ * Min receive window.  We want it to be large enough to accommodate receive
+ * coalescing, handle jumbo frames, and not trigger sender SWS avoidance.
+ */
+#define MIN_RCV_WND (24 * 1024U)
+#define LOOPBACK(x)     (((x) & htonl(0xff000000)) == htonl(0x7f000000))
+
+/* for TX: a skb must have a headroom of at least TX_HEADER_LEN bytes */
+#define TX_HEADER_LEN \
+	(sizeof(struct fw_ofld_tx_data_wr) + sizeof(struct sge_opaque_hdr))
+#define TX_TLSHDR_LEN \
+	(sizeof(struct fw_tlstx_data_wr) + sizeof(struct cpl_tx_tls_sfo) + \
+	 sizeof(struct sge_opaque_hdr))
+#define TXDATA_SKB_LEN 128
+
+enum {
+	CPL_TX_TLS_SFO_TYPE_CCS,
+	CPL_TX_TLS_SFO_TYPE_ALERT,
+	CPL_TX_TLS_SFO_TYPE_HANDSHAKE,
+	CPL_TX_TLS_SFO_TYPE_DATA,
+	CPL_TX_TLS_SFO_TYPE_HEARTBEAT,
+};
+
+enum {
+	TLS_HDR_TYPE_CCS = 20,
+	TLS_HDR_TYPE_ALERT,
+	TLS_HDR_TYPE_HANDSHAKE,
+	TLS_HDR_TYPE_RECORD,
+	TLS_HDR_TYPE_HEARTBEAT,
+};
+
+typedef void (*defer_handler_t)(struct chtls_dev *dev, struct sk_buff *skb);
+extern struct request_sock_ops chtls_rsk_ops;
+extern struct request_sock_ops chtls_rsk_opsv6;
+
+struct deferred_skb_cb {
+	defer_handler_t handler;
+	struct chtls_dev *dev;
+};
+
+#define DEFERRED_SKB_CB(skb) ((struct deferred_skb_cb *)(skb)->cb)
+#define failover_flowc_wr_len offsetof(struct fw_flowc_wr, mnemval[3])
+#define WR_SKB_CB(skb) ((struct wr_skb_cb *)(skb)->cb)
+#define ACCEPT_QUEUE(sk) (&inet_csk(sk)->icsk_accept_queue.rskq_accept_head)
+
+#define SND_WSCALE(tp) ((tp)->rx_opt.snd_wscale)
+#define RCV_WSCALE(tp) ((tp)->rx_opt.rcv_wscale)
+#define USER_MSS(tp) (READ_ONCE((tp)->rx_opt.user_mss))
+#define TS_RECENT_STAMP(tp) ((tp)->rx_opt.ts_recent_stamp)
+#define WSCALE_OK(tp) ((tp)->rx_opt.wscale_ok)
+#define TSTAMP_OK(tp) ((tp)->rx_opt.tstamp_ok)
+#define SACK_OK(tp) ((tp)->rx_opt.sack_ok)
+
+/* TLS SKB */
+#define skb_ulp_tls_inline(skb)      (ULP_SKB_CB(skb)->ulp.tls.ofld)
+#define skb_ulp_tls_iv_imm(skb)      (ULP_SKB_CB(skb)->ulp.tls.iv)
+
+void chtls_defer_reply(struct sk_buff *skb, struct chtls_dev *dev,
+		       defer_handler_t handler);
+
+/*
+ * Returns true if the socket is in one of the supplied states.
+ */
+static inline unsigned int sk_in_state(const struct sock *sk,
+				       unsigned int states)
+{
+	return states & (1 << sk->sk_state);
+}
+
+static void chtls_rsk_destructor(struct request_sock *req)
+{
+	/* do nothing */
+}
+
+static inline void chtls_init_rsk_ops(struct proto *chtls_tcp_prot,
+				      struct request_sock_ops *chtls_tcp_ops,
+				      struct proto *tcp_prot, int family)
+{
+	memset(chtls_tcp_ops, 0, sizeof(*chtls_tcp_ops));
+	chtls_tcp_ops->family = family;
+	chtls_tcp_ops->obj_size = sizeof(struct tcp_request_sock);
+	chtls_tcp_ops->destructor = chtls_rsk_destructor;
+	chtls_tcp_ops->slab = tcp_prot->rsk_prot->slab;
+	chtls_tcp_prot->rsk_prot = chtls_tcp_ops;
+}
+
+static inline void chtls_reqsk_free(struct request_sock *req)
+{
+	if (req->rsk_listener)
+		sock_put(req->rsk_listener);
+	kmem_cache_free(req->rsk_ops->slab, req);
+}
+
+#define DECLARE_TASK_FUNC(task, task_param) \
+		static void task(struct work_struct *task_param)
+
+static inline void sk_wakeup_sleepers(struct sock *sk, bool interruptable)
+{
+	struct socket_wq *wq;
+
+	rcu_read_lock();
+	wq = rcu_dereference(sk->sk_wq);
+	if (skwq_has_sleeper(wq)) {
+		if (interruptable)
+			wake_up_interruptible(sk_sleep(sk));
+		else
+			wake_up_all(sk_sleep(sk));
+	}
+	rcu_read_unlock();
+}
+
+static inline void chtls_set_req_port(struct request_sock *oreq,
+				      __be16 source, __be16 dest)
+{
+	inet_rsk(oreq)->ir_rmt_port = source;
+	inet_rsk(oreq)->ir_num = ntohs(dest);
+}
+
+static inline void chtls_set_req_addr(struct request_sock *oreq,
+				      __be32 local_ip, __be32 peer_ip)
+{
+	inet_rsk(oreq)->ir_loc_addr = local_ip;
+	inet_rsk(oreq)->ir_rmt_addr = peer_ip;
+}
+
+static inline void chtls_free_skb(struct sock *sk, struct sk_buff *skb)
+{
+	skb_dstref_steal(skb);
+	__skb_unlink(skb, &sk->sk_receive_queue);
+	__kfree_skb(skb);
+}
+
+static inline void chtls_kfree_skb(struct sock *sk, struct sk_buff *skb)
+{
+	skb_dstref_steal(skb);
+	__skb_unlink(skb, &sk->sk_receive_queue);
+	kfree_skb(skb);
+}
+
+static inline void chtls_reset_wr_list(struct chtls_sock *csk)
+{
+	csk->wr_skb_head = NULL;
+	csk->wr_skb_tail = NULL;
+}
+
+static inline void enqueue_wr(struct chtls_sock *csk, struct sk_buff *skb)
+{
+	WR_SKB_CB(skb)->next_wr = NULL;
+
+	skb_get(skb);
+
+	if (!csk->wr_skb_head)
+		csk->wr_skb_head = skb;
+	else
+		WR_SKB_CB(csk->wr_skb_tail)->next_wr = skb;
+	csk->wr_skb_tail = skb;
+}
+
+static inline struct sk_buff *dequeue_wr(struct sock *sk)
+{
+	struct chtls_sock *csk = rcu_dereference_sk_user_data(sk);
+	struct sk_buff *skb = NULL;
+
+	skb = csk->wr_skb_head;
+
+	if (likely(skb)) {
+	 /* Don't bother clearing the tail */
+		csk->wr_skb_head = WR_SKB_CB(skb)->next_wr;
+		WR_SKB_CB(skb)->next_wr = NULL;
+	}
+	return skb;
+}
+#endif
diff --git a/drivers/net/ethernet/chelsio/inline_crypto/chtls/chtls_hw.c b/drivers/net/ethernet/chelsio/inline_crypto/chtls/chtls_hw.c
new file mode 100644
index 000000000000..fab6df21f01c
--- /dev/null
+++ b/drivers/net/ethernet/chelsio/inline_crypto/chtls/chtls_hw.c
@@ -0,0 +1,462 @@
+// SPDX-License-Identifier: GPL-2.0-only
+/*
+ * Copyright (c) 2018 Chelsio Communications, Inc.
+ *
+ * Written by: Atul Gupta (atul.gupta@chelsio.com)
+ */
+
+#include <linux/module.h>
+#include <linux/list.h>
+#include <linux/workqueue.h>
+#include <linux/skbuff.h>
+#include <linux/timer.h>
+#include <linux/notifier.h>
+#include <linux/inetdevice.h>
+#include <linux/ip.h>
+#include <linux/tcp.h>
+#include <linux/tls.h>
+#include <net/tls.h>
+
+#include "chtls.h"
+#include "chtls_cm.h"
+
+static void __set_tcb_field_direct(struct chtls_sock *csk,
+				   struct cpl_set_tcb_field *req, u16 word,
+				   u64 mask, u64 val, u8 cookie, int no_reply)
+{
+	struct ulptx_idata *sc;
+
+	INIT_TP_WR_CPL(req, CPL_SET_TCB_FIELD, csk->tid);
+	req->wr.wr_mid |= htonl(FW_WR_FLOWID_V(csk->tid));
+	req->reply_ctrl = htons(NO_REPLY_V(no_reply) |
+				QUEUENO_V(csk->rss_qid));
+	req->word_cookie = htons(TCB_WORD_V(word) | TCB_COOKIE_V(cookie));
+	req->mask = cpu_to_be64(mask);
+	req->val = cpu_to_be64(val);
+	sc = (struct ulptx_idata *)(req + 1);
+	sc->cmd_more = htonl(ULPTX_CMD_V(ULP_TX_SC_NOOP));
+	sc->len = htonl(0);
+}
+
+static void __set_tcb_field(struct sock *sk, struct sk_buff *skb, u16 word,
+			    u64 mask, u64 val, u8 cookie, int no_reply)
+{
+	struct cpl_set_tcb_field *req;
+	struct chtls_sock *csk;
+	struct ulptx_idata *sc;
+	unsigned int wrlen;
+
+	wrlen = roundup(sizeof(*req) + sizeof(*sc), 16);
+	csk = rcu_dereference_sk_user_data(sk);
+
+	req = (struct cpl_set_tcb_field *)__skb_put(skb, wrlen);
+	__set_tcb_field_direct(csk, req, word, mask, val, cookie, no_reply);
+	set_wr_txq(skb, CPL_PRIORITY_CONTROL, csk->port_id);
+}
+
+/*
+ * Send control message to HW, message go as immediate data and packet
+ * is freed immediately.
+ */
+static int chtls_set_tcb_field(struct sock *sk, u16 word, u64 mask, u64 val)
+{
+	struct cpl_set_tcb_field *req;
+	unsigned int credits_needed;
+	struct chtls_sock *csk;
+	struct ulptx_idata *sc;
+	struct sk_buff *skb;
+	unsigned int wrlen;
+	int ret;
+
+	wrlen = roundup(sizeof(*req) + sizeof(*sc), 16);
+
+	skb = alloc_skb(wrlen, GFP_ATOMIC);
+	if (!skb)
+		return -ENOMEM;
+
+	credits_needed = DIV_ROUND_UP(wrlen, 16);
+	csk = rcu_dereference_sk_user_data(sk);
+
+	__set_tcb_field(sk, skb, word, mask, val, 0, 1);
+	skb_set_queue_mapping(skb, (csk->txq_idx << 1) | CPL_PRIORITY_DATA);
+	csk->wr_credits -= credits_needed;
+	csk->wr_unacked += credits_needed;
+	enqueue_wr(csk, skb);
+	ret = cxgb4_ofld_send(csk->egress_dev, skb);
+	if (ret < 0)
+		kfree_skb(skb);
+	return ret < 0 ? ret : 0;
+}
+
+void chtls_set_tcb_field_rpl_skb(struct sock *sk, u16 word,
+				 u64 mask, u64 val, u8 cookie,
+				 int through_l2t)
+{
+	struct sk_buff *skb;
+	unsigned int wrlen;
+
+	wrlen = sizeof(struct cpl_set_tcb_field) + sizeof(struct ulptx_idata);
+	wrlen = roundup(wrlen, 16);
+
+	skb = alloc_skb(wrlen, GFP_KERNEL | __GFP_NOFAIL);
+	if (!skb)
+		return;
+
+	__set_tcb_field(sk, skb, word, mask, val, cookie, 0);
+	send_or_defer(sk, tcp_sk(sk), skb, through_l2t);
+}
+
+static int chtls_set_tcb_keyid(struct sock *sk, int keyid)
+{
+	return chtls_set_tcb_field(sk, 31, 0xFFFFFFFFULL, keyid);
+}
+
+static int chtls_set_tcb_seqno(struct sock *sk)
+{
+	return chtls_set_tcb_field(sk, 28, ~0ULL, 0);
+}
+
+static int chtls_set_tcb_quiesce(struct sock *sk, int val)
+{
+	return chtls_set_tcb_field(sk, 1, (1ULL << TF_RX_QUIESCE_S),
+				   TF_RX_QUIESCE_V(val));
+}
+
+void chtls_set_quiesce_ctrl(struct sock *sk, int val)
+{
+	struct chtls_sock *csk;
+	struct sk_buff *skb;
+	unsigned int wrlen;
+	int ret;
+
+	wrlen = sizeof(struct cpl_set_tcb_field) + sizeof(struct ulptx_idata);
+	wrlen = roundup(wrlen, 16);
+
+	skb = alloc_skb(wrlen, GFP_ATOMIC);
+	if (!skb)
+		return;
+
+	csk = rcu_dereference_sk_user_data(sk);
+
+	__set_tcb_field(sk, skb, 1, TF_RX_QUIESCE_V(1), 0, 0, 1);
+	set_wr_txq(skb, CPL_PRIORITY_CONTROL, csk->port_id);
+	ret = cxgb4_ofld_send(csk->egress_dev, skb);
+	if (ret < 0)
+		kfree_skb(skb);
+}
+
+/* TLS Key bitmap processing */
+int chtls_init_kmap(struct chtls_dev *cdev, struct cxgb4_lld_info *lldi)
+{
+	unsigned int num_key_ctx, bsize;
+	int ksize;
+
+	num_key_ctx = (lldi->vr->key.size / TLS_KEY_CONTEXT_SZ);
+	bsize = BITS_TO_LONGS(num_key_ctx);
+
+	cdev->kmap.size = num_key_ctx;
+	cdev->kmap.available = bsize;
+	ksize = sizeof(*cdev->kmap.addr) * bsize;
+	cdev->kmap.addr = kvzalloc(ksize, GFP_KERNEL);
+	if (!cdev->kmap.addr)
+		return -ENOMEM;
+
+	cdev->kmap.start = lldi->vr->key.start;
+	spin_lock_init(&cdev->kmap.lock);
+	return 0;
+}
+
+static int get_new_keyid(struct chtls_sock *csk, u32 optname)
+{
+	struct net_device *dev = csk->egress_dev;
+	struct chtls_dev *cdev = csk->cdev;
+	struct chtls_hws *hws;
+	struct adapter *adap;
+	int keyid;
+
+	adap = netdev2adap(dev);
+	hws = &csk->tlshws;
+
+	spin_lock_bh(&cdev->kmap.lock);
+	keyid = find_first_zero_bit(cdev->kmap.addr, cdev->kmap.size);
+	if (keyid < cdev->kmap.size) {
+		__set_bit(keyid, cdev->kmap.addr);
+		if (optname == TLS_RX)
+			hws->rxkey = keyid;
+		else
+			hws->txkey = keyid;
+		atomic_inc(&adap->chcr_stats.tls_key);
+	} else {
+		keyid = -1;
+	}
+	spin_unlock_bh(&cdev->kmap.lock);
+	return keyid;
+}
+
+void free_tls_keyid(struct sock *sk)
+{
+	struct chtls_sock *csk = rcu_dereference_sk_user_data(sk);
+	struct net_device *dev = csk->egress_dev;
+	struct chtls_dev *cdev = csk->cdev;
+	struct chtls_hws *hws;
+	struct adapter *adap;
+
+	if (!cdev->kmap.addr)
+		return;
+
+	adap = netdev2adap(dev);
+	hws = &csk->tlshws;
+
+	spin_lock_bh(&cdev->kmap.lock);
+	if (hws->rxkey >= 0) {
+		__clear_bit(hws->rxkey, cdev->kmap.addr);
+		atomic_dec(&adap->chcr_stats.tls_key);
+		hws->rxkey = -1;
+	}
+	if (hws->txkey >= 0) {
+		__clear_bit(hws->txkey, cdev->kmap.addr);
+		atomic_dec(&adap->chcr_stats.tls_key);
+		hws->txkey = -1;
+	}
+	spin_unlock_bh(&cdev->kmap.lock);
+}
+
+unsigned int keyid_to_addr(int start_addr, int keyid)
+{
+	return (start_addr + (keyid * TLS_KEY_CONTEXT_SZ)) >> 5;
+}
+
+static void chtls_rxkey_ivauth(struct _key_ctx *kctx)
+{
+	kctx->iv_to_auth = cpu_to_be64(KEYCTX_TX_WR_IV_V(6ULL) |
+				  KEYCTX_TX_WR_AAD_V(1ULL) |
+				  KEYCTX_TX_WR_AADST_V(5ULL) |
+				  KEYCTX_TX_WR_CIPHER_V(14ULL) |
+				  KEYCTX_TX_WR_CIPHERST_V(0ULL) |
+				  KEYCTX_TX_WR_AUTH_V(14ULL) |
+				  KEYCTX_TX_WR_AUTHST_V(16ULL) |
+				  KEYCTX_TX_WR_AUTHIN_V(16ULL));
+}
+
+static int chtls_key_info(struct chtls_sock *csk,
+			  struct _key_ctx *kctx,
+			  u32 keylen, u32 optname,
+			  int cipher_type)
+{
+	unsigned char key[AES_MAX_KEY_SIZE];
+	unsigned char *key_p, *salt;
+	unsigned char ghash_h[AEAD_H_SIZE];
+	int ck_size, key_ctx_size, kctx_mackey_size, salt_size;
+	struct crypto_aes_ctx aes;
+	int ret;
+
+	key_ctx_size = sizeof(struct _key_ctx) +
+		       roundup(keylen, 16) + AEAD_H_SIZE;
+
+	/* GCM mode of AES supports 128 and 256 bit encryption, so
+	 * prepare key context base on GCM cipher type
+	 */
+	switch (cipher_type) {
+	case TLS_CIPHER_AES_GCM_128: {
+		struct tls12_crypto_info_aes_gcm_128 *gcm_ctx_128 =
+			(struct tls12_crypto_info_aes_gcm_128 *)
+					&csk->tlshws.crypto_info;
+		memcpy(key, gcm_ctx_128->key, keylen);
+
+		key_p            = gcm_ctx_128->key;
+		salt             = gcm_ctx_128->salt;
+		ck_size          = CHCR_KEYCTX_CIPHER_KEY_SIZE_128;
+		salt_size        = TLS_CIPHER_AES_GCM_128_SALT_SIZE;
+		kctx_mackey_size = CHCR_KEYCTX_MAC_KEY_SIZE_128;
+		break;
+	}
+	case TLS_CIPHER_AES_GCM_256: {
+		struct tls12_crypto_info_aes_gcm_256 *gcm_ctx_256 =
+			(struct tls12_crypto_info_aes_gcm_256 *)
+					&csk->tlshws.crypto_info;
+		memcpy(key, gcm_ctx_256->key, keylen);
+
+		key_p            = gcm_ctx_256->key;
+		salt             = gcm_ctx_256->salt;
+		ck_size          = CHCR_KEYCTX_CIPHER_KEY_SIZE_256;
+		salt_size        = TLS_CIPHER_AES_GCM_256_SALT_SIZE;
+		kctx_mackey_size = CHCR_KEYCTX_MAC_KEY_SIZE_256;
+		break;
+	}
+	default:
+		pr_err("GCM: Invalid key length %d\n", keylen);
+		return -EINVAL;
+	}
+
+	/* Calculate the H = CIPH(K, 0 repeated 16 times).
+	 * It will go in key context
+	 */
+	ret = aes_expandkey(&aes, key, keylen);
+	if (ret)
+		return ret;
+
+	memset(ghash_h, 0, AEAD_H_SIZE);
+	aes_encrypt(&aes, ghash_h, ghash_h);
+	memzero_explicit(&aes, sizeof(aes));
+	csk->tlshws.keylen = key_ctx_size;
+
+	/* Copy the Key context */
+	if (optname == TLS_RX) {
+		int key_ctx;
+
+		key_ctx = ((key_ctx_size >> 4) << 3);
+		kctx->ctx_hdr = FILL_KEY_CRX_HDR(ck_size,
+						 kctx_mackey_size,
+						 0, 0, key_ctx);
+		chtls_rxkey_ivauth(kctx);
+	} else {
+		kctx->ctx_hdr = FILL_KEY_CTX_HDR(ck_size,
+						 kctx_mackey_size,
+						 0, 0, key_ctx_size >> 4);
+	}
+
+	memcpy(kctx->salt, salt, salt_size);
+	memcpy(kctx->key, key_p, keylen);
+	memcpy(kctx->key + keylen, ghash_h, AEAD_H_SIZE);
+	/* erase key info from driver */
+	memset(key_p, 0, keylen);
+
+	return 0;
+}
+
+static void chtls_set_scmd(struct chtls_sock *csk)
+{
+	struct chtls_hws *hws = &csk->tlshws;
+
+	hws->scmd.seqno_numivs =
+		SCMD_SEQ_NO_CTRL_V(3) |
+		SCMD_PROTO_VERSION_V(0) |
+		SCMD_ENC_DEC_CTRL_V(0) |
+		SCMD_CIPH_AUTH_SEQ_CTRL_V(1) |
+		SCMD_CIPH_MODE_V(2) |
+		SCMD_AUTH_MODE_V(4) |
+		SCMD_HMAC_CTRL_V(0) |
+		SCMD_IV_SIZE_V(4) |
+		SCMD_NUM_IVS_V(1);
+
+	hws->scmd.ivgen_hdrlen =
+		SCMD_IV_GEN_CTRL_V(1) |
+		SCMD_KEY_CTX_INLINE_V(0) |
+		SCMD_TLS_FRAG_ENABLE_V(1);
+}
+
+int chtls_setkey(struct chtls_sock *csk, u32 keylen,
+		 u32 optname, int cipher_type)
+{
+	struct tls_key_req *kwr;
+	struct chtls_dev *cdev;
+	struct _key_ctx *kctx;
+	int wrlen, klen, len;
+	struct sk_buff *skb;
+	struct sock *sk;
+	int keyid;
+	int kaddr;
+	int ret;
+
+	cdev = csk->cdev;
+	sk = csk->sk;
+
+	klen = roundup((keylen + AEAD_H_SIZE) + sizeof(*kctx), 32);
+	wrlen = roundup(sizeof(*kwr), 16);
+	len = klen + wrlen;
+
+	/* Flush out-standing data before new key takes effect */
+	if (optname == TLS_TX) {
+		lock_sock(sk);
+		if (skb_queue_len(&csk->txq))
+			chtls_push_frames(csk, 0);
+		release_sock(sk);
+	}
+
+	skb = alloc_skb(len, GFP_KERNEL);
+	if (!skb)
+		return -ENOMEM;
+
+	keyid = get_new_keyid(csk, optname);
+	if (keyid < 0) {
+		ret = -ENOSPC;
+		goto out_nokey;
+	}
+
+	kaddr = keyid_to_addr(cdev->kmap.start, keyid);
+	kwr = (struct tls_key_req *)__skb_put_zero(skb, len);
+	kwr->wr.op_to_compl =
+		cpu_to_be32(FW_WR_OP_V(FW_ULPTX_WR) | FW_WR_COMPL_F |
+		      FW_WR_ATOMIC_V(1U));
+	kwr->wr.flowid_len16 =
+		cpu_to_be32(FW_WR_LEN16_V(DIV_ROUND_UP(len, 16) |
+			    FW_WR_FLOWID_V(csk->tid)));
+	kwr->wr.protocol = 0;
+	kwr->wr.mfs = htons(TLS_MFS);
+	kwr->wr.reneg_to_write_rx = optname;
+
+	/* ulptx command */
+	kwr->req.cmd = cpu_to_be32(ULPTX_CMD_V(ULP_TX_MEM_WRITE) |
+			    T5_ULP_MEMIO_ORDER_V(1) |
+			    T5_ULP_MEMIO_IMM_V(1));
+	kwr->req.len16 = cpu_to_be32((csk->tid << 8) |
+			      DIV_ROUND_UP(len - sizeof(kwr->wr), 16));
+	kwr->req.dlen = cpu_to_be32(ULP_MEMIO_DATA_LEN_V(klen >> 5));
+	kwr->req.lock_addr = cpu_to_be32(ULP_MEMIO_ADDR_V(kaddr));
+
+	/* sub command */
+	kwr->sc_imm.cmd_more = cpu_to_be32(ULPTX_CMD_V(ULP_TX_SC_IMM));
+	kwr->sc_imm.len = cpu_to_be32(klen);
+
+	lock_sock(sk);
+	/* key info */
+	kctx = (struct _key_ctx *)(kwr + 1);
+	ret = chtls_key_info(csk, kctx, keylen, optname, cipher_type);
+	if (ret)
+		goto out_notcb;
+
+	if (unlikely(csk_flag(sk, CSK_ABORT_SHUTDOWN)))
+		goto out_notcb;
+
+	set_wr_txq(skb, CPL_PRIORITY_DATA, csk->tlshws.txqid);
+	csk->wr_credits -= DIV_ROUND_UP(len, 16);
+	csk->wr_unacked += DIV_ROUND_UP(len, 16);
+	enqueue_wr(csk, skb);
+	cxgb4_ofld_send(csk->egress_dev, skb);
+	skb = NULL;
+
+	chtls_set_scmd(csk);
+	/* Clear quiesce for Rx key */
+	if (optname == TLS_RX) {
+		ret = chtls_set_tcb_keyid(sk, keyid);
+		if (ret)
+			goto out_notcb;
+		ret = chtls_set_tcb_field(sk, 0,
+					  TCB_ULP_RAW_V(TCB_ULP_RAW_M),
+					  TCB_ULP_RAW_V((TF_TLS_KEY_SIZE_V(1) |
+					  TF_TLS_CONTROL_V(1) |
+					  TF_TLS_ACTIVE_V(1) |
+					  TF_TLS_ENABLE_V(1))));
+		if (ret)
+			goto out_notcb;
+		ret = chtls_set_tcb_seqno(sk);
+		if (ret)
+			goto out_notcb;
+		ret = chtls_set_tcb_quiesce(sk, 0);
+		if (ret)
+			goto out_notcb;
+		csk->tlshws.rxkey = keyid;
+	} else {
+		csk->tlshws.tx_seq_no = 0;
+		csk->tlshws.txkey = keyid;
+	}
+
+	release_sock(sk);
+	return ret;
+out_notcb:
+	release_sock(sk);
+	free_tls_keyid(sk);
+out_nokey:
+	kfree_skb(skb);
+	return ret;
+}
diff --git a/drivers/net/ethernet/chelsio/inline_crypto/chtls/chtls_io.c b/drivers/net/ethernet/chelsio/inline_crypto/chtls/chtls_io.c
new file mode 100644
index 000000000000..4036db466e18
--- /dev/null
+++ b/drivers/net/ethernet/chelsio/inline_crypto/chtls/chtls_io.c
@@ -0,0 +1,1842 @@
+// SPDX-License-Identifier: GPL-2.0-only
+/*
+ * Copyright (c) 2018 Chelsio Communications, Inc.
+ *
+ * Written by: Atul Gupta (atul.gupta@chelsio.com)
+ */
+
+#include <linux/module.h>
+#include <linux/list.h>
+#include <linux/workqueue.h>
+#include <linux/skbuff.h>
+#include <linux/timer.h>
+#include <linux/notifier.h>
+#include <linux/inetdevice.h>
+#include <linux/ip.h>
+#include <linux/tcp.h>
+#include <linux/sched/signal.h>
+#include <net/tcp.h>
+#include <net/busy_poll.h>
+#include <crypto/aes.h>
+
+#include "chtls.h"
+#include "chtls_cm.h"
+
+static bool is_tls_tx(struct chtls_sock *csk)
+{
+	return csk->tlshws.txkey >= 0;
+}
+
+static bool is_tls_rx(struct chtls_sock *csk)
+{
+	return csk->tlshws.rxkey >= 0;
+}
+
+static int data_sgl_len(const struct sk_buff *skb)
+{
+	unsigned int cnt;
+
+	cnt = skb_shinfo(skb)->nr_frags;
+	return sgl_len(cnt) * 8;
+}
+
+static int nos_ivs(struct sock *sk, unsigned int size)
+{
+	struct chtls_sock *csk = rcu_dereference_sk_user_data(sk);
+
+	return DIV_ROUND_UP(size, csk->tlshws.mfs);
+}
+
+static int set_ivs_imm(struct sock *sk, const struct sk_buff *skb)
+{
+	int ivs_size = nos_ivs(sk, skb->len) * CIPHER_BLOCK_SIZE;
+	int hlen = TLS_WR_CPL_LEN + data_sgl_len(skb);
+
+	if ((hlen + KEY_ON_MEM_SZ + ivs_size) <
+	    MAX_IMM_OFLD_TX_DATA_WR_LEN) {
+		ULP_SKB_CB(skb)->ulp.tls.iv = 1;
+		return 1;
+	}
+	ULP_SKB_CB(skb)->ulp.tls.iv = 0;
+	return 0;
+}
+
+static int max_ivs_size(struct sock *sk, int size)
+{
+	return nos_ivs(sk, size) * CIPHER_BLOCK_SIZE;
+}
+
+static int ivs_size(struct sock *sk, const struct sk_buff *skb)
+{
+	return set_ivs_imm(sk, skb) ? (nos_ivs(sk, skb->len) *
+		 CIPHER_BLOCK_SIZE) : 0;
+}
+
+static int flowc_wr_credits(int nparams, int *flowclenp)
+{
+	int flowclen16, flowclen;
+
+	flowclen = offsetof(struct fw_flowc_wr, mnemval[nparams]);
+	flowclen16 = DIV_ROUND_UP(flowclen, 16);
+	flowclen = flowclen16 * 16;
+
+	if (flowclenp)
+		*flowclenp = flowclen;
+
+	return flowclen16;
+}
+
+static struct sk_buff *create_flowc_wr_skb(struct sock *sk,
+					   struct fw_flowc_wr *flowc,
+					   int flowclen)
+{
+	struct chtls_sock *csk = rcu_dereference_sk_user_data(sk);
+	struct sk_buff *skb;
+
+	skb = alloc_skb(flowclen, GFP_ATOMIC);
+	if (!skb)
+		return NULL;
+
+	__skb_put_data(skb, flowc, flowclen);
+	skb_set_queue_mapping(skb, (csk->txq_idx << 1) | CPL_PRIORITY_DATA);
+
+	return skb;
+}
+
+static int send_flowc_wr(struct sock *sk, struct fw_flowc_wr *flowc,
+			 int flowclen)
+{
+	struct chtls_sock *csk = rcu_dereference_sk_user_data(sk);
+	struct tcp_sock *tp = tcp_sk(sk);
+	struct sk_buff *skb;
+	int flowclen16;
+	int ret;
+
+	flowclen16 = flowclen / 16;
+
+	if (csk_flag(sk, CSK_TX_DATA_SENT)) {
+		skb = create_flowc_wr_skb(sk, flowc, flowclen);
+		if (!skb)
+			return -ENOMEM;
+
+		skb_entail(sk, skb,
+			   ULPCB_FLAG_NO_HDR | ULPCB_FLAG_NO_APPEND);
+		return 0;
+	}
+
+	ret = cxgb4_immdata_send(csk->egress_dev,
+				 csk->txq_idx,
+				 flowc, flowclen);
+	if (!ret)
+		return flowclen16;
+	skb = create_flowc_wr_skb(sk, flowc, flowclen);
+	if (!skb)
+		return -ENOMEM;
+	send_or_defer(sk, tp, skb, 0);
+	return flowclen16;
+}
+
+static u8 tcp_state_to_flowc_state(u8 state)
+{
+	switch (state) {
+	case TCP_ESTABLISHED:
+		return FW_FLOWC_MNEM_TCPSTATE_ESTABLISHED;
+	case TCP_CLOSE_WAIT:
+		return FW_FLOWC_MNEM_TCPSTATE_CLOSEWAIT;
+	case TCP_FIN_WAIT1:
+		return FW_FLOWC_MNEM_TCPSTATE_FINWAIT1;
+	case TCP_CLOSING:
+		return FW_FLOWC_MNEM_TCPSTATE_CLOSING;
+	case TCP_LAST_ACK:
+		return FW_FLOWC_MNEM_TCPSTATE_LASTACK;
+	case TCP_FIN_WAIT2:
+		return FW_FLOWC_MNEM_TCPSTATE_FINWAIT2;
+	}
+
+	return FW_FLOWC_MNEM_TCPSTATE_ESTABLISHED;
+}
+
+int send_tx_flowc_wr(struct sock *sk, int compl,
+		     u32 snd_nxt, u32 rcv_nxt)
+{
+	struct flowc_packed {
+		struct fw_flowc_wr fc;
+		struct fw_flowc_mnemval mnemval[FW_FLOWC_MNEM_MAX];
+	} __packed sflowc;
+	int nparams, paramidx, flowclen16, flowclen;
+	struct fw_flowc_wr *flowc;
+	struct chtls_sock *csk;
+	struct tcp_sock *tp;
+
+	csk = rcu_dereference_sk_user_data(sk);
+	tp = tcp_sk(sk);
+	memset(&sflowc, 0, sizeof(sflowc));
+	flowc = &sflowc.fc;
+
+#define FLOWC_PARAM(__m, __v) \
+	do { \
+		flowc->mnemval[paramidx].mnemonic = FW_FLOWC_MNEM_##__m; \
+		flowc->mnemval[paramidx].val = cpu_to_be32(__v); \
+		paramidx++; \
+	} while (0)
+
+	paramidx = 0;
+
+	FLOWC_PARAM(PFNVFN, FW_PFVF_CMD_PFN_V(csk->cdev->lldi->pf));
+	FLOWC_PARAM(CH, csk->tx_chan);
+	FLOWC_PARAM(PORT, csk->tx_chan);
+	FLOWC_PARAM(IQID, csk->rss_qid);
+	FLOWC_PARAM(SNDNXT, tp->snd_nxt);
+	FLOWC_PARAM(RCVNXT, tp->rcv_nxt);
+	FLOWC_PARAM(SNDBUF, csk->sndbuf);
+	FLOWC_PARAM(MSS, tp->mss_cache);
+	FLOWC_PARAM(TCPSTATE, tcp_state_to_flowc_state(sk->sk_state));
+
+	if (SND_WSCALE(tp))
+		FLOWC_PARAM(RCV_SCALE, SND_WSCALE(tp));
+
+	if (csk->ulp_mode == ULP_MODE_TLS)
+		FLOWC_PARAM(ULD_MODE, ULP_MODE_TLS);
+
+	if (csk->tlshws.fcplenmax)
+		FLOWC_PARAM(TXDATAPLEN_MAX, csk->tlshws.fcplenmax);
+
+	nparams = paramidx;
+#undef FLOWC_PARAM
+
+	flowclen16 = flowc_wr_credits(nparams, &flowclen);
+	flowc->op_to_nparams =
+		cpu_to_be32(FW_WR_OP_V(FW_FLOWC_WR) |
+			    FW_WR_COMPL_V(compl) |
+			    FW_FLOWC_WR_NPARAMS_V(nparams));
+	flowc->flowid_len16 = cpu_to_be32(FW_WR_LEN16_V(flowclen16) |
+					  FW_WR_FLOWID_V(csk->tid));
+
+	return send_flowc_wr(sk, flowc, flowclen);
+}
+
+/* Copy IVs to WR */
+static int tls_copy_ivs(struct sock *sk, struct sk_buff *skb)
+
+{
+	struct chtls_sock *csk;
+	unsigned char *iv_loc;
+	struct chtls_hws *hws;
+	unsigned char *ivs;
+	u16 number_of_ivs;
+	struct page *page;
+	int err = 0;
+
+	csk = rcu_dereference_sk_user_data(sk);
+	hws = &csk->tlshws;
+	number_of_ivs = nos_ivs(sk, skb->len);
+
+	if (number_of_ivs > MAX_IVS_PAGE) {
+		pr_warn("MAX IVs in PAGE exceeded %d\n", number_of_ivs);
+		return -ENOMEM;
+	}
+
+	/* generate the  IVs */
+	ivs = kmalloc_array(CIPHER_BLOCK_SIZE, number_of_ivs, GFP_ATOMIC);
+	if (!ivs)
+		return -ENOMEM;
+	get_random_bytes(ivs, number_of_ivs * CIPHER_BLOCK_SIZE);
+
+	if (skb_ulp_tls_iv_imm(skb)) {
+		/* send the IVs as immediate data in the WR */
+		iv_loc = (unsigned char *)__skb_push(skb, number_of_ivs *
+						CIPHER_BLOCK_SIZE);
+		if (iv_loc)
+			memcpy(iv_loc, ivs, number_of_ivs * CIPHER_BLOCK_SIZE);
+
+		hws->ivsize = number_of_ivs * CIPHER_BLOCK_SIZE;
+	} else {
+		/* Send the IVs as sgls */
+		/* Already accounted IV DSGL for credits */
+		skb_shinfo(skb)->nr_frags--;
+		page = alloc_pages(sk->sk_allocation | __GFP_COMP, 0);
+		if (!page) {
+			pr_info("%s : Page allocation for IVs failed\n",
+				__func__);
+			err = -ENOMEM;
+			goto out;
+		}
+		memcpy(page_address(page), ivs, number_of_ivs *
+		       CIPHER_BLOCK_SIZE);
+		skb_fill_page_desc(skb, skb_shinfo(skb)->nr_frags, page, 0,
+				   number_of_ivs * CIPHER_BLOCK_SIZE);
+		hws->ivsize = 0;
+	}
+out:
+	kfree(ivs);
+	return err;
+}
+
+/* Copy Key to WR */
+static void tls_copy_tx_key(struct sock *sk, struct sk_buff *skb)
+{
+	struct ulptx_sc_memrd *sc_memrd;
+	struct chtls_sock *csk;
+	struct chtls_dev *cdev;
+	struct ulptx_idata *sc;
+	struct chtls_hws *hws;
+	u32 immdlen;
+	int kaddr;
+
+	csk = rcu_dereference_sk_user_data(sk);
+	hws = &csk->tlshws;
+	cdev = csk->cdev;
+
+	immdlen = sizeof(*sc) + sizeof(*sc_memrd);
+	kaddr = keyid_to_addr(cdev->kmap.start, hws->txkey);
+	sc = (struct ulptx_idata *)__skb_push(skb, immdlen);
+	if (sc) {
+		sc->cmd_more = htonl(ULPTX_CMD_V(ULP_TX_SC_NOOP));
+		sc->len = htonl(0);
+		sc_memrd = (struct ulptx_sc_memrd *)(sc + 1);
+		sc_memrd->cmd_to_len =
+				htonl(ULPTX_CMD_V(ULP_TX_SC_MEMRD) |
+				ULP_TX_SC_MORE_V(1) |
+				ULPTX_LEN16_V(hws->keylen >> 4));
+		sc_memrd->addr = htonl(kaddr);
+	}
+}
+
+static u64 tlstx_incr_seqnum(struct chtls_hws *hws)
+{
+	return hws->tx_seq_no++;
+}
+
+static bool is_sg_request(const struct sk_buff *skb)
+{
+	return skb->peeked ||
+		(skb->len > MAX_IMM_ULPTX_WR_LEN);
+}
+
+/*
+ * Returns true if an sk_buff carries urgent data.
+ */
+static bool skb_urgent(struct sk_buff *skb)
+{
+	return ULP_SKB_CB(skb)->flags & ULPCB_FLAG_URG;
+}
+
+/* TLS content type for CPL SFO */
+static unsigned char tls_content_type(unsigned char content_type)
+{
+	switch (content_type) {
+	case TLS_HDR_TYPE_CCS:
+		return CPL_TX_TLS_SFO_TYPE_CCS;
+	case TLS_HDR_TYPE_ALERT:
+		return CPL_TX_TLS_SFO_TYPE_ALERT;
+	case TLS_HDR_TYPE_HANDSHAKE:
+		return CPL_TX_TLS_SFO_TYPE_HANDSHAKE;
+	case TLS_HDR_TYPE_HEARTBEAT:
+		return CPL_TX_TLS_SFO_TYPE_HEARTBEAT;
+	}
+	return CPL_TX_TLS_SFO_TYPE_DATA;
+}
+
+static void tls_tx_data_wr(struct sock *sk, struct sk_buff *skb,
+			   int dlen, int tls_immd, u32 credits,
+			   int expn, int pdus)
+{
+	struct fw_tlstx_data_wr *req_wr;
+	struct cpl_tx_tls_sfo *req_cpl;
+	unsigned int wr_ulp_mode_force;
+	struct tls_scmd *updated_scmd;
+	unsigned char data_type;
+	struct chtls_sock *csk;
+	struct net_device *dev;
+	struct chtls_hws *hws;
+	struct tls_scmd *scmd;
+	struct adapter *adap;
+	unsigned char *req;
+	int immd_len;
+	int iv_imm;
+	int len;
+
+	csk = rcu_dereference_sk_user_data(sk);
+	iv_imm = skb_ulp_tls_iv_imm(skb);
+	dev = csk->egress_dev;
+	adap = netdev2adap(dev);
+	hws = &csk->tlshws;
+	scmd = &hws->scmd;
+	len = dlen + expn;
+
+	dlen = (dlen < hws->mfs) ? dlen : hws->mfs;
+	atomic_inc(&adap->chcr_stats.tls_pdu_tx);
+
+	updated_scmd = scmd;
+	updated_scmd->seqno_numivs &= 0xffffff80;
+	updated_scmd->seqno_numivs |= SCMD_NUM_IVS_V(pdus);
+	hws->scmd = *updated_scmd;
+
+	req = (unsigned char *)__skb_push(skb, sizeof(struct cpl_tx_tls_sfo));
+	req_cpl = (struct cpl_tx_tls_sfo *)req;
+	req = (unsigned char *)__skb_push(skb, (sizeof(struct
+				fw_tlstx_data_wr)));
+
+	req_wr = (struct fw_tlstx_data_wr *)req;
+	immd_len = (tls_immd ? dlen : 0);
+	req_wr->op_to_immdlen =
+		htonl(FW_WR_OP_V(FW_TLSTX_DATA_WR) |
+		FW_TLSTX_DATA_WR_COMPL_V(1) |
+		FW_TLSTX_DATA_WR_IMMDLEN_V(immd_len));
+	req_wr->flowid_len16 = htonl(FW_TLSTX_DATA_WR_FLOWID_V(csk->tid) |
+				     FW_TLSTX_DATA_WR_LEN16_V(credits));
+	wr_ulp_mode_force = TX_ULP_MODE_V(ULP_MODE_TLS);
+
+	if (is_sg_request(skb))
+		wr_ulp_mode_force |= FW_OFLD_TX_DATA_WR_ALIGNPLD_F |
+			((tcp_sk(sk)->nonagle & TCP_NAGLE_OFF) ? 0 :
+			FW_OFLD_TX_DATA_WR_SHOVE_F);
+
+	req_wr->lsodisable_to_flags =
+			htonl(TX_ULP_MODE_V(ULP_MODE_TLS) |
+			      TX_URG_V(skb_urgent(skb)) |
+			      T6_TX_FORCE_F | wr_ulp_mode_force |
+			      TX_SHOVE_V((!csk_flag(sk, CSK_TX_MORE_DATA)) &&
+					 skb_queue_empty(&csk->txq)));
+
+	req_wr->ctxloc_to_exp =
+			htonl(FW_TLSTX_DATA_WR_NUMIVS_V(pdus) |
+			      FW_TLSTX_DATA_WR_EXP_V(expn) |
+			      FW_TLSTX_DATA_WR_CTXLOC_V(CHTLS_KEY_CONTEXT_DDR) |
+			      FW_TLSTX_DATA_WR_IVDSGL_V(!iv_imm) |
+			      FW_TLSTX_DATA_WR_KEYSIZE_V(hws->keylen >> 4));
+
+	/* Fill in the length */
+	req_wr->plen = htonl(len);
+	req_wr->mfs = htons(hws->mfs);
+	req_wr->adjustedplen_pkd =
+		htons(FW_TLSTX_DATA_WR_ADJUSTEDPLEN_V(hws->adjustlen));
+	req_wr->expinplenmax_pkd =
+		htons(FW_TLSTX_DATA_WR_EXPINPLENMAX_V(hws->expansion));
+	req_wr->pdusinplenmax_pkd =
+		FW_TLSTX_DATA_WR_PDUSINPLENMAX_V(hws->pdus);
+	req_wr->r10 = 0;
+
+	data_type = tls_content_type(ULP_SKB_CB(skb)->ulp.tls.type);
+	req_cpl->op_to_seg_len = htonl(CPL_TX_TLS_SFO_OPCODE_V(CPL_TX_TLS_SFO) |
+				       CPL_TX_TLS_SFO_DATA_TYPE_V(data_type) |
+				       CPL_TX_TLS_SFO_CPL_LEN_V(2) |
+				       CPL_TX_TLS_SFO_SEG_LEN_V(dlen));
+	req_cpl->pld_len = htonl(len - expn);
+
+	req_cpl->type_protover = htonl(CPL_TX_TLS_SFO_TYPE_V
+		((data_type == CPL_TX_TLS_SFO_TYPE_HEARTBEAT) ?
+		TLS_HDR_TYPE_HEARTBEAT : 0) |
+		CPL_TX_TLS_SFO_PROTOVER_V(0));
+
+	/* create the s-command */
+	req_cpl->r1_lo = 0;
+	req_cpl->seqno_numivs  = cpu_to_be32(hws->scmd.seqno_numivs);
+	req_cpl->ivgen_hdrlen = cpu_to_be32(hws->scmd.ivgen_hdrlen);
+	req_cpl->scmd1 = cpu_to_be64(tlstx_incr_seqnum(hws));
+}
+
+/*
+ * Calculate the TLS data expansion size
+ */
+static int chtls_expansion_size(struct sock *sk, int data_len,
+				int fullpdu,
+				unsigned short *pducnt)
+{
+	struct chtls_sock *csk = rcu_dereference_sk_user_data(sk);
+	struct chtls_hws *hws = &csk->tlshws;
+	struct tls_scmd *scmd = &hws->scmd;
+	int fragsize = hws->mfs;
+	int expnsize = 0;
+	int fragleft;
+	int fragcnt;
+	int expppdu;
+
+	if (SCMD_CIPH_MODE_G(scmd->seqno_numivs) ==
+	    SCMD_CIPH_MODE_AES_GCM) {
+		expppdu = GCM_TAG_SIZE + AEAD_EXPLICIT_DATA_SIZE +
+			  TLS_HEADER_LENGTH;
+
+		if (fullpdu) {
+			*pducnt = data_len / (expppdu + fragsize);
+			if (*pducnt > 32)
+				*pducnt = 32;
+			else if (!*pducnt)
+				*pducnt = 1;
+			expnsize = (*pducnt) * expppdu;
+			return expnsize;
+		}
+		fragcnt = (data_len / fragsize);
+		expnsize =  fragcnt * expppdu;
+		fragleft = data_len % fragsize;
+		if (fragleft > 0)
+			expnsize += expppdu;
+	}
+	return expnsize;
+}
+
+/* WR with IV, KEY and CPL SFO added */
+static void make_tlstx_data_wr(struct sock *sk, struct sk_buff *skb,
+			       int tls_tx_imm, int tls_len, u32 credits)
+{
+	unsigned short pdus_per_ulp = 0;
+	struct chtls_sock *csk;
+	struct chtls_hws *hws;
+	int expn_sz;
+	int pdus;
+
+	csk = rcu_dereference_sk_user_data(sk);
+	hws = &csk->tlshws;
+	pdus = DIV_ROUND_UP(tls_len, hws->mfs);
+	expn_sz = chtls_expansion_size(sk, tls_len, 0, NULL);
+	if (!hws->compute) {
+		hws->expansion = chtls_expansion_size(sk,
+						      hws->fcplenmax,
+						      1, &pdus_per_ulp);
+		hws->pdus = pdus_per_ulp;
+		hws->adjustlen = hws->pdus *
+			((hws->expansion / hws->pdus) + hws->mfs);
+		hws->compute = 1;
+	}
+	if (tls_copy_ivs(sk, skb))
+		return;
+	tls_copy_tx_key(sk, skb);
+	tls_tx_data_wr(sk, skb, tls_len, tls_tx_imm, credits, expn_sz, pdus);
+	hws->tx_seq_no += (pdus - 1);
+}
+
+static void make_tx_data_wr(struct sock *sk, struct sk_buff *skb,
+			    unsigned int immdlen, int len,
+			    u32 credits, u32 compl)
+{
+	struct fw_ofld_tx_data_wr *req;
+	unsigned int wr_ulp_mode_force;
+	struct chtls_sock *csk;
+	unsigned int opcode;
+
+	csk = rcu_dereference_sk_user_data(sk);
+	opcode = FW_OFLD_TX_DATA_WR;
+
+	req = (struct fw_ofld_tx_data_wr *)__skb_push(skb, sizeof(*req));
+	req->op_to_immdlen = htonl(WR_OP_V(opcode) |
+				FW_WR_COMPL_V(compl) |
+				FW_WR_IMMDLEN_V(immdlen));
+	req->flowid_len16 = htonl(FW_WR_FLOWID_V(csk->tid) |
+				FW_WR_LEN16_V(credits));
+
+	wr_ulp_mode_force = TX_ULP_MODE_V(csk->ulp_mode);
+	if (is_sg_request(skb))
+		wr_ulp_mode_force |= FW_OFLD_TX_DATA_WR_ALIGNPLD_F |
+			((tcp_sk(sk)->nonagle & TCP_NAGLE_OFF) ? 0 :
+				FW_OFLD_TX_DATA_WR_SHOVE_F);
+
+	req->tunnel_to_proxy = htonl(wr_ulp_mode_force |
+			TX_URG_V(skb_urgent(skb)) |
+			TX_SHOVE_V((!csk_flag(sk, CSK_TX_MORE_DATA)) &&
+				   skb_queue_empty(&csk->txq)));
+	req->plen = htonl(len);
+}
+
+static int chtls_wr_size(struct chtls_sock *csk, const struct sk_buff *skb,
+			 bool size)
+{
+	int wr_size;
+
+	wr_size = TLS_WR_CPL_LEN;
+	wr_size += KEY_ON_MEM_SZ;
+	wr_size += ivs_size(csk->sk, skb);
+
+	if (size)
+		return wr_size;
+
+	/* frags counted for IV dsgl */
+	if (!skb_ulp_tls_iv_imm(skb))
+		skb_shinfo(skb)->nr_frags++;
+
+	return wr_size;
+}
+
+static bool is_ofld_imm(struct chtls_sock *csk, const struct sk_buff *skb)
+{
+	int length = skb->len;
+
+	if (skb->peeked || skb->len > MAX_IMM_ULPTX_WR_LEN)
+		return false;
+
+	if (likely(ULP_SKB_CB(skb)->flags & ULPCB_FLAG_NEED_HDR)) {
+		/* Check TLS header len for Immediate */
+		if (csk->ulp_mode == ULP_MODE_TLS &&
+		    skb_ulp_tls_inline(skb))
+			length += chtls_wr_size(csk, skb, true);
+		else
+			length += sizeof(struct fw_ofld_tx_data_wr);
+
+		return length <= MAX_IMM_OFLD_TX_DATA_WR_LEN;
+	}
+	return true;
+}
+
+static unsigned int calc_tx_flits(const struct sk_buff *skb,
+				  unsigned int immdlen)
+{
+	unsigned int flits, cnt;
+
+	flits = immdlen / 8;   /* headers */
+	cnt = skb_shinfo(skb)->nr_frags;
+	if (skb_tail_pointer(skb) != skb_transport_header(skb))
+		cnt++;
+	return flits + sgl_len(cnt);
+}
+
+static void arp_failure_discard(void *handle, struct sk_buff *skb)
+{
+	kfree_skb(skb);
+}
+
+int chtls_push_frames(struct chtls_sock *csk, int comp)
+{
+	struct chtls_hws *hws = &csk->tlshws;
+	struct tcp_sock *tp;
+	struct sk_buff *skb;
+	int total_size = 0;
+	struct sock *sk;
+	int wr_size;
+
+	wr_size = sizeof(struct fw_ofld_tx_data_wr);
+	sk = csk->sk;
+	tp = tcp_sk(sk);
+
+	if (unlikely(sk_in_state(sk, TCPF_SYN_SENT | TCPF_CLOSE)))
+		return 0;
+
+	if (unlikely(csk_flag(sk, CSK_ABORT_SHUTDOWN)))
+		return 0;
+
+	while (csk->wr_credits && (skb = skb_peek(&csk->txq)) &&
+	       (!(ULP_SKB_CB(skb)->flags & ULPCB_FLAG_HOLD) ||
+		skb_queue_len(&csk->txq) > 1)) {
+		unsigned int credit_len = skb->len;
+		unsigned int credits_needed;
+		unsigned int completion = 0;
+		int tls_len = skb->len;/* TLS data len before IV/key */
+		unsigned int immdlen;
+		int len = skb->len;    /* length [ulp bytes] inserted by hw */
+		int flowclen16 = 0;
+		int tls_tx_imm = 0;
+
+		immdlen = skb->len;
+		if (!is_ofld_imm(csk, skb)) {
+			immdlen = skb_transport_offset(skb);
+			if (skb_ulp_tls_inline(skb))
+				wr_size = chtls_wr_size(csk, skb, false);
+			credit_len = 8 * calc_tx_flits(skb, immdlen);
+		} else {
+			if (skb_ulp_tls_inline(skb)) {
+				wr_size = chtls_wr_size(csk, skb, false);
+				tls_tx_imm = 1;
+			}
+		}
+		if (likely(ULP_SKB_CB(skb)->flags & ULPCB_FLAG_NEED_HDR))
+			credit_len += wr_size;
+		credits_needed = DIV_ROUND_UP(credit_len, 16);
+		if (!csk_flag_nochk(csk, CSK_TX_DATA_SENT)) {
+			flowclen16 = send_tx_flowc_wr(sk, 1, tp->snd_nxt,
+						      tp->rcv_nxt);
+			if (flowclen16 <= 0)
+				break;
+			csk->wr_credits -= flowclen16;
+			csk->wr_unacked += flowclen16;
+			csk->wr_nondata += flowclen16;
+			csk_set_flag(csk, CSK_TX_DATA_SENT);
+		}
+
+		if (csk->wr_credits < credits_needed) {
+			if (skb_ulp_tls_inline(skb) &&
+			    !skb_ulp_tls_iv_imm(skb))
+				skb_shinfo(skb)->nr_frags--;
+			break;
+		}
+
+		__skb_unlink(skb, &csk->txq);
+		skb_set_queue_mapping(skb, (csk->txq_idx << 1) |
+				      CPL_PRIORITY_DATA);
+		if (hws->ofld)
+			hws->txqid = (skb->queue_mapping >> 1);
+		skb->csum = (__force __wsum)(credits_needed + csk->wr_nondata);
+		csk->wr_credits -= credits_needed;
+		csk->wr_unacked += credits_needed;
+		csk->wr_nondata = 0;
+		enqueue_wr(csk, skb);
+
+		if (likely(ULP_SKB_CB(skb)->flags & ULPCB_FLAG_NEED_HDR)) {
+			if ((comp && csk->wr_unacked == credits_needed) ||
+			    (ULP_SKB_CB(skb)->flags & ULPCB_FLAG_COMPL) ||
+			    csk->wr_unacked >= csk->wr_max_credits / 2) {
+				completion = 1;
+				csk->wr_unacked = 0;
+			}
+			if (skb_ulp_tls_inline(skb))
+				make_tlstx_data_wr(sk, skb, tls_tx_imm,
+						   tls_len, credits_needed);
+			else
+				make_tx_data_wr(sk, skb, immdlen, len,
+						credits_needed, completion);
+			tp->snd_nxt += len;
+			tp->lsndtime = tcp_jiffies32;
+			if (completion)
+				ULP_SKB_CB(skb)->flags &= ~ULPCB_FLAG_NEED_HDR;
+		} else {
+			struct cpl_close_con_req *req = cplhdr(skb);
+			unsigned int cmd  = CPL_OPCODE_G(ntohl
+					     (OPCODE_TID(req)));
+
+			if (cmd == CPL_CLOSE_CON_REQ)
+				csk_set_flag(csk,
+					     CSK_CLOSE_CON_REQUESTED);
+
+			if ((ULP_SKB_CB(skb)->flags & ULPCB_FLAG_COMPL) &&
+			    (csk->wr_unacked >= csk->wr_max_credits / 2)) {
+				req->wr.wr_hi |= htonl(FW_WR_COMPL_F);
+				csk->wr_unacked = 0;
+			}
+		}
+		total_size += skb->truesize;
+		if (ULP_SKB_CB(skb)->flags & ULPCB_FLAG_BARRIER)
+			csk_set_flag(csk, CSK_TX_WAIT_IDLE);
+		t4_set_arp_err_handler(skb, NULL, arp_failure_discard);
+		cxgb4_l2t_send(csk->egress_dev, skb, csk->l2t_entry);
+	}
+	sk->sk_wmem_queued -= total_size;
+	return total_size;
+}
+
+static void mark_urg(struct tcp_sock *tp, int flags,
+		     struct sk_buff *skb)
+{
+	if (unlikely(flags & MSG_OOB)) {
+		tp->snd_up = tp->write_seq;
+		ULP_SKB_CB(skb)->flags = ULPCB_FLAG_URG |
+					 ULPCB_FLAG_BARRIER |
+					 ULPCB_FLAG_NO_APPEND |
+					 ULPCB_FLAG_NEED_HDR;
+	}
+}
+
+/*
+ * Returns true if a connection should send more data to TCP engine
+ */
+static bool should_push(struct sock *sk)
+{
+	struct chtls_sock *csk = rcu_dereference_sk_user_data(sk);
+	struct chtls_dev *cdev = csk->cdev;
+	struct tcp_sock *tp = tcp_sk(sk);
+
+	/*
+	 * If we've released our offload resources there's nothing to do ...
+	 */
+	if (!cdev)
+		return false;
+
+	/*
+	 * If there aren't any work requests in flight, or there isn't enough
+	 * data in flight, or Nagle is off then send the current TX_DATA
+	 * otherwise hold it and wait to accumulate more data.
+	 */
+	return csk->wr_credits == csk->wr_max_credits ||
+		(tp->nonagle & TCP_NAGLE_OFF);
+}
+
+/*
+ * Returns true if a TCP socket is corked.
+ */
+static bool corked(const struct tcp_sock *tp, int flags)
+{
+	return (flags & MSG_MORE) || (tp->nonagle & TCP_NAGLE_CORK);
+}
+
+/*
+ * Returns true if a send should try to push new data.
+ */
+static bool send_should_push(struct sock *sk, int flags)
+{
+	return should_push(sk) && !corked(tcp_sk(sk), flags);
+}
+
+void chtls_tcp_push(struct sock *sk, int flags)
+{
+	struct chtls_sock *csk = rcu_dereference_sk_user_data(sk);
+	int qlen = skb_queue_len(&csk->txq);
+
+	if (likely(qlen)) {
+		struct sk_buff *skb = skb_peek_tail(&csk->txq);
+		struct tcp_sock *tp = tcp_sk(sk);
+
+		mark_urg(tp, flags, skb);
+
+		if (!(ULP_SKB_CB(skb)->flags & ULPCB_FLAG_NO_APPEND) &&
+		    corked(tp, flags)) {
+			ULP_SKB_CB(skb)->flags |= ULPCB_FLAG_HOLD;
+			return;
+		}
+
+		ULP_SKB_CB(skb)->flags &= ~ULPCB_FLAG_HOLD;
+		if (qlen == 1 &&
+		    ((ULP_SKB_CB(skb)->flags & ULPCB_FLAG_NO_APPEND) ||
+		     should_push(sk)))
+			chtls_push_frames(csk, 1);
+	}
+}
+
+/*
+ * Calculate the size for a new send sk_buff.  It's maximum size so we can
+ * pack lots of data into it, unless we plan to send it immediately, in which
+ * case we size it more tightly.
+ *
+ * Note: we don't bother compensating for MSS < PAGE_SIZE because it doesn't
+ * arise in normal cases and when it does we are just wasting memory.
+ */
+static int select_size(struct sock *sk, int io_len, int flags, int len)
+{
+	const int pgbreak = SKB_MAX_HEAD(len);
+
+	/*
+	 * If the data wouldn't fit in the main body anyway, put only the
+	 * header in the main body so it can use immediate data and place all
+	 * the payload in page fragments.
+	 */
+	if (io_len > pgbreak)
+		return 0;
+
+	/*
+	 * If we will be accumulating payload get a large main body.
+	 */
+	if (!send_should_push(sk, flags))
+		return pgbreak;
+
+	return io_len;
+}
+
+void skb_entail(struct sock *sk, struct sk_buff *skb, int flags)
+{
+	struct chtls_sock *csk = rcu_dereference_sk_user_data(sk);
+	struct tcp_sock *tp = tcp_sk(sk);
+
+	ULP_SKB_CB(skb)->seq = tp->write_seq;
+	ULP_SKB_CB(skb)->flags = flags;
+	__skb_queue_tail(&csk->txq, skb);
+	sk->sk_wmem_queued += skb->truesize;
+
+	if (TCP_PAGE(sk) && TCP_OFF(sk)) {
+		put_page(TCP_PAGE(sk));
+		TCP_PAGE(sk) = NULL;
+		TCP_OFF(sk) = 0;
+	}
+}
+
+static struct sk_buff *get_tx_skb(struct sock *sk, int size)
+{
+	struct sk_buff *skb;
+
+	skb = alloc_skb(size + TX_HEADER_LEN, sk->sk_allocation);
+	if (likely(skb)) {
+		skb_reserve(skb, TX_HEADER_LEN);
+		skb_entail(sk, skb, ULPCB_FLAG_NEED_HDR);
+		skb_reset_transport_header(skb);
+	}
+	return skb;
+}
+
+static struct sk_buff *get_record_skb(struct sock *sk, int size, bool zcopy)
+{
+	struct chtls_sock *csk = rcu_dereference_sk_user_data(sk);
+	struct sk_buff *skb;
+
+	skb = alloc_skb(((zcopy ? 0 : size) + TX_TLSHDR_LEN +
+			KEY_ON_MEM_SZ + max_ivs_size(sk, size)),
+			sk->sk_allocation);
+	if (likely(skb)) {
+		skb_reserve(skb, (TX_TLSHDR_LEN +
+			    KEY_ON_MEM_SZ + max_ivs_size(sk, size)));
+		skb_entail(sk, skb, ULPCB_FLAG_NEED_HDR);
+		skb_reset_transport_header(skb);
+		ULP_SKB_CB(skb)->ulp.tls.ofld = 1;
+		ULP_SKB_CB(skb)->ulp.tls.type = csk->tlshws.type;
+	}
+	return skb;
+}
+
+static void tx_skb_finalize(struct sk_buff *skb)
+{
+	struct ulp_skb_cb *cb = ULP_SKB_CB(skb);
+
+	if (!(cb->flags & ULPCB_FLAG_NO_HDR))
+		cb->flags = ULPCB_FLAG_NEED_HDR;
+	cb->flags |= ULPCB_FLAG_NO_APPEND;
+}
+
+static void push_frames_if_head(struct sock *sk)
+{
+	struct chtls_sock *csk = rcu_dereference_sk_user_data(sk);
+
+	if (skb_queue_len(&csk->txq) == 1)
+		chtls_push_frames(csk, 1);
+}
+
+static int chtls_skb_copy_to_page_nocache(struct sock *sk,
+					  struct iov_iter *from,
+					  struct sk_buff *skb,
+					  struct page *page,
+					  int off, int copy)
+{
+	int err;
+
+	err = skb_do_copy_data_nocache(sk, skb, from, page_address(page) +
+				       off, copy, skb->len);
+	if (err)
+		return err;
+
+	skb->len             += copy;
+	skb->data_len        += copy;
+	skb->truesize        += copy;
+	sk->sk_wmem_queued   += copy;
+	return 0;
+}
+
+static bool csk_mem_free(struct chtls_dev *cdev, struct sock *sk)
+{
+	return (cdev->max_host_sndbuf - sk->sk_wmem_queued > 0);
+}
+
+static int csk_wait_memory(struct chtls_dev *cdev,
+			   struct sock *sk, long *timeo_p)
+{
+	DEFINE_WAIT_FUNC(wait, woken_wake_function);
+	int ret, err = 0;
+	long current_timeo;
+	long vm_wait = 0;
+	bool noblock;
+
+	current_timeo = *timeo_p;
+	noblock = (*timeo_p ? false : true);
+	if (csk_mem_free(cdev, sk)) {
+		current_timeo = get_random_u32_below(HZ / 5) + 2;
+		vm_wait = get_random_u32_below(HZ / 5) + 2;
+	}
+
+	add_wait_queue(sk_sleep(sk), &wait);
+	while (1) {
+		sk_set_bit(SOCKWQ_ASYNC_NOSPACE, sk);
+
+		if (sk->sk_err || (sk->sk_shutdown & SEND_SHUTDOWN))
+			goto do_error;
+		if (!*timeo_p) {
+			if (noblock)
+				set_bit(SOCK_NOSPACE, &sk->sk_socket->flags);
+			goto do_nonblock;
+		}
+		if (signal_pending(current))
+			goto do_interrupted;
+		sk_clear_bit(SOCKWQ_ASYNC_NOSPACE, sk);
+		if (csk_mem_free(cdev, sk) && !vm_wait)
+			break;
+
+		set_bit(SOCK_NOSPACE, &sk->sk_socket->flags);
+		sk->sk_write_pending++;
+		ret = sk_wait_event(sk, &current_timeo, sk->sk_err ||
+				    (sk->sk_shutdown & SEND_SHUTDOWN) ||
+				    (csk_mem_free(cdev, sk) && !vm_wait),
+				    &wait);
+		sk->sk_write_pending--;
+		if (ret < 0)
+			goto do_error;
+
+		if (vm_wait) {
+			vm_wait -= current_timeo;
+			current_timeo = *timeo_p;
+			if (current_timeo != MAX_SCHEDULE_TIMEOUT) {
+				current_timeo -= vm_wait;
+				if (current_timeo < 0)
+					current_timeo = 0;
+			}
+			vm_wait = 0;
+		}
+		*timeo_p = current_timeo;
+	}
+do_rm_wq:
+	remove_wait_queue(sk_sleep(sk), &wait);
+	return err;
+do_error:
+	err = -EPIPE;
+	goto do_rm_wq;
+do_nonblock:
+	err = -EAGAIN;
+	goto do_rm_wq;
+do_interrupted:
+	err = sock_intr_errno(*timeo_p);
+	goto do_rm_wq;
+}
+
+static int chtls_proccess_cmsg(struct sock *sk, struct msghdr *msg,
+			       unsigned char *record_type)
+{
+	struct cmsghdr *cmsg;
+	int rc = -EINVAL;
+
+	for_each_cmsghdr(cmsg, msg) {
+		if (!CMSG_OK(msg, cmsg))
+			return -EINVAL;
+		if (cmsg->cmsg_level != SOL_TLS)
+			continue;
+
+		switch (cmsg->cmsg_type) {
+		case TLS_SET_RECORD_TYPE:
+			if (cmsg->cmsg_len < CMSG_LEN(sizeof(*record_type)))
+				return -EINVAL;
+
+			if (msg->msg_flags & MSG_MORE)
+				return -EINVAL;
+
+			*record_type = *(unsigned char *)CMSG_DATA(cmsg);
+			rc = 0;
+			break;
+		default:
+			return -EINVAL;
+		}
+	}
+
+	return rc;
+}
+
+int chtls_sendmsg(struct sock *sk, struct msghdr *msg, size_t size)
+{
+	struct chtls_sock *csk = rcu_dereference_sk_user_data(sk);
+	struct chtls_dev *cdev = csk->cdev;
+	struct tcp_sock *tp = tcp_sk(sk);
+	struct sk_buff *skb;
+	int mss, flags, err;
+	int recordsz = 0;
+	int copied = 0;
+	long timeo;
+
+	lock_sock(sk);
+	flags = msg->msg_flags;
+	timeo = sock_sndtimeo(sk, flags & MSG_DONTWAIT);
+
+	if (!sk_in_state(sk, TCPF_ESTABLISHED | TCPF_CLOSE_WAIT)) {
+		err = sk_stream_wait_connect(sk, &timeo);
+		if (err)
+			goto out_err;
+	}
+
+	sk_clear_bit(SOCKWQ_ASYNC_NOSPACE, sk);
+	err = -EPIPE;
+	if (sk->sk_err || (sk->sk_shutdown & SEND_SHUTDOWN))
+		goto out_err;
+
+	mss = csk->mss;
+	csk_set_flag(csk, CSK_TX_MORE_DATA);
+
+	while (msg_data_left(msg)) {
+		int copy = 0;
+
+		skb = skb_peek_tail(&csk->txq);
+		if (skb) {
+			copy = mss - skb->len;
+			skb->ip_summed = CHECKSUM_UNNECESSARY;
+		}
+		if (!csk_mem_free(cdev, sk))
+			goto wait_for_sndbuf;
+
+		if (is_tls_tx(csk) && !csk->tlshws.txleft) {
+			unsigned char record_type = TLS_RECORD_TYPE_DATA;
+
+			if (unlikely(msg->msg_controllen)) {
+				err = chtls_proccess_cmsg(sk, msg,
+							  &record_type);
+				if (err)
+					goto out_err;
+
+				/* Avoid appending tls handshake, alert to tls data */
+				if (skb)
+					tx_skb_finalize(skb);
+			}
+
+			recordsz = size;
+			csk->tlshws.txleft = recordsz;
+			csk->tlshws.type = record_type;
+		}
+
+		if (!skb || (ULP_SKB_CB(skb)->flags & ULPCB_FLAG_NO_APPEND) ||
+		    copy <= 0) {
+new_buf:
+			if (skb) {
+				tx_skb_finalize(skb);
+				push_frames_if_head(sk);
+			}
+
+			if (is_tls_tx(csk)) {
+				skb = get_record_skb(sk,
+						     select_size(sk,
+								 recordsz,
+								 flags,
+								 TX_TLSHDR_LEN),
+								 false);
+			} else {
+				skb = get_tx_skb(sk,
+						 select_size(sk, size, flags,
+							     TX_HEADER_LEN));
+			}
+			if (unlikely(!skb))
+				goto wait_for_memory;
+
+			skb->ip_summed = CHECKSUM_UNNECESSARY;
+			copy = mss;
+		}
+		if (copy > size)
+			copy = size;
+
+		if (msg->msg_flags & MSG_SPLICE_PAGES) {
+			err = skb_splice_from_iter(skb, &msg->msg_iter, copy);
+			if (err < 0) {
+				if (err == -EMSGSIZE)
+					goto new_buf;
+				goto do_fault;
+			}
+			copy = err;
+			sk_wmem_queued_add(sk, copy);
+		} else if (skb_tailroom(skb) > 0) {
+			copy = min(copy, skb_tailroom(skb));
+			if (is_tls_tx(csk))
+				copy = min_t(int, copy, csk->tlshws.txleft);
+			err = skb_add_data_nocache(sk, skb,
+						   &msg->msg_iter, copy);
+			if (err)
+				goto do_fault;
+		} else {
+			int i = skb_shinfo(skb)->nr_frags;
+			struct page *page = TCP_PAGE(sk);
+			int pg_size = PAGE_SIZE;
+			int off = TCP_OFF(sk);
+			bool merge;
+
+			if (page)
+				pg_size = page_size(page);
+			if (off < pg_size &&
+			    skb_can_coalesce(skb, i, page, off)) {
+				merge = true;
+				goto copy;
+			}
+			merge = false;
+			if (i == (is_tls_tx(csk) ? (MAX_SKB_FRAGS - 1) :
+			    MAX_SKB_FRAGS))
+				goto new_buf;
+
+			if (page && off == pg_size) {
+				put_page(page);
+				TCP_PAGE(sk) = page = NULL;
+				pg_size = PAGE_SIZE;
+			}
+
+			if (!page) {
+				gfp_t gfp = sk->sk_allocation;
+				int order = cdev->send_page_order;
+
+				if (order) {
+					page = alloc_pages(gfp | __GFP_COMP |
+							   __GFP_NOWARN |
+							   __GFP_NORETRY,
+							   order);
+					if (page)
+						pg_size <<= order;
+				}
+				if (!page) {
+					page = alloc_page(gfp);
+					pg_size = PAGE_SIZE;
+				}
+				if (!page)
+					goto wait_for_memory;
+				off = 0;
+			}
+copy:
+			if (copy > pg_size - off)
+				copy = pg_size - off;
+			if (is_tls_tx(csk))
+				copy = min_t(int, copy, csk->tlshws.txleft);
+
+			err = chtls_skb_copy_to_page_nocache(sk, &msg->msg_iter,
+							     skb, page,
+							     off, copy);
+			if (unlikely(err)) {
+				if (!TCP_PAGE(sk)) {
+					TCP_PAGE(sk) = page;
+					TCP_OFF(sk) = 0;
+				}
+				goto do_fault;
+			}
+			/* Update the skb. */
+			if (merge) {
+				skb_frag_size_add(
+						&skb_shinfo(skb)->frags[i - 1],
+						copy);
+			} else {
+				skb_fill_page_desc(skb, i, page, off, copy);
+				if (off + copy < pg_size) {
+					/* space left keep page */
+					get_page(page);
+					TCP_PAGE(sk) = page;
+				} else {
+					TCP_PAGE(sk) = NULL;
+				}
+			}
+			TCP_OFF(sk) = off + copy;
+		}
+		if (unlikely(skb->len == mss))
+			tx_skb_finalize(skb);
+		tp->write_seq += copy;
+		copied += copy;
+		size -= copy;
+
+		if (is_tls_tx(csk))
+			csk->tlshws.txleft -= copy;
+
+		if (corked(tp, flags) &&
+		    (sk_stream_wspace(sk) < sk_stream_min_wspace(sk)))
+			ULP_SKB_CB(skb)->flags |= ULPCB_FLAG_NO_APPEND;
+
+		if (size == 0)
+			goto out;
+
+		if (ULP_SKB_CB(skb)->flags & ULPCB_FLAG_NO_APPEND)
+			push_frames_if_head(sk);
+		continue;
+wait_for_sndbuf:
+		set_bit(SOCK_NOSPACE, &sk->sk_socket->flags);
+wait_for_memory:
+		err = csk_wait_memory(cdev, sk, &timeo);
+		if (err)
+			goto do_error;
+	}
+out:
+	csk_reset_flag(csk, CSK_TX_MORE_DATA);
+	if (copied)
+		chtls_tcp_push(sk, flags);
+done:
+	release_sock(sk);
+	return copied;
+do_fault:
+	if (!skb->len) {
+		__skb_unlink(skb, &csk->txq);
+		sk->sk_wmem_queued -= skb->truesize;
+		__kfree_skb(skb);
+	}
+do_error:
+	if (copied)
+		goto out;
+out_err:
+	if (csk_conn_inline(csk))
+		csk_reset_flag(csk, CSK_TX_MORE_DATA);
+	copied = sk_stream_error(sk, flags, err);
+	goto done;
+}
+
+void chtls_splice_eof(struct socket *sock)
+{
+	struct sock *sk = sock->sk;
+
+	lock_sock(sk);
+	chtls_tcp_push(sk, 0);
+	release_sock(sk);
+}
+
+static void chtls_select_window(struct sock *sk)
+{
+	struct chtls_sock *csk = rcu_dereference_sk_user_data(sk);
+	struct tcp_sock *tp = tcp_sk(sk);
+	unsigned int wnd = tp->rcv_wnd;
+
+	wnd = max_t(unsigned int, wnd, tcp_full_space(sk));
+	wnd = max_t(unsigned int, MIN_RCV_WND, wnd);
+
+	if (wnd > MAX_RCV_WND)
+		wnd = MAX_RCV_WND;
+
+/*
+ * Check if we need to grow the receive window in response to an increase in
+ * the socket's receive buffer size.  Some applications increase the buffer
+ * size dynamically and rely on the window to grow accordingly.
+ */
+
+	if (wnd > tp->rcv_wnd) {
+		tp->rcv_wup -= wnd - tp->rcv_wnd;
+		tp->rcv_wnd = wnd;
+		/* Mark the receive window as updated */
+		csk_reset_flag(csk, CSK_UPDATE_RCV_WND);
+	}
+}
+
+/*
+ * Send RX credits through an RX_DATA_ACK CPL message.  We are permitted
+ * to return without sending the message in case we cannot allocate
+ * an sk_buff.  Returns the number of credits sent.
+ */
+static u32 send_rx_credits(struct chtls_sock *csk, u32 credits)
+{
+	struct cpl_rx_data_ack *req;
+	struct sk_buff *skb;
+
+	skb = alloc_skb(sizeof(*req), GFP_ATOMIC);
+	if (!skb)
+		return 0;
+	__skb_put(skb, sizeof(*req));
+	req = (struct cpl_rx_data_ack *)skb->head;
+
+	set_wr_txq(skb, CPL_PRIORITY_ACK, csk->port_id);
+	INIT_TP_WR(req, csk->tid);
+	OPCODE_TID(req) = cpu_to_be32(MK_OPCODE_TID(CPL_RX_DATA_ACK,
+						    csk->tid));
+	req->credit_dack = cpu_to_be32(RX_CREDITS_V(credits) |
+				       RX_FORCE_ACK_F);
+	cxgb4_ofld_send(csk->cdev->ports[csk->port_id], skb);
+	return credits;
+}
+
+#define CREDIT_RETURN_STATE (TCPF_ESTABLISHED | \
+			     TCPF_FIN_WAIT1 | \
+			     TCPF_FIN_WAIT2)
+
+/*
+ * Called after some received data has been read.  It returns RX credits
+ * to the HW for the amount of data processed.
+ */
+static void chtls_cleanup_rbuf(struct sock *sk, int copied)
+{
+	struct chtls_sock *csk = rcu_dereference_sk_user_data(sk);
+	struct tcp_sock *tp;
+	int must_send;
+	u32 credits;
+	u32 thres;
+
+	thres = 15 * 1024;
+
+	if (!sk_in_state(sk, CREDIT_RETURN_STATE))
+		return;
+
+	chtls_select_window(sk);
+	tp = tcp_sk(sk);
+	credits = tp->copied_seq - tp->rcv_wup;
+	if (unlikely(!credits))
+		return;
+
+/*
+ * For coalescing to work effectively ensure the receive window has
+ * at least 16KB left.
+ */
+	must_send = credits + 16384 >= tp->rcv_wnd;
+
+	if (must_send || credits >= thres)
+		tp->rcv_wup += send_rx_credits(csk, credits);
+}
+
+static int chtls_pt_recvmsg(struct sock *sk, struct msghdr *msg, size_t len,
+			    int flags, int *addr_len)
+{
+	struct chtls_sock *csk = rcu_dereference_sk_user_data(sk);
+	struct chtls_hws *hws = &csk->tlshws;
+	struct net_device *dev = csk->egress_dev;
+	struct adapter *adap = netdev2adap(dev);
+	struct tcp_sock *tp = tcp_sk(sk);
+	unsigned long avail;
+	int buffers_freed;
+	int copied = 0;
+	int target;
+	long timeo;
+	int ret;
+
+	buffers_freed = 0;
+
+	timeo = sock_rcvtimeo(sk, flags & MSG_DONTWAIT);
+	target = sock_rcvlowat(sk, flags & MSG_WAITALL, len);
+
+	if (unlikely(csk_flag(sk, CSK_UPDATE_RCV_WND)))
+		chtls_cleanup_rbuf(sk, copied);
+
+	do {
+		struct sk_buff *skb;
+		u32 offset = 0;
+
+		if (unlikely(tp->urg_data &&
+			     tp->urg_seq == tp->copied_seq)) {
+			if (copied)
+				break;
+			if (signal_pending(current)) {
+				copied = timeo ? sock_intr_errno(timeo) :
+					-EAGAIN;
+				break;
+			}
+		}
+		skb = skb_peek(&sk->sk_receive_queue);
+		if (skb)
+			goto found_ok_skb;
+		if (csk->wr_credits &&
+		    skb_queue_len(&csk->txq) &&
+		    chtls_push_frames(csk, csk->wr_credits ==
+				      csk->wr_max_credits))
+			sk->sk_write_space(sk);
+
+		if (copied >= target && !READ_ONCE(sk->sk_backlog.tail))
+			break;
+
+		if (copied) {
+			if (sk->sk_err || sk->sk_state == TCP_CLOSE ||
+			    (sk->sk_shutdown & RCV_SHUTDOWN) ||
+			    signal_pending(current))
+				break;
+
+			if (!timeo)
+				break;
+		} else {
+			if (sock_flag(sk, SOCK_DONE))
+				break;
+			if (sk->sk_err) {
+				copied = sock_error(sk);
+				break;
+			}
+			if (sk->sk_shutdown & RCV_SHUTDOWN)
+				break;
+			if (sk->sk_state == TCP_CLOSE) {
+				copied = -ENOTCONN;
+				break;
+			}
+			if (!timeo) {
+				copied = -EAGAIN;
+				break;
+			}
+			if (signal_pending(current)) {
+				copied = sock_intr_errno(timeo);
+				break;
+			}
+		}
+		if (READ_ONCE(sk->sk_backlog.tail)) {
+			release_sock(sk);
+			lock_sock(sk);
+			chtls_cleanup_rbuf(sk, copied);
+			continue;
+		}
+
+		if (copied >= target)
+			break;
+		chtls_cleanup_rbuf(sk, copied);
+		ret = sk_wait_data(sk, &timeo, NULL);
+		if (ret < 0) {
+			copied = copied ? : ret;
+			goto unlock;
+		}
+		continue;
+found_ok_skb:
+		if (!skb->len) {
+			skb_dstref_steal(skb);
+			__skb_unlink(skb, &sk->sk_receive_queue);
+			kfree_skb(skb);
+
+			if (!copied && !timeo) {
+				copied = -EAGAIN;
+				break;
+			}
+
+			if (copied < target) {
+				release_sock(sk);
+				lock_sock(sk);
+				continue;
+			}
+			break;
+		}
+		offset = hws->copied_seq;
+		avail = skb->len - offset;
+		if (len < avail)
+			avail = len;
+
+		if (unlikely(tp->urg_data)) {
+			u32 urg_offset = tp->urg_seq - tp->copied_seq;
+
+			if (urg_offset < avail) {
+				if (urg_offset) {
+					avail = urg_offset;
+				} else if (!sock_flag(sk, SOCK_URGINLINE)) {
+					/* First byte is urgent, skip */
+					tp->copied_seq++;
+					offset++;
+					avail--;
+					if (!avail)
+						goto skip_copy;
+				}
+			}
+		}
+		/* Set record type if not already done. For a non-data record,
+		 * do not proceed if record type could not be copied.
+		 */
+		if (ULP_SKB_CB(skb)->flags & ULPCB_FLAG_TLS_HDR) {
+			struct tls_hdr *thdr = (struct tls_hdr *)skb->data;
+			int cerr = 0;
+
+			cerr = put_cmsg(msg, SOL_TLS, TLS_GET_RECORD_TYPE,
+					sizeof(thdr->type), &thdr->type);
+
+			if (cerr && thdr->type != TLS_RECORD_TYPE_DATA) {
+				copied = -EIO;
+				break;
+			}
+			/*  don't send tls header, skip copy */
+			goto skip_copy;
+		}
+
+		if (skb_copy_datagram_msg(skb, offset, msg, avail)) {
+			if (!copied) {
+				copied = -EFAULT;
+				break;
+			}
+		}
+
+		copied += avail;
+		len -= avail;
+		hws->copied_seq += avail;
+skip_copy:
+		if (tp->urg_data && after(tp->copied_seq, tp->urg_seq))
+			tp->urg_data = 0;
+
+		if ((avail + offset) >= skb->len) {
+			struct sk_buff *next_skb;
+			if (ULP_SKB_CB(skb)->flags & ULPCB_FLAG_TLS_HDR) {
+				tp->copied_seq += skb->len;
+				hws->rcvpld = skb->hdr_len;
+			} else {
+				atomic_inc(&adap->chcr_stats.tls_pdu_rx);
+				tp->copied_seq += hws->rcvpld;
+			}
+			chtls_free_skb(sk, skb);
+			buffers_freed++;
+			hws->copied_seq = 0;
+			next_skb = skb_peek(&sk->sk_receive_queue);
+			if (copied >= target && !next_skb)
+				break;
+			if (ULP_SKB_CB(next_skb)->flags & ULPCB_FLAG_TLS_HDR)
+				break;
+		}
+	} while (len > 0);
+
+	if (buffers_freed)
+		chtls_cleanup_rbuf(sk, copied);
+
+unlock:
+	release_sock(sk);
+	return copied;
+}
+
+/*
+ * Peek at data in a socket's receive buffer.
+ */
+static int peekmsg(struct sock *sk, struct msghdr *msg,
+		   size_t len, int flags)
+{
+	struct tcp_sock *tp = tcp_sk(sk);
+	u32 peek_seq, offset;
+	struct sk_buff *skb;
+	int copied = 0;
+	size_t avail;          /* amount of available data in current skb */
+	long timeo;
+	int ret;
+
+	lock_sock(sk);
+	timeo = sock_rcvtimeo(sk, flags & MSG_DONTWAIT);
+	peek_seq = tp->copied_seq;
+
+	do {
+		if (unlikely(tp->urg_data && tp->urg_seq == peek_seq)) {
+			if (copied)
+				break;
+			if (signal_pending(current)) {
+				copied = timeo ? sock_intr_errno(timeo) :
+				-EAGAIN;
+				break;
+			}
+		}
+
+		skb_queue_walk(&sk->sk_receive_queue, skb) {
+			offset = peek_seq - ULP_SKB_CB(skb)->seq;
+			if (offset < skb->len)
+				goto found_ok_skb;
+		}
+
+		/* empty receive queue */
+		if (copied)
+			break;
+		if (sock_flag(sk, SOCK_DONE))
+			break;
+		if (sk->sk_err) {
+			copied = sock_error(sk);
+			break;
+		}
+		if (sk->sk_shutdown & RCV_SHUTDOWN)
+			break;
+		if (sk->sk_state == TCP_CLOSE) {
+			copied = -ENOTCONN;
+			break;
+		}
+		if (!timeo) {
+			copied = -EAGAIN;
+			break;
+		}
+		if (signal_pending(current)) {
+			copied = sock_intr_errno(timeo);
+			break;
+		}
+
+		if (READ_ONCE(sk->sk_backlog.tail)) {
+			/* Do not sleep, just process backlog. */
+			release_sock(sk);
+			lock_sock(sk);
+		} else {
+			ret = sk_wait_data(sk, &timeo, NULL);
+			if (ret < 0) {
+				/* here 'copied' is 0 due to previous checks */
+				copied = ret;
+				break;
+			}
+		}
+
+		if (unlikely(peek_seq != tp->copied_seq)) {
+			if (net_ratelimit())
+				pr_info("TCP(%s:%d), race in MSG_PEEK.\n",
+					current->comm, current->pid);
+			peek_seq = tp->copied_seq;
+		}
+		continue;
+
+found_ok_skb:
+		avail = skb->len - offset;
+		if (len < avail)
+			avail = len;
+		/*
+		 * Do we have urgent data here?  We need to skip over the
+		 * urgent byte.
+		 */
+		if (unlikely(tp->urg_data)) {
+			u32 urg_offset = tp->urg_seq - peek_seq;
+
+			if (urg_offset < avail) {
+				/*
+				 * The amount of data we are preparing to copy
+				 * contains urgent data.
+				 */
+				if (!urg_offset) { /* First byte is urgent */
+					if (!sock_flag(sk, SOCK_URGINLINE)) {
+						peek_seq++;
+						offset++;
+						avail--;
+					}
+					if (!avail)
+						continue;
+				} else {
+					/* stop short of the urgent data */
+					avail = urg_offset;
+				}
+			}
+		}
+
+		/*
+		 * If MSG_TRUNC is specified the data is discarded.
+		 */
+		if (likely(!(flags & MSG_TRUNC)))
+			if (skb_copy_datagram_msg(skb, offset, msg, len)) {
+				if (!copied) {
+					copied = -EFAULT;
+					break;
+				}
+			}
+		peek_seq += avail;
+		copied += avail;
+		len -= avail;
+	} while (len > 0);
+
+	release_sock(sk);
+	return copied;
+}
+
+int chtls_recvmsg(struct sock *sk, struct msghdr *msg, size_t len,
+		  int flags, int *addr_len)
+{
+	struct tcp_sock *tp = tcp_sk(sk);
+	struct chtls_sock *csk;
+	unsigned long avail;    /* amount of available data in current skb */
+	int buffers_freed;
+	int copied = 0;
+	long timeo;
+	int target;             /* Read at least this many bytes */
+	int ret;
+
+	buffers_freed = 0;
+
+	if (unlikely(flags & MSG_OOB))
+		return tcp_prot.recvmsg(sk, msg, len, flags, addr_len);
+
+	if (unlikely(flags & MSG_PEEK))
+		return peekmsg(sk, msg, len, flags);
+
+	if (sk_can_busy_loop(sk) &&
+	    skb_queue_empty_lockless(&sk->sk_receive_queue) &&
+	    sk->sk_state == TCP_ESTABLISHED)
+		sk_busy_loop(sk, flags & MSG_DONTWAIT);
+
+	lock_sock(sk);
+	csk = rcu_dereference_sk_user_data(sk);
+
+	if (is_tls_rx(csk))
+		return chtls_pt_recvmsg(sk, msg, len, flags, addr_len);
+
+	timeo = sock_rcvtimeo(sk, flags & MSG_DONTWAIT);
+	target = sock_rcvlowat(sk, flags & MSG_WAITALL, len);
+
+	if (unlikely(csk_flag(sk, CSK_UPDATE_RCV_WND)))
+		chtls_cleanup_rbuf(sk, copied);
+
+	do {
+		struct sk_buff *skb;
+		u32 offset;
+
+		if (unlikely(tp->urg_data && tp->urg_seq == tp->copied_seq)) {
+			if (copied)
+				break;
+			if (signal_pending(current)) {
+				copied = timeo ? sock_intr_errno(timeo) :
+					-EAGAIN;
+				break;
+			}
+		}
+
+		skb = skb_peek(&sk->sk_receive_queue);
+		if (skb)
+			goto found_ok_skb;
+
+		if (csk->wr_credits &&
+		    skb_queue_len(&csk->txq) &&
+		    chtls_push_frames(csk, csk->wr_credits ==
+				      csk->wr_max_credits))
+			sk->sk_write_space(sk);
+
+		if (copied >= target && !READ_ONCE(sk->sk_backlog.tail))
+			break;
+
+		if (copied) {
+			if (sk->sk_err || sk->sk_state == TCP_CLOSE ||
+			    (sk->sk_shutdown & RCV_SHUTDOWN) ||
+			    signal_pending(current))
+				break;
+		} else {
+			if (sock_flag(sk, SOCK_DONE))
+				break;
+			if (sk->sk_err) {
+				copied = sock_error(sk);
+				break;
+			}
+			if (sk->sk_shutdown & RCV_SHUTDOWN)
+				break;
+			if (sk->sk_state == TCP_CLOSE) {
+				copied = -ENOTCONN;
+				break;
+			}
+			if (!timeo) {
+				copied = -EAGAIN;
+				break;
+			}
+			if (signal_pending(current)) {
+				copied = sock_intr_errno(timeo);
+				break;
+			}
+		}
+
+		if (READ_ONCE(sk->sk_backlog.tail)) {
+			release_sock(sk);
+			lock_sock(sk);
+			chtls_cleanup_rbuf(sk, copied);
+			continue;
+		}
+
+		if (copied >= target)
+			break;
+		chtls_cleanup_rbuf(sk, copied);
+		ret = sk_wait_data(sk, &timeo, NULL);
+		if (ret < 0) {
+			copied = copied ? : ret;
+			goto unlock;
+		}
+		continue;
+
+found_ok_skb:
+		if (!skb->len) {
+			chtls_kfree_skb(sk, skb);
+			if (!copied && !timeo) {
+				copied = -EAGAIN;
+				break;
+			}
+
+			if (copied < target)
+				continue;
+
+			break;
+		}
+
+		offset = tp->copied_seq - ULP_SKB_CB(skb)->seq;
+		avail = skb->len - offset;
+		if (len < avail)
+			avail = len;
+
+		if (unlikely(tp->urg_data)) {
+			u32 urg_offset = tp->urg_seq - tp->copied_seq;
+
+			if (urg_offset < avail) {
+				if (urg_offset) {
+					avail = urg_offset;
+				} else if (!sock_flag(sk, SOCK_URGINLINE)) {
+					tp->copied_seq++;
+					offset++;
+					avail--;
+					if (!avail)
+						goto skip_copy;
+				}
+			}
+		}
+
+		if (likely(!(flags & MSG_TRUNC))) {
+			if (skb_copy_datagram_msg(skb, offset,
+						  msg, avail)) {
+				if (!copied) {
+					copied = -EFAULT;
+					break;
+				}
+			}
+		}
+
+		tp->copied_seq += avail;
+		copied += avail;
+		len -= avail;
+
+skip_copy:
+		if (tp->urg_data && after(tp->copied_seq, tp->urg_seq))
+			tp->urg_data = 0;
+
+		if (avail + offset >= skb->len) {
+			chtls_free_skb(sk, skb);
+			buffers_freed++;
+
+			if  (copied >= target &&
+			     !skb_peek(&sk->sk_receive_queue))
+				break;
+		}
+	} while (len > 0);
+
+	if (buffers_freed)
+		chtls_cleanup_rbuf(sk, copied);
+
+unlock:
+	release_sock(sk);
+	return copied;
+}
diff --git a/drivers/net/ethernet/chelsio/inline_crypto/chtls/chtls_main.c b/drivers/net/ethernet/chelsio/inline_crypto/chtls/chtls_main.c
new file mode 100644
index 000000000000..daa1ebaef511
--- /dev/null
+++ b/drivers/net/ethernet/chelsio/inline_crypto/chtls/chtls_main.c
@@ -0,0 +1,642 @@
+// SPDX-License-Identifier: GPL-2.0-only
+/*
+ * Copyright (c) 2018 Chelsio Communications, Inc.
+ *
+ * Written by: Atul Gupta (atul.gupta@chelsio.com)
+ */
+#include <linux/kernel.h>
+#include <linux/module.h>
+#include <linux/skbuff.h>
+#include <linux/socket.h>
+#include <linux/hash.h>
+#include <linux/in.h>
+#include <linux/net.h>
+#include <linux/ip.h>
+#include <linux/tcp.h>
+#include <net/ipv6.h>
+#include <net/transp_v6.h>
+#include <net/tcp.h>
+#include <net/tls.h>
+
+#include "chtls.h"
+#include "chtls_cm.h"
+
+#define DRV_NAME "chtls"
+
+/*
+ * chtls device management
+ * maintains a list of the chtls devices
+ */
+static LIST_HEAD(cdev_list);
+static DEFINE_MUTEX(cdev_mutex);
+
+static DEFINE_MUTEX(notify_mutex);
+static RAW_NOTIFIER_HEAD(listen_notify_list);
+static struct proto chtls_cpl_prot, chtls_cpl_protv6;
+struct request_sock_ops chtls_rsk_ops, chtls_rsk_opsv6;
+static uint send_page_order = (14 - PAGE_SHIFT < 0) ? 0 : 14 - PAGE_SHIFT;
+
+static void register_listen_notifier(struct notifier_block *nb)
+{
+	mutex_lock(&notify_mutex);
+	raw_notifier_chain_register(&listen_notify_list, nb);
+	mutex_unlock(&notify_mutex);
+}
+
+static void unregister_listen_notifier(struct notifier_block *nb)
+{
+	mutex_lock(&notify_mutex);
+	raw_notifier_chain_unregister(&listen_notify_list, nb);
+	mutex_unlock(&notify_mutex);
+}
+
+static int listen_notify_handler(struct notifier_block *this,
+				 unsigned long event, void *data)
+{
+	struct chtls_listen *clisten;
+	int ret = NOTIFY_DONE;
+
+	clisten = (struct chtls_listen *)data;
+
+	switch (event) {
+	case CHTLS_LISTEN_START:
+		ret = chtls_listen_start(clisten->cdev, clisten->sk);
+		kfree(clisten);
+		break;
+	case CHTLS_LISTEN_STOP:
+		chtls_listen_stop(clisten->cdev, clisten->sk);
+		kfree(clisten);
+		break;
+	}
+	return ret;
+}
+
+static struct notifier_block listen_notifier = {
+	.notifier_call = listen_notify_handler
+};
+
+static int listen_backlog_rcv(struct sock *sk, struct sk_buff *skb)
+{
+	if (likely(skb_transport_header(skb) != skb_network_header(skb)))
+		return tcp_v4_do_rcv(sk, skb);
+	BLOG_SKB_CB(skb)->backlog_rcv(sk, skb);
+	return 0;
+}
+
+static int chtls_start_listen(struct chtls_dev *cdev, struct sock *sk)
+{
+	struct chtls_listen *clisten;
+
+	if (sk->sk_protocol != IPPROTO_TCP)
+		return -EPROTONOSUPPORT;
+
+	if (sk->sk_family == PF_INET &&
+	    LOOPBACK(inet_sk(sk)->inet_rcv_saddr))
+		return -EADDRNOTAVAIL;
+
+	sk->sk_backlog_rcv = listen_backlog_rcv;
+	clisten = kmalloc(sizeof(*clisten), GFP_KERNEL);
+	if (!clisten)
+		return -ENOMEM;
+	clisten->cdev = cdev;
+	clisten->sk = sk;
+	mutex_lock(&notify_mutex);
+	raw_notifier_call_chain(&listen_notify_list,
+				      CHTLS_LISTEN_START, clisten);
+	mutex_unlock(&notify_mutex);
+	return 0;
+}
+
+static void chtls_stop_listen(struct chtls_dev *cdev, struct sock *sk)
+{
+	struct chtls_listen *clisten;
+
+	if (sk->sk_protocol != IPPROTO_TCP)
+		return;
+
+	clisten = kmalloc(sizeof(*clisten), GFP_KERNEL);
+	if (!clisten)
+		return;
+	clisten->cdev = cdev;
+	clisten->sk = sk;
+	mutex_lock(&notify_mutex);
+	raw_notifier_call_chain(&listen_notify_list,
+				CHTLS_LISTEN_STOP, clisten);
+	mutex_unlock(&notify_mutex);
+}
+
+static int chtls_inline_feature(struct tls_toe_device *dev)
+{
+	struct net_device *netdev;
+	struct chtls_dev *cdev;
+	int i;
+
+	cdev = to_chtls_dev(dev);
+
+	for (i = 0; i < cdev->lldi->nports; i++) {
+		netdev = cdev->ports[i];
+		if (netdev->features & NETIF_F_HW_TLS_RECORD)
+			return 1;
+	}
+	return 0;
+}
+
+static int chtls_create_hash(struct tls_toe_device *dev, struct sock *sk)
+{
+	struct chtls_dev *cdev = to_chtls_dev(dev);
+
+	if (sk->sk_state == TCP_LISTEN)
+		return chtls_start_listen(cdev, sk);
+	return 0;
+}
+
+static void chtls_destroy_hash(struct tls_toe_device *dev, struct sock *sk)
+{
+	struct chtls_dev *cdev = to_chtls_dev(dev);
+
+	if (sk->sk_state == TCP_LISTEN)
+		chtls_stop_listen(cdev, sk);
+}
+
+static void chtls_free_uld(struct chtls_dev *cdev)
+{
+	int i;
+
+	tls_toe_unregister_device(&cdev->tlsdev);
+	kvfree(cdev->kmap.addr);
+	idr_destroy(&cdev->hwtid_idr);
+	for (i = 0; i < (1 << RSPQ_HASH_BITS); i++)
+		kfree_skb(cdev->rspq_skb_cache[i]);
+	kfree(cdev->lldi);
+	kfree_skb(cdev->askb);
+	kfree(cdev);
+}
+
+static inline void chtls_dev_release(struct kref *kref)
+{
+	struct tls_toe_device *dev;
+	struct chtls_dev *cdev;
+	struct adapter *adap;
+
+	dev = container_of(kref, struct tls_toe_device, kref);
+	cdev = to_chtls_dev(dev);
+
+	/* Reset tls rx/tx stats */
+	adap = pci_get_drvdata(cdev->pdev);
+	atomic_set(&adap->chcr_stats.tls_pdu_tx, 0);
+	atomic_set(&adap->chcr_stats.tls_pdu_rx, 0);
+
+	chtls_free_uld(cdev);
+}
+
+static void chtls_register_dev(struct chtls_dev *cdev)
+{
+	struct tls_toe_device *tlsdev = &cdev->tlsdev;
+
+	strscpy(tlsdev->name, "chtls", TLS_TOE_DEVICE_NAME_MAX);
+	strlcat(tlsdev->name, cdev->lldi->ports[0]->name,
+		TLS_TOE_DEVICE_NAME_MAX);
+	tlsdev->feature = chtls_inline_feature;
+	tlsdev->hash = chtls_create_hash;
+	tlsdev->unhash = chtls_destroy_hash;
+	tlsdev->release = chtls_dev_release;
+	kref_init(&tlsdev->kref);
+	tls_toe_register_device(tlsdev);
+	cdev->cdev_state = CHTLS_CDEV_STATE_UP;
+}
+
+static void process_deferq(struct work_struct *task_param)
+{
+	struct chtls_dev *cdev = container_of(task_param,
+				struct chtls_dev, deferq_task);
+	struct sk_buff *skb;
+
+	spin_lock_bh(&cdev->deferq.lock);
+	while ((skb = __skb_dequeue(&cdev->deferq)) != NULL) {
+		spin_unlock_bh(&cdev->deferq.lock);
+		DEFERRED_SKB_CB(skb)->handler(cdev, skb);
+		spin_lock_bh(&cdev->deferq.lock);
+	}
+	spin_unlock_bh(&cdev->deferq.lock);
+}
+
+static int chtls_get_skb(struct chtls_dev *cdev)
+{
+	cdev->askb = alloc_skb(sizeof(struct tcphdr), GFP_KERNEL);
+	if (!cdev->askb)
+		return -ENOMEM;
+
+	skb_put(cdev->askb, sizeof(struct tcphdr));
+	skb_reset_transport_header(cdev->askb);
+	memset(cdev->askb->data, 0, cdev->askb->len);
+	return 0;
+}
+
+static void *chtls_uld_add(const struct cxgb4_lld_info *info)
+{
+	struct cxgb4_lld_info *lldi;
+	struct chtls_dev *cdev;
+	int i, j;
+
+	cdev = kzalloc(sizeof(*cdev), GFP_KERNEL);
+	if (!cdev)
+		goto out;
+
+	lldi = kzalloc(sizeof(*lldi), GFP_KERNEL);
+	if (!lldi)
+		goto out_lldi;
+
+	if (chtls_get_skb(cdev))
+		goto out_skb;
+
+	*lldi = *info;
+	cdev->lldi = lldi;
+	cdev->pdev = lldi->pdev;
+	cdev->tids = lldi->tids;
+	cdev->ports = lldi->ports;
+	cdev->mtus = lldi->mtus;
+	cdev->tids = lldi->tids;
+	cdev->pfvf = FW_VIID_PFN_G(cxgb4_port_viid(lldi->ports[0]))
+			<< FW_VIID_PFN_S;
+
+	for (i = 0; i < (1 << RSPQ_HASH_BITS); i++) {
+		unsigned int size = 64 - sizeof(struct rsp_ctrl) - 8;
+
+		cdev->rspq_skb_cache[i] = __alloc_skb(size,
+						      gfp_any(), 0,
+						      lldi->nodeid);
+		if (unlikely(!cdev->rspq_skb_cache[i]))
+			goto out_rspq_skb;
+	}
+
+	idr_init(&cdev->hwtid_idr);
+	INIT_WORK(&cdev->deferq_task, process_deferq);
+	spin_lock_init(&cdev->listen_lock);
+	spin_lock_init(&cdev->idr_lock);
+	cdev->send_page_order = min_t(uint, get_order(32768),
+				      send_page_order);
+	cdev->max_host_sndbuf = 48 * 1024;
+
+	if (lldi->vr->key.size)
+		if (chtls_init_kmap(cdev, lldi))
+			goto out_rspq_skb;
+
+	mutex_lock(&cdev_mutex);
+	list_add_tail(&cdev->list, &cdev_list);
+	mutex_unlock(&cdev_mutex);
+
+	return cdev;
+out_rspq_skb:
+	for (j = 0; j < i; j++)
+		kfree_skb(cdev->rspq_skb_cache[j]);
+	kfree_skb(cdev->askb);
+out_skb:
+	kfree(lldi);
+out_lldi:
+	kfree(cdev);
+out:
+	return NULL;
+}
+
+static void chtls_free_all_uld(void)
+{
+	struct chtls_dev *cdev, *tmp;
+
+	mutex_lock(&cdev_mutex);
+	list_for_each_entry_safe(cdev, tmp, &cdev_list, list) {
+		if (cdev->cdev_state == CHTLS_CDEV_STATE_UP) {
+			list_del(&cdev->list);
+			kref_put(&cdev->tlsdev.kref, cdev->tlsdev.release);
+		}
+	}
+	mutex_unlock(&cdev_mutex);
+}
+
+static int chtls_uld_state_change(void *handle, enum cxgb4_state new_state)
+{
+	struct chtls_dev *cdev = handle;
+
+	switch (new_state) {
+	case CXGB4_STATE_UP:
+		chtls_register_dev(cdev);
+		break;
+	case CXGB4_STATE_DOWN:
+		break;
+	case CXGB4_STATE_START_RECOVERY:
+		break;
+	case CXGB4_STATE_DETACH:
+		mutex_lock(&cdev_mutex);
+		list_del(&cdev->list);
+		mutex_unlock(&cdev_mutex);
+		kref_put(&cdev->tlsdev.kref, cdev->tlsdev.release);
+		break;
+	default:
+		break;
+	}
+	return 0;
+}
+
+static struct sk_buff *copy_gl_to_skb_pkt(const struct pkt_gl *gl,
+					  const __be64 *rsp,
+					  u32 pktshift)
+{
+	struct sk_buff *skb;
+
+	/* Allocate space for cpl_pass_accept_req which will be synthesized by
+	 * driver. Once driver synthesizes cpl_pass_accept_req the skb will go
+	 * through the regular cpl_pass_accept_req processing in TOM.
+	 */
+	skb = alloc_skb(size_add(gl->tot_len,
+				 sizeof(struct cpl_pass_accept_req)) -
+			pktshift, GFP_ATOMIC);
+	if (unlikely(!skb))
+		return NULL;
+	__skb_put(skb, gl->tot_len + sizeof(struct cpl_pass_accept_req)
+		   - pktshift);
+	/* For now we will copy  cpl_rx_pkt in the skb */
+	skb_copy_to_linear_data(skb, rsp, sizeof(struct cpl_rx_pkt));
+	skb_copy_to_linear_data_offset(skb, sizeof(struct cpl_pass_accept_req)
+				       , gl->va + pktshift,
+				       gl->tot_len - pktshift);
+
+	return skb;
+}
+
+static int chtls_recv_packet(struct chtls_dev *cdev,
+			     const struct pkt_gl *gl, const __be64 *rsp)
+{
+	unsigned int opcode = *(u8 *)rsp;
+	struct sk_buff *skb;
+	int ret;
+
+	skb = copy_gl_to_skb_pkt(gl, rsp, cdev->lldi->sge_pktshift);
+	if (!skb)
+		return -ENOMEM;
+
+	ret = chtls_handlers[opcode](cdev, skb);
+	if (ret & CPL_RET_BUF_DONE)
+		kfree_skb(skb);
+
+	return 0;
+}
+
+static int chtls_recv_rsp(struct chtls_dev *cdev, const __be64 *rsp)
+{
+	unsigned long rspq_bin;
+	unsigned int opcode;
+	struct sk_buff *skb;
+	unsigned int len;
+	int ret;
+
+	len = 64 - sizeof(struct rsp_ctrl) - 8;
+	opcode = *(u8 *)rsp;
+
+	rspq_bin = hash_ptr((void *)rsp, RSPQ_HASH_BITS);
+	skb = cdev->rspq_skb_cache[rspq_bin];
+	if (skb && !skb_is_nonlinear(skb) &&
+	    !skb_shared(skb) && !skb_cloned(skb)) {
+		refcount_inc(&skb->users);
+		if (refcount_read(&skb->users) == 2) {
+			__skb_trim(skb, 0);
+			if (skb_tailroom(skb) >= len)
+				goto copy_out;
+		}
+		refcount_dec(&skb->users);
+	}
+	skb = alloc_skb(len, GFP_ATOMIC);
+	if (unlikely(!skb))
+		return -ENOMEM;
+
+copy_out:
+	__skb_put(skb, len);
+	skb_copy_to_linear_data(skb, rsp, len);
+	skb_reset_network_header(skb);
+	skb_reset_transport_header(skb);
+	ret = chtls_handlers[opcode](cdev, skb);
+
+	if (ret & CPL_RET_BUF_DONE)
+		kfree_skb(skb);
+	return 0;
+}
+
+static void chtls_recv(struct chtls_dev *cdev,
+		       struct sk_buff **skbs, const __be64 *rsp)
+{
+	struct sk_buff *skb = *skbs;
+	unsigned int opcode;
+	int ret;
+
+	opcode = *(u8 *)rsp;
+
+	__skb_push(skb, sizeof(struct rss_header));
+	skb_copy_to_linear_data(skb, rsp, sizeof(struct rss_header));
+
+	ret = chtls_handlers[opcode](cdev, skb);
+	if (ret & CPL_RET_BUF_DONE)
+		kfree_skb(skb);
+}
+
+static int chtls_uld_rx_handler(void *handle, const __be64 *rsp,
+				const struct pkt_gl *gl)
+{
+	struct chtls_dev *cdev = handle;
+	unsigned int opcode;
+	struct sk_buff *skb;
+
+	opcode = *(u8 *)rsp;
+
+	if (unlikely(opcode == CPL_RX_PKT)) {
+		if (chtls_recv_packet(cdev, gl, rsp) < 0)
+			goto nomem;
+		return 0;
+	}
+
+	if (!gl)
+		return chtls_recv_rsp(cdev, rsp);
+
+#define RX_PULL_LEN 128
+	skb = cxgb4_pktgl_to_skb(gl, RX_PULL_LEN, RX_PULL_LEN);
+	if (unlikely(!skb))
+		goto nomem;
+	chtls_recv(cdev, &skb, rsp);
+	return 0;
+
+nomem:
+	return -ENOMEM;
+}
+
+static int do_chtls_getsockopt(struct sock *sk, char __user *optval,
+			       int __user *optlen)
+{
+	struct tls_crypto_info crypto_info = { 0 };
+
+	crypto_info.version = TLS_1_2_VERSION;
+	if (copy_to_user(optval, &crypto_info, sizeof(struct tls_crypto_info)))
+		return -EFAULT;
+	return 0;
+}
+
+static int chtls_getsockopt(struct sock *sk, int level, int optname,
+			    char __user *optval, int __user *optlen)
+{
+	struct tls_context *ctx = tls_get_ctx(sk);
+
+	if (level != SOL_TLS)
+		return ctx->sk_proto->getsockopt(sk, level,
+						 optname, optval, optlen);
+
+	return do_chtls_getsockopt(sk, optval, optlen);
+}
+
+static int do_chtls_setsockopt(struct sock *sk, int optname,
+			       sockptr_t optval, unsigned int optlen)
+{
+	struct tls_crypto_info *crypto_info, tmp_crypto_info;
+	struct chtls_sock *csk;
+	int keylen;
+	int cipher_type;
+	int rc = 0;
+
+	csk = rcu_dereference_sk_user_data(sk);
+
+	if (sockptr_is_null(optval) || optlen < sizeof(*crypto_info)) {
+		rc = -EINVAL;
+		goto out;
+	}
+
+	rc = copy_from_sockptr(&tmp_crypto_info, optval, sizeof(*crypto_info));
+	if (rc) {
+		rc = -EFAULT;
+		goto out;
+	}
+
+	/* check version */
+	if (tmp_crypto_info.version != TLS_1_2_VERSION) {
+		rc = -ENOTSUPP;
+		goto out;
+	}
+
+	crypto_info = (struct tls_crypto_info *)&csk->tlshws.crypto_info;
+
+	/* GCM mode of AES supports 128 and 256 bit encryption, so
+	 * copy keys from user based on GCM cipher type.
+	 */
+	switch (tmp_crypto_info.cipher_type) {
+	case TLS_CIPHER_AES_GCM_128: {
+		/* Obtain version and type from previous copy */
+		crypto_info[0] = tmp_crypto_info;
+		/* Now copy the following data */
+		rc = copy_from_sockptr_offset((char *)crypto_info +
+				sizeof(*crypto_info),
+				optval, sizeof(*crypto_info),
+				sizeof(struct tls12_crypto_info_aes_gcm_128)
+				- sizeof(*crypto_info));
+
+		if (rc) {
+			rc = -EFAULT;
+			goto out;
+		}
+
+		keylen = TLS_CIPHER_AES_GCM_128_KEY_SIZE;
+		cipher_type = TLS_CIPHER_AES_GCM_128;
+		break;
+	}
+	case TLS_CIPHER_AES_GCM_256: {
+		crypto_info[0] = tmp_crypto_info;
+		rc = copy_from_sockptr_offset((char *)crypto_info +
+				sizeof(*crypto_info),
+				optval, sizeof(*crypto_info),
+				sizeof(struct tls12_crypto_info_aes_gcm_256)
+				- sizeof(*crypto_info));
+
+		if (rc) {
+			rc = -EFAULT;
+			goto out;
+		}
+
+		keylen = TLS_CIPHER_AES_GCM_256_KEY_SIZE;
+		cipher_type = TLS_CIPHER_AES_GCM_256;
+		break;
+	}
+	default:
+		rc = -EINVAL;
+		goto out;
+	}
+	rc = chtls_setkey(csk, keylen, optname, cipher_type);
+out:
+	return rc;
+}
+
+static int chtls_setsockopt(struct sock *sk, int level, int optname,
+			    sockptr_t optval, unsigned int optlen)
+{
+	struct tls_context *ctx = tls_get_ctx(sk);
+
+	if (level != SOL_TLS)
+		return ctx->sk_proto->setsockopt(sk, level,
+						 optname, optval, optlen);
+
+	return do_chtls_setsockopt(sk, optname, optval, optlen);
+}
+
+static struct cxgb4_uld_info chtls_uld_info = {
+	.name = DRV_NAME,
+	.nrxq = MAX_ULD_QSETS,
+	.ntxq = MAX_ULD_QSETS,
+	.rxq_size = 1024,
+	.add = chtls_uld_add,
+	.state_change = chtls_uld_state_change,
+	.rx_handler = chtls_uld_rx_handler,
+};
+
+void chtls_install_cpl_ops(struct sock *sk)
+{
+	if (sk->sk_family == AF_INET)
+		sk->sk_prot = &chtls_cpl_prot;
+	else
+		sk->sk_prot = &chtls_cpl_protv6;
+}
+
+static void __init chtls_init_ulp_ops(void)
+{
+	chtls_cpl_prot			= tcp_prot;
+	chtls_init_rsk_ops(&chtls_cpl_prot, &chtls_rsk_ops,
+			   &tcp_prot, PF_INET);
+	chtls_cpl_prot.close		= chtls_close;
+	chtls_cpl_prot.disconnect	= chtls_disconnect;
+	chtls_cpl_prot.destroy		= chtls_destroy_sock;
+	chtls_cpl_prot.shutdown		= chtls_shutdown;
+	chtls_cpl_prot.sendmsg		= chtls_sendmsg;
+	chtls_cpl_prot.splice_eof	= chtls_splice_eof;
+	chtls_cpl_prot.recvmsg		= chtls_recvmsg;
+	chtls_cpl_prot.setsockopt	= chtls_setsockopt;
+	chtls_cpl_prot.getsockopt	= chtls_getsockopt;
+#if IS_ENABLED(CONFIG_IPV6)
+	chtls_cpl_protv6		= chtls_cpl_prot;
+	chtls_init_rsk_ops(&chtls_cpl_protv6, &chtls_rsk_opsv6,
+			   &tcpv6_prot, PF_INET6);
+#endif
+}
+
+static int __init chtls_register(void)
+{
+	chtls_init_ulp_ops();
+	register_listen_notifier(&listen_notifier);
+	cxgb4_register_uld(CXGB4_ULD_TLS, &chtls_uld_info);
+	return 0;
+}
+
+static void __exit chtls_unregister(void)
+{
+	unregister_listen_notifier(&listen_notifier);
+	chtls_free_all_uld();
+	cxgb4_unregister_uld(CXGB4_ULD_TLS);
+}
+
+module_init(chtls_register);
+module_exit(chtls_unregister);
+
+MODULE_DESCRIPTION("Chelsio TLS Inline driver");
+MODULE_LICENSE("GPL");
+MODULE_AUTHOR("Chelsio Communications");
+MODULE_VERSION(CHTLS_DRV_VERSION);
diff --git a/drivers/net/ethernet/chelsio/libcxgb/Makefile b/drivers/net/ethernet/chelsio/libcxgb/Makefile
index 2534e30a1560..fbedc31674b3 100644
--- a/drivers/net/ethernet/chelsio/libcxgb/Makefile
+++ b/drivers/net/ethernet/chelsio/libcxgb/Makefile
@@ -1,4 +1,5 @@
-ccflags-y := -Idrivers/net/ethernet/chelsio/cxgb4
+# SPDX-License-Identifier: GPL-2.0-only
+ccflags-y := -I $(src)/../cxgb4
 
 obj-$(CONFIG_CHELSIO_LIB) += libcxgb.o
 
diff --git a/drivers/net/ethernet/chelsio/libcxgb/libcxgb_cm.c b/drivers/net/ethernet/chelsio/libcxgb/libcxgb_cm.c
index d04a6c163445..da8d10475a08 100644
--- a/drivers/net/ethernet/chelsio/libcxgb/libcxgb_cm.c
+++ b/drivers/net/ethernet/chelsio/libcxgb/libcxgb_cm.c
@@ -32,6 +32,7 @@
 
 #include <linux/tcp.h>
 #include <linux/ipv6.h>
+#include <net/inet_ecn.h>
 #include <net/route.h>
 #include <net/ip6_route.h>
 
@@ -99,7 +100,7 @@ cxgb_find_route(struct cxgb4_lld_info *lldi,
 
 	rt = ip_route_output_ports(&init_net, &fl4, NULL, peer_ip, local_ip,
 				   peer_port, local_port, IPPROTO_TCP,
-				   tos, 0);
+				   tos & ~INET_ECN_MASK, 0);
 	if (IS_ERR(rt))
 		return NULL;
 	n = dst_neigh_lookup(&rt->dst, &peer_ip);
diff --git a/drivers/net/ethernet/chelsio/libcxgb/libcxgb_cm.h b/drivers/net/ethernet/chelsio/libcxgb/libcxgb_cm.h
index 515b94ff9080..240ba9d4c399 100644
--- a/drivers/net/ethernet/chelsio/libcxgb/libcxgb_cm.h
+++ b/drivers/net/ethernet/chelsio/libcxgb/libcxgb_cm.h
@@ -90,8 +90,7 @@ cxgb_mk_tid_release(struct sk_buff *skb, u32 len, u32 tid, u16 chan)
 {
 	struct cpl_tid_release *req;
 
-	req = (struct cpl_tid_release *)__skb_put(skb, len);
-	memset(req, 0, len);
+	req = __skb_put_zero(skb, len);
 
 	INIT_TP_WR(req, tid);
 	OPCODE_TID(req) = cpu_to_be32(MK_OPCODE_TID(CPL_TID_RELEASE, tid));
@@ -104,8 +103,7 @@ cxgb_mk_close_con_req(struct sk_buff *skb, u32 len, u32 tid, u16 chan,
 {
 	struct cpl_close_con_req *req;
 
-	req = (struct cpl_close_con_req *)__skb_put(skb, len);
-	memset(req, 0, len);
+	req = __skb_put_zero(skb, len);
 
 	INIT_TP_WR(req, tid);
 	OPCODE_TID(req) = cpu_to_be32(MK_OPCODE_TID(CPL_CLOSE_CON_REQ, tid));
@@ -119,8 +117,7 @@ cxgb_mk_abort_req(struct sk_buff *skb, u32 len, u32 tid, u16 chan,
 {
 	struct cpl_abort_req *req;
 
-	req = (struct cpl_abort_req *)__skb_put(skb, len);
-	memset(req, 0, len);
+	req = __skb_put_zero(skb, len);
 
 	INIT_TP_WR(req, tid);
 	OPCODE_TID(req) = cpu_to_be32(MK_OPCODE_TID(CPL_ABORT_REQ, tid));
@@ -134,8 +131,7 @@ cxgb_mk_abort_rpl(struct sk_buff *skb, u32 len, u32 tid, u16 chan)
 {
 	struct cpl_abort_rpl *rpl;
 
-	rpl = (struct cpl_abort_rpl *)__skb_put(skb, len);
-	memset(rpl, 0, len);
+	rpl = __skb_put_zero(skb, len);
 
 	INIT_TP_WR(rpl, tid);
 	OPCODE_TID(rpl) = cpu_to_be32(MK_OPCODE_TID(CPL_ABORT_RPL, tid));
@@ -149,8 +145,7 @@ cxgb_mk_rx_data_ack(struct sk_buff *skb, u32 len, u32 tid, u16 chan,
 {
 	struct cpl_rx_data_ack *req;
 
-	req = (struct cpl_rx_data_ack *)__skb_put(skb, len);
-	memset(req, 0, len);
+	req = __skb_put_zero(skb, len);
 
 	INIT_TP_WR(req, tid);
 	OPCODE_TID(req) = cpu_to_be32(MK_OPCODE_TID(CPL_RX_DATA_ACK, tid));
diff --git a/drivers/net/ethernet/chelsio/libcxgb/libcxgb_ppm.c b/drivers/net/ethernet/chelsio/libcxgb/libcxgb_ppm.c
index 0ed161642371..2e3973a32d9d 100644
--- a/drivers/net/ethernet/chelsio/libcxgb/libcxgb_ppm.c
+++ b/drivers/net/ethernet/chelsio/libcxgb/libcxgb_ppm.c
@@ -35,7 +35,6 @@
  */
 
 #define DRV_NAME "libcxgb"
-#define DRV_VERSION "1.0.0-ko"
 #define pr_fmt(fmt) DRV_NAME ": " fmt
 
 #include <linux/kernel.h>
@@ -123,6 +122,9 @@ static int ppm_get_cpu_entries(struct cxgbi_ppm *ppm, unsigned int count,
 	unsigned int cpu;
 	int i;
 
+	if (!ppm->pool)
+		return -EINVAL;
+
 	cpu = get_cpu();
 	pool = per_cpu_ptr(ppm->pool, cpu);
 	spin_lock_bh(&pool->lock);
@@ -169,7 +171,9 @@ static int ppm_get_entries(struct cxgbi_ppm *ppm, unsigned int count,
 	}
 
 	ppm->next = i + count;
-	if (ppm->next >= ppm->bmap_index_max)
+	if (ppm->max_index_in_edram && (ppm->next >= ppm->max_index_in_edram))
+		ppm->next = 0;
+	else if (ppm->next >= ppm->bmap_index_max)
 		ppm->next = 0;
 
 	spin_unlock_bh(&ppm->map_lock);
@@ -338,10 +342,10 @@ int cxgbi_ppm_release(struct cxgbi_ppm *ppm)
 }
 EXPORT_SYMBOL(cxgbi_ppm_release);
 
-static struct cxgbi_ppm_pool *ppm_alloc_cpu_pool(unsigned int *total,
-						 unsigned int *pcpu_ppmax)
+static struct cxgbi_ppm_pool __percpu *
+ppm_alloc_cpu_pool(unsigned int *total, unsigned int *pcpu_ppmax)
 {
-	struct cxgbi_ppm_pool *pools;
+	struct cxgbi_ppm_pool __percpu *pools;
 	unsigned int ppmax = (*total) / num_possible_cpus();
 	unsigned int max = (PCPU_MIN_UNIT_SIZE - sizeof(*pools)) << 3;
 	unsigned int bmap;
@@ -354,7 +358,10 @@ static struct cxgbi_ppm_pool *ppm_alloc_cpu_pool(unsigned int *total,
 		ppmax = max;
 
 	/* pool size must be multiple of unsigned long */
-	bmap = BITS_TO_LONGS(ppmax);
+	bmap = ppmax / BITS_PER_TYPE(unsigned long);
+	if (!bmap)
+		return NULL;
+
 	ppmax = (bmap * sizeof(unsigned long)) << 3;
 
 	alloc_sz = sizeof(*pools) + sizeof(unsigned long) * bmap;
@@ -379,18 +386,36 @@ static struct cxgbi_ppm_pool *ppm_alloc_cpu_pool(unsigned int *total,
 
 int cxgbi_ppm_init(void **ppm_pp, struct net_device *ndev,
 		   struct pci_dev *pdev, void *lldev,
-		   struct cxgbi_tag_format *tformat,
-		   unsigned int ppmax,
-		   unsigned int llimit,
-		   unsigned int start,
-		   unsigned int reserve_factor)
+		   struct cxgbi_tag_format *tformat, unsigned int iscsi_size,
+		   unsigned int llimit, unsigned int start,
+		   unsigned int reserve_factor, unsigned int iscsi_edram_start,
+		   unsigned int iscsi_edram_size)
 {
 	struct cxgbi_ppm *ppm = (struct cxgbi_ppm *)(*ppm_pp);
-	struct cxgbi_ppm_pool *pool = NULL;
-	unsigned int ppmax_pool = 0;
+	struct cxgbi_ppm_pool __percpu *pool = NULL;
 	unsigned int pool_index_max = 0;
-	unsigned int alloc_sz;
+	unsigned int ppmax_pool = 0;
 	unsigned int ppod_bmap_size;
+	unsigned int alloc_sz;
+	unsigned int ppmax;
+
+	if (!iscsi_edram_start)
+		iscsi_edram_size = 0;
+
+	if (iscsi_edram_size &&
+	    ((iscsi_edram_start + iscsi_edram_size) != start)) {
+		pr_err("iscsi ppod region not contiguous: EDRAM start 0x%x "
+			"size 0x%x DDR start 0x%x\n",
+			iscsi_edram_start, iscsi_edram_size, start);
+		return -EINVAL;
+	}
+
+	if (iscsi_edram_size) {
+		reserve_factor = 0;
+		start = iscsi_edram_start;
+	}
+
+	ppmax = (iscsi_edram_size + iscsi_size) >> PPOD_SIZE_SHIFT;
 
 	if (ppm) {
 		pr_info("ippm: %s, ppm 0x%p,0x%p already initialized, %u/%u.\n",
@@ -402,6 +427,10 @@ int cxgbi_ppm_init(void **ppm_pp, struct net_device *ndev,
 	if (reserve_factor) {
 		ppmax_pool = ppmax / reserve_factor;
 		pool = ppm_alloc_cpu_pool(&ppmax_pool, &pool_index_max);
+		if (!pool) {
+			ppmax_pool = 0;
+			reserve_factor = 0;
+		}
 
 		pr_debug("%s: ppmax %u, cpu total %u, per cpu %u.\n",
 			 ndev->name, ppmax, ppmax_pool, pool_index_max);
@@ -412,12 +441,10 @@ int cxgbi_ppm_init(void **ppm_pp, struct net_device *ndev,
 			ppmax * (sizeof(struct cxgbi_ppod_data)) +
 			ppod_bmap_size * sizeof(unsigned long);
 
-	ppm = vmalloc(alloc_sz);
+	ppm = vzalloc(alloc_sz);
 	if (!ppm)
 		goto release_ppm_pool;
 
-	memset(ppm, 0, alloc_sz);
-
 	ppm->ppod_bmap = (unsigned long *)(&ppm->ppod_data[ppmax]);
 
 	if ((ppod_bmap_size >> 3) > (ppmax - ppmax_pool)) {
@@ -429,6 +456,14 @@ int cxgbi_ppm_init(void **ppm_pp, struct net_device *ndev,
 			__func__, ppmax, ppmax_pool, ppod_bmap_size, start,
 			end);
 	}
+	if (iscsi_edram_size) {
+		unsigned int first_ddr_idx =
+				iscsi_edram_size >> PPOD_SIZE_SHIFT;
+
+		ppm->max_index_in_edram = first_ddr_idx - 1;
+		bitmap_set(ppm->ppod_bmap, first_ddr_idx, 1);
+		pr_debug("reserved %u ppod in bitmap\n", first_ddr_idx);
+	}
 
 	spin_lock_init(&ppm->map_lock);
 	kref_init(&ppm->refcnt);
@@ -494,5 +529,4 @@ EXPORT_SYMBOL(cxgbi_tagmask_set);
 
 MODULE_AUTHOR("Chelsio Communications");
 MODULE_DESCRIPTION("Chelsio common library");
-MODULE_VERSION(DRV_VERSION);
 MODULE_LICENSE("Dual BSD/GPL");
diff --git a/drivers/net/ethernet/chelsio/libcxgb/libcxgb_ppm.h b/drivers/net/ethernet/chelsio/libcxgb/libcxgb_ppm.h
index a91ad766cef0..1b4156461ba1 100644
--- a/drivers/net/ethernet/chelsio/libcxgb/libcxgb_ppm.h
+++ b/drivers/net/ethernet/chelsio/libcxgb/libcxgb_ppm.h
@@ -122,7 +122,7 @@ struct cxgbi_ppm_pool {
 	unsigned int base;		/* base index */
 	unsigned int next;		/* next possible free index */
 	spinlock_t lock;		/* ppm pool lock */
-	unsigned long bmap[0];
+	unsigned long bmap[];
 } ____cacheline_aligned_in_smp;
 
 struct cxgbi_ppm {
@@ -143,8 +143,9 @@ struct cxgbi_ppm {
 	spinlock_t map_lock;		/* ppm map lock */
 	unsigned int bmap_index_max;
 	unsigned int next;
+	unsigned int max_index_in_edram;
 	unsigned long *ppod_bmap;
-	struct cxgbi_ppod_data ppod_data[0];
+	struct cxgbi_ppod_data ppod_data[];
 };
 
 #define DDP_THRESHOLD		512
@@ -324,9 +325,9 @@ int cxgbi_ppm_ppods_reserve(struct cxgbi_ppm *, unsigned short nr_pages,
 			    unsigned long caller_data);
 int cxgbi_ppm_init(void **ppm_pp, struct net_device *, struct pci_dev *,
 		   void *lldev, struct cxgbi_tag_format *,
-		   unsigned int ppmax, unsigned int llimit,
-		   unsigned int start,
-		   unsigned int reserve_factor);
+		   unsigned int iscsi_size, unsigned int llimit,
+		   unsigned int start, unsigned int reserve_factor,
+		   unsigned int edram_start, unsigned int edram_size);
 int cxgbi_ppm_release(struct cxgbi_ppm *ppm);
 void cxgbi_tagmask_check(unsigned int tagmask, struct cxgbi_tag_format *);
 unsigned int cxgbi_tagmask_set(unsigned int ppmax);
diff --git a/drivers/net/ethernet/cirrus/Kconfig b/drivers/net/ethernet/cirrus/Kconfig
index 5ab912937aff..5bdf731d9503 100644
--- a/drivers/net/ethernet/cirrus/Kconfig
+++ b/drivers/net/ethernet/cirrus/Kconfig
@@ -1,3 +1,4 @@
+# SPDX-License-Identifier: GPL-2.0-only
 #
 # Cirrus network device configuration
 #
@@ -5,8 +6,8 @@
 config NET_VENDOR_CIRRUS
 	bool "Cirrus devices"
 	default y
-	depends on ISA || EISA || ARM || MAC
-	---help---
+	depends on ISA || EISA || ARM || MAC || COMPILE_TEST
+	help
 	  If you have a network (Ethernet) card belonging to this class, say Y.
 
 	  Note that the answer to this question doesn't directly affect the
@@ -17,30 +18,38 @@ config NET_VENDOR_CIRRUS
 if NET_VENDOR_CIRRUS
 
 config CS89x0
-	tristate "CS89x0 support"
-	depends on ISA || EISA || ARM
-	---help---
+	tristate
+
+config CS89x0_ISA
+	tristate "CS89x0 ISA driver support"
+	depends on HAS_IOPORT_MAP
+	depends on ISA
+	depends on !PPC32
+	depends on CS89x0_PLATFORM=n
+	select NETDEV_LEGACY_INIT
+	select CS89x0
+	help
 	  Support for CS89x0 chipset based Ethernet cards. If you have a
 	  network (Ethernet) card of this type, say Y and read the file
-	  <file:Documentation/networking/cs89x0.txt>.
+	  <file:Documentation/networking/device_drivers/ethernet/cirrus/cs89x0.rst>.
 
 	  To compile this driver as a module, choose M here. The module
 	  will be called cs89x0.
 
 config CS89x0_PLATFORM
-	bool "CS89x0 platform driver support" if HAS_IOPORT_MAP
-	default !HAS_IOPORT_MAP
-	depends on CS89x0
+	tristate "CS89x0 platform driver support"
+	depends on ARM || (COMPILE_TEST && !PPC)
+	select CS89x0
 	help
-	  Say Y to compile the cs89x0 driver as a platform driver. This
-	  makes this driver suitable for use on certain evaluation boards
-	  such as the iMX21ADS.
+	  Say Y to compile the cs89x0 platform driver. This makes this driver
+	  suitable for use on certain evaluation boards such as the iMX21ADS.
 
-	  If you are unsure, say N.
+	  To compile this driver as a module, choose M here. The module
+	  will be called cs89x0.
 
 config EP93XX_ETH
 	tristate "EP93xx Ethernet support"
-	depends on ARM && ARCH_EP93XX
+	depends on (ARM && ARCH_EP93XX) || COMPILE_TEST
 	select MII
 	help
 	  This is a driver for the ethernet hardware included in EP93xx CPUs.
@@ -49,7 +58,7 @@ config EP93XX_ETH
 config MAC89x0
 	tristate "Macintosh CS89x0 based ethernet cards"
 	depends on MAC
-	---help---
+	help
 	  Support for CS89x0 chipset based Ethernet cards.  If you have a
 	  Nubus or LC-PDS network (Ethernet) card of this type, say Y here.
 
diff --git a/drivers/net/ethernet/cirrus/Makefile b/drivers/net/ethernet/cirrus/Makefile
index ca245e2b5d98..84865e593788 100644
--- a/drivers/net/ethernet/cirrus/Makefile
+++ b/drivers/net/ethernet/cirrus/Makefile
@@ -1,3 +1,4 @@
+# SPDX-License-Identifier: GPL-2.0-only
 #
 # Makefile for the Cirrus network device drivers.
 #
diff --git a/drivers/net/ethernet/cirrus/cs89x0.c b/drivers/net/ethernet/cirrus/cs89x0.c
index da5b58b853e2..fa5857923db4 100644
--- a/drivers/net/ethernet/cirrus/cs89x0.c
+++ b/drivers/net/ethernet/cirrus/cs89x0.c
@@ -54,7 +54,6 @@
 #include <linux/netdevice.h>
 #include <linux/etherdevice.h>
 #include <linux/of.h>
-#include <linux/of_device.h>
 #include <linux/platform_device.h>
 #include <linux/kernel.h>
 #include <linux/types.h>
@@ -72,6 +71,8 @@
 #include <linux/gfp.h>
 #include <linux/io.h>
 
+#include <net/Space.h>
+
 #include <asm/irq.h>
 #include <linux/atomic.h>
 #if ALLOW_DMA
@@ -104,7 +105,7 @@ static char version[] __initdata =
  * them to system IRQ numbers. This mapping is card specific and is set to
  * the configuration of the Cirrus Eval board for this chip.
  */
-#ifndef CONFIG_CS89x0_PLATFORM
+#if IS_ENABLED(CONFIG_CS89x0_ISA)
 static unsigned int netcard_portlist[] __used __initdata = {
 	0x300, 0x320, 0x340, 0x360, 0x200, 0x220, 0x240,
 	0x260, 0x280, 0x2a0, 0x2c0, 0x2e0, 0
@@ -292,7 +293,7 @@ write_irq(struct net_device *dev, int chip_type, int irq)
 	int i;
 
 	if (chip_type == CS8900) {
-#ifndef CONFIG_CS89x0_PLATFORM
+#if IS_ENABLED(CONFIG_CS89x0_ISA)
 		/* Search the mapping table for the corresponding IRQ pin. */
 		for (i = 0; i != ARRAY_SIZE(cs8900_irq_map); i++)
 			if (cs8900_irq_map[i] == irq)
@@ -450,11 +451,10 @@ skip_this_frame:
 
 	if (bp + length > lp->end_dma_buff) {
 		int semi_cnt = lp->end_dma_buff - bp;
-		memcpy(skb_put(skb, semi_cnt), bp, semi_cnt);
-		memcpy(skb_put(skb, length - semi_cnt), lp->dma_buff,
-		       length - semi_cnt);
+		skb_put_data(skb, bp, semi_cnt);
+		skb_put_data(skb, lp->dma_buff, length - semi_cnt);
 	} else {
-		memcpy(skb_put(skb, length), bp, length);
+		skb_put_data(skb, bp, length);
 	}
 	bp += (length + 3) & ~3;
 	if (bp >= lp->end_dma_buff)
@@ -860,7 +860,7 @@ net_open(struct net_device *dev)
 			goto bad_out;
 		}
 	} else {
-#if !defined(CONFIG_CS89x0_PLATFORM)
+#if IS_ENABLED(CONFIG_CS89x0_ISA)
 		if (((1 << dev->irq) & lp->irq_map) == 0) {
 			pr_err("%s: IRQ %d is not in our map of allowable IRQs, which is %x\n",
 			       dev->name, dev->irq, lp->irq_map);
@@ -986,7 +986,7 @@ release_irq:
 		if (result == DETECTED_NONE) {
 			pr_warn("%s: 10Base-5 (AUI) has no cable\n", dev->name);
 			if (lp->auto_neg_cnf & IMM_BIT) /* check "ignore missing media" bit */
-				result = DETECTED_AUI; /* Yes! I don't care if I see a carrrier */
+				result = DETECTED_AUI; /* Yes! I don't care if I see a carrier */
 		}
 		break;
 	case A_CNF_MEDIA_10B_2:
@@ -1129,7 +1129,7 @@ net_get_stats(struct net_device *dev)
 	return &dev->stats;
 }
 
-static void net_timeout(struct net_device *dev)
+static void net_timeout(struct net_device *dev, unsigned int txqueue)
 {
 	/* If we get here, some higher level has decided we are broken.
 	   There should really be a "kick me" function call instead. */
@@ -1228,7 +1228,7 @@ static int set_mac_address(struct net_device *dev, void *p)
 	if (netif_running(dev))
 		return -EBUSY;
 
-	memcpy(dev->dev_addr, addr->sa_data, dev->addr_len);
+	eth_hw_addr_set(dev, addr->sa_data);
 
 	cs89_dbg(0, debug, "%s: Setting MAC address to %pM\n",
 		 dev->name, dev->dev_addr);
@@ -1315,6 +1315,7 @@ cs89x0_probe1(struct net_device *dev, void __iomem *ioaddr, int modular)
 	int tmp;
 	unsigned rev_type = 0;
 	int eeprom_buff[CHKSUM_LEN];
+	u8 addr[ETH_ALEN];
 	int retval;
 
 	/* Initialize the device structure. */
@@ -1388,9 +1389,10 @@ cs89x0_probe1(struct net_device *dev, void __iomem *ioaddr, int modular)
 		for (i = 0; i < ETH_ALEN / 2; i++) {
 			unsigned int Addr;
 			Addr = readreg(dev, PP_IA + i * 2);
-			dev->dev_addr[i * 2] = Addr & 0xFF;
-			dev->dev_addr[i * 2 + 1] = Addr >> 8;
+			addr[i * 2] = Addr & 0xFF;
+			addr[i * 2 + 1] = Addr >> 8;
 		}
+		eth_hw_addr_set(dev, addr);
 
 		/* Load the Adapter Configuration.
 		 * Note:  Barring any more specific information from some
@@ -1465,9 +1467,10 @@ cs89x0_probe1(struct net_device *dev, void __iomem *ioaddr, int modular)
 		/* eeprom_buff has 32-bit ints, so we can't just memcpy it */
 		/* store the initial memory base address */
 		for (i = 0; i < ETH_ALEN / 2; i++) {
-			dev->dev_addr[i * 2] = eeprom_buff[i];
-			dev->dev_addr[i * 2 + 1] = eeprom_buff[i] >> 8;
+			addr[i * 2] = eeprom_buff[i];
+			addr[i * 2 + 1] = eeprom_buff[i] >> 8;
 		}
+		eth_hw_addr_set(dev, addr);
 		cs89_dbg(1, debug, "%s: new adapter_cnf: 0x%x\n",
 			 dev->name, lp->adapter_cnf);
 	}
@@ -1524,7 +1527,7 @@ cs89x0_probe1(struct net_device *dev, void __iomem *ioaddr, int modular)
 			dev->irq = i;
 	} else {
 		i = lp->isa_config & INT_NO_MASK;
-#ifndef CONFIG_CS89x0_PLATFORM
+#if IS_ENABLED(CONFIG_CS89x0_ISA)
 		if (lp->chip_type == CS8900) {
 			/* Translate the IRQ using the IRQ mapping table. */
 			if (i >= ARRAY_SIZE(cs8900_irq_map))
@@ -1577,7 +1580,7 @@ out1:
 	return retval;
 }
 
-#ifndef CONFIG_CS89x0_PLATFORM
+#if IS_ENABLED(CONFIG_CS89x0_ISA)
 /*
  * This function converts the I/O port address used by the cs89x0_probe() and
  * init_module() functions to the I/O memory address used by the
@@ -1683,11 +1686,7 @@ out:
 	pr_warn("no cs8900 or cs8920 detected.  Be sure to disable PnP with SETUP\n");
 	return ERR_PTR(err);
 }
-#endif
-#endif
-
-#if defined(MODULE) && !defined(CONFIG_CS89x0_PLATFORM)
-
+#else
 static struct net_device *dev_cs89x0;
 
 /* Support the 'debug' module parm even if we're compiled for non-debug to
@@ -1758,9 +1757,9 @@ MODULE_LICENSE("GPL");
  * (hw or software util)
  */
 
-int __init init_module(void)
+static int __init cs89x0_isa_init_module(void)
 {
-	struct net_device *dev = alloc_etherdev(sizeof(struct net_local));
+	struct net_device *dev;
 	struct net_local *lp;
 	int ret = 0;
 
@@ -1769,6 +1768,7 @@ int __init init_module(void)
 #else
 	debug = 0;
 #endif
+	dev = alloc_etherdev(sizeof(struct net_local));
 	if (!dev)
 		return -ENOMEM;
 
@@ -1827,9 +1827,9 @@ out:
 	free_netdev(dev);
 	return ret;
 }
+module_init(cs89x0_isa_init_module);
 
-void __exit
-cleanup_module(void)
+static void __exit cs89x0_isa_cleanup_module(void)
 {
 	struct net_local *lp = netdev_priv(dev_cs89x0);
 
@@ -1839,31 +1839,27 @@ cleanup_module(void)
 	release_region(dev_cs89x0->base_addr, NETCARD_IO_EXTENT);
 	free_netdev(dev_cs89x0);
 }
-#endif /* MODULE && !CONFIG_CS89x0_PLATFORM */
+module_exit(cs89x0_isa_cleanup_module);
+#endif /* MODULE */
+#endif /* CONFIG_CS89x0_ISA */
 
-#ifdef CONFIG_CS89x0_PLATFORM
+#if IS_ENABLED(CONFIG_CS89x0_PLATFORM)
 static int __init cs89x0_platform_probe(struct platform_device *pdev)
 {
 	struct net_device *dev = alloc_etherdev(sizeof(struct net_local));
-	struct net_local *lp;
-	struct resource *mem_res;
 	void __iomem *virt_addr;
 	int err;
 
 	if (!dev)
 		return -ENOMEM;
 
-	lp = netdev_priv(dev);
-
 	dev->irq = platform_get_irq(pdev, 0);
-	if (dev->irq <= 0) {
-		dev_warn(&dev->dev, "interrupt resource missing\n");
-		err = -ENXIO;
+	if (dev->irq < 0) {
+		err = dev->irq;
 		goto free;
 	}
 
-	mem_res = platform_get_resource(pdev, IORESOURCE_MEM, 0);
-	virt_addr = devm_ioremap_resource(&pdev->dev, mem_res);
+	virt_addr = devm_platform_ioremap_resource(pdev, 0);
 	if (IS_ERR(virt_addr)) {
 		err = PTR_ERR(virt_addr);
 		goto free;
@@ -1883,7 +1879,7 @@ free:
 	return err;
 }
 
-static int cs89x0_platform_remove(struct platform_device *pdev)
+static void cs89x0_platform_remove(struct platform_device *pdev)
 {
 	struct net_device *dev = platform_get_drvdata(pdev);
 
@@ -1893,7 +1889,6 @@ static int cs89x0_platform_remove(struct platform_device *pdev)
 	 */
 	unregister_netdev(dev);
 	free_netdev(dev);
-	return 0;
 }
 
 static const struct of_device_id __maybe_unused cs89x0_match[] = {
@@ -1908,9 +1903,13 @@ static struct platform_driver cs89x0_driver = {
 		.name		= DRV_NAME,
 		.of_match_table	= of_match_ptr(cs89x0_match),
 	},
-	.remove	= cs89x0_platform_remove,
+	.remove = cs89x0_platform_remove,
 };
 
 module_platform_driver_probe(cs89x0_driver, cs89x0_platform_probe);
 
 #endif /* CONFIG_CS89x0_PLATFORM */
+
+MODULE_LICENSE("GPL");
+MODULE_DESCRIPTION("Crystal Semiconductor (Now Cirrus Logic) CS89[02]0 network driver");
+MODULE_AUTHOR("Russell Nelson <nelson@crynwr.com>");
diff --git a/drivers/net/ethernet/cirrus/cs89x0.h b/drivers/net/ethernet/cirrus/cs89x0.h
index 91423b70bb45..210f9ec9af4b 100644
--- a/drivers/net/ethernet/cirrus/cs89x0.h
+++ b/drivers/net/ethernet/cirrus/cs89x0.h
@@ -459,7 +459,3 @@
 #define PNP_CNF_INT 0x70
 #define PNP_CNF_DMA 0x74
 #define PNP_CNF_MEM 0x48
-
-#define BIT0 1
-#define BIT15 0x8000
-
diff --git a/drivers/net/ethernet/cirrus/ep93xx_eth.c b/drivers/net/ethernet/cirrus/ep93xx_eth.c
index 7a7c02f1f8b9..a4972457edd9 100644
--- a/drivers/net/ethernet/cirrus/ep93xx_eth.c
+++ b/drivers/net/ethernet/cirrus/ep93xx_eth.c
@@ -1,12 +1,8 @@
+// SPDX-License-Identifier: GPL-2.0-or-later
 /*
  * EP93xx ethernet network device driver
  * Copyright (C) 2006 Lennert Buytenhek <buytenh@wantstofly.org>
  * Dedicated to Marija Kulikova.
- *
- * This program is free software; you can redistribute it and/or modify
- * it under the terms of the GNU General Public License as published by
- * the Free Software Foundation; either version 2 of the License, or
- * (at your option) any later version.
  */
 
 #define pr_fmt(fmt) KBUILD_MODNAME ":%s: " fmt, __func__
@@ -20,15 +16,13 @@
 #include <linux/ethtool.h>
 #include <linux/interrupt.h>
 #include <linux/moduleparam.h>
+#include <linux/of.h>
 #include <linux/platform_device.h>
 #include <linux/delay.h>
 #include <linux/io.h>
 #include <linux/slab.h>
 
-#include <mach/hardware.h>
-
 #define DRV_MODULE_NAME		"ep93xx-eth"
-#define DRV_MODULE_VERSION	"0.1"
 
 #define RX_QUEUE_ENTRIES	64
 #define TX_QUEUE_ENTRIES	8
@@ -332,7 +326,7 @@ static int ep93xx_poll(struct napi_struct *napi, int budget)
 	return rx;
 }
 
-static int ep93xx_xmit(struct sk_buff *skb, struct net_device *dev)
+static netdev_tx_t ep93xx_xmit(struct sk_buff *skb, struct net_device *dev)
 {
 	struct ep93xx_priv *ep = netdev_priv(dev);
 	struct ep93xx_tdesc *txd;
@@ -694,15 +688,17 @@ static int ep93xx_ioctl(struct net_device *dev, struct ifreq *ifr, int cmd)
 
 static void ep93xx_get_drvinfo(struct net_device *dev, struct ethtool_drvinfo *info)
 {
-	strlcpy(info->driver, DRV_MODULE_NAME, sizeof(info->driver));
-	strlcpy(info->version, DRV_MODULE_VERSION, sizeof(info->version));
+	strscpy(info->driver, DRV_MODULE_NAME, sizeof(info->driver));
 }
 
 static int ep93xx_get_link_ksettings(struct net_device *dev,
 				     struct ethtool_link_ksettings *cmd)
 {
 	struct ep93xx_priv *ep = netdev_priv(dev);
-	return mii_ethtool_get_link_ksettings(&ep->mii, cmd);
+
+	mii_ethtool_get_link_ksettings(&ep->mii, cmd);
+
+	return 0;
 }
 
 static int ep93xx_set_link_ksettings(struct net_device *dev,
@@ -736,38 +732,20 @@ static const struct net_device_ops ep93xx_netdev_ops = {
 	.ndo_open		= ep93xx_open,
 	.ndo_stop		= ep93xx_close,
 	.ndo_start_xmit		= ep93xx_xmit,
-	.ndo_do_ioctl		= ep93xx_ioctl,
+	.ndo_eth_ioctl		= ep93xx_ioctl,
 	.ndo_validate_addr	= eth_validate_addr,
 	.ndo_set_mac_address	= eth_mac_addr,
 };
 
-static struct net_device *ep93xx_dev_alloc(struct ep93xx_eth_data *data)
-{
-	struct net_device *dev;
-
-	dev = alloc_etherdev(sizeof(struct ep93xx_priv));
-	if (dev == NULL)
-		return NULL;
-
-	memcpy(dev->dev_addr, data->dev_addr, ETH_ALEN);
-
-	dev->ethtool_ops = &ep93xx_ethtool_ops;
-	dev->netdev_ops = &ep93xx_netdev_ops;
-
-	dev->features |= NETIF_F_SG | NETIF_F_HW_CSUM;
-
-	return dev;
-}
-
-
-static int ep93xx_eth_remove(struct platform_device *pdev)
+static void ep93xx_eth_remove(struct platform_device *pdev)
 {
 	struct net_device *dev;
 	struct ep93xx_priv *ep;
+	struct resource *mem;
 
 	dev = platform_get_drvdata(pdev);
 	if (dev == NULL)
-		return 0;
+		return;
 
 	ep = netdev_priv(dev);
 
@@ -779,42 +757,62 @@ static int ep93xx_eth_remove(struct platform_device *pdev)
 		iounmap(ep->base_addr);
 
 	if (ep->res != NULL) {
-		release_resource(ep->res);
-		kfree(ep->res);
+		mem = platform_get_resource(pdev, IORESOURCE_MEM, 0);
+		release_mem_region(mem->start, resource_size(mem));
 	}
 
 	free_netdev(dev);
-
-	return 0;
 }
 
 static int ep93xx_eth_probe(struct platform_device *pdev)
 {
-	struct ep93xx_eth_data *data;
 	struct net_device *dev;
 	struct ep93xx_priv *ep;
 	struct resource *mem;
+	void __iomem *base_addr;
+	struct device_node *np;
+	u8 addr[ETH_ALEN];
+	u32 phy_id;
 	int irq;
 	int err;
 
 	if (pdev == NULL)
 		return -ENODEV;
-	data = dev_get_platdata(&pdev->dev);
 
 	mem = platform_get_resource(pdev, IORESOURCE_MEM, 0);
 	irq = platform_get_irq(pdev, 0);
 	if (!mem || irq < 0)
 		return -ENXIO;
 
-	dev = ep93xx_dev_alloc(data);
+	base_addr = ioremap(mem->start, resource_size(mem));
+	if (!base_addr)
+		return dev_err_probe(&pdev->dev, -EIO, "Failed to ioremap ethernet registers\n");
+
+	np = of_parse_phandle(pdev->dev.of_node, "phy-handle", 0);
+	if (!np)
+		return dev_err_probe(&pdev->dev, -ENODEV, "Please provide \"phy-handle\"\n");
+
+	err = of_property_read_u32(np, "reg", &phy_id);
+	of_node_put(np);
+	if (err)
+		return dev_err_probe(&pdev->dev, -ENOENT, "Failed to locate \"phy_id\"\n");
+
+	dev = alloc_etherdev(sizeof(struct ep93xx_priv));
 	if (dev == NULL) {
 		err = -ENOMEM;
 		goto err_out;
 	}
+
+	memcpy_fromio(addr, base_addr + 0x50, ETH_ALEN);
+	eth_hw_addr_set(dev, addr);
+	dev->ethtool_ops = &ep93xx_ethtool_ops;
+	dev->netdev_ops = &ep93xx_netdev_ops;
+	dev->features |= NETIF_F_SG | NETIF_F_HW_CSUM;
+
 	ep = netdev_priv(dev);
 	ep->dev = dev;
 	SET_NETDEV_DEV(dev, &pdev->dev);
-	netif_napi_add(dev, &ep->napi, ep93xx_poll, 64);
+	netif_napi_add(dev, &ep->napi, ep93xx_poll);
 
 	platform_set_drvdata(pdev, dev);
 
@@ -826,15 +824,10 @@ static int ep93xx_eth_probe(struct platform_device *pdev)
 		goto err_out;
 	}
 
-	ep->base_addr = ioremap(mem->start, resource_size(mem));
-	if (ep->base_addr == NULL) {
-		dev_err(&pdev->dev, "Failed to ioremap ethernet registers\n");
-		err = -EIO;
-		goto err_out;
-	}
+	ep->base_addr = base_addr;
 	ep->irq = irq;
 
-	ep->mii.phy_id = data->phy_id;
+	ep->mii.phy_id = phy_id;
 	ep->mii.phy_id_mask = 0x1f;
 	ep->mii.reg_num_mask = 0x1f;
 	ep->mii.dev = dev;
@@ -861,16 +854,23 @@ err_out:
 	return err;
 }
 
+static const struct of_device_id ep93xx_eth_of_ids[] = {
+	{ .compatible = "cirrus,ep9301-eth" },
+	{ /* sentinel */ }
+};
+MODULE_DEVICE_TABLE(of, ep93xx_eth_of_ids);
 
 static struct platform_driver ep93xx_eth_driver = {
 	.probe		= ep93xx_eth_probe,
 	.remove		= ep93xx_eth_remove,
 	.driver		= {
 		.name	= "ep93xx-eth",
+		.of_match_table = ep93xx_eth_of_ids,
 	},
 };
 
 module_platform_driver(ep93xx_eth_driver);
 
+MODULE_DESCRIPTION("Cirrus EP93xx Ethernet driver");
 MODULE_LICENSE("GPL");
 MODULE_ALIAS("platform:ep93xx-eth");
diff --git a/drivers/net/ethernet/cirrus/mac89x0.c b/drivers/net/ethernet/cirrus/mac89x0.c
index f910f0f386d6..6723df9b65d9 100644
--- a/drivers/net/ethernet/cirrus/mac89x0.c
+++ b/drivers/net/ethernet/cirrus/mac89x0.c
@@ -56,21 +56,11 @@
   local_irq_{dis,en}able()
 */
 
+#define pr_fmt(fmt) KBUILD_MODNAME ": " fmt
+
 static const char version[] =
 "cs89x0.c:v1.02 11/26/96 Russell Nelson <nelson@crynwr.com>\n";
 
-/* ======================= configure the driver here ======================= */
-
-/* use 0 for production, 1 for verification, >2 for debug */
-#ifndef NET_DEBUG
-#define NET_DEBUG 0
-#endif
-
-/* ======================= end of configuration ======================= */
-
-
-/* Always include 'config.h' first in case the user wants to turn on
-   or override something. */
 #include <linux/module.h>
 
 /*
@@ -93,6 +83,7 @@ static const char version[] =
 #include <linux/errno.h>
 #include <linux/init.h>
 #include <linux/netdevice.h>
+#include <linux/platform_device.h>
 #include <linux/etherdevice.h>
 #include <linux/skbuff.h>
 #include <linux/delay.h>
@@ -105,26 +96,24 @@ static const char version[] =
 
 #include "cs89x0.h"
 
-static unsigned int net_debug = NET_DEBUG;
+static int debug = -1;
+module_param(debug, int, 0);
+MODULE_PARM_DESC(debug, "debug message level");
 
 /* Information that need to be kept for each board. */
 struct net_local {
+	int msg_enable;
 	int chip_type;		/* one of: CS8900, CS8920, CS8920M */
 	char chip_revision;	/* revision letter of the chip ('A'...) */
 	int send_cmd;		/* the propercommand used to send a packet. */
 	int rx_mode;
 	int curr_rx_cfg;
         int send_underrun;      /* keep track of how many underruns in a row we get */
-	struct sk_buff *skb;
 };
 
 /* Index to functions, as function prototypes. */
-
-#if 0
-extern void reset_chip(struct net_device *dev);
-#endif
 static int net_open(struct net_device *dev);
-static int net_send_packet(struct sk_buff *skb, struct net_device *dev);
+static netdev_tx_t net_send_packet(struct sk_buff *skb, struct net_device *dev);
 static irqreturn_t net_interrupt(int irq, void *dev_id);
 static void set_multicast_list(struct net_device *dev);
 static void net_rx(struct net_device *dev);
@@ -132,10 +121,6 @@ static int net_close(struct net_device *dev);
 static struct net_device_stats *net_get_stats(struct net_device *dev);
 static int set_mac_address(struct net_device *dev, void *addr);
 
-
-/* Example routines you must write ;->. */
-#define tx_done(dev) 1
-
 /* For reading/writing registers ISA-style */
 static inline int
 readreg_io(struct net_device *dev, int portno)
@@ -176,39 +161,27 @@ static const struct net_device_ops mac89x0_netdev_ops = {
 
 /* Probe for the CS8900 card in slot E.  We won't bother looking
    anywhere else until we have a really good reason to do so. */
-struct net_device * __init mac89x0_probe(int unit)
+static int mac89x0_device_probe(struct platform_device *pdev)
 {
 	struct net_device *dev;
-	static int once_is_enough;
 	struct net_local *lp;
-	static unsigned version_printed;
 	int i, slot;
 	unsigned rev_type = 0;
 	unsigned long ioaddr;
 	unsigned short sig;
 	int err = -ENODEV;
-
-	if (!MACH_IS_MAC)
-		return ERR_PTR(-ENODEV);
+	struct nubus_rsrc *fres;
 
 	dev = alloc_etherdev(sizeof(struct net_local));
 	if (!dev)
-		return ERR_PTR(-ENOMEM);
-
-	if (unit >= 0) {
-		sprintf(dev->name, "eth%d", unit);
-		netdev_boot_setup_check(dev);
-	}
-
-	if (once_is_enough)
-		goto out;
-	once_is_enough = 1;
+		return -ENOMEM;
 
 	/* We might have to parameterize this later */
 	slot = 0xE;
 	/* Get out now if there's a real NuBus card in slot E */
-	if (nubus_find_slot(slot, NULL) != NULL)
-		goto out;
+	for_each_func_rsrc(fres)
+		if (fres->board->slot == slot)
+			goto out;
 
 	/* The pseudo-ISA bits always live at offset 0x300 (gee,
            wonder why...) */
@@ -228,9 +201,13 @@ struct net_device * __init mac89x0_probe(int unit)
 	if (sig != swab16(CHIP_EISA_ID_SIG))
 		goto out;
 
+	SET_NETDEV_DEV(dev, &pdev->dev);
+
 	/* Initialize the net_device structure. */
 	lp = netdev_priv(dev);
 
+	lp->msg_enable = netif_msg_init(debug, 0);
+
 	/* Fill in the 'dev' fields. */
 	dev->base_addr = ioaddr;
 	dev->mem_start = (unsigned long)
@@ -253,66 +230,50 @@ struct net_device * __init mac89x0_probe(int unit)
 	if (lp->chip_type != CS8900 && lp->chip_revision >= 'C')
 		lp->send_cmd = TX_NOW;
 
-	if (net_debug && version_printed++ == 0)
-		printk(version);
+	netif_dbg(lp, drv, dev, "%s", version);
 
-	printk(KERN_INFO "%s: cs89%c0%s rev %c found at %#8lx",
-	       dev->name,
-	       lp->chip_type==CS8900?'0':'2',
-	       lp->chip_type==CS8920M?"M":"",
-	       lp->chip_revision,
-	       dev->base_addr);
+	pr_info("cs89%c0%s rev %c found at %#8lx\n",
+		lp->chip_type == CS8900 ? '0' : '2',
+		lp->chip_type == CS8920M ? "M" : "",
+		lp->chip_revision, dev->base_addr);
 
 	/* Try to read the MAC address */
 	if ((readreg(dev, PP_SelfST) & (EEPROM_PRESENT | EEPROM_OK)) == 0) {
-		printk("\nmac89x0: No EEPROM, giving up now.\n");
+		pr_info("No EEPROM, giving up now.\n");
 		goto out1;
         } else {
+		u8 addr[ETH_ALEN];
+
                 for (i = 0; i < ETH_ALEN; i += 2) {
 			/* Big-endian (why??!) */
 			unsigned short s = readreg(dev, PP_IA + i);
-                        dev->dev_addr[i] = s >> 8;
-                        dev->dev_addr[i+1] = s & 0xff;
+			addr[i] = s >> 8;
+			addr[i+1] = s & 0xff;
                 }
+		eth_hw_addr_set(dev, addr);
         }
 
 	dev->irq = SLOT2IRQ(slot);
 
 	/* print the IRQ and ethernet address. */
 
-	printk(" IRQ %d ADDR %pM\n", dev->irq, dev->dev_addr);
+	pr_info("MAC %pM, IRQ %d\n", dev->dev_addr, dev->irq);
 
 	dev->netdev_ops		= &mac89x0_netdev_ops;
 
 	err = register_netdev(dev);
 	if (err)
 		goto out1;
-	return NULL;
+
+	platform_set_drvdata(pdev, dev);
+	return 0;
 out1:
 	nubus_writew(0, dev->base_addr + ADD_PORT);
 out:
 	free_netdev(dev);
-	return ERR_PTR(err);
+	return err;
 }
 
-#if 0
-/* This is useful for something, but I don't know what yet. */
-void __init reset_chip(struct net_device *dev)
-{
-	int reset_start_time;
-
-	writereg(dev, PP_SelfCTL, readreg(dev, PP_SelfCTL) | POWER_ON_RESET);
-
-	/* wait 30 ms */
-	msleep_interruptible(30);
-
-	/* Wait until the chip is reset */
-	reset_start_time = jiffies;
-	while( (readreg(dev, PP_SelfST) & INIT_DONE) == 0 && jiffies - reset_start_time < 2)
-		;
-}
-#endif
-
 /* Open/initialize the board.  This is called (in the current kernel)
    sometime after booting when the 'ifconfig' program is run.
 
@@ -366,17 +327,15 @@ net_open(struct net_device *dev)
 	return 0;
 }
 
-static int
+static netdev_tx_t
 net_send_packet(struct sk_buff *skb, struct net_device *dev)
 {
 	struct net_local *lp = netdev_priv(dev);
 	unsigned long flags;
 
-	if (net_debug > 3)
-		printk("%s: sent %d byte packet of type %x\n",
-		       dev->name, skb->len,
-		       (skb->data[ETH_ALEN+ETH_ALEN] << 8)
-		       | skb->data[ETH_ALEN+ETH_ALEN+1]);
+	netif_dbg(lp, tx_queued, dev, "sent %d byte packet of type %x\n",
+		  skb->len, skb->data[ETH_ALEN + ETH_ALEN] << 8 |
+		  skb->data[ETH_ALEN + ETH_ALEN + 1]);
 
 	/* keep the upload from being interrupted, since we
 	   ask the chip to start transmitting before the
@@ -414,11 +373,6 @@ static irqreturn_t net_interrupt(int irq, void *dev_id)
 	struct net_local *lp;
 	int ioaddr, status;
 
-	if (dev == NULL) {
-		printk ("net_interrupt(): irq %d for unknown device.\n", irq);
-		return IRQ_NONE;
-	}
-
 	ioaddr = dev->base_addr;
 	lp = netdev_priv(dev);
 
@@ -430,7 +384,7 @@ static irqreturn_t net_interrupt(int irq, void *dev_id)
            faster than you can read them off, you're screwed.  Hasta la
            vista, baby!  */
 	while ((status = swab16(nubus_readw(dev->base_addr + ISQ_PORT)))) {
-		if (net_debug > 4)printk("%s: event=%04x\n", dev->name, status);
+		netif_dbg(lp, intr, dev, "status=%04x\n", status);
 		switch(status & ISQ_EVENT_MASK) {
 		case ISQ_RECEIVER_EVENT:
 			/* Got a packet(s). */
@@ -460,7 +414,7 @@ static irqreturn_t net_interrupt(int irq, void *dev_id)
 				netif_wake_queue(dev);
 			}
 			if (status & TX_UNDERRUN) {
-				if (net_debug > 0) printk("%s: transmit underrun\n", dev->name);
+				netif_dbg(lp, tx_err, dev, "transmit underrun\n");
                                 lp->send_underrun++;
                                 if (lp->send_underrun == 3) lp->send_cmd = TX_AFTER_381;
                                 else if (lp->send_underrun == 6) lp->send_cmd = TX_AFTER_ALL;
@@ -481,6 +435,7 @@ static irqreturn_t net_interrupt(int irq, void *dev_id)
 static void
 net_rx(struct net_device *dev)
 {
+	struct net_local *lp = netdev_priv(dev);
 	struct sk_buff *skb;
 	int status, length;
 
@@ -504,7 +459,6 @@ net_rx(struct net_device *dev)
 	/* Malloc up new buffer. */
 	skb = alloc_skb(length, GFP_ATOMIC);
 	if (skb == NULL) {
-		printk("%s: Memory squeeze, dropping packet.\n", dev->name);
 		dev->stats.rx_dropped++;
 		return;
 	}
@@ -513,10 +467,9 @@ net_rx(struct net_device *dev)
 	skb_copy_to_linear_data(skb, (void *)(dev->mem_start + PP_RxFrame),
 				length);
 
-	if (net_debug > 3)printk("%s: received %d byte packet of type %x\n",
-                                 dev->name, length,
-                                 (skb->data[ETH_ALEN+ETH_ALEN] << 8)
-				 | skb->data[ETH_ALEN+ETH_ALEN+1]);
+	netif_dbg(lp, rx_status, dev, "received %d byte packet of type %x\n",
+		  length, skb->data[ETH_ALEN + ETH_ALEN] << 8 |
+		  skb->data[ETH_ALEN + ETH_ALEN + 1]);
 
         skb->protocol=eth_type_trans(skb,dev);
 	netif_rx(skb);
@@ -591,8 +544,8 @@ static int set_mac_address(struct net_device *dev, void *addr)
 	if (!is_valid_ether_addr(saddr->sa_data))
 		return -EADDRNOTAVAIL;
 
-	memcpy(dev->dev_addr, saddr->sa_data, ETH_ALEN);
-	printk("%s: Setting MAC address to %pM\n", dev->name, dev->dev_addr);
+	eth_hw_addr_set(dev, saddr->sa_data);
+	netdev_info(dev, "Setting MAC address to %pM\n", dev->dev_addr);
 
 	/* set the Ethernet address */
 	for (i=0; i < ETH_ALEN/2; i++)
@@ -601,32 +554,24 @@ static int set_mac_address(struct net_device *dev, void *addr)
 	return 0;
 }
 
-#ifdef MODULE
-
-static struct net_device *dev_cs89x0;
-static int debug;
-
-module_param(debug, int, 0);
-MODULE_PARM_DESC(debug, "CS89[02]0 debug level (0-5)");
+MODULE_DESCRIPTION("Macintosh CS89x0-based Ethernet driver");
 MODULE_LICENSE("GPL");
 
-int __init
-init_module(void)
+static void mac89x0_device_remove(struct platform_device *pdev)
 {
-	net_debug = debug;
-        dev_cs89x0 = mac89x0_probe(-1);
-	if (IS_ERR(dev_cs89x0)) {
-                printk(KERN_WARNING "mac89x0.c: No card found\n");
-		return PTR_ERR(dev_cs89x0);
-	}
-	return 0;
-}
+	struct net_device *dev = platform_get_drvdata(pdev);
 
-void
-cleanup_module(void)
-{
-	unregister_netdev(dev_cs89x0);
-	nubus_writew(0, dev_cs89x0->base_addr + ADD_PORT);
-	free_netdev(dev_cs89x0);
+	unregister_netdev(dev);
+	nubus_writew(0, dev->base_addr + ADD_PORT);
+	free_netdev(dev);
 }
-#endif /* MODULE */
+
+static struct platform_driver mac89x0_platform_driver = {
+	.probe = mac89x0_device_probe,
+	.remove = mac89x0_device_remove,
+	.driver = {
+		.name = "mac89x0",
+	},
+};
+
+module_platform_driver(mac89x0_platform_driver);
diff --git a/drivers/net/ethernet/cisco/Kconfig b/drivers/net/ethernet/cisco/Kconfig
index 15b713a89620..18c3a0718d6f 100644
--- a/drivers/net/ethernet/cisco/Kconfig
+++ b/drivers/net/ethernet/cisco/Kconfig
@@ -1,3 +1,4 @@
+# SPDX-License-Identifier: GPL-2.0-only
 #
 # Cisco device configuration
 #
@@ -6,7 +7,7 @@ config NET_VENDOR_CISCO
 	bool "Cisco devices"
 	default y
 	depends on PCI
-	---help---
+	help
 	  If you have a network (Ethernet) card belonging to this class, say Y.
 
 	  Note that the answer to this question doesn't directly affect the
diff --git a/drivers/net/ethernet/cisco/Makefile b/drivers/net/ethernet/cisco/Makefile
index 6c7437bc4a92..074635beec82 100644
--- a/drivers/net/ethernet/cisco/Makefile
+++ b/drivers/net/ethernet/cisco/Makefile
@@ -1,3 +1,4 @@
+# SPDX-License-Identifier: GPL-2.0-only
 #
 # Makefile for the Cisco device drivers.
 #
diff --git a/drivers/net/ethernet/cisco/enic/Kconfig b/drivers/net/ethernet/cisco/enic/Kconfig
index b63f8d8a4261..96709875fe4f 100644
--- a/drivers/net/ethernet/cisco/enic/Kconfig
+++ b/drivers/net/ethernet/cisco/enic/Kconfig
@@ -1,3 +1,4 @@
+# SPDX-License-Identifier: GPL-2.0-only
 #
 # Cisco device configuration
 #
@@ -5,5 +6,6 @@
 config ENIC
 	tristate "Cisco VIC Ethernet NIC Support"
 	depends on PCI
-	---help---
+	select PAGE_POOL
+	help
 	  This enables the support for the Cisco VIC Ethernet card.
diff --git a/drivers/net/ethernet/cisco/enic/Makefile b/drivers/net/ethernet/cisco/enic/Makefile
index aadcaf7876ce..a96b8332e6e2 100644
--- a/drivers/net/ethernet/cisco/enic/Makefile
+++ b/drivers/net/ethernet/cisco/enic/Makefile
@@ -1,6 +1,7 @@
+# SPDX-License-Identifier: GPL-2.0-only
 obj-$(CONFIG_ENIC) := enic.o
 
 enic-y := enic_main.o vnic_cq.o vnic_intr.o vnic_wq.o \
 	enic_res.o enic_dev.o enic_pp.o vnic_dev.o vnic_rq.o vnic_vic.o \
-	enic_ethtool.o enic_api.o enic_clsf.o
+	enic_ethtool.o enic_api.o enic_clsf.o enic_rq.o enic_wq.o
 
diff --git a/drivers/net/ethernet/cisco/enic/cq_desc.h b/drivers/net/ethernet/cisco/enic/cq_desc.h
index d6dd1b4edf6e..bfb3f14e89f5 100644
--- a/drivers/net/ethernet/cisco/enic/cq_desc.h
+++ b/drivers/net/ethernet/cisco/enic/cq_desc.h
@@ -1,20 +1,7 @@
+/* SPDX-License-Identifier: GPL-2.0-only */
 /*
  * Copyright 2008-2010 Cisco Systems, Inc.  All rights reserved.
  * Copyright 2007 Nuova Systems, Inc.  All rights reserved.
- *
- * This program is free software; you may redistribute it and/or modify
- * it under the terms of the GNU General Public License as published by
- * the Free Software Foundation; version 2 of the License.
- *
- * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND,
- * EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF
- * MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND
- * NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS
- * BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN
- * ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN
- * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
- * SOFTWARE.
- *
  */
 
 #ifndef _CQ_DESC_H_
@@ -53,28 +40,7 @@ struct cq_desc {
 #define CQ_DESC_COMP_NDX_BITS    12
 #define CQ_DESC_COMP_NDX_MASK    ((1 << CQ_DESC_COMP_NDX_BITS) - 1)
 
-static inline void cq_desc_dec(const struct cq_desc *desc_arg,
-	u8 *type, u8 *color, u16 *q_number, u16 *completed_index)
-{
-	const struct cq_desc *desc = desc_arg;
-	const u8 type_color = desc->type_color;
-
-	*color = (type_color >> CQ_DESC_COLOR_SHIFT) & CQ_DESC_COLOR_MASK;
-
-	/*
-	 * Make sure color bit is read from desc *before* other fields
-	 * are read from desc.  Hardware guarantees color bit is last
-	 * bit (byte) written.  Adding the rmb() prevents the compiler
-	 * and/or CPU from reordering the reads which would potentially
-	 * result in reading stale values.
-	 */
-
-	rmb();
-
-	*type = type_color & CQ_DESC_TYPE_MASK;
-	*q_number = le16_to_cpu(desc->q_number) & CQ_DESC_Q_NUM_MASK;
-	*completed_index = le16_to_cpu(desc->completed_index) &
-		CQ_DESC_COMP_NDX_MASK;
-}
+#define CQ_DESC_32_FI_MASK (BIT(0) | BIT(1))
+#define CQ_DESC_64_FI_MASK (BIT(0) | BIT(1))
 
 #endif /* _CQ_DESC_H_ */
diff --git a/drivers/net/ethernet/cisco/enic/cq_enet_desc.h b/drivers/net/ethernet/cisco/enic/cq_enet_desc.h
index ac37cacc6136..50787cff29db 100644
--- a/drivers/net/ethernet/cisco/enic/cq_enet_desc.h
+++ b/drivers/net/ethernet/cisco/enic/cq_enet_desc.h
@@ -1,20 +1,7 @@
+/* SPDX-License-Identifier: GPL-2.0-only */
 /*
  * Copyright 2008-2010 Cisco Systems, Inc.  All rights reserved.
  * Copyright 2007 Nuova Systems, Inc.  All rights reserved.
- *
- * This program is free software; you may redistribute it and/or modify
- * it under the terms of the GNU General Public License as published by
- * the Free Software Foundation; version 2 of the License.
- *
- * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND,
- * EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF
- * MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND
- * NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS
- * BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN
- * ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN
- * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
- * SOFTWARE.
- *
  */
 
 #ifndef _CQ_ENET_DESC_H_
@@ -30,12 +17,22 @@ struct cq_enet_wq_desc {
 	u8 type_color;
 };
 
-static inline void cq_enet_wq_desc_dec(struct cq_enet_wq_desc *desc,
-	u8 *type, u8 *color, u16 *q_number, u16 *completed_index)
-{
-	cq_desc_dec((struct cq_desc *)desc, type,
-		color, q_number, completed_index);
-}
+/*
+ * Defines and Capabilities for CMD_CQ_ENTRY_SIZE_SET
+ */
+#define VNIC_RQ_ALL (~0ULL)
+
+#define VNIC_RQ_CQ_ENTRY_SIZE_16 0
+#define VNIC_RQ_CQ_ENTRY_SIZE_32 1
+#define VNIC_RQ_CQ_ENTRY_SIZE_64 2
+
+#define VNIC_RQ_CQ_ENTRY_SIZE_16_CAPABLE BIT(VNIC_RQ_CQ_ENTRY_SIZE_16)
+#define VNIC_RQ_CQ_ENTRY_SIZE_32_CAPABLE BIT(VNIC_RQ_CQ_ENTRY_SIZE_32)
+#define VNIC_RQ_CQ_ENTRY_SIZE_64_CAPABLE BIT(VNIC_RQ_CQ_ENTRY_SIZE_64)
+
+#define VNIC_RQ_CQ_ENTRY_SIZE_ALL_BIT  (VNIC_RQ_CQ_ENTRY_SIZE_16_CAPABLE | \
+					VNIC_RQ_CQ_ENTRY_SIZE_32_CAPABLE | \
+					VNIC_RQ_CQ_ENTRY_SIZE_64_CAPABLE)
 
 /* Completion queue descriptor: Ethernet receive queue, 16B */
 struct cq_enet_rq_desc {
@@ -49,6 +46,45 @@ struct cq_enet_rq_desc {
 	u8 type_color;
 };
 
+/* Completion queue descriptor: Ethernet receive queue, 32B */
+struct cq_enet_rq_desc_32 {
+	__le16 completed_index_flags;
+	__le16 q_number_rss_type_flags;
+	__le32 rss_hash;
+	__le16 bytes_written_flags;
+	__le16 vlan;
+	__le16 checksum_fcoe;
+	u8 flags;
+	u8 fetch_index_flags;
+	__le32 time_stamp;
+	__le16 time_stamp2;
+	__le16 pie_info;
+	__le32 pie_info2;
+	__le16 pie_info3;
+	u8 pie_info4;
+	u8 type_color;
+};
+
+/* Completion queue descriptor: Ethernet receive queue, 64B */
+struct cq_enet_rq_desc_64 {
+	__le16 completed_index_flags;
+	__le16 q_number_rss_type_flags;
+	__le32 rss_hash;
+	__le16 bytes_written_flags;
+	__le16 vlan;
+	__le16 checksum_fcoe;
+	u8 flags;
+	u8 fetch_index_flags;
+	__le32 time_stamp;
+	__le16 time_stamp2;
+	__le16 pie_info;
+	__le32 pie_info2;
+	__le16 pie_info3;
+	u8 pie_info4;
+	u8 reserved[32];
+	u8 type_color;
+};
+
 #define CQ_ENET_RQ_DESC_FLAGS_INGRESS_PORT          (0x1 << 12)
 #define CQ_ENET_RQ_DESC_FLAGS_FCOE                  (0x1 << 13)
 #define CQ_ENET_RQ_DESC_FLAGS_EOP                   (0x1 << 14)
@@ -101,85 +137,4 @@ struct cq_enet_rq_desc {
 #define CQ_ENET_RQ_DESC_FLAGS_IPV4_FRAGMENT         (0x1 << 6)
 #define CQ_ENET_RQ_DESC_FLAGS_FCS_OK                (0x1 << 7)
 
-static inline void cq_enet_rq_desc_dec(struct cq_enet_rq_desc *desc,
-	u8 *type, u8 *color, u16 *q_number, u16 *completed_index,
-	u8 *ingress_port, u8 *fcoe, u8 *eop, u8 *sop, u8 *rss_type,
-	u8 *csum_not_calc, u32 *rss_hash, u16 *bytes_written, u8 *packet_error,
-	u8 *vlan_stripped, u16 *vlan_tci, u16 *checksum, u8 *fcoe_sof,
-	u8 *fcoe_fc_crc_ok, u8 *fcoe_enc_error, u8 *fcoe_eof,
-	u8 *tcp_udp_csum_ok, u8 *udp, u8 *tcp, u8 *ipv4_csum_ok,
-	u8 *ipv6, u8 *ipv4, u8 *ipv4_fragment, u8 *fcs_ok)
-{
-	u16 completed_index_flags;
-	u16 q_number_rss_type_flags;
-	u16 bytes_written_flags;
-
-	cq_desc_dec((struct cq_desc *)desc, type,
-		color, q_number, completed_index);
-
-	completed_index_flags = le16_to_cpu(desc->completed_index_flags);
-	q_number_rss_type_flags =
-		le16_to_cpu(desc->q_number_rss_type_flags);
-	bytes_written_flags = le16_to_cpu(desc->bytes_written_flags);
-
-	*ingress_port = (completed_index_flags &
-		CQ_ENET_RQ_DESC_FLAGS_INGRESS_PORT) ? 1 : 0;
-	*fcoe = (completed_index_flags & CQ_ENET_RQ_DESC_FLAGS_FCOE) ?
-		1 : 0;
-	*eop = (completed_index_flags & CQ_ENET_RQ_DESC_FLAGS_EOP) ?
-		1 : 0;
-	*sop = (completed_index_flags & CQ_ENET_RQ_DESC_FLAGS_SOP) ?
-		1 : 0;
-
-	*rss_type = (u8)((q_number_rss_type_flags >> CQ_DESC_Q_NUM_BITS) &
-		CQ_ENET_RQ_DESC_RSS_TYPE_MASK);
-	*csum_not_calc = (q_number_rss_type_flags &
-		CQ_ENET_RQ_DESC_FLAGS_CSUM_NOT_CALC) ? 1 : 0;
-
-	*rss_hash = le32_to_cpu(desc->rss_hash);
-
-	*bytes_written = bytes_written_flags &
-		CQ_ENET_RQ_DESC_BYTES_WRITTEN_MASK;
-	*packet_error = (bytes_written_flags &
-		CQ_ENET_RQ_DESC_FLAGS_TRUNCATED) ? 1 : 0;
-	*vlan_stripped = (bytes_written_flags &
-		CQ_ENET_RQ_DESC_FLAGS_VLAN_STRIPPED) ? 1 : 0;
-
-	/*
-	 * Tag Control Information(16) = user_priority(3) + cfi(1) + vlan(12)
-	 */
-	*vlan_tci = le16_to_cpu(desc->vlan);
-
-	if (*fcoe) {
-		*fcoe_sof = (u8)(le16_to_cpu(desc->checksum_fcoe) &
-			CQ_ENET_RQ_DESC_FCOE_SOF_MASK);
-		*fcoe_fc_crc_ok = (desc->flags &
-			CQ_ENET_RQ_DESC_FCOE_FC_CRC_OK) ? 1 : 0;
-		*fcoe_enc_error = (desc->flags &
-			CQ_ENET_RQ_DESC_FCOE_ENC_ERROR) ? 1 : 0;
-		*fcoe_eof = (u8)((le16_to_cpu(desc->checksum_fcoe) >>
-			CQ_ENET_RQ_DESC_FCOE_EOF_SHIFT) &
-			CQ_ENET_RQ_DESC_FCOE_EOF_MASK);
-		*checksum = 0;
-	} else {
-		*fcoe_sof = 0;
-		*fcoe_fc_crc_ok = 0;
-		*fcoe_enc_error = 0;
-		*fcoe_eof = 0;
-		*checksum = le16_to_cpu(desc->checksum_fcoe);
-	}
-
-	*tcp_udp_csum_ok =
-		(desc->flags & CQ_ENET_RQ_DESC_FLAGS_TCP_UDP_CSUM_OK) ? 1 : 0;
-	*udp = (desc->flags & CQ_ENET_RQ_DESC_FLAGS_UDP) ? 1 : 0;
-	*tcp = (desc->flags & CQ_ENET_RQ_DESC_FLAGS_TCP) ? 1 : 0;
-	*ipv4_csum_ok =
-		(desc->flags & CQ_ENET_RQ_DESC_FLAGS_IPV4_CSUM_OK) ? 1 : 0;
-	*ipv6 = (desc->flags & CQ_ENET_RQ_DESC_FLAGS_IPV6) ? 1 : 0;
-	*ipv4 = (desc->flags & CQ_ENET_RQ_DESC_FLAGS_IPV4) ? 1 : 0;
-	*ipv4_fragment =
-		(desc->flags & CQ_ENET_RQ_DESC_FLAGS_IPV4_FRAGMENT) ? 1 : 0;
-	*fcs_ok = (desc->flags & CQ_ENET_RQ_DESC_FLAGS_FCS_OK) ? 1 : 0;
-}
-
 #endif /* _CQ_ENET_DESC_H_ */
diff --git a/drivers/net/ethernet/cisco/enic/enic.h b/drivers/net/ethernet/cisco/enic/enic.h
index 2b23f46b34d3..301b3f3114af 100644
--- a/drivers/net/ethernet/cisco/enic/enic.h
+++ b/drivers/net/ethernet/cisco/enic/enic.h
@@ -1,20 +1,7 @@
+/* SPDX-License-Identifier: GPL-2.0-only */
 /*
  * Copyright 2008-2010 Cisco Systems, Inc.  All rights reserved.
  * Copyright 2007 Nuova Systems, Inc.  All rights reserved.
- *
- * This program is free software; you may redistribute it and/or modify
- * it under the terms of the GNU General Public License as published by
- * the Free Software Foundation; version 2 of the License.
- *
- * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND,
- * EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF
- * MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND
- * NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS
- * BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN
- * ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN
- * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
- * SOFTWARE.
- *
  */
 
 #ifndef _ENIC_H_
@@ -30,24 +17,31 @@
 #include "vnic_nic.h"
 #include "vnic_rss.h"
 #include <linux/irq.h>
+#include <net/page_pool/helpers.h>
 
 #define DRV_NAME		"enic"
 #define DRV_DESCRIPTION		"Cisco VIC Ethernet NIC Driver"
-#define DRV_VERSION		"2.3.0.31"
-#define DRV_COPYRIGHT		"Copyright 2008-2013 Cisco Systems, Inc"
 
 #define ENIC_BARS_MAX		6
 
-#define ENIC_WQ_MAX		8
-#define ENIC_RQ_MAX		8
-#define ENIC_CQ_MAX		(ENIC_WQ_MAX + ENIC_RQ_MAX)
-#define ENIC_INTR_MAX		(ENIC_CQ_MAX + 2)
+#define ENIC_WQ_MAX		256
+#define ENIC_RQ_MAX		256
+#define ENIC_RQ_MIN_DEFAULT	8
+
+#define ENIC_WQ_NAPI_BUDGET	256
 
 #define ENIC_AIC_LARGE_PKT_DIFF	3
 
+enum ext_cq {
+	ENIC_RQ_CQ_ENTRY_SIZE_16,
+	ENIC_RQ_CQ_ENTRY_SIZE_32,
+	ENIC_RQ_CQ_ENTRY_SIZE_64,
+	ENIC_RQ_CQ_ENTRY_SIZE_MAX,
+};
+
 struct enic_msix_entry {
 	int requested;
-	char devname[IFNAMSIZ];
+	char devname[IFNAMSIZ + 8];
 	irqreturn_t (*isr)(int, void *);
 	void *devid;
 	cpumask_var_t affinity_mask;
@@ -90,6 +84,10 @@ struct enic_rx_coal {
 #define ENIC_SET_INSTANCE		(1 << 3)
 #define ENIC_SET_HOST			(1 << 4)
 
+#define MAX_TSO			BIT(16)
+#define WQ_ENET_MAX_DESC_LEN	BIT(WQ_ENET_LEN_BITS)
+#define ENIC_DESC_MAX_SPLITS	(MAX_TSO / WQ_ENET_MAX_DESC_LEN + 1)
+
 struct enic_port_profile {
 	u32 set;
 	u8 request;
@@ -138,8 +136,56 @@ struct enic_rfs_flw_tbl {
 struct vxlan_offload {
 	u16 vxlan_udp_port_number;
 	u8 patch_level;
+	u8 flags;
+};
+
+struct enic_wq_stats {
+	u64 packets;		/* pkts queued for Tx */
+	u64 stopped;		/* Tx ring almost full, queue stopped */
+	u64 wake;		/* Tx ring no longer full, queue woken up*/
+	u64 tso;		/* non-encap tso pkt */
+	u64 encap_tso;		/* encap tso pkt */
+	u64 encap_csum;		/* encap HW csum */
+	u64 csum_partial;	/* skb->ip_summed = CHECKSUM_PARTIAL */
+	u64 csum_none;		/* HW csum not required */
+	u64 bytes;		/* bytes queued for Tx */
+	u64 add_vlan;		/* HW adds vlan tag */
+	u64 cq_work;		/* Tx completions processed */
+	u64 cq_bytes;		/* Tx bytes processed */
+	u64 null_pkt;		/* skb length <= 0 */
+	u64 skb_linear_fail;	/* linearize failures */
+	u64 desc_full_awake;	/* TX ring full while queue awake */
 };
 
+struct enic_rq_stats {
+	u64 packets;			/* pkts received */
+	u64 bytes;			/* bytes received */
+	u64 l4_rss_hash;		/* hashed on l4 */
+	u64 l3_rss_hash;		/* hashed on l3 */
+	u64 csum_unnecessary;		/* HW verified csum */
+	u64 csum_unnecessary_encap;	/* HW verified csum on encap packet */
+	u64 vlan_stripped;		/* HW stripped vlan */
+	u64 napi_complete;		/* napi complete intr reenabled */
+	u64 napi_repoll;		/* napi poll again */
+	u64 bad_fcs;			/* bad pkts */
+	u64 pkt_truncated;		/* truncated pkts */
+	u64 no_skb;			/* out of skbs */
+	u64 desc_skip;			/* Rx pkt went into later buffer */
+	u64 pp_alloc_fail;		/* page pool alloc failure */
+};
+
+struct enic_wq {
+	spinlock_t lock;		/* spinlock for wq */
+	struct vnic_wq vwq;
+	struct enic_wq_stats stats;
+} ____cacheline_aligned;
+
+struct enic_rq {
+	struct vnic_rq vrq;
+	struct enic_rq_stats stats;
+	struct page_pool *pool;
+} ____cacheline_aligned;
+
 /* Per-instance private data structure */
 struct enic {
 	struct net_device *netdev;
@@ -151,8 +197,8 @@ struct enic {
 	struct work_struct reset;
 	struct work_struct tx_hang_reset;
 	struct work_struct change_mtu_work;
-	struct msix_entry msix_entry[ENIC_INTR_MAX];
-	struct enic_msix_entry msix[ENIC_INTR_MAX];
+	struct msix_entry *msix_entry;
+	struct enic_msix_entry *msix;
 	u32 msg_enable;
 	spinlock_t devcmd_lock;
 	u8 mac_addr[ETH_ALEN];
@@ -168,35 +214,33 @@ struct enic {
 	u16 num_vfs;
 #endif
 	spinlock_t enic_api_lock;
+	bool enic_api_busy;
 	struct enic_port_profile *pp;
 
-	/* work queue cache line section */
-	____cacheline_aligned struct vnic_wq wq[ENIC_WQ_MAX];
-	spinlock_t wq_lock[ENIC_WQ_MAX];
+	struct enic_wq *wq;
+	unsigned int wq_avail;
 	unsigned int wq_count;
 	u16 loop_enable;
 	u16 loop_tag;
 
-	/* receive queue cache line section */
-	____cacheline_aligned struct vnic_rq rq[ENIC_RQ_MAX];
+	struct enic_rq *rq;
+	unsigned int rq_avail;
 	unsigned int rq_count;
 	struct vxlan_offload vxlan;
-	u64 rq_truncated_pkts;
-	u64 rq_bad_fcs;
-	struct napi_struct napi[ENIC_RQ_MAX + ENIC_WQ_MAX];
+	struct napi_struct *napi;
 
-	/* interrupt resource cache line section */
-	____cacheline_aligned struct vnic_intr intr[ENIC_INTR_MAX];
+	struct vnic_intr *intr;
+	unsigned int intr_avail;
 	unsigned int intr_count;
 	u32 __iomem *legacy_pba;		/* memory-mapped */
 
-	/* completion queue cache line section */
-	____cacheline_aligned struct vnic_cq cq[ENIC_CQ_MAX];
+	struct vnic_cq *cq;
+	unsigned int cq_avail;
 	unsigned int cq_count;
 	struct enic_rfs_flw_tbl rfs_h;
-	u32 rx_copybreak;
 	u8 rss_key[ENIC_RSS_LEN];
 	struct vnic_gen_stats gen_stats;
+	enum ext_cq ext_cq;
 };
 
 static inline struct net_device *vnic_get_netdev(struct vnic_dev *vdev)
@@ -237,21 +281,6 @@ static inline unsigned int enic_cq_wq(struct enic *enic, unsigned int wq)
 	return enic->rq_count + wq;
 }
 
-static inline unsigned int enic_legacy_io_intr(void)
-{
-	return 0;
-}
-
-static inline unsigned int enic_legacy_err_intr(void)
-{
-	return 1;
-}
-
-static inline unsigned int enic_legacy_notify_intr(void)
-{
-	return 2;
-}
-
 static inline unsigned int enic_msix_rq_intr(struct enic *enic,
 	unsigned int rq)
 {
@@ -264,21 +293,35 @@ static inline unsigned int enic_msix_wq_intr(struct enic *enic,
 	return enic->cq[enic_cq_wq(enic, wq)].interrupt_offset;
 }
 
+/* MSIX interrupts are organized as the error interrupt, then the notify
+ * interrupt followed by all the I/O interrupts.  The error interrupt needs
+ * to fit in 7 bits due to hardware constraints
+ */
+#define ENIC_MSIX_RESERVED_INTR 2
+#define ENIC_MSIX_ERR_INTR	0
+#define ENIC_MSIX_NOTIFY_INTR	1
+#define ENIC_MSIX_IO_INTR_BASE	ENIC_MSIX_RESERVED_INTR
+#define ENIC_MSIX_MIN_INTR	(ENIC_MSIX_RESERVED_INTR + 2)
+
+#define ENIC_LEGACY_IO_INTR	0
+#define ENIC_LEGACY_ERR_INTR	1
+#define ENIC_LEGACY_NOTIFY_INTR	2
+
 static inline unsigned int enic_msix_err_intr(struct enic *enic)
 {
-	return enic->rq_count + enic->wq_count;
+	return ENIC_MSIX_ERR_INTR;
 }
 
 static inline unsigned int enic_msix_notify_intr(struct enic *enic)
 {
-	return enic->rq_count + enic->wq_count + 1;
+	return ENIC_MSIX_NOTIFY_INTR;
 }
 
 static inline bool enic_is_err_intr(struct enic *enic, int intr)
 {
 	switch (vnic_dev_get_intr_mode(enic->vdev)) {
 	case VNIC_DEV_INTR_MODE_INTX:
-		return intr == enic_legacy_err_intr();
+		return intr == ENIC_LEGACY_ERR_INTR;
 	case VNIC_DEV_INTR_MODE_MSIX:
 		return intr == enic_msix_err_intr(enic);
 	case VNIC_DEV_INTR_MODE_MSI:
@@ -291,7 +334,7 @@ static inline bool enic_is_notify_intr(struct enic *enic, int intr)
 {
 	switch (vnic_dev_get_intr_mode(enic->vdev)) {
 	case VNIC_DEV_INTR_MODE_INTX:
-		return intr == enic_legacy_notify_intr();
+		return intr == ENIC_LEGACY_NOTIFY_INTR;
 	case VNIC_DEV_INTR_MODE_MSIX:
 		return intr == enic_msix_notify_intr(enic);
 	case VNIC_DEV_INTR_MODE_MSI:
@@ -302,7 +345,7 @@ static inline bool enic_is_notify_intr(struct enic *enic, int intr)
 
 static inline int enic_dma_map_check(struct enic *enic, dma_addr_t dma_addr)
 {
-	if (unlikely(pci_dma_mapping_error(enic->pdev, dma_addr))) {
+	if (unlikely(dma_mapping_error(&enic->pdev->dev, dma_addr))) {
 		net_warn_ratelimited("%s: PCI dma mapping failed!\n",
 				     enic->netdev->name);
 		enic->gen_stats.dma_map_error++;
@@ -319,5 +362,6 @@ int enic_is_valid_vf(struct enic *enic, int vf);
 int enic_is_dynamic(struct enic *enic);
 void enic_set_ethtool_ops(struct net_device *netdev);
 int __enic_set_rsskey(struct enic *enic);
+void enic_ext_cq(struct enic *enic);
 
 #endif /* _ENIC_H_ */
diff --git a/drivers/net/ethernet/cisco/enic/enic_api.c b/drivers/net/ethernet/cisco/enic/enic_api.c
index b161f24522b8..e3b700c28bc4 100644
--- a/drivers/net/ethernet/cisco/enic/enic_api.c
+++ b/drivers/net/ethernet/cisco/enic/enic_api.c
@@ -1,20 +1,5 @@
-/**
- * Copyright 2013 Cisco Systems, Inc.  All rights reserved.
- *
- * This program is free software; you may redistribute it and/or modify
- * it under the terms of the GNU General Public License as published by
- * the Free Software Foundation; version 2 of the License.
- *
- * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND,
- * EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF
- * MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND
- * NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS
- * BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN
- * ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN
- * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
- * SOFTWARE.
- *
- */
+// SPDX-License-Identifier: GPL-2.0-only
+// Copyright 2013 Cisco Systems, Inc.  All rights reserved.
 
 #include <linux/netdevice.h>
 #include <linux/spinlock.h>
@@ -34,6 +19,12 @@ int enic_api_devcmd_proxy_by_index(struct net_device *netdev, int vf,
 	struct vnic_dev *vdev = enic->vdev;
 
 	spin_lock(&enic->enic_api_lock);
+	while (enic->enic_api_busy) {
+		spin_unlock(&enic->enic_api_lock);
+		cpu_relax();
+		spin_lock(&enic->enic_api_lock);
+	}
+
 	spin_lock_bh(&enic->devcmd_lock);
 
 	vnic_dev_cmd_proxy_by_index_start(vdev, vf);
diff --git a/drivers/net/ethernet/cisco/enic/enic_api.h b/drivers/net/ethernet/cisco/enic/enic_api.h
index 6b9f9255af28..e01790fb0415 100644
--- a/drivers/net/ethernet/cisco/enic/enic_api.h
+++ b/drivers/net/ethernet/cisco/enic/enic_api.h
@@ -1,20 +1,5 @@
-/**
- * Copyright 2013 Cisco Systems, Inc.  All rights reserved.
- *
- * This program is free software; you may redistribute it and/or modify
- * it under the terms of the GNU General Public License as published by
- * the Free Software Foundation; version 2 of the License.
- *
- * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND,
- * EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF
- * MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND
- * NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS
- * BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN
- * ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN
- * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
- * SOFTWARE.
- *
- */
+/* SPDX-License-Identifier: GPL-2.0-only */
+/* Copyright 2013 Cisco Systems, Inc.  All rights reserved. */
 
 #ifndef __ENIC_API_H__
 #define __ENIC_API_H__
diff --git a/drivers/net/ethernet/cisco/enic/enic_clsf.c b/drivers/net/ethernet/cisco/enic/enic_clsf.c
index 3c677ed3c29e..837f954873ee 100644
--- a/drivers/net/ethernet/cisco/enic/enic_clsf.c
+++ b/drivers/net/ethernet/cisco/enic/enic_clsf.c
@@ -1,3 +1,4 @@
+// SPDX-License-Identifier: GPL-2.0
 #include <linux/if.h>
 #include <linux/if_ether.h>
 #include <linux/if_link.h>
@@ -31,7 +32,8 @@ int enic_addfltr_5t(struct enic *enic, struct flow_keys *keys, u16 rq)
 		break;
 	default:
 		return -EPROTONOSUPPORT;
-	};
+	}
+
 	data.type = FILTER_IPV4_5TUPLE;
 	data.u.ipv4.src_addr = ntohl(keys->addrs.v4addrs.src);
 	data.u.ipv4.dst_addr = ntohl(keys->addrs.v4addrs.dst);
@@ -78,7 +80,6 @@ void enic_rfs_flw_tbl_init(struct enic *enic)
 	enic->rfs_h.max = enic->config.num_arfs;
 	enic->rfs_h.free = enic->rfs_h.max;
 	enic->rfs_h.toclean = 0;
-	enic_rfs_timer_start(enic);
 }
 
 void enic_rfs_flw_tbl_free(struct enic *enic)
@@ -87,7 +88,6 @@ void enic_rfs_flw_tbl_free(struct enic *enic)
 
 	enic_rfs_timer_stop(enic);
 	spin_lock_bh(&enic->rfs_h.lock);
-	enic->rfs_h.free = 0;
 	for (i = 0; i < (1 << ENIC_RFS_FLW_BITSHIFT); i++) {
 		struct hlist_head *hhead;
 		struct hlist_node *tmp;
@@ -98,6 +98,7 @@ void enic_rfs_flw_tbl_free(struct enic *enic)
 			enic_delfltr(enic, n->fltr_id);
 			hlist_del(&n->node);
 			kfree(n);
+			enic->rfs_h.free++;
 		}
 	}
 	spin_unlock_bh(&enic->rfs_h.lock);
@@ -122,9 +123,9 @@ struct enic_rfs_fltr_node *htbl_fltr_search(struct enic *enic, u16 fltr_id)
 }
 
 #ifdef CONFIG_RFS_ACCEL
-void enic_flow_may_expire(unsigned long data)
+void enic_flow_may_expire(struct timer_list *t)
 {
-	struct enic *enic = (struct enic *)data;
+	struct enic *enic = timer_container_of(enic, t, rfs_h.rfs_may_expire);
 	bool res;
 	int j;
 
diff --git a/drivers/net/ethernet/cisco/enic/enic_clsf.h b/drivers/net/ethernet/cisco/enic/enic_clsf.h
index 6aa9f89d073b..5f5284102fb0 100644
--- a/drivers/net/ethernet/cisco/enic/enic_clsf.h
+++ b/drivers/net/ethernet/cisco/enic/enic_clsf.h
@@ -1,3 +1,4 @@
+/* SPDX-License-Identifier: GPL-2.0 */
 #ifndef _ENIC_CLSF_H_
 #define _ENIC_CLSF_H_
 
@@ -15,19 +16,17 @@ struct enic_rfs_fltr_node *htbl_fltr_search(struct enic *enic, u16 fltr_id);
 #ifdef CONFIG_RFS_ACCEL
 int enic_rx_flow_steer(struct net_device *dev, const struct sk_buff *skb,
 		       u16 rxq_index, u32 flow_id);
-void enic_flow_may_expire(unsigned long data);
+void enic_flow_may_expire(struct timer_list *t);
 
 static inline void enic_rfs_timer_start(struct enic *enic)
 {
-	init_timer(&enic->rfs_h.rfs_may_expire);
-	enic->rfs_h.rfs_may_expire.function = enic_flow_may_expire;
-	enic->rfs_h.rfs_may_expire.data = (unsigned long)enic;
+	timer_setup(&enic->rfs_h.rfs_may_expire, enic_flow_may_expire, 0);
 	mod_timer(&enic->rfs_h.rfs_may_expire, jiffies + HZ/4);
 }
 
 static inline void enic_rfs_timer_stop(struct enic *enic)
 {
-	del_timer_sync(&enic->rfs_h.rfs_may_expire);
+	timer_delete_sync(&enic->rfs_h.rfs_may_expire);
 }
 #else
 static inline void enic_rfs_timer_start(struct enic *enic) {}
diff --git a/drivers/net/ethernet/cisco/enic/enic_dev.c b/drivers/net/ethernet/cisco/enic/enic_dev.c
index f8d2a6a34282..2cbae7c6cc3d 100644
--- a/drivers/net/ethernet/cisco/enic/enic_dev.c
+++ b/drivers/net/ethernet/cisco/enic/enic_dev.c
@@ -1,20 +1,5 @@
-/*
- * Copyright 2011 Cisco Systems, Inc.  All rights reserved.
- *
- * This program is free software; you may redistribute it and/or modify
- * it under the terms of the GNU General Public License as published by
- * the Free Software Foundation; version 2 of the License.
- *
- * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND,
- * EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF
- * MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND
- * NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS
- * BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN
- * ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN
- * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
- * SOFTWARE.
- *
- */
+// SPDX-License-Identifier: GPL-2.0-only
+// Copyright 2011 Cisco Systems, Inc.  All rights reserved.
 
 #include <linux/pci.h>
 #include <linux/etherdevice.h>
diff --git a/drivers/net/ethernet/cisco/enic/enic_dev.h b/drivers/net/ethernet/cisco/enic/enic_dev.h
index f5bb058b3f96..698d0cb02064 100644
--- a/drivers/net/ethernet/cisco/enic/enic_dev.h
+++ b/drivers/net/ethernet/cisco/enic/enic_dev.h
@@ -1,20 +1,5 @@
-/*
- * Copyright 2011 Cisco Systems, Inc.  All rights reserved.
- *
- * This program is free software; you may redistribute it and/or modify
- * it under the terms of the GNU General Public License as published by
- * the Free Software Foundation; version 2 of the License.
- *
- * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND,
- * EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF
- * MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND
- * NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS
- * BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN
- * ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN
- * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
- * SOFTWARE.
- *
- */
+/* SPDX-License-Identifier: GPL-2.0-only */
+/* Copyright 2011 Cisco Systems, Inc.  All rights reserved. */
 
 #ifndef _ENIC_DEV_H_
 #define _ENIC_DEV_H_
diff --git a/drivers/net/ethernet/cisco/enic/enic_ethtool.c b/drivers/net/ethernet/cisco/enic/enic_ethtool.c
index fd3980cc1e34..a50f5dad34d5 100644
--- a/drivers/net/ethernet/cisco/enic/enic_ethtool.c
+++ b/drivers/net/ethernet/cisco/enic/enic_ethtool.c
@@ -1,23 +1,9 @@
-/**
- * Copyright 2013 Cisco Systems, Inc.  All rights reserved.
- *
- * This program is free software; you may redistribute it and/or modify
- * it under the terms of the GNU General Public License as published by
- * the Free Software Foundation; version 2 of the License.
- *
- * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND,
- * EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF
- * MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND
- * NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS
- * BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN
- * ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN
- * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
- * SOFTWARE.
- *
- */
+// SPDX-License-Identifier: GPL-2.0-only
+// Copyright 2013 Cisco Systems, Inc.  All rights reserved.
 
 #include <linux/netdevice.h>
 #include <linux/ethtool.h>
+#include <linux/net_tstamp.h>
 
 #include "enic_res.h"
 #include "enic.h"
@@ -46,6 +32,41 @@ struct enic_stat {
 	.index = offsetof(struct vnic_gen_stats, stat) / sizeof(u64)\
 }
 
+#define ENIC_PER_RQ_STAT(stat) { \
+	.name = "rq[%d]_"#stat, \
+	.index = offsetof(struct enic_rq_stats, stat) / sizeof(u64) \
+}
+
+#define ENIC_PER_WQ_STAT(stat) { \
+	.name = "wq[%d]_"#stat, \
+	.index = offsetof(struct enic_wq_stats, stat) / sizeof(u64) \
+}
+
+static const struct enic_stat enic_per_rq_stats[] = {
+	ENIC_PER_RQ_STAT(l4_rss_hash),
+	ENIC_PER_RQ_STAT(l3_rss_hash),
+	ENIC_PER_RQ_STAT(csum_unnecessary_encap),
+	ENIC_PER_RQ_STAT(vlan_stripped),
+	ENIC_PER_RQ_STAT(napi_complete),
+	ENIC_PER_RQ_STAT(napi_repoll),
+	ENIC_PER_RQ_STAT(no_skb),
+	ENIC_PER_RQ_STAT(desc_skip),
+};
+
+#define NUM_ENIC_PER_RQ_STATS   ARRAY_SIZE(enic_per_rq_stats)
+
+static const struct enic_stat enic_per_wq_stats[] = {
+	ENIC_PER_WQ_STAT(encap_tso),
+	ENIC_PER_WQ_STAT(encap_csum),
+	ENIC_PER_WQ_STAT(add_vlan),
+	ENIC_PER_WQ_STAT(cq_work),
+	ENIC_PER_WQ_STAT(cq_bytes),
+	ENIC_PER_WQ_STAT(null_pkt),
+	ENIC_PER_WQ_STAT(skb_linear_fail),
+	ENIC_PER_WQ_STAT(desc_full_awake),
+};
+
+#define NUM_ENIC_PER_WQ_STATS   ARRAY_SIZE(enic_per_wq_stats)
 static const struct enic_stat enic_tx_stats[] = {
 	ENIC_TX_STAT(tx_frames_ok),
 	ENIC_TX_STAT(tx_unicast_frames_ok),
@@ -60,6 +81,8 @@ static const struct enic_stat enic_tx_stats[] = {
 	ENIC_TX_STAT(tx_tso),
 };
 
+#define NUM_ENIC_TX_STATS	ARRAY_SIZE(enic_tx_stats)
+
 static const struct enic_stat enic_rx_stats[] = {
 	ENIC_RX_STAT(rx_frames_ok),
 	ENIC_RX_STAT(rx_frames_total),
@@ -84,13 +107,13 @@ static const struct enic_stat enic_rx_stats[] = {
 	ENIC_RX_STAT(rx_frames_to_max),
 };
 
+#define NUM_ENIC_RX_STATS	ARRAY_SIZE(enic_rx_stats)
+
 static const struct enic_stat enic_gen_stats[] = {
 	ENIC_GEN_STAT(dma_map_error),
 };
 
-static const unsigned int enic_n_tx_stats = ARRAY_SIZE(enic_tx_stats);
-static const unsigned int enic_n_rx_stats = ARRAY_SIZE(enic_rx_stats);
-static const unsigned int enic_n_gen_stats = ARRAY_SIZE(enic_gen_stats);
+#define NUM_ENIC_GEN_STATS	ARRAY_SIZE(enic_gen_stats)
 
 static void enic_intr_coal_set_rx(struct enic *enic, u32 timer)
 {
@@ -138,49 +161,153 @@ static void enic_get_drvinfo(struct net_device *netdev,
 	int err;
 
 	err = enic_dev_fw_info(enic, &fw_info);
-	/* return only when pci_zalloc_consistent fails in vnic_dev_fw_info
+	/* return only when dma_alloc_coherent fails in vnic_dev_fw_info
 	 * For other failures, like devcmd failure, we return previously
 	 * recorded info.
 	 */
 	if (err == -ENOMEM)
 		return;
 
-	strlcpy(drvinfo->driver, DRV_NAME, sizeof(drvinfo->driver));
-	strlcpy(drvinfo->version, DRV_VERSION, sizeof(drvinfo->version));
-	strlcpy(drvinfo->fw_version, fw_info->fw_version,
+	strscpy(drvinfo->driver, DRV_NAME, sizeof(drvinfo->driver));
+	strscpy(drvinfo->fw_version, fw_info->fw_version,
 		sizeof(drvinfo->fw_version));
-	strlcpy(drvinfo->bus_info, pci_name(enic->pdev),
+	strscpy(drvinfo->bus_info, pci_name(enic->pdev),
 		sizeof(drvinfo->bus_info));
 }
 
 static void enic_get_strings(struct net_device *netdev, u32 stringset,
 	u8 *data)
 {
+	struct enic *enic = netdev_priv(netdev);
 	unsigned int i;
+	unsigned int j;
 
 	switch (stringset) {
 	case ETH_SS_STATS:
-		for (i = 0; i < enic_n_tx_stats; i++) {
+		for (i = 0; i < NUM_ENIC_TX_STATS; i++) {
 			memcpy(data, enic_tx_stats[i].name, ETH_GSTRING_LEN);
 			data += ETH_GSTRING_LEN;
 		}
-		for (i = 0; i < enic_n_rx_stats; i++) {
+		for (i = 0; i < NUM_ENIC_RX_STATS; i++) {
 			memcpy(data, enic_rx_stats[i].name, ETH_GSTRING_LEN);
 			data += ETH_GSTRING_LEN;
 		}
-		for (i = 0; i < enic_n_gen_stats; i++) {
+		for (i = 0; i < NUM_ENIC_GEN_STATS; i++) {
 			memcpy(data, enic_gen_stats[i].name, ETH_GSTRING_LEN);
 			data += ETH_GSTRING_LEN;
 		}
+		for (i = 0; i < enic->rq_count; i++) {
+			for (j = 0; j < NUM_ENIC_PER_RQ_STATS; j++) {
+				snprintf(data, ETH_GSTRING_LEN,
+					 enic_per_rq_stats[j].name, i);
+				data += ETH_GSTRING_LEN;
+			}
+		}
+		for (i = 0; i < enic->wq_count; i++) {
+			for (j = 0; j < NUM_ENIC_PER_WQ_STATS; j++) {
+				snprintf(data, ETH_GSTRING_LEN,
+					 enic_per_wq_stats[j].name, i);
+				data += ETH_GSTRING_LEN;
+			}
+		}
 		break;
 	}
 }
 
+static void enic_get_ringparam(struct net_device *netdev,
+			       struct ethtool_ringparam *ring,
+			       struct kernel_ethtool_ringparam *kernel_ring,
+			       struct netlink_ext_ack *extack)
+{
+	struct enic *enic = netdev_priv(netdev);
+	struct vnic_enet_config *c = &enic->config;
+
+	ring->rx_max_pending = c->max_rq_ring;
+	ring->rx_pending = c->rq_desc_count;
+	ring->tx_max_pending = c->max_wq_ring;
+	ring->tx_pending = c->wq_desc_count;
+}
+
+static int enic_set_ringparam(struct net_device *netdev,
+			      struct ethtool_ringparam *ring,
+			      struct kernel_ethtool_ringparam *kernel_ring,
+			      struct netlink_ext_ack *extack)
+{
+	struct enic *enic = netdev_priv(netdev);
+	struct vnic_enet_config *c = &enic->config;
+	int running = netif_running(netdev);
+	unsigned int rx_pending;
+	unsigned int tx_pending;
+	int err = 0;
+
+	if (ring->rx_mini_max_pending || ring->rx_mini_pending) {
+		netdev_info(netdev,
+			    "modifying mini ring params is not supported");
+		return -EINVAL;
+	}
+	if (ring->rx_jumbo_max_pending || ring->rx_jumbo_pending) {
+		netdev_info(netdev,
+			    "modifying jumbo ring params is not supported");
+		return -EINVAL;
+	}
+	rx_pending = c->rq_desc_count;
+	tx_pending = c->wq_desc_count;
+	if (ring->rx_pending > c->max_rq_ring ||
+	    ring->rx_pending < ENIC_MIN_RQ_DESCS) {
+		netdev_info(netdev, "rx pending (%u) not in range [%u,%u]",
+			    ring->rx_pending, ENIC_MIN_RQ_DESCS,
+	      c->max_rq_ring);
+		return -EINVAL;
+	}
+	if (ring->tx_pending > c->max_wq_ring ||
+	    ring->tx_pending < ENIC_MIN_WQ_DESCS) {
+		netdev_info(netdev, "tx pending (%u) not in range [%u,%u]",
+			    ring->tx_pending, ENIC_MIN_WQ_DESCS,
+			c->max_wq_ring);
+		return -EINVAL;
+	}
+	if (running)
+		dev_close(netdev);
+	c->rq_desc_count =
+		ring->rx_pending & 0xffffffe0; /* must be aligned to groups of 32 */
+	c->wq_desc_count =
+		ring->tx_pending & 0xffffffe0; /* must be aligned to groups of 32 */
+	enic_free_vnic_resources(enic);
+	err = enic_alloc_vnic_resources(enic);
+	if (err) {
+		netdev_err(netdev,
+			   "Failed to alloc vNIC resources, aborting\n");
+		enic_free_vnic_resources(enic);
+		goto err_out;
+	}
+	enic_init_vnic_resources(enic);
+	if (running) {
+		err = dev_open(netdev, NULL);
+		if (err)
+			goto err_out;
+	}
+	return 0;
+err_out:
+	c->rq_desc_count = rx_pending;
+	c->wq_desc_count = tx_pending;
+	return err;
+}
+
 static int enic_get_sset_count(struct net_device *netdev, int sset)
 {
+	struct enic *enic = netdev_priv(netdev);
+	unsigned int n_per_rq_stats;
+	unsigned int n_per_wq_stats;
+	unsigned int n_stats;
+
 	switch (sset) {
 	case ETH_SS_STATS:
-		return enic_n_tx_stats + enic_n_rx_stats + enic_n_gen_stats;
+		n_per_rq_stats = NUM_ENIC_PER_RQ_STATS * enic->rq_count;
+		n_per_wq_stats = NUM_ENIC_PER_WQ_STATS * enic->wq_count;
+		n_stats = NUM_ENIC_TX_STATS + NUM_ENIC_RX_STATS +
+			NUM_ENIC_GEN_STATS +
+			n_per_rq_stats + n_per_wq_stats;
+		return n_stats;
 	default:
 		return -EOPNOTSUPP;
 	}
@@ -192,22 +319,41 @@ static void enic_get_ethtool_stats(struct net_device *netdev,
 	struct enic *enic = netdev_priv(netdev);
 	struct vnic_stats *vstats;
 	unsigned int i;
+	unsigned int j;
 	int err;
 
 	err = enic_dev_stats_dump(enic, &vstats);
-	/* return only when pci_zalloc_consistent fails in vnic_dev_stats_dump
+	/* return only when dma_alloc_coherent fails in vnic_dev_stats_dump
 	 * For other failures, like devcmd failure, we return previously
 	 * recorded stats.
 	 */
 	if (err == -ENOMEM)
 		return;
 
-	for (i = 0; i < enic_n_tx_stats; i++)
+	for (i = 0; i < NUM_ENIC_TX_STATS; i++)
 		*(data++) = ((u64 *)&vstats->tx)[enic_tx_stats[i].index];
-	for (i = 0; i < enic_n_rx_stats; i++)
+	for (i = 0; i < NUM_ENIC_RX_STATS; i++)
 		*(data++) = ((u64 *)&vstats->rx)[enic_rx_stats[i].index];
-	for (i = 0; i < enic_n_gen_stats; i++)
+	for (i = 0; i < NUM_ENIC_GEN_STATS; i++)
 		*(data++) = ((u64 *)&enic->gen_stats)[enic_gen_stats[i].index];
+	for (i = 0; i < enic->rq_count; i++) {
+		struct enic_rq_stats *rqstats = &enic->rq[i].stats;
+		int index;
+
+		for (j = 0; j < NUM_ENIC_PER_RQ_STATS; j++) {
+			index = enic_per_rq_stats[j].index;
+			*(data++) = ((u64 *)rqstats)[index];
+		}
+	}
+	for (i = 0; i < enic->wq_count; i++) {
+		struct enic_wq_stats *wqstats = &enic->wq[i].stats;
+		int index;
+
+		for (j = 0; j < NUM_ENIC_PER_WQ_STATS; j++) {
+			index = enic_per_wq_stats[j].index;
+			*(data++) = ((u64 *)wqstats)[index];
+		}
+	}
 }
 
 static u32 enic_get_msglevel(struct net_device *netdev)
@@ -223,7 +369,9 @@ static void enic_set_msglevel(struct net_device *netdev, u32 value)
 }
 
 static int enic_get_coalesce(struct net_device *netdev,
-	struct ethtool_coalesce *ecmd)
+			     struct ethtool_coalesce *ecmd,
+			     struct kernel_ethtool_coalesce *kernel_coal,
+			     struct netlink_ext_ack *extack)
 {
 	struct enic *enic = netdev_priv(netdev);
 	struct enic_rx_coal *rxcoal = &enic->rx_coalesce_setting;
@@ -248,25 +396,6 @@ static int enic_coalesce_valid(struct enic *enic,
 	u32 rx_coalesce_usecs_low = min_t(u32, coalesce_usecs_max,
 					  ec->rx_coalesce_usecs_low);
 
-	if (ec->rx_max_coalesced_frames		||
-	    ec->rx_coalesce_usecs_irq		||
-	    ec->rx_max_coalesced_frames_irq	||
-	    ec->tx_max_coalesced_frames		||
-	    ec->tx_coalesce_usecs_irq		||
-	    ec->tx_max_coalesced_frames_irq	||
-	    ec->stats_block_coalesce_usecs	||
-	    ec->use_adaptive_tx_coalesce	||
-	    ec->pkt_rate_low			||
-	    ec->rx_max_coalesced_frames_low	||
-	    ec->tx_coalesce_usecs_low		||
-	    ec->tx_max_coalesced_frames_low	||
-	    ec->pkt_rate_high			||
-	    ec->rx_max_coalesced_frames_high	||
-	    ec->tx_coalesce_usecs_high		||
-	    ec->tx_max_coalesced_frames_high	||
-	    ec->rate_sample_interval)
-		return -EINVAL;
-
 	if ((vnic_dev_get_intr_mode(enic->vdev) != VNIC_DEV_INTR_MODE_MSIX) &&
 	    ec->tx_coalesce_usecs)
 		return -EINVAL;
@@ -287,7 +416,9 @@ static int enic_coalesce_valid(struct enic *enic,
 }
 
 static int enic_set_coalesce(struct net_device *netdev,
-	struct ethtool_coalesce *ecmd)
+			     struct ethtool_coalesce *ecmd,
+			     struct kernel_ethtool_coalesce *kernel_coal,
+			     struct netlink_ext_ack *extack)
 {
 	struct enic *enic = netdev_priv(netdev);
 	u32 tx_coalesce_usecs;
@@ -378,7 +509,6 @@ static int enic_grxclsrule(struct enic *enic, struct ethtool_rxnfc *cmd)
 		break;
 	default:
 		return -EINVAL;
-		break;
 	}
 
 	fsp->h_u.tcp_ip4_spec.ip4src = flow_get_u32_src(&n->keys);
@@ -398,6 +528,51 @@ static int enic_grxclsrule(struct enic *enic, struct ethtool_rxnfc *cmd)
 	return 0;
 }
 
+static int enic_get_rx_flow_hash(struct net_device *dev,
+				 struct ethtool_rxfh_fields *cmd)
+{
+	struct enic *enic = netdev_priv(dev);
+	u8 rss_hash_type = 0;
+	cmd->data = 0;
+
+	spin_lock_bh(&enic->devcmd_lock);
+	(void)vnic_dev_capable_rss_hash_type(enic->vdev, &rss_hash_type);
+	spin_unlock_bh(&enic->devcmd_lock);
+	switch (cmd->flow_type) {
+	case TCP_V6_FLOW:
+	case TCP_V4_FLOW:
+		cmd->data |= RXH_L4_B_0_1 | RXH_L4_B_2_3 |
+			     RXH_IP_SRC | RXH_IP_DST;
+		break;
+	case UDP_V6_FLOW:
+		cmd->data |= RXH_IP_SRC | RXH_IP_DST;
+		if (rss_hash_type & NIC_CFG_RSS_HASH_TYPE_UDP_IPV6)
+			cmd->data |= RXH_L4_B_0_1 | RXH_L4_B_2_3;
+		break;
+	case UDP_V4_FLOW:
+		cmd->data |= RXH_IP_SRC | RXH_IP_DST;
+		if (rss_hash_type & NIC_CFG_RSS_HASH_TYPE_UDP_IPV4)
+			cmd->data |= RXH_L4_B_0_1 | RXH_L4_B_2_3;
+		break;
+	case SCTP_V4_FLOW:
+	case AH_ESP_V4_FLOW:
+	case AH_V4_FLOW:
+	case ESP_V4_FLOW:
+	case SCTP_V6_FLOW:
+	case AH_ESP_V6_FLOW:
+	case AH_V6_FLOW:
+	case ESP_V6_FLOW:
+	case IPV4_FLOW:
+	case IPV6_FLOW:
+		cmd->data |= RXH_IP_SRC | RXH_IP_DST;
+		break;
+	default:
+		return -EINVAL;
+	}
+
+	return 0;
+}
+
 static int enic_get_rxnfc(struct net_device *dev, struct ethtool_rxnfc *cmd,
 			  u32 *rule_locs)
 {
@@ -432,94 +607,95 @@ static int enic_get_rxnfc(struct net_device *dev, struct ethtool_rxnfc *cmd,
 	return ret;
 }
 
-static int enic_get_tunable(struct net_device *dev,
-			    const struct ethtool_tunable *tuna, void *data)
-{
-	struct enic *enic = netdev_priv(dev);
-	int ret = 0;
-
-	switch (tuna->id) {
-	case ETHTOOL_RX_COPYBREAK:
-		*(u32 *)data = enic->rx_copybreak;
-		break;
-	default:
-		ret = -EINVAL;
-		break;
-	}
-
-	return ret;
-}
-
-static int enic_set_tunable(struct net_device *dev,
-			    const struct ethtool_tunable *tuna,
-			    const void *data)
-{
-	struct enic *enic = netdev_priv(dev);
-	int ret = 0;
-
-	switch (tuna->id) {
-	case ETHTOOL_RX_COPYBREAK:
-		enic->rx_copybreak = *(u32 *)data;
-		break;
-	default:
-		ret = -EINVAL;
-		break;
-	}
-
-	return ret;
-}
-
 static u32 enic_get_rxfh_key_size(struct net_device *netdev)
 {
 	return ENIC_RSS_LEN;
 }
 
-static int enic_get_rxfh(struct net_device *netdev, u32 *indir, u8 *hkey,
-			 u8 *hfunc)
+static int enic_get_rxfh(struct net_device *netdev,
+			 struct ethtool_rxfh_param *rxfh)
 {
 	struct enic *enic = netdev_priv(netdev);
 
-	if (hkey)
-		memcpy(hkey, enic->rss_key, ENIC_RSS_LEN);
+	if (rxfh->key)
+		memcpy(rxfh->key, enic->rss_key, ENIC_RSS_LEN);
 
-	if (hfunc)
-		*hfunc = ETH_RSS_HASH_TOP;
+	rxfh->hfunc = ETH_RSS_HASH_TOP;
 
 	return 0;
 }
 
-static int enic_set_rxfh(struct net_device *netdev, const u32 *indir,
-			 const u8 *hkey, const u8 hfunc)
+static int enic_set_rxfh(struct net_device *netdev,
+			 struct ethtool_rxfh_param *rxfh,
+			 struct netlink_ext_ack *extack)
 {
 	struct enic *enic = netdev_priv(netdev);
 
-	if ((hfunc != ETH_RSS_HASH_NO_CHANGE && hfunc != ETH_RSS_HASH_TOP) ||
-	    indir)
+	if (rxfh->indir ||
+	    (rxfh->hfunc != ETH_RSS_HASH_NO_CHANGE &&
+	     rxfh->hfunc != ETH_RSS_HASH_TOP))
 		return -EINVAL;
 
-	if (hkey)
-		memcpy(enic->rss_key, hkey, ENIC_RSS_LEN);
+	if (rxfh->key)
+		memcpy(enic->rss_key, rxfh->key, ENIC_RSS_LEN);
 
 	return __enic_set_rsskey(enic);
 }
 
+static int enic_get_ts_info(struct net_device *netdev,
+			    struct kernel_ethtool_ts_info *info)
+{
+	info->so_timestamping = SOF_TIMESTAMPING_TX_SOFTWARE;
+
+	return 0;
+}
+
+static void enic_get_channels(struct net_device *netdev,
+			      struct ethtool_channels *channels)
+{
+	struct enic *enic = netdev_priv(netdev);
+
+	switch (vnic_dev_get_intr_mode(enic->vdev)) {
+	case VNIC_DEV_INTR_MODE_MSIX:
+		channels->max_rx = min(enic->rq_avail, ENIC_RQ_MAX);
+		channels->max_tx = min(enic->wq_avail, ENIC_WQ_MAX);
+		channels->rx_count = enic->rq_count;
+		channels->tx_count = enic->wq_count;
+		break;
+	case VNIC_DEV_INTR_MODE_MSI:
+	case VNIC_DEV_INTR_MODE_INTX:
+		channels->max_combined = 1;
+		channels->combined_count = 1;
+		break;
+	default:
+		break;
+	}
+}
+
 static const struct ethtool_ops enic_ethtool_ops = {
+	.supported_coalesce_params = ETHTOOL_COALESCE_USECS |
+				     ETHTOOL_COALESCE_USE_ADAPTIVE_RX |
+				     ETHTOOL_COALESCE_RX_USECS_LOW |
+				     ETHTOOL_COALESCE_RX_USECS_HIGH,
 	.get_drvinfo = enic_get_drvinfo,
 	.get_msglevel = enic_get_msglevel,
 	.set_msglevel = enic_set_msglevel,
 	.get_link = ethtool_op_get_link,
 	.get_strings = enic_get_strings,
+	.get_ringparam = enic_get_ringparam,
+	.set_ringparam = enic_set_ringparam,
 	.get_sset_count = enic_get_sset_count,
 	.get_ethtool_stats = enic_get_ethtool_stats,
 	.get_coalesce = enic_get_coalesce,
 	.set_coalesce = enic_set_coalesce,
 	.get_rxnfc = enic_get_rxnfc,
-	.get_tunable = enic_get_tunable,
-	.set_tunable = enic_set_tunable,
 	.get_rxfh_key_size = enic_get_rxfh_key_size,
 	.get_rxfh = enic_get_rxfh,
 	.set_rxfh = enic_set_rxfh,
+	.get_rxfh_fields = enic_get_rx_flow_hash,
 	.get_link_ksettings = enic_get_ksettings,
+	.get_ts_info = enic_get_ts_info,
+	.get_channels = enic_get_channels,
 };
 
 void enic_set_ethtool_ops(struct net_device *netdev)
diff --git a/drivers/net/ethernet/cisco/enic/enic_main.c b/drivers/net/ethernet/cisco/enic/enic_main.c
index 6a9c8878aca0..6bc8dfdb3d4b 100644
--- a/drivers/net/ethernet/cisco/enic/enic_main.c
+++ b/drivers/net/ethernet/cisco/enic/enic_main.c
@@ -46,6 +46,7 @@
 #include <linux/crash_dump.h>
 #include <net/busy_poll.h>
 #include <net/vxlan.h>
+#include <net/netdev_queues.h>
 
 #include "cq_enet_desc.h"
 #include "vnic_dev.h"
@@ -57,18 +58,15 @@
 #include "enic_dev.h"
 #include "enic_pp.h"
 #include "enic_clsf.h"
+#include "enic_rq.h"
+#include "enic_wq.h"
 
 #define ENIC_NOTIFY_TIMER_PERIOD	(2 * HZ)
-#define WQ_ENET_MAX_DESC_LEN		(1 << WQ_ENET_LEN_BITS)
-#define MAX_TSO				(1 << 16)
-#define ENIC_DESC_MAX_SPLITS		(MAX_TSO / WQ_ENET_MAX_DESC_LEN + 1)
 
 #define PCI_DEVICE_ID_CISCO_VIC_ENET         0x0043  /* ethernet vnic */
 #define PCI_DEVICE_ID_CISCO_VIC_ENET_DYN     0x0044  /* enet dynamic vnic */
 #define PCI_DEVICE_ID_CISCO_VIC_ENET_VF      0x0071  /* enet SRIOV VF */
 
-#define RX_COPYBREAK_DEFAULT		256
-
 /* Supported devices */
 static const struct pci_device_id enic_id_table[] = {
 	{ PCI_VDEVICE(CISCO, PCI_DEVICE_ID_CISCO_VIC_ENET) },
@@ -80,7 +78,6 @@ static const struct pci_device_id enic_id_table[] = {
 MODULE_DESCRIPTION(DRV_DESCRIPTION);
 MODULE_AUTHOR("Scott Feldman <scofeldm@cisco.com>");
 MODULE_LICENSE("GPL");
-MODULE_VERSION(DRV_VERSION);
 MODULE_DEVICE_TABLE(pci, enic_id_table);
 
 #define ENIC_LARGE_PKT_THRESHOLD		1000
@@ -109,7 +106,7 @@ static struct enic_intr_mod_table mod_table[ENIC_MAX_COALESCE_TIMERS + 1] = {
 static struct enic_intr_mod_range mod_range[ENIC_MAX_LINK_SPEEDS] = {
 	{0,  0}, /* 0  - 4  Gbps */
 	{0,  3}, /* 4  - 10 Gbps */
-	{3,  6}, /* 10 - 40 Gbps */
+	{3,  6}, /* 10+ Gbps */
 };
 
 static void enic_init_affinity_hint(struct enic *enic)
@@ -119,7 +116,7 @@ static void enic_init_affinity_hint(struct enic *enic)
 
 	for (i = 0; i < enic->intr_count; i++) {
 		if (enic_is_err_intr(enic, i) || enic_is_notify_intr(enic, i) ||
-		    (enic->msix[i].affinity_mask &&
+		    (cpumask_available(enic->msix[i].affinity_mask) &&
 		     !cpumask_empty(enic->msix[i].affinity_mask)))
 			continue;
 		if (zalloc_cpumask_var(&enic->msix[i].affinity_mask,
@@ -148,20 +145,20 @@ static void enic_set_affinity_hint(struct enic *enic)
 	for (i = 0; i < enic->intr_count; i++) {
 		if (enic_is_err_intr(enic, i)		||
 		    enic_is_notify_intr(enic, i)	||
-		    !enic->msix[i].affinity_mask	||
+		    !cpumask_available(enic->msix[i].affinity_mask) ||
 		    cpumask_empty(enic->msix[i].affinity_mask))
 			continue;
-		err = irq_set_affinity_hint(enic->msix_entry[i].vector,
-					    enic->msix[i].affinity_mask);
+		err = irq_update_affinity_hint(enic->msix_entry[i].vector,
+					       enic->msix[i].affinity_mask);
 		if (err)
-			netdev_warn(enic->netdev, "irq_set_affinity_hint failed, err %d\n",
+			netdev_warn(enic->netdev, "irq_update_affinity_hint failed, err %d\n",
 				    err);
 	}
 
 	for (i = 0; i < enic->wq_count; i++) {
 		int wq_intr = enic_msix_wq_intr(enic, i);
 
-		if (enic->msix[wq_intr].affinity_mask &&
+		if (cpumask_available(enic->msix[wq_intr].affinity_mask) &&
 		    !cpumask_empty(enic->msix[wq_intr].affinity_mask))
 			netif_set_xps_queue(enic->netdev,
 					    enic->msix[wq_intr].affinity_mask,
@@ -174,40 +171,21 @@ static void enic_unset_affinity_hint(struct enic *enic)
 	int i;
 
 	for (i = 0; i < enic->intr_count; i++)
-		irq_set_affinity_hint(enic->msix_entry[i].vector, NULL);
+		irq_update_affinity_hint(enic->msix_entry[i].vector, NULL);
 }
 
-static void enic_udp_tunnel_add(struct net_device *netdev,
-				struct udp_tunnel_info *ti)
+static int enic_udp_tunnel_set_port(struct net_device *netdev,
+				    unsigned int table, unsigned int entry,
+				    struct udp_tunnel_info *ti)
 {
 	struct enic *enic = netdev_priv(netdev);
-	__be16 port = ti->port;
 	int err;
 
 	spin_lock_bh(&enic->devcmd_lock);
 
-	if (ti->type != UDP_TUNNEL_TYPE_VXLAN) {
-		netdev_info(netdev, "udp_tnl: only vxlan tunnel offload supported");
-		goto error;
-	}
-
-	if (ti->sa_family != AF_INET) {
-		netdev_info(netdev, "vxlan: only IPv4 offload supported");
-		goto error;
-	}
-
-	if (enic->vxlan.vxlan_udp_port_number) {
-		if (ntohs(port) == enic->vxlan.vxlan_udp_port_number)
-			netdev_warn(netdev, "vxlan: udp port already offloaded");
-		else
-			netdev_info(netdev, "vxlan: offload supported for only one UDP port");
-
-		goto error;
-	}
-
 	err = vnic_dev_overlay_offload_cfg(enic->vdev,
 					   OVERLAY_CFG_VXLAN_PORT_UPDATE,
-					   ntohs(port));
+					   ntohs(ti->port));
 	if (err)
 		goto error;
 
@@ -216,53 +194,50 @@ static void enic_udp_tunnel_add(struct net_device *netdev,
 	if (err)
 		goto error;
 
-	enic->vxlan.vxlan_udp_port_number = ntohs(port);
-
-	netdev_info(netdev, "vxlan fw-vers-%d: offload enabled for udp port: %d, sa_family: %d ",
-		    (int)enic->vxlan.patch_level, ntohs(port), ti->sa_family);
-
-	goto unlock;
-
+	enic->vxlan.vxlan_udp_port_number = ntohs(ti->port);
 error:
-	netdev_info(netdev, "failed to offload udp port: %d, sa_family: %d, type: %d",
-		    ntohs(port), ti->sa_family, ti->type);
-unlock:
 	spin_unlock_bh(&enic->devcmd_lock);
+
+	return err;
 }
 
-static void enic_udp_tunnel_del(struct net_device *netdev,
-				struct udp_tunnel_info *ti)
+static int enic_udp_tunnel_unset_port(struct net_device *netdev,
+				      unsigned int table, unsigned int entry,
+				      struct udp_tunnel_info *ti)
 {
 	struct enic *enic = netdev_priv(netdev);
 	int err;
 
 	spin_lock_bh(&enic->devcmd_lock);
 
-	if ((ti->sa_family != AF_INET) ||
-	    ((ntohs(ti->port) != enic->vxlan.vxlan_udp_port_number)) ||
-	    (ti->type != UDP_TUNNEL_TYPE_VXLAN)) {
-		netdev_info(netdev, "udp_tnl: port:%d, sa_family: %d, type: %d not offloaded",
-			    ntohs(ti->port), ti->sa_family, ti->type);
-		goto unlock;
-	}
-
 	err = vnic_dev_overlay_offload_ctrl(enic->vdev, OVERLAY_FEATURE_VXLAN,
 					    OVERLAY_OFFLOAD_DISABLE);
-	if (err) {
-		netdev_err(netdev, "vxlan: del offload udp port: %d failed",
-			   ntohs(ti->port));
+	if (err)
 		goto unlock;
-	}
 
 	enic->vxlan.vxlan_udp_port_number = 0;
 
-	netdev_info(netdev, "vxlan: del offload udp port %d, family %d\n",
-		    ntohs(ti->port), ti->sa_family);
-
 unlock:
 	spin_unlock_bh(&enic->devcmd_lock);
+
+	return err;
 }
 
+static const struct udp_tunnel_nic_info enic_udp_tunnels = {
+	.set_port	= enic_udp_tunnel_set_port,
+	.unset_port	= enic_udp_tunnel_unset_port,
+	.tables		= {
+		{ .n_entries = 1, .tunnel_types = UDP_TUNNEL_TYPE_VXLAN, },
+	},
+}, enic_udp_tunnels_v4 = {
+	.set_port	= enic_udp_tunnel_set_port,
+	.unset_port	= enic_udp_tunnel_unset_port,
+	.flags		= UDP_TUNNEL_NIC_INFO_IPV4_ONLY,
+	.tables		= {
+		{ .n_entries = 1, .tunnel_types = UDP_TUNNEL_TYPE_VXLAN, },
+	},
+};
+
 static netdev_features_t enic_features_check(struct sk_buff *skb,
 					     struct net_device *dev,
 					     netdev_features_t features)
@@ -271,22 +246,37 @@ static netdev_features_t enic_features_check(struct sk_buff *skb,
 	struct enic *enic = netdev_priv(dev);
 	struct udphdr *udph;
 	u16 port = 0;
-	u16 proto;
+	u8 proto;
 
 	if (!skb->encapsulation)
 		return features;
 
 	features = vxlan_features_check(skb, features);
 
-	/* hardware only supports IPv4 vxlan tunnel */
-	if (vlan_get_protocol(skb) != htons(ETH_P_IP))
+	switch (vlan_get_protocol(skb)) {
+	case htons(ETH_P_IPV6):
+		if (!(enic->vxlan.flags & ENIC_VXLAN_OUTER_IPV6))
+			goto out;
+		proto = ipv6_hdr(skb)->nexthdr;
+		break;
+	case htons(ETH_P_IP):
+		proto = ip_hdr(skb)->protocol;
+		break;
+	default:
 		goto out;
+	}
 
-	/* hardware does not support offload of ipv6 inner pkt */
-	if (eth->h_proto != ntohs(ETH_P_IP))
+	switch (eth->h_proto) {
+	case ntohs(ETH_P_IPV6):
+		if (!(enic->vxlan.flags & ENIC_VXLAN_INNER_IPV6))
+			goto out;
+		fallthrough;
+	case ntohs(ETH_P_IP):
+		break;
+	default:
 		goto out;
+	}
 
-	proto = ip_hdr(skb)->protocol;
 
 	if (proto == IPPROTO_UDP) {
 		udph = udp_hdr(skb);
@@ -329,48 +319,6 @@ int enic_is_valid_vf(struct enic *enic, int vf)
 #endif
 }
 
-static void enic_free_wq_buf(struct vnic_wq *wq, struct vnic_wq_buf *buf)
-{
-	struct enic *enic = vnic_dev_priv(wq->vdev);
-
-	if (buf->sop)
-		pci_unmap_single(enic->pdev, buf->dma_addr,
-			buf->len, PCI_DMA_TODEVICE);
-	else
-		pci_unmap_page(enic->pdev, buf->dma_addr,
-			buf->len, PCI_DMA_TODEVICE);
-
-	if (buf->os_buf)
-		dev_kfree_skb_any(buf->os_buf);
-}
-
-static void enic_wq_free_buf(struct vnic_wq *wq,
-	struct cq_desc *cq_desc, struct vnic_wq_buf *buf, void *opaque)
-{
-	enic_free_wq_buf(wq, buf);
-}
-
-static int enic_wq_service(struct vnic_dev *vdev, struct cq_desc *cq_desc,
-	u8 type, u16 q_number, u16 completed_index, void *opaque)
-{
-	struct enic *enic = vnic_dev_priv(vdev);
-
-	spin_lock(&enic->wq_lock[q_number]);
-
-	vnic_wq_service(&enic->wq[q_number], cq_desc,
-		completed_index, enic_wq_free_buf,
-		opaque);
-
-	if (netif_tx_queue_stopped(netdev_get_tx_queue(enic->netdev, q_number)) &&
-	    vnic_wq_desc_avail(&enic->wq[q_number]) >=
-	    (MAX_SKB_FRAGS + ENIC_DESC_MAX_SPLITS))
-		netif_wake_subqueue(enic->netdev, q_number);
-
-	spin_unlock(&enic->wq_lock[q_number]);
-
-	return 0;
-}
-
 static bool enic_log_q_error(struct enic *enic)
 {
 	unsigned int i;
@@ -378,7 +326,7 @@ static bool enic_log_q_error(struct enic *enic)
 	bool err = false;
 
 	for (i = 0; i < enic->wq_count; i++) {
-		error_status = vnic_wq_error_status(&enic->wq[i]);
+		error_status = vnic_wq_error_status(&enic->wq[i].vwq);
 		err |= error_status;
 		if (error_status)
 			netdev_err(enic->netdev, "WQ[%d] error_status %d\n",
@@ -386,7 +334,7 @@ static bool enic_log_q_error(struct enic *enic)
 	}
 
 	for (i = 0; i < enic->rq_count; i++) {
-		error_status = vnic_rq_error_status(&enic->rq[i]);
+		error_status = vnic_rq_error_status(&enic->rq[i].vrq);
 		err |= error_status;
 		if (error_status)
 			netdev_err(enic->netdev, "RQ[%d] error_status %d\n",
@@ -429,6 +377,36 @@ static void enic_mtu_check(struct enic *enic)
 	}
 }
 
+static void enic_set_rx_coal_setting(struct enic *enic)
+{
+	unsigned int speed;
+	int index = -1;
+	struct enic_rx_coal *rx_coal = &enic->rx_coalesce_setting;
+
+	/* 1. Read the link speed from fw
+	 * 2. Pick the default range for the speed
+	 * 3. Update it in enic->rx_coalesce_setting
+	 */
+	speed = vnic_dev_port_speed(enic->vdev);
+	if (speed > ENIC_LINK_SPEED_10G)
+		index = ENIC_LINK_40G_INDEX;
+	else if (speed > ENIC_LINK_SPEED_4G)
+		index = ENIC_LINK_10G_INDEX;
+	else
+		index = ENIC_LINK_4G_INDEX;
+
+	rx_coal->small_pkt_range_start = mod_range[index].small_pkt_range_start;
+	rx_coal->large_pkt_range_start = mod_range[index].large_pkt_range_start;
+	rx_coal->range_end = ENIC_RX_COALESCE_RANGE_END;
+
+	/* Start with the value provided by UCSM */
+	for (index = 0; index < enic->rq_count; index++)
+		enic->cq[index].cur_rx_coal_timeval =
+				enic->config.intr_timer_usec;
+
+	rx_coal->use_adaptive_rx_coalesce = 1;
+}
+
 static void enic_link_check(struct enic *enic)
 {
 	int link_status = vnic_dev_link_status(enic->vdev);
@@ -437,6 +415,7 @@ static void enic_link_check(struct enic *enic)
 	if (link_status && !carrier_ok) {
 		netdev_info(enic->netdev, "Link UP\n");
 		netif_carrier_on(enic->netdev);
+		enic_set_rx_coal_setting(enic);
 	} else if (!link_status && carrier_ok) {
 		netdev_info(enic->netdev, "Link DOWN\n");
 		netif_carrier_off(enic->netdev);
@@ -456,9 +435,9 @@ static irqreturn_t enic_isr_legacy(int irq, void *data)
 {
 	struct net_device *netdev = data;
 	struct enic *enic = netdev_priv(netdev);
-	unsigned int io_intr = enic_legacy_io_intr();
-	unsigned int err_intr = enic_legacy_err_intr();
-	unsigned int notify_intr = enic_legacy_notify_intr();
+	unsigned int io_intr = ENIC_LEGACY_IO_INTR;
+	unsigned int err_intr = ENIC_LEGACY_ERR_INTR;
+	unsigned int notify_intr = ENIC_LEGACY_NOTIFY_INTR;
 	u32 pba;
 
 	vnic_intr_mask(&enic->intr[io_intr]);
@@ -582,8 +561,8 @@ static int enic_queue_wq_skb_vlan(struct enic *enic, struct vnic_wq *wq,
 	dma_addr_t dma_addr;
 	int err = 0;
 
-	dma_addr = pci_map_single(enic->pdev, skb->data, head_len,
-				  PCI_DMA_TODEVICE);
+	dma_addr = dma_map_single(&enic->pdev->dev, skb->data, head_len,
+				  DMA_TO_DEVICE);
 	if (unlikely(enic_dma_map_check(enic, dma_addr)))
 		return -ENOMEM;
 
@@ -598,6 +577,11 @@ static int enic_queue_wq_skb_vlan(struct enic *enic, struct vnic_wq *wq,
 	if (!eop)
 		err = enic_queue_wq_skb_cont(enic, wq, skb, len_left, loopback);
 
+	/* The enic_queue_wq_desc() above does not do HW checksum */
+	enic->wq[wq->index].stats.csum_none++;
+	enic->wq[wq->index].stats.packets++;
+	enic->wq[wq->index].stats.bytes += skb->len;
+
 	return err;
 }
 
@@ -613,8 +597,8 @@ static int enic_queue_wq_skb_csum_l4(struct enic *enic, struct vnic_wq *wq,
 	dma_addr_t dma_addr;
 	int err = 0;
 
-	dma_addr = pci_map_single(enic->pdev, skb->data, head_len,
-				  PCI_DMA_TODEVICE);
+	dma_addr = dma_map_single(&enic->pdev->dev, skb->data, head_len,
+				  DMA_TO_DEVICE);
 	if (unlikely(enic_dma_map_check(enic, dma_addr)))
 		return -ENOMEM;
 
@@ -630,17 +614,34 @@ static int enic_queue_wq_skb_csum_l4(struct enic *enic, struct vnic_wq *wq,
 	if (!eop)
 		err = enic_queue_wq_skb_cont(enic, wq, skb, len_left, loopback);
 
+	enic->wq[wq->index].stats.csum_partial++;
+	enic->wq[wq->index].stats.packets++;
+	enic->wq[wq->index].stats.bytes += skb->len;
+
 	return err;
 }
 
 static void enic_preload_tcp_csum_encap(struct sk_buff *skb)
 {
-	if (skb->protocol == cpu_to_be16(ETH_P_IP)) {
+	const struct ethhdr *eth = (struct ethhdr *)skb_inner_mac_header(skb);
+
+	switch (eth->h_proto) {
+	case ntohs(ETH_P_IP):
 		inner_ip_hdr(skb)->check = 0;
 		inner_tcp_hdr(skb)->check =
 			~csum_tcpudp_magic(inner_ip_hdr(skb)->saddr,
 					   inner_ip_hdr(skb)->daddr, 0,
 					   IPPROTO_TCP, 0);
+		break;
+	case ntohs(ETH_P_IPV6):
+		inner_tcp_hdr(skb)->check =
+			~csum_ipv6_magic(&inner_ipv6_hdr(skb)->saddr,
+					 &inner_ipv6_hdr(skb)->daddr, 0,
+					 IPPROTO_TCP, 0);
+		break;
+	default:
+		WARN_ONCE(1, "Non ipv4/ipv6 inner pkt for encap offload");
+		break;
 	}
 }
 
@@ -656,8 +657,7 @@ static void enic_preload_tcp_csum(struct sk_buff *skb)
 		tcp_hdr(skb)->check = ~csum_tcpudp_magic(ip_hdr(skb)->saddr,
 			ip_hdr(skb)->daddr, 0, IPPROTO_TCP, 0);
 	} else if (skb->protocol == cpu_to_be16(ETH_P_IPV6)) {
-		tcp_hdr(skb)->check = ~csum_ipv6_magic(&ipv6_hdr(skb)->saddr,
-			&ipv6_hdr(skb)->daddr, 0, IPPROTO_TCP, 0);
+		tcp_v6_gso_csum_prep(skb);
 	}
 }
 
@@ -672,16 +672,18 @@ static int enic_queue_wq_skb_tso(struct enic *enic, struct vnic_wq *wq,
 	unsigned int offset = 0;
 	unsigned int hdr_len;
 	dma_addr_t dma_addr;
+	unsigned int pkts;
 	unsigned int len;
 	skb_frag_t *frag;
 
 	if (skb->encapsulation) {
-		hdr_len = skb_inner_transport_header(skb) - skb->data;
-		hdr_len += inner_tcp_hdrlen(skb);
+		hdr_len = skb_inner_tcp_all_headers(skb);
 		enic_preload_tcp_csum_encap(skb);
+		enic->wq[wq->index].stats.encap_tso++;
 	} else {
-		hdr_len = skb_transport_offset(skb) + tcp_hdrlen(skb);
+		hdr_len = skb_tcp_all_headers(skb);
 		enic_preload_tcp_csum(skb);
+		enic->wq[wq->index].stats.tso++;
 	}
 
 	/* Queue WQ_ENET_MAX_DESC_LEN length descriptors
@@ -689,8 +691,9 @@ static int enic_queue_wq_skb_tso(struct enic *enic, struct vnic_wq *wq,
 	 */
 	while (frag_len_left) {
 		len = min(frag_len_left, (unsigned int)WQ_ENET_MAX_DESC_LEN);
-		dma_addr = pci_map_single(enic->pdev, skb->data + offset, len,
-					  PCI_DMA_TODEVICE);
+		dma_addr = dma_map_single(&enic->pdev->dev,
+					  skb->data + offset, len,
+					  DMA_TO_DEVICE);
 		if (unlikely(enic_dma_map_check(enic, dma_addr)))
 			return -ENOMEM;
 		enic_queue_wq_desc_tso(wq, skb, dma_addr, len, mss, hdr_len,
@@ -701,7 +704,7 @@ static int enic_queue_wq_skb_tso(struct enic *enic, struct vnic_wq *wq,
 	}
 
 	if (eop)
-		return 0;
+		goto tso_out_stats;
 
 	/* Queue WQ_ENET_MAX_DESC_LEN length descriptors
 	 * for additional data fragments
@@ -728,6 +731,15 @@ static int enic_queue_wq_skb_tso(struct enic *enic, struct vnic_wq *wq,
 		}
 	}
 
+tso_out_stats:
+	/* calculate how many packets tso sent */
+	len = skb->len - hdr_len;
+	pkts = len / mss;
+	if ((len % mss) > 0)
+		pkts++;
+	enic->wq[wq->index].stats.packets += pkts;
+	enic->wq[wq->index].stats.bytes += (len + (pkts * hdr_len));
+
 	return 0;
 }
 
@@ -748,8 +760,8 @@ static inline int enic_queue_wq_skb_encap(struct enic *enic, struct vnic_wq *wq,
 	dma_addr_t dma_addr;
 	int err = 0;
 
-	dma_addr = pci_map_single(enic->pdev, skb->data, head_len,
-				  PCI_DMA_TODEVICE);
+	dma_addr = dma_map_single(&enic->pdev->dev, skb->data, head_len,
+				  DMA_TO_DEVICE);
 	if (unlikely(enic_dma_map_check(enic, dma_addr)))
 		return -ENOMEM;
 
@@ -760,10 +772,14 @@ static inline int enic_queue_wq_skb_encap(struct enic *enic, struct vnic_wq *wq,
 	if (!eop)
 		err = enic_queue_wq_skb_cont(enic, wq, skb, len_left, loopback);
 
+	enic->wq[wq->index].stats.encap_csum++;
+	enic->wq[wq->index].stats.packets++;
+	enic->wq[wq->index].stats.bytes += skb->len;
+
 	return err;
 }
 
-static inline void enic_queue_wq_skb(struct enic *enic,
+static inline int enic_queue_wq_skb(struct enic *enic,
 	struct vnic_wq *wq, struct sk_buff *skb)
 {
 	unsigned int mss = skb_shinfo(skb)->gso_size;
@@ -776,6 +792,7 @@ static inline void enic_queue_wq_skb(struct enic *enic,
 		/* VLAN tag from trunking driver */
 		vlan_tag_insert = 1;
 		vlan_tag = skb_vlan_tag_get(skb);
+		enic->wq[wq->index].stats.add_vlan++;
 	} else if (enic->loop_enable) {
 		vlan_tag = enic->loop_tag;
 		loopback = 1;
@@ -788,7 +805,7 @@ static inline void enic_queue_wq_skb(struct enic *enic,
 	else if (skb->encapsulation)
 		err = enic_queue_wq_skb_encap(enic, wq, skb, vlan_tag_insert,
 					      vlan_tag, loopback);
-	else if	(skb->ip_summed == CHECKSUM_PARTIAL)
+	else if (skb->ip_summed == CHECKSUM_PARTIAL)
 		err = enic_queue_wq_skb_csum_l4(enic, wq, skb, vlan_tag_insert,
 						vlan_tag, loopback);
 	else
@@ -809,6 +826,7 @@ static inline void enic_queue_wq_skb(struct enic *enic,
 		wq->to_use = buf->next;
 		dev_kfree_skb(skb);
 	}
+	return err;
 }
 
 /* netif_tx_lock held, process context with BHs disabled, or BH */
@@ -820,13 +838,15 @@ static netdev_tx_t enic_hard_start_xmit(struct sk_buff *skb,
 	unsigned int txq_map;
 	struct netdev_queue *txq;
 
+	txq_map = skb_get_queue_mapping(skb) % enic->wq_count;
+	wq = &enic->wq[txq_map].vwq;
+
 	if (skb->len <= 0) {
 		dev_kfree_skb_any(skb);
+		enic->wq[wq->index].stats.null_pkt++;
 		return NETDEV_TX_OK;
 	}
 
-	txq_map = skb_get_queue_mapping(skb) % enic->wq_count;
-	wq = &enic->wq[txq_map];
 	txq = netdev_get_tx_queue(netdev, txq_map);
 
 	/* Non-TSO sends must fit within ENIC_NON_TSO_MAX_DESC descs,
@@ -838,42 +858,52 @@ static netdev_tx_t enic_hard_start_xmit(struct sk_buff *skb,
 	    skb_shinfo(skb)->nr_frags + 1 > ENIC_NON_TSO_MAX_DESC &&
 	    skb_linearize(skb)) {
 		dev_kfree_skb_any(skb);
+		enic->wq[wq->index].stats.skb_linear_fail++;
 		return NETDEV_TX_OK;
 	}
 
-	spin_lock(&enic->wq_lock[txq_map]);
+	spin_lock(&enic->wq[txq_map].lock);
 
 	if (vnic_wq_desc_avail(wq) <
 	    skb_shinfo(skb)->nr_frags + ENIC_DESC_MAX_SPLITS) {
 		netif_tx_stop_queue(txq);
 		/* This is a hard error, log it */
 		netdev_err(netdev, "BUG! Tx ring full when queue awake!\n");
-		spin_unlock(&enic->wq_lock[txq_map]);
+		spin_unlock(&enic->wq[txq_map].lock);
+		enic->wq[wq->index].stats.desc_full_awake++;
 		return NETDEV_TX_BUSY;
 	}
 
-	enic_queue_wq_skb(enic, wq, skb);
+	if (enic_queue_wq_skb(enic, wq, skb))
+		goto error;
 
-	if (vnic_wq_desc_avail(wq) < MAX_SKB_FRAGS + ENIC_DESC_MAX_SPLITS)
+	if (vnic_wq_desc_avail(wq) < MAX_SKB_FRAGS + ENIC_DESC_MAX_SPLITS) {
 		netif_tx_stop_queue(txq);
-	if (!skb->xmit_more || netif_xmit_stopped(txq))
+		enic->wq[wq->index].stats.stopped++;
+	}
+	skb_tx_timestamp(skb);
+	if (!netdev_xmit_more() || netif_xmit_stopped(txq))
 		vnic_wq_doorbell(wq);
 
-	spin_unlock(&enic->wq_lock[txq_map]);
+error:
+	spin_unlock(&enic->wq[txq_map].lock);
 
 	return NETDEV_TX_OK;
 }
 
-/* dev_base_lock rwlock held, nominally process context */
+/* rcu_read_lock potentially held, nominally process context */
 static void enic_get_stats(struct net_device *netdev,
 			   struct rtnl_link_stats64 *net_stats)
 {
 	struct enic *enic = netdev_priv(netdev);
 	struct vnic_stats *stats;
+	u64 pkt_truncated = 0;
+	u64 bad_fcs = 0;
 	int err;
+	int i;
 
 	err = enic_dev_stats_dump(enic, &stats);
-	/* return only when pci_zalloc_consistent fails in vnic_dev_stats_dump
+	/* return only when dma_alloc_coherent fails in vnic_dev_stats_dump
 	 * For other failures, like devcmd failure, we return previously
 	 * recorded stats.
 	 */
@@ -889,8 +919,17 @@ static void enic_get_stats(struct net_device *netdev,
 	net_stats->rx_bytes = stats->rx.rx_bytes_ok;
 	net_stats->rx_errors = stats->rx.rx_errors;
 	net_stats->multicast = stats->rx.rx_multicast_frames_ok;
-	net_stats->rx_over_errors = enic->rq_truncated_pkts;
-	net_stats->rx_crc_errors = enic->rq_bad_fcs;
+
+	for (i = 0; i < enic->rq_count; i++) {
+		struct enic_rq_stats *rqs = &enic->rq[i].stats;
+
+		if (!enic->rq[i].vrq.ctrl)
+			break;
+		pkt_truncated += rqs->pkt_truncated;
+		bad_fcs += rqs->bad_fcs;
+	}
+	net_stats->rx_over_errors = pkt_truncated;
+	net_stats->rx_crc_errors = bad_fcs;
 	net_stats->rx_dropped = stats->rx.rx_no_bufs + stats->rx.rx_drop;
 }
 
@@ -976,7 +1015,7 @@ static int enic_set_mac_addr(struct net_device *netdev, char *addr)
 			return -EADDRNOTAVAIL;
 	}
 
-	memcpy(netdev->dev_addr, addr, netdev->addr_len);
+	eth_hw_addr_set(netdev, addr);
 
 	return 0;
 }
@@ -1054,7 +1093,7 @@ static void enic_set_rx_mode(struct net_device *netdev)
 }
 
 /* netif_tx_lock held, BHs disabled */
-static void enic_tx_timeout(struct net_device *netdev)
+static void enic_tx_timeout(struct net_device *netdev, unsigned int txqueue)
 {
 	struct enic *enic = netdev_priv(netdev);
 	schedule_work(&enic->tx_hang_reset);
@@ -1089,6 +1128,7 @@ static int enic_set_vf_mac(struct net_device *netdev, int vf, u8 *mac)
 static int enic_set_vf_port(struct net_device *netdev, int vf,
 	struct nlattr *port[])
 {
+	static const u8 zero_addr[ETH_ALEN] = {};
 	struct enic *enic = netdev_priv(netdev);
 	struct enic_port_profile prev_pp;
 	struct enic_port_profile *pp;
@@ -1108,18 +1148,30 @@ static int enic_set_vf_port(struct net_device *netdev, int vf,
 	pp->request = nla_get_u8(port[IFLA_PORT_REQUEST]);
 
 	if (port[IFLA_PORT_PROFILE]) {
+		if (nla_len(port[IFLA_PORT_PROFILE]) != PORT_PROFILE_MAX) {
+			memcpy(pp, &prev_pp, sizeof(*pp));
+			return -EINVAL;
+		}
 		pp->set |= ENIC_SET_NAME;
 		memcpy(pp->name, nla_data(port[IFLA_PORT_PROFILE]),
 			PORT_PROFILE_MAX);
 	}
 
 	if (port[IFLA_PORT_INSTANCE_UUID]) {
+		if (nla_len(port[IFLA_PORT_INSTANCE_UUID]) != PORT_UUID_MAX) {
+			memcpy(pp, &prev_pp, sizeof(*pp));
+			return -EINVAL;
+		}
 		pp->set |= ENIC_SET_INSTANCE;
 		memcpy(pp->instance_uuid,
 			nla_data(port[IFLA_PORT_INSTANCE_UUID]), PORT_UUID_MAX);
 	}
 
 	if (port[IFLA_PORT_HOST_UUID]) {
+		if (nla_len(port[IFLA_PORT_HOST_UUID]) != PORT_UUID_MAX) {
+			memcpy(pp, &prev_pp, sizeof(*pp));
+			return -EINVAL;
+		}
 		pp->set |= ENIC_SET_HOST;
 		memcpy(pp->host_uuid,
 			nla_data(port[IFLA_PORT_HOST_UUID]), PORT_UUID_MAX);
@@ -1153,7 +1205,7 @@ static int enic_set_vf_port(struct net_device *netdev, int vf,
 		} else {
 			memset(pp, 0, sizeof(*pp));
 			if (vf == PORT_SELF_VF)
-				eth_zero_addr(netdev->dev_addr);
+				eth_hw_addr_set(netdev, zero_addr);
 		}
 	} else {
 		/* Set flag to indicate that the port assoc/disassoc
@@ -1165,7 +1217,7 @@ static int enic_set_vf_port(struct net_device *netdev, int vf,
 		if (pp->request == PORT_REQUEST_DISASSOCIATE) {
 			eth_zero_addr(pp->mac_addr);
 			if (vf == PORT_SELF_VF)
-				eth_zero_addr(netdev->dev_addr);
+				eth_hw_addr_set(netdev, zero_addr);
 		}
 	}
 
@@ -1210,229 +1262,6 @@ nla_put_failure:
 	return -EMSGSIZE;
 }
 
-static void enic_free_rq_buf(struct vnic_rq *rq, struct vnic_rq_buf *buf)
-{
-	struct enic *enic = vnic_dev_priv(rq->vdev);
-
-	if (!buf->os_buf)
-		return;
-
-	pci_unmap_single(enic->pdev, buf->dma_addr,
-		buf->len, PCI_DMA_FROMDEVICE);
-	dev_kfree_skb_any(buf->os_buf);
-	buf->os_buf = NULL;
-}
-
-static int enic_rq_alloc_buf(struct vnic_rq *rq)
-{
-	struct enic *enic = vnic_dev_priv(rq->vdev);
-	struct net_device *netdev = enic->netdev;
-	struct sk_buff *skb;
-	unsigned int len = netdev->mtu + VLAN_ETH_HLEN;
-	unsigned int os_buf_index = 0;
-	dma_addr_t dma_addr;
-	struct vnic_rq_buf *buf = rq->to_use;
-
-	if (buf->os_buf) {
-		enic_queue_rq_desc(rq, buf->os_buf, os_buf_index, buf->dma_addr,
-				   buf->len);
-
-		return 0;
-	}
-	skb = netdev_alloc_skb_ip_align(netdev, len);
-	if (!skb)
-		return -ENOMEM;
-
-	dma_addr = pci_map_single(enic->pdev, skb->data, len,
-				  PCI_DMA_FROMDEVICE);
-	if (unlikely(enic_dma_map_check(enic, dma_addr))) {
-		dev_kfree_skb(skb);
-		return -ENOMEM;
-	}
-
-	enic_queue_rq_desc(rq, skb, os_buf_index,
-		dma_addr, len);
-
-	return 0;
-}
-
-static void enic_intr_update_pkt_size(struct vnic_rx_bytes_counter *pkt_size,
-				      u32 pkt_len)
-{
-	if (ENIC_LARGE_PKT_THRESHOLD <= pkt_len)
-		pkt_size->large_pkt_bytes_cnt += pkt_len;
-	else
-		pkt_size->small_pkt_bytes_cnt += pkt_len;
-}
-
-static bool enic_rxcopybreak(struct net_device *netdev, struct sk_buff **skb,
-			     struct vnic_rq_buf *buf, u16 len)
-{
-	struct enic *enic = netdev_priv(netdev);
-	struct sk_buff *new_skb;
-
-	if (len > enic->rx_copybreak)
-		return false;
-	new_skb = netdev_alloc_skb_ip_align(netdev, len);
-	if (!new_skb)
-		return false;
-	pci_dma_sync_single_for_cpu(enic->pdev, buf->dma_addr, len,
-				    DMA_FROM_DEVICE);
-	memcpy(new_skb->data, (*skb)->data, len);
-	*skb = new_skb;
-
-	return true;
-}
-
-static void enic_rq_indicate_buf(struct vnic_rq *rq,
-	struct cq_desc *cq_desc, struct vnic_rq_buf *buf,
-	int skipped, void *opaque)
-{
-	struct enic *enic = vnic_dev_priv(rq->vdev);
-	struct net_device *netdev = enic->netdev;
-	struct sk_buff *skb;
-	struct vnic_cq *cq = &enic->cq[enic_cq_rq(enic, rq->index)];
-
-	u8 type, color, eop, sop, ingress_port, vlan_stripped;
-	u8 fcoe, fcoe_sof, fcoe_fc_crc_ok, fcoe_enc_error, fcoe_eof;
-	u8 tcp_udp_csum_ok, udp, tcp, ipv4_csum_ok;
-	u8 ipv6, ipv4, ipv4_fragment, fcs_ok, rss_type, csum_not_calc;
-	u8 packet_error;
-	u16 q_number, completed_index, bytes_written, vlan_tci, checksum;
-	u32 rss_hash;
-	bool outer_csum_ok = true, encap = false;
-
-	if (skipped)
-		return;
-
-	skb = buf->os_buf;
-
-	cq_enet_rq_desc_dec((struct cq_enet_rq_desc *)cq_desc,
-		&type, &color, &q_number, &completed_index,
-		&ingress_port, &fcoe, &eop, &sop, &rss_type,
-		&csum_not_calc, &rss_hash, &bytes_written,
-		&packet_error, &vlan_stripped, &vlan_tci, &checksum,
-		&fcoe_sof, &fcoe_fc_crc_ok, &fcoe_enc_error,
-		&fcoe_eof, &tcp_udp_csum_ok, &udp, &tcp,
-		&ipv4_csum_ok, &ipv6, &ipv4, &ipv4_fragment,
-		&fcs_ok);
-
-	if (packet_error) {
-
-		if (!fcs_ok) {
-			if (bytes_written > 0)
-				enic->rq_bad_fcs++;
-			else if (bytes_written == 0)
-				enic->rq_truncated_pkts++;
-		}
-
-		pci_unmap_single(enic->pdev, buf->dma_addr, buf->len,
-				 PCI_DMA_FROMDEVICE);
-		dev_kfree_skb_any(skb);
-		buf->os_buf = NULL;
-
-		return;
-	}
-
-	if (eop && bytes_written > 0) {
-
-		/* Good receive
-		 */
-
-		if (!enic_rxcopybreak(netdev, &skb, buf, bytes_written)) {
-			buf->os_buf = NULL;
-			pci_unmap_single(enic->pdev, buf->dma_addr, buf->len,
-					 PCI_DMA_FROMDEVICE);
-		}
-		prefetch(skb->data - NET_IP_ALIGN);
-
-		skb_put(skb, bytes_written);
-		skb->protocol = eth_type_trans(skb, netdev);
-		skb_record_rx_queue(skb, q_number);
-		if ((netdev->features & NETIF_F_RXHASH) && rss_hash &&
-		    (type == 3)) {
-			switch (rss_type) {
-			case CQ_ENET_RQ_DESC_RSS_TYPE_TCP_IPv4:
-			case CQ_ENET_RQ_DESC_RSS_TYPE_TCP_IPv6:
-			case CQ_ENET_RQ_DESC_RSS_TYPE_TCP_IPv6_EX:
-				skb_set_hash(skb, rss_hash, PKT_HASH_TYPE_L4);
-				break;
-			case CQ_ENET_RQ_DESC_RSS_TYPE_IPv4:
-			case CQ_ENET_RQ_DESC_RSS_TYPE_IPv6:
-			case CQ_ENET_RQ_DESC_RSS_TYPE_IPv6_EX:
-				skb_set_hash(skb, rss_hash, PKT_HASH_TYPE_L3);
-				break;
-			}
-		}
-		if (enic->vxlan.vxlan_udp_port_number) {
-			switch (enic->vxlan.patch_level) {
-			case 0:
-				if (fcoe) {
-					encap = true;
-					outer_csum_ok = fcoe_fc_crc_ok;
-				}
-				break;
-			case 2:
-				if ((type == 7) &&
-				    (rss_hash & BIT(0))) {
-					encap = true;
-					outer_csum_ok = (rss_hash & BIT(1)) &&
-							(rss_hash & BIT(2));
-				}
-				break;
-			}
-		}
-
-		/* Hardware does not provide whole packet checksum. It only
-		 * provides pseudo checksum. Since hw validates the packet
-		 * checksum but not provide us the checksum value. use
-		 * CHECSUM_UNNECESSARY.
-		 *
-		 * In case of encap pkt tcp_udp_csum_ok/tcp_udp_csum_ok is
-		 * inner csum_ok. outer_csum_ok is set by hw when outer udp
-		 * csum is correct or is zero.
-		 */
-		if ((netdev->features & NETIF_F_RXCSUM) && !csum_not_calc &&
-		    tcp_udp_csum_ok && ipv4_csum_ok && outer_csum_ok) {
-			skb->ip_summed = CHECKSUM_UNNECESSARY;
-			skb->csum_level = encap;
-		}
-
-		if (vlan_stripped)
-			__vlan_hwaccel_put_tag(skb, htons(ETH_P_8021Q), vlan_tci);
-
-		skb_mark_napi_id(skb, &enic->napi[rq->index]);
-		if (!(netdev->features & NETIF_F_GRO))
-			netif_receive_skb(skb);
-		else
-			napi_gro_receive(&enic->napi[q_number], skb);
-		if (enic->rx_coalesce_setting.use_adaptive_rx_coalesce)
-			enic_intr_update_pkt_size(&cq->pkt_size_counter,
-						  bytes_written);
-	} else {
-
-		/* Buffer overflow
-		 */
-
-		pci_unmap_single(enic->pdev, buf->dma_addr, buf->len,
-				 PCI_DMA_FROMDEVICE);
-		dev_kfree_skb_any(skb);
-		buf->os_buf = NULL;
-	}
-}
-
-static int enic_rq_service(struct vnic_dev *vdev, struct cq_desc *cq_desc,
-	u8 type, u16 q_number, u16 completed_index, void *opaque)
-{
-	struct enic *enic = vnic_dev_priv(vdev);
-
-	vnic_rq_service(&enic->rq[q_number], cq_desc,
-		completed_index, VNIC_RQ_RETURN_DESC,
-		enic_rq_indicate_buf, opaque);
-
-	return 0;
-}
-
 static void enic_set_int_moderation(struct enic *enic, struct vnic_rq *rq)
 {
 	unsigned int intr = enic_msix_rq_intr(enic, rq->index);
@@ -1497,18 +1326,16 @@ static int enic_poll(struct napi_struct *napi, int budget)
 	struct enic *enic = netdev_priv(netdev);
 	unsigned int cq_rq = enic_cq_rq(enic, 0);
 	unsigned int cq_wq = enic_cq_wq(enic, 0);
-	unsigned int intr = enic_legacy_io_intr();
+	unsigned int intr = ENIC_LEGACY_IO_INTR;
 	unsigned int rq_work_to_do = budget;
-	unsigned int wq_work_to_do = -1; /* no limit */
+	unsigned int wq_work_to_do = ENIC_WQ_NAPI_BUDGET;
 	unsigned int  work_done, rq_work_done = 0, wq_work_done;
 	int err;
 
-	wq_work_done = vnic_cq_service(&enic->cq[cq_wq], wq_work_to_do,
-				       enic_wq_service, NULL);
+	wq_work_done = enic_wq_cq_service(enic, cq_wq, wq_work_to_do);
 
 	if (budget > 0)
-		rq_work_done = vnic_cq_service(&enic->cq[cq_rq],
-			rq_work_to_do, enic_rq_service, NULL);
+		rq_work_done = enic_rq_cq_service(enic, cq_rq, rq_work_to_do);
 
 	/* Accumulate intr event credits for this polling
 	 * cycle.  An intr event is the completion of a
@@ -1523,7 +1350,7 @@ static int enic_poll(struct napi_struct *napi, int budget)
 			0 /* don't unmask intr */,
 			0 /* don't reset intr timer */);
 
-	err = vnic_rq_fill(&enic->rq[0], enic_rq_alloc_buf);
+	err = vnic_rq_fill(&enic->rq[0].vrq, enic_rq_alloc_buf);
 
 	/* Buffer allocation failed. Stay in polling
 	 * mode so we can try to fill the ring again.
@@ -1535,7 +1362,7 @@ static int enic_poll(struct napi_struct *napi, int budget)
 		/* Call the function which refreshes the intr coalescing timer
 		 * value based on the traffic.
 		 */
-		enic_calc_int_moderation(enic, &enic->rq[0]);
+		enic_calc_int_moderation(enic, &enic->rq[0].vrq);
 
 	if ((rq_work_done < budget) && napi_complete_done(napi, rq_work_done)) {
 
@@ -1544,8 +1371,11 @@ static int enic_poll(struct napi_struct *napi, int budget)
 		 */
 
 		if (enic->rx_coalesce_setting.use_adaptive_rx_coalesce)
-			enic_set_int_moderation(enic, &enic->rq[0]);
+			enic_set_int_moderation(enic, &enic->rq[0].vrq);
 		vnic_intr_unmask(&enic->intr[intr]);
+		enic->rq[0].stats.napi_complete++;
+	} else {
+		enic->rq[0].stats.napi_repoll++;
 	}
 
 	return rq_work_done;
@@ -1594,18 +1424,18 @@ static int enic_poll_msix_wq(struct napi_struct *napi, int budget)
 	struct net_device *netdev = napi->dev;
 	struct enic *enic = netdev_priv(netdev);
 	unsigned int wq_index = (napi - &enic->napi[0]) - enic->rq_count;
-	struct vnic_wq *wq = &enic->wq[wq_index];
+	struct vnic_wq *wq = &enic->wq[wq_index].vwq;
 	unsigned int cq;
 	unsigned int intr;
-	unsigned int wq_work_to_do = -1; /* clean all desc possible */
+	unsigned int wq_work_to_do = ENIC_WQ_NAPI_BUDGET;
 	unsigned int wq_work_done;
 	unsigned int wq_irq;
 
 	wq_irq = wq->index;
 	cq = enic_cq_wq(enic, wq_irq);
 	intr = enic_msix_wq_intr(enic, wq_irq);
-	wq_work_done = vnic_cq_service(&enic->cq[cq], wq_work_to_do,
-				       enic_wq_service, NULL);
+
+	wq_work_done = enic_wq_cq_service(enic, cq, wq_work_to_do);
 
 	vnic_intr_return_credits(&enic->intr[intr], wq_work_done,
 				 0 /* don't unmask intr */,
@@ -1634,8 +1464,7 @@ static int enic_poll_msix_rq(struct napi_struct *napi, int budget)
 	 */
 
 	if (budget > 0)
-		work_done = vnic_cq_service(&enic->cq[cq],
-			work_to_do, enic_rq_service, NULL);
+		work_done = enic_rq_cq_service(enic, cq, work_to_do);
 
 	/* Return intr event credits for this polling
 	 * cycle.  An intr event is the completion of a
@@ -1648,7 +1477,7 @@ static int enic_poll_msix_rq(struct napi_struct *napi, int budget)
 			0 /* don't unmask intr */,
 			0 /* don't reset intr timer */);
 
-	err = vnic_rq_fill(&enic->rq[rq], enic_rq_alloc_buf);
+	err = vnic_rq_fill(&enic->rq[rq].vrq, enic_rq_alloc_buf);
 
 	/* Buffer allocation failed. Stay in polling mode
 	 * so we can try to fill the ring again.
@@ -1660,7 +1489,7 @@ static int enic_poll_msix_rq(struct napi_struct *napi, int budget)
 		/* Call the function which refreshes the intr coalescing timer
 		 * value based on the traffic.
 		 */
-		enic_calc_int_moderation(enic, &enic->rq[rq]);
+		enic_calc_int_moderation(enic, &enic->rq[rq].vrq);
 
 	if ((work_done < budget) && napi_complete_done(napi, work_done)) {
 
@@ -1669,16 +1498,19 @@ static int enic_poll_msix_rq(struct napi_struct *napi, int budget)
 		 */
 
 		if (enic->rx_coalesce_setting.use_adaptive_rx_coalesce)
-			enic_set_int_moderation(enic, &enic->rq[rq]);
+			enic_set_int_moderation(enic, &enic->rq[rq].vrq);
 		vnic_intr_unmask(&enic->intr[intr]);
+		enic->rq[rq].stats.napi_complete++;
+	} else {
+		enic->rq[rq].stats.napi_repoll++;
 	}
 
 	return work_done;
 }
 
-static void enic_notify_timer(unsigned long data)
+static void enic_notify_timer(struct timer_list *t)
 {
-	struct enic *enic = (struct enic *)data;
+	struct enic *enic = timer_container_of(enic, t, notify_timer);
 
 	enic_notify_check(enic);
 
@@ -1700,7 +1532,7 @@ static void enic_free_intr(struct enic *enic)
 		free_irq(enic->pdev->irq, enic);
 		break;
 	case VNIC_DEV_INTR_MODE_MSIX:
-		for (i = 0; i < ARRAY_SIZE(enic->msix); i++)
+		for (i = 0; i < enic->intr_count; i++)
 			if (enic->msix[i].requested)
 				free_irq(enic->msix_entry[i].vector,
 					enic->msix[i].devid);
@@ -1737,7 +1569,7 @@ static int enic_request_intr(struct enic *enic)
 			intr = enic_msix_rq_intr(enic, i);
 			snprintf(enic->msix[intr].devname,
 				sizeof(enic->msix[intr].devname),
-				"%.11s-rx-%u", netdev->name, i);
+				"%s-rx-%u", netdev->name, i);
 			enic->msix[intr].isr = enic_isr_msix;
 			enic->msix[intr].devid = &enic->napi[i];
 		}
@@ -1748,7 +1580,7 @@ static int enic_request_intr(struct enic *enic)
 			intr = enic_msix_wq_intr(enic, i);
 			snprintf(enic->msix[intr].devname,
 				sizeof(enic->msix[intr].devname),
-				"%.11s-tx-%u", netdev->name, i);
+				"%s-tx-%u", netdev->name, i);
 			enic->msix[intr].isr = enic_isr_msix;
 			enic->msix[intr].devid = &enic->napi[wq];
 		}
@@ -1756,18 +1588,18 @@ static int enic_request_intr(struct enic *enic)
 		intr = enic_msix_err_intr(enic);
 		snprintf(enic->msix[intr].devname,
 			sizeof(enic->msix[intr].devname),
-			"%.11s-err", netdev->name);
+			"%s-err", netdev->name);
 		enic->msix[intr].isr = enic_isr_msix_err;
 		enic->msix[intr].devid = enic;
 
 		intr = enic_msix_notify_intr(enic);
 		snprintf(enic->msix[intr].devname,
 			sizeof(enic->msix[intr].devname),
-			"%.11s-notify", netdev->name);
+			"%s-notify", netdev->name);
 		enic->msix[intr].isr = enic_isr_msix_notify;
 		enic->msix[intr].devid = enic;
 
-		for (i = 0; i < ARRAY_SIZE(enic->msix); i++)
+		for (i = 0; i < enic->intr_count; i++)
 			enic->msix[i].requested = 0;
 
 		for (i = 0; i < enic->intr_count; i++) {
@@ -1809,36 +1641,6 @@ static void enic_synchronize_irqs(struct enic *enic)
 	}
 }
 
-static void enic_set_rx_coal_setting(struct enic *enic)
-{
-	unsigned int speed;
-	int index = -1;
-	struct enic_rx_coal *rx_coal = &enic->rx_coalesce_setting;
-
-	/* 1. Read the link speed from fw
-	 * 2. Pick the default range for the speed
-	 * 3. Update it in enic->rx_coalesce_setting
-	 */
-	speed = vnic_dev_port_speed(enic->vdev);
-	if (ENIC_LINK_SPEED_10G < speed)
-		index = ENIC_LINK_40G_INDEX;
-	else if (ENIC_LINK_SPEED_4G < speed)
-		index = ENIC_LINK_10G_INDEX;
-	else
-		index = ENIC_LINK_4G_INDEX;
-
-	rx_coal->small_pkt_range_start = mod_range[index].small_pkt_range_start;
-	rx_coal->large_pkt_range_start = mod_range[index].large_pkt_range_start;
-	rx_coal->range_end = ENIC_RX_COALESCE_RANGE_END;
-
-	/* Start with the value provided by UCSM */
-	for (index = 0; index < enic->rq_count; index++)
-		enic->cq[index].cur_rx_coal_timeval =
-				enic->config.intr_timer_usec;
-
-	rx_coal->use_adaptive_rx_coalesce = 1;
-}
-
 static int enic_dev_notify_set(struct enic *enic)
 {
 	int err;
@@ -1846,8 +1648,7 @@ static int enic_dev_notify_set(struct enic *enic)
 	spin_lock_bh(&enic->devcmd_lock);
 	switch (vnic_dev_get_intr_mode(enic->vdev)) {
 	case VNIC_DEV_INTR_MODE_INTX:
-		err = vnic_dev_notify_set(enic->vdev,
-			enic_legacy_notify_intr());
+		err = vnic_dev_notify_set(enic->vdev, ENIC_LEGACY_NOTIFY_INTR);
 		break;
 	case VNIC_DEV_INTR_MODE_MSIX:
 		err = vnic_dev_notify_set(enic->vdev,
@@ -1879,7 +1680,18 @@ static int enic_open(struct net_device *netdev)
 {
 	struct enic *enic = netdev_priv(netdev);
 	unsigned int i;
-	int err;
+	int err, ret;
+	unsigned int max_pkt_len = netdev->mtu + VLAN_ETH_HLEN;
+	struct page_pool_params pp_params = {
+		.order = get_order(max_pkt_len),
+		.pool_size = enic->config.rq_desc_count,
+		.nid = dev_to_node(&enic->pdev->dev),
+		.dev = &enic->pdev->dev,
+		.dma_dir = DMA_FROM_DEVICE,
+		.max_len = (max_pkt_len > PAGE_SIZE) ? max_pkt_len : PAGE_SIZE,
+		.netdev = netdev,
+		.flags = PP_FLAG_DMA_MAP | PP_FLAG_DMA_SYNC_DEV,
+	};
 
 	err = enic_request_intr(enic);
 	if (err) {
@@ -1897,9 +1709,21 @@ static int enic_open(struct net_device *netdev)
 	}
 
 	for (i = 0; i < enic->rq_count; i++) {
-		vnic_rq_fill(&enic->rq[i], enic_rq_alloc_buf);
+		/* create a page pool for each RQ */
+		pp_params.napi = &enic->napi[i];
+		pp_params.queue_idx = i;
+		enic->rq[i].pool = page_pool_create(&pp_params);
+		if (IS_ERR(enic->rq[i].pool)) {
+			err = PTR_ERR(enic->rq[i].pool);
+			enic->rq[i].pool = NULL;
+			goto err_out_free_rq;
+		}
+
+		/* enable rq before updating rq desc */
+		vnic_rq_enable(&enic->rq[i].vrq);
+		vnic_rq_fill(&enic->rq[i].vrq, enic_rq_alloc_buf);
 		/* Need at least one buffer on ring to get going */
-		if (vnic_rq_desc_used(&enic->rq[i]) == 0) {
+		if (vnic_rq_desc_used(&enic->rq[i].vrq) == 0) {
 			netdev_err(netdev, "Unable to alloc receive buffers\n");
 			err = -ENOMEM;
 			goto err_out_free_rq;
@@ -1907,9 +1731,7 @@ static int enic_open(struct net_device *netdev)
 	}
 
 	for (i = 0; i < enic->wq_count; i++)
-		vnic_wq_enable(&enic->wq[i]);
-	for (i = 0; i < enic->rq_count; i++)
-		vnic_rq_enable(&enic->rq[i]);
+		vnic_wq_enable(&enic->wq[i].vwq);
 
 	if (!enic_is_dynamic(enic) && !enic_is_sriov_vf(enic))
 		enic_dev_add_station_addr(enic);
@@ -1930,13 +1752,19 @@ static int enic_open(struct net_device *netdev)
 		vnic_intr_unmask(&enic->intr[i]);
 
 	enic_notify_timer_start(enic);
-	enic_rfs_flw_tbl_init(enic);
+	enic_rfs_timer_start(enic);
 
 	return 0;
 
 err_out_free_rq:
-	for (i = 0; i < enic->rq_count; i++)
-		vnic_rq_clean(&enic->rq[i], enic_free_rq_buf);
+	for (i = 0; i < enic->rq_count; i++) {
+		ret = vnic_rq_disable(&enic->rq[i].vrq);
+		if (!ret) {
+			vnic_rq_clean(&enic->rq[i].vrq, enic_free_rq_buf);
+			page_pool_destroy(enic->rq[i].pool);
+			enic->rq[i].pool = NULL;
+		}
+	}
 	enic_dev_notify_unset(enic);
 err_out_free_intr:
 	enic_unset_affinity_hint(enic);
@@ -1959,7 +1787,7 @@ static int enic_stop(struct net_device *netdev)
 
 	enic_synchronize_irqs(enic);
 
-	del_timer_sync(&enic->notify_timer);
+	timer_delete_sync(&enic->notify_timer);
 	enic_rfs_flw_tbl_free(enic);
 
 	enic_dev_disable(enic);
@@ -1968,21 +1796,21 @@ static int enic_stop(struct net_device *netdev)
 		napi_disable(&enic->napi[i]);
 
 	netif_carrier_off(netdev);
-	netif_tx_disable(netdev);
 	if (vnic_dev_get_intr_mode(enic->vdev) == VNIC_DEV_INTR_MODE_MSIX)
 		for (i = 0; i < enic->wq_count; i++)
 			napi_disable(&enic->napi[enic_cq_wq(enic, i)]);
+	netif_tx_disable(netdev);
 
 	if (!enic_is_dynamic(enic) && !enic_is_sriov_vf(enic))
 		enic_dev_del_station_addr(enic);
 
 	for (i = 0; i < enic->wq_count; i++) {
-		err = vnic_wq_disable(&enic->wq[i]);
+		err = vnic_wq_disable(&enic->wq[i].vwq);
 		if (err)
 			return err;
 	}
 	for (i = 0; i < enic->rq_count; i++) {
-		err = vnic_rq_disable(&enic->rq[i]);
+		err = vnic_rq_disable(&enic->rq[i].vrq);
 		if (err)
 			return err;
 	}
@@ -1992,9 +1820,12 @@ static int enic_stop(struct net_device *netdev)
 	enic_free_intr(enic);
 
 	for (i = 0; i < enic->wq_count; i++)
-		vnic_wq_clean(&enic->wq[i], enic_free_wq_buf);
-	for (i = 0; i < enic->rq_count; i++)
-		vnic_rq_clean(&enic->rq[i], enic_free_rq_buf);
+		vnic_wq_clean(&enic->wq[i].vwq, enic_free_wq_buf);
+	for (i = 0; i < enic->rq_count; i++) {
+		vnic_rq_clean(&enic->rq[i].vrq, enic_free_rq_buf);
+		page_pool_destroy(enic->rq[i].pool);
+		enic->rq[i].pool = NULL;
+	}
 	for (i = 0; i < enic->cq_count; i++)
 		vnic_cq_clean(&enic->cq[i]);
 	for (i = 0; i < enic->intr_count; i++)
@@ -2003,28 +1834,42 @@ static int enic_stop(struct net_device *netdev)
 	return 0;
 }
 
+static int _enic_change_mtu(struct net_device *netdev, int new_mtu)
+{
+	bool running = netif_running(netdev);
+	int err = 0;
+
+	ASSERT_RTNL();
+	if (running) {
+		err = enic_stop(netdev);
+		if (err)
+			return err;
+	}
+
+	WRITE_ONCE(netdev->mtu, new_mtu);
+
+	if (running) {
+		err = enic_open(netdev);
+		if (err)
+			return err;
+	}
+
+	return 0;
+}
+
 static int enic_change_mtu(struct net_device *netdev, int new_mtu)
 {
 	struct enic *enic = netdev_priv(netdev);
-	int running = netif_running(netdev);
 
 	if (enic_is_dynamic(enic) || enic_is_sriov_vf(enic))
 		return -EOPNOTSUPP;
 
-	if (running)
-		enic_stop(netdev);
-
-	netdev->mtu = new_mtu;
-
-	if (netdev->mtu > enic->port_mtu)
+	if (new_mtu > enic->port_mtu)
 		netdev_warn(netdev,
-			"interface MTU (%d) set higher than port MTU (%d)\n",
-			netdev->mtu, enic->port_mtu);
-
-	if (running)
-		enic_open(netdev);
+			    "interface MTU (%d) set higher than port MTU (%d)\n",
+			    new_mtu, enic->port_mtu);
 
-	return 0;
+	return _enic_change_mtu(netdev, new_mtu);
 }
 
 static void enic_change_mtu_work(struct work_struct *work)
@@ -2032,47 +1877,9 @@ static void enic_change_mtu_work(struct work_struct *work)
 	struct enic *enic = container_of(work, struct enic, change_mtu_work);
 	struct net_device *netdev = enic->netdev;
 	int new_mtu = vnic_dev_mtu(enic->vdev);
-	int err;
-	unsigned int i;
-
-	new_mtu = max_t(int, ENIC_MIN_MTU, min_t(int, ENIC_MAX_MTU, new_mtu));
 
 	rtnl_lock();
-
-	/* Stop RQ */
-	del_timer_sync(&enic->notify_timer);
-
-	for (i = 0; i < enic->rq_count; i++)
-		napi_disable(&enic->napi[i]);
-
-	vnic_intr_mask(&enic->intr[0]);
-	enic_synchronize_irqs(enic);
-	err = vnic_rq_disable(&enic->rq[0]);
-	if (err) {
-		rtnl_unlock();
-		netdev_err(netdev, "Unable to disable RQ.\n");
-		return;
-	}
-	vnic_rq_clean(&enic->rq[0], enic_free_rq_buf);
-	vnic_cq_clean(&enic->cq[0]);
-	vnic_intr_clean(&enic->intr[0]);
-
-	/* Fill RQ with new_mtu-sized buffers */
-	netdev->mtu = new_mtu;
-	vnic_rq_fill(&enic->rq[0], enic_rq_alloc_buf);
-	/* Need at least one buffer on ring to get going */
-	if (vnic_rq_desc_used(&enic->rq[0]) == 0) {
-		rtnl_unlock();
-		netdev_err(netdev, "Unable to alloc receive buffers.\n");
-		return;
-	}
-
-	/* Start RQ */
-	vnic_rq_enable(&enic->rq[0]);
-	napi_enable(&enic->napi[0]);
-	vnic_intr_unmask(&enic->intr[0]);
-	enic_notify_timer_start(enic);
-
+	(void)_enic_change_mtu(netdev, new_mtu);
 	rtnl_unlock();
 
 	netdev_info(netdev, "interface MTU set as %d\n", netdev->mtu);
@@ -2121,8 +1928,6 @@ static int enic_dev_wait(struct vnic_dev *vdev,
 	int done;
 	int err;
 
-	BUG_ON(in_interrupt());
-
 	err = start(vdev, arg);
 	if (err)
 		return err;
@@ -2150,9 +1955,10 @@ static int enic_dev_wait(struct vnic_dev *vdev,
 static int enic_dev_open(struct enic *enic)
 {
 	int err;
+	u32 flags = CMD_OPENF_IG_DESCCACHE;
 
 	err = enic_dev_wait(enic->vdev, vnic_dev_open,
-		vnic_dev_open_done, 0);
+		vnic_dev_open_done, flags);
 	if (err)
 		dev_err(enic_get_dev(enic), "vNIC device open failed, err %d\n",
 			err);
@@ -2192,9 +1998,9 @@ int __enic_set_rsskey(struct enic *enic)
 	dma_addr_t rss_key_buf_pa;
 	int i, kidx, bidx, err;
 
-	rss_key_buf_va = pci_zalloc_consistent(enic->pdev,
-					       sizeof(union vnic_rss_key),
-					       &rss_key_buf_pa);
+	rss_key_buf_va = dma_alloc_coherent(&enic->pdev->dev,
+					    sizeof(union vnic_rss_key),
+					    &rss_key_buf_pa, GFP_ATOMIC);
 	if (!rss_key_buf_va)
 		return -ENOMEM;
 
@@ -2209,8 +2015,8 @@ int __enic_set_rsskey(struct enic *enic)
 		sizeof(union vnic_rss_key));
 	spin_unlock_bh(&enic->devcmd_lock);
 
-	pci_free_consistent(enic->pdev, sizeof(union vnic_rss_key),
-		rss_key_buf_va, rss_key_buf_pa);
+	dma_free_coherent(&enic->pdev->dev, sizeof(union vnic_rss_key),
+			  rss_key_buf_va, rss_key_buf_pa);
 
 	return err;
 }
@@ -2229,8 +2035,9 @@ static int enic_set_rsscpu(struct enic *enic, u8 rss_hash_bits)
 	unsigned int i;
 	int err;
 
-	rss_cpu_buf_va = pci_alloc_consistent(enic->pdev,
-		sizeof(union vnic_rss_cpu), &rss_cpu_buf_pa);
+	rss_cpu_buf_va = dma_alloc_coherent(&enic->pdev->dev,
+					    sizeof(union vnic_rss_cpu),
+					    &rss_cpu_buf_pa, GFP_ATOMIC);
 	if (!rss_cpu_buf_va)
 		return -ENOMEM;
 
@@ -2243,8 +2050,8 @@ static int enic_set_rsscpu(struct enic *enic, u8 rss_hash_bits)
 		sizeof(union vnic_rss_cpu));
 	spin_unlock_bh(&enic->devcmd_lock);
 
-	pci_free_consistent(enic->pdev, sizeof(union vnic_rss_cpu),
-		rss_cpu_buf_va, rss_cpu_buf_pa);
+	dma_free_coherent(&enic->pdev->dev, sizeof(union vnic_rss_cpu),
+			  rss_cpu_buf_va, rss_cpu_buf_pa);
 
 	return err;
 }
@@ -2274,14 +2081,24 @@ static int enic_set_rss_nic_cfg(struct enic *enic)
 {
 	struct device *dev = enic_get_dev(enic);
 	const u8 rss_default_cpu = 0;
-	const u8 rss_hash_type = NIC_CFG_RSS_HASH_TYPE_IPV4 |
-		NIC_CFG_RSS_HASH_TYPE_TCP_IPV4 |
-		NIC_CFG_RSS_HASH_TYPE_IPV6 |
-		NIC_CFG_RSS_HASH_TYPE_TCP_IPV6;
 	const u8 rss_hash_bits = 7;
 	const u8 rss_base_cpu = 0;
+	u8 rss_hash_type;
+	int res;
 	u8 rss_enable = ENIC_SETTING(enic, RSS) && (enic->rq_count > 1);
 
+	spin_lock_bh(&enic->devcmd_lock);
+	res = vnic_dev_capable_rss_hash_type(enic->vdev, &rss_hash_type);
+	spin_unlock_bh(&enic->devcmd_lock);
+	if (res) {
+		/* defaults for old adapters
+		 */
+		rss_hash_type = NIC_CFG_RSS_HASH_TYPE_IPV4	|
+				NIC_CFG_RSS_HASH_TYPE_TCP_IPV4	|
+				NIC_CFG_RSS_HASH_TYPE_IPV6	|
+				NIC_CFG_RSS_HASH_TYPE_TCP_IPV6;
+	}
+
 	if (rss_enable) {
 		if (!enic_set_rsskey(enic)) {
 			if (enic_set_rsscpu(enic, rss_hash_bits)) {
@@ -2299,6 +2116,13 @@ static int enic_set_rss_nic_cfg(struct enic *enic)
 		rss_hash_bits, rss_base_cpu, rss_enable);
 }
 
+static void enic_set_api_busy(struct enic *enic, bool busy)
+{
+	spin_lock(&enic->enic_api_lock);
+	enic->enic_api_busy = busy;
+	spin_unlock(&enic->enic_api_lock);
+}
+
 static void enic_reset(struct work_struct *work)
 {
 	struct enic *enic = container_of(work, struct enic, reset);
@@ -2308,15 +2132,21 @@ static void enic_reset(struct work_struct *work)
 
 	rtnl_lock();
 
-	spin_lock(&enic->enic_api_lock);
+	/* Stop any activity from infiniband */
+	enic_set_api_busy(enic, true);
+
 	enic_stop(enic->netdev);
 	enic_dev_soft_reset(enic);
 	enic_reset_addr_lists(enic);
 	enic_init_vnic_resources(enic);
 	enic_set_rss_nic_cfg(enic);
 	enic_dev_set_ig_vlan_rewrite_mode(enic);
+	enic_ext_cq(enic);
 	enic_open(enic->netdev);
-	spin_unlock(&enic->enic_api_lock);
+
+	/* Allow infiniband to fiddle with the device again */
+	enic_set_api_busy(enic, false);
+
 	call_netdevice_notifiers(NETDEV_REBOOT, enic->netdev);
 
 	rtnl_unlock();
@@ -2328,7 +2158,9 @@ static void enic_tx_hang_reset(struct work_struct *work)
 
 	rtnl_lock();
 
-	spin_lock(&enic->enic_api_lock);
+	/* Stop any activity from infiniband */
+	enic_set_api_busy(enic, true);
+
 	enic_dev_hang_notify(enic);
 	enic_stop(enic->netdev);
 	enic_dev_hang_reset(enic);
@@ -2336,8 +2168,12 @@ static void enic_tx_hang_reset(struct work_struct *work)
 	enic_init_vnic_resources(enic);
 	enic_set_rss_nic_cfg(enic);
 	enic_dev_set_ig_vlan_rewrite_mode(enic);
+	enic_ext_cq(enic);
 	enic_open(enic->netdev);
-	spin_unlock(&enic->enic_api_lock);
+
+	/* Allow infiniband to fiddle with the device again */
+	enic_set_api_busy(enic, false);
+
 	call_netdevice_notifiers(NETDEV_REBOOT, enic->netdev);
 
 	rtnl_unlock();
@@ -2345,112 +2181,56 @@ static void enic_tx_hang_reset(struct work_struct *work)
 
 static int enic_set_intr_mode(struct enic *enic)
 {
-	unsigned int n = min_t(unsigned int, enic->rq_count, ENIC_RQ_MAX);
-	unsigned int m = min_t(unsigned int, enic->wq_count, ENIC_WQ_MAX);
 	unsigned int i;
+	int num_intr;
 
 	/* Set interrupt mode (INTx, MSI, MSI-X) depending
 	 * on system capabilities.
 	 *
 	 * Try MSI-X first
-	 *
-	 * We need n RQs, m WQs, n+m CQs, and n+m+2 INTRs
-	 * (the second to last INTR is used for WQ/RQ errors)
-	 * (the last INTR is used for notifications)
 	 */
 
-	BUG_ON(ARRAY_SIZE(enic->msix_entry) < n + m + 2);
-	for (i = 0; i < n + m + 2; i++)
-		enic->msix_entry[i].entry = i;
-
-	/* Use multiple RQs if RSS is enabled
-	 */
-
-	if (ENIC_SETTING(enic, RSS) &&
-	    enic->config.intr_mode < 1 &&
-	    enic->rq_count >= n &&
-	    enic->wq_count >= m &&
-	    enic->cq_count >= n + m &&
-	    enic->intr_count >= n + m + 2) {
-
-		if (pci_enable_msix_range(enic->pdev, enic->msix_entry,
-					  n + m + 2, n + m + 2) > 0) {
-
-			enic->rq_count = n;
-			enic->wq_count = m;
-			enic->cq_count = n + m;
-			enic->intr_count = n + m + 2;
-
-			vnic_dev_set_intr_mode(enic->vdev,
-				VNIC_DEV_INTR_MODE_MSIX);
-
-			return 0;
-		}
-	}
-
 	if (enic->config.intr_mode < 1 &&
-	    enic->rq_count >= 1 &&
-	    enic->wq_count >= m &&
-	    enic->cq_count >= 1 + m &&
-	    enic->intr_count >= 1 + m + 2) {
-		if (pci_enable_msix_range(enic->pdev, enic->msix_entry,
-					  1 + m + 2, 1 + m + 2) > 0) {
-
-			enic->rq_count = 1;
-			enic->wq_count = m;
-			enic->cq_count = 1 + m;
-			enic->intr_count = 1 + m + 2;
-
+	    enic->intr_avail >= ENIC_MSIX_MIN_INTR) {
+		for (i = 0; i < enic->intr_avail; i++)
+			enic->msix_entry[i].entry = i;
+
+		num_intr = pci_enable_msix_range(enic->pdev, enic->msix_entry,
+						 ENIC_MSIX_MIN_INTR,
+						 enic->intr_avail);
+		if (num_intr > 0) {
 			vnic_dev_set_intr_mode(enic->vdev,
-				VNIC_DEV_INTR_MODE_MSIX);
-
+					       VNIC_DEV_INTR_MODE_MSIX);
+			enic->intr_avail = num_intr;
 			return 0;
 		}
 	}
 
 	/* Next try MSI
 	 *
-	 * We need 1 RQ, 1 WQ, 2 CQs, and 1 INTR
+	 * We need 1 INTR
 	 */
 
 	if (enic->config.intr_mode < 2 &&
-	    enic->rq_count >= 1 &&
-	    enic->wq_count >= 1 &&
-	    enic->cq_count >= 2 &&
-	    enic->intr_count >= 1 &&
+	    enic->intr_avail >= 1 &&
 	    !pci_enable_msi(enic->pdev)) {
-
-		enic->rq_count = 1;
-		enic->wq_count = 1;
-		enic->cq_count = 2;
-		enic->intr_count = 1;
-
+		enic->intr_avail = 1;
 		vnic_dev_set_intr_mode(enic->vdev, VNIC_DEV_INTR_MODE_MSI);
-
 		return 0;
 	}
 
 	/* Next try INTx
 	 *
-	 * We need 1 RQ, 1 WQ, 2 CQs, and 3 INTRs
+	 * We need 3 INTRs
 	 * (the first INTR is used for WQ/RQ)
 	 * (the second INTR is used for WQ/RQ errors)
 	 * (the last INTR is used for notifications)
 	 */
 
 	if (enic->config.intr_mode < 3 &&
-	    enic->rq_count >= 1 &&
-	    enic->wq_count >= 1 &&
-	    enic->cq_count >= 2 &&
-	    enic->intr_count >= 3) {
-
-		enic->rq_count = 1;
-		enic->wq_count = 1;
-		enic->cq_count = 2;
-		enic->intr_count = 3;
-
+	    enic->intr_avail >= 3) {
+		enic->intr_avail = 3;
 		vnic_dev_set_intr_mode(enic->vdev, VNIC_DEV_INTR_MODE_INTX);
-
 		return 0;
 	}
 
@@ -2475,6 +2255,127 @@ static void enic_clear_intr_mode(struct enic *enic)
 	vnic_dev_set_intr_mode(enic->vdev, VNIC_DEV_INTR_MODE_UNKNOWN);
 }
 
+static int enic_adjust_resources(struct enic *enic)
+{
+	unsigned int max_queues;
+	unsigned int rq_default;
+	unsigned int rq_avail;
+	unsigned int wq_avail;
+
+	if (enic->rq_avail < 1 || enic->wq_avail < 1 || enic->cq_avail < 2) {
+		dev_err(enic_get_dev(enic),
+			"Not enough resources available rq: %d wq: %d cq: %d\n",
+			enic->rq_avail, enic->wq_avail,
+			enic->cq_avail);
+		return -ENOSPC;
+	}
+
+	if (is_kdump_kernel()) {
+		dev_info(enic_get_dev(enic), "Running from within kdump kernel. Using minimal resources\n");
+		enic->rq_avail = 1;
+		enic->wq_avail = 1;
+		enic->config.rq_desc_count = ENIC_MIN_RQ_DESCS;
+		enic->config.wq_desc_count = ENIC_MIN_WQ_DESCS;
+		enic->config.mtu = min_t(u16, 1500, enic->config.mtu);
+	}
+
+	/* if RSS isn't set, then we can only use one RQ */
+	if (!ENIC_SETTING(enic, RSS))
+		enic->rq_avail = 1;
+
+	switch (vnic_dev_get_intr_mode(enic->vdev)) {
+	case VNIC_DEV_INTR_MODE_INTX:
+	case VNIC_DEV_INTR_MODE_MSI:
+		enic->rq_count = 1;
+		enic->wq_count = 1;
+		enic->cq_count = 2;
+		enic->intr_count = enic->intr_avail;
+		break;
+	case VNIC_DEV_INTR_MODE_MSIX:
+		/* Adjust the number of wqs/rqs/cqs/interrupts that will be
+		 * used based on which resource is the most constrained
+		 */
+		wq_avail = min(enic->wq_avail, ENIC_WQ_MAX);
+		rq_default = max(netif_get_num_default_rss_queues(),
+				 ENIC_RQ_MIN_DEFAULT);
+		rq_avail = min3(enic->rq_avail, ENIC_RQ_MAX, rq_default);
+		max_queues = min(enic->cq_avail,
+				 enic->intr_avail - ENIC_MSIX_RESERVED_INTR);
+		if (wq_avail + rq_avail <= max_queues) {
+			enic->rq_count = rq_avail;
+			enic->wq_count = wq_avail;
+		} else {
+			/* recalculate wq/rq count */
+			if (rq_avail < wq_avail) {
+				enic->rq_count = min(rq_avail, max_queues / 2);
+				enic->wq_count = max_queues - enic->rq_count;
+			} else {
+				enic->wq_count = min(wq_avail, max_queues / 2);
+				enic->rq_count = max_queues - enic->wq_count;
+			}
+		}
+		enic->cq_count = enic->rq_count + enic->wq_count;
+		enic->intr_count = enic->cq_count + ENIC_MSIX_RESERVED_INTR;
+
+		break;
+	default:
+		dev_err(enic_get_dev(enic), "Unknown interrupt mode\n");
+		return -EINVAL;
+	}
+
+	return 0;
+}
+
+static void enic_get_queue_stats_rx(struct net_device *dev, int idx,
+				    struct netdev_queue_stats_rx *rxs)
+{
+	struct enic *enic = netdev_priv(dev);
+	struct enic_rq_stats *rqstats = &enic->rq[idx].stats;
+
+	rxs->bytes = rqstats->bytes;
+	rxs->packets = rqstats->packets;
+	rxs->hw_drops = rqstats->bad_fcs + rqstats->pkt_truncated;
+	rxs->hw_drop_overruns = rqstats->pkt_truncated;
+	rxs->csum_unnecessary = rqstats->csum_unnecessary +
+				rqstats->csum_unnecessary_encap;
+	rxs->alloc_fail = rqstats->pp_alloc_fail;
+}
+
+static void enic_get_queue_stats_tx(struct net_device *dev, int idx,
+				    struct netdev_queue_stats_tx *txs)
+{
+	struct enic *enic = netdev_priv(dev);
+	struct enic_wq_stats *wqstats = &enic->wq[idx].stats;
+
+	txs->bytes = wqstats->bytes;
+	txs->packets = wqstats->packets;
+	txs->csum_none = wqstats->csum_none;
+	txs->needs_csum = wqstats->csum_partial + wqstats->encap_csum +
+			  wqstats->tso;
+	txs->hw_gso_packets = wqstats->tso;
+	txs->stop = wqstats->stopped;
+	txs->wake = wqstats->wake;
+}
+
+static void enic_get_base_stats(struct net_device *dev,
+				struct netdev_queue_stats_rx *rxs,
+				struct netdev_queue_stats_tx *txs)
+{
+	rxs->bytes = 0;
+	rxs->packets = 0;
+	rxs->hw_drops = 0;
+	rxs->hw_drop_overruns = 0;
+	rxs->csum_unnecessary = 0;
+	rxs->alloc_fail = 0;
+	txs->bytes = 0;
+	txs->packets = 0;
+	txs->csum_none = 0;
+	txs->needs_csum = 0;
+	txs->hw_gso_packets = 0;
+	txs->stop = 0;
+	txs->wake = 0;
+}
+
 static const struct net_device_ops enic_netdev_dynamic_ops = {
 	.ndo_open		= enic_open,
 	.ndo_stop		= enic_stop,
@@ -2496,8 +2397,6 @@ static const struct net_device_ops enic_netdev_dynamic_ops = {
 #ifdef CONFIG_RFS_ACCEL
 	.ndo_rx_flow_steer	= enic_rx_flow_steer,
 #endif
-	.ndo_udp_tunnel_add	= enic_udp_tunnel_add,
-	.ndo_udp_tunnel_del	= enic_udp_tunnel_del,
 	.ndo_features_check	= enic_features_check,
 };
 
@@ -2522,38 +2421,98 @@ static const struct net_device_ops enic_netdev_ops = {
 #ifdef CONFIG_RFS_ACCEL
 	.ndo_rx_flow_steer	= enic_rx_flow_steer,
 #endif
-	.ndo_udp_tunnel_add	= enic_udp_tunnel_add,
-	.ndo_udp_tunnel_del	= enic_udp_tunnel_del,
 	.ndo_features_check	= enic_features_check,
 };
 
+static const struct netdev_stat_ops enic_netdev_stat_ops = {
+	.get_queue_stats_rx	= enic_get_queue_stats_rx,
+	.get_queue_stats_tx	= enic_get_queue_stats_tx,
+	.get_base_stats		= enic_get_base_stats,
+};
+
+static void enic_free_enic_resources(struct enic *enic)
+{
+	kfree(enic->wq);
+	enic->wq = NULL;
+
+	kfree(enic->rq);
+	enic->rq = NULL;
+
+	kfree(enic->cq);
+	enic->cq = NULL;
+
+	kfree(enic->napi);
+	enic->napi = NULL;
+
+	kfree(enic->msix_entry);
+	enic->msix_entry = NULL;
+
+	kfree(enic->msix);
+	enic->msix = NULL;
+
+	kfree(enic->intr);
+	enic->intr = NULL;
+}
+
+static int enic_alloc_enic_resources(struct enic *enic)
+{
+	enic->wq = kcalloc(enic->wq_avail, sizeof(struct enic_wq), GFP_KERNEL);
+	if (!enic->wq)
+		goto free_queues;
+
+	enic->rq = kcalloc(enic->rq_avail, sizeof(struct enic_rq), GFP_KERNEL);
+	if (!enic->rq)
+		goto free_queues;
+
+	enic->cq = kcalloc(enic->cq_avail, sizeof(struct vnic_cq), GFP_KERNEL);
+	if (!enic->cq)
+		goto free_queues;
+
+	enic->napi = kcalloc(enic->wq_avail + enic->rq_avail,
+			     sizeof(struct napi_struct), GFP_KERNEL);
+	if (!enic->napi)
+		goto free_queues;
+
+	enic->msix_entry = kcalloc(enic->intr_avail, sizeof(struct msix_entry),
+				   GFP_KERNEL);
+	if (!enic->msix_entry)
+		goto free_queues;
+
+	enic->msix = kcalloc(enic->intr_avail, sizeof(struct enic_msix_entry),
+			     GFP_KERNEL);
+	if (!enic->msix)
+		goto free_queues;
+
+	enic->intr = kcalloc(enic->intr_avail, sizeof(struct vnic_intr),
+			     GFP_KERNEL);
+	if (!enic->intr)
+		goto free_queues;
+
+	return 0;
+
+free_queues:
+	enic_free_enic_resources(enic);
+	return -ENOMEM;
+}
+
 static void enic_dev_deinit(struct enic *enic)
 {
 	unsigned int i;
 
-	for (i = 0; i < enic->rq_count; i++) {
-		napi_hash_del(&enic->napi[i]);
-		netif_napi_del(&enic->napi[i]);
-	}
+	for (i = 0; i < enic->rq_count; i++)
+		__netif_napi_del(&enic->napi[i]);
+
 	if (vnic_dev_get_intr_mode(enic->vdev) == VNIC_DEV_INTR_MODE_MSIX)
 		for (i = 0; i < enic->wq_count; i++)
-			netif_napi_del(&enic->napi[enic_cq_wq(enic, i)]);
+			__netif_napi_del(&enic->napi[enic_cq_wq(enic, i)]);
+
+	/* observe RCU grace period after __netif_napi_del() calls */
+	synchronize_net();
 
 	enic_free_vnic_resources(enic);
 	enic_clear_intr_mode(enic);
 	enic_free_affinity_hint(enic);
-}
-
-static void enic_kdump_kernel_config(struct enic *enic)
-{
-	if (is_kdump_kernel()) {
-		dev_info(enic_get_dev(enic), "Running from within kdump kernel. Using minimal resources\n");
-		enic->rq_count = 1;
-		enic->wq_count = 1;
-		enic->config.rq_desc_count = ENIC_MIN_RQ_DESCS;
-		enic->config.wq_desc_count = ENIC_MIN_WQ_DESCS;
-		enic->config.mtu = min_t(u16, 1500, enic->config.mtu);
-	}
+	enic_free_enic_resources(enic);
 }
 
 static int enic_dev_init(struct enic *enic)
@@ -2585,19 +2544,28 @@ static int enic_dev_init(struct enic *enic)
 
 	enic_get_res_counts(enic);
 
-	/* modify resource count if we are in kdump_kernel
-	 */
-	enic_kdump_kernel_config(enic);
+	enic_ext_cq(enic);
 
-	/* Set interrupt mode based on resource counts and system
-	 * capabilities
-	 */
+	err = enic_alloc_enic_resources(enic);
+	if (err) {
+		dev_err(dev, "Failed to allocate enic resources\n");
+		return err;
+	}
+
+	/* Set interrupt mode based on system capabilities */
 
 	err = enic_set_intr_mode(enic);
 	if (err) {
 		dev_err(dev, "Failed to set intr mode based on resource "
 			"counts and system capabilities, aborting\n");
-		return err;
+		goto err_out_free_vnic_resources;
+	}
+
+	/* Adjust resource counts based on most constrained resources */
+	err = enic_adjust_resources(enic);
+	if (err) {
+		dev_err(dev, "Failed to adjust resources\n");
+		goto err_out_free_vnic_resources;
 	}
 
 	/* Allocate and configure vNIC resources
@@ -2619,16 +2587,17 @@ static int enic_dev_init(struct enic *enic)
 
 	switch (vnic_dev_get_intr_mode(enic->vdev)) {
 	default:
-		netif_napi_add(netdev, &enic->napi[0], enic_poll, 64);
+		netif_napi_add(netdev, &enic->napi[0], enic_poll);
 		break;
 	case VNIC_DEV_INTR_MODE_MSIX:
 		for (i = 0; i < enic->rq_count; i++) {
 			netif_napi_add(netdev, &enic->napi[i],
-				enic_poll_msix_rq, NAPI_POLL_WEIGHT);
+				       enic_poll_msix_rq);
 		}
 		for (i = 0; i < enic->wq_count; i++)
-			netif_napi_add(netdev, &enic->napi[enic_cq_wq(enic, i)],
-				       enic_poll_msix_wq, NAPI_POLL_WEIGHT);
+			netif_napi_add(netdev,
+				       &enic->napi[enic_cq_wq(enic, i)],
+				       enic_poll_msix_wq);
 		break;
 	}
 
@@ -2638,6 +2607,7 @@ err_out_free_vnic_resources:
 	enic_free_affinity_hint(enic);
 	enic_clear_intr_mode(enic);
 	enic_free_vnic_resources(enic);
+	enic_free_enic_resources(enic);
 
 	return err;
 }
@@ -2699,30 +2669,18 @@ static int enic_probe(struct pci_dev *pdev, const struct pci_device_id *ent)
 	pci_set_master(pdev);
 
 	/* Query PCI controller on system for DMA addressing
-	 * limitation for the device.  Try 64-bit first, and
+	 * limitation for the device.  Try 47-bit first, and
 	 * fail to 32-bit.
 	 */
 
-	err = pci_set_dma_mask(pdev, DMA_BIT_MASK(64));
+	err = dma_set_mask_and_coherent(&pdev->dev, DMA_BIT_MASK(47));
 	if (err) {
-		err = pci_set_dma_mask(pdev, DMA_BIT_MASK(32));
+		err = dma_set_mask_and_coherent(&pdev->dev, DMA_BIT_MASK(32));
 		if (err) {
 			dev_err(dev, "No usable DMA configuration, aborting\n");
 			goto err_out_release_regions;
 		}
-		err = pci_set_consistent_dma_mask(pdev, DMA_BIT_MASK(32));
-		if (err) {
-			dev_err(dev, "Unable to obtain %u-bit DMA "
-				"for consistent allocations, aborting\n", 32);
-			goto err_out_release_regions;
-		}
 	} else {
-		err = pci_set_consistent_dma_mask(pdev, DMA_BIT_MASK(64));
-		if (err) {
-			dev_err(dev, "Unable to obtain %u-bit DMA "
-				"for consistent allocations, aborting\n", 64);
-			goto err_out_release_regions;
-		}
 		using_dac = 1;
 	}
 
@@ -2846,23 +2804,20 @@ static int enic_probe(struct pci_dev *pdev, const struct pci_device_id *ent)
 	/* Setup notification timer, HW reset task, and wq locks
 	 */
 
-	init_timer(&enic->notify_timer);
-	enic->notify_timer.function = enic_notify_timer;
-	enic->notify_timer.data = (unsigned long)enic;
+	timer_setup(&enic->notify_timer, enic_notify_timer, 0);
 
-	enic_set_rx_coal_setting(enic);
+	enic_rfs_flw_tbl_init(enic);
 	INIT_WORK(&enic->reset, enic_reset);
 	INIT_WORK(&enic->tx_hang_reset, enic_tx_hang_reset);
 	INIT_WORK(&enic->change_mtu_work, enic_change_mtu_work);
 
 	for (i = 0; i < enic->wq_count; i++)
-		spin_lock_init(&enic->wq_lock[i]);
+		spin_lock_init(&enic->wq[i].lock);
 
 	/* Register net device
 	 */
 
 	enic->port_mtu = enic->config.mtu;
-	(void)enic_change_mtu(netdev, enic->port_mtu);
 
 	err = enic_set_mac_addr(netdev, enic->mac_addr);
 	if (err) {
@@ -2880,6 +2835,7 @@ static int enic_probe(struct pci_dev *pdev, const struct pci_device_id *ent)
 		netdev->netdev_ops = &enic_netdev_dynamic_ops;
 	else
 		netdev->netdev_ops = &enic_netdev_ops;
+	netdev->stat_ops = &enic_netdev_stat_ops;
 
 	netdev->watchdog_timeo = 2 * HZ;
 	enic_set_ethtool_ops(netdev);
@@ -2902,9 +2858,11 @@ static int enic_probe(struct pci_dev *pdev, const struct pci_device_id *ent)
 		netdev->hw_features |= NETIF_F_RXCSUM;
 	if (ENIC_SETTING(enic, VXLAN)) {
 		u64 patch_level;
+		u64 a1 = 0;
 
 		netdev->hw_enc_features |= NETIF_F_RXCSUM		|
 					   NETIF_F_TSO			|
+					   NETIF_F_TSO6			|
 					   NETIF_F_TSO_ECN		|
 					   NETIF_F_GSO_UDP_TUNNEL	|
 					   NETIF_F_HW_CSUM		|
@@ -2923,15 +2881,23 @@ static int enic_probe(struct pci_dev *pdev, const struct pci_device_id *ent)
 		 */
 		err = vnic_dev_get_supported_feature_ver(enic->vdev,
 							 VIC_FEATURE_VXLAN,
-							 &patch_level);
+							 &patch_level, &a1);
 		if (err)
 			patch_level = 0;
+		enic->vxlan.flags = (u8)a1;
 		/* mask bits that are supported by driver
 		 */
 		patch_level &= BIT_ULL(0) | BIT_ULL(2);
 		patch_level = fls(patch_level);
 		patch_level = patch_level ? patch_level - 1 : 0;
 		enic->vxlan.patch_level = patch_level;
+
+		if (vnic_dev_get_res_count(enic->vdev, RES_TYPE_WQ) == 1 ||
+		    enic->vxlan.flags & ENIC_VXLAN_MULTI_WQ) {
+			netdev->udp_tunnel_nic_info = &enic_udp_tunnels_v4;
+			if (enic->vxlan.flags & ENIC_VXLAN_OUTER_IPV6)
+				netdev->udp_tunnel_nic_info = &enic_udp_tunnels;
+		}
 	}
 
 	netdev->features |= netdev->hw_features;
@@ -2949,13 +2915,13 @@ static int enic_probe(struct pci_dev *pdev, const struct pci_device_id *ent)
 	/* MTU range: 68 - 9000 */
 	netdev->min_mtu = ENIC_MIN_MTU;
 	netdev->max_mtu = ENIC_MAX_MTU;
+	netdev->mtu	= enic->port_mtu;
 
 	err = register_netdev(netdev);
 	if (err) {
 		dev_err(dev, "Cannot register net device, aborting\n");
 		goto err_out_dev_deinit;
 	}
-	enic->rx_copybreak = RX_COPYBREAK_DEFAULT;
 
 	return 0;
 
@@ -3020,17 +2986,4 @@ static struct pci_driver enic_driver = {
 	.remove = enic_remove,
 };
 
-static int __init enic_init_module(void)
-{
-	pr_info("%s, ver %s\n", DRV_DESCRIPTION, DRV_VERSION);
-
-	return pci_register_driver(&enic_driver);
-}
-
-static void __exit enic_cleanup_module(void)
-{
-	pci_unregister_driver(&enic_driver);
-}
-
-module_init(enic_init_module);
-module_exit(enic_cleanup_module);
+module_pci_driver(enic_driver);
diff --git a/drivers/net/ethernet/cisco/enic/enic_pp.c b/drivers/net/ethernet/cisco/enic/enic_pp.c
index e6a83198c3dd..4720a952725d 100644
--- a/drivers/net/ethernet/cisco/enic/enic_pp.c
+++ b/drivers/net/ethernet/cisco/enic/enic_pp.c
@@ -1,20 +1,5 @@
-/*
- * Copyright 2011 Cisco Systems, Inc.  All rights reserved.
- *
- * This program is free software; you may redistribute it and/or modify
- * it under the terms of the GNU General Public License as published by
- * the Free Software Foundation; version 2 of the License.
- *
- * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND,
- * EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF
- * MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND
- * NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS
- * BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN
- * ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN
- * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
- * SOFTWARE.
- *
- */
+// SPDX-License-Identifier: GPL-2.0-only
+// Copyright 2011 Cisco Systems, Inc.  All rights reserved.
 
 #include <linux/kernel.h>
 #include <linux/string.h>
@@ -73,9 +58,9 @@ static int enic_set_port_profile(struct enic *enic, int vf)
 	struct vic_provinfo *vp;
 	const u8 oui[3] = VIC_PROVINFO_CISCO_OUI;
 	const __be16 os_type = htons(VIC_GENERIC_PROV_OS_TYPE_LINUX);
+	const u8 *client_mac;
 	char uuid_str[38];
 	char client_mac_str[18];
-	u8 *client_mac;
 	int err;
 
 	ENIC_PP_BY_INDEX(enic, vf, pp, &err);
diff --git a/drivers/net/ethernet/cisco/enic/enic_pp.h b/drivers/net/ethernet/cisco/enic/enic_pp.h
index a09ff392c1c6..20a2687713ef 100644
--- a/drivers/net/ethernet/cisco/enic/enic_pp.h
+++ b/drivers/net/ethernet/cisco/enic/enic_pp.h
@@ -1,20 +1,5 @@
-/*
- * Copyright 2011 Cisco Systems, Inc.  All rights reserved.
- *
- * This program is free software; you may redistribute it and/or modify
- * it under the terms of the GNU General Public License as published by
- * the Free Software Foundation; version 2 of the License.
- *
- * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND,
- * EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF
- * MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND
- * NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS
- * BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN
- * ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN
- * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
- * SOFTWARE.
- *
- */
+/* SPDX-License-Identifier: GPL-2.0-only */
+/* Copyright 2011 Cisco Systems, Inc.  All rights reserved. */
 
 #ifndef _ENIC_PP_H_
 #define _ENIC_PP_H_
diff --git a/drivers/net/ethernet/cisco/enic/enic_res.c b/drivers/net/ethernet/cisco/enic/enic_res.c
index 9c96911fb2c8..bbd3143ed73e 100644
--- a/drivers/net/ethernet/cisco/enic/enic_res.c
+++ b/drivers/net/ethernet/cisco/enic/enic_res.c
@@ -1,20 +1,7 @@
+// SPDX-License-Identifier: GPL-2.0-only
 /*
  * Copyright 2008-2010 Cisco Systems, Inc.  All rights reserved.
  * Copyright 2007 Nuova Systems, Inc.  All rights reserved.
- *
- * This program is free software; you may redistribute it and/or modify
- * it under the terms of the GNU General Public License as published by
- * the Free Software Foundation; version 2 of the License.
- *
- * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND,
- * EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF
- * MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND
- * NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS
- * BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN
- * ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN
- * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
- * SOFTWARE.
- *
  */
 
 #include <linux/kernel.h>
@@ -72,31 +59,38 @@ int enic_get_vnic_config(struct enic *enic)
 	GET_CONFIG(intr_timer_usec);
 	GET_CONFIG(loop_tag);
 	GET_CONFIG(num_arfs);
+	GET_CONFIG(max_rq_ring);
+	GET_CONFIG(max_wq_ring);
+	GET_CONFIG(max_cq_ring);
+
+	if (!c->max_wq_ring)
+		c->max_wq_ring = ENIC_MAX_WQ_DESCS_DEFAULT;
+	if (!c->max_rq_ring)
+		c->max_rq_ring = ENIC_MAX_RQ_DESCS_DEFAULT;
+	if (!c->max_cq_ring)
+		c->max_cq_ring = ENIC_MAX_CQ_DESCS_DEFAULT;
 
 	c->wq_desc_count =
-		min_t(u32, ENIC_MAX_WQ_DESCS,
-		max_t(u32, ENIC_MIN_WQ_DESCS,
-		c->wq_desc_count));
+		min_t(u32, c->max_wq_ring,
+		      max_t(u32, ENIC_MIN_WQ_DESCS, c->wq_desc_count));
 	c->wq_desc_count &= 0xffffffe0; /* must be aligned to groups of 32 */
 
 	c->rq_desc_count =
-		min_t(u32, ENIC_MAX_RQ_DESCS,
-		max_t(u32, ENIC_MIN_RQ_DESCS,
-		c->rq_desc_count));
+		min_t(u32, c->max_rq_ring,
+		      max_t(u32, ENIC_MIN_RQ_DESCS, c->rq_desc_count));
 	c->rq_desc_count &= 0xffffffe0; /* must be aligned to groups of 32 */
 
 	if (c->mtu == 0)
 		c->mtu = 1500;
-	c->mtu = min_t(u16, ENIC_MAX_MTU,
-		max_t(u16, ENIC_MIN_MTU,
-		c->mtu));
+	c->mtu = min_t(u16, ENIC_MAX_MTU, max_t(u16, ENIC_MIN_MTU, c->mtu));
 
 	c->intr_timer_usec = min_t(u32, c->intr_timer_usec,
 		vnic_dev_get_intr_coal_timer_max(enic->vdev));
 
 	dev_info(enic_get_dev(enic),
-		"vNIC MAC addr %pM wq/rq %d/%d mtu %d\n",
-		enic->mac_addr, c->wq_desc_count, c->rq_desc_count, c->mtu);
+		 "vNIC MAC addr %pM wq/rq %d/%d max wq/rq/cq %d/%d/%d mtu %d\n",
+		 enic->mac_addr, c->wq_desc_count, c->rq_desc_count,
+		 c->max_wq_ring, c->max_rq_ring, c->max_cq_ring, c->mtu);
 
 	dev_info(enic_get_dev(enic), "vNIC csum tx/rx %s/%s "
 		"tso/lro %s/%s rss %s intr mode %s type %s timer %d usec "
@@ -149,6 +143,7 @@ int enic_set_nic_cfg(struct enic *enic, u8 rss_default_cpu, u8 rss_hash_type,
 	u8 rss_hash_bits, u8 rss_base_cpu, u8 rss_enable, u8 tso_ipid_split_en,
 	u8 ig_vlan_strip_en)
 {
+	enum vnic_devcmd_cmd cmd = CMD_NIC_CFG;
 	u64 a0, a1;
 	u32 nic_cfg;
 	int wait = 1000;
@@ -160,7 +155,11 @@ int enic_set_nic_cfg(struct enic *enic, u8 rss_default_cpu, u8 rss_hash_type,
 	a0 = nic_cfg;
 	a1 = 0;
 
-	return vnic_dev_cmd(enic->vdev, CMD_NIC_CFG, &a0, &a1, wait);
+	if (rss_hash_type & (NIC_CFG_RSS_HASH_TYPE_UDP_IPV4 |
+			     NIC_CFG_RSS_HASH_TYPE_UDP_IPV6))
+		cmd = CMD_NIC_CFG_CHK;
+
+	return vnic_dev_cmd(enic->vdev, cmd, &a0, &a1, wait);
 }
 
 int enic_set_rss_key(struct enic *enic, dma_addr_t key_pa, u64 len)
@@ -184,9 +183,9 @@ void enic_free_vnic_resources(struct enic *enic)
 	unsigned int i;
 
 	for (i = 0; i < enic->wq_count; i++)
-		vnic_wq_free(&enic->wq[i]);
+		vnic_wq_free(&enic->wq[i].vwq);
 	for (i = 0; i < enic->rq_count; i++)
-		vnic_rq_free(&enic->rq[i]);
+		vnic_rq_free(&enic->rq[i].vrq);
 	for (i = 0; i < enic->cq_count; i++)
 		vnic_cq_free(&enic->cq[i]);
 	for (i = 0; i < enic->intr_count; i++)
@@ -195,16 +194,21 @@ void enic_free_vnic_resources(struct enic *enic)
 
 void enic_get_res_counts(struct enic *enic)
 {
-	enic->wq_count = vnic_dev_get_res_count(enic->vdev, RES_TYPE_WQ);
-	enic->rq_count = vnic_dev_get_res_count(enic->vdev, RES_TYPE_RQ);
-	enic->cq_count = vnic_dev_get_res_count(enic->vdev, RES_TYPE_CQ);
-	enic->intr_count = vnic_dev_get_res_count(enic->vdev,
-		RES_TYPE_INTR_CTRL);
+	enic->wq_avail = vnic_dev_get_res_count(enic->vdev, RES_TYPE_WQ);
+	enic->rq_avail = vnic_dev_get_res_count(enic->vdev, RES_TYPE_RQ);
+	enic->cq_avail = vnic_dev_get_res_count(enic->vdev, RES_TYPE_CQ);
+	enic->intr_avail = vnic_dev_get_res_count(enic->vdev,
+						  RES_TYPE_INTR_CTRL);
+
+	enic->wq_count = enic->wq_avail;
+	enic->rq_count = enic->rq_avail;
+	enic->cq_count = enic->cq_avail;
+	enic->intr_count = enic->intr_avail;
 
 	dev_info(enic_get_dev(enic),
 		"vNIC resources avail: wq %d rq %d cq %d intr %d\n",
-		enic->wq_count, enic->rq_count,
-		enic->cq_count, enic->intr_count);
+		enic->wq_avail, enic->rq_avail,
+		enic->cq_avail, enic->intr_avail);
 }
 
 void enic_init_vnic_resources(struct enic *enic)
@@ -229,9 +233,12 @@ void enic_init_vnic_resources(struct enic *enic)
 
 	switch (intr_mode) {
 	case VNIC_DEV_INTR_MODE_INTX:
+		error_interrupt_enable = 1;
+		error_interrupt_offset = ENIC_LEGACY_ERR_INTR;
+		break;
 	case VNIC_DEV_INTR_MODE_MSIX:
 		error_interrupt_enable = 1;
-		error_interrupt_offset = enic->intr_count - 2;
+		error_interrupt_offset = enic_msix_err_intr(enic);
 		break;
 	default:
 		error_interrupt_enable = 0;
@@ -241,7 +248,7 @@ void enic_init_vnic_resources(struct enic *enic)
 
 	for (i = 0; i < enic->rq_count; i++) {
 		cq_index = i;
-		vnic_rq_init(&enic->rq[i],
+		vnic_rq_init(&enic->rq[i].vrq,
 			cq_index,
 			error_interrupt_enable,
 			error_interrupt_offset);
@@ -249,7 +256,7 @@ void enic_init_vnic_resources(struct enic *enic)
 
 	for (i = 0; i < enic->wq_count; i++) {
 		cq_index = enic->rq_count + i;
-		vnic_wq_init(&enic->wq[i],
+		vnic_wq_init(&enic->wq[i].vwq,
 			cq_index,
 			error_interrupt_enable,
 			error_interrupt_offset);
@@ -257,15 +264,15 @@ void enic_init_vnic_resources(struct enic *enic)
 
 	/* Init CQ resources
 	 *
-	 * CQ[0 - n+m-1] point to INTR[0] for INTx, MSI
-	 * CQ[0 - n+m-1] point to INTR[0 - n+m-1] for MSI-X
+	 * All CQs point to INTR[0] for INTx, MSI
+	 * CQ[i] point to INTR[ENIC_MSIX_IO_INTR_BASE + i] for MSI-X
 	 */
 
 	for (i = 0; i < enic->cq_count; i++) {
 
 		switch (intr_mode) {
 		case VNIC_DEV_INTR_MODE_MSIX:
-			interrupt_offset = i;
+			interrupt_offset = ENIC_MSIX_IO_INTR_BASE + i;
 			break;
 		default:
 			interrupt_offset = 0;
@@ -312,6 +319,7 @@ void enic_init_vnic_resources(struct enic *enic)
 int enic_alloc_vnic_resources(struct enic *enic)
 {
 	enum vnic_dev_intr_mode intr_mode;
+	int rq_cq_desc_size;
 	unsigned int i;
 	int err;
 
@@ -326,11 +334,29 @@ int enic_alloc_vnic_resources(struct enic *enic)
 		intr_mode == VNIC_DEV_INTR_MODE_MSIX ? "MSI-X" :
 		"unknown");
 
+	switch (enic->ext_cq) {
+	case ENIC_RQ_CQ_ENTRY_SIZE_16:
+		rq_cq_desc_size = 16;
+		break;
+	case ENIC_RQ_CQ_ENTRY_SIZE_32:
+		rq_cq_desc_size = 32;
+		break;
+	case ENIC_RQ_CQ_ENTRY_SIZE_64:
+		rq_cq_desc_size = 64;
+		break;
+	default:
+		dev_err(enic_get_dev(enic),
+			"Unable to determine rq cq desc size: %d",
+			enic->ext_cq);
+		err = -ENODEV;
+		goto err_out;
+	}
+
 	/* Allocate queue resources
 	 */
 
 	for (i = 0; i < enic->wq_count; i++) {
-		err = vnic_wq_alloc(enic->vdev, &enic->wq[i], i,
+		err = vnic_wq_alloc(enic->vdev, &enic->wq[i].vwq, i,
 			enic->config.wq_desc_count,
 			sizeof(struct wq_enet_desc));
 		if (err)
@@ -338,7 +364,7 @@ int enic_alloc_vnic_resources(struct enic *enic)
 	}
 
 	for (i = 0; i < enic->rq_count; i++) {
-		err = vnic_rq_alloc(enic->vdev, &enic->rq[i], i,
+		err = vnic_rq_alloc(enic->vdev, &enic->rq[i].vrq, i,
 			enic->config.rq_desc_count,
 			sizeof(struct rq_enet_desc));
 		if (err)
@@ -348,8 +374,8 @@ int enic_alloc_vnic_resources(struct enic *enic)
 	for (i = 0; i < enic->cq_count; i++) {
 		if (i < enic->rq_count)
 			err = vnic_cq_alloc(enic->vdev, &enic->cq[i], i,
-				enic->config.rq_desc_count,
-				sizeof(struct cq_enet_rq_desc));
+					enic->config.rq_desc_count,
+					rq_cq_desc_size);
 		else
 			err = vnic_cq_alloc(enic->vdev, &enic->cq[i], i,
 				enic->config.wq_desc_count,
@@ -380,6 +406,39 @@ int enic_alloc_vnic_resources(struct enic *enic)
 
 err_out_cleanup:
 	enic_free_vnic_resources(enic);
-
+err_out:
 	return err;
 }
+
+/*
+ * CMD_CQ_ENTRY_SIZE_SET can fail on older hw generations that don't support
+ * that command
+ */
+void enic_ext_cq(struct enic *enic)
+{
+	u64 a0 = CMD_CQ_ENTRY_SIZE_SET, a1 = 0;
+	int wait = 1000;
+	int ret;
+
+	spin_lock_bh(&enic->devcmd_lock);
+	ret = vnic_dev_cmd(enic->vdev, CMD_CAPABILITY, &a0, &a1, wait);
+	if (ret || a0) {
+		dev_info(&enic->pdev->dev,
+			 "CMD_CQ_ENTRY_SIZE_SET not supported.");
+		enic->ext_cq = ENIC_RQ_CQ_ENTRY_SIZE_16;
+		goto out;
+	}
+	a1 &= VNIC_RQ_CQ_ENTRY_SIZE_ALL_BIT;
+	enic->ext_cq = fls(a1) - 1;
+	a0 = VNIC_RQ_ALL;
+	a1 = enic->ext_cq;
+	ret = vnic_dev_cmd(enic->vdev, CMD_CQ_ENTRY_SIZE_SET, &a0, &a1, wait);
+	if (ret) {
+		dev_info(&enic->pdev->dev, "CMD_CQ_ENTRY_SIZE_SET failed.");
+		enic->ext_cq = ENIC_RQ_CQ_ENTRY_SIZE_16;
+	}
+out:
+	spin_unlock_bh(&enic->devcmd_lock);
+	dev_info(&enic->pdev->dev, "CQ entry size set to %d bytes",
+		 16 << enic->ext_cq);
+}
diff --git a/drivers/net/ethernet/cisco/enic/enic_res.h b/drivers/net/ethernet/cisco/enic/enic_res.h
index 81f98a8b60e9..02dca1ae4a22 100644
--- a/drivers/net/ethernet/cisco/enic/enic_res.h
+++ b/drivers/net/ethernet/cisco/enic/enic_res.h
@@ -1,20 +1,7 @@
+/* SPDX-License-Identifier: GPL-2.0-only */
 /*
  * Copyright 2008-2010 Cisco Systems, Inc.  All rights reserved.
  * Copyright 2007 Nuova Systems, Inc.  All rights reserved.
- *
- * This program is free software; you may redistribute it and/or modify
- * it under the terms of the GNU General Public License as published by
- * the Free Software Foundation; version 2 of the License.
- *
- * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND,
- * EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF
- * MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND
- * NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS
- * BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN
- * ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN
- * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
- * SOFTWARE.
- *
  */
 
 #ifndef _ENIC_RES_H_
@@ -25,10 +12,13 @@
 #include "vnic_wq.h"
 #include "vnic_rq.h"
 
-#define ENIC_MIN_WQ_DESCS		64
-#define ENIC_MAX_WQ_DESCS		4096
-#define ENIC_MIN_RQ_DESCS		64
-#define ENIC_MAX_RQ_DESCS		4096
+#define ENIC_MIN_WQ_DESCS 64
+#define ENIC_MAX_WQ_DESCS_DEFAULT 4096
+#define ENIC_MAX_WQ_DESCS 16384
+#define ENIC_MIN_RQ_DESCS 64
+#define ENIC_MAX_RQ_DESCS 16384
+#define ENIC_MAX_RQ_DESCS_DEFAULT 4096
+#define ENIC_MAX_CQ_DESCS_DEFAULT (64 * 1024)
 
 #define ENIC_MIN_MTU			ETH_MIN_MTU
 #define ENIC_MAX_MTU			9000
diff --git a/drivers/net/ethernet/cisco/enic/enic_rq.c b/drivers/net/ethernet/cisco/enic/enic_rq.c
new file mode 100644
index 000000000000..ccbf5c9a21d0
--- /dev/null
+++ b/drivers/net/ethernet/cisco/enic/enic_rq.c
@@ -0,0 +1,436 @@
+// SPDX-License-Identifier: GPL-2.0-only
+// Copyright 2024 Cisco Systems, Inc.  All rights reserved.
+
+#include <linux/skbuff.h>
+#include <linux/if_vlan.h>
+#include <net/busy_poll.h>
+#include "enic.h"
+#include "enic_res.h"
+#include "enic_rq.h"
+#include "vnic_rq.h"
+#include "cq_enet_desc.h"
+
+#define ENIC_LARGE_PKT_THRESHOLD                1000
+
+static void enic_intr_update_pkt_size(struct vnic_rx_bytes_counter *pkt_size,
+				      u32 pkt_len)
+{
+	if (pkt_len > ENIC_LARGE_PKT_THRESHOLD)
+		pkt_size->large_pkt_bytes_cnt += pkt_len;
+	else
+		pkt_size->small_pkt_bytes_cnt += pkt_len;
+}
+
+static void enic_rq_cq_desc_dec(void *cq_desc, u8 cq_desc_size, u8 *type,
+				u8 *color, u16 *q_number, u16 *completed_index)
+{
+	/* type_color is the last field for all cq structs */
+	u8 type_color;
+
+	switch (cq_desc_size) {
+	case VNIC_RQ_CQ_ENTRY_SIZE_16: {
+		struct cq_enet_rq_desc *desc =
+			(struct cq_enet_rq_desc *)cq_desc;
+		type_color = desc->type_color;
+
+		/* Make sure color bit is read from desc *before* other fields
+		 * are read from desc.  Hardware guarantees color bit is last
+		 * bit (byte) written.  Adding the rmb() prevents the compiler
+		 * and/or CPU from reordering the reads which would potentially
+		 * result in reading stale values.
+		 */
+		rmb();
+
+		*q_number = le16_to_cpu(desc->q_number_rss_type_flags) &
+			    CQ_DESC_Q_NUM_MASK;
+		*completed_index = le16_to_cpu(desc->completed_index_flags) &
+				   CQ_DESC_COMP_NDX_MASK;
+		break;
+	}
+	case VNIC_RQ_CQ_ENTRY_SIZE_32: {
+		struct cq_enet_rq_desc_32 *desc =
+			(struct cq_enet_rq_desc_32 *)cq_desc;
+		type_color = desc->type_color;
+
+		/* Make sure color bit is read from desc *before* other fields
+		 * are read from desc.  Hardware guarantees color bit is last
+		 * bit (byte) written.  Adding the rmb() prevents the compiler
+		 * and/or CPU from reordering the reads which would potentially
+		 * result in reading stale values.
+		 */
+		rmb();
+
+		*q_number = le16_to_cpu(desc->q_number_rss_type_flags) &
+			    CQ_DESC_Q_NUM_MASK;
+		*completed_index = le16_to_cpu(desc->completed_index_flags) &
+				   CQ_DESC_COMP_NDX_MASK;
+		*completed_index |= (desc->fetch_index_flags & CQ_DESC_32_FI_MASK) <<
+				CQ_DESC_COMP_NDX_BITS;
+		break;
+	}
+	case VNIC_RQ_CQ_ENTRY_SIZE_64: {
+		struct cq_enet_rq_desc_64 *desc =
+			(struct cq_enet_rq_desc_64 *)cq_desc;
+		type_color = desc->type_color;
+
+		/* Make sure color bit is read from desc *before* other fields
+		 * are read from desc.  Hardware guarantees color bit is last
+		 * bit (byte) written.  Adding the rmb() prevents the compiler
+		 * and/or CPU from reordering the reads which would potentially
+		 * result in reading stale values.
+		 */
+		rmb();
+
+		*q_number = le16_to_cpu(desc->q_number_rss_type_flags) &
+			    CQ_DESC_Q_NUM_MASK;
+		*completed_index = le16_to_cpu(desc->completed_index_flags) &
+				   CQ_DESC_COMP_NDX_MASK;
+		*completed_index |= (desc->fetch_index_flags & CQ_DESC_64_FI_MASK) <<
+				CQ_DESC_COMP_NDX_BITS;
+		break;
+	}
+	}
+
+	*color = (type_color >> CQ_DESC_COLOR_SHIFT) & CQ_DESC_COLOR_MASK;
+	*type = type_color & CQ_DESC_TYPE_MASK;
+}
+
+static void enic_rq_set_skb_flags(struct vnic_rq *vrq, u8 type, u32 rss_hash,
+				  u8 rss_type, u8 fcoe, u8 fcoe_fc_crc_ok,
+				  u8 vlan_stripped, u8 csum_not_calc,
+				  u8 tcp_udp_csum_ok, u8 ipv6, u8 ipv4_csum_ok,
+				  u16 vlan_tci, struct sk_buff *skb)
+{
+	struct enic *enic = vnic_dev_priv(vrq->vdev);
+	struct net_device *netdev = enic->netdev;
+	struct enic_rq_stats *rqstats =  &enic->rq[vrq->index].stats;
+	bool outer_csum_ok = true, encap = false;
+
+	if ((netdev->features & NETIF_F_RXHASH) && rss_hash && type == 3) {
+		switch (rss_type) {
+		case CQ_ENET_RQ_DESC_RSS_TYPE_TCP_IPv4:
+		case CQ_ENET_RQ_DESC_RSS_TYPE_TCP_IPv6:
+		case CQ_ENET_RQ_DESC_RSS_TYPE_TCP_IPv6_EX:
+			skb_set_hash(skb, rss_hash, PKT_HASH_TYPE_L4);
+			rqstats->l4_rss_hash++;
+			break;
+		case CQ_ENET_RQ_DESC_RSS_TYPE_IPv4:
+		case CQ_ENET_RQ_DESC_RSS_TYPE_IPv6:
+		case CQ_ENET_RQ_DESC_RSS_TYPE_IPv6_EX:
+			skb_set_hash(skb, rss_hash, PKT_HASH_TYPE_L3);
+			rqstats->l3_rss_hash++;
+			break;
+		}
+	}
+	if (enic->vxlan.vxlan_udp_port_number) {
+		switch (enic->vxlan.patch_level) {
+		case 0:
+			if (fcoe) {
+				encap = true;
+				outer_csum_ok = fcoe_fc_crc_ok;
+			}
+			break;
+		case 2:
+			if (type == 7 && (rss_hash & BIT(0))) {
+				encap = true;
+				outer_csum_ok = (rss_hash & BIT(1)) &&
+						(rss_hash & BIT(2));
+			}
+			break;
+		}
+	}
+
+	/* Hardware does not provide whole packet checksum. It only
+	 * provides pseudo checksum. Since hw validates the packet
+	 * checksum but not provide us the checksum value. use
+	 * CHECSUM_UNNECESSARY.
+	 *
+	 * In case of encap pkt tcp_udp_csum_ok/tcp_udp_csum_ok is
+	 * inner csum_ok. outer_csum_ok is set by hw when outer udp
+	 * csum is correct or is zero.
+	 */
+	if ((netdev->features & NETIF_F_RXCSUM) && !csum_not_calc &&
+	    tcp_udp_csum_ok && outer_csum_ok && (ipv4_csum_ok || ipv6)) {
+		skb->ip_summed = CHECKSUM_UNNECESSARY;
+		skb->csum_level = encap;
+		if (encap)
+			rqstats->csum_unnecessary_encap++;
+		else
+			rqstats->csum_unnecessary++;
+	}
+
+	if (vlan_stripped) {
+		__vlan_hwaccel_put_tag(skb, htons(ETH_P_8021Q), vlan_tci);
+		rqstats->vlan_stripped++;
+	}
+}
+
+/*
+ * cq_enet_rq_desc accesses section uses only the 1st 15 bytes of the cq which
+ * is identical for all type (16,32 and 64 byte) of cqs.
+ */
+static void cq_enet_rq_desc_dec(struct cq_enet_rq_desc *desc, u8 *ingress_port,
+				u8 *fcoe, u8 *eop, u8 *sop, u8 *rss_type,
+				u8 *csum_not_calc, u32 *rss_hash,
+				u16 *bytes_written, u8 *packet_error,
+				u8 *vlan_stripped, u16 *vlan_tci,
+				u16 *checksum, u8 *fcoe_sof,
+				u8 *fcoe_fc_crc_ok, u8 *fcoe_enc_error,
+				u8 *fcoe_eof, u8 *tcp_udp_csum_ok, u8 *udp,
+				u8 *tcp, u8 *ipv4_csum_ok, u8 *ipv6, u8 *ipv4,
+				u8 *ipv4_fragment, u8 *fcs_ok)
+{
+	u16 completed_index_flags;
+	u16 q_number_rss_type_flags;
+	u16 bytes_written_flags;
+
+	completed_index_flags = le16_to_cpu(desc->completed_index_flags);
+	q_number_rss_type_flags =
+		le16_to_cpu(desc->q_number_rss_type_flags);
+	bytes_written_flags = le16_to_cpu(desc->bytes_written_flags);
+
+	*ingress_port = (completed_index_flags &
+		CQ_ENET_RQ_DESC_FLAGS_INGRESS_PORT) ? 1 : 0;
+	*fcoe = (completed_index_flags & CQ_ENET_RQ_DESC_FLAGS_FCOE) ?
+		1 : 0;
+	*eop = (completed_index_flags & CQ_ENET_RQ_DESC_FLAGS_EOP) ?
+		1 : 0;
+	*sop = (completed_index_flags & CQ_ENET_RQ_DESC_FLAGS_SOP) ?
+		1 : 0;
+
+	*rss_type = (u8)((q_number_rss_type_flags >> CQ_DESC_Q_NUM_BITS) &
+		CQ_ENET_RQ_DESC_RSS_TYPE_MASK);
+	*csum_not_calc = (q_number_rss_type_flags &
+		CQ_ENET_RQ_DESC_FLAGS_CSUM_NOT_CALC) ? 1 : 0;
+
+	*rss_hash = le32_to_cpu(desc->rss_hash);
+
+	*bytes_written = bytes_written_flags &
+		CQ_ENET_RQ_DESC_BYTES_WRITTEN_MASK;
+	*packet_error = (bytes_written_flags &
+		CQ_ENET_RQ_DESC_FLAGS_TRUNCATED) ? 1 : 0;
+	*vlan_stripped = (bytes_written_flags &
+		CQ_ENET_RQ_DESC_FLAGS_VLAN_STRIPPED) ? 1 : 0;
+
+	/*
+	 * Tag Control Information(16) = user_priority(3) + cfi(1) + vlan(12)
+	 */
+	*vlan_tci = le16_to_cpu(desc->vlan);
+
+	if (*fcoe) {
+		*fcoe_sof = (u8)(le16_to_cpu(desc->checksum_fcoe) &
+			CQ_ENET_RQ_DESC_FCOE_SOF_MASK);
+		*fcoe_fc_crc_ok = (desc->flags &
+			CQ_ENET_RQ_DESC_FCOE_FC_CRC_OK) ? 1 : 0;
+		*fcoe_enc_error = (desc->flags &
+			CQ_ENET_RQ_DESC_FCOE_ENC_ERROR) ? 1 : 0;
+		*fcoe_eof = (u8)((le16_to_cpu(desc->checksum_fcoe) >>
+			CQ_ENET_RQ_DESC_FCOE_EOF_SHIFT) &
+			CQ_ENET_RQ_DESC_FCOE_EOF_MASK);
+		*checksum = 0;
+	} else {
+		*fcoe_sof = 0;
+		*fcoe_fc_crc_ok = 0;
+		*fcoe_enc_error = 0;
+		*fcoe_eof = 0;
+		*checksum = le16_to_cpu(desc->checksum_fcoe);
+	}
+
+	*tcp_udp_csum_ok =
+		(desc->flags & CQ_ENET_RQ_DESC_FLAGS_TCP_UDP_CSUM_OK) ? 1 : 0;
+	*udp = (desc->flags & CQ_ENET_RQ_DESC_FLAGS_UDP) ? 1 : 0;
+	*tcp = (desc->flags & CQ_ENET_RQ_DESC_FLAGS_TCP) ? 1 : 0;
+	*ipv4_csum_ok =
+		(desc->flags & CQ_ENET_RQ_DESC_FLAGS_IPV4_CSUM_OK) ? 1 : 0;
+	*ipv6 = (desc->flags & CQ_ENET_RQ_DESC_FLAGS_IPV6) ? 1 : 0;
+	*ipv4 = (desc->flags & CQ_ENET_RQ_DESC_FLAGS_IPV4) ? 1 : 0;
+	*ipv4_fragment =
+		(desc->flags & CQ_ENET_RQ_DESC_FLAGS_IPV4_FRAGMENT) ? 1 : 0;
+	*fcs_ok = (desc->flags & CQ_ENET_RQ_DESC_FLAGS_FCS_OK) ? 1 : 0;
+}
+
+static bool enic_rq_pkt_error(struct vnic_rq *vrq, u8 packet_error, u8 fcs_ok,
+			      u16 bytes_written)
+{
+	struct enic *enic = vnic_dev_priv(vrq->vdev);
+	struct enic_rq_stats *rqstats = &enic->rq[vrq->index].stats;
+
+	if (packet_error) {
+		if (!fcs_ok) {
+			if (bytes_written > 0)
+				rqstats->bad_fcs++;
+			else if (bytes_written == 0)
+				rqstats->pkt_truncated++;
+		}
+		return true;
+	}
+	return false;
+}
+
+int enic_rq_alloc_buf(struct vnic_rq *rq)
+{
+	struct enic *enic = vnic_dev_priv(rq->vdev);
+	struct net_device *netdev = enic->netdev;
+	struct enic_rq *erq = &enic->rq[rq->index];
+	struct enic_rq_stats *rqstats = &erq->stats;
+	unsigned int offset = 0;
+	unsigned int len = netdev->mtu + VLAN_ETH_HLEN;
+	unsigned int os_buf_index = 0;
+	dma_addr_t dma_addr;
+	struct vnic_rq_buf *buf = rq->to_use;
+	struct page *page;
+	unsigned int truesize = len;
+
+	if (buf->os_buf) {
+		enic_queue_rq_desc(rq, buf->os_buf, os_buf_index, buf->dma_addr,
+				   buf->len);
+
+		return 0;
+	}
+
+	page = page_pool_dev_alloc(erq->pool, &offset, &truesize);
+	if (unlikely(!page)) {
+		rqstats->pp_alloc_fail++;
+		return -ENOMEM;
+	}
+	buf->offset = offset;
+	buf->truesize = truesize;
+	dma_addr = page_pool_get_dma_addr(page) + offset;
+	enic_queue_rq_desc(rq, (void *)page, os_buf_index, dma_addr, len);
+
+	return 0;
+}
+
+void enic_free_rq_buf(struct vnic_rq *rq, struct vnic_rq_buf *buf)
+{
+	struct enic *enic = vnic_dev_priv(rq->vdev);
+	struct enic_rq *erq = &enic->rq[rq->index];
+
+	if (!buf->os_buf)
+		return;
+
+	page_pool_put_full_page(erq->pool, (struct page *)buf->os_buf, true);
+	buf->os_buf = NULL;
+}
+
+static void enic_rq_indicate_buf(struct enic *enic, struct vnic_rq *rq,
+				 struct vnic_rq_buf *buf, void *cq_desc,
+				 u8 type, u16 q_number, u16 completed_index)
+{
+	struct sk_buff *skb;
+	struct vnic_cq *cq = &enic->cq[enic_cq_rq(enic, rq->index)];
+	struct enic_rq_stats *rqstats = &enic->rq[rq->index].stats;
+	struct napi_struct *napi;
+
+	u8 eop, sop, ingress_port, vlan_stripped;
+	u8 fcoe, fcoe_sof, fcoe_fc_crc_ok, fcoe_enc_error, fcoe_eof;
+	u8 tcp_udp_csum_ok, udp, tcp, ipv4_csum_ok;
+	u8 ipv6, ipv4, ipv4_fragment, fcs_ok, rss_type, csum_not_calc;
+	u8 packet_error;
+	u16 bytes_written, vlan_tci, checksum;
+	u32 rss_hash;
+
+	rqstats->packets++;
+
+	cq_enet_rq_desc_dec((struct cq_enet_rq_desc *)cq_desc, &ingress_port,
+			    &fcoe, &eop, &sop, &rss_type, &csum_not_calc,
+			    &rss_hash, &bytes_written, &packet_error,
+			    &vlan_stripped, &vlan_tci, &checksum, &fcoe_sof,
+			    &fcoe_fc_crc_ok, &fcoe_enc_error, &fcoe_eof,
+			    &tcp_udp_csum_ok, &udp, &tcp, &ipv4_csum_ok, &ipv6,
+			    &ipv4, &ipv4_fragment, &fcs_ok);
+
+	if (enic_rq_pkt_error(rq, packet_error, fcs_ok, bytes_written))
+		return;
+
+	if (eop && bytes_written > 0) {
+		/* Good receive
+		 */
+		rqstats->bytes += bytes_written;
+		napi = &enic->napi[rq->index];
+		skb = napi_get_frags(napi);
+		if (unlikely(!skb)) {
+			net_warn_ratelimited("%s: skb alloc error rq[%d], desc[%d]\n",
+					     enic->netdev->name, rq->index,
+					     completed_index);
+			rqstats->no_skb++;
+			return;
+		}
+
+		prefetch(skb->data - NET_IP_ALIGN);
+
+		dma_sync_single_for_cpu(&enic->pdev->dev, buf->dma_addr,
+					bytes_written, DMA_FROM_DEVICE);
+		skb_add_rx_frag(skb, skb_shinfo(skb)->nr_frags,
+				(struct page *)buf->os_buf, buf->offset,
+				bytes_written, buf->truesize);
+		skb_record_rx_queue(skb, q_number);
+		enic_rq_set_skb_flags(rq, type, rss_hash, rss_type, fcoe,
+				      fcoe_fc_crc_ok, vlan_stripped,
+				      csum_not_calc, tcp_udp_csum_ok, ipv6,
+				      ipv4_csum_ok, vlan_tci, skb);
+		skb_mark_for_recycle(skb);
+		napi_gro_frags(napi);
+		if (enic->rx_coalesce_setting.use_adaptive_rx_coalesce)
+			enic_intr_update_pkt_size(&cq->pkt_size_counter,
+						  bytes_written);
+		buf->os_buf = NULL;
+		buf->dma_addr = 0;
+		buf = buf->next;
+	} else {
+		/* Buffer overflow
+		 */
+		rqstats->pkt_truncated++;
+	}
+}
+
+static void enic_rq_service(struct enic *enic, void *cq_desc, u8 type,
+			    u16 q_number, u16 completed_index)
+{
+	struct enic_rq_stats *rqstats = &enic->rq[q_number].stats;
+	struct vnic_rq *vrq = &enic->rq[q_number].vrq;
+	struct vnic_rq_buf *vrq_buf = vrq->to_clean;
+	int skipped;
+
+	while (1) {
+		skipped = (vrq_buf->index != completed_index);
+		if (!skipped)
+			enic_rq_indicate_buf(enic, vrq, vrq_buf, cq_desc, type,
+					     q_number, completed_index);
+		else
+			rqstats->desc_skip++;
+
+		vrq->ring.desc_avail++;
+		vrq->to_clean = vrq_buf->next;
+		vrq_buf = vrq_buf->next;
+		if (!skipped)
+			break;
+	}
+}
+
+unsigned int enic_rq_cq_service(struct enic *enic, unsigned int cq_index,
+				unsigned int work_to_do)
+{
+	struct vnic_cq *cq = &enic->cq[cq_index];
+	void *cq_desc = vnic_cq_to_clean(cq);
+	u16 q_number, completed_index;
+	unsigned int work_done = 0;
+	u8 type, color;
+
+	enic_rq_cq_desc_dec(cq_desc, enic->ext_cq, &type, &color, &q_number,
+			    &completed_index);
+
+	while (color != cq->last_color) {
+		enic_rq_service(enic, cq_desc, type, q_number, completed_index);
+		vnic_cq_inc_to_clean(cq);
+
+		if (++work_done >= work_to_do)
+			break;
+
+		cq_desc = vnic_cq_to_clean(cq);
+		enic_rq_cq_desc_dec(cq_desc, enic->ext_cq, &type, &color,
+				    &q_number, &completed_index);
+	}
+
+	return work_done;
+}
diff --git a/drivers/net/ethernet/cisco/enic/enic_rq.h b/drivers/net/ethernet/cisco/enic/enic_rq.h
new file mode 100644
index 000000000000..98476a7297af
--- /dev/null
+++ b/drivers/net/ethernet/cisco/enic/enic_rq.h
@@ -0,0 +1,8 @@
+/* SPDX-License-Identifier: GPL-2.0-only
+ * Copyright 2024 Cisco Systems, Inc.  All rights reserved.
+ */
+
+unsigned int enic_rq_cq_service(struct enic *enic, unsigned int cq_index,
+				unsigned int work_to_do);
+int enic_rq_alloc_buf(struct vnic_rq *rq);
+void enic_free_rq_buf(struct vnic_rq *rq, struct vnic_rq_buf *buf);
diff --git a/drivers/net/ethernet/cisco/enic/enic_wq.c b/drivers/net/ethernet/cisco/enic/enic_wq.c
new file mode 100644
index 000000000000..07936f8b4231
--- /dev/null
+++ b/drivers/net/ethernet/cisco/enic/enic_wq.c
@@ -0,0 +1,117 @@
+// SPDX-License-Identifier: GPL-2.0-only
+// Copyright 2025 Cisco Systems, Inc.  All rights reserved.
+
+#include <net/netdev_queues.h>
+#include "enic_res.h"
+#include "enic.h"
+#include "enic_wq.h"
+
+#define ENET_CQ_DESC_COMP_NDX_BITS 14
+#define ENET_CQ_DESC_COMP_NDX_MASK GENMASK(ENET_CQ_DESC_COMP_NDX_BITS - 1, 0)
+
+static void enic_wq_cq_desc_dec(const struct cq_desc *desc_arg, bool ext_wq,
+				u8 *type, u8 *color, u16 *q_number,
+				u16 *completed_index)
+{
+	const struct cq_desc *desc = desc_arg;
+	const u8 type_color = desc->type_color;
+
+	*color = (type_color >> CQ_DESC_COLOR_SHIFT) & CQ_DESC_COLOR_MASK;
+
+	/*
+	 * Make sure color bit is read from desc *before* other fields
+	 * are read from desc.  Hardware guarantees color bit is last
+	 * bit (byte) written.  Adding the rmb() prevents the compiler
+	 * and/or CPU from reordering the reads which would potentially
+	 * result in reading stale values.
+	 */
+	rmb();
+
+	*type = type_color & CQ_DESC_TYPE_MASK;
+	*q_number = le16_to_cpu(desc->q_number) & CQ_DESC_Q_NUM_MASK;
+
+	if (ext_wq)
+		*completed_index = le16_to_cpu(desc->completed_index) &
+			ENET_CQ_DESC_COMP_NDX_MASK;
+	else
+		*completed_index = le16_to_cpu(desc->completed_index) &
+			CQ_DESC_COMP_NDX_MASK;
+}
+
+void enic_free_wq_buf(struct vnic_wq *wq, struct vnic_wq_buf *buf)
+{
+	struct enic *enic = vnic_dev_priv(wq->vdev);
+
+	if (buf->sop)
+		dma_unmap_single(&enic->pdev->dev, buf->dma_addr, buf->len,
+				 DMA_TO_DEVICE);
+	else
+		dma_unmap_page(&enic->pdev->dev, buf->dma_addr, buf->len,
+			       DMA_TO_DEVICE);
+
+	if (buf->os_buf)
+		dev_kfree_skb_any(buf->os_buf);
+}
+
+static void enic_wq_free_buf(struct vnic_wq *wq, struct cq_desc *cq_desc,
+			     struct vnic_wq_buf *buf, void *opaque)
+{
+	struct enic *enic = vnic_dev_priv(wq->vdev);
+
+	enic->wq[wq->index].stats.cq_work++;
+	enic->wq[wq->index].stats.cq_bytes += buf->len;
+	enic_free_wq_buf(wq, buf);
+}
+
+static void enic_wq_service(struct vnic_dev *vdev, struct cq_desc *cq_desc,
+			    u8 type, u16 q_number, u16 completed_index)
+{
+	struct enic *enic = vnic_dev_priv(vdev);
+
+	spin_lock(&enic->wq[q_number].lock);
+
+	vnic_wq_service(&enic->wq[q_number].vwq, cq_desc,
+			completed_index, enic_wq_free_buf, NULL);
+
+	if (netif_tx_queue_stopped(netdev_get_tx_queue(enic->netdev, q_number))
+	    && vnic_wq_desc_avail(&enic->wq[q_number].vwq) >=
+	    (MAX_SKB_FRAGS + ENIC_DESC_MAX_SPLITS)) {
+		netif_wake_subqueue(enic->netdev, q_number);
+		enic->wq[q_number].stats.wake++;
+	}
+
+	spin_unlock(&enic->wq[q_number].lock);
+}
+
+unsigned int enic_wq_cq_service(struct enic *enic, unsigned int cq_index,
+				unsigned int work_to_do)
+{
+	struct vnic_cq *cq = &enic->cq[cq_index];
+	u16 q_number, completed_index;
+	unsigned int work_done = 0;
+	struct cq_desc *cq_desc;
+	u8 type, color;
+	bool ext_wq;
+
+	ext_wq = cq->ring.size > ENIC_MAX_WQ_DESCS_DEFAULT;
+
+	cq_desc = (struct cq_desc *)vnic_cq_to_clean(cq);
+	enic_wq_cq_desc_dec(cq_desc, ext_wq, &type, &color,
+			    &q_number, &completed_index);
+
+	while (color != cq->last_color) {
+		enic_wq_service(cq->vdev, cq_desc, type, q_number,
+				completed_index);
+
+		vnic_cq_inc_to_clean(cq);
+
+		if (++work_done >= work_to_do)
+			break;
+
+		cq_desc = (struct cq_desc *)vnic_cq_to_clean(cq);
+		enic_wq_cq_desc_dec(cq_desc, ext_wq, &type, &color,
+				    &q_number, &completed_index);
+	}
+
+	return work_done;
+}
diff --git a/drivers/net/ethernet/cisco/enic/enic_wq.h b/drivers/net/ethernet/cisco/enic/enic_wq.h
new file mode 100644
index 000000000000..12acb3f2fbc9
--- /dev/null
+++ b/drivers/net/ethernet/cisco/enic/enic_wq.h
@@ -0,0 +1,7 @@
+/* SPDX-License-Identifier: GPL-2.0-only
+ * Copyright 2025 Cisco Systems, Inc.  All rights reserved.
+ */
+
+void enic_free_wq_buf(struct vnic_wq *wq, struct vnic_wq_buf *buf);
+unsigned int enic_wq_cq_service(struct enic *enic, unsigned int cq_index,
+				unsigned int work_to_do);
diff --git a/drivers/net/ethernet/cisco/enic/rq_enet_desc.h b/drivers/net/ethernet/cisco/enic/rq_enet_desc.h
index e6dd30988d6f..0ab5fd6b8d46 100644
--- a/drivers/net/ethernet/cisco/enic/rq_enet_desc.h
+++ b/drivers/net/ethernet/cisco/enic/rq_enet_desc.h
@@ -1,20 +1,7 @@
+/* SPDX-License-Identifier: GPL-2.0-only */
 /*
  * Copyright 2008-2010 Cisco Systems, Inc.  All rights reserved.
  * Copyright 2007 Nuova Systems, Inc.  All rights reserved.
- *
- * This program is free software; you may redistribute it and/or modify
- * it under the terms of the GNU General Public License as published by
- * the Free Software Foundation; version 2 of the License.
- *
- * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND,
- * EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF
- * MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND
- * NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS
- * BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN
- * ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN
- * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
- * SOFTWARE.
- *
  */
 
 #ifndef _RQ_ENET_DESC_H_
diff --git a/drivers/net/ethernet/cisco/enic/vnic_cq.c b/drivers/net/ethernet/cisco/enic/vnic_cq.c
index 9c682aff3834..27c885e91552 100644
--- a/drivers/net/ethernet/cisco/enic/vnic_cq.c
+++ b/drivers/net/ethernet/cisco/enic/vnic_cq.c
@@ -1,20 +1,7 @@
+// SPDX-License-Identifier: GPL-2.0-only
 /*
  * Copyright 2008-2010 Cisco Systems, Inc.  All rights reserved.
  * Copyright 2007 Nuova Systems, Inc.  All rights reserved.
- *
- * This program is free software; you may redistribute it and/or modify
- * it under the terms of the GNU General Public License as published by
- * the Free Software Foundation; version 2 of the License.
- *
- * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND,
- * EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF
- * MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND
- * NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS
- * BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN
- * ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN
- * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
- * SOFTWARE.
- *
  */
 
 #include <linux/kernel.h>
@@ -36,8 +23,6 @@ void vnic_cq_free(struct vnic_cq *cq)
 int vnic_cq_alloc(struct vnic_dev *vdev, struct vnic_cq *cq, unsigned int index,
 	unsigned int desc_count, unsigned int desc_size)
 {
-	int err;
-
 	cq->index = index;
 	cq->vdev = vdev;
 
@@ -47,11 +32,7 @@ int vnic_cq_alloc(struct vnic_dev *vdev, struct vnic_cq *cq, unsigned int index,
 		return -EINVAL;
 	}
 
-	err = vnic_dev_alloc_desc_ring(vdev, &cq->ring, desc_count, desc_size);
-	if (err)
-		return err;
-
-	return 0;
+	return vnic_dev_alloc_desc_ring(vdev, &cq->ring, desc_count, desc_size);
 }
 
 void vnic_cq_init(struct vnic_cq *cq, unsigned int flow_control_enable,
diff --git a/drivers/net/ethernet/cisco/enic/vnic_cq.h b/drivers/net/ethernet/cisco/enic/vnic_cq.h
index 4e6aa65857f7..0e37f5d5e527 100644
--- a/drivers/net/ethernet/cisco/enic/vnic_cq.h
+++ b/drivers/net/ethernet/cisco/enic/vnic_cq.h
@@ -1,20 +1,7 @@
+/* SPDX-License-Identifier: GPL-2.0-only */
 /*
  * Copyright 2008-2010 Cisco Systems, Inc.  All rights reserved.
  * Copyright 2007 Nuova Systems, Inc.  All rights reserved.
- *
- * This program is free software; you may redistribute it and/or modify
- * it under the terms of the GNU General Public License as published by
- * the Free Software Foundation; version 2 of the License.
- *
- * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND,
- * EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF
- * MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND
- * NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS
- * BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN
- * ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN
- * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
- * SOFTWARE.
- *
  */
 
 #ifndef _VNIC_CQ_H_
@@ -69,45 +56,18 @@ struct vnic_cq {
 	ktime_t prev_ts;
 };
 
-static inline unsigned int vnic_cq_service(struct vnic_cq *cq,
-	unsigned int work_to_do,
-	int (*q_service)(struct vnic_dev *vdev, struct cq_desc *cq_desc,
-	u8 type, u16 q_number, u16 completed_index, void *opaque),
-	void *opaque)
+static inline void *vnic_cq_to_clean(struct vnic_cq *cq)
 {
-	struct cq_desc *cq_desc;
-	unsigned int work_done = 0;
-	u16 q_number, completed_index;
-	u8 type, color;
-
-	cq_desc = (struct cq_desc *)((u8 *)cq->ring.descs +
-		cq->ring.desc_size * cq->to_clean);
-	cq_desc_dec(cq_desc, &type, &color,
-		&q_number, &completed_index);
-
-	while (color != cq->last_color) {
-
-		if ((*q_service)(cq->vdev, cq_desc, type,
-			q_number, completed_index, opaque))
-			break;
-
-		cq->to_clean++;
-		if (cq->to_clean == cq->ring.desc_count) {
-			cq->to_clean = 0;
-			cq->last_color = cq->last_color ? 0 : 1;
-		}
-
-		cq_desc = (struct cq_desc *)((u8 *)cq->ring.descs +
-			cq->ring.desc_size * cq->to_clean);
-		cq_desc_dec(cq_desc, &type, &color,
-			&q_number, &completed_index);
+	return ((u8 *)cq->ring.descs + cq->ring.desc_size * cq->to_clean);
+}
 
-		work_done++;
-		if (work_done >= work_to_do)
-			break;
+static inline void vnic_cq_inc_to_clean(struct vnic_cq *cq)
+{
+	cq->to_clean++;
+	if (cq->to_clean == cq->ring.desc_count) {
+		cq->to_clean = 0;
+		cq->last_color = cq->last_color ? 0 : 1;
 	}
-
-	return work_done;
 }
 
 void vnic_cq_free(struct vnic_cq *cq);
diff --git a/drivers/net/ethernet/cisco/enic/vnic_dev.c b/drivers/net/ethernet/cisco/enic/vnic_dev.c
index 1841ad45d215..9f6089e81608 100644
--- a/drivers/net/ethernet/cisco/enic/vnic_dev.c
+++ b/drivers/net/ethernet/cisco/enic/vnic_dev.c
@@ -1,20 +1,7 @@
+// SPDX-License-Identifier: GPL-2.0-only
 /*
  * Copyright 2008-2010 Cisco Systems, Inc.  All rights reserved.
  * Copyright 2007 Nuova Systems, Inc.  All rights reserved.
- *
- * This program is free software; you may redistribute it and/or modify
- * it under the terms of the GNU General Public License as published by
- * the Free Software Foundation; version 2 of the License.
- *
- * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND,
- * EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF
- * MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND
- * NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS
- * BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN
- * ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN
- * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
- * SOFTWARE.
- *
  */
 
 #include <linux/kernel.h>
@@ -159,23 +146,19 @@ EXPORT_SYMBOL(vnic_dev_get_res);
 static unsigned int vnic_dev_desc_ring_size(struct vnic_dev_ring *ring,
 	unsigned int desc_count, unsigned int desc_size)
 {
-	/* The base address of the desc rings must be 512 byte aligned.
-	 * Descriptor count is aligned to groups of 32 descriptors.  A
-	 * count of 0 means the maximum 4096 descriptors.  Descriptor
-	 * size is aligned to 16 bytes.
-	 */
-
-	unsigned int count_align = 32;
-	unsigned int desc_align = 16;
 
-	ring->base_align = 512;
+	/* Descriptor ring base address alignment in bytes*/
+	ring->base_align = VNIC_DESC_BASE_ALIGN;
 
+	/* A count of 0 means the maximum descriptors */
 	if (desc_count == 0)
-		desc_count = 4096;
+		desc_count = VNIC_DESC_MAX_COUNT;
 
-	ring->desc_count = ALIGN(desc_count, count_align);
+	/* Descriptor count aligned in groups of VNIC_DESC_COUNT_ALIGN descriptors */
+	ring->desc_count = ALIGN(desc_count, VNIC_DESC_COUNT_ALIGN);
 
-	ring->desc_size = ALIGN(desc_size, desc_align);
+	/* Descriptor size alignment in bytes */
+	ring->desc_size = ALIGN(desc_size, VNIC_DESC_SIZE_ALIGN);
 
 	ring->size = ring->desc_count * ring->desc_size;
 	ring->size_unaligned = ring->size + ring->base_align;
@@ -193,9 +176,10 @@ int vnic_dev_alloc_desc_ring(struct vnic_dev *vdev, struct vnic_dev_ring *ring,
 {
 	vnic_dev_desc_ring_size(ring, desc_count, desc_size);
 
-	ring->descs_unaligned = pci_alloc_consistent(vdev->pdev,
-		ring->size_unaligned,
-		&ring->base_addr_unaligned);
+	ring->descs_unaligned = dma_alloc_coherent(&vdev->pdev->dev,
+						   ring->size_unaligned,
+						   &ring->base_addr_unaligned,
+						   GFP_KERNEL);
 
 	if (!ring->descs_unaligned) {
 		vdev_err(vdev, "Failed to allocate ring (size=%d), aborting\n",
@@ -218,10 +202,9 @@ int vnic_dev_alloc_desc_ring(struct vnic_dev *vdev, struct vnic_dev_ring *ring,
 void vnic_dev_free_desc_ring(struct vnic_dev *vdev, struct vnic_dev_ring *ring)
 {
 	if (ring->descs) {
-		pci_free_consistent(vdev->pdev,
-			ring->size_unaligned,
-			ring->descs_unaligned,
-			ring->base_addr_unaligned);
+		dma_free_coherent(&vdev->pdev->dev, ring->size_unaligned,
+				  ring->descs_unaligned,
+				  ring->base_addr_unaligned);
 		ring->descs = NULL;
 	}
 }
@@ -402,8 +385,8 @@ static int vnic_dev_init_devcmd2(struct vnic_dev *vdev)
 	fetch_index = ioread32(&vdev->devcmd2->wq.ctrl->fetch_index);
 	if (fetch_index == 0xFFFFFFFF) { /* check for hardware gone  */
 		vdev_err(vdev, "Fatal error in devcmd2 init - hardware surprise removal\n");
-
-		return -ENODEV;
+		err = -ENODEV;
+		goto err_free_wq;
 	}
 
 	enic_wq_init_start(&vdev->devcmd2->wq, 0, fetch_index, fetch_index, 0,
@@ -414,7 +397,7 @@ static int vnic_dev_init_devcmd2(struct vnic_dev *vdev)
 	err = vnic_dev_alloc_desc_ring(vdev, &vdev->devcmd2->results_ring,
 				       DEVCMD2_RING_SIZE, DEVCMD2_DESC_SIZE);
 	if (err)
-		goto err_free_wq;
+		goto err_disable_wq;
 
 	vdev->devcmd2->result = vdev->devcmd2->results_ring.descs;
 	vdev->devcmd2->cmd_ring = vdev->devcmd2->wq.ring.descs;
@@ -433,8 +416,9 @@ static int vnic_dev_init_devcmd2(struct vnic_dev *vdev)
 
 err_free_desc_ring:
 	vnic_dev_free_desc_ring(vdev, &vdev->devcmd2->results_ring);
-err_free_wq:
+err_disable_wq:
 	vnic_wq_disable(&vdev->devcmd2->wq);
+err_free_wq:
 	vnic_wq_free(&vdev->devcmd2->wq);
 err_free_devcmd2:
 	kfree(vdev->devcmd2);
@@ -550,9 +534,9 @@ int vnic_dev_fw_info(struct vnic_dev *vdev,
 	int err = 0;
 
 	if (!vdev->fw_info) {
-		vdev->fw_info = pci_zalloc_consistent(vdev->pdev,
-						      sizeof(struct vnic_devcmd_fw_info),
-						      &vdev->fw_info_pa);
+		vdev->fw_info = dma_alloc_coherent(&vdev->pdev->dev,
+						   sizeof(struct vnic_devcmd_fw_info),
+						   &vdev->fw_info_pa, GFP_ATOMIC);
 		if (!vdev->fw_info)
 			return -ENOMEM;
 
@@ -602,8 +586,9 @@ int vnic_dev_stats_dump(struct vnic_dev *vdev, struct vnic_stats **stats)
 	int wait = 1000;
 
 	if (!vdev->stats) {
-		vdev->stats = pci_alloc_consistent(vdev->pdev,
-			sizeof(struct vnic_stats), &vdev->stats_pa);
+		vdev->stats = dma_alloc_coherent(&vdev->pdev->dev,
+						 sizeof(struct vnic_stats),
+						 &vdev->stats_pa, GFP_ATOMIC);
 		if (!vdev->stats)
 			return -ENOMEM;
 	}
@@ -851,9 +836,9 @@ int vnic_dev_notify_set(struct vnic_dev *vdev, u16 intr)
 		return -EINVAL;
 	}
 
-	notify_addr = pci_alloc_consistent(vdev->pdev,
-			sizeof(struct vnic_devcmd_notify),
-			&notify_pa);
+	notify_addr = dma_alloc_coherent(&vdev->pdev->dev,
+					 sizeof(struct vnic_devcmd_notify),
+					 &notify_pa, GFP_ATOMIC);
 	if (!notify_addr)
 		return -ENOMEM;
 
@@ -881,10 +866,9 @@ static int vnic_dev_notify_unsetcmd(struct vnic_dev *vdev)
 int vnic_dev_notify_unset(struct vnic_dev *vdev)
 {
 	if (vdev->notify) {
-		pci_free_consistent(vdev->pdev,
-			sizeof(struct vnic_devcmd_notify),
-			vdev->notify,
-			vdev->notify_pa);
+		dma_free_coherent(&vdev->pdev->dev,
+				  sizeof(struct vnic_devcmd_notify),
+				  vdev->notify, vdev->notify_pa);
 	}
 
 	return vnic_dev_notify_unsetcmd(vdev);
@@ -1045,18 +1029,17 @@ void vnic_dev_unregister(struct vnic_dev *vdev)
 {
 	if (vdev) {
 		if (vdev->notify)
-			pci_free_consistent(vdev->pdev,
-				sizeof(struct vnic_devcmd_notify),
-				vdev->notify,
-				vdev->notify_pa);
+			dma_free_coherent(&vdev->pdev->dev,
+					  sizeof(struct vnic_devcmd_notify),
+					  vdev->notify, vdev->notify_pa);
 		if (vdev->stats)
-			pci_free_consistent(vdev->pdev,
-				sizeof(struct vnic_stats),
-				vdev->stats, vdev->stats_pa);
+			dma_free_coherent(&vdev->pdev->dev,
+					  sizeof(struct vnic_stats),
+					  vdev->stats, vdev->stats_pa);
 		if (vdev->fw_info)
-			pci_free_consistent(vdev->pdev,
-				sizeof(struct vnic_devcmd_fw_info),
-				vdev->fw_info, vdev->fw_info_pa);
+			dma_free_coherent(&vdev->pdev->dev,
+					  sizeof(struct vnic_devcmd_fw_info),
+					  vdev->fw_info, vdev->fw_info_pa);
 		if (vdev->devcmd2)
 			vnic_dev_deinit_devcmd2(vdev);
 
@@ -1070,7 +1053,7 @@ struct vnic_dev *vnic_dev_register(struct vnic_dev *vdev,
 	unsigned int num_bars)
 {
 	if (!vdev) {
-		vdev = kzalloc(sizeof(struct vnic_dev), GFP_ATOMIC);
+		vdev = kzalloc(sizeof(struct vnic_dev), GFP_KERNEL);
 		if (!vdev)
 			return NULL;
 	}
@@ -1126,7 +1109,7 @@ int vnic_dev_init_prov2(struct vnic_dev *vdev, u8 *buf, u32 len)
 	void *prov_buf;
 	int ret;
 
-	prov_buf = pci_alloc_consistent(vdev->pdev, len, &prov_pa);
+	prov_buf = dma_alloc_coherent(&vdev->pdev->dev, len, &prov_pa, GFP_ATOMIC);
 	if (!prov_buf)
 		return -ENOMEM;
 
@@ -1136,7 +1119,7 @@ int vnic_dev_init_prov2(struct vnic_dev *vdev, u8 *buf, u32 len)
 
 	ret = vnic_dev_cmd(vdev, CMD_INIT_PROV_INFO2, &a0, &a1, wait);
 
-	pci_free_consistent(vdev->pdev, len, prov_buf, prov_pa);
+	dma_free_coherent(&vdev->pdev->dev, len, prov_buf, prov_pa);
 
 	return ret;
 }
@@ -1216,7 +1199,8 @@ int vnic_dev_classifier(struct vnic_dev *vdev, u8 cmd, u16 *entry,
 		tlv_size = sizeof(struct filter) +
 			   sizeof(struct filter_action) +
 			   2 * sizeof(struct filter_tlv);
-		tlv_va = pci_alloc_consistent(vdev->pdev, tlv_size, &tlv_pa);
+		tlv_va = dma_alloc_coherent(&vdev->pdev->dev, tlv_size,
+					    &tlv_pa, GFP_ATOMIC);
 		if (!tlv_va)
 			return -ENOMEM;
 		tlv = tlv_va;
@@ -1239,7 +1223,7 @@ int vnic_dev_classifier(struct vnic_dev *vdev, u8 cmd, u16 *entry,
 
 		ret = vnic_dev_cmd(vdev, CMD_ADD_FILTER, &a0, &a1, wait);
 		*entry = (u16)a0;
-		pci_free_consistent(vdev->pdev, tlv_size, tlv_va, tlv_pa);
+		dma_free_coherent(&vdev->pdev->dev, tlv_size, tlv_va, tlv_pa);
 	} else if (cmd == CLSF_DEL) {
 		a0 = *entry;
 		ret = vnic_dev_cmd(vdev, CMD_DEL_FILTER, &a0, &a1, wait);
@@ -1268,16 +1252,36 @@ int vnic_dev_overlay_offload_cfg(struct vnic_dev *vdev, u8 overlay,
 }
 
 int vnic_dev_get_supported_feature_ver(struct vnic_dev *vdev, u8 feature,
-				       u64 *supported_versions)
+				       u64 *supported_versions, u64 *a1)
 {
 	u64 a0 = feature;
 	int wait = 1000;
-	u64 a1 = 0;
 	int ret;
 
-	ret = vnic_dev_cmd(vdev, CMD_GET_SUPP_FEATURE_VER, &a0, &a1, wait);
+	ret = vnic_dev_cmd(vdev, CMD_GET_SUPP_FEATURE_VER, &a0, a1, wait);
 	if (!ret)
 		*supported_versions = a0;
 
 	return ret;
 }
+
+int vnic_dev_capable_rss_hash_type(struct vnic_dev *vdev, u8 *rss_hash_type)
+{
+	u64 a0 = CMD_NIC_CFG, a1 = 0;
+	int wait = 1000;
+	int err;
+
+	err = vnic_dev_cmd(vdev, CMD_CAPABILITY, &a0, &a1, wait);
+	/* rss_hash_type is valid only when a0 is 1. Adapter which does not
+	 * support CMD_CAPABILITY for rss_hash_type has a0 = 0
+	 */
+	if (err || (a0 != 1))
+		return -EOPNOTSUPP;
+
+	a1 = (a1 >> NIC_CFG_RSS_HASH_TYPE_SHIFT) &
+	     NIC_CFG_RSS_HASH_TYPE_MASK_FIELD;
+
+	*rss_hash_type = (u8)a1;
+
+	return 0;
+}
diff --git a/drivers/net/ethernet/cisco/enic/vnic_dev.h b/drivers/net/ethernet/cisco/enic/vnic_dev.h
index 9d43d6bb9907..7fdd8c661c99 100644
--- a/drivers/net/ethernet/cisco/enic/vnic_dev.h
+++ b/drivers/net/ethernet/cisco/enic/vnic_dev.h
@@ -1,20 +1,7 @@
+/* SPDX-License-Identifier: GPL-2.0-only */
 /*
  * Copyright 2008-2010 Cisco Systems, Inc.  All rights reserved.
  * Copyright 2007 Nuova Systems, Inc.  All rights reserved.
- *
- * This program is free software; you may redistribute it and/or modify
- * it under the terms of the GNU General Public License as published by
- * the Free Software Foundation; version 2 of the License.
- *
- * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND,
- * EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF
- * MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND
- * NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS
- * BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN
- * ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN
- * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
- * SOFTWARE.
- *
  */
 
 #ifndef _VNIC_DEV_H_
@@ -44,6 +31,11 @@ static inline void writeq(u64 val, void __iomem *reg)
 #undef pr_fmt
 #define pr_fmt(fmt) KBUILD_MODNAME ": " fmt
 
+#define VNIC_DESC_SIZE_ALIGN	16
+#define VNIC_DESC_COUNT_ALIGN	32
+#define VNIC_DESC_BASE_ALIGN	512
+#define VNIC_DESC_MAX_COUNT	4096
+
 enum vnic_dev_intr_mode {
 	VNIC_DEV_INTR_MODE_UNKNOWN,
 	VNIC_DEV_INTR_MODE_INTX,
@@ -183,6 +175,7 @@ int vnic_dev_overlay_offload_ctrl(struct vnic_dev *vdev, u8 overlay, u8 config);
 int vnic_dev_overlay_offload_cfg(struct vnic_dev *vdev, u8 overlay,
 				 u16 vxlan_udp_port_number);
 int vnic_dev_get_supported_feature_ver(struct vnic_dev *vdev, u8 feature,
-				       u64 *supported_versions);
+				       u64 *supported_versions, u64 *a1);
+int vnic_dev_capable_rss_hash_type(struct vnic_dev *vdev, u8 *rss_hash_type);
 
 #endif /* _VNIC_DEV_H_ */
diff --git a/drivers/net/ethernet/cisco/enic/vnic_devcmd.h b/drivers/net/ethernet/cisco/enic/vnic_devcmd.h
index d83880b0d468..605ef17f967e 100644
--- a/drivers/net/ethernet/cisco/enic/vnic_devcmd.h
+++ b/drivers/net/ethernet/cisco/enic/vnic_devcmd.h
@@ -1,20 +1,7 @@
+/* SPDX-License-Identifier: GPL-2.0-only */
 /*
  * Copyright 2008-2010 Cisco Systems, Inc.  All rights reserved.
  * Copyright 2007 Nuova Systems, Inc.  All rights reserved.
- *
- * This program is free software; you may redistribute it and/or modify
- * it under the terms of the GNU General Public License as published by
- * the Free Software Foundation; version 2 of the License.
- *
- * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND,
- * EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF
- * MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND
- * NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS
- * BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN
- * ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN
- * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
- * SOFTWARE.
- *
  */
 
 #ifndef _VNIC_DEVCMD_H_
@@ -148,8 +135,26 @@ enum vnic_devcmd_cmd {
 	/* del VLAN id in (u16)a0 */
 	CMD_VLAN_DEL            = _CMDCNW(_CMD_DIR_WRITE, _CMD_VTYPE_ENET, 15),
 
-	/* nic_cfg in (u32)a0 */
+	/* nic_cfg (no wait, always succeeds)
+	 * in: (u32)a0
+	 *
+	 * Capability query:
+	 * out: (u64) a0 = 1 if a1 is valid
+	 *	(u64) a1 = (NIC_CFG bits supported) | (flags << 32)
+	 *
+	 * flags are CMD_NIC_CFG_CAPF_xxx
+	 */
 	CMD_NIC_CFG             = _CMDCNW(_CMD_DIR_WRITE, _CMD_VTYPE_ALL, 16),
+	/* nic_cfg_chk (will return error if flags are invalid)
+	 * in: (u32)a0
+	 *
+	 * Capability query:
+	 * out: (u64) a0 = 1 if a1 is valid
+	 *	(u64) a1 = (NIC_CFG bits supported) | (flags << 32)
+	 *
+	 * flags are CMD_NIC_CFG_CAPF_xxx
+	 */
+	CMD_NIC_CFG_CHK		= _CMDC(_CMD_DIR_WRITE, _CMD_VTYPE_ALL, 16),
 
 	/* union vnic_rss_key in mem: (u64)a0=paddr, (u16)a1=len */
 	CMD_RSS_KEY             = _CMDC(_CMD_DIR_WRITE, _CMD_VTYPE_ENET, 17),
@@ -431,6 +436,25 @@ enum vnic_devcmd_cmd {
 	 * in: (u16) a2 = unsigned short int port information
 	 */
 	CMD_OVERLAY_OFFLOAD_CFG = _CMDC(_CMD_DIR_WRITE, _CMD_VTYPE_ENET, 73),
+
+	/*
+	 * Set extended CQ field in MREGS of RQ (or all RQs)
+	 * for given vNIC
+	 * in: (u64) a0 = RQ selection (VNIC_RQ_ALL for all RQs)
+	 *     (u32) a1 = CQ entry size
+	 *         VNIC_RQ_CQ_ENTRY_SIZE_16 --> 16 bytes
+	 *         VNIC_RQ_CQ_ENTRY_SIZE_32 --> 32 bytes
+	 *         VNIC_RQ_CQ_ENTRY_SIZE_64 --> 64 bytes
+	 *
+	 * Capability query:
+	 * out: (u32) a0 = errno, 0:valid cmd
+	 *      (u32) a1 = value consisting of supported entries
+	 *         bit 0: 16 bytes
+	 *         bit 1: 32 bytes
+	 *         bit 2: 64 bytes
+	 */
+	CMD_CQ_ENTRY_SIZE_SET = _CMDC(_CMD_DIR_WRITE, _CMD_VTYPE_ENET, 90),
+
 };
 
 /* CMD_ENABLE2 flags */
@@ -439,6 +463,7 @@ enum vnic_devcmd_cmd {
 
 /* flags for CMD_OPEN */
 #define CMD_OPENF_OPROM		0x1	/* open coming from option rom */
+#define CMD_OPENF_IG_DESCCACHE	0x2	/* Do not flush IG DESC cache */
 
 /* flags for CMD_INIT */
 #define CMD_INITF_DEFAULT_MAC	0x1	/* init with default mac addr */
@@ -522,7 +547,7 @@ struct vnic_devcmd_notify {
 struct vnic_devcmd_provinfo {
 	u8 oui[3];
 	u8 type;
-	u8 data[0];
+	u8 data[];
 };
 
 /* These are used in flags field of different filters to denote
@@ -629,9 +654,9 @@ enum {
 #define FILTER_MAX_BUF_SIZE 100
 
 struct filter_tlv {
-	u_int32_t type;
-	u_int32_t length;
-	u_int32_t val[0];
+	u32 type;
+	u32 length;
+	u32 val[];
 };
 
 enum {
@@ -697,6 +722,10 @@ enum overlay_ofld_cmd {
 
 #define OVERLAY_CFG_VXLAN_PORT_UPDATE	0
 
+#define ENIC_VXLAN_INNER_IPV6		BIT(0)
+#define ENIC_VXLAN_OUTER_IPV6		BIT(1)
+#define ENIC_VXLAN_MULTI_WQ		BIT(2)
+
 /* Use this enum to get the supported versions for each of these features
  * If you need to use the devcmd_get_supported_feature_version(), add
  * the new feature into this enum and install function handler in devcmd.c
diff --git a/drivers/net/ethernet/cisco/enic/vnic_enet.h b/drivers/net/ethernet/cisco/enic/vnic_enet.h
index 7d6fbb5635a4..9e8e86262a3f 100644
--- a/drivers/net/ethernet/cisco/enic/vnic_enet.h
+++ b/drivers/net/ethernet/cisco/enic/vnic_enet.h
@@ -1,20 +1,7 @@
+/* SPDX-License-Identifier: GPL-2.0-only */
 /*
  * Copyright 2008-2010 Cisco Systems, Inc.  All rights reserved.
  * Copyright 2007 Nuova Systems, Inc.  All rights reserved.
- *
- * This program is free software; you may redistribute it and/or modify
- * it under the terms of the GNU General Public License as published by
- * the Free Software Foundation; version 2 of the License.
- *
- * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND,
- * EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF
- * MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND
- * NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS
- * BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN
- * ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN
- * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
- * SOFTWARE.
- *
  */
 
 #ifndef _VNIC_ENIC_H_
@@ -34,6 +21,11 @@ struct vnic_enet_config {
 	u16 loop_tag;
 	u16 vf_rq_count;
 	u16 num_arfs;
+	u8 reserved[66];
+	u32 max_rq_ring;	// MAX RQ ring size
+	u32 max_wq_ring;	// MAX WQ ring size
+	u32 max_cq_ring;	// MAX CQ ring size
+	u32 rdma_rsvd_lkey;	// Reserved (privileged) LKey
 };
 
 #define VENETF_TSO		0x1	/* TSO enabled */
diff --git a/drivers/net/ethernet/cisco/enic/vnic_intr.c b/drivers/net/ethernet/cisco/enic/vnic_intr.c
index 23604e3d4455..25319f072a04 100644
--- a/drivers/net/ethernet/cisco/enic/vnic_intr.c
+++ b/drivers/net/ethernet/cisco/enic/vnic_intr.c
@@ -1,20 +1,7 @@
+// SPDX-License-Identifier: GPL-2.0-only
 /*
  * Copyright 2008-2010 Cisco Systems, Inc.  All rights reserved.
  * Copyright 2007 Nuova Systems, Inc.  All rights reserved.
- *
- * This program is free software; you may redistribute it and/or modify
- * it under the terms of the GNU General Public License as published by
- * the Free Software Foundation; version 2 of the License.
- *
- * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND,
- * EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF
- * MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND
- * NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS
- * BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN
- * ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN
- * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
- * SOFTWARE.
- *
  */
 
 #include <linux/kernel.h>
diff --git a/drivers/net/ethernet/cisco/enic/vnic_intr.h b/drivers/net/ethernet/cisco/enic/vnic_intr.h
index 2b1636392294..33a72aa10b26 100644
--- a/drivers/net/ethernet/cisco/enic/vnic_intr.h
+++ b/drivers/net/ethernet/cisco/enic/vnic_intr.h
@@ -1,20 +1,7 @@
+/* SPDX-License-Identifier: GPL-2.0-only */
 /*
  * Copyright 2008-2010 Cisco Systems, Inc.  All rights reserved.
  * Copyright 2007 Nuova Systems, Inc.  All rights reserved.
- *
- * This program is free software; you may redistribute it and/or modify
- * it under the terms of the GNU General Public License as published by
- * the Free Software Foundation; version 2 of the License.
- *
- * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND,
- * EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF
- * MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND
- * NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS
- * BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN
- * ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN
- * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
- * SOFTWARE.
- *
  */
 
 #ifndef _VNIC_INTR_H_
diff --git a/drivers/net/ethernet/cisco/enic/vnic_nic.h b/drivers/net/ethernet/cisco/enic/vnic_nic.h
index 995a50dd4c99..04fee45b5d39 100644
--- a/drivers/net/ethernet/cisco/enic/vnic_nic.h
+++ b/drivers/net/ethernet/cisco/enic/vnic_nic.h
@@ -1,20 +1,7 @@
+/* SPDX-License-Identifier: GPL-2.0-only */
 /*
  * Copyright 2008-2010 Cisco Systems, Inc.  All rights reserved.
  * Copyright 2007 Nuova Systems, Inc.  All rights reserved.
- *
- * This program is free software; you may redistribute it and/or modify
- * it under the terms of the GNU General Public License as published by
- * the Free Software Foundation; version 2 of the License.
- *
- * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND,
- * EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF
- * MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND
- * NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS
- * BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN
- * ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN
- * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
- * SOFTWARE.
- *
  */
 
 #ifndef _VNIC_NIC_H_
@@ -41,12 +28,14 @@
 #define NIC_CFG_IG_VLAN_STRIP_EN_MASK_FIELD	1UL
 #define NIC_CFG_IG_VLAN_STRIP_EN_SHIFT		24
 
+#define NIC_CFG_RSS_HASH_TYPE_UDP_IPV4		(1 << 0)
 #define NIC_CFG_RSS_HASH_TYPE_IPV4		(1 << 1)
 #define NIC_CFG_RSS_HASH_TYPE_TCP_IPV4		(1 << 2)
 #define NIC_CFG_RSS_HASH_TYPE_IPV6		(1 << 3)
 #define NIC_CFG_RSS_HASH_TYPE_TCP_IPV6		(1 << 4)
 #define NIC_CFG_RSS_HASH_TYPE_IPV6_EX		(1 << 5)
 #define NIC_CFG_RSS_HASH_TYPE_TCP_IPV6_EX	(1 << 6)
+#define NIC_CFG_RSS_HASH_TYPE_UDP_IPV6		(1 << 7)
 
 static inline void vnic_set_nic_cfg(u32 *nic_cfg,
 	u8 rss_default_cpu, u8 rss_hash_type,
diff --git a/drivers/net/ethernet/cisco/enic/vnic_resource.h b/drivers/net/ethernet/cisco/enic/vnic_resource.h
index 4e45f88ac1d4..b4776e334d63 100644
--- a/drivers/net/ethernet/cisco/enic/vnic_resource.h
+++ b/drivers/net/ethernet/cisco/enic/vnic_resource.h
@@ -1,20 +1,7 @@
+/* SPDX-License-Identifier: GPL-2.0-only */
 /*
  * Copyright 2008-2010 Cisco Systems, Inc.  All rights reserved.
  * Copyright 2007 Nuova Systems, Inc.  All rights reserved.
- *
- * This program is free software; you may redistribute it and/or modify
- * it under the terms of the GNU General Public License as published by
- * the Free Software Foundation; version 2 of the License.
- *
- * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND,
- * EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF
- * MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND
- * NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS
- * BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN
- * ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN
- * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
- * SOFTWARE.
- *
  */
 
 #ifndef _VNIC_RESOURCE_H_
diff --git a/drivers/net/ethernet/cisco/enic/vnic_rq.c b/drivers/net/ethernet/cisco/enic/vnic_rq.c
index 36bc2c71fba9..5ae80551f17c 100644
--- a/drivers/net/ethernet/cisco/enic/vnic_rq.c
+++ b/drivers/net/ethernet/cisco/enic/vnic_rq.c
@@ -1,20 +1,7 @@
+// SPDX-License-Identifier: GPL-2.0-only
 /*
  * Copyright 2008-2010 Cisco Systems, Inc.  All rights reserved.
  * Copyright 2007 Nuova Systems, Inc.  All rights reserved.
- *
- * This program is free software; you may redistribute it and/or modify
- * it under the terms of the GNU General Public License as published by
- * the Free Software Foundation; version 2 of the License.
- *
- * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND,
- * EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF
- * MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND
- * NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS
- * BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN
- * ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN
- * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
- * SOFTWARE.
- *
  */
 
 #include <linux/kernel.h>
@@ -35,7 +22,7 @@ static int vnic_rq_alloc_bufs(struct vnic_rq *rq)
 	unsigned int blks = VNIC_RQ_BUF_BLKS_NEEDED(count);
 
 	for (i = 0; i < blks; i++) {
-		rq->bufs[i] = kzalloc(VNIC_RQ_BUF_BLK_SZ(count), GFP_ATOMIC);
+		rq->bufs[i] = kzalloc(VNIC_RQ_BUF_BLK_SZ(count), GFP_KERNEL);
 		if (!rq->bufs[i])
 			return -ENOMEM;
 	}
@@ -139,20 +126,8 @@ void vnic_rq_init(struct vnic_rq *rq, unsigned int cq_index,
 	unsigned int error_interrupt_enable,
 	unsigned int error_interrupt_offset)
 {
-	u32 fetch_index = 0;
-
-	/* Use current fetch_index as the ring starting point */
-	fetch_index = ioread32(&rq->ctrl->fetch_index);
-
-	if (fetch_index == 0xFFFFFFFF) { /* check for hardware gone  */
-		/* Hardware surprise removal: reset fetch_index */
-		fetch_index = 0;
-	}
-
-	vnic_rq_init_start(rq, cq_index,
-		fetch_index, fetch_index,
-		error_interrupt_enable,
-		error_interrupt_offset);
+	vnic_rq_init_start(rq, cq_index, 0, 0, error_interrupt_enable,
+			   error_interrupt_offset);
 }
 
 unsigned int vnic_rq_error_status(struct vnic_rq *rq)
@@ -228,4 +203,3 @@ void vnic_rq_clean(struct vnic_rq *rq,
 
 	vnic_dev_clear_desc_ring(&rq->ring);
 }
-
diff --git a/drivers/net/ethernet/cisco/enic/vnic_rq.h b/drivers/net/ethernet/cisco/enic/vnic_rq.h
index 0413103ebe94..a1cdd729caec 100644
--- a/drivers/net/ethernet/cisco/enic/vnic_rq.h
+++ b/drivers/net/ethernet/cisco/enic/vnic_rq.h
@@ -1,20 +1,7 @@
+/* SPDX-License-Identifier: GPL-2.0-only */
 /*
  * Copyright 2008-2010 Cisco Systems, Inc.  All rights reserved.
  * Copyright 2007 Nuova Systems, Inc.  All rights reserved.
- *
- * This program is free software; you may redistribute it and/or modify
- * it under the terms of the GNU General Public License as published by
- * the Free Software Foundation; version 2 of the License.
- *
- * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND,
- * EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF
- * MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND
- * NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS
- * BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN
- * ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN
- * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
- * SOFTWARE.
- *
  */
 
 #ifndef _VNIC_RQ_H_
@@ -63,7 +50,7 @@ struct vnic_rq_ctrl {
 	(VNIC_RQ_BUF_BLK_ENTRIES(entries) * sizeof(struct vnic_rq_buf))
 #define VNIC_RQ_BUF_BLKS_NEEDED(entries) \
 	DIV_ROUND_UP(entries, VNIC_RQ_BUF_BLK_ENTRIES(entries))
-#define VNIC_RQ_BUF_BLKS_MAX VNIC_RQ_BUF_BLKS_NEEDED(4096)
+#define VNIC_RQ_BUF_BLKS_MAX VNIC_RQ_BUF_BLKS_NEEDED(16384)
 
 struct vnic_rq_buf {
 	struct vnic_rq_buf *next;
@@ -74,6 +61,8 @@ struct vnic_rq_buf {
 	unsigned int index;
 	void *desc;
 	uint64_t wr_id;
+	unsigned int offset;
+	unsigned int truesize;
 };
 
 enum enic_poll_state {
diff --git a/drivers/net/ethernet/cisco/enic/vnic_rss.h b/drivers/net/ethernet/cisco/enic/vnic_rss.h
index 881fa18542b3..4dcf0e61cb13 100644
--- a/drivers/net/ethernet/cisco/enic/vnic_rss.h
+++ b/drivers/net/ethernet/cisco/enic/vnic_rss.h
@@ -1,19 +1,7 @@
+/* SPDX-License-Identifier: GPL-2.0-only */
 /*
  * Copyright 2008-2010 Cisco Systems, Inc.  All rights reserved.
  * Copyright 2007 Nuova Systems, Inc.  All rights reserved.
- *
- * This program is free software; you may redistribute it and/or modify
- * it under the terms of the GNU General Public License as published by
- * the Free Software Foundation; version 2 of the License.
- *
- * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND,
- * EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF
- * MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND
- * NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS
- * BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN
- * ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN
- * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
- * SOFTWARE.
  */
 
 #ifndef _VNIC_RSS_H_
diff --git a/drivers/net/ethernet/cisco/enic/vnic_stats.h b/drivers/net/ethernet/cisco/enic/vnic_stats.h
index 74c81ed6fdab..2dd04322d760 100644
--- a/drivers/net/ethernet/cisco/enic/vnic_stats.h
+++ b/drivers/net/ethernet/cisco/enic/vnic_stats.h
@@ -1,20 +1,7 @@
+/* SPDX-License-Identifier: GPL-2.0-only */
 /*
  * Copyright 2008-2010 Cisco Systems, Inc.  All rights reserved.
  * Copyright 2007 Nuova Systems, Inc.  All rights reserved.
- *
- * This program is free software; you may redistribute it and/or modify
- * it under the terms of the GNU General Public License as published by
- * the Free Software Foundation; version 2 of the License.
- *
- * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND,
- * EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF
- * MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND
- * NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS
- * BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN
- * ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN
- * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
- * SOFTWARE.
- *
  */
 
 #ifndef _VNIC_STATS_H_
diff --git a/drivers/net/ethernet/cisco/enic/vnic_vic.c b/drivers/net/ethernet/cisco/enic/vnic_vic.c
index 24ef8cd40545..66b577835338 100644
--- a/drivers/net/ethernet/cisco/enic/vnic_vic.c
+++ b/drivers/net/ethernet/cisco/enic/vnic_vic.c
@@ -1,20 +1,5 @@
-/*
- * Copyright 2010 Cisco Systems, Inc.  All rights reserved.
- *
- * This program is free software; you may redistribute it and/or modify
- * it under the terms of the GNU General Public License as published by
- * the Free Software Foundation; version 2 of the License.
- *
- * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND,
- * EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF
- * MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND
- * NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS
- * BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN
- * ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN
- * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
- * SOFTWARE.
- *
- */
+// SPDX-License-Identifier: GPL-2.0-only
+// Copyright 2010 Cisco Systems, Inc.  All rights reserved.
 
 #include <linux/kernel.h>
 #include <linux/errno.h>
@@ -64,7 +49,8 @@ int vic_provinfo_add_tlv(struct vic_provinfo *vp, u16 type, u16 length,
 
 	tlv->type = htons(type);
 	tlv->length = htons(length);
-	memcpy(tlv->value, value, length);
+	unsafe_memcpy(tlv->value, value, length,
+		      /* Flexible array of flexible arrays */);
 
 	vp->num_tlvs = htonl(ntohl(vp->num_tlvs) + 1);
 	vp->length = htonl(ntohl(vp->length) +
diff --git a/drivers/net/ethernet/cisco/enic/vnic_vic.h b/drivers/net/ethernet/cisco/enic/vnic_vic.h
index 9ef81f148351..b51c1c52f8bf 100644
--- a/drivers/net/ethernet/cisco/enic/vnic_vic.h
+++ b/drivers/net/ethernet/cisco/enic/vnic_vic.h
@@ -1,20 +1,5 @@
-/*
- * Copyright 2010 Cisco Systems, Inc.  All rights reserved.
- *
- * This program is free software; you may redistribute it and/or modify
- * it under the terms of the GNU General Public License as published by
- * the Free Software Foundation; version 2 of the License.
- *
- * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND,
- * EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF
- * MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND
- * NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS
- * BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN
- * ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN
- * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
- * SOFTWARE.
- *
- */
+/* SPDX-License-Identifier: GPL-2.0-only */
+/* Copyright 2010 Cisco Systems, Inc.  All rights reserved. */
 
 #ifndef _VNIC_VIC_H_
 #define _VNIC_VIC_H_
@@ -59,7 +44,7 @@ struct vic_provinfo {
 		u16 type;
 		u16 length;
 		u8 value[0];
-	} tlv[0];
+	} tlv[];
 } __packed;
 
 #define VIC_PROVINFO_ADD_TLV(vp, tlvtype, tlvlen, data) \
diff --git a/drivers/net/ethernet/cisco/enic/vnic_wq.c b/drivers/net/ethernet/cisco/enic/vnic_wq.c
index 090cc65658a3..29c7900349b2 100644
--- a/drivers/net/ethernet/cisco/enic/vnic_wq.c
+++ b/drivers/net/ethernet/cisco/enic/vnic_wq.c
@@ -1,20 +1,7 @@
+// SPDX-License-Identifier: GPL-2.0-only
 /*
  * Copyright 2008-2010 Cisco Systems, Inc.  All rights reserved.
  * Copyright 2007 Nuova Systems, Inc.  All rights reserved.
- *
- * This program is free software; you may redistribute it and/or modify
- * it under the terms of the GNU General Public License as published by
- * the Free Software Foundation; version 2 of the License.
- *
- * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND,
- * EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF
- * MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND
- * NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS
- * BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN
- * ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN
- * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
- * SOFTWARE.
- *
  */
 
 #include <linux/kernel.h>
@@ -35,7 +22,7 @@ static int vnic_wq_alloc_bufs(struct vnic_wq *wq)
 	unsigned int blks = VNIC_WQ_BUF_BLKS_NEEDED(count);
 
 	for (i = 0; i < blks; i++) {
-		wq->bufs[i] = kzalloc(VNIC_WQ_BUF_BLK_SZ(count), GFP_ATOMIC);
+		wq->bufs[i] = kzalloc(VNIC_WQ_BUF_BLK_SZ(count), GFP_KERNEL);
 		if (!wq->bufs[i])
 			return -ENOMEM;
 	}
diff --git a/drivers/net/ethernet/cisco/enic/vnic_wq.h b/drivers/net/ethernet/cisco/enic/vnic_wq.h
index 01209613d57d..3bb4758100ba 100644
--- a/drivers/net/ethernet/cisco/enic/vnic_wq.h
+++ b/drivers/net/ethernet/cisco/enic/vnic_wq.h
@@ -1,20 +1,7 @@
+/* SPDX-License-Identifier: GPL-2.0-only */
 /*
  * Copyright 2008-2010 Cisco Systems, Inc.  All rights reserved.
  * Copyright 2007 Nuova Systems, Inc.  All rights reserved.
- *
- * This program is free software; you may redistribute it and/or modify
- * it under the terms of the GNU General Public License as published by
- * the Free Software Foundation; version 2 of the License.
- *
- * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND,
- * EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF
- * MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND
- * NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS
- * BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN
- * ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN
- * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
- * SOFTWARE.
- *
  */
 
 #ifndef _VNIC_WQ_H_
@@ -75,7 +62,7 @@ struct vnic_wq_buf {
 	(VNIC_WQ_BUF_BLK_ENTRIES(entries) * sizeof(struct vnic_wq_buf))
 #define VNIC_WQ_BUF_BLKS_NEEDED(entries) \
 	DIV_ROUND_UP(entries, VNIC_WQ_BUF_BLK_ENTRIES(entries))
-#define VNIC_WQ_BUF_BLKS_MAX VNIC_WQ_BUF_BLKS_NEEDED(4096)
+#define VNIC_WQ_BUF_BLKS_MAX VNIC_WQ_BUF_BLKS_NEEDED(16384)
 
 struct vnic_wq {
 	unsigned int index;
diff --git a/drivers/net/ethernet/cisco/enic/wq_enet_desc.h b/drivers/net/ethernet/cisco/enic/wq_enet_desc.h
index c7021e3a631f..425e46a804ee 100644
--- a/drivers/net/ethernet/cisco/enic/wq_enet_desc.h
+++ b/drivers/net/ethernet/cisco/enic/wq_enet_desc.h
@@ -1,20 +1,7 @@
+/* SPDX-License-Identifier: GPL-2.0-only */
 /*
  * Copyright 2008-2010 Cisco Systems, Inc.  All rights reserved.
  * Copyright 2007 Nuova Systems, Inc.  All rights reserved.
- *
- * This program is free software; you may redistribute it and/or modify
- * it under the terms of the GNU General Public License as published by
- * the Free Software Foundation; version 2 of the License.
- *
- * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND,
- * EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF
- * MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND
- * NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS
- * BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN
- * ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN
- * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
- * SOFTWARE.
- *
  */
 
 #ifndef _WQ_ENET_DESC_H_
diff --git a/drivers/net/ethernet/cortina/Kconfig b/drivers/net/ethernet/cortina/Kconfig
new file mode 100644
index 000000000000..aaf9e294b70b
--- /dev/null
+++ b/drivers/net/ethernet/cortina/Kconfig
@@ -0,0 +1,23 @@
+# SPDX-License-Identifier: GPL-2.0
+# Cortina ethernet devices
+
+config NET_VENDOR_CORTINA
+	bool "Cortina Gemini devices"
+	default y
+	help
+	  If you have a network (Ethernet) card belonging to this class, say Y
+	  and read the Ethernet-HOWTO, available from
+	  <https://www.tldp.org/docs.html#howto>.
+
+if NET_VENDOR_CORTINA
+
+config GEMINI_ETHERNET
+	tristate "Gemini Gigabit Ethernet support"
+	depends on OF
+	depends on HAS_IOMEM
+	select PHYLIB
+	select CRC32
+	help
+	  This driver supports StorLink SL351x (Gemini) dual Gigabit Ethernet.
+
+endif # NET_VENDOR_CORTINA
diff --git a/drivers/net/ethernet/cortina/Makefile b/drivers/net/ethernet/cortina/Makefile
new file mode 100644
index 000000000000..4e86d398a89c
--- /dev/null
+++ b/drivers/net/ethernet/cortina/Makefile
@@ -0,0 +1,4 @@
+# SPDX-License-Identifier: GPL-2.0
+# Makefile for the Cortina Gemini network device drivers.
+
+obj-$(CONFIG_GEMINI_ETHERNET) += gemini.o
diff --git a/drivers/net/ethernet/cortina/gemini.c b/drivers/net/ethernet/cortina/gemini.c
new file mode 100644
index 000000000000..6a2004bbe87f
--- /dev/null
+++ b/drivers/net/ethernet/cortina/gemini.c
@@ -0,0 +1,2692 @@
+// SPDX-License-Identifier: GPL-2.0
+/* Ethernet device driver for Cortina Systems Gemini SoC
+ * Also known as the StorLink SL3512 and SL3516 (SL351x) or Lepus
+ * Net Engine and Gigabit Ethernet MAC (GMAC)
+ * This hardware contains a TCP Offload Engine (TOE) but currently the
+ * driver does not make use of it.
+ *
+ * Authors:
+ * Linus Walleij <linus.walleij@linaro.org>
+ * Tobias Waldvogel <tobias.waldvogel@gmail.com> (OpenWRT)
+ * Michał Mirosław <mirq-linux@rere.qmqm.pl>
+ * Paulius Zaleckas <paulius.zaleckas@gmail.com>
+ * Giuseppe De Robertis <Giuseppe.DeRobertis@ba.infn.it>
+ * Gary Chen & Ch Hsu Storlink Semiconductor
+ */
+#include <linux/kernel.h>
+#include <linux/init.h>
+#include <linux/module.h>
+#include <linux/platform_device.h>
+#include <linux/spinlock.h>
+#include <linux/slab.h>
+#include <linux/dma-mapping.h>
+#include <linux/cache.h>
+#include <linux/interrupt.h>
+#include <linux/reset.h>
+#include <linux/clk.h>
+#include <linux/of.h>
+#include <linux/of_mdio.h>
+#include <linux/of_net.h>
+#include <linux/of_platform.h>
+#include <linux/etherdevice.h>
+#include <linux/if_vlan.h>
+#include <linux/skbuff.h>
+#include <linux/phy.h>
+#include <linux/crc32.h>
+#include <linux/ethtool.h>
+#include <linux/tcp.h>
+#include <linux/u64_stats_sync.h>
+
+#include <linux/in.h>
+#include <linux/ip.h>
+#include <linux/ipv6.h>
+#include <net/gro.h>
+
+#include "gemini.h"
+
+#define DRV_NAME		"gmac-gemini"
+
+#define DEFAULT_MSG_ENABLE (NETIF_MSG_DRV | NETIF_MSG_PROBE | NETIF_MSG_LINK)
+static int debug = -1;
+module_param(debug, int, 0);
+MODULE_PARM_DESC(debug, "Debug level (0=none,...,16=all)");
+
+#define HSIZE_8			0x00
+#define HSIZE_16		0x01
+#define HSIZE_32		0x02
+
+#define HBURST_SINGLE		0x00
+#define HBURST_INCR		0x01
+#define HBURST_INCR4		0x02
+#define HBURST_INCR8		0x03
+
+#define HPROT_DATA_CACHE	BIT(0)
+#define HPROT_PRIVILIGED	BIT(1)
+#define HPROT_BUFFERABLE	BIT(2)
+#define HPROT_CACHABLE		BIT(3)
+
+#define DEFAULT_RX_COALESCE_NSECS	0
+#define DEFAULT_GMAC_RXQ_ORDER		9
+#define DEFAULT_GMAC_TXQ_ORDER		8
+#define DEFAULT_RX_BUF_ORDER		11
+#define TX_MAX_FRAGS			16
+#define TX_QUEUE_NUM			1	/* max: 6 */
+#define RX_MAX_ALLOC_ORDER		2
+
+#define GMAC0_IRQ0_2 (GMAC0_TXDERR_INT_BIT | GMAC0_TXPERR_INT_BIT | \
+		      GMAC0_RXDERR_INT_BIT | GMAC0_RXPERR_INT_BIT)
+#define GMAC0_IRQ0_TXQ0_INTS (GMAC0_SWTQ00_EOF_INT_BIT | \
+			      GMAC0_SWTQ00_FIN_INT_BIT)
+#define GMAC0_IRQ4_8 (GMAC0_MIB_INT_BIT | GMAC0_RX_OVERRUN_INT_BIT)
+
+#define GMAC_OFFLOAD_FEATURES (NETIF_F_SG | NETIF_F_IP_CSUM | \
+			       NETIF_F_IPV6_CSUM | NETIF_F_RXCSUM | \
+			       NETIF_F_TSO | NETIF_F_TSO_ECN | NETIF_F_TSO6)
+
+/**
+ * struct gmac_queue_page - page buffer per-page info
+ * @page: the page struct
+ * @mapping: the dma address handle
+ */
+struct gmac_queue_page {
+	struct page *page;
+	dma_addr_t mapping;
+};
+
+struct gmac_txq {
+	struct gmac_txdesc *ring;
+	struct sk_buff	**skb;
+	unsigned int	cptr;
+	unsigned int	noirq_packets;
+};
+
+struct gemini_ethernet;
+
+struct gemini_ethernet_port {
+	u8 id; /* 0 or 1 */
+
+	struct gemini_ethernet *geth;
+	struct net_device *netdev;
+	struct device *dev;
+	void __iomem *dma_base;
+	void __iomem *gmac_base;
+	struct clk *pclk;
+	struct reset_control *reset;
+	int irq;
+	__le32 mac_addr[3];
+
+	void __iomem		*rxq_rwptr;
+	struct gmac_rxdesc	*rxq_ring;
+	unsigned int		rxq_order;
+
+	struct napi_struct	napi;
+	struct hrtimer		rx_coalesce_timer;
+	unsigned int		rx_coalesce_nsecs;
+	unsigned int		freeq_refill;
+	struct gmac_txq		txq[TX_QUEUE_NUM];
+	unsigned int		txq_order;
+	unsigned int		irq_every_tx_packets;
+
+	dma_addr_t		rxq_dma_base;
+	dma_addr_t		txq_dma_base;
+
+	unsigned int		msg_enable;
+	spinlock_t		config_lock; /* Locks config register */
+
+	struct u64_stats_sync	tx_stats_syncp;
+	struct u64_stats_sync	rx_stats_syncp;
+	struct u64_stats_sync	ir_stats_syncp;
+
+	struct rtnl_link_stats64 stats;
+	u64			hw_stats[RX_STATS_NUM];
+	u64			rx_stats[RX_STATUS_NUM];
+	u64			rx_csum_stats[RX_CHKSUM_NUM];
+	u64			rx_napi_exits;
+	u64			tx_frag_stats[TX_MAX_FRAGS];
+	u64			tx_frags_linearized;
+	u64			tx_hw_csummed;
+};
+
+struct gemini_ethernet {
+	struct device *dev;
+	void __iomem *base;
+	struct gemini_ethernet_port *port0;
+	struct gemini_ethernet_port *port1;
+	bool initialized;
+
+	spinlock_t	irq_lock; /* Locks IRQ-related registers */
+	unsigned int	freeq_order;
+	unsigned int	freeq_frag_order;
+	struct gmac_rxdesc *freeq_ring;
+	dma_addr_t	freeq_dma_base;
+	struct gmac_queue_page	*freeq_pages;
+	unsigned int	num_freeq_pages;
+	spinlock_t	freeq_lock; /* Locks queue from reentrance */
+};
+
+#define GMAC_STATS_NUM	( \
+	RX_STATS_NUM + RX_STATUS_NUM + RX_CHKSUM_NUM + 1 + \
+	TX_MAX_FRAGS + 2)
+
+static const char gmac_stats_strings[GMAC_STATS_NUM][ETH_GSTRING_LEN] = {
+	"GMAC_IN_DISCARDS",
+	"GMAC_IN_ERRORS",
+	"GMAC_IN_MCAST",
+	"GMAC_IN_BCAST",
+	"GMAC_IN_MAC1",
+	"GMAC_IN_MAC2",
+	"RX_STATUS_GOOD_FRAME",
+	"RX_STATUS_TOO_LONG_GOOD_CRC",
+	"RX_STATUS_RUNT_FRAME",
+	"RX_STATUS_SFD_NOT_FOUND",
+	"RX_STATUS_CRC_ERROR",
+	"RX_STATUS_TOO_LONG_BAD_CRC",
+	"RX_STATUS_ALIGNMENT_ERROR",
+	"RX_STATUS_TOO_LONG_BAD_ALIGN",
+	"RX_STATUS_RX_ERR",
+	"RX_STATUS_DA_FILTERED",
+	"RX_STATUS_BUFFER_FULL",
+	"RX_STATUS_11",
+	"RX_STATUS_12",
+	"RX_STATUS_13",
+	"RX_STATUS_14",
+	"RX_STATUS_15",
+	"RX_CHKSUM_IP_UDP_TCP_OK",
+	"RX_CHKSUM_IP_OK_ONLY",
+	"RX_CHKSUM_NONE",
+	"RX_CHKSUM_3",
+	"RX_CHKSUM_IP_ERR_UNKNOWN",
+	"RX_CHKSUM_IP_ERR",
+	"RX_CHKSUM_TCP_UDP_ERR",
+	"RX_CHKSUM_7",
+	"RX_NAPI_EXITS",
+	"TX_FRAGS[1]",
+	"TX_FRAGS[2]",
+	"TX_FRAGS[3]",
+	"TX_FRAGS[4]",
+	"TX_FRAGS[5]",
+	"TX_FRAGS[6]",
+	"TX_FRAGS[7]",
+	"TX_FRAGS[8]",
+	"TX_FRAGS[9]",
+	"TX_FRAGS[10]",
+	"TX_FRAGS[11]",
+	"TX_FRAGS[12]",
+	"TX_FRAGS[13]",
+	"TX_FRAGS[14]",
+	"TX_FRAGS[15]",
+	"TX_FRAGS[16+]",
+	"TX_FRAGS_LINEARIZED",
+	"TX_HW_CSUMMED",
+};
+
+static void gmac_dump_dma_state(struct net_device *netdev);
+
+static void gmac_update_config0_reg(struct net_device *netdev,
+				    u32 val, u32 vmask)
+{
+	struct gemini_ethernet_port *port = netdev_priv(netdev);
+	unsigned long flags;
+	u32 reg;
+
+	spin_lock_irqsave(&port->config_lock, flags);
+
+	reg = readl(port->gmac_base + GMAC_CONFIG0);
+	reg = (reg & ~vmask) | val;
+	writel(reg, port->gmac_base + GMAC_CONFIG0);
+
+	spin_unlock_irqrestore(&port->config_lock, flags);
+}
+
+static void gmac_enable_tx_rx(struct net_device *netdev)
+{
+	struct gemini_ethernet_port *port = netdev_priv(netdev);
+	unsigned long flags;
+	u32 reg;
+
+	spin_lock_irqsave(&port->config_lock, flags);
+
+	reg = readl(port->gmac_base + GMAC_CONFIG0);
+	reg &= ~CONFIG0_TX_RX_DISABLE;
+	writel(reg, port->gmac_base + GMAC_CONFIG0);
+
+	spin_unlock_irqrestore(&port->config_lock, flags);
+}
+
+static void gmac_disable_tx_rx(struct net_device *netdev)
+{
+	struct gemini_ethernet_port *port = netdev_priv(netdev);
+	unsigned long flags;
+	u32 val;
+
+	spin_lock_irqsave(&port->config_lock, flags);
+
+	val = readl(port->gmac_base + GMAC_CONFIG0);
+	val |= CONFIG0_TX_RX_DISABLE;
+	writel(val, port->gmac_base + GMAC_CONFIG0);
+
+	spin_unlock_irqrestore(&port->config_lock, flags);
+
+	mdelay(10);	/* let GMAC consume packet */
+}
+
+static void gmac_set_flow_control(struct net_device *netdev, bool tx, bool rx)
+{
+	struct gemini_ethernet_port *port = netdev_priv(netdev);
+	unsigned long flags;
+	u32 val;
+
+	spin_lock_irqsave(&port->config_lock, flags);
+
+	val = readl(port->gmac_base + GMAC_CONFIG0);
+	val &= ~CONFIG0_FLOW_CTL;
+	if (tx)
+		val |= CONFIG0_FLOW_TX;
+	if (rx)
+		val |= CONFIG0_FLOW_RX;
+	writel(val, port->gmac_base + GMAC_CONFIG0);
+
+	spin_unlock_irqrestore(&port->config_lock, flags);
+}
+
+static void gmac_adjust_link(struct net_device *netdev)
+{
+	struct gemini_ethernet_port *port = netdev_priv(netdev);
+	struct phy_device *phydev = netdev->phydev;
+	union gmac_status status, old_status;
+	bool pause_tx = false;
+	bool pause_rx = false;
+
+	status.bits32 = readl(port->gmac_base + GMAC_STATUS);
+	old_status.bits32 = status.bits32;
+	status.bits.link = phydev->link;
+	status.bits.duplex = phydev->duplex;
+
+	switch (phydev->speed) {
+	case 1000:
+		status.bits.speed = GMAC_SPEED_1000;
+		if (phy_interface_mode_is_rgmii(phydev->interface))
+			status.bits.mii_rmii = GMAC_PHY_RGMII_1000;
+		netdev_dbg(netdev, "connect %s to RGMII @ 1Gbit\n",
+			   phydev_name(phydev));
+		break;
+	case 100:
+		status.bits.speed = GMAC_SPEED_100;
+		if (phy_interface_mode_is_rgmii(phydev->interface))
+			status.bits.mii_rmii = GMAC_PHY_RGMII_100_10;
+		netdev_dbg(netdev, "connect %s to RGMII @ 100 Mbit\n",
+			   phydev_name(phydev));
+		break;
+	case 10:
+		status.bits.speed = GMAC_SPEED_10;
+		if (phy_interface_mode_is_rgmii(phydev->interface))
+			status.bits.mii_rmii = GMAC_PHY_RGMII_100_10;
+		netdev_dbg(netdev, "connect %s to RGMII @ 10 Mbit\n",
+			   phydev_name(phydev));
+		break;
+	default:
+		netdev_warn(netdev, "Unsupported PHY speed (%d) on %s\n",
+			    phydev->speed, phydev_name(phydev));
+	}
+
+	if (phydev->duplex == DUPLEX_FULL) {
+		phy_get_pause(phydev, &pause_tx, &pause_rx);
+		netdev_dbg(netdev, "set negotiated pause params pause TX = %s, pause RX = %s\n",
+			   pause_tx ? "ON" : "OFF", pause_rx ? "ON" : "OFF");
+	}
+
+	gmac_set_flow_control(netdev, pause_tx, pause_rx);
+
+	if (old_status.bits32 == status.bits32)
+		return;
+
+	if (netif_msg_link(port)) {
+		phy_print_status(phydev);
+		netdev_info(netdev, "link flow control: %s\n",
+			    phydev->pause
+			    ? (phydev->asym_pause ? "tx" : "both")
+			    : (phydev->asym_pause ? "rx" : "none")
+		);
+	}
+
+	gmac_disable_tx_rx(netdev);
+	writel(status.bits32, port->gmac_base + GMAC_STATUS);
+	gmac_enable_tx_rx(netdev);
+}
+
+static int gmac_setup_phy(struct net_device *netdev)
+{
+	struct gemini_ethernet_port *port = netdev_priv(netdev);
+	union gmac_status status = { .bits32 = 0 };
+	struct device *dev = port->dev;
+	struct phy_device *phy;
+
+	phy = of_phy_get_and_connect(netdev,
+				     dev->of_node,
+				     gmac_adjust_link);
+	if (!phy)
+		return -ENODEV;
+	netdev->phydev = phy;
+
+	phy_set_max_speed(phy, SPEED_1000);
+	phy_support_asym_pause(phy);
+
+	/* set PHY interface type */
+	switch (phy->interface) {
+	case PHY_INTERFACE_MODE_MII:
+		netdev_dbg(netdev,
+			   "MII: set GMAC0 to GMII mode, GMAC1 disabled\n");
+		status.bits.mii_rmii = GMAC_PHY_MII;
+		break;
+	case PHY_INTERFACE_MODE_GMII:
+		netdev_dbg(netdev,
+			   "GMII: set GMAC0 to GMII mode, GMAC1 disabled\n");
+		status.bits.mii_rmii = GMAC_PHY_GMII;
+		break;
+	case PHY_INTERFACE_MODE_RGMII:
+	case PHY_INTERFACE_MODE_RGMII_ID:
+	case PHY_INTERFACE_MODE_RGMII_TXID:
+	case PHY_INTERFACE_MODE_RGMII_RXID:
+		netdev_dbg(netdev,
+			   "RGMII: set GMAC0 and GMAC1 to MII/RGMII mode\n");
+		status.bits.mii_rmii = GMAC_PHY_RGMII_100_10;
+		break;
+	default:
+		netdev_err(netdev, "Unsupported MII interface\n");
+		phy_disconnect(phy);
+		netdev->phydev = NULL;
+		return -EINVAL;
+	}
+	writel(status.bits32, port->gmac_base + GMAC_STATUS);
+
+	if (netif_msg_link(port))
+		phy_attached_info(phy);
+
+	return 0;
+}
+
+/* The maximum frame length is not logically enumerated in the
+ * hardware, so we do a table lookup to find the applicable max
+ * frame length.
+ */
+struct gmac_max_framelen {
+	unsigned int max_l3_len;
+	u8 val;
+};
+
+static const struct gmac_max_framelen gmac_maxlens[] = {
+	{
+		.max_l3_len = 1518,
+		.val = CONFIG0_MAXLEN_1518,
+	},
+	{
+		.max_l3_len = 1522,
+		.val = CONFIG0_MAXLEN_1522,
+	},
+	{
+		.max_l3_len = 1536,
+		.val = CONFIG0_MAXLEN_1536,
+	},
+	{
+		.max_l3_len = 1548,
+		.val = CONFIG0_MAXLEN_1548,
+	},
+	{
+		.max_l3_len = 9212,
+		.val = CONFIG0_MAXLEN_9k,
+	},
+	{
+		.max_l3_len = 10236,
+		.val = CONFIG0_MAXLEN_10k,
+	},
+};
+
+static int gmac_pick_rx_max_len(unsigned int max_l3_len)
+{
+	const struct gmac_max_framelen *maxlen;
+	int maxtot;
+	int i;
+
+	maxtot = max_l3_len + ETH_HLEN + VLAN_HLEN;
+
+	for (i = 0; i < ARRAY_SIZE(gmac_maxlens); i++) {
+		maxlen = &gmac_maxlens[i];
+		if (maxtot <= maxlen->max_l3_len)
+			return maxlen->val;
+	}
+
+	return -1;
+}
+
+static int gmac_init(struct net_device *netdev)
+{
+	struct gemini_ethernet_port *port = netdev_priv(netdev);
+	union gmac_config0 config0 = { .bits = {
+		.dis_tx = 1,
+		.dis_rx = 1,
+		.ipv4_rx_chksum = 1,
+		.ipv6_rx_chksum = 1,
+		.rx_err_detect = 1,
+		.rgmm_edge = 1,
+		.port0_chk_hwq = 1,
+		.port1_chk_hwq = 1,
+		.port0_chk_toeq = 1,
+		.port1_chk_toeq = 1,
+		.port0_chk_classq = 1,
+		.port1_chk_classq = 1,
+	} };
+	union gmac_ahb_weight ahb_weight = { .bits = {
+		.rx_weight = 1,
+		.tx_weight = 1,
+		.hash_weight = 1,
+		.pre_req = 0x1f,
+		.tq_dv_threshold = 0,
+	} };
+	union gmac_tx_wcr0 hw_weigh = { .bits = {
+		.hw_tq3 = 1,
+		.hw_tq2 = 1,
+		.hw_tq1 = 1,
+		.hw_tq0 = 1,
+	} };
+	union gmac_tx_wcr1 sw_weigh = { .bits = {
+		.sw_tq5 = 1,
+		.sw_tq4 = 1,
+		.sw_tq3 = 1,
+		.sw_tq2 = 1,
+		.sw_tq1 = 1,
+		.sw_tq0 = 1,
+	} };
+	union gmac_config1 config1 = { .bits = {
+		.set_threshold = 16,
+		.rel_threshold = 24,
+	} };
+	union gmac_config2 config2 = { .bits = {
+		.set_threshold = 16,
+		.rel_threshold = 32,
+	} };
+	union gmac_config3 config3 = { .bits = {
+		.set_threshold = 0,
+		.rel_threshold = 0,
+	} };
+	union gmac_config0 tmp;
+
+	config0.bits.max_len = gmac_pick_rx_max_len(netdev->mtu);
+	tmp.bits32 = readl(port->gmac_base + GMAC_CONFIG0);
+	config0.bits.reserved = tmp.bits.reserved;
+	writel(config0.bits32, port->gmac_base + GMAC_CONFIG0);
+	writel(config1.bits32, port->gmac_base + GMAC_CONFIG1);
+	writel(config2.bits32, port->gmac_base + GMAC_CONFIG2);
+	writel(config3.bits32, port->gmac_base + GMAC_CONFIG3);
+
+	readl(port->dma_base + GMAC_AHB_WEIGHT_REG);
+	writel(ahb_weight.bits32, port->dma_base + GMAC_AHB_WEIGHT_REG);
+
+	writel(hw_weigh.bits32,
+	       port->dma_base + GMAC_TX_WEIGHTING_CTRL_0_REG);
+	writel(sw_weigh.bits32,
+	       port->dma_base + GMAC_TX_WEIGHTING_CTRL_1_REG);
+
+	port->rxq_order = DEFAULT_GMAC_RXQ_ORDER;
+	port->txq_order = DEFAULT_GMAC_TXQ_ORDER;
+	port->rx_coalesce_nsecs = DEFAULT_RX_COALESCE_NSECS;
+
+	/* Mark every quarter of the queue a packet for interrupt
+	 * in order to be able to wake up the queue if it was stopped
+	 */
+	port->irq_every_tx_packets = 1 << (port->txq_order - 2);
+
+	return 0;
+}
+
+static int gmac_setup_txqs(struct net_device *netdev)
+{
+	struct gemini_ethernet_port *port = netdev_priv(netdev);
+	unsigned int n_txq = netdev->num_tx_queues;
+	struct gemini_ethernet *geth = port->geth;
+	size_t entries = 1 << port->txq_order;
+	struct gmac_txq *txq = port->txq;
+	struct gmac_txdesc *desc_ring;
+	size_t len = n_txq * entries;
+	struct sk_buff **skb_tab;
+	void __iomem *rwptr_reg;
+	unsigned int r;
+	int i;
+
+	rwptr_reg = port->dma_base + GMAC_SW_TX_QUEUE0_PTR_REG;
+
+	skb_tab = kcalloc(len, sizeof(*skb_tab), GFP_KERNEL);
+	if (!skb_tab)
+		return -ENOMEM;
+
+	desc_ring = dma_alloc_coherent(geth->dev, len * sizeof(*desc_ring),
+				       &port->txq_dma_base, GFP_KERNEL);
+
+	if (!desc_ring) {
+		kfree(skb_tab);
+		return -ENOMEM;
+	}
+
+	if (port->txq_dma_base & ~DMA_Q_BASE_MASK) {
+		dev_warn(geth->dev, "TX queue base is not aligned\n");
+		dma_free_coherent(geth->dev, len * sizeof(*desc_ring),
+				  desc_ring, port->txq_dma_base);
+		kfree(skb_tab);
+		return -ENOMEM;
+	}
+
+	writel(port->txq_dma_base | port->txq_order,
+	       port->dma_base + GMAC_SW_TX_QUEUE_BASE_REG);
+
+	for (i = 0; i < n_txq; i++) {
+		txq->ring = desc_ring;
+		txq->skb = skb_tab;
+		txq->noirq_packets = 0;
+
+		r = readw(rwptr_reg);
+		rwptr_reg += 2;
+		writew(r, rwptr_reg);
+		rwptr_reg += 2;
+		txq->cptr = r;
+
+		txq++;
+		desc_ring += entries;
+		skb_tab += entries;
+	}
+
+	return 0;
+}
+
+static void gmac_clean_txq(struct net_device *netdev, struct gmac_txq *txq,
+			   unsigned int r)
+{
+	struct gemini_ethernet_port *port = netdev_priv(netdev);
+	unsigned int m = (1 << port->txq_order) - 1;
+	struct gemini_ethernet *geth = port->geth;
+	unsigned int c = txq->cptr;
+	union gmac_txdesc_0 word0;
+	union gmac_txdesc_1 word1;
+	unsigned int hwchksum = 0;
+	unsigned long bytes = 0;
+	struct gmac_txdesc *txd;
+	unsigned short nfrags;
+	unsigned int errs = 0;
+	unsigned int pkts = 0;
+	unsigned int word3;
+	dma_addr_t mapping;
+
+	if (c == r)
+		return;
+
+	while (c != r) {
+		txd = txq->ring + c;
+		word0 = txd->word0;
+		word1 = txd->word1;
+		mapping = txd->word2.buf_adr;
+		word3 = txd->word3.bits32;
+
+		dma_unmap_single(geth->dev, mapping,
+				 word0.bits.buffer_size, DMA_TO_DEVICE);
+
+		if (word3 & EOF_BIT)
+			dev_kfree_skb(txq->skb[c]);
+
+		c++;
+		c &= m;
+
+		if (!(word3 & SOF_BIT))
+			continue;
+
+		if (!word0.bits.status_tx_ok) {
+			errs++;
+			continue;
+		}
+
+		pkts++;
+		bytes += txd->word1.bits.byte_count;
+
+		if (word1.bits32 & TSS_CHECKUM_ENABLE)
+			hwchksum++;
+
+		nfrags = word0.bits.desc_count - 1;
+		if (nfrags) {
+			if (nfrags >= TX_MAX_FRAGS)
+				nfrags = TX_MAX_FRAGS - 1;
+
+			u64_stats_update_begin(&port->tx_stats_syncp);
+			port->tx_frag_stats[nfrags]++;
+			u64_stats_update_end(&port->tx_stats_syncp);
+		}
+	}
+
+	u64_stats_update_begin(&port->ir_stats_syncp);
+	port->stats.tx_errors += errs;
+	port->stats.tx_packets += pkts;
+	port->stats.tx_bytes += bytes;
+	port->tx_hw_csummed += hwchksum;
+	u64_stats_update_end(&port->ir_stats_syncp);
+
+	txq->cptr = c;
+}
+
+static void gmac_cleanup_txqs(struct net_device *netdev)
+{
+	struct gemini_ethernet_port *port = netdev_priv(netdev);
+	unsigned int n_txq = netdev->num_tx_queues;
+	struct gemini_ethernet *geth = port->geth;
+	void __iomem *rwptr_reg;
+	unsigned int r, i;
+
+	rwptr_reg = port->dma_base + GMAC_SW_TX_QUEUE0_PTR_REG;
+
+	for (i = 0; i < n_txq; i++) {
+		r = readw(rwptr_reg);
+		rwptr_reg += 2;
+		writew(r, rwptr_reg);
+		rwptr_reg += 2;
+
+		gmac_clean_txq(netdev, port->txq + i, r);
+	}
+	writel(0, port->dma_base + GMAC_SW_TX_QUEUE_BASE_REG);
+
+	kfree(port->txq->skb);
+	dma_free_coherent(geth->dev,
+			  n_txq * sizeof(*port->txq->ring) << port->txq_order,
+			  port->txq->ring, port->txq_dma_base);
+}
+
+static int gmac_setup_rxq(struct net_device *netdev)
+{
+	struct gemini_ethernet_port *port = netdev_priv(netdev);
+	struct gemini_ethernet *geth = port->geth;
+	struct nontoe_qhdr __iomem *qhdr;
+
+	qhdr = geth->base + TOE_DEFAULT_Q_HDR_BASE(netdev->dev_id);
+	port->rxq_rwptr = &qhdr->word1;
+
+	/* Remap a slew of memory to use for the RX queue */
+	port->rxq_ring = dma_alloc_coherent(geth->dev,
+				sizeof(*port->rxq_ring) << port->rxq_order,
+				&port->rxq_dma_base, GFP_KERNEL);
+	if (!port->rxq_ring)
+		return -ENOMEM;
+	if (port->rxq_dma_base & ~NONTOE_QHDR0_BASE_MASK) {
+		dev_warn(geth->dev, "RX queue base is not aligned\n");
+		return -ENOMEM;
+	}
+
+	writel(port->rxq_dma_base | port->rxq_order, &qhdr->word0);
+	writel(0, port->rxq_rwptr);
+	return 0;
+}
+
+static struct gmac_queue_page *
+gmac_get_queue_page(struct gemini_ethernet *geth,
+		    struct gemini_ethernet_port *port,
+		    dma_addr_t addr)
+{
+	struct gmac_queue_page *gpage;
+	dma_addr_t mapping;
+	int i;
+
+	/* Only look for even pages */
+	mapping = addr & PAGE_MASK;
+
+	if (!geth->freeq_pages) {
+		dev_err(geth->dev, "try to get page with no page list\n");
+		return NULL;
+	}
+
+	/* Look up a ring buffer page from virtual mapping */
+	for (i = 0; i < geth->num_freeq_pages; i++) {
+		gpage = &geth->freeq_pages[i];
+		if (gpage->mapping == mapping)
+			return gpage;
+	}
+
+	return NULL;
+}
+
+static void gmac_cleanup_rxq(struct net_device *netdev)
+{
+	struct gemini_ethernet_port *port = netdev_priv(netdev);
+	struct gemini_ethernet *geth = port->geth;
+	struct gmac_rxdesc *rxd = port->rxq_ring;
+	static struct gmac_queue_page *gpage;
+	struct nontoe_qhdr __iomem *qhdr;
+	void __iomem *dma_reg;
+	void __iomem *ptr_reg;
+	dma_addr_t mapping;
+	union dma_rwptr rw;
+	unsigned int r, w;
+
+	qhdr = geth->base +
+		TOE_DEFAULT_Q_HDR_BASE(netdev->dev_id);
+	dma_reg = &qhdr->word0;
+	ptr_reg = &qhdr->word1;
+
+	rw.bits32 = readl(ptr_reg);
+	r = rw.bits.rptr;
+	w = rw.bits.wptr;
+	writew(r, ptr_reg + 2);
+
+	writel(0, dma_reg);
+
+	/* Loop from read pointer to write pointer of the RX queue
+	 * and free up all pages by the queue.
+	 */
+	while (r != w) {
+		mapping = rxd[r].word2.buf_adr;
+		r++;
+		r &= ((1 << port->rxq_order) - 1);
+
+		if (!mapping)
+			continue;
+
+		/* Freeq pointers are one page off */
+		gpage = gmac_get_queue_page(geth, port, mapping + PAGE_SIZE);
+		if (!gpage) {
+			dev_err(geth->dev, "could not find page\n");
+			continue;
+		}
+		/* Release the RX queue reference to the page */
+		put_page(gpage->page);
+	}
+
+	dma_free_coherent(geth->dev, sizeof(*port->rxq_ring) << port->rxq_order,
+			  port->rxq_ring, port->rxq_dma_base);
+}
+
+static struct page *geth_freeq_alloc_map_page(struct gemini_ethernet *geth,
+					      int pn)
+{
+	struct gmac_rxdesc *freeq_entry;
+	struct gmac_queue_page *gpage;
+	unsigned int fpp_order;
+	unsigned int frag_len;
+	dma_addr_t mapping;
+	struct page *page;
+	int i;
+
+	/* First allocate and DMA map a single page */
+	page = alloc_page(GFP_ATOMIC);
+	if (!page)
+		return NULL;
+
+	mapping = dma_map_single(geth->dev, page_address(page),
+				 PAGE_SIZE, DMA_FROM_DEVICE);
+	if (dma_mapping_error(geth->dev, mapping)) {
+		put_page(page);
+		return NULL;
+	}
+
+	/* The assign the page mapping (physical address) to the buffer address
+	 * in the hardware queue. PAGE_SHIFT on ARM is 12 (1 page is 4096 bytes,
+	 * 4k), and the default RX frag order is 11 (fragments are up 20 2048
+	 * bytes, 2k) so fpp_order (fragments per page order) is default 1. Thus
+	 * each page normally needs two entries in the queue.
+	 */
+	frag_len = 1 << geth->freeq_frag_order; /* Usually 2048 */
+	fpp_order = PAGE_SHIFT - geth->freeq_frag_order;
+	freeq_entry = geth->freeq_ring + (pn << fpp_order);
+	dev_dbg(geth->dev, "allocate page %d fragment length %d fragments per page %d, freeq entry %p\n",
+		 pn, frag_len, (1 << fpp_order), freeq_entry);
+	for (i = (1 << fpp_order); i > 0; i--) {
+		freeq_entry->word2.buf_adr = mapping;
+		freeq_entry++;
+		mapping += frag_len;
+	}
+
+	/* If the freeq entry already has a page mapped, then unmap it. */
+	gpage = &geth->freeq_pages[pn];
+	if (gpage->page) {
+		mapping = geth->freeq_ring[pn << fpp_order].word2.buf_adr;
+		dma_unmap_single(geth->dev, mapping, frag_len, DMA_FROM_DEVICE);
+		/* This should be the last reference to the page so it gets
+		 * released
+		 */
+		put_page(gpage->page);
+	}
+
+	/* Then put our new mapping into the page table */
+	dev_dbg(geth->dev, "page %d, DMA addr: %08x, page %p\n",
+		pn, (unsigned int)mapping, page);
+	gpage->mapping = mapping;
+	gpage->page = page;
+
+	return page;
+}
+
+/**
+ * geth_fill_freeq() - Fill the freeq with empty fragments to use
+ * @geth: the ethernet adapter
+ * @refill: whether to reset the queue by filling in all freeq entries or
+ * just refill it, usually the interrupt to refill the queue happens when
+ * the queue is half empty.
+ */
+static unsigned int geth_fill_freeq(struct gemini_ethernet *geth, bool refill)
+{
+	unsigned int fpp_order = PAGE_SHIFT - geth->freeq_frag_order;
+	unsigned int count = 0;
+	unsigned int pn, epn;
+	unsigned long flags;
+	union dma_rwptr rw;
+	unsigned int m_pn;
+
+	/* Mask for page */
+	m_pn = (1 << (geth->freeq_order - fpp_order)) - 1;
+
+	spin_lock_irqsave(&geth->freeq_lock, flags);
+
+	rw.bits32 = readl(geth->base + GLOBAL_SWFQ_RWPTR_REG);
+	pn = (refill ? rw.bits.wptr : rw.bits.rptr) >> fpp_order;
+	epn = (rw.bits.rptr >> fpp_order) - 1;
+	epn &= m_pn;
+
+	/* Loop over the freeq ring buffer entries */
+	while (pn != epn) {
+		struct gmac_queue_page *gpage;
+		struct page *page;
+
+		gpage = &geth->freeq_pages[pn];
+		page = gpage->page;
+
+		dev_dbg(geth->dev, "fill entry %d page ref count %d add %d refs\n",
+			pn, page_ref_count(page), 1 << fpp_order);
+
+		if (page_ref_count(page) > 1) {
+			unsigned int fl = (pn - epn) & m_pn;
+
+			if (fl > 64 >> fpp_order)
+				break;
+
+			page = geth_freeq_alloc_map_page(geth, pn);
+			if (!page)
+				break;
+		}
+
+		/* Add one reference per fragment in the page */
+		page_ref_add(page, 1 << fpp_order);
+		count += 1 << fpp_order;
+		pn++;
+		pn &= m_pn;
+	}
+
+	writew(pn << fpp_order, geth->base + GLOBAL_SWFQ_RWPTR_REG + 2);
+
+	spin_unlock_irqrestore(&geth->freeq_lock, flags);
+
+	return count;
+}
+
+static int geth_setup_freeq(struct gemini_ethernet *geth)
+{
+	unsigned int fpp_order = PAGE_SHIFT - geth->freeq_frag_order;
+	unsigned int frag_len = 1 << geth->freeq_frag_order;
+	unsigned int len = 1 << geth->freeq_order;
+	unsigned int pages = len >> fpp_order;
+	union queue_threshold qt;
+	union dma_skb_size skbsz;
+	unsigned int filled;
+	unsigned int pn;
+
+	geth->freeq_ring = dma_alloc_coherent(geth->dev,
+		sizeof(*geth->freeq_ring) << geth->freeq_order,
+		&geth->freeq_dma_base, GFP_KERNEL);
+	if (!geth->freeq_ring)
+		return -ENOMEM;
+	if (geth->freeq_dma_base & ~DMA_Q_BASE_MASK) {
+		dev_warn(geth->dev, "queue ring base is not aligned\n");
+		goto err_freeq;
+	}
+
+	/* Allocate a mapping to page look-up index */
+	geth->freeq_pages = kcalloc(pages, sizeof(*geth->freeq_pages),
+				    GFP_KERNEL);
+	if (!geth->freeq_pages)
+		goto err_freeq;
+	geth->num_freeq_pages = pages;
+
+	dev_info(geth->dev, "allocate %d pages for queue\n", pages);
+	for (pn = 0; pn < pages; pn++)
+		if (!geth_freeq_alloc_map_page(geth, pn))
+			goto err_freeq_alloc;
+
+	filled = geth_fill_freeq(geth, false);
+	if (!filled)
+		goto err_freeq_alloc;
+
+	qt.bits32 = readl(geth->base + GLOBAL_QUEUE_THRESHOLD_REG);
+	qt.bits.swfq_empty = 32;
+	writel(qt.bits32, geth->base + GLOBAL_QUEUE_THRESHOLD_REG);
+
+	skbsz.bits.sw_skb_size = 1 << geth->freeq_frag_order;
+	writel(skbsz.bits32, geth->base + GLOBAL_DMA_SKB_SIZE_REG);
+	writel(geth->freeq_dma_base | geth->freeq_order,
+	       geth->base + GLOBAL_SW_FREEQ_BASE_SIZE_REG);
+
+	return 0;
+
+err_freeq_alloc:
+	while (pn > 0) {
+		struct gmac_queue_page *gpage;
+		dma_addr_t mapping;
+
+		--pn;
+		mapping = geth->freeq_ring[pn << fpp_order].word2.buf_adr;
+		dma_unmap_single(geth->dev, mapping, frag_len, DMA_FROM_DEVICE);
+		gpage = &geth->freeq_pages[pn];
+		put_page(gpage->page);
+	}
+
+	kfree(geth->freeq_pages);
+err_freeq:
+	dma_free_coherent(geth->dev,
+			  sizeof(*geth->freeq_ring) << geth->freeq_order,
+			  geth->freeq_ring, geth->freeq_dma_base);
+	geth->freeq_ring = NULL;
+	return -ENOMEM;
+}
+
+/**
+ * geth_cleanup_freeq() - cleanup the DMA mappings and free the queue
+ * @geth: the Gemini global ethernet state
+ */
+static void geth_cleanup_freeq(struct gemini_ethernet *geth)
+{
+	unsigned int fpp_order = PAGE_SHIFT - geth->freeq_frag_order;
+	unsigned int frag_len = 1 << geth->freeq_frag_order;
+	unsigned int len = 1 << geth->freeq_order;
+	unsigned int pages = len >> fpp_order;
+	unsigned int pn;
+
+	writew(readw(geth->base + GLOBAL_SWFQ_RWPTR_REG),
+	       geth->base + GLOBAL_SWFQ_RWPTR_REG + 2);
+	writel(0, geth->base + GLOBAL_SW_FREEQ_BASE_SIZE_REG);
+
+	for (pn = 0; pn < pages; pn++) {
+		struct gmac_queue_page *gpage;
+		dma_addr_t mapping;
+
+		mapping = geth->freeq_ring[pn << fpp_order].word2.buf_adr;
+		dma_unmap_single(geth->dev, mapping, frag_len, DMA_FROM_DEVICE);
+
+		gpage = &geth->freeq_pages[pn];
+		while (page_ref_count(gpage->page) > 0)
+			put_page(gpage->page);
+	}
+
+	kfree(geth->freeq_pages);
+
+	dma_free_coherent(geth->dev,
+			  sizeof(*geth->freeq_ring) << geth->freeq_order,
+			  geth->freeq_ring, geth->freeq_dma_base);
+}
+
+/**
+ * geth_resize_freeq() - resize the software queue depth
+ * @port: the port requesting the change
+ *
+ * This gets called at least once during probe() so the device queue gets
+ * "resized" from the hardware defaults. Since both ports/net devices share
+ * the same hardware queue, some synchronization between the ports is
+ * needed.
+ */
+static int geth_resize_freeq(struct gemini_ethernet_port *port)
+{
+	struct gemini_ethernet *geth = port->geth;
+	struct net_device *netdev = port->netdev;
+	struct gemini_ethernet_port *other_port;
+	struct net_device *other_netdev;
+	unsigned int new_size = 0;
+	unsigned int new_order;
+	unsigned long flags;
+	u32 en;
+	int ret;
+
+	if (netdev->dev_id == 0)
+		other_netdev = geth->port1->netdev;
+	else
+		other_netdev = geth->port0->netdev;
+
+	if (other_netdev && netif_running(other_netdev))
+		return -EBUSY;
+
+	new_size = 1 << (port->rxq_order + 1);
+	netdev_dbg(netdev, "port %d size: %d order %d\n",
+		   netdev->dev_id,
+		   new_size,
+		   port->rxq_order);
+	if (other_netdev) {
+		other_port = netdev_priv(other_netdev);
+		new_size += 1 << (other_port->rxq_order + 1);
+		netdev_dbg(other_netdev, "port %d size: %d order %d\n",
+			   other_netdev->dev_id,
+			   (1 << (other_port->rxq_order + 1)),
+			   other_port->rxq_order);
+	}
+
+	new_order = min(15, ilog2(new_size - 1) + 1);
+	dev_dbg(geth->dev, "set shared queue to size %d order %d\n",
+		new_size, new_order);
+	if (geth->freeq_order == new_order)
+		return 0;
+
+	spin_lock_irqsave(&geth->irq_lock, flags);
+
+	/* Disable the software queue IRQs */
+	en = readl(geth->base + GLOBAL_INTERRUPT_ENABLE_4_REG);
+	en &= ~SWFQ_EMPTY_INT_BIT;
+	writel(en, geth->base + GLOBAL_INTERRUPT_ENABLE_4_REG);
+	spin_unlock_irqrestore(&geth->irq_lock, flags);
+
+	/* Drop the old queue */
+	if (geth->freeq_ring)
+		geth_cleanup_freeq(geth);
+
+	/* Allocate a new queue with the desired order */
+	geth->freeq_order = new_order;
+	ret = geth_setup_freeq(geth);
+
+	/* Restart the interrupts - NOTE if this is the first resize
+	 * after probe(), this is where the interrupts get turned on
+	 * in the first place.
+	 */
+	spin_lock_irqsave(&geth->irq_lock, flags);
+	en |= SWFQ_EMPTY_INT_BIT;
+	writel(en, geth->base + GLOBAL_INTERRUPT_ENABLE_4_REG);
+	spin_unlock_irqrestore(&geth->irq_lock, flags);
+
+	return ret;
+}
+
+static void gmac_tx_irq_enable(struct net_device *netdev,
+			       unsigned int txq, int en)
+{
+	struct gemini_ethernet_port *port = netdev_priv(netdev);
+	struct gemini_ethernet *geth = port->geth;
+	unsigned long flags;
+	u32 val, mask;
+
+	netdev_dbg(netdev, "%s device %d\n", __func__, netdev->dev_id);
+
+	spin_lock_irqsave(&geth->irq_lock, flags);
+
+	mask = GMAC0_IRQ0_TXQ0_INTS << (6 * netdev->dev_id + txq);
+
+	if (en)
+		writel(mask, geth->base + GLOBAL_INTERRUPT_STATUS_0_REG);
+
+	val = readl(geth->base + GLOBAL_INTERRUPT_ENABLE_0_REG);
+	val = en ? val | mask : val & ~mask;
+	writel(val, geth->base + GLOBAL_INTERRUPT_ENABLE_0_REG);
+
+	spin_unlock_irqrestore(&geth->irq_lock, flags);
+}
+
+static void gmac_tx_irq(struct net_device *netdev, unsigned int txq_num)
+{
+	struct netdev_queue *ntxq = netdev_get_tx_queue(netdev, txq_num);
+
+	gmac_tx_irq_enable(netdev, txq_num, 0);
+	netif_tx_wake_queue(ntxq);
+}
+
+static int gmac_map_tx_bufs(struct net_device *netdev, struct sk_buff *skb,
+			    struct gmac_txq *txq, unsigned short *desc)
+{
+	struct gemini_ethernet_port *port = netdev_priv(netdev);
+	struct skb_shared_info *skb_si =  skb_shinfo(skb);
+	unsigned short m = (1 << port->txq_order) - 1;
+	short frag, last_frag = skb_si->nr_frags - 1;
+	struct gemini_ethernet *geth = port->geth;
+	unsigned int word1, word3, buflen;
+	unsigned short w = *desc;
+	struct gmac_txdesc *txd;
+	skb_frag_t *skb_frag;
+	dma_addr_t mapping;
+	bool tcp = false;
+	void *buffer;
+	u16 mss;
+	int ret;
+
+	word1 = skb->len;
+	word3 = SOF_BIT;
+
+	/* Determine if we are doing TCP */
+	if (skb->protocol == htons(ETH_P_IP))
+		tcp = (ip_hdr(skb)->protocol == IPPROTO_TCP);
+	else
+		/* IPv6 */
+		tcp = (ipv6_hdr(skb)->nexthdr == IPPROTO_TCP);
+
+	mss = skb_shinfo(skb)->gso_size;
+	if (mss) {
+		/* This means we are dealing with TCP and skb->len is the
+		 * sum total of all the segments. The TSO will deal with
+		 * chopping this up for us.
+		 */
+		/* The accelerator needs the full frame size here */
+		mss += skb_tcp_all_headers(skb);
+		netdev_dbg(netdev, "segment offloading mss = %04x len=%04x\n",
+			   mss, skb->len);
+		word1 |= TSS_MTU_ENABLE_BIT;
+		word3 |= mss;
+	} else if (tcp) {
+		/* Even if we are not using TSO, use the hardware offloader
+		 * for transferring the TCP frame: this hardware has partial
+		 * TCP awareness (called TOE - TCP Offload Engine) and will
+		 * according to the datasheet put packets belonging to the
+		 * same TCP connection in the same queue for the TOE/TSO
+		 * engine to process. The engine will deal with chopping
+		 * up frames that exceed ETH_DATA_LEN which the
+		 * checksumming engine cannot handle (see below) into
+		 * manageable chunks. It flawlessly deals with quite big
+		 * frames and frames containing custom DSA EtherTypes.
+		 */
+		mss = netdev->mtu + skb_tcp_all_headers(skb);
+		mss = min(mss, skb->len);
+		netdev_dbg(netdev, "TOE/TSO len %04x mtu %04x mss %04x\n",
+			   skb->len, netdev->mtu, mss);
+		word1 |= TSS_MTU_ENABLE_BIT;
+		word3 |= mss;
+	} else if (skb->len >= ETH_FRAME_LEN) {
+		/* Hardware offloaded checksumming isn't working on non-TCP frames
+		 * bigger than 1514 bytes. A hypothesis about this is that the
+		 * checksum buffer is only 1518 bytes, so when the frames get
+		 * bigger they get truncated, or the last few bytes get
+		 * overwritten by the FCS.
+		 *
+		 * Just use software checksumming and bypass on bigger frames.
+		 */
+		if (skb->ip_summed == CHECKSUM_PARTIAL) {
+			ret = skb_checksum_help(skb);
+			if (ret)
+				return ret;
+		}
+		word1 |= TSS_BYPASS_BIT;
+	}
+
+	if (skb->ip_summed == CHECKSUM_PARTIAL) {
+		/* We do not switch off the checksumming on non TCP/UDP
+		 * frames: as is shown from tests, the checksumming engine
+		 * is smart enough to see that a frame is not actually TCP
+		 * or UDP and then just pass it through without any changes
+		 * to the frame.
+		 */
+		if (skb->protocol == htons(ETH_P_IP))
+			word1 |= TSS_IP_CHKSUM_BIT;
+		else
+			word1 |= TSS_IPV6_ENABLE_BIT;
+
+		word1 |= tcp ? TSS_TCP_CHKSUM_BIT : TSS_UDP_CHKSUM_BIT;
+	}
+
+	frag = -1;
+	while (frag <= last_frag) {
+		if (frag == -1) {
+			buffer = skb->data;
+			buflen = skb_headlen(skb);
+		} else {
+			skb_frag = skb_si->frags + frag;
+			buffer = skb_frag_address(skb_frag);
+			buflen = skb_frag_size(skb_frag);
+		}
+
+		if (frag == last_frag) {
+			word3 |= EOF_BIT;
+			txq->skb[w] = skb;
+		}
+
+		mapping = dma_map_single(geth->dev, buffer, buflen,
+					 DMA_TO_DEVICE);
+		if (dma_mapping_error(geth->dev, mapping))
+			goto map_error;
+
+		txd = txq->ring + w;
+		txd->word0.bits32 = buflen;
+		txd->word1.bits32 = word1;
+		txd->word2.buf_adr = mapping;
+		txd->word3.bits32 = word3;
+
+		word3 &= MTU_SIZE_BIT_MASK;
+		w++;
+		w &= m;
+		frag++;
+	}
+
+	*desc = w;
+	return 0;
+
+map_error:
+	while (w != *desc) {
+		w--;
+		w &= m;
+
+		dma_unmap_page(geth->dev, txq->ring[w].word2.buf_adr,
+			       txq->ring[w].word0.bits.buffer_size,
+			       DMA_TO_DEVICE);
+	}
+	return -ENOMEM;
+}
+
+static netdev_tx_t gmac_start_xmit(struct sk_buff *skb,
+				   struct net_device *netdev)
+{
+	struct gemini_ethernet_port *port = netdev_priv(netdev);
+	unsigned short m = (1 << port->txq_order) - 1;
+	struct netdev_queue *ntxq;
+	unsigned short r, w, d;
+	void __iomem *ptr_reg;
+	struct gmac_txq *txq;
+	int txq_num, nfrags;
+	union dma_rwptr rw;
+
+	if (skb->len >= 0x10000)
+		goto out_drop_free;
+
+	txq_num = skb_get_queue_mapping(skb);
+	ptr_reg = port->dma_base + GMAC_SW_TX_QUEUE_PTR_REG(txq_num);
+	txq = &port->txq[txq_num];
+	ntxq = netdev_get_tx_queue(netdev, txq_num);
+	nfrags = skb_shinfo(skb)->nr_frags;
+
+	rw.bits32 = readl(ptr_reg);
+	r = rw.bits.rptr;
+	w = rw.bits.wptr;
+
+	d = txq->cptr - w - 1;
+	d &= m;
+
+	if (d < nfrags + 2) {
+		gmac_clean_txq(netdev, txq, r);
+		d = txq->cptr - w - 1;
+		d &= m;
+
+		if (d < nfrags + 2) {
+			netif_tx_stop_queue(ntxq);
+
+			d = txq->cptr + nfrags + 16;
+			d &= m;
+			txq->ring[d].word3.bits.eofie = 1;
+			gmac_tx_irq_enable(netdev, txq_num, 1);
+
+			u64_stats_update_begin(&port->tx_stats_syncp);
+			netdev->stats.tx_fifo_errors++;
+			u64_stats_update_end(&port->tx_stats_syncp);
+			return NETDEV_TX_BUSY;
+		}
+	}
+
+	if (gmac_map_tx_bufs(netdev, skb, txq, &w)) {
+		if (skb_linearize(skb))
+			goto out_drop;
+
+		u64_stats_update_begin(&port->tx_stats_syncp);
+		port->tx_frags_linearized++;
+		u64_stats_update_end(&port->tx_stats_syncp);
+
+		if (gmac_map_tx_bufs(netdev, skb, txq, &w))
+			goto out_drop_free;
+	}
+
+	writew(w, ptr_reg + 2);
+
+	gmac_clean_txq(netdev, txq, r);
+	return NETDEV_TX_OK;
+
+out_drop_free:
+	dev_kfree_skb(skb);
+out_drop:
+	u64_stats_update_begin(&port->tx_stats_syncp);
+	port->stats.tx_dropped++;
+	u64_stats_update_end(&port->tx_stats_syncp);
+	return NETDEV_TX_OK;
+}
+
+static void gmac_tx_timeout(struct net_device *netdev, unsigned int txqueue)
+{
+	netdev_err(netdev, "Tx timeout\n");
+	gmac_dump_dma_state(netdev);
+}
+
+static void gmac_enable_irq(struct net_device *netdev, int enable)
+{
+	struct gemini_ethernet_port *port = netdev_priv(netdev);
+	struct gemini_ethernet *geth = port->geth;
+	unsigned long flags;
+	u32 val, mask;
+
+	netdev_dbg(netdev, "%s device %d %s\n", __func__,
+		   netdev->dev_id, enable ? "enable" : "disable");
+	spin_lock_irqsave(&geth->irq_lock, flags);
+
+	mask = GMAC0_IRQ0_2 << (netdev->dev_id * 2);
+	val = readl(geth->base + GLOBAL_INTERRUPT_ENABLE_0_REG);
+	val = enable ? (val | mask) : (val & ~mask);
+	writel(val, geth->base + GLOBAL_INTERRUPT_ENABLE_0_REG);
+
+	mask = DEFAULT_Q0_INT_BIT << netdev->dev_id;
+	val = readl(geth->base + GLOBAL_INTERRUPT_ENABLE_1_REG);
+	val = enable ? (val | mask) : (val & ~mask);
+	writel(val, geth->base + GLOBAL_INTERRUPT_ENABLE_1_REG);
+
+	mask = GMAC0_IRQ4_8 << (netdev->dev_id * 8);
+	val = readl(geth->base + GLOBAL_INTERRUPT_ENABLE_4_REG);
+	val = enable ? (val | mask) : (val & ~mask);
+	writel(val, geth->base + GLOBAL_INTERRUPT_ENABLE_4_REG);
+
+	spin_unlock_irqrestore(&geth->irq_lock, flags);
+}
+
+static void gmac_enable_rx_irq(struct net_device *netdev, int enable)
+{
+	struct gemini_ethernet_port *port = netdev_priv(netdev);
+	struct gemini_ethernet *geth = port->geth;
+	unsigned long flags;
+	u32 val, mask;
+
+	netdev_dbg(netdev, "%s device %d %s\n", __func__, netdev->dev_id,
+		   enable ? "enable" : "disable");
+	spin_lock_irqsave(&geth->irq_lock, flags);
+	mask = DEFAULT_Q0_INT_BIT << netdev->dev_id;
+
+	val = readl(geth->base + GLOBAL_INTERRUPT_ENABLE_1_REG);
+	val = enable ? (val | mask) : (val & ~mask);
+	writel(val, geth->base + GLOBAL_INTERRUPT_ENABLE_1_REG);
+
+	spin_unlock_irqrestore(&geth->irq_lock, flags);
+}
+
+static struct sk_buff *gmac_skb_if_good_frame(struct gemini_ethernet_port *port,
+					      union gmac_rxdesc_0 word0,
+					      unsigned int frame_len)
+{
+	unsigned int rx_csum = word0.bits.chksum_status;
+	unsigned int rx_status = word0.bits.status;
+	struct sk_buff *skb = NULL;
+
+	port->rx_stats[rx_status]++;
+	port->rx_csum_stats[rx_csum]++;
+
+	if (word0.bits.derr || word0.bits.perr ||
+	    rx_status || frame_len < ETH_ZLEN ||
+	    rx_csum >= RX_CHKSUM_IP_ERR_UNKNOWN) {
+		port->stats.rx_errors++;
+
+		if (frame_len < ETH_ZLEN || RX_ERROR_LENGTH(rx_status))
+			port->stats.rx_length_errors++;
+		if (RX_ERROR_OVER(rx_status))
+			port->stats.rx_over_errors++;
+		if (RX_ERROR_CRC(rx_status))
+			port->stats.rx_crc_errors++;
+		if (RX_ERROR_FRAME(rx_status))
+			port->stats.rx_frame_errors++;
+		return NULL;
+	}
+
+	skb = napi_get_frags(&port->napi);
+	if (!skb)
+		goto update_exit;
+
+	if (rx_csum == RX_CHKSUM_IP_UDP_TCP_OK)
+		skb->ip_summed = CHECKSUM_UNNECESSARY;
+
+update_exit:
+	port->stats.rx_bytes += frame_len;
+	port->stats.rx_packets++;
+	return skb;
+}
+
+static unsigned int gmac_rx(struct net_device *netdev, unsigned int budget)
+{
+	struct gemini_ethernet_port *port = netdev_priv(netdev);
+	unsigned short m = (1 << port->rxq_order) - 1;
+	struct gemini_ethernet *geth = port->geth;
+	void __iomem *ptr_reg = port->rxq_rwptr;
+	unsigned int frame_len, frag_len;
+	struct gmac_rxdesc *rx = NULL;
+	struct gmac_queue_page *gpage;
+	static struct sk_buff *skb;
+	union gmac_rxdesc_0 word0;
+	union gmac_rxdesc_1 word1;
+	union gmac_rxdesc_3 word3;
+	struct page *page = NULL;
+	unsigned int page_offs;
+	unsigned long flags;
+	unsigned short r, w;
+	union dma_rwptr rw;
+	dma_addr_t mapping;
+	int frag_nr = 0;
+
+	spin_lock_irqsave(&geth->irq_lock, flags);
+	rw.bits32 = readl(ptr_reg);
+	/* Reset interrupt as all packages until here are taken into account */
+	writel(DEFAULT_Q0_INT_BIT << netdev->dev_id,
+	       geth->base + GLOBAL_INTERRUPT_STATUS_1_REG);
+	spin_unlock_irqrestore(&geth->irq_lock, flags);
+
+	r = rw.bits.rptr;
+	w = rw.bits.wptr;
+
+	while (budget && w != r) {
+		rx = port->rxq_ring + r;
+		word0 = rx->word0;
+		word1 = rx->word1;
+		mapping = rx->word2.buf_adr;
+		word3 = rx->word3;
+
+		r++;
+		r &= m;
+
+		frag_len = word0.bits.buffer_size;
+		frame_len = word1.bits.byte_count;
+		page_offs = mapping & ~PAGE_MASK;
+
+		if (!mapping) {
+			netdev_err(netdev,
+				   "rxq[%u]: HW BUG: zero DMA desc\n", r);
+			goto err_drop;
+		}
+
+		/* Freeq pointers are one page off */
+		gpage = gmac_get_queue_page(geth, port, mapping + PAGE_SIZE);
+		if (!gpage) {
+			dev_err(geth->dev, "could not find mapping\n");
+			continue;
+		}
+		page = gpage->page;
+
+		if (word3.bits32 & SOF_BIT) {
+			if (skb) {
+				napi_free_frags(&port->napi);
+				port->stats.rx_dropped++;
+			}
+
+			skb = gmac_skb_if_good_frame(port, word0, frame_len);
+			if (!skb)
+				goto err_drop;
+
+			page_offs += NET_IP_ALIGN;
+			frag_len -= NET_IP_ALIGN;
+			frag_nr = 0;
+
+		} else if (!skb) {
+			put_page(page);
+			continue;
+		}
+
+		if (word3.bits32 & EOF_BIT)
+			frag_len = frame_len - skb->len;
+
+		/* append page frag to skb */
+		if (frag_nr == MAX_SKB_FRAGS)
+			goto err_drop;
+
+		if (frag_len == 0)
+			netdev_err(netdev, "Received fragment with len = 0\n");
+
+		skb_fill_page_desc(skb, frag_nr, page, page_offs, frag_len);
+		skb->len += frag_len;
+		skb->data_len += frag_len;
+		skb->truesize += frag_len;
+		frag_nr++;
+
+		if (word3.bits32 & EOF_BIT) {
+			napi_gro_frags(&port->napi);
+			skb = NULL;
+			--budget;
+		}
+		continue;
+
+err_drop:
+		if (skb) {
+			napi_free_frags(&port->napi);
+			skb = NULL;
+		}
+
+		if (mapping)
+			put_page(page);
+
+		port->stats.rx_dropped++;
+	}
+
+	writew(r, ptr_reg);
+	return budget;
+}
+
+static int gmac_napi_poll(struct napi_struct *napi, int budget)
+{
+	struct gemini_ethernet_port *port = netdev_priv(napi->dev);
+	struct gemini_ethernet *geth = port->geth;
+	unsigned int freeq_threshold;
+	unsigned int received;
+
+	freeq_threshold = 1 << (geth->freeq_order - 1);
+	u64_stats_update_begin(&port->rx_stats_syncp);
+
+	received = gmac_rx(napi->dev, budget);
+	if (received < budget) {
+		napi_gro_flush(napi, false);
+		napi_complete_done(napi, received);
+		gmac_enable_rx_irq(napi->dev, 1);
+		++port->rx_napi_exits;
+	}
+
+	port->freeq_refill += (budget - received);
+	if (port->freeq_refill > freeq_threshold) {
+		port->freeq_refill -= freeq_threshold;
+		geth_fill_freeq(geth, true);
+	}
+
+	u64_stats_update_end(&port->rx_stats_syncp);
+	return received;
+}
+
+static void gmac_dump_dma_state(struct net_device *netdev)
+{
+	struct gemini_ethernet_port *port = netdev_priv(netdev);
+	struct gemini_ethernet *geth = port->geth;
+	void __iomem *ptr_reg;
+	u32 reg[5];
+
+	/* Interrupt status */
+	reg[0] = readl(geth->base + GLOBAL_INTERRUPT_STATUS_0_REG);
+	reg[1] = readl(geth->base + GLOBAL_INTERRUPT_STATUS_1_REG);
+	reg[2] = readl(geth->base + GLOBAL_INTERRUPT_STATUS_2_REG);
+	reg[3] = readl(geth->base + GLOBAL_INTERRUPT_STATUS_3_REG);
+	reg[4] = readl(geth->base + GLOBAL_INTERRUPT_STATUS_4_REG);
+	netdev_err(netdev, "IRQ status: 0x%08x 0x%08x 0x%08x 0x%08x 0x%08x\n",
+		   reg[0], reg[1], reg[2], reg[3], reg[4]);
+
+	/* Interrupt enable */
+	reg[0] = readl(geth->base + GLOBAL_INTERRUPT_ENABLE_0_REG);
+	reg[1] = readl(geth->base + GLOBAL_INTERRUPT_ENABLE_1_REG);
+	reg[2] = readl(geth->base + GLOBAL_INTERRUPT_ENABLE_2_REG);
+	reg[3] = readl(geth->base + GLOBAL_INTERRUPT_ENABLE_3_REG);
+	reg[4] = readl(geth->base + GLOBAL_INTERRUPT_ENABLE_4_REG);
+	netdev_err(netdev, "IRQ enable: 0x%08x 0x%08x 0x%08x 0x%08x 0x%08x\n",
+		   reg[0], reg[1], reg[2], reg[3], reg[4]);
+
+	/* RX DMA status */
+	reg[0] = readl(port->dma_base + GMAC_DMA_RX_FIRST_DESC_REG);
+	reg[1] = readl(port->dma_base + GMAC_DMA_RX_CURR_DESC_REG);
+	reg[2] = GET_RPTR(port->rxq_rwptr);
+	reg[3] = GET_WPTR(port->rxq_rwptr);
+	netdev_err(netdev, "RX DMA regs: 0x%08x 0x%08x, ptr: %u %u\n",
+		   reg[0], reg[1], reg[2], reg[3]);
+
+	reg[0] = readl(port->dma_base + GMAC_DMA_RX_DESC_WORD0_REG);
+	reg[1] = readl(port->dma_base + GMAC_DMA_RX_DESC_WORD1_REG);
+	reg[2] = readl(port->dma_base + GMAC_DMA_RX_DESC_WORD2_REG);
+	reg[3] = readl(port->dma_base + GMAC_DMA_RX_DESC_WORD3_REG);
+	netdev_err(netdev, "RX DMA descriptor: 0x%08x 0x%08x 0x%08x 0x%08x\n",
+		   reg[0], reg[1], reg[2], reg[3]);
+
+	/* TX DMA status */
+	ptr_reg = port->dma_base + GMAC_SW_TX_QUEUE0_PTR_REG;
+
+	reg[0] = readl(port->dma_base + GMAC_DMA_TX_FIRST_DESC_REG);
+	reg[1] = readl(port->dma_base + GMAC_DMA_TX_CURR_DESC_REG);
+	reg[2] = GET_RPTR(ptr_reg);
+	reg[3] = GET_WPTR(ptr_reg);
+	netdev_err(netdev, "TX DMA regs: 0x%08x 0x%08x, ptr: %u %u\n",
+		   reg[0], reg[1], reg[2], reg[3]);
+
+	reg[0] = readl(port->dma_base + GMAC_DMA_TX_DESC_WORD0_REG);
+	reg[1] = readl(port->dma_base + GMAC_DMA_TX_DESC_WORD1_REG);
+	reg[2] = readl(port->dma_base + GMAC_DMA_TX_DESC_WORD2_REG);
+	reg[3] = readl(port->dma_base + GMAC_DMA_TX_DESC_WORD3_REG);
+	netdev_err(netdev, "TX DMA descriptor: 0x%08x 0x%08x 0x%08x 0x%08x\n",
+		   reg[0], reg[1], reg[2], reg[3]);
+
+	/* FREE queues status */
+	ptr_reg = geth->base + GLOBAL_SWFQ_RWPTR_REG;
+
+	reg[0] = GET_RPTR(ptr_reg);
+	reg[1] = GET_WPTR(ptr_reg);
+
+	ptr_reg = geth->base + GLOBAL_HWFQ_RWPTR_REG;
+
+	reg[2] = GET_RPTR(ptr_reg);
+	reg[3] = GET_WPTR(ptr_reg);
+	netdev_err(netdev, "FQ SW ptr: %u %u, HW ptr: %u %u\n",
+		   reg[0], reg[1], reg[2], reg[3]);
+}
+
+static void gmac_update_hw_stats(struct net_device *netdev)
+{
+	struct gemini_ethernet_port *port = netdev_priv(netdev);
+	unsigned int rx_discards, rx_mcast, rx_bcast;
+	struct gemini_ethernet *geth = port->geth;
+	unsigned long flags;
+
+	spin_lock_irqsave(&geth->irq_lock, flags);
+	u64_stats_update_begin(&port->ir_stats_syncp);
+
+	rx_discards = readl(port->gmac_base + GMAC_IN_DISCARDS);
+	port->hw_stats[0] += rx_discards;
+	port->hw_stats[1] += readl(port->gmac_base + GMAC_IN_ERRORS);
+	rx_mcast = readl(port->gmac_base + GMAC_IN_MCAST);
+	port->hw_stats[2] += rx_mcast;
+	rx_bcast = readl(port->gmac_base + GMAC_IN_BCAST);
+	port->hw_stats[3] += rx_bcast;
+	port->hw_stats[4] += readl(port->gmac_base + GMAC_IN_MAC1);
+	port->hw_stats[5] += readl(port->gmac_base + GMAC_IN_MAC2);
+
+	port->stats.rx_missed_errors += rx_discards;
+	port->stats.multicast += rx_mcast;
+	port->stats.multicast += rx_bcast;
+
+	writel(GMAC0_MIB_INT_BIT << (netdev->dev_id * 8),
+	       geth->base + GLOBAL_INTERRUPT_STATUS_4_REG);
+
+	u64_stats_update_end(&port->ir_stats_syncp);
+	spin_unlock_irqrestore(&geth->irq_lock, flags);
+}
+
+/**
+ * gmac_get_intr_flags() - get interrupt status flags for a port from
+ * @netdev: the net device for the port to get flags from
+ * @i: the interrupt status register 0..4
+ */
+static u32 gmac_get_intr_flags(struct net_device *netdev, int i)
+{
+	struct gemini_ethernet_port *port = netdev_priv(netdev);
+	struct gemini_ethernet *geth = port->geth;
+	void __iomem *irqif_reg, *irqen_reg;
+	unsigned int offs, val;
+
+	/* Calculate the offset using the stride of the status registers */
+	offs = i * (GLOBAL_INTERRUPT_STATUS_1_REG -
+		    GLOBAL_INTERRUPT_STATUS_0_REG);
+
+	irqif_reg = geth->base + GLOBAL_INTERRUPT_STATUS_0_REG + offs;
+	irqen_reg = geth->base + GLOBAL_INTERRUPT_ENABLE_0_REG + offs;
+
+	val = readl(irqif_reg) & readl(irqen_reg);
+	return val;
+}
+
+static enum hrtimer_restart gmac_coalesce_delay_expired(struct hrtimer *timer)
+{
+	struct gemini_ethernet_port *port =
+		container_of(timer, struct gemini_ethernet_port,
+			     rx_coalesce_timer);
+
+	napi_schedule(&port->napi);
+	return HRTIMER_NORESTART;
+}
+
+static irqreturn_t gmac_irq(int irq, void *data)
+{
+	struct gemini_ethernet_port *port;
+	struct net_device *netdev = data;
+	struct gemini_ethernet *geth;
+	u32 val, orr = 0;
+
+	port = netdev_priv(netdev);
+	geth = port->geth;
+
+	val = gmac_get_intr_flags(netdev, 0);
+	orr |= val;
+
+	if (val & (GMAC0_IRQ0_2 << (netdev->dev_id * 2))) {
+		/* Oh, crap */
+		netdev_err(netdev, "hw failure/sw bug\n");
+		gmac_dump_dma_state(netdev);
+
+		/* don't know how to recover, just reduce losses */
+		gmac_enable_irq(netdev, 0);
+		return IRQ_HANDLED;
+	}
+
+	if (val & (GMAC0_IRQ0_TXQ0_INTS << (netdev->dev_id * 6)))
+		gmac_tx_irq(netdev, 0);
+
+	val = gmac_get_intr_flags(netdev, 1);
+	orr |= val;
+
+	if (val & (DEFAULT_Q0_INT_BIT << netdev->dev_id)) {
+		gmac_enable_rx_irq(netdev, 0);
+
+		if (!port->rx_coalesce_nsecs) {
+			napi_schedule(&port->napi);
+		} else {
+			ktime_t ktime;
+
+			ktime = ktime_set(0, port->rx_coalesce_nsecs);
+			hrtimer_start(&port->rx_coalesce_timer, ktime,
+				      HRTIMER_MODE_REL);
+		}
+	}
+
+	val = gmac_get_intr_flags(netdev, 4);
+	orr |= val;
+
+	if (val & (GMAC0_MIB_INT_BIT << (netdev->dev_id * 8)))
+		gmac_update_hw_stats(netdev);
+
+	if (val & (GMAC0_RX_OVERRUN_INT_BIT << (netdev->dev_id * 8))) {
+		spin_lock(&geth->irq_lock);
+		writel(GMAC0_RXDERR_INT_BIT << (netdev->dev_id * 8),
+		       geth->base + GLOBAL_INTERRUPT_STATUS_4_REG);
+		u64_stats_update_begin(&port->ir_stats_syncp);
+		++port->stats.rx_fifo_errors;
+		u64_stats_update_end(&port->ir_stats_syncp);
+		spin_unlock(&geth->irq_lock);
+	}
+
+	return orr ? IRQ_HANDLED : IRQ_NONE;
+}
+
+static void gmac_start_dma(struct gemini_ethernet_port *port)
+{
+	void __iomem *dma_ctrl_reg = port->dma_base + GMAC_DMA_CTRL_REG;
+	union gmac_dma_ctrl dma_ctrl;
+
+	dma_ctrl.bits32 = readl(dma_ctrl_reg);
+	dma_ctrl.bits.rd_enable = 1;
+	dma_ctrl.bits.td_enable = 1;
+	dma_ctrl.bits.loopback = 0;
+	dma_ctrl.bits.drop_small_ack = 0;
+	dma_ctrl.bits.rd_insert_bytes = NET_IP_ALIGN;
+	dma_ctrl.bits.rd_prot = HPROT_DATA_CACHE | HPROT_PRIVILIGED;
+	dma_ctrl.bits.rd_burst_size = HBURST_INCR8;
+	dma_ctrl.bits.rd_bus = HSIZE_8;
+	dma_ctrl.bits.td_prot = HPROT_DATA_CACHE;
+	dma_ctrl.bits.td_burst_size = HBURST_INCR8;
+	dma_ctrl.bits.td_bus = HSIZE_8;
+
+	writel(dma_ctrl.bits32, dma_ctrl_reg);
+}
+
+static void gmac_stop_dma(struct gemini_ethernet_port *port)
+{
+	void __iomem *dma_ctrl_reg = port->dma_base + GMAC_DMA_CTRL_REG;
+	union gmac_dma_ctrl dma_ctrl;
+
+	dma_ctrl.bits32 = readl(dma_ctrl_reg);
+	dma_ctrl.bits.rd_enable = 0;
+	dma_ctrl.bits.td_enable = 0;
+	writel(dma_ctrl.bits32, dma_ctrl_reg);
+}
+
+static int gmac_open(struct net_device *netdev)
+{
+	struct gemini_ethernet_port *port = netdev_priv(netdev);
+	int err;
+
+	err = request_irq(netdev->irq, gmac_irq,
+			  IRQF_SHARED, netdev->name, netdev);
+	if (err) {
+		netdev_err(netdev, "no IRQ\n");
+		return err;
+	}
+
+	netif_carrier_off(netdev);
+	phy_start(netdev->phydev);
+
+	err = geth_resize_freeq(port);
+	/* It's fine if it's just busy, the other port has set up
+	 * the freeq in that case.
+	 */
+	if (err && (err != -EBUSY)) {
+		netdev_err(netdev, "could not resize freeq\n");
+		goto err_stop_phy;
+	}
+
+	err = gmac_setup_rxq(netdev);
+	if (err) {
+		netdev_err(netdev, "could not setup RXQ\n");
+		goto err_stop_phy;
+	}
+
+	err = gmac_setup_txqs(netdev);
+	if (err) {
+		netdev_err(netdev, "could not setup TXQs\n");
+		gmac_cleanup_rxq(netdev);
+		goto err_stop_phy;
+	}
+
+	napi_enable(&port->napi);
+
+	gmac_start_dma(port);
+	gmac_enable_irq(netdev, 1);
+	gmac_enable_tx_rx(netdev);
+	netif_tx_start_all_queues(netdev);
+
+	hrtimer_setup(&port->rx_coalesce_timer, &gmac_coalesce_delay_expired, CLOCK_MONOTONIC,
+		      HRTIMER_MODE_REL);
+
+	netdev_dbg(netdev, "opened\n");
+
+	return 0;
+
+err_stop_phy:
+	phy_stop(netdev->phydev);
+	free_irq(netdev->irq, netdev);
+	return err;
+}
+
+static int gmac_stop(struct net_device *netdev)
+{
+	struct gemini_ethernet_port *port = netdev_priv(netdev);
+
+	hrtimer_cancel(&port->rx_coalesce_timer);
+	netif_tx_stop_all_queues(netdev);
+	gmac_disable_tx_rx(netdev);
+	gmac_stop_dma(port);
+	napi_disable(&port->napi);
+
+	gmac_enable_irq(netdev, 0);
+	gmac_cleanup_rxq(netdev);
+	gmac_cleanup_txqs(netdev);
+
+	phy_stop(netdev->phydev);
+	free_irq(netdev->irq, netdev);
+
+	gmac_update_hw_stats(netdev);
+	return 0;
+}
+
+static void gmac_set_rx_mode(struct net_device *netdev)
+{
+	struct gemini_ethernet_port *port = netdev_priv(netdev);
+	union gmac_rx_fltr filter = { .bits = {
+		.broadcast = 1,
+		.multicast = 1,
+		.unicast = 1,
+	} };
+	struct netdev_hw_addr *ha;
+	unsigned int bit_nr;
+	u32 mc_filter[2];
+
+	mc_filter[1] = 0;
+	mc_filter[0] = 0;
+
+	if (netdev->flags & IFF_PROMISC) {
+		filter.bits.error = 1;
+		filter.bits.promiscuous = 1;
+		mc_filter[1] = ~0;
+		mc_filter[0] = ~0;
+	} else if (netdev->flags & IFF_ALLMULTI) {
+		mc_filter[1] = ~0;
+		mc_filter[0] = ~0;
+	} else {
+		netdev_for_each_mc_addr(ha, netdev) {
+			bit_nr = ~crc32_le(~0, ha->addr, ETH_ALEN) & 0x3f;
+			mc_filter[bit_nr >> 5] |= 1 << (bit_nr & 0x1f);
+		}
+	}
+
+	writel(mc_filter[0], port->gmac_base + GMAC_MCAST_FIL0);
+	writel(mc_filter[1], port->gmac_base + GMAC_MCAST_FIL1);
+	writel(filter.bits32, port->gmac_base + GMAC_RX_FLTR);
+}
+
+static void gmac_write_mac_address(struct net_device *netdev)
+{
+	struct gemini_ethernet_port *port = netdev_priv(netdev);
+	__le32 addr[3];
+
+	memset(addr, 0, sizeof(addr));
+	memcpy(addr, netdev->dev_addr, ETH_ALEN);
+
+	writel(le32_to_cpu(addr[0]), port->gmac_base + GMAC_STA_ADD0);
+	writel(le32_to_cpu(addr[1]), port->gmac_base + GMAC_STA_ADD1);
+	writel(le32_to_cpu(addr[2]), port->gmac_base + GMAC_STA_ADD2);
+}
+
+static int gmac_set_mac_address(struct net_device *netdev, void *addr)
+{
+	struct sockaddr *sa = addr;
+
+	eth_hw_addr_set(netdev, sa->sa_data);
+	gmac_write_mac_address(netdev);
+
+	return 0;
+}
+
+static void gmac_clear_hw_stats(struct net_device *netdev)
+{
+	struct gemini_ethernet_port *port = netdev_priv(netdev);
+
+	readl(port->gmac_base + GMAC_IN_DISCARDS);
+	readl(port->gmac_base + GMAC_IN_ERRORS);
+	readl(port->gmac_base + GMAC_IN_MCAST);
+	readl(port->gmac_base + GMAC_IN_BCAST);
+	readl(port->gmac_base + GMAC_IN_MAC1);
+	readl(port->gmac_base + GMAC_IN_MAC2);
+}
+
+static void gmac_get_stats64(struct net_device *netdev,
+			     struct rtnl_link_stats64 *stats)
+{
+	struct gemini_ethernet_port *port = netdev_priv(netdev);
+	unsigned int start;
+
+	gmac_update_hw_stats(netdev);
+
+	/* Racing with RX NAPI */
+	do {
+		start = u64_stats_fetch_begin(&port->rx_stats_syncp);
+
+		stats->rx_packets = port->stats.rx_packets;
+		stats->rx_bytes = port->stats.rx_bytes;
+		stats->rx_errors = port->stats.rx_errors;
+		stats->rx_dropped = port->stats.rx_dropped;
+
+		stats->rx_length_errors = port->stats.rx_length_errors;
+		stats->rx_over_errors = port->stats.rx_over_errors;
+		stats->rx_crc_errors = port->stats.rx_crc_errors;
+		stats->rx_frame_errors = port->stats.rx_frame_errors;
+
+	} while (u64_stats_fetch_retry(&port->rx_stats_syncp, start));
+
+	/* Racing with MIB and TX completion interrupts */
+	do {
+		start = u64_stats_fetch_begin(&port->ir_stats_syncp);
+
+		stats->tx_errors = port->stats.tx_errors;
+		stats->tx_packets = port->stats.tx_packets;
+		stats->tx_bytes = port->stats.tx_bytes;
+
+		stats->multicast = port->stats.multicast;
+		stats->rx_missed_errors = port->stats.rx_missed_errors;
+		stats->rx_fifo_errors = port->stats.rx_fifo_errors;
+
+	} while (u64_stats_fetch_retry(&port->ir_stats_syncp, start));
+
+	/* Racing with hard_start_xmit */
+	do {
+		start = u64_stats_fetch_begin(&port->tx_stats_syncp);
+
+		stats->tx_dropped = port->stats.tx_dropped;
+
+	} while (u64_stats_fetch_retry(&port->tx_stats_syncp, start));
+
+	stats->rx_dropped += stats->rx_missed_errors;
+}
+
+static int gmac_change_mtu(struct net_device *netdev, int new_mtu)
+{
+	int max_len = gmac_pick_rx_max_len(new_mtu);
+
+	if (max_len < 0)
+		return -EINVAL;
+
+	gmac_disable_tx_rx(netdev);
+
+	WRITE_ONCE(netdev->mtu, new_mtu);
+	gmac_update_config0_reg(netdev, max_len << CONFIG0_MAXLEN_SHIFT,
+				CONFIG0_MAXLEN_MASK);
+
+	netdev_update_features(netdev);
+
+	gmac_enable_tx_rx(netdev);
+
+	return 0;
+}
+
+static int gmac_set_features(struct net_device *netdev,
+			     netdev_features_t features)
+{
+	struct gemini_ethernet_port *port = netdev_priv(netdev);
+	int enable = features & NETIF_F_RXCSUM;
+	unsigned long flags;
+	u32 reg;
+
+	spin_lock_irqsave(&port->config_lock, flags);
+
+	reg = readl(port->gmac_base + GMAC_CONFIG0);
+	reg = enable ? reg | CONFIG0_RX_CHKSUM : reg & ~CONFIG0_RX_CHKSUM;
+	writel(reg, port->gmac_base + GMAC_CONFIG0);
+
+	spin_unlock_irqrestore(&port->config_lock, flags);
+	return 0;
+}
+
+static int gmac_get_sset_count(struct net_device *netdev, int sset)
+{
+	return sset == ETH_SS_STATS ? GMAC_STATS_NUM : 0;
+}
+
+static void gmac_get_strings(struct net_device *netdev, u32 stringset, u8 *data)
+{
+	if (stringset != ETH_SS_STATS)
+		return;
+
+	memcpy(data, gmac_stats_strings, sizeof(gmac_stats_strings));
+}
+
+static void gmac_get_ethtool_stats(struct net_device *netdev,
+				   struct ethtool_stats *estats, u64 *values)
+{
+	struct gemini_ethernet_port *port = netdev_priv(netdev);
+	unsigned int start;
+	u64 *p;
+	int i;
+
+	gmac_update_hw_stats(netdev);
+
+	/* Racing with MIB interrupt */
+	do {
+		p = values;
+		start = u64_stats_fetch_begin(&port->ir_stats_syncp);
+
+		for (i = 0; i < RX_STATS_NUM; i++)
+			*p++ = port->hw_stats[i];
+
+	} while (u64_stats_fetch_retry(&port->ir_stats_syncp, start));
+	values = p;
+
+	/* Racing with RX NAPI */
+	do {
+		p = values;
+		start = u64_stats_fetch_begin(&port->rx_stats_syncp);
+
+		for (i = 0; i < RX_STATUS_NUM; i++)
+			*p++ = port->rx_stats[i];
+		for (i = 0; i < RX_CHKSUM_NUM; i++)
+			*p++ = port->rx_csum_stats[i];
+		*p++ = port->rx_napi_exits;
+
+	} while (u64_stats_fetch_retry(&port->rx_stats_syncp, start));
+	values = p;
+
+	/* Racing with TX start_xmit */
+	do {
+		p = values;
+		start = u64_stats_fetch_begin(&port->tx_stats_syncp);
+
+		for (i = 0; i < TX_MAX_FRAGS; i++) {
+			*values++ = port->tx_frag_stats[i];
+			port->tx_frag_stats[i] = 0;
+		}
+		*values++ = port->tx_frags_linearized;
+		*values++ = port->tx_hw_csummed;
+
+	} while (u64_stats_fetch_retry(&port->tx_stats_syncp, start));
+}
+
+static int gmac_get_ksettings(struct net_device *netdev,
+			      struct ethtool_link_ksettings *cmd)
+{
+	if (!netdev->phydev)
+		return -ENXIO;
+	phy_ethtool_ksettings_get(netdev->phydev, cmd);
+
+	return 0;
+}
+
+static int gmac_set_ksettings(struct net_device *netdev,
+			      const struct ethtool_link_ksettings *cmd)
+{
+	if (!netdev->phydev)
+		return -ENXIO;
+	return phy_ethtool_ksettings_set(netdev->phydev, cmd);
+}
+
+static int gmac_nway_reset(struct net_device *netdev)
+{
+	if (!netdev->phydev)
+		return -ENXIO;
+	return phy_start_aneg(netdev->phydev);
+}
+
+static void gmac_get_pauseparam(struct net_device *netdev,
+				struct ethtool_pauseparam *pparam)
+{
+	struct gemini_ethernet_port *port = netdev_priv(netdev);
+	union gmac_config0 config0;
+
+	config0.bits32 = readl(port->gmac_base + GMAC_CONFIG0);
+
+	pparam->rx_pause = config0.bits.rx_fc_en;
+	pparam->tx_pause = config0.bits.tx_fc_en;
+	pparam->autoneg = true;
+}
+
+static int gmac_set_pauseparam(struct net_device *netdev,
+			       struct ethtool_pauseparam *pparam)
+{
+	struct phy_device *phydev = netdev->phydev;
+
+	if (!pparam->autoneg)
+		return -EOPNOTSUPP;
+
+	phy_set_asym_pause(phydev, pparam->rx_pause, pparam->tx_pause);
+
+	return 0;
+}
+
+static void gmac_get_ringparam(struct net_device *netdev,
+			       struct ethtool_ringparam *rp,
+			       struct kernel_ethtool_ringparam *kernel_rp,
+			       struct netlink_ext_ack *extack)
+{
+	struct gemini_ethernet_port *port = netdev_priv(netdev);
+
+	readl(port->gmac_base + GMAC_CONFIG0);
+
+	rp->rx_max_pending = 1 << 15;
+	rp->rx_mini_max_pending = 0;
+	rp->rx_jumbo_max_pending = 0;
+	rp->tx_max_pending = 1 << 15;
+
+	rp->rx_pending = 1 << port->rxq_order;
+	rp->rx_mini_pending = 0;
+	rp->rx_jumbo_pending = 0;
+	rp->tx_pending = 1 << port->txq_order;
+}
+
+static int gmac_set_ringparam(struct net_device *netdev,
+			      struct ethtool_ringparam *rp,
+			      struct kernel_ethtool_ringparam *kernel_rp,
+			      struct netlink_ext_ack *extack)
+{
+	struct gemini_ethernet_port *port = netdev_priv(netdev);
+	int err = 0;
+
+	if (netif_running(netdev))
+		return -EBUSY;
+
+	if (rp->rx_pending) {
+		port->rxq_order = min(15, ilog2(rp->rx_pending - 1) + 1);
+		err = geth_resize_freeq(port);
+	}
+	if (rp->tx_pending) {
+		port->txq_order = min(15, ilog2(rp->tx_pending - 1) + 1);
+		port->irq_every_tx_packets = 1 << (port->txq_order - 2);
+	}
+
+	return err;
+}
+
+static int gmac_get_coalesce(struct net_device *netdev,
+			     struct ethtool_coalesce *ecmd,
+			     struct kernel_ethtool_coalesce *kernel_coal,
+			     struct netlink_ext_ack *extack)
+{
+	struct gemini_ethernet_port *port = netdev_priv(netdev);
+
+	ecmd->rx_max_coalesced_frames = 1;
+	ecmd->tx_max_coalesced_frames = port->irq_every_tx_packets;
+	ecmd->rx_coalesce_usecs = port->rx_coalesce_nsecs / 1000;
+
+	return 0;
+}
+
+static int gmac_set_coalesce(struct net_device *netdev,
+			     struct ethtool_coalesce *ecmd,
+			     struct kernel_ethtool_coalesce *kernel_coal,
+			     struct netlink_ext_ack *extack)
+{
+	struct gemini_ethernet_port *port = netdev_priv(netdev);
+
+	if (ecmd->tx_max_coalesced_frames < 1)
+		return -EINVAL;
+	if (ecmd->tx_max_coalesced_frames >= 1 << port->txq_order)
+		return -EINVAL;
+
+	port->irq_every_tx_packets = ecmd->tx_max_coalesced_frames;
+	port->rx_coalesce_nsecs = ecmd->rx_coalesce_usecs * 1000;
+
+	return 0;
+}
+
+static u32 gmac_get_msglevel(struct net_device *netdev)
+{
+	struct gemini_ethernet_port *port = netdev_priv(netdev);
+
+	return port->msg_enable;
+}
+
+static void gmac_set_msglevel(struct net_device *netdev, u32 level)
+{
+	struct gemini_ethernet_port *port = netdev_priv(netdev);
+
+	port->msg_enable = level;
+}
+
+static void gmac_get_drvinfo(struct net_device *netdev,
+			     struct ethtool_drvinfo *info)
+{
+	strcpy(info->driver,  DRV_NAME);
+	strcpy(info->bus_info, netdev->dev_id ? "1" : "0");
+}
+
+static const struct net_device_ops gmac_351x_ops = {
+	.ndo_init		= gmac_init,
+	.ndo_open		= gmac_open,
+	.ndo_stop		= gmac_stop,
+	.ndo_start_xmit		= gmac_start_xmit,
+	.ndo_tx_timeout		= gmac_tx_timeout,
+	.ndo_set_rx_mode	= gmac_set_rx_mode,
+	.ndo_set_mac_address	= gmac_set_mac_address,
+	.ndo_get_stats64	= gmac_get_stats64,
+	.ndo_change_mtu		= gmac_change_mtu,
+	.ndo_set_features	= gmac_set_features,
+};
+
+static const struct ethtool_ops gmac_351x_ethtool_ops = {
+	.supported_coalesce_params = ETHTOOL_COALESCE_RX_USECS |
+				     ETHTOOL_COALESCE_MAX_FRAMES,
+	.get_sset_count	= gmac_get_sset_count,
+	.get_strings	= gmac_get_strings,
+	.get_ethtool_stats = gmac_get_ethtool_stats,
+	.get_link	= ethtool_op_get_link,
+	.get_link_ksettings = gmac_get_ksettings,
+	.set_link_ksettings = gmac_set_ksettings,
+	.nway_reset	= gmac_nway_reset,
+	.get_pauseparam	= gmac_get_pauseparam,
+	.set_pauseparam = gmac_set_pauseparam,
+	.get_ringparam	= gmac_get_ringparam,
+	.set_ringparam	= gmac_set_ringparam,
+	.get_coalesce	= gmac_get_coalesce,
+	.set_coalesce	= gmac_set_coalesce,
+	.get_msglevel	= gmac_get_msglevel,
+	.set_msglevel	= gmac_set_msglevel,
+	.get_drvinfo	= gmac_get_drvinfo,
+};
+
+static irqreturn_t gemini_port_irq_thread(int irq, void *data)
+{
+	unsigned long irqmask = SWFQ_EMPTY_INT_BIT;
+	struct gemini_ethernet_port *port = data;
+	struct gemini_ethernet *geth;
+	unsigned long flags;
+
+	geth = port->geth;
+	/* The queue is half empty so refill it */
+	geth_fill_freeq(geth, true);
+
+	spin_lock_irqsave(&geth->irq_lock, flags);
+	/* ACK queue interrupt */
+	writel(irqmask, geth->base + GLOBAL_INTERRUPT_STATUS_4_REG);
+	/* Enable queue interrupt again */
+	irqmask |= readl(geth->base + GLOBAL_INTERRUPT_ENABLE_4_REG);
+	writel(irqmask, geth->base + GLOBAL_INTERRUPT_ENABLE_4_REG);
+	spin_unlock_irqrestore(&geth->irq_lock, flags);
+
+	return IRQ_HANDLED;
+}
+
+static irqreturn_t gemini_port_irq(int irq, void *data)
+{
+	struct gemini_ethernet_port *port = data;
+	struct gemini_ethernet *geth;
+	irqreturn_t ret = IRQ_NONE;
+	u32 val, en;
+
+	geth = port->geth;
+	spin_lock(&geth->irq_lock);
+
+	val = readl(geth->base + GLOBAL_INTERRUPT_STATUS_4_REG);
+	en = readl(geth->base + GLOBAL_INTERRUPT_ENABLE_4_REG);
+
+	if (val & en & SWFQ_EMPTY_INT_BIT) {
+		/* Disable the queue empty interrupt while we work on
+		 * processing the queue. Also disable overrun interrupts
+		 * as there is not much we can do about it here.
+		 */
+		en &= ~(SWFQ_EMPTY_INT_BIT | GMAC0_RX_OVERRUN_INT_BIT
+					   | GMAC1_RX_OVERRUN_INT_BIT);
+		writel(en, geth->base + GLOBAL_INTERRUPT_ENABLE_4_REG);
+		ret = IRQ_WAKE_THREAD;
+	}
+
+	spin_unlock(&geth->irq_lock);
+
+	return ret;
+}
+
+static void gemini_port_remove(struct gemini_ethernet_port *port)
+{
+	if (port->netdev) {
+		phy_disconnect(port->netdev->phydev);
+		unregister_netdev(port->netdev);
+	}
+	clk_disable_unprepare(port->pclk);
+	geth_cleanup_freeq(port->geth);
+}
+
+static void gemini_ethernet_init(struct gemini_ethernet *geth)
+{
+	/* Only do this once both ports are online */
+	if (geth->initialized)
+		return;
+	if (geth->port0 && geth->port1)
+		geth->initialized = true;
+	else
+		return;
+
+	writel(0, geth->base + GLOBAL_INTERRUPT_ENABLE_0_REG);
+	writel(0, geth->base + GLOBAL_INTERRUPT_ENABLE_1_REG);
+	writel(0, geth->base + GLOBAL_INTERRUPT_ENABLE_2_REG);
+	writel(0, geth->base + GLOBAL_INTERRUPT_ENABLE_3_REG);
+	writel(0, geth->base + GLOBAL_INTERRUPT_ENABLE_4_REG);
+
+	/* Interrupt config:
+	 *
+	 *	GMAC0 intr bits ------> int0 ----> eth0
+	 *	GMAC1 intr bits ------> int1 ----> eth1
+	 *	TOE intr -------------> int1 ----> eth1
+	 *	Classification Intr --> int0 ----> eth0
+	 *	Default Q0 -----------> int0 ----> eth0
+	 *	Default Q1 -----------> int1 ----> eth1
+	 *	FreeQ intr -----------> int1 ----> eth1
+	 */
+	writel(0xCCFC0FC0, geth->base + GLOBAL_INTERRUPT_SELECT_0_REG);
+	writel(0x00F00002, geth->base + GLOBAL_INTERRUPT_SELECT_1_REG);
+	writel(0xFFFFFFFF, geth->base + GLOBAL_INTERRUPT_SELECT_2_REG);
+	writel(0xFFFFFFFF, geth->base + GLOBAL_INTERRUPT_SELECT_3_REG);
+	writel(0xFF000003, geth->base + GLOBAL_INTERRUPT_SELECT_4_REG);
+
+	/* edge-triggered interrupts packed to level-triggered one... */
+	writel(~0, geth->base + GLOBAL_INTERRUPT_STATUS_0_REG);
+	writel(~0, geth->base + GLOBAL_INTERRUPT_STATUS_1_REG);
+	writel(~0, geth->base + GLOBAL_INTERRUPT_STATUS_2_REG);
+	writel(~0, geth->base + GLOBAL_INTERRUPT_STATUS_3_REG);
+	writel(~0, geth->base + GLOBAL_INTERRUPT_STATUS_4_REG);
+
+	/* Set up queue */
+	writel(0, geth->base + GLOBAL_SW_FREEQ_BASE_SIZE_REG);
+	writel(0, geth->base + GLOBAL_HW_FREEQ_BASE_SIZE_REG);
+	writel(0, geth->base + GLOBAL_SWFQ_RWPTR_REG);
+	writel(0, geth->base + GLOBAL_HWFQ_RWPTR_REG);
+
+	geth->freeq_frag_order = DEFAULT_RX_BUF_ORDER;
+	/* This makes the queue resize on probe() so that we
+	 * set up and enable the queue IRQ. FIXME: fragile.
+	 */
+	geth->freeq_order = 1;
+}
+
+static void gemini_port_save_mac_addr(struct gemini_ethernet_port *port)
+{
+	port->mac_addr[0] =
+		cpu_to_le32(readl(port->gmac_base + GMAC_STA_ADD0));
+	port->mac_addr[1] =
+		cpu_to_le32(readl(port->gmac_base + GMAC_STA_ADD1));
+	port->mac_addr[2] =
+		cpu_to_le32(readl(port->gmac_base + GMAC_STA_ADD2));
+}
+
+static int gemini_ethernet_port_probe(struct platform_device *pdev)
+{
+	char *port_names[2] = { "ethernet0", "ethernet1" };
+	struct device_node *np = pdev->dev.of_node;
+	struct gemini_ethernet_port *port;
+	struct device *dev = &pdev->dev;
+	struct gemini_ethernet *geth;
+	struct net_device *netdev;
+	struct device *parent;
+	u8 mac[ETH_ALEN];
+	unsigned int id;
+	int irq;
+	int ret;
+
+	parent = dev->parent;
+	geth = dev_get_drvdata(parent);
+
+	if (!strcmp(dev_name(dev), "60008000.ethernet-port"))
+		id = 0;
+	else if (!strcmp(dev_name(dev), "6000c000.ethernet-port"))
+		id = 1;
+	else
+		return -ENODEV;
+
+	dev_info(dev, "probe %s ID %d\n", dev_name(dev), id);
+
+	netdev = devm_alloc_etherdev_mqs(dev, sizeof(*port), TX_QUEUE_NUM, TX_QUEUE_NUM);
+	if (!netdev) {
+		dev_err(dev, "Can't allocate ethernet device #%d\n", id);
+		return -ENOMEM;
+	}
+
+	port = netdev_priv(netdev);
+	SET_NETDEV_DEV(netdev, dev);
+	port->netdev = netdev;
+	port->id = id;
+	port->geth = geth;
+	port->dev = dev;
+	port->msg_enable = netif_msg_init(debug, DEFAULT_MSG_ENABLE);
+
+	/* DMA memory */
+	port->dma_base = devm_platform_get_and_ioremap_resource(pdev, 0, NULL);
+	if (IS_ERR(port->dma_base)) {
+		dev_err(dev, "get DMA address failed\n");
+		return PTR_ERR(port->dma_base);
+	}
+
+	/* GMAC config memory */
+	port->gmac_base = devm_platform_get_and_ioremap_resource(pdev, 1, NULL);
+	if (IS_ERR(port->gmac_base)) {
+		dev_err(dev, "get GMAC address failed\n");
+		return PTR_ERR(port->gmac_base);
+	}
+
+	/* Interrupt */
+	irq = platform_get_irq(pdev, 0);
+	if (irq < 0)
+		return irq;
+	port->irq = irq;
+
+	/* Clock the port */
+	port->pclk = devm_clk_get(dev, "PCLK");
+	if (IS_ERR(port->pclk)) {
+		dev_err(dev, "no PCLK\n");
+		return PTR_ERR(port->pclk);
+	}
+	ret = clk_prepare_enable(port->pclk);
+	if (ret)
+		return ret;
+
+	/* Maybe there is a nice ethernet address we should use */
+	gemini_port_save_mac_addr(port);
+
+	/* Reset the port */
+	port->reset = devm_reset_control_get_exclusive(dev, NULL);
+	if (IS_ERR(port->reset)) {
+		dev_err(dev, "no reset\n");
+		ret = PTR_ERR(port->reset);
+		goto unprepare;
+	}
+	reset_control_reset(port->reset);
+	usleep_range(100, 500);
+
+	/* Assign pointer in the main state container */
+	if (!id)
+		geth->port0 = port;
+	else
+		geth->port1 = port;
+
+	/* This will just be done once both ports are up and reset */
+	gemini_ethernet_init(geth);
+
+	platform_set_drvdata(pdev, port);
+
+	/* Set up and register the netdev */
+	netdev->dev_id = port->id;
+	netdev->irq = irq;
+	netdev->netdev_ops = &gmac_351x_ops;
+	netdev->ethtool_ops = &gmac_351x_ethtool_ops;
+
+	spin_lock_init(&port->config_lock);
+	gmac_clear_hw_stats(netdev);
+
+	netdev->hw_features = GMAC_OFFLOAD_FEATURES;
+	netdev->features |= GMAC_OFFLOAD_FEATURES | NETIF_F_GRO;
+	/* We can receive jumbo frames up to 10236 bytes but only
+	 * transmit 2047 bytes so, let's accept payloads of 2047
+	 * bytes minus VLAN and ethernet header
+	 */
+	netdev->min_mtu = ETH_MIN_MTU;
+	netdev->max_mtu = MTU_SIZE_BIT_MASK - VLAN_ETH_HLEN;
+
+	port->freeq_refill = 0;
+	netif_napi_add(netdev, &port->napi, gmac_napi_poll);
+
+	ret = of_get_mac_address(np, mac);
+	if (!ret) {
+		dev_info(dev, "Setting macaddr from DT %pM\n", mac);
+		memcpy(port->mac_addr, mac, ETH_ALEN);
+	}
+
+	if (is_valid_ether_addr((void *)port->mac_addr)) {
+		eth_hw_addr_set(netdev, (u8 *)port->mac_addr);
+	} else {
+		dev_dbg(dev, "ethernet address 0x%08x%08x%08x invalid\n",
+			port->mac_addr[0], port->mac_addr[1],
+			port->mac_addr[2]);
+		dev_info(dev, "using a random ethernet address\n");
+		eth_hw_addr_random(netdev);
+	}
+	gmac_write_mac_address(netdev);
+
+	ret = devm_request_threaded_irq(port->dev,
+					port->irq,
+					gemini_port_irq,
+					gemini_port_irq_thread,
+					IRQF_SHARED,
+					port_names[port->id],
+					port);
+	if (ret)
+		goto unprepare;
+
+	ret = gmac_setup_phy(netdev);
+	if (ret) {
+		netdev_err(netdev,
+			   "PHY init failed\n");
+		goto unprepare;
+	}
+
+	ret = register_netdev(netdev);
+	if (ret)
+		goto unprepare;
+
+	return 0;
+
+unprepare:
+	clk_disable_unprepare(port->pclk);
+	return ret;
+}
+
+static void gemini_ethernet_port_remove(struct platform_device *pdev)
+{
+	struct gemini_ethernet_port *port = platform_get_drvdata(pdev);
+
+	gemini_port_remove(port);
+}
+
+static const struct of_device_id gemini_ethernet_port_of_match[] = {
+	{
+		.compatible = "cortina,gemini-ethernet-port",
+	},
+	{},
+};
+MODULE_DEVICE_TABLE(of, gemini_ethernet_port_of_match);
+
+static struct platform_driver gemini_ethernet_port_driver = {
+	.driver = {
+		.name = "gemini-ethernet-port",
+		.of_match_table = gemini_ethernet_port_of_match,
+	},
+	.probe = gemini_ethernet_port_probe,
+	.remove = gemini_ethernet_port_remove,
+};
+
+static int gemini_ethernet_probe(struct platform_device *pdev)
+{
+	struct device *dev = &pdev->dev;
+	struct gemini_ethernet *geth;
+	unsigned int retry = 5;
+	u32 val;
+
+	/* Global registers */
+	geth = devm_kzalloc(dev, sizeof(*geth), GFP_KERNEL);
+	if (!geth)
+		return -ENOMEM;
+	geth->base = devm_platform_get_and_ioremap_resource(pdev, 0, NULL);
+	if (IS_ERR(geth->base))
+		return PTR_ERR(geth->base);
+	geth->dev = dev;
+
+	/* Wait for ports to stabilize */
+	do {
+		udelay(2);
+		val = readl(geth->base + GLOBAL_TOE_VERSION_REG);
+		barrier();
+	} while (!val && --retry);
+	if (!retry) {
+		dev_err(dev, "failed to reset ethernet\n");
+		return -EIO;
+	}
+	dev_info(dev, "Ethernet device ID: 0x%03x, revision 0x%01x\n",
+		 (val >> 4) & 0xFFFU, val & 0xFU);
+
+	spin_lock_init(&geth->irq_lock);
+	spin_lock_init(&geth->freeq_lock);
+
+	/* The children will use this */
+	platform_set_drvdata(pdev, geth);
+
+	/* Spawn child devices for the two ports */
+	return devm_of_platform_populate(dev);
+}
+
+static void gemini_ethernet_remove(struct platform_device *pdev)
+{
+	struct gemini_ethernet *geth = platform_get_drvdata(pdev);
+
+	geth_cleanup_freeq(geth);
+	geth->initialized = false;
+}
+
+static const struct of_device_id gemini_ethernet_of_match[] = {
+	{
+		.compatible = "cortina,gemini-ethernet",
+	},
+	{},
+};
+MODULE_DEVICE_TABLE(of, gemini_ethernet_of_match);
+
+static struct platform_driver gemini_ethernet_driver = {
+	.driver = {
+		.name = DRV_NAME,
+		.of_match_table = gemini_ethernet_of_match,
+	},
+	.probe = gemini_ethernet_probe,
+	.remove = gemini_ethernet_remove,
+};
+
+static int __init gemini_ethernet_module_init(void)
+{
+	int ret;
+
+	ret = platform_driver_register(&gemini_ethernet_port_driver);
+	if (ret)
+		return ret;
+
+	ret = platform_driver_register(&gemini_ethernet_driver);
+	if (ret) {
+		platform_driver_unregister(&gemini_ethernet_port_driver);
+		return ret;
+	}
+
+	return 0;
+}
+module_init(gemini_ethernet_module_init);
+
+static void __exit gemini_ethernet_module_exit(void)
+{
+	platform_driver_unregister(&gemini_ethernet_driver);
+	platform_driver_unregister(&gemini_ethernet_port_driver);
+}
+module_exit(gemini_ethernet_module_exit);
+
+MODULE_AUTHOR("Linus Walleij <linus.walleij@linaro.org>");
+MODULE_DESCRIPTION("StorLink SL351x (Gemini) ethernet driver");
+MODULE_LICENSE("GPL");
+MODULE_ALIAS("platform:" DRV_NAME);
diff --git a/drivers/net/ethernet/cortina/gemini.h b/drivers/net/ethernet/cortina/gemini.h
new file mode 100644
index 000000000000..24bb989981f2
--- /dev/null
+++ b/drivers/net/ethernet/cortina/gemini.h
@@ -0,0 +1,958 @@
+/* SPDX-License-Identifier: GPL-2.0 */
+/* Register definitions for Gemini GMAC Ethernet device driver
+ *
+ * Copyright (C) 2006 Storlink, Corp.
+ * Copyright (C) 2008-2009 Paulius Zaleckas <paulius.zaleckas@teltonika.lt>
+ * Copyright (C) 2010 Michał Mirosław <mirq-linux@rere.qmqm.pl>
+ * Copytight (C) 2017 Linus Walleij <linus.walleij@linaro.org>
+ */
+#ifndef _GEMINI_ETHERNET_H
+#define _GEMINI_ETHERNET_H
+
+#include <linux/bitops.h>
+
+/* Base Registers */
+#define TOE_NONTOE_QUE_HDR_BASE		0x2000
+#define TOE_TOE_QUE_HDR_BASE		0x3000
+
+/* Queue ID */
+#define TOE_SW_FREE_QID			0x00
+#define TOE_HW_FREE_QID			0x01
+#define TOE_GMAC0_SW_TXQ0_QID		0x02
+#define TOE_GMAC0_SW_TXQ1_QID		0x03
+#define TOE_GMAC0_SW_TXQ2_QID		0x04
+#define TOE_GMAC0_SW_TXQ3_QID		0x05
+#define TOE_GMAC0_SW_TXQ4_QID		0x06
+#define TOE_GMAC0_SW_TXQ5_QID		0x07
+#define TOE_GMAC0_HW_TXQ0_QID		0x08
+#define TOE_GMAC0_HW_TXQ1_QID		0x09
+#define TOE_GMAC0_HW_TXQ2_QID		0x0A
+#define TOE_GMAC0_HW_TXQ3_QID		0x0B
+#define TOE_GMAC1_SW_TXQ0_QID		0x12
+#define TOE_GMAC1_SW_TXQ1_QID		0x13
+#define TOE_GMAC1_SW_TXQ2_QID		0x14
+#define TOE_GMAC1_SW_TXQ3_QID		0x15
+#define TOE_GMAC1_SW_TXQ4_QID		0x16
+#define TOE_GMAC1_SW_TXQ5_QID		0x17
+#define TOE_GMAC1_HW_TXQ0_QID		0x18
+#define TOE_GMAC1_HW_TXQ1_QID		0x19
+#define TOE_GMAC1_HW_TXQ2_QID		0x1A
+#define TOE_GMAC1_HW_TXQ3_QID		0x1B
+#define TOE_GMAC0_DEFAULT_QID		0x20
+#define TOE_GMAC1_DEFAULT_QID		0x21
+#define TOE_CLASSIFICATION_QID(x)	(0x22 + x)	/* 0x22 ~ 0x2F */
+#define TOE_TOE_QID(x)			(0x40 + x)	/* 0x40 ~ 0x7F */
+
+/* TOE DMA Queue Size should be 2^n, n = 6...12
+ * TOE DMA Queues are the following queue types:
+ *		SW Free Queue, HW Free Queue,
+ *		GMAC 0/1 SW TX Q0-5, and GMAC 0/1 HW TX Q0-5
+ * The base address and descriptor number are configured at
+ * DMA Queues Descriptor Ring Base Address/Size Register (offset 0x0004)
+ */
+#define GET_WPTR(addr)			readw((addr) + 2)
+#define GET_RPTR(addr)			readw((addr))
+#define SET_WPTR(addr, data)		writew((data), (addr) + 2)
+#define SET_RPTR(addr, data)		writew((data), (addr))
+#define __RWPTR_NEXT(x, mask)		(((unsigned int)(x) + 1) & (mask))
+#define __RWPTR_PREV(x, mask)		(((unsigned int)(x) - 1) & (mask))
+#define __RWPTR_DISTANCE(r, w, mask)	(((unsigned int)(w) - (r)) & (mask))
+#define __RWPTR_MASK(order)		((1 << (order)) - 1)
+#define RWPTR_NEXT(x, order)		__RWPTR_NEXT((x), __RWPTR_MASK((order)))
+#define RWPTR_PREV(x, order)		__RWPTR_PREV((x), __RWPTR_MASK((order)))
+#define RWPTR_DISTANCE(r, w, order)	__RWPTR_DISTANCE((r), (w), \
+						__RWPTR_MASK((order)))
+
+/* Global registers */
+#define GLOBAL_TOE_VERSION_REG		0x0000
+#define GLOBAL_SW_FREEQ_BASE_SIZE_REG	0x0004
+#define GLOBAL_HW_FREEQ_BASE_SIZE_REG	0x0008
+#define GLOBAL_DMA_SKB_SIZE_REG		0x0010
+#define GLOBAL_SWFQ_RWPTR_REG		0x0014
+#define GLOBAL_HWFQ_RWPTR_REG		0x0018
+#define GLOBAL_INTERRUPT_STATUS_0_REG	0x0020
+#define GLOBAL_INTERRUPT_ENABLE_0_REG	0x0024
+#define GLOBAL_INTERRUPT_SELECT_0_REG	0x0028
+#define GLOBAL_INTERRUPT_STATUS_1_REG	0x0030
+#define GLOBAL_INTERRUPT_ENABLE_1_REG	0x0034
+#define GLOBAL_INTERRUPT_SELECT_1_REG	0x0038
+#define GLOBAL_INTERRUPT_STATUS_2_REG	0x0040
+#define GLOBAL_INTERRUPT_ENABLE_2_REG	0x0044
+#define GLOBAL_INTERRUPT_SELECT_2_REG	0x0048
+#define GLOBAL_INTERRUPT_STATUS_3_REG	0x0050
+#define GLOBAL_INTERRUPT_ENABLE_3_REG	0x0054
+#define GLOBAL_INTERRUPT_SELECT_3_REG	0x0058
+#define GLOBAL_INTERRUPT_STATUS_4_REG	0x0060
+#define GLOBAL_INTERRUPT_ENABLE_4_REG	0x0064
+#define GLOBAL_INTERRUPT_SELECT_4_REG	0x0068
+#define GLOBAL_HASH_TABLE_BASE_REG	0x006C
+#define GLOBAL_QUEUE_THRESHOLD_REG	0x0070
+
+/* GMAC 0/1 DMA/TOE register */
+#define GMAC_DMA_CTRL_REG		0x0000
+#define GMAC_TX_WEIGHTING_CTRL_0_REG	0x0004
+#define GMAC_TX_WEIGHTING_CTRL_1_REG	0x0008
+#define GMAC_SW_TX_QUEUE0_PTR_REG	0x000C
+#define GMAC_SW_TX_QUEUE1_PTR_REG	0x0010
+#define GMAC_SW_TX_QUEUE2_PTR_REG	0x0014
+#define GMAC_SW_TX_QUEUE3_PTR_REG	0x0018
+#define GMAC_SW_TX_QUEUE4_PTR_REG	0x001C
+#define GMAC_SW_TX_QUEUE5_PTR_REG	0x0020
+#define GMAC_SW_TX_QUEUE_PTR_REG(i)	(GMAC_SW_TX_QUEUE0_PTR_REG + 4 * (i))
+#define GMAC_HW_TX_QUEUE0_PTR_REG	0x0024
+#define GMAC_HW_TX_QUEUE1_PTR_REG	0x0028
+#define GMAC_HW_TX_QUEUE2_PTR_REG	0x002C
+#define GMAC_HW_TX_QUEUE3_PTR_REG	0x0030
+#define GMAC_HW_TX_QUEUE_PTR_REG(i)	(GMAC_HW_TX_QUEUE0_PTR_REG + 4 * (i))
+#define GMAC_DMA_TX_FIRST_DESC_REG	0x0038
+#define GMAC_DMA_TX_CURR_DESC_REG	0x003C
+#define GMAC_DMA_TX_DESC_WORD0_REG	0x0040
+#define GMAC_DMA_TX_DESC_WORD1_REG	0x0044
+#define GMAC_DMA_TX_DESC_WORD2_REG	0x0048
+#define GMAC_DMA_TX_DESC_WORD3_REG	0x004C
+#define GMAC_SW_TX_QUEUE_BASE_REG	0x0050
+#define GMAC_HW_TX_QUEUE_BASE_REG	0x0054
+#define GMAC_DMA_RX_FIRST_DESC_REG	0x0058
+#define GMAC_DMA_RX_CURR_DESC_REG	0x005C
+#define GMAC_DMA_RX_DESC_WORD0_REG	0x0060
+#define GMAC_DMA_RX_DESC_WORD1_REG	0x0064
+#define GMAC_DMA_RX_DESC_WORD2_REG	0x0068
+#define GMAC_DMA_RX_DESC_WORD3_REG	0x006C
+#define GMAC_HASH_ENGINE_REG0		0x0070
+#define GMAC_HASH_ENGINE_REG1		0x0074
+/* matching rule 0 Control register 0 */
+#define GMAC_MR0CR0			0x0078
+#define GMAC_MR0CR1			0x007C
+#define GMAC_MR0CR2			0x0080
+#define GMAC_MR1CR0			0x0084
+#define GMAC_MR1CR1			0x0088
+#define GMAC_MR1CR2			0x008C
+#define GMAC_MR2CR0			0x0090
+#define GMAC_MR2CR1			0x0094
+#define GMAC_MR2CR2			0x0098
+#define GMAC_MR3CR0			0x009C
+#define GMAC_MR3CR1			0x00A0
+#define GMAC_MR3CR2			0x00A4
+/* Support Protocol Register 0 */
+#define GMAC_SPR0			0x00A8
+#define GMAC_SPR1			0x00AC
+#define GMAC_SPR2			0x00B0
+#define GMAC_SPR3			0x00B4
+#define GMAC_SPR4			0x00B8
+#define GMAC_SPR5			0x00BC
+#define GMAC_SPR6			0x00C0
+#define GMAC_SPR7			0x00C4
+/* GMAC Hash/Rx/Tx AHB Weighting register */
+#define GMAC_AHB_WEIGHT_REG		0x00C8
+
+/* TOE GMAC 0/1 register */
+#define GMAC_STA_ADD0			0x0000
+#define GMAC_STA_ADD1			0x0004
+#define GMAC_STA_ADD2			0x0008
+#define GMAC_RX_FLTR			0x000c
+#define GMAC_MCAST_FIL0			0x0010
+#define GMAC_MCAST_FIL1			0x0014
+#define GMAC_CONFIG0			0x0018
+#define GMAC_CONFIG1			0x001c
+#define GMAC_CONFIG2			0x0020
+#define GMAC_CONFIG3			0x0024
+#define GMAC_RESERVED			0x0028
+#define GMAC_STATUS			0x002c
+#define GMAC_IN_DISCARDS		0x0030
+#define GMAC_IN_ERRORS			0x0034
+#define GMAC_IN_MCAST			0x0038
+#define GMAC_IN_BCAST			0x003c
+#define GMAC_IN_MAC1			0x0040	/* for STA 1 MAC Address */
+#define GMAC_IN_MAC2			0x0044	/* for STA 2 MAC Address */
+
+#define RX_STATS_NUM	6
+
+/* DMA Queues description Ring Base Address/Size Register (offset 0x0004) */
+union dma_q_base_size {
+	unsigned int bits32;
+	unsigned int base_size;
+};
+
+#define DMA_Q_BASE_MASK		(~0x0f)
+
+/* DMA SKB Buffer register (offset 0x0008) */
+union dma_skb_size {
+	unsigned int bits32;
+	struct bit_0008 {
+		unsigned int sw_skb_size : 16;	/* SW Free poll SKB Size */
+		unsigned int hw_skb_size : 16;	/* HW Free poll SKB Size */
+	} bits;
+};
+
+/* DMA SW Free Queue Read/Write Pointer Register (offset 0x000c) */
+union dma_rwptr {
+	unsigned int bits32;
+	struct bit_000c {
+		unsigned int rptr	: 16;	/* Read Ptr, RO */
+		unsigned int wptr	: 16;	/* Write Ptr, RW */
+	} bits;
+};
+
+/* Interrupt Status Register 0	(offset 0x0020)
+ * Interrupt Mask Register 0	(offset 0x0024)
+ * Interrupt Select Register 0	(offset 0x0028)
+ */
+#define GMAC1_TXDERR_INT_BIT		BIT(31)
+#define GMAC1_TXPERR_INT_BIT		BIT(30)
+#define GMAC0_TXDERR_INT_BIT		BIT(29)
+#define GMAC0_TXPERR_INT_BIT		BIT(28)
+#define GMAC1_RXDERR_INT_BIT		BIT(27)
+#define GMAC1_RXPERR_INT_BIT		BIT(26)
+#define GMAC0_RXDERR_INT_BIT		BIT(25)
+#define GMAC0_RXPERR_INT_BIT		BIT(24)
+#define GMAC1_SWTQ15_FIN_INT_BIT	BIT(23)
+#define GMAC1_SWTQ14_FIN_INT_BIT	BIT(22)
+#define GMAC1_SWTQ13_FIN_INT_BIT	BIT(21)
+#define GMAC1_SWTQ12_FIN_INT_BIT	BIT(20)
+#define GMAC1_SWTQ11_FIN_INT_BIT	BIT(19)
+#define GMAC1_SWTQ10_FIN_INT_BIT	BIT(18)
+#define GMAC0_SWTQ05_FIN_INT_BIT	BIT(17)
+#define GMAC0_SWTQ04_FIN_INT_BIT	BIT(16)
+#define GMAC0_SWTQ03_FIN_INT_BIT	BIT(15)
+#define GMAC0_SWTQ02_FIN_INT_BIT	BIT(14)
+#define GMAC0_SWTQ01_FIN_INT_BIT	BIT(13)
+#define GMAC0_SWTQ00_FIN_INT_BIT	BIT(12)
+#define GMAC1_SWTQ15_EOF_INT_BIT	BIT(11)
+#define GMAC1_SWTQ14_EOF_INT_BIT	BIT(10)
+#define GMAC1_SWTQ13_EOF_INT_BIT	BIT(9)
+#define GMAC1_SWTQ12_EOF_INT_BIT	BIT(8)
+#define GMAC1_SWTQ11_EOF_INT_BIT	BIT(7)
+#define GMAC1_SWTQ10_EOF_INT_BIT	BIT(6)
+#define GMAC0_SWTQ05_EOF_INT_BIT	BIT(5)
+#define GMAC0_SWTQ04_EOF_INT_BIT	BIT(4)
+#define GMAC0_SWTQ03_EOF_INT_BIT	BIT(3)
+#define GMAC0_SWTQ02_EOF_INT_BIT	BIT(2)
+#define GMAC0_SWTQ01_EOF_INT_BIT	BIT(1)
+#define GMAC0_SWTQ00_EOF_INT_BIT	BIT(0)
+
+/* Interrupt Status Register 1	(offset 0x0030)
+ * Interrupt Mask Register 1	(offset 0x0034)
+ * Interrupt Select Register 1	(offset 0x0038)
+ */
+#define TOE_IQ3_FULL_INT_BIT		BIT(31)
+#define TOE_IQ2_FULL_INT_BIT		BIT(30)
+#define TOE_IQ1_FULL_INT_BIT		BIT(29)
+#define TOE_IQ0_FULL_INT_BIT		BIT(28)
+#define TOE_IQ3_INT_BIT			BIT(27)
+#define TOE_IQ2_INT_BIT			BIT(26)
+#define TOE_IQ1_INT_BIT			BIT(25)
+#define TOE_IQ0_INT_BIT			BIT(24)
+#define GMAC1_HWTQ13_EOF_INT_BIT	BIT(23)
+#define GMAC1_HWTQ12_EOF_INT_BIT	BIT(22)
+#define GMAC1_HWTQ11_EOF_INT_BIT	BIT(21)
+#define GMAC1_HWTQ10_EOF_INT_BIT	BIT(20)
+#define GMAC0_HWTQ03_EOF_INT_BIT	BIT(19)
+#define GMAC0_HWTQ02_EOF_INT_BIT	BIT(18)
+#define GMAC0_HWTQ01_EOF_INT_BIT	BIT(17)
+#define GMAC0_HWTQ00_EOF_INT_BIT	BIT(16)
+#define CLASS_RX_INT_BIT(x)		BIT((x + 2))
+#define DEFAULT_Q1_INT_BIT		BIT(1)
+#define DEFAULT_Q0_INT_BIT		BIT(0)
+
+#define TOE_IQ_INT_BITS		(TOE_IQ0_INT_BIT | TOE_IQ1_INT_BIT | \
+				 TOE_IQ2_INT_BIT | TOE_IQ3_INT_BIT)
+#define	TOE_IQ_FULL_BITS	(TOE_IQ0_FULL_INT_BIT | TOE_IQ1_FULL_INT_BIT | \
+				 TOE_IQ2_FULL_INT_BIT | TOE_IQ3_FULL_INT_BIT)
+#define	TOE_IQ_ALL_BITS		(TOE_IQ_INT_BITS | TOE_IQ_FULL_BITS)
+#define TOE_CLASS_RX_INT_BITS	0xfffc
+
+/* Interrupt Status Register 2	(offset 0x0040)
+ * Interrupt Mask Register 2	(offset 0x0044)
+ * Interrupt Select Register 2	(offset 0x0048)
+ */
+#define TOE_QL_FULL_INT_BIT(x)		BIT(x)
+
+/* Interrupt Status Register 3	(offset 0x0050)
+ * Interrupt Mask Register 3	(offset 0x0054)
+ * Interrupt Select Register 3	(offset 0x0058)
+ */
+#define TOE_QH_FULL_INT_BIT(x)		BIT(x - 32)
+
+/* Interrupt Status Register 4	(offset 0x0060)
+ * Interrupt Mask Register 4	(offset 0x0064)
+ * Interrupt Select Register 4	(offset 0x0068)
+ */
+#define GMAC1_RESERVED_INT_BIT		BIT(31)
+#define GMAC1_MIB_INT_BIT		BIT(30)
+#define GMAC1_RX_PAUSE_ON_INT_BIT	BIT(29)
+#define GMAC1_TX_PAUSE_ON_INT_BIT	BIT(28)
+#define GMAC1_RX_PAUSE_OFF_INT_BIT	BIT(27)
+#define GMAC1_TX_PAUSE_OFF_INT_BIT	BIT(26)
+#define GMAC1_RX_OVERRUN_INT_BIT	BIT(25)
+#define GMAC1_STATUS_CHANGE_INT_BIT	BIT(24)
+#define GMAC0_RESERVED_INT_BIT		BIT(23)
+#define GMAC0_MIB_INT_BIT		BIT(22)
+#define GMAC0_RX_PAUSE_ON_INT_BIT	BIT(21)
+#define GMAC0_TX_PAUSE_ON_INT_BIT	BIT(20)
+#define GMAC0_RX_PAUSE_OFF_INT_BIT	BIT(19)
+#define GMAC0_TX_PAUSE_OFF_INT_BIT	BIT(18)
+#define GMAC0_RX_OVERRUN_INT_BIT	BIT(17)
+#define GMAC0_STATUS_CHANGE_INT_BIT	BIT(16)
+#define CLASS_RX_FULL_INT_BIT(x)	BIT(x + 2)
+#define HWFQ_EMPTY_INT_BIT		BIT(1)
+#define SWFQ_EMPTY_INT_BIT		BIT(0)
+
+#define GMAC0_INT_BITS	(GMAC0_RESERVED_INT_BIT | GMAC0_MIB_INT_BIT | \
+			 GMAC0_RX_PAUSE_ON_INT_BIT | \
+			 GMAC0_TX_PAUSE_ON_INT_BIT | \
+			 GMAC0_RX_PAUSE_OFF_INT_BIT | \
+			 GMAC0_TX_PAUSE_OFF_INT_BIT | \
+			 GMAC0_RX_OVERRUN_INT_BIT | \
+			 GMAC0_STATUS_CHANGE_INT_BIT)
+#define GMAC1_INT_BITS	(GMAC1_RESERVED_INT_BIT | GMAC1_MIB_INT_BIT | \
+			 GMAC1_RX_PAUSE_ON_INT_BIT | \
+			 GMAC1_TX_PAUSE_ON_INT_BIT | \
+			 GMAC1_RX_PAUSE_OFF_INT_BIT | \
+			 GMAC1_TX_PAUSE_OFF_INT_BIT | \
+			 GMAC1_RX_OVERRUN_INT_BIT | \
+			 GMAC1_STATUS_CHANGE_INT_BIT)
+
+#define CLASS_RX_FULL_INT_BITS		0xfffc
+
+/* GLOBAL_QUEUE_THRESHOLD_REG	(offset 0x0070) */
+union queue_threshold {
+	unsigned int bits32;
+	struct bit_0070_2 {
+		/*  7:0 Software Free Queue Empty Threshold */
+		unsigned int swfq_empty:8;
+		/* 15:8 Hardware Free Queue Empty Threshold */
+		unsigned int hwfq_empty:8;
+		/* 23:16 */
+		unsigned int intrq:8;
+		/* 31:24 */
+		unsigned int toe_class:8;
+	} bits;
+};
+
+/* GMAC DMA Control Register
+ * GMAC0 offset 0x8000
+ * GMAC1 offset 0xC000
+ */
+union gmac_dma_ctrl {
+	unsigned int bits32;
+	struct bit_8000 {
+		/* bit 1:0 Peripheral Bus Width */
+		unsigned int td_bus:2;
+		/* bit 3:2 TxDMA max burst size for every AHB request */
+		unsigned int td_burst_size:2;
+		/* bit 7:4 TxDMA protection control */
+		unsigned int td_prot:4;
+		/* bit 9:8 Peripheral Bus Width */
+		unsigned int rd_bus:2;
+		/* bit 11:10 DMA max burst size for every AHB request */
+		unsigned int rd_burst_size:2;
+		/* bit 15:12 DMA Protection Control */
+		unsigned int rd_prot:4;
+		/* bit 17:16 */
+		unsigned int rd_insert_bytes:2;
+		/* bit 27:18 */
+		unsigned int reserved:10;
+		/* bit 28 1: Drop, 0: Accept */
+		unsigned int drop_small_ack:1;
+		/* bit 29 Loopback TxDMA to RxDMA */
+		unsigned int loopback:1;
+		/* bit 30 Tx DMA Enable */
+		unsigned int td_enable:1;
+		/* bit 31 Rx DMA Enable */
+		unsigned int rd_enable:1;
+	} bits;
+};
+
+/* GMAC Tx Weighting Control Register 0
+ * GMAC0 offset 0x8004
+ * GMAC1 offset 0xC004
+ */
+union gmac_tx_wcr0 {
+	unsigned int bits32;
+	struct bit_8004 {
+		/* bit 5:0 HW TX Queue 3 */
+		unsigned int hw_tq0:6;
+		/* bit 11:6 HW TX Queue 2 */
+		unsigned int hw_tq1:6;
+		/* bit 17:12 HW TX Queue 1 */
+		unsigned int hw_tq2:6;
+		/* bit 23:18 HW TX Queue 0 */
+		unsigned int hw_tq3:6;
+		/* bit 31:24 */
+		unsigned int reserved:8;
+	} bits;
+};
+
+/* GMAC Tx Weighting Control Register 1
+ * GMAC0 offset 0x8008
+ * GMAC1 offset 0xC008
+ */
+union gmac_tx_wcr1 {
+	unsigned int bits32;
+	struct bit_8008 {
+		/* bit 4:0 SW TX Queue 0 */
+		unsigned int sw_tq0:5;
+		/* bit 9:5 SW TX Queue 1 */
+		unsigned int sw_tq1:5;
+		/* bit 14:10 SW TX Queue 2 */
+		unsigned int sw_tq2:5;
+		/* bit 19:15 SW TX Queue 3 */
+		unsigned int sw_tq3:5;
+		/* bit 24:20 SW TX Queue 4 */
+		unsigned int sw_tq4:5;
+		/* bit 29:25 SW TX Queue 5 */
+		unsigned int sw_tq5:5;
+		/* bit 31:30 */
+		unsigned int reserved:2;
+	} bits;
+};
+
+/* GMAC DMA Tx Description Word 0 Register
+ * GMAC0 offset 0x8040
+ * GMAC1 offset 0xC040
+ */
+union gmac_txdesc_0 {
+	unsigned int bits32;
+	struct bit_8040 {
+		/* bit 15:0 Transfer size */
+		unsigned int buffer_size:16;
+		/* bit 21:16 number of descriptors used for the current frame */
+		unsigned int desc_count:6;
+		/* bit 22 Tx Status, 1: Successful 0: Failed */
+		unsigned int status_tx_ok:1;
+		/* bit 28:23 Tx Status, Reserved bits */
+		unsigned int status_rvd:6;
+		/* bit 29 protocol error during processing this descriptor */
+		unsigned int perr:1;
+		/* bit 30 data error during processing this descriptor */
+		unsigned int derr:1;
+		/* bit 31 */
+		unsigned int reserved:1;
+	} bits;
+};
+
+/* GMAC DMA Tx Description Word 1 Register
+ * GMAC0 offset 0x8044
+ * GMAC1 offset 0xC044
+ */
+union gmac_txdesc_1 {
+	unsigned int bits32;
+	struct txdesc_word1 {
+		/* bit 15: 0 Tx Frame Byte Count */
+		unsigned int byte_count:16;
+		/* bit 16 TSS segmentation use MTU setting */
+		unsigned int mtu_enable:1;
+		/* bit 17 IPV4 Header Checksum Enable */
+		unsigned int ip_chksum:1;
+		/* bit 18 IPV6 Tx Enable */
+		unsigned int ipv6_enable:1;
+		/* bit 19 TCP Checksum Enable */
+		unsigned int tcp_chksum:1;
+		/* bit 20 UDP Checksum Enable */
+		unsigned int udp_chksum:1;
+		/* bit 21 Bypass HW offload engine */
+		unsigned int bypass_tss:1;
+		/* bit 22 Don't update IP length field */
+		unsigned int ip_fixed_len:1;
+		/* bit 31:23 Tx Flag, Reserved */
+		unsigned int reserved:9;
+	} bits;
+};
+
+#define TSS_IP_FIXED_LEN_BIT	BIT(22)
+#define TSS_BYPASS_BIT		BIT(21)
+#define TSS_UDP_CHKSUM_BIT	BIT(20)
+#define TSS_TCP_CHKSUM_BIT	BIT(19)
+#define TSS_IPV6_ENABLE_BIT	BIT(18)
+#define TSS_IP_CHKSUM_BIT	BIT(17)
+#define TSS_MTU_ENABLE_BIT	BIT(16)
+
+#define TSS_CHECKUM_ENABLE	\
+	(TSS_IP_CHKSUM_BIT | TSS_IPV6_ENABLE_BIT | \
+	 TSS_TCP_CHKSUM_BIT | TSS_UDP_CHKSUM_BIT)
+
+/* GMAC DMA Tx Description Word 2 Register
+ * GMAC0 offset 0x8048
+ * GMAC1 offset 0xC048
+ */
+union gmac_txdesc_2 {
+	unsigned int	bits32;
+	unsigned int	buf_adr;
+};
+
+/* GMAC DMA Tx Description Word 3 Register
+ * GMAC0 offset 0x804C
+ * GMAC1 offset 0xC04C
+ */
+union gmac_txdesc_3 {
+	unsigned int bits32;
+	struct txdesc_word3 {
+		/* bit 12: 0 Tx Frame Byte Count */
+		unsigned int mtu_size:13;
+		/* bit 28:13 */
+		unsigned int reserved:16;
+		/* bit 29 End of frame interrupt enable */
+		unsigned int eofie:1;
+		/* bit 31:30 11: only one, 10: first, 01: last, 00: linking */
+		unsigned int sof_eof:2;
+	} bits;
+};
+
+#define SOF_EOF_BIT_MASK	0x3fffffff
+#define SOF_BIT			0x80000000
+#define EOF_BIT			0x40000000
+#define EOFIE_BIT		BIT(29)
+#define MTU_SIZE_BIT_MASK	0x7ff /* Max MTU 2047 bytes */
+
+/* GMAC Tx Descriptor */
+struct gmac_txdesc {
+	union gmac_txdesc_0 word0;
+	union gmac_txdesc_1 word1;
+	union gmac_txdesc_2 word2;
+	union gmac_txdesc_3 word3;
+};
+
+/* GMAC DMA Rx Description Word 0 Register
+ * GMAC0 offset 0x8060
+ * GMAC1 offset 0xC060
+ */
+union gmac_rxdesc_0 {
+	unsigned int bits32;
+	struct bit_8060 {
+		/* bit 15:0 number of descriptors used for the current frame */
+		unsigned int buffer_size:16;
+		/* bit 21:16 number of descriptors used for the current frame */
+		unsigned int desc_count:6;
+		/* bit 24:22 Status of rx frame */
+		unsigned int status:4;
+		/* bit 28:26 Check Sum Status */
+		unsigned int chksum_status:3;
+		/* bit 29 protocol error during processing this descriptor */
+		unsigned int perr:1;
+		/* bit 30 data error during processing this descriptor */
+		unsigned int derr:1;
+		/* bit 31 TOE/CIS Queue Full dropped packet to default queue */
+		unsigned int drop:1;
+	} bits;
+};
+
+#define	GMAC_RXDESC_0_T_derr			BIT(30)
+#define	GMAC_RXDESC_0_T_perr			BIT(29)
+#define	GMAC_RXDESC_0_T_chksum_status(x)	BIT(x + 26)
+#define	GMAC_RXDESC_0_T_status(x)		BIT(x + 22)
+#define	GMAC_RXDESC_0_T_desc_count(x)		BIT(x + 16)
+
+#define	RX_CHKSUM_IP_UDP_TCP_OK			0
+#define	RX_CHKSUM_IP_OK_ONLY			1
+#define	RX_CHKSUM_NONE				2
+#define	RX_CHKSUM_IP_ERR_UNKNOWN		4
+#define	RX_CHKSUM_IP_ERR			5
+#define	RX_CHKSUM_TCP_UDP_ERR			6
+#define RX_CHKSUM_NUM				8
+
+#define RX_STATUS_GOOD_FRAME			0
+#define RX_STATUS_TOO_LONG_GOOD_CRC		1
+#define RX_STATUS_RUNT_FRAME			2
+#define RX_STATUS_SFD_NOT_FOUND			3
+#define RX_STATUS_CRC_ERROR			4
+#define RX_STATUS_TOO_LONG_BAD_CRC		5
+#define RX_STATUS_ALIGNMENT_ERROR		6
+#define RX_STATUS_TOO_LONG_BAD_ALIGN		7
+#define RX_STATUS_RX_ERR			8
+#define RX_STATUS_DA_FILTERED			9
+#define RX_STATUS_BUFFER_FULL			10
+#define RX_STATUS_NUM				16
+
+#define RX_ERROR_LENGTH(s) \
+	((s) == RX_STATUS_TOO_LONG_GOOD_CRC || \
+	 (s) == RX_STATUS_TOO_LONG_BAD_CRC || \
+	 (s) == RX_STATUS_TOO_LONG_BAD_ALIGN)
+#define RX_ERROR_OVER(s) \
+	((s) == RX_STATUS_BUFFER_FULL)
+#define RX_ERROR_CRC(s) \
+	((s) == RX_STATUS_CRC_ERROR || \
+	 (s) == RX_STATUS_TOO_LONG_BAD_CRC)
+#define RX_ERROR_FRAME(s) \
+	((s) == RX_STATUS_ALIGNMENT_ERROR || \
+	 (s) == RX_STATUS_TOO_LONG_BAD_ALIGN)
+#define RX_ERROR_FIFO(s) \
+	(0)
+
+/* GMAC DMA Rx Description Word 1 Register
+ * GMAC0 offset 0x8064
+ * GMAC1 offset 0xC064
+ */
+union gmac_rxdesc_1 {
+	unsigned int bits32;
+	struct rxdesc_word1 {
+		/* bit 15: 0 Rx Frame Byte Count */
+		unsigned int byte_count:16;
+		/* bit 31:16 Software ID */
+		unsigned int sw_id:16;
+	} bits;
+};
+
+/* GMAC DMA Rx Description Word 2 Register
+ * GMAC0 offset 0x8068
+ * GMAC1 offset 0xC068
+ */
+union gmac_rxdesc_2 {
+	unsigned int	bits32;
+	unsigned int	buf_adr;
+};
+
+#define RX_INSERT_NONE		0
+#define RX_INSERT_1_BYTE	1
+#define RX_INSERT_2_BYTE	2
+#define RX_INSERT_3_BYTE	3
+
+/* GMAC DMA Rx Description Word 3 Register
+ * GMAC0 offset 0x806C
+ * GMAC1 offset 0xC06C
+ */
+union gmac_rxdesc_3 {
+	unsigned int bits32;
+	struct rxdesc_word3 {
+		/* bit 7: 0 L3 data offset */
+		unsigned int l3_offset:8;
+		/* bit 15: 8 L4 data offset */
+		unsigned int l4_offset:8;
+		/* bit 23: 16 L7 data offset */
+		unsigned int l7_offset:8;
+		/* bit 24 Duplicated ACK detected */
+		unsigned int dup_ack:1;
+		/* bit 25 abnormal case found */
+		unsigned int abnormal:1;
+		/* bit 26 IPV4 option or IPV6 extension header */
+		unsigned int option:1;
+		/* bit 27 Out of Sequence packet */
+		unsigned int out_of_seq:1;
+		/* bit 28 Control Flag is present */
+		unsigned int ctrl_flag:1;
+		/* bit 29 End of frame interrupt enable */
+		unsigned int eofie:1;
+		/* bit 31:30 11: only one, 10: first, 01: last, 00: linking */
+		unsigned int sof_eof:2;
+	} bits;
+};
+
+/* GMAC Rx Descriptor, this is simply fitted over the queue registers */
+struct gmac_rxdesc {
+	union gmac_rxdesc_0 word0;
+	union gmac_rxdesc_1 word1;
+	union gmac_rxdesc_2 word2;
+	union gmac_rxdesc_3 word3;
+};
+
+/* GMAC Matching Rule Control Register 0
+ * GMAC0 offset 0x8078
+ * GMAC1 offset 0xC078
+ */
+#define MR_L2_BIT		BIT(31)
+#define MR_L3_BIT		BIT(30)
+#define MR_L4_BIT		BIT(29)
+#define MR_L7_BIT		BIT(28)
+#define MR_PORT_BIT		BIT(27)
+#define MR_PRIORITY_BIT		BIT(26)
+#define MR_DA_BIT		BIT(23)
+#define MR_SA_BIT		BIT(22)
+#define MR_ETHER_TYPE_BIT	BIT(21)
+#define MR_VLAN_BIT		BIT(20)
+#define MR_PPPOE_BIT		BIT(19)
+#define MR_IP_VER_BIT		BIT(15)
+#define MR_IP_HDR_LEN_BIT	BIT(14)
+#define MR_FLOW_LABLE_BIT	BIT(13)
+#define MR_TOS_TRAFFIC_BIT	BIT(12)
+#define MR_SPR_BIT(x)		BIT(x)
+#define MR_SPR_BITS		0xff
+
+/* GMAC_AHB_WEIGHT registers
+ * GMAC0 offset 0x80C8
+ * GMAC1 offset 0xC0C8
+ */
+union gmac_ahb_weight {
+	unsigned int bits32;
+	struct bit_80C8 {
+		/* 4:0 */
+		unsigned int hash_weight:5;
+		/* 9:5 */
+		unsigned int rx_weight:5;
+		/* 14:10 */
+		unsigned int tx_weight:5;
+		/* 19:15 Rx Data Pre Request FIFO Threshold */
+		unsigned int pre_req:5;
+		/* 24:20 DMA TqCtrl to Start tqDV FIFO Threshold */
+		unsigned int tq_dv_threshold:5;
+		/* 31:25 */
+		unsigned int reserved:7;
+	} bits;
+};
+
+/* GMAC RX FLTR
+ * GMAC0 Offset 0xA00C
+ * GMAC1 Offset 0xE00C
+ */
+union gmac_rx_fltr {
+	unsigned int bits32;
+	struct bit1_000c {
+		/* Enable receive of unicast frames that are sent to STA
+		 * address
+		 */
+		unsigned int unicast:1;
+		/* Enable receive of multicast frames that pass multicast
+		 * filter
+		 */
+		unsigned int multicast:1;
+		/* Enable receive of broadcast frames */
+		unsigned int broadcast:1;
+		/* Enable receive of all frames */
+		unsigned int promiscuous:1;
+		/* Enable receive of all error frames */
+		unsigned int error:1;
+		unsigned int reserved:27;
+	} bits;
+};
+
+/* GMAC Configuration 0
+ * GMAC0 Offset 0xA018
+ * GMAC1 Offset 0xE018
+ */
+union gmac_config0 {
+	unsigned int bits32;
+	struct bit1_0018 {
+		/* 0: disable transmit */
+		unsigned int dis_tx:1;
+		/* 1: disable receive */
+		unsigned int dis_rx:1;
+		/* 2: transmit data loopback enable */
+		unsigned int loop_back:1;
+		/* 3: flow control also trigged by Rx queues */
+		unsigned int flow_ctrl:1;
+		/* 4-7: adjust IFG from 96+/-56 */
+		unsigned int adj_ifg:4;
+		/* 8-10 maximum receive frame length allowed */
+		unsigned int max_len:3;
+		/* 11: disable back-off function */
+		unsigned int dis_bkoff:1;
+		/* 12: disable 16 collisions abort function */
+		unsigned int dis_col:1;
+		/* 13: speed up timers in simulation */
+		unsigned int sim_test:1;
+		/* 14: RX flow control enable */
+		unsigned int rx_fc_en:1;
+		/* 15: TX flow control enable */
+		unsigned int tx_fc_en:1;
+		/* 16: RGMII in-band status enable */
+		unsigned int rgmii_en:1;
+		/* 17: IPv4 RX Checksum enable */
+		unsigned int ipv4_rx_chksum:1;
+		/* 18: IPv6 RX Checksum enable */
+		unsigned int ipv6_rx_chksum:1;
+		/* 19: Remove Rx VLAN tag */
+		unsigned int rx_tag_remove:1;
+		/* 20 */
+		unsigned int rgmm_edge:1;
+		/* 21 */
+		unsigned int rxc_inv:1;
+		/* 22 */
+		unsigned int ipv6_exthdr_order:1;
+		/* 23 */
+		unsigned int rx_err_detect:1;
+		/* 24 */
+		unsigned int port0_chk_hwq:1;
+		/* 25 */
+		unsigned int port1_chk_hwq:1;
+		/* 26 */
+		unsigned int port0_chk_toeq:1;
+		/* 27 */
+		unsigned int port1_chk_toeq:1;
+		/* 28 */
+		unsigned int port0_chk_classq:1;
+		/* 29 */
+		unsigned int port1_chk_classq:1;
+		/* 30, 31 */
+		unsigned int reserved:2;
+	} bits;
+};
+
+#define CONFIG0_TX_RX_DISABLE	(BIT(1) | BIT(0))
+#define CONFIG0_RX_CHKSUM	(BIT(18) | BIT(17))
+#define CONFIG0_FLOW_RX		BIT(14)
+#define CONFIG0_FLOW_TX		BIT(15)
+#define CONFIG0_FLOW_TX_RX	(BIT(14) | BIT(15))
+#define CONFIG0_FLOW_CTL	(BIT(14) | BIT(15))
+
+#define CONFIG0_MAXLEN_SHIFT	8
+#define CONFIG0_MAXLEN_MASK	(7 << CONFIG0_MAXLEN_SHIFT)
+#define  CONFIG0_MAXLEN_1536	0
+#define  CONFIG0_MAXLEN_1518	1
+#define  CONFIG0_MAXLEN_1522	2
+#define  CONFIG0_MAXLEN_1548	3
+#define  CONFIG0_MAXLEN_9k	4	/* 9212 */
+#define  CONFIG0_MAXLEN_10k	5	/* 10236 */
+#define  CONFIG0_MAXLEN_1518__6	6
+#define  CONFIG0_MAXLEN_1518__7	7
+
+/* GMAC Configuration 1
+ * GMAC0 Offset 0xA01C
+ * GMAC1 Offset 0xE01C
+ */
+union gmac_config1 {
+	unsigned int bits32;
+	struct bit1_001c {
+		/* Flow control set threshold */
+		unsigned int set_threshold:8;
+		/* Flow control release threshold */
+		unsigned int rel_threshold:8;
+		unsigned int reserved:16;
+	} bits;
+};
+
+#define GMAC_FLOWCTRL_SET_MAX		32
+#define GMAC_FLOWCTRL_SET_MIN		0
+#define GMAC_FLOWCTRL_RELEASE_MAX	32
+#define GMAC_FLOWCTRL_RELEASE_MIN	0
+
+/* GMAC Configuration 2
+ * GMAC0 Offset 0xA020
+ * GMAC1 Offset 0xE020
+ */
+union gmac_config2 {
+	unsigned int bits32;
+	struct bit1_0020 {
+		/* Flow control set threshold */
+		unsigned int set_threshold:16;
+		/* Flow control release threshold */
+		unsigned int rel_threshold:16;
+	} bits;
+};
+
+/* GMAC Configuration 3
+ * GMAC0 Offset 0xA024
+ * GMAC1 Offset 0xE024
+ */
+union gmac_config3 {
+	unsigned int bits32;
+	struct bit1_0024 {
+		/* Flow control set threshold */
+		unsigned int set_threshold:16;
+		/* Flow control release threshold */
+		unsigned int rel_threshold:16;
+	} bits;
+};
+
+/* GMAC STATUS
+ * GMAC0 Offset 0xA02C
+ * GMAC1 Offset 0xE02C
+ */
+union gmac_status {
+	unsigned int bits32;
+	struct bit1_002c {
+		/* Link status */
+		unsigned int link:1;
+		/* Link speed(00->2.5M 01->25M 10->125M) */
+		unsigned int speed:2;
+		/* Duplex mode */
+		unsigned int duplex:1;
+		unsigned int reserved_1:1;
+		/* PHY interface type */
+		unsigned int mii_rmii:2;
+		unsigned int reserved_2:25;
+	} bits;
+};
+
+#define GMAC_SPEED_10			0
+#define GMAC_SPEED_100			1
+#define GMAC_SPEED_1000			2
+
+#define GMAC_PHY_MII			0
+#define GMAC_PHY_GMII			1
+#define GMAC_PHY_RGMII_100_10		2
+#define GMAC_PHY_RGMII_1000		3
+
+/* Queue Header
+ *	(1) TOE Queue Header
+ *	(2) Non-TOE Queue Header
+ *	(3) Interrupt Queue Header
+ *
+ * memory Layout
+ *	TOE Queue Header
+ *		     0x60003000 +---------------------------+ 0x0000
+ *				|     TOE Queue 0 Header    |
+ *				|         8 * 4 Bytes	    |
+ *				+---------------------------+ 0x0020
+ *				|     TOE Queue 1 Header    |
+ *				|         8 * 4 Bytes	    |
+ *				+---------------------------+ 0x0040
+ *				|          ......           |
+ *				|                           |
+ *				+---------------------------+
+ *
+ *	Non TOE Queue Header
+ *		     0x60002000 +---------------------------+ 0x0000
+ *				|   Default Queue 0 Header  |
+ *				|         2 * 4 Bytes       |
+ *				+---------------------------+ 0x0008
+ *				|   Default Queue 1 Header  |
+ *				|         2 * 4 Bytes       |
+ *				+---------------------------+ 0x0010
+ *				|   Classification Queue 0  |
+ *				|	  2 * 4 Bytes       |
+ *				+---------------------------+
+ *				|   Classification Queue 1  |
+ *				|	  2 * 4 Bytes       |
+ *				+---------------------------+ (n * 8 + 0x10)
+ *				|		...	    |
+ *				|	  2 * 4 Bytes	    |
+ *				+---------------------------+ (13 * 8 + 0x10)
+ *				|   Classification Queue 13 |
+ *				|	  2 * 4 Bytes	    |
+ *				+---------------------------+ 0x80
+ *				|      Interrupt Queue 0    |
+ *				|	  2 * 4 Bytes	    |
+ *				+---------------------------+
+ *				|      Interrupt Queue 1    |
+ *				|	  2 * 4 Bytes	    |
+ *				+---------------------------+
+ *				|      Interrupt Queue 2    |
+ *				|	  2 * 4 Bytes	    |
+ *				+---------------------------+
+ *				|      Interrupt Queue 3    |
+ *				|	  2 * 4 Bytes	    |
+ *				+---------------------------+
+ *
+ */
+#define TOE_QUEUE_HDR_ADDR(n)	(TOE_TOE_QUE_HDR_BASE + n * 32)
+#define TOE_Q_HDR_AREA_END	(TOE_QUEUE_HDR_ADDR(TOE_TOE_QUEUE_MAX + 1))
+#define TOE_DEFAULT_Q_HDR_BASE(x) (TOE_NONTOE_QUE_HDR_BASE + 0x08 * (x))
+#define TOE_CLASS_Q_HDR_BASE	(TOE_NONTOE_QUE_HDR_BASE + 0x10)
+#define TOE_INTR_Q_HDR_BASE	(TOE_NONTOE_QUE_HDR_BASE + 0x80)
+#define INTERRUPT_QUEUE_HDR_ADDR(n) (TOE_INTR_Q_HDR_BASE + n * 8)
+#define NONTOE_Q_HDR_AREA_END (INTERRUPT_QUEUE_HDR_ADDR(TOE_INTR_QUEUE_MAX + 1))
+
+/* NONTOE Queue Header Word 0 */
+union nontoe_qhdr0 {
+	unsigned int bits32;
+	unsigned int base_size;
+};
+
+#define NONTOE_QHDR0_BASE_MASK	(~0x0f)
+
+/* NONTOE Queue Header Word 1 */
+union nontoe_qhdr1 {
+	unsigned int bits32;
+	struct bit_nonqhdr1 {
+		/* bit 15:0 */
+		unsigned int rptr:16;
+		/* bit 31:16 */
+		unsigned int wptr:16;
+	} bits;
+};
+
+/* Non-TOE Queue Header */
+struct nontoe_qhdr {
+	union nontoe_qhdr0 word0;
+	union nontoe_qhdr1 word1;
+};
+
+#endif /* _GEMINI_ETHERNET_H */
diff --git a/drivers/net/ethernet/davicom/Kconfig b/drivers/net/ethernet/davicom/Kconfig
index 7ec2d74f94d3..02e0caff98e3 100644
--- a/drivers/net/ethernet/davicom/Kconfig
+++ b/drivers/net/ethernet/davicom/Kconfig
@@ -1,13 +1,27 @@
+# SPDX-License-Identifier: GPL-2.0-only
 #
 # Davicom device configuration
 #
 
+config NET_VENDOR_DAVICOM
+	bool "Davicom devices"
+	default y
+	help
+	  If you have a network (Ethernet) card belonging to this class, say Y.
+
+	  Note that the answer to this question doesn't directly affect the
+	  kernel: saying N will just cause the configurator to skip all
+	  the questions about Davicom devices. If you say Y, you will be asked
+	  for your specific card in the following selections.
+
+if NET_VENDOR_DAVICOM
+
 config DM9000
 	tristate "DM9000 support"
-	depends on ARM || BLACKFIN || MIPS || COLDFIRE || NIOS2
+	depends on ARM || MIPS || COLDFIRE || NIOS2 || COMPILE_TEST
 	select CRC32
 	select MII
-	---help---
+	help
 	  Support for DM9000 chipset.
 
 	  To compile this driver as a module, choose M here.  The module
@@ -16,8 +30,26 @@ config DM9000
 config DM9000_FORCE_SIMPLE_PHY_POLL
 	bool "Force simple NSR based PHY polling"
 	depends on DM9000
-	---help---
+	help
 	  This configuration forces the DM9000 to use the NSR's LinkStatus
 	  bit to determine if the link is up or down instead of the more
 	  costly MII PHY reads. Note, this will not work if the chip is
 	  operating with an external PHY.
+
+config DM9051
+	tristate "DM9051 SPI support"
+	depends on SPI
+	select CRC32
+	select MDIO
+	select PHYLIB
+	select REGMAP_SPI
+	help
+	  Support for DM9051 SPI chipset.
+
+	  To compile this driver as a module, choose M here.  The module
+	  will be called dm9051.
+
+	  The SPI mode for the host's SPI master to access DM9051 is mode
+	  0 on the SPI bus.
+
+endif # NET_VENDOR_DAVICOM
diff --git a/drivers/net/ethernet/davicom/Makefile b/drivers/net/ethernet/davicom/Makefile
index 74b31f0ebe18..225f85bc1f53 100644
--- a/drivers/net/ethernet/davicom/Makefile
+++ b/drivers/net/ethernet/davicom/Makefile
@@ -1,5 +1,7 @@
+# SPDX-License-Identifier: GPL-2.0-only
 #
 # Makefile for the Davicom device drivers.
 #
 
 obj-$(CONFIG_DM9000) += dm9000.o
+obj-$(CONFIG_DM9051) += dm9051.o
diff --git a/drivers/net/ethernet/davicom/dm9000.c b/drivers/net/ethernet/davicom/dm9000.c
index 008dc8161775..b87eaf0c250c 100644
--- a/drivers/net/ethernet/davicom/dm9000.c
+++ b/drivers/net/ethernet/davicom/dm9000.c
@@ -1,17 +1,8 @@
+// SPDX-License-Identifier: GPL-2.0-or-later
 /*
  *      Davicom DM9000 Fast Ethernet driver for Linux.
  * 	Copyright (C) 1997  Sten Wang
  *
- * 	This program is free software; you can redistribute it and/or
- * 	modify it under the terms of the GNU General Public License
- * 	as published by the Free Software Foundation; either version 2
- * 	of the License, or (at your option) any later version.
- *
- * 	This program is distributed in the hope that it will be useful,
- * 	but WITHOUT ANY WARRANTY; without even the implied warranty of
- * 	MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
- * 	GNU General Public License for more details.
- *
  * (C) Copyright 1997-1998 DAVICOM Semiconductor,Inc. All Rights Reserved.
  *
  * Additional updates, Copyright:
@@ -37,8 +28,7 @@
 #include <linux/irq.h>
 #include <linux/slab.h>
 #include <linux/regulator/consumer.h>
-#include <linux/gpio.h>
-#include <linux/of_gpio.h>
+#include <linux/gpio/consumer.h>
 
 #include <asm/delay.h>
 #include <asm/irq.h>
@@ -51,7 +41,6 @@
 #define DM9000_PHY		0x40	/* PHY address 0x01 */
 
 #define CARDNAME	"dm9000"
-#define DRV_VERSION	"1.31"
 
 /*
  * Transmit timeout, default 5 seconds.
@@ -65,7 +54,7 @@ MODULE_PARM_DESC(watchdog, "transmit timeout in milliseconds");
  */
 static int debug;
 module_param(debug, int, 0644);
-MODULE_PARM_DESC(debug, "dm9000 debug level (0-4)");
+MODULE_PARM_DESC(debug, "dm9000 debug level (0-6)");
 
 /* DM9000 register address locking.
  *
@@ -143,6 +132,8 @@ struct board_info {
 	u32		wake_state;
 
 	int		ip_summed;
+
+	struct regulator *power_supply;
 };
 
 /* debug code */
@@ -242,32 +233,29 @@ static void dm9000_inblk_32bit(void __iomem *reg, void *data, int count)
 static void dm9000_dumpblk_8bit(void __iomem *reg, int count)
 {
 	int i;
-	int tmp;
 
 	for (i = 0; i < count; i++)
-		tmp = readb(reg);
+		readb(reg);
 }
 
 static void dm9000_dumpblk_16bit(void __iomem *reg, int count)
 {
 	int i;
-	int tmp;
 
 	count = (count + 1) >> 1;
 
 	for (i = 0; i < count; i++)
-		tmp = readw(reg);
+		readw(reg);
 }
 
 static void dm9000_dumpblk_32bit(void __iomem *reg, int count)
 {
 	int i;
-	int tmp;
 
 	count = (count + 3) >> 2;
 
 	for (i = 0; i < count; i++)
-		tmp = readl(reg);
+		readl(reg);
 }
 
 /*
@@ -395,6 +383,7 @@ static void dm9000_set_io(struct board_info *db, int byte_width)
 
 	case 3:
 		dev_dbg(db->dev, ": 3 byte IO, falling back to 16bit\n");
+		fallthrough;
 	case 2:
 		db->dumpblk = dm9000_dumpblk_16bit;
 		db->outblk  = dm9000_outblk_16bit;
@@ -550,9 +539,8 @@ static void dm9000_get_drvinfo(struct net_device *dev,
 {
 	struct board_info *dm = to_dm9000_board(dev);
 
-	strlcpy(info->driver, CARDNAME, sizeof(info->driver));
-	strlcpy(info->version, DRV_VERSION, sizeof(info->version));
-	strlcpy(info->bus_info, to_platform_device(dm->dev)->name,
+	strscpy(info->driver, CARDNAME, sizeof(info->driver));
+	strscpy(info->bus_info, to_platform_device(dm->dev)->name,
 		sizeof(info->bus_info));
 }
 
@@ -972,7 +960,7 @@ dm9000_init_dm9000(struct net_device *dev)
 }
 
 /* Our watchdog timed out. Called by the networking layer */
-static void dm9000_timeout(struct net_device *dev)
+static void dm9000_timeout(struct net_device *dev, unsigned int txqueue)
 {
 	struct board_info *db = netdev_priv(dev);
 	u8 reg_save;
@@ -1023,7 +1011,7 @@ static void dm9000_send_packet(struct net_device *dev,
  *  Hardware start transmission.
  *  Send a packet to media from the upper layer.
  */
-static int
+static netdev_tx_t
 dm9000_start_xmit(struct sk_buff *skb, struct net_device *dev)
 {
 	unsigned long flags;
@@ -1171,7 +1159,7 @@ dm9000_rx(struct net_device *dev)
 		if (GoodPacket &&
 		    ((skb = netdev_alloc_skb(dev, RxLen + 4)) != NULL)) {
 			skb_reserve(skb, 2);
-			rdptr = (u8 *) skb_put(skb, RxLen - 4);
+			rdptr = skb_put(skb, RxLen - 4);
 
 			/* Read received packet from RX SRAM */
 
@@ -1383,7 +1371,7 @@ static const struct net_device_ops dm9000_netdev_ops = {
 	.ndo_start_xmit		= dm9000_start_xmit,
 	.ndo_tx_timeout		= dm9000_timeout,
 	.ndo_set_rx_mode	= dm9000_hash_table,
-	.ndo_do_ioctl		= dm9000_ioctl,
+	.ndo_eth_ioctl		= dm9000_ioctl,
 	.ndo_set_features	= dm9000_set_features,
 	.ndo_validate_addr	= eth_validate_addr,
 	.ndo_set_mac_address	= eth_mac_addr,
@@ -1396,7 +1384,7 @@ static struct dm9000_plat_data *dm9000_parse_dt(struct device *dev)
 {
 	struct dm9000_plat_data *pdata;
 	struct device_node *np = dev->of_node;
-	const void *mac_addr;
+	int ret;
 
 	if (!IS_ENABLED(CONFIG_OF) || !np)
 		return ERR_PTR(-ENXIO);
@@ -1405,14 +1393,14 @@ static struct dm9000_plat_data *dm9000_parse_dt(struct device *dev)
 	if (!pdata)
 		return ERR_PTR(-ENOMEM);
 
-	if (of_find_property(np, "davicom,ext-phy", NULL))
+	if (of_property_read_bool(np, "davicom,ext-phy"))
 		pdata->flags |= DM9000_PLATF_EXT_PHY;
-	if (of_find_property(np, "davicom,no-eeprom", NULL))
+	if (of_property_read_bool(np, "davicom,no-eeprom"))
 		pdata->flags |= DM9000_PLATF_NO_EEPROM;
 
-	mac_addr = of_get_mac_address(np);
-	if (mac_addr)
-		memcpy(pdata->dev_addr, mac_addr, sizeof(pdata->dev_addr));
+	ret = of_get_mac_address(np, pdata->dev_addr);
+	if (ret == -EPROBE_DEFER)
+		return ERR_PTR(ret);
 
 	return pdata;
 }
@@ -1432,10 +1420,10 @@ dm9000_probe(struct platform_device *pdev)
 	int iosize;
 	int i;
 	u32 id_val;
-	int reset_gpios;
-	enum of_gpio_flags flags;
+	struct gpio_desc *reset_gpio;
 	struct regulator *power;
 	bool inv_mac_addr = false;
+	u8 addr[ETH_ALEN];
 
 	power = devm_regulator_get(dev, "vcc");
 	if (IS_ERR(power)) {
@@ -1452,34 +1440,42 @@ dm9000_probe(struct platform_device *pdev)
 		dev_dbg(dev, "regulator enabled\n");
 	}
 
-	reset_gpios = of_get_named_gpio_flags(dev->of_node, "reset-gpios", 0,
-					      &flags);
-	if (gpio_is_valid(reset_gpios)) {
-		ret = devm_gpio_request_one(dev, reset_gpios, flags,
-					    "dm9000_reset");
+	reset_gpio = devm_gpiod_get_optional(dev, "reset", GPIOD_OUT_HIGH);
+	ret = PTR_ERR_OR_ZERO(reset_gpio);
+	if (ret) {
+		dev_err(dev, "failed to request reset gpio: %d\n", ret);
+		goto out_regulator_disable;
+	}
+
+	if (reset_gpio) {
+		ret = gpiod_set_consumer_name(reset_gpio, "dm9000_reset");
 		if (ret) {
-			dev_err(dev, "failed to request reset gpio %d: %d\n",
-				reset_gpios, ret);
-			return -ENODEV;
+			dev_err(dev, "failed to set reset gpio name: %d\n",
+				ret);
+			goto out_regulator_disable;
 		}
 
 		/* According to manual PWRST# Low Period Min 1ms */
 		msleep(2);
-		gpio_set_value(reset_gpios, 1);
+		gpiod_set_value_cansleep(reset_gpio, 0);
 		/* Needs 3ms to read eeprom when PWRST is deasserted */
 		msleep(4);
 	}
 
 	if (!pdata) {
 		pdata = dm9000_parse_dt(&pdev->dev);
-		if (IS_ERR(pdata))
-			return PTR_ERR(pdata);
+		if (IS_ERR(pdata)) {
+			ret = PTR_ERR(pdata);
+			goto out_regulator_disable;
+		}
 	}
 
 	/* Init network device */
 	ndev = alloc_etherdev(sizeof(struct board_info));
-	if (!ndev)
-		return -ENOMEM;
+	if (!ndev) {
+		ret = -ENOMEM;
+		goto out_regulator_disable;
+	}
 
 	SET_NETDEV_DEV(ndev, &pdev->dev);
 
@@ -1490,6 +1486,8 @@ dm9000_probe(struct platform_device *pdev)
 
 	db->dev = &pdev->dev;
 	db->ndev = ndev;
+	if (!IS_ERR(power))
+		db->power_supply = power;
 
 	spin_lock_init(&db->lock);
 	mutex_init(&db->addr_lock);
@@ -1508,13 +1506,11 @@ dm9000_probe(struct platform_device *pdev)
 
 	ndev->irq = platform_get_irq(pdev, 0);
 	if (ndev->irq < 0) {
-		dev_err(db->dev, "interrupt resource unavailable: %d\n",
-			ndev->irq);
 		ret = ndev->irq;
 		goto out;
 	}
 
-	db->irq_wake = platform_get_irq(pdev, 1);
+	db->irq_wake = platform_get_irq_optional(pdev, 1);
 	if (db->irq_wake >= 0) {
 		dev_dbg(db->dev, "wakeup irq %d\n", db->irq_wake);
 
@@ -1529,7 +1525,6 @@ dm9000_probe(struct platform_device *pdev)
 			if (ret) {
 				dev_err(db->dev, "irq %d cannot set wakeup (%d)\n",
 					db->irq_wake, ret);
-				ret = 0;
 			} else {
 				irq_set_irq_wake(db->irq_wake, 0);
 				db->wake_supported = 1;
@@ -1674,11 +1669,12 @@ dm9000_probe(struct platform_device *pdev)
 
 	/* try reading the node address from the attached EEPROM */
 	for (i = 0; i < 6; i += 2)
-		dm9000_read_eeprom(db, i / 2, ndev->dev_addr+i);
+		dm9000_read_eeprom(db, i / 2, addr + i);
+	eth_hw_addr_set(ndev, addr);
 
 	if (!is_valid_ether_addr(ndev->dev_addr) && pdata != NULL) {
 		mac_src = "platform data";
-		memcpy(ndev->dev_addr, pdata->dev_addr, ETH_ALEN);
+		eth_hw_addr_set(ndev, pdata->dev_addr);
 	}
 
 	if (!is_valid_ether_addr(ndev->dev_addr)) {
@@ -1686,7 +1682,8 @@ dm9000_probe(struct platform_device *pdev)
 
 		mac_src = "chip";
 		for (i = 0; i < 6; i++)
-			ndev->dev_addr[i] = ior(db, i+DM9000_PAR);
+			addr[i] = ior(db, i + DM9000_PAR);
+		eth_hw_addr_set(ndev, pdata->dev_addr);
 	}
 
 	if (!is_valid_ether_addr(ndev->dev_addr)) {
@@ -1716,14 +1713,17 @@ out:
 	dm9000_release_board(pdev, db);
 	free_netdev(ndev);
 
+out_regulator_disable:
+	if (!IS_ERR(power))
+		regulator_disable(power);
+
 	return ret;
 }
 
 static int
 dm9000_drv_suspend(struct device *dev)
 {
-	struct platform_device *pdev = to_platform_device(dev);
-	struct net_device *ndev = platform_get_drvdata(pdev);
+	struct net_device *ndev = dev_get_drvdata(dev);
 	struct board_info *db;
 
 	if (ndev) {
@@ -1745,8 +1745,7 @@ dm9000_drv_suspend(struct device *dev)
 static int
 dm9000_drv_resume(struct device *dev)
 {
-	struct platform_device *pdev = to_platform_device(dev);
-	struct net_device *ndev = platform_get_drvdata(pdev);
+	struct net_device *ndev = dev_get_drvdata(dev);
 	struct board_info *db = netdev_priv(ndev);
 
 	if (ndev) {
@@ -1771,17 +1770,19 @@ static const struct dev_pm_ops dm9000_drv_pm_ops = {
 	.resume		= dm9000_drv_resume,
 };
 
-static int
-dm9000_drv_remove(struct platform_device *pdev)
+static void dm9000_drv_remove(struct platform_device *pdev)
 {
 	struct net_device *ndev = platform_get_drvdata(pdev);
+	struct board_info *dm = to_dm9000_board(ndev);
 
 	unregister_netdev(ndev);
-	dm9000_release_board(pdev, netdev_priv(ndev));
+	dm9000_release_board(pdev, dm);
+	if (dm->power_supply)
+		regulator_disable(dm->power_supply);
+
 	free_netdev(ndev);		/* free device structure */
 
 	dev_dbg(&pdev->dev, "released and freed device\n");
-	return 0;
 }
 
 #ifdef CONFIG_OF
@@ -1799,7 +1800,7 @@ static struct platform_driver dm9000_driver = {
 		.of_match_table = of_match_ptr(dm9000_of_matches),
 	},
 	.probe   = dm9000_probe,
-	.remove  = dm9000_drv_remove,
+	.remove = dm9000_drv_remove,
 };
 
 module_platform_driver(dm9000_driver);
diff --git a/drivers/net/ethernet/davicom/dm9000.h b/drivers/net/ethernet/davicom/dm9000.h
index 9ce058adabab..581b35ad44ef 100644
--- a/drivers/net/ethernet/davicom/dm9000.h
+++ b/drivers/net/ethernet/davicom/dm9000.h
@@ -1,3 +1,4 @@
+/* SPDX-License-Identifier: GPL-2.0 */
 /*
  * dm9000 Ethernet
  */
diff --git a/drivers/net/ethernet/davicom/dm9051.c b/drivers/net/ethernet/davicom/dm9051.c
new file mode 100644
index 000000000000..59ea48d4c9de
--- /dev/null
+++ b/drivers/net/ethernet/davicom/dm9051.c
@@ -0,0 +1,1258 @@
+// SPDX-License-Identifier: GPL-2.0-only
+/*
+ * Copyright (c) 2022 Davicom Semiconductor,Inc.
+ * Davicom DM9051 SPI Fast Ethernet Linux driver
+ */
+
+#include <linux/etherdevice.h>
+#include <linux/ethtool.h>
+#include <linux/interrupt.h>
+#include <linux/iopoll.h>
+#include <linux/irq.h>
+#include <linux/mii.h>
+#include <linux/module.h>
+#include <linux/netdevice.h>
+#include <linux/phy.h>
+#include <linux/regmap.h>
+#include <linux/skbuff.h>
+#include <linux/spinlock.h>
+#include <linux/spi/spi.h>
+#include <linux/types.h>
+
+#include "dm9051.h"
+
+#define DRVNAME_9051	"dm9051"
+
+/**
+ * struct rx_ctl_mach - rx activities record
+ * @status_err_counter: rx status error counter
+ * @large_err_counter: rx get large packet length error counter
+ * @rx_err_counter: receive packet error counter
+ * @tx_err_counter: transmit packet error counter
+ * @fifo_rst_counter: reset operation counter
+ *
+ * To keep track for the driver operation statistics
+ */
+struct rx_ctl_mach {
+	u16				status_err_counter;
+	u16				large_err_counter;
+	u16				rx_err_counter;
+	u16				tx_err_counter;
+	u16				fifo_rst_counter;
+};
+
+/**
+ * struct dm9051_rxctrl - dm9051 driver rx control
+ * @hash_table: Multicast hash-table data
+ * @rcr_all: KS_RXCR1 register setting
+ *
+ * The settings needs to control the receive filtering
+ * such as the multicast hash-filter and the receive register settings
+ */
+struct dm9051_rxctrl {
+	u16				hash_table[4];
+	u8				rcr_all;
+};
+
+/**
+ * struct dm9051_rxhdr - rx packet data header
+ * @headbyte: lead byte equal to 0x01 notifies a valid packet
+ * @status: status bits for the received packet
+ * @rxlen: packet length
+ *
+ * The Rx packed, entered into the FIFO memory, start with these
+ * four bytes which is the Rx header, followed by the ethernet
+ * packet data and ends with an appended 4-byte CRC data.
+ * Both Rx packet and CRC data are for check purpose and finally
+ * are dropped by this driver
+ */
+struct dm9051_rxhdr {
+	u8				headbyte;
+	u8				status;
+	__le16				rxlen;
+};
+
+/**
+ * struct board_info - maintain the saved data
+ * @spidev: spi device structure
+ * @ndev: net device structure
+ * @mdiobus: mii bus structure
+ * @phydev: phy device structure
+ * @txq: tx queue structure
+ * @regmap_dm: regmap for register read/write
+ * @regmap_dmbulk: extra regmap for bulk read/write
+ * @rxctrl_work: Work queue for updating RX mode and multicast lists
+ * @tx_work: Work queue for tx packets
+ * @pause: ethtool pause parameter structure
+ * @spi_lockm: between threads lock structure
+ * @reg_mutex: regmap access lock structure
+ * @bc: rx control statistics structure
+ * @rxhdr: rx header structure
+ * @rctl: rx control setting structure
+ * @msg_enable: message level value
+ * @imr_all: to store operating imr value for register DM9051_IMR
+ * @lcr_all: to store operating rcr value for register DM9051_LMCR
+ *
+ * The saved data variables, keep up to date for retrieval back to use
+ */
+struct board_info {
+	u32				msg_enable;
+	struct spi_device		*spidev;
+	struct net_device		*ndev;
+	struct mii_bus			*mdiobus;
+	struct phy_device		*phydev;
+	struct sk_buff_head		txq;
+	struct regmap			*regmap_dm;
+	struct regmap			*regmap_dmbulk;
+	struct work_struct		rxctrl_work;
+	struct work_struct		tx_work;
+	struct ethtool_pauseparam	pause;
+	struct mutex			spi_lockm;
+	struct mutex			reg_mutex;
+	struct rx_ctl_mach		bc;
+	struct dm9051_rxhdr		rxhdr;
+	struct dm9051_rxctrl		rctl;
+	u8				imr_all;
+	u8				lcr_all;
+};
+
+static int dm9051_set_reg(struct board_info *db, unsigned int reg, unsigned int val)
+{
+	int ret;
+
+	ret = regmap_write(db->regmap_dm, reg, val);
+	if (ret < 0)
+		netif_err(db, drv, db->ndev, "%s: error %d set reg %02x\n",
+			  __func__, ret, reg);
+	return ret;
+}
+
+static int dm9051_update_bits(struct board_info *db, unsigned int reg, unsigned int mask,
+			      unsigned int val)
+{
+	int ret;
+
+	ret = regmap_update_bits(db->regmap_dm, reg, mask, val);
+	if (ret < 0)
+		netif_err(db, drv, db->ndev, "%s: error %d update bits reg %02x\n",
+			  __func__, ret, reg);
+	return ret;
+}
+
+/* skb buffer exhausted, just discard the received data
+ */
+static int dm9051_dumpblk(struct board_info *db, u8 reg, size_t count)
+{
+	struct net_device *ndev = db->ndev;
+	unsigned int rb;
+	int ret;
+
+	/* no skb buffer,
+	 * both reg and &rb must be noinc,
+	 * read once one byte via regmap_read
+	 */
+	do {
+		ret = regmap_read(db->regmap_dm, reg, &rb);
+		if (ret < 0) {
+			netif_err(db, drv, ndev, "%s: error %d dumping read reg %02x\n",
+				  __func__, ret, reg);
+			break;
+		}
+	} while (--count);
+
+	return ret;
+}
+
+static int dm9051_set_regs(struct board_info *db, unsigned int reg, const void *val,
+			   size_t val_count)
+{
+	int ret;
+
+	ret = regmap_bulk_write(db->regmap_dmbulk, reg, val, val_count);
+	if (ret < 0)
+		netif_err(db, drv, db->ndev, "%s: error %d bulk writing regs %02x\n",
+			  __func__, ret, reg);
+	return ret;
+}
+
+static int dm9051_get_regs(struct board_info *db, unsigned int reg, void *val,
+			   size_t val_count)
+{
+	int ret;
+
+	ret = regmap_bulk_read(db->regmap_dmbulk, reg, val, val_count);
+	if (ret < 0)
+		netif_err(db, drv, db->ndev, "%s: error %d bulk reading regs %02x\n",
+			  __func__, ret, reg);
+	return ret;
+}
+
+static int dm9051_write_mem(struct board_info *db, unsigned int reg, const void *buff,
+			    size_t len)
+{
+	int ret;
+
+	ret = regmap_noinc_write(db->regmap_dm, reg, buff, len);
+	if (ret < 0)
+		netif_err(db, drv, db->ndev, "%s: error %d noinc writing regs %02x\n",
+			  __func__, ret, reg);
+	return ret;
+}
+
+static int dm9051_read_mem(struct board_info *db, unsigned int reg, void *buff,
+			   size_t len)
+{
+	int ret;
+
+	ret = regmap_noinc_read(db->regmap_dm, reg, buff, len);
+	if (ret < 0)
+		netif_err(db, drv, db->ndev, "%s: error %d noinc reading regs %02x\n",
+			  __func__, ret, reg);
+	return ret;
+}
+
+/* waiting tx-end rather than tx-req
+ * got faster
+ */
+static int dm9051_nsr_poll(struct board_info *db)
+{
+	unsigned int mval;
+	int ret;
+
+	ret = regmap_read_poll_timeout(db->regmap_dm, DM9051_NSR, mval,
+				       mval & (NSR_TX2END | NSR_TX1END), 1, 20);
+	if (ret == -ETIMEDOUT)
+		netdev_err(db->ndev, "timeout in checking for tx end\n");
+	return ret;
+}
+
+static int dm9051_epcr_poll(struct board_info *db)
+{
+	unsigned int mval;
+	int ret;
+
+	ret = regmap_read_poll_timeout(db->regmap_dm, DM9051_EPCR, mval,
+				       !(mval & EPCR_ERRE), 100, 10000);
+	if (ret == -ETIMEDOUT)
+		netdev_err(db->ndev, "eeprom/phy in processing get timeout\n");
+	return ret;
+}
+
+static int dm9051_irq_flag(struct board_info *db)
+{
+	struct spi_device *spi = db->spidev;
+	int irq_type = irq_get_trigger_type(spi->irq);
+
+	if (irq_type)
+		return irq_type;
+
+	return IRQF_TRIGGER_LOW;
+}
+
+static unsigned int dm9051_intcr_value(struct board_info *db)
+{
+	return (dm9051_irq_flag(db) == IRQF_TRIGGER_LOW) ?
+		INTCR_POL_LOW : INTCR_POL_HIGH;
+}
+
+static int dm9051_set_fcr(struct board_info *db)
+{
+	u8 fcr = 0;
+
+	if (db->pause.rx_pause)
+		fcr |= FCR_BKPM | FCR_FLCE;
+	if (db->pause.tx_pause)
+		fcr |= FCR_TXPEN;
+
+	return dm9051_set_reg(db, DM9051_FCR, fcr);
+}
+
+static int dm9051_set_recv(struct board_info *db)
+{
+	int ret;
+
+	ret = dm9051_set_regs(db, DM9051_MAR, db->rctl.hash_table, sizeof(db->rctl.hash_table));
+	if (ret)
+		return ret;
+
+	return dm9051_set_reg(db, DM9051_RCR, db->rctl.rcr_all); /* enable rx */
+}
+
+static int dm9051_core_reset(struct board_info *db)
+{
+	int ret;
+
+	db->bc.fifo_rst_counter++;
+
+	ret = regmap_write(db->regmap_dm, DM9051_NCR, NCR_RST); /* NCR reset */
+	if (ret)
+		return ret;
+	ret = regmap_write(db->regmap_dm, DM9051_MBNDRY, MBNDRY_BYTE); /* MemBound */
+	if (ret)
+		return ret;
+	ret = regmap_write(db->regmap_dm, DM9051_PPCR, PPCR_PAUSE_COUNT); /* Pause Count */
+	if (ret)
+		return ret;
+	ret = regmap_write(db->regmap_dm, DM9051_LMCR, db->lcr_all); /* LEDMode1 */
+	if (ret)
+		return ret;
+
+	return dm9051_set_reg(db, DM9051_INTCR, dm9051_intcr_value(db));
+}
+
+static int dm9051_update_fcr(struct board_info *db)
+{
+	u8 fcr = 0;
+
+	if (db->pause.rx_pause)
+		fcr |= FCR_BKPM | FCR_FLCE;
+	if (db->pause.tx_pause)
+		fcr |= FCR_TXPEN;
+
+	return dm9051_update_bits(db, DM9051_FCR, FCR_RXTX_BITS, fcr);
+}
+
+static int dm9051_disable_interrupt(struct board_info *db)
+{
+	return dm9051_set_reg(db, DM9051_IMR, IMR_PAR); /* disable int */
+}
+
+static int dm9051_enable_interrupt(struct board_info *db)
+{
+	return dm9051_set_reg(db, DM9051_IMR, db->imr_all); /* enable int */
+}
+
+static int dm9051_stop_mrcmd(struct board_info *db)
+{
+	return dm9051_set_reg(db, DM9051_ISR, ISR_STOP_MRCMD); /* to stop mrcmd */
+}
+
+static int dm9051_clear_interrupt(struct board_info *db)
+{
+	return dm9051_update_bits(db, DM9051_ISR, ISR_CLR_INT, ISR_CLR_INT);
+}
+
+static int dm9051_eeprom_read(struct board_info *db, int offset, u8 *to)
+{
+	int ret;
+
+	ret = regmap_write(db->regmap_dm, DM9051_EPAR, offset);
+	if (ret)
+		return ret;
+
+	ret = regmap_write(db->regmap_dm, DM9051_EPCR, EPCR_ERPRR);
+	if (ret)
+		return ret;
+
+	ret = dm9051_epcr_poll(db);
+	if (ret)
+		return ret;
+
+	ret = regmap_write(db->regmap_dm, DM9051_EPCR, 0);
+	if (ret)
+		return ret;
+
+	return regmap_bulk_read(db->regmap_dmbulk, DM9051_EPDRL, to, 2);
+}
+
+static int dm9051_eeprom_write(struct board_info *db, int offset, u8 *data)
+{
+	int ret;
+
+	ret = regmap_write(db->regmap_dm, DM9051_EPAR, offset);
+	if (ret)
+		return ret;
+
+	ret = regmap_bulk_write(db->regmap_dmbulk, DM9051_EPDRL, data, 2);
+	if (ret < 0)
+		return ret;
+
+	ret = regmap_write(db->regmap_dm, DM9051_EPCR, EPCR_WEP | EPCR_ERPRW);
+	if (ret)
+		return ret;
+
+	ret = dm9051_epcr_poll(db);
+	if (ret)
+		return ret;
+
+	return regmap_write(db->regmap_dm, DM9051_EPCR, 0);
+}
+
+static int dm9051_phyread(void *context, unsigned int reg, unsigned int *val)
+{
+	struct board_info *db = context;
+	int ret;
+
+	ret = regmap_write(db->regmap_dm, DM9051_EPAR, DM9051_PHY | reg);
+	if (ret)
+		return ret;
+
+	ret = regmap_write(db->regmap_dm, DM9051_EPCR, EPCR_ERPRR | EPCR_EPOS);
+	if (ret)
+		return ret;
+
+	ret = dm9051_epcr_poll(db);
+	if (ret)
+		return ret;
+
+	ret = regmap_write(db->regmap_dm, DM9051_EPCR, 0);
+	if (ret)
+		return ret;
+
+	/* this is a 4 bytes data, clear to zero since following regmap_bulk_read
+	 * only fill lower 2 bytes
+	 */
+	*val = 0;
+	return regmap_bulk_read(db->regmap_dmbulk, DM9051_EPDRL, val, 2);
+}
+
+static int dm9051_phywrite(void *context, unsigned int reg, unsigned int val)
+{
+	struct board_info *db = context;
+	int ret;
+
+	ret = regmap_write(db->regmap_dm, DM9051_EPAR, DM9051_PHY | reg);
+	if (ret)
+		return ret;
+
+	ret = regmap_bulk_write(db->regmap_dmbulk, DM9051_EPDRL, &val, 2);
+	if (ret < 0)
+		return ret;
+
+	ret = regmap_write(db->regmap_dm, DM9051_EPCR, EPCR_EPOS | EPCR_ERPRW);
+	if (ret)
+		return ret;
+
+	ret = dm9051_epcr_poll(db);
+	if (ret)
+		return ret;
+
+	return regmap_write(db->regmap_dm, DM9051_EPCR, 0);
+}
+
+static int dm9051_mdio_read(struct mii_bus *bus, int addr, int regnum)
+{
+	struct board_info *db = bus->priv;
+	unsigned int val = 0xffff;
+	int ret;
+
+	if (addr == DM9051_PHY_ADDR) {
+		ret = dm9051_phyread(db, regnum, &val);
+		if (ret)
+			return ret;
+	}
+
+	return val;
+}
+
+static int dm9051_mdio_write(struct mii_bus *bus, int addr, int regnum, u16 val)
+{
+	struct board_info *db = bus->priv;
+
+	if (addr == DM9051_PHY_ADDR)
+		return dm9051_phywrite(db, regnum, val);
+
+	return -ENODEV;
+}
+
+static void dm9051_reg_lock_mutex(void *dbcontext)
+{
+	struct board_info *db = dbcontext;
+
+	mutex_lock(&db->reg_mutex);
+}
+
+static void dm9051_reg_unlock_mutex(void *dbcontext)
+{
+	struct board_info *db = dbcontext;
+
+	mutex_unlock(&db->reg_mutex);
+}
+
+static struct regmap_config regconfigdm = {
+	.reg_bits = 8,
+	.val_bits = 8,
+	.max_register = 0xff,
+	.reg_stride = 1,
+	.cache_type = REGCACHE_NONE,
+	.read_flag_mask = 0,
+	.write_flag_mask = DM_SPI_WR,
+	.val_format_endian = REGMAP_ENDIAN_LITTLE,
+	.lock = dm9051_reg_lock_mutex,
+	.unlock = dm9051_reg_unlock_mutex,
+};
+
+static struct regmap_config regconfigdmbulk = {
+	.reg_bits = 8,
+	.val_bits = 8,
+	.max_register = 0xff,
+	.reg_stride = 1,
+	.cache_type = REGCACHE_NONE,
+	.read_flag_mask = 0,
+	.write_flag_mask = DM_SPI_WR,
+	.val_format_endian = REGMAP_ENDIAN_LITTLE,
+	.lock = dm9051_reg_lock_mutex,
+	.unlock = dm9051_reg_unlock_mutex,
+	.use_single_read = true,
+	.use_single_write = true,
+};
+
+static int dm9051_map_init(struct spi_device *spi, struct board_info *db)
+{
+	/* create two regmap instances,
+	 * split read/write and bulk_read/bulk_write to individual regmap
+	 * to resolve regmap execution confliction problem
+	 */
+	regconfigdm.lock_arg = db;
+	db->regmap_dm = devm_regmap_init_spi(db->spidev, &regconfigdm);
+	if (IS_ERR(db->regmap_dm))
+		return PTR_ERR(db->regmap_dm);
+
+	regconfigdmbulk.lock_arg = db;
+	db->regmap_dmbulk = devm_regmap_init_spi(db->spidev, &regconfigdmbulk);
+	return PTR_ERR_OR_ZERO(db->regmap_dmbulk);
+}
+
+static int dm9051_map_chipid(struct board_info *db)
+{
+	struct device *dev = &db->spidev->dev;
+	unsigned short wid;
+	u8 buff[6];
+	int ret;
+
+	ret = dm9051_get_regs(db, DM9051_VIDL, buff, sizeof(buff));
+	if (ret < 0)
+		return ret;
+
+	wid = get_unaligned_le16(buff + 2);
+	if (wid != DM9051_ID) {
+		dev_err(dev, "chipid error as %04x !\n", wid);
+		return -ENODEV;
+	}
+
+	dev_info(dev, "chip %04x found\n", wid);
+	return 0;
+}
+
+/* Read DM9051_PAR registers which is the mac address loaded from EEPROM while power-on
+ */
+static int dm9051_map_etherdev_par(struct net_device *ndev, struct board_info *db)
+{
+	u8 addr[ETH_ALEN];
+	int ret;
+
+	ret = dm9051_get_regs(db, DM9051_PAR, addr, sizeof(addr));
+	if (ret < 0)
+		return ret;
+
+	if (!is_valid_ether_addr(addr)) {
+		eth_hw_addr_random(ndev);
+
+		ret = dm9051_set_regs(db, DM9051_PAR, ndev->dev_addr, sizeof(ndev->dev_addr));
+		if (ret < 0)
+			return ret;
+
+		dev_dbg(&db->spidev->dev, "Use random MAC address\n");
+		return 0;
+	}
+
+	eth_hw_addr_set(ndev, addr);
+	return 0;
+}
+
+/* ethtool-ops
+ */
+static void dm9051_get_drvinfo(struct net_device *dev, struct ethtool_drvinfo *info)
+{
+	strscpy(info->driver, DRVNAME_9051, sizeof(info->driver));
+}
+
+static void dm9051_set_msglevel(struct net_device *ndev, u32 value)
+{
+	struct board_info *db = to_dm9051_board(ndev);
+
+	db->msg_enable = value;
+}
+
+static u32 dm9051_get_msglevel(struct net_device *ndev)
+{
+	struct board_info *db = to_dm9051_board(ndev);
+
+	return db->msg_enable;
+}
+
+static int dm9051_get_eeprom_len(struct net_device *dev)
+{
+	return 128;
+}
+
+static int dm9051_get_eeprom(struct net_device *ndev,
+			     struct ethtool_eeprom *ee, u8 *data)
+{
+	struct board_info *db = to_dm9051_board(ndev);
+	int offset = ee->offset;
+	int len = ee->len;
+	int i, ret;
+
+	if ((len | offset) & 1)
+		return -EINVAL;
+
+	ee->magic = DM_EEPROM_MAGIC;
+
+	for (i = 0; i < len; i += 2) {
+		ret = dm9051_eeprom_read(db, (offset + i) / 2, data + i);
+		if (ret)
+			break;
+	}
+	return ret;
+}
+
+static int dm9051_set_eeprom(struct net_device *ndev,
+			     struct ethtool_eeprom *ee, u8 *data)
+{
+	struct board_info *db = to_dm9051_board(ndev);
+	int offset = ee->offset;
+	int len = ee->len;
+	int i, ret;
+
+	if ((len | offset) & 1)
+		return -EINVAL;
+
+	if (ee->magic != DM_EEPROM_MAGIC)
+		return -EINVAL;
+
+	for (i = 0; i < len; i += 2) {
+		ret = dm9051_eeprom_write(db, (offset + i) / 2, data + i);
+		if (ret)
+			break;
+	}
+	return ret;
+}
+
+static void dm9051_get_pauseparam(struct net_device *ndev,
+				  struct ethtool_pauseparam *pause)
+{
+	struct board_info *db = to_dm9051_board(ndev);
+
+	*pause = db->pause;
+}
+
+static int dm9051_set_pauseparam(struct net_device *ndev,
+				 struct ethtool_pauseparam *pause)
+{
+	struct board_info *db = to_dm9051_board(ndev);
+
+	db->pause = *pause;
+
+	if (pause->autoneg == AUTONEG_DISABLE)
+		return dm9051_update_fcr(db);
+
+	phy_set_sym_pause(db->phydev, pause->rx_pause, pause->tx_pause,
+			  pause->autoneg);
+	phy_start_aneg(db->phydev);
+	return 0;
+}
+
+static const struct ethtool_ops dm9051_ethtool_ops = {
+	.get_drvinfo = dm9051_get_drvinfo,
+	.get_link_ksettings = phy_ethtool_get_link_ksettings,
+	.set_link_ksettings = phy_ethtool_set_link_ksettings,
+	.get_msglevel = dm9051_get_msglevel,
+	.set_msglevel = dm9051_set_msglevel,
+	.nway_reset = phy_ethtool_nway_reset,
+	.get_link = ethtool_op_get_link,
+	.get_eeprom_len = dm9051_get_eeprom_len,
+	.get_eeprom = dm9051_get_eeprom,
+	.set_eeprom = dm9051_set_eeprom,
+	.get_pauseparam = dm9051_get_pauseparam,
+	.set_pauseparam = dm9051_set_pauseparam,
+};
+
+static int dm9051_all_start(struct board_info *db)
+{
+	int ret;
+
+	/* GPR power on of the internal phy
+	 */
+	ret = dm9051_set_reg(db, DM9051_GPR, 0);
+	if (ret)
+		return ret;
+
+	/* dm9051 chip registers could not be accessed within 1 ms
+	 * after GPR power on, delay 1 ms is essential
+	 */
+	msleep(1);
+
+	ret = dm9051_core_reset(db);
+	if (ret)
+		return ret;
+
+	return dm9051_enable_interrupt(db);
+}
+
+static int dm9051_all_stop(struct board_info *db)
+{
+	int ret;
+
+	/* GPR power off of the internal phy,
+	 * The internal phy still could be accessed after this GPR power off control
+	 */
+	ret = dm9051_set_reg(db, DM9051_GPR, GPR_PHY_OFF);
+	if (ret)
+		return ret;
+
+	return dm9051_set_reg(db, DM9051_RCR, RCR_RX_DISABLE);
+}
+
+/* fifo reset while rx error found
+ */
+static int dm9051_all_restart(struct board_info *db)
+{
+	struct net_device *ndev = db->ndev;
+	int ret;
+
+	ret = dm9051_core_reset(db);
+	if (ret)
+		return ret;
+
+	ret = dm9051_enable_interrupt(db);
+	if (ret)
+		return ret;
+
+	netdev_dbg(ndev, " rxstatus_Er & rxlen_Er %d, RST_c %d\n",
+		   db->bc.status_err_counter + db->bc.large_err_counter,
+		   db->bc.fifo_rst_counter);
+
+	ret = dm9051_set_recv(db);
+	if (ret)
+		return ret;
+
+	return dm9051_set_fcr(db);
+}
+
+/* read packets from the fifo memory
+ * return value,
+ *  > 0 - read packet number, caller can repeat the rx operation
+ *    0 - no error, caller need stop further rx operation
+ *  -EBUSY - read data error, caller escape from rx operation
+ */
+static int dm9051_loop_rx(struct board_info *db)
+{
+	struct net_device *ndev = db->ndev;
+	unsigned int rxbyte;
+	int ret, rxlen;
+	struct sk_buff *skb;
+	u8 *rdptr;
+	int scanrr = 0;
+
+	do {
+		ret = dm9051_read_mem(db, DM_SPI_MRCMDX, &rxbyte, 2);
+		if (ret)
+			return ret;
+
+		if ((rxbyte & GENMASK(7, 0)) != DM9051_PKT_RDY)
+			break; /* exhaust-empty */
+
+		ret = dm9051_read_mem(db, DM_SPI_MRCMD, &db->rxhdr, DM_RXHDR_SIZE);
+		if (ret)
+			return ret;
+
+		ret = dm9051_stop_mrcmd(db);
+		if (ret)
+			return ret;
+
+		rxlen = le16_to_cpu(db->rxhdr.rxlen);
+		if (db->rxhdr.status & RSR_ERR_BITS || rxlen > DM9051_PKT_MAX) {
+			netdev_dbg(ndev, "rxhdr-byte (%02x)\n",
+				   db->rxhdr.headbyte);
+
+			if (db->rxhdr.status & RSR_ERR_BITS) {
+				db->bc.status_err_counter++;
+				netdev_dbg(ndev, "check rxstatus-error (%02x)\n",
+					   db->rxhdr.status);
+			} else {
+				db->bc.large_err_counter++;
+				netdev_dbg(ndev, "check rxlen large-error (%d > %d)\n",
+					   rxlen, DM9051_PKT_MAX);
+			}
+			return dm9051_all_restart(db);
+		}
+
+		skb = dev_alloc_skb(rxlen);
+		if (!skb) {
+			ret = dm9051_dumpblk(db, DM_SPI_MRCMD, rxlen);
+			if (ret)
+				return ret;
+			return scanrr;
+		}
+
+		rdptr = skb_put(skb, rxlen - 4);
+		ret = dm9051_read_mem(db, DM_SPI_MRCMD, rdptr, rxlen);
+		if (ret) {
+			db->bc.rx_err_counter++;
+			dev_kfree_skb(skb);
+			return ret;
+		}
+
+		ret = dm9051_stop_mrcmd(db);
+		if (ret) {
+			dev_kfree_skb(skb);
+			return ret;
+		}
+
+		skb->protocol = eth_type_trans(skb, db->ndev);
+		if (db->ndev->features & NETIF_F_RXCSUM)
+			skb_checksum_none_assert(skb);
+		netif_rx(skb);
+		db->ndev->stats.rx_bytes += rxlen;
+		db->ndev->stats.rx_packets++;
+		scanrr++;
+	} while (!ret);
+
+	return scanrr;
+}
+
+/* transmit a packet,
+ * return value,
+ *   0 - succeed
+ *  -ETIMEDOUT - timeout error
+ */
+static int dm9051_single_tx(struct board_info *db, u8 *buff, unsigned int len)
+{
+	int ret;
+
+	ret = dm9051_nsr_poll(db);
+	if (ret)
+		return ret;
+
+	ret = dm9051_write_mem(db, DM_SPI_MWCMD, buff, len);
+	if (ret)
+		return ret;
+
+	ret = dm9051_set_regs(db, DM9051_TXPLL, &len, 2);
+	if (ret < 0)
+		return ret;
+
+	return dm9051_set_reg(db, DM9051_TCR, TCR_TXREQ);
+}
+
+static int dm9051_loop_tx(struct board_info *db)
+{
+	struct net_device *ndev = db->ndev;
+	int ntx = 0;
+	int ret;
+
+	while (!skb_queue_empty(&db->txq)) {
+		struct sk_buff *skb;
+		unsigned int len;
+
+		skb = skb_dequeue(&db->txq);
+		if (skb) {
+			ntx++;
+			ret = dm9051_single_tx(db, skb->data, skb->len);
+			len = skb->len;
+			dev_kfree_skb(skb);
+			if (ret < 0) {
+				db->bc.tx_err_counter++;
+				return 0;
+			}
+			ndev->stats.tx_bytes += len;
+			ndev->stats.tx_packets++;
+		}
+
+		if (netif_queue_stopped(ndev) &&
+		    (skb_queue_len(&db->txq) < DM9051_TX_QUE_LO_WATER))
+			netif_wake_queue(ndev);
+	}
+
+	return ntx;
+}
+
+static irqreturn_t dm9051_rx_threaded_irq(int irq, void *pw)
+{
+	struct board_info *db = pw;
+	int result, result_tx;
+
+	mutex_lock(&db->spi_lockm);
+
+	result = dm9051_disable_interrupt(db);
+	if (result)
+		goto out_unlock;
+
+	result = dm9051_clear_interrupt(db);
+	if (result)
+		goto out_unlock;
+
+	do {
+		result = dm9051_loop_rx(db); /* threaded irq rx */
+		if (result < 0)
+			goto out_unlock;
+		result_tx = dm9051_loop_tx(db); /* more tx better performance */
+		if (result_tx < 0)
+			goto out_unlock;
+	} while (result > 0);
+
+	dm9051_enable_interrupt(db);
+
+	/* To exit and has mutex unlock while rx or tx error
+	 */
+out_unlock:
+	mutex_unlock(&db->spi_lockm);
+
+	return IRQ_HANDLED;
+}
+
+static void dm9051_tx_delay(struct work_struct *work)
+{
+	struct board_info *db = container_of(work, struct board_info, tx_work);
+	int result;
+
+	mutex_lock(&db->spi_lockm);
+
+	result = dm9051_loop_tx(db);
+	if (result < 0)
+		netdev_err(db->ndev, "transmit packet error\n");
+
+	mutex_unlock(&db->spi_lockm);
+}
+
+static void dm9051_rxctl_delay(struct work_struct *work)
+{
+	struct board_info *db = container_of(work, struct board_info, rxctrl_work);
+	struct net_device *ndev = db->ndev;
+	int result;
+
+	mutex_lock(&db->spi_lockm);
+
+	result = dm9051_set_regs(db, DM9051_PAR, ndev->dev_addr, sizeof(ndev->dev_addr));
+	if (result < 0)
+		goto out_unlock;
+
+	dm9051_set_recv(db);
+
+	/* To has mutex unlock and return from this function if regmap function fail
+	 */
+out_unlock:
+	mutex_unlock(&db->spi_lockm);
+}
+
+/* Open network device
+ * Called when the network device is marked active, such as a user executing
+ * 'ifconfig up' on the device
+ */
+static int dm9051_open(struct net_device *ndev)
+{
+	struct board_info *db = to_dm9051_board(ndev);
+	struct spi_device *spi = db->spidev;
+	int ret;
+
+	db->imr_all = IMR_PAR | IMR_PRM;
+	db->lcr_all = LMCR_MODE1;
+	db->rctl.rcr_all = RCR_DIS_LONG | RCR_DIS_CRC | RCR_RXEN;
+	memset(db->rctl.hash_table, 0, sizeof(db->rctl.hash_table));
+
+	ndev->irq = spi->irq; /* by dts */
+	ret = request_threaded_irq(spi->irq, NULL, dm9051_rx_threaded_irq,
+				   dm9051_irq_flag(db) | IRQF_ONESHOT,
+				   ndev->name, db);
+	if (ret < 0) {
+		netdev_err(ndev, "failed to get irq\n");
+		return ret;
+	}
+
+	phy_support_sym_pause(db->phydev);
+	phy_start(db->phydev);
+
+	/* flow control parameters init */
+	db->pause.rx_pause = true;
+	db->pause.tx_pause = true;
+	db->pause.autoneg = AUTONEG_DISABLE;
+
+	if (db->phydev->autoneg)
+		db->pause.autoneg = AUTONEG_ENABLE;
+
+	ret = dm9051_all_start(db);
+	if (ret) {
+		phy_stop(db->phydev);
+		free_irq(spi->irq, db);
+		return ret;
+	}
+
+	netif_wake_queue(ndev);
+
+	return 0;
+}
+
+/* Close network device
+ * Called to close down a network device which has been active. Cancel any
+ * work, shutdown the RX and TX process and then place the chip into a low
+ * power state while it is not being used
+ */
+static int dm9051_stop(struct net_device *ndev)
+{
+	struct board_info *db = to_dm9051_board(ndev);
+	int ret;
+
+	ret = dm9051_all_stop(db);
+	if (ret)
+		return ret;
+
+	flush_work(&db->tx_work);
+	flush_work(&db->rxctrl_work);
+
+	phy_stop(db->phydev);
+
+	free_irq(db->spidev->irq, db);
+
+	netif_stop_queue(ndev);
+
+	skb_queue_purge(&db->txq);
+
+	return 0;
+}
+
+/* event: play a schedule starter in condition
+ */
+static netdev_tx_t dm9051_start_xmit(struct sk_buff *skb, struct net_device *ndev)
+{
+	struct board_info *db = to_dm9051_board(ndev);
+
+	skb_queue_tail(&db->txq, skb);
+	if (skb_queue_len(&db->txq) > DM9051_TX_QUE_HI_WATER)
+		netif_stop_queue(ndev); /* enforce limit queue size */
+
+	schedule_work(&db->tx_work);
+
+	return NETDEV_TX_OK;
+}
+
+/* event: play with a schedule starter
+ */
+static void dm9051_set_rx_mode(struct net_device *ndev)
+{
+	struct board_info *db = to_dm9051_board(ndev);
+	struct dm9051_rxctrl rxctrl;
+	struct netdev_hw_addr *ha;
+	u8 rcr = RCR_DIS_LONG | RCR_DIS_CRC | RCR_RXEN;
+	u32 hash_val;
+
+	memset(&rxctrl, 0, sizeof(rxctrl));
+
+	/* rx control */
+	if (ndev->flags & IFF_PROMISC) {
+		rcr |= RCR_PRMSC;
+		netdev_dbg(ndev, "set_multicast rcr |= RCR_PRMSC, rcr= %02x\n", rcr);
+	}
+
+	if (ndev->flags & IFF_ALLMULTI) {
+		rcr |= RCR_ALL;
+		netdev_dbg(ndev, "set_multicast rcr |= RCR_ALLMULTI, rcr= %02x\n", rcr);
+	}
+
+	rxctrl.rcr_all = rcr;
+
+	/* broadcast address */
+	rxctrl.hash_table[0] = 0;
+	rxctrl.hash_table[1] = 0;
+	rxctrl.hash_table[2] = 0;
+	rxctrl.hash_table[3] = 0x8000;
+
+	/* the multicast address in Hash Table : 64 bits */
+	netdev_for_each_mc_addr(ha, ndev) {
+		hash_val = ether_crc_le(ETH_ALEN, ha->addr) & GENMASK(5, 0);
+		rxctrl.hash_table[hash_val / 16] |= BIT(0) << (hash_val % 16);
+	}
+
+	/* schedule work to do the actual set of the data if needed */
+
+	if (memcmp(&db->rctl, &rxctrl, sizeof(rxctrl))) {
+		memcpy(&db->rctl, &rxctrl, sizeof(rxctrl));
+		schedule_work(&db->rxctrl_work);
+	}
+}
+
+/* event: write into the mac registers and eeprom directly
+ */
+static int dm9051_set_mac_address(struct net_device *ndev, void *p)
+{
+	struct board_info *db = to_dm9051_board(ndev);
+	int ret;
+
+	ret = eth_prepare_mac_addr_change(ndev, p);
+	if (ret < 0)
+		return ret;
+
+	eth_commit_mac_addr_change(ndev, p);
+	return dm9051_set_regs(db, DM9051_PAR, ndev->dev_addr, sizeof(ndev->dev_addr));
+}
+
+static const struct net_device_ops dm9051_netdev_ops = {
+	.ndo_open = dm9051_open,
+	.ndo_stop = dm9051_stop,
+	.ndo_start_xmit = dm9051_start_xmit,
+	.ndo_set_rx_mode = dm9051_set_rx_mode,
+	.ndo_validate_addr = eth_validate_addr,
+	.ndo_set_mac_address = dm9051_set_mac_address,
+};
+
+static void dm9051_operation_clear(struct board_info *db)
+{
+	db->bc.status_err_counter = 0;
+	db->bc.large_err_counter = 0;
+	db->bc.rx_err_counter = 0;
+	db->bc.tx_err_counter = 0;
+	db->bc.fifo_rst_counter = 0;
+}
+
+static int dm9051_mdio_register(struct board_info *db)
+{
+	struct spi_device *spi = db->spidev;
+	int ret;
+
+	db->mdiobus = devm_mdiobus_alloc(&spi->dev);
+	if (!db->mdiobus)
+		return -ENOMEM;
+
+	db->mdiobus->priv = db;
+	db->mdiobus->read = dm9051_mdio_read;
+	db->mdiobus->write = dm9051_mdio_write;
+	db->mdiobus->name = "dm9051-mdiobus";
+	db->mdiobus->phy_mask = (u32)~BIT(1);
+	db->mdiobus->parent = &spi->dev;
+	snprintf(db->mdiobus->id, MII_BUS_ID_SIZE,
+		 "dm9051-%s.%u", dev_name(&spi->dev), spi_get_chipselect(spi, 0));
+
+	ret = devm_mdiobus_register(&spi->dev, db->mdiobus);
+	if (ret)
+		dev_err(&spi->dev, "Could not register MDIO bus\n");
+
+	return ret;
+}
+
+static void dm9051_handle_link_change(struct net_device *ndev)
+{
+	struct board_info *db = to_dm9051_board(ndev);
+
+	phy_print_status(db->phydev);
+
+	/* only write pause settings to mac. since mac and phy are integrated
+	 * together, such as link state, speed and duplex are sync already
+	 */
+	if (db->phydev->link) {
+		if (db->phydev->pause) {
+			db->pause.rx_pause = true;
+			db->pause.tx_pause = true;
+		}
+		dm9051_update_fcr(db);
+	}
+}
+
+/* phy connect as poll mode
+ */
+static int dm9051_phy_connect(struct board_info *db)
+{
+	char phy_id[MII_BUS_ID_SIZE + 3];
+
+	snprintf(phy_id, sizeof(phy_id), PHY_ID_FMT,
+		 db->mdiobus->id, DM9051_PHY_ADDR);
+
+	db->phydev = phy_connect(db->ndev, phy_id, dm9051_handle_link_change,
+				 PHY_INTERFACE_MODE_MII);
+	return PTR_ERR_OR_ZERO(db->phydev);
+}
+
+static int dm9051_probe(struct spi_device *spi)
+{
+	struct device *dev = &spi->dev;
+	struct net_device *ndev;
+	struct board_info *db;
+	int ret;
+
+	ndev = devm_alloc_etherdev(dev, sizeof(struct board_info));
+	if (!ndev)
+		return -ENOMEM;
+
+	SET_NETDEV_DEV(ndev, dev);
+	dev_set_drvdata(dev, ndev);
+
+	db = netdev_priv(ndev);
+
+	db->msg_enable = 0;
+	db->spidev = spi;
+	db->ndev = ndev;
+
+	ndev->netdev_ops = &dm9051_netdev_ops;
+	ndev->ethtool_ops = &dm9051_ethtool_ops;
+
+	mutex_init(&db->spi_lockm);
+	mutex_init(&db->reg_mutex);
+
+	INIT_WORK(&db->rxctrl_work, dm9051_rxctl_delay);
+	INIT_WORK(&db->tx_work, dm9051_tx_delay);
+
+	ret = dm9051_map_init(spi, db);
+	if (ret)
+		return ret;
+
+	ret = dm9051_map_chipid(db);
+	if (ret)
+		return ret;
+
+	ret = dm9051_map_etherdev_par(ndev, db);
+	if (ret < 0)
+		return ret;
+
+	ret = dm9051_mdio_register(db);
+	if (ret)
+		return ret;
+
+	ret = dm9051_phy_connect(db);
+	if (ret)
+		return ret;
+
+	dm9051_operation_clear(db);
+	skb_queue_head_init(&db->txq);
+
+	ret = devm_register_netdev(dev, ndev);
+	if (ret) {
+		phy_disconnect(db->phydev);
+		return dev_err_probe(dev, ret, "device register failed");
+	}
+
+	return 0;
+}
+
+static void dm9051_drv_remove(struct spi_device *spi)
+{
+	struct device *dev = &spi->dev;
+	struct net_device *ndev = dev_get_drvdata(dev);
+	struct board_info *db = to_dm9051_board(ndev);
+
+	phy_disconnect(db->phydev);
+}
+
+static const struct of_device_id dm9051_match_table[] = {
+	{ .compatible = "davicom,dm9051" },
+	{}
+};
+MODULE_DEVICE_TABLE(of, dm9051_match_table);
+
+static const struct spi_device_id dm9051_id_table[] = {
+	{ "dm9051", 0 },
+	{}
+};
+
+static struct spi_driver dm9051_driver = {
+	.driver = {
+		.name = DRVNAME_9051,
+		.of_match_table = dm9051_match_table,
+	},
+	.probe = dm9051_probe,
+	.remove = dm9051_drv_remove,
+	.id_table = dm9051_id_table,
+};
+module_spi_driver(dm9051_driver);
+
+MODULE_AUTHOR("Joseph CHANG <joseph_chang@davicom.com.tw>");
+MODULE_DESCRIPTION("Davicom DM9051 network SPI driver");
+MODULE_LICENSE("GPL");
diff --git a/drivers/net/ethernet/davicom/dm9051.h b/drivers/net/ethernet/davicom/dm9051.h
new file mode 100644
index 000000000000..fef3120edd7c
--- /dev/null
+++ b/drivers/net/ethernet/davicom/dm9051.h
@@ -0,0 +1,162 @@
+/* SPDX-License-Identifier: GPL-2.0-only */
+/*
+ * Copyright (c) 2022 Davicom Semiconductor,Inc.
+ * Davicom DM9051 SPI Fast Ethernet Linux driver
+ */
+
+#ifndef _DM9051_H_
+#define _DM9051_H_
+
+#include <linux/bits.h>
+#include <linux/netdevice.h>
+#include <linux/types.h>
+
+#define DM9051_ID		0x9051
+
+#define DM9051_NCR		0x00
+#define DM9051_NSR		0x01
+#define DM9051_TCR		0x02
+#define DM9051_RCR		0x05
+#define DM9051_BPTR		0x08
+#define DM9051_FCR		0x0A
+#define DM9051_EPCR		0x0B
+#define DM9051_EPAR		0x0C
+#define DM9051_EPDRL		0x0D
+#define DM9051_EPDRH		0x0E
+#define DM9051_PAR		0x10
+#define DM9051_MAR		0x16
+#define DM9051_GPCR		0x1E
+#define DM9051_GPR		0x1F
+
+#define DM9051_VIDL		0x28
+#define DM9051_VIDH		0x29
+#define DM9051_PIDL		0x2A
+#define DM9051_PIDH		0x2B
+#define DM9051_SMCR		0x2F
+#define	DM9051_ATCR		0x30
+#define	DM9051_SPIBCR		0x38
+#define DM9051_INTCR		0x39
+#define DM9051_PPCR		0x3D
+
+#define DM9051_MPCR		0x55
+#define DM9051_LMCR		0x57
+#define DM9051_MBNDRY		0x5E
+
+#define DM9051_MRRL		0x74
+#define DM9051_MRRH		0x75
+#define DM9051_MWRL		0x7A
+#define DM9051_MWRH		0x7B
+#define DM9051_TXPLL		0x7C
+#define DM9051_TXPLH		0x7D
+#define DM9051_ISR		0x7E
+#define DM9051_IMR		0x7F
+
+#define DM_SPI_MRCMDX		0x70
+#define DM_SPI_MRCMD		0x72
+#define DM_SPI_MWCMD		0x78
+
+#define DM_SPI_WR		0x80
+
+/* dm9051 Ethernet controller registers bits
+ */
+/* 0x00 */
+#define NCR_WAKEEN		BIT(6)
+#define NCR_FDX			BIT(3)
+#define NCR_RST			BIT(0)
+/* 0x01 */
+#define NSR_SPEED		BIT(7)
+#define NSR_LINKST		BIT(6)
+#define NSR_WAKEST		BIT(5)
+#define NSR_TX2END		BIT(3)
+#define NSR_TX1END		BIT(2)
+/* 0x02 */
+#define TCR_DIS_JABBER_TIMER	BIT(6) /* for Jabber Packet support */
+#define TCR_TXREQ		BIT(0)
+/* 0x05 */
+#define RCR_DIS_WATCHDOG_TIMER	BIT(6)  /* for Jabber Packet support */
+#define RCR_DIS_LONG		BIT(5)
+#define RCR_DIS_CRC		BIT(4)
+#define RCR_ALL			BIT(3)
+#define RCR_PRMSC		BIT(1)
+#define RCR_RXEN		BIT(0)
+#define RCR_RX_DISABLE		(RCR_DIS_LONG | RCR_DIS_CRC)
+/* 0x06 */
+#define RSR_RF			BIT(7)
+#define RSR_MF			BIT(6)
+#define RSR_LCS			BIT(5)
+#define RSR_RWTO		BIT(4)
+#define RSR_PLE			BIT(3)
+#define RSR_AE			BIT(2)
+#define RSR_CE			BIT(1)
+#define RSR_FOE			BIT(0)
+#define	RSR_ERR_BITS		(RSR_RF | RSR_LCS | RSR_RWTO | RSR_PLE | \
+				 RSR_AE | RSR_CE | RSR_FOE)
+/* 0x0A */
+#define FCR_TXPEN		BIT(5)
+#define FCR_BKPM		BIT(3)
+#define FCR_FLCE		BIT(0)
+#define FCR_RXTX_BITS		(FCR_TXPEN | FCR_BKPM | FCR_FLCE)
+/* 0x0B */
+#define EPCR_WEP		BIT(4)
+#define EPCR_EPOS		BIT(3)
+#define EPCR_ERPRR		BIT(2)
+#define EPCR_ERPRW		BIT(1)
+#define EPCR_ERRE		BIT(0)
+/* 0x1E */
+#define GPCR_GEP_CNTL		BIT(0)
+/* 0x1F */
+#define GPR_PHY_OFF		BIT(0)
+/* 0x30 */
+#define	ATCR_AUTO_TX		BIT(7)
+/* 0x39 */
+#define INTCR_POL_LOW		(1 << 0)
+#define INTCR_POL_HIGH		(0 << 0)
+/* 0x3D */
+/* Pause Packet Control Register - default = 1 */
+#define PPCR_PAUSE_COUNT	0x08
+/* 0x55 */
+#define MPCR_RSTTX		BIT(1)
+#define MPCR_RSTRX		BIT(0)
+/* 0x57 */
+/* LEDMode Control Register - LEDMode1 */
+/* Value 0x81 : bit[7] = 1, bit[2] = 0, bit[1:0] = 01b */
+#define LMCR_NEWMOD		BIT(7)
+#define LMCR_TYPED1		BIT(1)
+#define LMCR_TYPED0		BIT(0)
+#define LMCR_MODE1		(LMCR_NEWMOD | LMCR_TYPED0)
+/* 0x5E */
+#define MBNDRY_BYTE		BIT(7)
+/* 0xFE */
+#define ISR_MBS			BIT(7)
+#define ISR_LNKCHG		BIT(5)
+#define ISR_ROOS		BIT(3)
+#define ISR_ROS			BIT(2)
+#define ISR_PTS			BIT(1)
+#define ISR_PRS			BIT(0)
+#define ISR_CLR_INT		(ISR_LNKCHG | ISR_ROOS | ISR_ROS | \
+				 ISR_PTS | ISR_PRS)
+#define ISR_STOP_MRCMD		(ISR_MBS)
+/* 0xFF */
+#define IMR_PAR			BIT(7)
+#define IMR_LNKCHGI		BIT(5)
+#define IMR_PTM			BIT(1)
+#define IMR_PRM			BIT(0)
+
+/* Const
+ */
+#define DM9051_PHY_ADDR			1	/* PHY id */
+#define DM9051_PHY			0x40	/* PHY address 0x01 */
+#define DM9051_PKT_RDY			0x01	/* Packet ready to receive */
+#define DM9051_PKT_MAX			1536	/* Received packet max size */
+#define DM9051_TX_QUE_HI_WATER		50
+#define DM9051_TX_QUE_LO_WATER		25
+#define DM_EEPROM_MAGIC			0x9051
+
+#define	DM_RXHDR_SIZE			sizeof(struct dm9051_rxhdr)
+
+static inline struct board_info *to_dm9051_board(struct net_device *ndev)
+{
+	return netdev_priv(ndev);
+}
+
+#endif /* _DM9051_H_ */
diff --git a/drivers/net/ethernet/dec/Kconfig b/drivers/net/ethernet/dec/Kconfig
index 740bbad5ed38..9e5e5f10bd19 100644
--- a/drivers/net/ethernet/dec/Kconfig
+++ b/drivers/net/ethernet/dec/Kconfig
@@ -1,3 +1,4 @@
+# SPDX-License-Identifier: GPL-2.0-only
 #
 # Digital Equipment Inc network device configuration
 #
@@ -6,7 +7,7 @@ config NET_VENDOR_DEC
 	bool "Digital Equipment devices"
 	default y
 	depends on PCI || EISA || CARDBUS
-	---help---
+	help
 	  If you have a network (Ethernet) card belonging to this class, say Y.
 
 	  Note that the answer to this question doesn't directly affect the
diff --git a/drivers/net/ethernet/dec/Makefile b/drivers/net/ethernet/dec/Makefile
index 32993fccbbfd..e8aa12c8492a 100644
--- a/drivers/net/ethernet/dec/Makefile
+++ b/drivers/net/ethernet/dec/Makefile
@@ -1,3 +1,4 @@
+# SPDX-License-Identifier: GPL-2.0-only
 #
 # Makefile for the Digital Equipment Inc. network device drivers.
 #
diff --git a/drivers/net/ethernet/dec/tulip/21142.c b/drivers/net/ethernet/dec/tulip/21142.c
index 369858272650..76767dec216d 100644
--- a/drivers/net/ethernet/dec/tulip/21142.c
+++ b/drivers/net/ethernet/dec/tulip/21142.c
@@ -216,7 +216,7 @@ void t21142_lnk_change(struct net_device *dev, int csr5)
 		    (csr12 & 2) == 2) ||
 		   (tp->nway && (csr5 & (TPLnkFail)))) {
 		/* Link blew? Maybe restart NWay. */
-		del_timer_sync(&tp->timer);
+		timer_delete_sync(&tp->timer);
 		t21142_start_nway(dev);
 		tp->timer.expires = RUN_AT(3*HZ);
 		add_timer(&tp->timer);
@@ -226,7 +226,7 @@ void t21142_lnk_change(struct net_device *dev, int csr5)
 				 medianame[dev->if_port],
 				 (csr12 & 2) ? "failed" : "good");
 		if ((csr12 & 2)  &&  ! tp->medialock) {
-			del_timer_sync(&tp->timer);
+			timer_delete_sync(&tp->timer);
 			t21142_start_nway(dev);
 			tp->timer.expires = RUN_AT(3*HZ);
 			add_timer(&tp->timer);
diff --git a/drivers/net/ethernet/dec/tulip/Kconfig b/drivers/net/ethernet/dec/tulip/Kconfig
index 1003201b5d80..078a12f07e96 100644
--- a/drivers/net/ethernet/dec/tulip/Kconfig
+++ b/drivers/net/ethernet/dec/tulip/Kconfig
@@ -1,3 +1,4 @@
+# SPDX-License-Identifier: GPL-2.0-only
 #
 # Tulip family network device configuration
 #
@@ -5,7 +6,7 @@
 config NET_TULIP
 	bool "DEC - Tulip devices"
 	depends on (PCI || EISA || CARDBUS)
-	---help---
+	help
 	  This selects the "Tulip" family of EISA/PCI network cards.
 
 if NET_TULIP
@@ -14,7 +15,7 @@ config DE2104X
 	tristate "Early DECchip Tulip (dc2104x) PCI support"
 	depends on PCI
 	select CRC32
-	---help---
+	help
 	  This driver is developed for the SMC EtherPower series Ethernet
 	  cards and also works with cards based on the DECchip
 	  21040 (Tulip series) chips.  Some LinkSys PCI cards are
@@ -31,7 +32,7 @@ config DE2104X_DSL
 	depends on DE2104X
 	range 0 31
 	default 0
-	---help---
+	help
 	  Setting this value allows to align ring buffer descriptors into their
 	  own cache lines. Value of 4 corresponds to the typical 32 byte line
 	  (the descriptor is 16 bytes). This is necessary on systems that lack
@@ -42,7 +43,7 @@ config TULIP
 	tristate "DECchip Tulip (dc2114x) PCI support"
 	depends on PCI
 	select CRC32
-	---help---
+	help
 	  This driver is developed for the SMC EtherPower series Ethernet
 	  cards and also works with cards based on the DECchip 
 	  21140 (Tulip series) chips.  Some LinkSys PCI cards are
@@ -57,7 +58,7 @@ config TULIP
 config TULIP_MWI
 	bool "New bus configuration"
 	depends on TULIP
-	---help---
+	help
 	  This configures your Tulip card specifically for the card and
 	  system cache line size type you are using.
 
@@ -68,7 +69,7 @@ config TULIP_MWI
 config TULIP_MMIO
 	bool "Use PCI shared mem for NIC registers"
 	depends on TULIP
-	---help---
+	help
 	  Use PCI shared memory for the NIC registers, rather than going through
 	  the Tulip's PIO (programmed I/O ports).  Faster, but could produce
 	  obscure bugs if your mainboard has memory controller timing issues.
@@ -77,7 +78,7 @@ config TULIP_MMIO
 config TULIP_NAPI
 	bool "Use RX polling (NAPI)"
 	depends on TULIP
-	---help---
+	help
 	  NAPI is a new driver API designed to reduce CPU and interrupt load
 	  when the driver is receiving lots of packets from the card. It is
 	  still somewhat experimental and thus not yet enabled by default.
@@ -91,7 +92,7 @@ config TULIP_NAPI
 config TULIP_NAPI_HW_MITIGATION
 	bool "Use Interrupt Mitigation"
 	depends on TULIP_NAPI
-	---help---
+	help
 	  Use HW to reduce RX interrupts. Not strictly necessary since NAPI
 	  reduces RX interrupts by itself. Interrupt mitigation reduces RX
 	  interrupts even at low levels of traffic at the cost of a small
@@ -103,27 +104,12 @@ config TULIP_DM910X
 	def_bool y
 	depends on TULIP && SPARC
 
-config DE4X5
-	tristate "Generic DECchip & DIGITAL EtherWORKS PCI/EISA"
-	depends on (PCI || EISA)
-	depends on VIRT_TO_BUS || ALPHA || PPC || SPARC
-	select CRC32
-	---help---
-	  This is support for the DIGITAL series of PCI/EISA Ethernet cards.
-	  These include the DE425, DE434, DE435, DE450 and DE500 models.  If
-	  you have a network card of this type, say Y.  More specific
-	  information is contained in
-	  <file:Documentation/networking/de4x5.txt>.
-
-	  To compile this driver as a module, choose M here. The module will
-	  be called de4x5.
-
 config WINBOND_840
 	tristate "Winbond W89c840 Ethernet support"
 	depends on PCI
 	select CRC32
 	select MII
-	---help---
+	help
 	  This driver is for the Winbond W89c840 chip.  It also works with 
 	  the TX9882 chip on the Compex RL100-ATX board.
 	  More specific information and updates are available from
@@ -133,11 +119,11 @@ config DM9102
 	tristate "Davicom DM910x/DM980x support"
 	depends on PCI
 	select CRC32
-	---help---
+	help
 	  This driver is for DM9102(A)/DM9132/DM9801 compatible PCI cards from
 	  Davicom (<http://www.davicom.com.tw/>).  If you have such a network
 	  (Ethernet) card, say Y.  Some information is contained in the file
-	  <file:Documentation/networking/dmfe.txt>.
+	  <file:Documentation/networking/device_drivers/ethernet/dec/dmfe.rst>.
 
 	  To compile this driver as a module, choose M here. The module will
 	  be called dmfe.
@@ -146,7 +132,7 @@ config ULI526X
 	tristate "ULi M526x controller support"
 	depends on PCI
 	select CRC32
-	---help---
+	help
 	  This driver is for ULi M5261/M5263 10/100M Ethernet Controller
 	  (<http://www.nvidia.com/page/uli_drivers.html>).
 
@@ -156,7 +142,7 @@ config ULI526X
 config PCMCIA_XIRCOM
 	tristate "Xircom CardBus support"
 	depends on CARDBUS
-	---help---
+	help
 	  This driver is for the Digital "Tulip" Ethernet CardBus adapters.
 	  It should work with most DEC 21*4*-based chips/ethercards, as well
 	  as with work-alike chips from Lite-On (PNIC) and Macronix (MXIC) and
diff --git a/drivers/net/ethernet/dec/tulip/Makefile b/drivers/net/ethernet/dec/tulip/Makefile
index 5e8be38b45bb..d4f1d21d29a0 100644
--- a/drivers/net/ethernet/dec/tulip/Makefile
+++ b/drivers/net/ethernet/dec/tulip/Makefile
@@ -1,3 +1,4 @@
+# SPDX-License-Identifier: GPL-2.0
 #
 # Makefile for the Linux "Tulip" family network device drivers.
 #
@@ -9,7 +10,6 @@ obj-$(CONFIG_DM9102)		+= dmfe.o
 obj-$(CONFIG_WINBOND_840)	+= winbond-840.o
 obj-$(CONFIG_DE2104X)		+= de2104x.o
 obj-$(CONFIG_TULIP)		+= tulip.o
-obj-$(CONFIG_DE4X5)		+= de4x5.o
 obj-$(CONFIG_ULI526X)		+= uli526x.o
 
 # Declare multi-part drivers.
diff --git a/drivers/net/ethernet/dec/tulip/de2104x.c b/drivers/net/ethernet/dec/tulip/de2104x.c
index 91b8f6f5a765..f9504f340c4a 100644
--- a/drivers/net/ethernet/dec/tulip/de2104x.c
+++ b/drivers/net/ethernet/dec/tulip/de2104x.c
@@ -30,7 +30,6 @@
 #define pr_fmt(fmt) KBUILD_MODNAME ": " fmt
 
 #define DRV_NAME		"de2104x"
-#define DRV_VERSION		"0.7"
 #define DRV_RELDATE		"Mar 17, 2004"
 
 #include <linux/module.h>
@@ -50,16 +49,11 @@
 #include <asm/io.h>
 #include <asm/irq.h>
 #include <linux/uaccess.h>
-#include <asm/unaligned.h>
-
-/* These identify the driver base version and may not be removed. */
-static char version[] =
-"PCI Ethernet driver v" DRV_VERSION " (" DRV_RELDATE ")";
+#include <linux/unaligned.h>
 
 MODULE_AUTHOR("Jeff Garzik <jgarzik@pobox.com>");
 MODULE_DESCRIPTION("Intel/Digital 21040/1 series PCI Ethernet driver");
 MODULE_LICENSE("GPL");
-MODULE_VERSION(DRV_VERSION);
 
 static int debug = -1;
 module_param (debug, int, 0);
@@ -91,7 +85,7 @@ MODULE_PARM_DESC (rx_copybreak, "de2104x Breakpoint at which Rx packets are copi
 #define DSL			CONFIG_DE2104X_DSL
 #endif
 
-#define DE_RX_RING_SIZE		64
+#define DE_RX_RING_SIZE		128
 #define DE_TX_RING_SIZE		64
 #define DE_RING_BYTES		\
 		((sizeof(struct de_desc) * DE_RX_RING_SIZE) +	\
@@ -333,8 +327,8 @@ static void de_set_rx_mode (struct net_device *dev);
 static void de_tx (struct de_private *de);
 static void de_clean_rings (struct de_private *de);
 static void de_media_interrupt (struct de_private *de, u32 status);
-static void de21040_media_timer (unsigned long data);
-static void de21041_media_timer (unsigned long data);
+static void de21040_media_timer (struct timer_list *t);
+static void de21041_media_timer (struct timer_list *t);
 static unsigned int de_ok_to_advertise (struct de_private *de, u32 new_media);
 
 
@@ -417,7 +411,10 @@ static void de_rx (struct de_private *de)
 		if (status & DescOwn)
 			break;
 
-		len = ((status >> 16) & 0x7ff) - 4;
+		/* the length is actually a 15 bit value here according
+		 * to Table 4-1 in the DE2104x spec so mask is 0x7fff
+		 */
+		len = ((status >> 16) & 0x7fff) - 4;
 		mapping = de->rx_skb[rx_tail].mapping;
 
 		if (unlikely(drop)) {
@@ -446,21 +443,23 @@ static void de_rx (struct de_private *de)
 		}
 
 		if (!copying_skb) {
-			pci_unmap_single(de->pdev, mapping,
-					 buflen, PCI_DMA_FROMDEVICE);
+			dma_unmap_single(&de->pdev->dev, mapping, buflen,
+					 DMA_FROM_DEVICE);
 			skb_put(skb, len);
 
 			mapping =
 			de->rx_skb[rx_tail].mapping =
-				pci_map_single(de->pdev, copy_skb->data,
-					       buflen, PCI_DMA_FROMDEVICE);
+				dma_map_single(&de->pdev->dev, copy_skb->data,
+					       buflen, DMA_FROM_DEVICE);
 			de->rx_skb[rx_tail].skb = copy_skb;
 		} else {
-			pci_dma_sync_single_for_cpu(de->pdev, mapping, len, PCI_DMA_FROMDEVICE);
+			dma_sync_single_for_cpu(&de->pdev->dev, mapping, len,
+						DMA_FROM_DEVICE);
 			skb_reserve(copy_skb, RX_OFFSET);
 			skb_copy_from_linear_data(skb, skb_put(copy_skb, len),
 						  len);
-			pci_dma_sync_single_for_device(de->pdev, mapping, len, PCI_DMA_FROMDEVICE);
+			dma_sync_single_for_device(&de->pdev->dev, mapping,
+						   len, DMA_FROM_DEVICE);
 
 			/* We'll reuse the original ring buffer. */
 			skb = copy_skb;
@@ -557,13 +556,15 @@ static void de_tx (struct de_private *de)
 			goto next;
 
 		if (unlikely(skb == DE_SETUP_SKB)) {
-			pci_unmap_single(de->pdev, de->tx_skb[tx_tail].mapping,
-					 sizeof(de->setup_frame), PCI_DMA_TODEVICE);
+			dma_unmap_single(&de->pdev->dev,
+					 de->tx_skb[tx_tail].mapping,
+					 sizeof(de->setup_frame),
+					 DMA_TO_DEVICE);
 			goto next;
 		}
 
-		pci_unmap_single(de->pdev, de->tx_skb[tx_tail].mapping,
-				 skb->len, PCI_DMA_TODEVICE);
+		dma_unmap_single(&de->pdev->dev, de->tx_skb[tx_tail].mapping,
+				 skb->len, DMA_TO_DEVICE);
 
 		if (status & LastFrag) {
 			if (status & TxError) {
@@ -585,7 +586,7 @@ static void de_tx (struct de_private *de)
 				netif_dbg(de, tx_done, de->dev,
 					  "tx done, slot %d\n", tx_tail);
 			}
-			dev_kfree_skb_irq(skb);
+			dev_consume_skb_irq(skb);
 		}
 
 next:
@@ -623,7 +624,8 @@ static netdev_tx_t de_start_xmit (struct sk_buff *skb,
 	txd = &de->tx_ring[entry];
 
 	len = skb->len;
-	mapping = pci_map_single(de->pdev, skb->data, len, PCI_DMA_TODEVICE);
+	mapping = dma_map_single(&de->pdev->dev, skb->data, len,
+				 DMA_TO_DEVICE);
 	if (entry == (DE_TX_RING_SIZE - 1))
 		flags |= RingEnd;
 	if (!tx_free || (tx_free == (DE_TX_RING_SIZE / 2)))
@@ -664,8 +666,8 @@ static void build_setup_frame_hash(u16 *setup_frm, struct net_device *dev)
 	struct de_private *de = netdev_priv(dev);
 	u16 hash_table[32];
 	struct netdev_hw_addr *ha;
+	const u16 *eaddrs;
 	int i;
-	u16 *eaddrs;
 
 	memset(hash_table, 0, sizeof(hash_table));
 	__set_bit_le(255, hash_table);			/* Broadcast entry */
@@ -683,7 +685,7 @@ static void build_setup_frame_hash(u16 *setup_frm, struct net_device *dev)
 	setup_frm = &de->setup_frame[13*6];
 
 	/* Fill the final entry with our physical address. */
-	eaddrs = (u16 *)dev->dev_addr;
+	eaddrs = (const u16 *)dev->dev_addr;
 	*setup_frm++ = eaddrs[0]; *setup_frm++ = eaddrs[0];
 	*setup_frm++ = eaddrs[1]; *setup_frm++ = eaddrs[1];
 	*setup_frm++ = eaddrs[2]; *setup_frm++ = eaddrs[2];
@@ -693,7 +695,7 @@ static void build_setup_frame_perfect(u16 *setup_frm, struct net_device *dev)
 {
 	struct de_private *de = netdev_priv(dev);
 	struct netdev_hw_addr *ha;
-	u16 *eaddrs;
+	const u16 *eaddrs;
 
 	/* We have <= 14 addresses so we can use the wonderful
 	   16 address perfect filtering of the Tulip. */
@@ -708,7 +710,7 @@ static void build_setup_frame_perfect(u16 *setup_frm, struct net_device *dev)
 	setup_frm = &de->setup_frame[15*6];
 
 	/* Fill the final entry with our physical address. */
-	eaddrs = (u16 *)dev->dev_addr;
+	eaddrs = (const u16 *)dev->dev_addr;
 	*setup_frm++ = eaddrs[0]; *setup_frm++ = eaddrs[0];
 	*setup_frm++ = eaddrs[1]; *setup_frm++ = eaddrs[1];
 	*setup_frm++ = eaddrs[2]; *setup_frm++ = eaddrs[2];
@@ -766,8 +768,8 @@ static void __de_set_rx_mode (struct net_device *dev)
 
 	de->tx_skb[entry].skb = DE_SETUP_SKB;
 	de->tx_skb[entry].mapping = mapping =
-	    pci_map_single (de->pdev, de->setup_frame,
-			    sizeof (de->setup_frame), PCI_DMA_TODEVICE);
+	    dma_map_single(&de->pdev->dev, de->setup_frame,
+			   sizeof(de->setup_frame), DMA_TO_DEVICE);
 
 	/* Put the setup frame on the Tx list. */
 	txd = &de->tx_ring[entry];
@@ -830,8 +832,8 @@ static struct net_device_stats *de_get_stats(struct net_device *dev)
 
 	/* The chip only need report frame silently dropped. */
 	spin_lock_irq(&de->lock);
- 	if (netif_running(dev) && netif_device_present(dev))
- 		__de_get_stats(de);
+	if (netif_running(dev) && netif_device_present(dev))
+		__de_get_stats(de);
 	spin_unlock_irq(&de->lock);
 
 	return &dev->stats;
@@ -959,9 +961,9 @@ static void de_next_media (struct de_private *de, const u32 *media,
 	}
 }
 
-static void de21040_media_timer (unsigned long data)
+static void de21040_media_timer (struct timer_list *t)
 {
-	struct de_private *de = (struct de_private *) data;
+	struct de_private *de = timer_container_of(de, t, media_timer);
 	struct net_device *dev = de->dev;
 	u32 status = dr32(SIAStatus);
 	unsigned int carrier;
@@ -1040,9 +1042,9 @@ static unsigned int de_ok_to_advertise (struct de_private *de, u32 new_media)
 	return 1;
 }
 
-static void de21041_media_timer (unsigned long data)
+static void de21041_media_timer (struct timer_list *t)
 {
-	struct de_private *de = (struct de_private *) data;
+	struct de_private *de = timer_container_of(de, t, media_timer);
 	struct net_device *dev = de->dev;
 	u32 status = dr32(SIAStatus);
 	unsigned int carrier;
@@ -1282,8 +1284,10 @@ static int de_refill_rx (struct de_private *de)
 		if (!skb)
 			goto err_out;
 
-		de->rx_skb[i].mapping = pci_map_single(de->pdev,
-			skb->data, de->rx_buf_sz, PCI_DMA_FROMDEVICE);
+		de->rx_skb[i].mapping = dma_map_single(&de->pdev->dev,
+						       skb->data,
+						       de->rx_buf_sz,
+						       DMA_FROM_DEVICE);
 		de->rx_skb[i].skb = skb;
 
 		de->rx_ring[i].opts1 = cpu_to_le32(DescOwn);
@@ -1316,7 +1320,8 @@ static int de_init_rings (struct de_private *de)
 
 static int de_alloc_rings (struct de_private *de)
 {
-	de->rx_ring = pci_alloc_consistent(de->pdev, DE_RING_BYTES, &de->ring_dma);
+	de->rx_ring = dma_alloc_coherent(&de->pdev->dev, DE_RING_BYTES,
+					 &de->ring_dma, GFP_KERNEL);
 	if (!de->rx_ring)
 		return -ENOMEM;
 	de->tx_ring = &de->rx_ring[DE_RX_RING_SIZE];
@@ -1336,8 +1341,9 @@ static void de_clean_rings (struct de_private *de)
 
 	for (i = 0; i < DE_RX_RING_SIZE; i++) {
 		if (de->rx_skb[i].skb) {
-			pci_unmap_single(de->pdev, de->rx_skb[i].mapping,
-					 de->rx_buf_sz, PCI_DMA_FROMDEVICE);
+			dma_unmap_single(&de->pdev->dev,
+					 de->rx_skb[i].mapping, de->rx_buf_sz,
+					 DMA_FROM_DEVICE);
 			dev_kfree_skb(de->rx_skb[i].skb);
 		}
 	}
@@ -1347,15 +1353,15 @@ static void de_clean_rings (struct de_private *de)
 		if ((skb) && (skb != DE_DUMMY_SKB)) {
 			if (skb != DE_SETUP_SKB) {
 				de->dev->stats.tx_dropped++;
-				pci_unmap_single(de->pdev,
-					de->tx_skb[i].mapping,
-					skb->len, PCI_DMA_TODEVICE);
+				dma_unmap_single(&de->pdev->dev,
+						 de->tx_skb[i].mapping,
+						 skb->len, DMA_TO_DEVICE);
 				dev_kfree_skb(skb);
 			} else {
-				pci_unmap_single(de->pdev,
-					de->tx_skb[i].mapping,
-					sizeof(de->setup_frame),
-					PCI_DMA_TODEVICE);
+				dma_unmap_single(&de->pdev->dev,
+						 de->tx_skb[i].mapping,
+						 sizeof(de->setup_frame),
+						 DMA_TO_DEVICE);
 			}
 		}
 	}
@@ -1367,7 +1373,8 @@ static void de_clean_rings (struct de_private *de)
 static void de_free_rings (struct de_private *de)
 {
 	de_clean_rings(de);
-	pci_free_consistent(de->pdev, DE_RING_BYTES, de->rx_ring, de->ring_dma);
+	dma_free_coherent(&de->pdev->dev, DE_RING_BYTES, de->rx_ring,
+			  de->ring_dma);
 	de->rx_ring = NULL;
 	de->tx_ring = NULL;
 }
@@ -1421,7 +1428,7 @@ static int de_close (struct net_device *dev)
 
 	netif_dbg(de, ifdown, dev, "disabling interface\n");
 
-	del_timer_sync(&de->media_timer);
+	timer_delete_sync(&de->media_timer);
 
 	spin_lock_irqsave(&de->lock, flags);
 	de_stop_hw(de);
@@ -1436,7 +1443,7 @@ static int de_close (struct net_device *dev)
 	return 0;
 }
 
-static void de_tx_timeout (struct net_device *dev)
+static void de_tx_timeout (struct net_device *dev, unsigned int txqueue)
 {
 	struct de_private *de = netdev_priv(dev);
 	const int irq = de->pdev->irq;
@@ -1445,7 +1452,7 @@ static void de_tx_timeout (struct net_device *dev)
 		   dr32(MacStatus), dr32(MacMode), dr32(SIAStatus),
 		   de->rx_tail, de->tx_head, de->tx_tail);
 
-	del_timer_sync(&de->media_timer);
+	timer_delete_sync(&de->media_timer);
 
 	disable_irq(irq);
 	spin_lock_irq(&de->lock);
@@ -1483,8 +1490,8 @@ static void __de_get_regs(struct de_private *de, u8 *buf)
 	de_rx_missed(de, rbuf[8]);
 }
 
-static int __de_get_link_ksettings(struct de_private *de,
-				   struct ethtool_link_ksettings *cmd)
+static void __de_get_link_ksettings(struct de_private *de,
+				    struct ethtool_link_ksettings *cmd)
 {
 	ethtool_convert_legacy_u32_to_link_mode(cmd->link_modes.supported,
 						de->media_supported);
@@ -1517,8 +1524,6 @@ static int __de_get_link_ksettings(struct de_private *de,
 		cmd->base.autoneg = AUTONEG_ENABLE;
 
 	/* ignore maxtxpkt, maxrxpkt for now */
-
-	return 0;
 }
 
 static int __de_set_link_ksettings(struct de_private *de,
@@ -1601,9 +1606,8 @@ static void de_get_drvinfo (struct net_device *dev,struct ethtool_drvinfo *info)
 {
 	struct de_private *de = netdev_priv(dev);
 
-	strlcpy(info->driver, DRV_NAME, sizeof(info->driver));
-	strlcpy(info->version, DRV_VERSION, sizeof(info->version));
-	strlcpy(info->bus_info, pci_name(de->pdev), sizeof(info->bus_info));
+	strscpy(info->driver, DRV_NAME, sizeof(info->driver));
+	strscpy(info->bus_info, pci_name(de->pdev), sizeof(info->bus_info));
 }
 
 static int de_get_regs_len(struct net_device *dev)
@@ -1615,13 +1619,12 @@ static int de_get_link_ksettings(struct net_device *dev,
 				 struct ethtool_link_ksettings *cmd)
 {
 	struct de_private *de = netdev_priv(dev);
-	int rc;
 
 	spin_lock_irq(&de->lock);
-	rc = __de_get_link_ksettings(de, cmd);
+	__de_get_link_ksettings(de, cmd);
 	spin_unlock_irq(&de->lock);
 
-	return rc;
+	return 0;
 }
 
 static int de_set_link_ksettings(struct net_device *dev,
@@ -1710,6 +1713,7 @@ static const struct ethtool_ops de_ethtool_ops = {
 
 static void de21040_get_mac_address(struct de_private *de)
 {
+	u8 addr[ETH_ALEN];
 	unsigned i;
 
 	dw32 (ROMCmd, 0);	/* Reset the pointer with a dummy write. */
@@ -1721,12 +1725,13 @@ static void de21040_get_mac_address(struct de_private *de)
 			value = dr32(ROMCmd);
 			rmb();
 		} while (value < 0 && --boguscnt > 0);
-		de->dev->dev_addr[i] = value;
+		addr[i] = value;
 		udelay(1);
 		if (boguscnt <= 0)
 			pr_warn("timeout reading 21040 MAC address byte %u\n",
 				i);
 	}
+	eth_hw_addr_set(de->dev, addr);
 }
 
 static void de21040_get_media_info(struct de_private *de)
@@ -1818,8 +1823,7 @@ static void de21041_get_srom_info(struct de_private *de)
 #endif
 
 	/* store MAC address */
-	for (i = 0; i < 6; i ++)
-		de->dev->dev_addr[i] = ee_data[i + sa_offset];
+	eth_hw_addr_set(de->dev, &ee_data[sa_offset]);
 
 	/* get offset of controller 0 info leaf.  ignore 2nd byte. */
 	ofs = ee_data[SROMC0InfoLeaf];
@@ -1980,11 +1984,6 @@ static int de_init_one(struct pci_dev *pdev, const struct pci_device_id *ent)
 
 	board_idx++;
 
-#ifndef MODULE
-	if (board_idx == 0)
-		pr_info("%s\n", version);
-#endif
-
 	/* allocate a new ethernet device structure, and fill in defaults */
 	dev = alloc_etherdev(sizeof(struct de_private));
 	if (!dev)
@@ -2002,12 +2001,9 @@ static int de_init_one(struct pci_dev *pdev, const struct pci_device_id *ent)
 	de->msg_enable = (debug < 0 ? DE_DEF_MSG_ENABLE : debug);
 	de->board_idx = board_idx;
 	spin_lock_init (&de->lock);
-	init_timer(&de->media_timer);
-	if (de->de21040)
-		de->media_timer.function = de21040_media_timer;
-	else
-		de->media_timer.function = de21041_media_timer;
-	de->media_timer.data = (unsigned long) de;
+	timer_setup(&de->media_timer,
+		    de->de21040 ? de21040_media_timer : de21041_media_timer,
+		    0);
 
 	netif_carrier_off(dev);
 
@@ -2045,7 +2041,7 @@ static int de_init_one(struct pci_dev *pdev, const struct pci_device_id *ent)
 	}
 
 	/* remap CSR registers */
-	regs = ioremap_nocache(pciaddr, DE_REGS_SIZE);
+	regs = ioremap(pciaddr, DE_REGS_SIZE);
 	if (!regs) {
 		rc = -EIO;
 		pr_err("Cannot map PCI MMIO (%llx@%lx) on pci dev %s\n",
@@ -2120,18 +2116,17 @@ static void de_remove_one(struct pci_dev *pdev)
 	free_netdev(dev);
 }
 
-#ifdef CONFIG_PM
-
-static int de_suspend (struct pci_dev *pdev, pm_message_t state)
+static int __maybe_unused de_suspend(struct device *dev_d)
 {
-	struct net_device *dev = pci_get_drvdata (pdev);
+	struct pci_dev *pdev = to_pci_dev(dev_d);
+	struct net_device *dev = pci_get_drvdata(pdev);
 	struct de_private *de = netdev_priv(dev);
 
 	rtnl_lock();
 	if (netif_running (dev)) {
 		const int irq = pdev->irq;
 
-		del_timer_sync(&de->media_timer);
+		timer_delete_sync(&de->media_timer);
 
 		disable_irq(irq);
 		spin_lock_irq(&de->lock);
@@ -2151,7 +2146,6 @@ static int de_suspend (struct pci_dev *pdev, pm_message_t state)
 		de_clean_rings(de);
 
 		de_adapter_sleep(de);
-		pci_disable_device(pdev);
 	} else {
 		netif_device_detach(dev);
 	}
@@ -2159,21 +2153,17 @@ static int de_suspend (struct pci_dev *pdev, pm_message_t state)
 	return 0;
 }
 
-static int de_resume (struct pci_dev *pdev)
+static int __maybe_unused de_resume(struct device *dev_d)
 {
-	struct net_device *dev = pci_get_drvdata (pdev);
+	struct pci_dev *pdev = to_pci_dev(dev_d);
+	struct net_device *dev = pci_get_drvdata(pdev);
 	struct de_private *de = netdev_priv(dev);
-	int retval = 0;
 
 	rtnl_lock();
 	if (netif_device_present(dev))
 		goto out;
 	if (!netif_running(dev))
 		goto out_attach;
-	if ((retval = pci_enable_device(pdev))) {
-		netdev_err(dev, "pci_enable_device failed in resume\n");
-		goto out;
-	}
 	pci_set_master(pdev);
 	de_init_rings(de);
 	de_init_hw(de);
@@ -2184,31 +2174,24 @@ out:
 	return 0;
 }
 
-#endif /* CONFIG_PM */
+static SIMPLE_DEV_PM_OPS(de_pm_ops, de_suspend, de_resume);
+
+static void de_shutdown(struct pci_dev *pdev)
+{
+	struct net_device *dev = pci_get_drvdata(pdev);
+
+	rtnl_lock();
+	dev_close(dev);
+	rtnl_unlock();
+}
 
 static struct pci_driver de_driver = {
 	.name		= DRV_NAME,
 	.id_table	= de_pci_tbl,
 	.probe		= de_init_one,
 	.remove		= de_remove_one,
-#ifdef CONFIG_PM
-	.suspend	= de_suspend,
-	.resume		= de_resume,
-#endif
+	.shutdown	= de_shutdown,
+	.driver.pm	= &de_pm_ops,
 };
 
-static int __init de_init (void)
-{
-#ifdef MODULE
-	pr_info("%s\n", version);
-#endif
-	return pci_register_driver(&de_driver);
-}
-
-static void __exit de_exit (void)
-{
-	pci_unregister_driver (&de_driver);
-}
-
-module_init(de_init);
-module_exit(de_exit);
+module_pci_driver(de_driver);
diff --git a/drivers/net/ethernet/dec/tulip/de4x5.c b/drivers/net/ethernet/dec/tulip/de4x5.c
deleted file mode 100644
index fd6bcf024729..000000000000
--- a/drivers/net/ethernet/dec/tulip/de4x5.c
+++ /dev/null
@@ -1,5577 +0,0 @@
-/*  de4x5.c: A DIGITAL DC21x4x DECchip and DE425/DE434/DE435/DE450/DE500
-             ethernet driver for Linux.
-
-    Copyright 1994, 1995 Digital Equipment Corporation.
-
-    Testing resources for this driver have been made available
-    in part by NASA Ames Research Center (mjacob@nas.nasa.gov).
-
-    The author may be reached at davies@maniac.ultranet.com.
-
-    This program is free software; you can redistribute  it and/or modify it
-    under  the terms of  the GNU General  Public License as published by the
-    Free Software Foundation;  either version 2 of the  License, or (at your
-    option) any later version.
-
-    THIS  SOFTWARE  IS PROVIDED   ``AS  IS'' AND   ANY  EXPRESS OR   IMPLIED
-    WARRANTIES,   INCLUDING, BUT NOT  LIMITED  TO, THE IMPLIED WARRANTIES OF
-    MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE DISCLAIMED.  IN
-    NO  EVENT  SHALL   THE AUTHOR  BE    LIABLE FOR ANY   DIRECT,  INDIRECT,
-    INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT
-    NOT LIMITED   TO, PROCUREMENT OF  SUBSTITUTE GOODS  OR SERVICES; LOSS OF
-    USE, DATA,  OR PROFITS; OR  BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON
-    ANY THEORY OF LIABILITY, WHETHER IN  CONTRACT, STRICT LIABILITY, OR TORT
-    (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF
-    THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
-
-    You should have received a copy of the  GNU General Public License along
-    with this program; if not, write  to the Free Software Foundation, Inc.,
-    675 Mass Ave, Cambridge, MA 02139, USA.
-
-    Originally,   this  driver  was    written  for the  Digital   Equipment
-    Corporation series of EtherWORKS ethernet cards:
-
-        DE425 TP/COAX EISA
-	DE434 TP PCI
-	DE435 TP/COAX/AUI PCI
-	DE450 TP/COAX/AUI PCI
-	DE500 10/100 PCI Fasternet
-
-    but it  will  now attempt  to  support all  cards which   conform to the
-    Digital Semiconductor   SROM   Specification.    The  driver   currently
-    recognises the following chips:
-
-        DC21040  (no SROM)
-	DC21041[A]
-	DC21140[A]
-	DC21142
-	DC21143
-
-    So far the driver is known to work with the following cards:
-
-        KINGSTON
-	Linksys
-	ZNYX342
-	SMC8432
-	SMC9332 (w/new SROM)
-	ZNYX31[45]
-	ZNYX346 10/100 4 port (can act as a 10/100 bridge!)
-
-    The driver has been tested on a relatively busy network using the DE425,
-    DE434, DE435 and DE500 cards and benchmarked with 'ttcp': it transferred
-    16M of data to a DECstation 5000/200 as follows:
-
-                TCP           UDP
-             TX     RX     TX     RX
-    DE425   1030k  997k   1170k  1128k
-    DE434   1063k  995k   1170k  1125k
-    DE435   1063k  995k   1170k  1125k
-    DE500   1063k  998k   1170k  1125k  in 10Mb/s mode
-
-    All  values are typical (in   kBytes/sec) from a  sample  of 4 for  each
-    measurement. Their error is +/-20k on a quiet (private) network and also
-    depend on what load the CPU has.
-
-    =========================================================================
-    This driver  has been written substantially  from  scratch, although its
-    inheritance of style and stack interface from 'ewrk3.c' and in turn from
-    Donald Becker's 'lance.c' should be obvious. With the module autoload of
-    every  usable DECchip board,  I  pinched Donald's 'next_module' field to
-    link my modules together.
-
-    Up to 15 EISA cards can be supported under this driver, limited primarily
-    by the available IRQ lines.  I have  checked different configurations of
-    multiple depca, EtherWORKS 3 cards and de4x5 cards and  have not found a
-    problem yet (provided you have at least depca.c v0.38) ...
-
-    PCI support has been added  to allow the driver  to work with the DE434,
-    DE435, DE450 and DE500 cards. The I/O accesses are a bit of a kludge due
-    to the differences in the EISA and PCI CSR address offsets from the base
-    address.
-
-    The ability to load this  driver as a loadable  module has been included
-    and used extensively  during the driver development  (to save those long
-    reboot sequences).  Loadable module support  under PCI and EISA has been
-    achieved by letting the driver autoprobe as if it were compiled into the
-    kernel. Do make sure  you're not sharing  interrupts with anything  that
-    cannot accommodate  interrupt  sharing!
-
-    To utilise this ability, you have to do 8 things:
-
-    0) have a copy of the loadable modules code installed on your system.
-    1) copy de4x5.c from the  /linux/drivers/net directory to your favourite
-    temporary directory.
-    2) for fixed  autoprobes (not  recommended),  edit the source code  near
-    line 5594 to reflect the I/O address  you're using, or assign these when
-    loading by:
-
-                   insmod de4x5 io=0xghh           where g = bus number
-		                                        hh = device number
-
-       NB: autoprobing for modules is now supported by default. You may just
-           use:
-
-                   insmod de4x5
-
-           to load all available boards. For a specific board, still use
-	   the 'io=?' above.
-    3) compile  de4x5.c, but include -DMODULE in  the command line to ensure
-    that the correct bits are compiled (see end of source code).
-    4) if you are wanting to add a new  card, goto 5. Otherwise, recompile a
-    kernel with the de4x5 configuration turned off and reboot.
-    5) insmod de4x5 [io=0xghh]
-    6) run the net startup bits for your new eth?? interface(s) manually
-    (usually /etc/rc.inet[12] at boot time).
-    7) enjoy!
-
-    To unload a module, turn off the associated interface(s)
-    'ifconfig eth?? down' then 'rmmod de4x5'.
-
-    Automedia detection is included so that in  principal you can disconnect
-    from, e.g.  TP, reconnect  to BNC  and  things will still work  (after a
-    pause whilst the   driver figures out   where its media went).  My tests
-    using ping showed that it appears to work....
-
-    By  default,  the driver will  now   autodetect any  DECchip based card.
-    Should you have a need to restrict the driver to DIGITAL only cards, you
-    can compile with a  DEC_ONLY define, or if  loading as a module, use the
-    'dec_only=1'  parameter.
-
-    I've changed the timing routines to  use the kernel timer and scheduling
-    functions  so that the  hangs  and other assorted problems that occurred
-    while autosensing the  media  should be gone.  A  bonus  for the DC21040
-    auto  media sense algorithm is  that it can now  use one that is more in
-    line with the  rest (the DC21040  chip doesn't  have a hardware  timer).
-    The downside is the 1 'jiffies' (10ms) resolution.
-
-    IEEE 802.3u MII interface code has  been added in anticipation that some
-    products may use it in the future.
-
-    The SMC9332 card  has a non-compliant SROM  which needs fixing -  I have
-    patched this  driver to detect it  because the SROM format used complies
-    to a previous DEC-STD format.
-
-    I have removed the buffer copies needed for receive on Intels.  I cannot
-    remove them for   Alphas since  the  Tulip hardware   only does longword
-    aligned  DMA transfers  and  the  Alphas get   alignment traps with  non
-    longword aligned data copies (which makes them really slow). No comment.
-
-    I  have added SROM decoding  routines to make this  driver work with any
-    card that  supports the Digital  Semiconductor SROM spec. This will help
-    all  cards running the dc2114x  series chips in particular.  Cards using
-    the dc2104x  chips should run correctly with  the basic  driver.  I'm in
-    debt to <mjacob@feral.com> for the  testing and feedback that helped get
-    this feature working.  So far we have  tested KINGSTON, SMC8432, SMC9332
-    (with the latest SROM complying  with the SROM spec  V3: their first was
-    broken), ZNYX342  and  LinkSys. ZYNX314 (dual  21041  MAC) and  ZNYX 315
-    (quad 21041 MAC)  cards also  appear  to work despite their  incorrectly
-    wired IRQs.
-
-    I have added a temporary fix for interrupt problems when some SCSI cards
-    share the same interrupt as the DECchip based  cards. The problem occurs
-    because  the SCSI card wants to  grab the interrupt  as a fast interrupt
-    (runs the   service routine with interrupts turned   off) vs.  this card
-    which really needs to run the service routine with interrupts turned on.
-    This driver will  now   add the interrupt service   routine  as  a  fast
-    interrupt if it   is bounced from the   slow interrupt.  THIS IS NOT   A
-    RECOMMENDED WAY TO RUN THE DRIVER  and has been done  for a limited time
-    until  people   sort  out their  compatibility    issues and the  kernel
-    interrupt  service code  is  fixed.   YOU  SHOULD SEPARATE OUT  THE FAST
-    INTERRUPT CARDS FROM THE SLOW INTERRUPT CARDS to ensure that they do not
-    run on the same interrupt. PCMCIA/CardBus is another can of worms...
-
-    Finally, I think  I have really  fixed  the module  loading problem with
-    more than one DECchip based  card.  As a  side effect, I don't mess with
-    the  device structure any  more which means that  if more than 1 card in
-    2.0.x is    installed (4  in   2.1.x),  the  user   will have   to  edit
-    linux/drivers/net/Space.c  to make room for  them. Hence, module loading
-    is  the preferred way to use   this driver, since  it  doesn't have this
-    limitation.
-
-    Where SROM media  detection is used and  full duplex is specified in the
-    SROM,  the feature is  ignored unless  lp->params.fdx  is set at compile
-    time  OR during  a   module load  (insmod  de4x5   args='eth??:fdx' [see
-    below]).  This is because there  is no way  to automatically detect full
-    duplex   links  except through   autonegotiation.    When I  include the
-    autonegotiation feature in  the SROM autoconf  code, this detection will
-    occur automatically for that case.
-
-    Command  line arguments are  now  allowed, similar  to passing arguments
-    through LILO. This will allow a per adapter board  set up of full duplex
-    and media. The only lexical constraints  are: the board name (dev->name)
-    appears in the list before its  parameters.  The list of parameters ends
-    either at the end of the parameter list or with another board name.  The
-    following parameters are allowed:
-
-            fdx        for full duplex
-	    autosense  to set the media/speed; with the following
-	               sub-parameters:
-		       TP, TP_NW, BNC, AUI, BNC_AUI, 100Mb, 10Mb, AUTO
-
-    Case sensitivity is important  for  the sub-parameters. They *must*   be
-    upper case. Examples:
-
-        insmod de4x5 args='eth1:fdx autosense=BNC eth0:autosense=100Mb'.
-
-    For a compiled in driver, at or above line 548, place e.g.
-	#define DE4X5_PARM "eth0:fdx autosense=AUI eth2:autosense=TP"
-
-    Yes,  I know full duplex isn't  permissible on BNC  or AUI; they're just
-    examples. By default, full duplex is turned off and  AUTO is the default
-    autosense setting.  In reality, I expect only  the full duplex option to
-    be used. Note the use of single quotes in the two examples above and the
-    lack of commas to separate items. ALSO, you must get the requested media
-    correct in relation to what the adapter SROM says it has. There's no way
-    to  determine this in  advance other than by  trial and error and common
-    sense, e.g. call a BNC connectored port 'BNC', not '10Mb'.
-
-    Changed the bus probing.  EISA used to be  done first,  followed by PCI.
-    Most people probably don't even know  what a de425 is today and the EISA
-    probe has messed  up some SCSI cards  in the past,  so now PCI is always
-    probed  first  followed by  EISA if  a) the architecture allows EISA and
-    either  b) there have been no PCI cards detected or  c) an EISA probe is
-    forced by  the user.  To force  a probe  include  "force_eisa"  in  your
-    insmod "args" line;  for built-in kernels either change the driver to do
-    this  automatically  or include  #define DE4X5_FORCE_EISA  on or  before
-    line 1040 in the driver.
-
-    TO DO:
-    ------
-
-    Revision History
-    ----------------
-
-    Version   Date        Description
-
-      0.1     17-Nov-94   Initial writing. ALPHA code release.
-      0.2     13-Jan-95   Added PCI support for DE435's.
-      0.21    19-Jan-95   Added auto media detection.
-      0.22    10-Feb-95   Fix interrupt handler call <chris@cosy.sbg.ac.at>.
-                          Fix recognition bug reported by <bkm@star.rl.ac.uk>.
-			  Add request/release_region code.
-			  Add loadable modules support for PCI.
-			  Clean up loadable modules support.
-      0.23    28-Feb-95   Added DC21041 and DC21140 support.
-                          Fix missed frame counter value and initialisation.
-			  Fixed EISA probe.
-      0.24    11-Apr-95   Change delay routine to use <linux/udelay>.
-                          Change TX_BUFFS_AVAIL macro.
-			  Change media autodetection to allow manual setting.
-			  Completed DE500 (DC21140) support.
-      0.241   18-Apr-95   Interim release without DE500 Autosense Algorithm.
-      0.242   10-May-95   Minor changes.
-      0.30    12-Jun-95   Timer fix for DC21140.
-                          Portability changes.
-			  Add ALPHA changes from <jestabro@ant.tay1.dec.com>.
-			  Add DE500 semi automatic autosense.
-			  Add Link Fail interrupt TP failure detection.
-			  Add timer based link change detection.
-			  Plugged a memory leak in de4x5_queue_pkt().
-      0.31    13-Jun-95   Fixed PCI stuff for 1.3.1.
-      0.32    26-Jun-95   Added verify_area() calls in de4x5_ioctl() from a
-                          suggestion by <heiko@colossus.escape.de>.
-      0.33     8-Aug-95   Add shared interrupt support (not released yet).
-      0.331   21-Aug-95   Fix de4x5_open() with fast CPUs.
-                          Fix de4x5_interrupt().
-                          Fix dc21140_autoconf() mess.
-			  No shared interrupt support.
-      0.332   11-Sep-95   Added MII management interface routines.
-      0.40     5-Mar-96   Fix setup frame timeout <maartenb@hpkuipc.cern.ch>.
-                          Add kernel timer code (h/w is too flaky).
-			  Add MII based PHY autosense.
-			  Add new multicasting code.
-			  Add new autosense algorithms for media/mode
-			  selection using kernel scheduling/timing.
-			  Re-formatted.
-			  Made changes suggested by <jeff@router.patch.net>:
-			    Change driver to detect all DECchip based cards
-			    with DEC_ONLY restriction a special case.
-			    Changed driver to autoprobe as a module. No irq
-			    checking is done now - assume BIOS is good!
-			  Added SMC9332 detection <manabe@Roy.dsl.tutics.ac.jp>
-      0.41    21-Mar-96   Don't check for get_hw_addr checksum unless DEC card
-                          only <niles@axp745gsfc.nasa.gov>
-			  Fix for multiple PCI cards reported by <jos@xos.nl>
-			  Duh, put the IRQF_SHARED flag into request_interrupt().
-			  Fix SMC ethernet address in enet_det[].
-			  Print chip name instead of "UNKNOWN" during boot.
-      0.42    26-Apr-96   Fix MII write TA bit error.
-                          Fix bug in dc21040 and dc21041 autosense code.
-			  Remove buffer copies on receive for Intels.
-			  Change sk_buff handling during media disconnects to
-			   eliminate DUP packets.
-			  Add dynamic TX thresholding.
-			  Change all chips to use perfect multicast filtering.
-			  Fix alloc_device() bug <jari@markkus2.fimr.fi>
-      0.43   21-Jun-96    Fix unconnected media TX retry bug.
-                          Add Accton to the list of broken cards.
-			  Fix TX under-run bug for non DC21140 chips.
-			  Fix boot command probe bug in alloc_device() as
-			   reported by <koen.gadeyne@barco.com> and
-			   <orava@nether.tky.hut.fi>.
-			  Add cache locks to prevent a race condition as
-			   reported by <csd@microplex.com> and
-			   <baba@beckman.uiuc.edu>.
-			  Upgraded alloc_device() code.
-      0.431  28-Jun-96    Fix potential bug in queue_pkt() from discussion
-                          with <csd@microplex.com>
-      0.44   13-Aug-96    Fix RX overflow bug in 2114[023] chips.
-                          Fix EISA probe bugs reported by <os2@kpi.kharkov.ua>
-			  and <michael@compurex.com>.
-      0.441   9-Sep-96    Change dc21041_autoconf() to probe quiet BNC media
-                           with a loopback packet.
-      0.442   9-Sep-96    Include AUI in dc21041 media printout. Bug reported
-                           by <bhat@mundook.cs.mu.OZ.AU>
-      0.45    8-Dec-96    Include endian functions for PPC use, from work
-                           by <cort@cs.nmt.edu> and <g.thomas@opengroup.org>.
-      0.451  28-Dec-96    Added fix to allow autoprobe for modules after
-                           suggestion from <mjacob@feral.com>.
-      0.5    30-Jan-97    Added SROM decoding functions.
-                          Updated debug flags.
-			  Fix sleep/wakeup calls for PCI cards, bug reported
-			   by <cross@gweep.lkg.dec.com>.
-			  Added multi-MAC, one SROM feature from discussion
-			   with <mjacob@feral.com>.
-			  Added full module autoprobe capability.
-			  Added attempt to use an SMC9332 with broken SROM.
-			  Added fix for ZYNX multi-mac cards that didn't
-			   get their IRQs wired correctly.
-      0.51   13-Feb-97    Added endian fixes for the SROM accesses from
-			   <paubert@iram.es>
-			  Fix init_connection() to remove extra device reset.
-			  Fix MAC/PHY reset ordering in dc21140m_autoconf().
-			  Fix initialisation problem with lp->timeout in
-			   typeX_infoblock() from <paubert@iram.es>.
-			  Fix MII PHY reset problem from work done by
-			   <paubert@iram.es>.
-      0.52   26-Apr-97    Some changes may not credit the right people -
-                           a disk crash meant I lost some mail.
-			  Change RX interrupt routine to drop rather than
-			   defer packets to avoid hang reported by
-			   <g.thomas@opengroup.org>.
-			  Fix srom_exec() to return for COMPACT and type 1
-			   infoblocks.
-			  Added DC21142 and DC21143 functions.
-			  Added byte counters from <phil@tazenda.demon.co.uk>
-			  Added IRQF_DISABLED temporary fix from
-			   <mjacob@feral.com>.
-      0.53   12-Nov-97    Fix the *_probe() to include 'eth??' name during
-                           module load: bug reported by
-			   <Piete.Brooks@cl.cam.ac.uk>
-			  Fix multi-MAC, one SROM, to work with 2114x chips:
-			   bug reported by <cmetz@inner.net>.
-			  Make above search independent of BIOS device scan
-			   direction.
-			  Completed DC2114[23] autosense functions.
-      0.531  21-Dec-97    Fix DE500-XA 100Mb/s bug reported by
-                           <robin@intercore.com
-			  Fix type1_infoblock() bug introduced in 0.53, from
-			   problem reports by
-			   <parmee@postecss.ncrfran.france.ncr.com> and
-			   <jo@ice.dillingen.baynet.de>.
-			  Added argument list to set up each board from either
-			   a module's command line or a compiled in #define.
-			  Added generic MII PHY functionality to deal with
-			   newer PHY chips.
-			  Fix the mess in 2.1.67.
-      0.532   5-Jan-98    Fix bug in mii_get_phy() reported by
-                           <redhat@cococo.net>.
-                          Fix bug in pci_probe() for 64 bit systems reported
-			   by <belliott@accessone.com>.
-      0.533   9-Jan-98    Fix more 64 bit bugs reported by <jal@cs.brown.edu>.
-      0.534  24-Jan-98    Fix last (?) endian bug from <geert@linux-m68k.org>
-      0.535  21-Feb-98    Fix Ethernet Address PROM reset bug for DC21040.
-      0.536  21-Mar-98    Change pci_probe() to use the pci_dev structure.
-			  **Incompatible with 2.0.x from here.**
-      0.540   5-Jul-98    Atomicize assertion of dev->interrupt for SMP
-                           from <lma@varesearch.com>
-			  Add TP, AUI and BNC cases to 21140m_autoconf() for
-			   case where a 21140 under SROM control uses, e.g. AUI
-			   from problem report by <delchini@lpnp09.in2p3.fr>
-			  Add MII parallel detection to 2114x_autoconf() for
-			   case where no autonegotiation partner exists from
-			   problem report by <mlapsley@ndirect.co.uk>.
-			  Add ability to force connection type directly even
-			   when using SROM control from problem report by
-			   <earl@exis.net>.
-			  Updated the PCI interface to conform with the latest
-			   version. I hope nothing is broken...
-          		  Add TX done interrupt modification from suggestion
-			   by <Austin.Donnelly@cl.cam.ac.uk>.
-			  Fix is_anc_capable() bug reported by
-			   <Austin.Donnelly@cl.cam.ac.uk>.
-			  Fix type[13]_infoblock() bug: during MII search, PHY
-			   lp->rst not run because lp->ibn not initialised -
-			   from report & fix by <paubert@iram.es>.
-			  Fix probe bug with EISA & PCI cards present from
-                           report by <eirik@netcom.com>.
-      0.541  24-Aug-98    Fix compiler problems associated with i386-string
-                           ops from multiple bug reports and temporary fix
-			   from <paubert@iram.es>.
-			  Fix pci_probe() to correctly emulate the old
-			   pcibios_find_class() function.
-			  Add an_exception() for old ZYNX346 and fix compile
-			   warning on PPC & SPARC, from <ecd@skynet.be>.
-			  Fix lastPCI to correctly work with compiled in
-			   kernels and modules from bug report by
-			   <Zlatko.Calusic@CARNet.hr> et al.
-      0.542  15-Sep-98    Fix dc2114x_autoconf() to stop multiple messages
-                           when media is unconnected.
-			  Change dev->interrupt to lp->interrupt to ensure
-			   alignment for Alpha's and avoid their unaligned
-			   access traps. This flag is merely for log messages:
-			   should do something more definitive though...
-      0.543  30-Dec-98    Add SMP spin locking.
-      0.544   8-May-99    Fix for buggy SROM in Motorola embedded boards using
-                           a 21143 by <mmporter@home.com>.
-			  Change PCI/EISA bus probing order.
-      0.545  28-Nov-99    Further Moto SROM bug fix from
-                           <mporter@eng.mcd.mot.com>
-                          Remove double checking for DEBUG_RX in de4x5_dbg_rx()
-			   from report by <geert@linux-m68k.org>
-      0.546  22-Feb-01    Fixes Alpha XP1000 oops.  The srom_search function
-                           was causing a page fault when initializing the
-                           variable 'pb', on a non de4x5 PCI device, in this
-                           case a PCI bridge (DEC chip 21152). The value of
-                           'pb' is now only initialized if a de4x5 chip is
-                           present.
-                           <france@handhelds.org>
-      0.547  08-Nov-01    Use library crc32 functions by <Matt_Domsch@dell.com>
-      0.548  30-Aug-03    Big 2.6 cleanup. Ported to PCI/EISA probing and
-                           generic DMA APIs. Fixed DE425 support on Alpha.
-			   <maz@wild-wind.fr.eu.org>
-    =========================================================================
-*/
-
-#include <linux/module.h>
-#include <linux/kernel.h>
-#include <linux/string.h>
-#include <linux/interrupt.h>
-#include <linux/ptrace.h>
-#include <linux/errno.h>
-#include <linux/ioport.h>
-#include <linux/pci.h>
-#include <linux/eisa.h>
-#include <linux/delay.h>
-#include <linux/init.h>
-#include <linux/spinlock.h>
-#include <linux/crc32.h>
-#include <linux/netdevice.h>
-#include <linux/etherdevice.h>
-#include <linux/skbuff.h>
-#include <linux/time.h>
-#include <linux/types.h>
-#include <linux/unistd.h>
-#include <linux/ctype.h>
-#include <linux/dma-mapping.h>
-#include <linux/moduleparam.h>
-#include <linux/bitops.h>
-#include <linux/gfp.h>
-
-#include <asm/io.h>
-#include <asm/dma.h>
-#include <asm/byteorder.h>
-#include <asm/unaligned.h>
-#include <linux/uaccess.h>
-#ifdef CONFIG_PPC_PMAC
-#include <asm/machdep.h>
-#endif /* CONFIG_PPC_PMAC */
-
-#include "de4x5.h"
-
-static const char version[] =
-	KERN_INFO "de4x5.c:V0.546 2001/02/22 davies@maniac.ultranet.com\n";
-
-#define c_char const char
-
-/*
-** MII Information
-*/
-struct phy_table {
-    int reset;              /* Hard reset required?                         */
-    int id;                 /* IEEE OUI                                     */
-    int ta;                 /* One cycle TA time - 802.3u is confusing here */
-    struct {                /* Non autonegotiation (parallel) speed det.    */
-	int reg;
-	int mask;
-	int value;
-    } spd;
-};
-
-struct mii_phy {
-    int reset;              /* Hard reset required?                      */
-    int id;                 /* IEEE OUI                                  */
-    int ta;                 /* One cycle TA time                         */
-    struct {                /* Non autonegotiation (parallel) speed det. */
-	int reg;
-	int mask;
-	int value;
-    } spd;
-    int addr;               /* MII address for the PHY                   */
-    u_char  *gep;           /* Start of GEP sequence block in SROM       */
-    u_char  *rst;           /* Start of reset sequence in SROM           */
-    u_int mc;               /* Media Capabilities                        */
-    u_int ana;              /* NWay Advertisement                        */
-    u_int fdx;              /* Full DupleX capabilities for each media   */
-    u_int ttm;              /* Transmit Threshold Mode for each media    */
-    u_int mci;              /* 21142 MII Connector Interrupt info        */
-};
-
-#define DE4X5_MAX_PHY 8     /* Allow up to 8 attached PHY devices per board */
-
-struct sia_phy {
-    u_char mc;              /* Media Code                                */
-    u_char ext;             /* csr13-15 valid when set                   */
-    int csr13;              /* SIA Connectivity Register                 */
-    int csr14;              /* SIA TX/RX Register                        */
-    int csr15;              /* SIA General Register                      */
-    int gepc;               /* SIA GEP Control Information               */
-    int gep;                /* SIA GEP Data                              */
-};
-
-/*
-** Define the know universe of PHY devices that can be
-** recognised by this driver.
-*/
-static struct phy_table phy_info[] = {
-    {0, NATIONAL_TX, 1, {0x19, 0x40, 0x00}},       /* National TX      */
-    {1, BROADCOM_T4, 1, {0x10, 0x02, 0x02}},       /* Broadcom T4      */
-    {0, SEEQ_T4    , 1, {0x12, 0x10, 0x10}},       /* SEEQ T4          */
-    {0, CYPRESS_T4 , 1, {0x05, 0x20, 0x20}},       /* Cypress T4       */
-    {0, 0x7810     , 1, {0x14, 0x0800, 0x0800}}    /* Level One LTX970 */
-};
-
-/*
-** These GENERIC values assumes that the PHY devices follow 802.3u and
-** allow parallel detection to set the link partner ability register.
-** Detection of 100Base-TX [H/F Duplex] and 100Base-T4 is supported.
-*/
-#define GENERIC_REG   0x05      /* Autoneg. Link Partner Advertisement Reg. */
-#define GENERIC_MASK  MII_ANLPA_100M /* All 100Mb/s Technologies            */
-#define GENERIC_VALUE MII_ANLPA_100M /* 100B-TX, 100B-TX FDX, 100B-T4       */
-
-/*
-** Define special SROM detection cases
-*/
-static c_char enet_det[][ETH_ALEN] = {
-    {0x00, 0x00, 0xc0, 0x00, 0x00, 0x00},
-    {0x00, 0x00, 0xe8, 0x00, 0x00, 0x00}
-};
-
-#define SMC    1
-#define ACCTON 2
-
-/*
-** SROM Repair definitions. If a broken SROM is detected a card may
-** use this information to help figure out what to do. This is a
-** "stab in the dark" and so far for SMC9332's only.
-*/
-static c_char srom_repair_info[][100] = {
-    {0x00,0x1e,0x00,0x00,0x00,0x08,             /* SMC9332 */
-     0x1f,0x01,0x8f,0x01,0x00,0x01,0x00,0x02,
-     0x01,0x00,0x00,0x78,0xe0,0x01,0x00,0x50,
-     0x00,0x18,}
-};
-
-
-#ifdef DE4X5_DEBUG
-static int de4x5_debug = DE4X5_DEBUG;
-#else
-/*static int de4x5_debug = (DEBUG_MII | DEBUG_SROM | DEBUG_PCICFG | DEBUG_MEDIA | DEBUG_VERSION);*/
-static int de4x5_debug = (DEBUG_MEDIA | DEBUG_VERSION);
-#endif
-
-/*
-** Allow per adapter set up. For modules this is simply a command line
-** parameter, e.g.:
-** insmod de4x5 args='eth1:fdx autosense=BNC eth0:autosense=100Mb'.
-**
-** For a compiled in driver, place e.g.
-**     #define DE4X5_PARM "eth0:fdx autosense=AUI eth2:autosense=TP"
-** here
-*/
-#ifdef DE4X5_PARM
-static char *args = DE4X5_PARM;
-#else
-static char *args;
-#endif
-
-struct parameters {
-    bool fdx;
-    int autosense;
-};
-
-#define DE4X5_AUTOSENSE_MS 250      /* msec autosense tick (DE500) */
-
-#define DE4X5_NDA 0xffe0            /* No Device (I/O) Address */
-
-/*
-** Ethernet PROM defines
-*/
-#define PROBE_LENGTH    32
-#define ETH_PROM_SIG    0xAA5500FFUL
-
-/*
-** Ethernet Info
-*/
-#define PKT_BUF_SZ	1536            /* Buffer size for each Tx/Rx buffer */
-#define IEEE802_3_SZ    1518            /* Packet + CRC */
-#define MAX_PKT_SZ   	1514            /* Maximum ethernet packet length */
-#define MAX_DAT_SZ   	1500            /* Maximum ethernet data length */
-#define MIN_DAT_SZ   	1               /* Minimum ethernet data length */
-#define PKT_HDR_LEN     14              /* Addresses and data length info */
-#define FAKE_FRAME_LEN  (MAX_PKT_SZ + 1)
-#define QUEUE_PKT_TIMEOUT (3*HZ)        /* 3 second timeout */
-
-
-/*
-** EISA bus defines
-*/
-#define DE4X5_EISA_IO_PORTS   0x0c00    /* I/O port base address, slot 0 */
-#define DE4X5_EISA_TOTAL_SIZE 0x100     /* I/O address extent */
-
-#define EISA_ALLOWED_IRQ_LIST  {5, 9, 10, 11}
-
-#define DE4X5_SIGNATURE {"DE425","DE434","DE435","DE450","DE500"}
-#define DE4X5_NAME_LENGTH 8
-
-static c_char *de4x5_signatures[] = DE4X5_SIGNATURE;
-
-/*
-** Ethernet PROM defines for DC21040
-*/
-#define PROBE_LENGTH    32
-#define ETH_PROM_SIG    0xAA5500FFUL
-
-/*
-** PCI Bus defines
-*/
-#define PCI_MAX_BUS_NUM      8
-#define DE4X5_PCI_TOTAL_SIZE 0x80       /* I/O address extent */
-#define DE4X5_CLASS_CODE     0x00020000 /* Network controller, Ethernet */
-
-/*
-** Memory Alignment. Each descriptor is 4 longwords long. To force a
-** particular alignment on the TX descriptor, adjust DESC_SKIP_LEN and
-** DESC_ALIGN. ALIGN aligns the start address of the private memory area
-** and hence the RX descriptor ring's first entry.
-*/
-#define DE4X5_ALIGN4      ((u_long)4 - 1)     /* 1 longword align */
-#define DE4X5_ALIGN8      ((u_long)8 - 1)     /* 2 longword align */
-#define DE4X5_ALIGN16     ((u_long)16 - 1)    /* 4 longword align */
-#define DE4X5_ALIGN32     ((u_long)32 - 1)    /* 8 longword align */
-#define DE4X5_ALIGN64     ((u_long)64 - 1)    /* 16 longword align */
-#define DE4X5_ALIGN128    ((u_long)128 - 1)   /* 32 longword align */
-
-#define DE4X5_ALIGN         DE4X5_ALIGN32           /* Keep the DC21040 happy... */
-#define DE4X5_CACHE_ALIGN   CAL_16LONG
-#define DESC_SKIP_LEN DSL_0             /* Must agree with DESC_ALIGN */
-/*#define DESC_ALIGN    u32 dummy[4];  / * Must agree with DESC_SKIP_LEN */
-#define DESC_ALIGN
-
-#ifndef DEC_ONLY                        /* See README.de4x5 for using this */
-static int dec_only;
-#else
-static int dec_only = 1;
-#endif
-
-/*
-** DE4X5 IRQ ENABLE/DISABLE
-*/
-#define ENABLE_IRQs { \
-    imr |= lp->irq_en;\
-    outl(imr, DE4X5_IMR);               /* Enable the IRQs */\
-}
-
-#define DISABLE_IRQs {\
-    imr = inl(DE4X5_IMR);\
-    imr &= ~lp->irq_en;\
-    outl(imr, DE4X5_IMR);               /* Disable the IRQs */\
-}
-
-#define UNMASK_IRQs {\
-    imr |= lp->irq_mask;\
-    outl(imr, DE4X5_IMR);               /* Unmask the IRQs */\
-}
-
-#define MASK_IRQs {\
-    imr = inl(DE4X5_IMR);\
-    imr &= ~lp->irq_mask;\
-    outl(imr, DE4X5_IMR);               /* Mask the IRQs */\
-}
-
-/*
-** DE4X5 START/STOP
-*/
-#define START_DE4X5 {\
-    omr = inl(DE4X5_OMR);\
-    omr |= OMR_ST | OMR_SR;\
-    outl(omr, DE4X5_OMR);               /* Enable the TX and/or RX */\
-}
-
-#define STOP_DE4X5 {\
-    omr = inl(DE4X5_OMR);\
-    omr &= ~(OMR_ST|OMR_SR);\
-    outl(omr, DE4X5_OMR);               /* Disable the TX and/or RX */ \
-}
-
-/*
-** DE4X5 SIA RESET
-*/
-#define RESET_SIA outl(0, DE4X5_SICR);  /* Reset SIA connectivity regs */
-
-/*
-** DE500 AUTOSENSE TIMER INTERVAL (MILLISECS)
-*/
-#define DE4X5_AUTOSENSE_MS  250
-
-/*
-** SROM Structure
-*/
-struct de4x5_srom {
-    char sub_vendor_id[2];
-    char sub_system_id[2];
-    char reserved[12];
-    char id_block_crc;
-    char reserved2;
-    char version;
-    char num_controllers;
-    char ieee_addr[6];
-    char info[100];
-    short chksum;
-};
-#define SUB_VENDOR_ID 0x500a
-
-/*
-** DE4X5 Descriptors. Make sure that all the RX buffers are contiguous
-** and have sizes of both a power of 2 and a multiple of 4.
-** A size of 256 bytes for each buffer could be chosen because over 90% of
-** all packets in our network are <256 bytes long and 64 longword alignment
-** is possible. 1536 showed better 'ttcp' performance. Take your pick. 32 TX
-** descriptors are needed for machines with an ALPHA CPU.
-*/
-#define NUM_RX_DESC 8                   /* Number of RX descriptors   */
-#define NUM_TX_DESC 32                  /* Number of TX descriptors   */
-#define RX_BUFF_SZ  1536                /* Power of 2 for kmalloc and */
-                                        /* Multiple of 4 for DC21040  */
-                                        /* Allows 512 byte alignment  */
-struct de4x5_desc {
-    volatile __le32 status;
-    __le32 des1;
-    __le32 buf;
-    __le32 next;
-    DESC_ALIGN
-};
-
-/*
-** The DE4X5 private structure
-*/
-#define DE4X5_PKT_STAT_SZ 16
-#define DE4X5_PKT_BIN_SZ  128            /* Should be >=100 unless you
-                                            increase DE4X5_PKT_STAT_SZ */
-
-struct pkt_stats {
-	u_int bins[DE4X5_PKT_STAT_SZ];      /* Private stats counters       */
-	u_int unicast;
-	u_int multicast;
-	u_int broadcast;
-	u_int excessive_collisions;
-	u_int tx_underruns;
-	u_int excessive_underruns;
-	u_int rx_runt_frames;
-	u_int rx_collision;
-	u_int rx_dribble;
-	u_int rx_overflow;
-};
-
-struct de4x5_private {
-    char adapter_name[80];                  /* Adapter name                 */
-    u_long interrupt;                       /* Aligned ISR flag             */
-    struct de4x5_desc *rx_ring;		    /* RX descriptor ring           */
-    struct de4x5_desc *tx_ring;		    /* TX descriptor ring           */
-    struct sk_buff *tx_skb[NUM_TX_DESC];    /* TX skb for freeing when sent */
-    struct sk_buff *rx_skb[NUM_RX_DESC];    /* RX skb's                     */
-    int rx_new, rx_old;                     /* RX descriptor ring pointers  */
-    int tx_new, tx_old;                     /* TX descriptor ring pointers  */
-    char setup_frame[SETUP_FRAME_LEN];      /* Holds MCA and PA info.       */
-    char frame[64];                         /* Min sized packet for loopback*/
-    spinlock_t lock;                        /* Adapter specific spinlock    */
-    struct net_device_stats stats;          /* Public stats                 */
-    struct pkt_stats pktStats;	            /* Private stats counters	    */
-    char rxRingSize;
-    char txRingSize;
-    int  bus;                               /* EISA or PCI                  */
-    int  bus_num;                           /* PCI Bus number               */
-    int  device;                            /* Device number on PCI bus     */
-    int  state;                             /* Adapter OPENED or CLOSED     */
-    int  chipset;                           /* DC21040, DC21041 or DC21140  */
-    s32  irq_mask;                          /* Interrupt Mask (Enable) bits */
-    s32  irq_en;                            /* Summary interrupt bits       */
-    int  media;                             /* Media (eg TP), mode (eg 100B)*/
-    int  c_media;                           /* Remember the last media conn */
-    bool fdx;                               /* media full duplex flag       */
-    int  linkOK;                            /* Link is OK                   */
-    int  autosense;                         /* Allow/disallow autosensing   */
-    bool tx_enable;                         /* Enable descriptor polling    */
-    int  setup_f;                           /* Setup frame filtering type   */
-    int  local_state;                       /* State within a 'media' state */
-    struct mii_phy phy[DE4X5_MAX_PHY];      /* List of attached PHY devices */
-    struct sia_phy sia;                     /* SIA PHY Information          */
-    int  active;                            /* Index to active PHY device   */
-    int  mii_cnt;                           /* Number of attached PHY's     */
-    int  timeout;                           /* Scheduling counter           */
-    struct timer_list timer;                /* Timer info for kernel        */
-    int tmp;                                /* Temporary global per card    */
-    struct {
-	u_long lock;                        /* Lock the cache accesses      */
-	s32 csr0;                           /* Saved Bus Mode Register      */
-	s32 csr6;                           /* Saved Operating Mode Reg.    */
-	s32 csr7;                           /* Saved IRQ Mask Register      */
-	s32 gep;                            /* Saved General Purpose Reg.   */
-	s32 gepc;                           /* Control info for GEP         */
-	s32 csr13;                          /* Saved SIA Connectivity Reg.  */
-	s32 csr14;                          /* Saved SIA TX/RX Register     */
-	s32 csr15;                          /* Saved SIA General Register   */
-	int save_cnt;                       /* Flag if state already saved  */
-	struct sk_buff_head queue;          /* Save the (re-ordered) skb's  */
-    } cache;
-    struct de4x5_srom srom;                 /* A copy of the SROM           */
-    int cfrv;				    /* Card CFRV copy */
-    int rx_ovf;                             /* Check for 'RX overflow' tag  */
-    bool useSROM;                           /* For non-DEC card use SROM    */
-    bool useMII;                            /* Infoblock using the MII      */
-    int asBitValid;                         /* Autosense bits in GEP?       */
-    int asPolarity;                         /* 0 => asserted high           */
-    int asBit;                              /* Autosense bit number in GEP  */
-    int defMedium;                          /* SROM default medium          */
-    int tcount;                             /* Last infoblock number        */
-    int infoblock_init;                     /* Initialised this infoblock?  */
-    int infoleaf_offset;                    /* SROM infoleaf for controller */
-    s32 infoblock_csr6;                     /* csr6 value in SROM infoblock */
-    int infoblock_media;                    /* infoblock media              */
-    int (*infoleaf_fn)(struct net_device *);    /* Pointer to infoleaf function */
-    u_char *rst;                            /* Pointer to Type 5 reset info */
-    u_char  ibn;                            /* Infoblock number             */
-    struct parameters params;               /* Command line/ #defined params */
-    struct device *gendev;	            /* Generic device */
-    dma_addr_t dma_rings;		    /* DMA handle for rings	    */
-    int dma_size;			    /* Size of the DMA area	    */
-    char *rx_bufs;			    /* rx bufs on alpha, sparc, ... */
-};
-
-/*
-** To get around certain poxy cards that don't provide an SROM
-** for the second and more DECchip, I have to key off the first
-** chip's address. I'll assume there's not a bad SROM iff:
-**
-**      o the chipset is the same
-**      o the bus number is the same and > 0
-**      o the sum of all the returned hw address bytes is 0 or 0x5fa
-**
-** Also have to save the irq for those cards whose hardware designers
-** can't follow the PCI to PCI Bridge Architecture spec.
-*/
-static struct {
-    int chipset;
-    int bus;
-    int irq;
-    u_char addr[ETH_ALEN];
-} last = {0,};
-
-/*
-** The transmit ring full condition is described by the tx_old and tx_new
-** pointers by:
-**    tx_old            = tx_new    Empty ring
-**    tx_old            = tx_new+1  Full ring
-**    tx_old+txRingSize = tx_new+1  Full ring  (wrapped condition)
-*/
-#define TX_BUFFS_AVAIL ((lp->tx_old<=lp->tx_new)?\
-			lp->tx_old+lp->txRingSize-lp->tx_new-1:\
-			lp->tx_old               -lp->tx_new-1)
-
-#define TX_PKT_PENDING (lp->tx_old != lp->tx_new)
-
-/*
-** Public Functions
-*/
-static int     de4x5_open(struct net_device *dev);
-static netdev_tx_t de4x5_queue_pkt(struct sk_buff *skb,
-					 struct net_device *dev);
-static irqreturn_t de4x5_interrupt(int irq, void *dev_id);
-static int     de4x5_close(struct net_device *dev);
-static struct  net_device_stats *de4x5_get_stats(struct net_device *dev);
-static void    de4x5_local_stats(struct net_device *dev, char *buf, int pkt_len);
-static void    set_multicast_list(struct net_device *dev);
-static int     de4x5_ioctl(struct net_device *dev, struct ifreq *rq, int cmd);
-
-/*
-** Private functions
-*/
-static int     de4x5_hw_init(struct net_device *dev, u_long iobase, struct device *gendev);
-static int     de4x5_init(struct net_device *dev);
-static int     de4x5_sw_reset(struct net_device *dev);
-static int     de4x5_rx(struct net_device *dev);
-static int     de4x5_tx(struct net_device *dev);
-static void    de4x5_ast(struct net_device *dev);
-static int     de4x5_txur(struct net_device *dev);
-static int     de4x5_rx_ovfc(struct net_device *dev);
-
-static int     autoconf_media(struct net_device *dev);
-static void    create_packet(struct net_device *dev, char *frame, int len);
-static void    load_packet(struct net_device *dev, char *buf, u32 flags, struct sk_buff *skb);
-static int     dc21040_autoconf(struct net_device *dev);
-static int     dc21041_autoconf(struct net_device *dev);
-static int     dc21140m_autoconf(struct net_device *dev);
-static int     dc2114x_autoconf(struct net_device *dev);
-static int     srom_autoconf(struct net_device *dev);
-static int     de4x5_suspect_state(struct net_device *dev, int timeout, int prev_state, int (*fn)(struct net_device *, int), int (*asfn)(struct net_device *));
-static int     dc21040_state(struct net_device *dev, int csr13, int csr14, int csr15, int timeout, int next_state, int suspect_state, int (*fn)(struct net_device *, int));
-static int     test_media(struct net_device *dev, s32 irqs, s32 irq_mask, s32 csr13, s32 csr14, s32 csr15, s32 msec);
-static int     test_for_100Mb(struct net_device *dev, int msec);
-static int     wait_for_link(struct net_device *dev);
-static int     test_mii_reg(struct net_device *dev, int reg, int mask, bool pol, long msec);
-static int     is_spd_100(struct net_device *dev);
-static int     is_100_up(struct net_device *dev);
-static int     is_10_up(struct net_device *dev);
-static int     is_anc_capable(struct net_device *dev);
-static int     ping_media(struct net_device *dev, int msec);
-static struct sk_buff *de4x5_alloc_rx_buff(struct net_device *dev, int index, int len);
-static void    de4x5_free_rx_buffs(struct net_device *dev);
-static void    de4x5_free_tx_buffs(struct net_device *dev);
-static void    de4x5_save_skbs(struct net_device *dev);
-static void    de4x5_rst_desc_ring(struct net_device *dev);
-static void    de4x5_cache_state(struct net_device *dev, int flag);
-static void    de4x5_put_cache(struct net_device *dev, struct sk_buff *skb);
-static void    de4x5_putb_cache(struct net_device *dev, struct sk_buff *skb);
-static struct  sk_buff *de4x5_get_cache(struct net_device *dev);
-static void    de4x5_setup_intr(struct net_device *dev);
-static void    de4x5_init_connection(struct net_device *dev);
-static int     de4x5_reset_phy(struct net_device *dev);
-static void    reset_init_sia(struct net_device *dev, s32 sicr, s32 strr, s32 sigr);
-static int     test_ans(struct net_device *dev, s32 irqs, s32 irq_mask, s32 msec);
-static int     test_tp(struct net_device *dev, s32 msec);
-static int     EISA_signature(char *name, struct device *device);
-static int     PCI_signature(char *name, struct de4x5_private *lp);
-static void    DevicePresent(struct net_device *dev, u_long iobase);
-static void    enet_addr_rst(u_long aprom_addr);
-static int     de4x5_bad_srom(struct de4x5_private *lp);
-static short   srom_rd(u_long address, u_char offset);
-static void    srom_latch(u_int command, u_long address);
-static void    srom_command(u_int command, u_long address);
-static void    srom_address(u_int command, u_long address, u_char offset);
-static short   srom_data(u_int command, u_long address);
-/*static void    srom_busy(u_int command, u_long address);*/
-static void    sendto_srom(u_int command, u_long addr);
-static int     getfrom_srom(u_long addr);
-static int     srom_map_media(struct net_device *dev);
-static int     srom_infoleaf_info(struct net_device *dev);
-static void    srom_init(struct net_device *dev);
-static void    srom_exec(struct net_device *dev, u_char *p);
-static int     mii_rd(u_char phyreg, u_char phyaddr, u_long ioaddr);
-static void    mii_wr(int data, u_char phyreg, u_char phyaddr, u_long ioaddr);
-static int     mii_rdata(u_long ioaddr);
-static void    mii_wdata(int data, int len, u_long ioaddr);
-static void    mii_ta(u_long rw, u_long ioaddr);
-static int     mii_swap(int data, int len);
-static void    mii_address(u_char addr, u_long ioaddr);
-static void    sendto_mii(u32 command, int data, u_long ioaddr);
-static int     getfrom_mii(u32 command, u_long ioaddr);
-static int     mii_get_oui(u_char phyaddr, u_long ioaddr);
-static int     mii_get_phy(struct net_device *dev);
-static void    SetMulticastFilter(struct net_device *dev);
-static int     get_hw_addr(struct net_device *dev);
-static void    srom_repair(struct net_device *dev, int card);
-static int     test_bad_enet(struct net_device *dev, int status);
-static int     an_exception(struct de4x5_private *lp);
-static char    *build_setup_frame(struct net_device *dev, int mode);
-static void    disable_ast(struct net_device *dev);
-static long    de4x5_switch_mac_port(struct net_device *dev);
-static int     gep_rd(struct net_device *dev);
-static void    gep_wr(s32 data, struct net_device *dev);
-static void    yawn(struct net_device *dev, int state);
-static void    de4x5_parse_params(struct net_device *dev);
-static void    de4x5_dbg_open(struct net_device *dev);
-static void    de4x5_dbg_mii(struct net_device *dev, int k);
-static void    de4x5_dbg_media(struct net_device *dev);
-static void    de4x5_dbg_srom(struct de4x5_srom *p);
-static void    de4x5_dbg_rx(struct sk_buff *skb, int len);
-static int     dc21041_infoleaf(struct net_device *dev);
-static int     dc21140_infoleaf(struct net_device *dev);
-static int     dc21142_infoleaf(struct net_device *dev);
-static int     dc21143_infoleaf(struct net_device *dev);
-static int     type0_infoblock(struct net_device *dev, u_char count, u_char *p);
-static int     type1_infoblock(struct net_device *dev, u_char count, u_char *p);
-static int     type2_infoblock(struct net_device *dev, u_char count, u_char *p);
-static int     type3_infoblock(struct net_device *dev, u_char count, u_char *p);
-static int     type4_infoblock(struct net_device *dev, u_char count, u_char *p);
-static int     type5_infoblock(struct net_device *dev, u_char count, u_char *p);
-static int     compact_infoblock(struct net_device *dev, u_char count, u_char *p);
-
-/*
-** Note now that module autoprobing is allowed under EISA and PCI. The
-** IRQ lines will not be auto-detected; instead I'll rely on the BIOSes
-** to "do the right thing".
-*/
-
-static int io=0x0;/* EDIT THIS LINE FOR YOUR CONFIGURATION IF NEEDED        */
-
-module_param_hw(io, int, ioport, 0);
-module_param(de4x5_debug, int, 0);
-module_param(dec_only, int, 0);
-module_param(args, charp, 0);
-
-MODULE_PARM_DESC(io, "de4x5 I/O base address");
-MODULE_PARM_DESC(de4x5_debug, "de4x5 debug mask");
-MODULE_PARM_DESC(dec_only, "de4x5 probe only for Digital boards (0-1)");
-MODULE_PARM_DESC(args, "de4x5 full duplex and media type settings; see de4x5.c for details");
-MODULE_LICENSE("GPL");
-
-/*
-** List the SROM infoleaf functions and chipsets
-*/
-struct InfoLeaf {
-    int chipset;
-    int (*fn)(struct net_device *);
-};
-static struct InfoLeaf infoleaf_array[] = {
-    {DC21041, dc21041_infoleaf},
-    {DC21140, dc21140_infoleaf},
-    {DC21142, dc21142_infoleaf},
-    {DC21143, dc21143_infoleaf}
-};
-#define INFOLEAF_SIZE ARRAY_SIZE(infoleaf_array)
-
-/*
-** List the SROM info block functions
-*/
-static int (*dc_infoblock[])(struct net_device *dev, u_char, u_char *) = {
-    type0_infoblock,
-    type1_infoblock,
-    type2_infoblock,
-    type3_infoblock,
-    type4_infoblock,
-    type5_infoblock,
-    compact_infoblock
-};
-
-#define COMPACT (ARRAY_SIZE(dc_infoblock) - 1)
-
-/*
-** Miscellaneous defines...
-*/
-#define RESET_DE4X5 {\
-    int i;\
-    i=inl(DE4X5_BMR);\
-    mdelay(1);\
-    outl(i | BMR_SWR, DE4X5_BMR);\
-    mdelay(1);\
-    outl(i, DE4X5_BMR);\
-    mdelay(1);\
-    for (i=0;i<5;i++) {inl(DE4X5_BMR); mdelay(1);}\
-    mdelay(1);\
-}
-
-#define PHY_HARD_RESET {\
-    outl(GEP_HRST, DE4X5_GEP);           /* Hard RESET the PHY dev. */\
-    mdelay(1);                           /* Assert for 1ms */\
-    outl(0x00, DE4X5_GEP);\
-    mdelay(2);                           /* Wait for 2ms */\
-}
-
-static const struct net_device_ops de4x5_netdev_ops = {
-    .ndo_open		= de4x5_open,
-    .ndo_stop		= de4x5_close,
-    .ndo_start_xmit	= de4x5_queue_pkt,
-    .ndo_get_stats	= de4x5_get_stats,
-    .ndo_set_rx_mode	= set_multicast_list,
-    .ndo_do_ioctl	= de4x5_ioctl,
-    .ndo_set_mac_address= eth_mac_addr,
-    .ndo_validate_addr	= eth_validate_addr,
-};
-
-
-static int
-de4x5_hw_init(struct net_device *dev, u_long iobase, struct device *gendev)
-{
-    char name[DE4X5_NAME_LENGTH + 1];
-    struct de4x5_private *lp = netdev_priv(dev);
-    struct pci_dev *pdev = NULL;
-    int i, status=0;
-
-    dev_set_drvdata(gendev, dev);
-
-    /* Ensure we're not sleeping */
-    if (lp->bus == EISA) {
-	outb(WAKEUP, PCI_CFPM);
-    } else {
-	pdev = to_pci_dev (gendev);
-	pci_write_config_byte(pdev, PCI_CFDA_PSM, WAKEUP);
-    }
-    mdelay(10);
-
-    RESET_DE4X5;
-
-    if ((inl(DE4X5_STS) & (STS_TS | STS_RS)) != 0) {
-	return -ENXIO;                       /* Hardware could not reset */
-    }
-
-    /*
-    ** Now find out what kind of DC21040/DC21041/DC21140 board we have.
-    */
-    lp->useSROM = false;
-    if (lp->bus == PCI) {
-	PCI_signature(name, lp);
-    } else {
-	EISA_signature(name, gendev);
-    }
-
-    if (*name == '\0') {                     /* Not found a board signature */
-	return -ENXIO;
-    }
-
-    dev->base_addr = iobase;
-    printk ("%s: %s at 0x%04lx", dev_name(gendev), name, iobase);
-
-    status = get_hw_addr(dev);
-    printk(", h/w address %pM\n", dev->dev_addr);
-
-    if (status != 0) {
-	printk("      which has an Ethernet PROM CRC error.\n");
-	return -ENXIO;
-    } else {
-	skb_queue_head_init(&lp->cache.queue);
-	lp->cache.gepc = GEP_INIT;
-	lp->asBit = GEP_SLNK;
-	lp->asPolarity = GEP_SLNK;
-	lp->asBitValid = ~0;
-	lp->timeout = -1;
-	lp->gendev = gendev;
-	spin_lock_init(&lp->lock);
-	init_timer(&lp->timer);
-	lp->timer.function = (void (*)(unsigned long))de4x5_ast;
-	lp->timer.data = (unsigned long)dev;
-	de4x5_parse_params(dev);
-
-	/*
-	** Choose correct autosensing in case someone messed up
-	*/
-        lp->autosense = lp->params.autosense;
-        if (lp->chipset != DC21140) {
-            if ((lp->chipset==DC21040) && (lp->params.autosense&TP_NW)) {
-                lp->params.autosense = TP;
-            }
-            if ((lp->chipset==DC21041) && (lp->params.autosense&BNC_AUI)) {
-                lp->params.autosense = BNC;
-            }
-        }
-	lp->fdx = lp->params.fdx;
-	sprintf(lp->adapter_name,"%s (%s)", name, dev_name(gendev));
-
-	lp->dma_size = (NUM_RX_DESC + NUM_TX_DESC) * sizeof(struct de4x5_desc);
-#if defined(__alpha__) || defined(__powerpc__) || defined(CONFIG_SPARC) || defined(DE4X5_DO_MEMCPY)
-	lp->dma_size += RX_BUFF_SZ * NUM_RX_DESC + DE4X5_ALIGN;
-#endif
-	lp->rx_ring = dma_alloc_coherent(gendev, lp->dma_size,
-					 &lp->dma_rings, GFP_ATOMIC);
-	if (lp->rx_ring == NULL) {
-	    return -ENOMEM;
-	}
-
-	lp->tx_ring = lp->rx_ring + NUM_RX_DESC;
-
-	/*
-	** Set up the RX descriptor ring (Intels)
-	** Allocate contiguous receive buffers, long word aligned (Alphas)
-	*/
-#if !defined(__alpha__) && !defined(__powerpc__) && !defined(CONFIG_SPARC) && !defined(DE4X5_DO_MEMCPY)
-	for (i=0; i<NUM_RX_DESC; i++) {
-	    lp->rx_ring[i].status = 0;
-	    lp->rx_ring[i].des1 = cpu_to_le32(RX_BUFF_SZ);
-	    lp->rx_ring[i].buf = 0;
-	    lp->rx_ring[i].next = 0;
-	    lp->rx_skb[i] = (struct sk_buff *) 1;     /* Dummy entry */
-	}
-
-#else
-	{
-		dma_addr_t dma_rx_bufs;
-
-		dma_rx_bufs = lp->dma_rings + (NUM_RX_DESC + NUM_TX_DESC)
-		      	* sizeof(struct de4x5_desc);
-		dma_rx_bufs = (dma_rx_bufs + DE4X5_ALIGN) & ~DE4X5_ALIGN;
-		lp->rx_bufs = (char *)(((long)(lp->rx_ring + NUM_RX_DESC
-		      	+ NUM_TX_DESC) + DE4X5_ALIGN) & ~DE4X5_ALIGN);
-		for (i=0; i<NUM_RX_DESC; i++) {
-	    		lp->rx_ring[i].status = 0;
-	    		lp->rx_ring[i].des1 = cpu_to_le32(RX_BUFF_SZ);
-	    		lp->rx_ring[i].buf =
-				cpu_to_le32(dma_rx_bufs+i*RX_BUFF_SZ);
-	    		lp->rx_ring[i].next = 0;
-	    		lp->rx_skb[i] = (struct sk_buff *) 1; /* Dummy entry */
-		}
-
-	}
-#endif
-
-	barrier();
-
-	lp->rxRingSize = NUM_RX_DESC;
-	lp->txRingSize = NUM_TX_DESC;
-
-	/* Write the end of list marker to the descriptor lists */
-	lp->rx_ring[lp->rxRingSize - 1].des1 |= cpu_to_le32(RD_RER);
-	lp->tx_ring[lp->txRingSize - 1].des1 |= cpu_to_le32(TD_TER);
-
-	/* Tell the adapter where the TX/RX rings are located. */
-	outl(lp->dma_rings, DE4X5_RRBA);
-	outl(lp->dma_rings + NUM_RX_DESC * sizeof(struct de4x5_desc),
-	     DE4X5_TRBA);
-
-	/* Initialise the IRQ mask and Enable/Disable */
-	lp->irq_mask = IMR_RIM | IMR_TIM | IMR_TUM | IMR_UNM;
-	lp->irq_en   = IMR_NIM | IMR_AIM;
-
-	/* Create a loopback packet frame for later media probing */
-	create_packet(dev, lp->frame, sizeof(lp->frame));
-
-	/* Check if the RX overflow bug needs testing for */
-	i = lp->cfrv & 0x000000fe;
-	if ((lp->chipset == DC21140) && (i == 0x20)) {
-	    lp->rx_ovf = 1;
-	}
-
-	/* Initialise the SROM pointers if possible */
-	if (lp->useSROM) {
-	    lp->state = INITIALISED;
-	    if (srom_infoleaf_info(dev)) {
-	        dma_free_coherent (gendev, lp->dma_size,
-			       lp->rx_ring, lp->dma_rings);
-		return -ENXIO;
-	    }
-	    srom_init(dev);
-	}
-
-	lp->state = CLOSED;
-
-	/*
-	** Check for an MII interface
-	*/
-	if ((lp->chipset != DC21040) && (lp->chipset != DC21041)) {
-	    mii_get_phy(dev);
-	}
-
-	printk("      and requires IRQ%d (provided by %s).\n", dev->irq,
-	       ((lp->bus == PCI) ? "PCI BIOS" : "EISA CNFG"));
-    }
-
-    if (de4x5_debug & DEBUG_VERSION) {
-	printk(version);
-    }
-
-    /* The DE4X5-specific entries in the device structure. */
-    SET_NETDEV_DEV(dev, gendev);
-    dev->netdev_ops = &de4x5_netdev_ops;
-    dev->mem_start = 0;
-
-    /* Fill in the generic fields of the device structure. */
-    if ((status = register_netdev (dev))) {
-	    dma_free_coherent (gendev, lp->dma_size,
-			       lp->rx_ring, lp->dma_rings);
-	    return status;
-    }
-
-    /* Let the adapter sleep to save power */
-    yawn(dev, SLEEP);
-
-    return status;
-}
-
-
-static int
-de4x5_open(struct net_device *dev)
-{
-    struct de4x5_private *lp = netdev_priv(dev);
-    u_long iobase = dev->base_addr;
-    int i, status = 0;
-    s32 omr;
-
-    /* Allocate the RX buffers */
-    for (i=0; i<lp->rxRingSize; i++) {
-	if (de4x5_alloc_rx_buff(dev, i, 0) == NULL) {
-	    de4x5_free_rx_buffs(dev);
-	    return -EAGAIN;
-	}
-    }
-
-    /*
-    ** Wake up the adapter
-    */
-    yawn(dev, WAKEUP);
-
-    /*
-    ** Re-initialize the DE4X5...
-    */
-    status = de4x5_init(dev);
-    spin_lock_init(&lp->lock);
-    lp->state = OPEN;
-    de4x5_dbg_open(dev);
-
-    if (request_irq(dev->irq, de4x5_interrupt, IRQF_SHARED,
-		                                     lp->adapter_name, dev)) {
-	printk("de4x5_open(): Requested IRQ%d is busy - attempting FAST/SHARE...", dev->irq);
-	if (request_irq(dev->irq, de4x5_interrupt, IRQF_SHARED,
-			                             lp->adapter_name, dev)) {
-	    printk("\n              Cannot get IRQ- reconfigure your hardware.\n");
-	    disable_ast(dev);
-	    de4x5_free_rx_buffs(dev);
-	    de4x5_free_tx_buffs(dev);
-	    yawn(dev, SLEEP);
-	    lp->state = CLOSED;
-	    return -EAGAIN;
-	} else {
-	    printk("\n              Succeeded, but you should reconfigure your hardware to avoid this.\n");
-	    printk("WARNING: there may be IRQ related problems in heavily loaded systems.\n");
-	}
-    }
-
-    lp->interrupt = UNMASK_INTERRUPTS;
-    netif_trans_update(dev); /* prevent tx timeout */
-
-    START_DE4X5;
-
-    de4x5_setup_intr(dev);
-
-    if (de4x5_debug & DEBUG_OPEN) {
-	printk("\tsts:  0x%08x\n", inl(DE4X5_STS));
-	printk("\tbmr:  0x%08x\n", inl(DE4X5_BMR));
-	printk("\timr:  0x%08x\n", inl(DE4X5_IMR));
-	printk("\tomr:  0x%08x\n", inl(DE4X5_OMR));
-	printk("\tsisr: 0x%08x\n", inl(DE4X5_SISR));
-	printk("\tsicr: 0x%08x\n", inl(DE4X5_SICR));
-	printk("\tstrr: 0x%08x\n", inl(DE4X5_STRR));
-	printk("\tsigr: 0x%08x\n", inl(DE4X5_SIGR));
-    }
-
-    return status;
-}
-
-/*
-** Initialize the DE4X5 operating conditions. NB: a chip problem with the
-** DC21140 requires using perfect filtering mode for that chip. Since I can't
-** see why I'd want > 14 multicast addresses, I have changed all chips to use
-** the perfect filtering mode. Keep the DMA burst length at 8: there seems
-** to be data corruption problems if it is larger (UDP errors seen from a
-** ttcp source).
-*/
-static int
-de4x5_init(struct net_device *dev)
-{
-    /* Lock out other processes whilst setting up the hardware */
-    netif_stop_queue(dev);
-
-    de4x5_sw_reset(dev);
-
-    /* Autoconfigure the connected port */
-    autoconf_media(dev);
-
-    return 0;
-}
-
-static int
-de4x5_sw_reset(struct net_device *dev)
-{
-    struct de4x5_private *lp = netdev_priv(dev);
-    u_long iobase = dev->base_addr;
-    int i, j, status = 0;
-    s32 bmr, omr;
-
-    /* Select the MII or SRL port now and RESET the MAC */
-    if (!lp->useSROM) {
-	if (lp->phy[lp->active].id != 0) {
-	    lp->infoblock_csr6 = OMR_SDP | OMR_PS | OMR_HBD;
-	} else {
-	    lp->infoblock_csr6 = OMR_SDP | OMR_TTM;
-	}
-	de4x5_switch_mac_port(dev);
-    }
-
-    /*
-    ** Set the programmable burst length to 8 longwords for all the DC21140
-    ** Fasternet chips and 4 longwords for all others: DMA errors result
-    ** without these values. Cache align 16 long.
-    */
-    bmr = (lp->chipset==DC21140 ? PBL_8 : PBL_4) | DESC_SKIP_LEN | DE4X5_CACHE_ALIGN;
-    bmr |= ((lp->chipset & ~0x00ff)==DC2114x ? BMR_RML : 0);
-    outl(bmr, DE4X5_BMR);
-
-    omr = inl(DE4X5_OMR) & ~OMR_PR;             /* Turn off promiscuous mode */
-    if (lp->chipset == DC21140) {
-	omr |= (OMR_SDP | OMR_SB);
-    }
-    lp->setup_f = PERFECT;
-    outl(lp->dma_rings, DE4X5_RRBA);
-    outl(lp->dma_rings + NUM_RX_DESC * sizeof(struct de4x5_desc),
-	 DE4X5_TRBA);
-
-    lp->rx_new = lp->rx_old = 0;
-    lp->tx_new = lp->tx_old = 0;
-
-    for (i = 0; i < lp->rxRingSize; i++) {
-	lp->rx_ring[i].status = cpu_to_le32(R_OWN);
-    }
-
-    for (i = 0; i < lp->txRingSize; i++) {
-	lp->tx_ring[i].status = cpu_to_le32(0);
-    }
-
-    barrier();
-
-    /* Build the setup frame depending on filtering mode */
-    SetMulticastFilter(dev);
-
-    load_packet(dev, lp->setup_frame, PERFECT_F|TD_SET|SETUP_FRAME_LEN, (struct sk_buff *)1);
-    outl(omr|OMR_ST, DE4X5_OMR);
-
-    /* Poll for setup frame completion (adapter interrupts are disabled now) */
-
-    for (j=0, i=0;(i<500) && (j==0);i++) {       /* Up to 500ms delay */
-	mdelay(1);
-	if ((s32)le32_to_cpu(lp->tx_ring[lp->tx_new].status) >= 0) j=1;
-    }
-    outl(omr, DE4X5_OMR);                        /* Stop everything! */
-
-    if (j == 0) {
-	printk("%s: Setup frame timed out, status %08x\n", dev->name,
-	       inl(DE4X5_STS));
-	status = -EIO;
-    }
-
-    lp->tx_new = (lp->tx_new + 1) % lp->txRingSize;
-    lp->tx_old = lp->tx_new;
-
-    return status;
-}
-
-/*
-** Writes a socket buffer address to the next available transmit descriptor.
-*/
-static netdev_tx_t
-de4x5_queue_pkt(struct sk_buff *skb, struct net_device *dev)
-{
-    struct de4x5_private *lp = netdev_priv(dev);
-    u_long iobase = dev->base_addr;
-    u_long flags = 0;
-
-    netif_stop_queue(dev);
-    if (!lp->tx_enable)                   /* Cannot send for now */
-		goto tx_err;
-
-    /*
-    ** Clean out the TX ring asynchronously to interrupts - sometimes the
-    ** interrupts are lost by delayed descriptor status updates relative to
-    ** the irq assertion, especially with a busy PCI bus.
-    */
-    spin_lock_irqsave(&lp->lock, flags);
-    de4x5_tx(dev);
-    spin_unlock_irqrestore(&lp->lock, flags);
-
-    /* Test if cache is already locked - requeue skb if so */
-    if (test_and_set_bit(0, (void *)&lp->cache.lock) && !lp->interrupt)
-		goto tx_err;
-
-    /* Transmit descriptor ring full or stale skb */
-    if (netif_queue_stopped(dev) || (u_long) lp->tx_skb[lp->tx_new] > 1) {
-	if (lp->interrupt) {
-	    de4x5_putb_cache(dev, skb);          /* Requeue the buffer */
-	} else {
-	    de4x5_put_cache(dev, skb);
-	}
-	if (de4x5_debug & DEBUG_TX) {
-	    printk("%s: transmit busy, lost media or stale skb found:\n  STS:%08x\n  tbusy:%d\n  IMR:%08x\n  OMR:%08x\n Stale skb: %s\n",dev->name, inl(DE4X5_STS), netif_queue_stopped(dev), inl(DE4X5_IMR), inl(DE4X5_OMR), ((u_long) lp->tx_skb[lp->tx_new] > 1) ? "YES" : "NO");
-	}
-    } else if (skb->len > 0) {
-	/* If we already have stuff queued locally, use that first */
-	if (!skb_queue_empty(&lp->cache.queue) && !lp->interrupt) {
-	    de4x5_put_cache(dev, skb);
-	    skb = de4x5_get_cache(dev);
-	}
-
-	while (skb && !netif_queue_stopped(dev) &&
-	       (u_long) lp->tx_skb[lp->tx_new] <= 1) {
-	    spin_lock_irqsave(&lp->lock, flags);
-	    netif_stop_queue(dev);
-	    load_packet(dev, skb->data, TD_IC | TD_LS | TD_FS | skb->len, skb);
- 	    lp->stats.tx_bytes += skb->len;
-	    outl(POLL_DEMAND, DE4X5_TPD);/* Start the TX */
-
-	    lp->tx_new = (lp->tx_new + 1) % lp->txRingSize;
-
-	    if (TX_BUFFS_AVAIL) {
-		netif_start_queue(dev);         /* Another pkt may be queued */
-	    }
-	    skb = de4x5_get_cache(dev);
-	    spin_unlock_irqrestore(&lp->lock, flags);
-	}
-	if (skb) de4x5_putb_cache(dev, skb);
-    }
-
-    lp->cache.lock = 0;
-
-    return NETDEV_TX_OK;
-tx_err:
-	dev_kfree_skb_any(skb);
-	return NETDEV_TX_OK;
-}
-
-/*
-** The DE4X5 interrupt handler.
-**
-** I/O Read/Writes through intermediate PCI bridges are never 'posted',
-** so that the asserted interrupt always has some real data to work with -
-** if these I/O accesses are ever changed to memory accesses, ensure the
-** STS write is read immediately to complete the transaction if the adapter
-** is not on bus 0. Lost interrupts can still occur when the PCI bus load
-** is high and descriptor status bits cannot be set before the associated
-** interrupt is asserted and this routine entered.
-*/
-static irqreturn_t
-de4x5_interrupt(int irq, void *dev_id)
-{
-    struct net_device *dev = dev_id;
-    struct de4x5_private *lp;
-    s32 imr, omr, sts, limit;
-    u_long iobase;
-    unsigned int handled = 0;
-
-    lp = netdev_priv(dev);
-    spin_lock(&lp->lock);
-    iobase = dev->base_addr;
-
-    DISABLE_IRQs;                        /* Ensure non re-entrancy */
-
-    if (test_and_set_bit(MASK_INTERRUPTS, (void*) &lp->interrupt))
-	printk("%s: Re-entering the interrupt handler.\n", dev->name);
-
-    synchronize_irq(dev->irq);
-
-    for (limit=0; limit<8; limit++) {
-	sts = inl(DE4X5_STS);            /* Read IRQ status */
-	outl(sts, DE4X5_STS);            /* Reset the board interrupts */
-
-	if (!(sts & lp->irq_mask)) break;/* All done */
-	handled = 1;
-
-	if (sts & (STS_RI | STS_RU))     /* Rx interrupt (packet[s] arrived) */
-	  de4x5_rx(dev);
-
-	if (sts & (STS_TI | STS_TU))     /* Tx interrupt (packet sent) */
-	  de4x5_tx(dev);
-
-	if (sts & STS_LNF) {             /* TP Link has failed */
-	    lp->irq_mask &= ~IMR_LFM;
-	}
-
-	if (sts & STS_UNF) {             /* Transmit underrun */
-	    de4x5_txur(dev);
-	}
-
-	if (sts & STS_SE) {              /* Bus Error */
-	    STOP_DE4X5;
-	    printk("%s: Fatal bus error occurred, sts=%#8x, device stopped.\n",
-		   dev->name, sts);
-	    spin_unlock(&lp->lock);
-	    return IRQ_HANDLED;
-	}
-    }
-
-    /* Load the TX ring with any locally stored packets */
-    if (!test_and_set_bit(0, (void *)&lp->cache.lock)) {
-	while (!skb_queue_empty(&lp->cache.queue) && !netif_queue_stopped(dev) && lp->tx_enable) {
-	    de4x5_queue_pkt(de4x5_get_cache(dev), dev);
-	}
-	lp->cache.lock = 0;
-    }
-
-    lp->interrupt = UNMASK_INTERRUPTS;
-    ENABLE_IRQs;
-    spin_unlock(&lp->lock);
-
-    return IRQ_RETVAL(handled);
-}
-
-static int
-de4x5_rx(struct net_device *dev)
-{
-    struct de4x5_private *lp = netdev_priv(dev);
-    u_long iobase = dev->base_addr;
-    int entry;
-    s32 status;
-
-    for (entry=lp->rx_new; (s32)le32_to_cpu(lp->rx_ring[entry].status)>=0;
-	                                                    entry=lp->rx_new) {
-	status = (s32)le32_to_cpu(lp->rx_ring[entry].status);
-
-	if (lp->rx_ovf) {
-	    if (inl(DE4X5_MFC) & MFC_FOCM) {
-		de4x5_rx_ovfc(dev);
-		break;
-	    }
-	}
-
-	if (status & RD_FS) {                 /* Remember the start of frame */
-	    lp->rx_old = entry;
-	}
-
-	if (status & RD_LS) {                 /* Valid frame status */
-	    if (lp->tx_enable) lp->linkOK++;
-	    if (status & RD_ES) {	      /* There was an error. */
-		lp->stats.rx_errors++;        /* Update the error stats. */
-		if (status & (RD_RF | RD_TL)) lp->stats.rx_frame_errors++;
-		if (status & RD_CE)           lp->stats.rx_crc_errors++;
-		if (status & RD_OF)           lp->stats.rx_fifo_errors++;
-		if (status & RD_TL)           lp->stats.rx_length_errors++;
-		if (status & RD_RF)           lp->pktStats.rx_runt_frames++;
-		if (status & RD_CS)           lp->pktStats.rx_collision++;
-		if (status & RD_DB)           lp->pktStats.rx_dribble++;
-		if (status & RD_OF)           lp->pktStats.rx_overflow++;
-	    } else {                          /* A valid frame received */
-		struct sk_buff *skb;
-		short pkt_len = (short)(le32_to_cpu(lp->rx_ring[entry].status)
-					                            >> 16) - 4;
-
-		if ((skb = de4x5_alloc_rx_buff(dev, entry, pkt_len)) == NULL) {
-		    printk("%s: Insufficient memory; nuking packet.\n",
-			                                            dev->name);
-		    lp->stats.rx_dropped++;
-		} else {
-		    de4x5_dbg_rx(skb, pkt_len);
-
-		    /* Push up the protocol stack */
-		    skb->protocol=eth_type_trans(skb,dev);
-		    de4x5_local_stats(dev, skb->data, pkt_len);
-		    netif_rx(skb);
-
-		    /* Update stats */
-		    lp->stats.rx_packets++;
- 		    lp->stats.rx_bytes += pkt_len;
-		}
-	    }
-
-	    /* Change buffer ownership for this frame, back to the adapter */
-	    for (;lp->rx_old!=entry;lp->rx_old=(lp->rx_old + 1)%lp->rxRingSize) {
-		lp->rx_ring[lp->rx_old].status = cpu_to_le32(R_OWN);
-		barrier();
-	    }
-	    lp->rx_ring[entry].status = cpu_to_le32(R_OWN);
-	    barrier();
-	}
-
-	/*
-	** Update entry information
-	*/
-	lp->rx_new = (lp->rx_new + 1) % lp->rxRingSize;
-    }
-
-    return 0;
-}
-
-static inline void
-de4x5_free_tx_buff(struct de4x5_private *lp, int entry)
-{
-    dma_unmap_single(lp->gendev, le32_to_cpu(lp->tx_ring[entry].buf),
-		     le32_to_cpu(lp->tx_ring[entry].des1) & TD_TBS1,
-		     DMA_TO_DEVICE);
-    if ((u_long) lp->tx_skb[entry] > 1)
-	dev_kfree_skb_irq(lp->tx_skb[entry]);
-    lp->tx_skb[entry] = NULL;
-}
-
-/*
-** Buffer sent - check for TX buffer errors.
-*/
-static int
-de4x5_tx(struct net_device *dev)
-{
-    struct de4x5_private *lp = netdev_priv(dev);
-    u_long iobase = dev->base_addr;
-    int entry;
-    s32 status;
-
-    for (entry = lp->tx_old; entry != lp->tx_new; entry = lp->tx_old) {
-	status = (s32)le32_to_cpu(lp->tx_ring[entry].status);
-	if (status < 0) {                     /* Buffer not sent yet */
-	    break;
-	} else if (status != 0x7fffffff) {    /* Not setup frame */
-	    if (status & TD_ES) {             /* An error happened */
-		lp->stats.tx_errors++;
-		if (status & TD_NC) lp->stats.tx_carrier_errors++;
-		if (status & TD_LC) lp->stats.tx_window_errors++;
-		if (status & TD_UF) lp->stats.tx_fifo_errors++;
-		if (status & TD_EC) lp->pktStats.excessive_collisions++;
-		if (status & TD_DE) lp->stats.tx_aborted_errors++;
-
-		if (TX_PKT_PENDING) {
-		    outl(POLL_DEMAND, DE4X5_TPD);/* Restart a stalled TX */
-		}
-	    } else {                      /* Packet sent */
-		lp->stats.tx_packets++;
-		if (lp->tx_enable) lp->linkOK++;
-	    }
-	    /* Update the collision counter */
-	    lp->stats.collisions += ((status & TD_EC) ? 16 :
-				                      ((status & TD_CC) >> 3));
-
-	    /* Free the buffer. */
-	    if (lp->tx_skb[entry] != NULL)
-	    	de4x5_free_tx_buff(lp, entry);
-	}
-
-	/* Update all the pointers */
-	lp->tx_old = (lp->tx_old + 1) % lp->txRingSize;
-    }
-
-    /* Any resources available? */
-    if (TX_BUFFS_AVAIL && netif_queue_stopped(dev)) {
-	if (lp->interrupt)
-	    netif_wake_queue(dev);
-	else
-	    netif_start_queue(dev);
-    }
-
-    return 0;
-}
-
-static void
-de4x5_ast(struct net_device *dev)
-{
-	struct de4x5_private *lp = netdev_priv(dev);
-	int next_tick = DE4X5_AUTOSENSE_MS;
-	int dt;
-
-	if (lp->useSROM)
-		next_tick = srom_autoconf(dev);
-	else if (lp->chipset == DC21140)
-		next_tick = dc21140m_autoconf(dev);
-	else if (lp->chipset == DC21041)
-		next_tick = dc21041_autoconf(dev);
-	else if (lp->chipset == DC21040)
-		next_tick = dc21040_autoconf(dev);
-	lp->linkOK = 0;
-
-	dt = (next_tick * HZ) / 1000;
-
-	if (!dt)
-		dt = 1;
-
-	mod_timer(&lp->timer, jiffies + dt);
-}
-
-static int
-de4x5_txur(struct net_device *dev)
-{
-    struct de4x5_private *lp = netdev_priv(dev);
-    u_long iobase = dev->base_addr;
-    int omr;
-
-    omr = inl(DE4X5_OMR);
-    if (!(omr & OMR_SF) || (lp->chipset==DC21041) || (lp->chipset==DC21040)) {
-	omr &= ~(OMR_ST|OMR_SR);
-	outl(omr, DE4X5_OMR);
-	while (inl(DE4X5_STS) & STS_TS);
-	if ((omr & OMR_TR) < OMR_TR) {
-	    omr += 0x4000;
-	} else {
-	    omr |= OMR_SF;
-	}
-	outl(omr | OMR_ST | OMR_SR, DE4X5_OMR);
-    }
-
-    return 0;
-}
-
-static int
-de4x5_rx_ovfc(struct net_device *dev)
-{
-    struct de4x5_private *lp = netdev_priv(dev);
-    u_long iobase = dev->base_addr;
-    int omr;
-
-    omr = inl(DE4X5_OMR);
-    outl(omr & ~OMR_SR, DE4X5_OMR);
-    while (inl(DE4X5_STS) & STS_RS);
-
-    for (; (s32)le32_to_cpu(lp->rx_ring[lp->rx_new].status)>=0;) {
-	lp->rx_ring[lp->rx_new].status = cpu_to_le32(R_OWN);
-	lp->rx_new = (lp->rx_new + 1) % lp->rxRingSize;
-    }
-
-    outl(omr, DE4X5_OMR);
-
-    return 0;
-}
-
-static int
-de4x5_close(struct net_device *dev)
-{
-    struct de4x5_private *lp = netdev_priv(dev);
-    u_long iobase = dev->base_addr;
-    s32 imr, omr;
-
-    disable_ast(dev);
-
-    netif_stop_queue(dev);
-
-    if (de4x5_debug & DEBUG_CLOSE) {
-	printk("%s: Shutting down ethercard, status was %8.8x.\n",
-	       dev->name, inl(DE4X5_STS));
-    }
-
-    /*
-    ** We stop the DE4X5 here... mask interrupts and stop TX & RX
-    */
-    DISABLE_IRQs;
-    STOP_DE4X5;
-
-    /* Free the associated irq */
-    free_irq(dev->irq, dev);
-    lp->state = CLOSED;
-
-    /* Free any socket buffers */
-    de4x5_free_rx_buffs(dev);
-    de4x5_free_tx_buffs(dev);
-
-    /* Put the adapter to sleep to save power */
-    yawn(dev, SLEEP);
-
-    return 0;
-}
-
-static struct net_device_stats *
-de4x5_get_stats(struct net_device *dev)
-{
-    struct de4x5_private *lp = netdev_priv(dev);
-    u_long iobase = dev->base_addr;
-
-    lp->stats.rx_missed_errors = (int)(inl(DE4X5_MFC) & (MFC_OVFL | MFC_CNTR));
-
-    return &lp->stats;
-}
-
-static void
-de4x5_local_stats(struct net_device *dev, char *buf, int pkt_len)
-{
-    struct de4x5_private *lp = netdev_priv(dev);
-    int i;
-
-    for (i=1; i<DE4X5_PKT_STAT_SZ-1; i++) {
-        if (pkt_len < (i*DE4X5_PKT_BIN_SZ)) {
-	    lp->pktStats.bins[i]++;
-	    i = DE4X5_PKT_STAT_SZ;
-	}
-    }
-    if (is_multicast_ether_addr(buf)) {
-        if (is_broadcast_ether_addr(buf)) {
-	    lp->pktStats.broadcast++;
-	} else {
-	    lp->pktStats.multicast++;
-	}
-    } else if (ether_addr_equal(buf, dev->dev_addr)) {
-        lp->pktStats.unicast++;
-    }
-
-    lp->pktStats.bins[0]++;       /* Duplicates stats.rx_packets */
-    if (lp->pktStats.bins[0] == 0) { /* Reset counters */
-        memset((char *)&lp->pktStats, 0, sizeof(lp->pktStats));
-    }
-}
-
-/*
-** Removes the TD_IC flag from previous descriptor to improve TX performance.
-** If the flag is changed on a descriptor that is being read by the hardware,
-** I assume PCI transaction ordering will mean you are either successful or
-** just miss asserting the change to the hardware. Anyway you're messing with
-** a descriptor you don't own, but this shouldn't kill the chip provided
-** the descriptor register is read only to the hardware.
-*/
-static void
-load_packet(struct net_device *dev, char *buf, u32 flags, struct sk_buff *skb)
-{
-    struct de4x5_private *lp = netdev_priv(dev);
-    int entry = (lp->tx_new ? lp->tx_new-1 : lp->txRingSize-1);
-    dma_addr_t buf_dma = dma_map_single(lp->gendev, buf, flags & TD_TBS1, DMA_TO_DEVICE);
-
-    lp->tx_ring[lp->tx_new].buf = cpu_to_le32(buf_dma);
-    lp->tx_ring[lp->tx_new].des1 &= cpu_to_le32(TD_TER);
-    lp->tx_ring[lp->tx_new].des1 |= cpu_to_le32(flags);
-    lp->tx_skb[lp->tx_new] = skb;
-    lp->tx_ring[entry].des1 &= cpu_to_le32(~TD_IC);
-    barrier();
-
-    lp->tx_ring[lp->tx_new].status = cpu_to_le32(T_OWN);
-    barrier();
-}
-
-/*
-** Set or clear the multicast filter for this adaptor.
-*/
-static void
-set_multicast_list(struct net_device *dev)
-{
-    struct de4x5_private *lp = netdev_priv(dev);
-    u_long iobase = dev->base_addr;
-
-    /* First, double check that the adapter is open */
-    if (lp->state == OPEN) {
-	if (dev->flags & IFF_PROMISC) {         /* set promiscuous mode */
-	    u32 omr;
-	    omr = inl(DE4X5_OMR);
-	    omr |= OMR_PR;
-	    outl(omr, DE4X5_OMR);
-	} else {
-	    SetMulticastFilter(dev);
-	    load_packet(dev, lp->setup_frame, TD_IC | PERFECT_F | TD_SET |
-			                                SETUP_FRAME_LEN, (struct sk_buff *)1);
-
-	    lp->tx_new = (lp->tx_new + 1) % lp->txRingSize;
-	    outl(POLL_DEMAND, DE4X5_TPD);       /* Start the TX */
-	    netif_trans_update(dev); /* prevent tx timeout */
-	}
-    }
-}
-
-/*
-** Calculate the hash code and update the logical address filter
-** from a list of ethernet multicast addresses.
-** Little endian crc one liner from Matt Thomas, DEC.
-*/
-static void
-SetMulticastFilter(struct net_device *dev)
-{
-    struct de4x5_private *lp = netdev_priv(dev);
-    struct netdev_hw_addr *ha;
-    u_long iobase = dev->base_addr;
-    int i, bit, byte;
-    u16 hashcode;
-    u32 omr, crc;
-    char *pa;
-    unsigned char *addrs;
-
-    omr = inl(DE4X5_OMR);
-    omr &= ~(OMR_PR | OMR_PM);
-    pa = build_setup_frame(dev, ALL);        /* Build the basic frame */
-
-    if ((dev->flags & IFF_ALLMULTI) || (netdev_mc_count(dev) > 14)) {
-	omr |= OMR_PM;                       /* Pass all multicasts */
-    } else if (lp->setup_f == HASH_PERF) {   /* Hash Filtering */
-	netdev_for_each_mc_addr(ha, dev) {
-		crc = ether_crc_le(ETH_ALEN, ha->addr);
-		hashcode = crc & DE4X5_HASH_BITS;  /* hashcode is 9 LSb of CRC */
-
-		byte = hashcode >> 3;        /* bit[3-8] -> byte in filter */
-		bit = 1 << (hashcode & 0x07);/* bit[0-2] -> bit in byte */
-
-		byte <<= 1;                  /* calc offset into setup frame */
-		if (byte & 0x02) {
-		    byte -= 1;
-		}
-		lp->setup_frame[byte] |= bit;
-	}
-    } else {                                 /* Perfect filtering */
-	netdev_for_each_mc_addr(ha, dev) {
-	    addrs = ha->addr;
-	    for (i=0; i<ETH_ALEN; i++) {
-		*(pa + (i&1)) = *addrs++;
-		if (i & 0x01) pa += 4;
-	    }
-	}
-    }
-    outl(omr, DE4X5_OMR);
-}
-
-#ifdef CONFIG_EISA
-
-static u_char de4x5_irq[] = EISA_ALLOWED_IRQ_LIST;
-
-static int de4x5_eisa_probe(struct device *gendev)
-{
-	struct eisa_device *edev;
-	u_long iobase;
-	u_char irq, regval;
-	u_short vendor;
-	u32 cfid;
-	int status, device;
-	struct net_device *dev;
-	struct de4x5_private *lp;
-
-	edev = to_eisa_device (gendev);
-	iobase = edev->base_addr;
-
-	if (!request_region (iobase, DE4X5_EISA_TOTAL_SIZE, "de4x5"))
-		return -EBUSY;
-
-	if (!request_region (iobase + DE4X5_EISA_IO_PORTS,
-			     DE4X5_EISA_TOTAL_SIZE, "de4x5")) {
-		status = -EBUSY;
-		goto release_reg_1;
-	}
-
-	if (!(dev = alloc_etherdev (sizeof (struct de4x5_private)))) {
-		status = -ENOMEM;
-		goto release_reg_2;
-	}
-	lp = netdev_priv(dev);
-
-	cfid = (u32) inl(PCI_CFID);
-	lp->cfrv = (u_short) inl(PCI_CFRV);
-	device = (cfid >> 8) & 0x00ffff00;
-	vendor = (u_short) cfid;
-
-	/* Read the EISA Configuration Registers */
-	regval = inb(EISA_REG0) & (ER0_INTL | ER0_INTT);
-#ifdef CONFIG_ALPHA
-	/* Looks like the Jensen firmware (rev 2.2) doesn't really
-	 * care about the EISA configuration, and thus doesn't
-	 * configure the PLX bridge properly. Oh well... Simply mimic
-	 * the EISA config file to sort it out. */
-
-	/* EISA REG1: Assert DecChip 21040 HW Reset */
-	outb (ER1_IAM | 1, EISA_REG1);
-	mdelay (1);
-
-        /* EISA REG1: Deassert DecChip 21040 HW Reset */
-	outb (ER1_IAM, EISA_REG1);
-	mdelay (1);
-
-	/* EISA REG3: R/W Burst Transfer Enable */
-	outb (ER3_BWE | ER3_BRE, EISA_REG3);
-
-	/* 32_bit slave/master, Preempt Time=23 bclks, Unlatched Interrupt */
-	outb (ER0_BSW | ER0_BMW | ER0_EPT | regval, EISA_REG0);
-#endif
-	irq = de4x5_irq[(regval >> 1) & 0x03];
-
-	if (is_DC2114x) {
-	    device = ((lp->cfrv & CFRV_RN) < DC2114x_BRK ? DC21142 : DC21143);
-	}
-	lp->chipset = device;
-	lp->bus = EISA;
-
-	/* Write the PCI Configuration Registers */
-	outl(PCI_COMMAND_IO | PCI_COMMAND_MASTER, PCI_CFCS);
-	outl(0x00006000, PCI_CFLT);
-	outl(iobase, PCI_CBIO);
-
-	DevicePresent(dev, EISA_APROM);
-
-	dev->irq = irq;
-
-	if (!(status = de4x5_hw_init (dev, iobase, gendev))) {
-		return 0;
-	}
-
-	free_netdev (dev);
- release_reg_2:
-	release_region (iobase + DE4X5_EISA_IO_PORTS, DE4X5_EISA_TOTAL_SIZE);
- release_reg_1:
-	release_region (iobase, DE4X5_EISA_TOTAL_SIZE);
-
-	return status;
-}
-
-static int de4x5_eisa_remove(struct device *device)
-{
-	struct net_device *dev;
-	u_long iobase;
-
-	dev = dev_get_drvdata(device);
-	iobase = dev->base_addr;
-
-	unregister_netdev (dev);
-	free_netdev (dev);
-	release_region (iobase + DE4X5_EISA_IO_PORTS, DE4X5_EISA_TOTAL_SIZE);
-	release_region (iobase, DE4X5_EISA_TOTAL_SIZE);
-
-	return 0;
-}
-
-static struct eisa_device_id de4x5_eisa_ids[] = {
-        { "DEC4250", 0 },	/* 0 is the board name index... */
-        { "" }
-};
-MODULE_DEVICE_TABLE(eisa, de4x5_eisa_ids);
-
-static struct eisa_driver de4x5_eisa_driver = {
-        .id_table = de4x5_eisa_ids,
-        .driver   = {
-                .name    = "de4x5",
-                .probe   = de4x5_eisa_probe,
-		.remove  = de4x5_eisa_remove,
-        }
-};
-MODULE_DEVICE_TABLE(eisa, de4x5_eisa_ids);
-#endif
-
-#ifdef CONFIG_PCI
-
-/*
-** This function searches the current bus (which is >0) for a DECchip with an
-** SROM, so that in multiport cards that have one SROM shared between multiple
-** DECchips, we can find the base SROM irrespective of the BIOS scan direction.
-** For single port cards this is a time waster...
-*/
-static void
-srom_search(struct net_device *dev, struct pci_dev *pdev)
-{
-    u_char pb;
-    u_short vendor, status;
-    u_int irq = 0, device;
-    u_long iobase = 0;                     /* Clear upper 32 bits in Alphas */
-    int i, j;
-    struct de4x5_private *lp = netdev_priv(dev);
-    struct pci_dev *this_dev;
-
-    list_for_each_entry(this_dev, &pdev->bus->devices, bus_list) {
-	vendor = this_dev->vendor;
-	device = this_dev->device << 8;
-	if (!(is_DC21040 || is_DC21041 || is_DC21140 || is_DC2114x)) continue;
-
-	/* Get the chip configuration revision register */
-	pb = this_dev->bus->number;
-
-	/* Set the device number information */
-	lp->device = PCI_SLOT(this_dev->devfn);
-	lp->bus_num = pb;
-
-	/* Set the chipset information */
-	if (is_DC2114x) {
-	    device = ((this_dev->revision & CFRV_RN) < DC2114x_BRK
-		      ? DC21142 : DC21143);
-	}
-	lp->chipset = device;
-
-	/* Get the board I/O address (64 bits on sparc64) */
-	iobase = pci_resource_start(this_dev, 0);
-
-	/* Fetch the IRQ to be used */
-	irq = this_dev->irq;
-	if ((irq == 0) || (irq == 0xff) || ((int)irq == -1)) continue;
-
-	/* Check if I/O accesses are enabled */
-	pci_read_config_word(this_dev, PCI_COMMAND, &status);
-	if (!(status & PCI_COMMAND_IO)) continue;
-
-	/* Search for a valid SROM attached to this DECchip */
-	DevicePresent(dev, DE4X5_APROM);
-	for (j=0, i=0; i<ETH_ALEN; i++) {
-	    j += (u_char) *((u_char *)&lp->srom + SROM_HWADD + i);
-	}
-	if (j != 0 && j != 6 * 0xff) {
-	    last.chipset = device;
-	    last.bus = pb;
-	    last.irq = irq;
-	    for (i=0; i<ETH_ALEN; i++) {
-		last.addr[i] = (u_char)*((u_char *)&lp->srom + SROM_HWADD + i);
-	    }
-	    return;
-	}
-    }
-}
-
-/*
-** PCI bus I/O device probe
-** NB: PCI I/O accesses and Bus Mastering are enabled by the PCI BIOS, not
-** the driver. Some PCI BIOS's, pre V2.1, need the slot + features to be
-** enabled by the user first in the set up utility. Hence we just check for
-** enabled features and silently ignore the card if they're not.
-**
-** STOP PRESS: Some BIOS's __require__ the driver to enable the bus mastering
-** bit. Here, check for I/O accesses and then set BM. If you put the card in
-** a non BM slot, you're on your own (and complain to the PC vendor that your
-** PC doesn't conform to the PCI standard)!
-**
-** This function is only compatible with the *latest* 2.1.x kernels. For 2.0.x
-** kernels use the V0.535[n] drivers.
-*/
-
-static int de4x5_pci_probe(struct pci_dev *pdev,
-			   const struct pci_device_id *ent)
-{
-	u_char pb, pbus = 0, dev_num, dnum = 0, timer;
-	u_short vendor, status;
-	u_int irq = 0, device;
-	u_long iobase = 0;	/* Clear upper 32 bits in Alphas */
-	int error;
-	struct net_device *dev;
-	struct de4x5_private *lp;
-
-	dev_num = PCI_SLOT(pdev->devfn);
-	pb = pdev->bus->number;
-
-	if (io) { /* probe a single PCI device */
-		pbus = (u_short)(io >> 8);
-		dnum = (u_short)(io & 0xff);
-		if ((pbus != pb) || (dnum != dev_num))
-			return -ENODEV;
-	}
-
-	vendor = pdev->vendor;
-	device = pdev->device << 8;
-	if (!(is_DC21040 || is_DC21041 || is_DC21140 || is_DC2114x))
-		return -ENODEV;
-
-	/* Ok, the device seems to be for us. */
-	if ((error = pci_enable_device (pdev)))
-		return error;
-
-	if (!(dev = alloc_etherdev (sizeof (struct de4x5_private)))) {
-		error = -ENOMEM;
-		goto disable_dev;
-	}
-
-	lp = netdev_priv(dev);
-	lp->bus = PCI;
-	lp->bus_num = 0;
-
-	/* Search for an SROM on this bus */
-	if (lp->bus_num != pb) {
-	    lp->bus_num = pb;
-	    srom_search(dev, pdev);
-	}
-
-	/* Get the chip configuration revision register */
-	lp->cfrv = pdev->revision;
-
-	/* Set the device number information */
-	lp->device = dev_num;
-	lp->bus_num = pb;
-
-	/* Set the chipset information */
-	if (is_DC2114x) {
-	    device = ((lp->cfrv & CFRV_RN) < DC2114x_BRK ? DC21142 : DC21143);
-	}
-	lp->chipset = device;
-
-	/* Get the board I/O address (64 bits on sparc64) */
-	iobase = pci_resource_start(pdev, 0);
-
-	/* Fetch the IRQ to be used */
-	irq = pdev->irq;
-	if ((irq == 0) || (irq == 0xff) || ((int)irq == -1)) {
-		error = -ENODEV;
-		goto free_dev;
-	}
-
-	/* Check if I/O accesses and Bus Mastering are enabled */
-	pci_read_config_word(pdev, PCI_COMMAND, &status);
-#ifdef __powerpc__
-	if (!(status & PCI_COMMAND_IO)) {
-	    status |= PCI_COMMAND_IO;
-	    pci_write_config_word(pdev, PCI_COMMAND, status);
-	    pci_read_config_word(pdev, PCI_COMMAND, &status);
-	}
-#endif /* __powerpc__ */
-	if (!(status & PCI_COMMAND_IO)) {
-		error = -ENODEV;
-		goto free_dev;
-	}
-
-	if (!(status & PCI_COMMAND_MASTER)) {
-	    status |= PCI_COMMAND_MASTER;
-	    pci_write_config_word(pdev, PCI_COMMAND, status);
-	    pci_read_config_word(pdev, PCI_COMMAND, &status);
-	}
-	if (!(status & PCI_COMMAND_MASTER)) {
-		error = -ENODEV;
-		goto free_dev;
-	}
-
-	/* Check the latency timer for values >= 0x60 */
-	pci_read_config_byte(pdev, PCI_LATENCY_TIMER, &timer);
-	if (timer < 0x60) {
-	    pci_write_config_byte(pdev, PCI_LATENCY_TIMER, 0x60);
-	}
-
-	DevicePresent(dev, DE4X5_APROM);
-
-	if (!request_region (iobase, DE4X5_PCI_TOTAL_SIZE, "de4x5")) {
-		error = -EBUSY;
-		goto free_dev;
-	}
-
-	dev->irq = irq;
-
-	if ((error = de4x5_hw_init(dev, iobase, &pdev->dev))) {
-		goto release;
-	}
-
-	return 0;
-
- release:
-	release_region (iobase, DE4X5_PCI_TOTAL_SIZE);
- free_dev:
-	free_netdev (dev);
- disable_dev:
-	pci_disable_device (pdev);
-	return error;
-}
-
-static void de4x5_pci_remove(struct pci_dev *pdev)
-{
-	struct net_device *dev;
-	u_long iobase;
-
-	dev = pci_get_drvdata(pdev);
-	iobase = dev->base_addr;
-
-	unregister_netdev (dev);
-	free_netdev (dev);
-	release_region (iobase, DE4X5_PCI_TOTAL_SIZE);
-	pci_disable_device (pdev);
-}
-
-static const struct pci_device_id de4x5_pci_tbl[] = {
-        { PCI_VENDOR_ID_DEC, PCI_DEVICE_ID_DEC_TULIP,
-          PCI_ANY_ID, PCI_ANY_ID, 0, 0, 0 },
-        { PCI_VENDOR_ID_DEC, PCI_DEVICE_ID_DEC_TULIP_PLUS,
-          PCI_ANY_ID, PCI_ANY_ID, 0, 0, 1 },
-        { PCI_VENDOR_ID_DEC, PCI_DEVICE_ID_DEC_TULIP_FAST,
-	  PCI_ANY_ID, PCI_ANY_ID, 0, 0, 2 },
-        { PCI_VENDOR_ID_DEC, PCI_DEVICE_ID_DEC_21142,
-	  PCI_ANY_ID, PCI_ANY_ID, 0, 0, 3 },
-        { },
-};
-
-static struct pci_driver de4x5_pci_driver = {
-        .name           = "de4x5",
-        .id_table       = de4x5_pci_tbl,
-        .probe          = de4x5_pci_probe,
-	.remove         = de4x5_pci_remove,
-};
-
-#endif
-
-/*
-** Auto configure the media here rather than setting the port at compile
-** time. This routine is called by de4x5_init() and when a loss of media is
-** detected (excessive collisions, loss of carrier, no carrier or link fail
-** [TP] or no recent receive activity) to check whether the user has been
-** sneaky and changed the port on us.
-*/
-static int
-autoconf_media(struct net_device *dev)
-{
-	struct de4x5_private *lp = netdev_priv(dev);
-	u_long iobase = dev->base_addr;
-
-	disable_ast(dev);
-
-	lp->c_media = AUTO;                     /* Bogus last media */
-	inl(DE4X5_MFC);                         /* Zero the lost frames counter */
-	lp->media = INIT;
-	lp->tcount = 0;
-
-	de4x5_ast(dev);
-
-	return lp->media;
-}
-
-/*
-** Autoconfigure the media when using the DC21040. AUI cannot be distinguished
-** from BNC as the port has a jumper to set thick or thin wire. When set for
-** BNC, the BNC port will indicate activity if it's not terminated correctly.
-** The only way to test for that is to place a loopback packet onto the
-** network and watch for errors. Since we're messing with the interrupt mask
-** register, disable the board interrupts and do not allow any more packets to
-** be queued to the hardware. Re-enable everything only when the media is
-** found.
-** I may have to "age out" locally queued packets so that the higher layer
-** timeouts don't effectively duplicate packets on the network.
-*/
-static int
-dc21040_autoconf(struct net_device *dev)
-{
-    struct de4x5_private *lp = netdev_priv(dev);
-    u_long iobase = dev->base_addr;
-    int next_tick = DE4X5_AUTOSENSE_MS;
-    s32 imr;
-
-    switch (lp->media) {
-    case INIT:
-	DISABLE_IRQs;
-	lp->tx_enable = false;
-	lp->timeout = -1;
-	de4x5_save_skbs(dev);
-	if ((lp->autosense == AUTO) || (lp->autosense == TP)) {
-	    lp->media = TP;
-	} else if ((lp->autosense == BNC) || (lp->autosense == AUI) || (lp->autosense == BNC_AUI)) {
-	    lp->media = BNC_AUI;
-	} else if (lp->autosense == EXT_SIA) {
-	    lp->media = EXT_SIA;
-	} else {
-	    lp->media = NC;
-	}
-	lp->local_state = 0;
-	next_tick = dc21040_autoconf(dev);
-	break;
-
-    case TP:
-	next_tick = dc21040_state(dev, 0x8f01, 0xffff, 0x0000, 3000, BNC_AUI,
-		                                         TP_SUSPECT, test_tp);
-	break;
-
-    case TP_SUSPECT:
-	next_tick = de4x5_suspect_state(dev, 1000, TP, test_tp, dc21040_autoconf);
-	break;
-
-    case BNC:
-    case AUI:
-    case BNC_AUI:
-	next_tick = dc21040_state(dev, 0x8f09, 0x0705, 0x0006, 3000, EXT_SIA,
-		                                  BNC_AUI_SUSPECT, ping_media);
-	break;
-
-    case BNC_AUI_SUSPECT:
-	next_tick = de4x5_suspect_state(dev, 1000, BNC_AUI, ping_media, dc21040_autoconf);
-	break;
-
-    case EXT_SIA:
-	next_tick = dc21040_state(dev, 0x3041, 0x0000, 0x0006, 3000,
-		                              NC, EXT_SIA_SUSPECT, ping_media);
-	break;
-
-    case EXT_SIA_SUSPECT:
-	next_tick = de4x5_suspect_state(dev, 1000, EXT_SIA, ping_media, dc21040_autoconf);
-	break;
-
-    case NC:
-	/* default to TP for all */
-	reset_init_sia(dev, 0x8f01, 0xffff, 0x0000);
-	if (lp->media != lp->c_media) {
-	    de4x5_dbg_media(dev);
-	    lp->c_media = lp->media;
-	}
-	lp->media = INIT;
-	lp->tx_enable = false;
-	break;
-    }
-
-    return next_tick;
-}
-
-static int
-dc21040_state(struct net_device *dev, int csr13, int csr14, int csr15, int timeout,
-	      int next_state, int suspect_state,
-	      int (*fn)(struct net_device *, int))
-{
-    struct de4x5_private *lp = netdev_priv(dev);
-    int next_tick = DE4X5_AUTOSENSE_MS;
-    int linkBad;
-
-    switch (lp->local_state) {
-    case 0:
-	reset_init_sia(dev, csr13, csr14, csr15);
-	lp->local_state++;
-	next_tick = 500;
-	break;
-
-    case 1:
-	if (!lp->tx_enable) {
-	    linkBad = fn(dev, timeout);
-	    if (linkBad < 0) {
-		next_tick = linkBad & ~TIMER_CB;
-	    } else {
-		if (linkBad && (lp->autosense == AUTO)) {
-		    lp->local_state = 0;
-		    lp->media = next_state;
-		} else {
-		    de4x5_init_connection(dev);
-		}
-	    }
-	} else if (!lp->linkOK && (lp->autosense == AUTO)) {
-	    lp->media = suspect_state;
-	    next_tick = 3000;
-	}
-	break;
-    }
-
-    return next_tick;
-}
-
-static int
-de4x5_suspect_state(struct net_device *dev, int timeout, int prev_state,
-		      int (*fn)(struct net_device *, int),
-		      int (*asfn)(struct net_device *))
-{
-    struct de4x5_private *lp = netdev_priv(dev);
-    int next_tick = DE4X5_AUTOSENSE_MS;
-    int linkBad;
-
-    switch (lp->local_state) {
-    case 1:
-	if (lp->linkOK) {
-	    lp->media = prev_state;
-	} else {
-	    lp->local_state++;
-	    next_tick = asfn(dev);
-	}
-	break;
-
-    case 2:
-	linkBad = fn(dev, timeout);
-	if (linkBad < 0) {
-	    next_tick = linkBad & ~TIMER_CB;
-	} else if (!linkBad) {
-	    lp->local_state--;
-	    lp->media = prev_state;
-	} else {
-	    lp->media = INIT;
-	    lp->tcount++;
-	}
-    }
-
-    return next_tick;
-}
-
-/*
-** Autoconfigure the media when using the DC21041. AUI needs to be tested
-** before BNC, because the BNC port will indicate activity if it's not
-** terminated correctly. The only way to test for that is to place a loopback
-** packet onto the network and watch for errors. Since we're messing with
-** the interrupt mask register, disable the board interrupts and do not allow
-** any more packets to be queued to the hardware. Re-enable everything only
-** when the media is found.
-*/
-static int
-dc21041_autoconf(struct net_device *dev)
-{
-    struct de4x5_private *lp = netdev_priv(dev);
-    u_long iobase = dev->base_addr;
-    s32 sts, irqs, irq_mask, imr, omr;
-    int next_tick = DE4X5_AUTOSENSE_MS;
-
-    switch (lp->media) {
-    case INIT:
-	DISABLE_IRQs;
-	lp->tx_enable = false;
-	lp->timeout = -1;
-	de4x5_save_skbs(dev);          /* Save non transmitted skb's */
-	if ((lp->autosense == AUTO) || (lp->autosense == TP_NW)) {
-	    lp->media = TP;            /* On chip auto negotiation is broken */
-	} else if (lp->autosense == TP) {
-	    lp->media = TP;
-	} else if (lp->autosense == BNC) {
-	    lp->media = BNC;
-	} else if (lp->autosense == AUI) {
-	    lp->media = AUI;
-	} else {
-	    lp->media = NC;
-	}
-	lp->local_state = 0;
-	next_tick = dc21041_autoconf(dev);
-	break;
-
-    case TP_NW:
-	if (lp->timeout < 0) {
-	    omr = inl(DE4X5_OMR);/* Set up full duplex for the autonegotiate */
-	    outl(omr | OMR_FDX, DE4X5_OMR);
-	}
-	irqs = STS_LNF | STS_LNP;
-	irq_mask = IMR_LFM | IMR_LPM;
-	sts = test_media(dev, irqs, irq_mask, 0xef01, 0xffff, 0x0008, 2400);
-	if (sts < 0) {
-	    next_tick = sts & ~TIMER_CB;
-	} else {
-	    if (sts & STS_LNP) {
-		lp->media = ANS;
-	    } else {
-		lp->media = AUI;
-	    }
-	    next_tick = dc21041_autoconf(dev);
-	}
-	break;
-
-    case ANS:
-	if (!lp->tx_enable) {
-	    irqs = STS_LNP;
-	    irq_mask = IMR_LPM;
-	    sts = test_ans(dev, irqs, irq_mask, 3000);
-	    if (sts < 0) {
-		next_tick = sts & ~TIMER_CB;
-	    } else {
-		if (!(sts & STS_LNP) && (lp->autosense == AUTO)) {
-		    lp->media = TP;
-		    next_tick = dc21041_autoconf(dev);
-		} else {
-		    lp->local_state = 1;
-		    de4x5_init_connection(dev);
-		}
-	    }
-	} else if (!lp->linkOK && (lp->autosense == AUTO)) {
-	    lp->media = ANS_SUSPECT;
-	    next_tick = 3000;
-	}
-	break;
-
-    case ANS_SUSPECT:
-	next_tick = de4x5_suspect_state(dev, 1000, ANS, test_tp, dc21041_autoconf);
-	break;
-
-    case TP:
-	if (!lp->tx_enable) {
-	    if (lp->timeout < 0) {
-		omr = inl(DE4X5_OMR);          /* Set up half duplex for TP */
-		outl(omr & ~OMR_FDX, DE4X5_OMR);
-	    }
-	    irqs = STS_LNF | STS_LNP;
-	    irq_mask = IMR_LFM | IMR_LPM;
-	    sts = test_media(dev,irqs, irq_mask, 0xef01, 0xff3f, 0x0008, 2400);
-	    if (sts < 0) {
-		next_tick = sts & ~TIMER_CB;
-	    } else {
-		if (!(sts & STS_LNP) && (lp->autosense == AUTO)) {
-		    if (inl(DE4X5_SISR) & SISR_NRA) {
-			lp->media = AUI;       /* Non selected port activity */
-		    } else {
-			lp->media = BNC;
-		    }
-		    next_tick = dc21041_autoconf(dev);
-		} else {
-		    lp->local_state = 1;
-		    de4x5_init_connection(dev);
-		}
-	    }
-	} else if (!lp->linkOK && (lp->autosense == AUTO)) {
-	    lp->media = TP_SUSPECT;
-	    next_tick = 3000;
-	}
-	break;
-
-    case TP_SUSPECT:
-	next_tick = de4x5_suspect_state(dev, 1000, TP, test_tp, dc21041_autoconf);
-	break;
-
-    case AUI:
-	if (!lp->tx_enable) {
-	    if (lp->timeout < 0) {
-		omr = inl(DE4X5_OMR);          /* Set up half duplex for AUI */
-		outl(omr & ~OMR_FDX, DE4X5_OMR);
-	    }
-	    irqs = 0;
-	    irq_mask = 0;
-	    sts = test_media(dev,irqs, irq_mask, 0xef09, 0xf73d, 0x000e, 1000);
-	    if (sts < 0) {
-		next_tick = sts & ~TIMER_CB;
-	    } else {
-		if (!(inl(DE4X5_SISR) & SISR_SRA) && (lp->autosense == AUTO)) {
-		    lp->media = BNC;
-		    next_tick = dc21041_autoconf(dev);
-		} else {
-		    lp->local_state = 1;
-		    de4x5_init_connection(dev);
-		}
-	    }
-	} else if (!lp->linkOK && (lp->autosense == AUTO)) {
-	    lp->media = AUI_SUSPECT;
-	    next_tick = 3000;
-	}
-	break;
-
-    case AUI_SUSPECT:
-	next_tick = de4x5_suspect_state(dev, 1000, AUI, ping_media, dc21041_autoconf);
-	break;
-
-    case BNC:
-	switch (lp->local_state) {
-	case 0:
-	    if (lp->timeout < 0) {
-		omr = inl(DE4X5_OMR);          /* Set up half duplex for BNC */
-		outl(omr & ~OMR_FDX, DE4X5_OMR);
-	    }
-	    irqs = 0;
-	    irq_mask = 0;
-	    sts = test_media(dev,irqs, irq_mask, 0xef09, 0xf73d, 0x0006, 1000);
-	    if (sts < 0) {
-		next_tick = sts & ~TIMER_CB;
-	    } else {
-		lp->local_state++;             /* Ensure media connected */
-		next_tick = dc21041_autoconf(dev);
-	    }
-	    break;
-
-	case 1:
-	    if (!lp->tx_enable) {
-		if ((sts = ping_media(dev, 3000)) < 0) {
-		    next_tick = sts & ~TIMER_CB;
-		} else {
-		    if (sts) {
-			lp->local_state = 0;
-			lp->media = NC;
-		    } else {
-			de4x5_init_connection(dev);
-		    }
-		}
-	    } else if (!lp->linkOK && (lp->autosense == AUTO)) {
-		lp->media = BNC_SUSPECT;
-		next_tick = 3000;
-	    }
-	    break;
-	}
-	break;
-
-    case BNC_SUSPECT:
-	next_tick = de4x5_suspect_state(dev, 1000, BNC, ping_media, dc21041_autoconf);
-	break;
-
-    case NC:
-	omr = inl(DE4X5_OMR);    /* Set up full duplex for the autonegotiate */
-	outl(omr | OMR_FDX, DE4X5_OMR);
-	reset_init_sia(dev, 0xef01, 0xffff, 0x0008);/* Initialise the SIA */
-	if (lp->media != lp->c_media) {
-	    de4x5_dbg_media(dev);
-	    lp->c_media = lp->media;
-	}
-	lp->media = INIT;
-	lp->tx_enable = false;
-	break;
-    }
-
-    return next_tick;
-}
-
-/*
-** Some autonegotiation chips are broken in that they do not return the
-** acknowledge bit (anlpa & MII_ANLPA_ACK) in the link partner advertisement
-** register, except at the first power up negotiation.
-*/
-static int
-dc21140m_autoconf(struct net_device *dev)
-{
-    struct de4x5_private *lp = netdev_priv(dev);
-    int ana, anlpa, cap, cr, slnk, sr;
-    int next_tick = DE4X5_AUTOSENSE_MS;
-    u_long imr, omr, iobase = dev->base_addr;
-
-    switch(lp->media) {
-    case INIT:
-        if (lp->timeout < 0) {
-	    DISABLE_IRQs;
-	    lp->tx_enable = false;
-	    lp->linkOK = 0;
-	    de4x5_save_skbs(dev);          /* Save non transmitted skb's */
-	}
-	if ((next_tick = de4x5_reset_phy(dev)) < 0) {
-	    next_tick &= ~TIMER_CB;
-	} else {
-	    if (lp->useSROM) {
-		if (srom_map_media(dev) < 0) {
-		    lp->tcount++;
-		    return next_tick;
-		}
-		srom_exec(dev, lp->phy[lp->active].gep);
-		if (lp->infoblock_media == ANS) {
-		    ana = lp->phy[lp->active].ana | MII_ANA_CSMA;
-		    mii_wr(ana, MII_ANA, lp->phy[lp->active].addr, DE4X5_MII);
-		}
-	    } else {
-		lp->tmp = MII_SR_ASSC;     /* Fake out the MII speed set */
-		SET_10Mb;
-		if (lp->autosense == _100Mb) {
-		    lp->media = _100Mb;
-		} else if (lp->autosense == _10Mb) {
-		    lp->media = _10Mb;
-		} else if ((lp->autosense == AUTO) &&
-			            ((sr=is_anc_capable(dev)) & MII_SR_ANC)) {
-		    ana = (((sr >> 6) & MII_ANA_TAF) | MII_ANA_CSMA);
-		    ana &= (lp->fdx ? ~0 : ~MII_ANA_FDAM);
-		    mii_wr(ana, MII_ANA, lp->phy[lp->active].addr, DE4X5_MII);
-		    lp->media = ANS;
-		} else if (lp->autosense == AUTO) {
-		    lp->media = SPD_DET;
-		} else if (is_spd_100(dev) && is_100_up(dev)) {
-		    lp->media = _100Mb;
-		} else {
-		    lp->media = NC;
-		}
-	    }
-	    lp->local_state = 0;
-	    next_tick = dc21140m_autoconf(dev);
-	}
-	break;
-
-    case ANS:
-	switch (lp->local_state) {
-	case 0:
-	    if (lp->timeout < 0) {
-		mii_wr(MII_CR_ASSE | MII_CR_RAN, MII_CR, lp->phy[lp->active].addr, DE4X5_MII);
-	    }
-	    cr = test_mii_reg(dev, MII_CR, MII_CR_RAN, false, 500);
-	    if (cr < 0) {
-		next_tick = cr & ~TIMER_CB;
-	    } else {
-		if (cr) {
-		    lp->local_state = 0;
-		    lp->media = SPD_DET;
-		} else {
-		    lp->local_state++;
-		}
-		next_tick = dc21140m_autoconf(dev);
-	    }
-	    break;
-
-	case 1:
-	    if ((sr=test_mii_reg(dev, MII_SR, MII_SR_ASSC, true, 2000)) < 0) {
-		next_tick = sr & ~TIMER_CB;
-	    } else {
-		lp->media = SPD_DET;
-		lp->local_state = 0;
-		if (sr) {                         /* Success! */
-		    lp->tmp = MII_SR_ASSC;
-		    anlpa = mii_rd(MII_ANLPA, lp->phy[lp->active].addr, DE4X5_MII);
-		    ana = mii_rd(MII_ANA, lp->phy[lp->active].addr, DE4X5_MII);
-		    if (!(anlpa & MII_ANLPA_RF) &&
-			 (cap = anlpa & MII_ANLPA_TAF & ana)) {
-			if (cap & MII_ANA_100M) {
-			    lp->fdx = (ana & anlpa & MII_ANA_FDAM & MII_ANA_100M) != 0;
-			    lp->media = _100Mb;
-			} else if (cap & MII_ANA_10M) {
-			    lp->fdx = (ana & anlpa & MII_ANA_FDAM & MII_ANA_10M) != 0;
-
-			    lp->media = _10Mb;
-			}
-		    }
-		}                       /* Auto Negotiation failed to finish */
-		next_tick = dc21140m_autoconf(dev);
-	    }                           /* Auto Negotiation failed to start */
-	    break;
-	}
-	break;
-
-    case SPD_DET:                              /* Choose 10Mb/s or 100Mb/s */
-        if (lp->timeout < 0) {
-	    lp->tmp = (lp->phy[lp->active].id ? MII_SR_LKS :
-		                                  (~gep_rd(dev) & GEP_LNP));
-	    SET_100Mb_PDET;
-	}
-        if ((slnk = test_for_100Mb(dev, 6500)) < 0) {
-	    next_tick = slnk & ~TIMER_CB;
-	} else {
-	    if (is_spd_100(dev) && is_100_up(dev)) {
-		lp->media = _100Mb;
-	    } else if ((!is_spd_100(dev) && (is_10_up(dev) & lp->tmp))) {
-		lp->media = _10Mb;
-	    } else {
-		lp->media = NC;
-	    }
-	    next_tick = dc21140m_autoconf(dev);
-	}
-	break;
-
-    case _100Mb:                               /* Set 100Mb/s */
-        next_tick = 3000;
-	if (!lp->tx_enable) {
-	    SET_100Mb;
-	    de4x5_init_connection(dev);
-	} else {
-	    if (!lp->linkOK && (lp->autosense == AUTO)) {
-		if (!is_100_up(dev) || (!lp->useSROM && !is_spd_100(dev))) {
-		    lp->media = INIT;
-		    lp->tcount++;
-		    next_tick = DE4X5_AUTOSENSE_MS;
-		}
-	    }
-	}
-	break;
-
-    case BNC:
-    case AUI:
-    case _10Mb:                                /* Set 10Mb/s */
-        next_tick = 3000;
-	if (!lp->tx_enable) {
-	    SET_10Mb;
-	    de4x5_init_connection(dev);
-	} else {
-	    if (!lp->linkOK && (lp->autosense == AUTO)) {
-		if (!is_10_up(dev) || (!lp->useSROM && is_spd_100(dev))) {
-		    lp->media = INIT;
-		    lp->tcount++;
-		    next_tick = DE4X5_AUTOSENSE_MS;
-		}
-	    }
-	}
-	break;
-
-    case NC:
-        if (lp->media != lp->c_media) {
-	    de4x5_dbg_media(dev);
-	    lp->c_media = lp->media;
-	}
-	lp->media = INIT;
-	lp->tx_enable = false;
-	break;
-    }
-
-    return next_tick;
-}
-
-/*
-** This routine may be merged into dc21140m_autoconf() sometime as I'm
-** changing how I figure out the media - but trying to keep it backwards
-** compatible with the de500-xa and de500-aa.
-** Whether it's BNC, AUI, SYM or MII is sorted out in the infoblock
-** functions and set during de4x5_mac_port() and/or de4x5_reset_phy().
-** This routine just has to figure out whether 10Mb/s or 100Mb/s is
-** active.
-** When autonegotiation is working, the ANS part searches the SROM for
-** the highest common speed (TP) link that both can run and if that can
-** be full duplex. That infoblock is executed and then the link speed set.
-**
-** Only _10Mb and _100Mb are tested here.
-*/
-static int
-dc2114x_autoconf(struct net_device *dev)
-{
-    struct de4x5_private *lp = netdev_priv(dev);
-    u_long iobase = dev->base_addr;
-    s32 cr, anlpa, ana, cap, irqs, irq_mask, imr, omr, slnk, sr, sts;
-    int next_tick = DE4X5_AUTOSENSE_MS;
-
-    switch (lp->media) {
-    case INIT:
-        if (lp->timeout < 0) {
-	    DISABLE_IRQs;
-	    lp->tx_enable = false;
-	    lp->linkOK = 0;
-            lp->timeout = -1;
-	    de4x5_save_skbs(dev);            /* Save non transmitted skb's */
-	    if (lp->params.autosense & ~AUTO) {
-		srom_map_media(dev);         /* Fixed media requested      */
-		if (lp->media != lp->params.autosense) {
-		    lp->tcount++;
-		    lp->media = INIT;
-		    return next_tick;
-		}
-		lp->media = INIT;
-	    }
-	}
-	if ((next_tick = de4x5_reset_phy(dev)) < 0) {
-	    next_tick &= ~TIMER_CB;
-	} else {
-	    if (lp->autosense == _100Mb) {
-		lp->media = _100Mb;
-	    } else if (lp->autosense == _10Mb) {
-		lp->media = _10Mb;
-	    } else if (lp->autosense == TP) {
-		lp->media = TP;
-	    } else if (lp->autosense == BNC) {
-		lp->media = BNC;
-	    } else if (lp->autosense == AUI) {
-		lp->media = AUI;
-	    } else {
-		lp->media = SPD_DET;
-		if ((lp->infoblock_media == ANS) &&
-		                    ((sr=is_anc_capable(dev)) & MII_SR_ANC)) {
-		    ana = (((sr >> 6) & MII_ANA_TAF) | MII_ANA_CSMA);
-		    ana &= (lp->fdx ? ~0 : ~MII_ANA_FDAM);
-		    mii_wr(ana, MII_ANA, lp->phy[lp->active].addr, DE4X5_MII);
-		    lp->media = ANS;
-		}
-	    }
-	    lp->local_state = 0;
-	    next_tick = dc2114x_autoconf(dev);
-        }
-	break;
-
-    case ANS:
-	switch (lp->local_state) {
-	case 0:
-	    if (lp->timeout < 0) {
-		mii_wr(MII_CR_ASSE | MII_CR_RAN, MII_CR, lp->phy[lp->active].addr, DE4X5_MII);
-	    }
-	    cr = test_mii_reg(dev, MII_CR, MII_CR_RAN, false, 500);
-	    if (cr < 0) {
-		next_tick = cr & ~TIMER_CB;
-	    } else {
-		if (cr) {
-		    lp->local_state = 0;
-		    lp->media = SPD_DET;
-		} else {
-		    lp->local_state++;
-		}
-		next_tick = dc2114x_autoconf(dev);
-	    }
-	    break;
-
-	case 1:
-	    sr = test_mii_reg(dev, MII_SR, MII_SR_ASSC, true, 2000);
-	    if (sr < 0) {
-		next_tick = sr & ~TIMER_CB;
-	    } else {
-		lp->media = SPD_DET;
-		lp->local_state = 0;
-		if (sr) {                         /* Success! */
-		    lp->tmp = MII_SR_ASSC;
-		    anlpa = mii_rd(MII_ANLPA, lp->phy[lp->active].addr, DE4X5_MII);
-		    ana = mii_rd(MII_ANA, lp->phy[lp->active].addr, DE4X5_MII);
-		    if (!(anlpa & MII_ANLPA_RF) &&
-			 (cap = anlpa & MII_ANLPA_TAF & ana)) {
-			if (cap & MII_ANA_100M) {
-			    lp->fdx = (ana & anlpa & MII_ANA_FDAM & MII_ANA_100M) != 0;
-			    lp->media = _100Mb;
-			} else if (cap & MII_ANA_10M) {
-			    lp->fdx = (ana & anlpa & MII_ANA_FDAM & MII_ANA_10M) != 0;
-			    lp->media = _10Mb;
-			}
-		    }
-		}                       /* Auto Negotiation failed to finish */
-		next_tick = dc2114x_autoconf(dev);
-	    }                           /* Auto Negotiation failed to start  */
-	    break;
-	}
-	break;
-
-    case AUI:
-	if (!lp->tx_enable) {
-	    if (lp->timeout < 0) {
-		omr = inl(DE4X5_OMR);   /* Set up half duplex for AUI        */
-		outl(omr & ~OMR_FDX, DE4X5_OMR);
-	    }
-	    irqs = 0;
-	    irq_mask = 0;
-	    sts = test_media(dev,irqs, irq_mask, 0, 0, 0, 1000);
-	    if (sts < 0) {
-		next_tick = sts & ~TIMER_CB;
-	    } else {
-		if (!(inl(DE4X5_SISR) & SISR_SRA) && (lp->autosense == AUTO)) {
-		    lp->media = BNC;
-		    next_tick = dc2114x_autoconf(dev);
-		} else {
-		    lp->local_state = 1;
-		    de4x5_init_connection(dev);
-		}
-	    }
-	} else if (!lp->linkOK && (lp->autosense == AUTO)) {
-	    lp->media = AUI_SUSPECT;
-	    next_tick = 3000;
-	}
-	break;
-
-    case AUI_SUSPECT:
-	next_tick = de4x5_suspect_state(dev, 1000, AUI, ping_media, dc2114x_autoconf);
-	break;
-
-    case BNC:
-	switch (lp->local_state) {
-	case 0:
-	    if (lp->timeout < 0) {
-		omr = inl(DE4X5_OMR);          /* Set up half duplex for BNC */
-		outl(omr & ~OMR_FDX, DE4X5_OMR);
-	    }
-	    irqs = 0;
-	    irq_mask = 0;
-	    sts = test_media(dev,irqs, irq_mask, 0, 0, 0, 1000);
-	    if (sts < 0) {
-		next_tick = sts & ~TIMER_CB;
-	    } else {
-		lp->local_state++;             /* Ensure media connected */
-		next_tick = dc2114x_autoconf(dev);
-	    }
-	    break;
-
-	case 1:
-	    if (!lp->tx_enable) {
-		if ((sts = ping_media(dev, 3000)) < 0) {
-		    next_tick = sts & ~TIMER_CB;
-		} else {
-		    if (sts) {
-			lp->local_state = 0;
-			lp->tcount++;
-			lp->media = INIT;
-		    } else {
-			de4x5_init_connection(dev);
-		    }
-		}
-	    } else if (!lp->linkOK && (lp->autosense == AUTO)) {
-		lp->media = BNC_SUSPECT;
-		next_tick = 3000;
-	    }
-	    break;
-	}
-	break;
-
-    case BNC_SUSPECT:
-	next_tick = de4x5_suspect_state(dev, 1000, BNC, ping_media, dc2114x_autoconf);
-	break;
-
-    case SPD_DET:                              /* Choose 10Mb/s or 100Mb/s */
-	  if (srom_map_media(dev) < 0) {
-	      lp->tcount++;
-	      lp->media = INIT;
-	      return next_tick;
-	  }
-	  if (lp->media == _100Mb) {
-	      if ((slnk = test_for_100Mb(dev, 6500)) < 0) {
-		  lp->media = SPD_DET;
-		  return slnk & ~TIMER_CB;
-	      }
-	  } else {
-	      if (wait_for_link(dev) < 0) {
-		  lp->media = SPD_DET;
-		  return PDET_LINK_WAIT;
-	      }
-	  }
-	  if (lp->media == ANS) {           /* Do MII parallel detection */
-	      if (is_spd_100(dev)) {
-		  lp->media = _100Mb;
-	      } else {
-		  lp->media = _10Mb;
-	      }
-	      next_tick = dc2114x_autoconf(dev);
-	  } else if (((lp->media == _100Mb) && is_100_up(dev)) ||
-		     (((lp->media == _10Mb) || (lp->media == TP) ||
-		       (lp->media == BNC)   || (lp->media == AUI)) &&
-		      is_10_up(dev))) {
-	      next_tick = dc2114x_autoconf(dev);
-	  } else {
-	      lp->tcount++;
-	      lp->media = INIT;
-	  }
-	  break;
-
-    case _10Mb:
-        next_tick = 3000;
-	if (!lp->tx_enable) {
-	    SET_10Mb;
-	    de4x5_init_connection(dev);
-	} else {
-	    if (!lp->linkOK && (lp->autosense == AUTO)) {
-		if (!is_10_up(dev) || (!lp->useSROM && is_spd_100(dev))) {
-		    lp->media = INIT;
-		    lp->tcount++;
-		    next_tick = DE4X5_AUTOSENSE_MS;
-		}
-	    }
-	}
-	break;
-
-    case _100Mb:
-        next_tick = 3000;
-	if (!lp->tx_enable) {
-	    SET_100Mb;
-	    de4x5_init_connection(dev);
-	} else {
-	    if (!lp->linkOK && (lp->autosense == AUTO)) {
-		if (!is_100_up(dev) || (!lp->useSROM && !is_spd_100(dev))) {
-		    lp->media = INIT;
-		    lp->tcount++;
-		    next_tick = DE4X5_AUTOSENSE_MS;
-		}
-	    }
-	}
-	break;
-
-    default:
-	lp->tcount++;
-printk("Huh?: media:%02x\n", lp->media);
-	lp->media = INIT;
-	break;
-    }
-
-    return next_tick;
-}
-
-static int
-srom_autoconf(struct net_device *dev)
-{
-    struct de4x5_private *lp = netdev_priv(dev);
-
-    return lp->infoleaf_fn(dev);
-}
-
-/*
-** This mapping keeps the original media codes and FDX flag unchanged.
-** While it isn't strictly necessary, it helps me for the moment...
-** The early return avoids a media state / SROM media space clash.
-*/
-static int
-srom_map_media(struct net_device *dev)
-{
-    struct de4x5_private *lp = netdev_priv(dev);
-
-    lp->fdx = false;
-    if (lp->infoblock_media == lp->media)
-      return 0;
-
-    switch(lp->infoblock_media) {
-      case SROM_10BASETF:
-	if (!lp->params.fdx) return -1;
-	lp->fdx = true;
-      case SROM_10BASET:
-	if (lp->params.fdx && !lp->fdx) return -1;
-	if ((lp->chipset == DC21140) || ((lp->chipset & ~0x00ff) == DC2114x)) {
-	    lp->media = _10Mb;
-	} else {
-	    lp->media = TP;
-	}
-	break;
-
-      case SROM_10BASE2:
-	lp->media = BNC;
-	break;
-
-      case SROM_10BASE5:
-	lp->media = AUI;
-	break;
-
-      case SROM_100BASETF:
-        if (!lp->params.fdx) return -1;
-	lp->fdx = true;
-      case SROM_100BASET:
-	if (lp->params.fdx && !lp->fdx) return -1;
-	lp->media = _100Mb;
-	break;
-
-      case SROM_100BASET4:
-	lp->media = _100Mb;
-	break;
-
-      case SROM_100BASEFF:
-	if (!lp->params.fdx) return -1;
-	lp->fdx = true;
-      case SROM_100BASEF:
-	if (lp->params.fdx && !lp->fdx) return -1;
-	lp->media = _100Mb;
-	break;
-
-      case ANS:
-	lp->media = ANS;
-	lp->fdx = lp->params.fdx;
-	break;
-
-      default:
-	printk("%s: Bad media code [%d] detected in SROM!\n", dev->name,
-	                                                  lp->infoblock_media);
-	return -1;
-    }
-
-    return 0;
-}
-
-static void
-de4x5_init_connection(struct net_device *dev)
-{
-    struct de4x5_private *lp = netdev_priv(dev);
-    u_long iobase = dev->base_addr;
-    u_long flags = 0;
-
-    if (lp->media != lp->c_media) {
-        de4x5_dbg_media(dev);
-	lp->c_media = lp->media;          /* Stop scrolling media messages */
-    }
-
-    spin_lock_irqsave(&lp->lock, flags);
-    de4x5_rst_desc_ring(dev);
-    de4x5_setup_intr(dev);
-    lp->tx_enable = true;
-    spin_unlock_irqrestore(&lp->lock, flags);
-    outl(POLL_DEMAND, DE4X5_TPD);
-
-    netif_wake_queue(dev);
-}
-
-/*
-** General PHY reset function. Some MII devices don't reset correctly
-** since their MII address pins can float at voltages that are dependent
-** on the signal pin use. Do a double reset to ensure a reset.
-*/
-static int
-de4x5_reset_phy(struct net_device *dev)
-{
-    struct de4x5_private *lp = netdev_priv(dev);
-    u_long iobase = dev->base_addr;
-    int next_tick = 0;
-
-    if ((lp->useSROM) || (lp->phy[lp->active].id)) {
-	if (lp->timeout < 0) {
-	    if (lp->useSROM) {
-		if (lp->phy[lp->active].rst) {
-		    srom_exec(dev, lp->phy[lp->active].rst);
-		    srom_exec(dev, lp->phy[lp->active].rst);
-		} else if (lp->rst) {          /* Type 5 infoblock reset */
-		    srom_exec(dev, lp->rst);
-		    srom_exec(dev, lp->rst);
-		}
-	    } else {
-		PHY_HARD_RESET;
-	    }
-	    if (lp->useMII) {
-	        mii_wr(MII_CR_RST, MII_CR, lp->phy[lp->active].addr, DE4X5_MII);
-            }
-        }
-	if (lp->useMII) {
-	    next_tick = test_mii_reg(dev, MII_CR, MII_CR_RST, false, 500);
-	}
-    } else if (lp->chipset == DC21140) {
-	PHY_HARD_RESET;
-    }
-
-    return next_tick;
-}
-
-static int
-test_media(struct net_device *dev, s32 irqs, s32 irq_mask, s32 csr13, s32 csr14, s32 csr15, s32 msec)
-{
-    struct de4x5_private *lp = netdev_priv(dev);
-    u_long iobase = dev->base_addr;
-    s32 sts, csr12;
-
-    if (lp->timeout < 0) {
-	lp->timeout = msec/100;
-	if (!lp->useSROM) {      /* Already done if by SROM, else dc2104[01] */
-	    reset_init_sia(dev, csr13, csr14, csr15);
-	}
-
-	/* set up the interrupt mask */
-	outl(irq_mask, DE4X5_IMR);
-
-	/* clear all pending interrupts */
-	sts = inl(DE4X5_STS);
-	outl(sts, DE4X5_STS);
-
-	/* clear csr12 NRA and SRA bits */
-	if ((lp->chipset == DC21041) || lp->useSROM) {
-	    csr12 = inl(DE4X5_SISR);
-	    outl(csr12, DE4X5_SISR);
-	}
-    }
-
-    sts = inl(DE4X5_STS) & ~TIMER_CB;
-
-    if (!(sts & irqs) && --lp->timeout) {
-	sts = 100 | TIMER_CB;
-    } else {
-	lp->timeout = -1;
-    }
-
-    return sts;
-}
-
-static int
-test_tp(struct net_device *dev, s32 msec)
-{
-    struct de4x5_private *lp = netdev_priv(dev);
-    u_long iobase = dev->base_addr;
-    int sisr;
-
-    if (lp->timeout < 0) {
-	lp->timeout = msec/100;
-    }
-
-    sisr = (inl(DE4X5_SISR) & ~TIMER_CB) & (SISR_LKF | SISR_NCR);
-
-    if (sisr && --lp->timeout) {
-	sisr = 100 | TIMER_CB;
-    } else {
-	lp->timeout = -1;
-    }
-
-    return sisr;
-}
-
-/*
-** Samples the 100Mb Link State Signal. The sample interval is important
-** because too fast a rate can give erroneous results and confuse the
-** speed sense algorithm.
-*/
-#define SAMPLE_INTERVAL 500  /* ms */
-#define SAMPLE_DELAY    2000 /* ms */
-static int
-test_for_100Mb(struct net_device *dev, int msec)
-{
-    struct de4x5_private *lp = netdev_priv(dev);
-    int gep = 0, ret = ((lp->chipset & ~0x00ff)==DC2114x? -1 :GEP_SLNK);
-
-    if (lp->timeout < 0) {
-	if ((msec/SAMPLE_INTERVAL) <= 0) return 0;
-	if (msec > SAMPLE_DELAY) {
-	    lp->timeout = (msec - SAMPLE_DELAY)/SAMPLE_INTERVAL;
-	    gep = SAMPLE_DELAY | TIMER_CB;
-	    return gep;
-	} else {
-	    lp->timeout = msec/SAMPLE_INTERVAL;
-	}
-    }
-
-    if (lp->phy[lp->active].id || lp->useSROM) {
-	gep = is_100_up(dev) | is_spd_100(dev);
-    } else {
-	gep = (~gep_rd(dev) & (GEP_SLNK | GEP_LNP));
-    }
-    if (!(gep & ret) && --lp->timeout) {
-	gep = SAMPLE_INTERVAL | TIMER_CB;
-    } else {
-	lp->timeout = -1;
-    }
-
-    return gep;
-}
-
-static int
-wait_for_link(struct net_device *dev)
-{
-    struct de4x5_private *lp = netdev_priv(dev);
-
-    if (lp->timeout < 0) {
-	lp->timeout = 1;
-    }
-
-    if (lp->timeout--) {
-	return TIMER_CB;
-    } else {
-	lp->timeout = -1;
-    }
-
-    return 0;
-}
-
-/*
-**
-**
-*/
-static int
-test_mii_reg(struct net_device *dev, int reg, int mask, bool pol, long msec)
-{
-    struct de4x5_private *lp = netdev_priv(dev);
-    int test;
-    u_long iobase = dev->base_addr;
-
-    if (lp->timeout < 0) {
-	lp->timeout = msec/100;
-    }
-
-    reg = mii_rd((u_char)reg, lp->phy[lp->active].addr, DE4X5_MII) & mask;
-    test = (reg ^ (pol ? ~0 : 0)) & mask;
-
-    if (test && --lp->timeout) {
-	reg = 100 | TIMER_CB;
-    } else {
-	lp->timeout = -1;
-    }
-
-    return reg;
-}
-
-static int
-is_spd_100(struct net_device *dev)
-{
-    struct de4x5_private *lp = netdev_priv(dev);
-    u_long iobase = dev->base_addr;
-    int spd;
-
-    if (lp->useMII) {
-	spd = mii_rd(lp->phy[lp->active].spd.reg, lp->phy[lp->active].addr, DE4X5_MII);
-	spd = ~(spd ^ lp->phy[lp->active].spd.value);
-	spd &= lp->phy[lp->active].spd.mask;
-    } else if (!lp->useSROM) {                      /* de500-xa */
-	spd = ((~gep_rd(dev)) & GEP_SLNK);
-    } else {
-	if ((lp->ibn == 2) || !lp->asBitValid)
-	    return (lp->chipset == DC21143) ? (~inl(DE4X5_SISR)&SISR_LS100) : 0;
-
-	spd = (lp->asBitValid & (lp->asPolarity ^ (gep_rd(dev) & lp->asBit))) |
-	          (lp->linkOK & ~lp->asBitValid);
-    }
-
-    return spd;
-}
-
-static int
-is_100_up(struct net_device *dev)
-{
-    struct de4x5_private *lp = netdev_priv(dev);
-    u_long iobase = dev->base_addr;
-
-    if (lp->useMII) {
-	/* Double read for sticky bits & temporary drops */
-	mii_rd(MII_SR, lp->phy[lp->active].addr, DE4X5_MII);
-	return mii_rd(MII_SR, lp->phy[lp->active].addr, DE4X5_MII) & MII_SR_LKS;
-    } else if (!lp->useSROM) {                       /* de500-xa */
-	return (~gep_rd(dev)) & GEP_SLNK;
-    } else {
-	if ((lp->ibn == 2) || !lp->asBitValid)
-	    return (lp->chipset == DC21143) ? (~inl(DE4X5_SISR)&SISR_LS100) : 0;
-
-        return (lp->asBitValid&(lp->asPolarity^(gep_rd(dev)&lp->asBit))) |
-		(lp->linkOK & ~lp->asBitValid);
-    }
-}
-
-static int
-is_10_up(struct net_device *dev)
-{
-    struct de4x5_private *lp = netdev_priv(dev);
-    u_long iobase = dev->base_addr;
-
-    if (lp->useMII) {
-	/* Double read for sticky bits & temporary drops */
-	mii_rd(MII_SR, lp->phy[lp->active].addr, DE4X5_MII);
-	return mii_rd(MII_SR, lp->phy[lp->active].addr, DE4X5_MII) & MII_SR_LKS;
-    } else if (!lp->useSROM) {                       /* de500-xa */
-	return (~gep_rd(dev)) & GEP_LNP;
-    } else {
-	if ((lp->ibn == 2) || !lp->asBitValid)
-	    return ((lp->chipset & ~0x00ff) == DC2114x) ?
-		    (~inl(DE4X5_SISR)&SISR_LS10):
-		    0;
-
-	return	(lp->asBitValid&(lp->asPolarity^(gep_rd(dev)&lp->asBit))) |
-		(lp->linkOK & ~lp->asBitValid);
-    }
-}
-
-static int
-is_anc_capable(struct net_device *dev)
-{
-    struct de4x5_private *lp = netdev_priv(dev);
-    u_long iobase = dev->base_addr;
-
-    if (lp->phy[lp->active].id && (!lp->useSROM || lp->useMII)) {
-	return mii_rd(MII_SR, lp->phy[lp->active].addr, DE4X5_MII);
-    } else if ((lp->chipset & ~0x00ff) == DC2114x) {
-	return (inl(DE4X5_SISR) & SISR_LPN) >> 12;
-    } else {
-	return 0;
-    }
-}
-
-/*
-** Send a packet onto the media and watch for send errors that indicate the
-** media is bad or unconnected.
-*/
-static int
-ping_media(struct net_device *dev, int msec)
-{
-    struct de4x5_private *lp = netdev_priv(dev);
-    u_long iobase = dev->base_addr;
-    int sisr;
-
-    if (lp->timeout < 0) {
-	lp->timeout = msec/100;
-
-	lp->tmp = lp->tx_new;                /* Remember the ring position */
-	load_packet(dev, lp->frame, TD_LS | TD_FS | sizeof(lp->frame), (struct sk_buff *)1);
-	lp->tx_new = (lp->tx_new + 1) % lp->txRingSize;
-	outl(POLL_DEMAND, DE4X5_TPD);
-    }
-
-    sisr = inl(DE4X5_SISR);
-
-    if ((!(sisr & SISR_NCR)) &&
-	((s32)le32_to_cpu(lp->tx_ring[lp->tmp].status) < 0) &&
-	 (--lp->timeout)) {
-	sisr = 100 | TIMER_CB;
-    } else {
-	if ((!(sisr & SISR_NCR)) &&
-	    !(le32_to_cpu(lp->tx_ring[lp->tmp].status) & (T_OWN | TD_ES)) &&
-	    lp->timeout) {
-	    sisr = 0;
-	} else {
-	    sisr = 1;
-	}
-	lp->timeout = -1;
-    }
-
-    return sisr;
-}
-
-/*
-** This function does 2 things: on Intels it kmalloc's another buffer to
-** replace the one about to be passed up. On Alpha's it kmallocs a buffer
-** into which the packet is copied.
-*/
-static struct sk_buff *
-de4x5_alloc_rx_buff(struct net_device *dev, int index, int len)
-{
-    struct de4x5_private *lp = netdev_priv(dev);
-    struct sk_buff *p;
-
-#if !defined(__alpha__) && !defined(__powerpc__) && !defined(CONFIG_SPARC) && !defined(DE4X5_DO_MEMCPY)
-    struct sk_buff *ret;
-    u_long i=0, tmp;
-
-    p = netdev_alloc_skb(dev, IEEE802_3_SZ + DE4X5_ALIGN + 2);
-    if (!p) return NULL;
-
-    tmp = virt_to_bus(p->data);
-    i = ((tmp + DE4X5_ALIGN) & ~DE4X5_ALIGN) - tmp;
-    skb_reserve(p, i);
-    lp->rx_ring[index].buf = cpu_to_le32(tmp + i);
-
-    ret = lp->rx_skb[index];
-    lp->rx_skb[index] = p;
-
-    if ((u_long) ret > 1) {
-	skb_put(ret, len);
-    }
-
-    return ret;
-
-#else
-    if (lp->state != OPEN) return (struct sk_buff *)1; /* Fake out the open */
-
-    p = netdev_alloc_skb(dev, len + 2);
-    if (!p) return NULL;
-
-    skb_reserve(p, 2);	                               /* Align */
-    if (index < lp->rx_old) {                          /* Wrapped buffer */
-	short tlen = (lp->rxRingSize - lp->rx_old) * RX_BUFF_SZ;
-	memcpy(skb_put(p,tlen),lp->rx_bufs + lp->rx_old * RX_BUFF_SZ,tlen);
-	memcpy(skb_put(p,len-tlen),lp->rx_bufs,len-tlen);
-    } else {                                           /* Linear buffer */
-	memcpy(skb_put(p,len),lp->rx_bufs + lp->rx_old * RX_BUFF_SZ,len);
-    }
-
-    return p;
-#endif
-}
-
-static void
-de4x5_free_rx_buffs(struct net_device *dev)
-{
-    struct de4x5_private *lp = netdev_priv(dev);
-    int i;
-
-    for (i=0; i<lp->rxRingSize; i++) {
-	if ((u_long) lp->rx_skb[i] > 1) {
-	    dev_kfree_skb(lp->rx_skb[i]);
-	}
-	lp->rx_ring[i].status = 0;
-	lp->rx_skb[i] = (struct sk_buff *)1;    /* Dummy entry */
-    }
-}
-
-static void
-de4x5_free_tx_buffs(struct net_device *dev)
-{
-    struct de4x5_private *lp = netdev_priv(dev);
-    int i;
-
-    for (i=0; i<lp->txRingSize; i++) {
-	if (lp->tx_skb[i])
-	    de4x5_free_tx_buff(lp, i);
-	lp->tx_ring[i].status = 0;
-    }
-
-    /* Unload the locally queued packets */
-    __skb_queue_purge(&lp->cache.queue);
-}
-
-/*
-** When a user pulls a connection, the DECchip can end up in a
-** 'running - waiting for end of transmission' state. This means that we
-** have to perform a chip soft reset to ensure that we can synchronize
-** the hardware and software and make any media probes using a loopback
-** packet meaningful.
-*/
-static void
-de4x5_save_skbs(struct net_device *dev)
-{
-    struct de4x5_private *lp = netdev_priv(dev);
-    u_long iobase = dev->base_addr;
-    s32 omr;
-
-    if (!lp->cache.save_cnt) {
-	STOP_DE4X5;
-	de4x5_tx(dev);                          /* Flush any sent skb's */
-	de4x5_free_tx_buffs(dev);
-	de4x5_cache_state(dev, DE4X5_SAVE_STATE);
-	de4x5_sw_reset(dev);
-	de4x5_cache_state(dev, DE4X5_RESTORE_STATE);
-	lp->cache.save_cnt++;
-	START_DE4X5;
-    }
-}
-
-static void
-de4x5_rst_desc_ring(struct net_device *dev)
-{
-    struct de4x5_private *lp = netdev_priv(dev);
-    u_long iobase = dev->base_addr;
-    int i;
-    s32 omr;
-
-    if (lp->cache.save_cnt) {
-	STOP_DE4X5;
-	outl(lp->dma_rings, DE4X5_RRBA);
-	outl(lp->dma_rings + NUM_RX_DESC * sizeof(struct de4x5_desc),
-	     DE4X5_TRBA);
-
-	lp->rx_new = lp->rx_old = 0;
-	lp->tx_new = lp->tx_old = 0;
-
-	for (i = 0; i < lp->rxRingSize; i++) {
-	    lp->rx_ring[i].status = cpu_to_le32(R_OWN);
-	}
-
-	for (i = 0; i < lp->txRingSize; i++) {
-	    lp->tx_ring[i].status = cpu_to_le32(0);
-	}
-
-	barrier();
-	lp->cache.save_cnt--;
-	START_DE4X5;
-    }
-}
-
-static void
-de4x5_cache_state(struct net_device *dev, int flag)
-{
-    struct de4x5_private *lp = netdev_priv(dev);
-    u_long iobase = dev->base_addr;
-
-    switch(flag) {
-      case DE4X5_SAVE_STATE:
-	lp->cache.csr0 = inl(DE4X5_BMR);
-	lp->cache.csr6 = (inl(DE4X5_OMR) & ~(OMR_ST | OMR_SR));
-	lp->cache.csr7 = inl(DE4X5_IMR);
-	break;
-
-      case DE4X5_RESTORE_STATE:
-	outl(lp->cache.csr0, DE4X5_BMR);
-	outl(lp->cache.csr6, DE4X5_OMR);
-	outl(lp->cache.csr7, DE4X5_IMR);
-	if (lp->chipset == DC21140) {
-	    gep_wr(lp->cache.gepc, dev);
-	    gep_wr(lp->cache.gep, dev);
-	} else {
-	    reset_init_sia(dev, lp->cache.csr13, lp->cache.csr14,
-			                                      lp->cache.csr15);
-	}
-	break;
-    }
-}
-
-static void
-de4x5_put_cache(struct net_device *dev, struct sk_buff *skb)
-{
-    struct de4x5_private *lp = netdev_priv(dev);
-
-    __skb_queue_tail(&lp->cache.queue, skb);
-}
-
-static void
-de4x5_putb_cache(struct net_device *dev, struct sk_buff *skb)
-{
-    struct de4x5_private *lp = netdev_priv(dev);
-
-    __skb_queue_head(&lp->cache.queue, skb);
-}
-
-static struct sk_buff *
-de4x5_get_cache(struct net_device *dev)
-{
-    struct de4x5_private *lp = netdev_priv(dev);
-
-    return __skb_dequeue(&lp->cache.queue);
-}
-
-/*
-** Check the Auto Negotiation State. Return OK when a link pass interrupt
-** is received and the auto-negotiation status is NWAY OK.
-*/
-static int
-test_ans(struct net_device *dev, s32 irqs, s32 irq_mask, s32 msec)
-{
-    struct de4x5_private *lp = netdev_priv(dev);
-    u_long iobase = dev->base_addr;
-    s32 sts, ans;
-
-    if (lp->timeout < 0) {
-	lp->timeout = msec/100;
-	outl(irq_mask, DE4X5_IMR);
-
-	/* clear all pending interrupts */
-	sts = inl(DE4X5_STS);
-	outl(sts, DE4X5_STS);
-    }
-
-    ans = inl(DE4X5_SISR) & SISR_ANS;
-    sts = inl(DE4X5_STS) & ~TIMER_CB;
-
-    if (!(sts & irqs) && (ans ^ ANS_NWOK) && --lp->timeout) {
-	sts = 100 | TIMER_CB;
-    } else {
-	lp->timeout = -1;
-    }
-
-    return sts;
-}
-
-static void
-de4x5_setup_intr(struct net_device *dev)
-{
-    struct de4x5_private *lp = netdev_priv(dev);
-    u_long iobase = dev->base_addr;
-    s32 imr, sts;
-
-    if (inl(DE4X5_OMR) & OMR_SR) {   /* Only unmask if TX/RX is enabled */
-	imr = 0;
-	UNMASK_IRQs;
-	sts = inl(DE4X5_STS);        /* Reset any pending (stale) interrupts */
-	outl(sts, DE4X5_STS);
-	ENABLE_IRQs;
-    }
-}
-
-/*
-**
-*/
-static void
-reset_init_sia(struct net_device *dev, s32 csr13, s32 csr14, s32 csr15)
-{
-    struct de4x5_private *lp = netdev_priv(dev);
-    u_long iobase = dev->base_addr;
-
-    RESET_SIA;
-    if (lp->useSROM) {
-	if (lp->ibn == 3) {
-	    srom_exec(dev, lp->phy[lp->active].rst);
-	    srom_exec(dev, lp->phy[lp->active].gep);
-	    outl(1, DE4X5_SICR);
-	    return;
-	} else {
-	    csr15 = lp->cache.csr15;
-	    csr14 = lp->cache.csr14;
-	    csr13 = lp->cache.csr13;
-	    outl(csr15 | lp->cache.gepc, DE4X5_SIGR);
-	    outl(csr15 | lp->cache.gep, DE4X5_SIGR);
-	}
-    } else {
-	outl(csr15, DE4X5_SIGR);
-    }
-    outl(csr14, DE4X5_STRR);
-    outl(csr13, DE4X5_SICR);
-
-    mdelay(10);
-}
-
-/*
-** Create a loopback ethernet packet
-*/
-static void
-create_packet(struct net_device *dev, char *frame, int len)
-{
-    int i;
-    char *buf = frame;
-
-    for (i=0; i<ETH_ALEN; i++) {             /* Use this source address */
-	*buf++ = dev->dev_addr[i];
-    }
-    for (i=0; i<ETH_ALEN; i++) {             /* Use this destination address */
-	*buf++ = dev->dev_addr[i];
-    }
-
-    *buf++ = 0;                              /* Packet length (2 bytes) */
-    *buf++ = 1;
-}
-
-/*
-** Look for a particular board name in the EISA configuration space
-*/
-static int
-EISA_signature(char *name, struct device *device)
-{
-    int i, status = 0, siglen = ARRAY_SIZE(de4x5_signatures);
-    struct eisa_device *edev;
-
-    *name = '\0';
-    edev = to_eisa_device (device);
-    i = edev->id.driver_data;
-
-    if (i >= 0 && i < siglen) {
-	    strcpy (name, de4x5_signatures[i]);
-	    status = 1;
-    }
-
-    return status;                         /* return the device name string */
-}
-
-/*
-** Look for a particular board name in the PCI configuration space
-*/
-static int
-PCI_signature(char *name, struct de4x5_private *lp)
-{
-    int i, status = 0, siglen = ARRAY_SIZE(de4x5_signatures);
-
-    if (lp->chipset == DC21040) {
-	strcpy(name, "DE434/5");
-	return status;
-    } else {                           /* Search for a DEC name in the SROM */
-	int tmp = *((char *)&lp->srom + 19) * 3;
-	strncpy(name, (char *)&lp->srom + 26 + tmp, 8);
-    }
-    name[8] = '\0';
-    for (i=0; i<siglen; i++) {
-	if (strstr(name,de4x5_signatures[i])!=NULL) break;
-    }
-    if (i == siglen) {
-	if (dec_only) {
-	    *name = '\0';
-	} else {                        /* Use chip name to avoid confusion */
-	    strcpy(name, (((lp->chipset == DC21040) ? "DC21040" :
-			   ((lp->chipset == DC21041) ? "DC21041" :
-			    ((lp->chipset == DC21140) ? "DC21140" :
-			     ((lp->chipset == DC21142) ? "DC21142" :
-			      ((lp->chipset == DC21143) ? "DC21143" : "UNKNOWN"
-			     )))))));
-	}
-	if (lp->chipset != DC21041) {
-	    lp->useSROM = true;             /* card is not recognisably DEC */
-	}
-    } else if ((lp->chipset & ~0x00ff) == DC2114x) {
-	lp->useSROM = true;
-    }
-
-    return status;
-}
-
-/*
-** Set up the Ethernet PROM counter to the start of the Ethernet address on
-** the DC21040, else  read the SROM for the other chips.
-** The SROM may not be present in a multi-MAC card, so first read the
-** MAC address and check for a bad address. If there is a bad one then exit
-** immediately with the prior srom contents intact (the h/w address will
-** be fixed up later).
-*/
-static void
-DevicePresent(struct net_device *dev, u_long aprom_addr)
-{
-    int i, j=0;
-    struct de4x5_private *lp = netdev_priv(dev);
-
-    if (lp->chipset == DC21040) {
-	if (lp->bus == EISA) {
-	    enet_addr_rst(aprom_addr); /* Reset Ethernet Address ROM Pointer */
-	} else {
-	    outl(0, aprom_addr);       /* Reset Ethernet Address ROM Pointer */
-	}
-    } else {                           /* Read new srom */
-	u_short tmp;
-	__le16 *p = (__le16 *)((char *)&lp->srom + SROM_HWADD);
-	for (i=0; i<(ETH_ALEN>>1); i++) {
-	    tmp = srom_rd(aprom_addr, (SROM_HWADD>>1) + i);
-	    j += tmp;	/* for check for 0:0:0:0:0:0 or ff:ff:ff:ff:ff:ff */
-	    *p = cpu_to_le16(tmp);
-	}
-	if (j == 0 || j == 3 * 0xffff) {
-		/* could get 0 only from all-0 and 3 * 0xffff only from all-1 */
-		return;
-	}
-
-	p = (__le16 *)&lp->srom;
-	for (i=0; i<(sizeof(struct de4x5_srom)>>1); i++) {
-	    tmp = srom_rd(aprom_addr, i);
-	    *p++ = cpu_to_le16(tmp);
-	}
-	de4x5_dbg_srom(&lp->srom);
-    }
-}
-
-/*
-** Since the write on the Enet PROM register doesn't seem to reset the PROM
-** pointer correctly (at least on my DE425 EISA card), this routine should do
-** it...from depca.c.
-*/
-static void
-enet_addr_rst(u_long aprom_addr)
-{
-    union {
-	struct {
-	    u32 a;
-	    u32 b;
-	} llsig;
-	char Sig[sizeof(u32) << 1];
-    } dev;
-    short sigLength=0;
-    s8 data;
-    int i, j;
-
-    dev.llsig.a = ETH_PROM_SIG;
-    dev.llsig.b = ETH_PROM_SIG;
-    sigLength = sizeof(u32) << 1;
-
-    for (i=0,j=0;j<sigLength && i<PROBE_LENGTH+sigLength-1;i++) {
-	data = inb(aprom_addr);
-	if (dev.Sig[j] == data) {    /* track signature */
-	    j++;
-	} else {                     /* lost signature; begin search again */
-	    if (data == dev.Sig[0]) {  /* rare case.... */
-		j=1;
-	    } else {
-		j=0;
-	    }
-	}
-    }
-}
-
-/*
-** For the bad status case and no SROM, then add one to the previous
-** address. However, need to add one backwards in case we have 0xff
-** as one or more of the bytes. Only the last 3 bytes should be checked
-** as the first three are invariant - assigned to an organisation.
-*/
-static int
-get_hw_addr(struct net_device *dev)
-{
-    u_long iobase = dev->base_addr;
-    int broken, i, k, tmp, status = 0;
-    u_short j,chksum;
-    struct de4x5_private *lp = netdev_priv(dev);
-
-    broken = de4x5_bad_srom(lp);
-
-    for (i=0,k=0,j=0;j<3;j++) {
-	k <<= 1;
-	if (k > 0xffff) k-=0xffff;
-
-	if (lp->bus == PCI) {
-	    if (lp->chipset == DC21040) {
-		while ((tmp = inl(DE4X5_APROM)) < 0);
-		k += (u_char) tmp;
-		dev->dev_addr[i++] = (u_char) tmp;
-		while ((tmp = inl(DE4X5_APROM)) < 0);
-		k += (u_short) (tmp << 8);
-		dev->dev_addr[i++] = (u_char) tmp;
-	    } else if (!broken) {
-		dev->dev_addr[i] = (u_char) lp->srom.ieee_addr[i]; i++;
-		dev->dev_addr[i] = (u_char) lp->srom.ieee_addr[i]; i++;
-	    } else if ((broken == SMC) || (broken == ACCTON)) {
-		dev->dev_addr[i] = *((u_char *)&lp->srom + i); i++;
-		dev->dev_addr[i] = *((u_char *)&lp->srom + i); i++;
-	    }
-	} else {
-	    k += (u_char) (tmp = inb(EISA_APROM));
-	    dev->dev_addr[i++] = (u_char) tmp;
-	    k += (u_short) ((tmp = inb(EISA_APROM)) << 8);
-	    dev->dev_addr[i++] = (u_char) tmp;
-	}
-
-	if (k > 0xffff) k-=0xffff;
-    }
-    if (k == 0xffff) k=0;
-
-    if (lp->bus == PCI) {
-	if (lp->chipset == DC21040) {
-	    while ((tmp = inl(DE4X5_APROM)) < 0);
-	    chksum = (u_char) tmp;
-	    while ((tmp = inl(DE4X5_APROM)) < 0);
-	    chksum |= (u_short) (tmp << 8);
-	    if ((k != chksum) && (dec_only)) status = -1;
-	}
-    } else {
-	chksum = (u_char) inb(EISA_APROM);
-	chksum |= (u_short) (inb(EISA_APROM) << 8);
-	if ((k != chksum) && (dec_only)) status = -1;
-    }
-
-    /* If possible, try to fix a broken card - SMC only so far */
-    srom_repair(dev, broken);
-
-#ifdef CONFIG_PPC_PMAC
-    /*
-    ** If the address starts with 00 a0, we have to bit-reverse
-    ** each byte of the address.
-    */
-    if ( machine_is(powermac) &&
-	 (dev->dev_addr[0] == 0) &&
-	 (dev->dev_addr[1] == 0xa0) )
-    {
-	    for (i = 0; i < ETH_ALEN; ++i)
-	    {
-		    int x = dev->dev_addr[i];
-		    x = ((x & 0xf) << 4) + ((x & 0xf0) >> 4);
-		    x = ((x & 0x33) << 2) + ((x & 0xcc) >> 2);
-		    dev->dev_addr[i] = ((x & 0x55) << 1) + ((x & 0xaa) >> 1);
-	    }
-    }
-#endif /* CONFIG_PPC_PMAC */
-
-    /* Test for a bad enet address */
-    status = test_bad_enet(dev, status);
-
-    return status;
-}
-
-/*
-** Test for enet addresses in the first 32 bytes.
-*/
-static int
-de4x5_bad_srom(struct de4x5_private *lp)
-{
-    int i, status = 0;
-
-    for (i = 0; i < ARRAY_SIZE(enet_det); i++) {
-	if (!memcmp(&lp->srom, &enet_det[i], 3) &&
-	    !memcmp((char *)&lp->srom+0x10, &enet_det[i], 3)) {
-	    if (i == 0) {
-		status = SMC;
-	    } else if (i == 1) {
-		status = ACCTON;
-	    }
-	    break;
-	}
-    }
-
-    return status;
-}
-
-static void
-srom_repair(struct net_device *dev, int card)
-{
-    struct de4x5_private *lp = netdev_priv(dev);
-
-    switch(card) {
-      case SMC:
-	memset((char *)&lp->srom, 0, sizeof(struct de4x5_srom));
-	memcpy(lp->srom.ieee_addr, (char *)dev->dev_addr, ETH_ALEN);
-	memcpy(lp->srom.info, (char *)&srom_repair_info[SMC-1], 100);
-	lp->useSROM = true;
-	break;
-    }
-}
-
-/*
-** Assume that the irq's do not follow the PCI spec - this is seems
-** to be true so far (2 for 2).
-*/
-static int
-test_bad_enet(struct net_device *dev, int status)
-{
-    struct de4x5_private *lp = netdev_priv(dev);
-    int i, tmp;
-
-    for (tmp=0,i=0; i<ETH_ALEN; i++) tmp += (u_char)dev->dev_addr[i];
-    if ((tmp == 0) || (tmp == 0x5fa)) {
-	if ((lp->chipset == last.chipset) &&
-	    (lp->bus_num == last.bus) && (lp->bus_num > 0)) {
-	    for (i=0; i<ETH_ALEN; i++) dev->dev_addr[i] = last.addr[i];
-	    for (i=ETH_ALEN-1; i>2; --i) {
-		dev->dev_addr[i] += 1;
-		if (dev->dev_addr[i] != 0) break;
-	    }
-	    for (i=0; i<ETH_ALEN; i++) last.addr[i] = dev->dev_addr[i];
-	    if (!an_exception(lp)) {
-		dev->irq = last.irq;
-	    }
-
-	    status = 0;
-	}
-    } else if (!status) {
-	last.chipset = lp->chipset;
-	last.bus = lp->bus_num;
-	last.irq = dev->irq;
-	for (i=0; i<ETH_ALEN; i++) last.addr[i] = dev->dev_addr[i];
-    }
-
-    return status;
-}
-
-/*
-** List of board exceptions with correctly wired IRQs
-*/
-static int
-an_exception(struct de4x5_private *lp)
-{
-    if ((*(u_short *)lp->srom.sub_vendor_id == 0x00c0) &&
-	(*(u_short *)lp->srom.sub_system_id == 0x95e0)) {
-	return -1;
-    }
-
-    return 0;
-}
-
-/*
-** SROM Read
-*/
-static short
-srom_rd(u_long addr, u_char offset)
-{
-    sendto_srom(SROM_RD | SROM_SR, addr);
-
-    srom_latch(SROM_RD | SROM_SR | DT_CS, addr);
-    srom_command(SROM_RD | SROM_SR | DT_IN | DT_CS, addr);
-    srom_address(SROM_RD | SROM_SR | DT_CS, addr, offset);
-
-    return srom_data(SROM_RD | SROM_SR | DT_CS, addr);
-}
-
-static void
-srom_latch(u_int command, u_long addr)
-{
-    sendto_srom(command, addr);
-    sendto_srom(command | DT_CLK, addr);
-    sendto_srom(command, addr);
-}
-
-static void
-srom_command(u_int command, u_long addr)
-{
-    srom_latch(command, addr);
-    srom_latch(command, addr);
-    srom_latch((command & 0x0000ff00) | DT_CS, addr);
-}
-
-static void
-srom_address(u_int command, u_long addr, u_char offset)
-{
-    int i, a;
-
-    a = offset << 2;
-    for (i=0; i<6; i++, a <<= 1) {
-	srom_latch(command | ((a & 0x80) ? DT_IN : 0), addr);
-    }
-    udelay(1);
-
-    i = (getfrom_srom(addr) >> 3) & 0x01;
-}
-
-static short
-srom_data(u_int command, u_long addr)
-{
-    int i;
-    short word = 0;
-    s32 tmp;
-
-    for (i=0; i<16; i++) {
-	sendto_srom(command  | DT_CLK, addr);
-	tmp = getfrom_srom(addr);
-	sendto_srom(command, addr);
-
-	word = (word << 1) | ((tmp >> 3) & 0x01);
-    }
-
-    sendto_srom(command & 0x0000ff00, addr);
-
-    return word;
-}
-
-/*
-static void
-srom_busy(u_int command, u_long addr)
-{
-   sendto_srom((command & 0x0000ff00) | DT_CS, addr);
-
-   while (!((getfrom_srom(addr) >> 3) & 0x01)) {
-       mdelay(1);
-   }
-
-   sendto_srom(command & 0x0000ff00, addr);
-}
-*/
-
-static void
-sendto_srom(u_int command, u_long addr)
-{
-    outl(command, addr);
-    udelay(1);
-}
-
-static int
-getfrom_srom(u_long addr)
-{
-    s32 tmp;
-
-    tmp = inl(addr);
-    udelay(1);
-
-    return tmp;
-}
-
-static int
-srom_infoleaf_info(struct net_device *dev)
-{
-    struct de4x5_private *lp = netdev_priv(dev);
-    int i, count;
-    u_char *p;
-
-    /* Find the infoleaf decoder function that matches this chipset */
-    for (i=0; i<INFOLEAF_SIZE; i++) {
-	if (lp->chipset == infoleaf_array[i].chipset) break;
-    }
-    if (i == INFOLEAF_SIZE) {
-	lp->useSROM = false;
-	printk("%s: Cannot find correct chipset for SROM decoding!\n",
-	                                                          dev->name);
-	return -ENXIO;
-    }
-
-    lp->infoleaf_fn = infoleaf_array[i].fn;
-
-    /* Find the information offset that this function should use */
-    count = *((u_char *)&lp->srom + 19);
-    p  = (u_char *)&lp->srom + 26;
-
-    if (count > 1) {
-	for (i=count; i; --i, p+=3) {
-	    if (lp->device == *p) break;
-	}
-	if (i == 0) {
-	    lp->useSROM = false;
-	    printk("%s: Cannot find correct PCI device [%d] for SROM decoding!\n",
-	                                               dev->name, lp->device);
-	    return -ENXIO;
-	}
-    }
-
-	lp->infoleaf_offset = get_unaligned_le16(p + 1);
-
-    return 0;
-}
-
-/*
-** This routine loads any type 1 or 3 MII info into the mii device
-** struct and executes any type 5 code to reset PHY devices for this
-** controller.
-** The info for the MII devices will be valid since the index used
-** will follow the discovery process from MII address 1-31 then 0.
-*/
-static void
-srom_init(struct net_device *dev)
-{
-    struct de4x5_private *lp = netdev_priv(dev);
-    u_char *p = (u_char *)&lp->srom + lp->infoleaf_offset;
-    u_char count;
-
-    p+=2;
-    if (lp->chipset == DC21140) {
-	lp->cache.gepc = (*p++ | GEP_CTRL);
-	gep_wr(lp->cache.gepc, dev);
-    }
-
-    /* Block count */
-    count = *p++;
-
-    /* Jump the infoblocks to find types */
-    for (;count; --count) {
-	if (*p < 128) {
-	    p += COMPACT_LEN;
-	} else if (*(p+1) == 5) {
-	    type5_infoblock(dev, 1, p);
-	    p += ((*p & BLOCK_LEN) + 1);
-	} else if (*(p+1) == 4) {
-	    p += ((*p & BLOCK_LEN) + 1);
-	} else if (*(p+1) == 3) {
-	    type3_infoblock(dev, 1, p);
-	    p += ((*p & BLOCK_LEN) + 1);
-	} else if (*(p+1) == 2) {
-	    p += ((*p & BLOCK_LEN) + 1);
-	} else if (*(p+1) == 1) {
-	    type1_infoblock(dev, 1, p);
-	    p += ((*p & BLOCK_LEN) + 1);
-	} else {
-	    p += ((*p & BLOCK_LEN) + 1);
-	}
-    }
-}
-
-/*
-** A generic routine that writes GEP control, data and reset information
-** to the GEP register (21140) or csr15 GEP portion (2114[23]).
-*/
-static void
-srom_exec(struct net_device *dev, u_char *p)
-{
-    struct de4x5_private *lp = netdev_priv(dev);
-    u_long iobase = dev->base_addr;
-    u_char count = (p ? *p++ : 0);
-    u_short *w = (u_short *)p;
-
-    if (((lp->ibn != 1) && (lp->ibn != 3) && (lp->ibn != 5)) || !count) return;
-
-    if (lp->chipset != DC21140) RESET_SIA;
-
-    while (count--) {
-	gep_wr(((lp->chipset==DC21140) && (lp->ibn!=5) ?
-		                                   *p++ : get_unaligned_le16(w++)), dev);
-	mdelay(2);                          /* 2ms per action */
-    }
-
-    if (lp->chipset != DC21140) {
-	outl(lp->cache.csr14, DE4X5_STRR);
-	outl(lp->cache.csr13, DE4X5_SICR);
-    }
-}
-
-/*
-** Basically this function is a NOP since it will never be called,
-** unless I implement the DC21041 SROM functions. There's no need
-** since the existing code will be satisfactory for all boards.
-*/
-static int
-dc21041_infoleaf(struct net_device *dev)
-{
-    return DE4X5_AUTOSENSE_MS;
-}
-
-static int
-dc21140_infoleaf(struct net_device *dev)
-{
-    struct de4x5_private *lp = netdev_priv(dev);
-    u_char count = 0;
-    u_char *p = (u_char *)&lp->srom + lp->infoleaf_offset;
-    int next_tick = DE4X5_AUTOSENSE_MS;
-
-    /* Read the connection type */
-    p+=2;
-
-    /* GEP control */
-    lp->cache.gepc = (*p++ | GEP_CTRL);
-
-    /* Block count */
-    count = *p++;
-
-    /* Recursively figure out the info blocks */
-    if (*p < 128) {
-	next_tick = dc_infoblock[COMPACT](dev, count, p);
-    } else {
-	next_tick = dc_infoblock[*(p+1)](dev, count, p);
-    }
-
-    if (lp->tcount == count) {
-	lp->media = NC;
-        if (lp->media != lp->c_media) {
-	    de4x5_dbg_media(dev);
-	    lp->c_media = lp->media;
-	}
-	lp->media = INIT;
-	lp->tcount = 0;
-	lp->tx_enable = false;
-    }
-
-    return next_tick & ~TIMER_CB;
-}
-
-static int
-dc21142_infoleaf(struct net_device *dev)
-{
-    struct de4x5_private *lp = netdev_priv(dev);
-    u_char count = 0;
-    u_char *p = (u_char *)&lp->srom + lp->infoleaf_offset;
-    int next_tick = DE4X5_AUTOSENSE_MS;
-
-    /* Read the connection type */
-    p+=2;
-
-    /* Block count */
-    count = *p++;
-
-    /* Recursively figure out the info blocks */
-    if (*p < 128) {
-	next_tick = dc_infoblock[COMPACT](dev, count, p);
-    } else {
-	next_tick = dc_infoblock[*(p+1)](dev, count, p);
-    }
-
-    if (lp->tcount == count) {
-	lp->media = NC;
-        if (lp->media != lp->c_media) {
-	    de4x5_dbg_media(dev);
-	    lp->c_media = lp->media;
-	}
-	lp->media = INIT;
-	lp->tcount = 0;
-	lp->tx_enable = false;
-    }
-
-    return next_tick & ~TIMER_CB;
-}
-
-static int
-dc21143_infoleaf(struct net_device *dev)
-{
-    struct de4x5_private *lp = netdev_priv(dev);
-    u_char count = 0;
-    u_char *p = (u_char *)&lp->srom + lp->infoleaf_offset;
-    int next_tick = DE4X5_AUTOSENSE_MS;
-
-    /* Read the connection type */
-    p+=2;
-
-    /* Block count */
-    count = *p++;
-
-    /* Recursively figure out the info blocks */
-    if (*p < 128) {
-	next_tick = dc_infoblock[COMPACT](dev, count, p);
-    } else {
-	next_tick = dc_infoblock[*(p+1)](dev, count, p);
-    }
-    if (lp->tcount == count) {
-	lp->media = NC;
-        if (lp->media != lp->c_media) {
-	    de4x5_dbg_media(dev);
-	    lp->c_media = lp->media;
-	}
-	lp->media = INIT;
-	lp->tcount = 0;
-	lp->tx_enable = false;
-    }
-
-    return next_tick & ~TIMER_CB;
-}
-
-/*
-** The compact infoblock is only designed for DC21140[A] chips, so
-** we'll reuse the dc21140m_autoconf function. Non MII media only.
-*/
-static int
-compact_infoblock(struct net_device *dev, u_char count, u_char *p)
-{
-    struct de4x5_private *lp = netdev_priv(dev);
-    u_char flags, csr6;
-
-    /* Recursively figure out the info blocks */
-    if (--count > lp->tcount) {
-	if (*(p+COMPACT_LEN) < 128) {
-	    return dc_infoblock[COMPACT](dev, count, p+COMPACT_LEN);
-	} else {
-	    return dc_infoblock[*(p+COMPACT_LEN+1)](dev, count, p+COMPACT_LEN);
-	}
-    }
-
-    if ((lp->media == INIT) && (lp->timeout < 0)) {
-        lp->ibn = COMPACT;
-        lp->active = 0;
-	gep_wr(lp->cache.gepc, dev);
-	lp->infoblock_media = (*p++) & COMPACT_MC;
-	lp->cache.gep = *p++;
-	csr6 = *p++;
-	flags = *p++;
-
-	lp->asBitValid = (flags & 0x80) ? 0 : -1;
-	lp->defMedium = (flags & 0x40) ? -1 : 0;
-	lp->asBit = 1 << ((csr6 >> 1) & 0x07);
-	lp->asPolarity = ((csr6 & 0x80) ? -1 : 0) & lp->asBit;
-	lp->infoblock_csr6 = OMR_DEF | ((csr6 & 0x71) << 18);
-	lp->useMII = false;
-
-	de4x5_switch_mac_port(dev);
-    }
-
-    return dc21140m_autoconf(dev);
-}
-
-/*
-** This block describes non MII media for the DC21140[A] only.
-*/
-static int
-type0_infoblock(struct net_device *dev, u_char count, u_char *p)
-{
-    struct de4x5_private *lp = netdev_priv(dev);
-    u_char flags, csr6, len = (*p & BLOCK_LEN)+1;
-
-    /* Recursively figure out the info blocks */
-    if (--count > lp->tcount) {
-	if (*(p+len) < 128) {
-	    return dc_infoblock[COMPACT](dev, count, p+len);
-	} else {
-	    return dc_infoblock[*(p+len+1)](dev, count, p+len);
-	}
-    }
-
-    if ((lp->media == INIT) && (lp->timeout < 0)) {
-        lp->ibn = 0;
-        lp->active = 0;
-        gep_wr(lp->cache.gepc, dev);
-	p+=2;
-	lp->infoblock_media = (*p++) & BLOCK0_MC;
-	lp->cache.gep = *p++;
-	csr6 = *p++;
-	flags = *p++;
-
-	lp->asBitValid = (flags & 0x80) ? 0 : -1;
-	lp->defMedium = (flags & 0x40) ? -1 : 0;
-	lp->asBit = 1 << ((csr6 >> 1) & 0x07);
-	lp->asPolarity = ((csr6 & 0x80) ? -1 : 0) & lp->asBit;
-	lp->infoblock_csr6 = OMR_DEF | ((csr6 & 0x71) << 18);
-	lp->useMII = false;
-
-	de4x5_switch_mac_port(dev);
-    }
-
-    return dc21140m_autoconf(dev);
-}
-
-/* These functions are under construction! */
-
-static int
-type1_infoblock(struct net_device *dev, u_char count, u_char *p)
-{
-    struct de4x5_private *lp = netdev_priv(dev);
-    u_char len = (*p & BLOCK_LEN)+1;
-
-    /* Recursively figure out the info blocks */
-    if (--count > lp->tcount) {
-	if (*(p+len) < 128) {
-	    return dc_infoblock[COMPACT](dev, count, p+len);
-	} else {
-	    return dc_infoblock[*(p+len+1)](dev, count, p+len);
-	}
-    }
-
-    p += 2;
-    if (lp->state == INITIALISED) {
-        lp->ibn = 1;
-	lp->active = *p++;
-	lp->phy[lp->active].gep = (*p ? p : NULL); p += (*p + 1);
-	lp->phy[lp->active].rst = (*p ? p : NULL); p += (*p + 1);
-	lp->phy[lp->active].mc  = get_unaligned_le16(p); p += 2;
-	lp->phy[lp->active].ana = get_unaligned_le16(p); p += 2;
-	lp->phy[lp->active].fdx = get_unaligned_le16(p); p += 2;
-	lp->phy[lp->active].ttm = get_unaligned_le16(p);
-	return 0;
-    } else if ((lp->media == INIT) && (lp->timeout < 0)) {
-        lp->ibn = 1;
-        lp->active = *p;
-	lp->infoblock_csr6 = OMR_MII_100;
-	lp->useMII = true;
-	lp->infoblock_media = ANS;
-
-	de4x5_switch_mac_port(dev);
-    }
-
-    return dc21140m_autoconf(dev);
-}
-
-static int
-type2_infoblock(struct net_device *dev, u_char count, u_char *p)
-{
-    struct de4x5_private *lp = netdev_priv(dev);
-    u_char len = (*p & BLOCK_LEN)+1;
-
-    /* Recursively figure out the info blocks */
-    if (--count > lp->tcount) {
-	if (*(p+len) < 128) {
-	    return dc_infoblock[COMPACT](dev, count, p+len);
-	} else {
-	    return dc_infoblock[*(p+len+1)](dev, count, p+len);
-	}
-    }
-
-    if ((lp->media == INIT) && (lp->timeout < 0)) {
-        lp->ibn = 2;
-        lp->active = 0;
-	p += 2;
-	lp->infoblock_media = (*p) & MEDIA_CODE;
-
-        if ((*p++) & EXT_FIELD) {
-	    lp->cache.csr13 = get_unaligned_le16(p); p += 2;
-	    lp->cache.csr14 = get_unaligned_le16(p); p += 2;
-	    lp->cache.csr15 = get_unaligned_le16(p); p += 2;
-	} else {
-	    lp->cache.csr13 = CSR13;
-	    lp->cache.csr14 = CSR14;
-	    lp->cache.csr15 = CSR15;
-	}
-        lp->cache.gepc = ((s32)(get_unaligned_le16(p)) << 16); p += 2;
-        lp->cache.gep  = ((s32)(get_unaligned_le16(p)) << 16);
-	lp->infoblock_csr6 = OMR_SIA;
-	lp->useMII = false;
-
-	de4x5_switch_mac_port(dev);
-    }
-
-    return dc2114x_autoconf(dev);
-}
-
-static int
-type3_infoblock(struct net_device *dev, u_char count, u_char *p)
-{
-    struct de4x5_private *lp = netdev_priv(dev);
-    u_char len = (*p & BLOCK_LEN)+1;
-
-    /* Recursively figure out the info blocks */
-    if (--count > lp->tcount) {
-	if (*(p+len) < 128) {
-	    return dc_infoblock[COMPACT](dev, count, p+len);
-	} else {
-	    return dc_infoblock[*(p+len+1)](dev, count, p+len);
-	}
-    }
-
-    p += 2;
-    if (lp->state == INITIALISED) {
-        lp->ibn = 3;
-        lp->active = *p++;
-	if (MOTO_SROM_BUG) lp->active = 0;
-	lp->phy[lp->active].gep = (*p ? p : NULL); p += (2 * (*p) + 1);
-	lp->phy[lp->active].rst = (*p ? p : NULL); p += (2 * (*p) + 1);
-	lp->phy[lp->active].mc  = get_unaligned_le16(p); p += 2;
-	lp->phy[lp->active].ana = get_unaligned_le16(p); p += 2;
-	lp->phy[lp->active].fdx = get_unaligned_le16(p); p += 2;
-	lp->phy[lp->active].ttm = get_unaligned_le16(p); p += 2;
-	lp->phy[lp->active].mci = *p;
-	return 0;
-    } else if ((lp->media == INIT) && (lp->timeout < 0)) {
-        lp->ibn = 3;
-	lp->active = *p;
-	if (MOTO_SROM_BUG) lp->active = 0;
-	lp->infoblock_csr6 = OMR_MII_100;
-	lp->useMII = true;
-	lp->infoblock_media = ANS;
-
-	de4x5_switch_mac_port(dev);
-    }
-
-    return dc2114x_autoconf(dev);
-}
-
-static int
-type4_infoblock(struct net_device *dev, u_char count, u_char *p)
-{
-    struct de4x5_private *lp = netdev_priv(dev);
-    u_char flags, csr6, len = (*p & BLOCK_LEN)+1;
-
-    /* Recursively figure out the info blocks */
-    if (--count > lp->tcount) {
-	if (*(p+len) < 128) {
-	    return dc_infoblock[COMPACT](dev, count, p+len);
-	} else {
-	    return dc_infoblock[*(p+len+1)](dev, count, p+len);
-	}
-    }
-
-    if ((lp->media == INIT) && (lp->timeout < 0)) {
-        lp->ibn = 4;
-        lp->active = 0;
-	p+=2;
-	lp->infoblock_media = (*p++) & MEDIA_CODE;
-        lp->cache.csr13 = CSR13;              /* Hard coded defaults */
-	lp->cache.csr14 = CSR14;
-	lp->cache.csr15 = CSR15;
-        lp->cache.gepc = ((s32)(get_unaligned_le16(p)) << 16); p += 2;
-        lp->cache.gep  = ((s32)(get_unaligned_le16(p)) << 16); p += 2;
-	csr6 = *p++;
-	flags = *p++;
-
-	lp->asBitValid = (flags & 0x80) ? 0 : -1;
-	lp->defMedium = (flags & 0x40) ? -1 : 0;
-	lp->asBit = 1 << ((csr6 >> 1) & 0x07);
-	lp->asPolarity = ((csr6 & 0x80) ? -1 : 0) & lp->asBit;
-	lp->infoblock_csr6 = OMR_DEF | ((csr6 & 0x71) << 18);
-	lp->useMII = false;
-
-	de4x5_switch_mac_port(dev);
-    }
-
-    return dc2114x_autoconf(dev);
-}
-
-/*
-** This block type provides information for resetting external devices
-** (chips) through the General Purpose Register.
-*/
-static int
-type5_infoblock(struct net_device *dev, u_char count, u_char *p)
-{
-    struct de4x5_private *lp = netdev_priv(dev);
-    u_char len = (*p & BLOCK_LEN)+1;
-
-    /* Recursively figure out the info blocks */
-    if (--count > lp->tcount) {
-	if (*(p+len) < 128) {
-	    return dc_infoblock[COMPACT](dev, count, p+len);
-	} else {
-	    return dc_infoblock[*(p+len+1)](dev, count, p+len);
-	}
-    }
-
-    /* Must be initializing to run this code */
-    if ((lp->state == INITIALISED) || (lp->media == INIT)) {
-	p+=2;
-        lp->rst = p;
-        srom_exec(dev, lp->rst);
-    }
-
-    return DE4X5_AUTOSENSE_MS;
-}
-
-/*
-** MII Read/Write
-*/
-
-static int
-mii_rd(u_char phyreg, u_char phyaddr, u_long ioaddr)
-{
-    mii_wdata(MII_PREAMBLE,  2, ioaddr);   /* Start of 34 bit preamble...    */
-    mii_wdata(MII_PREAMBLE, 32, ioaddr);   /* ...continued                   */
-    mii_wdata(MII_STRD, 4, ioaddr);        /* SFD and Read operation         */
-    mii_address(phyaddr, ioaddr);          /* PHY address to be accessed     */
-    mii_address(phyreg, ioaddr);           /* PHY Register to read           */
-    mii_ta(MII_STRD, ioaddr);              /* Turn around time - 2 MDC       */
-
-    return mii_rdata(ioaddr);              /* Read data                      */
-}
-
-static void
-mii_wr(int data, u_char phyreg, u_char phyaddr, u_long ioaddr)
-{
-    mii_wdata(MII_PREAMBLE,  2, ioaddr);   /* Start of 34 bit preamble...    */
-    mii_wdata(MII_PREAMBLE, 32, ioaddr);   /* ...continued                   */
-    mii_wdata(MII_STWR, 4, ioaddr);        /* SFD and Write operation        */
-    mii_address(phyaddr, ioaddr);          /* PHY address to be accessed     */
-    mii_address(phyreg, ioaddr);           /* PHY Register to write          */
-    mii_ta(MII_STWR, ioaddr);              /* Turn around time - 2 MDC       */
-    data = mii_swap(data, 16);             /* Swap data bit ordering         */
-    mii_wdata(data, 16, ioaddr);           /* Write data                     */
-}
-
-static int
-mii_rdata(u_long ioaddr)
-{
-    int i;
-    s32 tmp = 0;
-
-    for (i=0; i<16; i++) {
-	tmp <<= 1;
-	tmp |= getfrom_mii(MII_MRD | MII_RD, ioaddr);
-    }
-
-    return tmp;
-}
-
-static void
-mii_wdata(int data, int len, u_long ioaddr)
-{
-    int i;
-
-    for (i=0; i<len; i++) {
-	sendto_mii(MII_MWR | MII_WR, data, ioaddr);
-	data >>= 1;
-    }
-}
-
-static void
-mii_address(u_char addr, u_long ioaddr)
-{
-    int i;
-
-    addr = mii_swap(addr, 5);
-    for (i=0; i<5; i++) {
-	sendto_mii(MII_MWR | MII_WR, addr, ioaddr);
-	addr >>= 1;
-    }
-}
-
-static void
-mii_ta(u_long rw, u_long ioaddr)
-{
-    if (rw == MII_STWR) {
-	sendto_mii(MII_MWR | MII_WR, 1, ioaddr);
-	sendto_mii(MII_MWR | MII_WR, 0, ioaddr);
-    } else {
-	getfrom_mii(MII_MRD | MII_RD, ioaddr);        /* Tri-state MDIO */
-    }
-}
-
-static int
-mii_swap(int data, int len)
-{
-    int i, tmp = 0;
-
-    for (i=0; i<len; i++) {
-	tmp <<= 1;
-	tmp |= (data & 1);
-	data >>= 1;
-    }
-
-    return tmp;
-}
-
-static void
-sendto_mii(u32 command, int data, u_long ioaddr)
-{
-    u32 j;
-
-    j = (data & 1) << 17;
-    outl(command | j, ioaddr);
-    udelay(1);
-    outl(command | MII_MDC | j, ioaddr);
-    udelay(1);
-}
-
-static int
-getfrom_mii(u32 command, u_long ioaddr)
-{
-    outl(command, ioaddr);
-    udelay(1);
-    outl(command | MII_MDC, ioaddr);
-    udelay(1);
-
-    return (inl(ioaddr) >> 19) & 1;
-}
-
-/*
-** Here's 3 ways to calculate the OUI from the ID registers.
-*/
-static int
-mii_get_oui(u_char phyaddr, u_long ioaddr)
-{
-/*
-    union {
-	u_short reg;
-	u_char breg[2];
-    } a;
-    int i, r2, r3, ret=0;*/
-    int r2, r3;
-
-    /* Read r2 and r3 */
-    r2 = mii_rd(MII_ID0, phyaddr, ioaddr);
-    r3 = mii_rd(MII_ID1, phyaddr, ioaddr);
-                                                /* SEEQ and Cypress way * /
-    / * Shuffle r2 and r3 * /
-    a.reg=0;
-    r3 = ((r3>>10)|(r2<<6))&0x0ff;
-    r2 = ((r2>>2)&0x3fff);
-
-    / * Bit reverse r3 * /
-    for (i=0;i<8;i++) {
-	ret<<=1;
-	ret |= (r3&1);
-	r3>>=1;
-    }
-
-    / * Bit reverse r2 * /
-    for (i=0;i<16;i++) {
-	a.reg<<=1;
-	a.reg |= (r2&1);
-	r2>>=1;
-    }
-
-    / * Swap r2 bytes * /
-    i=a.breg[0];
-    a.breg[0]=a.breg[1];
-    a.breg[1]=i;
-
-    return (a.reg<<8)|ret; */                 /* SEEQ and Cypress way */
-/*    return (r2<<6)|(u_int)(r3>>10); */      /* NATIONAL and BROADCOM way */
-    return r2;                                  /* (I did it) My way */
-}
-
-/*
-** The SROM spec forces us to search addresses [1-31 0]. Bummer.
-*/
-static int
-mii_get_phy(struct net_device *dev)
-{
-    struct de4x5_private *lp = netdev_priv(dev);
-    u_long iobase = dev->base_addr;
-    int i, j, k, n, limit=ARRAY_SIZE(phy_info);
-    int id;
-
-    lp->active = 0;
-    lp->useMII = true;
-
-    /* Search the MII address space for possible PHY devices */
-    for (n=0, lp->mii_cnt=0, i=1; !((i==1) && (n==1)); i=(i+1)%DE4X5_MAX_MII) {
-	lp->phy[lp->active].addr = i;
-	if (i==0) n++;                             /* Count cycles */
-	while (de4x5_reset_phy(dev)<0) udelay(100);/* Wait for reset */
-	id = mii_get_oui(i, DE4X5_MII);
-	if ((id == 0) || (id == 65535)) continue;  /* Valid ID? */
-	for (j=0; j<limit; j++) {                  /* Search PHY table */
-	    if (id != phy_info[j].id) continue;    /* ID match? */
-	    for (k=0; k < DE4X5_MAX_PHY && lp->phy[k].id; k++);
-	    if (k < DE4X5_MAX_PHY) {
-		memcpy((char *)&lp->phy[k],
-		       (char *)&phy_info[j], sizeof(struct phy_table));
-		lp->phy[k].addr = i;
-		lp->mii_cnt++;
-		lp->active++;
-	    } else {
-		goto purgatory;                    /* Stop the search */
-	    }
-	    break;
-	}
-	if ((j == limit) && (i < DE4X5_MAX_MII)) {
-	    for (k=0; k < DE4X5_MAX_PHY && lp->phy[k].id; k++);
-	    lp->phy[k].addr = i;
-	    lp->phy[k].id = id;
-	    lp->phy[k].spd.reg = GENERIC_REG;      /* ANLPA register         */
-	    lp->phy[k].spd.mask = GENERIC_MASK;    /* 100Mb/s technologies   */
-	    lp->phy[k].spd.value = GENERIC_VALUE;  /* TX & T4, H/F Duplex    */
-	    lp->mii_cnt++;
-	    lp->active++;
-	    printk("%s: Using generic MII device control. If the board doesn't operate,\nplease mail the following dump to the author:\n", dev->name);
-	    j = de4x5_debug;
-	    de4x5_debug |= DEBUG_MII;
-	    de4x5_dbg_mii(dev, k);
-	    de4x5_debug = j;
-	    printk("\n");
-	}
-    }
-  purgatory:
-    lp->active = 0;
-    if (lp->phy[0].id) {                           /* Reset the PHY devices */
-	for (k=0; k < DE4X5_MAX_PHY && lp->phy[k].id; k++) { /*For each PHY*/
-	    mii_wr(MII_CR_RST, MII_CR, lp->phy[k].addr, DE4X5_MII);
-	    while (mii_rd(MII_CR, lp->phy[k].addr, DE4X5_MII) & MII_CR_RST);
-
-	    de4x5_dbg_mii(dev, k);
-	}
-    }
-    if (!lp->mii_cnt) lp->useMII = false;
-
-    return lp->mii_cnt;
-}
-
-static char *
-build_setup_frame(struct net_device *dev, int mode)
-{
-    struct de4x5_private *lp = netdev_priv(dev);
-    int i;
-    char *pa = lp->setup_frame;
-
-    /* Initialise the setup frame */
-    if (mode == ALL) {
-	memset(lp->setup_frame, 0, SETUP_FRAME_LEN);
-    }
-
-    if (lp->setup_f == HASH_PERF) {
-	for (pa=lp->setup_frame+IMPERF_PA_OFFSET, i=0; i<ETH_ALEN; i++) {
-	    *(pa + i) = dev->dev_addr[i];                 /* Host address */
-	    if (i & 0x01) pa += 2;
-	}
-	*(lp->setup_frame + (DE4X5_HASH_TABLE_LEN >> 3) - 3) = 0x80;
-    } else {
-	for (i=0; i<ETH_ALEN; i++) { /* Host address */
-	    *(pa + (i&1)) = dev->dev_addr[i];
-	    if (i & 0x01) pa += 4;
-	}
-	for (i=0; i<ETH_ALEN; i++) { /* Broadcast address */
-	    *(pa + (i&1)) = (char) 0xff;
-	    if (i & 0x01) pa += 4;
-	}
-    }
-
-    return pa;                     /* Points to the next entry */
-}
-
-static void
-disable_ast(struct net_device *dev)
-{
-	struct de4x5_private *lp = netdev_priv(dev);
-	del_timer_sync(&lp->timer);
-}
-
-static long
-de4x5_switch_mac_port(struct net_device *dev)
-{
-    struct de4x5_private *lp = netdev_priv(dev);
-    u_long iobase = dev->base_addr;
-    s32 omr;
-
-    STOP_DE4X5;
-
-    /* Assert the OMR_PS bit in CSR6 */
-    omr = (inl(DE4X5_OMR) & ~(OMR_PS | OMR_HBD | OMR_TTM | OMR_PCS | OMR_SCR |
-			                                             OMR_FDX));
-    omr |= lp->infoblock_csr6;
-    if (omr & OMR_PS) omr |= OMR_HBD;
-    outl(omr, DE4X5_OMR);
-
-    /* Soft Reset */
-    RESET_DE4X5;
-
-    /* Restore the GEP - especially for COMPACT and Type 0 Infoblocks */
-    if (lp->chipset == DC21140) {
-	gep_wr(lp->cache.gepc, dev);
-	gep_wr(lp->cache.gep, dev);
-    } else if ((lp->chipset & ~0x0ff) == DC2114x) {
-	reset_init_sia(dev, lp->cache.csr13, lp->cache.csr14, lp->cache.csr15);
-    }
-
-    /* Restore CSR6 */
-    outl(omr, DE4X5_OMR);
-
-    /* Reset CSR8 */
-    inl(DE4X5_MFC);
-
-    return omr;
-}
-
-static void
-gep_wr(s32 data, struct net_device *dev)
-{
-    struct de4x5_private *lp = netdev_priv(dev);
-    u_long iobase = dev->base_addr;
-
-    if (lp->chipset == DC21140) {
-	outl(data, DE4X5_GEP);
-    } else if ((lp->chipset & ~0x00ff) == DC2114x) {
-	outl((data<<16) | lp->cache.csr15, DE4X5_SIGR);
-    }
-}
-
-static int
-gep_rd(struct net_device *dev)
-{
-    struct de4x5_private *lp = netdev_priv(dev);
-    u_long iobase = dev->base_addr;
-
-    if (lp->chipset == DC21140) {
-	return inl(DE4X5_GEP);
-    } else if ((lp->chipset & ~0x00ff) == DC2114x) {
-	return inl(DE4X5_SIGR) & 0x000fffff;
-    }
-
-    return 0;
-}
-
-static void
-yawn(struct net_device *dev, int state)
-{
-    struct de4x5_private *lp = netdev_priv(dev);
-    u_long iobase = dev->base_addr;
-
-    if ((lp->chipset == DC21040) || (lp->chipset == DC21140)) return;
-
-    if(lp->bus == EISA) {
-	switch(state) {
-	  case WAKEUP:
-	    outb(WAKEUP, PCI_CFPM);
-	    mdelay(10);
-	    break;
-
-	  case SNOOZE:
-	    outb(SNOOZE, PCI_CFPM);
-	    break;
-
-	  case SLEEP:
-	    outl(0, DE4X5_SICR);
-	    outb(SLEEP, PCI_CFPM);
-	    break;
-	}
-    } else {
-	struct pci_dev *pdev = to_pci_dev (lp->gendev);
-	switch(state) {
-	  case WAKEUP:
-	    pci_write_config_byte(pdev, PCI_CFDA_PSM, WAKEUP);
-	    mdelay(10);
-	    break;
-
-	  case SNOOZE:
-	    pci_write_config_byte(pdev, PCI_CFDA_PSM, SNOOZE);
-	    break;
-
-	  case SLEEP:
-	    outl(0, DE4X5_SICR);
-	    pci_write_config_byte(pdev, PCI_CFDA_PSM, SLEEP);
-	    break;
-	}
-    }
-}
-
-static void
-de4x5_parse_params(struct net_device *dev)
-{
-    struct de4x5_private *lp = netdev_priv(dev);
-    char *p, *q, t;
-
-    lp->params.fdx = false;
-    lp->params.autosense = AUTO;
-
-    if (args == NULL) return;
-
-    if ((p = strstr(args, dev->name))) {
-	if (!(q = strstr(p+strlen(dev->name), "eth"))) q = p + strlen(p);
-	t = *q;
-	*q = '\0';
-
-	if (strstr(p, "fdx") || strstr(p, "FDX")) lp->params.fdx = true;
-
-	if (strstr(p, "autosense") || strstr(p, "AUTOSENSE")) {
-	    if (strstr(p, "TP_NW")) {
-		lp->params.autosense = TP_NW;
-	    } else if (strstr(p, "TP")) {
-		lp->params.autosense = TP;
-	    } else if (strstr(p, "BNC_AUI")) {
-		lp->params.autosense = BNC;
-	    } else if (strstr(p, "BNC")) {
-		lp->params.autosense = BNC;
-	    } else if (strstr(p, "AUI")) {
-		lp->params.autosense = AUI;
-	    } else if (strstr(p, "10Mb")) {
-		lp->params.autosense = _10Mb;
-	    } else if (strstr(p, "100Mb")) {
-		lp->params.autosense = _100Mb;
-	    } else if (strstr(p, "AUTO")) {
-		lp->params.autosense = AUTO;
-	    }
-	}
-	*q = t;
-    }
-}
-
-static void
-de4x5_dbg_open(struct net_device *dev)
-{
-    struct de4x5_private *lp = netdev_priv(dev);
-    int i;
-
-    if (de4x5_debug & DEBUG_OPEN) {
-	printk("%s: de4x5 opening with irq %d\n",dev->name,dev->irq);
-	printk("\tphysical address: %pM\n", dev->dev_addr);
-	printk("Descriptor head addresses:\n");
-	printk("\t0x%8.8lx  0x%8.8lx\n",(u_long)lp->rx_ring,(u_long)lp->tx_ring);
-	printk("Descriptor addresses:\nRX: ");
-	for (i=0;i<lp->rxRingSize-1;i++){
-	    if (i < 3) {
-		printk("0x%8.8lx  ",(u_long)&lp->rx_ring[i].status);
-	    }
-	}
-	printk("...0x%8.8lx\n",(u_long)&lp->rx_ring[i].status);
-	printk("TX: ");
-	for (i=0;i<lp->txRingSize-1;i++){
-	    if (i < 3) {
-		printk("0x%8.8lx  ", (u_long)&lp->tx_ring[i].status);
-	    }
-	}
-	printk("...0x%8.8lx\n", (u_long)&lp->tx_ring[i].status);
-	printk("Descriptor buffers:\nRX: ");
-	for (i=0;i<lp->rxRingSize-1;i++){
-	    if (i < 3) {
-		printk("0x%8.8x  ",le32_to_cpu(lp->rx_ring[i].buf));
-	    }
-	}
-	printk("...0x%8.8x\n",le32_to_cpu(lp->rx_ring[i].buf));
-	printk("TX: ");
-	for (i=0;i<lp->txRingSize-1;i++){
-	    if (i < 3) {
-		printk("0x%8.8x  ", le32_to_cpu(lp->tx_ring[i].buf));
-	    }
-	}
-	printk("...0x%8.8x\n", le32_to_cpu(lp->tx_ring[i].buf));
-	printk("Ring size:\nRX: %d\nTX: %d\n",
-	       (short)lp->rxRingSize,
-	       (short)lp->txRingSize);
-    }
-}
-
-static void
-de4x5_dbg_mii(struct net_device *dev, int k)
-{
-    struct de4x5_private *lp = netdev_priv(dev);
-    u_long iobase = dev->base_addr;
-
-    if (de4x5_debug & DEBUG_MII) {
-	printk("\nMII device address: %d\n", lp->phy[k].addr);
-	printk("MII CR:  %x\n",mii_rd(MII_CR,lp->phy[k].addr,DE4X5_MII));
-	printk("MII SR:  %x\n",mii_rd(MII_SR,lp->phy[k].addr,DE4X5_MII));
-	printk("MII ID0: %x\n",mii_rd(MII_ID0,lp->phy[k].addr,DE4X5_MII));
-	printk("MII ID1: %x\n",mii_rd(MII_ID1,lp->phy[k].addr,DE4X5_MII));
-	if (lp->phy[k].id != BROADCOM_T4) {
-	    printk("MII ANA: %x\n",mii_rd(0x04,lp->phy[k].addr,DE4X5_MII));
-	    printk("MII ANC: %x\n",mii_rd(0x05,lp->phy[k].addr,DE4X5_MII));
-	}
-	printk("MII 16:  %x\n",mii_rd(0x10,lp->phy[k].addr,DE4X5_MII));
-	if (lp->phy[k].id != BROADCOM_T4) {
-	    printk("MII 17:  %x\n",mii_rd(0x11,lp->phy[k].addr,DE4X5_MII));
-	    printk("MII 18:  %x\n",mii_rd(0x12,lp->phy[k].addr,DE4X5_MII));
-	} else {
-	    printk("MII 20:  %x\n",mii_rd(0x14,lp->phy[k].addr,DE4X5_MII));
-	}
-    }
-}
-
-static void
-de4x5_dbg_media(struct net_device *dev)
-{
-    struct de4x5_private *lp = netdev_priv(dev);
-
-    if (lp->media != lp->c_media) {
-	if (de4x5_debug & DEBUG_MEDIA) {
-	    printk("%s: media is %s%s\n", dev->name,
-		   (lp->media == NC  ? "unconnected, link down or incompatible connection" :
-		    (lp->media == TP  ? "TP" :
-		     (lp->media == ANS ? "TP/Nway" :
-		      (lp->media == BNC ? "BNC" :
-		       (lp->media == AUI ? "AUI" :
-			(lp->media == BNC_AUI ? "BNC/AUI" :
-			 (lp->media == EXT_SIA ? "EXT SIA" :
-			  (lp->media == _100Mb  ? "100Mb/s" :
-			   (lp->media == _10Mb   ? "10Mb/s" :
-			    "???"
-			    ))))))))), (lp->fdx?" full duplex.":"."));
-	}
-	lp->c_media = lp->media;
-    }
-}
-
-static void
-de4x5_dbg_srom(struct de4x5_srom *p)
-{
-    int i;
-
-    if (de4x5_debug & DEBUG_SROM) {
-	printk("Sub-system Vendor ID: %04x\n", *((u_short *)p->sub_vendor_id));
-	printk("Sub-system ID:        %04x\n", *((u_short *)p->sub_system_id));
-	printk("ID Block CRC:         %02x\n", (u_char)(p->id_block_crc));
-	printk("SROM version:         %02x\n", (u_char)(p->version));
-	printk("# controllers:        %02x\n", (u_char)(p->num_controllers));
-
-	printk("Hardware Address:     %pM\n", p->ieee_addr);
-	printk("CRC checksum:         %04x\n", (u_short)(p->chksum));
-	for (i=0; i<64; i++) {
-	    printk("%3d %04x\n", i<<1, (u_short)*((u_short *)p+i));
-	}
-    }
-}
-
-static void
-de4x5_dbg_rx(struct sk_buff *skb, int len)
-{
-    int i, j;
-
-    if (de4x5_debug & DEBUG_RX) {
-	printk("R: %pM <- %pM len/SAP:%02x%02x [%d]\n",
-	       skb->data, &skb->data[6],
-	       (u_char)skb->data[12],
-	       (u_char)skb->data[13],
-	       len);
-	for (j=0; len>0;j+=16, len-=16) {
-	  printk("    %03x: ",j);
-	  for (i=0; i<16 && i<len; i++) {
-	    printk("%02x ",(u_char)skb->data[i+j]);
-	  }
-	  printk("\n");
-	}
-    }
-}
-
-/*
-** Perform IOCTL call functions here. Some are privileged operations and the
-** effective uid is checked in those cases. In the normal course of events
-** this function is only used for my testing.
-*/
-static int
-de4x5_ioctl(struct net_device *dev, struct ifreq *rq, int cmd)
-{
-    struct de4x5_private *lp = netdev_priv(dev);
-    struct de4x5_ioctl *ioc = (struct de4x5_ioctl *) &rq->ifr_ifru;
-    u_long iobase = dev->base_addr;
-    int i, j, status = 0;
-    s32 omr;
-    union {
-	u8  addr[144];
-	u16 sval[72];
-	u32 lval[36];
-    } tmp;
-    u_long flags = 0;
-
-    switch(ioc->cmd) {
-    case DE4X5_GET_HWADDR:           /* Get the hardware address */
-	ioc->len = ETH_ALEN;
-	for (i=0; i<ETH_ALEN; i++) {
-	    tmp.addr[i] = dev->dev_addr[i];
-	}
-	if (copy_to_user(ioc->data, tmp.addr, ioc->len)) return -EFAULT;
-	break;
-
-    case DE4X5_SET_HWADDR:           /* Set the hardware address */
-	if (!capable(CAP_NET_ADMIN)) return -EPERM;
-	if (copy_from_user(tmp.addr, ioc->data, ETH_ALEN)) return -EFAULT;
-	if (netif_queue_stopped(dev))
-		return -EBUSY;
-	netif_stop_queue(dev);
-	for (i=0; i<ETH_ALEN; i++) {
-	    dev->dev_addr[i] = tmp.addr[i];
-	}
-	build_setup_frame(dev, PHYS_ADDR_ONLY);
-	/* Set up the descriptor and give ownership to the card */
-	load_packet(dev, lp->setup_frame, TD_IC | PERFECT_F | TD_SET |
-		                                       SETUP_FRAME_LEN, (struct sk_buff *)1);
-	lp->tx_new = (lp->tx_new + 1) % lp->txRingSize;
-	outl(POLL_DEMAND, DE4X5_TPD);                /* Start the TX */
-	netif_wake_queue(dev);                      /* Unlock the TX ring */
-	break;
-
-    case DE4X5_SAY_BOO:              /* Say "Boo!" to the kernel log file */
-	if (!capable(CAP_NET_ADMIN)) return -EPERM;
-	printk("%s: Boo!\n", dev->name);
-	break;
-
-    case DE4X5_MCA_EN:               /* Enable pass all multicast addressing */
-	if (!capable(CAP_NET_ADMIN)) return -EPERM;
-	omr = inl(DE4X5_OMR);
-	omr |= OMR_PM;
-	outl(omr, DE4X5_OMR);
-	break;
-
-    case DE4X5_GET_STATS:            /* Get the driver statistics */
-    {
-        struct pkt_stats statbuf;
-	ioc->len = sizeof(statbuf);
-	spin_lock_irqsave(&lp->lock, flags);
-	memcpy(&statbuf, &lp->pktStats, ioc->len);
-	spin_unlock_irqrestore(&lp->lock, flags);
-	if (copy_to_user(ioc->data, &statbuf, ioc->len))
-		return -EFAULT;
-	break;
-    }
-    case DE4X5_CLR_STATS:            /* Zero out the driver statistics */
-	if (!capable(CAP_NET_ADMIN)) return -EPERM;
-	spin_lock_irqsave(&lp->lock, flags);
-	memset(&lp->pktStats, 0, sizeof(lp->pktStats));
-	spin_unlock_irqrestore(&lp->lock, flags);
-	break;
-
-    case DE4X5_GET_OMR:              /* Get the OMR Register contents */
-	tmp.addr[0] = inl(DE4X5_OMR);
-	if (copy_to_user(ioc->data, tmp.addr, 1)) return -EFAULT;
-	break;
-
-    case DE4X5_SET_OMR:              /* Set the OMR Register contents */
-	if (!capable(CAP_NET_ADMIN)) return -EPERM;
-	if (copy_from_user(tmp.addr, ioc->data, 1)) return -EFAULT;
-	outl(tmp.addr[0], DE4X5_OMR);
-	break;
-
-    case DE4X5_GET_REG:              /* Get the DE4X5 Registers */
-	j = 0;
-	tmp.lval[0] = inl(DE4X5_STS); j+=4;
-	tmp.lval[1] = inl(DE4X5_BMR); j+=4;
-	tmp.lval[2] = inl(DE4X5_IMR); j+=4;
-	tmp.lval[3] = inl(DE4X5_OMR); j+=4;
-	tmp.lval[4] = inl(DE4X5_SISR); j+=4;
-	tmp.lval[5] = inl(DE4X5_SICR); j+=4;
-	tmp.lval[6] = inl(DE4X5_STRR); j+=4;
-	tmp.lval[7] = inl(DE4X5_SIGR); j+=4;
-	ioc->len = j;
-	if (copy_to_user(ioc->data, tmp.lval, ioc->len))
-		return -EFAULT;
-	break;
-
-#define DE4X5_DUMP              0x0f /* Dump the DE4X5 Status */
-/*
-      case DE4X5_DUMP:
-	j = 0;
-	tmp.addr[j++] = dev->irq;
-	for (i=0; i<ETH_ALEN; i++) {
-	    tmp.addr[j++] = dev->dev_addr[i];
-	}
-	tmp.addr[j++] = lp->rxRingSize;
-	tmp.lval[j>>2] = (long)lp->rx_ring; j+=4;
-	tmp.lval[j>>2] = (long)lp->tx_ring; j+=4;
-
-	for (i=0;i<lp->rxRingSize-1;i++){
-	    if (i < 3) {
-		tmp.lval[j>>2] = (long)&lp->rx_ring[i].status; j+=4;
-	    }
-	}
-	tmp.lval[j>>2] = (long)&lp->rx_ring[i].status; j+=4;
-	for (i=0;i<lp->txRingSize-1;i++){
-	    if (i < 3) {
-		tmp.lval[j>>2] = (long)&lp->tx_ring[i].status; j+=4;
-	    }
-	}
-	tmp.lval[j>>2] = (long)&lp->tx_ring[i].status; j+=4;
-
-	for (i=0;i<lp->rxRingSize-1;i++){
-	    if (i < 3) {
-		tmp.lval[j>>2] = (s32)le32_to_cpu(lp->rx_ring[i].buf); j+=4;
-	    }
-	}
-	tmp.lval[j>>2] = (s32)le32_to_cpu(lp->rx_ring[i].buf); j+=4;
-	for (i=0;i<lp->txRingSize-1;i++){
-	    if (i < 3) {
-		tmp.lval[j>>2] = (s32)le32_to_cpu(lp->tx_ring[i].buf); j+=4;
-	    }
-	}
-	tmp.lval[j>>2] = (s32)le32_to_cpu(lp->tx_ring[i].buf); j+=4;
-
-	for (i=0;i<lp->rxRingSize;i++){
-	    tmp.lval[j>>2] = le32_to_cpu(lp->rx_ring[i].status); j+=4;
-	}
-	for (i=0;i<lp->txRingSize;i++){
-	    tmp.lval[j>>2] = le32_to_cpu(lp->tx_ring[i].status); j+=4;
-	}
-
-	tmp.lval[j>>2] = inl(DE4X5_BMR);  j+=4;
-	tmp.lval[j>>2] = inl(DE4X5_TPD);  j+=4;
-	tmp.lval[j>>2] = inl(DE4X5_RPD);  j+=4;
-	tmp.lval[j>>2] = inl(DE4X5_RRBA); j+=4;
-	tmp.lval[j>>2] = inl(DE4X5_TRBA); j+=4;
-	tmp.lval[j>>2] = inl(DE4X5_STS);  j+=4;
-	tmp.lval[j>>2] = inl(DE4X5_OMR);  j+=4;
-	tmp.lval[j>>2] = inl(DE4X5_IMR);  j+=4;
-	tmp.lval[j>>2] = lp->chipset; j+=4;
-	if (lp->chipset == DC21140) {
-	    tmp.lval[j>>2] = gep_rd(dev);  j+=4;
-	} else {
-	    tmp.lval[j>>2] = inl(DE4X5_SISR); j+=4;
-	    tmp.lval[j>>2] = inl(DE4X5_SICR); j+=4;
-	    tmp.lval[j>>2] = inl(DE4X5_STRR); j+=4;
-	    tmp.lval[j>>2] = inl(DE4X5_SIGR); j+=4;
-	}
-	tmp.lval[j>>2] = lp->phy[lp->active].id; j+=4;
-	if (lp->phy[lp->active].id && (!lp->useSROM || lp->useMII)) {
-	    tmp.lval[j>>2] = lp->active; j+=4;
-	    tmp.lval[j>>2]=mii_rd(MII_CR,lp->phy[lp->active].addr,DE4X5_MII); j+=4;
-	    tmp.lval[j>>2]=mii_rd(MII_SR,lp->phy[lp->active].addr,DE4X5_MII); j+=4;
-	    tmp.lval[j>>2]=mii_rd(MII_ID0,lp->phy[lp->active].addr,DE4X5_MII); j+=4;
-	    tmp.lval[j>>2]=mii_rd(MII_ID1,lp->phy[lp->active].addr,DE4X5_MII); j+=4;
-	    if (lp->phy[lp->active].id != BROADCOM_T4) {
-		tmp.lval[j>>2]=mii_rd(MII_ANA,lp->phy[lp->active].addr,DE4X5_MII); j+=4;
-		tmp.lval[j>>2]=mii_rd(MII_ANLPA,lp->phy[lp->active].addr,DE4X5_MII); j+=4;
-	    }
-	    tmp.lval[j>>2]=mii_rd(0x10,lp->phy[lp->active].addr,DE4X5_MII); j+=4;
-	    if (lp->phy[lp->active].id != BROADCOM_T4) {
-		tmp.lval[j>>2]=mii_rd(0x11,lp->phy[lp->active].addr,DE4X5_MII); j+=4;
-		tmp.lval[j>>2]=mii_rd(0x12,lp->phy[lp->active].addr,DE4X5_MII); j+=4;
-	    } else {
-		tmp.lval[j>>2]=mii_rd(0x14,lp->phy[lp->active].addr,DE4X5_MII); j+=4;
-	    }
-	}
-
-	tmp.addr[j++] = lp->txRingSize;
-	tmp.addr[j++] = netif_queue_stopped(dev);
-
-	ioc->len = j;
-	if (copy_to_user(ioc->data, tmp.addr, ioc->len)) return -EFAULT;
-	break;
-
-*/
-    default:
-	return -EOPNOTSUPP;
-    }
-
-    return status;
-}
-
-static int __init de4x5_module_init (void)
-{
-	int err = 0;
-
-#ifdef CONFIG_PCI
-	err = pci_register_driver(&de4x5_pci_driver);
-#endif
-#ifdef CONFIG_EISA
-	err |= eisa_driver_register (&de4x5_eisa_driver);
-#endif
-
-	return err;
-}
-
-static void __exit de4x5_module_exit (void)
-{
-#ifdef CONFIG_PCI
-	pci_unregister_driver (&de4x5_pci_driver);
-#endif
-#ifdef CONFIG_EISA
-	eisa_driver_unregister (&de4x5_eisa_driver);
-#endif
-}
-
-module_init (de4x5_module_init);
-module_exit (de4x5_module_exit);
diff --git a/drivers/net/ethernet/dec/tulip/de4x5.h b/drivers/net/ethernet/dec/tulip/de4x5.h
deleted file mode 100644
index 1bfdc9b117f6..000000000000
--- a/drivers/net/ethernet/dec/tulip/de4x5.h
+++ /dev/null
@@ -1,1017 +0,0 @@
-/*
-    Copyright 1994 Digital Equipment Corporation.
-
-    This software may be used and distributed according to  the terms of the
-    GNU General Public License, incorporated herein by reference.
-
-    The author may    be  reached as davies@wanton.lkg.dec.com  or   Digital
-    Equipment Corporation, 550 King Street, Littleton MA 01460.
-
-    =========================================================================
-*/
-
-/*
-** DC21040 CSR<1..15> Register Address Map
-*/
-#define DE4X5_BMR    iobase+(0x000 << lp->bus)  /* Bus Mode Register */
-#define DE4X5_TPD    iobase+(0x008 << lp->bus)  /* Transmit Poll Demand Reg */
-#define DE4X5_RPD    iobase+(0x010 << lp->bus)  /* Receive Poll Demand Reg */
-#define DE4X5_RRBA   iobase+(0x018 << lp->bus)  /* RX Ring Base Address Reg */
-#define DE4X5_TRBA   iobase+(0x020 << lp->bus)  /* TX Ring Base Address Reg */
-#define DE4X5_STS    iobase+(0x028 << lp->bus)  /* Status Register */
-#define DE4X5_OMR    iobase+(0x030 << lp->bus)  /* Operation Mode Register */
-#define DE4X5_IMR    iobase+(0x038 << lp->bus)  /* Interrupt Mask Register */
-#define DE4X5_MFC    iobase+(0x040 << lp->bus)  /* Missed Frame Counter */
-#define DE4X5_APROM  iobase+(0x048 << lp->bus)  /* Ethernet Address PROM */
-#define DE4X5_BROM   iobase+(0x048 << lp->bus)  /* Boot ROM Register */
-#define DE4X5_SROM   iobase+(0x048 << lp->bus)  /* Serial ROM Register */
-#define DE4X5_MII    iobase+(0x048 << lp->bus)  /* MII Interface Register */
-#define DE4X5_DDR    iobase+(0x050 << lp->bus)  /* Data Diagnostic Register */
-#define DE4X5_FDR    iobase+(0x058 << lp->bus)  /* Full Duplex Register */
-#define DE4X5_GPT    iobase+(0x058 << lp->bus)  /* General Purpose Timer Reg.*/
-#define DE4X5_GEP    iobase+(0x060 << lp->bus)  /* General Purpose Register */
-#define DE4X5_SISR   iobase+(0x060 << lp->bus)  /* SIA Status Register */
-#define DE4X5_SICR   iobase+(0x068 << lp->bus)  /* SIA Connectivity Register */
-#define DE4X5_STRR   iobase+(0x070 << lp->bus)  /* SIA TX/RX Register */
-#define DE4X5_SIGR   iobase+(0x078 << lp->bus)  /* SIA General Register */
-
-/*
-** EISA Register Address Map
-*/
-#define EISA_ID      iobase+0x0c80   /* EISA ID Registers */
-#define EISA_ID0     iobase+0x0c80   /* EISA ID Register 0 */
-#define EISA_ID1     iobase+0x0c81   /* EISA ID Register 1 */
-#define EISA_ID2     iobase+0x0c82   /* EISA ID Register 2 */
-#define EISA_ID3     iobase+0x0c83   /* EISA ID Register 3 */
-#define EISA_CR      iobase+0x0c84   /* EISA Control Register */
-#define EISA_REG0    iobase+0x0c88   /* EISA Configuration Register 0 */
-#define EISA_REG1    iobase+0x0c89   /* EISA Configuration Register 1 */
-#define EISA_REG2    iobase+0x0c8a   /* EISA Configuration Register 2 */
-#define EISA_REG3    iobase+0x0c8f   /* EISA Configuration Register 3 */
-#define EISA_APROM   iobase+0x0c90   /* Ethernet Address PROM */
-
-/*
-** PCI/EISA Configuration Registers Address Map
-*/
-#define PCI_CFID     iobase+0x0008   /* PCI Configuration ID Register */
-#define PCI_CFCS     iobase+0x000c   /* PCI Command/Status Register */
-#define PCI_CFRV     iobase+0x0018   /* PCI Revision Register */
-#define PCI_CFLT     iobase+0x001c   /* PCI Latency Timer Register */
-#define PCI_CBIO     iobase+0x0028   /* PCI Base I/O Register */
-#define PCI_CBMA     iobase+0x002c   /* PCI Base Memory Address Register */
-#define PCI_CBER     iobase+0x0030   /* PCI Expansion ROM Base Address Reg. */
-#define PCI_CFIT     iobase+0x003c   /* PCI Configuration Interrupt Register */
-#define PCI_CFDA     iobase+0x0040   /* PCI Driver Area Register */
-#define PCI_CFDD     iobase+0x0041   /* PCI Driver Dependent Area Register */
-#define PCI_CFPM     iobase+0x0043   /* PCI Power Management Area Register */
-
-/*
-** EISA Configuration Register 0 bit definitions
-*/
-#define ER0_BSW       0x80           /* EISA Bus Slave Width, 1: 32 bits */
-#define ER0_BMW       0x40           /* EISA Bus Master Width, 1: 32 bits */
-#define ER0_EPT       0x20           /* EISA PREEMPT Time, 0: 23 BCLKs */
-#define ER0_ISTS      0x10           /* Interrupt Status (X) */
-#define ER0_LI        0x08           /* Latch Interrupts */
-#define ER0_INTL      0x06           /* INTerrupt Level */
-#define ER0_INTT      0x01           /* INTerrupt Type, 0: Level, 1: Edge */
-
-/*
-** EISA Configuration Register 1 bit definitions
-*/
-#define ER1_IAM       0xe0           /* ISA Address Mode */
-#define ER1_IAE       0x10           /* ISA Addressing Enable */
-#define ER1_UPIN      0x0f           /* User Pins */
-
-/*
-** EISA Configuration Register 2 bit definitions
-*/
-#define ER2_BRS       0xc0           /* Boot ROM Size */
-#define ER2_BRA       0x3c           /* Boot ROM Address <16:13> */
-
-/*
-** EISA Configuration Register 3 bit definitions
-*/
-#define ER3_BWE       0x40           /* Burst Write Enable */
-#define ER3_BRE       0x04           /* Burst Read Enable */
-#define ER3_LSR       0x02           /* Local Software Reset */
-
-/*
-** PCI Configuration ID Register (PCI_CFID). The Device IDs are left
-** shifted 8 bits to allow detection of DC21142 and DC21143 variants with
-** the configuration revision register step number.
-*/
-#define CFID_DID    0xff00           /* Device ID */
-#define CFID_VID    0x00ff           /* Vendor ID */
-#define DC21040_DID 0x0200           /* Unique Device ID # */
-#define DC21040_VID 0x1011           /* DC21040 Manufacturer */
-#define DC21041_DID 0x1400           /* Unique Device ID # */
-#define DC21041_VID 0x1011           /* DC21041 Manufacturer */
-#define DC21140_DID 0x0900           /* Unique Device ID # */
-#define DC21140_VID 0x1011           /* DC21140 Manufacturer */
-#define DC2114x_DID 0x1900           /* Unique Device ID # */
-#define DC2114x_VID 0x1011           /* DC2114[23] Manufacturer */
-
-/*
-** Chipset defines
-*/
-#define DC21040     DC21040_DID
-#define DC21041     DC21041_DID
-#define DC21140     DC21140_DID
-#define DC2114x     DC2114x_DID
-#define DC21142     (DC2114x_DID | 0x0010)
-#define DC21143     (DC2114x_DID | 0x0030)
-#define DC2114x_BRK 0x0020           /* CFRV break between DC21142 & DC21143 */
-
-#define is_DC21040 ((vendor == DC21040_VID) && (device == DC21040_DID))
-#define is_DC21041 ((vendor == DC21041_VID) && (device == DC21041_DID))
-#define is_DC21140 ((vendor == DC21140_VID) && (device == DC21140_DID))
-#define is_DC2114x ((vendor == DC2114x_VID) && (device == DC2114x_DID))
-#define is_DC21142 ((vendor == DC2114x_VID) && (device == DC21142))
-#define is_DC21143 ((vendor == DC2114x_VID) && (device == DC21143))
-
-/*
-** PCI Configuration Command/Status Register (PCI_CFCS)
-*/
-#define CFCS_DPE    0x80000000       /* Detected Parity Error (S) */
-#define CFCS_SSE    0x40000000       /* Signal System Error   (S) */
-#define CFCS_RMA    0x20000000       /* Receive Master Abort  (S) */
-#define CFCS_RTA    0x10000000       /* Receive Target Abort  (S) */
-#define CFCS_DST    0x06000000       /* DEVSEL Timing         (S) */
-#define CFCS_DPR    0x01000000       /* Data Parity Report    (S) */
-#define CFCS_FBB    0x00800000       /* Fast Back-To-Back     (S) */
-#define CFCS_SEE    0x00000100       /* System Error Enable   (C) */
-#define CFCS_PER    0x00000040       /* Parity Error Response (C) */
-#define CFCS_MO     0x00000004       /* Master Operation      (C) */
-#define CFCS_MSA    0x00000002       /* Memory Space Access   (C) */
-#define CFCS_IOSA   0x00000001       /* I/O Space Access      (C) */
-
-/*
-** PCI Configuration Revision Register (PCI_CFRV)
-*/
-#define CFRV_BC     0xff000000       /* Base Class */
-#define CFRV_SC     0x00ff0000       /* Subclass */
-#define CFRV_RN     0x000000f0       /* Revision Number */
-#define CFRV_SN     0x0000000f       /* Step Number */
-#define BASE_CLASS  0x02000000       /* Indicates Network Controller */
-#define SUB_CLASS   0x00000000       /* Indicates Ethernet Controller */
-#define STEP_NUMBER 0x00000020       /* Increments for future chips */
-#define REV_NUMBER  0x00000003       /* 0x00, 0x01, 0x02, 0x03: Rev in Step */
-#define CFRV_MASK   0xffff0000       /* Register mask */
-
-/*
-** PCI Configuration Latency Timer Register (PCI_CFLT)
-*/
-#define CFLT_BC     0x0000ff00       /* Latency Timer bits */
-
-/*
-** PCI Configuration Base I/O Address Register (PCI_CBIO)
-*/
-#define CBIO_MASK   -128             /* Base I/O Address Mask */
-#define CBIO_IOSI   0x00000001       /* I/O Space Indicator (RO, value is 1) */
-
-/*
-** PCI Configuration Card Information Structure Register (PCI_CCIS)
-*/
-#define CCIS_ROMI   0xf0000000       /* ROM Image */
-#define CCIS_ASO    0x0ffffff8       /* Address Space Offset */
-#define CCIS_ASI    0x00000007       /* Address Space Indicator */
-
-/*
-** PCI Configuration Subsystem ID Register (PCI_SSID)
-*/
-#define SSID_SSID   0xffff0000       /* Subsystem ID */
-#define SSID_SVID   0x0000ffff       /* Subsystem Vendor ID */
-
-/*
-** PCI Configuration Expansion ROM Base Address Register (PCI_CBER)
-*/
-#define CBER_MASK   0xfffffc00       /* Expansion ROM Base Address Mask */
-#define CBER_ROME   0x00000001       /* ROM Enable */
-
-/*
-** PCI Configuration Interrupt Register (PCI_CFIT)
-*/
-#define CFIT_MXLT   0xff000000       /* MAX_LAT Value (0.25us periods) */
-#define CFIT_MNGT   0x00ff0000       /* MIN_GNT Value (0.25us periods) */
-#define CFIT_IRQP   0x0000ff00       /* Interrupt Pin */
-#define CFIT_IRQL   0x000000ff       /* Interrupt Line */
-
-/*
-** PCI Configuration Power Management Area Register (PCI_CFPM)
-*/
-#define SLEEP       0x80             /* Power Saving Sleep Mode */
-#define SNOOZE      0x40             /* Power Saving Snooze Mode */
-#define WAKEUP      0x00             /* Power Saving Wakeup */
-
-#define PCI_CFDA_DSU 0x41            /* 8 bit Configuration Space Address */
-#define PCI_CFDA_PSM 0x43            /* 8 bit Configuration Space Address */
-
-/*
-** DC21040 Bus Mode Register (DE4X5_BMR)
-*/
-#define BMR_RML    0x00200000       /* [Memory] Read Multiple */
-#define BMR_DBO    0x00100000       /* Descriptor Byte Ordering (Endian) */
-#define BMR_TAP    0x000e0000       /* Transmit Automatic Polling */
-#define BMR_DAS    0x00010000       /* Diagnostic Address Space */
-#define BMR_CAL    0x0000c000       /* Cache Alignment */
-#define BMR_PBL    0x00003f00       /* Programmable Burst Length */
-#define BMR_BLE    0x00000080       /* Big/Little Endian */
-#define BMR_DSL    0x0000007c       /* Descriptor Skip Length */
-#define BMR_BAR    0x00000002       /* Bus ARbitration */
-#define BMR_SWR    0x00000001       /* Software Reset */
-
-                                    /* Timings here are for 10BASE-T/AUI only*/
-#define TAP_NOPOLL 0x00000000       /* No automatic polling */
-#define TAP_200US  0x00020000       /* TX automatic polling every 200us */
-#define TAP_800US  0x00040000       /* TX automatic polling every 800us */
-#define TAP_1_6MS  0x00060000       /* TX automatic polling every 1.6ms */
-#define TAP_12_8US 0x00080000       /* TX automatic polling every 12.8us */
-#define TAP_25_6US 0x000a0000       /* TX automatic polling every 25.6us */
-#define TAP_51_2US 0x000c0000       /* TX automatic polling every 51.2us */
-#define TAP_102_4US 0x000e0000      /* TX automatic polling every 102.4us */
-
-#define CAL_NOUSE  0x00000000       /* Not used */
-#define CAL_8LONG  0x00004000       /* 8-longword alignment */
-#define CAL_16LONG 0x00008000       /* 16-longword alignment */
-#define CAL_32LONG 0x0000c000       /* 32-longword alignment */
-
-#define PBL_0      0x00000000       /*  DMA burst length = amount in RX FIFO */
-#define PBL_1      0x00000100       /*  1 longword  DMA burst length */
-#define PBL_2      0x00000200       /*  2 longwords DMA burst length */
-#define PBL_4      0x00000400       /*  4 longwords DMA burst length */
-#define PBL_8      0x00000800       /*  8 longwords DMA burst length */
-#define PBL_16     0x00001000       /* 16 longwords DMA burst length */
-#define PBL_32     0x00002000       /* 32 longwords DMA burst length */
-
-#define DSL_0      0x00000000       /*  0 longword  / descriptor */
-#define DSL_1      0x00000004       /*  1 longword  / descriptor */
-#define DSL_2      0x00000008       /*  2 longwords / descriptor */
-#define DSL_4      0x00000010       /*  4 longwords / descriptor */
-#define DSL_8      0x00000020       /*  8 longwords / descriptor */
-#define DSL_16     0x00000040       /* 16 longwords / descriptor */
-#define DSL_32     0x00000080       /* 32 longwords / descriptor */
-
-/*
-** DC21040 Transmit Poll Demand Register (DE4X5_TPD)
-*/
-#define TPD        0x00000001       /* Transmit Poll Demand */
-
-/*
-** DC21040 Receive Poll Demand Register (DE4X5_RPD)
-*/
-#define RPD        0x00000001       /* Receive Poll Demand */
-
-/*
-** DC21040 Receive Ring Base Address Register (DE4X5_RRBA)
-*/
-#define RRBA       0xfffffffc       /* RX Descriptor List Start Address */
-
-/*
-** DC21040 Transmit Ring Base Address Register (DE4X5_TRBA)
-*/
-#define TRBA       0xfffffffc       /* TX Descriptor List Start Address */
-
-/*
-** Status Register (DE4X5_STS)
-*/
-#define STS_GPI    0x04000000       /* General Purpose Port Interrupt */
-#define STS_BE     0x03800000       /* Bus Error Bits */
-#define STS_TS     0x00700000       /* Transmit Process State */
-#define STS_RS     0x000e0000       /* Receive Process State */
-#define STS_NIS    0x00010000       /* Normal Interrupt Summary */
-#define STS_AIS    0x00008000       /* Abnormal Interrupt Summary */
-#define STS_ER     0x00004000       /* Early Receive */
-#define STS_FBE    0x00002000       /* Fatal Bus Error */
-#define STS_SE     0x00002000       /* System Error */
-#define STS_LNF    0x00001000       /* Link Fail */
-#define STS_FD     0x00000800       /* Full-Duplex Short Frame Received */
-#define STS_TM     0x00000800       /* Timer Expired (DC21041) */
-#define STS_ETI    0x00000400       /* Early Transmit Interrupt */
-#define STS_AT     0x00000400       /* AUI/TP Pin */
-#define STS_RWT    0x00000200       /* Receive Watchdog Time-Out */
-#define STS_RPS    0x00000100       /* Receive Process Stopped */
-#define STS_RU     0x00000080       /* Receive Buffer Unavailable */
-#define STS_RI     0x00000040       /* Receive Interrupt */
-#define STS_UNF    0x00000020       /* Transmit Underflow */
-#define STS_LNP    0x00000010       /* Link Pass */
-#define STS_ANC    0x00000010       /* Autonegotiation Complete */
-#define STS_TJT    0x00000008       /* Transmit Jabber Time-Out */
-#define STS_TU     0x00000004       /* Transmit Buffer Unavailable */
-#define STS_TPS    0x00000002       /* Transmit Process Stopped */
-#define STS_TI     0x00000001       /* Transmit Interrupt */
-
-#define EB_PAR     0x00000000       /* Parity Error */
-#define EB_MA      0x00800000       /* Master Abort */
-#define EB_TA      0x01000000       /* Target Abort */
-#define EB_RES0    0x01800000       /* Reserved */
-#define EB_RES1    0x02000000       /* Reserved */
-
-#define TS_STOP    0x00000000       /* Stopped */
-#define TS_FTD     0x00100000       /* Fetch Transmit Descriptor */
-#define TS_WEOT    0x00200000       /* Wait for End Of Transmission */
-#define TS_QDAT    0x00300000       /* Queue skb data into TX FIFO */
-#define TS_RES     0x00400000       /* Reserved */
-#define TS_SPKT    0x00500000       /* Setup Packet */
-#define TS_SUSP    0x00600000       /* Suspended */
-#define TS_CLTD    0x00700000       /* Close Transmit Descriptor */
-
-#define RS_STOP    0x00000000       /* Stopped */
-#define RS_FRD     0x00020000       /* Fetch Receive Descriptor */
-#define RS_CEOR    0x00040000       /* Check for End of Receive Packet */
-#define RS_WFRP    0x00060000       /* Wait for Receive Packet */
-#define RS_SUSP    0x00080000       /* Suspended */
-#define RS_CLRD    0x000a0000       /* Close Receive Descriptor */
-#define RS_FLUSH   0x000c0000       /* Flush RX FIFO */
-#define RS_QRFS    0x000e0000       /* Queue RX FIFO into RX Skb */
-
-#define INT_CANCEL 0x0001ffff       /* For zeroing all interrupt sources */
-
-/*
-** Operation Mode Register (DE4X5_OMR)
-*/
-#define OMR_SC     0x80000000       /* Special Capture Effect Enable */
-#define OMR_RA     0x40000000       /* Receive All */
-#define OMR_SDP    0x02000000       /* SD Polarity - MUST BE ASSERTED */
-#define OMR_SCR    0x01000000       /* Scrambler Mode */
-#define OMR_PCS    0x00800000       /* PCS Function */
-#define OMR_TTM    0x00400000       /* Transmit Threshold Mode */
-#define OMR_SF     0x00200000       /* Store and Forward */
-#define OMR_HBD    0x00080000       /* HeartBeat Disable */
-#define OMR_PS     0x00040000       /* Port Select */
-#define OMR_CA     0x00020000       /* Capture Effect Enable */
-#define OMR_BP     0x00010000       /* Back Pressure */
-#define OMR_TR     0x0000c000       /* Threshold Control Bits */
-#define OMR_ST     0x00002000       /* Start/Stop Transmission Command */
-#define OMR_FC     0x00001000       /* Force Collision Mode */
-#define OMR_OM     0x00000c00       /* Operating Mode */
-#define OMR_FDX    0x00000200       /* Full Duplex Mode */
-#define OMR_FKD    0x00000100       /* Flaky Oscillator Disable */
-#define OMR_PM     0x00000080       /* Pass All Multicast */
-#define OMR_PR     0x00000040       /* Promiscuous Mode */
-#define OMR_SB     0x00000020       /* Start/Stop Backoff Counter */
-#define OMR_IF     0x00000010       /* Inverse Filtering */
-#define OMR_PB     0x00000008       /* Pass Bad Frames */
-#define OMR_HO     0x00000004       /* Hash Only Filtering Mode */
-#define OMR_SR     0x00000002       /* Start/Stop Receive */
-#define OMR_HP     0x00000001       /* Hash/Perfect Receive Filtering Mode */
-
-#define TR_72      0x00000000       /* Threshold set to 72 (128) bytes */
-#define TR_96      0x00004000       /* Threshold set to 96 (256) bytes */
-#define TR_128     0x00008000       /* Threshold set to 128 (512) bytes */
-#define TR_160     0x0000c000       /* Threshold set to 160 (1024) bytes */
-
-#define OMR_DEF     (OMR_SDP)
-#define OMR_SIA     (OMR_SDP | OMR_TTM)
-#define OMR_SYM     (OMR_SDP | OMR_SCR | OMR_PCS | OMR_HBD | OMR_PS)
-#define OMR_MII_10  (OMR_SDP | OMR_TTM | OMR_PS)
-#define OMR_MII_100 (OMR_SDP | OMR_HBD | OMR_PS)
-
-/*
-** DC21040 Interrupt Mask Register (DE4X5_IMR)
-*/
-#define IMR_GPM    0x04000000       /* General Purpose Port Mask */
-#define IMR_NIM    0x00010000       /* Normal Interrupt Summary Mask */
-#define IMR_AIM    0x00008000       /* Abnormal Interrupt Summary Mask */
-#define IMR_ERM    0x00004000       /* Early Receive Mask */
-#define IMR_FBM    0x00002000       /* Fatal Bus Error Mask */
-#define IMR_SEM    0x00002000       /* System Error Mask */
-#define IMR_LFM    0x00001000       /* Link Fail Mask */
-#define IMR_FDM    0x00000800       /* Full-Duplex (Short Frame) Mask */
-#define IMR_TMM    0x00000800       /* Timer Expired Mask (DC21041) */
-#define IMR_ETM    0x00000400       /* Early Transmit Interrupt Mask */
-#define IMR_ATM    0x00000400       /* AUI/TP Switch Mask */
-#define IMR_RWM    0x00000200       /* Receive Watchdog Time-Out Mask */
-#define IMR_RSM    0x00000100       /* Receive Stopped Mask */
-#define IMR_RUM    0x00000080       /* Receive Buffer Unavailable Mask */
-#define IMR_RIM    0x00000040       /* Receive Interrupt Mask */
-#define IMR_UNM    0x00000020       /* Underflow Interrupt Mask */
-#define IMR_ANM    0x00000010       /* Autonegotiation Complete Mask */
-#define IMR_LPM    0x00000010       /* Link Pass */
-#define IMR_TJM    0x00000008       /* Transmit Time-Out Jabber Mask */
-#define IMR_TUM    0x00000004       /* Transmit Buffer Unavailable Mask */
-#define IMR_TSM    0x00000002       /* Transmission Stopped Mask */
-#define IMR_TIM    0x00000001       /* Transmit Interrupt Mask */
-
-/*
-** Missed Frames and FIFO Overflow Counters (DE4X5_MFC)
-*/
-#define MFC_FOCO   0x10000000       /* FIFO Overflow Counter Overflow Bit */
-#define MFC_FOC    0x0ffe0000       /* FIFO Overflow Counter Bits */
-#define MFC_OVFL   0x00010000       /* Missed Frames Counter Overflow Bit */
-#define MFC_CNTR   0x0000ffff       /* Missed Frames Counter Bits */
-#define MFC_FOCM   0x1ffe0000       /* FIFO Overflow Counter Mask */
-
-/*
-** DC21040 Ethernet Address PROM (DE4X5_APROM)
-*/
-#define APROM_DN   0x80000000       /* Data Not Valid */
-#define APROM_DT   0x000000ff       /* Address Byte */
-
-/*
-** DC21041 Boot/Ethernet Address ROM (DE4X5_BROM)
-*/
-#define BROM_MODE 0x00008000       /* MODE_1: 0,  MODE_0: 1  (read only) */
-#define BROM_RD   0x00004000       /* Read from Boot ROM */
-#define BROM_WR   0x00002000       /* Write to Boot ROM */
-#define BROM_BR   0x00001000       /* Select Boot ROM when set */
-#define BROM_SR   0x00000800       /* Select Serial ROM when set */
-#define BROM_REG  0x00000400       /* External Register Select */
-#define BROM_DT   0x000000ff       /* Data Byte */
-
-/*
-** DC21041 Serial/Ethernet Address ROM (DE4X5_SROM, DE4X5_MII)
-*/
-#define MII_MDI   0x00080000       /* MII Management Data In */
-#define MII_MDO   0x00060000       /* MII Management Mode/Data Out */
-#define MII_MRD   0x00040000       /* MII Management Define Read Mode */
-#define MII_MWR   0x00000000       /* MII Management Define Write Mode */
-#define MII_MDT   0x00020000       /* MII Management Data Out */
-#define MII_MDC   0x00010000       /* MII Management Clock */
-#define MII_RD    0x00004000       /* Read from MII */
-#define MII_WR    0x00002000       /* Write to MII */
-#define MII_SEL   0x00000800       /* Select MII when RESET */
-
-#define SROM_MODE 0x00008000       /* MODE_1: 0,  MODE_0: 1  (read only) */
-#define SROM_RD   0x00004000       /* Read from Boot ROM */
-#define SROM_WR   0x00002000       /* Write to Boot ROM */
-#define SROM_BR   0x00001000       /* Select Boot ROM when set */
-#define SROM_SR   0x00000800       /* Select Serial ROM when set */
-#define SROM_REG  0x00000400       /* External Register Select */
-#define SROM_DT   0x000000ff       /* Data Byte */
-
-#define DT_OUT    0x00000008       /* Serial Data Out */
-#define DT_IN     0x00000004       /* Serial Data In */
-#define DT_CLK    0x00000002       /* Serial ROM Clock */
-#define DT_CS     0x00000001       /* Serial ROM Chip Select */
-
-#define MII_PREAMBLE 0xffffffff    /* MII Management Preamble */
-#define MII_TEST     0xaaaaaaaa    /* MII Test Signal */
-#define MII_STRD     0x06          /* Start of Frame+Op Code: use low nibble */
-#define MII_STWR     0x0a          /* Start of Frame+Op Code: use low nibble */
-
-#define MII_CR       0x00          /* MII Management Control Register */
-#define MII_SR       0x01          /* MII Management Status Register */
-#define MII_ID0      0x02          /* PHY Identifier Register 0 */
-#define MII_ID1      0x03          /* PHY Identifier Register 1 */
-#define MII_ANA      0x04          /* Auto Negotiation Advertisement */
-#define MII_ANLPA    0x05          /* Auto Negotiation Link Partner Ability */
-#define MII_ANE      0x06          /* Auto Negotiation Expansion */
-#define MII_ANP      0x07          /* Auto Negotiation Next Page TX */
-
-#define DE4X5_MAX_MII 32           /* Maximum address of MII PHY devices */
-
-/*
-** MII Management Control Register
-*/
-#define MII_CR_RST  0x8000         /* RESET the PHY chip */
-#define MII_CR_LPBK 0x4000         /* Loopback enable */
-#define MII_CR_SPD  0x2000         /* 0: 10Mb/s; 1: 100Mb/s */
-#define MII_CR_10   0x0000         /* Set 10Mb/s */
-#define MII_CR_100  0x2000         /* Set 100Mb/s */
-#define MII_CR_ASSE 0x1000         /* Auto Speed Select Enable */
-#define MII_CR_PD   0x0800         /* Power Down */
-#define MII_CR_ISOL 0x0400         /* Isolate Mode */
-#define MII_CR_RAN  0x0200         /* Restart Auto Negotiation */
-#define MII_CR_FDM  0x0100         /* Full Duplex Mode */
-#define MII_CR_CTE  0x0080         /* Collision Test Enable */
-
-/*
-** MII Management Status Register
-*/
-#define MII_SR_T4C  0x8000         /* 100BASE-T4 capable */
-#define MII_SR_TXFD 0x4000         /* 100BASE-TX Full Duplex capable */
-#define MII_SR_TXHD 0x2000         /* 100BASE-TX Half Duplex capable */
-#define MII_SR_TFD  0x1000         /* 10BASE-T Full Duplex capable */
-#define MII_SR_THD  0x0800         /* 10BASE-T Half Duplex capable */
-#define MII_SR_ASSC 0x0020         /* Auto Speed Selection Complete*/
-#define MII_SR_RFD  0x0010         /* Remote Fault Detected */
-#define MII_SR_ANC  0x0008         /* Auto Negotiation capable */
-#define MII_SR_LKS  0x0004         /* Link Status */
-#define MII_SR_JABD 0x0002         /* Jabber Detect */
-#define MII_SR_XC   0x0001         /* Extended Capabilities */
-
-/*
-** MII Management Auto Negotiation Advertisement Register
-*/
-#define MII_ANA_TAF  0x03e0        /* Technology Ability Field */
-#define MII_ANA_T4AM 0x0200        /* T4 Technology Ability Mask */
-#define MII_ANA_TXAM 0x0180        /* TX Technology Ability Mask */
-#define MII_ANA_FDAM 0x0140        /* Full Duplex Technology Ability Mask */
-#define MII_ANA_HDAM 0x02a0        /* Half Duplex Technology Ability Mask */
-#define MII_ANA_100M 0x0380        /* 100Mb Technology Ability Mask */
-#define MII_ANA_10M  0x0060        /* 10Mb Technology Ability Mask */
-#define MII_ANA_CSMA 0x0001        /* CSMA-CD Capable */
-
-/*
-** MII Management Auto Negotiation Remote End Register
-*/
-#define MII_ANLPA_NP   0x8000      /* Next Page (Enable) */
-#define MII_ANLPA_ACK  0x4000      /* Remote Acknowledge */
-#define MII_ANLPA_RF   0x2000      /* Remote Fault */
-#define MII_ANLPA_TAF  0x03e0      /* Technology Ability Field */
-#define MII_ANLPA_T4AM 0x0200      /* T4 Technology Ability Mask */
-#define MII_ANLPA_TXAM 0x0180      /* TX Technology Ability Mask */
-#define MII_ANLPA_FDAM 0x0140      /* Full Duplex Technology Ability Mask */
-#define MII_ANLPA_HDAM 0x02a0      /* Half Duplex Technology Ability Mask */
-#define MII_ANLPA_100M 0x0380      /* 100Mb Technology Ability Mask */
-#define MII_ANLPA_10M  0x0060      /* 10Mb Technology Ability Mask */
-#define MII_ANLPA_CSMA 0x0001      /* CSMA-CD Capable */
-
-/*
-** SROM Media Definitions (ABG SROM Section)
-*/
-#define MEDIA_NWAY     0x0080      /* Nway (Auto Negotiation) on PHY */
-#define MEDIA_MII      0x0040      /* MII Present on the adapter */
-#define MEDIA_FIBRE    0x0008      /* Fibre Media present */
-#define MEDIA_AUI      0x0004      /* AUI Media present */
-#define MEDIA_TP       0x0002      /* TP Media present */
-#define MEDIA_BNC      0x0001      /* BNC Media present */
-
-/*
-** SROM Definitions (Digital Semiconductor Format)
-*/
-#define SROM_SSVID     0x0000      /* Sub-system Vendor ID offset */
-#define SROM_SSID      0x0002      /* Sub-system ID offset */
-#define SROM_CISPL     0x0004      /* CardBus CIS Pointer low offset */
-#define SROM_CISPH     0x0006      /* CardBus CIS Pointer high offset */
-#define SROM_IDCRC     0x0010      /* ID Block CRC offset*/
-#define SROM_RSVD2     0x0011      /* ID Reserved 2 offset */
-#define SROM_SFV       0x0012      /* SROM Format Version offset */
-#define SROM_CCNT      0x0013      /* Controller Count offset */
-#define SROM_HWADD     0x0014      /* Hardware Address offset */
-#define SROM_MRSVD     0x007c      /* Manufacturer Reserved offset*/
-#define SROM_CRC       0x007e      /* SROM CRC offset */
-
-/*
-** SROM Media Connection Definitions
-*/
-#define SROM_10BT      0x0000      /*  10BASE-T half duplex */
-#define SROM_10BTN     0x0100      /*  10BASE-T with Nway */
-#define SROM_10BTF     0x0204      /*  10BASE-T full duplex */
-#define SROM_10BTNLP   0x0400      /*  10BASE-T without Link Pass test */
-#define SROM_10B2      0x0001      /*  10BASE-2 (BNC) */
-#define SROM_10B5      0x0002      /*  10BASE-5 (AUI) */
-#define SROM_100BTH    0x0003      /*  100BASE-T half duplex */
-#define SROM_100BTF    0x0205      /*  100BASE-T full duplex */
-#define SROM_100BT4    0x0006      /*  100BASE-T4 */
-#define SROM_100BFX    0x0007      /*  100BASE-FX half duplex (Fiber) */
-#define SROM_M10BT     0x0009      /*  MII 10BASE-T half duplex */
-#define SROM_M10BTF    0x020a      /*  MII 10BASE-T full duplex */
-#define SROM_M100BT    0x000d      /*  MII 100BASE-T half duplex */
-#define SROM_M100BTF   0x020e      /*  MII 100BASE-T full duplex */
-#define SROM_M100BT4   0x000f      /*  MII 100BASE-T4 */
-#define SROM_M100BF    0x0010      /*  MII 100BASE-FX half duplex */
-#define SROM_M100BFF   0x0211      /*  MII 100BASE-FX full duplex */
-#define SROM_PDA       0x0800      /*  Powerup & Dynamic Autosense */
-#define SROM_PAO       0x8800      /*  Powerup Autosense Only */
-#define SROM_NSMI      0xffff      /*  No Selected Media Information */
-
-/*
-** SROM Media Definitions
-*/
-#define SROM_10BASET   0x0000      /*  10BASE-T half duplex */
-#define SROM_10BASE2   0x0001      /*  10BASE-2 (BNC) */
-#define SROM_10BASE5   0x0002      /*  10BASE-5 (AUI) */
-#define SROM_100BASET  0x0003      /*  100BASE-T half duplex */
-#define SROM_10BASETF  0x0004      /*  10BASE-T full duplex */
-#define SROM_100BASETF 0x0005      /*  100BASE-T full duplex */
-#define SROM_100BASET4 0x0006      /*  100BASE-T4 */
-#define SROM_100BASEF  0x0007      /*  100BASE-FX half duplex */
-#define SROM_100BASEFF 0x0008      /*  100BASE-FX full duplex */
-
-#define BLOCK_LEN      0x7f        /* Extended blocks length mask */
-#define EXT_FIELD      0x40        /* Extended blocks extension field bit */
-#define MEDIA_CODE     0x3f        /* Extended blocks media code mask */
-
-/*
-** SROM Compact Format Block Masks
-*/
-#define COMPACT_FI      0x80       /* Format Indicator */
-#define COMPACT_LEN     0x04       /* Length */
-#define COMPACT_MC      0x3f       /* Media Code */
-
-/*
-** SROM Extended Format Block Type 0 Masks
-*/
-#define BLOCK0_FI      0x80        /* Format Indicator */
-#define BLOCK0_MCS     0x80        /* Media Code byte Sign */
-#define BLOCK0_MC      0x3f        /* Media Code */
-
-/*
-** DC21040 Full Duplex Register (DE4X5_FDR)
-*/
-#define FDR_FDACV  0x0000ffff      /* Full Duplex Auto Configuration Value */
-
-/*
-** DC21041 General Purpose Timer Register (DE4X5_GPT)
-*/
-#define GPT_CON  0x00010000        /* One shot: 0,  Continuous: 1 */
-#define GPT_VAL  0x0000ffff        /* Timer Value */
-
-/*
-** DC21140 General Purpose Register (DE4X5_GEP) (hardware dependent bits)
-*/
-/* Valid ONLY for DE500 hardware */
-#define GEP_LNP  0x00000080        /* Link Pass               (input)        */
-#define GEP_SLNK 0x00000040        /* SYM LINK                (input)        */
-#define GEP_SDET 0x00000020        /* Signal Detect           (input)        */
-#define GEP_HRST 0x00000010        /* Hard RESET (to PHY)     (output)       */
-#define GEP_FDXD 0x00000008        /* Full Duplex Disable     (output)       */
-#define GEP_PHYL 0x00000004        /* PHY Loopback            (output)       */
-#define GEP_FLED 0x00000002        /* Force Activity LED on   (output)       */
-#define GEP_MODE 0x00000001        /* 0: 10Mb/s,  1: 100Mb/s                 */
-#define GEP_INIT 0x0000011f        /* Setup inputs (0) and outputs (1)       */
-#define GEP_CTRL 0x00000100        /* GEP control bit                        */
-
-/*
-** SIA Register Defaults
-*/
-#define CSR13 0x00000001
-#define CSR14 0x0003ff7f           /* Autonegotiation disabled               */
-#define CSR15 0x00000008
-
-/*
-** SIA Status Register (DE4X5_SISR)
-*/
-#define SISR_LPC   0xffff0000      /* Link Partner's Code Word               */
-#define SISR_LPN   0x00008000      /* Link Partner Negotiable                */
-#define SISR_ANS   0x00007000      /* Auto Negotiation Arbitration State     */
-#define SISR_NSN   0x00000800      /* Non Stable NLPs Detected (DC21041)     */
-#define SISR_TRF   0x00000800      /* Transmit Remote Fault                  */
-#define SISR_NSND  0x00000400      /* Non Stable NLPs Detected (DC21142)     */
-#define SISR_ANR_FDS 0x00000400    /* Auto Negotiate Restart/Full Duplex Sel.*/
-#define SISR_TRA   0x00000200      /* 10BASE-T Receive Port Activity         */
-#define SISR_NRA   0x00000200      /* Non Selected Port Receive Activity     */
-#define SISR_ARA   0x00000100      /* AUI Receive Port Activity              */
-#define SISR_SRA   0x00000100      /* Selected Port Receive Activity         */
-#define SISR_DAO   0x00000080      /* PLL All One                            */
-#define SISR_DAZ   0x00000040      /* PLL All Zero                           */
-#define SISR_DSP   0x00000020      /* PLL Self-Test Pass                     */
-#define SISR_DSD   0x00000010      /* PLL Self-Test Done                     */
-#define SISR_APS   0x00000008      /* Auto Polarity State                    */
-#define SISR_LKF   0x00000004      /* Link Fail Status                       */
-#define SISR_LS10  0x00000004      /* 10Mb/s Link Fail Status                */
-#define SISR_NCR   0x00000002      /* Network Connection Error               */
-#define SISR_LS100 0x00000002      /* 100Mb/s Link Fail Status               */
-#define SISR_PAUI  0x00000001      /* AUI_TP Indication                      */
-#define SISR_MRA   0x00000001      /* MII Receive Port Activity              */
-
-#define ANS_NDIS   0x00000000      /* Nway disable                           */
-#define ANS_TDIS   0x00001000      /* Transmit Disable                       */
-#define ANS_ADET   0x00002000      /* Ability Detect                         */
-#define ANS_ACK    0x00003000      /* Acknowledge                            */
-#define ANS_CACK   0x00004000      /* Complete Acknowledge                   */
-#define ANS_NWOK   0x00005000      /* Nway OK - FLP Link Good                */
-#define ANS_LCHK   0x00006000      /* Link Check                             */
-
-#define SISR_RST   0x00000301      /* CSR12 reset                            */
-#define SISR_ANR   0x00001301      /* Autonegotiation restart                */
-
-/*
-** SIA Connectivity Register (DE4X5_SICR)
-*/
-#define SICR_SDM   0xffff0000       /* SIA Diagnostics Mode */
-#define SICR_OE57  0x00008000       /* Output Enable 5 6 7 */
-#define SICR_OE24  0x00004000       /* Output Enable 2 4 */
-#define SICR_OE13  0x00002000       /* Output Enable 1 3 */
-#define SICR_IE    0x00001000       /* Input Enable */
-#define SICR_EXT   0x00000000       /* SIA MUX Select External SIA Mode */
-#define SICR_D_SIA 0x00000400       /* SIA MUX Select Diagnostics - SIA Sigs */
-#define SICR_DPLL  0x00000800       /* SIA MUX Select Diagnostics - DPLL Sigs*/
-#define SICR_APLL  0x00000a00       /* SIA MUX Select Diagnostics - DPLL Sigs*/
-#define SICR_D_RxM 0x00000c00       /* SIA MUX Select Diagnostics - RxM Sigs */
-#define SICR_M_RxM 0x00000d00       /* SIA MUX Select Diagnostics - RxM Sigs */
-#define SICR_LNKT  0x00000e00       /* SIA MUX Select Diagnostics - Link Test*/
-#define SICR_SEL   0x00000f00       /* SIA MUX Select AUI or TP with LEDs */
-#define SICR_ASE   0x00000080       /* APLL Start Enable*/
-#define SICR_SIM   0x00000040       /* Serial Interface Input Multiplexer */
-#define SICR_ENI   0x00000020       /* Encoder Input Multiplexer */
-#define SICR_EDP   0x00000010       /* SIA PLL External Input Enable */
-#define SICR_AUI   0x00000008       /* 10Base-T (0) or AUI (1) */
-#define SICR_CAC   0x00000004       /* CSR Auto Configuration */
-#define SICR_PS    0x00000002       /* Pin AUI/TP Selection */
-#define SICR_SRL   0x00000001       /* SIA Reset */
-#define SIA_RESET  0x00000000       /* SIA Reset Value */
-
-/*
-** SIA Transmit and Receive Register (DE4X5_STRR)
-*/
-#define STRR_TAS   0x00008000       /* 10Base-T/AUI Autosensing Enable */
-#define STRR_SPP   0x00004000       /* Set Polarity Plus */
-#define STRR_APE   0x00002000       /* Auto Polarity Enable */
-#define STRR_LTE   0x00001000       /* Link Test Enable */
-#define STRR_SQE   0x00000800       /* Signal Quality Enable */
-#define STRR_CLD   0x00000400       /* Collision Detect Enable */
-#define STRR_CSQ   0x00000200       /* Collision Squelch Enable */
-#define STRR_RSQ   0x00000100       /* Receive Squelch Enable */
-#define STRR_ANE   0x00000080       /* Auto Negotiate Enable */
-#define STRR_HDE   0x00000040       /* Half Duplex Enable */
-#define STRR_CPEN  0x00000030       /* Compensation Enable */
-#define STRR_LSE   0x00000008       /* Link Pulse Send Enable */
-#define STRR_DREN  0x00000004       /* Driver Enable */
-#define STRR_LBK   0x00000002       /* Loopback Enable */
-#define STRR_ECEN  0x00000001       /* Encoder Enable */
-#define STRR_RESET 0xffffffff       /* Reset value for STRR */
-
-/*
-** SIA General Register (DE4X5_SIGR)
-*/
-#define SIGR_RMI   0x40000000       /* Receive Match Interrupt */
-#define SIGR_GI1   0x20000000       /* General Port Interrupt 1 */
-#define SIGR_GI0   0x10000000       /* General Port Interrupt 0 */
-#define SIGR_CWE   0x08000000       /* Control Write Enable */
-#define SIGR_RME   0x04000000       /* Receive Match Enable */
-#define SIGR_GEI1  0x02000000       /* GEP Interrupt Enable on Port 1 */
-#define SIGR_GEI0  0x01000000       /* GEP Interrupt Enable on Port 0 */
-#define SIGR_LGS3  0x00800000       /* LED/GEP3 Select */
-#define SIGR_LGS2  0x00400000       /* LED/GEP2 Select */
-#define SIGR_LGS1  0x00200000       /* LED/GEP1 Select */
-#define SIGR_LGS0  0x00100000       /* LED/GEP0 Select */
-#define SIGR_MD    0x000f0000       /* General Purpose Mode and Data */
-#define SIGR_LV2   0x00008000       /* General Purpose LED2 value */
-#define SIGR_LE2   0x00004000       /* General Purpose LED2 enable */
-#define SIGR_FRL   0x00002000       /* Force Receiver Low */
-#define SIGR_DPST  0x00001000       /* PLL Self Test Start */
-#define SIGR_LSD   0x00000800       /* LED Stretch Disable */
-#define SIGR_FLF   0x00000400       /* Force Link Fail */
-#define SIGR_FUSQ  0x00000200       /* Force Unsquelch */
-#define SIGR_TSCK  0x00000100       /* Test Clock */
-#define SIGR_LV1   0x00000080       /* General Purpose LED1 value */
-#define SIGR_LE1   0x00000040       /* General Purpose LED1 enable */
-#define SIGR_RWR   0x00000020       /* Receive Watchdog Release */
-#define SIGR_RWD   0x00000010       /* Receive Watchdog Disable */
-#define SIGR_ABM   0x00000008       /* BNC: 0,  AUI:1 */
-#define SIGR_JCK   0x00000004       /* Jabber Clock */
-#define SIGR_HUJ   0x00000002       /* Host Unjab */
-#define SIGR_JBD   0x00000001       /* Jabber Disable */
-#define SIGR_RESET 0xffff0000       /* Reset value for SIGR */
-
-/*
-** Receive Descriptor Bit Summary
-*/
-#define R_OWN      0x80000000       /* Own Bit */
-#define RD_FF      0x40000000       /* Filtering Fail */
-#define RD_FL      0x3fff0000       /* Frame Length */
-#define RD_ES      0x00008000       /* Error Summary */
-#define RD_LE      0x00004000       /* Length Error */
-#define RD_DT      0x00003000       /* Data Type */
-#define RD_RF      0x00000800       /* Runt Frame */
-#define RD_MF      0x00000400       /* Multicast Frame */
-#define RD_FS      0x00000200       /* First Descriptor */
-#define RD_LS      0x00000100       /* Last Descriptor */
-#define RD_TL      0x00000080       /* Frame Too Long */
-#define RD_CS      0x00000040       /* Collision Seen */
-#define RD_FT      0x00000020       /* Frame Type */
-#define RD_RJ      0x00000010       /* Receive Watchdog */
-#define RD_RE      0x00000008       /* Report on MII Error */
-#define RD_DB      0x00000004       /* Dribbling Bit */
-#define RD_CE      0x00000002       /* CRC Error */
-#define RD_OF      0x00000001       /* Overflow */
-
-#define RD_RER     0x02000000       /* Receive End Of Ring */
-#define RD_RCH     0x01000000       /* Second Address Chained */
-#define RD_RBS2    0x003ff800       /* Buffer 2 Size */
-#define RD_RBS1    0x000007ff       /* Buffer 1 Size */
-
-/*
-** Transmit Descriptor Bit Summary
-*/
-#define T_OWN      0x80000000       /* Own Bit */
-#define TD_ES      0x00008000       /* Error Summary */
-#define TD_TO      0x00004000       /* Transmit Jabber Time-Out */
-#define TD_LO      0x00000800       /* Loss Of Carrier */
-#define TD_NC      0x00000400       /* No Carrier */
-#define TD_LC      0x00000200       /* Late Collision */
-#define TD_EC      0x00000100       /* Excessive Collisions */
-#define TD_HF      0x00000080       /* Heartbeat Fail */
-#define TD_CC      0x00000078       /* Collision Counter */
-#define TD_LF      0x00000004       /* Link Fail */
-#define TD_UF      0x00000002       /* Underflow Error */
-#define TD_DE      0x00000001       /* Deferred */
-
-#define TD_IC      0x80000000       /* Interrupt On Completion */
-#define TD_LS      0x40000000       /* Last Segment */
-#define TD_FS      0x20000000       /* First Segment */
-#define TD_FT1     0x10000000       /* Filtering Type */
-#define TD_SET     0x08000000       /* Setup Packet */
-#define TD_AC      0x04000000       /* Add CRC Disable */
-#define TD_TER     0x02000000       /* Transmit End Of Ring */
-#define TD_TCH     0x01000000       /* Second Address Chained */
-#define TD_DPD     0x00800000       /* Disabled Padding */
-#define TD_FT0     0x00400000       /* Filtering Type */
-#define TD_TBS2    0x003ff800       /* Buffer 2 Size */
-#define TD_TBS1    0x000007ff       /* Buffer 1 Size */
-
-#define PERFECT_F  0x00000000
-#define HASH_F     TD_FT0
-#define INVERSE_F  TD_FT1
-#define HASH_O_F   (TD_FT1 | TD_F0)
-
-/*
-** Media / mode state machine definitions
-** User selectable:
-*/
-#define TP              0x0040     /* 10Base-T (now equiv to _10Mb)        */
-#define TP_NW           0x0002     /* 10Base-T with Nway                   */
-#define BNC             0x0004     /* Thinwire                             */
-#define AUI             0x0008     /* Thickwire                            */
-#define BNC_AUI         0x0010     /* BNC/AUI on DC21040 indistinguishable */
-#define _10Mb           0x0040     /* 10Mb/s Ethernet                      */
-#define _100Mb          0x0080     /* 100Mb/s Ethernet                     */
-#define AUTO            0x4000     /* Auto sense the media or speed        */
-
-/*
-** Internal states
-*/
-#define NC              0x0000     /* No Connection                        */
-#define ANS             0x0020     /* Intermediate AutoNegotiation State   */
-#define SPD_DET         0x0100     /* Parallel speed detection             */
-#define INIT            0x0200     /* Initial state                        */
-#define EXT_SIA         0x0400     /* External SIA for motherboard chip    */
-#define ANS_SUSPECT     0x0802     /* Suspect the ANS (TP) port is down    */
-#define TP_SUSPECT      0x0803     /* Suspect the TP port is down          */
-#define BNC_AUI_SUSPECT 0x0804     /* Suspect the BNC or AUI port is down  */
-#define EXT_SIA_SUSPECT 0x0805     /* Suspect the EXT SIA port is down     */
-#define BNC_SUSPECT     0x0806     /* Suspect the BNC port is down         */
-#define AUI_SUSPECT     0x0807     /* Suspect the AUI port is down         */
-#define MII             0x1000     /* MII on the 21143                     */
-
-#define TIMER_CB        0x80000000 /* Timer callback detection             */
-
-/*
-** DE4X5 DEBUG Options
-*/
-#define DEBUG_NONE      0x0000     /* No DEBUG messages */
-#define DEBUG_VERSION   0x0001     /* Print version message */
-#define DEBUG_MEDIA     0x0002     /* Print media messages */
-#define DEBUG_TX        0x0004     /* Print TX (queue_pkt) messages */
-#define DEBUG_RX        0x0008     /* Print RX (de4x5_rx) messages */
-#define DEBUG_SROM      0x0010     /* Print SROM messages */
-#define DEBUG_MII       0x0020     /* Print MII messages */
-#define DEBUG_OPEN      0x0040     /* Print de4x5_open() messages */
-#define DEBUG_CLOSE     0x0080     /* Print de4x5_close() messages */
-#define DEBUG_PCICFG    0x0100
-#define DEBUG_ALL       0x01ff
-
-/*
-** Miscellaneous
-*/
-#define PCI  0
-#define EISA 1
-
-#define DE4X5_HASH_TABLE_LEN   512       /* Bits */
-#define DE4X5_HASH_BITS        0x01ff    /* 9 LS bits */
-
-#define SETUP_FRAME_LEN  192       /* Bytes */
-#define IMPERF_PA_OFFSET 156       /* Bytes */
-
-#define POLL_DEMAND          1
-
-#define LOST_MEDIA_THRESHOLD 3
-
-#define MASK_INTERRUPTS      1
-#define UNMASK_INTERRUPTS    0
-
-#define DE4X5_STRLEN         8
-
-#define DE4X5_INIT           0     /* Initialisation time */
-#define DE4X5_RUN            1     /* Run time */
-
-#define DE4X5_SAVE_STATE     0
-#define DE4X5_RESTORE_STATE  1
-
-/*
-** Address Filtering Modes
-*/
-#define PERFECT              0     /* 16 perfect physical addresses */
-#define HASH_PERF            1     /* 1 perfect, 512 multicast addresses */
-#define PERFECT_REJ          2     /* Reject 16 perfect physical addresses */
-#define ALL_HASH             3     /* Hashes all physical & multicast addrs */
-
-#define ALL                  0     /* Clear out all the setup frame */
-#define PHYS_ADDR_ONLY       1     /* Update the physical address only */
-
-/*
-** Adapter state
-*/
-#define INITIALISED          0     /* After h/w initialised and mem alloc'd */
-#define CLOSED               1     /* Ready for opening */
-#define OPEN                 2     /* Running */
-
-/*
-** Various wait times
-*/
-#define PDET_LINK_WAIT    1200    /* msecs to wait for link detect bits     */
-#define ANS_FINISH_WAIT   1000    /* msecs to wait for link detect bits     */
-
-/*
-** IEEE OUIs for various PHY vendor/chip combos - Reg 2 values only. Since
-** the vendors seem split 50-50 on how to calculate the OUI register values
-** anyway, just reading Reg2 seems reasonable for now [see de4x5_get_oui()].
-*/
-#define NATIONAL_TX 0x2000
-#define BROADCOM_T4 0x03e0
-#define SEEQ_T4     0x0016
-#define CYPRESS_T4  0x0014
-
-/*
-** Speed Selection stuff
-*/
-#define SET_10Mb {\
-  if ((lp->phy[lp->active].id) && (!lp->useSROM || lp->useMII)) {\
-    omr = inl(DE4X5_OMR) & ~(OMR_TTM | OMR_PCS | OMR_SCR | OMR_FDX);\
-    if ((lp->tmp != MII_SR_ASSC) || (lp->autosense != AUTO)) {\
-      mii_wr(MII_CR_10|(lp->fdx?MII_CR_FDM:0), MII_CR, lp->phy[lp->active].addr, DE4X5_MII);\
-    }\
-    omr |= ((lp->fdx ? OMR_FDX : 0) | OMR_TTM);\
-    outl(omr, DE4X5_OMR);\
-    if (!lp->useSROM) lp->cache.gep = 0;\
-  } else if (lp->useSROM && !lp->useMII) {\
-    omr = (inl(DE4X5_OMR) & ~(OMR_PS | OMR_HBD | OMR_TTM | OMR_PCS | OMR_SCR | OMR_FDX));\
-    omr |= (lp->fdx ? OMR_FDX : 0);\
-    outl(omr | (lp->infoblock_csr6 & ~(OMR_SCR | OMR_HBD)), DE4X5_OMR);\
-  } else {\
-    omr = (inl(DE4X5_OMR) & ~(OMR_PS | OMR_HBD | OMR_TTM | OMR_PCS | OMR_SCR | OMR_FDX));\
-    omr |= (lp->fdx ? OMR_FDX : 0);\
-    outl(omr | OMR_SDP | OMR_TTM, DE4X5_OMR);\
-    lp->cache.gep = (lp->fdx ? 0 : GEP_FDXD);\
-    gep_wr(lp->cache.gep, dev);\
-  }\
-}
-
-#define SET_100Mb {\
-  if ((lp->phy[lp->active].id) && (!lp->useSROM || lp->useMII)) {\
-    int fdx=0;\
-    if (lp->phy[lp->active].id == NATIONAL_TX) {\
-        mii_wr(mii_rd(0x18, lp->phy[lp->active].addr, DE4X5_MII) & ~0x2000,\
-                      0x18, lp->phy[lp->active].addr, DE4X5_MII);\
-    }\
-    omr = inl(DE4X5_OMR) & ~(OMR_TTM | OMR_PCS | OMR_SCR | OMR_FDX);\
-    sr = mii_rd(MII_SR, lp->phy[lp->active].addr, DE4X5_MII);\
-    if (!(sr & MII_ANA_T4AM) && lp->fdx) fdx=1;\
-    if ((lp->tmp != MII_SR_ASSC) || (lp->autosense != AUTO)) {\
-      mii_wr(MII_CR_100|(fdx?MII_CR_FDM:0), MII_CR, lp->phy[lp->active].addr, DE4X5_MII);\
-    }\
-    if (fdx) omr |= OMR_FDX;\
-    outl(omr, DE4X5_OMR);\
-    if (!lp->useSROM) lp->cache.gep = 0;\
-  } else if (lp->useSROM && !lp->useMII) {\
-    omr = (inl(DE4X5_OMR) & ~(OMR_PS | OMR_HBD | OMR_TTM | OMR_PCS | OMR_SCR | OMR_FDX));\
-    omr |= (lp->fdx ? OMR_FDX : 0);\
-    outl(omr | lp->infoblock_csr6, DE4X5_OMR);\
-  } else {\
-    omr = (inl(DE4X5_OMR) & ~(OMR_PS | OMR_HBD | OMR_TTM | OMR_PCS | OMR_SCR | OMR_FDX));\
-    omr |= (lp->fdx ? OMR_FDX : 0);\
-    outl(omr | OMR_SDP | OMR_PS | OMR_HBD | OMR_PCS | OMR_SCR, DE4X5_OMR);\
-    lp->cache.gep = (lp->fdx ? 0 : GEP_FDXD) | GEP_MODE;\
-    gep_wr(lp->cache.gep, dev);\
-  }\
-}
-
-/* FIX ME so I don't jam 10Mb networks */
-#define SET_100Mb_PDET {\
-  if ((lp->phy[lp->active].id) && (!lp->useSROM || lp->useMII)) {\
-    mii_wr(MII_CR_100|MII_CR_ASSE, MII_CR, lp->phy[lp->active].addr, DE4X5_MII);\
-    omr = (inl(DE4X5_OMR) & ~(OMR_TTM | OMR_PCS | OMR_SCR | OMR_FDX));\
-    outl(omr, DE4X5_OMR);\
-  } else if (lp->useSROM && !lp->useMII) {\
-    omr = (inl(DE4X5_OMR) & ~(OMR_TTM | OMR_PCS | OMR_SCR | OMR_FDX));\
-    outl(omr, DE4X5_OMR);\
-  } else {\
-    omr = (inl(DE4X5_OMR) & ~(OMR_PS | OMR_HBD | OMR_TTM | OMR_PCS | OMR_SCR | OMR_FDX));\
-    outl(omr | OMR_SDP | OMR_PS | OMR_HBD | OMR_PCS, DE4X5_OMR);\
-    lp->cache.gep = (GEP_FDXD | GEP_MODE);\
-    gep_wr(lp->cache.gep, dev);\
-  }\
-}
-
-/*
-** Include the IOCTL stuff
-*/
-#include <linux/sockios.h>
-
-struct de4x5_ioctl {
-	unsigned short cmd;                /* Command to run */
-	unsigned short len;                /* Length of the data buffer */
-	unsigned char  __user *data;       /* Pointer to the data buffer */
-};
-
-/*
-** Recognised commands for the driver
-*/
-#define DE4X5_GET_HWADDR	0x01 /* Get the hardware address */
-#define DE4X5_SET_HWADDR	0x02 /* Set the hardware address */
-/* 0x03 and 0x04 were used before and are obsoleted now. Don't use them. */
-#define DE4X5_SAY_BOO	        0x05 /* Say "Boo!" to the kernel log file */
-#define DE4X5_GET_MCA   	0x06 /* Get a multicast address */
-#define DE4X5_SET_MCA   	0x07 /* Set a multicast address */
-#define DE4X5_CLR_MCA    	0x08 /* Clear a multicast address */
-#define DE4X5_MCA_EN    	0x09 /* Enable a multicast address group */
-#define DE4X5_GET_STATS  	0x0a /* Get the driver statistics */
-#define DE4X5_CLR_STATS 	0x0b /* Zero out the driver statistics */
-#define DE4X5_GET_OMR           0x0c /* Get the OMR Register contents */
-#define DE4X5_SET_OMR           0x0d /* Set the OMR Register contents */
-#define DE4X5_GET_REG           0x0e /* Get the DE4X5 Registers */
-
-#define MOTO_SROM_BUG    (lp->active == 8 && (get_unaligned_le32(dev->dev_addr) & 0x00ffffff) == 0x3e0008)
diff --git a/drivers/net/ethernet/dec/tulip/dmfe.c b/drivers/net/ethernet/dec/tulip/dmfe.c
index 07e10a45beaa..2d3bd343b6e6 100644
--- a/drivers/net/ethernet/dec/tulip/dmfe.c
+++ b/drivers/net/ethernet/dec/tulip/dmfe.c
@@ -1,17 +1,9 @@
+// SPDX-License-Identifier: GPL-2.0-or-later
 /*
     A Davicom DM9102/DM9102A/DM9102A+DM9801/DM9102A+DM9802 NIC fast
     ethernet driver for Linux.
     Copyright (C) 1997  Sten Wang
 
-    This program is free software; you can redistribute it and/or
-    modify it under the terms of the GNU General Public License
-    as published by the Free Software Foundation; either version 2
-    of the License, or (at your option) any later version.
-
-    This program is distributed in the hope that it will be useful,
-    but WITHOUT ANY WARRANTY; without even the implied warranty of
-    MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
-    GNU General Public License for more details.
 
     DAVICOM Web-Site: www.davicom.com.tw
 
@@ -64,8 +56,6 @@
 #define pr_fmt(fmt) KBUILD_MODNAME ": " fmt
 
 #define DRV_NAME	"dmfe"
-#define DRV_VERSION	"1.36.4"
-#define DRV_RELDATE	"2002-01-17"
 
 #include <linux/module.h>
 #include <linux/kernel.h>
@@ -288,10 +278,6 @@ enum dmfe_CR6_bits {
 };
 
 /* Global variable declaration ----------------------------- */
-static int printed_version;
-static const char version[] =
-	"Davicom DM9xxx net driver, version " DRV_VERSION " (" DRV_RELDATE ")";
-
 static int dmfe_debug;
 static unsigned char dmfe_media_mode = DMFE_AUTO;
 static u32 dmfe_cr6_user_set;
@@ -331,7 +317,7 @@ static void dmfe_phy_write_1bit(void __iomem *, u32);
 static u16 dmfe_phy_read_1bit(void __iomem *);
 static u8 dmfe_sense_speed(struct dmfe_board_info *);
 static void dmfe_process_mode(struct dmfe_board_info *);
-static void dmfe_timer(unsigned long);
+static void dmfe_timer(struct timer_list *);
 static inline u32 cal_CRC(unsigned char *, unsigned int, u8);
 static void dmfe_rx_packet(struct net_device *, struct dmfe_board_info *);
 static void dmfe_free_tx_pkt(struct net_device *, struct dmfe_board_info *);
@@ -372,9 +358,6 @@ static int dmfe_init_one(struct pci_dev *pdev, const struct pci_device_id *ent)
 
 	DMFE_DBUG(0, "dmfe_init_one()", 0);
 
-	if (!printed_version++)
-		pr_info("%s\n", version);
-
 	/*
 	 *	SPARC on-board DM910x chips should be handled by the main
 	 *	tulip driver, except for early DM9100s.
@@ -397,7 +380,7 @@ static int dmfe_init_one(struct pci_dev *pdev, const struct pci_device_id *ent)
 		return -ENOMEM;
 	SET_NETDEV_DEV(dev, &pdev->dev);
 
-	if (pci_set_dma_mask(pdev, DMA_BIT_MASK(32))) {
+	if (dma_set_mask(&pdev->dev, DMA_BIT_MASK(32))) {
 		pr_warn("32-bit PCI DMA not available\n");
 		err = -ENODEV;
 		goto err_out_free;
@@ -439,15 +422,17 @@ static int dmfe_init_one(struct pci_dev *pdev, const struct pci_device_id *ent)
 	db = netdev_priv(dev);
 
 	/* Allocate Tx/Rx descriptor memory */
-	db->desc_pool_ptr = pci_alloc_consistent(pdev, sizeof(struct tx_desc) *
-			DESC_ALL_CNT + 0x20, &db->desc_pool_dma_ptr);
+	db->desc_pool_ptr = dma_alloc_coherent(&pdev->dev,
+					       sizeof(struct tx_desc) * DESC_ALL_CNT + 0x20,
+					       &db->desc_pool_dma_ptr, GFP_KERNEL);
 	if (!db->desc_pool_ptr) {
 		err = -ENOMEM;
 		goto err_out_res;
 	}
 
-	db->buf_pool_ptr = pci_alloc_consistent(pdev, TX_BUF_ALLOC *
-			TX_DESC_CNT + 4, &db->buf_pool_dma_ptr);
+	db->buf_pool_ptr = dma_alloc_coherent(&pdev->dev,
+					      TX_BUF_ALLOC * TX_DESC_CNT + 4,
+					      &db->buf_pool_dma_ptr, GFP_KERNEL);
 	if (!db->buf_pool_ptr) {
 		err = -ENOMEM;
 		goto err_out_free_desc;
@@ -491,8 +476,7 @@ static int dmfe_init_one(struct pci_dev *pdev, const struct pci_device_id *ent)
 	}
 
 	/* Set Node address */
-	for (i = 0; i < 6; i++)
-		dev->dev_addr[i] = db->srom[20 + i];
+	eth_hw_addr_set(dev, &db->srom[20]);
 
 	err = register_netdev (dev);
 	if (err)
@@ -509,11 +493,12 @@ static int dmfe_init_one(struct pci_dev *pdev, const struct pci_device_id *ent)
 err_out_unmap:
 	pci_iounmap(pdev, db->ioaddr);
 err_out_free_buf:
-	pci_free_consistent(pdev, TX_BUF_ALLOC * TX_DESC_CNT + 4,
-			    db->buf_pool_ptr, db->buf_pool_dma_ptr);
+	dma_free_coherent(&pdev->dev, TX_BUF_ALLOC * TX_DESC_CNT + 4,
+			  db->buf_pool_ptr, db->buf_pool_dma_ptr);
 err_out_free_desc:
-	pci_free_consistent(pdev, sizeof(struct tx_desc) * DESC_ALL_CNT + 0x20,
-			    db->desc_pool_ptr, db->desc_pool_dma_ptr);
+	dma_free_coherent(&pdev->dev,
+			  sizeof(struct tx_desc) * DESC_ALL_CNT + 0x20,
+			  db->desc_pool_ptr, db->desc_pool_dma_ptr);
 err_out_res:
 	pci_release_regions(pdev);
 err_out_disable:
@@ -532,15 +517,16 @@ static void dmfe_remove_one(struct pci_dev *pdev)
 
 	DMFE_DBUG(0, "dmfe_remove_one()", 0);
 
- 	if (dev) {
+	if (dev) {
 
 		unregister_netdev(dev);
 		pci_iounmap(db->pdev, db->ioaddr);
-		pci_free_consistent(db->pdev, sizeof(struct tx_desc) *
-					DESC_ALL_CNT + 0x20, db->desc_pool_ptr,
- 					db->desc_pool_dma_ptr);
-		pci_free_consistent(db->pdev, TX_BUF_ALLOC * TX_DESC_CNT + 4,
-					db->buf_pool_ptr, db->buf_pool_dma_ptr);
+		dma_free_coherent(&db->pdev->dev,
+				  sizeof(struct tx_desc) * DESC_ALL_CNT + 0x20,
+				  db->desc_pool_ptr, db->desc_pool_dma_ptr);
+		dma_free_coherent(&db->pdev->dev,
+				  TX_BUF_ALLOC * TX_DESC_CNT + 4,
+				  db->buf_pool_ptr, db->buf_pool_dma_ptr);
 		pci_release_regions(pdev);
 		free_netdev(dev);	/* free board information */
 	}
@@ -580,10 +566,10 @@ static int dmfe_open(struct net_device *dev)
 	/* CR6 operation mode decision */
 	if ( !chkmode || (db->chip_id == PCI_DM9132_ID) ||
 		(db->chip_revision >= 0x30) ) {
-    		db->cr6_data |= DMFE_TXTH_256;
+		db->cr6_data |= DMFE_TXTH_256;
 		db->cr0_data = CR0_DEFAULT;
 		db->dm910x_chk_mode=4;		/* Enter the normal mode */
- 	} else {
+	} else {
 		db->cr6_data |= CR6_SFT;	/* Store & Forward mode */
 		db->cr0_data = 0;
 		db->dm910x_chk_mode = 1;	/* Enter the check mode */
@@ -596,10 +582,8 @@ static int dmfe_open(struct net_device *dev)
 	netif_wake_queue(dev);
 
 	/* set and active a timer process */
-	init_timer(&db->timer);
+	timer_setup(&db->timer, dmfe_timer, 0);
 	db->timer.expires = DMFE_TIMER_WUT + HZ * 2;
-	db->timer.data = (unsigned long)dev;
-	db->timer.function = dmfe_timer;
 	add_timer(&db->timer);
 
 	return 0;
@@ -761,7 +745,7 @@ static int dmfe_stop(struct net_device *dev)
 	netif_stop_queue(dev);
 
 	/* deleted timer */
-	del_timer_sync(&db->timer);
+	timer_delete_sync(&db->timer);
 
 	/* Reset & stop DM910X board */
 	dw32(DCR0, DM910X_RESET);
@@ -918,7 +902,7 @@ static void dmfe_free_tx_pkt(struct net_device *dev, struct dmfe_board_info *db)
 			}
 		}
 
-    		txptr = txptr->next_tx_desc;
+		txptr = txptr->next_tx_desc;
 	}/* End of while */
 
 	/* Update TX remove pointer to next */
@@ -974,8 +958,8 @@ static void dmfe_rx_packet(struct net_device *dev, struct dmfe_board_info *db)
 		db->rx_avail_cnt--;
 		db->interval_rx_cnt++;
 
-		pci_unmap_single(db->pdev, le32_to_cpu(rxptr->rdes2),
-				 RX_ALLOC_SIZE, PCI_DMA_FROMDEVICE);
+		dma_unmap_single(&db->pdev->dev, le32_to_cpu(rxptr->rdes2),
+				 RX_ALLOC_SIZE, DMA_FROM_DEVICE);
 
 		if ( (rdes0 & 0x300) != 0x300) {
 			/* A packet without First/Last flag */
@@ -1090,9 +1074,8 @@ static void dmfe_ethtool_get_drvinfo(struct net_device *dev,
 {
 	struct dmfe_board_info *np = netdev_priv(dev);
 
-	strlcpy(info->driver, DRV_NAME, sizeof(info->driver));
-	strlcpy(info->version, DRV_VERSION, sizeof(info->version));
-	strlcpy(info->bus_info, pci_name(np->pdev), sizeof(info->bus_info));
+	strscpy(info->driver, DRV_NAME, sizeof(info->driver));
+	strscpy(info->bus_info, pci_name(np->pdev), sizeof(info->bus_info));
 }
 
 static int dmfe_ethtool_set_wol(struct net_device *dev,
@@ -1130,14 +1113,14 @@ static const struct ethtool_ops netdev_ethtool_ops = {
  *	Dynamic media sense, allocate Rx buffer...
  */
 
-static void dmfe_timer(unsigned long data)
+static void dmfe_timer(struct timer_list *t)
 {
-	struct net_device *dev = (struct net_device *)data;
-	struct dmfe_board_info *db = netdev_priv(dev);
+	struct dmfe_board_info *db = timer_container_of(db, t, timer);
+	struct net_device *dev = pci_get_drvdata(db->pdev);
 	void __iomem *ioaddr = db->ioaddr;
 	u32 tmp_cr8;
 	unsigned char tmp_cr12;
- 	unsigned long flags;
+	unsigned long flags;
 
 	int link_ok, link_ok_phy;
 
@@ -1233,7 +1216,7 @@ static void dmfe_timer(unsigned long data)
 	if (link_ok_phy != link_ok) {
 		DMFE_DBUG (0, "PHY and chip report different link status", 0);
 		link_ok = link_ok | link_ok_phy;
- 	}
+	}
 
 	if ( !link_ok && netif_carrier_ok(dev)) {
 		/* Link Failed */
@@ -1349,8 +1332,8 @@ static void dmfe_reuse_skb(struct dmfe_board_info *db, struct sk_buff * skb)
 
 	if (!(rxptr->rdes0 & cpu_to_le32(0x80000000))) {
 		rxptr->rx_skb_ptr = skb;
-		rxptr->rdes2 = cpu_to_le32( pci_map_single(db->pdev,
-			    skb->data, RX_ALLOC_SIZE, PCI_DMA_FROMDEVICE) );
+		rxptr->rdes2 = cpu_to_le32(dma_map_single(&db->pdev->dev, skb->data,
+							  RX_ALLOC_SIZE, DMA_FROM_DEVICE));
 		wmb();
 		rxptr->rdes0 = cpu_to_le32(0x80000000);
 		db->rx_avail_cnt++;
@@ -1452,9 +1435,9 @@ static void update_cr6(u32 cr6_data, void __iomem *ioaddr)
 
 static void dm9132_id_table(struct net_device *dev)
 {
+	const u16 *addrptr = (const u16 *)dev->dev_addr;
 	struct dmfe_board_info *db = netdev_priv(dev);
 	void __iomem *ioaddr = db->ioaddr + 0xc0;
-	u16 *addrptr = (u16 *)dev->dev_addr;
 	struct netdev_hw_addr *ha;
 	u16 i, hash_table[4];
 
@@ -1493,7 +1476,7 @@ static void send_filter_frame(struct net_device *dev)
 	struct dmfe_board_info *db = netdev_priv(dev);
 	struct netdev_hw_addr *ha;
 	struct tx_desc *txptr;
-	u16 * addrptr;
+	const u16 * addrptr;
 	u32 * suptr;
 	int i;
 
@@ -1503,7 +1486,7 @@ static void send_filter_frame(struct net_device *dev)
 	suptr = (u32 *) txptr->tx_buf_ptr;
 
 	/* Node address */
-	addrptr = (u16 *) dev->dev_addr;
+	addrptr = (const u16 *) dev->dev_addr;
 	*suptr++ = addrptr[0];
 	*suptr++ = addrptr[1];
 	*suptr++ = addrptr[2];
@@ -1564,8 +1547,8 @@ static void allocate_rx_buffer(struct net_device *dev)
 		if ( ( skb = netdev_alloc_skb(dev, RX_ALLOC_SIZE) ) == NULL )
 			break;
 		rxptr->rx_skb_ptr = skb; /* FIXME (?) */
-		rxptr->rdes2 = cpu_to_le32( pci_map_single(db->pdev, skb->data,
-				    RX_ALLOC_SIZE, PCI_DMA_FROMDEVICE) );
+		rxptr->rdes2 = cpu_to_le32(dma_map_single(&db->pdev->dev, skb->data,
+							  RX_ALLOC_SIZE, DMA_FROM_DEVICE));
 		wmb();
 		rxptr->rdes0 = cpu_to_le32(0x80000000);
 		rxptr = rxptr->next_rx_desc;
@@ -1715,14 +1698,14 @@ static void dmfe_set_phyxcer(struct dmfe_board_info *db)
 		if (db->chip_id == PCI_DM9009_ID) phy_reg &= 0x61;
 	}
 
-  	/* Write new capability to Phyxcer Reg4 */
+	/* Write new capability to Phyxcer Reg4 */
 	if ( !(phy_reg & 0x01e0)) {
 		phy_reg|=db->PHY_reg4;
 		db->media_mode|=DMFE_AUTO;
 	}
 	dmfe_phy_write(db->ioaddr, db->phy_addr, 4, phy_reg, db->chip_id);
 
- 	/* Restart Auto-Negotiation */
+	/* Restart Auto-Negotiation */
 	if ( db->chip_type && (db->chip_id == PCI_DM9102_ID) )
 		dmfe_phy_write(db->ioaddr, db->phy_addr, 0, 0x1800, db->chip_id);
 	if ( !db->chip_type )
@@ -1770,7 +1753,7 @@ static void dmfe_process_mode(struct dmfe_board_info *db)
 			}
 			dmfe_phy_write(db->ioaddr,
 				       db->phy_addr, 0, phy_reg, db->chip_id);
-       			if ( db->chip_type && (db->chip_id == PCI_DM9102_ID) )
+			if ( db->chip_type && (db->chip_id == PCI_DM9102_ID) )
 				mdelay(20);
 			dmfe_phy_write(db->ioaddr,
 				       db->phy_addr, 0, phy_reg, db->chip_id);
@@ -2101,14 +2084,11 @@ static const struct pci_device_id dmfe_pci_tbl[] = {
 };
 MODULE_DEVICE_TABLE(pci, dmfe_pci_tbl);
 
-
-#ifdef CONFIG_PM
-static int dmfe_suspend(struct pci_dev *pci_dev, pm_message_t state)
+static int __maybe_unused dmfe_suspend(struct device *dev_d)
 {
-	struct net_device *dev = pci_get_drvdata(pci_dev);
+	struct net_device *dev = dev_get_drvdata(dev_d);
 	struct dmfe_board_info *db = netdev_priv(dev);
 	void __iomem *ioaddr = db->ioaddr;
-	u32 tmp;
 
 	/* Disable upper layer interface */
 	netif_device_detach(dev);
@@ -2125,69 +2105,40 @@ static int dmfe_suspend(struct pci_dev *pci_dev, pm_message_t state)
 	dmfe_free_rxbuffer(db);
 
 	/* Enable WOL */
-	pci_read_config_dword(pci_dev, 0x40, &tmp);
-	tmp &= ~(DMFE_WOL_LINKCHANGE|DMFE_WOL_MAGICPACKET);
-
-	if (db->wol_mode & WAKE_PHY)
-		tmp |= DMFE_WOL_LINKCHANGE;
-	if (db->wol_mode & WAKE_MAGIC)
-		tmp |= DMFE_WOL_MAGICPACKET;
-
-	pci_write_config_dword(pci_dev, 0x40, tmp);
-
-	pci_enable_wake(pci_dev, PCI_D3hot, 1);
-	pci_enable_wake(pci_dev, PCI_D3cold, 1);
-
-	/* Power down device*/
-	pci_save_state(pci_dev);
-	pci_set_power_state(pci_dev, pci_choose_state (pci_dev, state));
+	device_wakeup_enable(dev_d);
 
 	return 0;
 }
 
-static int dmfe_resume(struct pci_dev *pci_dev)
+static int __maybe_unused dmfe_resume(struct device *dev_d)
 {
-	struct net_device *dev = pci_get_drvdata(pci_dev);
-	u32 tmp;
-
-	pci_set_power_state(pci_dev, PCI_D0);
-	pci_restore_state(pci_dev);
+	struct net_device *dev = dev_get_drvdata(dev_d);
 
 	/* Re-initialize DM910X board */
 	dmfe_init_dm910x(dev);
 
 	/* Disable WOL */
-	pci_read_config_dword(pci_dev, 0x40, &tmp);
-
-	tmp &= ~(DMFE_WOL_LINKCHANGE | DMFE_WOL_MAGICPACKET);
-	pci_write_config_dword(pci_dev, 0x40, tmp);
-
-	pci_enable_wake(pci_dev, PCI_D3hot, 0);
-	pci_enable_wake(pci_dev, PCI_D3cold, 0);
+	device_wakeup_disable(dev_d);
 
 	/* Restart upper layer interface */
 	netif_device_attach(dev);
 
 	return 0;
 }
-#else
-#define dmfe_suspend NULL
-#define dmfe_resume NULL
-#endif
+
+static SIMPLE_DEV_PM_OPS(dmfe_pm_ops, dmfe_suspend, dmfe_resume);
 
 static struct pci_driver dmfe_driver = {
 	.name		= "dmfe",
 	.id_table	= dmfe_pci_tbl,
 	.probe		= dmfe_init_one,
 	.remove		= dmfe_remove_one,
-	.suspend        = dmfe_suspend,
-	.resume         = dmfe_resume
+	.driver.pm	= &dmfe_pm_ops,
 };
 
 MODULE_AUTHOR("Sten Wang, sten_wang@davicom.com.tw");
 MODULE_DESCRIPTION("Davicom DM910X fast ethernet driver");
 MODULE_LICENSE("GPL");
-MODULE_VERSION(DRV_VERSION);
 
 module_param(debug, int, 0);
 module_param(mode, byte, 0);
@@ -2214,9 +2165,6 @@ static int __init dmfe_init_module(void)
 {
 	int rc;
 
-	pr_info("%s\n", version);
-	printed_version = 1;
-
 	DMFE_DBUG(0, "init_module() ", debug);
 
 	if (debug)
@@ -2224,15 +2172,16 @@ static int __init dmfe_init_module(void)
 	if (cr6set)
 		dmfe_cr6_user_set = cr6set;
 
- 	switch(mode) {
-   	case DMFE_10MHF:
+	switch (mode) {
+	case DMFE_10MHF:
 	case DMFE_100MHF:
 	case DMFE_10MFD:
 	case DMFE_100MFD:
 	case DMFE_1M_HPNA:
 		dmfe_media_mode = mode;
 		break;
-	default:dmfe_media_mode = DMFE_AUTO;
+	default:
+		dmfe_media_mode = DMFE_AUTO;
 		break;
 	}
 
diff --git a/drivers/net/ethernet/dec/tulip/eeprom.c b/drivers/net/ethernet/dec/tulip/eeprom.c
index 1812f4916917..71ff9e6db209 100644
--- a/drivers/net/ethernet/dec/tulip/eeprom.c
+++ b/drivers/net/ethernet/dec/tulip/eeprom.c
@@ -13,7 +13,7 @@
 #include <linux/pci.h>
 #include <linux/slab.h>
 #include "tulip.h"
-#include <asm/unaligned.h>
+#include <linux/unaligned.h>
 
 
 
@@ -117,8 +117,8 @@ static void tulip_build_fake_mediatable(struct tulip_private *tp)
 			  0x00, 0x06  /* ttm bit map */
 			};
 
-		tp->mtable = kmalloc(sizeof(struct mediatable) +
-				     sizeof(struct medialeaf), GFP_KERNEL);
+		tp->mtable = devm_kmalloc(&tp->pdev->dev, sizeof(struct mediatable) +
+					  sizeof(struct medialeaf), GFP_KERNEL);
 
 		if (tp->mtable == NULL)
 			return; /* Horrible, impossible failure. */
@@ -224,9 +224,8 @@ subsequent_board:
 		        return;
 		}
 
-		mtable = kmalloc(sizeof(struct mediatable) +
-				 count * sizeof(struct medialeaf),
-				 GFP_KERNEL);
+		mtable = devm_kmalloc(&tp->pdev->dev, struct_size(mtable, mleaf, count),
+				      GFP_KERNEL);
 		if (mtable == NULL)
 			return;				/* Horrible, impossible failure. */
 		last_mediatable = tp->mtable = mtable;
diff --git a/drivers/net/ethernet/dec/tulip/interrupt.c b/drivers/net/ethernet/dec/tulip/interrupt.c
index ba6ae24acf62..0a12cb9b3ba7 100644
--- a/drivers/net/ethernet/dec/tulip/interrupt.c
+++ b/drivers/net/ethernet/dec/tulip/interrupt.c
@@ -74,8 +74,8 @@ int tulip_refill_rx(struct net_device *dev)
 			if (skb == NULL)
 				break;
 
-			mapping = pci_map_single(tp->pdev, skb->data, PKT_BUF_SZ,
-						 PCI_DMA_FROMDEVICE);
+			mapping = dma_map_single(&tp->pdev->dev, skb->data,
+						 PKT_BUF_SZ, DMA_FROM_DEVICE);
 			if (dma_mapping_error(&tp->pdev->dev, mapping)) {
 				dev_kfree_skb(skb);
 				tp->rx_buffers[entry].skb = NULL;
@@ -102,10 +102,10 @@ int tulip_refill_rx(struct net_device *dev)
 
 #ifdef CONFIG_TULIP_NAPI
 
-void oom_timer(unsigned long data)
+void oom_timer(struct timer_list *t)
 {
-        struct net_device *dev = (struct net_device *)data;
-	struct tulip_private *tp = netdev_priv(dev);
+	struct tulip_private *tp = timer_container_of(tp, t, oom_timer);
+
 	napi_schedule(&tp->napi);
 }
 
@@ -210,21 +210,23 @@ int tulip_poll(struct napi_struct *napi, int budget)
                                if (pkt_len < tulip_rx_copybreak &&
                                    (skb = netdev_alloc_skb(dev, pkt_len + 2)) != NULL) {
                                        skb_reserve(skb, 2);    /* 16 byte align the IP header */
-                                       pci_dma_sync_single_for_cpu(tp->pdev,
-								   tp->rx_buffers[entry].mapping,
-								   pkt_len, PCI_DMA_FROMDEVICE);
+					dma_sync_single_for_cpu(&tp->pdev->dev,
+								tp->rx_buffers[entry].mapping,
+								pkt_len,
+								DMA_FROM_DEVICE);
 #if ! defined(__alpha__)
                                        skb_copy_to_linear_data(skb, tp->rx_buffers[entry].skb->data,
                                                         pkt_len);
                                        skb_put(skb, pkt_len);
 #else
-                                       memcpy(skb_put(skb, pkt_len),
-                                              tp->rx_buffers[entry].skb->data,
-                                              pkt_len);
+                                       skb_put_data(skb,
+                                                    tp->rx_buffers[entry].skb->data,
+                                                    pkt_len);
 #endif
-                                       pci_dma_sync_single_for_device(tp->pdev,
-								      tp->rx_buffers[entry].mapping,
-								      pkt_len, PCI_DMA_FROMDEVICE);
+					dma_sync_single_for_device(&tp->pdev->dev,
+								   tp->rx_buffers[entry].mapping,
+								   pkt_len,
+								   DMA_FROM_DEVICE);
                                } else {        /* Pass up the skb already on the Rx ring. */
                                        char *temp = skb_put(skb = tp->rx_buffers[entry].skb,
                                                             pkt_len);
@@ -240,8 +242,10 @@ int tulip_poll(struct napi_struct *napi, int budget)
                                        }
 #endif
 
-                                       pci_unmap_single(tp->pdev, tp->rx_buffers[entry].mapping,
-                                                        PKT_BUF_SZ, PCI_DMA_FROMDEVICE);
+					dma_unmap_single(&tp->pdev->dev,
+							 tp->rx_buffers[entry].mapping,
+							 PKT_BUF_SZ,
+							 DMA_FROM_DEVICE);
 
                                        tp->rx_buffers[entry].skb = NULL;
                                        tp->rx_buffers[entry].mapping = 0;
@@ -436,21 +440,23 @@ static int tulip_rx(struct net_device *dev)
 			if (pkt_len < tulip_rx_copybreak &&
 			    (skb = netdev_alloc_skb(dev, pkt_len + 2)) != NULL) {
 				skb_reserve(skb, 2);	/* 16 byte align the IP header */
-				pci_dma_sync_single_for_cpu(tp->pdev,
-							    tp->rx_buffers[entry].mapping,
-							    pkt_len, PCI_DMA_FROMDEVICE);
+				dma_sync_single_for_cpu(&tp->pdev->dev,
+							tp->rx_buffers[entry].mapping,
+							pkt_len,
+							DMA_FROM_DEVICE);
 #if ! defined(__alpha__)
 				skb_copy_to_linear_data(skb, tp->rx_buffers[entry].skb->data,
 						 pkt_len);
 				skb_put(skb, pkt_len);
 #else
-				memcpy(skb_put(skb, pkt_len),
-				       tp->rx_buffers[entry].skb->data,
-				       pkt_len);
+				skb_put_data(skb,
+					     tp->rx_buffers[entry].skb->data,
+					     pkt_len);
 #endif
-				pci_dma_sync_single_for_device(tp->pdev,
-							       tp->rx_buffers[entry].mapping,
-							       pkt_len, PCI_DMA_FROMDEVICE);
+				dma_sync_single_for_device(&tp->pdev->dev,
+							   tp->rx_buffers[entry].mapping,
+							   pkt_len,
+							   DMA_FROM_DEVICE);
 			} else { 	/* Pass up the skb already on the Rx ring. */
 				char *temp = skb_put(skb = tp->rx_buffers[entry].skb,
 						     pkt_len);
@@ -466,8 +472,9 @@ static int tulip_rx(struct net_device *dev)
 				}
 #endif
 
-				pci_unmap_single(tp->pdev, tp->rx_buffers[entry].mapping,
-						 PKT_BUF_SZ, PCI_DMA_FROMDEVICE);
+				dma_unmap_single(&tp->pdev->dev,
+						 tp->rx_buffers[entry].mapping,
+						 PKT_BUF_SZ, DMA_FROM_DEVICE);
 
 				tp->rx_buffers[entry].skb = NULL;
 				tp->rx_buffers[entry].mapping = 0;
@@ -597,10 +604,10 @@ irqreturn_t tulip_interrupt(int irq, void *dev_instance)
 				if (tp->tx_buffers[entry].skb == NULL) {
 					/* test because dummy frames not mapped */
 					if (tp->tx_buffers[entry].mapping)
-						pci_unmap_single(tp->pdev,
-							 tp->tx_buffers[entry].mapping,
-							 sizeof(tp->setup_frame),
-							 PCI_DMA_TODEVICE);
+						dma_unmap_single(&tp->pdev->dev,
+								 tp->tx_buffers[entry].mapping,
+								 sizeof(tp->setup_frame),
+								 DMA_TO_DEVICE);
 					continue;
 				}
 
@@ -629,9 +636,10 @@ irqreturn_t tulip_interrupt(int irq, void *dev_instance)
 					dev->stats.tx_packets++;
 				}
 
-				pci_unmap_single(tp->pdev, tp->tx_buffers[entry].mapping,
+				dma_unmap_single(&tp->pdev->dev,
+						 tp->tx_buffers[entry].mapping,
 						 tp->tx_buffers[entry].skb->len,
-						 PCI_DMA_TODEVICE);
+						 DMA_TO_DEVICE);
 
 				/* Free the original skb. */
 				dev_kfree_skb_irq(tp->tx_buffers[entry].skb);
@@ -691,8 +699,8 @@ irqreturn_t tulip_interrupt(int irq, void *dev_instance)
 				tulip_start_rxtx(tp);
 			}
 			/*
-			 * NB: t21142_lnk_change() does a del_timer_sync(), so be careful if this
-			 * call is ever done under the spinlock
+			 * NB: t21142_lnk_change() does a timer_delete_sync(), so be careful
+			 * if this call is ever done under the spinlock
 			 */
 			if (csr5 & (TPLnkPass | TPLnkFail | 0x08000000)) {
 				if (tp->link_change)
diff --git a/drivers/net/ethernet/dec/tulip/media.c b/drivers/net/ethernet/dec/tulip/media.c
index dcf21a36a9cf..55d6fc99f40b 100644
--- a/drivers/net/ethernet/dec/tulip/media.c
+++ b/drivers/net/ethernet/dec/tulip/media.c
@@ -319,13 +319,8 @@ void tulip_select_media(struct net_device *dev, int startup)
 			break;
 		}
 		case 5: case 6: {
-			u16 setup[5];
-
 			new_csr6 = 0; /* FIXME */
 
-			for (i = 0; i < 5; i++)
-				setup[i] = get_u16(&p[i*2 + 1]);
-
 			if (startup && mtable->has_reset) {
 				struct medialeaf *rleaf = &mtable->mleaf[mtable->has_reset];
 				unsigned char *rst = rleaf->leafdata;
@@ -367,7 +362,7 @@ void tulip_select_media(struct net_device *dev, int startup)
 			iowrite32(0x33, ioaddr + CSR12);
 			new_csr6 = 0x01860000;
 			/* Trigger autonegotiation. */
-			iowrite32(startup ? 0x0201F868 : 0x0001F868, ioaddr + 0xB8);
+			iowrite32(0x0001F868, ioaddr + 0xB8);
 		} else {
 			iowrite32(0x32, ioaddr + CSR12);
 			new_csr6 = 0x00420000;
diff --git a/drivers/net/ethernet/dec/tulip/pnic.c b/drivers/net/ethernet/dec/tulip/pnic.c
index 7bcccf5cac7a..1de5ed967070 100644
--- a/drivers/net/ethernet/dec/tulip/pnic.c
+++ b/drivers/net/ethernet/dec/tulip/pnic.c
@@ -21,7 +21,7 @@ void pnic_do_nway(struct net_device *dev)
 	struct tulip_private *tp = netdev_priv(dev);
 	void __iomem *ioaddr = tp->base_addr;
 	u32 phy_reg = ioread32(ioaddr + 0xB8);
-	u32 new_csr6 = tp->csr6 & ~0x40C40200;
+	u32 new_csr6;
 
 	if (phy_reg & 0x78000000) { /* Ignore baseT4 */
 		if (phy_reg & 0x20000000)		dev->if_port = 5;
@@ -84,10 +84,10 @@ void pnic_lnk_change(struct net_device *dev, int csr5)
 	}
 }
 
-void pnic_timer(unsigned long data)
+void pnic_timer(struct timer_list *t)
 {
-	struct net_device *dev = (struct net_device *)data;
-	struct tulip_private *tp = netdev_priv(dev);
+	struct tulip_private *tp = timer_container_of(tp, t, timer);
+	struct net_device *dev = tp->dev;
 	void __iomem *ioaddr = tp->base_addr;
 	int next_tick = 60*HZ;
 
diff --git a/drivers/net/ethernet/dec/tulip/pnic2.c b/drivers/net/ethernet/dec/tulip/pnic2.c
index 5895fc43f6e0..39c410bf224e 100644
--- a/drivers/net/ethernet/dec/tulip/pnic2.c
+++ b/drivers/net/ethernet/dec/tulip/pnic2.c
@@ -76,10 +76,10 @@
 #include <linux/delay.h>
 
 
-void pnic2_timer(unsigned long data)
+void pnic2_timer(struct timer_list *t)
 {
-	struct net_device *dev = (struct net_device *)data;
-	struct tulip_private *tp = netdev_priv(dev);
+	struct tulip_private *tp = timer_container_of(tp, t, timer);
+	struct net_device *dev = tp->dev;
 	void __iomem *ioaddr = tp->base_addr;
 	int next_tick = 60*HZ;
 
@@ -323,7 +323,7 @@ void pnic2_lnk_change(struct net_device *dev, int csr5)
 		if (tulip_debug > 2)
 			netdev_dbg(dev, "Ugh! Link blew?\n");
 
-		del_timer_sync(&tp->timer);
+		timer_delete_sync(&tp->timer);
 		pnic2_start_nway(dev);
 		tp->timer.expires = RUN_AT(3*HZ);
 		add_timer(&tp->timer);
@@ -348,10 +348,10 @@ void pnic2_lnk_change(struct net_device *dev, int csr5)
 
                 /* if failed then try doing an nway to get in sync */
 		if ((csr12 & 2)  &&  ! tp->medialock) {
-			del_timer_sync(&tp->timer);
+			timer_delete_sync(&tp->timer);
 			pnic2_start_nway(dev);
 			tp->timer.expires = RUN_AT(3*HZ);
-       			add_timer(&tp->timer);
+			add_timer(&tp->timer);
                 }
 
                 return;
@@ -372,10 +372,10 @@ void pnic2_lnk_change(struct net_device *dev, int csr5)
 
                 /* if failed, try doing an nway to get in sync */
 		if ((csr12 & 4)  &&  ! tp->medialock) {
-			del_timer_sync(&tp->timer);
+			timer_delete_sync(&tp->timer);
 			pnic2_start_nway(dev);
 			tp->timer.expires = RUN_AT(3*HZ);
-       			add_timer(&tp->timer);
+			add_timer(&tp->timer);
                 }
 
                 return;
diff --git a/drivers/net/ethernet/dec/tulip/timer.c b/drivers/net/ethernet/dec/tulip/timer.c
index 523d9dde50a2..ca0c509b601c 100644
--- a/drivers/net/ethernet/dec/tulip/timer.c
+++ b/drivers/net/ethernet/dec/tulip/timer.c
@@ -137,10 +137,10 @@ void tulip_media_task(struct work_struct *work)
 }
 
 
-void mxic_timer(unsigned long data)
+void mxic_timer(struct timer_list *t)
 {
-	struct net_device *dev = (struct net_device *)data;
-	struct tulip_private *tp = netdev_priv(dev);
+	struct tulip_private *tp = timer_container_of(tp, t, timer);
+	struct net_device *dev = tp->dev;
 	void __iomem *ioaddr = tp->base_addr;
 	int next_tick = 60*HZ;
 
@@ -154,10 +154,10 @@ void mxic_timer(unsigned long data)
 }
 
 
-void comet_timer(unsigned long data)
+void comet_timer(struct timer_list *t)
 {
-	struct net_device *dev = (struct net_device *)data;
-	struct tulip_private *tp = netdev_priv(dev);
+	struct tulip_private *tp = timer_container_of(tp, t, timer);
+	struct net_device *dev = tp->dev;
 	int next_tick = 2*HZ;
 
 	if (tulip_debug > 1)
diff --git a/drivers/net/ethernet/dec/tulip/tulip.h b/drivers/net/ethernet/dec/tulip/tulip.h
index 38431a155f09..5e010e1fa6f7 100644
--- a/drivers/net/ethernet/dec/tulip/tulip.h
+++ b/drivers/net/ethernet/dec/tulip/tulip.h
@@ -23,7 +23,7 @@
 #include <linux/pci.h>
 #include <asm/io.h>
 #include <asm/irq.h>
-#include <asm/unaligned.h>
+#include <linux/unaligned.h>
 
 
 
@@ -43,7 +43,7 @@ struct tulip_chip_table {
 	int io_size;
 	int valid_intrs;	/* CSR7 interrupt enable settings */
 	int flags;
-	void (*media_timer) (unsigned long);
+	void (*media_timer) (struct timer_list *);
 	work_func_t media_task;
 };
 
@@ -381,7 +381,7 @@ struct mediatable {
 	unsigned has_reset:6;
 	u32 csr15dir;
 	u32 csr15val;		/* 21143 NWay setting. */
-	struct medialeaf mleaf[0];
+	struct medialeaf mleaf[] __counted_by(leafcount);
 };
 
 
@@ -476,9 +476,8 @@ void t21142_lnk_change(struct net_device *dev, int csr5);
 
 /* PNIC2.c */
 void pnic2_lnk_change(struct net_device *dev, int csr5);
-void pnic2_timer(unsigned long data);
+void pnic2_timer(struct timer_list *t);
 void pnic2_start_nway(struct net_device *dev);
-void pnic2_lnk_change(struct net_device *dev, int csr5);
 
 /* eeprom.c */
 void tulip_parse_eeprom(struct net_device *dev);
@@ -504,19 +503,19 @@ void tulip_find_mii (struct net_device *dev, int board_idx);
 /* pnic.c */
 void pnic_do_nway(struct net_device *dev);
 void pnic_lnk_change(struct net_device *dev, int csr5);
-void pnic_timer(unsigned long data);
+void pnic_timer(struct timer_list *t);
 
 /* timer.c */
 void tulip_media_task(struct work_struct *work);
-void mxic_timer(unsigned long data);
-void comet_timer(unsigned long data);
+void mxic_timer(struct timer_list *t);
+void comet_timer(struct timer_list *t);
 
 /* tulip_core.c */
 extern int tulip_debug;
 extern const char * const medianame[];
 extern const char tulip_media_cap[];
-extern struct tulip_chip_table tulip_tbl[];
-void oom_timer(unsigned long data);
+extern const struct tulip_chip_table tulip_tbl[];
+void oom_timer(struct timer_list *t);
 extern u8 t21040_csr13[];
 
 static inline void tulip_start_rxtx(struct tulip_private *tp)
diff --git a/drivers/net/ethernet/dec/tulip/tulip_core.c b/drivers/net/ethernet/dec/tulip/tulip_core.c
index 17e566a8b345..b608585f1954 100644
--- a/drivers/net/ethernet/dec/tulip/tulip_core.c
+++ b/drivers/net/ethernet/dec/tulip/tulip_core.c
@@ -12,13 +12,6 @@
 #define pr_fmt(fmt) "tulip: " fmt
 
 #define DRV_NAME	"tulip"
-#ifdef CONFIG_TULIP_NAPI
-#define DRV_VERSION    "1.1.15-NAPI" /* Keep at least for test */
-#else
-#define DRV_VERSION	"1.1.15"
-#endif
-#define DRV_RELDATE	"Feb 27, 2007"
-
 
 #include <linux/module.h>
 #include <linux/pci.h>
@@ -30,16 +23,13 @@
 #include <linux/delay.h>
 #include <linux/mii.h>
 #include <linux/crc32.h>
-#include <asm/unaligned.h>
+#include <linux/unaligned.h>
 #include <linux/uaccess.h>
 
 #ifdef CONFIG_SPARC
 #include <asm/prom.h>
 #endif
 
-static char version[] =
-	"Linux Tulip driver version " DRV_VERSION " (" DRV_RELDATE ")\n";
-
 /* A few user-configurable values. */
 
 /* Maximum events (Rx packets, etc.) to handle at each interrupt. */
@@ -109,7 +99,6 @@ static int csr0;
 MODULE_AUTHOR("The Linux Kernel Team");
 MODULE_DESCRIPTION("Digital 21*4* Tulip ethernet driver");
 MODULE_LICENSE("GPL");
-MODULE_VERSION(DRV_VERSION);
 module_param(tulip_debug, int, 0);
 module_param(max_interrupt_work, int, 0);
 module_param(rx_copybreak, int, 0);
@@ -123,10 +112,10 @@ int tulip_debug = TULIP_DEBUG;
 int tulip_debug = 1;
 #endif
 
-static void tulip_timer(unsigned long data)
+static void tulip_timer(struct timer_list *t)
 {
-	struct net_device *dev = (struct net_device *)data;
-	struct tulip_private *tp = netdev_priv(dev);
+	struct tulip_private *tp = timer_container_of(tp, t, timer);
+	struct net_device *dev = tp->dev;
 
 	if (netif_running(dev))
 		schedule_work(&tp->media_work);
@@ -138,7 +127,7 @@ static void tulip_timer(unsigned long data)
  * It is indexed via the values in 'enum chips'
  */
 
-struct tulip_chip_table tulip_tbl[] = {
+const struct tulip_chip_table tulip_tbl[] = {
   { }, /* placeholder for array, slot unused currently */
   { }, /* placeholder for array, slot unused currently */
 
@@ -255,7 +244,7 @@ MODULE_DEVICE_TABLE(pci, tulip_pci_tbl);
 const char tulip_media_cap[32] =
 {0,0,0,16,  3,19,16,24,  27,4,7,5, 0,20,23,20,  28,31,0,0, };
 
-static void tulip_tx_timeout(struct net_device *dev);
+static void tulip_tx_timeout(struct net_device *dev, unsigned int txqueue);
 static void tulip_init_ring(struct net_device *dev);
 static void tulip_free_ring(struct net_device *dev);
 static netdev_tx_t tulip_start_xmit(struct sk_buff *skb,
@@ -350,7 +339,7 @@ static void tulip_up(struct net_device *dev)
 		}
 	} else {
 		/* This is set_rx_mode(), but without starting the transmitter. */
-		u16 *eaddrs = (u16 *)dev->dev_addr;
+		const u16 *eaddrs = (const u16 *)dev->dev_addr;
 		u16 *setup_frm = &tp->setup_frame[15*6];
 		dma_addr_t mapping;
 
@@ -361,9 +350,9 @@ static void tulip_up(struct net_device *dev)
 		*setup_frm++ = eaddrs[1]; *setup_frm++ = eaddrs[1];
 		*setup_frm++ = eaddrs[2]; *setup_frm++ = eaddrs[2];
 
-		mapping = pci_map_single(tp->pdev, tp->setup_frame,
+		mapping = dma_map_single(&tp->pdev->dev, tp->setup_frame,
 					 sizeof(tp->setup_frame),
-					 PCI_DMA_TODEVICE);
+					 DMA_TO_DEVICE);
 		tp->tx_buffers[tp->cur_tx].skb = NULL;
 		tp->tx_buffers[tp->cur_tx].mapping = mapping;
 
@@ -505,7 +494,7 @@ media_picked:
 	tp->timer.expires = RUN_AT(next_tick);
 	add_timer(&tp->timer);
 #ifdef CONFIG_TULIP_NAPI
-	setup_timer(&tp->oom_timer, oom_timer, (unsigned long)dev);
+	timer_setup(&tp->oom_timer, oom_timer, 0);
 #endif
 }
 
@@ -534,7 +523,7 @@ free_ring:
 }
 
 
-static void tulip_tx_timeout(struct net_device *dev)
+static void tulip_tx_timeout(struct net_device *dev, unsigned int txqueue)
 {
 	struct tulip_private *tp = netdev_priv(dev);
 	void __iomem *ioaddr = tp->base_addr;
@@ -641,8 +630,8 @@ static void tulip_init_ring(struct net_device *dev)
 		tp->rx_buffers[i].skb = skb;
 		if (skb == NULL)
 			break;
-		mapping = pci_map_single(tp->pdev, skb->data,
-					 PKT_BUF_SZ, PCI_DMA_FROMDEVICE);
+		mapping = dma_map_single(&tp->pdev->dev, skb->data,
+					 PKT_BUF_SZ, DMA_FROM_DEVICE);
 		tp->rx_buffers[i].mapping = mapping;
 		tp->rx_ring[i].status = cpu_to_le32(DescOwned);	/* Owned by Tulip chip */
 		tp->rx_ring[i].buffer1 = cpu_to_le32(mapping);
@@ -675,8 +664,8 @@ tulip_start_xmit(struct sk_buff *skb, struct net_device *dev)
 	entry = tp->cur_tx % TX_RING_SIZE;
 
 	tp->tx_buffers[entry].skb = skb;
-	mapping = pci_map_single(tp->pdev, skb->data,
-				 skb->len, PCI_DMA_TODEVICE);
+	mapping = dma_map_single(&tp->pdev->dev, skb->data, skb->len,
+				 DMA_TO_DEVICE);
 	tp->tx_buffers[entry].mapping = mapping;
 	tp->tx_ring[entry].buffer1 = cpu_to_le32(mapping);
 
@@ -727,16 +716,17 @@ static void tulip_clean_tx_ring(struct tulip_private *tp)
 		if (tp->tx_buffers[entry].skb == NULL) {
 			/* test because dummy frames not mapped */
 			if (tp->tx_buffers[entry].mapping)
-				pci_unmap_single(tp->pdev,
-					tp->tx_buffers[entry].mapping,
-					sizeof(tp->setup_frame),
-					PCI_DMA_TODEVICE);
+				dma_unmap_single(&tp->pdev->dev,
+						 tp->tx_buffers[entry].mapping,
+						 sizeof(tp->setup_frame),
+						 DMA_TO_DEVICE);
 			continue;
 		}
 
-		pci_unmap_single(tp->pdev, tp->tx_buffers[entry].mapping,
-				tp->tx_buffers[entry].skb->len,
-				PCI_DMA_TODEVICE);
+		dma_unmap_single(&tp->pdev->dev,
+				 tp->tx_buffers[entry].mapping,
+				 tp->tx_buffers[entry].skb->len,
+				 DMA_TO_DEVICE);
 
 		/* Free the original skb. */
 		dev_kfree_skb_irq(tp->tx_buffers[entry].skb);
@@ -757,9 +747,9 @@ static void tulip_down (struct net_device *dev)
 	napi_disable(&tp->napi);
 #endif
 
-	del_timer_sync (&tp->timer);
+	timer_delete_sync(&tp->timer);
 #ifdef CONFIG_TULIP_NAPI
-	del_timer_sync (&tp->oom_timer);
+	timer_delete_sync(&tp->oom_timer);
 #endif
 	spin_lock_irqsave (&tp->lock, flags);
 
@@ -780,8 +770,7 @@ static void tulip_down (struct net_device *dev)
 
 	spin_unlock_irqrestore (&tp->lock, flags);
 
-	setup_timer(&tp->timer, tulip_tbl[tp->chip_id].media_timer,
-		    (unsigned long)dev);
+	timer_setup(&tp->timer, tulip_tbl[tp->chip_id].media_timer, 0);
 
 	dev->if_port = tp->saved_if_port;
 
@@ -807,8 +796,8 @@ static void tulip_free_ring (struct net_device *dev)
 		/* An invalid address. */
 		tp->rx_ring[i].buffer1 = cpu_to_le32(0xBADF00D0);
 		if (skb) {
-			pci_unmap_single(tp->pdev, mapping, PKT_BUF_SZ,
-					 PCI_DMA_FROMDEVICE);
+			dma_unmap_single(&tp->pdev->dev, mapping, PKT_BUF_SZ,
+					 DMA_FROM_DEVICE);
 			dev_kfree_skb (skb);
 		}
 	}
@@ -817,8 +806,9 @@ static void tulip_free_ring (struct net_device *dev)
 		struct sk_buff *skb = tp->tx_buffers[i].skb;
 
 		if (skb != NULL) {
-			pci_unmap_single(tp->pdev, tp->tx_buffers[i].mapping,
-					 skb->len, PCI_DMA_TODEVICE);
+			dma_unmap_single(&tp->pdev->dev,
+					 tp->tx_buffers[i].mapping, skb->len,
+					 DMA_TO_DEVICE);
 			dev_kfree_skb (skb);
 		}
 		tp->tx_buffers[i].skb = NULL;
@@ -868,9 +858,8 @@ static struct net_device_stats *tulip_get_stats(struct net_device *dev)
 static void tulip_get_drvinfo(struct net_device *dev, struct ethtool_drvinfo *info)
 {
 	struct tulip_private *np = netdev_priv(dev);
-	strlcpy(info->driver, DRV_NAME, sizeof(info->driver));
-	strlcpy(info->version, DRV_VERSION, sizeof(info->version));
-	strlcpy(info->bus_info, pci_name(np->pdev), sizeof(info->bus_info));
+	strscpy(info->driver, DRV_NAME, sizeof(info->driver));
+	strscpy(info->bus_info, pci_name(np->pdev), sizeof(info->bus_info));
 }
 
 
@@ -924,6 +913,7 @@ static int private_ioctl (struct net_device *dev, struct ifreq *rq, int cmd)
 			data->phy_id = 1;
 		else
 			return -ENODEV;
+		fallthrough;
 
 	case SIOCGMIIREG:		/* Read MII PHY register. */
 		if (data->phy_id == 32 && (tp->flags & HAS_NWAY)) {
@@ -1011,8 +1001,8 @@ static void build_setup_frame_hash(u16 *setup_frm, struct net_device *dev)
 	struct tulip_private *tp = netdev_priv(dev);
 	u16 hash_table[32];
 	struct netdev_hw_addr *ha;
+	const u16 *eaddrs;
 	int i;
-	u16 *eaddrs;
 
 	memset(hash_table, 0, sizeof(hash_table));
 	__set_bit_le(255, hash_table);			/* Broadcast entry */
@@ -1029,7 +1019,7 @@ static void build_setup_frame_hash(u16 *setup_frm, struct net_device *dev)
 	setup_frm = &tp->setup_frame[13*6];
 
 	/* Fill the final entry with our physical address. */
-	eaddrs = (u16 *)dev->dev_addr;
+	eaddrs = (const u16 *)dev->dev_addr;
 	*setup_frm++ = eaddrs[0]; *setup_frm++ = eaddrs[0];
 	*setup_frm++ = eaddrs[1]; *setup_frm++ = eaddrs[1];
 	*setup_frm++ = eaddrs[2]; *setup_frm++ = eaddrs[2];
@@ -1039,7 +1029,7 @@ static void build_setup_frame_perfect(u16 *setup_frm, struct net_device *dev)
 {
 	struct tulip_private *tp = netdev_priv(dev);
 	struct netdev_hw_addr *ha;
-	u16 *eaddrs;
+	const u16 *eaddrs;
 
 	/* We have <= 14 addresses so we can use the wonderful
 	   16 address perfect filtering of the Tulip. */
@@ -1054,7 +1044,7 @@ static void build_setup_frame_perfect(u16 *setup_frm, struct net_device *dev)
 	setup_frm = &tp->setup_frame[15*6];
 
 	/* Fill the final entry with our physical address. */
-	eaddrs = (u16 *)dev->dev_addr;
+	eaddrs = (const u16 *)dev->dev_addr;
 	*setup_frm++ = eaddrs[0]; *setup_frm++ = eaddrs[0];
 	*setup_frm++ = eaddrs[1]; *setup_frm++ = eaddrs[1];
 	*setup_frm++ = eaddrs[2]; *setup_frm++ = eaddrs[2];
@@ -1161,9 +1151,10 @@ static void set_rx_mode(struct net_device *dev)
 
 			tp->tx_buffers[entry].skb = NULL;
 			tp->tx_buffers[entry].mapping =
-				pci_map_single(tp->pdev, tp->setup_frame,
+				dma_map_single(&tp->pdev->dev,
+					       tp->setup_frame,
 					       sizeof(tp->setup_frame),
-					       PCI_DMA_TODEVICE);
+					       DMA_TO_DEVICE);
 			/* Put the setup frame on the Tx list. */
 			if (entry == TX_RING_SIZE-1)
 				tx_flags |= DESC_RING_WRAP;		/* Wrap ring. */
@@ -1186,7 +1177,6 @@ static void set_rx_mode(struct net_device *dev)
 	iowrite32(csr6, ioaddr + CSR6);
 }
 
-#ifdef CONFIG_TULIP_MWI
 static void tulip_mwi_config(struct pci_dev *pdev, struct net_device *dev)
 {
 	struct tulip_private *tp = netdev_priv(dev);
@@ -1260,7 +1250,6 @@ out:
 		netdev_dbg(dev, "MWI config cacheline=%d, csr0=%08x\n",
 			   cache, csr0);
 }
-#endif
 
 /*
  *	Chips that have the MRM/reserved bit quirk and the burst quirk. That
@@ -1280,7 +1269,7 @@ static const struct net_device_ops tulip_netdev_ops = {
 	.ndo_tx_timeout		= tulip_tx_timeout,
 	.ndo_stop		= tulip_close,
 	.ndo_get_stats		= tulip_get_stats,
-	.ndo_do_ioctl 		= private_ioctl,
+	.ndo_eth_ioctl		= private_ioctl,
 	.ndo_set_rx_mode	= set_rx_mode,
 	.ndo_set_mac_address	= eth_mac_addr,
 	.ndo_validate_addr	= eth_validate_addr,
@@ -1289,7 +1278,7 @@ static const struct net_device_ops tulip_netdev_ops = {
 #endif
 };
 
-const struct pci_device_id early_486_chipsets[] = {
+static const struct pci_device_id early_486_chipsets[] = {
 	{ PCI_DEVICE(PCI_VENDOR_ID_INTEL, PCI_DEVICE_ID_INTEL_82424) },
 	{ PCI_DEVICE(PCI_VENDOR_ID_SI, PCI_DEVICE_ID_SI_496) },
 	{ },
@@ -1302,8 +1291,9 @@ static int tulip_init_one(struct pci_dev *pdev, const struct pci_device_id *ent)
 	static unsigned char last_phys_addr[ETH_ALEN] = {
 		0x00, 'L', 'i', 'n', 'u', 'x'
 	};
+#if defined(__i386__) || defined(__x86_64__)	/* Patch up x86 BIOS bug. */
 	static int last_irq;
-	static int multiport_cnt;	/* For four-port boards w/one EEPROM */
+#endif
 	int i, irq;
 	unsigned short sum;
 	unsigned char *ee_data;
@@ -1313,13 +1303,9 @@ static int tulip_init_one(struct pci_dev *pdev, const struct pci_device_id *ent)
 	int chip_idx = ent->driver_data;
 	const char *chip_name = tulip_tbl[chip_idx].chip_name;
 	unsigned int eeprom_missing = 0;
+	u8 addr[ETH_ALEN] __aligned(2);
 	unsigned int force_csr0 = 0;
 
-#ifndef MODULE
-	if (tulip_debug > 0)
-		printk_once(KERN_INFO "%s", version);
-#endif
-
 	board_idx++;
 
 	/*
@@ -1401,7 +1387,7 @@ static int tulip_init_one(struct pci_dev *pdev, const struct pci_device_id *ent)
 	 *	And back to business
 	 */
 
-	i = pci_enable_device(pdev);
+	i = pcim_enable_device(pdev);
 	if (i) {
 		pr_err("Cannot enable tulip board #%d, aborting\n", board_idx);
 		return i;
@@ -1410,7 +1396,7 @@ static int tulip_init_one(struct pci_dev *pdev, const struct pci_device_id *ent)
 	irq = pdev->irq;
 
 	/* alloc_etherdev ensures aligned and zeroed private structures */
-	dev = alloc_etherdev (sizeof (*tp));
+	dev = devm_alloc_etherdev(&pdev->dev, sizeof(*tp));
 	if (!dev)
 		return -ENOMEM;
 
@@ -1420,18 +1406,18 @@ static int tulip_init_one(struct pci_dev *pdev, const struct pci_device_id *ent)
 		       pci_name(pdev),
 		       (unsigned long long)pci_resource_len (pdev, 0),
 		       (unsigned long long)pci_resource_start (pdev, 0));
-		goto err_out_free_netdev;
+		return -ENODEV;
 	}
 
 	/* grab all resources from both PIO and MMIO regions, as we
 	 * don't want anyone else messing around with our hardware */
-	if (pci_request_regions (pdev, DRV_NAME))
-		goto err_out_free_netdev;
+	if (pcim_request_all_regions(pdev, DRV_NAME))
+		return -ENODEV;
 
-	ioaddr =  pci_iomap(pdev, TULIP_BAR, tulip_tbl[chip_idx].io_size);
+	ioaddr = pcim_iomap(pdev, TULIP_BAR, tulip_tbl[chip_idx].io_size);
 
 	if (!ioaddr)
-		goto err_out_free_res;
+		return -ENODEV;
 
 	/*
 	 * initialize private data structure 'tp'
@@ -1440,12 +1426,12 @@ static int tulip_init_one(struct pci_dev *pdev, const struct pci_device_id *ent)
 	tp = netdev_priv(dev);
 	tp->dev = dev;
 
-	tp->rx_ring = pci_alloc_consistent(pdev,
-					   sizeof(struct tulip_rx_desc) * RX_RING_SIZE +
-					   sizeof(struct tulip_tx_desc) * TX_RING_SIZE,
-					   &tp->rx_ring_dma);
+	tp->rx_ring = dmam_alloc_coherent(&pdev->dev,
+					  sizeof(struct tulip_rx_desc) * RX_RING_SIZE +
+					  sizeof(struct tulip_tx_desc) * TX_RING_SIZE,
+					  &tp->rx_ring_dma, GFP_KERNEL);
 	if (!tp->rx_ring)
-		goto err_out_mtable;
+		return -ENODEV;
 	tp->tx_ring = (struct tulip_tx_desc *)(tp->rx_ring + RX_RING_SIZE);
 	tp->tx_ring_dma = tp->rx_ring_dma + sizeof(struct tulip_rx_desc) * RX_RING_SIZE;
 
@@ -1471,15 +1457,13 @@ static int tulip_init_one(struct pci_dev *pdev, const struct pci_device_id *ent)
 	tp->csr0 = csr0;
 	spin_lock_init(&tp->lock);
 	spin_lock_init(&tp->mii_lock);
-	setup_timer(&tp->timer, tulip_tbl[tp->chip_id].media_timer,
-		    (unsigned long)dev);
+	timer_setup(&tp->timer, tulip_tbl[tp->chip_id].media_timer, 0);
 
 	INIT_WORK(&tp->media_work, tulip_tbl[tp->chip_id].media_task);
 
-#ifdef CONFIG_TULIP_MWI
-	if (!force_csr0 && (tp->flags & HAS_PCI_MWI))
+	if (IS_ENABLED(CONFIG_TULIP_MWI) && !force_csr0 &&
+	    (tp->flags & HAS_PCI_MWI))
 		tulip_mwi_config (pdev, dev);
-#endif
 
 	/* Stop the chip's Tx and Rx processes. */
 	tulip_stop_rxtx(tp);
@@ -1520,13 +1504,15 @@ static int tulip_init_one(struct pci_dev *pdev, const struct pci_device_id *ent)
 			do {
 				value = ioread32(ioaddr + CSR9);
 			} while (value < 0  && --boguscnt > 0);
-			put_unaligned_le16(value, ((__le16 *)dev->dev_addr) + i);
+			put_unaligned_le16(value, ((__le16 *)addr) + i);
 			sum += value & 0xffff;
 		}
+		eth_hw_addr_set(dev, addr);
 	} else if (chip_idx == COMET) {
 		/* No need to read the EEPROM. */
-		put_unaligned_le32(ioread32(ioaddr + 0xA4), dev->dev_addr);
-		put_unaligned_le16(ioread32(ioaddr + 0xA8), dev->dev_addr + 4);
+		put_unaligned_le32(ioread32(ioaddr + 0xA4), addr);
+		put_unaligned_le16(ioread32(ioaddr + 0xA8), addr + 4);
+		eth_hw_addr_set(dev, addr);
 		for (i = 0; i < 6; i ++)
 			sum += dev->dev_addr[i];
 	} else {
@@ -1557,7 +1543,6 @@ static int tulip_init_one(struct pci_dev *pdev, const struct pci_device_id *ent)
 		} else if (ee_data[0] == 0xff  &&  ee_data[1] == 0xff &&
 				   ee_data[2] == 0) {
 			sa_offset = 2;		/* Grrr, damn Matrox boards. */
-			multiport_cnt = 4;
 		}
 #ifdef CONFIG_MIPS_COBALT
                if ((pdev->bus->number == 0) &&
@@ -1565,7 +1550,7 @@ static int tulip_init_one(struct pci_dev *pdev, const struct pci_device_id *ent)
                     (PCI_SLOT(pdev->devfn) == 12))) {
                        /* Cobalt MAC address in first EEPROM locations. */
                        sa_offset = 0;
-		       /* Ensure our media table fixup get's applied */
+		       /* Ensure our media table fixup gets applied */
 		       memcpy(ee_data + 16, ee_data, 8);
                }
 #endif
@@ -1590,20 +1575,23 @@ static int tulip_init_one(struct pci_dev *pdev, const struct pci_device_id *ent)
 #endif
 
 		for (i = 0; i < 6; i ++) {
-			dev->dev_addr[i] = ee_data[i + sa_offset];
+			addr[i] = ee_data[i + sa_offset];
 			sum += ee_data[i + sa_offset];
 		}
+		eth_hw_addr_set(dev, addr);
 	}
 	/* Lite-On boards have the address byte-swapped. */
 	if ((dev->dev_addr[0] == 0xA0 ||
 	     dev->dev_addr[0] == 0xC0 ||
 	     dev->dev_addr[0] == 0x02) &&
-	    dev->dev_addr[1] == 0x00)
+	    dev->dev_addr[1] == 0x00) {
 		for (i = 0; i < 6; i+=2) {
-			char tmp = dev->dev_addr[i];
-			dev->dev_addr[i] = dev->dev_addr[i+1];
-			dev->dev_addr[i+1] = tmp;
+			addr[i] = dev->dev_addr[i+1];
+			addr[i+1] = dev->dev_addr[i];
 		}
+		eth_hw_addr_set(dev, addr);
+	}
+
 	/* On the Zynx 315 Etherarray and other multiport boards only the
 	   first Tulip has an EEPROM.
 	   On Sparc systems the mac address is held in the OBP property
@@ -1614,17 +1602,18 @@ static int tulip_init_one(struct pci_dev *pdev, const struct pci_device_id *ent)
 	if (sum == 0  || sum == 6*0xff) {
 #if defined(CONFIG_SPARC)
 		struct device_node *dp = pci_device_to_OF_node(pdev);
-		const unsigned char *addr;
+		const unsigned char *addr2;
 		int len;
 #endif
 		eeprom_missing = 1;
 		for (i = 0; i < 5; i++)
-			dev->dev_addr[i] = last_phys_addr[i];
-		dev->dev_addr[i] = last_phys_addr[i] + 1;
+			addr[i] = last_phys_addr[i];
+		addr[i] = last_phys_addr[i] + 1;
+		eth_hw_addr_set(dev, addr);
 #if defined(CONFIG_SPARC)
-		addr = of_get_property(dp, "local-mac-address", &len);
-		if (addr && len == ETH_ALEN)
-			memcpy(dev->dev_addr, addr, ETH_ALEN);
+		addr2 = of_get_property(dp, "local-mac-address", &len);
+		if (addr2 && len == ETH_ALEN)
+			eth_hw_addr_set(dev, addr2);
 #endif
 #if defined(__i386__) || defined(__x86_64__)	/* Patch up x86 BIOS bug. */
 		if (last_irq)
@@ -1634,7 +1623,9 @@ static int tulip_init_one(struct pci_dev *pdev, const struct pci_device_id *ent)
 
 	for (i = 0; i < 6; i++)
 		last_phys_addr[i] = dev->dev_addr[i];
+#if defined(__i386__) || defined(__x86_64__)	/* Patch up x86 BIOS bug. */
 	last_irq = irq;
+#endif
 
 	/* The lower four bits are the media type. */
 	if (board_idx >= 0  &&  board_idx < MAX_UNITS) {
@@ -1695,12 +1686,13 @@ static int tulip_init_one(struct pci_dev *pdev, const struct pci_device_id *ent)
 	dev->netdev_ops = &tulip_netdev_ops;
 	dev->watchdog_timeo = TX_TIMEOUT;
 #ifdef CONFIG_TULIP_NAPI
-	netif_napi_add(dev, &tp->napi, tulip_poll, 16);
+	netif_napi_add_weight(dev, &tp->napi, tulip_poll, 16);
 #endif
 	dev->ethtool_ops = &ops;
 
-	if (register_netdev(dev))
-		goto err_out_free_ring;
+	i = register_netdev(dev);
+	if (i)
+		return i;
 
 	pci_set_drvdata(pdev, dev);
 
@@ -1775,23 +1767,6 @@ static int tulip_init_one(struct pci_dev *pdev, const struct pci_device_id *ent)
 	tulip_set_power_state (tp, 0, 1);
 
 	return 0;
-
-err_out_free_ring:
-	pci_free_consistent (pdev,
-			     sizeof (struct tulip_rx_desc) * RX_RING_SIZE +
-			     sizeof (struct tulip_tx_desc) * TX_RING_SIZE,
-			     tp->rx_ring, tp->rx_ring_dma);
-
-err_out_mtable:
-	kfree (tp->mtable);
-	pci_iounmap(pdev, ioaddr);
-
-err_out_free_res:
-	pci_release_regions (pdev);
-
-err_out_free_netdev:
-	free_netdev (dev);
-	return -ENODEV;
 }
 
 
@@ -1803,14 +1778,13 @@ static void tulip_set_wolopts (struct pci_dev *pdev, u32 wolopts)
 	void __iomem *ioaddr = tp->base_addr;
 
 	if (tp->flags & COMET_PM) {
-	  
 		unsigned int tmp;
-			
+
 		tmp = ioread32(ioaddr + CSR18);
 		tmp &= ~(comet_csr18_pmes_sticky | comet_csr18_apm_mode | comet_csr18_d3a);
 		tmp |= comet_csr18_pm_mode;
 		iowrite32(tmp, ioaddr + CSR18);
-			
+
 		/* Set the Wake-up Control/Status Register to the given WOL options*/
 		tmp = ioread32(ioaddr + CSR13);
 		tmp &= ~(comet_csr13_linkoffe | comet_csr13_linkone | comet_csr13_wfre | comet_csr13_lsce | comet_csr13_mpre);
@@ -1824,13 +1798,9 @@ static void tulip_set_wolopts (struct pci_dev *pdev, u32 wolopts)
 	}
 }
 
-#ifdef CONFIG_PM
-
-
-static int tulip_suspend (struct pci_dev *pdev, pm_message_t state)
+static int __maybe_unused tulip_suspend(struct device *dev_d)
 {
-	pci_power_t pstate;
-	struct net_device *dev = pci_get_drvdata(pdev);
+	struct net_device *dev = dev_get_drvdata(dev_d);
 	struct tulip_private *tp = netdev_priv(dev);
 
 	if (!dev)
@@ -1846,45 +1816,27 @@ static int tulip_suspend (struct pci_dev *pdev, pm_message_t state)
 	free_irq(tp->pdev->irq, dev);
 
 save_state:
-	pci_save_state(pdev);
-	pci_disable_device(pdev);
-	pstate = pci_choose_state(pdev, state);
-	if (state.event == PM_EVENT_SUSPEND && pstate != PCI_D0) {
-		int rc;
-
-		tulip_set_wolopts(pdev, tp->wolinfo.wolopts);
-		rc = pci_enable_wake(pdev, pstate, tp->wolinfo.wolopts);
-		if (rc)
-			pr_err("pci_enable_wake failed (%d)\n", rc);
-	}
-	pci_set_power_state(pdev, pstate);
+	tulip_set_wolopts(to_pci_dev(dev_d), tp->wolinfo.wolopts);
+	device_set_wakeup_enable(dev_d, !!tp->wolinfo.wolopts);
 
 	return 0;
 }
 
-
-static int tulip_resume(struct pci_dev *pdev)
+static int __maybe_unused tulip_resume(struct device *dev_d)
 {
-	struct net_device *dev = pci_get_drvdata(pdev);
+	struct pci_dev *pdev = to_pci_dev(dev_d);
+	struct net_device *dev = dev_get_drvdata(dev_d);
 	struct tulip_private *tp = netdev_priv(dev);
 	void __iomem *ioaddr = tp->base_addr;
-	int retval;
 	unsigned int tmp;
+	int retval = 0;
 
 	if (!dev)
 		return -EINVAL;
 
-	pci_set_power_state(pdev, PCI_D0);
-	pci_restore_state(pdev);
-
 	if (!netif_running(dev))
 		return 0;
 
-	if ((retval = pci_enable_device(pdev))) {
-		pr_err("pci_enable_device failed in resume\n");
-		return retval;
-	}
-
 	retval = request_irq(pdev->irq, tulip_interrupt, IRQF_SHARED,
 			     dev->name, dev);
 	if (retval) {
@@ -1893,8 +1845,7 @@ static int tulip_resume(struct pci_dev *pdev)
 	}
 
 	if (tp->flags & COMET_PM) {
-		pci_enable_wake(pdev, PCI_D3hot, 0);
-		pci_enable_wake(pdev, PCI_D3cold, 0);
+		device_set_wakeup_enable(dev_d, 0);
 
 		/* Clear the PMES flag */
 		tmp = ioread32(ioaddr + CSR20);
@@ -1912,30 +1863,14 @@ static int tulip_resume(struct pci_dev *pdev)
 	return 0;
 }
 
-#endif /* CONFIG_PM */
-
-
 static void tulip_remove_one(struct pci_dev *pdev)
 {
 	struct net_device *dev = pci_get_drvdata (pdev);
-	struct tulip_private *tp;
 
 	if (!dev)
 		return;
 
-	tp = netdev_priv(dev);
 	unregister_netdev(dev);
-	pci_free_consistent (pdev,
-			     sizeof (struct tulip_rx_desc) * RX_RING_SIZE +
-			     sizeof (struct tulip_tx_desc) * TX_RING_SIZE,
-			     tp->rx_ring, tp->rx_ring_dma);
-	kfree (tp->mtable);
-	pci_iounmap(pdev, tp->base_addr);
-	free_netdev (dev);
-	pci_release_regions (pdev);
-	pci_disable_device(pdev);
-
-	/* pci_power_off (pdev, -1); */
 }
 
 #ifdef CONFIG_NET_POLL_CONTROLLER
@@ -1958,24 +1893,19 @@ static void poll_tulip (struct net_device *dev)
 }
 #endif
 
+static SIMPLE_DEV_PM_OPS(tulip_pm_ops, tulip_suspend, tulip_resume);
+
 static struct pci_driver tulip_driver = {
 	.name		= DRV_NAME,
 	.id_table	= tulip_pci_tbl,
 	.probe		= tulip_init_one,
 	.remove		= tulip_remove_one,
-#ifdef CONFIG_PM
-	.suspend	= tulip_suspend,
-	.resume		= tulip_resume,
-#endif /* CONFIG_PM */
+	.driver.pm	= &tulip_pm_ops,
 };
 
 
 static int __init tulip_init (void)
 {
-#ifdef MODULE
-	pr_info("%s", version);
-#endif
-
 	if (!csr0) {
 		pr_warn("tulip: unknown CPU architecture, using default csr0\n");
 		/* default to 8 longword cache line alignment */
diff --git a/drivers/net/ethernet/dec/tulip/uli526x.c b/drivers/net/ethernet/dec/tulip/uli526x.c
index 8d98b259d1ba..6e4d8d31aba9 100644
--- a/drivers/net/ethernet/dec/tulip/uli526x.c
+++ b/drivers/net/ethernet/dec/tulip/uli526x.c
@@ -1,13 +1,5 @@
+// SPDX-License-Identifier: GPL-2.0-or-later
 /*
-    This program is free software; you can redistribute it and/or
-    modify it under the terms of the GNU General Public License
-    as published by the Free Software Foundation; either version 2
-    of the License, or (at your option) any later version.
-
-    This program is distributed in the hope that it will be useful,
-    but WITHOUT ANY WARRANTY; without even the implied warranty of
-    MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
-    GNU General Public License for more details.
 
 
 */
@@ -15,8 +7,6 @@
 #define pr_fmt(fmt) KBUILD_MODNAME ": " fmt
 
 #define DRV_NAME	"uli526x"
-#define DRV_VERSION	"0.9.3"
-#define DRV_RELDATE	"2005-7-29"
 
 #include <linux/module.h>
 
@@ -204,10 +194,6 @@ enum uli526x_CR6_bits {
 };
 
 /* Global variable declaration ----------------------------- */
-static int printed_version;
-static const char version[] =
-	"ULi M5261/M5263 net driver, version " DRV_VERSION " (" DRV_RELDATE ")";
-
 static int uli526x_debug;
 static unsigned char uli526x_media_mode = ULI526X_AUTO;
 static u32 uli526x_cr6_user_set;
@@ -241,7 +227,7 @@ static void phy_write_1bit(struct uli526x_board_info *db, u32);
 static u16 phy_read_1bit(struct uli526x_board_info *db);
 static u8 uli526x_sense_speed(struct uli526x_board_info *);
 static void uli526x_process_mode(struct uli526x_board_info *);
-static void uli526x_timer(unsigned long);
+static void uli526x_timer(struct timer_list *t);
 static void uli526x_rx_packet(struct net_device *, struct uli526x_board_info *);
 static void uli526x_free_tx_pkt(struct net_device *, struct uli526x_board_info *);
 static void uli526x_reuse_skb(struct uli526x_board_info *, struct sk_buff *);
@@ -286,20 +272,18 @@ static int uli526x_init_one(struct pci_dev *pdev,
 	struct uli526x_board_info *db;	/* board information structure */
 	struct net_device *dev;
 	void __iomem *ioaddr;
+	u8 addr[ETH_ALEN];
 	int i, err;
 
 	ULI526X_DBUG(0, "uli526x_init_one()", 0);
 
-	if (!printed_version++)
-		pr_info("%s\n", version);
-
 	/* Init network device */
 	dev = alloc_etherdev(sizeof(*db));
 	if (dev == NULL)
 		return -ENOMEM;
 	SET_NETDEV_DEV(dev, &pdev->dev);
 
-	if (pci_set_dma_mask(pdev, DMA_BIT_MASK(32))) {
+	if (dma_set_mask(&pdev->dev, DMA_BIT_MASK(32))) {
 		pr_warn("32-bit PCI DMA not available\n");
 		err = -ENODEV;
 		goto err_out_free;
@@ -334,11 +318,15 @@ static int uli526x_init_one(struct pci_dev *pdev,
 	/* Allocate Tx/Rx descriptor memory */
 	err = -ENOMEM;
 
-	db->desc_pool_ptr = pci_alloc_consistent(pdev, sizeof(struct tx_desc) * DESC_ALL_CNT + 0x20, &db->desc_pool_dma_ptr);
+	db->desc_pool_ptr = dma_alloc_coherent(&pdev->dev,
+					       sizeof(struct tx_desc) * DESC_ALL_CNT + 0x20,
+					       &db->desc_pool_dma_ptr, GFP_KERNEL);
 	if (!db->desc_pool_ptr)
 		goto err_out_release;
 
-	db->buf_pool_ptr = pci_alloc_consistent(pdev, TX_BUF_ALLOC * TX_DESC_CNT + 4, &db->buf_pool_dma_ptr);
+	db->buf_pool_ptr = dma_alloc_coherent(&pdev->dev,
+					      TX_BUF_ALLOC * TX_DESC_CNT + 4,
+					      &db->buf_pool_dma_ptr, GFP_KERNEL);
 	if (!db->buf_pool_ptr)
 		goto err_out_free_tx_desc;
 
@@ -392,7 +380,7 @@ static int uli526x_init_one(struct pci_dev *pdev,
 		uw32(DCR13, 0x1b0);	//Select ID Table access port
 		//Read MAC address from CR14
 		for (i = 0; i < 6; i++)
-			dev->dev_addr[i] = ur32(DCR14);
+			addr[i] = ur32(DCR14);
 		//Read end
 		uw32(DCR13, 0);		//Clear CR13
 		uw32(DCR0, 0);		//Clear CR0
@@ -401,8 +389,10 @@ static int uli526x_init_one(struct pci_dev *pdev,
 	else		/*Exist SROM*/
 	{
 		for (i = 0; i < 6; i++)
-			dev->dev_addr[i] = db->srom[20 + i];
+			addr[i] = db->srom[20 + i];
 	}
+	eth_hw_addr_set(dev, addr);
+
 	err = register_netdev (dev);
 	if (err)
 		goto err_out_unmap;
@@ -418,11 +408,12 @@ static int uli526x_init_one(struct pci_dev *pdev,
 err_out_unmap:
 	pci_iounmap(pdev, db->ioaddr);
 err_out_free_tx_buf:
-	pci_free_consistent(pdev, TX_BUF_ALLOC * TX_DESC_CNT + 4,
-			    db->buf_pool_ptr, db->buf_pool_dma_ptr);
+	dma_free_coherent(&pdev->dev, TX_BUF_ALLOC * TX_DESC_CNT + 4,
+			  db->buf_pool_ptr, db->buf_pool_dma_ptr);
 err_out_free_tx_desc:
-	pci_free_consistent(pdev, sizeof(struct tx_desc) * DESC_ALL_CNT + 0x20,
-			    db->desc_pool_ptr, db->desc_pool_dma_ptr);
+	dma_free_coherent(&pdev->dev,
+			  sizeof(struct tx_desc) * DESC_ALL_CNT + 0x20,
+			  db->desc_pool_ptr, db->desc_pool_dma_ptr);
 err_out_release:
 	pci_release_regions(pdev);
 err_out_disable:
@@ -441,11 +432,11 @@ static void uli526x_remove_one(struct pci_dev *pdev)
 
 	unregister_netdev(dev);
 	pci_iounmap(pdev, db->ioaddr);
-	pci_free_consistent(db->pdev, sizeof(struct tx_desc) *
-				DESC_ALL_CNT + 0x20, db->desc_pool_ptr,
- 				db->desc_pool_dma_ptr);
-	pci_free_consistent(db->pdev, TX_BUF_ALLOC * TX_DESC_CNT + 4,
-				db->buf_pool_ptr, db->buf_pool_dma_ptr);
+	dma_free_coherent(&db->pdev->dev,
+			  sizeof(struct tx_desc) * DESC_ALL_CNT + 0x20,
+			  db->desc_pool_ptr, db->desc_pool_dma_ptr);
+	dma_free_coherent(&db->pdev->dev, TX_BUF_ALLOC * TX_DESC_CNT + 4,
+			  db->buf_pool_ptr, db->buf_pool_dma_ptr);
 	pci_release_regions(pdev);
 	pci_disable_device(pdev);
 	free_netdev(dev);
@@ -491,10 +482,8 @@ static int uli526x_open(struct net_device *dev)
 	netif_wake_queue(dev);
 
 	/* set and active a timer process */
-	init_timer(&db->timer);
+	timer_setup(&db->timer, uli526x_timer, 0);
 	db->timer.expires = ULI526X_TIMER_WUT + HZ * 2;
-	db->timer.data = (unsigned long)dev;
-	db->timer.function = uli526x_timer;
 	add_timer(&db->timer);
 
 	return 0;
@@ -667,7 +656,7 @@ static int uli526x_stop(struct net_device *dev)
 	netif_stop_queue(dev);
 
 	/* deleted timer */
-	del_timer_sync(&db->timer);
+	timer_delete_sync(&db->timer);
 
 	/* Reset & stop ULI526X board */
 	uw32(DCR0, ULI526X_RESET);
@@ -794,7 +783,7 @@ static void uli526x_free_tx_pkt(struct net_device *dev,
 			}
 		}
 
-    		txptr = txptr->next_tx_desc;
+		txptr = txptr->next_tx_desc;
 	}/* End of while */
 
 	/* Update TX remove pointer to next */
@@ -829,7 +818,8 @@ static void uli526x_rx_packet(struct net_device *dev, struct uli526x_board_info
 		db->rx_avail_cnt--;
 		db->interval_rx_cnt++;
 
-		pci_unmap_single(db->pdev, le32_to_cpu(rxptr->rdes2), RX_ALLOC_SIZE, PCI_DMA_FROMDEVICE);
+		dma_unmap_single(&db->pdev->dev, le32_to_cpu(rxptr->rdes2),
+				 RX_ALLOC_SIZE, DMA_FROM_DEVICE);
 		if ( (rdes0 & 0x300) != 0x300) {
 			/* A packet without First/Last flag */
 			/* reuse this SKB */
@@ -864,9 +854,9 @@ static void uli526x_rx_packet(struct net_device *dev, struct uli526x_board_info
 					skb = new_skb;
 					/* size less than COPY_SIZE, allocate a rxlen SKB */
 					skb_reserve(skb, 2); /* 16byte align */
-					memcpy(skb_put(skb, rxlen),
-					       skb_tail_pointer(rxptr->rx_skb_ptr),
-					       rxlen);
+					skb_put_data(skb,
+						     skb_tail_pointer(rxptr->rx_skb_ptr),
+						     rxlen);
 					uli526x_reuse_skb(db, rxptr->rx_skb_ptr);
 				} else
 					skb_put(skb, rxlen);
@@ -981,9 +971,8 @@ static void netdev_get_drvinfo(struct net_device *dev,
 {
 	struct uli526x_board_info *np = netdev_priv(dev);
 
-	strlcpy(info->driver, DRV_NAME, sizeof(info->driver));
-	strlcpy(info->version, DRV_VERSION, sizeof(info->version));
-	strlcpy(info->bus_info, pci_name(np->pdev), sizeof(info->bus_info));
+	strscpy(info->driver, DRV_NAME, sizeof(info->driver));
+	strscpy(info->bus_info, pci_name(np->pdev), sizeof(info->bus_info));
 }
 
 static int netdev_get_link_ksettings(struct net_device *dev,
@@ -1023,13 +1012,13 @@ static const struct ethtool_ops netdev_ethtool_ops = {
  *	Dynamic media sense, allocate Rx buffer...
  */
 
-static void uli526x_timer(unsigned long data)
+static void uli526x_timer(struct timer_list *t)
 {
-	struct net_device *dev = (struct net_device *) data;
-	struct uli526x_board_info *db = netdev_priv(dev);
+	struct uli526x_board_info *db = timer_container_of(db, t, timer);
+	struct net_device *dev = pci_get_drvdata(db->pdev);
 	struct uli_phy_ops *phy = &db->phy;
 	void __iomem *ioaddr = db->ioaddr;
- 	unsigned long flags;
+	unsigned long flags;
 	u8 tmp_cr12 = 0;
 	u32 tmp_cr8;
 
@@ -1183,65 +1172,41 @@ static void uli526x_dynamic_reset(struct net_device *dev)
 	netif_wake_queue(dev);
 }
 
-
-#ifdef CONFIG_PM
-
 /*
  *	Suspend the interface.
  */
 
-static int uli526x_suspend(struct pci_dev *pdev, pm_message_t state)
+static int __maybe_unused uli526x_suspend(struct device *dev_d)
 {
-	struct net_device *dev = pci_get_drvdata(pdev);
-	pci_power_t power_state;
-	int err;
+	struct net_device *dev = dev_get_drvdata(dev_d);
 
 	ULI526X_DBUG(0, "uli526x_suspend", 0);
 
-	pci_save_state(pdev);
-
 	if (!netif_running(dev))
 		return 0;
 
 	netif_device_detach(dev);
 	uli526x_reset_prepare(dev);
 
-	power_state = pci_choose_state(pdev, state);
-	pci_enable_wake(pdev, power_state, 0);
-	err = pci_set_power_state(pdev, power_state);
-	if (err) {
-		netif_device_attach(dev);
-		/* Re-initialize ULI526X board */
-		uli526x_init(dev);
-		/* Restart upper layer interface */
-		netif_wake_queue(dev);
-	}
+	device_set_wakeup_enable(dev_d, 0);
 
-	return err;
+	return 0;
 }
 
 /*
  *	Resume the interface.
  */
 
-static int uli526x_resume(struct pci_dev *pdev)
+static int __maybe_unused uli526x_resume(struct device *dev_d)
 {
-	struct net_device *dev = pci_get_drvdata(pdev);
-	int err;
+	struct net_device *dev = dev_get_drvdata(dev_d);
 
 	ULI526X_DBUG(0, "uli526x_resume", 0);
 
-	pci_restore_state(pdev);
 
 	if (!netif_running(dev))
 		return 0;
 
-	err = pci_set_power_state(pdev, PCI_D0);
-	if (err) {
-		netdev_warn(dev, "Could not put device into D0\n");
-		return err;
-	}
-
 	netif_device_attach(dev);
 	/* Re-initialize ULI526X board */
 	uli526x_init(dev);
@@ -1251,14 +1216,6 @@ static int uli526x_resume(struct pci_dev *pdev)
 	return 0;
 }
 
-#else /* !CONFIG_PM */
-
-#define uli526x_suspend	NULL
-#define uli526x_resume	NULL
-
-#endif /* !CONFIG_PM */
-
-
 /*
  *	free all allocated rx buffer
  */
@@ -1286,10 +1243,8 @@ static void uli526x_reuse_skb(struct uli526x_board_info *db, struct sk_buff * sk
 
 	if (!(rxptr->rdes0 & cpu_to_le32(0x80000000))) {
 		rxptr->rx_skb_ptr = skb;
-		rxptr->rdes2 = cpu_to_le32(pci_map_single(db->pdev,
-							  skb_tail_pointer(skb),
-							  RX_ALLOC_SIZE,
-							  PCI_DMA_FROMDEVICE));
+		rxptr->rdes2 = cpu_to_le32(dma_map_single(&db->pdev->dev, skb_tail_pointer(skb),
+							  RX_ALLOC_SIZE, DMA_FROM_DEVICE));
 		wmb();
 		rxptr->rdes0 = cpu_to_le32(0x80000000);
 		db->rx_avail_cnt++;
@@ -1391,7 +1346,7 @@ static void send_filter_frame(struct net_device *dev, int mc_cnt)
 	void __iomem *ioaddr = db->ioaddr;
 	struct netdev_hw_addr *ha;
 	struct tx_desc *txptr;
-	u16 * addrptr;
+	const u16 * addrptr;
 	u32 * suptr;
 	int i;
 
@@ -1401,7 +1356,7 @@ static void send_filter_frame(struct net_device *dev, int mc_cnt)
 	suptr = (u32 *) txptr->tx_buf_ptr;
 
 	/* Node address */
-	addrptr = (u16 *) dev->dev_addr;
+	addrptr = (const u16 *) dev->dev_addr;
 	*suptr++ = addrptr[0] << FLT_SHIFT;
 	*suptr++ = addrptr[1] << FLT_SHIFT;
 	*suptr++ = addrptr[2] << FLT_SHIFT;
@@ -1461,10 +1416,8 @@ static void allocate_rx_buffer(struct net_device *dev)
 		if (skb == NULL)
 			break;
 		rxptr->rx_skb_ptr = skb; /* FIXME (?) */
-		rxptr->rdes2 = cpu_to_le32(pci_map_single(db->pdev,
-							  skb_tail_pointer(skb),
-							  RX_ALLOC_SIZE,
-							  PCI_DMA_FROMDEVICE));
+		rxptr->rdes2 = cpu_to_le32(dma_map_single(&db->pdev->dev, skb_tail_pointer(skb),
+							  RX_ALLOC_SIZE, DMA_FROM_DEVICE));
 		wmb();
 		rxptr->rdes0 = cpu_to_le32(0x80000000);
 		rxptr = rxptr->next_rx_desc;
@@ -1585,14 +1538,14 @@ static void uli526x_set_phyxcer(struct uli526x_board_info *db)
 
 	}
 
-  	/* Write new capability to Phyxcer Reg4 */
+	/* Write new capability to Phyxcer Reg4 */
 	if ( !(phy_reg & 0x01e0)) {
 		phy_reg|=db->PHY_reg4;
 		db->media_mode|=ULI526X_AUTO;
 	}
 	phy->write(db, db->phy_addr, 4, phy_reg);
 
- 	/* Restart Auto-Negotiation */
+	/* Restart Auto-Negotiation */
 	phy->write(db, db->phy_addr, 0, 0x1200);
 	udelay(50);
 }
@@ -1600,7 +1553,7 @@ static void uli526x_set_phyxcer(struct uli526x_board_info *db)
 
 /*
  *	Process op-mode
- 	AUTO mode : PHY controller in Auto-negotiation Mode
+	AUTO mode : PHY controller in Auto-negotiation Mode
  *	Force mode: PHY controller in force mode with HUB
  *			N-way force capability with SWITCH
  */
@@ -1781,14 +1734,14 @@ static const struct pci_device_id uli526x_pci_tbl[] = {
 };
 MODULE_DEVICE_TABLE(pci, uli526x_pci_tbl);
 
+static SIMPLE_DEV_PM_OPS(uli526x_pm_ops, uli526x_suspend, uli526x_resume);
 
 static struct pci_driver uli526x_driver = {
 	.name		= "uli526x",
 	.id_table	= uli526x_pci_tbl,
 	.probe		= uli526x_init_one,
 	.remove		= uli526x_remove_one,
-	.suspend	= uli526x_suspend,
-	.resume		= uli526x_resume,
+	.driver.pm	= &uli526x_pm_ops,
 };
 
 MODULE_AUTHOR("Peer Chen, peer.chen@uli.com.tw");
@@ -1809,9 +1762,6 @@ MODULE_PARM_DESC(mode, "ULi M5261/M5263: Bit 0: 10/100Mbps, bit 2: duplex, bit 8
 static int __init uli526x_init_module(void)
 {
 
-	pr_info("%s\n", version);
-	printed_version = 1;
-
 	ULI526X_DBUG(0, "init_module() ", debug);
 
 	if (debug)
@@ -1819,8 +1769,8 @@ static int __init uli526x_init_module(void)
 	if (cr6set)
 		uli526x_cr6_user_set = cr6set;
 
- 	switch (mode) {
-   	case ULI526X_10MHF:
+	switch (mode) {
+	case ULI526X_10MHF:
 	case ULI526X_100MHF:
 	case ULI526X_10MFD:
 	case ULI526X_100MFD:
diff --git a/drivers/net/ethernet/dec/tulip/winbond-840.c b/drivers/net/ethernet/dec/tulip/winbond-840.c
index d1f2f3cc7cfa..a24a25a5f73d 100644
--- a/drivers/net/ethernet/dec/tulip/winbond-840.c
+++ b/drivers/net/ethernet/dec/tulip/winbond-840.c
@@ -36,7 +36,7 @@
 		power management.
 		support for big endian descriptors
 			Copyright (C) 2001 Manfred Spraul
-  	* ethtool support (jgarzik)
+	* ethtool support (jgarzik)
 	* Replace some MII-related magic numbers with constants (jgarzik)
 
 	TODO:
@@ -47,9 +47,6 @@
 #define pr_fmt(fmt) KBUILD_MODNAME ": " fmt
 
 #define DRV_NAME	"winbond-840"
-#define DRV_VERSION	"1.01-e"
-#define DRV_RELDATE	"Sep-11-2006"
-
 
 /* Automatically extracted configuration info:
 probe-func: winbond840_probe
@@ -139,16 +136,9 @@ static int full_duplex[MAX_UNITS] = {-1, -1, -1, -1, -1, -1, -1, -1};
 #undef PKT_BUF_SZ			/* tulip.h also defines this */
 #define PKT_BUF_SZ		1536	/* Size of each temporary Rx buffer.*/
 
-/* These identify the driver base version and may not be removed. */
-static const char version[] __initconst =
-	"v" DRV_VERSION " (2.4 port) "
-	DRV_RELDATE "  Donald Becker <becker@scyld.com>\n"
-	"  http://www.scyld.com/network/drivers.html\n";
-
 MODULE_AUTHOR("Donald Becker <becker@scyld.com>");
 MODULE_DESCRIPTION("Winbond W89c840 Ethernet driver");
 MODULE_LICENSE("GPL");
-MODULE_VERSION(DRV_VERSION);
 
 module_param(max_interrupt_work, int, 0);
 module_param(debug, int, 0);
@@ -327,11 +317,11 @@ static int  mdio_read(struct net_device *dev, int phy_id, int location);
 static void mdio_write(struct net_device *dev, int phy_id, int location, int value);
 static int  netdev_open(struct net_device *dev);
 static int  update_link(struct net_device *dev);
-static void netdev_timer(unsigned long data);
+static void netdev_timer(struct timer_list *t);
 static void init_rxtx_rings(struct net_device *dev);
 static void free_rxtx_rings(struct netdev_private *np);
 static void init_registers(struct net_device *dev);
-static void tx_timeout(struct net_device *dev);
+static void tx_timeout(struct net_device *dev, unsigned int txqueue);
 static int alloc_ringdesc(struct net_device *dev);
 static void free_ringdesc(struct netdev_private *np);
 static netdev_tx_t start_tx(struct sk_buff *skb, struct net_device *dev);
@@ -351,7 +341,7 @@ static const struct net_device_ops netdev_ops = {
 	.ndo_start_xmit		= start_tx,
 	.ndo_get_stats		= get_stats,
 	.ndo_set_rx_mode	= set_rx_mode,
-	.ndo_do_ioctl		= netdev_ioctl,
+	.ndo_eth_ioctl		= netdev_ioctl,
 	.ndo_tx_timeout		= tx_timeout,
 	.ndo_set_mac_address	= eth_mac_addr,
 	.ndo_validate_addr	= eth_validate_addr,
@@ -365,16 +355,17 @@ static int w840_probe1(struct pci_dev *pdev, const struct pci_device_id *ent)
 	int chip_idx = ent->driver_data;
 	int irq;
 	int i, option = find_cnt < MAX_UNITS ? options[find_cnt] : 0;
+	__le16 addr[ETH_ALEN / 2];
 	void __iomem *ioaddr;
 
-	i = pci_enable_device(pdev);
+	i = pcim_enable_device(pdev);
 	if (i) return i;
 
 	pci_set_master(pdev);
 
 	irq = pdev->irq;
 
-	if (pci_set_dma_mask(pdev, DMA_BIT_MASK(32))) {
+	if (dma_set_mask(&pdev->dev, DMA_BIT_MASK(32))) {
 		pr_warn("Device %s disabled due to DMA limitations\n",
 			pci_name(pdev));
 		return -EIO;
@@ -384,15 +375,16 @@ static int w840_probe1(struct pci_dev *pdev, const struct pci_device_id *ent)
 		return -ENOMEM;
 	SET_NETDEV_DEV(dev, &pdev->dev);
 
-	if (pci_request_regions(pdev, DRV_NAME))
+	if (pcim_request_all_regions(pdev, DRV_NAME))
 		goto err_out_netdev;
 
 	ioaddr = pci_iomap(pdev, TULIP_BAR, netdev_res_size);
 	if (!ioaddr)
-		goto err_out_free_res;
+		goto err_out_netdev;
 
 	for (i = 0; i < 3; i++)
-		((__le16 *)dev->dev_addr)[i] = cpu_to_le16(eeprom_read(ioaddr, i));
+		addr[i] = cpu_to_le16(eeprom_read(ioaddr, i));
+	eth_hw_addr_set(dev, (u8 *)addr);
 
 	/* Reset the chip to erase previous misconfiguration.
 	   No hold time required! */
@@ -468,8 +460,6 @@ static int w840_probe1(struct pci_dev *pdev, const struct pci_device_id *ent)
 
 err_out_cleardev:
 	pci_iounmap(pdev, ioaddr);
-err_out_free_res:
-	pci_release_regions(pdev);
 err_out_netdev:
 	free_netdev (dev);
 	return -ENODEV;
@@ -484,8 +474,6 @@ err_out_netdev:
    No extra delay is needed with 33Mhz PCI, but future 66Mhz access may need
    a delay.  Note that pre-2.0.34 kernels had a cache-alignment bug that
    made udelay() unreliable.
-   The old method of using an ISA access as a delay, __SLOW_DOWN_IO__, is
-   deprecated.
 */
 #define eeprom_delay(ee_addr)	ioread32(ee_addr)
 
@@ -640,9 +628,10 @@ static int netdev_open(struct net_device *dev)
 		goto out_err;
 
 	if (debug > 1)
-		netdev_dbg(dev, "w89c840_open() irq %d\n", irq);
+		netdev_dbg(dev, "%s() irq %d\n", __func__, irq);
 
-	if((i=alloc_ringdesc(dev)))
+	i = alloc_ringdesc(dev);
+	if (i)
 		goto out_err;
 
 	spin_lock_irq(&np->lock);
@@ -652,13 +641,11 @@ static int netdev_open(struct net_device *dev)
 
 	netif_start_queue(dev);
 	if (debug > 2)
-		netdev_dbg(dev, "Done netdev_open()\n");
+		netdev_dbg(dev, "Done %s()\n", __func__);
 
 	/* Set the timer to check for link beat. */
-	init_timer(&np->timer);
+	timer_setup(&np->timer, netdev_timer, 0);
 	np->timer.expires = jiffies + 1*HZ;
-	np->timer.data = (unsigned long)dev;
-	np->timer.function = netdev_timer;				/* timer handler */
 	add_timer(&np->timer);
 	return 0;
 out_err:
@@ -774,10 +761,10 @@ static inline void update_csr6(struct net_device *dev, int new)
 		np->mii_if.full_duplex = 1;
 }
 
-static void netdev_timer(unsigned long data)
+static void netdev_timer(struct timer_list *t)
 {
-	struct net_device *dev = (struct net_device *)data;
-	struct netdev_private *np = netdev_priv(dev);
+	struct netdev_private *np = timer_container_of(np, t, timer);
+	struct net_device *dev = pci_get_drvdata(np->pci_dev);
 	void __iomem *ioaddr = np->base_addr;
 
 	if (debug > 2)
@@ -814,8 +801,9 @@ static void init_rxtx_rings(struct net_device *dev)
 		np->rx_skbuff[i] = skb;
 		if (skb == NULL)
 			break;
-		np->rx_addr[i] = pci_map_single(np->pci_dev,skb->data,
-					np->rx_buf_sz,PCI_DMA_FROMDEVICE);
+		np->rx_addr[i] = dma_map_single(&np->pci_dev->dev, skb->data,
+						np->rx_buf_sz,
+						DMA_FROM_DEVICE);
 
 		np->rx_ring[i].buffer1 = np->rx_addr[i];
 		np->rx_ring[i].status = DescOwned;
@@ -845,20 +833,17 @@ static void free_rxtx_rings(struct netdev_private* np)
 	for (i = 0; i < RX_RING_SIZE; i++) {
 		np->rx_ring[i].status = 0;
 		if (np->rx_skbuff[i]) {
-			pci_unmap_single(np->pci_dev,
-						np->rx_addr[i],
-						np->rx_skbuff[i]->len,
-						PCI_DMA_FROMDEVICE);
+			dma_unmap_single(&np->pci_dev->dev, np->rx_addr[i],
+					 np->rx_skbuff[i]->len,
+					 DMA_FROM_DEVICE);
 			dev_kfree_skb(np->rx_skbuff[i]);
 		}
 		np->rx_skbuff[i] = NULL;
 	}
 	for (i = 0; i < TX_RING_SIZE; i++) {
 		if (np->tx_skbuff[i]) {
-			pci_unmap_single(np->pci_dev,
-						np->tx_addr[i],
-						np->tx_skbuff[i]->len,
-						PCI_DMA_TODEVICE);
+			dma_unmap_single(&np->pci_dev->dev, np->tx_addr[i],
+					 np->tx_skbuff[i]->len, DMA_TO_DEVICE);
 			dev_kfree_skb(np->tx_skbuff[i]);
 		}
 		np->tx_skbuff[i] = NULL;
@@ -892,7 +877,7 @@ static void init_registers(struct net_device *dev)
 		8000	16 longwords		0200 2 longwords	2000 32 longwords
 		C000	32  longwords		0400 4 longwords */
 
-#if defined (__i386__) && !defined(MODULE)
+#if defined (__i386__) && !defined(MODULE) && !defined(CONFIG_UML)
 	/* When not a module we can work around broken '486 PCI boards. */
 	if (boot_cpu_data.x86 <= 4) {
 		i |= 0x4800;
@@ -923,7 +908,7 @@ static void init_registers(struct net_device *dev)
 	iowrite32(0, ioaddr + RxStartDemand);
 }
 
-static void tx_timeout(struct net_device *dev)
+static void tx_timeout(struct net_device *dev, unsigned int txqueue)
 {
 	struct netdev_private *np = netdev_priv(dev);
 	void __iomem *ioaddr = np->base_addr;
@@ -976,10 +961,10 @@ static int alloc_ringdesc(struct net_device *dev)
 
 	np->rx_buf_sz = (dev->mtu <= 1500 ? PKT_BUF_SZ : dev->mtu + 32);
 
-	np->rx_ring = pci_alloc_consistent(np->pci_dev,
-			sizeof(struct w840_rx_desc)*RX_RING_SIZE +
-			sizeof(struct w840_tx_desc)*TX_RING_SIZE,
-			&np->ring_dma_addr);
+	np->rx_ring = dma_alloc_coherent(&np->pci_dev->dev,
+					 sizeof(struct w840_rx_desc) * RX_RING_SIZE +
+					 sizeof(struct w840_tx_desc) * TX_RING_SIZE,
+					 &np->ring_dma_addr, GFP_KERNEL);
 	if(!np->rx_ring)
 		return -ENOMEM;
 	init_rxtx_rings(dev);
@@ -988,10 +973,10 @@ static int alloc_ringdesc(struct net_device *dev)
 
 static void free_ringdesc(struct netdev_private *np)
 {
-	pci_free_consistent(np->pci_dev,
-			sizeof(struct w840_rx_desc)*RX_RING_SIZE +
-			sizeof(struct w840_tx_desc)*TX_RING_SIZE,
-			np->rx_ring, np->ring_dma_addr);
+	dma_free_coherent(&np->pci_dev->dev,
+			  sizeof(struct w840_rx_desc) * RX_RING_SIZE +
+			  sizeof(struct w840_tx_desc) * TX_RING_SIZE,
+			  np->rx_ring, np->ring_dma_addr);
 
 }
 
@@ -1006,8 +991,8 @@ static netdev_tx_t start_tx(struct sk_buff *skb, struct net_device *dev)
 	/* Calculate the next Tx descriptor entry. */
 	entry = np->cur_tx % TX_RING_SIZE;
 
-	np->tx_addr[entry] = pci_map_single(np->pci_dev,
-				skb->data,skb->len, PCI_DMA_TODEVICE);
+	np->tx_addr[entry] = dma_map_single(&np->pci_dev->dev, skb->data,
+					    skb->len, DMA_TO_DEVICE);
 	np->tx_skbuff[entry] = skb;
 
 	np->tx_ring[entry].buffer1 = np->tx_addr[entry];
@@ -1090,9 +1075,8 @@ static void netdev_tx_done(struct net_device *dev)
 			np->stats.tx_packets++;
 		}
 		/* Free the original skb. */
-		pci_unmap_single(np->pci_dev,np->tx_addr[entry],
-					np->tx_skbuff[entry]->len,
-					PCI_DMA_TODEVICE);
+		dma_unmap_single(&np->pci_dev->dev, np->tx_addr[entry],
+				 np->tx_skbuff[entry]->len, DMA_TO_DEVICE);
 		np->tx_q_bytes -= np->tx_skbuff[entry]->len;
 		dev_kfree_skb_irq(np->tx_skbuff[entry]);
 		np->tx_skbuff[entry] = NULL;
@@ -1229,18 +1213,21 @@ static int netdev_rx(struct net_device *dev)
 			if (pkt_len < rx_copybreak &&
 			    (skb = netdev_alloc_skb(dev, pkt_len + 2)) != NULL) {
 				skb_reserve(skb, 2);	/* 16 byte align the IP header */
-				pci_dma_sync_single_for_cpu(np->pci_dev,np->rx_addr[entry],
-							    np->rx_skbuff[entry]->len,
-							    PCI_DMA_FROMDEVICE);
+				dma_sync_single_for_cpu(&np->pci_dev->dev,
+							np->rx_addr[entry],
+							np->rx_skbuff[entry]->len,
+							DMA_FROM_DEVICE);
 				skb_copy_to_linear_data(skb, np->rx_skbuff[entry]->data, pkt_len);
 				skb_put(skb, pkt_len);
-				pci_dma_sync_single_for_device(np->pci_dev,np->rx_addr[entry],
-							       np->rx_skbuff[entry]->len,
-							       PCI_DMA_FROMDEVICE);
+				dma_sync_single_for_device(&np->pci_dev->dev,
+							   np->rx_addr[entry],
+							   np->rx_skbuff[entry]->len,
+							   DMA_FROM_DEVICE);
 			} else {
-				pci_unmap_single(np->pci_dev,np->rx_addr[entry],
-							np->rx_skbuff[entry]->len,
-							PCI_DMA_FROMDEVICE);
+				dma_unmap_single(&np->pci_dev->dev,
+						 np->rx_addr[entry],
+						 np->rx_skbuff[entry]->len,
+						 DMA_FROM_DEVICE);
 				skb_put(skb = np->rx_skbuff[entry], pkt_len);
 				np->rx_skbuff[entry] = NULL;
 			}
@@ -1270,9 +1257,10 @@ static int netdev_rx(struct net_device *dev)
 			np->rx_skbuff[entry] = skb;
 			if (skb == NULL)
 				break;			/* Better luck next round. */
-			np->rx_addr[entry] = pci_map_single(np->pci_dev,
-							skb->data,
-							np->rx_buf_sz, PCI_DMA_FROMDEVICE);
+			np->rx_addr[entry] = dma_map_single(&np->pci_dev->dev,
+							    skb->data,
+							    np->rx_buf_sz,
+							    DMA_FROM_DEVICE);
 			np->rx_ring[entry].buffer1 = np->rx_addr[entry];
 		}
 		wmb();
@@ -1386,22 +1374,20 @@ static void netdev_get_drvinfo (struct net_device *dev, struct ethtool_drvinfo *
 {
 	struct netdev_private *np = netdev_priv(dev);
 
-	strlcpy(info->driver, DRV_NAME, sizeof(info->driver));
-	strlcpy(info->version, DRV_VERSION, sizeof(info->version));
-	strlcpy(info->bus_info, pci_name(np->pci_dev), sizeof(info->bus_info));
+	strscpy(info->driver, DRV_NAME, sizeof(info->driver));
+	strscpy(info->bus_info, pci_name(np->pci_dev), sizeof(info->bus_info));
 }
 
 static int netdev_get_link_ksettings(struct net_device *dev,
 				     struct ethtool_link_ksettings *cmd)
 {
 	struct netdev_private *np = netdev_priv(dev);
-	int rc;
 
 	spin_lock_irq(&np->lock);
-	rc = mii_ethtool_get_link_ksettings(&np->mii_if, cmd);
+	mii_ethtool_get_link_ksettings(&np->mii_if, cmd);
 	spin_unlock_irq(&np->lock);
 
-	return rc;
+	return 0;
 }
 
 static int netdev_set_link_ksettings(struct net_device *dev,
@@ -1457,7 +1443,7 @@ static int netdev_ioctl(struct net_device *dev, struct ifreq *rq, int cmd)
 	switch(cmd) {
 	case SIOCGMIIPHY:		/* Get address of MII PHY in use. */
 		data->phy_id = ((struct netdev_private *)netdev_priv(dev))->phys[0] & 0x1f;
-		/* Fall Through */
+		fallthrough;
 
 	case SIOCGMIIREG:		/* Read MII PHY register. */
 		spin_lock_irq(&np->lock);
@@ -1491,7 +1477,7 @@ static int netdev_close(struct net_device *dev)
 			   np->cur_rx, np->dirty_rx);
 	}
 
- 	/* Stop the chip's Tx and Rx processes. */
+	/* Stop the chip's Tx and Rx processes. */
 	spin_lock_irq(&np->lock);
 	netif_device_detach(dev);
 	update_csr6(dev, 0);
@@ -1523,7 +1509,7 @@ static int netdev_close(struct net_device *dev)
 	}
 #endif /* __i386__ debugging only */
 
-	del_timer_sync(&np->timer);
+	timer_delete_sync(&np->timer);
 
 	free_rxtx_rings(np);
 	free_ringdesc(np);
@@ -1538,14 +1524,11 @@ static void w840_remove1(struct pci_dev *pdev)
 	if (dev) {
 		struct netdev_private *np = netdev_priv(dev);
 		unregister_netdev(dev);
-		pci_release_regions(pdev);
 		pci_iounmap(pdev, np->base_addr);
 		free_netdev(dev);
 	}
 }
 
-#ifdef CONFIG_PM
-
 /*
  * suspend/resume synchronization:
  * - open, close, do_ioctl:
@@ -1569,15 +1552,15 @@ static void w840_remove1(struct pci_dev *pdev)
  * Detach must occur under spin_unlock_irq(), interrupts from a detached
  * device would cause an irq storm.
  */
-static int w840_suspend (struct pci_dev *pdev, pm_message_t state)
+static int __maybe_unused w840_suspend(struct device *dev_d)
 {
-	struct net_device *dev = pci_get_drvdata (pdev);
+	struct net_device *dev = dev_get_drvdata(dev_d);
 	struct netdev_private *np = netdev_priv(dev);
 	void __iomem *ioaddr = np->base_addr;
 
 	rtnl_lock();
 	if (netif_running (dev)) {
-		del_timer_sync(&np->timer);
+		timer_delete_sync(&np->timer);
 
 		spin_lock_irq(&np->lock);
 		netif_device_detach(dev);
@@ -1604,21 +1587,15 @@ static int w840_suspend (struct pci_dev *pdev, pm_message_t state)
 	return 0;
 }
 
-static int w840_resume (struct pci_dev *pdev)
+static int __maybe_unused w840_resume(struct device *dev_d)
 {
-	struct net_device *dev = pci_get_drvdata (pdev);
+	struct net_device *dev = dev_get_drvdata(dev_d);
 	struct netdev_private *np = netdev_priv(dev);
-	int retval = 0;
 
 	rtnl_lock();
 	if (netif_device_present(dev))
 		goto out; /* device not suspended */
 	if (netif_running(dev)) {
-		if ((retval = pci_enable_device(pdev))) {
-			dev_err(&dev->dev,
-				"pci_enable_device failed in resume\n");
-			goto out;
-		}
 		spin_lock_irq(&np->lock);
 		iowrite32(1, np->base_addr+PCIBusCfg);
 		ioread32(np->base_addr+PCIBusCfg);
@@ -1636,31 +1613,17 @@ static int w840_resume (struct pci_dev *pdev)
 	}
 out:
 	rtnl_unlock();
-	return retval;
+	return 0;
 }
-#endif
+
+static SIMPLE_DEV_PM_OPS(w840_pm_ops, w840_suspend, w840_resume);
 
 static struct pci_driver w840_driver = {
 	.name		= DRV_NAME,
 	.id_table	= w840_pci_tbl,
 	.probe		= w840_probe1,
 	.remove		= w840_remove1,
-#ifdef CONFIG_PM
-	.suspend	= w840_suspend,
-	.resume		= w840_resume,
-#endif
+	.driver.pm	= &w840_pm_ops,
 };
 
-static int __init w840_init(void)
-{
-	printk(version);
-	return pci_register_driver(&w840_driver);
-}
-
-static void __exit w840_exit(void)
-{
-	pci_unregister_driver(&w840_driver);
-}
-
-module_init(w840_init);
-module_exit(w840_exit);
+module_pci_driver(w840_driver);
diff --git a/drivers/net/ethernet/dec/tulip/xircom_cb.c b/drivers/net/ethernet/dec/tulip/xircom_cb.c
index a8de79355578..e5d2ede13845 100644
--- a/drivers/net/ethernet/dec/tulip/xircom_cb.c
+++ b/drivers/net/ethernet/dec/tulip/xircom_cb.c
@@ -143,7 +143,7 @@ static const struct pci_device_id xircom_pci_table[] = {
 };
 MODULE_DEVICE_TABLE(pci, xircom_pci_table);
 
-static struct pci_driver xircom_ops = {
+static struct pci_driver xircom_driver = {
 	.name		= "xircom_cb",
 	.id_table	= xircom_pci_table,
 	.probe		= xircom_probe,
@@ -1015,12 +1015,14 @@ static void read_mac_address(struct xircom_private *card)
 		xw32(CSR10, i + 3);
 		data_count = xr32(CSR9);
 		if ((tuple == 0x22) && (data_id == 0x04) && (data_count == 0x06)) {
+			u8 addr[ETH_ALEN];
 			int j;
 
 			for (j = 0; j < 6; j++) {
 				xw32(CSR10, i + j + 4);
-				card->dev->dev_addr[j] = xr32(CSR9) & 0xff;
+				addr[j] = xr32(CSR9) & 0xff;
 			}
+			eth_hw_addr_set(card->dev, addr);
 			break;
 		} else if (link == 0) {
 			break;
@@ -1167,4 +1169,4 @@ investigate_write_descriptor(struct net_device *dev,
 	}
 }
 
-module_pci_driver(xircom_ops);
+module_pci_driver(xircom_driver);
diff --git a/drivers/net/ethernet/dlink/Kconfig b/drivers/net/ethernet/dlink/Kconfig
index ebdc83247bb6..0d77f84c8e7b 100644
--- a/drivers/net/ethernet/dlink/Kconfig
+++ b/drivers/net/ethernet/dlink/Kconfig
@@ -1,3 +1,4 @@
+# SPDX-License-Identifier: GPL-2.0-only
 #
 # D-Link device configuration
 #
@@ -6,7 +7,7 @@ config NET_VENDOR_DLINK
 	bool "D-Link devices"
 	default y
 	depends on PCI
-	---help---
+	help
 	  If you have a network (Ethernet) card belonging to this class, say Y.
 
 	  Note that the answer to this question doesn't directly affect the
@@ -20,7 +21,7 @@ config DL2K
 	tristate "DL2000/TC902x/IP1000A-based Gigabit Ethernet support"
 	depends on PCI
 	select CRC32
-	---help---
+	help
 	  This driver supports DL2000/TC902x/IP1000A-based Gigabit ethernet cards,
 	  which includes
 	  D-Link DGE-550T Gigabit Ethernet Adapter.
@@ -36,7 +37,7 @@ config SUNDANCE
 	depends on PCI
 	select CRC32
 	select MII
-	---help---
+	help
 	  This driver is for the Sundance "Alta" chip.
 	  More specific information and updates are available from
 	  <http://www.scyld.com/network/sundance.html>.
@@ -44,7 +45,7 @@ config SUNDANCE
 config SUNDANCE_MMIO
 	bool "Use MMIO instead of PIO"
 	depends on SUNDANCE
-	---help---
+	help
 	  Enable memory-mapped I/O for interaction with Sundance NIC registers.
 	  Do NOT enable this by default, PIO (enabled when MMIO is disabled)
 	  is known to solve bugs on certain chips.
diff --git a/drivers/net/ethernet/dlink/Makefile b/drivers/net/ethernet/dlink/Makefile
index 40085f67157b..3ff503c747db 100644
--- a/drivers/net/ethernet/dlink/Makefile
+++ b/drivers/net/ethernet/dlink/Makefile
@@ -1,3 +1,4 @@
+# SPDX-License-Identifier: GPL-2.0-only
 #
 # Makefile for the D-Link network device drivers.
 #
diff --git a/drivers/net/ethernet/dlink/dl2k.c b/drivers/net/ethernet/dlink/dl2k.c
index 778f974e2928..1996d2e4e3e2 100644
--- a/drivers/net/ethernet/dlink/dl2k.c
+++ b/drivers/net/ethernet/dlink/dl2k.c
@@ -1,18 +1,12 @@
+// SPDX-License-Identifier: GPL-2.0-or-later
 /*  D-Link DL2000-based Gigabit Ethernet Adapter Linux driver */
 /*
     Copyright (c) 2001, 2002 by D-Link Corporation
     Written by Edward Peng.<edward_peng@dlink.com.tw>
     Created 03-May-2001, base on Linux' sundance.c.
 
-    This program is free software; you can redistribute it and/or modify
-    it under the terms of the GNU General Public License as published by
-    the Free Software Foundation; either version 2 of the License, or
-    (at your option) any later version.
 */
 
-#define DRV_NAME	"DL2000/TC902x-based linux driver"
-#define DRV_VERSION	"v1.19"
-#define DRV_RELDATE	"2007/08/12"
 #include "dl2k.h"
 #include <linux/dma-mapping.h>
 
@@ -23,8 +17,6 @@
 #define dr16(reg)	ioread16(ioaddr + (reg))
 #define dr8(reg)	ioread8(ioaddr + (reg))
 
-static char version[] =
-      KERN_INFO DRV_NAME " " DRV_VERSION " " DRV_RELDATE "\n";
 #define MAX_UNITS 8
 static int mtu[MAX_UNITS];
 static int vlan[MAX_UNITS];
@@ -68,8 +60,8 @@ static const int max_intrloop = 50;
 static const int multicast_filter_limit = 0x40;
 
 static int rio_open (struct net_device *dev);
-static void rio_timer (unsigned long data);
-static void rio_tx_timeout (struct net_device *dev);
+static void rio_timer (struct timer_list *t);
+static void rio_tx_timeout (struct net_device *dev, unsigned int txqueue);
 static netdev_tx_t start_xmit (struct sk_buff *skb, struct net_device *dev);
 static irqreturn_t rio_interrupt (int irq, void *dev_instance);
 static void rio_free_tx (struct net_device *dev, int irq);
@@ -103,10 +95,17 @@ static const struct net_device_ops netdev_ops = {
 	.ndo_validate_addr	= eth_validate_addr,
 	.ndo_set_mac_address 	= eth_mac_addr,
 	.ndo_set_rx_mode	= set_multicast,
-	.ndo_do_ioctl		= rio_ioctl,
+	.ndo_eth_ioctl		= rio_ioctl,
 	.ndo_tx_timeout		= rio_tx_timeout,
 };
 
+static bool is_support_rmon_mmio(struct pci_dev *pdev)
+{
+	return pdev->vendor == PCI_VENDOR_ID_DLINK &&
+	       pdev->device == 0x4000 &&
+	       pdev->revision == 0x0c;
+}
+
 static int
 rio_probe1 (struct pci_dev *pdev, const struct pci_device_id *ent)
 {
@@ -116,13 +115,9 @@ rio_probe1 (struct pci_dev *pdev, const struct pci_device_id *ent)
 	int chip_idx = ent->driver_data;
 	int err, irq;
 	void __iomem *ioaddr;
-	static int version_printed;
 	void *ring_space;
 	dma_addr_t ring_dma;
 
-	if (!version_printed++)
-		printk ("%s", version);
-
 	err = pci_enable_device (pdev);
 	if (err)
 		return err;
@@ -143,21 +138,27 @@ rio_probe1 (struct pci_dev *pdev, const struct pci_device_id *ent)
 
 	np = netdev_priv(dev);
 
+	if (is_support_rmon_mmio(pdev))
+		np->rmon_enable = true;
+
 	/* IO registers range. */
 	ioaddr = pci_iomap(pdev, 0, 0);
 	if (!ioaddr)
 		goto err_out_dev;
 	np->eeprom_addr = ioaddr;
 
-#ifdef MEM_MAPPING
-	/* MM registers range. */
-	ioaddr = pci_iomap(pdev, 1, 0);
-	if (!ioaddr)
-		goto err_out_iounmap;
-#endif
+	if (np->rmon_enable) {
+		/* MM registers range. */
+		ioaddr = pci_iomap(pdev, 1, 0);
+		if (!ioaddr)
+			goto err_out_iounmap;
+	}
+
 	np->ioaddr = ioaddr;
 	np->chip_id = chip_idx;
 	np->pdev = pdev;
+
+	spin_lock_init(&np->stats_lock);
 	spin_lock_init (&np->tx_lock);
 	spin_lock_init (&np->rx_lock);
 
@@ -234,13 +235,15 @@ rio_probe1 (struct pci_dev *pdev, const struct pci_device_id *ent)
 
 	pci_set_drvdata (pdev, dev);
 
-	ring_space = pci_alloc_consistent (pdev, TX_TOTAL_SIZE, &ring_dma);
+	ring_space = dma_alloc_coherent(&pdev->dev, TX_TOTAL_SIZE, &ring_dma,
+					GFP_KERNEL);
 	if (!ring_space)
 		goto err_out_iounmap;
 	np->tx_ring = ring_space;
 	np->tx_ring_dma = ring_dma;
 
-	ring_space = pci_alloc_consistent (pdev, RX_TOTAL_SIZE, &ring_dma);
+	ring_space = dma_alloc_coherent(&pdev->dev, RX_TOTAL_SIZE, &ring_dma,
+					GFP_KERNEL);
 	if (!ring_space)
 		goto err_out_unmap_tx;
 	np->rx_ring = ring_space;
@@ -291,13 +294,14 @@ rio_probe1 (struct pci_dev *pdev, const struct pci_device_id *ent)
 	return 0;
 
 err_out_unmap_rx:
-	pci_free_consistent (pdev, RX_TOTAL_SIZE, np->rx_ring, np->rx_ring_dma);
+	dma_free_coherent(&pdev->dev, RX_TOTAL_SIZE, np->rx_ring,
+			  np->rx_ring_dma);
 err_out_unmap_tx:
-	pci_free_consistent (pdev, TX_TOTAL_SIZE, np->tx_ring, np->tx_ring_dma);
+	dma_free_coherent(&pdev->dev, TX_TOTAL_SIZE, np->tx_ring,
+			  np->tx_ring_dma);
 err_out_iounmap:
-#ifdef MEM_MAPPING
-	pci_iounmap(pdev, np->ioaddr);
-#endif
+	if (np->rmon_enable)
+		pci_iounmap(pdev, np->ioaddr);
 	pci_iounmap(pdev, np->eeprom_addr);
 err_out_dev:
 	free_netdev (dev);
@@ -313,7 +317,7 @@ find_miiphy (struct net_device *dev)
 {
 	struct netdev_private *np = netdev_priv(dev);
 	int i, phy_found = 0;
-	np = netdev_priv(dev);
+
 	np->phy_addr = 1;
 
 	for (i = 31; i >= 0; i--) {
@@ -357,11 +361,10 @@ parse_eeprom (struct net_device *dev)
 	}
 
 	/* Set MAC address */
-	for (i = 0; i < 6; i++)
-		dev->dev_addr[i] = psrom->mac_addr[i];
+	eth_hw_addr_set(dev, psrom->mac_addr);
 
 	if (np->chip_id == CHIP_IP1000A) {
-		np->led_mode = psrom->led_mode;
+		np->led_mode = le16_to_cpu(psrom->led_mode);
 		return 0;
 	}
 
@@ -447,8 +450,9 @@ static void free_list(struct net_device *dev)
 	for (i = 0; i < RX_RING_SIZE; i++) {
 		skb = np->rx_skbuff[i];
 		if (skb) {
-			pci_unmap_single(np->pdev, desc_to_dma(&np->rx_ring[i]),
-					 skb->len, PCI_DMA_FROMDEVICE);
+			dma_unmap_single(&np->pdev->dev,
+					 desc_to_dma(&np->rx_ring[i]),
+					 skb->len, DMA_FROM_DEVICE);
 			dev_kfree_skb(skb);
 			np->rx_skbuff[i] = NULL;
 		}
@@ -458,8 +462,9 @@ static void free_list(struct net_device *dev)
 	for (i = 0; i < TX_RING_SIZE; i++) {
 		skb = np->tx_skbuff[i];
 		if (skb) {
-			pci_unmap_single(np->pdev, desc_to_dma(&np->tx_ring[i]),
-					 skb->len, PCI_DMA_TODEVICE);
+			dma_unmap_single(&np->pdev->dev,
+					 desc_to_dma(&np->tx_ring[i]),
+					 skb->len, DMA_TO_DEVICE);
 			dev_kfree_skb(skb);
 			np->tx_skbuff[i] = NULL;
 		}
@@ -516,9 +521,8 @@ static int alloc_list(struct net_device *dev)
 						sizeof(struct netdev_desc));
 		/* Rubicon now supports 40 bits of addressing space. */
 		np->rx_ring[i].fraginfo =
-		    cpu_to_le64(pci_map_single(
-				  np->pdev, skb->data, np->rx_buf_sz,
-				  PCI_DMA_FROMDEVICE));
+		    cpu_to_le64(dma_map_single(&np->pdev->dev, skb->data,
+					       np->rx_buf_sz, DMA_FROM_DEVICE));
 		np->rx_ring[i].fraginfo |= cpu_to_le64((u64)np->rx_buf_sz << 48);
 	}
 
@@ -573,8 +577,7 @@ static void rio_hw_init(struct net_device *dev)
 	 * too. However, it doesn't work on IP1000A so we use 16-bit access.
 	 */
 	for (i = 0; i < 3; i++)
-		dw16(StationAddr0 + 2 * i,
-		     cpu_to_le16(((u16 *)dev->dev_addr)[i]));
+		dw16(StationAddr0 + 2 * i, get_unaligned_le16(&dev->dev_addr[2 * i]));
 
 	set_multicast (dev);
 	if (np->coalesce) {
@@ -585,7 +588,8 @@ static void rio_hw_init(struct net_device *dev)
 	dw8(TxDMAPollPeriod, 0xff);
 	dw8(RxDMABurstThresh, 0x30);
 	dw8(RxDMAUrgentThresh, 0x30);
-	dw32(RmonStatMask, 0x0007ffff);
+	if (!np->rmon_enable)
+		dw32(RmonStatMask, 0x0007ffff);
 	/* clear statistics */
 	clear_stats (dev);
 
@@ -644,7 +648,7 @@ static int rio_open(struct net_device *dev)
 		return i;
 	}
 
-	setup_timer(&np->timer, rio_timer, (unsigned long)dev);
+	timer_setup(&np->timer, rio_timer, 0);
 	np->timer.expires = jiffies + 1 * HZ;
 	add_timer(&np->timer);
 
@@ -655,10 +659,10 @@ static int rio_open(struct net_device *dev)
 }
 
 static void
-rio_timer (unsigned long data)
+rio_timer (struct timer_list *t)
 {
-	struct net_device *dev = (struct net_device *)data;
-	struct netdev_private *np = netdev_priv(dev);
+	struct netdev_private *np = timer_container_of(np, t, timer);
+	struct net_device *dev = pci_get_drvdata(np->pdev);
 	unsigned int entry;
 	int next_tick = 1*HZ;
 	unsigned long flags;
@@ -684,9 +688,8 @@ rio_timer (unsigned long data)
 				}
 				np->rx_skbuff[entry] = skb;
 				np->rx_ring[entry].fraginfo =
-				    cpu_to_le64 (pci_map_single
-					 (np->pdev, skb->data, np->rx_buf_sz,
-					  PCI_DMA_FROMDEVICE));
+				    cpu_to_le64 (dma_map_single(&np->pdev->dev, skb->data,
+								np->rx_buf_sz, DMA_FROM_DEVICE));
 			}
 			np->rx_ring[entry].fraginfo |=
 			    cpu_to_le64((u64)np->rx_buf_sz << 48);
@@ -699,7 +702,7 @@ rio_timer (unsigned long data)
 }
 
 static void
-rio_tx_timeout (struct net_device *dev)
+rio_tx_timeout (struct net_device *dev, unsigned int txqueue)
 {
 	struct netdev_private *np = netdev_priv(dev);
 	void __iomem *ioaddr = np->ioaddr;
@@ -740,9 +743,8 @@ start_xmit (struct sk_buff *skb, struct net_device *dev)
 		    ((u64)np->vlan << 32) |
 		    ((u64)skb->priority << 45);
 	}
-	txdesc->fraginfo = cpu_to_le64 (pci_map_single (np->pdev, skb->data,
-							skb->len,
-							PCI_DMA_TODEVICE));
+	txdesc->fraginfo = cpu_to_le64 (dma_map_single(&np->pdev->dev, skb->data,
+						       skb->len, DMA_TO_DEVICE));
 	txdesc->fraginfo |= cpu_to_le64((u64)skb->len << 48);
 
 	/* DL2K bug: DMA fails to get next descriptor ptr in 10Mbps mode
@@ -824,7 +826,6 @@ rio_free_tx (struct net_device *dev, int irq)
 {
 	struct netdev_private *np = netdev_priv(dev);
 	int entry = np->old_tx % TX_RING_SIZE;
-	int tx_use = 0;
 	unsigned long flag = 0;
 
 	if (irq)
@@ -839,17 +840,16 @@ rio_free_tx (struct net_device *dev, int irq)
 		if (!(np->tx_ring[entry].status & cpu_to_le64(TFDDone)))
 			break;
 		skb = np->tx_skbuff[entry];
-		pci_unmap_single (np->pdev,
-				  desc_to_dma(&np->tx_ring[entry]),
-				  skb->len, PCI_DMA_TODEVICE);
+		dma_unmap_single(&np->pdev->dev,
+				 desc_to_dma(&np->tx_ring[entry]), skb->len,
+				 DMA_TO_DEVICE);
 		if (irq)
-			dev_kfree_skb_irq (skb);
+			dev_consume_skb_irq(skb);
 		else
-			dev_kfree_skb (skb);
+			dev_kfree_skb(skb);
 
 		np->tx_skbuff[entry] = NULL;
 		entry = (entry + 1) % TX_RING_SIZE;
-		tx_use++;
 	}
 	if (irq)
 		spin_unlock(&np->tx_lock);
@@ -878,7 +878,6 @@ tx_error (struct net_device *dev, int tx_status)
 	frame_id = (tx_status & 0xffff0000);
 	printk (KERN_ERR "%s: Transmit error, TxStatus %4.4x, FrameId %d.\n",
 		dev->name, tx_status, frame_id);
-	dev->stats.tx_errors++;
 	/* Ttransmit Underrun */
 	if (tx_status & 0x10) {
 		dev->stats.tx_fifo_errors++;
@@ -915,9 +914,15 @@ tx_error (struct net_device *dev, int tx_status)
 		rio_set_led_mode(dev);
 		/* Let TxStartThresh stay default value */
 	}
+
+	spin_lock(&np->stats_lock);
 	/* Maximum Collisions */
 	if (tx_status & 0x08)
 		dev->stats.collisions++;
+
+	dev->stats.tx_errors++;
+	spin_unlock(&np->stats_lock);
+
 	/* Restart the Tx */
 	dw32(MACCtrl, dr16(MACCtrl) | TxEnable);
 }
@@ -959,27 +964,30 @@ receive_packet (struct net_device *dev)
 		} else {
 			struct sk_buff *skb;
 
+			skb = NULL;
 			/* Small skbuffs for short packets */
-			if (pkt_len > copy_thresh) {
-				pci_unmap_single (np->pdev,
-						  desc_to_dma(desc),
-						  np->rx_buf_sz,
-						  PCI_DMA_FROMDEVICE);
+			if (pkt_len <= copy_thresh)
+				skb = netdev_alloc_skb_ip_align(dev, pkt_len);
+			if (!skb) {
+				dma_unmap_single(&np->pdev->dev,
+						 desc_to_dma(desc),
+						 np->rx_buf_sz,
+						 DMA_FROM_DEVICE);
 				skb_put (skb = np->rx_skbuff[entry], pkt_len);
 				np->rx_skbuff[entry] = NULL;
-			} else if ((skb = netdev_alloc_skb_ip_align(dev, pkt_len))) {
-				pci_dma_sync_single_for_cpu(np->pdev,
-							    desc_to_dma(desc),
-							    np->rx_buf_sz,
-							    PCI_DMA_FROMDEVICE);
+			} else {
+				dma_sync_single_for_cpu(&np->pdev->dev,
+							desc_to_dma(desc),
+							np->rx_buf_sz,
+							DMA_FROM_DEVICE);
 				skb_copy_to_linear_data (skb,
 						  np->rx_skbuff[entry]->data,
 						  pkt_len);
 				skb_put (skb, pkt_len);
-				pci_dma_sync_single_for_device(np->pdev,
-							       desc_to_dma(desc),
-							       np->rx_buf_sz,
-							       PCI_DMA_FROMDEVICE);
+				dma_sync_single_for_device(&np->pdev->dev,
+							   desc_to_dma(desc),
+							   np->rx_buf_sz,
+							   DMA_FROM_DEVICE);
 			}
 			skb->protocol = eth_type_trans (skb, dev);
 #if 0
@@ -1012,9 +1020,8 @@ receive_packet (struct net_device *dev)
 			}
 			np->rx_skbuff[entry] = skb;
 			np->rx_ring[entry].fraginfo =
-			    cpu_to_le64 (pci_map_single
-					 (np->pdev, skb->data, np->rx_buf_sz,
-					  PCI_DMA_FROMDEVICE));
+			    cpu_to_le64(dma_map_single(&np->pdev->dev, skb->data,
+						       np->rx_buf_sz, DMA_FROM_DEVICE));
 		}
 		np->rx_ring[entry].fraginfo |=
 		    cpu_to_le64((u64)np->rx_buf_sz << 48);
@@ -1083,11 +1090,10 @@ get_stats (struct net_device *dev)
 {
 	struct netdev_private *np = netdev_priv(dev);
 	void __iomem *ioaddr = np->ioaddr;
-#ifdef MEM_MAPPING
-	int i;
-#endif
 	unsigned int stat_reg;
+	unsigned long flags;
 
+	spin_lock_irqsave(&np->stats_lock, flags);
 	/* All statistics registers need to be acknowledged,
 	   else statistic overflow could cause problems */
 
@@ -1096,7 +1102,7 @@ get_stats (struct net_device *dev)
 	dev->stats.rx_bytes += dr32(OctetRcvOk);
 	dev->stats.tx_bytes += dr32(OctetXmtOk);
 
-	dev->stats.multicast = dr32(McstFramesRcvdOk);
+	dev->stats.multicast += dr32(McstFramesRcvdOk);
 	dev->stats.collisions += dr32(SingleColFrames)
 			     +  dr32(MultiColFrames);
 
@@ -1128,15 +1134,18 @@ get_stats (struct net_device *dev)
 	dr16(MacControlFramesXmtd);
 	dr16(FramesWEXDeferal);
 
-#ifdef MEM_MAPPING
-	for (i = 0x100; i <= 0x150; i += 4)
-		dr32(i);
-#endif
+	if (np->rmon_enable)
+		for (int i = 0x100; i <= 0x150; i += 4)
+			dr32(i);
+
 	dr16(TxJumboFrames);
 	dr16(RxJumboFrames);
 	dr16(TCPCheckSumErrors);
 	dr16(UDPCheckSumErrors);
 	dr16(IPCheckSumErrors);
+
+	spin_unlock_irqrestore(&np->stats_lock, flags);
+
 	return &dev->stats;
 }
 
@@ -1145,9 +1154,6 @@ clear_stats (struct net_device *dev)
 {
 	struct netdev_private *np = netdev_priv(dev);
 	void __iomem *ioaddr = np->ioaddr;
-#ifdef MEM_MAPPING
-	int i;
-#endif
 
 	/* All statistics registers need to be acknowledged,
 	   else statistic overflow could cause problems */
@@ -1183,10 +1189,9 @@ clear_stats (struct net_device *dev)
 	dr16(BcstFramesXmtdOk);
 	dr16(MacControlFramesXmtd);
 	dr16(FramesWEXDeferal);
-#ifdef MEM_MAPPING
-	for (i = 0x100; i <= 0x150; i += 4)
-		dr32(i);
-#endif
+	if (np->rmon_enable)
+		for (int i = 0x100; i <= 0x150; i += 4)
+			dr32(i);
 	dr16(TxJumboFrames);
 	dr16(RxJumboFrames);
 	dr16(TCPCheckSumErrors);
@@ -1246,9 +1251,8 @@ static void rio_get_drvinfo(struct net_device *dev, struct ethtool_drvinfo *info
 {
 	struct netdev_private *np = netdev_priv(dev);
 
-	strlcpy(info->driver, "dl2k", sizeof(info->driver));
-	strlcpy(info->version, DRV_VERSION, sizeof(info->version));
-	strlcpy(info->bus_info, pci_name(np->pdev), sizeof(info->bus_info));
+	strscpy(info->driver, "dl2k", sizeof(info->driver));
+	strscpy(info->bus_info, pci_name(np->pdev), sizeof(info->bus_info));
 }
 
 static int rio_get_link_ksettings(struct net_device *dev,
@@ -1793,7 +1797,7 @@ rio_close (struct net_device *dev)
 	rio_hw_stop(dev);
 
 	free_irq(pdev->irq, dev);
-	del_timer_sync (&np->timer);
+	timer_delete_sync(&np->timer);
 
 	free_list(dev);
 
@@ -1809,13 +1813,12 @@ rio_remove1 (struct pci_dev *pdev)
 		struct netdev_private *np = netdev_priv(dev);
 
 		unregister_netdev (dev);
-		pci_free_consistent (pdev, RX_TOTAL_SIZE, np->rx_ring,
-				     np->rx_ring_dma);
-		pci_free_consistent (pdev, TX_TOTAL_SIZE, np->tx_ring,
-				     np->tx_ring_dma);
-#ifdef MEM_MAPPING
-		pci_iounmap(pdev, np->ioaddr);
-#endif
+		dma_free_coherent(&pdev->dev, RX_TOTAL_SIZE, np->rx_ring,
+				  np->rx_ring_dma);
+		dma_free_coherent(&pdev->dev, TX_TOTAL_SIZE, np->tx_ring,
+				  np->tx_ring_dma);
+		if (np->rmon_enable)
+			pci_iounmap(pdev, np->ioaddr);
 		pci_iounmap(pdev, np->eeprom_addr);
 		free_netdev (dev);
 		pci_release_regions (pdev);
@@ -1833,7 +1836,7 @@ static int rio_suspend(struct device *device)
 		return 0;
 
 	netif_device_detach(dev);
-	del_timer_sync(&np->timer);
+	timer_delete_sync(&np->timer);
 	rio_hw_stop(dev);
 
 	return 0;
@@ -1857,7 +1860,7 @@ static int rio_resume(struct device *device)
 	return 0;
 }
 
-static SIMPLE_DEV_PM_OPS(rio_pm_ops, rio_suspend, rio_resume);
+static DEFINE_SIMPLE_DEV_PM_OPS(rio_pm_ops, rio_suspend, rio_resume);
 #define RIO_PM_OPS    (&rio_pm_ops)
 
 #else
@@ -1875,13 +1878,5 @@ static struct pci_driver rio_driver = {
 };
 
 module_pci_driver(rio_driver);
-/*
-
-Compile command:
-
-gcc -D__KERNEL__ -DMODULE -I/usr/src/linux/include -Wall -Wstrict-prototypes -O2 -c dl2k.c
-
-Read Documentation/networking/dl2k.txt for details.
-
-*/
 
+/* Read Documentation/networking/device_drivers/ethernet/dlink/dl2k.rst. */
diff --git a/drivers/net/ethernet/dlink/dl2k.h b/drivers/net/ethernet/dlink/dl2k.h
index 10e98ba33ebf..4788cc94639d 100644
--- a/drivers/net/ethernet/dlink/dl2k.h
+++ b/drivers/net/ethernet/dlink/dl2k.h
@@ -1,13 +1,10 @@
+/* SPDX-License-Identifier: GPL-2.0-or-later */
 /*  D-Link DL2000-based Gigabit Ethernet Adapter Linux driver */
 /*
     Copyright (c) 2001, 2002 by D-Link Corporation
     Written by Edward Peng.<edward_peng@dlink.com.tw>
     Created 03-May-2001, base on Linux' sundance.c.
 
-    This program is free software; you can redistribute it and/or modify
-    it under the terms of the GNU General Public License as published by
-    the Free Software Foundation; either version 2 of the License, or
-    (at your option) any later version.
 */
 
 #ifndef __DL2K_H__
@@ -332,18 +329,18 @@ enum _pcs_anlpar {
 };
 
 typedef struct t_SROM {
-	u16 config_param;	/* 0x00 */
-	u16 asic_ctrl;		/* 0x02 */
-	u16 sub_vendor_id;	/* 0x04 */
-	u16 sub_system_id;	/* 0x06 */
-	u16 pci_base_1;		/* 0x08 (IP1000A only) */
-	u16 pci_base_2;		/* 0x0a (IP1000A only) */
-	u16 led_mode;		/* 0x0c (IP1000A only) */
-	u16 reserved1[9];	/* 0x0e-0x1f */
+	__le16 config_param;	/* 0x00 */
+	__le16 asic_ctrl;	/* 0x02 */
+	__le16 sub_vendor_id;	/* 0x04 */
+	__le16 sub_system_id;	/* 0x06 */
+	__le16 pci_base_1;	/* 0x08 (IP1000A only) */
+	__le16 pci_base_2;	/* 0x0a (IP1000A only) */
+	__le16 led_mode;	/* 0x0c (IP1000A only) */
+	__le16 reserved1[9];	/* 0x0e-0x1f */
 	u8 mac_addr[6];		/* 0x20-0x25 */
 	u8 reserved2[10];	/* 0x26-0x2f */
 	u8 sib[204];		/* 0x30-0xfb */
-	u32 crc;		/* 0xfc-0xff */
+	__le32 crc;		/* 0xfc-0xff */
 } SROM_t, *PSROM_t;
 
 /* Ioctl custom data */
@@ -375,6 +372,8 @@ struct netdev_private {
 	struct pci_dev *pdev;
 	void __iomem *ioaddr;
 	void __iomem *eeprom_addr;
+	// To ensure synchronization when stats are updated.
+	spinlock_t stats_lock;
 	spinlock_t tx_lock;
 	spinlock_t rx_lock;
 	unsigned int rx_buf_sz;		/* Based on MTU+slack. */
@@ -404,6 +403,8 @@ struct netdev_private {
 	u16 negotiate;		/* Negotiated media */
 	int phy_addr;		/* PHY addresses. */
 	u16 led_mode;		/* LED mode read from EEPROM (IP1000A only) */
+
+	bool rmon_enable;
 };
 
 /* The station address location in the EEPROM. */
diff --git a/drivers/net/ethernet/dlink/sundance.c b/drivers/net/ethernet/dlink/sundance.c
index 2704bcf023be..277c50ef773f 100644
--- a/drivers/net/ethernet/dlink/sundance.c
+++ b/drivers/net/ethernet/dlink/sundance.c
@@ -18,14 +18,11 @@
 	http://www.scyld.com/network/sundance.html
 	[link no longer provides useful info -jgarzik]
 	Archives of the mailing list are still available at
-	http://www.beowulf.org/pipermail/netdrivers/
+	https://www.beowulf.org/pipermail/netdrivers/
 
 */
 
 #define DRV_NAME	"sundance"
-#define DRV_VERSION	"1.2"
-#define DRV_RELDATE	"11-Sep-2006"
-
 
 /* The user-configurable values.
    These may be modified when a driver module is loaded.*/
@@ -101,11 +98,6 @@ static char *media[MAX_UNITS];
 #include <linux/ethtool.h>
 #include <linux/mii.h>
 
-/* These identify the driver base version and may not be removed. */
-static const char version[] =
-	KERN_INFO DRV_NAME ".c:v" DRV_VERSION " " DRV_RELDATE
-	" Written by Donald Becker\n";
-
 MODULE_AUTHOR("Donald Becker <becker@scyld.com>");
 MODULE_DESCRIPTION("Sundance Alta Ethernet driver");
 MODULE_LICENSE("GPL");
@@ -348,7 +340,7 @@ enum wake_event_bits {
 struct netdev_desc {
 	__le32 next_desc;
 	__le32 status;
-	struct desc_frag { __le32 addr, length; } frag[1];
+	struct desc_frag { __le32 addr, length; } frag;
 };
 
 /* Bits in netdev_desc.status */
@@ -375,6 +367,7 @@ struct netdev_private {
         dma_addr_t tx_ring_dma;
         dma_addr_t rx_ring_dma;
 	struct timer_list timer;		/* Media monitoring timer. */
+	struct net_device *ndev;		/* backpointer */
 	/* ethtool extra stats */
 	struct {
 		u64 tx_multiple_collisions;
@@ -431,14 +424,14 @@ static void mdio_write(struct net_device *dev, int phy_id, int location, int val
 static int  mdio_wait_link(struct net_device *dev, int wait);
 static int  netdev_open(struct net_device *dev);
 static void check_duplex(struct net_device *dev);
-static void netdev_timer(unsigned long data);
-static void tx_timeout(struct net_device *dev);
+static void netdev_timer(struct timer_list *t);
+static void tx_timeout(struct net_device *dev, unsigned int txqueue);
 static void init_ring(struct net_device *dev);
 static netdev_tx_t start_tx(struct sk_buff *skb, struct net_device *dev);
 static int reset_tx (struct net_device *dev);
 static irqreturn_t intr_handler(int irq, void *dev_instance);
-static void rx_poll(unsigned long data);
-static void tx_poll(unsigned long data);
+static void rx_poll(struct tasklet_struct *t);
+static void tx_poll(struct tasklet_struct *t);
 static void refill_rx (struct net_device *dev);
 static void netdev_error(struct net_device *dev, int intr_status);
 static void netdev_error(struct net_device *dev, int intr_status);
@@ -486,7 +479,7 @@ static const struct net_device_ops netdev_ops = {
 	.ndo_start_xmit		= start_tx,
 	.ndo_get_stats 		= get_stats,
 	.ndo_set_rx_mode	= set_rx_mode,
-	.ndo_do_ioctl 		= netdev_ioctl,
+	.ndo_eth_ioctl		= netdev_ioctl,
 	.ndo_tx_timeout		= tx_timeout,
 	.ndo_change_mtu		= change_mtu,
 	.ndo_set_mac_address 	= sundance_set_mac_addr,
@@ -515,13 +508,7 @@ static int sundance_probe1(struct pci_dev *pdev,
 	int bar = 1;
 #endif
 	int phy, phy_end, phy_idx = 0;
-
-/* when built into the kernel, we only print version if device is found */
-#ifndef MODULE
-	static int printed_version;
-	if (!printed_version++)
-		printk(version);
-#endif
+	__le16 addr[ETH_ALEN / 2];
 
 	if (pci_enable_device(pdev))
 		return -EIO;
@@ -542,18 +529,20 @@ static int sundance_probe1(struct pci_dev *pdev,
 		goto err_out_res;
 
 	for (i = 0; i < 3; i++)
-		((__le16 *)dev->dev_addr)[i] =
+		addr[i] =
 			cpu_to_le16(eeprom_read(ioaddr, i + EEPROM_SA_OFFSET));
+	eth_hw_addr_set(dev, (u8 *)addr);
 
 	np = netdev_priv(dev);
+	np->ndev = dev;
 	np->base = ioaddr;
 	np->pci_dev = pdev;
 	np->chip_id = chip_idx;
 	np->msg_enable = (1 << debug) - 1;
 	spin_lock_init(&np->lock);
 	spin_lock_init(&np->statlock);
-	tasklet_init(&np->rx_tasklet, rx_poll, (unsigned long)dev);
-	tasklet_init(&np->tx_tasklet, tx_poll, (unsigned long)dev);
+	tasklet_setup(&np->rx_tasklet, rx_poll);
+	tasklet_setup(&np->tx_tasklet, tx_poll);
 
 	ring_space = dma_alloc_coherent(&pdev->dev, TX_TOTAL_SIZE,
 			&ring_dma, GFP_KERNEL);
@@ -719,7 +708,7 @@ static int change_mtu(struct net_device *dev, int new_mtu)
 {
 	if (netif_running(dev))
 		return -EBUSY;
-	dev->mtu = new_mtu;
+	WRITE_ONCE(dev->mtu, new_mtu);
 	return 0;
 }
 
@@ -913,10 +902,8 @@ static int netdev_open(struct net_device *dev)
 			   ioread16(ioaddr + MACCtrl1), ioread16(ioaddr + MACCtrl0));
 
 	/* Set the timer to check for link beat. */
-	init_timer(&np->timer);
+	timer_setup(&np->timer, netdev_timer, 0);
 	np->timer.expires = jiffies + 3*HZ;
-	np->timer.data = (unsigned long)dev;
-	np->timer.function = netdev_timer;				/* timer handler */
 	add_timer(&np->timer);
 
 	/* Enable interrupts by setting the interrupt mask. */
@@ -953,10 +940,10 @@ static void check_duplex(struct net_device *dev)
 	}
 }
 
-static void netdev_timer(unsigned long data)
+static void netdev_timer(struct timer_list *t)
 {
-	struct net_device *dev = (struct net_device *)data;
-	struct netdev_private *np = netdev_priv(dev);
+	struct netdev_private *np = timer_container_of(np, t, timer);
+	struct net_device *dev = np->mii_if.dev;
 	void __iomem *ioaddr = np->base;
 	int next_tick = 10*HZ;
 
@@ -971,14 +958,14 @@ static void netdev_timer(unsigned long data)
 	add_timer(&np->timer);
 }
 
-static void tx_timeout(struct net_device *dev)
+static void tx_timeout(struct net_device *dev, unsigned int txqueue)
 {
 	struct netdev_private *np = netdev_priv(dev);
 	void __iomem *ioaddr = np->base;
 	unsigned long flag;
 
 	netif_stop_queue(dev);
-	tasklet_disable(&np->tx_tasklet);
+	tasklet_disable_in_atomic(&np->tx_tasklet);
 	iowrite16(0, ioaddr + IntrEnable);
 	printk(KERN_WARNING "%s: Transmit timed out, TxStatus %2.2x "
 		   "TxFrameId %2.2x,"
@@ -993,8 +980,8 @@ static void tx_timeout(struct net_device *dev)
 				le32_to_cpu(np->tx_ring[i].next_desc),
 				le32_to_cpu(np->tx_ring[i].status),
 				(le32_to_cpu(np->tx_ring[i].status) >> 2) & 0xff,
-				le32_to_cpu(np->tx_ring[i].frag[0].addr),
-				le32_to_cpu(np->tx_ring[i].frag[0].length));
+				le32_to_cpu(np->tx_ring[i].frag.addr),
+				le32_to_cpu(np->tx_ring[i].frag.length));
 		}
 		printk(KERN_DEBUG "TxListPtr=%08x netif_queue_stopped=%d\n",
 			ioread32(np->base + TxListPtr),
@@ -1040,28 +1027,29 @@ static void init_ring(struct net_device *dev)
 		np->rx_ring[i].next_desc = cpu_to_le32(np->rx_ring_dma +
 			((i+1)%RX_RING_SIZE)*sizeof(*np->rx_ring));
 		np->rx_ring[i].status = 0;
-		np->rx_ring[i].frag[0].length = 0;
+		np->rx_ring[i].frag.length = 0;
 		np->rx_skbuff[i] = NULL;
 	}
 
 	/* Fill in the Rx buffers.  Handle allocation failure gracefully. */
 	for (i = 0; i < RX_RING_SIZE; i++) {
+		dma_addr_t addr;
+
 		struct sk_buff *skb =
 			netdev_alloc_skb(dev, np->rx_buf_sz + 2);
 		np->rx_skbuff[i] = skb;
 		if (skb == NULL)
 			break;
 		skb_reserve(skb, 2);	/* 16 byte align the IP header. */
-		np->rx_ring[i].frag[0].addr = cpu_to_le32(
-			dma_map_single(&np->pci_dev->dev, skb->data,
-				np->rx_buf_sz, DMA_FROM_DEVICE));
-		if (dma_mapping_error(&np->pci_dev->dev,
-					np->rx_ring[i].frag[0].addr)) {
+		addr = dma_map_single(&np->pci_dev->dev, skb->data,
+				      np->rx_buf_sz, DMA_FROM_DEVICE);
+		if (dma_mapping_error(&np->pci_dev->dev, addr)) {
 			dev_kfree_skb(skb);
 			np->rx_skbuff[i] = NULL;
 			break;
 		}
-		np->rx_ring[i].frag[0].length = cpu_to_le32(np->rx_buf_sz | LastFrag);
+		np->rx_ring[i].frag.addr = cpu_to_le32(addr);
+		np->rx_ring[i].frag.length = cpu_to_le32(np->rx_buf_sz | LastFrag);
 	}
 	np->dirty_rx = (unsigned int)(i - RX_RING_SIZE);
 
@@ -1071,10 +1059,9 @@ static void init_ring(struct net_device *dev)
 	}
 }
 
-static void tx_poll (unsigned long data)
+static void tx_poll(struct tasklet_struct *t)
 {
-	struct net_device *dev = (struct net_device *)data;
-	struct netdev_private *np = netdev_priv(dev);
+	struct netdev_private *np = from_tasklet(np, t, tx_tasklet);
 	unsigned head = np->cur_task % TX_RING_SIZE;
 	struct netdev_desc *txdesc =
 		&np->tx_ring[(np->cur_tx - 1) % TX_RING_SIZE];
@@ -1102,6 +1089,7 @@ start_tx (struct sk_buff *skb, struct net_device *dev)
 {
 	struct netdev_private *np = netdev_priv(dev);
 	struct netdev_desc *txdesc;
+	dma_addr_t addr;
 	unsigned entry;
 
 	/* Calculate the next Tx descriptor entry. */
@@ -1109,14 +1097,15 @@ start_tx (struct sk_buff *skb, struct net_device *dev)
 	np->tx_skbuff[entry] = skb;
 	txdesc = &np->tx_ring[entry];
 
+	addr = dma_map_single(&np->pci_dev->dev, skb->data, skb->len,
+			      DMA_TO_DEVICE);
+	if (dma_mapping_error(&np->pci_dev->dev, addr))
+		goto drop_frame;
+
 	txdesc->next_desc = 0;
 	txdesc->status = cpu_to_le32 ((entry << 2) | DisableAlign);
-	txdesc->frag[0].addr = cpu_to_le32(dma_map_single(&np->pci_dev->dev,
-				skb->data, skb->len, DMA_TO_DEVICE));
-	if (dma_mapping_error(&np->pci_dev->dev,
-				txdesc->frag[0].addr))
-			goto drop_frame;
-	txdesc->frag[0].length = cpu_to_le32 (skb->len | LastFrag);
+	txdesc->frag.addr = cpu_to_le32(addr);
+	txdesc->frag.length = cpu_to_le32 (skb->len | LastFrag);
 
 	/* Increment cur_tx before tasklet_schedule() */
 	np->cur_tx++;
@@ -1165,7 +1154,7 @@ reset_tx (struct net_device *dev)
 		skb = np->tx_skbuff[i];
 		if (skb) {
 			dma_unmap_single(&np->pci_dev->dev,
-				le32_to_cpu(np->tx_ring[i].frag[0].addr),
+				le32_to_cpu(np->tx_ring[i].frag.addr),
 				skb->len, DMA_TO_DEVICE);
 			dev_kfree_skb_any(skb);
 			np->tx_skbuff[i] = NULL;
@@ -1195,7 +1184,6 @@ static irqreturn_t intr_handler(int irq, void *dev_instance)
 	int handled = 0;
 	int i;
 
-
 	do {
 		int intr_status = ioread16(ioaddr + IntrStatus);
 		iowrite16(intr_status, ioaddr + IntrStatus);
@@ -1286,12 +1274,12 @@ static irqreturn_t intr_handler(int irq, void *dev_instance)
 				skb = np->tx_skbuff[entry];
 				/* Free the original skb. */
 				dma_unmap_single(&np->pci_dev->dev,
-					le32_to_cpu(np->tx_ring[entry].frag[0].addr),
+					le32_to_cpu(np->tx_ring[entry].frag.addr),
 					skb->len, DMA_TO_DEVICE);
-				dev_kfree_skb_irq (np->tx_skbuff[entry]);
+				dev_consume_skb_irq(np->tx_skbuff[entry]);
 				np->tx_skbuff[entry] = NULL;
-				np->tx_ring[entry].frag[0].addr = 0;
-				np->tx_ring[entry].frag[0].length = 0;
+				np->tx_ring[entry].frag.addr = 0;
+				np->tx_ring[entry].frag.length = 0;
 			}
 			spin_unlock(&np->lock);
 		} else {
@@ -1305,12 +1293,12 @@ static irqreturn_t intr_handler(int irq, void *dev_instance)
 				skb = np->tx_skbuff[entry];
 				/* Free the original skb. */
 				dma_unmap_single(&np->pci_dev->dev,
-					le32_to_cpu(np->tx_ring[entry].frag[0].addr),
+					le32_to_cpu(np->tx_ring[entry].frag.addr),
 					skb->len, DMA_TO_DEVICE);
-				dev_kfree_skb_irq (np->tx_skbuff[entry]);
+				dev_consume_skb_irq(np->tx_skbuff[entry]);
 				np->tx_skbuff[entry] = NULL;
-				np->tx_ring[entry].frag[0].addr = 0;
-				np->tx_ring[entry].frag[0].length = 0;
+				np->tx_ring[entry].frag.addr = 0;
+				np->tx_ring[entry].frag.length = 0;
 			}
 			spin_unlock(&np->lock);
 		}
@@ -1330,10 +1318,10 @@ static irqreturn_t intr_handler(int irq, void *dev_instance)
 	return IRQ_RETVAL(handled);
 }
 
-static void rx_poll(unsigned long data)
+static void rx_poll(struct tasklet_struct *t)
 {
-	struct net_device *dev = (struct net_device *)data;
-	struct netdev_private *np = netdev_priv(dev);
+	struct netdev_private *np = from_tasklet(np, t, rx_tasklet);
+	struct net_device *dev = np->ndev;
 	int entry = np->cur_rx % RX_RING_SIZE;
 	int boguscnt = np->budget;
 	void __iomem *ioaddr = np->base;
@@ -1387,16 +1375,16 @@ static void rx_poll(unsigned long data)
 			    (skb = netdev_alloc_skb(dev, pkt_len + 2)) != NULL) {
 				skb_reserve(skb, 2);	/* 16 byte align the IP header */
 				dma_sync_single_for_cpu(&np->pci_dev->dev,
-						le32_to_cpu(desc->frag[0].addr),
+						le32_to_cpu(desc->frag.addr),
 						np->rx_buf_sz, DMA_FROM_DEVICE);
 				skb_copy_to_linear_data(skb, np->rx_skbuff[entry]->data, pkt_len);
 				dma_sync_single_for_device(&np->pci_dev->dev,
-						le32_to_cpu(desc->frag[0].addr),
+						le32_to_cpu(desc->frag.addr),
 						np->rx_buf_sz, DMA_FROM_DEVICE);
 				skb_put(skb, pkt_len);
 			} else {
 				dma_unmap_single(&np->pci_dev->dev,
-					le32_to_cpu(desc->frag[0].addr),
+					le32_to_cpu(desc->frag.addr),
 					np->rx_buf_sz, DMA_FROM_DEVICE);
 				skb_put(skb = np->rx_skbuff[entry], pkt_len);
 				np->rx_skbuff[entry] = NULL;
@@ -1429,12 +1417,13 @@ static void refill_rx (struct net_device *dev)
 {
 	struct netdev_private *np = netdev_priv(dev);
 	int entry;
-	int cnt = 0;
 
 	/* Refill the Rx ring buffers. */
 	for (;(np->cur_rx - np->dirty_rx + RX_RING_SIZE) % RX_RING_SIZE > 0;
 		np->dirty_rx = (np->dirty_rx + 1) % RX_RING_SIZE) {
 		struct sk_buff *skb;
+		dma_addr_t addr;
+
 		entry = np->dirty_rx % RX_RING_SIZE;
 		if (np->rx_skbuff[entry] == NULL) {
 			skb = netdev_alloc_skb(dev, np->rx_buf_sz + 2);
@@ -1442,21 +1431,20 @@ static void refill_rx (struct net_device *dev)
 			if (skb == NULL)
 				break;		/* Better luck next round. */
 			skb_reserve(skb, 2);	/* Align IP on 16 byte boundaries */
-			np->rx_ring[entry].frag[0].addr = cpu_to_le32(
-				dma_map_single(&np->pci_dev->dev, skb->data,
-					np->rx_buf_sz, DMA_FROM_DEVICE));
-			if (dma_mapping_error(&np->pci_dev->dev,
-				    np->rx_ring[entry].frag[0].addr)) {
+			addr = dma_map_single(&np->pci_dev->dev, skb->data,
+					      np->rx_buf_sz, DMA_FROM_DEVICE);
+			if (dma_mapping_error(&np->pci_dev->dev, addr)) {
 			    dev_kfree_skb_irq(skb);
 			    np->rx_skbuff[entry] = NULL;
 			    break;
 			}
+
+			np->rx_ring[entry].frag.addr = cpu_to_le32(addr);
 		}
 		/* Perhaps we need not reset this field. */
-		np->rx_ring[entry].frag[0].length =
+		np->rx_ring[entry].frag.length =
 			cpu_to_le32(np->rx_buf_sz | LastFrag);
 		np->rx_ring[entry].status = 0;
-		cnt++;
 	}
 }
 static void netdev_error(struct net_device *dev, int intr_status)
@@ -1628,7 +1616,7 @@ static int sundance_set_mac_addr(struct net_device *dev, void *data)
 
 	if (!is_valid_ether_addr(addr->sa_data))
 		return -EADDRNOTAVAIL;
-	memcpy(dev->dev_addr, addr->sa_data, ETH_ALEN);
+	eth_hw_addr_set(dev, addr->sa_data);
 	__set_mac_addr(dev);
 
 	return 0;
@@ -1659,9 +1647,8 @@ static int check_if_running(struct net_device *dev)
 static void get_drvinfo(struct net_device *dev, struct ethtool_drvinfo *info)
 {
 	struct netdev_private *np = netdev_priv(dev);
-	strlcpy(info->driver, DRV_NAME, sizeof(info->driver));
-	strlcpy(info->version, DRV_VERSION, sizeof(info->version));
-	strlcpy(info->bus_info, pci_name(np->pci_dev), sizeof(info->bus_info));
+	strscpy(info->driver, DRV_NAME, sizeof(info->driver));
+	strscpy(info->bus_info, pci_name(np->pci_dev), sizeof(info->bus_info));
 }
 
 static int get_link_ksettings(struct net_device *dev,
@@ -1865,20 +1852,20 @@ static int netdev_close(struct net_device *dev)
 	/* Stop the chip's Tx and Rx processes. */
 	iowrite16(TxDisable | RxDisable | StatsDisable, ioaddr + MACCtrl1);
 
-    	for (i = 2000; i > 0; i--) {
- 		if ((ioread32(ioaddr + DMACtrl) & 0xc000) == 0)
+	for (i = 2000; i > 0; i--) {
+		if ((ioread32(ioaddr + DMACtrl) & 0xc000) == 0)
 			break;
 		mdelay(1);
-    	}
+	}
 
-    	iowrite16(GlobalReset | DMAReset | FIFOReset | NetworkReset,
+	iowrite16(GlobalReset | DMAReset | FIFOReset | NetworkReset,
 			ioaddr + ASIC_HI_WORD(ASICCtrl));
 
-    	for (i = 2000; i > 0; i--) {
+	for (i = 2000; i > 0; i--) {
 		if ((ioread16(ioaddr + ASIC_HI_WORD(ASICCtrl)) & ResetBusy) == 0)
 			break;
 		mdelay(1);
-    	}
+	}
 
 #ifdef __i386__
 	if (netif_msg_hw(np)) {
@@ -1886,21 +1873,21 @@ static int netdev_close(struct net_device *dev)
 			   (int)(np->tx_ring_dma));
 		for (i = 0; i < TX_RING_SIZE; i++)
 			printk(KERN_DEBUG " #%d desc. %4.4x %8.8x %8.8x.\n",
-				   i, np->tx_ring[i].status, np->tx_ring[i].frag[0].addr,
-				   np->tx_ring[i].frag[0].length);
+				   i, np->tx_ring[i].status, np->tx_ring[i].frag.addr,
+				   np->tx_ring[i].frag.length);
 		printk(KERN_DEBUG "  Rx ring %8.8x:\n",
 			   (int)(np->rx_ring_dma));
 		for (i = 0; i < /*RX_RING_SIZE*/4 ; i++) {
 			printk(KERN_DEBUG " #%d desc. %4.4x %4.4x %8.8x\n",
-				   i, np->rx_ring[i].status, np->rx_ring[i].frag[0].addr,
-				   np->rx_ring[i].frag[0].length);
+				   i, np->rx_ring[i].status, np->rx_ring[i].frag.addr,
+				   np->rx_ring[i].frag.length);
 		}
 	}
 #endif /* __i386__ debugging only */
 
 	free_irq(np->pci_dev->irq, dev);
 
-	del_timer_sync(&np->timer);
+	timer_delete_sync(&np->timer);
 
 	/* Free all the skbuffs in the Rx queue. */
 	for (i = 0; i < RX_RING_SIZE; i++) {
@@ -1908,19 +1895,19 @@ static int netdev_close(struct net_device *dev)
 		skb = np->rx_skbuff[i];
 		if (skb) {
 			dma_unmap_single(&np->pci_dev->dev,
-				le32_to_cpu(np->rx_ring[i].frag[0].addr),
+				le32_to_cpu(np->rx_ring[i].frag.addr),
 				np->rx_buf_sz, DMA_FROM_DEVICE);
 			dev_kfree_skb(skb);
 			np->rx_skbuff[i] = NULL;
 		}
-		np->rx_ring[i].frag[0].addr = cpu_to_le32(0xBADF00D0); /* poison */
+		np->rx_ring[i].frag.addr = cpu_to_le32(0xBADF00D0); /* poison */
 	}
 	for (i = 0; i < TX_RING_SIZE; i++) {
 		np->tx_ring[i].next_desc = 0;
 		skb = np->tx_skbuff[i];
 		if (skb) {
 			dma_unmap_single(&np->pci_dev->dev,
-				le32_to_cpu(np->tx_ring[i].frag[0].addr),
+				le32_to_cpu(np->tx_ring[i].frag.addr),
 				skb->len, DMA_TO_DEVICE);
 			dev_kfree_skb(skb);
 			np->tx_skbuff[i] = NULL;
@@ -1947,11 +1934,9 @@ static void sundance_remove1(struct pci_dev *pdev)
 	}
 }
 
-#ifdef CONFIG_PM
-
-static int sundance_suspend(struct pci_dev *pci_dev, pm_message_t state)
+static int __maybe_unused sundance_suspend(struct device *dev_d)
 {
-	struct net_device *dev = pci_get_drvdata(pci_dev);
+	struct net_device *dev = dev_get_drvdata(dev_d);
 	struct netdev_private *np = netdev_priv(dev);
 	void __iomem *ioaddr = np->base;
 
@@ -1961,30 +1946,24 @@ static int sundance_suspend(struct pci_dev *pci_dev, pm_message_t state)
 	netdev_close(dev);
 	netif_device_detach(dev);
 
-	pci_save_state(pci_dev);
 	if (np->wol_enabled) {
 		iowrite8(AcceptBroadcast | AcceptMyPhys, ioaddr + RxMode);
 		iowrite16(RxEnable, ioaddr + MACCtrl1);
 	}
-	pci_enable_wake(pci_dev, pci_choose_state(pci_dev, state),
-			np->wol_enabled);
-	pci_set_power_state(pci_dev, pci_choose_state(pci_dev, state));
+
+	device_set_wakeup_enable(dev_d, np->wol_enabled);
 
 	return 0;
 }
 
-static int sundance_resume(struct pci_dev *pci_dev)
+static int __maybe_unused sundance_resume(struct device *dev_d)
 {
-	struct net_device *dev = pci_get_drvdata(pci_dev);
+	struct net_device *dev = dev_get_drvdata(dev_d);
 	int err = 0;
 
 	if (!netif_running(dev))
 		return 0;
 
-	pci_set_power_state(pci_dev, PCI_D0);
-	pci_restore_state(pci_dev);
-	pci_enable_wake(pci_dev, PCI_D0, 0);
-
 	err = netdev_open(dev);
 	if (err) {
 		printk(KERN_ERR "%s: Can't resume interface!\n",
@@ -1998,34 +1977,14 @@ out:
 	return err;
 }
 
-#endif /* CONFIG_PM */
+static SIMPLE_DEV_PM_OPS(sundance_pm_ops, sundance_suspend, sundance_resume);
 
 static struct pci_driver sundance_driver = {
 	.name		= DRV_NAME,
 	.id_table	= sundance_pci_tbl,
 	.probe		= sundance_probe1,
 	.remove		= sundance_remove1,
-#ifdef CONFIG_PM
-	.suspend	= sundance_suspend,
-	.resume		= sundance_resume,
-#endif /* CONFIG_PM */
+	.driver.pm	= &sundance_pm_ops,
 };
 
-static int __init sundance_init(void)
-{
-/* when a module, this is printed whether or not devices are found in probe */
-#ifdef MODULE
-	printk(version);
-#endif
-	return pci_register_driver(&sundance_driver);
-}
-
-static void __exit sundance_exit(void)
-{
-	pci_unregister_driver(&sundance_driver);
-}
-
-module_init(sundance_init);
-module_exit(sundance_exit);
-
-
+module_pci_driver(sundance_driver);
diff --git a/drivers/net/ethernet/dnet.c b/drivers/net/ethernet/dnet.c
index 3e77dd863175..0de3cd660ec8 100644
--- a/drivers/net/ethernet/dnet.c
+++ b/drivers/net/ethernet/dnet.c
@@ -1,12 +1,9 @@
+// SPDX-License-Identifier: GPL-2.0-only
 /*
  * Dave DNET Ethernet Controller driver
  *
  * Copyright (C) 2008 Dave S.r.l. <www.dave.eu>
  * Copyright (C) 2009 Ilya Yanok, Emcraft Systems Ltd, <yanok@emcraft.com>
- *
- * This program is free software; you can redistribute it and/or modify
- * it under the terms of the GNU General Public License version 2 as
- * published by the Free Software Foundation.
  */
 #include <linux/io.h>
 #include <linux/module.h>
@@ -63,11 +60,11 @@ static void __dnet_set_hwaddr(struct dnet *bp)
 {
 	u16 tmp;
 
-	tmp = be16_to_cpup((__be16 *)bp->dev->dev_addr);
+	tmp = be16_to_cpup((const __be16 *)bp->dev->dev_addr);
 	dnet_writew_mac(bp, DNET_INTERNAL_MAC_ADDR_0_REG, tmp);
-	tmp = be16_to_cpup((__be16 *)(bp->dev->dev_addr + 2));
+	tmp = be16_to_cpup((const __be16 *)(bp->dev->dev_addr + 2));
 	dnet_writew_mac(bp, DNET_INTERNAL_MAC_ADDR_1_REG, tmp);
-	tmp = be16_to_cpup((__be16 *)(bp->dev->dev_addr + 4));
+	tmp = be16_to_cpup((const __be16 *)(bp->dev->dev_addr + 4));
 	dnet_writew_mac(bp, DNET_INTERNAL_MAC_ADDR_2_REG, tmp);
 }
 
@@ -96,7 +93,7 @@ static void dnet_get_hwaddr(struct dnet *bp)
 	*((__be16 *)(addr + 4)) = cpu_to_be16(tmp);
 
 	if (is_valid_ether_addr(addr))
-		memcpy(bp->dev->dev_addr, addr, sizeof(addr));
+		eth_hw_addr_set(bp->dev, addr);
 }
 
 static int dnet_mdio_read(struct mii_bus *bus, int mii_id, int regnum)
@@ -284,13 +281,11 @@ static int dnet_mii_probe(struct net_device *dev)
 
 	/* mask with MAC supported features */
 	if (bp->capabilities & DNET_HAS_GIGABIT)
-		phydev->supported &= PHY_GBIT_FEATURES;
+		phy_set_max_speed(phydev, SPEED_1000);
 	else
-		phydev->supported &= PHY_BASIC_FEATURES;
-
-	phydev->supported |= SUPPORTED_Asym_Pause | SUPPORTED_Pause;
+		phy_set_max_speed(phydev, SPEED_100);
 
-	phydev->advertising = phydev->supported;
+	phy_support_asym_pause(phydev);
 
 	bp->link = 0;
 	bp->speed = 0;
@@ -399,7 +394,7 @@ static int dnet_poll(struct napi_struct *napi, int budget)
 			 * 'skb_put()' points to the start of sk_buff
 			 * data area.
 			 */
-			data_ptr = (unsigned int *)skb_put(skb, pkt_len);
+			data_ptr = skb_put(skb, pkt_len);
 			for (i = 0; i < (pkt_len + 3) >> 2; i++)
 				*data_ptr++ = dnet_readl(bp, RX_DATA_FIFO);
 			skb->protocol = eth_type_trans(skb, dev);
@@ -512,23 +507,20 @@ static netdev_tx_t dnet_start_xmit(struct sk_buff *skb, struct net_device *dev)
 {
 
 	struct dnet *bp = netdev_priv(dev);
-	u32 tx_status, irq_enable;
-	unsigned int len, i, tx_cmd, wrsz;
+	unsigned int i, tx_cmd, wrsz;
 	unsigned long flags;
 	unsigned int *bufp;
+	u32 irq_enable;
 
-	tx_status = dnet_readl(bp, TX_STATUS);
+	dnet_readl(bp, TX_STATUS);
 
 	pr_debug("start_xmit: len %u head %p data %p\n",
 	       skb->len, skb->head, skb->data);
 	dnet_print_skb(skb);
 
-	/* frame size (words) */
-	len = (skb->len + 3) >> 2;
-
 	spin_lock_irqsave(&bp->lock, flags);
 
-	tx_status = dnet_readl(bp, TX_STATUS);
+	dnet_readl(bp, TX_STATUS);
 
 	bufp = (unsigned int *)(((unsigned long) skb->data) & ~0x3UL);
 	wrsz = (u32) skb->len + 3;
@@ -550,7 +542,7 @@ static netdev_tx_t dnet_start_xmit(struct sk_buff *skb, struct net_device *dev)
 
 	if (dnet_readl(bp, TX_FIFO_WCNT) > DNET_FIFO_TX_DATA_AF_TH) {
 		netif_stop_queue(dev);
-		tx_status = dnet_readl(bp, INTR_SRC);
+		dnet_readl(bp, INTR_SRC);
 		irq_enable = dnet_readl(bp, INTR_ENB);
 		irq_enable |= DNET_INTR_ENB_TX_FIFOAE;
 		dnet_writel(bp, irq_enable, INTR_ENB);
@@ -558,11 +550,11 @@ static netdev_tx_t dnet_start_xmit(struct sk_buff *skb, struct net_device *dev)
 
 	skb_tx_timestamp(skb);
 
+	spin_unlock_irqrestore(&bp->lock, flags);
+
 	/* free the buffer */
 	dev_kfree_skb(skb);
 
-	spin_unlock_irqrestore(&bp->lock, flags);
-
 	return NETDEV_TX_OK;
 }
 
@@ -730,25 +722,11 @@ static struct net_device_stats *dnet_get_stats(struct net_device *dev)
 	return nstat;
 }
 
-static int dnet_ioctl(struct net_device *dev, struct ifreq *rq, int cmd)
-{
-	struct phy_device *phydev = dev->phydev;
-
-	if (!netif_running(dev))
-		return -EINVAL;
-
-	if (!phydev)
-		return -ENODEV;
-
-	return phy_mii_ioctl(phydev, rq, cmd);
-}
-
 static void dnet_get_drvinfo(struct net_device *dev,
 			     struct ethtool_drvinfo *info)
 {
-	strlcpy(info->driver, DRV_NAME, sizeof(info->driver));
-	strlcpy(info->version, DRV_VERSION, sizeof(info->version));
-	strlcpy(info->bus_info, "0", sizeof(info->bus_info));
+	strscpy(info->driver, DRV_NAME, sizeof(info->driver));
+	strscpy(info->bus_info, "0", sizeof(info->bus_info));
 }
 
 static const struct ethtool_ops dnet_ethtool_ops = {
@@ -764,7 +742,7 @@ static const struct net_device_ops dnet_netdev_ops = {
 	.ndo_stop		= dnet_close,
 	.ndo_get_stats		= dnet_get_stats,
 	.ndo_start_xmit		= dnet_start_xmit,
-	.ndo_do_ioctl		= dnet_ioctl,
+	.ndo_eth_ioctl		= phy_do_ioctl_running,
 	.ndo_set_mac_address	= eth_mac_addr,
 	.ndo_validate_addr	= eth_validate_addr,
 };
@@ -795,8 +773,7 @@ static int dnet_probe(struct platform_device *pdev)
 
 	spin_lock_init(&bp->lock);
 
-	res = platform_get_resource(pdev, IORESOURCE_MEM, 0);
-	bp->regs = devm_ioremap_resource(&pdev->dev, res);
+	bp->regs = devm_platform_get_and_ioremap_resource(pdev, 0, &res);
 	if (IS_ERR(bp->regs)) {
 		err = PTR_ERR(bp->regs);
 		goto err_out_free_dev;
@@ -811,7 +788,7 @@ static int dnet_probe(struct platform_device *pdev)
 	}
 
 	dev->netdev_ops = &dnet_netdev_ops;
-	netif_napi_add(dev, &bp->napi, dnet_poll, 64);
+	netif_napi_add(dev, &bp->napi, dnet_poll);
 	dev->ethtool_ops = &dnet_ethtool_ops;
 
 	dev->base_addr = (unsigned long)bp->regs;
@@ -864,7 +841,7 @@ err_out_free_dev:
 	return err;
 }
 
-static int dnet_remove(struct platform_device *pdev)
+static void dnet_remove(struct platform_device *pdev)
 {
 
 	struct net_device *dev;
@@ -882,8 +859,6 @@ static int dnet_remove(struct platform_device *pdev)
 		free_irq(dev->irq, dev);
 		free_netdev(dev);
 	}
-
-	return 0;
 }
 
 static struct platform_driver dnet_driver = {
diff --git a/drivers/net/ethernet/dnet.h b/drivers/net/ethernet/dnet.h
index d985080bbd5d..030724484b49 100644
--- a/drivers/net/ethernet/dnet.h
+++ b/drivers/net/ethernet/dnet.h
@@ -1,17 +1,13 @@
+/* SPDX-License-Identifier: GPL-2.0-only */
 /*
  * Dave DNET Ethernet Controller driver
  *
  * Copyright (C) 2008 Dave S.r.l. <www.dave.eu>
- *
- * This program is free software; you can redistribute it and/or modify
- * it under the terms of the GNU General Public License version 2 as
- * published by the Free Software Foundation.
  */
 #ifndef _DNET_H
 #define _DNET_H
 
 #define DRV_NAME		"dnet"
-#define DRV_VERSION		"0.9.1"
 #define PFX				DRV_NAME ": "
 
 /* Register access macros */
diff --git a/drivers/net/ethernet/ec_bhf.c b/drivers/net/ethernet/ec_bhf.c
index 278f139f2a22..67275aa4f65b 100644
--- a/drivers/net/ethernet/ec_bhf.c
+++ b/drivers/net/ethernet/ec_bhf.c
@@ -1,17 +1,8 @@
+// SPDX-License-Identifier: GPL-2.0-only
  /*
  * drivers/net/ethernet/ec_bhf.c
  *
  * Copyright (C) 2014 Darek Marcinkiewicz <reksio@newterm.pl>
- *
- * This software is licensed under the terms of the GNU General Public
- * License version 2, as published by the Free Software Foundation, and
- * may be copied, distributed, and modified under those terms.
- *
- * This program is distributed in the hope that it will be useful,
- * but WITHOUT ANY WARRANTY; without even the implied warranty of
- * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
- * GNU General Public License for more details.
- *
  */
 
 /* This is a driver for EtherCAT master module present on CCAT FPGA.
@@ -73,7 +64,7 @@
 
 #define ETHERCAT_MASTER_ID	0x14
 
-static struct pci_device_id ids[] = {
+static const struct pci_device_id ids[] = {
 	{ PCI_DEVICE(0x15ec, 0x5000), },
 	{ 0, }
 };
@@ -223,7 +214,7 @@ static void ec_bhf_process_rx(struct ec_bhf_priv *priv)
 
 		skb = netdev_alloc_skb_ip_align(priv->net_dev, pkt_size);
 		if (skb) {
-			memcpy(skb_put(skb, pkt_size), data, pkt_size);
+			skb_put_data(skb, data, pkt_size);
 			skb->protocol = eth_type_trans(skb, priv->net_dev);
 			priv->stat_rx_bytes += pkt_size;
 
@@ -425,8 +416,7 @@ static int ec_bhf_open(struct net_device *net_dev)
 
 	netif_start_queue(net_dev);
 
-	hrtimer_init(&priv->hrtimer, CLOCK_MONOTONIC, HRTIMER_MODE_REL);
-	priv->hrtimer.function = ec_bhf_timer_fun;
+	hrtimer_setup(&priv->hrtimer, ec_bhf_timer_fun, CLOCK_MONOTONIC, HRTIMER_MODE_REL);
 	hrtimer_start(&priv->hrtimer, polling_frequency, HRTIMER_MODE_REL);
 
 	return 0;
@@ -488,6 +478,7 @@ static int ec_bhf_probe(struct pci_dev *dev, const struct pci_device_id *id)
 	struct net_device *net_dev;
 	struct ec_bhf_priv *priv;
 	void __iomem *dma_io;
+	u8 addr[ETH_ALEN];
 	void __iomem *io;
 	int err = 0;
 
@@ -497,15 +488,7 @@ static int ec_bhf_probe(struct pci_dev *dev, const struct pci_device_id *id)
 
 	pci_set_master(dev);
 
-	err = pci_set_dma_mask(dev, DMA_BIT_MASK(32));
-	if (err) {
-		dev_err(&dev->dev,
-			"Required dma mask not supported, failed to initialize device\n");
-		err = -EIO;
-		goto err_disable_dev;
-	}
-
-	err = pci_set_consistent_dma_mask(dev, DMA_BIT_MASK(32));
+	err = dma_set_mask_and_coherent(&dev->dev, DMA_BIT_MASK(32));
 	if (err) {
 		dev_err(&dev->dev,
 			"Required dma mask not supported, failed to initialize device\n");
@@ -556,7 +539,8 @@ static int ec_bhf_probe(struct pci_dev *dev, const struct pci_device_id *id)
 	if (err < 0)
 		goto err_free_net_dev;
 
-	memcpy_fromio(net_dev->dev_addr, priv->mii_io + MII_MAC_ADDR, 6);
+	memcpy_fromio(addr, priv->mii_io + MII_MAC_ADDR, ETH_ALEN);
+	eth_hw_addr_set(net_dev, addr);
 
 	err = register_netdev(net_dev);
 	if (err < 0)
@@ -573,7 +557,6 @@ err_unmap:
 err_release_regions:
 	pci_release_regions(dev);
 err_disable_dev:
-	pci_clear_master(dev);
 	pci_disable_device(dev);
 
 	return err;
@@ -585,12 +568,13 @@ static void ec_bhf_remove(struct pci_dev *dev)
 	struct ec_bhf_priv *priv = netdev_priv(net_dev);
 
 	unregister_netdev(net_dev);
-	free_netdev(net_dev);
 
 	pci_iounmap(dev, priv->dma_io);
 	pci_iounmap(dev, priv->io);
+
+	free_netdev(net_dev);
+
 	pci_release_regions(dev);
-	pci_clear_master(dev);
 	pci_disable_device(dev);
 }
 
@@ -602,8 +586,9 @@ static struct pci_driver pci_driver = {
 };
 module_pci_driver(pci_driver);
 
-module_param(polling_frequency, long, S_IRUGO);
+module_param(polling_frequency, long, 0444);
 MODULE_PARM_DESC(polling_frequency, "Polling timer frequency in ns");
 
+MODULE_DESCRIPTION("Beckhoff CX5020 EtherCAT Ethernet driver");
 MODULE_LICENSE("GPL");
 MODULE_AUTHOR("Dariusz Marcinkiewicz <reksio@newterm.pl>");
diff --git a/drivers/net/ethernet/emulex/Kconfig b/drivers/net/ethernet/emulex/Kconfig
index fdbb27ceb02f..5797a76dc731 100644
--- a/drivers/net/ethernet/emulex/Kconfig
+++ b/drivers/net/ethernet/emulex/Kconfig
@@ -1,3 +1,4 @@
+# SPDX-License-Identifier: GPL-2.0-only
 #
 # Emulex driver configuration
 #
@@ -6,7 +7,7 @@ config NET_VENDOR_EMULEX
 	bool "Emulex devices"
 	default y
 	depends on PCI
-	---help---
+	help
 	  If you have a network (Ethernet) card belonging to this class, say Y.
 
 	  Note that the answer to this question doesn't directly affect the
diff --git a/drivers/net/ethernet/emulex/Makefile b/drivers/net/ethernet/emulex/Makefile
index ea8ec574d45a..1a7c5aed6f65 100644
--- a/drivers/net/ethernet/emulex/Makefile
+++ b/drivers/net/ethernet/emulex/Makefile
@@ -1,3 +1,4 @@
+# SPDX-License-Identifier: GPL-2.0-only
 #
 # Makefile for the Emulex device drivers.
 #
diff --git a/drivers/net/ethernet/emulex/benet/Kconfig b/drivers/net/ethernet/emulex/benet/Kconfig
index b4853ec9de8d..f51dca1526c6 100644
--- a/drivers/net/ethernet/emulex/benet/Kconfig
+++ b/drivers/net/ethernet/emulex/benet/Kconfig
@@ -1,7 +1,8 @@
+# SPDX-License-Identifier: GPL-2.0-only
 config BE2NET
 	tristate "ServerEngines' 10Gbps NIC - BladeEngine"
 	depends on PCI
-	---help---
+	help
 	  This driver implements the NIC functionality for ServerEngines'
 	  10Gbps network adapter - BladeEngine.
 
@@ -10,6 +11,42 @@ config BE2NET_HWMON
 	depends on BE2NET && HWMON
 	depends on !(BE2NET=y && HWMON=m)
 	default y
-	---help---
+	help
 	  Say Y here if you want to expose thermal sensor data on
 	  be2net network adapter.
+
+config BE2NET_BE2
+	bool "Support for BE2 chipsets"
+	depends on BE2NET
+	default y
+	help
+	  Say Y here if you want to use devices based on BE2
+	  chipsets. (e.g. OneConnect OCe10xxx)
+
+config BE2NET_BE3
+	bool "Support for BE3 chipsets"
+	depends on BE2NET
+	default y
+	help
+	  Say Y here if you want to use devices based on BE3
+	  chipsets. (e.g. OneConnect OCe11xxx)
+
+config BE2NET_LANCER
+	bool "Support for Lancer chipsets"
+	depends on BE2NET
+	default y
+	help
+	  Say Y here if you want to use devices based on Lancer
+	  chipsets. (e.g LightPulse LPe12xxx)
+
+config BE2NET_SKYHAWK
+	bool "Support for Skyhawk chipsets"
+	depends on BE2NET
+	default y
+	help
+	  Say Y here if you want to use devices based on Skyhawk
+	  chipsets. (e.g. OneConnect OCe14xxx)
+
+comment "WARNING: be2net is useless without any enabled chip"
+	depends on BE2NET_BE2=n && BE2NET_BE3=n && BE2NET_LANCER=n && \
+		BE2NET_SKYHAWK=n && BE2NET
diff --git a/drivers/net/ethernet/emulex/benet/Makefile b/drivers/net/ethernet/emulex/benet/Makefile
index 1a91b276940d..1a238ec7fe1a 100644
--- a/drivers/net/ethernet/emulex/benet/Makefile
+++ b/drivers/net/ethernet/emulex/benet/Makefile
@@ -1,3 +1,4 @@
+# SPDX-License-Identifier: GPL-2.0-only
 #
 # Makefile to build the network driver for ServerEngine's BladeEngine.
 #
diff --git a/drivers/net/ethernet/emulex/benet/be.h b/drivers/net/ethernet/emulex/benet/be.h
index 674cf9d13b98..270ff9aab335 100644
--- a/drivers/net/ethernet/emulex/benet/be.h
+++ b/drivers/net/ethernet/emulex/benet/be.h
@@ -1,12 +1,8 @@
+/* SPDX-License-Identifier: GPL-2.0-only */
 /*
  * Copyright (C) 2005 - 2016 Broadcom
  * All rights reserved.
  *
- * This program is free software; you can redistribute it and/or
- * modify it under the terms of the GNU General Public License version 2
- * as published by the Free Software Foundation.  The full GNU General
- * Public License is included in this distribution in the file called COPYING.
- *
  * Contact Information:
  * linux-drivers@emulex.com
  *
@@ -37,7 +33,6 @@
 #include "be_hw.h"
 #include "be_roce.h"
 
-#define DRV_VER			"11.4.0.0"
 #define DRV_NAME		"be2net"
 #define BE_NAME			"Emulex BladeEngine2"
 #define BE3_NAME		"Emulex BladeEngine3"
@@ -106,8 +101,7 @@
 #define MAX_ROCE_EQS		5
 #define MAX_MSIX_VECTORS	32
 #define MIN_MSIX_VECTORS	1
-#define BE_NAPI_WEIGHT		64
-#define MAX_RX_POST		BE_NAPI_WEIGHT /* Frags posted at a time */
+#define MAX_RX_POST		NAPI_POLL_WEIGHT /* Frags posted at a time */
 #define RX_FRAGS_REFILL_WM	(RX_Q_LEN - MAX_RX_POST)
 #define MAX_NUM_POST_ERX_DB	255u
 
@@ -187,36 +181,15 @@ struct be_eq_obj {
 	struct be_queue_info q;
 	char desc[32];
 
-	/* Adaptive interrupt coalescing (AIC) info */
-	bool enable_aic;
-	u32 min_eqd;		/* in usecs */
-	u32 max_eqd;		/* in usecs */
-	u32 eqd;		/* configured val when aic is off */
-	u32 cur_eqd;		/* in usecs */
-
+	struct be_adapter *adapter;
+	struct napi_struct napi;
 	u8 idx;			/* array index */
 	u8 msix_idx;
 	u16 spurious_intr;
-	struct napi_struct napi;
-	struct be_adapter *adapter;
 	cpumask_var_t  affinity_mask;
-
-#ifdef CONFIG_NET_RX_BUSY_POLL
-#define BE_EQ_IDLE		0
-#define BE_EQ_NAPI		1	/* napi owns this EQ */
-#define BE_EQ_POLL		2	/* poll owns this EQ */
-#define BE_EQ_LOCKED		(BE_EQ_NAPI | BE_EQ_POLL)
-#define BE_EQ_NAPI_YIELD	4	/* napi yielded this EQ */
-#define BE_EQ_POLL_YIELD	8	/* poll yielded this EQ */
-#define BE_EQ_YIELD		(BE_EQ_NAPI_YIELD | BE_EQ_POLL_YIELD)
-#define BE_EQ_USER_PEND		(BE_EQ_POLL | BE_EQ_POLL_YIELD)
-	unsigned int state;
-	spinlock_t lock;	/* lock to serialize napi and busy-poll */
-#endif  /* CONFIG_NET_RX_BUSY_POLL */
 } ____cacheline_aligned_in_smp;
 
 struct be_aic_obj {		/* Adaptive interrupt coalescing (AIC) info */
-	bool enable;
 	u32 min_eqd;		/* in usecs */
 	u32 max_eqd;		/* in usecs */
 	u32 prev_eqd;		/* in usecs */
@@ -238,7 +211,6 @@ struct be_tx_stats {
 	u64 tx_vxlan_offload_pkts;
 	u64 tx_reqs;
 	u64 tx_compl;
-	ulong tx_jiffies;
 	u32 tx_stops;
 	u32 tx_drv_drops;	/* pkts dropped by driver */
 	/* the error counters are described in be_ethtool.c */
@@ -248,6 +220,7 @@ struct be_tx_stats {
 	u32 tx_spoof_check_err;
 	u32 tx_qinq_err;
 	u32 tx_internal_parity_err;
+	u32 tx_sge_err;
 	struct u64_stats_sync sync;
 	struct u64_stats_sync sync_compl;
 };
@@ -260,9 +233,9 @@ struct be_tx_compl_info {
 
 struct be_tx_obj {
 	u32 db_offset;
+	struct be_tx_compl_info txcp;
 	struct be_queue_info q;
 	struct be_queue_info cq;
-	struct be_tx_compl_info txcp;
 	/* Remember the skbs that were transmitted */
 	struct sk_buff *sent_skb_list[TX_Q_LEN];
 	struct be_tx_stats stats;
@@ -457,10 +430,10 @@ struct be_port_resources {
 #define be_is_os2bmc_enabled(adapter) (adapter->flags & BE_FLAGS_OS2BMC)
 
 struct rss_info {
-	u64 rss_flags;
 	u8 rsstable[RSS_INDIR_TABLE_LEN];
 	u8 rss_queue[RSS_INDIR_TABLE_LEN];
 	u8 rss_hkey[RSS_HASH_KEY_LEN];
+	u64 rss_flags;
 };
 
 #define BE_INVALID_DIE_TEMP	0xFF
@@ -543,11 +516,13 @@ enum {
 };
 
 struct be_error_recovery {
-	/* Lancer error recovery variables */
-	u8 recovery_retries;
+	union {
+		u8 recovery_retries;	/* used for Lancer		*/
+		u8 recovery_state;	/* used for BEx and Skyhawk	*/
+	};
 
 	/* BEx/Skyhawk error recovery variables */
-	u8 recovery_state;
+	bool recovery_supported;
 	u16 ue_to_reset_time;		/* Time after UE, to soft reset
 					 * the chip - PF0 only
 					 */
@@ -555,7 +530,6 @@ struct be_error_recovery {
 					 * of SLIPORT_SEMAPHORE reg
 					 */
 	u16 last_err_code;
-	bool recovery_supported;
 	unsigned long probe_time;
 	unsigned long last_recovery_time;
 
@@ -588,7 +562,7 @@ struct be_adapter {
 	struct be_dma_mem mbox_mem_alloced;
 
 	struct be_mcc_obj mcc_obj;
-	struct mutex mcc_lock;	/* For serializing mcc cmds to BE card */
+	spinlock_t mcc_lock;	/* For serializing mcc cmds to BE card */
 	spinlock_t mcc_cq_lock;
 
 	u16 cfg_num_rx_irqs;		/* configured via set-channels */
@@ -612,6 +586,7 @@ struct be_adapter {
 
 	struct be_drv_stats drv_stats;
 	struct be_aic_obj aic_obj[MAX_EVT_QS];
+	bool aic_enabled;
 	u8 vlan_prio_bmap;	/* Available Priority BitMap */
 	u16 recommended_prio_bits;/* Recommended Priority bits in vlan tag */
 	struct be_dma_mem rx_filter; /* Cmd DMA mem for rx-filter */
@@ -678,8 +653,6 @@ struct be_adapter {
 	u8 hba_port_num;
 	u16 pvid;
 	__be16 vxlan_port;		/* offloaded vxlan port num */
-	int vxlan_port_count;		/* active vxlan port count */
-	struct list_head vxlan_port_list;	/* vxlan port list */
 	struct phy_info phy;
 	u8 wol_cap;
 	bool wol_en;
@@ -703,9 +676,6 @@ struct be_adapter {
 struct be_cmd_work {
 	struct work_struct work;
 	struct be_adapter *adapter;
-	union {
-		__be16 vxlan_port;
-	} info;
 };
 
 #define be_physfn(adapter)		(!adapter->virtfn)
@@ -772,17 +742,33 @@ static inline u16 be_max_any_irqs(struct be_adapter *adapter)
 /* Is BE in QNQ multi-channel mode */
 #define be_is_qnq_mode(adapter)		(adapter->function_mode & QNQ_MODE)
 
+#ifdef CONFIG_BE2NET_LANCER
 #define lancer_chip(adapter)	(adapter->pdev->device == OC_DEVICE_ID3 || \
 				 adapter->pdev->device == OC_DEVICE_ID4)
+#else
+#define lancer_chip(adapter)	(0)
+#endif /* CONFIG_BE2NET_LANCER */
 
+#ifdef CONFIG_BE2NET_SKYHAWK
 #define skyhawk_chip(adapter)	(adapter->pdev->device == OC_DEVICE_ID5 || \
 				 adapter->pdev->device == OC_DEVICE_ID6)
+#else
+#define skyhawk_chip(adapter)	(0)
+#endif /* CONFIG_BE2NET_SKYHAWK */
 
+#ifdef CONFIG_BE2NET_BE3
 #define BE3_chip(adapter)	(adapter->pdev->device == BE_DEVICE_ID2 || \
 				 adapter->pdev->device == OC_DEVICE_ID2)
+#else
+#define BE3_chip(adapter)	(0)
+#endif /* CONFIG_BE2NET_BE3 */
 
+#ifdef CONFIG_BE2NET_BE2
 #define BE2_chip(adapter)	(adapter->pdev->device == BE_DEVICE_ID1 || \
 				 adapter->pdev->device == OC_DEVICE_ID1)
+#else
+#define BE2_chip(adapter)	(0)
+#endif /* CONFIG_BE2NET_BE2 */
 
 #define BEx_chip(adapter)	(BE3_chip(adapter) || BE2_chip(adapter))
 
@@ -930,14 +916,24 @@ static inline bool is_ipv4_pkt(struct sk_buff *skb)
 	return skb->protocol == htons(ETH_P_IP) && ip_hdr(skb)->version == 4;
 }
 
+static inline bool is_ipv6_ext_hdr(struct sk_buff *skb)
+{
+	if (ip_hdr(skb)->version == 6)
+		return ipv6_ext_hdr(ipv6_hdr(skb)->nexthdr);
+	else
+		return false;
+}
+
 #define be_error_recovering(adapter)	\
 		(adapter->flags & BE_FLAGS_TRY_RECOVERY)
 
 #define BE_ERROR_EEH		1
 #define BE_ERROR_UE		BIT(1)
 #define BE_ERROR_FW		BIT(2)
-#define BE_ERROR_HW		(BE_ERROR_EEH | BE_ERROR_UE)
-#define BE_ERROR_ANY		(BE_ERROR_EEH | BE_ERROR_UE | BE_ERROR_FW)
+#define BE_ERROR_TX		BIT(3)
+#define BE_ERROR_HW		(BE_ERROR_EEH | BE_ERROR_UE | BE_ERROR_TX)
+#define BE_ERROR_ANY		(BE_ERROR_EEH | BE_ERROR_UE | BE_ERROR_FW | \
+				 BE_ERROR_TX)
 #define BE_CLEAR_ALL		0xFF
 
 static inline u8 be_check_error(struct be_adapter *adapter, u32 err_type)
@@ -970,9 +966,7 @@ void be_cq_notify(struct be_adapter *adapter, u16 qid, bool arm,
 void be_link_status_update(struct be_adapter *adapter, u8 link_status);
 void be_parse_stats(struct be_adapter *adapter);
 int be_load_fw(struct be_adapter *adapter, u8 *func);
-bool be_is_wol_supported(struct be_adapter *adapter);
 bool be_pause_supported(struct be_adapter *adapter);
-u32 be_get_fw_log_level(struct be_adapter *adapter);
 int be_update_queues(struct be_adapter *adapter);
 int be_poll(struct napi_struct *napi, int budget);
 void be_eqd_update(struct be_adapter *adapter, bool force_update);
diff --git a/drivers/net/ethernet/emulex/benet/be_cmds.c b/drivers/net/ethernet/emulex/benet/be_cmds.c
index 02dd5246dfae..bb5d2fa15736 100644
--- a/drivers/net/ethernet/emulex/benet/be_cmds.c
+++ b/drivers/net/ethernet/emulex/benet/be_cmds.c
@@ -1,12 +1,8 @@
+// SPDX-License-Identifier: GPL-2.0-only
 /*
  * Copyright (C) 2005 - 2016 Broadcom
  * All rights reserved.
  *
- * This program is free software; you can redistribute it and/or
- * modify it under the terms of the GNU General Public License version 2
- * as published by the Free Software Foundation.  The full GNU General
- * Public License is included in this distribution in the file called COPYING.
- *
  * Contact Information:
  * linux-drivers@emulex.com
  *
@@ -19,7 +15,7 @@
 #include "be.h"
 #include "be_cmds.h"
 
-char *be_misconfig_evt_port_state[] = {
+const char * const be_misconfig_evt_port_state[] = {
 	"Physical Link is functional",
 	"Optics faulted/incorrectly installed/not installed - Reseat optics. If issue not resolved, replace.",
 	"Optics of two types installed – Remove one optic or install matching pair of optics.",
@@ -103,7 +99,7 @@ static struct be_cmd_priv_map cmd_priv_map[] = {
 static bool be_cmd_allowed(struct be_adapter *adapter, u8 opcode, u8 subsystem)
 {
 	int i;
-	int num_entries = sizeof(cmd_priv_map)/sizeof(struct be_cmd_priv_map);
+	int num_entries = ARRAY_SIZE(cmd_priv_map);
 	u32 cmd_privileges = adapter->cmd_privileges;
 
 	for (i = 0; i < num_entries; i++)
@@ -139,7 +135,8 @@ static int be_mcc_notify(struct be_adapter *adapter)
 
 /* To check if valid bit is set, check the entire word as we don't know
  * the endianness of the data (old entry is host endian while a new entry is
- * little endian) */
+ * little endian)
+ */
 static inline bool be_mcc_compl_is_new(struct be_mcc_compl *compl)
 {
 	u32 flags;
@@ -252,7 +249,8 @@ static int be_mcc_compl_process(struct be_adapter *adapter,
 	u8 opcode = 0, subsystem = 0;
 
 	/* Just swap the status to host endian; mcc tag is opaquely copied
-	 * from mcc_wrb */
+	 * from mcc_wrb
+	 */
 	be_dws_le_to_cpu(compl, 4);
 
 	base_status = base_status(compl->status);
@@ -577,7 +575,7 @@ int be_process_mcc(struct be_adapter *adapter)
 /* Wait till no more pending mcc requests are present */
 static int be_mcc_wait_compl(struct be_adapter *adapter)
 {
-#define mcc_timeout		12000 /* 12s timeout */
+#define mcc_timeout		120000 /* 12s timeout */
 	int i, status = 0;
 	struct be_mcc_obj *mcc_obj = &adapter->mcc_obj;
 
@@ -591,7 +589,7 @@ static int be_mcc_wait_compl(struct be_adapter *adapter)
 
 		if (atomic_read(&mcc_obj->q.used) == 0)
 			break;
-		usleep_range(500, 1000);
+		udelay(100);
 	}
 	if (i == mcc_timeout) {
 		dev_err(&adapter->pdev->dev, "FW not responding\n");
@@ -661,8 +659,7 @@ static int be_mbox_db_ready_wait(struct be_adapter *adapter, void __iomem *db)
 	return 0;
 }
 
-/*
- * Insert the mailbox address into the doorbell in two steps
+/* Insert the mailbox address into the doorbell in two steps
  * Polls on the mbox doorbell till a command completion (or a timeout) occurs
  */
 static int be_mbox_notify_wait(struct be_adapter *adapter)
@@ -806,7 +803,7 @@ static void be_wrb_cmd_hdr_prepare(struct be_cmd_req_hdr *req_hdr,
 	req_hdr->subsystem = subsystem;
 	req_hdr->request_length = cpu_to_le32(cmd_len - sizeof(*req_hdr));
 	req_hdr->version = 0;
-	fill_wrb_tags(wrb, (ulong) req_hdr);
+	fill_wrb_tags(wrb, (ulong)req_hdr);
 	wrb->payload_length = cmd_len;
 	if (mem) {
 		wrb->embedded |= (1 & MCC_WRB_SGE_CNT_MASK) <<
@@ -836,8 +833,8 @@ static void be_cmd_page_addrs_prepare(struct phys_addr *pages, u32 max_pages,
 static inline struct be_mcc_wrb *wrb_from_mbox(struct be_adapter *adapter)
 {
 	struct be_dma_mem *mbox_mem = &adapter->mbox_mem;
-	struct be_mcc_wrb *wrb
-		= &((struct be_mcc_mailbox *)(mbox_mem->va))->wrb;
+	struct be_mcc_wrb *wrb = &((struct be_mcc_mailbox *)(mbox_mem->va))->wrb;
+
 	memset(wrb, 0, sizeof(*wrb));
 	return wrb;
 }
@@ -869,7 +866,7 @@ static bool use_mcc(struct be_adapter *adapter)
 static int be_cmd_lock(struct be_adapter *adapter)
 {
 	if (use_mcc(adapter)) {
-		mutex_lock(&adapter->mcc_lock);
+		spin_lock_bh(&adapter->mcc_lock);
 		return 0;
 	} else {
 		return mutex_lock_interruptible(&adapter->mbox_lock);
@@ -880,7 +877,7 @@ static int be_cmd_lock(struct be_adapter *adapter)
 static void be_cmd_unlock(struct be_adapter *adapter)
 {
 	if (use_mcc(adapter))
-		return mutex_unlock(&adapter->mcc_lock);
+		return spin_unlock_bh(&adapter->mcc_lock);
 	else
 		return mutex_unlock(&adapter->mbox_lock);
 }
@@ -900,7 +897,7 @@ static struct be_mcc_wrb *be_cmd_copy(struct be_adapter *adapter,
 
 	memcpy(dest_wrb, wrb, sizeof(*wrb));
 	if (wrb->embedded & cpu_to_le32(MCC_WRB_EMBEDDED_MASK))
-		fill_wrb_tags(dest_wrb, (ulong) embedded_payload(wrb));
+		fill_wrb_tags(dest_wrb, (ulong)embedded_payload(wrb));
 
 	return dest_wrb;
 }
@@ -1050,7 +1047,7 @@ int be_cmd_mac_addr_query(struct be_adapter *adapter, u8 *mac_addr,
 	struct be_cmd_req_mac_query *req;
 	int status;
 
-	mutex_lock(&adapter->mcc_lock);
+	spin_lock_bh(&adapter->mcc_lock);
 
 	wrb = wrb_from_mccq(adapter);
 	if (!wrb) {
@@ -1079,19 +1076,19 @@ int be_cmd_mac_addr_query(struct be_adapter *adapter, u8 *mac_addr,
 	}
 
 err:
-	mutex_unlock(&adapter->mcc_lock);
+	spin_unlock_bh(&adapter->mcc_lock);
 	return status;
 }
 
 /* Uses synchronous MCCQ */
-int be_cmd_pmac_add(struct be_adapter *adapter, u8 *mac_addr,
+int be_cmd_pmac_add(struct be_adapter *adapter, const u8 *mac_addr,
 		    u32 if_id, u32 *pmac_id, u32 domain)
 {
 	struct be_mcc_wrb *wrb;
 	struct be_cmd_req_pmac_add *req;
 	int status;
 
-	mutex_lock(&adapter->mcc_lock);
+	spin_lock_bh(&adapter->mcc_lock);
 
 	wrb = wrb_from_mccq(adapter);
 	if (!wrb) {
@@ -1116,9 +1113,9 @@ int be_cmd_pmac_add(struct be_adapter *adapter, u8 *mac_addr,
 	}
 
 err:
-	mutex_unlock(&adapter->mcc_lock);
+	spin_unlock_bh(&adapter->mcc_lock);
 
-	 if (base_status(status) == MCC_STATUS_UNAUTHORIZED_REQUEST)
+	if (base_status(status) == MCC_STATUS_UNAUTHORIZED_REQUEST)
 		status = -EPERM;
 
 	return status;
@@ -1134,7 +1131,7 @@ int be_cmd_pmac_del(struct be_adapter *adapter, u32 if_id, int pmac_id, u32 dom)
 	if (pmac_id == -1)
 		return 0;
 
-	mutex_lock(&adapter->mcc_lock);
+	spin_lock_bh(&adapter->mcc_lock);
 
 	wrb = wrb_from_mccq(adapter);
 	if (!wrb) {
@@ -1154,7 +1151,7 @@ int be_cmd_pmac_del(struct be_adapter *adapter, u32 if_id, int pmac_id, u32 dom)
 	status = be_mcc_notify_wait(adapter);
 
 err:
-	mutex_unlock(&adapter->mcc_lock);
+	spin_unlock_bh(&adapter->mcc_lock);
 	return status;
 }
 
@@ -1417,7 +1414,7 @@ int be_cmd_rxq_create(struct be_adapter *adapter,
 	struct be_dma_mem *q_mem = &rxq->dma_mem;
 	int status;
 
-	mutex_lock(&adapter->mcc_lock);
+	spin_lock_bh(&adapter->mcc_lock);
 
 	wrb = wrb_from_mccq(adapter);
 	if (!wrb) {
@@ -1447,7 +1444,7 @@ int be_cmd_rxq_create(struct be_adapter *adapter,
 	}
 
 err:
-	mutex_unlock(&adapter->mcc_lock);
+	spin_unlock_bh(&adapter->mcc_lock);
 	return status;
 }
 
@@ -1511,7 +1508,7 @@ int be_cmd_rxq_destroy(struct be_adapter *adapter, struct be_queue_info *q)
 	struct be_cmd_req_q_destroy *req;
 	int status;
 
-	mutex_lock(&adapter->mcc_lock);
+	spin_lock_bh(&adapter->mcc_lock);
 
 	wrb = wrb_from_mccq(adapter);
 	if (!wrb) {
@@ -1528,7 +1525,7 @@ int be_cmd_rxq_destroy(struct be_adapter *adapter, struct be_queue_info *q)
 	q->created = false;
 
 err:
-	mutex_unlock(&adapter->mcc_lock);
+	spin_unlock_bh(&adapter->mcc_lock);
 	return status;
 }
 
@@ -1596,7 +1593,7 @@ int be_cmd_get_stats(struct be_adapter *adapter, struct be_dma_mem *nonemb_cmd)
 	struct be_cmd_req_hdr *hdr;
 	int status = 0;
 
-	mutex_lock(&adapter->mcc_lock);
+	spin_lock_bh(&adapter->mcc_lock);
 
 	wrb = wrb_from_mccq(adapter);
 	if (!wrb) {
@@ -1612,7 +1609,7 @@ int be_cmd_get_stats(struct be_adapter *adapter, struct be_dma_mem *nonemb_cmd)
 	/* version 1 of the cmd is not supported only by BE2 */
 	if (BE2_chip(adapter))
 		hdr->version = 0;
-	if (BE3_chip(adapter) || lancer_chip(adapter))
+	else if (BE3_chip(adapter) || lancer_chip(adapter))
 		hdr->version = 1;
 	else
 		hdr->version = 2;
@@ -1624,7 +1621,7 @@ int be_cmd_get_stats(struct be_adapter *adapter, struct be_dma_mem *nonemb_cmd)
 	adapter->stats_cmd_sent = true;
 
 err:
-	mutex_unlock(&adapter->mcc_lock);
+	spin_unlock_bh(&adapter->mcc_lock);
 	return status;
 }
 
@@ -1640,7 +1637,7 @@ int lancer_cmd_get_pport_stats(struct be_adapter *adapter,
 			    CMD_SUBSYSTEM_ETH))
 		return -EPERM;
 
-	mutex_lock(&adapter->mcc_lock);
+	spin_lock_bh(&adapter->mcc_lock);
 
 	wrb = wrb_from_mccq(adapter);
 	if (!wrb) {
@@ -1663,7 +1660,7 @@ int lancer_cmd_get_pport_stats(struct be_adapter *adapter,
 	adapter->stats_cmd_sent = true;
 
 err:
-	mutex_unlock(&adapter->mcc_lock);
+	spin_unlock_bh(&adapter->mcc_lock);
 	return status;
 }
 
@@ -1700,7 +1697,7 @@ int be_cmd_link_status_query(struct be_adapter *adapter, u16 *link_speed,
 	struct be_cmd_req_link_status *req;
 	int status;
 
-	mutex_lock(&adapter->mcc_lock);
+	spin_lock_bh(&adapter->mcc_lock);
 
 	if (link_status)
 		*link_status = LINK_DOWN;
@@ -1739,7 +1736,7 @@ int be_cmd_link_status_query(struct be_adapter *adapter, u16 *link_speed,
 	}
 
 err:
-	mutex_unlock(&adapter->mcc_lock);
+	spin_unlock_bh(&adapter->mcc_lock);
 	return status;
 }
 
@@ -1750,7 +1747,7 @@ int be_cmd_get_die_temperature(struct be_adapter *adapter)
 	struct be_cmd_req_get_cntl_addnl_attribs *req;
 	int status = 0;
 
-	mutex_lock(&adapter->mcc_lock);
+	spin_lock_bh(&adapter->mcc_lock);
 
 	wrb = wrb_from_mccq(adapter);
 	if (!wrb) {
@@ -1765,7 +1762,7 @@ int be_cmd_get_die_temperature(struct be_adapter *adapter)
 
 	status = be_mcc_notify(adapter);
 err:
-	mutex_unlock(&adapter->mcc_lock);
+	spin_unlock_bh(&adapter->mcc_lock);
 	return status;
 }
 
@@ -1807,17 +1804,17 @@ int be_cmd_get_fat_dump(struct be_adapter *adapter, u32 buf_len, void *buf)
 
 	total_size = buf_len;
 
-	get_fat_cmd.size = sizeof(struct be_cmd_req_get_fat) + 60*1024;
-	get_fat_cmd.va = dma_zalloc_coherent(&adapter->pdev->dev,
-					     get_fat_cmd.size,
-					     &get_fat_cmd.dma, GFP_ATOMIC);
+	get_fat_cmd.size = sizeof(struct be_cmd_req_get_fat) + 60 * 1024;
+	get_fat_cmd.va = dma_alloc_coherent(&adapter->pdev->dev,
+					    get_fat_cmd.size,
+					    &get_fat_cmd.dma, GFP_ATOMIC);
 	if (!get_fat_cmd.va)
 		return -ENOMEM;
 
-	mutex_lock(&adapter->mcc_lock);
+	spin_lock_bh(&adapter->mcc_lock);
 
 	while (total_size) {
-		buf_size = min(total_size, (u32)60*1024);
+		buf_size = min(total_size, (u32)60 * 1024);
 		total_size -= buf_size;
 
 		wrb = wrb_from_mccq(adapter);
@@ -1852,9 +1849,9 @@ int be_cmd_get_fat_dump(struct be_adapter *adapter, u32 buf_len, void *buf)
 		log_offset += buf_size;
 	}
 err:
+	spin_unlock_bh(&adapter->mcc_lock);
 	dma_free_coherent(&adapter->pdev->dev, get_fat_cmd.size,
 			  get_fat_cmd.va, get_fat_cmd.dma);
-	mutex_unlock(&adapter->mcc_lock);
 	return status;
 }
 
@@ -1865,7 +1862,7 @@ int be_cmd_get_fw_ver(struct be_adapter *adapter)
 	struct be_cmd_req_get_fw_version *req;
 	int status;
 
-	mutex_lock(&adapter->mcc_lock);
+	spin_lock_bh(&adapter->mcc_lock);
 
 	wrb = wrb_from_mccq(adapter);
 	if (!wrb) {
@@ -1882,13 +1879,13 @@ int be_cmd_get_fw_ver(struct be_adapter *adapter)
 	if (!status) {
 		struct be_cmd_resp_get_fw_version *resp = embedded_payload(wrb);
 
-		strlcpy(adapter->fw_ver, resp->firmware_version_string,
+		strscpy(adapter->fw_ver, resp->firmware_version_string,
 			sizeof(adapter->fw_ver));
-		strlcpy(adapter->fw_on_flash, resp->fw_on_flash_version_string,
+		strscpy(adapter->fw_on_flash, resp->fw_on_flash_version_string,
 			sizeof(adapter->fw_on_flash));
 	}
 err:
-	mutex_unlock(&adapter->mcc_lock);
+	spin_unlock_bh(&adapter->mcc_lock);
 	return status;
 }
 
@@ -1902,7 +1899,7 @@ static int __be_cmd_modify_eqd(struct be_adapter *adapter,
 	struct be_cmd_req_modify_eq_delay *req;
 	int status = 0, i;
 
-	mutex_lock(&adapter->mcc_lock);
+	spin_lock_bh(&adapter->mcc_lock);
 
 	wrb = wrb_from_mccq(adapter);
 	if (!wrb) {
@@ -1925,7 +1922,7 @@ static int __be_cmd_modify_eqd(struct be_adapter *adapter,
 
 	status = be_mcc_notify(adapter);
 err:
-	mutex_unlock(&adapter->mcc_lock);
+	spin_unlock_bh(&adapter->mcc_lock);
 	return status;
 }
 
@@ -1952,7 +1949,7 @@ int be_cmd_vlan_config(struct be_adapter *adapter, u32 if_id, u16 *vtag_array,
 	struct be_cmd_req_vlan_config *req;
 	int status;
 
-	mutex_lock(&adapter->mcc_lock);
+	spin_lock_bh(&adapter->mcc_lock);
 
 	wrb = wrb_from_mccq(adapter);
 	if (!wrb) {
@@ -1974,7 +1971,7 @@ int be_cmd_vlan_config(struct be_adapter *adapter, u32 if_id, u16 *vtag_array,
 
 	status = be_mcc_notify_wait(adapter);
 err:
-	mutex_unlock(&adapter->mcc_lock);
+	spin_unlock_bh(&adapter->mcc_lock);
 	return status;
 }
 
@@ -1985,7 +1982,7 @@ static int __be_cmd_rx_filter(struct be_adapter *adapter, u32 flags, u32 value)
 	struct be_cmd_req_rx_filter *req = mem->va;
 	int status;
 
-	mutex_lock(&adapter->mcc_lock);
+	spin_lock_bh(&adapter->mcc_lock);
 
 	wrb = wrb_from_mccq(adapter);
 	if (!wrb) {
@@ -2018,7 +2015,7 @@ static int __be_cmd_rx_filter(struct be_adapter *adapter, u32 flags, u32 value)
 
 	status = be_mcc_notify_wait(adapter);
 err:
-	mutex_unlock(&adapter->mcc_lock);
+	spin_unlock_bh(&adapter->mcc_lock);
 	return status;
 }
 
@@ -2049,7 +2046,7 @@ int be_cmd_set_flow_control(struct be_adapter *adapter, u32 tx_fc, u32 rx_fc)
 			    CMD_SUBSYSTEM_COMMON))
 		return -EPERM;
 
-	mutex_lock(&adapter->mcc_lock);
+	spin_lock_bh(&adapter->mcc_lock);
 
 	wrb = wrb_from_mccq(adapter);
 	if (!wrb) {
@@ -2069,7 +2066,7 @@ int be_cmd_set_flow_control(struct be_adapter *adapter, u32 tx_fc, u32 rx_fc)
 	status = be_mcc_notify_wait(adapter);
 
 err:
-	mutex_unlock(&adapter->mcc_lock);
+	spin_unlock_bh(&adapter->mcc_lock);
 
 	if (base_status(status) == MCC_STATUS_FEATURE_NOT_SUPPORTED)
 		return  -EOPNOTSUPP;
@@ -2088,7 +2085,7 @@ int be_cmd_get_flow_control(struct be_adapter *adapter, u32 *tx_fc, u32 *rx_fc)
 			    CMD_SUBSYSTEM_COMMON))
 		return -EPERM;
 
-	mutex_lock(&adapter->mcc_lock);
+	spin_lock_bh(&adapter->mcc_lock);
 
 	wrb = wrb_from_mccq(adapter);
 	if (!wrb) {
@@ -2111,7 +2108,7 @@ int be_cmd_get_flow_control(struct be_adapter *adapter, u32 *tx_fc, u32 *rx_fc)
 	}
 
 err:
-	mutex_unlock(&adapter->mcc_lock);
+	spin_unlock_bh(&adapter->mcc_lock);
 	return status;
 }
 
@@ -2192,7 +2189,7 @@ int be_cmd_rss_config(struct be_adapter *adapter, u8 *rsstable,
 	if (!(be_if_cap_flags(adapter) & BE_IF_FLAGS_RSS))
 		return 0;
 
-	mutex_lock(&adapter->mcc_lock);
+	spin_lock_bh(&adapter->mcc_lock);
 
 	wrb = wrb_from_mccq(adapter);
 	if (!wrb) {
@@ -2217,7 +2214,7 @@ int be_cmd_rss_config(struct be_adapter *adapter, u8 *rsstable,
 
 	status = be_mcc_notify_wait(adapter);
 err:
-	mutex_unlock(&adapter->mcc_lock);
+	spin_unlock_bh(&adapter->mcc_lock);
 	return status;
 }
 
@@ -2229,7 +2226,7 @@ int be_cmd_set_beacon_state(struct be_adapter *adapter, u8 port_num,
 	struct be_cmd_req_enable_disable_beacon *req;
 	int status;
 
-	mutex_lock(&adapter->mcc_lock);
+	spin_lock_bh(&adapter->mcc_lock);
 
 	wrb = wrb_from_mccq(adapter);
 	if (!wrb) {
@@ -2250,7 +2247,7 @@ int be_cmd_set_beacon_state(struct be_adapter *adapter, u8 port_num,
 	status = be_mcc_notify_wait(adapter);
 
 err:
-	mutex_unlock(&adapter->mcc_lock);
+	spin_unlock_bh(&adapter->mcc_lock);
 	return status;
 }
 
@@ -2261,7 +2258,7 @@ int be_cmd_get_beacon_state(struct be_adapter *adapter, u8 port_num, u32 *state)
 	struct be_cmd_req_get_beacon_state *req;
 	int status;
 
-	mutex_lock(&adapter->mcc_lock);
+	spin_lock_bh(&adapter->mcc_lock);
 
 	wrb = wrb_from_mccq(adapter);
 	if (!wrb) {
@@ -2285,13 +2282,13 @@ int be_cmd_get_beacon_state(struct be_adapter *adapter, u8 port_num, u32 *state)
 	}
 
 err:
-	mutex_unlock(&adapter->mcc_lock);
+	spin_unlock_bh(&adapter->mcc_lock);
 	return status;
 }
 
 /* Uses sync mcc */
 int be_cmd_read_port_transceiver_data(struct be_adapter *adapter,
-				      u8 page_num, u8 *data)
+				      u8 page_num, u32 off, u32 len, u8 *data)
 {
 	struct be_dma_mem cmd;
 	struct be_mcc_wrb *wrb;
@@ -2302,14 +2299,14 @@ int be_cmd_read_port_transceiver_data(struct be_adapter *adapter,
 		return -EINVAL;
 
 	cmd.size = sizeof(struct be_cmd_resp_port_type);
-	cmd.va = dma_zalloc_coherent(&adapter->pdev->dev, cmd.size, &cmd.dma,
-				     GFP_ATOMIC);
+	cmd.va = dma_alloc_coherent(&adapter->pdev->dev, cmd.size, &cmd.dma,
+				    GFP_ATOMIC);
 	if (!cmd.va) {
 		dev_err(&adapter->pdev->dev, "Memory allocation failed\n");
 		return -ENOMEM;
 	}
 
-	mutex_lock(&adapter->mcc_lock);
+	spin_lock_bh(&adapter->mcc_lock);
 
 	wrb = wrb_from_mccq(adapter);
 	if (!wrb) {
@@ -2325,13 +2322,13 @@ int be_cmd_read_port_transceiver_data(struct be_adapter *adapter,
 	req->port = cpu_to_le32(adapter->hba_port_num);
 	req->page_num = cpu_to_le32(page_num);
 	status = be_mcc_notify_wait(adapter);
-	if (!status) {
+	if (!status && len > 0) {
 		struct be_cmd_resp_port_type *resp = cmd.va;
 
-		memcpy(data, resp->page_data, PAGE_DATA_LEN);
+		memcpy(data, resp->page_data + off, len);
 	}
 err:
-	mutex_unlock(&adapter->mcc_lock);
+	spin_unlock_bh(&adapter->mcc_lock);
 	dma_free_coherent(&adapter->pdev->dev, cmd.size, cmd.va, cmd.dma);
 	return status;
 }
@@ -2348,7 +2345,7 @@ static int lancer_cmd_write_object(struct be_adapter *adapter,
 	void *ctxt = NULL;
 	int status;
 
-	mutex_lock(&adapter->mcc_lock);
+	spin_lock_bh(&adapter->mcc_lock);
 	adapter->flash_status = 0;
 
 	wrb = wrb_from_mccq(adapter);
@@ -2377,7 +2374,7 @@ static int lancer_cmd_write_object(struct be_adapter *adapter,
 
 	be_dws_cpu_to_le(ctxt, sizeof(req->context));
 	req->write_offset = cpu_to_le32(data_offset);
-	strlcpy(req->object_name, obj_name, sizeof(req->object_name));
+	strscpy(req->object_name, obj_name, sizeof(req->object_name));
 	req->descriptor_count = cpu_to_le32(1);
 	req->buf_len = cpu_to_le32(data_size);
 	req->addr_low = cpu_to_le32((cmd->dma +
@@ -2390,7 +2387,7 @@ static int lancer_cmd_write_object(struct be_adapter *adapter,
 	if (status)
 		goto err_unlock;
 
-	mutex_unlock(&adapter->mcc_lock);
+	spin_unlock_bh(&adapter->mcc_lock);
 
 	if (!wait_for_completion_timeout(&adapter->et_cmd_compl,
 					 msecs_to_jiffies(60000)))
@@ -2409,7 +2406,7 @@ static int lancer_cmd_write_object(struct be_adapter *adapter,
 	return status;
 
 err_unlock:
-	mutex_unlock(&adapter->mcc_lock);
+	spin_unlock_bh(&adapter->mcc_lock);
 	return status;
 }
 
@@ -2419,7 +2416,7 @@ int be_cmd_query_cable_type(struct be_adapter *adapter)
 	int status;
 
 	status = be_cmd_read_port_transceiver_data(adapter, TR_PAGE_A0,
-						   page_data);
+						   0, PAGE_DATA_LEN, page_data);
 	if (!status) {
 		switch (adapter->phy.interface_type) {
 		case PHY_TYPE_QSFP:
@@ -2444,11 +2441,11 @@ int be_cmd_query_sfp_info(struct be_adapter *adapter)
 	int status;
 
 	status = be_cmd_read_port_transceiver_data(adapter, TR_PAGE_A0,
-						   page_data);
+						   0, PAGE_DATA_LEN, page_data);
 	if (!status) {
-		strlcpy(adapter->phy.vendor_name, page_data +
+		strscpy(adapter->phy.vendor_name, page_data +
 			SFP_VENDOR_NAME_OFFSET, SFP_VENDOR_NAME_LEN - 1);
-		strlcpy(adapter->phy.vendor_pn,
+		strscpy(adapter->phy.vendor_pn,
 			page_data + SFP_VENDOR_PN_OFFSET,
 			SFP_VENDOR_NAME_LEN - 1);
 	}
@@ -2463,7 +2460,7 @@ static int lancer_cmd_delete_object(struct be_adapter *adapter,
 	struct be_mcc_wrb *wrb;
 	int status;
 
-	mutex_lock(&adapter->mcc_lock);
+	spin_lock_bh(&adapter->mcc_lock);
 
 	wrb = wrb_from_mccq(adapter);
 	if (!wrb) {
@@ -2477,11 +2474,11 @@ static int lancer_cmd_delete_object(struct be_adapter *adapter,
 			       OPCODE_COMMON_DELETE_OBJECT,
 			       sizeof(*req), wrb, NULL);
 
-	strlcpy(req->object_name, obj_name, sizeof(req->object_name));
+	strscpy(req->object_name, obj_name, sizeof(req->object_name));
 
 	status = be_mcc_notify_wait(adapter);
 err:
-	mutex_unlock(&adapter->mcc_lock);
+	spin_unlock_bh(&adapter->mcc_lock);
 	return status;
 }
 
@@ -2494,7 +2491,7 @@ int lancer_cmd_read_object(struct be_adapter *adapter, struct be_dma_mem *cmd,
 	struct lancer_cmd_resp_read_object *resp;
 	int status;
 
-	mutex_lock(&adapter->mcc_lock);
+	spin_lock_bh(&adapter->mcc_lock);
 
 	wrb = wrb_from_mccq(adapter);
 	if (!wrb) {
@@ -2528,7 +2525,7 @@ int lancer_cmd_read_object(struct be_adapter *adapter, struct be_dma_mem *cmd,
 	}
 
 err_unlock:
-	mutex_unlock(&adapter->mcc_lock);
+	spin_unlock_bh(&adapter->mcc_lock);
 	return status;
 }
 
@@ -2540,7 +2537,7 @@ static int be_cmd_write_flashrom(struct be_adapter *adapter,
 	struct be_cmd_write_flashrom *req;
 	int status;
 
-	mutex_lock(&adapter->mcc_lock);
+	spin_lock_bh(&adapter->mcc_lock);
 	adapter->flash_status = 0;
 
 	wrb = wrb_from_mccq(adapter);
@@ -2565,7 +2562,7 @@ static int be_cmd_write_flashrom(struct be_adapter *adapter,
 	if (status)
 		goto err_unlock;
 
-	mutex_unlock(&adapter->mcc_lock);
+	spin_unlock_bh(&adapter->mcc_lock);
 
 	if (!wait_for_completion_timeout(&adapter->et_cmd_compl,
 					 msecs_to_jiffies(40000)))
@@ -2576,7 +2573,7 @@ static int be_cmd_write_flashrom(struct be_adapter *adapter,
 	return status;
 
 err_unlock:
-	mutex_unlock(&adapter->mcc_lock);
+	spin_unlock_bh(&adapter->mcc_lock);
 	return status;
 }
 
@@ -2587,7 +2584,7 @@ static int be_cmd_get_flash_crc(struct be_adapter *adapter, u8 *flashed_crc,
 	struct be_mcc_wrb *wrb;
 	int status;
 
-	mutex_lock(&adapter->mcc_lock);
+	spin_lock_bh(&adapter->mcc_lock);
 
 	wrb = wrb_from_mccq(adapter);
 	if (!wrb) {
@@ -2614,11 +2611,15 @@ static int be_cmd_get_flash_crc(struct be_adapter *adapter, u8 *flashed_crc,
 		memcpy(flashed_crc, req->crc, 4);
 
 err:
-	mutex_unlock(&adapter->mcc_lock);
+	spin_unlock_bh(&adapter->mcc_lock);
 	return status;
 }
 
-static char flash_cookie[2][16] = {"*** SE FLAS", "H DIRECTORY *** "};
+/*
+ * Since the cookie is text, add a parsing-skipped space to keep it from
+ * ever being matched on storage holding this source file.
+ */
+static const char flash_cookie[32] __nonstring = "*** SE FLAS" "H DIRECTORY *** ";
 
 static bool phy_flashing_required(struct be_adapter *adapter)
 {
@@ -2762,7 +2763,7 @@ static int be_flash_BEx(struct be_adapter *adapter,
 	bool crc_match;
 	const u8 *p;
 
-	struct flash_comp gen3_flash_types[] = {
+	static const struct flash_comp gen3_flash_types[] = {
 		{ BE3_ISCSI_PRIMARY_IMAGE_START, OPTYPE_ISCSI_ACTIVE,
 			BE3_COMP_MAX_SIZE, IMAGE_FIRMWARE_ISCSI},
 		{ BE3_REDBOOT_START, OPTYPE_REDBOOT,
@@ -2785,7 +2786,7 @@ static int be_flash_BEx(struct be_adapter *adapter,
 			BE3_PHY_FW_COMP_MAX_SIZE, IMAGE_FIRMWARE_PHY}
 	};
 
-	struct flash_comp gen2_flash_types[] = {
+	static const struct flash_comp gen2_flash_types[] = {
 		{ BE2_ISCSI_PRIMARY_IMAGE_START, OPTYPE_ISCSI_ACTIVE,
 			BE2_COMP_MAX_SIZE, IMAGE_FIRMWARE_ISCSI},
 		{ BE2_REDBOOT_START, OPTYPE_REDBOOT,
@@ -3066,8 +3067,8 @@ int lancer_fw_download(struct be_adapter *adapter,
 
 	flash_cmd.size = sizeof(struct lancer_cmd_req_write_object)
 				+ LANCER_FW_DOWNLOAD_CHUNK;
-	flash_cmd.va = dma_zalloc_coherent(dev, flash_cmd.size,
-					   &flash_cmd.dma, GFP_KERNEL);
+	flash_cmd.va = dma_alloc_coherent(dev, flash_cmd.size, &flash_cmd.dma,
+					  GFP_KERNEL);
 	if (!flash_cmd.va)
 		return -ENOMEM;
 
@@ -3184,8 +3185,8 @@ int be_fw_download(struct be_adapter *adapter, const struct firmware *fw)
 	}
 
 	flash_cmd.size = sizeof(struct be_cmd_write_flashrom);
-	flash_cmd.va = dma_zalloc_coherent(dev, flash_cmd.size, &flash_cmd.dma,
-					   GFP_KERNEL);
+	flash_cmd.va = dma_alloc_coherent(dev, flash_cmd.size, &flash_cmd.dma,
+					  GFP_KERNEL);
 	if (!flash_cmd.va)
 		return -ENOMEM;
 
@@ -3220,7 +3221,7 @@ int be_cmd_enable_magic_wol(struct be_adapter *adapter, u8 *mac,
 	struct be_cmd_req_acpi_wol_magic_config *req;
 	int status;
 
-	mutex_lock(&adapter->mcc_lock);
+	spin_lock_bh(&adapter->mcc_lock);
 
 	wrb = wrb_from_mccq(adapter);
 	if (!wrb) {
@@ -3237,7 +3238,7 @@ int be_cmd_enable_magic_wol(struct be_adapter *adapter, u8 *mac,
 	status = be_mcc_notify_wait(adapter);
 
 err:
-	mutex_unlock(&adapter->mcc_lock);
+	spin_unlock_bh(&adapter->mcc_lock);
 	return status;
 }
 
@@ -3252,7 +3253,7 @@ int be_cmd_set_loopback(struct be_adapter *adapter, u8 port_num,
 			    CMD_SUBSYSTEM_LOWLEVEL))
 		return -EPERM;
 
-	mutex_lock(&adapter->mcc_lock);
+	spin_lock_bh(&adapter->mcc_lock);
 
 	wrb = wrb_from_mccq(adapter);
 	if (!wrb) {
@@ -3275,7 +3276,7 @@ int be_cmd_set_loopback(struct be_adapter *adapter, u8 port_num,
 	if (status)
 		goto err_unlock;
 
-	mutex_unlock(&adapter->mcc_lock);
+	spin_unlock_bh(&adapter->mcc_lock);
 
 	if (!wait_for_completion_timeout(&adapter->et_cmd_compl,
 					 msecs_to_jiffies(SET_LB_MODE_TIMEOUT)))
@@ -3284,7 +3285,7 @@ int be_cmd_set_loopback(struct be_adapter *adapter, u8 port_num,
 	return status;
 
 err_unlock:
-	mutex_unlock(&adapter->mcc_lock);
+	spin_unlock_bh(&adapter->mcc_lock);
 	return status;
 }
 
@@ -3301,7 +3302,7 @@ int be_cmd_loopback_test(struct be_adapter *adapter, u32 port_num,
 			    CMD_SUBSYSTEM_LOWLEVEL))
 		return -EPERM;
 
-	mutex_lock(&adapter->mcc_lock);
+	spin_lock_bh(&adapter->mcc_lock);
 
 	wrb = wrb_from_mccq(adapter);
 	if (!wrb) {
@@ -3327,7 +3328,7 @@ int be_cmd_loopback_test(struct be_adapter *adapter, u32 port_num,
 	if (status)
 		goto err;
 
-	mutex_unlock(&adapter->mcc_lock);
+	spin_unlock_bh(&adapter->mcc_lock);
 
 	wait_for_completion(&adapter->et_cmd_compl);
 	resp = embedded_payload(wrb);
@@ -3335,7 +3336,7 @@ int be_cmd_loopback_test(struct be_adapter *adapter, u32 port_num,
 
 	return status;
 err:
-	mutex_unlock(&adapter->mcc_lock);
+	spin_unlock_bh(&adapter->mcc_lock);
 	return status;
 }
 
@@ -3351,7 +3352,7 @@ int be_cmd_ddr_dma_test(struct be_adapter *adapter, u64 pattern,
 			    CMD_SUBSYSTEM_LOWLEVEL))
 		return -EPERM;
 
-	mutex_lock(&adapter->mcc_lock);
+	spin_lock_bh(&adapter->mcc_lock);
 
 	wrb = wrb_from_mccq(adapter);
 	if (!wrb) {
@@ -3366,7 +3367,7 @@ int be_cmd_ddr_dma_test(struct be_adapter *adapter, u64 pattern,
 	req->pattern = cpu_to_le64(pattern);
 	req->byte_count = cpu_to_le32(byte_cnt);
 	for (i = 0; i < byte_cnt; i++) {
-		req->snd_buff[i] = (u8)(pattern >> (j*8));
+		req->snd_buff[i] = (u8)(pattern >> (j * 8));
 		j++;
 		if (j > 7)
 			j = 0;
@@ -3385,7 +3386,7 @@ int be_cmd_ddr_dma_test(struct be_adapter *adapter, u64 pattern,
 	}
 
 err:
-	mutex_unlock(&adapter->mcc_lock);
+	spin_unlock_bh(&adapter->mcc_lock);
 	return status;
 }
 
@@ -3396,7 +3397,7 @@ int be_cmd_get_seeprom_data(struct be_adapter *adapter,
 	struct be_cmd_req_seeprom_read *req;
 	int status;
 
-	mutex_lock(&adapter->mcc_lock);
+	spin_lock_bh(&adapter->mcc_lock);
 
 	wrb = wrb_from_mccq(adapter);
 	if (!wrb) {
@@ -3412,7 +3413,7 @@ int be_cmd_get_seeprom_data(struct be_adapter *adapter,
 	status = be_mcc_notify_wait(adapter);
 
 err:
-	mutex_unlock(&adapter->mcc_lock);
+	spin_unlock_bh(&adapter->mcc_lock);
 	return status;
 }
 
@@ -3427,7 +3428,7 @@ int be_cmd_get_phy_info(struct be_adapter *adapter)
 			    CMD_SUBSYSTEM_COMMON))
 		return -EPERM;
 
-	mutex_lock(&adapter->mcc_lock);
+	spin_lock_bh(&adapter->mcc_lock);
 
 	wrb = wrb_from_mccq(adapter);
 	if (!wrb) {
@@ -3435,8 +3436,8 @@ int be_cmd_get_phy_info(struct be_adapter *adapter)
 		goto err;
 	}
 	cmd.size = sizeof(struct be_cmd_req_get_phy_info);
-	cmd.va = dma_zalloc_coherent(&adapter->pdev->dev, cmd.size, &cmd.dma,
-				     GFP_ATOMIC);
+	cmd.va = dma_alloc_coherent(&adapter->pdev->dev, cmd.size, &cmd.dma,
+				    GFP_ATOMIC);
 	if (!cmd.va) {
 		dev_err(&adapter->pdev->dev, "Memory alloc failure\n");
 		status = -ENOMEM;
@@ -3472,7 +3473,7 @@ int be_cmd_get_phy_info(struct be_adapter *adapter)
 	}
 	dma_free_coherent(&adapter->pdev->dev, cmd.size, cmd.va, cmd.dma);
 err:
-	mutex_unlock(&adapter->mcc_lock);
+	spin_unlock_bh(&adapter->mcc_lock);
 	return status;
 }
 
@@ -3482,7 +3483,7 @@ static int be_cmd_set_qos(struct be_adapter *adapter, u32 bps, u32 domain)
 	struct be_cmd_req_set_qos *req;
 	int status;
 
-	mutex_lock(&adapter->mcc_lock);
+	spin_lock_bh(&adapter->mcc_lock);
 
 	wrb = wrb_from_mccq(adapter);
 	if (!wrb) {
@@ -3502,7 +3503,7 @@ static int be_cmd_set_qos(struct be_adapter *adapter, u32 bps, u32 domain)
 	status = be_mcc_notify_wait(adapter);
 
 err:
-	mutex_unlock(&adapter->mcc_lock);
+	spin_unlock_bh(&adapter->mcc_lock);
 	return status;
 }
 
@@ -3522,9 +3523,9 @@ int be_cmd_get_cntl_attributes(struct be_adapter *adapter)
 
 	memset(&attribs_cmd, 0, sizeof(struct be_dma_mem));
 	attribs_cmd.size = sizeof(struct be_cmd_resp_cntl_attribs);
-	attribs_cmd.va = dma_zalloc_coherent(&adapter->pdev->dev,
-					     attribs_cmd.size,
-					     &attribs_cmd.dma, GFP_ATOMIC);
+	attribs_cmd.va = dma_alloc_coherent(&adapter->pdev->dev,
+					    attribs_cmd.size,
+					    &attribs_cmd.dma, GFP_ATOMIC);
 	if (!attribs_cmd.va) {
 		dev_err(&adapter->pdev->dev, "Memory allocation failure\n");
 		status = -ENOMEM;
@@ -3614,7 +3615,7 @@ int be_cmd_get_fn_privileges(struct be_adapter *adapter, u32 *privilege,
 	struct be_cmd_req_get_fn_privileges *req;
 	int status;
 
-	mutex_lock(&adapter->mcc_lock);
+	spin_lock_bh(&adapter->mcc_lock);
 
 	wrb = wrb_from_mccq(adapter);
 	if (!wrb) {
@@ -3646,7 +3647,7 @@ int be_cmd_get_fn_privileges(struct be_adapter *adapter, u32 *privilege,
 	}
 
 err:
-	mutex_unlock(&adapter->mcc_lock);
+	spin_unlock_bh(&adapter->mcc_lock);
 	return status;
 }
 
@@ -3658,7 +3659,7 @@ int be_cmd_set_fn_privileges(struct be_adapter *adapter, u32 privileges,
 	struct be_cmd_req_set_fn_privileges *req;
 	int status;
 
-	mutex_lock(&adapter->mcc_lock);
+	spin_lock_bh(&adapter->mcc_lock);
 
 	wrb = wrb_from_mccq(adapter);
 	if (!wrb) {
@@ -3678,7 +3679,7 @@ int be_cmd_set_fn_privileges(struct be_adapter *adapter, u32 privileges,
 
 	status = be_mcc_notify_wait(adapter);
 err:
-	mutex_unlock(&adapter->mcc_lock);
+	spin_unlock_bh(&adapter->mcc_lock);
 	return status;
 }
 
@@ -3699,10 +3700,10 @@ int be_cmd_get_mac_from_list(struct be_adapter *adapter, u8 *mac,
 
 	memset(&get_mac_list_cmd, 0, sizeof(struct be_dma_mem));
 	get_mac_list_cmd.size = sizeof(struct be_cmd_resp_get_mac_list);
-	get_mac_list_cmd.va = dma_zalloc_coherent(&adapter->pdev->dev,
-						  get_mac_list_cmd.size,
-						  &get_mac_list_cmd.dma,
-						  GFP_ATOMIC);
+	get_mac_list_cmd.va = dma_alloc_coherent(&adapter->pdev->dev,
+						 get_mac_list_cmd.size,
+						 &get_mac_list_cmd.dma,
+						 GFP_ATOMIC);
 
 	if (!get_mac_list_cmd.va) {
 		dev_err(&adapter->pdev->dev,
@@ -3710,7 +3711,7 @@ int be_cmd_get_mac_from_list(struct be_adapter *adapter, u8 *mac,
 		return -ENOMEM;
 	}
 
-	mutex_lock(&adapter->mcc_lock);
+	spin_lock_bh(&adapter->mcc_lock);
 
 	wrb = wrb_from_mccq(adapter);
 	if (!wrb) {
@@ -3774,7 +3775,7 @@ int be_cmd_get_mac_from_list(struct be_adapter *adapter, u8 *mac,
 	}
 
 out:
-	mutex_unlock(&adapter->mcc_lock);
+	spin_unlock_bh(&adapter->mcc_lock);
 	dma_free_coherent(&adapter->pdev->dev, get_mac_list_cmd.size,
 			  get_mac_list_cmd.va, get_mac_list_cmd.dma);
 	return status;
@@ -3829,12 +3830,12 @@ int be_cmd_set_mac_list(struct be_adapter *adapter, u8 *mac_array,
 
 	memset(&cmd, 0, sizeof(struct be_dma_mem));
 	cmd.size = sizeof(struct be_cmd_req_set_mac_list);
-	cmd.va = dma_zalloc_coherent(&adapter->pdev->dev, cmd.size, &cmd.dma,
-				     GFP_KERNEL);
+	cmd.va = dma_alloc_coherent(&adapter->pdev->dev, cmd.size, &cmd.dma,
+				    GFP_KERNEL);
 	if (!cmd.va)
 		return -ENOMEM;
 
-	mutex_lock(&adapter->mcc_lock);
+	spin_lock_bh(&adapter->mcc_lock);
 
 	wrb = wrb_from_mccq(adapter);
 	if (!wrb) {
@@ -3850,13 +3851,13 @@ int be_cmd_set_mac_list(struct be_adapter *adapter, u8 *mac_array,
 	req->hdr.domain = domain;
 	req->mac_count = mac_count;
 	if (mac_count)
-		memcpy(req->mac, mac_array, ETH_ALEN*mac_count);
+		memcpy(req->mac, mac_array, ETH_ALEN * mac_count);
 
 	status = be_mcc_notify_wait(adapter);
 
 err:
+	spin_unlock_bh(&adapter->mcc_lock);
 	dma_free_coherent(&adapter->pdev->dev, cmd.size, cmd.va, cmd.dma);
-	mutex_unlock(&adapter->mcc_lock);
 	return status;
 }
 
@@ -3892,7 +3893,7 @@ int be_cmd_set_hsw_config(struct be_adapter *adapter, u16 pvid,
 			    CMD_SUBSYSTEM_COMMON))
 		return -EPERM;
 
-	mutex_lock(&adapter->mcc_lock);
+	spin_lock_bh(&adapter->mcc_lock);
 
 	wrb = wrb_from_mccq(adapter);
 	if (!wrb) {
@@ -3933,7 +3934,7 @@ int be_cmd_set_hsw_config(struct be_adapter *adapter, u16 pvid,
 	status = be_mcc_notify_wait(adapter);
 
 err:
-	mutex_unlock(&adapter->mcc_lock);
+	spin_unlock_bh(&adapter->mcc_lock);
 	return status;
 }
 
@@ -3947,7 +3948,7 @@ int be_cmd_get_hsw_config(struct be_adapter *adapter, u16 *pvid,
 	int status;
 	u16 vid;
 
-	mutex_lock(&adapter->mcc_lock);
+	spin_lock_bh(&adapter->mcc_lock);
 
 	wrb = wrb_from_mccq(adapter);
 	if (!wrb) {
@@ -3994,7 +3995,7 @@ int be_cmd_get_hsw_config(struct be_adapter *adapter, u16 *pvid,
 	}
 
 err:
-	mutex_unlock(&adapter->mcc_lock);
+	spin_unlock_bh(&adapter->mcc_lock);
 	return status;
 }
 
@@ -4035,8 +4036,8 @@ int be_cmd_get_acpi_wol_cap(struct be_adapter *adapter)
 
 	memset(&cmd, 0, sizeof(struct be_dma_mem));
 	cmd.size = sizeof(struct be_cmd_resp_acpi_wol_magic_config_v1);
-	cmd.va = dma_zalloc_coherent(&adapter->pdev->dev, cmd.size, &cmd.dma,
-				     GFP_ATOMIC);
+	cmd.va = dma_alloc_coherent(&adapter->pdev->dev, cmd.size, &cmd.dma,
+				    GFP_ATOMIC);
 	if (!cmd.va) {
 		dev_err(&adapter->pdev->dev, "Memory allocation failure\n");
 		status = -ENOMEM;
@@ -4089,9 +4090,9 @@ int be_cmd_set_fw_log_level(struct be_adapter *adapter, u32 level)
 
 	memset(&extfat_cmd, 0, sizeof(struct be_dma_mem));
 	extfat_cmd.size = sizeof(struct be_cmd_resp_get_ext_fat_caps);
-	extfat_cmd.va = dma_zalloc_coherent(&adapter->pdev->dev,
-					    extfat_cmd.size, &extfat_cmd.dma,
-					    GFP_ATOMIC);
+	extfat_cmd.va = dma_alloc_coherent(&adapter->pdev->dev,
+					   extfat_cmd.size, &extfat_cmd.dma,
+					   GFP_ATOMIC);
 	if (!extfat_cmd.va)
 		return -ENOMEM;
 
@@ -4127,9 +4128,9 @@ int be_cmd_get_fw_log_level(struct be_adapter *adapter)
 
 	memset(&extfat_cmd, 0, sizeof(struct be_dma_mem));
 	extfat_cmd.size = sizeof(struct be_cmd_resp_get_ext_fat_caps);
-	extfat_cmd.va = dma_zalloc_coherent(&adapter->pdev->dev,
-					    extfat_cmd.size, &extfat_cmd.dma,
-					    GFP_ATOMIC);
+	extfat_cmd.va = dma_alloc_coherent(&adapter->pdev->dev,
+					   extfat_cmd.size, &extfat_cmd.dma,
+					   GFP_ATOMIC);
 
 	if (!extfat_cmd.va) {
 		dev_err(&adapter->pdev->dev, "%s: Memory allocation failure\n",
@@ -4193,7 +4194,7 @@ int be_cmd_set_ext_fat_capabilites(struct be_adapter *adapter,
 	struct be_cmd_req_set_ext_fat_caps *req;
 	int status;
 
-	mutex_lock(&adapter->mcc_lock);
+	spin_lock_bh(&adapter->mcc_lock);
 
 	wrb = wrb_from_mccq(adapter);
 	if (!wrb) {
@@ -4209,7 +4210,7 @@ int be_cmd_set_ext_fat_capabilites(struct be_adapter *adapter,
 
 	status = be_mcc_notify_wait(adapter);
 err:
-	mutex_unlock(&adapter->mcc_lock);
+	spin_unlock_bh(&adapter->mcc_lock);
 	return status;
 }
 
@@ -4354,8 +4355,8 @@ int be_cmd_get_func_config(struct be_adapter *adapter, struct be_resources *res)
 
 	memset(&cmd, 0, sizeof(struct be_dma_mem));
 	cmd.size = sizeof(struct be_cmd_resp_get_func_config);
-	cmd.va = dma_zalloc_coherent(&adapter->pdev->dev, cmd.size, &cmd.dma,
-				     GFP_ATOMIC);
+	cmd.va = dma_alloc_coherent(&adapter->pdev->dev, cmd.size, &cmd.dma,
+				    GFP_ATOMIC);
 	if (!cmd.va) {
 		dev_err(&adapter->pdev->dev, "Memory alloc failure\n");
 		status = -ENOMEM;
@@ -4452,8 +4453,8 @@ int be_cmd_get_profile_config(struct be_adapter *adapter,
 
 	memset(&cmd, 0, sizeof(struct be_dma_mem));
 	cmd.size = sizeof(struct be_cmd_resp_get_profile_config);
-	cmd.va = dma_zalloc_coherent(&adapter->pdev->dev, cmd.size, &cmd.dma,
-				     GFP_ATOMIC);
+	cmd.va = dma_alloc_coherent(&adapter->pdev->dev, cmd.size, &cmd.dma,
+				    GFP_ATOMIC);
 	if (!cmd.va)
 		return -ENOMEM;
 
@@ -4500,7 +4501,7 @@ int be_cmd_get_profile_config(struct be_adapter *adapter,
 				port_res->max_vfs += le16_to_cpu(pcie->num_vfs);
 			}
 		}
-		return status;
+		goto err;
 	}
 
 	pcie = be_get_pcie_desc(resp->func_param, desc_count,
@@ -4539,8 +4540,8 @@ static int be_cmd_set_profile_config(struct be_adapter *adapter, void *desc,
 
 	memset(&cmd, 0, sizeof(struct be_dma_mem));
 	cmd.size = sizeof(struct be_cmd_req_set_profile_config);
-	cmd.va = dma_zalloc_coherent(&adapter->pdev->dev, cmd.size, &cmd.dma,
-				     GFP_ATOMIC);
+	cmd.va = dma_alloc_coherent(&adapter->pdev->dev, cmd.size, &cmd.dma,
+				    GFP_ATOMIC);
 	if (!cmd.va)
 		return -ENOMEM;
 
@@ -4687,7 +4688,7 @@ int be_cmd_manage_iface(struct be_adapter *adapter, u32 iface, u8 op)
 	if (iface == 0xFFFFFFFF)
 		return -1;
 
-	mutex_lock(&adapter->mcc_lock);
+	spin_lock_bh(&adapter->mcc_lock);
 
 	wrb = wrb_from_mccq(adapter);
 	if (!wrb) {
@@ -4704,7 +4705,7 @@ int be_cmd_manage_iface(struct be_adapter *adapter, u32 iface, u8 op)
 
 	status = be_mcc_notify_wait(adapter);
 err:
-	mutex_unlock(&adapter->mcc_lock);
+	spin_unlock_bh(&adapter->mcc_lock);
 	return status;
 }
 
@@ -4738,7 +4739,7 @@ int be_cmd_get_if_id(struct be_adapter *adapter, struct be_vf_cfg *vf_cfg,
 	struct be_cmd_resp_get_iface_list *resp;
 	int status;
 
-	mutex_lock(&adapter->mcc_lock);
+	spin_lock_bh(&adapter->mcc_lock);
 
 	wrb = wrb_from_mccq(adapter);
 	if (!wrb) {
@@ -4759,7 +4760,7 @@ int be_cmd_get_if_id(struct be_adapter *adapter, struct be_vf_cfg *vf_cfg,
 	}
 
 err:
-	mutex_unlock(&adapter->mcc_lock);
+	spin_unlock_bh(&adapter->mcc_lock);
 	return status;
 }
 
@@ -4853,7 +4854,7 @@ int be_cmd_enable_vf(struct be_adapter *adapter, u8 domain)
 	if (BEx_chip(adapter))
 		return 0;
 
-	mutex_lock(&adapter->mcc_lock);
+	spin_lock_bh(&adapter->mcc_lock);
 
 	wrb = wrb_from_mccq(adapter);
 	if (!wrb) {
@@ -4871,7 +4872,7 @@ int be_cmd_enable_vf(struct be_adapter *adapter, u8 domain)
 	req->enable = 1;
 	status = be_mcc_notify_wait(adapter);
 err:
-	mutex_unlock(&adapter->mcc_lock);
+	spin_unlock_bh(&adapter->mcc_lock);
 	return status;
 }
 
@@ -4944,7 +4945,7 @@ __be_cmd_set_logical_link_config(struct be_adapter *adapter,
 	u32 link_config = 0;
 	int status;
 
-	mutex_lock(&adapter->mcc_lock);
+	spin_lock_bh(&adapter->mcc_lock);
 
 	wrb = wrb_from_mccq(adapter);
 	if (!wrb) {
@@ -4972,7 +4973,7 @@ __be_cmd_set_logical_link_config(struct be_adapter *adapter,
 
 	status = be_mcc_notify_wait(adapter);
 err:
-	mutex_unlock(&adapter->mcc_lock);
+	spin_unlock_bh(&adapter->mcc_lock);
 	return status;
 }
 
@@ -5003,8 +5004,7 @@ int be_cmd_set_features(struct be_adapter *adapter)
 	struct be_mcc_wrb *wrb;
 	int status;
 
-	if (mutex_lock_interruptible(&adapter->mcc_lock))
-		return -1;
+	spin_lock_bh(&adapter->mcc_lock);
 
 	wrb = wrb_from_mccq(adapter);
 	if (!wrb) {
@@ -5042,7 +5042,7 @@ err:
 		dev_info(&adapter->pdev->dev,
 			 "Adapter does not support HW error recovery\n");
 
-	mutex_unlock(&adapter->mcc_lock);
+	spin_unlock_bh(&adapter->mcc_lock);
 	return status;
 }
 
@@ -5056,7 +5056,7 @@ int be_roce_mcc_cmd(void *netdev_handle, void *wrb_payload,
 	struct be_cmd_resp_hdr *resp;
 	int status;
 
-	mutex_lock(&adapter->mcc_lock);
+	spin_lock_bh(&adapter->mcc_lock);
 
 	wrb = wrb_from_mccq(adapter);
 	if (!wrb) {
@@ -5079,7 +5079,7 @@ int be_roce_mcc_cmd(void *netdev_handle, void *wrb_payload,
 	memcpy(wrb_payload, resp, sizeof(*resp) + resp->response_length);
 	be_dws_le_to_cpu(wrb_payload, sizeof(*resp) + resp->response_length);
 err:
-	mutex_unlock(&adapter->mcc_lock);
+	spin_unlock_bh(&adapter->mcc_lock);
 	return status;
 }
 EXPORT_SYMBOL(be_roce_mcc_cmd);
diff --git a/drivers/net/ethernet/emulex/benet/be_cmds.h b/drivers/net/ethernet/emulex/benet/be_cmds.h
index 09da2d82c2f0..5e2d3ddb5d43 100644
--- a/drivers/net/ethernet/emulex/benet/be_cmds.h
+++ b/drivers/net/ethernet/emulex/benet/be_cmds.h
@@ -1,12 +1,8 @@
+/* SPDX-License-Identifier: GPL-2.0-only */
 /*
  * Copyright (C) 2005 - 2016 Broadcom
  * All rights reserved.
  *
- * This program is free software; you can redistribute it and/or
- * modify it under the terms of the GNU General Public License version 2
- * as published by the Free Software Foundation.  The full GNU General
- * Public License is included in this distribution in the file called COPYING.
- *
  * Contact Information:
  * linux-drivers@emulex.com
  *
@@ -201,7 +197,7 @@ enum {
 			 phy_state == BE_PHY_UNQUALIFIED ||	\
 			 phy_state == BE_PHY_UNCERTIFIED)
 
-extern  char *be_misconfig_evt_port_state[];
+extern const  char * const be_misconfig_evt_port_state[];
 
 /* async event indicating misconfigured port */
 struct be_async_event_misconfig_port {
@@ -1419,7 +1415,7 @@ struct flash_section_entry {
 } __packed;
 
 struct flash_section_info {
-	u8 cookie[32];
+	u8 cookie[32] __nonstring;
 	struct flash_section_hdr fsec_hdr;
 	struct flash_section_entry fsec_entry[32];
 } __packed;
@@ -2385,11 +2381,10 @@ struct be_cmd_req_manage_iface_filters {
 } __packed;
 
 u16 be_POST_stage_get(struct be_adapter *adapter);
-int be_pci_fnum_get(struct be_adapter *adapter);
 int be_fw_wait_ready(struct be_adapter *adapter);
 int be_cmd_mac_addr_query(struct be_adapter *adapter, u8 *mac_addr,
 			  bool permanent, u32 if_handle, u32 pmac_id);
-int be_cmd_pmac_add(struct be_adapter *adapter, u8 *mac_addr, u32 if_id,
+int be_cmd_pmac_add(struct be_adapter *adapter, const u8 *mac_addr, u32 if_id,
 		    u32 *pmac_id, u32 domain);
 int be_cmd_pmac_del(struct be_adapter *adapter, u32 if_id, int pmac_id,
 		    u32 domain);
@@ -2410,7 +2405,6 @@ int be_cmd_q_destroy(struct be_adapter *adapter, struct be_queue_info *q,
 int be_cmd_rxq_destroy(struct be_adapter *adapter, struct be_queue_info *q);
 int be_cmd_link_status_query(struct be_adapter *adapter, u16 *link_speed,
 			     u8 *link_status, u32 dom);
-int be_cmd_reset(struct be_adapter *adapter);
 int be_cmd_get_stats(struct be_adapter *adapter, struct be_dma_mem *nonemb_cmd);
 int lancer_cmd_get_pport_stats(struct be_adapter *adapter,
 			       struct be_dma_mem *nonemb_cmd);
@@ -2431,7 +2425,7 @@ int be_cmd_set_beacon_state(struct be_adapter *adapter, u8 port_num, u8 beacon,
 int be_cmd_get_beacon_state(struct be_adapter *adapter, u8 port_num,
 			    u32 *state);
 int be_cmd_read_port_transceiver_data(struct be_adapter *adapter,
-				      u8 page_num, u8 *data);
+				      u8 page_num, u32 off, u32 len, u8 *data);
 int be_cmd_query_cable_type(struct be_adapter *adapter);
 int be_cmd_query_sfp_info(struct be_adapter *adapter);
 int lancer_cmd_read_object(struct be_adapter *adapter, struct be_dma_mem *cmd,
@@ -2492,7 +2486,6 @@ int lancer_physdev_ctrl(struct be_adapter *adapter, u32 mask);
 int lancer_initiate_dump(struct be_adapter *adapter);
 int lancer_delete_dump(struct be_adapter *adapter);
 bool dump_present(struct be_adapter *adapter);
-int lancer_test_and_set_rdy_state(struct be_adapter *adapter);
 int be_cmd_query_port_name(struct be_adapter *adapter);
 int be_cmd_get_func_config(struct be_adapter *adapter,
 			   struct be_resources *res);
diff --git a/drivers/net/ethernet/emulex/benet/be_ethtool.c b/drivers/net/ethernet/emulex/benet/be_ethtool.c
index 7d1819c9e8cc..f9216326bdfe 100644
--- a/drivers/net/ethernet/emulex/benet/be_ethtool.c
+++ b/drivers/net/ethernet/emulex/benet/be_ethtool.c
@@ -1,12 +1,8 @@
+// SPDX-License-Identifier: GPL-2.0-only
 /*
  * Copyright (C) 2005 - 2016 Broadcom
  * All rights reserved.
  *
- * This program is free software; you can redistribute it and/or
- * modify it under the terms of the GNU General Public License version 2
- * as published by the Free Software Foundation.  The full GNU General
- * Public License is included in this distribution in the file called COPYING.
- *
  * Contact Information:
  * linux-drivers@emulex.com
  *
@@ -27,7 +23,7 @@ struct be_ethtool_stat {
 };
 
 enum {DRVSTAT_TX, DRVSTAT_RX, DRVSTAT};
-#define FIELDINFO(_struct, field) FIELD_SIZEOF(_struct, field), \
+#define FIELDINFO(_struct, field) sizeof_field(_struct, field), \
 					offsetof(_struct, field)
 #define DRVSTAT_TX_INFO(field)	#field, DRVSTAT_TX,\
 					FIELDINFO(struct be_tx_stats, field)
@@ -189,6 +185,7 @@ static const struct be_ethtool_stat et_tx_stats[] = {
 	 * packet data. This counter is applicable only for Lancer adapters.
 	 */
 	{DRVSTAT_TX_INFO(tx_internal_parity_err)},
+	{DRVSTAT_TX_INFO(tx_sge_err)},
 	{DRVSTAT_TX_INFO(tx_bytes)},
 	{DRVSTAT_TX_INFO(tx_pkts)},
 	{DRVSTAT_TX_INFO(tx_vxlan_offload_pkts)},
@@ -223,16 +220,15 @@ static void be_get_drvinfo(struct net_device *netdev,
 {
 	struct be_adapter *adapter = netdev_priv(netdev);
 
-	strlcpy(drvinfo->driver, DRV_NAME, sizeof(drvinfo->driver));
-	strlcpy(drvinfo->version, DRV_VER, sizeof(drvinfo->version));
+	strscpy(drvinfo->driver, DRV_NAME, sizeof(drvinfo->driver));
 	if (!memcmp(adapter->fw_ver, adapter->fw_on_flash, FW_VER_LEN))
-		strlcpy(drvinfo->fw_version, adapter->fw_ver,
+		strscpy(drvinfo->fw_version, adapter->fw_ver,
 			sizeof(drvinfo->fw_version));
 	else
 		snprintf(drvinfo->fw_version, sizeof(drvinfo->fw_version),
 			 "%s [%s]", adapter->fw_ver, adapter->fw_on_flash);
 
-	strlcpy(drvinfo->bus_info, pci_name(adapter->pdev),
+	strscpy(drvinfo->bus_info, pci_name(adapter->pdev),
 		sizeof(drvinfo->bus_info));
 }
 
@@ -273,8 +269,8 @@ static int lancer_cmd_read_file(struct be_adapter *adapter, u8 *file_name,
 	int status = 0;
 
 	read_cmd.size = LANCER_READ_FILE_CHUNK;
-	read_cmd.va = dma_zalloc_coherent(&adapter->pdev->dev, read_cmd.size,
-					  &read_cmd.dma, GFP_ATOMIC);
+	read_cmd.va = dma_alloc_coherent(&adapter->pdev->dev, read_cmd.size,
+					 &read_cmd.dma, GFP_ATOMIC);
 
 	if (!read_cmd.va) {
 		dev_err(&adapter->pdev->dev,
@@ -319,7 +315,9 @@ static int be_read_dump_data(struct be_adapter *adapter, u32 dump_len,
 }
 
 static int be_get_coalesce(struct net_device *netdev,
-			   struct ethtool_coalesce *et)
+			   struct ethtool_coalesce *et,
+			   struct kernel_ethtool_coalesce *kernel_coal,
+			   struct netlink_ext_ack *extack)
 {
 	struct be_adapter *adapter = netdev_priv(netdev);
 	struct be_aic_obj *aic = &adapter->aic_obj[0];
@@ -332,8 +330,8 @@ static int be_get_coalesce(struct net_device *netdev,
 	et->tx_coalesce_usecs_high = aic->max_eqd;
 	et->tx_coalesce_usecs_low = aic->min_eqd;
 
-	et->use_adaptive_rx_coalesce = aic->enable;
-	et->use_adaptive_tx_coalesce = aic->enable;
+	et->use_adaptive_rx_coalesce = adapter->aic_enabled;
+	et->use_adaptive_tx_coalesce = adapter->aic_enabled;
 
 	return 0;
 }
@@ -342,15 +340,18 @@ static int be_get_coalesce(struct net_device *netdev,
  * eqd cmd is issued in the worker thread.
  */
 static int be_set_coalesce(struct net_device *netdev,
-			   struct ethtool_coalesce *et)
+			   struct ethtool_coalesce *et,
+			   struct kernel_ethtool_coalesce *kernel_coal,
+			   struct netlink_ext_ack *extack)
 {
 	struct be_adapter *adapter = netdev_priv(netdev);
 	struct be_aic_obj *aic = &adapter->aic_obj[0];
 	struct be_eq_obj *eqo;
 	int i;
 
+	adapter->aic_enabled = et->use_adaptive_rx_coalesce;
+
 	for_all_evt_queues(adapter, eqo, i) {
-		aic->enable = et->use_adaptive_rx_coalesce;
 		aic->max_eqd = min(et->rx_coalesce_usecs_high, BE_MAX_EQD);
 		aic->min_eqd = min(et->rx_coalesce_usecs_low, aic->max_eqd);
 		aic->et_eqd = min(et->rx_coalesce_usecs, aic->max_eqd);
@@ -388,10 +389,10 @@ static void be_get_ethtool_stats(struct net_device *netdev,
 		struct be_rx_stats *stats = rx_stats(rxo);
 
 		do {
-			start = u64_stats_fetch_begin_irq(&stats->sync);
+			start = u64_stats_fetch_begin(&stats->sync);
 			data[base] = stats->rx_bytes;
 			data[base + 1] = stats->rx_pkts;
-		} while (u64_stats_fetch_retry_irq(&stats->sync, start));
+		} while (u64_stats_fetch_retry(&stats->sync, start));
 
 		for (i = 2; i < ETHTOOL_RXSTATS_NUM; i++) {
 			p = (u8 *)stats + et_rx_stats[i].offset;
@@ -404,19 +405,19 @@ static void be_get_ethtool_stats(struct net_device *netdev,
 		struct be_tx_stats *stats = tx_stats(txo);
 
 		do {
-			start = u64_stats_fetch_begin_irq(&stats->sync_compl);
+			start = u64_stats_fetch_begin(&stats->sync_compl);
 			data[base] = stats->tx_compl;
-		} while (u64_stats_fetch_retry_irq(&stats->sync_compl, start));
+		} while (u64_stats_fetch_retry(&stats->sync_compl, start));
 
 		do {
-			start = u64_stats_fetch_begin_irq(&stats->sync);
+			start = u64_stats_fetch_begin(&stats->sync);
 			for (i = 1; i < ETHTOOL_TXSTATS_NUM; i++) {
 				p = (u8 *)stats + et_tx_stats[i].offset;
 				data[base + i] =
 					(et_tx_stats[i].size == sizeof(u64)) ?
 						*(u64 *)p : *(u32 *)p;
 			}
-		} while (u64_stats_fetch_retry_irq(&stats->sync, start));
+		} while (u64_stats_fetch_retry(&stats->sync, start));
 		base += ETHTOOL_TXSTATS_NUM;
 	}
 }
@@ -574,6 +575,7 @@ static u32 convert_to_et_setting(struct be_adapter *adapter, u32 if_speeds)
 				break;
 			}
 		}
+		fallthrough;
 	case PHY_TYPE_SFP_PLUS_10GB:
 	case PHY_TYPE_XFP_10GB:
 	case PHY_TYPE_SFP_1GB:
@@ -681,7 +683,9 @@ static int be_get_link_ksettings(struct net_device *netdev,
 }
 
 static void be_get_ringparam(struct net_device *netdev,
-			     struct ethtool_ringparam *ring)
+			     struct ethtool_ringparam *ring,
+			     struct kernel_ethtool_ringparam *kernel_ring,
+			     struct netlink_ext_ack *extack)
 {
 	struct be_adapter *adapter = netdev_priv(netdev);
 
@@ -813,7 +817,7 @@ static int be_set_wol(struct net_device *netdev, struct ethtool_wolinfo *wol)
 	}
 
 	cmd.size = sizeof(struct be_cmd_req_acpi_wol_magic_config);
-	cmd.va = dma_zalloc_coherent(dev, cmd.size, &cmd.dma, GFP_KERNEL);
+	cmd.va = dma_alloc_coherent(dev, cmd.size, &cmd.dma, GFP_KERNEL);
 	if (!cmd.va)
 		return -ENOMEM;
 
@@ -849,9 +853,9 @@ static int be_test_ddr_dma(struct be_adapter *adapter)
 	};
 
 	ddrdma_cmd.size = sizeof(struct be_cmd_req_ddrdma_test);
-	ddrdma_cmd.va = dma_zalloc_coherent(&adapter->pdev->dev,
-					    ddrdma_cmd.size, &ddrdma_cmd.dma,
-					    GFP_KERNEL);
+	ddrdma_cmd.va = dma_alloc_coherent(&adapter->pdev->dev,
+					   ddrdma_cmd.size, &ddrdma_cmd.dma,
+					   GFP_KERNEL);
 	if (!ddrdma_cmd.va)
 		return -ENOMEM;
 
@@ -893,7 +897,7 @@ static void be_self_test(struct net_device *netdev, struct ethtool_test *test,
 			 u64 *data)
 {
 	struct be_adapter *adapter = netdev_priv(netdev);
-	int status;
+	int status, cnt;
 	u8 link_status = 0;
 
 	if (adapter->function_caps & BE_FUNCTION_CAPS_SUPER_NIC) {
@@ -904,6 +908,9 @@ static void be_self_test(struct net_device *netdev, struct ethtool_test *test,
 
 	memset(data, 0, sizeof(u64) * ETHTOOL_TESTS_NUM);
 
+	/* check link status before offline tests */
+	link_status = netif_carrier_ok(netdev);
+
 	if (test->flags & ETH_TEST_FL_OFFLINE) {
 		if (be_loopback_test(adapter, BE_MAC_LOOPBACK, &data[0]) != 0)
 			test->flags |= ETH_TEST_FL_FAILED;
@@ -924,13 +931,26 @@ static void be_self_test(struct net_device *netdev, struct ethtool_test *test,
 		test->flags |= ETH_TEST_FL_FAILED;
 	}
 
-	status = be_cmd_link_status_query(adapter, NULL, &link_status, 0);
-	if (status) {
-		test->flags |= ETH_TEST_FL_FAILED;
-		data[4] = -1;
-	} else if (!link_status) {
+	/* link status was down prior to test */
+	if (!link_status) {
 		test->flags |= ETH_TEST_FL_FAILED;
 		data[4] = 1;
+		return;
+	}
+
+	for (cnt = 10; cnt; cnt--) {
+		status = be_cmd_link_status_query(adapter, NULL, &link_status,
+						  0);
+		if (status) {
+			test->flags |= ETH_TEST_FL_FAILED;
+			data[4] = -1;
+			break;
+		}
+
+		if (link_status)
+			break;
+
+		msleep_interruptible(500);
 	}
 }
 
@@ -1012,9 +1032,9 @@ static int be_read_eeprom(struct net_device *netdev,
 
 	memset(&eeprom_cmd, 0, sizeof(struct be_dma_mem));
 	eeprom_cmd.size = sizeof(struct be_cmd_req_seeprom_read);
-	eeprom_cmd.va = dma_zalloc_coherent(&adapter->pdev->dev,
-					    eeprom_cmd.size, &eeprom_cmd.dma,
-					    GFP_KERNEL);
+	eeprom_cmd.va = dma_alloc_coherent(&adapter->pdev->dev,
+					   eeprom_cmd.size, &eeprom_cmd.dma,
+					   GFP_KERNEL);
 
 	if (!eeprom_cmd.va)
 		return -ENOMEM;
@@ -1053,10 +1073,19 @@ static void be_set_msg_level(struct net_device *netdev, u32 level)
 	adapter->msg_enable = level;
 }
 
-static u64 be_get_rss_hash_opts(struct be_adapter *adapter, u64 flow_type)
+static int be_get_rxfh_fields(struct net_device *netdev,
+			      struct ethtool_rxfh_fields *cmd)
 {
+	struct be_adapter *adapter = netdev_priv(netdev);
+	u64 flow_type = cmd->flow_type;
 	u64 data = 0;
 
+	if (!be_multi_rxq(adapter)) {
+		dev_info(&adapter->pdev->dev,
+			 "ethtool::get_rxfh: RX flow hashing is disabled\n");
+		return -EINVAL;
+	}
+
 	switch (flow_type) {
 	case TCP_V4_FLOW:
 		if (adapter->rss_info.rss_flags & RSS_ENABLE_IPV4)
@@ -1084,7 +1113,8 @@ static u64 be_get_rss_hash_opts(struct be_adapter *adapter, u64 flow_type)
 		break;
 	}
 
-	return data;
+	cmd->data = data;
+	return 0;
 }
 
 static int be_get_rxnfc(struct net_device *netdev, struct ethtool_rxnfc *cmd,
@@ -1099,11 +1129,8 @@ static int be_get_rxnfc(struct net_device *netdev, struct ethtool_rxnfc *cmd,
 	}
 
 	switch (cmd->cmd) {
-	case ETHTOOL_GRXFH:
-		cmd->data = be_get_rss_hash_opts(adapter, cmd->flow_type);
-		break;
 	case ETHTOOL_GRXRINGS:
-		cmd->data = adapter->num_rx_qs - 1;
+		cmd->data = adapter->num_rx_qs;
 		break;
 	default:
 		return -EINVAL;
@@ -1112,11 +1139,19 @@ static int be_get_rxnfc(struct net_device *netdev, struct ethtool_rxnfc *cmd,
 	return 0;
 }
 
-static int be_set_rss_hash_opts(struct be_adapter *adapter,
-				struct ethtool_rxnfc *cmd)
+static int be_set_rxfh_fields(struct net_device *netdev,
+			      const struct ethtool_rxfh_fields *cmd,
+			      struct netlink_ext_ack *extack)
 {
-	int status;
+	struct be_adapter *adapter = netdev_priv(netdev);
 	u32 rss_flags = adapter->rss_info.rss_flags;
+	int status;
+
+	if (!be_multi_rxq(adapter)) {
+		dev_err(&adapter->pdev->dev,
+			"ethtool::set_rxfh: RX flow hashing is disabled\n");
+		return -EINVAL;
+	}
 
 	if (cmd->data != L3_RSS_FLAGS &&
 	    cmd->data != (L3_RSS_FLAGS | L4_RSS_FLAGS))
@@ -1175,28 +1210,6 @@ static int be_set_rss_hash_opts(struct be_adapter *adapter,
 	return be_cmd_status(status);
 }
 
-static int be_set_rxnfc(struct net_device *netdev, struct ethtool_rxnfc *cmd)
-{
-	struct be_adapter *adapter = netdev_priv(netdev);
-	int status = 0;
-
-	if (!be_multi_rxq(adapter)) {
-		dev_err(&adapter->pdev->dev,
-			"ethtool::set_rxnfc: RX flow hashing is disabled\n");
-		return -EINVAL;
-	}
-
-	switch (cmd->cmd) {
-	case ETHTOOL_SRXFH:
-		status = be_set_rss_hash_opts(adapter, cmd);
-		break;
-	default:
-		return -EINVAL;
-	}
-
-	return status;
-}
-
 static void be_get_channels(struct net_device *netdev,
 			    struct ethtool_channels *ch)
 {
@@ -1251,43 +1264,45 @@ static u32 be_get_rxfh_key_size(struct net_device *netdev)
 	return RSS_HASH_KEY_LEN;
 }
 
-static int be_get_rxfh(struct net_device *netdev, u32 *indir, u8 *hkey,
-		       u8 *hfunc)
+static int be_get_rxfh(struct net_device *netdev,
+		       struct ethtool_rxfh_param *rxfh)
 {
 	struct be_adapter *adapter = netdev_priv(netdev);
 	int i;
 	struct rss_info *rss = &adapter->rss_info;
 
-	if (indir) {
+	if (rxfh->indir) {
 		for (i = 0; i < RSS_INDIR_TABLE_LEN; i++)
-			indir[i] = rss->rss_queue[i];
+			rxfh->indir[i] = rss->rss_queue[i];
 	}
 
-	if (hkey)
-		memcpy(hkey, rss->rss_hkey, RSS_HASH_KEY_LEN);
+	if (rxfh->key)
+		memcpy(rxfh->key, rss->rss_hkey, RSS_HASH_KEY_LEN);
 
-	if (hfunc)
-		*hfunc = ETH_RSS_HASH_TOP;
+	rxfh->hfunc = ETH_RSS_HASH_TOP;
 
 	return 0;
 }
 
-static int be_set_rxfh(struct net_device *netdev, const u32 *indir,
-		       const u8 *hkey, const u8 hfunc)
+static int be_set_rxfh(struct net_device *netdev,
+		       struct ethtool_rxfh_param *rxfh,
+		       struct netlink_ext_ack *extack)
 {
 	int rc = 0, i, j;
 	struct be_adapter *adapter = netdev_priv(netdev);
+	u8 *hkey = rxfh->key;
 	u8 rsstable[RSS_INDIR_TABLE_LEN];
 
 	/* We do not allow change in unsupported parameters */
-	if (hfunc != ETH_RSS_HASH_NO_CHANGE && hfunc != ETH_RSS_HASH_TOP)
+	if (rxfh->hfunc != ETH_RSS_HASH_NO_CHANGE &&
+	    rxfh->hfunc != ETH_RSS_HASH_TOP)
 		return -EOPNOTSUPP;
 
-	if (indir) {
+	if (rxfh->indir) {
 		struct be_rx_obj *rxo;
 
 		for (i = 0; i < RSS_INDIR_TABLE_LEN; i++) {
-			j = indir[i];
+			j = rxfh->indir[i];
 			rxo = &adapter->rx_obj[j];
 			rsstable[i] = rxo->rss_id;
 			adapter->rss_info.rss_queue[i] = j;
@@ -1324,7 +1339,7 @@ static int be_get_module_info(struct net_device *netdev,
 		return -EOPNOTSUPP;
 
 	status = be_cmd_read_port_transceiver_data(adapter, TR_PAGE_A0,
-						   page_data);
+						   0, PAGE_DATA_LEN, page_data);
 	if (!status) {
 		if (!page_data[SFP_PLUS_SFF_8472_COMP]) {
 			modinfo->type = ETH_MODULE_SFF_8079;
@@ -1342,25 +1357,32 @@ static int be_get_module_eeprom(struct net_device *netdev,
 {
 	struct be_adapter *adapter = netdev_priv(netdev);
 	int status;
+	u32 begin, end;
 
 	if (!check_privilege(adapter, MAX_PRIVILEGES))
 		return -EOPNOTSUPP;
 
-	status = be_cmd_read_port_transceiver_data(adapter, TR_PAGE_A0,
-						   data);
-	if (status)
-		goto err;
+	begin = eeprom->offset;
+	end = eeprom->offset + eeprom->len;
+
+	if (begin < PAGE_DATA_LEN) {
+		status = be_cmd_read_port_transceiver_data(adapter, TR_PAGE_A0, begin,
+							   min_t(u32, end, PAGE_DATA_LEN) - begin,
+							   data);
+		if (status)
+			goto err;
+
+		data += PAGE_DATA_LEN - begin;
+		begin = PAGE_DATA_LEN;
+	}
 
-	if (eeprom->offset + eeprom->len > PAGE_DATA_LEN) {
-		status = be_cmd_read_port_transceiver_data(adapter,
-							   TR_PAGE_A2,
-							   data +
-							   PAGE_DATA_LEN);
+	if (end > PAGE_DATA_LEN) {
+		status = be_cmd_read_port_transceiver_data(adapter, TR_PAGE_A2,
+							   begin - PAGE_DATA_LEN,
+							   end - begin, data);
 		if (status)
 			goto err;
 	}
-	if (eeprom->offset)
-		memcpy(data, data + eeprom->offset, eeprom->len);
 err:
 	return be_cmd_status(status);
 }
@@ -1394,6 +1416,9 @@ static int be_set_priv_flags(struct net_device *netdev, u32 flags)
 }
 
 const struct ethtool_ops be_ethtool_ops = {
+	.supported_coalesce_params = ETHTOOL_COALESCE_USECS |
+				     ETHTOOL_COALESCE_USE_ADAPTIVE |
+				     ETHTOOL_COALESCE_USECS_LOW_HIGH,
 	.get_drvinfo = be_get_drvinfo,
 	.get_wol = be_get_wol,
 	.set_wol = be_set_wol,
@@ -1417,7 +1442,8 @@ const struct ethtool_ops be_ethtool_ops = {
 	.flash_device = be_do_flash,
 	.self_test = be_self_test,
 	.get_rxnfc = be_get_rxnfc,
-	.set_rxnfc = be_set_rxnfc,
+	.get_rxfh_fields = be_get_rxfh_fields,
+	.set_rxfh_fields = be_set_rxfh_fields,
 	.get_rxfh_indir_size = be_get_rxfh_indir_size,
 	.get_rxfh_key_size = be_get_rxfh_key_size,
 	.get_rxfh = be_get_rxfh,
diff --git a/drivers/net/ethernet/emulex/benet/be_hw.h b/drivers/net/ethernet/emulex/benet/be_hw.h
index c967f45705d9..3476194f0855 100644
--- a/drivers/net/ethernet/emulex/benet/be_hw.h
+++ b/drivers/net/ethernet/emulex/benet/be_hw.h
@@ -1,12 +1,8 @@
+/* SPDX-License-Identifier: GPL-2.0-only */
 /*
  * Copyright (C) 2005-2016 Broadcom.
  * All rights reserved.
  *
- * This program is free software; you can redistribute it and/or
- * modify it under the terms of the GNU General Public License version 2
- * as published by the Free Software Foundation.  The full GNU General
- * Public License is included in this distribution in the file called COPYING.
- *
  * Contact Information:
  * linux-drivers@emulex.com
  *
@@ -261,6 +257,7 @@ struct be_eth_hdr_wrb {
 #define LANCER_TX_COMP_HSW_DROP_MAC_ERR		0x3
 #define LANCER_TX_COMP_HSW_DROP_VLAN_ERR	0x5
 #define LANCER_TX_COMP_QINQ_ERR			0x7
+#define LANCER_TX_COMP_SGE_ERR			0x9
 #define LANCER_TX_COMP_PARITY_ERR		0xb
 #define LANCER_TX_COMP_DMA_ERR			0xd
 
diff --git a/drivers/net/ethernet/emulex/benet/be_main.c b/drivers/net/ethernet/emulex/benet/be_main.c
index 319eee36649b..cb004fd16252 100644
--- a/drivers/net/ethernet/emulex/benet/be_main.c
+++ b/drivers/net/ethernet/emulex/benet/be_main.c
@@ -1,12 +1,8 @@
+// SPDX-License-Identifier: GPL-2.0-only
 /*
  * Copyright (C) 2005 - 2016 Broadcom
  * All rights reserved.
  *
- * This program is free software; you can redistribute it and/or
- * modify it under the terms of the GNU General Public License version 2
- * as published by the Free Software Foundation.  The full GNU General
- * Public License is included in this distribution in the file called COPYING.
- *
  * Contact Information:
  * linux-drivers@emulex.com
  *
@@ -20,13 +16,11 @@
 #include "be.h"
 #include "be_cmds.h"
 #include <asm/div64.h>
-#include <linux/aer.h>
 #include <linux/if_bridge.h>
 #include <net/busy_poll.h>
 #include <net/vxlan.h>
 
-MODULE_VERSION(DRV_VER);
-MODULE_DESCRIPTION(DRV_DESC " " DRV_VER);
+MODULE_DESCRIPTION(DRV_DESC);
 MODULE_AUTHOR("Emulex Corporation");
 MODULE_LICENSE("GPL");
 
@@ -34,11 +28,11 @@ MODULE_LICENSE("GPL");
  * Use sysfs method to enable/disable VFs.
  */
 static unsigned int num_vfs;
-module_param(num_vfs, uint, S_IRUGO);
+module_param(num_vfs, uint, 0444);
 MODULE_PARM_DESC(num_vfs, "Number of PCI VFs to initialize");
 
 static ushort rx_frag_size = 2048;
-module_param(rx_frag_size, ushort, S_IRUGO);
+module_param(rx_frag_size, ushort, 0444);
 MODULE_PARM_DESC(rx_frag_size, "Size of a fragment that holds rcvd data.");
 
 /* Per-module error detection/recovery workq shared across all functions.
@@ -47,14 +41,22 @@ MODULE_PARM_DESC(rx_frag_size, "Size of a fragment that holds rcvd data.");
 static struct workqueue_struct *be_err_recovery_workq;
 
 static const struct pci_device_id be_dev_ids[] = {
+#ifdef CONFIG_BE2NET_BE2
 	{ PCI_DEVICE(BE_VENDOR_ID, BE_DEVICE_ID1) },
-	{ PCI_DEVICE(BE_VENDOR_ID, BE_DEVICE_ID2) },
 	{ PCI_DEVICE(BE_VENDOR_ID, OC_DEVICE_ID1) },
+#endif /* CONFIG_BE2NET_BE2 */
+#ifdef CONFIG_BE2NET_BE3
+	{ PCI_DEVICE(BE_VENDOR_ID, BE_DEVICE_ID2) },
 	{ PCI_DEVICE(BE_VENDOR_ID, OC_DEVICE_ID2) },
+#endif /* CONFIG_BE2NET_BE3 */
+#ifdef CONFIG_BE2NET_LANCER
 	{ PCI_DEVICE(EMULEX_VENDOR_ID, OC_DEVICE_ID3)},
 	{ PCI_DEVICE(EMULEX_VENDOR_ID, OC_DEVICE_ID4)},
+#endif /* CONFIG_BE2NET_LANCER */
+#ifdef CONFIG_BE2NET_SKYHAWK
 	{ PCI_DEVICE(EMULEX_VENDOR_ID, OC_DEVICE_ID5)},
 	{ PCI_DEVICE(EMULEX_VENDOR_ID, OC_DEVICE_ID6)},
+#endif /* CONFIG_BE2NET_SKYHAWK */
 	{ 0 }
 };
 MODULE_DEVICE_TABLE(pci, be_dev_ids);
@@ -159,8 +161,8 @@ static int be_queue_alloc(struct be_adapter *adapter, struct be_queue_info *q,
 	q->len = len;
 	q->entry_size = entry_size;
 	mem->size = len * entry_size;
-	mem->va = dma_zalloc_coherent(&adapter->pdev->dev, mem->size, &mem->dma,
-				      GFP_KERNEL);
+	mem->va = dma_alloc_coherent(&adapter->pdev->dev, mem->size,
+				     &mem->dma, GFP_KERNEL);
 	if (!mem->va)
 		return -ENOMEM;
 	return 0;
@@ -269,7 +271,7 @@ void be_cq_notify(struct be_adapter *adapter, u16 qid, bool arm, u16 num_popped)
 	iowrite32(val, adapter->db + DB_CQ_OFFSET);
 }
 
-static int be_dev_mac_add(struct be_adapter *adapter, u8 *mac)
+static int be_dev_mac_add(struct be_adapter *adapter, const u8 *mac)
 {
 	int i;
 
@@ -366,7 +368,7 @@ static int be_mac_addr_set(struct net_device *netdev, void *p)
 	/* Remember currently programmed MAC */
 	ether_addr_copy(adapter->dev_mac, addr->sa_data);
 done:
-	ether_addr_copy(netdev->dev_addr, addr->sa_data);
+	eth_hw_addr_set(netdev, addr->sa_data);
 	dev_info(dev, "MAC address changed to %pM\n", addr->sa_data);
 	return 0;
 err:
@@ -605,7 +607,7 @@ static void accumulate_16bit_val(u32 *acc, u16 val)
 
 	if (wrapped)
 		newacc += 65536;
-	ACCESS_ONCE(*acc) = newacc;
+	WRITE_ONCE(*acc, newacc);
 }
 
 static void populate_erx_stats(struct be_adapter *adapter,
@@ -662,10 +664,10 @@ static void be_get_stats64(struct net_device *netdev,
 		const struct be_rx_stats *rx_stats = rx_stats(rxo);
 
 		do {
-			start = u64_stats_fetch_begin_irq(&rx_stats->sync);
+			start = u64_stats_fetch_begin(&rx_stats->sync);
 			pkts = rx_stats(rxo)->rx_pkts;
 			bytes = rx_stats(rxo)->rx_bytes;
-		} while (u64_stats_fetch_retry_irq(&rx_stats->sync, start));
+		} while (u64_stats_fetch_retry(&rx_stats->sync, start));
 		stats->rx_packets += pkts;
 		stats->rx_bytes += bytes;
 		stats->multicast += rx_stats(rxo)->rx_mcast_pkts;
@@ -677,10 +679,10 @@ static void be_get_stats64(struct net_device *netdev,
 		const struct be_tx_stats *tx_stats = tx_stats(txo);
 
 		do {
-			start = u64_stats_fetch_begin_irq(&tx_stats->sync);
+			start = u64_stats_fetch_begin(&tx_stats->sync);
 			pkts = tx_stats(txo)->tx_pkts;
 			bytes = tx_stats(txo)->tx_bytes;
-		} while (u64_stats_fetch_retry_irq(&tx_stats->sync, start));
+		} while (u64_stats_fetch_retry(&tx_stats->sync, start));
 		stats->tx_packets += pkts;
 		stats->tx_bytes += bytes;
 	}
@@ -734,9 +736,9 @@ void be_link_status_update(struct be_adapter *adapter, u8 link_status)
 static int be_gso_hdr_len(struct sk_buff *skb)
 {
 	if (skb->encapsulation)
-		return skb_inner_transport_offset(skb) +
-		       inner_tcp_hdrlen(skb);
-	return skb_transport_offset(skb) + tcp_hdrlen(skb);
+		return skb_inner_tcp_all_headers(skb);
+
+	return skb_tcp_all_headers(skb);
 }
 
 static void be_tx_stats_update(struct be_tx_obj *txo, struct sk_buff *skb)
@@ -788,7 +790,7 @@ static inline u16 be_get_tx_vlan_tag(struct be_adapter *adapter,
 	u16 vlan_tag;
 
 	vlan_tag = skb_vlan_tag_get(skb);
-	vlan_prio = (vlan_tag & VLAN_PRIO_MASK) >> VLAN_PRIO_SHIFT;
+	vlan_prio = skb_vlan_tag_get_prio(skb);
 	/* If vlan priority provided by OS is NOT in available bmap */
 	if (!(adapter->vlan_prio_bmap & (1 << vlan_prio)))
 		vlan_tag = (vlan_tag & ~VLAN_PRIO_MASK) |
@@ -991,9 +993,8 @@ static u32 be_xmit_enqueue(struct be_adapter *adapter, struct be_tx_obj *txo,
 {
 	u32 i, copied = 0, wrb_cnt = skb_wrb_cnt(skb);
 	struct device *dev = &adapter->pdev->dev;
-	struct be_queue_info *txq = &txo->q;
 	bool map_single = false;
-	u32 head = txq->head;
+	u32 head;
 	dma_addr_t busaddr;
 	int len;
 
@@ -1011,7 +1012,7 @@ static u32 be_xmit_enqueue(struct be_adapter *adapter, struct be_tx_obj *txo,
 	}
 
 	for (i = 0; i < skb_shinfo(skb)->nr_frags; i++) {
-		const struct skb_frag_struct *frag = &skb_shinfo(skb)->frags[i];
+		const skb_frag_t *frag = &skb_shinfo(skb)->frags[i];
 		len = skb_frag_size(frag);
 
 		busaddr = skb_frag_dma_map(dev, frag, 0, len, DMA_TO_DEVICE);
@@ -1042,30 +1043,35 @@ static struct sk_buff *be_insert_vlan_in_pkt(struct be_adapter *adapter,
 					     struct be_wrb_params
 					     *wrb_params)
 {
+	bool insert_vlan = false;
 	u16 vlan_tag = 0;
 
 	skb = skb_share_check(skb, GFP_ATOMIC);
 	if (unlikely(!skb))
 		return skb;
 
-	if (skb_vlan_tag_present(skb))
+	if (skb_vlan_tag_present(skb)) {
 		vlan_tag = be_get_tx_vlan_tag(adapter, skb);
+		insert_vlan = true;
+	}
 
 	if (qnq_async_evt_rcvd(adapter) && adapter->pvid) {
-		if (!vlan_tag)
+		if (!insert_vlan) {
 			vlan_tag = adapter->pvid;
+			insert_vlan = true;
+		}
 		/* f/w workaround to set skip_hw_vlan = 1, informs the F/W to
 		 * skip VLAN insertion
 		 */
 		BE_WRB_F_SET(wrb_params->features, VLAN_SKIP_HW, 1);
 	}
 
-	if (vlan_tag) {
+	if (insert_vlan) {
 		skb = vlan_insert_tag_set_proto(skb, htons(ETH_P_8021Q),
 						vlan_tag);
 		if (unlikely(!skb))
 			return skb;
-		skb->vlan_tci = 0;
+		__vlan_hwaccel_clear_tag(skb);
 	}
 
 	/* Insert the outer VLAN, if any */
@@ -1118,7 +1124,7 @@ static struct sk_buff *be_lancer_xmit_workarounds(struct be_adapter *adapter,
 						  struct be_wrb_params
 						  *wrb_params)
 {
-	struct vlan_ethhdr *veh = (struct vlan_ethhdr *)skb->data;
+	struct vlan_ethhdr *veh = skb_vlan_eth_hdr(skb);
 	unsigned int eth_hdr_len;
 	struct iphdr *ip;
 
@@ -1129,10 +1135,11 @@ static struct sk_buff *be_lancer_xmit_workarounds(struct be_adapter *adapter,
 	eth_hdr_len = ntohs(skb->protocol) == ETH_P_8021Q ?
 						VLAN_ETH_HLEN : ETH_HLEN;
 	if (skb->len <= 60 &&
-	    (lancer_chip(adapter) || skb_vlan_tag_present(skb)) &&
-	    is_ipv4_pkt(skb)) {
+	    (lancer_chip(adapter) || BE3_chip(adapter) ||
+	     skb_vlan_tag_present(skb)) && is_ipv4_pkt(skb)) {
 		ip = (struct iphdr *)ip_hdr(skb);
-		pskb_trim(skb, eth_hdr_len + ntohs(ip->tot_len));
+		if (unlikely(pskb_trim(skb, eth_hdr_len + ntohs(ip->tot_len))))
+			goto tx_drop;
 	}
 
 	/* If vlan tag is already inlined in the packet, skip HW VLAN
@@ -1258,10 +1265,6 @@ static void be_xmit_flush(struct be_adapter *adapter, struct be_tx_obj *txo)
 #define is_arp_allowed_on_bmc(adapter, skb)	\
 	(is_arp(skb) && is_arp_filt_enabled(adapter))
 
-#define is_broadcast_packet(eh, adapter)	\
-		(is_multicast_ether_addr(eh->h_dest) && \
-		!compare_ether_addr(eh->h_dest, adapter->netdev->broadcast))
-
 #define is_arp(skb)	(skb->protocol == htons(ETH_P_ARP))
 
 #define is_arp_filt_enabled(adapter)	\
@@ -1368,7 +1371,7 @@ static netdev_tx_t be_xmit(struct sk_buff *skb, struct net_device *netdev)
 	u16 q_idx = skb_get_queue_mapping(skb);
 	struct be_tx_obj *txo = &adapter->tx_obj[q_idx];
 	struct be_wrb_params wrb_params = { 0 };
-	bool flush = !skb->xmit_more;
+	bool flush = !netdev_xmit_more();
 	u16 wrb_cnt;
 
 	skb = be_xmit_workarounds(adapter, skb, &wrb_params);
@@ -1378,10 +1381,8 @@ static netdev_tx_t be_xmit(struct sk_buff *skb, struct net_device *netdev)
 	be_get_wrb_params_from_skb(adapter, skb, &wrb_params);
 
 	wrb_cnt = be_xmit_enqueue(adapter, txo, skb, &wrb_params);
-	if (unlikely(!wrb_cnt)) {
-		dev_kfree_skb_any(skb);
-		goto drop;
-	}
+	if (unlikely(!wrb_cnt))
+		goto drop_skb;
 
 	/* if os2bmc is enabled and if the pkt is destined to bmc,
 	 * enqueue the pkt a 2nd time with mgmt bit set.
@@ -1390,7 +1391,7 @@ static netdev_tx_t be_xmit(struct sk_buff *skb, struct net_device *netdev)
 		BE_WRB_F_SET(wrb_params.features, OS2BMC, 1);
 		wrb_cnt = be_xmit_enqueue(adapter, txo, skb, &wrb_params);
 		if (unlikely(!wrb_cnt))
-			goto drop;
+			goto drop_skb;
 		else
 			skb_get(skb);
 	}
@@ -1404,6 +1405,8 @@ static netdev_tx_t be_xmit(struct sk_buff *skb, struct net_device *netdev)
 		be_xmit_flush(adapter, txo);
 
 	return NETDEV_TX_OK;
+drop_skb:
+	dev_kfree_skb_any(skb);
 drop:
 	tx_stats(txo)->tx_drv_drops++;
 	/* Flush the already enqueued tx requests */
@@ -1413,6 +1416,83 @@ drop:
 	return NETDEV_TX_OK;
 }
 
+static void be_tx_timeout(struct net_device *netdev, unsigned int txqueue)
+{
+	struct be_adapter *adapter = netdev_priv(netdev);
+	struct device *dev = &adapter->pdev->dev;
+	struct be_tx_obj *txo;
+	struct sk_buff *skb;
+	struct tcphdr *tcphdr;
+	struct udphdr *udphdr;
+	u32 *entry;
+	int status;
+	int i, j;
+
+	for_all_tx_queues(adapter, txo, i) {
+		dev_info(dev, "TXQ Dump: %d H: %d T: %d used: %d, qid: 0x%x\n",
+			 i, txo->q.head, txo->q.tail,
+			 atomic_read(&txo->q.used), txo->q.id);
+
+		entry = txo->q.dma_mem.va;
+		for (j = 0; j < TX_Q_LEN * 4; j += 4) {
+			if (entry[j] != 0 || entry[j + 1] != 0 ||
+			    entry[j + 2] != 0 || entry[j + 3] != 0) {
+				dev_info(dev, "Entry %d 0x%x 0x%x 0x%x 0x%x\n",
+					 j, entry[j], entry[j + 1],
+					 entry[j + 2], entry[j + 3]);
+			}
+		}
+
+		entry = txo->cq.dma_mem.va;
+		dev_info(dev, "TXCQ Dump: %d  H: %d T: %d used: %d\n",
+			 i, txo->cq.head, txo->cq.tail,
+			 atomic_read(&txo->cq.used));
+		for (j = 0; j < TX_CQ_LEN * 4; j += 4) {
+			if (entry[j] != 0 || entry[j + 1] != 0 ||
+			    entry[j + 2] != 0 || entry[j + 3] != 0) {
+				dev_info(dev, "Entry %d 0x%x 0x%x 0x%x 0x%x\n",
+					 j, entry[j], entry[j + 1],
+					 entry[j + 2], entry[j + 3]);
+			}
+		}
+
+		for (j = 0; j < TX_Q_LEN; j++) {
+			if (txo->sent_skb_list[j]) {
+				skb = txo->sent_skb_list[j];
+				if (ip_hdr(skb)->protocol == IPPROTO_TCP) {
+					tcphdr = tcp_hdr(skb);
+					dev_info(dev, "TCP source port %d\n",
+						 ntohs(tcphdr->source));
+					dev_info(dev, "TCP dest port %d\n",
+						 ntohs(tcphdr->dest));
+					dev_info(dev, "TCP sequence num %u\n",
+						 ntohl(tcphdr->seq));
+					dev_info(dev, "TCP ack_seq %u\n",
+						 ntohl(tcphdr->ack_seq));
+				} else if (ip_hdr(skb)->protocol ==
+					   IPPROTO_UDP) {
+					udphdr = udp_hdr(skb);
+					dev_info(dev, "UDP source port %d\n",
+						 ntohs(udphdr->source));
+					dev_info(dev, "UDP dest port %d\n",
+						 ntohs(udphdr->dest));
+				}
+				dev_info(dev, "skb[%d] %p len %d proto 0x%x\n",
+					 j, skb, skb->len, skb->protocol);
+			}
+		}
+	}
+
+	if (lancer_chip(adapter)) {
+		dev_info(dev, "Initiating reset due to tx timeout\n");
+		dev_info(dev, "Resetting adapter\n");
+		status = lancer_physdev_ctrl(adapter,
+					     PHYSDEV_CONTROL_FW_RESET_MASK);
+		if (status)
+			dev_err(dev, "Reset failed .. Reboot server\n");
+	}
+}
+
 static inline bool be_in_all_promisc(struct be_adapter *adapter)
 {
 	return (adapter->if_flags & BE_IF_FLAGS_ALL_PROMISCUOUS) ==
@@ -2066,7 +2146,7 @@ static int be_get_new_eqd(struct be_eq_obj *eqo)
 	int i;
 
 	aic = &adapter->aic_obj[eqo->idx];
-	if (!aic->enable) {
+	if (!adapter->aic_enabled) {
 		if (aic->jiffies)
 			aic->jiffies = 0;
 		eqd = aic->et_eqd;
@@ -2075,16 +2155,16 @@ static int be_get_new_eqd(struct be_eq_obj *eqo)
 
 	for_all_rx_queues_on_eq(adapter, eqo, rxo, i) {
 		do {
-			start = u64_stats_fetch_begin_irq(&rxo->stats.sync);
+			start = u64_stats_fetch_begin(&rxo->stats.sync);
 			rx_pkts += rxo->stats.rx_pkts;
-		} while (u64_stats_fetch_retry_irq(&rxo->stats.sync, start));
+		} while (u64_stats_fetch_retry(&rxo->stats.sync, start));
 	}
 
 	for_all_tx_queues_on_eq(adapter, eqo, txo, i) {
 		do {
-			start = u64_stats_fetch_begin_irq(&txo->stats.sync);
+			start = u64_stats_fetch_begin(&txo->stats.sync);
 			tx_pkts += txo->stats.tx_reqs;
-		} while (u64_stats_fetch_retry_irq(&txo->stats.sync, start));
+		} while (u64_stats_fetch_retry(&txo->stats.sync, start));
 	}
 
 	/* Skip, if wrapped around or first calculation */
@@ -2123,7 +2203,7 @@ static u32 be_get_eq_delay_mult_enc(struct be_eq_obj *eqo)
 	int eqd;
 	u32 mult_enc;
 
-	if (!aic->enable)
+	if (!adapter->aic_enabled)
 		return 0;
 
 	if (jiffies_to_msecs(now - aic->jiffies) < 1)
@@ -2264,11 +2344,10 @@ static void skb_fill_rx_data(struct be_rx_obj *rxo, struct sk_buff *skb,
 		hdr_len = ETH_HLEN;
 		memcpy(skb->data, start, hdr_len);
 		skb_shinfo(skb)->nr_frags = 1;
-		skb_frag_set_page(skb, 0, page_info->page);
-		skb_shinfo(skb)->frags[0].page_offset =
-					page_info->page_offset + hdr_len;
-		skb_frag_size_set(&skb_shinfo(skb)->frags[0],
-				  curr_frag_len - hdr_len);
+		skb_frag_fill_page_desc(&skb_shinfo(skb)->frags[0],
+					page_info->page,
+					page_info->page_offset + hdr_len,
+					curr_frag_len - hdr_len);
 		skb->data_len = curr_frag_len - hdr_len;
 		skb->truesize += rx_frag_size;
 		skb->tail += hdr_len;
@@ -2290,16 +2369,17 @@ static void skb_fill_rx_data(struct be_rx_obj *rxo, struct sk_buff *skb,
 		if (page_info->page_offset == 0) {
 			/* Fresh page */
 			j++;
-			skb_frag_set_page(skb, j, page_info->page);
-			skb_shinfo(skb)->frags[j].page_offset =
-							page_info->page_offset;
-			skb_frag_size_set(&skb_shinfo(skb)->frags[j], 0);
+			skb_frag_fill_page_desc(&skb_shinfo(skb)->frags[j],
+						page_info->page,
+						page_info->page_offset,
+						curr_frag_len);
 			skb_shinfo(skb)->nr_frags++;
 		} else {
 			put_page(page_info->page);
+			skb_frag_size_add(&skb_shinfo(skb)->frags[j],
+					  curr_frag_len);
 		}
 
-		skb_frag_size_add(&skb_shinfo(skb)->frags[j], curr_frag_len);
 		skb->len += curr_frag_len;
 		skb->data_len += curr_frag_len;
 		skb->truesize += rx_frag_size;
@@ -2372,14 +2452,16 @@ static void be_rx_compl_process_gro(struct be_rx_obj *rxo,
 		if (i == 0 || page_info->page_offset == 0) {
 			/* First frag or Fresh page */
 			j++;
-			skb_frag_set_page(skb, j, page_info->page);
-			skb_shinfo(skb)->frags[j].page_offset =
-							page_info->page_offset;
-			skb_frag_size_set(&skb_shinfo(skb)->frags[j], 0);
+			skb_frag_fill_page_desc(&skb_shinfo(skb)->frags[j],
+						page_info->page,
+						page_info->page_offset,
+						curr_frag_len);
 		} else {
 			put_page(page_info->page);
+			skb_frag_size_add(&skb_shinfo(skb)->frags[j],
+					  curr_frag_len);
 		}
-		skb_frag_size_add(&skb_shinfo(skb)->frags[j], curr_frag_len);
+
 		skb->truesize += rx_frag_size;
 		remaining -= curr_frag_len;
 		memset(page_info, 0, sizeof(*page_info));
@@ -2584,7 +2666,48 @@ static void be_post_rx_frags(struct be_rx_obj *rxo, gfp_t gfp, u32 frags_needed)
 	}
 }
 
-static struct be_tx_compl_info *be_tx_compl_get(struct be_tx_obj *txo)
+static inline void be_update_tx_err(struct be_tx_obj *txo, u8 status)
+{
+	switch (status) {
+	case BE_TX_COMP_HDR_PARSE_ERR:
+		tx_stats(txo)->tx_hdr_parse_err++;
+		break;
+	case BE_TX_COMP_NDMA_ERR:
+		tx_stats(txo)->tx_dma_err++;
+		break;
+	case BE_TX_COMP_ACL_ERR:
+		tx_stats(txo)->tx_spoof_check_err++;
+		break;
+	}
+}
+
+static inline void lancer_update_tx_err(struct be_tx_obj *txo, u8 status)
+{
+	switch (status) {
+	case LANCER_TX_COMP_LSO_ERR:
+		tx_stats(txo)->tx_tso_err++;
+		break;
+	case LANCER_TX_COMP_HSW_DROP_MAC_ERR:
+	case LANCER_TX_COMP_HSW_DROP_VLAN_ERR:
+		tx_stats(txo)->tx_spoof_check_err++;
+		break;
+	case LANCER_TX_COMP_QINQ_ERR:
+		tx_stats(txo)->tx_qinq_err++;
+		break;
+	case LANCER_TX_COMP_PARITY_ERR:
+		tx_stats(txo)->tx_internal_parity_err++;
+		break;
+	case LANCER_TX_COMP_DMA_ERR:
+		tx_stats(txo)->tx_dma_err++;
+		break;
+	case LANCER_TX_COMP_SGE_ERR:
+		tx_stats(txo)->tx_sge_err++;
+		break;
+	}
+}
+
+static struct be_tx_compl_info *be_tx_compl_get(struct be_adapter *adapter,
+						struct be_tx_obj *txo)
 {
 	struct be_queue_info *tx_cq = &txo->cq;
 	struct be_tx_compl_info *txcp = &txo->txcp;
@@ -2600,6 +2723,24 @@ static struct be_tx_compl_info *be_tx_compl_get(struct be_tx_obj *txo)
 	txcp->status = GET_TX_COMPL_BITS(status, compl);
 	txcp->end_index = GET_TX_COMPL_BITS(wrb_index, compl);
 
+	if (txcp->status) {
+		if (lancer_chip(adapter)) {
+			lancer_update_tx_err(txo, txcp->status);
+			/* Reset the adapter incase of TSO,
+			 * SGE or Parity error
+			 */
+			if (txcp->status == LANCER_TX_COMP_LSO_ERR ||
+			    txcp->status == LANCER_TX_COMP_PARITY_ERR ||
+			    txcp->status == LANCER_TX_COMP_SGE_ERR)
+				be_set_error(adapter, BE_ERROR_TX);
+		} else {
+			be_update_tx_err(txo, txcp->status);
+		}
+	}
+
+	if (be_check_error(adapter, BE_ERROR_TX))
+		return NULL;
+
 	compl->dw[offsetof(struct amap_eth_tx_compl, valid) / 32] = 0;
 	queue_tail_inc(tx_cq);
 	return txcp;
@@ -2742,7 +2883,7 @@ static void be_tx_compl_clean(struct be_adapter *adapter)
 			cmpl = 0;
 			num_wrbs = 0;
 			txq = &txo->q;
-			while ((txcp = be_tx_compl_get(txo))) {
+			while ((txcp = be_tx_compl_get(adapter, txo))) {
 				num_wrbs +=
 					be_tx_compl_process(adapter, txo,
 							    txcp->end_index);
@@ -2819,6 +2960,8 @@ static int be_evt_queues_create(struct be_adapter *adapter)
 				    max(adapter->cfg_num_rx_irqs,
 					adapter->cfg_num_tx_irqs));
 
+	adapter->aic_enabled = true;
+
 	for_all_evt_queues(adapter, eqo, i) {
 		int numa_node = dev_to_node(&adapter->pdev->dev);
 
@@ -2826,7 +2969,6 @@ static int be_evt_queues_create(struct be_adapter *adapter)
 		eqo->adapter = adapter;
 		eqo->idx = i;
 		aic->max_eqd = BE_MAX_EQD;
-		aic->enable = true;
 
 		eq = &eqo->q;
 		rc = be_queue_alloc(adapter, eq, EVNT_Q_LEN,
@@ -2842,8 +2984,7 @@ static int be_evt_queues_create(struct be_adapter *adapter)
 			return -ENOMEM;
 		cpumask_set_cpu(cpumask_local_spread(i, numa_node),
 				eqo->affinity_mask);
-		netif_napi_add(adapter->netdev, &eqo->napi, be_poll,
-			       BE_NAPI_WEIGHT);
+		netif_napi_add(adapter->netdev, &eqo->napi, be_poll);
 	}
 	return 0;
 }
@@ -3038,7 +3179,7 @@ static irqreturn_t be_intx(int irq, void *dev)
 	}
 	be_eq_notify(adapter, eqo->q.id, false, true, num_evts, 0);
 
-	/* Return IRQ_HANDLED only for the the first spurious intr
+	/* Return IRQ_HANDLED only for the first spurious intr
 	 * after a valid intr to stop the kernel from branding
 	 * this irq as a bad one!
 	 */
@@ -3121,42 +3262,6 @@ loop_continue:
 	return work_done;
 }
 
-static inline void be_update_tx_err(struct be_tx_obj *txo, u8 status)
-{
-	switch (status) {
-	case BE_TX_COMP_HDR_PARSE_ERR:
-		tx_stats(txo)->tx_hdr_parse_err++;
-		break;
-	case BE_TX_COMP_NDMA_ERR:
-		tx_stats(txo)->tx_dma_err++;
-		break;
-	case BE_TX_COMP_ACL_ERR:
-		tx_stats(txo)->tx_spoof_check_err++;
-		break;
-	}
-}
-
-static inline void lancer_update_tx_err(struct be_tx_obj *txo, u8 status)
-{
-	switch (status) {
-	case LANCER_TX_COMP_LSO_ERR:
-		tx_stats(txo)->tx_tso_err++;
-		break;
-	case LANCER_TX_COMP_HSW_DROP_MAC_ERR:
-	case LANCER_TX_COMP_HSW_DROP_VLAN_ERR:
-		tx_stats(txo)->tx_spoof_check_err++;
-		break;
-	case LANCER_TX_COMP_QINQ_ERR:
-		tx_stats(txo)->tx_qinq_err++;
-		break;
-	case LANCER_TX_COMP_PARITY_ERR:
-		tx_stats(txo)->tx_internal_parity_err++;
-		break;
-	case LANCER_TX_COMP_DMA_ERR:
-		tx_stats(txo)->tx_dma_err++;
-		break;
-	}
-}
 
 static void be_process_tx(struct be_adapter *adapter, struct be_tx_obj *txo,
 			  int idx)
@@ -3164,16 +3269,9 @@ static void be_process_tx(struct be_adapter *adapter, struct be_tx_obj *txo,
 	int num_wrbs = 0, work_done = 0;
 	struct be_tx_compl_info *txcp;
 
-	while ((txcp = be_tx_compl_get(txo))) {
+	while ((txcp = be_tx_compl_get(adapter, txo))) {
 		num_wrbs += be_tx_compl_process(adapter, txo, txcp->end_index);
 		work_done++;
-
-		if (txcp->status) {
-			if (lancer_chip(adapter))
-				lancer_update_tx_err(txo, txcp->status);
-			else
-				be_update_tx_err(txo, txcp->status);
-		}
 	}
 
 	if (work_done) {
@@ -3259,7 +3357,7 @@ void be_detect_error(struct be_adapter *adapter)
 			/* Do not log error messages if its a FW reset */
 			if (sliport_err1 == SLIPORT_ERROR_FW_RESET1 &&
 			    sliport_err2 == SLIPORT_ERROR_FW_RESET2) {
-				dev_info(dev, "Firmware update in progress\n");
+				dev_info(dev, "Reset is in progress\n");
 			} else {
 				dev_err(dev, "Error detected in the card\n");
 				dev_err(dev, "ERR: sliport status 0x%x\n",
@@ -3294,7 +3392,9 @@ void be_detect_error(struct be_adapter *adapter)
 				if ((val & POST_STAGE_FAT_LOG_START)
 				     != POST_STAGE_FAT_LOG_START &&
 				    (val & POST_STAGE_ARMFW_UE)
-				     != POST_STAGE_ARMFW_UE)
+				     != POST_STAGE_ARMFW_UE &&
+				    (val & POST_STAGE_RECOVERABLE_ERR)
+				     != POST_STAGE_RECOVERABLE_ERR)
 					return;
 			}
 
@@ -3392,7 +3492,7 @@ static int be_msix_register(struct be_adapter *adapter)
 		if (status)
 			goto err_msix;
 
-		irq_set_affinity_hint(vec, eqo->affinity_mask);
+		irq_update_affinity_hint(vec, eqo->affinity_mask);
 	}
 
 	return 0;
@@ -3453,7 +3553,7 @@ static void be_irq_unregister(struct be_adapter *adapter)
 	/* MSIx */
 	for_all_evt_queues(adapter, eqo, i) {
 		vec = be_msix_vec_get(adapter, eqo);
-		irq_set_affinity_hint(vec, NULL);
+		irq_update_affinity_hint(vec, NULL);
 		free_irq(vec, eqo);
 	}
 
@@ -3730,8 +3830,8 @@ static int be_open(struct net_device *netdev)
 		be_link_status_update(adapter, link_status);
 
 	netif_tx_start_all_queues(netdev);
-	if (skyhawk_chip(adapter))
-		udp_tunnel_get_rx_info(netdev);
+
+	udp_tunnel_nic_reset_ntf(netdev);
 
 	return 0;
 err:
@@ -3868,18 +3968,23 @@ static void be_cancel_err_detection(struct be_adapter *adapter)
 	}
 }
 
-static int be_enable_vxlan_offloads(struct be_adapter *adapter)
+/* VxLAN offload Notes:
+ *
+ * The stack defines tunnel offload flags (hw_enc_features) for IP and doesn't
+ * distinguish various types of transports (VxLAN, GRE, NVGRE ..). So, offload
+ * is expected to work across all types of IP tunnels once exported. Skyhawk
+ * supports offloads for either VxLAN or NVGRE, exclusively. So we export VxLAN
+ * offloads in hw_enc_features only when a VxLAN port is added. If other (non
+ * VxLAN) tunnels are configured while VxLAN offloads are enabled, offloads for
+ * those other tunnels are unexported on the fly through ndo_features_check().
+ */
+static int be_vxlan_set_port(struct net_device *netdev, unsigned int table,
+			     unsigned int entry, struct udp_tunnel_info *ti)
 {
-	struct net_device *netdev = adapter->netdev;
+	struct be_adapter *adapter = netdev_priv(netdev);
 	struct device *dev = &adapter->pdev->dev;
-	struct be_vxlan_port *vxlan_port;
-	__be16 port;
 	int status;
 
-	vxlan_port = list_first_entry(&adapter->vxlan_port_list,
-				      struct be_vxlan_port, list);
-	port = vxlan_port->port;
-
 	status = be_cmd_manage_iface(adapter, adapter->if_handle,
 				     OP_CONVERT_NORMAL_TO_TUNNEL);
 	if (status) {
@@ -3888,27 +3993,26 @@ static int be_enable_vxlan_offloads(struct be_adapter *adapter)
 	}
 	adapter->flags |= BE_FLAGS_VXLAN_OFFLOADS;
 
-	status = be_cmd_set_vxlan_port(adapter, port);
+	status = be_cmd_set_vxlan_port(adapter, ti->port);
 	if (status) {
 		dev_warn(dev, "Failed to add VxLAN port\n");
 		return status;
 	}
-	adapter->vxlan_port = port;
+	adapter->vxlan_port = ti->port;
 
 	netdev->hw_enc_features |= NETIF_F_IP_CSUM | NETIF_F_IPV6_CSUM |
 				   NETIF_F_TSO | NETIF_F_TSO6 |
 				   NETIF_F_GSO_UDP_TUNNEL;
-	netdev->hw_features |= NETIF_F_GSO_UDP_TUNNEL;
-	netdev->features |= NETIF_F_GSO_UDP_TUNNEL;
 
 	dev_info(dev, "Enabled VxLAN offloads for UDP port %d\n",
-		 be16_to_cpu(port));
+		 be16_to_cpu(ti->port));
 	return 0;
 }
 
-static void be_disable_vxlan_offloads(struct be_adapter *adapter)
+static int be_vxlan_unset_port(struct net_device *netdev, unsigned int table,
+			       unsigned int entry, struct udp_tunnel_info *ti)
 {
-	struct net_device *netdev = adapter->netdev;
+	struct be_adapter *adapter = netdev_priv(netdev);
 
 	if (adapter->flags & BE_FLAGS_VXLAN_OFFLOADS)
 		be_cmd_manage_iface(adapter, adapter->if_handle,
@@ -3921,10 +4025,18 @@ static void be_disable_vxlan_offloads(struct be_adapter *adapter)
 	adapter->vxlan_port = 0;
 
 	netdev->hw_enc_features = 0;
-	netdev->hw_features &= ~(NETIF_F_GSO_UDP_TUNNEL);
-	netdev->features &= ~(NETIF_F_GSO_UDP_TUNNEL);
+	return 0;
 }
 
+static const struct udp_tunnel_nic_info be_udp_tunnels = {
+	.set_port	= be_vxlan_set_port,
+	.unset_port	= be_vxlan_unset_port,
+	.flags		= UDP_TUNNEL_NIC_INFO_OPEN_ONLY,
+	.tables		= {
+		{ .n_entries = 1, .tunnel_types = UDP_TUNNEL_TYPE_VXLAN, },
+	},
+};
+
 static void be_calculate_vf_res(struct be_adapter *adapter, u16 num_vfs,
 				struct be_resources *vft_res)
 {
@@ -4040,7 +4152,7 @@ static int be_clear(struct be_adapter *adapter)
 					&vft_res);
 	}
 
-	be_disable_vxlan_offloads(adapter);
+	be_vxlan_unset_port(adapter->netdev, 0, 0, NULL);
 
 	be_if_destroy(adapter);
 
@@ -4487,7 +4599,7 @@ static int be_mac_setup(struct be_adapter *adapter)
 		if (status)
 			return status;
 
-		memcpy(adapter->netdev->dev_addr, mac, ETH_ALEN);
+		eth_hw_addr_set(adapter->netdev, mac);
 		memcpy(adapter->netdev->perm_addr, mac, ETH_ALEN);
 
 		/* Initial MAC for BE3 VFs is already programmed by PF */
@@ -4509,7 +4621,6 @@ static void be_destroy_err_recovery_workq(void)
 	if (!be_err_recovery_workq)
 		return;
 
-	flush_workqueue(be_err_recovery_workq);
 	destroy_workqueue(be_err_recovery_workq);
 	be_err_recovery_workq = NULL;
 }
@@ -4565,7 +4676,6 @@ static int be_if_create(struct be_adapter *adapter)
 {
 	u32 en_flags = BE_IF_FLAGS_RSS | BE_IF_FLAGS_DEFQ_RSS;
 	u32 cap_flags = be_if_cap_flags(adapter);
-	int status;
 
 	/* alloc required memory for other filtering fields */
 	adapter->pmac_id = kcalloc(be_max_uc(adapter),
@@ -4588,13 +4698,8 @@ static int be_if_create(struct be_adapter *adapter)
 
 	en_flags &= cap_flags;
 	/* will enable all the needed filter flags in be_open() */
-	status = be_cmd_if_create(adapter, be_if_cap_flags(adapter), en_flags,
+	return be_cmd_if_create(adapter, be_if_cap_flags(adapter), en_flags,
 				  &adapter->if_handle, 0);
-
-	if (status)
-		return status;
-
-	return 0;
 }
 
 int be_update_queues(struct be_adapter *adapter)
@@ -4602,8 +4707,17 @@ int be_update_queues(struct be_adapter *adapter)
 	struct net_device *netdev = adapter->netdev;
 	int status;
 
-	if (netif_running(netdev))
+	if (netif_running(netdev)) {
+		/* be_tx_timeout() must not run concurrently with this
+		 * function, synchronize with an already-running dev_watchdog
+		 */
+		netif_tx_lock_bh(netdev);
+		/* device cannot transmit now, avoid dev_watchdog timeouts */
+		netif_carrier_off(netdev);
+		netif_tx_unlock_bh(netdev);
+
 		be_close(netdev);
+	}
 
 	be_cancel_worker(adapter);
 
@@ -4634,6 +4748,14 @@ int be_update_queues(struct be_adapter *adapter)
 
 	be_schedule_worker(adapter);
 
+	/* The IF was destroyed and re-created. We need to clear
+	 * all promiscuous flags valid for the destroyed IF.
+	 * Without this promisc mode is not restored during
+	 * be_open() because the driver thinks that it is
+	 * already enabled in HW.
+	 */
+	adapter->if_flags &= ~BE_IF_FLAGS_ALL_PROMISCUOUS;
+
 	if (netif_running(netdev))
 		status = be_open(netdev);
 
@@ -4844,7 +4966,7 @@ fw_exit:
 }
 
 static int be_ndo_bridge_setlink(struct net_device *dev, struct nlmsghdr *nlh,
-				 u16 flags)
+				 u16 flags, struct netlink_ext_ack *extack)
 {
 	struct be_adapter *adapter = netdev_priv(dev);
 	struct nlattr *attr, *br_spec;
@@ -4859,13 +4981,7 @@ static int be_ndo_bridge_setlink(struct net_device *dev, struct nlmsghdr *nlh,
 	if (!br_spec)
 		return -EINVAL;
 
-	nla_for_each_nested(attr, br_spec, rem) {
-		if (nla_type(attr) != IFLA_BRIDGE_MODE)
-			continue;
-
-		if (nla_len(attr) < sizeof(mode))
-			return -EINVAL;
-
+	nla_for_each_nested_type(attr, IFLA_BRIDGE_MODE, br_spec, rem) {
 		mode = nla_get_u16(attr);
 		if (BE3_chip(adapter) && mode == BRIDGE_MODE_VEPA)
 			return -EOPNOTSUPP;
@@ -4941,147 +5057,6 @@ static struct be_cmd_work *be_alloc_work(struct be_adapter *adapter,
 	return work;
 }
 
-/* VxLAN offload Notes:
- *
- * The stack defines tunnel offload flags (hw_enc_features) for IP and doesn't
- * distinguish various types of transports (VxLAN, GRE, NVGRE ..). So, offload
- * is expected to work across all types of IP tunnels once exported. Skyhawk
- * supports offloads for either VxLAN or NVGRE, exclusively. So we export VxLAN
- * offloads in hw_enc_features only when a VxLAN port is added. If other (non
- * VxLAN) tunnels are configured while VxLAN offloads are enabled, offloads for
- * those other tunnels are unexported on the fly through ndo_features_check().
- *
- * Skyhawk supports VxLAN offloads only for one UDP dport. So, if the stack
- * adds more than one port, disable offloads and re-enable them again when
- * there's only one port left. We maintain a list of ports for this purpose.
- */
-static void be_work_add_vxlan_port(struct work_struct *work)
-{
-	struct be_cmd_work *cmd_work =
-				container_of(work, struct be_cmd_work, work);
-	struct be_adapter *adapter = cmd_work->adapter;
-	struct device *dev = &adapter->pdev->dev;
-	__be16 port = cmd_work->info.vxlan_port;
-	struct be_vxlan_port *vxlan_port;
-	int status;
-
-	/* Bump up the alias count if it is an existing port */
-	list_for_each_entry(vxlan_port, &adapter->vxlan_port_list, list) {
-		if (vxlan_port->port == port) {
-			vxlan_port->port_aliases++;
-			goto done;
-		}
-	}
-
-	/* Add a new port to our list. We don't need a lock here since port
-	 * add/delete are done only in the context of a single-threaded work
-	 * queue (be_wq).
-	 */
-	vxlan_port = kzalloc(sizeof(*vxlan_port), GFP_KERNEL);
-	if (!vxlan_port)
-		goto done;
-
-	vxlan_port->port = port;
-	INIT_LIST_HEAD(&vxlan_port->list);
-	list_add_tail(&vxlan_port->list, &adapter->vxlan_port_list);
-	adapter->vxlan_port_count++;
-
-	if (adapter->flags & BE_FLAGS_VXLAN_OFFLOADS) {
-		dev_info(dev,
-			 "Only one UDP port supported for VxLAN offloads\n");
-		dev_info(dev, "Disabling VxLAN offloads\n");
-		goto err;
-	}
-
-	if (adapter->vxlan_port_count > 1)
-		goto done;
-
-	status = be_enable_vxlan_offloads(adapter);
-	if (!status)
-		goto done;
-
-err:
-	be_disable_vxlan_offloads(adapter);
-done:
-	kfree(cmd_work);
-	return;
-}
-
-static void be_work_del_vxlan_port(struct work_struct *work)
-{
-	struct be_cmd_work *cmd_work =
-				container_of(work, struct be_cmd_work, work);
-	struct be_adapter *adapter = cmd_work->adapter;
-	__be16 port = cmd_work->info.vxlan_port;
-	struct be_vxlan_port *vxlan_port;
-
-	/* Nothing to be done if a port alias is being deleted */
-	list_for_each_entry(vxlan_port, &adapter->vxlan_port_list, list) {
-		if (vxlan_port->port == port) {
-			if (vxlan_port->port_aliases) {
-				vxlan_port->port_aliases--;
-				goto done;
-			}
-			break;
-		}
-	}
-
-	/* No port aliases left; delete the port from the list */
-	list_del(&vxlan_port->list);
-	adapter->vxlan_port_count--;
-
-	/* Disable VxLAN offload if this is the offloaded port */
-	if (adapter->vxlan_port == vxlan_port->port) {
-		WARN_ON(adapter->vxlan_port_count);
-		be_disable_vxlan_offloads(adapter);
-		dev_info(&adapter->pdev->dev,
-			 "Disabled VxLAN offloads for UDP port %d\n",
-			 be16_to_cpu(port));
-		goto out;
-	}
-
-	/* If only 1 port is left, re-enable VxLAN offload */
-	if (adapter->vxlan_port_count == 1)
-		be_enable_vxlan_offloads(adapter);
-
-out:
-	kfree(vxlan_port);
-done:
-	kfree(cmd_work);
-}
-
-static void be_cfg_vxlan_port(struct net_device *netdev,
-			      struct udp_tunnel_info *ti,
-			      void (*func)(struct work_struct *))
-{
-	struct be_adapter *adapter = netdev_priv(netdev);
-	struct be_cmd_work *cmd_work;
-
-	if (ti->type != UDP_TUNNEL_TYPE_VXLAN)
-		return;
-
-	if (lancer_chip(adapter) || BEx_chip(adapter) || be_is_mc(adapter))
-		return;
-
-	cmd_work = be_alloc_work(adapter, func);
-	if (cmd_work) {
-		cmd_work->info.vxlan_port = ti->port;
-		queue_work(be_wq, &cmd_work->work);
-	}
-}
-
-static void be_del_vxlan_port(struct net_device *netdev,
-			      struct udp_tunnel_info *ti)
-{
-	be_cfg_vxlan_port(netdev, ti, be_work_del_vxlan_port);
-}
-
-static void be_add_vxlan_port(struct net_device *netdev,
-			      struct udp_tunnel_info *ti)
-{
-	be_cfg_vxlan_port(netdev, ti, be_work_add_vxlan_port);
-}
-
 static netdev_features_t be_features_check(struct sk_buff *skb,
 					   struct net_device *dev,
 					   netdev_features_t features)
@@ -5089,6 +5064,23 @@ static netdev_features_t be_features_check(struct sk_buff *skb,
 	struct be_adapter *adapter = netdev_priv(dev);
 	u8 l4_hdr = 0;
 
+	if (skb_is_gso(skb)) {
+		/* IPv6 TSO requests with extension hdrs are a problem
+		 * to Lancer and BE3 HW. Disable TSO6 feature.
+		 */
+		if (!skyhawk_chip(adapter) && is_ipv6_ext_hdr(skb))
+			features &= ~NETIF_F_TSO6;
+
+		/* Lancer cannot handle the packet with MSS less than 256.
+		 * Also it can't handle a TSO packet with a single segment
+		 * Disable the GSO support in such cases
+		 */
+		if (lancer_chip(adapter) &&
+		    (skb_shinfo(skb)->gso_size < 256 ||
+		     skb_shinfo(skb)->gso_segs == 1))
+			features &= ~NETIF_F_GSO_MASK;
+	}
+
 	/* The code below restricts offload features for some tunneled and
 	 * Q-in-Q packets.
 	 * Offload features for normal (non tunnel) packets are unchanged.
@@ -5174,13 +5166,12 @@ static const struct net_device_ops be_netdev_ops = {
 	.ndo_get_vf_config	= be_get_vf_config,
 	.ndo_set_vf_link_state  = be_set_vf_link_state,
 	.ndo_set_vf_spoofchk    = be_set_vf_spoofchk,
+	.ndo_tx_timeout		= be_tx_timeout,
 #ifdef CONFIG_NET_POLL_CONTROLLER
 	.ndo_poll_controller	= be_netpoll,
 #endif
 	.ndo_bridge_setlink	= be_ndo_bridge_setlink,
 	.ndo_bridge_getlink	= be_ndo_bridge_getlink,
-	.ndo_udp_tunnel_add	= be_add_vxlan_port,
-	.ndo_udp_tunnel_del	= be_del_vxlan_port,
 	.ndo_features_check	= be_features_check,
 	.ndo_get_phys_port_id   = be_get_phys_port_id,
 };
@@ -5190,13 +5181,15 @@ static void be_netdev_init(struct net_device *netdev)
 	struct be_adapter *adapter = netdev_priv(netdev);
 
 	netdev->hw_features |= NETIF_F_SG | NETIF_F_TSO | NETIF_F_TSO6 |
+		NETIF_F_GSO_UDP_TUNNEL |
 		NETIF_F_IP_CSUM | NETIF_F_IPV6_CSUM | NETIF_F_RXCSUM |
 		NETIF_F_HW_VLAN_CTAG_TX;
 	if ((be_if_cap_flags(adapter) & BE_IF_FLAGS_RSS))
 		netdev->hw_features |= NETIF_F_RXHASH;
 
 	netdev->features |= netdev->hw_features |
-		NETIF_F_HW_VLAN_CTAG_RX | NETIF_F_HW_VLAN_CTAG_FILTER;
+		NETIF_F_HW_VLAN_CTAG_RX | NETIF_F_HW_VLAN_CTAG_FILTER |
+		NETIF_F_HIGHDMA;
 
 	netdev->vlan_features |= NETIF_F_SG | NETIF_F_TSO | NETIF_F_TSO6 |
 		NETIF_F_IP_CSUM | NETIF_F_IPV6_CSUM;
@@ -5205,12 +5198,15 @@ static void be_netdev_init(struct net_device *netdev)
 
 	netdev->flags |= IFF_MULTICAST;
 
-	netif_set_gso_max_size(netdev, BE_MAX_GSO_SIZE - ETH_HLEN);
+	netif_set_tso_max_size(netdev, BE_MAX_GSO_SIZE - ETH_HLEN);
 
 	netdev->netdev_ops = &be_netdev_ops;
 
 	netdev->ethtool_ops = &be_ethtool_ops;
 
+	if (!lancer_chip(adapter) && !BEx_chip(adapter) && !be_is_mc(adapter))
+		netdev->udp_tunnel_nic_info = &be_udp_tunnels;
+
 	/* MTU range: 256 - 9000 */
 	netdev->min_mtu = BE_MIN_MTU;
 	netdev->max_mtu = BE_MAX_MTU;
@@ -5636,9 +5632,9 @@ static int be_drv_init(struct be_adapter *adapter)
 	int status = 0;
 
 	mbox_mem_alloc->size = sizeof(struct be_mcc_mailbox) + 16;
-	mbox_mem_alloc->va = dma_zalloc_coherent(dev, mbox_mem_alloc->size,
-						 &mbox_mem_alloc->dma,
-						 GFP_KERNEL);
+	mbox_mem_alloc->va = dma_alloc_coherent(dev, mbox_mem_alloc->size,
+						&mbox_mem_alloc->dma,
+						GFP_KERNEL);
 	if (!mbox_mem_alloc->va)
 		return -ENOMEM;
 
@@ -5647,8 +5643,8 @@ static int be_drv_init(struct be_adapter *adapter)
 	mbox_mem_align->dma = PTR_ALIGN(mbox_mem_alloc->dma, 16);
 
 	rx_filter->size = sizeof(struct be_cmd_req_rx_filter);
-	rx_filter->va = dma_zalloc_coherent(dev, rx_filter->size,
-					    &rx_filter->dma, GFP_KERNEL);
+	rx_filter->va = dma_alloc_coherent(dev, rx_filter->size,
+					   &rx_filter->dma, GFP_KERNEL);
 	if (!rx_filter->va) {
 		status = -ENOMEM;
 		goto free_mbox;
@@ -5662,16 +5658,16 @@ static int be_drv_init(struct be_adapter *adapter)
 		stats_cmd->size = sizeof(struct be_cmd_req_get_stats_v1);
 	else
 		stats_cmd->size = sizeof(struct be_cmd_req_get_stats_v2);
-	stats_cmd->va = dma_zalloc_coherent(dev, stats_cmd->size,
-					    &stats_cmd->dma, GFP_KERNEL);
+	stats_cmd->va = dma_alloc_coherent(dev, stats_cmd->size,
+					   &stats_cmd->dma, GFP_KERNEL);
 	if (!stats_cmd->va) {
 		status = -ENOMEM;
 		goto free_rx_filter;
 	}
 
 	mutex_init(&adapter->mbox_lock);
-	mutex_init(&adapter->mcc_lock);
 	mutex_init(&adapter->rx_filter_lock);
+	spin_lock_init(&adapter->mcc_lock);
 	spin_lock_init(&adapter->mcc_cq_lock);
 	init_completion(&adapter->et_cmd_compl);
 
@@ -5690,7 +5686,6 @@ static int be_drv_init(struct be_adapter *adapter)
 	/* Must be a power of 2 or else MODULO will BUG_ON */
 	adapter->be_get_temp_freq = 64;
 
-	INIT_LIST_HEAD(&adapter->vxlan_port_list);
 	return 0;
 
 free_rx_filter:
@@ -5726,8 +5721,6 @@ static void be_remove(struct pci_dev *pdev)
 	be_unmap_pci_bars(adapter);
 	be_drv_cleanup(adapter);
 
-	pci_disable_pcie_error_reporting(pdev);
-
 	pci_release_regions(pdev);
 	pci_disable_device(pdev);
 
@@ -5748,7 +5741,7 @@ static ssize_t be_hwmon_show_temp(struct device *dev,
 			       adapter->hwmon_info.be_on_die_temp * 1000);
 }
 
-static SENSOR_DEVICE_ATTR(temp1_input, S_IRUGO,
+static SENSOR_DEVICE_ATTR(temp1_input, 0444,
 			  be_hwmon_show_temp, NULL, 1);
 
 static struct attribute *be_hwmon_attrs[] = {
@@ -5819,8 +5812,6 @@ static int be_probe(struct pci_dev *pdev, const struct pci_device_id *pdev_id)
 	struct net_device *netdev;
 	int status = 0;
 
-	dev_info(&pdev->dev, "%s version is %s\n", DRV_NAME, DRV_VER);
-
 	status = pci_enable_device(pdev);
 	if (status)
 		goto do_none;
@@ -5842,20 +5833,11 @@ static int be_probe(struct pci_dev *pdev, const struct pci_device_id *pdev_id)
 	SET_NETDEV_DEV(netdev, &pdev->dev);
 
 	status = dma_set_mask_and_coherent(&pdev->dev, DMA_BIT_MASK(64));
-	if (!status) {
-		netdev->features |= NETIF_F_HIGHDMA;
-	} else {
-		status = dma_set_mask_and_coherent(&pdev->dev, DMA_BIT_MASK(32));
-		if (status) {
-			dev_err(&pdev->dev, "Could not set PCI DMA Mask\n");
-			goto free_netdev;
-		}
+	if (status) {
+		dev_err(&pdev->dev, "Could not set PCI DMA Mask\n");
+		goto free_netdev;
 	}
 
-	status = pci_enable_pcie_error_reporting(pdev);
-	if (!status)
-		dev_info(&pdev->dev, "PCIe error reporting enabled\n");
-
 	status = be_map_pci_bars(adapter);
 	if (status)
 		goto free_netdev;
@@ -5910,32 +5892,23 @@ do_none:
 	return status;
 }
 
-static int be_suspend(struct pci_dev *pdev, pm_message_t state)
+static int __maybe_unused be_suspend(struct device *dev_d)
 {
-	struct be_adapter *adapter = pci_get_drvdata(pdev);
+	struct be_adapter *adapter = dev_get_drvdata(dev_d);
 
 	be_intr_set(adapter, false);
 	be_cancel_err_detection(adapter);
 
 	be_cleanup(adapter);
 
-	pci_save_state(pdev);
-	pci_disable_device(pdev);
-	pci_set_power_state(pdev, pci_choose_state(pdev, state));
 	return 0;
 }
 
-static int be_pci_resume(struct pci_dev *pdev)
+static int __maybe_unused be_pci_resume(struct device *dev_d)
 {
-	struct be_adapter *adapter = pci_get_drvdata(pdev);
+	struct be_adapter *adapter = dev_get_drvdata(dev_d);
 	int status = 0;
 
-	status = pci_enable_device(pdev);
-	if (status)
-		return status;
-
-	pci_restore_state(pdev);
-
 	status = be_resume(adapter);
 	if (status)
 		return status;
@@ -6021,7 +5994,6 @@ static pci_ers_result_t be_eeh_reset(struct pci_dev *pdev)
 	if (status)
 		return PCI_ERS_RESULT_DISCONNECT;
 
-	pci_cleanup_aer_uncorrect_error_status(pdev);
 	be_clear_error(adapter, BE_CLEAR_ALL);
 	return PCI_ERS_RESULT_RECOVERED;
 }
@@ -6108,13 +6080,14 @@ static const struct pci_error_handlers be_eeh_handlers = {
 	.resume = be_eeh_resume,
 };
 
+static SIMPLE_DEV_PM_OPS(be_pci_pm_ops, be_suspend, be_pci_resume);
+
 static struct pci_driver be_driver = {
 	.name = DRV_NAME,
 	.id_table = be_dev_ids,
 	.probe = be_probe,
 	.remove = be_remove,
-	.suspend = be_suspend,
-	.resume = be_pci_resume,
+	.driver.pm = &be_pci_pm_ops,
 	.shutdown = be_shutdown,
 	.sriov_configure = be_pci_sriov_configure,
 	.err_handler = &be_eeh_handlers
diff --git a/drivers/net/ethernet/emulex/benet/be_roce.c b/drivers/net/ethernet/emulex/benet/be_roce.c
index 2b62841c4c63..521c4c2b4887 100644
--- a/drivers/net/ethernet/emulex/benet/be_roce.c
+++ b/drivers/net/ethernet/emulex/benet/be_roce.c
@@ -1,12 +1,8 @@
+// SPDX-License-Identifier: GPL-2.0-only
 /*
  * Copyright (C) 2005 - 2016 Broadcom
  * All rights reserved.
  *
- * This program is free software; you can redistribute it and/or
- * modify it under the terms of the GNU General Public License version 2
- * as published by the Free Software Foundation. The full GNU General
- * Public License is included in this distribution in the file called COPYING.
- *
  * Contact Information:
  * linux-drivers@emulex.com
  *
@@ -139,10 +135,7 @@ int be_roce_register_driver(struct ocrdma_driver *drv)
 	}
 	ocrdma_drv = drv;
 	list_for_each_entry(dev, &be_adapter_list, entry) {
-		struct net_device *netdev;
-
 		_be_roce_dev_add(dev);
-		netdev = dev->netdev;
 	}
 	mutex_unlock(&be_adapter_list_lock);
 	return 0;
diff --git a/drivers/net/ethernet/emulex/benet/be_roce.h b/drivers/net/ethernet/emulex/benet/be_roce.h
index e51719a7307f..801e10522129 100644
--- a/drivers/net/ethernet/emulex/benet/be_roce.h
+++ b/drivers/net/ethernet/emulex/benet/be_roce.h
@@ -1,12 +1,8 @@
+/* SPDX-License-Identifier: GPL-2.0-only */
 /*
  * Copyright (C) 2005 - 2016 Broadcom
  * All rights reserved.
  *
- * This program is free software; you can redistribute it and/or
- * modify it under the terms of the GNU General Public License version 2
- * as published by the Free Software Foundation. The full GNU General
- * Public License is included in this distribution in the file called COPYING.
- *
  * Contact Information:
  * linux-drivers@emulex.com
  *
diff --git a/drivers/net/ethernet/engleder/Kconfig b/drivers/net/ethernet/engleder/Kconfig
new file mode 100644
index 000000000000..3df6bf476ae7
--- /dev/null
+++ b/drivers/net/ethernet/engleder/Kconfig
@@ -0,0 +1,40 @@
+# SPDX-License-Identifier: GPL-2.0
+#
+# Engleder network device configuration
+#
+
+config NET_VENDOR_ENGLEDER
+	bool "Engleder devices"
+	default y
+	help
+	  If you have a network (Ethernet) card belonging to this class, say Y.
+
+	  Note that the answer to this question doesn't directly affect the
+	  kernel: saying N will just cause the configurator to skip all
+	  the questions about Engleder devices. If you say Y, you will be asked
+	  for your specific card in the following questions.
+
+if NET_VENDOR_ENGLEDER
+
+config TSNEP
+	tristate "TSN endpoint support"
+	depends on HAS_IOMEM && HAS_DMA
+	depends on PTP_1588_CLOCK_OPTIONAL
+	select PHYLIB
+	select PAGE_POOL
+	help
+	  Support for the Engleder TSN endpoint Ethernet MAC IP Core.
+
+	  To compile this driver as a module, choose M here. The module will be
+	  called tsnep.
+
+config TSNEP_SELFTESTS
+	bool "TSN endpoint self test support"
+	default n
+	depends on TSNEP
+	help
+	  This enables self test support within the TSN endpoint driver.
+
+	  If unsure, say N.
+
+endif # NET_VENDOR_ENGLEDER
diff --git a/drivers/net/ethernet/engleder/Makefile b/drivers/net/ethernet/engleder/Makefile
new file mode 100644
index 000000000000..b98135f65eb7
--- /dev/null
+++ b/drivers/net/ethernet/engleder/Makefile
@@ -0,0 +1,10 @@
+# SPDX-License-Identifier: GPL-2.0
+#
+# Makefile for the Engleder Ethernet drivers
+#
+
+obj-$(CONFIG_TSNEP) += tsnep.o
+
+tsnep-objs := tsnep_main.o tsnep_ethtool.o tsnep_ptp.o tsnep_tc.o \
+	      tsnep_rxnfc.o tsnep_xdp.o
+tsnep-$(CONFIG_TSNEP_SELFTESTS) += tsnep_selftests.o
diff --git a/drivers/net/ethernet/engleder/tsnep.h b/drivers/net/ethernet/engleder/tsnep.h
new file mode 100644
index 000000000000..f188fba021a6
--- /dev/null
+++ b/drivers/net/ethernet/engleder/tsnep.h
@@ -0,0 +1,260 @@
+/* SPDX-License-Identifier: GPL-2.0 */
+/* Copyright (C) 2021 Gerhard Engleder <gerhard@engleder-embedded.com> */
+
+#ifndef _TSNEP_H
+#define _TSNEP_H
+
+#include "tsnep_hw.h"
+
+#include <linux/platform_device.h>
+#include <linux/dma-mapping.h>
+#include <linux/etherdevice.h>
+#include <linux/phy.h>
+#include <linux/ethtool.h>
+#include <linux/net_tstamp.h>
+#include <linux/ptp_clock_kernel.h>
+#include <linux/miscdevice.h>
+#include <net/xdp.h>
+
+#define TSNEP "tsnep"
+
+#define TSNEP_RING_SIZE 256
+#define TSNEP_RING_MASK (TSNEP_RING_SIZE - 1)
+#define TSNEP_RING_RX_REFILL 16
+#define TSNEP_RING_RX_REUSE (TSNEP_RING_SIZE - TSNEP_RING_SIZE / 4)
+#define TSNEP_RING_ENTRIES_PER_PAGE (PAGE_SIZE / TSNEP_DESC_SIZE)
+#define TSNEP_RING_PAGE_COUNT (TSNEP_RING_SIZE / TSNEP_RING_ENTRIES_PER_PAGE)
+
+struct tsnep_gcl {
+	void __iomem *addr;
+
+	u64 base_time;
+	u64 cycle_time;
+	u64 cycle_time_extension;
+
+	struct tsnep_gcl_operation operation[TSNEP_GCL_COUNT];
+	int count;
+
+	u64 change_limit;
+
+	u64 start_time;
+	bool change;
+};
+
+enum tsnep_rxnfc_filter_type {
+	TSNEP_RXNFC_ETHER_TYPE,
+};
+
+struct tsnep_rxnfc_filter {
+	enum tsnep_rxnfc_filter_type type;
+	union {
+		u16 ether_type;
+	};
+};
+
+struct tsnep_rxnfc_rule {
+	struct list_head list;
+	struct tsnep_rxnfc_filter filter;
+	int queue_index;
+	int location;
+};
+
+struct tsnep_tx_entry {
+	struct tsnep_tx_desc *desc;
+	struct tsnep_tx_desc_wb *desc_wb;
+	dma_addr_t desc_dma;
+	bool owner_user_flag;
+
+	u32 properties;
+
+	u32 type;
+	union {
+		struct sk_buff *skb;
+		struct xdp_frame *xdpf;
+		bool zc;
+	};
+	size_t len;
+	DEFINE_DMA_UNMAP_ADDR(dma);
+};
+
+struct tsnep_tx {
+	struct tsnep_adapter *adapter;
+	void __iomem *addr;
+	int queue_index;
+
+	void *page[TSNEP_RING_PAGE_COUNT];
+	dma_addr_t page_dma[TSNEP_RING_PAGE_COUNT];
+
+	struct tsnep_tx_entry entry[TSNEP_RING_SIZE];
+	int write;
+	int read;
+	u32 owner_counter;
+	int increment_owner_counter;
+	struct xsk_buff_pool *xsk_pool;
+
+	u32 packets;
+	u32 bytes;
+	u32 dropped;
+};
+
+struct tsnep_rx_entry {
+	struct tsnep_rx_desc *desc;
+	struct tsnep_rx_desc_wb *desc_wb;
+	dma_addr_t desc_dma;
+
+	u32 properties;
+
+	union {
+		struct page *page;
+		struct xdp_buff *xdp;
+	};
+	size_t len;
+	dma_addr_t dma;
+};
+
+struct tsnep_rx {
+	struct tsnep_adapter *adapter;
+	void __iomem *addr;
+	int queue_index;
+	int tx_queue_index;
+
+	void *page[TSNEP_RING_PAGE_COUNT];
+	dma_addr_t page_dma[TSNEP_RING_PAGE_COUNT];
+
+	struct tsnep_rx_entry entry[TSNEP_RING_SIZE];
+	int write;
+	int read;
+	u32 owner_counter;
+	int increment_owner_counter;
+	struct page_pool *page_pool;
+	struct page **page_buffer;
+	struct xsk_buff_pool *xsk_pool;
+	struct xdp_buff **xdp_batch;
+
+	u32 packets;
+	u32 bytes;
+	u32 dropped;
+	u32 multicast;
+	u32 alloc_failed;
+
+	struct xdp_rxq_info xdp_rxq;
+	struct xdp_rxq_info xdp_rxq_zc;
+};
+
+struct tsnep_queue {
+	struct tsnep_adapter *adapter;
+	char name[IFNAMSIZ + 16];
+
+	struct tsnep_tx *tx;
+	struct tsnep_rx *rx;
+
+	struct napi_struct napi;
+
+	int irq;
+	u32 irq_mask;
+	void __iomem *irq_delay_addr;
+	u8 irq_delay;
+};
+
+struct tsnep_adapter {
+	struct net_device *netdev;
+	u8 mac_address[ETH_ALEN];
+	struct mii_bus *mdiobus;
+	bool suppress_preamble;
+	phy_interface_t phy_mode;
+	struct phy_device *phydev;
+	int msg_enable;
+
+	struct platform_device *pdev;
+	struct device *dmadev;
+	void __iomem *addr;
+
+	bool gate_control;
+	/* gate control lock */
+	struct mutex gate_control_lock;
+	bool gate_control_active;
+	struct tsnep_gcl gcl[2];
+	int next_gcl;
+
+	struct hwtstamp_config hwtstamp_config;
+	struct ptp_clock *ptp_clock;
+	struct ptp_clock_info ptp_clock_info;
+	/* ptp clock lock */
+	spinlock_t ptp_lock;
+
+	/* RX flow classification rules lock */
+	struct mutex rxnfc_lock;
+	struct list_head rxnfc_rules;
+	int rxnfc_count;
+	int rxnfc_max;
+
+	struct bpf_prog *xdp_prog;
+
+	int num_tx_queues;
+	struct tsnep_tx tx[TSNEP_MAX_QUEUES];
+	int num_rx_queues;
+	struct tsnep_rx rx[TSNEP_MAX_QUEUES];
+
+	int num_queues;
+	struct tsnep_queue queue[TSNEP_MAX_QUEUES];
+};
+
+extern const struct ethtool_ops tsnep_ethtool_ops;
+
+int tsnep_ptp_init(struct tsnep_adapter *adapter);
+void tsnep_ptp_cleanup(struct tsnep_adapter *adapter);
+int tsnep_ptp_ioctl(struct net_device *netdev, struct ifreq *ifr, int cmd);
+
+int tsnep_tc_init(struct tsnep_adapter *adapter);
+void tsnep_tc_cleanup(struct tsnep_adapter *adapter);
+int tsnep_tc_setup(struct net_device *netdev, enum tc_setup_type type,
+		   void *type_data);
+
+int tsnep_rxnfc_init(struct tsnep_adapter *adapter);
+void tsnep_rxnfc_cleanup(struct tsnep_adapter *adapter);
+int tsnep_rxnfc_get_rule(struct tsnep_adapter *adapter,
+			 struct ethtool_rxnfc *cmd);
+int tsnep_rxnfc_get_all(struct tsnep_adapter *adapter,
+			struct ethtool_rxnfc *cmd,
+			u32 *rule_locs);
+int tsnep_rxnfc_add_rule(struct tsnep_adapter *adapter,
+			 struct ethtool_rxnfc *cmd);
+int tsnep_rxnfc_del_rule(struct tsnep_adapter *adapter,
+			 struct ethtool_rxnfc *cmd);
+
+int tsnep_xdp_setup_prog(struct tsnep_adapter *adapter, struct bpf_prog *prog,
+			 struct netlink_ext_ack *extack);
+int tsnep_xdp_setup_pool(struct tsnep_adapter *adapter,
+			 struct xsk_buff_pool *pool, u16 queue_id);
+
+#if IS_ENABLED(CONFIG_TSNEP_SELFTESTS)
+int tsnep_ethtool_get_test_count(void);
+void tsnep_ethtool_get_test_strings(u8 *data);
+void tsnep_ethtool_self_test(struct net_device *netdev,
+			     struct ethtool_test *eth_test, u64 *data);
+#else
+static inline int tsnep_ethtool_get_test_count(void)
+{
+	return -EOPNOTSUPP;
+}
+
+static inline void tsnep_ethtool_get_test_strings(u8 *data)
+{
+	/* not enabled */
+}
+
+static inline void tsnep_ethtool_self_test(struct net_device *dev,
+					   struct ethtool_test *eth_test,
+					   u64 *data)
+{
+	/* not enabled */
+}
+#endif /* CONFIG_TSNEP_SELFTESTS */
+
+void tsnep_get_system_time(struct tsnep_adapter *adapter, u64 *time);
+int tsnep_set_irq_coalesce(struct tsnep_queue *queue, u32 usecs);
+u32 tsnep_get_irq_coalesce(struct tsnep_queue *queue);
+int tsnep_enable_xsk(struct tsnep_queue *queue, struct xsk_buff_pool *pool);
+void tsnep_disable_xsk(struct tsnep_queue *queue);
+
+#endif /* _TSNEP_H */
diff --git a/drivers/net/ethernet/engleder/tsnep_ethtool.c b/drivers/net/ethernet/engleder/tsnep_ethtool.c
new file mode 100644
index 000000000000..228a638eae16
--- /dev/null
+++ b/drivers/net/ethernet/engleder/tsnep_ethtool.c
@@ -0,0 +1,480 @@
+// SPDX-License-Identifier: GPL-2.0
+/* Copyright (C) 2021 Gerhard Engleder <gerhard@engleder-embedded.com> */
+
+#include "tsnep.h"
+
+static const char tsnep_stats_strings[][ETH_GSTRING_LEN] = {
+	"rx_packets",
+	"rx_bytes",
+	"rx_dropped",
+	"rx_multicast",
+	"rx_alloc_failed",
+	"rx_phy_errors",
+	"rx_forwarded_phy_errors",
+	"rx_invalid_frame_errors",
+	"tx_packets",
+	"tx_bytes",
+	"tx_dropped",
+};
+
+struct tsnep_stats {
+	u64 rx_packets;
+	u64 rx_bytes;
+	u64 rx_dropped;
+	u64 rx_multicast;
+	u64 rx_alloc_failed;
+	u64 rx_phy_errors;
+	u64 rx_forwarded_phy_errors;
+	u64 rx_invalid_frame_errors;
+	u64 tx_packets;
+	u64 tx_bytes;
+	u64 tx_dropped;
+};
+
+#define TSNEP_STATS_COUNT (sizeof(struct tsnep_stats) / sizeof(u64))
+
+static const char tsnep_rx_queue_stats_strings[][ETH_GSTRING_LEN] = {
+	"rx_%d_packets",
+	"rx_%d_bytes",
+	"rx_%d_dropped",
+	"rx_%d_multicast",
+	"rx_%d_alloc_failed",
+	"rx_%d_no_descriptor_errors",
+	"rx_%d_buffer_too_small_errors",
+	"rx_%d_fifo_overflow_errors",
+	"rx_%d_invalid_frame_errors",
+};
+
+struct tsnep_rx_queue_stats {
+	u64 rx_packets;
+	u64 rx_bytes;
+	u64 rx_dropped;
+	u64 rx_multicast;
+	u64 rx_alloc_failed;
+	u64 rx_no_descriptor_errors;
+	u64 rx_buffer_too_small_errors;
+	u64 rx_fifo_overflow_errors;
+	u64 rx_invalid_frame_errors;
+};
+
+#define TSNEP_RX_QUEUE_STATS_COUNT (sizeof(struct tsnep_rx_queue_stats) / \
+				    sizeof(u64))
+
+static const char tsnep_tx_queue_stats_strings[][ETH_GSTRING_LEN] = {
+	"tx_%d_packets",
+	"tx_%d_bytes",
+	"tx_%d_dropped",
+};
+
+struct tsnep_tx_queue_stats {
+	u64 tx_packets;
+	u64 tx_bytes;
+	u64 tx_dropped;
+};
+
+#define TSNEP_TX_QUEUE_STATS_COUNT (sizeof(struct tsnep_tx_queue_stats) / \
+				    sizeof(u64))
+
+static void tsnep_ethtool_get_drvinfo(struct net_device *netdev,
+				      struct ethtool_drvinfo *drvinfo)
+{
+	struct tsnep_adapter *adapter = netdev_priv(netdev);
+
+	strscpy(drvinfo->driver, TSNEP, sizeof(drvinfo->driver));
+	strscpy(drvinfo->bus_info, dev_name(&adapter->pdev->dev),
+		sizeof(drvinfo->bus_info));
+}
+
+static int tsnep_ethtool_get_regs_len(struct net_device *netdev)
+{
+	struct tsnep_adapter *adapter = netdev_priv(netdev);
+	int len;
+	int num_additional_queues;
+
+	len = TSNEP_MAC_SIZE;
+
+	/* first queue pair is within TSNEP_MAC_SIZE, only queues additional to
+	 * the first queue pair extend the register length by TSNEP_QUEUE_SIZE
+	 */
+	num_additional_queues =
+		max(adapter->num_tx_queues, adapter->num_rx_queues) - 1;
+	len += TSNEP_QUEUE_SIZE * num_additional_queues;
+
+	return len;
+}
+
+static void tsnep_ethtool_get_regs(struct net_device *netdev,
+				   struct ethtool_regs *regs,
+				   void *p)
+{
+	struct tsnep_adapter *adapter = netdev_priv(netdev);
+
+	regs->version = 1;
+
+	memcpy_fromio(p, adapter->addr, regs->len);
+}
+
+static u32 tsnep_ethtool_get_msglevel(struct net_device *netdev)
+{
+	struct tsnep_adapter *adapter = netdev_priv(netdev);
+
+	return adapter->msg_enable;
+}
+
+static void tsnep_ethtool_set_msglevel(struct net_device *netdev, u32 data)
+{
+	struct tsnep_adapter *adapter = netdev_priv(netdev);
+
+	adapter->msg_enable = data;
+}
+
+static void tsnep_ethtool_get_strings(struct net_device *netdev, u32 stringset,
+				      u8 *data)
+{
+	struct tsnep_adapter *adapter = netdev_priv(netdev);
+	int rx_count = adapter->num_rx_queues;
+	int tx_count = adapter->num_tx_queues;
+	int i, j;
+
+	switch (stringset) {
+	case ETH_SS_STATS:
+		memcpy(data, tsnep_stats_strings, sizeof(tsnep_stats_strings));
+		data += sizeof(tsnep_stats_strings);
+
+		for (i = 0; i < rx_count; i++) {
+			for (j = 0; j < TSNEP_RX_QUEUE_STATS_COUNT; j++) {
+				snprintf(data, ETH_GSTRING_LEN,
+					 tsnep_rx_queue_stats_strings[j], i);
+				data += ETH_GSTRING_LEN;
+			}
+		}
+
+		for (i = 0; i < tx_count; i++) {
+			for (j = 0; j < TSNEP_TX_QUEUE_STATS_COUNT; j++) {
+				snprintf(data, ETH_GSTRING_LEN,
+					 tsnep_tx_queue_stats_strings[j], i);
+				data += ETH_GSTRING_LEN;
+			}
+		}
+		break;
+	case ETH_SS_TEST:
+		tsnep_ethtool_get_test_strings(data);
+		break;
+	}
+}
+
+static void tsnep_ethtool_get_ethtool_stats(struct net_device *netdev,
+					    struct ethtool_stats *stats,
+					    u64 *data)
+{
+	struct tsnep_adapter *adapter = netdev_priv(netdev);
+	int rx_count = adapter->num_rx_queues;
+	int tx_count = adapter->num_tx_queues;
+	struct tsnep_stats tsnep_stats;
+	struct tsnep_rx_queue_stats tsnep_rx_queue_stats;
+	struct tsnep_tx_queue_stats tsnep_tx_queue_stats;
+	u32 reg;
+	int i;
+
+	memset(&tsnep_stats, 0, sizeof(tsnep_stats));
+	for (i = 0; i < adapter->num_rx_queues; i++) {
+		tsnep_stats.rx_packets += adapter->rx[i].packets;
+		tsnep_stats.rx_bytes += adapter->rx[i].bytes;
+		tsnep_stats.rx_dropped += adapter->rx[i].dropped;
+		tsnep_stats.rx_multicast += adapter->rx[i].multicast;
+		tsnep_stats.rx_alloc_failed += adapter->rx[i].alloc_failed;
+	}
+	reg = ioread32(adapter->addr + ECM_STAT);
+	tsnep_stats.rx_phy_errors =
+		(reg & ECM_STAT_RX_ERR_MASK) >> ECM_STAT_RX_ERR_SHIFT;
+	tsnep_stats.rx_forwarded_phy_errors =
+		(reg & ECM_STAT_FWD_RX_ERR_MASK) >> ECM_STAT_FWD_RX_ERR_SHIFT;
+	tsnep_stats.rx_invalid_frame_errors =
+		(reg & ECM_STAT_INV_FRM_MASK) >> ECM_STAT_INV_FRM_SHIFT;
+	for (i = 0; i < adapter->num_tx_queues; i++) {
+		tsnep_stats.tx_packets += adapter->tx[i].packets;
+		tsnep_stats.tx_bytes += adapter->tx[i].bytes;
+		tsnep_stats.tx_dropped += adapter->tx[i].dropped;
+	}
+	memcpy(data, &tsnep_stats, sizeof(tsnep_stats));
+	data += TSNEP_STATS_COUNT;
+
+	for (i = 0; i < rx_count; i++) {
+		memset(&tsnep_rx_queue_stats, 0, sizeof(tsnep_rx_queue_stats));
+		tsnep_rx_queue_stats.rx_packets = adapter->rx[i].packets;
+		tsnep_rx_queue_stats.rx_bytes = adapter->rx[i].bytes;
+		tsnep_rx_queue_stats.rx_dropped = adapter->rx[i].dropped;
+		tsnep_rx_queue_stats.rx_multicast = adapter->rx[i].multicast;
+		tsnep_rx_queue_stats.rx_alloc_failed =
+			adapter->rx[i].alloc_failed;
+		reg = ioread32(adapter->addr + TSNEP_QUEUE(i) +
+			       TSNEP_RX_STATISTIC);
+		tsnep_rx_queue_stats.rx_no_descriptor_errors =
+			(reg & TSNEP_RX_STATISTIC_NO_DESC_MASK) >>
+			TSNEP_RX_STATISTIC_NO_DESC_SHIFT;
+		tsnep_rx_queue_stats.rx_buffer_too_small_errors =
+			(reg & TSNEP_RX_STATISTIC_BUFFER_TOO_SMALL_MASK) >>
+			TSNEP_RX_STATISTIC_BUFFER_TOO_SMALL_SHIFT;
+		tsnep_rx_queue_stats.rx_fifo_overflow_errors =
+			(reg & TSNEP_RX_STATISTIC_FIFO_OVERFLOW_MASK) >>
+			TSNEP_RX_STATISTIC_FIFO_OVERFLOW_SHIFT;
+		tsnep_rx_queue_stats.rx_invalid_frame_errors =
+			(reg & TSNEP_RX_STATISTIC_INVALID_FRAME_MASK) >>
+			TSNEP_RX_STATISTIC_INVALID_FRAME_SHIFT;
+		memcpy(data, &tsnep_rx_queue_stats,
+		       sizeof(tsnep_rx_queue_stats));
+		data += TSNEP_RX_QUEUE_STATS_COUNT;
+	}
+
+	for (i = 0; i < tx_count; i++) {
+		memset(&tsnep_tx_queue_stats, 0, sizeof(tsnep_tx_queue_stats));
+		tsnep_tx_queue_stats.tx_packets += adapter->tx[i].packets;
+		tsnep_tx_queue_stats.tx_bytes += adapter->tx[i].bytes;
+		tsnep_tx_queue_stats.tx_dropped += adapter->tx[i].dropped;
+		memcpy(data, &tsnep_tx_queue_stats,
+		       sizeof(tsnep_tx_queue_stats));
+		data += TSNEP_TX_QUEUE_STATS_COUNT;
+	}
+}
+
+static int tsnep_ethtool_get_sset_count(struct net_device *netdev, int sset)
+{
+	struct tsnep_adapter *adapter = netdev_priv(netdev);
+	int rx_count;
+	int tx_count;
+
+	switch (sset) {
+	case ETH_SS_STATS:
+		rx_count = adapter->num_rx_queues;
+		tx_count = adapter->num_tx_queues;
+		return TSNEP_STATS_COUNT +
+		       TSNEP_RX_QUEUE_STATS_COUNT * rx_count +
+		       TSNEP_TX_QUEUE_STATS_COUNT * tx_count;
+	case ETH_SS_TEST:
+		return tsnep_ethtool_get_test_count();
+	default:
+		return -EOPNOTSUPP;
+	}
+}
+
+static int tsnep_ethtool_get_rxnfc(struct net_device *netdev,
+				   struct ethtool_rxnfc *cmd, u32 *rule_locs)
+{
+	struct tsnep_adapter *adapter = netdev_priv(netdev);
+
+	switch (cmd->cmd) {
+	case ETHTOOL_GRXRINGS:
+		cmd->data = adapter->num_rx_queues;
+		return 0;
+	case ETHTOOL_GRXCLSRLCNT:
+		cmd->rule_cnt = adapter->rxnfc_count;
+		cmd->data = adapter->rxnfc_max;
+		cmd->data |= RX_CLS_LOC_SPECIAL;
+		return 0;
+	case ETHTOOL_GRXCLSRULE:
+		return tsnep_rxnfc_get_rule(adapter, cmd);
+	case ETHTOOL_GRXCLSRLALL:
+		return tsnep_rxnfc_get_all(adapter, cmd, rule_locs);
+	default:
+		return -EOPNOTSUPP;
+	}
+}
+
+static int tsnep_ethtool_set_rxnfc(struct net_device *netdev,
+				   struct ethtool_rxnfc *cmd)
+{
+	struct tsnep_adapter *adapter = netdev_priv(netdev);
+
+	switch (cmd->cmd) {
+	case ETHTOOL_SRXCLSRLINS:
+		return tsnep_rxnfc_add_rule(adapter, cmd);
+	case ETHTOOL_SRXCLSRLDEL:
+		return tsnep_rxnfc_del_rule(adapter, cmd);
+	default:
+		return -EOPNOTSUPP;
+	}
+}
+
+static void tsnep_ethtool_get_channels(struct net_device *netdev,
+				       struct ethtool_channels *ch)
+{
+	struct tsnep_adapter *adapter = netdev_priv(netdev);
+
+	ch->max_combined = adapter->num_queues;
+	ch->combined_count = adapter->num_queues;
+}
+
+static int tsnep_ethtool_get_ts_info(struct net_device *netdev,
+				     struct kernel_ethtool_ts_info *info)
+{
+	struct tsnep_adapter *adapter = netdev_priv(netdev);
+
+	info->so_timestamping = SOF_TIMESTAMPING_TX_SOFTWARE |
+				SOF_TIMESTAMPING_TX_HARDWARE |
+				SOF_TIMESTAMPING_RX_HARDWARE |
+				SOF_TIMESTAMPING_RAW_HARDWARE;
+
+	if (adapter->ptp_clock)
+		info->phc_index = ptp_clock_index(adapter->ptp_clock);
+
+	info->tx_types = BIT(HWTSTAMP_TX_OFF) |
+			 BIT(HWTSTAMP_TX_ON);
+	info->rx_filters = BIT(HWTSTAMP_FILTER_NONE) |
+			   BIT(HWTSTAMP_FILTER_ALL);
+
+	return 0;
+}
+
+static struct tsnep_queue *tsnep_get_queue_with_tx(struct tsnep_adapter *adapter,
+						   int index)
+{
+	int i;
+
+	for (i = 0; i < adapter->num_queues; i++) {
+		if (adapter->queue[i].tx) {
+			if (index == 0)
+				return &adapter->queue[i];
+
+			index--;
+		}
+	}
+
+	return NULL;
+}
+
+static struct tsnep_queue *tsnep_get_queue_with_rx(struct tsnep_adapter *adapter,
+						   int index)
+{
+	int i;
+
+	for (i = 0; i < adapter->num_queues; i++) {
+		if (adapter->queue[i].rx) {
+			if (index == 0)
+				return &adapter->queue[i];
+
+			index--;
+		}
+	}
+
+	return NULL;
+}
+
+static int tsnep_ethtool_get_coalesce(struct net_device *netdev,
+				      struct ethtool_coalesce *ec,
+				      struct kernel_ethtool_coalesce *kernel_coal,
+				      struct netlink_ext_ack *extack)
+{
+	struct tsnep_adapter *adapter = netdev_priv(netdev);
+	struct tsnep_queue *queue;
+
+	queue = tsnep_get_queue_with_rx(adapter, 0);
+	if (queue)
+		ec->rx_coalesce_usecs = tsnep_get_irq_coalesce(queue);
+
+	queue = tsnep_get_queue_with_tx(adapter, 0);
+	if (queue)
+		ec->tx_coalesce_usecs = tsnep_get_irq_coalesce(queue);
+
+	return 0;
+}
+
+static int tsnep_ethtool_set_coalesce(struct net_device *netdev,
+				      struct ethtool_coalesce *ec,
+				      struct kernel_ethtool_coalesce *kernel_coal,
+				      struct netlink_ext_ack *extack)
+{
+	struct tsnep_adapter *adapter = netdev_priv(netdev);
+	int i;
+	int retval;
+
+	for (i = 0; i < adapter->num_queues; i++) {
+		/* RX coalesce has priority for queues with TX and RX */
+		if (adapter->queue[i].rx)
+			retval = tsnep_set_irq_coalesce(&adapter->queue[i],
+							ec->rx_coalesce_usecs);
+		else
+			retval = tsnep_set_irq_coalesce(&adapter->queue[i],
+							ec->tx_coalesce_usecs);
+		if (retval != 0)
+			return retval;
+	}
+
+	return 0;
+}
+
+static int tsnep_ethtool_get_per_queue_coalesce(struct net_device *netdev,
+						u32 queue,
+						struct ethtool_coalesce *ec)
+{
+	struct tsnep_adapter *adapter = netdev_priv(netdev);
+	struct tsnep_queue *queue_with_rx;
+	struct tsnep_queue *queue_with_tx;
+
+	if (queue >= max(adapter->num_tx_queues, adapter->num_rx_queues))
+		return -EINVAL;
+
+	queue_with_rx = tsnep_get_queue_with_rx(adapter, queue);
+	if (queue_with_rx)
+		ec->rx_coalesce_usecs = tsnep_get_irq_coalesce(queue_with_rx);
+
+	queue_with_tx = tsnep_get_queue_with_tx(adapter, queue);
+	if (queue_with_tx)
+		ec->tx_coalesce_usecs = tsnep_get_irq_coalesce(queue_with_tx);
+
+	return 0;
+}
+
+static int tsnep_ethtool_set_per_queue_coalesce(struct net_device *netdev,
+						u32 queue,
+						struct ethtool_coalesce *ec)
+{
+	struct tsnep_adapter *adapter = netdev_priv(netdev);
+	struct tsnep_queue *queue_with_rx;
+	struct tsnep_queue *queue_with_tx;
+	int retval;
+
+	if (queue >= max(adapter->num_tx_queues, adapter->num_rx_queues))
+		return -EINVAL;
+
+	queue_with_rx = tsnep_get_queue_with_rx(adapter, queue);
+	if (queue_with_rx) {
+		retval = tsnep_set_irq_coalesce(queue_with_rx, ec->rx_coalesce_usecs);
+		if (retval != 0)
+			return retval;
+	}
+
+	/* RX coalesce has priority for queues with TX and RX */
+	queue_with_tx = tsnep_get_queue_with_tx(adapter, queue);
+	if (queue_with_tx && !queue_with_tx->rx) {
+		retval = tsnep_set_irq_coalesce(queue_with_tx, ec->tx_coalesce_usecs);
+		if (retval != 0)
+			return retval;
+	}
+
+	return 0;
+}
+
+const struct ethtool_ops tsnep_ethtool_ops = {
+	.supported_coalesce_params = ETHTOOL_COALESCE_USECS,
+	.get_drvinfo = tsnep_ethtool_get_drvinfo,
+	.get_regs_len = tsnep_ethtool_get_regs_len,
+	.get_regs = tsnep_ethtool_get_regs,
+	.get_msglevel = tsnep_ethtool_get_msglevel,
+	.set_msglevel = tsnep_ethtool_set_msglevel,
+	.nway_reset = phy_ethtool_nway_reset,
+	.get_link = ethtool_op_get_link,
+	.self_test = tsnep_ethtool_self_test,
+	.get_strings = tsnep_ethtool_get_strings,
+	.get_ethtool_stats = tsnep_ethtool_get_ethtool_stats,
+	.get_sset_count = tsnep_ethtool_get_sset_count,
+	.get_rxnfc = tsnep_ethtool_get_rxnfc,
+	.set_rxnfc = tsnep_ethtool_set_rxnfc,
+	.get_channels = tsnep_ethtool_get_channels,
+	.get_ts_info = tsnep_ethtool_get_ts_info,
+	.get_coalesce = tsnep_ethtool_get_coalesce,
+	.set_coalesce = tsnep_ethtool_set_coalesce,
+	.get_per_queue_coalesce = tsnep_ethtool_get_per_queue_coalesce,
+	.set_per_queue_coalesce = tsnep_ethtool_set_per_queue_coalesce,
+	.get_link_ksettings = phy_ethtool_get_link_ksettings,
+	.set_link_ksettings = phy_ethtool_set_link_ksettings,
+};
diff --git a/drivers/net/ethernet/engleder/tsnep_hw.h b/drivers/net/ethernet/engleder/tsnep_hw.h
new file mode 100644
index 000000000000..64c97eb66f67
--- /dev/null
+++ b/drivers/net/ethernet/engleder/tsnep_hw.h
@@ -0,0 +1,248 @@
+/* SPDX-License-Identifier: GPL-2.0 */
+/* Copyright (C) 2021 Gerhard Engleder <gerhard@engleder-embedded.com> */
+
+/* Hardware definition of TSNEP and EtherCAT MAC device */
+
+#ifndef _TSNEP_HW_H
+#define _TSNEP_HW_H
+
+#include <linux/types.h>
+
+/* type */
+#define ECM_TYPE 0x0000
+#define ECM_REVISION_MASK 0x000000FF
+#define ECM_REVISION_SHIFT 0
+#define ECM_VERSION_MASK 0x0000FF00
+#define ECM_VERSION_SHIFT 8
+#define ECM_QUEUE_COUNT_MASK 0x00070000
+#define ECM_QUEUE_COUNT_SHIFT 16
+#define ECM_GATE_CONTROL 0x02000000
+
+/* system time */
+#define ECM_SYSTEM_TIME_LOW 0x0008
+#define ECM_SYSTEM_TIME_HIGH 0x000C
+
+/* clock */
+#define ECM_CLOCK_RATE 0x0010
+#define ECM_CLOCK_RATE_OFFSET_MASK 0x7FFFFFFF
+#define ECM_CLOCK_RATE_OFFSET_SIGN 0x80000000
+
+/* interrupt */
+#define ECM_INT_ENABLE 0x0018
+#define ECM_INT_ACTIVE 0x001C
+#define ECM_INT_ACKNOWLEDGE 0x001C
+#define ECM_INT_LINK 0x00000020
+#define ECM_INT_TX_0 0x00000100
+#define ECM_INT_RX_0 0x00000200
+#define ECM_INT_TXRX_SHIFT 2
+#define ECM_INT_ALL 0x7FFFFFFF
+#define ECM_INT_DISABLE 0x80000000
+
+/* reset */
+#define ECM_RESET 0x0020
+#define ECM_RESET_COMMON 0x00000001
+#define ECM_RESET_CHANNEL 0x00000100
+#define ECM_RESET_TXRX 0x00010000
+
+/* counter */
+#define ECM_COUNTER_LOW 0x0028
+#define ECM_COUNTER_HIGH 0x002C
+
+/* interrupt delay */
+#define ECM_INT_DELAY 0x0030
+#define ECM_INT_DELAY_MASK 0xF0
+#define ECM_INT_DELAY_SHIFT 4
+#define ECM_INT_DELAY_BASE_US 16
+#define ECM_INT_DELAY_OFFSET 1
+
+/* control and status */
+#define ECM_STATUS 0x0080
+#define ECM_LINK_MODE_OFF 0x01000000
+#define ECM_LINK_MODE_100 0x02000000
+#define ECM_LINK_MODE_1000 0x04000000
+#define ECM_NO_LINK 0x01000000
+#define ECM_LINK_MODE_MASK 0x06000000
+
+/* management data */
+#define ECM_MD_CONTROL 0x0084
+#define ECM_MD_STATUS 0x0084
+#define ECM_MD_PREAMBLE 0x00000001
+#define ECM_MD_READ 0x00000004
+#define ECM_MD_WRITE 0x00000002
+#define ECM_MD_ADDR_MASK 0x000000F8
+#define ECM_MD_ADDR_SHIFT 3
+#define ECM_MD_PHY_ADDR_MASK 0x00001F00
+#define ECM_MD_PHY_ADDR_SHIFT 8
+#define ECM_MD_BUSY 0x00000001
+#define ECM_MD_DATA_MASK 0xFFFF0000
+#define ECM_MD_DATA_SHIFT 16
+
+/* statistic */
+#define ECM_STAT 0x00B0
+#define ECM_STAT_RX_ERR_MASK 0x000000FF
+#define ECM_STAT_RX_ERR_SHIFT 0
+#define ECM_STAT_INV_FRM_MASK 0x0000FF00
+#define ECM_STAT_INV_FRM_SHIFT 8
+#define ECM_STAT_FWD_RX_ERR_MASK 0x00FF0000
+#define ECM_STAT_FWD_RX_ERR_SHIFT 16
+
+/* tsnep */
+#define TSNEP_MAC_SIZE 0x4000
+#define TSNEP_QUEUE_SIZE 0x1000
+#define TSNEP_QUEUE(n) ({ typeof(n) __n = (n); \
+			  (__n) == 0 ? \
+			  0 : \
+			  TSNEP_MAC_SIZE + TSNEP_QUEUE_SIZE * ((__n) - 1); })
+#define TSNEP_MAX_QUEUES 8
+#define TSNEP_MAX_FRAME_SIZE (2 * 1024) /* hardware supports actually 16k */
+#define TSNEP_DESC_SIZE 256
+#define TSNEP_DESC_OFFSET 128
+
+/* tsnep register */
+#define TSNEP_INFO 0x0100
+#define TSNEP_INFO_TX_TIME 0x00010000
+#define TSNEP_CONTROL 0x0108
+#define TSNEP_CONTROL_TX_RESET 0x00000001
+#define TSNEP_CONTROL_TX_ENABLE 0x00000002
+#define TSNEP_CONTROL_TX_DMA_ERROR 0x00000010
+#define TSNEP_CONTROL_TX_DESC_ERROR 0x00000020
+#define TSNEP_CONTROL_RX_RESET 0x00000100
+#define TSNEP_CONTROL_RX_ENABLE 0x00000200
+#define TSNEP_CONTROL_RX_DISABLE 0x00000400
+#define TSNEP_CONTROL_RX_DMA_ERROR 0x00001000
+#define TSNEP_CONTROL_RX_DESC_ERROR 0x00002000
+#define TSNEP_TX_DESC_ADDR_LOW 0x0140
+#define TSNEP_TX_DESC_ADDR_HIGH 0x0144
+#define TSNEP_RX_DESC_ADDR_LOW 0x0180
+#define TSNEP_RX_DESC_ADDR_HIGH 0x0184
+#define TSNEP_RESET_OWNER_COUNTER 0x01
+#define TSNEP_RX_STATISTIC 0x0190
+#define TSNEP_RX_STATISTIC_NO_DESC_MASK 0x000000FF
+#define TSNEP_RX_STATISTIC_NO_DESC_SHIFT 0
+#define TSNEP_RX_STATISTIC_BUFFER_TOO_SMALL_MASK 0x0000FF00
+#define TSNEP_RX_STATISTIC_BUFFER_TOO_SMALL_SHIFT 8
+#define TSNEP_RX_STATISTIC_FIFO_OVERFLOW_MASK 0x00FF0000
+#define TSNEP_RX_STATISTIC_FIFO_OVERFLOW_SHIFT 16
+#define TSNEP_RX_STATISTIC_INVALID_FRAME_MASK 0xFF000000
+#define TSNEP_RX_STATISTIC_INVALID_FRAME_SHIFT 24
+#define TSNEP_RX_STATISTIC_NO_DESC 0x0190
+#define TSNEP_RX_STATISTIC_BUFFER_TOO_SMALL 0x0191
+#define TSNEP_RX_STATISTIC_FIFO_OVERFLOW 0x0192
+#define TSNEP_RX_STATISTIC_INVALID_FRAME 0x0193
+#define TSNEP_MAC_ADDRESS_LOW 0x0800
+#define TSNEP_MAC_ADDRESS_HIGH 0x0804
+#define TSNEP_RX_FILTER 0x0806
+#define TSNEP_RX_FILTER_ACCEPT_ALL_MULTICASTS 0x0001
+#define TSNEP_RX_FILTER_ACCEPT_ALL_UNICASTS 0x0002
+#define TSNEP_GC 0x0808
+#define TSNEP_GC_ENABLE_A 0x00000002
+#define TSNEP_GC_ENABLE_B 0x00000004
+#define TSNEP_GC_DISABLE 0x00000008
+#define TSNEP_GC_ENABLE_TIMEOUT 0x00000010
+#define TSNEP_GC_ACTIVE_A 0x00000002
+#define TSNEP_GC_ACTIVE_B 0x00000004
+#define TSNEP_GC_CHANGE_AB 0x00000008
+#define TSNEP_GC_TIMEOUT_ACTIVE 0x00000010
+#define TSNEP_GC_TIMEOUT_SIGNAL 0x00000020
+#define TSNEP_GC_LIST_ERROR 0x00000080
+#define TSNEP_GC_OPEN 0x00FF0000
+#define TSNEP_GC_OPEN_SHIFT 16
+#define TSNEP_GC_NEXT_OPEN 0xFF000000
+#define TSNEP_GC_NEXT_OPEN_SHIFT 24
+#define TSNEP_GC_TIMEOUT 131072
+#define TSNEP_GC_TIME 0x080C
+#define TSNEP_GC_CHANGE 0x0810
+#define TSNEP_GCL_A 0x2000
+#define TSNEP_GCL_B 0x2800
+#define TSNEP_GCL_SIZE SZ_2K
+#define TSNEP_RX_ASSIGN 0x0840
+#define TSNEP_RX_ASSIGN_ACTIVE 0x00000001
+#define TSNEP_RX_ASSIGN_QUEUE_MASK 0x00000006
+#define TSNEP_RX_ASSIGN_QUEUE_SHIFT 1
+#define TSNEP_RX_ASSIGN_OFFSET 1
+#define TSNEP_RX_ASSIGN_ETHER_TYPE 0x0880
+#define TSNEP_RX_ASSIGN_ETHER_TYPE_OFFSET 2
+#define TSNEP_RX_ASSIGN_ETHER_TYPE_COUNT 2
+
+/* tsnep gate control list operation */
+struct tsnep_gcl_operation {
+	u32 properties;
+	u32 interval;
+};
+
+#define TSNEP_GCL_COUNT (TSNEP_GCL_SIZE / sizeof(struct tsnep_gcl_operation))
+#define TSNEP_GCL_MASK 0x000000FF
+#define TSNEP_GCL_INSERT 0x20000000
+#define TSNEP_GCL_CHANGE 0x40000000
+#define TSNEP_GCL_LAST 0x80000000
+#define TSNEP_GCL_MIN_INTERVAL 32
+
+/* tsnep TX/RX descriptor */
+#define TSNEP_DESC_SIZE 256
+#define TSNEP_DESC_SIZE_DATA_AFTER 2048
+#define TSNEP_DESC_OFFSET 128
+#define TSNEP_DESC_SIZE_DATA_AFTER_INLINE (64 - sizeof(struct tsnep_tx_desc) + \
+					   sizeof_field(struct tsnep_tx_desc, tx))
+#define TSNEP_DESC_OWNER_COUNTER_MASK 0xC0000000
+#define TSNEP_DESC_OWNER_COUNTER_SHIFT 30
+#define TSNEP_DESC_LENGTH_MASK 0x00003FFF
+#define TSNEP_DESC_INTERRUPT_FLAG 0x00040000
+#define TSNEP_DESC_EXTENDED_WRITEBACK_FLAG 0x00080000
+#define TSNEP_DESC_NO_LINK_FLAG 0x01000000
+
+/* tsnep TX descriptor */
+struct tsnep_tx_desc {
+	__le32 properties;
+	__le32 more_properties;
+	__le32 reserved[2];
+	__le64 next;
+	__le64 tx;
+};
+
+#define TSNEP_TX_DESC_OWNER_MASK 0xE0000000
+#define TSNEP_TX_DESC_OWNER_USER_FLAG 0x20000000
+#define TSNEP_TX_DESC_LAST_FRAGMENT_FLAG 0x00010000
+#define TSNEP_TX_DESC_DATA_AFTER_DESC_FLAG 0x00020000
+
+/* tsnep TX descriptor writeback */
+struct tsnep_tx_desc_wb {
+	__le32 properties;
+	__le32 reserved1;
+	__le64 counter;
+	__le64 timestamp;
+	__le32 dma_delay;
+	__le32 reserved2;
+};
+
+#define TSNEP_TX_DESC_UNDERRUN_ERROR_FLAG 0x00010000
+#define TSNEP_TX_DESC_DMA_DELAY_FIRST_DATA_MASK 0x0000FFFC
+#define TSNEP_TX_DESC_DMA_DELAY_FIRST_DATA_SHIFT 2
+#define TSNEP_TX_DESC_DMA_DELAY_LAST_DATA_MASK 0xFFFC0000
+#define TSNEP_TX_DESC_DMA_DELAY_LAST_DATA_SHIFT 18
+#define TSNEP_TX_DESC_DMA_DELAY_NS 64
+
+/* tsnep RX descriptor */
+struct tsnep_rx_desc {
+	__le32 properties;
+	__le32 reserved[3];
+	__le64 next;
+	__le64 rx;
+};
+
+#define TSNEP_RX_DESC_BUFFER_SIZE_MASK 0x00003FFC
+
+/* tsnep RX descriptor writeback */
+struct tsnep_rx_desc_wb {
+	__le32 properties;
+	__le32 reserved[7];
+};
+
+/* tsnep RX inline meta */
+struct tsnep_rx_inline {
+	__le64 counter;
+	__le64 timestamp;
+};
+
+#define TSNEP_RX_INLINE_METADATA_SIZE (sizeof(struct tsnep_rx_inline))
+
+#endif /* _TSNEP_HW_H */
diff --git a/drivers/net/ethernet/engleder/tsnep_main.c b/drivers/net/ethernet/engleder/tsnep_main.c
new file mode 100644
index 000000000000..eba73246f986
--- /dev/null
+++ b/drivers/net/ethernet/engleder/tsnep_main.c
@@ -0,0 +1,2723 @@
+// SPDX-License-Identifier: GPL-2.0
+/* Copyright (C) 2021 Gerhard Engleder <gerhard@engleder-embedded.com> */
+
+/* TSN endpoint Ethernet MAC driver
+ *
+ * The TSN endpoint Ethernet MAC is a FPGA based network device for real-time
+ * communication. It is designed for endpoints within TSN (Time Sensitive
+ * Networking) networks; e.g., for PLCs in the industrial automation case.
+ *
+ * It supports multiple TX/RX queue pairs. The first TX/RX queue pair is used
+ * by the driver.
+ *
+ * More information can be found here:
+ * - www.embedded-experts.at/tsn
+ * - www.engleder-embedded.com
+ */
+
+#include "tsnep.h"
+#include "tsnep_hw.h"
+
+#include <linux/module.h>
+#include <linux/of.h>
+#include <linux/of_net.h>
+#include <linux/of_mdio.h>
+#include <linux/interrupt.h>
+#include <linux/etherdevice.h>
+#include <linux/phy.h>
+#include <linux/iopoll.h>
+#include <linux/bpf.h>
+#include <linux/bpf_trace.h>
+#include <net/page_pool/helpers.h>
+#include <net/xdp_sock_drv.h>
+
+#define TSNEP_RX_OFFSET (max(NET_SKB_PAD, XDP_PACKET_HEADROOM) + NET_IP_ALIGN)
+#define TSNEP_HEADROOM ALIGN(TSNEP_RX_OFFSET, 4)
+#define TSNEP_MAX_RX_BUF_SIZE (PAGE_SIZE - TSNEP_HEADROOM - \
+			       SKB_DATA_ALIGN(sizeof(struct skb_shared_info)))
+/* XSK buffer shall store at least Q-in-Q frame */
+#define TSNEP_XSK_RX_BUF_SIZE (ALIGN(TSNEP_RX_INLINE_METADATA_SIZE + \
+				     ETH_FRAME_LEN + ETH_FCS_LEN + \
+				     VLAN_HLEN * 2, 4))
+
+#ifdef CONFIG_ARCH_DMA_ADDR_T_64BIT
+#define DMA_ADDR_HIGH(dma_addr) ((u32)(((dma_addr) >> 32) & 0xFFFFFFFF))
+#else
+#define DMA_ADDR_HIGH(dma_addr) ((u32)(0))
+#endif
+#define DMA_ADDR_LOW(dma_addr) ((u32)((dma_addr) & 0xFFFFFFFF))
+
+#define TSNEP_COALESCE_USECS_DEFAULT 64
+#define TSNEP_COALESCE_USECS_MAX     ((ECM_INT_DELAY_MASK >> ECM_INT_DELAY_SHIFT) * \
+				      ECM_INT_DELAY_BASE_US + ECM_INT_DELAY_BASE_US - 1)
+
+/* mapping type */
+#define TSNEP_TX_TYPE_MAP		BIT(0)
+#define TSNEP_TX_TYPE_MAP_PAGE		BIT(1)
+#define TSNEP_TX_TYPE_INLINE		BIT(2)
+/* buffer type */
+#define TSNEP_TX_TYPE_SKB		BIT(8)
+#define TSNEP_TX_TYPE_SKB_MAP		(TSNEP_TX_TYPE_SKB | TSNEP_TX_TYPE_MAP)
+#define TSNEP_TX_TYPE_SKB_INLINE	(TSNEP_TX_TYPE_SKB | TSNEP_TX_TYPE_INLINE)
+#define TSNEP_TX_TYPE_SKB_FRAG		BIT(9)
+#define TSNEP_TX_TYPE_SKB_FRAG_MAP_PAGE	(TSNEP_TX_TYPE_SKB_FRAG | TSNEP_TX_TYPE_MAP_PAGE)
+#define TSNEP_TX_TYPE_SKB_FRAG_INLINE	(TSNEP_TX_TYPE_SKB_FRAG | TSNEP_TX_TYPE_INLINE)
+#define TSNEP_TX_TYPE_XDP_TX		BIT(10)
+#define TSNEP_TX_TYPE_XDP_NDO		BIT(11)
+#define TSNEP_TX_TYPE_XDP_NDO_MAP_PAGE	(TSNEP_TX_TYPE_XDP_NDO | TSNEP_TX_TYPE_MAP_PAGE)
+#define TSNEP_TX_TYPE_XDP		(TSNEP_TX_TYPE_XDP_TX | TSNEP_TX_TYPE_XDP_NDO)
+#define TSNEP_TX_TYPE_XSK		BIT(12)
+#define TSNEP_TX_TYPE_TSTAMP		BIT(13)
+#define TSNEP_TX_TYPE_SKB_TSTAMP	(TSNEP_TX_TYPE_SKB | TSNEP_TX_TYPE_TSTAMP)
+
+#define TSNEP_XDP_TX		BIT(0)
+#define TSNEP_XDP_REDIRECT	BIT(1)
+
+static void tsnep_enable_irq(struct tsnep_adapter *adapter, u32 mask)
+{
+	iowrite32(mask, adapter->addr + ECM_INT_ENABLE);
+}
+
+static void tsnep_disable_irq(struct tsnep_adapter *adapter, u32 mask)
+{
+	mask |= ECM_INT_DISABLE;
+	iowrite32(mask, adapter->addr + ECM_INT_ENABLE);
+}
+
+static irqreturn_t tsnep_irq(int irq, void *arg)
+{
+	struct tsnep_adapter *adapter = arg;
+	u32 active = ioread32(adapter->addr + ECM_INT_ACTIVE);
+
+	/* acknowledge interrupt */
+	if (active != 0)
+		iowrite32(active, adapter->addr + ECM_INT_ACKNOWLEDGE);
+
+	/* handle link interrupt */
+	if ((active & ECM_INT_LINK) != 0)
+		phy_mac_interrupt(adapter->netdev->phydev);
+
+	/* handle TX/RX queue 0 interrupt */
+	if ((active & adapter->queue[0].irq_mask) != 0) {
+		if (napi_schedule_prep(&adapter->queue[0].napi)) {
+			tsnep_disable_irq(adapter, adapter->queue[0].irq_mask);
+			/* schedule after masking to avoid races */
+			__napi_schedule(&adapter->queue[0].napi);
+		}
+	}
+
+	return IRQ_HANDLED;
+}
+
+static irqreturn_t tsnep_irq_txrx(int irq, void *arg)
+{
+	struct tsnep_queue *queue = arg;
+
+	/* handle TX/RX queue interrupt */
+	if (napi_schedule_prep(&queue->napi)) {
+		tsnep_disable_irq(queue->adapter, queue->irq_mask);
+		/* schedule after masking to avoid races */
+		__napi_schedule(&queue->napi);
+	}
+
+	return IRQ_HANDLED;
+}
+
+int tsnep_set_irq_coalesce(struct tsnep_queue *queue, u32 usecs)
+{
+	if (usecs > TSNEP_COALESCE_USECS_MAX)
+		return -ERANGE;
+
+	usecs /= ECM_INT_DELAY_BASE_US;
+	usecs <<= ECM_INT_DELAY_SHIFT;
+	usecs &= ECM_INT_DELAY_MASK;
+
+	queue->irq_delay &= ~ECM_INT_DELAY_MASK;
+	queue->irq_delay |= usecs;
+	iowrite8(queue->irq_delay, queue->irq_delay_addr);
+
+	return 0;
+}
+
+u32 tsnep_get_irq_coalesce(struct tsnep_queue *queue)
+{
+	u32 usecs;
+
+	usecs = (queue->irq_delay & ECM_INT_DELAY_MASK);
+	usecs >>= ECM_INT_DELAY_SHIFT;
+	usecs *= ECM_INT_DELAY_BASE_US;
+
+	return usecs;
+}
+
+static int tsnep_mdiobus_read(struct mii_bus *bus, int addr, int regnum)
+{
+	struct tsnep_adapter *adapter = bus->priv;
+	u32 md;
+	int retval;
+
+	md = ECM_MD_READ;
+	if (!adapter->suppress_preamble)
+		md |= ECM_MD_PREAMBLE;
+	md |= (regnum << ECM_MD_ADDR_SHIFT) & ECM_MD_ADDR_MASK;
+	md |= (addr << ECM_MD_PHY_ADDR_SHIFT) & ECM_MD_PHY_ADDR_MASK;
+	iowrite32(md, adapter->addr + ECM_MD_CONTROL);
+	retval = readl_poll_timeout_atomic(adapter->addr + ECM_MD_STATUS, md,
+					   !(md & ECM_MD_BUSY), 16, 1000);
+	if (retval != 0)
+		return retval;
+
+	return (md & ECM_MD_DATA_MASK) >> ECM_MD_DATA_SHIFT;
+}
+
+static int tsnep_mdiobus_write(struct mii_bus *bus, int addr, int regnum,
+			       u16 val)
+{
+	struct tsnep_adapter *adapter = bus->priv;
+	u32 md;
+	int retval;
+
+	md = ECM_MD_WRITE;
+	if (!adapter->suppress_preamble)
+		md |= ECM_MD_PREAMBLE;
+	md |= (regnum << ECM_MD_ADDR_SHIFT) & ECM_MD_ADDR_MASK;
+	md |= (addr << ECM_MD_PHY_ADDR_SHIFT) & ECM_MD_PHY_ADDR_MASK;
+	md |= ((u32)val << ECM_MD_DATA_SHIFT) & ECM_MD_DATA_MASK;
+	iowrite32(md, adapter->addr + ECM_MD_CONTROL);
+	retval = readl_poll_timeout_atomic(adapter->addr + ECM_MD_STATUS, md,
+					   !(md & ECM_MD_BUSY), 16, 1000);
+	if (retval != 0)
+		return retval;
+
+	return 0;
+}
+
+static void tsnep_set_link_mode(struct tsnep_adapter *adapter)
+{
+	u32 mode;
+
+	switch (adapter->phydev->speed) {
+	case SPEED_100:
+		mode = ECM_LINK_MODE_100;
+		break;
+	case SPEED_1000:
+		mode = ECM_LINK_MODE_1000;
+		break;
+	default:
+		mode = ECM_LINK_MODE_OFF;
+		break;
+	}
+	iowrite32(mode, adapter->addr + ECM_STATUS);
+}
+
+static void tsnep_phy_link_status_change(struct net_device *netdev)
+{
+	struct tsnep_adapter *adapter = netdev_priv(netdev);
+	struct phy_device *phydev = netdev->phydev;
+
+	if (phydev->link)
+		tsnep_set_link_mode(adapter);
+
+	phy_print_status(netdev->phydev);
+}
+
+static int tsnep_phy_loopback(struct tsnep_adapter *adapter, bool enable)
+{
+	int speed;
+
+	if (enable) {
+		if (adapter->phydev->autoneg == AUTONEG_DISABLE &&
+		    adapter->phydev->speed == SPEED_100)
+			speed = SPEED_100;
+		else
+			speed = SPEED_1000;
+	} else {
+		speed = 0;
+	}
+
+	return phy_loopback(adapter->phydev, enable, speed);
+}
+
+static int tsnep_phy_open(struct tsnep_adapter *adapter)
+{
+	struct phy_device *phydev;
+	struct ethtool_keee ethtool_keee;
+	int retval;
+
+	retval = phy_connect_direct(adapter->netdev, adapter->phydev,
+				    tsnep_phy_link_status_change,
+				    adapter->phy_mode);
+	if (retval)
+		return retval;
+	phydev = adapter->netdev->phydev;
+
+	/* MAC supports only 100Mbps|1000Mbps full duplex
+	 * SPE (Single Pair Ethernet) is also an option but not implemented yet
+	 */
+	phy_remove_link_mode(phydev, ETHTOOL_LINK_MODE_10baseT_Half_BIT);
+	phy_remove_link_mode(phydev, ETHTOOL_LINK_MODE_10baseT_Full_BIT);
+	phy_remove_link_mode(phydev, ETHTOOL_LINK_MODE_100baseT_Half_BIT);
+	phy_remove_link_mode(phydev, ETHTOOL_LINK_MODE_1000baseT_Half_BIT);
+
+	/* disable EEE autoneg, EEE not supported by TSNEP */
+	memset(&ethtool_keee, 0, sizeof(ethtool_keee));
+	phy_ethtool_set_eee(adapter->phydev, &ethtool_keee);
+
+	adapter->phydev->irq = PHY_MAC_INTERRUPT;
+	phy_start(adapter->phydev);
+
+	return 0;
+}
+
+static void tsnep_phy_close(struct tsnep_adapter *adapter)
+{
+	phy_stop(adapter->netdev->phydev);
+	phy_disconnect(adapter->netdev->phydev);
+}
+
+static void tsnep_tx_ring_cleanup(struct tsnep_tx *tx)
+{
+	struct device *dmadev = tx->adapter->dmadev;
+	int i;
+
+	memset(tx->entry, 0, sizeof(tx->entry));
+
+	for (i = 0; i < TSNEP_RING_PAGE_COUNT; i++) {
+		if (tx->page[i]) {
+			dma_free_coherent(dmadev, PAGE_SIZE, tx->page[i],
+					  tx->page_dma[i]);
+			tx->page[i] = NULL;
+			tx->page_dma[i] = 0;
+		}
+	}
+}
+
+static int tsnep_tx_ring_create(struct tsnep_tx *tx)
+{
+	struct device *dmadev = tx->adapter->dmadev;
+	struct tsnep_tx_entry *entry;
+	struct tsnep_tx_entry *next_entry;
+	int i, j;
+	int retval;
+
+	for (i = 0; i < TSNEP_RING_PAGE_COUNT; i++) {
+		tx->page[i] =
+			dma_alloc_coherent(dmadev, PAGE_SIZE, &tx->page_dma[i],
+					   GFP_KERNEL);
+		if (!tx->page[i]) {
+			retval = -ENOMEM;
+			goto alloc_failed;
+		}
+		for (j = 0; j < TSNEP_RING_ENTRIES_PER_PAGE; j++) {
+			entry = &tx->entry[TSNEP_RING_ENTRIES_PER_PAGE * i + j];
+			entry->desc_wb = (struct tsnep_tx_desc_wb *)
+				(((u8 *)tx->page[i]) + TSNEP_DESC_SIZE * j);
+			entry->desc = (struct tsnep_tx_desc *)
+				(((u8 *)entry->desc_wb) + TSNEP_DESC_OFFSET);
+			entry->desc_dma = tx->page_dma[i] + TSNEP_DESC_SIZE * j;
+			entry->owner_user_flag = false;
+		}
+	}
+	for (i = 0; i < TSNEP_RING_SIZE; i++) {
+		entry = &tx->entry[i];
+		next_entry = &tx->entry[(i + 1) & TSNEP_RING_MASK];
+		entry->desc->next = __cpu_to_le64(next_entry->desc_dma);
+	}
+
+	return 0;
+
+alloc_failed:
+	tsnep_tx_ring_cleanup(tx);
+	return retval;
+}
+
+static void tsnep_tx_init(struct tsnep_tx *tx)
+{
+	dma_addr_t dma;
+
+	dma = tx->entry[0].desc_dma | TSNEP_RESET_OWNER_COUNTER;
+	iowrite32(DMA_ADDR_LOW(dma), tx->addr + TSNEP_TX_DESC_ADDR_LOW);
+	iowrite32(DMA_ADDR_HIGH(dma), tx->addr + TSNEP_TX_DESC_ADDR_HIGH);
+	tx->write = 0;
+	tx->read = 0;
+	tx->owner_counter = 1;
+	tx->increment_owner_counter = TSNEP_RING_SIZE - 1;
+}
+
+static void tsnep_tx_enable(struct tsnep_tx *tx)
+{
+	struct netdev_queue *nq;
+
+	nq = netdev_get_tx_queue(tx->adapter->netdev, tx->queue_index);
+
+	__netif_tx_lock_bh(nq);
+	netif_tx_wake_queue(nq);
+	__netif_tx_unlock_bh(nq);
+}
+
+static void tsnep_tx_disable(struct tsnep_tx *tx, struct napi_struct *napi)
+{
+	struct netdev_queue *nq;
+	u32 val;
+
+	nq = netdev_get_tx_queue(tx->adapter->netdev, tx->queue_index);
+
+	__netif_tx_lock_bh(nq);
+	netif_tx_stop_queue(nq);
+	__netif_tx_unlock_bh(nq);
+
+	/* wait until TX is done in hardware */
+	readx_poll_timeout(ioread32, tx->addr + TSNEP_CONTROL, val,
+			   ((val & TSNEP_CONTROL_TX_ENABLE) == 0), 10000,
+			   1000000);
+
+	/* wait until TX is also done in software */
+	while (READ_ONCE(tx->read) != tx->write) {
+		napi_schedule(napi);
+		napi_synchronize(napi);
+	}
+}
+
+static void tsnep_tx_activate(struct tsnep_tx *tx, int index, int length,
+			      bool last)
+{
+	struct tsnep_tx_entry *entry = &tx->entry[index];
+
+	entry->properties = 0;
+	/* xdpf and zc are union with skb */
+	if (entry->skb) {
+		entry->properties = length & TSNEP_DESC_LENGTH_MASK;
+		entry->properties |= TSNEP_DESC_INTERRUPT_FLAG;
+		if ((entry->type & TSNEP_TX_TYPE_SKB_TSTAMP) == TSNEP_TX_TYPE_SKB_TSTAMP)
+			entry->properties |= TSNEP_DESC_EXTENDED_WRITEBACK_FLAG;
+
+		/* toggle user flag to prevent false acknowledge
+		 *
+		 * Only the first fragment is acknowledged. For all other
+		 * fragments no acknowledge is done and the last written owner
+		 * counter stays in the writeback descriptor. Therefore, it is
+		 * possible that the last written owner counter is identical to
+		 * the new incremented owner counter and a false acknowledge is
+		 * detected before the real acknowledge has been done by
+		 * hardware.
+		 *
+		 * The user flag is used to prevent this situation. The user
+		 * flag is copied to the writeback descriptor by the hardware
+		 * and is used as additional acknowledge data. By toggeling the
+		 * user flag only for the first fragment (which is
+		 * acknowledged), it is guaranteed that the last acknowledge
+		 * done for this descriptor has used a different user flag and
+		 * cannot be detected as false acknowledge.
+		 */
+		entry->owner_user_flag = !entry->owner_user_flag;
+	}
+	if (last)
+		entry->properties |= TSNEP_TX_DESC_LAST_FRAGMENT_FLAG;
+	if (index == tx->increment_owner_counter) {
+		tx->owner_counter++;
+		if (tx->owner_counter == 4)
+			tx->owner_counter = 1;
+		tx->increment_owner_counter--;
+		if (tx->increment_owner_counter < 0)
+			tx->increment_owner_counter = TSNEP_RING_SIZE - 1;
+	}
+	entry->properties |=
+		(tx->owner_counter << TSNEP_DESC_OWNER_COUNTER_SHIFT) &
+		TSNEP_DESC_OWNER_COUNTER_MASK;
+	if (entry->owner_user_flag)
+		entry->properties |= TSNEP_TX_DESC_OWNER_USER_FLAG;
+	entry->desc->more_properties =
+		__cpu_to_le32(entry->len & TSNEP_DESC_LENGTH_MASK);
+	if (entry->type & TSNEP_TX_TYPE_INLINE)
+		entry->properties |= TSNEP_TX_DESC_DATA_AFTER_DESC_FLAG;
+
+	/* descriptor properties shall be written last, because valid data is
+	 * signaled there
+	 */
+	dma_wmb();
+
+	entry->desc->properties = __cpu_to_le32(entry->properties);
+}
+
+static int tsnep_tx_desc_available(struct tsnep_tx *tx)
+{
+	if (tx->read <= tx->write)
+		return TSNEP_RING_SIZE - tx->write + tx->read - 1;
+	else
+		return tx->read - tx->write - 1;
+}
+
+static int tsnep_tx_map_frag(skb_frag_t *frag, struct tsnep_tx_entry *entry,
+			     struct device *dmadev, dma_addr_t *dma)
+{
+	unsigned int len;
+	int mapped;
+
+	len = skb_frag_size(frag);
+	if (likely(len > TSNEP_DESC_SIZE_DATA_AFTER_INLINE)) {
+		*dma = skb_frag_dma_map(dmadev, frag, 0, len, DMA_TO_DEVICE);
+		if (dma_mapping_error(dmadev, *dma))
+			return -ENOMEM;
+		entry->type = TSNEP_TX_TYPE_SKB_FRAG_MAP_PAGE;
+		mapped = 1;
+	} else {
+		void *fragdata = skb_frag_address_safe(frag);
+
+		if (likely(fragdata)) {
+			memcpy(&entry->desc->tx, fragdata, len);
+		} else {
+			struct page *page = skb_frag_page(frag);
+
+			fragdata = kmap_local_page(page);
+			memcpy(&entry->desc->tx, fragdata + skb_frag_off(frag),
+			       len);
+			kunmap_local(fragdata);
+		}
+		entry->type = TSNEP_TX_TYPE_SKB_FRAG_INLINE;
+		mapped = 0;
+	}
+
+	return mapped;
+}
+
+static int tsnep_tx_map(struct sk_buff *skb, struct tsnep_tx *tx, int count,
+			bool do_tstamp)
+{
+	struct device *dmadev = tx->adapter->dmadev;
+	struct tsnep_tx_entry *entry;
+	unsigned int len;
+	int map_len = 0;
+	dma_addr_t dma;
+	int i, mapped;
+
+	for (i = 0; i < count; i++) {
+		entry = &tx->entry[(tx->write + i) & TSNEP_RING_MASK];
+
+		if (!i) {
+			len = skb_headlen(skb);
+			if (likely(len > TSNEP_DESC_SIZE_DATA_AFTER_INLINE)) {
+				dma = dma_map_single(dmadev, skb->data, len,
+						     DMA_TO_DEVICE);
+				if (dma_mapping_error(dmadev, dma))
+					return -ENOMEM;
+				entry->type = TSNEP_TX_TYPE_SKB_MAP;
+				mapped = 1;
+			} else {
+				memcpy(&entry->desc->tx, skb->data, len);
+				entry->type = TSNEP_TX_TYPE_SKB_INLINE;
+				mapped = 0;
+			}
+
+			if (do_tstamp)
+				entry->type |= TSNEP_TX_TYPE_TSTAMP;
+		} else {
+			skb_frag_t *frag = &skb_shinfo(skb)->frags[i - 1];
+
+			len = skb_frag_size(frag);
+			mapped = tsnep_tx_map_frag(frag, entry, dmadev, &dma);
+			if (mapped < 0)
+				return mapped;
+		}
+
+		entry->len = len;
+		if (likely(mapped)) {
+			dma_unmap_addr_set(entry, dma, dma);
+			entry->desc->tx = __cpu_to_le64(dma);
+		}
+
+		map_len += len;
+	}
+
+	return map_len;
+}
+
+static int tsnep_tx_unmap(struct tsnep_tx *tx, int index, int count)
+{
+	struct device *dmadev = tx->adapter->dmadev;
+	struct tsnep_tx_entry *entry;
+	int map_len = 0;
+	int i;
+
+	for (i = 0; i < count; i++) {
+		entry = &tx->entry[(index + i) & TSNEP_RING_MASK];
+
+		if (entry->len) {
+			if (entry->type & TSNEP_TX_TYPE_MAP)
+				dma_unmap_single(dmadev,
+						 dma_unmap_addr(entry, dma),
+						 dma_unmap_len(entry, len),
+						 DMA_TO_DEVICE);
+			else if (entry->type & TSNEP_TX_TYPE_MAP_PAGE)
+				dma_unmap_page(dmadev,
+					       dma_unmap_addr(entry, dma),
+					       dma_unmap_len(entry, len),
+					       DMA_TO_DEVICE);
+			map_len += entry->len;
+			entry->len = 0;
+		}
+	}
+
+	return map_len;
+}
+
+static netdev_tx_t tsnep_xmit_frame_ring(struct sk_buff *skb,
+					 struct tsnep_tx *tx)
+{
+	struct tsnep_tx_entry *entry;
+	bool do_tstamp = false;
+	int count = 1;
+	int length;
+	int retval;
+	int i;
+
+	if (skb_shinfo(skb)->nr_frags > 0)
+		count += skb_shinfo(skb)->nr_frags;
+
+	if (tsnep_tx_desc_available(tx) < count) {
+		/* ring full, shall not happen because queue is stopped if full
+		 * below
+		 */
+		netif_stop_subqueue(tx->adapter->netdev, tx->queue_index);
+
+		return NETDEV_TX_BUSY;
+	}
+
+	entry = &tx->entry[tx->write];
+	entry->skb = skb;
+
+	if (unlikely(skb_shinfo(skb)->tx_flags & SKBTX_HW_TSTAMP) &&
+	    tx->adapter->hwtstamp_config.tx_type == HWTSTAMP_TX_ON) {
+		skb_shinfo(skb)->tx_flags |= SKBTX_IN_PROGRESS;
+		do_tstamp = true;
+	}
+
+	retval = tsnep_tx_map(skb, tx, count, do_tstamp);
+	if (retval < 0) {
+		tsnep_tx_unmap(tx, tx->write, count);
+		dev_kfree_skb_any(entry->skb);
+		entry->skb = NULL;
+
+		tx->dropped++;
+
+		return NETDEV_TX_OK;
+	}
+	length = retval;
+
+	for (i = 0; i < count; i++)
+		tsnep_tx_activate(tx, (tx->write + i) & TSNEP_RING_MASK, length,
+				  i == count - 1);
+	tx->write = (tx->write + count) & TSNEP_RING_MASK;
+
+	skb_tx_timestamp(skb);
+
+	/* descriptor properties shall be valid before hardware is notified */
+	dma_wmb();
+
+	iowrite32(TSNEP_CONTROL_TX_ENABLE, tx->addr + TSNEP_CONTROL);
+
+	if (tsnep_tx_desc_available(tx) < (MAX_SKB_FRAGS + 1)) {
+		/* ring can get full with next frame */
+		netif_stop_subqueue(tx->adapter->netdev, tx->queue_index);
+	}
+
+	return NETDEV_TX_OK;
+}
+
+static int tsnep_xdp_tx_map(struct xdp_frame *xdpf, struct tsnep_tx *tx,
+			    struct skb_shared_info *shinfo, int count, u32 type)
+{
+	struct device *dmadev = tx->adapter->dmadev;
+	struct tsnep_tx_entry *entry;
+	struct page *page;
+	skb_frag_t *frag;
+	unsigned int len;
+	int map_len = 0;
+	dma_addr_t dma;
+	void *data;
+	int i;
+
+	frag = NULL;
+	len = xdpf->len;
+	for (i = 0; i < count; i++) {
+		entry = &tx->entry[(tx->write + i) & TSNEP_RING_MASK];
+		if (type & TSNEP_TX_TYPE_XDP_NDO) {
+			data = unlikely(frag) ? skb_frag_address(frag) :
+						xdpf->data;
+			dma = dma_map_single(dmadev, data, len, DMA_TO_DEVICE);
+			if (dma_mapping_error(dmadev, dma))
+				return -ENOMEM;
+
+			entry->type = TSNEP_TX_TYPE_XDP_NDO_MAP_PAGE;
+		} else {
+			page = unlikely(frag) ? skb_frag_page(frag) :
+						virt_to_page(xdpf->data);
+			dma = page_pool_get_dma_addr(page);
+			if (unlikely(frag))
+				dma += skb_frag_off(frag);
+			else
+				dma += sizeof(*xdpf) + xdpf->headroom;
+			dma_sync_single_for_device(dmadev, dma, len,
+						   DMA_BIDIRECTIONAL);
+
+			entry->type = TSNEP_TX_TYPE_XDP_TX;
+		}
+
+		entry->len = len;
+		dma_unmap_addr_set(entry, dma, dma);
+
+		entry->desc->tx = __cpu_to_le64(dma);
+
+		map_len += len;
+
+		if (i + 1 < count) {
+			frag = &shinfo->frags[i];
+			len = skb_frag_size(frag);
+		}
+	}
+
+	return map_len;
+}
+
+/* This function requires __netif_tx_lock is held by the caller. */
+static bool tsnep_xdp_xmit_frame_ring(struct xdp_frame *xdpf,
+				      struct tsnep_tx *tx, u32 type)
+{
+	struct skb_shared_info *shinfo = xdp_get_shared_info_from_frame(xdpf);
+	struct tsnep_tx_entry *entry;
+	int count, length, retval, i;
+
+	count = 1;
+	if (unlikely(xdp_frame_has_frags(xdpf)))
+		count += shinfo->nr_frags;
+
+	/* ensure that TX ring is not filled up by XDP, always MAX_SKB_FRAGS
+	 * will be available for normal TX path and queue is stopped there if
+	 * necessary
+	 */
+	if (tsnep_tx_desc_available(tx) < (MAX_SKB_FRAGS + 1 + count))
+		return false;
+
+	entry = &tx->entry[tx->write];
+	entry->xdpf = xdpf;
+
+	retval = tsnep_xdp_tx_map(xdpf, tx, shinfo, count, type);
+	if (retval < 0) {
+		tsnep_tx_unmap(tx, tx->write, count);
+		entry->xdpf = NULL;
+
+		tx->dropped++;
+
+		return false;
+	}
+	length = retval;
+
+	for (i = 0; i < count; i++)
+		tsnep_tx_activate(tx, (tx->write + i) & TSNEP_RING_MASK, length,
+				  i == count - 1);
+	tx->write = (tx->write + count) & TSNEP_RING_MASK;
+
+	/* descriptor properties shall be valid before hardware is notified */
+	dma_wmb();
+
+	return true;
+}
+
+static void tsnep_xdp_xmit_flush(struct tsnep_tx *tx)
+{
+	iowrite32(TSNEP_CONTROL_TX_ENABLE, tx->addr + TSNEP_CONTROL);
+}
+
+static bool tsnep_xdp_xmit_back(struct tsnep_adapter *adapter,
+				struct xdp_buff *xdp,
+				struct netdev_queue *tx_nq, struct tsnep_tx *tx,
+				bool zc)
+{
+	struct xdp_frame *xdpf = xdp_convert_buff_to_frame(xdp);
+	bool xmit;
+	u32 type;
+
+	if (unlikely(!xdpf))
+		return false;
+
+	/* no page pool for zero copy */
+	if (zc)
+		type = TSNEP_TX_TYPE_XDP_NDO;
+	else
+		type = TSNEP_TX_TYPE_XDP_TX;
+
+	__netif_tx_lock(tx_nq, smp_processor_id());
+
+	xmit = tsnep_xdp_xmit_frame_ring(xdpf, tx, type);
+
+	/* Avoid transmit queue timeout since we share it with the slow path */
+	if (xmit)
+		txq_trans_cond_update(tx_nq);
+
+	__netif_tx_unlock(tx_nq);
+
+	return xmit;
+}
+
+static int tsnep_xdp_tx_map_zc(struct xdp_desc *xdpd, struct tsnep_tx *tx)
+{
+	struct tsnep_tx_entry *entry;
+	dma_addr_t dma;
+
+	entry = &tx->entry[tx->write];
+	entry->zc = true;
+
+	dma = xsk_buff_raw_get_dma(tx->xsk_pool, xdpd->addr);
+	xsk_buff_raw_dma_sync_for_device(tx->xsk_pool, dma, xdpd->len);
+
+	entry->type = TSNEP_TX_TYPE_XSK;
+	entry->len = xdpd->len;
+
+	entry->desc->tx = __cpu_to_le64(dma);
+
+	return xdpd->len;
+}
+
+static void tsnep_xdp_xmit_frame_ring_zc(struct xdp_desc *xdpd,
+					 struct tsnep_tx *tx)
+{
+	int length;
+
+	length = tsnep_xdp_tx_map_zc(xdpd, tx);
+
+	tsnep_tx_activate(tx, tx->write, length, true);
+	tx->write = (tx->write + 1) & TSNEP_RING_MASK;
+}
+
+static void tsnep_xdp_xmit_zc(struct tsnep_tx *tx)
+{
+	int desc_available = tsnep_tx_desc_available(tx);
+	struct xdp_desc *descs = tx->xsk_pool->tx_descs;
+	int batch, i;
+
+	/* ensure that TX ring is not filled up by XDP, always MAX_SKB_FRAGS
+	 * will be available for normal TX path and queue is stopped there if
+	 * necessary
+	 */
+	if (desc_available <= (MAX_SKB_FRAGS + 1))
+		return;
+	desc_available -= MAX_SKB_FRAGS + 1;
+
+	batch = xsk_tx_peek_release_desc_batch(tx->xsk_pool, desc_available);
+	for (i = 0; i < batch; i++)
+		tsnep_xdp_xmit_frame_ring_zc(&descs[i], tx);
+
+	if (batch) {
+		/* descriptor properties shall be valid before hardware is
+		 * notified
+		 */
+		dma_wmb();
+
+		tsnep_xdp_xmit_flush(tx);
+	}
+}
+
+static bool tsnep_tx_poll(struct tsnep_tx *tx, int napi_budget)
+{
+	struct tsnep_tx_entry *entry;
+	struct netdev_queue *nq;
+	int xsk_frames = 0;
+	int budget = 128;
+	int length;
+	int count;
+
+	nq = netdev_get_tx_queue(tx->adapter->netdev, tx->queue_index);
+	__netif_tx_lock(nq, smp_processor_id());
+
+	do {
+		if (tx->read == tx->write)
+			break;
+
+		entry = &tx->entry[tx->read];
+		if ((__le32_to_cpu(entry->desc_wb->properties) &
+		     TSNEP_TX_DESC_OWNER_MASK) !=
+		    (entry->properties & TSNEP_TX_DESC_OWNER_MASK))
+			break;
+
+		/* descriptor properties shall be read first, because valid data
+		 * is signaled there
+		 */
+		dma_rmb();
+
+		count = 1;
+		if ((entry->type & TSNEP_TX_TYPE_SKB) &&
+		    skb_shinfo(entry->skb)->nr_frags > 0)
+			count += skb_shinfo(entry->skb)->nr_frags;
+		else if ((entry->type & TSNEP_TX_TYPE_XDP) &&
+			 xdp_frame_has_frags(entry->xdpf))
+			count += xdp_get_shared_info_from_frame(entry->xdpf)->nr_frags;
+
+		length = tsnep_tx_unmap(tx, tx->read, count);
+
+		if (((entry->type & TSNEP_TX_TYPE_SKB_TSTAMP) == TSNEP_TX_TYPE_SKB_TSTAMP) &&
+		    (__le32_to_cpu(entry->desc_wb->properties) &
+		     TSNEP_DESC_EXTENDED_WRITEBACK_FLAG)) {
+			struct skb_shared_hwtstamps hwtstamps;
+			u64 timestamp;
+
+			if (entry->skb->sk &&
+			    READ_ONCE(entry->skb->sk->sk_tsflags) & SOF_TIMESTAMPING_BIND_PHC)
+				timestamp =
+					__le64_to_cpu(entry->desc_wb->counter);
+			else
+				timestamp =
+					__le64_to_cpu(entry->desc_wb->timestamp);
+
+			memset(&hwtstamps, 0, sizeof(hwtstamps));
+			hwtstamps.hwtstamp = ns_to_ktime(timestamp);
+
+			skb_tstamp_tx(entry->skb, &hwtstamps);
+		}
+
+		if (entry->type & TSNEP_TX_TYPE_SKB)
+			napi_consume_skb(entry->skb, napi_budget);
+		else if (entry->type & TSNEP_TX_TYPE_XDP)
+			xdp_return_frame_rx_napi(entry->xdpf);
+		else
+			xsk_frames++;
+		/* xdpf and zc are union with skb */
+		entry->skb = NULL;
+
+		tx->read = (tx->read + count) & TSNEP_RING_MASK;
+
+		tx->packets++;
+		tx->bytes += length + ETH_FCS_LEN;
+
+		budget--;
+	} while (likely(budget));
+
+	if (tx->xsk_pool) {
+		if (xsk_frames)
+			xsk_tx_completed(tx->xsk_pool, xsk_frames);
+		if (xsk_uses_need_wakeup(tx->xsk_pool))
+			xsk_set_tx_need_wakeup(tx->xsk_pool);
+		tsnep_xdp_xmit_zc(tx);
+	}
+
+	if ((tsnep_tx_desc_available(tx) >= ((MAX_SKB_FRAGS + 1) * 2)) &&
+	    netif_tx_queue_stopped(nq)) {
+		netif_tx_wake_queue(nq);
+	}
+
+	__netif_tx_unlock(nq);
+
+	return budget != 0;
+}
+
+static bool tsnep_tx_pending(struct tsnep_tx *tx)
+{
+	struct tsnep_tx_entry *entry;
+	struct netdev_queue *nq;
+	bool pending = false;
+
+	nq = netdev_get_tx_queue(tx->adapter->netdev, tx->queue_index);
+	__netif_tx_lock(nq, smp_processor_id());
+
+	if (tx->read != tx->write) {
+		entry = &tx->entry[tx->read];
+		if ((__le32_to_cpu(entry->desc_wb->properties) &
+		     TSNEP_TX_DESC_OWNER_MASK) ==
+		    (entry->properties & TSNEP_TX_DESC_OWNER_MASK))
+			pending = true;
+	}
+
+	__netif_tx_unlock(nq);
+
+	return pending;
+}
+
+static int tsnep_tx_open(struct tsnep_tx *tx)
+{
+	int retval;
+
+	retval = tsnep_tx_ring_create(tx);
+	if (retval)
+		return retval;
+
+	tsnep_tx_init(tx);
+
+	return 0;
+}
+
+static void tsnep_tx_close(struct tsnep_tx *tx)
+{
+	tsnep_tx_ring_cleanup(tx);
+}
+
+static void tsnep_rx_ring_cleanup(struct tsnep_rx *rx)
+{
+	struct device *dmadev = rx->adapter->dmadev;
+	struct tsnep_rx_entry *entry;
+	int i;
+
+	for (i = 0; i < TSNEP_RING_SIZE; i++) {
+		entry = &rx->entry[i];
+		if (!rx->xsk_pool && entry->page)
+			page_pool_put_full_page(rx->page_pool, entry->page,
+						false);
+		if (rx->xsk_pool && entry->xdp)
+			xsk_buff_free(entry->xdp);
+		/* xdp is union with page */
+		entry->page = NULL;
+	}
+
+	if (rx->page_pool)
+		page_pool_destroy(rx->page_pool);
+
+	memset(rx->entry, 0, sizeof(rx->entry));
+
+	for (i = 0; i < TSNEP_RING_PAGE_COUNT; i++) {
+		if (rx->page[i]) {
+			dma_free_coherent(dmadev, PAGE_SIZE, rx->page[i],
+					  rx->page_dma[i]);
+			rx->page[i] = NULL;
+			rx->page_dma[i] = 0;
+		}
+	}
+}
+
+static int tsnep_rx_ring_create(struct tsnep_rx *rx)
+{
+	struct device *dmadev = rx->adapter->dmadev;
+	struct tsnep_rx_entry *entry;
+	struct page_pool_params pp_params = { 0 };
+	struct tsnep_rx_entry *next_entry;
+	int i, j;
+	int retval;
+
+	for (i = 0; i < TSNEP_RING_PAGE_COUNT; i++) {
+		rx->page[i] =
+			dma_alloc_coherent(dmadev, PAGE_SIZE, &rx->page_dma[i],
+					   GFP_KERNEL);
+		if (!rx->page[i]) {
+			retval = -ENOMEM;
+			goto failed;
+		}
+		for (j = 0; j < TSNEP_RING_ENTRIES_PER_PAGE; j++) {
+			entry = &rx->entry[TSNEP_RING_ENTRIES_PER_PAGE * i + j];
+			entry->desc_wb = (struct tsnep_rx_desc_wb *)
+				(((u8 *)rx->page[i]) + TSNEP_DESC_SIZE * j);
+			entry->desc = (struct tsnep_rx_desc *)
+				(((u8 *)entry->desc_wb) + TSNEP_DESC_OFFSET);
+			entry->desc_dma = rx->page_dma[i] + TSNEP_DESC_SIZE * j;
+		}
+	}
+
+	pp_params.flags = PP_FLAG_DMA_MAP | PP_FLAG_DMA_SYNC_DEV;
+	pp_params.order = 0;
+	pp_params.pool_size = TSNEP_RING_SIZE;
+	pp_params.nid = dev_to_node(dmadev);
+	pp_params.dev = dmadev;
+	pp_params.dma_dir = DMA_BIDIRECTIONAL;
+	pp_params.max_len = TSNEP_MAX_RX_BUF_SIZE;
+	pp_params.offset = TSNEP_RX_OFFSET;
+	rx->page_pool = page_pool_create(&pp_params);
+	if (IS_ERR(rx->page_pool)) {
+		retval = PTR_ERR(rx->page_pool);
+		rx->page_pool = NULL;
+		goto failed;
+	}
+
+	for (i = 0; i < TSNEP_RING_SIZE; i++) {
+		entry = &rx->entry[i];
+		next_entry = &rx->entry[(i + 1) & TSNEP_RING_MASK];
+		entry->desc->next = __cpu_to_le64(next_entry->desc_dma);
+	}
+
+	return 0;
+
+failed:
+	tsnep_rx_ring_cleanup(rx);
+	return retval;
+}
+
+static void tsnep_rx_init(struct tsnep_rx *rx)
+{
+	dma_addr_t dma;
+
+	dma = rx->entry[0].desc_dma | TSNEP_RESET_OWNER_COUNTER;
+	iowrite32(DMA_ADDR_LOW(dma), rx->addr + TSNEP_RX_DESC_ADDR_LOW);
+	iowrite32(DMA_ADDR_HIGH(dma), rx->addr + TSNEP_RX_DESC_ADDR_HIGH);
+	rx->write = 0;
+	rx->read = 0;
+	rx->owner_counter = 1;
+	rx->increment_owner_counter = TSNEP_RING_SIZE - 1;
+}
+
+static void tsnep_rx_enable(struct tsnep_rx *rx)
+{
+	/* descriptor properties shall be valid before hardware is notified */
+	dma_wmb();
+
+	iowrite32(TSNEP_CONTROL_RX_ENABLE, rx->addr + TSNEP_CONTROL);
+}
+
+static void tsnep_rx_disable(struct tsnep_rx *rx)
+{
+	u32 val;
+
+	iowrite32(TSNEP_CONTROL_RX_DISABLE, rx->addr + TSNEP_CONTROL);
+	readx_poll_timeout(ioread32, rx->addr + TSNEP_CONTROL, val,
+			   ((val & TSNEP_CONTROL_RX_ENABLE) == 0), 10000,
+			   1000000);
+}
+
+static int tsnep_rx_desc_available(struct tsnep_rx *rx)
+{
+	if (rx->read <= rx->write)
+		return TSNEP_RING_SIZE - rx->write + rx->read - 1;
+	else
+		return rx->read - rx->write - 1;
+}
+
+static void tsnep_rx_free_page_buffer(struct tsnep_rx *rx)
+{
+	struct page **page;
+
+	/* last entry of page_buffer is always zero, because ring cannot be
+	 * filled completely
+	 */
+	page = rx->page_buffer;
+	while (*page) {
+		page_pool_put_full_page(rx->page_pool, *page, false);
+		*page = NULL;
+		page++;
+	}
+}
+
+static int tsnep_rx_alloc_page_buffer(struct tsnep_rx *rx)
+{
+	int i;
+
+	/* alloc for all ring entries except the last one, because ring cannot
+	 * be filled completely
+	 */
+	for (i = 0; i < TSNEP_RING_SIZE - 1; i++) {
+		rx->page_buffer[i] = page_pool_dev_alloc_pages(rx->page_pool);
+		if (!rx->page_buffer[i]) {
+			tsnep_rx_free_page_buffer(rx);
+
+			return -ENOMEM;
+		}
+	}
+
+	return 0;
+}
+
+static void tsnep_rx_set_page(struct tsnep_rx *rx, struct tsnep_rx_entry *entry,
+			      struct page *page)
+{
+	entry->page = page;
+	entry->len = TSNEP_MAX_RX_BUF_SIZE;
+	entry->dma = page_pool_get_dma_addr(entry->page);
+	entry->desc->rx = __cpu_to_le64(entry->dma + TSNEP_RX_OFFSET);
+}
+
+static int tsnep_rx_alloc_buffer(struct tsnep_rx *rx, int index)
+{
+	struct tsnep_rx_entry *entry = &rx->entry[index];
+	struct page *page;
+
+	page = page_pool_dev_alloc_pages(rx->page_pool);
+	if (unlikely(!page))
+		return -ENOMEM;
+	tsnep_rx_set_page(rx, entry, page);
+
+	return 0;
+}
+
+static void tsnep_rx_reuse_buffer(struct tsnep_rx *rx, int index)
+{
+	struct tsnep_rx_entry *entry = &rx->entry[index];
+	struct tsnep_rx_entry *read = &rx->entry[rx->read];
+
+	tsnep_rx_set_page(rx, entry, read->page);
+	read->page = NULL;
+}
+
+static void tsnep_rx_activate(struct tsnep_rx *rx, int index)
+{
+	struct tsnep_rx_entry *entry = &rx->entry[index];
+
+	/* TSNEP_MAX_RX_BUF_SIZE and TSNEP_XSK_RX_BUF_SIZE are multiple of 4 */
+	entry->properties = entry->len & TSNEP_DESC_LENGTH_MASK;
+	entry->properties |= TSNEP_DESC_INTERRUPT_FLAG;
+	if (index == rx->increment_owner_counter) {
+		rx->owner_counter++;
+		if (rx->owner_counter == 4)
+			rx->owner_counter = 1;
+		rx->increment_owner_counter--;
+		if (rx->increment_owner_counter < 0)
+			rx->increment_owner_counter = TSNEP_RING_SIZE - 1;
+	}
+	entry->properties |=
+		(rx->owner_counter << TSNEP_DESC_OWNER_COUNTER_SHIFT) &
+		TSNEP_DESC_OWNER_COUNTER_MASK;
+
+	/* descriptor properties shall be written last, because valid data is
+	 * signaled there
+	 */
+	dma_wmb();
+
+	entry->desc->properties = __cpu_to_le32(entry->properties);
+}
+
+static int tsnep_rx_alloc(struct tsnep_rx *rx, int count, bool reuse)
+{
+	bool alloc_failed = false;
+	int i, index;
+
+	for (i = 0; i < count && !alloc_failed; i++) {
+		index = (rx->write + i) & TSNEP_RING_MASK;
+
+		if (unlikely(tsnep_rx_alloc_buffer(rx, index))) {
+			rx->alloc_failed++;
+			alloc_failed = true;
+
+			/* reuse only if no other allocation was successful */
+			if (i == 0 && reuse)
+				tsnep_rx_reuse_buffer(rx, index);
+			else
+				break;
+		}
+
+		tsnep_rx_activate(rx, index);
+	}
+
+	if (i)
+		rx->write = (rx->write + i) & TSNEP_RING_MASK;
+
+	return i;
+}
+
+static int tsnep_rx_refill(struct tsnep_rx *rx, int count, bool reuse)
+{
+	int desc_refilled;
+
+	desc_refilled = tsnep_rx_alloc(rx, count, reuse);
+	if (desc_refilled)
+		tsnep_rx_enable(rx);
+
+	return desc_refilled;
+}
+
+static void tsnep_rx_set_xdp(struct tsnep_rx *rx, struct tsnep_rx_entry *entry,
+			     struct xdp_buff *xdp)
+{
+	entry->xdp = xdp;
+	entry->len = TSNEP_XSK_RX_BUF_SIZE;
+	entry->dma = xsk_buff_xdp_get_dma(entry->xdp);
+	entry->desc->rx = __cpu_to_le64(entry->dma);
+}
+
+static void tsnep_rx_reuse_buffer_zc(struct tsnep_rx *rx, int index)
+{
+	struct tsnep_rx_entry *entry = &rx->entry[index];
+	struct tsnep_rx_entry *read = &rx->entry[rx->read];
+
+	tsnep_rx_set_xdp(rx, entry, read->xdp);
+	read->xdp = NULL;
+}
+
+static int tsnep_rx_alloc_zc(struct tsnep_rx *rx, int count, bool reuse)
+{
+	u32 allocated;
+	int i;
+
+	allocated = xsk_buff_alloc_batch(rx->xsk_pool, rx->xdp_batch, count);
+	for (i = 0; i < allocated; i++) {
+		int index = (rx->write + i) & TSNEP_RING_MASK;
+		struct tsnep_rx_entry *entry = &rx->entry[index];
+
+		tsnep_rx_set_xdp(rx, entry, rx->xdp_batch[i]);
+		tsnep_rx_activate(rx, index);
+	}
+	if (i == 0) {
+		rx->alloc_failed++;
+
+		if (reuse) {
+			tsnep_rx_reuse_buffer_zc(rx, rx->write);
+			tsnep_rx_activate(rx, rx->write);
+		}
+	}
+
+	if (i)
+		rx->write = (rx->write + i) & TSNEP_RING_MASK;
+
+	return i;
+}
+
+static void tsnep_rx_free_zc(struct tsnep_rx *rx)
+{
+	int i;
+
+	for (i = 0; i < TSNEP_RING_SIZE; i++) {
+		struct tsnep_rx_entry *entry = &rx->entry[i];
+
+		if (entry->xdp)
+			xsk_buff_free(entry->xdp);
+		entry->xdp = NULL;
+	}
+}
+
+static int tsnep_rx_refill_zc(struct tsnep_rx *rx, int count, bool reuse)
+{
+	int desc_refilled;
+
+	desc_refilled = tsnep_rx_alloc_zc(rx, count, reuse);
+	if (desc_refilled)
+		tsnep_rx_enable(rx);
+
+	return desc_refilled;
+}
+
+static void tsnep_xsk_rx_need_wakeup(struct tsnep_rx *rx, int desc_available)
+{
+	if (desc_available)
+		xsk_set_rx_need_wakeup(rx->xsk_pool);
+	else
+		xsk_clear_rx_need_wakeup(rx->xsk_pool);
+}
+
+static bool tsnep_xdp_run_prog(struct tsnep_rx *rx, struct bpf_prog *prog,
+			       struct xdp_buff *xdp, int *status,
+			       struct netdev_queue *tx_nq, struct tsnep_tx *tx)
+{
+	unsigned int length;
+	unsigned int sync;
+	u32 act;
+
+	length = xdp->data_end - xdp->data_hard_start - XDP_PACKET_HEADROOM;
+
+	act = bpf_prog_run_xdp(prog, xdp);
+	switch (act) {
+	case XDP_PASS:
+		return false;
+	case XDP_TX:
+		if (!tsnep_xdp_xmit_back(rx->adapter, xdp, tx_nq, tx, false))
+			goto out_failure;
+		*status |= TSNEP_XDP_TX;
+		return true;
+	case XDP_REDIRECT:
+		if (xdp_do_redirect(rx->adapter->netdev, xdp, prog) < 0)
+			goto out_failure;
+		*status |= TSNEP_XDP_REDIRECT;
+		return true;
+	default:
+		bpf_warn_invalid_xdp_action(rx->adapter->netdev, prog, act);
+		fallthrough;
+	case XDP_ABORTED:
+out_failure:
+		trace_xdp_exception(rx->adapter->netdev, prog, act);
+		fallthrough;
+	case XDP_DROP:
+		/* Due xdp_adjust_tail: DMA sync for_device cover max len CPU
+		 * touch
+		 */
+		sync = xdp->data_end - xdp->data_hard_start -
+		       XDP_PACKET_HEADROOM;
+		sync = max(sync, length);
+		page_pool_put_page(rx->page_pool, virt_to_head_page(xdp->data),
+				   sync, true);
+		return true;
+	}
+}
+
+static bool tsnep_xdp_run_prog_zc(struct tsnep_rx *rx, struct bpf_prog *prog,
+				  struct xdp_buff *xdp, int *status,
+				  struct netdev_queue *tx_nq,
+				  struct tsnep_tx *tx)
+{
+	u32 act;
+
+	act = bpf_prog_run_xdp(prog, xdp);
+
+	/* XDP_REDIRECT is the main action for zero-copy */
+	if (likely(act == XDP_REDIRECT)) {
+		if (xdp_do_redirect(rx->adapter->netdev, xdp, prog) < 0)
+			goto out_failure;
+		*status |= TSNEP_XDP_REDIRECT;
+		return true;
+	}
+
+	switch (act) {
+	case XDP_PASS:
+		return false;
+	case XDP_TX:
+		if (!tsnep_xdp_xmit_back(rx->adapter, xdp, tx_nq, tx, true))
+			goto out_failure;
+		*status |= TSNEP_XDP_TX;
+		return true;
+	default:
+		bpf_warn_invalid_xdp_action(rx->adapter->netdev, prog, act);
+		fallthrough;
+	case XDP_ABORTED:
+out_failure:
+		trace_xdp_exception(rx->adapter->netdev, prog, act);
+		fallthrough;
+	case XDP_DROP:
+		xsk_buff_free(xdp);
+		return true;
+	}
+}
+
+static void tsnep_finalize_xdp(struct tsnep_adapter *adapter, int status,
+			       struct netdev_queue *tx_nq, struct tsnep_tx *tx)
+{
+	if (status & TSNEP_XDP_TX) {
+		__netif_tx_lock(tx_nq, smp_processor_id());
+		tsnep_xdp_xmit_flush(tx);
+		__netif_tx_unlock(tx_nq);
+	}
+
+	if (status & TSNEP_XDP_REDIRECT)
+		xdp_do_flush();
+}
+
+static struct sk_buff *tsnep_build_skb(struct tsnep_rx *rx, struct page *page,
+				       int length)
+{
+	struct sk_buff *skb;
+
+	skb = napi_build_skb(page_address(page), PAGE_SIZE);
+	if (unlikely(!skb))
+		return NULL;
+
+	/* update pointers within the skb to store the data */
+	skb_reserve(skb, TSNEP_RX_OFFSET + TSNEP_RX_INLINE_METADATA_SIZE);
+	__skb_put(skb, length - ETH_FCS_LEN);
+
+	if (rx->adapter->hwtstamp_config.rx_filter == HWTSTAMP_FILTER_ALL) {
+		struct skb_shared_hwtstamps *hwtstamps = skb_hwtstamps(skb);
+		struct tsnep_rx_inline *rx_inline =
+			(struct tsnep_rx_inline *)(page_address(page) +
+						   TSNEP_RX_OFFSET);
+
+		skb_shinfo(skb)->tx_flags |=
+			SKBTX_HW_TSTAMP_NETDEV;
+		memset(hwtstamps, 0, sizeof(*hwtstamps));
+		hwtstamps->netdev_data = rx_inline;
+	}
+
+	skb_record_rx_queue(skb, rx->queue_index);
+	skb->protocol = eth_type_trans(skb, rx->adapter->netdev);
+
+	return skb;
+}
+
+static void tsnep_rx_page(struct tsnep_rx *rx, struct napi_struct *napi,
+			  struct page *page, int length)
+{
+	struct sk_buff *skb;
+
+	skb = tsnep_build_skb(rx, page, length);
+	if (skb) {
+		skb_mark_for_recycle(skb);
+
+		rx->packets++;
+		rx->bytes += length;
+		if (skb->pkt_type == PACKET_MULTICAST)
+			rx->multicast++;
+
+		napi_gro_receive(napi, skb);
+	} else {
+		page_pool_recycle_direct(rx->page_pool, page);
+
+		rx->dropped++;
+	}
+}
+
+static int tsnep_rx_poll(struct tsnep_rx *rx, struct napi_struct *napi,
+			 int budget)
+{
+	struct device *dmadev = rx->adapter->dmadev;
+	enum dma_data_direction dma_dir;
+	struct tsnep_rx_entry *entry;
+	struct netdev_queue *tx_nq;
+	struct bpf_prog *prog;
+	struct xdp_buff xdp;
+	struct tsnep_tx *tx;
+	int desc_available;
+	int xdp_status = 0;
+	int done = 0;
+	int length;
+
+	desc_available = tsnep_rx_desc_available(rx);
+	dma_dir = page_pool_get_dma_dir(rx->page_pool);
+	prog = READ_ONCE(rx->adapter->xdp_prog);
+	if (prog) {
+		tx_nq = netdev_get_tx_queue(rx->adapter->netdev,
+					    rx->tx_queue_index);
+		tx = &rx->adapter->tx[rx->tx_queue_index];
+
+		xdp_init_buff(&xdp, PAGE_SIZE, &rx->xdp_rxq);
+	}
+
+	while (likely(done < budget) && (rx->read != rx->write)) {
+		entry = &rx->entry[rx->read];
+		if ((__le32_to_cpu(entry->desc_wb->properties) &
+		     TSNEP_DESC_OWNER_COUNTER_MASK) !=
+		    (entry->properties & TSNEP_DESC_OWNER_COUNTER_MASK))
+			break;
+		done++;
+
+		if (desc_available >= TSNEP_RING_RX_REFILL) {
+			bool reuse = desc_available >= TSNEP_RING_RX_REUSE;
+
+			desc_available -= tsnep_rx_refill(rx, desc_available,
+							  reuse);
+			if (!entry->page) {
+				/* buffer has been reused for refill to prevent
+				 * empty RX ring, thus buffer cannot be used for
+				 * RX processing
+				 */
+				rx->read = (rx->read + 1) & TSNEP_RING_MASK;
+				desc_available++;
+
+				rx->dropped++;
+
+				continue;
+			}
+		}
+
+		/* descriptor properties shall be read first, because valid data
+		 * is signaled there
+		 */
+		dma_rmb();
+
+		prefetch(page_address(entry->page) + TSNEP_RX_OFFSET);
+		length = __le32_to_cpu(entry->desc_wb->properties) &
+			 TSNEP_DESC_LENGTH_MASK;
+		dma_sync_single_range_for_cpu(dmadev, entry->dma,
+					      TSNEP_RX_OFFSET, length, dma_dir);
+
+		/* RX metadata with timestamps is in front of actual data,
+		 * subtract metadata size to get length of actual data and
+		 * consider metadata size as offset of actual data during RX
+		 * processing
+		 */
+		length -= TSNEP_RX_INLINE_METADATA_SIZE;
+
+		rx->read = (rx->read + 1) & TSNEP_RING_MASK;
+		desc_available++;
+
+		if (prog) {
+			bool consume;
+
+			xdp_prepare_buff(&xdp, page_address(entry->page),
+					 XDP_PACKET_HEADROOM + TSNEP_RX_INLINE_METADATA_SIZE,
+					 length - ETH_FCS_LEN, false);
+
+			consume = tsnep_xdp_run_prog(rx, prog, &xdp,
+						     &xdp_status, tx_nq, tx);
+			if (consume) {
+				rx->packets++;
+				rx->bytes += length;
+
+				entry->page = NULL;
+
+				continue;
+			}
+		}
+
+		tsnep_rx_page(rx, napi, entry->page, length);
+		entry->page = NULL;
+	}
+
+	if (xdp_status)
+		tsnep_finalize_xdp(rx->adapter, xdp_status, tx_nq, tx);
+
+	if (desc_available)
+		tsnep_rx_refill(rx, desc_available, false);
+
+	return done;
+}
+
+static int tsnep_rx_poll_zc(struct tsnep_rx *rx, struct napi_struct *napi,
+			    int budget)
+{
+	struct tsnep_rx_entry *entry;
+	struct netdev_queue *tx_nq;
+	struct bpf_prog *prog;
+	struct tsnep_tx *tx;
+	int desc_available;
+	int xdp_status = 0;
+	struct page *page;
+	int done = 0;
+	int length;
+
+	desc_available = tsnep_rx_desc_available(rx);
+	prog = READ_ONCE(rx->adapter->xdp_prog);
+	if (prog) {
+		tx_nq = netdev_get_tx_queue(rx->adapter->netdev,
+					    rx->tx_queue_index);
+		tx = &rx->adapter->tx[rx->tx_queue_index];
+	}
+
+	while (likely(done < budget) && (rx->read != rx->write)) {
+		entry = &rx->entry[rx->read];
+		if ((__le32_to_cpu(entry->desc_wb->properties) &
+		     TSNEP_DESC_OWNER_COUNTER_MASK) !=
+		    (entry->properties & TSNEP_DESC_OWNER_COUNTER_MASK))
+			break;
+		done++;
+
+		if (desc_available >= TSNEP_RING_RX_REFILL) {
+			bool reuse = desc_available >= TSNEP_RING_RX_REUSE;
+
+			desc_available -= tsnep_rx_refill_zc(rx, desc_available,
+							     reuse);
+			if (!entry->xdp) {
+				/* buffer has been reused for refill to prevent
+				 * empty RX ring, thus buffer cannot be used for
+				 * RX processing
+				 */
+				rx->read = (rx->read + 1) & TSNEP_RING_MASK;
+				desc_available++;
+
+				rx->dropped++;
+
+				continue;
+			}
+		}
+
+		/* descriptor properties shall be read first, because valid data
+		 * is signaled there
+		 */
+		dma_rmb();
+
+		prefetch(entry->xdp->data);
+		length = __le32_to_cpu(entry->desc_wb->properties) &
+			 TSNEP_DESC_LENGTH_MASK;
+		xsk_buff_set_size(entry->xdp, length - ETH_FCS_LEN);
+		xsk_buff_dma_sync_for_cpu(entry->xdp);
+
+		/* RX metadata with timestamps is in front of actual data,
+		 * subtract metadata size to get length of actual data and
+		 * consider metadata size as offset of actual data during RX
+		 * processing
+		 */
+		length -= TSNEP_RX_INLINE_METADATA_SIZE;
+
+		rx->read = (rx->read + 1) & TSNEP_RING_MASK;
+		desc_available++;
+
+		if (prog) {
+			bool consume;
+
+			entry->xdp->data += TSNEP_RX_INLINE_METADATA_SIZE;
+			entry->xdp->data_meta += TSNEP_RX_INLINE_METADATA_SIZE;
+
+			consume = tsnep_xdp_run_prog_zc(rx, prog, entry->xdp,
+							&xdp_status, tx_nq, tx);
+			if (consume) {
+				rx->packets++;
+				rx->bytes += length;
+
+				entry->xdp = NULL;
+
+				continue;
+			}
+		}
+
+		page = page_pool_dev_alloc_pages(rx->page_pool);
+		if (page) {
+			memcpy(page_address(page) + TSNEP_RX_OFFSET,
+			       entry->xdp->data - TSNEP_RX_INLINE_METADATA_SIZE,
+			       length + TSNEP_RX_INLINE_METADATA_SIZE);
+			tsnep_rx_page(rx, napi, page, length);
+		} else {
+			rx->dropped++;
+		}
+		xsk_buff_free(entry->xdp);
+		entry->xdp = NULL;
+	}
+
+	if (xdp_status)
+		tsnep_finalize_xdp(rx->adapter, xdp_status, tx_nq, tx);
+
+	if (desc_available)
+		desc_available -= tsnep_rx_refill_zc(rx, desc_available, false);
+
+	if (xsk_uses_need_wakeup(rx->xsk_pool)) {
+		tsnep_xsk_rx_need_wakeup(rx, desc_available);
+
+		return done;
+	}
+
+	return desc_available ? budget : done;
+}
+
+static bool tsnep_rx_pending(struct tsnep_rx *rx)
+{
+	struct tsnep_rx_entry *entry;
+
+	if (rx->read != rx->write) {
+		entry = &rx->entry[rx->read];
+		if ((__le32_to_cpu(entry->desc_wb->properties) &
+		     TSNEP_DESC_OWNER_COUNTER_MASK) ==
+		    (entry->properties & TSNEP_DESC_OWNER_COUNTER_MASK))
+			return true;
+	}
+
+	return false;
+}
+
+static int tsnep_rx_open(struct tsnep_rx *rx)
+{
+	int desc_available;
+	int retval;
+
+	retval = tsnep_rx_ring_create(rx);
+	if (retval)
+		return retval;
+
+	tsnep_rx_init(rx);
+
+	desc_available = tsnep_rx_desc_available(rx);
+	if (rx->xsk_pool)
+		retval = tsnep_rx_alloc_zc(rx, desc_available, false);
+	else
+		retval = tsnep_rx_alloc(rx, desc_available, false);
+	if (retval != desc_available) {
+		retval = -ENOMEM;
+
+		goto alloc_failed;
+	}
+
+	/* prealloc pages to prevent allocation failures when XSK pool is
+	 * disabled at runtime
+	 */
+	if (rx->xsk_pool) {
+		retval = tsnep_rx_alloc_page_buffer(rx);
+		if (retval)
+			goto alloc_failed;
+	}
+
+	return 0;
+
+alloc_failed:
+	tsnep_rx_ring_cleanup(rx);
+	return retval;
+}
+
+static void tsnep_rx_close(struct tsnep_rx *rx)
+{
+	if (rx->xsk_pool)
+		tsnep_rx_free_page_buffer(rx);
+
+	tsnep_rx_ring_cleanup(rx);
+}
+
+static void tsnep_rx_reopen(struct tsnep_rx *rx)
+{
+	struct page **page = rx->page_buffer;
+	int i;
+
+	tsnep_rx_init(rx);
+
+	for (i = 0; i < TSNEP_RING_SIZE; i++) {
+		struct tsnep_rx_entry *entry = &rx->entry[i];
+
+		/* defined initial values for properties are required for
+		 * correct owner counter checking
+		 */
+		entry->desc->properties = 0;
+		entry->desc_wb->properties = 0;
+
+		/* prevent allocation failures by reusing kept pages */
+		if (*page) {
+			tsnep_rx_set_page(rx, entry, *page);
+			tsnep_rx_activate(rx, rx->write);
+			rx->write++;
+
+			*page = NULL;
+			page++;
+		}
+	}
+}
+
+static void tsnep_rx_reopen_xsk(struct tsnep_rx *rx)
+{
+	struct page **page = rx->page_buffer;
+	u32 allocated;
+	int i;
+
+	tsnep_rx_init(rx);
+
+	/* alloc all ring entries except the last one, because ring cannot be
+	 * filled completely, as many buffers as possible is enough as wakeup is
+	 * done if new buffers are available
+	 */
+	allocated = xsk_buff_alloc_batch(rx->xsk_pool, rx->xdp_batch,
+					 TSNEP_RING_SIZE - 1);
+
+	for (i = 0; i < TSNEP_RING_SIZE; i++) {
+		struct tsnep_rx_entry *entry = &rx->entry[i];
+
+		/* keep pages to prevent allocation failures when xsk is
+		 * disabled
+		 */
+		if (entry->page) {
+			*page = entry->page;
+			entry->page = NULL;
+
+			page++;
+		}
+
+		/* defined initial values for properties are required for
+		 * correct owner counter checking
+		 */
+		entry->desc->properties = 0;
+		entry->desc_wb->properties = 0;
+
+		if (allocated) {
+			tsnep_rx_set_xdp(rx, entry,
+					 rx->xdp_batch[allocated - 1]);
+			tsnep_rx_activate(rx, rx->write);
+			rx->write++;
+
+			allocated--;
+		}
+	}
+
+	/* set need wakeup flag immediately if ring is not filled completely,
+	 * first polling would be too late as need wakeup signalisation would
+	 * be delayed for an indefinite time
+	 */
+	if (xsk_uses_need_wakeup(rx->xsk_pool))
+		tsnep_xsk_rx_need_wakeup(rx, tsnep_rx_desc_available(rx));
+}
+
+static bool tsnep_pending(struct tsnep_queue *queue)
+{
+	if (queue->tx && tsnep_tx_pending(queue->tx))
+		return true;
+
+	if (queue->rx && tsnep_rx_pending(queue->rx))
+		return true;
+
+	return false;
+}
+
+static int tsnep_poll(struct napi_struct *napi, int budget)
+{
+	struct tsnep_queue *queue = container_of(napi, struct tsnep_queue,
+						 napi);
+	bool complete = true;
+	int done = 0;
+
+	if (queue->tx)
+		complete = tsnep_tx_poll(queue->tx, budget);
+
+	/* handle case where we are called by netpoll with a budget of 0 */
+	if (unlikely(budget <= 0))
+		return budget;
+
+	if (queue->rx) {
+		done = queue->rx->xsk_pool ?
+		       tsnep_rx_poll_zc(queue->rx, napi, budget) :
+		       tsnep_rx_poll(queue->rx, napi, budget);
+		if (done >= budget)
+			complete = false;
+	}
+
+	/* if all work not completed, return budget and keep polling */
+	if (!complete)
+		return budget;
+
+	if (likely(napi_complete_done(napi, done))) {
+		tsnep_enable_irq(queue->adapter, queue->irq_mask);
+
+		/* reschedule if work is already pending, prevent rotten packets
+		 * which are transmitted or received after polling but before
+		 * interrupt enable
+		 */
+		if (tsnep_pending(queue)) {
+			tsnep_disable_irq(queue->adapter, queue->irq_mask);
+			napi_schedule(napi);
+		}
+	}
+
+	return min(done, budget - 1);
+}
+
+static int tsnep_request_irq(struct tsnep_queue *queue, bool first)
+{
+	const char *name = netdev_name(queue->adapter->netdev);
+	irq_handler_t handler;
+	void *dev;
+	int retval;
+
+	if (first) {
+		sprintf(queue->name, "%s-mac", name);
+		handler = tsnep_irq;
+		dev = queue->adapter;
+	} else {
+		if (queue->tx && queue->rx)
+			snprintf(queue->name, sizeof(queue->name), "%s-txrx-%d",
+				 name, queue->rx->queue_index);
+		else if (queue->tx)
+			snprintf(queue->name, sizeof(queue->name), "%s-tx-%d",
+				 name, queue->tx->queue_index);
+		else
+			snprintf(queue->name, sizeof(queue->name), "%s-rx-%d",
+				 name, queue->rx->queue_index);
+		handler = tsnep_irq_txrx;
+		dev = queue;
+	}
+
+	retval = request_irq(queue->irq, handler, 0, queue->name, dev);
+	if (retval) {
+		/* if name is empty, then interrupt won't be freed */
+		memset(queue->name, 0, sizeof(queue->name));
+	}
+
+	return retval;
+}
+
+static void tsnep_free_irq(struct tsnep_queue *queue, bool first)
+{
+	void *dev;
+
+	if (!strlen(queue->name))
+		return;
+
+	if (first)
+		dev = queue->adapter;
+	else
+		dev = queue;
+
+	free_irq(queue->irq, dev);
+	memset(queue->name, 0, sizeof(queue->name));
+}
+
+static void tsnep_queue_close(struct tsnep_queue *queue, bool first)
+{
+	struct tsnep_rx *rx = queue->rx;
+
+	tsnep_free_irq(queue, first);
+
+	if (rx) {
+		if (xdp_rxq_info_is_reg(&rx->xdp_rxq))
+			xdp_rxq_info_unreg(&rx->xdp_rxq);
+		if (xdp_rxq_info_is_reg(&rx->xdp_rxq_zc))
+			xdp_rxq_info_unreg(&rx->xdp_rxq_zc);
+	}
+
+	netif_napi_del(&queue->napi);
+}
+
+static int tsnep_queue_open(struct tsnep_adapter *adapter,
+			    struct tsnep_queue *queue, bool first)
+{
+	struct tsnep_rx *rx = queue->rx;
+	struct tsnep_tx *tx = queue->tx;
+	int retval;
+
+	netif_napi_add(adapter->netdev, &queue->napi, tsnep_poll);
+
+	if (rx) {
+		/* choose TX queue for XDP_TX */
+		if (tx)
+			rx->tx_queue_index = tx->queue_index;
+		else if (rx->queue_index < adapter->num_tx_queues)
+			rx->tx_queue_index = rx->queue_index;
+		else
+			rx->tx_queue_index = 0;
+
+		/* prepare both memory models to eliminate possible registration
+		 * errors when memory model is switched between page pool and
+		 * XSK pool during runtime
+		 */
+		retval = xdp_rxq_info_reg(&rx->xdp_rxq, adapter->netdev,
+					  rx->queue_index, queue->napi.napi_id);
+		if (retval)
+			goto failed;
+		retval = xdp_rxq_info_reg_mem_model(&rx->xdp_rxq,
+						    MEM_TYPE_PAGE_POOL,
+						    rx->page_pool);
+		if (retval)
+			goto failed;
+		retval = xdp_rxq_info_reg(&rx->xdp_rxq_zc, adapter->netdev,
+					  rx->queue_index, queue->napi.napi_id);
+		if (retval)
+			goto failed;
+		retval = xdp_rxq_info_reg_mem_model(&rx->xdp_rxq_zc,
+						    MEM_TYPE_XSK_BUFF_POOL,
+						    NULL);
+		if (retval)
+			goto failed;
+		if (rx->xsk_pool)
+			xsk_pool_set_rxq_info(rx->xsk_pool, &rx->xdp_rxq_zc);
+	}
+
+	retval = tsnep_request_irq(queue, first);
+	if (retval) {
+		netif_err(adapter, drv, adapter->netdev,
+			  "can't get assigned irq %d.\n", queue->irq);
+		goto failed;
+	}
+
+	return 0;
+
+failed:
+	tsnep_queue_close(queue, first);
+
+	return retval;
+}
+
+static void tsnep_queue_enable(struct tsnep_queue *queue)
+{
+	struct tsnep_adapter *adapter = queue->adapter;
+
+	netif_napi_set_irq(&queue->napi, queue->irq);
+	napi_enable(&queue->napi);
+	tsnep_enable_irq(adapter, queue->irq_mask);
+
+	if (queue->tx) {
+		netif_queue_set_napi(adapter->netdev, queue->tx->queue_index,
+				     NETDEV_QUEUE_TYPE_TX, &queue->napi);
+		tsnep_tx_enable(queue->tx);
+	}
+
+	if (queue->rx) {
+		netif_queue_set_napi(adapter->netdev, queue->rx->queue_index,
+				     NETDEV_QUEUE_TYPE_RX, &queue->napi);
+		tsnep_rx_enable(queue->rx);
+	}
+}
+
+static void tsnep_queue_disable(struct tsnep_queue *queue)
+{
+	struct tsnep_adapter *adapter = queue->adapter;
+
+	if (queue->rx)
+		netif_queue_set_napi(adapter->netdev, queue->rx->queue_index,
+				     NETDEV_QUEUE_TYPE_RX, NULL);
+
+	if (queue->tx) {
+		tsnep_tx_disable(queue->tx, &queue->napi);
+		netif_queue_set_napi(adapter->netdev, queue->tx->queue_index,
+				     NETDEV_QUEUE_TYPE_TX, NULL);
+	}
+
+	napi_disable(&queue->napi);
+	tsnep_disable_irq(adapter, queue->irq_mask);
+
+	/* disable RX after NAPI polling has been disabled, because RX can be
+	 * enabled during NAPI polling
+	 */
+	if (queue->rx)
+		tsnep_rx_disable(queue->rx);
+}
+
+static int tsnep_netdev_open(struct net_device *netdev)
+{
+	struct tsnep_adapter *adapter = netdev_priv(netdev);
+	int i, retval;
+
+	for (i = 0; i < adapter->num_queues; i++) {
+		if (adapter->queue[i].tx) {
+			retval = tsnep_tx_open(adapter->queue[i].tx);
+			if (retval)
+				goto failed;
+		}
+		if (adapter->queue[i].rx) {
+			retval = tsnep_rx_open(adapter->queue[i].rx);
+			if (retval)
+				goto failed;
+		}
+
+		retval = tsnep_queue_open(adapter, &adapter->queue[i], i == 0);
+		if (retval)
+			goto failed;
+	}
+
+	retval = netif_set_real_num_tx_queues(adapter->netdev,
+					      adapter->num_tx_queues);
+	if (retval)
+		goto failed;
+	retval = netif_set_real_num_rx_queues(adapter->netdev,
+					      adapter->num_rx_queues);
+	if (retval)
+		goto failed;
+
+	tsnep_enable_irq(adapter, ECM_INT_LINK);
+	retval = tsnep_phy_open(adapter);
+	if (retval)
+		goto phy_failed;
+
+	for (i = 0; i < adapter->num_queues; i++)
+		tsnep_queue_enable(&adapter->queue[i]);
+
+	return 0;
+
+phy_failed:
+	tsnep_disable_irq(adapter, ECM_INT_LINK);
+failed:
+	for (i = 0; i < adapter->num_queues; i++) {
+		tsnep_queue_close(&adapter->queue[i], i == 0);
+
+		if (adapter->queue[i].rx)
+			tsnep_rx_close(adapter->queue[i].rx);
+		if (adapter->queue[i].tx)
+			tsnep_tx_close(adapter->queue[i].tx);
+	}
+	return retval;
+}
+
+static int tsnep_netdev_close(struct net_device *netdev)
+{
+	struct tsnep_adapter *adapter = netdev_priv(netdev);
+	int i;
+
+	tsnep_disable_irq(adapter, ECM_INT_LINK);
+	tsnep_phy_close(adapter);
+
+	for (i = 0; i < adapter->num_queues; i++) {
+		tsnep_queue_disable(&adapter->queue[i]);
+
+		tsnep_queue_close(&adapter->queue[i], i == 0);
+
+		if (adapter->queue[i].rx)
+			tsnep_rx_close(adapter->queue[i].rx);
+		if (adapter->queue[i].tx)
+			tsnep_tx_close(adapter->queue[i].tx);
+	}
+
+	return 0;
+}
+
+int tsnep_enable_xsk(struct tsnep_queue *queue, struct xsk_buff_pool *pool)
+{
+	bool running = netif_running(queue->adapter->netdev);
+	u32 frame_size;
+
+	frame_size = xsk_pool_get_rx_frame_size(pool);
+	if (frame_size < TSNEP_XSK_RX_BUF_SIZE)
+		return -EOPNOTSUPP;
+
+	queue->rx->page_buffer = kcalloc(TSNEP_RING_SIZE,
+					 sizeof(*queue->rx->page_buffer),
+					 GFP_KERNEL);
+	if (!queue->rx->page_buffer)
+		return -ENOMEM;
+	queue->rx->xdp_batch = kcalloc(TSNEP_RING_SIZE,
+				       sizeof(*queue->rx->xdp_batch),
+				       GFP_KERNEL);
+	if (!queue->rx->xdp_batch) {
+		kfree(queue->rx->page_buffer);
+		queue->rx->page_buffer = NULL;
+
+		return -ENOMEM;
+	}
+
+	xsk_pool_set_rxq_info(pool, &queue->rx->xdp_rxq_zc);
+
+	if (running)
+		tsnep_queue_disable(queue);
+
+	queue->tx->xsk_pool = pool;
+	queue->rx->xsk_pool = pool;
+
+	if (running) {
+		tsnep_rx_reopen_xsk(queue->rx);
+		tsnep_queue_enable(queue);
+	}
+
+	return 0;
+}
+
+void tsnep_disable_xsk(struct tsnep_queue *queue)
+{
+	bool running = netif_running(queue->adapter->netdev);
+
+	if (running)
+		tsnep_queue_disable(queue);
+
+	tsnep_rx_free_zc(queue->rx);
+
+	queue->rx->xsk_pool = NULL;
+	queue->tx->xsk_pool = NULL;
+
+	if (running) {
+		tsnep_rx_reopen(queue->rx);
+		tsnep_queue_enable(queue);
+	}
+
+	kfree(queue->rx->xdp_batch);
+	queue->rx->xdp_batch = NULL;
+	kfree(queue->rx->page_buffer);
+	queue->rx->page_buffer = NULL;
+}
+
+static netdev_tx_t tsnep_netdev_xmit_frame(struct sk_buff *skb,
+					   struct net_device *netdev)
+{
+	struct tsnep_adapter *adapter = netdev_priv(netdev);
+	u16 queue_mapping = skb_get_queue_mapping(skb);
+
+	if (queue_mapping >= adapter->num_tx_queues)
+		queue_mapping = 0;
+
+	return tsnep_xmit_frame_ring(skb, &adapter->tx[queue_mapping]);
+}
+
+static int tsnep_netdev_ioctl(struct net_device *netdev, struct ifreq *ifr,
+			      int cmd)
+{
+	if (!netif_running(netdev))
+		return -EINVAL;
+	if (cmd == SIOCSHWTSTAMP || cmd == SIOCGHWTSTAMP)
+		return tsnep_ptp_ioctl(netdev, ifr, cmd);
+	return phy_mii_ioctl(netdev->phydev, ifr, cmd);
+}
+
+static void tsnep_netdev_set_multicast(struct net_device *netdev)
+{
+	struct tsnep_adapter *adapter = netdev_priv(netdev);
+
+	u16 rx_filter = 0;
+
+	/* configured MAC address and broadcasts are never filtered */
+	if (netdev->flags & IFF_PROMISC) {
+		rx_filter |= TSNEP_RX_FILTER_ACCEPT_ALL_MULTICASTS;
+		rx_filter |= TSNEP_RX_FILTER_ACCEPT_ALL_UNICASTS;
+	} else if (!netdev_mc_empty(netdev) || (netdev->flags & IFF_ALLMULTI)) {
+		rx_filter |= TSNEP_RX_FILTER_ACCEPT_ALL_MULTICASTS;
+	}
+	iowrite16(rx_filter, adapter->addr + TSNEP_RX_FILTER);
+}
+
+static void tsnep_netdev_get_stats64(struct net_device *netdev,
+				     struct rtnl_link_stats64 *stats)
+{
+	struct tsnep_adapter *adapter = netdev_priv(netdev);
+	u32 reg;
+	u32 val;
+	int i;
+
+	for (i = 0; i < adapter->num_tx_queues; i++) {
+		stats->tx_packets += adapter->tx[i].packets;
+		stats->tx_bytes += adapter->tx[i].bytes;
+		stats->tx_dropped += adapter->tx[i].dropped;
+	}
+	for (i = 0; i < adapter->num_rx_queues; i++) {
+		stats->rx_packets += adapter->rx[i].packets;
+		stats->rx_bytes += adapter->rx[i].bytes;
+		stats->rx_dropped += adapter->rx[i].dropped;
+		stats->multicast += adapter->rx[i].multicast;
+
+		reg = ioread32(adapter->addr + TSNEP_QUEUE(i) +
+			       TSNEP_RX_STATISTIC);
+		val = (reg & TSNEP_RX_STATISTIC_NO_DESC_MASK) >>
+		      TSNEP_RX_STATISTIC_NO_DESC_SHIFT;
+		stats->rx_dropped += val;
+		val = (reg & TSNEP_RX_STATISTIC_BUFFER_TOO_SMALL_MASK) >>
+		      TSNEP_RX_STATISTIC_BUFFER_TOO_SMALL_SHIFT;
+		stats->rx_dropped += val;
+		val = (reg & TSNEP_RX_STATISTIC_FIFO_OVERFLOW_MASK) >>
+		      TSNEP_RX_STATISTIC_FIFO_OVERFLOW_SHIFT;
+		stats->rx_errors += val;
+		stats->rx_fifo_errors += val;
+		val = (reg & TSNEP_RX_STATISTIC_INVALID_FRAME_MASK) >>
+		      TSNEP_RX_STATISTIC_INVALID_FRAME_SHIFT;
+		stats->rx_errors += val;
+		stats->rx_frame_errors += val;
+	}
+
+	reg = ioread32(adapter->addr + ECM_STAT);
+	val = (reg & ECM_STAT_RX_ERR_MASK) >> ECM_STAT_RX_ERR_SHIFT;
+	stats->rx_errors += val;
+	val = (reg & ECM_STAT_INV_FRM_MASK) >> ECM_STAT_INV_FRM_SHIFT;
+	stats->rx_errors += val;
+	stats->rx_crc_errors += val;
+	val = (reg & ECM_STAT_FWD_RX_ERR_MASK) >> ECM_STAT_FWD_RX_ERR_SHIFT;
+	stats->rx_errors += val;
+}
+
+static void tsnep_mac_set_address(struct tsnep_adapter *adapter, u8 *addr)
+{
+	iowrite32(*(u32 *)addr, adapter->addr + TSNEP_MAC_ADDRESS_LOW);
+	iowrite16(*(u16 *)(addr + sizeof(u32)),
+		  adapter->addr + TSNEP_MAC_ADDRESS_HIGH);
+
+	ether_addr_copy(adapter->mac_address, addr);
+	netif_info(adapter, drv, adapter->netdev, "MAC address set to %pM\n",
+		   addr);
+}
+
+static int tsnep_netdev_set_mac_address(struct net_device *netdev, void *addr)
+{
+	struct tsnep_adapter *adapter = netdev_priv(netdev);
+	struct sockaddr *sock_addr = addr;
+	int retval;
+
+	retval = eth_prepare_mac_addr_change(netdev, sock_addr);
+	if (retval)
+		return retval;
+	eth_hw_addr_set(netdev, sock_addr->sa_data);
+	tsnep_mac_set_address(adapter, sock_addr->sa_data);
+
+	return 0;
+}
+
+static int tsnep_netdev_set_features(struct net_device *netdev,
+				     netdev_features_t features)
+{
+	struct tsnep_adapter *adapter = netdev_priv(netdev);
+	netdev_features_t changed = netdev->features ^ features;
+	bool enable;
+	int retval = 0;
+
+	if (changed & NETIF_F_LOOPBACK) {
+		enable = !!(features & NETIF_F_LOOPBACK);
+		retval = tsnep_phy_loopback(adapter, enable);
+	}
+
+	return retval;
+}
+
+static ktime_t tsnep_netdev_get_tstamp(struct net_device *netdev,
+				       const struct skb_shared_hwtstamps *hwtstamps,
+				       bool cycles)
+{
+	struct tsnep_rx_inline *rx_inline = hwtstamps->netdev_data;
+	u64 timestamp;
+
+	if (cycles)
+		timestamp = __le64_to_cpu(rx_inline->counter);
+	else
+		timestamp = __le64_to_cpu(rx_inline->timestamp);
+
+	return ns_to_ktime(timestamp);
+}
+
+static int tsnep_netdev_bpf(struct net_device *dev, struct netdev_bpf *bpf)
+{
+	struct tsnep_adapter *adapter = netdev_priv(dev);
+
+	switch (bpf->command) {
+	case XDP_SETUP_PROG:
+		return tsnep_xdp_setup_prog(adapter, bpf->prog, bpf->extack);
+	case XDP_SETUP_XSK_POOL:
+		return tsnep_xdp_setup_pool(adapter, bpf->xsk.pool,
+					    bpf->xsk.queue_id);
+	default:
+		return -EOPNOTSUPP;
+	}
+}
+
+static struct tsnep_tx *tsnep_xdp_get_tx(struct tsnep_adapter *adapter, u32 cpu)
+{
+	if (cpu >= TSNEP_MAX_QUEUES)
+		cpu &= TSNEP_MAX_QUEUES - 1;
+
+	while (cpu >= adapter->num_tx_queues)
+		cpu -= adapter->num_tx_queues;
+
+	return &adapter->tx[cpu];
+}
+
+static int tsnep_netdev_xdp_xmit(struct net_device *dev, int n,
+				 struct xdp_frame **xdp, u32 flags)
+{
+	struct tsnep_adapter *adapter = netdev_priv(dev);
+	u32 cpu = smp_processor_id();
+	struct netdev_queue *nq;
+	struct tsnep_tx *tx;
+	int nxmit;
+	bool xmit;
+
+	if (unlikely(flags & ~XDP_XMIT_FLAGS_MASK))
+		return -EINVAL;
+
+	tx = tsnep_xdp_get_tx(adapter, cpu);
+	nq = netdev_get_tx_queue(adapter->netdev, tx->queue_index);
+
+	__netif_tx_lock(nq, cpu);
+
+	for (nxmit = 0; nxmit < n; nxmit++) {
+		xmit = tsnep_xdp_xmit_frame_ring(xdp[nxmit], tx,
+						 TSNEP_TX_TYPE_XDP_NDO);
+		if (!xmit)
+			break;
+
+		/* avoid transmit queue timeout since we share it with the slow
+		 * path
+		 */
+		txq_trans_cond_update(nq);
+	}
+
+	if (flags & XDP_XMIT_FLUSH)
+		tsnep_xdp_xmit_flush(tx);
+
+	__netif_tx_unlock(nq);
+
+	return nxmit;
+}
+
+static int tsnep_netdev_xsk_wakeup(struct net_device *dev, u32 queue_id,
+				   u32 flags)
+{
+	struct tsnep_adapter *adapter = netdev_priv(dev);
+	struct tsnep_queue *queue;
+
+	if (queue_id >= adapter->num_rx_queues ||
+	    queue_id >= adapter->num_tx_queues)
+		return -EINVAL;
+
+	queue = &adapter->queue[queue_id];
+
+	if (!napi_if_scheduled_mark_missed(&queue->napi))
+		napi_schedule(&queue->napi);
+
+	return 0;
+}
+
+static const struct net_device_ops tsnep_netdev_ops = {
+	.ndo_open = tsnep_netdev_open,
+	.ndo_stop = tsnep_netdev_close,
+	.ndo_start_xmit = tsnep_netdev_xmit_frame,
+	.ndo_eth_ioctl = tsnep_netdev_ioctl,
+	.ndo_set_rx_mode = tsnep_netdev_set_multicast,
+	.ndo_get_stats64 = tsnep_netdev_get_stats64,
+	.ndo_set_mac_address = tsnep_netdev_set_mac_address,
+	.ndo_set_features = tsnep_netdev_set_features,
+	.ndo_get_tstamp = tsnep_netdev_get_tstamp,
+	.ndo_setup_tc = tsnep_tc_setup,
+	.ndo_bpf = tsnep_netdev_bpf,
+	.ndo_xdp_xmit = tsnep_netdev_xdp_xmit,
+	.ndo_xsk_wakeup = tsnep_netdev_xsk_wakeup,
+};
+
+static int tsnep_mac_init(struct tsnep_adapter *adapter)
+{
+	int retval;
+
+	/* initialize RX filtering, at least configured MAC address and
+	 * broadcast are not filtered
+	 */
+	iowrite16(0, adapter->addr + TSNEP_RX_FILTER);
+
+	/* try to get MAC address in the following order:
+	 * - device tree
+	 * - valid MAC address already set
+	 * - MAC address register if valid
+	 * - random MAC address
+	 */
+	retval = of_get_mac_address(adapter->pdev->dev.of_node,
+				    adapter->mac_address);
+	if (retval == -EPROBE_DEFER)
+		return retval;
+	if (retval && !is_valid_ether_addr(adapter->mac_address)) {
+		*(u32 *)adapter->mac_address =
+			ioread32(adapter->addr + TSNEP_MAC_ADDRESS_LOW);
+		*(u16 *)(adapter->mac_address + sizeof(u32)) =
+			ioread16(adapter->addr + TSNEP_MAC_ADDRESS_HIGH);
+		if (!is_valid_ether_addr(adapter->mac_address))
+			eth_random_addr(adapter->mac_address);
+	}
+
+	tsnep_mac_set_address(adapter, adapter->mac_address);
+	eth_hw_addr_set(adapter->netdev, adapter->mac_address);
+
+	return 0;
+}
+
+static int tsnep_mdio_init(struct tsnep_adapter *adapter)
+{
+	struct device_node *np = adapter->pdev->dev.of_node;
+	int retval;
+
+	if (np) {
+		np = of_get_child_by_name(np, "mdio");
+		if (!np)
+			return 0;
+
+		adapter->suppress_preamble =
+			of_property_read_bool(np, "suppress-preamble");
+	}
+
+	adapter->mdiobus = devm_mdiobus_alloc(&adapter->pdev->dev);
+	if (!adapter->mdiobus) {
+		retval = -ENOMEM;
+
+		goto out;
+	}
+
+	adapter->mdiobus->priv = (void *)adapter;
+	adapter->mdiobus->parent = &adapter->pdev->dev;
+	adapter->mdiobus->read = tsnep_mdiobus_read;
+	adapter->mdiobus->write = tsnep_mdiobus_write;
+	adapter->mdiobus->name = TSNEP "-mdiobus";
+	snprintf(adapter->mdiobus->id, MII_BUS_ID_SIZE, "%s",
+		 adapter->pdev->name);
+
+	/* do not scan broadcast address */
+	adapter->mdiobus->phy_mask = 0x0000001;
+
+	retval = of_mdiobus_register(adapter->mdiobus, np);
+
+out:
+	of_node_put(np);
+
+	return retval;
+}
+
+static int tsnep_phy_init(struct tsnep_adapter *adapter)
+{
+	struct device_node *phy_node;
+	int retval;
+
+	retval = of_get_phy_mode(adapter->pdev->dev.of_node,
+				 &adapter->phy_mode);
+	if (retval)
+		adapter->phy_mode = PHY_INTERFACE_MODE_GMII;
+
+	phy_node = of_parse_phandle(adapter->pdev->dev.of_node, "phy-handle",
+				    0);
+	adapter->phydev = of_phy_find_device(phy_node);
+	of_node_put(phy_node);
+	if (!adapter->phydev && adapter->mdiobus)
+		adapter->phydev = phy_find_first(adapter->mdiobus);
+	if (!adapter->phydev)
+		return -EIO;
+
+	return 0;
+}
+
+static int tsnep_queue_init(struct tsnep_adapter *adapter, int queue_count)
+{
+	u32 irq_mask = ECM_INT_TX_0 | ECM_INT_RX_0;
+	char name[8];
+	int i;
+	int retval;
+
+	/* one TX/RX queue pair for netdev is mandatory */
+	if (platform_irq_count(adapter->pdev) == 1)
+		retval = platform_get_irq(adapter->pdev, 0);
+	else
+		retval = platform_get_irq_byname(adapter->pdev, "mac");
+	if (retval < 0)
+		return retval;
+	adapter->num_tx_queues = 1;
+	adapter->num_rx_queues = 1;
+	adapter->num_queues = 1;
+	adapter->queue[0].adapter = adapter;
+	adapter->queue[0].irq = retval;
+	adapter->queue[0].tx = &adapter->tx[0];
+	adapter->queue[0].tx->adapter = adapter;
+	adapter->queue[0].tx->addr = adapter->addr + TSNEP_QUEUE(0);
+	adapter->queue[0].tx->queue_index = 0;
+	adapter->queue[0].rx = &adapter->rx[0];
+	adapter->queue[0].rx->adapter = adapter;
+	adapter->queue[0].rx->addr = adapter->addr + TSNEP_QUEUE(0);
+	adapter->queue[0].rx->queue_index = 0;
+	adapter->queue[0].irq_mask = irq_mask;
+	adapter->queue[0].irq_delay_addr = adapter->addr + ECM_INT_DELAY;
+	retval = tsnep_set_irq_coalesce(&adapter->queue[0],
+					TSNEP_COALESCE_USECS_DEFAULT);
+	if (retval < 0)
+		return retval;
+
+	adapter->netdev->irq = adapter->queue[0].irq;
+
+	/* add additional TX/RX queue pairs only if dedicated interrupt is
+	 * available
+	 */
+	for (i = 1; i < queue_count; i++) {
+		sprintf(name, "txrx-%d", i);
+		retval = platform_get_irq_byname_optional(adapter->pdev, name);
+		if (retval < 0)
+			break;
+
+		adapter->num_tx_queues++;
+		adapter->num_rx_queues++;
+		adapter->num_queues++;
+		adapter->queue[i].adapter = adapter;
+		adapter->queue[i].irq = retval;
+		adapter->queue[i].tx = &adapter->tx[i];
+		adapter->queue[i].tx->adapter = adapter;
+		adapter->queue[i].tx->addr = adapter->addr + TSNEP_QUEUE(i);
+		adapter->queue[i].tx->queue_index = i;
+		adapter->queue[i].rx = &adapter->rx[i];
+		adapter->queue[i].rx->adapter = adapter;
+		adapter->queue[i].rx->addr = adapter->addr + TSNEP_QUEUE(i);
+		adapter->queue[i].rx->queue_index = i;
+		adapter->queue[i].irq_mask =
+			irq_mask << (ECM_INT_TXRX_SHIFT * i);
+		adapter->queue[i].irq_delay_addr =
+			adapter->addr + ECM_INT_DELAY + ECM_INT_DELAY_OFFSET * i;
+		retval = tsnep_set_irq_coalesce(&adapter->queue[i],
+						TSNEP_COALESCE_USECS_DEFAULT);
+		if (retval < 0)
+			return retval;
+	}
+
+	return 0;
+}
+
+static int tsnep_probe(struct platform_device *pdev)
+{
+	struct tsnep_adapter *adapter;
+	struct net_device *netdev;
+	struct resource *io;
+	u32 type;
+	int revision;
+	int version;
+	int queue_count;
+	int retval;
+
+	netdev = devm_alloc_etherdev_mqs(&pdev->dev,
+					 sizeof(struct tsnep_adapter),
+					 TSNEP_MAX_QUEUES, TSNEP_MAX_QUEUES);
+	if (!netdev)
+		return -ENODEV;
+	SET_NETDEV_DEV(netdev, &pdev->dev);
+	adapter = netdev_priv(netdev);
+	platform_set_drvdata(pdev, adapter);
+	adapter->pdev = pdev;
+	adapter->dmadev = &pdev->dev;
+	adapter->netdev = netdev;
+	adapter->msg_enable = NETIF_MSG_DRV | NETIF_MSG_PROBE |
+			      NETIF_MSG_LINK | NETIF_MSG_IFUP |
+			      NETIF_MSG_IFDOWN | NETIF_MSG_TX_QUEUED;
+
+	netdev->min_mtu = ETH_MIN_MTU;
+	netdev->max_mtu = TSNEP_MAX_FRAME_SIZE;
+
+	mutex_init(&adapter->gate_control_lock);
+	mutex_init(&adapter->rxnfc_lock);
+	INIT_LIST_HEAD(&adapter->rxnfc_rules);
+
+	adapter->addr = devm_platform_get_and_ioremap_resource(pdev, 0, &io);
+	if (IS_ERR(adapter->addr))
+		return PTR_ERR(adapter->addr);
+	netdev->mem_start = io->start;
+	netdev->mem_end = io->end;
+
+	type = ioread32(adapter->addr + ECM_TYPE);
+	revision = (type & ECM_REVISION_MASK) >> ECM_REVISION_SHIFT;
+	version = (type & ECM_VERSION_MASK) >> ECM_VERSION_SHIFT;
+	queue_count = (type & ECM_QUEUE_COUNT_MASK) >> ECM_QUEUE_COUNT_SHIFT;
+	adapter->gate_control = type & ECM_GATE_CONTROL;
+	adapter->rxnfc_max = TSNEP_RX_ASSIGN_ETHER_TYPE_COUNT;
+
+	tsnep_disable_irq(adapter, ECM_INT_ALL);
+
+	retval = tsnep_queue_init(adapter, queue_count);
+	if (retval)
+		return retval;
+
+	retval = dma_set_mask_and_coherent(&adapter->pdev->dev,
+					   DMA_BIT_MASK(64));
+	if (retval) {
+		dev_err(&adapter->pdev->dev, "no usable DMA configuration.\n");
+		return retval;
+	}
+
+	retval = tsnep_mac_init(adapter);
+	if (retval)
+		return retval;
+
+	retval = tsnep_mdio_init(adapter);
+	if (retval)
+		goto mdio_init_failed;
+
+	retval = tsnep_phy_init(adapter);
+	if (retval)
+		goto phy_init_failed;
+
+	retval = tsnep_ptp_init(adapter);
+	if (retval)
+		goto ptp_init_failed;
+
+	retval = tsnep_tc_init(adapter);
+	if (retval)
+		goto tc_init_failed;
+
+	retval = tsnep_rxnfc_init(adapter);
+	if (retval)
+		goto rxnfc_init_failed;
+
+	netdev->netdev_ops = &tsnep_netdev_ops;
+	netdev->ethtool_ops = &tsnep_ethtool_ops;
+	netdev->features = NETIF_F_SG;
+	netdev->hw_features = netdev->features | NETIF_F_LOOPBACK;
+
+	netdev->xdp_features = NETDEV_XDP_ACT_BASIC | NETDEV_XDP_ACT_REDIRECT |
+			       NETDEV_XDP_ACT_NDO_XMIT |
+			       NETDEV_XDP_ACT_NDO_XMIT_SG |
+			       NETDEV_XDP_ACT_XSK_ZEROCOPY;
+
+	/* carrier off reporting is important to ethtool even BEFORE open */
+	netif_carrier_off(netdev);
+
+	retval = register_netdev(netdev);
+	if (retval)
+		goto register_failed;
+
+	dev_info(&adapter->pdev->dev, "device version %d.%02d\n", version,
+		 revision);
+	if (adapter->gate_control)
+		dev_info(&adapter->pdev->dev, "gate control detected\n");
+
+	return 0;
+
+register_failed:
+	tsnep_rxnfc_cleanup(adapter);
+rxnfc_init_failed:
+	tsnep_tc_cleanup(adapter);
+tc_init_failed:
+	tsnep_ptp_cleanup(adapter);
+ptp_init_failed:
+phy_init_failed:
+	if (adapter->mdiobus)
+		mdiobus_unregister(adapter->mdiobus);
+mdio_init_failed:
+	return retval;
+}
+
+static void tsnep_remove(struct platform_device *pdev)
+{
+	struct tsnep_adapter *adapter = platform_get_drvdata(pdev);
+
+	unregister_netdev(adapter->netdev);
+
+	tsnep_rxnfc_cleanup(adapter);
+
+	tsnep_tc_cleanup(adapter);
+
+	tsnep_ptp_cleanup(adapter);
+
+	if (adapter->mdiobus)
+		mdiobus_unregister(adapter->mdiobus);
+
+	tsnep_disable_irq(adapter, ECM_INT_ALL);
+}
+
+static const struct of_device_id tsnep_of_match[] = {
+	{ .compatible = "engleder,tsnep", },
+{ },
+};
+MODULE_DEVICE_TABLE(of, tsnep_of_match);
+
+static struct platform_driver tsnep_driver = {
+	.driver = {
+		.name = TSNEP,
+		.of_match_table = tsnep_of_match,
+	},
+	.probe = tsnep_probe,
+	.remove = tsnep_remove,
+};
+module_platform_driver(tsnep_driver);
+
+MODULE_AUTHOR("Gerhard Engleder <gerhard@engleder-embedded.com>");
+MODULE_DESCRIPTION("TSN endpoint Ethernet MAC driver");
+MODULE_LICENSE("GPL");
diff --git a/drivers/net/ethernet/engleder/tsnep_ptp.c b/drivers/net/ethernet/engleder/tsnep_ptp.c
new file mode 100644
index 000000000000..54fbf0126815
--- /dev/null
+++ b/drivers/net/ethernet/engleder/tsnep_ptp.c
@@ -0,0 +1,246 @@
+// SPDX-License-Identifier: GPL-2.0
+/* Copyright (C) 2021 Gerhard Engleder <gerhard@engleder-embedded.com> */
+
+#include "tsnep.h"
+
+void tsnep_get_system_time(struct tsnep_adapter *adapter, u64 *time)
+{
+	u32 high_before;
+	u32 low;
+	u32 high;
+
+	/* read high dword twice to detect overrun */
+	high = ioread32(adapter->addr + ECM_SYSTEM_TIME_HIGH);
+	do {
+		low = ioread32(adapter->addr + ECM_SYSTEM_TIME_LOW);
+		high_before = high;
+		high = ioread32(adapter->addr + ECM_SYSTEM_TIME_HIGH);
+	} while (high != high_before);
+	*time = (((u64)high) << 32) | ((u64)low);
+}
+
+int tsnep_ptp_ioctl(struct net_device *netdev, struct ifreq *ifr, int cmd)
+{
+	struct tsnep_adapter *adapter = netdev_priv(netdev);
+	struct hwtstamp_config config;
+
+	if (!ifr)
+		return -EINVAL;
+
+	if (cmd == SIOCSHWTSTAMP) {
+		if (copy_from_user(&config, ifr->ifr_data, sizeof(config)))
+			return -EFAULT;
+
+		switch (config.tx_type) {
+		case HWTSTAMP_TX_OFF:
+		case HWTSTAMP_TX_ON:
+			break;
+		default:
+			return -ERANGE;
+		}
+
+		switch (config.rx_filter) {
+		case HWTSTAMP_FILTER_NONE:
+			break;
+		case HWTSTAMP_FILTER_ALL:
+		case HWTSTAMP_FILTER_PTP_V1_L4_EVENT:
+		case HWTSTAMP_FILTER_PTP_V1_L4_SYNC:
+		case HWTSTAMP_FILTER_PTP_V1_L4_DELAY_REQ:
+		case HWTSTAMP_FILTER_PTP_V2_L4_EVENT:
+		case HWTSTAMP_FILTER_PTP_V2_L4_SYNC:
+		case HWTSTAMP_FILTER_PTP_V2_L4_DELAY_REQ:
+		case HWTSTAMP_FILTER_PTP_V2_L2_EVENT:
+		case HWTSTAMP_FILTER_PTP_V2_L2_SYNC:
+		case HWTSTAMP_FILTER_PTP_V2_L2_DELAY_REQ:
+		case HWTSTAMP_FILTER_PTP_V2_EVENT:
+		case HWTSTAMP_FILTER_PTP_V2_SYNC:
+		case HWTSTAMP_FILTER_PTP_V2_DELAY_REQ:
+		case HWTSTAMP_FILTER_NTP_ALL:
+			config.rx_filter = HWTSTAMP_FILTER_ALL;
+			break;
+		default:
+			return -ERANGE;
+		}
+
+		memcpy(&adapter->hwtstamp_config, &config,
+		       sizeof(adapter->hwtstamp_config));
+	}
+
+	if (copy_to_user(ifr->ifr_data, &adapter->hwtstamp_config,
+			 sizeof(adapter->hwtstamp_config)))
+		return -EFAULT;
+
+	return 0;
+}
+
+static int tsnep_ptp_adjfine(struct ptp_clock_info *ptp, long scaled_ppm)
+{
+	struct tsnep_adapter *adapter = container_of(ptp, struct tsnep_adapter,
+						     ptp_clock_info);
+	bool negative = false;
+	u64 rate_offset;
+
+	if (scaled_ppm < 0) {
+		scaled_ppm = -scaled_ppm;
+		negative = true;
+	}
+
+	/* convert from 16 bit to 32 bit binary fractional, divide by 1000000 to
+	 * eliminate ppm, multiply with 8 to compensate 8ns clock cycle time,
+	 * simplify calculation because 15625 * 8 = 1000000 / 8
+	 */
+	rate_offset = scaled_ppm;
+	rate_offset <<= 16 - 3;
+	rate_offset = div_u64(rate_offset, 15625);
+
+	rate_offset &= ECM_CLOCK_RATE_OFFSET_MASK;
+	if (negative)
+		rate_offset |= ECM_CLOCK_RATE_OFFSET_SIGN;
+	iowrite32(rate_offset & 0xFFFFFFFF, adapter->addr + ECM_CLOCK_RATE);
+
+	return 0;
+}
+
+static int tsnep_ptp_adjtime(struct ptp_clock_info *ptp, s64 delta)
+{
+	struct tsnep_adapter *adapter = container_of(ptp, struct tsnep_adapter,
+						     ptp_clock_info);
+	u64 system_time;
+	unsigned long flags;
+
+	spin_lock_irqsave(&adapter->ptp_lock, flags);
+
+	tsnep_get_system_time(adapter, &system_time);
+
+	system_time += delta;
+
+	/* high dword is buffered in hardware and synchronously written to
+	 * system time when low dword is written
+	 */
+	iowrite32(system_time >> 32, adapter->addr + ECM_SYSTEM_TIME_HIGH);
+	iowrite32(system_time & 0xFFFFFFFF,
+		  adapter->addr + ECM_SYSTEM_TIME_LOW);
+
+	spin_unlock_irqrestore(&adapter->ptp_lock, flags);
+
+	return 0;
+}
+
+static int tsnep_ptp_gettimex64(struct ptp_clock_info *ptp,
+				struct timespec64 *ts,
+				struct ptp_system_timestamp *sts)
+{
+	struct tsnep_adapter *adapter = container_of(ptp, struct tsnep_adapter,
+						     ptp_clock_info);
+	u32 high_before;
+	u32 low;
+	u32 high;
+	u64 system_time;
+
+	/* read high dword twice to detect overrun */
+	high = ioread32(adapter->addr + ECM_SYSTEM_TIME_HIGH);
+	do {
+		ptp_read_system_prets(sts);
+		low = ioread32(adapter->addr + ECM_SYSTEM_TIME_LOW);
+		ptp_read_system_postts(sts);
+		high_before = high;
+		high = ioread32(adapter->addr + ECM_SYSTEM_TIME_HIGH);
+	} while (high != high_before);
+	system_time = (((u64)high) << 32) | ((u64)low);
+
+	*ts = ns_to_timespec64(system_time);
+
+	return 0;
+}
+
+static int tsnep_ptp_settime64(struct ptp_clock_info *ptp,
+			       const struct timespec64 *ts)
+{
+	struct tsnep_adapter *adapter = container_of(ptp, struct tsnep_adapter,
+						     ptp_clock_info);
+	u64 system_time = timespec64_to_ns(ts);
+	unsigned long flags;
+
+	spin_lock_irqsave(&adapter->ptp_lock, flags);
+
+	/* high dword is buffered in hardware and synchronously written to
+	 * system time when low dword is written
+	 */
+	iowrite32(system_time >> 32, adapter->addr + ECM_SYSTEM_TIME_HIGH);
+	iowrite32(system_time & 0xFFFFFFFF,
+		  adapter->addr + ECM_SYSTEM_TIME_LOW);
+
+	spin_unlock_irqrestore(&adapter->ptp_lock, flags);
+
+	return 0;
+}
+
+static int tsnep_ptp_getcyclesx64(struct ptp_clock_info *ptp,
+				  struct timespec64 *ts,
+				  struct ptp_system_timestamp *sts)
+{
+	struct tsnep_adapter *adapter = container_of(ptp, struct tsnep_adapter,
+						     ptp_clock_info);
+	u32 high_before;
+	u32 low;
+	u32 high;
+	u64 counter;
+
+	/* read high dword twice to detect overrun */
+	high = ioread32(adapter->addr + ECM_COUNTER_HIGH);
+	do {
+		ptp_read_system_prets(sts);
+		low = ioread32(adapter->addr + ECM_COUNTER_LOW);
+		ptp_read_system_postts(sts);
+		high_before = high;
+		high = ioread32(adapter->addr + ECM_COUNTER_HIGH);
+	} while (high != high_before);
+	counter = (((u64)high) << 32) | ((u64)low);
+
+	*ts = ns_to_timespec64(counter);
+
+	return 0;
+}
+
+int tsnep_ptp_init(struct tsnep_adapter *adapter)
+{
+	int retval = 0;
+
+	adapter->hwtstamp_config.rx_filter = HWTSTAMP_FILTER_NONE;
+	adapter->hwtstamp_config.tx_type = HWTSTAMP_TX_OFF;
+
+	snprintf(adapter->ptp_clock_info.name, 16, "%s", TSNEP);
+	adapter->ptp_clock_info.owner = THIS_MODULE;
+	/* at most 2^-1ns adjustment every clock cycle for 8ns clock cycle time,
+	 * stay slightly below because only bits below 2^-1ns are supported
+	 */
+	adapter->ptp_clock_info.max_adj = (500000000 / 8 - 1);
+	adapter->ptp_clock_info.adjfine = tsnep_ptp_adjfine;
+	adapter->ptp_clock_info.adjtime = tsnep_ptp_adjtime;
+	adapter->ptp_clock_info.gettimex64 = tsnep_ptp_gettimex64;
+	adapter->ptp_clock_info.settime64 = tsnep_ptp_settime64;
+	adapter->ptp_clock_info.getcyclesx64 = tsnep_ptp_getcyclesx64;
+
+	spin_lock_init(&adapter->ptp_lock);
+
+	adapter->ptp_clock = ptp_clock_register(&adapter->ptp_clock_info,
+						&adapter->pdev->dev);
+	if (IS_ERR(adapter->ptp_clock)) {
+		netdev_err(adapter->netdev, "ptp_clock_register failed\n");
+
+		retval = PTR_ERR(adapter->ptp_clock);
+		adapter->ptp_clock = NULL;
+	} else if (adapter->ptp_clock) {
+		netdev_info(adapter->netdev, "PHC added\n");
+	}
+
+	return retval;
+}
+
+void tsnep_ptp_cleanup(struct tsnep_adapter *adapter)
+{
+	if (adapter->ptp_clock) {
+		ptp_clock_unregister(adapter->ptp_clock);
+		netdev_info(adapter->netdev, "PHC removed\n");
+	}
+}
diff --git a/drivers/net/ethernet/engleder/tsnep_rxnfc.c b/drivers/net/ethernet/engleder/tsnep_rxnfc.c
new file mode 100644
index 000000000000..9ac2a0cf3833
--- /dev/null
+++ b/drivers/net/ethernet/engleder/tsnep_rxnfc.c
@@ -0,0 +1,307 @@
+// SPDX-License-Identifier: GPL-2.0
+/* Copyright (C) 2022 Gerhard Engleder <gerhard@engleder-embedded.com> */
+
+#include "tsnep.h"
+
+#define ETHER_TYPE_FULL_MASK ((__force __be16)~0)
+
+static void tsnep_enable_rule(struct tsnep_adapter *adapter,
+			      struct tsnep_rxnfc_rule *rule)
+{
+	u8 rx_assign;
+	void __iomem *addr;
+
+	rx_assign = TSNEP_RX_ASSIGN_ACTIVE;
+	rx_assign |= (rule->queue_index << TSNEP_RX_ASSIGN_QUEUE_SHIFT) &
+		     TSNEP_RX_ASSIGN_QUEUE_MASK;
+
+	addr = adapter->addr + TSNEP_RX_ASSIGN_ETHER_TYPE +
+	       TSNEP_RX_ASSIGN_ETHER_TYPE_OFFSET * rule->location;
+	iowrite16(rule->filter.ether_type, addr);
+
+	/* enable rule after all settings are done */
+	addr = adapter->addr + TSNEP_RX_ASSIGN +
+	       TSNEP_RX_ASSIGN_OFFSET * rule->location;
+	iowrite8(rx_assign, addr);
+}
+
+static void tsnep_disable_rule(struct tsnep_adapter *adapter,
+			       struct tsnep_rxnfc_rule *rule)
+{
+	void __iomem *addr;
+
+	addr = adapter->addr + TSNEP_RX_ASSIGN +
+	       TSNEP_RX_ASSIGN_OFFSET * rule->location;
+	iowrite8(0, addr);
+}
+
+static struct tsnep_rxnfc_rule *tsnep_get_rule(struct tsnep_adapter *adapter,
+					       int location)
+{
+	struct tsnep_rxnfc_rule *rule;
+
+	list_for_each_entry(rule, &adapter->rxnfc_rules, list) {
+		if (rule->location == location)
+			return rule;
+		if (rule->location > location)
+			break;
+	}
+
+	return NULL;
+}
+
+static void tsnep_add_rule(struct tsnep_adapter *adapter,
+			   struct tsnep_rxnfc_rule *rule)
+{
+	struct tsnep_rxnfc_rule *pred, *cur;
+
+	tsnep_enable_rule(adapter, rule);
+
+	pred = NULL;
+	list_for_each_entry(cur, &adapter->rxnfc_rules, list) {
+		if (cur->location >= rule->location)
+			break;
+		pred = cur;
+	}
+
+	list_add(&rule->list, pred ? &pred->list : &adapter->rxnfc_rules);
+	adapter->rxnfc_count++;
+}
+
+static void tsnep_delete_rule(struct tsnep_adapter *adapter,
+			      struct tsnep_rxnfc_rule *rule)
+{
+	tsnep_disable_rule(adapter, rule);
+
+	list_del(&rule->list);
+	adapter->rxnfc_count--;
+
+	kfree(rule);
+}
+
+static void tsnep_flush_rules(struct tsnep_adapter *adapter)
+{
+	struct tsnep_rxnfc_rule *rule, *tmp;
+
+	mutex_lock(&adapter->rxnfc_lock);
+
+	list_for_each_entry_safe(rule, tmp, &adapter->rxnfc_rules, list)
+		tsnep_delete_rule(adapter, rule);
+
+	mutex_unlock(&adapter->rxnfc_lock);
+}
+
+int tsnep_rxnfc_get_rule(struct tsnep_adapter *adapter,
+			 struct ethtool_rxnfc *cmd)
+{
+	struct ethtool_rx_flow_spec *fsp = &cmd->fs;
+	struct tsnep_rxnfc_rule *rule = NULL;
+
+	cmd->data = adapter->rxnfc_max;
+
+	mutex_lock(&adapter->rxnfc_lock);
+
+	rule = tsnep_get_rule(adapter, fsp->location);
+	if (!rule) {
+		mutex_unlock(&adapter->rxnfc_lock);
+
+		return -ENOENT;
+	}
+
+	fsp->flow_type = ETHER_FLOW;
+	fsp->ring_cookie = rule->queue_index;
+
+	if (rule->filter.type == TSNEP_RXNFC_ETHER_TYPE) {
+		fsp->h_u.ether_spec.h_proto = htons(rule->filter.ether_type);
+		fsp->m_u.ether_spec.h_proto = ETHER_TYPE_FULL_MASK;
+	}
+
+	mutex_unlock(&adapter->rxnfc_lock);
+
+	return 0;
+}
+
+int tsnep_rxnfc_get_all(struct tsnep_adapter *adapter,
+			struct ethtool_rxnfc *cmd,
+			u32 *rule_locs)
+{
+	struct tsnep_rxnfc_rule *rule;
+	int count = 0;
+
+	cmd->data = adapter->rxnfc_max;
+
+	mutex_lock(&adapter->rxnfc_lock);
+
+	list_for_each_entry(rule, &adapter->rxnfc_rules, list) {
+		if (count == cmd->rule_cnt) {
+			mutex_unlock(&adapter->rxnfc_lock);
+
+			return -EMSGSIZE;
+		}
+
+		rule_locs[count] = rule->location;
+		count++;
+	}
+
+	mutex_unlock(&adapter->rxnfc_lock);
+
+	cmd->rule_cnt = count;
+
+	return 0;
+}
+
+static int tsnep_rxnfc_find_location(struct tsnep_adapter *adapter)
+{
+	struct tsnep_rxnfc_rule *tmp;
+	int location = 0;
+
+	list_for_each_entry(tmp, &adapter->rxnfc_rules, list) {
+		if (tmp->location == location)
+			location++;
+		else
+			return location;
+	}
+
+	if (location >= adapter->rxnfc_max)
+		return -ENOSPC;
+
+	return location;
+}
+
+static void tsnep_rxnfc_init_rule(struct tsnep_rxnfc_rule *rule,
+				  const struct ethtool_rx_flow_spec *fsp)
+{
+	INIT_LIST_HEAD(&rule->list);
+
+	rule->queue_index = fsp->ring_cookie;
+	rule->location = fsp->location;
+
+	rule->filter.type = TSNEP_RXNFC_ETHER_TYPE;
+	rule->filter.ether_type = ntohs(fsp->h_u.ether_spec.h_proto);
+}
+
+static int tsnep_rxnfc_check_rule(struct tsnep_adapter *adapter,
+				  struct tsnep_rxnfc_rule *rule)
+{
+	struct net_device *dev = adapter->netdev;
+	struct tsnep_rxnfc_rule *tmp;
+
+	list_for_each_entry(tmp, &adapter->rxnfc_rules, list) {
+		if (!memcmp(&rule->filter, &tmp->filter, sizeof(rule->filter)) &&
+		    tmp->location != rule->location) {
+			netdev_dbg(dev, "rule already exists\n");
+
+			return -EEXIST;
+		}
+	}
+
+	return 0;
+}
+
+int tsnep_rxnfc_add_rule(struct tsnep_adapter *adapter,
+			 struct ethtool_rxnfc *cmd)
+{
+	struct net_device *netdev = adapter->netdev;
+	struct ethtool_rx_flow_spec *fsp =
+		(struct ethtool_rx_flow_spec *)&cmd->fs;
+	struct tsnep_rxnfc_rule *rule, *old_rule;
+	int retval;
+
+	/* only EtherType is supported */
+	if (fsp->flow_type != ETHER_FLOW ||
+	    !is_zero_ether_addr(fsp->m_u.ether_spec.h_dest) ||
+	    !is_zero_ether_addr(fsp->m_u.ether_spec.h_source) ||
+	    fsp->m_u.ether_spec.h_proto != ETHER_TYPE_FULL_MASK) {
+		netdev_dbg(netdev, "only ethernet protocol is supported\n");
+
+		return -EOPNOTSUPP;
+	}
+
+	if (fsp->ring_cookie >
+	    (TSNEP_RX_ASSIGN_QUEUE_MASK >> TSNEP_RX_ASSIGN_QUEUE_SHIFT)) {
+		netdev_dbg(netdev, "invalid action\n");
+
+		return -EINVAL;
+	}
+
+	if (fsp->location != RX_CLS_LOC_ANY &&
+	    fsp->location >= adapter->rxnfc_max) {
+		netdev_dbg(netdev, "invalid location\n");
+
+		return -EINVAL;
+	}
+
+	rule = kzalloc(sizeof(*rule), GFP_KERNEL);
+	if (!rule)
+		return -ENOMEM;
+
+	mutex_lock(&adapter->rxnfc_lock);
+
+	if (fsp->location == RX_CLS_LOC_ANY) {
+		retval = tsnep_rxnfc_find_location(adapter);
+		if (retval < 0)
+			goto failed;
+		fsp->location = retval;
+	}
+
+	tsnep_rxnfc_init_rule(rule, fsp);
+
+	retval = tsnep_rxnfc_check_rule(adapter, rule);
+	if (retval)
+		goto failed;
+
+	old_rule = tsnep_get_rule(adapter, fsp->location);
+	if (old_rule)
+		tsnep_delete_rule(adapter, old_rule);
+
+	tsnep_add_rule(adapter, rule);
+
+	mutex_unlock(&adapter->rxnfc_lock);
+
+	return 0;
+
+failed:
+	mutex_unlock(&adapter->rxnfc_lock);
+	kfree(rule);
+	return retval;
+}
+
+int tsnep_rxnfc_del_rule(struct tsnep_adapter *adapter,
+			 struct ethtool_rxnfc *cmd)
+{
+	struct ethtool_rx_flow_spec *fsp =
+		(struct ethtool_rx_flow_spec *)&cmd->fs;
+	struct tsnep_rxnfc_rule *rule;
+
+	mutex_lock(&adapter->rxnfc_lock);
+
+	rule = tsnep_get_rule(adapter, fsp->location);
+	if (!rule) {
+		mutex_unlock(&adapter->rxnfc_lock);
+
+		return -ENOENT;
+	}
+
+	tsnep_delete_rule(adapter, rule);
+
+	mutex_unlock(&adapter->rxnfc_lock);
+
+	return 0;
+}
+
+int tsnep_rxnfc_init(struct tsnep_adapter *adapter)
+{
+	int i;
+
+	/* disable all rules */
+	for (i = 0; i < adapter->rxnfc_max;
+	     i += sizeof(u32) / TSNEP_RX_ASSIGN_OFFSET)
+		iowrite32(0, adapter->addr + TSNEP_RX_ASSIGN + i);
+
+	return 0;
+}
+
+void tsnep_rxnfc_cleanup(struct tsnep_adapter *adapter)
+{
+	tsnep_flush_rules(adapter);
+}
diff --git a/drivers/net/ethernet/engleder/tsnep_selftests.c b/drivers/net/ethernet/engleder/tsnep_selftests.c
new file mode 100644
index 000000000000..8a9145f93147
--- /dev/null
+++ b/drivers/net/ethernet/engleder/tsnep_selftests.c
@@ -0,0 +1,811 @@
+// SPDX-License-Identifier: GPL-2.0
+/* Copyright (C) 2021 Gerhard Engleder <gerhard@engleder-embedded.com> */
+
+#include "tsnep.h"
+
+#include <net/pkt_sched.h>
+
+enum tsnep_test {
+	TSNEP_TEST_ENABLE = 0,
+	TSNEP_TEST_TAPRIO,
+	TSNEP_TEST_TAPRIO_CHANGE,
+	TSNEP_TEST_TAPRIO_EXTENSION,
+};
+
+static const char tsnep_test_strings[][ETH_GSTRING_LEN] = {
+	"Enable timeout        (offline)",
+	"TAPRIO                (offline)",
+	"TAPRIO change         (offline)",
+	"TAPRIO extension      (offline)",
+};
+
+#define TSNEP_TEST_COUNT (sizeof(tsnep_test_strings) / ETH_GSTRING_LEN)
+
+static bool enable_gc_timeout(struct tsnep_adapter *adapter)
+{
+	iowrite8(TSNEP_GC_ENABLE_TIMEOUT, adapter->addr + TSNEP_GC);
+	if (!(ioread32(adapter->addr + TSNEP_GC) & TSNEP_GC_TIMEOUT_ACTIVE))
+		return false;
+
+	return true;
+}
+
+static bool gc_timeout_signaled(struct tsnep_adapter *adapter)
+{
+	if (ioread32(adapter->addr + TSNEP_GC) & TSNEP_GC_TIMEOUT_SIGNAL)
+		return true;
+
+	return false;
+}
+
+static bool ack_gc_timeout(struct tsnep_adapter *adapter)
+{
+	iowrite8(TSNEP_GC_ENABLE_TIMEOUT, adapter->addr + TSNEP_GC);
+	if (ioread32(adapter->addr + TSNEP_GC) &
+	    (TSNEP_GC_TIMEOUT_ACTIVE | TSNEP_GC_TIMEOUT_SIGNAL))
+		return false;
+	return true;
+}
+
+static bool enable_gc(struct tsnep_adapter *adapter, bool a)
+{
+	u8 enable;
+	u8 active;
+
+	if (a) {
+		enable = TSNEP_GC_ENABLE_A;
+		active = TSNEP_GC_ACTIVE_A;
+	} else {
+		enable = TSNEP_GC_ENABLE_B;
+		active = TSNEP_GC_ACTIVE_B;
+	}
+
+	iowrite8(enable, adapter->addr + TSNEP_GC);
+	if (!(ioread32(adapter->addr + TSNEP_GC) & active))
+		return false;
+
+	return true;
+}
+
+static bool disable_gc(struct tsnep_adapter *adapter)
+{
+	iowrite8(TSNEP_GC_DISABLE, adapter->addr + TSNEP_GC);
+	if (ioread32(adapter->addr + TSNEP_GC) &
+	    (TSNEP_GC_ACTIVE_A | TSNEP_GC_ACTIVE_B))
+		return false;
+
+	return true;
+}
+
+static bool gc_delayed_enable(struct tsnep_adapter *adapter, bool a, int delay)
+{
+	u64 before, after;
+	u32 time;
+	bool enabled;
+
+	if (!disable_gc(adapter))
+		return false;
+
+	before = ktime_get_ns();
+
+	if (!enable_gc_timeout(adapter))
+		return false;
+
+	/* for start time after timeout, the timeout can guarantee, that enable
+	 * is blocked if too late
+	 */
+	time = ioread32(adapter->addr + ECM_SYSTEM_TIME_LOW);
+	time += TSNEP_GC_TIMEOUT;
+	iowrite32(time, adapter->addr + TSNEP_GC_TIME);
+
+	ndelay(delay);
+
+	enabled = enable_gc(adapter, a);
+	after = ktime_get_ns();
+
+	if (delay > TSNEP_GC_TIMEOUT) {
+		/* timeout must have blocked enable */
+		if (enabled)
+			return false;
+	} else if ((after - before) < TSNEP_GC_TIMEOUT * 14 / 16) {
+		/* timeout must not have blocked enable */
+		if (!enabled)
+			return false;
+	}
+
+	if (enabled) {
+		if (gc_timeout_signaled(adapter))
+			return false;
+	} else {
+		if (!gc_timeout_signaled(adapter))
+			return false;
+		if (!ack_gc_timeout(adapter))
+			return false;
+	}
+
+	if (!disable_gc(adapter))
+		return false;
+
+	return true;
+}
+
+static bool tsnep_test_gc_enable(struct tsnep_adapter *adapter)
+{
+	int i;
+
+	iowrite32(0x80000001, adapter->addr + TSNEP_GCL_A + 0);
+	iowrite32(100000, adapter->addr + TSNEP_GCL_A + 4);
+
+	for (i = 0; i < 200000; i += 100) {
+		if (!gc_delayed_enable(adapter, true, i))
+			return false;
+	}
+
+	iowrite32(0x80000001, adapter->addr + TSNEP_GCL_B + 0);
+	iowrite32(100000, adapter->addr + TSNEP_GCL_B + 4);
+
+	for (i = 0; i < 200000; i += 100) {
+		if (!gc_delayed_enable(adapter, false, i))
+			return false;
+	}
+
+	return true;
+}
+
+static void delay_base_time(struct tsnep_adapter *adapter,
+			    struct tc_taprio_qopt_offload *qopt, s64 ms)
+{
+	u64 system_time;
+	u64 base_time = ktime_to_ns(qopt->base_time);
+	u64 n;
+
+	tsnep_get_system_time(adapter, &system_time);
+	system_time += ms * 1000000;
+	n = div64_u64(system_time - base_time, qopt->cycle_time);
+
+	qopt->base_time = ktime_add_ns(qopt->base_time,
+				       (n + 1) * qopt->cycle_time);
+}
+
+static void get_gate_state(struct tsnep_adapter *adapter, u32 *gc, u32 *gc_time,
+			   u64 *system_time)
+{
+	u32 time_high_before;
+	u32 time_low;
+	u32 time_high;
+	u32 gc_time_before;
+
+	time_high = ioread32(adapter->addr + ECM_SYSTEM_TIME_HIGH);
+	*gc_time = ioread32(adapter->addr + TSNEP_GC_TIME);
+	do {
+		time_low = ioread32(adapter->addr + ECM_SYSTEM_TIME_LOW);
+		*gc = ioread32(adapter->addr + TSNEP_GC);
+
+		gc_time_before = *gc_time;
+		*gc_time = ioread32(adapter->addr + TSNEP_GC_TIME);
+		time_high_before = time_high;
+		time_high = ioread32(adapter->addr + ECM_SYSTEM_TIME_HIGH);
+	} while ((time_high != time_high_before) ||
+		 (*gc_time != gc_time_before));
+
+	*system_time = (((u64)time_high) << 32) | ((u64)time_low);
+}
+
+static int get_operation(struct tsnep_gcl *gcl, u64 system_time, u64 *next)
+{
+	u64 n = div64_u64(system_time - gcl->base_time, gcl->cycle_time);
+	u64 cycle_start = gcl->base_time + gcl->cycle_time * n;
+	int i;
+
+	*next = cycle_start;
+	for (i = 0; i < gcl->count; i++) {
+		*next += gcl->operation[i].interval;
+		if (*next > system_time)
+			break;
+	}
+
+	return i;
+}
+
+static bool check_gate(struct tsnep_adapter *adapter)
+{
+	u32 gc_time;
+	u32 gc;
+	u64 system_time;
+	struct tsnep_gcl *curr;
+	struct tsnep_gcl *prev;
+	u64 next_time;
+	u8 gate_open;
+	u8 next_gate_open;
+
+	get_gate_state(adapter, &gc, &gc_time, &system_time);
+
+	if (gc & TSNEP_GC_ACTIVE_A) {
+		curr = &adapter->gcl[0];
+		prev = &adapter->gcl[1];
+	} else if (gc & TSNEP_GC_ACTIVE_B) {
+		curr = &adapter->gcl[1];
+		prev = &adapter->gcl[0];
+	} else {
+		return false;
+	}
+	if (curr->start_time <= system_time) {
+		/* GCL is already active */
+		int index;
+
+		index = get_operation(curr, system_time, &next_time);
+		gate_open = curr->operation[index].properties & TSNEP_GCL_MASK;
+		if (index == curr->count - 1)
+			index = 0;
+		else
+			index++;
+		next_gate_open =
+			curr->operation[index].properties & TSNEP_GCL_MASK;
+	} else if (curr->change) {
+		/* operation of previous GCL is active */
+		int index;
+		u64 start_before;
+		u64 n;
+
+		index = get_operation(prev, system_time, &next_time);
+		next_time = curr->start_time;
+		start_before = prev->base_time;
+		n = div64_u64(curr->start_time - start_before,
+			      prev->cycle_time);
+		start_before += n * prev->cycle_time;
+		if (curr->start_time == start_before)
+			start_before -= prev->cycle_time;
+		if (((start_before + prev->cycle_time_extension) >=
+		     curr->start_time) &&
+		    (curr->start_time - prev->cycle_time_extension <=
+		     system_time)) {
+			/* extend */
+			index = prev->count - 1;
+		}
+		gate_open = prev->operation[index].properties & TSNEP_GCL_MASK;
+		next_gate_open =
+			curr->operation[0].properties & TSNEP_GCL_MASK;
+	} else {
+		/* GCL is waiting for start */
+		next_time = curr->start_time;
+		gate_open = 0xFF;
+		next_gate_open = curr->operation[0].properties & TSNEP_GCL_MASK;
+	}
+
+	if (gc_time != (next_time & 0xFFFFFFFF)) {
+		dev_err(&adapter->pdev->dev, "gate control time 0x%x!=0x%llx\n",
+			gc_time, next_time);
+		return false;
+	}
+	if (((gc & TSNEP_GC_OPEN) >> TSNEP_GC_OPEN_SHIFT) != gate_open) {
+		dev_err(&adapter->pdev->dev,
+			"gate control open 0x%02x!=0x%02x\n",
+			((gc & TSNEP_GC_OPEN) >> TSNEP_GC_OPEN_SHIFT),
+			gate_open);
+		return false;
+	}
+	if (((gc & TSNEP_GC_NEXT_OPEN) >> TSNEP_GC_NEXT_OPEN_SHIFT) !=
+	    next_gate_open) {
+		dev_err(&adapter->pdev->dev,
+			"gate control next open 0x%02x!=0x%02x\n",
+			((gc & TSNEP_GC_NEXT_OPEN) >> TSNEP_GC_NEXT_OPEN_SHIFT),
+			next_gate_open);
+		return false;
+	}
+
+	return true;
+}
+
+static bool check_gate_duration(struct tsnep_adapter *adapter, s64 ms)
+{
+	ktime_t start = ktime_get();
+
+	do {
+		if (!check_gate(adapter))
+			return false;
+	} while (ktime_ms_delta(ktime_get(), start) < ms);
+
+	return true;
+}
+
+static bool enable_check_taprio(struct tsnep_adapter *adapter,
+				struct tc_taprio_qopt_offload *qopt, s64 ms)
+{
+	int retval;
+
+	retval = tsnep_tc_setup(adapter->netdev, TC_SETUP_QDISC_TAPRIO, qopt);
+	if (retval)
+		return false;
+
+	if (!check_gate_duration(adapter, ms))
+		return false;
+
+	return true;
+}
+
+static bool disable_taprio(struct tsnep_adapter *adapter)
+{
+	struct tc_taprio_qopt_offload qopt;
+	int retval;
+
+	memset(&qopt, 0, sizeof(qopt));
+	qopt.cmd = TAPRIO_CMD_DESTROY;
+	retval = tsnep_tc_setup(adapter->netdev, TC_SETUP_QDISC_TAPRIO, &qopt);
+	if (retval)
+		return false;
+
+	return true;
+}
+
+static bool run_taprio(struct tsnep_adapter *adapter,
+		       struct tc_taprio_qopt_offload *qopt, s64 ms)
+{
+	if (!enable_check_taprio(adapter, qopt, ms))
+		return false;
+
+	if (!disable_taprio(adapter))
+		return false;
+
+	return true;
+}
+
+static bool tsnep_test_taprio(struct tsnep_adapter *adapter)
+{
+	struct tc_taprio_qopt_offload *qopt;
+	int i;
+
+	qopt = kzalloc(struct_size(qopt, entries, 255), GFP_KERNEL);
+	if (!qopt)
+		return false;
+	for (i = 0; i < 255; i++)
+		qopt->entries[i].command = TC_TAPRIO_CMD_SET_GATES;
+
+	qopt->cmd = TAPRIO_CMD_REPLACE;
+	qopt->base_time = ktime_set(0, 0);
+	qopt->cycle_time = 1500000;
+	qopt->cycle_time_extension = 0;
+	qopt->entries[0].gate_mask = 0x02;
+	qopt->entries[0].interval = 200000;
+	qopt->entries[1].gate_mask = 0x03;
+	qopt->entries[1].interval = 800000;
+	qopt->entries[2].gate_mask = 0x07;
+	qopt->entries[2].interval = 240000;
+	qopt->entries[3].gate_mask = 0x01;
+	qopt->entries[3].interval = 80000;
+	qopt->entries[4].gate_mask = 0x04;
+	qopt->entries[4].interval = 70000;
+	qopt->entries[5].gate_mask = 0x06;
+	qopt->entries[5].interval = 60000;
+	qopt->entries[6].gate_mask = 0x0F;
+	qopt->entries[6].interval = 50000;
+	qopt->num_entries = 7;
+	if (!run_taprio(adapter, qopt, 100))
+		goto failed;
+
+	qopt->cmd = TAPRIO_CMD_REPLACE;
+	qopt->base_time = ktime_set(0, 0);
+	qopt->cycle_time = 411854;
+	qopt->cycle_time_extension = 0;
+	qopt->entries[0].gate_mask = 0x17;
+	qopt->entries[0].interval = 23842;
+	qopt->entries[1].gate_mask = 0x16;
+	qopt->entries[1].interval = 13482;
+	qopt->entries[2].gate_mask = 0x15;
+	qopt->entries[2].interval = 49428;
+	qopt->entries[3].gate_mask = 0x14;
+	qopt->entries[3].interval = 38189;
+	qopt->entries[4].gate_mask = 0x13;
+	qopt->entries[4].interval = 92321;
+	qopt->entries[5].gate_mask = 0x12;
+	qopt->entries[5].interval = 71239;
+	qopt->entries[6].gate_mask = 0x11;
+	qopt->entries[6].interval = 69932;
+	qopt->entries[7].gate_mask = 0x10;
+	qopt->entries[7].interval = 53421;
+	qopt->num_entries = 8;
+	if (!run_taprio(adapter, qopt, 100))
+		goto failed;
+
+	qopt->cmd = TAPRIO_CMD_REPLACE;
+	qopt->base_time = ktime_set(0, 0);
+	delay_base_time(adapter, qopt, 12);
+	qopt->cycle_time = 125000;
+	qopt->cycle_time_extension = 0;
+	qopt->entries[0].gate_mask = 0x27;
+	qopt->entries[0].interval = 15000;
+	qopt->entries[1].gate_mask = 0x26;
+	qopt->entries[1].interval = 15000;
+	qopt->entries[2].gate_mask = 0x25;
+	qopt->entries[2].interval = 12500;
+	qopt->entries[3].gate_mask = 0x24;
+	qopt->entries[3].interval = 17500;
+	qopt->entries[4].gate_mask = 0x23;
+	qopt->entries[4].interval = 10000;
+	qopt->entries[5].gate_mask = 0x22;
+	qopt->entries[5].interval = 11000;
+	qopt->entries[6].gate_mask = 0x21;
+	qopt->entries[6].interval = 9000;
+	qopt->entries[7].gate_mask = 0x20;
+	qopt->entries[7].interval = 10000;
+	qopt->entries[8].gate_mask = 0x20;
+	qopt->entries[8].interval = 12500;
+	qopt->entries[9].gate_mask = 0x20;
+	qopt->entries[9].interval = 12500;
+	qopt->num_entries = 10;
+	if (!run_taprio(adapter, qopt, 100))
+		goto failed;
+
+	kfree(qopt);
+
+	return true;
+
+failed:
+	disable_taprio(adapter);
+	kfree(qopt);
+
+	return false;
+}
+
+static bool tsnep_test_taprio_change(struct tsnep_adapter *adapter)
+{
+	struct tc_taprio_qopt_offload *qopt;
+	int i;
+
+	qopt = kzalloc(struct_size(qopt, entries, 255), GFP_KERNEL);
+	if (!qopt)
+		return false;
+	for (i = 0; i < 255; i++)
+		qopt->entries[i].command = TC_TAPRIO_CMD_SET_GATES;
+
+	qopt->cmd = TAPRIO_CMD_REPLACE;
+	qopt->base_time = ktime_set(0, 0);
+	qopt->cycle_time = 100000;
+	qopt->cycle_time_extension = 0;
+	qopt->entries[0].gate_mask = 0x30;
+	qopt->entries[0].interval = 20000;
+	qopt->entries[1].gate_mask = 0x31;
+	qopt->entries[1].interval = 80000;
+	qopt->num_entries = 2;
+	if (!enable_check_taprio(adapter, qopt, 100))
+		goto failed;
+
+	/* change to identical */
+	if (!enable_check_taprio(adapter, qopt, 100))
+		goto failed;
+	delay_base_time(adapter, qopt, 17);
+	if (!enable_check_taprio(adapter, qopt, 100))
+		goto failed;
+
+	/* change to same cycle time */
+	qopt->base_time = ktime_set(0, 0);
+	qopt->entries[0].gate_mask = 0x42;
+	qopt->entries[1].gate_mask = 0x43;
+	delay_base_time(adapter, qopt, 2);
+	if (!enable_check_taprio(adapter, qopt, 100))
+		goto failed;
+	qopt->base_time = ktime_set(0, 0);
+	qopt->entries[0].gate_mask = 0x54;
+	qopt->entries[0].interval = 33333;
+	qopt->entries[1].gate_mask = 0x55;
+	qopt->entries[1].interval = 66667;
+	delay_base_time(adapter, qopt, 23);
+	if (!enable_check_taprio(adapter, qopt, 100))
+		goto failed;
+	qopt->base_time = ktime_set(0, 0);
+	qopt->entries[0].gate_mask = 0x66;
+	qopt->entries[0].interval = 50000;
+	qopt->entries[1].gate_mask = 0x67;
+	qopt->entries[1].interval = 25000;
+	qopt->entries[2].gate_mask = 0x68;
+	qopt->entries[2].interval = 25000;
+	qopt->num_entries = 3;
+	delay_base_time(adapter, qopt, 11);
+	if (!enable_check_taprio(adapter, qopt, 100))
+		goto failed;
+
+	/* change to multiple of cycle time */
+	qopt->base_time = ktime_set(0, 0);
+	qopt->cycle_time = 200000;
+	qopt->entries[0].gate_mask = 0x79;
+	qopt->entries[0].interval = 50000;
+	qopt->entries[1].gate_mask = 0x7A;
+	qopt->entries[1].interval = 150000;
+	qopt->num_entries = 2;
+	delay_base_time(adapter, qopt, 11);
+	if (!enable_check_taprio(adapter, qopt, 100))
+		goto failed;
+	qopt->base_time = ktime_set(0, 0);
+	qopt->cycle_time = 1000000;
+	qopt->entries[0].gate_mask = 0x7B;
+	qopt->entries[0].interval = 125000;
+	qopt->entries[1].gate_mask = 0x7C;
+	qopt->entries[1].interval = 250000;
+	qopt->entries[2].gate_mask = 0x7D;
+	qopt->entries[2].interval = 375000;
+	qopt->entries[3].gate_mask = 0x7E;
+	qopt->entries[3].interval = 250000;
+	qopt->num_entries = 4;
+	delay_base_time(adapter, qopt, 3);
+	if (!enable_check_taprio(adapter, qopt, 100))
+		goto failed;
+
+	/* change to shorter cycle time */
+	qopt->base_time = ktime_set(0, 0);
+	qopt->cycle_time = 333333;
+	qopt->entries[0].gate_mask = 0x8F;
+	qopt->entries[0].interval = 166666;
+	qopt->entries[1].gate_mask = 0x80;
+	qopt->entries[1].interval = 166667;
+	qopt->num_entries = 2;
+	delay_base_time(adapter, qopt, 11);
+	if (!enable_check_taprio(adapter, qopt, 100))
+		goto failed;
+	qopt->base_time = ktime_set(0, 0);
+	qopt->cycle_time = 62500;
+	qopt->entries[0].gate_mask = 0x81;
+	qopt->entries[0].interval = 31250;
+	qopt->entries[1].gate_mask = 0x82;
+	qopt->entries[1].interval = 15625;
+	qopt->entries[2].gate_mask = 0x83;
+	qopt->entries[2].interval = 15625;
+	qopt->num_entries = 3;
+	delay_base_time(adapter, qopt, 1);
+	if (!enable_check_taprio(adapter, qopt, 100))
+		goto failed;
+
+	/* change to longer cycle time */
+	qopt->base_time = ktime_set(0, 0);
+	qopt->cycle_time = 400000;
+	qopt->entries[0].gate_mask = 0x84;
+	qopt->entries[0].interval = 100000;
+	qopt->entries[1].gate_mask = 0x85;
+	qopt->entries[1].interval = 100000;
+	qopt->entries[2].gate_mask = 0x86;
+	qopt->entries[2].interval = 100000;
+	qopt->entries[3].gate_mask = 0x87;
+	qopt->entries[3].interval = 100000;
+	qopt->num_entries = 4;
+	delay_base_time(adapter, qopt, 7);
+	if (!enable_check_taprio(adapter, qopt, 100))
+		goto failed;
+	qopt->base_time = ktime_set(0, 0);
+	qopt->cycle_time = 1700000;
+	qopt->entries[0].gate_mask = 0x88;
+	qopt->entries[0].interval = 200000;
+	qopt->entries[1].gate_mask = 0x89;
+	qopt->entries[1].interval = 300000;
+	qopt->entries[2].gate_mask = 0x8A;
+	qopt->entries[2].interval = 600000;
+	qopt->entries[3].gate_mask = 0x8B;
+	qopt->entries[3].interval = 100000;
+	qopt->entries[4].gate_mask = 0x8C;
+	qopt->entries[4].interval = 500000;
+	qopt->num_entries = 5;
+	delay_base_time(adapter, qopt, 6);
+	if (!enable_check_taprio(adapter, qopt, 100))
+		goto failed;
+
+	if (!disable_taprio(adapter))
+		goto failed;
+
+	kfree(qopt);
+
+	return true;
+
+failed:
+	disable_taprio(adapter);
+	kfree(qopt);
+
+	return false;
+}
+
+static bool tsnep_test_taprio_extension(struct tsnep_adapter *adapter)
+{
+	struct tc_taprio_qopt_offload *qopt;
+	int i;
+
+	qopt = kzalloc(struct_size(qopt, entries, 255), GFP_KERNEL);
+	if (!qopt)
+		return false;
+	for (i = 0; i < 255; i++)
+		qopt->entries[i].command = TC_TAPRIO_CMD_SET_GATES;
+
+	qopt->cmd = TAPRIO_CMD_REPLACE;
+	qopt->base_time = ktime_set(0, 0);
+	qopt->cycle_time = 100000;
+	qopt->cycle_time_extension = 50000;
+	qopt->entries[0].gate_mask = 0x90;
+	qopt->entries[0].interval = 20000;
+	qopt->entries[1].gate_mask = 0x91;
+	qopt->entries[1].interval = 80000;
+	qopt->num_entries = 2;
+	if (!enable_check_taprio(adapter, qopt, 100))
+		goto failed;
+
+	/* change to different phase */
+	qopt->base_time = ktime_set(0, 50000);
+	qopt->entries[0].gate_mask = 0x92;
+	qopt->entries[0].interval = 33000;
+	qopt->entries[1].gate_mask = 0x93;
+	qopt->entries[1].interval = 67000;
+	qopt->num_entries = 2;
+	delay_base_time(adapter, qopt, 2);
+	if (!enable_check_taprio(adapter, qopt, 100))
+		goto failed;
+
+	/* change to different phase and longer cycle time */
+	qopt->base_time = ktime_set(0, 0);
+	qopt->cycle_time = 1000000;
+	qopt->cycle_time_extension = 700000;
+	qopt->entries[0].gate_mask = 0x94;
+	qopt->entries[0].interval = 400000;
+	qopt->entries[1].gate_mask = 0x95;
+	qopt->entries[1].interval = 600000;
+	qopt->num_entries = 2;
+	delay_base_time(adapter, qopt, 7);
+	if (!enable_check_taprio(adapter, qopt, 100))
+		goto failed;
+	qopt->base_time = ktime_set(0, 700000);
+	qopt->cycle_time = 2000000;
+	qopt->cycle_time_extension = 1900000;
+	qopt->entries[0].gate_mask = 0x96;
+	qopt->entries[0].interval = 400000;
+	qopt->entries[1].gate_mask = 0x97;
+	qopt->entries[1].interval = 1600000;
+	qopt->num_entries = 2;
+	delay_base_time(adapter, qopt, 9);
+	if (!enable_check_taprio(adapter, qopt, 100))
+		goto failed;
+
+	/* change to different phase and shorter cycle time */
+	qopt->base_time = ktime_set(0, 0);
+	qopt->cycle_time = 1500000;
+	qopt->cycle_time_extension = 700000;
+	qopt->entries[0].gate_mask = 0x98;
+	qopt->entries[0].interval = 400000;
+	qopt->entries[1].gate_mask = 0x99;
+	qopt->entries[1].interval = 600000;
+	qopt->entries[2].gate_mask = 0x9A;
+	qopt->entries[2].interval = 500000;
+	qopt->num_entries = 3;
+	delay_base_time(adapter, qopt, 3);
+	if (!enable_check_taprio(adapter, qopt, 100))
+		goto failed;
+	qopt->base_time = ktime_set(0, 100000);
+	qopt->cycle_time = 500000;
+	qopt->cycle_time_extension = 300000;
+	qopt->entries[0].gate_mask = 0x9B;
+	qopt->entries[0].interval = 150000;
+	qopt->entries[1].gate_mask = 0x9C;
+	qopt->entries[1].interval = 350000;
+	qopt->num_entries = 2;
+	delay_base_time(adapter, qopt, 9);
+	if (!enable_check_taprio(adapter, qopt, 100))
+		goto failed;
+
+	/* change to different cycle time */
+	qopt->base_time = ktime_set(0, 0);
+	qopt->cycle_time = 1000000;
+	qopt->cycle_time_extension = 700000;
+	qopt->entries[0].gate_mask = 0xAD;
+	qopt->entries[0].interval = 400000;
+	qopt->entries[1].gate_mask = 0xAE;
+	qopt->entries[1].interval = 300000;
+	qopt->entries[2].gate_mask = 0xAF;
+	qopt->entries[2].interval = 300000;
+	qopt->num_entries = 3;
+	if (!enable_check_taprio(adapter, qopt, 100))
+		goto failed;
+	qopt->base_time = ktime_set(0, 0);
+	qopt->cycle_time = 400000;
+	qopt->cycle_time_extension = 100000;
+	qopt->entries[0].gate_mask = 0xA0;
+	qopt->entries[0].interval = 200000;
+	qopt->entries[1].gate_mask = 0xA1;
+	qopt->entries[1].interval = 200000;
+	qopt->num_entries = 2;
+	delay_base_time(adapter, qopt, 19);
+	if (!enable_check_taprio(adapter, qopt, 100))
+		goto failed;
+	qopt->base_time = ktime_set(0, 0);
+	qopt->cycle_time = 500000;
+	qopt->cycle_time_extension = 499999;
+	qopt->entries[0].gate_mask = 0xB2;
+	qopt->entries[0].interval = 100000;
+	qopt->entries[1].gate_mask = 0xB3;
+	qopt->entries[1].interval = 100000;
+	qopt->entries[2].gate_mask = 0xB4;
+	qopt->entries[2].interval = 100000;
+	qopt->entries[3].gate_mask = 0xB5;
+	qopt->entries[3].interval = 200000;
+	qopt->num_entries = 4;
+	delay_base_time(adapter, qopt, 19);
+	if (!enable_check_taprio(adapter, qopt, 100))
+		goto failed;
+	qopt->base_time = ktime_set(0, 0);
+	qopt->cycle_time = 6000000;
+	qopt->cycle_time_extension = 5999999;
+	qopt->entries[0].gate_mask = 0xC6;
+	qopt->entries[0].interval = 1000000;
+	qopt->entries[1].gate_mask = 0xC7;
+	qopt->entries[1].interval = 1000000;
+	qopt->entries[2].gate_mask = 0xC8;
+	qopt->entries[2].interval = 1000000;
+	qopt->entries[3].gate_mask = 0xC9;
+	qopt->entries[3].interval = 1500000;
+	qopt->entries[4].gate_mask = 0xCA;
+	qopt->entries[4].interval = 1500000;
+	qopt->num_entries = 5;
+	delay_base_time(adapter, qopt, 1);
+	if (!enable_check_taprio(adapter, qopt, 100))
+		goto failed;
+
+	if (!disable_taprio(adapter))
+		goto failed;
+
+	kfree(qopt);
+
+	return true;
+
+failed:
+	disable_taprio(adapter);
+	kfree(qopt);
+
+	return false;
+}
+
+int tsnep_ethtool_get_test_count(void)
+{
+	return TSNEP_TEST_COUNT;
+}
+
+void tsnep_ethtool_get_test_strings(u8 *data)
+{
+	memcpy(data, tsnep_test_strings, sizeof(tsnep_test_strings));
+}
+
+void tsnep_ethtool_self_test(struct net_device *netdev,
+			     struct ethtool_test *eth_test, u64 *data)
+{
+	struct tsnep_adapter *adapter = netdev_priv(netdev);
+
+	eth_test->len = TSNEP_TEST_COUNT;
+
+	if (eth_test->flags != ETH_TEST_FL_OFFLINE) {
+		/* no tests are done online */
+		data[TSNEP_TEST_ENABLE] = 0;
+		data[TSNEP_TEST_TAPRIO] = 0;
+		data[TSNEP_TEST_TAPRIO_CHANGE] = 0;
+		data[TSNEP_TEST_TAPRIO_EXTENSION] = 0;
+
+		return;
+	}
+
+	if (tsnep_test_gc_enable(adapter)) {
+		data[TSNEP_TEST_ENABLE] = 0;
+	} else {
+		eth_test->flags |= ETH_TEST_FL_FAILED;
+		data[TSNEP_TEST_ENABLE] = 1;
+	}
+
+	if (tsnep_test_taprio(adapter)) {
+		data[TSNEP_TEST_TAPRIO] = 0;
+	} else {
+		eth_test->flags |= ETH_TEST_FL_FAILED;
+		data[TSNEP_TEST_TAPRIO] = 1;
+	}
+
+	if (tsnep_test_taprio_change(adapter)) {
+		data[TSNEP_TEST_TAPRIO_CHANGE] = 0;
+	} else {
+		eth_test->flags |= ETH_TEST_FL_FAILED;
+		data[TSNEP_TEST_TAPRIO_CHANGE] = 1;
+	}
+
+	if (tsnep_test_taprio_extension(adapter)) {
+		data[TSNEP_TEST_TAPRIO_EXTENSION] = 0;
+	} else {
+		eth_test->flags |= ETH_TEST_FL_FAILED;
+		data[TSNEP_TEST_TAPRIO_EXTENSION] = 1;
+	}
+}
diff --git a/drivers/net/ethernet/engleder/tsnep_tc.c b/drivers/net/ethernet/engleder/tsnep_tc.c
new file mode 100644
index 000000000000..745b191a5540
--- /dev/null
+++ b/drivers/net/ethernet/engleder/tsnep_tc.c
@@ -0,0 +1,466 @@
+// SPDX-License-Identifier: GPL-2.0
+/* Copyright (C) 2021 Gerhard Engleder <gerhard@engleder-embedded.com> */
+
+#include "tsnep.h"
+
+#include <net/pkt_sched.h>
+
+/* save one operation at the end for additional operation at list change */
+#define TSNEP_MAX_GCL_NUM (TSNEP_GCL_COUNT - 1)
+
+static int tsnep_validate_gcl(struct tc_taprio_qopt_offload *qopt)
+{
+	int i;
+	u64 cycle_time;
+
+	if (!qopt->cycle_time)
+		return -ERANGE;
+	if (qopt->num_entries > TSNEP_MAX_GCL_NUM)
+		return -EINVAL;
+	cycle_time = 0;
+	for (i = 0; i < qopt->num_entries; i++) {
+		if (qopt->entries[i].command != TC_TAPRIO_CMD_SET_GATES)
+			return -EINVAL;
+		if (qopt->entries[i].gate_mask & ~TSNEP_GCL_MASK)
+			return -EINVAL;
+		if (qopt->entries[i].interval < TSNEP_GCL_MIN_INTERVAL)
+			return -EINVAL;
+		cycle_time += qopt->entries[i].interval;
+	}
+	if (qopt->cycle_time != cycle_time)
+		return -EINVAL;
+	if (qopt->cycle_time_extension >= qopt->cycle_time)
+		return -EINVAL;
+
+	return 0;
+}
+
+static void tsnep_write_gcl_operation(struct tsnep_gcl *gcl, int index,
+				      u32 properties, u32 interval, bool flush)
+{
+	void __iomem *addr = gcl->addr +
+			     sizeof(struct tsnep_gcl_operation) * index;
+
+	gcl->operation[index].properties = properties;
+	gcl->operation[index].interval = interval;
+
+	iowrite32(properties, addr);
+	iowrite32(interval, addr + sizeof(u32));
+
+	if (flush) {
+		/* flush write with read access */
+		ioread32(addr);
+	}
+}
+
+static u64 tsnep_change_duration(struct tsnep_gcl *gcl, int index)
+{
+	u64 duration;
+	int count;
+
+	/* change needs to be triggered one or two operations before start of
+	 * new gate control list
+	 * - change is triggered at start of operation (minimum one operation)
+	 * - operation with adjusted interval is inserted on demand to exactly
+	 *   meet the start of the new gate control list (optional)
+	 *
+	 * additionally properties are read directly after start of previous
+	 * operation
+	 *
+	 * therefore, three operations needs to be considered for the limit
+	 */
+	duration = 0;
+	count = 3;
+	while (count) {
+		duration += gcl->operation[index].interval;
+
+		index--;
+		if (index < 0)
+			index = gcl->count - 1;
+
+		count--;
+	}
+
+	return duration;
+}
+
+static void tsnep_write_gcl(struct tsnep_gcl *gcl,
+			    struct tc_taprio_qopt_offload *qopt)
+{
+	int i;
+	u32 properties;
+	u64 extend;
+	u64 cut;
+
+	gcl->base_time = ktime_to_ns(qopt->base_time);
+	gcl->cycle_time = qopt->cycle_time;
+	gcl->cycle_time_extension = qopt->cycle_time_extension;
+
+	for (i = 0; i < qopt->num_entries; i++) {
+		properties = qopt->entries[i].gate_mask;
+		if (i == (qopt->num_entries - 1))
+			properties |= TSNEP_GCL_LAST;
+
+		tsnep_write_gcl_operation(gcl, i, properties,
+					  qopt->entries[i].interval, true);
+	}
+	gcl->count = qopt->num_entries;
+
+	/* calculate change limit; i.e., the time needed between enable and
+	 * start of new gate control list
+	 */
+
+	/* case 1: extend cycle time for change
+	 * - change duration of last operation
+	 * - cycle time extension
+	 */
+	extend = tsnep_change_duration(gcl, gcl->count - 1);
+	extend += gcl->cycle_time_extension;
+
+	/* case 2: cut cycle time for change
+	 * - maximum change duration
+	 */
+	cut = 0;
+	for (i = 0; i < gcl->count; i++)
+		cut = max(cut, tsnep_change_duration(gcl, i));
+
+	/* use maximum, because the actual case (extend or cut) can be
+	 * determined only after limit is known (chicken-and-egg problem)
+	 */
+	gcl->change_limit = max(extend, cut);
+}
+
+static u64 tsnep_gcl_start_after(struct tsnep_gcl *gcl, u64 limit)
+{
+	u64 start = gcl->base_time;
+	u64 n;
+
+	if (start <= limit) {
+		n = div64_u64(limit - start, gcl->cycle_time);
+		start += (n + 1) * gcl->cycle_time;
+	}
+
+	return start;
+}
+
+static u64 tsnep_gcl_start_before(struct tsnep_gcl *gcl, u64 limit)
+{
+	u64 start = gcl->base_time;
+	u64 n;
+
+	n = div64_u64(limit - start, gcl->cycle_time);
+	start += n * gcl->cycle_time;
+	if (start == limit)
+		start -= gcl->cycle_time;
+
+	return start;
+}
+
+static u64 tsnep_set_gcl_change(struct tsnep_gcl *gcl, int index, u64 change,
+				bool insert)
+{
+	/* previous operation triggers change and properties are evaluated at
+	 * start of operation
+	 */
+	if (index == 0)
+		index = gcl->count - 1;
+	else
+		index = index - 1;
+	change -= gcl->operation[index].interval;
+
+	/* optionally change to new list with additional operation in between */
+	if (insert) {
+		void __iomem *addr = gcl->addr +
+				     sizeof(struct tsnep_gcl_operation) * index;
+
+		gcl->operation[index].properties |= TSNEP_GCL_INSERT;
+		iowrite32(gcl->operation[index].properties, addr);
+	}
+
+	return change;
+}
+
+static void tsnep_clean_gcl(struct tsnep_gcl *gcl)
+{
+	int i;
+	u32 mask = TSNEP_GCL_LAST | TSNEP_GCL_MASK;
+	void __iomem *addr;
+
+	/* search for insert operation and reset properties */
+	for (i = 0; i < gcl->count; i++) {
+		if (gcl->operation[i].properties & ~mask) {
+			addr = gcl->addr +
+			       sizeof(struct tsnep_gcl_operation) * i;
+
+			gcl->operation[i].properties &= mask;
+			iowrite32(gcl->operation[i].properties, addr);
+
+			break;
+		}
+	}
+}
+
+static u64 tsnep_insert_gcl_operation(struct tsnep_gcl *gcl, int ref,
+				      u64 change, u32 interval)
+{
+	u32 properties;
+
+	properties = gcl->operation[ref].properties & TSNEP_GCL_MASK;
+	/* change to new list directly after inserted operation */
+	properties |= TSNEP_GCL_CHANGE;
+
+	/* last operation of list is reserved to insert operation */
+	tsnep_write_gcl_operation(gcl, TSNEP_GCL_COUNT - 1, properties,
+				  interval, false);
+
+	return tsnep_set_gcl_change(gcl, ref, change, true);
+}
+
+static u64 tsnep_extend_gcl(struct tsnep_gcl *gcl, u64 start, u32 extension)
+{
+	int ref = gcl->count - 1;
+	u32 interval = gcl->operation[ref].interval + extension;
+
+	start -= gcl->operation[ref].interval;
+
+	return tsnep_insert_gcl_operation(gcl, ref, start, interval);
+}
+
+static u64 tsnep_cut_gcl(struct tsnep_gcl *gcl, u64 start, u64 cycle_time)
+{
+	u64 sum = 0;
+	int i;
+
+	/* find operation which shall be cutted */
+	for (i = 0; i < gcl->count; i++) {
+		u64 sum_tmp = sum + gcl->operation[i].interval;
+		u64 interval;
+
+		/* sum up operations as long as cycle time is not exceeded */
+		if (sum_tmp > cycle_time)
+			break;
+
+		/* remaining interval must be big enough for hardware */
+		interval = cycle_time - sum_tmp;
+		if (interval > 0 && interval < TSNEP_GCL_MIN_INTERVAL)
+			break;
+
+		sum = sum_tmp;
+	}
+	if (sum == cycle_time) {
+		/* no need to cut operation itself or whole cycle
+		 * => change exactly at operation
+		 */
+		return tsnep_set_gcl_change(gcl, i, start + sum, false);
+	}
+	return tsnep_insert_gcl_operation(gcl, i, start + sum,
+					  cycle_time - sum);
+}
+
+static int tsnep_enable_gcl(struct tsnep_adapter *adapter,
+			    struct tsnep_gcl *gcl, struct tsnep_gcl *curr)
+{
+	u64 system_time;
+	u64 timeout;
+	u64 limit;
+
+	/* estimate timeout limit after timeout enable, actually timeout limit
+	 * in hardware will be earlier than estimate so we are on the safe side
+	 */
+	tsnep_get_system_time(adapter, &system_time);
+	timeout = system_time + TSNEP_GC_TIMEOUT;
+
+	if (curr)
+		limit = timeout + curr->change_limit;
+	else
+		limit = timeout;
+
+	gcl->start_time = tsnep_gcl_start_after(gcl, limit);
+
+	/* gate control time register is only 32bit => time shall be in the near
+	 * future (no driver support for far future implemented)
+	 */
+	if ((gcl->start_time - system_time) >= U32_MAX)
+		return -EAGAIN;
+
+	if (curr) {
+		/* change gate control list */
+		u64 last;
+		u64 change;
+
+		last = tsnep_gcl_start_before(curr, gcl->start_time);
+		if ((last + curr->cycle_time) == gcl->start_time)
+			change = tsnep_cut_gcl(curr, last,
+					       gcl->start_time - last);
+		else if (((gcl->start_time - last) <=
+			  curr->cycle_time_extension) ||
+			 ((gcl->start_time - last) <= TSNEP_GCL_MIN_INTERVAL))
+			change = tsnep_extend_gcl(curr, last,
+						  gcl->start_time - last);
+		else
+			change = tsnep_cut_gcl(curr, last,
+					       gcl->start_time - last);
+
+		WARN_ON(change <= timeout);
+		gcl->change = true;
+		iowrite32(change & 0xFFFFFFFF, adapter->addr + TSNEP_GC_CHANGE);
+	} else {
+		/* start gate control list */
+		WARN_ON(gcl->start_time <= timeout);
+		gcl->change = false;
+		iowrite32(gcl->start_time & 0xFFFFFFFF,
+			  adapter->addr + TSNEP_GC_TIME);
+	}
+
+	return 0;
+}
+
+static int tsnep_taprio(struct tsnep_adapter *adapter,
+			struct tc_taprio_qopt_offload *qopt)
+{
+	struct tsnep_gcl *gcl;
+	struct tsnep_gcl *curr;
+	int retval;
+
+	if (!adapter->gate_control)
+		return -EOPNOTSUPP;
+
+	if (qopt->cmd == TAPRIO_CMD_DESTROY) {
+		/* disable gate control if active */
+		mutex_lock(&adapter->gate_control_lock);
+
+		if (adapter->gate_control_active) {
+			iowrite8(TSNEP_GC_DISABLE, adapter->addr + TSNEP_GC);
+			adapter->gate_control_active = false;
+		}
+
+		mutex_unlock(&adapter->gate_control_lock);
+
+		return 0;
+	} else if (qopt->cmd != TAPRIO_CMD_REPLACE) {
+		return -EOPNOTSUPP;
+	}
+
+	retval = tsnep_validate_gcl(qopt);
+	if (retval)
+		return retval;
+
+	mutex_lock(&adapter->gate_control_lock);
+
+	gcl = &adapter->gcl[adapter->next_gcl];
+	tsnep_write_gcl(gcl, qopt);
+
+	/* select current gate control list if active */
+	if (adapter->gate_control_active) {
+		if (adapter->next_gcl == 0)
+			curr = &adapter->gcl[1];
+		else
+			curr = &adapter->gcl[0];
+	} else {
+		curr = NULL;
+	}
+
+	for (;;) {
+		/* start timeout which discards late enable, this helps ensuring
+		 * that start/change time are in the future at enable
+		 */
+		iowrite8(TSNEP_GC_ENABLE_TIMEOUT, adapter->addr + TSNEP_GC);
+
+		retval = tsnep_enable_gcl(adapter, gcl, curr);
+		if (retval) {
+			mutex_unlock(&adapter->gate_control_lock);
+
+			return retval;
+		}
+
+		/* enable gate control list */
+		if (adapter->next_gcl == 0)
+			iowrite8(TSNEP_GC_ENABLE_A, adapter->addr + TSNEP_GC);
+		else
+			iowrite8(TSNEP_GC_ENABLE_B, adapter->addr + TSNEP_GC);
+
+		/* done if timeout did not happen */
+		if (!(ioread32(adapter->addr + TSNEP_GC) &
+		      TSNEP_GC_TIMEOUT_SIGNAL))
+			break;
+
+		/* timeout is acknowledged with any enable */
+		iowrite8(TSNEP_GC_ENABLE_A, adapter->addr + TSNEP_GC);
+
+		if (curr)
+			tsnep_clean_gcl(curr);
+
+		/* retry because of timeout */
+	}
+
+	adapter->gate_control_active = true;
+
+	if (adapter->next_gcl == 0)
+		adapter->next_gcl = 1;
+	else
+		adapter->next_gcl = 0;
+
+	mutex_unlock(&adapter->gate_control_lock);
+
+	return 0;
+}
+
+static int tsnep_tc_query_caps(struct tsnep_adapter *adapter,
+			       struct tc_query_caps_base *base)
+{
+	switch (base->type) {
+	case TC_SETUP_QDISC_TAPRIO: {
+		struct tc_taprio_caps *caps = base->caps;
+
+		if (!adapter->gate_control)
+			return -EOPNOTSUPP;
+
+		caps->gate_mask_per_txq = true;
+
+		return 0;
+	}
+	default:
+		return -EOPNOTSUPP;
+	}
+}
+
+int tsnep_tc_setup(struct net_device *netdev, enum tc_setup_type type,
+		   void *type_data)
+{
+	struct tsnep_adapter *adapter = netdev_priv(netdev);
+
+	switch (type) {
+	case TC_QUERY_CAPS:
+		return tsnep_tc_query_caps(adapter, type_data);
+	case TC_SETUP_QDISC_TAPRIO:
+		return tsnep_taprio(adapter, type_data);
+	default:
+		return -EOPNOTSUPP;
+	}
+}
+
+int tsnep_tc_init(struct tsnep_adapter *adapter)
+{
+	if (!adapter->gate_control)
+		return 0;
+
+	/* open all gates */
+	iowrite8(TSNEP_GC_DISABLE, adapter->addr + TSNEP_GC);
+	iowrite32(TSNEP_GC_OPEN | TSNEP_GC_NEXT_OPEN, adapter->addr + TSNEP_GC);
+
+	adapter->gcl[0].addr = adapter->addr + TSNEP_GCL_A;
+	adapter->gcl[1].addr = adapter->addr + TSNEP_GCL_B;
+
+	return 0;
+}
+
+void tsnep_tc_cleanup(struct tsnep_adapter *adapter)
+{
+	if (!adapter->gate_control)
+		return;
+
+	if (adapter->gate_control_active) {
+		iowrite8(TSNEP_GC_DISABLE, adapter->addr + TSNEP_GC);
+		adapter->gate_control_active = false;
+	}
+}
diff --git a/drivers/net/ethernet/engleder/tsnep_xdp.c b/drivers/net/ethernet/engleder/tsnep_xdp.c
new file mode 100644
index 000000000000..c0513848c547
--- /dev/null
+++ b/drivers/net/ethernet/engleder/tsnep_xdp.c
@@ -0,0 +1,85 @@
+// SPDX-License-Identifier: GPL-2.0
+/* Copyright (C) 2022 Gerhard Engleder <gerhard@engleder-embedded.com> */
+
+#include <linux/if_vlan.h>
+#include <net/xdp_sock_drv.h>
+
+#include "tsnep.h"
+
+int tsnep_xdp_setup_prog(struct tsnep_adapter *adapter, struct bpf_prog *prog,
+			 struct netlink_ext_ack *extack)
+{
+	struct bpf_prog *old_prog;
+
+	old_prog = xchg(&adapter->xdp_prog, prog);
+	if (old_prog)
+		bpf_prog_put(old_prog);
+
+	return 0;
+}
+
+static int tsnep_xdp_enable_pool(struct tsnep_adapter *adapter,
+				 struct xsk_buff_pool *pool, u16 queue_id)
+{
+	struct tsnep_queue *queue;
+	int retval;
+
+	if (queue_id >= adapter->num_rx_queues ||
+	    queue_id >= adapter->num_tx_queues)
+		return -EINVAL;
+
+	queue = &adapter->queue[queue_id];
+	if (queue->rx->queue_index != queue_id ||
+	    queue->tx->queue_index != queue_id) {
+		netdev_err(adapter->netdev,
+			   "XSK support only for TX/RX queue pairs\n");
+
+		return -EOPNOTSUPP;
+	}
+
+	retval = xsk_pool_dma_map(pool, adapter->dmadev,
+				  DMA_ATTR_SKIP_CPU_SYNC);
+	if (retval) {
+		netdev_err(adapter->netdev, "failed to map XSK pool\n");
+
+		return retval;
+	}
+
+	retval = tsnep_enable_xsk(queue, pool);
+	if (retval) {
+		xsk_pool_dma_unmap(pool, DMA_ATTR_SKIP_CPU_SYNC);
+
+		return retval;
+	}
+
+	return 0;
+}
+
+static int tsnep_xdp_disable_pool(struct tsnep_adapter *adapter, u16 queue_id)
+{
+	struct xsk_buff_pool *pool;
+	struct tsnep_queue *queue;
+
+	if (queue_id >= adapter->num_rx_queues ||
+	    queue_id >= adapter->num_tx_queues)
+		return -EINVAL;
+
+	pool = xsk_get_pool_from_qid(adapter->netdev, queue_id);
+	if (!pool)
+		return -EINVAL;
+
+	queue = &adapter->queue[queue_id];
+
+	tsnep_disable_xsk(queue);
+
+	xsk_pool_dma_unmap(pool, DMA_ATTR_SKIP_CPU_SYNC);
+
+	return 0;
+}
+
+int tsnep_xdp_setup_pool(struct tsnep_adapter *adapter,
+			 struct xsk_buff_pool *pool, u16 queue_id)
+{
+	return pool ? tsnep_xdp_enable_pool(adapter, pool, queue_id) :
+		      tsnep_xdp_disable_pool(adapter, queue_id);
+}
diff --git a/drivers/net/ethernet/ethoc.c b/drivers/net/ethernet/ethoc.c
index e863ba74d005..0c418557264c 100644
--- a/drivers/net/ethernet/ethoc.c
+++ b/drivers/net/ethernet/ethoc.c
@@ -1,13 +1,10 @@
+// SPDX-License-Identifier: GPL-2.0-only
 /*
  * linux/drivers/net/ethernet/ethoc.c
  *
  * Copyright (C) 2007-2008 Avionic Design Development GmbH
  * Copyright (C) 2008-2009 Avionic Design GmbH
  *
- * This program is free software; you can redistribute it and/or modify
- * it under the terms of the GNU General Public License version 2 as
- * published by the Free Software Foundation.
- *
  * Written by Thierry Reding <thierry.reding@avionic-design.de>
  */
 
@@ -180,6 +177,7 @@ MODULE_PARM_DESC(buffer_size, "DMA buffer allocation size");
  * struct ethoc - driver-private device structure
  * @iobase:	pointer to I/O memory region
  * @membase:	pointer to buffer memory region
+ * @big_endian: just big or little (endian)
  * @num_bd:	number of buffer descriptors
  * @num_tx:	number of send buffers
  * @cur_tx:	last send buffer written
@@ -192,7 +190,10 @@ MODULE_PARM_DESC(buffer_size, "DMA buffer allocation size");
  * @msg_enable:	device state flags
  * @lock:	device lock
  * @mdio:	MDIO bus for PHY access
+ * @clk:	clock
  * @phy_id:	address of attached PHY
+ * @old_link:	previous link info
+ * @old_duplex: previous duplex info
  */
 struct ethoc {
 	void __iomem *iobase;
@@ -706,25 +707,18 @@ static int ethoc_mdio_probe(struct net_device *dev)
 	else
 		phy = phy_find_first(priv->mdio);
 
-	if (!phy) {
-		dev_err(&dev->dev, "no PHY found\n");
-		return -ENXIO;
-	}
+	if (!phy)
+		return dev_err_probe(&dev->dev, -ENXIO, "no PHY found\n");
 
 	priv->old_duplex = -1;
 	priv->old_link = -1;
 
 	err = phy_connect_direct(dev, phy, ethoc_mdio_poll,
 				 PHY_INTERFACE_MODE_GMII);
-	if (err) {
-		dev_err(&dev->dev, "could not attach to PHY\n");
-		return err;
-	}
+	if (err)
+		return dev_err_probe(&dev->dev, err, "could not attach to PHY\n");
 
-	phy->advertising &= ~(ADVERTISED_1000baseT_Full |
-			      ADVERTISED_1000baseT_Half);
-	phy->supported &= ~(SUPPORTED_1000baseT_Full |
-			    SUPPORTED_1000baseT_Half);
+	phy_set_max_speed(phy, SPEED_100);
 
 	return 0;
 }
@@ -739,6 +733,8 @@ static int ethoc_open(struct net_device *dev)
 	if (ret)
 		return ret;
 
+	napi_enable(&priv->napi);
+
 	ethoc_init_ring(priv, dev->mem_start);
 	ethoc_reset(priv);
 
@@ -754,7 +750,6 @@ static int ethoc_open(struct net_device *dev)
 	priv->old_duplex = -1;
 
 	phy_start(dev->phydev);
-	napi_enable(&priv->napi);
 
 	if (netif_msg_ifup(priv)) {
 		dev_info(&dev->dev, "I/O: %08lx Memory: %08lx-%08lx\n",
@@ -807,8 +802,8 @@ static int ethoc_ioctl(struct net_device *dev, struct ifreq *ifr, int cmd)
 
 static void ethoc_do_set_mac_address(struct net_device *dev)
 {
+	const unsigned char *mac = dev->dev_addr;
 	struct ethoc *priv = netdev_priv(dev);
-	unsigned char *mac = dev->dev_addr;
 
 	ethoc_write(priv, MAC_ADDR0, (mac[2] << 24) | (mac[3] << 16) |
 				     (mac[4] <<  8) | (mac[5] <<  0));
@@ -821,7 +816,7 @@ static int ethoc_set_mac_address(struct net_device *dev, void *p)
 
 	if (!is_valid_ether_addr(addr->sa_data))
 		return -EADDRNOTAVAIL;
-	memcpy(dev->dev_addr, addr->sa_data, ETH_ALEN);
+	eth_hw_addr_set(dev, addr->sa_data);
 	ethoc_do_set_mac_address(dev);
 	return 0;
 }
@@ -874,7 +869,7 @@ static int ethoc_change_mtu(struct net_device *dev, int new_mtu)
 	return -ENOSYS;
 }
 
-static void ethoc_tx_timeout(struct net_device *dev)
+static void ethoc_tx_timeout(struct net_device *dev, unsigned int txqueue)
 {
 	struct ethoc *priv = netdev_priv(dev);
 	u32 pending = ethoc_read(priv, INT_SOURCE);
@@ -950,7 +945,9 @@ static void ethoc_get_regs(struct net_device *dev, struct ethtool_regs *regs,
 }
 
 static void ethoc_get_ringparam(struct net_device *dev,
-				struct ethtool_ringparam *ring)
+				struct ethtool_ringparam *ring,
+				struct kernel_ethtool_ringparam *kernel_ring,
+				struct netlink_ext_ack *extack)
 {
 	struct ethoc *priv = netdev_priv(dev);
 
@@ -966,7 +963,9 @@ static void ethoc_get_ringparam(struct net_device *dev,
 }
 
 static int ethoc_set_ringparam(struct net_device *dev,
-			       struct ethtool_ringparam *ring)
+			       struct ethtool_ringparam *ring,
+			       struct kernel_ethtool_ringparam *kernel_ring,
+			       struct netlink_ext_ack *extack)
 {
 	struct ethoc *priv = netdev_priv(dev);
 
@@ -1010,7 +1009,7 @@ static const struct ethtool_ops ethoc_ethtool_ops = {
 static const struct net_device_ops ethoc_netdev_ops = {
 	.ndo_open = ethoc_open,
 	.ndo_stop = ethoc_stop,
-	.ndo_do_ioctl = ethoc_ioctl,
+	.ndo_eth_ioctl = ethoc_ioctl,
 	.ndo_set_mac_address = ethoc_set_mac_address,
 	.ndo_set_rx_mode = ethoc_set_multicast_list,
 	.ndo_change_mtu = ethoc_change_mtu,
@@ -1020,7 +1019,7 @@ static const struct net_device_ops ethoc_netdev_ops = {
 
 /**
  * ethoc_probe - initialize OpenCores ethernet MAC
- * pdev:	platform device
+ * @pdev:	platform device
  */
 static int ethoc_probe(struct platform_device *pdev)
 {
@@ -1079,20 +1078,17 @@ static int ethoc_probe(struct platform_device *pdev)
 
 
 	/* obtain device IRQ number */
-	res = platform_get_resource(pdev, IORESOURCE_IRQ, 0);
-	if (!res) {
-		dev_err(&pdev->dev, "cannot obtain IRQ\n");
-		ret = -ENXIO;
+	ret = platform_get_irq(pdev, 0);
+	if (ret < 0)
 		goto free;
-	}
 
-	netdev->irq = res->start;
+	netdev->irq = ret;
 
 	/* setup driver-private data */
 	priv = netdev_priv(netdev);
 	priv->netdev = netdev;
 
-	priv->iobase = devm_ioremap_nocache(&pdev->dev, netdev->base_addr,
+	priv->iobase = devm_ioremap(&pdev->dev, netdev->base_addr,
 			resource_size(mmio));
 	if (!priv->iobase) {
 		dev_err(&pdev->dev, "cannot remap I/O memory space\n");
@@ -1101,7 +1097,7 @@ static int ethoc_probe(struct platform_device *pdev)
 	}
 
 	if (netdev->mem_end) {
-		priv->membase = devm_ioremap_nocache(&pdev->dev,
+		priv->membase = devm_ioremap(&pdev->dev,
 			netdev->mem_start, resource_size(mem));
 		if (!priv->membase) {
 			dev_err(&pdev->dev, "cannot remap memory space\n");
@@ -1140,7 +1136,8 @@ static int ethoc_probe(struct platform_device *pdev)
 	dev_dbg(&pdev->dev, "ethoc: num_tx: %d num_rx: %d\n",
 		priv->num_tx, priv->num_rx);
 
-	priv->vma = devm_kzalloc(&pdev->dev, num_bd*sizeof(void *), GFP_KERNEL);
+	priv->vma = devm_kcalloc(&pdev->dev, num_bd, sizeof(void *),
+				 GFP_KERNEL);
 	if (!priv->vma) {
 		ret = -ENOMEM;
 		goto free;
@@ -1148,22 +1145,22 @@ static int ethoc_probe(struct platform_device *pdev)
 
 	/* Allow the platform setup code to pass in a MAC address. */
 	if (pdata) {
-		ether_addr_copy(netdev->dev_addr, pdata->hwaddr);
+		eth_hw_addr_set(netdev, pdata->hwaddr);
 		priv->phy_id = pdata->phy_id;
 	} else {
-		const void *mac;
-
-		mac = of_get_mac_address(pdev->dev.of_node);
-		if (mac)
-			ether_addr_copy(netdev->dev_addr, mac);
+		of_get_ethdev_address(pdev->dev.of_node, netdev);
 		priv->phy_id = -1;
 	}
 
 	/* Check that the given MAC address is valid. If it isn't, read the
 	 * current MAC from the controller.
 	 */
-	if (!is_valid_ether_addr(netdev->dev_addr))
-		ethoc_get_mac_address(netdev, netdev->dev_addr);
+	if (!is_valid_ether_addr(netdev->dev_addr)) {
+		u8 addr[ETH_ALEN];
+
+		ethoc_get_mac_address(netdev, addr);
+		eth_hw_addr_set(netdev, addr);
+	}
 
 	/* Check the MAC again for validity, if it still isn't choose and
 	 * program a random one.
@@ -1211,7 +1208,7 @@ static int ethoc_probe(struct platform_device *pdev)
 	ret = mdiobus_register(priv->mdio);
 	if (ret) {
 		dev_err(&netdev->dev, "failed to register MDIO bus\n");
-		goto free2;
+		goto free3;
 	}
 
 	ret = ethoc_mdio_probe(netdev);
@@ -1227,7 +1224,7 @@ static int ethoc_probe(struct platform_device *pdev)
 	netdev->ethtool_ops = &ethoc_ethtool_ops;
 
 	/* setup NAPI */
-	netif_napi_add(netdev, &priv->napi, ethoc_poll, 64);
+	netif_napi_add(netdev, &priv->napi, ethoc_poll);
 
 	spin_lock_init(&priv->lock);
 
@@ -1243,10 +1240,10 @@ error2:
 	netif_napi_del(&priv->napi);
 error:
 	mdiobus_unregister(priv->mdio);
+free3:
 	mdiobus_free(priv->mdio);
 free2:
-	if (priv->clk)
-		clk_disable_unprepare(priv->clk);
+	clk_disable_unprepare(priv->clk);
 free:
 	free_netdev(netdev);
 out:
@@ -1257,7 +1254,7 @@ out:
  * ethoc_remove - shutdown OpenCores ethernet MAC
  * @pdev:	platform device
  */
-static int ethoc_remove(struct platform_device *pdev)
+static void ethoc_remove(struct platform_device *pdev)
 {
 	struct net_device *netdev = platform_get_drvdata(pdev);
 	struct ethoc *priv = netdev_priv(netdev);
@@ -1270,13 +1267,10 @@ static int ethoc_remove(struct platform_device *pdev)
 			mdiobus_unregister(priv->mdio);
 			mdiobus_free(priv->mdio);
 		}
-		if (priv->clk)
-			clk_disable_unprepare(priv->clk);
+		clk_disable_unprepare(priv->clk);
 		unregister_netdev(netdev);
 		free_netdev(netdev);
 	}
-
-	return 0;
 }
 
 #ifdef CONFIG_PM
@@ -1302,7 +1296,7 @@ MODULE_DEVICE_TABLE(of, ethoc_match);
 
 static struct platform_driver ethoc_driver = {
 	.probe   = ethoc_probe,
-	.remove  = ethoc_remove,
+	.remove = ethoc_remove,
 	.suspend = ethoc_suspend,
 	.resume  = ethoc_resume,
 	.driver  = {
diff --git a/drivers/net/ethernet/ezchip/Kconfig b/drivers/net/ethernet/ezchip/Kconfig
index b423ad380b6a..9241b9b1c7a3 100644
--- a/drivers/net/ethernet/ezchip/Kconfig
+++ b/drivers/net/ethernet/ezchip/Kconfig
@@ -1,3 +1,4 @@
+# SPDX-License-Identifier: GPL-2.0-only
 #
 # EZchip network device configuration
 #
@@ -5,7 +6,7 @@
 config NET_VENDOR_EZCHIP
 	bool "EZchip devices"
 	default y
-	---help---
+	help
 	  If you have a network (Ethernet) device belonging to this class, say Y.
 
 	  Note that the answer to this question doesn't directly affect the
@@ -17,9 +18,9 @@ if NET_VENDOR_EZCHIP
 
 config EZCHIP_NPS_MANAGEMENT_ENET
 	tristate "EZchip NPS management enet support"
-	depends on OF_IRQ && OF_NET
+	depends on OF_IRQ
 	depends on HAS_IOMEM
-	---help---
+	help
 	  Simple LAN device for debug or management purposes.
 	  Device supports interrupts for RX and TX(completion).
 	  Device does not have DMA ability.
diff --git a/drivers/net/ethernet/ezchip/Makefile b/drivers/net/ethernet/ezchip/Makefile
index e490176a8137..444570f35d45 100644
--- a/drivers/net/ethernet/ezchip/Makefile
+++ b/drivers/net/ethernet/ezchip/Makefile
@@ -1 +1,2 @@
+# SPDX-License-Identifier: GPL-2.0-only
 obj-$(CONFIG_EZCHIP_NPS_MANAGEMENT_ENET) += nps_enet.o
diff --git a/drivers/net/ethernet/ezchip/nps_enet.c b/drivers/net/ethernet/ezchip/nps_enet.c
index 659f1ad37e96..5cb478e98697 100644
--- a/drivers/net/ethernet/ezchip/nps_enet.c
+++ b/drivers/net/ethernet/ezchip/nps_enet.c
@@ -1,26 +1,14 @@
+// SPDX-License-Identifier: GPL-2.0-only
 /*
  * Copyright(c) 2015 EZchip Technologies.
- *
- * This program is free software; you can redistribute it and/or modify it
- * under the terms and conditions of the GNU General Public License,
- * version 2, as published by the Free Software Foundation.
- *
- * This program is distributed in the hope it will be useful, but WITHOUT
- * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
- * FITNESS FOR A PARTICULAR PURPOSE.  See the GNU General Public License for
- * more details.
- *
- * The full GNU General Public License is included in this distribution in
- * the file called "COPYING".
  */
 
 #include <linux/module.h>
 #include <linux/etherdevice.h>
 #include <linux/interrupt.h>
-#include <linux/of_address.h>
-#include <linux/of_irq.h>
+#include <linux/mod_devicetable.h>
 #include <linux/of_net.h>
-#include <linux/of_platform.h>
+#include <linux/platform_device.h>
 #include "nps_enet.h"
 
 #define DRV_NAME			"nps_mgt_enet"
@@ -210,7 +198,7 @@ static int nps_enet_poll(struct napi_struct *napi, int budget)
 		 */
 		if (nps_enet_is_tx_pending(priv)) {
 			nps_enet_reg_set(priv, NPS_ENET_REG_BUF_INT_ENABLE, 0);
-			napi_reschedule(napi);
+			napi_schedule(napi);
 		}
 	}
 
@@ -432,7 +420,7 @@ static s32 nps_enet_set_mac_address(struct net_device *ndev, void *p)
 
 	res = eth_mac_addr(ndev, p);
 	if (!res) {
-		ether_addr_copy(ndev->dev_addr, addr->sa_data);
+		eth_hw_addr_set(ndev, addr->sa_data);
 		nps_enet_set_hw_mac_address(ndev);
 	}
 
@@ -586,8 +574,6 @@ static s32 nps_enet_probe(struct platform_device *pdev)
 	struct net_device *ndev;
 	struct nps_enet_priv *priv;
 	s32 err = 0;
-	const char *mac_addr;
-	struct resource *res_regs;
 
 	if (!dev->of_node)
 		return -ENODEV;
@@ -606,8 +592,7 @@ static s32 nps_enet_probe(struct platform_device *pdev)
 	/* FIXME :: no multicast support yet */
 	ndev->flags &= ~IFF_MULTICAST;
 
-	res_regs = platform_get_resource(pdev, IORESOURCE_MEM, 0);
-	priv->regs_base = devm_ioremap_resource(dev, res_regs);
+	priv->regs_base = devm_platform_ioremap_resource(pdev, 0);
 	if (IS_ERR(priv->regs_base)) {
 		err = PTR_ERR(priv->regs_base);
 		goto out_netdev;
@@ -615,22 +600,19 @@ static s32 nps_enet_probe(struct platform_device *pdev)
 	dev_dbg(dev, "Registers base address is 0x%p\n", priv->regs_base);
 
 	/* set kernel MAC address to dev */
-	mac_addr = of_get_mac_address(dev->of_node);
-	if (mac_addr)
-		ether_addr_copy(ndev->dev_addr, mac_addr);
-	else
+	err = of_get_ethdev_address(dev->of_node, ndev);
+	if (err)
 		eth_hw_addr_random(ndev);
 
 	/* Get IRQ number */
 	priv->irq = platform_get_irq(pdev, 0);
-	if (!priv->irq) {
-		dev_err(dev, "failed to retrieve <irq Rx-Tx> value from device tree\n");
+	if (priv->irq < 0) {
 		err = -ENODEV;
 		goto out_netdev;
 	}
 
-	netif_napi_add(ndev, &priv->napi, nps_enet_poll,
-		       NPS_ENET_NAPI_POLL_WEIGHT);
+	netif_napi_add_weight(ndev, &priv->napi, nps_enet_poll,
+			      NPS_ENET_NAPI_POLL_WEIGHT);
 
 	/* Register the driver. Should be the last thing in probe */
 	err = register_netdev(ndev);
@@ -646,22 +628,19 @@ static s32 nps_enet_probe(struct platform_device *pdev)
 out_netif_api:
 	netif_napi_del(&priv->napi);
 out_netdev:
-	if (err)
-		free_netdev(ndev);
+	free_netdev(ndev);
 
 	return err;
 }
 
-static s32 nps_enet_remove(struct platform_device *pdev)
+static void nps_enet_remove(struct platform_device *pdev)
 {
 	struct net_device *ndev = platform_get_drvdata(pdev);
 	struct nps_enet_priv *priv = netdev_priv(ndev);
 
 	unregister_netdev(ndev);
-	free_netdev(ndev);
 	netif_napi_del(&priv->napi);
-
-	return 0;
+	free_netdev(ndev);
 }
 
 static const struct of_device_id nps_enet_dt_ids[] = {
@@ -682,4 +661,5 @@ static struct platform_driver nps_enet_driver = {
 module_platform_driver(nps_enet_driver);
 
 MODULE_AUTHOR("EZchip Semiconductor");
+MODULE_DESCRIPTION("EZchip NPS Ethernet driver");
 MODULE_LICENSE("GPL v2");
diff --git a/drivers/net/ethernet/ezchip/nps_enet.h b/drivers/net/ethernet/ezchip/nps_enet.h
index 3939ca20cc9f..092da2d90026 100644
--- a/drivers/net/ethernet/ezchip/nps_enet.h
+++ b/drivers/net/ethernet/ezchip/nps_enet.h
@@ -1,17 +1,6 @@
+/* SPDX-License-Identifier: GPL-2.0-only */
 /*
  * Copyright(c) 2015 EZchip Technologies.
- *
- * This program is free software; you can redistribute it and/or modify it
- * under the terms and conditions of the GNU General Public License,
- * version 2, as published by the Free Software Foundation.
- *
- * This program is distributed in the hope it will be useful, but WITHOUT
- * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
- * FITNESS FOR A PARTICULAR PURPOSE.  See the GNU General Public License for
- * more details.
- *
- * The full GNU General Public License is included in this distribution in
- * the file called "COPYING".
  */
 
 #ifndef _NPS_ENET_H
@@ -178,7 +167,7 @@ struct nps_enet_priv {
 };
 
 /**
- * nps_reg_set - Sets ENET register with provided value.
+ * nps_enet_reg_set - Sets ENET register with provided value.
  * @priv:       Pointer to EZchip ENET private data structure.
  * @reg:        Register offset from base address.
  * @value:      Value to set in register.
@@ -190,7 +179,7 @@ static inline void nps_enet_reg_set(struct nps_enet_priv *priv,
 }
 
 /**
- * nps_reg_get - Gets value of specified ENET register.
+ * nps_enet_reg_get - Gets value of specified ENET register.
  * @priv:       Pointer to EZchip ENET private data structure.
  * @reg:        Register offset from base address.
  *
diff --git a/drivers/net/ethernet/faraday/Kconfig b/drivers/net/ethernet/faraday/Kconfig
index 040c7f163325..474073c7f94d 100644
--- a/drivers/net/ethernet/faraday/Kconfig
+++ b/drivers/net/ethernet/faraday/Kconfig
@@ -1,3 +1,4 @@
+# SPDX-License-Identifier: GPL-2.0-only
 #
 # Faraday device configuration
 #
@@ -5,8 +6,8 @@
 config NET_VENDOR_FARADAY
 	bool "Faraday devices"
 	default y
-	depends on ARM
-	---help---
+	depends on ARM || COMPILE_TEST
+	help
 	  If you have a network (Ethernet) card belonging to this class, say Y.
 
 	  Note that the answer to this question doesn't directly affect the
@@ -18,20 +19,23 @@ if NET_VENDOR_FARADAY
 
 config FTMAC100
 	tristate "Faraday FTMAC100 10/100 Ethernet support"
-	depends on ARM
+	depends on ARM || COMPILE_TEST
+	depends on !64BIT || BROKEN
 	select MII
-	---help---
+	help
 	  This driver supports the FTMAC100 10/100 Ethernet controller
-	  from Faraday. It is used on Faraday A320, Andes AG101 and some
-	  other ARM/NDS32 SoC's.
+	  from Faraday. It is used on Faraday A320 and some other ARM SoC's.
 
 config FTGMAC100
 	tristate "Faraday FTGMAC100 Gigabit Ethernet support"
-	depends on ARM
+	depends on ARM || COMPILE_TEST
+	depends on !64BIT || BROKEN
 	select PHYLIB
-	---help---
+	select FIXED_PHY
+	select MDIO_ASPEED if MACH_ASPEED_G6
+	select CRC32
+	help
 	  This driver supports the FTGMAC100 Gigabit Ethernet controller
-	  from Faraday. It is used on Faraday A369, Andes AG102 and some
-	  other ARM/NDS32 SoC's.
+	  from Faraday. It is used on Faraday A369 and some other ARM SoC's.
 
 endif # NET_VENDOR_FARADAY
diff --git a/drivers/net/ethernet/faraday/Makefile b/drivers/net/ethernet/faraday/Makefile
index 408b53980d53..f16f58467868 100644
--- a/drivers/net/ethernet/faraday/Makefile
+++ b/drivers/net/ethernet/faraday/Makefile
@@ -1,3 +1,4 @@
+# SPDX-License-Identifier: GPL-2.0-only
 #
 # Makefile for the Faraday device drivers.
 #
diff --git a/drivers/net/ethernet/faraday/ftgmac100.c b/drivers/net/ethernet/faraday/ftgmac100.c
index 95bf5e89cfd1..a863f7841210 100644
--- a/drivers/net/ethernet/faraday/ftgmac100.c
+++ b/drivers/net/ethernet/faraday/ftgmac100.c
@@ -1,26 +1,15 @@
+// SPDX-License-Identifier: GPL-2.0-or-later
 /*
  * Faraday FTGMAC100 Gigabit Ethernet
  *
  * (C) Copyright 2009-2011 Faraday Technology
  * Po-Yu Chuang <ratbert@faraday-tech.com>
- *
- * This program is free software; you can redistribute it and/or modify
- * it under the terms of the GNU General Public License as published by
- * the Free Software Foundation; either version 2 of the License, or
- * (at your option) any later version.
- *
- * This program is distributed in the hope that it will be useful,
- * but WITHOUT ANY WARRANTY; without even the implied warranty of
- * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
- * GNU General Public License for more details.
- *
- * You should have received a copy of the GNU General Public License
- * along with this program; if not, write to the Free Software
- * Foundation, Inc., 675 Mass Ave, Cambridge, MA 02139, USA.
  */
 
 #define pr_fmt(fmt)	KBUILD_MODNAME ": " fmt
 
+#include <linux/clk.h>
+#include <linux/reset.h>
 #include <linux/dma-mapping.h>
 #include <linux/etherdevice.h>
 #include <linux/ethtool.h>
@@ -29,19 +18,20 @@
 #include <linux/module.h>
 #include <linux/netdevice.h>
 #include <linux/of.h>
+#include <linux/of_mdio.h>
 #include <linux/phy.h>
 #include <linux/platform_device.h>
 #include <linux/property.h>
 #include <linux/crc32.h>
 #include <linux/if_vlan.h>
 #include <linux/of_net.h>
+#include <linux/phy_fixed.h>
 #include <net/ip.h>
 #include <net/ncsi.h>
 
 #include "ftgmac100.h"
 
 #define DRV_NAME	"ftgmac100"
-#define DRV_VERSION	"0.7"
 
 /* Arbitrary values, I am not sure the HW has limits */
 #define MAX_RX_QUEUE_ENTRIES	1024
@@ -59,6 +49,18 @@
 /* Min number of tx ring entries before stopping queue */
 #define TX_THRESHOLD		(MAX_SKB_FRAGS + 1)
 
+#define FTGMAC_100MHZ		100000000
+#define FTGMAC_25MHZ		25000000
+
+/* For NC-SI to register a fixed-link phy device */
+static struct fixed_phy_status ncsi_phy_status = {
+	.link = 1,
+	.speed = SPEED_100,
+	.duplex = DUPLEX_FULL,
+	.pause = 0,
+	.asym_pause = 0
+};
+
 struct ftgmac100 {
 	/* Registers */
 	struct resource *res;
@@ -96,6 +98,12 @@ struct ftgmac100 {
 	struct napi_struct napi;
 	struct work_struct reset_task;
 	struct mii_bus *mii_bus;
+	struct clk *clk;
+
+	/* AST2500/AST2600 RMII ref clock gate */
+	struct clk *rclk;
+	/* Aspeed reset control */
+	struct reset_control *rst;
 
 	/* Link management */
 	int cur_speed;
@@ -125,7 +133,7 @@ static int ftgmac100_reset_mac(struct ftgmac100 *priv, u32 maccr)
 	iowrite32(maccr, priv->base + FTGMAC100_OFFSET_MACCR);
 	iowrite32(maccr | FTGMAC100_MACCR_SW_RST,
 		  priv->base + FTGMAC100_OFFSET_MACCR);
-	for (i = 0; i < 50; i++) {
+	for (i = 0; i < 200; i++) {
 		unsigned int maccr;
 
 		maccr = ioread32(priv->base + FTGMAC100_OFFSET_MACCR);
@@ -143,6 +151,23 @@ static int ftgmac100_reset_and_config_mac(struct ftgmac100 *priv)
 {
 	u32 maccr = 0;
 
+	/* Aspeed RMII needs SCU reset to clear status */
+	if (priv->is_aspeed && priv->netdev->phydev->interface == PHY_INTERFACE_MODE_RMII) {
+		int err;
+
+		err = reset_control_assert(priv->rst);
+		if (err) {
+			dev_err(priv->dev, "Failed to reset mac (%d)\n", err);
+			return err;
+		}
+		usleep_range(10000, 20000);
+		err = reset_control_deassert(priv->rst);
+		if (err) {
+			dev_err(priv->dev, "Failed to deassert mac reset (%d)\n", err);
+			return err;
+		}
+	}
+
 	switch (priv->cur_speed) {
 	case SPEED_10:
 	case 0: /* no link */
@@ -182,19 +207,20 @@ static void ftgmac100_write_mac_addr(struct ftgmac100 *priv, const u8 *mac)
 	iowrite32(laddr, priv->base + FTGMAC100_OFFSET_MAC_LADR);
 }
 
-static void ftgmac100_initial_mac(struct ftgmac100 *priv)
+static int ftgmac100_initial_mac(struct ftgmac100 *priv)
 {
 	u8 mac[ETH_ALEN];
 	unsigned int m;
 	unsigned int l;
-	void *addr;
+	int err;
 
-	addr = device_get_mac_address(priv->dev, mac, ETH_ALEN);
-	if (addr) {
-		ether_addr_copy(priv->netdev->dev_addr, mac);
+	err = of_get_ethdev_address(priv->dev->of_node, priv->netdev);
+	if (err == -EPROBE_DEFER)
+		return err;
+	if (!err) {
 		dev_info(priv->dev, "Read MAC address %pM from device tree\n",
-			 mac);
-		return;
+			 priv->netdev->dev_addr);
+		return 0;
 	}
 
 	m = ioread32(priv->base + FTGMAC100_OFFSET_MAC_MADR);
@@ -208,13 +234,15 @@ static void ftgmac100_initial_mac(struct ftgmac100 *priv)
 	mac[5] = l & 0xff;
 
 	if (is_valid_ether_addr(mac)) {
-		ether_addr_copy(priv->netdev->dev_addr, mac);
+		eth_hw_addr_set(priv->netdev, mac);
 		dev_info(priv->dev, "Read MAC address %pM from chip\n", mac);
 	} else {
 		eth_hw_addr_random(priv->netdev);
 		dev_info(priv->dev, "Generated random MAC address %pM\n",
 			 priv->netdev->dev_addr);
 	}
+
+	return 0;
 }
 
 static int ftgmac100_set_mac_addr(struct net_device *dev, void *p)
@@ -392,7 +420,7 @@ static int ftgmac100_alloc_rx_buf(struct ftgmac100 *priv, unsigned int entry,
 	struct net_device *netdev = priv->netdev;
 	struct sk_buff *skb;
 	dma_addr_t map;
-	int err;
+	int err = 0;
 
 	skb = netdev_alloc_skb_ip_align(netdev, RX_BUF_SIZE);
 	if (unlikely(!skb)) {
@@ -428,7 +456,7 @@ static int ftgmac100_alloc_rx_buf(struct ftgmac100 *priv, unsigned int entry,
 	else
 		rxdes->rxdes0 = 0;
 
-	return 0;
+	return err;
 }
 
 static unsigned int ftgmac100_next_rx_pointer(struct ftgmac100 *priv,
@@ -574,7 +602,7 @@ static bool ftgmac100_rx_packet(struct ftgmac100 *priv, int *processed)
 	(*processed)++;
 	return true;
 
- drop:
+drop:
 	/* Clean rxdes0 (which resets own bit) */
 	rxdes->rxdes0 = cpu_to_le32(status & priv->rxdes0_edorr_mask);
 	priv->rx_pointer = ftgmac100_next_rx_pointer(priv, pointer);
@@ -658,6 +686,11 @@ static bool ftgmac100_tx_complete_packet(struct ftgmac100 *priv)
 	ftgmac100_free_tx_packet(priv, pointer, skb, txdes, ctl_stat);
 	txdes->txdes0 = cpu_to_le32(ctl_stat & priv->txdes0_edotr_mask);
 
+	/* Ensure the descriptor config is visible before setting the tx
+	 * pointer.
+	 */
+	smp_wmb();
+
 	priv->tx_clean_pointer = ftgmac100_next_tx_pointer(priv, pointer);
 
 	return true;
@@ -707,8 +740,8 @@ static bool ftgmac100_prep_tx_csum(struct sk_buff *skb, u32 *csum_vlan)
 	return skb_checksum_help(skb) == 0;
 }
 
-static int ftgmac100_hard_start_xmit(struct sk_buff *skb,
-				     struct net_device *netdev)
+static netdev_tx_t ftgmac100_hard_start_xmit(struct sk_buff *skb,
+					     struct net_device *netdev)
 {
 	struct ftgmac100 *priv = netdev_priv(netdev);
 	struct ftgmac100_txdes *txdes, *first;
@@ -734,6 +767,18 @@ static int ftgmac100_hard_start_xmit(struct sk_buff *skb,
 	 */
 	nfrags = skb_shinfo(skb)->nr_frags;
 
+	/* Setup HW checksumming */
+	csum_vlan = 0;
+	if (skb->ip_summed == CHECKSUM_PARTIAL &&
+	    !ftgmac100_prep_tx_csum(skb, &csum_vlan))
+		goto drop;
+
+	/* Add VLAN tag */
+	if (skb_vlan_tag_present(skb)) {
+		csum_vlan |= FTGMAC100_TXDES1_INS_VLANTAG;
+		csum_vlan |= skb_vlan_tag_get(skb) & 0xffff;
+	}
+
 	/* Get header len */
 	len = skb_headlen(skb);
 
@@ -760,19 +805,6 @@ static int ftgmac100_hard_start_xmit(struct sk_buff *skb,
 	if (nfrags == 0)
 		f_ctl_stat |= FTGMAC100_TXDES0_LTS;
 	txdes->txdes3 = cpu_to_le32(map);
-
-	/* Setup HW checksumming */
-	csum_vlan = 0;
-	if (skb->ip_summed == CHECKSUM_PARTIAL &&
-	    !ftgmac100_prep_tx_csum(skb, &csum_vlan))
-		goto drop;
-
-	/* Add VLAN tag */
-	if (skb_vlan_tag_present(skb)) {
-		csum_vlan |= FTGMAC100_TXDES1_INS_VLANTAG;
-		csum_vlan |= skb_vlan_tag_get(skb) & 0xffff;
-	}
-
 	txdes->txdes1 = cpu_to_le32(csum_vlan);
 
 	/* Next descriptor */
@@ -782,7 +814,7 @@ static int ftgmac100_hard_start_xmit(struct sk_buff *skb,
 	for (i = 0; i < nfrags; i++) {
 		skb_frag_t *frag = &skb_shinfo(skb)->frags[i];
 
-		len = frag->size;
+		len = skb_frag_size(frag);
 
 		/* Map it */
 		map = skb_frag_dma_map(priv->dev, frag, 0, len,
@@ -812,6 +844,11 @@ static int ftgmac100_hard_start_xmit(struct sk_buff *skb,
 	dma_wmb();
 	first->txdes0 = cpu_to_le32(f_ctl_stat);
 
+	/* Ensure the descriptor config is visible before setting the tx
+	 * pointer.
+	 */
+	smp_wmb();
+
 	/* Update next TX pointer */
 	priv->tx_pointer = pointer;
 
@@ -832,7 +869,7 @@ static int ftgmac100_hard_start_xmit(struct sk_buff *skb,
 
 	return NETDEV_TX_OK;
 
- dma_err:
+dma_err:
 	if (net_ratelimit())
 		netdev_err(netdev, "map tx fragment failed\n");
 
@@ -854,7 +891,7 @@ static int ftgmac100_hard_start_xmit(struct sk_buff *skb,
 	 * last fragment, so we know ftgmac100_free_tx_packet()
 	 * hasn't freed the skb yet.
 	 */
- drop:
+drop:
 	/* Drop the packet */
 	dev_kfree_skb_any(skb);
 	netdev->stats.tx_dropped++;
@@ -930,16 +967,14 @@ static int ftgmac100_alloc_rings(struct ftgmac100 *priv)
 		return -ENOMEM;
 
 	/* Allocate descriptors */
-	priv->rxdes = dma_zalloc_coherent(priv->dev,
-					  MAX_RX_QUEUE_ENTRIES *
-					  sizeof(struct ftgmac100_rxdes),
-					  &priv->rxdes_dma, GFP_KERNEL);
+	priv->rxdes = dma_alloc_coherent(priv->dev,
+					 MAX_RX_QUEUE_ENTRIES * sizeof(struct ftgmac100_rxdes),
+					 &priv->rxdes_dma, GFP_KERNEL);
 	if (!priv->rxdes)
 		return -ENOMEM;
-	priv->txdes = dma_zalloc_coherent(priv->dev,
-					  MAX_TX_QUEUE_ENTRIES *
-					  sizeof(struct ftgmac100_txdes),
-					  &priv->txdes_dma, GFP_KERNEL);
+	priv->txdes = dma_alloc_coherent(priv->dev,
+					 MAX_TX_QUEUE_ENTRIES * sizeof(struct ftgmac100_txdes),
+					 &priv->txdes_dma, GFP_KERNEL);
 	if (!priv->txdes)
 		return -ENOMEM;
 
@@ -1000,89 +1035,6 @@ static int ftgmac100_alloc_rx_buffers(struct ftgmac100 *priv)
 	return 0;
 }
 
-static void ftgmac100_adjust_link(struct net_device *netdev)
-{
-	struct ftgmac100 *priv = netdev_priv(netdev);
-	struct phy_device *phydev = netdev->phydev;
-	bool tx_pause, rx_pause;
-	int new_speed;
-
-	/* We store "no link" as speed 0 */
-	if (!phydev->link)
-		new_speed = 0;
-	else
-		new_speed = phydev->speed;
-
-	/* Grab pause settings from PHY if configured to do so */
-	if (priv->aneg_pause) {
-		rx_pause = tx_pause = phydev->pause;
-		if (phydev->asym_pause)
-			tx_pause = !rx_pause;
-	} else {
-		rx_pause = priv->rx_pause;
-		tx_pause = priv->tx_pause;
-	}
-
-	/* Link hasn't changed, do nothing */
-	if (phydev->speed == priv->cur_speed &&
-	    phydev->duplex == priv->cur_duplex &&
-	    rx_pause == priv->rx_pause &&
-	    tx_pause == priv->tx_pause)
-		return;
-
-	/* Print status if we have a link or we had one and just lost it,
-	 * don't print otherwise.
-	 */
-	if (new_speed || priv->cur_speed)
-		phy_print_status(phydev);
-
-	priv->cur_speed = new_speed;
-	priv->cur_duplex = phydev->duplex;
-	priv->rx_pause = rx_pause;
-	priv->tx_pause = tx_pause;
-
-	/* Link is down, do nothing else */
-	if (!new_speed)
-		return;
-
-	/* Disable all interrupts */
-	iowrite32(0, priv->base + FTGMAC100_OFFSET_IER);
-
-	/* Reset the adapter asynchronously */
-	schedule_work(&priv->reset_task);
-}
-
-static int ftgmac100_mii_probe(struct ftgmac100 *priv, phy_interface_t intf)
-{
-	struct net_device *netdev = priv->netdev;
-	struct phy_device *phydev;
-
-	phydev = phy_find_first(priv->mii_bus);
-	if (!phydev) {
-		netdev_info(netdev, "%s: no PHY found\n", netdev->name);
-		return -ENODEV;
-	}
-
-	phydev = phy_connect(netdev, phydev_name(phydev),
-			     &ftgmac100_adjust_link, intf);
-
-	if (IS_ERR(phydev)) {
-		netdev_err(netdev, "%s: Could not attach to PHY\n", netdev->name);
-		return PTR_ERR(phydev);
-	}
-
-	/* Indicate that we support PAUSE frames (see comment in
-	 * Documentation/networking/phy.txt)
-	 */
-	phydev->supported |= SUPPORTED_Pause | SUPPORTED_Asym_Pause;
-	phydev->advertising = phydev->supported;
-
-	/* Display what we found */
-	phy_attached_info(phydev);
-
-	return 0;
-}
-
 static int ftgmac100_mdiobus_read(struct mii_bus *bus, int phy_addr, int regnum)
 {
 	struct net_device *netdev = bus->priv;
@@ -1157,13 +1109,15 @@ static int ftgmac100_mdiobus_write(struct mii_bus *bus, int phy_addr,
 static void ftgmac100_get_drvinfo(struct net_device *netdev,
 				  struct ethtool_drvinfo *info)
 {
-	strlcpy(info->driver, DRV_NAME, sizeof(info->driver));
-	strlcpy(info->version, DRV_VERSION, sizeof(info->version));
-	strlcpy(info->bus_info, dev_name(&netdev->dev), sizeof(info->bus_info));
+	strscpy(info->driver, DRV_NAME, sizeof(info->driver));
+	strscpy(info->bus_info, dev_name(&netdev->dev), sizeof(info->bus_info));
 }
 
-static void ftgmac100_get_ringparam(struct net_device *netdev,
-				    struct ethtool_ringparam *ering)
+static void
+ftgmac100_get_ringparam(struct net_device *netdev,
+			struct ethtool_ringparam *ering,
+			struct kernel_ethtool_ringparam *kernel_ering,
+			struct netlink_ext_ack *extack)
 {
 	struct ftgmac100 *priv = netdev_priv(netdev);
 
@@ -1174,8 +1128,11 @@ static void ftgmac100_get_ringparam(struct net_device *netdev,
 	ering->tx_pending = priv->tx_q_entries;
 }
 
-static int ftgmac100_set_ringparam(struct net_device *netdev,
-				   struct ethtool_ringparam *ering)
+static int
+ftgmac100_set_ringparam(struct net_device *netdev,
+			struct ethtool_ringparam *ering,
+			struct kernel_ethtool_ringparam *kernel_ering,
+			struct netlink_ext_ack *extack)
 {
 	struct ftgmac100 *priv = netdev_priv(netdev);
 
@@ -1215,22 +1172,11 @@ static int ftgmac100_set_pauseparam(struct net_device *netdev,
 	priv->tx_pause = pause->tx_pause;
 	priv->rx_pause = pause->rx_pause;
 
-	if (phydev) {
-		phydev->advertising &= ~ADVERTISED_Pause;
-		phydev->advertising &= ~ADVERTISED_Asym_Pause;
+	if (phydev)
+		phy_set_asym_pause(phydev, pause->rx_pause, pause->tx_pause);
 
-		if (pause->rx_pause) {
-			phydev->advertising |= ADVERTISED_Pause;
-			phydev->advertising |= ADVERTISED_Asym_Pause;
-		}
-
-		if (pause->tx_pause)
-			phydev->advertising ^= ADVERTISED_Asym_Pause;
-	}
 	if (netif_running(netdev)) {
-		if (phydev && priv->aneg_pause)
-			phy_start_aneg(phydev);
-		else
+		if (!(phydev && priv->aneg_pause))
 			ftgmac100_config_pause(priv);
 	}
 
@@ -1329,6 +1275,7 @@ static int ftgmac100_poll(struct napi_struct *napi, int budget)
 	 */
 	if (unlikely(priv->need_mac_restart)) {
 		ftgmac100_start_hw(priv);
+		priv->need_mac_restart = false;
 
 		/* Re-enable "bad" interrupts */
 		iowrite32(FTGMAC100_INT_BAD,
@@ -1398,10 +1345,8 @@ static int ftgmac100_init_all(struct ftgmac100 *priv, bool ignore_alloc_err)
 	return err;
 }
 
-static void ftgmac100_reset_task(struct work_struct *work)
+static void ftgmac100_reset(struct ftgmac100 *priv)
 {
-	struct ftgmac100 *priv = container_of(work, struct ftgmac100,
-					      reset_task);
 	struct net_device *netdev = priv->netdev;
 	int err;
 
@@ -1439,7 +1384,7 @@ static void ftgmac100_reset_task(struct work_struct *work)
 	ftgmac100_init_all(priv, true);
 
 	netdev_dbg(netdev, "Reset done !\n");
- bail:
+bail:
 	if (priv->mii_bus)
 		mutex_unlock(&priv->mii_bus->mdio_lock);
 	if (netdev->phydev)
@@ -1447,6 +1392,134 @@ static void ftgmac100_reset_task(struct work_struct *work)
 	rtnl_unlock();
 }
 
+static void ftgmac100_reset_task(struct work_struct *work)
+{
+	struct ftgmac100 *priv = container_of(work, struct ftgmac100,
+					      reset_task);
+
+	ftgmac100_reset(priv);
+}
+
+static void ftgmac100_adjust_link(struct net_device *netdev)
+{
+	struct ftgmac100 *priv = netdev_priv(netdev);
+	struct phy_device *phydev = netdev->phydev;
+	bool tx_pause, rx_pause;
+	int new_speed;
+
+	/* We store "no link" as speed 0 */
+	if (!phydev->link)
+		new_speed = 0;
+	else
+		new_speed = phydev->speed;
+
+	/* Grab pause settings from PHY if configured to do so */
+	if (priv->aneg_pause) {
+		rx_pause = tx_pause = phydev->pause;
+		if (phydev->asym_pause)
+			tx_pause = !rx_pause;
+	} else {
+		rx_pause = priv->rx_pause;
+		tx_pause = priv->tx_pause;
+	}
+
+	/* Link hasn't changed, do nothing */
+	if (phydev->speed == priv->cur_speed &&
+	    phydev->duplex == priv->cur_duplex &&
+	    rx_pause == priv->rx_pause &&
+	    tx_pause == priv->tx_pause)
+		return;
+
+	/* Print status if we have a link or we had one and just lost it,
+	 * don't print otherwise.
+	 */
+	if (new_speed || priv->cur_speed)
+		phy_print_status(phydev);
+
+	priv->cur_speed = new_speed;
+	priv->cur_duplex = phydev->duplex;
+	priv->rx_pause = rx_pause;
+	priv->tx_pause = tx_pause;
+
+	/* Link is down, do nothing else */
+	if (!new_speed)
+		return;
+
+	/* Disable all interrupts */
+	iowrite32(0, priv->base + FTGMAC100_OFFSET_IER);
+
+	/* Release phy lock to allow ftgmac100_reset to acquire it, keeping lock
+	 * order consistent to prevent dead lock.
+	 */
+	if (netdev->phydev)
+		mutex_unlock(&netdev->phydev->lock);
+
+	ftgmac100_reset(priv);
+
+	if (netdev->phydev)
+		mutex_lock(&netdev->phydev->lock);
+
+}
+
+static int ftgmac100_mii_probe(struct net_device *netdev)
+{
+	struct ftgmac100 *priv = netdev_priv(netdev);
+	struct platform_device *pdev = to_platform_device(priv->dev);
+	struct device_node *np = pdev->dev.of_node;
+	struct phy_device *phydev;
+	phy_interface_t phy_intf;
+	int err;
+
+	/* Default to RGMII. It's a gigabit part after all */
+	err = of_get_phy_mode(np, &phy_intf);
+	if (err)
+		phy_intf = PHY_INTERFACE_MODE_RGMII;
+
+	/* Aspeed only supports these. I don't know about other IP
+	 * block vendors so I'm going to just let them through for
+	 * now. Note that this is only a warning if for some obscure
+	 * reason the DT really means to lie about it or it's a newer
+	 * part we don't know about.
+	 *
+	 * On the Aspeed SoC there are additionally straps and SCU
+	 * control bits that could tell us what the interface is
+	 * (or allow us to configure it while the IP block is held
+	 * in reset). For now I chose to keep this driver away from
+	 * those SoC specific bits and assume the device-tree is
+	 * right and the SCU has been configured properly by pinmux
+	 * or the firmware.
+	 */
+	if (priv->is_aspeed && !(phy_interface_mode_is_rgmii(phy_intf))) {
+		netdev_warn(netdev,
+			    "Unsupported PHY mode %s !\n",
+			    phy_modes(phy_intf));
+	}
+
+	phydev = phy_find_first(priv->mii_bus);
+	if (!phydev) {
+		netdev_info(netdev, "%s: no PHY found\n", netdev->name);
+		return -ENODEV;
+	}
+
+	phydev = phy_connect(netdev, phydev_name(phydev),
+			     &ftgmac100_adjust_link, phy_intf);
+
+	if (IS_ERR(phydev)) {
+		netdev_err(netdev, "%s: Could not attach to PHY\n", netdev->name);
+		return PTR_ERR(phydev);
+	}
+
+	/* Indicate that we support PAUSE frames (see comment in
+	 * Documentation/networking/phy.rst)
+	 */
+	phy_support_asym_pause(phydev);
+
+	/* Display what we found */
+	phy_attached_info(phydev);
+
+	return 0;
+}
+
 static int ftgmac100_open(struct net_device *netdev)
 {
 	struct ftgmac100 *priv = netdev_priv(netdev);
@@ -1479,7 +1552,7 @@ static int ftgmac100_open(struct net_device *netdev)
 		goto err_hw;
 
 	/* Initialize NAPI */
-	netif_napi_add(netdev, &priv->napi, ftgmac100_poll, 64);
+	netif_napi_add(netdev, &priv->napi, ftgmac100_poll);
 
 	/* Grab our interrupt */
 	err = request_irq(netdev->irq, ftgmac100_interrupt, 0, netdev->name, netdev);
@@ -1498,7 +1571,8 @@ static int ftgmac100_open(struct net_device *netdev)
 	if (netdev->phydev) {
 		/* If we have a PHY, start polling */
 		phy_start(netdev->phydev);
-	} else if (priv->use_ncsi) {
+	}
+	if (priv->use_ncsi) {
 		/* If using NC-SI, set our carrier on and start the stack */
 		netif_carrier_on(netdev);
 
@@ -1510,15 +1584,16 @@ static int ftgmac100_open(struct net_device *netdev)
 
 	return 0;
 
- err_ncsi:
+err_ncsi:
+	phy_stop(netdev->phydev);
 	napi_disable(&priv->napi);
 	netif_stop_queue(netdev);
- err_alloc:
+err_alloc:
 	ftgmac100_free_buffers(priv);
 	free_irq(netdev->irq, netdev);
- err_irq:
+err_irq:
 	netif_napi_del(&priv->napi);
- err_hw:
+err_hw:
 	iowrite32(0, priv->base + FTGMAC100_OFFSET_IER);
 	ftgmac100_free_rings(priv);
 	return err;
@@ -1544,7 +1619,7 @@ static int ftgmac100_stop(struct net_device *netdev)
 	netif_napi_del(&priv->napi);
 	if (netdev->phydev)
 		phy_stop(netdev->phydev);
-	else if (priv->use_ncsi)
+	if (priv->use_ncsi)
 		ncsi_stop_dev(priv->ndev);
 
 	ftgmac100_stop_hw(priv);
@@ -1555,16 +1630,7 @@ static int ftgmac100_stop(struct net_device *netdev)
 	return 0;
 }
 
-/* optional */
-static int ftgmac100_do_ioctl(struct net_device *netdev, struct ifreq *ifr, int cmd)
-{
-	if (!netdev->phydev)
-		return -ENXIO;
-
-	return phy_mii_ioctl(netdev->phydev, ifr, cmd);
-}
-
-static void ftgmac100_tx_timeout(struct net_device *netdev)
+static void ftgmac100_tx_timeout(struct net_device *netdev, unsigned int txqueue)
 {
 	struct ftgmac100 *priv = netdev_priv(netdev);
 
@@ -1616,21 +1682,23 @@ static const struct net_device_ops ftgmac100_netdev_ops = {
 	.ndo_start_xmit		= ftgmac100_hard_start_xmit,
 	.ndo_set_mac_address	= ftgmac100_set_mac_addr,
 	.ndo_validate_addr	= eth_validate_addr,
-	.ndo_do_ioctl		= ftgmac100_do_ioctl,
+	.ndo_eth_ioctl		= phy_do_ioctl,
 	.ndo_tx_timeout		= ftgmac100_tx_timeout,
 	.ndo_set_rx_mode	= ftgmac100_set_rx_mode,
 	.ndo_set_features	= ftgmac100_set_features,
 #ifdef CONFIG_NET_POLL_CONTROLLER
 	.ndo_poll_controller	= ftgmac100_poll_controller,
 #endif
+	.ndo_vlan_rx_add_vid	= ncsi_vlan_rx_add_vid,
+	.ndo_vlan_rx_kill_vid	= ncsi_vlan_rx_kill_vid,
 };
 
 static int ftgmac100_setup_mdio(struct net_device *netdev)
 {
 	struct ftgmac100 *priv = netdev_priv(netdev);
 	struct platform_device *pdev = to_platform_device(priv->dev);
-	int phy_intf = PHY_INTERFACE_MODE_RGMII;
 	struct device_node *np = pdev->dev.of_node;
+	struct device_node *mdio_np;
 	int i, err = 0;
 	u32 reg;
 
@@ -1639,49 +1707,22 @@ static int ftgmac100_setup_mdio(struct net_device *netdev)
 	if (!priv->mii_bus)
 		return -EIO;
 
-	if (priv->is_aspeed) {
-		/* This driver supports the old MDIO interface */
+	if (of_device_is_compatible(np, "aspeed,ast2400-mac") ||
+	    of_device_is_compatible(np, "aspeed,ast2500-mac")) {
+		/* The AST2600 has a separate MDIO controller */
+
+		/* For the AST2400 and AST2500 this driver only supports the
+		 * old MDIO interface
+		 */
 		reg = ioread32(priv->base + FTGMAC100_OFFSET_REVR);
 		reg &= ~FTGMAC100_REVR_NEW_MDIO_INTERFACE;
 		iowrite32(reg, priv->base + FTGMAC100_OFFSET_REVR);
-	};
-
-	/* Get PHY mode from device-tree */
-	if (np) {
-		/* Default to RGMII. It's a gigabit part after all */
-		phy_intf = of_get_phy_mode(np);
-		if (phy_intf < 0)
-			phy_intf = PHY_INTERFACE_MODE_RGMII;
-
-		/* Aspeed only supports these. I don't know about other IP
-		 * block vendors so I'm going to just let them through for
-		 * now. Note that this is only a warning if for some obscure
-		 * reason the DT really means to lie about it or it's a newer
-		 * part we don't know about.
-		 *
-		 * On the Aspeed SoC there are additionally straps and SCU
-		 * control bits that could tell us what the interface is
-		 * (or allow us to configure it while the IP block is held
-		 * in reset). For now I chose to keep this driver away from
-		 * those SoC specific bits and assume the device-tree is
-		 * right and the SCU has been configured properly by pinmux
-		 * or the firmware.
-		 */
-		if (priv->is_aspeed &&
-		    phy_intf != PHY_INTERFACE_MODE_RMII &&
-		    phy_intf != PHY_INTERFACE_MODE_RGMII &&
-		    phy_intf != PHY_INTERFACE_MODE_RGMII_ID &&
-		    phy_intf != PHY_INTERFACE_MODE_RGMII_RXID &&
-		    phy_intf != PHY_INTERFACE_MODE_RGMII_TXID) {
-			netdev_warn(netdev,
-				   "Unsupported PHY mode %s !\n",
-				   phy_modes(phy_intf));
-		}
 	}
 
 	priv->mii_bus->name = "ftgmac100_mdio";
 	snprintf(priv->mii_bus->id, MII_BUS_ID_SIZE, "%s-%d",
 		 pdev->name, pdev->id);
+	priv->mii_bus->parent = priv->dev;
 	priv->mii_bus->priv = priv->netdev;
 	priv->mii_bus->read = ftgmac100_mdiobus_read;
 	priv->mii_bus->write = ftgmac100_mdiobus_write;
@@ -1689,35 +1730,46 @@ static int ftgmac100_setup_mdio(struct net_device *netdev)
 	for (i = 0; i < PHY_MAX_ADDR; i++)
 		priv->mii_bus->irq[i] = PHY_POLL;
 
-	err = mdiobus_register(priv->mii_bus);
+	mdio_np = of_get_child_by_name(np, "mdio");
+
+	err = of_mdiobus_register(priv->mii_bus, mdio_np);
 	if (err) {
 		dev_err(priv->dev, "Cannot register MDIO bus!\n");
 		goto err_register_mdiobus;
 	}
 
-	err = ftgmac100_mii_probe(priv, phy_intf);
-	if (err) {
-		dev_err(priv->dev, "MII Probe failed!\n");
-		goto err_mii_probe;
-	}
+	of_node_put(mdio_np);
 
 	return 0;
 
-err_mii_probe:
-	mdiobus_unregister(priv->mii_bus);
 err_register_mdiobus:
 	mdiobus_free(priv->mii_bus);
 	return err;
 }
 
+static void ftgmac100_phy_disconnect(struct net_device *netdev)
+{
+	struct ftgmac100 *priv = netdev_priv(netdev);
+	struct phy_device *phydev = netdev->phydev;
+
+	if (!phydev)
+		return;
+
+	phy_disconnect(phydev);
+	if (of_phy_is_fixed_link(priv->dev->of_node))
+		of_phy_deregister_fixed_link(priv->dev->of_node);
+
+	if (priv->use_ncsi)
+		fixed_phy_unregister(phydev);
+}
+
 static void ftgmac100_destroy_mdio(struct net_device *netdev)
 {
 	struct ftgmac100 *priv = netdev_priv(netdev);
 
-	if (!netdev->phydev)
+	if (!priv->mii_bus)
 		return;
 
-	phy_disconnect(netdev->phydev);
 	mdiobus_unregister(priv->mii_bus);
 	mdiobus_free(priv->mii_bus);
 }
@@ -1727,8 +1779,58 @@ static void ftgmac100_ncsi_handler(struct ncsi_dev *nd)
 	if (unlikely(nd->state != ncsi_dev_state_functional))
 		return;
 
-	netdev_info(nd->dev, "NCSI interface %s\n",
-		    nd->link_up ? "up" : "down");
+	netdev_dbg(nd->dev, "NCSI interface %s\n",
+		   nd->link_up ? "up" : "down");
+}
+
+static int ftgmac100_setup_clk(struct ftgmac100 *priv)
+{
+	struct clk *clk;
+	int rc;
+
+	clk = devm_clk_get(priv->dev, NULL /* MACCLK */);
+	if (IS_ERR(clk))
+		return PTR_ERR(clk);
+	priv->clk = clk;
+	rc = clk_prepare_enable(priv->clk);
+	if (rc)
+		return rc;
+
+	/* Aspeed specifies a 100MHz clock is required for up to
+	 * 1000Mbit link speeds. As NCSI is limited to 100Mbit, 25MHz
+	 * is sufficient
+	 */
+	rc = clk_set_rate(priv->clk, priv->use_ncsi ? FTGMAC_25MHZ :
+			  FTGMAC_100MHZ);
+	if (rc)
+		goto cleanup_clk;
+
+	/* RCLK is for RMII, typically used for NCSI. Optional because it's not
+	 * necessary if it's the AST2400 MAC, or the MAC is configured for
+	 * RGMII, or the controller is not an ASPEED-based controller.
+	 */
+	priv->rclk = devm_clk_get_optional(priv->dev, "RCLK");
+	rc = clk_prepare_enable(priv->rclk);
+	if (!rc)
+		return 0;
+
+cleanup_clk:
+	clk_disable_unprepare(priv->clk);
+
+	return rc;
+}
+
+static bool ftgmac100_has_child_node(struct device_node *np, const char *name)
+{
+	struct device_node *child_np = of_get_child_by_name(np, name);
+	bool ret = false;
+
+	if (child_np) {
+		ret = true;
+		of_node_put(child_np);
+	}
+
+	return ret;
 }
 
 static int ftgmac100_probe(struct platform_device *pdev)
@@ -1736,13 +1838,11 @@ static int ftgmac100_probe(struct platform_device *pdev)
 	struct resource *res;
 	int irq;
 	struct net_device *netdev;
+	struct phy_device *phydev;
 	struct ftgmac100 *priv;
 	struct device_node *np;
 	int err = 0;
 
-	if (!pdev)
-		return -ENODEV;
-
 	res = platform_get_resource(pdev, IORESOURCE_MEM, 0);
 	if (!res)
 		return -ENXIO;
@@ -1796,11 +1896,14 @@ static int ftgmac100_probe(struct platform_device *pdev)
 	priv->aneg_pause = true;
 
 	/* MAC address from chip or random one */
-	ftgmac100_initial_mac(priv);
+	err = ftgmac100_initial_mac(priv);
+	if (err)
+		goto err_phy_connect;
 
 	np = pdev->dev.of_node;
 	if (np && (of_device_is_compatible(np, "aspeed,ast2400-mac") ||
-		   of_device_is_compatible(np, "aspeed,ast2500-mac"))) {
+		   of_device_is_compatible(np, "aspeed,ast2500-mac") ||
+		   of_device_is_compatible(np, "aspeed,ast2600-mac"))) {
 		priv->rxdes0_edorr_mask = BIT(30);
 		priv->txdes0_edotr_mask = BIT(30);
 		priv->is_aspeed = true;
@@ -1812,19 +1915,95 @@ static int ftgmac100_probe(struct platform_device *pdev)
 	if (np && of_get_property(np, "use-ncsi", NULL)) {
 		if (!IS_ENABLED(CONFIG_NET_NCSI)) {
 			dev_err(&pdev->dev, "NCSI stack not enabled\n");
-			goto err_ncsi_dev;
+			err = -EINVAL;
+			goto err_phy_connect;
 		}
 
 		dev_info(&pdev->dev, "Using NCSI interface\n");
 		priv->use_ncsi = true;
 		priv->ndev = ncsi_register_dev(netdev, ftgmac100_ncsi_handler);
-		if (!priv->ndev)
-			goto err_ncsi_dev;
-	} else {
+		if (!priv->ndev) {
+			err = -EINVAL;
+			goto err_phy_connect;
+		}
+
+		phydev = fixed_phy_register(&ncsi_phy_status, np);
+		if (IS_ERR(phydev)) {
+			dev_err(&pdev->dev, "failed to register fixed PHY device\n");
+			err = PTR_ERR(phydev);
+			goto err_phy_connect;
+		}
+		err = phy_connect_direct(netdev, phydev, ftgmac100_adjust_link,
+					 PHY_INTERFACE_MODE_RMII);
+		if (err) {
+			dev_err(&pdev->dev, "Connecting PHY failed\n");
+			goto err_phy_connect;
+		}
+	} else if (np && (of_phy_is_fixed_link(np) ||
+			  of_get_property(np, "phy-handle", NULL))) {
+		struct phy_device *phy;
+
+		/* Support "mdio"/"phy" child nodes for ast2400/2500 with
+		 * an embedded MDIO controller. Automatically scan the DTS for
+		 * available PHYs and register them.
+		 */
+		if (of_get_property(np, "phy-handle", NULL) &&
+		    (of_device_is_compatible(np, "aspeed,ast2400-mac") ||
+		     of_device_is_compatible(np, "aspeed,ast2500-mac"))) {
+			err = ftgmac100_setup_mdio(netdev);
+			if (err)
+				goto err_setup_mdio;
+		}
+
+		phy = of_phy_get_and_connect(priv->netdev, np,
+					     &ftgmac100_adjust_link);
+		if (!phy) {
+			dev_err(&pdev->dev, "Failed to connect to phy\n");
+			err = -EINVAL;
+			goto err_phy_connect;
+		}
+
+		/* Indicate that we support PAUSE frames (see comment in
+		 * Documentation/networking/phy.rst)
+		 */
+		phy_support_asym_pause(phy);
+
+		/* Display what we found */
+		phy_attached_info(phy);
+	} else if (np && !ftgmac100_has_child_node(np, "mdio")) {
+		/* Support legacy ASPEED devicetree descriptions that decribe a
+		 * MAC with an embedded MDIO controller but have no "mdio"
+		 * child node. Automatically scan the MDIO bus for available
+		 * PHYs.
+		 */
 		priv->use_ncsi = false;
 		err = ftgmac100_setup_mdio(netdev);
 		if (err)
 			goto err_setup_mdio;
+
+		err = ftgmac100_mii_probe(netdev);
+		if (err) {
+			dev_err(priv->dev, "MII probe failed!\n");
+			goto err_ncsi_dev;
+		}
+
+	}
+
+	priv->rst = devm_reset_control_get_optional_exclusive(priv->dev, NULL);
+	if (IS_ERR(priv->rst)) {
+		err = PTR_ERR(priv->rst);
+		goto err_phy_connect;
+	}
+
+	if (priv->is_aspeed) {
+		err = ftgmac100_setup_clk(priv);
+		if (err)
+			goto err_phy_connect;
+
+		/* Disable ast2600 problematic HW arbitration */
+		if (of_device_is_compatible(np, "aspeed,ast2600-mac"))
+			iowrite32(FTGMAC100_TM_DEFAULT,
+				  priv->base + FTGMAC100_OFFSET_TM);
 	}
 
 	/* Default ring sizes */
@@ -1836,9 +2015,17 @@ static int ftgmac100_probe(struct platform_device *pdev)
 		NETIF_F_GRO | NETIF_F_SG | NETIF_F_HW_VLAN_CTAG_RX |
 		NETIF_F_HW_VLAN_CTAG_TX;
 
+	if (priv->use_ncsi)
+		netdev->hw_features |= NETIF_F_HW_VLAN_CTAG_FILTER;
+
 	/* AST2400  doesn't have working HW checksum generation */
 	if (np && (of_device_is_compatible(np, "aspeed,ast2400-mac")))
 		netdev->hw_features &= ~NETIF_F_HW_CSUM;
+
+	/* AST2600 tx checksum with NCSI is broken */
+	if (priv->use_ncsi && of_device_is_compatible(np, "aspeed,ast2600-mac"))
+		netdev->hw_features &= ~NETIF_F_HW_CSUM;
+
 	if (np && of_get_property(np, "no-hw-checksum", NULL))
 		netdev->hw_features &= ~(NETIF_F_HW_CSUM | NETIF_F_RXCSUM);
 	netdev->features |= netdev->hw_features;
@@ -1854,21 +2041,26 @@ static int ftgmac100_probe(struct platform_device *pdev)
 
 	return 0;
 
-err_ncsi_dev:
 err_register_netdev:
+	clk_disable_unprepare(priv->rclk);
+	clk_disable_unprepare(priv->clk);
+err_phy_connect:
+	ftgmac100_phy_disconnect(netdev);
+err_ncsi_dev:
+	if (priv->ndev)
+		ncsi_unregister_dev(priv->ndev);
 	ftgmac100_destroy_mdio(netdev);
 err_setup_mdio:
 	iounmap(priv->base);
 err_ioremap:
 	release_resource(priv->res);
 err_req_mem:
-	netif_napi_del(&priv->napi);
 	free_netdev(netdev);
 err_alloc_etherdev:
 	return err;
 }
 
-static int ftgmac100_remove(struct platform_device *pdev)
+static void ftgmac100_remove(struct platform_device *pdev)
 {
 	struct net_device *netdev;
 	struct ftgmac100 *priv;
@@ -1876,13 +2068,19 @@ static int ftgmac100_remove(struct platform_device *pdev)
 	netdev = platform_get_drvdata(pdev);
 	priv = netdev_priv(netdev);
 
+	if (priv->ndev)
+		ncsi_unregister_dev(priv->ndev);
 	unregister_netdev(netdev);
 
+	clk_disable_unprepare(priv->rclk);
+	clk_disable_unprepare(priv->clk);
+
 	/* There's a small chance the reset task will have been re-queued,
 	 * during stop, make sure it's gone before we free the structure.
 	 */
 	cancel_work_sync(&priv->reset_task);
 
+	ftgmac100_phy_disconnect(netdev);
 	ftgmac100_destroy_mdio(netdev);
 
 	iounmap(priv->base);
@@ -1890,7 +2088,6 @@ static int ftgmac100_remove(struct platform_device *pdev)
 
 	netif_napi_del(&priv->napi);
 	free_netdev(netdev);
-	return 0;
 }
 
 static const struct of_device_id ftgmac100_of_match[] = {
@@ -1901,7 +2098,7 @@ MODULE_DEVICE_TABLE(of, ftgmac100_of_match);
 
 static struct platform_driver ftgmac100_driver = {
 	.probe	= ftgmac100_probe,
-	.remove	= ftgmac100_remove,
+	.remove = ftgmac100_remove,
 	.driver	= {
 		.name		= DRV_NAME,
 		.of_match_table	= ftgmac100_of_match,
diff --git a/drivers/net/ethernet/faraday/ftgmac100.h b/drivers/net/ethernet/faraday/ftgmac100.h
index 0653d8176e6a..4968f6f0bdbc 100644
--- a/drivers/net/ethernet/faraday/ftgmac100.h
+++ b/drivers/net/ethernet/faraday/ftgmac100.h
@@ -1,22 +1,9 @@
+/* SPDX-License-Identifier: GPL-2.0-or-later */
 /*
  * Faraday FTGMAC100 Gigabit Ethernet
  *
  * (C) Copyright 2009-2011 Faraday Technology
  * Po-Yu Chuang <ratbert@faraday-tech.com>
- *
- * This program is free software; you can redistribute it and/or modify
- * it under the terms of the GNU General Public License as published by
- * the Free Software Foundation; either version 2 of the License, or
- * (at your option) any later version.
- *
- * This program is distributed in the hope that it will be useful,
- * but WITHOUT ANY WARRANTY; without even the implied warranty of
- * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
- * GNU General Public License for more details.
- *
- * You should have received a copy of the GNU General Public License
- * along with this program; if not, write to the Free Software
- * Foundation, Inc., 675 Mass Ave, Cambridge, MA 02139, USA.
  */
 
 #ifndef __FTGMAC100_H
@@ -97,7 +84,7 @@
 			    FTGMAC100_INT_RPKT_BUF)
 
 /* All the interrupts we care about */
-#define FTGMAC100_INT_ALL (FTGMAC100_INT_RPKT_BUF  |  \
+#define FTGMAC100_INT_ALL (FTGMAC100_INT_RXTX  |  \
 			   FTGMAC100_INT_BAD)
 
 /*
@@ -183,6 +170,14 @@
 #define FTGMAC100_MACCR_SW_RST		(1 << 31)
 
 /*
+ * test mode control register
+ */
+#define FTGMAC100_TM_RQ_TX_VALID_DIS (1 << 28)
+#define FTGMAC100_TM_RQ_RR_IDLE_PREV (1 << 27)
+#define FTGMAC100_TM_DEFAULT                                                   \
+	(FTGMAC100_TM_RQ_TX_VALID_DIS | FTGMAC100_TM_RQ_RR_IDLE_PREV)
+
+/*
  * PHY control register
  */
 #define FTGMAC100_PHYCR_MDC_CYCTHR_MASK	0x3f
diff --git a/drivers/net/ethernet/faraday/ftmac100.c b/drivers/net/ethernet/faraday/ftmac100.c
index 1536356e2ea8..5803a382f0ba 100644
--- a/drivers/net/ethernet/faraday/ftmac100.c
+++ b/drivers/net/ethernet/faraday/ftmac100.c
@@ -1,22 +1,9 @@
+// SPDX-License-Identifier: GPL-2.0-or-later
 /*
  * Faraday FTMAC100 10/100 Ethernet
  *
  * (C) Copyright 2009-2011 Faraday Technology
  * Po-Yu Chuang <ratbert@faraday-tech.com>
- *
- * This program is free software; you can redistribute it and/or modify
- * it under the terms of the GNU General Public License as published by
- * the Free Software Foundation; either version 2 of the License, or
- * (at your option) any later version.
- *
- * This program is distributed in the hope that it will be useful,
- * but WITHOUT ANY WARRANTY; without even the implied warranty of
- * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
- * GNU General Public License for more details.
- *
- * You should have received a copy of the GNU General Public License
- * along with this program; if not, write to the Free Software
- * Foundation, Inc., 675 Mass Ave, Cambridge, MA 02139, USA.
  */
 
 #define pr_fmt(fmt)	KBUILD_MODNAME ": " fmt
@@ -24,24 +11,26 @@
 #include <linux/dma-mapping.h>
 #include <linux/etherdevice.h>
 #include <linux/ethtool.h>
+#include <linux/if_ether.h>
+#include <linux/if_vlan.h>
 #include <linux/init.h>
 #include <linux/interrupt.h>
 #include <linux/io.h>
 #include <linux/mii.h>
 #include <linux/module.h>
+#include <linux/mod_devicetable.h>
 #include <linux/netdevice.h>
 #include <linux/platform_device.h>
 
 #include "ftmac100.h"
 
 #define DRV_NAME	"ftmac100"
-#define DRV_VERSION	"0.2"
 
 #define RX_QUEUE_ENTRIES	128	/* must be power of 2 */
 #define TX_QUEUE_ENTRIES	16	/* must be power of 2 */
 
-#define MAX_PKT_SIZE		1518
 #define RX_BUF_SIZE		2044	/* must be smaller than 0x7ff */
+#define MAX_PKT_SIZE		RX_BUF_SIZE /* multi-segment not supported */
 
 #if MAX_PKT_SIZE > 0x7ff
 #error invalid MAX_PKT_SIZE
@@ -160,6 +149,40 @@ static void ftmac100_set_mac(struct ftmac100 *priv, const unsigned char *mac)
 	iowrite32(laddr, priv->base + FTMAC100_OFFSET_MAC_LADR);
 }
 
+static void ftmac100_setup_mc_ht(struct ftmac100 *priv)
+{
+	struct netdev_hw_addr *ha;
+	u64 maht = 0; /* Multicast Address Hash Table */
+
+	netdev_for_each_mc_addr(ha, priv->netdev) {
+		u32 hash = ether_crc(ETH_ALEN, ha->addr) >> 26;
+
+		maht |= BIT_ULL(hash);
+	}
+
+	iowrite32(lower_32_bits(maht), priv->base + FTMAC100_OFFSET_MAHT0);
+	iowrite32(upper_32_bits(maht), priv->base + FTMAC100_OFFSET_MAHT1);
+}
+
+static void ftmac100_set_rx_bits(struct ftmac100 *priv, unsigned int *maccr)
+{
+	struct net_device *netdev = priv->netdev;
+
+	/* Clear all */
+	*maccr &= ~(FTMAC100_MACCR_RCV_ALL | FTMAC100_MACCR_RX_MULTIPKT |
+		   FTMAC100_MACCR_HT_MULTI_EN);
+
+	/* Set the requested bits */
+	if (netdev->flags & IFF_PROMISC)
+		*maccr |= FTMAC100_MACCR_RCV_ALL;
+	if (netdev->flags & IFF_ALLMULTI)
+		*maccr |= FTMAC100_MACCR_RX_MULTIPKT;
+	else if (netdev_mc_count(netdev)) {
+		*maccr |= FTMAC100_MACCR_HT_MULTI_EN;
+		ftmac100_setup_mc_ht(priv);
+	}
+}
+
 #define MACCR_ENABLE_ALL	(FTMAC100_MACCR_XMT_EN	| \
 				 FTMAC100_MACCR_RCV_EN	| \
 				 FTMAC100_MACCR_XDMA_EN	| \
@@ -172,6 +195,7 @@ static void ftmac100_set_mac(struct ftmac100 *priv, const unsigned char *mac)
 static int ftmac100_start_hw(struct ftmac100 *priv)
 {
 	struct net_device *netdev = priv->netdev;
+	unsigned int maccr = MACCR_ENABLE_ALL;
 
 	if (ftmac100_reset(priv))
 		return -EIO;
@@ -188,7 +212,13 @@ static int ftmac100_start_hw(struct ftmac100 *priv)
 
 	ftmac100_set_mac(priv, netdev->dev_addr);
 
-	iowrite32(MACCR_ENABLE_ALL, priv->base + FTMAC100_OFFSET_MACCR);
+	 /* See ftmac100_change_mtu() */
+	if (netdev->mtu > ETH_DATA_LEN)
+		maccr |= FTMAC100_MACCR_RX_FTL;
+
+	ftmac100_set_rx_bits(priv, &maccr);
+
+	iowrite32(maccr, priv->base + FTMAC100_OFFSET_MACCR);
 	return 0;
 }
 
@@ -231,11 +261,6 @@ static bool ftmac100_rxdes_crc_error(struct ftmac100_rxdes *rxdes)
 	return rxdes->rxdes0 & cpu_to_le32(FTMAC100_RXDES0_CRC_ERR);
 }
 
-static bool ftmac100_rxdes_frame_too_long(struct ftmac100_rxdes *rxdes)
-{
-	return rxdes->rxdes0 & cpu_to_le32(FTMAC100_RXDES0_FTL);
-}
-
 static bool ftmac100_rxdes_runt(struct ftmac100_rxdes *rxdes)
 {
 	return rxdes->rxdes0 & cpu_to_le32(FTMAC100_RXDES0_RUNT);
@@ -350,13 +375,7 @@ static bool ftmac100_rx_packet_error(struct ftmac100 *priv,
 		error = true;
 	}
 
-	if (unlikely(ftmac100_rxdes_frame_too_long(rxdes))) {
-		if (net_ratelimit())
-			netdev_info(netdev, "rx frame too long\n");
-
-		netdev->stats.rx_length_errors++;
-		error = true;
-	} else if (unlikely(ftmac100_rxdes_runt(rxdes))) {
+	if (unlikely(ftmac100_rxdes_runt(rxdes))) {
 		if (net_ratelimit())
 			netdev_info(netdev, "rx runt\n");
 
@@ -369,6 +388,11 @@ static bool ftmac100_rx_packet_error(struct ftmac100 *priv,
 		netdev->stats.rx_length_errors++;
 		error = true;
 	}
+	/*
+	 * FTMAC100_RXDES0_FTL is not an error, it just indicates that the
+	 * frame is longer than 1518 octets. Receiving these is possible when
+	 * we told the hardware not to drop them, via FTMAC100_MACCR_RX_FTL.
+	 */
 
 	return error;
 }
@@ -402,6 +426,7 @@ static bool ftmac100_rx_packet(struct ftmac100 *priv, int *processed)
 	struct page *page;
 	dma_addr_t map;
 	int length;
+	bool ret;
 
 	rxdes = ftmac100_rx_locate_first_segment(priv);
 	if (!rxdes)
@@ -412,12 +437,13 @@ static bool ftmac100_rx_packet(struct ftmac100 *priv, int *processed)
 		return true;
 	}
 
-	/*
-	 * It is impossible to get multi-segment packets
-	 * because we always provide big enough receive buffers.
-	 */
-	if (unlikely(!ftmac100_rxdes_last_segment(rxdes)))
-		BUG();
+	/* We don't support multi-segment packets for now, so drop them. */
+	ret = ftmac100_rxdes_last_segment(rxdes);
+	if (unlikely(!ret)) {
+		netdev->stats.rx_length_errors++;
+		ftmac100_rx_drop_packet(priv);
+		return true;
+	}
 
 	/* start processing */
 	skb = netdev_alloc_skb_ip_align(netdev, 128);
@@ -632,8 +658,8 @@ static void ftmac100_tx_complete(struct ftmac100 *priv)
 		;
 }
 
-static int ftmac100_xmit(struct ftmac100 *priv, struct sk_buff *skb,
-			 dma_addr_t map)
+static netdev_tx_t ftmac100_xmit(struct ftmac100 *priv, struct sk_buff *skb,
+				 dma_addr_t map)
 {
 	struct net_device *netdev = priv->netdev;
 	struct ftmac100_txdes *txdes;
@@ -732,10 +758,9 @@ static int ftmac100_alloc_buffers(struct ftmac100 *priv)
 {
 	int i;
 
-	priv->descs = dma_zalloc_coherent(priv->dev,
-					  sizeof(struct ftmac100_descs),
-					  &priv->descs_dma_addr,
-					  GFP_KERNEL);
+	priv->descs = dma_alloc_coherent(priv->dev,
+					 sizeof(struct ftmac100_descs),
+					 &priv->descs_dma_addr, GFP_KERNEL);
 	if (!priv->descs)
 		return -ENOMEM;
 
@@ -820,16 +845,18 @@ static void ftmac100_mdio_write(struct net_device *netdev, int phy_id, int reg,
 static void ftmac100_get_drvinfo(struct net_device *netdev,
 				 struct ethtool_drvinfo *info)
 {
-	strlcpy(info->driver, DRV_NAME, sizeof(info->driver));
-	strlcpy(info->version, DRV_VERSION, sizeof(info->version));
-	strlcpy(info->bus_info, dev_name(&netdev->dev), sizeof(info->bus_info));
+	strscpy(info->driver, DRV_NAME, sizeof(info->driver));
+	strscpy(info->bus_info, dev_name(&netdev->dev), sizeof(info->bus_info));
 }
 
 static int ftmac100_get_link_ksettings(struct net_device *netdev,
 				       struct ethtool_link_ksettings *cmd)
 {
 	struct ftmac100 *priv = netdev_priv(netdev);
-	return mii_ethtool_get_link_ksettings(&priv->mii, cmd);
+
+	mii_ethtool_get_link_ksettings(&priv->mii, cmd);
+
+	return 0;
 }
 
 static int ftmac100_set_link_ksettings(struct net_device *netdev,
@@ -867,11 +894,10 @@ static irqreturn_t ftmac100_interrupt(int irq, void *dev_id)
 	struct net_device *netdev = dev_id;
 	struct ftmac100 *priv = netdev_priv(netdev);
 
-	if (likely(netif_running(netdev))) {
-		/* Disable interrupts for polling */
-		ftmac100_disable_all_int(priv);
+	/* Disable interrupts for polling */
+	ftmac100_disable_all_int(priv);
+	if (likely(netif_running(netdev)))
 		napi_schedule(&priv->napi);
-	}
 
 	return IRQ_HANDLED;
 }
@@ -1011,7 +1037,8 @@ static int ftmac100_stop(struct net_device *netdev)
 	return 0;
 }
 
-static int ftmac100_hard_start_xmit(struct sk_buff *skb, struct net_device *netdev)
+static netdev_tx_t
+ftmac100_hard_start_xmit(struct sk_buff *skb, struct net_device *netdev)
 {
 	struct ftmac100 *priv = netdev_priv(netdev);
 	dma_addr_t map;
@@ -1048,13 +1075,46 @@ static int ftmac100_do_ioctl(struct net_device *netdev, struct ifreq *ifr, int c
 	return generic_mii_ioctl(&priv->mii, data, cmd, NULL);
 }
 
+static int ftmac100_change_mtu(struct net_device *netdev, int mtu)
+{
+	struct ftmac100 *priv = netdev_priv(netdev);
+	unsigned int maccr;
+
+	maccr = ioread32(priv->base + FTMAC100_OFFSET_MACCR);
+	if (mtu > ETH_DATA_LEN) {
+		/* process long packets in the driver */
+		maccr |= FTMAC100_MACCR_RX_FTL;
+	} else {
+		/* Let the controller drop incoming packets greater
+		 * than 1518 (that is 1500 + 14 Ethernet + 4 FCS).
+		 */
+		maccr &= ~FTMAC100_MACCR_RX_FTL;
+	}
+	iowrite32(maccr, priv->base + FTMAC100_OFFSET_MACCR);
+
+	WRITE_ONCE(netdev->mtu, mtu);
+
+	return 0;
+}
+
+static void ftmac100_set_rx_mode(struct net_device *netdev)
+{
+	struct ftmac100 *priv = netdev_priv(netdev);
+	unsigned int maccr = ioread32(priv->base + FTMAC100_OFFSET_MACCR);
+
+	ftmac100_set_rx_bits(priv, &maccr);
+	iowrite32(maccr, priv->base + FTMAC100_OFFSET_MACCR);
+}
+
 static const struct net_device_ops ftmac100_netdev_ops = {
 	.ndo_open		= ftmac100_open,
 	.ndo_stop		= ftmac100_stop,
 	.ndo_start_xmit		= ftmac100_hard_start_xmit,
 	.ndo_set_mac_address	= eth_mac_addr,
 	.ndo_validate_addr	= eth_validate_addr,
-	.ndo_do_ioctl		= ftmac100_do_ioctl,
+	.ndo_eth_ioctl		= ftmac100_do_ioctl,
+	.ndo_change_mtu		= ftmac100_change_mtu,
+	.ndo_set_rx_mode	= ftmac100_set_rx_mode,
 };
 
 /******************************************************************************
@@ -1068,9 +1128,6 @@ static int ftmac100_probe(struct platform_device *pdev)
 	struct ftmac100 *priv;
 	int err;
 
-	if (!pdev)
-		return -ENODEV;
-
 	res = platform_get_resource(pdev, IORESOURCE_MEM, 0);
 	if (!res)
 		return -ENXIO;
@@ -1089,6 +1146,11 @@ static int ftmac100_probe(struct platform_device *pdev)
 	SET_NETDEV_DEV(netdev, &pdev->dev);
 	netdev->ethtool_ops = &ftmac100_ethtool_ops;
 	netdev->netdev_ops = &ftmac100_netdev_ops;
+	netdev->max_mtu = MAX_PKT_SIZE - VLAN_ETH_HLEN;
+
+	err = platform_get_ethdev_address(&pdev->dev, netdev);
+	if (err == -EPROBE_DEFER)
+		goto defer_get_mac;
 
 	platform_set_drvdata(pdev, netdev);
 
@@ -1100,7 +1162,7 @@ static int ftmac100_probe(struct platform_device *pdev)
 	spin_lock_init(&priv->tx_lock);
 
 	/* initialize NAPI */
-	netif_napi_add(netdev, &priv->napi, ftmac100_poll, 64);
+	netif_napi_add(netdev, &priv->napi, ftmac100_poll);
 
 	/* map io memory */
 	priv->res = request_mem_region(res->start, resource_size(res),
@@ -1151,12 +1213,13 @@ err_ioremap:
 	release_resource(priv->res);
 err_req_mem:
 	netif_napi_del(&priv->napi);
+defer_get_mac:
 	free_netdev(netdev);
 err_alloc_etherdev:
 	return err;
 }
 
-static int ftmac100_remove(struct platform_device *pdev)
+static void ftmac100_remove(struct platform_device *pdev)
 {
 	struct net_device *netdev;
 	struct ftmac100 *priv;
@@ -1171,7 +1234,6 @@ static int ftmac100_remove(struct platform_device *pdev)
 
 	netif_napi_del(&priv->napi);
 	free_netdev(netdev);
-	return 0;
 }
 
 static const struct of_device_id ftmac100_of_ids[] = {
@@ -1191,19 +1253,7 @@ static struct platform_driver ftmac100_driver = {
 /******************************************************************************
  * initialization / finalization
  *****************************************************************************/
-static int __init ftmac100_init(void)
-{
-	pr_info("Loading version " DRV_VERSION " ...\n");
-	return platform_driver_register(&ftmac100_driver);
-}
-
-static void __exit ftmac100_exit(void)
-{
-	platform_driver_unregister(&ftmac100_driver);
-}
-
-module_init(ftmac100_init);
-module_exit(ftmac100_exit);
+module_platform_driver(ftmac100_driver);
 
 MODULE_AUTHOR("Po-Yu Chuang <ratbert@faraday-tech.com>");
 MODULE_DESCRIPTION("FTMAC100 driver");
diff --git a/drivers/net/ethernet/faraday/ftmac100.h b/drivers/net/ethernet/faraday/ftmac100.h
index 46a0c47b1ee1..8af32f9070f4 100644
--- a/drivers/net/ethernet/faraday/ftmac100.h
+++ b/drivers/net/ethernet/faraday/ftmac100.h
@@ -1,22 +1,9 @@
+/* SPDX-License-Identifier: GPL-2.0-or-later */
 /*
  * Faraday FTMAC100 10/100 Ethernet
  *
  * (C) Copyright 2009-2011 Faraday Technology
  * Po-Yu Chuang <ratbert@faraday-tech.com>
- *
- * This program is free software; you can redistribute it and/or modify
- * it under the terms of the GNU General Public License as published by
- * the Free Software Foundation; either version 2 of the License, or
- * (at your option) any later version.
- *
- * This program is distributed in the hope that it will be useful,
- * but WITHOUT ANY WARRANTY; without even the implied warranty of
- * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
- * GNU General Public License for more details.
- *
- * You should have received a copy of the GNU General Public License
- * along with this program; if not, write to the Free Software
- * Foundation, Inc., 675 Mass Ave, Cambridge, MA 02139, USA.
  */
 
 #ifndef __FTMAC100_H
@@ -135,9 +122,9 @@
  * Transmit descriptor, aligned to 16 bytes
  */
 struct ftmac100_txdes {
-	unsigned int	txdes0;
-	unsigned int	txdes1;
-	unsigned int	txdes2;	/* TXBUF_BADR */
+	__le32		txdes0;
+	__le32		txdes1;
+	__le32		txdes2;	/* TXBUF_BADR */
 	unsigned int	txdes3;	/* not used by HW */
 } __attribute__ ((aligned(16)));
 
@@ -156,9 +143,9 @@ struct ftmac100_txdes {
  * Receive descriptor, aligned to 16 bytes
  */
 struct ftmac100_rxdes {
-	unsigned int	rxdes0;
-	unsigned int	rxdes1;
-	unsigned int	rxdes2;	/* RXBUF_BADR */
+	__le32		rxdes0;
+	__le32		rxdes1;
+	__le32		rxdes2;	/* RXBUF_BADR */
 	unsigned int	rxdes3;	/* not used by HW */
 } __attribute__ ((aligned(16)));
 
diff --git a/drivers/net/ethernet/fealnx.c b/drivers/net/ethernet/fealnx.c
index 766636a7c25e..6ac8547ef9b8 100644
--- a/drivers/net/ethernet/fealnx.c
+++ b/drivers/net/ethernet/fealnx.c
@@ -25,8 +25,6 @@
 */
 
 #define DRV_NAME	"fealnx"
-#define DRV_VERSION	"2.52"
-#define DRV_RELDATE	"Sep-11-2006"
 
 static int debug;		/* 1-> print debug message */
 static int max_interrupt_work = 20;
@@ -91,11 +89,6 @@ static int full_duplex[MAX_UNITS] = { -1, -1, -1, -1, -1, -1, -1, -1 };
 #include <linux/uaccess.h>
 #include <asm/byteorder.h>
 
-/* These identify the driver base version and may not be removed. */
-static const char version[] =
-	KERN_INFO DRV_NAME ".c:v" DRV_VERSION " " DRV_RELDATE "\n";
-
-
 /* This driver was written to use PCI memory space, however some x86 systems
    work only with I/O space accesses. */
 #ifndef __alpha__
@@ -151,7 +144,7 @@ struct chip_info {
 };
 
 static const struct chip_info skel_netdrv_tbl[] = {
- 	{ "100/10M Ethernet PCI Adapter",	HAS_MII_XCVR },
+	{ "100/10M Ethernet PCI Adapter",	HAS_MII_XCVR },
 	{ "100/10M Ethernet PCI Adapter",	HAS_CHIP_XCVR },
 	{ "1000/100/10M Ethernet PCI Adapter",	HAS_MII_XCVR },
 };
@@ -257,8 +250,8 @@ enum rx_desc_status_bits {
 	RXFSD = 0x00000800,	/* first descriptor */
 	RXLSD = 0x00000400,	/* last descriptor */
 	ErrorSummary = 0x80,	/* error summary */
-	RUNT = 0x40,		/* runt packet received */
-	LONG = 0x20,		/* long packet received */
+	RUNTPKT = 0x40,		/* runt packet received */
+	LONGPKT = 0x20,		/* long packet received */
 	FAE = 0x10,		/* frame align error */
 	CRC = 0x08,		/* crc error */
 	RXER = 0x04,		/* receive error */
@@ -426,9 +419,9 @@ static void mdio_write(struct net_device *dev, int phy_id, int location, int val
 static int netdev_open(struct net_device *dev);
 static void getlinktype(struct net_device *dev);
 static void getlinkstatus(struct net_device *dev);
-static void netdev_timer(unsigned long data);
-static void reset_timer(unsigned long data);
-static void fealnx_tx_timeout(struct net_device *dev);
+static void netdev_timer(struct timer_list *t);
+static void reset_timer(struct timer_list *t);
+static void fealnx_tx_timeout(struct net_device *dev, unsigned int txqueue);
 static void init_ring(struct net_device *dev);
 static netdev_tx_t start_tx(struct sk_buff *skb, struct net_device *dev);
 static irqreturn_t intr_handler(int irq, void *dev_instance);
@@ -470,7 +463,7 @@ static const struct net_device_ops netdev_ops = {
 	.ndo_start_xmit		= start_tx,
 	.ndo_get_stats 		= get_stats,
 	.ndo_set_rx_mode	= set_rx_mode,
-	.ndo_do_ioctl		= mii_ioctl,
+	.ndo_eth_ioctl		= mii_ioctl,
 	.ndo_tx_timeout		= fealnx_tx_timeout,
 	.ndo_set_mac_address 	= eth_mac_addr,
 	.ndo_validate_addr	= eth_validate_addr,
@@ -489,19 +482,13 @@ static int fealnx_init_one(struct pci_dev *pdev,
 	struct net_device *dev;
 	void *ring_space;
 	dma_addr_t ring_dma;
+	u8 addr[ETH_ALEN];
 #ifdef USE_IO_OPS
 	int bar = 0;
 #else
 	int bar = 1;
 #endif
 
-/* when built into the kernel, we only print version if device is found */
-#ifndef MODULE
-	static int printed_version;
-	if (!printed_version++)
-		printk(version);
-#endif
-
 	card_idx++;
 	sprintf(boardname, "fealnx%d", card_idx);
 
@@ -539,7 +526,8 @@ static int fealnx_init_one(struct pci_dev *pdev,
 
 	/* read ethernet id */
 	for (i = 0; i < 6; ++i)
-		dev->dev_addr[i] = ioread8(ioaddr + PAR0 + i);
+		addr[i] = ioread8(ioaddr + PAR0 + i);
+	eth_hw_addr_set(dev, addr);
 
 	/* Reset the chip to erase previous misconfiguration. */
 	iowrite32(0x00000001, ioaddr + BCR);
@@ -557,7 +545,8 @@ static int fealnx_init_one(struct pci_dev *pdev,
 	np->mii.phy_id_mask = 0x1f;
 	np->mii.reg_num_mask = 0x1f;
 
-	ring_space = pci_alloc_consistent(pdev, RX_TOTAL_SIZE, &ring_dma);
+	ring_space = dma_alloc_coherent(&pdev->dev, RX_TOTAL_SIZE, &ring_dma,
+					GFP_KERNEL);
 	if (!ring_space) {
 		err = -ENOMEM;
 		goto err_out_free_dev;
@@ -565,7 +554,8 @@ static int fealnx_init_one(struct pci_dev *pdev,
 	np->rx_ring = ring_space;
 	np->rx_ring_dma = ring_dma;
 
-	ring_space = pci_alloc_consistent(pdev, TX_TOTAL_SIZE, &ring_dma);
+	ring_space = dma_alloc_coherent(&pdev->dev, TX_TOTAL_SIZE, &ring_dma,
+					GFP_KERNEL);
 	if (!ring_space) {
 		err = -ENOMEM;
 		goto err_out_free_rx;
@@ -670,9 +660,11 @@ static int fealnx_init_one(struct pci_dev *pdev,
 	return 0;
 
 err_out_free_tx:
-	pci_free_consistent(pdev, TX_TOTAL_SIZE, np->tx_ring, np->tx_ring_dma);
+	dma_free_coherent(&pdev->dev, TX_TOTAL_SIZE, np->tx_ring,
+			  np->tx_ring_dma);
 err_out_free_rx:
-	pci_free_consistent(pdev, RX_TOTAL_SIZE, np->rx_ring, np->rx_ring_dma);
+	dma_free_coherent(&pdev->dev, RX_TOTAL_SIZE, np->rx_ring,
+			  np->rx_ring_dma);
 err_out_free_dev:
 	free_netdev(dev);
 err_out_unmap:
@@ -690,10 +682,10 @@ static void fealnx_remove_one(struct pci_dev *pdev)
 	if (dev) {
 		struct netdev_private *np = netdev_priv(dev);
 
-		pci_free_consistent(pdev, TX_TOTAL_SIZE, np->tx_ring,
-			np->tx_ring_dma);
-		pci_free_consistent(pdev, RX_TOTAL_SIZE, np->rx_ring,
-			np->rx_ring_dma);
+		dma_free_coherent(&pdev->dev, TX_TOTAL_SIZE, np->tx_ring,
+				  np->tx_ring_dma);
+		dma_free_coherent(&pdev->dev, RX_TOTAL_SIZE, np->rx_ring,
+				  np->rx_ring_dma);
 		unregister_netdev(dev);
 		pci_iounmap(pdev, np->mem);
 		free_netdev(dev);
@@ -837,7 +829,7 @@ static int netdev_open(struct net_device *dev)
 		return -EAGAIN;
 
 	for (i = 0; i < 3; i++)
-		iowrite16(((unsigned short*)dev->dev_addr)[i],
+		iowrite16(((const unsigned short *)dev->dev_addr)[i],
 				ioaddr + PAR0 + i*2);
 
 	init_ring(dev);
@@ -867,7 +859,7 @@ static int netdev_open(struct net_device *dev)
 	np->bcrvalue |= 0x04;	/* big-endian */
 #endif
 
-#if defined(__i386__) && !defined(MODULE)
+#if defined(__i386__) && !defined(MODULE) && !defined(CONFIG_UML)
 	if (boot_cpu_data.x86 <= 4)
 		np->crvalue = 0xa00;
 	else
@@ -909,17 +901,13 @@ static int netdev_open(struct net_device *dev)
 		printk(KERN_DEBUG "%s: Done netdev_open().\n", dev->name);
 
 	/* Set the timer to check for link beat. */
-	init_timer(&np->timer);
+	timer_setup(&np->timer, netdev_timer, 0);
 	np->timer.expires = RUN_AT(3 * HZ);
-	np->timer.data = (unsigned long) dev;
-	np->timer.function = netdev_timer;
 
 	/* timer handler */
 	add_timer(&np->timer);
 
-	init_timer(&np->reset_timer);
-	np->reset_timer.data = (unsigned long) dev;
-	np->reset_timer.function = reset_timer;
+	timer_setup(&np->reset_timer, reset_timer, 0);
 	np->reset_timer_armed = 0;
 	return rc;
 }
@@ -1074,18 +1062,20 @@ static void allocate_rx_buffers(struct net_device *dev)
 			np->lack_rxbuf = np->lack_rxbuf->next_desc_logical;
 
 		np->lack_rxbuf->skbuff = skb;
-		np->lack_rxbuf->buffer = pci_map_single(np->pci_dev, skb->data,
-			np->rx_buf_sz, PCI_DMA_FROMDEVICE);
+		np->lack_rxbuf->buffer = dma_map_single(&np->pci_dev->dev,
+							skb->data,
+							np->rx_buf_sz,
+							DMA_FROM_DEVICE);
 		np->lack_rxbuf->status = RXOWN;
 		++np->really_rx_count;
 	}
 }
 
 
-static void netdev_timer(unsigned long data)
+static void netdev_timer(struct timer_list *t)
 {
-	struct net_device *dev = (struct net_device *) data;
-	struct netdev_private *np = netdev_priv(dev);
+	struct netdev_private *np = timer_container_of(np, t, timer);
+	struct net_device *dev = np->mii.dev;
 	void __iomem *ioaddr = np->mem;
 	int old_crvalue = np->crvalue;
 	unsigned int old_linkok = np->linkok;
@@ -1171,10 +1161,10 @@ static void enable_rxtx(struct net_device *dev)
 }
 
 
-static void reset_timer(unsigned long data)
+static void reset_timer(struct timer_list *t)
 {
-	struct net_device *dev = (struct net_device *) data;
-	struct netdev_private *np = netdev_priv(dev);
+	struct netdev_private *np = timer_container_of(np, t, reset_timer);
+	struct net_device *dev = np->mii.dev;
 	unsigned long flags;
 
 	printk(KERN_WARNING "%s: resetting tx and rx machinery\n", dev->name);
@@ -1195,7 +1185,7 @@ static void reset_timer(unsigned long data)
 }
 
 
-static void fealnx_tx_timeout(struct net_device *dev)
+static void fealnx_tx_timeout(struct net_device *dev, unsigned int txqueue)
 {
 	struct netdev_private *np = netdev_priv(dev);
 	void __iomem *ioaddr = np->mem;
@@ -1269,8 +1259,10 @@ static void init_ring(struct net_device *dev)
 
 		++np->really_rx_count;
 		np->rx_ring[i].skbuff = skb;
-		np->rx_ring[i].buffer = pci_map_single(np->pci_dev, skb->data,
-			np->rx_buf_sz, PCI_DMA_FROMDEVICE);
+		np->rx_ring[i].buffer = dma_map_single(&np->pci_dev->dev,
+						       skb->data,
+						       np->rx_buf_sz,
+						       DMA_FROM_DEVICE);
 		np->rx_ring[i].status = RXOWN;
 		np->rx_ring[i].control |= RXIC;
 	}
@@ -1308,8 +1300,8 @@ static netdev_tx_t start_tx(struct sk_buff *skb, struct net_device *dev)
 #define one_buffer
 #define BPT 1022
 #if defined(one_buffer)
-	np->cur_tx_copy->buffer = pci_map_single(np->pci_dev, skb->data,
-		skb->len, PCI_DMA_TODEVICE);
+	np->cur_tx_copy->buffer = dma_map_single(&np->pci_dev->dev, skb->data,
+						 skb->len, DMA_TO_DEVICE);
 	np->cur_tx_copy->control = TXIC | TXLD | TXFD | CRCEnable | PADEnable;
 	np->cur_tx_copy->control |= (skb->len << PKTSShift);	/* pkt size */
 	np->cur_tx_copy->control |= (skb->len << TBSShift);	/* buffer size */
@@ -1324,8 +1316,9 @@ static netdev_tx_t start_tx(struct sk_buff *skb, struct net_device *dev)
 		struct fealnx_desc *next;
 
 		/* for the first descriptor */
-		np->cur_tx_copy->buffer = pci_map_single(np->pci_dev, skb->data,
-			BPT, PCI_DMA_TODEVICE);
+		np->cur_tx_copy->buffer = dma_map_single(&np->pci_dev->dev,
+							 skb->data, BPT,
+							 DMA_TO_DEVICE);
 		np->cur_tx_copy->control = TXIC | TXFD | CRCEnable | PADEnable;
 		np->cur_tx_copy->control |= (skb->len << PKTSShift);	/* pkt size */
 		np->cur_tx_copy->control |= (BPT << TBSShift);	/* buffer size */
@@ -1339,8 +1332,9 @@ static netdev_tx_t start_tx(struct sk_buff *skb, struct net_device *dev)
 // 89/12/29 add,
 		if (np->pci_dev->device == 0x891)
 			np->cur_tx_copy->control |= ETIControl | RetryTxLC;
-		next->buffer = pci_map_single(ep->pci_dev, skb->data + BPT,
-                                skb->len - BPT, PCI_DMA_TODEVICE);
+		next->buffer = dma_map_single(&ep->pci_dev->dev,
+					      skb->data + BPT, skb->len - BPT,
+					      DMA_TO_DEVICE);
 
 		next->status = TXOWN;
 		np->cur_tx_copy->status = TXOWN;
@@ -1348,8 +1342,9 @@ static netdev_tx_t start_tx(struct sk_buff *skb, struct net_device *dev)
 		np->cur_tx_copy = next->next_desc_logical;
 		np->free_tx_count -= 2;
 	} else {
-		np->cur_tx_copy->buffer = pci_map_single(np->pci_dev, skb->data,
-			skb->len, PCI_DMA_TODEVICE);
+		np->cur_tx_copy->buffer = dma_map_single(&np->pci_dev->dev,
+							 skb->data, skb->len,
+							 DMA_TO_DEVICE);
 		np->cur_tx_copy->control = TXIC | TXLD | TXFD | CRCEnable | PADEnable;
 		np->cur_tx_copy->control |= (skb->len << PKTSShift);	/* pkt size */
 		np->cur_tx_copy->control |= (skb->len << TBSShift);	/* buffer size */
@@ -1389,8 +1384,8 @@ static void reset_tx_descriptors(struct net_device *dev)
 	for (i = 0; i < TX_RING_SIZE; i++) {
 		cur = &np->tx_ring[i];
 		if (cur->skbuff) {
-			pci_unmap_single(np->pci_dev, cur->buffer,
-				cur->skbuff->len, PCI_DMA_TODEVICE);
+			dma_unmap_single(&np->pci_dev->dev, cur->buffer,
+					 cur->skbuff->len, DMA_TO_DEVICE);
 			dev_kfree_skb_any(cur->skbuff);
 			cur->skbuff = NULL;
 		}
@@ -1533,9 +1528,11 @@ static irqreturn_t intr_handler(int irq, void *dev_instance)
 			}
 
 			/* Free the original skb. */
-			pci_unmap_single(np->pci_dev, np->cur_tx->buffer,
-				np->cur_tx->skbuff->len, PCI_DMA_TODEVICE);
-			dev_kfree_skb_irq(np->cur_tx->skbuff);
+			dma_unmap_single(&np->pci_dev->dev,
+					 np->cur_tx->buffer,
+					 np->cur_tx->skbuff->len,
+					 DMA_TO_DEVICE);
+			dev_consume_skb_irq(np->cur_tx->skbuff);
 			np->cur_tx->skbuff = NULL;
 			--np->really_tx_count;
 			if (np->cur_tx->control & TXLD) {
@@ -1632,7 +1629,7 @@ static int netdev_rx(struct net_device *dev)
 					       dev->name, rx_status);
 
 				dev->stats.rx_errors++;	/* end of a packet. */
-				if (rx_status & (LONG | RUNT))
+				if (rx_status & (LONGPKT | RUNTPKT))
 					dev->stats.rx_length_errors++;
 				if (rx_status & RXER)
 					dev->stats.rx_frame_errors++;
@@ -1700,10 +1697,10 @@ static int netdev_rx(struct net_device *dev)
 			if (pkt_len < rx_copybreak &&
 			    (skb = netdev_alloc_skb(dev, pkt_len + 2)) != NULL) {
 				skb_reserve(skb, 2);	/* 16 byte align the IP header */
-				pci_dma_sync_single_for_cpu(np->pci_dev,
-							    np->cur_rx->buffer,
-							    np->rx_buf_sz,
-							    PCI_DMA_FROMDEVICE);
+				dma_sync_single_for_cpu(&np->pci_dev->dev,
+							np->cur_rx->buffer,
+							np->rx_buf_sz,
+							DMA_FROM_DEVICE);
 				/* Call copy + cksum if available. */
 
 #if ! defined(__alpha__)
@@ -1711,18 +1708,18 @@ static int netdev_rx(struct net_device *dev)
 					np->cur_rx->skbuff->data, pkt_len);
 				skb_put(skb, pkt_len);
 #else
-				memcpy(skb_put(skb, pkt_len),
-					np->cur_rx->skbuff->data, pkt_len);
+				skb_put_data(skb, np->cur_rx->skbuff->data,
+					     pkt_len);
 #endif
-				pci_dma_sync_single_for_device(np->pci_dev,
-							       np->cur_rx->buffer,
-							       np->rx_buf_sz,
-							       PCI_DMA_FROMDEVICE);
+				dma_sync_single_for_device(&np->pci_dev->dev,
+							   np->cur_rx->buffer,
+							   np->rx_buf_sz,
+							   DMA_FROM_DEVICE);
 			} else {
-				pci_unmap_single(np->pci_dev,
+				dma_unmap_single(&np->pci_dev->dev,
 						 np->cur_rx->buffer,
 						 np->rx_buf_sz,
-						 PCI_DMA_FROMDEVICE);
+						 DMA_FROM_DEVICE);
 				skb_put(skb = np->cur_rx->skbuff, pkt_len);
 				np->cur_rx->skbuff = NULL;
 				--np->really_rx_count;
@@ -1812,22 +1809,20 @@ static void netdev_get_drvinfo(struct net_device *dev, struct ethtool_drvinfo *i
 {
 	struct netdev_private *np = netdev_priv(dev);
 
-	strlcpy(info->driver, DRV_NAME, sizeof(info->driver));
-	strlcpy(info->version, DRV_VERSION, sizeof(info->version));
-	strlcpy(info->bus_info, pci_name(np->pci_dev), sizeof(info->bus_info));
+	strscpy(info->driver, DRV_NAME, sizeof(info->driver));
+	strscpy(info->bus_info, pci_name(np->pci_dev), sizeof(info->bus_info));
 }
 
 static int netdev_get_link_ksettings(struct net_device *dev,
 				     struct ethtool_link_ksettings *cmd)
 {
 	struct netdev_private *np = netdev_priv(dev);
-	int rc;
 
 	spin_lock_irq(&np->lock);
-	rc = mii_ethtool_get_link_ksettings(&np->mii, cmd);
+	mii_ethtool_get_link_ksettings(&np->mii, cmd);
 	spin_unlock_irq(&np->lock);
 
-	return rc;
+	return 0;
 }
 
 static int netdev_set_link_ksettings(struct net_device *dev,
@@ -1905,8 +1900,8 @@ static int netdev_close(struct net_device *dev)
 	/* Stop the chip's Tx and Rx processes. */
 	stop_nic_rxtx(ioaddr, 0);
 
-	del_timer_sync(&np->timer);
-	del_timer_sync(&np->reset_timer);
+	timer_delete_sync(&np->timer);
+	timer_delete_sync(&np->reset_timer);
 
 	free_irq(np->pci_dev->irq, dev);
 
@@ -1916,8 +1911,9 @@ static int netdev_close(struct net_device *dev)
 
 		np->rx_ring[i].status = 0;
 		if (skb) {
-			pci_unmap_single(np->pci_dev, np->rx_ring[i].buffer,
-				np->rx_buf_sz, PCI_DMA_FROMDEVICE);
+			dma_unmap_single(&np->pci_dev->dev,
+					 np->rx_ring[i].buffer, np->rx_buf_sz,
+					 DMA_FROM_DEVICE);
 			dev_kfree_skb(skb);
 			np->rx_ring[i].skbuff = NULL;
 		}
@@ -1927,8 +1923,9 @@ static int netdev_close(struct net_device *dev)
 		struct sk_buff *skb = np->tx_ring[i].skbuff;
 
 		if (skb) {
-			pci_unmap_single(np->pci_dev, np->tx_ring[i].buffer,
-				skb->len, PCI_DMA_TODEVICE);
+			dma_unmap_single(&np->pci_dev->dev,
+					 np->tx_ring[i].buffer, skb->len,
+					 DMA_TO_DEVICE);
 			dev_kfree_skb(skb);
 			np->tx_ring[i].skbuff = NULL;
 		}
@@ -1953,20 +1950,4 @@ static struct pci_driver fealnx_driver = {
 	.remove		= fealnx_remove_one,
 };
 
-static int __init fealnx_init(void)
-{
-/* when a module, this is printed whether or not devices are found in probe */
-#ifdef MODULE
-	printk(version);
-#endif
-
-	return pci_register_driver(&fealnx_driver);
-}
-
-static void __exit fealnx_exit(void)
-{
-	pci_unregister_driver(&fealnx_driver);
-}
-
-module_init(fealnx_init);
-module_exit(fealnx_exit);
+module_pci_driver(fealnx_driver);
diff --git a/drivers/net/ethernet/freescale/Kconfig b/drivers/net/ethernet/freescale/Kconfig
index 6e490fd2345d..bbef47c3480c 100644
--- a/drivers/net/ethernet/freescale/Kconfig
+++ b/drivers/net/ethernet/freescale/Kconfig
@@ -1,3 +1,4 @@
+# SPDX-License-Identifier: GPL-2.0-only
 #
 # Freescale device configuration
 #
@@ -8,8 +9,8 @@ config NET_VENDOR_FREESCALE
 	depends on FSL_SOC || QUICC_ENGINE || CPM1 || CPM2 || PPC_MPC512x || \
 		   M523x || M527x || M5272 || M528x || M520x || M532x || \
 		   ARCH_MXC || ARCH_MXS || (PPC_MPC52xx && PPC_BESTCOMM) || \
-		   ARCH_LAYERSCAPE || COMPILE_TEST
-	---help---
+		   ARCH_LAYERSCAPE || ARCH_S32 || COMPILE_TEST
+	help
 	  If you have a network (Ethernet) card belonging to this class, say Y.
 
 	  Note that the answer to this question doesn't directly affect the
@@ -22,13 +23,17 @@ if NET_VENDOR_FREESCALE
 config FEC
 	tristate "FEC ethernet controller (of ColdFire and some i.MX CPUs)"
 	depends on (M523x || M527x || M5272 || M528x || M520x || M532x || \
-		   ARCH_MXC || SOC_IMX28)
+		   ARCH_MXC || ARCH_S32 || SOC_IMX28 || COMPILE_TEST)
 	default ARCH_MXC || SOC_IMX28 if ARM
+	depends on PTP_1588_CLOCK_OPTIONAL
+	select CRC32
 	select PHYLIB
-	imply PTP_1588_CLOCK
-	---help---
+	select PAGE_POOL
+	imply PAGE_POOL_STATS
+	imply NET_SELFTESTS
+	help
 	  Say Y here if you want to use the built-in 10/100 Fast ethernet
-	  controller on some Motorola ColdFire and Freescale i.MX processors.
+	  controller on some Motorola ColdFire and Freescale i.MX/S32 processors.
 
 config FEC_MPC52xx
 	tristate "FEC MPC52xx driver"
@@ -36,7 +41,7 @@ config FEC_MPC52xx
 	select CRC32
 	select PHYLIB
 	select PPC_BESTCOMM_FEC
-	---help---
+	help
 	  This option enables support for the MPC5200's on-chip
 	  Fast Ethernet Controller
 	  If compiled as module, it will be called fec_mpc52xx.
@@ -45,7 +50,7 @@ config FEC_MPC52xx_MDIO
 	bool "FEC MPC52xx MDIO bus driver"
 	depends on FEC_MPC52xx
 	default y
-	---help---
+	help
 	  The MPC5200's FEC can connect to the Ethernet either with
 	  an external MII PHY chip or 10 Mbps 7-wire interface
 	  (Motorola? industry standard).
@@ -59,7 +64,7 @@ source "drivers/net/ethernet/freescale/fman/Kconfig"
 config FSL_PQ_MDIO
 	tristate "Freescale PQ MDIO"
 	select PHYLIB
-	---help---
+	help
 	  This driver supports the MDIO bus used by the gianfar and UCC drivers.
 
 config FSL_XGMAC_MDIO
@@ -67,16 +72,16 @@ config FSL_XGMAC_MDIO
 	select PHYLIB
 	depends on OF
 	select OF_MDIO
-	---help---
+	help
 	  This driver supports the MDIO bus on the Fman 10G Ethernet MACs, and
 	  on the FMan mEMAC (which supports both Clauses 22 and 45)
 
 config UCC_GETH
 	tristate "Freescale QE Gigabit Ethernet"
-	depends on QUICC_ENGINE
+	depends on QUICC_ENGINE && PPC32
 	select FSL_PQ_MDIO
-	select PHYLIB
-	---help---
+	select PHYLINK
+	help
 	  This driver supports the Gigabit Ethernet mode of the QUICC Engine,
 	  which is available on some Freescale SOCs.
 
@@ -89,12 +94,15 @@ config GIANFAR
 	depends on HAS_DMA
 	select FSL_PQ_MDIO
 	select PHYLIB
+	select FIXED_PHY
 	select CRC32
-	---help---
+	help
 	  This driver supports the Gigabit TSEC on the MPC83xx, MPC85xx,
 	  and MPC86xx family of chips, the eTSEC on LS1021A and the FEC
 	  on the 8540.
 
 source "drivers/net/ethernet/freescale/dpaa/Kconfig"
+source "drivers/net/ethernet/freescale/dpaa2/Kconfig"
+source "drivers/net/ethernet/freescale/enetc/Kconfig"
 
 endif # NET_VENDOR_FREESCALE
diff --git a/drivers/net/ethernet/freescale/Makefile b/drivers/net/ethernet/freescale/Makefile
index c46df5c82af5..de7b31842233 100644
--- a/drivers/net/ethernet/freescale/Makefile
+++ b/drivers/net/ethernet/freescale/Makefile
@@ -1,3 +1,4 @@
+# SPDX-License-Identifier: GPL-2.0
 #
 # Makefile for the Freescale network device drivers.
 #
@@ -13,7 +14,6 @@ obj-$(CONFIG_FS_ENET) += fs_enet/
 obj-$(CONFIG_FSL_PQ_MDIO) += fsl_pq_mdio.o
 obj-$(CONFIG_FSL_XGMAC_MDIO) += xgmac_mdio.o
 obj-$(CONFIG_GIANFAR) += gianfar_driver.o
-obj-$(CONFIG_PTP_1588_CLOCK_GIANFAR) += gianfar_ptp.o
 gianfar_driver-objs := gianfar.o \
 		gianfar_ethtool.o
 obj-$(CONFIG_UCC_GETH) += ucc_geth_driver.o
@@ -21,3 +21,7 @@ ucc_geth_driver-objs := ucc_geth.o ucc_geth_ethtool.o
 
 obj-$(CONFIG_FSL_FMAN) += fman/
 obj-$(CONFIG_FSL_DPAA_ETH) += dpaa/
+
+obj-$(CONFIG_FSL_DPAA2_ETH) += dpaa2/
+
+obj-y += enetc/
diff --git a/drivers/net/ethernet/freescale/dpaa/Kconfig b/drivers/net/ethernet/freescale/dpaa/Kconfig
index a654736237a9..2b560661c82a 100644
--- a/drivers/net/ethernet/freescale/dpaa/Kconfig
+++ b/drivers/net/ethernet/freescale/dpaa/Kconfig
@@ -1,9 +1,10 @@
+# SPDX-License-Identifier: GPL-2.0-only
 menuconfig FSL_DPAA_ETH
 	tristate "DPAA Ethernet"
 	depends on FSL_DPAA && FSL_FMAN
-	select PHYLIB
-	select FSL_FMAN_MAC
-	---help---
+	select PHYLINK
+	select PCS_LYNX
+	help
 	  Data Path Acceleration Architecture Ethernet driver,
 	  supporting the Freescale QorIQ chips.
 	  Depends on Freescale Buffer Manager and Queue Manager
diff --git a/drivers/net/ethernet/freescale/dpaa/Makefile b/drivers/net/ethernet/freescale/dpaa/Makefile
index 7db50bccb137..4f23e79232fa 100644
--- a/drivers/net/ethernet/freescale/dpaa/Makefile
+++ b/drivers/net/ethernet/freescale/dpaa/Makefile
@@ -1,3 +1,4 @@
+# SPDX-License-Identifier: GPL-2.0
 #
 # Makefile for the Freescale DPAA Ethernet controllers
 #
diff --git a/drivers/net/ethernet/freescale/dpaa/dpaa_eth.c b/drivers/net/ethernet/freescale/dpaa/dpaa_eth.c
index 9a520e4f0df9..3edc8d142dd5 100644
--- a/drivers/net/ethernet/freescale/dpaa/dpaa_eth.c
+++ b/drivers/net/ethernet/freescale/dpaa/dpaa_eth.c
@@ -1,38 +1,14 @@
-/* Copyright 2008 - 2016 Freescale Semiconductor Inc.
- *
- * Redistribution and use in source and binary forms, with or without
- * modification, are permitted provided that the following conditions are met:
- *     * Redistributions of source code must retain the above copyright
- *	 notice, this list of conditions and the following disclaimer.
- *     * Redistributions in binary form must reproduce the above copyright
- *	 notice, this list of conditions and the following disclaimer in the
- *	 documentation and/or other materials provided with the distribution.
- *     * Neither the name of Freescale Semiconductor nor the
- *	 names of its contributors may be used to endorse or promote products
- *	 derived from this software without specific prior written permission.
- *
- * ALTERNATIVELY, this software may be distributed under the terms of the
- * GNU General Public License ("GPL") as published by the Free Software
- * Foundation, either version 2 of that License or (at your option) any
- * later version.
- *
- * THIS SOFTWARE IS PROVIDED BY Freescale Semiconductor ``AS IS'' AND ANY
- * EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED
- * WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE
- * DISCLAIMED. IN NO EVENT SHALL Freescale Semiconductor BE LIABLE FOR ANY
- * DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES
- * (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES;
- * LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND
- * ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
- * (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS
- * SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
+// SPDX-License-Identifier: BSD-3-Clause OR GPL-2.0-or-later
+/*
+ * Copyright 2008 - 2016 Freescale Semiconductor Inc.
+ * Copyright 2020 NXP
  */
 
 #define pr_fmt(fmt) KBUILD_MODNAME ": " fmt
 
 #include <linux/init.h>
+#include <linux/mod_devicetable.h>
 #include <linux/module.h>
-#include <linux/of_platform.h>
 #include <linux/of_mdio.h>
 #include <linux/of_net.h>
 #include <linux/io.h>
@@ -41,6 +17,7 @@
 #include <linux/icmp.h>
 #include <linux/ip.h>
 #include <linux/ipv6.h>
+#include <linux/platform_device.h>
 #include <linux/udp.h>
 #include <linux/tcp.h>
 #include <linux/net.h>
@@ -51,9 +28,10 @@
 #include <linux/percpu.h>
 #include <linux/dma-mapping.h>
 #include <linux/sort.h>
+#include <linux/bpf.h>
+#include <linux/bpf_trace.h>
 #include <soc/fsl/bman.h>
 #include <soc/fsl/qman.h>
-
 #include "fman.h"
 #include "fman_port.h"
 #include "mac.h"
@@ -86,7 +64,7 @@ MODULE_PARM_DESC(tx_timeout, "The Tx timeout in ms");
 
 #define DPAA_MSG_DEFAULT (NETIF_MSG_DRV | NETIF_MSG_PROBE | \
 			  NETIF_MSG_LINK | NETIF_MSG_IFUP | \
-			  NETIF_MSG_IFDOWN)
+			  NETIF_MSG_IFDOWN | NETIF_MSG_HW)
 
 #define DPAA_INGRESS_CS_THRESHOLD 0x10000000
 /* Ingress congestion threshold on FMan ports
@@ -123,7 +101,25 @@ MODULE_PARM_DESC(tx_timeout, "The Tx timeout in ms");
 #define FSL_QMAN_MAX_OAL	127
 
 /* Default alignment for start of data in an Rx FD */
+#ifdef CONFIG_DPAA_ERRATUM_A050385
+/* aligning data start to 64 avoids DMA transaction splits, unless the buffer
+ * is crossing a 4k page boundary
+ */
+#define DPAA_FD_DATA_ALIGNMENT  (fman_has_errata_a050385() ? 64 : 16)
+/* aligning to 256 avoids DMA transaction splits caused by 4k page boundary
+ * crossings; also, all SG fragments except the last must have a size multiple
+ * of 256 to avoid DMA transaction splits
+ */
+#define DPAA_A050385_ALIGN 256
+#define DPAA_FD_RX_DATA_ALIGNMENT (fman_has_errata_a050385() ? \
+				   DPAA_A050385_ALIGN : 16)
+#else
 #define DPAA_FD_DATA_ALIGNMENT  16
+#define DPAA_FD_RX_DATA_ALIGNMENT DPAA_FD_DATA_ALIGNMENT
+#endif
+
+/* The DPAA requires 256 bytes reserved and mapped for the SGT */
+#define DPAA_SGT_SIZE 256
 
 /* Values for the L3R field of the FM Parse Results
  */
@@ -155,10 +151,20 @@ MODULE_PARM_DESC(tx_timeout, "The Tx timeout in ms");
 #define DPAA_PARSE_RESULTS_SIZE sizeof(struct fman_prs_result)
 #define DPAA_TIME_STAMP_SIZE 8
 #define DPAA_HASH_RESULTS_SIZE 8
-#define DPAA_RX_PRIV_DATA_SIZE	(u16)(DPAA_TX_PRIV_DATA_SIZE + \
-					dpaa_rx_extra_headroom)
-
-#define DPAA_ETH_RX_QUEUES	128
+#define DPAA_HWA_SIZE (DPAA_PARSE_RESULTS_SIZE + DPAA_TIME_STAMP_SIZE \
+		       + DPAA_HASH_RESULTS_SIZE)
+#define DPAA_RX_PRIV_DATA_DEFAULT_SIZE (DPAA_TX_PRIV_DATA_SIZE + \
+					XDP_PACKET_HEADROOM - DPAA_HWA_SIZE)
+#ifdef CONFIG_DPAA_ERRATUM_A050385
+#define DPAA_RX_PRIV_DATA_A050385_SIZE (DPAA_A050385_ALIGN - DPAA_HWA_SIZE)
+#define DPAA_RX_PRIV_DATA_SIZE (fman_has_errata_a050385() ? \
+				DPAA_RX_PRIV_DATA_A050385_SIZE : \
+				DPAA_RX_PRIV_DATA_DEFAULT_SIZE)
+#else
+#define DPAA_RX_PRIV_DATA_SIZE DPAA_RX_PRIV_DATA_DEFAULT_SIZE
+#endif
+
+#define DPAA_ETH_PCD_RXQ_NUM	128
 
 #define DPAA_ENQUEUE_RETRIES	100000
 
@@ -169,36 +175,20 @@ struct fm_port_fqs {
 	struct dpaa_fq *tx_errq;
 	struct dpaa_fq *rx_defq;
 	struct dpaa_fq *rx_errq;
+	struct dpaa_fq *rx_pcdq;
 };
 
 /* All the dpa bps in use at any moment */
 static struct dpaa_bp *dpaa_bp_array[BM_MAX_NUM_OF_POOLS];
 
-/* The raw buffer size must be cacheline aligned */
 #define DPAA_BP_RAW_SIZE 4096
-/* When using more than one buffer pool, the raw sizes are as follows:
- * 1 bp: 4KB
- * 2 bp: 2KB, 4KB
- * 3 bp: 1KB, 2KB, 4KB
- * 4 bp: 1KB, 2KB, 4KB, 8KB
- */
-static inline size_t bpool_buffer_raw_size(u8 index, u8 cnt)
-{
-	size_t res = DPAA_BP_RAW_SIZE / 4;
-	u8 i;
 
-	for (i = (cnt < 3) ? cnt : 3; i < 3 + index; i++)
-		res *= 2;
-	return res;
-}
-
-/* FMan-DMA requires 16-byte alignment for Rx buffers, but SKB_DATA_ALIGN is
- * even stronger (SMP_CACHE_BYTES-aligned), so we just get away with that,
- * via SKB_WITH_OVERHEAD(). We can't rely on netdev_alloc_frag() giving us
- * half-page-aligned buffers, so we reserve some more space for start-of-buffer
- * alignment.
- */
-#define dpaa_bp_size(raw_size) SKB_WITH_OVERHEAD((raw_size) - SMP_CACHE_BYTES)
+#ifdef CONFIG_DPAA_ERRATUM_A050385
+#define dpaa_bp_size(raw_size) (SKB_WITH_OVERHEAD(raw_size) & \
+				~(DPAA_A050385_ALIGN - 1))
+#else
+#define dpaa_bp_size(raw_size) SKB_WITH_OVERHEAD(raw_size)
+#endif
 
 static int dpaa_max_frm;
 
@@ -207,12 +197,15 @@ static int dpaa_rx_extra_headroom;
 #define dpaa_get_max_mtu()	\
 	(dpaa_max_frm - (VLAN_ETH_HLEN + ETH_FCS_LEN))
 
+static void dpaa_eth_cgr_set_speed(struct mac_device *mac_dev, int speed);
+
 static int dpaa_netdev_init(struct net_device *net_dev,
 			    const struct net_device_ops *dpaa_ops,
 			    u16 tx_timeout)
 {
 	struct dpaa_priv *priv = netdev_priv(net_dev);
 	struct device *dev = net_dev->dev.parent;
+	struct mac_device *mac_dev = priv->mac_dev;
 	struct dpaa_percpu_priv *percpu_priv;
 	const u8 *mac_addr;
 	int i, err;
@@ -226,16 +219,16 @@ static int dpaa_netdev_init(struct net_device *net_dev,
 	}
 
 	net_dev->netdev_ops = dpaa_ops;
-	mac_addr = priv->mac_dev->addr;
+	mac_addr = mac_dev->addr;
 
-	net_dev->mem_start = priv->mac_dev->res->start;
-	net_dev->mem_end = priv->mac_dev->res->end;
+	net_dev->mem_start = (unsigned long)priv->mac_dev->res->start;
+	net_dev->mem_end = (unsigned long)priv->mac_dev->res->end;
 
 	net_dev->min_mtu = ETH_MIN_MTU;
 	net_dev->max_mtu = dpaa_get_max_mtu();
 
 	net_dev->hw_features |= (NETIF_F_IP_CSUM | NETIF_F_IPV6_CSUM |
-				 NETIF_F_LLTX);
+				 NETIF_F_RXHASH);
 
 	net_dev->hw_features |= NETIF_F_SG | NETIF_F_HIGHDMA;
 	/* The kernels enables GSO automatically, if we declare NETIF_F_SG.
@@ -245,26 +238,58 @@ static int dpaa_netdev_init(struct net_device *net_dev,
 	net_dev->features |= NETIF_F_RXCSUM;
 
 	net_dev->priv_flags |= IFF_LIVE_ADDR_CHANGE;
+	net_dev->lltx = true;
 	/* we do not want shared skbs on TX */
 	net_dev->priv_flags &= ~IFF_TX_SKB_SHARING;
 
 	net_dev->features |= net_dev->hw_features;
 	net_dev->vlan_features = net_dev->features;
 
-	memcpy(net_dev->perm_addr, mac_addr, net_dev->addr_len);
-	memcpy(net_dev->dev_addr, mac_addr, net_dev->addr_len);
+	net_dev->xdp_features = NETDEV_XDP_ACT_BASIC |
+				NETDEV_XDP_ACT_REDIRECT |
+				NETDEV_XDP_ACT_NDO_XMIT;
+
+	if (is_valid_ether_addr(mac_addr)) {
+		memcpy(net_dev->perm_addr, mac_addr, net_dev->addr_len);
+		eth_hw_addr_set(net_dev, mac_addr);
+	} else {
+		eth_hw_addr_random(net_dev);
+		err = mac_dev->change_addr(mac_dev->fman_mac,
+			(const enet_addr_t *)net_dev->dev_addr);
+		if (err) {
+			dev_err(dev, "Failed to set random MAC address\n");
+			return -EINVAL;
+		}
+		dev_info(dev, "Using random MAC address: %pM\n",
+			 net_dev->dev_addr);
+	}
 
 	net_dev->ethtool_ops = &dpaa_ethtool_ops;
 
 	net_dev->needed_headroom = priv->tx_headroom;
 	net_dev->watchdog_timeo = msecs_to_jiffies(tx_timeout);
 
+	/* The rest of the config is filled in by the mac device already */
+	mac_dev->phylink_config.dev = &net_dev->dev;
+	mac_dev->phylink_config.type = PHYLINK_NETDEV;
+	mac_dev->update_speed = dpaa_eth_cgr_set_speed;
+	mac_dev->phylink = phylink_create(&mac_dev->phylink_config,
+					  dev_fwnode(mac_dev->dev),
+					  mac_dev->phy_if,
+					  mac_dev->phylink_ops);
+	if (IS_ERR(mac_dev->phylink)) {
+		err = PTR_ERR(mac_dev->phylink);
+		dev_err_probe(dev, err, "Could not create phylink\n");
+		return err;
+	}
+
 	/* start without the RUNNING flag, phylib controls it later */
 	netif_carrier_off(net_dev);
 
 	err = register_netdev(net_dev);
 	if (err < 0) {
 		dev_err(dev, "register_netdev() = %d\n", err);
+		phylink_destroy(mac_dev->phylink);
 		return err;
 	}
 
@@ -275,7 +300,8 @@ static int dpaa_stop(struct net_device *net_dev)
 {
 	struct mac_device *mac_dev;
 	struct dpaa_priv *priv;
-	int i, err, error;
+	int i, error;
+	int err = 0;
 
 	priv = netdev_priv(net_dev);
 	mac_dev = priv->mac_dev;
@@ -284,12 +310,10 @@ static int dpaa_stop(struct net_device *net_dev)
 	/* Allow the Fman (Tx) port to process in-flight frames before we
 	 * try switching it off.
 	 */
-	usleep_range(5000, 10000);
+	msleep(200);
 
-	err = mac_dev->stop(mac_dev);
-	if (err < 0)
-		netif_err(priv, ifdown, net_dev, "mac_dev->stop() = %d\n",
-			  err);
+	phylink_stop(mac_dev->phylink);
+	mac_dev->disable(mac_dev->fman_mac);
 
 	for (i = 0; i < ARRAY_SIZE(mac_dev->port); i++) {
 		error = fman_port_disable(mac_dev->port[i]);
@@ -297,14 +321,15 @@ static int dpaa_stop(struct net_device *net_dev)
 			err = error;
 	}
 
-	if (net_dev->phydev)
-		phy_disconnect(net_dev->phydev);
+	phylink_disconnect_phy(mac_dev->phylink);
 	net_dev->phydev = NULL;
 
+	msleep(200);
+
 	return err;
 }
 
-static void dpaa_tx_timeout(struct net_device *net_dev)
+static void dpaa_tx_timeout(struct net_device *net_dev, unsigned int txqueue)
 {
 	struct dpaa_percpu_priv *percpu_priv;
 	const struct dpaa_priv	*priv;
@@ -342,18 +367,20 @@ static void dpaa_get_stats64(struct net_device *net_dev,
 	}
 }
 
-static int dpaa_setup_tc(struct net_device *net_dev, u32 handle, __be16 proto,
-			 struct tc_to_netdev *tc)
+static int dpaa_setup_tc(struct net_device *net_dev, enum tc_setup_type type,
+			 void *type_data)
 {
 	struct dpaa_priv *priv = netdev_priv(net_dev);
+	int num_txqs_per_tc = dpaa_num_txqs_per_tc();
+	struct tc_mqprio_qopt *mqprio = type_data;
 	u8 num_tc;
 	int i;
 
-	if (tc->type != TC_SETUP_MQPRIO)
-		return -EINVAL;
+	if (type != TC_SETUP_QDISC_MQPRIO)
+		return -EOPNOTSUPP;
 
-	tc->mqprio->hw = TC_MQPRIO_HW_OFFLOAD_TCS;
-	num_tc = tc->mqprio->num_tc;
+	mqprio->hw = TC_MQPRIO_HW_OFFLOAD_TCS;
+	num_tc = mqprio->num_tc;
 
 	if (num_tc == priv->num_tc)
 		return 0;
@@ -372,45 +399,30 @@ static int dpaa_setup_tc(struct net_device *net_dev, u32 handle, __be16 proto,
 	netdev_set_num_tc(net_dev, num_tc);
 
 	for (i = 0; i < num_tc; i++)
-		netdev_set_tc_queue(net_dev, i, DPAA_TC_TXQ_NUM,
-				    i * DPAA_TC_TXQ_NUM);
+		netdev_set_tc_queue(net_dev, i, num_txqs_per_tc,
+				    i * num_txqs_per_tc);
 
 out:
 	priv->num_tc = num_tc ? : 1;
-	netif_set_real_num_tx_queues(net_dev, priv->num_tc * DPAA_TC_TXQ_NUM);
+	netif_set_real_num_tx_queues(net_dev, priv->num_tc * num_txqs_per_tc);
 	return 0;
 }
 
 static struct mac_device *dpaa_mac_dev_get(struct platform_device *pdev)
 {
-	struct platform_device *of_dev;
 	struct dpaa_eth_data *eth_data;
-	struct device *dpaa_dev, *dev;
-	struct device_node *mac_node;
+	struct device *dpaa_dev;
 	struct mac_device *mac_dev;
 
 	dpaa_dev = &pdev->dev;
 	eth_data = dpaa_dev->platform_data;
-	if (!eth_data)
+	if (!eth_data) {
+		dev_err(dpaa_dev, "eth_data missing\n");
 		return ERR_PTR(-ENODEV);
-
-	mac_node = eth_data->mac_node;
-
-	of_dev = of_find_device_by_node(mac_node);
-	if (!of_dev) {
-		dev_err(dpaa_dev, "of_find_device_by_node(%s) failed\n",
-			mac_node->full_name);
-		of_node_put(mac_node);
-		return ERR_PTR(-EINVAL);
 	}
-	of_node_put(mac_node);
-
-	dev = &of_dev->dev;
-
-	mac_dev = dev_get_drvdata(dev);
+	mac_dev = eth_data->mac_dev;
 	if (!mac_dev) {
-		dev_err(dpaa_dev, "dev_get_drvdata(%s) failed\n",
-			dev_name(dev));
+		dev_err(dpaa_dev, "mac_dev missing\n");
 		return ERR_PTR(-EINVAL);
 	}
 
@@ -437,7 +449,7 @@ static int dpaa_set_mac_address(struct net_device *net_dev, void *addr)
 	mac_dev = priv->mac_dev;
 
 	err = mac_dev->change_addr(mac_dev->fman_mac,
-				   (enet_addr_t *)net_dev->dev_addr);
+				   (const enet_addr_t *)net_dev->dev_addr);
 	if (err < 0) {
 		netif_err(priv, drv, net_dev, "mac_dev->change_addr() = %d\n",
 			  err);
@@ -450,6 +462,22 @@ static int dpaa_set_mac_address(struct net_device *net_dev, void *addr)
 	return 0;
 }
 
+static int dpaa_addr_sync(struct net_device *net_dev, const u8 *addr)
+{
+	const struct dpaa_priv *priv = netdev_priv(net_dev);
+
+	return priv->mac_dev->add_hash_mac_addr(priv->mac_dev->fman_mac,
+						(enet_addr_t *)addr);
+}
+
+static int dpaa_addr_unsync(struct net_device *net_dev, const u8 *addr)
+{
+	const struct dpaa_priv *priv = netdev_priv(net_dev);
+
+	return priv->mac_dev->remove_hash_mac_addr(priv->mac_dev->fman_mac,
+						   (enet_addr_t *)addr);
+}
+
 static void dpaa_set_rx_mode(struct net_device *net_dev)
 {
 	const struct dpaa_priv	*priv;
@@ -467,9 +495,19 @@ static void dpaa_set_rx_mode(struct net_device *net_dev)
 				  err);
 	}
 
-	err = priv->mac_dev->set_multi(net_dev, priv->mac_dev);
+	if (!!(net_dev->flags & IFF_ALLMULTI) != priv->mac_dev->allmulti) {
+		priv->mac_dev->allmulti = !priv->mac_dev->allmulti;
+		err = priv->mac_dev->set_allmulti(priv->mac_dev->fman_mac,
+						  priv->mac_dev->allmulti);
+		if (err < 0)
+			netif_err(priv, drv, net_dev,
+				  "mac_dev->set_allmulti() = %d\n",
+				  err);
+	}
+
+	err = __dev_mc_sync(net_dev, dpaa_addr_sync, dpaa_addr_unsync);
 	if (err < 0)
-		netif_err(priv, drv, net_dev, "mac_dev->set_multi() = %d\n",
+		netif_err(priv, drv, net_dev, "dpaa_addr_sync() = %d\n",
 			  err);
 }
 
@@ -485,7 +523,7 @@ static struct dpaa_bp *dpaa_bpid2pool(int bpid)
 static bool dpaa_bpid2pool_use(int bpid)
 {
 	if (dpaa_bpid2pool(bpid)) {
-		atomic_inc(&dpaa_bp_array[bpid]->refs);
+		refcount_inc(&dpaa_bp_array[bpid]->refs);
 		return true;
 	}
 
@@ -496,7 +534,7 @@ static bool dpaa_bpid2pool_use(int bpid)
 static void dpaa_bpid2pool_map(int bpid, struct dpaa_bp *dpaa_bp)
 {
 	dpaa_bp_array[bpid] = dpaa_bp;
-	atomic_set(&dpaa_bp->refs, 1);
+	refcount_set(&dpaa_bp->refs, 1);
 }
 
 static int dpaa_bp_alloc_pool(struct dpaa_bp *dpaa_bp)
@@ -584,7 +622,7 @@ static void dpaa_bp_free(struct dpaa_bp *dpaa_bp)
 	if (!bp)
 		return;
 
-	if (!atomic_dec_and_test(&bp->refs))
+	if (!refcount_dec_and_test(&bp->refs))
 		return;
 
 	if (bp->free_buf_cb)
@@ -596,10 +634,7 @@ static void dpaa_bp_free(struct dpaa_bp *dpaa_bp)
 
 static void dpaa_bps_free(struct dpaa_priv *priv)
 {
-	int i;
-
-	for (i = 0; i < DPAA_BPS_NUM; i++)
-		dpaa_bp_free(priv->dpaa_bps[i]);
+	dpaa_bp_free(priv->dpaa_bp);
 }
 
 /* Use multiple WQs for FQ assignment:
@@ -627,10 +662,11 @@ static inline void dpaa_assign_wq(struct dpaa_fq *fq, int idx)
 		fq->wq = 5;
 		break;
 	case FQ_TYPE_RX_DEFAULT:
+	case FQ_TYPE_RX_PCD:
 		fq->wq = 6;
 		break;
 	case FQ_TYPE_TX:
-		switch (idx / DPAA_TC_TXQ_NUM) {
+		switch (idx / dpaa_num_txqs_per_tc()) {
 		case 0:
 			/* Low priority (best effort) */
 			fq->wq = 6;
@@ -648,8 +684,8 @@ static inline void dpaa_assign_wq(struct dpaa_fq *fq, int idx)
 			fq->wq = 0;
 			break;
 		default:
-			WARN(1, "Too many TX FQs: more than %d!\n",
-			     DPAA_ETH_TXQ_NUM);
+			WARN(1, "Too many TX FQs: more than %zu!\n",
+			     dpaa_max_num_txqs());
 		}
 		break;
 	default:
@@ -666,7 +702,7 @@ static struct dpaa_fq *dpaa_fq_alloc(struct device *dev,
 	struct dpaa_fq *dpaa_fq;
 	int i;
 
-	dpaa_fq = devm_kzalloc(dev, sizeof(*dpaa_fq) * count,
+	dpaa_fq = devm_kcalloc(dev, count, sizeof(*dpaa_fq),
 			       GFP_KERNEL);
 	if (!dpaa_fq)
 		return NULL;
@@ -687,6 +723,7 @@ static int dpaa_alloc_all_fqs(struct device *dev, struct list_head *list,
 			      struct fm_port_fqs *port_fqs)
 {
 	struct dpaa_fq *dpaa_fq;
+	u32 fq_base, fq_base_aligned, i;
 
 	dpaa_fq = dpaa_fq_alloc(dev, 0, 1, list, FQ_TYPE_RX_ERROR);
 	if (!dpaa_fq)
@@ -700,7 +737,28 @@ static int dpaa_alloc_all_fqs(struct device *dev, struct list_head *list,
 
 	port_fqs->rx_defq = &dpaa_fq[0];
 
-	if (!dpaa_fq_alloc(dev, 0, DPAA_ETH_TXQ_NUM, list, FQ_TYPE_TX_CONF_MQ))
+	/* the PCD FQIDs range needs to be aligned for correct operation */
+	if (qman_alloc_fqid_range(&fq_base, 2 * DPAA_ETH_PCD_RXQ_NUM))
+		goto fq_alloc_failed;
+
+	fq_base_aligned = ALIGN(fq_base, DPAA_ETH_PCD_RXQ_NUM);
+
+	for (i = fq_base; i < fq_base_aligned; i++)
+		qman_release_fqid(i);
+
+	for (i = fq_base_aligned + DPAA_ETH_PCD_RXQ_NUM;
+	     i < (fq_base + 2 * DPAA_ETH_PCD_RXQ_NUM); i++)
+		qman_release_fqid(i);
+
+	dpaa_fq = dpaa_fq_alloc(dev, fq_base_aligned, DPAA_ETH_PCD_RXQ_NUM,
+				list, FQ_TYPE_RX_PCD);
+	if (!dpaa_fq)
+		goto fq_alloc_failed;
+
+	port_fqs->rx_pcdq = &dpaa_fq[0];
+
+	if (!dpaa_fq_alloc(dev, 0, dpaa_max_num_txqs(), list,
+			   FQ_TYPE_TX_CONF_MQ))
 		goto fq_alloc_failed;
 
 	dpaa_fq = dpaa_fq_alloc(dev, 0, 1, list, FQ_TYPE_TX_ERROR);
@@ -715,7 +773,7 @@ static int dpaa_alloc_all_fqs(struct device *dev, struct list_head *list,
 
 	port_fqs->tx_defq = &dpaa_fq[0];
 
-	if (!dpaa_fq_alloc(dev, 0, DPAA_ETH_TXQ_NUM, list, FQ_TYPE_TX))
+	if (!dpaa_fq_alloc(dev, 0, dpaa_max_num_txqs(), list, FQ_TYPE_TX))
 		goto fq_alloc_failed;
 
 	return 0;
@@ -751,16 +809,17 @@ static void dpaa_release_channel(void)
 	qman_release_pool(rx_pool_channel);
 }
 
-static void dpaa_eth_add_channel(u16 channel)
+static void dpaa_eth_add_channel(u16 channel, struct device *dev)
 {
 	u32 pool = QM_SDQCR_CHANNELS_POOL_CONV(channel);
 	const cpumask_t *cpus = qman_affine_cpus();
 	struct qman_portal *portal;
 	int cpu;
 
-	for_each_cpu(cpu, cpus) {
+	for_each_cpu_and(cpu, cpus, cpu_online_mask) {
 		portal = qman_get_affine_portal(cpu);
 		qman_p_static_dequeue_add(portal, pool);
+		qman_start_using_portal(portal, dev);
 	}
 }
 
@@ -806,12 +865,12 @@ static int dpaa_eth_cgr_init(struct dpaa_priv *priv)
 	initcgr.we_mask = cpu_to_be16(QM_CGR_WE_CSCN_EN | QM_CGR_WE_CS_THRES);
 	initcgr.cgr.cscn_en = QM_CGR_EN;
 
-	/* Set different thresholds based on the MAC speed.
-	 * This may turn suboptimal if the MAC is reconfigured at a speed
-	 * lower than its max, e.g. if a dTSEC later negotiates a 100Mbps link.
-	 * In such cases, we ought to reconfigure the threshold, too.
+	/* Set different thresholds based on the configured MAC speed.
+	 * This may turn suboptimal if the MAC is reconfigured at another
+	 * speed, so MACs must call dpaa_eth_cgr_set_speed in their link_up
+	 * callback.
 	 */
-	if (priv->mac_dev->if_support & SUPPORTED_10000baseT_Full)
+	if (priv->mac_dev->phylink_config.mac_capabilities & MAC_10000FD)
 		cs_th = DPAA_CS_THRESHOLD_10G;
 	else
 		cs_th = DPAA_CS_THRESHOLD_1G;
@@ -838,6 +897,31 @@ out_error:
 	return err;
 }
 
+static void dpaa_eth_cgr_set_speed(struct mac_device *mac_dev, int speed)
+{
+	struct net_device *net_dev = to_net_dev(mac_dev->phylink_config.dev);
+	struct dpaa_priv *priv = netdev_priv(net_dev);
+	struct qm_mcc_initcgr opts = { };
+	u32 cs_th;
+	int err;
+
+	opts.we_mask = cpu_to_be16(QM_CGR_WE_CS_THRES);
+	switch (speed) {
+	case SPEED_10000:
+		cs_th = DPAA_CS_THRESHOLD_10G;
+		break;
+	case SPEED_1000:
+	default:
+		cs_th = DPAA_CS_THRESHOLD_1G;
+		break;
+	}
+	qm_cgr_cs_thres_set64(&opts.cgr.cs_thres, cs_th, 1);
+
+	err = qman_update_cgr_safe(&priv->cgr_data.cgr, &opts);
+	if (err)
+		netdev_err(net_dev, "could not update speed: %d\n", err);
+}
+
 static inline void dpaa_setup_ingress(const struct dpaa_priv *priv,
 				      struct dpaa_fq *fq,
 				      const struct qman_fq *template)
@@ -865,20 +949,25 @@ static inline void dpaa_setup_egress(const struct dpaa_priv *priv,
 	}
 }
 
-static void dpaa_fq_setup(struct dpaa_priv *priv,
-			  const struct dpaa_fq_cbs *fq_cbs,
-			  struct fman_port *tx_port)
+static int dpaa_fq_setup(struct dpaa_priv *priv,
+			 const struct dpaa_fq_cbs *fq_cbs,
+			 struct fman_port *tx_port)
 {
-	int egress_cnt = 0, conf_cnt = 0, num_portals = 0, cpu;
+	int egress_cnt = 0, conf_cnt = 0, num_portals = 0, portal_cnt = 0, cpu;
 	const cpumask_t *affine_cpus = qman_affine_cpus();
-	u16 portals[NR_CPUS];
 	struct dpaa_fq *fq;
+	u16 *channels;
+
+	channels = kcalloc(num_possible_cpus(), sizeof(u16), GFP_KERNEL);
+	if (!channels)
+		return -ENOMEM;
+
+	for_each_cpu_and(cpu, affine_cpus, cpu_online_mask)
+		channels[num_portals++] = qman_affine_channel(cpu);
 
-	for_each_cpu(cpu, affine_cpus)
-		portals[num_portals++] = qman_affine_channel(cpu);
 	if (num_portals == 0)
 		dev_err(priv->net_dev->dev.parent,
-			"No Qman software (affine) channels found");
+			"No Qman software (affine) channels found\n");
 
 	/* Initialize each FQ in the list */
 	list_for_each_entry(fq, &priv->dpaa_fq_list, list) {
@@ -889,18 +978,20 @@ static void dpaa_fq_setup(struct dpaa_priv *priv,
 		case FQ_TYPE_RX_ERROR:
 			dpaa_setup_ingress(priv, fq, &fq_cbs->rx_errq);
 			break;
+		case FQ_TYPE_RX_PCD:
+			if (!num_portals)
+				continue;
+			dpaa_setup_ingress(priv, fq, &fq_cbs->rx_defq);
+			fq->channel = channels[portal_cnt++ % num_portals];
+			break;
 		case FQ_TYPE_TX:
 			dpaa_setup_egress(priv, fq, tx_port,
 					  &fq_cbs->egress_ern);
-			/* If we have more Tx queues than the number of cores,
-			 * just ignore the extra ones.
-			 */
-			if (egress_cnt < DPAA_ETH_TXQ_NUM)
-				priv->egress_fqs[egress_cnt++] = &fq->fq_base;
+			priv->egress_fqs[egress_cnt++] = &fq->fq_base;
 			break;
 		case FQ_TYPE_TX_CONF_MQ:
 			priv->conf_fqs[conf_cnt++] = &fq->fq_base;
-			/* fall through */
+			fallthrough;
 		case FQ_TYPE_TX_CONFIRM:
 			dpaa_setup_ingress(priv, fq, &fq_cbs->tx_defq);
 			break;
@@ -914,16 +1005,9 @@ static void dpaa_fq_setup(struct dpaa_priv *priv,
 		}
 	}
 
-	 /* Make sure all CPUs receive a corresponding Tx queue. */
-	while (egress_cnt < DPAA_ETH_TXQ_NUM) {
-		list_for_each_entry(fq, &priv->dpaa_fq_list, list) {
-			if (fq->fq_type != FQ_TYPE_TX)
-				continue;
-			priv->egress_fqs[egress_cnt++] = &fq->fq_base;
-			if (egress_cnt == DPAA_ETH_TXQ_NUM)
-				break;
-		}
-	}
+	kfree(channels);
+
+	return 0;
 }
 
 static inline int dpaa_tx_fq_to_id(const struct dpaa_priv *priv,
@@ -931,7 +1015,7 @@ static inline int dpaa_tx_fq_to_id(const struct dpaa_priv *priv,
 {
 	int i;
 
-	for (i = 0; i < DPAA_ETH_TXQ_NUM; i++)
+	for (i = 0; i < dpaa_max_num_txqs(); i++)
 		if (priv->egress_fqs[i] == tx_fq)
 			return i;
 
@@ -1038,7 +1122,8 @@ static int dpaa_fq_init(struct dpaa_fq *dpaa_fq, bool td_enable)
 		/* Put all the ingress queues in our "ingress CGR". */
 		if (priv->use_ingress_cgr &&
 		    (dpaa_fq->fq_type == FQ_TYPE_RX_DEFAULT ||
-		     dpaa_fq->fq_type == FQ_TYPE_RX_ERROR)) {
+		     dpaa_fq->fq_type == FQ_TYPE_RX_ERROR ||
+		     dpaa_fq->fq_type == FQ_TYPE_RX_PCD)) {
 			initfq.we_mask |= cpu_to_be16(QM_INITFQ_WE_CGID);
 			initfq.fqd.fq_ctrl |= cpu_to_be16(QM_FQCTRL_CGE);
 			initfq.fqd.cgid = (u8)priv->ingress_cgr.cgrid;
@@ -1077,6 +1162,25 @@ static int dpaa_fq_init(struct dpaa_fq *dpaa_fq, bool td_enable)
 
 	dpaa_fq->fqid = qman_fq_fqid(fq);
 
+	if (dpaa_fq->fq_type == FQ_TYPE_RX_DEFAULT ||
+	    dpaa_fq->fq_type == FQ_TYPE_RX_PCD) {
+		err = xdp_rxq_info_reg(&dpaa_fq->xdp_rxq, dpaa_fq->net_dev,
+				       dpaa_fq->fqid, 0);
+		if (err) {
+			dev_err(dev, "xdp_rxq_info_reg() = %d\n", err);
+			return err;
+		}
+
+		err = xdp_rxq_info_reg_mem_model(&dpaa_fq->xdp_rxq,
+						 MEM_TYPE_PAGE_ORDER0, NULL);
+		if (err) {
+			dev_err(dev, "xdp_rxq_info_reg_mem_model() = %d\n",
+				err);
+			xdp_rxq_info_unreg(&dpaa_fq->xdp_rxq);
+			return err;
+		}
+	}
+
 	return 0;
 }
 
@@ -1106,6 +1210,11 @@ static int dpaa_fq_free_entry(struct device *dev, struct qman_fq *fq)
 		}
 	}
 
+	if ((dpaa_fq->fq_type == FQ_TYPE_RX_DEFAULT ||
+	     dpaa_fq->fq_type == FQ_TYPE_RX_PCD) &&
+	    xdp_rxq_info_is_reg(&dpaa_fq->xdp_rxq))
+		xdp_rxq_info_unreg(&dpaa_fq->xdp_rxq);
+
 	qman_destroy_fq(fq);
 	list_del(&dpaa_fq->list);
 
@@ -1141,7 +1250,7 @@ static int dpaa_eth_init_tx_port(struct fman_port *port, struct dpaa_fq *errq,
 	buf_prefix_content.priv_data_size = buf_layout->priv_data_size;
 	buf_prefix_content.pass_prs_result = true;
 	buf_prefix_content.pass_hash_result = true;
-	buf_prefix_content.pass_time_stamp = false;
+	buf_prefix_content.pass_time_stamp = true;
 	buf_prefix_content.data_align = DPAA_FD_DATA_ALIGNMENT;
 
 	params.specific_params.non_rx_params.err_fqid = errq->fqid;
@@ -1167,15 +1276,15 @@ static int dpaa_eth_init_tx_port(struct fman_port *port, struct dpaa_fq *errq,
 	return err;
 }
 
-static int dpaa_eth_init_rx_port(struct fman_port *port, struct dpaa_bp **bps,
-				 size_t count, struct dpaa_fq *errq,
-				 struct dpaa_fq *defq,
+static int dpaa_eth_init_rx_port(struct fman_port *port, struct dpaa_bp *bp,
+				 struct dpaa_fq *errq,
+				 struct dpaa_fq *defq, struct dpaa_fq *pcdq,
 				 struct dpaa_buffer_layout *buf_layout)
 {
 	struct fman_buffer_prefix_content buf_prefix_content;
 	struct fman_port_rx_params *rx_p;
 	struct fman_port_params params;
-	int i, err;
+	int err;
 
 	memset(&params, 0, sizeof(params));
 	memset(&buf_prefix_content, 0, sizeof(buf_prefix_content));
@@ -1183,20 +1292,21 @@ static int dpaa_eth_init_rx_port(struct fman_port *port, struct dpaa_bp **bps,
 	buf_prefix_content.priv_data_size = buf_layout->priv_data_size;
 	buf_prefix_content.pass_prs_result = true;
 	buf_prefix_content.pass_hash_result = true;
-	buf_prefix_content.pass_time_stamp = false;
-	buf_prefix_content.data_align = DPAA_FD_DATA_ALIGNMENT;
+	buf_prefix_content.pass_time_stamp = true;
+	buf_prefix_content.data_align = DPAA_FD_RX_DATA_ALIGNMENT;
 
 	rx_p = &params.specific_params.rx_params;
 	rx_p->err_fqid = errq->fqid;
 	rx_p->dflt_fqid = defq->fqid;
-
-	count = min(ARRAY_SIZE(rx_p->ext_buf_pools.ext_buf_pool), count);
-	rx_p->ext_buf_pools.num_of_pools_used = (u8)count;
-	for (i = 0; i < count; i++) {
-		rx_p->ext_buf_pools.ext_buf_pool[i].id =  bps[i]->bpid;
-		rx_p->ext_buf_pools.ext_buf_pool[i].size = (u16)bps[i]->size;
+	if (pcdq) {
+		rx_p->pcd_base_fqid = pcdq->fqid;
+		rx_p->pcd_fqs_count = DPAA_ETH_PCD_RXQ_NUM;
 	}
 
+	rx_p->ext_buf_pools.num_of_pools_used = 1;
+	rx_p->ext_buf_pools.ext_buf_pool[0].id =  bp->bpid;
+	rx_p->ext_buf_pools.ext_buf_pool[0].size = (u16)bp->size;
+
 	err = fman_port_config(port, &params);
 	if (err) {
 		pr_err("%s: fman_port_config failed\n", __func__);
@@ -1218,7 +1328,7 @@ static int dpaa_eth_init_rx_port(struct fman_port *port, struct dpaa_bp **bps,
 }
 
 static int dpaa_eth_init_ports(struct mac_device *mac_dev,
-			       struct dpaa_bp **bps, size_t count,
+			       struct dpaa_bp *bp,
 			       struct fm_port_fqs *port_fqs,
 			       struct dpaa_buffer_layout *buf_layout,
 			       struct device *dev)
@@ -1232,8 +1342,9 @@ static int dpaa_eth_init_ports(struct mac_device *mac_dev,
 	if (err)
 		return err;
 
-	err = dpaa_eth_init_rx_port(rxport, bps, count, port_fqs->rx_errq,
-				    port_fqs->rx_defq, &buf_layout[RX]);
+	err = dpaa_eth_init_rx_port(rxport, bp, port_fqs->rx_errq,
+				    port_fqs->rx_defq, port_fqs->rx_pcdq,
+				    &buf_layout[RX]);
 
 	return err;
 }
@@ -1245,7 +1356,7 @@ static int dpaa_bman_release(const struct dpaa_bp *dpaa_bp,
 
 	err = bman_release(dpaa_bp->pool, bmb, cnt);
 	/* Should never occur, address anyway to avoid leaking the buffers */
-	if (unlikely(WARN_ON(err)) && dpaa_bp->free_buf_cb)
+	if (WARN_ON(err) && dpaa_bp->free_buf_cb)
 		while (cnt-- > 0)
 			dpaa_bp->free_buf_cb(dpaa_bp, &bmb[cnt]);
 
@@ -1300,15 +1411,16 @@ static void dpaa_fd_release(const struct net_device *net_dev,
 		vaddr = phys_to_virt(qm_fd_addr(fd));
 		sgt = vaddr + qm_fd_get_offset(fd);
 
-		dma_unmap_single(dpaa_bp->dev, qm_fd_addr(fd), dpaa_bp->size,
-				 DMA_FROM_DEVICE);
+		dma_unmap_page(dpaa_bp->priv->rx_dma_dev, qm_fd_addr(fd),
+			       DPAA_BP_RAW_SIZE, DMA_FROM_DEVICE);
 
 		dpaa_release_sgt_members(sgt);
 
-		addr = dma_map_single(dpaa_bp->dev, vaddr, dpaa_bp->size,
-				      DMA_FROM_DEVICE);
-		if (dma_mapping_error(dpaa_bp->dev, addr)) {
-			dev_err(dpaa_bp->dev, "DMA mapping failed");
+		addr = dma_map_page(dpaa_bp->priv->rx_dma_dev,
+				    virt_to_page(vaddr), 0, DPAA_BP_RAW_SIZE,
+				    DMA_FROM_DEVICE);
+		if (dma_mapping_error(dpaa_bp->priv->rx_dma_dev, addr)) {
+			netdev_err(net_dev, "DMA mapping failed\n");
 			return;
 		}
 		bm_buffer_set64(&bmb, addr);
@@ -1361,7 +1473,7 @@ static void count_ern(struct dpaa_percpu_priv *percpu_priv,
 static int dpaa_enable_tx_csum(struct dpaa_priv *priv,
 			       struct sk_buff *skb,
 			       struct qm_fd *fd,
-			       char *parse_results)
+			       void *parse_results)
 {
 	struct fman_prs_result *parse_result;
 	u16 ethertype = ntohs(skb->protocol);
@@ -1383,13 +1495,8 @@ static int dpaa_enable_tx_csum(struct dpaa_priv *priv,
 	parse_result = (struct fman_prs_result *)parse_results;
 
 	/* If we're dealing with VLAN, get the real Ethernet type */
-	if (ethertype == ETH_P_8021Q) {
-		/* We can't always assume the MAC header is set correctly
-		 * by the stack, so reset to beginning of skb->data
-		 */
-		skb_reset_mac_header(skb);
-		ethertype = ntohs(vlan_eth_hdr(skb)->h_vlan_encapsulated_proto);
-	}
+	if (ethertype == ETH_P_8021Q)
+		ethertype = ntohs(skb_vlan_eth_hdr(skb)->h_vlan_encapsulated_proto);
 
 	/* Fill in the relevant L3 parse result fields
 	 * and read the L4 protocol type
@@ -1453,25 +1560,24 @@ return_error:
 
 static int dpaa_bp_add_8_bufs(const struct dpaa_bp *dpaa_bp)
 {
-	struct device *dev = dpaa_bp->dev;
+	struct net_device *net_dev = dpaa_bp->priv->net_dev;
 	struct bm_buffer bmb[8];
 	dma_addr_t addr;
-	void *new_buf;
+	struct page *p;
 	u8 i;
 
 	for (i = 0; i < 8; i++) {
-		new_buf = netdev_alloc_frag(dpaa_bp->raw_size);
-		if (unlikely(!new_buf)) {
-			dev_err(dev, "netdev_alloc_frag() failed, size %zu\n",
-				dpaa_bp->raw_size);
+		p = dev_alloc_pages(0);
+		if (unlikely(!p)) {
+			netdev_err(net_dev, "dev_alloc_pages() failed\n");
 			goto release_previous_buffs;
 		}
-		new_buf = PTR_ALIGN(new_buf, SMP_CACHE_BYTES);
 
-		addr = dma_map_single(dev, new_buf,
-				      dpaa_bp->size, DMA_FROM_DEVICE);
-		if (unlikely(dma_mapping_error(dev, addr))) {
-			dev_err(dpaa_bp->dev, "DMA map failed");
+		addr = dma_map_page(dpaa_bp->priv->rx_dma_dev, p, 0,
+				    DPAA_BP_RAW_SIZE, DMA_FROM_DEVICE);
+		if (unlikely(dma_mapping_error(dpaa_bp->priv->rx_dma_dev,
+					       addr))) {
+			netdev_err(net_dev, "DMA map failed\n");
 			goto release_previous_buffs;
 		}
 
@@ -1546,18 +1652,13 @@ static int dpaa_eth_refill_bpools(struct dpaa_priv *priv)
 {
 	struct dpaa_bp *dpaa_bp;
 	int *countptr;
-	int res, i;
 
-	for (i = 0; i < DPAA_BPS_NUM; i++) {
-		dpaa_bp = priv->dpaa_bps[i];
-		if (!dpaa_bp)
-			return -EINVAL;
-		countptr = this_cpu_ptr(dpaa_bp->percpu_count);
-		res  = dpaa_eth_refill_bpool(dpaa_bp, countptr);
-		if (res)
-			return res;
-	}
-	return 0;
+	dpaa_bp = priv->dpaa_bp;
+	if (!dpaa_bp)
+		return -EINVAL;
+	countptr = this_cpu_ptr(dpaa_bp->percpu_count);
+
+	return dpaa_eth_refill_bpool(dpaa_bp, countptr);
 }
 
 /* Cleanup function for outgoing frame descriptors that were built on Tx path,
@@ -1565,54 +1666,87 @@ static int dpaa_eth_refill_bpools(struct dpaa_priv *priv)
  * Skb freeing is not handled here.
  *
  * This function may be called on error paths in the Tx function, so guard
- * against cases when not all fd relevant fields were filled in.
+ * against cases when not all fd relevant fields were filled in. To avoid
+ * reading the invalid transmission timestamp for the error paths set ts to
+ * false.
  *
  * Return the skb backpointer, since for S/G frames the buffer containing it
  * gets freed here.
+ *
+ * No skb backpointer is set when transmitting XDP frames. Cleanup the buffer
+ * and return NULL in this case.
  */
 static struct sk_buff *dpaa_cleanup_tx_fd(const struct dpaa_priv *priv,
-					  const struct qm_fd *fd)
+					  const struct qm_fd *fd, bool ts)
 {
 	const enum dma_data_direction dma_dir = DMA_TO_DEVICE;
 	struct device *dev = priv->net_dev->dev.parent;
+	struct skb_shared_hwtstamps shhwtstamps;
 	dma_addr_t addr = qm_fd_addr(fd);
+	void *vaddr = phys_to_virt(addr);
 	const struct qm_sg_entry *sgt;
-	struct sk_buff **skbh, *skb;
-	int nr_frags, i;
-
-	skbh = (struct sk_buff **)phys_to_virt(addr);
-	skb = *skbh;
+	struct dpaa_eth_swbp *swbp;
+	struct sk_buff *skb;
+	u64 ns;
+	int i;
 
 	if (unlikely(qm_fd_get_format(fd) == qm_fd_sg)) {
-		nr_frags = skb_shinfo(skb)->nr_frags;
-		dma_unmap_single(dev, addr, qm_fd_get_offset(fd) +
-				 sizeof(struct qm_sg_entry) * (1 + nr_frags),
-				 dma_dir);
+		dma_unmap_page(priv->tx_dma_dev, addr,
+			       qm_fd_get_offset(fd) + DPAA_SGT_SIZE,
+			       dma_dir);
 
 		/* The sgt buffer has been allocated with netdev_alloc_frag(),
 		 * it's from lowmem.
 		 */
-		sgt = phys_to_virt(addr + qm_fd_get_offset(fd));
+		sgt = vaddr + qm_fd_get_offset(fd);
 
 		/* sgt[0] is from lowmem, was dma_map_single()-ed */
-		dma_unmap_single(dev, qm_sg_addr(&sgt[0]),
+		dma_unmap_single(priv->tx_dma_dev, qm_sg_addr(&sgt[0]),
 				 qm_sg_entry_get_len(&sgt[0]), dma_dir);
 
 		/* remaining pages were mapped with skb_frag_dma_map() */
-		for (i = 1; i < nr_frags; i++) {
+		for (i = 1; (i < DPAA_SGT_MAX_ENTRIES) &&
+		     !qm_sg_entry_is_final(&sgt[i - 1]); i++) {
 			WARN_ON(qm_sg_entry_is_ext(&sgt[i]));
 
-			dma_unmap_page(dev, qm_sg_addr(&sgt[i]),
+			dma_unmap_page(priv->tx_dma_dev, qm_sg_addr(&sgt[i]),
 				       qm_sg_entry_get_len(&sgt[i]), dma_dir);
 		}
-
-		/* Free the page frag that we allocated on Tx */
-		skb_free_frag(phys_to_virt(addr));
 	} else {
-		dma_unmap_single(dev, addr,
-				 skb_tail_pointer(skb) - (u8 *)skbh, dma_dir);
+		dma_unmap_single(priv->tx_dma_dev, addr,
+				 qm_fd_get_offset(fd) + qm_fd_get_length(fd),
+				 dma_dir);
+	}
+
+	swbp = (struct dpaa_eth_swbp *)vaddr;
+	skb = swbp->skb;
+
+	/* No skb backpointer is set when running XDP. An xdp_frame
+	 * backpointer is saved instead.
+	 */
+	if (!skb) {
+		xdp_return_frame(swbp->xdpf);
+		return NULL;
+	}
+
+	/* DMA unmapping is required before accessing the HW provided info */
+	if (ts && priv->tx_tstamp &&
+	    skb_shinfo(skb)->tx_flags & SKBTX_HW_TSTAMP) {
+		memset(&shhwtstamps, 0, sizeof(shhwtstamps));
+
+		if (!fman_port_get_tstamp(priv->mac_dev->port[TX], vaddr,
+					  &ns)) {
+			shhwtstamps.hwtstamp = ns_to_ktime(ns);
+			skb_tstamp_tx(skb, &shhwtstamps);
+		} else {
+			dev_warn(dev, "fman_port_get_tstamp failed!\n");
+		}
 	}
 
+	if (qm_fd_get_format(fd) == qm_fd_sg)
+		/* Free the page that we allocated on Tx for the SGT */
+		free_pages((unsigned long)vaddr, 0);
+
 	return skb;
 }
 
@@ -1633,6 +1767,8 @@ static u8 rx_csum_offload(const struct dpaa_priv *priv, const struct qm_fd *fd)
 	return CHECKSUM_NONE;
 }
 
+#define PTR_IS_ALIGNED(x, a) (IS_ALIGNED((unsigned long)(x), (a)))
+
 /* Build a linear skb around the received buffer.
  * We are guaranteed there is enough room at the end of the data buffer to
  * accommodate the shared info area of the skb.
@@ -1655,11 +1791,8 @@ static struct sk_buff *contig_fd_to_skb(const struct dpaa_priv *priv,
 
 	skb = build_skb(vaddr, dpaa_bp->size +
 			SKB_DATA_ALIGN(sizeof(struct skb_shared_info)));
-	if (unlikely(!skb)) {
-		WARN_ONCE(1, "Build skb failure on Rx\n");
+	if (WARN_ONCE(!skb, "Build skb failure on Rx\n"))
 		goto free_buffer;
-	}
-	WARN_ON(fd_off != priv->rx_headroom);
 	skb_reserve(skb, fd_off);
 	skb_put(skb, qm_fd_get_length(fd));
 
@@ -1668,7 +1801,7 @@ static struct sk_buff *contig_fd_to_skb(const struct dpaa_priv *priv,
 	return skb;
 
 free_buffer:
-	skb_free_frag(vaddr);
+	free_pages((unsigned long)vaddr, 0);
 	return NULL;
 }
 
@@ -1686,41 +1819,40 @@ static struct sk_buff *sg_fd_to_skb(const struct dpaa_priv *priv,
 	struct page *page, *head_page;
 	struct dpaa_bp *dpaa_bp;
 	void *vaddr, *sg_vaddr;
-	int frag_off, frag_len;
 	struct sk_buff *skb;
 	dma_addr_t sg_addr;
 	int page_offset;
 	unsigned int sz;
 	int *count_ptr;
-	int i;
+	int i, j;
 
 	vaddr = phys_to_virt(addr);
 	WARN_ON(!IS_ALIGNED((unsigned long)vaddr, SMP_CACHE_BYTES));
 
 	/* Iterate through the SGT entries and add data buffers to the skb */
 	sgt = vaddr + fd_off;
+	skb = NULL;
 	for (i = 0; i < DPAA_SGT_MAX_ENTRIES; i++) {
 		/* Extension bit is not supported */
 		WARN_ON(qm_sg_entry_is_ext(&sgt[i]));
 
 		sg_addr = qm_sg_addr(&sgt[i]);
 		sg_vaddr = phys_to_virt(sg_addr);
-		WARN_ON(!IS_ALIGNED((unsigned long)sg_vaddr,
-				    SMP_CACHE_BYTES));
+		WARN_ON(!PTR_IS_ALIGNED(sg_vaddr, SMP_CACHE_BYTES));
+
+		dma_unmap_page(priv->rx_dma_dev, sg_addr,
+			       DPAA_BP_RAW_SIZE, DMA_FROM_DEVICE);
 
 		/* We may use multiple Rx pools */
 		dpaa_bp = dpaa_bpid2pool(sgt[i].bpid);
 		if (!dpaa_bp)
 			goto free_buffers;
 
-		count_ptr = this_cpu_ptr(dpaa_bp->percpu_count);
-		dma_unmap_single(dpaa_bp->dev, sg_addr, dpaa_bp->size,
-				 DMA_FROM_DEVICE);
-		if (i == 0) {
+		if (!skb) {
 			sz = dpaa_bp->size +
 				SKB_DATA_ALIGN(sizeof(struct skb_shared_info));
 			skb = build_skb(sg_vaddr, sz);
-			if (WARN_ON(unlikely(!skb)))
+			if (WARN_ON(!skb))
 				goto free_buffers;
 
 			skb->ip_summed = rx_csum_offload(priv, fd);
@@ -1729,6 +1861,11 @@ static struct sk_buff *sg_fd_to_skb(const struct dpaa_priv *priv,
 			 * on Tx, if extra headers are added.
 			 */
 			WARN_ON(fd_off != priv->rx_headroom);
+			/* The offset to data start within the buffer holding
+			 * the SGT should always be equal to the offset to data
+			 * start within the first buffer holding the frame.
+			 */
+			WARN_ON_ONCE(fd_off != qm_sg_entry_get_off(&sgt[i]));
 			skb_reserve(skb, fd_off);
 			skb_put(skb, qm_sg_entry_get_len(&sgt[i]));
 		} else {
@@ -1742,23 +1879,27 @@ static struct sk_buff *sg_fd_to_skb(const struct dpaa_priv *priv,
 			page = virt_to_page(sg_vaddr);
 			head_page = virt_to_head_page(sg_vaddr);
 
-			/* Compute offset in (possibly tail) page */
+			/* Compute offset of sg_vaddr in (possibly tail) page */
 			page_offset = ((unsigned long)sg_vaddr &
 					(PAGE_SIZE - 1)) +
 				(page_address(page) - page_address(head_page));
-			/* page_offset only refers to the beginning of sgt[i];
-			 * but the buffer itself may have an internal offset.
+
+			/* Non-initial SGT entries should not have a buffer
+			 * offset.
 			 */
-			frag_off = qm_sg_entry_get_off(&sgt[i]) + page_offset;
-			frag_len = qm_sg_entry_get_len(&sgt[i]);
+			WARN_ON_ONCE(qm_sg_entry_get_off(&sgt[i]));
+
 			/* skb_add_rx_frag() does no checking on the page; if
 			 * we pass it a tail page, we'll end up with
-			 * bad page accounting and eventually with segafults.
+			 * bad page accounting and eventually with segfaults.
 			 */
-			skb_add_rx_frag(skb, i - 1, head_page, frag_off,
-					frag_len, dpaa_bp->size);
+			skb_add_rx_frag(skb, i - 1, head_page, page_offset,
+					qm_sg_entry_get_len(&sgt[i]),
+					dpaa_bp->size);
 		}
+
 		/* Update the pool count for the current {cpu x bpool} */
+		count_ptr = this_cpu_ptr(dpaa_bp->percpu_count);
 		(*count_ptr)--;
 
 		if (qm_sg_entry_is_final(&sgt[i]))
@@ -1767,35 +1908,34 @@ static struct sk_buff *sg_fd_to_skb(const struct dpaa_priv *priv,
 	WARN_ONCE(i == DPAA_SGT_MAX_ENTRIES, "No final bit on SGT\n");
 
 	/* free the SG table buffer */
-	skb_free_frag(vaddr);
+	free_pages((unsigned long)vaddr, 0);
 
 	return skb;
 
 free_buffers:
-	/* compensate sw bpool counter changes */
-	for (i--; i >= 0; i--) {
-		dpaa_bp = dpaa_bpid2pool(sgt[i].bpid);
-		if (dpaa_bp) {
-			count_ptr = this_cpu_ptr(dpaa_bp->percpu_count);
-			(*count_ptr)++;
-		}
-	}
 	/* free all the SG entries */
-	for (i = 0; i < DPAA_SGT_MAX_ENTRIES ; i++) {
-		sg_addr = qm_sg_addr(&sgt[i]);
+	for (j = 0; j < DPAA_SGT_MAX_ENTRIES ; j++) {
+		sg_addr = qm_sg_addr(&sgt[j]);
 		sg_vaddr = phys_to_virt(sg_addr);
-		skb_free_frag(sg_vaddr);
-		dpaa_bp = dpaa_bpid2pool(sgt[i].bpid);
-		if (dpaa_bp) {
-			count_ptr = this_cpu_ptr(dpaa_bp->percpu_count);
-			(*count_ptr)--;
+		/* all pages 0..i were unmaped */
+		if (j > i)
+			dma_unmap_page(priv->rx_dma_dev, qm_sg_addr(&sgt[j]),
+				       DPAA_BP_RAW_SIZE, DMA_FROM_DEVICE);
+		free_pages((unsigned long)sg_vaddr, 0);
+		/* counters 0..i-1 were decremented */
+		if (j >= i) {
+			dpaa_bp = dpaa_bpid2pool(sgt[j].bpid);
+			if (dpaa_bp) {
+				count_ptr = this_cpu_ptr(dpaa_bp->percpu_count);
+				(*count_ptr)--;
+			}
 		}
 
-		if (qm_sg_entry_is_final(&sgt[i]))
+		if (qm_sg_entry_is_final(&sgt[j]))
 			break;
 	}
 	/* free the SGT fragment */
-	skb_free_frag(vaddr);
+	free_pages((unsigned long)vaddr, 0);
 
 	return NULL;
 }
@@ -1805,10 +1945,9 @@ static int skb_to_contig_fd(struct dpaa_priv *priv,
 			    int *offset)
 {
 	struct net_device *net_dev = priv->net_dev;
-	struct device *dev = net_dev->dev.parent;
 	enum dma_data_direction dma_dir;
-	unsigned char *buffer_start;
-	struct sk_buff **skbh;
+	struct dpaa_eth_swbp *swbp;
+	unsigned char *buff_start;
 	dma_addr_t addr;
 	int err;
 
@@ -1816,11 +1955,11 @@ static int skb_to_contig_fd(struct dpaa_priv *priv,
 	 * available, so just use that for offset.
 	 */
 	fd->bpid = FSL_DPAA_BPID_INV;
-	buffer_start = skb->data - priv->tx_headroom;
+	buff_start = skb->data - priv->tx_headroom;
 	dma_dir = DMA_TO_DEVICE;
 
-	skbh = (struct sk_buff **)buffer_start;
-	*skbh = skb;
+	swbp = (struct dpaa_eth_swbp *)buff_start;
+	swbp->skb = skb;
 
 	/* Enable L3/L4 hardware checksum computation.
 	 *
@@ -1828,7 +1967,7 @@ static int skb_to_contig_fd(struct dpaa_priv *priv,
 	 * need to write into the skb.
 	 */
 	err = dpaa_enable_tx_csum(priv, skb, fd,
-				  ((char *)skbh) + DPAA_TX_PRIV_DATA_SIZE);
+				  buff_start + DPAA_TX_PRIV_DATA_SIZE);
 	if (unlikely(err < 0)) {
 		if (net_ratelimit())
 			netif_err(priv, tx_err, net_dev, "HW csum error: %d\n",
@@ -1841,9 +1980,9 @@ static int skb_to_contig_fd(struct dpaa_priv *priv,
 	fd->cmd |= cpu_to_be32(FM_FD_CMD_FCO);
 
 	/* Map the entire buffer size that may be seen by FMan, but no more */
-	addr = dma_map_single(dev, skbh,
-			      skb_tail_pointer(skb) - buffer_start, dma_dir);
-	if (unlikely(dma_mapping_error(dev, addr))) {
+	addr = dma_map_single(priv->tx_dma_dev, buff_start,
+			      priv->tx_headroom + skb->len, dma_dir);
+	if (unlikely(dma_mapping_error(priv->tx_dma_dev, addr))) {
 		if (net_ratelimit())
 			netif_err(priv, tx_err, net_dev, "dma_map_single() failed\n");
 		return -EINVAL;
@@ -1859,25 +1998,22 @@ static int skb_to_sg_fd(struct dpaa_priv *priv,
 	const enum dma_data_direction dma_dir = DMA_TO_DEVICE;
 	const int nr_frags = skb_shinfo(skb)->nr_frags;
 	struct net_device *net_dev = priv->net_dev;
-	struct device *dev = net_dev->dev.parent;
+	struct dpaa_eth_swbp *swbp;
 	struct qm_sg_entry *sgt;
-	struct sk_buff **skbh;
-	int i, j, err, sz;
-	void *buffer_start;
+	void *buff_start;
 	skb_frag_t *frag;
 	dma_addr_t addr;
 	size_t frag_len;
-	void *sgt_buf;
-
-	/* get a page frag to store the SGTable */
-	sz = SKB_DATA_ALIGN(priv->tx_headroom +
-		sizeof(struct qm_sg_entry) * (1 + nr_frags));
-	sgt_buf = netdev_alloc_frag(sz);
-	if (unlikely(!sgt_buf)) {
-		netdev_err(net_dev, "netdev_alloc_frag() failed for size %d\n",
-			   sz);
+	struct page *p;
+	int i, j, err;
+
+	/* get a page to store the SGTable */
+	p = dev_alloc_pages(0);
+	if (unlikely(!p)) {
+		netdev_err(net_dev, "dev_alloc_pages() failed\n");
 		return -ENOMEM;
 	}
+	buff_start = page_address(p);
 
 	/* Enable L3/L4 hardware checksum computation.
 	 *
@@ -1885,7 +2021,7 @@ static int skb_to_sg_fd(struct dpaa_priv *priv,
 	 * need to write into the skb.
 	 */
 	err = dpaa_enable_tx_csum(priv, skb, fd,
-				  sgt_buf + DPAA_TX_PRIV_DATA_SIZE);
+				  buff_start + DPAA_TX_PRIV_DATA_SIZE);
 	if (unlikely(err < 0)) {
 		if (net_ratelimit())
 			netif_err(priv, tx_err, net_dev, "HW csum error: %d\n",
@@ -1893,54 +2029,56 @@ static int skb_to_sg_fd(struct dpaa_priv *priv,
 		goto csum_failed;
 	}
 
-	sgt = (struct qm_sg_entry *)(sgt_buf + priv->tx_headroom);
-	qm_sg_entry_set_len(&sgt[0], skb_headlen(skb));
+	/* SGT[0] is used by the linear part */
+	sgt = (struct qm_sg_entry *)(buff_start + priv->tx_headroom);
+	frag_len = skb_headlen(skb);
+	qm_sg_entry_set_len(&sgt[0], frag_len);
 	sgt[0].bpid = FSL_DPAA_BPID_INV;
 	sgt[0].offset = 0;
-	addr = dma_map_single(dev, skb->data,
+	addr = dma_map_single(priv->tx_dma_dev, skb->data,
 			      skb_headlen(skb), dma_dir);
-	if (unlikely(dma_mapping_error(dev, addr))) {
-		dev_err(dev, "DMA mapping failed");
+	if (unlikely(dma_mapping_error(priv->tx_dma_dev, addr))) {
+		netdev_err(priv->net_dev, "DMA mapping failed\n");
 		err = -EINVAL;
 		goto sg0_map_failed;
 	}
 	qm_sg_entry_set64(&sgt[0], addr);
 
 	/* populate the rest of SGT entries */
-	frag = &skb_shinfo(skb)->frags[0];
-	frag_len = frag->size;
-	for (i = 1; i <= nr_frags; i++, frag++) {
+	for (i = 0; i < nr_frags; i++) {
+		frag = &skb_shinfo(skb)->frags[i];
+		frag_len = skb_frag_size(frag);
 		WARN_ON(!skb_frag_page(frag));
-		addr = skb_frag_dma_map(dev, frag, 0,
+		addr = skb_frag_dma_map(priv->tx_dma_dev, frag, 0,
 					frag_len, dma_dir);
-		if (unlikely(dma_mapping_error(dev, addr))) {
-			dev_err(dev, "DMA mapping failed");
+		if (unlikely(dma_mapping_error(priv->tx_dma_dev, addr))) {
+			netdev_err(priv->net_dev, "DMA mapping failed\n");
 			err = -EINVAL;
 			goto sg_map_failed;
 		}
 
-		qm_sg_entry_set_len(&sgt[i], frag_len);
-		sgt[i].bpid = FSL_DPAA_BPID_INV;
-		sgt[i].offset = 0;
+		qm_sg_entry_set_len(&sgt[i + 1], frag_len);
+		sgt[i + 1].bpid = FSL_DPAA_BPID_INV;
+		sgt[i + 1].offset = 0;
 
 		/* keep the offset in the address */
-		qm_sg_entry_set64(&sgt[i], addr);
-		frag_len = frag->size;
+		qm_sg_entry_set64(&sgt[i + 1], addr);
 	}
-	qm_sg_entry_set_f(&sgt[i - 1], frag_len);
 
+	/* Set the final bit in the last used entry of the SGT */
+	qm_sg_entry_set_f(&sgt[nr_frags], frag_len);
+
+	/* set fd offset to priv->tx_headroom */
 	qm_fd_set_sg(fd, priv->tx_headroom, skb->len);
 
 	/* DMA map the SGT page */
-	buffer_start = (void *)sgt - priv->tx_headroom;
-	skbh = (struct sk_buff **)buffer_start;
-	*skbh = skb;
-
-	addr = dma_map_single(dev, buffer_start, priv->tx_headroom +
-			      sizeof(struct qm_sg_entry) * (1 + nr_frags),
-			      dma_dir);
-	if (unlikely(dma_mapping_error(dev, addr))) {
-		dev_err(dev, "DMA mapping failed");
+	swbp = (struct dpaa_eth_swbp *)buff_start;
+	swbp->skb = skb;
+
+	addr = dma_map_page(priv->tx_dma_dev, p, 0,
+			    priv->tx_headroom + DPAA_SGT_SIZE, dma_dir);
+	if (unlikely(dma_mapping_error(priv->tx_dma_dev, addr))) {
+		netdev_err(priv->net_dev, "DMA mapping failed\n");
 		err = -EINVAL;
 		goto sgt_map_failed;
 	}
@@ -1954,11 +2092,11 @@ static int skb_to_sg_fd(struct dpaa_priv *priv,
 sgt_map_failed:
 sg_map_failed:
 	for (j = 0; j < i; j++)
-		dma_unmap_page(dev, qm_sg_addr(&sgt[j]),
+		dma_unmap_page(priv->tx_dma_dev, qm_sg_addr(&sgt[j]),
 			       qm_sg_entry_get_len(&sgt[j]), dma_dir);
 sg0_map_failed:
 csum_failed:
-	skb_free_frag(sgt_buf);
+	free_pages((unsigned long)buff_start, 0);
 
 	return err;
 }
@@ -1985,7 +2123,6 @@ static inline int dpaa_xmit(struct dpaa_priv *priv,
 	}
 
 	if (unlikely(err < 0)) {
-		percpu_stats->tx_errors++;
 		percpu_stats->tx_fifo_errors++;
 		return err;
 	}
@@ -1996,14 +2133,173 @@ static inline int dpaa_xmit(struct dpaa_priv *priv,
 	return 0;
 }
 
-static int dpaa_start_xmit(struct sk_buff *skb, struct net_device *net_dev)
+#ifdef CONFIG_DPAA_ERRATUM_A050385
+static int dpaa_a050385_wa_skb(struct net_device *net_dev, struct sk_buff **s)
+{
+	struct dpaa_priv *priv = netdev_priv(net_dev);
+	struct sk_buff *new_skb, *skb = *s;
+	unsigned char *start, i;
+
+	/* check linear buffer alignment */
+	if (!PTR_IS_ALIGNED(skb->data, DPAA_A050385_ALIGN))
+		goto workaround;
+
+	/* linear buffers just need to have an aligned start */
+	if (!skb_is_nonlinear(skb))
+		return 0;
+
+	/* linear data size for nonlinear skbs needs to be aligned */
+	if (!IS_ALIGNED(skb_headlen(skb), DPAA_A050385_ALIGN))
+		goto workaround;
+
+	for (i = 0; i < skb_shinfo(skb)->nr_frags; i++) {
+		skb_frag_t *frag = &skb_shinfo(skb)->frags[i];
+
+		/* all fragments need to have aligned start addresses */
+		if (!IS_ALIGNED(skb_frag_off(frag), DPAA_A050385_ALIGN))
+			goto workaround;
+
+		/* all but last fragment need to have aligned sizes */
+		if (!IS_ALIGNED(skb_frag_size(frag), DPAA_A050385_ALIGN) &&
+		    (i < skb_shinfo(skb)->nr_frags - 1))
+			goto workaround;
+	}
+
+	return 0;
+
+workaround:
+	/* copy all the skb content into a new linear buffer */
+	new_skb = netdev_alloc_skb(net_dev, skb->len + DPAA_A050385_ALIGN - 1 +
+						priv->tx_headroom);
+	if (!new_skb)
+		return -ENOMEM;
+
+	/* NET_SKB_PAD bytes already reserved, adding up to tx_headroom */
+	skb_reserve(new_skb, priv->tx_headroom - NET_SKB_PAD);
+
+	/* Workaround for DPAA_A050385 requires data start to be aligned */
+	start = PTR_ALIGN(new_skb->data, DPAA_A050385_ALIGN);
+	if (start - new_skb->data)
+		skb_reserve(new_skb, start - new_skb->data);
+
+	skb_put(new_skb, skb->len);
+	skb_copy_bits(skb, 0, new_skb->data, skb->len);
+	skb_copy_header(new_skb, skb);
+	new_skb->dev = skb->dev;
+
+	/* Copy relevant timestamp info from the old skb to the new */
+	if (priv->tx_tstamp) {
+		skb_shinfo(new_skb)->tx_flags = skb_shinfo(skb)->tx_flags;
+		skb_shinfo(new_skb)->hwtstamps = skb_shinfo(skb)->hwtstamps;
+		skb_shinfo(new_skb)->tskey = skb_shinfo(skb)->tskey;
+		if (skb->sk)
+			skb_set_owner_w(new_skb, skb->sk);
+	}
+
+	/* We move the headroom when we align it so we have to reset the
+	 * network and transport header offsets relative to the new data
+	 * pointer. The checksum offload relies on these offsets.
+	 */
+	skb_set_network_header(new_skb, skb_network_offset(skb));
+	skb_set_transport_header(new_skb, skb_transport_offset(skb));
+
+	dev_kfree_skb(skb);
+	*s = new_skb;
+
+	return 0;
+}
+
+static int dpaa_a050385_wa_xdpf(struct dpaa_priv *priv,
+				struct xdp_frame **init_xdpf)
+{
+	struct xdp_frame *new_xdpf, *xdpf = *init_xdpf;
+	void *new_buff, *aligned_data;
+	struct page *p;
+	u32 data_shift;
+	int headroom;
+
+	/* Check the data alignment and make sure the headroom is large
+	 * enough to store the xdpf backpointer. Use an aligned headroom
+	 * value.
+	 *
+	 * Due to alignment constraints, we give XDP access to the full 256
+	 * byte frame headroom. If the XDP program uses all of it, copy the
+	 * data to a new buffer and make room for storing the backpointer.
+	 */
+	if (PTR_IS_ALIGNED(xdpf->data, DPAA_FD_DATA_ALIGNMENT) &&
+	    xdpf->headroom >= priv->tx_headroom) {
+		xdpf->headroom = priv->tx_headroom;
+		return 0;
+	}
+
+	/* Try to move the data inside the buffer just enough to align it and
+	 * store the xdpf backpointer. If the available headroom isn't large
+	 * enough, resort to allocating a new buffer and copying the data.
+	 */
+	aligned_data = PTR_ALIGN_DOWN(xdpf->data, DPAA_FD_DATA_ALIGNMENT);
+	data_shift = xdpf->data - aligned_data;
+
+	/* The XDP frame's headroom needs to be large enough to accommodate
+	 * shifting the data as well as storing the xdpf backpointer.
+	 */
+	if (xdpf->headroom  >= data_shift + priv->tx_headroom) {
+		memmove(aligned_data, xdpf->data, xdpf->len);
+		xdpf->data = aligned_data;
+		xdpf->headroom = priv->tx_headroom;
+		return 0;
+	}
+
+	/* The new xdp_frame is stored in the new buffer. Reserve enough space
+	 * in the headroom for storing it along with the driver's private
+	 * info. The headroom needs to be aligned to DPAA_FD_DATA_ALIGNMENT to
+	 * guarantee the data's alignment in the buffer.
+	 */
+	headroom = ALIGN(sizeof(*new_xdpf) + priv->tx_headroom,
+			 DPAA_FD_DATA_ALIGNMENT);
+
+	/* Assure the extended headroom and data don't overflow the buffer,
+	 * while maintaining the mandatory tailroom.
+	 */
+	if (headroom + xdpf->len > DPAA_BP_RAW_SIZE -
+			SKB_DATA_ALIGN(sizeof(struct skb_shared_info)))
+		return -ENOMEM;
+
+	p = dev_alloc_pages(0);
+	if (unlikely(!p))
+		return -ENOMEM;
+
+	/* Copy the data to the new buffer at a properly aligned offset */
+	new_buff = page_address(p);
+	memcpy(new_buff + headroom, xdpf->data, xdpf->len);
+
+	/* Create an XDP frame around the new buffer in a similar fashion
+	 * to xdp_convert_buff_to_frame.
+	 */
+	new_xdpf = new_buff;
+	new_xdpf->data = new_buff + headroom;
+	new_xdpf->len = xdpf->len;
+	new_xdpf->headroom = priv->tx_headroom;
+	new_xdpf->frame_sz = DPAA_BP_RAW_SIZE;
+	new_xdpf->mem_type = MEM_TYPE_PAGE_ORDER0;
+
+	/* Release the initial buffer */
+	xdp_return_frame_rx_napi(xdpf);
+
+	*init_xdpf = new_xdpf;
+	return 0;
+}
+#endif
+
+static netdev_tx_t
+dpaa_start_xmit(struct sk_buff *skb, struct net_device *net_dev)
 {
 	const int queue_mapping = skb_get_queue_mapping(skb);
-	bool nonlinear = skb_is_nonlinear(skb);
 	struct rtnl_link_stats64 *percpu_stats;
 	struct dpaa_percpu_priv *percpu_priv;
+	struct netdev_queue *txq;
 	struct dpaa_priv *priv;
 	struct qm_fd fd;
+	bool nonlinear;
 	int offset = 0;
 	int err = 0;
 
@@ -2013,6 +2309,13 @@ static int dpaa_start_xmit(struct sk_buff *skb, struct net_device *net_dev)
 
 	qm_fd_clear_fd(&fd);
 
+	/* Packet data is always read as 32-bit words, so zero out any part of
+	 * the skb which might be sent if we have to pad the packet
+	 */
+	if (__skb_put_padto(skb, ETH_ZLEN, false))
+		goto enomem;
+
+	nonlinear = skb_is_nonlinear(skb);
 	if (!nonlinear) {
 		/* We're going to store the skb backpointer at the beginning
 		 * of the data buffer, so we need a privately owned skb
@@ -2029,28 +2332,50 @@ static int dpaa_start_xmit(struct sk_buff *skb, struct net_device *net_dev)
 	/* MAX_SKB_FRAGS is equal or larger than our dpaa_SGT_MAX_ENTRIES;
 	 * make sure we don't feed FMan with more fragments than it supports.
 	 */
-	if (nonlinear &&
-	    likely(skb_shinfo(skb)->nr_frags < DPAA_SGT_MAX_ENTRIES)) {
-		/* Just create a S/G fd based on the skb */
-		err = skb_to_sg_fd(priv, skb, &fd);
-		percpu_priv->tx_frag_skbuffs++;
-	} else {
+	if (unlikely(nonlinear &&
+		     (skb_shinfo(skb)->nr_frags >= DPAA_SGT_MAX_ENTRIES))) {
 		/* If the egress skb contains more fragments than we support
 		 * we have no choice but to linearize it ourselves.
 		 */
-		if (unlikely(nonlinear) && __skb_linearize(skb))
+		if (__skb_linearize(skb))
 			goto enomem;
 
-		/* Finally, create a contig FD from this skb */
+		nonlinear = skb_is_nonlinear(skb);
+	}
+
+#ifdef CONFIG_DPAA_ERRATUM_A050385
+	if (unlikely(fman_has_errata_a050385())) {
+		if (dpaa_a050385_wa_skb(net_dev, &skb))
+			goto enomem;
+		nonlinear = skb_is_nonlinear(skb);
+	}
+#endif
+
+	if (nonlinear) {
+		/* Just create a S/G fd based on the skb */
+		err = skb_to_sg_fd(priv, skb, &fd);
+		percpu_priv->tx_frag_skbuffs++;
+	} else {
+		/* Create a contig FD from this skb */
 		err = skb_to_contig_fd(priv, skb, &fd, &offset);
 	}
 	if (unlikely(err < 0))
 		goto skb_to_fd_failed;
 
+	txq = netdev_get_tx_queue(net_dev, queue_mapping);
+
+	/* LLTX requires to do our own update of trans_start */
+	txq_trans_cond_update(txq);
+
+	if (priv->tx_tstamp && skb_shinfo(skb)->tx_flags & SKBTX_HW_TSTAMP) {
+		fd.cmd |= cpu_to_be32(FM_FD_CMD_UPD);
+		skb_shinfo(skb)->tx_flags |= SKBTX_IN_PROGRESS;
+	}
+
 	if (likely(dpaa_xmit(priv, percpu_stats, queue_mapping, &fd) == 0))
 		return NETDEV_TX_OK;
 
-	dpaa_cleanup_tx_fd(priv, &fd);
+	dpaa_cleanup_tx_fd(priv, &fd, false);
 skb_to_fd_failed:
 enomem:
 	percpu_stats->tx_errors++;
@@ -2096,7 +2421,7 @@ static void dpaa_tx_error(struct net_device *net_dev,
 
 	percpu_priv->stats.tx_errors++;
 
-	skb = dpaa_cleanup_tx_fd(priv, fd);
+	skb = dpaa_cleanup_tx_fd(priv, fd, false);
 	dev_kfree_skb(skb);
 }
 
@@ -2104,13 +2429,18 @@ static int dpaa_eth_poll(struct napi_struct *napi, int budget)
 {
 	struct dpaa_napi_portal *np =
 			container_of(napi, struct dpaa_napi_portal, napi);
+	int cleaned;
+
+	np->xdp_act = 0;
+
+	cleaned = qman_p_poll_dqrr(np->p, budget);
 
-	int cleaned = qman_p_poll_dqrr(np->p, budget);
+	if (np->xdp_act & XDP_REDIRECT)
+		xdp_do_flush();
 
 	if (cleaned < budget) {
 		napi_complete_done(napi, cleaned);
 		qman_p_irqsource_add(np->p, QM_PIRQ_DQRI);
-
 	} else if (np->down) {
 		qman_p_irqsource_add(np->p, QM_PIRQ_DQRI);
 	}
@@ -2137,15 +2467,15 @@ static void dpaa_tx_conf(struct net_device *net_dev,
 
 	percpu_priv->tx_confirm++;
 
-	skb = dpaa_cleanup_tx_fd(priv, fd);
+	skb = dpaa_cleanup_tx_fd(priv, fd, true);
 
 	consume_skb(skb);
 }
 
 static inline int dpaa_eth_napi_schedule(struct dpaa_percpu_priv *percpu_priv,
-					 struct qman_portal *portal)
+					 struct qman_portal *portal, bool sched_napi)
 {
-	if (unlikely(in_irq() || !in_serving_softirq())) {
+	if (sched_napi) {
 		/* Disable QMan IRQ and invoke NAPI */
 		qman_p_irqsource_remove(portal, QM_PIRQ_DQRI);
 
@@ -2159,7 +2489,8 @@ static inline int dpaa_eth_napi_schedule(struct dpaa_percpu_priv *percpu_priv,
 
 static enum qman_cb_dqrr_result rx_error_dqrr(struct qman_portal *portal,
 					      struct qman_fq *fq,
-					      const struct qm_dqrr_entry *dq)
+					      const struct qm_dqrr_entry *dq,
+					      bool sched_napi)
 {
 	struct dpaa_fq *dpaa_fq = container_of(fq, struct dpaa_fq, fq_base);
 	struct dpaa_percpu_priv *percpu_priv;
@@ -2175,41 +2506,208 @@ static enum qman_cb_dqrr_result rx_error_dqrr(struct qman_portal *portal,
 
 	percpu_priv = this_cpu_ptr(priv->percpu_priv);
 
-	if (dpaa_eth_napi_schedule(percpu_priv, portal))
+	if (dpaa_eth_napi_schedule(percpu_priv, portal, sched_napi))
 		return qman_cb_dqrr_stop;
 
-	if (dpaa_eth_refill_bpools(priv))
-		/* Unable to refill the buffer pool due to insufficient
-		 * system memory. Just release the frame back into the pool,
-		 * otherwise we'll soon end up with an empty buffer pool.
-		 */
-		dpaa_fd_release(net_dev, &dq->fd);
-	else
-		dpaa_rx_error(net_dev, priv, percpu_priv, &dq->fd, fq->fqid);
+	dpaa_eth_refill_bpools(priv);
+	dpaa_rx_error(net_dev, priv, percpu_priv, &dq->fd, fq->fqid);
 
 	return qman_cb_dqrr_consume;
 }
 
+static int dpaa_xdp_xmit_frame(struct net_device *net_dev,
+			       struct xdp_frame *xdpf)
+{
+	struct dpaa_priv *priv = netdev_priv(net_dev);
+	struct rtnl_link_stats64 *percpu_stats;
+	struct dpaa_percpu_priv *percpu_priv;
+	struct dpaa_eth_swbp *swbp;
+	struct netdev_queue *txq;
+	void *buff_start;
+	struct qm_fd fd;
+	dma_addr_t addr;
+	int err;
+
+	percpu_priv = this_cpu_ptr(priv->percpu_priv);
+	percpu_stats = &percpu_priv->stats;
+
+#ifdef CONFIG_DPAA_ERRATUM_A050385
+	if (unlikely(fman_has_errata_a050385())) {
+		if (dpaa_a050385_wa_xdpf(priv, &xdpf)) {
+			err = -ENOMEM;
+			goto out_error;
+		}
+	}
+#endif
+
+	if (xdpf->headroom < DPAA_TX_PRIV_DATA_SIZE) {
+		err = -EINVAL;
+		goto out_error;
+	}
+
+	buff_start = xdpf->data - xdpf->headroom;
+
+	/* Leave empty the skb backpointer at the start of the buffer.
+	 * Save the XDP frame for easy cleanup on confirmation.
+	 */
+	swbp = (struct dpaa_eth_swbp *)buff_start;
+	swbp->skb = NULL;
+	swbp->xdpf = xdpf;
+
+	qm_fd_clear_fd(&fd);
+	fd.bpid = FSL_DPAA_BPID_INV;
+	fd.cmd |= cpu_to_be32(FM_FD_CMD_FCO);
+	qm_fd_set_contig(&fd, xdpf->headroom, xdpf->len);
+
+	addr = dma_map_single(priv->tx_dma_dev, buff_start,
+			      xdpf->headroom + xdpf->len,
+			      DMA_TO_DEVICE);
+	if (unlikely(dma_mapping_error(priv->tx_dma_dev, addr))) {
+		err = -EINVAL;
+		goto out_error;
+	}
+
+	qm_fd_addr_set64(&fd, addr);
+
+	/* Bump the trans_start */
+	txq = netdev_get_tx_queue(net_dev, smp_processor_id());
+	txq_trans_cond_update(txq);
+
+	err = dpaa_xmit(priv, percpu_stats, smp_processor_id(), &fd);
+	if (err) {
+		dma_unmap_single(priv->tx_dma_dev, addr,
+				 qm_fd_get_offset(&fd) + qm_fd_get_length(&fd),
+				 DMA_TO_DEVICE);
+		goto out_error;
+	}
+
+	return 0;
+
+out_error:
+	percpu_stats->tx_errors++;
+	return err;
+}
+
+static u32 dpaa_run_xdp(struct dpaa_priv *priv, struct qm_fd *fd, void *vaddr,
+			struct dpaa_fq *dpaa_fq, unsigned int *xdp_meta_len)
+{
+	ssize_t fd_off = qm_fd_get_offset(fd);
+	struct bpf_prog *xdp_prog;
+	struct xdp_frame *xdpf;
+	struct xdp_buff xdp;
+	u32 xdp_act;
+	int err;
+
+	xdp_prog = READ_ONCE(priv->xdp_prog);
+	if (!xdp_prog)
+		return XDP_PASS;
+
+	xdp_init_buff(&xdp, DPAA_BP_RAW_SIZE - DPAA_TX_PRIV_DATA_SIZE,
+		      &dpaa_fq->xdp_rxq);
+	xdp_prepare_buff(&xdp, vaddr + fd_off - XDP_PACKET_HEADROOM,
+			 XDP_PACKET_HEADROOM, qm_fd_get_length(fd), true);
+
+	/* We reserve a fixed headroom of 256 bytes under the erratum and we
+	 * offer it all to XDP programs to use. If no room is left for the
+	 * xdpf backpointer on TX, we will need to copy the data.
+	 * Disable metadata support since data realignments might be required
+	 * and the information can be lost.
+	 */
+#ifdef CONFIG_DPAA_ERRATUM_A050385
+	if (unlikely(fman_has_errata_a050385())) {
+		xdp_set_data_meta_invalid(&xdp);
+		xdp.data_hard_start = vaddr;
+		xdp.frame_sz = DPAA_BP_RAW_SIZE;
+	}
+#endif
+
+	xdp_act = bpf_prog_run_xdp(xdp_prog, &xdp);
+
+	/* Update the length and the offset of the FD */
+	qm_fd_set_contig(fd, xdp.data - vaddr, xdp.data_end - xdp.data);
+
+	switch (xdp_act) {
+	case XDP_PASS:
+#ifdef CONFIG_DPAA_ERRATUM_A050385
+		*xdp_meta_len = xdp_data_meta_unsupported(&xdp) ? 0 :
+				xdp.data - xdp.data_meta;
+#else
+		*xdp_meta_len = xdp.data - xdp.data_meta;
+#endif
+		break;
+	case XDP_TX:
+		/* We can access the full headroom when sending the frame
+		 * back out
+		 */
+		xdp.data_hard_start = vaddr;
+		xdp.frame_sz = DPAA_BP_RAW_SIZE;
+		xdpf = xdp_convert_buff_to_frame(&xdp);
+		if (unlikely(!xdpf)) {
+			free_pages((unsigned long)vaddr, 0);
+			break;
+		}
+
+		if (dpaa_xdp_xmit_frame(priv->net_dev, xdpf))
+			xdp_return_frame_rx_napi(xdpf);
+
+		break;
+	case XDP_REDIRECT:
+		/* Allow redirect to use the full headroom */
+		xdp.data_hard_start = vaddr;
+		xdp.frame_sz = DPAA_BP_RAW_SIZE;
+
+		err = xdp_do_redirect(priv->net_dev, &xdp, xdp_prog);
+		if (err) {
+			trace_xdp_exception(priv->net_dev, xdp_prog, xdp_act);
+			free_pages((unsigned long)vaddr, 0);
+		}
+		break;
+	default:
+		bpf_warn_invalid_xdp_action(priv->net_dev, xdp_prog, xdp_act);
+		fallthrough;
+	case XDP_ABORTED:
+		trace_xdp_exception(priv->net_dev, xdp_prog, xdp_act);
+		fallthrough;
+	case XDP_DROP:
+		/* Free the buffer */
+		free_pages((unsigned long)vaddr, 0);
+		break;
+	}
+
+	return xdp_act;
+}
+
 static enum qman_cb_dqrr_result rx_default_dqrr(struct qman_portal *portal,
 						struct qman_fq *fq,
-						const struct qm_dqrr_entry *dq)
+						const struct qm_dqrr_entry *dq,
+						bool sched_napi)
 {
+	bool ts_valid = false, hash_valid = false;
+	struct skb_shared_hwtstamps *shhwtstamps;
+	unsigned int skb_len, xdp_meta_len = 0;
 	struct rtnl_link_stats64 *percpu_stats;
 	struct dpaa_percpu_priv *percpu_priv;
 	const struct qm_fd *fd = &dq->fd;
 	dma_addr_t addr = qm_fd_addr(fd);
+	struct dpaa_napi_portal *np;
 	enum qm_fd_format fd_format;
 	struct net_device *net_dev;
-	u32 fd_status;
+	u32 fd_status, hash_offset;
+	struct qm_sg_entry *sgt;
 	struct dpaa_bp *dpaa_bp;
+	struct dpaa_fq *dpaa_fq;
 	struct dpaa_priv *priv;
-	unsigned int skb_len;
 	struct sk_buff *skb;
 	int *count_ptr;
+	u32 xdp_act;
+	void *vaddr;
+	u32 hash;
+	u64 ns;
 
+	dpaa_fq = container_of(fq, struct dpaa_fq, fq_base);
 	fd_status = be32_to_cpu(fd->status);
 	fd_format = qm_fd_get_format(fd);
-	net_dev = ((struct dpaa_fq *)fq)->net_dev;
+	net_dev = dpaa_fq->net_dev;
 	priv = netdev_priv(net_dev);
 	dpaa_bp = dpaa_bpid2pool(dq->fd.bpid);
 	if (!dpaa_bp)
@@ -2220,8 +2718,9 @@ static enum qman_cb_dqrr_result rx_default_dqrr(struct qman_portal *portal,
 
 	percpu_priv = this_cpu_ptr(priv->percpu_priv);
 	percpu_stats = &percpu_priv->stats;
+	np = &percpu_priv->np;
 
-	if (unlikely(dpaa_eth_napi_schedule(percpu_priv, portal)))
+	if (unlikely(dpaa_eth_napi_schedule(percpu_priv, portal, sched_napi)))
 		return qman_cb_dqrr_stop;
 
 	/* Make sure we didn't run out of buffers */
@@ -2244,16 +2743,13 @@ static enum qman_cb_dqrr_result rx_default_dqrr(struct qman_portal *portal,
 		return qman_cb_dqrr_consume;
 	}
 
-	dpaa_bp = dpaa_bpid2pool(fd->bpid);
-	if (!dpaa_bp)
-		return qman_cb_dqrr_consume;
-
-	dma_unmap_single(dpaa_bp->dev, addr, dpaa_bp->size, DMA_FROM_DEVICE);
+	dma_unmap_page(dpaa_bp->priv->rx_dma_dev, addr, DPAA_BP_RAW_SIZE,
+		       DMA_FROM_DEVICE);
 
 	/* prefetch the first 64 bytes of the frame or the SGT start */
-	prefetch(phys_to_virt(addr) + qm_fd_get_offset(fd));
+	vaddr = phys_to_virt(addr);
+	prefetch(vaddr + qm_fd_get_offset(fd));
 
-	fd_format = qm_fd_get_format(fd);
 	/* The only FD types that we may receive are contig and S/G */
 	WARN_ON((fd_format != qm_fd_contig) && (fd_format != qm_fd_sg));
 
@@ -2263,19 +2759,76 @@ static enum qman_cb_dqrr_result rx_default_dqrr(struct qman_portal *portal,
 	count_ptr = this_cpu_ptr(dpaa_bp->percpu_count);
 	(*count_ptr)--;
 
-	if (likely(fd_format == qm_fd_contig))
+	/* Extract the timestamp stored in the headroom before running XDP */
+	if (priv->rx_tstamp) {
+		if (!fman_port_get_tstamp(priv->mac_dev->port[RX], vaddr, &ns))
+			ts_valid = true;
+		else
+			WARN_ONCE(1, "fman_port_get_tstamp failed!\n");
+	}
+
+	/* Extract the hash stored in the headroom before running XDP */
+	if (net_dev->features & NETIF_F_RXHASH && priv->keygen_in_use &&
+	    !fman_port_get_hash_result_offset(priv->mac_dev->port[RX],
+					      &hash_offset)) {
+		hash = be32_to_cpu(*(__be32 *)(vaddr + hash_offset));
+		hash_valid = true;
+	}
+
+	if (likely(fd_format == qm_fd_contig)) {
+		xdp_act = dpaa_run_xdp(priv, (struct qm_fd *)fd, vaddr,
+				       dpaa_fq, &xdp_meta_len);
+		np->xdp_act |= xdp_act;
+		if (xdp_act != XDP_PASS) {
+			percpu_stats->rx_packets++;
+			percpu_stats->rx_bytes += qm_fd_get_length(fd);
+			return qman_cb_dqrr_consume;
+		}
 		skb = contig_fd_to_skb(priv, fd);
-	else
+	} else {
+		/* XDP doesn't support S/G frames. Return the fragments to the
+		 * buffer pool and release the SGT.
+		 */
+		if (READ_ONCE(priv->xdp_prog)) {
+			WARN_ONCE(1, "S/G frames not supported under XDP\n");
+			sgt = vaddr + qm_fd_get_offset(fd);
+			dpaa_release_sgt_members(sgt);
+			free_pages((unsigned long)vaddr, 0);
+			return qman_cb_dqrr_consume;
+		}
 		skb = sg_fd_to_skb(priv, fd);
+	}
 	if (!skb)
 		return qman_cb_dqrr_consume;
 
+	if (xdp_meta_len)
+		skb_metadata_set(skb, xdp_meta_len);
+
+	/* Set the previously extracted timestamp */
+	if (ts_valid) {
+		shhwtstamps = skb_hwtstamps(skb);
+		memset(shhwtstamps, 0, sizeof(*shhwtstamps));
+		shhwtstamps->hwtstamp = ns_to_ktime(ns);
+	}
+
 	skb->protocol = eth_type_trans(skb, net_dev);
 
+	/* Set the previously extracted hash */
+	if (hash_valid) {
+		enum pkt_hash_types type;
+
+		/* if L4 exists, it was used in the hash generation */
+		type = be32_to_cpu(fd->status) & FM_FD_STAT_L4CV ?
+			PKT_HASH_TYPE_L4 : PKT_HASH_TYPE_L3;
+		skb_set_hash(skb, hash, type);
+	}
+
 	skb_len = skb->len;
 
-	if (unlikely(netif_receive_skb(skb) == NET_RX_DROP))
+	if (unlikely(netif_receive_skb(skb) == NET_RX_DROP)) {
+		percpu_stats->rx_dropped++;
 		return qman_cb_dqrr_consume;
+	}
 
 	percpu_stats->rx_packets++;
 	percpu_stats->rx_bytes += skb_len;
@@ -2285,7 +2838,8 @@ static enum qman_cb_dqrr_result rx_default_dqrr(struct qman_portal *portal,
 
 static enum qman_cb_dqrr_result conf_error_dqrr(struct qman_portal *portal,
 						struct qman_fq *fq,
-						const struct qm_dqrr_entry *dq)
+						const struct qm_dqrr_entry *dq,
+						bool sched_napi)
 {
 	struct dpaa_percpu_priv *percpu_priv;
 	struct net_device *net_dev;
@@ -2296,7 +2850,7 @@ static enum qman_cb_dqrr_result conf_error_dqrr(struct qman_portal *portal,
 
 	percpu_priv = this_cpu_ptr(priv->percpu_priv);
 
-	if (dpaa_eth_napi_schedule(percpu_priv, portal))
+	if (dpaa_eth_napi_schedule(percpu_priv, portal, sched_napi))
 		return qman_cb_dqrr_stop;
 
 	dpaa_tx_error(net_dev, priv, percpu_priv, &dq->fd, fq->fqid);
@@ -2306,7 +2860,8 @@ static enum qman_cb_dqrr_result conf_error_dqrr(struct qman_portal *portal,
 
 static enum qman_cb_dqrr_result conf_dflt_dqrr(struct qman_portal *portal,
 					       struct qman_fq *fq,
-					       const struct qm_dqrr_entry *dq)
+					       const struct qm_dqrr_entry *dq,
+					       bool sched_napi)
 {
 	struct dpaa_percpu_priv *percpu_priv;
 	struct net_device *net_dev;
@@ -2320,7 +2875,7 @@ static enum qman_cb_dqrr_result conf_dflt_dqrr(struct qman_portal *portal,
 
 	percpu_priv = this_cpu_ptr(priv->percpu_priv);
 
-	if (dpaa_eth_napi_schedule(percpu_priv, portal))
+	if (dpaa_eth_napi_schedule(percpu_priv, portal, sched_napi))
 		return qman_cb_dqrr_stop;
 
 	dpaa_tx_conf(net_dev, priv, percpu_priv, &dq->fd, fq->fqid);
@@ -2346,7 +2901,7 @@ static void egress_ern(struct qman_portal *portal,
 	percpu_priv->stats.tx_fifo_errors++;
 	count_ern(percpu_priv, msg);
 
-	skb = dpaa_cleanup_tx_fd(priv, fd);
+	skb = dpaa_cleanup_tx_fd(priv, fd, false);
 	dev_kfree_skb_any(skb);
 }
 
@@ -2363,10 +2918,10 @@ static void dpaa_eth_napi_enable(struct dpaa_priv *priv)
 	struct dpaa_percpu_priv *percpu_priv;
 	int i;
 
-	for_each_possible_cpu(i) {
+	for_each_online_cpu(i) {
 		percpu_priv = per_cpu_ptr(priv->percpu_priv, i);
 
-		percpu_priv->np.down = 0;
+		percpu_priv->np.down = false;
 		napi_enable(&percpu_priv->np.napi);
 	}
 }
@@ -2376,10 +2931,10 @@ static void dpaa_eth_napi_disable(struct dpaa_priv *priv)
 	struct dpaa_percpu_priv *percpu_priv;
 	int i;
 
-	for_each_possible_cpu(i) {
+	for_each_online_cpu(i) {
 		percpu_priv = per_cpu_ptr(priv->percpu_priv, i);
 
-		percpu_priv->np.down = 1;
+		percpu_priv->np.down = true;
 		napi_disable(&percpu_priv->np.napi);
 	}
 }
@@ -2394,12 +2949,10 @@ static int dpaa_open(struct net_device *net_dev)
 	mac_dev = priv->mac_dev;
 	dpaa_eth_napi_enable(priv);
 
-	net_dev->phydev = mac_dev->init_phy(net_dev, priv->mac_dev);
-	if (!net_dev->phydev) {
-		netif_err(priv, ifup, net_dev, "init_phy() failed\n");
-		err = -ENODEV;
+	err = phylink_of_phy_connect(mac_dev->phylink,
+				     mac_dev->dev->of_node, 0);
+	if (err)
 		goto phy_init_failed;
-	}
 
 	for (i = 0; i < ARRAY_SIZE(mac_dev->port); i++) {
 		err = fman_port_enable(mac_dev->port[i]);
@@ -2407,11 +2960,12 @@ static int dpaa_open(struct net_device *net_dev)
 			goto mac_start_failed;
 	}
 
-	err = priv->mac_dev->start(mac_dev);
+	err = priv->mac_dev->enable(mac_dev->fman_mac);
 	if (err < 0) {
-		netif_err(priv, ifup, net_dev, "mac_dev->start() = %d\n", err);
+		netif_err(priv, ifup, net_dev, "mac_dev->enable() = %d\n", err);
 		goto mac_start_failed;
 	}
+	phylink_start(mac_dev->phylink);
 
 	netif_tx_start_all_queues(net_dev);
 
@@ -2420,6 +2974,7 @@ static int dpaa_open(struct net_device *net_dev)
 mac_start_failed:
 	for (i = 0; i < ARRAY_SIZE(mac_dev->port); i++)
 		fman_port_disable(mac_dev->port[i]);
+	phylink_disconnect_phy(mac_dev->phylink);
 
 phy_init_failed:
 	dpaa_eth_napi_disable(priv);
@@ -2440,11 +2995,152 @@ static int dpaa_eth_stop(struct net_device *net_dev)
 	return err;
 }
 
-static int dpaa_ioctl(struct net_device *net_dev, struct ifreq *rq, int cmd)
+static bool xdp_validate_mtu(struct dpaa_priv *priv, int mtu)
+{
+	int max_contig_data = priv->dpaa_bp->size - priv->rx_headroom;
+
+	/* We do not support S/G fragments when XDP is enabled.
+	 * Limit the MTU in relation to the buffer size.
+	 */
+	if (mtu + VLAN_ETH_HLEN + ETH_FCS_LEN > max_contig_data) {
+		dev_warn(priv->net_dev->dev.parent,
+			 "The maximum MTU for XDP is %d\n",
+			 max_contig_data - VLAN_ETH_HLEN - ETH_FCS_LEN);
+		return false;
+	}
+
+	return true;
+}
+
+static int dpaa_change_mtu(struct net_device *net_dev, int new_mtu)
+{
+	struct dpaa_priv *priv = netdev_priv(net_dev);
+
+	if (priv->xdp_prog && !xdp_validate_mtu(priv, new_mtu))
+		return -EINVAL;
+
+	WRITE_ONCE(net_dev->mtu, new_mtu);
+	return 0;
+}
+
+static int dpaa_setup_xdp(struct net_device *net_dev, struct netdev_bpf *bpf)
+{
+	struct dpaa_priv *priv = netdev_priv(net_dev);
+	struct bpf_prog *old_prog;
+	int err;
+	bool up;
+
+	/* S/G fragments are not supported in XDP-mode */
+	if (bpf->prog && !xdp_validate_mtu(priv, net_dev->mtu)) {
+		NL_SET_ERR_MSG_MOD(bpf->extack, "MTU too large for XDP");
+		return -EINVAL;
+	}
+
+	up = netif_running(net_dev);
+
+	if (up)
+		dpaa_eth_stop(net_dev);
+
+	old_prog = xchg(&priv->xdp_prog, bpf->prog);
+	if (old_prog)
+		bpf_prog_put(old_prog);
+
+	if (up) {
+		err = dpaa_open(net_dev);
+		if (err) {
+			NL_SET_ERR_MSG_MOD(bpf->extack, "dpaa_open() failed");
+			return err;
+		}
+	}
+
+	return 0;
+}
+
+static int dpaa_xdp(struct net_device *net_dev, struct netdev_bpf *xdp)
 {
-	if (!net_dev->phydev)
+	switch (xdp->command) {
+	case XDP_SETUP_PROG:
+		return dpaa_setup_xdp(net_dev, xdp);
+	default:
 		return -EINVAL;
-	return phy_mii_ioctl(net_dev->phydev, rq, cmd);
+	}
+}
+
+static int dpaa_xdp_xmit(struct net_device *net_dev, int n,
+			 struct xdp_frame **frames, u32 flags)
+{
+	struct xdp_frame *xdpf;
+	int i, nxmit = 0;
+
+	if (unlikely(flags & ~XDP_XMIT_FLAGS_MASK))
+		return -EINVAL;
+
+	if (!netif_running(net_dev))
+		return -ENETDOWN;
+
+	for (i = 0; i < n; i++) {
+		xdpf = frames[i];
+		if (dpaa_xdp_xmit_frame(net_dev, xdpf))
+			break;
+		nxmit++;
+	}
+
+	return nxmit;
+}
+
+static int dpaa_hwtstamp_get(struct net_device *dev,
+			     struct kernel_hwtstamp_config *config)
+{
+	struct dpaa_priv *priv = netdev_priv(dev);
+
+	config->tx_type = priv->tx_tstamp ? HWTSTAMP_TX_ON : HWTSTAMP_TX_OFF;
+	config->rx_filter = priv->rx_tstamp ? HWTSTAMP_FILTER_ALL :
+			    HWTSTAMP_FILTER_NONE;
+
+	return 0;
+}
+
+static int dpaa_hwtstamp_set(struct net_device *dev,
+			     struct kernel_hwtstamp_config *config,
+			     struct netlink_ext_ack *extack)
+{
+	struct dpaa_priv *priv = netdev_priv(dev);
+
+	switch (config->tx_type) {
+	case HWTSTAMP_TX_OFF:
+		/* Couldn't disable rx/tx timestamping separately.
+		 * Do nothing here.
+		 */
+		priv->tx_tstamp = false;
+		break;
+	case HWTSTAMP_TX_ON:
+		priv->mac_dev->set_tstamp(priv->mac_dev->fman_mac, true);
+		priv->tx_tstamp = true;
+		break;
+	default:
+		return -ERANGE;
+	}
+
+	if (config->rx_filter == HWTSTAMP_FILTER_NONE) {
+		/* Couldn't disable rx/tx timestamping separately.
+		 * Do nothing here.
+		 */
+		priv->rx_tstamp = false;
+	} else {
+		priv->mac_dev->set_tstamp(priv->mac_dev->fman_mac, true);
+		priv->rx_tstamp = true;
+		/* TS is set for all frame types, not only those requested */
+		config->rx_filter = HWTSTAMP_FILTER_ALL;
+	}
+
+	return 0;
+}
+
+static int dpaa_ioctl(struct net_device *net_dev, struct ifreq *rq, int cmd)
+{
+	struct dpaa_priv *priv = netdev_priv(net_dev);
+
+	return phylink_mii_ioctl(priv->mac_dev->phylink, rq, cmd);
 }
 
 static const struct net_device_ops dpaa_ops = {
@@ -2456,8 +3152,13 @@ static const struct net_device_ops dpaa_ops = {
 	.ndo_set_mac_address = dpaa_set_mac_address,
 	.ndo_validate_addr = eth_validate_addr,
 	.ndo_set_rx_mode = dpaa_set_rx_mode,
-	.ndo_do_ioctl = dpaa_ioctl,
+	.ndo_eth_ioctl = dpaa_ioctl,
 	.ndo_setup_tc = dpaa_setup_tc,
+	.ndo_change_mtu = dpaa_change_mtu,
+	.ndo_bpf = dpaa_xdp,
+	.ndo_xdp_xmit = dpaa_xdp_xmit,
+	.ndo_hwtstamp_get = dpaa_hwtstamp_get,
+	.ndo_hwtstamp_set = dpaa_hwtstamp_set,
 };
 
 static int dpaa_napi_add(struct net_device *net_dev)
@@ -2469,8 +3170,7 @@ static int dpaa_napi_add(struct net_device *net_dev)
 	for_each_possible_cpu(cpu) {
 		percpu_priv = per_cpu_ptr(priv->percpu_priv, cpu);
 
-		netif_napi_add(net_dev, &percpu_priv->np.napi,
-			       dpaa_eth_poll, NAPI_POLL_WEIGHT);
+		netif_napi_add(net_dev, &percpu_priv->np.napi, dpaa_eth_poll);
 	}
 
 	return 0;
@@ -2485,8 +3185,9 @@ static void dpaa_napi_del(struct net_device *net_dev)
 	for_each_possible_cpu(cpu) {
 		percpu_priv = per_cpu_ptr(priv->percpu_priv, cpu);
 
-		netif_napi_del(&percpu_priv->np.napi);
+		__netif_napi_del(&percpu_priv->np.napi);
 	}
+	synchronize_net();
 }
 
 static inline void dpaa_bp_free_pf(const struct dpaa_bp *bp,
@@ -2494,7 +3195,8 @@ static inline void dpaa_bp_free_pf(const struct dpaa_bp *bp,
 {
 	dma_addr_t addr = bm_buf_addr(bmb);
 
-	dma_unmap_single(bp->dev, addr, bp->size, DMA_FROM_DEVICE);
+	dma_unmap_page(bp->priv->rx_dma_dev, addr, DPAA_BP_RAW_SIZE,
+		       DMA_FROM_DEVICE);
 
 	skb_free_frag(phys_to_virt(addr));
 }
@@ -2510,6 +3212,9 @@ static struct dpaa_bp *dpaa_bp_alloc(struct device *dev)
 
 	dpaa_bp->bpid = FSL_DPAA_BPID_INV;
 	dpaa_bp->percpu_count = devm_alloc_percpu(dev, *dpaa_bp->percpu_count);
+	if (!dpaa_bp->percpu_count)
+		return ERR_PTR(-ENOMEM);
+
 	dpaa_bp->config_count = FSL_DPAA_ETH_MAX_BUF_COUNT;
 
 	dpaa_bp->seed_cb = dpaa_bp_seed;
@@ -2568,9 +3273,8 @@ out_error:
 	return err;
 }
 
-static const struct of_device_id dpaa_match[];
-
-static inline u16 dpaa_get_headroom(struct dpaa_buffer_layout *bl)
+static u16 dpaa_get_headroom(struct dpaa_buffer_layout *bl,
+			     enum port_type port)
 {
 	u16 headroom;
 
@@ -2584,39 +3288,75 @@ static inline u16 dpaa_get_headroom(struct dpaa_buffer_layout *bl)
 	 *
 	 * Also make sure the headroom is a multiple of data_align bytes
 	 */
-	headroom = (u16)(bl->priv_data_size + DPAA_PARSE_RESULTS_SIZE +
-		DPAA_TIME_STAMP_SIZE + DPAA_HASH_RESULTS_SIZE);
+	headroom = (u16)(bl[port].priv_data_size + DPAA_HWA_SIZE);
 
-	return DPAA_FD_DATA_ALIGNMENT ? ALIGN(headroom,
-					      DPAA_FD_DATA_ALIGNMENT) :
-					headroom;
+	if (port == RX) {
+#ifdef CONFIG_DPAA_ERRATUM_A050385
+		if (unlikely(fman_has_errata_a050385()))
+			headroom = XDP_PACKET_HEADROOM;
+#endif
+
+		return ALIGN(headroom, DPAA_FD_RX_DATA_ALIGNMENT);
+	} else {
+		return ALIGN(headroom, DPAA_FD_DATA_ALIGNMENT);
+	}
 }
 
 static int dpaa_eth_probe(struct platform_device *pdev)
 {
-	struct dpaa_bp *dpaa_bps[DPAA_BPS_NUM] = {NULL};
-	struct dpaa_percpu_priv *percpu_priv;
 	struct net_device *net_dev = NULL;
+	struct dpaa_bp *dpaa_bp = NULL;
 	struct dpaa_fq *dpaa_fq, *tmp;
 	struct dpaa_priv *priv = NULL;
 	struct fm_port_fqs port_fqs;
 	struct mac_device *mac_dev;
-	int err = 0, i, channel;
+	int err = 0, channel;
 	struct device *dev;
 
 	dev = &pdev->dev;
 
+	err = bman_is_probed();
+	if (!err)
+		return -EPROBE_DEFER;
+	if (err < 0) {
+		dev_err(dev, "failing probe due to bman probe error\n");
+		return -ENODEV;
+	}
+	err = qman_is_probed();
+	if (!err)
+		return -EPROBE_DEFER;
+	if (err < 0) {
+		dev_err(dev, "failing probe due to qman probe error\n");
+		return -ENODEV;
+	}
+	err = bman_portals_probed();
+	if (!err)
+		return -EPROBE_DEFER;
+	if (err < 0) {
+		dev_err(dev,
+			"failing probe due to bman portals probe error\n");
+		return -ENODEV;
+	}
+	err = qman_portals_probed();
+	if (!err)
+		return -EPROBE_DEFER;
+	if (err < 0) {
+		dev_err(dev,
+			"failing probe due to qman portals probe error\n");
+		return -ENODEV;
+	}
+
 	/* Allocate this early, so we can store relevant information in
 	 * the private area
 	 */
-	net_dev = alloc_etherdev_mq(sizeof(*priv), DPAA_ETH_TXQ_NUM);
+	net_dev = alloc_etherdev_mq(sizeof(*priv), dpaa_max_num_txqs());
 	if (!net_dev) {
 		dev_err(dev, "alloc_etherdev_mq() failed\n");
-		goto alloc_etherdev_mq_failed;
+		return -ENOMEM;
 	}
 
 	/* Do this here, so we can be verbose early */
-	SET_NETDEV_DEV(net_dev, dev);
+	SET_NETDEV_DEV(net_dev, dev->parent);
 	dev_set_drvdata(dev, net_dev);
 
 	priv = netdev_priv(net_dev);
@@ -2624,11 +3364,39 @@ static int dpaa_eth_probe(struct platform_device *pdev)
 
 	priv->msg_enable = netif_msg_init(debug, DPAA_MSG_DEFAULT);
 
+	priv->egress_fqs = devm_kcalloc(dev, dpaa_max_num_txqs(),
+					sizeof(*priv->egress_fqs),
+					GFP_KERNEL);
+	if (!priv->egress_fqs) {
+		err = -ENOMEM;
+		goto free_netdev;
+	}
+
+	priv->conf_fqs = devm_kcalloc(dev, dpaa_max_num_txqs(),
+				      sizeof(*priv->conf_fqs),
+				      GFP_KERNEL);
+	if (!priv->conf_fqs) {
+		err = -ENOMEM;
+		goto free_netdev;
+	}
+
 	mac_dev = dpaa_mac_dev_get(pdev);
 	if (IS_ERR(mac_dev)) {
-		dev_err(dev, "dpaa_mac_dev_get() failed\n");
+		netdev_err(net_dev, "dpaa_mac_dev_get() failed\n");
 		err = PTR_ERR(mac_dev);
-		goto mac_probe_failed;
+		goto free_netdev;
+	}
+
+	/* Devices used for DMA mapping */
+	priv->rx_dma_dev = fman_port_get_device(mac_dev->port[RX]);
+	priv->tx_dma_dev = fman_port_get_device(mac_dev->port[TX]);
+	err = dma_coerce_mask_and_coherent(priv->rx_dma_dev, DMA_BIT_MASK(40));
+	if (!err)
+		err = dma_coerce_mask_and_coherent(priv->tx_dma_dev,
+						   DMA_BIT_MASK(40));
+	if (err) {
+		netdev_err(net_dev, "dma_coerce_mask_and_coherent() failed\n");
+		goto free_netdev;
 	}
 
 	/* If fsl_fm_max_frm is set to a higher value than the all-common 1500,
@@ -2646,35 +3414,22 @@ static int dpaa_eth_probe(struct platform_device *pdev)
 	priv->buf_layout[RX].priv_data_size = DPAA_RX_PRIV_DATA_SIZE; /* Rx */
 	priv->buf_layout[TX].priv_data_size = DPAA_TX_PRIV_DATA_SIZE; /* Tx */
 
-	/* device used for DMA mapping */
-	arch_setup_dma_ops(dev, 0, 0, NULL, false);
-	err = dma_coerce_mask_and_coherent(dev, DMA_BIT_MASK(40));
-	if (err) {
-		dev_err(dev, "dma_coerce_mask_and_coherent() failed\n");
-		goto dev_mask_failed;
-	}
-
 	/* bp init */
-	for (i = 0; i < DPAA_BPS_NUM; i++) {
-		int err;
-
-		dpaa_bps[i] = dpaa_bp_alloc(dev);
-		if (IS_ERR(dpaa_bps[i]))
-			return PTR_ERR(dpaa_bps[i]);
-		/* the raw size of the buffers used for reception */
-		dpaa_bps[i]->raw_size = bpool_buffer_raw_size(i, DPAA_BPS_NUM);
-		/* avoid runtime computations by keeping the usable size here */
-		dpaa_bps[i]->size = dpaa_bp_size(dpaa_bps[i]->raw_size);
-		dpaa_bps[i]->dev = dev;
-
-		err = dpaa_bp_alloc_pool(dpaa_bps[i]);
-		if (err < 0) {
-			dpaa_bps_free(priv);
-			priv->dpaa_bps[i] = NULL;
-			goto bp_create_failed;
-		}
-		priv->dpaa_bps[i] = dpaa_bps[i];
-	}
+	dpaa_bp = dpaa_bp_alloc(dev);
+	if (IS_ERR(dpaa_bp)) {
+		err = PTR_ERR(dpaa_bp);
+		goto free_dpaa_bps;
+	}
+	/* the raw size of the buffers used for reception */
+	dpaa_bp->raw_size = DPAA_BP_RAW_SIZE;
+	/* avoid runtime computations by keeping the usable size here */
+	dpaa_bp->size = dpaa_bp_size(dpaa_bp->raw_size);
+	dpaa_bp->priv = priv;
+
+	err = dpaa_bp_alloc_pool(dpaa_bp);
+	if (err < 0)
+		goto free_dpaa_bps;
+	priv->dpaa_bp = dpaa_bp;
 
 	INIT_LIST_HEAD(&priv->dpaa_fq_list);
 
@@ -2683,7 +3438,7 @@ static int dpaa_eth_probe(struct platform_device *pdev)
 	err = dpaa_alloc_all_fqs(dev, &priv->dpaa_fq_list, &port_fqs);
 	if (err < 0) {
 		dev_err(dev, "dpaa_alloc_all_fqs() failed\n");
-		goto fq_probe_failed;
+		goto free_dpaa_bps;
 	}
 
 	priv->mac_dev = mac_dev;
@@ -2692,17 +3447,19 @@ static int dpaa_eth_probe(struct platform_device *pdev)
 	if (channel < 0) {
 		dev_err(dev, "dpaa_get_channel() failed\n");
 		err = channel;
-		goto get_channel_failed;
+		goto free_dpaa_bps;
 	}
 
 	priv->channel = (u16)channel;
 
-	/* Start a thread that will walk the CPUs with affine portals
+	/* Walk the CPUs with affine portals
 	 * and add this pool channel to each's dequeue mask.
 	 */
-	dpaa_eth_add_channel(priv->channel);
+	dpaa_eth_add_channel(priv->channel, &pdev->dev);
 
-	dpaa_fq_setup(priv, &dpaa_fq_cbs, priv->mac_dev->port[TX]);
+	err = dpaa_fq_setup(priv, &dpaa_fq_cbs, priv->mac_dev->port[TX]);
+	if (err)
+		goto free_dpaa_bps;
 
 	/* Create a congestion group for this netdev, with
 	 * dynamically-allocated CGR ID.
@@ -2712,53 +3469,53 @@ static int dpaa_eth_probe(struct platform_device *pdev)
 	err = dpaa_eth_cgr_init(priv);
 	if (err < 0) {
 		dev_err(dev, "Error initializing CGR\n");
-		goto tx_cgr_init_failed;
+		goto free_dpaa_bps;
 	}
 
 	err = dpaa_ingress_cgr_init(priv);
 	if (err < 0) {
 		dev_err(dev, "Error initializing ingress CGR\n");
-		goto rx_cgr_init_failed;
+		goto delete_egress_cgr;
 	}
 
 	/* Add the FQs to the interface, and make them active */
 	list_for_each_entry_safe(dpaa_fq, tmp, &priv->dpaa_fq_list, list) {
 		err = dpaa_fq_init(dpaa_fq, false);
 		if (err < 0)
-			goto fq_alloc_failed;
+			goto free_dpaa_fqs;
 	}
 
-	priv->tx_headroom = dpaa_get_headroom(&priv->buf_layout[TX]);
-	priv->rx_headroom = dpaa_get_headroom(&priv->buf_layout[RX]);
+	priv->tx_headroom = dpaa_get_headroom(priv->buf_layout, TX);
+	priv->rx_headroom = dpaa_get_headroom(priv->buf_layout, RX);
 
 	/* All real interfaces need their ports initialized */
-	err = dpaa_eth_init_ports(mac_dev, dpaa_bps, DPAA_BPS_NUM, &port_fqs,
+	err = dpaa_eth_init_ports(mac_dev, dpaa_bp, &port_fqs,
 				  &priv->buf_layout[0], dev);
 	if (err)
-		goto init_ports_failed;
+		goto free_dpaa_fqs;
+
+	/* Rx traffic distribution based on keygen hashing defaults to on */
+	priv->keygen_in_use = true;
 
 	priv->percpu_priv = devm_alloc_percpu(dev, *priv->percpu_priv);
 	if (!priv->percpu_priv) {
 		dev_err(dev, "devm_alloc_percpu() failed\n");
 		err = -ENOMEM;
-		goto alloc_percpu_failed;
-	}
-	for_each_possible_cpu(i) {
-		percpu_priv = per_cpu_ptr(priv->percpu_priv, i);
-		memset(percpu_priv, 0, sizeof(*percpu_priv));
+		goto free_dpaa_fqs;
 	}
 
 	priv->num_tc = 1;
-	netif_set_real_num_tx_queues(net_dev, priv->num_tc * DPAA_TC_TXQ_NUM);
+	netif_set_real_num_tx_queues(net_dev,
+				     priv->num_tc * dpaa_num_txqs_per_tc());
 
 	/* Initialize NAPI */
 	err = dpaa_napi_add(net_dev);
 	if (err < 0)
-		goto napi_add_failed;
+		goto delete_dpaa_napi;
 
 	err = dpaa_netdev_init(net_dev, &dpaa_ops, tx_timeout);
 	if (err < 0)
-		goto netdev_init_failed;
+		goto delete_dpaa_napi;
 
 	dpaa_eth_sysfs_init(&net_dev->dev);
 
@@ -2767,36 +3524,25 @@ static int dpaa_eth_probe(struct platform_device *pdev)
 
 	return 0;
 
-netdev_init_failed:
-napi_add_failed:
+delete_dpaa_napi:
 	dpaa_napi_del(net_dev);
-alloc_percpu_failed:
-init_ports_failed:
+free_dpaa_fqs:
 	dpaa_fq_free(dev, &priv->dpaa_fq_list);
-fq_alloc_failed:
 	qman_delete_cgr_safe(&priv->ingress_cgr);
 	qman_release_cgrid(priv->ingress_cgr.cgrid);
-rx_cgr_init_failed:
+delete_egress_cgr:
 	qman_delete_cgr_safe(&priv->cgr_data.cgr);
 	qman_release_cgrid(priv->cgr_data.cgr.cgrid);
-tx_cgr_init_failed:
-get_channel_failed:
+free_dpaa_bps:
 	dpaa_bps_free(priv);
-bp_create_failed:
-fq_probe_failed:
-dev_mask_failed:
-mac_probe_failed:
+free_netdev:
 	dev_set_drvdata(dev, NULL);
 	free_netdev(net_dev);
-alloc_etherdev_mq_failed:
-	for (i = 0; i < DPAA_BPS_NUM && dpaa_bps[i]; i++) {
-		if (atomic_read(&dpaa_bps[i]->refs) == 0)
-			devm_kfree(dev, dpaa_bps[i]);
-	}
+
 	return err;
 }
 
-static int dpaa_remove(struct platform_device *pdev)
+static void dpaa_remove(struct platform_device *pdev)
 {
 	struct net_device *net_dev;
 	struct dpaa_priv *priv;
@@ -2812,8 +3558,12 @@ static int dpaa_remove(struct platform_device *pdev)
 
 	dev_set_drvdata(dev, NULL);
 	unregister_netdev(net_dev);
+	phylink_destroy(priv->mac_dev->phylink);
 
 	err = dpaa_fq_free(dev, &priv->dpaa_fq_list);
+	if (err)
+		dev_err(dev, "Failed to free FQs on remove (%pE)\n",
+			ERR_PTR(err));
 
 	qman_delete_cgr_safe(&priv->ingress_cgr);
 	qman_release_cgrid(priv->ingress_cgr.cgrid);
@@ -2825,11 +3575,9 @@ static int dpaa_remove(struct platform_device *pdev)
 	dpaa_bps_free(priv);
 
 	free_netdev(net_dev);
-
-	return err;
 }
 
-static struct platform_device_id dpaa_devtype[] = {
+static const struct platform_device_id dpaa_devtype[] = {
 	{
 		.name = "dpaa-ethernet",
 		.driver_data = 0,
diff --git a/drivers/net/ethernet/freescale/dpaa/dpaa_eth.h b/drivers/net/ethernet/freescale/dpaa/dpaa_eth.h
index 9941a7866ebe..7ed659eb08de 100644
--- a/drivers/net/ethernet/freescale/dpaa/dpaa_eth.h
+++ b/drivers/net/ethernet/freescale/dpaa/dpaa_eth.h
@@ -1,37 +1,14 @@
-/* Copyright 2008 - 2016 Freescale Semiconductor Inc.
- *
- * Redistribution and use in source and binary forms, with or without
- * modification, are permitted provided that the following conditions are met:
- *     * Redistributions of source code must retain the above copyright
- *	 notice, this list of conditions and the following disclaimer.
- *     * Redistributions in binary form must reproduce the above copyright
- *	 notice, this list of conditions and the following disclaimer in the
- *	 documentation and/or other materials provided with the distribution.
- *     * Neither the name of Freescale Semiconductor nor the
- *	 names of its contributors may be used to endorse or promote products
- *	 derived from this software without specific prior written permission.
- *
- * ALTERNATIVELY, this software may be distributed under the terms of the
- * GNU General Public License ("GPL") as published by the Free Software
- * Foundation, either version 2 of that License or (at your option) any
- * later version.
- *
- * THIS SOFTWARE IS PROVIDED BY Freescale Semiconductor ``AS IS'' AND ANY
- * EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED
- * WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE
- * DISCLAIMED. IN NO EVENT SHALL Freescale Semiconductor BE LIABLE FOR ANY
- * DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES
- * (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES;
- * LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND
- * ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
- * (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS
- * SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
+/* SPDX-License-Identifier: BSD-3-Clause OR GPL-2.0-or-later */
+/*
+ * Copyright 2008 - 2016 Freescale Semiconductor Inc.
  */
 
 #ifndef __DPAA_H
 #define __DPAA_H
 
 #include <linux/netdevice.h>
+#include <linux/refcount.h>
+#include <net/xdp.h>
 #include <soc/fsl/qman.h>
 #include <soc/fsl/bman.h>
 
@@ -41,17 +18,12 @@
 
 /* Number of prioritised traffic classes */
 #define DPAA_TC_NUM		4
-/* Number of Tx queues per traffic class */
-#define DPAA_TC_TXQ_NUM		NR_CPUS
-/* Total number of Tx queues */
-#define DPAA_ETH_TXQ_NUM	(DPAA_TC_NUM * DPAA_TC_TXQ_NUM)
-
-#define DPAA_BPS_NUM 3 /* number of bpools per interface */
 
 /* More detailed FQ types - used for fine-grained WQ assignments */
 enum dpaa_fq_type {
 	FQ_TYPE_RX_DEFAULT = 1, /* Rx Default FQs */
 	FQ_TYPE_RX_ERROR,	/* Rx Error FQs */
+	FQ_TYPE_RX_PCD,		/* Rx Parse Classify Distribute FQs */
 	FQ_TYPE_TX,		/* "Real" Tx FQs */
 	FQ_TYPE_TX_CONFIRM,	/* Tx default Conf FQ (actually an Rx FQ) */
 	FQ_TYPE_TX_CONF_MQ,	/* Tx conf FQs (one for each Tx FQ) */
@@ -68,6 +40,7 @@ struct dpaa_fq {
 	u16 channel;
 	u8 wq;
 	enum dpaa_fq_type fq_type;
+	struct xdp_rxq_info xdp_rxq;
 };
 
 struct dpaa_fq_cbs {
@@ -78,9 +51,11 @@ struct dpaa_fq_cbs {
 	struct qman_fq egress_ern;
 };
 
+struct dpaa_priv;
+
 struct dpaa_bp {
-	/* device used in the DMA mapping operations */
-	struct device *dev;
+	/* used in the DMA mapping operations */
+	struct dpaa_priv *priv;
 	/* current number of buffers in the buffer pool alloted to each CPU */
 	int __percpu *percpu_count;
 	/* all buffers allocated for this pool have this raw size */
@@ -98,7 +73,7 @@ struct dpaa_bp {
 	int (*seed_cb)(struct dpaa_bp *);
 	/* bpool can be emptied before freeing by this cb */
 	void (*free_buf_cb)(const struct dpaa_bp *, struct bm_buffer *);
-	atomic_t refs;
+	refcount_t refs;
 };
 
 struct dpaa_rx_errors {
@@ -124,6 +99,7 @@ struct dpaa_napi_portal {
 	struct napi_struct napi;
 	struct qman_portal *p;
 	bool down;
+	int xdp_act;
 };
 
 struct dpaa_percpu_priv {
@@ -142,22 +118,34 @@ struct dpaa_buffer_layout {
 	u16 priv_data_size;
 };
 
+/* Information to be used on the Tx confirmation path. Stored just
+ * before the start of the transmit buffer. Maximum size allowed
+ * is DPAA_TX_PRIV_DATA_SIZE bytes.
+ */
+struct dpaa_eth_swbp {
+	struct sk_buff *skb;
+	struct xdp_frame *xdpf;
+};
+
 struct dpaa_priv {
 	struct dpaa_percpu_priv __percpu *percpu_priv;
-	struct dpaa_bp *dpaa_bps[DPAA_BPS_NUM];
+	struct dpaa_bp *dpaa_bp;
 	/* Store here the needed Tx headroom for convenience and speed
 	 * (even though it can be computed based on the fields of buf_layout)
 	 */
 	u16 tx_headroom;
 	struct net_device *net_dev;
 	struct mac_device *mac_dev;
-	struct qman_fq *egress_fqs[DPAA_ETH_TXQ_NUM];
-	struct qman_fq *conf_fqs[DPAA_ETH_TXQ_NUM];
+	struct device *rx_dma_dev;
+	struct device *tx_dma_dev;
+	struct qman_fq **egress_fqs;
+	struct qman_fq **conf_fqs;
 
 	u16 channel;
 	struct list_head dpaa_fq_list;
 
 	u8 num_tc;
+	bool keygen_in_use;
 	u32 msg_enable;	/* net_device message level */
 
 	struct {
@@ -180,6 +168,11 @@ struct dpaa_priv {
 
 	struct dpaa_buffer_layout buf_layout[2];
 	u16 rx_headroom;
+
+	bool tx_tstamp; /* Tx timestamping enabled */
+	bool rx_tstamp; /* Rx timestamping enabled */
+
+	struct bpf_prog *xdp_prog;
 };
 
 /* from dpaa_ethtool.c */
@@ -188,4 +181,16 @@ extern const struct ethtool_ops dpaa_ethtool_ops;
 /* from dpaa_eth_sysfs.c */
 void dpaa_eth_sysfs_remove(struct device *dev);
 void dpaa_eth_sysfs_init(struct device *dev);
+
+static inline size_t dpaa_num_txqs_per_tc(void)
+{
+	return num_possible_cpus();
+}
+
+/* Total number of Tx queues */
+static inline size_t dpaa_max_num_txqs(void)
+{
+	return DPAA_TC_NUM * dpaa_num_txqs_per_tc();
+}
+
 #endif	/* __DPAA_H */
diff --git a/drivers/net/ethernet/freescale/dpaa/dpaa_eth_sysfs.c b/drivers/net/ethernet/freescale/dpaa/dpaa_eth_sysfs.c
index ec75d1c6fa89..aad470e9caea 100644
--- a/drivers/net/ethernet/freescale/dpaa/dpaa_eth_sysfs.c
+++ b/drivers/net/ethernet/freescale/dpaa/dpaa_eth_sysfs.c
@@ -1,32 +1,6 @@
-/* Copyright 2008-2016 Freescale Semiconductor Inc.
- *
- * Redistribution and use in source and binary forms, with or without
- * modification, are permitted provided that the following conditions are met:
- *     * Redistributions of source code must retain the above copyright
- *	 notice, this list of conditions and the following disclaimer.
- *     * Redistributions in binary form must reproduce the above copyright
- *	 notice, this list of conditions and the following disclaimer in the
- *	 documentation and/or other materials provided with the distribution.
- *     * Neither the name of Freescale Semiconductor nor the
- *	 names of its contributors may be used to endorse or promote products
- *	 derived from this software without specific prior written permission.
- *
- *
- * ALTERNATIVELY, this software may be distributed under the terms of the
- * GNU General Public License ("GPL") as published by the Free Software
- * Foundation, either version 2 of that License or (at your option) any
- * later version.
- *
- * THIS SOFTWARE IS PROVIDED BY Freescale Semiconductor ``AS IS'' AND ANY
- * EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED
- * WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE
- * DISCLAIMED. IN NO EVENT SHALL Freescale Semiconductor BE LIABLE FOR ANY
- * DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES
- * (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES;
- * LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND
- * ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
- * (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS
- * SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
+// SPDX-License-Identifier: BSD-3-Clause OR GPL-2.0-or-later
+/*
+ * Copyright 2008 - 2016 Freescale Semiconductor Inc.
  */
 
 #include <linux/init.h>
@@ -61,7 +35,6 @@ static ssize_t dpaa_eth_show_fqids(struct device *dev,
 	u32 last_fqid = 0;
 	ssize_t bytes = 0;
 	char *str;
-	int i = 0;
 
 	list_for_each_entry_safe(fq, tmp, &priv->dpaa_fq_list, list) {
 		switch (fq->fq_type) {
@@ -71,6 +44,9 @@ static ssize_t dpaa_eth_show_fqids(struct device *dev,
 		case FQ_TYPE_RX_ERROR:
 			str = "Rx error";
 			break;
+		case FQ_TYPE_RX_PCD:
+			str = "Rx PCD";
+			break;
 		case FQ_TYPE_TX_CONFIRM:
 			str = "Tx default confirmation";
 			break;
@@ -108,7 +84,6 @@ static ssize_t dpaa_eth_show_fqids(struct device *dev,
 
 		prev = fq;
 		prevstr = str;
-		i++;
 	}
 
 	if (prev) {
@@ -128,11 +103,9 @@ static ssize_t dpaa_eth_show_bpids(struct device *dev,
 {
 	struct dpaa_priv *priv = netdev_priv(to_net_dev(dev));
 	ssize_t bytes = 0;
-	int i = 0;
 
-	for (i = 0; i < DPAA_BPS_NUM; i++)
-		bytes += snprintf(buf + bytes, PAGE_SIZE - bytes, "%u\n",
-				  priv->dpaa_bps[i]->bpid);
+	bytes += snprintf(buf + bytes, PAGE_SIZE - bytes, "%u\n",
+				  priv->dpaa_bp->bpid);
 
 	return bytes;
 }
diff --git a/drivers/net/ethernet/freescale/dpaa/dpaa_eth_trace.h b/drivers/net/ethernet/freescale/dpaa/dpaa_eth_trace.h
index 409c1dc39430..9e1d44ae92cc 100644
--- a/drivers/net/ethernet/freescale/dpaa/dpaa_eth_trace.h
+++ b/drivers/net/ethernet/freescale/dpaa/dpaa_eth_trace.h
@@ -1,32 +1,6 @@
-/* Copyright 2013-2015 Freescale Semiconductor Inc.
- *
- * Redistribution and use in source and binary forms, with or without
- * modification, are permitted provided that the following conditions are met:
- *     * Redistributions of source code must retain the above copyright
- *	 notice, this list of conditions and the following disclaimer.
- *     * Redistributions in binary form must reproduce the above copyright
- *	 notice, this list of conditions and the following disclaimer in the
- *	 documentation and/or other materials provided with the distribution.
- *     * Neither the name of Freescale Semiconductor nor the
- *	 names of its contributors may be used to endorse or promote products
- *	 derived from this software without specific prior written permission.
- *
- *
- * ALTERNATIVELY, this software may be distributed under the terms of the
- * GNU General Public License ("GPL") as published by the Free Software
- * Foundation, either version 2 of that License or (at your option) any
- * later version.
- *
- * THIS SOFTWARE IS PROVIDED BY Freescale Semiconductor ``AS IS'' AND ANY
- * EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED
- * WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE
- * DISCLAIMED. IN NO EVENT SHALL Freescale Semiconductor BE LIABLE FOR ANY
- * DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES
- * (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES;
- * LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND
- * ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
- * (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS
- * SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
+/* SPDX-License-Identifier: BSD-3-Clause OR GPL-2.0-or-later */
+/*
+ * Copyright 2013-2015 Freescale Semiconductor Inc.
  */
 
 #undef TRACE_SYSTEM
@@ -82,8 +56,8 @@ DECLARE_EVENT_CLASS(dpaa_eth_fd,
 		__entry->fd_format = qm_fd_get_format(fd);
 		__entry->fd_offset = qm_fd_get_offset(fd);
 		__entry->fd_length = qm_fd_get_length(fd);
-		__entry->fd_status = fd->status;
-		__assign_str(name, netdev->name);
+		__entry->fd_status = __be32_to_cpu(fd->status);
+		__assign_str(name);
 	),
 
 	/* This is what gets printed when the trace event is triggered */
diff --git a/drivers/net/ethernet/freescale/dpaa/dpaa_ethtool.c b/drivers/net/ethernet/freescale/dpaa/dpaa_ethtool.c
index 15571e251fb9..d09e456f14c0 100644
--- a/drivers/net/ethernet/freescale/dpaa/dpaa_ethtool.c
+++ b/drivers/net/ethernet/freescale/dpaa/dpaa_ethtool.c
@@ -1,37 +1,16 @@
-/* Copyright 2008-2016 Freescale Semiconductor, Inc.
- *
- * Redistribution and use in source and binary forms, with or without
- * modification, are permitted provided that the following conditions are met:
- *     * Redistributions of source code must retain the above copyright
- *	 notice, this list of conditions and the following disclaimer.
- *     * Redistributions in binary form must reproduce the above copyright
- *	 notice, this list of conditions and the following disclaimer in the
- *	 documentation and/or other materials provided with the distribution.
- *     * Neither the name of Freescale Semiconductor nor the
- *	 names of its contributors may be used to endorse or promote products
- *	 derived from this software without specific prior written permission.
- *
- *
- * ALTERNATIVELY, this software may be distributed under the terms of the
- * GNU General Public License ("GPL") as published by the Free Software
- * Foundation, either version 2 of that License or (at your option) any
- * later version.
- *
- * THIS SOFTWARE IS PROVIDED BY Freescale Semiconductor ``AS IS'' AND ANY
- * EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED
- * WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE
- * DISCLAIMED. IN NO EVENT SHALL Freescale Semiconductor BE LIABLE FOR ANY
- * DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES
- * (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES;
- * LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND
- * ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
- * (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS
- * SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
+// SPDX-License-Identifier: BSD-3-Clause OR GPL-2.0-or-later
+/*
+ * Copyright 2008 - 2016 Freescale Semiconductor Inc.
  */
 
 #define pr_fmt(fmt) KBUILD_MODNAME ": " fmt
 
 #include <linux/string.h>
+#include <linux/of.h>
+#include <linux/of_platform.h>
+#include <linux/platform_device.h>
+#include <linux/net_tstamp.h>
+#include <linux/fsl/ptp_qoriq.h>
 
 #include "dpaa_eth.h"
 #include "mac.h"
@@ -44,6 +23,8 @@ static const char dpaa_stats_percpu[][ETH_GSTRING_LEN] = {
 	"tx S/G",
 	"tx error",
 	"rx error",
+	"rx dropped",
+	"tx dropped",
 };
 
 static char dpaa_stats_global[][ETH_GSTRING_LEN] = {
@@ -75,52 +56,27 @@ static char dpaa_stats_global[][ETH_GSTRING_LEN] = {
 static int dpaa_get_link_ksettings(struct net_device *net_dev,
 				   struct ethtool_link_ksettings *cmd)
 {
-	int err;
+	struct dpaa_priv *priv = netdev_priv(net_dev);
+	struct mac_device *mac_dev = priv->mac_dev;
 
-	if (!net_dev->phydev) {
-		netdev_dbg(net_dev, "phy device not initialized\n");
-		return 0;
-	}
-
-	err = phy_ethtool_ksettings_get(net_dev->phydev, cmd);
-
-	return err;
+	return phylink_ethtool_ksettings_get(mac_dev->phylink, cmd);
 }
 
 static int dpaa_set_link_ksettings(struct net_device *net_dev,
 				   const struct ethtool_link_ksettings *cmd)
 {
-	int err;
-
-	if (!net_dev->phydev) {
-		netdev_err(net_dev, "phy device not initialized\n");
-		return -ENODEV;
-	}
-
-	err = phy_ethtool_ksettings_set(net_dev->phydev, cmd);
-	if (err < 0)
-		netdev_err(net_dev, "phy_ethtool_ksettings_set() = %d\n", err);
+	struct dpaa_priv *priv = netdev_priv(net_dev);
+	struct mac_device *mac_dev = priv->mac_dev;
 
-	return err;
+	return phylink_ethtool_ksettings_set(mac_dev->phylink, cmd);
 }
 
 static void dpaa_get_drvinfo(struct net_device *net_dev,
 			     struct ethtool_drvinfo *drvinfo)
 {
-	int len;
-
-	strlcpy(drvinfo->driver, KBUILD_MODNAME,
+	strscpy(drvinfo->driver, KBUILD_MODNAME,
 		sizeof(drvinfo->driver));
-	len = snprintf(drvinfo->version, sizeof(drvinfo->version),
-		       "%X", 0);
-	len = snprintf(drvinfo->fw_version, sizeof(drvinfo->fw_version),
-		       "%X", 0);
-
-	if (len >= sizeof(drvinfo->fw_version)) {
-		/* Truncated output */
-		netdev_notice(net_dev, "snprintf() = %d\n", len);
-	}
-	strlcpy(drvinfo->bus_info, dev_name(net_dev->dev.parent->parent),
+	strscpy(drvinfo->bus_info, dev_name(net_dev->dev.parent->parent),
 		sizeof(drvinfo->bus_info));
 }
 
@@ -137,108 +93,28 @@ static void dpaa_set_msglevel(struct net_device *net_dev,
 
 static int dpaa_nway_reset(struct net_device *net_dev)
 {
-	int err;
-
-	if (!net_dev->phydev) {
-		netdev_err(net_dev, "phy device not initialized\n");
-		return -ENODEV;
-	}
-
-	err = 0;
-	if (net_dev->phydev->autoneg) {
-		err = phy_start_aneg(net_dev->phydev);
-		if (err < 0)
-			netdev_err(net_dev, "phy_start_aneg() = %d\n",
-				   err);
-	}
+	struct dpaa_priv *priv = netdev_priv(net_dev);
+	struct mac_device *mac_dev = priv->mac_dev;
 
-	return err;
+	return phylink_ethtool_nway_reset(mac_dev->phylink);
 }
 
 static void dpaa_get_pauseparam(struct net_device *net_dev,
 				struct ethtool_pauseparam *epause)
 {
-	struct mac_device *mac_dev;
-	struct dpaa_priv *priv;
-
-	priv = netdev_priv(net_dev);
-	mac_dev = priv->mac_dev;
+	struct dpaa_priv *priv = netdev_priv(net_dev);
+	struct mac_device *mac_dev = priv->mac_dev;
 
-	if (!net_dev->phydev) {
-		netdev_err(net_dev, "phy device not initialized\n");
-		return;
-	}
-
-	epause->autoneg = mac_dev->autoneg_pause;
-	epause->rx_pause = mac_dev->rx_pause_active;
-	epause->tx_pause = mac_dev->tx_pause_active;
+	phylink_ethtool_get_pauseparam(mac_dev->phylink, epause);
 }
 
 static int dpaa_set_pauseparam(struct net_device *net_dev,
 			       struct ethtool_pauseparam *epause)
 {
-	struct mac_device *mac_dev;
-	struct phy_device *phydev;
-	bool rx_pause, tx_pause;
-	struct dpaa_priv *priv;
-	u32 newadv, oldadv;
-	int err;
-
-	priv = netdev_priv(net_dev);
-	mac_dev = priv->mac_dev;
-
-	phydev = net_dev->phydev;
-	if (!phydev) {
-		netdev_err(net_dev, "phy device not initialized\n");
-		return -ENODEV;
-	}
-
-	if (!(phydev->supported & SUPPORTED_Pause) ||
-	    (!(phydev->supported & SUPPORTED_Asym_Pause) &&
-	    (epause->rx_pause != epause->tx_pause)))
-		return -EINVAL;
-
-	/* The MAC should know how to handle PAUSE frame autonegotiation before
-	 * adjust_link is triggered by a forced renegotiation of sym/asym PAUSE
-	 * settings.
-	 */
-	mac_dev->autoneg_pause = !!epause->autoneg;
-	mac_dev->rx_pause_req = !!epause->rx_pause;
-	mac_dev->tx_pause_req = !!epause->tx_pause;
-
-	/* Determine the sym/asym advertised PAUSE capabilities from the desired
-	 * rx/tx pause settings.
-	 */
-	newadv = 0;
-	if (epause->rx_pause)
-		newadv = ADVERTISED_Pause | ADVERTISED_Asym_Pause;
-	if (epause->tx_pause)
-		newadv |= ADVERTISED_Asym_Pause;
-
-	oldadv = phydev->advertising &
-			(ADVERTISED_Pause | ADVERTISED_Asym_Pause);
-
-	/* If there are differences between the old and the new advertised
-	 * values, restart PHY autonegotiation and advertise the new values.
-	 */
-	if (oldadv != newadv) {
-		phydev->advertising &= ~(ADVERTISED_Pause
-				| ADVERTISED_Asym_Pause);
-		phydev->advertising |= newadv;
-		if (phydev->autoneg) {
-			err = phy_start_aneg(phydev);
-			if (err < 0)
-				netdev_err(net_dev, "phy_start_aneg() = %d\n",
-					   err);
-		}
-	}
-
-	fman_get_pause_cfg(mac_dev, &rx_pause, &tx_pause);
-	err = fman_set_mac_active_pause(mac_dev, rx_pause, tx_pause);
-	if (err < 0)
-		netdev_err(net_dev, "set_mac_active_pause() = %d\n", err);
+	struct dpaa_priv *priv = netdev_priv(net_dev);
+	struct mac_device *mac_dev = priv->mac_dev;
 
-	return err;
+	return phylink_ethtool_set_pauseparam(mac_dev->phylink, epause);
 }
 
 static int dpaa_get_sset_count(struct net_device *net_dev, int type)
@@ -246,7 +122,7 @@ static int dpaa_get_sset_count(struct net_device *net_dev, int type)
 	unsigned int total_stats, num_stats;
 
 	num_stats   = num_online_cpus() + 1;
-	total_stats = num_stats * (DPAA_STATS_PERCPU_LEN + DPAA_BPS_NUM) +
+	total_stats = num_stats * (DPAA_STATS_PERCPU_LEN + 1) +
 			DPAA_STATS_GLOBAL_LEN;
 
 	switch (type) {
@@ -258,10 +134,10 @@ static int dpaa_get_sset_count(struct net_device *net_dev, int type)
 }
 
 static void copy_stats(struct dpaa_percpu_priv *percpu_priv, int num_cpus,
-		       int crr_cpu, u64 *bp_count, u64 *data)
+		       int crr_cpu, u64 bp_count, u64 *data)
 {
 	int num_values = num_cpus + 1;
-	int crr = 0, j;
+	int crr = 0;
 
 	/* update current CPU's stats and also add them to the total values */
 	data[crr * num_values + crr_cpu] = percpu_priv->in_interrupt;
@@ -285,23 +161,27 @@ static void copy_stats(struct dpaa_percpu_priv *percpu_priv, int num_cpus,
 	data[crr * num_values + crr_cpu] = percpu_priv->stats.rx_errors;
 	data[crr++ * num_values + num_cpus] += percpu_priv->stats.rx_errors;
 
-	for (j = 0; j < DPAA_BPS_NUM; j++) {
-		data[crr * num_values + crr_cpu] = bp_count[j];
-		data[crr++ * num_values + num_cpus] += bp_count[j];
-	}
+	data[crr * num_values + crr_cpu] = percpu_priv->stats.rx_dropped;
+	data[crr++ * num_values + num_cpus] += percpu_priv->stats.rx_dropped;
+
+	data[crr * num_values + crr_cpu] = percpu_priv->stats.tx_dropped;
+	data[crr++ * num_values + num_cpus] += percpu_priv->stats.tx_dropped;
+
+	data[crr * num_values + crr_cpu] = bp_count;
+	data[crr++ * num_values + num_cpus] += bp_count;
 }
 
 static void dpaa_get_ethtool_stats(struct net_device *net_dev,
 				   struct ethtool_stats *stats, u64 *data)
 {
-	u64 bp_count[DPAA_BPS_NUM], cg_time, cg_num;
 	struct dpaa_percpu_priv *percpu_priv;
 	struct dpaa_rx_errors rx_errors;
 	unsigned int num_cpus, offset;
+	u64 bp_count, cg_time, cg_num;
 	struct dpaa_ern_cnt ern_cnt;
 	struct dpaa_bp *dpaa_bp;
 	struct dpaa_priv *priv;
-	int total_stats, i, j;
+	int total_stats, i;
 	bool cg_status;
 
 	total_stats = dpaa_get_sset_count(net_dev, ETH_SS_STATS);
@@ -315,12 +195,10 @@ static void dpaa_get_ethtool_stats(struct net_device *net_dev,
 
 	for_each_online_cpu(i) {
 		percpu_priv = per_cpu_ptr(priv->percpu_priv, i);
-		for (j = 0; j < DPAA_BPS_NUM; j++) {
-			dpaa_bp = priv->dpaa_bps[j];
-			if (!dpaa_bp->percpu_count)
-				continue;
-			bp_count[j] = *(per_cpu_ptr(dpaa_bp->percpu_count, i));
-		}
+		dpaa_bp = priv->dpaa_bp;
+		if (!dpaa_bp->percpu_count)
+			continue;
+		bp_count = *(per_cpu_ptr(dpaa_bp->percpu_count, i));
 		rx_errors.dme += percpu_priv->rx_errors.dme;
 		rx_errors.fpe += percpu_priv->rx_errors.fpe;
 		rx_errors.fse += percpu_priv->rx_errors.fse;
@@ -338,7 +216,7 @@ static void dpaa_get_ethtool_stats(struct net_device *net_dev,
 		copy_stats(percpu_priv, num_cpus, i, bp_count, data);
 	}
 
-	offset = (num_cpus + 1) * (DPAA_STATS_PERCPU_LEN + DPAA_BPS_NUM);
+	offset = (num_cpus + 1) * (DPAA_STATS_PERCPU_LEN + 1);
 	memcpy(data + offset, &rx_errors, sizeof(struct dpaa_rx_errors));
 
 	offset += sizeof(struct dpaa_rx_errors) / sizeof(u64);
@@ -346,7 +224,7 @@ static void dpaa_get_ethtool_stats(struct net_device *net_dev,
 
 	/* gather congestion related counters */
 	cg_num    = 0;
-	cg_status = 0;
+	cg_status = false;
 	cg_time   = jiffies_to_msecs(priv->cgr_data.congested_jiffies);
 	if (qman_query_cgr_congested(&priv->cgr_data.cgr, &cg_status) == 0) {
 		cg_num    = priv->cgr_data.cgr_congested_count;
@@ -365,43 +243,233 @@ static void dpaa_get_ethtool_stats(struct net_device *net_dev,
 static void dpaa_get_strings(struct net_device *net_dev, u32 stringset,
 			     u8 *data)
 {
-	unsigned int i, j, num_cpus, size;
-	char string_cpu[ETH_GSTRING_LEN];
-	u8 *strings;
+	unsigned int i, j, num_cpus;
 
-	memset(string_cpu, 0, sizeof(string_cpu));
-	strings   = data;
-	num_cpus  = num_online_cpus();
-	size      = DPAA_STATS_GLOBAL_LEN * ETH_GSTRING_LEN;
+	num_cpus = num_online_cpus();
 
 	for (i = 0; i < DPAA_STATS_PERCPU_LEN; i++) {
-		for (j = 0; j < num_cpus; j++) {
-			snprintf(string_cpu, ETH_GSTRING_LEN, "%s [CPU %d]",
-				 dpaa_stats_percpu[i], j);
-			memcpy(strings, string_cpu, ETH_GSTRING_LEN);
-			strings += ETH_GSTRING_LEN;
-		}
-		snprintf(string_cpu, ETH_GSTRING_LEN, "%s [TOTAL]",
-			 dpaa_stats_percpu[i]);
-		memcpy(strings, string_cpu, ETH_GSTRING_LEN);
-		strings += ETH_GSTRING_LEN;
+		for (j = 0; j < num_cpus; j++)
+			ethtool_sprintf(&data, "%s [CPU %d]",
+					dpaa_stats_percpu[i], j);
+
+		ethtool_sprintf(&data, "%s [TOTAL]", dpaa_stats_percpu[i]);
+	}
+	for (i = 0; i < num_cpus; i++)
+		ethtool_sprintf(&data, "bpool [CPU %d]", i);
+
+	ethtool_puts(&data, "bpool [TOTAL]");
+
+	for (i = 0; i < DPAA_STATS_GLOBAL_LEN; i++)
+		ethtool_puts(&data, dpaa_stats_global[i]);
+}
+
+static int dpaa_get_rxfh_fields(struct net_device *dev,
+				struct ethtool_rxfh_fields *cmd)
+{
+	struct dpaa_priv *priv = netdev_priv(dev);
+
+	cmd->data = 0;
+
+	switch (cmd->flow_type) {
+	case TCP_V4_FLOW:
+	case TCP_V6_FLOW:
+	case UDP_V4_FLOW:
+	case UDP_V6_FLOW:
+		if (priv->keygen_in_use)
+			cmd->data |= RXH_L4_B_0_1 | RXH_L4_B_2_3;
+		fallthrough;
+	case IPV4_FLOW:
+	case IPV6_FLOW:
+	case SCTP_V4_FLOW:
+	case SCTP_V6_FLOW:
+	case AH_ESP_V4_FLOW:
+	case AH_ESP_V6_FLOW:
+	case AH_V4_FLOW:
+	case AH_V6_FLOW:
+	case ESP_V4_FLOW:
+	case ESP_V6_FLOW:
+		if (priv->keygen_in_use)
+			cmd->data |= RXH_IP_SRC | RXH_IP_DST;
+		break;
+	default:
+		cmd->data = 0;
+		break;
 	}
-	for (i = 0; i < DPAA_BPS_NUM; i++) {
-		for (j = 0; j < num_cpus; j++) {
-			snprintf(string_cpu, ETH_GSTRING_LEN,
-				 "bpool %c [CPU %d]", 'a' + i, j);
-			memcpy(strings, string_cpu, ETH_GSTRING_LEN);
-			strings += ETH_GSTRING_LEN;
+
+	return 0;
+}
+
+static void dpaa_set_hash(struct net_device *net_dev, bool enable)
+{
+	struct mac_device *mac_dev;
+	struct fman_port *rxport;
+	struct dpaa_priv *priv;
+
+	priv = netdev_priv(net_dev);
+	mac_dev = priv->mac_dev;
+	rxport = mac_dev->port[0];
+
+	fman_port_use_kg_hash(rxport, enable);
+	priv->keygen_in_use = enable;
+}
+
+static int dpaa_set_rxfh_fields(struct net_device *dev,
+				const struct ethtool_rxfh_fields *nfc,
+				struct netlink_ext_ack *extack)
+{
+	int ret = -EINVAL;
+
+	/* we support hashing on IPv4/v6 src/dest IP and L4 src/dest port */
+	if (nfc->data &
+	    ~(RXH_IP_SRC | RXH_IP_DST | RXH_L4_B_0_1 | RXH_L4_B_2_3))
+		return -EINVAL;
+
+	switch (nfc->flow_type) {
+	case TCP_V4_FLOW:
+	case TCP_V6_FLOW:
+	case UDP_V4_FLOW:
+	case UDP_V6_FLOW:
+	case IPV4_FLOW:
+	case IPV6_FLOW:
+	case SCTP_V4_FLOW:
+	case SCTP_V6_FLOW:
+	case AH_ESP_V4_FLOW:
+	case AH_ESP_V6_FLOW:
+	case AH_V4_FLOW:
+	case AH_V6_FLOW:
+	case ESP_V4_FLOW:
+	case ESP_V6_FLOW:
+		dpaa_set_hash(dev, !!nfc->data);
+		ret = 0;
+		break;
+	default:
+		break;
+	}
+
+	return ret;
+}
+
+static int dpaa_get_ts_info(struct net_device *net_dev,
+			    struct kernel_ethtool_ts_info *info)
+{
+	struct device *dev = net_dev->dev.parent;
+	struct device_node *mac_node = dev->of_node;
+	struct device_node *fman_node = NULL, *ptp_node = NULL;
+	struct platform_device *ptp_dev = NULL;
+	struct ptp_qoriq *ptp = NULL;
+
+	info->phc_index = -1;
+
+	fman_node = of_get_parent(mac_node);
+	if (fman_node) {
+		ptp_node = of_parse_phandle(fman_node, "ptimer-handle", 0);
+		of_node_put(fman_node);
+	}
+
+	if (ptp_node) {
+		ptp_dev = of_find_device_by_node(ptp_node);
+		of_node_put(ptp_node);
+	}
+
+	if (ptp_dev) {
+		ptp = platform_get_drvdata(ptp_dev);
+		put_device(&ptp_dev->dev);
+	}
+
+	if (ptp)
+		info->phc_index = ptp->phc_index;
+
+	info->so_timestamping = SOF_TIMESTAMPING_TX_HARDWARE |
+				SOF_TIMESTAMPING_RX_HARDWARE |
+				SOF_TIMESTAMPING_RAW_HARDWARE;
+	info->tx_types = (1 << HWTSTAMP_TX_OFF) |
+			 (1 << HWTSTAMP_TX_ON);
+	info->rx_filters = (1 << HWTSTAMP_FILTER_NONE) |
+			   (1 << HWTSTAMP_FILTER_ALL);
+
+	return 0;
+}
+
+static int dpaa_get_coalesce(struct net_device *dev,
+			     struct ethtool_coalesce *c,
+			     struct kernel_ethtool_coalesce *kernel_coal,
+			     struct netlink_ext_ack *extack)
+{
+	struct qman_portal *portal;
+	u32 period;
+	u8 thresh;
+
+	portal = qman_get_affine_portal(smp_processor_id());
+	qman_portal_get_iperiod(portal, &period);
+	qman_dqrr_get_ithresh(portal, &thresh);
+
+	c->rx_coalesce_usecs = period;
+	c->rx_max_coalesced_frames = thresh;
+
+	return 0;
+}
+
+static int dpaa_set_coalesce(struct net_device *dev,
+			     struct ethtool_coalesce *c,
+			     struct kernel_ethtool_coalesce *kernel_coal,
+			     struct netlink_ext_ack *extack)
+{
+	const cpumask_t *cpus = qman_affine_cpus();
+	struct qman_portal *portal;
+	u32 period, prev_period;
+	u8 thresh, prev_thresh;
+	bool *needs_revert;
+	int cpu, res;
+
+	needs_revert = kcalloc(num_possible_cpus(), sizeof(bool), GFP_KERNEL);
+	if (!needs_revert)
+		return -ENOMEM;
+
+	period = c->rx_coalesce_usecs;
+	thresh = c->rx_max_coalesced_frames;
+
+	/* save previous values */
+	portal = qman_get_affine_portal(smp_processor_id());
+	qman_portal_get_iperiod(portal, &prev_period);
+	qman_dqrr_get_ithresh(portal, &prev_thresh);
+
+	/* set new values */
+	for_each_cpu_and(cpu, cpus, cpu_online_mask) {
+		portal = qman_get_affine_portal(cpu);
+		res = qman_portal_set_iperiod(portal, period);
+		if (res)
+			goto revert_values;
+		res = qman_dqrr_set_ithresh(portal, thresh);
+		if (res) {
+			qman_portal_set_iperiod(portal, prev_period);
+			goto revert_values;
 		}
-		snprintf(string_cpu, ETH_GSTRING_LEN, "bpool %c [TOTAL]",
-			 'a' + i);
-		memcpy(strings, string_cpu, ETH_GSTRING_LEN);
-		strings += ETH_GSTRING_LEN;
+		needs_revert[cpu] = true;
 	}
-	memcpy(strings, dpaa_stats_global, size);
+
+	kfree(needs_revert);
+
+	return 0;
+
+revert_values:
+	/* restore previous values */
+	for_each_cpu_and(cpu, cpus, cpu_online_mask) {
+		if (!needs_revert[cpu])
+			continue;
+		portal = qman_get_affine_portal(cpu);
+		/* previous values will not fail, ignore return value */
+		qman_portal_set_iperiod(portal, prev_period);
+		qman_dqrr_set_ithresh(portal, prev_thresh);
+	}
+
+	kfree(needs_revert);
+
+	return res;
 }
 
 const struct ethtool_ops dpaa_ethtool_ops = {
+	.supported_coalesce_params = ETHTOOL_COALESCE_RX_USECS |
+				     ETHTOOL_COALESCE_RX_MAX_FRAMES,
 	.get_drvinfo = dpaa_get_drvinfo,
 	.get_msglevel = dpaa_get_msglevel,
 	.set_msglevel = dpaa_set_msglevel,
@@ -414,4 +482,9 @@ const struct ethtool_ops dpaa_ethtool_ops = {
 	.get_strings = dpaa_get_strings,
 	.get_link_ksettings = dpaa_get_link_ksettings,
 	.set_link_ksettings = dpaa_set_link_ksettings,
+	.get_rxfh_fields = dpaa_get_rxfh_fields,
+	.set_rxfh_fields = dpaa_set_rxfh_fields,
+	.get_ts_info = dpaa_get_ts_info,
+	.get_coalesce = dpaa_get_coalesce,
+	.set_coalesce = dpaa_set_coalesce,
 };
diff --git a/drivers/net/ethernet/freescale/dpaa2/Kconfig b/drivers/net/ethernet/freescale/dpaa2/Kconfig
new file mode 100644
index 000000000000..d029b69c3f18
--- /dev/null
+++ b/drivers/net/ethernet/freescale/dpaa2/Kconfig
@@ -0,0 +1,39 @@
+# SPDX-License-Identifier: GPL-2.0-only
+config FSL_DPAA2_ETH
+	tristate "Freescale DPAA2 Ethernet"
+	depends on FSL_MC_BUS && FSL_MC_DPIO
+	select PHYLINK
+	select PCS_LYNX
+	select FSL_XGMAC_MDIO
+	select NET_DEVLINK
+	help
+	  This is the DPAA2 Ethernet driver supporting Freescale SoCs
+	  with DPAA2 (DataPath Acceleration Architecture v2).
+	  The driver manages network objects discovered on the Freescale
+	  MC bus.
+
+if FSL_DPAA2_ETH
+config FSL_DPAA2_ETH_DCB
+	bool "Data Center Bridging (DCB) Support"
+	default n
+	depends on DCB
+	help
+	  Enable Priority-Based Flow Control (PFC) support for DPAA2 Ethernet
+	  devices.
+endif
+
+config FSL_DPAA2_PTP_CLOCK
+	tristate "Freescale DPAA2 PTP Clock"
+	depends on FSL_DPAA2_ETH && PTP_1588_CLOCK_QORIQ
+	default y
+	help
+	  This driver adds support for using the DPAA2 1588 timer module
+	  as a PTP clock.
+
+config FSL_DPAA2_SWITCH
+	tristate "Freescale DPAA2 Ethernet Switch"
+	depends on BRIDGE || BRIDGE=n
+	depends on NET_SWITCHDEV
+	help
+	  Driver for Freescale DPAA2 Ethernet Switch. This driver manages
+	  switch objects discovered on the Freeescale MC bus.
diff --git a/drivers/net/ethernet/freescale/dpaa2/Makefile b/drivers/net/ethernet/freescale/dpaa2/Makefile
new file mode 100644
index 000000000000..1b05ba8d1cbf
--- /dev/null
+++ b/drivers/net/ethernet/freescale/dpaa2/Makefile
@@ -0,0 +1,17 @@
+# SPDX-License-Identifier: GPL-2.0
+#
+# Makefile for the Freescale DPAA2 Ethernet controller
+#
+
+obj-$(CONFIG_FSL_DPAA2_ETH)		+= fsl-dpaa2-eth.o
+obj-$(CONFIG_FSL_DPAA2_PTP_CLOCK)	+= fsl-dpaa2-ptp.o
+obj-$(CONFIG_FSL_DPAA2_SWITCH)		+= fsl-dpaa2-switch.o
+
+fsl-dpaa2-eth-objs	:= dpaa2-eth.o dpaa2-ethtool.o dpni.o dpaa2-mac.o dpmac.o dpaa2-eth-devlink.o dpaa2-xsk.o
+fsl-dpaa2-eth-${CONFIG_FSL_DPAA2_ETH_DCB} += dpaa2-eth-dcb.o
+fsl-dpaa2-eth-${CONFIG_DEBUG_FS} += dpaa2-eth-debugfs.o
+fsl-dpaa2-ptp-objs	:= dpaa2-ptp.o dprtc.o
+fsl-dpaa2-switch-objs	:= dpaa2-switch.o dpaa2-switch-ethtool.o dpsw.o dpaa2-switch-flower.o dpaa2-mac.o dpmac.o
+
+# Needed by the tracing framework
+CFLAGS_dpaa2-eth.o := -I$(src)
diff --git a/drivers/net/ethernet/freescale/dpaa2/dpaa2-eth-dcb.c b/drivers/net/ethernet/freescale/dpaa2/dpaa2-eth-dcb.c
new file mode 100644
index 000000000000..84de0644168d
--- /dev/null
+++ b/drivers/net/ethernet/freescale/dpaa2/dpaa2-eth-dcb.c
@@ -0,0 +1,150 @@
+// SPDX-License-Identifier: (GPL-2.0+ OR BSD-3-Clause)
+/* Copyright 2020 NXP */
+
+#include "dpaa2-eth.h"
+
+static int dpaa2_eth_dcbnl_ieee_getpfc(struct net_device *net_dev,
+				       struct ieee_pfc *pfc)
+{
+	struct dpaa2_eth_priv *priv = netdev_priv(net_dev);
+
+	if (!(priv->link_state.options & DPNI_LINK_OPT_PFC_PAUSE))
+		return 0;
+
+	memcpy(pfc, &priv->pfc, sizeof(priv->pfc));
+	pfc->pfc_cap = dpaa2_eth_tc_count(priv);
+
+	return 0;
+}
+
+static inline bool dpaa2_eth_is_prio_enabled(u8 pfc_en, u8 tc)
+{
+	return !!(pfc_en & (1 << tc));
+}
+
+static int dpaa2_eth_set_pfc_cn(struct dpaa2_eth_priv *priv, u8 pfc_en)
+{
+	struct dpni_congestion_notification_cfg cfg = {0};
+	int i, err;
+
+	cfg.notification_mode = DPNI_CONG_OPT_FLOW_CONTROL;
+	cfg.units = DPNI_CONGESTION_UNIT_FRAMES;
+	cfg.message_iova = 0ULL;
+	cfg.message_ctx = 0ULL;
+
+	for (i = 0; i < dpaa2_eth_tc_count(priv); i++) {
+		if (dpaa2_eth_is_prio_enabled(pfc_en, i)) {
+			cfg.threshold_entry = DPAA2_ETH_CN_THRESH_ENTRY(priv);
+			cfg.threshold_exit = DPAA2_ETH_CN_THRESH_EXIT(priv);
+		} else {
+			/* For priorities not set in the pfc_en mask, we leave
+			 * the congestion thresholds at zero, which effectively
+			 * disables generation of PFC frames for them
+			 */
+			cfg.threshold_entry = 0;
+			cfg.threshold_exit = 0;
+		}
+
+		err = dpni_set_congestion_notification(priv->mc_io, 0,
+						       priv->mc_token,
+						       DPNI_QUEUE_RX, i, &cfg);
+		if (err) {
+			netdev_err(priv->net_dev,
+				   "dpni_set_congestion_notification failed\n");
+			return err;
+		}
+	}
+
+	return 0;
+}
+
+static int dpaa2_eth_dcbnl_ieee_setpfc(struct net_device *net_dev,
+				       struct ieee_pfc *pfc)
+{
+	struct dpaa2_eth_priv *priv = netdev_priv(net_dev);
+	struct dpni_link_cfg link_cfg = {0};
+	bool tx_pause;
+	int err;
+
+	if (pfc->mbc || pfc->delay)
+		return -EOPNOTSUPP;
+
+	/* If same PFC enabled mask, nothing to do */
+	if (priv->pfc.pfc_en == pfc->pfc_en)
+		return 0;
+
+	/* We allow PFC configuration even if it won't have any effect until
+	 * general pause frames are enabled
+	 */
+	tx_pause = dpaa2_eth_tx_pause_enabled(priv->link_state.options);
+	if (!dpaa2_eth_rx_pause_enabled(priv->link_state.options) || !tx_pause)
+		netdev_warn(net_dev, "Pause support must be enabled in order for PFC to work!\n");
+
+	link_cfg.rate = priv->link_state.rate;
+	link_cfg.options = priv->link_state.options;
+	if (pfc->pfc_en)
+		link_cfg.options |= DPNI_LINK_OPT_PFC_PAUSE;
+	else
+		link_cfg.options &= ~DPNI_LINK_OPT_PFC_PAUSE;
+	err = dpni_set_link_cfg(priv->mc_io, 0, priv->mc_token, &link_cfg);
+	if (err) {
+		netdev_err(net_dev, "dpni_set_link_cfg failed\n");
+		return err;
+	}
+
+	/* Configure congestion notifications for the enabled priorities */
+	err = dpaa2_eth_set_pfc_cn(priv, pfc->pfc_en);
+	if (err)
+		return err;
+
+	memcpy(&priv->pfc, pfc, sizeof(priv->pfc));
+	priv->pfc_enabled = !!pfc->pfc_en;
+
+	dpaa2_eth_set_rx_taildrop(priv, tx_pause, priv->pfc_enabled);
+
+	return 0;
+}
+
+static u8 dpaa2_eth_dcbnl_getdcbx(struct net_device *net_dev)
+{
+	struct dpaa2_eth_priv *priv = netdev_priv(net_dev);
+
+	return priv->dcbx_mode;
+}
+
+static u8 dpaa2_eth_dcbnl_setdcbx(struct net_device *net_dev, u8 mode)
+{
+	struct dpaa2_eth_priv *priv = netdev_priv(net_dev);
+
+	return (mode != (priv->dcbx_mode)) ? 1 : 0;
+}
+
+static u8 dpaa2_eth_dcbnl_getcap(struct net_device *net_dev, int capid, u8 *cap)
+{
+	struct dpaa2_eth_priv *priv = netdev_priv(net_dev);
+
+	switch (capid) {
+	case DCB_CAP_ATTR_PFC:
+		*cap = true;
+		break;
+	case DCB_CAP_ATTR_PFC_TCS:
+		*cap = 1 << (dpaa2_eth_tc_count(priv) - 1);
+		break;
+	case DCB_CAP_ATTR_DCBX:
+		*cap = priv->dcbx_mode;
+		break;
+	default:
+		*cap = false;
+		break;
+	}
+
+	return 0;
+}
+
+const struct dcbnl_rtnl_ops dpaa2_eth_dcbnl_ops = {
+	.ieee_getpfc	= dpaa2_eth_dcbnl_ieee_getpfc,
+	.ieee_setpfc	= dpaa2_eth_dcbnl_ieee_setpfc,
+	.getdcbx	= dpaa2_eth_dcbnl_getdcbx,
+	.setdcbx	= dpaa2_eth_dcbnl_setdcbx,
+	.getcap		= dpaa2_eth_dcbnl_getcap,
+};
diff --git a/drivers/net/ethernet/freescale/dpaa2/dpaa2-eth-debugfs.c b/drivers/net/ethernet/freescale/dpaa2/dpaa2-eth-debugfs.c
new file mode 100644
index 000000000000..1af254caeb0d
--- /dev/null
+++ b/drivers/net/ethernet/freescale/dpaa2/dpaa2-eth-debugfs.c
@@ -0,0 +1,207 @@
+// SPDX-License-Identifier: (GPL-2.0+ OR BSD-3-Clause)
+/* Copyright 2015 Freescale Semiconductor Inc.
+ * Copyright 2018-2019 NXP
+ */
+#include <linux/module.h>
+#include <linux/debugfs.h>
+#include "dpaa2-eth.h"
+#include "dpaa2-eth-debugfs.h"
+
+#define DPAA2_ETH_DBG_ROOT "dpaa2-eth"
+
+static struct dentry *dpaa2_dbg_root;
+
+static int dpaa2_dbg_cpu_show(struct seq_file *file, void *offset)
+{
+	struct dpaa2_eth_priv *priv = (struct dpaa2_eth_priv *)file->private;
+	struct rtnl_link_stats64 *stats;
+	struct dpaa2_eth_drv_stats *extras;
+	int i;
+
+	seq_printf(file, "Per-CPU stats for %s\n", priv->net_dev->name);
+	seq_printf(file, "%s%16s%16s%16s%16s%16s%16s%16s%16s%16s\n",
+		   "CPU", "Rx", "Rx Err", "Rx SG", "Tx", "Tx Err", "Tx conf",
+		   "Tx SG", "Tx converted to SG", "Enq busy");
+
+	for_each_online_cpu(i) {
+		stats = per_cpu_ptr(priv->percpu_stats, i);
+		extras = per_cpu_ptr(priv->percpu_extras, i);
+		seq_printf(file, "%3d%16llu%16llu%16llu%16llu%16llu%16llu%16llu%16llu%16llu\n",
+			   i,
+			   stats->rx_packets,
+			   stats->rx_errors,
+			   extras->rx_sg_frames,
+			   stats->tx_packets,
+			   stats->tx_errors,
+			   extras->tx_conf_frames,
+			   extras->tx_sg_frames,
+			   extras->tx_converted_sg_frames,
+			   extras->tx_portal_busy);
+	}
+
+	return 0;
+}
+
+DEFINE_SHOW_ATTRIBUTE(dpaa2_dbg_cpu);
+
+static char *fq_type_to_str(struct dpaa2_eth_fq *fq)
+{
+	switch (fq->type) {
+	case DPAA2_RX_FQ:
+		return "Rx";
+	case DPAA2_TX_CONF_FQ:
+		return "Tx conf";
+	default:
+		return "N/A";
+	}
+}
+
+static int dpaa2_dbg_fqs_show(struct seq_file *file, void *offset)
+{
+	struct dpaa2_eth_priv *priv = (struct dpaa2_eth_priv *)file->private;
+	struct dpaa2_eth_fq *fq;
+	u32 fcnt, bcnt;
+	int i, err;
+
+	seq_printf(file, "FQ stats for %s:\n", priv->net_dev->name);
+	seq_printf(file, "%s%16s%16s%16s%16s%16s\n",
+		   "VFQID", "CPU", "TC", "Type", "Frames", "Pending frames");
+
+	for (i = 0; i <  priv->num_fqs; i++) {
+		fq = &priv->fq[i];
+		err = dpaa2_io_query_fq_count(NULL, fq->fqid, &fcnt, &bcnt);
+		if (err)
+			fcnt = 0;
+
+		/* Skip FQs with no traffic */
+		if (!fq->stats.frames && !fcnt)
+			continue;
+
+		seq_printf(file, "%5d%16d%16d%16s%16llu%16u\n",
+			   fq->fqid,
+			   fq->target_cpu,
+			   fq->tc,
+			   fq_type_to_str(fq),
+			   fq->stats.frames,
+			   fcnt);
+	}
+
+	return 0;
+}
+
+DEFINE_SHOW_ATTRIBUTE(dpaa2_dbg_fqs);
+
+static int dpaa2_dbg_ch_show(struct seq_file *file, void *offset)
+{
+	struct dpaa2_eth_priv *priv = (struct dpaa2_eth_priv *)file->private;
+	struct dpaa2_eth_channel *ch;
+	int i;
+
+	seq_printf(file, "Channel stats for %s:\n", priv->net_dev->name);
+	seq_printf(file, "%s  %5s%16s%16s%16s%16s%16s%16s\n",
+		   "IDX", "CHID", "CPU", "Deq busy", "Frames", "CDANs",
+		   "Avg Frm/CDAN", "Buf count");
+
+	for (i = 0; i < priv->num_channels; i++) {
+		ch = priv->channel[i];
+		seq_printf(file, "%3s%d%6d%16d%16llu%16llu%16llu%16llu%16d\n",
+			   "CH#", i, ch->ch_id,
+			   ch->nctx.desired_cpu,
+			   ch->stats.dequeue_portal_busy,
+			   ch->stats.frames,
+			   ch->stats.cdan,
+			   div64_u64(ch->stats.frames, ch->stats.cdan),
+			   ch->buf_count);
+	}
+
+	return 0;
+}
+
+DEFINE_SHOW_ATTRIBUTE(dpaa2_dbg_ch);
+
+static int dpaa2_dbg_bp_show(struct seq_file *file, void *offset)
+{
+	struct dpaa2_eth_priv *priv = (struct dpaa2_eth_priv *)file->private;
+	int i, j, num_queues, buf_cnt;
+	struct dpaa2_eth_bp *bp;
+	char ch_name[10];
+	int err;
+
+	/* Print out the header */
+	seq_printf(file, "Buffer pool info for %s:\n", priv->net_dev->name);
+	seq_printf(file, "%s  %10s%15s", "IDX", "BPID", "Buf count");
+	num_queues = dpaa2_eth_queue_count(priv);
+	for (i = 0; i < num_queues; i++) {
+		snprintf(ch_name, sizeof(ch_name), "CH#%d", i);
+		seq_printf(file, "%10s", ch_name);
+	}
+	seq_printf(file, "\n");
+
+	/* For each buffer pool, print out its BPID, the number of buffers in
+	 * that buffer pool and the channels which are using it.
+	 */
+	for (i = 0; i < priv->num_bps; i++) {
+		bp = priv->bp[i];
+
+		err = dpaa2_io_query_bp_count(NULL, bp->bpid, &buf_cnt);
+		if (err) {
+			netdev_warn(priv->net_dev, "Buffer count query error %d\n", err);
+			return err;
+		}
+
+		seq_printf(file, "%3s%d%10d%15d", "BP#", i, bp->bpid, buf_cnt);
+		for (j = 0; j < num_queues; j++) {
+			if (priv->channel[j]->bp == bp)
+				seq_printf(file, "%10s", "x");
+			else
+				seq_printf(file, "%10s", "");
+		}
+		seq_printf(file, "\n");
+	}
+
+	return 0;
+}
+
+DEFINE_SHOW_ATTRIBUTE(dpaa2_dbg_bp);
+
+void dpaa2_dbg_add(struct dpaa2_eth_priv *priv)
+{
+	struct fsl_mc_device *dpni_dev;
+	struct dentry *dir;
+	char name[10];
+
+	/* Create a directory for the interface */
+	dpni_dev = to_fsl_mc_device(priv->net_dev->dev.parent);
+	snprintf(name, 10, "dpni.%d", dpni_dev->obj_desc.id);
+	dir = debugfs_create_dir(name, dpaa2_dbg_root);
+	priv->dbg.dir = dir;
+
+	/* per-cpu stats file */
+	debugfs_create_file("cpu_stats", 0444, dir, priv, &dpaa2_dbg_cpu_fops);
+
+	/* per-fq stats file */
+	debugfs_create_file("fq_stats", 0444, dir, priv, &dpaa2_dbg_fqs_fops);
+
+	/* per-fq stats file */
+	debugfs_create_file("ch_stats", 0444, dir, priv, &dpaa2_dbg_ch_fops);
+
+	/* per buffer pool stats file */
+	debugfs_create_file("bp_stats", 0444, dir, priv, &dpaa2_dbg_bp_fops);
+
+}
+
+void dpaa2_dbg_remove(struct dpaa2_eth_priv *priv)
+{
+	debugfs_remove_recursive(priv->dbg.dir);
+}
+
+void dpaa2_eth_dbg_init(void)
+{
+	dpaa2_dbg_root = debugfs_create_dir(DPAA2_ETH_DBG_ROOT, NULL);
+	pr_debug("DPAA2-ETH: debugfs created\n");
+}
+
+void dpaa2_eth_dbg_exit(void)
+{
+	debugfs_remove(dpaa2_dbg_root);
+}
diff --git a/drivers/net/ethernet/freescale/dpaa2/dpaa2-eth-debugfs.h b/drivers/net/ethernet/freescale/dpaa2/dpaa2-eth-debugfs.h
new file mode 100644
index 000000000000..15598b28f03b
--- /dev/null
+++ b/drivers/net/ethernet/freescale/dpaa2/dpaa2-eth-debugfs.h
@@ -0,0 +1,28 @@
+/* SPDX-License-Identifier: (GPL-2.0+ OR BSD-3-Clause) */
+/* Copyright 2015 Freescale Semiconductor Inc.
+ * Copyright 2018-2019 NXP
+ */
+#ifndef DPAA2_ETH_DEBUGFS_H
+#define DPAA2_ETH_DEBUGFS_H
+
+#include <linux/dcache.h>
+
+struct dpaa2_eth_priv;
+
+struct dpaa2_debugfs {
+	struct dentry *dir;
+};
+
+#ifdef CONFIG_DEBUG_FS
+void dpaa2_eth_dbg_init(void);
+void dpaa2_eth_dbg_exit(void);
+void dpaa2_dbg_add(struct dpaa2_eth_priv *priv);
+void dpaa2_dbg_remove(struct dpaa2_eth_priv *priv);
+#else
+static inline void dpaa2_eth_dbg_init(void) {}
+static inline void dpaa2_eth_dbg_exit(void) {}
+static inline void dpaa2_dbg_add(struct dpaa2_eth_priv *priv) {}
+static inline void dpaa2_dbg_remove(struct dpaa2_eth_priv *priv) {}
+#endif /* CONFIG_DEBUG_FS */
+
+#endif /* DPAA2_ETH_DEBUGFS_H */
diff --git a/drivers/net/ethernet/freescale/dpaa2/dpaa2-eth-devlink.c b/drivers/net/ethernet/freescale/dpaa2/dpaa2-eth-devlink.c
new file mode 100644
index 000000000000..76f808d38066
--- /dev/null
+++ b/drivers/net/ethernet/freescale/dpaa2/dpaa2-eth-devlink.c
@@ -0,0 +1,289 @@
+// SPDX-License-Identifier: (GPL-2.0+ OR BSD-3-Clause)
+#include "dpaa2-eth.h"
+/* Copyright 2020 NXP
+ */
+
+#define DPAA2_ETH_TRAP_DROP(_id, _group_id)					\
+	DEVLINK_TRAP_GENERIC(DROP, DROP, _id,					\
+			     DEVLINK_TRAP_GROUP_GENERIC_ID_##_group_id, 0)
+
+static const struct devlink_trap_group dpaa2_eth_trap_groups_arr[] = {
+	DEVLINK_TRAP_GROUP_GENERIC(PARSER_ERROR_DROPS, 0),
+};
+
+static const struct devlink_trap dpaa2_eth_traps_arr[] = {
+	DPAA2_ETH_TRAP_DROP(VXLAN_PARSING, PARSER_ERROR_DROPS),
+	DPAA2_ETH_TRAP_DROP(LLC_SNAP_PARSING, PARSER_ERROR_DROPS),
+	DPAA2_ETH_TRAP_DROP(VLAN_PARSING, PARSER_ERROR_DROPS),
+	DPAA2_ETH_TRAP_DROP(PPPOE_PPP_PARSING, PARSER_ERROR_DROPS),
+	DPAA2_ETH_TRAP_DROP(MPLS_PARSING, PARSER_ERROR_DROPS),
+	DPAA2_ETH_TRAP_DROP(ARP_PARSING, PARSER_ERROR_DROPS),
+	DPAA2_ETH_TRAP_DROP(IP_1_PARSING, PARSER_ERROR_DROPS),
+	DPAA2_ETH_TRAP_DROP(IP_N_PARSING, PARSER_ERROR_DROPS),
+	DPAA2_ETH_TRAP_DROP(GRE_PARSING, PARSER_ERROR_DROPS),
+	DPAA2_ETH_TRAP_DROP(UDP_PARSING, PARSER_ERROR_DROPS),
+	DPAA2_ETH_TRAP_DROP(TCP_PARSING, PARSER_ERROR_DROPS),
+	DPAA2_ETH_TRAP_DROP(IPSEC_PARSING, PARSER_ERROR_DROPS),
+	DPAA2_ETH_TRAP_DROP(SCTP_PARSING, PARSER_ERROR_DROPS),
+	DPAA2_ETH_TRAP_DROP(DCCP_PARSING, PARSER_ERROR_DROPS),
+	DPAA2_ETH_TRAP_DROP(GTP_PARSING, PARSER_ERROR_DROPS),
+	DPAA2_ETH_TRAP_DROP(ESP_PARSING, PARSER_ERROR_DROPS),
+};
+
+static int dpaa2_eth_dl_info_get(struct devlink *devlink,
+				 struct devlink_info_req *req,
+				 struct netlink_ext_ack *extack)
+{
+	struct dpaa2_eth_devlink_priv *dl_priv = devlink_priv(devlink);
+	struct dpaa2_eth_priv *priv = dl_priv->dpaa2_priv;
+	char buf[10];
+
+	scnprintf(buf, 10, "%d.%d", priv->dpni_ver_major, priv->dpni_ver_minor);
+	return devlink_info_version_running_put(req, "dpni", buf);
+}
+
+static struct dpaa2_eth_trap_item *
+dpaa2_eth_dl_trap_item_lookup(struct dpaa2_eth_priv *priv, u16 trap_id)
+{
+	struct dpaa2_eth_trap_data *dpaa2_eth_trap_data = priv->trap_data;
+	int i;
+
+	for (i = 0; i < ARRAY_SIZE(dpaa2_eth_traps_arr); i++) {
+		if (dpaa2_eth_traps_arr[i].id == trap_id)
+			return &dpaa2_eth_trap_data->trap_items_arr[i];
+	}
+
+	return NULL;
+}
+
+struct dpaa2_eth_trap_item *dpaa2_eth_dl_get_trap(struct dpaa2_eth_priv *priv,
+						  struct dpaa2_fapr *fapr)
+{
+	static const struct dpaa2_faf_error_bit {
+		int position;
+		enum devlink_trap_generic_id trap_id;
+	} faf_bits[] = {
+		{ .position = 5,  .trap_id = DEVLINK_TRAP_GENERIC_ID_VXLAN_PARSING },
+		{ .position = 20, .trap_id = DEVLINK_TRAP_GENERIC_ID_LLC_SNAP_PARSING },
+		{ .position = 24, .trap_id = DEVLINK_TRAP_GENERIC_ID_VLAN_PARSING },
+		{ .position = 26, .trap_id = DEVLINK_TRAP_GENERIC_ID_PPPOE_PPP_PARSING },
+		{ .position = 29, .trap_id = DEVLINK_TRAP_GENERIC_ID_MPLS_PARSING },
+		{ .position = 31, .trap_id = DEVLINK_TRAP_GENERIC_ID_ARP_PARSING },
+		{ .position = 52, .trap_id = DEVLINK_TRAP_GENERIC_ID_IP_1_PARSING },
+		{ .position = 61, .trap_id = DEVLINK_TRAP_GENERIC_ID_IP_N_PARSING },
+		{ .position = 67, .trap_id = DEVLINK_TRAP_GENERIC_ID_GRE_PARSING },
+		{ .position = 71, .trap_id = DEVLINK_TRAP_GENERIC_ID_UDP_PARSING },
+		{ .position = 76, .trap_id = DEVLINK_TRAP_GENERIC_ID_TCP_PARSING },
+		{ .position = 80, .trap_id = DEVLINK_TRAP_GENERIC_ID_IPSEC_PARSING },
+		{ .position = 82, .trap_id = DEVLINK_TRAP_GENERIC_ID_SCTP_PARSING },
+		{ .position = 84, .trap_id = DEVLINK_TRAP_GENERIC_ID_DCCP_PARSING },
+		{ .position = 88, .trap_id = DEVLINK_TRAP_GENERIC_ID_GTP_PARSING },
+		{ .position = 90, .trap_id = DEVLINK_TRAP_GENERIC_ID_ESP_PARSING },
+	};
+	u64 faf_word;
+	u64 mask;
+	int i;
+
+	for (i = 0; i < ARRAY_SIZE(faf_bits); i++) {
+		if (faf_bits[i].position < 32) {
+			/* Low part of FAF.
+			 * position ranges from 31 to 0, mask from 0 to 31.
+			 */
+			mask = 1ull << (31 - faf_bits[i].position);
+			faf_word = __le32_to_cpu(fapr->faf_lo);
+		} else {
+			/* High part of FAF.
+			 * position ranges from 95 to 32, mask from 0 to 63.
+			 */
+			mask = 1ull << (63 - (faf_bits[i].position - 32));
+			faf_word = __le64_to_cpu(fapr->faf_hi);
+		}
+		if (faf_word & mask)
+			return dpaa2_eth_dl_trap_item_lookup(priv, faf_bits[i].trap_id);
+	}
+	return NULL;
+}
+
+static int dpaa2_eth_dl_trap_init(struct devlink *devlink,
+				  const struct devlink_trap *trap,
+				  void *trap_ctx)
+{
+	struct dpaa2_eth_devlink_priv *dl_priv = devlink_priv(devlink);
+	struct dpaa2_eth_priv *priv = dl_priv->dpaa2_priv;
+	struct dpaa2_eth_trap_item *dpaa2_eth_trap_item;
+
+	dpaa2_eth_trap_item = dpaa2_eth_dl_trap_item_lookup(priv, trap->id);
+	if (WARN_ON(!dpaa2_eth_trap_item))
+		return -ENOENT;
+
+	dpaa2_eth_trap_item->trap_ctx = trap_ctx;
+
+	return 0;
+}
+
+static int dpaa2_eth_dl_trap_action_set(struct devlink *devlink,
+					const struct devlink_trap *trap,
+					enum devlink_trap_action action,
+					struct netlink_ext_ack *extack)
+{
+	/* No support for changing the action of an independent packet trap,
+	 * only per trap group - parser error drops
+	 */
+	NL_SET_ERR_MSG_MOD(extack,
+			   "Cannot change trap action independently of group");
+	return -EOPNOTSUPP;
+}
+
+static int dpaa2_eth_dl_trap_group_action_set(struct devlink *devlink,
+					      const struct devlink_trap_group *group,
+					      enum devlink_trap_action action,
+					      struct netlink_ext_ack *extack)
+{
+	struct dpaa2_eth_devlink_priv *dl_priv = devlink_priv(devlink);
+	struct dpaa2_eth_priv *priv = dl_priv->dpaa2_priv;
+	struct net_device *net_dev = priv->net_dev;
+	struct device *dev = net_dev->dev.parent;
+	struct dpni_error_cfg err_cfg = {0};
+	int err;
+
+	if (group->id != DEVLINK_TRAP_GROUP_GENERIC_ID_PARSER_ERROR_DROPS)
+		return -EOPNOTSUPP;
+
+	/* Configure handling of frames marked as errors from the parser */
+	err_cfg.errors = DPAA2_FAS_RX_ERR_MASK;
+	err_cfg.set_frame_annotation = 1;
+
+	switch (action) {
+	case DEVLINK_TRAP_ACTION_DROP:
+		err_cfg.error_action = DPNI_ERROR_ACTION_DISCARD;
+		break;
+	case DEVLINK_TRAP_ACTION_TRAP:
+		err_cfg.error_action = DPNI_ERROR_ACTION_SEND_TO_ERROR_QUEUE;
+		break;
+	default:
+		return -EOPNOTSUPP;
+	}
+
+	err = dpni_set_errors_behavior(priv->mc_io, 0, priv->mc_token, &err_cfg);
+	if (err) {
+		dev_err(dev, "dpni_set_errors_behavior failed\n");
+		return err;
+	}
+
+	return 0;
+}
+
+static const struct devlink_ops dpaa2_eth_devlink_ops = {
+	.info_get = dpaa2_eth_dl_info_get,
+	.trap_init = dpaa2_eth_dl_trap_init,
+	.trap_action_set = dpaa2_eth_dl_trap_action_set,
+	.trap_group_action_set = dpaa2_eth_dl_trap_group_action_set,
+};
+
+int dpaa2_eth_dl_alloc(struct dpaa2_eth_priv *priv)
+{
+	struct net_device *net_dev = priv->net_dev;
+	struct device *dev = net_dev->dev.parent;
+	struct dpaa2_eth_devlink_priv *dl_priv;
+
+	priv->devlink =
+		devlink_alloc(&dpaa2_eth_devlink_ops, sizeof(*dl_priv), dev);
+	if (!priv->devlink) {
+		dev_err(dev, "devlink_alloc failed\n");
+		return -ENOMEM;
+	}
+	dl_priv = devlink_priv(priv->devlink);
+	dl_priv->dpaa2_priv = priv;
+	return 0;
+}
+
+void dpaa2_eth_dl_free(struct dpaa2_eth_priv *priv)
+{
+	devlink_free(priv->devlink);
+}
+
+
+void dpaa2_eth_dl_register(struct dpaa2_eth_priv *priv)
+{
+	devlink_register(priv->devlink);
+}
+
+void dpaa2_eth_dl_unregister(struct dpaa2_eth_priv *priv)
+{
+	devlink_unregister(priv->devlink);
+}
+
+int dpaa2_eth_dl_port_add(struct dpaa2_eth_priv *priv)
+{
+	struct devlink_port *devlink_port = &priv->devlink_port;
+	struct devlink_port_attrs attrs = {};
+
+	attrs.flavour = DEVLINK_PORT_FLAVOUR_PHYSICAL;
+	devlink_port_attrs_set(devlink_port, &attrs);
+	return devlink_port_register(priv->devlink, devlink_port, 0);
+}
+
+void dpaa2_eth_dl_port_del(struct dpaa2_eth_priv *priv)
+{
+	struct devlink_port *devlink_port = &priv->devlink_port;
+
+	devlink_port_unregister(devlink_port);
+}
+
+int dpaa2_eth_dl_traps_register(struct dpaa2_eth_priv *priv)
+{
+	struct dpaa2_eth_trap_data *dpaa2_eth_trap_data;
+	struct net_device *net_dev = priv->net_dev;
+	struct device *dev = net_dev->dev.parent;
+	int err;
+
+	dpaa2_eth_trap_data = kzalloc(sizeof(*dpaa2_eth_trap_data), GFP_KERNEL);
+	if (!dpaa2_eth_trap_data)
+		return -ENOMEM;
+	priv->trap_data = dpaa2_eth_trap_data;
+
+	dpaa2_eth_trap_data->trap_items_arr = kcalloc(ARRAY_SIZE(dpaa2_eth_traps_arr),
+						      sizeof(struct dpaa2_eth_trap_item),
+						      GFP_KERNEL);
+	if (!dpaa2_eth_trap_data->trap_items_arr) {
+		err = -ENOMEM;
+		goto trap_data_free;
+	}
+
+	err = devlink_trap_groups_register(priv->devlink, dpaa2_eth_trap_groups_arr,
+					   ARRAY_SIZE(dpaa2_eth_trap_groups_arr));
+	if (err) {
+		dev_err(dev, "devlink_trap_groups_register() = %d\n", err);
+		goto trap_items_arr_free;
+	}
+
+	err = devlink_traps_register(priv->devlink, dpaa2_eth_traps_arr,
+				     ARRAY_SIZE(dpaa2_eth_traps_arr), priv);
+	if (err) {
+		dev_err(dev, "devlink_traps_register() = %d\n", err);
+		goto trap_groups_unregiser;
+	}
+
+	return 0;
+
+trap_groups_unregiser:
+	devlink_trap_groups_unregister(priv->devlink, dpaa2_eth_trap_groups_arr,
+				       ARRAY_SIZE(dpaa2_eth_trap_groups_arr));
+trap_items_arr_free:
+	kfree(dpaa2_eth_trap_data->trap_items_arr);
+trap_data_free:
+	kfree(dpaa2_eth_trap_data);
+	priv->trap_data = NULL;
+
+	return err;
+}
+
+void dpaa2_eth_dl_traps_unregister(struct dpaa2_eth_priv *priv)
+{
+	devlink_traps_unregister(priv->devlink, dpaa2_eth_traps_arr,
+				 ARRAY_SIZE(dpaa2_eth_traps_arr));
+	devlink_trap_groups_unregister(priv->devlink, dpaa2_eth_trap_groups_arr,
+				       ARRAY_SIZE(dpaa2_eth_trap_groups_arr));
+	kfree(priv->trap_data->trap_items_arr);
+	kfree(priv->trap_data);
+}
diff --git a/drivers/net/ethernet/freescale/dpaa2/dpaa2-eth-trace.h b/drivers/net/ethernet/freescale/dpaa2/dpaa2-eth-trace.h
new file mode 100644
index 000000000000..956767e0869c
--- /dev/null
+++ b/drivers/net/ethernet/freescale/dpaa2/dpaa2-eth-trace.h
@@ -0,0 +1,197 @@
+/* SPDX-License-Identifier: (GPL-2.0+ OR BSD-3-Clause) */
+/* Copyright 2014-2015 Freescale Semiconductor Inc.
+ */
+
+#undef TRACE_SYSTEM
+#define TRACE_SYSTEM	dpaa2_eth
+
+#if !defined(_DPAA2_ETH_TRACE_H) || defined(TRACE_HEADER_MULTI_READ)
+#define _DPAA2_ETH_TRACE_H
+
+#include <linux/skbuff.h>
+#include <linux/netdevice.h>
+#include <linux/tracepoint.h>
+
+#define TR_FMT "[%s] fd: addr=0x%llx, len=%u, off=%u"
+/* trace_printk format for raw buffer event class */
+#define TR_BUF_FMT "[%s] vaddr=%p size=%zu dma_addr=%pad map_size=%zu bpid=%d"
+
+/* This is used to declare a class of events.
+ * individual events of this type will be defined below.
+ */
+
+/* Store details about a frame descriptor */
+DECLARE_EVENT_CLASS(dpaa2_eth_fd,
+		    /* Trace function prototype */
+		    TP_PROTO(struct net_device *netdev,
+			     const struct dpaa2_fd *fd),
+
+		    /* Repeat argument list here */
+		    TP_ARGS(netdev, fd),
+
+		    /* A structure containing the relevant information we want
+		     * to record. Declare name and type for each normal element,
+		     * name, type and size for arrays. Use __string for variable
+		     * length strings.
+		     */
+		    TP_STRUCT__entry(
+				     __field(u64, fd_addr)
+				     __field(u32, fd_len)
+				     __field(u16, fd_offset)
+				     __string(name, netdev->name)
+		    ),
+
+		    /* The function that assigns values to the above declared
+		     * fields
+		     */
+		    TP_fast_assign(
+				   __entry->fd_addr = dpaa2_fd_get_addr(fd);
+				   __entry->fd_len = dpaa2_fd_get_len(fd);
+				   __entry->fd_offset = dpaa2_fd_get_offset(fd);
+				   __assign_str(name);
+		    ),
+
+		    /* This is what gets printed when the trace event is
+		     * triggered.
+		     */
+		    TP_printk(TR_FMT,
+			      __get_str(name),
+			      __entry->fd_addr,
+			      __entry->fd_len,
+			      __entry->fd_offset)
+);
+
+/* Now declare events of the above type. Format is:
+ * DEFINE_EVENT(class, name, proto, args), with proto and args same as for class
+ */
+
+/* Tx (egress) fd */
+DEFINE_EVENT(dpaa2_eth_fd, dpaa2_tx_fd,
+	     TP_PROTO(struct net_device *netdev,
+		      const struct dpaa2_fd *fd),
+
+	     TP_ARGS(netdev, fd)
+);
+
+/* Tx (egress) XSK fd */
+DEFINE_EVENT(dpaa2_eth_fd, dpaa2_tx_xsk_fd,
+	     TP_PROTO(struct net_device *netdev,
+		      const struct dpaa2_fd *fd),
+
+	     TP_ARGS(netdev, fd)
+);
+
+/* Rx fd */
+DEFINE_EVENT(dpaa2_eth_fd, dpaa2_rx_fd,
+	     TP_PROTO(struct net_device *netdev,
+		      const struct dpaa2_fd *fd),
+
+	     TP_ARGS(netdev, fd)
+);
+
+/* Rx XSK fd */
+DEFINE_EVENT(dpaa2_eth_fd, dpaa2_rx_xsk_fd,
+	     TP_PROTO(struct net_device *netdev,
+		      const struct dpaa2_fd *fd),
+
+	     TP_ARGS(netdev, fd)
+);
+
+/* Tx confirmation fd */
+DEFINE_EVENT(dpaa2_eth_fd, dpaa2_tx_conf_fd,
+	     TP_PROTO(struct net_device *netdev,
+		      const struct dpaa2_fd *fd),
+
+	     TP_ARGS(netdev, fd)
+);
+
+/* Log data about raw buffers. Useful for tracing DPBP content. */
+DECLARE_EVENT_CLASS(dpaa2_eth_buf,
+		    /* Trace function prototype */
+		    TP_PROTO(struct net_device *netdev,
+			     /* virtual address and size */
+			    void *vaddr,
+			    size_t size,
+			    /* dma map address and size */
+			    dma_addr_t dma_addr,
+			    size_t map_size,
+			    /* buffer pool id, if relevant */
+			    u16 bpid),
+
+		    /* Repeat argument list here */
+		    TP_ARGS(netdev, vaddr, size, dma_addr, map_size, bpid),
+
+		    /* A structure containing the relevant information we want
+		     * to record. Declare name and type for each normal element,
+		     * name, type and size for arrays. Use __string for variable
+		     * length strings.
+		     */
+		    TP_STRUCT__entry(
+				      __field(void *, vaddr)
+				      __field(size_t, size)
+				      __field(dma_addr_t, dma_addr)
+				      __field(size_t, map_size)
+				      __field(u16, bpid)
+				      __string(name, netdev->name)
+		    ),
+
+		    /* The function that assigns values to the above declared
+		     * fields
+		     */
+		    TP_fast_assign(
+				   __entry->vaddr = vaddr;
+				   __entry->size = size;
+				   __entry->dma_addr = dma_addr;
+				   __entry->map_size = map_size;
+				   __entry->bpid = bpid;
+				   __assign_str(name);
+		    ),
+
+		    /* This is what gets printed when the trace event is
+		     * triggered.
+		     */
+		    TP_printk(TR_BUF_FMT,
+			      __get_str(name),
+			      __entry->vaddr,
+			      __entry->size,
+			      &__entry->dma_addr,
+			      __entry->map_size,
+			      __entry->bpid)
+);
+
+/* Main memory buff seeding */
+DEFINE_EVENT(dpaa2_eth_buf, dpaa2_eth_buf_seed,
+	     TP_PROTO(struct net_device *netdev,
+		      void *vaddr,
+		      size_t size,
+		      dma_addr_t dma_addr,
+		      size_t map_size,
+		      u16 bpid),
+
+	     TP_ARGS(netdev, vaddr, size, dma_addr, map_size, bpid)
+);
+
+/* UMEM buff seeding on AF_XDP fast path */
+DEFINE_EVENT(dpaa2_eth_buf, dpaa2_xsk_buf_seed,
+	     TP_PROTO(struct net_device *netdev,
+		      void *vaddr,
+		      size_t size,
+		      dma_addr_t dma_addr,
+		      size_t map_size,
+		      u16 bpid),
+
+	     TP_ARGS(netdev, vaddr, size, dma_addr, map_size, bpid)
+);
+
+/* If only one event of a certain type needs to be declared, use TRACE_EVENT().
+ * The syntax is the same as for DECLARE_EVENT_CLASS().
+ */
+
+#endif /* _DPAA2_ETH_TRACE_H */
+
+/* This must be outside ifdef _DPAA2_ETH_TRACE_H */
+#undef TRACE_INCLUDE_PATH
+#define TRACE_INCLUDE_PATH .
+#undef TRACE_INCLUDE_FILE
+#define TRACE_INCLUDE_FILE	dpaa2-eth-trace
+#include <trace/define_trace.h>
diff --git a/drivers/net/ethernet/freescale/dpaa2/dpaa2-eth.c b/drivers/net/ethernet/freescale/dpaa2/dpaa2-eth.c
new file mode 100644
index 000000000000..c96d1d6ba8fe
--- /dev/null
+++ b/drivers/net/ethernet/freescale/dpaa2/dpaa2-eth.c
@@ -0,0 +1,5168 @@
+// SPDX-License-Identifier: (GPL-2.0+ OR BSD-3-Clause)
+/* Copyright 2014-2016 Freescale Semiconductor Inc.
+ * Copyright 2016-2022 NXP
+ */
+#include <linux/init.h>
+#include <linux/module.h>
+#include <linux/platform_device.h>
+#include <linux/etherdevice.h>
+#include <linux/of_net.h>
+#include <linux/interrupt.h>
+#include <linux/kthread.h>
+#include <linux/iommu.h>
+#include <linux/fsl/mc.h>
+#include <linux/bpf.h>
+#include <linux/bpf_trace.h>
+#include <linux/fsl/ptp_qoriq.h>
+#include <linux/ptp_classify.h>
+#include <net/pkt_cls.h>
+#include <net/sock.h>
+#include <net/tso.h>
+#include <net/xdp_sock_drv.h>
+
+#include "dpaa2-eth.h"
+
+/* CREATE_TRACE_POINTS only needs to be defined once. Other dpa files
+ * using trace events only need to #include <trace/events/sched.h>
+ */
+#define CREATE_TRACE_POINTS
+#include "dpaa2-eth-trace.h"
+
+MODULE_LICENSE("Dual BSD/GPL");
+MODULE_AUTHOR("Freescale Semiconductor, Inc");
+MODULE_DESCRIPTION("Freescale DPAA2 Ethernet Driver");
+
+struct ptp_qoriq *dpaa2_ptp;
+EXPORT_SYMBOL(dpaa2_ptp);
+
+static void dpaa2_eth_detect_features(struct dpaa2_eth_priv *priv)
+{
+	priv->features = 0;
+
+	if (dpaa2_eth_cmp_dpni_ver(priv, DPNI_PTP_ONESTEP_VER_MAJOR,
+				   DPNI_PTP_ONESTEP_VER_MINOR) >= 0)
+		priv->features |= DPAA2_ETH_FEATURE_ONESTEP_CFG_DIRECT;
+}
+
+static void dpaa2_update_ptp_onestep_indirect(struct dpaa2_eth_priv *priv,
+					      u32 offset, u8 udp)
+{
+	struct dpni_single_step_cfg cfg;
+
+	cfg.en = 1;
+	cfg.ch_update = udp;
+	cfg.offset = offset;
+	cfg.peer_delay = 0;
+
+	if (dpni_set_single_step_cfg(priv->mc_io, 0, priv->mc_token, &cfg))
+		WARN_ONCE(1, "Failed to set single step register");
+}
+
+static void dpaa2_update_ptp_onestep_direct(struct dpaa2_eth_priv *priv,
+					    u32 offset, u8 udp)
+{
+	u32 val = 0;
+
+	val = DPAA2_PTP_SINGLE_STEP_ENABLE |
+	       DPAA2_PTP_SINGLE_CORRECTION_OFF(offset);
+
+	if (udp)
+		val |= DPAA2_PTP_SINGLE_STEP_CH;
+
+	if (priv->onestep_reg_base)
+		writel(val, priv->onestep_reg_base);
+}
+
+static void dpaa2_ptp_onestep_reg_update_method(struct dpaa2_eth_priv *priv)
+{
+	struct device *dev = priv->net_dev->dev.parent;
+	struct dpni_single_step_cfg ptp_cfg;
+
+	priv->dpaa2_set_onestep_params_cb = dpaa2_update_ptp_onestep_indirect;
+
+	if (!(priv->features & DPAA2_ETH_FEATURE_ONESTEP_CFG_DIRECT))
+		return;
+
+	if (dpni_get_single_step_cfg(priv->mc_io, 0,
+				     priv->mc_token, &ptp_cfg)) {
+		dev_err(dev, "dpni_get_single_step_cfg cannot retrieve onestep reg, falling back to indirect update\n");
+		return;
+	}
+
+	if (!ptp_cfg.ptp_onestep_reg_base) {
+		dev_err(dev, "1588 onestep reg not available, falling back to indirect update\n");
+		return;
+	}
+
+	priv->onestep_reg_base = ioremap(ptp_cfg.ptp_onestep_reg_base,
+					 sizeof(u32));
+	if (!priv->onestep_reg_base) {
+		dev_err(dev, "1588 onestep reg cannot be mapped, falling back to indirect update\n");
+		return;
+	}
+
+	priv->dpaa2_set_onestep_params_cb = dpaa2_update_ptp_onestep_direct;
+}
+
+void *dpaa2_iova_to_virt(struct iommu_domain *domain,
+			 dma_addr_t iova_addr)
+{
+	phys_addr_t phys_addr;
+
+	phys_addr = domain ? iommu_iova_to_phys(domain, iova_addr) : iova_addr;
+
+	return phys_to_virt(phys_addr);
+}
+
+static void dpaa2_eth_validate_rx_csum(struct dpaa2_eth_priv *priv,
+				       u32 fd_status,
+				       struct sk_buff *skb)
+{
+	skb_checksum_none_assert(skb);
+
+	/* HW checksum validation is disabled, nothing to do here */
+	if (!(priv->net_dev->features & NETIF_F_RXCSUM))
+		return;
+
+	/* Read checksum validation bits */
+	if (!((fd_status & DPAA2_FAS_L3CV) &&
+	      (fd_status & DPAA2_FAS_L4CV)))
+		return;
+
+	/* Inform the stack there's no need to compute L3/L4 csum anymore */
+	skb->ip_summed = CHECKSUM_UNNECESSARY;
+}
+
+/* Free a received FD.
+ * Not to be used for Tx conf FDs or on any other paths.
+ */
+static void dpaa2_eth_free_rx_fd(struct dpaa2_eth_priv *priv,
+				 const struct dpaa2_fd *fd,
+				 void *vaddr)
+{
+	struct device *dev = priv->net_dev->dev.parent;
+	dma_addr_t addr = dpaa2_fd_get_addr(fd);
+	u8 fd_format = dpaa2_fd_get_format(fd);
+	struct dpaa2_sg_entry *sgt;
+	void *sg_vaddr;
+	int i;
+
+	/* If single buffer frame, just free the data buffer */
+	if (fd_format == dpaa2_fd_single)
+		goto free_buf;
+	else if (fd_format != dpaa2_fd_sg)
+		/* We don't support any other format */
+		return;
+
+	/* For S/G frames, we first need to free all SG entries
+	 * except the first one, which was taken care of already
+	 */
+	sgt = vaddr + dpaa2_fd_get_offset(fd);
+	for (i = 1; i < DPAA2_ETH_MAX_SG_ENTRIES; i++) {
+		addr = dpaa2_sg_get_addr(&sgt[i]);
+		sg_vaddr = dpaa2_iova_to_virt(priv->iommu_domain, addr);
+		dma_unmap_page(dev, addr, priv->rx_buf_size,
+			       DMA_BIDIRECTIONAL);
+
+		free_pages((unsigned long)sg_vaddr, 0);
+		if (dpaa2_sg_is_final(&sgt[i]))
+			break;
+	}
+
+free_buf:
+	free_pages((unsigned long)vaddr, 0);
+}
+
+/* Build a linear skb based on a single-buffer frame descriptor */
+static struct sk_buff *dpaa2_eth_build_linear_skb(struct dpaa2_eth_channel *ch,
+						  const struct dpaa2_fd *fd,
+						  void *fd_vaddr)
+{
+	struct sk_buff *skb = NULL;
+	u16 fd_offset = dpaa2_fd_get_offset(fd);
+	u32 fd_length = dpaa2_fd_get_len(fd);
+
+	ch->buf_count--;
+
+	skb = build_skb(fd_vaddr, DPAA2_ETH_RX_BUF_RAW_SIZE);
+	if (unlikely(!skb))
+		return NULL;
+
+	skb_reserve(skb, fd_offset);
+	skb_put(skb, fd_length);
+
+	return skb;
+}
+
+/* Build a non linear (fragmented) skb based on a S/G table */
+static struct sk_buff *dpaa2_eth_build_frag_skb(struct dpaa2_eth_priv *priv,
+						struct dpaa2_eth_channel *ch,
+						struct dpaa2_sg_entry *sgt)
+{
+	struct sk_buff *skb = NULL;
+	struct device *dev = priv->net_dev->dev.parent;
+	void *sg_vaddr;
+	dma_addr_t sg_addr;
+	u16 sg_offset;
+	u32 sg_length;
+	struct page *page, *head_page;
+	int page_offset;
+	int i;
+
+	for (i = 0; i < DPAA2_ETH_MAX_SG_ENTRIES; i++) {
+		struct dpaa2_sg_entry *sge = &sgt[i];
+
+		/* NOTE: We only support SG entries in dpaa2_sg_single format,
+		 * but this is the only format we may receive from HW anyway
+		 */
+
+		/* Get the address and length from the S/G entry */
+		sg_addr = dpaa2_sg_get_addr(sge);
+		sg_vaddr = dpaa2_iova_to_virt(priv->iommu_domain, sg_addr);
+		dma_unmap_page(dev, sg_addr, priv->rx_buf_size,
+			       DMA_BIDIRECTIONAL);
+
+		sg_length = dpaa2_sg_get_len(sge);
+
+		if (i == 0) {
+			/* We build the skb around the first data buffer */
+			skb = build_skb(sg_vaddr, DPAA2_ETH_RX_BUF_RAW_SIZE);
+			if (unlikely(!skb)) {
+				/* Free the first SG entry now, since we already
+				 * unmapped it and obtained the virtual address
+				 */
+				free_pages((unsigned long)sg_vaddr, 0);
+
+				/* We still need to subtract the buffers used
+				 * by this FD from our software counter
+				 */
+				while (!dpaa2_sg_is_final(&sgt[i]) &&
+				       i < DPAA2_ETH_MAX_SG_ENTRIES)
+					i++;
+				break;
+			}
+
+			sg_offset = dpaa2_sg_get_offset(sge);
+			skb_reserve(skb, sg_offset);
+			skb_put(skb, sg_length);
+		} else {
+			/* Rest of the data buffers are stored as skb frags */
+			page = virt_to_page(sg_vaddr);
+			head_page = virt_to_head_page(sg_vaddr);
+
+			/* Offset in page (which may be compound).
+			 * Data in subsequent SG entries is stored from the
+			 * beginning of the buffer, so we don't need to add the
+			 * sg_offset.
+			 */
+			page_offset = ((unsigned long)sg_vaddr &
+				(PAGE_SIZE - 1)) +
+				(page_address(page) - page_address(head_page));
+
+			skb_add_rx_frag(skb, i - 1, head_page, page_offset,
+					sg_length, priv->rx_buf_size);
+		}
+
+		if (dpaa2_sg_is_final(sge))
+			break;
+	}
+
+	WARN_ONCE(i == DPAA2_ETH_MAX_SG_ENTRIES, "Final bit not set in SGT");
+
+	/* Count all data buffers + SG table buffer */
+	ch->buf_count -= i + 2;
+
+	return skb;
+}
+
+/* Free buffers acquired from the buffer pool or which were meant to
+ * be released in the pool
+ */
+static void dpaa2_eth_free_bufs(struct dpaa2_eth_priv *priv, u64 *buf_array,
+				int count, bool xsk_zc)
+{
+	struct device *dev = priv->net_dev->dev.parent;
+	struct dpaa2_eth_swa *swa;
+	struct xdp_buff *xdp_buff;
+	void *vaddr;
+	int i;
+
+	for (i = 0; i < count; i++) {
+		vaddr = dpaa2_iova_to_virt(priv->iommu_domain, buf_array[i]);
+
+		if (!xsk_zc) {
+			dma_unmap_page(dev, buf_array[i], priv->rx_buf_size,
+				       DMA_BIDIRECTIONAL);
+			free_pages((unsigned long)vaddr, 0);
+		} else {
+			swa = (struct dpaa2_eth_swa *)
+				(vaddr + DPAA2_ETH_RX_HWA_SIZE);
+			xdp_buff = swa->xsk.xdp_buff;
+			xsk_buff_free(xdp_buff);
+		}
+	}
+}
+
+void dpaa2_eth_recycle_buf(struct dpaa2_eth_priv *priv,
+			   struct dpaa2_eth_channel *ch,
+			   dma_addr_t addr)
+{
+	int retries = 0;
+	int err;
+
+	ch->recycled_bufs[ch->recycled_bufs_cnt++] = addr;
+	if (ch->recycled_bufs_cnt < DPAA2_ETH_BUFS_PER_CMD)
+		return;
+
+	while ((err = dpaa2_io_service_release(ch->dpio, ch->bp->bpid,
+					       ch->recycled_bufs,
+					       ch->recycled_bufs_cnt)) == -EBUSY) {
+		if (retries++ >= DPAA2_ETH_SWP_BUSY_RETRIES)
+			break;
+		cpu_relax();
+	}
+
+	if (err) {
+		dpaa2_eth_free_bufs(priv, ch->recycled_bufs,
+				    ch->recycled_bufs_cnt, ch->xsk_zc);
+		ch->buf_count -= ch->recycled_bufs_cnt;
+	}
+
+	ch->recycled_bufs_cnt = 0;
+}
+
+static int dpaa2_eth_xdp_flush(struct dpaa2_eth_priv *priv,
+			       struct dpaa2_eth_fq *fq,
+			       struct dpaa2_eth_xdp_fds *xdp_fds)
+{
+	int total_enqueued = 0, retries = 0, enqueued;
+	struct dpaa2_eth_drv_stats *percpu_extras;
+	int num_fds, err, max_retries;
+	struct dpaa2_fd *fds;
+
+	percpu_extras = this_cpu_ptr(priv->percpu_extras);
+
+	/* try to enqueue all the FDs until the max number of retries is hit */
+	fds = xdp_fds->fds;
+	num_fds = xdp_fds->num;
+	max_retries = num_fds * DPAA2_ETH_ENQUEUE_RETRIES;
+	while (total_enqueued < num_fds && retries < max_retries) {
+		err = priv->enqueue(priv, fq, &fds[total_enqueued],
+				    0, num_fds - total_enqueued, &enqueued);
+		if (err == -EBUSY) {
+			percpu_extras->tx_portal_busy += ++retries;
+			continue;
+		}
+		total_enqueued += enqueued;
+	}
+	xdp_fds->num = 0;
+
+	return total_enqueued;
+}
+
+static void dpaa2_eth_xdp_tx_flush(struct dpaa2_eth_priv *priv,
+				   struct dpaa2_eth_channel *ch,
+				   struct dpaa2_eth_fq *fq)
+{
+	struct rtnl_link_stats64 *percpu_stats;
+	struct dpaa2_fd *fds;
+	int enqueued, i;
+
+	percpu_stats = this_cpu_ptr(priv->percpu_stats);
+
+	// enqueue the array of XDP_TX frames
+	enqueued = dpaa2_eth_xdp_flush(priv, fq, &fq->xdp_tx_fds);
+
+	/* update statistics */
+	percpu_stats->tx_packets += enqueued;
+	fds = fq->xdp_tx_fds.fds;
+	for (i = 0; i < enqueued; i++) {
+		percpu_stats->tx_bytes += dpaa2_fd_get_len(&fds[i]);
+		ch->stats.xdp_tx++;
+	}
+	for (i = enqueued; i < fq->xdp_tx_fds.num; i++) {
+		dpaa2_eth_recycle_buf(priv, ch, dpaa2_fd_get_addr(&fds[i]));
+		percpu_stats->tx_errors++;
+		ch->stats.xdp_tx_err++;
+	}
+	fq->xdp_tx_fds.num = 0;
+}
+
+void dpaa2_eth_xdp_enqueue(struct dpaa2_eth_priv *priv,
+			   struct dpaa2_eth_channel *ch,
+			   struct dpaa2_fd *fd,
+			   void *buf_start, u16 queue_id)
+{
+	struct dpaa2_faead *faead;
+	struct dpaa2_fd *dest_fd;
+	struct dpaa2_eth_fq *fq;
+	u32 ctrl, frc;
+
+	/* Mark the egress frame hardware annotation area as valid */
+	frc = dpaa2_fd_get_frc(fd);
+	dpaa2_fd_set_frc(fd, frc | DPAA2_FD_FRC_FAEADV);
+	dpaa2_fd_set_ctrl(fd, DPAA2_FD_CTRL_ASAL);
+
+	/* Instruct hardware to release the FD buffer directly into
+	 * the buffer pool once transmission is completed, instead of
+	 * sending a Tx confirmation frame to us
+	 */
+	ctrl = DPAA2_FAEAD_A4V | DPAA2_FAEAD_A2V | DPAA2_FAEAD_EBDDV;
+	faead = dpaa2_get_faead(buf_start, false);
+	faead->ctrl = cpu_to_le32(ctrl);
+	faead->conf_fqid = 0;
+
+	fq = &priv->fq[queue_id];
+	dest_fd = &fq->xdp_tx_fds.fds[fq->xdp_tx_fds.num++];
+	memcpy(dest_fd, fd, sizeof(*dest_fd));
+
+	if (fq->xdp_tx_fds.num < DEV_MAP_BULK_SIZE)
+		return;
+
+	dpaa2_eth_xdp_tx_flush(priv, ch, fq);
+}
+
+static u32 dpaa2_eth_run_xdp(struct dpaa2_eth_priv *priv,
+			     struct dpaa2_eth_channel *ch,
+			     struct dpaa2_eth_fq *rx_fq,
+			     struct dpaa2_fd *fd, void *vaddr)
+{
+	dma_addr_t addr = dpaa2_fd_get_addr(fd);
+	struct bpf_prog *xdp_prog;
+	struct xdp_buff xdp;
+	u32 xdp_act = XDP_PASS;
+	int err, offset;
+
+	xdp_prog = READ_ONCE(ch->xdp.prog);
+	if (!xdp_prog)
+		goto out;
+
+	offset = dpaa2_fd_get_offset(fd) - XDP_PACKET_HEADROOM;
+	xdp_init_buff(&xdp, DPAA2_ETH_RX_BUF_RAW_SIZE - offset, &ch->xdp_rxq);
+	xdp_prepare_buff(&xdp, vaddr + offset, XDP_PACKET_HEADROOM,
+			 dpaa2_fd_get_len(fd), false);
+
+	xdp_act = bpf_prog_run_xdp(xdp_prog, &xdp);
+
+	/* xdp.data pointer may have changed */
+	dpaa2_fd_set_offset(fd, xdp.data - vaddr);
+	dpaa2_fd_set_len(fd, xdp.data_end - xdp.data);
+
+	switch (xdp_act) {
+	case XDP_PASS:
+		break;
+	case XDP_TX:
+		dpaa2_eth_xdp_enqueue(priv, ch, fd, vaddr, rx_fq->flowid);
+		break;
+	default:
+		bpf_warn_invalid_xdp_action(priv->net_dev, xdp_prog, xdp_act);
+		fallthrough;
+	case XDP_ABORTED:
+		trace_xdp_exception(priv->net_dev, xdp_prog, xdp_act);
+		fallthrough;
+	case XDP_DROP:
+		dpaa2_eth_recycle_buf(priv, ch, addr);
+		ch->stats.xdp_drop++;
+		break;
+	case XDP_REDIRECT:
+		dma_unmap_page(priv->net_dev->dev.parent, addr,
+			       priv->rx_buf_size, DMA_BIDIRECTIONAL);
+		ch->buf_count--;
+
+		/* Allow redirect use of full headroom */
+		xdp.data_hard_start = vaddr;
+		xdp.frame_sz = DPAA2_ETH_RX_BUF_RAW_SIZE;
+
+		err = xdp_do_redirect(priv->net_dev, &xdp, xdp_prog);
+		if (unlikely(err)) {
+			addr = dma_map_page(priv->net_dev->dev.parent,
+					    virt_to_page(vaddr), 0,
+					    priv->rx_buf_size, DMA_BIDIRECTIONAL);
+			if (unlikely(dma_mapping_error(priv->net_dev->dev.parent, addr))) {
+				free_pages((unsigned long)vaddr, 0);
+			} else {
+				ch->buf_count++;
+				dpaa2_eth_recycle_buf(priv, ch, addr);
+			}
+			ch->stats.xdp_drop++;
+		} else {
+			ch->stats.xdp_redirect++;
+		}
+		break;
+	}
+
+	ch->xdp.res |= xdp_act;
+out:
+	return xdp_act;
+}
+
+struct sk_buff *dpaa2_eth_alloc_skb(struct dpaa2_eth_priv *priv,
+				    struct dpaa2_eth_channel *ch,
+				    const struct dpaa2_fd *fd, u32 fd_length,
+				    void *fd_vaddr)
+{
+	u16 fd_offset = dpaa2_fd_get_offset(fd);
+	struct sk_buff *skb = NULL;
+	unsigned int skb_len;
+
+	skb_len = fd_length + dpaa2_eth_needed_headroom(NULL);
+
+	skb = napi_alloc_skb(&ch->napi, skb_len);
+	if (!skb)
+		return NULL;
+
+	skb_reserve(skb, dpaa2_eth_needed_headroom(NULL));
+	skb_put(skb, fd_length);
+
+	memcpy(skb->data, fd_vaddr + fd_offset, fd_length);
+
+	return skb;
+}
+
+static struct sk_buff *dpaa2_eth_copybreak(struct dpaa2_eth_channel *ch,
+					   const struct dpaa2_fd *fd,
+					   void *fd_vaddr)
+{
+	struct dpaa2_eth_priv *priv = ch->priv;
+	u32 fd_length = dpaa2_fd_get_len(fd);
+
+	if (fd_length > priv->rx_copybreak)
+		return NULL;
+
+	return dpaa2_eth_alloc_skb(priv, ch, fd, fd_length, fd_vaddr);
+}
+
+void dpaa2_eth_receive_skb(struct dpaa2_eth_priv *priv,
+			   struct dpaa2_eth_channel *ch,
+			   const struct dpaa2_fd *fd, void *vaddr,
+			   struct dpaa2_eth_fq *fq,
+			   struct rtnl_link_stats64 *percpu_stats,
+			   struct sk_buff *skb)
+{
+	struct dpaa2_fas *fas;
+	u32 status = 0;
+
+	fas = dpaa2_get_fas(vaddr, false);
+	prefetch(fas);
+	prefetch(skb->data);
+
+	/* Get the timestamp value */
+	if (priv->rx_tstamp) {
+		struct skb_shared_hwtstamps *shhwtstamps = skb_hwtstamps(skb);
+		__le64 *ts = dpaa2_get_ts(vaddr, false);
+		u64 ns;
+
+		memset(shhwtstamps, 0, sizeof(*shhwtstamps));
+
+		ns = DPAA2_PTP_CLK_PERIOD_NS * le64_to_cpup(ts);
+		shhwtstamps->hwtstamp = ns_to_ktime(ns);
+	}
+
+	/* Check if we need to validate the L4 csum */
+	if (likely(dpaa2_fd_get_frc(fd) & DPAA2_FD_FRC_FASV)) {
+		status = le32_to_cpu(fas->status);
+		dpaa2_eth_validate_rx_csum(priv, status, skb);
+	}
+
+	skb->protocol = eth_type_trans(skb, priv->net_dev);
+	skb_record_rx_queue(skb, fq->flowid);
+
+	percpu_stats->rx_packets++;
+	percpu_stats->rx_bytes += dpaa2_fd_get_len(fd);
+	ch->stats.bytes_per_cdan += dpaa2_fd_get_len(fd);
+
+	list_add_tail(&skb->list, ch->rx_list);
+}
+
+/* Main Rx frame processing routine */
+void dpaa2_eth_rx(struct dpaa2_eth_priv *priv,
+		  struct dpaa2_eth_channel *ch,
+		  const struct dpaa2_fd *fd,
+		  struct dpaa2_eth_fq *fq)
+{
+	dma_addr_t addr = dpaa2_fd_get_addr(fd);
+	u8 fd_format = dpaa2_fd_get_format(fd);
+	void *vaddr;
+	struct sk_buff *skb;
+	struct rtnl_link_stats64 *percpu_stats;
+	struct dpaa2_eth_drv_stats *percpu_extras;
+	struct device *dev = priv->net_dev->dev.parent;
+	bool recycle_rx_buf = false;
+	void *buf_data;
+	u32 xdp_act;
+
+	/* Tracing point */
+	trace_dpaa2_rx_fd(priv->net_dev, fd);
+
+	vaddr = dpaa2_iova_to_virt(priv->iommu_domain, addr);
+	dma_sync_single_for_cpu(dev, addr, priv->rx_buf_size,
+				DMA_BIDIRECTIONAL);
+
+	buf_data = vaddr + dpaa2_fd_get_offset(fd);
+	prefetch(buf_data);
+
+	percpu_stats = this_cpu_ptr(priv->percpu_stats);
+	percpu_extras = this_cpu_ptr(priv->percpu_extras);
+
+	if (fd_format == dpaa2_fd_single) {
+		xdp_act = dpaa2_eth_run_xdp(priv, ch, fq, (struct dpaa2_fd *)fd, vaddr);
+		if (xdp_act != XDP_PASS) {
+			percpu_stats->rx_packets++;
+			percpu_stats->rx_bytes += dpaa2_fd_get_len(fd);
+			return;
+		}
+
+		skb = dpaa2_eth_copybreak(ch, fd, vaddr);
+		if (!skb) {
+			dma_unmap_page(dev, addr, priv->rx_buf_size,
+				       DMA_BIDIRECTIONAL);
+			skb = dpaa2_eth_build_linear_skb(ch, fd, vaddr);
+		} else {
+			recycle_rx_buf = true;
+		}
+	} else if (fd_format == dpaa2_fd_sg) {
+		WARN_ON(priv->xdp_prog);
+
+		dma_unmap_page(dev, addr, priv->rx_buf_size,
+			       DMA_BIDIRECTIONAL);
+		skb = dpaa2_eth_build_frag_skb(priv, ch, buf_data);
+		free_pages((unsigned long)vaddr, 0);
+		percpu_extras->rx_sg_frames++;
+		percpu_extras->rx_sg_bytes += dpaa2_fd_get_len(fd);
+	} else {
+		/* We don't support any other format */
+		goto err_frame_format;
+	}
+
+	if (unlikely(!skb))
+		goto err_build_skb;
+
+	dpaa2_eth_receive_skb(priv, ch, fd, vaddr, fq, percpu_stats, skb);
+
+	if (recycle_rx_buf)
+		dpaa2_eth_recycle_buf(priv, ch, dpaa2_fd_get_addr(fd));
+	return;
+
+err_build_skb:
+	dpaa2_eth_free_rx_fd(priv, fd, vaddr);
+err_frame_format:
+	percpu_stats->rx_dropped++;
+}
+
+/* Processing of Rx frames received on the error FQ
+ * We check and print the error bits and then free the frame
+ */
+static void dpaa2_eth_rx_err(struct dpaa2_eth_priv *priv,
+			     struct dpaa2_eth_channel *ch,
+			     const struct dpaa2_fd *fd,
+			     struct dpaa2_eth_fq *fq __always_unused)
+{
+	struct device *dev = priv->net_dev->dev.parent;
+	dma_addr_t addr = dpaa2_fd_get_addr(fd);
+	u8 fd_format = dpaa2_fd_get_format(fd);
+	struct rtnl_link_stats64 *percpu_stats;
+	struct dpaa2_eth_trap_item *trap_item;
+	struct dpaa2_fapr *fapr;
+	struct sk_buff *skb;
+	void *buf_data;
+	void *vaddr;
+
+	vaddr = dpaa2_iova_to_virt(priv->iommu_domain, addr);
+	dma_sync_single_for_cpu(dev, addr, priv->rx_buf_size,
+				DMA_BIDIRECTIONAL);
+
+	buf_data = vaddr + dpaa2_fd_get_offset(fd);
+
+	if (fd_format == dpaa2_fd_single) {
+		dma_unmap_page(dev, addr, priv->rx_buf_size,
+			       DMA_BIDIRECTIONAL);
+		skb = dpaa2_eth_build_linear_skb(ch, fd, vaddr);
+	} else if (fd_format == dpaa2_fd_sg) {
+		dma_unmap_page(dev, addr, priv->rx_buf_size,
+			       DMA_BIDIRECTIONAL);
+		skb = dpaa2_eth_build_frag_skb(priv, ch, buf_data);
+		free_pages((unsigned long)vaddr, 0);
+	} else {
+		/* We don't support any other format */
+		dpaa2_eth_free_rx_fd(priv, fd, vaddr);
+		goto err_frame_format;
+	}
+
+	fapr = dpaa2_get_fapr(vaddr, false);
+	trap_item = dpaa2_eth_dl_get_trap(priv, fapr);
+	if (trap_item)
+		devlink_trap_report(priv->devlink, skb, trap_item->trap_ctx,
+				    &priv->devlink_port, NULL);
+	consume_skb(skb);
+
+err_frame_format:
+	percpu_stats = this_cpu_ptr(priv->percpu_stats);
+	percpu_stats->rx_errors++;
+	ch->buf_count--;
+}
+
+/* Consume all frames pull-dequeued into the store. This is the simplest way to
+ * make sure we don't accidentally issue another volatile dequeue which would
+ * overwrite (leak) frames already in the store.
+ *
+ * Observance of NAPI budget is not our concern, leaving that to the caller.
+ */
+static int dpaa2_eth_consume_frames(struct dpaa2_eth_channel *ch,
+				    struct dpaa2_eth_fq **src)
+{
+	struct dpaa2_eth_priv *priv = ch->priv;
+	struct dpaa2_eth_fq *fq = NULL;
+	struct dpaa2_dq *dq;
+	const struct dpaa2_fd *fd;
+	int cleaned = 0, retries = 0;
+	int is_last;
+
+	do {
+		dq = dpaa2_io_store_next(ch->store, &is_last);
+		if (unlikely(!dq)) {
+			/* If we're here, we *must* have placed a
+			 * volatile dequeue comnmand, so keep reading through
+			 * the store until we get some sort of valid response
+			 * token (either a valid frame or an "empty dequeue")
+			 */
+			if (retries++ >= DPAA2_ETH_SWP_BUSY_RETRIES) {
+				netdev_err_once(priv->net_dev,
+						"Unable to read a valid dequeue response\n");
+				return -ETIMEDOUT;
+			}
+			continue;
+		}
+
+		fd = dpaa2_dq_fd(dq);
+		fq = (struct dpaa2_eth_fq *)(uintptr_t)dpaa2_dq_fqd_ctx(dq);
+
+		fq->consume(priv, ch, fd, fq);
+		cleaned++;
+		retries = 0;
+	} while (!is_last);
+
+	if (!cleaned)
+		return 0;
+
+	fq->stats.frames += cleaned;
+	ch->stats.frames += cleaned;
+	ch->stats.frames_per_cdan += cleaned;
+
+	/* A dequeue operation only pulls frames from a single queue
+	 * into the store. Return the frame queue as an out param.
+	 */
+	if (src)
+		*src = fq;
+
+	return cleaned;
+}
+
+static int dpaa2_eth_ptp_parse(struct sk_buff *skb,
+			       u8 *msgtype, u8 *twostep, u8 *udp,
+			       u16 *correction_offset,
+			       u16 *origintimestamp_offset)
+{
+	unsigned int ptp_class;
+	struct ptp_header *hdr;
+	unsigned int type;
+	u8 *base;
+
+	ptp_class = ptp_classify_raw(skb);
+	if (ptp_class == PTP_CLASS_NONE)
+		return -EINVAL;
+
+	hdr = ptp_parse_header(skb, ptp_class);
+	if (!hdr)
+		return -EINVAL;
+
+	*msgtype = ptp_get_msgtype(hdr, ptp_class);
+	*twostep = hdr->flag_field[0] & 0x2;
+
+	type = ptp_class & PTP_CLASS_PMASK;
+	if (type == PTP_CLASS_IPV4 ||
+	    type == PTP_CLASS_IPV6)
+		*udp = 1;
+	else
+		*udp = 0;
+
+	base = skb_mac_header(skb);
+	*correction_offset = (u8 *)&hdr->correction - base;
+	*origintimestamp_offset = (u8 *)hdr + sizeof(struct ptp_header) - base;
+
+	return 0;
+}
+
+/* Configure the egress frame annotation for timestamp update */
+static void dpaa2_eth_enable_tx_tstamp(struct dpaa2_eth_priv *priv,
+				       struct dpaa2_fd *fd,
+				       void *buf_start,
+				       struct sk_buff *skb)
+{
+	struct ptp_tstamp origin_timestamp;
+	u8 msgtype, twostep, udp;
+	struct dpaa2_faead *faead;
+	struct dpaa2_fas *fas;
+	struct timespec64 ts;
+	u16 offset1, offset2;
+	u32 ctrl, frc;
+	__le64 *ns;
+	u8 *data;
+
+	/* Mark the egress frame annotation area as valid */
+	frc = dpaa2_fd_get_frc(fd);
+	dpaa2_fd_set_frc(fd, frc | DPAA2_FD_FRC_FAEADV);
+
+	/* Set hardware annotation size */
+	ctrl = dpaa2_fd_get_ctrl(fd);
+	dpaa2_fd_set_ctrl(fd, ctrl | DPAA2_FD_CTRL_ASAL);
+
+	/* enable UPD (update prepanded data) bit in FAEAD field of
+	 * hardware frame annotation area
+	 */
+	ctrl = DPAA2_FAEAD_A2V | DPAA2_FAEAD_UPDV | DPAA2_FAEAD_UPD;
+	faead = dpaa2_get_faead(buf_start, true);
+	faead->ctrl = cpu_to_le32(ctrl);
+
+	if (skb->cb[0] == TX_TSTAMP_ONESTEP_SYNC) {
+		if (dpaa2_eth_ptp_parse(skb, &msgtype, &twostep, &udp,
+					&offset1, &offset2) ||
+		    msgtype != PTP_MSGTYPE_SYNC || twostep) {
+			WARN_ONCE(1, "Bad packet for one-step timestamping\n");
+			return;
+		}
+
+		/* Mark the frame annotation status as valid */
+		frc = dpaa2_fd_get_frc(fd);
+		dpaa2_fd_set_frc(fd, frc | DPAA2_FD_FRC_FASV);
+
+		/* Mark the PTP flag for one step timestamping */
+		fas = dpaa2_get_fas(buf_start, true);
+		fas->status = cpu_to_le32(DPAA2_FAS_PTP);
+
+		dpaa2_ptp->caps.gettime64(&dpaa2_ptp->caps, &ts);
+		ns = dpaa2_get_ts(buf_start, true);
+		*ns = cpu_to_le64(timespec64_to_ns(&ts) /
+				  DPAA2_PTP_CLK_PERIOD_NS);
+
+		/* Update current time to PTP message originTimestamp field */
+		ns_to_ptp_tstamp(&origin_timestamp, le64_to_cpup(ns));
+		data = skb_mac_header(skb);
+		*(__be16 *)(data + offset2) = htons(origin_timestamp.sec_msb);
+		*(__be32 *)(data + offset2 + 2) =
+			htonl(origin_timestamp.sec_lsb);
+		*(__be32 *)(data + offset2 + 6) = htonl(origin_timestamp.nsec);
+
+		if (priv->ptp_correction_off == offset1)
+			return;
+
+		priv->dpaa2_set_onestep_params_cb(priv, offset1, udp);
+		priv->ptp_correction_off = offset1;
+
+	}
+}
+
+void *dpaa2_eth_sgt_get(struct dpaa2_eth_priv *priv)
+{
+	struct dpaa2_eth_sgt_cache *sgt_cache;
+	void *sgt_buf = NULL;
+	int sgt_buf_size;
+
+	sgt_cache = this_cpu_ptr(priv->sgt_cache);
+	sgt_buf_size = priv->tx_data_offset +
+		DPAA2_ETH_SG_ENTRIES_MAX * sizeof(struct dpaa2_sg_entry);
+
+	if (sgt_cache->count == 0)
+		sgt_buf = napi_alloc_frag_align(sgt_buf_size, DPAA2_ETH_TX_BUF_ALIGN);
+	else
+		sgt_buf = sgt_cache->buf[--sgt_cache->count];
+	if (!sgt_buf)
+		return NULL;
+
+	memset(sgt_buf, 0, sgt_buf_size);
+
+	return sgt_buf;
+}
+
+void dpaa2_eth_sgt_recycle(struct dpaa2_eth_priv *priv, void *sgt_buf)
+{
+	struct dpaa2_eth_sgt_cache *sgt_cache;
+
+	sgt_cache = this_cpu_ptr(priv->sgt_cache);
+	if (sgt_cache->count >= DPAA2_ETH_SGT_CACHE_SIZE)
+		skb_free_frag(sgt_buf);
+	else
+		sgt_cache->buf[sgt_cache->count++] = sgt_buf;
+}
+
+/* Create a frame descriptor based on a fragmented skb */
+static int dpaa2_eth_build_sg_fd(struct dpaa2_eth_priv *priv,
+				 struct sk_buff *skb,
+				 struct dpaa2_fd *fd,
+				 void **swa_addr)
+{
+	struct device *dev = priv->net_dev->dev.parent;
+	void *sgt_buf = NULL;
+	dma_addr_t addr;
+	int nr_frags = skb_shinfo(skb)->nr_frags;
+	struct dpaa2_sg_entry *sgt;
+	int i, err;
+	int sgt_buf_size;
+	struct scatterlist *scl, *crt_scl;
+	int num_sg;
+	int num_dma_bufs;
+	struct dpaa2_eth_swa *swa;
+
+	/* Create and map scatterlist.
+	 * We don't advertise NETIF_F_FRAGLIST, so skb_to_sgvec() will not have
+	 * to go beyond nr_frags+1.
+	 * Note: We don't support chained scatterlists
+	 */
+	if (unlikely(PAGE_SIZE / sizeof(struct scatterlist) < nr_frags + 1))
+		return -EINVAL;
+
+	scl = kmalloc_array(nr_frags + 1, sizeof(struct scatterlist), GFP_ATOMIC);
+	if (unlikely(!scl))
+		return -ENOMEM;
+
+	sg_init_table(scl, nr_frags + 1);
+	num_sg = skb_to_sgvec(skb, scl, 0, skb->len);
+	if (unlikely(num_sg < 0)) {
+		err = -ENOMEM;
+		goto dma_map_sg_failed;
+	}
+	num_dma_bufs = dma_map_sg(dev, scl, num_sg, DMA_BIDIRECTIONAL);
+	if (unlikely(!num_dma_bufs)) {
+		err = -ENOMEM;
+		goto dma_map_sg_failed;
+	}
+
+	/* Prepare the HW SGT structure */
+	sgt_buf_size = priv->tx_data_offset +
+		       sizeof(struct dpaa2_sg_entry) *  num_dma_bufs;
+	sgt_buf = dpaa2_eth_sgt_get(priv);
+	if (unlikely(!sgt_buf)) {
+		err = -ENOMEM;
+		goto sgt_buf_alloc_failed;
+	}
+
+	sgt = (struct dpaa2_sg_entry *)(sgt_buf + priv->tx_data_offset);
+
+	/* Fill in the HW SGT structure.
+	 *
+	 * sgt_buf is zeroed out, so the following fields are implicit
+	 * in all sgt entries:
+	 *   - offset is 0
+	 *   - format is 'dpaa2_sg_single'
+	 */
+	for_each_sg(scl, crt_scl, num_dma_bufs, i) {
+		dpaa2_sg_set_addr(&sgt[i], sg_dma_address(crt_scl));
+		dpaa2_sg_set_len(&sgt[i], sg_dma_len(crt_scl));
+	}
+	dpaa2_sg_set_final(&sgt[i - 1], true);
+
+	/* Store the skb backpointer in the SGT buffer.
+	 * Fit the scatterlist and the number of buffers alongside the
+	 * skb backpointer in the software annotation area. We'll need
+	 * all of them on Tx Conf.
+	 */
+	*swa_addr = (void *)sgt_buf;
+	swa = (struct dpaa2_eth_swa *)sgt_buf;
+	swa->type = DPAA2_ETH_SWA_SG;
+	swa->sg.skb = skb;
+	swa->sg.scl = scl;
+	swa->sg.num_sg = num_sg;
+	swa->sg.sgt_size = sgt_buf_size;
+
+	/* Separately map the SGT buffer */
+	addr = dma_map_single(dev, sgt_buf, sgt_buf_size, DMA_BIDIRECTIONAL);
+	if (unlikely(dma_mapping_error(dev, addr))) {
+		err = -ENOMEM;
+		goto dma_map_single_failed;
+	}
+	memset(fd, 0, sizeof(struct dpaa2_fd));
+	dpaa2_fd_set_offset(fd, priv->tx_data_offset);
+	dpaa2_fd_set_format(fd, dpaa2_fd_sg);
+	dpaa2_fd_set_addr(fd, addr);
+	dpaa2_fd_set_len(fd, skb->len);
+	dpaa2_fd_set_ctrl(fd, FD_CTRL_PTA);
+
+	return 0;
+
+dma_map_single_failed:
+	dpaa2_eth_sgt_recycle(priv, sgt_buf);
+sgt_buf_alloc_failed:
+	dma_unmap_sg(dev, scl, num_sg, DMA_BIDIRECTIONAL);
+dma_map_sg_failed:
+	kfree(scl);
+	return err;
+}
+
+/* Create a SG frame descriptor based on a linear skb.
+ *
+ * This function is used on the Tx path when the skb headroom is not large
+ * enough for the HW requirements, thus instead of realloc-ing the skb we
+ * create a SG frame descriptor with only one entry.
+ */
+static int dpaa2_eth_build_sg_fd_single_buf(struct dpaa2_eth_priv *priv,
+					    struct sk_buff *skb,
+					    struct dpaa2_fd *fd,
+					    void **swa_addr)
+{
+	struct device *dev = priv->net_dev->dev.parent;
+	struct dpaa2_sg_entry *sgt;
+	struct dpaa2_eth_swa *swa;
+	dma_addr_t addr, sgt_addr;
+	void *sgt_buf = NULL;
+	int sgt_buf_size;
+	int err;
+
+	/* Prepare the HW SGT structure */
+	sgt_buf_size = priv->tx_data_offset + sizeof(struct dpaa2_sg_entry);
+	sgt_buf = dpaa2_eth_sgt_get(priv);
+	if (unlikely(!sgt_buf))
+		return -ENOMEM;
+	sgt = (struct dpaa2_sg_entry *)(sgt_buf + priv->tx_data_offset);
+
+	addr = dma_map_single(dev, skb->data, skb->len, DMA_BIDIRECTIONAL);
+	if (unlikely(dma_mapping_error(dev, addr))) {
+		err = -ENOMEM;
+		goto data_map_failed;
+	}
+
+	/* Fill in the HW SGT structure */
+	dpaa2_sg_set_addr(sgt, addr);
+	dpaa2_sg_set_len(sgt, skb->len);
+	dpaa2_sg_set_final(sgt, true);
+
+	/* Store the skb backpointer in the SGT buffer */
+	*swa_addr = (void *)sgt_buf;
+	swa = (struct dpaa2_eth_swa *)sgt_buf;
+	swa->type = DPAA2_ETH_SWA_SINGLE;
+	swa->single.skb = skb;
+	swa->single.sgt_size = sgt_buf_size;
+
+	/* Separately map the SGT buffer */
+	sgt_addr = dma_map_single(dev, sgt_buf, sgt_buf_size, DMA_BIDIRECTIONAL);
+	if (unlikely(dma_mapping_error(dev, sgt_addr))) {
+		err = -ENOMEM;
+		goto sgt_map_failed;
+	}
+
+	memset(fd, 0, sizeof(struct dpaa2_fd));
+	dpaa2_fd_set_offset(fd, priv->tx_data_offset);
+	dpaa2_fd_set_format(fd, dpaa2_fd_sg);
+	dpaa2_fd_set_addr(fd, sgt_addr);
+	dpaa2_fd_set_len(fd, skb->len);
+	dpaa2_fd_set_ctrl(fd, FD_CTRL_PTA);
+
+	return 0;
+
+sgt_map_failed:
+	dma_unmap_single(dev, addr, skb->len, DMA_BIDIRECTIONAL);
+data_map_failed:
+	dpaa2_eth_sgt_recycle(priv, sgt_buf);
+
+	return err;
+}
+
+/* Create a frame descriptor based on a linear skb */
+static int dpaa2_eth_build_single_fd(struct dpaa2_eth_priv *priv,
+				     struct sk_buff *skb,
+				     struct dpaa2_fd *fd,
+				     void **swa_addr)
+{
+	struct device *dev = priv->net_dev->dev.parent;
+	u8 *buffer_start, *aligned_start;
+	struct dpaa2_eth_swa *swa;
+	dma_addr_t addr;
+
+	buffer_start = skb->data - dpaa2_eth_needed_headroom(skb);
+	aligned_start = PTR_ALIGN(buffer_start - DPAA2_ETH_TX_BUF_ALIGN,
+				  DPAA2_ETH_TX_BUF_ALIGN);
+	if (aligned_start >= skb->head)
+		buffer_start = aligned_start;
+	else
+		return -ENOMEM;
+
+	/* Store a backpointer to the skb at the beginning of the buffer
+	 * (in the private data area) such that we can release it
+	 * on Tx confirm
+	 */
+	*swa_addr = (void *)buffer_start;
+	swa = (struct dpaa2_eth_swa *)buffer_start;
+	swa->type = DPAA2_ETH_SWA_SINGLE;
+	swa->single.skb = skb;
+
+	addr = dma_map_single(dev, buffer_start,
+			      skb_tail_pointer(skb) - buffer_start,
+			      DMA_BIDIRECTIONAL);
+	if (unlikely(dma_mapping_error(dev, addr)))
+		return -ENOMEM;
+
+	memset(fd, 0, sizeof(struct dpaa2_fd));
+	dpaa2_fd_set_addr(fd, addr);
+	dpaa2_fd_set_offset(fd, (u16)(skb->data - buffer_start));
+	dpaa2_fd_set_len(fd, skb->len);
+	dpaa2_fd_set_format(fd, dpaa2_fd_single);
+	dpaa2_fd_set_ctrl(fd, FD_CTRL_PTA);
+
+	return 0;
+}
+
+/* FD freeing routine on the Tx path
+ *
+ * DMA-unmap and free FD and possibly SGT buffer allocated on Tx. The skb
+ * back-pointed to is also freed.
+ * This can be called either from dpaa2_eth_tx_conf() or on the error path of
+ * dpaa2_eth_tx().
+ */
+void dpaa2_eth_free_tx_fd(struct dpaa2_eth_priv *priv,
+			  struct dpaa2_eth_channel *ch,
+			  struct dpaa2_eth_fq *fq,
+			  const struct dpaa2_fd *fd, bool in_napi)
+{
+	struct device *dev = priv->net_dev->dev.parent;
+	dma_addr_t fd_addr, sg_addr;
+	struct sk_buff *skb = NULL;
+	unsigned char *buffer_start;
+	struct dpaa2_eth_swa *swa;
+	u8 fd_format = dpaa2_fd_get_format(fd);
+	u32 fd_len = dpaa2_fd_get_len(fd);
+	struct dpaa2_sg_entry *sgt;
+	int should_free_skb = 1;
+	void *tso_hdr;
+	int i;
+
+	fd_addr = dpaa2_fd_get_addr(fd);
+	buffer_start = dpaa2_iova_to_virt(priv->iommu_domain, fd_addr);
+	swa = (struct dpaa2_eth_swa *)buffer_start;
+
+	if (fd_format == dpaa2_fd_single) {
+		if (swa->type == DPAA2_ETH_SWA_SINGLE) {
+			skb = swa->single.skb;
+			/* Accessing the skb buffer is safe before dma unmap,
+			 * because we didn't map the actual skb shell.
+			 */
+			dma_unmap_single(dev, fd_addr,
+					 skb_tail_pointer(skb) - buffer_start,
+					 DMA_BIDIRECTIONAL);
+		} else {
+			WARN_ONCE(swa->type != DPAA2_ETH_SWA_XDP, "Wrong SWA type");
+			dma_unmap_single(dev, fd_addr, swa->xdp.dma_size,
+					 DMA_BIDIRECTIONAL);
+		}
+	} else if (fd_format == dpaa2_fd_sg) {
+		if (swa->type == DPAA2_ETH_SWA_SG) {
+			skb = swa->sg.skb;
+
+			/* Unmap the scatterlist */
+			dma_unmap_sg(dev, swa->sg.scl, swa->sg.num_sg,
+				     DMA_BIDIRECTIONAL);
+			kfree(swa->sg.scl);
+
+			/* Unmap the SGT buffer */
+			dma_unmap_single(dev, fd_addr, swa->sg.sgt_size,
+					 DMA_BIDIRECTIONAL);
+		} else if (swa->type == DPAA2_ETH_SWA_SW_TSO) {
+			skb = swa->tso.skb;
+
+			sgt = (struct dpaa2_sg_entry *)(buffer_start +
+							priv->tx_data_offset);
+
+			/* Unmap the SGT buffer */
+			dma_unmap_single(dev, fd_addr, swa->tso.sgt_size,
+					 DMA_BIDIRECTIONAL);
+
+			/* Unmap and free the header */
+			tso_hdr = dpaa2_iova_to_virt(priv->iommu_domain, dpaa2_sg_get_addr(sgt));
+			dma_unmap_single(dev, dpaa2_sg_get_addr(sgt), TSO_HEADER_SIZE,
+					 DMA_TO_DEVICE);
+			kfree(tso_hdr);
+
+			/* Unmap the other SG entries for the data */
+			for (i = 1; i < swa->tso.num_sg; i++)
+				dma_unmap_single(dev, dpaa2_sg_get_addr(&sgt[i]),
+						 dpaa2_sg_get_len(&sgt[i]), DMA_TO_DEVICE);
+
+			if (!swa->tso.is_last_fd)
+				should_free_skb = 0;
+		} else if (swa->type == DPAA2_ETH_SWA_XSK) {
+			/* Unmap the SGT Buffer */
+			dma_unmap_single(dev, fd_addr, swa->xsk.sgt_size,
+					 DMA_BIDIRECTIONAL);
+		} else {
+			skb = swa->single.skb;
+
+			/* Unmap the SGT Buffer */
+			dma_unmap_single(dev, fd_addr, swa->single.sgt_size,
+					 DMA_BIDIRECTIONAL);
+
+			sgt = (struct dpaa2_sg_entry *)(buffer_start +
+							priv->tx_data_offset);
+			sg_addr = dpaa2_sg_get_addr(sgt);
+			dma_unmap_single(dev, sg_addr, skb->len, DMA_BIDIRECTIONAL);
+		}
+	} else {
+		netdev_dbg(priv->net_dev, "Invalid FD format\n");
+		return;
+	}
+
+	if (swa->type == DPAA2_ETH_SWA_XSK) {
+		ch->xsk_tx_pkts_sent++;
+		dpaa2_eth_sgt_recycle(priv, buffer_start);
+		return;
+	}
+
+	if (swa->type != DPAA2_ETH_SWA_XDP && in_napi) {
+		fq->dq_frames++;
+		fq->dq_bytes += fd_len;
+	}
+
+	if (swa->type == DPAA2_ETH_SWA_XDP) {
+		xdp_return_frame(swa->xdp.xdpf);
+		return;
+	}
+
+	/* Get the timestamp value */
+	if (swa->type != DPAA2_ETH_SWA_SW_TSO) {
+		if (skb->cb[0] == TX_TSTAMP) {
+			struct skb_shared_hwtstamps shhwtstamps;
+			__le64 *ts = dpaa2_get_ts(buffer_start, true);
+			u64 ns;
+
+			memset(&shhwtstamps, 0, sizeof(shhwtstamps));
+
+			ns = DPAA2_PTP_CLK_PERIOD_NS * le64_to_cpup(ts);
+			shhwtstamps.hwtstamp = ns_to_ktime(ns);
+			skb_tstamp_tx(skb, &shhwtstamps);
+		} else if (skb->cb[0] == TX_TSTAMP_ONESTEP_SYNC) {
+			mutex_unlock(&priv->onestep_tstamp_lock);
+		}
+	}
+
+	/* Free SGT buffer allocated on tx */
+	if (fd_format != dpaa2_fd_single)
+		dpaa2_eth_sgt_recycle(priv, buffer_start);
+
+	/* Move on with skb release. If we are just confirming multiple FDs
+	 * from the same TSO skb then only the last one will need to free the
+	 * skb.
+	 */
+	if (should_free_skb)
+		napi_consume_skb(skb, in_napi);
+}
+
+static int dpaa2_eth_build_gso_fd(struct dpaa2_eth_priv *priv,
+				  struct sk_buff *skb, struct dpaa2_fd *fd,
+				  int *num_fds, u32 *total_fds_len)
+{
+	struct device *dev = priv->net_dev->dev.parent;
+	int hdr_len, total_len, data_left, fd_len;
+	int num_sge, err, i, sgt_buf_size;
+	struct dpaa2_fd *fd_start = fd;
+	struct dpaa2_sg_entry *sgt;
+	struct dpaa2_eth_swa *swa;
+	dma_addr_t sgt_addr, addr;
+	dma_addr_t tso_hdr_dma;
+	unsigned int index = 0;
+	struct tso_t tso;
+	char *tso_hdr;
+	void *sgt_buf;
+
+	/* Initialize the TSO handler, and prepare the first payload */
+	hdr_len = tso_start(skb, &tso);
+	*total_fds_len = 0;
+
+	total_len = skb->len - hdr_len;
+	while (total_len > 0) {
+		/* Prepare the HW SGT structure for this frame */
+		sgt_buf = dpaa2_eth_sgt_get(priv);
+		if (unlikely(!sgt_buf)) {
+			netdev_err(priv->net_dev, "dpaa2_eth_sgt_get() failed\n");
+			err = -ENOMEM;
+			goto err_sgt_get;
+		}
+		sgt = (struct dpaa2_sg_entry *)(sgt_buf + priv->tx_data_offset);
+
+		/* Determine the data length of this frame */
+		data_left = min_t(int, skb_shinfo(skb)->gso_size, total_len);
+		total_len -= data_left;
+		fd_len = data_left + hdr_len;
+
+		/* Prepare packet headers: MAC + IP + TCP */
+		tso_hdr = kmalloc(TSO_HEADER_SIZE, GFP_ATOMIC);
+		if (!tso_hdr) {
+			err =  -ENOMEM;
+			goto err_alloc_tso_hdr;
+		}
+
+		tso_build_hdr(skb, tso_hdr, &tso, data_left, total_len == 0);
+		tso_hdr_dma = dma_map_single(dev, tso_hdr, TSO_HEADER_SIZE, DMA_TO_DEVICE);
+		if (dma_mapping_error(dev, tso_hdr_dma)) {
+			netdev_err(priv->net_dev, "dma_map_single(tso_hdr) failed\n");
+			err = -ENOMEM;
+			goto err_map_tso_hdr;
+		}
+
+		/* Setup the SG entry for the header */
+		dpaa2_sg_set_addr(sgt, tso_hdr_dma);
+		dpaa2_sg_set_len(sgt, hdr_len);
+		dpaa2_sg_set_final(sgt, data_left <= 0);
+
+		/* Compose the SG entries for each fragment of data */
+		num_sge = 1;
+		while (data_left > 0) {
+			int size;
+
+			/* Move to the next SG entry */
+			sgt++;
+			size = min_t(int, tso.size, data_left);
+
+			addr = dma_map_single(dev, tso.data, size, DMA_TO_DEVICE);
+			if (dma_mapping_error(dev, addr)) {
+				netdev_err(priv->net_dev, "dma_map_single(tso.data) failed\n");
+				err = -ENOMEM;
+				goto err_map_data;
+			}
+			dpaa2_sg_set_addr(sgt, addr);
+			dpaa2_sg_set_len(sgt, size);
+			dpaa2_sg_set_final(sgt, size == data_left);
+
+			num_sge++;
+
+			/* Build the data for the __next__ fragment */
+			data_left -= size;
+			tso_build_data(skb, &tso, size);
+		}
+
+		/* Store the skb backpointer in the SGT buffer */
+		sgt_buf_size = priv->tx_data_offset + num_sge * sizeof(struct dpaa2_sg_entry);
+		swa = (struct dpaa2_eth_swa *)sgt_buf;
+		swa->type = DPAA2_ETH_SWA_SW_TSO;
+		swa->tso.skb = skb;
+		swa->tso.num_sg = num_sge;
+		swa->tso.sgt_size = sgt_buf_size;
+		swa->tso.is_last_fd = total_len == 0 ? 1 : 0;
+
+		/* Separately map the SGT buffer */
+		sgt_addr = dma_map_single(dev, sgt_buf, sgt_buf_size, DMA_BIDIRECTIONAL);
+		if (unlikely(dma_mapping_error(dev, sgt_addr))) {
+			netdev_err(priv->net_dev, "dma_map_single(sgt_buf) failed\n");
+			err = -ENOMEM;
+			goto err_map_sgt;
+		}
+
+		/* Setup the frame descriptor */
+		memset(fd, 0, sizeof(struct dpaa2_fd));
+		dpaa2_fd_set_offset(fd, priv->tx_data_offset);
+		dpaa2_fd_set_format(fd, dpaa2_fd_sg);
+		dpaa2_fd_set_addr(fd, sgt_addr);
+		dpaa2_fd_set_len(fd, fd_len);
+		dpaa2_fd_set_ctrl(fd, FD_CTRL_PTA);
+
+		*total_fds_len += fd_len;
+		/* Advance to the next frame descriptor */
+		fd++;
+		index++;
+	}
+
+	*num_fds = index;
+
+	return 0;
+
+err_map_sgt:
+err_map_data:
+	/* Unmap all the data S/G entries for the current FD */
+	sgt = (struct dpaa2_sg_entry *)(sgt_buf + priv->tx_data_offset);
+	for (i = 1; i < num_sge; i++)
+		dma_unmap_single(dev, dpaa2_sg_get_addr(&sgt[i]),
+				 dpaa2_sg_get_len(&sgt[i]), DMA_TO_DEVICE);
+
+	/* Unmap the header entry */
+	dma_unmap_single(dev, tso_hdr_dma, TSO_HEADER_SIZE, DMA_TO_DEVICE);
+err_map_tso_hdr:
+	kfree(tso_hdr);
+err_alloc_tso_hdr:
+	dpaa2_eth_sgt_recycle(priv, sgt_buf);
+err_sgt_get:
+	/* Free all the other FDs that were already fully created */
+	for (i = 0; i < index; i++)
+		dpaa2_eth_free_tx_fd(priv, NULL, NULL, &fd_start[i], false);
+
+	return err;
+}
+
+static netdev_tx_t __dpaa2_eth_tx(struct sk_buff *skb,
+				  struct net_device *net_dev)
+{
+	struct dpaa2_eth_priv *priv = netdev_priv(net_dev);
+	int total_enqueued = 0, retries = 0, enqueued;
+	struct dpaa2_eth_drv_stats *percpu_extras;
+	struct rtnl_link_stats64 *percpu_stats;
+	unsigned int needed_headroom;
+	int num_fds = 1, max_retries;
+	struct dpaa2_eth_fq *fq;
+	struct netdev_queue *nq;
+	struct dpaa2_fd *fd;
+	u16 queue_mapping;
+	void *swa = NULL;
+	u8 prio = 0;
+	int err, i;
+	u32 fd_len;
+
+	percpu_stats = this_cpu_ptr(priv->percpu_stats);
+	percpu_extras = this_cpu_ptr(priv->percpu_extras);
+	fd = (this_cpu_ptr(priv->fd))->array;
+
+	needed_headroom = dpaa2_eth_needed_headroom(skb);
+
+	/* We'll be holding a back-reference to the skb until Tx Confirmation;
+	 * we don't want that overwritten by a concurrent Tx with a cloned skb.
+	 */
+	skb = skb_unshare(skb, GFP_ATOMIC);
+	if (unlikely(!skb)) {
+		/* skb_unshare() has already freed the skb */
+		percpu_stats->tx_dropped++;
+		return NETDEV_TX_OK;
+	}
+
+	/* Setup the FD fields */
+
+	if (skb_is_gso(skb)) {
+		err = dpaa2_eth_build_gso_fd(priv, skb, fd, &num_fds, &fd_len);
+		percpu_extras->tx_sg_frames += num_fds;
+		percpu_extras->tx_sg_bytes += fd_len;
+		percpu_extras->tx_tso_frames += num_fds;
+		percpu_extras->tx_tso_bytes += fd_len;
+	} else if (skb_is_nonlinear(skb)) {
+		err = dpaa2_eth_build_sg_fd(priv, skb, fd, &swa);
+		percpu_extras->tx_sg_frames++;
+		percpu_extras->tx_sg_bytes += skb->len;
+		fd_len = dpaa2_fd_get_len(fd);
+	} else if (skb_headroom(skb) < needed_headroom) {
+		err = dpaa2_eth_build_sg_fd_single_buf(priv, skb, fd, &swa);
+		percpu_extras->tx_sg_frames++;
+		percpu_extras->tx_sg_bytes += skb->len;
+		percpu_extras->tx_converted_sg_frames++;
+		percpu_extras->tx_converted_sg_bytes += skb->len;
+		fd_len = dpaa2_fd_get_len(fd);
+	} else {
+		err = dpaa2_eth_build_single_fd(priv, skb, fd, &swa);
+		fd_len = dpaa2_fd_get_len(fd);
+	}
+
+	if (unlikely(err)) {
+		percpu_stats->tx_dropped++;
+		goto err_build_fd;
+	}
+
+	if (swa && skb->cb[0])
+		dpaa2_eth_enable_tx_tstamp(priv, fd, swa, skb);
+
+	/* Tracing point */
+	for (i = 0; i < num_fds; i++)
+		trace_dpaa2_tx_fd(net_dev, &fd[i]);
+
+	/* TxConf FQ selection relies on queue id from the stack.
+	 * In case of a forwarded frame from another DPNI interface, we choose
+	 * a queue affined to the same core that processed the Rx frame
+	 */
+	queue_mapping = skb_get_queue_mapping(skb);
+
+	if (net_dev->num_tc) {
+		prio = netdev_txq_to_tc(net_dev, queue_mapping);
+		/* Hardware interprets priority level 0 as being the highest,
+		 * so we need to do a reverse mapping to the netdev tc index
+		 */
+		prio = net_dev->num_tc - prio - 1;
+		/* We have only one FQ array entry for all Tx hardware queues
+		 * with the same flow id (but different priority levels)
+		 */
+		queue_mapping %= dpaa2_eth_queue_count(priv);
+	}
+	fq = &priv->fq[queue_mapping];
+	nq = netdev_get_tx_queue(net_dev, queue_mapping);
+	netdev_tx_sent_queue(nq, fd_len);
+
+	/* Everything that happens after this enqueues might race with
+	 * the Tx confirmation callback for this frame
+	 */
+	max_retries = num_fds * DPAA2_ETH_ENQUEUE_RETRIES;
+	while (total_enqueued < num_fds && retries < max_retries) {
+		err = priv->enqueue(priv, fq, &fd[total_enqueued],
+				    prio, num_fds - total_enqueued, &enqueued);
+		if (err == -EBUSY) {
+			retries++;
+			continue;
+		}
+
+		total_enqueued += enqueued;
+	}
+	percpu_extras->tx_portal_busy += retries;
+
+	if (unlikely(err < 0)) {
+		percpu_stats->tx_errors++;
+		/* Clean up everything, including freeing the skb */
+		dpaa2_eth_free_tx_fd(priv, NULL, fq, fd, false);
+		netdev_tx_completed_queue(nq, 1, fd_len);
+	} else {
+		percpu_stats->tx_packets += total_enqueued;
+		percpu_stats->tx_bytes += fd_len;
+	}
+
+	return NETDEV_TX_OK;
+
+err_build_fd:
+	dev_kfree_skb(skb);
+
+	return NETDEV_TX_OK;
+}
+
+static void dpaa2_eth_tx_onestep_tstamp(struct work_struct *work)
+{
+	struct dpaa2_eth_priv *priv = container_of(work, struct dpaa2_eth_priv,
+						   tx_onestep_tstamp);
+	struct sk_buff *skb;
+
+	while (true) {
+		skb = skb_dequeue(&priv->tx_skbs);
+		if (!skb)
+			return;
+
+		/* Lock just before TX one-step timestamping packet,
+		 * and release the lock in dpaa2_eth_free_tx_fd when
+		 * confirm the packet has been sent on hardware, or
+		 * when clean up during transmit failure.
+		 */
+		mutex_lock(&priv->onestep_tstamp_lock);
+		__dpaa2_eth_tx(skb, priv->net_dev);
+	}
+}
+
+static netdev_tx_t dpaa2_eth_tx(struct sk_buff *skb, struct net_device *net_dev)
+{
+	struct dpaa2_eth_priv *priv = netdev_priv(net_dev);
+	u8 msgtype, twostep, udp;
+	u16 offset1, offset2;
+
+	/* Utilize skb->cb[0] for timestamping request per skb */
+	skb->cb[0] = 0;
+
+	if ((skb_shinfo(skb)->tx_flags & SKBTX_HW_TSTAMP) && dpaa2_ptp) {
+		if (priv->tx_tstamp_type == HWTSTAMP_TX_ON)
+			skb->cb[0] = TX_TSTAMP;
+		else if (priv->tx_tstamp_type == HWTSTAMP_TX_ONESTEP_SYNC)
+			skb->cb[0] = TX_TSTAMP_ONESTEP_SYNC;
+	}
+
+	/* TX for one-step timestamping PTP Sync packet */
+	if (skb->cb[0] == TX_TSTAMP_ONESTEP_SYNC) {
+		if (!dpaa2_eth_ptp_parse(skb, &msgtype, &twostep, &udp,
+					 &offset1, &offset2))
+			if (msgtype == PTP_MSGTYPE_SYNC && twostep == 0) {
+				skb_queue_tail(&priv->tx_skbs, skb);
+				queue_work(priv->dpaa2_ptp_wq,
+					   &priv->tx_onestep_tstamp);
+				return NETDEV_TX_OK;
+			}
+		/* Use two-step timestamping if not one-step timestamping
+		 * PTP Sync packet
+		 */
+		skb->cb[0] = TX_TSTAMP;
+	}
+
+	/* TX for other packets */
+	return __dpaa2_eth_tx(skb, net_dev);
+}
+
+/* Tx confirmation frame processing routine */
+static void dpaa2_eth_tx_conf(struct dpaa2_eth_priv *priv,
+			      struct dpaa2_eth_channel *ch,
+			      const struct dpaa2_fd *fd,
+			      struct dpaa2_eth_fq *fq)
+{
+	struct rtnl_link_stats64 *percpu_stats;
+	struct dpaa2_eth_drv_stats *percpu_extras;
+	u32 fd_len = dpaa2_fd_get_len(fd);
+	u32 fd_errors;
+
+	/* Tracing point */
+	trace_dpaa2_tx_conf_fd(priv->net_dev, fd);
+
+	percpu_extras = this_cpu_ptr(priv->percpu_extras);
+	percpu_extras->tx_conf_frames++;
+	percpu_extras->tx_conf_bytes += fd_len;
+	ch->stats.bytes_per_cdan += fd_len;
+
+	/* Check frame errors in the FD field */
+	fd_errors = dpaa2_fd_get_ctrl(fd) & DPAA2_FD_TX_ERR_MASK;
+	dpaa2_eth_free_tx_fd(priv, ch, fq, fd, true);
+
+	if (likely(!fd_errors))
+		return;
+
+	if (net_ratelimit())
+		netdev_dbg(priv->net_dev, "TX frame FD error: 0x%08x\n",
+			   fd_errors);
+
+	percpu_stats = this_cpu_ptr(priv->percpu_stats);
+	/* Tx-conf logically pertains to the egress path. */
+	percpu_stats->tx_errors++;
+}
+
+static int dpaa2_eth_set_rx_vlan_filtering(struct dpaa2_eth_priv *priv,
+					   bool enable)
+{
+	int err;
+
+	err = dpni_enable_vlan_filter(priv->mc_io, 0, priv->mc_token, enable);
+
+	if (err) {
+		netdev_err(priv->net_dev,
+			   "dpni_enable_vlan_filter failed\n");
+		return err;
+	}
+
+	return 0;
+}
+
+static int dpaa2_eth_set_rx_csum(struct dpaa2_eth_priv *priv, bool enable)
+{
+	int err;
+
+	err = dpni_set_offload(priv->mc_io, 0, priv->mc_token,
+			       DPNI_OFF_RX_L3_CSUM, enable);
+	if (err) {
+		netdev_err(priv->net_dev,
+			   "dpni_set_offload(RX_L3_CSUM) failed\n");
+		return err;
+	}
+
+	err = dpni_set_offload(priv->mc_io, 0, priv->mc_token,
+			       DPNI_OFF_RX_L4_CSUM, enable);
+	if (err) {
+		netdev_err(priv->net_dev,
+			   "dpni_set_offload(RX_L4_CSUM) failed\n");
+		return err;
+	}
+
+	return 0;
+}
+
+static int dpaa2_eth_set_tx_csum(struct dpaa2_eth_priv *priv, bool enable)
+{
+	int err;
+
+	err = dpni_set_offload(priv->mc_io, 0, priv->mc_token,
+			       DPNI_OFF_TX_L3_CSUM, enable);
+	if (err) {
+		netdev_err(priv->net_dev, "dpni_set_offload(TX_L3_CSUM) failed\n");
+		return err;
+	}
+
+	err = dpni_set_offload(priv->mc_io, 0, priv->mc_token,
+			       DPNI_OFF_TX_L4_CSUM, enable);
+	if (err) {
+		netdev_err(priv->net_dev, "dpni_set_offload(TX_L4_CSUM) failed\n");
+		return err;
+	}
+
+	return 0;
+}
+
+/* Perform a single release command to add buffers
+ * to the specified buffer pool
+ */
+static int dpaa2_eth_add_bufs(struct dpaa2_eth_priv *priv,
+			      struct dpaa2_eth_channel *ch)
+{
+	struct xdp_buff *xdp_buffs[DPAA2_ETH_BUFS_PER_CMD];
+	struct device *dev = priv->net_dev->dev.parent;
+	u64 buf_array[DPAA2_ETH_BUFS_PER_CMD];
+	struct dpaa2_eth_swa *swa;
+	struct page *page;
+	dma_addr_t addr;
+	int retries = 0;
+	int i = 0, err;
+	u32 batch;
+
+	/* Allocate buffers visible to WRIOP */
+	if (!ch->xsk_zc) {
+		for (i = 0; i < DPAA2_ETH_BUFS_PER_CMD; i++) {
+			/* Also allocate skb shared info and alignment padding.
+			 * There is one page for each Rx buffer. WRIOP sees
+			 * the entire page except for a tailroom reserved for
+			 * skb shared info
+			 */
+			page = dev_alloc_pages(0);
+			if (!page)
+				goto err_alloc;
+
+			addr = dma_map_page(dev, page, 0, priv->rx_buf_size,
+					    DMA_BIDIRECTIONAL);
+			if (unlikely(dma_mapping_error(dev, addr)))
+				goto err_map;
+
+			buf_array[i] = addr;
+
+			/* tracing point */
+			trace_dpaa2_eth_buf_seed(priv->net_dev,
+						 page_address(page),
+						 DPAA2_ETH_RX_BUF_RAW_SIZE,
+						 addr, priv->rx_buf_size,
+						 ch->bp->bpid);
+		}
+	} else if (xsk_buff_can_alloc(ch->xsk_pool, DPAA2_ETH_BUFS_PER_CMD)) {
+		/* Allocate XSK buffers for AF_XDP fast path in batches
+		 * of DPAA2_ETH_BUFS_PER_CMD. Bail out if the UMEM cannot
+		 * provide enough buffers at the moment
+		 */
+		batch = xsk_buff_alloc_batch(ch->xsk_pool, xdp_buffs,
+					     DPAA2_ETH_BUFS_PER_CMD);
+		if (!batch)
+			goto err_alloc;
+
+		for (i = 0; i < batch; i++) {
+			swa = (struct dpaa2_eth_swa *)(xdp_buffs[i]->data_hard_start +
+						       DPAA2_ETH_RX_HWA_SIZE);
+			swa->xsk.xdp_buff = xdp_buffs[i];
+
+			addr = xsk_buff_xdp_get_frame_dma(xdp_buffs[i]);
+			if (unlikely(dma_mapping_error(dev, addr)))
+				goto err_map;
+
+			buf_array[i] = addr;
+
+			trace_dpaa2_xsk_buf_seed(priv->net_dev,
+						 xdp_buffs[i]->data_hard_start,
+						 DPAA2_ETH_RX_BUF_RAW_SIZE,
+						 addr, priv->rx_buf_size,
+						 ch->bp->bpid);
+		}
+	}
+
+release_bufs:
+	/* In case the portal is busy, retry until successful */
+	while ((err = dpaa2_io_service_release(ch->dpio, ch->bp->bpid,
+					       buf_array, i)) == -EBUSY) {
+		if (retries++ >= DPAA2_ETH_SWP_BUSY_RETRIES)
+			break;
+		cpu_relax();
+	}
+
+	/* If release command failed, clean up and bail out;
+	 * not much else we can do about it
+	 */
+	if (err) {
+		dpaa2_eth_free_bufs(priv, buf_array, i, ch->xsk_zc);
+		return 0;
+	}
+
+	return i;
+
+err_map:
+	if (!ch->xsk_zc) {
+		__free_pages(page, 0);
+	} else {
+		for (; i < batch; i++)
+			xsk_buff_free(xdp_buffs[i]);
+	}
+err_alloc:
+	/* If we managed to allocate at least some buffers,
+	 * release them to hardware
+	 */
+	if (i)
+		goto release_bufs;
+
+	return 0;
+}
+
+static int dpaa2_eth_seed_pool(struct dpaa2_eth_priv *priv,
+			       struct dpaa2_eth_channel *ch)
+{
+	int i;
+	int new_count;
+
+	for (i = 0; i < DPAA2_ETH_NUM_BUFS; i += DPAA2_ETH_BUFS_PER_CMD) {
+		new_count = dpaa2_eth_add_bufs(priv, ch);
+		ch->buf_count += new_count;
+
+		if (new_count < DPAA2_ETH_BUFS_PER_CMD)
+			return -ENOMEM;
+	}
+
+	return 0;
+}
+
+static void dpaa2_eth_seed_pools(struct dpaa2_eth_priv *priv)
+{
+	struct net_device *net_dev = priv->net_dev;
+	struct dpaa2_eth_channel *channel;
+	int i, err = 0;
+
+	for (i = 0; i < priv->num_channels; i++) {
+		channel = priv->channel[i];
+
+		err = dpaa2_eth_seed_pool(priv, channel);
+
+		/* Not much to do; the buffer pool, though not filled up,
+		 * may still contain some buffers which would enable us
+		 * to limp on.
+		 */
+		if (err)
+			netdev_err(net_dev, "Buffer seeding failed for DPBP %d (bpid=%d)\n",
+				   channel->bp->dev->obj_desc.id,
+				   channel->bp->bpid);
+	}
+}
+
+/*
+ * Drain the specified number of buffers from one of the DPNI's private buffer
+ * pools.
+ * @count must not exceeed DPAA2_ETH_BUFS_PER_CMD
+ */
+static void dpaa2_eth_drain_bufs(struct dpaa2_eth_priv *priv, int bpid,
+				 int count)
+{
+	u64 buf_array[DPAA2_ETH_BUFS_PER_CMD];
+	bool xsk_zc = false;
+	int retries = 0;
+	int i, ret;
+
+	for (i = 0; i < priv->num_channels; i++)
+		if (priv->channel[i]->bp->bpid == bpid)
+			xsk_zc = priv->channel[i]->xsk_zc;
+
+	do {
+		ret = dpaa2_io_service_acquire(NULL, bpid, buf_array, count);
+		if (ret < 0) {
+			if (ret == -EBUSY &&
+			    retries++ < DPAA2_ETH_SWP_BUSY_RETRIES)
+				continue;
+			netdev_err(priv->net_dev, "dpaa2_io_service_acquire() failed\n");
+			return;
+		}
+		dpaa2_eth_free_bufs(priv, buf_array, ret, xsk_zc);
+		retries = 0;
+	} while (ret);
+}
+
+static void dpaa2_eth_drain_pool(struct dpaa2_eth_priv *priv, int bpid)
+{
+	int i;
+
+	/* Drain the buffer pool */
+	dpaa2_eth_drain_bufs(priv, bpid, DPAA2_ETH_BUFS_PER_CMD);
+	dpaa2_eth_drain_bufs(priv, bpid, 1);
+
+	/* Setup to zero the buffer count of all channels which were
+	 * using this buffer pool.
+	 */
+	for (i = 0; i < priv->num_channels; i++)
+		if (priv->channel[i]->bp->bpid == bpid)
+			priv->channel[i]->buf_count = 0;
+}
+
+static void dpaa2_eth_drain_pools(struct dpaa2_eth_priv *priv)
+{
+	int i;
+
+	for (i = 0; i < priv->num_bps; i++)
+		dpaa2_eth_drain_pool(priv, priv->bp[i]->bpid);
+}
+
+/* Function is called from softirq context only, so we don't need to guard
+ * the access to percpu count
+ */
+static int dpaa2_eth_refill_pool(struct dpaa2_eth_priv *priv,
+				 struct dpaa2_eth_channel *ch)
+{
+	int new_count;
+
+	if (likely(ch->buf_count >= DPAA2_ETH_REFILL_THRESH))
+		return 0;
+
+	do {
+		new_count = dpaa2_eth_add_bufs(priv, ch);
+		if (unlikely(!new_count)) {
+			/* Out of memory; abort for now, we'll try later on */
+			break;
+		}
+		ch->buf_count += new_count;
+	} while (ch->buf_count < DPAA2_ETH_NUM_BUFS);
+
+	if (unlikely(ch->buf_count < DPAA2_ETH_NUM_BUFS))
+		return -ENOMEM;
+
+	return 0;
+}
+
+static void dpaa2_eth_sgt_cache_drain(struct dpaa2_eth_priv *priv)
+{
+	struct dpaa2_eth_sgt_cache *sgt_cache;
+	u16 count;
+	int k, i;
+
+	for_each_possible_cpu(k) {
+		sgt_cache = per_cpu_ptr(priv->sgt_cache, k);
+		count = sgt_cache->count;
+
+		for (i = 0; i < count; i++)
+			skb_free_frag(sgt_cache->buf[i]);
+		sgt_cache->count = 0;
+	}
+}
+
+static int dpaa2_eth_pull_channel(struct dpaa2_eth_channel *ch)
+{
+	int err;
+	int dequeues = -1;
+
+	/* Retry while portal is busy */
+	do {
+		err = dpaa2_io_service_pull_channel(ch->dpio, ch->ch_id,
+						    ch->store);
+		dequeues++;
+		cpu_relax();
+	} while (err == -EBUSY && dequeues < DPAA2_ETH_SWP_BUSY_RETRIES);
+
+	ch->stats.dequeue_portal_busy += dequeues;
+	if (unlikely(err))
+		ch->stats.pull_err++;
+
+	return err;
+}
+
+/* NAPI poll routine
+ *
+ * Frames are dequeued from the QMan channel associated with this NAPI context.
+ * Rx, Tx confirmation and (if configured) Rx error frames all count
+ * towards the NAPI budget.
+ */
+static int dpaa2_eth_poll(struct napi_struct *napi, int budget)
+{
+	struct dpaa2_eth_channel *ch;
+	struct dpaa2_eth_priv *priv;
+	int rx_cleaned = 0, txconf_cleaned = 0;
+	struct dpaa2_eth_fq *fq, *txc_fq = NULL;
+	struct netdev_queue *nq;
+	int store_cleaned, work_done;
+	bool work_done_zc = false;
+	struct list_head rx_list;
+	int retries = 0;
+	u16 flowid;
+	int err;
+
+	ch = container_of(napi, struct dpaa2_eth_channel, napi);
+	ch->xdp.res = 0;
+	priv = ch->priv;
+
+	INIT_LIST_HEAD(&rx_list);
+	ch->rx_list = &rx_list;
+
+	if (ch->xsk_zc) {
+		work_done_zc = dpaa2_xsk_tx(priv, ch);
+		/* If we reached the XSK Tx per NAPI threshold, we're done */
+		if (work_done_zc) {
+			work_done = budget;
+			goto out;
+		}
+	}
+
+	do {
+		err = dpaa2_eth_pull_channel(ch);
+		if (unlikely(err))
+			break;
+
+		/* Refill pool if appropriate */
+		dpaa2_eth_refill_pool(priv, ch);
+
+		store_cleaned = dpaa2_eth_consume_frames(ch, &fq);
+		if (store_cleaned <= 0)
+			break;
+		if (fq->type == DPAA2_RX_FQ) {
+			rx_cleaned += store_cleaned;
+			flowid = fq->flowid;
+		} else {
+			txconf_cleaned += store_cleaned;
+			/* We have a single Tx conf FQ on this channel */
+			txc_fq = fq;
+		}
+
+		/* If we either consumed the whole NAPI budget with Rx frames
+		 * or we reached the Tx confirmations threshold, we're done.
+		 */
+		if (rx_cleaned >= budget ||
+		    txconf_cleaned >= DPAA2_ETH_TXCONF_PER_NAPI) {
+			work_done = budget;
+			if (ch->xdp.res & XDP_REDIRECT)
+				xdp_do_flush();
+			goto out;
+		}
+	} while (store_cleaned);
+
+	if (ch->xdp.res & XDP_REDIRECT)
+		xdp_do_flush();
+
+	/* Update NET DIM with the values for this CDAN */
+	dpaa2_io_update_net_dim(ch->dpio, ch->stats.frames_per_cdan,
+				ch->stats.bytes_per_cdan);
+	ch->stats.frames_per_cdan = 0;
+	ch->stats.bytes_per_cdan = 0;
+
+	/* We didn't consume the entire budget, so finish napi and
+	 * re-enable data availability notifications
+	 */
+	napi_complete_done(napi, rx_cleaned);
+	do {
+		err = dpaa2_io_service_rearm(ch->dpio, &ch->nctx);
+		cpu_relax();
+	} while (err == -EBUSY && retries++ < DPAA2_ETH_SWP_BUSY_RETRIES);
+	WARN_ONCE(err, "CDAN notifications rearm failed on core %d",
+		  ch->nctx.desired_cpu);
+
+	work_done = max(rx_cleaned, 1);
+
+out:
+	netif_receive_skb_list(ch->rx_list);
+
+	if (ch->xsk_tx_pkts_sent) {
+		xsk_tx_completed(ch->xsk_pool, ch->xsk_tx_pkts_sent);
+		ch->xsk_tx_pkts_sent = 0;
+	}
+
+	if (txc_fq && txc_fq->dq_frames) {
+		nq = netdev_get_tx_queue(priv->net_dev, txc_fq->flowid);
+		netdev_tx_completed_queue(nq, txc_fq->dq_frames,
+					  txc_fq->dq_bytes);
+		txc_fq->dq_frames = 0;
+		txc_fq->dq_bytes = 0;
+	}
+
+	if (rx_cleaned && ch->xdp.res & XDP_TX)
+		dpaa2_eth_xdp_tx_flush(priv, ch, &priv->fq[flowid]);
+
+	return work_done;
+}
+
+static void dpaa2_eth_enable_ch_napi(struct dpaa2_eth_priv *priv)
+{
+	struct dpaa2_eth_channel *ch;
+	int i;
+
+	for (i = 0; i < priv->num_channels; i++) {
+		ch = priv->channel[i];
+		napi_enable(&ch->napi);
+	}
+}
+
+static void dpaa2_eth_disable_ch_napi(struct dpaa2_eth_priv *priv)
+{
+	struct dpaa2_eth_channel *ch;
+	int i;
+
+	for (i = 0; i < priv->num_channels; i++) {
+		ch = priv->channel[i];
+		napi_disable(&ch->napi);
+	}
+}
+
+void dpaa2_eth_set_rx_taildrop(struct dpaa2_eth_priv *priv,
+			       bool tx_pause, bool pfc)
+{
+	struct dpni_taildrop td = {0};
+	struct dpaa2_eth_fq *fq;
+	int i, err;
+
+	/* FQ taildrop: threshold is in bytes, per frame queue. Enabled if
+	 * flow control is disabled (as it might interfere with either the
+	 * buffer pool depletion trigger for pause frames or with the group
+	 * congestion trigger for PFC frames)
+	 */
+	td.enable = !tx_pause;
+	if (priv->rx_fqtd_enabled == td.enable)
+		goto set_cgtd;
+
+	td.threshold = DPAA2_ETH_FQ_TAILDROP_THRESH;
+	td.units = DPNI_CONGESTION_UNIT_BYTES;
+
+	for (i = 0; i < priv->num_fqs; i++) {
+		fq = &priv->fq[i];
+		if (fq->type != DPAA2_RX_FQ)
+			continue;
+		err = dpni_set_taildrop(priv->mc_io, 0, priv->mc_token,
+					DPNI_CP_QUEUE, DPNI_QUEUE_RX,
+					fq->tc, fq->flowid, &td);
+		if (err) {
+			netdev_err(priv->net_dev,
+				   "dpni_set_taildrop(FQ) failed\n");
+			return;
+		}
+	}
+
+	priv->rx_fqtd_enabled = td.enable;
+
+set_cgtd:
+	/* Congestion group taildrop: threshold is in frames, per group
+	 * of FQs belonging to the same traffic class
+	 * Enabled if general Tx pause disabled or if PFCs are enabled
+	 * (congestion group threhsold for PFC generation is lower than the
+	 * CG taildrop threshold, so it won't interfere with it; we also
+	 * want frames in non-PFC enabled traffic classes to be kept in check)
+	 */
+	td.enable = !tx_pause || pfc;
+	if (priv->rx_cgtd_enabled == td.enable)
+		return;
+
+	td.threshold = DPAA2_ETH_CG_TAILDROP_THRESH(priv);
+	td.units = DPNI_CONGESTION_UNIT_FRAMES;
+	for (i = 0; i < dpaa2_eth_tc_count(priv); i++) {
+		err = dpni_set_taildrop(priv->mc_io, 0, priv->mc_token,
+					DPNI_CP_GROUP, DPNI_QUEUE_RX,
+					i, 0, &td);
+		if (err) {
+			netdev_err(priv->net_dev,
+				   "dpni_set_taildrop(CG) failed\n");
+			return;
+		}
+	}
+
+	priv->rx_cgtd_enabled = td.enable;
+}
+
+static int dpaa2_eth_link_state_update(struct dpaa2_eth_priv *priv)
+{
+	struct dpni_link_state state = {0};
+	bool tx_pause;
+	int err;
+
+	err = dpni_get_link_state(priv->mc_io, 0, priv->mc_token, &state);
+	if (unlikely(err)) {
+		netdev_err(priv->net_dev,
+			   "dpni_get_link_state() failed\n");
+		return err;
+	}
+
+	/* If Tx pause frame settings have changed, we need to update
+	 * Rx FQ taildrop configuration as well. We configure taildrop
+	 * only when pause frame generation is disabled.
+	 */
+	tx_pause = dpaa2_eth_tx_pause_enabled(state.options);
+	dpaa2_eth_set_rx_taildrop(priv, tx_pause, priv->pfc_enabled);
+
+	/* When we manage the MAC/PHY using phylink there is no need
+	 * to manually update the netif_carrier.
+	 * We can avoid locking because we are called from the "link changed"
+	 * IRQ handler, which is the same as the "endpoint changed" IRQ handler
+	 * (the writer to priv->mac), so we cannot race with it.
+	 */
+	if (dpaa2_mac_is_type_phy(priv->mac))
+		goto out;
+
+	/* Chech link state; speed / duplex changes are not treated yet */
+	if (priv->link_state.up == state.up)
+		goto out;
+
+	if (state.up) {
+		netif_carrier_on(priv->net_dev);
+		netif_tx_start_all_queues(priv->net_dev);
+	} else {
+		netif_tx_stop_all_queues(priv->net_dev);
+		netif_carrier_off(priv->net_dev);
+	}
+
+	netdev_info(priv->net_dev, "Link Event: state %s\n",
+		    state.up ? "up" : "down");
+
+out:
+	priv->link_state = state;
+
+	return 0;
+}
+
+static int dpaa2_eth_open(struct net_device *net_dev)
+{
+	struct dpaa2_eth_priv *priv = netdev_priv(net_dev);
+	int err;
+
+	dpaa2_eth_seed_pools(priv);
+
+	mutex_lock(&priv->mac_lock);
+
+	if (!dpaa2_eth_is_type_phy(priv)) {
+		/* We'll only start the txqs when the link is actually ready;
+		 * make sure we don't race against the link up notification,
+		 * which may come immediately after dpni_enable();
+		 */
+		netif_tx_stop_all_queues(net_dev);
+
+		/* Also, explicitly set carrier off, otherwise
+		 * netif_carrier_ok() will return true and cause 'ip link show'
+		 * to report the LOWER_UP flag, even though the link
+		 * notification wasn't even received.
+		 */
+		netif_carrier_off(net_dev);
+	}
+	dpaa2_eth_enable_ch_napi(priv);
+
+	err = dpni_enable(priv->mc_io, 0, priv->mc_token);
+	if (err < 0) {
+		mutex_unlock(&priv->mac_lock);
+		netdev_err(net_dev, "dpni_enable() failed\n");
+		goto enable_err;
+	}
+
+	if (dpaa2_eth_is_type_phy(priv))
+		dpaa2_mac_start(priv->mac);
+
+	mutex_unlock(&priv->mac_lock);
+
+	return 0;
+
+enable_err:
+	dpaa2_eth_disable_ch_napi(priv);
+	dpaa2_eth_drain_pools(priv);
+	return err;
+}
+
+/* Total number of in-flight frames on ingress queues */
+static u32 dpaa2_eth_ingress_fq_count(struct dpaa2_eth_priv *priv)
+{
+	struct dpaa2_eth_fq *fq;
+	u32 fcnt = 0, bcnt = 0, total = 0;
+	int i, err;
+
+	for (i = 0; i < priv->num_fqs; i++) {
+		fq = &priv->fq[i];
+		err = dpaa2_io_query_fq_count(NULL, fq->fqid, &fcnt, &bcnt);
+		if (err) {
+			netdev_warn(priv->net_dev, "query_fq_count failed");
+			break;
+		}
+		total += fcnt;
+	}
+
+	return total;
+}
+
+static void dpaa2_eth_wait_for_ingress_fq_empty(struct dpaa2_eth_priv *priv)
+{
+	int retries = 10;
+	u32 pending;
+
+	do {
+		pending = dpaa2_eth_ingress_fq_count(priv);
+		if (pending)
+			msleep(100);
+	} while (pending && --retries);
+}
+
+#define DPNI_TX_PENDING_VER_MAJOR	7
+#define DPNI_TX_PENDING_VER_MINOR	13
+static void dpaa2_eth_wait_for_egress_fq_empty(struct dpaa2_eth_priv *priv)
+{
+	union dpni_statistics stats;
+	int retries = 10;
+	int err;
+
+	if (dpaa2_eth_cmp_dpni_ver(priv, DPNI_TX_PENDING_VER_MAJOR,
+				   DPNI_TX_PENDING_VER_MINOR) < 0)
+		goto out;
+
+	do {
+		err = dpni_get_statistics(priv->mc_io, 0, priv->mc_token, 6,
+					  &stats);
+		if (err)
+			goto out;
+		if (stats.page_6.tx_pending_frames == 0)
+			return;
+	} while (--retries);
+
+out:
+	msleep(500);
+}
+
+static int dpaa2_eth_stop(struct net_device *net_dev)
+{
+	struct dpaa2_eth_priv *priv = netdev_priv(net_dev);
+	int dpni_enabled = 0;
+	int retries = 10;
+
+	mutex_lock(&priv->mac_lock);
+
+	if (dpaa2_eth_is_type_phy(priv)) {
+		dpaa2_mac_stop(priv->mac);
+	} else {
+		netif_tx_stop_all_queues(net_dev);
+		netif_carrier_off(net_dev);
+	}
+
+	mutex_unlock(&priv->mac_lock);
+
+	/* On dpni_disable(), the MC firmware will:
+	 * - stop MAC Rx and wait for all Rx frames to be enqueued to software
+	 * - cut off WRIOP dequeues from egress FQs and wait until transmission
+	 * of all in flight Tx frames is finished (and corresponding Tx conf
+	 * frames are enqueued back to software)
+	 *
+	 * Before calling dpni_disable(), we wait for all Tx frames to arrive
+	 * on WRIOP. After it finishes, wait until all remaining frames on Rx
+	 * and Tx conf queues are consumed on NAPI poll.
+	 */
+	dpaa2_eth_wait_for_egress_fq_empty(priv);
+
+	do {
+		dpni_disable(priv->mc_io, 0, priv->mc_token);
+		dpni_is_enabled(priv->mc_io, 0, priv->mc_token, &dpni_enabled);
+		if (dpni_enabled)
+			/* Allow the hardware some slack */
+			msleep(100);
+	} while (dpni_enabled && --retries);
+	if (!retries) {
+		netdev_warn(net_dev, "Retry count exceeded disabling DPNI\n");
+		/* Must go on and disable NAPI nonetheless, so we don't crash at
+		 * the next "ifconfig up"
+		 */
+	}
+
+	dpaa2_eth_wait_for_ingress_fq_empty(priv);
+	dpaa2_eth_disable_ch_napi(priv);
+
+	/* Empty the buffer pool */
+	dpaa2_eth_drain_pools(priv);
+
+	/* Empty the Scatter-Gather Buffer cache */
+	dpaa2_eth_sgt_cache_drain(priv);
+
+	return 0;
+}
+
+static int dpaa2_eth_set_addr(struct net_device *net_dev, void *addr)
+{
+	struct dpaa2_eth_priv *priv = netdev_priv(net_dev);
+	struct device *dev = net_dev->dev.parent;
+	int err;
+
+	err = eth_mac_addr(net_dev, addr);
+	if (err < 0) {
+		dev_err(dev, "eth_mac_addr() failed (%d)\n", err);
+		return err;
+	}
+
+	err = dpni_set_primary_mac_addr(priv->mc_io, 0, priv->mc_token,
+					net_dev->dev_addr);
+	if (err) {
+		dev_err(dev, "dpni_set_primary_mac_addr() failed (%d)\n", err);
+		return err;
+	}
+
+	return 0;
+}
+
+/** Fill in counters maintained by the GPP driver. These may be different from
+ * the hardware counters obtained by ethtool.
+ */
+static void dpaa2_eth_get_stats(struct net_device *net_dev,
+				struct rtnl_link_stats64 *stats)
+{
+	struct dpaa2_eth_priv *priv = netdev_priv(net_dev);
+	struct rtnl_link_stats64 *percpu_stats;
+	u64 *cpustats;
+	u64 *netstats = (u64 *)stats;
+	int i, j;
+	int num = sizeof(struct rtnl_link_stats64) / sizeof(u64);
+
+	for_each_possible_cpu(i) {
+		percpu_stats = per_cpu_ptr(priv->percpu_stats, i);
+		cpustats = (u64 *)percpu_stats;
+		for (j = 0; j < num; j++)
+			netstats[j] += cpustats[j];
+	}
+}
+
+/* Copy mac unicast addresses from @net_dev to @priv.
+ * Its sole purpose is to make dpaa2_eth_set_rx_mode() more readable.
+ */
+static void dpaa2_eth_add_uc_hw_addr(const struct net_device *net_dev,
+				     struct dpaa2_eth_priv *priv)
+{
+	struct netdev_hw_addr *ha;
+	int err;
+
+	netdev_for_each_uc_addr(ha, net_dev) {
+		err = dpni_add_mac_addr(priv->mc_io, 0, priv->mc_token,
+					ha->addr);
+		if (err)
+			netdev_warn(priv->net_dev,
+				    "Could not add ucast MAC %pM to the filtering table (err %d)\n",
+				    ha->addr, err);
+	}
+}
+
+/* Copy mac multicast addresses from @net_dev to @priv
+ * Its sole purpose is to make dpaa2_eth_set_rx_mode() more readable.
+ */
+static void dpaa2_eth_add_mc_hw_addr(const struct net_device *net_dev,
+				     struct dpaa2_eth_priv *priv)
+{
+	struct netdev_hw_addr *ha;
+	int err;
+
+	netdev_for_each_mc_addr(ha, net_dev) {
+		err = dpni_add_mac_addr(priv->mc_io, 0, priv->mc_token,
+					ha->addr);
+		if (err)
+			netdev_warn(priv->net_dev,
+				    "Could not add mcast MAC %pM to the filtering table (err %d)\n",
+				    ha->addr, err);
+	}
+}
+
+static int dpaa2_eth_rx_add_vid(struct net_device *net_dev,
+				__be16 vlan_proto, u16 vid)
+{
+	struct dpaa2_eth_priv *priv = netdev_priv(net_dev);
+	int err;
+
+	err = dpni_add_vlan_id(priv->mc_io, 0, priv->mc_token,
+			       vid, 0, 0, 0);
+
+	if (err) {
+		netdev_warn(priv->net_dev,
+			    "Could not add the vlan id %u\n",
+			    vid);
+		return err;
+	}
+
+	return 0;
+}
+
+static int dpaa2_eth_rx_kill_vid(struct net_device *net_dev,
+				 __be16 vlan_proto, u16 vid)
+{
+	struct dpaa2_eth_priv *priv = netdev_priv(net_dev);
+	int err;
+
+	err = dpni_remove_vlan_id(priv->mc_io, 0, priv->mc_token, vid);
+
+	if (err) {
+		netdev_warn(priv->net_dev,
+			    "Could not remove the vlan id %u\n",
+			    vid);
+		return err;
+	}
+
+	return 0;
+}
+
+static void dpaa2_eth_set_rx_mode(struct net_device *net_dev)
+{
+	struct dpaa2_eth_priv *priv = netdev_priv(net_dev);
+	int uc_count = netdev_uc_count(net_dev);
+	int mc_count = netdev_mc_count(net_dev);
+	u8 max_mac = priv->dpni_attrs.mac_filter_entries;
+	u32 options = priv->dpni_attrs.options;
+	u16 mc_token = priv->mc_token;
+	struct fsl_mc_io *mc_io = priv->mc_io;
+	int err;
+
+	/* Basic sanity checks; these probably indicate a misconfiguration */
+	if (options & DPNI_OPT_NO_MAC_FILTER && max_mac != 0)
+		netdev_info(net_dev,
+			    "mac_filter_entries=%d, DPNI_OPT_NO_MAC_FILTER option must be disabled\n",
+			    max_mac);
+
+	/* Force promiscuous if the uc or mc counts exceed our capabilities. */
+	if (uc_count > max_mac) {
+		netdev_info(net_dev,
+			    "Unicast addr count reached %d, max allowed is %d; forcing promisc\n",
+			    uc_count, max_mac);
+		goto force_promisc;
+	}
+	if (mc_count + uc_count > max_mac) {
+		netdev_info(net_dev,
+			    "Unicast + multicast addr count reached %d, max allowed is %d; forcing promisc\n",
+			    uc_count + mc_count, max_mac);
+		goto force_mc_promisc;
+	}
+
+	/* Adjust promisc settings due to flag combinations */
+	if (net_dev->flags & IFF_PROMISC)
+		goto force_promisc;
+	if (net_dev->flags & IFF_ALLMULTI) {
+		/* First, rebuild unicast filtering table. This should be done
+		 * in promisc mode, in order to avoid frame loss while we
+		 * progressively add entries to the table.
+		 * We don't know whether we had been in promisc already, and
+		 * making an MC call to find out is expensive; so set uc promisc
+		 * nonetheless.
+		 */
+		err = dpni_set_unicast_promisc(mc_io, 0, mc_token, 1);
+		if (err)
+			netdev_warn(net_dev, "Can't set uc promisc\n");
+
+		/* Actual uc table reconstruction. */
+		err = dpni_clear_mac_filters(mc_io, 0, mc_token, 1, 0);
+		if (err)
+			netdev_warn(net_dev, "Can't clear uc filters\n");
+		dpaa2_eth_add_uc_hw_addr(net_dev, priv);
+
+		/* Finally, clear uc promisc and set mc promisc as requested. */
+		err = dpni_set_unicast_promisc(mc_io, 0, mc_token, 0);
+		if (err)
+			netdev_warn(net_dev, "Can't clear uc promisc\n");
+		goto force_mc_promisc;
+	}
+
+	/* Neither unicast, nor multicast promisc will be on... eventually.
+	 * For now, rebuild mac filtering tables while forcing both of them on.
+	 */
+	err = dpni_set_unicast_promisc(mc_io, 0, mc_token, 1);
+	if (err)
+		netdev_warn(net_dev, "Can't set uc promisc (%d)\n", err);
+	err = dpni_set_multicast_promisc(mc_io, 0, mc_token, 1);
+	if (err)
+		netdev_warn(net_dev, "Can't set mc promisc (%d)\n", err);
+
+	/* Actual mac filtering tables reconstruction */
+	err = dpni_clear_mac_filters(mc_io, 0, mc_token, 1, 1);
+	if (err)
+		netdev_warn(net_dev, "Can't clear mac filters\n");
+	dpaa2_eth_add_mc_hw_addr(net_dev, priv);
+	dpaa2_eth_add_uc_hw_addr(net_dev, priv);
+
+	/* Now we can clear both ucast and mcast promisc, without risking
+	 * to drop legitimate frames anymore.
+	 */
+	err = dpni_set_unicast_promisc(mc_io, 0, mc_token, 0);
+	if (err)
+		netdev_warn(net_dev, "Can't clear ucast promisc\n");
+	err = dpni_set_multicast_promisc(mc_io, 0, mc_token, 0);
+	if (err)
+		netdev_warn(net_dev, "Can't clear mcast promisc\n");
+
+	return;
+
+force_promisc:
+	err = dpni_set_unicast_promisc(mc_io, 0, mc_token, 1);
+	if (err)
+		netdev_warn(net_dev, "Can't set ucast promisc\n");
+force_mc_promisc:
+	err = dpni_set_multicast_promisc(mc_io, 0, mc_token, 1);
+	if (err)
+		netdev_warn(net_dev, "Can't set mcast promisc\n");
+}
+
+static int dpaa2_eth_set_features(struct net_device *net_dev,
+				  netdev_features_t features)
+{
+	struct dpaa2_eth_priv *priv = netdev_priv(net_dev);
+	netdev_features_t changed = features ^ net_dev->features;
+	bool enable;
+	int err;
+
+	if (changed & NETIF_F_HW_VLAN_CTAG_FILTER) {
+		enable = !!(features & NETIF_F_HW_VLAN_CTAG_FILTER);
+		err = dpaa2_eth_set_rx_vlan_filtering(priv, enable);
+		if (err)
+			return err;
+	}
+
+	if (changed & NETIF_F_RXCSUM) {
+		enable = !!(features & NETIF_F_RXCSUM);
+		err = dpaa2_eth_set_rx_csum(priv, enable);
+		if (err)
+			return err;
+	}
+
+	if (changed & (NETIF_F_IP_CSUM | NETIF_F_IPV6_CSUM)) {
+		enable = !!(features & (NETIF_F_IP_CSUM | NETIF_F_IPV6_CSUM));
+		err = dpaa2_eth_set_tx_csum(priv, enable);
+		if (err)
+			return err;
+	}
+
+	return 0;
+}
+
+static int dpaa2_eth_hwtstamp_set(struct net_device *dev,
+				  struct kernel_hwtstamp_config *config,
+				  struct netlink_ext_ack *extack)
+{
+	struct dpaa2_eth_priv *priv = netdev_priv(dev);
+
+	if (!dpaa2_ptp)
+		return -EINVAL;
+
+	switch (config->tx_type) {
+	case HWTSTAMP_TX_OFF:
+	case HWTSTAMP_TX_ON:
+	case HWTSTAMP_TX_ONESTEP_SYNC:
+		priv->tx_tstamp_type = config->tx_type;
+		break;
+	default:
+		return -ERANGE;
+	}
+
+	if (config->rx_filter == HWTSTAMP_FILTER_NONE) {
+		priv->rx_tstamp = false;
+	} else {
+		priv->rx_tstamp = true;
+		/* TS is set for all frame types, not only those requested */
+		config->rx_filter = HWTSTAMP_FILTER_ALL;
+	}
+
+	if (priv->tx_tstamp_type == HWTSTAMP_TX_ONESTEP_SYNC)
+		dpaa2_ptp_onestep_reg_update_method(priv);
+
+	return 0;
+}
+
+static int dpaa2_eth_hwtstamp_get(struct net_device *dev,
+				  struct kernel_hwtstamp_config *config)
+{
+	struct dpaa2_eth_priv *priv = netdev_priv(dev);
+
+	if (!dpaa2_ptp)
+		return -EINVAL;
+
+	config->tx_type = priv->tx_tstamp_type;
+	config->rx_filter = priv->rx_tstamp ? HWTSTAMP_FILTER_ALL :
+			    HWTSTAMP_FILTER_NONE;
+
+	return 0;
+}
+
+static int dpaa2_eth_ioctl(struct net_device *dev, struct ifreq *rq, int cmd)
+{
+	struct dpaa2_eth_priv *priv = netdev_priv(dev);
+	int err;
+
+	mutex_lock(&priv->mac_lock);
+
+	if (dpaa2_eth_is_type_phy(priv)) {
+		err = phylink_mii_ioctl(priv->mac->phylink, rq, cmd);
+		mutex_unlock(&priv->mac_lock);
+		return err;
+	}
+
+	mutex_unlock(&priv->mac_lock);
+
+	return -EOPNOTSUPP;
+}
+
+static bool xdp_mtu_valid(struct dpaa2_eth_priv *priv, int mtu)
+{
+	int mfl, linear_mfl;
+
+	mfl = DPAA2_ETH_L2_MAX_FRM(mtu);
+	linear_mfl = priv->rx_buf_size - DPAA2_ETH_RX_HWA_SIZE -
+		     dpaa2_eth_rx_head_room(priv) - XDP_PACKET_HEADROOM;
+
+	if (mfl > linear_mfl) {
+		netdev_warn(priv->net_dev, "Maximum MTU for XDP is %d\n",
+			    linear_mfl - VLAN_ETH_HLEN);
+		return false;
+	}
+
+	return true;
+}
+
+static int dpaa2_eth_set_rx_mfl(struct dpaa2_eth_priv *priv, int mtu, bool has_xdp)
+{
+	int mfl, err;
+
+	/* We enforce a maximum Rx frame length based on MTU only if we have
+	 * an XDP program attached (in order to avoid Rx S/G frames).
+	 * Otherwise, we accept all incoming frames as long as they are not
+	 * larger than maximum size supported in hardware
+	 */
+	if (has_xdp)
+		mfl = DPAA2_ETH_L2_MAX_FRM(mtu);
+	else
+		mfl = DPAA2_ETH_MFL;
+
+	err = dpni_set_max_frame_length(priv->mc_io, 0, priv->mc_token, mfl);
+	if (err) {
+		netdev_err(priv->net_dev, "dpni_set_max_frame_length failed\n");
+		return err;
+	}
+
+	return 0;
+}
+
+static int dpaa2_eth_change_mtu(struct net_device *dev, int new_mtu)
+{
+	struct dpaa2_eth_priv *priv = netdev_priv(dev);
+	int err;
+
+	if (!priv->xdp_prog)
+		goto out;
+
+	if (!xdp_mtu_valid(priv, new_mtu))
+		return -EINVAL;
+
+	err = dpaa2_eth_set_rx_mfl(priv, new_mtu, true);
+	if (err)
+		return err;
+
+out:
+	WRITE_ONCE(dev->mtu, new_mtu);
+	return 0;
+}
+
+static int dpaa2_eth_update_rx_buffer_headroom(struct dpaa2_eth_priv *priv, bool has_xdp)
+{
+	struct dpni_buffer_layout buf_layout = {0};
+	int err;
+
+	err = dpni_get_buffer_layout(priv->mc_io, 0, priv->mc_token,
+				     DPNI_QUEUE_RX, &buf_layout);
+	if (err) {
+		netdev_err(priv->net_dev, "dpni_get_buffer_layout failed\n");
+		return err;
+	}
+
+	/* Reserve extra headroom for XDP header size changes */
+	buf_layout.data_head_room = dpaa2_eth_rx_head_room(priv) +
+				    (has_xdp ? XDP_PACKET_HEADROOM : 0);
+	buf_layout.options = DPNI_BUF_LAYOUT_OPT_DATA_HEAD_ROOM;
+	err = dpni_set_buffer_layout(priv->mc_io, 0, priv->mc_token,
+				     DPNI_QUEUE_RX, &buf_layout);
+	if (err) {
+		netdev_err(priv->net_dev, "dpni_set_buffer_layout failed\n");
+		return err;
+	}
+
+	return 0;
+}
+
+static int dpaa2_eth_setup_xdp(struct net_device *dev, struct bpf_prog *prog)
+{
+	struct dpaa2_eth_priv *priv = netdev_priv(dev);
+	struct dpaa2_eth_channel *ch;
+	struct bpf_prog *old;
+	bool up, need_update;
+	int i, err;
+
+	if (prog && !xdp_mtu_valid(priv, dev->mtu))
+		return -EINVAL;
+
+	if (prog)
+		bpf_prog_add(prog, priv->num_channels);
+
+	up = netif_running(dev);
+	need_update = (!!priv->xdp_prog != !!prog);
+
+	if (up)
+		dev_close(dev);
+
+	/* While in xdp mode, enforce a maximum Rx frame size based on MTU.
+	 * Also, when switching between xdp/non-xdp modes we need to reconfigure
+	 * our Rx buffer layout. Buffer pool was drained on dpaa2_eth_stop,
+	 * so we are sure no old format buffers will be used from now on.
+	 */
+	if (need_update) {
+		err = dpaa2_eth_set_rx_mfl(priv, dev->mtu, !!prog);
+		if (err)
+			goto out_err;
+		err = dpaa2_eth_update_rx_buffer_headroom(priv, !!prog);
+		if (err)
+			goto out_err;
+	}
+
+	old = xchg(&priv->xdp_prog, prog);
+	if (old)
+		bpf_prog_put(old);
+
+	for (i = 0; i < priv->num_channels; i++) {
+		ch = priv->channel[i];
+		old = xchg(&ch->xdp.prog, prog);
+		if (old)
+			bpf_prog_put(old);
+	}
+
+	if (up) {
+		err = dev_open(dev, NULL);
+		if (err)
+			return err;
+	}
+
+	return 0;
+
+out_err:
+	if (prog)
+		bpf_prog_sub(prog, priv->num_channels);
+	if (up)
+		dev_open(dev, NULL);
+
+	return err;
+}
+
+static int dpaa2_eth_xdp(struct net_device *dev, struct netdev_bpf *xdp)
+{
+	switch (xdp->command) {
+	case XDP_SETUP_PROG:
+		return dpaa2_eth_setup_xdp(dev, xdp->prog);
+	case XDP_SETUP_XSK_POOL:
+		return dpaa2_xsk_setup_pool(dev, xdp->xsk.pool, xdp->xsk.queue_id);
+	default:
+		return -EINVAL;
+	}
+
+	return 0;
+}
+
+static int dpaa2_eth_xdp_create_fd(struct net_device *net_dev,
+				   struct xdp_frame *xdpf,
+				   struct dpaa2_fd *fd)
+{
+	struct device *dev = net_dev->dev.parent;
+	unsigned int needed_headroom;
+	struct dpaa2_eth_swa *swa;
+	void *buffer_start, *aligned_start;
+	dma_addr_t addr;
+
+	/* We require a minimum headroom to be able to transmit the frame.
+	 * Otherwise return an error and let the original net_device handle it
+	 */
+	needed_headroom = dpaa2_eth_needed_headroom(NULL);
+	if (xdpf->headroom < needed_headroom)
+		return -EINVAL;
+
+	/* Setup the FD fields */
+	memset(fd, 0, sizeof(*fd));
+
+	/* Align FD address, if possible */
+	buffer_start = xdpf->data - needed_headroom;
+	aligned_start = PTR_ALIGN(buffer_start - DPAA2_ETH_TX_BUF_ALIGN,
+				  DPAA2_ETH_TX_BUF_ALIGN);
+	if (aligned_start >= xdpf->data - xdpf->headroom)
+		buffer_start = aligned_start;
+
+	swa = (struct dpaa2_eth_swa *)buffer_start;
+	/* fill in necessary fields here */
+	swa->type = DPAA2_ETH_SWA_XDP;
+	swa->xdp.dma_size = xdpf->data + xdpf->len - buffer_start;
+	swa->xdp.xdpf = xdpf;
+
+	addr = dma_map_single(dev, buffer_start,
+			      swa->xdp.dma_size,
+			      DMA_BIDIRECTIONAL);
+	if (unlikely(dma_mapping_error(dev, addr)))
+		return -ENOMEM;
+
+	dpaa2_fd_set_addr(fd, addr);
+	dpaa2_fd_set_offset(fd, xdpf->data - buffer_start);
+	dpaa2_fd_set_len(fd, xdpf->len);
+	dpaa2_fd_set_format(fd, dpaa2_fd_single);
+	dpaa2_fd_set_ctrl(fd, FD_CTRL_PTA);
+
+	return 0;
+}
+
+static int dpaa2_eth_xdp_xmit(struct net_device *net_dev, int n,
+			      struct xdp_frame **frames, u32 flags)
+{
+	struct dpaa2_eth_priv *priv = netdev_priv(net_dev);
+	struct dpaa2_eth_xdp_fds *xdp_redirect_fds;
+	struct rtnl_link_stats64 *percpu_stats;
+	struct dpaa2_eth_fq *fq;
+	struct dpaa2_fd *fds;
+	int enqueued, i, err;
+
+	if (unlikely(flags & ~XDP_XMIT_FLAGS_MASK))
+		return -EINVAL;
+
+	if (!netif_running(net_dev))
+		return -ENETDOWN;
+
+	fq = &priv->fq[smp_processor_id()];
+	xdp_redirect_fds = &fq->xdp_redirect_fds;
+	fds = xdp_redirect_fds->fds;
+
+	percpu_stats = this_cpu_ptr(priv->percpu_stats);
+
+	/* create a FD for each xdp_frame in the list received */
+	for (i = 0; i < n; i++) {
+		err = dpaa2_eth_xdp_create_fd(net_dev, frames[i], &fds[i]);
+		if (err)
+			break;
+	}
+	xdp_redirect_fds->num = i;
+
+	/* enqueue all the frame descriptors */
+	enqueued = dpaa2_eth_xdp_flush(priv, fq, xdp_redirect_fds);
+
+	/* update statistics */
+	percpu_stats->tx_packets += enqueued;
+	for (i = 0; i < enqueued; i++)
+		percpu_stats->tx_bytes += dpaa2_fd_get_len(&fds[i]);
+
+	return enqueued;
+}
+
+static int update_xps(struct dpaa2_eth_priv *priv)
+{
+	struct net_device *net_dev = priv->net_dev;
+	int i, num_queues, netdev_queues;
+	struct dpaa2_eth_fq *fq;
+	cpumask_var_t xps_mask;
+	int err = 0;
+
+	if (!alloc_cpumask_var(&xps_mask, GFP_KERNEL))
+		return -ENOMEM;
+
+	num_queues = dpaa2_eth_queue_count(priv);
+	netdev_queues = (net_dev->num_tc ? : 1) * num_queues;
+
+	/* The first <num_queues> entries in priv->fq array are Tx/Tx conf
+	 * queues, so only process those
+	 */
+	for (i = 0; i < netdev_queues; i++) {
+		fq = &priv->fq[i % num_queues];
+
+		cpumask_clear(xps_mask);
+		cpumask_set_cpu(fq->target_cpu, xps_mask);
+
+		err = netif_set_xps_queue(net_dev, xps_mask, i);
+		if (err) {
+			netdev_warn_once(net_dev, "Error setting XPS queue\n");
+			break;
+		}
+	}
+
+	free_cpumask_var(xps_mask);
+	return err;
+}
+
+static int dpaa2_eth_setup_mqprio(struct net_device *net_dev,
+				  struct tc_mqprio_qopt *mqprio)
+{
+	struct dpaa2_eth_priv *priv = netdev_priv(net_dev);
+	u8 num_tc, num_queues;
+	int i;
+
+	mqprio->hw = TC_MQPRIO_HW_OFFLOAD_TCS;
+	num_queues = dpaa2_eth_queue_count(priv);
+	num_tc = mqprio->num_tc;
+
+	if (num_tc == net_dev->num_tc)
+		return 0;
+
+	if (num_tc  > dpaa2_eth_tc_count(priv)) {
+		netdev_err(net_dev, "Max %d traffic classes supported\n",
+			   dpaa2_eth_tc_count(priv));
+		return -EOPNOTSUPP;
+	}
+
+	if (!num_tc) {
+		netdev_reset_tc(net_dev);
+		netif_set_real_num_tx_queues(net_dev, num_queues);
+		goto out;
+	}
+
+	netdev_set_num_tc(net_dev, num_tc);
+	netif_set_real_num_tx_queues(net_dev, num_tc * num_queues);
+
+	for (i = 0; i < num_tc; i++)
+		netdev_set_tc_queue(net_dev, i, num_queues, i * num_queues);
+
+out:
+	update_xps(priv);
+
+	return 0;
+}
+
+#define bps_to_mbits(rate) (div_u64((rate), 1000000) * 8)
+
+static int dpaa2_eth_setup_tbf(struct net_device *net_dev, struct tc_tbf_qopt_offload *p)
+{
+	struct tc_tbf_qopt_offload_replace_params *cfg = &p->replace_params;
+	struct dpaa2_eth_priv *priv = netdev_priv(net_dev);
+	struct dpni_tx_shaping_cfg tx_cr_shaper = { 0 };
+	struct dpni_tx_shaping_cfg tx_er_shaper = { 0 };
+	int err;
+
+	if (p->command == TC_TBF_STATS)
+		return -EOPNOTSUPP;
+
+	/* Only per port Tx shaping */
+	if (p->parent != TC_H_ROOT)
+		return -EOPNOTSUPP;
+
+	if (p->command == TC_TBF_REPLACE) {
+		if (cfg->max_size > DPAA2_ETH_MAX_BURST_SIZE) {
+			netdev_err(net_dev, "burst size cannot be greater than %d\n",
+				   DPAA2_ETH_MAX_BURST_SIZE);
+			return -EINVAL;
+		}
+
+		tx_cr_shaper.max_burst_size = cfg->max_size;
+		/* The TBF interface is in bytes/s, whereas DPAA2 expects the
+		 * rate in Mbits/s
+		 */
+		tx_cr_shaper.rate_limit = bps_to_mbits(cfg->rate.rate_bytes_ps);
+	}
+
+	err = dpni_set_tx_shaping(priv->mc_io, 0, priv->mc_token, &tx_cr_shaper,
+				  &tx_er_shaper, 0);
+	if (err) {
+		netdev_err(net_dev, "dpni_set_tx_shaping() = %d\n", err);
+		return err;
+	}
+
+	return 0;
+}
+
+static int dpaa2_eth_setup_tc(struct net_device *net_dev,
+			      enum tc_setup_type type, void *type_data)
+{
+	switch (type) {
+	case TC_SETUP_QDISC_MQPRIO:
+		return dpaa2_eth_setup_mqprio(net_dev, type_data);
+	case TC_SETUP_QDISC_TBF:
+		return dpaa2_eth_setup_tbf(net_dev, type_data);
+	default:
+		return -EOPNOTSUPP;
+	}
+}
+
+static const struct net_device_ops dpaa2_eth_ops = {
+	.ndo_open = dpaa2_eth_open,
+	.ndo_start_xmit = dpaa2_eth_tx,
+	.ndo_stop = dpaa2_eth_stop,
+	.ndo_set_mac_address = dpaa2_eth_set_addr,
+	.ndo_get_stats64 = dpaa2_eth_get_stats,
+	.ndo_set_rx_mode = dpaa2_eth_set_rx_mode,
+	.ndo_set_features = dpaa2_eth_set_features,
+	.ndo_eth_ioctl = dpaa2_eth_ioctl,
+	.ndo_change_mtu = dpaa2_eth_change_mtu,
+	.ndo_bpf = dpaa2_eth_xdp,
+	.ndo_xdp_xmit = dpaa2_eth_xdp_xmit,
+	.ndo_xsk_wakeup = dpaa2_xsk_wakeup,
+	.ndo_setup_tc = dpaa2_eth_setup_tc,
+	.ndo_vlan_rx_add_vid = dpaa2_eth_rx_add_vid,
+	.ndo_vlan_rx_kill_vid = dpaa2_eth_rx_kill_vid,
+	.ndo_hwtstamp_get = dpaa2_eth_hwtstamp_get,
+	.ndo_hwtstamp_set = dpaa2_eth_hwtstamp_set,
+};
+
+static void dpaa2_eth_cdan_cb(struct dpaa2_io_notification_ctx *ctx)
+{
+	struct dpaa2_eth_channel *ch;
+
+	ch = container_of(ctx, struct dpaa2_eth_channel, nctx);
+
+	/* Update NAPI statistics */
+	ch->stats.cdan++;
+
+	/* NAPI can also be scheduled from the AF_XDP Tx path. Mark a missed
+	 * so that it can be rescheduled again.
+	 */
+	if (!napi_if_scheduled_mark_missed(&ch->napi))
+		napi_schedule(&ch->napi);
+}
+
+/* Allocate and configure a DPCON object */
+static struct fsl_mc_device *dpaa2_eth_setup_dpcon(struct dpaa2_eth_priv *priv)
+{
+	struct fsl_mc_device *dpcon;
+	struct device *dev = priv->net_dev->dev.parent;
+	int err;
+
+	err = fsl_mc_object_allocate(to_fsl_mc_device(dev),
+				     FSL_MC_POOL_DPCON, &dpcon);
+	if (err) {
+		if (err == -ENXIO) {
+			dev_dbg(dev, "Waiting for DPCON\n");
+			err = -EPROBE_DEFER;
+		} else {
+			dev_info(dev, "Not enough DPCONs, will go on as-is\n");
+		}
+		return ERR_PTR(err);
+	}
+
+	err = dpcon_open(priv->mc_io, 0, dpcon->obj_desc.id, &dpcon->mc_handle);
+	if (err) {
+		dev_err(dev, "dpcon_open() failed\n");
+		goto free;
+	}
+
+	err = dpcon_reset(priv->mc_io, 0, dpcon->mc_handle);
+	if (err) {
+		dev_err(dev, "dpcon_reset() failed\n");
+		goto close;
+	}
+
+	err = dpcon_enable(priv->mc_io, 0, dpcon->mc_handle);
+	if (err) {
+		dev_err(dev, "dpcon_enable() failed\n");
+		goto close;
+	}
+
+	return dpcon;
+
+close:
+	dpcon_close(priv->mc_io, 0, dpcon->mc_handle);
+free:
+	fsl_mc_object_free(dpcon);
+
+	return ERR_PTR(err);
+}
+
+static void dpaa2_eth_free_dpcon(struct dpaa2_eth_priv *priv,
+				 struct fsl_mc_device *dpcon)
+{
+	dpcon_disable(priv->mc_io, 0, dpcon->mc_handle);
+	dpcon_close(priv->mc_io, 0, dpcon->mc_handle);
+	fsl_mc_object_free(dpcon);
+}
+
+static struct dpaa2_eth_channel *dpaa2_eth_alloc_channel(struct dpaa2_eth_priv *priv)
+{
+	struct dpaa2_eth_channel *channel;
+	struct dpcon_attr attr;
+	struct device *dev = priv->net_dev->dev.parent;
+	int err;
+
+	channel = kzalloc(sizeof(*channel), GFP_KERNEL);
+	if (!channel)
+		return NULL;
+
+	channel->dpcon = dpaa2_eth_setup_dpcon(priv);
+	if (IS_ERR(channel->dpcon)) {
+		err = PTR_ERR(channel->dpcon);
+		goto err_setup;
+	}
+
+	err = dpcon_get_attributes(priv->mc_io, 0, channel->dpcon->mc_handle,
+				   &attr);
+	if (err) {
+		dev_err(dev, "dpcon_get_attributes() failed\n");
+		goto err_get_attr;
+	}
+
+	channel->dpcon_id = attr.id;
+	channel->ch_id = attr.qbman_ch_id;
+	channel->priv = priv;
+
+	return channel;
+
+err_get_attr:
+	dpaa2_eth_free_dpcon(priv, channel->dpcon);
+err_setup:
+	kfree(channel);
+	return ERR_PTR(err);
+}
+
+static void dpaa2_eth_free_channel(struct dpaa2_eth_priv *priv,
+				   struct dpaa2_eth_channel *channel)
+{
+	dpaa2_eth_free_dpcon(priv, channel->dpcon);
+	kfree(channel);
+}
+
+/* DPIO setup: allocate and configure QBMan channels, setup core affinity
+ * and register data availability notifications
+ */
+static int dpaa2_eth_setup_dpio(struct dpaa2_eth_priv *priv)
+{
+	struct dpaa2_io_notification_ctx *nctx;
+	struct dpaa2_eth_channel *channel;
+	struct dpcon_notification_cfg dpcon_notif_cfg;
+	struct device *dev = priv->net_dev->dev.parent;
+	int i, err;
+
+	/* We want the ability to spread ingress traffic (RX, TX conf) to as
+	 * many cores as possible, so we need one channel for each core
+	 * (unless there's fewer queues than cores, in which case the extra
+	 * channels would be wasted).
+	 * Allocate one channel per core and register it to the core's
+	 * affine DPIO. If not enough channels are available for all cores
+	 * or if some cores don't have an affine DPIO, there will be no
+	 * ingress frame processing on those cores.
+	 */
+	cpumask_clear(&priv->dpio_cpumask);
+	for_each_online_cpu(i) {
+		/* Try to allocate a channel */
+		channel = dpaa2_eth_alloc_channel(priv);
+		if (IS_ERR_OR_NULL(channel)) {
+			err = PTR_ERR_OR_ZERO(channel);
+			if (err == -EPROBE_DEFER)
+				dev_dbg(dev, "waiting for affine channel\n");
+			else
+				dev_info(dev,
+					 "No affine channel for cpu %d and above\n", i);
+			goto err_alloc_ch;
+		}
+
+		priv->channel[priv->num_channels] = channel;
+
+		nctx = &channel->nctx;
+		nctx->is_cdan = 1;
+		nctx->cb = dpaa2_eth_cdan_cb;
+		nctx->id = channel->ch_id;
+		nctx->desired_cpu = i;
+
+		/* Register the new context */
+		channel->dpio = dpaa2_io_service_select(i);
+		err = dpaa2_io_service_register(channel->dpio, nctx, dev);
+		if (err) {
+			dev_dbg(dev, "No affine DPIO for cpu %d\n", i);
+			/* If no affine DPIO for this core, there's probably
+			 * none available for next cores either. Signal we want
+			 * to retry later, in case the DPIO devices weren't
+			 * probed yet.
+			 */
+			err = -EPROBE_DEFER;
+			goto err_service_reg;
+		}
+
+		/* Register DPCON notification with MC */
+		dpcon_notif_cfg.dpio_id = nctx->dpio_id;
+		dpcon_notif_cfg.priority = 0;
+		dpcon_notif_cfg.user_ctx = nctx->qman64;
+		err = dpcon_set_notification(priv->mc_io, 0,
+					     channel->dpcon->mc_handle,
+					     &dpcon_notif_cfg);
+		if (err) {
+			dev_err(dev, "dpcon_set_notification failed()\n");
+			goto err_set_cdan;
+		}
+
+		/* If we managed to allocate a channel and also found an affine
+		 * DPIO for this core, add it to the final mask
+		 */
+		cpumask_set_cpu(i, &priv->dpio_cpumask);
+		priv->num_channels++;
+
+		/* Stop if we already have enough channels to accommodate all
+		 * RX and TX conf queues
+		 */
+		if (priv->num_channels == priv->dpni_attrs.num_queues)
+			break;
+	}
+
+	return 0;
+
+err_set_cdan:
+	dpaa2_io_service_deregister(channel->dpio, nctx, dev);
+err_service_reg:
+	dpaa2_eth_free_channel(priv, channel);
+err_alloc_ch:
+	if (err == -EPROBE_DEFER) {
+		for (i = 0; i < priv->num_channels; i++) {
+			channel = priv->channel[i];
+			nctx = &channel->nctx;
+			dpaa2_io_service_deregister(channel->dpio, nctx, dev);
+			dpaa2_eth_free_channel(priv, channel);
+		}
+		priv->num_channels = 0;
+		return err;
+	}
+
+	if (cpumask_empty(&priv->dpio_cpumask)) {
+		dev_err(dev, "No cpu with an affine DPIO/DPCON\n");
+		return -ENODEV;
+	}
+
+	dev_info(dev, "Cores %*pbl available for processing ingress traffic\n",
+		 cpumask_pr_args(&priv->dpio_cpumask));
+
+	return 0;
+}
+
+static void dpaa2_eth_free_dpio(struct dpaa2_eth_priv *priv)
+{
+	struct device *dev = priv->net_dev->dev.parent;
+	struct dpaa2_eth_channel *ch;
+	int i;
+
+	/* deregister CDAN notifications and free channels */
+	for (i = 0; i < priv->num_channels; i++) {
+		ch = priv->channel[i];
+		dpaa2_io_service_deregister(ch->dpio, &ch->nctx, dev);
+		dpaa2_eth_free_channel(priv, ch);
+	}
+}
+
+static struct dpaa2_eth_channel *dpaa2_eth_get_affine_channel(struct dpaa2_eth_priv *priv,
+							      int cpu)
+{
+	struct device *dev = priv->net_dev->dev.parent;
+	int i;
+
+	for (i = 0; i < priv->num_channels; i++)
+		if (priv->channel[i]->nctx.desired_cpu == cpu)
+			return priv->channel[i];
+
+	/* We should never get here. Issue a warning and return
+	 * the first channel, because it's still better than nothing
+	 */
+	dev_warn(dev, "No affine channel found for cpu %d\n", cpu);
+
+	return priv->channel[0];
+}
+
+static void dpaa2_eth_set_fq_affinity(struct dpaa2_eth_priv *priv)
+{
+	struct device *dev = priv->net_dev->dev.parent;
+	struct dpaa2_eth_fq *fq;
+	int rx_cpu, txc_cpu;
+	int i;
+
+	/* For each FQ, pick one channel/CPU to deliver frames to.
+	 * This may well change at runtime, either through irqbalance or
+	 * through direct user intervention.
+	 */
+	rx_cpu = txc_cpu = cpumask_first(&priv->dpio_cpumask);
+
+	for (i = 0; i < priv->num_fqs; i++) {
+		fq = &priv->fq[i];
+		switch (fq->type) {
+		case DPAA2_RX_FQ:
+		case DPAA2_RX_ERR_FQ:
+			fq->target_cpu = rx_cpu;
+			rx_cpu = cpumask_next(rx_cpu, &priv->dpio_cpumask);
+			if (rx_cpu >= nr_cpu_ids)
+				rx_cpu = cpumask_first(&priv->dpio_cpumask);
+			break;
+		case DPAA2_TX_CONF_FQ:
+			fq->target_cpu = txc_cpu;
+			txc_cpu = cpumask_next(txc_cpu, &priv->dpio_cpumask);
+			if (txc_cpu >= nr_cpu_ids)
+				txc_cpu = cpumask_first(&priv->dpio_cpumask);
+			break;
+		default:
+			dev_err(dev, "Unknown FQ type: %d\n", fq->type);
+		}
+		fq->channel = dpaa2_eth_get_affine_channel(priv, fq->target_cpu);
+	}
+
+	update_xps(priv);
+}
+
+static void dpaa2_eth_setup_fqs(struct dpaa2_eth_priv *priv)
+{
+	int i, j;
+
+	/* We have one TxConf FQ per Tx flow.
+	 * The number of Tx and Rx queues is the same.
+	 * Tx queues come first in the fq array.
+	 */
+	for (i = 0; i < dpaa2_eth_queue_count(priv); i++) {
+		priv->fq[priv->num_fqs].type = DPAA2_TX_CONF_FQ;
+		priv->fq[priv->num_fqs].consume = dpaa2_eth_tx_conf;
+		priv->fq[priv->num_fqs++].flowid = (u16)i;
+	}
+
+	for (j = 0; j < dpaa2_eth_tc_count(priv); j++) {
+		for (i = 0; i < dpaa2_eth_queue_count(priv); i++) {
+			priv->fq[priv->num_fqs].type = DPAA2_RX_FQ;
+			priv->fq[priv->num_fqs].consume = dpaa2_eth_rx;
+			priv->fq[priv->num_fqs].tc = (u8)j;
+			priv->fq[priv->num_fqs++].flowid = (u16)i;
+		}
+	}
+
+	/* We have exactly one Rx error queue per DPNI */
+	priv->fq[priv->num_fqs].type = DPAA2_RX_ERR_FQ;
+	priv->fq[priv->num_fqs++].consume = dpaa2_eth_rx_err;
+
+	/* For each FQ, decide on which core to process incoming frames */
+	dpaa2_eth_set_fq_affinity(priv);
+}
+
+/* Allocate and configure a buffer pool */
+struct dpaa2_eth_bp *dpaa2_eth_allocate_dpbp(struct dpaa2_eth_priv *priv)
+{
+	struct device *dev = priv->net_dev->dev.parent;
+	struct fsl_mc_device *dpbp_dev;
+	struct dpbp_attr dpbp_attrs;
+	struct dpaa2_eth_bp *bp;
+	int err;
+
+	err = fsl_mc_object_allocate(to_fsl_mc_device(dev), FSL_MC_POOL_DPBP,
+				     &dpbp_dev);
+	if (err) {
+		if (err == -ENXIO)
+			err = -EPROBE_DEFER;
+		else
+			dev_err(dev, "DPBP device allocation failed\n");
+		return ERR_PTR(err);
+	}
+
+	bp = kzalloc(sizeof(*bp), GFP_KERNEL);
+	if (!bp) {
+		err = -ENOMEM;
+		goto err_alloc;
+	}
+
+	err = dpbp_open(priv->mc_io, 0, dpbp_dev->obj_desc.id,
+			&dpbp_dev->mc_handle);
+	if (err) {
+		dev_err(dev, "dpbp_open() failed\n");
+		goto err_open;
+	}
+
+	err = dpbp_reset(priv->mc_io, 0, dpbp_dev->mc_handle);
+	if (err) {
+		dev_err(dev, "dpbp_reset() failed\n");
+		goto err_reset;
+	}
+
+	err = dpbp_enable(priv->mc_io, 0, dpbp_dev->mc_handle);
+	if (err) {
+		dev_err(dev, "dpbp_enable() failed\n");
+		goto err_enable;
+	}
+
+	err = dpbp_get_attributes(priv->mc_io, 0, dpbp_dev->mc_handle,
+				  &dpbp_attrs);
+	if (err) {
+		dev_err(dev, "dpbp_get_attributes() failed\n");
+		goto err_get_attr;
+	}
+
+	bp->dev = dpbp_dev;
+	bp->bpid = dpbp_attrs.bpid;
+
+	return bp;
+
+err_get_attr:
+	dpbp_disable(priv->mc_io, 0, dpbp_dev->mc_handle);
+err_enable:
+err_reset:
+	dpbp_close(priv->mc_io, 0, dpbp_dev->mc_handle);
+err_open:
+	kfree(bp);
+err_alloc:
+	fsl_mc_object_free(dpbp_dev);
+
+	return ERR_PTR(err);
+}
+
+static int dpaa2_eth_setup_default_dpbp(struct dpaa2_eth_priv *priv)
+{
+	struct dpaa2_eth_bp *bp;
+	int i;
+
+	bp = dpaa2_eth_allocate_dpbp(priv);
+	if (IS_ERR(bp))
+		return PTR_ERR(bp);
+
+	priv->bp[DPAA2_ETH_DEFAULT_BP_IDX] = bp;
+	priv->num_bps++;
+
+	for (i = 0; i < priv->num_channels; i++)
+		priv->channel[i]->bp = bp;
+
+	return 0;
+}
+
+void dpaa2_eth_free_dpbp(struct dpaa2_eth_priv *priv, struct dpaa2_eth_bp *bp)
+{
+	int idx_bp;
+
+	/* Find the index at which this BP is stored */
+	for (idx_bp = 0; idx_bp < priv->num_bps; idx_bp++)
+		if (priv->bp[idx_bp] == bp)
+			break;
+
+	/* Drain the pool and disable the associated MC object */
+	dpaa2_eth_drain_pool(priv, bp->bpid);
+	dpbp_disable(priv->mc_io, 0, bp->dev->mc_handle);
+	dpbp_close(priv->mc_io, 0, bp->dev->mc_handle);
+	fsl_mc_object_free(bp->dev);
+	kfree(bp);
+
+	/* Move the last in use DPBP over in this position */
+	priv->bp[idx_bp] = priv->bp[priv->num_bps - 1];
+	priv->num_bps--;
+}
+
+static void dpaa2_eth_free_dpbps(struct dpaa2_eth_priv *priv)
+{
+	int i;
+
+	for (i = 0; i < priv->num_bps; i++)
+		dpaa2_eth_free_dpbp(priv, priv->bp[i]);
+}
+
+static int dpaa2_eth_set_buffer_layout(struct dpaa2_eth_priv *priv)
+{
+	struct device *dev = priv->net_dev->dev.parent;
+	struct dpni_buffer_layout buf_layout = {0};
+	u16 rx_buf_align;
+	int err;
+
+	/* We need to check for WRIOP version 1.0.0, but depending on the MC
+	 * version, this number is not always provided correctly on rev1.
+	 * We need to check for both alternatives in this situation.
+	 */
+	if (priv->dpni_attrs.wriop_version == DPAA2_WRIOP_VERSION(0, 0, 0) ||
+	    priv->dpni_attrs.wriop_version == DPAA2_WRIOP_VERSION(1, 0, 0))
+		rx_buf_align = DPAA2_ETH_RX_BUF_ALIGN_REV1;
+	else
+		rx_buf_align = DPAA2_ETH_RX_BUF_ALIGN;
+
+	/* We need to ensure that the buffer size seen by WRIOP is a multiple
+	 * of 64 or 256 bytes depending on the WRIOP version.
+	 */
+	priv->rx_buf_size = ALIGN_DOWN(DPAA2_ETH_RX_BUF_SIZE, rx_buf_align);
+
+	/* tx buffer */
+	buf_layout.private_data_size = DPAA2_ETH_SWA_SIZE;
+	buf_layout.pass_timestamp = true;
+	buf_layout.pass_frame_status = true;
+	buf_layout.options = DPNI_BUF_LAYOUT_OPT_PRIVATE_DATA_SIZE |
+			     DPNI_BUF_LAYOUT_OPT_TIMESTAMP |
+			     DPNI_BUF_LAYOUT_OPT_FRAME_STATUS;
+	err = dpni_set_buffer_layout(priv->mc_io, 0, priv->mc_token,
+				     DPNI_QUEUE_TX, &buf_layout);
+	if (err) {
+		dev_err(dev, "dpni_set_buffer_layout(TX) failed\n");
+		return err;
+	}
+
+	/* tx-confirm buffer */
+	buf_layout.options = DPNI_BUF_LAYOUT_OPT_TIMESTAMP |
+			     DPNI_BUF_LAYOUT_OPT_FRAME_STATUS;
+	err = dpni_set_buffer_layout(priv->mc_io, 0, priv->mc_token,
+				     DPNI_QUEUE_TX_CONFIRM, &buf_layout);
+	if (err) {
+		dev_err(dev, "dpni_set_buffer_layout(TX_CONF) failed\n");
+		return err;
+	}
+
+	/* Now that we've set our tx buffer layout, retrieve the minimum
+	 * required tx data offset.
+	 */
+	err = dpni_get_tx_data_offset(priv->mc_io, 0, priv->mc_token,
+				      &priv->tx_data_offset);
+	if (err) {
+		dev_err(dev, "dpni_get_tx_data_offset() failed\n");
+		return err;
+	}
+
+	if ((priv->tx_data_offset % 64) != 0)
+		dev_warn(dev, "Tx data offset (%d) not a multiple of 64B\n",
+			 priv->tx_data_offset);
+
+	/* rx buffer */
+	buf_layout.pass_frame_status = true;
+	buf_layout.pass_parser_result = true;
+	buf_layout.data_align = rx_buf_align;
+	buf_layout.data_head_room = dpaa2_eth_rx_head_room(priv);
+	buf_layout.private_data_size = 0;
+	buf_layout.options = DPNI_BUF_LAYOUT_OPT_PARSER_RESULT |
+			     DPNI_BUF_LAYOUT_OPT_FRAME_STATUS |
+			     DPNI_BUF_LAYOUT_OPT_DATA_ALIGN |
+			     DPNI_BUF_LAYOUT_OPT_DATA_HEAD_ROOM |
+			     DPNI_BUF_LAYOUT_OPT_TIMESTAMP;
+	err = dpni_set_buffer_layout(priv->mc_io, 0, priv->mc_token,
+				     DPNI_QUEUE_RX, &buf_layout);
+	if (err) {
+		dev_err(dev, "dpni_set_buffer_layout(RX) failed\n");
+		return err;
+	}
+
+	return 0;
+}
+
+#define DPNI_ENQUEUE_FQID_VER_MAJOR	7
+#define DPNI_ENQUEUE_FQID_VER_MINOR	9
+
+static inline int dpaa2_eth_enqueue_qd(struct dpaa2_eth_priv *priv,
+				       struct dpaa2_eth_fq *fq,
+				       struct dpaa2_fd *fd, u8 prio,
+				       u32 num_frames __always_unused,
+				       int *frames_enqueued)
+{
+	int err;
+
+	err = dpaa2_io_service_enqueue_qd(fq->channel->dpio,
+					  priv->tx_qdid, prio,
+					  fq->tx_qdbin, fd);
+	if (!err && frames_enqueued)
+		*frames_enqueued = 1;
+	return err;
+}
+
+static inline int dpaa2_eth_enqueue_fq_multiple(struct dpaa2_eth_priv *priv,
+						struct dpaa2_eth_fq *fq,
+						struct dpaa2_fd *fd,
+						u8 prio, u32 num_frames,
+						int *frames_enqueued)
+{
+	int err;
+
+	err = dpaa2_io_service_enqueue_multiple_fq(fq->channel->dpio,
+						   fq->tx_fqid[prio],
+						   fd, num_frames);
+
+	if (err == 0)
+		return -EBUSY;
+
+	if (frames_enqueued)
+		*frames_enqueued = err;
+	return 0;
+}
+
+static void dpaa2_eth_set_enqueue_mode(struct dpaa2_eth_priv *priv)
+{
+	if (dpaa2_eth_cmp_dpni_ver(priv, DPNI_ENQUEUE_FQID_VER_MAJOR,
+				   DPNI_ENQUEUE_FQID_VER_MINOR) < 0)
+		priv->enqueue = dpaa2_eth_enqueue_qd;
+	else
+		priv->enqueue = dpaa2_eth_enqueue_fq_multiple;
+}
+
+static int dpaa2_eth_set_pause(struct dpaa2_eth_priv *priv)
+{
+	struct device *dev = priv->net_dev->dev.parent;
+	struct dpni_link_cfg link_cfg = {0};
+	int err;
+
+	/* Get the default link options so we don't override other flags */
+	err = dpni_get_link_cfg(priv->mc_io, 0, priv->mc_token, &link_cfg);
+	if (err) {
+		dev_err(dev, "dpni_get_link_cfg() failed\n");
+		return err;
+	}
+
+	/* By default, enable both Rx and Tx pause frames */
+	link_cfg.options |= DPNI_LINK_OPT_PAUSE;
+	link_cfg.options &= ~DPNI_LINK_OPT_ASYM_PAUSE;
+	err = dpni_set_link_cfg(priv->mc_io, 0, priv->mc_token, &link_cfg);
+	if (err) {
+		dev_err(dev, "dpni_set_link_cfg() failed\n");
+		return err;
+	}
+
+	priv->link_state.options = link_cfg.options;
+
+	return 0;
+}
+
+static void dpaa2_eth_update_tx_fqids(struct dpaa2_eth_priv *priv)
+{
+	struct dpni_queue_id qid = {0};
+	struct dpaa2_eth_fq *fq;
+	struct dpni_queue queue;
+	int i, j, err;
+
+	/* We only use Tx FQIDs for FQID-based enqueue, so check
+	 * if DPNI version supports it before updating FQIDs
+	 */
+	if (dpaa2_eth_cmp_dpni_ver(priv, DPNI_ENQUEUE_FQID_VER_MAJOR,
+				   DPNI_ENQUEUE_FQID_VER_MINOR) < 0)
+		return;
+
+	for (i = 0; i < priv->num_fqs; i++) {
+		fq = &priv->fq[i];
+		if (fq->type != DPAA2_TX_CONF_FQ)
+			continue;
+		for (j = 0; j < dpaa2_eth_tc_count(priv); j++) {
+			err = dpni_get_queue(priv->mc_io, 0, priv->mc_token,
+					     DPNI_QUEUE_TX, j, fq->flowid,
+					     &queue, &qid);
+			if (err)
+				goto out_err;
+
+			fq->tx_fqid[j] = qid.fqid;
+			if (fq->tx_fqid[j] == 0)
+				goto out_err;
+		}
+	}
+
+	priv->enqueue = dpaa2_eth_enqueue_fq_multiple;
+
+	return;
+
+out_err:
+	netdev_info(priv->net_dev,
+		    "Error reading Tx FQID, fallback to QDID-based enqueue\n");
+	priv->enqueue = dpaa2_eth_enqueue_qd;
+}
+
+/* Configure ingress classification based on VLAN PCP */
+static int dpaa2_eth_set_vlan_qos(struct dpaa2_eth_priv *priv)
+{
+	struct device *dev = priv->net_dev->dev.parent;
+	struct dpkg_profile_cfg kg_cfg = {0};
+	struct dpni_qos_tbl_cfg qos_cfg = {0};
+	struct dpni_rule_cfg key_params;
+	void *dma_mem, *key, *mask;
+	u8 key_size = 2;	/* VLAN TCI field */
+	int i, pcp, err;
+
+	/* VLAN-based classification only makes sense if we have multiple
+	 * traffic classes.
+	 * Also, we need to extract just the 3-bit PCP field from the VLAN
+	 * header and we can only do that by using a mask
+	 */
+	if (dpaa2_eth_tc_count(priv) == 1 || !dpaa2_eth_fs_mask_enabled(priv)) {
+		dev_dbg(dev, "VLAN-based QoS classification not supported\n");
+		return -EOPNOTSUPP;
+	}
+
+	dma_mem = kzalloc(DPAA2_CLASSIFIER_DMA_SIZE, GFP_KERNEL);
+	if (!dma_mem)
+		return -ENOMEM;
+
+	kg_cfg.num_extracts = 1;
+	kg_cfg.extracts[0].type = DPKG_EXTRACT_FROM_HDR;
+	kg_cfg.extracts[0].extract.from_hdr.prot = NET_PROT_VLAN;
+	kg_cfg.extracts[0].extract.from_hdr.type = DPKG_FULL_FIELD;
+	kg_cfg.extracts[0].extract.from_hdr.field = NH_FLD_VLAN_TCI;
+
+	err = dpni_prepare_key_cfg(&kg_cfg, dma_mem);
+	if (err) {
+		dev_err(dev, "dpni_prepare_key_cfg failed\n");
+		goto out_free_tbl;
+	}
+
+	/* set QoS table */
+	qos_cfg.default_tc = 0;
+	qos_cfg.discard_on_miss = 0;
+	qos_cfg.key_cfg_iova = dma_map_single(dev, dma_mem,
+					      DPAA2_CLASSIFIER_DMA_SIZE,
+					      DMA_TO_DEVICE);
+	if (dma_mapping_error(dev, qos_cfg.key_cfg_iova)) {
+		dev_err(dev, "QoS table DMA mapping failed\n");
+		err = -ENOMEM;
+		goto out_free_tbl;
+	}
+
+	err = dpni_set_qos_table(priv->mc_io, 0, priv->mc_token, &qos_cfg);
+	if (err) {
+		dev_err(dev, "dpni_set_qos_table failed\n");
+		goto out_unmap_tbl;
+	}
+
+	/* Add QoS table entries */
+	key = kzalloc(key_size * 2, GFP_KERNEL);
+	if (!key) {
+		err = -ENOMEM;
+		goto out_unmap_tbl;
+	}
+	mask = key + key_size;
+	*(__be16 *)mask = cpu_to_be16(VLAN_PRIO_MASK);
+
+	key_params.key_iova = dma_map_single(dev, key, key_size * 2,
+					     DMA_TO_DEVICE);
+	if (dma_mapping_error(dev, key_params.key_iova)) {
+		dev_err(dev, "Qos table entry DMA mapping failed\n");
+		err = -ENOMEM;
+		goto out_free_key;
+	}
+
+	key_params.mask_iova = key_params.key_iova + key_size;
+	key_params.key_size = key_size;
+
+	/* We add rules for PCP-based distribution starting with highest
+	 * priority (VLAN PCP = 7). If this DPNI doesn't have enough traffic
+	 * classes to accommodate all priority levels, the lowest ones end up
+	 * on TC 0 which was configured as default
+	 */
+	for (i = dpaa2_eth_tc_count(priv) - 1, pcp = 7; i >= 0; i--, pcp--) {
+		*(__be16 *)key = cpu_to_be16(pcp << VLAN_PRIO_SHIFT);
+		dma_sync_single_for_device(dev, key_params.key_iova,
+					   key_size * 2, DMA_TO_DEVICE);
+
+		err = dpni_add_qos_entry(priv->mc_io, 0, priv->mc_token,
+					 &key_params, i, i);
+		if (err) {
+			dev_err(dev, "dpni_add_qos_entry failed\n");
+			dpni_clear_qos_table(priv->mc_io, 0, priv->mc_token);
+			goto out_unmap_key;
+		}
+	}
+
+	priv->vlan_cls_enabled = true;
+
+	/* Table and key memory is not persistent, clean everything up after
+	 * configuration is finished
+	 */
+out_unmap_key:
+	dma_unmap_single(dev, key_params.key_iova, key_size * 2, DMA_TO_DEVICE);
+out_free_key:
+	kfree(key);
+out_unmap_tbl:
+	dma_unmap_single(dev, qos_cfg.key_cfg_iova, DPAA2_CLASSIFIER_DMA_SIZE,
+			 DMA_TO_DEVICE);
+out_free_tbl:
+	kfree(dma_mem);
+
+	return err;
+}
+
+/* Configure the DPNI object this interface is associated with */
+static int dpaa2_eth_setup_dpni(struct fsl_mc_device *ls_dev)
+{
+	struct device *dev = &ls_dev->dev;
+	struct dpaa2_eth_priv *priv;
+	struct net_device *net_dev;
+	int err;
+
+	net_dev = dev_get_drvdata(dev);
+	priv = netdev_priv(net_dev);
+
+	/* get a handle for the DPNI object */
+	err = dpni_open(priv->mc_io, 0, ls_dev->obj_desc.id, &priv->mc_token);
+	if (err) {
+		dev_err(dev, "dpni_open() failed\n");
+		return err;
+	}
+
+	/* Check if we can work with this DPNI object */
+	err = dpni_get_api_version(priv->mc_io, 0, &priv->dpni_ver_major,
+				   &priv->dpni_ver_minor);
+	if (err) {
+		dev_err(dev, "dpni_get_api_version() failed\n");
+		goto close;
+	}
+	if (dpaa2_eth_cmp_dpni_ver(priv, DPNI_VER_MAJOR, DPNI_VER_MINOR) < 0) {
+		dev_err(dev, "DPNI version %u.%u not supported, need >= %u.%u\n",
+			priv->dpni_ver_major, priv->dpni_ver_minor,
+			DPNI_VER_MAJOR, DPNI_VER_MINOR);
+		err = -EOPNOTSUPP;
+		goto close;
+	}
+
+	ls_dev->mc_io = priv->mc_io;
+	ls_dev->mc_handle = priv->mc_token;
+
+	err = dpni_reset(priv->mc_io, 0, priv->mc_token);
+	if (err) {
+		dev_err(dev, "dpni_reset() failed\n");
+		goto close;
+	}
+
+	err = dpni_get_attributes(priv->mc_io, 0, priv->mc_token,
+				  &priv->dpni_attrs);
+	if (err) {
+		dev_err(dev, "dpni_get_attributes() failed (err=%d)\n", err);
+		goto close;
+	}
+
+	err = dpaa2_eth_set_buffer_layout(priv);
+	if (err)
+		goto close;
+
+	dpaa2_eth_set_enqueue_mode(priv);
+
+	/* Enable pause frame support */
+	if (dpaa2_eth_has_pause_support(priv)) {
+		err = dpaa2_eth_set_pause(priv);
+		if (err)
+			goto close;
+	}
+
+	err = dpaa2_eth_set_vlan_qos(priv);
+	if (err && err != -EOPNOTSUPP)
+		goto close;
+
+	priv->cls_rules = devm_kcalloc(dev, dpaa2_eth_fs_count(priv),
+				       sizeof(struct dpaa2_eth_cls_rule),
+				       GFP_KERNEL);
+	if (!priv->cls_rules) {
+		err = -ENOMEM;
+		goto close;
+	}
+
+	return 0;
+
+close:
+	dpni_close(priv->mc_io, 0, priv->mc_token);
+
+	return err;
+}
+
+static void dpaa2_eth_free_dpni(struct dpaa2_eth_priv *priv)
+{
+	int err;
+
+	err = dpni_reset(priv->mc_io, 0, priv->mc_token);
+	if (err)
+		netdev_warn(priv->net_dev, "dpni_reset() failed (err %d)\n",
+			    err);
+
+	dpni_close(priv->mc_io, 0, priv->mc_token);
+}
+
+static int dpaa2_eth_setup_rx_flow(struct dpaa2_eth_priv *priv,
+				   struct dpaa2_eth_fq *fq)
+{
+	struct device *dev = priv->net_dev->dev.parent;
+	struct dpni_queue queue;
+	struct dpni_queue_id qid;
+	int err;
+
+	err = dpni_get_queue(priv->mc_io, 0, priv->mc_token,
+			     DPNI_QUEUE_RX, fq->tc, fq->flowid, &queue, &qid);
+	if (err) {
+		dev_err(dev, "dpni_get_queue(RX) failed\n");
+		return err;
+	}
+
+	fq->fqid = qid.fqid;
+
+	queue.destination.id = fq->channel->dpcon_id;
+	queue.destination.type = DPNI_DEST_DPCON;
+	queue.destination.priority = 1;
+	queue.user_context = (u64)(uintptr_t)fq;
+	err = dpni_set_queue(priv->mc_io, 0, priv->mc_token,
+			     DPNI_QUEUE_RX, fq->tc, fq->flowid,
+			     DPNI_QUEUE_OPT_USER_CTX | DPNI_QUEUE_OPT_DEST,
+			     &queue);
+	if (err) {
+		dev_err(dev, "dpni_set_queue(RX) failed\n");
+		return err;
+	}
+
+	/* xdp_rxq setup */
+	/* only once for each channel */
+	if (fq->tc > 0)
+		return 0;
+
+	err = xdp_rxq_info_reg(&fq->channel->xdp_rxq, priv->net_dev,
+			       fq->flowid, 0);
+	if (err) {
+		dev_err(dev, "xdp_rxq_info_reg failed\n");
+		return err;
+	}
+
+	err = xdp_rxq_info_reg_mem_model(&fq->channel->xdp_rxq,
+					 MEM_TYPE_PAGE_ORDER0, NULL);
+	if (err) {
+		dev_err(dev, "xdp_rxq_info_reg_mem_model failed\n");
+		xdp_rxq_info_unreg(&fq->channel->xdp_rxq);
+		return err;
+	}
+
+	return 0;
+}
+
+static int dpaa2_eth_setup_tx_flow(struct dpaa2_eth_priv *priv,
+				   struct dpaa2_eth_fq *fq)
+{
+	struct device *dev = priv->net_dev->dev.parent;
+	struct dpni_queue queue;
+	struct dpni_queue_id qid;
+	int i, err;
+
+	for (i = 0; i < dpaa2_eth_tc_count(priv); i++) {
+		err = dpni_get_queue(priv->mc_io, 0, priv->mc_token,
+				     DPNI_QUEUE_TX, i, fq->flowid,
+				     &queue, &qid);
+		if (err) {
+			dev_err(dev, "dpni_get_queue(TX) failed\n");
+			return err;
+		}
+		fq->tx_fqid[i] = qid.fqid;
+	}
+
+	/* All Tx queues belonging to the same flowid have the same qdbin */
+	fq->tx_qdbin = qid.qdbin;
+
+	err = dpni_get_queue(priv->mc_io, 0, priv->mc_token,
+			     DPNI_QUEUE_TX_CONFIRM, 0, fq->flowid,
+			     &queue, &qid);
+	if (err) {
+		dev_err(dev, "dpni_get_queue(TX_CONF) failed\n");
+		return err;
+	}
+
+	fq->fqid = qid.fqid;
+
+	queue.destination.id = fq->channel->dpcon_id;
+	queue.destination.type = DPNI_DEST_DPCON;
+	queue.destination.priority = 0;
+	queue.user_context = (u64)(uintptr_t)fq;
+	err = dpni_set_queue(priv->mc_io, 0, priv->mc_token,
+			     DPNI_QUEUE_TX_CONFIRM, 0, fq->flowid,
+			     DPNI_QUEUE_OPT_USER_CTX | DPNI_QUEUE_OPT_DEST,
+			     &queue);
+	if (err) {
+		dev_err(dev, "dpni_set_queue(TX_CONF) failed\n");
+		return err;
+	}
+
+	return 0;
+}
+
+static int setup_rx_err_flow(struct dpaa2_eth_priv *priv,
+			     struct dpaa2_eth_fq *fq)
+{
+	struct device *dev = priv->net_dev->dev.parent;
+	struct dpni_queue q = { { 0 } };
+	struct dpni_queue_id qid;
+	u8 q_opt = DPNI_QUEUE_OPT_USER_CTX | DPNI_QUEUE_OPT_DEST;
+	int err;
+
+	err = dpni_get_queue(priv->mc_io, 0, priv->mc_token,
+			     DPNI_QUEUE_RX_ERR, 0, 0, &q, &qid);
+	if (err) {
+		dev_err(dev, "dpni_get_queue() failed (%d)\n", err);
+		return err;
+	}
+
+	fq->fqid = qid.fqid;
+
+	q.destination.id = fq->channel->dpcon_id;
+	q.destination.type = DPNI_DEST_DPCON;
+	q.destination.priority = 1;
+	q.user_context = (u64)(uintptr_t)fq;
+	err = dpni_set_queue(priv->mc_io, 0, priv->mc_token,
+			     DPNI_QUEUE_RX_ERR, 0, 0, q_opt, &q);
+	if (err) {
+		dev_err(dev, "dpni_set_queue() failed (%d)\n", err);
+		return err;
+	}
+
+	return 0;
+}
+
+/* Supported header fields for Rx hash distribution key */
+static const struct dpaa2_eth_dist_fields dist_fields[] = {
+	{
+		/* L2 header */
+		.rxnfc_field = RXH_L2DA,
+		.cls_prot = NET_PROT_ETH,
+		.cls_field = NH_FLD_ETH_DA,
+		.id = DPAA2_ETH_DIST_ETHDST,
+		.size = 6,
+	}, {
+		.cls_prot = NET_PROT_ETH,
+		.cls_field = NH_FLD_ETH_SA,
+		.id = DPAA2_ETH_DIST_ETHSRC,
+		.size = 6,
+	}, {
+		/* This is the last ethertype field parsed:
+		 * depending on frame format, it can be the MAC ethertype
+		 * or the VLAN etype.
+		 */
+		.cls_prot = NET_PROT_ETH,
+		.cls_field = NH_FLD_ETH_TYPE,
+		.id = DPAA2_ETH_DIST_ETHTYPE,
+		.size = 2,
+	}, {
+		/* VLAN header */
+		.rxnfc_field = RXH_VLAN,
+		.cls_prot = NET_PROT_VLAN,
+		.cls_field = NH_FLD_VLAN_TCI,
+		.id = DPAA2_ETH_DIST_VLAN,
+		.size = 2,
+	}, {
+		/* IP header */
+		.rxnfc_field = RXH_IP_SRC,
+		.cls_prot = NET_PROT_IP,
+		.cls_field = NH_FLD_IP_SRC,
+		.id = DPAA2_ETH_DIST_IPSRC,
+		.size = 4,
+	}, {
+		.rxnfc_field = RXH_IP_DST,
+		.cls_prot = NET_PROT_IP,
+		.cls_field = NH_FLD_IP_DST,
+		.id = DPAA2_ETH_DIST_IPDST,
+		.size = 4,
+	}, {
+		.rxnfc_field = RXH_L3_PROTO,
+		.cls_prot = NET_PROT_IP,
+		.cls_field = NH_FLD_IP_PROTO,
+		.id = DPAA2_ETH_DIST_IPPROTO,
+		.size = 1,
+	}, {
+		/* Using UDP ports, this is functionally equivalent to raw
+		 * byte pairs from L4 header.
+		 */
+		.rxnfc_field = RXH_L4_B_0_1,
+		.cls_prot = NET_PROT_UDP,
+		.cls_field = NH_FLD_UDP_PORT_SRC,
+		.id = DPAA2_ETH_DIST_L4SRC,
+		.size = 2,
+	}, {
+		.rxnfc_field = RXH_L4_B_2_3,
+		.cls_prot = NET_PROT_UDP,
+		.cls_field = NH_FLD_UDP_PORT_DST,
+		.id = DPAA2_ETH_DIST_L4DST,
+		.size = 2,
+	},
+};
+
+/* Configure the Rx hash key using the legacy API */
+static int dpaa2_eth_config_legacy_hash_key(struct dpaa2_eth_priv *priv, dma_addr_t key)
+{
+	struct device *dev = priv->net_dev->dev.parent;
+	struct dpni_rx_tc_dist_cfg dist_cfg;
+	int i, err = 0;
+
+	memset(&dist_cfg, 0, sizeof(dist_cfg));
+
+	dist_cfg.key_cfg_iova = key;
+	dist_cfg.dist_size = dpaa2_eth_queue_count(priv);
+	dist_cfg.dist_mode = DPNI_DIST_MODE_HASH;
+
+	for (i = 0; i < dpaa2_eth_tc_count(priv); i++) {
+		err = dpni_set_rx_tc_dist(priv->mc_io, 0, priv->mc_token,
+					  i, &dist_cfg);
+		if (err) {
+			dev_err(dev, "dpni_set_rx_tc_dist failed\n");
+			break;
+		}
+	}
+
+	return err;
+}
+
+/* Configure the Rx hash key using the new API */
+static int dpaa2_eth_config_hash_key(struct dpaa2_eth_priv *priv, dma_addr_t key)
+{
+	struct device *dev = priv->net_dev->dev.parent;
+	struct dpni_rx_dist_cfg dist_cfg;
+	int i, err = 0;
+
+	memset(&dist_cfg, 0, sizeof(dist_cfg));
+
+	dist_cfg.key_cfg_iova = key;
+	dist_cfg.dist_size = dpaa2_eth_queue_count(priv);
+	dist_cfg.enable = 1;
+
+	for (i = 0; i < dpaa2_eth_tc_count(priv); i++) {
+		dist_cfg.tc = i;
+		err = dpni_set_rx_hash_dist(priv->mc_io, 0, priv->mc_token,
+					    &dist_cfg);
+		if (err) {
+			dev_err(dev, "dpni_set_rx_hash_dist failed\n");
+			break;
+		}
+
+		/* If the flow steering / hashing key is shared between all
+		 * traffic classes, install it just once
+		 */
+		if (priv->dpni_attrs.options & DPNI_OPT_SHARED_FS)
+			break;
+	}
+
+	return err;
+}
+
+/* Configure the Rx flow classification key */
+static int dpaa2_eth_config_cls_key(struct dpaa2_eth_priv *priv, dma_addr_t key)
+{
+	struct device *dev = priv->net_dev->dev.parent;
+	struct dpni_rx_dist_cfg dist_cfg;
+	int i, err = 0;
+
+	memset(&dist_cfg, 0, sizeof(dist_cfg));
+
+	dist_cfg.key_cfg_iova = key;
+	dist_cfg.dist_size = dpaa2_eth_queue_count(priv);
+	dist_cfg.enable = 1;
+
+	for (i = 0; i < dpaa2_eth_tc_count(priv); i++) {
+		dist_cfg.tc = i;
+		err = dpni_set_rx_fs_dist(priv->mc_io, 0, priv->mc_token,
+					  &dist_cfg);
+		if (err) {
+			dev_err(dev, "dpni_set_rx_fs_dist failed\n");
+			break;
+		}
+
+		/* If the flow steering / hashing key is shared between all
+		 * traffic classes, install it just once
+		 */
+		if (priv->dpni_attrs.options & DPNI_OPT_SHARED_FS)
+			break;
+	}
+
+	return err;
+}
+
+/* Size of the Rx flow classification key */
+int dpaa2_eth_cls_key_size(u64 fields)
+{
+	int i, size = 0;
+
+	for (i = 0; i < ARRAY_SIZE(dist_fields); i++) {
+		if (!(fields & dist_fields[i].id))
+			continue;
+		size += dist_fields[i].size;
+	}
+
+	return size;
+}
+
+/* Offset of header field in Rx classification key */
+int dpaa2_eth_cls_fld_off(int prot, int field)
+{
+	int i, off = 0;
+
+	for (i = 0; i < ARRAY_SIZE(dist_fields); i++) {
+		if (dist_fields[i].cls_prot == prot &&
+		    dist_fields[i].cls_field == field)
+			return off;
+		off += dist_fields[i].size;
+	}
+
+	WARN_ONCE(1, "Unsupported header field used for Rx flow cls\n");
+	return 0;
+}
+
+/* Prune unused fields from the classification rule.
+ * Used when masking is not supported
+ */
+void dpaa2_eth_cls_trim_rule(void *key_mem, u64 fields)
+{
+	int off = 0, new_off = 0;
+	int i, size;
+
+	for (i = 0; i < ARRAY_SIZE(dist_fields); i++) {
+		size = dist_fields[i].size;
+		if (dist_fields[i].id & fields) {
+			memcpy(key_mem + new_off, key_mem + off, size);
+			new_off += size;
+		}
+		off += size;
+	}
+}
+
+/* Set Rx distribution (hash or flow classification) key
+ * flags is a combination of RXH_ bits
+ */
+static int dpaa2_eth_set_dist_key(struct net_device *net_dev,
+				  enum dpaa2_eth_rx_dist type, u64 flags)
+{
+	struct device *dev = net_dev->dev.parent;
+	struct dpaa2_eth_priv *priv = netdev_priv(net_dev);
+	struct dpkg_profile_cfg cls_cfg;
+	u32 rx_hash_fields = 0;
+	dma_addr_t key_iova;
+	u8 *dma_mem;
+	int i;
+	int err = 0;
+
+	memset(&cls_cfg, 0, sizeof(cls_cfg));
+
+	for (i = 0; i < ARRAY_SIZE(dist_fields); i++) {
+		struct dpkg_extract *key =
+			&cls_cfg.extracts[cls_cfg.num_extracts];
+
+		/* For both Rx hashing and classification keys
+		 * we set only the selected fields.
+		 */
+		if (!(flags & dist_fields[i].id))
+			continue;
+		if (type == DPAA2_ETH_RX_DIST_HASH)
+			rx_hash_fields |= dist_fields[i].rxnfc_field;
+
+		if (cls_cfg.num_extracts >= DPKG_MAX_NUM_OF_EXTRACTS) {
+			dev_err(dev, "error adding key extraction rule, too many rules?\n");
+			return -E2BIG;
+		}
+
+		key->type = DPKG_EXTRACT_FROM_HDR;
+		key->extract.from_hdr.prot = dist_fields[i].cls_prot;
+		key->extract.from_hdr.type = DPKG_FULL_FIELD;
+		key->extract.from_hdr.field = dist_fields[i].cls_field;
+		cls_cfg.num_extracts++;
+	}
+
+	dma_mem = kzalloc(DPAA2_CLASSIFIER_DMA_SIZE, GFP_KERNEL);
+	if (!dma_mem)
+		return -ENOMEM;
+
+	err = dpni_prepare_key_cfg(&cls_cfg, dma_mem);
+	if (err) {
+		dev_err(dev, "dpni_prepare_key_cfg error %d\n", err);
+		goto free_key;
+	}
+
+	/* Prepare for setting the rx dist */
+	key_iova = dma_map_single(dev, dma_mem, DPAA2_CLASSIFIER_DMA_SIZE,
+				  DMA_TO_DEVICE);
+	if (dma_mapping_error(dev, key_iova)) {
+		dev_err(dev, "DMA mapping failed\n");
+		err = -ENOMEM;
+		goto free_key;
+	}
+
+	if (type == DPAA2_ETH_RX_DIST_HASH) {
+		if (dpaa2_eth_has_legacy_dist(priv))
+			err = dpaa2_eth_config_legacy_hash_key(priv, key_iova);
+		else
+			err = dpaa2_eth_config_hash_key(priv, key_iova);
+	} else {
+		err = dpaa2_eth_config_cls_key(priv, key_iova);
+	}
+
+	dma_unmap_single(dev, key_iova, DPAA2_CLASSIFIER_DMA_SIZE,
+			 DMA_TO_DEVICE);
+	if (!err && type == DPAA2_ETH_RX_DIST_HASH)
+		priv->rx_hash_fields = rx_hash_fields;
+
+free_key:
+	kfree(dma_mem);
+	return err;
+}
+
+int dpaa2_eth_set_hash(struct net_device *net_dev, u64 flags)
+{
+	struct dpaa2_eth_priv *priv = netdev_priv(net_dev);
+	u64 key = 0;
+	int i;
+
+	if (!dpaa2_eth_hash_enabled(priv))
+		return -EOPNOTSUPP;
+
+	for (i = 0; i < ARRAY_SIZE(dist_fields); i++)
+		if (dist_fields[i].rxnfc_field & flags)
+			key |= dist_fields[i].id;
+
+	return dpaa2_eth_set_dist_key(net_dev, DPAA2_ETH_RX_DIST_HASH, key);
+}
+
+int dpaa2_eth_set_cls(struct net_device *net_dev, u64 flags)
+{
+	return dpaa2_eth_set_dist_key(net_dev, DPAA2_ETH_RX_DIST_CLS, flags);
+}
+
+static int dpaa2_eth_set_default_cls(struct dpaa2_eth_priv *priv)
+{
+	struct device *dev = priv->net_dev->dev.parent;
+	int err;
+
+	/* Check if we actually support Rx flow classification */
+	if (dpaa2_eth_has_legacy_dist(priv)) {
+		dev_dbg(dev, "Rx cls not supported by current MC version\n");
+		return -EOPNOTSUPP;
+	}
+
+	if (!dpaa2_eth_fs_enabled(priv)) {
+		dev_dbg(dev, "Rx cls disabled in DPNI options\n");
+		return -EOPNOTSUPP;
+	}
+
+	if (!dpaa2_eth_hash_enabled(priv)) {
+		dev_dbg(dev, "Rx cls disabled for single queue DPNIs\n");
+		return -EOPNOTSUPP;
+	}
+
+	/* If there is no support for masking in the classification table,
+	 * we don't set a default key, as it will depend on the rules
+	 * added by the user at runtime.
+	 */
+	if (!dpaa2_eth_fs_mask_enabled(priv))
+		goto out;
+
+	err = dpaa2_eth_set_cls(priv->net_dev, DPAA2_ETH_DIST_ALL);
+	if (err)
+		return err;
+
+out:
+	priv->rx_cls_enabled = 1;
+
+	return 0;
+}
+
+/* Bind the DPNI to its needed objects and resources: buffer pool, DPIOs,
+ * frame queues and channels
+ */
+static int dpaa2_eth_bind_dpni(struct dpaa2_eth_priv *priv)
+{
+	struct dpaa2_eth_bp *bp = priv->bp[DPAA2_ETH_DEFAULT_BP_IDX];
+	struct net_device *net_dev = priv->net_dev;
+	struct dpni_pools_cfg pools_params = { 0 };
+	struct device *dev = net_dev->dev.parent;
+	struct dpni_error_cfg err_cfg;
+	int err = 0;
+	int i;
+
+	pools_params.num_dpbp = 1;
+	pools_params.pools[0].dpbp_id = bp->dev->obj_desc.id;
+	pools_params.pools[0].backup_pool = 0;
+	pools_params.pools[0].buffer_size = priv->rx_buf_size;
+	err = dpni_set_pools(priv->mc_io, 0, priv->mc_token, &pools_params);
+	if (err) {
+		dev_err(dev, "dpni_set_pools() failed\n");
+		return err;
+	}
+
+	/* have the interface implicitly distribute traffic based on
+	 * the default hash key
+	 */
+	err = dpaa2_eth_set_hash(net_dev, DPAA2_RXH_DEFAULT);
+	if (err && err != -EOPNOTSUPP)
+		dev_err(dev, "Failed to configure hashing\n");
+
+	/* Configure the flow classification key; it includes all
+	 * supported header fields and cannot be modified at runtime
+	 */
+	err = dpaa2_eth_set_default_cls(priv);
+	if (err && err != -EOPNOTSUPP)
+		dev_err(dev, "Failed to configure Rx classification key\n");
+
+	/* Configure handling of error frames */
+	err_cfg.errors = DPAA2_FAS_RX_ERR_MASK;
+	err_cfg.set_frame_annotation = 1;
+	err_cfg.error_action = DPNI_ERROR_ACTION_DISCARD;
+	err = dpni_set_errors_behavior(priv->mc_io, 0, priv->mc_token,
+				       &err_cfg);
+	if (err) {
+		dev_err(dev, "dpni_set_errors_behavior failed\n");
+		return err;
+	}
+
+	/* Configure Rx and Tx conf queues to generate CDANs */
+	for (i = 0; i < priv->num_fqs; i++) {
+		switch (priv->fq[i].type) {
+		case DPAA2_RX_FQ:
+			err = dpaa2_eth_setup_rx_flow(priv, &priv->fq[i]);
+			break;
+		case DPAA2_TX_CONF_FQ:
+			err = dpaa2_eth_setup_tx_flow(priv, &priv->fq[i]);
+			break;
+		case DPAA2_RX_ERR_FQ:
+			err = setup_rx_err_flow(priv, &priv->fq[i]);
+			break;
+		default:
+			dev_err(dev, "Invalid FQ type %d\n", priv->fq[i].type);
+			return -EINVAL;
+		}
+		if (err)
+			goto out;
+	}
+
+	err = dpni_get_qdid(priv->mc_io, 0, priv->mc_token,
+			    DPNI_QUEUE_TX, &priv->tx_qdid);
+	if (err) {
+		dev_err(dev, "dpni_get_qdid() failed\n");
+		goto out;
+	}
+
+	return 0;
+
+out:
+	while (i--) {
+		if (priv->fq[i].type == DPAA2_RX_FQ &&
+		    xdp_rxq_info_is_reg(&priv->fq[i].channel->xdp_rxq))
+			xdp_rxq_info_unreg(&priv->fq[i].channel->xdp_rxq);
+	}
+	return err;
+}
+
+/* Allocate rings for storing incoming frame descriptors */
+static int dpaa2_eth_alloc_rings(struct dpaa2_eth_priv *priv)
+{
+	struct net_device *net_dev = priv->net_dev;
+	struct device *dev = net_dev->dev.parent;
+	int i;
+
+	for (i = 0; i < priv->num_channels; i++) {
+		priv->channel[i]->store =
+			dpaa2_io_store_create(DPAA2_ETH_STORE_SIZE, dev);
+		if (!priv->channel[i]->store) {
+			netdev_err(net_dev, "dpaa2_io_store_create() failed\n");
+			goto err_ring;
+		}
+	}
+
+	return 0;
+
+err_ring:
+	for (i = 0; i < priv->num_channels; i++) {
+		if (!priv->channel[i]->store)
+			break;
+		dpaa2_io_store_destroy(priv->channel[i]->store);
+	}
+
+	return -ENOMEM;
+}
+
+static void dpaa2_eth_free_rings(struct dpaa2_eth_priv *priv)
+{
+	int i;
+
+	for (i = 0; i < priv->num_channels; i++)
+		dpaa2_io_store_destroy(priv->channel[i]->store);
+}
+
+static int dpaa2_eth_set_mac_addr(struct dpaa2_eth_priv *priv)
+{
+	struct net_device *net_dev = priv->net_dev;
+	struct device *dev = net_dev->dev.parent;
+	u8 mac_addr[ETH_ALEN], dpni_mac_addr[ETH_ALEN];
+	int err;
+
+	/* Get firmware address, if any */
+	err = dpni_get_port_mac_addr(priv->mc_io, 0, priv->mc_token, mac_addr);
+	if (err) {
+		dev_err(dev, "dpni_get_port_mac_addr() failed\n");
+		return err;
+	}
+
+	/* Get DPNI attributes address, if any */
+	err = dpni_get_primary_mac_addr(priv->mc_io, 0, priv->mc_token,
+					dpni_mac_addr);
+	if (err) {
+		dev_err(dev, "dpni_get_primary_mac_addr() failed\n");
+		return err;
+	}
+
+	/* First check if firmware has any address configured by bootloader */
+	if (!is_zero_ether_addr(mac_addr)) {
+		/* If the DPMAC addr != DPNI addr, update it */
+		if (!ether_addr_equal(mac_addr, dpni_mac_addr)) {
+			err = dpni_set_primary_mac_addr(priv->mc_io, 0,
+							priv->mc_token,
+							mac_addr);
+			if (err) {
+				dev_err(dev, "dpni_set_primary_mac_addr() failed\n");
+				return err;
+			}
+		}
+		eth_hw_addr_set(net_dev, mac_addr);
+	} else if (is_zero_ether_addr(dpni_mac_addr)) {
+		/* No MAC address configured, fill in net_dev->dev_addr
+		 * with a random one
+		 */
+		eth_hw_addr_random(net_dev);
+		dev_dbg_once(dev, "device(s) have all-zero hwaddr, replaced with random\n");
+
+		err = dpni_set_primary_mac_addr(priv->mc_io, 0, priv->mc_token,
+						net_dev->dev_addr);
+		if (err) {
+			dev_err(dev, "dpni_set_primary_mac_addr() failed\n");
+			return err;
+		}
+
+		/* Override NET_ADDR_RANDOM set by eth_hw_addr_random(); for all
+		 * practical purposes, this will be our "permanent" mac address,
+		 * at least until the next reboot. This move will also permit
+		 * register_netdevice() to properly fill up net_dev->perm_addr.
+		 */
+		net_dev->addr_assign_type = NET_ADDR_PERM;
+	} else {
+		/* NET_ADDR_PERM is default, all we have to do is
+		 * fill in the device addr.
+		 */
+		eth_hw_addr_set(net_dev, dpni_mac_addr);
+	}
+
+	return 0;
+}
+
+static int dpaa2_eth_netdev_init(struct net_device *net_dev)
+{
+	struct device *dev = net_dev->dev.parent;
+	struct dpaa2_eth_priv *priv = netdev_priv(net_dev);
+	u32 options = priv->dpni_attrs.options;
+	u64 supported = 0, not_supported = 0;
+	u8 bcast_addr[ETH_ALEN];
+	u8 num_queues;
+	int err;
+
+	net_dev->netdev_ops = &dpaa2_eth_ops;
+	net_dev->ethtool_ops = &dpaa2_ethtool_ops;
+
+	err = dpaa2_eth_set_mac_addr(priv);
+	if (err)
+		return err;
+
+	/* Explicitly add the broadcast address to the MAC filtering table */
+	eth_broadcast_addr(bcast_addr);
+	err = dpni_add_mac_addr(priv->mc_io, 0, priv->mc_token, bcast_addr);
+	if (err) {
+		dev_err(dev, "dpni_add_mac_addr() failed\n");
+		return err;
+	}
+
+	/* Set MTU upper limit; lower limit is 68B (default value) */
+	net_dev->max_mtu = DPAA2_ETH_MAX_MTU;
+	err = dpni_set_max_frame_length(priv->mc_io, 0, priv->mc_token,
+					DPAA2_ETH_MFL);
+	if (err) {
+		dev_err(dev, "dpni_set_max_frame_length() failed\n");
+		return err;
+	}
+
+	/* Set actual number of queues in the net device */
+	num_queues = dpaa2_eth_queue_count(priv);
+	err = netif_set_real_num_tx_queues(net_dev, num_queues);
+	if (err) {
+		dev_err(dev, "netif_set_real_num_tx_queues() failed\n");
+		return err;
+	}
+	err = netif_set_real_num_rx_queues(net_dev, num_queues);
+	if (err) {
+		dev_err(dev, "netif_set_real_num_rx_queues() failed\n");
+		return err;
+	}
+
+	dpaa2_eth_detect_features(priv);
+
+	/* Capabilities listing */
+	supported |= IFF_LIVE_ADDR_CHANGE;
+
+	if (options & DPNI_OPT_NO_MAC_FILTER)
+		not_supported |= IFF_UNICAST_FLT;
+	else
+		supported |= IFF_UNICAST_FLT;
+
+	net_dev->priv_flags |= supported;
+	net_dev->priv_flags &= ~not_supported;
+	net_dev->lltx = true;
+
+	/* Features */
+	net_dev->features = NETIF_F_RXCSUM |
+			    NETIF_F_IP_CSUM | NETIF_F_IPV6_CSUM |
+			    NETIF_F_SG | NETIF_F_HIGHDMA |
+			    NETIF_F_HW_TC | NETIF_F_TSO;
+	net_dev->gso_max_segs = DPAA2_ETH_ENQUEUE_MAX_FDS;
+	net_dev->hw_features = net_dev->features;
+	net_dev->xdp_features = NETDEV_XDP_ACT_BASIC |
+				NETDEV_XDP_ACT_REDIRECT |
+				NETDEV_XDP_ACT_NDO_XMIT;
+	if (priv->dpni_attrs.wriop_version >= DPAA2_WRIOP_VERSION(3, 0, 0) &&
+	    priv->dpni_attrs.num_queues <= 8)
+		net_dev->xdp_features |= NETDEV_XDP_ACT_XSK_ZEROCOPY;
+
+	if (priv->dpni_attrs.vlan_filter_entries)
+		net_dev->hw_features |= NETIF_F_HW_VLAN_CTAG_FILTER;
+
+	return 0;
+}
+
+static int dpaa2_eth_poll_link_state(void *arg)
+{
+	struct dpaa2_eth_priv *priv = (struct dpaa2_eth_priv *)arg;
+	int err;
+
+	while (!kthread_should_stop()) {
+		err = dpaa2_eth_link_state_update(priv);
+		if (unlikely(err))
+			return err;
+
+		msleep(DPAA2_ETH_LINK_STATE_REFRESH);
+	}
+
+	return 0;
+}
+
+static int dpaa2_eth_connect_mac(struct dpaa2_eth_priv *priv)
+{
+	struct fsl_mc_device *dpni_dev, *dpmac_dev;
+	struct dpaa2_mac *mac;
+	int err;
+
+	dpni_dev = to_fsl_mc_device(priv->net_dev->dev.parent);
+	dpmac_dev = fsl_mc_get_endpoint(dpni_dev, 0);
+
+	if (PTR_ERR(dpmac_dev) == -EPROBE_DEFER) {
+		netdev_dbg(priv->net_dev, "waiting for mac\n");
+		return PTR_ERR(dpmac_dev);
+	}
+
+	if (IS_ERR(dpmac_dev))
+		return 0;
+
+	if (dpmac_dev->dev.type != &fsl_mc_bus_dpmac_type) {
+		err = 0;
+		goto out_put_device;
+	}
+
+	mac = kzalloc(sizeof(struct dpaa2_mac), GFP_KERNEL);
+	if (!mac) {
+		err = -ENOMEM;
+		goto out_put_device;
+	}
+
+	mac->mc_dev = dpmac_dev;
+	mac->mc_io = priv->mc_io;
+	mac->net_dev = priv->net_dev;
+
+	err = dpaa2_mac_open(mac);
+	if (err)
+		goto err_free_mac;
+
+	if (dpaa2_mac_is_type_phy(mac)) {
+		err = dpaa2_mac_connect(mac);
+		if (err) {
+			if (err == -EPROBE_DEFER)
+				netdev_dbg(priv->net_dev,
+					   "could not connect to MAC\n");
+			else
+				netdev_err(priv->net_dev,
+					   "Error connecting to the MAC endpoint: %pe",
+					   ERR_PTR(err));
+			goto err_close_mac;
+		}
+	}
+
+	mutex_lock(&priv->mac_lock);
+	priv->mac = mac;
+	mutex_unlock(&priv->mac_lock);
+
+	return 0;
+
+err_close_mac:
+	dpaa2_mac_close(mac);
+err_free_mac:
+	kfree(mac);
+out_put_device:
+	put_device(&dpmac_dev->dev);
+	return err;
+}
+
+static void dpaa2_eth_disconnect_mac(struct dpaa2_eth_priv *priv)
+{
+	struct dpaa2_mac *mac;
+
+	mutex_lock(&priv->mac_lock);
+	mac = priv->mac;
+	priv->mac = NULL;
+	mutex_unlock(&priv->mac_lock);
+
+	if (!mac)
+		return;
+
+	if (dpaa2_mac_is_type_phy(mac))
+		dpaa2_mac_disconnect(mac);
+
+	dpaa2_mac_close(mac);
+	kfree(mac);
+}
+
+static irqreturn_t dpni_irq0_handler_thread(int irq_num, void *arg)
+{
+	u32 status = ~0;
+	struct device *dev = (struct device *)arg;
+	struct fsl_mc_device *dpni_dev = to_fsl_mc_device(dev);
+	struct net_device *net_dev = dev_get_drvdata(dev);
+	struct dpaa2_eth_priv *priv = netdev_priv(net_dev);
+	bool had_mac;
+	int err;
+
+	err = dpni_get_irq_status(dpni_dev->mc_io, 0, dpni_dev->mc_handle,
+				  DPNI_IRQ_INDEX, &status);
+	if (unlikely(err)) {
+		netdev_err(net_dev, "Can't get irq status (err %d)\n", err);
+		return IRQ_HANDLED;
+	}
+
+	if (status & DPNI_IRQ_EVENT_LINK_CHANGED)
+		dpaa2_eth_link_state_update(netdev_priv(net_dev));
+
+	if (status & DPNI_IRQ_EVENT_ENDPOINT_CHANGED) {
+		dpaa2_eth_set_mac_addr(netdev_priv(net_dev));
+		dpaa2_eth_update_tx_fqids(priv);
+
+		/* We can avoid locking because the "endpoint changed" IRQ
+		 * handler is the only one who changes priv->mac at runtime,
+		 * so we are not racing with anyone.
+		 */
+		had_mac = !!priv->mac;
+		if (had_mac)
+			dpaa2_eth_disconnect_mac(priv);
+		else
+			dpaa2_eth_connect_mac(priv);
+	}
+
+	return IRQ_HANDLED;
+}
+
+static int dpaa2_eth_setup_irqs(struct fsl_mc_device *ls_dev)
+{
+	int err = 0;
+	struct fsl_mc_device_irq *irq;
+
+	err = fsl_mc_allocate_irqs(ls_dev);
+	if (err) {
+		dev_err(&ls_dev->dev, "MC irqs allocation failed\n");
+		return err;
+	}
+
+	irq = ls_dev->irqs[0];
+	err = devm_request_threaded_irq(&ls_dev->dev, irq->virq,
+					NULL, dpni_irq0_handler_thread,
+					IRQF_NO_SUSPEND | IRQF_ONESHOT,
+					dev_name(&ls_dev->dev), &ls_dev->dev);
+	if (err < 0) {
+		dev_err(&ls_dev->dev, "devm_request_threaded_irq(): %d\n", err);
+		goto free_mc_irq;
+	}
+
+	err = dpni_set_irq_mask(ls_dev->mc_io, 0, ls_dev->mc_handle,
+				DPNI_IRQ_INDEX, DPNI_IRQ_EVENT_LINK_CHANGED |
+				DPNI_IRQ_EVENT_ENDPOINT_CHANGED);
+	if (err < 0) {
+		dev_err(&ls_dev->dev, "dpni_set_irq_mask(): %d\n", err);
+		goto free_irq;
+	}
+
+	err = dpni_set_irq_enable(ls_dev->mc_io, 0, ls_dev->mc_handle,
+				  DPNI_IRQ_INDEX, 1);
+	if (err < 0) {
+		dev_err(&ls_dev->dev, "dpni_set_irq_enable(): %d\n", err);
+		goto free_irq;
+	}
+
+	return 0;
+
+free_irq:
+	devm_free_irq(&ls_dev->dev, irq->virq, &ls_dev->dev);
+free_mc_irq:
+	fsl_mc_free_irqs(ls_dev);
+
+	return err;
+}
+
+static void dpaa2_eth_add_ch_napi(struct dpaa2_eth_priv *priv)
+{
+	int i;
+	struct dpaa2_eth_channel *ch;
+
+	for (i = 0; i < priv->num_channels; i++) {
+		ch = priv->channel[i];
+		/* NAPI weight *MUST* be a multiple of DPAA2_ETH_STORE_SIZE */
+		netif_napi_add(priv->net_dev, &ch->napi, dpaa2_eth_poll);
+	}
+}
+
+static void dpaa2_eth_del_ch_napi(struct dpaa2_eth_priv *priv)
+{
+	int i;
+	struct dpaa2_eth_channel *ch;
+
+	for (i = 0; i < priv->num_channels; i++) {
+		ch = priv->channel[i];
+		netif_napi_del(&ch->napi);
+	}
+}
+
+static void dpaa2_eth_free_rx_xdp_rxq(struct dpaa2_eth_priv *priv)
+{
+	int i;
+
+	for (i = 0; i < priv->num_fqs; i++) {
+		if (priv->fq[i].type == DPAA2_RX_FQ &&
+		    xdp_rxq_info_is_reg(&priv->fq[i].channel->xdp_rxq))
+			xdp_rxq_info_unreg(&priv->fq[i].channel->xdp_rxq);
+	}
+}
+
+static int dpaa2_eth_probe(struct fsl_mc_device *dpni_dev)
+{
+	struct device *dev;
+	struct net_device *net_dev = NULL;
+	struct dpaa2_eth_priv *priv = NULL;
+	int err = 0;
+
+	dev = &dpni_dev->dev;
+
+	/* Net device */
+	net_dev = alloc_etherdev_mq(sizeof(*priv), DPAA2_ETH_MAX_NETDEV_QUEUES);
+	if (!net_dev) {
+		dev_err(dev, "alloc_etherdev_mq() failed\n");
+		return -ENOMEM;
+	}
+
+	SET_NETDEV_DEV(net_dev, dev);
+	dev_set_drvdata(dev, net_dev);
+
+	priv = netdev_priv(net_dev);
+	priv->net_dev = net_dev;
+	SET_NETDEV_DEVLINK_PORT(net_dev, &priv->devlink_port);
+
+	mutex_init(&priv->mac_lock);
+
+	priv->iommu_domain = iommu_get_domain_for_dev(dev);
+
+	priv->tx_tstamp_type = HWTSTAMP_TX_OFF;
+	priv->rx_tstamp = false;
+
+	priv->dpaa2_ptp_wq = alloc_workqueue("dpaa2_ptp_wq", WQ_PERCPU, 0);
+	if (!priv->dpaa2_ptp_wq) {
+		err = -ENOMEM;
+		goto err_wq_alloc;
+	}
+
+	INIT_WORK(&priv->tx_onestep_tstamp, dpaa2_eth_tx_onestep_tstamp);
+	mutex_init(&priv->onestep_tstamp_lock);
+	skb_queue_head_init(&priv->tx_skbs);
+
+	priv->rx_copybreak = DPAA2_ETH_DEFAULT_COPYBREAK;
+
+	/* Obtain a MC portal */
+	err = fsl_mc_portal_allocate(dpni_dev, FSL_MC_IO_ATOMIC_CONTEXT_PORTAL,
+				     &priv->mc_io);
+	if (err) {
+		if (err == -ENXIO) {
+			dev_dbg(dev, "waiting for MC portal\n");
+			err = -EPROBE_DEFER;
+		} else {
+			dev_err(dev, "MC portal allocation failed\n");
+		}
+		goto err_portal_alloc;
+	}
+
+	/* MC objects initialization and configuration */
+	err = dpaa2_eth_setup_dpni(dpni_dev);
+	if (err)
+		goto err_dpni_setup;
+
+	err = dpaa2_eth_setup_dpio(priv);
+	if (err)
+		goto err_dpio_setup;
+
+	dpaa2_eth_setup_fqs(priv);
+
+	err = dpaa2_eth_setup_default_dpbp(priv);
+	if (err)
+		goto err_dpbp_setup;
+
+	err = dpaa2_eth_bind_dpni(priv);
+	if (err)
+		goto err_bind;
+
+	/* Add a NAPI context for each channel */
+	dpaa2_eth_add_ch_napi(priv);
+
+	/* Percpu statistics */
+	priv->percpu_stats = alloc_percpu(*priv->percpu_stats);
+	if (!priv->percpu_stats) {
+		dev_err(dev, "alloc_percpu(percpu_stats) failed\n");
+		err = -ENOMEM;
+		goto err_alloc_percpu_stats;
+	}
+	priv->percpu_extras = alloc_percpu(*priv->percpu_extras);
+	if (!priv->percpu_extras) {
+		dev_err(dev, "alloc_percpu(percpu_extras) failed\n");
+		err = -ENOMEM;
+		goto err_alloc_percpu_extras;
+	}
+
+	priv->sgt_cache = alloc_percpu(*priv->sgt_cache);
+	if (!priv->sgt_cache) {
+		dev_err(dev, "alloc_percpu(sgt_cache) failed\n");
+		err = -ENOMEM;
+		goto err_alloc_sgt_cache;
+	}
+
+	priv->fd = alloc_percpu(*priv->fd);
+	if (!priv->fd) {
+		dev_err(dev, "alloc_percpu(fds) failed\n");
+		err = -ENOMEM;
+		goto err_alloc_fds;
+	}
+
+	err = dpaa2_eth_netdev_init(net_dev);
+	if (err)
+		goto err_netdev_init;
+
+	/* Configure checksum offload based on current interface flags */
+	err = dpaa2_eth_set_rx_csum(priv, !!(net_dev->features & NETIF_F_RXCSUM));
+	if (err)
+		goto err_csum;
+
+	err = dpaa2_eth_set_tx_csum(priv,
+				    !!(net_dev->features & (NETIF_F_IP_CSUM | NETIF_F_IPV6_CSUM)));
+	if (err)
+		goto err_csum;
+
+	err = dpaa2_eth_alloc_rings(priv);
+	if (err)
+		goto err_alloc_rings;
+
+#ifdef CONFIG_FSL_DPAA2_ETH_DCB
+	if (dpaa2_eth_has_pause_support(priv) && priv->vlan_cls_enabled) {
+		priv->dcbx_mode = DCB_CAP_DCBX_HOST | DCB_CAP_DCBX_VER_IEEE;
+		net_dev->dcbnl_ops = &dpaa2_eth_dcbnl_ops;
+	} else {
+		dev_dbg(dev, "PFC not supported\n");
+	}
+#endif
+
+	err = dpaa2_eth_connect_mac(priv);
+	if (err)
+		goto err_connect_mac;
+
+	err = dpaa2_eth_setup_irqs(dpni_dev);
+	if (err) {
+		netdev_warn(net_dev, "Failed to set link interrupt, fall back to polling\n");
+		priv->poll_thread = kthread_run(dpaa2_eth_poll_link_state, priv,
+						"%s_poll_link", net_dev->name);
+		if (IS_ERR(priv->poll_thread)) {
+			dev_err(dev, "Error starting polling thread\n");
+			goto err_poll_thread;
+		}
+		priv->do_link_poll = true;
+	}
+
+	err = dpaa2_eth_dl_alloc(priv);
+	if (err)
+		goto err_dl_register;
+
+	err = dpaa2_eth_dl_traps_register(priv);
+	if (err)
+		goto err_dl_trap_register;
+
+	err = dpaa2_eth_dl_port_add(priv);
+	if (err)
+		goto err_dl_port_add;
+
+	net_dev->needed_headroom = DPAA2_ETH_SWA_SIZE + DPAA2_ETH_TX_BUF_ALIGN;
+
+	err = register_netdev(net_dev);
+	if (err < 0) {
+		dev_err(dev, "register_netdev() failed\n");
+		goto err_netdev_reg;
+	}
+
+#ifdef CONFIG_DEBUG_FS
+	dpaa2_dbg_add(priv);
+#endif
+
+	dpaa2_eth_dl_register(priv);
+	dev_info(dev, "Probed interface %s\n", net_dev->name);
+	return 0;
+
+err_netdev_reg:
+	dpaa2_eth_dl_port_del(priv);
+err_dl_port_add:
+	dpaa2_eth_dl_traps_unregister(priv);
+err_dl_trap_register:
+	dpaa2_eth_dl_free(priv);
+err_dl_register:
+	if (priv->do_link_poll)
+		kthread_stop(priv->poll_thread);
+	else
+		fsl_mc_free_irqs(dpni_dev);
+err_poll_thread:
+	dpaa2_eth_disconnect_mac(priv);
+err_connect_mac:
+	dpaa2_eth_free_rings(priv);
+err_alloc_rings:
+err_csum:
+err_netdev_init:
+	free_percpu(priv->fd);
+err_alloc_fds:
+	free_percpu(priv->sgt_cache);
+err_alloc_sgt_cache:
+	free_percpu(priv->percpu_extras);
+err_alloc_percpu_extras:
+	free_percpu(priv->percpu_stats);
+err_alloc_percpu_stats:
+	dpaa2_eth_del_ch_napi(priv);
+	dpaa2_eth_free_rx_xdp_rxq(priv);
+err_bind:
+	dpaa2_eth_free_dpbps(priv);
+err_dpbp_setup:
+	dpaa2_eth_free_dpio(priv);
+err_dpio_setup:
+	dpaa2_eth_free_dpni(priv);
+err_dpni_setup:
+	fsl_mc_portal_free(priv->mc_io);
+err_portal_alloc:
+	destroy_workqueue(priv->dpaa2_ptp_wq);
+err_wq_alloc:
+	dev_set_drvdata(dev, NULL);
+	free_netdev(net_dev);
+
+	return err;
+}
+
+static void dpaa2_eth_remove(struct fsl_mc_device *ls_dev)
+{
+	struct device *dev;
+	struct net_device *net_dev;
+	struct dpaa2_eth_priv *priv;
+
+	dev = &ls_dev->dev;
+	net_dev = dev_get_drvdata(dev);
+	priv = netdev_priv(net_dev);
+
+	dpaa2_eth_dl_unregister(priv);
+
+#ifdef CONFIG_DEBUG_FS
+	dpaa2_dbg_remove(priv);
+#endif
+
+	unregister_netdev(net_dev);
+
+	dpaa2_eth_dl_port_del(priv);
+	dpaa2_eth_dl_traps_unregister(priv);
+	dpaa2_eth_dl_free(priv);
+
+	if (priv->do_link_poll)
+		kthread_stop(priv->poll_thread);
+	else
+		fsl_mc_free_irqs(ls_dev);
+
+	dpaa2_eth_disconnect_mac(priv);
+	dpaa2_eth_free_rings(priv);
+	free_percpu(priv->fd);
+	free_percpu(priv->sgt_cache);
+	free_percpu(priv->percpu_stats);
+	free_percpu(priv->percpu_extras);
+
+	dpaa2_eth_del_ch_napi(priv);
+	dpaa2_eth_free_rx_xdp_rxq(priv);
+	dpaa2_eth_free_dpbps(priv);
+	dpaa2_eth_free_dpio(priv);
+	dpaa2_eth_free_dpni(priv);
+	if (priv->onestep_reg_base)
+		iounmap(priv->onestep_reg_base);
+
+	fsl_mc_portal_free(priv->mc_io);
+
+	destroy_workqueue(priv->dpaa2_ptp_wq);
+
+	dev_dbg(net_dev->dev.parent, "Removed interface %s\n", net_dev->name);
+
+	free_netdev(net_dev);
+}
+
+static const struct fsl_mc_device_id dpaa2_eth_match_id_table[] = {
+	{
+		.vendor = FSL_MC_VENDOR_FREESCALE,
+		.obj_type = "dpni",
+	},
+	{ .vendor = 0x0 }
+};
+MODULE_DEVICE_TABLE(fslmc, dpaa2_eth_match_id_table);
+
+static struct fsl_mc_driver dpaa2_eth_driver = {
+	.driver = {
+		.name = KBUILD_MODNAME,
+	},
+	.probe = dpaa2_eth_probe,
+	.remove = dpaa2_eth_remove,
+	.match_id_table = dpaa2_eth_match_id_table
+};
+
+static int __init dpaa2_eth_driver_init(void)
+{
+	int err;
+
+	dpaa2_eth_dbg_init();
+	err = fsl_mc_driver_register(&dpaa2_eth_driver);
+	if (err) {
+		dpaa2_eth_dbg_exit();
+		return err;
+	}
+
+	return 0;
+}
+
+static void __exit dpaa2_eth_driver_exit(void)
+{
+	dpaa2_eth_dbg_exit();
+	fsl_mc_driver_unregister(&dpaa2_eth_driver);
+}
+
+module_init(dpaa2_eth_driver_init);
+module_exit(dpaa2_eth_driver_exit);
diff --git a/drivers/net/ethernet/freescale/dpaa2/dpaa2-eth.h b/drivers/net/ethernet/freescale/dpaa2/dpaa2-eth.h
new file mode 100644
index 000000000000..834cba8c3a41
--- /dev/null
+++ b/drivers/net/ethernet/freescale/dpaa2/dpaa2-eth.h
@@ -0,0 +1,861 @@
+/* SPDX-License-Identifier: (GPL-2.0+ OR BSD-3-Clause) */
+/* Copyright 2014-2016 Freescale Semiconductor Inc.
+ * Copyright 2016-2022 NXP
+ */
+
+#ifndef __DPAA2_ETH_H
+#define __DPAA2_ETH_H
+
+#include <linux/dcbnl.h>
+#include <linux/netdevice.h>
+#include <linux/if_vlan.h>
+#include <linux/fsl/mc.h>
+#include <linux/net_tstamp.h>
+#include <net/devlink.h>
+#include <net/xdp.h>
+
+#include <soc/fsl/dpaa2-io.h>
+#include <soc/fsl/dpaa2-fd.h>
+#include "dpni.h"
+#include "dpni-cmd.h"
+
+#include "dpaa2-eth-trace.h"
+#include "dpaa2-eth-debugfs.h"
+#include "dpaa2-mac.h"
+
+#define DPAA2_WRIOP_VERSION(x, y, z) ((x) << 10 | (y) << 5 | (z) << 0)
+
+#define DPAA2_ETH_STORE_SIZE		16
+
+/* Maximum number of scatter-gather entries in an ingress frame,
+ * considering the maximum receive frame size is 64K
+ */
+#define DPAA2_ETH_MAX_SG_ENTRIES	((64 * 1024) / DPAA2_ETH_RX_BUF_SIZE)
+
+/* Maximum acceptable MTU value. It is in direct relation with the hardware
+ * enforced Max Frame Length (currently 10k).
+ */
+#define DPAA2_ETH_MFL			(10 * 1024)
+#define DPAA2_ETH_MAX_MTU		(DPAA2_ETH_MFL - VLAN_ETH_HLEN)
+/* Convert L3 MTU to L2 MFL */
+#define DPAA2_ETH_L2_MAX_FRM(mtu)	((mtu) + VLAN_ETH_HLEN)
+
+/* Set the taildrop threshold (in bytes) to allow the enqueue of a large
+ * enough number of jumbo frames in the Rx queues (length of the current
+ * frame is not taken into account when making the taildrop decision)
+ */
+#define DPAA2_ETH_FQ_TAILDROP_THRESH	(1024 * 1024)
+
+/* Maximum burst size value for Tx shaping */
+#define DPAA2_ETH_MAX_BURST_SIZE	0xF7FF
+
+/* Maximum number of Tx confirmation frames to be processed
+ * in a single NAPI call
+ */
+#define DPAA2_ETH_TXCONF_PER_NAPI	256
+
+/* Maximum number of Tx frames to be processed in a single NAPI
+ * call when AF_XDP is running. Bind it to DPAA2_ETH_TXCONF_PER_NAPI
+ * to maximize the throughput.
+ */
+#define DPAA2_ETH_TX_ZC_PER_NAPI	DPAA2_ETH_TXCONF_PER_NAPI
+
+/* Buffer qouta per channel. We want to keep in check number of ingress frames
+ * in flight: for small sized frames, congestion group taildrop may kick in
+ * first; for large sizes, Rx FQ taildrop threshold will ensure only a
+ * reasonable number of frames will be pending at any given time.
+ * Ingress frame drop due to buffer pool depletion should be a corner case only
+ */
+#define DPAA2_ETH_NUM_BUFS		1280
+#define DPAA2_ETH_REFILL_THRESH \
+	(DPAA2_ETH_NUM_BUFS - DPAA2_ETH_BUFS_PER_CMD)
+
+/* Congestion group taildrop threshold: number of frames allowed to accumulate
+ * at any moment in a group of Rx queues belonging to the same traffic class.
+ * Choose value such that we don't risk depleting the buffer pool before the
+ * taildrop kicks in
+ */
+#define DPAA2_ETH_CG_TAILDROP_THRESH(priv)				\
+	(1024 * dpaa2_eth_queue_count(priv) / dpaa2_eth_tc_count(priv))
+
+/* Congestion group notification threshold: when this many frames accumulate
+ * on the Rx queues belonging to the same TC, the MAC is instructed to send
+ * PFC frames for that TC.
+ * When number of pending frames drops below exit threshold transmission of
+ * PFC frames is stopped.
+ */
+#define DPAA2_ETH_CN_THRESH_ENTRY(priv) \
+	(DPAA2_ETH_CG_TAILDROP_THRESH(priv) / 2)
+#define DPAA2_ETH_CN_THRESH_EXIT(priv) \
+	(DPAA2_ETH_CN_THRESH_ENTRY(priv) * 3 / 4)
+
+/* Maximum number of buffers that can be acquired/released through a single
+ * QBMan command
+ */
+#define DPAA2_ETH_BUFS_PER_CMD		7
+
+/* Hardware requires alignment for ingress/egress buffer addresses */
+#define DPAA2_ETH_TX_BUF_ALIGN		64
+
+#define DPAA2_ETH_RX_BUF_RAW_SIZE	PAGE_SIZE
+#define DPAA2_ETH_RX_BUF_TAILROOM \
+	SKB_DATA_ALIGN(sizeof(struct skb_shared_info))
+#define DPAA2_ETH_RX_BUF_SIZE \
+	(DPAA2_ETH_RX_BUF_RAW_SIZE - DPAA2_ETH_RX_BUF_TAILROOM)
+
+/* Hardware annotation area in RX/TX buffers */
+#define DPAA2_ETH_RX_HWA_SIZE		64
+#define DPAA2_ETH_TX_HWA_SIZE		128
+
+/* PTP nominal frequency 1GHz */
+#define DPAA2_PTP_CLK_PERIOD_NS		1
+
+/* Due to a limitation in WRIOP 1.0.0, the RX buffer data must be aligned
+ * to 256B. For newer revisions, the requirement is only for 64B alignment
+ */
+#define DPAA2_ETH_RX_BUF_ALIGN_REV1	256
+#define DPAA2_ETH_RX_BUF_ALIGN		64
+
+/* The firmware allows assigning multiple buffer pools to a single DPNI -
+ * maximum 8 DPBP objects. By default, only the first DPBP (idx 0) is used for
+ * all queues. Thus, when enabling AF_XDP we must accommodate up to 9 DPBPs
+ * object: the default and 8 other distinct buffer pools, one for each queue.
+ */
+#define DPAA2_ETH_DEFAULT_BP_IDX	0
+#define DPAA2_ETH_MAX_BPS		9
+
+/* We are accommodating a skb backpointer and some S/G info
+ * in the frame's software annotation. The hardware
+ * options are either 0 or 64, so we choose the latter.
+ */
+#define DPAA2_ETH_SWA_SIZE		64
+
+/* We store different information in the software annotation area of a Tx frame
+ * based on what type of frame it is
+ */
+enum dpaa2_eth_swa_type {
+	DPAA2_ETH_SWA_SINGLE,
+	DPAA2_ETH_SWA_SG,
+	DPAA2_ETH_SWA_XDP,
+	DPAA2_ETH_SWA_XSK,
+	DPAA2_ETH_SWA_SW_TSO,
+};
+
+/* Must keep this struct smaller than DPAA2_ETH_SWA_SIZE */
+struct dpaa2_eth_swa {
+	enum dpaa2_eth_swa_type type;
+	union {
+		struct {
+			struct sk_buff *skb;
+			int sgt_size;
+		} single;
+		struct {
+			struct sk_buff *skb;
+			struct scatterlist *scl;
+			int num_sg;
+			int sgt_size;
+		} sg;
+		struct {
+			int dma_size;
+			struct xdp_frame *xdpf;
+		} xdp;
+		struct {
+			struct xdp_buff *xdp_buff;
+			int sgt_size;
+		} xsk;
+		struct {
+			struct sk_buff *skb;
+			int num_sg;
+			int sgt_size;
+			int is_last_fd;
+		} tso;
+	};
+};
+
+/* Annotation valid bits in FD FRC */
+#define DPAA2_FD_FRC_FASV		0x8000
+#define DPAA2_FD_FRC_FAEADV		0x4000
+#define DPAA2_FD_FRC_FAPRV		0x2000
+#define DPAA2_FD_FRC_FAIADV		0x1000
+#define DPAA2_FD_FRC_FASWOV		0x0800
+#define DPAA2_FD_FRC_FAICFDV		0x0400
+
+/* Error bits in FD CTRL */
+#define DPAA2_FD_RX_ERR_MASK		(FD_CTRL_SBE | FD_CTRL_FAERR)
+#define DPAA2_FD_TX_ERR_MASK		(FD_CTRL_UFD	| \
+					 FD_CTRL_SBE	| \
+					 FD_CTRL_FSE	| \
+					 FD_CTRL_FAERR)
+
+/* Annotation bits in FD CTRL */
+#define DPAA2_FD_CTRL_ASAL		0x00020000	/* ASAL = 128B */
+
+/* Frame annotation status */
+struct dpaa2_fas {
+	u8 reserved;
+	u8 ppid;
+	__le16 ifpid;
+	__le32 status;
+};
+
+/* Frame annotation status word is located in the first 8 bytes
+ * of the buffer's hardware annoatation area
+ */
+#define DPAA2_FAS_OFFSET		0
+#define DPAA2_FAS_SIZE			(sizeof(struct dpaa2_fas))
+
+/* Timestamp is located in the next 8 bytes of the buffer's
+ * hardware annotation area
+ */
+#define DPAA2_TS_OFFSET			0x8
+
+/* Frame annotation parse results */
+struct dpaa2_fapr {
+	/* 64-bit word 1 */
+	__le32 faf_lo;
+	__le16 faf_ext;
+	__le16 nxt_hdr;
+	/* 64-bit word 2 */
+	__le64 faf_hi;
+	/* 64-bit word 3 */
+	u8 last_ethertype_offset;
+	u8 vlan_tci_offset_n;
+	u8 vlan_tci_offset_1;
+	u8 llc_snap_offset;
+	u8 eth_offset;
+	u8 ip1_pid_offset;
+	u8 shim_offset_2;
+	u8 shim_offset_1;
+	/* 64-bit word 4 */
+	u8 l5_offset;
+	u8 l4_offset;
+	u8 gre_offset;
+	u8 l3_offset_n;
+	u8 l3_offset_1;
+	u8 mpls_offset_n;
+	u8 mpls_offset_1;
+	u8 pppoe_offset;
+	/* 64-bit word 5 */
+	__le16 running_sum;
+	__le16 gross_running_sum;
+	u8 ipv6_frag_offset;
+	u8 nxt_hdr_offset;
+	u8 routing_hdr_offset_2;
+	u8 routing_hdr_offset_1;
+	/* 64-bit word 6 */
+	u8 reserved[5]; /* Soft-parsing context */
+	u8 ip_proto_offset_n;
+	u8 nxt_hdr_frag_offset;
+	u8 parse_error_code;
+};
+
+#define DPAA2_FAPR_OFFSET		0x10
+#define DPAA2_FAPR_SIZE			sizeof((struct dpaa2_fapr))
+
+/* Frame annotation egress action descriptor */
+#define DPAA2_FAEAD_OFFSET		0x58
+
+struct dpaa2_faead {
+	__le32 conf_fqid;
+	__le32 ctrl;
+};
+
+#define DPAA2_FAEAD_A2V			0x20000000
+#define DPAA2_FAEAD_A4V			0x08000000
+#define DPAA2_FAEAD_UPDV		0x00001000
+#define DPAA2_FAEAD_EBDDV		0x00002000
+#define DPAA2_FAEAD_UPD			0x00000010
+
+struct ptp_tstamp {
+	u16 sec_msb;
+	u32 sec_lsb;
+	u32 nsec;
+};
+
+static inline void ns_to_ptp_tstamp(struct ptp_tstamp *tstamp, u64 ns)
+{
+	u64 sec, nsec;
+
+	sec = ns;
+	nsec = do_div(sec, 1000000000);
+
+	tstamp->sec_lsb = sec & 0xFFFFFFFF;
+	tstamp->sec_msb = (sec >> 32) & 0xFFFF;
+	tstamp->nsec = nsec;
+}
+
+/* Accessors for the hardware annotation fields that we use */
+static inline void *dpaa2_get_hwa(void *buf_addr, bool swa)
+{
+	return buf_addr + (swa ? DPAA2_ETH_SWA_SIZE : 0);
+}
+
+static inline struct dpaa2_fas *dpaa2_get_fas(void *buf_addr, bool swa)
+{
+	return dpaa2_get_hwa(buf_addr, swa) + DPAA2_FAS_OFFSET;
+}
+
+static inline __le64 *dpaa2_get_ts(void *buf_addr, bool swa)
+{
+	return dpaa2_get_hwa(buf_addr, swa) + DPAA2_TS_OFFSET;
+}
+
+static inline struct dpaa2_fapr *dpaa2_get_fapr(void *buf_addr, bool swa)
+{
+	return dpaa2_get_hwa(buf_addr, swa) + DPAA2_FAPR_OFFSET;
+}
+
+static inline struct dpaa2_faead *dpaa2_get_faead(void *buf_addr, bool swa)
+{
+	return dpaa2_get_hwa(buf_addr, swa) + DPAA2_FAEAD_OFFSET;
+}
+
+/* Error and status bits in the frame annotation status word */
+/* Debug frame, otherwise supposed to be discarded */
+#define DPAA2_FAS_DISC			0x80000000
+/* MACSEC frame */
+#define DPAA2_FAS_MS			0x40000000
+#define DPAA2_FAS_PTP			0x08000000
+/* Ethernet multicast frame */
+#define DPAA2_FAS_MC			0x04000000
+/* Ethernet broadcast frame */
+#define DPAA2_FAS_BC			0x02000000
+#define DPAA2_FAS_KSE			0x00040000
+#define DPAA2_FAS_EOFHE			0x00020000
+#define DPAA2_FAS_MNLE			0x00010000
+#define DPAA2_FAS_TIDE			0x00008000
+#define DPAA2_FAS_PIEE			0x00004000
+/* Frame length error */
+#define DPAA2_FAS_FLE			0x00002000
+/* Frame physical error */
+#define DPAA2_FAS_FPE			0x00001000
+#define DPAA2_FAS_PTE			0x00000080
+#define DPAA2_FAS_ISP			0x00000040
+#define DPAA2_FAS_PHE			0x00000020
+#define DPAA2_FAS_BLE			0x00000010
+/* L3 csum validation performed */
+#define DPAA2_FAS_L3CV			0x00000008
+/* L3 csum error */
+#define DPAA2_FAS_L3CE			0x00000004
+/* L4 csum validation performed */
+#define DPAA2_FAS_L4CV			0x00000002
+/* L4 csum error */
+#define DPAA2_FAS_L4CE			0x00000001
+/* Possible errors on the ingress path */
+#define DPAA2_FAS_RX_ERR_MASK		(DPAA2_FAS_KSE		| \
+					 DPAA2_FAS_EOFHE	| \
+					 DPAA2_FAS_MNLE		| \
+					 DPAA2_FAS_TIDE		| \
+					 DPAA2_FAS_PIEE		| \
+					 DPAA2_FAS_FLE		| \
+					 DPAA2_FAS_FPE		| \
+					 DPAA2_FAS_PTE		| \
+					 DPAA2_FAS_ISP		| \
+					 DPAA2_FAS_PHE		| \
+					 DPAA2_FAS_BLE		| \
+					 DPAA2_FAS_L3CE		| \
+					 DPAA2_FAS_L4CE)
+
+/* Time in milliseconds between link state updates */
+#define DPAA2_ETH_LINK_STATE_REFRESH	1000
+
+/* Number of times to retry a frame enqueue before giving up.
+ * Value determined empirically, in order to minimize the number
+ * of frames dropped on Tx
+ */
+#define DPAA2_ETH_ENQUEUE_RETRIES	10
+
+/* Number of times to retry DPIO portal operations while waiting
+ * for portal to finish executing current command and become
+ * available. We want to avoid being stuck in a while loop in case
+ * hardware becomes unresponsive, but not give up too easily if
+ * the portal really is busy for valid reasons
+ */
+#define DPAA2_ETH_SWP_BUSY_RETRIES	1000
+
+/* Driver statistics, other than those in struct rtnl_link_stats64.
+ * These are usually collected per-CPU and aggregated by ethtool.
+ */
+struct dpaa2_eth_drv_stats {
+	__u64	tx_conf_frames;
+	__u64	tx_conf_bytes;
+	__u64	tx_sg_frames;
+	__u64	tx_sg_bytes;
+	__u64	tx_tso_frames;
+	__u64	tx_tso_bytes;
+	__u64	rx_sg_frames;
+	__u64	rx_sg_bytes;
+	/* Linear skbs sent as a S/G FD due to insufficient headroom */
+	__u64	tx_converted_sg_frames;
+	__u64	tx_converted_sg_bytes;
+	/* Enqueues retried due to portal busy */
+	__u64	tx_portal_busy;
+};
+
+/* Per-FQ statistics */
+struct dpaa2_eth_fq_stats {
+	/* Number of frames received on this queue */
+	__u64 frames;
+};
+
+/* Per-channel statistics */
+struct dpaa2_eth_ch_stats {
+	/* Volatile dequeues retried due to portal busy */
+	__u64 dequeue_portal_busy;
+	/* Pull errors */
+	__u64 pull_err;
+	/* Number of CDANs; useful to estimate avg NAPI len */
+	__u64 cdan;
+	/* XDP counters */
+	__u64 xdp_drop;
+	__u64 xdp_tx;
+	__u64 xdp_tx_err;
+	__u64 xdp_redirect;
+	/* Must be last, does not show up in ethtool stats */
+	__u64 frames;
+	__u64 frames_per_cdan;
+	__u64 bytes_per_cdan;
+};
+
+#define DPAA2_ETH_CH_STATS	7
+
+/* Maximum number of queues associated with a DPNI */
+#define DPAA2_ETH_MAX_TCS		8
+#define DPAA2_ETH_MAX_RX_QUEUES_PER_TC	16
+#define DPAA2_ETH_MAX_RX_QUEUES		\
+	(DPAA2_ETH_MAX_RX_QUEUES_PER_TC * DPAA2_ETH_MAX_TCS)
+#define DPAA2_ETH_MAX_TX_QUEUES		16
+#define DPAA2_ETH_MAX_RX_ERR_QUEUES	1
+#define DPAA2_ETH_MAX_QUEUES		(DPAA2_ETH_MAX_RX_QUEUES + \
+					DPAA2_ETH_MAX_TX_QUEUES + \
+					DPAA2_ETH_MAX_RX_ERR_QUEUES)
+#define DPAA2_ETH_MAX_NETDEV_QUEUES	\
+	(DPAA2_ETH_MAX_TX_QUEUES * DPAA2_ETH_MAX_TCS)
+
+#define DPAA2_ETH_MAX_DPCONS		16
+
+enum dpaa2_eth_fq_type {
+	DPAA2_RX_FQ = 0,
+	DPAA2_TX_CONF_FQ,
+	DPAA2_RX_ERR_FQ
+};
+
+struct dpaa2_eth_priv;
+struct dpaa2_eth_channel;
+struct dpaa2_eth_fq;
+
+struct dpaa2_eth_xdp_fds {
+	struct dpaa2_fd fds[DEV_MAP_BULK_SIZE];
+	ssize_t num;
+};
+
+typedef void dpaa2_eth_consume_cb_t(struct dpaa2_eth_priv *priv,
+				    struct dpaa2_eth_channel *ch,
+				    const struct dpaa2_fd *fd,
+				    struct dpaa2_eth_fq *fq);
+
+struct dpaa2_eth_fq {
+	u32 fqid;
+	u32 tx_qdbin;
+	u32 tx_fqid[DPAA2_ETH_MAX_TCS];
+	u16 flowid;
+	u8 tc;
+	int target_cpu;
+	u32 dq_frames;
+	u32 dq_bytes;
+	struct dpaa2_eth_channel *channel;
+	enum dpaa2_eth_fq_type type;
+
+	dpaa2_eth_consume_cb_t *consume;
+	struct dpaa2_eth_fq_stats stats;
+
+	struct dpaa2_eth_xdp_fds xdp_redirect_fds;
+	struct dpaa2_eth_xdp_fds xdp_tx_fds;
+};
+
+struct dpaa2_eth_ch_xdp {
+	struct bpf_prog *prog;
+	unsigned int res;
+};
+
+struct dpaa2_eth_bp {
+	struct fsl_mc_device *dev;
+	int bpid;
+};
+
+struct dpaa2_eth_channel {
+	struct dpaa2_io_notification_ctx nctx;
+	struct fsl_mc_device *dpcon;
+	int dpcon_id;
+	int ch_id;
+	struct napi_struct napi;
+	struct dpaa2_io *dpio;
+	struct dpaa2_io_store *store;
+	struct dpaa2_eth_priv *priv;
+	int buf_count;
+	struct dpaa2_eth_ch_stats stats;
+	struct dpaa2_eth_ch_xdp xdp;
+	struct xdp_rxq_info xdp_rxq;
+	struct list_head *rx_list;
+
+	/* Buffers to be recycled back in the buffer pool */
+	u64 recycled_bufs[DPAA2_ETH_BUFS_PER_CMD];
+	int recycled_bufs_cnt;
+
+	bool xsk_zc;
+	int xsk_tx_pkts_sent;
+	struct xsk_buff_pool *xsk_pool;
+	struct dpaa2_eth_bp *bp;
+};
+
+struct dpaa2_eth_dist_fields {
+	u64 rxnfc_field;
+	enum net_prot cls_prot;
+	int cls_field;
+	int size;
+	u64 id;
+};
+
+struct dpaa2_eth_cls_rule {
+	struct ethtool_rx_flow_spec fs;
+	u8 in_use;
+};
+
+#define DPAA2_ETH_SGT_CACHE_SIZE	256
+struct dpaa2_eth_sgt_cache {
+	void *buf[DPAA2_ETH_SGT_CACHE_SIZE];
+	u16 count;
+};
+
+struct dpaa2_eth_trap_item {
+	void *trap_ctx;
+};
+
+struct dpaa2_eth_trap_data {
+	struct dpaa2_eth_trap_item *trap_items_arr;
+	struct dpaa2_eth_priv *priv;
+};
+
+#define DPAA2_ETH_SG_ENTRIES_MAX	(PAGE_SIZE / sizeof(struct scatterlist))
+
+#define DPAA2_ETH_DEFAULT_COPYBREAK	512
+
+#define DPAA2_ETH_ENQUEUE_MAX_FDS	256
+struct dpaa2_eth_fds {
+	struct dpaa2_fd array[DPAA2_ETH_ENQUEUE_MAX_FDS];
+};
+
+/* Driver private data */
+struct dpaa2_eth_priv {
+	struct net_device *net_dev;
+
+	u8 num_fqs;
+	struct dpaa2_eth_fq fq[DPAA2_ETH_MAX_QUEUES];
+	int (*enqueue)(struct dpaa2_eth_priv *priv,
+		       struct dpaa2_eth_fq *fq,
+		       struct dpaa2_fd *fd, u8 prio,
+		       u32 num_frames,
+		       int *frames_enqueued);
+
+	u8 num_channels;
+	struct dpaa2_eth_channel *channel[DPAA2_ETH_MAX_DPCONS];
+	struct dpaa2_eth_sgt_cache __percpu *sgt_cache;
+	unsigned long features;
+	struct dpni_attr dpni_attrs;
+	u16 dpni_ver_major;
+	u16 dpni_ver_minor;
+	u16 tx_data_offset;
+	void __iomem *onestep_reg_base;
+	u8 ptp_correction_off;
+	void (*dpaa2_set_onestep_params_cb)(struct dpaa2_eth_priv *priv,
+					    u32 offset, u8 udp);
+	u16 rx_buf_size;
+	struct iommu_domain *iommu_domain;
+
+	enum hwtstamp_tx_types tx_tstamp_type;	/* Tx timestamping type */
+	bool rx_tstamp;				/* Rx timestamping enabled */
+
+	/* Buffer pool management */
+	struct dpaa2_eth_bp *bp[DPAA2_ETH_MAX_BPS];
+	int num_bps;
+
+	u16 tx_qdid;
+	struct fsl_mc_io *mc_io;
+	/* Cores which have an affine DPIO/DPCON.
+	 * This is the cpu set on which Rx and Tx conf frames are processed
+	 */
+	struct cpumask dpio_cpumask;
+
+	/* Standard statistics */
+	struct rtnl_link_stats64 __percpu *percpu_stats;
+	/* Extra stats, in addition to the ones known by the kernel */
+	struct dpaa2_eth_drv_stats __percpu *percpu_extras;
+
+	u16 mc_token;
+	u8 rx_fqtd_enabled;
+	u8 rx_cgtd_enabled;
+
+	struct dpni_link_state link_state;
+	bool do_link_poll;
+	struct task_struct *poll_thread;
+
+	/* enabled ethtool hashing bits */
+	u64 rx_hash_fields;
+	u64 rx_cls_fields;
+	struct dpaa2_eth_cls_rule *cls_rules;
+	u8 rx_cls_enabled;
+	u8 vlan_cls_enabled;
+	u8 pfc_enabled;
+#ifdef CONFIG_FSL_DPAA2_ETH_DCB
+	u8 dcbx_mode;
+	struct ieee_pfc pfc;
+#endif
+	struct bpf_prog *xdp_prog;
+#ifdef CONFIG_DEBUG_FS
+	struct dpaa2_debugfs dbg;
+#endif
+
+	struct dpaa2_mac *mac;
+	/* Serializes changes to priv->mac */
+	struct mutex		mac_lock;
+	struct workqueue_struct	*dpaa2_ptp_wq;
+	struct work_struct	tx_onestep_tstamp;
+	struct sk_buff_head	tx_skbs;
+	/* The one-step timestamping configuration on hardware
+	 * registers could only be done when no one-step
+	 * timestamping frames are in flight. So we use a mutex
+	 * lock here to make sure the lock is released by last
+	 * one-step timestamping packet through TX confirmation
+	 * queue before transmit current packet.
+	 */
+	struct mutex		onestep_tstamp_lock;
+	struct devlink *devlink;
+	struct dpaa2_eth_trap_data *trap_data;
+	struct devlink_port devlink_port;
+
+	u32 rx_copybreak;
+
+	struct dpaa2_eth_fds __percpu *fd;
+};
+
+struct dpaa2_eth_devlink_priv {
+	struct dpaa2_eth_priv *dpaa2_priv;
+};
+
+#define TX_TSTAMP		0x1
+#define TX_TSTAMP_ONESTEP_SYNC	0x2
+
+#define DPAA2_RXH_SUPPORTED	(RXH_L2DA | RXH_VLAN | RXH_L3_PROTO \
+				| RXH_IP_SRC | RXH_IP_DST | RXH_L4_B_0_1 \
+				| RXH_L4_B_2_3)
+
+/* default Rx hash options, set during probing */
+#define DPAA2_RXH_DEFAULT	(RXH_L3_PROTO | RXH_IP_SRC | RXH_IP_DST | \
+				 RXH_L4_B_0_1 | RXH_L4_B_2_3)
+
+#define dpaa2_eth_hash_enabled(priv)	\
+	((priv)->dpni_attrs.num_queues > 1)
+
+/* Required by struct dpni_rx_tc_dist_cfg::key_cfg_iova */
+#define DPAA2_CLASSIFIER_DMA_SIZE 256
+
+extern const struct ethtool_ops dpaa2_ethtool_ops;
+extern int dpaa2_phc_index;
+extern struct ptp_qoriq *dpaa2_ptp;
+
+static inline int dpaa2_eth_cmp_dpni_ver(struct dpaa2_eth_priv *priv,
+					 u16 ver_major, u16 ver_minor)
+{
+	if (priv->dpni_ver_major == ver_major)
+		return priv->dpni_ver_minor - ver_minor;
+	return priv->dpni_ver_major - ver_major;
+}
+
+/* Minimum firmware version that supports a more flexible API
+ * for configuring the Rx flow hash key
+ */
+#define DPNI_RX_DIST_KEY_VER_MAJOR	7
+#define DPNI_RX_DIST_KEY_VER_MINOR	5
+
+#define dpaa2_eth_has_legacy_dist(priv)					\
+	(dpaa2_eth_cmp_dpni_ver((priv), DPNI_RX_DIST_KEY_VER_MAJOR,	\
+				DPNI_RX_DIST_KEY_VER_MINOR) < 0)
+
+#define dpaa2_eth_fs_enabled(priv)	\
+	(!((priv)->dpni_attrs.options & DPNI_OPT_NO_FS))
+
+#define dpaa2_eth_fs_mask_enabled(priv)	\
+	((priv)->dpni_attrs.options & DPNI_OPT_HAS_KEY_MASKING)
+
+#define dpaa2_eth_fs_count(priv)        \
+	((priv)->dpni_attrs.fs_entries)
+
+#define dpaa2_eth_tc_count(priv)	\
+	((priv)->dpni_attrs.num_tcs)
+
+/* We have exactly one {Rx, Tx conf} queue per channel */
+#define dpaa2_eth_queue_count(priv)     \
+	((priv)->num_channels)
+
+enum dpaa2_eth_rx_dist {
+	DPAA2_ETH_RX_DIST_HASH,
+	DPAA2_ETH_RX_DIST_CLS
+};
+
+/* Unique IDs for the supported Rx classification header fields */
+#define DPAA2_ETH_DIST_ETHDST		BIT(0)
+#define DPAA2_ETH_DIST_ETHSRC		BIT(1)
+#define DPAA2_ETH_DIST_ETHTYPE		BIT(2)
+#define DPAA2_ETH_DIST_VLAN		BIT(3)
+#define DPAA2_ETH_DIST_IPSRC		BIT(4)
+#define DPAA2_ETH_DIST_IPDST		BIT(5)
+#define DPAA2_ETH_DIST_IPPROTO		BIT(6)
+#define DPAA2_ETH_DIST_L4SRC		BIT(7)
+#define DPAA2_ETH_DIST_L4DST		BIT(8)
+#define DPAA2_ETH_DIST_ALL		(~0ULL)
+
+#define DPNI_PTP_ONESTEP_VER_MAJOR 8
+#define DPNI_PTP_ONESTEP_VER_MINOR 2
+#define DPAA2_ETH_FEATURE_ONESTEP_CFG_DIRECT BIT(0)
+#define DPAA2_PTP_SINGLE_STEP_ENABLE	BIT(31)
+#define DPAA2_PTP_SINGLE_STEP_CH	BIT(7)
+#define DPAA2_PTP_SINGLE_CORRECTION_OFF(v) ((v) << 8)
+
+#define DPNI_PAUSE_VER_MAJOR		7
+#define DPNI_PAUSE_VER_MINOR		13
+#define dpaa2_eth_has_pause_support(priv)			\
+	(dpaa2_eth_cmp_dpni_ver((priv), DPNI_PAUSE_VER_MAJOR,	\
+				DPNI_PAUSE_VER_MINOR) >= 0)
+
+static inline bool dpaa2_eth_tx_pause_enabled(u64 link_options)
+{
+	return !!(link_options & DPNI_LINK_OPT_PAUSE) ^
+	       !!(link_options & DPNI_LINK_OPT_ASYM_PAUSE);
+}
+
+static inline bool dpaa2_eth_rx_pause_enabled(u64 link_options)
+{
+	return !!(link_options & DPNI_LINK_OPT_PAUSE);
+}
+
+static inline unsigned int dpaa2_eth_needed_headroom(struct sk_buff *skb)
+{
+	unsigned int headroom = DPAA2_ETH_SWA_SIZE + DPAA2_ETH_TX_BUF_ALIGN;
+
+	/* If we don't have an skb (e.g. XDP buffer), we only need space for
+	 * the software annotation area
+	 */
+	if (!skb)
+		return headroom;
+
+	/* For non-linear skbs we have no headroom requirement, as we build a
+	 * SG frame with a newly allocated SGT buffer
+	 */
+	if (skb_is_nonlinear(skb))
+		return 0;
+
+	/* If we have Tx timestamping, need 128B hardware annotation */
+	if (skb->cb[0])
+		headroom += DPAA2_ETH_TX_HWA_SIZE;
+
+	return headroom;
+}
+
+/* Extra headroom space requested to hardware, in order to make sure there's
+ * no realloc'ing in forwarding scenarios
+ */
+static inline unsigned int dpaa2_eth_rx_head_room(struct dpaa2_eth_priv *priv)
+{
+	return priv->tx_data_offset - DPAA2_ETH_RX_HWA_SIZE;
+}
+
+static inline bool dpaa2_eth_is_type_phy(struct dpaa2_eth_priv *priv)
+{
+	lockdep_assert_held(&priv->mac_lock);
+
+	return dpaa2_mac_is_type_phy(priv->mac);
+}
+
+static inline bool dpaa2_eth_has_mac(struct dpaa2_eth_priv *priv)
+{
+	lockdep_assert_held(&priv->mac_lock);
+
+	return priv->mac ? true : false;
+}
+
+int dpaa2_eth_set_hash(struct net_device *net_dev, u64 flags);
+int dpaa2_eth_set_cls(struct net_device *net_dev, u64 key);
+int dpaa2_eth_cls_key_size(u64 key);
+int dpaa2_eth_cls_fld_off(int prot, int field);
+void dpaa2_eth_cls_trim_rule(void *key_mem, u64 fields);
+
+void dpaa2_eth_set_rx_taildrop(struct dpaa2_eth_priv *priv,
+			       bool tx_pause, bool pfc);
+
+extern const struct dcbnl_rtnl_ops dpaa2_eth_dcbnl_ops;
+
+int dpaa2_eth_dl_alloc(struct dpaa2_eth_priv *priv);
+void dpaa2_eth_dl_free(struct dpaa2_eth_priv *priv);
+
+void dpaa2_eth_dl_register(struct dpaa2_eth_priv *priv);
+void dpaa2_eth_dl_unregister(struct dpaa2_eth_priv *priv);
+
+int dpaa2_eth_dl_port_add(struct dpaa2_eth_priv *priv);
+void dpaa2_eth_dl_port_del(struct dpaa2_eth_priv *priv);
+
+int dpaa2_eth_dl_traps_register(struct dpaa2_eth_priv *priv);
+void dpaa2_eth_dl_traps_unregister(struct dpaa2_eth_priv *priv);
+
+struct dpaa2_eth_trap_item *dpaa2_eth_dl_get_trap(struct dpaa2_eth_priv *priv,
+						  struct dpaa2_fapr *fapr);
+
+struct dpaa2_eth_bp *dpaa2_eth_allocate_dpbp(struct dpaa2_eth_priv *priv);
+void dpaa2_eth_free_dpbp(struct dpaa2_eth_priv *priv, struct dpaa2_eth_bp *bp);
+
+struct sk_buff *dpaa2_eth_alloc_skb(struct dpaa2_eth_priv *priv,
+				    struct dpaa2_eth_channel *ch,
+				    const struct dpaa2_fd *fd, u32 fd_length,
+				    void *fd_vaddr);
+
+void dpaa2_eth_receive_skb(struct dpaa2_eth_priv *priv,
+			   struct dpaa2_eth_channel *ch,
+			   const struct dpaa2_fd *fd, void *vaddr,
+			   struct dpaa2_eth_fq *fq,
+			   struct rtnl_link_stats64 *percpu_stats,
+			   struct sk_buff *skb);
+
+void dpaa2_eth_rx(struct dpaa2_eth_priv *priv,
+		  struct dpaa2_eth_channel *ch,
+		  const struct dpaa2_fd *fd,
+		  struct dpaa2_eth_fq *fq);
+
+struct dpaa2_eth_bp *dpaa2_eth_allocate_dpbp(struct dpaa2_eth_priv *priv);
+void dpaa2_eth_free_dpbp(struct dpaa2_eth_priv *priv,
+			 struct dpaa2_eth_bp *bp);
+
+void *dpaa2_iova_to_virt(struct iommu_domain *domain, dma_addr_t iova_addr);
+void dpaa2_eth_recycle_buf(struct dpaa2_eth_priv *priv,
+			   struct dpaa2_eth_channel *ch,
+			   dma_addr_t addr);
+
+void dpaa2_eth_xdp_enqueue(struct dpaa2_eth_priv *priv,
+			   struct dpaa2_eth_channel *ch,
+			   struct dpaa2_fd *fd,
+			   void *buf_start, u16 queue_id);
+
+int dpaa2_xsk_wakeup(struct net_device *dev, u32 qid, u32 flags);
+int dpaa2_xsk_setup_pool(struct net_device *dev, struct xsk_buff_pool *pool, u16 qid);
+
+void dpaa2_eth_free_tx_fd(struct dpaa2_eth_priv *priv,
+			  struct dpaa2_eth_channel *ch,
+			  struct dpaa2_eth_fq *fq,
+			  const struct dpaa2_fd *fd, bool in_napi);
+bool dpaa2_xsk_tx(struct dpaa2_eth_priv *priv,
+		  struct dpaa2_eth_channel *ch);
+
+/* SGT (Scatter-Gather Table) cache management */
+void *dpaa2_eth_sgt_get(struct dpaa2_eth_priv *priv);
+
+void dpaa2_eth_sgt_recycle(struct dpaa2_eth_priv *priv, void *sgt_buf);
+
+#endif	/* __DPAA2_H */
diff --git a/drivers/net/ethernet/freescale/dpaa2/dpaa2-ethtool.c b/drivers/net/ethernet/freescale/dpaa2/dpaa2-ethtool.c
new file mode 100644
index 000000000000..00474ed11d53
--- /dev/null
+++ b/drivers/net/ethernet/freescale/dpaa2/dpaa2-ethtool.c
@@ -0,0 +1,960 @@
+// SPDX-License-Identifier: (GPL-2.0+ OR BSD-3-Clause)
+/* Copyright 2014-2016 Freescale Semiconductor Inc.
+ * Copyright 2016-2022 NXP
+ */
+
+#include <linux/net_tstamp.h>
+#include <linux/nospec.h>
+
+#include "dpni.h"	/* DPNI_LINK_OPT_* */
+#include "dpaa2-eth.h"
+
+/* To be kept in sync with DPNI statistics */
+static char dpaa2_ethtool_stats[][ETH_GSTRING_LEN] = {
+	"[hw] rx frames",
+	"[hw] rx bytes",
+	"[hw] rx mcast frames",
+	"[hw] rx mcast bytes",
+	"[hw] rx bcast frames",
+	"[hw] rx bcast bytes",
+	"[hw] tx frames",
+	"[hw] tx bytes",
+	"[hw] tx mcast frames",
+	"[hw] tx mcast bytes",
+	"[hw] tx bcast frames",
+	"[hw] tx bcast bytes",
+	"[hw] rx filtered frames",
+	"[hw] rx discarded frames",
+	"[hw] rx nobuffer discards",
+	"[hw] tx discarded frames",
+	"[hw] tx confirmed frames",
+	"[hw] tx dequeued bytes",
+	"[hw] tx dequeued frames",
+	"[hw] tx rejected bytes",
+	"[hw] tx rejected frames",
+	"[hw] tx pending frames",
+};
+
+#define DPAA2_ETH_NUM_STATS	ARRAY_SIZE(dpaa2_ethtool_stats)
+
+static char dpaa2_ethtool_extras[][ETH_GSTRING_LEN] = {
+	/* per-cpu stats */
+	"[drv] tx conf frames",
+	"[drv] tx conf bytes",
+	"[drv] tx sg frames",
+	"[drv] tx sg bytes",
+	"[drv] tx tso frames",
+	"[drv] tx tso bytes",
+	"[drv] rx sg frames",
+	"[drv] rx sg bytes",
+	"[drv] tx converted sg frames",
+	"[drv] tx converted sg bytes",
+	"[drv] enqueue portal busy",
+	/* Channel stats */
+	"[drv] dequeue portal busy",
+	"[drv] channel pull errors",
+	"[drv] cdan",
+	"[drv] xdp drop",
+	"[drv] xdp tx",
+	"[drv] xdp tx errors",
+	"[drv] xdp redirect",
+	/* FQ stats */
+	"[qbman] rx pending frames",
+	"[qbman] rx pending bytes",
+	"[qbman] tx conf pending frames",
+	"[qbman] tx conf pending bytes",
+	"[qbman] buffer count",
+};
+
+#define DPAA2_ETH_NUM_EXTRA_STATS	ARRAY_SIZE(dpaa2_ethtool_extras)
+
+static void dpaa2_eth_get_drvinfo(struct net_device *net_dev,
+				  struct ethtool_drvinfo *drvinfo)
+{
+	struct dpaa2_eth_priv *priv = netdev_priv(net_dev);
+
+	strscpy(drvinfo->driver, KBUILD_MODNAME, sizeof(drvinfo->driver));
+
+	snprintf(drvinfo->fw_version, sizeof(drvinfo->fw_version),
+		 "%u.%u", priv->dpni_ver_major, priv->dpni_ver_minor);
+
+	strscpy(drvinfo->bus_info, dev_name(net_dev->dev.parent->parent),
+		sizeof(drvinfo->bus_info));
+}
+
+static int dpaa2_eth_nway_reset(struct net_device *net_dev)
+{
+	struct dpaa2_eth_priv *priv = netdev_priv(net_dev);
+	int err = -EOPNOTSUPP;
+
+	mutex_lock(&priv->mac_lock);
+
+	if (dpaa2_eth_is_type_phy(priv))
+		err = phylink_ethtool_nway_reset(priv->mac->phylink);
+
+	mutex_unlock(&priv->mac_lock);
+
+	return err;
+}
+
+static int
+dpaa2_eth_get_link_ksettings(struct net_device *net_dev,
+			     struct ethtool_link_ksettings *link_settings)
+{
+	struct dpaa2_eth_priv *priv = netdev_priv(net_dev);
+	int err;
+
+	mutex_lock(&priv->mac_lock);
+
+	if (dpaa2_eth_is_type_phy(priv)) {
+		err = phylink_ethtool_ksettings_get(priv->mac->phylink,
+						    link_settings);
+		mutex_unlock(&priv->mac_lock);
+		return err;
+	}
+
+	mutex_unlock(&priv->mac_lock);
+
+	link_settings->base.autoneg = AUTONEG_DISABLE;
+	if (!(priv->link_state.options & DPNI_LINK_OPT_HALF_DUPLEX))
+		link_settings->base.duplex = DUPLEX_FULL;
+	link_settings->base.speed = priv->link_state.rate;
+
+	return 0;
+}
+
+static int
+dpaa2_eth_set_link_ksettings(struct net_device *net_dev,
+			     const struct ethtool_link_ksettings *link_settings)
+{
+	struct dpaa2_eth_priv *priv = netdev_priv(net_dev);
+	int err = -EOPNOTSUPP;
+
+	mutex_lock(&priv->mac_lock);
+
+	if (dpaa2_eth_is_type_phy(priv))
+		err = phylink_ethtool_ksettings_set(priv->mac->phylink,
+						    link_settings);
+
+	mutex_unlock(&priv->mac_lock);
+
+	return err;
+}
+
+static void dpaa2_eth_get_pauseparam(struct net_device *net_dev,
+				     struct ethtool_pauseparam *pause)
+{
+	struct dpaa2_eth_priv *priv = netdev_priv(net_dev);
+	u64 link_options = priv->link_state.options;
+
+	mutex_lock(&priv->mac_lock);
+
+	if (dpaa2_eth_is_type_phy(priv)) {
+		phylink_ethtool_get_pauseparam(priv->mac->phylink, pause);
+		mutex_unlock(&priv->mac_lock);
+		return;
+	}
+
+	mutex_unlock(&priv->mac_lock);
+
+	pause->rx_pause = dpaa2_eth_rx_pause_enabled(link_options);
+	pause->tx_pause = dpaa2_eth_tx_pause_enabled(link_options);
+	pause->autoneg = AUTONEG_DISABLE;
+}
+
+static int dpaa2_eth_set_pauseparam(struct net_device *net_dev,
+				    struct ethtool_pauseparam *pause)
+{
+	struct dpaa2_eth_priv *priv = netdev_priv(net_dev);
+	struct dpni_link_cfg cfg = {0};
+	int err;
+
+	if (!dpaa2_eth_has_pause_support(priv)) {
+		netdev_info(net_dev, "No pause frame support for DPNI version < %d.%d\n",
+			    DPNI_PAUSE_VER_MAJOR, DPNI_PAUSE_VER_MINOR);
+		return -EOPNOTSUPP;
+	}
+
+	mutex_lock(&priv->mac_lock);
+
+	if (dpaa2_eth_is_type_phy(priv)) {
+		err = phylink_ethtool_set_pauseparam(priv->mac->phylink,
+						     pause);
+		mutex_unlock(&priv->mac_lock);
+		return err;
+	}
+
+	mutex_unlock(&priv->mac_lock);
+
+	if (pause->autoneg)
+		return -EOPNOTSUPP;
+
+	cfg.rate = priv->link_state.rate;
+	cfg.options = priv->link_state.options;
+	if (pause->rx_pause)
+		cfg.options |= DPNI_LINK_OPT_PAUSE;
+	else
+		cfg.options &= ~DPNI_LINK_OPT_PAUSE;
+	if (!!pause->rx_pause ^ !!pause->tx_pause)
+		cfg.options |= DPNI_LINK_OPT_ASYM_PAUSE;
+	else
+		cfg.options &= ~DPNI_LINK_OPT_ASYM_PAUSE;
+
+	if (cfg.options == priv->link_state.options)
+		return 0;
+
+	err = dpni_set_link_cfg(priv->mc_io, 0, priv->mc_token, &cfg);
+	if (err) {
+		netdev_err(net_dev, "dpni_set_link_state failed\n");
+		return err;
+	}
+
+	priv->link_state.options = cfg.options;
+
+	return 0;
+}
+
+static void dpaa2_eth_get_strings(struct net_device *netdev, u32 stringset,
+				  u8 *data)
+{
+	int i;
+
+	switch (stringset) {
+	case ETH_SS_STATS:
+		for (i = 0; i < DPAA2_ETH_NUM_STATS; i++)
+			ethtool_puts(&data, dpaa2_ethtool_stats[i]);
+		for (i = 0; i < DPAA2_ETH_NUM_EXTRA_STATS; i++)
+			ethtool_puts(&data, dpaa2_ethtool_extras[i]);
+		dpaa2_mac_get_strings(&data);
+		break;
+	}
+}
+
+static int dpaa2_eth_get_sset_count(struct net_device *net_dev, int sset)
+{
+	switch (sset) {
+	case ETH_SS_STATS: /* ethtool_get_stats(), ethtool_get_drvinfo() */
+		return DPAA2_ETH_NUM_STATS + DPAA2_ETH_NUM_EXTRA_STATS +
+		       dpaa2_mac_get_sset_count();
+	default:
+		return -EOPNOTSUPP;
+	}
+}
+
+/** Fill in hardware counters, as returned by MC.
+ */
+static void dpaa2_eth_get_ethtool_stats(struct net_device *net_dev,
+					struct ethtool_stats *stats,
+					u64 *data)
+{
+	struct dpaa2_eth_priv *priv = netdev_priv(net_dev);
+	union dpni_statistics dpni_stats;
+	int dpni_stats_page_size[DPNI_STATISTICS_CNT] = {
+		sizeof(dpni_stats.page_0),
+		sizeof(dpni_stats.page_1),
+		sizeof(dpni_stats.page_2),
+		sizeof(dpni_stats.page_3),
+		sizeof(dpni_stats.page_4),
+		sizeof(dpni_stats.page_5),
+		sizeof(dpni_stats.page_6),
+	};
+	u32 fcnt_rx_total = 0, fcnt_tx_total = 0;
+	u32 bcnt_rx_total = 0, bcnt_tx_total = 0;
+	struct dpaa2_eth_ch_stats *ch_stats;
+	struct dpaa2_eth_drv_stats *extras;
+	u32 buf_cnt, buf_cnt_total = 0;
+	int j, k, err, num_cnt, i = 0;
+	u32 fcnt, bcnt;
+
+	memset(data, 0,
+	       sizeof(u64) * (DPAA2_ETH_NUM_STATS + DPAA2_ETH_NUM_EXTRA_STATS));
+
+	/* Print standard counters, from DPNI statistics */
+	for (j = 0; j <= 6; j++) {
+		/* We're not interested in pages 4 & 5 for now */
+		if (j == 4 || j == 5)
+			continue;
+		err = dpni_get_statistics(priv->mc_io, 0, priv->mc_token,
+					  j, &dpni_stats);
+		if (err == -EINVAL)
+			/* Older firmware versions don't support all pages */
+			memset(&dpni_stats, 0, sizeof(dpni_stats));
+		else if (err)
+			netdev_warn(net_dev, "dpni_get_stats(%d) failed\n", j);
+
+		num_cnt = dpni_stats_page_size[j] / sizeof(u64);
+		for (k = 0; k < num_cnt; k++)
+			*(data + i++) = dpni_stats.raw.counter[k];
+	}
+
+	/* Print per-cpu extra stats */
+	for_each_online_cpu(k) {
+		extras = per_cpu_ptr(priv->percpu_extras, k);
+		for (j = 0; j < sizeof(*extras) / sizeof(__u64); j++)
+			*((__u64 *)data + i + j) += *((__u64 *)extras + j);
+	}
+	i += j;
+
+	/* Per-channel stats */
+	for (k = 0; k < priv->num_channels; k++) {
+		ch_stats = &priv->channel[k]->stats;
+		for (j = 0; j < DPAA2_ETH_CH_STATS; j++)
+			*((__u64 *)data + i + j) += *((__u64 *)ch_stats + j);
+	}
+	i += j;
+
+	for (j = 0; j < priv->num_fqs; j++) {
+		/* Print FQ instantaneous counts */
+		err = dpaa2_io_query_fq_count(NULL, priv->fq[j].fqid,
+					      &fcnt, &bcnt);
+		if (err) {
+			netdev_warn(net_dev, "FQ query error %d", err);
+			return;
+		}
+
+		if (priv->fq[j].type == DPAA2_TX_CONF_FQ) {
+			fcnt_tx_total += fcnt;
+			bcnt_tx_total += bcnt;
+		} else {
+			fcnt_rx_total += fcnt;
+			bcnt_rx_total += bcnt;
+		}
+	}
+
+	*(data + i++) = fcnt_rx_total;
+	*(data + i++) = bcnt_rx_total;
+	*(data + i++) = fcnt_tx_total;
+	*(data + i++) = bcnt_tx_total;
+
+	for (j = 0; j < priv->num_bps; j++) {
+		err = dpaa2_io_query_bp_count(NULL, priv->bp[j]->bpid, &buf_cnt);
+		if (err) {
+			netdev_warn(net_dev, "Buffer count query error %d\n", err);
+			return;
+		}
+		buf_cnt_total += buf_cnt;
+	}
+	*(data + i++) = buf_cnt_total;
+
+	mutex_lock(&priv->mac_lock);
+
+	if (dpaa2_eth_has_mac(priv))
+		dpaa2_mac_get_ethtool_stats(priv->mac, data + i);
+
+	mutex_unlock(&priv->mac_lock);
+}
+
+static int dpaa2_eth_prep_eth_rule(struct ethhdr *eth_value, struct ethhdr *eth_mask,
+				   void *key, void *mask, u64 *fields)
+{
+	int off;
+
+	if (eth_mask->h_proto) {
+		off = dpaa2_eth_cls_fld_off(NET_PROT_ETH, NH_FLD_ETH_TYPE);
+		*(__be16 *)(key + off) = eth_value->h_proto;
+		*(__be16 *)(mask + off) = eth_mask->h_proto;
+		*fields |= DPAA2_ETH_DIST_ETHTYPE;
+	}
+
+	if (!is_zero_ether_addr(eth_mask->h_source)) {
+		off = dpaa2_eth_cls_fld_off(NET_PROT_ETH, NH_FLD_ETH_SA);
+		ether_addr_copy(key + off, eth_value->h_source);
+		ether_addr_copy(mask + off, eth_mask->h_source);
+		*fields |= DPAA2_ETH_DIST_ETHSRC;
+	}
+
+	if (!is_zero_ether_addr(eth_mask->h_dest)) {
+		off = dpaa2_eth_cls_fld_off(NET_PROT_ETH, NH_FLD_ETH_DA);
+		ether_addr_copy(key + off, eth_value->h_dest);
+		ether_addr_copy(mask + off, eth_mask->h_dest);
+		*fields |= DPAA2_ETH_DIST_ETHDST;
+	}
+
+	return 0;
+}
+
+static int dpaa2_eth_prep_uip_rule(struct ethtool_usrip4_spec *uip_value,
+				   struct ethtool_usrip4_spec *uip_mask,
+				   void *key, void *mask, u64 *fields)
+{
+	int off;
+	u32 tmp_value, tmp_mask;
+
+	if (uip_mask->tos || uip_mask->ip_ver)
+		return -EOPNOTSUPP;
+
+	if (uip_mask->ip4src) {
+		off = dpaa2_eth_cls_fld_off(NET_PROT_IP, NH_FLD_IP_SRC);
+		*(__be32 *)(key + off) = uip_value->ip4src;
+		*(__be32 *)(mask + off) = uip_mask->ip4src;
+		*fields |= DPAA2_ETH_DIST_IPSRC;
+	}
+
+	if (uip_mask->ip4dst) {
+		off = dpaa2_eth_cls_fld_off(NET_PROT_IP, NH_FLD_IP_DST);
+		*(__be32 *)(key + off) = uip_value->ip4dst;
+		*(__be32 *)(mask + off) = uip_mask->ip4dst;
+		*fields |= DPAA2_ETH_DIST_IPDST;
+	}
+
+	if (uip_mask->proto) {
+		off = dpaa2_eth_cls_fld_off(NET_PROT_IP, NH_FLD_IP_PROTO);
+		*(u8 *)(key + off) = uip_value->proto;
+		*(u8 *)(mask + off) = uip_mask->proto;
+		*fields |= DPAA2_ETH_DIST_IPPROTO;
+	}
+
+	if (uip_mask->l4_4_bytes) {
+		tmp_value = be32_to_cpu(uip_value->l4_4_bytes);
+		tmp_mask = be32_to_cpu(uip_mask->l4_4_bytes);
+
+		off = dpaa2_eth_cls_fld_off(NET_PROT_UDP, NH_FLD_UDP_PORT_SRC);
+		*(__be16 *)(key + off) = htons(tmp_value >> 16);
+		*(__be16 *)(mask + off) = htons(tmp_mask >> 16);
+		*fields |= DPAA2_ETH_DIST_L4SRC;
+
+		off = dpaa2_eth_cls_fld_off(NET_PROT_UDP, NH_FLD_UDP_PORT_DST);
+		*(__be16 *)(key + off) = htons(tmp_value & 0xFFFF);
+		*(__be16 *)(mask + off) = htons(tmp_mask & 0xFFFF);
+		*fields |= DPAA2_ETH_DIST_L4DST;
+	}
+
+	/* Only apply the rule for IPv4 frames */
+	off = dpaa2_eth_cls_fld_off(NET_PROT_ETH, NH_FLD_ETH_TYPE);
+	*(__be16 *)(key + off) = htons(ETH_P_IP);
+	*(__be16 *)(mask + off) = htons(0xFFFF);
+	*fields |= DPAA2_ETH_DIST_ETHTYPE;
+
+	return 0;
+}
+
+static int dpaa2_eth_prep_l4_rule(struct ethtool_tcpip4_spec *l4_value,
+				  struct ethtool_tcpip4_spec *l4_mask,
+				  void *key, void *mask, u8 l4_proto, u64 *fields)
+{
+	int off;
+
+	if (l4_mask->tos)
+		return -EOPNOTSUPP;
+
+	if (l4_mask->ip4src) {
+		off = dpaa2_eth_cls_fld_off(NET_PROT_IP, NH_FLD_IP_SRC);
+		*(__be32 *)(key + off) = l4_value->ip4src;
+		*(__be32 *)(mask + off) = l4_mask->ip4src;
+		*fields |= DPAA2_ETH_DIST_IPSRC;
+	}
+
+	if (l4_mask->ip4dst) {
+		off = dpaa2_eth_cls_fld_off(NET_PROT_IP, NH_FLD_IP_DST);
+		*(__be32 *)(key + off) = l4_value->ip4dst;
+		*(__be32 *)(mask + off) = l4_mask->ip4dst;
+		*fields |= DPAA2_ETH_DIST_IPDST;
+	}
+
+	if (l4_mask->psrc) {
+		off = dpaa2_eth_cls_fld_off(NET_PROT_UDP, NH_FLD_UDP_PORT_SRC);
+		*(__be16 *)(key + off) = l4_value->psrc;
+		*(__be16 *)(mask + off) = l4_mask->psrc;
+		*fields |= DPAA2_ETH_DIST_L4SRC;
+	}
+
+	if (l4_mask->pdst) {
+		off = dpaa2_eth_cls_fld_off(NET_PROT_UDP, NH_FLD_UDP_PORT_DST);
+		*(__be16 *)(key + off) = l4_value->pdst;
+		*(__be16 *)(mask + off) = l4_mask->pdst;
+		*fields |= DPAA2_ETH_DIST_L4DST;
+	}
+
+	/* Only apply the rule for IPv4 frames with the specified L4 proto */
+	off = dpaa2_eth_cls_fld_off(NET_PROT_ETH, NH_FLD_ETH_TYPE);
+	*(__be16 *)(key + off) = htons(ETH_P_IP);
+	*(__be16 *)(mask + off) = htons(0xFFFF);
+	*fields |= DPAA2_ETH_DIST_ETHTYPE;
+
+	off = dpaa2_eth_cls_fld_off(NET_PROT_IP, NH_FLD_IP_PROTO);
+	*(u8 *)(key + off) = l4_proto;
+	*(u8 *)(mask + off) = 0xFF;
+	*fields |= DPAA2_ETH_DIST_IPPROTO;
+
+	return 0;
+}
+
+static int dpaa2_eth_prep_ext_rule(struct ethtool_flow_ext *ext_value,
+				   struct ethtool_flow_ext *ext_mask,
+				   void *key, void *mask, u64 *fields)
+{
+	int off;
+
+	if (ext_mask->vlan_etype)
+		return -EOPNOTSUPP;
+
+	if (ext_mask->vlan_tci) {
+		off = dpaa2_eth_cls_fld_off(NET_PROT_VLAN, NH_FLD_VLAN_TCI);
+		*(__be16 *)(key + off) = ext_value->vlan_tci;
+		*(__be16 *)(mask + off) = ext_mask->vlan_tci;
+		*fields |= DPAA2_ETH_DIST_VLAN;
+	}
+
+	return 0;
+}
+
+static int dpaa2_eth_prep_mac_ext_rule(struct ethtool_flow_ext *ext_value,
+				       struct ethtool_flow_ext *ext_mask,
+				       void *key, void *mask, u64 *fields)
+{
+	int off;
+
+	if (!is_zero_ether_addr(ext_mask->h_dest)) {
+		off = dpaa2_eth_cls_fld_off(NET_PROT_ETH, NH_FLD_ETH_DA);
+		ether_addr_copy(key + off, ext_value->h_dest);
+		ether_addr_copy(mask + off, ext_mask->h_dest);
+		*fields |= DPAA2_ETH_DIST_ETHDST;
+	}
+
+	return 0;
+}
+
+static int dpaa2_eth_prep_cls_rule(struct ethtool_rx_flow_spec *fs, void *key,
+				   void *mask, u64 *fields)
+{
+	int err;
+
+	switch (fs->flow_type & 0xFF) {
+	case ETHER_FLOW:
+		err = dpaa2_eth_prep_eth_rule(&fs->h_u.ether_spec, &fs->m_u.ether_spec,
+					      key, mask, fields);
+		break;
+	case IP_USER_FLOW:
+		err = dpaa2_eth_prep_uip_rule(&fs->h_u.usr_ip4_spec,
+					      &fs->m_u.usr_ip4_spec, key, mask, fields);
+		break;
+	case TCP_V4_FLOW:
+		err = dpaa2_eth_prep_l4_rule(&fs->h_u.tcp_ip4_spec, &fs->m_u.tcp_ip4_spec,
+					     key, mask, IPPROTO_TCP, fields);
+		break;
+	case UDP_V4_FLOW:
+		err = dpaa2_eth_prep_l4_rule(&fs->h_u.udp_ip4_spec, &fs->m_u.udp_ip4_spec,
+					     key, mask, IPPROTO_UDP, fields);
+		break;
+	case SCTP_V4_FLOW:
+		err = dpaa2_eth_prep_l4_rule(&fs->h_u.sctp_ip4_spec,
+					     &fs->m_u.sctp_ip4_spec, key, mask,
+					     IPPROTO_SCTP, fields);
+		break;
+	default:
+		return -EOPNOTSUPP;
+	}
+
+	if (err)
+		return err;
+
+	if (fs->flow_type & FLOW_EXT) {
+		err = dpaa2_eth_prep_ext_rule(&fs->h_ext, &fs->m_ext, key, mask, fields);
+		if (err)
+			return err;
+	}
+
+	if (fs->flow_type & FLOW_MAC_EXT) {
+		err = dpaa2_eth_prep_mac_ext_rule(&fs->h_ext, &fs->m_ext, key,
+						  mask, fields);
+		if (err)
+			return err;
+	}
+
+	return 0;
+}
+
+static int dpaa2_eth_do_cls_rule(struct net_device *net_dev,
+				 struct ethtool_rx_flow_spec *fs,
+				 bool add)
+{
+	struct dpaa2_eth_priv *priv = netdev_priv(net_dev);
+	struct device *dev = net_dev->dev.parent;
+	struct dpni_rule_cfg rule_cfg = { 0 };
+	struct dpni_fs_action_cfg fs_act = { 0 };
+	dma_addr_t key_iova;
+	u64 fields = 0;
+	void *key_buf;
+	int i, err;
+
+	if (fs->ring_cookie != RX_CLS_FLOW_DISC &&
+	    fs->ring_cookie >= dpaa2_eth_queue_count(priv))
+		return -EINVAL;
+
+	rule_cfg.key_size = dpaa2_eth_cls_key_size(DPAA2_ETH_DIST_ALL);
+
+	/* allocate twice the key size, for the actual key and for mask */
+	key_buf = kzalloc(rule_cfg.key_size * 2, GFP_KERNEL);
+	if (!key_buf)
+		return -ENOMEM;
+
+	/* Fill the key and mask memory areas */
+	err = dpaa2_eth_prep_cls_rule(fs, key_buf, key_buf + rule_cfg.key_size, &fields);
+	if (err)
+		goto free_mem;
+
+	if (!dpaa2_eth_fs_mask_enabled(priv)) {
+		/* Masking allows us to configure a maximal key during init and
+		 * use it for all flow steering rules. Without it, we include
+		 * in the key only the fields actually used, so we need to
+		 * extract the others from the final key buffer.
+		 *
+		 * Program the FS key if needed, or return error if previously
+		 * set key can't be used for the current rule. User needs to
+		 * delete existing rules in this case to allow for the new one.
+		 */
+		if (!priv->rx_cls_fields) {
+			err = dpaa2_eth_set_cls(net_dev, fields);
+			if (err)
+				goto free_mem;
+
+			priv->rx_cls_fields = fields;
+		} else if (priv->rx_cls_fields != fields) {
+			netdev_err(net_dev, "No support for multiple FS keys, need to delete existing rules\n");
+			err = -EOPNOTSUPP;
+			goto free_mem;
+		}
+
+		dpaa2_eth_cls_trim_rule(key_buf, fields);
+		rule_cfg.key_size = dpaa2_eth_cls_key_size(fields);
+	}
+
+	key_iova = dma_map_single(dev, key_buf, rule_cfg.key_size * 2,
+				  DMA_TO_DEVICE);
+	if (dma_mapping_error(dev, key_iova)) {
+		err = -ENOMEM;
+		goto free_mem;
+	}
+
+	rule_cfg.key_iova = key_iova;
+	if (dpaa2_eth_fs_mask_enabled(priv))
+		rule_cfg.mask_iova = key_iova + rule_cfg.key_size;
+
+	if (add) {
+		if (fs->ring_cookie == RX_CLS_FLOW_DISC)
+			fs_act.options |= DPNI_FS_OPT_DISCARD;
+		else
+			fs_act.flow_id = fs->ring_cookie;
+	}
+	for (i = 0; i < dpaa2_eth_tc_count(priv); i++) {
+		if (add)
+			err = dpni_add_fs_entry(priv->mc_io, 0, priv->mc_token,
+						i, fs->location, &rule_cfg,
+						&fs_act);
+		else
+			err = dpni_remove_fs_entry(priv->mc_io, 0,
+						   priv->mc_token, i,
+						   &rule_cfg);
+		if (err || priv->dpni_attrs.options & DPNI_OPT_SHARED_FS)
+			break;
+	}
+
+	dma_unmap_single(dev, key_iova, rule_cfg.key_size * 2, DMA_TO_DEVICE);
+
+free_mem:
+	kfree(key_buf);
+
+	return err;
+}
+
+static int dpaa2_eth_num_cls_rules(struct dpaa2_eth_priv *priv)
+{
+	int i, rules = 0;
+
+	for (i = 0; i < dpaa2_eth_fs_count(priv); i++)
+		if (priv->cls_rules[i].in_use)
+			rules++;
+
+	return rules;
+}
+
+static int dpaa2_eth_update_cls_rule(struct net_device *net_dev,
+				     struct ethtool_rx_flow_spec *new_fs,
+				     unsigned int location)
+{
+	struct dpaa2_eth_priv *priv = netdev_priv(net_dev);
+	struct dpaa2_eth_cls_rule *rule;
+	int err = -EINVAL;
+
+	if (!priv->rx_cls_enabled)
+		return -EOPNOTSUPP;
+
+	if (location >= dpaa2_eth_fs_count(priv))
+		return -EINVAL;
+
+	rule = &priv->cls_rules[location];
+
+	/* If a rule is present at the specified location, delete it. */
+	if (rule->in_use) {
+		err = dpaa2_eth_do_cls_rule(net_dev, &rule->fs, false);
+		if (err)
+			return err;
+
+		rule->in_use = 0;
+
+		if (!dpaa2_eth_fs_mask_enabled(priv) &&
+		    !dpaa2_eth_num_cls_rules(priv))
+			priv->rx_cls_fields = 0;
+	}
+
+	/* If no new entry to add, return here */
+	if (!new_fs)
+		return err;
+
+	err = dpaa2_eth_do_cls_rule(net_dev, new_fs, true);
+	if (err)
+		return err;
+
+	rule->in_use = 1;
+	rule->fs = *new_fs;
+
+	return 0;
+}
+
+static int dpaa2_eth_get_rxnfc(struct net_device *net_dev,
+			       struct ethtool_rxnfc *rxnfc, u32 *rule_locs)
+{
+	struct dpaa2_eth_priv *priv = netdev_priv(net_dev);
+	int max_rules = dpaa2_eth_fs_count(priv);
+	int i, j = 0;
+
+	switch (rxnfc->cmd) {
+	case ETHTOOL_GRXRINGS:
+		rxnfc->data = dpaa2_eth_queue_count(priv);
+		break;
+	case ETHTOOL_GRXCLSRLCNT:
+		rxnfc->rule_cnt = 0;
+		rxnfc->rule_cnt = dpaa2_eth_num_cls_rules(priv);
+		rxnfc->data = max_rules;
+		break;
+	case ETHTOOL_GRXCLSRULE:
+		if (rxnfc->fs.location >= max_rules)
+			return -EINVAL;
+		rxnfc->fs.location = array_index_nospec(rxnfc->fs.location,
+							max_rules);
+		if (!priv->cls_rules[rxnfc->fs.location].in_use)
+			return -EINVAL;
+		rxnfc->fs = priv->cls_rules[rxnfc->fs.location].fs;
+		break;
+	case ETHTOOL_GRXCLSRLALL:
+		for (i = 0; i < max_rules; i++) {
+			if (!priv->cls_rules[i].in_use)
+				continue;
+			if (j == rxnfc->rule_cnt)
+				return -EMSGSIZE;
+			rule_locs[j++] = i;
+		}
+		rxnfc->rule_cnt = j;
+		rxnfc->data = max_rules;
+		break;
+	default:
+		return -EOPNOTSUPP;
+	}
+
+	return 0;
+}
+
+static int dpaa2_eth_set_rxnfc(struct net_device *net_dev,
+			       struct ethtool_rxnfc *rxnfc)
+{
+	int err = 0;
+
+	switch (rxnfc->cmd) {
+	case ETHTOOL_SRXCLSRLINS:
+		err = dpaa2_eth_update_cls_rule(net_dev, &rxnfc->fs, rxnfc->fs.location);
+		break;
+	case ETHTOOL_SRXCLSRLDEL:
+		err = dpaa2_eth_update_cls_rule(net_dev, NULL, rxnfc->fs.location);
+		break;
+	default:
+		err = -EOPNOTSUPP;
+	}
+
+	return err;
+}
+
+static int dpaa2_eth_get_rxfh_fields(struct net_device *net_dev,
+				     struct ethtool_rxfh_fields *rxnfc)
+{
+	struct dpaa2_eth_priv *priv = netdev_priv(net_dev);
+
+	/* we purposely ignore cmd->flow_type for now, because the
+	 * classifier only supports a single set of fields for all
+	 * protocols
+	 */
+	rxnfc->data = priv->rx_hash_fields;
+	return 0;
+}
+
+static int dpaa2_eth_set_rxfh_fields(struct net_device *net_dev,
+				     const struct ethtool_rxfh_fields *rxnfc,
+				     struct netlink_ext_ack *extack)
+{
+	if ((rxnfc->data & DPAA2_RXH_SUPPORTED) != rxnfc->data)
+		return -EOPNOTSUPP;
+	return dpaa2_eth_set_hash(net_dev, rxnfc->data);
+}
+
+int dpaa2_phc_index = -1;
+EXPORT_SYMBOL(dpaa2_phc_index);
+
+static int dpaa2_eth_get_ts_info(struct net_device *dev,
+				 struct kernel_ethtool_ts_info *info)
+{
+	if (!dpaa2_ptp)
+		return ethtool_op_get_ts_info(dev, info);
+
+	info->so_timestamping = SOF_TIMESTAMPING_TX_HARDWARE |
+				SOF_TIMESTAMPING_RX_HARDWARE |
+				SOF_TIMESTAMPING_RAW_HARDWARE;
+
+	info->phc_index = dpaa2_phc_index;
+
+	info->tx_types = (1 << HWTSTAMP_TX_OFF) |
+			 (1 << HWTSTAMP_TX_ON) |
+			 (1 << HWTSTAMP_TX_ONESTEP_SYNC);
+
+	info->rx_filters = (1 << HWTSTAMP_FILTER_NONE) |
+			   (1 << HWTSTAMP_FILTER_ALL);
+	return 0;
+}
+
+static int dpaa2_eth_get_tunable(struct net_device *net_dev,
+				 const struct ethtool_tunable *tuna,
+				 void *data)
+{
+	struct dpaa2_eth_priv *priv = netdev_priv(net_dev);
+	int err = 0;
+
+	switch (tuna->id) {
+	case ETHTOOL_RX_COPYBREAK:
+		*(u32 *)data = priv->rx_copybreak;
+		break;
+	default:
+		err = -EOPNOTSUPP;
+		break;
+	}
+
+	return err;
+}
+
+static int dpaa2_eth_set_tunable(struct net_device *net_dev,
+				 const struct ethtool_tunable *tuna,
+				 const void *data)
+{
+	struct dpaa2_eth_priv *priv = netdev_priv(net_dev);
+	int err = 0;
+
+	switch (tuna->id) {
+	case ETHTOOL_RX_COPYBREAK:
+		priv->rx_copybreak = *(u32 *)data;
+		break;
+	default:
+		err = -EOPNOTSUPP;
+		break;
+	}
+
+	return err;
+}
+
+static int dpaa2_eth_get_coalesce(struct net_device *dev,
+				  struct ethtool_coalesce *ic,
+				  struct kernel_ethtool_coalesce *kernel_coal,
+				  struct netlink_ext_ack *extack)
+{
+	struct dpaa2_eth_priv *priv = netdev_priv(dev);
+	struct dpaa2_io *dpio = priv->channel[0]->dpio;
+
+	dpaa2_io_get_irq_coalescing(dpio, &ic->rx_coalesce_usecs);
+	ic->use_adaptive_rx_coalesce = dpaa2_io_get_adaptive_coalescing(dpio);
+
+	return 0;
+}
+
+static int dpaa2_eth_set_coalesce(struct net_device *dev,
+				  struct ethtool_coalesce *ic,
+				  struct kernel_ethtool_coalesce *kernel_coal,
+				  struct netlink_ext_ack *extack)
+{
+	struct dpaa2_eth_priv *priv = netdev_priv(dev);
+	struct dpaa2_io *dpio;
+	int prev_adaptive;
+	u32 prev_rx_usecs;
+	int i, j, err;
+
+	/* Keep track of the previous value, just in case we fail */
+	dpio = priv->channel[0]->dpio;
+	dpaa2_io_get_irq_coalescing(dpio, &prev_rx_usecs);
+	prev_adaptive = dpaa2_io_get_adaptive_coalescing(dpio);
+
+	/* Setup new value for rx coalescing */
+	for (i = 0; i < priv->num_channels; i++) {
+		dpio = priv->channel[i]->dpio;
+
+		dpaa2_io_set_adaptive_coalescing(dpio,
+						 ic->use_adaptive_rx_coalesce);
+		err = dpaa2_io_set_irq_coalescing(dpio, ic->rx_coalesce_usecs);
+		if (err)
+			goto restore_rx_usecs;
+	}
+
+	return 0;
+
+restore_rx_usecs:
+	for (j = 0; j < i; j++) {
+		dpio = priv->channel[j]->dpio;
+
+		dpaa2_io_set_irq_coalescing(dpio, prev_rx_usecs);
+		dpaa2_io_set_adaptive_coalescing(dpio, prev_adaptive);
+	}
+
+	return err;
+}
+
+static void dpaa2_eth_get_channels(struct net_device *net_dev,
+				   struct ethtool_channels *channels)
+{
+	struct dpaa2_eth_priv *priv = netdev_priv(net_dev);
+	int queue_count = dpaa2_eth_queue_count(priv);
+
+	channels->max_rx = queue_count;
+	channels->max_tx = queue_count;
+	channels->rx_count = queue_count;
+	channels->tx_count = queue_count;
+
+	/* Tx confirmation and Rx error */
+	channels->max_other = queue_count + 1;
+	channels->max_combined = channels->max_rx +
+				 channels->max_tx +
+				 channels->max_other;
+	/* Tx conf and Rx err */
+	channels->other_count = queue_count + 1;
+	channels->combined_count = channels->rx_count +
+				   channels->tx_count +
+				   channels->other_count;
+}
+
+const struct ethtool_ops dpaa2_ethtool_ops = {
+	.supported_coalesce_params = ETHTOOL_COALESCE_RX_USECS |
+				     ETHTOOL_COALESCE_USE_ADAPTIVE_RX,
+	.get_drvinfo = dpaa2_eth_get_drvinfo,
+	.nway_reset = dpaa2_eth_nway_reset,
+	.get_link = ethtool_op_get_link,
+	.get_link_ksettings = dpaa2_eth_get_link_ksettings,
+	.set_link_ksettings = dpaa2_eth_set_link_ksettings,
+	.get_pauseparam = dpaa2_eth_get_pauseparam,
+	.set_pauseparam = dpaa2_eth_set_pauseparam,
+	.get_sset_count = dpaa2_eth_get_sset_count,
+	.get_ethtool_stats = dpaa2_eth_get_ethtool_stats,
+	.get_strings = dpaa2_eth_get_strings,
+	.get_rxnfc = dpaa2_eth_get_rxnfc,
+	.set_rxnfc = dpaa2_eth_set_rxnfc,
+	.get_rxfh_fields = dpaa2_eth_get_rxfh_fields,
+	.set_rxfh_fields = dpaa2_eth_set_rxfh_fields,
+	.get_ts_info = dpaa2_eth_get_ts_info,
+	.get_tunable = dpaa2_eth_get_tunable,
+	.set_tunable = dpaa2_eth_set_tunable,
+	.get_coalesce = dpaa2_eth_get_coalesce,
+	.set_coalesce = dpaa2_eth_set_coalesce,
+	.get_channels = dpaa2_eth_get_channels,
+};
diff --git a/drivers/net/ethernet/freescale/dpaa2/dpaa2-mac.c b/drivers/net/ethernet/freescale/dpaa2/dpaa2-mac.c
new file mode 100644
index 000000000000..422ce13a7c94
--- /dev/null
+++ b/drivers/net/ethernet/freescale/dpaa2/dpaa2-mac.c
@@ -0,0 +1,586 @@
+// SPDX-License-Identifier: (GPL-2.0+ OR BSD-3-Clause)
+/* Copyright 2019 NXP */
+
+#include <linux/acpi.h>
+#include <linux/pcs-lynx.h>
+#include <linux/phy/phy.h>
+#include <linux/property.h>
+
+#include "dpaa2-eth.h"
+#include "dpaa2-mac.h"
+
+#define phylink_to_dpaa2_mac(config) \
+	container_of((config), struct dpaa2_mac, phylink_config)
+
+#define DPMAC_PROTOCOL_CHANGE_VER_MAJOR		4
+#define DPMAC_PROTOCOL_CHANGE_VER_MINOR		8
+
+#define DPAA2_MAC_FEATURE_PROTOCOL_CHANGE	BIT(0)
+
+static int dpaa2_mac_cmp_ver(struct dpaa2_mac *mac,
+			     u16 ver_major, u16 ver_minor)
+{
+	if (mac->ver_major == ver_major)
+		return mac->ver_minor - ver_minor;
+	return mac->ver_major - ver_major;
+}
+
+static void dpaa2_mac_detect_features(struct dpaa2_mac *mac)
+{
+	mac->features = 0;
+
+	if (dpaa2_mac_cmp_ver(mac, DPMAC_PROTOCOL_CHANGE_VER_MAJOR,
+			      DPMAC_PROTOCOL_CHANGE_VER_MINOR) >= 0)
+		mac->features |= DPAA2_MAC_FEATURE_PROTOCOL_CHANGE;
+}
+
+static int phy_mode(enum dpmac_eth_if eth_if, phy_interface_t *if_mode)
+{
+	*if_mode = PHY_INTERFACE_MODE_NA;
+
+	switch (eth_if) {
+	case DPMAC_ETH_IF_RGMII:
+		*if_mode = PHY_INTERFACE_MODE_RGMII;
+		break;
+	case DPMAC_ETH_IF_USXGMII:
+		*if_mode = PHY_INTERFACE_MODE_USXGMII;
+		break;
+	case DPMAC_ETH_IF_QSGMII:
+		*if_mode = PHY_INTERFACE_MODE_QSGMII;
+		break;
+	case DPMAC_ETH_IF_SGMII:
+		*if_mode = PHY_INTERFACE_MODE_SGMII;
+		break;
+	case DPMAC_ETH_IF_XFI:
+		*if_mode = PHY_INTERFACE_MODE_10GBASER;
+		break;
+	case DPMAC_ETH_IF_CAUI:
+		*if_mode = PHY_INTERFACE_MODE_25GBASER;
+		break;
+	default:
+		return -EINVAL;
+	}
+
+	return 0;
+}
+
+static enum dpmac_eth_if dpmac_eth_if_mode(phy_interface_t if_mode)
+{
+	switch (if_mode) {
+	case PHY_INTERFACE_MODE_RGMII:
+	case PHY_INTERFACE_MODE_RGMII_ID:
+	case PHY_INTERFACE_MODE_RGMII_RXID:
+	case PHY_INTERFACE_MODE_RGMII_TXID:
+		return DPMAC_ETH_IF_RGMII;
+	case PHY_INTERFACE_MODE_USXGMII:
+		return DPMAC_ETH_IF_USXGMII;
+	case PHY_INTERFACE_MODE_QSGMII:
+		return DPMAC_ETH_IF_QSGMII;
+	case PHY_INTERFACE_MODE_SGMII:
+		return DPMAC_ETH_IF_SGMII;
+	case PHY_INTERFACE_MODE_10GBASER:
+		return DPMAC_ETH_IF_XFI;
+	case PHY_INTERFACE_MODE_1000BASEX:
+		return DPMAC_ETH_IF_1000BASEX;
+	case PHY_INTERFACE_MODE_25GBASER:
+		return DPMAC_ETH_IF_CAUI;
+	default:
+		return DPMAC_ETH_IF_MII;
+	}
+}
+
+static struct fwnode_handle *dpaa2_mac_get_node(struct device *dev,
+						u16 dpmac_id)
+{
+	struct fwnode_handle *fwnode, *parent = NULL, *child  = NULL;
+	struct device_node *dpmacs = NULL;
+	int err;
+	u32 id;
+
+	fwnode = dev_fwnode(dev->parent);
+	if (is_of_node(fwnode)) {
+		dpmacs = of_find_node_by_name(NULL, "dpmacs");
+		if (!dpmacs)
+			return NULL;
+		parent = of_fwnode_handle(dpmacs);
+	} else if (is_acpi_node(fwnode)) {
+		parent = fwnode;
+	} else {
+		/* The root dprc device didn't yet get to finalize it's probe,
+		 * thus the fwnode field is not yet set. Defer probe if we are
+		 * facing this situation.
+		 */
+		dev_dbg(dev, "dprc not finished probing\n");
+		return ERR_PTR(-EPROBE_DEFER);
+	}
+
+	fwnode_for_each_child_node(parent, child) {
+		err = -EINVAL;
+		if (is_acpi_device_node(child))
+			err = acpi_get_local_address(ACPI_HANDLE_FWNODE(child), &id);
+		else if (is_of_node(child))
+			err = of_property_read_u32(to_of_node(child), "reg", &id);
+		if (err)
+			continue;
+
+		if (id == dpmac_id) {
+			of_node_put(dpmacs);
+			return child;
+		}
+	}
+	of_node_put(dpmacs);
+	return NULL;
+}
+
+static int dpaa2_mac_get_if_mode(struct fwnode_handle *dpmac_node,
+				 struct dpmac_attr attr)
+{
+	phy_interface_t if_mode;
+	int err;
+
+	err = fwnode_get_phy_mode(dpmac_node);
+	if (err > 0)
+		return err;
+
+	err = phy_mode(attr.eth_if, &if_mode);
+	if (!err)
+		return if_mode;
+
+	return err;
+}
+
+static struct phylink_pcs *dpaa2_mac_select_pcs(struct phylink_config *config,
+						phy_interface_t interface)
+{
+	struct dpaa2_mac *mac = phylink_to_dpaa2_mac(config);
+
+	return mac->pcs;
+}
+
+static void dpaa2_mac_config(struct phylink_config *config, unsigned int mode,
+			     const struct phylink_link_state *state)
+{
+	struct dpaa2_mac *mac = phylink_to_dpaa2_mac(config);
+	struct dpmac_link_state *dpmac_state = &mac->state;
+	int err;
+
+	if (linkmode_test_bit(ETHTOOL_LINK_MODE_Autoneg_BIT,
+			      state->advertising))
+		dpmac_state->options |= DPMAC_LINK_OPT_AUTONEG;
+	else
+		dpmac_state->options &= ~DPMAC_LINK_OPT_AUTONEG;
+
+	err = dpmac_set_link_state(mac->mc_io, 0,
+				   mac->mc_dev->mc_handle, dpmac_state);
+	if (err)
+		netdev_err(mac->net_dev, "%s: dpmac_set_link_state() = %d\n",
+			   __func__, err);
+
+	if (!mac->serdes_phy)
+		return;
+
+	/* This happens only if we support changing of protocol at runtime */
+	err = dpmac_set_protocol(mac->mc_io, 0, mac->mc_dev->mc_handle,
+				 dpmac_eth_if_mode(state->interface));
+	if (err)
+		netdev_err(mac->net_dev,  "dpmac_set_protocol() = %d\n", err);
+
+	err = phy_set_mode_ext(mac->serdes_phy, PHY_MODE_ETHERNET, state->interface);
+	if (err)
+		netdev_err(mac->net_dev, "phy_set_mode_ext() = %d\n", err);
+}
+
+static void dpaa2_mac_link_up(struct phylink_config *config,
+			      struct phy_device *phy,
+			      unsigned int mode, phy_interface_t interface,
+			      int speed, int duplex,
+			      bool tx_pause, bool rx_pause)
+{
+	struct dpaa2_mac *mac = phylink_to_dpaa2_mac(config);
+	struct dpmac_link_state *dpmac_state = &mac->state;
+	int err;
+
+	dpmac_state->up = 1;
+
+	dpmac_state->rate = speed;
+
+	if (duplex == DUPLEX_HALF)
+		dpmac_state->options |= DPMAC_LINK_OPT_HALF_DUPLEX;
+	else if (duplex == DUPLEX_FULL)
+		dpmac_state->options &= ~DPMAC_LINK_OPT_HALF_DUPLEX;
+
+	if (rx_pause)
+		dpmac_state->options |= DPMAC_LINK_OPT_PAUSE;
+	else
+		dpmac_state->options &= ~DPMAC_LINK_OPT_PAUSE;
+
+	if (rx_pause ^ tx_pause)
+		dpmac_state->options |= DPMAC_LINK_OPT_ASYM_PAUSE;
+	else
+		dpmac_state->options &= ~DPMAC_LINK_OPT_ASYM_PAUSE;
+
+	err = dpmac_set_link_state(mac->mc_io, 0,
+				   mac->mc_dev->mc_handle, dpmac_state);
+	if (err)
+		netdev_err(mac->net_dev, "%s: dpmac_set_link_state() = %d\n",
+			   __func__, err);
+}
+
+static void dpaa2_mac_link_down(struct phylink_config *config,
+				unsigned int mode,
+				phy_interface_t interface)
+{
+	struct dpaa2_mac *mac = phylink_to_dpaa2_mac(config);
+	struct dpmac_link_state *dpmac_state = &mac->state;
+	int err;
+
+	dpmac_state->up = 0;
+	err = dpmac_set_link_state(mac->mc_io, 0,
+				   mac->mc_dev->mc_handle, dpmac_state);
+	if (err)
+		netdev_err(mac->net_dev, "dpmac_set_link_state() = %d\n", err);
+}
+
+static const struct phylink_mac_ops dpaa2_mac_phylink_ops = {
+	.mac_select_pcs = dpaa2_mac_select_pcs,
+	.mac_config = dpaa2_mac_config,
+	.mac_link_up = dpaa2_mac_link_up,
+	.mac_link_down = dpaa2_mac_link_down,
+};
+
+static int dpaa2_pcs_create(struct dpaa2_mac *mac,
+			    struct fwnode_handle *dpmac_node,
+			    int id)
+{
+	struct fwnode_handle *node;
+	struct phylink_pcs *pcs;
+
+	node = fwnode_find_reference(dpmac_node, "pcs-handle", 0);
+	if (IS_ERR(node)) {
+		/* do not error out on old DTS files */
+		netdev_warn(mac->net_dev, "pcs-handle node not found\n");
+		return 0;
+	}
+
+	pcs = lynx_pcs_create_fwnode(node);
+	fwnode_handle_put(node);
+
+	if (pcs == ERR_PTR(-EPROBE_DEFER)) {
+		netdev_dbg(mac->net_dev, "missing PCS device\n");
+		return -EPROBE_DEFER;
+	}
+
+	if (pcs == ERR_PTR(-ENODEV)) {
+		netdev_err(mac->net_dev, "pcs-handle node not available\n");
+		return PTR_ERR(pcs);
+	}
+
+	if (IS_ERR(pcs)) {
+		netdev_err(mac->net_dev,
+			   "lynx_pcs_create_fwnode() failed: %pe\n", pcs);
+		return PTR_ERR(pcs);
+	}
+
+	mac->pcs = pcs;
+
+	return 0;
+}
+
+static void dpaa2_pcs_destroy(struct dpaa2_mac *mac)
+{
+	struct phylink_pcs *phylink_pcs = mac->pcs;
+
+	if (phylink_pcs) {
+		lynx_pcs_destroy(phylink_pcs);
+		mac->pcs = NULL;
+	}
+}
+
+static void dpaa2_mac_set_supported_interfaces(struct dpaa2_mac *mac)
+{
+	int intf, err;
+
+	/* We support the current interface mode, and if we have a PCS
+	 * similar interface modes that do not require the SerDes lane to be
+	 * reconfigured.
+	 */
+	__set_bit(mac->if_mode, mac->phylink_config.supported_interfaces);
+	if (mac->pcs) {
+		switch (mac->if_mode) {
+		case PHY_INTERFACE_MODE_1000BASEX:
+		case PHY_INTERFACE_MODE_SGMII:
+			__set_bit(PHY_INTERFACE_MODE_1000BASEX,
+				  mac->phylink_config.supported_interfaces);
+			__set_bit(PHY_INTERFACE_MODE_SGMII,
+				  mac->phylink_config.supported_interfaces);
+			break;
+
+		default:
+			break;
+		}
+	}
+
+	if (!mac->serdes_phy)
+		return;
+
+	/* In case we have access to the SerDes phy/lane, then ask the SerDes
+	 * driver what interfaces are supported based on the current PLL
+	 * configuration.
+	 */
+	for (intf = 0; intf < PHY_INTERFACE_MODE_MAX; intf++) {
+		if (intf == PHY_INTERFACE_MODE_NA)
+			continue;
+
+		err = phy_validate(mac->serdes_phy, PHY_MODE_ETHERNET, intf, NULL);
+		if (err)
+			continue;
+
+		__set_bit(intf, mac->phylink_config.supported_interfaces);
+	}
+}
+
+void dpaa2_mac_start(struct dpaa2_mac *mac)
+{
+	ASSERT_RTNL();
+
+	if (mac->serdes_phy)
+		phy_power_on(mac->serdes_phy);
+
+	phylink_start(mac->phylink);
+}
+
+void dpaa2_mac_stop(struct dpaa2_mac *mac)
+{
+	ASSERT_RTNL();
+
+	phylink_stop(mac->phylink);
+
+	if (mac->serdes_phy)
+		phy_power_off(mac->serdes_phy);
+}
+
+int dpaa2_mac_connect(struct dpaa2_mac *mac)
+{
+	struct net_device *net_dev = mac->net_dev;
+	struct fwnode_handle *dpmac_node;
+	struct phy *serdes_phy = NULL;
+	struct phylink *phylink;
+	int err;
+
+	mac->if_link_type = mac->attr.link_type;
+
+	dpmac_node = mac->fw_node;
+	if (!dpmac_node) {
+		netdev_err(net_dev, "No dpmac@%d node found.\n", mac->attr.id);
+		return -ENODEV;
+	}
+
+	err = dpaa2_mac_get_if_mode(dpmac_node, mac->attr);
+	if (err < 0)
+		return -EINVAL;
+	mac->if_mode = err;
+
+	if (mac->features & DPAA2_MAC_FEATURE_PROTOCOL_CHANGE &&
+	    !phy_interface_mode_is_rgmii(mac->if_mode) &&
+	    is_of_node(dpmac_node)) {
+		serdes_phy = of_phy_get(to_of_node(dpmac_node), NULL);
+
+		if (serdes_phy == ERR_PTR(-ENODEV))
+			serdes_phy = NULL;
+		else if (IS_ERR(serdes_phy))
+			return PTR_ERR(serdes_phy);
+		else
+			phy_init(serdes_phy);
+	}
+	mac->serdes_phy = serdes_phy;
+
+	/* The MAC does not have the capability to add RGMII delays so
+	 * error out if the interface mode requests them and there is no PHY
+	 * to act upon them
+	 */
+	if (of_phy_is_fixed_link(to_of_node(dpmac_node)) &&
+	    (mac->if_mode == PHY_INTERFACE_MODE_RGMII_ID ||
+	     mac->if_mode == PHY_INTERFACE_MODE_RGMII_RXID ||
+	     mac->if_mode == PHY_INTERFACE_MODE_RGMII_TXID)) {
+		netdev_err(net_dev, "RGMII delay not supported\n");
+		return -EINVAL;
+	}
+
+	if ((mac->attr.link_type == DPMAC_LINK_TYPE_PHY &&
+	     mac->attr.eth_if != DPMAC_ETH_IF_RGMII) ||
+	    mac->attr.link_type == DPMAC_LINK_TYPE_BACKPLANE) {
+		err = dpaa2_pcs_create(mac, dpmac_node, mac->attr.id);
+		if (err)
+			return err;
+	}
+
+	memset(&mac->phylink_config, 0, sizeof(mac->phylink_config));
+	mac->phylink_config.dev = &net_dev->dev;
+	mac->phylink_config.type = PHYLINK_NETDEV;
+
+	mac->phylink_config.mac_capabilities = MAC_SYM_PAUSE | MAC_ASYM_PAUSE |
+		MAC_10FD | MAC_100FD | MAC_1000FD | MAC_2500FD | MAC_5000FD |
+		MAC_10000FD | MAC_25000FD;
+
+	dpaa2_mac_set_supported_interfaces(mac);
+
+	phylink = phylink_create(&mac->phylink_config,
+				 dpmac_node, mac->if_mode,
+				 &dpaa2_mac_phylink_ops);
+	if (IS_ERR(phylink)) {
+		err = PTR_ERR(phylink);
+		goto err_pcs_destroy;
+	}
+	mac->phylink = phylink;
+
+	rtnl_lock();
+	err = phylink_fwnode_phy_connect(mac->phylink, dpmac_node, 0);
+	rtnl_unlock();
+	if (err) {
+		netdev_err(net_dev, "phylink_fwnode_phy_connect() = %d\n", err);
+		goto err_phylink_destroy;
+	}
+
+	return 0;
+
+err_phylink_destroy:
+	phylink_destroy(mac->phylink);
+err_pcs_destroy:
+	dpaa2_pcs_destroy(mac);
+
+	return err;
+}
+
+void dpaa2_mac_disconnect(struct dpaa2_mac *mac)
+{
+	rtnl_lock();
+	phylink_disconnect_phy(mac->phylink);
+	rtnl_unlock();
+
+	phylink_destroy(mac->phylink);
+	dpaa2_pcs_destroy(mac);
+	of_phy_put(mac->serdes_phy);
+	mac->serdes_phy = NULL;
+}
+
+int dpaa2_mac_open(struct dpaa2_mac *mac)
+{
+	struct fsl_mc_device *dpmac_dev = mac->mc_dev;
+	struct net_device *net_dev = mac->net_dev;
+	struct fwnode_handle *fw_node;
+	int err;
+
+	err = dpmac_open(mac->mc_io, 0, dpmac_dev->obj_desc.id,
+			 &dpmac_dev->mc_handle);
+	if (err || !dpmac_dev->mc_handle) {
+		netdev_err(net_dev, "dpmac_open() = %d\n", err);
+		return -ENODEV;
+	}
+
+	err = dpmac_get_attributes(mac->mc_io, 0, dpmac_dev->mc_handle,
+				   &mac->attr);
+	if (err) {
+		netdev_err(net_dev, "dpmac_get_attributes() = %d\n", err);
+		goto err_close_dpmac;
+	}
+
+	err = dpmac_get_api_version(mac->mc_io, 0, &mac->ver_major, &mac->ver_minor);
+	if (err) {
+		netdev_err(net_dev, "dpmac_get_api_version() = %d\n", err);
+		goto err_close_dpmac;
+	}
+
+	dpaa2_mac_detect_features(mac);
+
+	/* Find the device node representing the MAC device and link the device
+	 * behind the associated netdev to it.
+	 */
+	fw_node = dpaa2_mac_get_node(&mac->mc_dev->dev, mac->attr.id);
+	if (IS_ERR(fw_node)) {
+		err = PTR_ERR(fw_node);
+		goto err_close_dpmac;
+	}
+
+	mac->fw_node = fw_node;
+	net_dev->dev.of_node = to_of_node(mac->fw_node);
+
+	return 0;
+
+err_close_dpmac:
+	dpmac_close(mac->mc_io, 0, dpmac_dev->mc_handle);
+	return err;
+}
+
+void dpaa2_mac_close(struct dpaa2_mac *mac)
+{
+	struct fsl_mc_device *dpmac_dev = mac->mc_dev;
+
+	dpmac_close(mac->mc_io, 0, dpmac_dev->mc_handle);
+	if (mac->fw_node)
+		fwnode_handle_put(mac->fw_node);
+}
+
+static char dpaa2_mac_ethtool_stats[][ETH_GSTRING_LEN] = {
+	[DPMAC_CNT_ING_ALL_FRAME]		= "[mac] rx all frames",
+	[DPMAC_CNT_ING_GOOD_FRAME]		= "[mac] rx frames ok",
+	[DPMAC_CNT_ING_ERR_FRAME]		= "[mac] rx frame errors",
+	[DPMAC_CNT_ING_FRAME_DISCARD]		= "[mac] rx frame discards",
+	[DPMAC_CNT_ING_UCAST_FRAME]		= "[mac] rx u-cast",
+	[DPMAC_CNT_ING_BCAST_FRAME]		= "[mac] rx b-cast",
+	[DPMAC_CNT_ING_MCAST_FRAME]		= "[mac] rx m-cast",
+	[DPMAC_CNT_ING_FRAME_64]		= "[mac] rx 64 bytes",
+	[DPMAC_CNT_ING_FRAME_127]		= "[mac] rx 65-127 bytes",
+	[DPMAC_CNT_ING_FRAME_255]		= "[mac] rx 128-255 bytes",
+	[DPMAC_CNT_ING_FRAME_511]		= "[mac] rx 256-511 bytes",
+	[DPMAC_CNT_ING_FRAME_1023]		= "[mac] rx 512-1023 bytes",
+	[DPMAC_CNT_ING_FRAME_1518]		= "[mac] rx 1024-1518 bytes",
+	[DPMAC_CNT_ING_FRAME_1519_MAX]		= "[mac] rx 1519-max bytes",
+	[DPMAC_CNT_ING_FRAG]			= "[mac] rx frags",
+	[DPMAC_CNT_ING_JABBER]			= "[mac] rx jabber",
+	[DPMAC_CNT_ING_ALIGN_ERR]		= "[mac] rx align errors",
+	[DPMAC_CNT_ING_OVERSIZED]		= "[mac] rx oversized",
+	[DPMAC_CNT_ING_VALID_PAUSE_FRAME]	= "[mac] rx pause",
+	[DPMAC_CNT_ING_BYTE]			= "[mac] rx bytes",
+	[DPMAC_CNT_EGR_GOOD_FRAME]		= "[mac] tx frames ok",
+	[DPMAC_CNT_EGR_UCAST_FRAME]		= "[mac] tx u-cast",
+	[DPMAC_CNT_EGR_MCAST_FRAME]		= "[mac] tx m-cast",
+	[DPMAC_CNT_EGR_BCAST_FRAME]		= "[mac] tx b-cast",
+	[DPMAC_CNT_EGR_ERR_FRAME]		= "[mac] tx frame errors",
+	[DPMAC_CNT_EGR_UNDERSIZED]		= "[mac] tx undersized",
+	[DPMAC_CNT_EGR_VALID_PAUSE_FRAME]	= "[mac] tx b-pause",
+	[DPMAC_CNT_EGR_BYTE]			= "[mac] tx bytes",
+};
+
+#define DPAA2_MAC_NUM_STATS	ARRAY_SIZE(dpaa2_mac_ethtool_stats)
+
+int dpaa2_mac_get_sset_count(void)
+{
+	return DPAA2_MAC_NUM_STATS;
+}
+
+void dpaa2_mac_get_strings(u8 **data)
+{
+	int i;
+
+	for (i = 0; i < DPAA2_MAC_NUM_STATS; i++)
+		ethtool_puts(data, dpaa2_mac_ethtool_stats[i]);
+}
+
+void dpaa2_mac_get_ethtool_stats(struct dpaa2_mac *mac, u64 *data)
+{
+	struct fsl_mc_device *dpmac_dev = mac->mc_dev;
+	int i, err;
+	u64 value;
+
+	for (i = 0; i < DPAA2_MAC_NUM_STATS; i++) {
+		err = dpmac_get_counter(mac->mc_io, 0, dpmac_dev->mc_handle,
+					i, &value);
+		if (err) {
+			netdev_err_once(mac->net_dev,
+					"dpmac_get_counter error %d\n", err);
+			*(data + i) = U64_MAX;
+			continue;
+		}
+		*(data + i) = value;
+	}
+}
diff --git a/drivers/net/ethernet/freescale/dpaa2/dpaa2-mac.h b/drivers/net/ethernet/freescale/dpaa2/dpaa2-mac.h
new file mode 100644
index 000000000000..53f8d106d11e
--- /dev/null
+++ b/drivers/net/ethernet/freescale/dpaa2/dpaa2-mac.h
@@ -0,0 +1,60 @@
+/* SPDX-License-Identifier: (GPL-2.0+ OR BSD-3-Clause) */
+/* Copyright 2019 NXP */
+#ifndef DPAA2_MAC_H
+#define DPAA2_MAC_H
+
+#include <linux/of.h>
+#include <linux/of_mdio.h>
+#include <linux/of_net.h>
+#include <linux/phylink.h>
+
+#include "dpmac.h"
+#include "dpmac-cmd.h"
+
+struct dpaa2_mac {
+	struct fsl_mc_device *mc_dev;
+	struct dpmac_link_state state;
+	struct net_device *net_dev;
+	struct fsl_mc_io *mc_io;
+	struct dpmac_attr attr;
+	u16 ver_major, ver_minor;
+	unsigned long features;
+
+	struct phylink_config phylink_config;
+	struct phylink *phylink;
+	phy_interface_t if_mode;
+	enum dpmac_link_type if_link_type;
+	struct phylink_pcs *pcs;
+	struct fwnode_handle *fw_node;
+
+	struct phy *serdes_phy;
+};
+
+static inline bool dpaa2_mac_is_type_phy(struct dpaa2_mac *mac)
+{
+	if (!mac)
+		return false;
+
+	return mac->attr.link_type == DPMAC_LINK_TYPE_PHY ||
+	       mac->attr.link_type == DPMAC_LINK_TYPE_BACKPLANE;
+}
+
+int dpaa2_mac_open(struct dpaa2_mac *mac);
+
+void dpaa2_mac_close(struct dpaa2_mac *mac);
+
+int dpaa2_mac_connect(struct dpaa2_mac *mac);
+
+void dpaa2_mac_disconnect(struct dpaa2_mac *mac);
+
+int dpaa2_mac_get_sset_count(void);
+
+void dpaa2_mac_get_strings(u8 **data);
+
+void dpaa2_mac_get_ethtool_stats(struct dpaa2_mac *mac, u64 *data);
+
+void dpaa2_mac_start(struct dpaa2_mac *mac);
+
+void dpaa2_mac_stop(struct dpaa2_mac *mac);
+
+#endif /* DPAA2_MAC_H */
diff --git a/drivers/net/ethernet/freescale/dpaa2/dpaa2-ptp.c b/drivers/net/ethernet/freescale/dpaa2/dpaa2-ptp.c
new file mode 100644
index 000000000000..4497e3c0456d
--- /dev/null
+++ b/drivers/net/ethernet/freescale/dpaa2/dpaa2-ptp.c
@@ -0,0 +1,259 @@
+// SPDX-License-Identifier: GPL-2.0
+/*
+ * Copyright 2013-2016 Freescale Semiconductor Inc.
+ * Copyright 2016-2018 NXP
+ * Copyright 2020 NXP
+ */
+
+#include <linux/module.h>
+#include <linux/of.h>
+#include <linux/of_address.h>
+#include <linux/fsl/mc.h>
+
+#include "dpaa2-ptp.h"
+
+static int dpaa2_ptp_enable(struct ptp_clock_info *ptp,
+			    struct ptp_clock_request *rq, int on)
+{
+	struct ptp_qoriq *ptp_qoriq = container_of(ptp, struct ptp_qoriq, caps);
+	struct fsl_mc_device *mc_dev;
+	struct device *dev;
+	u32 mask = 0;
+	u32 bit;
+	int err;
+
+	dev = ptp_qoriq->dev;
+	mc_dev = to_fsl_mc_device(dev);
+
+	switch (rq->type) {
+	case PTP_CLK_REQ_EXTTS:
+		switch (rq->extts.index) {
+		case 0:
+			bit = DPRTC_EVENT_ETS1;
+			break;
+		case 1:
+			bit = DPRTC_EVENT_ETS2;
+			break;
+		default:
+			return -EINVAL;
+		}
+		if (on)
+			extts_clean_up(ptp_qoriq, rq->extts.index, false);
+		break;
+	case PTP_CLK_REQ_PPS:
+		bit = DPRTC_EVENT_PPS;
+		break;
+	default:
+		return -EOPNOTSUPP;
+	}
+
+	err = dprtc_get_irq_mask(mc_dev->mc_io, 0, mc_dev->mc_handle,
+				 DPRTC_IRQ_INDEX, &mask);
+	if (err < 0) {
+		dev_err(dev, "dprtc_get_irq_mask(): %d\n", err);
+		return err;
+	}
+
+	if (on)
+		mask |= bit;
+	else
+		mask &= ~bit;
+
+	err = dprtc_set_irq_mask(mc_dev->mc_io, 0, mc_dev->mc_handle,
+				 DPRTC_IRQ_INDEX, mask);
+	if (err < 0) {
+		dev_err(dev, "dprtc_set_irq_mask(): %d\n", err);
+		return err;
+	}
+
+	return 0;
+}
+
+static const struct ptp_clock_info dpaa2_ptp_caps = {
+	.owner		= THIS_MODULE,
+	.name		= "DPAA2 PTP Clock",
+	.max_adj	= 512000,
+	.n_alarm	= 2,
+	.n_ext_ts	= 2,
+	.n_per_out	= 3,
+	.n_pins		= 0,
+	.pps		= 1,
+	.adjfine	= ptp_qoriq_adjfine,
+	.adjtime	= ptp_qoriq_adjtime,
+	.gettime64	= ptp_qoriq_gettime,
+	.settime64	= ptp_qoriq_settime,
+	.enable		= dpaa2_ptp_enable,
+};
+
+static irqreturn_t dpaa2_ptp_irq_handler_thread(int irq, void *priv)
+{
+	struct ptp_qoriq *ptp_qoriq = priv;
+	struct ptp_clock_event event;
+	struct fsl_mc_device *mc_dev;
+	struct device *dev;
+	u32 status = 0;
+	int err;
+
+	dev = ptp_qoriq->dev;
+	mc_dev = to_fsl_mc_device(dev);
+
+	err = dprtc_get_irq_status(mc_dev->mc_io, 0, mc_dev->mc_handle,
+				   DPRTC_IRQ_INDEX, &status);
+	if (unlikely(err)) {
+		dev_err(dev, "dprtc_get_irq_status err %d\n", err);
+		return IRQ_NONE;
+	}
+
+	if (status & DPRTC_EVENT_PPS) {
+		event.type = PTP_CLOCK_PPS;
+		ptp_clock_event(ptp_qoriq->clock, &event);
+	}
+
+	if (status & DPRTC_EVENT_ETS1)
+		extts_clean_up(ptp_qoriq, 0, true);
+
+	if (status & DPRTC_EVENT_ETS2)
+		extts_clean_up(ptp_qoriq, 1, true);
+
+	err = dprtc_clear_irq_status(mc_dev->mc_io, 0, mc_dev->mc_handle,
+				     DPRTC_IRQ_INDEX, status);
+	if (unlikely(err)) {
+		dev_err(dev, "dprtc_clear_irq_status err %d\n", err);
+		return IRQ_NONE;
+	}
+
+	return IRQ_HANDLED;
+}
+
+static int dpaa2_ptp_probe(struct fsl_mc_device *mc_dev)
+{
+	struct device *dev = &mc_dev->dev;
+	struct ptp_qoriq *ptp_qoriq;
+	struct device_node *node;
+	void __iomem *base;
+	int err;
+
+	ptp_qoriq = devm_kzalloc(dev, sizeof(*ptp_qoriq), GFP_KERNEL);
+	if (!ptp_qoriq)
+		return -ENOMEM;
+
+	err = fsl_mc_portal_allocate(mc_dev, 0, &mc_dev->mc_io);
+	if (err) {
+		if (err == -ENXIO)
+			err = -EPROBE_DEFER;
+		else
+			dev_err(dev, "fsl_mc_portal_allocate err %d\n", err);
+		goto err_exit;
+	}
+
+	err = dprtc_open(mc_dev->mc_io, 0, mc_dev->obj_desc.id,
+			 &mc_dev->mc_handle);
+	if (err) {
+		dev_err(dev, "dprtc_open err %d\n", err);
+		goto err_free_mcp;
+	}
+
+	ptp_qoriq->dev = dev;
+
+	node = of_find_compatible_node(NULL, NULL, "fsl,dpaa2-ptp");
+	if (!node) {
+		err = -ENODEV;
+		goto err_close;
+	}
+
+	dev->of_node = node;
+
+	base = of_iomap(node, 0);
+	if (!base) {
+		err = -ENOMEM;
+		goto err_put;
+	}
+
+	err = fsl_mc_allocate_irqs(mc_dev);
+	if (err) {
+		dev_err(dev, "MC irqs allocation failed\n");
+		goto err_unmap;
+	}
+
+	ptp_qoriq->irq = mc_dev->irqs[0]->virq;
+
+	err = request_threaded_irq(ptp_qoriq->irq, NULL,
+				   dpaa2_ptp_irq_handler_thread,
+				   IRQF_NO_SUSPEND | IRQF_ONESHOT,
+				   dev_name(dev), ptp_qoriq);
+	if (err < 0) {
+		dev_err(dev, "devm_request_threaded_irq(): %d\n", err);
+		goto err_free_mc_irq;
+	}
+
+	err = dprtc_set_irq_enable(mc_dev->mc_io, 0, mc_dev->mc_handle,
+				   DPRTC_IRQ_INDEX, 1);
+	if (err < 0) {
+		dev_err(dev, "dprtc_set_irq_enable(): %d\n", err);
+		goto err_free_threaded_irq;
+	}
+
+	err = ptp_qoriq_init(ptp_qoriq, base, &dpaa2_ptp_caps);
+	if (err)
+		goto err_free_threaded_irq;
+
+	dpaa2_phc_index = ptp_qoriq->phc_index;
+	dpaa2_ptp = ptp_qoriq;
+	dev_set_drvdata(dev, ptp_qoriq);
+
+	return 0;
+
+err_free_threaded_irq:
+	free_irq(ptp_qoriq->irq, ptp_qoriq);
+err_free_mc_irq:
+	fsl_mc_free_irqs(mc_dev);
+err_unmap:
+	iounmap(base);
+err_put:
+	of_node_put(node);
+err_close:
+	dprtc_close(mc_dev->mc_io, 0, mc_dev->mc_handle);
+err_free_mcp:
+	fsl_mc_portal_free(mc_dev->mc_io);
+err_exit:
+	return err;
+}
+
+static void dpaa2_ptp_remove(struct fsl_mc_device *mc_dev)
+{
+	struct device *dev = &mc_dev->dev;
+	struct ptp_qoriq *ptp_qoriq;
+
+	ptp_qoriq = dev_get_drvdata(dev);
+
+	dpaa2_phc_index = -1;
+	ptp_qoriq_free(ptp_qoriq);
+
+	fsl_mc_free_irqs(mc_dev);
+	dprtc_close(mc_dev->mc_io, 0, mc_dev->mc_handle);
+	fsl_mc_portal_free(mc_dev->mc_io);
+}
+
+static const struct fsl_mc_device_id dpaa2_ptp_match_id_table[] = {
+	{
+		.vendor = FSL_MC_VENDOR_FREESCALE,
+		.obj_type = "dprtc",
+	},
+	{}
+};
+MODULE_DEVICE_TABLE(fslmc, dpaa2_ptp_match_id_table);
+
+static struct fsl_mc_driver dpaa2_ptp_drv = {
+	.driver = {
+		.name = KBUILD_MODNAME,
+		.owner = THIS_MODULE,
+	},
+	.probe = dpaa2_ptp_probe,
+	.remove = dpaa2_ptp_remove,
+	.match_id_table = dpaa2_ptp_match_id_table,
+};
+
+module_fsl_mc_driver(dpaa2_ptp_drv);
+
+MODULE_LICENSE("GPL v2");
+MODULE_DESCRIPTION("DPAA2 PTP Clock Driver");
diff --git a/drivers/net/ethernet/freescale/dpaa2/dpaa2-ptp.h b/drivers/net/ethernet/freescale/dpaa2/dpaa2-ptp.h
new file mode 100644
index 000000000000..e1023538b4c3
--- /dev/null
+++ b/drivers/net/ethernet/freescale/dpaa2/dpaa2-ptp.h
@@ -0,0 +1,18 @@
+/* SPDX-License-Identifier: GPL-2.0 */
+/*
+ * Copyright 2018 NXP
+ * Copyright 2020 NXP
+ */
+
+#ifndef __RTC_H
+#define __RTC_H
+
+#include <linux/fsl/ptp_qoriq.h>
+
+#include "dprtc.h"
+#include "dprtc-cmd.h"
+
+extern int dpaa2_phc_index;
+extern struct ptp_qoriq *dpaa2_ptp;
+
+#endif
diff --git a/drivers/net/ethernet/freescale/dpaa2/dpaa2-switch-ethtool.c b/drivers/net/ethernet/freescale/dpaa2/dpaa2-switch-ethtool.c
new file mode 100644
index 000000000000..a888f6e6e9b0
--- /dev/null
+++ b/drivers/net/ethernet/freescale/dpaa2/dpaa2-switch-ethtool.c
@@ -0,0 +1,221 @@
+// SPDX-License-Identifier: GPL-2.0
+/*
+ * DPAA2 Ethernet Switch ethtool support
+ *
+ * Copyright 2014-2016 Freescale Semiconductor Inc.
+ * Copyright 2017-2018 NXP
+ *
+ */
+
+#include <linux/ethtool.h>
+
+#include "dpaa2-switch.h"
+
+static struct {
+	enum dpsw_counter id;
+	char name[ETH_GSTRING_LEN];
+} dpaa2_switch_ethtool_counters[] =  {
+	{DPSW_CNT_ING_FRAME,		"[hw] rx frames"},
+	{DPSW_CNT_ING_BYTE,		"[hw] rx bytes"},
+	{DPSW_CNT_ING_FLTR_FRAME,	"[hw] rx filtered frames"},
+	{DPSW_CNT_ING_FRAME_DISCARD,	"[hw] rx discarded frames"},
+	{DPSW_CNT_ING_BCAST_FRAME,	"[hw] rx bcast frames"},
+	{DPSW_CNT_ING_BCAST_BYTES,	"[hw] rx bcast bytes"},
+	{DPSW_CNT_ING_MCAST_FRAME,	"[hw] rx mcast frames"},
+	{DPSW_CNT_ING_MCAST_BYTE,	"[hw] rx mcast bytes"},
+	{DPSW_CNT_EGR_FRAME,		"[hw] tx frames"},
+	{DPSW_CNT_EGR_BYTE,		"[hw] tx bytes"},
+	{DPSW_CNT_EGR_FRAME_DISCARD,	"[hw] tx discarded frames"},
+	{DPSW_CNT_ING_NO_BUFF_DISCARD,	"[hw] rx nobuffer discards"},
+};
+
+#define DPAA2_SWITCH_NUM_COUNTERS	ARRAY_SIZE(dpaa2_switch_ethtool_counters)
+
+static void dpaa2_switch_get_drvinfo(struct net_device *netdev,
+				     struct ethtool_drvinfo *drvinfo)
+{
+	struct ethsw_port_priv *port_priv = netdev_priv(netdev);
+	u16 version_major, version_minor;
+	int err;
+
+	strscpy(drvinfo->driver, KBUILD_MODNAME, sizeof(drvinfo->driver));
+
+	err = dpsw_get_api_version(port_priv->ethsw_data->mc_io, 0,
+				   &version_major,
+				   &version_minor);
+	if (err)
+		strscpy(drvinfo->fw_version, "N/A",
+			sizeof(drvinfo->fw_version));
+	else
+		snprintf(drvinfo->fw_version, sizeof(drvinfo->fw_version),
+			 "%u.%u", version_major, version_minor);
+
+	strscpy(drvinfo->bus_info, dev_name(netdev->dev.parent->parent),
+		sizeof(drvinfo->bus_info));
+}
+
+static int
+dpaa2_switch_get_link_ksettings(struct net_device *netdev,
+				struct ethtool_link_ksettings *link_ksettings)
+{
+	struct ethsw_port_priv *port_priv = netdev_priv(netdev);
+	struct dpsw_link_state state = {0};
+	int err;
+
+	mutex_lock(&port_priv->mac_lock);
+
+	if (dpaa2_switch_port_is_type_phy(port_priv)) {
+		err = phylink_ethtool_ksettings_get(port_priv->mac->phylink,
+						    link_ksettings);
+		mutex_unlock(&port_priv->mac_lock);
+		return err;
+	}
+
+	mutex_unlock(&port_priv->mac_lock);
+
+	err = dpsw_if_get_link_state(port_priv->ethsw_data->mc_io, 0,
+				     port_priv->ethsw_data->dpsw_handle,
+				     port_priv->idx,
+				     &state);
+	if (err) {
+		netdev_err(netdev, "ERROR %d getting link state\n", err);
+		goto out;
+	}
+
+	/* At the moment, we have no way of interrogating the DPMAC
+	 * from the DPSW side or there may not exist a DPMAC at all.
+	 * Report only autoneg state, duplexity and speed.
+	 */
+	if (state.options & DPSW_LINK_OPT_AUTONEG)
+		link_ksettings->base.autoneg = AUTONEG_ENABLE;
+	if (!(state.options & DPSW_LINK_OPT_HALF_DUPLEX))
+		link_ksettings->base.duplex = DUPLEX_FULL;
+	link_ksettings->base.speed = state.rate;
+
+out:
+	return err;
+}
+
+static int
+dpaa2_switch_set_link_ksettings(struct net_device *netdev,
+				const struct ethtool_link_ksettings *link_ksettings)
+{
+	struct ethsw_port_priv *port_priv = netdev_priv(netdev);
+	struct ethsw_core *ethsw = port_priv->ethsw_data;
+	struct dpsw_link_cfg cfg = {0};
+	bool if_running;
+	int err = 0, ret;
+
+	mutex_lock(&port_priv->mac_lock);
+
+	if (dpaa2_switch_port_is_type_phy(port_priv)) {
+		err = phylink_ethtool_ksettings_set(port_priv->mac->phylink,
+						    link_ksettings);
+		mutex_unlock(&port_priv->mac_lock);
+		return err;
+	}
+
+	mutex_unlock(&port_priv->mac_lock);
+
+	/* Interface needs to be down to change link settings */
+	if_running = netif_running(netdev);
+	if (if_running) {
+		err = dpsw_if_disable(ethsw->mc_io, 0,
+				      ethsw->dpsw_handle,
+				      port_priv->idx);
+		if (err) {
+			netdev_err(netdev, "dpsw_if_disable err %d\n", err);
+			return err;
+		}
+	}
+
+	cfg.rate = link_ksettings->base.speed;
+	if (link_ksettings->base.autoneg == AUTONEG_ENABLE)
+		cfg.options |= DPSW_LINK_OPT_AUTONEG;
+	else
+		cfg.options &= ~DPSW_LINK_OPT_AUTONEG;
+	if (link_ksettings->base.duplex  == DUPLEX_HALF)
+		cfg.options |= DPSW_LINK_OPT_HALF_DUPLEX;
+	else
+		cfg.options &= ~DPSW_LINK_OPT_HALF_DUPLEX;
+
+	err = dpsw_if_set_link_cfg(port_priv->ethsw_data->mc_io, 0,
+				   port_priv->ethsw_data->dpsw_handle,
+				   port_priv->idx,
+				   &cfg);
+
+	if (if_running) {
+		ret = dpsw_if_enable(ethsw->mc_io, 0,
+				     ethsw->dpsw_handle,
+				     port_priv->idx);
+		if (ret) {
+			netdev_err(netdev, "dpsw_if_enable err %d\n", ret);
+			return ret;
+		}
+	}
+	return err;
+}
+
+static int
+dpaa2_switch_ethtool_get_sset_count(struct net_device *netdev, int sset)
+{
+	switch (sset) {
+	case ETH_SS_STATS:
+		return DPAA2_SWITCH_NUM_COUNTERS + dpaa2_mac_get_sset_count();
+	default:
+		return -EOPNOTSUPP;
+	}
+}
+
+static void dpaa2_switch_ethtool_get_strings(struct net_device *netdev,
+					     u32 stringset, u8 *data)
+{
+	const char *str;
+	int i;
+
+	switch (stringset) {
+	case ETH_SS_STATS:
+		for (i = 0; i < DPAA2_SWITCH_NUM_COUNTERS; i++) {
+			str = dpaa2_switch_ethtool_counters[i].name;
+			ethtool_puts(&data, str);
+		}
+		dpaa2_mac_get_strings(&data);
+		break;
+	}
+}
+
+static void dpaa2_switch_ethtool_get_stats(struct net_device *netdev,
+					   struct ethtool_stats *stats,
+					   u64 *data)
+{
+	struct ethsw_port_priv *port_priv = netdev_priv(netdev);
+	int i, err;
+
+	for (i = 0; i < DPAA2_SWITCH_NUM_COUNTERS; i++) {
+		err = dpsw_if_get_counter(port_priv->ethsw_data->mc_io, 0,
+					  port_priv->ethsw_data->dpsw_handle,
+					  port_priv->idx,
+					  dpaa2_switch_ethtool_counters[i].id,
+					  &data[i]);
+		if (err)
+			netdev_err(netdev, "dpsw_if_get_counter[%s] err %d\n",
+				   dpaa2_switch_ethtool_counters[i].name, err);
+	}
+
+	mutex_lock(&port_priv->mac_lock);
+
+	if (dpaa2_switch_port_has_mac(port_priv))
+		dpaa2_mac_get_ethtool_stats(port_priv->mac, data + i);
+
+	mutex_unlock(&port_priv->mac_lock);
+}
+
+const struct ethtool_ops dpaa2_switch_port_ethtool_ops = {
+	.get_drvinfo		= dpaa2_switch_get_drvinfo,
+	.get_link		= ethtool_op_get_link,
+	.get_link_ksettings	= dpaa2_switch_get_link_ksettings,
+	.set_link_ksettings	= dpaa2_switch_set_link_ksettings,
+	.get_strings		= dpaa2_switch_ethtool_get_strings,
+	.get_ethtool_stats	= dpaa2_switch_ethtool_get_stats,
+	.get_sset_count		= dpaa2_switch_ethtool_get_sset_count,
+};
diff --git a/drivers/net/ethernet/freescale/dpaa2/dpaa2-switch-flower.c b/drivers/net/ethernet/freescale/dpaa2/dpaa2-switch-flower.c
new file mode 100644
index 000000000000..701a87370737
--- /dev/null
+++ b/drivers/net/ethernet/freescale/dpaa2/dpaa2-switch-flower.c
@@ -0,0 +1,891 @@
+// SPDX-License-Identifier: GPL-2.0
+/*
+ * DPAA2 Ethernet Switch flower support
+ *
+ * Copyright 2021 NXP
+ *
+ */
+
+#include "dpaa2-switch.h"
+
+static int dpaa2_switch_flower_parse_key(struct flow_cls_offload *cls,
+					 struct dpsw_acl_key *acl_key)
+{
+	struct flow_rule *rule = flow_cls_offload_flow_rule(cls);
+	struct flow_dissector *dissector = rule->match.dissector;
+	struct netlink_ext_ack *extack = cls->common.extack;
+	struct dpsw_acl_fields *acl_h, *acl_m;
+
+	if (dissector->used_keys &
+	    ~(BIT_ULL(FLOW_DISSECTOR_KEY_BASIC) |
+	      BIT_ULL(FLOW_DISSECTOR_KEY_CONTROL) |
+	      BIT_ULL(FLOW_DISSECTOR_KEY_ETH_ADDRS) |
+	      BIT_ULL(FLOW_DISSECTOR_KEY_VLAN) |
+	      BIT_ULL(FLOW_DISSECTOR_KEY_PORTS) |
+	      BIT_ULL(FLOW_DISSECTOR_KEY_IP) |
+	      BIT_ULL(FLOW_DISSECTOR_KEY_IPV6_ADDRS) |
+	      BIT_ULL(FLOW_DISSECTOR_KEY_IPV4_ADDRS))) {
+		NL_SET_ERR_MSG_MOD(extack,
+				   "Unsupported keys used");
+		return -EOPNOTSUPP;
+	}
+
+	acl_h = &acl_key->match;
+	acl_m = &acl_key->mask;
+
+	if (flow_rule_match_has_control_flags(rule, extack))
+		return -EOPNOTSUPP;
+
+	if (flow_rule_match_key(rule, FLOW_DISSECTOR_KEY_BASIC)) {
+		struct flow_match_basic match;
+
+		flow_rule_match_basic(rule, &match);
+		acl_h->l3_protocol = match.key->ip_proto;
+		acl_h->l2_ether_type = be16_to_cpu(match.key->n_proto);
+		acl_m->l3_protocol = match.mask->ip_proto;
+		acl_m->l2_ether_type = be16_to_cpu(match.mask->n_proto);
+	}
+
+	if (flow_rule_match_key(rule, FLOW_DISSECTOR_KEY_ETH_ADDRS)) {
+		struct flow_match_eth_addrs match;
+
+		flow_rule_match_eth_addrs(rule, &match);
+		ether_addr_copy(acl_h->l2_dest_mac, &match.key->dst[0]);
+		ether_addr_copy(acl_h->l2_source_mac, &match.key->src[0]);
+		ether_addr_copy(acl_m->l2_dest_mac, &match.mask->dst[0]);
+		ether_addr_copy(acl_m->l2_source_mac, &match.mask->src[0]);
+	}
+
+	if (flow_rule_match_key(rule, FLOW_DISSECTOR_KEY_VLAN)) {
+		struct flow_match_vlan match;
+
+		flow_rule_match_vlan(rule, &match);
+		acl_h->l2_vlan_id = match.key->vlan_id;
+		acl_h->l2_tpid = be16_to_cpu(match.key->vlan_tpid);
+		acl_h->l2_pcp_dei = match.key->vlan_priority << 1 |
+				    match.key->vlan_dei;
+
+		acl_m->l2_vlan_id = match.mask->vlan_id;
+		acl_m->l2_tpid = be16_to_cpu(match.mask->vlan_tpid);
+		acl_m->l2_pcp_dei = match.mask->vlan_priority << 1 |
+				    match.mask->vlan_dei;
+	}
+
+	if (flow_rule_match_key(rule, FLOW_DISSECTOR_KEY_IPV4_ADDRS)) {
+		struct flow_match_ipv4_addrs match;
+
+		flow_rule_match_ipv4_addrs(rule, &match);
+		acl_h->l3_source_ip = be32_to_cpu(match.key->src);
+		acl_h->l3_dest_ip = be32_to_cpu(match.key->dst);
+		acl_m->l3_source_ip = be32_to_cpu(match.mask->src);
+		acl_m->l3_dest_ip = be32_to_cpu(match.mask->dst);
+	}
+
+	if (flow_rule_match_key(rule, FLOW_DISSECTOR_KEY_PORTS)) {
+		struct flow_match_ports match;
+
+		flow_rule_match_ports(rule, &match);
+		acl_h->l4_source_port = be16_to_cpu(match.key->src);
+		acl_h->l4_dest_port = be16_to_cpu(match.key->dst);
+		acl_m->l4_source_port = be16_to_cpu(match.mask->src);
+		acl_m->l4_dest_port = be16_to_cpu(match.mask->dst);
+	}
+
+	if (flow_rule_match_key(rule, FLOW_DISSECTOR_KEY_IP)) {
+		struct flow_match_ip match;
+
+		flow_rule_match_ip(rule, &match);
+		if (match.mask->ttl != 0) {
+			NL_SET_ERR_MSG_MOD(extack,
+					   "Matching on TTL not supported");
+			return -EOPNOTSUPP;
+		}
+
+		if ((match.mask->tos & 0x3) != 0) {
+			NL_SET_ERR_MSG_MOD(extack,
+					   "Matching on ECN not supported, only DSCP");
+			return -EOPNOTSUPP;
+		}
+
+		acl_h->l3_dscp = match.key->tos >> 2;
+		acl_m->l3_dscp = match.mask->tos >> 2;
+	}
+
+	return 0;
+}
+
+int dpaa2_switch_acl_entry_add(struct dpaa2_switch_filter_block *filter_block,
+			       struct dpaa2_switch_acl_entry *entry)
+{
+	struct dpsw_acl_entry_cfg *acl_entry_cfg = &entry->cfg;
+	struct ethsw_core *ethsw = filter_block->ethsw;
+	struct dpsw_acl_key *acl_key = &entry->key;
+	struct device *dev = ethsw->dev;
+	u8 *cmd_buff;
+	int err;
+
+	cmd_buff = kzalloc(DPAA2_ETHSW_PORT_ACL_CMD_BUF_SIZE, GFP_KERNEL);
+	if (!cmd_buff)
+		return -ENOMEM;
+
+	dpsw_acl_prepare_entry_cfg(acl_key, cmd_buff);
+
+	acl_entry_cfg->key_iova = dma_map_single(dev, cmd_buff,
+						 DPAA2_ETHSW_PORT_ACL_CMD_BUF_SIZE,
+						 DMA_TO_DEVICE);
+	if (unlikely(dma_mapping_error(dev, acl_entry_cfg->key_iova))) {
+		dev_err(dev, "DMA mapping failed\n");
+		kfree(cmd_buff);
+		return -EFAULT;
+	}
+
+	err = dpsw_acl_add_entry(ethsw->mc_io, 0, ethsw->dpsw_handle,
+				 filter_block->acl_id, acl_entry_cfg);
+
+	dma_unmap_single(dev, acl_entry_cfg->key_iova,
+			 DPAA2_ETHSW_PORT_ACL_CMD_BUF_SIZE,
+			 DMA_TO_DEVICE);
+	if (err) {
+		dev_err(dev, "dpsw_acl_add_entry() failed %d\n", err);
+		kfree(cmd_buff);
+		return err;
+	}
+
+	kfree(cmd_buff);
+
+	return 0;
+}
+
+static int
+dpaa2_switch_acl_entry_remove(struct dpaa2_switch_filter_block *block,
+			      struct dpaa2_switch_acl_entry *entry)
+{
+	struct dpsw_acl_entry_cfg *acl_entry_cfg = &entry->cfg;
+	struct dpsw_acl_key *acl_key = &entry->key;
+	struct ethsw_core *ethsw = block->ethsw;
+	struct device *dev = ethsw->dev;
+	u8 *cmd_buff;
+	int err;
+
+	cmd_buff = kzalloc(DPAA2_ETHSW_PORT_ACL_CMD_BUF_SIZE, GFP_KERNEL);
+	if (!cmd_buff)
+		return -ENOMEM;
+
+	dpsw_acl_prepare_entry_cfg(acl_key, cmd_buff);
+
+	acl_entry_cfg->key_iova = dma_map_single(dev, cmd_buff,
+						 DPAA2_ETHSW_PORT_ACL_CMD_BUF_SIZE,
+						 DMA_TO_DEVICE);
+	if (unlikely(dma_mapping_error(dev, acl_entry_cfg->key_iova))) {
+		dev_err(dev, "DMA mapping failed\n");
+		kfree(cmd_buff);
+		return -EFAULT;
+	}
+
+	err = dpsw_acl_remove_entry(ethsw->mc_io, 0, ethsw->dpsw_handle,
+				    block->acl_id, acl_entry_cfg);
+
+	dma_unmap_single(dev, acl_entry_cfg->key_iova,
+			 DPAA2_ETHSW_PORT_ACL_CMD_BUF_SIZE, DMA_TO_DEVICE);
+	if (err) {
+		dev_err(dev, "dpsw_acl_remove_entry() failed %d\n", err);
+		kfree(cmd_buff);
+		return err;
+	}
+
+	kfree(cmd_buff);
+
+	return 0;
+}
+
+static int
+dpaa2_switch_acl_entry_add_to_list(struct dpaa2_switch_filter_block *block,
+				   struct dpaa2_switch_acl_entry *entry)
+{
+	struct dpaa2_switch_acl_entry *tmp;
+	struct list_head *pos, *n;
+	int index = 0;
+
+	if (list_empty(&block->acl_entries)) {
+		list_add(&entry->list, &block->acl_entries);
+		return index;
+	}
+
+	list_for_each_safe(pos, n, &block->acl_entries) {
+		tmp = list_entry(pos, struct dpaa2_switch_acl_entry, list);
+		if (entry->prio < tmp->prio)
+			break;
+		index++;
+	}
+	list_add(&entry->list, pos->prev);
+	return index;
+}
+
+static struct dpaa2_switch_acl_entry*
+dpaa2_switch_acl_entry_get_by_index(struct dpaa2_switch_filter_block *block,
+				    int index)
+{
+	struct dpaa2_switch_acl_entry *tmp;
+	int i = 0;
+
+	list_for_each_entry(tmp, &block->acl_entries, list) {
+		if (i == index)
+			return tmp;
+		++i;
+	}
+
+	return NULL;
+}
+
+static int
+dpaa2_switch_acl_entry_set_precedence(struct dpaa2_switch_filter_block *block,
+				      struct dpaa2_switch_acl_entry *entry,
+				      int precedence)
+{
+	int err;
+
+	err = dpaa2_switch_acl_entry_remove(block, entry);
+	if (err)
+		return err;
+
+	entry->cfg.precedence = precedence;
+	return dpaa2_switch_acl_entry_add(block, entry);
+}
+
+static int
+dpaa2_switch_acl_tbl_add_entry(struct dpaa2_switch_filter_block *block,
+			       struct dpaa2_switch_acl_entry *entry)
+{
+	struct dpaa2_switch_acl_entry *tmp;
+	int index, i, precedence, err;
+
+	/* Add the new ACL entry to the linked list and get its index */
+	index = dpaa2_switch_acl_entry_add_to_list(block, entry);
+
+	/* Move up in priority the ACL entries to make space
+	 * for the new filter.
+	 */
+	precedence = DPAA2_ETHSW_PORT_MAX_ACL_ENTRIES - block->num_acl_rules - 1;
+	for (i = 0; i < index; i++) {
+		tmp = dpaa2_switch_acl_entry_get_by_index(block, i);
+
+		err = dpaa2_switch_acl_entry_set_precedence(block, tmp,
+							    precedence);
+		if (err)
+			return err;
+
+		precedence++;
+	}
+
+	/* Add the new entry to hardware */
+	entry->cfg.precedence = precedence;
+	err = dpaa2_switch_acl_entry_add(block, entry);
+	block->num_acl_rules++;
+
+	return err;
+}
+
+static struct dpaa2_switch_acl_entry *
+dpaa2_switch_acl_tbl_find_entry_by_cookie(struct dpaa2_switch_filter_block *block,
+					  unsigned long cookie)
+{
+	struct dpaa2_switch_acl_entry *tmp, *n;
+
+	list_for_each_entry_safe(tmp, n, &block->acl_entries, list) {
+		if (tmp->cookie == cookie)
+			return tmp;
+	}
+	return NULL;
+}
+
+static int
+dpaa2_switch_acl_entry_get_index(struct dpaa2_switch_filter_block *block,
+				 struct dpaa2_switch_acl_entry *entry)
+{
+	struct dpaa2_switch_acl_entry *tmp, *n;
+	int index = 0;
+
+	list_for_each_entry_safe(tmp, n, &block->acl_entries, list) {
+		if (tmp->cookie == entry->cookie)
+			return index;
+		index++;
+	}
+	return -ENOENT;
+}
+
+static struct dpaa2_switch_mirror_entry *
+dpaa2_switch_mirror_find_entry_by_cookie(struct dpaa2_switch_filter_block *block,
+					 unsigned long cookie)
+{
+	struct dpaa2_switch_mirror_entry *tmp, *n;
+
+	list_for_each_entry_safe(tmp, n, &block->mirror_entries, list) {
+		if (tmp->cookie == cookie)
+			return tmp;
+	}
+	return NULL;
+}
+
+static int
+dpaa2_switch_acl_tbl_remove_entry(struct dpaa2_switch_filter_block *block,
+				  struct dpaa2_switch_acl_entry *entry)
+{
+	struct dpaa2_switch_acl_entry *tmp;
+	int index, i, precedence, err;
+
+	index = dpaa2_switch_acl_entry_get_index(block, entry);
+
+	/* Remove from hardware the ACL entry */
+	err = dpaa2_switch_acl_entry_remove(block, entry);
+	if (err)
+		return err;
+
+	block->num_acl_rules--;
+
+	/* Remove it from the list also */
+	list_del(&entry->list);
+
+	/* Move down in priority the entries over the deleted one */
+	precedence = entry->cfg.precedence;
+	for (i = index - 1; i >= 0; i--) {
+		tmp = dpaa2_switch_acl_entry_get_by_index(block, i);
+		err = dpaa2_switch_acl_entry_set_precedence(block, tmp,
+							    precedence);
+		if (err)
+			return err;
+
+		precedence--;
+	}
+
+	kfree(entry);
+
+	return 0;
+}
+
+static int dpaa2_switch_tc_parse_action_acl(struct ethsw_core *ethsw,
+					    struct flow_action_entry *cls_act,
+					    struct dpsw_acl_result *dpsw_act,
+					    struct netlink_ext_ack *extack)
+{
+	int err = 0;
+
+	switch (cls_act->id) {
+	case FLOW_ACTION_TRAP:
+		dpsw_act->action = DPSW_ACL_ACTION_REDIRECT_TO_CTRL_IF;
+		break;
+	case FLOW_ACTION_REDIRECT:
+		if (!dpaa2_switch_port_dev_check(cls_act->dev)) {
+			NL_SET_ERR_MSG_MOD(extack,
+					   "Destination not a DPAA2 switch port");
+			return -EOPNOTSUPP;
+		}
+
+		dpsw_act->if_id = dpaa2_switch_get_index(ethsw, cls_act->dev);
+		dpsw_act->action = DPSW_ACL_ACTION_REDIRECT;
+		break;
+	case FLOW_ACTION_DROP:
+		dpsw_act->action = DPSW_ACL_ACTION_DROP;
+		break;
+	default:
+		NL_SET_ERR_MSG_MOD(extack,
+				   "Action not supported");
+		err = -EOPNOTSUPP;
+		goto out;
+	}
+
+out:
+	return err;
+}
+
+static int
+dpaa2_switch_block_add_mirror(struct dpaa2_switch_filter_block *block,
+			      struct dpaa2_switch_mirror_entry *entry,
+			      u16 to, struct netlink_ext_ack *extack)
+{
+	unsigned long block_ports = block->ports;
+	struct ethsw_core *ethsw = block->ethsw;
+	struct ethsw_port_priv *port_priv;
+	unsigned long ports_added = 0;
+	u16 vlan = entry->cfg.vlan_id;
+	bool mirror_port_enabled;
+	int err, port;
+
+	/* Setup the mirroring port */
+	mirror_port_enabled = (ethsw->mirror_port != ethsw->sw_attr.num_ifs);
+	if (!mirror_port_enabled) {
+		err = dpsw_set_reflection_if(ethsw->mc_io, 0,
+					     ethsw->dpsw_handle, to);
+		if (err)
+			return err;
+		ethsw->mirror_port = to;
+	}
+
+	/* Setup the same egress mirroring configuration on all the switch
+	 * ports that share the same filter block.
+	 */
+	for_each_set_bit(port, &block_ports, ethsw->sw_attr.num_ifs) {
+		port_priv = ethsw->ports[port];
+
+		/* We cannot add a per VLAN mirroring rule if the VLAN in
+		 * question is not installed on the switch port.
+		 */
+		if (entry->cfg.filter == DPSW_REFLECTION_FILTER_INGRESS_VLAN &&
+		    !(port_priv->vlans[vlan] & ETHSW_VLAN_MEMBER)) {
+			NL_SET_ERR_MSG(extack,
+				       "VLAN must be installed on the switch port");
+			err = -EINVAL;
+			goto err_remove_filters;
+		}
+
+		err = dpsw_if_add_reflection(ethsw->mc_io, 0,
+					     ethsw->dpsw_handle,
+					     port, &entry->cfg);
+		if (err)
+			goto err_remove_filters;
+
+		ports_added |= BIT(port);
+	}
+
+	list_add(&entry->list, &block->mirror_entries);
+
+	return 0;
+
+err_remove_filters:
+	for_each_set_bit(port, &ports_added, ethsw->sw_attr.num_ifs) {
+		dpsw_if_remove_reflection(ethsw->mc_io, 0, ethsw->dpsw_handle,
+					  port, &entry->cfg);
+	}
+
+	if (!mirror_port_enabled)
+		ethsw->mirror_port = ethsw->sw_attr.num_ifs;
+
+	return err;
+}
+
+static int
+dpaa2_switch_block_remove_mirror(struct dpaa2_switch_filter_block *block,
+				 struct dpaa2_switch_mirror_entry *entry)
+{
+	struct dpsw_reflection_cfg *cfg = &entry->cfg;
+	unsigned long block_ports = block->ports;
+	struct ethsw_core *ethsw = block->ethsw;
+	int port;
+
+	/* Remove this mirroring configuration from all the ports belonging to
+	 * the filter block.
+	 */
+	for_each_set_bit(port, &block_ports, ethsw->sw_attr.num_ifs)
+		dpsw_if_remove_reflection(ethsw->mc_io, 0, ethsw->dpsw_handle,
+					  port, cfg);
+
+	/* Also remove it from the list of mirror filters */
+	list_del(&entry->list);
+	kfree(entry);
+
+	/* If this was the last mirror filter, then unset the mirror port */
+	if (list_empty(&block->mirror_entries))
+		ethsw->mirror_port =  ethsw->sw_attr.num_ifs;
+
+	return 0;
+}
+
+static int
+dpaa2_switch_cls_flower_replace_acl(struct dpaa2_switch_filter_block *block,
+				    struct flow_cls_offload *cls)
+{
+	struct flow_rule *rule = flow_cls_offload_flow_rule(cls);
+	struct netlink_ext_ack *extack = cls->common.extack;
+	struct dpaa2_switch_acl_entry *acl_entry;
+	struct ethsw_core *ethsw = block->ethsw;
+	struct flow_action_entry *act;
+	int err;
+
+	if (dpaa2_switch_acl_tbl_is_full(block)) {
+		NL_SET_ERR_MSG(extack, "Maximum filter capacity reached");
+		return -ENOMEM;
+	}
+
+	acl_entry = kzalloc(sizeof(*acl_entry), GFP_KERNEL);
+	if (!acl_entry)
+		return -ENOMEM;
+
+	err = dpaa2_switch_flower_parse_key(cls, &acl_entry->key);
+	if (err)
+		goto free_acl_entry;
+
+	act = &rule->action.entries[0];
+	err = dpaa2_switch_tc_parse_action_acl(ethsw, act,
+					       &acl_entry->cfg.result, extack);
+	if (err)
+		goto free_acl_entry;
+
+	acl_entry->prio = cls->common.prio;
+	acl_entry->cookie = cls->cookie;
+
+	err = dpaa2_switch_acl_tbl_add_entry(block, acl_entry);
+	if (err)
+		goto free_acl_entry;
+
+	return 0;
+
+free_acl_entry:
+	kfree(acl_entry);
+
+	return err;
+}
+
+static int dpaa2_switch_flower_parse_mirror_key(struct flow_cls_offload *cls,
+						u16 *vlan)
+{
+	struct flow_rule *rule = flow_cls_offload_flow_rule(cls);
+	struct flow_dissector *dissector = rule->match.dissector;
+	struct netlink_ext_ack *extack = cls->common.extack;
+	int ret = -EOPNOTSUPP;
+
+	if (dissector->used_keys &
+	    ~(BIT_ULL(FLOW_DISSECTOR_KEY_BASIC) |
+	      BIT_ULL(FLOW_DISSECTOR_KEY_CONTROL) |
+	      BIT_ULL(FLOW_DISSECTOR_KEY_VLAN))) {
+		NL_SET_ERR_MSG_MOD(extack,
+				   "Mirroring is supported only per VLAN");
+		return -EOPNOTSUPP;
+	}
+
+	if (flow_rule_match_has_control_flags(rule, extack))
+		return -EOPNOTSUPP;
+
+	if (flow_rule_match_key(rule, FLOW_DISSECTOR_KEY_VLAN)) {
+		struct flow_match_vlan match;
+
+		flow_rule_match_vlan(rule, &match);
+
+		if (match.mask->vlan_priority != 0 ||
+		    match.mask->vlan_dei != 0) {
+			NL_SET_ERR_MSG_MOD(extack,
+					   "Only matching on VLAN ID supported");
+			return -EOPNOTSUPP;
+		}
+
+		if (match.mask->vlan_id != 0xFFF) {
+			NL_SET_ERR_MSG_MOD(extack,
+					   "Masked matching not supported");
+			return -EOPNOTSUPP;
+		}
+
+		*vlan = (u16)match.key->vlan_id;
+		ret = 0;
+	}
+
+	return ret;
+}
+
+static int
+dpaa2_switch_cls_flower_replace_mirror(struct dpaa2_switch_filter_block *block,
+				       struct flow_cls_offload *cls)
+{
+	struct netlink_ext_ack *extack = cls->common.extack;
+	struct dpaa2_switch_mirror_entry *mirror_entry;
+	struct ethsw_core *ethsw = block->ethsw;
+	struct dpaa2_switch_mirror_entry *tmp;
+	struct flow_action_entry *cls_act;
+	struct list_head *pos, *n;
+	bool mirror_port_enabled;
+	u16 if_id, vlan;
+	int err;
+
+	mirror_port_enabled = (ethsw->mirror_port != ethsw->sw_attr.num_ifs);
+	cls_act = &cls->rule->action.entries[0];
+
+	/* Offload rules only when the destination is a DPAA2 switch port */
+	if (!dpaa2_switch_port_dev_check(cls_act->dev)) {
+		NL_SET_ERR_MSG_MOD(extack,
+				   "Destination not a DPAA2 switch port");
+		return -EOPNOTSUPP;
+	}
+	if_id = dpaa2_switch_get_index(ethsw, cls_act->dev);
+
+	/* We have a single mirror port but can configure egress mirroring on
+	 * all the other switch ports. We need to allow mirroring rules only
+	 * when the destination port is the same.
+	 */
+	if (mirror_port_enabled && ethsw->mirror_port != if_id) {
+		NL_SET_ERR_MSG_MOD(extack,
+				   "Multiple mirror ports not supported");
+		return -EBUSY;
+	}
+
+	/* Parse the key */
+	err = dpaa2_switch_flower_parse_mirror_key(cls, &vlan);
+	if (err)
+		return err;
+
+	/* Make sure that we don't already have a mirror rule with the same
+	 * configuration.
+	 */
+	list_for_each_safe(pos, n, &block->mirror_entries) {
+		tmp = list_entry(pos, struct dpaa2_switch_mirror_entry, list);
+
+		if (tmp->cfg.filter == DPSW_REFLECTION_FILTER_INGRESS_VLAN &&
+		    tmp->cfg.vlan_id == vlan) {
+			NL_SET_ERR_MSG_MOD(extack,
+					   "VLAN mirror filter already installed");
+			return -EBUSY;
+		}
+	}
+
+	mirror_entry = kzalloc(sizeof(*mirror_entry), GFP_KERNEL);
+	if (!mirror_entry)
+		return -ENOMEM;
+
+	mirror_entry->cfg.filter = DPSW_REFLECTION_FILTER_INGRESS_VLAN;
+	mirror_entry->cfg.vlan_id = vlan;
+	mirror_entry->cookie = cls->cookie;
+
+	return dpaa2_switch_block_add_mirror(block, mirror_entry, if_id,
+					     extack);
+}
+
+int dpaa2_switch_cls_flower_replace(struct dpaa2_switch_filter_block *block,
+				    struct flow_cls_offload *cls)
+{
+	struct flow_rule *rule = flow_cls_offload_flow_rule(cls);
+	struct netlink_ext_ack *extack = cls->common.extack;
+	struct flow_action_entry *act;
+
+	if (!flow_offload_has_one_action(&rule->action)) {
+		NL_SET_ERR_MSG(extack, "Only singular actions are supported");
+		return -EOPNOTSUPP;
+	}
+
+	act = &rule->action.entries[0];
+	switch (act->id) {
+	case FLOW_ACTION_REDIRECT:
+	case FLOW_ACTION_TRAP:
+	case FLOW_ACTION_DROP:
+		return dpaa2_switch_cls_flower_replace_acl(block, cls);
+	case FLOW_ACTION_MIRRED:
+		return dpaa2_switch_cls_flower_replace_mirror(block, cls);
+	default:
+		NL_SET_ERR_MSG_MOD(extack, "Action not supported");
+		return -EOPNOTSUPP;
+	}
+}
+
+int dpaa2_switch_cls_flower_destroy(struct dpaa2_switch_filter_block *block,
+				    struct flow_cls_offload *cls)
+{
+	struct dpaa2_switch_mirror_entry *mirror_entry;
+	struct dpaa2_switch_acl_entry *acl_entry;
+
+	/* If this filter is a an ACL one, remove it */
+	acl_entry = dpaa2_switch_acl_tbl_find_entry_by_cookie(block,
+							      cls->cookie);
+	if (acl_entry)
+		return dpaa2_switch_acl_tbl_remove_entry(block, acl_entry);
+
+	/* If not, then it has to be a mirror */
+	mirror_entry = dpaa2_switch_mirror_find_entry_by_cookie(block,
+								cls->cookie);
+	if (mirror_entry)
+		return dpaa2_switch_block_remove_mirror(block,
+							mirror_entry);
+
+	return 0;
+}
+
+static int
+dpaa2_switch_cls_matchall_replace_acl(struct dpaa2_switch_filter_block *block,
+				      struct tc_cls_matchall_offload *cls)
+{
+	struct netlink_ext_ack *extack = cls->common.extack;
+	struct ethsw_core *ethsw = block->ethsw;
+	struct dpaa2_switch_acl_entry *acl_entry;
+	struct flow_action_entry *act;
+	int err;
+
+	if (dpaa2_switch_acl_tbl_is_full(block)) {
+		NL_SET_ERR_MSG(extack, "Maximum filter capacity reached");
+		return -ENOMEM;
+	}
+
+	acl_entry = kzalloc(sizeof(*acl_entry), GFP_KERNEL);
+	if (!acl_entry)
+		return -ENOMEM;
+
+	act = &cls->rule->action.entries[0];
+	err = dpaa2_switch_tc_parse_action_acl(ethsw, act,
+					       &acl_entry->cfg.result, extack);
+	if (err)
+		goto free_acl_entry;
+
+	acl_entry->prio = cls->common.prio;
+	acl_entry->cookie = cls->cookie;
+
+	err = dpaa2_switch_acl_tbl_add_entry(block, acl_entry);
+	if (err)
+		goto free_acl_entry;
+
+	return 0;
+
+free_acl_entry:
+	kfree(acl_entry);
+
+	return err;
+}
+
+static int
+dpaa2_switch_cls_matchall_replace_mirror(struct dpaa2_switch_filter_block *block,
+					 struct tc_cls_matchall_offload *cls)
+{
+	struct netlink_ext_ack *extack = cls->common.extack;
+	struct dpaa2_switch_mirror_entry *mirror_entry;
+	struct ethsw_core *ethsw = block->ethsw;
+	struct dpaa2_switch_mirror_entry *tmp;
+	struct flow_action_entry *cls_act;
+	struct list_head *pos, *n;
+	bool mirror_port_enabled;
+	u16 if_id;
+
+	mirror_port_enabled = (ethsw->mirror_port != ethsw->sw_attr.num_ifs);
+	cls_act = &cls->rule->action.entries[0];
+
+	/* Offload rules only when the destination is a DPAA2 switch port */
+	if (!dpaa2_switch_port_dev_check(cls_act->dev)) {
+		NL_SET_ERR_MSG_MOD(extack,
+				   "Destination not a DPAA2 switch port");
+		return -EOPNOTSUPP;
+	}
+	if_id = dpaa2_switch_get_index(ethsw, cls_act->dev);
+
+	/* We have a single mirror port but can configure egress mirroring on
+	 * all the other switch ports. We need to allow mirroring rules only
+	 * when the destination port is the same.
+	 */
+	if (mirror_port_enabled && ethsw->mirror_port != if_id) {
+		NL_SET_ERR_MSG_MOD(extack,
+				   "Multiple mirror ports not supported");
+		return -EBUSY;
+	}
+
+	/* Make sure that we don't already have a mirror rule with the same
+	 * configuration. One matchall rule per block is the maximum.
+	 */
+	list_for_each_safe(pos, n, &block->mirror_entries) {
+		tmp = list_entry(pos, struct dpaa2_switch_mirror_entry, list);
+
+		if (tmp->cfg.filter == DPSW_REFLECTION_FILTER_INGRESS_ALL) {
+			NL_SET_ERR_MSG_MOD(extack,
+					   "Matchall mirror filter already installed");
+			return -EBUSY;
+		}
+	}
+
+	mirror_entry = kzalloc(sizeof(*mirror_entry), GFP_KERNEL);
+	if (!mirror_entry)
+		return -ENOMEM;
+
+	mirror_entry->cfg.filter = DPSW_REFLECTION_FILTER_INGRESS_ALL;
+	mirror_entry->cookie = cls->cookie;
+
+	return dpaa2_switch_block_add_mirror(block, mirror_entry, if_id,
+					     extack);
+}
+
+int dpaa2_switch_cls_matchall_replace(struct dpaa2_switch_filter_block *block,
+				      struct tc_cls_matchall_offload *cls)
+{
+	struct netlink_ext_ack *extack = cls->common.extack;
+	struct flow_action_entry *act;
+
+	if (!flow_offload_has_one_action(&cls->rule->action)) {
+		NL_SET_ERR_MSG(extack, "Only singular actions are supported");
+		return -EOPNOTSUPP;
+	}
+
+	act = &cls->rule->action.entries[0];
+	switch (act->id) {
+	case FLOW_ACTION_REDIRECT:
+	case FLOW_ACTION_TRAP:
+	case FLOW_ACTION_DROP:
+		return dpaa2_switch_cls_matchall_replace_acl(block, cls);
+	case FLOW_ACTION_MIRRED:
+		return dpaa2_switch_cls_matchall_replace_mirror(block, cls);
+	default:
+		NL_SET_ERR_MSG_MOD(extack, "Action not supported");
+		return -EOPNOTSUPP;
+	}
+}
+
+int dpaa2_switch_block_offload_mirror(struct dpaa2_switch_filter_block *block,
+				      struct ethsw_port_priv *port_priv)
+{
+	struct ethsw_core *ethsw = port_priv->ethsw_data;
+	struct dpaa2_switch_mirror_entry *tmp;
+	int err;
+
+	list_for_each_entry(tmp, &block->mirror_entries, list) {
+		err = dpsw_if_add_reflection(ethsw->mc_io, 0,
+					     ethsw->dpsw_handle,
+					     port_priv->idx, &tmp->cfg);
+		if (err)
+			goto unwind_add;
+	}
+
+	return 0;
+
+unwind_add:
+	list_for_each_entry(tmp, &block->mirror_entries, list)
+		dpsw_if_remove_reflection(ethsw->mc_io, 0,
+					  ethsw->dpsw_handle,
+					  port_priv->idx, &tmp->cfg);
+
+	return err;
+}
+
+int dpaa2_switch_block_unoffload_mirror(struct dpaa2_switch_filter_block *block,
+					struct ethsw_port_priv *port_priv)
+{
+	struct ethsw_core *ethsw = port_priv->ethsw_data;
+	struct dpaa2_switch_mirror_entry *tmp;
+	int err;
+
+	list_for_each_entry(tmp, &block->mirror_entries, list) {
+		err = dpsw_if_remove_reflection(ethsw->mc_io, 0,
+						ethsw->dpsw_handle,
+						port_priv->idx, &tmp->cfg);
+		if (err)
+			goto unwind_remove;
+	}
+
+	return 0;
+
+unwind_remove:
+	list_for_each_entry(tmp, &block->mirror_entries, list)
+		dpsw_if_add_reflection(ethsw->mc_io, 0, ethsw->dpsw_handle,
+				       port_priv->idx, &tmp->cfg);
+
+	return err;
+}
+
+int dpaa2_switch_cls_matchall_destroy(struct dpaa2_switch_filter_block *block,
+				      struct tc_cls_matchall_offload *cls)
+{
+	struct dpaa2_switch_mirror_entry *mirror_entry;
+	struct dpaa2_switch_acl_entry *acl_entry;
+
+	/* If this filter is a an ACL one, remove it */
+	acl_entry = dpaa2_switch_acl_tbl_find_entry_by_cookie(block,
+							      cls->cookie);
+	if (acl_entry)
+		return dpaa2_switch_acl_tbl_remove_entry(block,
+							 acl_entry);
+
+	/* If not, then it has to be a mirror */
+	mirror_entry = dpaa2_switch_mirror_find_entry_by_cookie(block,
+								cls->cookie);
+	if (mirror_entry)
+		return dpaa2_switch_block_remove_mirror(block,
+							mirror_entry);
+
+	return 0;
+}
diff --git a/drivers/net/ethernet/freescale/dpaa2/dpaa2-switch.c b/drivers/net/ethernet/freescale/dpaa2/dpaa2-switch.c
new file mode 100644
index 000000000000..b1e1ad9e4b48
--- /dev/null
+++ b/drivers/net/ethernet/freescale/dpaa2/dpaa2-switch.c
@@ -0,0 +1,3595 @@
+// SPDX-License-Identifier: GPL-2.0
+/*
+ * DPAA2 Ethernet Switch driver
+ *
+ * Copyright 2014-2016 Freescale Semiconductor Inc.
+ * Copyright 2017-2021 NXP
+ *
+ */
+
+#include <linux/module.h>
+
+#include <linux/interrupt.h>
+#include <linux/kthread.h>
+#include <linux/workqueue.h>
+#include <linux/iommu.h>
+#include <net/pkt_cls.h>
+
+#include <linux/fsl/mc.h>
+
+#include "dpaa2-switch.h"
+
+/* Minimal supported DPSW version */
+#define DPSW_MIN_VER_MAJOR		8
+#define DPSW_MIN_VER_MINOR		9
+
+#define DEFAULT_VLAN_ID			1
+
+static u16 dpaa2_switch_port_get_fdb_id(struct ethsw_port_priv *port_priv)
+{
+	return port_priv->fdb->fdb_id;
+}
+
+static struct dpaa2_switch_fdb *dpaa2_switch_fdb_get_unused(struct ethsw_core *ethsw)
+{
+	int i;
+
+	for (i = 0; i < ethsw->sw_attr.num_ifs; i++)
+		if (!ethsw->fdbs[i].in_use)
+			return &ethsw->fdbs[i];
+	return NULL;
+}
+
+static struct dpaa2_switch_filter_block *
+dpaa2_switch_filter_block_get_unused(struct ethsw_core *ethsw)
+{
+	int i;
+
+	for (i = 0; i < ethsw->sw_attr.num_ifs; i++)
+		if (!ethsw->filter_blocks[i].in_use)
+			return &ethsw->filter_blocks[i];
+	return NULL;
+}
+
+static u16 dpaa2_switch_port_set_fdb(struct ethsw_port_priv *port_priv,
+				     struct net_device *bridge_dev)
+{
+	struct ethsw_port_priv *other_port_priv = NULL;
+	struct dpaa2_switch_fdb *fdb;
+	struct net_device *other_dev;
+	struct list_head *iter;
+
+	/* If we leave a bridge (bridge_dev is NULL), find an unused
+	 * FDB and use that.
+	 */
+	if (!bridge_dev) {
+		fdb = dpaa2_switch_fdb_get_unused(port_priv->ethsw_data);
+
+		/* If there is no unused FDB, we must be the last port that
+		 * leaves the last bridge, all the others are standalone. We
+		 * can just keep the FDB that we already have.
+		 */
+
+		if (!fdb) {
+			port_priv->fdb->bridge_dev = NULL;
+			return 0;
+		}
+
+		port_priv->fdb = fdb;
+		port_priv->fdb->in_use = true;
+		port_priv->fdb->bridge_dev = NULL;
+		return 0;
+	}
+
+	/* The below call to netdev_for_each_lower_dev() demands the RTNL lock
+	 * being held. Assert on it so that it's easier to catch new code
+	 * paths that reach this point without the RTNL lock.
+	 */
+	ASSERT_RTNL();
+
+	/* If part of a bridge, use the FDB of the first dpaa2 switch interface
+	 * to be present in that bridge
+	 */
+	netdev_for_each_lower_dev(bridge_dev, other_dev, iter) {
+		if (!dpaa2_switch_port_dev_check(other_dev))
+			continue;
+
+		if (other_dev == port_priv->netdev)
+			continue;
+
+		other_port_priv = netdev_priv(other_dev);
+		break;
+	}
+
+	/* The current port is about to change its FDB to the one used by the
+	 * first port that joined the bridge.
+	 */
+	if (other_port_priv) {
+		/* The previous FDB is about to become unused, since the
+		 * interface is no longer standalone.
+		 */
+		port_priv->fdb->in_use = false;
+		port_priv->fdb->bridge_dev = NULL;
+
+		/* Get a reference to the new FDB */
+		port_priv->fdb = other_port_priv->fdb;
+	}
+
+	/* Keep track of the new upper bridge device */
+	port_priv->fdb->bridge_dev = bridge_dev;
+
+	return 0;
+}
+
+static void dpaa2_switch_fdb_get_flood_cfg(struct ethsw_core *ethsw, u16 fdb_id,
+					   enum dpsw_flood_type type,
+					   struct dpsw_egress_flood_cfg *cfg)
+{
+	int i = 0, j;
+
+	memset(cfg, 0, sizeof(*cfg));
+
+	/* Add all the DPAA2 switch ports found in the same bridging domain to
+	 * the egress flooding domain
+	 */
+	for (j = 0; j < ethsw->sw_attr.num_ifs; j++) {
+		if (!ethsw->ports[j])
+			continue;
+		if (ethsw->ports[j]->fdb->fdb_id != fdb_id)
+			continue;
+
+		if (type == DPSW_BROADCAST && ethsw->ports[j]->bcast_flood)
+			cfg->if_id[i++] = ethsw->ports[j]->idx;
+		else if (type == DPSW_FLOODING && ethsw->ports[j]->ucast_flood)
+			cfg->if_id[i++] = ethsw->ports[j]->idx;
+	}
+
+	/* Add the CTRL interface to the egress flooding domain */
+	cfg->if_id[i++] = ethsw->sw_attr.num_ifs;
+
+	cfg->fdb_id = fdb_id;
+	cfg->flood_type = type;
+	cfg->num_ifs = i;
+}
+
+static int dpaa2_switch_fdb_set_egress_flood(struct ethsw_core *ethsw, u16 fdb_id)
+{
+	struct dpsw_egress_flood_cfg flood_cfg;
+	int err;
+
+	/* Setup broadcast flooding domain */
+	dpaa2_switch_fdb_get_flood_cfg(ethsw, fdb_id, DPSW_BROADCAST, &flood_cfg);
+	err = dpsw_set_egress_flood(ethsw->mc_io, 0, ethsw->dpsw_handle,
+				    &flood_cfg);
+	if (err) {
+		dev_err(ethsw->dev, "dpsw_set_egress_flood() = %d\n", err);
+		return err;
+	}
+
+	/* Setup unknown flooding domain */
+	dpaa2_switch_fdb_get_flood_cfg(ethsw, fdb_id, DPSW_FLOODING, &flood_cfg);
+	err = dpsw_set_egress_flood(ethsw->mc_io, 0, ethsw->dpsw_handle,
+				    &flood_cfg);
+	if (err) {
+		dev_err(ethsw->dev, "dpsw_set_egress_flood() = %d\n", err);
+		return err;
+	}
+
+	return 0;
+}
+
+static void *dpaa2_iova_to_virt(struct iommu_domain *domain,
+				dma_addr_t iova_addr)
+{
+	phys_addr_t phys_addr;
+
+	phys_addr = domain ? iommu_iova_to_phys(domain, iova_addr) : iova_addr;
+
+	return phys_to_virt(phys_addr);
+}
+
+static int dpaa2_switch_add_vlan(struct ethsw_port_priv *port_priv, u16 vid)
+{
+	struct ethsw_core *ethsw = port_priv->ethsw_data;
+	struct dpsw_vlan_cfg vcfg = {0};
+	int err;
+
+	vcfg.fdb_id = dpaa2_switch_port_get_fdb_id(port_priv);
+	err = dpsw_vlan_add(ethsw->mc_io, 0,
+			    ethsw->dpsw_handle, vid, &vcfg);
+	if (err) {
+		dev_err(ethsw->dev, "dpsw_vlan_add err %d\n", err);
+		return err;
+	}
+	ethsw->vlans[vid] = ETHSW_VLAN_MEMBER;
+
+	return 0;
+}
+
+static bool dpaa2_switch_port_is_up(struct ethsw_port_priv *port_priv)
+{
+	struct net_device *netdev = port_priv->netdev;
+	struct dpsw_link_state state;
+	int err;
+
+	err = dpsw_if_get_link_state(port_priv->ethsw_data->mc_io, 0,
+				     port_priv->ethsw_data->dpsw_handle,
+				     port_priv->idx, &state);
+	if (err) {
+		netdev_err(netdev, "dpsw_if_get_link_state() err %d\n", err);
+		return true;
+	}
+
+	WARN_ONCE(state.up > 1, "Garbage read into link_state");
+
+	return state.up ? true : false;
+}
+
+static int dpaa2_switch_port_set_pvid(struct ethsw_port_priv *port_priv, u16 pvid)
+{
+	struct ethsw_core *ethsw = port_priv->ethsw_data;
+	struct net_device *netdev = port_priv->netdev;
+	struct dpsw_tci_cfg tci_cfg = { 0 };
+	bool up;
+	int err, ret;
+
+	err = dpsw_if_get_tci(ethsw->mc_io, 0, ethsw->dpsw_handle,
+			      port_priv->idx, &tci_cfg);
+	if (err) {
+		netdev_err(netdev, "dpsw_if_get_tci err %d\n", err);
+		return err;
+	}
+
+	tci_cfg.vlan_id = pvid;
+
+	/* Interface needs to be down to change PVID */
+	up = dpaa2_switch_port_is_up(port_priv);
+	if (up) {
+		err = dpsw_if_disable(ethsw->mc_io, 0,
+				      ethsw->dpsw_handle,
+				      port_priv->idx);
+		if (err) {
+			netdev_err(netdev, "dpsw_if_disable err %d\n", err);
+			return err;
+		}
+	}
+
+	err = dpsw_if_set_tci(ethsw->mc_io, 0, ethsw->dpsw_handle,
+			      port_priv->idx, &tci_cfg);
+	if (err) {
+		netdev_err(netdev, "dpsw_if_set_tci err %d\n", err);
+		goto set_tci_error;
+	}
+
+	/* Delete previous PVID info and mark the new one */
+	port_priv->vlans[port_priv->pvid] &= ~ETHSW_VLAN_PVID;
+	port_priv->vlans[pvid] |= ETHSW_VLAN_PVID;
+	port_priv->pvid = pvid;
+
+set_tci_error:
+	if (up) {
+		ret = dpsw_if_enable(ethsw->mc_io, 0,
+				     ethsw->dpsw_handle,
+				     port_priv->idx);
+		if (ret) {
+			netdev_err(netdev, "dpsw_if_enable err %d\n", ret);
+			return ret;
+		}
+	}
+
+	return err;
+}
+
+static int dpaa2_switch_port_add_vlan(struct ethsw_port_priv *port_priv,
+				      u16 vid, u16 flags)
+{
+	struct ethsw_core *ethsw = port_priv->ethsw_data;
+	struct net_device *netdev = port_priv->netdev;
+	struct dpsw_vlan_if_cfg vcfg = {0};
+	int err;
+
+	if (port_priv->vlans[vid]) {
+		netdev_err(netdev, "VLAN %d already configured\n", vid);
+		return -EEXIST;
+	}
+
+	/* If hit, this VLAN rule will lead the packet into the FDB table
+	 * specified in the vlan configuration below
+	 */
+	vcfg.num_ifs = 1;
+	vcfg.if_id[0] = port_priv->idx;
+	vcfg.fdb_id = dpaa2_switch_port_get_fdb_id(port_priv);
+	vcfg.options |= DPSW_VLAN_ADD_IF_OPT_FDB_ID;
+	err = dpsw_vlan_add_if(ethsw->mc_io, 0, ethsw->dpsw_handle, vid, &vcfg);
+	if (err) {
+		netdev_err(netdev, "dpsw_vlan_add_if err %d\n", err);
+		return err;
+	}
+
+	port_priv->vlans[vid] = ETHSW_VLAN_MEMBER;
+
+	if (flags & BRIDGE_VLAN_INFO_UNTAGGED) {
+		err = dpsw_vlan_add_if_untagged(ethsw->mc_io, 0,
+						ethsw->dpsw_handle,
+						vid, &vcfg);
+		if (err) {
+			netdev_err(netdev,
+				   "dpsw_vlan_add_if_untagged err %d\n", err);
+			return err;
+		}
+		port_priv->vlans[vid] |= ETHSW_VLAN_UNTAGGED;
+	}
+
+	if (flags & BRIDGE_VLAN_INFO_PVID) {
+		err = dpaa2_switch_port_set_pvid(port_priv, vid);
+		if (err)
+			return err;
+	}
+
+	return 0;
+}
+
+static enum dpsw_stp_state br_stp_state_to_dpsw(u8 state)
+{
+	switch (state) {
+	case BR_STATE_DISABLED:
+		return DPSW_STP_STATE_DISABLED;
+	case BR_STATE_LISTENING:
+		return DPSW_STP_STATE_LISTENING;
+	case BR_STATE_LEARNING:
+		return DPSW_STP_STATE_LEARNING;
+	case BR_STATE_FORWARDING:
+		return DPSW_STP_STATE_FORWARDING;
+	case BR_STATE_BLOCKING:
+		return DPSW_STP_STATE_BLOCKING;
+	default:
+		return DPSW_STP_STATE_DISABLED;
+	}
+}
+
+static int dpaa2_switch_port_set_stp_state(struct ethsw_port_priv *port_priv, u8 state)
+{
+	struct dpsw_stp_cfg stp_cfg = {0};
+	int err;
+	u16 vid;
+
+	if (!netif_running(port_priv->netdev) || state == port_priv->stp_state)
+		return 0;	/* Nothing to do */
+
+	stp_cfg.state = br_stp_state_to_dpsw(state);
+	for (vid = 0; vid <= VLAN_VID_MASK; vid++) {
+		if (port_priv->vlans[vid] & ETHSW_VLAN_MEMBER) {
+			stp_cfg.vlan_id = vid;
+			err = dpsw_if_set_stp(port_priv->ethsw_data->mc_io, 0,
+					      port_priv->ethsw_data->dpsw_handle,
+					      port_priv->idx, &stp_cfg);
+			if (err) {
+				netdev_err(port_priv->netdev,
+					   "dpsw_if_set_stp err %d\n", err);
+				return err;
+			}
+		}
+	}
+
+	port_priv->stp_state = state;
+
+	return 0;
+}
+
+static int dpaa2_switch_dellink(struct ethsw_core *ethsw, u16 vid)
+{
+	struct ethsw_port_priv *ppriv_local = NULL;
+	int i, err;
+
+	if (!ethsw->vlans[vid])
+		return -ENOENT;
+
+	err = dpsw_vlan_remove(ethsw->mc_io, 0, ethsw->dpsw_handle, vid);
+	if (err) {
+		dev_err(ethsw->dev, "dpsw_vlan_remove err %d\n", err);
+		return err;
+	}
+	ethsw->vlans[vid] = 0;
+
+	for (i = 0; i < ethsw->sw_attr.num_ifs; i++) {
+		ppriv_local = ethsw->ports[i];
+		if (ppriv_local)
+			ppriv_local->vlans[vid] = 0;
+	}
+
+	return 0;
+}
+
+static int dpaa2_switch_port_fdb_add_uc(struct ethsw_port_priv *port_priv,
+					const unsigned char *addr)
+{
+	struct dpsw_fdb_unicast_cfg entry = {0};
+	u16 fdb_id;
+	int err;
+
+	entry.if_egress = port_priv->idx;
+	entry.type = DPSW_FDB_ENTRY_STATIC;
+	ether_addr_copy(entry.mac_addr, addr);
+
+	fdb_id = dpaa2_switch_port_get_fdb_id(port_priv);
+	err = dpsw_fdb_add_unicast(port_priv->ethsw_data->mc_io, 0,
+				   port_priv->ethsw_data->dpsw_handle,
+				   fdb_id, &entry);
+	if (err)
+		netdev_err(port_priv->netdev,
+			   "dpsw_fdb_add_unicast err %d\n", err);
+	return err;
+}
+
+static int dpaa2_switch_port_fdb_del_uc(struct ethsw_port_priv *port_priv,
+					const unsigned char *addr)
+{
+	struct dpsw_fdb_unicast_cfg entry = {0};
+	u16 fdb_id;
+	int err;
+
+	entry.if_egress = port_priv->idx;
+	entry.type = DPSW_FDB_ENTRY_STATIC;
+	ether_addr_copy(entry.mac_addr, addr);
+
+	fdb_id = dpaa2_switch_port_get_fdb_id(port_priv);
+	err = dpsw_fdb_remove_unicast(port_priv->ethsw_data->mc_io, 0,
+				      port_priv->ethsw_data->dpsw_handle,
+				      fdb_id, &entry);
+	/* Silently discard error for calling multiple times the del command */
+	if (err && err != -ENXIO)
+		netdev_err(port_priv->netdev,
+			   "dpsw_fdb_remove_unicast err %d\n", err);
+	return err;
+}
+
+static int dpaa2_switch_port_fdb_add_mc(struct ethsw_port_priv *port_priv,
+					const unsigned char *addr)
+{
+	struct dpsw_fdb_multicast_cfg entry = {0};
+	u16 fdb_id;
+	int err;
+
+	ether_addr_copy(entry.mac_addr, addr);
+	entry.type = DPSW_FDB_ENTRY_STATIC;
+	entry.num_ifs = 1;
+	entry.if_id[0] = port_priv->idx;
+
+	fdb_id = dpaa2_switch_port_get_fdb_id(port_priv);
+	err = dpsw_fdb_add_multicast(port_priv->ethsw_data->mc_io, 0,
+				     port_priv->ethsw_data->dpsw_handle,
+				     fdb_id, &entry);
+	/* Silently discard error for calling multiple times the add command */
+	if (err && err != -ENXIO)
+		netdev_err(port_priv->netdev, "dpsw_fdb_add_multicast err %d\n",
+			   err);
+	return err;
+}
+
+static int dpaa2_switch_port_fdb_del_mc(struct ethsw_port_priv *port_priv,
+					const unsigned char *addr)
+{
+	struct dpsw_fdb_multicast_cfg entry = {0};
+	u16 fdb_id;
+	int err;
+
+	ether_addr_copy(entry.mac_addr, addr);
+	entry.type = DPSW_FDB_ENTRY_STATIC;
+	entry.num_ifs = 1;
+	entry.if_id[0] = port_priv->idx;
+
+	fdb_id = dpaa2_switch_port_get_fdb_id(port_priv);
+	err = dpsw_fdb_remove_multicast(port_priv->ethsw_data->mc_io, 0,
+					port_priv->ethsw_data->dpsw_handle,
+					fdb_id, &entry);
+	/* Silently discard error for calling multiple times the del command */
+	if (err && err != -ENAVAIL)
+		netdev_err(port_priv->netdev,
+			   "dpsw_fdb_remove_multicast err %d\n", err);
+	return err;
+}
+
+static void dpaa2_switch_port_get_stats(struct net_device *netdev,
+					struct rtnl_link_stats64 *stats)
+{
+	struct ethsw_port_priv *port_priv = netdev_priv(netdev);
+	u64 tmp;
+	int err;
+
+	err = dpsw_if_get_counter(port_priv->ethsw_data->mc_io, 0,
+				  port_priv->ethsw_data->dpsw_handle,
+				  port_priv->idx,
+				  DPSW_CNT_ING_FRAME, &stats->rx_packets);
+	if (err)
+		goto error;
+
+	err = dpsw_if_get_counter(port_priv->ethsw_data->mc_io, 0,
+				  port_priv->ethsw_data->dpsw_handle,
+				  port_priv->idx,
+				  DPSW_CNT_EGR_FRAME, &stats->tx_packets);
+	if (err)
+		goto error;
+
+	err = dpsw_if_get_counter(port_priv->ethsw_data->mc_io, 0,
+				  port_priv->ethsw_data->dpsw_handle,
+				  port_priv->idx,
+				  DPSW_CNT_ING_BYTE, &stats->rx_bytes);
+	if (err)
+		goto error;
+
+	err = dpsw_if_get_counter(port_priv->ethsw_data->mc_io, 0,
+				  port_priv->ethsw_data->dpsw_handle,
+				  port_priv->idx,
+				  DPSW_CNT_EGR_BYTE, &stats->tx_bytes);
+	if (err)
+		goto error;
+
+	err = dpsw_if_get_counter(port_priv->ethsw_data->mc_io, 0,
+				  port_priv->ethsw_data->dpsw_handle,
+				  port_priv->idx,
+				  DPSW_CNT_ING_FRAME_DISCARD,
+				  &stats->rx_dropped);
+	if (err)
+		goto error;
+
+	err = dpsw_if_get_counter(port_priv->ethsw_data->mc_io, 0,
+				  port_priv->ethsw_data->dpsw_handle,
+				  port_priv->idx,
+				  DPSW_CNT_ING_FLTR_FRAME,
+				  &tmp);
+	if (err)
+		goto error;
+	stats->rx_dropped += tmp;
+
+	err = dpsw_if_get_counter(port_priv->ethsw_data->mc_io, 0,
+				  port_priv->ethsw_data->dpsw_handle,
+				  port_priv->idx,
+				  DPSW_CNT_EGR_FRAME_DISCARD,
+				  &stats->tx_dropped);
+	if (err)
+		goto error;
+
+	return;
+
+error:
+	netdev_err(netdev, "dpsw_if_get_counter err %d\n", err);
+}
+
+static bool dpaa2_switch_port_has_offload_stats(const struct net_device *netdev,
+						int attr_id)
+{
+	return (attr_id == IFLA_OFFLOAD_XSTATS_CPU_HIT);
+}
+
+static int dpaa2_switch_port_get_offload_stats(int attr_id,
+					       const struct net_device *netdev,
+					       void *sp)
+{
+	switch (attr_id) {
+	case IFLA_OFFLOAD_XSTATS_CPU_HIT:
+		dpaa2_switch_port_get_stats((struct net_device *)netdev, sp);
+		return 0;
+	}
+
+	return -EINVAL;
+}
+
+static int dpaa2_switch_port_change_mtu(struct net_device *netdev, int mtu)
+{
+	struct ethsw_port_priv *port_priv = netdev_priv(netdev);
+	int err;
+
+	err = dpsw_if_set_max_frame_length(port_priv->ethsw_data->mc_io,
+					   0,
+					   port_priv->ethsw_data->dpsw_handle,
+					   port_priv->idx,
+					   (u16)ETHSW_L2_MAX_FRM(mtu));
+	if (err) {
+		netdev_err(netdev,
+			   "dpsw_if_set_max_frame_length() err %d\n", err);
+		return err;
+	}
+
+	WRITE_ONCE(netdev->mtu, mtu);
+	return 0;
+}
+
+static int dpaa2_switch_port_link_state_update(struct net_device *netdev)
+{
+	struct ethsw_port_priv *port_priv = netdev_priv(netdev);
+	struct dpsw_link_state state;
+	int err;
+
+	/* When we manage the MAC/PHY using phylink there is no need
+	 * to manually update the netif_carrier.
+	 * We can avoid locking because we are called from the "link changed"
+	 * IRQ handler, which is the same as the "endpoint changed" IRQ handler
+	 * (the writer to port_priv->mac), so we cannot race with it.
+	 */
+	if (dpaa2_mac_is_type_phy(port_priv->mac))
+		return 0;
+
+	/* Interrupts are received even though no one issued an 'ifconfig up'
+	 * on the switch interface. Ignore these link state update interrupts
+	 */
+	if (!netif_running(netdev))
+		return 0;
+
+	err = dpsw_if_get_link_state(port_priv->ethsw_data->mc_io, 0,
+				     port_priv->ethsw_data->dpsw_handle,
+				     port_priv->idx, &state);
+	if (err) {
+		netdev_err(netdev, "dpsw_if_get_link_state() err %d\n", err);
+		return err;
+	}
+
+	WARN_ONCE(state.up > 1, "Garbage read into link_state");
+
+	if (state.up != port_priv->link_state) {
+		if (state.up) {
+			netif_carrier_on(netdev);
+			netif_tx_start_all_queues(netdev);
+		} else {
+			netif_carrier_off(netdev);
+			netif_tx_stop_all_queues(netdev);
+		}
+		port_priv->link_state = state.up;
+	}
+
+	return 0;
+}
+
+/* Manage all NAPI instances for the control interface.
+ *
+ * We only have one RX queue and one Tx Conf queue for all
+ * switch ports. Therefore, we only need to enable the NAPI instance once, the
+ * first time one of the switch ports runs .dev_open().
+ */
+
+static void dpaa2_switch_enable_ctrl_if_napi(struct ethsw_core *ethsw)
+{
+	int i;
+
+	/* Access to the ethsw->napi_users relies on the RTNL lock */
+	ASSERT_RTNL();
+
+	/* a new interface is using the NAPI instance */
+	ethsw->napi_users++;
+
+	/* if there is already a user of the instance, return */
+	if (ethsw->napi_users > 1)
+		return;
+
+	for (i = 0; i < DPAA2_SWITCH_RX_NUM_FQS; i++)
+		napi_enable(&ethsw->fq[i].napi);
+}
+
+static void dpaa2_switch_disable_ctrl_if_napi(struct ethsw_core *ethsw)
+{
+	int i;
+
+	/* Access to the ethsw->napi_users relies on the RTNL lock */
+	ASSERT_RTNL();
+
+	/* If we are not the last interface using the NAPI, return */
+	ethsw->napi_users--;
+	if (ethsw->napi_users)
+		return;
+
+	for (i = 0; i < DPAA2_SWITCH_RX_NUM_FQS; i++)
+		napi_disable(&ethsw->fq[i].napi);
+}
+
+static int dpaa2_switch_port_open(struct net_device *netdev)
+{
+	struct ethsw_port_priv *port_priv = netdev_priv(netdev);
+	struct ethsw_core *ethsw = port_priv->ethsw_data;
+	int err;
+
+	mutex_lock(&port_priv->mac_lock);
+
+	if (!dpaa2_switch_port_is_type_phy(port_priv)) {
+		/* Explicitly set carrier off, otherwise
+		 * netif_carrier_ok() will return true and cause 'ip link show'
+		 * to report the LOWER_UP flag, even though the link
+		 * notification wasn't even received.
+		 */
+		netif_carrier_off(netdev);
+	}
+
+	err = dpsw_if_enable(port_priv->ethsw_data->mc_io, 0,
+			     port_priv->ethsw_data->dpsw_handle,
+			     port_priv->idx);
+	if (err) {
+		mutex_unlock(&port_priv->mac_lock);
+		netdev_err(netdev, "dpsw_if_enable err %d\n", err);
+		return err;
+	}
+
+	dpaa2_switch_enable_ctrl_if_napi(ethsw);
+
+	if (dpaa2_switch_port_is_type_phy(port_priv))
+		dpaa2_mac_start(port_priv->mac);
+
+	mutex_unlock(&port_priv->mac_lock);
+
+	return 0;
+}
+
+static int dpaa2_switch_port_stop(struct net_device *netdev)
+{
+	struct ethsw_port_priv *port_priv = netdev_priv(netdev);
+	struct ethsw_core *ethsw = port_priv->ethsw_data;
+	int err;
+
+	mutex_lock(&port_priv->mac_lock);
+
+	if (dpaa2_switch_port_is_type_phy(port_priv)) {
+		dpaa2_mac_stop(port_priv->mac);
+	} else {
+		netif_tx_stop_all_queues(netdev);
+		netif_carrier_off(netdev);
+	}
+
+	mutex_unlock(&port_priv->mac_lock);
+
+	err = dpsw_if_disable(port_priv->ethsw_data->mc_io, 0,
+			      port_priv->ethsw_data->dpsw_handle,
+			      port_priv->idx);
+	if (err) {
+		netdev_err(netdev, "dpsw_if_disable err %d\n", err);
+		return err;
+	}
+
+	dpaa2_switch_disable_ctrl_if_napi(ethsw);
+
+	return 0;
+}
+
+static int dpaa2_switch_port_parent_id(struct net_device *dev,
+				       struct netdev_phys_item_id *ppid)
+{
+	struct ethsw_port_priv *port_priv = netdev_priv(dev);
+
+	ppid->id_len = 1;
+	ppid->id[0] = port_priv->ethsw_data->dev_id;
+
+	return 0;
+}
+
+static int dpaa2_switch_port_get_phys_name(struct net_device *netdev, char *name,
+					   size_t len)
+{
+	struct ethsw_port_priv *port_priv = netdev_priv(netdev);
+	int err;
+
+	err = snprintf(name, len, "p%d", port_priv->idx);
+	if (err >= len)
+		return -EINVAL;
+
+	return 0;
+}
+
+struct ethsw_dump_ctx {
+	struct net_device *dev;
+	struct sk_buff *skb;
+	struct netlink_callback *cb;
+	int idx;
+};
+
+static int dpaa2_switch_fdb_dump_nl(struct fdb_dump_entry *entry,
+				    struct ethsw_dump_ctx *dump)
+{
+	struct ndo_fdb_dump_context *ctx = (void *)dump->cb->ctx;
+	int is_dynamic = entry->type & DPSW_FDB_ENTRY_DINAMIC;
+	u32 portid = NETLINK_CB(dump->cb->skb).portid;
+	u32 seq = dump->cb->nlh->nlmsg_seq;
+	struct nlmsghdr *nlh;
+	struct ndmsg *ndm;
+
+	if (dump->idx < ctx->fdb_idx)
+		goto skip;
+
+	nlh = nlmsg_put(dump->skb, portid, seq, RTM_NEWNEIGH,
+			sizeof(*ndm), NLM_F_MULTI);
+	if (!nlh)
+		return -EMSGSIZE;
+
+	ndm = nlmsg_data(nlh);
+	ndm->ndm_family  = AF_BRIDGE;
+	ndm->ndm_pad1    = 0;
+	ndm->ndm_pad2    = 0;
+	ndm->ndm_flags   = NTF_SELF;
+	ndm->ndm_type    = 0;
+	ndm->ndm_ifindex = dump->dev->ifindex;
+	ndm->ndm_state   = is_dynamic ? NUD_REACHABLE : NUD_NOARP;
+
+	if (nla_put(dump->skb, NDA_LLADDR, ETH_ALEN, entry->mac_addr))
+		goto nla_put_failure;
+
+	nlmsg_end(dump->skb, nlh);
+
+skip:
+	dump->idx++;
+	return 0;
+
+nla_put_failure:
+	nlmsg_cancel(dump->skb, nlh);
+	return -EMSGSIZE;
+}
+
+static int dpaa2_switch_port_fdb_valid_entry(struct fdb_dump_entry *entry,
+					     struct ethsw_port_priv *port_priv)
+{
+	int idx = port_priv->idx;
+	int valid;
+
+	if (entry->type & DPSW_FDB_ENTRY_TYPE_UNICAST)
+		valid = entry->if_info == port_priv->idx;
+	else
+		valid = entry->if_mask[idx / 8] & BIT(idx % 8);
+
+	return valid;
+}
+
+static int dpaa2_switch_fdb_iterate(struct ethsw_port_priv *port_priv,
+				    dpaa2_switch_fdb_cb_t cb, void *data)
+{
+	struct net_device *net_dev = port_priv->netdev;
+	struct ethsw_core *ethsw = port_priv->ethsw_data;
+	struct device *dev = net_dev->dev.parent;
+	struct fdb_dump_entry *fdb_entries;
+	struct fdb_dump_entry fdb_entry;
+	dma_addr_t fdb_dump_iova;
+	u16 num_fdb_entries;
+	u32 fdb_dump_size;
+	int err = 0, i;
+	u8 *dma_mem;
+	u16 fdb_id;
+
+	fdb_dump_size = ethsw->sw_attr.max_fdb_entries * sizeof(fdb_entry);
+	dma_mem = kzalloc(fdb_dump_size, GFP_KERNEL);
+	if (!dma_mem)
+		return -ENOMEM;
+
+	fdb_dump_iova = dma_map_single(dev, dma_mem, fdb_dump_size,
+				       DMA_FROM_DEVICE);
+	if (dma_mapping_error(dev, fdb_dump_iova)) {
+		netdev_err(net_dev, "dma_map_single() failed\n");
+		err = -ENOMEM;
+		goto err_map;
+	}
+
+	fdb_id = dpaa2_switch_port_get_fdb_id(port_priv);
+	err = dpsw_fdb_dump(ethsw->mc_io, 0, ethsw->dpsw_handle, fdb_id,
+			    fdb_dump_iova, fdb_dump_size, &num_fdb_entries);
+	if (err) {
+		netdev_err(net_dev, "dpsw_fdb_dump() = %d\n", err);
+		goto err_dump;
+	}
+
+	dma_unmap_single(dev, fdb_dump_iova, fdb_dump_size, DMA_FROM_DEVICE);
+
+	fdb_entries = (struct fdb_dump_entry *)dma_mem;
+	for (i = 0; i < num_fdb_entries; i++) {
+		fdb_entry = fdb_entries[i];
+
+		err = cb(port_priv, &fdb_entry, data);
+		if (err)
+			goto end;
+	}
+
+end:
+	kfree(dma_mem);
+
+	return 0;
+
+err_dump:
+	dma_unmap_single(dev, fdb_dump_iova, fdb_dump_size, DMA_TO_DEVICE);
+err_map:
+	kfree(dma_mem);
+	return err;
+}
+
+static int dpaa2_switch_fdb_entry_dump(struct ethsw_port_priv *port_priv,
+				       struct fdb_dump_entry *fdb_entry,
+				       void *data)
+{
+	if (!dpaa2_switch_port_fdb_valid_entry(fdb_entry, port_priv))
+		return 0;
+
+	return dpaa2_switch_fdb_dump_nl(fdb_entry, data);
+}
+
+static int dpaa2_switch_port_fdb_dump(struct sk_buff *skb, struct netlink_callback *cb,
+				      struct net_device *net_dev,
+				      struct net_device *filter_dev, int *idx)
+{
+	struct ethsw_port_priv *port_priv = netdev_priv(net_dev);
+	struct ethsw_dump_ctx dump = {
+		.dev = net_dev,
+		.skb = skb,
+		.cb = cb,
+		.idx = *idx,
+	};
+	int err;
+
+	err = dpaa2_switch_fdb_iterate(port_priv, dpaa2_switch_fdb_entry_dump, &dump);
+	*idx = dump.idx;
+
+	return err;
+}
+
+static int dpaa2_switch_fdb_entry_fast_age(struct ethsw_port_priv *port_priv,
+					   struct fdb_dump_entry *fdb_entry,
+					   void *data __always_unused)
+{
+	if (!dpaa2_switch_port_fdb_valid_entry(fdb_entry, port_priv))
+		return 0;
+
+	if (!(fdb_entry->type & DPSW_FDB_ENTRY_TYPE_DYNAMIC))
+		return 0;
+
+	if (fdb_entry->type & DPSW_FDB_ENTRY_TYPE_UNICAST)
+		dpaa2_switch_port_fdb_del_uc(port_priv, fdb_entry->mac_addr);
+	else
+		dpaa2_switch_port_fdb_del_mc(port_priv, fdb_entry->mac_addr);
+
+	return 0;
+}
+
+static void dpaa2_switch_port_fast_age(struct ethsw_port_priv *port_priv)
+{
+	dpaa2_switch_fdb_iterate(port_priv,
+				 dpaa2_switch_fdb_entry_fast_age, NULL);
+}
+
+static int dpaa2_switch_port_vlan_add(struct net_device *netdev, __be16 proto,
+				      u16 vid)
+{
+	struct switchdev_obj_port_vlan vlan = {
+		.obj.id = SWITCHDEV_OBJ_ID_PORT_VLAN,
+		.vid = vid,
+		.obj.orig_dev = netdev,
+		/* This API only allows programming tagged, non-PVID VIDs */
+		.flags = 0,
+	};
+
+	return dpaa2_switch_port_vlans_add(netdev, &vlan);
+}
+
+static int dpaa2_switch_port_vlan_kill(struct net_device *netdev, __be16 proto,
+				       u16 vid)
+{
+	struct switchdev_obj_port_vlan vlan = {
+		.obj.id = SWITCHDEV_OBJ_ID_PORT_VLAN,
+		.vid = vid,
+		.obj.orig_dev = netdev,
+		/* This API only allows programming tagged, non-PVID VIDs */
+		.flags = 0,
+	};
+
+	return dpaa2_switch_port_vlans_del(netdev, &vlan);
+}
+
+static int dpaa2_switch_port_set_mac_addr(struct ethsw_port_priv *port_priv)
+{
+	struct ethsw_core *ethsw = port_priv->ethsw_data;
+	struct net_device *net_dev = port_priv->netdev;
+	struct device *dev = net_dev->dev.parent;
+	u8 mac_addr[ETH_ALEN];
+	int err;
+
+	if (!(ethsw->features & ETHSW_FEATURE_MAC_ADDR))
+		return 0;
+
+	/* Get firmware address, if any */
+	err = dpsw_if_get_port_mac_addr(ethsw->mc_io, 0, ethsw->dpsw_handle,
+					port_priv->idx, mac_addr);
+	if (err) {
+		dev_err(dev, "dpsw_if_get_port_mac_addr() failed\n");
+		return err;
+	}
+
+	/* First check if firmware has any address configured by bootloader */
+	if (!is_zero_ether_addr(mac_addr)) {
+		eth_hw_addr_set(net_dev, mac_addr);
+	} else {
+		/* No MAC address configured, fill in net_dev->dev_addr
+		 * with a random one
+		 */
+		eth_hw_addr_random(net_dev);
+		dev_dbg_once(dev, "device(s) have all-zero hwaddr, replaced with random\n");
+
+		/* Override NET_ADDR_RANDOM set by eth_hw_addr_random(); for all
+		 * practical purposes, this will be our "permanent" mac address,
+		 * at least until the next reboot. This move will also permit
+		 * register_netdevice() to properly fill up net_dev->perm_addr.
+		 */
+		net_dev->addr_assign_type = NET_ADDR_PERM;
+	}
+
+	return 0;
+}
+
+static void dpaa2_switch_free_fd(const struct ethsw_core *ethsw,
+				 const struct dpaa2_fd *fd)
+{
+	struct device *dev = ethsw->dev;
+	unsigned char *buffer_start;
+	struct sk_buff **skbh, *skb;
+	dma_addr_t fd_addr;
+
+	fd_addr = dpaa2_fd_get_addr(fd);
+	skbh = dpaa2_iova_to_virt(ethsw->iommu_domain, fd_addr);
+
+	skb = *skbh;
+	buffer_start = (unsigned char *)skbh;
+
+	dma_unmap_single(dev, fd_addr,
+			 skb_tail_pointer(skb) - buffer_start,
+			 DMA_TO_DEVICE);
+
+	/* Move on with skb release */
+	dev_kfree_skb(skb);
+}
+
+static int dpaa2_switch_build_single_fd(struct ethsw_core *ethsw,
+					struct sk_buff *skb,
+					struct dpaa2_fd *fd)
+{
+	struct device *dev = ethsw->dev;
+	struct sk_buff **skbh;
+	dma_addr_t addr;
+	u8 *buff_start;
+	void *hwa;
+
+	buff_start = PTR_ALIGN(skb->data - DPAA2_SWITCH_TX_DATA_OFFSET -
+			       DPAA2_SWITCH_TX_BUF_ALIGN,
+			       DPAA2_SWITCH_TX_BUF_ALIGN);
+
+	/* Clear FAS to have consistent values for TX confirmation. It is
+	 * located in the first 8 bytes of the buffer's hardware annotation
+	 * area
+	 */
+	hwa = buff_start + DPAA2_SWITCH_SWA_SIZE;
+	memset(hwa, 0, 8);
+
+	/* Store a backpointer to the skb at the beginning of the buffer
+	 * (in the private data area) such that we can release it
+	 * on Tx confirm
+	 */
+	skbh = (struct sk_buff **)buff_start;
+	*skbh = skb;
+
+	addr = dma_map_single(dev, buff_start,
+			      skb_tail_pointer(skb) - buff_start,
+			      DMA_TO_DEVICE);
+	if (unlikely(dma_mapping_error(dev, addr)))
+		return -ENOMEM;
+
+	/* Setup the FD fields */
+	memset(fd, 0, sizeof(*fd));
+
+	dpaa2_fd_set_addr(fd, addr);
+	dpaa2_fd_set_offset(fd, (u16)(skb->data - buff_start));
+	dpaa2_fd_set_len(fd, skb->len);
+	dpaa2_fd_set_format(fd, dpaa2_fd_single);
+
+	return 0;
+}
+
+static netdev_tx_t dpaa2_switch_port_tx(struct sk_buff *skb,
+					struct net_device *net_dev)
+{
+	struct ethsw_port_priv *port_priv = netdev_priv(net_dev);
+	struct ethsw_core *ethsw = port_priv->ethsw_data;
+	int retries = DPAA2_SWITCH_SWP_BUSY_RETRIES;
+	struct dpaa2_fd fd;
+	int err;
+
+	if (unlikely(skb_headroom(skb) < DPAA2_SWITCH_NEEDED_HEADROOM)) {
+		struct sk_buff *ns;
+
+		ns = skb_realloc_headroom(skb, DPAA2_SWITCH_NEEDED_HEADROOM);
+		if (unlikely(!ns)) {
+			net_err_ratelimited("%s: Error reallocating skb headroom\n", net_dev->name);
+			goto err_free_skb;
+		}
+		dev_consume_skb_any(skb);
+		skb = ns;
+	}
+
+	/* We'll be holding a back-reference to the skb until Tx confirmation */
+	skb = skb_unshare(skb, GFP_ATOMIC);
+	if (unlikely(!skb)) {
+		/* skb_unshare() has already freed the skb */
+		net_err_ratelimited("%s: Error copying the socket buffer\n", net_dev->name);
+		goto err_exit;
+	}
+
+	/* At this stage, we do not support non-linear skbs so just try to
+	 * linearize the skb and if that's not working, just drop the packet.
+	 */
+	err = skb_linearize(skb);
+	if (err) {
+		net_err_ratelimited("%s: skb_linearize error (%d)!\n", net_dev->name, err);
+		goto err_free_skb;
+	}
+
+	err = dpaa2_switch_build_single_fd(ethsw, skb, &fd);
+	if (unlikely(err)) {
+		net_err_ratelimited("%s: ethsw_build_*_fd() %d\n", net_dev->name, err);
+		goto err_free_skb;
+	}
+
+	do {
+		err = dpaa2_io_service_enqueue_qd(NULL,
+						  port_priv->tx_qdid,
+						  8, 0, &fd);
+		retries--;
+	} while (err == -EBUSY && retries);
+
+	if (unlikely(err < 0)) {
+		dpaa2_switch_free_fd(ethsw, &fd);
+		goto err_exit;
+	}
+
+	return NETDEV_TX_OK;
+
+err_free_skb:
+	dev_kfree_skb(skb);
+err_exit:
+	return NETDEV_TX_OK;
+}
+
+static int
+dpaa2_switch_setup_tc_cls_flower(struct dpaa2_switch_filter_block *filter_block,
+				 struct flow_cls_offload *f)
+{
+	switch (f->command) {
+	case FLOW_CLS_REPLACE:
+		return dpaa2_switch_cls_flower_replace(filter_block, f);
+	case FLOW_CLS_DESTROY:
+		return dpaa2_switch_cls_flower_destroy(filter_block, f);
+	default:
+		return -EOPNOTSUPP;
+	}
+}
+
+static int
+dpaa2_switch_setup_tc_cls_matchall(struct dpaa2_switch_filter_block *block,
+				   struct tc_cls_matchall_offload *f)
+{
+	switch (f->command) {
+	case TC_CLSMATCHALL_REPLACE:
+		return dpaa2_switch_cls_matchall_replace(block, f);
+	case TC_CLSMATCHALL_DESTROY:
+		return dpaa2_switch_cls_matchall_destroy(block, f);
+	default:
+		return -EOPNOTSUPP;
+	}
+}
+
+static int dpaa2_switch_port_setup_tc_block_cb_ig(enum tc_setup_type type,
+						  void *type_data,
+						  void *cb_priv)
+{
+	switch (type) {
+	case TC_SETUP_CLSFLOWER:
+		return dpaa2_switch_setup_tc_cls_flower(cb_priv, type_data);
+	case TC_SETUP_CLSMATCHALL:
+		return dpaa2_switch_setup_tc_cls_matchall(cb_priv, type_data);
+	default:
+		return -EOPNOTSUPP;
+	}
+}
+
+static LIST_HEAD(dpaa2_switch_block_cb_list);
+
+static int
+dpaa2_switch_port_acl_tbl_bind(struct ethsw_port_priv *port_priv,
+			       struct dpaa2_switch_filter_block *block)
+{
+	struct ethsw_core *ethsw = port_priv->ethsw_data;
+	struct net_device *netdev = port_priv->netdev;
+	struct dpsw_acl_if_cfg acl_if_cfg;
+	int err;
+
+	if (port_priv->filter_block)
+		return -EINVAL;
+
+	acl_if_cfg.if_id[0] = port_priv->idx;
+	acl_if_cfg.num_ifs = 1;
+	err = dpsw_acl_add_if(ethsw->mc_io, 0, ethsw->dpsw_handle,
+			      block->acl_id, &acl_if_cfg);
+	if (err) {
+		netdev_err(netdev, "dpsw_acl_add_if err %d\n", err);
+		return err;
+	}
+
+	block->ports |= BIT(port_priv->idx);
+	port_priv->filter_block = block;
+
+	return 0;
+}
+
+static int
+dpaa2_switch_port_acl_tbl_unbind(struct ethsw_port_priv *port_priv,
+				 struct dpaa2_switch_filter_block *block)
+{
+	struct ethsw_core *ethsw = port_priv->ethsw_data;
+	struct net_device *netdev = port_priv->netdev;
+	struct dpsw_acl_if_cfg acl_if_cfg;
+	int err;
+
+	if (port_priv->filter_block != block)
+		return -EINVAL;
+
+	acl_if_cfg.if_id[0] = port_priv->idx;
+	acl_if_cfg.num_ifs = 1;
+	err = dpsw_acl_remove_if(ethsw->mc_io, 0, ethsw->dpsw_handle,
+				 block->acl_id, &acl_if_cfg);
+	if (err) {
+		netdev_err(netdev, "dpsw_acl_add_if err %d\n", err);
+		return err;
+	}
+
+	block->ports &= ~BIT(port_priv->idx);
+	port_priv->filter_block = NULL;
+	return 0;
+}
+
+static int dpaa2_switch_port_block_bind(struct ethsw_port_priv *port_priv,
+					struct dpaa2_switch_filter_block *block)
+{
+	struct dpaa2_switch_filter_block *old_block = port_priv->filter_block;
+	int err;
+
+	/* Offload all the mirror entries found in the block on this new port
+	 * joining it.
+	 */
+	err = dpaa2_switch_block_offload_mirror(block, port_priv);
+	if (err)
+		return err;
+
+	/* If the port is already bound to this ACL table then do nothing. This
+	 * can happen when this port is the first one to join a tc block
+	 */
+	if (port_priv->filter_block == block)
+		return 0;
+
+	err = dpaa2_switch_port_acl_tbl_unbind(port_priv, old_block);
+	if (err)
+		return err;
+
+	/* Mark the previous ACL table as being unused if this was the last
+	 * port that was using it.
+	 */
+	if (old_block->ports == 0)
+		old_block->in_use = false;
+
+	return dpaa2_switch_port_acl_tbl_bind(port_priv, block);
+}
+
+static int
+dpaa2_switch_port_block_unbind(struct ethsw_port_priv *port_priv,
+			       struct dpaa2_switch_filter_block *block)
+{
+	struct ethsw_core *ethsw = port_priv->ethsw_data;
+	struct dpaa2_switch_filter_block *new_block;
+	int err;
+
+	/* Unoffload all the mirror entries found in the block from the
+	 * port leaving it.
+	 */
+	err = dpaa2_switch_block_unoffload_mirror(block, port_priv);
+	if (err)
+		return err;
+
+	/* We are the last port that leaves a block (an ACL table).
+	 * We'll continue to use this table.
+	 */
+	if (block->ports == BIT(port_priv->idx))
+		return 0;
+
+	err = dpaa2_switch_port_acl_tbl_unbind(port_priv, block);
+	if (err)
+		return err;
+
+	if (block->ports == 0)
+		block->in_use = false;
+
+	new_block = dpaa2_switch_filter_block_get_unused(ethsw);
+	new_block->in_use = true;
+	return dpaa2_switch_port_acl_tbl_bind(port_priv, new_block);
+}
+
+static int dpaa2_switch_setup_tc_block_bind(struct net_device *netdev,
+					    struct flow_block_offload *f)
+{
+	struct ethsw_port_priv *port_priv = netdev_priv(netdev);
+	struct ethsw_core *ethsw = port_priv->ethsw_data;
+	struct dpaa2_switch_filter_block *filter_block;
+	struct flow_block_cb *block_cb;
+	bool register_block = false;
+	int err;
+
+	block_cb = flow_block_cb_lookup(f->block,
+					dpaa2_switch_port_setup_tc_block_cb_ig,
+					ethsw);
+
+	if (!block_cb) {
+		/* If the filter block is not already known, then this port
+		 * must be the first to join it. In this case, we can just
+		 * continue to use our private table
+		 */
+		filter_block = port_priv->filter_block;
+
+		block_cb = flow_block_cb_alloc(dpaa2_switch_port_setup_tc_block_cb_ig,
+					       ethsw, filter_block, NULL);
+		if (IS_ERR(block_cb))
+			return PTR_ERR(block_cb);
+
+		register_block = true;
+	} else {
+		filter_block = flow_block_cb_priv(block_cb);
+	}
+
+	flow_block_cb_incref(block_cb);
+	err = dpaa2_switch_port_block_bind(port_priv, filter_block);
+	if (err)
+		goto err_block_bind;
+
+	if (register_block) {
+		flow_block_cb_add(block_cb, f);
+		list_add_tail(&block_cb->driver_list,
+			      &dpaa2_switch_block_cb_list);
+	}
+
+	return 0;
+
+err_block_bind:
+	if (!flow_block_cb_decref(block_cb))
+		flow_block_cb_free(block_cb);
+	return err;
+}
+
+static void dpaa2_switch_setup_tc_block_unbind(struct net_device *netdev,
+					       struct flow_block_offload *f)
+{
+	struct ethsw_port_priv *port_priv = netdev_priv(netdev);
+	struct ethsw_core *ethsw = port_priv->ethsw_data;
+	struct dpaa2_switch_filter_block *filter_block;
+	struct flow_block_cb *block_cb;
+	int err;
+
+	block_cb = flow_block_cb_lookup(f->block,
+					dpaa2_switch_port_setup_tc_block_cb_ig,
+					ethsw);
+	if (!block_cb)
+		return;
+
+	filter_block = flow_block_cb_priv(block_cb);
+	err = dpaa2_switch_port_block_unbind(port_priv, filter_block);
+	if (!err && !flow_block_cb_decref(block_cb)) {
+		flow_block_cb_remove(block_cb, f);
+		list_del(&block_cb->driver_list);
+	}
+}
+
+static int dpaa2_switch_setup_tc_block(struct net_device *netdev,
+				       struct flow_block_offload *f)
+{
+	if (f->binder_type != FLOW_BLOCK_BINDER_TYPE_CLSACT_INGRESS)
+		return -EOPNOTSUPP;
+
+	f->driver_block_list = &dpaa2_switch_block_cb_list;
+
+	switch (f->command) {
+	case FLOW_BLOCK_BIND:
+		return dpaa2_switch_setup_tc_block_bind(netdev, f);
+	case FLOW_BLOCK_UNBIND:
+		dpaa2_switch_setup_tc_block_unbind(netdev, f);
+		return 0;
+	default:
+		return -EOPNOTSUPP;
+	}
+}
+
+static int dpaa2_switch_port_setup_tc(struct net_device *netdev,
+				      enum tc_setup_type type,
+				      void *type_data)
+{
+	switch (type) {
+	case TC_SETUP_BLOCK: {
+		return dpaa2_switch_setup_tc_block(netdev, type_data);
+	}
+	default:
+		return -EOPNOTSUPP;
+	}
+
+	return 0;
+}
+
+static const struct net_device_ops dpaa2_switch_port_ops = {
+	.ndo_open		= dpaa2_switch_port_open,
+	.ndo_stop		= dpaa2_switch_port_stop,
+
+	.ndo_set_mac_address	= eth_mac_addr,
+	.ndo_get_stats64	= dpaa2_switch_port_get_stats,
+	.ndo_change_mtu		= dpaa2_switch_port_change_mtu,
+	.ndo_has_offload_stats	= dpaa2_switch_port_has_offload_stats,
+	.ndo_get_offload_stats	= dpaa2_switch_port_get_offload_stats,
+	.ndo_fdb_dump		= dpaa2_switch_port_fdb_dump,
+	.ndo_vlan_rx_add_vid	= dpaa2_switch_port_vlan_add,
+	.ndo_vlan_rx_kill_vid	= dpaa2_switch_port_vlan_kill,
+
+	.ndo_start_xmit		= dpaa2_switch_port_tx,
+	.ndo_get_port_parent_id	= dpaa2_switch_port_parent_id,
+	.ndo_get_phys_port_name = dpaa2_switch_port_get_phys_name,
+	.ndo_setup_tc		= dpaa2_switch_port_setup_tc,
+};
+
+bool dpaa2_switch_port_dev_check(const struct net_device *netdev)
+{
+	return netdev->netdev_ops == &dpaa2_switch_port_ops;
+}
+
+static int dpaa2_switch_port_connect_mac(struct ethsw_port_priv *port_priv)
+{
+	struct fsl_mc_device *dpsw_port_dev, *dpmac_dev;
+	struct dpaa2_mac *mac;
+	int err;
+
+	dpsw_port_dev = to_fsl_mc_device(port_priv->netdev->dev.parent);
+	dpmac_dev = fsl_mc_get_endpoint(dpsw_port_dev, port_priv->idx);
+
+	if (PTR_ERR(dpmac_dev) == -EPROBE_DEFER)
+		return PTR_ERR(dpmac_dev);
+
+	if (IS_ERR(dpmac_dev))
+		return 0;
+
+	if (dpmac_dev->dev.type != &fsl_mc_bus_dpmac_type) {
+		err = 0;
+		goto out_put_device;
+	}
+
+	mac = kzalloc(sizeof(*mac), GFP_KERNEL);
+	if (!mac) {
+		err = -ENOMEM;
+		goto out_put_device;
+	}
+
+	mac->mc_dev = dpmac_dev;
+	mac->mc_io = port_priv->ethsw_data->mc_io;
+	mac->net_dev = port_priv->netdev;
+
+	err = dpaa2_mac_open(mac);
+	if (err)
+		goto err_free_mac;
+
+	if (dpaa2_mac_is_type_phy(mac)) {
+		err = dpaa2_mac_connect(mac);
+		if (err) {
+			netdev_err(port_priv->netdev,
+				   "Error connecting to the MAC endpoint %pe\n",
+				   ERR_PTR(err));
+			goto err_close_mac;
+		}
+	}
+
+	mutex_lock(&port_priv->mac_lock);
+	port_priv->mac = mac;
+	mutex_unlock(&port_priv->mac_lock);
+
+	return 0;
+
+err_close_mac:
+	dpaa2_mac_close(mac);
+err_free_mac:
+	kfree(mac);
+out_put_device:
+	put_device(&dpmac_dev->dev);
+	return err;
+}
+
+static void dpaa2_switch_port_disconnect_mac(struct ethsw_port_priv *port_priv)
+{
+	struct dpaa2_mac *mac;
+
+	mutex_lock(&port_priv->mac_lock);
+	mac = port_priv->mac;
+	port_priv->mac = NULL;
+	mutex_unlock(&port_priv->mac_lock);
+
+	if (!mac)
+		return;
+
+	if (dpaa2_mac_is_type_phy(mac))
+		dpaa2_mac_disconnect(mac);
+
+	dpaa2_mac_close(mac);
+	kfree(mac);
+}
+
+static irqreturn_t dpaa2_switch_irq0_handler_thread(int irq_num, void *arg)
+{
+	struct device *dev = (struct device *)arg;
+	struct ethsw_core *ethsw = dev_get_drvdata(dev);
+	struct ethsw_port_priv *port_priv;
+	int err, if_id;
+	bool had_mac;
+	u32 status;
+
+	err = dpsw_get_irq_status(ethsw->mc_io, 0, ethsw->dpsw_handle,
+				  DPSW_IRQ_INDEX_IF, &status);
+	if (err) {
+		dev_err(dev, "Can't get irq status (err %d)\n", err);
+		goto out;
+	}
+
+	if_id = (status & 0xFFFF0000) >> 16;
+	port_priv = ethsw->ports[if_id];
+
+	if (status & DPSW_IRQ_EVENT_LINK_CHANGED)
+		dpaa2_switch_port_link_state_update(port_priv->netdev);
+
+	if (status & DPSW_IRQ_EVENT_ENDPOINT_CHANGED) {
+		dpaa2_switch_port_set_mac_addr(port_priv);
+		/* We can avoid locking because the "endpoint changed" IRQ
+		 * handler is the only one who changes priv->mac at runtime,
+		 * so we are not racing with anyone.
+		 */
+		had_mac = !!port_priv->mac;
+		if (had_mac)
+			dpaa2_switch_port_disconnect_mac(port_priv);
+		else
+			dpaa2_switch_port_connect_mac(port_priv);
+	}
+
+	err = dpsw_clear_irq_status(ethsw->mc_io, 0, ethsw->dpsw_handle,
+				    DPSW_IRQ_INDEX_IF, status);
+	if (err)
+		dev_err(dev, "Can't clear irq status (err %d)\n", err);
+
+out:
+	return IRQ_HANDLED;
+}
+
+static int dpaa2_switch_setup_irqs(struct fsl_mc_device *sw_dev)
+{
+	u32 mask = DPSW_IRQ_EVENT_LINK_CHANGED | DPSW_IRQ_EVENT_ENDPOINT_CHANGED;
+	struct device *dev = &sw_dev->dev;
+	struct ethsw_core *ethsw = dev_get_drvdata(dev);
+	struct fsl_mc_device_irq *irq;
+	int err;
+
+	err = fsl_mc_allocate_irqs(sw_dev);
+	if (err) {
+		dev_err(dev, "MC irqs allocation failed\n");
+		return err;
+	}
+
+	if (WARN_ON(sw_dev->obj_desc.irq_count != DPSW_IRQ_NUM)) {
+		err = -EINVAL;
+		goto free_irq;
+	}
+
+	err = dpsw_set_irq_enable(ethsw->mc_io, 0, ethsw->dpsw_handle,
+				  DPSW_IRQ_INDEX_IF, 0);
+	if (err) {
+		dev_err(dev, "dpsw_set_irq_enable err %d\n", err);
+		goto free_irq;
+	}
+
+	irq = sw_dev->irqs[DPSW_IRQ_INDEX_IF];
+
+	err = devm_request_threaded_irq(dev, irq->virq, NULL,
+					dpaa2_switch_irq0_handler_thread,
+					IRQF_NO_SUSPEND | IRQF_ONESHOT,
+					dev_name(dev), dev);
+	if (err) {
+		dev_err(dev, "devm_request_threaded_irq(): %d\n", err);
+		goto free_irq;
+	}
+
+	err = dpsw_set_irq_mask(ethsw->mc_io, 0, ethsw->dpsw_handle,
+				DPSW_IRQ_INDEX_IF, mask);
+	if (err) {
+		dev_err(dev, "dpsw_set_irq_mask(): %d\n", err);
+		goto free_devm_irq;
+	}
+
+	err = dpsw_set_irq_enable(ethsw->mc_io, 0, ethsw->dpsw_handle,
+				  DPSW_IRQ_INDEX_IF, 1);
+	if (err) {
+		dev_err(dev, "dpsw_set_irq_enable(): %d\n", err);
+		goto free_devm_irq;
+	}
+
+	return 0;
+
+free_devm_irq:
+	devm_free_irq(dev, irq->virq, dev);
+free_irq:
+	fsl_mc_free_irqs(sw_dev);
+	return err;
+}
+
+static void dpaa2_switch_teardown_irqs(struct fsl_mc_device *sw_dev)
+{
+	struct device *dev = &sw_dev->dev;
+	struct ethsw_core *ethsw = dev_get_drvdata(dev);
+	int err;
+
+	err = dpsw_set_irq_enable(ethsw->mc_io, 0, ethsw->dpsw_handle,
+				  DPSW_IRQ_INDEX_IF, 0);
+	if (err)
+		dev_err(dev, "dpsw_set_irq_enable err %d\n", err);
+
+	fsl_mc_free_irqs(sw_dev);
+}
+
+static int dpaa2_switch_port_set_learning(struct ethsw_port_priv *port_priv, bool enable)
+{
+	struct ethsw_core *ethsw = port_priv->ethsw_data;
+	enum dpsw_learning_mode learn_mode;
+	int err;
+
+	if (enable)
+		learn_mode = DPSW_LEARNING_MODE_HW;
+	else
+		learn_mode = DPSW_LEARNING_MODE_DIS;
+
+	err = dpsw_if_set_learning_mode(ethsw->mc_io, 0, ethsw->dpsw_handle,
+					port_priv->idx, learn_mode);
+	if (err)
+		netdev_err(port_priv->netdev, "dpsw_if_set_learning_mode err %d\n", err);
+
+	if (!enable)
+		dpaa2_switch_port_fast_age(port_priv);
+
+	return err;
+}
+
+static int dpaa2_switch_port_attr_stp_state_set(struct net_device *netdev,
+						u8 state)
+{
+	struct ethsw_port_priv *port_priv = netdev_priv(netdev);
+	int err;
+
+	err = dpaa2_switch_port_set_stp_state(port_priv, state);
+	if (err)
+		return err;
+
+	switch (state) {
+	case BR_STATE_DISABLED:
+	case BR_STATE_BLOCKING:
+	case BR_STATE_LISTENING:
+		err = dpaa2_switch_port_set_learning(port_priv, false);
+		break;
+	case BR_STATE_LEARNING:
+	case BR_STATE_FORWARDING:
+		err = dpaa2_switch_port_set_learning(port_priv,
+						     port_priv->learn_ena);
+		break;
+	}
+
+	return err;
+}
+
+static int dpaa2_switch_port_flood(struct ethsw_port_priv *port_priv,
+				   struct switchdev_brport_flags flags)
+{
+	struct ethsw_core *ethsw = port_priv->ethsw_data;
+
+	if (flags.mask & BR_BCAST_FLOOD)
+		port_priv->bcast_flood = !!(flags.val & BR_BCAST_FLOOD);
+
+	if (flags.mask & BR_FLOOD)
+		port_priv->ucast_flood = !!(flags.val & BR_FLOOD);
+
+	return dpaa2_switch_fdb_set_egress_flood(ethsw, port_priv->fdb->fdb_id);
+}
+
+static int dpaa2_switch_port_pre_bridge_flags(struct net_device *netdev,
+					      struct switchdev_brport_flags flags,
+					      struct netlink_ext_ack *extack)
+{
+	if (flags.mask & ~(BR_LEARNING | BR_BCAST_FLOOD | BR_FLOOD |
+			   BR_MCAST_FLOOD))
+		return -EINVAL;
+
+	if (flags.mask & (BR_FLOOD | BR_MCAST_FLOOD)) {
+		bool multicast = !!(flags.val & BR_MCAST_FLOOD);
+		bool unicast = !!(flags.val & BR_FLOOD);
+
+		if (unicast != multicast) {
+			NL_SET_ERR_MSG_MOD(extack,
+					   "Cannot configure multicast flooding independently of unicast");
+			return -EINVAL;
+		}
+	}
+
+	return 0;
+}
+
+static int dpaa2_switch_port_bridge_flags(struct net_device *netdev,
+					  struct switchdev_brport_flags flags,
+					  struct netlink_ext_ack *extack)
+{
+	struct ethsw_port_priv *port_priv = netdev_priv(netdev);
+	int err;
+
+	if (flags.mask & BR_LEARNING) {
+		bool learn_ena = !!(flags.val & BR_LEARNING);
+
+		err = dpaa2_switch_port_set_learning(port_priv, learn_ena);
+		if (err)
+			return err;
+		port_priv->learn_ena = learn_ena;
+	}
+
+	if (flags.mask & (BR_BCAST_FLOOD | BR_FLOOD | BR_MCAST_FLOOD)) {
+		err = dpaa2_switch_port_flood(port_priv, flags);
+		if (err)
+			return err;
+	}
+
+	return 0;
+}
+
+static int dpaa2_switch_port_attr_set(struct net_device *netdev, const void *ctx,
+				      const struct switchdev_attr *attr,
+				      struct netlink_ext_ack *extack)
+{
+	int err = 0;
+
+	switch (attr->id) {
+	case SWITCHDEV_ATTR_ID_PORT_STP_STATE:
+		err = dpaa2_switch_port_attr_stp_state_set(netdev,
+							   attr->u.stp_state);
+		break;
+	case SWITCHDEV_ATTR_ID_BRIDGE_VLAN_FILTERING:
+		if (!attr->u.vlan_filtering) {
+			NL_SET_ERR_MSG_MOD(extack,
+					   "The DPAA2 switch does not support VLAN-unaware operation");
+			return -EOPNOTSUPP;
+		}
+		break;
+	case SWITCHDEV_ATTR_ID_PORT_PRE_BRIDGE_FLAGS:
+		err = dpaa2_switch_port_pre_bridge_flags(netdev, attr->u.brport_flags, extack);
+		break;
+	case SWITCHDEV_ATTR_ID_PORT_BRIDGE_FLAGS:
+		err = dpaa2_switch_port_bridge_flags(netdev, attr->u.brport_flags, extack);
+		break;
+	default:
+		err = -EOPNOTSUPP;
+		break;
+	}
+
+	return err;
+}
+
+int dpaa2_switch_port_vlans_add(struct net_device *netdev,
+				const struct switchdev_obj_port_vlan *vlan)
+{
+	struct ethsw_port_priv *port_priv = netdev_priv(netdev);
+	struct ethsw_core *ethsw = port_priv->ethsw_data;
+	struct dpsw_attr *attr = &ethsw->sw_attr;
+	int err = 0;
+
+	/* Make sure that the VLAN is not already configured
+	 * on the switch port
+	 */
+	if (port_priv->vlans[vlan->vid] & ETHSW_VLAN_MEMBER) {
+		netdev_err(netdev, "VLAN %d already configured\n", vlan->vid);
+		return -EEXIST;
+	}
+
+	/* Check if there is space for a new VLAN */
+	err = dpsw_get_attributes(ethsw->mc_io, 0, ethsw->dpsw_handle,
+				  &ethsw->sw_attr);
+	if (err) {
+		netdev_err(netdev, "dpsw_get_attributes err %d\n", err);
+		return err;
+	}
+	if (attr->max_vlans - attr->num_vlans < 1)
+		return -ENOSPC;
+
+	/* Check if there is space for a new VLAN */
+	err = dpsw_get_attributes(ethsw->mc_io, 0, ethsw->dpsw_handle,
+				  &ethsw->sw_attr);
+	if (err) {
+		netdev_err(netdev, "dpsw_get_attributes err %d\n", err);
+		return err;
+	}
+	if (attr->max_vlans - attr->num_vlans < 1)
+		return -ENOSPC;
+
+	if (!port_priv->ethsw_data->vlans[vlan->vid]) {
+		/* this is a new VLAN */
+		err = dpaa2_switch_add_vlan(port_priv, vlan->vid);
+		if (err)
+			return err;
+
+		port_priv->ethsw_data->vlans[vlan->vid] |= ETHSW_VLAN_GLOBAL;
+	}
+
+	return dpaa2_switch_port_add_vlan(port_priv, vlan->vid, vlan->flags);
+}
+
+static int dpaa2_switch_port_lookup_address(struct net_device *netdev, int is_uc,
+					    const unsigned char *addr)
+{
+	struct netdev_hw_addr_list *list = (is_uc) ? &netdev->uc : &netdev->mc;
+	struct netdev_hw_addr *ha;
+
+	netif_addr_lock_bh(netdev);
+	list_for_each_entry(ha, &list->list, list) {
+		if (ether_addr_equal(ha->addr, addr)) {
+			netif_addr_unlock_bh(netdev);
+			return 1;
+		}
+	}
+	netif_addr_unlock_bh(netdev);
+	return 0;
+}
+
+static int dpaa2_switch_port_mdb_add(struct net_device *netdev,
+				     const struct switchdev_obj_port_mdb *mdb)
+{
+	struct ethsw_port_priv *port_priv = netdev_priv(netdev);
+	int err;
+
+	/* Check if address is already set on this port */
+	if (dpaa2_switch_port_lookup_address(netdev, 0, mdb->addr))
+		return -EEXIST;
+
+	err = dpaa2_switch_port_fdb_add_mc(port_priv, mdb->addr);
+	if (err)
+		return err;
+
+	err = dev_mc_add(netdev, mdb->addr);
+	if (err) {
+		netdev_err(netdev, "dev_mc_add err %d\n", err);
+		dpaa2_switch_port_fdb_del_mc(port_priv, mdb->addr);
+	}
+
+	return err;
+}
+
+static int dpaa2_switch_port_obj_add(struct net_device *netdev,
+				     const struct switchdev_obj *obj)
+{
+	int err;
+
+	switch (obj->id) {
+	case SWITCHDEV_OBJ_ID_PORT_VLAN:
+		err = dpaa2_switch_port_vlans_add(netdev,
+						  SWITCHDEV_OBJ_PORT_VLAN(obj));
+		break;
+	case SWITCHDEV_OBJ_ID_PORT_MDB:
+		err = dpaa2_switch_port_mdb_add(netdev,
+						SWITCHDEV_OBJ_PORT_MDB(obj));
+		break;
+	default:
+		err = -EOPNOTSUPP;
+		break;
+	}
+
+	return err;
+}
+
+static int dpaa2_switch_port_del_vlan(struct ethsw_port_priv *port_priv, u16 vid)
+{
+	struct ethsw_core *ethsw = port_priv->ethsw_data;
+	struct net_device *netdev = port_priv->netdev;
+	struct dpsw_vlan_if_cfg vcfg;
+	int i, err;
+
+	if (!port_priv->vlans[vid])
+		return -ENOENT;
+
+	if (port_priv->vlans[vid] & ETHSW_VLAN_PVID) {
+		/* If we are deleting the PVID of a port, use VLAN 4095 instead
+		 * as we are sure that neither the bridge nor the 8021q module
+		 * will use it
+		 */
+		err = dpaa2_switch_port_set_pvid(port_priv, 4095);
+		if (err)
+			return err;
+	}
+
+	vcfg.num_ifs = 1;
+	vcfg.if_id[0] = port_priv->idx;
+	if (port_priv->vlans[vid] & ETHSW_VLAN_UNTAGGED) {
+		err = dpsw_vlan_remove_if_untagged(ethsw->mc_io, 0,
+						   ethsw->dpsw_handle,
+						   vid, &vcfg);
+		if (err) {
+			netdev_err(netdev,
+				   "dpsw_vlan_remove_if_untagged err %d\n",
+				   err);
+		}
+		port_priv->vlans[vid] &= ~ETHSW_VLAN_UNTAGGED;
+	}
+
+	if (port_priv->vlans[vid] & ETHSW_VLAN_MEMBER) {
+		err = dpsw_vlan_remove_if(ethsw->mc_io, 0, ethsw->dpsw_handle,
+					  vid, &vcfg);
+		if (err) {
+			netdev_err(netdev,
+				   "dpsw_vlan_remove_if err %d\n", err);
+			return err;
+		}
+		port_priv->vlans[vid] &= ~ETHSW_VLAN_MEMBER;
+
+		/* Delete VLAN from switch if it is no longer configured on
+		 * any port
+		 */
+		for (i = 0; i < ethsw->sw_attr.num_ifs; i++) {
+			if (ethsw->ports[i] &&
+			    ethsw->ports[i]->vlans[vid] & ETHSW_VLAN_MEMBER)
+				return 0; /* Found a port member in VID */
+		}
+
+		ethsw->vlans[vid] &= ~ETHSW_VLAN_GLOBAL;
+
+		err = dpaa2_switch_dellink(ethsw, vid);
+		if (err)
+			return err;
+	}
+
+	return 0;
+}
+
+int dpaa2_switch_port_vlans_del(struct net_device *netdev,
+				const struct switchdev_obj_port_vlan *vlan)
+{
+	struct ethsw_port_priv *port_priv = netdev_priv(netdev);
+
+	if (netif_is_bridge_master(vlan->obj.orig_dev))
+		return -EOPNOTSUPP;
+
+	return dpaa2_switch_port_del_vlan(port_priv, vlan->vid);
+}
+
+static int dpaa2_switch_port_mdb_del(struct net_device *netdev,
+				     const struct switchdev_obj_port_mdb *mdb)
+{
+	struct ethsw_port_priv *port_priv = netdev_priv(netdev);
+	int err;
+
+	if (!dpaa2_switch_port_lookup_address(netdev, 0, mdb->addr))
+		return -ENOENT;
+
+	err = dpaa2_switch_port_fdb_del_mc(port_priv, mdb->addr);
+	if (err)
+		return err;
+
+	err = dev_mc_del(netdev, mdb->addr);
+	if (err) {
+		netdev_err(netdev, "dev_mc_del err %d\n", err);
+		return err;
+	}
+
+	return err;
+}
+
+static int dpaa2_switch_port_obj_del(struct net_device *netdev,
+				     const struct switchdev_obj *obj)
+{
+	int err;
+
+	switch (obj->id) {
+	case SWITCHDEV_OBJ_ID_PORT_VLAN:
+		err = dpaa2_switch_port_vlans_del(netdev, SWITCHDEV_OBJ_PORT_VLAN(obj));
+		break;
+	case SWITCHDEV_OBJ_ID_PORT_MDB:
+		err = dpaa2_switch_port_mdb_del(netdev, SWITCHDEV_OBJ_PORT_MDB(obj));
+		break;
+	default:
+		err = -EOPNOTSUPP;
+		break;
+	}
+	return err;
+}
+
+static int dpaa2_switch_port_attr_set_event(struct net_device *netdev,
+					    struct switchdev_notifier_port_attr_info *ptr)
+{
+	int err;
+
+	err = switchdev_handle_port_attr_set(netdev, ptr,
+					     dpaa2_switch_port_dev_check,
+					     dpaa2_switch_port_attr_set);
+	return notifier_from_errno(err);
+}
+
+static int dpaa2_switch_port_bridge_join(struct net_device *netdev,
+					 struct net_device *upper_dev,
+					 struct netlink_ext_ack *extack)
+{
+	struct ethsw_port_priv *port_priv = netdev_priv(netdev);
+	struct dpaa2_switch_fdb *old_fdb = port_priv->fdb;
+	struct ethsw_core *ethsw = port_priv->ethsw_data;
+	bool learn_ena;
+	int err;
+
+	/* Delete the previously manually installed VLAN 1 */
+	err = dpaa2_switch_port_del_vlan(port_priv, 1);
+	if (err)
+		return err;
+
+	dpaa2_switch_port_set_fdb(port_priv, upper_dev);
+
+	/* Inherit the initial bridge port learning state */
+	learn_ena = br_port_flag_is_set(netdev, BR_LEARNING);
+	err = dpaa2_switch_port_set_learning(port_priv, learn_ena);
+	port_priv->learn_ena = learn_ena;
+
+	/* Setup the egress flood policy (broadcast, unknown unicast) */
+	err = dpaa2_switch_fdb_set_egress_flood(ethsw, port_priv->fdb->fdb_id);
+	if (err)
+		goto err_egress_flood;
+
+	/* Recreate the egress flood domain of the FDB that we just left. */
+	err = dpaa2_switch_fdb_set_egress_flood(ethsw, old_fdb->fdb_id);
+	if (err)
+		goto err_egress_flood;
+
+	err = switchdev_bridge_port_offload(netdev, netdev, NULL,
+					    NULL, NULL, false, extack);
+	if (err)
+		goto err_switchdev_offload;
+
+	return 0;
+
+err_switchdev_offload:
+err_egress_flood:
+	dpaa2_switch_port_set_fdb(port_priv, NULL);
+	return err;
+}
+
+static int dpaa2_switch_port_clear_rxvlan(struct net_device *vdev, int vid, void *arg)
+{
+	__be16 vlan_proto = htons(ETH_P_8021Q);
+
+	if (vdev)
+		vlan_proto = vlan_dev_vlan_proto(vdev);
+
+	return dpaa2_switch_port_vlan_kill(arg, vlan_proto, vid);
+}
+
+static int dpaa2_switch_port_restore_rxvlan(struct net_device *vdev, int vid, void *arg)
+{
+	__be16 vlan_proto = htons(ETH_P_8021Q);
+
+	if (vdev)
+		vlan_proto = vlan_dev_vlan_proto(vdev);
+
+	return dpaa2_switch_port_vlan_add(arg, vlan_proto, vid);
+}
+
+static void dpaa2_switch_port_pre_bridge_leave(struct net_device *netdev)
+{
+	switchdev_bridge_port_unoffload(netdev, NULL, NULL, NULL);
+}
+
+static int dpaa2_switch_port_bridge_leave(struct net_device *netdev)
+{
+	struct ethsw_port_priv *port_priv = netdev_priv(netdev);
+	struct dpaa2_switch_fdb *old_fdb = port_priv->fdb;
+	struct ethsw_core *ethsw = port_priv->ethsw_data;
+	int err;
+
+	/* First of all, fast age any learn FDB addresses on this switch port */
+	dpaa2_switch_port_fast_age(port_priv);
+
+	/* Clear all RX VLANs installed through vlan_vid_add() either as VLAN
+	 * upper devices or otherwise from the FDB table that we are about to
+	 * leave
+	 */
+	err = vlan_for_each(netdev, dpaa2_switch_port_clear_rxvlan, netdev);
+	if (err)
+		netdev_err(netdev, "Unable to clear RX VLANs from old FDB table, err (%d)\n", err);
+
+	dpaa2_switch_port_set_fdb(port_priv, NULL);
+
+	/* Restore all RX VLANs into the new FDB table that we just joined */
+	err = vlan_for_each(netdev, dpaa2_switch_port_restore_rxvlan, netdev);
+	if (err)
+		netdev_err(netdev, "Unable to restore RX VLANs to the new FDB, err (%d)\n", err);
+
+	/* Reset the flooding state to denote that this port can send any
+	 * packet in standalone mode. With this, we are also ensuring that any
+	 * later bridge join will have the flooding flag on.
+	 */
+	port_priv->bcast_flood = true;
+	port_priv->ucast_flood = true;
+
+	/* Setup the egress flood policy (broadcast, unknown unicast).
+	 * When the port is not under a bridge, only the CTRL interface is part
+	 * of the flooding domain besides the actual port
+	 */
+	err = dpaa2_switch_fdb_set_egress_flood(ethsw, port_priv->fdb->fdb_id);
+	if (err)
+		return err;
+
+	/* Recreate the egress flood domain of the FDB that we just left */
+	err = dpaa2_switch_fdb_set_egress_flood(ethsw, old_fdb->fdb_id);
+	if (err)
+		return err;
+
+	/* No HW learning when not under a bridge */
+	err = dpaa2_switch_port_set_learning(port_priv, false);
+	if (err)
+		return err;
+	port_priv->learn_ena = false;
+
+	/* Add the VLAN 1 as PVID when not under a bridge. We need this since
+	 * the dpaa2 switch interfaces are not capable to be VLAN unaware
+	 */
+	return dpaa2_switch_port_add_vlan(port_priv, DEFAULT_VLAN_ID,
+					  BRIDGE_VLAN_INFO_UNTAGGED | BRIDGE_VLAN_INFO_PVID);
+}
+
+static int dpaa2_switch_prevent_bridging_with_8021q_upper(struct net_device *netdev)
+{
+	struct net_device *upper_dev;
+	struct list_head *iter;
+
+	/* RCU read lock not necessary because we have write-side protection
+	 * (rtnl_mutex), however a non-rcu iterator does not exist.
+	 */
+	netdev_for_each_upper_dev_rcu(netdev, upper_dev, iter)
+		if (is_vlan_dev(upper_dev))
+			return -EOPNOTSUPP;
+
+	return 0;
+}
+
+static int
+dpaa2_switch_prechangeupper_sanity_checks(struct net_device *netdev,
+					  struct net_device *upper_dev,
+					  struct netlink_ext_ack *extack)
+{
+	struct ethsw_port_priv *port_priv = netdev_priv(netdev);
+	struct ethsw_port_priv *other_port_priv;
+	struct net_device *other_dev;
+	struct list_head *iter;
+	int err;
+
+	if (!br_vlan_enabled(upper_dev)) {
+		NL_SET_ERR_MSG_MOD(extack, "Cannot join a VLAN-unaware bridge");
+		return -EOPNOTSUPP;
+	}
+
+	err = dpaa2_switch_prevent_bridging_with_8021q_upper(netdev);
+	if (err) {
+		NL_SET_ERR_MSG_MOD(extack,
+				   "Cannot join a bridge while VLAN uppers are present");
+		return 0;
+	}
+
+	netdev_for_each_lower_dev(upper_dev, other_dev, iter) {
+		if (!dpaa2_switch_port_dev_check(other_dev))
+			continue;
+
+		other_port_priv = netdev_priv(other_dev);
+		if (other_port_priv->ethsw_data != port_priv->ethsw_data) {
+			NL_SET_ERR_MSG_MOD(extack,
+					   "Interface from a different DPSW is in the bridge already");
+			return -EINVAL;
+		}
+	}
+
+	return 0;
+}
+
+static int dpaa2_switch_port_prechangeupper(struct net_device *netdev,
+					    struct netdev_notifier_changeupper_info *info)
+{
+	struct netlink_ext_ack *extack;
+	struct net_device *upper_dev;
+	int err;
+
+	if (!dpaa2_switch_port_dev_check(netdev))
+		return 0;
+
+	extack = netdev_notifier_info_to_extack(&info->info);
+	upper_dev = info->upper_dev;
+	if (netif_is_bridge_master(upper_dev)) {
+		err = dpaa2_switch_prechangeupper_sanity_checks(netdev,
+								upper_dev,
+								extack);
+		if (err)
+			return err;
+
+		if (!info->linking)
+			dpaa2_switch_port_pre_bridge_leave(netdev);
+	}
+
+	return 0;
+}
+
+static int dpaa2_switch_port_changeupper(struct net_device *netdev,
+					 struct netdev_notifier_changeupper_info *info)
+{
+	struct netlink_ext_ack *extack;
+	struct net_device *upper_dev;
+
+	if (!dpaa2_switch_port_dev_check(netdev))
+		return 0;
+
+	extack = netdev_notifier_info_to_extack(&info->info);
+
+	upper_dev = info->upper_dev;
+	if (netif_is_bridge_master(upper_dev)) {
+		if (info->linking)
+			return dpaa2_switch_port_bridge_join(netdev,
+							     upper_dev,
+							     extack);
+		else
+			return dpaa2_switch_port_bridge_leave(netdev);
+	}
+
+	return 0;
+}
+
+static int dpaa2_switch_port_netdevice_event(struct notifier_block *nb,
+					     unsigned long event, void *ptr)
+{
+	struct net_device *netdev = netdev_notifier_info_to_dev(ptr);
+	int err = 0;
+
+	switch (event) {
+	case NETDEV_PRECHANGEUPPER:
+		err = dpaa2_switch_port_prechangeupper(netdev, ptr);
+		if (err)
+			return notifier_from_errno(err);
+
+		break;
+	case NETDEV_CHANGEUPPER:
+		err = dpaa2_switch_port_changeupper(netdev, ptr);
+		if (err)
+			return notifier_from_errno(err);
+
+		break;
+	}
+
+	return NOTIFY_DONE;
+}
+
+struct ethsw_switchdev_event_work {
+	struct work_struct work;
+	struct switchdev_notifier_fdb_info fdb_info;
+	struct net_device *dev;
+	unsigned long event;
+};
+
+static void dpaa2_switch_event_work(struct work_struct *work)
+{
+	struct ethsw_switchdev_event_work *switchdev_work =
+		container_of(work, struct ethsw_switchdev_event_work, work);
+	struct net_device *dev = switchdev_work->dev;
+	struct switchdev_notifier_fdb_info *fdb_info;
+	int err;
+
+	rtnl_lock();
+	fdb_info = &switchdev_work->fdb_info;
+
+	switch (switchdev_work->event) {
+	case SWITCHDEV_FDB_ADD_TO_DEVICE:
+		if (!fdb_info->added_by_user || fdb_info->is_local)
+			break;
+		if (is_unicast_ether_addr(fdb_info->addr))
+			err = dpaa2_switch_port_fdb_add_uc(netdev_priv(dev),
+							   fdb_info->addr);
+		else
+			err = dpaa2_switch_port_fdb_add_mc(netdev_priv(dev),
+							   fdb_info->addr);
+		if (err)
+			break;
+		fdb_info->offloaded = true;
+		call_switchdev_notifiers(SWITCHDEV_FDB_OFFLOADED, dev,
+					 &fdb_info->info, NULL);
+		break;
+	case SWITCHDEV_FDB_DEL_TO_DEVICE:
+		if (!fdb_info->added_by_user || fdb_info->is_local)
+			break;
+		if (is_unicast_ether_addr(fdb_info->addr))
+			dpaa2_switch_port_fdb_del_uc(netdev_priv(dev), fdb_info->addr);
+		else
+			dpaa2_switch_port_fdb_del_mc(netdev_priv(dev), fdb_info->addr);
+		break;
+	}
+
+	rtnl_unlock();
+	kfree(switchdev_work->fdb_info.addr);
+	kfree(switchdev_work);
+	dev_put(dev);
+}
+
+/* Called under rcu_read_lock() */
+static int dpaa2_switch_port_event(struct notifier_block *nb,
+				   unsigned long event, void *ptr)
+{
+	struct net_device *dev = switchdev_notifier_info_to_dev(ptr);
+	struct ethsw_port_priv *port_priv = netdev_priv(dev);
+	struct ethsw_switchdev_event_work *switchdev_work;
+	struct switchdev_notifier_fdb_info *fdb_info = ptr;
+	struct ethsw_core *ethsw = port_priv->ethsw_data;
+
+	if (event == SWITCHDEV_PORT_ATTR_SET)
+		return dpaa2_switch_port_attr_set_event(dev, ptr);
+
+	if (!dpaa2_switch_port_dev_check(dev))
+		return NOTIFY_DONE;
+
+	switchdev_work = kzalloc(sizeof(*switchdev_work), GFP_ATOMIC);
+	if (!switchdev_work)
+		return NOTIFY_BAD;
+
+	INIT_WORK(&switchdev_work->work, dpaa2_switch_event_work);
+	switchdev_work->dev = dev;
+	switchdev_work->event = event;
+
+	switch (event) {
+	case SWITCHDEV_FDB_ADD_TO_DEVICE:
+	case SWITCHDEV_FDB_DEL_TO_DEVICE:
+		memcpy(&switchdev_work->fdb_info, ptr,
+		       sizeof(switchdev_work->fdb_info));
+		switchdev_work->fdb_info.addr = kzalloc(ETH_ALEN, GFP_ATOMIC);
+		if (!switchdev_work->fdb_info.addr)
+			goto err_addr_alloc;
+
+		ether_addr_copy((u8 *)switchdev_work->fdb_info.addr,
+				fdb_info->addr);
+
+		/* Take a reference on the device to avoid being freed. */
+		dev_hold(dev);
+		break;
+	default:
+		kfree(switchdev_work);
+		return NOTIFY_DONE;
+	}
+
+	queue_work(ethsw->workqueue, &switchdev_work->work);
+
+	return NOTIFY_DONE;
+
+err_addr_alloc:
+	kfree(switchdev_work);
+	return NOTIFY_BAD;
+}
+
+static int dpaa2_switch_port_obj_event(unsigned long event,
+				       struct net_device *netdev,
+				       struct switchdev_notifier_port_obj_info *port_obj_info)
+{
+	int err = -EOPNOTSUPP;
+
+	if (!dpaa2_switch_port_dev_check(netdev))
+		return NOTIFY_DONE;
+
+	switch (event) {
+	case SWITCHDEV_PORT_OBJ_ADD:
+		err = dpaa2_switch_port_obj_add(netdev, port_obj_info->obj);
+		break;
+	case SWITCHDEV_PORT_OBJ_DEL:
+		err = dpaa2_switch_port_obj_del(netdev, port_obj_info->obj);
+		break;
+	}
+
+	port_obj_info->handled = true;
+	return notifier_from_errno(err);
+}
+
+static int dpaa2_switch_port_blocking_event(struct notifier_block *nb,
+					    unsigned long event, void *ptr)
+{
+	struct net_device *dev = switchdev_notifier_info_to_dev(ptr);
+
+	switch (event) {
+	case SWITCHDEV_PORT_OBJ_ADD:
+	case SWITCHDEV_PORT_OBJ_DEL:
+		return dpaa2_switch_port_obj_event(event, dev, ptr);
+	case SWITCHDEV_PORT_ATTR_SET:
+		return dpaa2_switch_port_attr_set_event(dev, ptr);
+	}
+
+	return NOTIFY_DONE;
+}
+
+/* Build a linear skb based on a single-buffer frame descriptor */
+static struct sk_buff *dpaa2_switch_build_linear_skb(struct ethsw_core *ethsw,
+						     const struct dpaa2_fd *fd)
+{
+	u16 fd_offset = dpaa2_fd_get_offset(fd);
+	dma_addr_t addr = dpaa2_fd_get_addr(fd);
+	u32 fd_length = dpaa2_fd_get_len(fd);
+	struct device *dev = ethsw->dev;
+	struct sk_buff *skb = NULL;
+	void *fd_vaddr;
+
+	fd_vaddr = dpaa2_iova_to_virt(ethsw->iommu_domain, addr);
+	dma_unmap_page(dev, addr, DPAA2_SWITCH_RX_BUF_SIZE,
+		       DMA_FROM_DEVICE);
+
+	skb = build_skb(fd_vaddr, DPAA2_SWITCH_RX_BUF_SIZE +
+			SKB_DATA_ALIGN(sizeof(struct skb_shared_info)));
+	if (unlikely(!skb)) {
+		dev_err(dev, "build_skb() failed\n");
+		return NULL;
+	}
+
+	skb_reserve(skb, fd_offset);
+	skb_put(skb, fd_length);
+
+	ethsw->buf_count--;
+
+	return skb;
+}
+
+static void dpaa2_switch_tx_conf(struct dpaa2_switch_fq *fq,
+				 const struct dpaa2_fd *fd)
+{
+	dpaa2_switch_free_fd(fq->ethsw, fd);
+}
+
+static void dpaa2_switch_rx(struct dpaa2_switch_fq *fq,
+			    const struct dpaa2_fd *fd)
+{
+	struct ethsw_core *ethsw = fq->ethsw;
+	struct ethsw_port_priv *port_priv;
+	struct net_device *netdev;
+	struct vlan_ethhdr *hdr;
+	struct sk_buff *skb;
+	u16 vlan_tci, vid;
+	int if_id, err;
+
+	/* get switch ingress interface ID */
+	if_id = upper_32_bits(dpaa2_fd_get_flc(fd)) & 0x0000FFFF;
+
+	if (if_id >= ethsw->sw_attr.num_ifs) {
+		dev_err(ethsw->dev, "Frame received from unknown interface!\n");
+		goto err_free_fd;
+	}
+	port_priv = ethsw->ports[if_id];
+	netdev = port_priv->netdev;
+
+	/* build the SKB based on the FD received */
+	if (dpaa2_fd_get_format(fd) != dpaa2_fd_single) {
+		if (net_ratelimit()) {
+			netdev_err(netdev, "Received invalid frame format\n");
+			goto err_free_fd;
+		}
+	}
+
+	skb = dpaa2_switch_build_linear_skb(ethsw, fd);
+	if (unlikely(!skb))
+		goto err_free_fd;
+
+	skb_reset_mac_header(skb);
+
+	/* Remove the VLAN header if the packet that we just received has a vid
+	 * equal to the port PVIDs. Since the dpaa2-switch can operate only in
+	 * VLAN-aware mode and no alterations are made on the packet when it's
+	 * redirected/mirrored to the control interface, we are sure that there
+	 * will always be a VLAN header present.
+	 */
+	hdr = vlan_eth_hdr(skb);
+	vid = ntohs(hdr->h_vlan_TCI) & VLAN_VID_MASK;
+	if (vid == port_priv->pvid) {
+		err = __skb_vlan_pop(skb, &vlan_tci);
+		if (err) {
+			dev_info(ethsw->dev, "__skb_vlan_pop() returned %d", err);
+			goto err_free_fd;
+		}
+	}
+
+	skb->dev = netdev;
+	skb->protocol = eth_type_trans(skb, skb->dev);
+
+	/* Setup the offload_fwd_mark only if the port is under a bridge */
+	skb->offload_fwd_mark = !!(port_priv->fdb->bridge_dev);
+
+	netif_receive_skb(skb);
+
+	return;
+
+err_free_fd:
+	dpaa2_switch_free_fd(ethsw, fd);
+}
+
+static void dpaa2_switch_detect_features(struct ethsw_core *ethsw)
+{
+	ethsw->features = 0;
+
+	if (ethsw->major > 8 || (ethsw->major == 8 && ethsw->minor >= 6))
+		ethsw->features |= ETHSW_FEATURE_MAC_ADDR;
+}
+
+static int dpaa2_switch_setup_fqs(struct ethsw_core *ethsw)
+{
+	struct dpsw_ctrl_if_attr ctrl_if_attr;
+	struct device *dev = ethsw->dev;
+	int i = 0;
+	int err;
+
+	err = dpsw_ctrl_if_get_attributes(ethsw->mc_io, 0, ethsw->dpsw_handle,
+					  &ctrl_if_attr);
+	if (err) {
+		dev_err(dev, "dpsw_ctrl_if_get_attributes() = %d\n", err);
+		return err;
+	}
+
+	ethsw->fq[i].fqid = ctrl_if_attr.rx_fqid;
+	ethsw->fq[i].ethsw = ethsw;
+	ethsw->fq[i++].type = DPSW_QUEUE_RX;
+
+	ethsw->fq[i].fqid = ctrl_if_attr.tx_err_conf_fqid;
+	ethsw->fq[i].ethsw = ethsw;
+	ethsw->fq[i++].type = DPSW_QUEUE_TX_ERR_CONF;
+
+	return 0;
+}
+
+/* Free buffers acquired from the buffer pool or which were meant to
+ * be released in the pool
+ */
+static void dpaa2_switch_free_bufs(struct ethsw_core *ethsw, u64 *buf_array, int count)
+{
+	struct device *dev = ethsw->dev;
+	void *vaddr;
+	int i;
+
+	for (i = 0; i < count; i++) {
+		vaddr = dpaa2_iova_to_virt(ethsw->iommu_domain, buf_array[i]);
+		dma_unmap_page(dev, buf_array[i], DPAA2_SWITCH_RX_BUF_SIZE,
+			       DMA_FROM_DEVICE);
+		free_pages((unsigned long)vaddr, 0);
+	}
+}
+
+/* Perform a single release command to add buffers
+ * to the specified buffer pool
+ */
+static int dpaa2_switch_add_bufs(struct ethsw_core *ethsw, u16 bpid)
+{
+	struct device *dev = ethsw->dev;
+	u64 buf_array[BUFS_PER_CMD];
+	struct page *page;
+	int retries = 0;
+	dma_addr_t addr;
+	int err;
+	int i;
+
+	for (i = 0; i < BUFS_PER_CMD; i++) {
+		/* Allocate one page for each Rx buffer. WRIOP sees
+		 * the entire page except for a tailroom reserved for
+		 * skb shared info
+		 */
+		page = dev_alloc_pages(0);
+		if (!page) {
+			dev_err(dev, "buffer allocation failed\n");
+			goto err_alloc;
+		}
+
+		addr = dma_map_page(dev, page, 0, DPAA2_SWITCH_RX_BUF_SIZE,
+				    DMA_FROM_DEVICE);
+		if (dma_mapping_error(dev, addr)) {
+			dev_err(dev, "dma_map_single() failed\n");
+			goto err_map;
+		}
+		buf_array[i] = addr;
+	}
+
+release_bufs:
+	/* In case the portal is busy, retry until successful or
+	 * max retries hit.
+	 */
+	while ((err = dpaa2_io_service_release(NULL, bpid,
+					       buf_array, i)) == -EBUSY) {
+		if (retries++ >= DPAA2_SWITCH_SWP_BUSY_RETRIES)
+			break;
+
+		cpu_relax();
+	}
+
+	/* If release command failed, clean up and bail out. */
+	if (err) {
+		dpaa2_switch_free_bufs(ethsw, buf_array, i);
+		return 0;
+	}
+
+	return i;
+
+err_map:
+	__free_pages(page, 0);
+err_alloc:
+	/* If we managed to allocate at least some buffers,
+	 * release them to hardware
+	 */
+	if (i)
+		goto release_bufs;
+
+	return 0;
+}
+
+static int dpaa2_switch_refill_bp(struct ethsw_core *ethsw)
+{
+	int *count = &ethsw->buf_count;
+	int new_count;
+	int err = 0;
+
+	if (unlikely(*count < DPAA2_ETHSW_REFILL_THRESH)) {
+		do {
+			new_count = dpaa2_switch_add_bufs(ethsw, ethsw->bpid);
+			if (unlikely(!new_count)) {
+				/* Out of memory; abort for now, we'll
+				 * try later on
+				 */
+				break;
+			}
+			*count += new_count;
+		} while (*count < DPAA2_ETHSW_NUM_BUFS);
+
+		if (unlikely(*count < DPAA2_ETHSW_NUM_BUFS))
+			err = -ENOMEM;
+	}
+
+	return err;
+}
+
+static int dpaa2_switch_seed_bp(struct ethsw_core *ethsw)
+{
+	int *count, ret, i;
+
+	for (i = 0; i < DPAA2_ETHSW_NUM_BUFS; i += BUFS_PER_CMD) {
+		ret = dpaa2_switch_add_bufs(ethsw, ethsw->bpid);
+		count = &ethsw->buf_count;
+		*count += ret;
+
+		if (unlikely(ret < BUFS_PER_CMD))
+			return -ENOMEM;
+	}
+
+	return 0;
+}
+
+static void dpaa2_switch_drain_bp(struct ethsw_core *ethsw)
+{
+	u64 buf_array[BUFS_PER_CMD];
+	int ret;
+
+	do {
+		ret = dpaa2_io_service_acquire(NULL, ethsw->bpid,
+					       buf_array, BUFS_PER_CMD);
+		if (ret < 0) {
+			dev_err(ethsw->dev,
+				"dpaa2_io_service_acquire() = %d\n", ret);
+			return;
+		}
+		dpaa2_switch_free_bufs(ethsw, buf_array, ret);
+
+	} while (ret);
+}
+
+static int dpaa2_switch_setup_dpbp(struct ethsw_core *ethsw)
+{
+	struct dpsw_ctrl_if_pools_cfg dpsw_ctrl_if_pools_cfg = { 0 };
+	struct device *dev = ethsw->dev;
+	struct fsl_mc_device *dpbp_dev;
+	struct dpbp_attr dpbp_attrs;
+	int err;
+
+	err = fsl_mc_object_allocate(to_fsl_mc_device(dev), FSL_MC_POOL_DPBP,
+				     &dpbp_dev);
+	if (err) {
+		if (err == -ENXIO)
+			err = -EPROBE_DEFER;
+		else
+			dev_err(dev, "DPBP device allocation failed\n");
+		return err;
+	}
+	ethsw->dpbp_dev = dpbp_dev;
+
+	err = dpbp_open(ethsw->mc_io, 0, dpbp_dev->obj_desc.id,
+			&dpbp_dev->mc_handle);
+	if (err) {
+		dev_err(dev, "dpbp_open() failed\n");
+		goto err_open;
+	}
+
+	err = dpbp_reset(ethsw->mc_io, 0, dpbp_dev->mc_handle);
+	if (err) {
+		dev_err(dev, "dpbp_reset() failed\n");
+		goto err_reset;
+	}
+
+	err = dpbp_enable(ethsw->mc_io, 0, dpbp_dev->mc_handle);
+	if (err) {
+		dev_err(dev, "dpbp_enable() failed\n");
+		goto err_enable;
+	}
+
+	err = dpbp_get_attributes(ethsw->mc_io, 0, dpbp_dev->mc_handle,
+				  &dpbp_attrs);
+	if (err) {
+		dev_err(dev, "dpbp_get_attributes() failed\n");
+		goto err_get_attr;
+	}
+
+	dpsw_ctrl_if_pools_cfg.num_dpbp = 1;
+	dpsw_ctrl_if_pools_cfg.pools[0].dpbp_id = dpbp_attrs.id;
+	dpsw_ctrl_if_pools_cfg.pools[0].buffer_size = DPAA2_SWITCH_RX_BUF_SIZE;
+	dpsw_ctrl_if_pools_cfg.pools[0].backup_pool = 0;
+
+	err = dpsw_ctrl_if_set_pools(ethsw->mc_io, 0, ethsw->dpsw_handle,
+				     &dpsw_ctrl_if_pools_cfg);
+	if (err) {
+		dev_err(dev, "dpsw_ctrl_if_set_pools() failed\n");
+		goto err_get_attr;
+	}
+	ethsw->bpid = dpbp_attrs.bpid;
+
+	return 0;
+
+err_get_attr:
+	dpbp_disable(ethsw->mc_io, 0, dpbp_dev->mc_handle);
+err_enable:
+err_reset:
+	dpbp_close(ethsw->mc_io, 0, dpbp_dev->mc_handle);
+err_open:
+	fsl_mc_object_free(dpbp_dev);
+	return err;
+}
+
+static void dpaa2_switch_free_dpbp(struct ethsw_core *ethsw)
+{
+	dpbp_disable(ethsw->mc_io, 0, ethsw->dpbp_dev->mc_handle);
+	dpbp_close(ethsw->mc_io, 0, ethsw->dpbp_dev->mc_handle);
+	fsl_mc_object_free(ethsw->dpbp_dev);
+}
+
+static int dpaa2_switch_alloc_rings(struct ethsw_core *ethsw)
+{
+	int i;
+
+	for (i = 0; i < DPAA2_SWITCH_RX_NUM_FQS; i++) {
+		ethsw->fq[i].store =
+			dpaa2_io_store_create(DPAA2_SWITCH_STORE_SIZE,
+					      ethsw->dev);
+		if (!ethsw->fq[i].store) {
+			dev_err(ethsw->dev, "dpaa2_io_store_create failed\n");
+			while (--i >= 0)
+				dpaa2_io_store_destroy(ethsw->fq[i].store);
+			return -ENOMEM;
+		}
+	}
+
+	return 0;
+}
+
+static void dpaa2_switch_destroy_rings(struct ethsw_core *ethsw)
+{
+	int i;
+
+	for (i = 0; i < DPAA2_SWITCH_RX_NUM_FQS; i++)
+		dpaa2_io_store_destroy(ethsw->fq[i].store);
+}
+
+static int dpaa2_switch_pull_fq(struct dpaa2_switch_fq *fq)
+{
+	int err, retries = 0;
+
+	/* Try to pull from the FQ while the portal is busy and we didn't hit
+	 * the maximum number fo retries
+	 */
+	do {
+		err = dpaa2_io_service_pull_fq(NULL, fq->fqid, fq->store);
+		cpu_relax();
+	} while (err == -EBUSY && retries++ < DPAA2_SWITCH_SWP_BUSY_RETRIES);
+
+	if (unlikely(err))
+		dev_err(fq->ethsw->dev, "dpaa2_io_service_pull err %d", err);
+
+	return err;
+}
+
+/* Consume all frames pull-dequeued into the store */
+static int dpaa2_switch_store_consume(struct dpaa2_switch_fq *fq)
+{
+	struct ethsw_core *ethsw = fq->ethsw;
+	int cleaned = 0, is_last;
+	struct dpaa2_dq *dq;
+	int retries = 0;
+
+	do {
+		/* Get the next available FD from the store */
+		dq = dpaa2_io_store_next(fq->store, &is_last);
+		if (unlikely(!dq)) {
+			if (retries++ >= DPAA2_SWITCH_SWP_BUSY_RETRIES) {
+				dev_err_once(ethsw->dev,
+					     "No valid dequeue response\n");
+				return -ETIMEDOUT;
+			}
+			continue;
+		}
+
+		if (fq->type == DPSW_QUEUE_RX)
+			dpaa2_switch_rx(fq, dpaa2_dq_fd(dq));
+		else
+			dpaa2_switch_tx_conf(fq, dpaa2_dq_fd(dq));
+		cleaned++;
+
+	} while (!is_last);
+
+	return cleaned;
+}
+
+/* NAPI poll routine */
+static int dpaa2_switch_poll(struct napi_struct *napi, int budget)
+{
+	int err, cleaned = 0, store_cleaned, work_done;
+	struct dpaa2_switch_fq *fq;
+	int retries = 0;
+
+	fq = container_of(napi, struct dpaa2_switch_fq, napi);
+
+	do {
+		err = dpaa2_switch_pull_fq(fq);
+		if (unlikely(err))
+			break;
+
+		/* Refill pool if appropriate */
+		dpaa2_switch_refill_bp(fq->ethsw);
+
+		store_cleaned = dpaa2_switch_store_consume(fq);
+		cleaned += store_cleaned;
+
+		if (cleaned >= budget) {
+			work_done = budget;
+			goto out;
+		}
+
+	} while (store_cleaned);
+
+	/* We didn't consume the entire budget, so finish napi and re-enable
+	 * data availability notifications
+	 */
+	napi_complete_done(napi, cleaned);
+	do {
+		err = dpaa2_io_service_rearm(NULL, &fq->nctx);
+		cpu_relax();
+	} while (err == -EBUSY && retries++ < DPAA2_SWITCH_SWP_BUSY_RETRIES);
+
+	work_done = max(cleaned, 1);
+out:
+
+	return work_done;
+}
+
+static void dpaa2_switch_fqdan_cb(struct dpaa2_io_notification_ctx *nctx)
+{
+	struct dpaa2_switch_fq *fq;
+
+	fq = container_of(nctx, struct dpaa2_switch_fq, nctx);
+
+	napi_schedule(&fq->napi);
+}
+
+static int dpaa2_switch_setup_dpio(struct ethsw_core *ethsw)
+{
+	struct dpsw_ctrl_if_queue_cfg queue_cfg;
+	struct dpaa2_io_notification_ctx *nctx;
+	int err, i, j;
+
+	for (i = 0; i < DPAA2_SWITCH_RX_NUM_FQS; i++) {
+		nctx = &ethsw->fq[i].nctx;
+
+		/* Register a new software context for the FQID.
+		 * By using NULL as the first parameter, we specify that we do
+		 * not care on which cpu are interrupts received for this queue
+		 */
+		nctx->is_cdan = 0;
+		nctx->id = ethsw->fq[i].fqid;
+		nctx->desired_cpu = DPAA2_IO_ANY_CPU;
+		nctx->cb = dpaa2_switch_fqdan_cb;
+		err = dpaa2_io_service_register(NULL, nctx, ethsw->dev);
+		if (err) {
+			err = -EPROBE_DEFER;
+			goto err_register;
+		}
+
+		queue_cfg.options = DPSW_CTRL_IF_QUEUE_OPT_DEST |
+				    DPSW_CTRL_IF_QUEUE_OPT_USER_CTX;
+		queue_cfg.dest_cfg.dest_type = DPSW_CTRL_IF_DEST_DPIO;
+		queue_cfg.dest_cfg.dest_id = nctx->dpio_id;
+		queue_cfg.dest_cfg.priority = 0;
+		queue_cfg.user_ctx = nctx->qman64;
+
+		err = dpsw_ctrl_if_set_queue(ethsw->mc_io, 0,
+					     ethsw->dpsw_handle,
+					     ethsw->fq[i].type,
+					     &queue_cfg);
+		if (err)
+			goto err_set_queue;
+	}
+
+	return 0;
+
+err_set_queue:
+	dpaa2_io_service_deregister(NULL, nctx, ethsw->dev);
+err_register:
+	for (j = 0; j < i; j++)
+		dpaa2_io_service_deregister(NULL, &ethsw->fq[j].nctx,
+					    ethsw->dev);
+
+	return err;
+}
+
+static void dpaa2_switch_free_dpio(struct ethsw_core *ethsw)
+{
+	int i;
+
+	for (i = 0; i < DPAA2_SWITCH_RX_NUM_FQS; i++)
+		dpaa2_io_service_deregister(NULL, &ethsw->fq[i].nctx,
+					    ethsw->dev);
+}
+
+static int dpaa2_switch_ctrl_if_setup(struct ethsw_core *ethsw)
+{
+	int err;
+
+	/* setup FQs for Rx and Tx Conf */
+	err = dpaa2_switch_setup_fqs(ethsw);
+	if (err)
+		return err;
+
+	/* setup the buffer pool needed on the Rx path */
+	err = dpaa2_switch_setup_dpbp(ethsw);
+	if (err)
+		return err;
+
+	err = dpaa2_switch_alloc_rings(ethsw);
+	if (err)
+		goto err_free_dpbp;
+
+	err = dpaa2_switch_setup_dpio(ethsw);
+	if (err)
+		goto err_destroy_rings;
+
+	err = dpaa2_switch_seed_bp(ethsw);
+	if (err)
+		goto err_deregister_dpio;
+
+	err = dpsw_ctrl_if_enable(ethsw->mc_io, 0, ethsw->dpsw_handle);
+	if (err) {
+		dev_err(ethsw->dev, "dpsw_ctrl_if_enable err %d\n", err);
+		goto err_drain_dpbp;
+	}
+
+	return 0;
+
+err_drain_dpbp:
+	dpaa2_switch_drain_bp(ethsw);
+err_deregister_dpio:
+	dpaa2_switch_free_dpio(ethsw);
+err_destroy_rings:
+	dpaa2_switch_destroy_rings(ethsw);
+err_free_dpbp:
+	dpaa2_switch_free_dpbp(ethsw);
+
+	return err;
+}
+
+static void dpaa2_switch_remove_port(struct ethsw_core *ethsw,
+				     u16 port_idx)
+{
+	struct ethsw_port_priv *port_priv = ethsw->ports[port_idx];
+
+	dpaa2_switch_port_disconnect_mac(port_priv);
+	free_netdev(port_priv->netdev);
+	ethsw->ports[port_idx] = NULL;
+}
+
+static int dpaa2_switch_init(struct fsl_mc_device *sw_dev)
+{
+	struct device *dev = &sw_dev->dev;
+	struct ethsw_core *ethsw = dev_get_drvdata(dev);
+	struct dpsw_vlan_if_cfg vcfg = {0};
+	struct dpsw_tci_cfg tci_cfg = {0};
+	struct dpsw_stp_cfg stp_cfg;
+	int err;
+	u16 i;
+
+	ethsw->dev_id = sw_dev->obj_desc.id;
+
+	err = dpsw_open(ethsw->mc_io, 0, ethsw->dev_id, &ethsw->dpsw_handle);
+	if (err) {
+		dev_err(dev, "dpsw_open err %d\n", err);
+		return err;
+	}
+
+	err = dpsw_get_attributes(ethsw->mc_io, 0, ethsw->dpsw_handle,
+				  &ethsw->sw_attr);
+	if (err) {
+		dev_err(dev, "dpsw_get_attributes err %d\n", err);
+		goto err_close;
+	}
+
+	err = dpsw_get_api_version(ethsw->mc_io, 0,
+				   &ethsw->major,
+				   &ethsw->minor);
+	if (err) {
+		dev_err(dev, "dpsw_get_api_version err %d\n", err);
+		goto err_close;
+	}
+
+	/* Minimum supported DPSW version check */
+	if (ethsw->major < DPSW_MIN_VER_MAJOR ||
+	    (ethsw->major == DPSW_MIN_VER_MAJOR &&
+	     ethsw->minor < DPSW_MIN_VER_MINOR)) {
+		dev_err(dev, "DPSW version %d:%d not supported. Use firmware 10.28.0 or greater.\n",
+			ethsw->major, ethsw->minor);
+		err = -EOPNOTSUPP;
+		goto err_close;
+	}
+
+	if (!dpaa2_switch_supports_cpu_traffic(ethsw)) {
+		err = -EOPNOTSUPP;
+		goto err_close;
+	}
+
+	dpaa2_switch_detect_features(ethsw);
+
+	err = dpsw_reset(ethsw->mc_io, 0, ethsw->dpsw_handle);
+	if (err) {
+		dev_err(dev, "dpsw_reset err %d\n", err);
+		goto err_close;
+	}
+
+	stp_cfg.vlan_id = DEFAULT_VLAN_ID;
+	stp_cfg.state = DPSW_STP_STATE_FORWARDING;
+
+	for (i = 0; i < ethsw->sw_attr.num_ifs; i++) {
+		err = dpsw_if_disable(ethsw->mc_io, 0, ethsw->dpsw_handle, i);
+		if (err) {
+			dev_err(dev, "dpsw_if_disable err %d\n", err);
+			goto err_close;
+		}
+
+		err = dpsw_if_set_stp(ethsw->mc_io, 0, ethsw->dpsw_handle, i,
+				      &stp_cfg);
+		if (err) {
+			dev_err(dev, "dpsw_if_set_stp err %d for port %d\n",
+				err, i);
+			goto err_close;
+		}
+
+		/* Switch starts with all ports configured to VLAN 1. Need to
+		 * remove this setting to allow configuration at bridge join
+		 */
+		vcfg.num_ifs = 1;
+		vcfg.if_id[0] = i;
+		err = dpsw_vlan_remove_if_untagged(ethsw->mc_io, 0, ethsw->dpsw_handle,
+						   DEFAULT_VLAN_ID, &vcfg);
+		if (err) {
+			dev_err(dev, "dpsw_vlan_remove_if_untagged err %d\n",
+				err);
+			goto err_close;
+		}
+
+		tci_cfg.vlan_id = 4095;
+		err = dpsw_if_set_tci(ethsw->mc_io, 0, ethsw->dpsw_handle, i, &tci_cfg);
+		if (err) {
+			dev_err(dev, "dpsw_if_set_tci err %d\n", err);
+			goto err_close;
+		}
+
+		err = dpsw_vlan_remove_if(ethsw->mc_io, 0, ethsw->dpsw_handle,
+					  DEFAULT_VLAN_ID, &vcfg);
+		if (err) {
+			dev_err(dev, "dpsw_vlan_remove_if err %d\n", err);
+			goto err_close;
+		}
+	}
+
+	err = dpsw_vlan_remove(ethsw->mc_io, 0, ethsw->dpsw_handle, DEFAULT_VLAN_ID);
+	if (err) {
+		dev_err(dev, "dpsw_vlan_remove err %d\n", err);
+		goto err_close;
+	}
+
+	ethsw->workqueue = alloc_ordered_workqueue("%s_%d_ordered",
+						   WQ_MEM_RECLAIM, "ethsw",
+						   ethsw->sw_attr.id);
+	if (!ethsw->workqueue) {
+		err = -ENOMEM;
+		goto err_close;
+	}
+
+	err = dpsw_fdb_remove(ethsw->mc_io, 0, ethsw->dpsw_handle, 0);
+	if (err)
+		goto err_destroy_ordered_workqueue;
+
+	err = dpaa2_switch_ctrl_if_setup(ethsw);
+	if (err)
+		goto err_destroy_ordered_workqueue;
+
+	return 0;
+
+err_destroy_ordered_workqueue:
+	destroy_workqueue(ethsw->workqueue);
+
+err_close:
+	dpsw_close(ethsw->mc_io, 0, ethsw->dpsw_handle);
+	return err;
+}
+
+/* Add an ACL to redirect frames with specific destination MAC address to
+ * control interface
+ */
+static int dpaa2_switch_port_trap_mac_addr(struct ethsw_port_priv *port_priv,
+					   const char *mac)
+{
+	struct dpaa2_switch_acl_entry acl_entry = {0};
+
+	/* Match on the destination MAC address */
+	ether_addr_copy(acl_entry.key.match.l2_dest_mac, mac);
+	eth_broadcast_addr(acl_entry.key.mask.l2_dest_mac);
+
+	/* Trap to CPU */
+	acl_entry.cfg.precedence = 0;
+	acl_entry.cfg.result.action = DPSW_ACL_ACTION_REDIRECT_TO_CTRL_IF;
+
+	return dpaa2_switch_acl_entry_add(port_priv->filter_block, &acl_entry);
+}
+
+static int dpaa2_switch_port_init(struct ethsw_port_priv *port_priv, u16 port)
+{
+	const char stpa[ETH_ALEN] = {0x01, 0x80, 0xc2, 0x00, 0x00, 0x00};
+	struct switchdev_obj_port_vlan vlan = {
+		.obj.id = SWITCHDEV_OBJ_ID_PORT_VLAN,
+		.vid = DEFAULT_VLAN_ID,
+		.flags = BRIDGE_VLAN_INFO_UNTAGGED | BRIDGE_VLAN_INFO_PVID,
+	};
+	struct net_device *netdev = port_priv->netdev;
+	struct ethsw_core *ethsw = port_priv->ethsw_data;
+	struct dpaa2_switch_filter_block *filter_block;
+	struct dpsw_fdb_cfg fdb_cfg = {0};
+	struct dpsw_if_attr dpsw_if_attr;
+	struct dpaa2_switch_fdb *fdb;
+	struct dpsw_acl_cfg acl_cfg;
+	u16 fdb_id, acl_tbl_id;
+	int err;
+
+	/* Get the Tx queue for this specific port */
+	err = dpsw_if_get_attributes(ethsw->mc_io, 0, ethsw->dpsw_handle,
+				     port_priv->idx, &dpsw_if_attr);
+	if (err) {
+		netdev_err(netdev, "dpsw_if_get_attributes err %d\n", err);
+		return err;
+	}
+	port_priv->tx_qdid = dpsw_if_attr.qdid;
+
+	/* Create a FDB table for this particular switch port */
+	fdb_cfg.num_fdb_entries = ethsw->sw_attr.max_fdb_entries / ethsw->sw_attr.num_ifs;
+	err = dpsw_fdb_add(ethsw->mc_io, 0, ethsw->dpsw_handle,
+			   &fdb_id, &fdb_cfg);
+	if (err) {
+		netdev_err(netdev, "dpsw_fdb_add err %d\n", err);
+		return err;
+	}
+
+	/* Find an unused dpaa2_switch_fdb structure and use it */
+	fdb = dpaa2_switch_fdb_get_unused(ethsw);
+	fdb->fdb_id = fdb_id;
+	fdb->in_use = true;
+	fdb->bridge_dev = NULL;
+	port_priv->fdb = fdb;
+
+	/* We need to add VLAN 1 as the PVID on this port until it is under a
+	 * bridge since the DPAA2 switch is not able to handle the traffic in a
+	 * VLAN unaware fashion
+	 */
+	err = dpaa2_switch_port_vlans_add(netdev, &vlan);
+	if (err)
+		return err;
+
+	/* Setup the egress flooding domains (broadcast, unknown unicast */
+	err = dpaa2_switch_fdb_set_egress_flood(ethsw, port_priv->fdb->fdb_id);
+	if (err)
+		return err;
+
+	/* Create an ACL table to be used by this switch port */
+	acl_cfg.max_entries = DPAA2_ETHSW_PORT_MAX_ACL_ENTRIES;
+	err = dpsw_acl_add(ethsw->mc_io, 0, ethsw->dpsw_handle,
+			   &acl_tbl_id, &acl_cfg);
+	if (err) {
+		netdev_err(netdev, "dpsw_acl_add err %d\n", err);
+		return err;
+	}
+
+	filter_block = dpaa2_switch_filter_block_get_unused(ethsw);
+	filter_block->ethsw = ethsw;
+	filter_block->acl_id = acl_tbl_id;
+	filter_block->in_use = true;
+	filter_block->num_acl_rules = 0;
+	INIT_LIST_HEAD(&filter_block->acl_entries);
+	INIT_LIST_HEAD(&filter_block->mirror_entries);
+
+	err = dpaa2_switch_port_acl_tbl_bind(port_priv, filter_block);
+	if (err)
+		return err;
+
+	err = dpaa2_switch_port_trap_mac_addr(port_priv, stpa);
+	if (err)
+		return err;
+
+	return err;
+}
+
+static void dpaa2_switch_ctrl_if_teardown(struct ethsw_core *ethsw)
+{
+	dpsw_ctrl_if_disable(ethsw->mc_io, 0, ethsw->dpsw_handle);
+	dpaa2_switch_free_dpio(ethsw);
+	dpaa2_switch_destroy_rings(ethsw);
+	dpaa2_switch_drain_bp(ethsw);
+	dpaa2_switch_free_dpbp(ethsw);
+}
+
+static void dpaa2_switch_teardown(struct fsl_mc_device *sw_dev)
+{
+	struct device *dev = &sw_dev->dev;
+	struct ethsw_core *ethsw = dev_get_drvdata(dev);
+	int err;
+
+	dpaa2_switch_ctrl_if_teardown(ethsw);
+
+	destroy_workqueue(ethsw->workqueue);
+
+	err = dpsw_close(ethsw->mc_io, 0, ethsw->dpsw_handle);
+	if (err)
+		dev_warn(dev, "dpsw_close err %d\n", err);
+}
+
+static void dpaa2_switch_remove(struct fsl_mc_device *sw_dev)
+{
+	struct ethsw_port_priv *port_priv;
+	struct ethsw_core *ethsw;
+	struct device *dev;
+	int i;
+
+	dev = &sw_dev->dev;
+	ethsw = dev_get_drvdata(dev);
+
+	dpaa2_switch_teardown_irqs(sw_dev);
+
+	dpsw_disable(ethsw->mc_io, 0, ethsw->dpsw_handle);
+
+	for (i = 0; i < ethsw->sw_attr.num_ifs; i++) {
+		port_priv = ethsw->ports[i];
+		unregister_netdev(port_priv->netdev);
+		dpaa2_switch_remove_port(ethsw, i);
+	}
+
+	kfree(ethsw->fdbs);
+	kfree(ethsw->filter_blocks);
+	kfree(ethsw->ports);
+
+	dpaa2_switch_teardown(sw_dev);
+
+	fsl_mc_portal_free(ethsw->mc_io);
+
+	kfree(ethsw);
+
+	dev_set_drvdata(dev, NULL);
+}
+
+static int dpaa2_switch_probe_port(struct ethsw_core *ethsw,
+				   u16 port_idx)
+{
+	struct ethsw_port_priv *port_priv;
+	struct device *dev = ethsw->dev;
+	struct net_device *port_netdev;
+	int err;
+
+	port_netdev = alloc_etherdev(sizeof(struct ethsw_port_priv));
+	if (!port_netdev) {
+		dev_err(dev, "alloc_etherdev error\n");
+		return -ENOMEM;
+	}
+
+	port_priv = netdev_priv(port_netdev);
+	port_priv->netdev = port_netdev;
+	port_priv->ethsw_data = ethsw;
+
+	mutex_init(&port_priv->mac_lock);
+
+	port_priv->idx = port_idx;
+	port_priv->stp_state = BR_STATE_FORWARDING;
+
+	SET_NETDEV_DEV(port_netdev, dev);
+	port_netdev->netdev_ops = &dpaa2_switch_port_ops;
+	port_netdev->ethtool_ops = &dpaa2_switch_port_ethtool_ops;
+
+	port_netdev->needed_headroom = DPAA2_SWITCH_NEEDED_HEADROOM;
+
+	port_priv->bcast_flood = true;
+	port_priv->ucast_flood = true;
+
+	/* Set MTU limits */
+	port_netdev->min_mtu = ETH_MIN_MTU;
+	port_netdev->max_mtu = ETHSW_MAX_FRAME_LENGTH;
+
+	/* Populate the private port structure so that later calls to
+	 * dpaa2_switch_port_init() can use it.
+	 */
+	ethsw->ports[port_idx] = port_priv;
+
+	/* The DPAA2 switch's ingress path depends on the VLAN table,
+	 * thus we are not able to disable VLAN filtering.
+	 */
+	port_netdev->features = NETIF_F_HW_VLAN_CTAG_FILTER |
+				NETIF_F_HW_VLAN_STAG_FILTER |
+				NETIF_F_HW_TC;
+	port_netdev->priv_flags |= IFF_LIVE_ADDR_CHANGE;
+
+	err = dpaa2_switch_port_init(port_priv, port_idx);
+	if (err)
+		goto err_port_probe;
+
+	err = dpaa2_switch_port_set_mac_addr(port_priv);
+	if (err)
+		goto err_port_probe;
+
+	err = dpaa2_switch_port_set_learning(port_priv, false);
+	if (err)
+		goto err_port_probe;
+	port_priv->learn_ena = false;
+
+	err = dpaa2_switch_port_connect_mac(port_priv);
+	if (err)
+		goto err_port_probe;
+
+	return 0;
+
+err_port_probe:
+	free_netdev(port_netdev);
+	ethsw->ports[port_idx] = NULL;
+
+	return err;
+}
+
+static int dpaa2_switch_probe(struct fsl_mc_device *sw_dev)
+{
+	struct device *dev = &sw_dev->dev;
+	struct ethsw_core *ethsw;
+	int i, err;
+
+	/* Allocate switch core*/
+	ethsw = kzalloc(sizeof(*ethsw), GFP_KERNEL);
+
+	if (!ethsw)
+		return -ENOMEM;
+
+	ethsw->dev = dev;
+	ethsw->iommu_domain = iommu_get_domain_for_dev(dev);
+	dev_set_drvdata(dev, ethsw);
+
+	err = fsl_mc_portal_allocate(sw_dev, FSL_MC_IO_ATOMIC_CONTEXT_PORTAL,
+				     &ethsw->mc_io);
+	if (err) {
+		if (err == -ENXIO)
+			err = -EPROBE_DEFER;
+		else
+			dev_err(dev, "fsl_mc_portal_allocate err %d\n", err);
+		goto err_free_drvdata;
+	}
+
+	err = dpaa2_switch_init(sw_dev);
+	if (err)
+		goto err_free_cmdport;
+
+	ethsw->ports = kcalloc(ethsw->sw_attr.num_ifs, sizeof(*ethsw->ports),
+			       GFP_KERNEL);
+	if (!(ethsw->ports)) {
+		err = -ENOMEM;
+		goto err_teardown;
+	}
+
+	ethsw->fdbs = kcalloc(ethsw->sw_attr.num_ifs, sizeof(*ethsw->fdbs),
+			      GFP_KERNEL);
+	if (!ethsw->fdbs) {
+		err = -ENOMEM;
+		goto err_free_ports;
+	}
+
+	ethsw->filter_blocks = kcalloc(ethsw->sw_attr.num_ifs,
+				       sizeof(*ethsw->filter_blocks),
+				       GFP_KERNEL);
+	if (!ethsw->filter_blocks) {
+		err = -ENOMEM;
+		goto err_free_fdbs;
+	}
+
+	for (i = 0; i < ethsw->sw_attr.num_ifs; i++) {
+		err = dpaa2_switch_probe_port(ethsw, i);
+		if (err)
+			goto err_free_netdev;
+	}
+
+	/* Add a NAPI instance for each of the Rx queues. The first port's
+	 * net_device will be associated with the instances since we do not have
+	 * different queues for each switch ports.
+	 */
+	for (i = 0; i < DPAA2_SWITCH_RX_NUM_FQS; i++)
+		netif_napi_add(ethsw->ports[0]->netdev, &ethsw->fq[i].napi,
+			       dpaa2_switch_poll);
+
+	/* Setup IRQs */
+	err = dpaa2_switch_setup_irqs(sw_dev);
+	if (err)
+		goto err_stop;
+
+	/* By convention, if the mirror port is equal to the number of switch
+	 * interfaces, then mirroring of any kind is disabled.
+	 */
+	ethsw->mirror_port =  ethsw->sw_attr.num_ifs;
+
+	/* Register the netdev only when the entire setup is done and the
+	 * switch port interfaces are ready to receive traffic
+	 */
+	for (i = 0; i < ethsw->sw_attr.num_ifs; i++) {
+		err = register_netdev(ethsw->ports[i]->netdev);
+		if (err < 0) {
+			dev_err(dev, "register_netdev error %d\n", err);
+			goto err_unregister_ports;
+		}
+	}
+
+	return 0;
+
+err_unregister_ports:
+	for (i--; i >= 0; i--)
+		unregister_netdev(ethsw->ports[i]->netdev);
+	dpaa2_switch_teardown_irqs(sw_dev);
+err_stop:
+	dpsw_disable(ethsw->mc_io, 0, ethsw->dpsw_handle);
+err_free_netdev:
+	for (i--; i >= 0; i--)
+		dpaa2_switch_remove_port(ethsw, i);
+	kfree(ethsw->filter_blocks);
+err_free_fdbs:
+	kfree(ethsw->fdbs);
+err_free_ports:
+	kfree(ethsw->ports);
+
+err_teardown:
+	dpaa2_switch_teardown(sw_dev);
+
+err_free_cmdport:
+	fsl_mc_portal_free(ethsw->mc_io);
+
+err_free_drvdata:
+	kfree(ethsw);
+	dev_set_drvdata(dev, NULL);
+
+	return err;
+}
+
+static const struct fsl_mc_device_id dpaa2_switch_match_id_table[] = {
+	{
+		.vendor = FSL_MC_VENDOR_FREESCALE,
+		.obj_type = "dpsw",
+	},
+	{ .vendor = 0x0 }
+};
+MODULE_DEVICE_TABLE(fslmc, dpaa2_switch_match_id_table);
+
+static struct fsl_mc_driver dpaa2_switch_drv = {
+	.driver = {
+		.name = KBUILD_MODNAME,
+	},
+	.probe = dpaa2_switch_probe,
+	.remove = dpaa2_switch_remove,
+	.match_id_table = dpaa2_switch_match_id_table
+};
+
+static struct notifier_block dpaa2_switch_port_nb __read_mostly = {
+	.notifier_call = dpaa2_switch_port_netdevice_event,
+};
+
+static struct notifier_block dpaa2_switch_port_switchdev_nb = {
+	.notifier_call = dpaa2_switch_port_event,
+};
+
+static struct notifier_block dpaa2_switch_port_switchdev_blocking_nb = {
+	.notifier_call = dpaa2_switch_port_blocking_event,
+};
+
+static int dpaa2_switch_register_notifiers(void)
+{
+	int err;
+
+	err = register_netdevice_notifier(&dpaa2_switch_port_nb);
+	if (err) {
+		pr_err("dpaa2-switch: failed to register net_device notifier (%d)\n", err);
+		return err;
+	}
+
+	err = register_switchdev_notifier(&dpaa2_switch_port_switchdev_nb);
+	if (err) {
+		pr_err("dpaa2-switch: failed to register switchdev notifier (%d)\n", err);
+		goto err_switchdev_nb;
+	}
+
+	err = register_switchdev_blocking_notifier(&dpaa2_switch_port_switchdev_blocking_nb);
+	if (err) {
+		pr_err("dpaa2-switch: failed to register switchdev blocking notifier (%d)\n", err);
+		goto err_switchdev_blocking_nb;
+	}
+
+	return 0;
+
+err_switchdev_blocking_nb:
+	unregister_switchdev_notifier(&dpaa2_switch_port_switchdev_nb);
+err_switchdev_nb:
+	unregister_netdevice_notifier(&dpaa2_switch_port_nb);
+
+	return err;
+}
+
+static void dpaa2_switch_unregister_notifiers(void)
+{
+	int err;
+
+	err = unregister_switchdev_blocking_notifier(&dpaa2_switch_port_switchdev_blocking_nb);
+	if (err)
+		pr_err("dpaa2-switch: failed to unregister switchdev blocking notifier (%d)\n",
+		       err);
+
+	err = unregister_switchdev_notifier(&dpaa2_switch_port_switchdev_nb);
+	if (err)
+		pr_err("dpaa2-switch: failed to unregister switchdev notifier (%d)\n", err);
+
+	err = unregister_netdevice_notifier(&dpaa2_switch_port_nb);
+	if (err)
+		pr_err("dpaa2-switch: failed to unregister net_device notifier (%d)\n", err);
+}
+
+static int __init dpaa2_switch_driver_init(void)
+{
+	int err;
+
+	err = fsl_mc_driver_register(&dpaa2_switch_drv);
+	if (err)
+		return err;
+
+	err = dpaa2_switch_register_notifiers();
+	if (err) {
+		fsl_mc_driver_unregister(&dpaa2_switch_drv);
+		return err;
+	}
+
+	return 0;
+}
+
+static void __exit dpaa2_switch_driver_exit(void)
+{
+	dpaa2_switch_unregister_notifiers();
+	fsl_mc_driver_unregister(&dpaa2_switch_drv);
+}
+
+module_init(dpaa2_switch_driver_init);
+module_exit(dpaa2_switch_driver_exit);
+
+MODULE_LICENSE("GPL v2");
+MODULE_DESCRIPTION("DPAA2 Ethernet Switch Driver");
diff --git a/drivers/net/ethernet/freescale/dpaa2/dpaa2-switch.h b/drivers/net/ethernet/freescale/dpaa2/dpaa2-switch.h
new file mode 100644
index 000000000000..42b3ca73f55d
--- /dev/null
+++ b/drivers/net/ethernet/freescale/dpaa2/dpaa2-switch.h
@@ -0,0 +1,277 @@
+/* SPDX-License-Identifier: GPL-2.0 */
+/*
+ * DPAA2 Ethernet Switch declarations
+ *
+ * Copyright 2014-2016 Freescale Semiconductor Inc.
+ * Copyright 2017-2021 NXP
+ *
+ */
+
+#ifndef __ETHSW_H
+#define __ETHSW_H
+
+#include <linux/netdevice.h>
+#include <linux/etherdevice.h>
+#include <linux/rtnetlink.h>
+#include <linux/if_vlan.h>
+#include <uapi/linux/if_bridge.h>
+#include <net/switchdev.h>
+#include <linux/if_bridge.h>
+#include <linux/fsl/mc.h>
+#include <net/pkt_cls.h>
+#include <soc/fsl/dpaa2-io.h>
+
+#include "dpaa2-mac.h"
+#include "dpsw.h"
+
+/* Number of IRQs supported */
+#define DPSW_IRQ_NUM	2
+
+/* Port is member of VLAN */
+#define ETHSW_VLAN_MEMBER	1
+/* VLAN to be treated as untagged on egress */
+#define ETHSW_VLAN_UNTAGGED	2
+/* Untagged frames will be assigned to this VLAN */
+#define ETHSW_VLAN_PVID		4
+/* VLAN configured on the switch */
+#define ETHSW_VLAN_GLOBAL	8
+
+/* Maximum Frame Length supported by HW (currently 10k) */
+#define DPAA2_MFL		(10 * 1024)
+#define ETHSW_MAX_FRAME_LENGTH	(DPAA2_MFL - VLAN_ETH_HLEN - ETH_FCS_LEN)
+#define ETHSW_L2_MAX_FRM(mtu)	((mtu) + VLAN_ETH_HLEN + ETH_FCS_LEN)
+
+#define ETHSW_FEATURE_MAC_ADDR	BIT(0)
+
+/* Number of receive queues (one RX and one TX_CONF) */
+#define DPAA2_SWITCH_RX_NUM_FQS	2
+
+/* Hardware requires alignment for ingress/egress buffer addresses */
+#define DPAA2_SWITCH_RX_BUF_RAW_SIZE	PAGE_SIZE
+#define DPAA2_SWITCH_RX_BUF_TAILROOM \
+	SKB_DATA_ALIGN(sizeof(struct skb_shared_info))
+#define DPAA2_SWITCH_RX_BUF_SIZE \
+	(DPAA2_SWITCH_RX_BUF_RAW_SIZE - DPAA2_SWITCH_RX_BUF_TAILROOM)
+
+#define DPAA2_SWITCH_STORE_SIZE 16
+
+/* Buffer management */
+#define BUFS_PER_CMD			7
+#define DPAA2_ETHSW_NUM_BUFS		(1024 * BUFS_PER_CMD)
+#define DPAA2_ETHSW_REFILL_THRESH	(DPAA2_ETHSW_NUM_BUFS * 5 / 6)
+
+/* Number of times to retry DPIO portal operations while waiting
+ * for portal to finish executing current command and become
+ * available. We want to avoid being stuck in a while loop in case
+ * hardware becomes unresponsive, but not give up too easily if
+ * the portal really is busy for valid reasons
+ */
+#define DPAA2_SWITCH_SWP_BUSY_RETRIES		1000
+
+/* Hardware annotation buffer size */
+#define DPAA2_SWITCH_HWA_SIZE			64
+/* Software annotation buffer size */
+#define DPAA2_SWITCH_SWA_SIZE			64
+
+#define DPAA2_SWITCH_TX_BUF_ALIGN		64
+
+#define DPAA2_SWITCH_TX_DATA_OFFSET \
+	(DPAA2_SWITCH_HWA_SIZE + DPAA2_SWITCH_SWA_SIZE)
+
+#define DPAA2_SWITCH_NEEDED_HEADROOM \
+	(DPAA2_SWITCH_TX_DATA_OFFSET + DPAA2_SWITCH_TX_BUF_ALIGN)
+
+#define DPAA2_ETHSW_PORT_MAX_ACL_ENTRIES	16
+#define DPAA2_ETHSW_PORT_DEFAULT_TRAPS		1
+
+#define DPAA2_ETHSW_PORT_ACL_CMD_BUF_SIZE	256
+
+extern const struct ethtool_ops dpaa2_switch_port_ethtool_ops;
+
+struct ethsw_core;
+
+struct dpaa2_switch_fq {
+	struct ethsw_core *ethsw;
+	enum dpsw_queue_type type;
+	struct dpaa2_io_store *store;
+	struct dpaa2_io_notification_ctx nctx;
+	struct napi_struct napi;
+	u32 fqid;
+};
+
+struct dpaa2_switch_fdb {
+	struct net_device	*bridge_dev;
+	u16			fdb_id;
+	bool			in_use;
+};
+
+struct dpaa2_switch_acl_entry {
+	struct list_head	list;
+	u16			prio;
+	unsigned long		cookie;
+
+	struct dpsw_acl_entry_cfg cfg;
+	struct dpsw_acl_key	key;
+};
+
+struct dpaa2_switch_mirror_entry {
+	struct list_head	list;
+	struct dpsw_reflection_cfg cfg;
+	unsigned long		cookie;
+	u16 if_id;
+};
+
+struct dpaa2_switch_filter_block {
+	struct ethsw_core	*ethsw;
+	u64			ports;
+	bool			in_use;
+
+	struct list_head	acl_entries;
+	u16			acl_id;
+	u8			num_acl_rules;
+
+	struct list_head	mirror_entries;
+};
+
+static inline bool
+dpaa2_switch_acl_tbl_is_full(struct dpaa2_switch_filter_block *filter_block)
+{
+	if ((filter_block->num_acl_rules + DPAA2_ETHSW_PORT_DEFAULT_TRAPS) >=
+	    DPAA2_ETHSW_PORT_MAX_ACL_ENTRIES)
+		return true;
+	return false;
+}
+
+/* Per port private data */
+struct ethsw_port_priv {
+	struct net_device	*netdev;
+	u16			idx;
+	struct ethsw_core	*ethsw_data;
+	u8			link_state;
+	u8			stp_state;
+
+	u8			vlans[VLAN_VID_MASK + 1];
+	u16			pvid;
+	u16			tx_qdid;
+
+	struct dpaa2_switch_fdb	*fdb;
+	bool			bcast_flood;
+	bool			ucast_flood;
+	bool			learn_ena;
+
+	struct dpaa2_switch_filter_block *filter_block;
+	struct dpaa2_mac	*mac;
+	/* Protects against changes to port_priv->mac */
+	struct mutex		mac_lock;
+};
+
+/* Switch data */
+struct ethsw_core {
+	struct device			*dev;
+	struct fsl_mc_io		*mc_io;
+	u16				dpsw_handle;
+	struct dpsw_attr		sw_attr;
+	u16				major, minor;
+	unsigned long			features;
+	int				dev_id;
+	struct ethsw_port_priv		**ports;
+	struct iommu_domain		*iommu_domain;
+
+	u8				vlans[VLAN_VID_MASK + 1];
+
+	struct workqueue_struct		*workqueue;
+
+	struct dpaa2_switch_fq		fq[DPAA2_SWITCH_RX_NUM_FQS];
+	struct fsl_mc_device		*dpbp_dev;
+	int				buf_count;
+	u16				bpid;
+	int				napi_users;
+
+	struct dpaa2_switch_fdb		*fdbs;
+	struct dpaa2_switch_filter_block *filter_blocks;
+	u16				mirror_port;
+};
+
+static inline int dpaa2_switch_get_index(struct ethsw_core *ethsw,
+					 struct net_device *netdev)
+{
+	int i;
+
+	for (i = 0; i < ethsw->sw_attr.num_ifs; i++)
+		if (ethsw->ports[i]->netdev == netdev)
+			return ethsw->ports[i]->idx;
+
+	return -EINVAL;
+}
+
+static inline bool dpaa2_switch_supports_cpu_traffic(struct ethsw_core *ethsw)
+{
+	if (ethsw->sw_attr.options & DPSW_OPT_CTRL_IF_DIS) {
+		dev_err(ethsw->dev, "Control Interface is disabled, cannot probe\n");
+		return false;
+	}
+
+	if (ethsw->sw_attr.flooding_cfg != DPSW_FLOODING_PER_FDB) {
+		dev_err(ethsw->dev, "Flooding domain is not per FDB, cannot probe\n");
+		return false;
+	}
+
+	if (ethsw->sw_attr.broadcast_cfg != DPSW_BROADCAST_PER_FDB) {
+		dev_err(ethsw->dev, "Broadcast domain is not per FDB, cannot probe\n");
+		return false;
+	}
+
+	if (ethsw->sw_attr.max_fdbs < ethsw->sw_attr.num_ifs) {
+		dev_err(ethsw->dev, "The number of FDBs is lower than the number of ports, cannot probe\n");
+		return false;
+	}
+
+	return true;
+}
+
+static inline bool
+dpaa2_switch_port_is_type_phy(struct ethsw_port_priv *port_priv)
+{
+	return dpaa2_mac_is_type_phy(port_priv->mac);
+}
+
+static inline bool dpaa2_switch_port_has_mac(struct ethsw_port_priv *port_priv)
+{
+	return port_priv->mac ? true : false;
+}
+
+bool dpaa2_switch_port_dev_check(const struct net_device *netdev);
+
+int dpaa2_switch_port_vlans_add(struct net_device *netdev,
+				const struct switchdev_obj_port_vlan *vlan);
+
+int dpaa2_switch_port_vlans_del(struct net_device *netdev,
+				const struct switchdev_obj_port_vlan *vlan);
+
+typedef int dpaa2_switch_fdb_cb_t(struct ethsw_port_priv *port_priv,
+				  struct fdb_dump_entry *fdb_entry,
+				  void *data);
+
+/* TC offload */
+
+int dpaa2_switch_cls_flower_replace(struct dpaa2_switch_filter_block *block,
+				    struct flow_cls_offload *cls);
+
+int dpaa2_switch_cls_flower_destroy(struct dpaa2_switch_filter_block *block,
+				    struct flow_cls_offload *cls);
+
+int dpaa2_switch_cls_matchall_replace(struct dpaa2_switch_filter_block *block,
+				      struct tc_cls_matchall_offload *cls);
+
+int dpaa2_switch_cls_matchall_destroy(struct dpaa2_switch_filter_block *block,
+				      struct tc_cls_matchall_offload *cls);
+
+int dpaa2_switch_acl_entry_add(struct dpaa2_switch_filter_block *block,
+			       struct dpaa2_switch_acl_entry *entry);
+
+int dpaa2_switch_block_offload_mirror(struct dpaa2_switch_filter_block *block,
+				      struct ethsw_port_priv *port_priv);
+
+int dpaa2_switch_block_unoffload_mirror(struct dpaa2_switch_filter_block *block,
+					struct ethsw_port_priv *port_priv);
+#endif	/* __ETHSW_H */
diff --git a/drivers/net/ethernet/freescale/dpaa2/dpaa2-xsk.c b/drivers/net/ethernet/freescale/dpaa2/dpaa2-xsk.c
new file mode 100644
index 000000000000..4b0ae7d9af92
--- /dev/null
+++ b/drivers/net/ethernet/freescale/dpaa2/dpaa2-xsk.c
@@ -0,0 +1,452 @@
+// SPDX-License-Identifier: (GPL-2.0+ OR BSD-3-Clause)
+/* Copyright 2022 NXP
+ */
+#include <linux/filter.h>
+#include <linux/compiler.h>
+#include <linux/bpf_trace.h>
+#include <net/xdp.h>
+#include <net/xdp_sock_drv.h>
+
+#include "dpaa2-eth.h"
+
+static void dpaa2_eth_setup_consume_func(struct dpaa2_eth_priv *priv,
+					 struct dpaa2_eth_channel *ch,
+					 enum dpaa2_eth_fq_type type,
+					 dpaa2_eth_consume_cb_t *consume)
+{
+	struct dpaa2_eth_fq *fq;
+	int i;
+
+	for (i = 0; i < priv->num_fqs; i++) {
+		fq = &priv->fq[i];
+
+		if (fq->type != type)
+			continue;
+		if (fq->channel != ch)
+			continue;
+
+		fq->consume = consume;
+	}
+}
+
+static u32 dpaa2_xsk_run_xdp(struct dpaa2_eth_priv *priv,
+			     struct dpaa2_eth_channel *ch,
+			     struct dpaa2_eth_fq *rx_fq,
+			     struct dpaa2_fd *fd, void *vaddr)
+{
+	dma_addr_t addr = dpaa2_fd_get_addr(fd);
+	struct bpf_prog *xdp_prog;
+	struct xdp_buff *xdp_buff;
+	struct dpaa2_eth_swa *swa;
+	u32 xdp_act = XDP_PASS;
+	int err;
+
+	xdp_prog = READ_ONCE(ch->xdp.prog);
+	if (!xdp_prog)
+		goto out;
+
+	swa = (struct dpaa2_eth_swa *)(vaddr + DPAA2_ETH_RX_HWA_SIZE +
+				       ch->xsk_pool->umem->headroom);
+	xdp_buff = swa->xsk.xdp_buff;
+
+	xdp_buff->data_hard_start = vaddr;
+	xdp_buff->data = vaddr + dpaa2_fd_get_offset(fd);
+	xdp_buff->data_end = xdp_buff->data + dpaa2_fd_get_len(fd);
+	xdp_set_data_meta_invalid(xdp_buff);
+	xdp_buff->rxq = &ch->xdp_rxq;
+
+	xsk_buff_dma_sync_for_cpu(xdp_buff);
+	xdp_act = bpf_prog_run_xdp(xdp_prog, xdp_buff);
+
+	/* xdp.data pointer may have changed */
+	dpaa2_fd_set_offset(fd, xdp_buff->data - vaddr);
+	dpaa2_fd_set_len(fd, xdp_buff->data_end - xdp_buff->data);
+
+	if (likely(xdp_act == XDP_REDIRECT)) {
+		err = xdp_do_redirect(priv->net_dev, xdp_buff, xdp_prog);
+		if (unlikely(err)) {
+			ch->stats.xdp_drop++;
+			dpaa2_eth_recycle_buf(priv, ch, addr);
+		} else {
+			ch->buf_count--;
+			ch->stats.xdp_redirect++;
+		}
+
+		goto xdp_redir;
+	}
+
+	switch (xdp_act) {
+	case XDP_PASS:
+		break;
+	case XDP_TX:
+		dpaa2_eth_xdp_enqueue(priv, ch, fd, vaddr, rx_fq->flowid);
+		break;
+	default:
+		bpf_warn_invalid_xdp_action(priv->net_dev, xdp_prog, xdp_act);
+		fallthrough;
+	case XDP_ABORTED:
+		trace_xdp_exception(priv->net_dev, xdp_prog, xdp_act);
+		fallthrough;
+	case XDP_DROP:
+		dpaa2_eth_recycle_buf(priv, ch, addr);
+		ch->stats.xdp_drop++;
+		break;
+	}
+
+xdp_redir:
+	ch->xdp.res |= xdp_act;
+out:
+	return xdp_act;
+}
+
+/* Rx frame processing routine for the AF_XDP fast path */
+static void dpaa2_xsk_rx(struct dpaa2_eth_priv *priv,
+			 struct dpaa2_eth_channel *ch,
+			 const struct dpaa2_fd *fd,
+			 struct dpaa2_eth_fq *fq)
+{
+	dma_addr_t addr = dpaa2_fd_get_addr(fd);
+	u8 fd_format = dpaa2_fd_get_format(fd);
+	struct rtnl_link_stats64 *percpu_stats;
+	u32 fd_length = dpaa2_fd_get_len(fd);
+	struct sk_buff *skb;
+	void *vaddr;
+	u32 xdp_act;
+
+	trace_dpaa2_rx_xsk_fd(priv->net_dev, fd);
+
+	vaddr = dpaa2_iova_to_virt(priv->iommu_domain, addr);
+	percpu_stats = this_cpu_ptr(priv->percpu_stats);
+
+	if (fd_format != dpaa2_fd_single) {
+		WARN_ON(priv->xdp_prog);
+		/* AF_XDP doesn't support any other formats */
+		goto err_frame_format;
+	}
+
+	xdp_act = dpaa2_xsk_run_xdp(priv, ch, fq, (struct dpaa2_fd *)fd, vaddr);
+	if (xdp_act != XDP_PASS) {
+		percpu_stats->rx_packets++;
+		percpu_stats->rx_bytes += dpaa2_fd_get_len(fd);
+		return;
+	}
+
+	/* Build skb */
+	skb = dpaa2_eth_alloc_skb(priv, ch, fd, fd_length, vaddr);
+	if (!skb)
+		/* Nothing else we can do, recycle the buffer and
+		 * drop the frame.
+		 */
+		goto err_alloc_skb;
+
+	/* Send the skb to the Linux networking stack */
+	dpaa2_eth_receive_skb(priv, ch, fd, vaddr, fq, percpu_stats, skb);
+
+	return;
+
+err_alloc_skb:
+	dpaa2_eth_recycle_buf(priv, ch, addr);
+err_frame_format:
+	percpu_stats->rx_dropped++;
+}
+
+static void dpaa2_xsk_set_bp_per_qdbin(struct dpaa2_eth_priv *priv,
+				       struct dpni_pools_cfg *pools_params)
+{
+	int curr_bp = 0, i, j;
+
+	pools_params->pool_options = DPNI_POOL_ASSOC_QDBIN;
+	for (i = 0; i < priv->num_bps; i++) {
+		for (j = 0; j < priv->num_channels; j++)
+			if (priv->bp[i] == priv->channel[j]->bp)
+				pools_params->pools[curr_bp].priority_mask |= (1 << j);
+		if (!pools_params->pools[curr_bp].priority_mask)
+			continue;
+
+		pools_params->pools[curr_bp].dpbp_id = priv->bp[i]->bpid;
+		pools_params->pools[curr_bp].buffer_size = priv->rx_buf_size;
+		pools_params->pools[curr_bp++].backup_pool = 0;
+	}
+	pools_params->num_dpbp = curr_bp;
+}
+
+static int dpaa2_xsk_disable_pool(struct net_device *dev, u16 qid)
+{
+	struct xsk_buff_pool *pool = xsk_get_pool_from_qid(dev, qid);
+	struct dpaa2_eth_priv *priv = netdev_priv(dev);
+	struct dpni_pools_cfg pools_params = { 0 };
+	struct dpaa2_eth_channel *ch;
+	int err;
+	bool up;
+
+	ch = priv->channel[qid];
+	if (!ch->xsk_pool)
+		return -EINVAL;
+
+	up = netif_running(dev);
+	if (up)
+		dev_close(dev);
+
+	xsk_pool_dma_unmap(pool, 0);
+	err = xdp_rxq_info_reg_mem_model(&ch->xdp_rxq,
+					 MEM_TYPE_PAGE_ORDER0, NULL);
+	if (err)
+		netdev_err(dev, "xsk_rxq_info_reg_mem_model() failed (err = %d)\n",
+			   err);
+
+	dpaa2_eth_free_dpbp(priv, ch->bp);
+
+	ch->xsk_zc = false;
+	ch->xsk_pool = NULL;
+	ch->xsk_tx_pkts_sent = 0;
+	ch->bp = priv->bp[DPAA2_ETH_DEFAULT_BP_IDX];
+
+	dpaa2_eth_setup_consume_func(priv, ch, DPAA2_RX_FQ, dpaa2_eth_rx);
+
+	dpaa2_xsk_set_bp_per_qdbin(priv, &pools_params);
+	err = dpni_set_pools(priv->mc_io, 0, priv->mc_token, &pools_params);
+	if (err)
+		netdev_err(dev, "dpni_set_pools() failed\n");
+
+	if (up) {
+		err = dev_open(dev, NULL);
+		if (err)
+			return err;
+	}
+
+	return 0;
+}
+
+static int dpaa2_xsk_enable_pool(struct net_device *dev,
+				 struct xsk_buff_pool *pool,
+				 u16 qid)
+{
+	struct dpaa2_eth_priv *priv = netdev_priv(dev);
+	struct dpni_pools_cfg pools_params = { 0 };
+	struct dpaa2_eth_channel *ch;
+	int err, err2;
+	bool up;
+
+	if (priv->dpni_attrs.wriop_version < DPAA2_WRIOP_VERSION(3, 0, 0)) {
+		netdev_err(dev, "AF_XDP zero-copy not supported on devices <= WRIOP(3, 0, 0)\n");
+		return -EOPNOTSUPP;
+	}
+
+	if (priv->dpni_attrs.num_queues > 8) {
+		netdev_err(dev, "AF_XDP zero-copy not supported on DPNI with more then 8 queues\n");
+		return -EOPNOTSUPP;
+	}
+
+	up = netif_running(dev);
+	if (up)
+		dev_close(dev);
+
+	err = xsk_pool_dma_map(pool, priv->net_dev->dev.parent, 0);
+	if (err) {
+		netdev_err(dev, "xsk_pool_dma_map() failed (err = %d)\n",
+			   err);
+		goto err_dma_unmap;
+	}
+
+	ch = priv->channel[qid];
+	err = xdp_rxq_info_reg_mem_model(&ch->xdp_rxq, MEM_TYPE_XSK_BUFF_POOL, NULL);
+	if (err) {
+		netdev_err(dev, "xdp_rxq_info_reg_mem_model() failed (err = %d)\n", err);
+		goto err_mem_model;
+	}
+	xsk_pool_set_rxq_info(pool, &ch->xdp_rxq);
+
+	priv->bp[priv->num_bps] = dpaa2_eth_allocate_dpbp(priv);
+	if (IS_ERR(priv->bp[priv->num_bps])) {
+		err = PTR_ERR(priv->bp[priv->num_bps]);
+		goto err_bp_alloc;
+	}
+	ch->xsk_zc = true;
+	ch->xsk_pool = pool;
+	ch->bp = priv->bp[priv->num_bps++];
+
+	dpaa2_eth_setup_consume_func(priv, ch, DPAA2_RX_FQ, dpaa2_xsk_rx);
+
+	dpaa2_xsk_set_bp_per_qdbin(priv, &pools_params);
+	err = dpni_set_pools(priv->mc_io, 0, priv->mc_token, &pools_params);
+	if (err) {
+		netdev_err(dev, "dpni_set_pools() failed\n");
+		goto err_set_pools;
+	}
+
+	if (up) {
+		err = dev_open(dev, NULL);
+		if (err)
+			return err;
+	}
+
+	return 0;
+
+err_set_pools:
+	err2 = dpaa2_xsk_disable_pool(dev, qid);
+	if (err2)
+		netdev_err(dev, "dpaa2_xsk_disable_pool() failed %d\n", err2);
+err_bp_alloc:
+	err2 = xdp_rxq_info_reg_mem_model(&priv->channel[qid]->xdp_rxq,
+					  MEM_TYPE_PAGE_ORDER0, NULL);
+	if (err2)
+		netdev_err(dev, "xsk_rxq_info_reg_mem_model() failed with %d)\n", err2);
+err_mem_model:
+	xsk_pool_dma_unmap(pool, 0);
+err_dma_unmap:
+	if (up)
+		dev_open(dev, NULL);
+
+	return err;
+}
+
+int dpaa2_xsk_setup_pool(struct net_device *dev, struct xsk_buff_pool *pool, u16 qid)
+{
+	return pool ? dpaa2_xsk_enable_pool(dev, pool, qid) :
+		      dpaa2_xsk_disable_pool(dev, qid);
+}
+
+int dpaa2_xsk_wakeup(struct net_device *dev, u32 qid, u32 flags)
+{
+	struct dpaa2_eth_priv *priv = netdev_priv(dev);
+	struct dpaa2_eth_channel *ch = priv->channel[qid];
+
+	if (!priv->link_state.up)
+		return -ENETDOWN;
+
+	if (!priv->xdp_prog)
+		return -EINVAL;
+
+	if (!ch->xsk_zc)
+		return -EINVAL;
+
+	/* We do not have access to a per channel SW interrupt, so instead we
+	 * schedule a NAPI instance.
+	 */
+	if (!napi_if_scheduled_mark_missed(&ch->napi))
+		napi_schedule(&ch->napi);
+
+	return 0;
+}
+
+static int dpaa2_xsk_tx_build_fd(struct dpaa2_eth_priv *priv,
+				 struct dpaa2_eth_channel *ch,
+				 struct dpaa2_fd *fd,
+				 struct xdp_desc *xdp_desc)
+{
+	struct device *dev = priv->net_dev->dev.parent;
+	struct dpaa2_sg_entry *sgt;
+	struct dpaa2_eth_swa *swa;
+	void *sgt_buf = NULL;
+	dma_addr_t sgt_addr;
+	int sgt_buf_size;
+	dma_addr_t addr;
+	int err = 0;
+
+	/* Prepare the HW SGT structure */
+	sgt_buf_size = priv->tx_data_offset + sizeof(struct dpaa2_sg_entry);
+	sgt_buf = dpaa2_eth_sgt_get(priv);
+	if (unlikely(!sgt_buf))
+		return -ENOMEM;
+	sgt = (struct dpaa2_sg_entry *)(sgt_buf + priv->tx_data_offset);
+
+	/* Get the address of the XSK Tx buffer */
+	addr = xsk_buff_raw_get_dma(ch->xsk_pool, xdp_desc->addr);
+	xsk_buff_raw_dma_sync_for_device(ch->xsk_pool, addr, xdp_desc->len);
+
+	/* Fill in the HW SGT structure */
+	dpaa2_sg_set_addr(sgt, addr);
+	dpaa2_sg_set_len(sgt, xdp_desc->len);
+	dpaa2_sg_set_final(sgt, true);
+
+	/* Store the necessary info in the SGT buffer */
+	swa = (struct dpaa2_eth_swa *)sgt_buf;
+	swa->type = DPAA2_ETH_SWA_XSK;
+	swa->xsk.sgt_size = sgt_buf_size;
+
+	/* Separately map the SGT buffer */
+	sgt_addr = dma_map_single(dev, sgt_buf, sgt_buf_size, DMA_BIDIRECTIONAL);
+	if (unlikely(dma_mapping_error(dev, sgt_addr))) {
+		err = -ENOMEM;
+		goto sgt_map_failed;
+	}
+
+	/* Initialize FD fields */
+	memset(fd, 0, sizeof(struct dpaa2_fd));
+	dpaa2_fd_set_offset(fd, priv->tx_data_offset);
+	dpaa2_fd_set_format(fd, dpaa2_fd_sg);
+	dpaa2_fd_set_addr(fd, sgt_addr);
+	dpaa2_fd_set_len(fd, xdp_desc->len);
+	dpaa2_fd_set_ctrl(fd, FD_CTRL_PTA);
+
+	return 0;
+
+sgt_map_failed:
+	dpaa2_eth_sgt_recycle(priv, sgt_buf);
+
+	return err;
+}
+
+bool dpaa2_xsk_tx(struct dpaa2_eth_priv *priv,
+		  struct dpaa2_eth_channel *ch)
+{
+	struct xdp_desc *xdp_descs = ch->xsk_pool->tx_descs;
+	struct dpaa2_eth_drv_stats *percpu_extras;
+	struct rtnl_link_stats64 *percpu_stats;
+	int budget = DPAA2_ETH_TX_ZC_PER_NAPI;
+	int total_enqueued, enqueued;
+	int retries, max_retries;
+	struct dpaa2_eth_fq *fq;
+	struct dpaa2_fd *fds;
+	int batch, i, err;
+
+	percpu_stats = this_cpu_ptr(priv->percpu_stats);
+	percpu_extras = this_cpu_ptr(priv->percpu_extras);
+	fds = (this_cpu_ptr(priv->fd))->array;
+
+	/* Use the FQ with the same idx as the affine CPU */
+	fq = &priv->fq[ch->nctx.desired_cpu];
+
+	batch = xsk_tx_peek_release_desc_batch(ch->xsk_pool, budget);
+	if (!batch)
+		return false;
+
+	/* Create a FD for each XSK frame to be sent */
+	for (i = 0; i < batch; i++) {
+		err = dpaa2_xsk_tx_build_fd(priv, ch, &fds[i], &xdp_descs[i]);
+		if (err) {
+			batch = i;
+			break;
+		}
+
+		trace_dpaa2_tx_xsk_fd(priv->net_dev, &fds[i]);
+	}
+
+	/* Enqueue all the created FDs */
+	max_retries = batch * DPAA2_ETH_ENQUEUE_RETRIES;
+	total_enqueued = 0;
+	enqueued = 0;
+	retries = 0;
+	while (total_enqueued < batch && retries < max_retries) {
+		err = priv->enqueue(priv, fq, &fds[total_enqueued], 0,
+				    batch - total_enqueued, &enqueued);
+		if (err == -EBUSY) {
+			retries++;
+			continue;
+		}
+
+		total_enqueued += enqueued;
+	}
+	percpu_extras->tx_portal_busy += retries;
+
+	/* Update statistics */
+	percpu_stats->tx_packets += total_enqueued;
+	for (i = 0; i < total_enqueued; i++)
+		percpu_stats->tx_bytes += dpaa2_fd_get_len(&fds[i]);
+	for (i = total_enqueued; i < batch; i++) {
+		dpaa2_eth_free_tx_fd(priv, ch, fq, &fds[i], false);
+		percpu_stats->tx_errors++;
+	}
+
+	return total_enqueued == budget;
+}
diff --git a/drivers/net/ethernet/freescale/dpaa2/dpkg.h b/drivers/net/ethernet/freescale/dpaa2/dpkg.h
new file mode 100644
index 000000000000..6f596a5fbeeb
--- /dev/null
+++ b/drivers/net/ethernet/freescale/dpaa2/dpkg.h
@@ -0,0 +1,481 @@
+/* SPDX-License-Identifier: (GPL-2.0+ OR BSD-3-Clause) */
+/* Copyright 2013-2015 Freescale Semiconductor Inc.
+ */
+#ifndef __FSL_DPKG_H_
+#define __FSL_DPKG_H_
+
+#include <linux/types.h>
+
+/* Data Path Key Generator API
+ * Contains initialization APIs and runtime APIs for the Key Generator
+ */
+
+/** Key Generator properties */
+
+/**
+ * DPKG_NUM_OF_MASKS - Number of masks per key extraction
+ */
+#define DPKG_NUM_OF_MASKS		4
+
+/**
+ * DPKG_MAX_NUM_OF_EXTRACTS - Number of extractions per key profile
+ */
+#define DPKG_MAX_NUM_OF_EXTRACTS	10
+
+/**
+ * enum dpkg_extract_from_hdr_type - Selecting extraction by header types
+ * @DPKG_FROM_HDR: Extract selected bytes from header, by offset
+ * @DPKG_FROM_FIELD: Extract selected bytes from header, by offset from field
+ * @DPKG_FULL_FIELD: Extract a full field
+ */
+enum dpkg_extract_from_hdr_type {
+	DPKG_FROM_HDR = 0,
+	DPKG_FROM_FIELD = 1,
+	DPKG_FULL_FIELD = 2
+};
+
+/**
+ * enum dpkg_extract_type - Enumeration for selecting extraction type
+ * @DPKG_EXTRACT_FROM_HDR: Extract from the header
+ * @DPKG_EXTRACT_FROM_DATA: Extract from data not in specific header
+ * @DPKG_EXTRACT_FROM_PARSE: Extract from parser-result;
+ *	e.g. can be used to extract header existence;
+ *	please refer to 'Parse Result definition' section in the parser BG
+ */
+enum dpkg_extract_type {
+	DPKG_EXTRACT_FROM_HDR = 0,
+	DPKG_EXTRACT_FROM_DATA = 1,
+	DPKG_EXTRACT_FROM_PARSE = 3
+};
+
+/**
+ * struct dpkg_mask - A structure for defining a single extraction mask
+ * @mask: Byte mask for the extracted content
+ * @offset: Offset within the extracted content
+ */
+struct dpkg_mask {
+	u8 mask;
+	u8 offset;
+};
+
+/* Protocol fields */
+
+/* Ethernet fields */
+#define NH_FLD_ETH_DA				BIT(0)
+#define NH_FLD_ETH_SA				BIT(1)
+#define NH_FLD_ETH_LENGTH			BIT(2)
+#define NH_FLD_ETH_TYPE				BIT(3)
+#define NH_FLD_ETH_FINAL_CKSUM			BIT(4)
+#define NH_FLD_ETH_PADDING			BIT(5)
+#define NH_FLD_ETH_ALL_FIELDS			(BIT(6) - 1)
+
+/* VLAN fields */
+#define NH_FLD_VLAN_VPRI			BIT(0)
+#define NH_FLD_VLAN_CFI				BIT(1)
+#define NH_FLD_VLAN_VID				BIT(2)
+#define NH_FLD_VLAN_LENGTH			BIT(3)
+#define NH_FLD_VLAN_TYPE			BIT(4)
+#define NH_FLD_VLAN_ALL_FIELDS			(BIT(5) - 1)
+
+#define NH_FLD_VLAN_TCI				(NH_FLD_VLAN_VPRI | \
+						 NH_FLD_VLAN_CFI | \
+						 NH_FLD_VLAN_VID)
+
+/* IP (generic) fields */
+#define NH_FLD_IP_VER				BIT(0)
+#define NH_FLD_IP_DSCP				BIT(2)
+#define NH_FLD_IP_ECN				BIT(3)
+#define NH_FLD_IP_PROTO				BIT(4)
+#define NH_FLD_IP_SRC				BIT(5)
+#define NH_FLD_IP_DST				BIT(6)
+#define NH_FLD_IP_TOS_TC			BIT(7)
+#define NH_FLD_IP_ID				BIT(8)
+#define NH_FLD_IP_ALL_FIELDS			(BIT(9) - 1)
+
+/* IPV4 fields */
+#define NH_FLD_IPV4_VER				BIT(0)
+#define NH_FLD_IPV4_HDR_LEN			BIT(1)
+#define NH_FLD_IPV4_TOS				BIT(2)
+#define NH_FLD_IPV4_TOTAL_LEN			BIT(3)
+#define NH_FLD_IPV4_ID				BIT(4)
+#define NH_FLD_IPV4_FLAG_D			BIT(5)
+#define NH_FLD_IPV4_FLAG_M			BIT(6)
+#define NH_FLD_IPV4_OFFSET			BIT(7)
+#define NH_FLD_IPV4_TTL				BIT(8)
+#define NH_FLD_IPV4_PROTO			BIT(9)
+#define NH_FLD_IPV4_CKSUM			BIT(10)
+#define NH_FLD_IPV4_SRC_IP			BIT(11)
+#define NH_FLD_IPV4_DST_IP			BIT(12)
+#define NH_FLD_IPV4_OPTS			BIT(13)
+#define NH_FLD_IPV4_OPTS_COUNT			BIT(14)
+#define NH_FLD_IPV4_ALL_FIELDS			(BIT(15) - 1)
+
+/* IPV6 fields */
+#define NH_FLD_IPV6_VER				BIT(0)
+#define NH_FLD_IPV6_TC				BIT(1)
+#define NH_FLD_IPV6_SRC_IP			BIT(2)
+#define NH_FLD_IPV6_DST_IP			BIT(3)
+#define NH_FLD_IPV6_NEXT_HDR			BIT(4)
+#define NH_FLD_IPV6_FL				BIT(5)
+#define NH_FLD_IPV6_HOP_LIMIT			BIT(6)
+#define NH_FLD_IPV6_ID				BIT(7)
+#define NH_FLD_IPV6_ALL_FIELDS			(BIT(8) - 1)
+
+/* ICMP fields */
+#define NH_FLD_ICMP_TYPE			BIT(0)
+#define NH_FLD_ICMP_CODE			BIT(1)
+#define NH_FLD_ICMP_CKSUM			BIT(2)
+#define NH_FLD_ICMP_ID				BIT(3)
+#define NH_FLD_ICMP_SQ_NUM			BIT(4)
+#define NH_FLD_ICMP_ALL_FIELDS			(BIT(5) - 1)
+
+/* IGMP fields */
+#define NH_FLD_IGMP_VERSION			BIT(0)
+#define NH_FLD_IGMP_TYPE			BIT(1)
+#define NH_FLD_IGMP_CKSUM			BIT(2)
+#define NH_FLD_IGMP_DATA			BIT(3)
+#define NH_FLD_IGMP_ALL_FIELDS			(BIT(4) - 1)
+
+/* TCP fields */
+#define NH_FLD_TCP_PORT_SRC			BIT(0)
+#define NH_FLD_TCP_PORT_DST			BIT(1)
+#define NH_FLD_TCP_SEQ				BIT(2)
+#define NH_FLD_TCP_ACK				BIT(3)
+#define NH_FLD_TCP_OFFSET			BIT(4)
+#define NH_FLD_TCP_FLAGS			BIT(5)
+#define NH_FLD_TCP_WINDOW			BIT(6)
+#define NH_FLD_TCP_CKSUM			BIT(7)
+#define NH_FLD_TCP_URGPTR			BIT(8)
+#define NH_FLD_TCP_OPTS				BIT(9)
+#define NH_FLD_TCP_OPTS_COUNT			BIT(10)
+#define NH_FLD_TCP_ALL_FIELDS			(BIT(11) - 1)
+
+/* UDP fields */
+#define NH_FLD_UDP_PORT_SRC			BIT(0)
+#define NH_FLD_UDP_PORT_DST			BIT(1)
+#define NH_FLD_UDP_LEN				BIT(2)
+#define NH_FLD_UDP_CKSUM			BIT(3)
+#define NH_FLD_UDP_ALL_FIELDS			(BIT(4) - 1)
+
+/* UDP-lite fields */
+#define NH_FLD_UDP_LITE_PORT_SRC		BIT(0)
+#define NH_FLD_UDP_LITE_PORT_DST		BIT(1)
+#define NH_FLD_UDP_LITE_ALL_FIELDS		(BIT(2) - 1)
+
+/* UDP-encap-ESP fields */
+#define NH_FLD_UDP_ENC_ESP_PORT_SRC		BIT(0)
+#define NH_FLD_UDP_ENC_ESP_PORT_DST		BIT(1)
+#define NH_FLD_UDP_ENC_ESP_LEN			BIT(2)
+#define NH_FLD_UDP_ENC_ESP_CKSUM		BIT(3)
+#define NH_FLD_UDP_ENC_ESP_SPI			BIT(4)
+#define NH_FLD_UDP_ENC_ESP_SEQUENCE_NUM		BIT(5)
+#define NH_FLD_UDP_ENC_ESP_ALL_FIELDS		(BIT(6) - 1)
+
+/* SCTP fields */
+#define NH_FLD_SCTP_PORT_SRC			BIT(0)
+#define NH_FLD_SCTP_PORT_DST			BIT(1)
+#define NH_FLD_SCTP_VER_TAG			BIT(2)
+#define NH_FLD_SCTP_CKSUM			BIT(3)
+#define NH_FLD_SCTP_ALL_FIELDS			(BIT(4) - 1)
+
+/* DCCP fields */
+#define NH_FLD_DCCP_PORT_SRC			BIT(0)
+#define NH_FLD_DCCP_PORT_DST			BIT(1)
+#define NH_FLD_DCCP_ALL_FIELDS			(BIT(2) - 1)
+
+/* IPHC fields */
+#define NH_FLD_IPHC_CID				BIT(0)
+#define NH_FLD_IPHC_CID_TYPE			BIT(1)
+#define NH_FLD_IPHC_HCINDEX			BIT(2)
+#define NH_FLD_IPHC_GEN				BIT(3)
+#define NH_FLD_IPHC_D_BIT			BIT(4)
+#define NH_FLD_IPHC_ALL_FIELDS			(BIT(5) - 1)
+
+/* SCTP fields */
+#define NH_FLD_SCTP_CHUNK_DATA_TYPE		BIT(0)
+#define NH_FLD_SCTP_CHUNK_DATA_FLAGS		BIT(1)
+#define NH_FLD_SCTP_CHUNK_DATA_LENGTH		BIT(2)
+#define NH_FLD_SCTP_CHUNK_DATA_TSN		BIT(3)
+#define NH_FLD_SCTP_CHUNK_DATA_STREAM_ID	BIT(4)
+#define NH_FLD_SCTP_CHUNK_DATA_STREAM_SQN	BIT(5)
+#define NH_FLD_SCTP_CHUNK_DATA_PAYLOAD_PID	BIT(6)
+#define NH_FLD_SCTP_CHUNK_DATA_UNORDERED	BIT(7)
+#define NH_FLD_SCTP_CHUNK_DATA_BEGGINING	BIT(8)
+#define NH_FLD_SCTP_CHUNK_DATA_END		BIT(9)
+#define NH_FLD_SCTP_CHUNK_DATA_ALL_FIELDS	(BIT(10) - 1)
+
+/* L2TPV2 fields */
+#define NH_FLD_L2TPV2_TYPE_BIT			BIT(0)
+#define NH_FLD_L2TPV2_LENGTH_BIT		BIT(1)
+#define NH_FLD_L2TPV2_SEQUENCE_BIT		BIT(2)
+#define NH_FLD_L2TPV2_OFFSET_BIT		BIT(3)
+#define NH_FLD_L2TPV2_PRIORITY_BIT		BIT(4)
+#define NH_FLD_L2TPV2_VERSION			BIT(5)
+#define NH_FLD_L2TPV2_LEN			BIT(6)
+#define NH_FLD_L2TPV2_TUNNEL_ID			BIT(7)
+#define NH_FLD_L2TPV2_SESSION_ID		BIT(8)
+#define NH_FLD_L2TPV2_NS			BIT(9)
+#define NH_FLD_L2TPV2_NR			BIT(10)
+#define NH_FLD_L2TPV2_OFFSET_SIZE		BIT(11)
+#define NH_FLD_L2TPV2_FIRST_BYTE		BIT(12)
+#define NH_FLD_L2TPV2_ALL_FIELDS		(BIT(13) - 1)
+
+/* L2TPV3 fields */
+#define NH_FLD_L2TPV3_CTRL_TYPE_BIT		BIT(0)
+#define NH_FLD_L2TPV3_CTRL_LENGTH_BIT		BIT(1)
+#define NH_FLD_L2TPV3_CTRL_SEQUENCE_BIT		BIT(2)
+#define NH_FLD_L2TPV3_CTRL_VERSION		BIT(3)
+#define NH_FLD_L2TPV3_CTRL_LENGTH		BIT(4)
+#define NH_FLD_L2TPV3_CTRL_CONTROL		BIT(5)
+#define NH_FLD_L2TPV3_CTRL_SENT			BIT(6)
+#define NH_FLD_L2TPV3_CTRL_RECV			BIT(7)
+#define NH_FLD_L2TPV3_CTRL_FIRST_BYTE		BIT(8)
+#define NH_FLD_L2TPV3_CTRL_ALL_FIELDS		(BIT(9) - 1)
+
+#define NH_FLD_L2TPV3_SESS_TYPE_BIT		BIT(0)
+#define NH_FLD_L2TPV3_SESS_VERSION		BIT(1)
+#define NH_FLD_L2TPV3_SESS_ID			BIT(2)
+#define NH_FLD_L2TPV3_SESS_COOKIE		BIT(3)
+#define NH_FLD_L2TPV3_SESS_ALL_FIELDS		(BIT(4) - 1)
+
+/* PPP fields */
+#define NH_FLD_PPP_PID				BIT(0)
+#define NH_FLD_PPP_COMPRESSED			BIT(1)
+#define NH_FLD_PPP_ALL_FIELDS			(BIT(2) - 1)
+
+/* PPPoE fields */
+#define NH_FLD_PPPOE_VER			BIT(0)
+#define NH_FLD_PPPOE_TYPE			BIT(1)
+#define NH_FLD_PPPOE_CODE			BIT(2)
+#define NH_FLD_PPPOE_SID			BIT(3)
+#define NH_FLD_PPPOE_LEN			BIT(4)
+#define NH_FLD_PPPOE_SESSION			BIT(5)
+#define NH_FLD_PPPOE_PID			BIT(6)
+#define NH_FLD_PPPOE_ALL_FIELDS			(BIT(7) - 1)
+
+/* PPP-Mux fields */
+#define NH_FLD_PPPMUX_PID			BIT(0)
+#define NH_FLD_PPPMUX_CKSUM			BIT(1)
+#define NH_FLD_PPPMUX_COMPRESSED		BIT(2)
+#define NH_FLD_PPPMUX_ALL_FIELDS		(BIT(3) - 1)
+
+/* PPP-Mux sub-frame fields */
+#define NH_FLD_PPPMUX_SUBFRM_PFF		BIT(0)
+#define NH_FLD_PPPMUX_SUBFRM_LXT		BIT(1)
+#define NH_FLD_PPPMUX_SUBFRM_LEN		BIT(2)
+#define NH_FLD_PPPMUX_SUBFRM_PID		BIT(3)
+#define NH_FLD_PPPMUX_SUBFRM_USE_PID		BIT(4)
+#define NH_FLD_PPPMUX_SUBFRM_ALL_FIELDS		(BIT(5) - 1)
+
+/* LLC fields */
+#define NH_FLD_LLC_DSAP				BIT(0)
+#define NH_FLD_LLC_SSAP				BIT(1)
+#define NH_FLD_LLC_CTRL				BIT(2)
+#define NH_FLD_LLC_ALL_FIELDS			(BIT(3) - 1)
+
+/* NLPID fields */
+#define NH_FLD_NLPID_NLPID			BIT(0)
+#define NH_FLD_NLPID_ALL_FIELDS			(BIT(1) - 1)
+
+/* SNAP fields */
+#define NH_FLD_SNAP_OUI				BIT(0)
+#define NH_FLD_SNAP_PID				BIT(1)
+#define NH_FLD_SNAP_ALL_FIELDS			(BIT(2) - 1)
+
+/* LLC SNAP fields */
+#define NH_FLD_LLC_SNAP_TYPE			BIT(0)
+#define NH_FLD_LLC_SNAP_ALL_FIELDS		(BIT(1) - 1)
+
+/* ARP fields */
+#define NH_FLD_ARP_HTYPE			BIT(0)
+#define NH_FLD_ARP_PTYPE			BIT(1)
+#define NH_FLD_ARP_HLEN				BIT(2)
+#define NH_FLD_ARP_PLEN				BIT(3)
+#define NH_FLD_ARP_OPER				BIT(4)
+#define NH_FLD_ARP_SHA				BIT(5)
+#define NH_FLD_ARP_SPA				BIT(6)
+#define NH_FLD_ARP_THA				BIT(7)
+#define NH_FLD_ARP_TPA				BIT(8)
+#define NH_FLD_ARP_ALL_FIELDS			(BIT(9) - 1)
+
+/* RFC2684 fields */
+#define NH_FLD_RFC2684_LLC			BIT(0)
+#define NH_FLD_RFC2684_NLPID			BIT(1)
+#define NH_FLD_RFC2684_OUI			BIT(2)
+#define NH_FLD_RFC2684_PID			BIT(3)
+#define NH_FLD_RFC2684_VPN_OUI			BIT(4)
+#define NH_FLD_RFC2684_VPN_IDX			BIT(5)
+#define NH_FLD_RFC2684_ALL_FIELDS		(BIT(6) - 1)
+
+/* User defined fields */
+#define NH_FLD_USER_DEFINED_SRCPORT		BIT(0)
+#define NH_FLD_USER_DEFINED_PCDID		BIT(1)
+#define NH_FLD_USER_DEFINED_ALL_FIELDS		(BIT(2) - 1)
+
+/* Payload fields */
+#define NH_FLD_PAYLOAD_BUFFER			BIT(0)
+#define NH_FLD_PAYLOAD_SIZE			BIT(1)
+#define NH_FLD_MAX_FRM_SIZE			BIT(2)
+#define NH_FLD_MIN_FRM_SIZE			BIT(3)
+#define NH_FLD_PAYLOAD_TYPE			BIT(4)
+#define NH_FLD_FRAME_SIZE			BIT(5)
+#define NH_FLD_PAYLOAD_ALL_FIELDS		(BIT(6) - 1)
+
+/* GRE fields */
+#define NH_FLD_GRE_TYPE				BIT(0)
+#define NH_FLD_GRE_ALL_FIELDS			(BIT(1) - 1)
+
+/* MINENCAP fields */
+#define NH_FLD_MINENCAP_SRC_IP			BIT(0)
+#define NH_FLD_MINENCAP_DST_IP			BIT(1)
+#define NH_FLD_MINENCAP_TYPE			BIT(2)
+#define NH_FLD_MINENCAP_ALL_FIELDS		(BIT(3) - 1)
+
+/* IPSEC AH fields */
+#define NH_FLD_IPSEC_AH_SPI			BIT(0)
+#define NH_FLD_IPSEC_AH_NH			BIT(1)
+#define NH_FLD_IPSEC_AH_ALL_FIELDS		(BIT(2) - 1)
+
+/* IPSEC ESP fields */
+#define NH_FLD_IPSEC_ESP_SPI			BIT(0)
+#define NH_FLD_IPSEC_ESP_SEQUENCE_NUM		BIT(1)
+#define NH_FLD_IPSEC_ESP_ALL_FIELDS		(BIT(2) - 1)
+
+/* MPLS fields */
+#define NH_FLD_MPLS_LABEL_STACK			BIT(0)
+#define NH_FLD_MPLS_LABEL_STACK_ALL_FIELDS	(BIT(1) - 1)
+
+/* MACSEC fields */
+#define NH_FLD_MACSEC_SECTAG			BIT(0)
+#define NH_FLD_MACSEC_ALL_FIELDS		(BIT(1) - 1)
+
+/* GTP fields */
+#define NH_FLD_GTP_TEID				BIT(0)
+
+/* Supported protocols */
+enum net_prot {
+	NET_PROT_NONE = 0,
+	NET_PROT_PAYLOAD,
+	NET_PROT_ETH,
+	NET_PROT_VLAN,
+	NET_PROT_IPV4,
+	NET_PROT_IPV6,
+	NET_PROT_IP,
+	NET_PROT_TCP,
+	NET_PROT_UDP,
+	NET_PROT_UDP_LITE,
+	NET_PROT_IPHC,
+	NET_PROT_SCTP,
+	NET_PROT_SCTP_CHUNK_DATA,
+	NET_PROT_PPPOE,
+	NET_PROT_PPP,
+	NET_PROT_PPPMUX,
+	NET_PROT_PPPMUX_SUBFRM,
+	NET_PROT_L2TPV2,
+	NET_PROT_L2TPV3_CTRL,
+	NET_PROT_L2TPV3_SESS,
+	NET_PROT_LLC,
+	NET_PROT_LLC_SNAP,
+	NET_PROT_NLPID,
+	NET_PROT_SNAP,
+	NET_PROT_MPLS,
+	NET_PROT_IPSEC_AH,
+	NET_PROT_IPSEC_ESP,
+	NET_PROT_UDP_ENC_ESP, /* RFC 3948 */
+	NET_PROT_MACSEC,
+	NET_PROT_GRE,
+	NET_PROT_MINENCAP,
+	NET_PROT_DCCP,
+	NET_PROT_ICMP,
+	NET_PROT_IGMP,
+	NET_PROT_ARP,
+	NET_PROT_CAPWAP_DATA,
+	NET_PROT_CAPWAP_CTRL,
+	NET_PROT_RFC2684,
+	NET_PROT_ICMPV6,
+	NET_PROT_FCOE,
+	NET_PROT_FIP,
+	NET_PROT_ISCSI,
+	NET_PROT_GTP,
+	NET_PROT_USER_DEFINED_L2,
+	NET_PROT_USER_DEFINED_L3,
+	NET_PROT_USER_DEFINED_L4,
+	NET_PROT_USER_DEFINED_L5,
+	NET_PROT_USER_DEFINED_SHIM1,
+	NET_PROT_USER_DEFINED_SHIM2,
+
+	NET_PROT_DUMMY_LAST
+};
+
+/**
+ * struct dpkg_extract - A structure for defining a single extraction
+ * @type: Determines how the union below is interpreted:
+ *	DPKG_EXTRACT_FROM_HDR: selects 'from_hdr';
+ *	DPKG_EXTRACT_FROM_DATA: selects 'from_data';
+ *	DPKG_EXTRACT_FROM_PARSE: selects 'from_parse'
+ * @extract: Selects extraction method
+ * @extract.from_hdr: Used when 'type = DPKG_EXTRACT_FROM_HDR'
+ * @extract.from_data: Used when 'type = DPKG_EXTRACT_FROM_DATA'
+ * @extract.from_parse:  Used when 'type = DPKG_EXTRACT_FROM_PARSE'
+ * @extract.from_hdr.prot: Any of the supported headers
+ * @extract.from_hdr.type: Defines the type of header extraction:
+ *	DPKG_FROM_HDR: use size & offset below;
+ *	DPKG_FROM_FIELD: use field, size and offset below;
+ *	DPKG_FULL_FIELD: use field below
+ * @extract.from_hdr.field: One of the supported fields (NH_FLD_)
+ * @extract.from_hdr.size: Size in bytes
+ * @extract.from_hdr.offset: Byte offset
+ * @extract.from_hdr.hdr_index: Clear for cases not listed below;
+ *	Used for protocols that may have more than a single
+ *	header, 0 indicates an outer header;
+ *	Supported protocols (possible values):
+ *	NET_PROT_VLAN (0, HDR_INDEX_LAST);
+ *	NET_PROT_MPLS (0, 1, HDR_INDEX_LAST);
+ *	NET_PROT_IP(0, HDR_INDEX_LAST);
+ *	NET_PROT_IPv4(0, HDR_INDEX_LAST);
+ *	NET_PROT_IPv6(0, HDR_INDEX_LAST);
+ * @extract.from_data.size: Size in bytes
+ * @extract.from_data.offset: Byte offset
+ * @extract.from_parse.size: Size in bytes
+ * @extract.from_parse.offset: Byte offset
+ * @num_of_byte_masks: Defines the number of valid entries in the array below;
+ *		This is	also the number of bytes to be used as masks
+ * @masks: Masks parameters
+ */
+struct dpkg_extract {
+	enum dpkg_extract_type type;
+	union {
+		struct {
+			enum net_prot			prot;
+			enum dpkg_extract_from_hdr_type type;
+			u32			field;
+			u8			size;
+			u8			offset;
+			u8			hdr_index;
+		} from_hdr;
+		struct {
+			u8 size;
+			u8 offset;
+		} from_data;
+		struct {
+			u8 size;
+			u8 offset;
+		} from_parse;
+	} extract;
+
+	u8		num_of_byte_masks;
+	struct dpkg_mask	masks[DPKG_NUM_OF_MASKS];
+};
+
+/**
+ * struct dpkg_profile_cfg - A structure for defining a full Key Generation
+ *				profile (rule)
+ * @num_extracts: Defines the number of valid entries in the array below
+ * @extracts: Array of required extractions
+ */
+struct dpkg_profile_cfg {
+	u8 num_extracts;
+	struct dpkg_extract extracts[DPKG_MAX_NUM_OF_EXTRACTS];
+};
+
+#endif /* __FSL_DPKG_H_ */
diff --git a/drivers/net/ethernet/freescale/dpaa2/dpmac-cmd.h b/drivers/net/ethernet/freescale/dpaa2/dpmac-cmd.h
new file mode 100644
index 000000000000..e9ac2ecef3be
--- /dev/null
+++ b/drivers/net/ethernet/freescale/dpaa2/dpmac-cmd.h
@@ -0,0 +1,85 @@
+/* SPDX-License-Identifier: (GPL-2.0+ OR BSD-3-Clause) */
+/* Copyright 2013-2016 Freescale Semiconductor Inc.
+ * Copyright 2019 NXP
+ */
+#ifndef _FSL_DPMAC_CMD_H
+#define _FSL_DPMAC_CMD_H
+
+/* DPMAC Version */
+#define DPMAC_VER_MAJOR				4
+#define DPMAC_VER_MINOR				4
+#define DPMAC_CMD_BASE_VERSION			1
+#define DPMAC_CMD_2ND_VERSION			2
+#define DPMAC_CMD_ID_OFFSET			4
+
+#define DPMAC_CMD(id)	(((id) << DPMAC_CMD_ID_OFFSET) | DPMAC_CMD_BASE_VERSION)
+#define DPMAC_CMD_V2(id) (((id) << DPMAC_CMD_ID_OFFSET) | DPMAC_CMD_2ND_VERSION)
+
+/* Command IDs */
+#define DPMAC_CMDID_CLOSE		DPMAC_CMD(0x800)
+#define DPMAC_CMDID_OPEN		DPMAC_CMD(0x80c)
+
+#define DPMAC_CMDID_GET_API_VERSION	DPMAC_CMD(0xa0c)
+
+#define DPMAC_CMDID_GET_ATTR		DPMAC_CMD(0x004)
+#define DPMAC_CMDID_SET_LINK_STATE	DPMAC_CMD_V2(0x0c3)
+
+#define DPMAC_CMDID_GET_COUNTER		DPMAC_CMD(0x0c4)
+
+#define DPMAC_CMDID_SET_PROTOCOL	DPMAC_CMD(0x0c7)
+
+/* Macros for accessing command fields smaller than 1byte */
+#define DPMAC_MASK(field)        \
+	GENMASK(DPMAC_##field##_SHIFT + DPMAC_##field##_SIZE - 1, \
+		DPMAC_##field##_SHIFT)
+
+#define dpmac_set_field(var, field, val) \
+	((var) |= (((val) << DPMAC_##field##_SHIFT) & DPMAC_MASK(field)))
+#define dpmac_get_field(var, field)      \
+	(((var) & DPMAC_MASK(field)) >> DPMAC_##field##_SHIFT)
+
+struct dpmac_cmd_open {
+	__le32 dpmac_id;
+};
+
+struct dpmac_rsp_get_attributes {
+	u8 eth_if;
+	u8 link_type;
+	__le16 id;
+	__le32 max_rate;
+};
+
+#define DPMAC_STATE_SIZE	1
+#define DPMAC_STATE_SHIFT	0
+#define DPMAC_STATE_VALID_SIZE	1
+#define DPMAC_STATE_VALID_SHIFT	1
+
+struct dpmac_cmd_set_link_state {
+	__le64 options;
+	__le32 rate;
+	__le32 pad0;
+	/* from lsb: up:1, state_valid:1 */
+	u8 state;
+	u8 pad1[7];
+	__le64 supported;
+	__le64 advertising;
+};
+
+struct dpmac_cmd_get_counter {
+	u8 id;
+};
+
+struct dpmac_rsp_get_counter {
+	__le64 pad;
+	__le64 counter;
+};
+
+struct dpmac_rsp_get_api_version {
+	__le16 major;
+	__le16 minor;
+};
+
+struct dpmac_cmd_set_protocol {
+	u8 eth_if;
+};
+#endif /* _FSL_DPMAC_CMD_H */
diff --git a/drivers/net/ethernet/freescale/dpaa2/dpmac.c b/drivers/net/ethernet/freescale/dpaa2/dpmac.c
new file mode 100644
index 000000000000..f440a4c3b70c
--- /dev/null
+++ b/drivers/net/ethernet/freescale/dpaa2/dpmac.c
@@ -0,0 +1,237 @@
+// SPDX-License-Identifier: (GPL-2.0+ OR BSD-3-Clause)
+/* Copyright 2013-2016 Freescale Semiconductor Inc.
+ * Copyright 2019 NXP
+ */
+#include <linux/fsl/mc.h>
+#include "dpmac.h"
+#include "dpmac-cmd.h"
+
+/**
+ * dpmac_open() - Open a control session for the specified object.
+ * @mc_io:	Pointer to MC portal's I/O object
+ * @cmd_flags:	Command flags; one or more of 'MC_CMD_FLAG_'
+ * @dpmac_id:	DPMAC unique ID
+ * @token:	Returned token; use in subsequent API calls
+ *
+ * This function can be used to open a control session for an
+ * already created object; an object may have been declared in
+ * the DPL or by calling the dpmac_create function.
+ * This function returns a unique authentication token,
+ * associated with the specific object ID and the specific MC
+ * portal; this token must be used in all subsequent commands for
+ * this specific object
+ *
+ * Return:	'0' on Success; Error code otherwise.
+ */
+int dpmac_open(struct fsl_mc_io *mc_io,
+	       u32 cmd_flags,
+	       int dpmac_id,
+	       u16 *token)
+{
+	struct dpmac_cmd_open *cmd_params;
+	struct fsl_mc_command cmd = { 0 };
+	int err;
+
+	/* prepare command */
+	cmd.header = mc_encode_cmd_header(DPMAC_CMDID_OPEN,
+					  cmd_flags,
+					  0);
+	cmd_params = (struct dpmac_cmd_open *)cmd.params;
+	cmd_params->dpmac_id = cpu_to_le32(dpmac_id);
+
+	/* send command to mc*/
+	err = mc_send_command(mc_io, &cmd);
+	if (err)
+		return err;
+
+	/* retrieve response parameters */
+	*token = mc_cmd_hdr_read_token(&cmd);
+
+	return err;
+}
+
+/**
+ * dpmac_close() - Close the control session of the object
+ * @mc_io:	Pointer to MC portal's I/O object
+ * @cmd_flags:	Command flags; one or more of 'MC_CMD_FLAG_'
+ * @token:	Token of DPMAC object
+ *
+ * After this function is called, no further operations are
+ * allowed on the object without opening a new control session.
+ *
+ * Return:	'0' on Success; Error code otherwise.
+ */
+int dpmac_close(struct fsl_mc_io *mc_io,
+		u32 cmd_flags,
+		u16 token)
+{
+	struct fsl_mc_command cmd = { 0 };
+
+	/* prepare command */
+	cmd.header = mc_encode_cmd_header(DPMAC_CMDID_CLOSE, cmd_flags,
+					  token);
+
+	/* send command to mc*/
+	return mc_send_command(mc_io, &cmd);
+}
+
+/**
+ * dpmac_get_attributes - Retrieve DPMAC attributes.
+ *
+ * @mc_io:	Pointer to MC portal's I/O object
+ * @cmd_flags:	Command flags; one or more of 'MC_CMD_FLAG_'
+ * @token:	Token of DPMAC object
+ * @attr:	Returned object's attributes
+ *
+ * Return:	'0' on Success; Error code otherwise.
+ */
+int dpmac_get_attributes(struct fsl_mc_io *mc_io,
+			 u32 cmd_flags,
+			 u16 token,
+			 struct dpmac_attr *attr)
+{
+	struct dpmac_rsp_get_attributes *rsp_params;
+	struct fsl_mc_command cmd = { 0 };
+	int err;
+
+	/* prepare command */
+	cmd.header = mc_encode_cmd_header(DPMAC_CMDID_GET_ATTR,
+					  cmd_flags,
+					  token);
+
+	/* send command to mc*/
+	err = mc_send_command(mc_io, &cmd);
+	if (err)
+		return err;
+
+	/* retrieve response parameters */
+	rsp_params = (struct dpmac_rsp_get_attributes *)cmd.params;
+	attr->eth_if = rsp_params->eth_if;
+	attr->link_type = rsp_params->link_type;
+	attr->id = le16_to_cpu(rsp_params->id);
+	attr->max_rate = le32_to_cpu(rsp_params->max_rate);
+
+	return 0;
+}
+
+/**
+ * dpmac_set_link_state() - Set the Ethernet link status
+ * @mc_io:      Pointer to opaque I/O object
+ * @cmd_flags:  Command flags; one or more of 'MC_CMD_FLAG_'
+ * @token:      Token of DPMAC object
+ * @link_state: Link state configuration
+ *
+ * Return:      '0' on Success; Error code otherwise.
+ */
+int dpmac_set_link_state(struct fsl_mc_io *mc_io,
+			 u32 cmd_flags,
+			 u16 token,
+			 struct dpmac_link_state *link_state)
+{
+	struct dpmac_cmd_set_link_state *cmd_params;
+	struct fsl_mc_command cmd = { 0 };
+
+	/* prepare command */
+	cmd.header = mc_encode_cmd_header(DPMAC_CMDID_SET_LINK_STATE,
+					  cmd_flags,
+					  token);
+	cmd_params = (struct dpmac_cmd_set_link_state *)cmd.params;
+	cmd_params->options = cpu_to_le64(link_state->options);
+	cmd_params->rate = cpu_to_le32(link_state->rate);
+	dpmac_set_field(cmd_params->state, STATE, link_state->up);
+	dpmac_set_field(cmd_params->state, STATE_VALID,
+			link_state->state_valid);
+	cmd_params->supported = cpu_to_le64(link_state->supported);
+	cmd_params->advertising = cpu_to_le64(link_state->advertising);
+
+	/* send command to mc*/
+	return mc_send_command(mc_io, &cmd);
+}
+
+/**
+ * dpmac_get_counter() - Read a specific DPMAC counter
+ * @mc_io:	Pointer to opaque I/O object
+ * @cmd_flags:	Command flags; one or more of 'MC_CMD_FLAG_'
+ * @token:	Token of DPMAC object
+ * @id:		The requested counter ID
+ * @value:	Returned counter value
+ *
+ * Return:	The requested counter; '0' otherwise.
+ */
+int dpmac_get_counter(struct fsl_mc_io *mc_io, u32 cmd_flags, u16 token,
+		      enum dpmac_counter_id id, u64 *value)
+{
+	struct dpmac_cmd_get_counter *dpmac_cmd;
+	struct dpmac_rsp_get_counter *dpmac_rsp;
+	struct fsl_mc_command cmd = { 0 };
+	int err = 0;
+
+	cmd.header = mc_encode_cmd_header(DPMAC_CMDID_GET_COUNTER,
+					  cmd_flags,
+					  token);
+	dpmac_cmd = (struct dpmac_cmd_get_counter *)cmd.params;
+	dpmac_cmd->id = id;
+
+	err = mc_send_command(mc_io, &cmd);
+	if (err)
+		return err;
+
+	dpmac_rsp = (struct dpmac_rsp_get_counter *)cmd.params;
+	*value = le64_to_cpu(dpmac_rsp->counter);
+
+	return 0;
+}
+
+/**
+ * dpmac_get_api_version() - Get Data Path MAC version
+ * @mc_io:	Pointer to MC portal's I/O object
+ * @cmd_flags:	Command flags; one or more of 'MC_CMD_FLAG_'
+ * @major_ver:	Major version of data path mac API
+ * @minor_ver:	Minor version of data path mac API
+ *
+ * Return:  '0' on Success; Error code otherwise.
+ */
+int dpmac_get_api_version(struct fsl_mc_io *mc_io, u32 cmd_flags,
+			  u16 *major_ver, u16 *minor_ver)
+{
+	struct dpmac_rsp_get_api_version *rsp_params;
+	struct fsl_mc_command cmd = { 0 };
+	int err;
+
+	cmd.header = mc_encode_cmd_header(DPMAC_CMDID_GET_API_VERSION,
+					  cmd_flags,
+					  0);
+
+	err = mc_send_command(mc_io, &cmd);
+	if (err)
+		return err;
+
+	rsp_params = (struct dpmac_rsp_get_api_version *)cmd.params;
+	*major_ver = le16_to_cpu(rsp_params->major);
+	*minor_ver = le16_to_cpu(rsp_params->minor);
+
+	return 0;
+}
+
+/**
+ * dpmac_set_protocol() - Reconfigure the DPMAC protocol
+ * @mc_io:      Pointer to opaque I/O object
+ * @cmd_flags:  Command flags; one or more of 'MC_CMD_FLAG_'
+ * @token:      Token of DPMAC object
+ * @protocol:   New protocol for the DPMAC to be reconfigured in.
+ *
+ * Return:      '0' on Success; Error code otherwise.
+ */
+int dpmac_set_protocol(struct fsl_mc_io *mc_io, u32 cmd_flags, u16 token,
+		       enum dpmac_eth_if protocol)
+{
+	struct dpmac_cmd_set_protocol *cmd_params;
+	struct fsl_mc_command cmd = { 0 };
+
+	cmd.header = mc_encode_cmd_header(DPMAC_CMDID_SET_PROTOCOL,
+					  cmd_flags, token);
+	cmd_params = (struct dpmac_cmd_set_protocol *)cmd.params;
+	cmd_params->eth_if = protocol;
+
+	return mc_send_command(mc_io, &cmd);
+}
diff --git a/drivers/net/ethernet/freescale/dpaa2/dpmac.h b/drivers/net/ethernet/freescale/dpaa2/dpmac.h
new file mode 100644
index 000000000000..17488819ef68
--- /dev/null
+++ b/drivers/net/ethernet/freescale/dpaa2/dpmac.h
@@ -0,0 +1,213 @@
+/* SPDX-License-Identifier: (GPL-2.0+ OR BSD-3-Clause) */
+/* Copyright 2013-2016 Freescale Semiconductor Inc.
+ * Copyright 2019 NXP
+ */
+#ifndef __FSL_DPMAC_H
+#define __FSL_DPMAC_H
+
+/* Data Path MAC API
+ * Contains initialization APIs and runtime control APIs for DPMAC
+ */
+
+struct fsl_mc_io;
+
+int dpmac_open(struct fsl_mc_io *mc_io,
+	       u32 cmd_flags,
+	       int dpmac_id,
+	       u16 *token);
+
+int dpmac_close(struct fsl_mc_io *mc_io,
+		u32 cmd_flags,
+		u16 token);
+
+/**
+ * enum dpmac_link_type -  DPMAC link type
+ * @DPMAC_LINK_TYPE_NONE: No link
+ * @DPMAC_LINK_TYPE_FIXED: Link is fixed type
+ * @DPMAC_LINK_TYPE_PHY: Link by PHY ID
+ * @DPMAC_LINK_TYPE_BACKPLANE: Backplane link type
+ */
+enum dpmac_link_type {
+	DPMAC_LINK_TYPE_NONE,
+	DPMAC_LINK_TYPE_FIXED,
+	DPMAC_LINK_TYPE_PHY,
+	DPMAC_LINK_TYPE_BACKPLANE
+};
+
+/**
+ * enum dpmac_eth_if - DPMAC Ethrnet interface
+ * @DPMAC_ETH_IF_MII: MII interface
+ * @DPMAC_ETH_IF_RMII: RMII interface
+ * @DPMAC_ETH_IF_SMII: SMII interface
+ * @DPMAC_ETH_IF_GMII: GMII interface
+ * @DPMAC_ETH_IF_RGMII: RGMII interface
+ * @DPMAC_ETH_IF_SGMII: SGMII interface
+ * @DPMAC_ETH_IF_QSGMII: QSGMII interface
+ * @DPMAC_ETH_IF_XAUI: XAUI interface
+ * @DPMAC_ETH_IF_XFI: XFI interface
+ * @DPMAC_ETH_IF_CAUI: CAUI interface
+ * @DPMAC_ETH_IF_1000BASEX: 1000BASEX interface
+ * @DPMAC_ETH_IF_USXGMII: USXGMII interface
+ */
+enum dpmac_eth_if {
+	DPMAC_ETH_IF_MII,
+	DPMAC_ETH_IF_RMII,
+	DPMAC_ETH_IF_SMII,
+	DPMAC_ETH_IF_GMII,
+	DPMAC_ETH_IF_RGMII,
+	DPMAC_ETH_IF_SGMII,
+	DPMAC_ETH_IF_QSGMII,
+	DPMAC_ETH_IF_XAUI,
+	DPMAC_ETH_IF_XFI,
+	DPMAC_ETH_IF_CAUI,
+	DPMAC_ETH_IF_1000BASEX,
+	DPMAC_ETH_IF_USXGMII,
+};
+
+/**
+ * struct dpmac_attr - Structure representing DPMAC attributes
+ * @id:		DPMAC object ID
+ * @max_rate:	Maximum supported rate - in Mbps
+ * @eth_if:	Ethernet interface
+ * @link_type:	link type
+ */
+struct dpmac_attr {
+	u16 id;
+	u32 max_rate;
+	enum dpmac_eth_if eth_if;
+	enum dpmac_link_type link_type;
+};
+
+int dpmac_get_attributes(struct fsl_mc_io *mc_io,
+			 u32 cmd_flags,
+			 u16 token,
+			 struct dpmac_attr *attr);
+
+/* DPMAC link configuration/state options */
+
+#define DPMAC_LINK_OPT_AUTONEG			BIT_ULL(0)
+#define DPMAC_LINK_OPT_HALF_DUPLEX		BIT_ULL(1)
+#define DPMAC_LINK_OPT_PAUSE			BIT_ULL(2)
+#define DPMAC_LINK_OPT_ASYM_PAUSE		BIT_ULL(3)
+
+/* Advertised link speeds */
+#define DPMAC_ADVERTISED_10BASET_FULL		BIT_ULL(0)
+#define DPMAC_ADVERTISED_100BASET_FULL		BIT_ULL(1)
+#define DPMAC_ADVERTISED_1000BASET_FULL		BIT_ULL(2)
+#define DPMAC_ADVERTISED_10000BASET_FULL	BIT_ULL(4)
+#define DPMAC_ADVERTISED_2500BASEX_FULL		BIT_ULL(5)
+
+/* Advertise auto-negotiation enable */
+#define DPMAC_ADVERTISED_AUTONEG		BIT_ULL(3)
+
+/**
+ * struct dpmac_link_state - DPMAC link configuration request
+ * @rate: Rate in Mbps
+ * @options: Enable/Disable DPMAC link cfg features (bitmap)
+ * @up: Link state
+ * @state_valid: Ignore/Update the state of the link
+ * @supported: Speeds capability of the phy (bitmap)
+ * @advertising: Speeds that are advertised for autoneg (bitmap)
+ */
+struct dpmac_link_state {
+	u32 rate;
+	u64 options;
+	int up;
+	int state_valid;
+	u64 supported;
+	u64 advertising;
+};
+
+int dpmac_set_link_state(struct fsl_mc_io *mc_io,
+			 u32 cmd_flags,
+			 u16 token,
+			 struct dpmac_link_state *link_state);
+
+/**
+ * enum dpmac_counter_id - DPMAC counter types
+ *
+ * @DPMAC_CNT_ING_FRAME_64: counts 64-bytes frames, good or bad.
+ * @DPMAC_CNT_ING_FRAME_127: counts 65- to 127-bytes frames, good or bad.
+ * @DPMAC_CNT_ING_FRAME_255: counts 128- to 255-bytes frames, good or bad.
+ * @DPMAC_CNT_ING_FRAME_511: counts 256- to 511-bytes frames, good or bad.
+ * @DPMAC_CNT_ING_FRAME_1023: counts 512- to 1023-bytes frames, good or bad.
+ * @DPMAC_CNT_ING_FRAME_1518: counts 1024- to 1518-bytes frames, good or bad.
+ * @DPMAC_CNT_ING_FRAME_1519_MAX: counts 1519-bytes frames and larger
+ *				  (up to max frame length specified),
+ *				  good or bad.
+ * @DPMAC_CNT_ING_FRAG: counts frames which are shorter than 64 bytes received
+ *			with a wrong CRC
+ * @DPMAC_CNT_ING_JABBER: counts frames longer than the maximum frame length
+ *			  specified, with a bad frame check sequence.
+ * @DPMAC_CNT_ING_FRAME_DISCARD: counts dropped frames due to internal errors.
+ *				 Occurs when a receive FIFO overflows.
+ *				 Includes also frames truncated as a result of
+ *				 the receive FIFO overflow.
+ * @DPMAC_CNT_ING_ALIGN_ERR: counts frames with an alignment error
+ *			     (optional used for wrong SFD).
+ * @DPMAC_CNT_EGR_UNDERSIZED: counts frames transmitted that was less than 64
+ *			      bytes long with a good CRC.
+ * @DPMAC_CNT_ING_OVERSIZED: counts frames longer than the maximum frame length
+ *			     specified, with a good frame check sequence.
+ * @DPMAC_CNT_ING_VALID_PAUSE_FRAME: counts valid pause frames (regular and PFC)
+ * @DPMAC_CNT_EGR_VALID_PAUSE_FRAME: counts valid pause frames transmitted
+ *				     (regular and PFC).
+ * @DPMAC_CNT_ING_BYTE: counts bytes received except preamble for all valid
+ *			frames and valid pause frames.
+ * @DPMAC_CNT_ING_MCAST_FRAME: counts received multicast frames.
+ * @DPMAC_CNT_ING_BCAST_FRAME: counts received broadcast frames.
+ * @DPMAC_CNT_ING_ALL_FRAME: counts each good or bad frames received.
+ * @DPMAC_CNT_ING_UCAST_FRAME: counts received unicast frames.
+ * @DPMAC_CNT_ING_ERR_FRAME: counts frames received with an error
+ *			     (except for undersized/fragment frame).
+ * @DPMAC_CNT_EGR_BYTE: counts bytes transmitted except preamble for all valid
+ *			frames and valid pause frames transmitted.
+ * @DPMAC_CNT_EGR_MCAST_FRAME: counts transmitted multicast frames.
+ * @DPMAC_CNT_EGR_BCAST_FRAME: counts transmitted broadcast frames.
+ * @DPMAC_CNT_EGR_UCAST_FRAME: counts transmitted unicast frames.
+ * @DPMAC_CNT_EGR_ERR_FRAME: counts frames transmitted with an error.
+ * @DPMAC_CNT_ING_GOOD_FRAME: counts frames received without error, including
+ *			      pause frames.
+ * @DPMAC_CNT_EGR_GOOD_FRAME: counts frames transmitted without error, including
+ *			      pause frames.
+ */
+enum dpmac_counter_id {
+	DPMAC_CNT_ING_FRAME_64,
+	DPMAC_CNT_ING_FRAME_127,
+	DPMAC_CNT_ING_FRAME_255,
+	DPMAC_CNT_ING_FRAME_511,
+	DPMAC_CNT_ING_FRAME_1023,
+	DPMAC_CNT_ING_FRAME_1518,
+	DPMAC_CNT_ING_FRAME_1519_MAX,
+	DPMAC_CNT_ING_FRAG,
+	DPMAC_CNT_ING_JABBER,
+	DPMAC_CNT_ING_FRAME_DISCARD,
+	DPMAC_CNT_ING_ALIGN_ERR,
+	DPMAC_CNT_EGR_UNDERSIZED,
+	DPMAC_CNT_ING_OVERSIZED,
+	DPMAC_CNT_ING_VALID_PAUSE_FRAME,
+	DPMAC_CNT_EGR_VALID_PAUSE_FRAME,
+	DPMAC_CNT_ING_BYTE,
+	DPMAC_CNT_ING_MCAST_FRAME,
+	DPMAC_CNT_ING_BCAST_FRAME,
+	DPMAC_CNT_ING_ALL_FRAME,
+	DPMAC_CNT_ING_UCAST_FRAME,
+	DPMAC_CNT_ING_ERR_FRAME,
+	DPMAC_CNT_EGR_BYTE,
+	DPMAC_CNT_EGR_MCAST_FRAME,
+	DPMAC_CNT_EGR_BCAST_FRAME,
+	DPMAC_CNT_EGR_UCAST_FRAME,
+	DPMAC_CNT_EGR_ERR_FRAME,
+	DPMAC_CNT_ING_GOOD_FRAME,
+	DPMAC_CNT_EGR_GOOD_FRAME
+};
+
+int dpmac_get_counter(struct fsl_mc_io *mc_io, u32 cmd_flags, u16 token,
+		      enum dpmac_counter_id id, u64 *value);
+
+int dpmac_get_api_version(struct fsl_mc_io *mc_io, u32 cmd_flags,
+			  u16 *major_ver, u16 *minor_ver);
+
+int dpmac_set_protocol(struct fsl_mc_io *mc_io, u32 cmd_flags, u16 token,
+		       enum dpmac_eth_if protocol);
+#endif /* __FSL_DPMAC_H */
diff --git a/drivers/net/ethernet/freescale/dpaa2/dpni-cmd.h b/drivers/net/ethernet/freescale/dpaa2/dpni-cmd.h
new file mode 100644
index 000000000000..be9492b8d5dc
--- /dev/null
+++ b/drivers/net/ethernet/freescale/dpaa2/dpni-cmd.h
@@ -0,0 +1,693 @@
+/* SPDX-License-Identifier: (GPL-2.0+ OR BSD-3-Clause) */
+/* Copyright 2013-2016 Freescale Semiconductor Inc.
+ * Copyright 2016 NXP
+ * Copyright 2020 NXP
+ */
+#ifndef _FSL_DPNI_CMD_H
+#define _FSL_DPNI_CMD_H
+
+#include "dpni.h"
+
+/* DPNI Version */
+#define DPNI_VER_MAJOR				7
+#define DPNI_VER_MINOR				0
+#define DPNI_CMD_BASE_VERSION			1
+#define DPNI_CMD_2ND_VERSION			2
+#define DPNI_CMD_3RD_VERSION			3
+#define DPNI_CMD_ID_OFFSET			4
+
+#define DPNI_CMD(id)	(((id) << DPNI_CMD_ID_OFFSET) | DPNI_CMD_BASE_VERSION)
+#define DPNI_CMD_V2(id)	(((id) << DPNI_CMD_ID_OFFSET) | DPNI_CMD_2ND_VERSION)
+#define DPNI_CMD_V3(id)	(((id) << DPNI_CMD_ID_OFFSET) | DPNI_CMD_3RD_VERSION)
+
+#define DPNI_CMDID_OPEN					DPNI_CMD(0x801)
+#define DPNI_CMDID_CLOSE				DPNI_CMD(0x800)
+#define DPNI_CMDID_CREATE				DPNI_CMD(0x901)
+#define DPNI_CMDID_DESTROY				DPNI_CMD(0x900)
+#define DPNI_CMDID_GET_API_VERSION			DPNI_CMD(0xa01)
+
+#define DPNI_CMDID_ENABLE				DPNI_CMD(0x002)
+#define DPNI_CMDID_DISABLE				DPNI_CMD(0x003)
+#define DPNI_CMDID_GET_ATTR				DPNI_CMD(0x004)
+#define DPNI_CMDID_RESET				DPNI_CMD(0x005)
+#define DPNI_CMDID_IS_ENABLED				DPNI_CMD(0x006)
+
+#define DPNI_CMDID_SET_IRQ				DPNI_CMD(0x010)
+#define DPNI_CMDID_GET_IRQ				DPNI_CMD(0x011)
+#define DPNI_CMDID_SET_IRQ_ENABLE			DPNI_CMD(0x012)
+#define DPNI_CMDID_GET_IRQ_ENABLE			DPNI_CMD(0x013)
+#define DPNI_CMDID_SET_IRQ_MASK				DPNI_CMD(0x014)
+#define DPNI_CMDID_GET_IRQ_MASK				DPNI_CMD(0x015)
+#define DPNI_CMDID_GET_IRQ_STATUS			DPNI_CMD(0x016)
+#define DPNI_CMDID_CLEAR_IRQ_STATUS			DPNI_CMD(0x017)
+
+#define DPNI_CMDID_SET_POOLS				DPNI_CMD_V3(0x200)
+#define DPNI_CMDID_SET_ERRORS_BEHAVIOR			DPNI_CMD(0x20B)
+
+#define DPNI_CMDID_GET_QDID				DPNI_CMD(0x210)
+#define DPNI_CMDID_GET_TX_DATA_OFFSET			DPNI_CMD(0x212)
+#define DPNI_CMDID_GET_LINK_STATE			DPNI_CMD(0x215)
+#define DPNI_CMDID_SET_MAX_FRAME_LENGTH			DPNI_CMD(0x216)
+#define DPNI_CMDID_GET_MAX_FRAME_LENGTH			DPNI_CMD(0x217)
+#define DPNI_CMDID_SET_LINK_CFG				DPNI_CMD(0x21A)
+#define DPNI_CMDID_SET_TX_SHAPING			DPNI_CMD_V2(0x21B)
+
+#define DPNI_CMDID_SET_MCAST_PROMISC			DPNI_CMD(0x220)
+#define DPNI_CMDID_GET_MCAST_PROMISC			DPNI_CMD(0x221)
+#define DPNI_CMDID_SET_UNICAST_PROMISC			DPNI_CMD(0x222)
+#define DPNI_CMDID_GET_UNICAST_PROMISC			DPNI_CMD(0x223)
+#define DPNI_CMDID_SET_PRIM_MAC				DPNI_CMD(0x224)
+#define DPNI_CMDID_GET_PRIM_MAC				DPNI_CMD(0x225)
+#define DPNI_CMDID_ADD_MAC_ADDR				DPNI_CMD(0x226)
+#define DPNI_CMDID_REMOVE_MAC_ADDR			DPNI_CMD(0x227)
+#define DPNI_CMDID_CLR_MAC_FILTERS			DPNI_CMD(0x228)
+
+#define DPNI_CMDID_SET_RX_TC_DIST			DPNI_CMD(0x235)
+
+#define DPNI_CMDID_ENABLE_VLAN_FILTER			DPNI_CMD(0x230)
+#define DPNI_CMDID_ADD_VLAN_ID				DPNI_CMD_V2(0x231)
+#define DPNI_CMDID_REMOVE_VLAN_ID			DPNI_CMD(0x232)
+
+#define DPNI_CMDID_SET_QOS_TBL				DPNI_CMD(0x240)
+#define DPNI_CMDID_ADD_QOS_ENT				DPNI_CMD(0x241)
+#define DPNI_CMDID_REMOVE_QOS_ENT			DPNI_CMD(0x242)
+#define DPNI_CMDID_CLR_QOS_TBL				DPNI_CMD(0x243)
+#define DPNI_CMDID_ADD_FS_ENT				DPNI_CMD(0x244)
+#define DPNI_CMDID_REMOVE_FS_ENT			DPNI_CMD(0x245)
+#define DPNI_CMDID_CLR_FS_ENT				DPNI_CMD(0x246)
+
+#define DPNI_CMDID_GET_STATISTICS			DPNI_CMD(0x25D)
+#define DPNI_CMDID_GET_QUEUE				DPNI_CMD(0x25F)
+#define DPNI_CMDID_SET_QUEUE				DPNI_CMD(0x260)
+#define DPNI_CMDID_GET_TAILDROP				DPNI_CMD(0x261)
+#define DPNI_CMDID_SET_TAILDROP				DPNI_CMD(0x262)
+
+#define DPNI_CMDID_GET_PORT_MAC_ADDR			DPNI_CMD(0x263)
+
+#define DPNI_CMDID_GET_BUFFER_LAYOUT			DPNI_CMD(0x264)
+#define DPNI_CMDID_SET_BUFFER_LAYOUT			DPNI_CMD(0x265)
+
+#define DPNI_CMDID_SET_TX_CONFIRMATION_MODE		DPNI_CMD(0x266)
+#define DPNI_CMDID_SET_CONGESTION_NOTIFICATION		DPNI_CMD(0x267)
+#define DPNI_CMDID_GET_CONGESTION_NOTIFICATION		DPNI_CMD(0x268)
+#define DPNI_CMDID_SET_EARLY_DROP			DPNI_CMD(0x269)
+#define DPNI_CMDID_GET_EARLY_DROP			DPNI_CMD(0x26A)
+#define DPNI_CMDID_GET_OFFLOAD				DPNI_CMD(0x26B)
+#define DPNI_CMDID_SET_OFFLOAD				DPNI_CMD(0x26C)
+
+#define DPNI_CMDID_SET_RX_FS_DIST			DPNI_CMD(0x273)
+#define DPNI_CMDID_SET_RX_HASH_DIST			DPNI_CMD(0x274)
+#define DPNI_CMDID_GET_LINK_CFG				DPNI_CMD(0x278)
+
+#define DPNI_CMDID_SET_SINGLE_STEP_CFG			DPNI_CMD(0x279)
+#define DPNI_CMDID_GET_SINGLE_STEP_CFG			DPNI_CMD_V2(0x27a)
+
+/* Macros for accessing command fields smaller than 1byte */
+#define DPNI_MASK(field)	\
+	GENMASK(DPNI_##field##_SHIFT + DPNI_##field##_SIZE - 1, \
+		DPNI_##field##_SHIFT)
+
+#define dpni_set_field(var, field, val)	\
+	((var) |= (((val) << DPNI_##field##_SHIFT) & DPNI_MASK(field)))
+#define dpni_get_field(var, field)	\
+	(((var) & DPNI_MASK(field)) >> DPNI_##field##_SHIFT)
+
+struct dpni_cmd_open {
+	__le32 dpni_id;
+};
+
+#define DPNI_BACKUP_POOL(val, order)	(((val) & 0x1) << (order))
+
+struct dpni_cmd_pool {
+	__le16 dpbp_id;
+	u8 priority_mask;
+	u8 pad;
+};
+
+struct dpni_cmd_set_pools {
+	u8 num_dpbp;
+	u8 backup_pool_mask;
+	u8 pad;
+	u8 pool_options;
+	struct dpni_cmd_pool pool[DPNI_MAX_DPBP];
+	__le16 buffer_size[DPNI_MAX_DPBP];
+};
+
+/* The enable indication is always the least significant bit */
+#define DPNI_ENABLE_SHIFT		0
+#define DPNI_ENABLE_SIZE		1
+
+struct dpni_rsp_is_enabled {
+	u8 enabled;
+};
+
+struct dpni_rsp_get_irq {
+	/* response word 0 */
+	__le32 irq_val;
+	__le32 pad;
+	/* response word 1 */
+	__le64 irq_addr;
+	/* response word 2 */
+	__le32 irq_num;
+	__le32 type;
+};
+
+struct dpni_cmd_set_irq_enable {
+	u8 enable;
+	u8 pad[3];
+	u8 irq_index;
+};
+
+struct dpni_cmd_get_irq_enable {
+	__le32 pad;
+	u8 irq_index;
+};
+
+struct dpni_rsp_get_irq_enable {
+	u8 enabled;
+};
+
+struct dpni_cmd_set_irq_mask {
+	__le32 mask;
+	u8 irq_index;
+};
+
+struct dpni_cmd_get_irq_mask {
+	__le32 pad;
+	u8 irq_index;
+};
+
+struct dpni_rsp_get_irq_mask {
+	__le32 mask;
+};
+
+struct dpni_cmd_get_irq_status {
+	__le32 status;
+	u8 irq_index;
+};
+
+struct dpni_rsp_get_irq_status {
+	__le32 status;
+};
+
+struct dpni_cmd_clear_irq_status {
+	__le32 status;
+	u8 irq_index;
+};
+
+struct dpni_rsp_get_attr {
+	/* response word 0 */
+	__le32 options;
+	u8 num_queues;
+	u8 num_tcs;
+	u8 mac_filter_entries;
+	u8 pad0;
+	/* response word 1 */
+	u8 vlan_filter_entries;
+	u8 pad1;
+	u8 qos_entries;
+	u8 pad2;
+	__le16 fs_entries;
+	__le16 pad3;
+	/* response word 2 */
+	u8 qos_key_size;
+	u8 fs_key_size;
+	__le16 wriop_version;
+};
+
+#define DPNI_ERROR_ACTION_SHIFT		0
+#define DPNI_ERROR_ACTION_SIZE		4
+#define DPNI_FRAME_ANN_SHIFT		4
+#define DPNI_FRAME_ANN_SIZE		1
+
+struct dpni_cmd_set_errors_behavior {
+	__le32 errors;
+	/* from least significant bit: error_action:4, set_frame_annotation:1 */
+	u8 flags;
+};
+
+/* There are 3 separate commands for configuring Rx, Tx and Tx confirmation
+ * buffer layouts, but they all share the same parameters.
+ * If one of the functions changes, below structure needs to be split.
+ */
+
+#define DPNI_PASS_TS_SHIFT		0
+#define DPNI_PASS_TS_SIZE		1
+#define DPNI_PASS_PR_SHIFT		1
+#define DPNI_PASS_PR_SIZE		1
+#define DPNI_PASS_FS_SHIFT		2
+#define DPNI_PASS_FS_SIZE		1
+
+struct dpni_cmd_get_buffer_layout {
+	u8 qtype;
+};
+
+struct dpni_rsp_get_buffer_layout {
+	/* response word 0 */
+	u8 pad0[6];
+	/* from LSB: pass_timestamp:1, parser_result:1, frame_status:1 */
+	u8 flags;
+	u8 pad1;
+	/* response word 1 */
+	__le16 private_data_size;
+	__le16 data_align;
+	__le16 head_room;
+	__le16 tail_room;
+};
+
+struct dpni_cmd_set_buffer_layout {
+	/* cmd word 0 */
+	u8 qtype;
+	u8 pad0[3];
+	__le16 options;
+	/* from LSB: pass_timestamp:1, parser_result:1, frame_status:1 */
+	u8 flags;
+	u8 pad1;
+	/* cmd word 1 */
+	__le16 private_data_size;
+	__le16 data_align;
+	__le16 head_room;
+	__le16 tail_room;
+};
+
+struct dpni_cmd_set_offload {
+	u8 pad[3];
+	u8 dpni_offload;
+	__le32 config;
+};
+
+struct dpni_cmd_get_offload {
+	u8 pad[3];
+	u8 dpni_offload;
+};
+
+struct dpni_rsp_get_offload {
+	__le32 pad;
+	__le32 config;
+};
+
+struct dpni_cmd_get_qdid {
+	u8 qtype;
+};
+
+struct dpni_rsp_get_qdid {
+	__le16 qdid;
+};
+
+struct dpni_rsp_get_tx_data_offset {
+	__le16 data_offset;
+};
+
+struct dpni_cmd_get_statistics {
+	u8 page_number;
+};
+
+struct dpni_rsp_get_statistics {
+	__le64 counter[DPNI_STATISTICS_CNT];
+};
+
+struct dpni_cmd_link_cfg {
+	/* cmd word 0 */
+	__le64 pad0;
+	/* cmd word 1 */
+	__le32 rate;
+	__le32 pad1;
+	/* cmd word 2 */
+	__le64 options;
+};
+
+#define DPNI_LINK_STATE_SHIFT		0
+#define DPNI_LINK_STATE_SIZE		1
+
+struct dpni_rsp_get_link_state {
+	/* response word 0 */
+	__le32 pad0;
+	/* from LSB: up:1 */
+	u8 flags;
+	u8 pad1[3];
+	/* response word 1 */
+	__le32 rate;
+	__le32 pad2;
+	/* response word 2 */
+	__le64 options;
+};
+
+struct dpni_cmd_set_max_frame_length {
+	__le16 max_frame_length;
+};
+
+struct dpni_rsp_get_max_frame_length {
+	__le16 max_frame_length;
+};
+
+struct dpni_cmd_set_multicast_promisc {
+	u8 enable;
+};
+
+struct dpni_rsp_get_multicast_promisc {
+	u8 enabled;
+};
+
+struct dpni_cmd_set_unicast_promisc {
+	u8 enable;
+};
+
+struct dpni_rsp_get_unicast_promisc {
+	u8 enabled;
+};
+
+struct dpni_cmd_set_primary_mac_addr {
+	__le16 pad;
+	u8 mac_addr[6];
+};
+
+struct dpni_rsp_get_primary_mac_addr {
+	__le16 pad;
+	u8 mac_addr[6];
+};
+
+struct dpni_rsp_get_port_mac_addr {
+	__le16 pad;
+	u8 mac_addr[6];
+};
+
+struct dpni_cmd_add_mac_addr {
+	__le16 pad;
+	u8 mac_addr[6];
+};
+
+struct dpni_cmd_remove_mac_addr {
+	__le16 pad;
+	u8 mac_addr[6];
+};
+
+#define DPNI_UNICAST_FILTERS_SHIFT	0
+#define DPNI_UNICAST_FILTERS_SIZE	1
+#define DPNI_MULTICAST_FILTERS_SHIFT	1
+#define DPNI_MULTICAST_FILTERS_SIZE	1
+
+struct dpni_cmd_clear_mac_filters {
+	/* from LSB: unicast:1, multicast:1 */
+	u8 flags;
+};
+
+#define DPNI_DIST_MODE_SHIFT		0
+#define DPNI_DIST_MODE_SIZE		4
+#define DPNI_MISS_ACTION_SHIFT		4
+#define DPNI_MISS_ACTION_SIZE		4
+
+struct dpni_cmd_set_rx_tc_dist {
+	/* cmd word 0 */
+	__le16 dist_size;
+	u8 tc_id;
+	/* from LSB: dist_mode:4, miss_action:4 */
+	u8 flags;
+	__le16 pad0;
+	__le16 default_flow_id;
+	/* cmd word 1..5 */
+	__le64 pad1[5];
+	/* cmd word 6 */
+	__le64 key_cfg_iova;
+};
+
+/* dpni_set_rx_tc_dist extension (structure of the DMA-able memory at
+ * key_cfg_iova)
+ */
+struct dpni_mask_cfg {
+	u8 mask;
+	u8 offset;
+};
+
+#define DPNI_EFH_TYPE_SHIFT		0
+#define DPNI_EFH_TYPE_SIZE		4
+#define DPNI_EXTRACT_TYPE_SHIFT		0
+#define DPNI_EXTRACT_TYPE_SIZE		4
+
+struct dpni_dist_extract {
+	/* word 0 */
+	u8 prot;
+	/* EFH type stored in the 4 least significant bits */
+	u8 efh_type;
+	u8 size;
+	u8 offset;
+	__le32 field;
+	/* word 1 */
+	u8 hdr_index;
+	u8 constant;
+	u8 num_of_repeats;
+	u8 num_of_byte_masks;
+	/* Extraction type is stored in the 4 LSBs */
+	u8 extract_type;
+	u8 pad[3];
+	/* word 2 */
+	struct dpni_mask_cfg masks[4];
+};
+
+struct dpni_ext_set_rx_tc_dist {
+	/* extension word 0 */
+	u8 num_extracts;
+	u8 pad[7];
+	/* words 1..25 */
+	struct dpni_dist_extract extracts[DPKG_MAX_NUM_OF_EXTRACTS];
+};
+
+struct dpni_cmd_get_queue {
+	u8 qtype;
+	u8 tc;
+	u8 index;
+};
+
+#define DPNI_DEST_TYPE_SHIFT		0
+#define DPNI_DEST_TYPE_SIZE		4
+#define DPNI_STASH_CTRL_SHIFT		6
+#define DPNI_STASH_CTRL_SIZE		1
+#define DPNI_HOLD_ACTIVE_SHIFT		7
+#define DPNI_HOLD_ACTIVE_SIZE		1
+
+struct dpni_rsp_get_queue {
+	/* response word 0 */
+	__le64 pad0;
+	/* response word 1 */
+	__le32 dest_id;
+	__le16 pad1;
+	u8 dest_prio;
+	/* From LSB: dest_type:4, pad:2, flc_stash_ctrl:1, hold_active:1 */
+	u8 flags;
+	/* response word 2 */
+	__le64 flc;
+	/* response word 3 */
+	__le64 user_context;
+	/* response word 4 */
+	__le32 fqid;
+	__le16 qdbin;
+};
+
+struct dpni_cmd_set_queue {
+	/* cmd word 0 */
+	u8 qtype;
+	u8 tc;
+	u8 index;
+	u8 options;
+	__le32 pad0;
+	/* cmd word 1 */
+	__le32 dest_id;
+	__le16 pad1;
+	u8 dest_prio;
+	u8 flags;
+	/* cmd word 2 */
+	__le64 flc;
+	/* cmd word 3 */
+	__le64 user_context;
+};
+
+struct dpni_cmd_set_taildrop {
+	/* cmd word 0 */
+	u8 congestion_point;
+	u8 qtype;
+	u8 tc;
+	u8 index;
+	__le32 pad0;
+	/* cmd word 1 */
+	/* Only least significant bit is relevant */
+	u8 enable;
+	u8 pad1;
+	u8 units;
+	u8 pad2;
+	__le32 threshold;
+};
+
+struct dpni_cmd_get_taildrop {
+	u8 congestion_point;
+	u8 qtype;
+	u8 tc;
+	u8 index;
+};
+
+struct dpni_rsp_get_taildrop {
+	/* cmd word 0 */
+	__le64 pad0;
+	/* cmd word 1 */
+	/* only least significant bit is relevant */
+	u8 enable;
+	u8 pad1;
+	u8 units;
+	u8 pad2;
+	__le32 threshold;
+};
+
+struct dpni_rsp_get_api_version {
+	__le16 major;
+	__le16 minor;
+};
+
+#define DPNI_RX_FS_DIST_ENABLE_SHIFT	0
+#define DPNI_RX_FS_DIST_ENABLE_SIZE	1
+struct dpni_cmd_set_rx_fs_dist {
+	__le16 dist_size;
+	u8 enable;
+	u8 tc;
+	__le16 miss_flow_id;
+	__le16 pad;
+	__le64 key_cfg_iova;
+};
+
+#define DPNI_RX_HASH_DIST_ENABLE_SHIFT	0
+#define DPNI_RX_HASH_DIST_ENABLE_SIZE	1
+struct dpni_cmd_set_rx_hash_dist {
+	__le16 dist_size;
+	u8 enable;
+	u8 tc;
+	__le32 pad;
+	__le64 key_cfg_iova;
+};
+
+struct dpni_cmd_add_fs_entry {
+	/* cmd word 0 */
+	__le16 options;
+	u8 tc_id;
+	u8 key_size;
+	__le16 index;
+	__le16 flow_id;
+	/* cmd word 1 */
+	__le64 key_iova;
+	/* cmd word 2 */
+	__le64 mask_iova;
+	/* cmd word 3 */
+	__le64 flc;
+};
+
+struct dpni_cmd_remove_fs_entry {
+	/* cmd word 0 */
+	__le16 pad0;
+	u8 tc_id;
+	u8 key_size;
+	__le32 pad1;
+	/* cmd word 1 */
+	__le64 key_iova;
+	/* cmd word 2 */
+	__le64 mask_iova;
+};
+
+#define DPNI_DISCARD_ON_MISS_SHIFT	0
+#define DPNI_DISCARD_ON_MISS_SIZE	1
+
+struct dpni_cmd_set_qos_table {
+	__le32 pad;
+	u8 default_tc;
+	/* only the LSB */
+	u8 discard_on_miss;
+	__le16 pad1[21];
+	__le64 key_cfg_iova;
+};
+
+struct dpni_cmd_add_qos_entry {
+	__le16 pad;
+	u8 tc_id;
+	u8 key_size;
+	__le16 index;
+	__le16 pad1;
+	__le64 key_iova;
+	__le64 mask_iova;
+};
+
+struct dpni_cmd_remove_qos_entry {
+	u8 pad[3];
+	u8 key_size;
+	__le32 pad1;
+	__le64 key_iova;
+	__le64 mask_iova;
+};
+
+#define DPNI_DEST_TYPE_SHIFT		0
+#define DPNI_DEST_TYPE_SIZE		4
+#define DPNI_CONG_UNITS_SHIFT		4
+#define DPNI_CONG_UNITS_SIZE		2
+
+struct dpni_cmd_set_congestion_notification {
+	/* cmd word 0 */
+	u8 qtype;
+	u8 tc;
+	u8 pad[6];
+	/* cmd word 1 */
+	__le32 dest_id;
+	__le16 notification_mode;
+	u8 dest_priority;
+	/* from LSB: dest_type: 4 units:2 */
+	u8 type_units;
+	/* cmd word 2 */
+	__le64 message_iova;
+	/* cmd word 3 */
+	__le64 message_ctx;
+	/* cmd word 4 */
+	__le32 threshold_entry;
+	__le32 threshold_exit;
+};
+
+#define DPNI_COUPLED_SHIFT	0
+#define DPNI_COUPLED_SIZE	1
+
+struct dpni_cmd_set_tx_shaping {
+	__le16 tx_cr_max_burst_size;
+	__le16 tx_er_max_burst_size;
+	__le32 pad;
+	__le32 tx_cr_rate_limit;
+	__le32 tx_er_rate_limit;
+	/* from LSB: coupled:1 */
+	u8 coupled;
+};
+
+#define DPNI_PTP_ENABLE_SHIFT			0
+#define DPNI_PTP_ENABLE_SIZE			1
+#define DPNI_PTP_CH_UPDATE_SHIFT		1
+#define DPNI_PTP_CH_UPDATE_SIZE			1
+
+struct dpni_cmd_single_step_cfg {
+	__le16 flags;
+	__le16 offset;
+	__le32 peer_delay;
+	__le32 ptp_onestep_reg_base;
+	__le32 pad0;
+};
+
+struct dpni_rsp_single_step_cfg {
+	__le16 flags;
+	__le16 offset;
+	__le32 peer_delay;
+	__le32 ptp_onestep_reg_base;
+	__le32 pad0;
+};
+
+struct dpni_cmd_enable_vlan_filter {
+	/* only the LSB */
+	u8 en;
+};
+
+struct dpni_cmd_vlan_id {
+	u8 flags;
+	u8 tc_id;
+	u8 flow_id;
+	u8 pad;
+	__le16 vlan_id;
+};
+
+#endif /* _FSL_DPNI_CMD_H */
diff --git a/drivers/net/ethernet/freescale/dpaa2/dpni.c b/drivers/net/ethernet/freescale/dpaa2/dpni.c
new file mode 100644
index 000000000000..02601a283b59
--- /dev/null
+++ b/drivers/net/ethernet/freescale/dpaa2/dpni.c
@@ -0,0 +1,2185 @@
+// SPDX-License-Identifier: (GPL-2.0+ OR BSD-3-Clause)
+/* Copyright 2013-2016 Freescale Semiconductor Inc.
+ * Copyright 2016 NXP
+ * Copyright 2020 NXP
+ */
+#include <linux/kernel.h>
+#include <linux/errno.h>
+#include <linux/fsl/mc.h>
+#include "dpni.h"
+#include "dpni-cmd.h"
+
+/**
+ * dpni_prepare_key_cfg() - function prepare extract parameters
+ * @cfg: defining a full Key Generation profile (rule)
+ * @key_cfg_buf: Zeroed 256 bytes of memory before mapping it to DMA
+ *
+ * This function has to be called before the following functions:
+ *	- dpni_set_rx_tc_dist()
+ *	- dpni_set_qos_table()
+ *
+ * Return:	'0' on Success; Error code otherwise.
+ */
+int dpni_prepare_key_cfg(const struct dpkg_profile_cfg *cfg, u8 *key_cfg_buf)
+{
+	int i, j;
+	struct dpni_ext_set_rx_tc_dist *dpni_ext;
+	struct dpni_dist_extract *extr;
+
+	if (cfg->num_extracts > DPKG_MAX_NUM_OF_EXTRACTS)
+		return -EINVAL;
+
+	dpni_ext = (struct dpni_ext_set_rx_tc_dist *)key_cfg_buf;
+	dpni_ext->num_extracts = cfg->num_extracts;
+
+	for (i = 0; i < cfg->num_extracts; i++) {
+		extr = &dpni_ext->extracts[i];
+
+		switch (cfg->extracts[i].type) {
+		case DPKG_EXTRACT_FROM_HDR:
+			extr->prot = cfg->extracts[i].extract.from_hdr.prot;
+			dpni_set_field(extr->efh_type, EFH_TYPE,
+				       cfg->extracts[i].extract.from_hdr.type);
+			extr->size = cfg->extracts[i].extract.from_hdr.size;
+			extr->offset = cfg->extracts[i].extract.from_hdr.offset;
+			extr->field = cpu_to_le32(
+				cfg->extracts[i].extract.from_hdr.field);
+			extr->hdr_index =
+				cfg->extracts[i].extract.from_hdr.hdr_index;
+			break;
+		case DPKG_EXTRACT_FROM_DATA:
+			extr->size = cfg->extracts[i].extract.from_data.size;
+			extr->offset =
+				cfg->extracts[i].extract.from_data.offset;
+			break;
+		case DPKG_EXTRACT_FROM_PARSE:
+			extr->size = cfg->extracts[i].extract.from_parse.size;
+			extr->offset =
+				cfg->extracts[i].extract.from_parse.offset;
+			break;
+		default:
+			return -EINVAL;
+		}
+
+		extr->num_of_byte_masks = cfg->extracts[i].num_of_byte_masks;
+		dpni_set_field(extr->extract_type, EXTRACT_TYPE,
+			       cfg->extracts[i].type);
+
+		for (j = 0; j < DPKG_NUM_OF_MASKS; j++) {
+			extr->masks[j].mask = cfg->extracts[i].masks[j].mask;
+			extr->masks[j].offset =
+				cfg->extracts[i].masks[j].offset;
+		}
+	}
+
+	return 0;
+}
+
+/**
+ * dpni_open() - Open a control session for the specified object
+ * @mc_io:	Pointer to MC portal's I/O object
+ * @cmd_flags:	Command flags; one or more of 'MC_CMD_FLAG_'
+ * @dpni_id:	DPNI unique ID
+ * @token:	Returned token; use in subsequent API calls
+ *
+ * This function can be used to open a control session for an
+ * already created object; an object may have been declared in
+ * the DPL or by calling the dpni_create() function.
+ * This function returns a unique authentication token,
+ * associated with the specific object ID and the specific MC
+ * portal; this token must be used in all subsequent commands for
+ * this specific object.
+ *
+ * Return:	'0' on Success; Error code otherwise.
+ */
+int dpni_open(struct fsl_mc_io *mc_io,
+	      u32 cmd_flags,
+	      int dpni_id,
+	      u16 *token)
+{
+	struct fsl_mc_command cmd = { 0 };
+	struct dpni_cmd_open *cmd_params;
+
+	int err;
+
+	/* prepare command */
+	cmd.header = mc_encode_cmd_header(DPNI_CMDID_OPEN,
+					  cmd_flags,
+					  0);
+	cmd_params = (struct dpni_cmd_open *)cmd.params;
+	cmd_params->dpni_id = cpu_to_le32(dpni_id);
+
+	/* send command to mc*/
+	err = mc_send_command(mc_io, &cmd);
+	if (err)
+		return err;
+
+	/* retrieve response parameters */
+	*token = mc_cmd_hdr_read_token(&cmd);
+
+	return 0;
+}
+
+/**
+ * dpni_close() - Close the control session of the object
+ * @mc_io:	Pointer to MC portal's I/O object
+ * @cmd_flags:	Command flags; one or more of 'MC_CMD_FLAG_'
+ * @token:	Token of DPNI object
+ *
+ * After this function is called, no further operations are
+ * allowed on the object without opening a new control session.
+ *
+ * Return:	'0' on Success; Error code otherwise.
+ */
+int dpni_close(struct fsl_mc_io *mc_io,
+	       u32 cmd_flags,
+	       u16 token)
+{
+	struct fsl_mc_command cmd = { 0 };
+
+	/* prepare command */
+	cmd.header = mc_encode_cmd_header(DPNI_CMDID_CLOSE,
+					  cmd_flags,
+					  token);
+
+	/* send command to mc*/
+	return mc_send_command(mc_io, &cmd);
+}
+
+/**
+ * dpni_set_pools() - Set buffer pools configuration
+ * @mc_io:	Pointer to MC portal's I/O object
+ * @cmd_flags:	Command flags; one or more of 'MC_CMD_FLAG_'
+ * @token:	Token of DPNI object
+ * @cfg:	Buffer pools configuration
+ *
+ * mandatory for DPNI operation
+ * warning:Allowed only when DPNI is disabled
+ *
+ * Return:	'0' on Success; Error code otherwise.
+ */
+int dpni_set_pools(struct fsl_mc_io *mc_io,
+		   u32 cmd_flags,
+		   u16 token,
+		   const struct dpni_pools_cfg *cfg)
+{
+	struct fsl_mc_command cmd = { 0 };
+	struct dpni_cmd_set_pools *cmd_params;
+	int i;
+
+	/* prepare command */
+	cmd.header = mc_encode_cmd_header(DPNI_CMDID_SET_POOLS,
+					  cmd_flags,
+					  token);
+	cmd_params = (struct dpni_cmd_set_pools *)cmd.params;
+	cmd_params->num_dpbp = cfg->num_dpbp;
+	cmd_params->pool_options = cfg->pool_options;
+	for (i = 0; i < DPNI_MAX_DPBP; i++) {
+		cmd_params->pool[i].dpbp_id =
+			cpu_to_le16(cfg->pools[i].dpbp_id);
+		cmd_params->pool[i].priority_mask =
+			cfg->pools[i].priority_mask;
+		cmd_params->buffer_size[i] =
+			cpu_to_le16(cfg->pools[i].buffer_size);
+		cmd_params->backup_pool_mask |=
+			DPNI_BACKUP_POOL(cfg->pools[i].backup_pool, i);
+	}
+
+	/* send command to mc*/
+	return mc_send_command(mc_io, &cmd);
+}
+
+/**
+ * dpni_enable() - Enable the DPNI, allow sending and receiving frames.
+ * @mc_io:	Pointer to MC portal's I/O object
+ * @cmd_flags:	Command flags; one or more of 'MC_CMD_FLAG_'
+ * @token:		Token of DPNI object
+ *
+ * Return:	'0' on Success; Error code otherwise.
+ */
+int dpni_enable(struct fsl_mc_io *mc_io,
+		u32 cmd_flags,
+		u16 token)
+{
+	struct fsl_mc_command cmd = { 0 };
+
+	/* prepare command */
+	cmd.header = mc_encode_cmd_header(DPNI_CMDID_ENABLE,
+					  cmd_flags,
+					  token);
+
+	/* send command to mc*/
+	return mc_send_command(mc_io, &cmd);
+}
+
+/**
+ * dpni_disable() - Disable the DPNI, stop sending and receiving frames.
+ * @mc_io:	Pointer to MC portal's I/O object
+ * @cmd_flags:	Command flags; one or more of 'MC_CMD_FLAG_'
+ * @token:	Token of DPNI object
+ *
+ * Return:	'0' on Success; Error code otherwise.
+ */
+int dpni_disable(struct fsl_mc_io *mc_io,
+		 u32 cmd_flags,
+		 u16 token)
+{
+	struct fsl_mc_command cmd = { 0 };
+
+	/* prepare command */
+	cmd.header = mc_encode_cmd_header(DPNI_CMDID_DISABLE,
+					  cmd_flags,
+					  token);
+
+	/* send command to mc*/
+	return mc_send_command(mc_io, &cmd);
+}
+
+/**
+ * dpni_is_enabled() - Check if the DPNI is enabled.
+ * @mc_io:	Pointer to MC portal's I/O object
+ * @cmd_flags:	Command flags; one or more of 'MC_CMD_FLAG_'
+ * @token:	Token of DPNI object
+ * @en:		Returns '1' if object is enabled; '0' otherwise
+ *
+ * Return:	'0' on Success; Error code otherwise.
+ */
+int dpni_is_enabled(struct fsl_mc_io *mc_io,
+		    u32 cmd_flags,
+		    u16 token,
+		    int *en)
+{
+	struct fsl_mc_command cmd = { 0 };
+	struct dpni_rsp_is_enabled *rsp_params;
+	int err;
+
+	/* prepare command */
+	cmd.header = mc_encode_cmd_header(DPNI_CMDID_IS_ENABLED,
+					  cmd_flags,
+					  token);
+
+	/* send command to mc*/
+	err = mc_send_command(mc_io, &cmd);
+	if (err)
+		return err;
+
+	/* retrieve response parameters */
+	rsp_params = (struct dpni_rsp_is_enabled *)cmd.params;
+	*en = dpni_get_field(rsp_params->enabled, ENABLE);
+
+	return 0;
+}
+
+/**
+ * dpni_reset() - Reset the DPNI, returns the object to initial state.
+ * @mc_io:	Pointer to MC portal's I/O object
+ * @cmd_flags:	Command flags; one or more of 'MC_CMD_FLAG_'
+ * @token:	Token of DPNI object
+ *
+ * Return:	'0' on Success; Error code otherwise.
+ */
+int dpni_reset(struct fsl_mc_io *mc_io,
+	       u32 cmd_flags,
+	       u16 token)
+{
+	struct fsl_mc_command cmd = { 0 };
+
+	/* prepare command */
+	cmd.header = mc_encode_cmd_header(DPNI_CMDID_RESET,
+					  cmd_flags,
+					  token);
+
+	/* send command to mc*/
+	return mc_send_command(mc_io, &cmd);
+}
+
+/**
+ * dpni_set_irq_enable() - Set overall interrupt state.
+ * @mc_io:	Pointer to MC portal's I/O object
+ * @cmd_flags:	Command flags; one or more of 'MC_CMD_FLAG_'
+ * @token:	Token of DPNI object
+ * @irq_index:	The interrupt index to configure
+ * @en:		Interrupt state: - enable = 1, disable = 0
+ *
+ * Allows GPP software to control when interrupts are generated.
+ * Each interrupt can have up to 32 causes.  The enable/disable control's the
+ * overall interrupt state. if the interrupt is disabled no causes will cause
+ * an interrupt.
+ *
+ * Return:	'0' on Success; Error code otherwise.
+ */
+int dpni_set_irq_enable(struct fsl_mc_io *mc_io,
+			u32 cmd_flags,
+			u16 token,
+			u8 irq_index,
+			u8 en)
+{
+	struct fsl_mc_command cmd = { 0 };
+	struct dpni_cmd_set_irq_enable *cmd_params;
+
+	/* prepare command */
+	cmd.header = mc_encode_cmd_header(DPNI_CMDID_SET_IRQ_ENABLE,
+					  cmd_flags,
+					  token);
+	cmd_params = (struct dpni_cmd_set_irq_enable *)cmd.params;
+	dpni_set_field(cmd_params->enable, ENABLE, en);
+	cmd_params->irq_index = irq_index;
+
+	/* send command to mc*/
+	return mc_send_command(mc_io, &cmd);
+}
+
+/**
+ * dpni_get_irq_enable() - Get overall interrupt state
+ * @mc_io:	Pointer to MC portal's I/O object
+ * @cmd_flags:	Command flags; one or more of 'MC_CMD_FLAG_'
+ * @token:	Token of DPNI object
+ * @irq_index:	The interrupt index to configure
+ * @en:		Returned interrupt state - enable = 1, disable = 0
+ *
+ * Return:	'0' on Success; Error code otherwise.
+ */
+int dpni_get_irq_enable(struct fsl_mc_io *mc_io,
+			u32 cmd_flags,
+			u16 token,
+			u8 irq_index,
+			u8 *en)
+{
+	struct fsl_mc_command cmd = { 0 };
+	struct dpni_cmd_get_irq_enable *cmd_params;
+	struct dpni_rsp_get_irq_enable *rsp_params;
+
+	int err;
+
+	/* prepare command */
+	cmd.header = mc_encode_cmd_header(DPNI_CMDID_GET_IRQ_ENABLE,
+					  cmd_flags,
+					  token);
+	cmd_params = (struct dpni_cmd_get_irq_enable *)cmd.params;
+	cmd_params->irq_index = irq_index;
+
+	/* send command to mc*/
+	err = mc_send_command(mc_io, &cmd);
+	if (err)
+		return err;
+
+	/* retrieve response parameters */
+	rsp_params = (struct dpni_rsp_get_irq_enable *)cmd.params;
+	*en = dpni_get_field(rsp_params->enabled, ENABLE);
+
+	return 0;
+}
+
+/**
+ * dpni_set_irq_mask() - Set interrupt mask.
+ * @mc_io:	Pointer to MC portal's I/O object
+ * @cmd_flags:	Command flags; one or more of 'MC_CMD_FLAG_'
+ * @token:	Token of DPNI object
+ * @irq_index:	The interrupt index to configure
+ * @mask:	event mask to trigger interrupt;
+ *			each bit:
+ *				0 = ignore event
+ *				1 = consider event for asserting IRQ
+ *
+ * Every interrupt can have up to 32 causes and the interrupt model supports
+ * masking/unmasking each cause independently
+ *
+ * Return:	'0' on Success; Error code otherwise.
+ */
+int dpni_set_irq_mask(struct fsl_mc_io *mc_io,
+		      u32 cmd_flags,
+		      u16 token,
+		      u8 irq_index,
+		      u32 mask)
+{
+	struct fsl_mc_command cmd = { 0 };
+	struct dpni_cmd_set_irq_mask *cmd_params;
+
+	/* prepare command */
+	cmd.header = mc_encode_cmd_header(DPNI_CMDID_SET_IRQ_MASK,
+					  cmd_flags,
+					  token);
+	cmd_params = (struct dpni_cmd_set_irq_mask *)cmd.params;
+	cmd_params->mask = cpu_to_le32(mask);
+	cmd_params->irq_index = irq_index;
+
+	/* send command to mc*/
+	return mc_send_command(mc_io, &cmd);
+}
+
+/**
+ * dpni_get_irq_mask() - Get interrupt mask.
+ * @mc_io:	Pointer to MC portal's I/O object
+ * @cmd_flags:	Command flags; one or more of 'MC_CMD_FLAG_'
+ * @token:	Token of DPNI object
+ * @irq_index:	The interrupt index to configure
+ * @mask:	Returned event mask to trigger interrupt
+ *
+ * Every interrupt can have up to 32 causes and the interrupt model supports
+ * masking/unmasking each cause independently
+ *
+ * Return:	'0' on Success; Error code otherwise.
+ */
+int dpni_get_irq_mask(struct fsl_mc_io *mc_io,
+		      u32 cmd_flags,
+		      u16 token,
+		      u8 irq_index,
+		      u32 *mask)
+{
+	struct fsl_mc_command cmd = { 0 };
+	struct dpni_cmd_get_irq_mask *cmd_params;
+	struct dpni_rsp_get_irq_mask *rsp_params;
+	int err;
+
+	/* prepare command */
+	cmd.header = mc_encode_cmd_header(DPNI_CMDID_GET_IRQ_MASK,
+					  cmd_flags,
+					  token);
+	cmd_params = (struct dpni_cmd_get_irq_mask *)cmd.params;
+	cmd_params->irq_index = irq_index;
+
+	/* send command to mc*/
+	err = mc_send_command(mc_io, &cmd);
+	if (err)
+		return err;
+
+	/* retrieve response parameters */
+	rsp_params = (struct dpni_rsp_get_irq_mask *)cmd.params;
+	*mask = le32_to_cpu(rsp_params->mask);
+
+	return 0;
+}
+
+/**
+ * dpni_get_irq_status() - Get the current status of any pending interrupts.
+ * @mc_io:	Pointer to MC portal's I/O object
+ * @cmd_flags:	Command flags; one or more of 'MC_CMD_FLAG_'
+ * @token:	Token of DPNI object
+ * @irq_index:	The interrupt index to configure
+ * @status:	Returned interrupts status - one bit per cause:
+ *			0 = no interrupt pending
+ *			1 = interrupt pending
+ *
+ * Return:	'0' on Success; Error code otherwise.
+ */
+int dpni_get_irq_status(struct fsl_mc_io *mc_io,
+			u32 cmd_flags,
+			u16 token,
+			u8 irq_index,
+			u32 *status)
+{
+	struct fsl_mc_command cmd = { 0 };
+	struct dpni_cmd_get_irq_status *cmd_params;
+	struct dpni_rsp_get_irq_status *rsp_params;
+	int err;
+
+	/* prepare command */
+	cmd.header = mc_encode_cmd_header(DPNI_CMDID_GET_IRQ_STATUS,
+					  cmd_flags,
+					  token);
+	cmd_params = (struct dpni_cmd_get_irq_status *)cmd.params;
+	cmd_params->status = cpu_to_le32(*status);
+	cmd_params->irq_index = irq_index;
+
+	/* send command to mc*/
+	err = mc_send_command(mc_io, &cmd);
+	if (err)
+		return err;
+
+	/* retrieve response parameters */
+	rsp_params = (struct dpni_rsp_get_irq_status *)cmd.params;
+	*status = le32_to_cpu(rsp_params->status);
+
+	return 0;
+}
+
+/**
+ * dpni_clear_irq_status() - Clear a pending interrupt's status
+ * @mc_io:	Pointer to MC portal's I/O object
+ * @cmd_flags:	Command flags; one or more of 'MC_CMD_FLAG_'
+ * @token:	Token of DPNI object
+ * @irq_index:	The interrupt index to configure
+ * @status:	bits to clear (W1C) - one bit per cause:
+ *			0 = don't change
+ *			1 = clear status bit
+ *
+ * Return:	'0' on Success; Error code otherwise.
+ */
+int dpni_clear_irq_status(struct fsl_mc_io *mc_io,
+			  u32 cmd_flags,
+			  u16 token,
+			  u8 irq_index,
+			  u32 status)
+{
+	struct fsl_mc_command cmd = { 0 };
+	struct dpni_cmd_clear_irq_status *cmd_params;
+
+	/* prepare command */
+	cmd.header = mc_encode_cmd_header(DPNI_CMDID_CLEAR_IRQ_STATUS,
+					  cmd_flags,
+					  token);
+	cmd_params = (struct dpni_cmd_clear_irq_status *)cmd.params;
+	cmd_params->irq_index = irq_index;
+	cmd_params->status = cpu_to_le32(status);
+
+	/* send command to mc*/
+	return mc_send_command(mc_io, &cmd);
+}
+
+/**
+ * dpni_get_attributes() - Retrieve DPNI attributes.
+ * @mc_io:	Pointer to MC portal's I/O object
+ * @cmd_flags:	Command flags; one or more of 'MC_CMD_FLAG_'
+ * @token:	Token of DPNI object
+ * @attr:	Object's attributes
+ *
+ * Return:	'0' on Success; Error code otherwise.
+ */
+int dpni_get_attributes(struct fsl_mc_io *mc_io,
+			u32 cmd_flags,
+			u16 token,
+			struct dpni_attr *attr)
+{
+	struct fsl_mc_command cmd = { 0 };
+	struct dpni_rsp_get_attr *rsp_params;
+
+	int err;
+
+	/* prepare command */
+	cmd.header = mc_encode_cmd_header(DPNI_CMDID_GET_ATTR,
+					  cmd_flags,
+					  token);
+
+	/* send command to mc*/
+	err = mc_send_command(mc_io, &cmd);
+	if (err)
+		return err;
+
+	/* retrieve response parameters */
+	rsp_params = (struct dpni_rsp_get_attr *)cmd.params;
+	attr->options = le32_to_cpu(rsp_params->options);
+	attr->num_queues = rsp_params->num_queues;
+	attr->num_tcs = rsp_params->num_tcs;
+	attr->mac_filter_entries = rsp_params->mac_filter_entries;
+	attr->vlan_filter_entries = rsp_params->vlan_filter_entries;
+	attr->qos_entries = rsp_params->qos_entries;
+	attr->fs_entries = le16_to_cpu(rsp_params->fs_entries);
+	attr->qos_key_size = rsp_params->qos_key_size;
+	attr->fs_key_size = rsp_params->fs_key_size;
+	attr->wriop_version = le16_to_cpu(rsp_params->wriop_version);
+
+	return 0;
+}
+
+/**
+ * dpni_set_errors_behavior() - Set errors behavior
+ * @mc_io:	Pointer to MC portal's I/O object
+ * @cmd_flags:	Command flags; one or more of 'MC_CMD_FLAG_'
+ * @token:	Token of DPNI object
+ * @cfg:	Errors configuration
+ *
+ * this function may be called numerous times with different
+ * error masks
+ *
+ * Return:	'0' on Success; Error code otherwise.
+ */
+int dpni_set_errors_behavior(struct fsl_mc_io *mc_io,
+			     u32 cmd_flags,
+			     u16 token,
+			     struct dpni_error_cfg *cfg)
+{
+	struct fsl_mc_command cmd = { 0 };
+	struct dpni_cmd_set_errors_behavior *cmd_params;
+
+	/* prepare command */
+	cmd.header = mc_encode_cmd_header(DPNI_CMDID_SET_ERRORS_BEHAVIOR,
+					  cmd_flags,
+					  token);
+	cmd_params = (struct dpni_cmd_set_errors_behavior *)cmd.params;
+	cmd_params->errors = cpu_to_le32(cfg->errors);
+	dpni_set_field(cmd_params->flags, ERROR_ACTION, cfg->error_action);
+	dpni_set_field(cmd_params->flags, FRAME_ANN, cfg->set_frame_annotation);
+
+	/* send command to mc*/
+	return mc_send_command(mc_io, &cmd);
+}
+
+/**
+ * dpni_get_buffer_layout() - Retrieve buffer layout attributes.
+ * @mc_io:	Pointer to MC portal's I/O object
+ * @cmd_flags:	Command flags; one or more of 'MC_CMD_FLAG_'
+ * @token:	Token of DPNI object
+ * @qtype:	Type of queue to retrieve configuration for
+ * @layout:	Returns buffer layout attributes
+ *
+ * Return:	'0' on Success; Error code otherwise.
+ */
+int dpni_get_buffer_layout(struct fsl_mc_io *mc_io,
+			   u32 cmd_flags,
+			   u16 token,
+			   enum dpni_queue_type qtype,
+			   struct dpni_buffer_layout *layout)
+{
+	struct fsl_mc_command cmd = { 0 };
+	struct dpni_cmd_get_buffer_layout *cmd_params;
+	struct dpni_rsp_get_buffer_layout *rsp_params;
+	int err;
+
+	/* prepare command */
+	cmd.header = mc_encode_cmd_header(DPNI_CMDID_GET_BUFFER_LAYOUT,
+					  cmd_flags,
+					  token);
+	cmd_params = (struct dpni_cmd_get_buffer_layout *)cmd.params;
+	cmd_params->qtype = qtype;
+
+	/* send command to mc*/
+	err = mc_send_command(mc_io, &cmd);
+	if (err)
+		return err;
+
+	/* retrieve response parameters */
+	rsp_params = (struct dpni_rsp_get_buffer_layout *)cmd.params;
+	layout->pass_timestamp = dpni_get_field(rsp_params->flags, PASS_TS);
+	layout->pass_parser_result = dpni_get_field(rsp_params->flags, PASS_PR);
+	layout->pass_frame_status = dpni_get_field(rsp_params->flags, PASS_FS);
+	layout->private_data_size = le16_to_cpu(rsp_params->private_data_size);
+	layout->data_align = le16_to_cpu(rsp_params->data_align);
+	layout->data_head_room = le16_to_cpu(rsp_params->head_room);
+	layout->data_tail_room = le16_to_cpu(rsp_params->tail_room);
+
+	return 0;
+}
+
+/**
+ * dpni_set_buffer_layout() - Set buffer layout configuration.
+ * @mc_io:	Pointer to MC portal's I/O object
+ * @cmd_flags:	Command flags; one or more of 'MC_CMD_FLAG_'
+ * @token:	Token of DPNI object
+ * @qtype:	Type of queue this configuration applies to
+ * @layout:	Buffer layout configuration
+ *
+ * Return:	'0' on Success; Error code otherwise.
+ *
+ * @warning	Allowed only when DPNI is disabled
+ */
+int dpni_set_buffer_layout(struct fsl_mc_io *mc_io,
+			   u32 cmd_flags,
+			   u16 token,
+			   enum dpni_queue_type qtype,
+			   const struct dpni_buffer_layout *layout)
+{
+	struct fsl_mc_command cmd = { 0 };
+	struct dpni_cmd_set_buffer_layout *cmd_params;
+
+	/* prepare command */
+	cmd.header = mc_encode_cmd_header(DPNI_CMDID_SET_BUFFER_LAYOUT,
+					  cmd_flags,
+					  token);
+	cmd_params = (struct dpni_cmd_set_buffer_layout *)cmd.params;
+	cmd_params->qtype = qtype;
+	cmd_params->options = cpu_to_le16(layout->options);
+	dpni_set_field(cmd_params->flags, PASS_TS, layout->pass_timestamp);
+	dpni_set_field(cmd_params->flags, PASS_PR, layout->pass_parser_result);
+	dpni_set_field(cmd_params->flags, PASS_FS, layout->pass_frame_status);
+	cmd_params->private_data_size = cpu_to_le16(layout->private_data_size);
+	cmd_params->data_align = cpu_to_le16(layout->data_align);
+	cmd_params->head_room = cpu_to_le16(layout->data_head_room);
+	cmd_params->tail_room = cpu_to_le16(layout->data_tail_room);
+
+	/* send command to mc*/
+	return mc_send_command(mc_io, &cmd);
+}
+
+/**
+ * dpni_set_offload() - Set DPNI offload configuration.
+ * @mc_io:	Pointer to MC portal's I/O object
+ * @cmd_flags:	Command flags; one or more of 'MC_CMD_FLAG_'
+ * @token:	Token of DPNI object
+ * @type:	Type of DPNI offload
+ * @config:	Offload configuration.
+ *		For checksum offloads, non-zero value enables the offload
+ *
+ * Return:     '0' on Success; Error code otherwise.
+ *
+ * @warning    Allowed only when DPNI is disabled
+ */
+
+int dpni_set_offload(struct fsl_mc_io *mc_io,
+		     u32 cmd_flags,
+		     u16 token,
+		     enum dpni_offload type,
+		     u32 config)
+{
+	struct fsl_mc_command cmd = { 0 };
+	struct dpni_cmd_set_offload *cmd_params;
+
+	cmd.header = mc_encode_cmd_header(DPNI_CMDID_SET_OFFLOAD,
+					  cmd_flags,
+					  token);
+	cmd_params = (struct dpni_cmd_set_offload *)cmd.params;
+	cmd_params->dpni_offload = type;
+	cmd_params->config = cpu_to_le32(config);
+
+	return mc_send_command(mc_io, &cmd);
+}
+
+int dpni_get_offload(struct fsl_mc_io *mc_io,
+		     u32 cmd_flags,
+		     u16 token,
+		     enum dpni_offload type,
+		     u32 *config)
+{
+	struct fsl_mc_command cmd = { 0 };
+	struct dpni_cmd_get_offload *cmd_params;
+	struct dpni_rsp_get_offload *rsp_params;
+	int err;
+
+	/* prepare command */
+	cmd.header = mc_encode_cmd_header(DPNI_CMDID_GET_OFFLOAD,
+					  cmd_flags,
+					  token);
+	cmd_params = (struct dpni_cmd_get_offload *)cmd.params;
+	cmd_params->dpni_offload = type;
+
+	/* send command to mc*/
+	err = mc_send_command(mc_io, &cmd);
+	if (err)
+		return err;
+
+	/* retrieve response parameters */
+	rsp_params = (struct dpni_rsp_get_offload *)cmd.params;
+	*config = le32_to_cpu(rsp_params->config);
+
+	return 0;
+}
+
+/**
+ * dpni_get_qdid() - Get the Queuing Destination ID (QDID) that should be used
+ *			for enqueue operations
+ * @mc_io:	Pointer to MC portal's I/O object
+ * @cmd_flags:	Command flags; one or more of 'MC_CMD_FLAG_'
+ * @token:	Token of DPNI object
+ * @qtype:	Type of queue to receive QDID for
+ * @qdid:	Returned virtual QDID value that should be used as an argument
+ *			in all enqueue operations
+ *
+ * Return:	'0' on Success; Error code otherwise.
+ */
+int dpni_get_qdid(struct fsl_mc_io *mc_io,
+		  u32 cmd_flags,
+		  u16 token,
+		  enum dpni_queue_type qtype,
+		  u16 *qdid)
+{
+	struct fsl_mc_command cmd = { 0 };
+	struct dpni_cmd_get_qdid *cmd_params;
+	struct dpni_rsp_get_qdid *rsp_params;
+	int err;
+
+	/* prepare command */
+	cmd.header = mc_encode_cmd_header(DPNI_CMDID_GET_QDID,
+					  cmd_flags,
+					  token);
+	cmd_params = (struct dpni_cmd_get_qdid *)cmd.params;
+	cmd_params->qtype = qtype;
+
+	/* send command to mc*/
+	err = mc_send_command(mc_io, &cmd);
+	if (err)
+		return err;
+
+	/* retrieve response parameters */
+	rsp_params = (struct dpni_rsp_get_qdid *)cmd.params;
+	*qdid = le16_to_cpu(rsp_params->qdid);
+
+	return 0;
+}
+
+/**
+ * dpni_get_tx_data_offset() - Get the Tx data offset (from start of buffer)
+ * @mc_io:	Pointer to MC portal's I/O object
+ * @cmd_flags:	Command flags; one or more of 'MC_CMD_FLAG_'
+ * @token:	Token of DPNI object
+ * @data_offset: Tx data offset (from start of buffer)
+ *
+ * Return:	'0' on Success; Error code otherwise.
+ */
+int dpni_get_tx_data_offset(struct fsl_mc_io *mc_io,
+			    u32 cmd_flags,
+			    u16 token,
+			    u16 *data_offset)
+{
+	struct fsl_mc_command cmd = { 0 };
+	struct dpni_rsp_get_tx_data_offset *rsp_params;
+	int err;
+
+	/* prepare command */
+	cmd.header = mc_encode_cmd_header(DPNI_CMDID_GET_TX_DATA_OFFSET,
+					  cmd_flags,
+					  token);
+
+	/* send command to mc*/
+	err = mc_send_command(mc_io, &cmd);
+	if (err)
+		return err;
+
+	/* retrieve response parameters */
+	rsp_params = (struct dpni_rsp_get_tx_data_offset *)cmd.params;
+	*data_offset = le16_to_cpu(rsp_params->data_offset);
+
+	return 0;
+}
+
+/**
+ * dpni_set_link_cfg() - set the link configuration.
+ * @mc_io:	Pointer to MC portal's I/O object
+ * @cmd_flags:	Command flags; one or more of 'MC_CMD_FLAG_'
+ * @token:	Token of DPNI object
+ * @cfg:	Link configuration
+ *
+ * Return:	'0' on Success; Error code otherwise.
+ */
+int dpni_set_link_cfg(struct fsl_mc_io *mc_io,
+		      u32 cmd_flags,
+		      u16 token,
+		      const struct dpni_link_cfg *cfg)
+{
+	struct fsl_mc_command cmd = { 0 };
+	struct dpni_cmd_link_cfg *cmd_params;
+
+	/* prepare command */
+	cmd.header = mc_encode_cmd_header(DPNI_CMDID_SET_LINK_CFG,
+					  cmd_flags,
+					  token);
+	cmd_params = (struct dpni_cmd_link_cfg *)cmd.params;
+	cmd_params->rate = cpu_to_le32(cfg->rate);
+	cmd_params->options = cpu_to_le64(cfg->options);
+
+	/* send command to mc*/
+	return mc_send_command(mc_io, &cmd);
+}
+
+/**
+ * dpni_get_link_cfg() - return the link configuration
+ * @mc_io:	Pointer to MC portal's I/O object
+ * @cmd_flags:	Command flags; one or more of 'MC_CMD_FLAG_'
+ * @token:	Token of DPNI object
+ * @cfg:	Link configuration from dpni object
+ *
+ * Return:	'0' on Success; Error code otherwise.
+ */
+int dpni_get_link_cfg(struct fsl_mc_io *mc_io,
+		      u32 cmd_flags,
+		      u16 token,
+		      struct dpni_link_cfg *cfg)
+{
+	struct fsl_mc_command cmd = { 0 };
+	struct dpni_cmd_link_cfg *rsp_params;
+	int err;
+
+	/* prepare command */
+	cmd.header = mc_encode_cmd_header(DPNI_CMDID_GET_LINK_CFG,
+					  cmd_flags,
+					  token);
+
+	/* send command to mc*/
+	err = mc_send_command(mc_io, &cmd);
+	if (err)
+		return err;
+
+	/* retrieve response parameters */
+	rsp_params = (struct dpni_cmd_link_cfg *)cmd.params;
+	cfg->rate = le32_to_cpu(rsp_params->rate);
+	cfg->options = le64_to_cpu(rsp_params->options);
+
+	return err;
+}
+
+/**
+ * dpni_get_link_state() - Return the link state (either up or down)
+ * @mc_io:	Pointer to MC portal's I/O object
+ * @cmd_flags:	Command flags; one or more of 'MC_CMD_FLAG_'
+ * @token:	Token of DPNI object
+ * @state:	Returned link state;
+ *
+ * Return:	'0' on Success; Error code otherwise.
+ */
+int dpni_get_link_state(struct fsl_mc_io *mc_io,
+			u32 cmd_flags,
+			u16 token,
+			struct dpni_link_state *state)
+{
+	struct fsl_mc_command cmd = { 0 };
+	struct dpni_rsp_get_link_state *rsp_params;
+	int err;
+
+	/* prepare command */
+	cmd.header = mc_encode_cmd_header(DPNI_CMDID_GET_LINK_STATE,
+					  cmd_flags,
+					  token);
+
+	/* send command to mc*/
+	err = mc_send_command(mc_io, &cmd);
+	if (err)
+		return err;
+
+	/* retrieve response parameters */
+	rsp_params = (struct dpni_rsp_get_link_state *)cmd.params;
+	state->up = dpni_get_field(rsp_params->flags, LINK_STATE);
+	state->rate = le32_to_cpu(rsp_params->rate);
+	state->options = le64_to_cpu(rsp_params->options);
+
+	return 0;
+}
+
+/**
+ * dpni_set_max_frame_length() - Set the maximum received frame length.
+ * @mc_io:	Pointer to MC portal's I/O object
+ * @cmd_flags:	Command flags; one or more of 'MC_CMD_FLAG_'
+ * @token:	Token of DPNI object
+ * @max_frame_length:	Maximum received frame length (in
+ *				bytes); frame is discarded if its
+ *				length exceeds this value
+ *
+ * Return:	'0' on Success; Error code otherwise.
+ */
+int dpni_set_max_frame_length(struct fsl_mc_io *mc_io,
+			      u32 cmd_flags,
+			      u16 token,
+			      u16 max_frame_length)
+{
+	struct fsl_mc_command cmd = { 0 };
+	struct dpni_cmd_set_max_frame_length *cmd_params;
+
+	/* prepare command */
+	cmd.header = mc_encode_cmd_header(DPNI_CMDID_SET_MAX_FRAME_LENGTH,
+					  cmd_flags,
+					  token);
+	cmd_params = (struct dpni_cmd_set_max_frame_length *)cmd.params;
+	cmd_params->max_frame_length = cpu_to_le16(max_frame_length);
+
+	/* send command to mc*/
+	return mc_send_command(mc_io, &cmd);
+}
+
+/**
+ * dpni_get_max_frame_length() - Get the maximum received frame length.
+ * @mc_io:	Pointer to MC portal's I/O object
+ * @cmd_flags:	Command flags; one or more of 'MC_CMD_FLAG_'
+ * @token:	Token of DPNI object
+ * @max_frame_length:	Maximum received frame length (in
+ *				bytes); frame is discarded if its
+ *				length exceeds this value
+ *
+ * Return:	'0' on Success; Error code otherwise.
+ */
+int dpni_get_max_frame_length(struct fsl_mc_io *mc_io,
+			      u32 cmd_flags,
+			      u16 token,
+			      u16 *max_frame_length)
+{
+	struct fsl_mc_command cmd = { 0 };
+	struct dpni_rsp_get_max_frame_length *rsp_params;
+	int err;
+
+	/* prepare command */
+	cmd.header = mc_encode_cmd_header(DPNI_CMDID_GET_MAX_FRAME_LENGTH,
+					  cmd_flags,
+					  token);
+
+	/* send command to mc*/
+	err = mc_send_command(mc_io, &cmd);
+	if (err)
+		return err;
+
+	/* retrieve response parameters */
+	rsp_params = (struct dpni_rsp_get_max_frame_length *)cmd.params;
+	*max_frame_length = le16_to_cpu(rsp_params->max_frame_length);
+
+	return 0;
+}
+
+/**
+ * dpni_set_multicast_promisc() - Enable/disable multicast promiscuous mode
+ * @mc_io:	Pointer to MC portal's I/O object
+ * @cmd_flags:	Command flags; one or more of 'MC_CMD_FLAG_'
+ * @token:	Token of DPNI object
+ * @en:		Set to '1' to enable; '0' to disable
+ *
+ * Return:	'0' on Success; Error code otherwise.
+ */
+int dpni_set_multicast_promisc(struct fsl_mc_io *mc_io,
+			       u32 cmd_flags,
+			       u16 token,
+			       int en)
+{
+	struct fsl_mc_command cmd = { 0 };
+	struct dpni_cmd_set_multicast_promisc *cmd_params;
+
+	/* prepare command */
+	cmd.header = mc_encode_cmd_header(DPNI_CMDID_SET_MCAST_PROMISC,
+					  cmd_flags,
+					  token);
+	cmd_params = (struct dpni_cmd_set_multicast_promisc *)cmd.params;
+	dpni_set_field(cmd_params->enable, ENABLE, en);
+
+	/* send command to mc*/
+	return mc_send_command(mc_io, &cmd);
+}
+
+/**
+ * dpni_get_multicast_promisc() - Get multicast promiscuous mode
+ * @mc_io:	Pointer to MC portal's I/O object
+ * @cmd_flags:	Command flags; one or more of 'MC_CMD_FLAG_'
+ * @token:	Token of DPNI object
+ * @en:		Returns '1' if enabled; '0' otherwise
+ *
+ * Return:	'0' on Success; Error code otherwise.
+ */
+int dpni_get_multicast_promisc(struct fsl_mc_io *mc_io,
+			       u32 cmd_flags,
+			       u16 token,
+			       int *en)
+{
+	struct fsl_mc_command cmd = { 0 };
+	struct dpni_rsp_get_multicast_promisc *rsp_params;
+	int err;
+
+	/* prepare command */
+	cmd.header = mc_encode_cmd_header(DPNI_CMDID_GET_MCAST_PROMISC,
+					  cmd_flags,
+					  token);
+
+	/* send command to mc*/
+	err = mc_send_command(mc_io, &cmd);
+	if (err)
+		return err;
+
+	/* retrieve response parameters */
+	rsp_params = (struct dpni_rsp_get_multicast_promisc *)cmd.params;
+	*en = dpni_get_field(rsp_params->enabled, ENABLE);
+
+	return 0;
+}
+
+/**
+ * dpni_set_unicast_promisc() - Enable/disable unicast promiscuous mode
+ * @mc_io:	Pointer to MC portal's I/O object
+ * @cmd_flags:	Command flags; one or more of 'MC_CMD_FLAG_'
+ * @token:	Token of DPNI object
+ * @en:		Set to '1' to enable; '0' to disable
+ *
+ * Return:	'0' on Success; Error code otherwise.
+ */
+int dpni_set_unicast_promisc(struct fsl_mc_io *mc_io,
+			     u32 cmd_flags,
+			     u16 token,
+			     int en)
+{
+	struct fsl_mc_command cmd = { 0 };
+	struct dpni_cmd_set_unicast_promisc *cmd_params;
+
+	/* prepare command */
+	cmd.header = mc_encode_cmd_header(DPNI_CMDID_SET_UNICAST_PROMISC,
+					  cmd_flags,
+					  token);
+	cmd_params = (struct dpni_cmd_set_unicast_promisc *)cmd.params;
+	dpni_set_field(cmd_params->enable, ENABLE, en);
+
+	/* send command to mc*/
+	return mc_send_command(mc_io, &cmd);
+}
+
+/**
+ * dpni_get_unicast_promisc() - Get unicast promiscuous mode
+ * @mc_io:	Pointer to MC portal's I/O object
+ * @cmd_flags:	Command flags; one or more of 'MC_CMD_FLAG_'
+ * @token:	Token of DPNI object
+ * @en:		Returns '1' if enabled; '0' otherwise
+ *
+ * Return:	'0' on Success; Error code otherwise.
+ */
+int dpni_get_unicast_promisc(struct fsl_mc_io *mc_io,
+			     u32 cmd_flags,
+			     u16 token,
+			     int *en)
+{
+	struct fsl_mc_command cmd = { 0 };
+	struct dpni_rsp_get_unicast_promisc *rsp_params;
+	int err;
+
+	/* prepare command */
+	cmd.header = mc_encode_cmd_header(DPNI_CMDID_GET_UNICAST_PROMISC,
+					  cmd_flags,
+					  token);
+
+	/* send command to mc*/
+	err = mc_send_command(mc_io, &cmd);
+	if (err)
+		return err;
+
+	/* retrieve response parameters */
+	rsp_params = (struct dpni_rsp_get_unicast_promisc *)cmd.params;
+	*en = dpni_get_field(rsp_params->enabled, ENABLE);
+
+	return 0;
+}
+
+/**
+ * dpni_set_primary_mac_addr() - Set the primary MAC address
+ * @mc_io:	Pointer to MC portal's I/O object
+ * @cmd_flags:	Command flags; one or more of 'MC_CMD_FLAG_'
+ * @token:	Token of DPNI object
+ * @mac_addr:	MAC address to set as primary address
+ *
+ * Return:	'0' on Success; Error code otherwise.
+ */
+int dpni_set_primary_mac_addr(struct fsl_mc_io *mc_io,
+			      u32 cmd_flags,
+			      u16 token,
+			      const u8 mac_addr[6])
+{
+	struct fsl_mc_command cmd = { 0 };
+	struct dpni_cmd_set_primary_mac_addr *cmd_params;
+	int i;
+
+	/* prepare command */
+	cmd.header = mc_encode_cmd_header(DPNI_CMDID_SET_PRIM_MAC,
+					  cmd_flags,
+					  token);
+	cmd_params = (struct dpni_cmd_set_primary_mac_addr *)cmd.params;
+	for (i = 0; i < 6; i++)
+		cmd_params->mac_addr[i] = mac_addr[5 - i];
+
+	/* send command to mc*/
+	return mc_send_command(mc_io, &cmd);
+}
+
+/**
+ * dpni_get_primary_mac_addr() - Get the primary MAC address
+ * @mc_io:	Pointer to MC portal's I/O object
+ * @cmd_flags:	Command flags; one or more of 'MC_CMD_FLAG_'
+ * @token:	Token of DPNI object
+ * @mac_addr:	Returned MAC address
+ *
+ * Return:	'0' on Success; Error code otherwise.
+ */
+int dpni_get_primary_mac_addr(struct fsl_mc_io *mc_io,
+			      u32 cmd_flags,
+			      u16 token,
+			      u8 mac_addr[6])
+{
+	struct fsl_mc_command cmd = { 0 };
+	struct dpni_rsp_get_primary_mac_addr *rsp_params;
+	int i, err;
+
+	/* prepare command */
+	cmd.header = mc_encode_cmd_header(DPNI_CMDID_GET_PRIM_MAC,
+					  cmd_flags,
+					  token);
+
+	/* send command to mc*/
+	err = mc_send_command(mc_io, &cmd);
+	if (err)
+		return err;
+
+	/* retrieve response parameters */
+	rsp_params = (struct dpni_rsp_get_primary_mac_addr *)cmd.params;
+	for (i = 0; i < 6; i++)
+		mac_addr[5 - i] = rsp_params->mac_addr[i];
+
+	return 0;
+}
+
+/**
+ * dpni_get_port_mac_addr() - Retrieve MAC address associated to the physical
+ *			port the DPNI is attached to
+ * @mc_io:	Pointer to MC portal's I/O object
+ * @cmd_flags:	Command flags; one or more of 'MC_CMD_FLAG_'
+ * @token:	Token of DPNI object
+ * @mac_addr:	MAC address of the physical port, if any, otherwise 0
+ *
+ * The primary MAC address is not cleared by this operation.
+ *
+ * Return:	'0' on Success; Error code otherwise.
+ */
+int dpni_get_port_mac_addr(struct fsl_mc_io *mc_io,
+			   u32 cmd_flags,
+			   u16 token,
+			   u8 mac_addr[6])
+{
+	struct fsl_mc_command cmd = { 0 };
+	struct dpni_rsp_get_port_mac_addr *rsp_params;
+	int i, err;
+
+	/* prepare command */
+	cmd.header = mc_encode_cmd_header(DPNI_CMDID_GET_PORT_MAC_ADDR,
+					  cmd_flags,
+					  token);
+
+	/* send command to mc*/
+	err = mc_send_command(mc_io, &cmd);
+	if (err)
+		return err;
+
+	/* retrieve response parameters */
+	rsp_params = (struct dpni_rsp_get_port_mac_addr *)cmd.params;
+	for (i = 0; i < 6; i++)
+		mac_addr[5 - i] = rsp_params->mac_addr[i];
+
+	return 0;
+}
+
+/**
+ * dpni_enable_vlan_filter() - Enable/disable VLAN filtering mode
+ * @mc_io:	Pointer to MC portal's I/O object
+ * @cmd_flags:	Command flags; one or more of 'MC_CMD_FLAG_'
+ * @token:	Token of DPNI object
+ * @en:		Set to '1' to enable; '0' to disable
+ *
+ * Return:	'0' on Success; Error code otherwise.
+ */
+int dpni_enable_vlan_filter(struct fsl_mc_io *mc_io,
+			    u32 cmd_flags,
+			    u16 token,
+			    u32 en)
+{
+	struct dpni_cmd_enable_vlan_filter *cmd_params;
+	struct fsl_mc_command cmd = { 0 };
+
+	/* prepare command */
+	cmd.header = mc_encode_cmd_header(DPNI_CMDID_ENABLE_VLAN_FILTER,
+					  cmd_flags,
+					  token);
+	cmd_params = (struct dpni_cmd_enable_vlan_filter *)cmd.params;
+	dpni_set_field(cmd_params->en, ENABLE, en);
+
+	/* send command to mc*/
+	return mc_send_command(mc_io, &cmd);
+}
+
+/**
+ * dpni_add_vlan_id() - Add VLAN ID filter
+ * @mc_io:	Pointer to MC portal's I/O object
+ * @cmd_flags:	Command flags; one or more of 'MC_CMD_FLAG_'
+ * @token:	Token of DPNI object
+ * @vlan_id:	VLAN ID to add
+ * @flags:   0 - tc_id and flow_id will be ignored.
+ * Pkt with this vlan_id will be passed to the next
+ * classification stages
+ * DPNI_VLAN_SET_QUEUE_ACTION
+ * Pkt with this vlan_id will be forward directly to
+ * queue defined by the tc_id and flow_id
+ *
+ * @tc_id: Traffic class selection (0-7)
+ * @flow_id: Selects the specific queue out of the set allocated for the
+ *           same as tc_id. Value must be in range 0 to NUM_QUEUES - 1
+ *
+ * Return:	'0' on Success; Error code otherwise.
+ */
+int dpni_add_vlan_id(struct fsl_mc_io *mc_io, u32 cmd_flags, u16 token,
+		     u16 vlan_id, u8 flags, u8 tc_id, u8 flow_id)
+{
+	struct dpni_cmd_vlan_id *cmd_params;
+	struct fsl_mc_command cmd = { 0 };
+
+	/* prepare command */
+	cmd.header = mc_encode_cmd_header(DPNI_CMDID_ADD_VLAN_ID,
+					  cmd_flags,
+					  token);
+	cmd_params = (struct dpni_cmd_vlan_id *)cmd.params;
+	cmd_params->flags = flags;
+	cmd_params->tc_id = tc_id;
+	cmd_params->flow_id =  flow_id;
+	cmd_params->vlan_id = cpu_to_le16(vlan_id);
+
+	/* send command to mc*/
+	return mc_send_command(mc_io, &cmd);
+}
+
+/**
+ * dpni_remove_vlan_id() - Remove VLAN ID filter
+ * @mc_io:	Pointer to MC portal's I/O object
+ * @cmd_flags:	Command flags; one or more of 'MC_CMD_FLAG_'
+ * @token:	Token of DPNI object
+ * @vlan_id:	VLAN ID to remove
+ *
+ * Return:	'0' on Success; Error code otherwise.
+ */
+int dpni_remove_vlan_id(struct fsl_mc_io *mc_io, u32 cmd_flags, u16 token,
+			u16 vlan_id)
+{
+	struct dpni_cmd_vlan_id *cmd_params;
+	struct fsl_mc_command cmd = { 0 };
+
+	/* prepare command */
+	cmd.header = mc_encode_cmd_header(DPNI_CMDID_REMOVE_VLAN_ID,
+					  cmd_flags,
+					  token);
+	cmd_params = (struct dpni_cmd_vlan_id *)cmd.params;
+	cmd_params->vlan_id = cpu_to_le16(vlan_id);
+
+	/* send command to mc*/
+	return mc_send_command(mc_io, &cmd);
+}
+
+/**
+ * dpni_add_mac_addr() - Add MAC address filter
+ * @mc_io:	Pointer to MC portal's I/O object
+ * @cmd_flags:	Command flags; one or more of 'MC_CMD_FLAG_'
+ * @token:	Token of DPNI object
+ * @mac_addr:	MAC address to add
+ *
+ * Return:	'0' on Success; Error code otherwise.
+ */
+int dpni_add_mac_addr(struct fsl_mc_io *mc_io,
+		      u32 cmd_flags,
+		      u16 token,
+		      const u8 mac_addr[6])
+{
+	struct fsl_mc_command cmd = { 0 };
+	struct dpni_cmd_add_mac_addr *cmd_params;
+	int i;
+
+	/* prepare command */
+	cmd.header = mc_encode_cmd_header(DPNI_CMDID_ADD_MAC_ADDR,
+					  cmd_flags,
+					  token);
+	cmd_params = (struct dpni_cmd_add_mac_addr *)cmd.params;
+	for (i = 0; i < 6; i++)
+		cmd_params->mac_addr[i] = mac_addr[5 - i];
+
+	/* send command to mc*/
+	return mc_send_command(mc_io, &cmd);
+}
+
+/**
+ * dpni_remove_mac_addr() - Remove MAC address filter
+ * @mc_io:	Pointer to MC portal's I/O object
+ * @cmd_flags:	Command flags; one or more of 'MC_CMD_FLAG_'
+ * @token:	Token of DPNI object
+ * @mac_addr:	MAC address to remove
+ *
+ * Return:	'0' on Success; Error code otherwise.
+ */
+int dpni_remove_mac_addr(struct fsl_mc_io *mc_io,
+			 u32 cmd_flags,
+			 u16 token,
+			 const u8 mac_addr[6])
+{
+	struct fsl_mc_command cmd = { 0 };
+	struct dpni_cmd_remove_mac_addr *cmd_params;
+	int i;
+
+	/* prepare command */
+	cmd.header = mc_encode_cmd_header(DPNI_CMDID_REMOVE_MAC_ADDR,
+					  cmd_flags,
+					  token);
+	cmd_params = (struct dpni_cmd_remove_mac_addr *)cmd.params;
+	for (i = 0; i < 6; i++)
+		cmd_params->mac_addr[i] = mac_addr[5 - i];
+
+	/* send command to mc*/
+	return mc_send_command(mc_io, &cmd);
+}
+
+/**
+ * dpni_clear_mac_filters() - Clear all unicast and/or multicast MAC filters
+ * @mc_io:	Pointer to MC portal's I/O object
+ * @cmd_flags:	Command flags; one or more of 'MC_CMD_FLAG_'
+ * @token:	Token of DPNI object
+ * @unicast:	Set to '1' to clear unicast addresses
+ * @multicast:	Set to '1' to clear multicast addresses
+ *
+ * The primary MAC address is not cleared by this operation.
+ *
+ * Return:	'0' on Success; Error code otherwise.
+ */
+int dpni_clear_mac_filters(struct fsl_mc_io *mc_io,
+			   u32 cmd_flags,
+			   u16 token,
+			   int unicast,
+			   int multicast)
+{
+	struct fsl_mc_command cmd = { 0 };
+	struct dpni_cmd_clear_mac_filters *cmd_params;
+
+	/* prepare command */
+	cmd.header = mc_encode_cmd_header(DPNI_CMDID_CLR_MAC_FILTERS,
+					  cmd_flags,
+					  token);
+	cmd_params = (struct dpni_cmd_clear_mac_filters *)cmd.params;
+	dpni_set_field(cmd_params->flags, UNICAST_FILTERS, unicast);
+	dpni_set_field(cmd_params->flags, MULTICAST_FILTERS, multicast);
+
+	/* send command to mc*/
+	return mc_send_command(mc_io, &cmd);
+}
+
+/**
+ * dpni_set_rx_tc_dist() - Set Rx traffic class distribution configuration
+ * @mc_io:	Pointer to MC portal's I/O object
+ * @cmd_flags:	Command flags; one or more of 'MC_CMD_FLAG_'
+ * @token:	Token of DPNI object
+ * @tc_id:	Traffic class selection (0-7)
+ * @cfg:	Traffic class distribution configuration
+ *
+ * warning: if 'dist_mode != DPNI_DIST_MODE_NONE', call dpni_prepare_key_cfg()
+ *			first to prepare the key_cfg_iova parameter
+ *
+ * Return:	'0' on Success; error code otherwise.
+ */
+int dpni_set_rx_tc_dist(struct fsl_mc_io *mc_io,
+			u32 cmd_flags,
+			u16 token,
+			u8 tc_id,
+			const struct dpni_rx_tc_dist_cfg *cfg)
+{
+	struct fsl_mc_command cmd = { 0 };
+	struct dpni_cmd_set_rx_tc_dist *cmd_params;
+
+	/* prepare command */
+	cmd.header = mc_encode_cmd_header(DPNI_CMDID_SET_RX_TC_DIST,
+					  cmd_flags,
+					  token);
+	cmd_params = (struct dpni_cmd_set_rx_tc_dist *)cmd.params;
+	cmd_params->dist_size = cpu_to_le16(cfg->dist_size);
+	cmd_params->tc_id = tc_id;
+	dpni_set_field(cmd_params->flags, DIST_MODE, cfg->dist_mode);
+	dpni_set_field(cmd_params->flags, MISS_ACTION, cfg->fs_cfg.miss_action);
+	cmd_params->default_flow_id = cpu_to_le16(cfg->fs_cfg.default_flow_id);
+	cmd_params->key_cfg_iova = cpu_to_le64(cfg->key_cfg_iova);
+
+	/* send command to mc*/
+	return mc_send_command(mc_io, &cmd);
+}
+
+/**
+ * dpni_set_congestion_notification() - Set traffic class congestion
+ *					notification configuration
+ * @mc_io:	Pointer to MC portal's I/O object
+ * @cmd_flags:	Command flags; one or more of 'MC_CMD_FLAG_'
+ * @token:	Token of DPNI object
+ * @qtype:	Type of queue - Rx, Tx and Tx confirm types are supported
+ * @tc_id:	Traffic class selection (0-7)
+ * @cfg:	Congestion notification configuration
+ *
+ * Return:	'0' on Success; error code otherwise.
+ */
+int dpni_set_congestion_notification(
+			struct fsl_mc_io *mc_io,
+			u32 cmd_flags,
+			u16 token,
+			enum dpni_queue_type qtype,
+			u8 tc_id,
+			const struct dpni_congestion_notification_cfg *cfg)
+{
+	struct dpni_cmd_set_congestion_notification *cmd_params;
+	struct fsl_mc_command cmd = { 0 };
+
+	/* prepare command */
+	cmd.header =
+		mc_encode_cmd_header(DPNI_CMDID_SET_CONGESTION_NOTIFICATION,
+				     cmd_flags,
+				     token);
+	cmd_params = (struct dpni_cmd_set_congestion_notification *)cmd.params;
+	cmd_params->qtype = qtype;
+	cmd_params->tc = tc_id;
+	cmd_params->dest_id = cpu_to_le32(cfg->dest_cfg.dest_id);
+	cmd_params->notification_mode = cpu_to_le16(cfg->notification_mode);
+	cmd_params->dest_priority = cfg->dest_cfg.priority;
+	dpni_set_field(cmd_params->type_units, DEST_TYPE,
+		       cfg->dest_cfg.dest_type);
+	dpni_set_field(cmd_params->type_units, CONG_UNITS, cfg->units);
+	cmd_params->message_iova = cpu_to_le64(cfg->message_iova);
+	cmd_params->message_ctx = cpu_to_le64(cfg->message_ctx);
+	cmd_params->threshold_entry = cpu_to_le32(cfg->threshold_entry);
+	cmd_params->threshold_exit = cpu_to_le32(cfg->threshold_exit);
+
+	/* send command to mc*/
+	return mc_send_command(mc_io, &cmd);
+}
+
+/**
+ * dpni_set_queue() - Set queue parameters
+ * @mc_io:	Pointer to MC portal's I/O object
+ * @cmd_flags:	Command flags; one or more of 'MC_CMD_FLAG_'
+ * @token:	Token of DPNI object
+ * @qtype:	Type of queue - all queue types are supported, although
+ *		the command is ignored for Tx
+ * @tc:		Traffic class, in range 0 to NUM_TCS - 1
+ * @index:	Selects the specific queue out of the set allocated for the
+ *		same TC. Value must be in range 0 to NUM_QUEUES - 1
+ * @options:	A combination of DPNI_QUEUE_OPT_ values that control what
+ *		configuration options are set on the queue
+ * @queue:	Queue structure
+ *
+ * Return:	'0' on Success; Error code otherwise.
+ */
+int dpni_set_queue(struct fsl_mc_io *mc_io,
+		   u32 cmd_flags,
+		   u16 token,
+		   enum dpni_queue_type qtype,
+		   u8 tc,
+		   u8 index,
+		   u8 options,
+		   const struct dpni_queue *queue)
+{
+	struct fsl_mc_command cmd = { 0 };
+	struct dpni_cmd_set_queue *cmd_params;
+
+	/* prepare command */
+	cmd.header = mc_encode_cmd_header(DPNI_CMDID_SET_QUEUE,
+					  cmd_flags,
+					  token);
+	cmd_params = (struct dpni_cmd_set_queue *)cmd.params;
+	cmd_params->qtype = qtype;
+	cmd_params->tc = tc;
+	cmd_params->index = index;
+	cmd_params->options = options;
+	cmd_params->dest_id = cpu_to_le32(queue->destination.id);
+	cmd_params->dest_prio = queue->destination.priority;
+	dpni_set_field(cmd_params->flags, DEST_TYPE, queue->destination.type);
+	dpni_set_field(cmd_params->flags, STASH_CTRL, queue->flc.stash_control);
+	dpni_set_field(cmd_params->flags, HOLD_ACTIVE,
+		       queue->destination.hold_active);
+	cmd_params->flc = cpu_to_le64(queue->flc.value);
+	cmd_params->user_context = cpu_to_le64(queue->user_context);
+
+	/* send command to mc */
+	return mc_send_command(mc_io, &cmd);
+}
+
+/**
+ * dpni_get_queue() - Get queue parameters
+ * @mc_io:	Pointer to MC portal's I/O object
+ * @cmd_flags:	Command flags; one or more of 'MC_CMD_FLAG_'
+ * @token:	Token of DPNI object
+ * @qtype:	Type of queue - all queue types are supported
+ * @tc:		Traffic class, in range 0 to NUM_TCS - 1
+ * @index:	Selects the specific queue out of the set allocated for the
+ *		same TC. Value must be in range 0 to NUM_QUEUES - 1
+ * @queue:	Queue configuration structure
+ * @qid:	Queue identification
+ *
+ * Return:	'0' on Success; Error code otherwise.
+ */
+int dpni_get_queue(struct fsl_mc_io *mc_io,
+		   u32 cmd_flags,
+		   u16 token,
+		   enum dpni_queue_type qtype,
+		   u8 tc,
+		   u8 index,
+		   struct dpni_queue *queue,
+		   struct dpni_queue_id *qid)
+{
+	struct fsl_mc_command cmd = { 0 };
+	struct dpni_cmd_get_queue *cmd_params;
+	struct dpni_rsp_get_queue *rsp_params;
+	int err;
+
+	/* prepare command */
+	cmd.header = mc_encode_cmd_header(DPNI_CMDID_GET_QUEUE,
+					  cmd_flags,
+					  token);
+	cmd_params = (struct dpni_cmd_get_queue *)cmd.params;
+	cmd_params->qtype = qtype;
+	cmd_params->tc = tc;
+	cmd_params->index = index;
+
+	/* send command to mc */
+	err = mc_send_command(mc_io, &cmd);
+	if (err)
+		return err;
+
+	/* retrieve response parameters */
+	rsp_params = (struct dpni_rsp_get_queue *)cmd.params;
+	queue->destination.id = le32_to_cpu(rsp_params->dest_id);
+	queue->destination.priority = rsp_params->dest_prio;
+	queue->destination.type = dpni_get_field(rsp_params->flags,
+						 DEST_TYPE);
+	queue->flc.stash_control = dpni_get_field(rsp_params->flags,
+						  STASH_CTRL);
+	queue->destination.hold_active = dpni_get_field(rsp_params->flags,
+							HOLD_ACTIVE);
+	queue->flc.value = le64_to_cpu(rsp_params->flc);
+	queue->user_context = le64_to_cpu(rsp_params->user_context);
+	qid->fqid = le32_to_cpu(rsp_params->fqid);
+	qid->qdbin = le16_to_cpu(rsp_params->qdbin);
+
+	return 0;
+}
+
+/**
+ * dpni_get_statistics() - Get DPNI statistics
+ * @mc_io:	Pointer to MC portal's I/O object
+ * @cmd_flags:	Command flags; one or more of 'MC_CMD_FLAG_'
+ * @token:	Token of DPNI object
+ * @page:	Selects the statistics page to retrieve, see
+ *		DPNI_GET_STATISTICS output. Pages are numbered 0 to 6.
+ * @stat:	Structure containing the statistics
+ *
+ * Return:	'0' on Success; Error code otherwise.
+ */
+int dpni_get_statistics(struct fsl_mc_io *mc_io,
+			u32 cmd_flags,
+			u16 token,
+			u8 page,
+			union dpni_statistics *stat)
+{
+	struct fsl_mc_command cmd = { 0 };
+	struct dpni_cmd_get_statistics *cmd_params;
+	struct dpni_rsp_get_statistics *rsp_params;
+	int i, err;
+
+	/* prepare command */
+	cmd.header = mc_encode_cmd_header(DPNI_CMDID_GET_STATISTICS,
+					  cmd_flags,
+					  token);
+	cmd_params = (struct dpni_cmd_get_statistics *)cmd.params;
+	cmd_params->page_number = page;
+
+	/* send command to mc */
+	err = mc_send_command(mc_io, &cmd);
+	if (err)
+		return err;
+
+	/* retrieve response parameters */
+	rsp_params = (struct dpni_rsp_get_statistics *)cmd.params;
+	for (i = 0; i < DPNI_STATISTICS_CNT; i++)
+		stat->raw.counter[i] = le64_to_cpu(rsp_params->counter[i]);
+
+	return 0;
+}
+
+/**
+ * dpni_set_taildrop() - Set taildrop per queue or TC
+ * @mc_io:	Pointer to MC portal's I/O object
+ * @cmd_flags:	Command flags; one or more of 'MC_CMD_FLAG_'
+ * @token:	Token of DPNI object
+ * @cg_point:	Congestion point
+ * @qtype:	Queue type on which the taildrop is configured.
+ *		Only Rx queues are supported for now
+ * @tc:		Traffic class to apply this taildrop to
+ * @index:	Index of the queue if the DPNI supports multiple queues for
+ *		traffic distribution. Ignored if CONGESTION_POINT is not 0.
+ * @taildrop:	Taildrop structure
+ *
+ * Return:	'0' on Success; Error code otherwise.
+ */
+int dpni_set_taildrop(struct fsl_mc_io *mc_io,
+		      u32 cmd_flags,
+		      u16 token,
+		      enum dpni_congestion_point cg_point,
+		      enum dpni_queue_type qtype,
+		      u8 tc,
+		      u8 index,
+		      struct dpni_taildrop *taildrop)
+{
+	struct fsl_mc_command cmd = { 0 };
+	struct dpni_cmd_set_taildrop *cmd_params;
+
+	/* prepare command */
+	cmd.header = mc_encode_cmd_header(DPNI_CMDID_SET_TAILDROP,
+					  cmd_flags,
+					  token);
+	cmd_params = (struct dpni_cmd_set_taildrop *)cmd.params;
+	cmd_params->congestion_point = cg_point;
+	cmd_params->qtype = qtype;
+	cmd_params->tc = tc;
+	cmd_params->index = index;
+	dpni_set_field(cmd_params->enable, ENABLE, taildrop->enable);
+	cmd_params->units = taildrop->units;
+	cmd_params->threshold = cpu_to_le32(taildrop->threshold);
+
+	/* send command to mc */
+	return mc_send_command(mc_io, &cmd);
+}
+
+/**
+ * dpni_get_taildrop() - Get taildrop information
+ * @mc_io:	Pointer to MC portal's I/O object
+ * @cmd_flags:	Command flags; one or more of 'MC_CMD_FLAG_'
+ * @token:	Token of DPNI object
+ * @cg_point:	Congestion point
+ * @qtype:	Queue type on which the taildrop is configured.
+ *		Only Rx queues are supported for now
+ * @tc:		Traffic class to apply this taildrop to
+ * @index:	Index of the queue if the DPNI supports multiple queues for
+ *		traffic distribution. Ignored if CONGESTION_POINT is not 0.
+ * @taildrop:	Taildrop structure
+ *
+ * Return:	'0' on Success; Error code otherwise.
+ */
+int dpni_get_taildrop(struct fsl_mc_io *mc_io,
+		      u32 cmd_flags,
+		      u16 token,
+		      enum dpni_congestion_point cg_point,
+		      enum dpni_queue_type qtype,
+		      u8 tc,
+		      u8 index,
+		      struct dpni_taildrop *taildrop)
+{
+	struct fsl_mc_command cmd = { 0 };
+	struct dpni_cmd_get_taildrop *cmd_params;
+	struct dpni_rsp_get_taildrop *rsp_params;
+	int err;
+
+	/* prepare command */
+	cmd.header = mc_encode_cmd_header(DPNI_CMDID_GET_TAILDROP,
+					  cmd_flags,
+					  token);
+	cmd_params = (struct dpni_cmd_get_taildrop *)cmd.params;
+	cmd_params->congestion_point = cg_point;
+	cmd_params->qtype = qtype;
+	cmd_params->tc = tc;
+	cmd_params->index = index;
+
+	/* send command to mc */
+	err = mc_send_command(mc_io, &cmd);
+	if (err)
+		return err;
+
+	/* retrieve response parameters */
+	rsp_params = (struct dpni_rsp_get_taildrop *)cmd.params;
+	taildrop->enable = dpni_get_field(rsp_params->enable, ENABLE);
+	taildrop->units = rsp_params->units;
+	taildrop->threshold = le32_to_cpu(rsp_params->threshold);
+
+	return 0;
+}
+
+/**
+ * dpni_get_api_version() - Get Data Path Network Interface API version
+ * @mc_io:	Pointer to MC portal's I/O object
+ * @cmd_flags:	Command flags; one or more of 'MC_CMD_FLAG_'
+ * @major_ver:	Major version of data path network interface API
+ * @minor_ver:	Minor version of data path network interface API
+ *
+ * Return:	'0' on Success; Error code otherwise.
+ */
+int dpni_get_api_version(struct fsl_mc_io *mc_io,
+			 u32 cmd_flags,
+			 u16 *major_ver,
+			 u16 *minor_ver)
+{
+	struct dpni_rsp_get_api_version *rsp_params;
+	struct fsl_mc_command cmd = { 0 };
+	int err;
+
+	cmd.header = mc_encode_cmd_header(DPNI_CMDID_GET_API_VERSION,
+					  cmd_flags, 0);
+
+	err = mc_send_command(mc_io, &cmd);
+	if (err)
+		return err;
+
+	rsp_params = (struct dpni_rsp_get_api_version *)cmd.params;
+	*major_ver = le16_to_cpu(rsp_params->major);
+	*minor_ver = le16_to_cpu(rsp_params->minor);
+
+	return 0;
+}
+
+/**
+ * dpni_set_rx_fs_dist() - Set Rx flow steering distribution
+ * @mc_io:	Pointer to MC portal's I/O object
+ * @cmd_flags:	Command flags; one or more of 'MC_CMD_FLAG_'
+ * @token:	Token of DPNI object
+ * @cfg: Distribution configuration
+ *
+ * If the FS is already enabled with a previous call the classification
+ * key will be changed but all the table rules are kept. If the
+ * existing rules do not match the key the results will not be
+ * predictable. It is the user responsibility to keep key integrity.
+ * If cfg.enable is set to 1 the command will create a flow steering table
+ * and will classify packets according to this table. The packets that
+ * miss all the table rules will be classified according to settings
+ * made in dpni_set_rx_hash_dist()
+ * If cfg.enable is set to 0 the command will clear flow steering table.
+ * The packets will be classified according to settings made in
+ * dpni_set_rx_hash_dist()
+ *
+ * Return:	'0' on Success; Error code otherwise.
+ */
+int dpni_set_rx_fs_dist(struct fsl_mc_io *mc_io,
+			u32 cmd_flags,
+			u16 token,
+			const struct dpni_rx_dist_cfg *cfg)
+{
+	struct dpni_cmd_set_rx_fs_dist *cmd_params;
+	struct fsl_mc_command cmd = { 0 };
+
+	/* prepare command */
+	cmd.header = mc_encode_cmd_header(DPNI_CMDID_SET_RX_FS_DIST,
+					  cmd_flags,
+					  token);
+	cmd_params = (struct dpni_cmd_set_rx_fs_dist *)cmd.params;
+	cmd_params->dist_size = cpu_to_le16(cfg->dist_size);
+	dpni_set_field(cmd_params->enable, RX_FS_DIST_ENABLE, cfg->enable);
+	cmd_params->tc = cfg->tc;
+	cmd_params->miss_flow_id = cpu_to_le16(cfg->fs_miss_flow_id);
+	cmd_params->key_cfg_iova = cpu_to_le64(cfg->key_cfg_iova);
+
+	/* send command to mc*/
+	return mc_send_command(mc_io, &cmd);
+}
+
+/**
+ * dpni_set_rx_hash_dist() - Set Rx hash distribution
+ * @mc_io:	Pointer to MC portal's I/O object
+ * @cmd_flags:	Command flags; one or more of 'MC_CMD_FLAG_'
+ * @token:	Token of DPNI object
+ * @cfg: Distribution configuration
+ * If cfg.enable is set to 1 the packets will be classified using a hash
+ * function based on the key received in cfg.key_cfg_iova parameter.
+ * If cfg.enable is set to 0 the packets will be sent to the default queue
+ *
+ * Return:	'0' on Success; Error code otherwise.
+ */
+int dpni_set_rx_hash_dist(struct fsl_mc_io *mc_io,
+			  u32 cmd_flags,
+			  u16 token,
+			  const struct dpni_rx_dist_cfg *cfg)
+{
+	struct dpni_cmd_set_rx_hash_dist *cmd_params;
+	struct fsl_mc_command cmd = { 0 };
+
+	/* prepare command */
+	cmd.header = mc_encode_cmd_header(DPNI_CMDID_SET_RX_HASH_DIST,
+					  cmd_flags,
+					  token);
+	cmd_params = (struct dpni_cmd_set_rx_hash_dist *)cmd.params;
+	cmd_params->dist_size = cpu_to_le16(cfg->dist_size);
+	dpni_set_field(cmd_params->enable, RX_HASH_DIST_ENABLE, cfg->enable);
+	cmd_params->tc = cfg->tc;
+	cmd_params->key_cfg_iova = cpu_to_le64(cfg->key_cfg_iova);
+
+	/* send command to mc*/
+	return mc_send_command(mc_io, &cmd);
+}
+
+/**
+ * dpni_add_fs_entry() - Add Flow Steering entry for a specific traffic class
+ *			(to select a flow ID)
+ * @mc_io:	Pointer to MC portal's I/O object
+ * @cmd_flags:	Command flags; one or more of 'MC_CMD_FLAG_'
+ * @token:	Token of DPNI object
+ * @tc_id:	Traffic class selection (0-7)
+ * @index:	Location in the FS table where to insert the entry.
+ *		Only relevant if MASKING is enabled for FS
+ *		classification on this DPNI, it is ignored for exact match.
+ * @cfg:	Flow steering rule to add
+ * @action:	Action to be taken as result of a classification hit
+ *
+ * Return:	'0' on Success; Error code otherwise.
+ */
+int dpni_add_fs_entry(struct fsl_mc_io *mc_io,
+		      u32 cmd_flags,
+		      u16 token,
+		      u8 tc_id,
+		      u16 index,
+		      const struct dpni_rule_cfg *cfg,
+		      const struct dpni_fs_action_cfg *action)
+{
+	struct dpni_cmd_add_fs_entry *cmd_params;
+	struct fsl_mc_command cmd = { 0 };
+
+	/* prepare command */
+	cmd.header = mc_encode_cmd_header(DPNI_CMDID_ADD_FS_ENT,
+					  cmd_flags,
+					  token);
+	cmd_params = (struct dpni_cmd_add_fs_entry *)cmd.params;
+	cmd_params->tc_id = tc_id;
+	cmd_params->key_size = cfg->key_size;
+	cmd_params->index = cpu_to_le16(index);
+	cmd_params->key_iova = cpu_to_le64(cfg->key_iova);
+	cmd_params->mask_iova = cpu_to_le64(cfg->mask_iova);
+	cmd_params->options = cpu_to_le16(action->options);
+	cmd_params->flow_id = cpu_to_le16(action->flow_id);
+	cmd_params->flc = cpu_to_le64(action->flc);
+
+	/* send command to mc*/
+	return mc_send_command(mc_io, &cmd);
+}
+
+/**
+ * dpni_remove_fs_entry() - Remove Flow Steering entry from a specific
+ *			    traffic class
+ * @mc_io:	Pointer to MC portal's I/O object
+ * @cmd_flags:	Command flags; one or more of 'MC_CMD_FLAG_'
+ * @token:	Token of DPNI object
+ * @tc_id:	Traffic class selection (0-7)
+ * @cfg:	Flow steering rule to remove
+ *
+ * Return:	'0' on Success; Error code otherwise.
+ */
+int dpni_remove_fs_entry(struct fsl_mc_io *mc_io,
+			 u32 cmd_flags,
+			 u16 token,
+			 u8 tc_id,
+			 const struct dpni_rule_cfg *cfg)
+{
+	struct dpni_cmd_remove_fs_entry *cmd_params;
+	struct fsl_mc_command cmd = { 0 };
+
+	/* prepare command */
+	cmd.header = mc_encode_cmd_header(DPNI_CMDID_REMOVE_FS_ENT,
+					  cmd_flags,
+					  token);
+	cmd_params = (struct dpni_cmd_remove_fs_entry *)cmd.params;
+	cmd_params->tc_id = tc_id;
+	cmd_params->key_size = cfg->key_size;
+	cmd_params->key_iova = cpu_to_le64(cfg->key_iova);
+	cmd_params->mask_iova = cpu_to_le64(cfg->mask_iova);
+
+	/* send command to mc*/
+	return mc_send_command(mc_io, &cmd);
+}
+
+/**
+ * dpni_set_qos_table() - Set QoS mapping table
+ * @mc_io:	Pointer to MC portal's I/O object
+ * @cmd_flags:	Command flags; one or more of 'MC_CMD_FLAG_'
+ * @token:	Token of DPNI object
+ * @cfg:	QoS table configuration
+ *
+ * This function and all QoS-related functions require that
+ *'max_tcs > 1' was set at DPNI creation.
+ *
+ * warning: Before calling this function, call dpkg_prepare_key_cfg() to
+ *			prepare the key_cfg_iova parameter
+ *
+ * Return:	'0' on Success; Error code otherwise.
+ */
+int dpni_set_qos_table(struct fsl_mc_io *mc_io,
+		       u32 cmd_flags,
+		       u16 token,
+		       const struct dpni_qos_tbl_cfg *cfg)
+{
+	struct dpni_cmd_set_qos_table *cmd_params;
+	struct fsl_mc_command cmd = { 0 };
+
+	/* prepare command */
+	cmd.header = mc_encode_cmd_header(DPNI_CMDID_SET_QOS_TBL,
+					  cmd_flags,
+					  token);
+	cmd_params = (struct dpni_cmd_set_qos_table *)cmd.params;
+	cmd_params->default_tc = cfg->default_tc;
+	cmd_params->key_cfg_iova = cpu_to_le64(cfg->key_cfg_iova);
+	dpni_set_field(cmd_params->discard_on_miss, DISCARD_ON_MISS,
+		       cfg->discard_on_miss);
+
+	/* send command to mc*/
+	return mc_send_command(mc_io, &cmd);
+}
+
+/**
+ * dpni_add_qos_entry() - Add QoS mapping entry (to select a traffic class)
+ * @mc_io:	Pointer to MC portal's I/O object
+ * @cmd_flags:	Command flags; one or more of 'MC_CMD_FLAG_'
+ * @token:	Token of DPNI object
+ * @cfg:	QoS rule to add
+ * @tc_id:	Traffic class selection (0-7)
+ * @index:	Location in the QoS table where to insert the entry.
+ *		Only relevant if MASKING is enabled for QoS classification on
+ *		this DPNI, it is ignored for exact match.
+ *
+ * Return:	'0' on Success; Error code otherwise.
+ */
+int dpni_add_qos_entry(struct fsl_mc_io *mc_io,
+		       u32 cmd_flags,
+		       u16 token,
+		       const struct dpni_rule_cfg *cfg,
+		       u8 tc_id,
+		       u16 index)
+{
+	struct dpni_cmd_add_qos_entry *cmd_params;
+	struct fsl_mc_command cmd = { 0 };
+
+	/* prepare command */
+	cmd.header = mc_encode_cmd_header(DPNI_CMDID_ADD_QOS_ENT,
+					  cmd_flags,
+					  token);
+	cmd_params = (struct dpni_cmd_add_qos_entry *)cmd.params;
+	cmd_params->tc_id = tc_id;
+	cmd_params->key_size = cfg->key_size;
+	cmd_params->index = cpu_to_le16(index);
+	cmd_params->key_iova = cpu_to_le64(cfg->key_iova);
+	cmd_params->mask_iova = cpu_to_le64(cfg->mask_iova);
+
+	/* send command to mc*/
+	return mc_send_command(mc_io, &cmd);
+}
+
+/**
+ * dpni_remove_qos_entry() - Remove QoS mapping entry
+ * @mc_io:	Pointer to MC portal's I/O object
+ * @cmd_flags:	Command flags; one or more of 'MC_CMD_FLAG_'
+ * @token:	Token of DPNI object
+ * @cfg:	QoS rule to remove
+ *
+ * Return:	'0' on Success; Error code otherwise.
+ */
+int dpni_remove_qos_entry(struct fsl_mc_io *mc_io,
+			  u32 cmd_flags,
+			  u16 token,
+			  const struct dpni_rule_cfg *cfg)
+{
+	struct dpni_cmd_remove_qos_entry *cmd_params;
+	struct fsl_mc_command cmd = { 0 };
+
+	/* prepare command */
+	cmd.header = mc_encode_cmd_header(DPNI_CMDID_REMOVE_QOS_ENT,
+					  cmd_flags,
+					  token);
+	cmd_params = (struct dpni_cmd_remove_qos_entry *)cmd.params;
+	cmd_params->key_size = cfg->key_size;
+	cmd_params->key_iova = cpu_to_le64(cfg->key_iova);
+	cmd_params->mask_iova = cpu_to_le64(cfg->mask_iova);
+
+	/* send command to mc*/
+	return mc_send_command(mc_io, &cmd);
+}
+
+/**
+ * dpni_clear_qos_table() - Clear all QoS mapping entries
+ * @mc_io:	Pointer to MC portal's I/O object
+ * @cmd_flags:	Command flags; one or more of 'MC_CMD_FLAG_'
+ * @token:	Token of DPNI object
+ *
+ * Following this function call, all frames are directed to
+ * the default traffic class (0)
+ *
+ * Return:	'0' on Success; Error code otherwise.
+ */
+int dpni_clear_qos_table(struct fsl_mc_io *mc_io,
+			 u32 cmd_flags,
+			 u16 token)
+{
+	struct fsl_mc_command cmd = { 0 };
+
+	/* prepare command */
+	cmd.header = mc_encode_cmd_header(DPNI_CMDID_CLR_QOS_TBL,
+					  cmd_flags,
+					  token);
+
+	/* send command to mc*/
+	return mc_send_command(mc_io, &cmd);
+}
+
+/**
+ * dpni_set_tx_shaping() - Set the transmit shaping
+ * @mc_io:		Pointer to MC portal's I/O object
+ * @cmd_flags:		Command flags; one or more of 'MC_CMD_FLAG_'
+ * @token:		Token of DPNI object
+ * @tx_cr_shaper:	TX committed rate shaping configuration
+ * @tx_er_shaper:	TX excess rate shaping configuration
+ * @coupled:		Committed and excess rate shapers are coupled
+ *
+ * Return:	'0' on Success; Error code otherwise.
+ */
+int dpni_set_tx_shaping(struct fsl_mc_io *mc_io,
+			u32 cmd_flags,
+			u16 token,
+			const struct dpni_tx_shaping_cfg *tx_cr_shaper,
+			const struct dpni_tx_shaping_cfg *tx_er_shaper,
+			int coupled)
+{
+	struct dpni_cmd_set_tx_shaping *cmd_params;
+	struct fsl_mc_command cmd = { 0 };
+
+	/* prepare command */
+	cmd.header = mc_encode_cmd_header(DPNI_CMDID_SET_TX_SHAPING,
+					  cmd_flags,
+					  token);
+	cmd_params = (struct dpni_cmd_set_tx_shaping *)cmd.params;
+	cmd_params->tx_cr_max_burst_size = cpu_to_le16(tx_cr_shaper->max_burst_size);
+	cmd_params->tx_er_max_burst_size = cpu_to_le16(tx_er_shaper->max_burst_size);
+	cmd_params->tx_cr_rate_limit = cpu_to_le32(tx_cr_shaper->rate_limit);
+	cmd_params->tx_er_rate_limit = cpu_to_le32(tx_er_shaper->rate_limit);
+	dpni_set_field(cmd_params->coupled, COUPLED, coupled);
+
+	/* send command to mc*/
+	return mc_send_command(mc_io, &cmd);
+}
+
+/**
+ * dpni_get_single_step_cfg() - return current configuration for
+ *                              single step PTP
+ * @mc_io:	Pointer to MC portal's I/O object
+ * @cmd_flags:	Command flags; one or more of 'MC_CMD_FLAG_'
+ * @token:	Token of DPNI object
+ * @ptp_cfg:	ptp single step configuration
+ *
+ * Return:	'0' on Success; Error code otherwise.
+ *
+ */
+int dpni_get_single_step_cfg(struct fsl_mc_io *mc_io,
+			     u32 cmd_flags,
+			     u16 token,
+			     struct dpni_single_step_cfg *ptp_cfg)
+{
+	struct dpni_rsp_single_step_cfg *rsp_params;
+	struct fsl_mc_command cmd = { 0 };
+	int err;
+
+	/* prepare command */
+	cmd.header = mc_encode_cmd_header(DPNI_CMDID_GET_SINGLE_STEP_CFG,
+					  cmd_flags, token);
+	/* send command to mc*/
+	err =  mc_send_command(mc_io, &cmd);
+	if (err)
+		return err;
+
+	/* read command response */
+	rsp_params = (struct dpni_rsp_single_step_cfg *)cmd.params;
+	ptp_cfg->offset = le16_to_cpu(rsp_params->offset);
+	ptp_cfg->en = dpni_get_field(le16_to_cpu(rsp_params->flags),
+				     PTP_ENABLE) ? 1 : 0;
+	ptp_cfg->ch_update = dpni_get_field(le16_to_cpu(rsp_params->flags),
+					    PTP_CH_UPDATE) ? 1 : 0;
+	ptp_cfg->peer_delay = le32_to_cpu(rsp_params->peer_delay);
+	ptp_cfg->ptp_onestep_reg_base =
+				  le32_to_cpu(rsp_params->ptp_onestep_reg_base);
+
+	return err;
+}
+
+/**
+ * dpni_set_single_step_cfg() - enable/disable and configure single step PTP
+ * @mc_io:	Pointer to MC portal's I/O object
+ * @cmd_flags:	Command flags; one or more of 'MC_CMD_FLAG_'
+ * @token:	Token of DPNI object
+ * @ptp_cfg:	ptp single step configuration
+ *
+ * Return:	'0' on Success; Error code otherwise.
+ *
+ * The function has effect only when dpni object is connected to a dpmac
+ * object. If the dpni is not connected to a dpmac the configuration will
+ * be stored inside and applied when connection is made.
+ */
+int dpni_set_single_step_cfg(struct fsl_mc_io *mc_io,
+			     u32 cmd_flags,
+			     u16 token,
+			     struct dpni_single_step_cfg *ptp_cfg)
+{
+	struct dpni_cmd_single_step_cfg *cmd_params;
+	struct fsl_mc_command cmd = { 0 };
+	u16 flags;
+
+	/* prepare command */
+	cmd.header = mc_encode_cmd_header(DPNI_CMDID_SET_SINGLE_STEP_CFG,
+					  cmd_flags, token);
+	cmd_params = (struct dpni_cmd_single_step_cfg *)cmd.params;
+	cmd_params->offset = cpu_to_le16(ptp_cfg->offset);
+	cmd_params->peer_delay = cpu_to_le32(ptp_cfg->peer_delay);
+
+	flags = le16_to_cpu(cmd_params->flags);
+	dpni_set_field(flags, PTP_ENABLE, !!ptp_cfg->en);
+	dpni_set_field(flags, PTP_CH_UPDATE, !!ptp_cfg->ch_update);
+	cmd_params->flags = cpu_to_le16(flags);
+
+	/* send command to mc*/
+	return mc_send_command(mc_io, &cmd);
+}
diff --git a/drivers/net/ethernet/freescale/dpaa2/dpni.h b/drivers/net/ethernet/freescale/dpaa2/dpni.h
new file mode 100644
index 000000000000..5c0a1d5ac934
--- /dev/null
+++ b/drivers/net/ethernet/freescale/dpaa2/dpni.h
@@ -0,0 +1,1119 @@
+/* SPDX-License-Identifier: (GPL-2.0+ OR BSD-3-Clause) */
+/* Copyright 2013-2016 Freescale Semiconductor Inc.
+ * Copyright 2016 NXP
+ * Copyright 2020 NXP
+ */
+#ifndef __FSL_DPNI_H
+#define __FSL_DPNI_H
+
+#include "dpkg.h"
+
+struct fsl_mc_io;
+
+/* Data Path Network Interface API
+ * Contains initialization APIs and runtime control APIs for DPNI
+ */
+
+/** General DPNI macros */
+
+/**
+ * DPNI_MAX_TC - Maximum number of traffic classes
+ */
+#define DPNI_MAX_TC				8
+/**
+ * DPNI_MAX_DPBP - Maximum number of buffer pools per DPNI
+ */
+#define DPNI_MAX_DPBP				8
+
+/**
+ * DPNI_ALL_TCS - All traffic classes considered; see dpni_set_queue()
+ */
+#define DPNI_ALL_TCS				(u8)(-1)
+/**
+ * DPNI_ALL_TC_FLOWS - All flows within traffic class considered; see
+ * dpni_set_queue()
+ */
+#define DPNI_ALL_TC_FLOWS			(u16)(-1)
+/**
+ * DPNI_NEW_FLOW_ID - Generate new flow ID; see dpni_set_queue()
+ */
+#define DPNI_NEW_FLOW_ID			(u16)(-1)
+
+/**
+ * DPNI_OPT_TX_FRM_RELEASE - Tx traffic is always released to a buffer pool on
+ * transmit, there are no resources allocated to have the frames confirmed back
+ * to the source after transmission.
+ */
+#define DPNI_OPT_TX_FRM_RELEASE			0x000001
+/**
+ * DPNI_OPT_NO_MAC_FILTER - Disables support for MAC address filtering for
+ * addresses other than primary MAC address. This affects both unicast and
+ * multicast. Promiscuous mode can still be enabled/disabled for both unicast
+ * and multicast. If promiscuous mode is disabled, only traffic matching the
+ * primary MAC address will be accepted.
+ */
+#define DPNI_OPT_NO_MAC_FILTER			0x000002
+/**
+ * DPNI_OPT_HAS_POLICING - Allocate policers for this DPNI. They can be used to
+ * rate-limit traffic per traffic class (TC) basis.
+ */
+#define DPNI_OPT_HAS_POLICING			0x000004
+/**
+ * DPNI_OPT_SHARED_CONGESTION - Congestion can be managed in several ways,
+ * allowing the buffer pool to deplete on ingress, taildrop on each queue or
+ * use congestion groups for sets of queues. If set, it configures a single
+ * congestion groups across all TCs.  If reset, a congestion group is allocated
+ * for each TC. Only relevant if the DPNI has multiple traffic classes.
+ */
+#define DPNI_OPT_SHARED_CONGESTION		0x000008
+/**
+ * DPNI_OPT_HAS_KEY_MASKING - Enables TCAM for Flow Steering and QoS look-ups.
+ * If not specified, all look-ups are exact match. Note that TCAM is not
+ * available on LS1088 and its variants. Setting this bit on these SoCs will
+ * trigger an error.
+ */
+#define DPNI_OPT_HAS_KEY_MASKING		0x000010
+/**
+ * DPNI_OPT_NO_FS - Disables the flow steering table.
+ */
+#define DPNI_OPT_NO_FS				0x000020
+/**
+ * DPNI_OPT_SHARED_FS - Flow steering table is shared between all traffic
+ * classes
+ */
+#define DPNI_OPT_SHARED_FS			0x001000
+
+int dpni_open(struct fsl_mc_io	*mc_io,
+	      u32		cmd_flags,
+	      int		dpni_id,
+	      u16		*token);
+
+int dpni_close(struct fsl_mc_io	*mc_io,
+	       u32		cmd_flags,
+	       u16		token);
+
+#define DPNI_POOL_ASSOC_QPRI	0
+#define DPNI_POOL_ASSOC_QDBIN	1
+
+/**
+ * struct dpni_pools_cfg - Structure representing buffer pools configuration
+ * @num_dpbp: Number of DPBPs
+ * @pool_options: Buffer assignment options.
+ *	This field is a combination of DPNI_POOL_ASSOC_flags
+ * @pools: Array of buffer pools parameters; The number of valid entries
+ *	must match 'num_dpbp' value
+ * @pools.dpbp_id: DPBP object ID
+ * @pools.priority: Priority mask that indicates TC's used with this buffer.
+ *	If set to 0x00 MC will assume value 0xff.
+ * @pools.buffer_size: Buffer size
+ * @pools.backup_pool: Backup pool
+ */
+struct dpni_pools_cfg {
+	u8		num_dpbp;
+	u8		pool_options;
+	struct {
+		int	dpbp_id;
+		u8	priority_mask;
+		u16	buffer_size;
+		int	backup_pool;
+	} pools[DPNI_MAX_DPBP];
+};
+
+int dpni_set_pools(struct fsl_mc_io		*mc_io,
+		   u32				cmd_flags,
+		   u16				token,
+		   const struct dpni_pools_cfg	*cfg);
+
+int dpni_enable(struct fsl_mc_io	*mc_io,
+		u32			cmd_flags,
+		u16			token);
+
+int dpni_disable(struct fsl_mc_io	*mc_io,
+		 u32			cmd_flags,
+		 u16			token);
+
+int dpni_is_enabled(struct fsl_mc_io	*mc_io,
+		    u32			cmd_flags,
+		    u16			token,
+		    int			*en);
+
+int dpni_reset(struct fsl_mc_io	*mc_io,
+	       u32		cmd_flags,
+	       u16		token);
+
+/* DPNI IRQ Index and Events */
+
+#define DPNI_IRQ_INDEX				0
+
+/* DPNI_IRQ_EVENT_LINK_CHANGED - indicates a change in link state */
+#define DPNI_IRQ_EVENT_LINK_CHANGED		0x00000001
+
+/* DPNI_IRQ_EVENT_ENDPOINT_CHANGED - indicates a change in endpoint */
+#define DPNI_IRQ_EVENT_ENDPOINT_CHANGED		0x00000002
+
+int dpni_set_irq_enable(struct fsl_mc_io	*mc_io,
+			u32			cmd_flags,
+			u16			token,
+			u8			irq_index,
+			u8			en);
+
+int dpni_get_irq_enable(struct fsl_mc_io	*mc_io,
+			u32			cmd_flags,
+			u16			token,
+			u8			irq_index,
+			u8			*en);
+
+int dpni_set_irq_mask(struct fsl_mc_io	*mc_io,
+		      u32		cmd_flags,
+		      u16		token,
+		      u8		irq_index,
+		      u32		mask);
+
+int dpni_get_irq_mask(struct fsl_mc_io	*mc_io,
+		      u32		cmd_flags,
+		      u16		token,
+		      u8		irq_index,
+		      u32		*mask);
+
+int dpni_get_irq_status(struct fsl_mc_io	*mc_io,
+			u32			cmd_flags,
+			u16			token,
+			u8			irq_index,
+			u32			*status);
+
+int dpni_clear_irq_status(struct fsl_mc_io	*mc_io,
+			  u32			cmd_flags,
+			  u16			token,
+			  u8			irq_index,
+			  u32			status);
+
+/**
+ * struct dpni_attr - Structure representing DPNI attributes
+ * @options: Any combination of the following options:
+ *		DPNI_OPT_TX_FRM_RELEASE
+ *		DPNI_OPT_NO_MAC_FILTER
+ *		DPNI_OPT_HAS_POLICING
+ *		DPNI_OPT_SHARED_CONGESTION
+ *		DPNI_OPT_HAS_KEY_MASKING
+ *		DPNI_OPT_NO_FS
+ * @num_queues: Number of Tx and Rx queues used for traffic distribution.
+ * @num_tcs: Number of traffic classes (TCs), reserved for the DPNI.
+ * @mac_filter_entries: Number of entries in the MAC address filtering table.
+ * @vlan_filter_entries: Number of entries in the VLAN address filtering table.
+ * @qos_entries: Number of entries in the QoS classification table.
+ * @fs_entries: Number of entries in the flow steering table.
+ * @qos_key_size: Size, in bytes, of the QoS look-up key. Defining a key larger
+ *		than this when adding QoS entries will result in an error.
+ * @fs_key_size: Size, in bytes, of the flow steering look-up key. Defining a
+ *		key larger than this when composing the hash + FS key will
+ *		result in an error.
+ * @wriop_version: Version of WRIOP HW block. The 3 version values are stored
+ *		on 6, 5, 5 bits respectively.
+ */
+struct dpni_attr {
+	u32 options;
+	u8 num_queues;
+	u8 num_tcs;
+	u8 mac_filter_entries;
+	u8 vlan_filter_entries;
+	u8 qos_entries;
+	u16 fs_entries;
+	u8 qos_key_size;
+	u8 fs_key_size;
+	u16 wriop_version;
+};
+
+int dpni_get_attributes(struct fsl_mc_io	*mc_io,
+			u32			cmd_flags,
+			u16			token,
+			struct dpni_attr	*attr);
+
+/* DPNI errors */
+
+/**
+ * DPNI_ERROR_EOFHE - Extract out of frame header error
+ */
+#define DPNI_ERROR_EOFHE	0x00020000
+/**
+ * DPNI_ERROR_FLE - Frame length error
+ */
+#define DPNI_ERROR_FLE		0x00002000
+/**
+ * DPNI_ERROR_FPE - Frame physical error
+ */
+#define DPNI_ERROR_FPE		0x00001000
+/**
+ * DPNI_ERROR_PHE - Parsing header error
+ */
+#define DPNI_ERROR_PHE		0x00000020
+/**
+ * DPNI_ERROR_L3CE - Parser L3 checksum error
+ */
+#define DPNI_ERROR_L3CE		0x00000004
+/**
+ * DPNI_ERROR_L4CE - Parser L3 checksum error
+ */
+#define DPNI_ERROR_L4CE		0x00000001
+
+/**
+ * enum dpni_error_action - Defines DPNI behavior for errors
+ * @DPNI_ERROR_ACTION_DISCARD: Discard the frame
+ * @DPNI_ERROR_ACTION_CONTINUE: Continue with the normal flow
+ * @DPNI_ERROR_ACTION_SEND_TO_ERROR_QUEUE: Send the frame to the error queue
+ */
+enum dpni_error_action {
+	DPNI_ERROR_ACTION_DISCARD = 0,
+	DPNI_ERROR_ACTION_CONTINUE = 1,
+	DPNI_ERROR_ACTION_SEND_TO_ERROR_QUEUE = 2
+};
+
+/**
+ * struct dpni_error_cfg - Structure representing DPNI errors treatment
+ * @errors: Errors mask; use 'DPNI_ERROR__<X>
+ * @error_action: The desired action for the errors mask
+ * @set_frame_annotation: Set to '1' to mark the errors in frame annotation
+ *		status (FAS); relevant only for the non-discard action
+ */
+struct dpni_error_cfg {
+	u32			errors;
+	enum dpni_error_action	error_action;
+	int			set_frame_annotation;
+};
+
+int dpni_set_errors_behavior(struct fsl_mc_io		*mc_io,
+			     u32			cmd_flags,
+			     u16			token,
+			     struct dpni_error_cfg	*cfg);
+
+/* DPNI buffer layout modification options */
+
+/**
+ * DPNI_BUF_LAYOUT_OPT_TIMESTAMP - Select to modify the time-stamp setting
+ */
+#define DPNI_BUF_LAYOUT_OPT_TIMESTAMP		0x00000001
+/**
+ * DPNI_BUF_LAYOUT_OPT_PARSER_RESULT - Select to modify the parser-result
+ * setting; not applicable for Tx
+ */
+#define DPNI_BUF_LAYOUT_OPT_PARSER_RESULT	0x00000002
+/**
+ * DPNI_BUF_LAYOUT_OPT_FRAME_STATUS - Select to modify the frame-status setting
+ */
+#define DPNI_BUF_LAYOUT_OPT_FRAME_STATUS	0x00000004
+/**
+ * DPNI_BUF_LAYOUT_OPT_PRIVATE_DATA_SIZE - Select to modify the private-data-size setting
+ */
+#define DPNI_BUF_LAYOUT_OPT_PRIVATE_DATA_SIZE	0x00000008
+/**
+ * DPNI_BUF_LAYOUT_OPT_DATA_ALIGN - Select to modify the data-alignment setting
+ */
+#define DPNI_BUF_LAYOUT_OPT_DATA_ALIGN		0x00000010
+/**
+ * DPNI_BUF_LAYOUT_OPT_DATA_HEAD_ROOM - Select to modify the data-head-room setting
+ */
+#define DPNI_BUF_LAYOUT_OPT_DATA_HEAD_ROOM	0x00000020
+/**
+ * DPNI_BUF_LAYOUT_OPT_DATA_TAIL_ROOM - Select to modify the data-tail-room setting
+ */
+#define DPNI_BUF_LAYOUT_OPT_DATA_TAIL_ROOM	0x00000040
+
+/**
+ * struct dpni_buffer_layout - Structure representing DPNI buffer layout
+ * @options: Flags representing the suggested modifications to the buffer
+ *		layout; Use any combination of 'DPNI_BUF_LAYOUT_OPT_<X>' flags
+ * @pass_timestamp: Pass timestamp value
+ * @pass_parser_result: Pass parser results
+ * @pass_frame_status: Pass frame status
+ * @private_data_size: Size kept for private data (in bytes)
+ * @data_align: Data alignment
+ * @data_head_room: Data head room
+ * @data_tail_room: Data tail room
+ */
+struct dpni_buffer_layout {
+	u32	options;
+	int	pass_timestamp;
+	int	pass_parser_result;
+	int	pass_frame_status;
+	u16	private_data_size;
+	u16	data_align;
+	u16	data_head_room;
+	u16	data_tail_room;
+};
+
+/**
+ * enum dpni_queue_type - Identifies a type of queue targeted by the command
+ * @DPNI_QUEUE_RX: Rx queue
+ * @DPNI_QUEUE_TX: Tx queue
+ * @DPNI_QUEUE_TX_CONFIRM: Tx confirmation queue
+ * @DPNI_QUEUE_RX_ERR: Rx error queue
+ */
+enum dpni_queue_type {
+	DPNI_QUEUE_RX,
+	DPNI_QUEUE_TX,
+	DPNI_QUEUE_TX_CONFIRM,
+	DPNI_QUEUE_RX_ERR,
+};
+
+int dpni_get_buffer_layout(struct fsl_mc_io		*mc_io,
+			   u32				cmd_flags,
+			   u16				token,
+			   enum dpni_queue_type		qtype,
+			   struct dpni_buffer_layout	*layout);
+
+int dpni_set_buffer_layout(struct fsl_mc_io		   *mc_io,
+			   u32				   cmd_flags,
+			   u16				   token,
+			   enum dpni_queue_type		   qtype,
+			   const struct dpni_buffer_layout *layout);
+
+/**
+ * enum dpni_offload - Identifies a type of offload targeted by the command
+ * @DPNI_OFF_RX_L3_CSUM: Rx L3 checksum validation
+ * @DPNI_OFF_RX_L4_CSUM: Rx L4 checksum validation
+ * @DPNI_OFF_TX_L3_CSUM: Tx L3 checksum generation
+ * @DPNI_OFF_TX_L4_CSUM: Tx L4 checksum generation
+ */
+enum dpni_offload {
+	DPNI_OFF_RX_L3_CSUM,
+	DPNI_OFF_RX_L4_CSUM,
+	DPNI_OFF_TX_L3_CSUM,
+	DPNI_OFF_TX_L4_CSUM,
+};
+
+int dpni_set_offload(struct fsl_mc_io	*mc_io,
+		     u32		cmd_flags,
+		     u16		token,
+		     enum dpni_offload	type,
+		     u32		config);
+
+int dpni_get_offload(struct fsl_mc_io	*mc_io,
+		     u32		cmd_flags,
+		     u16		token,
+		     enum dpni_offload	type,
+		     u32		*config);
+
+int dpni_get_qdid(struct fsl_mc_io	*mc_io,
+		  u32			cmd_flags,
+		  u16			token,
+		  enum dpni_queue_type	qtype,
+		  u16			*qdid);
+
+int dpni_get_tx_data_offset(struct fsl_mc_io	*mc_io,
+			    u32			cmd_flags,
+			    u16			token,
+			    u16			*data_offset);
+
+#define DPNI_STATISTICS_CNT		7
+
+/**
+ * union dpni_statistics - Union describing the DPNI statistics
+ * @page_0: Page_0 statistics structure
+ * @page_0.ingress_all_frames: Ingress frame count
+ * @page_0.ingress_all_bytes: Ingress byte count
+ * @page_0.ingress_multicast_frames: Ingress multicast frame count
+ * @page_0.ingress_multicast_bytes: Ingress multicast byte count
+ * @page_0.ingress_broadcast_frames: Ingress broadcast frame count
+ * @page_0.ingress_broadcast_bytes: Ingress broadcast byte count
+ * @page_1: Page_1 statistics structure
+ * @page_1.egress_all_frames: Egress frame count
+ * @page_1.egress_all_bytes: Egress byte count
+ * @page_1.egress_multicast_frames: Egress multicast frame count
+ * @page_1.egress_multicast_bytes: Egress multicast byte count
+ * @page_1.egress_broadcast_frames: Egress broadcast frame count
+ * @page_1.egress_broadcast_bytes: Egress broadcast byte count
+ * @page_2: Page_2 statistics structure
+ * @page_2.ingress_filtered_frames: Ingress filtered frame count
+ * @page_2.ingress_discarded_frames: Ingress discarded frame count
+ * @page_2.ingress_nobuffer_discards: Ingress discarded frame count due to
+ *	lack of buffers
+ * @page_2.egress_discarded_frames: Egress discarded frame count
+ * @page_2.egress_confirmed_frames: Egress confirmed frame count
+ * @page_3: Page_3 statistics structure
+ * @page_3.egress_dequeue_bytes: Cumulative count of the number of bytes
+ *	dequeued from egress FQs
+ * @page_3.egress_dequeue_frames: Cumulative count of the number of frames
+ *	dequeued from egress FQs
+ * @page_3.egress_reject_bytes: Cumulative count of the number of bytes in
+ *	egress frames whose enqueue was rejected
+ * @page_3.egress_reject_frames: Cumulative count of the number of egress
+ *	frames whose enqueue was rejected
+ * @page_4: Page_4 statistics structure: congestion points
+ * @page_4.cgr_reject_frames: number of rejected frames due to congestion point
+ * @page_4.cgr_reject_bytes: number of rejected bytes due to congestion point
+ * @page_5: Page_5 statistics structure: policer
+ * @page_5.policer_cnt_red: NUmber of red colored frames
+ * @page_5.policer_cnt_yellow: number of yellow colored frames
+ * @page_5.policer_cnt_green: number of green colored frames
+ * @page_5.policer_cnt_re_red: number of recolored red frames
+ * @page_5.policer_cnt_re_yellow: number of recolored yellow frames
+ * @page_6: Page_6 statistics structure
+ * @page_6.tx_pending_frames: total number of frames pending in egress FQs
+ * @raw: raw statistics structure, used to index counters
+ */
+union dpni_statistics {
+	struct {
+		u64 ingress_all_frames;
+		u64 ingress_all_bytes;
+		u64 ingress_multicast_frames;
+		u64 ingress_multicast_bytes;
+		u64 ingress_broadcast_frames;
+		u64 ingress_broadcast_bytes;
+	} page_0;
+	struct {
+		u64 egress_all_frames;
+		u64 egress_all_bytes;
+		u64 egress_multicast_frames;
+		u64 egress_multicast_bytes;
+		u64 egress_broadcast_frames;
+		u64 egress_broadcast_bytes;
+	} page_1;
+	struct {
+		u64 ingress_filtered_frames;
+		u64 ingress_discarded_frames;
+		u64 ingress_nobuffer_discards;
+		u64 egress_discarded_frames;
+		u64 egress_confirmed_frames;
+	} page_2;
+	struct {
+		u64 egress_dequeue_bytes;
+		u64 egress_dequeue_frames;
+		u64 egress_reject_bytes;
+		u64 egress_reject_frames;
+	} page_3;
+	struct {
+		u64 cgr_reject_frames;
+		u64 cgr_reject_bytes;
+	} page_4;
+	struct {
+		u64 policer_cnt_red;
+		u64 policer_cnt_yellow;
+		u64 policer_cnt_green;
+		u64 policer_cnt_re_red;
+		u64 policer_cnt_re_yellow;
+	} page_5;
+	struct {
+		u64 tx_pending_frames;
+	} page_6;
+	struct {
+		u64 counter[DPNI_STATISTICS_CNT];
+	} raw;
+};
+
+int dpni_get_statistics(struct fsl_mc_io	*mc_io,
+			u32			cmd_flags,
+			u16			token,
+			u8			page,
+			union dpni_statistics	*stat);
+
+#define DPNI_LINK_OPT_AUTONEG		0x0000000000000001ULL
+#define DPNI_LINK_OPT_HALF_DUPLEX	0x0000000000000002ULL
+#define DPNI_LINK_OPT_PAUSE		0x0000000000000004ULL
+#define DPNI_LINK_OPT_ASYM_PAUSE	0x0000000000000008ULL
+#define DPNI_LINK_OPT_PFC_PAUSE		0x0000000000000010ULL
+
+/**
+ * struct dpni_link_cfg - Structure representing DPNI link configuration
+ * @rate: Rate
+ * @options: Mask of available options; use 'DPNI_LINK_OPT_<X>' values
+ */
+struct dpni_link_cfg {
+	u32 rate;
+	u64 options;
+};
+
+int dpni_set_link_cfg(struct fsl_mc_io			*mc_io,
+		      u32				cmd_flags,
+		      u16				token,
+		      const struct dpni_link_cfg	*cfg);
+
+int dpni_get_link_cfg(struct fsl_mc_io			*mc_io,
+		      u32				cmd_flags,
+		      u16				token,
+		      struct dpni_link_cfg		*cfg);
+
+/**
+ * struct dpni_link_state - Structure representing DPNI link state
+ * @rate: Rate
+ * @options: Mask of available options; use 'DPNI_LINK_OPT_<X>' values
+ * @up: Link state; '0' for down, '1' for up
+ */
+struct dpni_link_state {
+	u32	rate;
+	u64	options;
+	int	up;
+};
+
+int dpni_get_link_state(struct fsl_mc_io	*mc_io,
+			u32			cmd_flags,
+			u16			token,
+			struct dpni_link_state	*state);
+
+int dpni_set_max_frame_length(struct fsl_mc_io	*mc_io,
+			      u32		cmd_flags,
+			      u16		token,
+			      u16		max_frame_length);
+
+int dpni_get_max_frame_length(struct fsl_mc_io	*mc_io,
+			      u32		cmd_flags,
+			      u16		token,
+			      u16		*max_frame_length);
+
+int dpni_set_multicast_promisc(struct fsl_mc_io *mc_io,
+			       u32		cmd_flags,
+			       u16		token,
+			       int		en);
+
+int dpni_get_multicast_promisc(struct fsl_mc_io *mc_io,
+			       u32		cmd_flags,
+			       u16		token,
+			       int		*en);
+
+int dpni_set_unicast_promisc(struct fsl_mc_io	*mc_io,
+			     u32		cmd_flags,
+			     u16		token,
+			     int		en);
+
+int dpni_get_unicast_promisc(struct fsl_mc_io	*mc_io,
+			     u32		cmd_flags,
+			     u16		token,
+			     int		*en);
+
+int dpni_set_primary_mac_addr(struct fsl_mc_io *mc_io,
+			      u32		cmd_flags,
+			      u16		token,
+			      const u8		mac_addr[6]);
+
+int dpni_get_primary_mac_addr(struct fsl_mc_io	*mc_io,
+			      u32		cmd_flags,
+			      u16		token,
+			      u8		mac_addr[6]);
+
+int dpni_get_port_mac_addr(struct fsl_mc_io	*mc_io,
+			   u32			cm_flags,
+			   u16			token,
+			   u8			mac_addr[6]);
+
+int dpni_add_mac_addr(struct fsl_mc_io	*mc_io,
+		      u32		cmd_flags,
+		      u16		token,
+		      const u8		mac_addr[6]);
+
+int dpni_remove_mac_addr(struct fsl_mc_io	*mc_io,
+			 u32			cmd_flags,
+			 u16			token,
+			 const u8		mac_addr[6]);
+
+int dpni_clear_mac_filters(struct fsl_mc_io	*mc_io,
+			   u32			cmd_flags,
+			   u16			token,
+			   int			unicast,
+			   int			multicast);
+
+/**
+ * enum dpni_dist_mode - DPNI distribution mode
+ * @DPNI_DIST_MODE_NONE: No distribution
+ * @DPNI_DIST_MODE_HASH: Use hash distribution; only relevant if
+ *		the 'DPNI_OPT_DIST_HASH' option was set at DPNI creation
+ * @DPNI_DIST_MODE_FS:  Use explicit flow steering; only relevant if
+ *	 the 'DPNI_OPT_DIST_FS' option was set at DPNI creation
+ */
+enum dpni_dist_mode {
+	DPNI_DIST_MODE_NONE = 0,
+	DPNI_DIST_MODE_HASH = 1,
+	DPNI_DIST_MODE_FS = 2
+};
+
+/**
+ * enum dpni_fs_miss_action -   DPNI Flow Steering miss action
+ * @DPNI_FS_MISS_DROP: In case of no-match, drop the frame
+ * @DPNI_FS_MISS_EXPLICIT_FLOWID: In case of no-match, use explicit flow-id
+ * @DPNI_FS_MISS_HASH: In case of no-match, distribute using hash
+ */
+enum dpni_fs_miss_action {
+	DPNI_FS_MISS_DROP = 0,
+	DPNI_FS_MISS_EXPLICIT_FLOWID = 1,
+	DPNI_FS_MISS_HASH = 2
+};
+
+/**
+ * struct dpni_fs_tbl_cfg - Flow Steering table configuration
+ * @miss_action: Miss action selection
+ * @default_flow_id: Used when 'miss_action = DPNI_FS_MISS_EXPLICIT_FLOWID'
+ */
+struct dpni_fs_tbl_cfg {
+	enum dpni_fs_miss_action	miss_action;
+	u16				default_flow_id;
+};
+
+int dpni_prepare_key_cfg(const struct dpkg_profile_cfg *cfg,
+			 u8 *key_cfg_buf);
+
+/**
+ * struct dpni_rx_tc_dist_cfg - Rx traffic class distribution configuration
+ * @dist_size: Set the distribution size;
+ *	supported values: 1,2,3,4,6,7,8,12,14,16,24,28,32,48,56,64,96,
+ *	112,128,192,224,256,384,448,512,768,896,1024
+ * @dist_mode: Distribution mode
+ * @key_cfg_iova: I/O virtual address of 256 bytes DMA-able memory filled with
+ *		the extractions to be used for the distribution key by calling
+ *		dpni_prepare_key_cfg() relevant only when
+ *		'dist_mode != DPNI_DIST_MODE_NONE', otherwise it can be '0'
+ * @fs_cfg: Flow Steering table configuration; only relevant if
+ *		'dist_mode = DPNI_DIST_MODE_FS'
+ */
+struct dpni_rx_tc_dist_cfg {
+	u16			dist_size;
+	enum dpni_dist_mode	dist_mode;
+	u64			key_cfg_iova;
+	struct dpni_fs_tbl_cfg	fs_cfg;
+};
+
+int dpni_set_rx_tc_dist(struct fsl_mc_io			*mc_io,
+			u32					cmd_flags,
+			u16					token,
+			u8					tc_id,
+			const struct dpni_rx_tc_dist_cfg	*cfg);
+
+/**
+ * DPNI_FS_MISS_DROP - When used for fs_miss_flow_id in function
+ * dpni_set_rx_dist, will signal to dpni to drop all unclassified frames
+ */
+#define DPNI_FS_MISS_DROP		((uint16_t)-1)
+
+/**
+ * struct dpni_rx_dist_cfg - Rx distribution configuration
+ * @dist_size:	distribution size
+ * @key_cfg_iova: I/O virtual address of 256 bytes DMA-able memory filled with
+ *		the extractions to be used for the distribution key by calling
+ *		dpni_prepare_key_cfg(); relevant only when enable!=0 otherwise
+ *		it can be '0'
+ * @enable: enable/disable the distribution.
+ * @tc: TC id for which distribution is set
+ * @fs_miss_flow_id: when packet misses all rules from flow steering table and
+ *		hash is disabled it will be put into this queue id; use
+ *		DPNI_FS_MISS_DROP to drop frames. The value of this field is
+ *		used only when flow steering distribution is enabled and hash
+ *		distribution is disabled
+ */
+struct dpni_rx_dist_cfg {
+	u16 dist_size;
+	u64 key_cfg_iova;
+	u8 enable;
+	u8 tc;
+	u16 fs_miss_flow_id;
+};
+
+int dpni_set_rx_fs_dist(struct fsl_mc_io *mc_io,
+			u32 cmd_flags,
+			u16 token,
+			const struct dpni_rx_dist_cfg *cfg);
+
+int dpni_set_rx_hash_dist(struct fsl_mc_io *mc_io,
+			  u32 cmd_flags,
+			  u16 token,
+			  const struct dpni_rx_dist_cfg *cfg);
+
+/**
+ * struct dpni_qos_tbl_cfg - Structure representing QOS table configuration
+ * @key_cfg_iova: I/O virtual address of 256 bytes DMA-able memory filled with
+ *		key extractions to be used as the QoS criteria by calling
+ *		dpkg_prepare_key_cfg()
+ * @discard_on_miss: Set to '1' to discard frames in case of no match (miss);
+ *		'0' to use the 'default_tc' in such cases
+ * @default_tc: Used in case of no-match and 'discard_on_miss'= 0
+ */
+struct dpni_qos_tbl_cfg {
+	u64 key_cfg_iova;
+	int discard_on_miss;
+	u8 default_tc;
+};
+
+int dpni_set_qos_table(struct fsl_mc_io *mc_io,
+		       u32 cmd_flags,
+		       u16 token,
+		       const struct dpni_qos_tbl_cfg *cfg);
+
+/**
+ * enum dpni_dest - DPNI destination types
+ * @DPNI_DEST_NONE: Unassigned destination; The queue is set in parked mode and
+ *		does not generate FQDAN notifications; user is expected to
+ *		dequeue from the queue based on polling or other user-defined
+ *		method
+ * @DPNI_DEST_DPIO: The queue is set in schedule mode and generates FQDAN
+ *		notifications to the specified DPIO; user is expected to dequeue
+ *		from the queue only after notification is received
+ * @DPNI_DEST_DPCON: The queue is set in schedule mode and does not generate
+ *		FQDAN notifications, but is connected to the specified DPCON
+ *		object; user is expected to dequeue from the DPCON channel
+ */
+enum dpni_dest {
+	DPNI_DEST_NONE = 0,
+	DPNI_DEST_DPIO = 1,
+	DPNI_DEST_DPCON = 2
+};
+
+/**
+ * struct dpni_queue - Queue structure
+ * @destination: - Destination structure
+ * @destination.id: ID of the destination, only relevant if DEST_TYPE is > 0.
+ *	Identifies either a DPIO or a DPCON object.
+ *	Not relevant for Tx queues.
+ * @destination.type:	May be one of the following:
+ *	0 - No destination, queue can be manually
+ *		queried, but will not push traffic or
+ *		notifications to a DPIO;
+ *	1 - The destination is a DPIO. When traffic
+ *		becomes available in the queue a FQDAN
+ *		(FQ data available notification) will be
+ *		generated to selected DPIO;
+ *	2 - The destination is a DPCON. The queue is
+ *		associated with a DPCON object for the
+ *		purpose of scheduling between multiple
+ *		queues. The DPCON may be independently
+ *		configured to generate notifications.
+ *		Not relevant for Tx queues.
+ * @destination.hold_active: Hold active, maintains a queue scheduled for longer
+ *	in a DPIO during dequeue to reduce spread of traffic.
+ *	Only relevant if queues are
+ *	not affined to a single DPIO.
+ * @user_context: User data, presented to the user along with any frames
+ *	from this queue. Not relevant for Tx queues.
+ * @flc: FD FLow Context structure
+ * @flc.value: Default FLC value for traffic dequeued from
+ *      this queue.  Please check description of FD
+ *      structure for more information.
+ *      Note that FLC values set using dpni_add_fs_entry,
+ *      if any, take precedence over values per queue.
+ * @flc.stash_control: Boolean, indicates whether the 6 lowest
+ *      - significant bits are used for stash control.
+ *      significant bits are used for stash control.  If set, the 6
+ *      least significant bits in value are interpreted as follows:
+ *      - bits 0-1: indicates the number of 64 byte units of context
+ *      that are stashed.  FLC value is interpreted as a memory address
+ *      in this case, excluding the 6 LS bits.
+ *      - bits 2-3: indicates the number of 64 byte units of frame
+ *      annotation to be stashed.  Annotation is placed at FD[ADDR].
+ *      - bits 4-5: indicates the number of 64 byte units of frame
+ *      data to be stashed.  Frame data is placed at FD[ADDR] +
+ *      FD[OFFSET].
+ *      For more details check the Frame Descriptor section in the
+ *      hardware documentation.
+ */
+struct dpni_queue {
+	struct {
+		u16 id;
+		enum dpni_dest type;
+		char hold_active;
+		u8 priority;
+	} destination;
+	u64 user_context;
+	struct {
+		u64 value;
+		char stash_control;
+	} flc;
+};
+
+/**
+ * struct dpni_queue_id - Queue identification, used for enqueue commands
+ *			or queue control
+ * @fqid: FQID used for enqueueing to and/or configuration of this specific FQ
+ * @qdbin: Queueing bin, used to enqueue using QDID, DQBIN, QPRI. Only relevant
+ *		for Tx queues.
+ */
+struct dpni_queue_id {
+	u32 fqid;
+	u16 qdbin;
+};
+
+/* Set User Context */
+#define DPNI_QUEUE_OPT_USER_CTX		0x00000001
+#define DPNI_QUEUE_OPT_DEST		0x00000002
+#define DPNI_QUEUE_OPT_FLC		0x00000004
+#define DPNI_QUEUE_OPT_HOLD_ACTIVE	0x00000008
+
+int dpni_set_queue(struct fsl_mc_io	*mc_io,
+		   u32			cmd_flags,
+		   u16			token,
+		   enum dpni_queue_type	qtype,
+		   u8			tc,
+		   u8			index,
+		   u8			options,
+		   const struct dpni_queue *queue);
+
+int dpni_get_queue(struct fsl_mc_io	*mc_io,
+		   u32			cmd_flags,
+		   u16			token,
+		   enum dpni_queue_type	qtype,
+		   u8			tc,
+		   u8			index,
+		   struct dpni_queue	*queue,
+		   struct dpni_queue_id	*qid);
+
+/**
+ * enum dpni_congestion_unit - DPNI congestion units
+ * @DPNI_CONGESTION_UNIT_BYTES: bytes units
+ * @DPNI_CONGESTION_UNIT_FRAMES: frames units
+ */
+enum dpni_congestion_unit {
+	DPNI_CONGESTION_UNIT_BYTES = 0,
+	DPNI_CONGESTION_UNIT_FRAMES
+};
+
+/**
+ * enum dpni_congestion_point - Structure representing congestion point
+ * @DPNI_CP_QUEUE: Set taildrop per queue, identified by QUEUE_TYPE, TC and
+ *		QUEUE_INDEX
+ * @DPNI_CP_GROUP: Set taildrop per queue group. Depending on options used to
+ *		define the DPNI this can be either per TC (default) or per
+ *		interface (DPNI_OPT_SHARED_CONGESTION set at DPNI create).
+ *		QUEUE_INDEX is ignored if this type is used.
+ */
+enum dpni_congestion_point {
+	DPNI_CP_QUEUE,
+	DPNI_CP_GROUP,
+};
+
+/**
+ * struct dpni_dest_cfg - Structure representing DPNI destination parameters
+ * @dest_type:	Destination type
+ * @dest_id:	Either DPIO ID or DPCON ID, depending on the destination type
+ * @priority:	Priority selection within the DPIO or DPCON channel; valid
+ *		values are 0-1 or 0-7, depending on the number of priorities
+ *		in that channel; not relevant for 'DPNI_DEST_NONE' option
+ */
+struct dpni_dest_cfg {
+	enum dpni_dest dest_type;
+	int dest_id;
+	u8 priority;
+};
+
+/* DPNI congestion options */
+
+/**
+ * DPNI_CONG_OPT_FLOW_CONTROL - This congestion will trigger flow control or
+ * priority flow control.  This will have effect only if flow control is
+ * enabled with dpni_set_link_cfg().
+ */
+#define DPNI_CONG_OPT_FLOW_CONTROL		0x00000040
+
+/**
+ * struct dpni_congestion_notification_cfg - congestion notification
+ *					configuration
+ * @units: Units type
+ * @threshold_entry: Above this threshold we enter a congestion state.
+ *		set it to '0' to disable it
+ * @threshold_exit: Below this threshold we exit the congestion state.
+ * @message_ctx: The context that will be part of the CSCN message
+ * @message_iova: I/O virtual address (must be in DMA-able memory),
+ *		must be 16B aligned; valid only if 'DPNI_CONG_OPT_WRITE_MEM_<X>'
+ *		is contained in 'options'
+ * @dest_cfg: CSCN can be send to either DPIO or DPCON WQ channel
+ * @notification_mode: Mask of available options; use 'DPNI_CONG_OPT_<X>' values
+ */
+
+struct dpni_congestion_notification_cfg {
+	enum dpni_congestion_unit units;
+	u32 threshold_entry;
+	u32 threshold_exit;
+	u64 message_ctx;
+	u64 message_iova;
+	struct dpni_dest_cfg dest_cfg;
+	u16 notification_mode;
+};
+
+int dpni_set_congestion_notification(
+			struct fsl_mc_io *mc_io,
+			u32 cmd_flags,
+			u16 token,
+			enum dpni_queue_type qtype,
+			u8 tc_id,
+			const struct dpni_congestion_notification_cfg *cfg);
+
+/**
+ * struct dpni_taildrop - Structure representing the taildrop
+ * @enable:	Indicates whether the taildrop is active or not.
+ * @units:	Indicates the unit of THRESHOLD. Queue taildrop only supports
+ *		byte units, this field is ignored and assumed = 0 if
+ *		CONGESTION_POINT is 0.
+ * @threshold:	Threshold value, in units identified by UNITS field. Value 0
+ *		cannot be used as a valid taildrop threshold, THRESHOLD must
+ *		be > 0 if the taildrop is enabled.
+ */
+struct dpni_taildrop {
+	char enable;
+	enum dpni_congestion_unit units;
+	u32 threshold;
+};
+
+int dpni_set_taildrop(struct fsl_mc_io *mc_io,
+		      u32 cmd_flags,
+		      u16 token,
+		      enum dpni_congestion_point cg_point,
+		      enum dpni_queue_type q_type,
+		      u8 tc,
+		      u8 q_index,
+		      struct dpni_taildrop *taildrop);
+
+int dpni_get_taildrop(struct fsl_mc_io *mc_io,
+		      u32 cmd_flags,
+		      u16 token,
+		      enum dpni_congestion_point cg_point,
+		      enum dpni_queue_type q_type,
+		      u8 tc,
+		      u8 q_index,
+		      struct dpni_taildrop *taildrop);
+
+/**
+ * struct dpni_rule_cfg - Rule configuration for table lookup
+ * @key_iova: I/O virtual address of the key (must be in DMA-able memory)
+ * @mask_iova: I/O virtual address of the mask (must be in DMA-able memory)
+ * @key_size: key and mask size (in bytes)
+ */
+struct dpni_rule_cfg {
+	u64	key_iova;
+	u64	mask_iova;
+	u8	key_size;
+};
+
+/**
+ * DPNI_FS_OPT_DISCARD - Discard matching traffic. If set, this takes
+ * precedence over any other configuration and matching traffic is always
+ * discarded.
+ */
+ #define DPNI_FS_OPT_DISCARD            0x1
+
+/**
+ * DPNI_FS_OPT_SET_FLC - Set FLC value. If set, flc member of struct
+ * dpni_fs_action_cfg is used to override the FLC value set per queue.
+ * For more details check the Frame Descriptor section in the hardware
+ * documentation.
+ */
+#define DPNI_FS_OPT_SET_FLC            0x2
+
+/**
+ * DPNI_FS_OPT_SET_STASH_CONTROL - Indicates whether the 6 lowest significant
+ * bits of FLC are used for stash control. If set, the 6 least significant bits
+ * in value are interpreted as follows:
+ *     - bits 0-1: indicates the number of 64 byte units of context that are
+ *     stashed. FLC value is interpreted as a memory address in this case,
+ *     excluding the 6 LS bits.
+ *     - bits 2-3: indicates the number of 64 byte units of frame annotation
+ *     to be stashed. Annotation is placed at FD[ADDR].
+ *     - bits 4-5: indicates the number of 64 byte units of frame data to be
+ *     stashed. Frame data is placed at FD[ADDR] + FD[OFFSET].
+ * This flag is ignored if DPNI_FS_OPT_SET_FLC is not specified.
+ */
+#define DPNI_FS_OPT_SET_STASH_CONTROL  0x4
+
+/**
+ * struct dpni_fs_action_cfg - Action configuration for table look-up
+ * @flc:	FLC value for traffic matching this rule. Please check the
+ *		Frame Descriptor section in the hardware documentation for
+ *		more information.
+ * @flow_id:	Identifies the Rx queue used for matching traffic. Supported
+ *		values are in range 0 to num_queue-1.
+ * @options:	Any combination of DPNI_FS_OPT_ values.
+ */
+struct dpni_fs_action_cfg {
+	u64 flc;
+	u16 flow_id;
+	u16 options;
+};
+
+int dpni_add_fs_entry(struct fsl_mc_io *mc_io,
+		      u32 cmd_flags,
+		      u16 token,
+		      u8 tc_id,
+		      u16 index,
+		      const struct dpni_rule_cfg *cfg,
+		      const struct dpni_fs_action_cfg *action);
+
+int dpni_remove_fs_entry(struct fsl_mc_io *mc_io,
+			 u32 cmd_flags,
+			 u16 token,
+			 u8 tc_id,
+			 const struct dpni_rule_cfg *cfg);
+
+int dpni_add_qos_entry(struct fsl_mc_io *mc_io,
+		       u32 cmd_flags,
+		       u16 token,
+		       const struct dpni_rule_cfg *cfg,
+		       u8 tc_id,
+		       u16 index);
+
+int dpni_remove_qos_entry(struct fsl_mc_io *mc_io,
+			  u32 cmd_flags,
+			  u16 token,
+			  const struct dpni_rule_cfg *cfg);
+
+int dpni_clear_qos_table(struct fsl_mc_io *mc_io,
+			 u32 cmd_flags,
+			 u16 token);
+
+int dpni_get_api_version(struct fsl_mc_io *mc_io,
+			 u32 cmd_flags,
+			 u16 *major_ver,
+			 u16 *minor_ver);
+/**
+ * struct dpni_tx_shaping_cfg - Structure representing DPNI tx shaping configuration
+ * @rate_limit:		Rate in Mbps
+ * @max_burst_size:	Burst size in bytes (up to 64KB)
+ */
+struct dpni_tx_shaping_cfg {
+	u32 rate_limit;
+	u16 max_burst_size;
+};
+
+int dpni_set_tx_shaping(struct fsl_mc_io *mc_io,
+			u32 cmd_flags,
+			u16 token,
+			const struct dpni_tx_shaping_cfg *tx_cr_shaper,
+			const struct dpni_tx_shaping_cfg *tx_er_shaper,
+			int coupled);
+
+/**
+ * struct dpni_single_step_cfg - configure single step PTP (IEEE 1588)
+ * @en:		enable single step PTP. When enabled the PTPv1 functionality
+ *		will not work. If the field is zero, offset and ch_update
+ *		parameters will be ignored
+ * @offset:	start offset from the beginning of the frame where
+ *		timestamp field is found. The offset must respect all MAC
+ *		headers, VLAN tags and other protocol headers
+ * @ch_update:	when set UDP checksum will be updated inside packet
+ * @peer_delay:	For peer-to-peer transparent clocks add this value to the
+ *		correction field in addition to the transient time update.
+ *		The value expresses nanoseconds.
+ * @ptp_onestep_reg_base: 1588 SINGLE_STEP register base address. This address
+ *			  is used to update directly the register contents.
+ *			  User has to create an address mapping for it.
+ *
+ *
+ */
+struct dpni_single_step_cfg {
+	u8	en;
+	u8	ch_update;
+	u16	offset;
+	u32	peer_delay;
+	u32	ptp_onestep_reg_base;
+};
+
+int dpni_set_single_step_cfg(struct fsl_mc_io *mc_io,
+			     u32 cmd_flags,
+			     u16 token,
+			     struct dpni_single_step_cfg *ptp_cfg);
+
+int dpni_get_single_step_cfg(struct fsl_mc_io *mc_io,
+			     u32 cmd_flags,
+			     u16 token,
+			     struct dpni_single_step_cfg *ptp_cfg);
+
+int dpni_enable_vlan_filter(struct fsl_mc_io *mc_io, u32 cmd_flags, u16 token,
+			    u32 en);
+
+int dpni_add_vlan_id(struct fsl_mc_io *mc_io, u32 cmd_flags, u16 token,
+		     u16 vlan_id, u8 flags, u8 tc_id, u8 flow_id);
+
+int dpni_remove_vlan_id(struct fsl_mc_io *mc_io, u32 cmd_flags, u16 token,
+			u16 vlan_id);
+
+#endif /* __FSL_DPNI_H */
diff --git a/drivers/net/ethernet/freescale/dpaa2/dprtc-cmd.h b/drivers/net/ethernet/freescale/dpaa2/dprtc-cmd.h
new file mode 100644
index 000000000000..96ffeb948f08
--- /dev/null
+++ b/drivers/net/ethernet/freescale/dpaa2/dprtc-cmd.h
@@ -0,0 +1,74 @@
+/* SPDX-License-Identifier: GPL-2.0 */
+/*
+ * Copyright 2013-2016 Freescale Semiconductor Inc.
+ * Copyright 2016-2018 NXP
+ */
+
+#ifndef _FSL_DPRTC_CMD_H
+#define _FSL_DPRTC_CMD_H
+
+/* Command versioning */
+#define DPRTC_CMD_BASE_VERSION		1
+#define DPRTC_CMD_VERSION_2		2
+#define DPRTC_CMD_ID_OFFSET		4
+
+#define DPRTC_CMD(id)	(((id) << DPRTC_CMD_ID_OFFSET) | DPRTC_CMD_BASE_VERSION)
+#define DPRTC_CMD_V2(id) (((id) << DPRTC_CMD_ID_OFFSET) | DPRTC_CMD_VERSION_2)
+
+/* Command IDs */
+#define DPRTC_CMDID_CLOSE			DPRTC_CMD(0x800)
+#define DPRTC_CMDID_OPEN			DPRTC_CMD(0x810)
+
+#define DPRTC_CMDID_SET_IRQ_ENABLE		DPRTC_CMD(0x012)
+#define DPRTC_CMDID_GET_IRQ_ENABLE		DPRTC_CMD(0x013)
+#define DPRTC_CMDID_SET_IRQ_MASK		DPRTC_CMD_V2(0x014)
+#define DPRTC_CMDID_GET_IRQ_MASK		DPRTC_CMD(0x015)
+#define DPRTC_CMDID_GET_IRQ_STATUS		DPRTC_CMD(0x016)
+#define DPRTC_CMDID_CLEAR_IRQ_STATUS		DPRTC_CMD(0x017)
+
+#pragma pack(push, 1)
+struct dprtc_cmd_open {
+	__le32 dprtc_id;
+};
+
+struct dprtc_cmd_get_irq {
+	__le32 pad;
+	u8 irq_index;
+};
+
+struct dprtc_cmd_set_irq_enable {
+	u8 en;
+	u8 pad[3];
+	u8 irq_index;
+};
+
+struct dprtc_rsp_get_irq_enable {
+	u8 en;
+};
+
+struct dprtc_cmd_set_irq_mask {
+	__le32 mask;
+	u8 irq_index;
+};
+
+struct dprtc_rsp_get_irq_mask {
+	__le32 mask;
+};
+
+struct dprtc_cmd_get_irq_status {
+	__le32 status;
+	u8 irq_index;
+};
+
+struct dprtc_rsp_get_irq_status {
+	__le32 status;
+};
+
+struct dprtc_cmd_clear_irq_status {
+	__le32 status;
+	u8 irq_index;
+};
+
+#pragma pack(pop)
+
+#endif /* _FSL_DPRTC_CMD_H */
diff --git a/drivers/net/ethernet/freescale/dpaa2/dprtc.c b/drivers/net/ethernet/freescale/dpaa2/dprtc.c
new file mode 100644
index 000000000000..ed52a34fa6a1
--- /dev/null
+++ b/drivers/net/ethernet/freescale/dpaa2/dprtc.c
@@ -0,0 +1,293 @@
+// SPDX-License-Identifier: GPL-2.0
+/*
+ * Copyright 2013-2016 Freescale Semiconductor Inc.
+ * Copyright 2016-2018 NXP
+ */
+
+#include <linux/fsl/mc.h>
+
+#include "dprtc.h"
+#include "dprtc-cmd.h"
+
+/**
+ * dprtc_open() - Open a control session for the specified object.
+ * @mc_io:	Pointer to MC portal's I/O object
+ * @cmd_flags:	Command flags; one or more of 'MC_CMD_FLAG_'
+ * @dprtc_id:	DPRTC unique ID
+ * @token:	Returned token; use in subsequent API calls
+ *
+ * This function can be used to open a control session for an
+ * already created object; an object may have been declared in
+ * the DPL or by calling the dprtc_create function.
+ * This function returns a unique authentication token,
+ * associated with the specific object ID and the specific MC
+ * portal; this token must be used in all subsequent commands for
+ * this specific object
+ *
+ * Return:	'0' on Success; Error code otherwise.
+ */
+int dprtc_open(struct fsl_mc_io *mc_io,
+	       u32 cmd_flags,
+	       int dprtc_id,
+	       u16 *token)
+{
+	struct dprtc_cmd_open *cmd_params;
+	struct fsl_mc_command cmd = { 0 };
+	int err;
+
+	cmd.header = mc_encode_cmd_header(DPRTC_CMDID_OPEN,
+					  cmd_flags,
+					  0);
+	cmd_params = (struct dprtc_cmd_open *)cmd.params;
+	cmd_params->dprtc_id = cpu_to_le32(dprtc_id);
+
+	err = mc_send_command(mc_io, &cmd);
+	if (err)
+		return err;
+
+	*token = mc_cmd_hdr_read_token(&cmd);
+
+	return 0;
+}
+
+/**
+ * dprtc_close() - Close the control session of the object
+ * @mc_io:	Pointer to MC portal's I/O object
+ * @cmd_flags:	Command flags; one or more of 'MC_CMD_FLAG_'
+ * @token:	Token of DPRTC object
+ *
+ * After this function is called, no further operations are
+ * allowed on the object without opening a new control session.
+ *
+ * Return:	'0' on Success; Error code otherwise.
+ */
+int dprtc_close(struct fsl_mc_io *mc_io,
+		u32 cmd_flags,
+		u16 token)
+{
+	struct fsl_mc_command cmd = { 0 };
+
+	cmd.header = mc_encode_cmd_header(DPRTC_CMDID_CLOSE, cmd_flags,
+					  token);
+
+	return mc_send_command(mc_io, &cmd);
+}
+
+/**
+ * dprtc_set_irq_enable() - Set overall interrupt state.
+ * @mc_io:	Pointer to MC portal's I/O object
+ * @cmd_flags:	Command flags; one or more of 'MC_CMD_FLAG_'
+ * @token:	Token of DPRTC object
+ * @irq_index:	The interrupt index to configure
+ * @en:		Interrupt state - enable = 1, disable = 0
+ *
+ * Allows GPP software to control when interrupts are generated.
+ * Each interrupt can have up to 32 causes.  The enable/disable control's the
+ * overall interrupt state. if the interrupt is disabled no causes will cause
+ * an interrupt.
+ *
+ * Return:	'0' on Success; Error code otherwise.
+ */
+int dprtc_set_irq_enable(struct fsl_mc_io *mc_io,
+			 u32 cmd_flags,
+			 u16 token,
+			 u8 irq_index,
+			 u8 en)
+{
+	struct dprtc_cmd_set_irq_enable *cmd_params;
+	struct fsl_mc_command cmd = { 0 };
+
+	cmd.header = mc_encode_cmd_header(DPRTC_CMDID_SET_IRQ_ENABLE,
+					  cmd_flags,
+					  token);
+	cmd_params = (struct dprtc_cmd_set_irq_enable *)cmd.params;
+	cmd_params->irq_index = irq_index;
+	cmd_params->en = en;
+
+	return mc_send_command(mc_io, &cmd);
+}
+
+/**
+ * dprtc_get_irq_enable() - Get overall interrupt state
+ * @mc_io:	Pointer to MC portal's I/O object
+ * @cmd_flags:	Command flags; one or more of 'MC_CMD_FLAG_'
+ * @token:	Token of DPRTC object
+ * @irq_index:	The interrupt index to configure
+ * @en:		Returned interrupt state - enable = 1, disable = 0
+ *
+ * Return:	'0' on Success; Error code otherwise.
+ */
+int dprtc_get_irq_enable(struct fsl_mc_io *mc_io,
+			 u32 cmd_flags,
+			 u16 token,
+			 u8 irq_index,
+			 u8 *en)
+{
+	struct dprtc_rsp_get_irq_enable *rsp_params;
+	struct dprtc_cmd_get_irq *cmd_params;
+	struct fsl_mc_command cmd = { 0 };
+	int err;
+
+	cmd.header = mc_encode_cmd_header(DPRTC_CMDID_GET_IRQ_ENABLE,
+					  cmd_flags,
+					  token);
+	cmd_params = (struct dprtc_cmd_get_irq *)cmd.params;
+	cmd_params->irq_index = irq_index;
+
+	err = mc_send_command(mc_io, &cmd);
+	if (err)
+		return err;
+
+	rsp_params = (struct dprtc_rsp_get_irq_enable *)cmd.params;
+	*en = rsp_params->en;
+
+	return 0;
+}
+
+/**
+ * dprtc_set_irq_mask() - Set interrupt mask.
+ * @mc_io:	Pointer to MC portal's I/O object
+ * @cmd_flags:	Command flags; one or more of 'MC_CMD_FLAG_'
+ * @token:	Token of DPRTC object
+ * @irq_index:	The interrupt index to configure
+ * @mask:	Event mask to trigger interrupt;
+ *		each bit:
+ *			0 = ignore event
+ *			1 = consider event for asserting IRQ
+ *
+ * Every interrupt can have up to 32 causes and the interrupt model supports
+ * masking/unmasking each cause independently
+ *
+ * Return:	'0' on Success; Error code otherwise.
+ */
+int dprtc_set_irq_mask(struct fsl_mc_io *mc_io,
+		       u32 cmd_flags,
+		       u16 token,
+		       u8 irq_index,
+		       u32 mask)
+{
+	struct dprtc_cmd_set_irq_mask *cmd_params;
+	struct fsl_mc_command cmd = { 0 };
+
+	cmd.header = mc_encode_cmd_header(DPRTC_CMDID_SET_IRQ_MASK,
+					  cmd_flags,
+					  token);
+	cmd_params = (struct dprtc_cmd_set_irq_mask *)cmd.params;
+	cmd_params->mask = cpu_to_le32(mask);
+	cmd_params->irq_index = irq_index;
+
+	return mc_send_command(mc_io, &cmd);
+}
+
+/**
+ * dprtc_get_irq_mask() - Get interrupt mask.
+ * @mc_io:	Pointer to MC portal's I/O object
+ * @cmd_flags:	Command flags; one or more of 'MC_CMD_FLAG_'
+ * @token:	Token of DPRTC object
+ * @irq_index:	The interrupt index to configure
+ * @mask:	Returned event mask to trigger interrupt
+ *
+ * Every interrupt can have up to 32 causes and the interrupt model supports
+ * masking/unmasking each cause independently
+ *
+ * Return:	'0' on Success; Error code otherwise.
+ */
+int dprtc_get_irq_mask(struct fsl_mc_io *mc_io,
+		       u32 cmd_flags,
+		       u16 token,
+		       u8 irq_index,
+		       u32 *mask)
+{
+	struct dprtc_rsp_get_irq_mask *rsp_params;
+	struct dprtc_cmd_get_irq *cmd_params;
+	struct fsl_mc_command cmd = { 0 };
+	int err;
+
+	cmd.header = mc_encode_cmd_header(DPRTC_CMDID_GET_IRQ_MASK,
+					  cmd_flags,
+					  token);
+	cmd_params = (struct dprtc_cmd_get_irq *)cmd.params;
+	cmd_params->irq_index = irq_index;
+
+	err = mc_send_command(mc_io, &cmd);
+	if (err)
+		return err;
+
+	rsp_params = (struct dprtc_rsp_get_irq_mask *)cmd.params;
+	*mask = le32_to_cpu(rsp_params->mask);
+
+	return 0;
+}
+
+/**
+ * dprtc_get_irq_status() - Get the current status of any pending interrupts.
+ *
+ * @mc_io:	Pointer to MC portal's I/O object
+ * @cmd_flags:	Command flags; one or more of 'MC_CMD_FLAG_'
+ * @token:	Token of DPRTC object
+ * @irq_index:	The interrupt index to configure
+ * @status:	Returned interrupts status - one bit per cause:
+ *			0 = no interrupt pending
+ *			1 = interrupt pending
+ *
+ * Return:	'0' on Success; Error code otherwise.
+ */
+int dprtc_get_irq_status(struct fsl_mc_io *mc_io,
+			 u32 cmd_flags,
+			 u16 token,
+			 u8 irq_index,
+			 u32 *status)
+{
+	struct dprtc_cmd_get_irq_status *cmd_params;
+	struct dprtc_rsp_get_irq_status *rsp_params;
+	struct fsl_mc_command cmd = { 0 };
+	int err;
+
+	cmd.header = mc_encode_cmd_header(DPRTC_CMDID_GET_IRQ_STATUS,
+					  cmd_flags,
+					  token);
+	cmd_params = (struct dprtc_cmd_get_irq_status *)cmd.params;
+	cmd_params->status = cpu_to_le32(*status);
+	cmd_params->irq_index = irq_index;
+
+	err = mc_send_command(mc_io, &cmd);
+	if (err)
+		return err;
+
+	rsp_params = (struct dprtc_rsp_get_irq_status *)cmd.params;
+	*status = le32_to_cpu(rsp_params->status);
+
+	return 0;
+}
+
+/**
+ * dprtc_clear_irq_status() - Clear a pending interrupt's status
+ *
+ * @mc_io:	Pointer to MC portal's I/O object
+ * @cmd_flags:	Command flags; one or more of 'MC_CMD_FLAG_'
+ * @token:	Token of DPRTC object
+ * @irq_index:	The interrupt index to configure
+ * @status:	Bits to clear (W1C) - one bit per cause:
+ *			0 = don't change
+ *			1 = clear status bit
+ *
+ * Return:	'0' on Success; Error code otherwise.
+ */
+int dprtc_clear_irq_status(struct fsl_mc_io *mc_io,
+			   u32 cmd_flags,
+			   u16 token,
+			   u8 irq_index,
+			   u32 status)
+{
+	struct dprtc_cmd_clear_irq_status *cmd_params;
+	struct fsl_mc_command cmd = { 0 };
+
+	cmd.header = mc_encode_cmd_header(DPRTC_CMDID_CLEAR_IRQ_STATUS,
+					  cmd_flags,
+					  token);
+	cmd_params = (struct dprtc_cmd_clear_irq_status *)cmd.params;
+	cmd_params->irq_index = irq_index;
+	cmd_params->status = cpu_to_le32(status);
+
+	return mc_send_command(mc_io, &cmd);
+}
diff --git a/drivers/net/ethernet/freescale/dpaa2/dprtc.h b/drivers/net/ethernet/freescale/dpaa2/dprtc.h
new file mode 100644
index 000000000000..01d77c685a5b
--- /dev/null
+++ b/drivers/net/ethernet/freescale/dpaa2/dprtc.h
@@ -0,0 +1,68 @@
+/* SPDX-License-Identifier: GPL-2.0 */
+/*
+ * Copyright 2013-2016 Freescale Semiconductor Inc.
+ * Copyright 2016-2018 NXP
+ */
+
+#ifndef __FSL_DPRTC_H
+#define __FSL_DPRTC_H
+
+/* Data Path Real Time Counter API
+ * Contains initialization APIs and runtime control APIs for RTC
+ */
+
+struct fsl_mc_io;
+
+#define DPRTC_MAX_IRQ_NUM	1
+#define DPRTC_IRQ_INDEX		0
+
+#define DPRTC_EVENT_PPS		0x08000000
+#define DPRTC_EVENT_ETS1	0x00800000
+#define DPRTC_EVENT_ETS2	0x00400000
+
+int dprtc_open(struct fsl_mc_io *mc_io,
+	       u32 cmd_flags,
+	       int dprtc_id,
+	       u16 *token);
+
+int dprtc_close(struct fsl_mc_io *mc_io,
+		u32 cmd_flags,
+		u16 token);
+
+int dprtc_set_irq_enable(struct fsl_mc_io *mc_io,
+			 u32 cmd_flags,
+			 u16 token,
+			 u8 irq_index,
+			 u8 en);
+
+int dprtc_get_irq_enable(struct fsl_mc_io *mc_io,
+			 u32 cmd_flags,
+			 u16 token,
+			 u8 irq_index,
+			 u8 *en);
+
+int dprtc_set_irq_mask(struct fsl_mc_io *mc_io,
+		       u32 cmd_flags,
+		       u16 token,
+		       u8 irq_index,
+		       u32 mask);
+
+int dprtc_get_irq_mask(struct fsl_mc_io *mc_io,
+		       u32 cmd_flags,
+		       u16 token,
+		       u8 irq_index,
+		       u32 *mask);
+
+int dprtc_get_irq_status(struct fsl_mc_io *mc_io,
+			 u32 cmd_flags,
+			 u16 token,
+			 u8 irq_index,
+			 u32 *status);
+
+int dprtc_clear_irq_status(struct fsl_mc_io *mc_io,
+			   u32 cmd_flags,
+			   u16 token,
+			   u8 irq_index,
+			   u32 status);
+
+#endif /* __FSL_DPRTC_H */
diff --git a/drivers/net/ethernet/freescale/dpaa2/dpsw-cmd.h b/drivers/net/ethernet/freescale/dpaa2/dpsw-cmd.h
new file mode 100644
index 000000000000..397d55f2bd99
--- /dev/null
+++ b/drivers/net/ethernet/freescale/dpaa2/dpsw-cmd.h
@@ -0,0 +1,556 @@
+/* SPDX-License-Identifier: GPL-2.0 */
+/*
+ * Copyright 2014-2016 Freescale Semiconductor Inc.
+ * Copyright 2017-2021 NXP
+ *
+ */
+
+#ifndef __FSL_DPSW_CMD_H
+#define __FSL_DPSW_CMD_H
+
+#include "dpsw.h"
+
+/* DPSW Version */
+#define DPSW_VER_MAJOR		8
+#define DPSW_VER_MINOR		9
+
+#define DPSW_CMD_BASE_VERSION	1
+#define DPSW_CMD_VERSION_2	2
+#define DPSW_CMD_ID_OFFSET	4
+
+#define DPSW_CMD_ID(id)	(((id) << DPSW_CMD_ID_OFFSET) | DPSW_CMD_BASE_VERSION)
+#define DPSW_CMD_V2(id) (((id) << DPSW_CMD_ID_OFFSET) | DPSW_CMD_VERSION_2)
+
+/* Command IDs */
+#define DPSW_CMDID_CLOSE                    DPSW_CMD_ID(0x800)
+#define DPSW_CMDID_OPEN                     DPSW_CMD_ID(0x802)
+
+#define DPSW_CMDID_GET_API_VERSION          DPSW_CMD_ID(0xa02)
+
+#define DPSW_CMDID_ENABLE                   DPSW_CMD_ID(0x002)
+#define DPSW_CMDID_DISABLE                  DPSW_CMD_ID(0x003)
+#define DPSW_CMDID_GET_ATTR                 DPSW_CMD_V2(0x004)
+#define DPSW_CMDID_RESET                    DPSW_CMD_ID(0x005)
+
+#define DPSW_CMDID_SET_IRQ_ENABLE           DPSW_CMD_ID(0x012)
+
+#define DPSW_CMDID_SET_IRQ_MASK             DPSW_CMD_ID(0x014)
+
+#define DPSW_CMDID_GET_IRQ_STATUS           DPSW_CMD_ID(0x016)
+#define DPSW_CMDID_CLEAR_IRQ_STATUS         DPSW_CMD_ID(0x017)
+
+#define DPSW_CMDID_SET_REFLECTION_IF        DPSW_CMD_ID(0x022)
+
+#define DPSW_CMDID_IF_SET_TCI               DPSW_CMD_ID(0x030)
+#define DPSW_CMDID_IF_SET_STP               DPSW_CMD_ID(0x031)
+
+#define DPSW_CMDID_IF_GET_COUNTER           DPSW_CMD_V2(0x034)
+
+#define DPSW_CMDID_IF_ADD_REFLECTION        DPSW_CMD_ID(0x037)
+#define DPSW_CMDID_IF_REMOVE_REFLECTION     DPSW_CMD_ID(0x038)
+
+#define DPSW_CMDID_IF_ENABLE                DPSW_CMD_ID(0x03D)
+#define DPSW_CMDID_IF_DISABLE               DPSW_CMD_ID(0x03E)
+
+#define DPSW_CMDID_IF_GET_ATTR              DPSW_CMD_ID(0x042)
+
+#define DPSW_CMDID_IF_SET_MAX_FRAME_LENGTH  DPSW_CMD_ID(0x044)
+
+#define DPSW_CMDID_IF_GET_LINK_STATE        DPSW_CMD_ID(0x046)
+
+#define DPSW_CMDID_IF_GET_TCI               DPSW_CMD_ID(0x04A)
+
+#define DPSW_CMDID_IF_SET_LINK_CFG          DPSW_CMD_ID(0x04C)
+
+#define DPSW_CMDID_VLAN_ADD                 DPSW_CMD_ID(0x060)
+#define DPSW_CMDID_VLAN_ADD_IF              DPSW_CMD_V2(0x061)
+#define DPSW_CMDID_VLAN_ADD_IF_UNTAGGED     DPSW_CMD_ID(0x062)
+
+#define DPSW_CMDID_VLAN_REMOVE_IF           DPSW_CMD_ID(0x064)
+#define DPSW_CMDID_VLAN_REMOVE_IF_UNTAGGED  DPSW_CMD_ID(0x065)
+#define DPSW_CMDID_VLAN_REMOVE_IF_FLOODING  DPSW_CMD_ID(0x066)
+#define DPSW_CMDID_VLAN_REMOVE              DPSW_CMD_ID(0x067)
+
+#define DPSW_CMDID_FDB_ADD                  DPSW_CMD_ID(0x082)
+#define DPSW_CMDID_FDB_REMOVE               DPSW_CMD_ID(0x083)
+#define DPSW_CMDID_FDB_ADD_UNICAST          DPSW_CMD_ID(0x084)
+#define DPSW_CMDID_FDB_REMOVE_UNICAST       DPSW_CMD_ID(0x085)
+#define DPSW_CMDID_FDB_ADD_MULTICAST        DPSW_CMD_ID(0x086)
+#define DPSW_CMDID_FDB_REMOVE_MULTICAST     DPSW_CMD_ID(0x087)
+#define DPSW_CMDID_FDB_DUMP                 DPSW_CMD_ID(0x08A)
+
+#define DPSW_CMDID_ACL_ADD                  DPSW_CMD_ID(0x090)
+#define DPSW_CMDID_ACL_REMOVE               DPSW_CMD_ID(0x091)
+#define DPSW_CMDID_ACL_ADD_ENTRY            DPSW_CMD_ID(0x092)
+#define DPSW_CMDID_ACL_REMOVE_ENTRY         DPSW_CMD_ID(0x093)
+#define DPSW_CMDID_ACL_ADD_IF               DPSW_CMD_ID(0x094)
+#define DPSW_CMDID_ACL_REMOVE_IF            DPSW_CMD_ID(0x095)
+
+#define DPSW_CMDID_IF_GET_PORT_MAC_ADDR     DPSW_CMD_ID(0x0A7)
+
+#define DPSW_CMDID_CTRL_IF_GET_ATTR         DPSW_CMD_ID(0x0A0)
+#define DPSW_CMDID_CTRL_IF_SET_POOLS        DPSW_CMD_ID(0x0A1)
+#define DPSW_CMDID_CTRL_IF_ENABLE           DPSW_CMD_ID(0x0A2)
+#define DPSW_CMDID_CTRL_IF_DISABLE          DPSW_CMD_ID(0x0A3)
+#define DPSW_CMDID_CTRL_IF_SET_QUEUE        DPSW_CMD_ID(0x0A6)
+
+#define DPSW_CMDID_SET_EGRESS_FLOOD         DPSW_CMD_ID(0x0AC)
+#define DPSW_CMDID_IF_SET_LEARNING_MODE     DPSW_CMD_ID(0x0AD)
+
+/* Macros for accessing command fields smaller than 1byte */
+#define DPSW_MASK(field)        \
+	GENMASK(DPSW_##field##_SHIFT + DPSW_##field##_SIZE - 1, \
+		DPSW_##field##_SHIFT)
+#define dpsw_set_field(var, field, val) \
+	((var) |= (((val) << DPSW_##field##_SHIFT) & DPSW_MASK(field)))
+#define dpsw_get_field(var, field)      \
+	(((var) & DPSW_MASK(field)) >> DPSW_##field##_SHIFT)
+#define dpsw_get_bit(var, bit) \
+	(((var)  >> (bit)) & GENMASK(0, 0))
+
+#pragma pack(push, 1)
+struct dpsw_cmd_open {
+	__le32 dpsw_id;
+};
+
+#define DPSW_COMPONENT_TYPE_SHIFT	0
+#define DPSW_COMPONENT_TYPE_SIZE	4
+
+struct dpsw_cmd_create {
+	/* cmd word 0 */
+	__le16 num_ifs;
+	u8 max_fdbs;
+	u8 max_meters_per_if;
+	/* from LSB: only the first 4 bits */
+	u8 component_type;
+	u8 pad[3];
+	/* cmd word 1 */
+	__le16 max_vlans;
+	__le16 max_fdb_entries;
+	__le16 fdb_aging_time;
+	__le16 max_fdb_mc_groups;
+	/* cmd word 2 */
+	__le64 options;
+};
+
+struct dpsw_cmd_destroy {
+	__le32 dpsw_id;
+};
+
+#define DPSW_ENABLE_SHIFT 0
+#define DPSW_ENABLE_SIZE  1
+
+struct dpsw_rsp_is_enabled {
+	/* from LSB: enable:1 */
+	u8 enabled;
+};
+
+struct dpsw_cmd_set_irq_enable {
+	u8 enable_state;
+	u8 pad[3];
+	u8 irq_index;
+};
+
+struct dpsw_cmd_get_irq_enable {
+	__le32 pad;
+	u8 irq_index;
+};
+
+struct dpsw_rsp_get_irq_enable {
+	u8 enable_state;
+};
+
+struct dpsw_cmd_set_irq_mask {
+	__le32 mask;
+	u8 irq_index;
+};
+
+struct dpsw_cmd_get_irq_mask {
+	__le32 pad;
+	u8 irq_index;
+};
+
+struct dpsw_rsp_get_irq_mask {
+	__le32 mask;
+};
+
+struct dpsw_cmd_get_irq_status {
+	__le32 status;
+	u8 irq_index;
+};
+
+struct dpsw_rsp_get_irq_status {
+	__le32 status;
+};
+
+struct dpsw_cmd_clear_irq_status {
+	__le32 status;
+	u8 irq_index;
+};
+
+#define DPSW_COMPONENT_TYPE_SHIFT	0
+#define DPSW_COMPONENT_TYPE_SIZE	4
+
+#define DPSW_FLOODING_CFG_SHIFT		0
+#define DPSW_FLOODING_CFG_SIZE		4
+
+#define DPSW_BROADCAST_CFG_SHIFT	4
+#define DPSW_BROADCAST_CFG_SIZE		4
+
+struct dpsw_rsp_get_attr {
+	/* cmd word 0 */
+	__le16 num_ifs;
+	u8 max_fdbs;
+	u8 num_fdbs;
+	__le16 max_vlans;
+	__le16 num_vlans;
+	/* cmd word 1 */
+	__le16 max_fdb_entries;
+	__le16 fdb_aging_time;
+	__le32 dpsw_id;
+	/* cmd word 2 */
+	__le16 mem_size;
+	__le16 max_fdb_mc_groups;
+	u8 max_meters_per_if;
+	/* from LSB only the first 4 bits */
+	u8 component_type;
+	/* [0:3] - flooding configuration
+	 * [4:7] - broadcast configuration
+	 */
+	u8 repl_cfg;
+	u8 pad;
+	/* cmd word 3 */
+	__le64 options;
+};
+
+#define DPSW_VLAN_ID_SHIFT	0
+#define DPSW_VLAN_ID_SIZE	12
+#define DPSW_DEI_SHIFT		12
+#define DPSW_DEI_SIZE		1
+#define DPSW_PCP_SHIFT		13
+#define DPSW_PCP_SIZE		3
+
+struct dpsw_cmd_if_set_tci {
+	__le16 if_id;
+	/* from LSB: VLAN_ID:12 DEI:1 PCP:3 */
+	__le16 conf;
+};
+
+struct dpsw_cmd_if_get_tci {
+	__le16 if_id;
+};
+
+struct dpsw_rsp_if_get_tci {
+	__le16 pad;
+	__le16 vlan_id;
+	u8 dei;
+	u8 pcp;
+};
+
+#define DPSW_STATE_SHIFT	0
+#define DPSW_STATE_SIZE		4
+
+struct dpsw_cmd_if_set_stp {
+	__le16 if_id;
+	__le16 vlan_id;
+	/* only the first LSB 4 bits */
+	u8 state;
+};
+
+#define DPSW_COUNTER_TYPE_SHIFT		0
+#define DPSW_COUNTER_TYPE_SIZE		5
+
+struct dpsw_cmd_if_get_counter {
+	__le16 if_id;
+	/* from LSB: type:5 */
+	u8 type;
+};
+
+struct dpsw_rsp_if_get_counter {
+	__le64 pad;
+	__le64 counter;
+};
+
+struct dpsw_cmd_if {
+	__le16 if_id;
+};
+
+#define DPSW_ADMIT_UNTAGGED_SHIFT	0
+#define DPSW_ADMIT_UNTAGGED_SIZE	4
+#define DPSW_ENABLED_SHIFT		5
+#define DPSW_ENABLED_SIZE		1
+#define DPSW_ACCEPT_ALL_VLAN_SHIFT	6
+#define DPSW_ACCEPT_ALL_VLAN_SIZE	1
+
+struct dpsw_rsp_if_get_attr {
+	/* cmd word 0 */
+	/* from LSB: admit_untagged:4 enabled:1 accept_all_vlan:1 */
+	u8 conf;
+	u8 pad1;
+	u8 num_tcs;
+	u8 pad2;
+	__le16 qdid;
+	/* cmd word 1 */
+	__le32 options;
+	__le32 pad3;
+	/* cmd word 2 */
+	__le32 rate;
+};
+
+struct dpsw_cmd_if_set_max_frame_length {
+	__le16 if_id;
+	__le16 frame_length;
+};
+
+struct dpsw_cmd_if_set_link_cfg {
+	/* cmd word 0 */
+	__le16 if_id;
+	u8 pad[6];
+	/* cmd word 1 */
+	__le32 rate;
+	__le32 pad1;
+	/* cmd word 2 */
+	__le64 options;
+};
+
+struct dpsw_cmd_if_get_link_state {
+	__le16 if_id;
+};
+
+#define DPSW_UP_SHIFT	0
+#define DPSW_UP_SIZE	1
+
+struct dpsw_rsp_if_get_link_state {
+	/* cmd word 0 */
+	__le32 pad0;
+	u8 up;
+	u8 pad1[3];
+	/* cmd word 1 */
+	__le32 rate;
+	__le32 pad2;
+	/* cmd word 2 */
+	__le64 options;
+};
+
+struct dpsw_vlan_add {
+	__le16 fdb_id;
+	__le16 vlan_id;
+};
+
+struct dpsw_cmd_vlan_add_if {
+	/* cmd word 0 */
+	__le16 options;
+	__le16 vlan_id;
+	__le16 fdb_id;
+	__le16 pad0;
+	/* cmd word 1-4 */
+	__le64 if_id;
+};
+
+struct dpsw_cmd_vlan_manage_if {
+	/* cmd word 0 */
+	__le16 pad0;
+	__le16 vlan_id;
+	__le32 pad1;
+	/* cmd word 1-4 */
+	__le64 if_id;
+};
+
+struct dpsw_cmd_vlan_remove {
+	__le16 pad;
+	__le16 vlan_id;
+};
+
+struct dpsw_cmd_fdb_add {
+	__le32 pad;
+	__le16 fdb_ageing_time;
+	__le16 num_fdb_entries;
+};
+
+struct dpsw_rsp_fdb_add {
+	__le16 fdb_id;
+};
+
+struct dpsw_cmd_fdb_remove {
+	__le16 fdb_id;
+};
+
+#define DPSW_ENTRY_TYPE_SHIFT	0
+#define DPSW_ENTRY_TYPE_SIZE	4
+
+struct dpsw_cmd_fdb_unicast_op {
+	/* cmd word 0 */
+	__le16 fdb_id;
+	u8 mac_addr[6];
+	/* cmd word 1 */
+	__le16 if_egress;
+	/* only the first 4 bits from LSB */
+	u8 type;
+};
+
+struct dpsw_cmd_fdb_multicast_op {
+	/* cmd word 0 */
+	__le16 fdb_id;
+	__le16 num_ifs;
+	/* only the first 4 bits from LSB */
+	u8 type;
+	u8 pad[3];
+	/* cmd word 1 */
+	u8 mac_addr[6];
+	__le16 pad2;
+	/* cmd word 2-5 */
+	__le64 if_id;
+};
+
+struct dpsw_cmd_fdb_dump {
+	__le16 fdb_id;
+	__le16 pad0;
+	__le32 pad1;
+	__le64 iova_addr;
+	__le32 iova_size;
+};
+
+struct dpsw_rsp_fdb_dump {
+	__le16 num_entries;
+};
+
+struct dpsw_rsp_ctrl_if_get_attr {
+	__le64 pad;
+	__le32 rx_fqid;
+	__le32 rx_err_fqid;
+	__le32 tx_err_conf_fqid;
+};
+
+#define DPSW_BACKUP_POOL(val, order)	(((val) & 0x1) << (order))
+struct dpsw_cmd_ctrl_if_set_pools {
+	u8 num_dpbp;
+	u8 backup_pool_mask;
+	__le16 pad;
+	__le32 dpbp_id[DPSW_MAX_DPBP];
+	__le16 buffer_size[DPSW_MAX_DPBP];
+};
+
+#define DPSW_DEST_TYPE_SHIFT	0
+#define DPSW_DEST_TYPE_SIZE	4
+
+struct dpsw_cmd_ctrl_if_set_queue {
+	__le32 dest_id;
+	u8 dest_priority;
+	u8 pad;
+	/* from LSB: dest_type:4 */
+	u8 dest_type;
+	u8 qtype;
+	__le64 user_ctx;
+	__le32 options;
+};
+
+struct dpsw_rsp_get_api_version {
+	__le16 version_major;
+	__le16 version_minor;
+};
+
+struct dpsw_rsp_if_get_mac_addr {
+	__le16 pad;
+	u8 mac_addr[6];
+};
+
+struct dpsw_cmd_set_egress_flood {
+	__le16 fdb_id;
+	u8 flood_type;
+	u8 pad[5];
+	__le64 if_id;
+};
+
+#define DPSW_LEARNING_MODE_SHIFT	0
+#define DPSW_LEARNING_MODE_SIZE		4
+
+struct dpsw_cmd_if_set_learning_mode {
+	__le16 if_id;
+	/* only the first 4 bits from LSB */
+	u8 mode;
+};
+
+struct dpsw_cmd_acl_add {
+	__le16 pad;
+	__le16 max_entries;
+};
+
+struct dpsw_rsp_acl_add {
+	__le16 acl_id;
+};
+
+struct dpsw_cmd_acl_remove {
+	__le16 acl_id;
+};
+
+struct dpsw_cmd_acl_if {
+	__le16 acl_id;
+	__le16 num_ifs;
+	__le32 pad;
+	__le64 if_id;
+};
+
+struct dpsw_prep_acl_entry {
+	u8 match_l2_dest_mac[6];
+	__le16 match_l2_tpid;
+
+	u8 match_l2_source_mac[6];
+	__le16 match_l2_vlan_id;
+
+	__le32 match_l3_dest_ip;
+	__le32 match_l3_source_ip;
+
+	__le16 match_l4_dest_port;
+	__le16 match_l4_source_port;
+	__le16 match_l2_ether_type;
+	u8 match_l2_pcp_dei;
+	u8 match_l3_dscp;
+
+	u8 mask_l2_dest_mac[6];
+	__le16 mask_l2_tpid;
+
+	u8 mask_l2_source_mac[6];
+	__le16 mask_l2_vlan_id;
+
+	__le32 mask_l3_dest_ip;
+	__le32 mask_l3_source_ip;
+
+	__le16 mask_l4_dest_port;
+	__le16 mask_l4_source_port;
+	__le16 mask_l2_ether_type;
+	u8 mask_l2_pcp_dei;
+	u8 mask_l3_dscp;
+
+	u8 match_l3_protocol;
+	u8 mask_l3_protocol;
+};
+
+#define DPSW_RESULT_ACTION_SHIFT	0
+#define DPSW_RESULT_ACTION_SIZE		4
+
+struct dpsw_cmd_acl_entry {
+	__le16 acl_id;
+	__le16 result_if_id;
+	__le32 precedence;
+	/* from LSB only the first 4 bits */
+	u8 result_action;
+	u8 pad[7];
+	__le64 pad2[4];
+	__le64 key_iova;
+};
+
+struct dpsw_cmd_set_reflection_if {
+	__le16 if_id;
+};
+
+#define DPSW_FILTER_SHIFT	0
+#define DPSW_FILTER_SIZE	2
+
+struct dpsw_cmd_if_reflection {
+	__le16 if_id;
+	__le16 vlan_id;
+	/* only 2 bits from the LSB */
+	u8 filter;
+};
+#pragma pack(pop)
+#endif /* __FSL_DPSW_CMD_H */
diff --git a/drivers/net/ethernet/freescale/dpaa2/dpsw.c b/drivers/net/ethernet/freescale/dpaa2/dpsw.c
new file mode 100644
index 000000000000..ab921d75deb2
--- /dev/null
+++ b/drivers/net/ethernet/freescale/dpaa2/dpsw.c
@@ -0,0 +1,1661 @@
+// SPDX-License-Identifier: GPL-2.0
+/*
+ * Copyright 2014-2016 Freescale Semiconductor Inc.
+ * Copyright 2017-2021 NXP
+ *
+ */
+
+#include <linux/fsl/mc.h>
+#include "dpsw.h"
+#include "dpsw-cmd.h"
+
+static void build_if_id_bitmap(__le64 *bmap, const u16 *id, const u16 num_ifs)
+{
+	int i;
+
+	for (i = 0; (i < num_ifs) && (i < DPSW_MAX_IF); i++) {
+		if (id[i] < DPSW_MAX_IF)
+			bmap[id[i] / 64] |= cpu_to_le64(BIT_MASK(id[i] % 64));
+	}
+}
+
+/**
+ * dpsw_open() - Open a control session for the specified object
+ * @mc_io:	Pointer to MC portal's I/O object
+ * @cmd_flags:	Command flags; one or more of 'MC_CMD_FLAG_'
+ * @dpsw_id:	DPSW unique ID
+ * @token:	Returned token; use in subsequent API calls
+ *
+ * This function can be used to open a control session for an
+ * already created object; an object may have been declared in
+ * the DPL or by calling the dpsw_create() function.
+ * This function returns a unique authentication token,
+ * associated with the specific object ID and the specific MC
+ * portal; this token must be used in all subsequent commands for
+ * this specific object
+ *
+ * Return:	'0' on Success; Error code otherwise.
+ */
+int dpsw_open(struct fsl_mc_io *mc_io, u32 cmd_flags, int dpsw_id, u16 *token)
+{
+	struct fsl_mc_command cmd = { 0 };
+	struct dpsw_cmd_open *cmd_params;
+	int err;
+
+	/* prepare command */
+	cmd.header = mc_encode_cmd_header(DPSW_CMDID_OPEN,
+					  cmd_flags,
+					  0);
+	cmd_params = (struct dpsw_cmd_open *)cmd.params;
+	cmd_params->dpsw_id = cpu_to_le32(dpsw_id);
+
+	/* send command to mc*/
+	err = mc_send_command(mc_io, &cmd);
+	if (err)
+		return err;
+
+	/* retrieve response parameters */
+	*token = mc_cmd_hdr_read_token(&cmd);
+
+	return 0;
+}
+
+/**
+ * dpsw_close() - Close the control session of the object
+ * @mc_io:	Pointer to MC portal's I/O object
+ * @cmd_flags:	Command flags; one or more of 'MC_CMD_FLAG_'
+ * @token:	Token of DPSW object
+ *
+ * After this function is called, no further operations are
+ * allowed on the object without opening a new control session.
+ *
+ * Return:	'0' on Success; Error code otherwise.
+ */
+int dpsw_close(struct fsl_mc_io *mc_io, u32 cmd_flags, u16 token)
+{
+	struct fsl_mc_command cmd = { 0 };
+
+	/* prepare command */
+	cmd.header = mc_encode_cmd_header(DPSW_CMDID_CLOSE,
+					  cmd_flags,
+					  token);
+
+	/* send command to mc*/
+	return mc_send_command(mc_io, &cmd);
+}
+
+/**
+ * dpsw_enable() - Enable DPSW functionality
+ * @mc_io:	Pointer to MC portal's I/O object
+ * @cmd_flags:	Command flags; one or more of 'MC_CMD_FLAG_'
+ * @token:	Token of DPSW object
+ *
+ * Return:	Completion status. '0' on Success; Error code otherwise.
+ */
+int dpsw_enable(struct fsl_mc_io *mc_io, u32 cmd_flags, u16 token)
+{
+	struct fsl_mc_command cmd = { 0 };
+
+	/* prepare command */
+	cmd.header = mc_encode_cmd_header(DPSW_CMDID_ENABLE,
+					  cmd_flags,
+					  token);
+
+	/* send command to mc*/
+	return mc_send_command(mc_io, &cmd);
+}
+
+/**
+ * dpsw_disable() - Disable DPSW functionality
+ * @mc_io:	Pointer to MC portal's I/O object
+ * @cmd_flags:	Command flags; one or more of 'MC_CMD_FLAG_'
+ * @token:	Token of DPSW object
+ *
+ * Return:	Completion status. '0' on Success; Error code otherwise.
+ */
+int dpsw_disable(struct fsl_mc_io *mc_io, u32 cmd_flags, u16 token)
+{
+	struct fsl_mc_command cmd = { 0 };
+
+	/* prepare command */
+	cmd.header = mc_encode_cmd_header(DPSW_CMDID_DISABLE,
+					  cmd_flags,
+					  token);
+
+	/* send command to mc*/
+	return mc_send_command(mc_io, &cmd);
+}
+
+/**
+ * dpsw_reset() - Reset the DPSW, returns the object to initial state.
+ * @mc_io:	Pointer to MC portal's I/O object
+ * @cmd_flags:	Command flags; one or more of 'MC_CMD_FLAG_'
+ * @token:	Token of DPSW object
+ *
+ * Return:	'0' on Success; Error code otherwise.
+ */
+int dpsw_reset(struct fsl_mc_io *mc_io, u32 cmd_flags, u16 token)
+{
+	struct fsl_mc_command cmd = { 0 };
+
+	/* prepare command */
+	cmd.header = mc_encode_cmd_header(DPSW_CMDID_RESET,
+					  cmd_flags,
+					  token);
+
+	/* send command to mc*/
+	return mc_send_command(mc_io, &cmd);
+}
+
+/**
+ * dpsw_set_irq_enable() - Set overall interrupt state.
+ * @mc_io:	Pointer to MC portal's I/O object
+ * @cmd_flags:	Command flags; one or more of 'MC_CMD_FLAG_'
+ * @token:	Token of DPCI object
+ * @irq_index:	The interrupt index to configure
+ * @en:		Interrupt state - enable = 1, disable = 0
+ *
+ * Allows GPP software to control when interrupts are generated.
+ * Each interrupt can have up to 32 causes.  The enable/disable control's the
+ * overall interrupt state. if the interrupt is disabled no causes will cause
+ * an interrupt
+ *
+ * Return:	'0' on Success; Error code otherwise.
+ */
+int dpsw_set_irq_enable(struct fsl_mc_io *mc_io, u32 cmd_flags, u16 token,
+			u8 irq_index, u8 en)
+{
+	struct fsl_mc_command cmd = { 0 };
+	struct dpsw_cmd_set_irq_enable *cmd_params;
+
+	/* prepare command */
+	cmd.header = mc_encode_cmd_header(DPSW_CMDID_SET_IRQ_ENABLE,
+					  cmd_flags,
+					  token);
+	cmd_params = (struct dpsw_cmd_set_irq_enable *)cmd.params;
+	dpsw_set_field(cmd_params->enable_state, ENABLE, en);
+	cmd_params->irq_index = irq_index;
+
+	/* send command to mc*/
+	return mc_send_command(mc_io, &cmd);
+}
+
+/**
+ * dpsw_set_irq_mask() - Set interrupt mask.
+ * @mc_io:	Pointer to MC portal's I/O object
+ * @cmd_flags:	Command flags; one or more of 'MC_CMD_FLAG_'
+ * @token:	Token of DPCI object
+ * @irq_index:	The interrupt index to configure
+ * @mask:	Event mask to trigger interrupt;
+ *		each bit:
+ *			0 = ignore event
+ *			1 = consider event for asserting IRQ
+ *
+ * Every interrupt can have up to 32 causes and the interrupt model supports
+ * masking/unmasking each cause independently
+ *
+ * Return:	'0' on Success; Error code otherwise.
+ */
+int dpsw_set_irq_mask(struct fsl_mc_io *mc_io, u32 cmd_flags, u16 token,
+		      u8 irq_index, u32 mask)
+{
+	struct fsl_mc_command cmd = { 0 };
+	struct dpsw_cmd_set_irq_mask *cmd_params;
+
+	/* prepare command */
+	cmd.header = mc_encode_cmd_header(DPSW_CMDID_SET_IRQ_MASK,
+					  cmd_flags,
+					  token);
+	cmd_params = (struct dpsw_cmd_set_irq_mask *)cmd.params;
+	cmd_params->mask = cpu_to_le32(mask);
+	cmd_params->irq_index = irq_index;
+
+	/* send command to mc*/
+	return mc_send_command(mc_io, &cmd);
+}
+
+/**
+ * dpsw_get_irq_status() - Get the current status of any pending interrupts
+ * @mc_io:	Pointer to MC portal's I/O object
+ * @cmd_flags:	Command flags; one or more of 'MC_CMD_FLAG_'
+ * @token:	Token of DPSW object
+ * @irq_index:	The interrupt index to configure
+ * @status:	Returned interrupts status - one bit per cause:
+ *			0 = no interrupt pending
+ *			1 = interrupt pending
+ *
+ * Return:	'0' on Success; Error code otherwise.
+ */
+int dpsw_get_irq_status(struct fsl_mc_io *mc_io, u32 cmd_flags, u16 token,
+			u8 irq_index, u32 *status)
+{
+	struct fsl_mc_command cmd = { 0 };
+	struct dpsw_cmd_get_irq_status *cmd_params;
+	struct dpsw_rsp_get_irq_status *rsp_params;
+	int err;
+
+	/* prepare command */
+	cmd.header = mc_encode_cmd_header(DPSW_CMDID_GET_IRQ_STATUS,
+					  cmd_flags,
+					  token);
+	cmd_params = (struct dpsw_cmd_get_irq_status *)cmd.params;
+	cmd_params->status = cpu_to_le32(*status);
+	cmd_params->irq_index = irq_index;
+
+	/* send command to mc*/
+	err = mc_send_command(mc_io, &cmd);
+	if (err)
+		return err;
+
+	/* retrieve response parameters */
+	rsp_params = (struct dpsw_rsp_get_irq_status *)cmd.params;
+	*status = le32_to_cpu(rsp_params->status);
+
+	return 0;
+}
+
+/**
+ * dpsw_clear_irq_status() - Clear a pending interrupt's status
+ * @mc_io:	Pointer to MC portal's I/O object
+ * @cmd_flags:	Command flags; one or more of 'MC_CMD_FLAG_'
+ * @token:	Token of DPCI object
+ * @irq_index:	The interrupt index to configure
+ * @status:	bits to clear (W1C) - one bit per cause:
+ *			0 = don't change
+ *			1 = clear status bit
+ *
+ * Return:	'0' on Success; Error code otherwise.
+ */
+int dpsw_clear_irq_status(struct fsl_mc_io *mc_io, u32 cmd_flags, u16 token,
+			  u8 irq_index, u32 status)
+{
+	struct fsl_mc_command cmd = { 0 };
+	struct dpsw_cmd_clear_irq_status *cmd_params;
+
+	/* prepare command */
+	cmd.header = mc_encode_cmd_header(DPSW_CMDID_CLEAR_IRQ_STATUS,
+					  cmd_flags,
+					  token);
+	cmd_params = (struct dpsw_cmd_clear_irq_status *)cmd.params;
+	cmd_params->status = cpu_to_le32(status);
+	cmd_params->irq_index = irq_index;
+
+	/* send command to mc*/
+	return mc_send_command(mc_io, &cmd);
+}
+
+/**
+ * dpsw_get_attributes() - Retrieve DPSW attributes
+ * @mc_io:	Pointer to MC portal's I/O object
+ * @cmd_flags:	Command flags; one or more of 'MC_CMD_FLAG_'
+ * @token:	Token of DPSW object
+ * @attr:	Returned DPSW attributes
+ *
+ * Return:	Completion status. '0' on Success; Error code otherwise.
+ */
+int dpsw_get_attributes(struct fsl_mc_io *mc_io, u32 cmd_flags, u16 token,
+			struct dpsw_attr *attr)
+{
+	struct fsl_mc_command cmd = { 0 };
+	struct dpsw_rsp_get_attr *rsp_params;
+	int err;
+
+	/* prepare command */
+	cmd.header = mc_encode_cmd_header(DPSW_CMDID_GET_ATTR,
+					  cmd_flags,
+					  token);
+
+	/* send command to mc*/
+	err = mc_send_command(mc_io, &cmd);
+	if (err)
+		return err;
+
+	/* retrieve response parameters */
+	rsp_params = (struct dpsw_rsp_get_attr *)cmd.params;
+	attr->num_ifs = le16_to_cpu(rsp_params->num_ifs);
+	attr->max_fdbs = rsp_params->max_fdbs;
+	attr->num_fdbs = rsp_params->num_fdbs;
+	attr->max_vlans = le16_to_cpu(rsp_params->max_vlans);
+	attr->num_vlans = le16_to_cpu(rsp_params->num_vlans);
+	attr->max_fdb_entries = le16_to_cpu(rsp_params->max_fdb_entries);
+	attr->fdb_aging_time = le16_to_cpu(rsp_params->fdb_aging_time);
+	attr->id = le32_to_cpu(rsp_params->dpsw_id);
+	attr->mem_size = le16_to_cpu(rsp_params->mem_size);
+	attr->max_fdb_mc_groups = le16_to_cpu(rsp_params->max_fdb_mc_groups);
+	attr->max_meters_per_if = rsp_params->max_meters_per_if;
+	attr->options = le64_to_cpu(rsp_params->options);
+	attr->component_type = dpsw_get_field(rsp_params->component_type, COMPONENT_TYPE);
+	attr->flooding_cfg = dpsw_get_field(rsp_params->repl_cfg, FLOODING_CFG);
+	attr->broadcast_cfg = dpsw_get_field(rsp_params->repl_cfg, BROADCAST_CFG);
+	return 0;
+}
+
+/**
+ * dpsw_if_set_link_cfg() - Set the link configuration.
+ * @mc_io:	Pointer to MC portal's I/O object
+ * @cmd_flags:	Command flags; one or more of 'MC_CMD_FLAG_'
+ * @token:	Token of DPSW object
+ * @if_id:	Interface id
+ * @cfg:	Link configuration
+ *
+ * Return:	'0' on Success; Error code otherwise.
+ */
+int dpsw_if_set_link_cfg(struct fsl_mc_io *mc_io, u32 cmd_flags, u16 token, u16 if_id,
+			 struct dpsw_link_cfg *cfg)
+{
+	struct fsl_mc_command cmd = { 0 };
+	struct dpsw_cmd_if_set_link_cfg *cmd_params;
+
+	/* prepare command */
+	cmd.header = mc_encode_cmd_header(DPSW_CMDID_IF_SET_LINK_CFG,
+					  cmd_flags,
+					  token);
+	cmd_params = (struct dpsw_cmd_if_set_link_cfg *)cmd.params;
+	cmd_params->if_id = cpu_to_le16(if_id);
+	cmd_params->rate = cpu_to_le32(cfg->rate);
+	cmd_params->options = cpu_to_le64(cfg->options);
+
+	/* send command to mc*/
+	return mc_send_command(mc_io, &cmd);
+}
+
+/**
+ * dpsw_if_get_link_state - Return the link state
+ * @mc_io:	Pointer to MC portal's I/O object
+ * @cmd_flags:	Command flags; one or more of 'MC_CMD_FLAG_'
+ * @token:	Token of DPSW object
+ * @if_id:	Interface id
+ * @state:	Link state	1 - linkup, 0 - link down or disconnected
+ *
+ * Return:	'0' on Success; Error code otherwise.
+ */
+int dpsw_if_get_link_state(struct fsl_mc_io *mc_io, u32 cmd_flags, u16 token,
+			   u16 if_id, struct dpsw_link_state *state)
+{
+	struct fsl_mc_command cmd = { 0 };
+	struct dpsw_cmd_if_get_link_state *cmd_params;
+	struct dpsw_rsp_if_get_link_state *rsp_params;
+	int err;
+
+	/* prepare command */
+	cmd.header = mc_encode_cmd_header(DPSW_CMDID_IF_GET_LINK_STATE,
+					  cmd_flags,
+					  token);
+	cmd_params = (struct dpsw_cmd_if_get_link_state *)cmd.params;
+	cmd_params->if_id = cpu_to_le16(if_id);
+
+	/* send command to mc*/
+	err = mc_send_command(mc_io, &cmd);
+	if (err)
+		return err;
+
+	/* retrieve response parameters */
+	rsp_params = (struct dpsw_rsp_if_get_link_state *)cmd.params;
+	state->rate = le32_to_cpu(rsp_params->rate);
+	state->options = le64_to_cpu(rsp_params->options);
+	state->up = dpsw_get_field(rsp_params->up, UP);
+
+	return 0;
+}
+
+/**
+ * dpsw_if_set_tci() - Set default VLAN Tag Control Information (TCI)
+ * @mc_io:	Pointer to MC portal's I/O object
+ * @cmd_flags:	Command flags; one or more of 'MC_CMD_FLAG_'
+ * @token:	Token of DPSW object
+ * @if_id:	Interface Identifier
+ * @cfg:	Tag Control Information Configuration
+ *
+ * Return:	Completion status. '0' on Success; Error code otherwise.
+ */
+int dpsw_if_set_tci(struct fsl_mc_io *mc_io, u32 cmd_flags, u16 token, u16 if_id,
+		    const struct dpsw_tci_cfg *cfg)
+{
+	struct fsl_mc_command cmd = { 0 };
+	struct dpsw_cmd_if_set_tci *cmd_params;
+	u16 tmp_conf = 0;
+
+	/* prepare command */
+	cmd.header = mc_encode_cmd_header(DPSW_CMDID_IF_SET_TCI,
+					  cmd_flags,
+					  token);
+	cmd_params = (struct dpsw_cmd_if_set_tci *)cmd.params;
+	cmd_params->if_id = cpu_to_le16(if_id);
+	dpsw_set_field(tmp_conf, VLAN_ID, cfg->vlan_id);
+	dpsw_set_field(tmp_conf, DEI, cfg->dei);
+	dpsw_set_field(tmp_conf, PCP, cfg->pcp);
+	cmd_params->conf = cpu_to_le16(tmp_conf);
+
+	/* send command to mc*/
+	return mc_send_command(mc_io, &cmd);
+}
+
+/**
+ * dpsw_if_get_tci() - Get default VLAN Tag Control Information (TCI)
+ * @mc_io:	Pointer to MC portal's I/O object
+ * @cmd_flags:	Command flags; one or more of 'MC_CMD_FLAG_'
+ * @token:	Token of DPSW object
+ * @if_id:	Interface Identifier
+ * @cfg:	Tag Control Information Configuration
+ *
+ * Return:	Completion status. '0' on Success; Error code otherwise.
+ */
+int dpsw_if_get_tci(struct fsl_mc_io *mc_io, u32 cmd_flags, u16 token, u16 if_id,
+		    struct dpsw_tci_cfg *cfg)
+{
+	struct fsl_mc_command cmd = { 0 };
+	struct dpsw_cmd_if_get_tci *cmd_params;
+	struct dpsw_rsp_if_get_tci *rsp_params;
+	int err;
+
+	/* prepare command */
+	cmd.header = mc_encode_cmd_header(DPSW_CMDID_IF_GET_TCI,
+					  cmd_flags,
+					  token);
+	cmd_params = (struct dpsw_cmd_if_get_tci *)cmd.params;
+	cmd_params->if_id = cpu_to_le16(if_id);
+
+	/* send command to mc*/
+	err = mc_send_command(mc_io, &cmd);
+	if (err)
+		return err;
+
+	/* retrieve response parameters */
+	rsp_params = (struct dpsw_rsp_if_get_tci *)cmd.params;
+	cfg->pcp = rsp_params->pcp;
+	cfg->dei = rsp_params->dei;
+	cfg->vlan_id = le16_to_cpu(rsp_params->vlan_id);
+
+	return 0;
+}
+
+/**
+ * dpsw_if_set_stp() - Function sets Spanning Tree Protocol (STP) state.
+ * @mc_io:	Pointer to MC portal's I/O object
+ * @cmd_flags:	Command flags; one or more of 'MC_CMD_FLAG_'
+ * @token:	Token of DPSW object
+ * @if_id:	Interface Identifier
+ * @cfg:	STP State configuration parameters
+ *
+ * The following STP states are supported -
+ * blocking, listening, learning, forwarding and disabled.
+ *
+ * Return:	Completion status. '0' on Success; Error code otherwise.
+ */
+int dpsw_if_set_stp(struct fsl_mc_io *mc_io, u32 cmd_flags, u16 token, u16 if_id,
+		    const struct dpsw_stp_cfg *cfg)
+{
+	struct fsl_mc_command cmd = { 0 };
+	struct dpsw_cmd_if_set_stp *cmd_params;
+
+	/* prepare command */
+	cmd.header = mc_encode_cmd_header(DPSW_CMDID_IF_SET_STP,
+					  cmd_flags,
+					  token);
+	cmd_params = (struct dpsw_cmd_if_set_stp *)cmd.params;
+	cmd_params->if_id = cpu_to_le16(if_id);
+	cmd_params->vlan_id = cpu_to_le16(cfg->vlan_id);
+	dpsw_set_field(cmd_params->state, STATE, cfg->state);
+
+	/* send command to mc*/
+	return mc_send_command(mc_io, &cmd);
+}
+
+/**
+ * dpsw_if_get_counter() - Get specific counter of particular interface
+ * @mc_io:	Pointer to MC portal's I/O object
+ * @cmd_flags:	Command flags; one or more of 'MC_CMD_FLAG_'
+ * @token:	Token of DPSW object
+ * @if_id:	Interface Identifier
+ * @type:	Counter type
+ * @counter:	return value
+ *
+ * Return:	Completion status. '0' on Success; Error code otherwise.
+ */
+int dpsw_if_get_counter(struct fsl_mc_io *mc_io, u32 cmd_flags, u16 token,
+			u16 if_id, enum dpsw_counter type, u64 *counter)
+{
+	struct fsl_mc_command cmd = { 0 };
+	struct dpsw_cmd_if_get_counter *cmd_params;
+	struct dpsw_rsp_if_get_counter *rsp_params;
+	int err;
+
+	/* prepare command */
+	cmd.header = mc_encode_cmd_header(DPSW_CMDID_IF_GET_COUNTER,
+					  cmd_flags,
+					  token);
+	cmd_params = (struct dpsw_cmd_if_get_counter *)cmd.params;
+	cmd_params->if_id = cpu_to_le16(if_id);
+	dpsw_set_field(cmd_params->type, COUNTER_TYPE, type);
+
+	/* send command to mc*/
+	err = mc_send_command(mc_io, &cmd);
+	if (err)
+		return err;
+
+	/* retrieve response parameters */
+	rsp_params = (struct dpsw_rsp_if_get_counter *)cmd.params;
+	*counter = le64_to_cpu(rsp_params->counter);
+
+	return 0;
+}
+
+/**
+ * dpsw_if_enable() - Enable Interface
+ * @mc_io:	Pointer to MC portal's I/O object
+ * @cmd_flags:	Command flags; one or more of 'MC_CMD_FLAG_'
+ * @token:	Token of DPSW object
+ * @if_id:	Interface Identifier
+ *
+ * Return:	Completion status. '0' on Success; Error code otherwise.
+ */
+int dpsw_if_enable(struct fsl_mc_io *mc_io, u32 cmd_flags, u16 token, u16 if_id)
+{
+	struct fsl_mc_command cmd = { 0 };
+	struct dpsw_cmd_if *cmd_params;
+
+	/* prepare command */
+	cmd.header = mc_encode_cmd_header(DPSW_CMDID_IF_ENABLE,
+					  cmd_flags,
+					  token);
+	cmd_params = (struct dpsw_cmd_if *)cmd.params;
+	cmd_params->if_id = cpu_to_le16(if_id);
+
+	/* send command to mc*/
+	return mc_send_command(mc_io, &cmd);
+}
+
+/**
+ * dpsw_if_disable() - Disable Interface
+ * @mc_io:	Pointer to MC portal's I/O object
+ * @cmd_flags:	Command flags; one or more of 'MC_CMD_FLAG_'
+ * @token:	Token of DPSW object
+ * @if_id:	Interface Identifier
+ *
+ * Return:	Completion status. '0' on Success; Error code otherwise.
+ */
+int dpsw_if_disable(struct fsl_mc_io *mc_io, u32 cmd_flags, u16 token, u16 if_id)
+{
+	struct fsl_mc_command cmd = { 0 };
+	struct dpsw_cmd_if *cmd_params;
+
+	/* prepare command */
+	cmd.header = mc_encode_cmd_header(DPSW_CMDID_IF_DISABLE,
+					  cmd_flags,
+					  token);
+	cmd_params = (struct dpsw_cmd_if *)cmd.params;
+	cmd_params->if_id = cpu_to_le16(if_id);
+
+	/* send command to mc*/
+	return mc_send_command(mc_io, &cmd);
+}
+
+/**
+ * dpsw_if_get_attributes() - Function obtains attributes of interface
+ * @mc_io:	Pointer to MC portal's I/O object
+ * @cmd_flags:	Command flags; one or more of 'MC_CMD_FLAG_'
+ * @token:	Token of DPSW object
+ * @if_id:	Interface Identifier
+ * @attr:	Returned interface attributes
+ *
+ * Return:	Completion status. '0' on Success; Error code otherwise.
+ */
+int dpsw_if_get_attributes(struct fsl_mc_io *mc_io, u32 cmd_flags, u16 token,
+			   u16 if_id, struct dpsw_if_attr *attr)
+{
+	struct dpsw_rsp_if_get_attr *rsp_params;
+	struct fsl_mc_command cmd = { 0 };
+	struct dpsw_cmd_if *cmd_params;
+	int err;
+
+	cmd.header = mc_encode_cmd_header(DPSW_CMDID_IF_GET_ATTR, cmd_flags,
+					  token);
+	cmd_params = (struct dpsw_cmd_if *)cmd.params;
+	cmd_params->if_id = cpu_to_le16(if_id);
+
+	err = mc_send_command(mc_io, &cmd);
+	if (err)
+		return err;
+
+	rsp_params = (struct dpsw_rsp_if_get_attr *)cmd.params;
+	attr->num_tcs = rsp_params->num_tcs;
+	attr->rate = le32_to_cpu(rsp_params->rate);
+	attr->options = le32_to_cpu(rsp_params->options);
+	attr->qdid = le16_to_cpu(rsp_params->qdid);
+	attr->enabled = dpsw_get_field(rsp_params->conf, ENABLED);
+	attr->accept_all_vlan = dpsw_get_field(rsp_params->conf,
+					       ACCEPT_ALL_VLAN);
+	attr->admit_untagged = dpsw_get_field(rsp_params->conf,
+					      ADMIT_UNTAGGED);
+
+	return 0;
+}
+
+/**
+ * dpsw_if_set_max_frame_length() - Set Maximum Receive frame length.
+ * @mc_io:		Pointer to MC portal's I/O object
+ * @cmd_flags:		Command flags; one or more of 'MC_CMD_FLAG_'
+ * @token:		Token of DPSW object
+ * @if_id:		Interface Identifier
+ * @frame_length:	Maximum Frame Length
+ *
+ * Return:	Completion status. '0' on Success; Error code otherwise.
+ */
+int dpsw_if_set_max_frame_length(struct fsl_mc_io *mc_io, u32 cmd_flags, u16 token,
+				 u16 if_id, u16 frame_length)
+{
+	struct fsl_mc_command cmd = { 0 };
+	struct dpsw_cmd_if_set_max_frame_length *cmd_params;
+
+	/* prepare command */
+	cmd.header = mc_encode_cmd_header(DPSW_CMDID_IF_SET_MAX_FRAME_LENGTH,
+					  cmd_flags,
+					  token);
+	cmd_params = (struct dpsw_cmd_if_set_max_frame_length *)cmd.params;
+	cmd_params->if_id = cpu_to_le16(if_id);
+	cmd_params->frame_length = cpu_to_le16(frame_length);
+
+	/* send command to mc*/
+	return mc_send_command(mc_io, &cmd);
+}
+
+/**
+ * dpsw_vlan_add() - Adding new VLAN to DPSW.
+ * @mc_io:	Pointer to MC portal's I/O object
+ * @cmd_flags:	Command flags; one or more of 'MC_CMD_FLAG_'
+ * @token:	Token of DPSW object
+ * @vlan_id:	VLAN Identifier
+ * @cfg:	VLAN configuration
+ *
+ * Only VLAN ID and FDB ID are required parameters here.
+ * 12 bit VLAN ID is defined in IEEE802.1Q.
+ * Adding a duplicate VLAN ID is not allowed.
+ * FDB ID can be shared across multiple VLANs. Shared learning
+ * is obtained by calling dpsw_vlan_add for multiple VLAN IDs
+ * with same fdb_id
+ *
+ * Return:	Completion status. '0' on Success; Error code otherwise.
+ */
+int dpsw_vlan_add(struct fsl_mc_io *mc_io, u32 cmd_flags, u16 token,
+		  u16 vlan_id, const struct dpsw_vlan_cfg *cfg)
+{
+	struct fsl_mc_command cmd = { 0 };
+	struct dpsw_vlan_add *cmd_params;
+
+	/* prepare command */
+	cmd.header = mc_encode_cmd_header(DPSW_CMDID_VLAN_ADD,
+					  cmd_flags,
+					  token);
+	cmd_params = (struct dpsw_vlan_add *)cmd.params;
+	cmd_params->fdb_id = cpu_to_le16(cfg->fdb_id);
+	cmd_params->vlan_id = cpu_to_le16(vlan_id);
+
+	/* send command to mc*/
+	return mc_send_command(mc_io, &cmd);
+}
+
+/**
+ * dpsw_vlan_add_if() - Adding a set of interfaces to an existing VLAN.
+ * @mc_io:	Pointer to MC portal's I/O object
+ * @cmd_flags:	Command flags; one or more of 'MC_CMD_FLAG_'
+ * @token:	Token of DPSW object
+ * @vlan_id:	VLAN Identifier
+ * @cfg:	Set of interfaces to add
+ *
+ * It adds only interfaces not belonging to this VLAN yet,
+ * otherwise an error is generated and an entire command is
+ * ignored. This function can be called numerous times always
+ * providing required interfaces delta.
+ *
+ * Return:	Completion status. '0' on Success; Error code otherwise.
+ */
+int dpsw_vlan_add_if(struct fsl_mc_io *mc_io, u32 cmd_flags, u16 token,
+		     u16 vlan_id, const struct dpsw_vlan_if_cfg *cfg)
+{
+	struct dpsw_cmd_vlan_add_if *cmd_params;
+	struct fsl_mc_command cmd = { 0 };
+
+	/* prepare command */
+	cmd.header = mc_encode_cmd_header(DPSW_CMDID_VLAN_ADD_IF,
+					  cmd_flags,
+					  token);
+	cmd_params = (struct dpsw_cmd_vlan_add_if *)cmd.params;
+	cmd_params->vlan_id = cpu_to_le16(vlan_id);
+	cmd_params->options = cpu_to_le16(cfg->options);
+	cmd_params->fdb_id = cpu_to_le16(cfg->fdb_id);
+	build_if_id_bitmap(&cmd_params->if_id, cfg->if_id, cfg->num_ifs);
+
+	/* send command to mc*/
+	return mc_send_command(mc_io, &cmd);
+}
+
+/**
+ * dpsw_vlan_add_if_untagged() - Defining a set of interfaces that should be
+ *				transmitted as untagged.
+ * @mc_io:	Pointer to MC portal's I/O object
+ * @cmd_flags:	Command flags; one or more of 'MC_CMD_FLAG_'
+ * @token:	Token of DPSW object
+ * @vlan_id:	VLAN Identifier
+ * @cfg:	Set of interfaces that should be transmitted as untagged
+ *
+ * These interfaces should already belong to this VLAN.
+ * By default all interfaces are transmitted as tagged.
+ * Providing un-existing interface or untagged interface that is
+ * configured untagged already generates an error and the entire
+ * command is ignored.
+ *
+ * Return:	Completion status. '0' on Success; Error code otherwise.
+ */
+int dpsw_vlan_add_if_untagged(struct fsl_mc_io *mc_io, u32 cmd_flags, u16 token,
+			      u16 vlan_id, const struct dpsw_vlan_if_cfg *cfg)
+{
+	struct fsl_mc_command cmd = { 0 };
+	struct dpsw_cmd_vlan_manage_if *cmd_params;
+
+	/* prepare command */
+	cmd.header = mc_encode_cmd_header(DPSW_CMDID_VLAN_ADD_IF_UNTAGGED,
+					  cmd_flags,
+					  token);
+	cmd_params = (struct dpsw_cmd_vlan_manage_if *)cmd.params;
+	cmd_params->vlan_id = cpu_to_le16(vlan_id);
+	build_if_id_bitmap(&cmd_params->if_id, cfg->if_id, cfg->num_ifs);
+
+	/* send command to mc*/
+	return mc_send_command(mc_io, &cmd);
+}
+
+/**
+ * dpsw_vlan_remove_if() - Remove interfaces from an existing VLAN.
+ * @mc_io:	Pointer to MC portal's I/O object
+ * @cmd_flags:	Command flags; one or more of 'MC_CMD_FLAG_'
+ * @token:	Token of DPSW object
+ * @vlan_id:	VLAN Identifier
+ * @cfg:	Set of interfaces that should be removed
+ *
+ * Interfaces must belong to this VLAN, otherwise an error
+ * is returned and an the command is ignored
+ *
+ * Return:	Completion status. '0' on Success; Error code otherwise.
+ */
+int dpsw_vlan_remove_if(struct fsl_mc_io *mc_io, u32 cmd_flags, u16 token,
+			u16 vlan_id, const struct dpsw_vlan_if_cfg *cfg)
+{
+	struct fsl_mc_command cmd = { 0 };
+	struct dpsw_cmd_vlan_manage_if *cmd_params;
+
+	/* prepare command */
+	cmd.header = mc_encode_cmd_header(DPSW_CMDID_VLAN_REMOVE_IF,
+					  cmd_flags,
+					  token);
+	cmd_params = (struct dpsw_cmd_vlan_manage_if *)cmd.params;
+	cmd_params->vlan_id = cpu_to_le16(vlan_id);
+	build_if_id_bitmap(&cmd_params->if_id, cfg->if_id, cfg->num_ifs);
+
+	/* send command to mc*/
+	return mc_send_command(mc_io, &cmd);
+}
+
+/**
+ * dpsw_vlan_remove_if_untagged() - Define a set of interfaces that should be
+ *		converted from transmitted as untagged to transmit as tagged.
+ * @mc_io:	Pointer to MC portal's I/O object
+ * @cmd_flags:	Command flags; one or more of 'MC_CMD_FLAG_'
+ * @token:	Token of DPSW object
+ * @vlan_id:	VLAN Identifier
+ * @cfg:	Set of interfaces that should be removed
+ *
+ * Interfaces provided by API have to belong to this VLAN and
+ * configured untagged, otherwise an error is returned and the
+ * command is ignored
+ *
+ * Return:	Completion status. '0' on Success; Error code otherwise.
+ */
+int dpsw_vlan_remove_if_untagged(struct fsl_mc_io *mc_io, u32 cmd_flags, u16 token,
+				 u16 vlan_id, const struct dpsw_vlan_if_cfg *cfg)
+{
+	struct fsl_mc_command cmd = { 0 };
+	struct dpsw_cmd_vlan_manage_if *cmd_params;
+
+	/* prepare command */
+	cmd.header = mc_encode_cmd_header(DPSW_CMDID_VLAN_REMOVE_IF_UNTAGGED,
+					  cmd_flags,
+					  token);
+	cmd_params = (struct dpsw_cmd_vlan_manage_if *)cmd.params;
+	cmd_params->vlan_id = cpu_to_le16(vlan_id);
+	build_if_id_bitmap(&cmd_params->if_id, cfg->if_id, cfg->num_ifs);
+
+	/* send command to mc*/
+	return mc_send_command(mc_io, &cmd);
+}
+
+/**
+ * dpsw_vlan_remove() - Remove an entire VLAN
+ * @mc_io:	Pointer to MC portal's I/O object
+ * @cmd_flags:	Command flags; one or more of 'MC_CMD_FLAG_'
+ * @token:	Token of DPSW object
+ * @vlan_id:	VLAN Identifier
+ *
+ * Return:	Completion status. '0' on Success; Error code otherwise.
+ */
+int dpsw_vlan_remove(struct fsl_mc_io *mc_io, u32 cmd_flags, u16 token,
+		     u16 vlan_id)
+{
+	struct fsl_mc_command cmd = { 0 };
+	struct dpsw_cmd_vlan_remove *cmd_params;
+
+	/* prepare command */
+	cmd.header = mc_encode_cmd_header(DPSW_CMDID_VLAN_REMOVE,
+					  cmd_flags,
+					  token);
+	cmd_params = (struct dpsw_cmd_vlan_remove *)cmd.params;
+	cmd_params->vlan_id = cpu_to_le16(vlan_id);
+
+	/* send command to mc*/
+	return mc_send_command(mc_io, &cmd);
+}
+
+/**
+ * dpsw_fdb_add() - Add FDB to switch and Returns handle to FDB table for
+ *		the reference
+ * @mc_io:	Pointer to MC portal's I/O object
+ * @cmd_flags:	Command flags; one or more of 'MC_CMD_FLAG_'
+ * @token:	Token of DPSW object
+ * @fdb_id:	Returned Forwarding Database Identifier
+ * @cfg:	FDB Configuration
+ *
+ * Return:	Completion status. '0' on Success; Error code otherwise.
+ */
+int dpsw_fdb_add(struct fsl_mc_io *mc_io, u32 cmd_flags, u16 token, u16 *fdb_id,
+		 const struct dpsw_fdb_cfg *cfg)
+{
+	struct dpsw_cmd_fdb_add *cmd_params;
+	struct dpsw_rsp_fdb_add *rsp_params;
+	struct fsl_mc_command cmd = { 0 };
+	int err;
+
+	cmd.header = mc_encode_cmd_header(DPSW_CMDID_FDB_ADD,
+					  cmd_flags,
+					  token);
+	cmd_params = (struct dpsw_cmd_fdb_add *)cmd.params;
+	cmd_params->fdb_ageing_time = cpu_to_le16(cfg->fdb_ageing_time);
+	cmd_params->num_fdb_entries = cpu_to_le16(cfg->num_fdb_entries);
+
+	err = mc_send_command(mc_io, &cmd);
+	if (err)
+		return err;
+
+	rsp_params = (struct dpsw_rsp_fdb_add *)cmd.params;
+	*fdb_id = le16_to_cpu(rsp_params->fdb_id);
+
+	return 0;
+}
+
+/**
+ * dpsw_fdb_remove() - Remove FDB from switch
+ * @mc_io:	Pointer to MC portal's I/O object
+ * @cmd_flags:	Command flags; one or more of 'MC_CMD_FLAG_'
+ * @token:	Token of DPSW object
+ * @fdb_id:	Forwarding Database Identifier
+ *
+ * Return:	Completion status. '0' on Success; Error code otherwise.
+ */
+int dpsw_fdb_remove(struct fsl_mc_io *mc_io, u32 cmd_flags, u16 token, u16 fdb_id)
+{
+	struct dpsw_cmd_fdb_remove *cmd_params;
+	struct fsl_mc_command cmd = { 0 };
+
+	/* prepare command */
+	cmd.header = mc_encode_cmd_header(DPSW_CMDID_FDB_REMOVE,
+					  cmd_flags,
+					  token);
+	cmd_params = (struct dpsw_cmd_fdb_remove *)cmd.params;
+	cmd_params->fdb_id = cpu_to_le16(fdb_id);
+
+	return mc_send_command(mc_io, &cmd);
+}
+
+/**
+ * dpsw_fdb_add_unicast() - Function adds an unicast entry into MAC lookup table
+ * @mc_io:	Pointer to MC portal's I/O object
+ * @cmd_flags:	Command flags; one or more of 'MC_CMD_FLAG_'
+ * @token:	Token of DPSW object
+ * @fdb_id:	Forwarding Database Identifier
+ * @cfg:	Unicast entry configuration
+ *
+ * Return:	Completion status. '0' on Success; Error code otherwise.
+ */
+int dpsw_fdb_add_unicast(struct fsl_mc_io *mc_io, u32 cmd_flags, u16 token,
+			 u16 fdb_id, const struct dpsw_fdb_unicast_cfg *cfg)
+{
+	struct fsl_mc_command cmd = { 0 };
+	struct dpsw_cmd_fdb_unicast_op *cmd_params;
+	int i;
+
+	/* prepare command */
+	cmd.header = mc_encode_cmd_header(DPSW_CMDID_FDB_ADD_UNICAST,
+					  cmd_flags,
+					  token);
+	cmd_params = (struct dpsw_cmd_fdb_unicast_op *)cmd.params;
+	cmd_params->fdb_id = cpu_to_le16(fdb_id);
+	cmd_params->if_egress = cpu_to_le16(cfg->if_egress);
+	for (i = 0; i < 6; i++)
+		cmd_params->mac_addr[i] = cfg->mac_addr[5 - i];
+	dpsw_set_field(cmd_params->type, ENTRY_TYPE, cfg->type);
+
+	/* send command to mc*/
+	return mc_send_command(mc_io, &cmd);
+}
+
+/**
+ * dpsw_fdb_dump() - Dump the content of FDB table into memory.
+ * @mc_io:	Pointer to MC portal's I/O object
+ * @cmd_flags:	Command flags; one or more of 'MC_CMD_FLAG_'
+ * @token:	Token of DPSW object
+ * @fdb_id:	Forwarding Database Identifier
+ * @iova_addr:	Data will be stored here as an array of struct fdb_dump_entry
+ * @iova_size:	Memory size allocated at iova_addr
+ * @num_entries:Number of entries written at iova_addr
+ *
+ * Return:	Completion status. '0' on Success; Error code otherwise.
+ *
+ * The memory allocated at iova_addr must be initialized with zero before
+ * command execution. If the FDB table does not fit into memory MC will stop
+ * after the memory is filled up.
+ * The struct fdb_dump_entry array must be parsed until the end of memory
+ * area or until an entry with mac_addr set to zero is found.
+ */
+int dpsw_fdb_dump(struct fsl_mc_io *mc_io, u32 cmd_flags, u16 token, u16 fdb_id,
+		  u64 iova_addr, u32 iova_size, u16 *num_entries)
+{
+	struct dpsw_cmd_fdb_dump *cmd_params;
+	struct dpsw_rsp_fdb_dump *rsp_params;
+	struct fsl_mc_command cmd = { 0 };
+	int err;
+
+	/* prepare command */
+	cmd.header = mc_encode_cmd_header(DPSW_CMDID_FDB_DUMP,
+					  cmd_flags,
+					  token);
+	cmd_params = (struct dpsw_cmd_fdb_dump *)cmd.params;
+	cmd_params->fdb_id = cpu_to_le16(fdb_id);
+	cmd_params->iova_addr = cpu_to_le64(iova_addr);
+	cmd_params->iova_size = cpu_to_le32(iova_size);
+
+	/* send command to mc */
+	err = mc_send_command(mc_io, &cmd);
+	if (err)
+		return err;
+
+	rsp_params = (struct dpsw_rsp_fdb_dump *)cmd.params;
+	*num_entries = le16_to_cpu(rsp_params->num_entries);
+
+	return 0;
+}
+
+/**
+ * dpsw_fdb_remove_unicast() - removes an entry from MAC lookup table
+ * @mc_io:	Pointer to MC portal's I/O object
+ * @cmd_flags:	Command flags; one or more of 'MC_CMD_FLAG_'
+ * @token:	Token of DPSW object
+ * @fdb_id:	Forwarding Database Identifier
+ * @cfg:	Unicast entry configuration
+ *
+ * Return:	Completion status. '0' on Success; Error code otherwise.
+ */
+int dpsw_fdb_remove_unicast(struct fsl_mc_io *mc_io, u32 cmd_flags, u16 token,
+			    u16 fdb_id, const struct dpsw_fdb_unicast_cfg *cfg)
+{
+	struct fsl_mc_command cmd = { 0 };
+	struct dpsw_cmd_fdb_unicast_op *cmd_params;
+	int i;
+
+	/* prepare command */
+	cmd.header = mc_encode_cmd_header(DPSW_CMDID_FDB_REMOVE_UNICAST,
+					  cmd_flags,
+					  token);
+	cmd_params = (struct dpsw_cmd_fdb_unicast_op *)cmd.params;
+	cmd_params->fdb_id = cpu_to_le16(fdb_id);
+	for (i = 0; i < 6; i++)
+		cmd_params->mac_addr[i] = cfg->mac_addr[5 - i];
+	cmd_params->if_egress = cpu_to_le16(cfg->if_egress);
+	dpsw_set_field(cmd_params->type, ENTRY_TYPE, cfg->type);
+
+	/* send command to mc*/
+	return mc_send_command(mc_io, &cmd);
+}
+
+/**
+ * dpsw_fdb_add_multicast() - Add a set of egress interfaces to multi-cast group
+ * @mc_io:	Pointer to MC portal's I/O object
+ * @cmd_flags:	Command flags; one or more of 'MC_CMD_FLAG_'
+ * @token:	Token of DPSW object
+ * @fdb_id:	Forwarding Database Identifier
+ * @cfg:	Multicast entry configuration
+ *
+ * If group doesn't exist, it will be created.
+ * It adds only interfaces not belonging to this multicast group
+ * yet, otherwise error will be generated and the command is
+ * ignored.
+ * This function may be called numerous times always providing
+ * required interfaces delta.
+ *
+ * Return:	Completion status. '0' on Success; Error code otherwise.
+ */
+int dpsw_fdb_add_multicast(struct fsl_mc_io *mc_io, u32 cmd_flags, u16 token,
+			   u16 fdb_id, const struct dpsw_fdb_multicast_cfg *cfg)
+{
+	struct fsl_mc_command cmd = { 0 };
+	struct dpsw_cmd_fdb_multicast_op *cmd_params;
+	int i;
+
+	/* prepare command */
+	cmd.header = mc_encode_cmd_header(DPSW_CMDID_FDB_ADD_MULTICAST,
+					  cmd_flags,
+					  token);
+	cmd_params = (struct dpsw_cmd_fdb_multicast_op *)cmd.params;
+	cmd_params->fdb_id = cpu_to_le16(fdb_id);
+	cmd_params->num_ifs = cpu_to_le16(cfg->num_ifs);
+	dpsw_set_field(cmd_params->type, ENTRY_TYPE, cfg->type);
+	build_if_id_bitmap(&cmd_params->if_id, cfg->if_id, cfg->num_ifs);
+	for (i = 0; i < 6; i++)
+		cmd_params->mac_addr[i] = cfg->mac_addr[5 - i];
+
+	/* send command to mc*/
+	return mc_send_command(mc_io, &cmd);
+}
+
+/**
+ * dpsw_fdb_remove_multicast() - Removing interfaces from an existing multicast
+ *				group.
+ * @mc_io:	Pointer to MC portal's I/O object
+ * @cmd_flags:	Command flags; one or more of 'MC_CMD_FLAG_'
+ * @token:	Token of DPSW object
+ * @fdb_id:	Forwarding Database Identifier
+ * @cfg:	Multicast entry configuration
+ *
+ * Interfaces provided by this API have to exist in the group,
+ * otherwise an error will be returned and an entire command
+ * ignored. If there is no interface left in the group,
+ * an entire group is deleted
+ *
+ * Return:	Completion status. '0' on Success; Error code otherwise.
+ */
+int dpsw_fdb_remove_multicast(struct fsl_mc_io *mc_io, u32 cmd_flags, u16 token,
+			      u16 fdb_id, const struct dpsw_fdb_multicast_cfg *cfg)
+{
+	struct fsl_mc_command cmd = { 0 };
+	struct dpsw_cmd_fdb_multicast_op *cmd_params;
+	int i;
+
+	/* prepare command */
+	cmd.header = mc_encode_cmd_header(DPSW_CMDID_FDB_REMOVE_MULTICAST,
+					  cmd_flags,
+					  token);
+	cmd_params = (struct dpsw_cmd_fdb_multicast_op *)cmd.params;
+	cmd_params->fdb_id = cpu_to_le16(fdb_id);
+	cmd_params->num_ifs = cpu_to_le16(cfg->num_ifs);
+	dpsw_set_field(cmd_params->type, ENTRY_TYPE, cfg->type);
+	build_if_id_bitmap(&cmd_params->if_id, cfg->if_id, cfg->num_ifs);
+	for (i = 0; i < 6; i++)
+		cmd_params->mac_addr[i] = cfg->mac_addr[5 - i];
+
+	/* send command to mc*/
+	return mc_send_command(mc_io, &cmd);
+}
+
+/**
+ * dpsw_ctrl_if_get_attributes() - Obtain control interface attributes
+ * @mc_io:	Pointer to MC portal's I/O object
+ * @cmd_flags:	Command flags; one or more of 'MC_CMD_FLAG_'
+ * @token:	Token of DPSW object
+ * @attr:	Returned control interface attributes
+ *
+ * Return:	'0' on Success; Error code otherwise.
+ */
+int dpsw_ctrl_if_get_attributes(struct fsl_mc_io *mc_io, u32 cmd_flags,
+				u16 token, struct dpsw_ctrl_if_attr *attr)
+{
+	struct dpsw_rsp_ctrl_if_get_attr *rsp_params;
+	struct fsl_mc_command cmd = { 0 };
+	int err;
+
+	cmd.header = mc_encode_cmd_header(DPSW_CMDID_CTRL_IF_GET_ATTR,
+					  cmd_flags, token);
+
+	err = mc_send_command(mc_io, &cmd);
+	if (err)
+		return err;
+
+	rsp_params = (struct dpsw_rsp_ctrl_if_get_attr *)cmd.params;
+	attr->rx_fqid = le32_to_cpu(rsp_params->rx_fqid);
+	attr->rx_err_fqid = le32_to_cpu(rsp_params->rx_err_fqid);
+	attr->tx_err_conf_fqid = le32_to_cpu(rsp_params->tx_err_conf_fqid);
+
+	return 0;
+}
+
+/**
+ * dpsw_ctrl_if_set_pools() - Set control interface buffer pools
+ * @mc_io:	Pointer to MC portal's I/O object
+ * @cmd_flags:	Command flags; one or more of 'MC_CMD_FLAG_'
+ * @token:	Token of DPSW object
+ * @cfg:	Buffer pools configuration
+ *
+ * Return:	'0' on Success; Error code otherwise.
+ */
+int dpsw_ctrl_if_set_pools(struct fsl_mc_io *mc_io, u32 cmd_flags, u16 token,
+			   const struct dpsw_ctrl_if_pools_cfg *cfg)
+{
+	struct dpsw_cmd_ctrl_if_set_pools *cmd_params;
+	struct fsl_mc_command cmd = { 0 };
+	int i;
+
+	cmd.header = mc_encode_cmd_header(DPSW_CMDID_CTRL_IF_SET_POOLS,
+					  cmd_flags, token);
+	cmd_params = (struct dpsw_cmd_ctrl_if_set_pools *)cmd.params;
+	cmd_params->num_dpbp = cfg->num_dpbp;
+	for (i = 0; i < DPSW_MAX_DPBP; i++) {
+		cmd_params->dpbp_id[i] = cpu_to_le32(cfg->pools[i].dpbp_id);
+		cmd_params->buffer_size[i] =
+			cpu_to_le16(cfg->pools[i].buffer_size);
+		cmd_params->backup_pool_mask |=
+			DPSW_BACKUP_POOL(cfg->pools[i].backup_pool, i);
+	}
+
+	return mc_send_command(mc_io, &cmd);
+}
+
+/**
+ * dpsw_ctrl_if_set_queue() - Set Rx queue configuration
+ * @mc_io:	Pointer to MC portal's I/O object
+ * @cmd_flags:	Command flags; one or more of 'MC_CMD_FLAG_'
+ * @token:	Token of dpsw object
+ * @qtype:	dpsw_queue_type of the targeted queue
+ * @cfg:	Rx queue configuration
+ *
+ * Return:	'0' on Success; Error code otherwise.
+ */
+int dpsw_ctrl_if_set_queue(struct fsl_mc_io *mc_io, u32 cmd_flags, u16 token,
+			   enum dpsw_queue_type qtype,
+			   const struct dpsw_ctrl_if_queue_cfg *cfg)
+{
+	struct dpsw_cmd_ctrl_if_set_queue *cmd_params;
+	struct fsl_mc_command cmd = { 0 };
+
+	cmd.header = mc_encode_cmd_header(DPSW_CMDID_CTRL_IF_SET_QUEUE,
+					  cmd_flags,
+					  token);
+	cmd_params = (struct dpsw_cmd_ctrl_if_set_queue *)cmd.params;
+	cmd_params->dest_id = cpu_to_le32(cfg->dest_cfg.dest_id);
+	cmd_params->dest_priority = cfg->dest_cfg.priority;
+	cmd_params->qtype = qtype;
+	cmd_params->user_ctx = cpu_to_le64(cfg->user_ctx);
+	cmd_params->options = cpu_to_le32(cfg->options);
+	dpsw_set_field(cmd_params->dest_type,
+		       DEST_TYPE,
+		       cfg->dest_cfg.dest_type);
+
+	return mc_send_command(mc_io, &cmd);
+}
+
+/**
+ * dpsw_get_api_version() - Get Data Path Switch API version
+ * @mc_io:	Pointer to MC portal's I/O object
+ * @cmd_flags:	Command flags; one or more of 'MC_CMD_FLAG_'
+ * @major_ver:	Major version of data path switch API
+ * @minor_ver:	Minor version of data path switch API
+ *
+ * Return:  '0' on Success; Error code otherwise.
+ */
+int dpsw_get_api_version(struct fsl_mc_io *mc_io, u32 cmd_flags,
+			 u16 *major_ver, u16 *minor_ver)
+{
+	struct fsl_mc_command cmd = { 0 };
+	struct dpsw_rsp_get_api_version *rsp_params;
+	int err;
+
+	cmd.header = mc_encode_cmd_header(DPSW_CMDID_GET_API_VERSION,
+					  cmd_flags,
+					  0);
+
+	err = mc_send_command(mc_io, &cmd);
+	if (err)
+		return err;
+
+	rsp_params = (struct dpsw_rsp_get_api_version *)cmd.params;
+	*major_ver = le16_to_cpu(rsp_params->version_major);
+	*minor_ver = le16_to_cpu(rsp_params->version_minor);
+
+	return 0;
+}
+
+/**
+ * dpsw_if_get_port_mac_addr() - Retrieve MAC address associated to the physical port
+ * @mc_io:	Pointer to MC portal's I/O object
+ * @cmd_flags:	Command flags; one or more of 'MC_CMD_FLAG_'
+ * @token:	Token of DPSW object
+ * @if_id:	Interface Identifier
+ * @mac_addr:	MAC address of the physical port, if any, otherwise 0
+ *
+ * Return:	Completion status. '0' on Success; Error code otherwise.
+ */
+int dpsw_if_get_port_mac_addr(struct fsl_mc_io *mc_io, u32 cmd_flags, u16 token,
+			      u16 if_id, u8 mac_addr[6])
+{
+	struct dpsw_rsp_if_get_mac_addr *rsp_params;
+	struct fsl_mc_command cmd = { 0 };
+	struct dpsw_cmd_if *cmd_params;
+	int err, i;
+
+	/* prepare command */
+	cmd.header = mc_encode_cmd_header(DPSW_CMDID_IF_GET_PORT_MAC_ADDR,
+					  cmd_flags,
+					  token);
+	cmd_params = (struct dpsw_cmd_if *)cmd.params;
+	cmd_params->if_id = cpu_to_le16(if_id);
+
+	/* send command to mc*/
+	err = mc_send_command(mc_io, &cmd);
+	if (err)
+		return err;
+
+	/* retrieve response parameters */
+	rsp_params = (struct dpsw_rsp_if_get_mac_addr *)cmd.params;
+	for (i = 0; i < 6; i++)
+		mac_addr[5 - i] = rsp_params->mac_addr[i];
+
+	return 0;
+}
+
+/**
+ * dpsw_ctrl_if_enable() - Enable control interface
+ * @mc_io:	Pointer to MC portal's I/O object
+ * @cmd_flags:	Command flags; one or more of 'MC_CMD_FLAG_'
+ * @token:	Token of DPSW object
+ *
+ * Return:	'0' on Success; Error code otherwise.
+ */
+int dpsw_ctrl_if_enable(struct fsl_mc_io *mc_io, u32 cmd_flags, u16 token)
+{
+	struct fsl_mc_command cmd = { 0 };
+
+	cmd.header = mc_encode_cmd_header(DPSW_CMDID_CTRL_IF_ENABLE, cmd_flags,
+					  token);
+
+	return mc_send_command(mc_io, &cmd);
+}
+
+/**
+ * dpsw_ctrl_if_disable() - Function disables control interface
+ * @mc_io:	Pointer to MC portal's I/O object
+ * @cmd_flags:	Command flags; one or more of 'MC_CMD_FLAG_'
+ * @token:	Token of DPSW object
+ *
+ * Return:	'0' on Success; Error code otherwise.
+ */
+int dpsw_ctrl_if_disable(struct fsl_mc_io *mc_io, u32 cmd_flags, u16 token)
+{
+	struct fsl_mc_command cmd = { 0 };
+
+	cmd.header = mc_encode_cmd_header(DPSW_CMDID_CTRL_IF_DISABLE,
+					  cmd_flags,
+					  token);
+
+	return mc_send_command(mc_io, &cmd);
+}
+
+/**
+ * dpsw_set_egress_flood() - Set egress parameters associated with an FDB ID
+ * @mc_io:	Pointer to MC portal's I/O object
+ * @cmd_flags:	Command flags; one or more of 'MC_CMD_FLAG_'
+ * @token:	Token of DPSW object
+ * @cfg:	Egress flooding configuration
+ *
+ * Return:	'0' on Success; Error code otherwise.
+ */
+int dpsw_set_egress_flood(struct fsl_mc_io *mc_io, u32 cmd_flags, u16 token,
+			  const struct dpsw_egress_flood_cfg *cfg)
+{
+	struct dpsw_cmd_set_egress_flood *cmd_params;
+	struct fsl_mc_command cmd = { 0 };
+
+	cmd.header = mc_encode_cmd_header(DPSW_CMDID_SET_EGRESS_FLOOD, cmd_flags, token);
+	cmd_params = (struct dpsw_cmd_set_egress_flood *)cmd.params;
+	cmd_params->fdb_id = cpu_to_le16(cfg->fdb_id);
+	cmd_params->flood_type = cfg->flood_type;
+	build_if_id_bitmap(&cmd_params->if_id, cfg->if_id, cfg->num_ifs);
+
+	return mc_send_command(mc_io, &cmd);
+}
+
+/**
+ * dpsw_if_set_learning_mode() - Configure the learning mode on an interface.
+ * If this API is used, it will take precedence over the FDB configuration.
+ * @mc_io:	Pointer to MC portal's I/O object
+ * @cmd_flags:	Command flags; one or more of 'MC_CMD_FLAG_'
+ * @token:	Token of DPSW object
+ * @if_id:	InterfaceID
+ * @mode:	Learning mode
+ *
+ * Return:	Completion status. '0' on Success; Error code otherwise.
+ */
+int dpsw_if_set_learning_mode(struct fsl_mc_io *mc_io, u32 cmd_flags, u16 token,
+			      u16 if_id, enum dpsw_learning_mode mode)
+{
+	struct dpsw_cmd_if_set_learning_mode *cmd_params;
+	struct fsl_mc_command cmd = { 0 };
+
+	cmd.header = mc_encode_cmd_header(DPSW_CMDID_IF_SET_LEARNING_MODE,
+					  cmd_flags,
+					  token);
+	cmd_params = (struct dpsw_cmd_if_set_learning_mode *)cmd.params;
+	cmd_params->if_id = cpu_to_le16(if_id);
+	dpsw_set_field(cmd_params->mode, LEARNING_MODE, mode);
+
+	return mc_send_command(mc_io, &cmd);
+}
+
+/**
+ * dpsw_acl_add() - Create an ACL table
+ * @mc_io:	Pointer to MC portal's I/O object
+ * @cmd_flags:	Command flags; one or more of 'MC_CMD_FLAG_'
+ * @token:	Token of DPSW object
+ * @acl_id:	Returned ACL ID, for future references
+ * @cfg:	ACL configuration
+ *
+ * Create Access Control List table. Multiple ACLs can be created and
+ * co-exist in L2 switch
+ *
+ * Return:	'0' on Success; Error code otherwise.
+ */
+int dpsw_acl_add(struct fsl_mc_io *mc_io, u32 cmd_flags, u16 token, u16 *acl_id,
+		 const struct dpsw_acl_cfg *cfg)
+{
+	struct dpsw_cmd_acl_add *cmd_params;
+	struct dpsw_rsp_acl_add *rsp_params;
+	struct fsl_mc_command cmd = { 0 };
+	int err;
+
+	cmd.header = mc_encode_cmd_header(DPSW_CMDID_ACL_ADD, cmd_flags, token);
+	cmd_params = (struct dpsw_cmd_acl_add *)cmd.params;
+	cmd_params->max_entries = cpu_to_le16(cfg->max_entries);
+
+	err = mc_send_command(mc_io, &cmd);
+	if (err)
+		return err;
+
+	rsp_params = (struct dpsw_rsp_acl_add *)cmd.params;
+	*acl_id = le16_to_cpu(rsp_params->acl_id);
+
+	return 0;
+}
+
+/**
+ * dpsw_acl_remove() - Remove an ACL table from L2 switch.
+ * @mc_io:	Pointer to MC portal's I/O object
+ * @cmd_flags:	Command flags; one or more of 'MC_CMD_FLAG_'
+ * @token:	Token of DPSW object
+ * @acl_id:	ACL ID
+ *
+ * Return:	'0' on Success; Error code otherwise.
+ */
+int dpsw_acl_remove(struct fsl_mc_io *mc_io, u32 cmd_flags, u16 token,
+		    u16 acl_id)
+{
+	struct dpsw_cmd_acl_remove *cmd_params;
+	struct fsl_mc_command cmd = { 0 };
+
+	cmd.header = mc_encode_cmd_header(DPSW_CMDID_ACL_REMOVE, cmd_flags,
+					  token);
+	cmd_params = (struct dpsw_cmd_acl_remove *)cmd.params;
+	cmd_params->acl_id = cpu_to_le16(acl_id);
+
+	return mc_send_command(mc_io, &cmd);
+}
+
+/**
+ * dpsw_acl_add_if() - Associate interface/interfaces with an ACL table.
+ * @mc_io:	Pointer to MC portal's I/O object
+ * @cmd_flags:	Command flags; one or more of 'MC_CMD_FLAG_'
+ * @token:	Token of DPSW object
+ * @acl_id:	ACL ID
+ * @cfg:	Interfaces list
+ *
+ * Return:	'0' on Success; Error code otherwise.
+ */
+int dpsw_acl_add_if(struct fsl_mc_io *mc_io, u32 cmd_flags, u16 token,
+		    u16 acl_id, const struct dpsw_acl_if_cfg *cfg)
+{
+	struct dpsw_cmd_acl_if *cmd_params;
+	struct fsl_mc_command cmd = { 0 };
+
+	cmd.header = mc_encode_cmd_header(DPSW_CMDID_ACL_ADD_IF, cmd_flags,
+					  token);
+	cmd_params = (struct dpsw_cmd_acl_if *)cmd.params;
+	cmd_params->acl_id = cpu_to_le16(acl_id);
+	cmd_params->num_ifs = cpu_to_le16(cfg->num_ifs);
+	build_if_id_bitmap(&cmd_params->if_id, cfg->if_id, cfg->num_ifs);
+
+	return mc_send_command(mc_io, &cmd);
+}
+
+/**
+ * dpsw_acl_remove_if() - De-associate interface/interfaces from an ACL table
+ * @mc_io:	Pointer to MC portal's I/O object
+ * @cmd_flags:	Command flags; one or more of 'MC_CMD_FLAG_'
+ * @token:	Token of DPSW object
+ * @acl_id:	ACL ID
+ * @cfg:	Interfaces list
+ *
+ * Return:	'0' on Success; Error code otherwise.
+ */
+int dpsw_acl_remove_if(struct fsl_mc_io *mc_io, u32 cmd_flags, u16 token,
+		       u16 acl_id, const struct dpsw_acl_if_cfg *cfg)
+{
+	struct dpsw_cmd_acl_if *cmd_params;
+	struct fsl_mc_command cmd = { 0 };
+
+	/* prepare command */
+	cmd.header = mc_encode_cmd_header(DPSW_CMDID_ACL_REMOVE_IF, cmd_flags,
+					  token);
+	cmd_params = (struct dpsw_cmd_acl_if *)cmd.params;
+	cmd_params->acl_id = cpu_to_le16(acl_id);
+	cmd_params->num_ifs = cpu_to_le16(cfg->num_ifs);
+	build_if_id_bitmap(&cmd_params->if_id, cfg->if_id, cfg->num_ifs);
+
+	/* send command to mc*/
+	return mc_send_command(mc_io, &cmd);
+}
+
+/**
+ * dpsw_acl_prepare_entry_cfg() - Setup an ACL entry
+ * @key:		Key
+ * @entry_cfg_buf:	Zeroed 256 bytes of memory before mapping it to DMA
+ *
+ * This function has to be called before adding or removing acl_entry
+ *
+ */
+void dpsw_acl_prepare_entry_cfg(const struct dpsw_acl_key *key,
+				u8 *entry_cfg_buf)
+{
+	struct dpsw_prep_acl_entry *ext_params;
+	int i;
+
+	ext_params = (struct dpsw_prep_acl_entry *)entry_cfg_buf;
+
+	for (i = 0; i < 6; i++) {
+		ext_params->match_l2_dest_mac[i] = key->match.l2_dest_mac[5 - i];
+		ext_params->match_l2_source_mac[i] = key->match.l2_source_mac[5 - i];
+		ext_params->mask_l2_dest_mac[i] = key->mask.l2_dest_mac[5 - i];
+		ext_params->mask_l2_source_mac[i] = key->mask.l2_source_mac[5 - i];
+	}
+
+	ext_params->match_l2_tpid = cpu_to_le16(key->match.l2_tpid);
+	ext_params->match_l2_vlan_id = cpu_to_le16(key->match.l2_vlan_id);
+	ext_params->match_l3_dest_ip = cpu_to_le32(key->match.l3_dest_ip);
+	ext_params->match_l3_source_ip = cpu_to_le32(key->match.l3_source_ip);
+	ext_params->match_l4_dest_port = cpu_to_le16(key->match.l4_dest_port);
+	ext_params->match_l4_source_port = cpu_to_le16(key->match.l4_source_port);
+	ext_params->match_l2_ether_type = cpu_to_le16(key->match.l2_ether_type);
+	ext_params->match_l2_pcp_dei = key->match.l2_pcp_dei;
+	ext_params->match_l3_dscp = key->match.l3_dscp;
+
+	ext_params->mask_l2_tpid = cpu_to_le16(key->mask.l2_tpid);
+	ext_params->mask_l2_vlan_id = cpu_to_le16(key->mask.l2_vlan_id);
+	ext_params->mask_l3_dest_ip = cpu_to_le32(key->mask.l3_dest_ip);
+	ext_params->mask_l3_source_ip = cpu_to_le32(key->mask.l3_source_ip);
+	ext_params->mask_l4_dest_port = cpu_to_le16(key->mask.l4_dest_port);
+	ext_params->mask_l4_source_port = cpu_to_le16(key->mask.l4_source_port);
+	ext_params->mask_l2_ether_type = cpu_to_le16(key->mask.l2_ether_type);
+	ext_params->mask_l2_pcp_dei = key->mask.l2_pcp_dei;
+	ext_params->mask_l3_dscp = key->mask.l3_dscp;
+	ext_params->match_l3_protocol = key->match.l3_protocol;
+	ext_params->mask_l3_protocol = key->mask.l3_protocol;
+}
+
+/**
+ * dpsw_acl_add_entry() - Add a rule to the ACL table.
+ * @mc_io:	Pointer to MC portal's I/O object
+ * @cmd_flags:	Command flags; one or more of 'MC_CMD_FLAG_'
+ * @token:	Token of DPSW object
+ * @acl_id:	ACL ID
+ * @cfg:	Entry configuration
+ *
+ * warning: This function has to be called after dpsw_acl_prepare_entry_cfg()
+ *
+ * Return:	'0' on Success; Error code otherwise.
+ */
+int dpsw_acl_add_entry(struct fsl_mc_io *mc_io, u32 cmd_flags, u16 token,
+		       u16 acl_id, const struct dpsw_acl_entry_cfg *cfg)
+{
+	struct dpsw_cmd_acl_entry *cmd_params;
+	struct fsl_mc_command cmd = { 0 };
+
+	cmd.header = mc_encode_cmd_header(DPSW_CMDID_ACL_ADD_ENTRY, cmd_flags,
+					  token);
+	cmd_params = (struct dpsw_cmd_acl_entry *)cmd.params;
+	cmd_params->acl_id = cpu_to_le16(acl_id);
+	cmd_params->result_if_id = cpu_to_le16(cfg->result.if_id);
+	cmd_params->precedence = cpu_to_le32(cfg->precedence);
+	cmd_params->key_iova = cpu_to_le64(cfg->key_iova);
+	dpsw_set_field(cmd_params->result_action,
+		       RESULT_ACTION,
+		       cfg->result.action);
+
+	return mc_send_command(mc_io, &cmd);
+}
+
+/**
+ * dpsw_acl_remove_entry() - Removes an entry from ACL.
+ * @mc_io:	Pointer to MC portal's I/O object
+ * @cmd_flags:	Command flags; one or more of 'MC_CMD_FLAG_'
+ * @token:	Token of DPSW object
+ * @acl_id:	ACL ID
+ * @cfg:	Entry configuration
+ *
+ * warning: This function has to be called after dpsw_acl_set_entry_cfg()
+ *
+ * Return:	'0' on Success; Error code otherwise.
+ */
+int dpsw_acl_remove_entry(struct fsl_mc_io *mc_io, u32 cmd_flags, u16 token,
+			  u16 acl_id, const struct dpsw_acl_entry_cfg *cfg)
+{
+	struct dpsw_cmd_acl_entry *cmd_params;
+	struct fsl_mc_command cmd = { 0 };
+
+	/* prepare command */
+	cmd.header = mc_encode_cmd_header(DPSW_CMDID_ACL_REMOVE_ENTRY,
+					  cmd_flags,
+					  token);
+	cmd_params = (struct dpsw_cmd_acl_entry *)cmd.params;
+	cmd_params->acl_id = cpu_to_le16(acl_id);
+	cmd_params->result_if_id = cpu_to_le16(cfg->result.if_id);
+	cmd_params->precedence = cpu_to_le32(cfg->precedence);
+	cmd_params->key_iova = cpu_to_le64(cfg->key_iova);
+	dpsw_set_field(cmd_params->result_action,
+		       RESULT_ACTION,
+		       cfg->result.action);
+
+	/* send command to mc*/
+	return mc_send_command(mc_io, &cmd);
+}
+
+/**
+ * dpsw_set_reflection_if() - Set target interface for traffic mirrored
+ * @mc_io:	Pointer to MC portal's I/O object
+ * @cmd_flags:	Command flags; one or more of 'MC_CMD_FLAG_'
+ * @token:	Token of DPSW object
+ * @if_id:	Interface Id
+ *
+ * Only one mirroring destination is allowed per switch
+ *
+ * Return:	Completion status. '0' on Success; Error code otherwise.
+ */
+int dpsw_set_reflection_if(struct fsl_mc_io *mc_io, u32 cmd_flags, u16 token,
+			   u16 if_id)
+{
+	struct dpsw_cmd_set_reflection_if *cmd_params;
+	struct fsl_mc_command cmd = { 0 };
+
+	cmd.header = mc_encode_cmd_header(DPSW_CMDID_SET_REFLECTION_IF,
+					  cmd_flags,
+					  token);
+	cmd_params = (struct dpsw_cmd_set_reflection_if *)cmd.params;
+	cmd_params->if_id = cpu_to_le16(if_id);
+
+	return mc_send_command(mc_io, &cmd);
+}
+
+/**
+ * dpsw_if_add_reflection() - Setup mirroring rule
+ * @mc_io:	Pointer to MC portal's I/O object
+ * @cmd_flags:	Command flags; one or more of 'MC_CMD_FLAG_'
+ * @token:	Token of DPSW object
+ * @if_id:	Interface Identifier
+ * @cfg:	Reflection configuration
+ *
+ * Return:	Completion status. '0' on Success; Error code otherwise.
+ */
+int dpsw_if_add_reflection(struct fsl_mc_io *mc_io, u32 cmd_flags, u16 token,
+			   u16 if_id, const struct dpsw_reflection_cfg *cfg)
+{
+	struct dpsw_cmd_if_reflection *cmd_params;
+	struct fsl_mc_command cmd = { 0 };
+
+	cmd.header = mc_encode_cmd_header(DPSW_CMDID_IF_ADD_REFLECTION,
+					  cmd_flags,
+					  token);
+	cmd_params = (struct dpsw_cmd_if_reflection *)cmd.params;
+	cmd_params->if_id = cpu_to_le16(if_id);
+	cmd_params->vlan_id = cpu_to_le16(cfg->vlan_id);
+	dpsw_set_field(cmd_params->filter, FILTER, cfg->filter);
+
+	return mc_send_command(mc_io, &cmd);
+}
+
+/**
+ * dpsw_if_remove_reflection() - Remove mirroring rule
+ * @mc_io:	Pointer to MC portal's I/O object
+ * @cmd_flags:	Command flags; one or more of 'MC_CMD_FLAG_'
+ * @token:	Token of DPSW object
+ * @if_id:	Interface Identifier
+ * @cfg:	Reflection configuration
+ *
+ * Return:	Completion status. '0' on Success; Error code otherwise.
+ */
+int dpsw_if_remove_reflection(struct fsl_mc_io *mc_io, u32 cmd_flags, u16 token,
+			      u16 if_id, const struct dpsw_reflection_cfg *cfg)
+{
+	struct dpsw_cmd_if_reflection *cmd_params;
+	struct fsl_mc_command cmd = { 0 };
+
+	cmd.header = mc_encode_cmd_header(DPSW_CMDID_IF_REMOVE_REFLECTION,
+					  cmd_flags,
+					  token);
+	cmd_params = (struct dpsw_cmd_if_reflection *)cmd.params;
+	cmd_params->if_id = cpu_to_le16(if_id);
+	cmd_params->vlan_id = cpu_to_le16(cfg->vlan_id);
+	dpsw_set_field(cmd_params->filter, FILTER, cfg->filter);
+
+	return mc_send_command(mc_io, &cmd);
+}
diff --git a/drivers/net/ethernet/freescale/dpaa2/dpsw.h b/drivers/net/ethernet/freescale/dpaa2/dpsw.h
new file mode 100644
index 000000000000..b90bd363f47a
--- /dev/null
+++ b/drivers/net/ethernet/freescale/dpaa2/dpsw.h
@@ -0,0 +1,791 @@
+/* SPDX-License-Identifier: GPL-2.0 */
+/*
+ * Copyright 2014-2016 Freescale Semiconductor Inc.
+ * Copyright 2017-2021 NXP
+ *
+ */
+
+#ifndef __FSL_DPSW_H
+#define __FSL_DPSW_H
+
+/* Data Path L2-Switch API
+ * Contains API for handling DPSW topology and functionality
+ */
+
+struct fsl_mc_io;
+
+/* DPSW general definitions */
+
+#define DPSW_MAX_PRIORITIES	8
+
+#define DPSW_MAX_IF		64
+
+int dpsw_open(struct fsl_mc_io *mc_io, u32 cmd_flags, int dpsw_id, u16 *token);
+
+int dpsw_close(struct fsl_mc_io *mc_io, u32 cmd_flags, u16 token);
+
+/* DPSW options */
+
+/**
+ * DPSW_OPT_FLOODING_DIS - Flooding was disabled at device create
+ */
+#define DPSW_OPT_FLOODING_DIS		0x0000000000000001ULL
+/**
+ * DPSW_OPT_MULTICAST_DIS - Multicast was disabled at device create
+ */
+#define DPSW_OPT_MULTICAST_DIS		0x0000000000000004ULL
+/**
+ * DPSW_OPT_CTRL_IF_DIS - Control interface support is disabled
+ */
+#define DPSW_OPT_CTRL_IF_DIS		0x0000000000000010ULL
+
+/**
+ * enum dpsw_component_type - component type of a bridge
+ * @DPSW_COMPONENT_TYPE_C_VLAN: A C-VLAN component of an
+ *   enterprise VLAN bridge or of a Provider Bridge used
+ *   to process C-tagged frames
+ * @DPSW_COMPONENT_TYPE_S_VLAN: An S-VLAN component of a
+ *   Provider Bridge
+ *
+ */
+enum dpsw_component_type {
+	DPSW_COMPONENT_TYPE_C_VLAN = 0,
+	DPSW_COMPONENT_TYPE_S_VLAN
+};
+
+/**
+ *  enum dpsw_flooding_cfg - flooding configuration requested
+ * @DPSW_FLOODING_PER_VLAN: Flooding replicators are allocated per VLAN and
+ * interfaces present in each of them can be configured using
+ * dpsw_vlan_add_if_flooding()/dpsw_vlan_remove_if_flooding().
+ * This is the default configuration.
+ *
+ * @DPSW_FLOODING_PER_FDB: Flooding replicators are allocated per FDB and
+ * interfaces present in each of them can be configured using
+ * dpsw_set_egress_flood().
+ */
+enum dpsw_flooding_cfg {
+	DPSW_FLOODING_PER_VLAN = 0,
+	DPSW_FLOODING_PER_FDB,
+};
+
+/**
+ * enum dpsw_broadcast_cfg - broadcast configuration requested
+ * @DPSW_BROADCAST_PER_OBJECT: There is only one broadcast replicator per DPSW
+ * object. This is the default configuration.
+ * @DPSW_BROADCAST_PER_FDB: Broadcast replicators are allocated per FDB and
+ * interfaces present in each of them can be configured using
+ * dpsw_set_egress_flood().
+ */
+enum dpsw_broadcast_cfg {
+	DPSW_BROADCAST_PER_OBJECT = 0,
+	DPSW_BROADCAST_PER_FDB,
+};
+
+int dpsw_enable(struct fsl_mc_io *mc_io, u32 cmd_flags, u16 token);
+
+int dpsw_disable(struct fsl_mc_io *mc_io, u32 cmd_flags, u16 token);
+
+int dpsw_reset(struct fsl_mc_io *mc_io, u32 cmd_flags, u16 token);
+
+/* DPSW IRQ Index and Events */
+
+#define DPSW_IRQ_INDEX_IF		0x0000
+#define DPSW_IRQ_INDEX_L2SW		0x0001
+
+/**
+ * DPSW_IRQ_EVENT_LINK_CHANGED - Indicates that the link state changed
+ */
+#define DPSW_IRQ_EVENT_LINK_CHANGED	0x0001
+
+/**
+ * DPSW_IRQ_EVENT_ENDPOINT_CHANGED - Indicates a change in endpoint
+ */
+#define DPSW_IRQ_EVENT_ENDPOINT_CHANGED	0x0002
+
+/**
+ * struct dpsw_irq_cfg - IRQ configuration
+ * @addr:	Address that must be written to signal a message-based interrupt
+ * @val:	Value to write into irq_addr address
+ * @irq_num: A user defined number associated with this IRQ
+ */
+struct dpsw_irq_cfg {
+	u64 addr;
+	u32 val;
+	int irq_num;
+};
+
+int dpsw_set_irq_enable(struct fsl_mc_io *mc_io, u32 cmd_flags, u16 token,
+			u8 irq_index, u8 en);
+
+int dpsw_set_irq_mask(struct fsl_mc_io *mc_io, u32 cmd_flags, u16 token,
+		      u8 irq_index, u32 mask);
+
+int dpsw_get_irq_status(struct fsl_mc_io *mc_io, u32 cmd_flags, u16 token,
+			u8 irq_index, u32 *status);
+
+int dpsw_clear_irq_status(struct fsl_mc_io *mc_io, u32 cmd_flags, u16 token,
+			  u8 irq_index, u32 status);
+
+/**
+ * struct dpsw_attr - Structure representing DPSW attributes
+ * @id: DPSW object ID
+ * @options: Enable/Disable DPSW features
+ * @max_vlans: Maximum Number of VLANs
+ * @max_meters_per_if:  Number of meters per interface
+ * @max_fdbs: Maximum Number of FDBs
+ * @max_fdb_entries: Number of FDB entries for default FDB table;
+ *			0 - indicates default 1024 entries.
+ * @fdb_aging_time: Default FDB aging time for default FDB table;
+ *			0 - indicates default 300 seconds
+ * @max_fdb_mc_groups: Number of multicast groups in each FDB table;
+ *			0 - indicates default 32
+ * @mem_size: DPSW frame storage memory size
+ * @num_ifs: Number of interfaces
+ * @num_vlans: Current number of VLANs
+ * @num_fdbs: Current number of FDBs
+ * @component_type: Component type of this bridge
+ * @flooding_cfg: Flooding configuration (PER_VLAN - default, PER_FDB)
+ * @broadcast_cfg: Broadcast configuration (PER_OBJECT - default, PER_FDB)
+ */
+struct dpsw_attr {
+	int id;
+	u64 options;
+	u16 max_vlans;
+	u8 max_meters_per_if;
+	u8 max_fdbs;
+	u16 max_fdb_entries;
+	u16 fdb_aging_time;
+	u16 max_fdb_mc_groups;
+	u16 num_ifs;
+	u16 mem_size;
+	u16 num_vlans;
+	u8 num_fdbs;
+	enum dpsw_component_type component_type;
+	enum dpsw_flooding_cfg flooding_cfg;
+	enum dpsw_broadcast_cfg broadcast_cfg;
+};
+
+int dpsw_get_attributes(struct fsl_mc_io *mc_io, u32 cmd_flags, u16 token,
+			struct dpsw_attr *attr);
+
+/**
+ * struct dpsw_ctrl_if_attr - Control interface attributes
+ * @rx_fqid:		Receive FQID
+ * @rx_err_fqid:	Receive error FQID
+ * @tx_err_conf_fqid:	Transmit error and confirmation FQID
+ */
+struct dpsw_ctrl_if_attr {
+	u32 rx_fqid;
+	u32 rx_err_fqid;
+	u32 tx_err_conf_fqid;
+};
+
+int dpsw_ctrl_if_get_attributes(struct fsl_mc_io *mc_io, u32 cmd_flags,
+				u16 token, struct dpsw_ctrl_if_attr *attr);
+
+enum dpsw_queue_type {
+	DPSW_QUEUE_RX,
+	DPSW_QUEUE_TX_ERR_CONF,
+	DPSW_QUEUE_RX_ERR,
+};
+
+#define DPSW_MAX_DPBP     8
+
+/**
+ * struct dpsw_ctrl_if_pools_cfg - Control interface buffer pools configuration
+ * @num_dpbp: Number of DPBPs
+ * @pools: Array of buffer pools parameters; The number of valid entries
+ *	must match 'num_dpbp' value
+ * @pools.dpbp_id: DPBP object ID
+ * @pools.buffer_size: Buffer size
+ * @pools.backup_pool: Backup pool
+ */
+struct dpsw_ctrl_if_pools_cfg {
+	u8 num_dpbp;
+	struct {
+		int dpbp_id;
+		u16 buffer_size;
+		int backup_pool;
+	} pools[DPSW_MAX_DPBP];
+};
+
+int dpsw_ctrl_if_set_pools(struct fsl_mc_io *mc_io, u32 cmd_flags, u16 token,
+			   const struct dpsw_ctrl_if_pools_cfg *cfg);
+
+#define DPSW_CTRL_IF_QUEUE_OPT_USER_CTX		0x00000001
+#define DPSW_CTRL_IF_QUEUE_OPT_DEST		0x00000002
+
+enum dpsw_ctrl_if_dest {
+	DPSW_CTRL_IF_DEST_NONE = 0,
+	DPSW_CTRL_IF_DEST_DPIO = 1,
+};
+
+struct dpsw_ctrl_if_dest_cfg {
+	enum dpsw_ctrl_if_dest dest_type;
+	int dest_id;
+	u8 priority;
+};
+
+struct dpsw_ctrl_if_queue_cfg {
+	u32 options;
+	u64 user_ctx;
+	struct dpsw_ctrl_if_dest_cfg dest_cfg;
+};
+
+int dpsw_ctrl_if_set_queue(struct fsl_mc_io *mc_io, u32 cmd_flags, u16 token,
+			   enum dpsw_queue_type qtype,
+			   const struct dpsw_ctrl_if_queue_cfg *cfg);
+
+int dpsw_ctrl_if_enable(struct fsl_mc_io *mc_io, u32 cmd_flags, u16 token);
+
+int dpsw_ctrl_if_disable(struct fsl_mc_io *mc_io, u32 cmd_flags, u16 token);
+
+/**
+ * enum dpsw_action - Action selection for special/control frames
+ * @DPSW_ACTION_DROP: Drop frame
+ * @DPSW_ACTION_REDIRECT: Redirect frame to control port
+ */
+enum dpsw_action {
+	DPSW_ACTION_DROP = 0,
+	DPSW_ACTION_REDIRECT = 1
+};
+
+#define DPSW_LINK_OPT_AUTONEG		0x0000000000000001ULL
+#define DPSW_LINK_OPT_HALF_DUPLEX	0x0000000000000002ULL
+#define DPSW_LINK_OPT_PAUSE		0x0000000000000004ULL
+#define DPSW_LINK_OPT_ASYM_PAUSE	0x0000000000000008ULL
+
+/**
+ * struct dpsw_link_cfg - Structure representing DPSW link configuration
+ * @rate: Rate
+ * @options: Mask of available options; use 'DPSW_LINK_OPT_<X>' values
+ */
+struct dpsw_link_cfg {
+	u32 rate;
+	u64 options;
+};
+
+int dpsw_if_set_link_cfg(struct fsl_mc_io *mc_io, u32 cmd_flags, u16 token, u16 if_id,
+			 struct dpsw_link_cfg *cfg);
+
+/**
+ * struct dpsw_link_state - Structure representing DPSW link state
+ * @rate: Rate
+ * @options: Mask of available options; use 'DPSW_LINK_OPT_<X>' values
+ * @up: 0 - covers two cases: down and disconnected, 1 - up
+ */
+struct dpsw_link_state {
+	u32 rate;
+	u64 options;
+	u8 up;
+};
+
+int dpsw_if_get_link_state(struct fsl_mc_io *mc_io, u32 cmd_flags, u16 token,
+			   u16 if_id, struct dpsw_link_state *state);
+
+/**
+ * struct dpsw_tci_cfg - Tag Control Information (TCI) configuration
+ * @pcp: Priority Code Point (PCP): a 3-bit field which refers
+ *		 to the IEEE 802.1p priority
+ * @dei: Drop Eligible Indicator (DEI): a 1-bit field. May be used
+ *		 separately or in conjunction with PCP to indicate frames
+ *		 eligible to be dropped in the presence of congestion
+ * @vlan_id: VLAN Identifier (VID): a 12-bit field specifying the VLAN
+ *			to which the frame belongs. The hexadecimal values
+ *			of 0x000 and 0xFFF are reserved;
+ *			all other values may be used as VLAN identifiers,
+ *			allowing up to 4,094 VLANs
+ */
+struct dpsw_tci_cfg {
+	u8 pcp;
+	u8 dei;
+	u16 vlan_id;
+};
+
+int dpsw_if_set_tci(struct fsl_mc_io *mc_io, u32 cmd_flags, u16 token, u16 if_id,
+		    const struct dpsw_tci_cfg *cfg);
+
+int dpsw_if_get_tci(struct fsl_mc_io *mc_io, u32 cmd_flags, u16 token, u16 if_id,
+		    struct dpsw_tci_cfg *cfg);
+
+/**
+ * enum dpsw_stp_state - Spanning Tree Protocol (STP) states
+ * @DPSW_STP_STATE_DISABLED: Disabled state
+ * @DPSW_STP_STATE_LISTENING: Listening state
+ * @DPSW_STP_STATE_LEARNING: Learning state
+ * @DPSW_STP_STATE_FORWARDING: Forwarding state
+ * @DPSW_STP_STATE_BLOCKING: Blocking state
+ *
+ */
+enum dpsw_stp_state {
+	DPSW_STP_STATE_DISABLED = 0,
+	DPSW_STP_STATE_LISTENING = 1,
+	DPSW_STP_STATE_LEARNING = 2,
+	DPSW_STP_STATE_FORWARDING = 3,
+	DPSW_STP_STATE_BLOCKING = 0
+};
+
+/**
+ * struct dpsw_stp_cfg - Spanning Tree Protocol (STP) Configuration
+ * @vlan_id: VLAN ID STP state
+ * @state: STP state
+ */
+struct dpsw_stp_cfg {
+	u16 vlan_id;
+	enum dpsw_stp_state state;
+};
+
+int dpsw_if_set_stp(struct fsl_mc_io *mc_io, u32 cmd_flags, u16 token, u16 if_id,
+		    const struct dpsw_stp_cfg *cfg);
+
+/**
+ * enum dpsw_accepted_frames - Types of frames to accept
+ * @DPSW_ADMIT_ALL: The device accepts VLAN tagged, untagged and
+ *			priority tagged frames
+ * @DPSW_ADMIT_ONLY_VLAN_TAGGED: The device discards untagged frames or
+ *			Priority-Tagged frames received on this interface.
+ *
+ */
+enum dpsw_accepted_frames {
+	DPSW_ADMIT_ALL = 1,
+	DPSW_ADMIT_ONLY_VLAN_TAGGED = 3
+};
+
+/**
+ * enum dpsw_counter  - Counters types
+ * @DPSW_CNT_ING_FRAME: Counts ingress frames
+ * @DPSW_CNT_ING_BYTE: Counts ingress bytes
+ * @DPSW_CNT_ING_FLTR_FRAME: Counts filtered ingress frames
+ * @DPSW_CNT_ING_FRAME_DISCARD: Counts discarded ingress frame
+ * @DPSW_CNT_ING_MCAST_FRAME: Counts ingress multicast frames
+ * @DPSW_CNT_ING_MCAST_BYTE: Counts ingress multicast bytes
+ * @DPSW_CNT_ING_BCAST_FRAME: Counts ingress broadcast frames
+ * @DPSW_CNT_ING_BCAST_BYTES: Counts ingress broadcast bytes
+ * @DPSW_CNT_EGR_FRAME: Counts egress frames
+ * @DPSW_CNT_EGR_BYTE: Counts egress bytes
+ * @DPSW_CNT_EGR_FRAME_DISCARD: Counts discarded egress frames
+ * @DPSW_CNT_EGR_STP_FRAME_DISCARD: Counts egress STP discarded frames
+ * @DPSW_CNT_ING_NO_BUFF_DISCARD: Counts ingress no buffer discarded frames
+ */
+enum dpsw_counter {
+	DPSW_CNT_ING_FRAME = 0x0,
+	DPSW_CNT_ING_BYTE = 0x1,
+	DPSW_CNT_ING_FLTR_FRAME = 0x2,
+	DPSW_CNT_ING_FRAME_DISCARD = 0x3,
+	DPSW_CNT_ING_MCAST_FRAME = 0x4,
+	DPSW_CNT_ING_MCAST_BYTE = 0x5,
+	DPSW_CNT_ING_BCAST_FRAME = 0x6,
+	DPSW_CNT_ING_BCAST_BYTES = 0x7,
+	DPSW_CNT_EGR_FRAME = 0x8,
+	DPSW_CNT_EGR_BYTE = 0x9,
+	DPSW_CNT_EGR_FRAME_DISCARD = 0xa,
+	DPSW_CNT_EGR_STP_FRAME_DISCARD = 0xb,
+	DPSW_CNT_ING_NO_BUFF_DISCARD = 0xc,
+};
+
+int dpsw_if_get_counter(struct fsl_mc_io *mc_io, u32 cmd_flags, u16 token,
+			u16 if_id, enum dpsw_counter type, u64 *counter);
+
+int dpsw_if_enable(struct fsl_mc_io *mc_io, u32 cmd_flags, u16 token, u16 if_id);
+
+int dpsw_if_disable(struct fsl_mc_io *mc_io, u32 cmd_flags, u16 token, u16 if_id);
+
+/**
+ * struct dpsw_if_attr - Structure representing DPSW interface attributes
+ * @num_tcs: Number of traffic classes
+ * @rate: Transmit rate in bits per second
+ * @options: Interface configuration options (bitmap)
+ * @enabled: Indicates if interface is enabled
+ * @accept_all_vlan: The device discards/accepts incoming frames
+ *		for VLANs that do not include this interface
+ * @admit_untagged: When set to 'DPSW_ADMIT_ONLY_VLAN_TAGGED', the device
+ *		discards untagged frames or priority-tagged frames received on
+ *		this interface;
+ *		When set to 'DPSW_ADMIT_ALL', untagged frames or priority-
+ *		tagged frames received on this interface are accepted
+ * @qdid: control frames transmit qdid
+ */
+struct dpsw_if_attr {
+	u8 num_tcs;
+	u32 rate;
+	u32 options;
+	int enabled;
+	int accept_all_vlan;
+	enum dpsw_accepted_frames admit_untagged;
+	u16 qdid;
+};
+
+int dpsw_if_get_attributes(struct fsl_mc_io *mc_io, u32 cmd_flags, u16 token,
+			   u16 if_id, struct dpsw_if_attr *attr);
+
+int dpsw_if_set_max_frame_length(struct fsl_mc_io *mc_io, u32 cmd_flags, u16 token,
+				 u16 if_id, u16 frame_length);
+
+/**
+ * struct dpsw_vlan_cfg - VLAN Configuration
+ * @fdb_id: Forwarding Data Base
+ */
+struct dpsw_vlan_cfg {
+	u16 fdb_id;
+};
+
+int dpsw_vlan_add(struct fsl_mc_io *mc_io, u32 cmd_flags, u16 token,
+		  u16 vlan_id, const struct dpsw_vlan_cfg *cfg);
+
+#define DPSW_VLAN_ADD_IF_OPT_FDB_ID            0x0001
+
+/**
+ * struct dpsw_vlan_if_cfg - Set of VLAN Interfaces
+ * @num_ifs: The number of interfaces that are assigned to the egress
+ *		list for this VLAN
+ * @if_id: The set of interfaces that are
+ *		assigned to the egress list for this VLAN
+ * @options: Options map for this command (DPSW_VLAN_ADD_IF_OPT_FDB_ID)
+ * @fdb_id: FDB id to be used by this VLAN on these specific interfaces
+ *		(taken into account only if the DPSW_VLAN_ADD_IF_OPT_FDB_ID is
+ *		specified in the options field)
+ */
+struct dpsw_vlan_if_cfg {
+	u16 num_ifs;
+	u16 options;
+	u16 if_id[DPSW_MAX_IF];
+	u16 fdb_id;
+};
+
+int dpsw_vlan_add_if(struct fsl_mc_io *mc_io, u32 cmd_flags, u16 token,
+		     u16 vlan_id, const struct dpsw_vlan_if_cfg *cfg);
+
+int dpsw_vlan_add_if_untagged(struct fsl_mc_io *mc_io, u32 cmd_flags, u16 token,
+			      u16 vlan_id, const struct dpsw_vlan_if_cfg *cfg);
+
+int dpsw_vlan_remove_if(struct fsl_mc_io *mc_io, u32 cmd_flags, u16 token,
+			u16 vlan_id, const struct dpsw_vlan_if_cfg *cfg);
+
+int dpsw_vlan_remove_if_untagged(struct fsl_mc_io *mc_io, u32 cmd_flags, u16 token,
+				 u16 vlan_id, const struct dpsw_vlan_if_cfg *cfg);
+
+int dpsw_vlan_remove(struct fsl_mc_io *mc_io, u32 cmd_flags, u16 token,
+		     u16 vlan_id);
+
+/**
+ * enum dpsw_fdb_entry_type - FDB Entry type - Static/Dynamic
+ * @DPSW_FDB_ENTRY_STATIC: Static entry
+ * @DPSW_FDB_ENTRY_DINAMIC: Dynamic entry
+ */
+enum dpsw_fdb_entry_type {
+	DPSW_FDB_ENTRY_STATIC = 0,
+	DPSW_FDB_ENTRY_DINAMIC = 1
+};
+
+/**
+ * struct dpsw_fdb_unicast_cfg - Unicast entry configuration
+ * @type: Select static or dynamic entry
+ * @mac_addr: MAC address
+ * @if_egress: Egress interface ID
+ */
+struct dpsw_fdb_unicast_cfg {
+	enum dpsw_fdb_entry_type type;
+	u8 mac_addr[6];
+	u16 if_egress;
+};
+
+int dpsw_fdb_add_unicast(struct fsl_mc_io *mc_io, u32 cmd_flags, u16 token,
+			 u16 fdb_id, const struct dpsw_fdb_unicast_cfg *cfg);
+
+int dpsw_fdb_remove_unicast(struct fsl_mc_io *mc_io, u32 cmd_flags, u16 token,
+			    u16 fdb_id, const struct dpsw_fdb_unicast_cfg *cfg);
+
+#define DPSW_FDB_ENTRY_TYPE_DYNAMIC  BIT(0)
+#define DPSW_FDB_ENTRY_TYPE_UNICAST  BIT(1)
+
+/**
+ * struct fdb_dump_entry - fdb snapshot entry
+ * @mac_addr: MAC address
+ * @type: bit0 - DINAMIC(1)/STATIC(0), bit1 - UNICAST(1)/MULTICAST(0)
+ * @if_info: unicast - egress interface, multicast - number of egress interfaces
+ * @if_mask: multicast - egress interface mask
+ */
+struct fdb_dump_entry {
+	u8 mac_addr[6];
+	u8 type;
+	u8 if_info;
+	u8 if_mask[8];
+};
+
+int dpsw_fdb_dump(struct fsl_mc_io *mc_io, u32 cmd_flags, u16 token, u16 fdb_id,
+		  u64 iova_addr, u32 iova_size, u16 *num_entries);
+
+/**
+ * struct dpsw_fdb_multicast_cfg - Multi-cast entry configuration
+ * @type: Select static or dynamic entry
+ * @mac_addr: MAC address
+ * @num_ifs: Number of external and internal interfaces
+ * @if_id: Egress interface IDs
+ */
+struct dpsw_fdb_multicast_cfg {
+	enum dpsw_fdb_entry_type type;
+	u8 mac_addr[6];
+	u16 num_ifs;
+	u16 if_id[DPSW_MAX_IF];
+};
+
+int dpsw_fdb_add_multicast(struct fsl_mc_io *mc_io, u32 cmd_flags, u16 token,
+			   u16 fdb_id, const struct dpsw_fdb_multicast_cfg *cfg);
+
+int dpsw_fdb_remove_multicast(struct fsl_mc_io *mc_io, u32 cmd_flags, u16 token,
+			      u16 fdb_id, const struct dpsw_fdb_multicast_cfg *cfg);
+
+/**
+ * enum dpsw_learning_mode - Auto-learning modes
+ * @DPSW_LEARNING_MODE_DIS: Disable Auto-learning
+ * @DPSW_LEARNING_MODE_HW: Enable HW auto-Learning
+ * @DPSW_LEARNING_MODE_NON_SECURE: Enable None secure learning by CPU
+ * @DPSW_LEARNING_MODE_SECURE: Enable secure learning by CPU
+ *
+ *	NONE - SECURE LEARNING
+ *	SMAC found	DMAC found	CTLU Action
+ *	v		v	Forward frame to
+ *						1.  DMAC destination
+ *	-		v	Forward frame to
+ *						1.  DMAC destination
+ *						2.  Control interface
+ *	v		-	Forward frame to
+ *						1.  Flooding list of interfaces
+ *	-		-	Forward frame to
+ *						1.  Flooding list of interfaces
+ *						2.  Control interface
+ *	SECURE LEARING
+ *	SMAC found	DMAC found	CTLU Action
+ *	v		v		Forward frame to
+ *						1.  DMAC destination
+ *	-		v		Forward frame to
+ *						1.  Control interface
+ *	v		-		Forward frame to
+ *						1.  Flooding list of interfaces
+ *	-		-		Forward frame to
+ *						1.  Control interface
+ */
+enum dpsw_learning_mode {
+	DPSW_LEARNING_MODE_DIS = 0,
+	DPSW_LEARNING_MODE_HW = 1,
+	DPSW_LEARNING_MODE_NON_SECURE = 2,
+	DPSW_LEARNING_MODE_SECURE = 3
+};
+
+/**
+ * struct dpsw_fdb_attr - FDB Attributes
+ * @max_fdb_entries: Number of FDB entries
+ * @fdb_ageing_time: Ageing time in seconds
+ * @learning_mode: Learning mode
+ * @num_fdb_mc_groups: Current number of multicast groups
+ * @max_fdb_mc_groups: Maximum number of multicast groups
+ */
+struct dpsw_fdb_attr {
+	u16 max_fdb_entries;
+	u16 fdb_ageing_time;
+	enum dpsw_learning_mode learning_mode;
+	u16 num_fdb_mc_groups;
+	u16 max_fdb_mc_groups;
+};
+
+int dpsw_get_api_version(struct fsl_mc_io *mc_io, u32 cmd_flags,
+			 u16 *major_ver, u16 *minor_ver);
+
+int dpsw_if_get_port_mac_addr(struct fsl_mc_io *mc_io, u32 cmd_flags, u16 token,
+			      u16 if_id, u8 mac_addr[6]);
+
+/**
+ * struct dpsw_fdb_cfg  - FDB Configuration
+ * @num_fdb_entries: Number of FDB entries
+ * @fdb_ageing_time: Ageing time in seconds
+ */
+struct dpsw_fdb_cfg {
+	u16 num_fdb_entries;
+	u16 fdb_ageing_time;
+};
+
+int dpsw_fdb_add(struct fsl_mc_io *mc_io, u32 cmd_flags, u16 token, u16 *fdb_id,
+		 const struct dpsw_fdb_cfg *cfg);
+
+int dpsw_fdb_remove(struct fsl_mc_io *mc_io, u32 cmd_flags, u16 token, u16 fdb_id);
+
+/**
+ * enum dpsw_flood_type - Define the flood type of a DPSW object
+ * @DPSW_BROADCAST: Broadcast flooding
+ * @DPSW_FLOODING: Unknown flooding
+ */
+enum dpsw_flood_type {
+	DPSW_BROADCAST = 0,
+	DPSW_FLOODING,
+};
+
+struct dpsw_egress_flood_cfg {
+	u16 fdb_id;
+	enum dpsw_flood_type flood_type;
+	u16 num_ifs;
+	u16 if_id[DPSW_MAX_IF];
+};
+
+int dpsw_set_egress_flood(struct fsl_mc_io *mc_io, u32 cmd_flags, u16 token,
+			  const struct dpsw_egress_flood_cfg *cfg);
+
+int dpsw_if_set_learning_mode(struct fsl_mc_io *mc_io, u32 cmd_flags, u16 token,
+			      u16 if_id, enum dpsw_learning_mode mode);
+
+/**
+ * struct dpsw_acl_cfg - ACL Configuration
+ * @max_entries: Number of ACL rules
+ */
+struct dpsw_acl_cfg {
+	u16 max_entries;
+};
+
+int dpsw_acl_add(struct fsl_mc_io *mc_io, u32 cmd_flags, u16 token, u16 *acl_id,
+		 const struct dpsw_acl_cfg *cfg);
+
+int dpsw_acl_remove(struct fsl_mc_io *mc_io, u32 cmd_flags, u16 token,
+		    u16 acl_id);
+
+/**
+ * struct dpsw_acl_if_cfg - List of interfaces to associate with an ACL table
+ * @num_ifs: Number of interfaces
+ * @if_id: List of interfaces
+ */
+struct dpsw_acl_if_cfg {
+	u16 num_ifs;
+	u16 if_id[DPSW_MAX_IF];
+};
+
+int dpsw_acl_add_if(struct fsl_mc_io *mc_io, u32 cmd_flags, u16 token,
+		    u16 acl_id, const struct dpsw_acl_if_cfg *cfg);
+
+int dpsw_acl_remove_if(struct fsl_mc_io *mc_io, u32 cmd_flags, u16 token,
+		       u16 acl_id, const struct dpsw_acl_if_cfg *cfg);
+
+/**
+ * struct dpsw_acl_fields - ACL fields.
+ * @l2_dest_mac: Destination MAC address: BPDU, Multicast, Broadcast, Unicast,
+ *			slow protocols, MVRP, STP
+ * @l2_source_mac: Source MAC address
+ * @l2_tpid: Layer 2 (Ethernet) protocol type, used to identify the following
+ *		protocols: MPLS, PTP, PFC, ARP, Jumbo frames, LLDP, IEEE802.1ae,
+ *		Q-in-Q, IPv4, IPv6, PPPoE
+ * @l2_pcp_dei: indicate which protocol is encapsulated in the payload
+ * @l2_vlan_id: layer 2 VLAN ID
+ * @l2_ether_type: layer 2 Ethernet type
+ * @l3_dscp: Layer 3 differentiated services code point
+ * @l3_protocol: Tells the Network layer at the destination host, to which
+ *		Protocol this packet belongs to. The following protocol are
+ *		supported: ICMP, IGMP, IPv4 (encapsulation), TCP, IPv6
+ *		(encapsulation), GRE, PTP
+ * @l3_source_ip: Source IPv4 IP
+ * @l3_dest_ip: Destination IPv4 IP
+ * @l4_source_port: Source TCP/UDP Port
+ * @l4_dest_port: Destination TCP/UDP Port
+ */
+struct dpsw_acl_fields {
+	u8 l2_dest_mac[6];
+	u8 l2_source_mac[6];
+	u16 l2_tpid;
+	u8 l2_pcp_dei;
+	u16 l2_vlan_id;
+	u16 l2_ether_type;
+	u8 l3_dscp;
+	u8 l3_protocol;
+	u32 l3_source_ip;
+	u32 l3_dest_ip;
+	u16 l4_source_port;
+	u16 l4_dest_port;
+};
+
+/**
+ * struct dpsw_acl_key - ACL key
+ * @match: Match fields
+ * @mask: Mask: b'1 - valid, b'0 don't care
+ */
+struct dpsw_acl_key {
+	struct dpsw_acl_fields match;
+	struct dpsw_acl_fields mask;
+};
+
+/**
+ * enum dpsw_acl_action - action to be run on the ACL rule match
+ * @DPSW_ACL_ACTION_DROP: Drop frame
+ * @DPSW_ACL_ACTION_REDIRECT: Redirect to certain port
+ * @DPSW_ACL_ACTION_ACCEPT: Accept frame
+ * @DPSW_ACL_ACTION_REDIRECT_TO_CTRL_IF: Redirect to control interface
+ */
+enum dpsw_acl_action {
+	DPSW_ACL_ACTION_DROP,
+	DPSW_ACL_ACTION_REDIRECT,
+	DPSW_ACL_ACTION_ACCEPT,
+	DPSW_ACL_ACTION_REDIRECT_TO_CTRL_IF
+};
+
+/**
+ * struct dpsw_acl_result - ACL action
+ * @action: Action should be taken when	ACL entry hit
+ * @if_id:  Interface IDs to redirect frame. Valid only if redirect selected for
+ *		 action
+ */
+struct dpsw_acl_result {
+	enum dpsw_acl_action action;
+	u16 if_id;
+};
+
+/**
+ * struct dpsw_acl_entry_cfg - ACL entry
+ * @key_iova: I/O virtual address of DMA-able memory filled with key after call
+ *				to dpsw_acl_prepare_entry_cfg()
+ * @result: Required action when entry hit occurs
+ * @precedence: Precedence inside ACL 0 is lowest; This priority can not change
+ *		during the lifetime of a Policy. It is user responsibility to
+ *		space the priorities according to consequent rule additions.
+ */
+struct dpsw_acl_entry_cfg {
+	u64 key_iova;
+	struct dpsw_acl_result result;
+	int precedence;
+};
+
+void dpsw_acl_prepare_entry_cfg(const struct dpsw_acl_key *key,
+				u8 *entry_cfg_buf);
+
+int dpsw_acl_add_entry(struct fsl_mc_io *mc_io, u32 cmd_flags, u16 token,
+		       u16 acl_id, const struct dpsw_acl_entry_cfg *cfg);
+
+int dpsw_acl_remove_entry(struct fsl_mc_io *mc_io, u32 cmd_flags, u16 token,
+			  u16 acl_id, const struct dpsw_acl_entry_cfg *cfg);
+
+/**
+ * enum dpsw_reflection_filter - Filter type for frames to be reflected
+ * @DPSW_REFLECTION_FILTER_INGRESS_ALL: Reflect all frames
+ * @DPSW_REFLECTION_FILTER_INGRESS_VLAN: Reflect only frames that belong to
+ *	the particular VLAN defined by vid parameter
+ *
+ */
+enum dpsw_reflection_filter {
+	DPSW_REFLECTION_FILTER_INGRESS_ALL = 0,
+	DPSW_REFLECTION_FILTER_INGRESS_VLAN = 1
+};
+
+/**
+ * struct dpsw_reflection_cfg - Structure representing the mirroring config
+ * @filter: Filter type for frames to be mirrored
+ * @vlan_id: VLAN ID to mirror; valid only when the type is DPSW_INGRESS_VLAN
+ */
+struct dpsw_reflection_cfg {
+	enum dpsw_reflection_filter filter;
+	u16 vlan_id;
+};
+
+int dpsw_set_reflection_if(struct fsl_mc_io *mc_io, u32 cmd_flags, u16 token,
+			   u16 if_id);
+
+int dpsw_if_add_reflection(struct fsl_mc_io *mc_io, u32 cmd_flags, u16 token,
+			   u16 if_id, const struct dpsw_reflection_cfg *cfg);
+
+int dpsw_if_remove_reflection(struct fsl_mc_io *mc_io, u32 cmd_flags, u16 token,
+			      u16 if_id, const struct dpsw_reflection_cfg *cfg);
+#endif /* __FSL_DPSW_H */
diff --git a/drivers/net/ethernet/freescale/enetc/Kconfig b/drivers/net/ethernet/freescale/enetc/Kconfig
new file mode 100644
index 000000000000..117038104b69
--- /dev/null
+++ b/drivers/net/ethernet/freescale/enetc/Kconfig
@@ -0,0 +1,130 @@
+# SPDX-License-Identifier: GPL-2.0
+config FSL_ENETC_CORE
+	tristate
+	select NXP_NETC_LIB if NXP_NTMP
+	help
+	  This module supports common functionality between the PF and VF
+	  drivers for the NXP ENETC controller.
+
+	  If compiled as module (M), the module name is fsl-enetc-core.
+
+config NXP_ENETC_PF_COMMON
+	tristate
+	help
+	  This module supports common functionality between drivers of
+	  different versions of NXP ENETC PF controllers.
+
+	  If compiled as module (M), the module name is nxp-enetc-pf-common.
+
+config NXP_NETC_LIB
+	tristate
+	help
+	  This module provides common functionalities for both ENETC and NETC
+	  Switch, such as NETC Table Management Protocol (NTMP) 2.0, common tc
+	  flower and debugfs interfaces and so on.
+
+config NXP_NTMP
+	bool
+
+config FSL_ENETC
+	tristate "ENETC PF driver"
+	depends on PTP_1588_CLOCK_OPTIONAL
+	depends on PCI_MSI
+	select FSL_ENETC_CORE
+	select FSL_ENETC_IERB
+	select FSL_ENETC_MDIO
+	select NXP_ENETC_PF_COMMON
+	select PHYLINK
+	select PCS_LYNX
+	select DIMLIB
+	help
+	  This driver supports NXP ENETC gigabit ethernet controller PCIe
+	  physical function (PF) devices, managing ENETC Ports at a privileged
+	  level.
+
+	  If compiled as module (M), the module name is fsl-enetc.
+
+config NXP_ENETC4
+	tristate "ENETC4 PF driver"
+	depends on PTP_1588_CLOCK_OPTIONAL
+	depends on PCI_MSI
+	select FSL_ENETC_CORE
+	select FSL_ENETC_MDIO
+	select NXP_ENETC_PF_COMMON
+	select NXP_NTMP
+	select PHYLINK
+	select DIMLIB
+	help
+	  This driver supports NXP ENETC devices with major revision 4. ENETC is
+	  as the NIC functionality in NETC, it supports virtualization/isolation
+	  based on PCIe Single Root IO Virtualization (SR-IOV) and a full range
+	  of TSN standards and NIC offload capabilities.
+
+	  If compiled as module (M), the module name is nxp-enetc4.
+
+config FSL_ENETC_VF
+	tristate "ENETC VF driver"
+	depends on PTP_1588_CLOCK_OPTIONAL
+	depends on PCI_MSI
+	select FSL_ENETC_CORE
+	select FSL_ENETC_MDIO
+	select PHYLINK
+	select DIMLIB
+	help
+	  This driver supports NXP ENETC gigabit ethernet controller PCIe
+	  virtual function (VF) devices enabled by the ENETC PF driver.
+
+	  If compiled as module (M), the module name is fsl-enetc-vf.
+
+config FSL_ENETC_IERB
+	tristate "ENETC IERB driver"
+	help
+	  This driver configures the Integrated Endpoint Register Block on NXP
+	  LS1028A.
+
+	  If compiled as module (M), the module name is fsl-enetc-ierb.
+
+config FSL_ENETC_MDIO
+	tristate "ENETC MDIO driver"
+	depends on PCI && PHYLIB
+	help
+	  This driver supports NXP ENETC Central MDIO controller as a PCIe
+	  physical function (PF) device.
+
+	  If compiled as module (M), the module name is fsl-enetc-mdio.
+
+config FSL_ENETC_PTP_CLOCK
+	tristate "ENETC PTP clock driver"
+	depends on PTP_1588_CLOCK_QORIQ && (FSL_ENETC || FSL_ENETC_VF)
+	default y
+	help
+	  This driver adds support for using the ENETC 1588 timer
+	  as a PTP clock. This clock is only useful if your PTP
+	  programs are getting hardware time stamps on the PTP Ethernet
+	  packets using the SO_TIMESTAMPING API.
+
+	  If compiled as module (M), the module name is fsl-enetc-ptp.
+
+config FSL_ENETC_QOS
+	bool "ENETC hardware Time-sensitive Network support"
+	depends on (FSL_ENETC || FSL_ENETC_VF) && (NET_SCH_TAPRIO || NET_SCH_CBS)
+	help
+	  There are Time-Sensitive Network(TSN) capabilities(802.1Qbv/802.1Qci
+	  /802.1Qbu etc.) supported by ENETC. These TSN capabilities can be set
+	  enable/disable from user space via Qos commands(tc). In the kernel
+	  side, it can be loaded by Qos driver. Currently, it is only support
+	  taprio(802.1Qbv) and Credit Based Shaper(802.1Qbu).
+
+config NXP_NETC_BLK_CTRL
+	tristate "NETC blocks control driver"
+	help
+	  This driver configures Integrated Endpoint Register Block (IERB) and
+	  Privileged Register Block (PRB) of NETC. For i.MX platforms, it also
+	  includes the configuration of NETCMIX block.
+	  The IERB contains registers that are used for pre-boot initialization,
+	  debug, and non-customer configuration. The PRB controls global reset
+	  and global error handling for NETC. The NETCMIX block is mainly used
+	  to set MII protocol and PCS protocol of the links, it also contains
+	  settings for some other functions.
+
+	  If compiled as module (M), the module name is nxp-netc-blk-ctrl.
diff --git a/drivers/net/ethernet/freescale/enetc/Makefile b/drivers/net/ethernet/freescale/enetc/Makefile
new file mode 100644
index 000000000000..f1c5ad45fd76
--- /dev/null
+++ b/drivers/net/ethernet/freescale/enetc/Makefile
@@ -0,0 +1,34 @@
+# SPDX-License-Identifier: GPL-2.0
+
+obj-$(CONFIG_FSL_ENETC_CORE) += fsl-enetc-core.o
+fsl-enetc-core-y := enetc.o enetc_cbdr.o enetc_ethtool.o
+
+obj-$(CONFIG_NXP_ENETC_PF_COMMON) += nxp-enetc-pf-common.o
+nxp-enetc-pf-common-y := enetc_pf_common.o
+
+obj-$(CONFIG_NXP_NETC_LIB) += nxp-netc-lib.o
+nxp-netc-lib-y := ntmp.o
+
+obj-$(CONFIG_FSL_ENETC) += fsl-enetc.o
+fsl-enetc-y := enetc_pf.o
+fsl-enetc-$(CONFIG_PCI_IOV) += enetc_msg.o
+fsl-enetc-$(CONFIG_FSL_ENETC_QOS) += enetc_qos.o
+
+obj-$(CONFIG_NXP_ENETC4) += nxp-enetc4.o
+nxp-enetc4-y := enetc4_pf.o
+nxp-enetc4-$(CONFIG_DEBUG_FS) += enetc4_debugfs.o
+
+obj-$(CONFIG_FSL_ENETC_VF) += fsl-enetc-vf.o
+fsl-enetc-vf-y := enetc_vf.o
+
+obj-$(CONFIG_FSL_ENETC_IERB) += fsl-enetc-ierb.o
+fsl-enetc-ierb-y := enetc_ierb.o
+
+obj-$(CONFIG_FSL_ENETC_MDIO) += fsl-enetc-mdio.o
+fsl-enetc-mdio-y := enetc_pci_mdio.o enetc_mdio.o
+
+obj-$(CONFIG_FSL_ENETC_PTP_CLOCK) += fsl-enetc-ptp.o
+fsl-enetc-ptp-y := enetc_ptp.o
+
+obj-$(CONFIG_NXP_NETC_BLK_CTRL) += nxp-netc-blk-ctrl.o
+nxp-netc-blk-ctrl-y := netc_blk_ctrl.o
diff --git a/drivers/net/ethernet/freescale/enetc/enetc.c b/drivers/net/ethernet/freescale/enetc/enetc.c
new file mode 100644
index 000000000000..aae462a0cf5a
--- /dev/null
+++ b/drivers/net/ethernet/freescale/enetc/enetc.c
@@ -0,0 +1,3734 @@
+// SPDX-License-Identifier: (GPL-2.0+ OR BSD-3-Clause)
+/* Copyright 2017-2019 NXP */
+
+#include "enetc.h"
+#include <linux/bpf_trace.h>
+#include <linux/clk.h>
+#include <linux/tcp.h>
+#include <linux/udp.h>
+#include <linux/vmalloc.h>
+#include <linux/ptp_classify.h>
+#include <net/ip6_checksum.h>
+#include <net/pkt_sched.h>
+#include <net/tso.h>
+
+u32 enetc_port_mac_rd(struct enetc_si *si, u32 reg)
+{
+	return enetc_port_rd(&si->hw, reg);
+}
+EXPORT_SYMBOL_GPL(enetc_port_mac_rd);
+
+void enetc_port_mac_wr(struct enetc_si *si, u32 reg, u32 val)
+{
+	enetc_port_wr(&si->hw, reg, val);
+	if (si->hw_features & ENETC_SI_F_QBU)
+		enetc_port_wr(&si->hw, reg + si->drvdata->pmac_offset, val);
+}
+EXPORT_SYMBOL_GPL(enetc_port_mac_wr);
+
+static void enetc_change_preemptible_tcs(struct enetc_ndev_priv *priv,
+					 u8 preemptible_tcs)
+{
+	if (!(priv->si->hw_features & ENETC_SI_F_QBU))
+		return;
+
+	priv->preemptible_tcs = preemptible_tcs;
+	enetc_mm_commit_preemptible_tcs(priv);
+}
+
+static int enetc_mac_addr_hash_idx(const u8 *addr)
+{
+	u64 fold = __swab64(ether_addr_to_u64(addr)) >> 16;
+	u64 mask = 0;
+	int res = 0;
+	int i;
+
+	for (i = 0; i < 8; i++)
+		mask |= BIT_ULL(i * 6);
+
+	for (i = 0; i < 6; i++)
+		res |= (hweight64(fold & (mask << i)) & 0x1) << i;
+
+	return res;
+}
+
+void enetc_add_mac_addr_ht_filter(struct enetc_mac_filter *filter,
+				  const unsigned char *addr)
+{
+	int idx = enetc_mac_addr_hash_idx(addr);
+
+	/* add hash table entry */
+	__set_bit(idx, filter->mac_hash_table);
+	filter->mac_addr_cnt++;
+}
+EXPORT_SYMBOL_GPL(enetc_add_mac_addr_ht_filter);
+
+void enetc_reset_mac_addr_filter(struct enetc_mac_filter *filter)
+{
+	filter->mac_addr_cnt = 0;
+
+	bitmap_zero(filter->mac_hash_table,
+		    ENETC_MADDR_HASH_TBL_SZ);
+}
+EXPORT_SYMBOL_GPL(enetc_reset_mac_addr_filter);
+
+static int enetc_num_stack_tx_queues(struct enetc_ndev_priv *priv)
+{
+	int num_tx_rings = priv->num_tx_rings;
+
+	if (priv->xdp_prog)
+		return num_tx_rings - num_possible_cpus();
+
+	return num_tx_rings;
+}
+
+static struct enetc_bdr *enetc_rx_ring_from_xdp_tx_ring(struct enetc_ndev_priv *priv,
+							struct enetc_bdr *tx_ring)
+{
+	int index = &priv->tx_ring[tx_ring->index] - priv->xdp_tx_ring;
+
+	return priv->rx_ring[index];
+}
+
+static struct sk_buff *enetc_tx_swbd_get_skb(struct enetc_tx_swbd *tx_swbd)
+{
+	if (tx_swbd->is_xdp_tx || tx_swbd->is_xdp_redirect)
+		return NULL;
+
+	return tx_swbd->skb;
+}
+
+static struct xdp_frame *
+enetc_tx_swbd_get_xdp_frame(struct enetc_tx_swbd *tx_swbd)
+{
+	if (tx_swbd->is_xdp_redirect)
+		return tx_swbd->xdp_frame;
+
+	return NULL;
+}
+
+static void enetc_unmap_tx_buff(struct enetc_bdr *tx_ring,
+				struct enetc_tx_swbd *tx_swbd)
+{
+	/* For XDP_TX, pages come from RX, whereas for the other contexts where
+	 * we have is_dma_page_set, those come from skb_frag_dma_map. We need
+	 * to match the DMA mapping length, so we need to differentiate those.
+	 */
+	if (tx_swbd->is_dma_page)
+		dma_unmap_page(tx_ring->dev, tx_swbd->dma,
+			       tx_swbd->is_xdp_tx ? PAGE_SIZE : tx_swbd->len,
+			       tx_swbd->dir);
+	else
+		dma_unmap_single(tx_ring->dev, tx_swbd->dma,
+				 tx_swbd->len, tx_swbd->dir);
+	tx_swbd->dma = 0;
+}
+
+static void enetc_free_tx_frame(struct enetc_bdr *tx_ring,
+				struct enetc_tx_swbd *tx_swbd)
+{
+	struct xdp_frame *xdp_frame = enetc_tx_swbd_get_xdp_frame(tx_swbd);
+	struct sk_buff *skb = enetc_tx_swbd_get_skb(tx_swbd);
+
+	if (tx_swbd->dma)
+		enetc_unmap_tx_buff(tx_ring, tx_swbd);
+
+	if (xdp_frame) {
+		xdp_return_frame(tx_swbd->xdp_frame);
+		tx_swbd->xdp_frame = NULL;
+	} else if (skb) {
+		dev_kfree_skb_any(skb);
+		tx_swbd->skb = NULL;
+	}
+}
+
+/* Let H/W know BD ring has been updated */
+static void enetc_update_tx_ring_tail(struct enetc_bdr *tx_ring)
+{
+	/* includes wmb() */
+	enetc_wr_reg_hot(tx_ring->tpir, tx_ring->next_to_use);
+}
+
+static int enetc_ptp_parse(struct sk_buff *skb, u8 *udp,
+			   u8 *msgtype, u8 *twostep,
+			   u16 *correction_offset, u16 *body_offset)
+{
+	unsigned int ptp_class;
+	struct ptp_header *hdr;
+	unsigned int type;
+	u8 *base;
+
+	ptp_class = ptp_classify_raw(skb);
+	if (ptp_class == PTP_CLASS_NONE)
+		return -EINVAL;
+
+	hdr = ptp_parse_header(skb, ptp_class);
+	if (!hdr)
+		return -EINVAL;
+
+	type = ptp_class & PTP_CLASS_PMASK;
+	if (type == PTP_CLASS_IPV4 || type == PTP_CLASS_IPV6)
+		*udp = 1;
+	else
+		*udp = 0;
+
+	*msgtype = ptp_get_msgtype(hdr, ptp_class);
+	*twostep = hdr->flag_field[0] & 0x2;
+
+	base = skb_mac_header(skb);
+	*correction_offset = (u8 *)&hdr->correction - base;
+	*body_offset = (u8 *)hdr + sizeof(struct ptp_header) - base;
+
+	return 0;
+}
+
+static bool enetc_tx_csum_offload_check(struct sk_buff *skb)
+{
+	switch (skb->csum_offset) {
+	case offsetof(struct tcphdr, check):
+	case offsetof(struct udphdr, check):
+		return true;
+	default:
+		return false;
+	}
+}
+
+static bool enetc_skb_is_ipv6(struct sk_buff *skb)
+{
+	return vlan_get_protocol(skb) == htons(ETH_P_IPV6);
+}
+
+static bool enetc_skb_is_tcp(struct sk_buff *skb)
+{
+	return skb->csum_offset == offsetof(struct tcphdr, check);
+}
+
+/**
+ * enetc_unwind_tx_frame() - Unwind the DMA mappings of a multi-buffer Tx frame
+ * @tx_ring: Pointer to the Tx ring on which the buffer descriptors are located
+ * @count: Number of Tx buffer descriptors which need to be unmapped
+ * @i: Index of the last successfully mapped Tx buffer descriptor
+ */
+static void enetc_unwind_tx_frame(struct enetc_bdr *tx_ring, int count, int i)
+{
+	while (count--) {
+		struct enetc_tx_swbd *tx_swbd = &tx_ring->tx_swbd[i];
+
+		enetc_free_tx_frame(tx_ring, tx_swbd);
+		if (i == 0)
+			i = tx_ring->bd_count;
+		i--;
+	}
+}
+
+static void enetc_set_one_step_ts(struct enetc_si *si, bool udp, int offset)
+{
+	u32 val = ENETC_PM0_SINGLE_STEP_EN;
+
+	val |= ENETC_SET_SINGLE_STEP_OFFSET(offset);
+	if (udp)
+		val |= ENETC_PM0_SINGLE_STEP_CH;
+
+	/* The "Correction" field of a packet is updated based on the
+	 * current time and the timestamp provided
+	 */
+	enetc_port_mac_wr(si, ENETC_PM0_SINGLE_STEP, val);
+}
+
+static void enetc4_set_one_step_ts(struct enetc_si *si, bool udp, int offset)
+{
+	u32 val = PM_SINGLE_STEP_EN;
+
+	val |= PM_SINGLE_STEP_OFFSET_SET(offset);
+	if (udp)
+		val |= PM_SINGLE_STEP_CH;
+
+	enetc_port_mac_wr(si, ENETC4_PM_SINGLE_STEP(0), val);
+}
+
+static u32 enetc_update_ptp_sync_msg(struct enetc_ndev_priv *priv,
+				     struct sk_buff *skb, bool csum_offload)
+{
+	struct enetc_skb_cb *enetc_cb = ENETC_SKB_CB(skb);
+	u16 tstamp_off = enetc_cb->origin_tstamp_off;
+	u16 corr_off = enetc_cb->correction_off;
+	struct enetc_si *si = priv->si;
+	struct enetc_hw *hw = &si->hw;
+	__be32 new_sec_l, new_nsec;
+	__be16 new_sec_h;
+	u32 lo, hi, nsec;
+	u8 *data;
+	u64 sec;
+
+	lo = enetc_rd_hot(hw, ENETC_SICTR0);
+	hi = enetc_rd_hot(hw, ENETC_SICTR1);
+	sec = (u64)hi << 32 | lo;
+	nsec = do_div(sec, 1000000000);
+
+	/* Update originTimestamp field of Sync packet
+	 * - 48 bits seconds field
+	 * - 32 bits nanseconds field
+	 *
+	 * In addition, if csum_offload is false, the UDP checksum needs
+	 * to be updated by software after updating originTimestamp field,
+	 * otherwise the hardware will calculate the wrong checksum when
+	 * updating the correction field and update it to the packet.
+	 */
+
+	data = skb_mac_header(skb);
+	new_sec_h = htons((sec >> 32) & 0xffff);
+	new_sec_l = htonl(sec & 0xffffffff);
+	new_nsec = htonl(nsec);
+	if (enetc_cb->udp && !csum_offload) {
+		struct udphdr *uh = udp_hdr(skb);
+		__be32 old_sec_l, old_nsec;
+		__be16 old_sec_h;
+
+		old_sec_h = *(__be16 *)(data + tstamp_off);
+		inet_proto_csum_replace2(&uh->check, skb, old_sec_h,
+					 new_sec_h, false);
+
+		old_sec_l = *(__be32 *)(data + tstamp_off + 2);
+		inet_proto_csum_replace4(&uh->check, skb, old_sec_l,
+					 new_sec_l, false);
+
+		old_nsec = *(__be32 *)(data + tstamp_off + 6);
+		inet_proto_csum_replace4(&uh->check, skb, old_nsec,
+					 new_nsec, false);
+	}
+
+	*(__be16 *)(data + tstamp_off) = new_sec_h;
+	*(__be32 *)(data + tstamp_off + 2) = new_sec_l;
+	*(__be32 *)(data + tstamp_off + 6) = new_nsec;
+
+	/* Configure single-step register */
+	if (is_enetc_rev1(si))
+		enetc_set_one_step_ts(si, enetc_cb->udp, corr_off);
+	else
+		enetc4_set_one_step_ts(si, enetc_cb->udp, corr_off);
+
+	return lo & ENETC_TXBD_TSTAMP;
+}
+
+static int enetc_map_tx_buffs(struct enetc_bdr *tx_ring, struct sk_buff *skb)
+{
+	bool do_vlan, do_onestep_tstamp = false, do_twostep_tstamp = false;
+	struct enetc_ndev_priv *priv = netdev_priv(tx_ring->ndev);
+	struct enetc_skb_cb *enetc_cb = ENETC_SKB_CB(skb);
+	struct enetc_tx_swbd *tx_swbd;
+	int len = skb_headlen(skb);
+	union enetc_tx_bd temp_bd;
+	bool csum_offload = false;
+	union enetc_tx_bd *txbd;
+	int i, count = 0;
+	skb_frag_t *frag;
+	unsigned int f;
+	dma_addr_t dma;
+	u8 flags = 0;
+	u32 tstamp;
+
+	enetc_clear_tx_bd(&temp_bd);
+	if (skb->ip_summed == CHECKSUM_PARTIAL) {
+		/* Can not support TSD and checksum offload at the same time */
+		if (priv->active_offloads & ENETC_F_TXCSUM &&
+		    enetc_tx_csum_offload_check(skb) && !tx_ring->tsd_enable) {
+			temp_bd.l3_aux0 = FIELD_PREP(ENETC_TX_BD_L3_START,
+						     skb_network_offset(skb));
+			temp_bd.l3_aux1 = FIELD_PREP(ENETC_TX_BD_L3_HDR_LEN,
+						     skb_network_header_len(skb) / 4);
+			temp_bd.l3_aux1 |= FIELD_PREP(ENETC_TX_BD_L3T,
+						      enetc_skb_is_ipv6(skb));
+			if (enetc_skb_is_tcp(skb))
+				temp_bd.l4_aux = FIELD_PREP(ENETC_TX_BD_L4T,
+							    ENETC_TXBD_L4T_TCP);
+			else
+				temp_bd.l4_aux = FIELD_PREP(ENETC_TX_BD_L4T,
+							    ENETC_TXBD_L4T_UDP);
+			flags |= ENETC_TXBD_FLAGS_CSUM_LSO | ENETC_TXBD_FLAGS_L4CS;
+			csum_offload = true;
+		} else if (skb_checksum_help(skb)) {
+			return 0;
+		}
+	}
+
+	if (enetc_cb->flag & ENETC_F_TX_ONESTEP_SYNC_TSTAMP) {
+		do_onestep_tstamp = true;
+		tstamp = enetc_update_ptp_sync_msg(priv, skb, csum_offload);
+	} else if (enetc_cb->flag & ENETC_F_TX_TSTAMP) {
+		do_twostep_tstamp = true;
+	}
+
+	i = tx_ring->next_to_use;
+	txbd = ENETC_TXBD(*tx_ring, i);
+	prefetchw(txbd);
+
+	dma = dma_map_single(tx_ring->dev, skb->data, len, DMA_TO_DEVICE);
+	if (unlikely(dma_mapping_error(tx_ring->dev, dma)))
+		goto dma_err;
+
+	temp_bd.addr = cpu_to_le64(dma);
+	temp_bd.buf_len = cpu_to_le16(len);
+
+	tx_swbd = &tx_ring->tx_swbd[i];
+	tx_swbd->dma = dma;
+	tx_swbd->len = len;
+	tx_swbd->is_dma_page = 0;
+	tx_swbd->dir = DMA_TO_DEVICE;
+	count++;
+
+	do_vlan = skb_vlan_tag_present(skb);
+	tx_swbd->do_twostep_tstamp = do_twostep_tstamp;
+	tx_swbd->qbv_en = !!(priv->active_offloads & ENETC_F_QBV);
+	tx_swbd->check_wb = tx_swbd->do_twostep_tstamp || tx_swbd->qbv_en;
+
+	if (do_vlan || do_onestep_tstamp || do_twostep_tstamp)
+		flags |= ENETC_TXBD_FLAGS_EX;
+
+	if (tx_ring->tsd_enable)
+		flags |= ENETC_TXBD_FLAGS_TSE | ENETC_TXBD_FLAGS_TXSTART;
+
+	/* first BD needs frm_len and offload flags set */
+	temp_bd.frm_len = cpu_to_le16(skb->len);
+	temp_bd.flags = flags;
+
+	if (flags & ENETC_TXBD_FLAGS_TSE)
+		temp_bd.txstart = enetc_txbd_set_tx_start(skb->skb_mstamp_ns,
+							  flags);
+
+	if (flags & ENETC_TXBD_FLAGS_EX) {
+		u8 e_flags = 0;
+		*txbd = temp_bd;
+		enetc_clear_tx_bd(&temp_bd);
+
+		/* add extension BD for VLAN and/or timestamping */
+		flags = 0;
+		tx_swbd++;
+		txbd++;
+		i++;
+		if (unlikely(i == tx_ring->bd_count)) {
+			i = 0;
+			tx_swbd = tx_ring->tx_swbd;
+			txbd = ENETC_TXBD(*tx_ring, 0);
+		}
+		prefetchw(txbd);
+
+		if (do_vlan) {
+			temp_bd.ext.vid = cpu_to_le16(skb_vlan_tag_get(skb));
+			temp_bd.ext.tpid = 0; /* < C-TAG */
+			e_flags |= ENETC_TXBD_E_FLAGS_VLAN_INS;
+		}
+
+		if (do_onestep_tstamp) {
+			/* Configure extension BD */
+			temp_bd.ext.tstamp = cpu_to_le32(tstamp);
+			e_flags |= ENETC_TXBD_E_FLAGS_ONE_STEP_PTP;
+		} else if (do_twostep_tstamp) {
+			skb_shinfo(skb)->tx_flags |= SKBTX_IN_PROGRESS;
+			e_flags |= ENETC_TXBD_E_FLAGS_TWO_STEP_PTP;
+		}
+
+		temp_bd.ext.e_flags = e_flags;
+		count++;
+	}
+
+	frag = &skb_shinfo(skb)->frags[0];
+	for (f = 0; f < skb_shinfo(skb)->nr_frags; f++, frag++) {
+		len = skb_frag_size(frag);
+		dma = skb_frag_dma_map(tx_ring->dev, frag, 0, len,
+				       DMA_TO_DEVICE);
+		if (dma_mapping_error(tx_ring->dev, dma))
+			goto dma_err;
+
+		*txbd = temp_bd;
+		enetc_clear_tx_bd(&temp_bd);
+
+		flags = 0;
+		tx_swbd++;
+		txbd++;
+		i++;
+		if (unlikely(i == tx_ring->bd_count)) {
+			i = 0;
+			tx_swbd = tx_ring->tx_swbd;
+			txbd = ENETC_TXBD(*tx_ring, 0);
+		}
+		prefetchw(txbd);
+
+		temp_bd.addr = cpu_to_le64(dma);
+		temp_bd.buf_len = cpu_to_le16(len);
+
+		tx_swbd->dma = dma;
+		tx_swbd->len = len;
+		tx_swbd->is_dma_page = 1;
+		tx_swbd->dir = DMA_TO_DEVICE;
+		count++;
+	}
+
+	/* last BD needs 'F' bit set */
+	flags |= ENETC_TXBD_FLAGS_F;
+	temp_bd.flags = flags;
+	*txbd = temp_bd;
+
+	tx_ring->tx_swbd[i].is_eof = true;
+	tx_ring->tx_swbd[i].skb = skb;
+
+	enetc_bdr_idx_inc(tx_ring, &i);
+	tx_ring->next_to_use = i;
+
+	skb_tx_timestamp(skb);
+
+	enetc_update_tx_ring_tail(tx_ring);
+
+	return count;
+
+dma_err:
+	dev_err(tx_ring->dev, "DMA map error");
+
+	enetc_unwind_tx_frame(tx_ring, count, i);
+
+	return 0;
+}
+
+static int enetc_map_tx_tso_hdr(struct enetc_bdr *tx_ring, struct sk_buff *skb,
+				struct enetc_tx_swbd *tx_swbd,
+				union enetc_tx_bd *txbd, int *i, int hdr_len,
+				int data_len)
+{
+	union enetc_tx_bd txbd_tmp;
+	u8 flags = 0, e_flags = 0;
+	dma_addr_t addr;
+	int count = 1;
+
+	enetc_clear_tx_bd(&txbd_tmp);
+	addr = tx_ring->tso_headers_dma + *i * TSO_HEADER_SIZE;
+
+	if (skb_vlan_tag_present(skb))
+		flags |= ENETC_TXBD_FLAGS_EX;
+
+	txbd_tmp.addr = cpu_to_le64(addr);
+	txbd_tmp.buf_len = cpu_to_le16(hdr_len);
+
+	/* first BD needs frm_len and offload flags set */
+	txbd_tmp.frm_len = cpu_to_le16(hdr_len + data_len);
+	txbd_tmp.flags = flags;
+
+	/* For the TSO header we do not set the dma address since we do not
+	 * want it unmapped when we do cleanup. We still set len so that we
+	 * count the bytes sent.
+	 */
+	tx_swbd->len = hdr_len;
+	tx_swbd->do_twostep_tstamp = false;
+	tx_swbd->check_wb = false;
+
+	/* Actually write the header in the BD */
+	*txbd = txbd_tmp;
+
+	/* Add extension BD for VLAN */
+	if (flags & ENETC_TXBD_FLAGS_EX) {
+		/* Get the next BD */
+		enetc_bdr_idx_inc(tx_ring, i);
+		txbd = ENETC_TXBD(*tx_ring, *i);
+		tx_swbd = &tx_ring->tx_swbd[*i];
+		prefetchw(txbd);
+
+		/* Setup the VLAN fields */
+		enetc_clear_tx_bd(&txbd_tmp);
+		txbd_tmp.ext.vid = cpu_to_le16(skb_vlan_tag_get(skb));
+		txbd_tmp.ext.tpid = 0; /* < C-TAG */
+		e_flags |= ENETC_TXBD_E_FLAGS_VLAN_INS;
+
+		/* Write the BD */
+		txbd_tmp.ext.e_flags = e_flags;
+		*txbd = txbd_tmp;
+		count++;
+	}
+
+	return count;
+}
+
+static int enetc_map_tx_tso_data(struct enetc_bdr *tx_ring, struct sk_buff *skb,
+				 struct enetc_tx_swbd *tx_swbd,
+				 union enetc_tx_bd *txbd, char *data,
+				 int size, bool last_bd)
+{
+	union enetc_tx_bd txbd_tmp;
+	dma_addr_t addr;
+	u8 flags = 0;
+
+	enetc_clear_tx_bd(&txbd_tmp);
+
+	addr = dma_map_single(tx_ring->dev, data, size, DMA_TO_DEVICE);
+	if (unlikely(dma_mapping_error(tx_ring->dev, addr))) {
+		netdev_err(tx_ring->ndev, "DMA map error\n");
+		return -ENOMEM;
+	}
+
+	if (last_bd) {
+		flags |= ENETC_TXBD_FLAGS_F;
+		tx_swbd->is_eof = 1;
+	}
+
+	txbd_tmp.addr = cpu_to_le64(addr);
+	txbd_tmp.buf_len = cpu_to_le16(size);
+	txbd_tmp.flags = flags;
+
+	tx_swbd->dma = addr;
+	tx_swbd->len = size;
+	tx_swbd->dir = DMA_TO_DEVICE;
+
+	*txbd = txbd_tmp;
+
+	return 0;
+}
+
+static __wsum enetc_tso_hdr_csum(struct tso_t *tso, struct sk_buff *skb,
+				 char *hdr, int hdr_len, int *l4_hdr_len)
+{
+	char *l4_hdr = hdr + skb_transport_offset(skb);
+	int mac_hdr_len = skb_network_offset(skb);
+
+	if (tso->tlen != sizeof(struct udphdr)) {
+		struct tcphdr *tcph = (struct tcphdr *)(l4_hdr);
+
+		tcph->check = 0;
+	} else {
+		struct udphdr *udph = (struct udphdr *)(l4_hdr);
+
+		udph->check = 0;
+	}
+
+	/* Compute the IP checksum. This is necessary since tso_build_hdr()
+	 * already incremented the IP ID field.
+	 */
+	if (!tso->ipv6) {
+		struct iphdr *iph = (void *)(hdr + mac_hdr_len);
+
+		iph->check = 0;
+		iph->check = ip_fast_csum((unsigned char *)iph, iph->ihl);
+	}
+
+	/* Compute the checksum over the L4 header. */
+	*l4_hdr_len = hdr_len - skb_transport_offset(skb);
+	return csum_partial(l4_hdr, *l4_hdr_len, 0);
+}
+
+static void enetc_tso_complete_csum(struct enetc_bdr *tx_ring, struct tso_t *tso,
+				    struct sk_buff *skb, char *hdr, int len,
+				    __wsum sum)
+{
+	char *l4_hdr = hdr + skb_transport_offset(skb);
+	__sum16 csum_final;
+
+	/* Complete the L4 checksum by appending the pseudo-header to the
+	 * already computed checksum.
+	 */
+	if (!tso->ipv6)
+		csum_final = csum_tcpudp_magic(ip_hdr(skb)->saddr,
+					       ip_hdr(skb)->daddr,
+					       len, ip_hdr(skb)->protocol, sum);
+	else
+		csum_final = csum_ipv6_magic(&ipv6_hdr(skb)->saddr,
+					     &ipv6_hdr(skb)->daddr,
+					     len, ipv6_hdr(skb)->nexthdr, sum);
+
+	if (tso->tlen != sizeof(struct udphdr)) {
+		struct tcphdr *tcph = (struct tcphdr *)(l4_hdr);
+
+		tcph->check = csum_final;
+	} else {
+		struct udphdr *udph = (struct udphdr *)(l4_hdr);
+
+		udph->check = csum_final;
+	}
+}
+
+static int enetc_lso_count_descs(const struct sk_buff *skb)
+{
+	/* 4 BDs: 1 BD for LSO header + 1 BD for extended BD + 1 BD
+	 * for linear area data but not include LSO header, namely
+	 * skb_headlen(skb) - lso_hdr_len (it may be 0, but that's
+	 * okay, we only need to consider the worst case). And 1 BD
+	 * for gap.
+	 */
+	return skb_shinfo(skb)->nr_frags + 4;
+}
+
+static int enetc_lso_get_hdr_len(const struct sk_buff *skb)
+{
+	int hdr_len, tlen;
+
+	tlen = skb_is_gso_tcp(skb) ? tcp_hdrlen(skb) : sizeof(struct udphdr);
+	hdr_len = skb_transport_offset(skb) + tlen;
+
+	return hdr_len;
+}
+
+static void enetc_lso_start(struct sk_buff *skb, struct enetc_lso_t *lso)
+{
+	lso->lso_seg_size = skb_shinfo(skb)->gso_size;
+	lso->ipv6 = enetc_skb_is_ipv6(skb);
+	lso->tcp = skb_is_gso_tcp(skb);
+	lso->l3_hdr_len = skb_network_header_len(skb);
+	lso->l3_start = skb_network_offset(skb);
+	lso->hdr_len = enetc_lso_get_hdr_len(skb);
+	lso->total_len = skb->len - lso->hdr_len;
+}
+
+static void enetc_lso_map_hdr(struct enetc_bdr *tx_ring, struct sk_buff *skb,
+			      int *i, struct enetc_lso_t *lso)
+{
+	union enetc_tx_bd txbd_tmp, *txbd;
+	struct enetc_tx_swbd *tx_swbd;
+	u16 frm_len, frm_len_ext;
+	u8 flags, e_flags = 0;
+	dma_addr_t addr;
+	char *hdr;
+
+	/* Get the first BD of the LSO BDs chain */
+	txbd = ENETC_TXBD(*tx_ring, *i);
+	tx_swbd = &tx_ring->tx_swbd[*i];
+	prefetchw(txbd);
+
+	/* Prepare LSO header: MAC + IP + TCP/UDP */
+	hdr = tx_ring->tso_headers + *i * TSO_HEADER_SIZE;
+	memcpy(hdr, skb->data, lso->hdr_len);
+	addr = tx_ring->tso_headers_dma + *i * TSO_HEADER_SIZE;
+
+	/* {frm_len_ext, frm_len} indicates the total length of
+	 * large transmit data unit. frm_len contains the 16 least
+	 * significant bits and frm_len_ext contains the 4 most
+	 * significant bits.
+	 */
+	frm_len = lso->total_len & 0xffff;
+	frm_len_ext = (lso->total_len >> 16) & 0xf;
+
+	/* Set the flags of the first BD */
+	flags = ENETC_TXBD_FLAGS_EX | ENETC_TXBD_FLAGS_CSUM_LSO |
+		ENETC_TXBD_FLAGS_LSO | ENETC_TXBD_FLAGS_L4CS;
+
+	enetc_clear_tx_bd(&txbd_tmp);
+	txbd_tmp.addr = cpu_to_le64(addr);
+	txbd_tmp.hdr_len = cpu_to_le16(lso->hdr_len);
+
+	/* first BD needs frm_len and offload flags set */
+	txbd_tmp.frm_len = cpu_to_le16(frm_len);
+	txbd_tmp.flags = flags;
+
+	txbd_tmp.l3_aux0 = FIELD_PREP(ENETC_TX_BD_L3_START, lso->l3_start);
+	/* l3_hdr_size in 32-bits (4 bytes) */
+	txbd_tmp.l3_aux1 = FIELD_PREP(ENETC_TX_BD_L3_HDR_LEN,
+				      lso->l3_hdr_len / 4);
+	if (lso->ipv6)
+		txbd_tmp.l3_aux1 |= ENETC_TX_BD_L3T;
+	else
+		txbd_tmp.l3_aux0 |= ENETC_TX_BD_IPCS;
+
+	txbd_tmp.l4_aux = FIELD_PREP(ENETC_TX_BD_L4T, lso->tcp ?
+				     ENETC_TXBD_L4T_TCP : ENETC_TXBD_L4T_UDP);
+
+	/* For the LSO header we do not set the dma address since
+	 * we do not want it unmapped when we do cleanup. We still
+	 * set len so that we count the bytes sent.
+	 */
+	tx_swbd->len = lso->hdr_len;
+	tx_swbd->do_twostep_tstamp = false;
+	tx_swbd->check_wb = false;
+
+	/* Actually write the header in the BD */
+	*txbd = txbd_tmp;
+
+	/* Get the next BD, and the next BD is extended BD */
+	enetc_bdr_idx_inc(tx_ring, i);
+	txbd = ENETC_TXBD(*tx_ring, *i);
+	tx_swbd = &tx_ring->tx_swbd[*i];
+	prefetchw(txbd);
+
+	enetc_clear_tx_bd(&txbd_tmp);
+	if (skb_vlan_tag_present(skb)) {
+		/* Setup the VLAN fields */
+		txbd_tmp.ext.vid = cpu_to_le16(skb_vlan_tag_get(skb));
+		txbd_tmp.ext.tpid = ENETC_TPID_8021Q;
+		e_flags = ENETC_TXBD_E_FLAGS_VLAN_INS;
+	}
+
+	/* Write the BD */
+	txbd_tmp.ext.e_flags = e_flags;
+	txbd_tmp.ext.lso_sg_size = cpu_to_le16(lso->lso_seg_size);
+	txbd_tmp.ext.frm_len_ext = cpu_to_le16(frm_len_ext);
+	*txbd = txbd_tmp;
+}
+
+static int enetc_lso_map_data(struct enetc_bdr *tx_ring, struct sk_buff *skb,
+			      int *i, struct enetc_lso_t *lso, int *count)
+{
+	union enetc_tx_bd txbd_tmp, *txbd = NULL;
+	struct enetc_tx_swbd *tx_swbd;
+	skb_frag_t *frag;
+	dma_addr_t dma;
+	u8 flags = 0;
+	int len, f;
+
+	len = skb_headlen(skb) - lso->hdr_len;
+	if (len > 0) {
+		dma = dma_map_single(tx_ring->dev, skb->data + lso->hdr_len,
+				     len, DMA_TO_DEVICE);
+		if (dma_mapping_error(tx_ring->dev, dma))
+			return -ENOMEM;
+
+		enetc_bdr_idx_inc(tx_ring, i);
+		txbd = ENETC_TXBD(*tx_ring, *i);
+		tx_swbd = &tx_ring->tx_swbd[*i];
+		prefetchw(txbd);
+		*count += 1;
+
+		enetc_clear_tx_bd(&txbd_tmp);
+		txbd_tmp.addr = cpu_to_le64(dma);
+		txbd_tmp.buf_len = cpu_to_le16(len);
+
+		tx_swbd->dma = dma;
+		tx_swbd->len = len;
+		tx_swbd->is_dma_page = 0;
+		tx_swbd->dir = DMA_TO_DEVICE;
+	}
+
+	frag = &skb_shinfo(skb)->frags[0];
+	for (f = 0; f < skb_shinfo(skb)->nr_frags; f++, frag++) {
+		if (txbd)
+			*txbd = txbd_tmp;
+
+		len = skb_frag_size(frag);
+		dma = skb_frag_dma_map(tx_ring->dev, frag);
+		if (dma_mapping_error(tx_ring->dev, dma))
+			return -ENOMEM;
+
+		/* Get the next BD */
+		enetc_bdr_idx_inc(tx_ring, i);
+		txbd = ENETC_TXBD(*tx_ring, *i);
+		tx_swbd = &tx_ring->tx_swbd[*i];
+		prefetchw(txbd);
+		*count += 1;
+
+		enetc_clear_tx_bd(&txbd_tmp);
+		txbd_tmp.addr = cpu_to_le64(dma);
+		txbd_tmp.buf_len = cpu_to_le16(len);
+
+		tx_swbd->dma = dma;
+		tx_swbd->len = len;
+		tx_swbd->is_dma_page = 1;
+		tx_swbd->dir = DMA_TO_DEVICE;
+	}
+
+	/* Last BD needs 'F' bit set */
+	flags |= ENETC_TXBD_FLAGS_F;
+	txbd_tmp.flags = flags;
+	*txbd = txbd_tmp;
+
+	tx_swbd->is_eof = 1;
+	tx_swbd->skb = skb;
+
+	return 0;
+}
+
+static int enetc_lso_hw_offload(struct enetc_bdr *tx_ring, struct sk_buff *skb)
+{
+	struct enetc_tx_swbd *tx_swbd;
+	struct enetc_lso_t lso = {0};
+	int err, i, count = 0;
+
+	/* Initialize the LSO handler */
+	enetc_lso_start(skb, &lso);
+	i = tx_ring->next_to_use;
+
+	enetc_lso_map_hdr(tx_ring, skb, &i, &lso);
+	/* First BD and an extend BD */
+	count += 2;
+
+	err = enetc_lso_map_data(tx_ring, skb, &i, &lso, &count);
+	if (err)
+		goto dma_err;
+
+	/* Go to the next BD */
+	enetc_bdr_idx_inc(tx_ring, &i);
+	tx_ring->next_to_use = i;
+	enetc_update_tx_ring_tail(tx_ring);
+
+	return count;
+
+dma_err:
+	do {
+		tx_swbd = &tx_ring->tx_swbd[i];
+		enetc_free_tx_frame(tx_ring, tx_swbd);
+		if (i == 0)
+			i = tx_ring->bd_count;
+		i--;
+	} while (--count);
+
+	return 0;
+}
+
+static int enetc_map_tx_tso_buffs(struct enetc_bdr *tx_ring, struct sk_buff *skb)
+{
+	struct enetc_ndev_priv *priv = netdev_priv(tx_ring->ndev);
+	int hdr_len, total_len, data_len;
+	struct enetc_tx_swbd *tx_swbd;
+	union enetc_tx_bd *txbd;
+	struct tso_t tso;
+	__wsum csum, csum2;
+	int count = 0, pos;
+	int err, i, bd_data_num;
+
+	/* Initialize the TSO handler, and prepare the first payload */
+	hdr_len = tso_start(skb, &tso);
+	total_len = skb->len - hdr_len;
+	i = tx_ring->next_to_use;
+
+	while (total_len > 0) {
+		char *hdr;
+
+		/* Get the BD */
+		txbd = ENETC_TXBD(*tx_ring, i);
+		tx_swbd = &tx_ring->tx_swbd[i];
+		prefetchw(txbd);
+
+		/* Determine the length of this packet */
+		data_len = min_t(int, skb_shinfo(skb)->gso_size, total_len);
+		total_len -= data_len;
+
+		/* prepare packet headers: MAC + IP + TCP */
+		hdr = tx_ring->tso_headers + i * TSO_HEADER_SIZE;
+		tso_build_hdr(skb, hdr, &tso, data_len, total_len == 0);
+
+		/* compute the csum over the L4 header */
+		csum = enetc_tso_hdr_csum(&tso, skb, hdr, hdr_len, &pos);
+		count += enetc_map_tx_tso_hdr(tx_ring, skb, tx_swbd, txbd,
+					      &i, hdr_len, data_len);
+		bd_data_num = 0;
+
+		while (data_len > 0) {
+			int size;
+
+			size = min_t(int, tso.size, data_len);
+
+			/* Advance the index in the BDR */
+			enetc_bdr_idx_inc(tx_ring, &i);
+			txbd = ENETC_TXBD(*tx_ring, i);
+			tx_swbd = &tx_ring->tx_swbd[i];
+			prefetchw(txbd);
+
+			/* Compute the checksum over this segment of data and
+			 * add it to the csum already computed (over the L4
+			 * header and possible other data segments).
+			 */
+			csum2 = csum_partial(tso.data, size, 0);
+			csum = csum_block_add(csum, csum2, pos);
+			pos += size;
+
+			err = enetc_map_tx_tso_data(tx_ring, skb, tx_swbd, txbd,
+						    tso.data, size,
+						    size == data_len);
+			if (err) {
+				if (i == 0)
+					i = tx_ring->bd_count;
+				i--;
+
+				goto err_map_data;
+			}
+
+			data_len -= size;
+			count++;
+			bd_data_num++;
+			tso_build_data(skb, &tso, size);
+
+			if (unlikely(bd_data_num >= priv->max_frags && data_len))
+				goto err_chained_bd;
+		}
+
+		enetc_tso_complete_csum(tx_ring, &tso, skb, hdr, pos, csum);
+
+		if (total_len == 0)
+			tx_swbd->skb = skb;
+
+		/* Go to the next BD */
+		enetc_bdr_idx_inc(tx_ring, &i);
+	}
+
+	tx_ring->next_to_use = i;
+	enetc_update_tx_ring_tail(tx_ring);
+
+	return count;
+
+err_map_data:
+	dev_err(tx_ring->dev, "DMA map error");
+
+err_chained_bd:
+	enetc_unwind_tx_frame(tx_ring, count, i);
+
+	return 0;
+}
+
+static netdev_tx_t enetc_start_xmit(struct sk_buff *skb,
+				    struct net_device *ndev)
+{
+	struct enetc_skb_cb *enetc_cb = ENETC_SKB_CB(skb);
+	struct enetc_ndev_priv *priv = netdev_priv(ndev);
+	struct enetc_bdr *tx_ring;
+	int count;
+
+	/* Queue one-step Sync packet if already locked */
+	if (enetc_cb->flag & ENETC_F_TX_ONESTEP_SYNC_TSTAMP) {
+		if (test_and_set_bit_lock(ENETC_TX_ONESTEP_TSTAMP_IN_PROGRESS,
+					  &priv->flags)) {
+			skb_queue_tail(&priv->tx_skbs, skb);
+			return NETDEV_TX_OK;
+		}
+	}
+
+	tx_ring = priv->tx_ring[skb->queue_mapping];
+
+	if (skb_is_gso(skb)) {
+		/* LSO data unit lengths of up to 256KB are supported */
+		if (priv->active_offloads & ENETC_F_LSO &&
+		    (skb->len - enetc_lso_get_hdr_len(skb)) <=
+		    ENETC_LSO_MAX_DATA_LEN) {
+			if (enetc_bd_unused(tx_ring) < enetc_lso_count_descs(skb)) {
+				netif_stop_subqueue(ndev, tx_ring->index);
+				return NETDEV_TX_BUSY;
+			}
+
+			count = enetc_lso_hw_offload(tx_ring, skb);
+		} else {
+			if (enetc_bd_unused(tx_ring) < tso_count_descs(skb)) {
+				netif_stop_subqueue(ndev, tx_ring->index);
+				return NETDEV_TX_BUSY;
+			}
+
+			enetc_lock_mdio();
+			count = enetc_map_tx_tso_buffs(tx_ring, skb);
+			enetc_unlock_mdio();
+		}
+	} else {
+		if (unlikely(skb_shinfo(skb)->nr_frags > priv->max_frags))
+			if (unlikely(skb_linearize(skb)))
+				goto drop_packet_err;
+
+		count = skb_shinfo(skb)->nr_frags + 1; /* fragments + head */
+		if (enetc_bd_unused(tx_ring) < ENETC_TXBDS_NEEDED(count)) {
+			netif_stop_subqueue(ndev, tx_ring->index);
+			return NETDEV_TX_BUSY;
+		}
+
+		enetc_lock_mdio();
+		count = enetc_map_tx_buffs(tx_ring, skb);
+		enetc_unlock_mdio();
+	}
+
+	if (unlikely(!count))
+		goto drop_packet_err;
+
+	if (enetc_bd_unused(tx_ring) < ENETC_TXBDS_MAX_NEEDED(priv->max_frags))
+		netif_stop_subqueue(ndev, tx_ring->index);
+
+	return NETDEV_TX_OK;
+
+drop_packet_err:
+	dev_kfree_skb_any(skb);
+	return NETDEV_TX_OK;
+}
+
+netdev_tx_t enetc_xmit(struct sk_buff *skb, struct net_device *ndev)
+{
+	struct enetc_skb_cb *enetc_cb = ENETC_SKB_CB(skb);
+	struct enetc_ndev_priv *priv = netdev_priv(ndev);
+	u8 udp, msgtype, twostep;
+	u16 offset1, offset2;
+
+	/* Mark tx timestamp type on enetc_cb->flag if requires */
+	if ((skb_shinfo(skb)->tx_flags & SKBTX_HW_TSTAMP) &&
+	    (priv->active_offloads & ENETC_F_TX_TSTAMP_MASK))
+		enetc_cb->flag = priv->active_offloads & ENETC_F_TX_TSTAMP_MASK;
+	else
+		enetc_cb->flag = 0;
+
+	/* Fall back to two-step timestamp if not one-step Sync packet */
+	if (enetc_cb->flag & ENETC_F_TX_ONESTEP_SYNC_TSTAMP) {
+		if (enetc_ptp_parse(skb, &udp, &msgtype, &twostep,
+				    &offset1, &offset2) ||
+		    msgtype != PTP_MSGTYPE_SYNC || twostep != 0) {
+			enetc_cb->flag = ENETC_F_TX_TSTAMP;
+		} else {
+			enetc_cb->udp = !!udp;
+			enetc_cb->correction_off = offset1;
+			enetc_cb->origin_tstamp_off = offset2;
+		}
+	}
+
+	return enetc_start_xmit(skb, ndev);
+}
+EXPORT_SYMBOL_GPL(enetc_xmit);
+
+static irqreturn_t enetc_msix(int irq, void *data)
+{
+	struct enetc_int_vector	*v = data;
+	int i;
+
+	enetc_lock_mdio();
+
+	/* disable interrupts */
+	enetc_wr_reg_hot(v->rbier, 0);
+	enetc_wr_reg_hot(v->ricr1, v->rx_ictt);
+
+	for_each_set_bit(i, &v->tx_rings_map, ENETC_MAX_NUM_TXQS)
+		enetc_wr_reg_hot(v->tbier_base + ENETC_BDR_OFF(i), 0);
+
+	enetc_unlock_mdio();
+
+	napi_schedule(&v->napi);
+
+	return IRQ_HANDLED;
+}
+
+static void enetc_rx_dim_work(struct work_struct *w)
+{
+	struct dim *dim = container_of(w, struct dim, work);
+	struct dim_cq_moder moder =
+		net_dim_get_rx_moderation(dim->mode, dim->profile_ix);
+	struct enetc_int_vector	*v =
+		container_of(dim, struct enetc_int_vector, rx_dim);
+	struct enetc_ndev_priv *priv = netdev_priv(v->rx_ring.ndev);
+
+	v->rx_ictt = enetc_usecs_to_cycles(moder.usec, priv->sysclk_freq);
+	dim->state = DIM_START_MEASURE;
+}
+
+static void enetc_rx_net_dim(struct enetc_int_vector *v)
+{
+	struct dim_sample dim_sample = {};
+
+	v->comp_cnt++;
+
+	if (!v->rx_napi_work)
+		return;
+
+	dim_update_sample(v->comp_cnt,
+			  v->rx_ring.stats.packets,
+			  v->rx_ring.stats.bytes,
+			  &dim_sample);
+	net_dim(&v->rx_dim, &dim_sample);
+}
+
+static int enetc_bd_ready_count(struct enetc_bdr *tx_ring, int ci)
+{
+	int pi = enetc_rd_reg_hot(tx_ring->tcir) & ENETC_TBCIR_IDX_MASK;
+
+	return pi >= ci ? pi - ci : tx_ring->bd_count - ci + pi;
+}
+
+static bool enetc_page_reusable(struct page *page)
+{
+	return (!page_is_pfmemalloc(page) && page_ref_count(page) == 1);
+}
+
+static void enetc_reuse_page(struct enetc_bdr *rx_ring,
+			     struct enetc_rx_swbd *old)
+{
+	struct enetc_rx_swbd *new;
+
+	new = &rx_ring->rx_swbd[rx_ring->next_to_alloc];
+
+	/* next buf that may reuse a page */
+	enetc_bdr_idx_inc(rx_ring, &rx_ring->next_to_alloc);
+
+	/* copy page reference */
+	*new = *old;
+}
+
+static void enetc_get_tx_tstamp(struct enetc_hw *hw, union enetc_tx_bd *txbd,
+				u64 *tstamp)
+{
+	u32 lo, hi, tstamp_lo;
+
+	lo = enetc_rd_hot(hw, ENETC_SICTR0);
+	hi = enetc_rd_hot(hw, ENETC_SICTR1);
+	tstamp_lo = le32_to_cpu(txbd->wb.tstamp);
+	if (lo <= tstamp_lo)
+		hi -= 1;
+	*tstamp = (u64)hi << 32 | tstamp_lo;
+}
+
+static void enetc_tstamp_tx(struct sk_buff *skb, u64 tstamp)
+{
+	struct skb_shared_hwtstamps shhwtstamps;
+
+	if (skb_shinfo(skb)->tx_flags & SKBTX_IN_PROGRESS) {
+		memset(&shhwtstamps, 0, sizeof(shhwtstamps));
+		shhwtstamps.hwtstamp = ns_to_ktime(tstamp);
+		skb_txtime_consumed(skb);
+		skb_tstamp_tx(skb, &shhwtstamps);
+	}
+}
+
+static void enetc_recycle_xdp_tx_buff(struct enetc_bdr *tx_ring,
+				      struct enetc_tx_swbd *tx_swbd)
+{
+	struct enetc_ndev_priv *priv = netdev_priv(tx_ring->ndev);
+	struct enetc_rx_swbd rx_swbd = {
+		.dma = tx_swbd->dma,
+		.page = tx_swbd->page,
+		.page_offset = tx_swbd->page_offset,
+		.dir = tx_swbd->dir,
+		.len = tx_swbd->len,
+	};
+	struct enetc_bdr *rx_ring;
+
+	rx_ring = enetc_rx_ring_from_xdp_tx_ring(priv, tx_ring);
+
+	if (likely(enetc_swbd_unused(rx_ring))) {
+		enetc_reuse_page(rx_ring, &rx_swbd);
+
+		/* sync for use by the device */
+		dma_sync_single_range_for_device(rx_ring->dev, rx_swbd.dma,
+						 rx_swbd.page_offset,
+						 ENETC_RXB_DMA_SIZE_XDP,
+						 rx_swbd.dir);
+
+		rx_ring->stats.recycles++;
+	} else {
+		/* RX ring is already full, we need to unmap and free the
+		 * page, since there's nothing useful we can do with it.
+		 */
+		rx_ring->stats.recycle_failures++;
+
+		dma_unmap_page(rx_ring->dev, rx_swbd.dma, PAGE_SIZE,
+			       rx_swbd.dir);
+		__free_page(rx_swbd.page);
+	}
+
+	rx_ring->xdp.xdp_tx_in_flight--;
+}
+
+static bool enetc_clean_tx_ring(struct enetc_bdr *tx_ring, int napi_budget)
+{
+	int tx_frm_cnt = 0, tx_byte_cnt = 0, tx_win_drop = 0;
+	struct net_device *ndev = tx_ring->ndev;
+	struct enetc_ndev_priv *priv = netdev_priv(ndev);
+	struct enetc_tx_swbd *tx_swbd;
+	int i, bds_to_clean;
+	bool do_twostep_tstamp;
+	u64 tstamp = 0;
+
+	i = tx_ring->next_to_clean;
+	tx_swbd = &tx_ring->tx_swbd[i];
+
+	bds_to_clean = enetc_bd_ready_count(tx_ring, i);
+
+	do_twostep_tstamp = false;
+
+	while (bds_to_clean && tx_frm_cnt < ENETC_DEFAULT_TX_WORK) {
+		struct xdp_frame *xdp_frame = enetc_tx_swbd_get_xdp_frame(tx_swbd);
+		struct sk_buff *skb = enetc_tx_swbd_get_skb(tx_swbd);
+		bool is_eof = tx_swbd->is_eof;
+
+		if (unlikely(tx_swbd->check_wb)) {
+			union enetc_tx_bd *txbd = ENETC_TXBD(*tx_ring, i);
+
+			if (txbd->flags & ENETC_TXBD_FLAGS_W &&
+			    tx_swbd->do_twostep_tstamp) {
+				enetc_get_tx_tstamp(&priv->si->hw, txbd,
+						    &tstamp);
+				do_twostep_tstamp = true;
+			}
+
+			if (tx_swbd->qbv_en &&
+			    txbd->wb.status & ENETC_TXBD_STATS_WIN)
+				tx_win_drop++;
+		}
+
+		if (tx_swbd->is_xdp_tx)
+			enetc_recycle_xdp_tx_buff(tx_ring, tx_swbd);
+		else if (likely(tx_swbd->dma))
+			enetc_unmap_tx_buff(tx_ring, tx_swbd);
+
+		if (xdp_frame) {
+			xdp_return_frame(xdp_frame);
+		} else if (skb) {
+			struct enetc_skb_cb *enetc_cb = ENETC_SKB_CB(skb);
+
+			if (unlikely(enetc_cb->flag & ENETC_F_TX_ONESTEP_SYNC_TSTAMP)) {
+				/* Start work to release lock for next one-step
+				 * timestamping packet. And send one skb in
+				 * tx_skbs queue if has.
+				 */
+				schedule_work(&priv->tx_onestep_tstamp);
+			} else if (unlikely(do_twostep_tstamp)) {
+				enetc_tstamp_tx(skb, tstamp);
+				do_twostep_tstamp = false;
+			}
+			napi_consume_skb(skb, napi_budget);
+		}
+
+		tx_byte_cnt += tx_swbd->len;
+		/* Scrub the swbd here so we don't have to do that
+		 * when we reuse it during xmit
+		 */
+		memset(tx_swbd, 0, sizeof(*tx_swbd));
+
+		bds_to_clean--;
+		tx_swbd++;
+		i++;
+		if (unlikely(i == tx_ring->bd_count)) {
+			i = 0;
+			tx_swbd = tx_ring->tx_swbd;
+		}
+
+		/* BD iteration loop end */
+		if (is_eof) {
+			tx_frm_cnt++;
+			/* re-arm interrupt source */
+			enetc_wr_reg_hot(tx_ring->idr, BIT(tx_ring->index) |
+					 BIT(16 + tx_ring->index));
+		}
+
+		if (unlikely(!bds_to_clean))
+			bds_to_clean = enetc_bd_ready_count(tx_ring, i);
+	}
+
+	tx_ring->next_to_clean = i;
+	tx_ring->stats.packets += tx_frm_cnt;
+	tx_ring->stats.bytes += tx_byte_cnt;
+	tx_ring->stats.win_drop += tx_win_drop;
+
+	if (unlikely(tx_frm_cnt && netif_carrier_ok(ndev) &&
+		     __netif_subqueue_stopped(ndev, tx_ring->index) &&
+		     !test_bit(ENETC_TX_DOWN, &priv->flags) &&
+		     (enetc_bd_unused(tx_ring) >=
+		      ENETC_TXBDS_MAX_NEEDED(priv->max_frags)))) {
+		netif_wake_subqueue(ndev, tx_ring->index);
+	}
+
+	return tx_frm_cnt != ENETC_DEFAULT_TX_WORK;
+}
+
+static bool enetc_new_page(struct enetc_bdr *rx_ring,
+			   struct enetc_rx_swbd *rx_swbd)
+{
+	bool xdp = !!(rx_ring->xdp.prog);
+	struct page *page;
+	dma_addr_t addr;
+
+	page = dev_alloc_page();
+	if (unlikely(!page))
+		return false;
+
+	/* For XDP_TX, we forgo dma_unmap -> dma_map */
+	rx_swbd->dir = xdp ? DMA_BIDIRECTIONAL : DMA_FROM_DEVICE;
+
+	addr = dma_map_page(rx_ring->dev, page, 0, PAGE_SIZE, rx_swbd->dir);
+	if (unlikely(dma_mapping_error(rx_ring->dev, addr))) {
+		__free_page(page);
+
+		return false;
+	}
+
+	rx_swbd->dma = addr;
+	rx_swbd->page = page;
+	rx_swbd->page_offset = rx_ring->buffer_offset;
+
+	return true;
+}
+
+static int enetc_refill_rx_ring(struct enetc_bdr *rx_ring, const int buff_cnt)
+{
+	struct enetc_rx_swbd *rx_swbd;
+	union enetc_rx_bd *rxbd;
+	int i, j;
+
+	i = rx_ring->next_to_use;
+	rx_swbd = &rx_ring->rx_swbd[i];
+	rxbd = enetc_rxbd(rx_ring, i);
+
+	for (j = 0; j < buff_cnt; j++) {
+		/* try reuse page */
+		if (unlikely(!rx_swbd->page)) {
+			if (unlikely(!enetc_new_page(rx_ring, rx_swbd))) {
+				rx_ring->stats.rx_alloc_errs++;
+				break;
+			}
+		}
+
+		/* update RxBD */
+		rxbd->w.addr = cpu_to_le64(rx_swbd->dma +
+					   rx_swbd->page_offset);
+		/* clear 'R" as well */
+		rxbd->r.lstatus = 0;
+
+		enetc_rxbd_next(rx_ring, &rxbd, &i);
+		rx_swbd = &rx_ring->rx_swbd[i];
+	}
+
+	if (likely(j)) {
+		rx_ring->next_to_alloc = i; /* keep track from page reuse */
+		rx_ring->next_to_use = i;
+
+		/* update ENETC's consumer index */
+		enetc_wr_reg_hot(rx_ring->rcir, rx_ring->next_to_use);
+	}
+
+	return j;
+}
+
+static void enetc_get_rx_tstamp(struct net_device *ndev,
+				union enetc_rx_bd *rxbd,
+				struct sk_buff *skb)
+{
+	struct skb_shared_hwtstamps *shhwtstamps = skb_hwtstamps(skb);
+	struct enetc_ndev_priv *priv = netdev_priv(ndev);
+	struct enetc_hw *hw = &priv->si->hw;
+	u32 lo, hi, tstamp_lo;
+	u64 tstamp;
+
+	if (le16_to_cpu(rxbd->r.flags) & ENETC_RXBD_FLAG_TSTMP) {
+		lo = enetc_rd_reg_hot(hw->reg + ENETC_SICTR0);
+		hi = enetc_rd_reg_hot(hw->reg + ENETC_SICTR1);
+		rxbd = enetc_rxbd_ext(rxbd);
+		tstamp_lo = le32_to_cpu(rxbd->ext.tstamp);
+		if (lo <= tstamp_lo)
+			hi -= 1;
+
+		tstamp = (u64)hi << 32 | tstamp_lo;
+		memset(shhwtstamps, 0, sizeof(*shhwtstamps));
+		shhwtstamps->hwtstamp = ns_to_ktime(tstamp);
+	}
+}
+
+static void enetc_get_offloads(struct enetc_bdr *rx_ring,
+			       union enetc_rx_bd *rxbd, struct sk_buff *skb)
+{
+	struct enetc_ndev_priv *priv = netdev_priv(rx_ring->ndev);
+
+	/* TODO: hashing */
+	if (rx_ring->ndev->features & NETIF_F_RXCSUM) {
+		u16 inet_csum = le16_to_cpu(rxbd->r.inet_csum);
+
+		skb->csum = csum_unfold((__force __sum16)~htons(inet_csum));
+		skb->ip_summed = CHECKSUM_COMPLETE;
+	}
+
+	if (le16_to_cpu(rxbd->r.flags) & ENETC_RXBD_FLAG_VLAN) {
+		struct enetc_hw *hw = &priv->si->hw;
+		__be16 tpid = 0;
+
+		switch (le16_to_cpu(rxbd->r.flags) & ENETC_RXBD_FLAG_TPID) {
+		case 0:
+			tpid = htons(ETH_P_8021Q);
+			break;
+		case 1:
+			tpid = htons(ETH_P_8021AD);
+			break;
+		case 2:
+			tpid = htons(enetc_rd_hot(hw, ENETC_SICVLANR1) &
+				     SICVLANR_ETYPE);
+			break;
+		case 3:
+			tpid = htons(enetc_rd_hot(hw, ENETC_SICVLANR2) &
+				     SICVLANR_ETYPE);
+		}
+
+		__vlan_hwaccel_put_tag(skb, tpid, le16_to_cpu(rxbd->r.vlan_opt));
+	}
+
+	if (priv->active_offloads & ENETC_F_RX_TSTAMP)
+		enetc_get_rx_tstamp(rx_ring->ndev, rxbd, skb);
+}
+
+/* This gets called during the non-XDP NAPI poll cycle as well as on XDP_PASS,
+ * so it needs to work with both DMA_FROM_DEVICE as well as DMA_BIDIRECTIONAL
+ * mapped buffers.
+ */
+static struct enetc_rx_swbd *enetc_get_rx_buff(struct enetc_bdr *rx_ring,
+					       int i, u16 size)
+{
+	struct enetc_rx_swbd *rx_swbd = &rx_ring->rx_swbd[i];
+
+	dma_sync_single_range_for_cpu(rx_ring->dev, rx_swbd->dma,
+				      rx_swbd->page_offset,
+				      size, rx_swbd->dir);
+	return rx_swbd;
+}
+
+/* Reuse the current page without performing half-page buffer flipping */
+static void enetc_put_rx_buff(struct enetc_bdr *rx_ring,
+			      struct enetc_rx_swbd *rx_swbd)
+{
+	size_t buffer_size = ENETC_RXB_TRUESIZE - rx_ring->buffer_offset;
+
+	enetc_reuse_page(rx_ring, rx_swbd);
+
+	dma_sync_single_range_for_device(rx_ring->dev, rx_swbd->dma,
+					 rx_swbd->page_offset,
+					 buffer_size, rx_swbd->dir);
+
+	rx_swbd->page = NULL;
+}
+
+/* Reuse the current page by performing half-page buffer flipping */
+static void enetc_flip_rx_buff(struct enetc_bdr *rx_ring,
+			       struct enetc_rx_swbd *rx_swbd)
+{
+	if (likely(enetc_page_reusable(rx_swbd->page))) {
+		rx_swbd->page_offset ^= ENETC_RXB_TRUESIZE;
+		page_ref_inc(rx_swbd->page);
+
+		enetc_put_rx_buff(rx_ring, rx_swbd);
+	} else {
+		dma_unmap_page(rx_ring->dev, rx_swbd->dma, PAGE_SIZE,
+			       rx_swbd->dir);
+		rx_swbd->page = NULL;
+	}
+}
+
+static struct sk_buff *enetc_map_rx_buff_to_skb(struct enetc_bdr *rx_ring,
+						int i, u16 size)
+{
+	struct enetc_rx_swbd *rx_swbd = enetc_get_rx_buff(rx_ring, i, size);
+	struct sk_buff *skb;
+	void *ba;
+
+	ba = page_address(rx_swbd->page) + rx_swbd->page_offset;
+	skb = build_skb(ba - rx_ring->buffer_offset, ENETC_RXB_TRUESIZE);
+	if (unlikely(!skb)) {
+		rx_ring->stats.rx_alloc_errs++;
+		return NULL;
+	}
+
+	skb_reserve(skb, rx_ring->buffer_offset);
+	__skb_put(skb, size);
+
+	enetc_flip_rx_buff(rx_ring, rx_swbd);
+
+	return skb;
+}
+
+static void enetc_add_rx_buff_to_skb(struct enetc_bdr *rx_ring, int i,
+				     u16 size, struct sk_buff *skb)
+{
+	struct enetc_rx_swbd *rx_swbd = enetc_get_rx_buff(rx_ring, i, size);
+
+	skb_add_rx_frag(skb, skb_shinfo(skb)->nr_frags, rx_swbd->page,
+			rx_swbd->page_offset, size, ENETC_RXB_TRUESIZE);
+
+	enetc_flip_rx_buff(rx_ring, rx_swbd);
+}
+
+static bool enetc_check_bd_errors_and_consume(struct enetc_bdr *rx_ring,
+					      u32 bd_status,
+					      union enetc_rx_bd **rxbd, int *i)
+{
+	if (likely(!(bd_status & ENETC_RXBD_LSTATUS(ENETC_RXBD_ERR_MASK))))
+		return false;
+
+	enetc_put_rx_buff(rx_ring, &rx_ring->rx_swbd[*i]);
+	enetc_rxbd_next(rx_ring, rxbd, i);
+
+	while (!(bd_status & ENETC_RXBD_LSTATUS_F)) {
+		dma_rmb();
+		bd_status = le32_to_cpu((*rxbd)->r.lstatus);
+
+		enetc_put_rx_buff(rx_ring, &rx_ring->rx_swbd[*i]);
+		enetc_rxbd_next(rx_ring, rxbd, i);
+	}
+
+	rx_ring->ndev->stats.rx_dropped++;
+	rx_ring->ndev->stats.rx_errors++;
+
+	return true;
+}
+
+static struct sk_buff *enetc_build_skb(struct enetc_bdr *rx_ring,
+				       u32 bd_status, union enetc_rx_bd **rxbd,
+				       int *i, int *cleaned_cnt, int buffer_size)
+{
+	struct sk_buff *skb;
+	u16 size;
+
+	size = le16_to_cpu((*rxbd)->r.buf_len);
+	skb = enetc_map_rx_buff_to_skb(rx_ring, *i, size);
+	if (!skb)
+		return NULL;
+
+	enetc_get_offloads(rx_ring, *rxbd, skb);
+
+	(*cleaned_cnt)++;
+
+	enetc_rxbd_next(rx_ring, rxbd, i);
+
+	/* not last BD in frame? */
+	while (!(bd_status & ENETC_RXBD_LSTATUS_F)) {
+		bd_status = le32_to_cpu((*rxbd)->r.lstatus);
+		size = buffer_size;
+
+		if (bd_status & ENETC_RXBD_LSTATUS_F) {
+			dma_rmb();
+			size = le16_to_cpu((*rxbd)->r.buf_len);
+		}
+
+		enetc_add_rx_buff_to_skb(rx_ring, *i, size, skb);
+
+		(*cleaned_cnt)++;
+
+		enetc_rxbd_next(rx_ring, rxbd, i);
+	}
+
+	skb_record_rx_queue(skb, rx_ring->index);
+	skb->protocol = eth_type_trans(skb, rx_ring->ndev);
+
+	return skb;
+}
+
+#define ENETC_RXBD_BUNDLE 16 /* # of BDs to update at once */
+
+static int enetc_clean_rx_ring(struct enetc_bdr *rx_ring,
+			       struct napi_struct *napi, int work_limit)
+{
+	int rx_frm_cnt = 0, rx_byte_cnt = 0;
+	int cleaned_cnt, i;
+
+	cleaned_cnt = enetc_bd_unused(rx_ring);
+	/* next descriptor to process */
+	i = rx_ring->next_to_clean;
+
+	while (likely(rx_frm_cnt < work_limit)) {
+		union enetc_rx_bd *rxbd;
+		struct sk_buff *skb;
+		u32 bd_status;
+
+		if (cleaned_cnt >= ENETC_RXBD_BUNDLE)
+			cleaned_cnt -= enetc_refill_rx_ring(rx_ring,
+							    cleaned_cnt);
+
+		rxbd = enetc_rxbd(rx_ring, i);
+		bd_status = le32_to_cpu(rxbd->r.lstatus);
+		if (!bd_status)
+			break;
+
+		enetc_wr_reg_hot(rx_ring->idr, BIT(rx_ring->index));
+		dma_rmb(); /* for reading other rxbd fields */
+
+		if (enetc_check_bd_errors_and_consume(rx_ring, bd_status,
+						      &rxbd, &i))
+			break;
+
+		skb = enetc_build_skb(rx_ring, bd_status, &rxbd, &i,
+				      &cleaned_cnt, ENETC_RXB_DMA_SIZE);
+		if (!skb)
+			break;
+
+		/* When set, the outer VLAN header is extracted and reported
+		 * in the receive buffer descriptor. So rx_byte_cnt should
+		 * add the length of the extracted VLAN header.
+		 */
+		if (bd_status & ENETC_RXBD_FLAG_VLAN)
+			rx_byte_cnt += VLAN_HLEN;
+		rx_byte_cnt += skb->len + ETH_HLEN;
+		rx_frm_cnt++;
+
+		napi_gro_receive(napi, skb);
+	}
+
+	rx_ring->next_to_clean = i;
+
+	rx_ring->stats.packets += rx_frm_cnt;
+	rx_ring->stats.bytes += rx_byte_cnt;
+
+	return rx_frm_cnt;
+}
+
+static void enetc_xdp_map_tx_buff(struct enetc_bdr *tx_ring, int i,
+				  struct enetc_tx_swbd *tx_swbd,
+				  int frm_len)
+{
+	union enetc_tx_bd *txbd = ENETC_TXBD(*tx_ring, i);
+
+	prefetchw(txbd);
+
+	enetc_clear_tx_bd(txbd);
+	txbd->addr = cpu_to_le64(tx_swbd->dma + tx_swbd->page_offset);
+	txbd->buf_len = cpu_to_le16(tx_swbd->len);
+	txbd->frm_len = cpu_to_le16(frm_len);
+
+	memcpy(&tx_ring->tx_swbd[i], tx_swbd, sizeof(*tx_swbd));
+}
+
+/* Puts in the TX ring one XDP frame, mapped as an array of TX software buffer
+ * descriptors.
+ */
+static bool enetc_xdp_tx(struct enetc_bdr *tx_ring,
+			 struct enetc_tx_swbd *xdp_tx_arr, int num_tx_swbd)
+{
+	struct enetc_tx_swbd *tmp_tx_swbd = xdp_tx_arr;
+	int i, k, frm_len = tmp_tx_swbd->len;
+
+	if (unlikely(enetc_bd_unused(tx_ring) < ENETC_TXBDS_NEEDED(num_tx_swbd)))
+		return false;
+
+	while (unlikely(!tmp_tx_swbd->is_eof)) {
+		tmp_tx_swbd++;
+		frm_len += tmp_tx_swbd->len;
+	}
+
+	i = tx_ring->next_to_use;
+
+	for (k = 0; k < num_tx_swbd; k++) {
+		struct enetc_tx_swbd *xdp_tx_swbd = &xdp_tx_arr[k];
+
+		enetc_xdp_map_tx_buff(tx_ring, i, xdp_tx_swbd, frm_len);
+
+		/* last BD needs 'F' bit set */
+		if (xdp_tx_swbd->is_eof) {
+			union enetc_tx_bd *txbd = ENETC_TXBD(*tx_ring, i);
+
+			txbd->flags = ENETC_TXBD_FLAGS_F;
+		}
+
+		enetc_bdr_idx_inc(tx_ring, &i);
+	}
+
+	tx_ring->next_to_use = i;
+
+	return true;
+}
+
+static int enetc_xdp_frame_to_xdp_tx_swbd(struct enetc_bdr *tx_ring,
+					  struct enetc_tx_swbd *xdp_tx_arr,
+					  struct xdp_frame *xdp_frame)
+{
+	struct enetc_tx_swbd *xdp_tx_swbd = &xdp_tx_arr[0];
+	struct skb_shared_info *shinfo;
+	void *data = xdp_frame->data;
+	int len = xdp_frame->len;
+	skb_frag_t *frag;
+	dma_addr_t dma;
+	unsigned int f;
+	int n = 0;
+
+	dma = dma_map_single(tx_ring->dev, data, len, DMA_TO_DEVICE);
+	if (unlikely(dma_mapping_error(tx_ring->dev, dma))) {
+		netdev_err(tx_ring->ndev, "DMA map error\n");
+		return -1;
+	}
+
+	xdp_tx_swbd->dma = dma;
+	xdp_tx_swbd->dir = DMA_TO_DEVICE;
+	xdp_tx_swbd->len = len;
+	xdp_tx_swbd->is_xdp_redirect = true;
+	xdp_tx_swbd->is_eof = false;
+	xdp_tx_swbd->xdp_frame = NULL;
+
+	n++;
+
+	if (!xdp_frame_has_frags(xdp_frame))
+		goto out;
+
+	xdp_tx_swbd = &xdp_tx_arr[n];
+
+	shinfo = xdp_get_shared_info_from_frame(xdp_frame);
+
+	for (f = 0, frag = &shinfo->frags[0]; f < shinfo->nr_frags;
+	     f++, frag++) {
+		data = skb_frag_address(frag);
+		len = skb_frag_size(frag);
+
+		dma = dma_map_single(tx_ring->dev, data, len, DMA_TO_DEVICE);
+		if (unlikely(dma_mapping_error(tx_ring->dev, dma))) {
+			/* Undo the DMA mapping for all fragments */
+			while (--n >= 0)
+				enetc_unmap_tx_buff(tx_ring, &xdp_tx_arr[n]);
+
+			netdev_err(tx_ring->ndev, "DMA map error\n");
+			return -1;
+		}
+
+		xdp_tx_swbd->dma = dma;
+		xdp_tx_swbd->dir = DMA_TO_DEVICE;
+		xdp_tx_swbd->len = len;
+		xdp_tx_swbd->is_xdp_redirect = true;
+		xdp_tx_swbd->is_eof = false;
+		xdp_tx_swbd->xdp_frame = NULL;
+
+		n++;
+		xdp_tx_swbd = &xdp_tx_arr[n];
+	}
+out:
+	xdp_tx_arr[n - 1].is_eof = true;
+	xdp_tx_arr[n - 1].xdp_frame = xdp_frame;
+
+	return n;
+}
+
+int enetc_xdp_xmit(struct net_device *ndev, int num_frames,
+		   struct xdp_frame **frames, u32 flags)
+{
+	struct enetc_tx_swbd xdp_redirect_arr[ENETC_MAX_SKB_FRAGS] = {0};
+	struct enetc_ndev_priv *priv = netdev_priv(ndev);
+	struct enetc_bdr *tx_ring;
+	int xdp_tx_bd_cnt, i, k;
+	int xdp_tx_frm_cnt = 0;
+
+	if (unlikely(test_bit(ENETC_TX_DOWN, &priv->flags)))
+		return -ENETDOWN;
+
+	enetc_lock_mdio();
+
+	tx_ring = priv->xdp_tx_ring[smp_processor_id()];
+
+	prefetchw(ENETC_TXBD(*tx_ring, tx_ring->next_to_use));
+
+	for (k = 0; k < num_frames; k++) {
+		xdp_tx_bd_cnt = enetc_xdp_frame_to_xdp_tx_swbd(tx_ring,
+							       xdp_redirect_arr,
+							       frames[k]);
+		if (unlikely(xdp_tx_bd_cnt < 0))
+			break;
+
+		if (unlikely(!enetc_xdp_tx(tx_ring, xdp_redirect_arr,
+					   xdp_tx_bd_cnt))) {
+			for (i = 0; i < xdp_tx_bd_cnt; i++)
+				enetc_unmap_tx_buff(tx_ring,
+						    &xdp_redirect_arr[i]);
+			tx_ring->stats.xdp_tx_drops++;
+			break;
+		}
+
+		xdp_tx_frm_cnt++;
+	}
+
+	if (unlikely((flags & XDP_XMIT_FLUSH) || k != xdp_tx_frm_cnt))
+		enetc_update_tx_ring_tail(tx_ring);
+
+	tx_ring->stats.xdp_tx += xdp_tx_frm_cnt;
+
+	enetc_unlock_mdio();
+
+	return xdp_tx_frm_cnt;
+}
+EXPORT_SYMBOL_GPL(enetc_xdp_xmit);
+
+static void enetc_map_rx_buff_to_xdp(struct enetc_bdr *rx_ring, int i,
+				     struct xdp_buff *xdp_buff, u16 size)
+{
+	struct enetc_rx_swbd *rx_swbd = enetc_get_rx_buff(rx_ring, i, size);
+	void *hard_start = page_address(rx_swbd->page) + rx_swbd->page_offset;
+
+	/* To be used for XDP_TX */
+	rx_swbd->len = size;
+
+	xdp_prepare_buff(xdp_buff, hard_start - rx_ring->buffer_offset,
+			 rx_ring->buffer_offset, size, false);
+}
+
+static void enetc_add_rx_buff_to_xdp(struct enetc_bdr *rx_ring, int i,
+				     u16 size, struct xdp_buff *xdp_buff)
+{
+	struct skb_shared_info *shinfo = xdp_get_shared_info_from_buff(xdp_buff);
+	struct enetc_rx_swbd *rx_swbd = enetc_get_rx_buff(rx_ring, i, size);
+	skb_frag_t *frag;
+
+	/* To be used for XDP_TX */
+	rx_swbd->len = size;
+
+	if (!xdp_buff_has_frags(xdp_buff)) {
+		xdp_buff_set_frags_flag(xdp_buff);
+		shinfo->xdp_frags_size = size;
+		shinfo->nr_frags = 0;
+	} else {
+		shinfo->xdp_frags_size += size;
+	}
+
+	if (page_is_pfmemalloc(rx_swbd->page))
+		xdp_buff_set_frag_pfmemalloc(xdp_buff);
+
+	frag = &shinfo->frags[shinfo->nr_frags];
+	skb_frag_fill_page_desc(frag, rx_swbd->page, rx_swbd->page_offset,
+				size);
+
+	shinfo->nr_frags++;
+}
+
+static void enetc_build_xdp_buff(struct enetc_bdr *rx_ring, u32 bd_status,
+				 union enetc_rx_bd **rxbd, int *i,
+				 int *cleaned_cnt, struct xdp_buff *xdp_buff)
+{
+	u16 size = le16_to_cpu((*rxbd)->r.buf_len);
+
+	xdp_init_buff(xdp_buff, ENETC_RXB_TRUESIZE, &rx_ring->xdp.rxq);
+
+	enetc_map_rx_buff_to_xdp(rx_ring, *i, xdp_buff, size);
+	(*cleaned_cnt)++;
+	enetc_rxbd_next(rx_ring, rxbd, i);
+
+	/* not last BD in frame? */
+	while (!(bd_status & ENETC_RXBD_LSTATUS_F)) {
+		bd_status = le32_to_cpu((*rxbd)->r.lstatus);
+		size = ENETC_RXB_DMA_SIZE_XDP;
+
+		if (bd_status & ENETC_RXBD_LSTATUS_F) {
+			dma_rmb();
+			size = le16_to_cpu((*rxbd)->r.buf_len);
+		}
+
+		enetc_add_rx_buff_to_xdp(rx_ring, *i, size, xdp_buff);
+		(*cleaned_cnt)++;
+		enetc_rxbd_next(rx_ring, rxbd, i);
+	}
+}
+
+/* Convert RX buffer descriptors to TX buffer descriptors. These will be
+ * recycled back into the RX ring in enetc_clean_tx_ring.
+ */
+static int enetc_rx_swbd_to_xdp_tx_swbd(struct enetc_tx_swbd *xdp_tx_arr,
+					struct enetc_bdr *rx_ring,
+					int rx_ring_first, int rx_ring_last)
+{
+	int n = 0;
+
+	for (; rx_ring_first != rx_ring_last;
+	     n++, enetc_bdr_idx_inc(rx_ring, &rx_ring_first)) {
+		struct enetc_rx_swbd *rx_swbd = &rx_ring->rx_swbd[rx_ring_first];
+		struct enetc_tx_swbd *tx_swbd = &xdp_tx_arr[n];
+
+		/* No need to dma_map, we already have DMA_BIDIRECTIONAL */
+		tx_swbd->dma = rx_swbd->dma;
+		tx_swbd->dir = rx_swbd->dir;
+		tx_swbd->page = rx_swbd->page;
+		tx_swbd->page_offset = rx_swbd->page_offset;
+		tx_swbd->len = rx_swbd->len;
+		tx_swbd->is_dma_page = true;
+		tx_swbd->is_xdp_tx = true;
+		tx_swbd->is_eof = false;
+	}
+
+	/* We rely on caller providing an rx_ring_last > rx_ring_first */
+	xdp_tx_arr[n - 1].is_eof = true;
+
+	return n;
+}
+
+static void enetc_xdp_drop(struct enetc_bdr *rx_ring, int rx_ring_first,
+			   int rx_ring_last)
+{
+	while (rx_ring_first != rx_ring_last) {
+		enetc_put_rx_buff(rx_ring,
+				  &rx_ring->rx_swbd[rx_ring_first]);
+		enetc_bdr_idx_inc(rx_ring, &rx_ring_first);
+	}
+}
+
+static void enetc_bulk_flip_buff(struct enetc_bdr *rx_ring, int rx_ring_first,
+				 int rx_ring_last)
+{
+	while (rx_ring_first != rx_ring_last) {
+		enetc_flip_rx_buff(rx_ring,
+				   &rx_ring->rx_swbd[rx_ring_first]);
+		enetc_bdr_idx_inc(rx_ring, &rx_ring_first);
+	}
+}
+
+static int enetc_clean_rx_ring_xdp(struct enetc_bdr *rx_ring,
+				   struct napi_struct *napi, int work_limit,
+				   struct bpf_prog *prog)
+{
+	int xdp_tx_bd_cnt, xdp_tx_frm_cnt = 0, xdp_redirect_frm_cnt = 0;
+	struct enetc_tx_swbd xdp_tx_arr[ENETC_MAX_SKB_FRAGS] = {0};
+	struct enetc_ndev_priv *priv = netdev_priv(rx_ring->ndev);
+	int rx_frm_cnt = 0, rx_byte_cnt = 0;
+	struct enetc_bdr *tx_ring;
+	int cleaned_cnt, i;
+	u32 xdp_act;
+
+	cleaned_cnt = enetc_bd_unused(rx_ring);
+	/* next descriptor to process */
+	i = rx_ring->next_to_clean;
+
+	while (likely(rx_frm_cnt < work_limit)) {
+		union enetc_rx_bd *rxbd, *orig_rxbd;
+		struct xdp_buff xdp_buff;
+		struct sk_buff *skb;
+		int orig_i, err;
+		u32 bd_status;
+
+		rxbd = enetc_rxbd(rx_ring, i);
+		bd_status = le32_to_cpu(rxbd->r.lstatus);
+		if (!bd_status)
+			break;
+
+		enetc_wr_reg_hot(rx_ring->idr, BIT(rx_ring->index));
+		dma_rmb(); /* for reading other rxbd fields */
+
+		if (enetc_check_bd_errors_and_consume(rx_ring, bd_status,
+						      &rxbd, &i))
+			break;
+
+		orig_rxbd = rxbd;
+		orig_i = i;
+
+		enetc_build_xdp_buff(rx_ring, bd_status, &rxbd, &i,
+				     &cleaned_cnt, &xdp_buff);
+
+		/* When set, the outer VLAN header is extracted and reported
+		 * in the receive buffer descriptor. So rx_byte_cnt should
+		 * add the length of the extracted VLAN header.
+		 */
+		if (bd_status & ENETC_RXBD_FLAG_VLAN)
+			rx_byte_cnt += VLAN_HLEN;
+		rx_byte_cnt += xdp_get_buff_len(&xdp_buff);
+
+		xdp_act = bpf_prog_run_xdp(prog, &xdp_buff);
+
+		switch (xdp_act) {
+		default:
+			bpf_warn_invalid_xdp_action(rx_ring->ndev, prog, xdp_act);
+			fallthrough;
+		case XDP_ABORTED:
+			trace_xdp_exception(rx_ring->ndev, prog, xdp_act);
+			fallthrough;
+		case XDP_DROP:
+			enetc_xdp_drop(rx_ring, orig_i, i);
+			rx_ring->stats.xdp_drops++;
+			break;
+		case XDP_PASS:
+			skb = xdp_build_skb_from_buff(&xdp_buff);
+			/* Probably under memory pressure, stop NAPI */
+			if (unlikely(!skb)) {
+				enetc_xdp_drop(rx_ring, orig_i, i);
+				rx_ring->stats.xdp_drops++;
+				goto out;
+			}
+
+			enetc_get_offloads(rx_ring, orig_rxbd, skb);
+
+			/* These buffers are about to be owned by the stack.
+			 * Update our buffer cache (the rx_swbd array elements)
+			 * with their other page halves.
+			 */
+			enetc_bulk_flip_buff(rx_ring, orig_i, i);
+
+			napi_gro_receive(napi, skb);
+			break;
+		case XDP_TX:
+			tx_ring = priv->xdp_tx_ring[rx_ring->index];
+			if (unlikely(test_bit(ENETC_TX_DOWN, &priv->flags))) {
+				enetc_xdp_drop(rx_ring, orig_i, i);
+				tx_ring->stats.xdp_tx_drops++;
+				break;
+			}
+
+			xdp_tx_bd_cnt = enetc_rx_swbd_to_xdp_tx_swbd(xdp_tx_arr,
+								     rx_ring,
+								     orig_i, i);
+
+			if (!enetc_xdp_tx(tx_ring, xdp_tx_arr, xdp_tx_bd_cnt)) {
+				enetc_xdp_drop(rx_ring, orig_i, i);
+				tx_ring->stats.xdp_tx_drops++;
+			} else {
+				tx_ring->stats.xdp_tx++;
+				rx_ring->xdp.xdp_tx_in_flight += xdp_tx_bd_cnt;
+				xdp_tx_frm_cnt++;
+				/* The XDP_TX enqueue was successful, so we
+				 * need to scrub the RX software BDs because
+				 * the ownership of the buffers no longer
+				 * belongs to the RX ring, and we must prevent
+				 * enetc_refill_rx_ring() from reusing
+				 * rx_swbd->page.
+				 */
+				while (orig_i != i) {
+					rx_ring->rx_swbd[orig_i].page = NULL;
+					enetc_bdr_idx_inc(rx_ring, &orig_i);
+				}
+			}
+			break;
+		case XDP_REDIRECT:
+			err = xdp_do_redirect(rx_ring->ndev, &xdp_buff, prog);
+			if (unlikely(err)) {
+				enetc_xdp_drop(rx_ring, orig_i, i);
+				rx_ring->stats.xdp_redirect_failures++;
+			} else {
+				enetc_bulk_flip_buff(rx_ring, orig_i, i);
+				xdp_redirect_frm_cnt++;
+				rx_ring->stats.xdp_redirect++;
+			}
+		}
+
+		rx_frm_cnt++;
+	}
+
+out:
+	rx_ring->next_to_clean = i;
+
+	rx_ring->stats.packets += rx_frm_cnt;
+	rx_ring->stats.bytes += rx_byte_cnt;
+
+	if (xdp_redirect_frm_cnt)
+		xdp_do_flush();
+
+	if (xdp_tx_frm_cnt)
+		enetc_update_tx_ring_tail(tx_ring);
+
+	if (cleaned_cnt > rx_ring->xdp.xdp_tx_in_flight)
+		enetc_refill_rx_ring(rx_ring, enetc_bd_unused(rx_ring) -
+				     rx_ring->xdp.xdp_tx_in_flight);
+
+	return rx_frm_cnt;
+}
+
+static int enetc_poll(struct napi_struct *napi, int budget)
+{
+	struct enetc_int_vector
+		*v = container_of(napi, struct enetc_int_vector, napi);
+	struct enetc_bdr *rx_ring = &v->rx_ring;
+	struct bpf_prog *prog;
+	bool complete = true;
+	int work_done;
+	int i;
+
+	enetc_lock_mdio();
+
+	for (i = 0; i < v->count_tx_rings; i++)
+		if (!enetc_clean_tx_ring(&v->tx_ring[i], budget))
+			complete = false;
+
+	prog = rx_ring->xdp.prog;
+	if (prog)
+		work_done = enetc_clean_rx_ring_xdp(rx_ring, napi, budget, prog);
+	else
+		work_done = enetc_clean_rx_ring(rx_ring, napi, budget);
+	if (work_done == budget)
+		complete = false;
+	if (work_done)
+		v->rx_napi_work = true;
+
+	if (!complete) {
+		enetc_unlock_mdio();
+		return budget;
+	}
+
+	napi_complete_done(napi, work_done);
+
+	if (likely(v->rx_dim_en))
+		enetc_rx_net_dim(v);
+
+	v->rx_napi_work = false;
+
+	/* enable interrupts */
+	enetc_wr_reg_hot(v->rbier, ENETC_RBIER_RXTIE);
+
+	for_each_set_bit(i, &v->tx_rings_map, ENETC_MAX_NUM_TXQS)
+		enetc_wr_reg_hot(v->tbier_base + ENETC_BDR_OFF(i),
+				 ENETC_TBIER_TXTIE);
+
+	enetc_unlock_mdio();
+
+	return work_done;
+}
+
+/* Probing and Init */
+#define ENETC_MAX_RFS_SIZE 64
+void enetc_get_si_caps(struct enetc_si *si)
+{
+	struct enetc_hw *hw = &si->hw;
+	u32 val;
+
+	/* find out how many of various resources we have to work with */
+	val = enetc_rd(hw, ENETC_SICAPR0);
+	si->num_rx_rings = (val >> 16) & 0xff;
+	si->num_tx_rings = val & 0xff;
+
+	val = enetc_rd(hw, ENETC_SIPCAPR0);
+	if (val & ENETC_SIPCAPR0_RFS) {
+		val = enetc_rd(hw, ENETC_SIRFSCAPR);
+		si->num_fs_entries = ENETC_SIRFSCAPR_GET_NUM_RFS(val);
+		si->num_fs_entries = min(si->num_fs_entries, ENETC_MAX_RFS_SIZE);
+	} else {
+		/* ENETC which not supports RFS */
+		si->num_fs_entries = 0;
+	}
+
+	si->num_rss = 0;
+	val = enetc_rd(hw, ENETC_SIPCAPR0);
+	if (val & ENETC_SIPCAPR0_RSS) {
+		u32 rss;
+
+		rss = enetc_rd(hw, ENETC_SIRSSCAPR);
+		si->num_rss = ENETC_SIRSSCAPR_GET_NUM_RSS(rss);
+	}
+
+	if (val & ENETC_SIPCAPR0_LSO)
+		si->hw_features |= ENETC_SI_F_LSO;
+}
+EXPORT_SYMBOL_GPL(enetc_get_si_caps);
+
+static int enetc_dma_alloc_bdr(struct enetc_bdr_resource *res)
+{
+	size_t bd_base_size = res->bd_count * res->bd_size;
+
+	res->bd_base = dma_alloc_coherent(res->dev, bd_base_size,
+					  &res->bd_dma_base, GFP_KERNEL);
+	if (!res->bd_base)
+		return -ENOMEM;
+
+	/* h/w requires 128B alignment */
+	if (!IS_ALIGNED(res->bd_dma_base, 128)) {
+		dma_free_coherent(res->dev, bd_base_size, res->bd_base,
+				  res->bd_dma_base);
+		return -EINVAL;
+	}
+
+	return 0;
+}
+
+static void enetc_dma_free_bdr(const struct enetc_bdr_resource *res)
+{
+	size_t bd_base_size = res->bd_count * res->bd_size;
+
+	dma_free_coherent(res->dev, bd_base_size, res->bd_base,
+			  res->bd_dma_base);
+}
+
+static int enetc_alloc_tx_resource(struct enetc_bdr_resource *res,
+				   struct device *dev, size_t bd_count)
+{
+	int err;
+
+	res->dev = dev;
+	res->bd_count = bd_count;
+	res->bd_size = sizeof(union enetc_tx_bd);
+
+	res->tx_swbd = vcalloc(bd_count, sizeof(*res->tx_swbd));
+	if (!res->tx_swbd)
+		return -ENOMEM;
+
+	err = enetc_dma_alloc_bdr(res);
+	if (err)
+		goto err_alloc_bdr;
+
+	res->tso_headers = dma_alloc_coherent(dev, bd_count * TSO_HEADER_SIZE,
+					      &res->tso_headers_dma,
+					      GFP_KERNEL);
+	if (!res->tso_headers) {
+		err = -ENOMEM;
+		goto err_alloc_tso;
+	}
+
+	return 0;
+
+err_alloc_tso:
+	enetc_dma_free_bdr(res);
+err_alloc_bdr:
+	vfree(res->tx_swbd);
+	res->tx_swbd = NULL;
+
+	return err;
+}
+
+static void enetc_free_tx_resource(const struct enetc_bdr_resource *res)
+{
+	dma_free_coherent(res->dev, res->bd_count * TSO_HEADER_SIZE,
+			  res->tso_headers, res->tso_headers_dma);
+	enetc_dma_free_bdr(res);
+	vfree(res->tx_swbd);
+}
+
+static struct enetc_bdr_resource *
+enetc_alloc_tx_resources(struct enetc_ndev_priv *priv)
+{
+	struct enetc_bdr_resource *tx_res;
+	int i, err;
+
+	tx_res = kcalloc(priv->num_tx_rings, sizeof(*tx_res), GFP_KERNEL);
+	if (!tx_res)
+		return ERR_PTR(-ENOMEM);
+
+	for (i = 0; i < priv->num_tx_rings; i++) {
+		struct enetc_bdr *tx_ring = priv->tx_ring[i];
+
+		err = enetc_alloc_tx_resource(&tx_res[i], tx_ring->dev,
+					      tx_ring->bd_count);
+		if (err)
+			goto fail;
+	}
+
+	return tx_res;
+
+fail:
+	while (i-- > 0)
+		enetc_free_tx_resource(&tx_res[i]);
+
+	kfree(tx_res);
+
+	return ERR_PTR(err);
+}
+
+static void enetc_free_tx_resources(const struct enetc_bdr_resource *tx_res,
+				    size_t num_resources)
+{
+	size_t i;
+
+	for (i = 0; i < num_resources; i++)
+		enetc_free_tx_resource(&tx_res[i]);
+
+	kfree(tx_res);
+}
+
+static int enetc_alloc_rx_resource(struct enetc_bdr_resource *res,
+				   struct device *dev, size_t bd_count,
+				   bool extended)
+{
+	int err;
+
+	res->dev = dev;
+	res->bd_count = bd_count;
+	res->bd_size = sizeof(union enetc_rx_bd);
+	if (extended)
+		res->bd_size *= 2;
+
+	res->rx_swbd = vcalloc(bd_count, sizeof(struct enetc_rx_swbd));
+	if (!res->rx_swbd)
+		return -ENOMEM;
+
+	err = enetc_dma_alloc_bdr(res);
+	if (err) {
+		vfree(res->rx_swbd);
+		return err;
+	}
+
+	return 0;
+}
+
+static void enetc_free_rx_resource(const struct enetc_bdr_resource *res)
+{
+	enetc_dma_free_bdr(res);
+	vfree(res->rx_swbd);
+}
+
+static struct enetc_bdr_resource *
+enetc_alloc_rx_resources(struct enetc_ndev_priv *priv, bool extended)
+{
+	struct enetc_bdr_resource *rx_res;
+	int i, err;
+
+	rx_res = kcalloc(priv->num_rx_rings, sizeof(*rx_res), GFP_KERNEL);
+	if (!rx_res)
+		return ERR_PTR(-ENOMEM);
+
+	for (i = 0; i < priv->num_rx_rings; i++) {
+		struct enetc_bdr *rx_ring = priv->rx_ring[i];
+
+		err = enetc_alloc_rx_resource(&rx_res[i], rx_ring->dev,
+					      rx_ring->bd_count, extended);
+		if (err)
+			goto fail;
+	}
+
+	return rx_res;
+
+fail:
+	while (i-- > 0)
+		enetc_free_rx_resource(&rx_res[i]);
+
+	kfree(rx_res);
+
+	return ERR_PTR(err);
+}
+
+static void enetc_free_rx_resources(const struct enetc_bdr_resource *rx_res,
+				    size_t num_resources)
+{
+	size_t i;
+
+	for (i = 0; i < num_resources; i++)
+		enetc_free_rx_resource(&rx_res[i]);
+
+	kfree(rx_res);
+}
+
+static void enetc_assign_tx_resource(struct enetc_bdr *tx_ring,
+				     const struct enetc_bdr_resource *res)
+{
+	tx_ring->bd_base = res ? res->bd_base : NULL;
+	tx_ring->bd_dma_base = res ? res->bd_dma_base : 0;
+	tx_ring->tx_swbd = res ? res->tx_swbd : NULL;
+	tx_ring->tso_headers = res ? res->tso_headers : NULL;
+	tx_ring->tso_headers_dma = res ? res->tso_headers_dma : 0;
+}
+
+static void enetc_assign_rx_resource(struct enetc_bdr *rx_ring,
+				     const struct enetc_bdr_resource *res)
+{
+	rx_ring->bd_base = res ? res->bd_base : NULL;
+	rx_ring->bd_dma_base = res ? res->bd_dma_base : 0;
+	rx_ring->rx_swbd = res ? res->rx_swbd : NULL;
+}
+
+static void enetc_assign_tx_resources(struct enetc_ndev_priv *priv,
+				      const struct enetc_bdr_resource *res)
+{
+	int i;
+
+	if (priv->tx_res)
+		enetc_free_tx_resources(priv->tx_res, priv->num_tx_rings);
+
+	for (i = 0; i < priv->num_tx_rings; i++) {
+		enetc_assign_tx_resource(priv->tx_ring[i],
+					 res ? &res[i] : NULL);
+	}
+
+	priv->tx_res = res;
+}
+
+static void enetc_assign_rx_resources(struct enetc_ndev_priv *priv,
+				      const struct enetc_bdr_resource *res)
+{
+	int i;
+
+	if (priv->rx_res)
+		enetc_free_rx_resources(priv->rx_res, priv->num_rx_rings);
+
+	for (i = 0; i < priv->num_rx_rings; i++) {
+		enetc_assign_rx_resource(priv->rx_ring[i],
+					 res ? &res[i] : NULL);
+	}
+
+	priv->rx_res = res;
+}
+
+static void enetc_free_tx_ring(struct enetc_bdr *tx_ring)
+{
+	int i;
+
+	for (i = 0; i < tx_ring->bd_count; i++) {
+		struct enetc_tx_swbd *tx_swbd = &tx_ring->tx_swbd[i];
+
+		enetc_free_tx_frame(tx_ring, tx_swbd);
+	}
+}
+
+static void enetc_free_rx_ring(struct enetc_bdr *rx_ring)
+{
+	int i;
+
+	for (i = 0; i < rx_ring->bd_count; i++) {
+		struct enetc_rx_swbd *rx_swbd = &rx_ring->rx_swbd[i];
+
+		if (!rx_swbd->page)
+			continue;
+
+		dma_unmap_page(rx_ring->dev, rx_swbd->dma, PAGE_SIZE,
+			       rx_swbd->dir);
+		__free_page(rx_swbd->page);
+		rx_swbd->page = NULL;
+	}
+}
+
+static void enetc_free_rxtx_rings(struct enetc_ndev_priv *priv)
+{
+	int i;
+
+	for (i = 0; i < priv->num_rx_rings; i++)
+		enetc_free_rx_ring(priv->rx_ring[i]);
+
+	for (i = 0; i < priv->num_tx_rings; i++)
+		enetc_free_tx_ring(priv->tx_ring[i]);
+}
+
+static int enetc_setup_default_rss_table(struct enetc_si *si, int num_groups)
+{
+	int *rss_table;
+	int i;
+
+	rss_table = kmalloc_array(si->num_rss, sizeof(*rss_table), GFP_KERNEL);
+	if (!rss_table)
+		return -ENOMEM;
+
+	/* Set up RSS table defaults */
+	for (i = 0; i < si->num_rss; i++)
+		rss_table[i] = i % num_groups;
+
+	si->ops->set_rss_table(si, rss_table, si->num_rss);
+
+	kfree(rss_table);
+
+	return 0;
+}
+
+static void enetc_set_lso_flags_mask(struct enetc_hw *hw)
+{
+	enetc_wr(hw, ENETC4_SILSOSFMR0,
+		 SILSOSFMR0_VAL_SET(ENETC4_TCP_NL_SEG_FLAGS_DMASK,
+				    ENETC4_TCP_NL_SEG_FLAGS_DMASK));
+	enetc_wr(hw, ENETC4_SILSOSFMR1, 0);
+}
+
+static void enetc_set_rss(struct net_device *ndev, int en)
+{
+	struct enetc_ndev_priv *priv = netdev_priv(ndev);
+	struct enetc_hw *hw = &priv->si->hw;
+	u32 reg;
+
+	enetc_wr(hw, ENETC_SIRBGCR, priv->num_rx_rings);
+
+	reg = enetc_rd(hw, ENETC_SIMR);
+	reg &= ~ENETC_SIMR_RSSE;
+	reg |= (en) ? ENETC_SIMR_RSSE : 0;
+	enetc_wr(hw, ENETC_SIMR, reg);
+}
+
+int enetc_configure_si(struct enetc_ndev_priv *priv)
+{
+	struct enetc_si *si = priv->si;
+	struct enetc_hw *hw = &si->hw;
+	int err;
+
+	/* set SI cache attributes */
+	enetc_wr(hw, ENETC_SICAR0,
+		 ENETC_SICAR_RD_COHERENT | ENETC_SICAR_WR_COHERENT);
+	enetc_wr(hw, ENETC_SICAR1, ENETC_SICAR_MSI);
+	/* enable SI */
+	enetc_wr(hw, ENETC_SIMR, ENETC_SIMR_EN);
+
+	if (si->hw_features & ENETC_SI_F_LSO)
+		enetc_set_lso_flags_mask(hw);
+
+	if (si->num_rss) {
+		err = enetc_setup_default_rss_table(si, priv->num_rx_rings);
+		if (err)
+			return err;
+
+		if (priv->ndev->features & NETIF_F_RXHASH)
+			enetc_set_rss(priv->ndev, true);
+	}
+
+	return 0;
+}
+EXPORT_SYMBOL_GPL(enetc_configure_si);
+
+void enetc_init_si_rings_params(struct enetc_ndev_priv *priv)
+{
+	struct enetc_si *si = priv->si;
+	int cpus = num_online_cpus();
+
+	priv->tx_bd_count = ENETC_TX_RING_DEFAULT_SIZE;
+	priv->rx_bd_count = ENETC_RX_RING_DEFAULT_SIZE;
+
+	/* Enable all available TX rings in order to configure as many
+	 * priorities as possible, when needed.
+	 * TODO: Make # of TX rings run-time configurable
+	 */
+	priv->num_rx_rings = min_t(int, cpus, si->num_rx_rings);
+	priv->num_tx_rings = si->num_tx_rings;
+	priv->bdr_int_num = priv->num_rx_rings;
+	priv->ic_mode = ENETC_IC_RX_ADAPTIVE | ENETC_IC_TX_MANUAL;
+	priv->tx_ictt = enetc_usecs_to_cycles(600, priv->sysclk_freq);
+}
+EXPORT_SYMBOL_GPL(enetc_init_si_rings_params);
+
+int enetc_alloc_si_resources(struct enetc_ndev_priv *priv)
+{
+	struct enetc_si *si = priv->si;
+
+	priv->cls_rules = kcalloc(si->num_fs_entries, sizeof(*priv->cls_rules),
+				  GFP_KERNEL);
+	if (!priv->cls_rules)
+		return -ENOMEM;
+
+	return 0;
+}
+EXPORT_SYMBOL_GPL(enetc_alloc_si_resources);
+
+void enetc_free_si_resources(struct enetc_ndev_priv *priv)
+{
+	kfree(priv->cls_rules);
+}
+EXPORT_SYMBOL_GPL(enetc_free_si_resources);
+
+static void enetc_setup_txbdr(struct enetc_hw *hw, struct enetc_bdr *tx_ring)
+{
+	int idx = tx_ring->index;
+	u32 tbmr;
+
+	enetc_txbdr_wr(hw, idx, ENETC_TBBAR0,
+		       lower_32_bits(tx_ring->bd_dma_base));
+
+	enetc_txbdr_wr(hw, idx, ENETC_TBBAR1,
+		       upper_32_bits(tx_ring->bd_dma_base));
+
+	WARN_ON(!IS_ALIGNED(tx_ring->bd_count, 64)); /* multiple of 64 */
+	enetc_txbdr_wr(hw, idx, ENETC_TBLENR,
+		       ENETC_RTBLENR_LEN(tx_ring->bd_count));
+
+	/* clearing PI/CI registers for Tx not supported, adjust sw indexes */
+	tx_ring->next_to_use = enetc_txbdr_rd(hw, idx, ENETC_TBPIR);
+	tx_ring->next_to_clean = enetc_txbdr_rd(hw, idx, ENETC_TBCIR);
+
+	/* enable Tx ints by setting pkt thr to 1 */
+	enetc_txbdr_wr(hw, idx, ENETC_TBICR0, ENETC_TBICR0_ICEN | 0x1);
+
+	tbmr = ENETC_TBMR_SET_PRIO(tx_ring->prio);
+	if (tx_ring->ndev->features & NETIF_F_HW_VLAN_CTAG_TX)
+		tbmr |= ENETC_TBMR_VIH;
+
+	/* enable ring */
+	enetc_txbdr_wr(hw, idx, ENETC_TBMR, tbmr);
+
+	tx_ring->tpir = hw->reg + ENETC_BDR(TX, idx, ENETC_TBPIR);
+	tx_ring->tcir = hw->reg + ENETC_BDR(TX, idx, ENETC_TBCIR);
+	tx_ring->idr = hw->reg + ENETC_SITXIDR;
+}
+
+static void enetc_setup_rxbdr(struct enetc_hw *hw, struct enetc_bdr *rx_ring,
+			      bool extended)
+{
+	int idx = rx_ring->index;
+	u32 rbmr = 0;
+
+	enetc_rxbdr_wr(hw, idx, ENETC_RBBAR0,
+		       lower_32_bits(rx_ring->bd_dma_base));
+
+	enetc_rxbdr_wr(hw, idx, ENETC_RBBAR1,
+		       upper_32_bits(rx_ring->bd_dma_base));
+
+	WARN_ON(!IS_ALIGNED(rx_ring->bd_count, 64)); /* multiple of 64 */
+	enetc_rxbdr_wr(hw, idx, ENETC_RBLENR,
+		       ENETC_RTBLENR_LEN(rx_ring->bd_count));
+
+	if (rx_ring->xdp.prog)
+		enetc_rxbdr_wr(hw, idx, ENETC_RBBSR, ENETC_RXB_DMA_SIZE_XDP);
+	else
+		enetc_rxbdr_wr(hw, idx, ENETC_RBBSR, ENETC_RXB_DMA_SIZE);
+
+	/* Also prepare the consumer index in case page allocation never
+	 * succeeds. In that case, hardware will never advance producer index
+	 * to match consumer index, and will drop all frames.
+	 */
+	enetc_rxbdr_wr(hw, idx, ENETC_RBPIR, 0);
+	enetc_rxbdr_wr(hw, idx, ENETC_RBCIR, 1);
+
+	/* enable Rx ints by setting pkt thr to 1 */
+	enetc_rxbdr_wr(hw, idx, ENETC_RBICR0, ENETC_RBICR0_ICEN | 0x1);
+
+	rx_ring->ext_en = extended;
+	if (rx_ring->ext_en)
+		rbmr |= ENETC_RBMR_BDS;
+
+	if (rx_ring->ndev->features & NETIF_F_HW_VLAN_CTAG_RX)
+		rbmr |= ENETC_RBMR_VTE;
+
+	rx_ring->rcir = hw->reg + ENETC_BDR(RX, idx, ENETC_RBCIR);
+	rx_ring->idr = hw->reg + ENETC_SIRXIDR;
+
+	rx_ring->next_to_clean = 0;
+	rx_ring->next_to_use = 0;
+	rx_ring->next_to_alloc = 0;
+
+	enetc_lock_mdio();
+	enetc_refill_rx_ring(rx_ring, enetc_bd_unused(rx_ring));
+	enetc_unlock_mdio();
+
+	enetc_rxbdr_wr(hw, idx, ENETC_RBMR, rbmr);
+}
+
+static void enetc_setup_bdrs(struct enetc_ndev_priv *priv, bool extended)
+{
+	struct enetc_hw *hw = &priv->si->hw;
+	int i;
+
+	for (i = 0; i < priv->num_tx_rings; i++)
+		enetc_setup_txbdr(hw, priv->tx_ring[i]);
+
+	for (i = 0; i < priv->num_rx_rings; i++)
+		enetc_setup_rxbdr(hw, priv->rx_ring[i], extended);
+}
+
+static void enetc_enable_txbdr(struct enetc_hw *hw, struct enetc_bdr *tx_ring)
+{
+	int idx = tx_ring->index;
+	u32 tbmr;
+
+	tbmr = enetc_txbdr_rd(hw, idx, ENETC_TBMR);
+	tbmr |= ENETC_TBMR_EN;
+	enetc_txbdr_wr(hw, idx, ENETC_TBMR, tbmr);
+}
+
+static void enetc_enable_rxbdr(struct enetc_hw *hw, struct enetc_bdr *rx_ring)
+{
+	int idx = rx_ring->index;
+	u32 rbmr;
+
+	rbmr = enetc_rxbdr_rd(hw, idx, ENETC_RBMR);
+	rbmr |= ENETC_RBMR_EN;
+	enetc_rxbdr_wr(hw, idx, ENETC_RBMR, rbmr);
+}
+
+static void enetc_enable_rx_bdrs(struct enetc_ndev_priv *priv)
+{
+	struct enetc_hw *hw = &priv->si->hw;
+	int i;
+
+	for (i = 0; i < priv->num_rx_rings; i++)
+		enetc_enable_rxbdr(hw, priv->rx_ring[i]);
+}
+
+static void enetc_enable_tx_bdrs(struct enetc_ndev_priv *priv)
+{
+	struct enetc_hw *hw = &priv->si->hw;
+	int i;
+
+	for (i = 0; i < priv->num_tx_rings; i++)
+		enetc_enable_txbdr(hw, priv->tx_ring[i]);
+}
+
+static void enetc_disable_rxbdr(struct enetc_hw *hw, struct enetc_bdr *rx_ring)
+{
+	int idx = rx_ring->index;
+
+	/* disable EN bit on ring */
+	enetc_rxbdr_wr(hw, idx, ENETC_RBMR, 0);
+}
+
+static void enetc_disable_txbdr(struct enetc_hw *hw, struct enetc_bdr *rx_ring)
+{
+	int idx = rx_ring->index;
+
+	/* disable EN bit on ring */
+	enetc_txbdr_wr(hw, idx, ENETC_TBMR, 0);
+}
+
+static void enetc_disable_rx_bdrs(struct enetc_ndev_priv *priv)
+{
+	struct enetc_hw *hw = &priv->si->hw;
+	int i;
+
+	for (i = 0; i < priv->num_rx_rings; i++)
+		enetc_disable_rxbdr(hw, priv->rx_ring[i]);
+}
+
+static void enetc_disable_tx_bdrs(struct enetc_ndev_priv *priv)
+{
+	struct enetc_hw *hw = &priv->si->hw;
+	int i;
+
+	for (i = 0; i < priv->num_tx_rings; i++)
+		enetc_disable_txbdr(hw, priv->tx_ring[i]);
+}
+
+static void enetc_wait_txbdr(struct enetc_hw *hw, struct enetc_bdr *tx_ring)
+{
+	int delay = 8, timeout = 100;
+	int idx = tx_ring->index;
+
+	/* wait for busy to clear */
+	while (delay < timeout &&
+	       enetc_txbdr_rd(hw, idx, ENETC_TBSR) & ENETC_TBSR_BUSY) {
+		msleep(delay);
+		delay *= 2;
+	}
+
+	if (delay >= timeout)
+		netdev_warn(tx_ring->ndev, "timeout for tx ring #%d clear\n",
+			    idx);
+}
+
+static void enetc_wait_bdrs(struct enetc_ndev_priv *priv)
+{
+	struct enetc_hw *hw = &priv->si->hw;
+	int i;
+
+	for (i = 0; i < priv->num_tx_rings; i++)
+		enetc_wait_txbdr(hw, priv->tx_ring[i]);
+}
+
+static int enetc_setup_irqs(struct enetc_ndev_priv *priv)
+{
+	struct pci_dev *pdev = priv->si->pdev;
+	struct enetc_hw *hw = &priv->si->hw;
+	int i, j, err;
+
+	for (i = 0; i < priv->bdr_int_num; i++) {
+		int irq = pci_irq_vector(pdev, ENETC_BDR_INT_BASE_IDX + i);
+		struct enetc_int_vector *v = priv->int_vector[i];
+		int entry = ENETC_BDR_INT_BASE_IDX + i;
+
+		snprintf(v->name, sizeof(v->name), "%s-rxtx%d",
+			 priv->ndev->name, i);
+		err = request_irq(irq, enetc_msix, IRQF_NO_AUTOEN, v->name, v);
+		if (err) {
+			dev_err(priv->dev, "request_irq() failed!\n");
+			goto irq_err;
+		}
+
+		v->tbier_base = hw->reg + ENETC_BDR(TX, 0, ENETC_TBIER);
+		v->rbier = hw->reg + ENETC_BDR(RX, i, ENETC_RBIER);
+		v->ricr1 = hw->reg + ENETC_BDR(RX, i, ENETC_RBICR1);
+
+		enetc_wr(hw, ENETC_SIMSIRRV(i), entry);
+
+		for (j = 0; j < v->count_tx_rings; j++) {
+			int idx = v->tx_ring[j].index;
+
+			enetc_wr(hw, ENETC_SIMSITRV(idx), entry);
+		}
+		irq_set_affinity_hint(irq, get_cpu_mask(i % num_online_cpus()));
+	}
+
+	return 0;
+
+irq_err:
+	while (i--) {
+		int irq = pci_irq_vector(pdev, ENETC_BDR_INT_BASE_IDX + i);
+
+		irq_set_affinity_hint(irq, NULL);
+		free_irq(irq, priv->int_vector[i]);
+	}
+
+	return err;
+}
+
+static void enetc_free_irqs(struct enetc_ndev_priv *priv)
+{
+	struct pci_dev *pdev = priv->si->pdev;
+	int i;
+
+	for (i = 0; i < priv->bdr_int_num; i++) {
+		int irq = pci_irq_vector(pdev, ENETC_BDR_INT_BASE_IDX + i);
+
+		irq_set_affinity_hint(irq, NULL);
+		free_irq(irq, priv->int_vector[i]);
+	}
+}
+
+static void enetc_setup_interrupts(struct enetc_ndev_priv *priv)
+{
+	struct enetc_hw *hw = &priv->si->hw;
+	u32 icpt, ictt;
+	int i;
+
+	/* enable Tx & Rx event indication */
+	if (priv->ic_mode &
+	    (ENETC_IC_RX_MANUAL | ENETC_IC_RX_ADAPTIVE)) {
+		icpt = ENETC_RBICR0_SET_ICPT(ENETC_RXIC_PKTTHR);
+		/* init to non-0 minimum, will be adjusted later */
+		ictt = 0x1;
+	} else {
+		icpt = 0x1; /* enable Rx ints by setting pkt thr to 1 */
+		ictt = 0;
+	}
+
+	for (i = 0; i < priv->num_rx_rings; i++) {
+		enetc_rxbdr_wr(hw, i, ENETC_RBICR1, ictt);
+		enetc_rxbdr_wr(hw, i, ENETC_RBICR0, ENETC_RBICR0_ICEN | icpt);
+		enetc_rxbdr_wr(hw, i, ENETC_RBIER, ENETC_RBIER_RXTIE);
+	}
+
+	if (priv->ic_mode & ENETC_IC_TX_MANUAL)
+		icpt = ENETC_TBICR0_SET_ICPT(ENETC_TXIC_PKTTHR);
+	else
+		icpt = 0x1; /* enable Tx ints by setting pkt thr to 1 */
+
+	for (i = 0; i < priv->num_tx_rings; i++) {
+		enetc_txbdr_wr(hw, i, ENETC_TBICR1, priv->tx_ictt);
+		enetc_txbdr_wr(hw, i, ENETC_TBICR0, ENETC_TBICR0_ICEN | icpt);
+		enetc_txbdr_wr(hw, i, ENETC_TBIER, ENETC_TBIER_TXTIE);
+	}
+}
+
+static void enetc_clear_interrupts(struct enetc_ndev_priv *priv)
+{
+	struct enetc_hw *hw = &priv->si->hw;
+	int i;
+
+	for (i = 0; i < priv->num_tx_rings; i++)
+		enetc_txbdr_wr(hw, i, ENETC_TBIER, 0);
+
+	for (i = 0; i < priv->num_rx_rings; i++)
+		enetc_rxbdr_wr(hw, i, ENETC_RBIER, 0);
+}
+
+static int enetc_phylink_connect(struct net_device *ndev)
+{
+	struct enetc_ndev_priv *priv = netdev_priv(ndev);
+	struct ethtool_keee edata;
+	int err;
+
+	if (!priv->phylink) {
+		/* phy-less mode */
+		netif_carrier_on(ndev);
+		return 0;
+	}
+
+	err = phylink_of_phy_connect(priv->phylink, priv->dev->of_node, 0);
+	if (err) {
+		dev_err(&ndev->dev, "could not attach to PHY\n");
+		return err;
+	}
+
+	/* disable EEE autoneg, until ENETC driver supports it */
+	memset(&edata, 0, sizeof(struct ethtool_keee));
+	phylink_ethtool_set_eee(priv->phylink, &edata);
+
+	phylink_start(priv->phylink);
+
+	return 0;
+}
+
+static void enetc_tx_onestep_tstamp(struct work_struct *work)
+{
+	struct enetc_ndev_priv *priv;
+	struct sk_buff *skb;
+
+	priv = container_of(work, struct enetc_ndev_priv, tx_onestep_tstamp);
+
+	netif_tx_lock_bh(priv->ndev);
+
+	clear_bit_unlock(ENETC_TX_ONESTEP_TSTAMP_IN_PROGRESS, &priv->flags);
+	skb = skb_dequeue(&priv->tx_skbs);
+	if (skb)
+		enetc_start_xmit(skb, priv->ndev);
+
+	netif_tx_unlock_bh(priv->ndev);
+}
+
+static void enetc_tx_onestep_tstamp_init(struct enetc_ndev_priv *priv)
+{
+	INIT_WORK(&priv->tx_onestep_tstamp, enetc_tx_onestep_tstamp);
+	skb_queue_head_init(&priv->tx_skbs);
+}
+
+void enetc_start(struct net_device *ndev)
+{
+	struct enetc_ndev_priv *priv = netdev_priv(ndev);
+	int i;
+
+	enetc_setup_interrupts(priv);
+
+	for (i = 0; i < priv->bdr_int_num; i++) {
+		int irq = pci_irq_vector(priv->si->pdev,
+					 ENETC_BDR_INT_BASE_IDX + i);
+
+		napi_enable(&priv->int_vector[i]->napi);
+		enable_irq(irq);
+	}
+
+	enetc_enable_tx_bdrs(priv);
+
+	enetc_enable_rx_bdrs(priv);
+
+	netif_tx_start_all_queues(ndev);
+
+	clear_bit(ENETC_TX_DOWN, &priv->flags);
+}
+EXPORT_SYMBOL_GPL(enetc_start);
+
+int enetc_open(struct net_device *ndev)
+{
+	struct enetc_ndev_priv *priv = netdev_priv(ndev);
+	struct enetc_bdr_resource *tx_res, *rx_res;
+	bool extended;
+	int err;
+
+	extended = !!(priv->active_offloads & ENETC_F_RX_TSTAMP);
+
+	err = clk_prepare_enable(priv->ref_clk);
+	if (err)
+		return err;
+
+	err = enetc_setup_irqs(priv);
+	if (err)
+		goto err_setup_irqs;
+
+	err = enetc_phylink_connect(ndev);
+	if (err)
+		goto err_phy_connect;
+
+	tx_res = enetc_alloc_tx_resources(priv);
+	if (IS_ERR(tx_res)) {
+		err = PTR_ERR(tx_res);
+		goto err_alloc_tx;
+	}
+
+	rx_res = enetc_alloc_rx_resources(priv, extended);
+	if (IS_ERR(rx_res)) {
+		err = PTR_ERR(rx_res);
+		goto err_alloc_rx;
+	}
+
+	enetc_tx_onestep_tstamp_init(priv);
+	enetc_assign_tx_resources(priv, tx_res);
+	enetc_assign_rx_resources(priv, rx_res);
+	enetc_setup_bdrs(priv, extended);
+	enetc_start(ndev);
+
+	return 0;
+
+err_alloc_rx:
+	enetc_free_tx_resources(tx_res, priv->num_tx_rings);
+err_alloc_tx:
+	if (priv->phylink)
+		phylink_disconnect_phy(priv->phylink);
+err_phy_connect:
+	enetc_free_irqs(priv);
+err_setup_irqs:
+	clk_disable_unprepare(priv->ref_clk);
+
+	return err;
+}
+EXPORT_SYMBOL_GPL(enetc_open);
+
+void enetc_stop(struct net_device *ndev)
+{
+	struct enetc_ndev_priv *priv = netdev_priv(ndev);
+	int i;
+
+	set_bit(ENETC_TX_DOWN, &priv->flags);
+
+	netif_tx_stop_all_queues(ndev);
+
+	enetc_disable_rx_bdrs(priv);
+
+	enetc_wait_bdrs(priv);
+
+	enetc_disable_tx_bdrs(priv);
+
+	for (i = 0; i < priv->bdr_int_num; i++) {
+		int irq = pci_irq_vector(priv->si->pdev,
+					 ENETC_BDR_INT_BASE_IDX + i);
+
+		disable_irq(irq);
+		napi_synchronize(&priv->int_vector[i]->napi);
+		napi_disable(&priv->int_vector[i]->napi);
+	}
+
+	enetc_clear_interrupts(priv);
+}
+EXPORT_SYMBOL_GPL(enetc_stop);
+
+int enetc_close(struct net_device *ndev)
+{
+	struct enetc_ndev_priv *priv = netdev_priv(ndev);
+
+	enetc_stop(ndev);
+
+	if (priv->phylink) {
+		phylink_stop(priv->phylink);
+		phylink_disconnect_phy(priv->phylink);
+	} else {
+		netif_carrier_off(ndev);
+	}
+
+	enetc_free_rxtx_rings(priv);
+
+	/* Avoids dangling pointers and also frees old resources */
+	enetc_assign_rx_resources(priv, NULL);
+	enetc_assign_tx_resources(priv, NULL);
+
+	enetc_free_irqs(priv);
+	clk_disable_unprepare(priv->ref_clk);
+
+	return 0;
+}
+EXPORT_SYMBOL_GPL(enetc_close);
+
+static int enetc_reconfigure(struct enetc_ndev_priv *priv, bool extended,
+			     int (*cb)(struct enetc_ndev_priv *priv, void *ctx),
+			     void *ctx)
+{
+	struct enetc_bdr_resource *tx_res, *rx_res;
+	int err;
+
+	ASSERT_RTNL();
+
+	/* If the interface is down, run the callback right away,
+	 * without reconfiguration.
+	 */
+	if (!netif_running(priv->ndev)) {
+		if (cb) {
+			err = cb(priv, ctx);
+			if (err)
+				return err;
+		}
+
+		return 0;
+	}
+
+	tx_res = enetc_alloc_tx_resources(priv);
+	if (IS_ERR(tx_res)) {
+		err = PTR_ERR(tx_res);
+		goto out;
+	}
+
+	rx_res = enetc_alloc_rx_resources(priv, extended);
+	if (IS_ERR(rx_res)) {
+		err = PTR_ERR(rx_res);
+		goto out_free_tx_res;
+	}
+
+	enetc_stop(priv->ndev);
+	enetc_free_rxtx_rings(priv);
+
+	/* Interface is down, run optional callback now */
+	if (cb) {
+		err = cb(priv, ctx);
+		if (err)
+			goto out_restart;
+	}
+
+	enetc_assign_tx_resources(priv, tx_res);
+	enetc_assign_rx_resources(priv, rx_res);
+	enetc_setup_bdrs(priv, extended);
+	enetc_start(priv->ndev);
+
+	return 0;
+
+out_restart:
+	enetc_setup_bdrs(priv, extended);
+	enetc_start(priv->ndev);
+	enetc_free_rx_resources(rx_res, priv->num_rx_rings);
+out_free_tx_res:
+	enetc_free_tx_resources(tx_res, priv->num_tx_rings);
+out:
+	return err;
+}
+
+static void enetc_debug_tx_ring_prios(struct enetc_ndev_priv *priv)
+{
+	int i;
+
+	for (i = 0; i < priv->num_tx_rings; i++)
+		netdev_dbg(priv->ndev, "TX ring %d prio %d\n", i,
+			   priv->tx_ring[i]->prio);
+}
+
+void enetc_reset_tc_mqprio(struct net_device *ndev)
+{
+	struct enetc_ndev_priv *priv = netdev_priv(ndev);
+	struct enetc_hw *hw = &priv->si->hw;
+	struct enetc_bdr *tx_ring;
+	int num_stack_tx_queues;
+	int i;
+
+	num_stack_tx_queues = enetc_num_stack_tx_queues(priv);
+
+	netdev_reset_tc(ndev);
+	netif_set_real_num_tx_queues(ndev, num_stack_tx_queues);
+	priv->min_num_stack_tx_queues = num_possible_cpus();
+
+	/* Reset all ring priorities to 0 */
+	for (i = 0; i < priv->num_tx_rings; i++) {
+		tx_ring = priv->tx_ring[i];
+		tx_ring->prio = 0;
+		enetc_set_bdr_prio(hw, tx_ring->index, tx_ring->prio);
+	}
+
+	enetc_debug_tx_ring_prios(priv);
+
+	enetc_change_preemptible_tcs(priv, 0);
+}
+EXPORT_SYMBOL_GPL(enetc_reset_tc_mqprio);
+
+int enetc_setup_tc_mqprio(struct net_device *ndev, void *type_data)
+{
+	struct tc_mqprio_qopt_offload *mqprio = type_data;
+	struct enetc_ndev_priv *priv = netdev_priv(ndev);
+	struct tc_mqprio_qopt *qopt = &mqprio->qopt;
+	struct enetc_hw *hw = &priv->si->hw;
+	int num_stack_tx_queues = 0;
+	struct enetc_bdr *tx_ring;
+	u8 num_tc = qopt->num_tc;
+	int offset, count;
+	int err, tc, q;
+
+	if (!num_tc) {
+		enetc_reset_tc_mqprio(ndev);
+		return 0;
+	}
+
+	err = netdev_set_num_tc(ndev, num_tc);
+	if (err)
+		return err;
+
+	for (tc = 0; tc < num_tc; tc++) {
+		offset = qopt->offset[tc];
+		count = qopt->count[tc];
+		num_stack_tx_queues += count;
+
+		err = netdev_set_tc_queue(ndev, tc, count, offset);
+		if (err)
+			goto err_reset_tc;
+
+		for (q = offset; q < offset + count; q++) {
+			tx_ring = priv->tx_ring[q];
+			/* The prio_tc_map is skb_tx_hash()'s way of selecting
+			 * between TX queues based on skb->priority. As such,
+			 * there's nothing to offload based on it.
+			 * Make the mqprio "traffic class" be the priority of
+			 * this ring group, and leave the Tx IPV to traffic
+			 * class mapping as its default mapping value of 1:1.
+			 */
+			tx_ring->prio = tc;
+			enetc_set_bdr_prio(hw, tx_ring->index, tx_ring->prio);
+		}
+	}
+
+	err = netif_set_real_num_tx_queues(ndev, num_stack_tx_queues);
+	if (err)
+		goto err_reset_tc;
+
+	priv->min_num_stack_tx_queues = num_stack_tx_queues;
+
+	enetc_debug_tx_ring_prios(priv);
+
+	enetc_change_preemptible_tcs(priv, mqprio->preemptible_tcs);
+
+	return 0;
+
+err_reset_tc:
+	enetc_reset_tc_mqprio(ndev);
+	return err;
+}
+EXPORT_SYMBOL_GPL(enetc_setup_tc_mqprio);
+
+static int enetc_reconfigure_xdp_cb(struct enetc_ndev_priv *priv, void *ctx)
+{
+	struct bpf_prog *old_prog, *prog = ctx;
+	int num_stack_tx_queues;
+	int err, i;
+
+	old_prog = xchg(&priv->xdp_prog, prog);
+
+	num_stack_tx_queues = enetc_num_stack_tx_queues(priv);
+	err = netif_set_real_num_tx_queues(priv->ndev, num_stack_tx_queues);
+	if (err) {
+		xchg(&priv->xdp_prog, old_prog);
+		return err;
+	}
+
+	if (old_prog)
+		bpf_prog_put(old_prog);
+
+	for (i = 0; i < priv->num_rx_rings; i++) {
+		struct enetc_bdr *rx_ring = priv->rx_ring[i];
+
+		rx_ring->xdp.prog = prog;
+
+		if (prog)
+			rx_ring->buffer_offset = XDP_PACKET_HEADROOM;
+		else
+			rx_ring->buffer_offset = ENETC_RXB_PAD;
+	}
+
+	return 0;
+}
+
+static int enetc_setup_xdp_prog(struct net_device *ndev, struct bpf_prog *prog,
+				struct netlink_ext_ack *extack)
+{
+	int num_xdp_tx_queues = prog ? num_possible_cpus() : 0;
+	struct enetc_ndev_priv *priv = netdev_priv(ndev);
+	bool extended;
+
+	if (priv->min_num_stack_tx_queues + num_xdp_tx_queues >
+	    priv->num_tx_rings) {
+		NL_SET_ERR_MSG_FMT_MOD(extack,
+				       "Reserving %d XDP TXQs leaves under %d for stack (total %d)",
+				       num_xdp_tx_queues,
+				       priv->min_num_stack_tx_queues,
+				       priv->num_tx_rings);
+		return -EBUSY;
+	}
+
+	extended = !!(priv->active_offloads & ENETC_F_RX_TSTAMP);
+
+	/* The buffer layout is changing, so we need to drain the old
+	 * RX buffers and seed new ones.
+	 */
+	return enetc_reconfigure(priv, extended, enetc_reconfigure_xdp_cb, prog);
+}
+
+int enetc_setup_bpf(struct net_device *ndev, struct netdev_bpf *bpf)
+{
+	switch (bpf->command) {
+	case XDP_SETUP_PROG:
+		return enetc_setup_xdp_prog(ndev, bpf->prog, bpf->extack);
+	default:
+		return -EINVAL;
+	}
+
+	return 0;
+}
+EXPORT_SYMBOL_GPL(enetc_setup_bpf);
+
+struct net_device_stats *enetc_get_stats(struct net_device *ndev)
+{
+	struct enetc_ndev_priv *priv = netdev_priv(ndev);
+	struct net_device_stats *stats = &ndev->stats;
+	unsigned long packets = 0, bytes = 0;
+	unsigned long tx_dropped = 0;
+	int i;
+
+	for (i = 0; i < priv->num_rx_rings; i++) {
+		packets += priv->rx_ring[i]->stats.packets;
+		bytes	+= priv->rx_ring[i]->stats.bytes;
+	}
+
+	stats->rx_packets = packets;
+	stats->rx_bytes = bytes;
+	bytes = 0;
+	packets = 0;
+
+	for (i = 0; i < priv->num_tx_rings; i++) {
+		packets += priv->tx_ring[i]->stats.packets;
+		bytes	+= priv->tx_ring[i]->stats.bytes;
+		tx_dropped += priv->tx_ring[i]->stats.win_drop;
+	}
+
+	stats->tx_packets = packets;
+	stats->tx_bytes = bytes;
+	stats->tx_dropped = tx_dropped;
+
+	return stats;
+}
+EXPORT_SYMBOL_GPL(enetc_get_stats);
+
+static void enetc_enable_rxvlan(struct net_device *ndev, bool en)
+{
+	struct enetc_ndev_priv *priv = netdev_priv(ndev);
+	struct enetc_hw *hw = &priv->si->hw;
+	int i;
+
+	for (i = 0; i < priv->num_rx_rings; i++)
+		enetc_bdr_enable_rxvlan(hw, i, en);
+}
+
+static void enetc_enable_txvlan(struct net_device *ndev, bool en)
+{
+	struct enetc_ndev_priv *priv = netdev_priv(ndev);
+	struct enetc_hw *hw = &priv->si->hw;
+	int i;
+
+	for (i = 0; i < priv->num_tx_rings; i++)
+		enetc_bdr_enable_txvlan(hw, i, en);
+}
+
+void enetc_set_features(struct net_device *ndev, netdev_features_t features)
+{
+	netdev_features_t changed = ndev->features ^ features;
+
+	if (changed & NETIF_F_RXHASH)
+		enetc_set_rss(ndev, !!(features & NETIF_F_RXHASH));
+
+	if (changed & NETIF_F_HW_VLAN_CTAG_RX)
+		enetc_enable_rxvlan(ndev,
+				    !!(features & NETIF_F_HW_VLAN_CTAG_RX));
+
+	if (changed & NETIF_F_HW_VLAN_CTAG_TX)
+		enetc_enable_txvlan(ndev,
+				    !!(features & NETIF_F_HW_VLAN_CTAG_TX));
+}
+EXPORT_SYMBOL_GPL(enetc_set_features);
+
+int enetc_hwtstamp_set(struct net_device *ndev,
+		       struct kernel_hwtstamp_config *config,
+		       struct netlink_ext_ack *extack)
+{
+	struct enetc_ndev_priv *priv = netdev_priv(ndev);
+	int err, new_offloads = priv->active_offloads;
+
+	if (!enetc_ptp_clock_is_enabled(priv->si))
+		return -EOPNOTSUPP;
+
+	switch (config->tx_type) {
+	case HWTSTAMP_TX_OFF:
+		new_offloads &= ~ENETC_F_TX_TSTAMP_MASK;
+		break;
+	case HWTSTAMP_TX_ON:
+		new_offloads &= ~ENETC_F_TX_TSTAMP_MASK;
+		new_offloads |= ENETC_F_TX_TSTAMP;
+		break;
+	case HWTSTAMP_TX_ONESTEP_SYNC:
+		if (!enetc_si_is_pf(priv->si))
+			return -EOPNOTSUPP;
+
+		new_offloads &= ~ENETC_F_TX_TSTAMP_MASK;
+		new_offloads |= ENETC_F_TX_ONESTEP_SYNC_TSTAMP;
+		break;
+	default:
+		return -ERANGE;
+	}
+
+	switch (config->rx_filter) {
+	case HWTSTAMP_FILTER_NONE:
+		new_offloads &= ~ENETC_F_RX_TSTAMP;
+		break;
+	default:
+		new_offloads |= ENETC_F_RX_TSTAMP;
+		config->rx_filter = HWTSTAMP_FILTER_ALL;
+	}
+
+	if ((new_offloads ^ priv->active_offloads) & ENETC_F_RX_TSTAMP) {
+		bool extended = !!(new_offloads & ENETC_F_RX_TSTAMP);
+
+		err = enetc_reconfigure(priv, extended, NULL, NULL);
+		if (err)
+			return err;
+	}
+
+	priv->active_offloads = new_offloads;
+
+	return 0;
+}
+EXPORT_SYMBOL_GPL(enetc_hwtstamp_set);
+
+int enetc_hwtstamp_get(struct net_device *ndev,
+		       struct kernel_hwtstamp_config *config)
+{
+	struct enetc_ndev_priv *priv = netdev_priv(ndev);
+
+	if (!enetc_ptp_clock_is_enabled(priv->si))
+		return -EOPNOTSUPP;
+
+	if (priv->active_offloads & ENETC_F_TX_ONESTEP_SYNC_TSTAMP)
+		config->tx_type = HWTSTAMP_TX_ONESTEP_SYNC;
+	else if (priv->active_offloads & ENETC_F_TX_TSTAMP)
+		config->tx_type = HWTSTAMP_TX_ON;
+	else
+		config->tx_type = HWTSTAMP_TX_OFF;
+
+	config->rx_filter = (priv->active_offloads & ENETC_F_RX_TSTAMP) ?
+			     HWTSTAMP_FILTER_ALL : HWTSTAMP_FILTER_NONE;
+
+	return 0;
+}
+EXPORT_SYMBOL_GPL(enetc_hwtstamp_get);
+
+int enetc_ioctl(struct net_device *ndev, struct ifreq *rq, int cmd)
+{
+	struct enetc_ndev_priv *priv = netdev_priv(ndev);
+
+	if (!priv->phylink)
+		return -EOPNOTSUPP;
+
+	return phylink_mii_ioctl(priv->phylink, rq, cmd);
+}
+EXPORT_SYMBOL_GPL(enetc_ioctl);
+
+static int enetc_int_vector_init(struct enetc_ndev_priv *priv, int i,
+				 int v_tx_rings)
+{
+	struct enetc_int_vector *v;
+	struct enetc_bdr *bdr;
+	int j, err;
+
+	v = kzalloc(struct_size(v, tx_ring, v_tx_rings), GFP_KERNEL);
+	if (!v)
+		return -ENOMEM;
+
+	priv->int_vector[i] = v;
+	bdr = &v->rx_ring;
+	bdr->index = i;
+	bdr->ndev = priv->ndev;
+	bdr->dev = priv->dev;
+	bdr->bd_count = priv->rx_bd_count;
+	bdr->buffer_offset = ENETC_RXB_PAD;
+	priv->rx_ring[i] = bdr;
+
+	err = __xdp_rxq_info_reg(&bdr->xdp.rxq, priv->ndev, i, 0,
+				 ENETC_RXB_DMA_SIZE_XDP);
+	if (err)
+		goto free_vector;
+
+	err = xdp_rxq_info_reg_mem_model(&bdr->xdp.rxq, MEM_TYPE_PAGE_SHARED,
+					 NULL);
+	if (err) {
+		xdp_rxq_info_unreg(&bdr->xdp.rxq);
+		goto free_vector;
+	}
+
+	/* init defaults for adaptive IC */
+	if (priv->ic_mode & ENETC_IC_RX_ADAPTIVE) {
+		v->rx_ictt = 0x1;
+		v->rx_dim_en = true;
+	}
+
+	INIT_WORK(&v->rx_dim.work, enetc_rx_dim_work);
+	netif_napi_add(priv->ndev, &v->napi, enetc_poll);
+	v->count_tx_rings = v_tx_rings;
+
+	for (j = 0; j < v_tx_rings; j++) {
+		int idx;
+
+		/* default tx ring mapping policy */
+		idx = priv->bdr_int_num * j + i;
+		__set_bit(idx, &v->tx_rings_map);
+		bdr = &v->tx_ring[j];
+		bdr->index = idx;
+		bdr->ndev = priv->ndev;
+		bdr->dev = priv->dev;
+		bdr->bd_count = priv->tx_bd_count;
+		priv->tx_ring[idx] = bdr;
+	}
+
+	return 0;
+
+free_vector:
+	priv->rx_ring[i] = NULL;
+	priv->int_vector[i] = NULL;
+	kfree(v);
+
+	return err;
+}
+
+static void enetc_int_vector_destroy(struct enetc_ndev_priv *priv, int i)
+{
+	struct enetc_int_vector *v = priv->int_vector[i];
+	struct enetc_bdr *rx_ring = &v->rx_ring;
+	int j, tx_ring_index;
+
+	xdp_rxq_info_unreg_mem_model(&rx_ring->xdp.rxq);
+	xdp_rxq_info_unreg(&rx_ring->xdp.rxq);
+	netif_napi_del(&v->napi);
+	cancel_work_sync(&v->rx_dim.work);
+
+	for (j = 0; j < v->count_tx_rings; j++) {
+		tx_ring_index = priv->bdr_int_num * j + i;
+		priv->tx_ring[tx_ring_index] = NULL;
+	}
+
+	priv->rx_ring[i] = NULL;
+	priv->int_vector[i] = NULL;
+	kfree(v);
+}
+
+int enetc_alloc_msix(struct enetc_ndev_priv *priv)
+{
+	struct pci_dev *pdev = priv->si->pdev;
+	int v_tx_rings, v_remainder;
+	int num_stack_tx_queues;
+	int first_xdp_tx_ring;
+	int i, n, err, nvec;
+
+	nvec = ENETC_BDR_INT_BASE_IDX + priv->bdr_int_num;
+	/* allocate MSIX for both messaging and Rx/Tx interrupts */
+	n = pci_alloc_irq_vectors(pdev, nvec, nvec, PCI_IRQ_MSIX);
+
+	if (n < 0)
+		return n;
+
+	if (n != nvec)
+		return -EPERM;
+
+	/* # of tx rings per int vector */
+	v_tx_rings = priv->num_tx_rings / priv->bdr_int_num;
+	v_remainder = priv->num_tx_rings % priv->bdr_int_num;
+
+	for (i = 0; i < priv->bdr_int_num; i++) {
+		/* Distribute the remaining TX rings to the first v_remainder
+		 * interrupt vectors
+		 */
+		int num_tx_rings = i < v_remainder ? v_tx_rings + 1 : v_tx_rings;
+
+		err = enetc_int_vector_init(priv, i, num_tx_rings);
+		if (err)
+			goto fail;
+	}
+
+	num_stack_tx_queues = enetc_num_stack_tx_queues(priv);
+
+	err = netif_set_real_num_tx_queues(priv->ndev, num_stack_tx_queues);
+	if (err)
+		goto fail;
+
+	err = netif_set_real_num_rx_queues(priv->ndev, priv->num_rx_rings);
+	if (err)
+		goto fail;
+
+	priv->min_num_stack_tx_queues = num_possible_cpus();
+	first_xdp_tx_ring = priv->num_tx_rings - num_possible_cpus();
+	priv->xdp_tx_ring = &priv->tx_ring[first_xdp_tx_ring];
+
+	return 0;
+
+fail:
+	while (i--)
+		enetc_int_vector_destroy(priv, i);
+
+	pci_free_irq_vectors(pdev);
+
+	return err;
+}
+EXPORT_SYMBOL_GPL(enetc_alloc_msix);
+
+void enetc_free_msix(struct enetc_ndev_priv *priv)
+{
+	int i;
+
+	for (i = 0; i < priv->bdr_int_num; i++)
+		enetc_int_vector_destroy(priv, i);
+
+	/* disable all MSIX for this device */
+	pci_free_irq_vectors(priv->si->pdev);
+}
+EXPORT_SYMBOL_GPL(enetc_free_msix);
+
+static void enetc_kfree_si(struct enetc_si *si)
+{
+	char *p = (char *)si - si->pad;
+
+	kfree(p);
+}
+
+static void enetc_detect_errata(struct enetc_si *si)
+{
+	if (si->pdev->revision == ENETC_REV1)
+		si->errata = ENETC_ERR_VLAN_ISOL | ENETC_ERR_UCMCSWP;
+}
+
+int enetc_pci_probe(struct pci_dev *pdev, const char *name, int sizeof_priv)
+{
+	struct enetc_si *si, *p;
+	struct enetc_hw *hw;
+	size_t alloc_size;
+	int err, len;
+
+	pcie_flr(pdev);
+	err = pci_enable_device_mem(pdev);
+	if (err)
+		return dev_err_probe(&pdev->dev, err, "device enable failed\n");
+
+	/* set up for high or low dma */
+	err = dma_set_mask_and_coherent(&pdev->dev, DMA_BIT_MASK(64));
+	if (err) {
+		dev_err(&pdev->dev, "DMA configuration failed: 0x%x\n", err);
+		goto err_dma;
+	}
+
+	err = pci_request_mem_regions(pdev, name);
+	if (err) {
+		dev_err(&pdev->dev, "pci_request_regions failed err=%d\n", err);
+		goto err_pci_mem_reg;
+	}
+
+	pci_set_master(pdev);
+
+	alloc_size = sizeof(struct enetc_si);
+	if (sizeof_priv) {
+		/* align priv to 32B */
+		alloc_size = ALIGN(alloc_size, ENETC_SI_ALIGN);
+		alloc_size += sizeof_priv;
+	}
+	/* force 32B alignment for enetc_si */
+	alloc_size += ENETC_SI_ALIGN - 1;
+
+	p = kzalloc(alloc_size, GFP_KERNEL);
+	if (!p) {
+		err = -ENOMEM;
+		goto err_alloc_si;
+	}
+
+	si = PTR_ALIGN(p, ENETC_SI_ALIGN);
+	si->pad = (char *)si - (char *)p;
+
+	pci_set_drvdata(pdev, si);
+	si->pdev = pdev;
+	hw = &si->hw;
+
+	len = pci_resource_len(pdev, ENETC_BAR_REGS);
+	hw->reg = ioremap(pci_resource_start(pdev, ENETC_BAR_REGS), len);
+	if (!hw->reg) {
+		err = -ENXIO;
+		dev_err(&pdev->dev, "ioremap() failed\n");
+		goto err_ioremap;
+	}
+	if (len > ENETC_PORT_BASE)
+		hw->port = hw->reg + ENETC_PORT_BASE;
+	if (len > ENETC_GLOBAL_BASE)
+		hw->global = hw->reg + ENETC_GLOBAL_BASE;
+
+	enetc_detect_errata(si);
+
+	return 0;
+
+err_ioremap:
+	enetc_kfree_si(si);
+err_alloc_si:
+	pci_release_mem_regions(pdev);
+err_pci_mem_reg:
+err_dma:
+	pci_disable_device(pdev);
+
+	return err;
+}
+EXPORT_SYMBOL_GPL(enetc_pci_probe);
+
+void enetc_pci_remove(struct pci_dev *pdev)
+{
+	struct enetc_si *si = pci_get_drvdata(pdev);
+	struct enetc_hw *hw = &si->hw;
+
+	iounmap(hw->reg);
+	enetc_kfree_si(si);
+	pci_release_mem_regions(pdev);
+	pci_disable_device(pdev);
+}
+EXPORT_SYMBOL_GPL(enetc_pci_remove);
+
+static const struct enetc_drvdata enetc_pf_data = {
+	.sysclk_freq = ENETC_CLK_400M,
+	.pmac_offset = ENETC_PMAC_OFFSET,
+	.max_frags = ENETC_MAX_SKB_FRAGS,
+	.eth_ops = &enetc_pf_ethtool_ops,
+};
+
+static const struct enetc_drvdata enetc4_pf_data = {
+	.sysclk_freq = ENETC_CLK_333M,
+	.tx_csum = true,
+	.max_frags = ENETC4_MAX_SKB_FRAGS,
+	.pmac_offset = ENETC4_PMAC_OFFSET,
+	.eth_ops = &enetc4_pf_ethtool_ops,
+};
+
+static const struct enetc_drvdata enetc_vf_data = {
+	.sysclk_freq = ENETC_CLK_400M,
+	.max_frags = ENETC_MAX_SKB_FRAGS,
+	.eth_ops = &enetc_vf_ethtool_ops,
+};
+
+static const struct enetc_platform_info enetc_info[] = {
+	{ .revision = ENETC_REV_1_0,
+	  .dev_id = ENETC_DEV_ID_PF,
+	  .data = &enetc_pf_data,
+	},
+	{ .revision = ENETC_REV_4_1,
+	  .dev_id = NXP_ENETC_PF_DEV_ID,
+	  .data = &enetc4_pf_data,
+	},
+	{ .revision = ENETC_REV_1_0,
+	  .dev_id = ENETC_DEV_ID_VF,
+	  .data = &enetc_vf_data,
+	},
+};
+
+int enetc_get_driver_data(struct enetc_si *si)
+{
+	u16 dev_id = si->pdev->device;
+	int i;
+
+	for (i = 0; i < ARRAY_SIZE(enetc_info); i++) {
+		if (si->revision == enetc_info[i].revision &&
+		    dev_id == enetc_info[i].dev_id) {
+			si->drvdata = enetc_info[i].data;
+
+			return 0;
+		}
+	}
+
+	return -ERANGE;
+}
+EXPORT_SYMBOL_GPL(enetc_get_driver_data);
+
+MODULE_DESCRIPTION("NXP ENETC Ethernet driver");
+MODULE_LICENSE("Dual BSD/GPL");
diff --git a/drivers/net/ethernet/freescale/enetc/enetc.h b/drivers/net/ethernet/freescale/enetc/enetc.h
new file mode 100644
index 000000000000..0ec010a7d640
--- /dev/null
+++ b/drivers/net/ethernet/freescale/enetc/enetc.h
@@ -0,0 +1,700 @@
+/* SPDX-License-Identifier: (GPL-2.0+ OR BSD-3-Clause) */
+/* Copyright 2017-2019 NXP */
+
+#include <linux/timer.h>
+#include <linux/pci.h>
+#include <linux/netdevice.h>
+#include <linux/etherdevice.h>
+#include <linux/dma-mapping.h>
+#include <linux/skbuff.h>
+#include <linux/ethtool.h>
+#include <linux/fsl/ntmp.h>
+#include <linux/if_vlan.h>
+#include <linux/phylink.h>
+#include <linux/dim.h>
+#include <net/xdp.h>
+
+#include "enetc_hw.h"
+#include "enetc4_hw.h"
+
+#define ENETC_MAC_MAXFRM_SIZE	9600
+#define ENETC_MAX_MTU		(ENETC_MAC_MAXFRM_SIZE - \
+				(ETH_FCS_LEN + ETH_HLEN + VLAN_HLEN))
+
+#define ENETC_CBD_DATA_MEM_ALIGN 64
+
+#define ENETC_MADDR_HASH_TBL_SZ	64
+
+enum enetc_mac_addr_type {UC, MC, MADDR_TYPE};
+
+struct enetc_mac_filter {
+	union {
+		char mac_addr[ETH_ALEN];
+		DECLARE_BITMAP(mac_hash_table, ENETC_MADDR_HASH_TBL_SZ);
+	};
+	int mac_addr_cnt;
+};
+
+struct enetc_tx_swbd {
+	union {
+		struct sk_buff *skb;
+		struct xdp_frame *xdp_frame;
+	};
+	dma_addr_t dma;
+	struct page *page;	/* valid only if is_xdp_tx */
+	u16 page_offset;	/* valid only if is_xdp_tx */
+	u16 len;
+	enum dma_data_direction dir;
+	u8 is_dma_page:1;
+	u8 check_wb:1;
+	u8 do_twostep_tstamp:1;
+	u8 is_eof:1;
+	u8 is_xdp_tx:1;
+	u8 is_xdp_redirect:1;
+	u8 qbv_en:1;
+};
+
+struct enetc_skb_cb {
+	u8 flag;
+	bool udp;
+	u16 correction_off;
+	u16 origin_tstamp_off;
+};
+
+#define ENETC_SKB_CB(skb) ((struct enetc_skb_cb *)((skb)->cb))
+
+struct enetc_lso_t {
+	bool	ipv6;
+	bool	tcp;
+	u8	l3_hdr_len;
+	u8	hdr_len; /* LSO header length */
+	u8	l3_start;
+	u16	lso_seg_size;
+	int	total_len; /* total data length, not include LSO header */
+};
+
+#define ENETC_LSO_MAX_DATA_LEN		SZ_256K
+
+#define ENETC_RX_MAXFRM_SIZE	ENETC_MAC_MAXFRM_SIZE
+#define ENETC_RXB_TRUESIZE	2048 /* PAGE_SIZE >> 1 */
+#define ENETC_RXB_PAD		NET_SKB_PAD /* add extra space if needed */
+#define ENETC_RXB_DMA_SIZE	\
+	(SKB_WITH_OVERHEAD(ENETC_RXB_TRUESIZE) - ENETC_RXB_PAD)
+#define ENETC_RXB_DMA_SIZE_XDP	\
+	(SKB_WITH_OVERHEAD(ENETC_RXB_TRUESIZE) - XDP_PACKET_HEADROOM)
+
+struct enetc_rx_swbd {
+	dma_addr_t dma;
+	struct page *page;
+	u16 page_offset;
+	enum dma_data_direction dir;
+	u16 len;
+};
+
+/* ENETC overhead: optional extension BD + 1 BD gap */
+#define ENETC_TXBDS_NEEDED(val)	((val) + 2)
+/* For LS1028A, max # of chained Tx BDs is 15, including head and
+ * extension BD.
+ */
+#define ENETC_MAX_SKB_FRAGS	13
+/* For ENETC v4 and later versions, max # of chained Tx BDs is 63,
+ * including head and extension BD, but the range of MAX_SKB_FRAGS
+ * is 17 ~ 45, so set ENETC4_MAX_SKB_FRAGS to MAX_SKB_FRAGS.
+ */
+#define ENETC4_MAX_SKB_FRAGS		MAX_SKB_FRAGS
+#define ENETC_TXBDS_MAX_NEEDED(x)	ENETC_TXBDS_NEEDED((x) + 1)
+
+struct enetc_ring_stats {
+	unsigned long packets;
+	unsigned long bytes;
+	unsigned long rx_alloc_errs;
+	unsigned long xdp_drops;
+	unsigned long xdp_tx;
+	unsigned long xdp_tx_drops;
+	unsigned long xdp_redirect;
+	unsigned long xdp_redirect_failures;
+	unsigned long recycles;
+	unsigned long recycle_failures;
+	unsigned long win_drop;
+};
+
+struct enetc_xdp_data {
+	struct xdp_rxq_info rxq;
+	struct bpf_prog *prog;
+	int xdp_tx_in_flight;
+};
+
+#define ENETC_RX_RING_DEFAULT_SIZE	2048
+#define ENETC_TX_RING_DEFAULT_SIZE	2048
+#define ENETC_DEFAULT_TX_WORK		(ENETC_TX_RING_DEFAULT_SIZE / 2)
+
+struct enetc_bdr_resource {
+	/* Input arguments saved for teardown */
+	struct device *dev; /* for DMA mapping */
+	size_t bd_count;
+	size_t bd_size;
+
+	/* Resource proper */
+	void *bd_base; /* points to Rx or Tx BD ring */
+	dma_addr_t bd_dma_base;
+	union {
+		struct enetc_tx_swbd *tx_swbd;
+		struct enetc_rx_swbd *rx_swbd;
+	};
+	char *tso_headers;
+	dma_addr_t tso_headers_dma;
+};
+
+struct enetc_bdr {
+	struct device *dev; /* for DMA mapping */
+	struct net_device *ndev;
+	void *bd_base; /* points to Rx or Tx BD ring */
+	union {
+		void __iomem *tpir;
+		void __iomem *rcir;
+	};
+	u16 index;
+	u16 prio;
+	int bd_count; /* # of BDs */
+	int next_to_use;
+	int next_to_clean;
+	union {
+		struct enetc_tx_swbd *tx_swbd;
+		struct enetc_rx_swbd *rx_swbd;
+	};
+	union {
+		void __iomem *tcir; /* Tx */
+		int next_to_alloc; /* Rx */
+	};
+	void __iomem *idr; /* Interrupt Detect Register pointer */
+
+	int buffer_offset;
+	struct enetc_xdp_data xdp;
+
+	struct enetc_ring_stats stats;
+
+	dma_addr_t bd_dma_base;
+	u8 tsd_enable; /* Time specific departure */
+	bool ext_en; /* enable h/w descriptor extensions */
+
+	/* DMA buffer for TSO headers */
+	char *tso_headers;
+	dma_addr_t tso_headers_dma;
+} ____cacheline_aligned_in_smp;
+
+static inline void enetc_bdr_idx_inc(struct enetc_bdr *bdr, int *i)
+{
+	if (unlikely(++*i == bdr->bd_count))
+		*i = 0;
+}
+
+static inline int enetc_bd_unused(struct enetc_bdr *bdr)
+{
+	if (bdr->next_to_clean > bdr->next_to_use)
+		return bdr->next_to_clean - bdr->next_to_use - 1;
+
+	return bdr->bd_count + bdr->next_to_clean - bdr->next_to_use - 1;
+}
+
+static inline int enetc_swbd_unused(struct enetc_bdr *bdr)
+{
+	if (bdr->next_to_clean > bdr->next_to_alloc)
+		return bdr->next_to_clean - bdr->next_to_alloc - 1;
+
+	return bdr->bd_count + bdr->next_to_clean - bdr->next_to_alloc - 1;
+}
+
+/* Control BD ring */
+#define ENETC_CBDR_DEFAULT_SIZE	64
+struct enetc_cbdr {
+	void *bd_base; /* points to Rx or Tx BD ring */
+	void __iomem *pir;
+	void __iomem *cir;
+	void __iomem *mr; /* mode register */
+
+	int bd_count; /* # of BDs */
+	int next_to_use;
+	int next_to_clean;
+
+	dma_addr_t bd_dma_base;
+	struct device *dma_dev;
+};
+
+#define ENETC_TXBD(BDR, i) (&(((union enetc_tx_bd *)((BDR).bd_base))[i]))
+
+static inline union enetc_rx_bd *enetc_rxbd(struct enetc_bdr *rx_ring, int i)
+{
+	int hw_idx = i;
+
+	if (rx_ring->ext_en)
+		hw_idx = 2 * i;
+
+	return &(((union enetc_rx_bd *)rx_ring->bd_base)[hw_idx]);
+}
+
+static inline void enetc_rxbd_next(struct enetc_bdr *rx_ring,
+				   union enetc_rx_bd **old_rxbd, int *old_index)
+{
+	union enetc_rx_bd *new_rxbd = *old_rxbd;
+	int new_index = *old_index;
+
+	new_rxbd++;
+
+	if (rx_ring->ext_en)
+		new_rxbd++;
+
+	if (unlikely(++new_index == rx_ring->bd_count)) {
+		new_rxbd = rx_ring->bd_base;
+		new_index = 0;
+	}
+
+	*old_rxbd = new_rxbd;
+	*old_index = new_index;
+}
+
+static inline union enetc_rx_bd *enetc_rxbd_ext(union enetc_rx_bd *rxbd)
+{
+	return ++rxbd;
+}
+
+struct enetc_msg_swbd {
+	void *vaddr;
+	dma_addr_t dma;
+	int size;
+};
+
+#define ENETC_REV1	0x1
+enum enetc_errata {
+	ENETC_ERR_VLAN_ISOL	= BIT(0),
+	ENETC_ERR_UCMCSWP	= BIT(1),
+};
+
+#define ENETC_SI_F_PSFP BIT(0)
+#define ENETC_SI_F_QBV  BIT(1)
+#define ENETC_SI_F_QBU  BIT(2)
+#define ENETC_SI_F_LSO	BIT(3)
+
+struct enetc_drvdata {
+	u32 pmac_offset; /* Only valid for PSI which supports 802.1Qbu */
+	u8 tx_csum:1;
+	u8 max_frags;
+	u64 sysclk_freq;
+	const struct ethtool_ops *eth_ops;
+};
+
+struct enetc_platform_info {
+	u16 revision;
+	u16 dev_id;
+	const struct enetc_drvdata *data;
+};
+
+struct enetc_si;
+
+/*
+ * This structure defines the some common hooks for ENETC PSI and VSI.
+ * In addition, since VSI only uses the struct enetc_si as its private
+ * driver data, so this structure also define some hooks specifically
+ * for VSI. For VSI-specific hooks, the format is ‘vf_*()’.
+ */
+struct enetc_si_ops {
+	int (*get_rss_table)(struct enetc_si *si, u32 *table, int count);
+	int (*set_rss_table)(struct enetc_si *si, const u32 *table, int count);
+};
+
+/* PCI IEP device data */
+struct enetc_si {
+	struct pci_dev *pdev;
+	struct enetc_hw hw;
+	enum enetc_errata errata;
+
+	struct net_device *ndev; /* back ref. */
+
+	union {
+		struct enetc_cbdr cbd_ring; /* Only ENETC 1.0 */
+		struct ntmp_user ntmp_user; /* ENETC 4.1 and later */
+	};
+
+	int num_rx_rings; /* how many rings are available in the SI */
+	int num_tx_rings;
+	int num_fs_entries;
+	int num_rss; /* number of RSS buckets */
+	unsigned short pad;
+	u16 revision;
+	int hw_features;
+	const struct enetc_drvdata *drvdata;
+	const struct enetc_si_ops *ops;
+
+	struct workqueue_struct *workqueue;
+	struct work_struct rx_mode_task;
+	struct dentry *debugfs_root;
+};
+
+#define ENETC_SI_ALIGN	32
+
+static inline bool is_enetc_rev1(struct enetc_si *si)
+{
+	return si->pdev->revision == ENETC_REV1;
+}
+
+static inline void *enetc_si_priv(const struct enetc_si *si)
+{
+	return (char *)si + ALIGN(sizeof(struct enetc_si), ENETC_SI_ALIGN);
+}
+
+static inline bool enetc_si_is_pf(struct enetc_si *si)
+{
+	return !!(si->hw.port);
+}
+
+static inline int enetc_pf_to_port(struct pci_dev *pf_pdev)
+{
+	switch (pf_pdev->devfn) {
+	case 0:
+		return 0;
+	case 1:
+		return 1;
+	case 2:
+		return 2;
+	case 6:
+		return 3;
+	default:
+		return -1;
+	}
+}
+
+#define ENETC_MAX_NUM_TXQS	8
+#define ENETC_INT_NAME_MAX	(IFNAMSIZ + 8)
+
+struct enetc_int_vector {
+	void __iomem *rbier;
+	void __iomem *tbier_base;
+	void __iomem *ricr1;
+	unsigned long tx_rings_map;
+	int count_tx_rings;
+	u32 rx_ictt;
+	u16 comp_cnt;
+	bool rx_dim_en, rx_napi_work;
+	struct napi_struct napi ____cacheline_aligned_in_smp;
+	struct dim rx_dim ____cacheline_aligned_in_smp;
+	char name[ENETC_INT_NAME_MAX];
+
+	struct enetc_bdr rx_ring;
+	struct enetc_bdr tx_ring[] __counted_by(count_tx_rings);
+} ____cacheline_aligned_in_smp;
+
+struct enetc_cls_rule {
+	struct ethtool_rx_flow_spec fs;
+	int used;
+};
+
+#define ENETC_MAX_BDR_INT	6 /* fixed to max # of available cpus */
+struct psfp_cap {
+	u32 max_streamid;
+	u32 max_psfp_filter;
+	u32 max_psfp_gate;
+	u32 max_psfp_gatelist;
+	u32 max_psfp_meter;
+};
+
+#define ENETC_F_TX_TSTAMP_MASK	0xff
+enum enetc_active_offloads {
+	/* 8 bits reserved for TX timestamp types (hwtstamp_tx_types) */
+	ENETC_F_TX_TSTAMP		= BIT(0),
+	ENETC_F_TX_ONESTEP_SYNC_TSTAMP	= BIT(1),
+
+	ENETC_F_RX_TSTAMP		= BIT(8),
+	ENETC_F_QBV			= BIT(9),
+	ENETC_F_QCI			= BIT(10),
+	ENETC_F_QBU			= BIT(11),
+	ENETC_F_TXCSUM			= BIT(12),
+	ENETC_F_LSO			= BIT(13),
+};
+
+enum enetc_flags_bit {
+	ENETC_TX_ONESTEP_TSTAMP_IN_PROGRESS = 0,
+	ENETC_TX_DOWN,
+};
+
+/* interrupt coalescing modes */
+enum enetc_ic_mode {
+	/* one interrupt per frame */
+	ENETC_IC_NONE = 0,
+	/* activated when int coalescing time is set to a non-0 value */
+	ENETC_IC_RX_MANUAL = BIT(0),
+	ENETC_IC_TX_MANUAL = BIT(1),
+	/* use dynamic interrupt moderation */
+	ENETC_IC_RX_ADAPTIVE = BIT(2),
+};
+
+#define ENETC_RXIC_PKTTHR	min_t(u32, 256, ENETC_RX_RING_DEFAULT_SIZE / 2)
+#define ENETC_TXIC_PKTTHR	min_t(u32, 128, ENETC_TX_RING_DEFAULT_SIZE / 2)
+
+struct enetc_ndev_priv {
+	struct net_device *ndev;
+	struct device *dev; /* dma-mapping device */
+	struct enetc_si *si;
+
+	int bdr_int_num; /* number of Rx/Tx ring interrupts */
+	struct enetc_int_vector *int_vector[ENETC_MAX_BDR_INT];
+	u16 num_rx_rings, num_tx_rings;
+	u16 rx_bd_count, tx_bd_count;
+
+	u16 msg_enable;
+
+	u8 preemptible_tcs;
+	u8 max_frags; /* The maximum number of BDs for fragments */
+
+	enum enetc_active_offloads active_offloads;
+
+	u32 speed; /* store speed for compare update pspeed */
+
+	struct enetc_bdr **xdp_tx_ring;
+	struct enetc_bdr *tx_ring[16];
+	struct enetc_bdr *rx_ring[16];
+	const struct enetc_bdr_resource *tx_res;
+	const struct enetc_bdr_resource *rx_res;
+
+	struct enetc_cls_rule *cls_rules;
+
+	struct psfp_cap psfp_cap;
+
+	/* Minimum number of TX queues required by the network stack */
+	unsigned int min_num_stack_tx_queues;
+
+	struct phylink *phylink;
+	int ic_mode;
+	u32 tx_ictt;
+
+	struct bpf_prog *xdp_prog;
+
+	unsigned long flags;
+
+	struct work_struct	tx_onestep_tstamp;
+	struct sk_buff_head	tx_skbs;
+
+	/* Serialize access to MAC Merge state between ethtool requests
+	 * and link state updates
+	 */
+	struct mutex		mm_lock;
+
+	struct clk *ref_clk; /* RGMII/RMII reference clock */
+	u64 sysclk_freq; /* NETC system clock frequency */
+};
+
+/* Messaging */
+
+/* VF-PF set primary MAC address message format */
+struct enetc_msg_cmd_set_primary_mac {
+	struct enetc_msg_cmd_header header;
+	struct sockaddr mac;
+};
+
+#define ENETC_CBD(R, i)	(&(((struct enetc_cbd *)((R).bd_base))[i]))
+
+#define ENETC_CBDR_TIMEOUT	1000 /* usecs */
+
+/* SI common */
+u32 enetc_port_mac_rd(struct enetc_si *si, u32 reg);
+void enetc_port_mac_wr(struct enetc_si *si, u32 reg, u32 val);
+int enetc_pci_probe(struct pci_dev *pdev, const char *name, int sizeof_priv);
+void enetc_pci_remove(struct pci_dev *pdev);
+int enetc_alloc_msix(struct enetc_ndev_priv *priv);
+void enetc_free_msix(struct enetc_ndev_priv *priv);
+void enetc_get_si_caps(struct enetc_si *si);
+void enetc_init_si_rings_params(struct enetc_ndev_priv *priv);
+int enetc_alloc_si_resources(struct enetc_ndev_priv *priv);
+void enetc_free_si_resources(struct enetc_ndev_priv *priv);
+int enetc_configure_si(struct enetc_ndev_priv *priv);
+int enetc_get_driver_data(struct enetc_si *si);
+void enetc_add_mac_addr_ht_filter(struct enetc_mac_filter *filter,
+				  const unsigned char *addr);
+void enetc_reset_mac_addr_filter(struct enetc_mac_filter *filter);
+
+int enetc_open(struct net_device *ndev);
+int enetc_close(struct net_device *ndev);
+void enetc_start(struct net_device *ndev);
+void enetc_stop(struct net_device *ndev);
+netdev_tx_t enetc_xmit(struct sk_buff *skb, struct net_device *ndev);
+struct net_device_stats *enetc_get_stats(struct net_device *ndev);
+void enetc_set_features(struct net_device *ndev, netdev_features_t features);
+int enetc_ioctl(struct net_device *ndev, struct ifreq *rq, int cmd);
+int enetc_setup_tc_mqprio(struct net_device *ndev, void *type_data);
+void enetc_reset_tc_mqprio(struct net_device *ndev);
+int enetc_setup_bpf(struct net_device *ndev, struct netdev_bpf *bpf);
+int enetc_xdp_xmit(struct net_device *ndev, int num_frames,
+		   struct xdp_frame **frames, u32 flags);
+
+int enetc_hwtstamp_get(struct net_device *ndev,
+		       struct kernel_hwtstamp_config *config);
+int enetc_hwtstamp_set(struct net_device *ndev,
+		       struct kernel_hwtstamp_config *config,
+		       struct netlink_ext_ack *extack);
+
+/* ethtool */
+extern const struct ethtool_ops enetc_pf_ethtool_ops;
+extern const struct ethtool_ops enetc4_pf_ethtool_ops;
+extern const struct ethtool_ops enetc_vf_ethtool_ops;
+void enetc_set_ethtool_ops(struct net_device *ndev);
+void enetc_mm_link_state_update(struct enetc_ndev_priv *priv, bool link);
+void enetc_mm_commit_preemptible_tcs(struct enetc_ndev_priv *priv);
+
+/* control buffer descriptor ring (CBDR) */
+int enetc_setup_cbdr(struct device *dev, struct enetc_hw *hw, int bd_count,
+		     struct enetc_cbdr *cbdr);
+void enetc_teardown_cbdr(struct enetc_cbdr *cbdr);
+int enetc4_setup_cbdr(struct enetc_si *si);
+void enetc4_teardown_cbdr(struct enetc_si *si);
+int enetc_set_mac_flt_entry(struct enetc_si *si, int index,
+			    char *mac_addr, int si_map);
+int enetc_clear_mac_flt_entry(struct enetc_si *si, int index);
+int enetc_set_fs_entry(struct enetc_si *si, struct enetc_cmd_rfse *rfse,
+		       int index);
+void enetc_set_rss_key(struct enetc_si *si, const u8 *bytes);
+int enetc_get_rss_table(struct enetc_si *si, u32 *table, int count);
+int enetc_set_rss_table(struct enetc_si *si, const u32 *table, int count);
+int enetc_send_cmd(struct enetc_si *si, struct enetc_cbd *cbd);
+int enetc4_get_rss_table(struct enetc_si *si, u32 *table, int count);
+int enetc4_set_rss_table(struct enetc_si *si, const u32 *table, int count);
+
+static inline void *enetc_cbd_alloc_data_mem(struct enetc_si *si,
+					     struct enetc_cbd *cbd,
+					     int size, dma_addr_t *dma,
+					     void **data_align)
+{
+	struct enetc_cbdr *ring = &si->cbd_ring;
+	dma_addr_t dma_align;
+	void *data;
+
+	data = dma_alloc_coherent(ring->dma_dev,
+				  size + ENETC_CBD_DATA_MEM_ALIGN,
+				  dma, GFP_KERNEL);
+	if (!data) {
+		dev_err(ring->dma_dev, "CBD alloc data memory failed!\n");
+		return NULL;
+	}
+
+	dma_align = ALIGN(*dma, ENETC_CBD_DATA_MEM_ALIGN);
+	*data_align = PTR_ALIGN(data, ENETC_CBD_DATA_MEM_ALIGN);
+
+	cbd->addr[0] = cpu_to_le32(lower_32_bits(dma_align));
+	cbd->addr[1] = cpu_to_le32(upper_32_bits(dma_align));
+	cbd->length = cpu_to_le16(size);
+
+	return data;
+}
+
+static inline void enetc_cbd_free_data_mem(struct enetc_si *si, int size,
+					   void *data, dma_addr_t *dma)
+{
+	struct enetc_cbdr *ring = &si->cbd_ring;
+
+	dma_free_coherent(ring->dma_dev, size + ENETC_CBD_DATA_MEM_ALIGN,
+			  data, *dma);
+}
+
+void enetc_reset_ptcmsdur(struct enetc_hw *hw);
+void enetc_set_ptcmsdur(struct enetc_hw *hw, u32 *queue_max_sdu);
+
+static inline bool enetc_ptp_clock_is_enabled(struct enetc_si *si)
+{
+	if (is_enetc_rev1(si))
+		return IS_ENABLED(CONFIG_FSL_ENETC_PTP_CLOCK);
+
+	return IS_ENABLED(CONFIG_PTP_NETC_V4_TIMER);
+}
+
+#ifdef CONFIG_FSL_ENETC_QOS
+int enetc_qos_query_caps(struct net_device *ndev, void *type_data);
+int enetc_setup_tc_taprio(struct net_device *ndev, void *type_data);
+void enetc_sched_speed_set(struct enetc_ndev_priv *priv, int speed);
+int enetc_setup_tc_cbs(struct net_device *ndev, void *type_data);
+int enetc_setup_tc_txtime(struct net_device *ndev, void *type_data);
+int enetc_setup_tc_block_cb(enum tc_setup_type type, void *type_data,
+			    void *cb_priv);
+int enetc_setup_tc_psfp(struct net_device *ndev, void *type_data);
+int enetc_psfp_init(struct enetc_ndev_priv *priv);
+int enetc_psfp_clean(struct enetc_ndev_priv *priv);
+int enetc_set_psfp(struct net_device *ndev, bool en);
+
+static inline void enetc_get_max_cap(struct enetc_ndev_priv *priv)
+{
+	struct enetc_hw *hw = &priv->si->hw;
+	u32 reg;
+
+	reg = enetc_port_rd(hw, ENETC_PSIDCAPR);
+	priv->psfp_cap.max_streamid = reg & ENETC_PSIDCAPR_MSK;
+	/* Port stream filter capability */
+	reg = enetc_port_rd(hw, ENETC_PSFCAPR);
+	priv->psfp_cap.max_psfp_filter = reg & ENETC_PSFCAPR_MSK;
+	/* Port stream gate capability */
+	reg = enetc_port_rd(hw, ENETC_PSGCAPR);
+	priv->psfp_cap.max_psfp_gate = (reg & ENETC_PSGCAPR_SGIT_MSK);
+	priv->psfp_cap.max_psfp_gatelist = (reg & ENETC_PSGCAPR_GCL_MSK) >> 16;
+	/* Port flow meter capability */
+	reg = enetc_port_rd(hw, ENETC_PFMCAPR);
+	priv->psfp_cap.max_psfp_meter = reg & ENETC_PFMCAPR_MSK;
+}
+
+static inline int enetc_psfp_enable(struct enetc_ndev_priv *priv)
+{
+	struct enetc_hw *hw = &priv->si->hw;
+	int err;
+
+	enetc_get_max_cap(priv);
+
+	err = enetc_psfp_init(priv);
+	if (err)
+		return err;
+
+	enetc_wr(hw, ENETC_PPSFPMR, enetc_rd(hw, ENETC_PPSFPMR) |
+		 ENETC_PPSFPMR_PSFPEN | ENETC_PPSFPMR_VS |
+		 ENETC_PPSFPMR_PVC | ENETC_PPSFPMR_PVZC);
+
+	return 0;
+}
+
+static inline int enetc_psfp_disable(struct enetc_ndev_priv *priv)
+{
+	struct enetc_hw *hw = &priv->si->hw;
+	int err;
+
+	err = enetc_psfp_clean(priv);
+	if (err)
+		return err;
+
+	enetc_wr(hw, ENETC_PPSFPMR, enetc_rd(hw, ENETC_PPSFPMR) &
+		 ~ENETC_PPSFPMR_PSFPEN & ~ENETC_PPSFPMR_VS &
+		 ~ENETC_PPSFPMR_PVC & ~ENETC_PPSFPMR_PVZC);
+
+	memset(&priv->psfp_cap, 0, sizeof(struct psfp_cap));
+
+	return 0;
+}
+
+#else
+#define enetc_qos_query_caps(ndev, type_data) -EOPNOTSUPP
+#define enetc_setup_tc_taprio(ndev, type_data) -EOPNOTSUPP
+#define enetc_sched_speed_set(priv, speed) (void)0
+#define enetc_setup_tc_cbs(ndev, type_data) -EOPNOTSUPP
+#define enetc_setup_tc_txtime(ndev, type_data) -EOPNOTSUPP
+#define enetc_setup_tc_psfp(ndev, type_data) -EOPNOTSUPP
+#define enetc_setup_tc_block_cb NULL
+
+#define enetc_get_max_cap(p)		\
+	memset(&((p)->psfp_cap), 0, sizeof(struct psfp_cap))
+
+static inline int enetc_psfp_enable(struct enetc_ndev_priv *priv)
+{
+	return 0;
+}
+
+static inline int enetc_psfp_disable(struct enetc_ndev_priv *priv)
+{
+	return 0;
+}
+
+static inline int enetc_set_psfp(struct net_device *ndev, bool en)
+{
+	return 0;
+}
+#endif
diff --git a/drivers/net/ethernet/freescale/enetc/enetc4_debugfs.c b/drivers/net/ethernet/freescale/enetc/enetc4_debugfs.c
new file mode 100644
index 000000000000..1b1591dce73d
--- /dev/null
+++ b/drivers/net/ethernet/freescale/enetc/enetc4_debugfs.c
@@ -0,0 +1,90 @@
+// SPDX-License-Identifier: GPL-2.0+
+/* Copyright 2025 NXP */
+
+#include <linux/device.h>
+#include <linux/debugfs.h>
+#include <linux/seq_file.h>
+#include <linux/string_choices.h>
+
+#include "enetc_pf.h"
+#include "enetc4_debugfs.h"
+
+static void enetc_show_si_mac_hash_filter(struct seq_file *s, int i)
+{
+	struct enetc_si *si = s->private;
+	struct enetc_hw *hw = &si->hw;
+	u32 hash_h, hash_l;
+
+	hash_l = enetc_port_rd(hw, ENETC4_PSIUMHFR0(i));
+	hash_h = enetc_port_rd(hw, ENETC4_PSIUMHFR1(i));
+	seq_printf(s, "SI %d unicast MAC hash filter: 0x%08x%08x\n",
+		   i, hash_h, hash_l);
+
+	hash_l = enetc_port_rd(hw, ENETC4_PSIMMHFR0(i));
+	hash_h = enetc_port_rd(hw, ENETC4_PSIMMHFR1(i));
+	seq_printf(s, "SI %d multicast MAC hash filter: 0x%08x%08x\n",
+		   i, hash_h, hash_l);
+}
+
+static int enetc_mac_filter_show(struct seq_file *s, void *data)
+{
+	struct enetc_si *si = s->private;
+	struct enetc_hw *hw = &si->hw;
+	struct maft_entry_data maft;
+	struct enetc_pf *pf;
+	int i, err, num_si;
+	u32 val;
+
+	pf = enetc_si_priv(si);
+	num_si = pf->caps.num_vsi + 1;
+
+	val = enetc_port_rd(hw, ENETC4_PSIPMMR);
+	for (i = 0; i < num_si; i++) {
+		seq_printf(s, "SI %d Unicast Promiscuous mode: %s\n", i,
+			   str_enabled_disabled(PSIPMMR_SI_MAC_UP(i) & val));
+		seq_printf(s, "SI %d Multicast Promiscuous mode: %s\n", i,
+			   str_enabled_disabled(PSIPMMR_SI_MAC_MP(i) & val));
+	}
+
+	/* MAC hash filter table */
+	for (i = 0; i < num_si; i++)
+		enetc_show_si_mac_hash_filter(s, i);
+
+	if (!pf->num_mfe)
+		return 0;
+
+	/* MAC address filter table */
+	seq_puts(s, "MAC address filter table\n");
+	for (i = 0; i < pf->num_mfe; i++) {
+		memset(&maft, 0, sizeof(maft));
+		err = ntmp_maft_query_entry(&si->ntmp_user, i, &maft);
+		if (err)
+			return err;
+
+		seq_printf(s, "Entry %d, MAC: %pM, SI bitmap: 0x%04x\n", i,
+			   maft.keye.mac_addr, le16_to_cpu(maft.cfge.si_bitmap));
+	}
+
+	return 0;
+}
+DEFINE_SHOW_ATTRIBUTE(enetc_mac_filter);
+
+void enetc_create_debugfs(struct enetc_si *si)
+{
+	struct net_device *ndev = si->ndev;
+	struct dentry *root;
+
+	root = debugfs_create_dir(netdev_name(ndev), NULL);
+	if (IS_ERR(root))
+		return;
+
+	si->debugfs_root = root;
+
+	debugfs_create_file("mac_filter", 0444, root, si, &enetc_mac_filter_fops);
+}
+
+void enetc_remove_debugfs(struct enetc_si *si)
+{
+	debugfs_remove(si->debugfs_root);
+	si->debugfs_root = NULL;
+}
diff --git a/drivers/net/ethernet/freescale/enetc/enetc4_debugfs.h b/drivers/net/ethernet/freescale/enetc/enetc4_debugfs.h
new file mode 100644
index 000000000000..96caca35f79d
--- /dev/null
+++ b/drivers/net/ethernet/freescale/enetc/enetc4_debugfs.h
@@ -0,0 +1,20 @@
+/* SPDX-License-Identifier: (GPL-2.0+ OR BSD-3-Clause) */
+/* Copyright 2025 NXP */
+
+#ifndef __ENETC4_DEBUGFS_H
+#define __ENETC4_DEBUGFS_H
+
+#if IS_ENABLED(CONFIG_DEBUG_FS)
+void enetc_create_debugfs(struct enetc_si *si);
+void enetc_remove_debugfs(struct enetc_si *si);
+#else
+static inline void enetc_create_debugfs(struct enetc_si *si)
+{
+}
+
+static inline void enetc_remove_debugfs(struct enetc_si *si)
+{
+}
+#endif
+
+#endif
diff --git a/drivers/net/ethernet/freescale/enetc/enetc4_hw.h b/drivers/net/ethernet/freescale/enetc/enetc4_hw.h
new file mode 100644
index 000000000000..19bf0e89cdc2
--- /dev/null
+++ b/drivers/net/ethernet/freescale/enetc/enetc4_hw.h
@@ -0,0 +1,196 @@
+/* SPDX-License-Identifier: (GPL-2.0+ OR BSD-3-Clause) */
+/*
+ * This header file defines the register offsets and bit fields
+ * of ENETC4 PF and VFs. Note that the same registers as ENETC
+ * version 1.0 are defined in the enetc_hw.h file.
+ *
+ * Copyright 2024 NXP
+ */
+#ifndef __ENETC4_HW_H_
+#define __ENETC4_HW_H_
+
+#define NXP_ENETC_VENDOR_ID		0x1131
+#define NXP_ENETC_PF_DEV_ID		0xe101
+
+/**********************Station interface registers************************/
+/* Station interface LSO segmentation flag mask register 0/1 */
+#define ENETC4_SILSOSFMR0		0x1300
+#define  SILSOSFMR0_TCP_MID_SEG		GENMASK(27, 16)
+#define  SILSOSFMR0_TCP_1ST_SEG		GENMASK(11, 0)
+#define  SILSOSFMR0_VAL_SET(first, mid)	(FIELD_PREP(SILSOSFMR0_TCP_MID_SEG, mid) | \
+					 FIELD_PREP(SILSOSFMR0_TCP_1ST_SEG, first))
+
+#define ENETC4_SILSOSFMR1		0x1304
+#define  SILSOSFMR1_TCP_LAST_SEG	GENMASK(11, 0)
+#define   ENETC4_TCP_FLAGS_FIN		BIT(0)
+#define   ENETC4_TCP_FLAGS_SYN		BIT(1)
+#define   ENETC4_TCP_FLAGS_RST		BIT(2)
+#define   ENETC4_TCP_FLAGS_PSH		BIT(3)
+#define   ENETC4_TCP_FLAGS_ACK		BIT(4)
+#define   ENETC4_TCP_FLAGS_URG		BIT(5)
+#define   ENETC4_TCP_FLAGS_ECE		BIT(6)
+#define   ENETC4_TCP_FLAGS_CWR		BIT(7)
+#define   ENETC4_TCP_FLAGS_NS		BIT(8)
+/* According to tso_build_hdr(), clear all special flags for not last packet. */
+#define ENETC4_TCP_NL_SEG_FLAGS_DMASK	(ENETC4_TCP_FLAGS_FIN | \
+					 ENETC4_TCP_FLAGS_RST | ENETC4_TCP_FLAGS_PSH)
+
+/***************************ENETC port registers**************************/
+#define ENETC4_ECAPR0			0x0
+#define  ECAPR0_RFS			BIT(2)
+#define  ECAPR0_TSD			BIT(5)
+#define  ECAPR0_RSS			BIT(8)
+#define  ECAPR0_RSC			BIT(9)
+#define  ECAPR0_LSO			BIT(10)
+#define  ECAPR0_WO			BIT(13)
+
+#define ENETC4_ECAPR1			0x4
+#define  ECAPR1_NUM_TCS			GENMASK(6, 4)
+#define  ECAPR1_NUM_MCH			GENMASK(9, 8)
+#define  ECAPR1_NUM_UCH			GENMASK(11, 10)
+#define  ECAPR1_NUM_MSIX		GENMASK(22, 12)
+#define  ECAPR1_NUM_VSI			GENMASK(27, 24)
+#define  ECAPR1_NUM_IPV			BIT(31)
+
+#define ENETC4_ECAPR2			0x8
+#define  ECAPR2_NUM_TX_BDR		GENMASK(9, 0)
+#define  ECAPR2_NUM_RX_BDR		GENMASK(25, 16)
+
+#define ENETC4_PMR			0x10
+#define  PMR_SI_EN(a)			BIT((16 + (a)))
+
+/* Port Pause ON/OFF threshold register */
+#define ENETC4_PPAUONTR			0x108
+#define ENETC4_PPAUOFFTR		0x10c
+
+/* Port Station interface promiscuous MAC mode register */
+#define ENETC4_PSIPMMR			0x200
+#define  PSIPMMR_SI_MAC_UP(a)		BIT(a) /* a = SI index */
+#define  PSIPMMR_SI_MAC_MP(a)		BIT((a) + 16)
+
+/* Port Station interface promiscuous VLAN mode register */
+#define ENETC4_PSIPVMR			0x204
+
+/* Port RSS key register n. n = 0,1,2,...,9 */
+#define ENETC4_PRSSKR(n)		((n) * 0x4 + 0x250)
+
+/* Port station interface MAC address filtering capability register */
+#define ENETC4_PSIMAFCAPR		0x280
+#define  PSIMAFCAPR_NUM_MAC_AFTE	GENMASK(11, 0)
+
+/* Port station interface VLAN filtering capability register */
+#define ENETC4_PSIVLANFCAPR		0x2c0
+#define  PSIVLANFCAPR_NUM_VLAN_FTE	GENMASK(11, 0)
+
+/* Port station interface VLAN filtering mode register */
+#define ENETC4_PSIVLANFMR		0x2c4
+#define  PSIVLANFMR_VS			BIT(0)
+
+/* Port Station interface a primary MAC address registers */
+#define ENETC4_PSIPMAR0(a)		((a) * 0x80 + 0x2000)
+#define ENETC4_PSIPMAR1(a)		((a) * 0x80 + 0x2004)
+
+/* Port station interface a configuration register 0/2 */
+#define ENETC4_PSICFGR0(a)		((a) * 0x80 + 0x2010)
+#define  PSICFGR0_VASE			BIT(13)
+#define  PSICFGR0_ASE			BIT(15)
+#define  PSICFGR0_ANTI_SPOOFING		(PSICFGR0_VASE | PSICFGR0_ASE)
+
+#define ENETC4_PSICFGR2(a)		((a) * 0x80 + 0x2018)
+#define  PSICFGR2_NUM_MSIX		GENMASK(5, 0)
+
+/* Port station interface a unicast MAC hash filter register 0/1 */
+#define ENETC4_PSIUMHFR0(a)		((a) * 0x80 + 0x2050)
+#define ENETC4_PSIUMHFR1(a)		((a) * 0x80 + 0x2054)
+
+/* Port station interface a multicast MAC hash filter register 0/1 */
+#define ENETC4_PSIMMHFR0(a)		((a) * 0x80 + 0x2058)
+#define ENETC4_PSIMMHFR1(a)		((a) * 0x80 + 0x205c)
+
+/* Port station interface a VLAN hash filter register 0/1 */
+#define ENETC4_PSIVHFR0(a)		((a) * 0x80 + 0x2060)
+#define ENETC4_PSIVHFR1(a)		((a) * 0x80 + 0x2064)
+
+#define ENETC4_PMCAPR			0x4004
+#define  PMCAPR_HD			BIT(8)
+#define  PMCAPR_FP			GENMASK(10, 9)
+
+/* Port configuration register */
+#define ENETC4_PCR			0x4010
+#define  PCR_HDR_FMT			BIT(0)
+#define  PCR_L2DOSE			BIT(4)
+#define  PCR_TIMER_CS			BIT(8)
+#define  PCR_PSPEED			GENMASK(29, 16)
+#define  PCR_PSPEED_VAL(speed)		(((speed) / 10 - 1) << 16)
+
+/* Port MAC address register 0/1 */
+#define ENETC4_PMAR0			0x4020
+#define ENETC4_PMAR1			0x4024
+
+/* Port operational register */
+#define ENETC4_POR			0x4100
+
+/* Port traffic class a transmit maximum SDU register */
+#define ENETC4_PTCTMSDUR(a)		((a) * 0x20 + 0x4208)
+#define  PTCTMSDUR_MAXSDU		GENMASK(15, 0)
+#define  PTCTMSDUR_SDU_TYPE		GENMASK(17, 16)
+#define   SDU_TYPE_PPDU			0
+#define   SDU_TYPE_MPDU			1
+#define   SDU_TYPE_MSDU			2
+
+#define ENETC4_PMAC_OFFSET		0x400
+#define ENETC4_PM_CMD_CFG(mac)		(0x5008 + (mac) * 0x400)
+#define  PM_CMD_CFG_TX_EN		BIT(0)
+#define  PM_CMD_CFG_RX_EN		BIT(1)
+#define  PM_CMD_CFG_PAUSE_FWD		BIT(7)
+#define  PM_CMD_CFG_PAUSE_IGN		BIT(8)
+#define  PM_CMD_CFG_TX_ADDR_INS		BIT(9)
+#define  PM_CMD_CFG_LOOP_EN		BIT(10)
+#define  PM_CMD_CFG_LPBK_MODE		GENMASK(12, 11)
+#define   LPBCK_MODE_EXT_TX_CLK		0
+#define   LPBCK_MODE_MAC_LEVEL		1
+#define   LPBCK_MODE_INT_TX_CLK		2
+#define  PM_CMD_CFG_CNT_FRM_EN		BIT(13)
+#define  PM_CMD_CFG_TXP			BIT(15)
+#define  PM_CMD_CFG_SEND_IDLE		BIT(16)
+#define  PM_CMD_CFG_HD_FCEN		BIT(18)
+#define  PM_CMD_CFG_SFD			BIT(21)
+#define  PM_CMD_CFG_TX_FLUSH		BIT(22)
+#define  PM_CMD_CFG_TX_LOWP_EN		BIT(23)
+#define  PM_CMD_CFG_RX_LOWP_EMPTY	BIT(24)
+#define  PM_CMD_CFG_SWR			BIT(26)
+#define  PM_CMD_CFG_TS_MODE		BIT(30)
+#define  PM_CMD_CFG_MG			BIT(31)
+
+/* Port MAC 0/1 Maximum Frame Length Register */
+#define ENETC4_PM_MAXFRM(mac)		(0x5014 + (mac) * 0x400)
+
+/* Port MAC 0/1 Pause Quanta Register */
+#define ENETC4_PM_PAUSE_QUANTA(mac)	(0x5054 + (mac) * 0x400)
+
+/* Port MAC 0/1 Pause Quanta Threshold Register */
+#define ENETC4_PM_PAUSE_THRESH(mac)	(0x5064 + (mac) * 0x400)
+
+#define ENETC4_PM_SINGLE_STEP(mac)	(0x50c0 + (mac) * 0x400)
+#define  PM_SINGLE_STEP_CH		BIT(6)
+#define  PM_SINGLE_STEP_OFFSET		GENMASK(15, 7)
+#define  PM_SINGLE_STEP_OFFSET_SET(o)	FIELD_PREP(PM_SINGLE_STEP_OFFSET, o)
+#define  PM_SINGLE_STEP_EN		BIT(31)
+
+/* Port MAC 0 Interface Mode Control Register */
+#define ENETC4_PM_IF_MODE(mac)		(0x5300 + (mac) * 0x400)
+#define  PM_IF_MODE_IFMODE		GENMASK(2, 0)
+#define   IFMODE_XGMII			0
+#define   IFMODE_RMII			3
+#define   IFMODE_RGMII			4
+#define   IFMODE_SGMII			5
+#define  PM_IF_MODE_REVMII		BIT(3)
+#define  PM_IF_MODE_M10			BIT(4)
+#define  PM_IF_MODE_HD			BIT(6)
+#define  PM_IF_MODE_SSP			GENMASK(14, 13)
+#define   SSP_100M			0
+#define   SSP_10M			1
+#define   SSP_1G			2
+#define  PM_IF_MODE_ENA			BIT(15)
+
+#endif
diff --git a/drivers/net/ethernet/freescale/enetc/enetc4_pf.c b/drivers/net/ethernet/freescale/enetc/enetc4_pf.c
new file mode 100644
index 000000000000..82c443b28b15
--- /dev/null
+++ b/drivers/net/ethernet/freescale/enetc/enetc4_pf.c
@@ -0,0 +1,1087 @@
+// SPDX-License-Identifier: (GPL-2.0+ OR BSD-3-Clause)
+/* Copyright 2024 NXP */
+
+#include <linux/clk.h>
+#include <linux/module.h>
+#include <linux/of_net.h>
+#include <linux/of_platform.h>
+#include <linux/unaligned.h>
+
+#include "enetc_pf_common.h"
+#include "enetc4_debugfs.h"
+
+#define ENETC_SI_MAX_RING_NUM	8
+
+#define ENETC_MAC_FILTER_TYPE_UC	BIT(0)
+#define ENETC_MAC_FILTER_TYPE_MC	BIT(1)
+#define ENETC_MAC_FILTER_TYPE_ALL	(ENETC_MAC_FILTER_TYPE_UC | \
+					 ENETC_MAC_FILTER_TYPE_MC)
+
+struct enetc_mac_addr {
+	u8 addr[ETH_ALEN];
+};
+
+static void enetc4_get_port_caps(struct enetc_pf *pf)
+{
+	struct enetc_hw *hw = &pf->si->hw;
+	u32 val;
+
+	val = enetc_port_rd(hw, ENETC4_ECAPR1);
+	pf->caps.num_vsi = (val & ECAPR1_NUM_VSI) >> 24;
+	pf->caps.num_msix = ((val & ECAPR1_NUM_MSIX) >> 12) + 1;
+
+	val = enetc_port_rd(hw, ENETC4_ECAPR2);
+	pf->caps.num_rx_bdr = (val & ECAPR2_NUM_RX_BDR) >> 16;
+	pf->caps.num_tx_bdr = val & ECAPR2_NUM_TX_BDR;
+
+	val = enetc_port_rd(hw, ENETC4_PMCAPR);
+	pf->caps.half_duplex = (val & PMCAPR_HD) ? 1 : 0;
+
+	val = enetc_port_rd(hw, ENETC4_PSIMAFCAPR);
+	pf->caps.mac_filter_num = val & PSIMAFCAPR_NUM_MAC_AFTE;
+}
+
+static void enetc4_pf_set_si_primary_mac(struct enetc_hw *hw, int si,
+					 const u8 *addr)
+{
+	u16 lower = get_unaligned_le16(addr + 4);
+	u32 upper = get_unaligned_le32(addr);
+
+	if (si != 0) {
+		__raw_writel(upper, hw->port + ENETC4_PSIPMAR0(si));
+		__raw_writew(lower, hw->port + ENETC4_PSIPMAR1(si));
+	} else {
+		__raw_writel(upper, hw->port + ENETC4_PMAR0);
+		__raw_writew(lower, hw->port + ENETC4_PMAR1);
+	}
+}
+
+static void enetc4_pf_get_si_primary_mac(struct enetc_hw *hw, int si,
+					 u8 *addr)
+{
+	u32 upper;
+	u16 lower;
+
+	upper = __raw_readl(hw->port + ENETC4_PSIPMAR0(si));
+	lower = __raw_readw(hw->port + ENETC4_PSIPMAR1(si));
+
+	put_unaligned_le32(upper, addr);
+	put_unaligned_le16(lower, addr + 4);
+}
+
+static void enetc4_pf_set_si_mac_promisc(struct enetc_hw *hw, int si,
+					 bool uc_promisc, bool mc_promisc)
+{
+	u32 val = enetc_port_rd(hw, ENETC4_PSIPMMR);
+
+	if (uc_promisc)
+		val |= PSIPMMR_SI_MAC_UP(si);
+	else
+		val &= ~PSIPMMR_SI_MAC_UP(si);
+
+	if (mc_promisc)
+		val |= PSIPMMR_SI_MAC_MP(si);
+	else
+		val &= ~PSIPMMR_SI_MAC_MP(si);
+
+	enetc_port_wr(hw, ENETC4_PSIPMMR, val);
+}
+
+static void enetc4_pf_set_si_uc_hash_filter(struct enetc_hw *hw, int si,
+					    u64 hash)
+{
+	enetc_port_wr(hw, ENETC4_PSIUMHFR0(si), lower_32_bits(hash));
+	enetc_port_wr(hw, ENETC4_PSIUMHFR1(si), upper_32_bits(hash));
+}
+
+static void enetc4_pf_set_si_mc_hash_filter(struct enetc_hw *hw, int si,
+					    u64 hash)
+{
+	enetc_port_wr(hw, ENETC4_PSIMMHFR0(si), lower_32_bits(hash));
+	enetc_port_wr(hw, ENETC4_PSIMMHFR1(si), upper_32_bits(hash));
+}
+
+static void enetc4_pf_set_loopback(struct net_device *ndev, bool en)
+{
+	struct enetc_ndev_priv *priv = netdev_priv(ndev);
+	struct enetc_si *si = priv->si;
+	u32 val;
+
+	val = enetc_port_mac_rd(si, ENETC4_PM_CMD_CFG(0));
+	val = u32_replace_bits(val, en ? 1 : 0, PM_CMD_CFG_LOOP_EN);
+	/* Default to select MAC level loopback mode if loopback is enabled. */
+	val = u32_replace_bits(val, en ? LPBCK_MODE_MAC_LEVEL : 0,
+			       PM_CMD_CFG_LPBK_MODE);
+
+	enetc_port_mac_wr(si, ENETC4_PM_CMD_CFG(0), val);
+}
+
+static void enetc4_pf_clear_maft_entries(struct enetc_pf *pf)
+{
+	int i;
+
+	for (i = 0; i < pf->num_mfe; i++)
+		ntmp_maft_delete_entry(&pf->si->ntmp_user, i);
+
+	pf->num_mfe = 0;
+}
+
+static int enetc4_pf_add_maft_entries(struct enetc_pf *pf,
+				      struct enetc_mac_addr *mac,
+				      int mac_cnt)
+{
+	struct maft_entry_data maft = {};
+	u16 si_bit = BIT(0);
+	int i, err;
+
+	maft.cfge.si_bitmap = cpu_to_le16(si_bit);
+	for (i = 0; i < mac_cnt; i++) {
+		ether_addr_copy(maft.keye.mac_addr, mac[i].addr);
+		err = ntmp_maft_add_entry(&pf->si->ntmp_user, i, &maft);
+		if (unlikely(err)) {
+			pf->num_mfe = i;
+			goto clear_maft_entries;
+		}
+	}
+
+	pf->num_mfe = mac_cnt;
+
+	return 0;
+
+clear_maft_entries:
+	enetc4_pf_clear_maft_entries(pf);
+
+	return  err;
+}
+
+static int enetc4_pf_set_uc_exact_filter(struct enetc_pf *pf)
+{
+	int max_num_mfe = pf->caps.mac_filter_num;
+	struct enetc_mac_filter mac_filter = {};
+	struct net_device *ndev = pf->si->ndev;
+	struct enetc_hw *hw = &pf->si->hw;
+	struct enetc_mac_addr *mac_tbl;
+	struct netdev_hw_addr *ha;
+	int i = 0, err;
+	int mac_cnt;
+
+	netif_addr_lock_bh(ndev);
+
+	mac_cnt = netdev_uc_count(ndev);
+	if (!mac_cnt) {
+		netif_addr_unlock_bh(ndev);
+		/* clear both MAC hash and exact filters */
+		enetc4_pf_set_si_uc_hash_filter(hw, 0, 0);
+		enetc4_pf_clear_maft_entries(pf);
+
+		return 0;
+	}
+
+	if (mac_cnt > max_num_mfe) {
+		err = -ENOSPC;
+		goto unlock_netif_addr;
+	}
+
+	mac_tbl = kcalloc(mac_cnt, sizeof(*mac_tbl), GFP_ATOMIC);
+	if (!mac_tbl) {
+		err = -ENOMEM;
+		goto unlock_netif_addr;
+	}
+
+	netdev_for_each_uc_addr(ha, ndev) {
+		enetc_add_mac_addr_ht_filter(&mac_filter, ha->addr);
+		ether_addr_copy(mac_tbl[i++].addr, ha->addr);
+	}
+
+	netif_addr_unlock_bh(ndev);
+
+	/* Set temporary unicast hash filters in case of Rx loss when
+	 * updating MAC address filter table
+	 */
+	enetc4_pf_set_si_uc_hash_filter(hw, 0, *mac_filter.mac_hash_table);
+	enetc4_pf_clear_maft_entries(pf);
+
+	if (!enetc4_pf_add_maft_entries(pf, mac_tbl, i))
+		enetc4_pf_set_si_uc_hash_filter(hw, 0, 0);
+
+	kfree(mac_tbl);
+
+	return 0;
+
+unlock_netif_addr:
+	netif_addr_unlock_bh(ndev);
+
+	return err;
+}
+
+static void enetc4_pf_set_mac_hash_filter(struct enetc_pf *pf, int type)
+{
+	struct net_device *ndev = pf->si->ndev;
+	struct enetc_mac_filter *mac_filter;
+	struct enetc_hw *hw = &pf->si->hw;
+	struct netdev_hw_addr *ha;
+
+	netif_addr_lock_bh(ndev);
+	if (type & ENETC_MAC_FILTER_TYPE_UC) {
+		mac_filter = &pf->mac_filter[UC];
+		enetc_reset_mac_addr_filter(mac_filter);
+		netdev_for_each_uc_addr(ha, ndev)
+			enetc_add_mac_addr_ht_filter(mac_filter, ha->addr);
+
+		enetc4_pf_set_si_uc_hash_filter(hw, 0,
+						*mac_filter->mac_hash_table);
+	}
+
+	if (type & ENETC_MAC_FILTER_TYPE_MC) {
+		mac_filter = &pf->mac_filter[MC];
+		enetc_reset_mac_addr_filter(mac_filter);
+		netdev_for_each_mc_addr(ha, ndev)
+			enetc_add_mac_addr_ht_filter(mac_filter, ha->addr);
+
+		enetc4_pf_set_si_mc_hash_filter(hw, 0,
+						*mac_filter->mac_hash_table);
+	}
+	netif_addr_unlock_bh(ndev);
+}
+
+static void enetc4_pf_set_mac_filter(struct enetc_pf *pf, int type)
+{
+	/* Currently, the MAC address filter table (MAFT) only has 4 entries,
+	 * and multiple multicast addresses for filtering will be configured
+	 * in the default network configuration, so MAFT is only suitable for
+	 * unicast filtering. If the number of unicast addresses exceeds the
+	 * table capacity, the MAC hash filter will be used.
+	 */
+	if (type & ENETC_MAC_FILTER_TYPE_UC && enetc4_pf_set_uc_exact_filter(pf)) {
+		/* Fall back to the MAC hash filter */
+		enetc4_pf_set_mac_hash_filter(pf, ENETC_MAC_FILTER_TYPE_UC);
+		/* Clear the old MAC exact filter */
+		enetc4_pf_clear_maft_entries(pf);
+	}
+
+	if (type & ENETC_MAC_FILTER_TYPE_MC)
+		enetc4_pf_set_mac_hash_filter(pf, ENETC_MAC_FILTER_TYPE_MC);
+}
+
+static const struct enetc_pf_ops enetc4_pf_ops = {
+	.set_si_primary_mac = enetc4_pf_set_si_primary_mac,
+	.get_si_primary_mac = enetc4_pf_get_si_primary_mac,
+};
+
+static int enetc4_pf_struct_init(struct enetc_si *si)
+{
+	struct enetc_pf *pf = enetc_si_priv(si);
+
+	pf->si = si;
+	pf->total_vfs = pci_sriov_get_totalvfs(si->pdev);
+	pf->ops = &enetc4_pf_ops;
+
+	enetc4_get_port_caps(pf);
+
+	return 0;
+}
+
+static u32 enetc4_psicfgr0_val_construct(bool is_vf, u32 num_tx_bdr, u32 num_rx_bdr)
+{
+	u32 val;
+
+	val = ENETC_PSICFGR0_SET_TXBDR(num_tx_bdr);
+	val |= ENETC_PSICFGR0_SET_RXBDR(num_rx_bdr);
+	val |= ENETC_PSICFGR0_SIVC(ENETC_VLAN_TYPE_C | ENETC_VLAN_TYPE_S);
+
+	if (is_vf)
+		val |= ENETC_PSICFGR0_VTE | ENETC_PSICFGR0_SIVIE;
+
+	return val;
+}
+
+static void enetc4_default_rings_allocation(struct enetc_pf *pf)
+{
+	struct enetc_hw *hw = &pf->si->hw;
+	u32 num_rx_bdr, num_tx_bdr, val;
+	u32 vf_tx_bdr, vf_rx_bdr;
+	int i, rx_rem, tx_rem;
+
+	if (pf->caps.num_rx_bdr < ENETC_SI_MAX_RING_NUM + pf->caps.num_vsi)
+		num_rx_bdr = pf->caps.num_rx_bdr - pf->caps.num_vsi;
+	else
+		num_rx_bdr = ENETC_SI_MAX_RING_NUM;
+
+	if (pf->caps.num_tx_bdr < ENETC_SI_MAX_RING_NUM + pf->caps.num_vsi)
+		num_tx_bdr = pf->caps.num_tx_bdr - pf->caps.num_vsi;
+	else
+		num_tx_bdr = ENETC_SI_MAX_RING_NUM;
+
+	val = enetc4_psicfgr0_val_construct(false, num_tx_bdr, num_rx_bdr);
+	enetc_port_wr(hw, ENETC4_PSICFGR0(0), val);
+
+	num_rx_bdr = pf->caps.num_rx_bdr - num_rx_bdr;
+	rx_rem = num_rx_bdr % pf->caps.num_vsi;
+	num_rx_bdr = num_rx_bdr / pf->caps.num_vsi;
+
+	num_tx_bdr = pf->caps.num_tx_bdr - num_tx_bdr;
+	tx_rem = num_tx_bdr % pf->caps.num_vsi;
+	num_tx_bdr = num_tx_bdr / pf->caps.num_vsi;
+
+	for (i = 0; i < pf->caps.num_vsi; i++) {
+		vf_tx_bdr = (i < tx_rem) ? num_tx_bdr + 1 : num_tx_bdr;
+		vf_rx_bdr = (i < rx_rem) ? num_rx_bdr + 1 : num_rx_bdr;
+		val = enetc4_psicfgr0_val_construct(true, vf_tx_bdr, vf_rx_bdr);
+		enetc_port_wr(hw, ENETC4_PSICFGR0(i + 1), val);
+	}
+}
+
+static void enetc4_allocate_si_rings(struct enetc_pf *pf)
+{
+	enetc4_default_rings_allocation(pf);
+}
+
+static void enetc4_pf_set_si_vlan_promisc(struct enetc_hw *hw, int si, bool en)
+{
+	u32 val = enetc_port_rd(hw, ENETC4_PSIPVMR);
+
+	if (en)
+		val |= BIT(si);
+	else
+		val &= ~BIT(si);
+
+	enetc_port_wr(hw, ENETC4_PSIPVMR, val);
+}
+
+static void enetc4_set_default_si_vlan_promisc(struct enetc_pf *pf)
+{
+	struct enetc_hw *hw = &pf->si->hw;
+	int num_si = pf->caps.num_vsi + 1;
+	int i;
+
+	/* enforce VLAN promiscuous mode for all SIs */
+	for (i = 0; i < num_si; i++)
+		enetc4_pf_set_si_vlan_promisc(hw, i, true);
+}
+
+/* Allocate the number of MSI-X vectors for per SI. */
+static void enetc4_set_si_msix_num(struct enetc_pf *pf)
+{
+	struct enetc_hw *hw = &pf->si->hw;
+	int i, num_msix, total_si;
+	u32 val;
+
+	total_si = pf->caps.num_vsi + 1;
+
+	num_msix = pf->caps.num_msix / total_si +
+		   pf->caps.num_msix % total_si - 1;
+	val = num_msix & PSICFGR2_NUM_MSIX;
+	enetc_port_wr(hw, ENETC4_PSICFGR2(0), val);
+
+	num_msix = pf->caps.num_msix / total_si - 1;
+	val = num_msix & PSICFGR2_NUM_MSIX;
+	for (i = 0; i < pf->caps.num_vsi; i++)
+		enetc_port_wr(hw, ENETC4_PSICFGR2(i + 1), val);
+}
+
+static void enetc4_enable_all_si(struct enetc_pf *pf)
+{
+	struct enetc_hw *hw = &pf->si->hw;
+	int num_si = pf->caps.num_vsi + 1;
+	u32 si_bitmap = 0;
+	int i;
+
+	/* Master enable for all SIs */
+	for (i = 0; i < num_si; i++)
+		si_bitmap |= PMR_SI_EN(i);
+
+	enetc_port_wr(hw, ENETC4_PMR, si_bitmap);
+}
+
+static void enetc4_configure_port_si(struct enetc_pf *pf)
+{
+	struct enetc_hw *hw = &pf->si->hw;
+
+	enetc4_allocate_si_rings(pf);
+
+	/* Outer VLAN tag will be used for VLAN filtering */
+	enetc_port_wr(hw, ENETC4_PSIVLANFMR, PSIVLANFMR_VS);
+
+	enetc4_set_default_si_vlan_promisc(pf);
+
+	/* Disable SI MAC multicast & unicast promiscuous */
+	enetc_port_wr(hw, ENETC4_PSIPMMR, 0);
+
+	enetc4_set_si_msix_num(pf);
+
+	enetc4_enable_all_si(pf);
+}
+
+static void enetc4_pf_reset_tc_msdu(struct enetc_hw *hw)
+{
+	u32 val = ENETC_MAC_MAXFRM_SIZE;
+	int tc;
+
+	val = u32_replace_bits(val, SDU_TYPE_MPDU, PTCTMSDUR_SDU_TYPE);
+
+	for (tc = 0; tc < ENETC_NUM_TC; tc++)
+		enetc_port_wr(hw, ENETC4_PTCTMSDUR(tc), val);
+}
+
+static void enetc4_set_trx_frame_size(struct enetc_pf *pf)
+{
+	struct enetc_si *si = pf->si;
+
+	enetc_port_mac_wr(si, ENETC4_PM_MAXFRM(0),
+			  ENETC_SET_MAXFRM(ENETC_MAC_MAXFRM_SIZE));
+
+	enetc4_pf_reset_tc_msdu(&si->hw);
+}
+
+static void enetc4_enable_trx(struct enetc_pf *pf)
+{
+	struct enetc_hw *hw = &pf->si->hw;
+
+	/* Enable port transmit/receive */
+	enetc_port_wr(hw, ENETC4_POR, 0);
+}
+
+static void enetc4_configure_port(struct enetc_pf *pf)
+{
+	enetc4_configure_port_si(pf);
+	enetc4_set_trx_frame_size(pf);
+	enetc_set_default_rss_key(pf);
+	enetc4_enable_trx(pf);
+}
+
+static int enetc4_init_ntmp_user(struct enetc_si *si)
+{
+	struct ntmp_user *user = &si->ntmp_user;
+
+	/* For ENETC 4.1, all table versions are 0 */
+	memset(&user->tbl, 0, sizeof(user->tbl));
+
+	return enetc4_setup_cbdr(si);
+}
+
+static void enetc4_free_ntmp_user(struct enetc_si *si)
+{
+	enetc4_teardown_cbdr(si);
+}
+
+static int enetc4_pf_init(struct enetc_pf *pf)
+{
+	struct device *dev = &pf->si->pdev->dev;
+	int err;
+
+	/* Initialize the MAC address for PF and VFs */
+	err = enetc_setup_mac_addresses(dev->of_node, pf);
+	if (err) {
+		dev_err(dev, "Failed to set MAC addresses\n");
+		return err;
+	}
+
+	err = enetc4_init_ntmp_user(pf->si);
+	if (err) {
+		dev_err(dev, "Failed to init CBDR\n");
+		return err;
+	}
+
+	enetc4_configure_port(pf);
+
+	return 0;
+}
+
+static void enetc4_pf_free(struct enetc_pf *pf)
+{
+	enetc4_free_ntmp_user(pf->si);
+}
+
+static void enetc4_psi_do_set_rx_mode(struct work_struct *work)
+{
+	struct enetc_si *si = container_of(work, struct enetc_si, rx_mode_task);
+	struct enetc_pf *pf = enetc_si_priv(si);
+	struct net_device *ndev = si->ndev;
+	struct enetc_hw *hw = &si->hw;
+	bool uc_promisc = false;
+	bool mc_promisc = false;
+	int type = 0;
+
+	rtnl_lock();
+
+	if (ndev->flags & IFF_PROMISC) {
+		uc_promisc = true;
+		mc_promisc = true;
+	} else if (ndev->flags & IFF_ALLMULTI) {
+		mc_promisc = true;
+		type = ENETC_MAC_FILTER_TYPE_UC;
+	} else {
+		type = ENETC_MAC_FILTER_TYPE_ALL;
+	}
+
+	enetc4_pf_set_si_mac_promisc(hw, 0, uc_promisc, mc_promisc);
+
+	if (uc_promisc) {
+		enetc4_pf_set_si_uc_hash_filter(hw, 0, 0);
+		enetc4_pf_clear_maft_entries(pf);
+	}
+
+	if (mc_promisc)
+		enetc4_pf_set_si_mc_hash_filter(hw, 0, 0);
+
+	/* Set new MAC filter */
+	enetc4_pf_set_mac_filter(pf, type);
+
+	rtnl_unlock();
+}
+
+static void enetc4_pf_set_rx_mode(struct net_device *ndev)
+{
+	struct enetc_ndev_priv *priv = netdev_priv(ndev);
+	struct enetc_si *si = priv->si;
+
+	queue_work(si->workqueue, &si->rx_mode_task);
+}
+
+static int enetc4_pf_set_features(struct net_device *ndev,
+				  netdev_features_t features)
+{
+	netdev_features_t changed = ndev->features ^ features;
+	struct enetc_ndev_priv *priv = netdev_priv(ndev);
+	struct enetc_hw *hw = &priv->si->hw;
+
+	if (changed & NETIF_F_HW_VLAN_CTAG_FILTER) {
+		bool promisc_en = !(features & NETIF_F_HW_VLAN_CTAG_FILTER);
+
+		enetc4_pf_set_si_vlan_promisc(hw, 0, promisc_en);
+	}
+
+	if (changed & NETIF_F_LOOPBACK)
+		enetc4_pf_set_loopback(ndev, !!(features & NETIF_F_LOOPBACK));
+
+	enetc_set_features(ndev, features);
+
+	return 0;
+}
+
+static const struct net_device_ops enetc4_ndev_ops = {
+	.ndo_open		= enetc_open,
+	.ndo_stop		= enetc_close,
+	.ndo_start_xmit		= enetc_xmit,
+	.ndo_get_stats		= enetc_get_stats,
+	.ndo_set_mac_address	= enetc_pf_set_mac_addr,
+	.ndo_set_rx_mode	= enetc4_pf_set_rx_mode,
+	.ndo_set_features	= enetc4_pf_set_features,
+	.ndo_vlan_rx_add_vid	= enetc_vlan_rx_add_vid,
+	.ndo_vlan_rx_kill_vid	= enetc_vlan_rx_del_vid,
+	.ndo_eth_ioctl		= enetc_ioctl,
+	.ndo_hwtstamp_get	= enetc_hwtstamp_get,
+	.ndo_hwtstamp_set	= enetc_hwtstamp_set,
+};
+
+static struct phylink_pcs *
+enetc4_pl_mac_select_pcs(struct phylink_config *config, phy_interface_t iface)
+{
+	struct enetc_pf *pf = phylink_to_enetc_pf(config);
+
+	return pf->pcs;
+}
+
+static void enetc4_mac_config(struct enetc_pf *pf, unsigned int mode,
+			      phy_interface_t phy_mode)
+{
+	struct enetc_ndev_priv *priv = netdev_priv(pf->si->ndev);
+	struct enetc_si *si = pf->si;
+	u32 val;
+
+	val = enetc_port_mac_rd(si, ENETC4_PM_IF_MODE(0));
+	val &= ~(PM_IF_MODE_IFMODE | PM_IF_MODE_ENA);
+
+	switch (phy_mode) {
+	case PHY_INTERFACE_MODE_RGMII:
+	case PHY_INTERFACE_MODE_RGMII_ID:
+	case PHY_INTERFACE_MODE_RGMII_RXID:
+	case PHY_INTERFACE_MODE_RGMII_TXID:
+		val |= IFMODE_RGMII;
+		/* We need to enable auto-negotiation for the MAC
+		 * if its RGMII interface support In-Band status.
+		 */
+		if (phylink_autoneg_inband(mode))
+			val |= PM_IF_MODE_ENA;
+		break;
+	case PHY_INTERFACE_MODE_RMII:
+		val |= IFMODE_RMII;
+		break;
+	case PHY_INTERFACE_MODE_SGMII:
+	case PHY_INTERFACE_MODE_2500BASEX:
+		val |= IFMODE_SGMII;
+		break;
+	case PHY_INTERFACE_MODE_10GBASER:
+	case PHY_INTERFACE_MODE_XGMII:
+	case PHY_INTERFACE_MODE_USXGMII:
+		val |= IFMODE_XGMII;
+		break;
+	default:
+		dev_err(priv->dev,
+			"Unsupported PHY mode:%d\n", phy_mode);
+		return;
+	}
+
+	enetc_port_mac_wr(si, ENETC4_PM_IF_MODE(0), val);
+}
+
+static void enetc4_pl_mac_config(struct phylink_config *config, unsigned int mode,
+				 const struct phylink_link_state *state)
+{
+	struct enetc_pf *pf = phylink_to_enetc_pf(config);
+
+	enetc4_mac_config(pf, mode, state->interface);
+}
+
+static void enetc4_set_port_speed(struct enetc_ndev_priv *priv, int speed)
+{
+	u32 old_speed = priv->speed;
+	u32 val;
+
+	if (speed == old_speed)
+		return;
+
+	val = enetc_port_rd(&priv->si->hw, ENETC4_PCR);
+	val &= ~PCR_PSPEED;
+
+	switch (speed) {
+	case SPEED_100:
+	case SPEED_1000:
+	case SPEED_2500:
+	case SPEED_10000:
+		val |= (PCR_PSPEED & PCR_PSPEED_VAL(speed));
+		break;
+	case SPEED_10:
+	default:
+		val |= (PCR_PSPEED & PCR_PSPEED_VAL(SPEED_10));
+	}
+
+	priv->speed = speed;
+	enetc_port_wr(&priv->si->hw, ENETC4_PCR, val);
+}
+
+static void enetc4_set_rgmii_mac(struct enetc_pf *pf, int speed, int duplex)
+{
+	struct enetc_si *si = pf->si;
+	u32 old_val, val;
+
+	old_val = enetc_port_mac_rd(si, ENETC4_PM_IF_MODE(0));
+	val = old_val & ~(PM_IF_MODE_ENA | PM_IF_MODE_M10 | PM_IF_MODE_REVMII);
+
+	switch (speed) {
+	case SPEED_1000:
+		val = u32_replace_bits(val, SSP_1G, PM_IF_MODE_SSP);
+		break;
+	case SPEED_100:
+		val = u32_replace_bits(val, SSP_100M, PM_IF_MODE_SSP);
+		break;
+	case SPEED_10:
+		val = u32_replace_bits(val, SSP_10M, PM_IF_MODE_SSP);
+	}
+
+	val = u32_replace_bits(val, duplex == DUPLEX_FULL ? 0 : 1,
+			       PM_IF_MODE_HD);
+
+	if (val == old_val)
+		return;
+
+	enetc_port_mac_wr(si, ENETC4_PM_IF_MODE(0), val);
+}
+
+static void enetc4_set_rmii_mac(struct enetc_pf *pf, int speed, int duplex)
+{
+	struct enetc_si *si = pf->si;
+	u32 old_val, val;
+
+	old_val = enetc_port_mac_rd(si, ENETC4_PM_IF_MODE(0));
+	val = old_val & ~(PM_IF_MODE_ENA | PM_IF_MODE_SSP);
+
+	switch (speed) {
+	case SPEED_100:
+		val &= ~PM_IF_MODE_M10;
+		break;
+	case SPEED_10:
+		val |= PM_IF_MODE_M10;
+	}
+
+	val = u32_replace_bits(val, duplex == DUPLEX_FULL ? 0 : 1,
+			       PM_IF_MODE_HD);
+
+	if (val == old_val)
+		return;
+
+	enetc_port_mac_wr(si, ENETC4_PM_IF_MODE(0), val);
+}
+
+static void enetc4_set_hd_flow_control(struct enetc_pf *pf, bool enable)
+{
+	struct enetc_si *si = pf->si;
+	u32 old_val, val;
+
+	if (!pf->caps.half_duplex)
+		return;
+
+	old_val = enetc_port_mac_rd(si, ENETC4_PM_CMD_CFG(0));
+	val = u32_replace_bits(old_val, enable ? 1 : 0, PM_CMD_CFG_HD_FCEN);
+	if (val == old_val)
+		return;
+
+	enetc_port_mac_wr(si, ENETC4_PM_CMD_CFG(0), val);
+}
+
+static void enetc4_set_rx_pause(struct enetc_pf *pf, bool rx_pause)
+{
+	struct enetc_si *si = pf->si;
+	u32 old_val, val;
+
+	old_val = enetc_port_mac_rd(si, ENETC4_PM_CMD_CFG(0));
+	val = u32_replace_bits(old_val, rx_pause ? 0 : 1, PM_CMD_CFG_PAUSE_IGN);
+	if (val == old_val)
+		return;
+
+	enetc_port_mac_wr(si, ENETC4_PM_CMD_CFG(0), val);
+}
+
+static void enetc4_set_tx_pause(struct enetc_pf *pf, int num_rxbdr, bool tx_pause)
+{
+	u32 pause_off_thresh = 0, pause_on_thresh = 0;
+	u32 init_quanta = 0, refresh_quanta = 0;
+	struct enetc_hw *hw = &pf->si->hw;
+	u32 rbmr, old_rbmr;
+	int i;
+
+	for (i = 0; i < num_rxbdr; i++) {
+		old_rbmr = enetc_rxbdr_rd(hw, i, ENETC_RBMR);
+		rbmr = u32_replace_bits(old_rbmr, tx_pause ? 1 : 0, ENETC_RBMR_CM);
+		if (rbmr == old_rbmr)
+			continue;
+
+		enetc_rxbdr_wr(hw, i, ENETC_RBMR, rbmr);
+	}
+
+	if (tx_pause) {
+		/* When the port first enters congestion, send a PAUSE request
+		 * with the maximum number of quanta. When the port exits
+		 * congestion, it will automatically send a PAUSE frame with
+		 * zero quanta.
+		 */
+		init_quanta = 0xffff;
+
+		/* Also, set up the refresh timer to send follow-up PAUSE
+		 * frames at half the quanta value, in case the congestion
+		 * condition persists.
+		 */
+		refresh_quanta = 0xffff / 2;
+
+		/* Start emitting PAUSE frames when 3 large frames (or more
+		 * smaller frames) have accumulated in the FIFO waiting to be
+		 * DMAed to the RX ring.
+		 */
+		pause_on_thresh = 3 * ENETC_MAC_MAXFRM_SIZE;
+		pause_off_thresh = 1 * ENETC_MAC_MAXFRM_SIZE;
+	}
+
+	enetc_port_mac_wr(pf->si, ENETC4_PM_PAUSE_QUANTA(0), init_quanta);
+	enetc_port_mac_wr(pf->si, ENETC4_PM_PAUSE_THRESH(0), refresh_quanta);
+	enetc_port_wr(hw, ENETC4_PPAUONTR, pause_on_thresh);
+	enetc_port_wr(hw, ENETC4_PPAUOFFTR, pause_off_thresh);
+}
+
+static void enetc4_enable_mac(struct enetc_pf *pf, bool en)
+{
+	struct enetc_si *si = pf->si;
+	u32 val;
+
+	val = enetc_port_mac_rd(si, ENETC4_PM_CMD_CFG(0));
+	val &= ~(PM_CMD_CFG_TX_EN | PM_CMD_CFG_RX_EN);
+	val |= en ? (PM_CMD_CFG_TX_EN | PM_CMD_CFG_RX_EN) : 0;
+
+	enetc_port_mac_wr(si, ENETC4_PM_CMD_CFG(0), val);
+}
+
+static void enetc4_pl_mac_link_up(struct phylink_config *config,
+				  struct phy_device *phy, unsigned int mode,
+				  phy_interface_t interface, int speed,
+				  int duplex, bool tx_pause, bool rx_pause)
+{
+	struct enetc_pf *pf = phylink_to_enetc_pf(config);
+	struct enetc_si *si = pf->si;
+	struct enetc_ndev_priv *priv;
+	bool hd_fc = false;
+
+	priv = netdev_priv(si->ndev);
+	enetc4_set_port_speed(priv, speed);
+
+	if (!phylink_autoneg_inband(mode) &&
+	    phy_interface_mode_is_rgmii(interface))
+		enetc4_set_rgmii_mac(pf, speed, duplex);
+
+	if (interface == PHY_INTERFACE_MODE_RMII)
+		enetc4_set_rmii_mac(pf, speed, duplex);
+
+	if (duplex == DUPLEX_FULL) {
+		/* When preemption is enabled, generation of PAUSE frames
+		 * must be disabled, as stated in the IEEE 802.3 standard.
+		 */
+		if (priv->active_offloads & ENETC_F_QBU)
+			tx_pause = false;
+	} else { /* DUPLEX_HALF */
+		if (tx_pause || rx_pause)
+			hd_fc = true;
+
+		/* As per 802.3 annex 31B, PAUSE frames are only supported
+		 * when the link is configured for full duplex operation.
+		 */
+		tx_pause = false;
+		rx_pause = false;
+	}
+
+	enetc4_set_hd_flow_control(pf, hd_fc);
+	enetc4_set_tx_pause(pf, priv->num_rx_rings, tx_pause);
+	enetc4_set_rx_pause(pf, rx_pause);
+	enetc4_enable_mac(pf, true);
+}
+
+static void enetc4_pl_mac_link_down(struct phylink_config *config,
+				    unsigned int mode,
+				    phy_interface_t interface)
+{
+	struct enetc_pf *pf = phylink_to_enetc_pf(config);
+
+	enetc4_enable_mac(pf, false);
+}
+
+static const struct phylink_mac_ops enetc_pl_mac_ops = {
+	.mac_select_pcs = enetc4_pl_mac_select_pcs,
+	.mac_config = enetc4_pl_mac_config,
+	.mac_link_up = enetc4_pl_mac_link_up,
+	.mac_link_down = enetc4_pl_mac_link_down,
+};
+
+static void enetc4_pci_remove(void *data)
+{
+	struct pci_dev *pdev = data;
+
+	enetc_pci_remove(pdev);
+}
+
+static int enetc4_link_init(struct enetc_ndev_priv *priv,
+			    struct device_node *node)
+{
+	struct enetc_pf *pf = enetc_si_priv(priv->si);
+	struct device *dev = priv->dev;
+	int err;
+
+	err = of_get_phy_mode(node, &pf->if_mode);
+	if (err) {
+		dev_err(dev, "Failed to get PHY mode\n");
+		return err;
+	}
+
+	err = enetc_mdiobus_create(pf, node);
+	if (err) {
+		dev_err(dev, "Failed to create MDIO bus\n");
+		return err;
+	}
+
+	err = enetc_phylink_create(priv, node, &enetc_pl_mac_ops);
+	if (err) {
+		dev_err(dev, "Failed to create phylink\n");
+		goto err_phylink_create;
+	}
+
+	return 0;
+
+err_phylink_create:
+	enetc_mdiobus_destroy(pf);
+
+	return err;
+}
+
+static void enetc4_link_deinit(struct enetc_ndev_priv *priv)
+{
+	struct enetc_pf *pf = enetc_si_priv(priv->si);
+
+	enetc_phylink_destroy(priv);
+	enetc_mdiobus_destroy(pf);
+}
+
+static int enetc4_psi_wq_task_init(struct enetc_si *si)
+{
+	char wq_name[24];
+
+	INIT_WORK(&si->rx_mode_task, enetc4_psi_do_set_rx_mode);
+	snprintf(wq_name, sizeof(wq_name), "enetc-%s", pci_name(si->pdev));
+	si->workqueue = create_singlethread_workqueue(wq_name);
+	if (!si->workqueue)
+		return -ENOMEM;
+
+	return 0;
+}
+
+static int enetc4_pf_netdev_create(struct enetc_si *si)
+{
+	struct device *dev = &si->pdev->dev;
+	struct enetc_ndev_priv *priv;
+	struct net_device *ndev;
+	int err;
+
+	ndev = alloc_etherdev_mqs(sizeof(struct enetc_ndev_priv),
+				  si->num_tx_rings, si->num_rx_rings);
+	if (!ndev)
+		return  -ENOMEM;
+
+	priv = netdev_priv(ndev);
+	priv->ref_clk = devm_clk_get_optional(dev, "ref");
+	if (IS_ERR(priv->ref_clk)) {
+		dev_err(dev, "Get reference clock failed\n");
+		err = PTR_ERR(priv->ref_clk);
+		goto err_clk_get;
+	}
+
+	enetc_pf_netdev_setup(si, ndev, &enetc4_ndev_ops);
+
+	enetc_init_si_rings_params(priv);
+
+	err = enetc_configure_si(priv);
+	if (err) {
+		dev_err(dev, "Failed to configure SI\n");
+		goto err_config_si;
+	}
+
+	err = enetc_alloc_msix(priv);
+	if (err) {
+		dev_err(dev, "Failed to alloc MSI-X\n");
+		goto err_alloc_msix;
+	}
+
+	err = enetc4_link_init(priv, dev->of_node);
+	if (err)
+		goto err_link_init;
+
+	err = enetc4_psi_wq_task_init(si);
+	if (err) {
+		dev_err(dev, "Failed to init workqueue\n");
+		goto err_wq_init;
+	}
+
+	err = register_netdev(ndev);
+	if (err) {
+		dev_err(dev, "Failed to register netdev\n");
+		goto err_reg_netdev;
+	}
+
+	return 0;
+
+err_reg_netdev:
+	destroy_workqueue(si->workqueue);
+err_wq_init:
+	enetc4_link_deinit(priv);
+err_link_init:
+	enetc_free_msix(priv);
+err_alloc_msix:
+err_config_si:
+err_clk_get:
+	free_netdev(ndev);
+
+	return err;
+}
+
+static void enetc4_pf_netdev_destroy(struct enetc_si *si)
+{
+	struct enetc_ndev_priv *priv = netdev_priv(si->ndev);
+	struct net_device *ndev = si->ndev;
+
+	unregister_netdev(ndev);
+	cancel_work(&si->rx_mode_task);
+	destroy_workqueue(si->workqueue);
+	enetc4_link_deinit(priv);
+	enetc_free_msix(priv);
+	free_netdev(ndev);
+}
+
+static const struct enetc_si_ops enetc4_psi_ops = {
+	.get_rss_table = enetc4_get_rss_table,
+	.set_rss_table = enetc4_set_rss_table,
+};
+
+static int enetc4_pf_probe(struct pci_dev *pdev,
+			   const struct pci_device_id *ent)
+{
+	struct device *dev = &pdev->dev;
+	struct enetc_si *si;
+	struct enetc_pf *pf;
+	int err;
+
+	err = enetc_pci_probe(pdev, KBUILD_MODNAME, sizeof(*pf));
+	if (err)
+		return dev_err_probe(dev, err, "PCIe probing failed\n");
+
+	err = devm_add_action_or_reset(dev, enetc4_pci_remove, pdev);
+	if (err)
+		return err;
+
+	/* si is the private data. */
+	si = pci_get_drvdata(pdev);
+	if (!si->hw.port || !si->hw.global)
+		return dev_err_probe(dev, -ENODEV,
+				     "Couldn't map PF only space\n");
+
+	si->revision = enetc_get_ip_revision(&si->hw);
+	si->ops = &enetc4_psi_ops;
+	err = enetc_get_driver_data(si);
+	if (err)
+		return dev_err_probe(dev, err,
+				     "Could not get PF driver data\n");
+
+	err = enetc4_pf_struct_init(si);
+	if (err)
+		return err;
+
+	pf = enetc_si_priv(si);
+	err = enetc4_pf_init(pf);
+	if (err)
+		return err;
+
+	enetc_get_si_caps(si);
+
+	err = enetc4_pf_netdev_create(si);
+	if (err)
+		goto err_netdev_create;
+
+	enetc_create_debugfs(si);
+
+	return 0;
+
+err_netdev_create:
+	enetc4_pf_free(pf);
+
+	return err;
+}
+
+static void enetc4_pf_remove(struct pci_dev *pdev)
+{
+	struct enetc_si *si = pci_get_drvdata(pdev);
+	struct enetc_pf *pf = enetc_si_priv(si);
+
+	enetc_remove_debugfs(si);
+	enetc4_pf_netdev_destroy(si);
+	enetc4_pf_free(pf);
+}
+
+static const struct pci_device_id enetc4_pf_id_table[] = {
+	{ PCI_DEVICE(NXP_ENETC_VENDOR_ID, NXP_ENETC_PF_DEV_ID) },
+	{ 0, } /* End of table. */
+};
+MODULE_DEVICE_TABLE(pci, enetc4_pf_id_table);
+
+static struct pci_driver enetc4_pf_driver = {
+	.name = KBUILD_MODNAME,
+	.id_table = enetc4_pf_id_table,
+	.probe = enetc4_pf_probe,
+	.remove = enetc4_pf_remove,
+};
+module_pci_driver(enetc4_pf_driver);
+
+MODULE_DESCRIPTION("ENETC4 PF Driver");
+MODULE_LICENSE("Dual BSD/GPL");
diff --git a/drivers/net/ethernet/freescale/enetc/enetc_cbdr.c b/drivers/net/ethernet/freescale/enetc/enetc_cbdr.c
new file mode 100644
index 000000000000..3d5f31879d5c
--- /dev/null
+++ b/drivers/net/ethernet/freescale/enetc/enetc_cbdr.c
@@ -0,0 +1,308 @@
+// SPDX-License-Identifier: (GPL-2.0+ OR BSD-3-Clause)
+/* Copyright 2017-2019 NXP */
+
+#include "enetc.h"
+
+int enetc_setup_cbdr(struct device *dev, struct enetc_hw *hw, int bd_count,
+		     struct enetc_cbdr *cbdr)
+{
+	int size = bd_count * sizeof(struct enetc_cbd);
+
+	cbdr->bd_base = dma_alloc_coherent(dev, size, &cbdr->bd_dma_base,
+					   GFP_KERNEL);
+	if (!cbdr->bd_base)
+		return -ENOMEM;
+
+	/* h/w requires 128B alignment */
+	if (!IS_ALIGNED(cbdr->bd_dma_base, 128)) {
+		dma_free_coherent(dev, size, cbdr->bd_base,
+				  cbdr->bd_dma_base);
+		return -EINVAL;
+	}
+
+	cbdr->next_to_clean = 0;
+	cbdr->next_to_use = 0;
+	cbdr->dma_dev = dev;
+	cbdr->bd_count = bd_count;
+
+	cbdr->pir = hw->reg + ENETC_SICBDRPIR;
+	cbdr->cir = hw->reg + ENETC_SICBDRCIR;
+	cbdr->mr = hw->reg + ENETC_SICBDRMR;
+
+	/* set CBDR cache attributes */
+	enetc_wr(hw, ENETC_SICAR2,
+		 ENETC_SICAR_RD_COHERENT | ENETC_SICAR_WR_COHERENT);
+
+	enetc_wr(hw, ENETC_SICBDRBAR0, lower_32_bits(cbdr->bd_dma_base));
+	enetc_wr(hw, ENETC_SICBDRBAR1, upper_32_bits(cbdr->bd_dma_base));
+	enetc_wr(hw, ENETC_SICBDRLENR, ENETC_RTBLENR_LEN(cbdr->bd_count));
+
+	enetc_wr_reg(cbdr->pir, cbdr->next_to_clean);
+	enetc_wr_reg(cbdr->cir, cbdr->next_to_use);
+	/* enable ring */
+	enetc_wr_reg(cbdr->mr, BIT(31));
+
+	return 0;
+}
+EXPORT_SYMBOL_GPL(enetc_setup_cbdr);
+
+void enetc_teardown_cbdr(struct enetc_cbdr *cbdr)
+{
+	int size = cbdr->bd_count * sizeof(struct enetc_cbd);
+
+	/* disable ring */
+	enetc_wr_reg(cbdr->mr, 0);
+
+	dma_free_coherent(cbdr->dma_dev, size, cbdr->bd_base,
+			  cbdr->bd_dma_base);
+	cbdr->bd_base = NULL;
+	cbdr->dma_dev = NULL;
+}
+EXPORT_SYMBOL_GPL(enetc_teardown_cbdr);
+
+int enetc4_setup_cbdr(struct enetc_si *si)
+{
+	struct ntmp_user *user = &si->ntmp_user;
+	struct device *dev = &si->pdev->dev;
+	struct enetc_hw *hw = &si->hw;
+	struct netc_cbdr_regs regs;
+
+	user->cbdr_num = 1;
+	user->dev = dev;
+	user->ring = devm_kcalloc(dev, user->cbdr_num,
+				  sizeof(struct netc_cbdr), GFP_KERNEL);
+	if (!user->ring)
+		return -ENOMEM;
+
+	/* set CBDR cache attributes */
+	enetc_wr(hw, ENETC_SICAR2,
+		 ENETC_SICAR_RD_COHERENT | ENETC_SICAR_WR_COHERENT);
+
+	regs.pir = hw->reg + ENETC_SICBDRPIR;
+	regs.cir = hw->reg + ENETC_SICBDRCIR;
+	regs.mr = hw->reg + ENETC_SICBDRMR;
+	regs.bar0 = hw->reg + ENETC_SICBDRBAR0;
+	regs.bar1 = hw->reg + ENETC_SICBDRBAR1;
+	regs.lenr = hw->reg + ENETC_SICBDRLENR;
+
+	return ntmp_init_cbdr(user->ring, dev, &regs);
+}
+EXPORT_SYMBOL_GPL(enetc4_setup_cbdr);
+
+void enetc4_teardown_cbdr(struct enetc_si *si)
+{
+	struct ntmp_user *user = &si->ntmp_user;
+
+	ntmp_free_cbdr(user->ring);
+	user->dev = NULL;
+}
+EXPORT_SYMBOL_GPL(enetc4_teardown_cbdr);
+
+static void enetc_clean_cbdr(struct enetc_cbdr *ring)
+{
+	struct enetc_cbd *dest_cbd;
+	int i, status;
+
+	i = ring->next_to_clean;
+
+	while (enetc_rd_reg(ring->cir) != i) {
+		dest_cbd = ENETC_CBD(*ring, i);
+		status = dest_cbd->status_flags & ENETC_CBD_STATUS_MASK;
+		if (status)
+			dev_warn(ring->dma_dev, "CMD err %04x for cmd %04x\n",
+				 status, dest_cbd->cmd);
+
+		memset(dest_cbd, 0, sizeof(*dest_cbd));
+
+		i = (i + 1) % ring->bd_count;
+	}
+
+	ring->next_to_clean = i;
+}
+
+static int enetc_cbd_unused(struct enetc_cbdr *r)
+{
+	return (r->next_to_clean - r->next_to_use - 1 + r->bd_count) %
+		r->bd_count;
+}
+
+int enetc_send_cmd(struct enetc_si *si, struct enetc_cbd *cbd)
+{
+	struct enetc_cbdr *ring = &si->cbd_ring;
+	int timeout = ENETC_CBDR_TIMEOUT;
+	struct enetc_cbd *dest_cbd;
+	int i;
+
+	if (unlikely(!ring->bd_base))
+		return -EIO;
+
+	if (unlikely(!enetc_cbd_unused(ring)))
+		enetc_clean_cbdr(ring);
+
+	i = ring->next_to_use;
+	dest_cbd = ENETC_CBD(*ring, i);
+
+	/* copy command to the ring */
+	*dest_cbd = *cbd;
+	i = (i + 1) % ring->bd_count;
+
+	ring->next_to_use = i;
+	/* let H/W know BD ring has been updated */
+	enetc_wr_reg(ring->pir, i);
+
+	do {
+		if (enetc_rd_reg(ring->cir) == i)
+			break;
+		udelay(10); /* cannot sleep, rtnl_lock() */
+		timeout -= 10;
+	} while (timeout);
+
+	if (!timeout)
+		return -EBUSY;
+
+	/* CBD may writeback data, feedback up level */
+	*cbd = *dest_cbd;
+
+	enetc_clean_cbdr(ring);
+
+	return 0;
+}
+EXPORT_SYMBOL_GPL(enetc_send_cmd);
+
+int enetc_clear_mac_flt_entry(struct enetc_si *si, int index)
+{
+	struct enetc_cbd cbd;
+
+	memset(&cbd, 0, sizeof(cbd));
+
+	cbd.cls = 1;
+	cbd.status_flags = ENETC_CBD_FLAGS_SF;
+	cbd.index = cpu_to_le16(index);
+
+	return enetc_send_cmd(si, &cbd);
+}
+EXPORT_SYMBOL_GPL(enetc_clear_mac_flt_entry);
+
+int enetc_set_mac_flt_entry(struct enetc_si *si, int index,
+			    char *mac_addr, int si_map)
+{
+	struct enetc_cbd cbd;
+	u32 upper;
+	u16 lower;
+
+	memset(&cbd, 0, sizeof(cbd));
+
+	/* fill up the "set" descriptor */
+	cbd.cls = 1;
+	cbd.status_flags = ENETC_CBD_FLAGS_SF;
+	cbd.index = cpu_to_le16(index);
+	cbd.opt[3] = cpu_to_le32(si_map);
+	/* enable entry */
+	cbd.opt[0] = cpu_to_le32(BIT(31));
+
+	upper = *(const u32 *)mac_addr;
+	lower = *(const u16 *)(mac_addr + 4);
+	cbd.addr[0] = cpu_to_le32(upper);
+	cbd.addr[1] = cpu_to_le32(lower);
+
+	return enetc_send_cmd(si, &cbd);
+}
+EXPORT_SYMBOL_GPL(enetc_set_mac_flt_entry);
+
+/* Set entry in RFS table */
+int enetc_set_fs_entry(struct enetc_si *si, struct enetc_cmd_rfse *rfse,
+		       int index)
+{
+	struct enetc_cbdr *ring = &si->cbd_ring;
+	struct enetc_cbd cbd = {.cmd = 0};
+	void *tmp, *tmp_align;
+	dma_addr_t dma;
+	int err;
+
+	/* fill up the "set" descriptor */
+	cbd.cmd = 0;
+	cbd.cls = 4;
+	cbd.index = cpu_to_le16(index);
+	cbd.opt[3] = cpu_to_le32(0); /* SI */
+
+	tmp = enetc_cbd_alloc_data_mem(si, &cbd, sizeof(*rfse),
+				       &dma, &tmp_align);
+	if (!tmp)
+		return -ENOMEM;
+
+	memcpy(tmp_align, rfse, sizeof(*rfse));
+
+	err = enetc_send_cmd(si, &cbd);
+	if (err)
+		dev_err(ring->dma_dev, "FS entry add failed (%d)!", err);
+
+	enetc_cbd_free_data_mem(si, sizeof(*rfse), tmp, &dma);
+
+	return err;
+}
+EXPORT_SYMBOL_GPL(enetc_set_fs_entry);
+
+static int enetc_cmd_rss_table(struct enetc_si *si, u32 *table, int count,
+			       bool read)
+{
+	struct enetc_cbdr *ring = &si->cbd_ring;
+	struct enetc_cbd cbd = {.cmd = 0};
+	u8 *tmp, *tmp_align;
+	dma_addr_t dma;
+	int err, i;
+
+	if (count < ENETC_CBD_DATA_MEM_ALIGN)
+		/* HW only takes in a full 64 entry table */
+		return -EINVAL;
+
+	tmp = enetc_cbd_alloc_data_mem(si, &cbd, count,
+				       &dma, (void *)&tmp_align);
+	if (!tmp)
+		return -ENOMEM;
+
+	if (!read)
+		for (i = 0; i < count; i++)
+			tmp_align[i] = (u8)(table[i]);
+
+	/* fill up the descriptor */
+	cbd.cmd = read ? 2 : 1;
+	cbd.cls = 3;
+
+	err = enetc_send_cmd(si, &cbd);
+	if (err)
+		dev_err(ring->dma_dev, "RSS cmd failed (%d)!", err);
+
+	if (read)
+		for (i = 0; i < count; i++)
+			table[i] = tmp_align[i];
+
+	enetc_cbd_free_data_mem(si, count, tmp, &dma);
+
+	return err;
+}
+
+/* Get RSS table */
+int enetc_get_rss_table(struct enetc_si *si, u32 *table, int count)
+{
+	return enetc_cmd_rss_table(si, table, count, true);
+}
+EXPORT_SYMBOL_GPL(enetc_get_rss_table);
+
+/* Set RSS table */
+int enetc_set_rss_table(struct enetc_si *si, const u32 *table, int count)
+{
+	return enetc_cmd_rss_table(si, (u32 *)table, count, false);
+}
+EXPORT_SYMBOL_GPL(enetc_set_rss_table);
+
+int enetc4_get_rss_table(struct enetc_si *si, u32 *table, int count)
+{
+	return ntmp_rsst_query_entry(&si->ntmp_user, table, count);
+}
+EXPORT_SYMBOL_GPL(enetc4_get_rss_table);
+
+int enetc4_set_rss_table(struct enetc_si *si, const u32 *table, int count)
+{
+	return ntmp_rsst_update_entry(&si->ntmp_user, table, count);
+}
+EXPORT_SYMBOL_GPL(enetc4_set_rss_table);
diff --git a/drivers/net/ethernet/freescale/enetc/enetc_ethtool.c b/drivers/net/ethernet/freescale/enetc/enetc_ethtool.c
new file mode 100644
index 000000000000..71d052de669a
--- /dev/null
+++ b/drivers/net/ethernet/freescale/enetc/enetc_ethtool.c
@@ -0,0 +1,1367 @@
+// SPDX-License-Identifier: (GPL-2.0+ OR BSD-3-Clause)
+/* Copyright 2017-2019 NXP */
+
+#include <linux/ethtool_netlink.h>
+#include <linux/net_tstamp.h>
+#include <linux/module.h>
+#include <linux/of.h>
+#include <linux/ptp_clock_kernel.h>
+
+#include "enetc.h"
+
+static const u32 enetc_si_regs[] = {
+	ENETC_SIMR, ENETC_SIPMAR0, ENETC_SIPMAR1, ENETC_SICBDRMR,
+	ENETC_SICBDRSR,	ENETC_SICBDRBAR0, ENETC_SICBDRBAR1, ENETC_SICBDRPIR,
+	ENETC_SICBDRCIR, ENETC_SICBDRLENR, ENETC_SICAPR0, ENETC_SICAPR1,
+	ENETC_SIUEFDCR
+};
+
+static const u32 enetc_txbdr_regs[] = {
+	ENETC_TBMR, ENETC_TBSR, ENETC_TBBAR0, ENETC_TBBAR1,
+	ENETC_TBPIR, ENETC_TBCIR, ENETC_TBLENR, ENETC_TBIER, ENETC_TBICR0,
+	ENETC_TBICR1
+};
+
+static const u32 enetc_rxbdr_regs[] = {
+	ENETC_RBMR, ENETC_RBSR, ENETC_RBBSR, ENETC_RBCIR, ENETC_RBBAR0,
+	ENETC_RBBAR1, ENETC_RBPIR, ENETC_RBLENR, ENETC_RBIER, ENETC_RBICR0,
+	ENETC_RBICR1
+};
+
+static const u32 enetc_port_regs[] = {
+	ENETC_PMR, ENETC_PSR, ENETC_PSIPMR, ENETC_PSIPMAR0(0),
+	ENETC_PSIPMAR1(0), ENETC_PTXMBAR, ENETC_PCAPR0, ENETC_PCAPR1,
+	ENETC_PSICFGR0(0), ENETC_PRFSCAPR, ENETC_PTCMSDUR(0),
+	ENETC_PM0_CMD_CFG, ENETC_PM0_MAXFRM, ENETC_PM0_IF_MODE
+};
+
+static const u32 enetc_port_mm_regs[] = {
+	ENETC_MMCSR, ENETC_PFPMR, ENETC_PTCFPR(0), ENETC_PTCFPR(1),
+	ENETC_PTCFPR(2), ENETC_PTCFPR(3), ENETC_PTCFPR(4), ENETC_PTCFPR(5),
+	ENETC_PTCFPR(6), ENETC_PTCFPR(7),
+};
+
+static int enetc_get_reglen(struct net_device *ndev)
+{
+	struct enetc_ndev_priv *priv = netdev_priv(ndev);
+	struct enetc_hw *hw = &priv->si->hw;
+	int len;
+
+	len = ARRAY_SIZE(enetc_si_regs);
+	len += ARRAY_SIZE(enetc_txbdr_regs) * priv->num_tx_rings;
+	len += ARRAY_SIZE(enetc_rxbdr_regs) * priv->num_rx_rings;
+
+	if (hw->port)
+		len += ARRAY_SIZE(enetc_port_regs);
+
+	if (hw->port && !!(priv->si->hw_features & ENETC_SI_F_QBU))
+		len += ARRAY_SIZE(enetc_port_mm_regs);
+
+	len *= sizeof(u32) * 2; /* store 2 entries per reg: addr and value */
+
+	return len;
+}
+
+static void enetc_get_regs(struct net_device *ndev, struct ethtool_regs *regs,
+			   void *regbuf)
+{
+	struct enetc_ndev_priv *priv = netdev_priv(ndev);
+	struct enetc_hw *hw = &priv->si->hw;
+	u32 *buf = (u32 *)regbuf;
+	int i, j;
+	u32 addr;
+
+	for (i = 0; i < ARRAY_SIZE(enetc_si_regs); i++) {
+		*buf++ = enetc_si_regs[i];
+		*buf++ = enetc_rd(hw, enetc_si_regs[i]);
+	}
+
+	for (i = 0; i < priv->num_tx_rings; i++) {
+		for (j = 0; j < ARRAY_SIZE(enetc_txbdr_regs); j++) {
+			addr = ENETC_BDR(TX, i, enetc_txbdr_regs[j]);
+
+			*buf++ = addr;
+			*buf++ = enetc_rd(hw, addr);
+		}
+	}
+
+	for (i = 0; i < priv->num_rx_rings; i++) {
+		for (j = 0; j < ARRAY_SIZE(enetc_rxbdr_regs); j++) {
+			addr = ENETC_BDR(RX, i, enetc_rxbdr_regs[j]);
+
+			*buf++ = addr;
+			*buf++ = enetc_rd(hw, addr);
+		}
+	}
+
+	if (!hw->port)
+		return;
+
+	for (i = 0; i < ARRAY_SIZE(enetc_port_regs); i++) {
+		addr = ENETC_PORT_BASE + enetc_port_regs[i];
+		*buf++ = addr;
+		*buf++ = enetc_rd(hw, addr);
+	}
+
+	if (priv->si->hw_features & ENETC_SI_F_QBU) {
+		for (i = 0; i < ARRAY_SIZE(enetc_port_mm_regs); i++) {
+			addr = ENETC_PORT_BASE + enetc_port_mm_regs[i];
+			*buf++ = addr;
+			*buf++ = enetc_rd(hw, addr);
+		}
+	}
+}
+
+static const struct {
+	int reg;
+	char name[ETH_GSTRING_LEN];
+} enetc_si_counters[] =  {
+	{ ENETC_SIROCT, "SI rx octets" },
+	{ ENETC_SIRFRM, "SI rx frames" },
+	{ ENETC_SIRUCA, "SI rx u-cast frames" },
+	{ ENETC_SIRMCA, "SI rx m-cast frames" },
+	{ ENETC_SITOCT, "SI tx octets" },
+	{ ENETC_SITFRM, "SI tx frames" },
+	{ ENETC_SITUCA, "SI tx u-cast frames" },
+	{ ENETC_SITMCA, "SI tx m-cast frames" },
+	{ ENETC_RBDCR(0), "Rx ring  0 discarded frames" },
+	{ ENETC_RBDCR(1), "Rx ring  1 discarded frames" },
+	{ ENETC_RBDCR(2), "Rx ring  2 discarded frames" },
+	{ ENETC_RBDCR(3), "Rx ring  3 discarded frames" },
+	{ ENETC_RBDCR(4), "Rx ring  4 discarded frames" },
+	{ ENETC_RBDCR(5), "Rx ring  5 discarded frames" },
+	{ ENETC_RBDCR(6), "Rx ring  6 discarded frames" },
+	{ ENETC_RBDCR(7), "Rx ring  7 discarded frames" },
+	{ ENETC_RBDCR(8), "Rx ring  8 discarded frames" },
+	{ ENETC_RBDCR(9), "Rx ring  9 discarded frames" },
+	{ ENETC_RBDCR(10), "Rx ring 10 discarded frames" },
+	{ ENETC_RBDCR(11), "Rx ring 11 discarded frames" },
+	{ ENETC_RBDCR(12), "Rx ring 12 discarded frames" },
+	{ ENETC_RBDCR(13), "Rx ring 13 discarded frames" },
+	{ ENETC_RBDCR(14), "Rx ring 14 discarded frames" },
+	{ ENETC_RBDCR(15), "Rx ring 15 discarded frames" },
+};
+
+static const struct {
+	int reg;
+	char name[ETH_GSTRING_LEN] __nonstring;
+} enetc_pm_counters[] = {
+	{ ENETC_PM_REOCT(0),	"MAC rx ethernet octets" },
+	{ ENETC_PM_RALN(0),	"MAC rx alignment errors" },
+	{ ENETC_PM_RXPF(0),	"MAC rx valid pause frames" },
+	{ ENETC_PM_RFRM(0),	"MAC rx valid frames" },
+	{ ENETC_PM_RFCS(0),	"MAC rx fcs errors" },
+	{ ENETC_PM_RVLAN(0),	"MAC rx VLAN frames" },
+	{ ENETC_PM_RERR(0),	"MAC rx frame errors" },
+	{ ENETC_PM_RUCA(0),	"MAC rx unicast frames" },
+	{ ENETC_PM_RMCA(0),	"MAC rx multicast frames" },
+	{ ENETC_PM_RBCA(0),	"MAC rx broadcast frames" },
+	{ ENETC_PM_RDRP(0),	"MAC rx dropped packets" },
+	{ ENETC_PM_RPKT(0),	"MAC rx packets" },
+	{ ENETC_PM_RUND(0),	"MAC rx undersized packets" },
+	{ ENETC_PM_R64(0),	"MAC rx 64 byte packets" },
+	{ ENETC_PM_R127(0),	"MAC rx 65-127 byte packets" },
+	{ ENETC_PM_R255(0),	"MAC rx 128-255 byte packets" },
+	{ ENETC_PM_R511(0),	"MAC rx 256-511 byte packets" },
+	{ ENETC_PM_R1023(0),	"MAC rx 512-1023 byte packets" },
+	{ ENETC_PM_R1522(0),	"MAC rx 1024-1522 byte packets" },
+	{ ENETC_PM_R1523X(0),	"MAC rx 1523 to max-octet packets" },
+	{ ENETC_PM_ROVR(0),	"MAC rx oversized packets" },
+	{ ENETC_PM_RJBR(0),	"MAC rx jabber packets" },
+	{ ENETC_PM_RFRG(0),	"MAC rx fragment packets" },
+	{ ENETC_PM_RCNP(0),	"MAC rx control packets" },
+	{ ENETC_PM_RDRNTP(0),	"MAC rx fifo drop" },
+	{ ENETC_PM_TEOCT(0),	"MAC tx ethernet octets" },
+	{ ENETC_PM_TOCT(0),	"MAC tx octets" },
+	{ ENETC_PM_TCRSE(0),	"MAC tx carrier sense errors" },
+	{ ENETC_PM_TXPF(0),	"MAC tx valid pause frames" },
+	{ ENETC_PM_TFRM(0),	"MAC tx frames" },
+	{ ENETC_PM_TFCS(0),	"MAC tx fcs errors" },
+	{ ENETC_PM_TVLAN(0),	"MAC tx VLAN frames" },
+	{ ENETC_PM_TERR(0),	"MAC tx frame errors" },
+	{ ENETC_PM_TUCA(0),	"MAC tx unicast frames" },
+	{ ENETC_PM_TMCA(0),	"MAC tx multicast frames" },
+	{ ENETC_PM_TBCA(0),	"MAC tx broadcast frames" },
+	{ ENETC_PM_TPKT(0),	"MAC tx packets" },
+	{ ENETC_PM_TUND(0),	"MAC tx undersized packets" },
+	{ ENETC_PM_T64(0),	"MAC tx 64 byte packets" },
+	{ ENETC_PM_T127(0),	"MAC tx 65-127 byte packets" },
+	{ ENETC_PM_T255(0),	"MAC tx 128-255 byte packets" },
+	{ ENETC_PM_T511(0),	"MAC tx 256-511 byte packets" },
+	{ ENETC_PM_T1023(0),	"MAC tx 512-1023 byte packets" },
+	{ ENETC_PM_T1522(0),	"MAC tx 1024-1522 byte packets" },
+	{ ENETC_PM_T1523X(0),	"MAC tx 1523 to max-octet packets" },
+	{ ENETC_PM_TCNP(0),	"MAC tx control packets" },
+	{ ENETC_PM_TDFR(0),	"MAC tx deferred packets" },
+	{ ENETC_PM_TMCOL(0),	"MAC tx multiple collisions" },
+	{ ENETC_PM_TSCOL(0),	"MAC tx single collisions" },
+	{ ENETC_PM_TLCOL(0),	"MAC tx late collisions" },
+	{ ENETC_PM_TECOL(0),	"MAC tx excessive collisions" },
+};
+
+static const struct {
+	int reg;
+	char name[ETH_GSTRING_LEN] __nonstring;
+} enetc_port_counters[] = {
+	{ ENETC_UFDMF,		"SI MAC nomatch u-cast discards" },
+	{ ENETC_MFDMF,		"SI MAC nomatch m-cast discards" },
+	{ ENETC_PBFDSIR,	"SI MAC nomatch b-cast discards" },
+	{ ENETC_PUFDVFR,	"SI VLAN nomatch u-cast discards" },
+	{ ENETC_PMFDVFR,	"SI VLAN nomatch m-cast discards" },
+	{ ENETC_PBFDVFR,	"SI VLAN nomatch b-cast discards" },
+	{ ENETC_PFDMSAPR,	"SI pruning discarded frames" },
+	{ ENETC_PICDR(0),	"ICM DR0 discarded frames" },
+	{ ENETC_PICDR(1),	"ICM DR1 discarded frames" },
+	{ ENETC_PICDR(2),	"ICM DR2 discarded frames" },
+	{ ENETC_PICDR(3),	"ICM DR3 discarded frames" },
+};
+
+static const char rx_ring_stats[][ETH_GSTRING_LEN] = {
+	"Rx ring %2d frames",
+	"Rx ring %2d alloc errors",
+	"Rx ring %2d XDP drops",
+	"Rx ring %2d recycles",
+	"Rx ring %2d recycle failures",
+	"Rx ring %2d redirects",
+	"Rx ring %2d redirect failures",
+};
+
+static const char tx_ring_stats[][ETH_GSTRING_LEN] = {
+	"Tx ring %2d frames",
+	"Tx ring %2d XDP frames",
+	"Tx ring %2d XDP drops",
+	"Tx window drop %2d frames",
+};
+
+static int enetc_get_sset_count(struct net_device *ndev, int sset)
+{
+	struct enetc_ndev_priv *priv = netdev_priv(ndev);
+	int len;
+
+	if (sset != ETH_SS_STATS)
+		return -EOPNOTSUPP;
+
+	len = ARRAY_SIZE(enetc_si_counters) +
+	      ARRAY_SIZE(tx_ring_stats) * priv->num_tx_rings +
+	      ARRAY_SIZE(rx_ring_stats) * priv->num_rx_rings;
+
+	if (!enetc_si_is_pf(priv->si))
+		return len;
+
+	len += ARRAY_SIZE(enetc_port_counters);
+	len += ARRAY_SIZE(enetc_pm_counters);
+
+	return len;
+}
+
+static void enetc_get_strings(struct net_device *ndev, u32 stringset, u8 *data)
+{
+	struct enetc_ndev_priv *priv = netdev_priv(ndev);
+	int i, j;
+
+	switch (stringset) {
+	case ETH_SS_STATS:
+		for (i = 0; i < ARRAY_SIZE(enetc_si_counters); i++)
+			ethtool_puts(&data, enetc_si_counters[i].name);
+		for (i = 0; i < priv->num_tx_rings; i++)
+			for (j = 0; j < ARRAY_SIZE(tx_ring_stats); j++)
+				ethtool_sprintf(&data, tx_ring_stats[j], i);
+		for (i = 0; i < priv->num_rx_rings; i++)
+			for (j = 0; j < ARRAY_SIZE(rx_ring_stats); j++)
+				ethtool_sprintf(&data, rx_ring_stats[j], i);
+
+		if (!enetc_si_is_pf(priv->si))
+			break;
+
+		for (i = 0; i < ARRAY_SIZE(enetc_port_counters); i++)
+			ethtool_cpy(&data, enetc_port_counters[i].name);
+
+		for (i = 0; i < ARRAY_SIZE(enetc_pm_counters); i++)
+			ethtool_cpy(&data, enetc_pm_counters[i].name);
+
+		break;
+	}
+}
+
+static void enetc_get_ethtool_stats(struct net_device *ndev,
+				    struct ethtool_stats *stats, u64 *data)
+{
+	struct enetc_ndev_priv *priv = netdev_priv(ndev);
+	struct enetc_hw *hw = &priv->si->hw;
+	int i, o = 0;
+
+	for (i = 0; i < ARRAY_SIZE(enetc_si_counters); i++)
+		data[o++] = enetc_rd64(hw, enetc_si_counters[i].reg);
+
+	for (i = 0; i < priv->num_tx_rings; i++) {
+		data[o++] = priv->tx_ring[i]->stats.packets;
+		data[o++] = priv->tx_ring[i]->stats.xdp_tx;
+		data[o++] = priv->tx_ring[i]->stats.xdp_tx_drops;
+		data[o++] = priv->tx_ring[i]->stats.win_drop;
+	}
+
+	for (i = 0; i < priv->num_rx_rings; i++) {
+		data[o++] = priv->rx_ring[i]->stats.packets;
+		data[o++] = priv->rx_ring[i]->stats.rx_alloc_errs;
+		data[o++] = priv->rx_ring[i]->stats.xdp_drops;
+		data[o++] = priv->rx_ring[i]->stats.recycles;
+		data[o++] = priv->rx_ring[i]->stats.recycle_failures;
+		data[o++] = priv->rx_ring[i]->stats.xdp_redirect;
+		data[o++] = priv->rx_ring[i]->stats.xdp_redirect_failures;
+	}
+
+	if (!enetc_si_is_pf(priv->si))
+		return;
+
+	for (i = 0; i < ARRAY_SIZE(enetc_port_counters); i++)
+		data[o++] = enetc_port_rd(hw, enetc_port_counters[i].reg);
+
+	for (i = 0; i < ARRAY_SIZE(enetc_pm_counters); i++)
+		data[o++] = enetc_port_rd64(hw, enetc_pm_counters[i].reg);
+}
+
+static void enetc_pause_stats(struct enetc_hw *hw, int mac,
+			      struct ethtool_pause_stats *pause_stats)
+{
+	pause_stats->tx_pause_frames = enetc_port_rd64(hw, ENETC_PM_TXPF(mac));
+	pause_stats->rx_pause_frames = enetc_port_rd64(hw, ENETC_PM_RXPF(mac));
+}
+
+static void enetc_get_pause_stats(struct net_device *ndev,
+				  struct ethtool_pause_stats *pause_stats)
+{
+	struct enetc_ndev_priv *priv = netdev_priv(ndev);
+	struct enetc_hw *hw = &priv->si->hw;
+	struct enetc_si *si = priv->si;
+
+	switch (pause_stats->src) {
+	case ETHTOOL_MAC_STATS_SRC_EMAC:
+		enetc_pause_stats(hw, 0, pause_stats);
+		break;
+	case ETHTOOL_MAC_STATS_SRC_PMAC:
+		if (si->hw_features & ENETC_SI_F_QBU)
+			enetc_pause_stats(hw, 1, pause_stats);
+		break;
+	case ETHTOOL_MAC_STATS_SRC_AGGREGATE:
+		ethtool_aggregate_pause_stats(ndev, pause_stats);
+		break;
+	}
+}
+
+static void enetc_mac_stats(struct enetc_hw *hw, int mac,
+			    struct ethtool_eth_mac_stats *s)
+{
+	s->FramesTransmittedOK = enetc_port_rd64(hw, ENETC_PM_TFRM(mac));
+	s->SingleCollisionFrames = enetc_port_rd64(hw, ENETC_PM_TSCOL(mac));
+	s->MultipleCollisionFrames = enetc_port_rd64(hw, ENETC_PM_TMCOL(mac));
+	s->FramesReceivedOK = enetc_port_rd64(hw, ENETC_PM_RFRM(mac));
+	s->FrameCheckSequenceErrors = enetc_port_rd64(hw, ENETC_PM_RFCS(mac));
+	s->AlignmentErrors = enetc_port_rd64(hw, ENETC_PM_RALN(mac));
+	s->OctetsTransmittedOK = enetc_port_rd64(hw, ENETC_PM_TEOCT(mac));
+	s->FramesWithDeferredXmissions = enetc_port_rd64(hw, ENETC_PM_TDFR(mac));
+	s->LateCollisions = enetc_port_rd64(hw, ENETC_PM_TLCOL(mac));
+	s->FramesAbortedDueToXSColls = enetc_port_rd64(hw, ENETC_PM_TECOL(mac));
+	s->FramesLostDueToIntMACXmitError = enetc_port_rd64(hw, ENETC_PM_TERR(mac));
+	s->CarrierSenseErrors = enetc_port_rd64(hw, ENETC_PM_TCRSE(mac));
+	s->OctetsReceivedOK = enetc_port_rd64(hw, ENETC_PM_REOCT(mac));
+	s->FramesLostDueToIntMACRcvError = enetc_port_rd64(hw, ENETC_PM_RDRNTP(mac));
+	s->MulticastFramesXmittedOK = enetc_port_rd64(hw, ENETC_PM_TMCA(mac));
+	s->BroadcastFramesXmittedOK = enetc_port_rd64(hw, ENETC_PM_TBCA(mac));
+	s->MulticastFramesReceivedOK = enetc_port_rd64(hw, ENETC_PM_RMCA(mac));
+	s->BroadcastFramesReceivedOK = enetc_port_rd64(hw, ENETC_PM_RBCA(mac));
+}
+
+static void enetc_ctrl_stats(struct enetc_hw *hw, int mac,
+			     struct ethtool_eth_ctrl_stats *s)
+{
+	s->MACControlFramesTransmitted = enetc_port_rd64(hw, ENETC_PM_TCNP(mac));
+	s->MACControlFramesReceived = enetc_port_rd64(hw, ENETC_PM_RCNP(mac));
+}
+
+static const struct ethtool_rmon_hist_range enetc_rmon_ranges[] = {
+	{   64,   64 },
+	{   65,  127 },
+	{  128,  255 },
+	{  256,  511 },
+	{  512, 1023 },
+	{ 1024, 1522 },
+	{ 1523, ENETC_MAC_MAXFRM_SIZE },
+	{},
+};
+
+static void enetc_rmon_stats(struct enetc_hw *hw, int mac,
+			     struct ethtool_rmon_stats *s)
+{
+	s->undersize_pkts = enetc_port_rd64(hw, ENETC_PM_RUND(mac));
+	s->oversize_pkts = enetc_port_rd64(hw, ENETC_PM_ROVR(mac));
+	s->fragments = enetc_port_rd64(hw, ENETC_PM_RFRG(mac));
+	s->jabbers = enetc_port_rd64(hw, ENETC_PM_RJBR(mac));
+
+	s->hist[0] = enetc_port_rd64(hw, ENETC_PM_R64(mac));
+	s->hist[1] = enetc_port_rd64(hw, ENETC_PM_R127(mac));
+	s->hist[2] = enetc_port_rd64(hw, ENETC_PM_R255(mac));
+	s->hist[3] = enetc_port_rd64(hw, ENETC_PM_R511(mac));
+	s->hist[4] = enetc_port_rd64(hw, ENETC_PM_R1023(mac));
+	s->hist[5] = enetc_port_rd64(hw, ENETC_PM_R1522(mac));
+	s->hist[6] = enetc_port_rd64(hw, ENETC_PM_R1523X(mac));
+
+	s->hist_tx[0] = enetc_port_rd64(hw, ENETC_PM_T64(mac));
+	s->hist_tx[1] = enetc_port_rd64(hw, ENETC_PM_T127(mac));
+	s->hist_tx[2] = enetc_port_rd64(hw, ENETC_PM_T255(mac));
+	s->hist_tx[3] = enetc_port_rd64(hw, ENETC_PM_T511(mac));
+	s->hist_tx[4] = enetc_port_rd64(hw, ENETC_PM_T1023(mac));
+	s->hist_tx[5] = enetc_port_rd64(hw, ENETC_PM_T1522(mac));
+	s->hist_tx[6] = enetc_port_rd64(hw, ENETC_PM_T1523X(mac));
+}
+
+static void enetc_get_eth_mac_stats(struct net_device *ndev,
+				    struct ethtool_eth_mac_stats *mac_stats)
+{
+	struct enetc_ndev_priv *priv = netdev_priv(ndev);
+	struct enetc_hw *hw = &priv->si->hw;
+	struct enetc_si *si = priv->si;
+
+	switch (mac_stats->src) {
+	case ETHTOOL_MAC_STATS_SRC_EMAC:
+		enetc_mac_stats(hw, 0, mac_stats);
+		break;
+	case ETHTOOL_MAC_STATS_SRC_PMAC:
+		if (si->hw_features & ENETC_SI_F_QBU)
+			enetc_mac_stats(hw, 1, mac_stats);
+		break;
+	case ETHTOOL_MAC_STATS_SRC_AGGREGATE:
+		ethtool_aggregate_mac_stats(ndev, mac_stats);
+		break;
+	}
+}
+
+static void enetc_get_eth_ctrl_stats(struct net_device *ndev,
+				     struct ethtool_eth_ctrl_stats *ctrl_stats)
+{
+	struct enetc_ndev_priv *priv = netdev_priv(ndev);
+	struct enetc_hw *hw = &priv->si->hw;
+	struct enetc_si *si = priv->si;
+
+	switch (ctrl_stats->src) {
+	case ETHTOOL_MAC_STATS_SRC_EMAC:
+		enetc_ctrl_stats(hw, 0, ctrl_stats);
+		break;
+	case ETHTOOL_MAC_STATS_SRC_PMAC:
+		if (si->hw_features & ENETC_SI_F_QBU)
+			enetc_ctrl_stats(hw, 1, ctrl_stats);
+		break;
+	case ETHTOOL_MAC_STATS_SRC_AGGREGATE:
+		ethtool_aggregate_ctrl_stats(ndev, ctrl_stats);
+		break;
+	}
+}
+
+static void enetc_get_rmon_stats(struct net_device *ndev,
+				 struct ethtool_rmon_stats *rmon_stats,
+				 const struct ethtool_rmon_hist_range **ranges)
+{
+	struct enetc_ndev_priv *priv = netdev_priv(ndev);
+	struct enetc_hw *hw = &priv->si->hw;
+	struct enetc_si *si = priv->si;
+
+	*ranges = enetc_rmon_ranges;
+
+	switch (rmon_stats->src) {
+	case ETHTOOL_MAC_STATS_SRC_EMAC:
+		enetc_rmon_stats(hw, 0, rmon_stats);
+		break;
+	case ETHTOOL_MAC_STATS_SRC_PMAC:
+		if (si->hw_features & ENETC_SI_F_QBU)
+			enetc_rmon_stats(hw, 1, rmon_stats);
+		break;
+	case ETHTOOL_MAC_STATS_SRC_AGGREGATE:
+		ethtool_aggregate_rmon_stats(ndev, rmon_stats);
+		break;
+	}
+}
+
+#define ENETC_RSSHASH_L3 (RXH_L2DA | RXH_VLAN | RXH_L3_PROTO | RXH_IP_SRC | \
+			  RXH_IP_DST)
+#define ENETC_RSSHASH_L4 (ENETC_RSSHASH_L3 | RXH_L4_B_0_1 | RXH_L4_B_2_3)
+static int enetc_get_rxfh_fields(struct net_device *netdev,
+				 struct ethtool_rxfh_fields *rxnfc)
+{
+	static const u32 rsshash[] = {
+			[TCP_V4_FLOW]    = ENETC_RSSHASH_L4,
+			[UDP_V4_FLOW]    = ENETC_RSSHASH_L4,
+			[SCTP_V4_FLOW]   = ENETC_RSSHASH_L4,
+			[AH_ESP_V4_FLOW] = ENETC_RSSHASH_L3,
+			[IPV4_FLOW]      = ENETC_RSSHASH_L3,
+			[TCP_V6_FLOW]    = ENETC_RSSHASH_L4,
+			[UDP_V6_FLOW]    = ENETC_RSSHASH_L4,
+			[SCTP_V6_FLOW]   = ENETC_RSSHASH_L4,
+			[AH_ESP_V6_FLOW] = ENETC_RSSHASH_L3,
+			[IPV6_FLOW]      = ENETC_RSSHASH_L3,
+			[ETHER_FLOW]     = 0,
+	};
+
+	if (rxnfc->flow_type >= ARRAY_SIZE(rsshash))
+		return -EINVAL;
+
+	rxnfc->data = rsshash[rxnfc->flow_type];
+
+	return 0;
+}
+
+/* current HW spec does byte reversal on everything including MAC addresses */
+static void ether_addr_copy_swap(u8 *dst, const u8 *src)
+{
+	int i;
+
+	for (i = 0; i < ETH_ALEN; i++)
+		dst[i] = src[ETH_ALEN - i - 1];
+}
+
+static int enetc_set_cls_entry(struct enetc_si *si,
+			       struct ethtool_rx_flow_spec *fs, bool en)
+{
+	struct ethtool_tcpip4_spec *l4ip4_h, *l4ip4_m;
+	struct ethtool_usrip4_spec *l3ip4_h, *l3ip4_m;
+	struct ethhdr *eth_h, *eth_m;
+	struct enetc_cmd_rfse rfse = { {0} };
+
+	if (!en)
+		goto done;
+
+	switch (fs->flow_type & 0xff) {
+	case TCP_V4_FLOW:
+		l4ip4_h = &fs->h_u.tcp_ip4_spec;
+		l4ip4_m = &fs->m_u.tcp_ip4_spec;
+		goto l4ip4;
+	case UDP_V4_FLOW:
+		l4ip4_h = &fs->h_u.udp_ip4_spec;
+		l4ip4_m = &fs->m_u.udp_ip4_spec;
+		goto l4ip4;
+	case SCTP_V4_FLOW:
+		l4ip4_h = &fs->h_u.sctp_ip4_spec;
+		l4ip4_m = &fs->m_u.sctp_ip4_spec;
+l4ip4:
+		rfse.sip_h[0] = l4ip4_h->ip4src;
+		rfse.sip_m[0] = l4ip4_m->ip4src;
+		rfse.dip_h[0] = l4ip4_h->ip4dst;
+		rfse.dip_m[0] = l4ip4_m->ip4dst;
+		rfse.sport_h = ntohs(l4ip4_h->psrc);
+		rfse.sport_m = ntohs(l4ip4_m->psrc);
+		rfse.dport_h = ntohs(l4ip4_h->pdst);
+		rfse.dport_m = ntohs(l4ip4_m->pdst);
+		if (l4ip4_m->tos)
+			netdev_warn(si->ndev, "ToS field is not supported and was ignored\n");
+		rfse.ethtype_h = ETH_P_IP; /* IPv4 */
+		rfse.ethtype_m = 0xffff;
+		break;
+	case IP_USER_FLOW:
+		l3ip4_h = &fs->h_u.usr_ip4_spec;
+		l3ip4_m = &fs->m_u.usr_ip4_spec;
+
+		rfse.sip_h[0] = l3ip4_h->ip4src;
+		rfse.sip_m[0] = l3ip4_m->ip4src;
+		rfse.dip_h[0] = l3ip4_h->ip4dst;
+		rfse.dip_m[0] = l3ip4_m->ip4dst;
+		if (l3ip4_m->tos)
+			netdev_warn(si->ndev, "ToS field is not supported and was ignored\n");
+		rfse.ethtype_h = ETH_P_IP; /* IPv4 */
+		rfse.ethtype_m = 0xffff;
+		break;
+	case ETHER_FLOW:
+		eth_h = &fs->h_u.ether_spec;
+		eth_m = &fs->m_u.ether_spec;
+
+		ether_addr_copy_swap(rfse.smac_h, eth_h->h_source);
+		ether_addr_copy_swap(rfse.smac_m, eth_m->h_source);
+		ether_addr_copy_swap(rfse.dmac_h, eth_h->h_dest);
+		ether_addr_copy_swap(rfse.dmac_m, eth_m->h_dest);
+		rfse.ethtype_h = ntohs(eth_h->h_proto);
+		rfse.ethtype_m = ntohs(eth_m->h_proto);
+		break;
+	default:
+		return -EOPNOTSUPP;
+	}
+
+	rfse.mode |= ENETC_RFSE_EN;
+	if (fs->ring_cookie != RX_CLS_FLOW_DISC) {
+		rfse.mode |= ENETC_RFSE_MODE_BD;
+		rfse.result = fs->ring_cookie;
+	}
+done:
+	return enetc_set_fs_entry(si, &rfse, fs->location);
+}
+
+static int enetc_get_rxnfc(struct net_device *ndev, struct ethtool_rxnfc *rxnfc,
+			   u32 *rule_locs)
+{
+	struct enetc_ndev_priv *priv = netdev_priv(ndev);
+	int i, j;
+
+	switch (rxnfc->cmd) {
+	case ETHTOOL_GRXRINGS:
+		rxnfc->data = priv->num_rx_rings;
+		break;
+	case ETHTOOL_GRXCLSRLCNT:
+		/* total number of entries */
+		rxnfc->data = priv->si->num_fs_entries;
+		/* number of entries in use */
+		rxnfc->rule_cnt = 0;
+		for (i = 0; i < priv->si->num_fs_entries; i++)
+			if (priv->cls_rules[i].used)
+				rxnfc->rule_cnt++;
+		break;
+	case ETHTOOL_GRXCLSRULE:
+		if (rxnfc->fs.location >= priv->si->num_fs_entries)
+			return -EINVAL;
+
+		/* get entry x */
+		rxnfc->fs = priv->cls_rules[rxnfc->fs.location].fs;
+		break;
+	case ETHTOOL_GRXCLSRLALL:
+		/* total number of entries */
+		rxnfc->data = priv->si->num_fs_entries;
+		/* array of indexes of used entries */
+		j = 0;
+		for (i = 0; i < priv->si->num_fs_entries; i++) {
+			if (!priv->cls_rules[i].used)
+				continue;
+			if (j == rxnfc->rule_cnt)
+				return -EMSGSIZE;
+			rule_locs[j++] = i;
+		}
+		/* number of entries in use */
+		rxnfc->rule_cnt = j;
+		break;
+	default:
+		return -EOPNOTSUPP;
+	}
+
+	return 0;
+}
+
+/* i.MX95 ENETC does not support RFS table, but we can use ingress port
+ * filter table to implement Wake-on-LAN filter or drop the matched flow,
+ * so the implementation will be different from enetc_get_rxnfc() and
+ * enetc_set_rxnfc(). Therefore, add enetc4_get_rxnfc() for ENETC v4 PF.
+ */
+static int enetc4_get_rxnfc(struct net_device *ndev, struct ethtool_rxnfc *rxnfc,
+			    u32 *rule_locs)
+{
+	struct enetc_ndev_priv *priv = netdev_priv(ndev);
+
+	switch (rxnfc->cmd) {
+	case ETHTOOL_GRXRINGS:
+		rxnfc->data = priv->num_rx_rings;
+		break;
+	default:
+		return -EOPNOTSUPP;
+	}
+
+	return 0;
+}
+
+static int enetc_set_rxnfc(struct net_device *ndev, struct ethtool_rxnfc *rxnfc)
+{
+	struct enetc_ndev_priv *priv = netdev_priv(ndev);
+	int err;
+
+	switch (rxnfc->cmd) {
+	case ETHTOOL_SRXCLSRLINS:
+		if (rxnfc->fs.location >= priv->si->num_fs_entries)
+			return -EINVAL;
+
+		if (rxnfc->fs.ring_cookie >= priv->num_rx_rings &&
+		    rxnfc->fs.ring_cookie != RX_CLS_FLOW_DISC)
+			return -EINVAL;
+
+		err = enetc_set_cls_entry(priv->si, &rxnfc->fs, true);
+		if (err)
+			return err;
+		priv->cls_rules[rxnfc->fs.location].fs = rxnfc->fs;
+		priv->cls_rules[rxnfc->fs.location].used = 1;
+		break;
+	case ETHTOOL_SRXCLSRLDEL:
+		if (rxnfc->fs.location >= priv->si->num_fs_entries)
+			return -EINVAL;
+
+		err = enetc_set_cls_entry(priv->si, &rxnfc->fs, false);
+		if (err)
+			return err;
+		priv->cls_rules[rxnfc->fs.location].used = 0;
+		break;
+	default:
+		return -EOPNOTSUPP;
+	}
+
+	return 0;
+}
+
+static u32 enetc_get_rxfh_key_size(struct net_device *ndev)
+{
+	struct enetc_ndev_priv *priv = netdev_priv(ndev);
+
+	/* return the size of the RX flow hash key.  PF only */
+	return (priv->si->hw.port) ? ENETC_RSSHASH_KEY_SIZE : 0;
+}
+
+static u32 enetc_get_rxfh_indir_size(struct net_device *ndev)
+{
+	struct enetc_ndev_priv *priv = netdev_priv(ndev);
+
+	/* return the size of the RX flow hash indirection table */
+	return priv->si->num_rss;
+}
+
+static int enetc_get_rss_key_base(struct enetc_si *si)
+{
+	if (is_enetc_rev1(si))
+		return ENETC_PRSSK(0);
+
+	return ENETC4_PRSSKR(0);
+}
+
+static void enetc_get_rss_key(struct enetc_si *si, const u8 *key)
+{
+	int base = enetc_get_rss_key_base(si);
+	struct enetc_hw *hw = &si->hw;
+	int i;
+
+	for (i = 0; i < ENETC_RSSHASH_KEY_SIZE / 4; i++)
+		((u32 *)key)[i] = enetc_port_rd(hw, base + i * 4);
+}
+
+static int enetc_get_rxfh(struct net_device *ndev,
+			  struct ethtool_rxfh_param *rxfh)
+{
+	struct enetc_ndev_priv *priv = netdev_priv(ndev);
+	struct enetc_si *si = priv->si;
+	int err = 0;
+
+	/* return hash function */
+	rxfh->hfunc = ETH_RSS_HASH_TOP;
+
+	/* return hash key */
+	if (rxfh->key && enetc_si_is_pf(si))
+		enetc_get_rss_key(si, rxfh->key);
+
+	/* return RSS table */
+	if (rxfh->indir)
+		err = si->ops->get_rss_table(si, rxfh->indir, si->num_rss);
+
+	return err;
+}
+
+void enetc_set_rss_key(struct enetc_si *si, const u8 *bytes)
+{
+	int base = enetc_get_rss_key_base(si);
+	struct enetc_hw *hw = &si->hw;
+	int i;
+
+	for (i = 0; i < ENETC_RSSHASH_KEY_SIZE / 4; i++)
+		enetc_port_wr(hw, base + i * 4, ((u32 *)bytes)[i]);
+}
+EXPORT_SYMBOL_GPL(enetc_set_rss_key);
+
+static int enetc_set_rxfh(struct net_device *ndev,
+			  struct ethtool_rxfh_param *rxfh,
+			  struct netlink_ext_ack *extack)
+{
+	struct enetc_ndev_priv *priv = netdev_priv(ndev);
+	struct enetc_si *si = priv->si;
+	int err = 0;
+
+	/* set hash key, if PF */
+	if (rxfh->key && enetc_si_is_pf(si))
+		enetc_set_rss_key(si, rxfh->key);
+
+	/* set RSS table */
+	if (rxfh->indir)
+		err = si->ops->set_rss_table(si, rxfh->indir, si->num_rss);
+
+	return err;
+}
+
+static void enetc_get_ringparam(struct net_device *ndev,
+				struct ethtool_ringparam *ring,
+				struct kernel_ethtool_ringparam *kernel_ring,
+				struct netlink_ext_ack *extack)
+{
+	struct enetc_ndev_priv *priv = netdev_priv(ndev);
+
+	ring->rx_pending = priv->rx_bd_count;
+	ring->tx_pending = priv->tx_bd_count;
+
+	/* do some h/w sanity checks for BDR length */
+	if (netif_running(ndev)) {
+		struct enetc_hw *hw = &priv->si->hw;
+		u32 val = enetc_rxbdr_rd(hw, 0, ENETC_RBLENR);
+
+		if (val != priv->rx_bd_count)
+			netif_err(priv, hw, ndev, "RxBDR[RBLENR] = %d!\n", val);
+
+		val = enetc_txbdr_rd(hw, 0, ENETC_TBLENR);
+
+		if (val != priv->tx_bd_count)
+			netif_err(priv, hw, ndev, "TxBDR[TBLENR] = %d!\n", val);
+	}
+}
+
+static int enetc_get_coalesce(struct net_device *ndev,
+			      struct ethtool_coalesce *ic,
+			      struct kernel_ethtool_coalesce *kernel_coal,
+			      struct netlink_ext_ack *extack)
+{
+	struct enetc_ndev_priv *priv = netdev_priv(ndev);
+	struct enetc_int_vector *v = priv->int_vector[0];
+	u64 clk_freq = priv->sysclk_freq;
+
+	ic->tx_coalesce_usecs = enetc_cycles_to_usecs(priv->tx_ictt, clk_freq);
+	ic->rx_coalesce_usecs = enetc_cycles_to_usecs(v->rx_ictt, clk_freq);
+
+	ic->tx_max_coalesced_frames = ENETC_TXIC_PKTTHR;
+	ic->rx_max_coalesced_frames = ENETC_RXIC_PKTTHR;
+
+	ic->use_adaptive_rx_coalesce = priv->ic_mode & ENETC_IC_RX_ADAPTIVE;
+
+	return 0;
+}
+
+static int enetc_set_coalesce(struct net_device *ndev,
+			      struct ethtool_coalesce *ic,
+			      struct kernel_ethtool_coalesce *kernel_coal,
+			      struct netlink_ext_ack *extack)
+{
+	struct enetc_ndev_priv *priv = netdev_priv(ndev);
+	u64 clk_freq = priv->sysclk_freq;
+	u32 rx_ictt, tx_ictt;
+	int i, ic_mode;
+	bool changed;
+
+	tx_ictt = enetc_usecs_to_cycles(ic->tx_coalesce_usecs, clk_freq);
+	rx_ictt = enetc_usecs_to_cycles(ic->rx_coalesce_usecs, clk_freq);
+
+	if (ic->rx_max_coalesced_frames != ENETC_RXIC_PKTTHR)
+		return -EOPNOTSUPP;
+
+	if (ic->tx_max_coalesced_frames != ENETC_TXIC_PKTTHR)
+		return -EOPNOTSUPP;
+
+	ic_mode = ENETC_IC_NONE;
+	if (ic->use_adaptive_rx_coalesce) {
+		ic_mode |= ENETC_IC_RX_ADAPTIVE;
+		rx_ictt = 0x1;
+	} else {
+		ic_mode |= rx_ictt ? ENETC_IC_RX_MANUAL : 0;
+	}
+
+	ic_mode |= tx_ictt ? ENETC_IC_TX_MANUAL : 0;
+
+	/* commit the settings */
+	changed = (ic_mode != priv->ic_mode) || (priv->tx_ictt != tx_ictt);
+
+	priv->ic_mode = ic_mode;
+	priv->tx_ictt = tx_ictt;
+
+	for (i = 0; i < priv->bdr_int_num; i++) {
+		struct enetc_int_vector *v = priv->int_vector[i];
+
+		v->rx_ictt = rx_ictt;
+		v->rx_dim_en = !!(ic_mode & ENETC_IC_RX_ADAPTIVE);
+	}
+
+	if (netif_running(ndev) && changed) {
+		/* reconfigure the operation mode of h/w interrupts,
+		 * traffic needs to be paused in the process
+		 */
+		enetc_stop(ndev);
+		enetc_start(ndev);
+	}
+
+	return 0;
+}
+
+static int enetc_get_phc_index_by_pdev(struct enetc_si *si)
+{
+	struct pci_bus *bus = si->pdev->bus;
+	struct pci_dev *timer_pdev;
+	unsigned int devfn;
+	int phc_index;
+
+	switch (si->revision) {
+	case ENETC_REV_1_0:
+		devfn = PCI_DEVFN(0, 4);
+		break;
+	case ENETC_REV_4_1:
+		devfn = PCI_DEVFN(24, 0);
+		break;
+	default:
+		return -1;
+	}
+
+	timer_pdev = pci_get_domain_bus_and_slot(pci_domain_nr(bus),
+						 bus->number, devfn);
+	if (!timer_pdev)
+		return -1;
+
+	phc_index = ptp_clock_index_by_dev(&timer_pdev->dev);
+	pci_dev_put(timer_pdev);
+
+	return phc_index;
+}
+
+static int enetc_get_phc_index(struct enetc_si *si)
+{
+	struct device_node *np = si->pdev->dev.of_node;
+	struct device_node *timer_np;
+	int phc_index;
+
+	if (!np)
+		return enetc_get_phc_index_by_pdev(si);
+
+	timer_np = of_parse_phandle(np, "ptp-timer", 0);
+	if (!timer_np)
+		return enetc_get_phc_index_by_pdev(si);
+
+	phc_index = ptp_clock_index_by_of_node(timer_np);
+	of_node_put(timer_np);
+
+	return phc_index;
+}
+
+static void enetc_get_ts_generic_info(struct net_device *ndev,
+				      struct kernel_ethtool_ts_info *info)
+{
+	struct enetc_ndev_priv *priv = netdev_priv(ndev);
+
+	info->so_timestamping = SOF_TIMESTAMPING_TX_HARDWARE |
+				SOF_TIMESTAMPING_RX_HARDWARE |
+				SOF_TIMESTAMPING_RAW_HARDWARE |
+				SOF_TIMESTAMPING_TX_SOFTWARE;
+
+	info->tx_types = (1 << HWTSTAMP_TX_OFF) |
+			 (1 << HWTSTAMP_TX_ON);
+
+	if (enetc_si_is_pf(priv->si))
+		info->tx_types |= (1 << HWTSTAMP_TX_ONESTEP_SYNC);
+
+	info->rx_filters = (1 << HWTSTAMP_FILTER_NONE) |
+			   (1 << HWTSTAMP_FILTER_ALL);
+}
+
+static int enetc_get_ts_info(struct net_device *ndev,
+			     struct kernel_ethtool_ts_info *info)
+{
+	struct enetc_ndev_priv *priv = netdev_priv(ndev);
+	struct enetc_si *si = priv->si;
+
+	if (!enetc_ptp_clock_is_enabled(si))
+		goto timestamp_tx_sw;
+
+	info->phc_index = enetc_get_phc_index(si);
+	if (info->phc_index < 0)
+		goto timestamp_tx_sw;
+
+	enetc_get_ts_generic_info(ndev, info);
+
+	return 0;
+
+timestamp_tx_sw:
+	info->so_timestamping = SOF_TIMESTAMPING_TX_SOFTWARE;
+
+	return 0;
+}
+
+static void enetc_get_wol(struct net_device *dev,
+			  struct ethtool_wolinfo *wol)
+{
+	wol->supported = 0;
+	wol->wolopts = 0;
+
+	if (dev->phydev)
+		phy_ethtool_get_wol(dev->phydev, wol);
+}
+
+static int enetc_set_wol(struct net_device *dev,
+			 struct ethtool_wolinfo *wol)
+{
+	int ret;
+
+	if (!dev->phydev)
+		return -EOPNOTSUPP;
+
+	ret = phy_ethtool_set_wol(dev->phydev, wol);
+	if (!ret)
+		device_set_wakeup_enable(&dev->dev, wol->wolopts);
+
+	return ret;
+}
+
+static void enetc_get_pauseparam(struct net_device *dev,
+				 struct ethtool_pauseparam *pause)
+{
+	struct enetc_ndev_priv *priv = netdev_priv(dev);
+
+	phylink_ethtool_get_pauseparam(priv->phylink, pause);
+}
+
+static int enetc_set_pauseparam(struct net_device *dev,
+				struct ethtool_pauseparam *pause)
+{
+	struct enetc_ndev_priv *priv = netdev_priv(dev);
+
+	return phylink_ethtool_set_pauseparam(priv->phylink, pause);
+}
+
+static int enetc_get_link_ksettings(struct net_device *dev,
+				    struct ethtool_link_ksettings *cmd)
+{
+	struct enetc_ndev_priv *priv = netdev_priv(dev);
+
+	if (!priv->phylink)
+		return -EOPNOTSUPP;
+
+	return phylink_ethtool_ksettings_get(priv->phylink, cmd);
+}
+
+static int enetc_set_link_ksettings(struct net_device *dev,
+				    const struct ethtool_link_ksettings *cmd)
+{
+	struct enetc_ndev_priv *priv = netdev_priv(dev);
+
+	if (!priv->phylink)
+		return -EOPNOTSUPP;
+
+	return phylink_ethtool_ksettings_set(priv->phylink, cmd);
+}
+
+static void enetc_get_mm_stats(struct net_device *ndev,
+			       struct ethtool_mm_stats *s)
+{
+	struct enetc_ndev_priv *priv = netdev_priv(ndev);
+	struct enetc_hw *hw = &priv->si->hw;
+	struct enetc_si *si = priv->si;
+
+	if (!(si->hw_features & ENETC_SI_F_QBU))
+		return;
+
+	s->MACMergeFrameAssErrorCount = enetc_port_rd(hw, ENETC_MMFAECR);
+	s->MACMergeFrameSmdErrorCount = enetc_port_rd(hw, ENETC_MMFSECR);
+	s->MACMergeFrameAssOkCount = enetc_port_rd(hw, ENETC_MMFAOCR);
+	s->MACMergeFragCountRx = enetc_port_rd(hw, ENETC_MMFCRXR);
+	s->MACMergeFragCountTx = enetc_port_rd(hw, ENETC_MMFCTXR);
+	s->MACMergeHoldCount = enetc_port_rd(hw, ENETC_MMHCR);
+}
+
+static int enetc_get_mm(struct net_device *ndev, struct ethtool_mm_state *state)
+{
+	struct enetc_ndev_priv *priv = netdev_priv(ndev);
+	struct enetc_si *si = priv->si;
+	struct enetc_hw *hw = &si->hw;
+	u32 lafs, rafs, val;
+
+	if (!(si->hw_features & ENETC_SI_F_QBU))
+		return -EOPNOTSUPP;
+
+	mutex_lock(&priv->mm_lock);
+
+	val = enetc_port_rd(hw, ENETC_PFPMR);
+	state->pmac_enabled = !!(val & ENETC_PFPMR_PMACE);
+
+	val = enetc_port_rd(hw, ENETC_MMCSR);
+
+	switch (ENETC_MMCSR_GET_VSTS(val)) {
+	case 0:
+		state->verify_status = ETHTOOL_MM_VERIFY_STATUS_DISABLED;
+		break;
+	case 2:
+		state->verify_status = ETHTOOL_MM_VERIFY_STATUS_VERIFYING;
+		break;
+	case 3:
+		state->verify_status = ETHTOOL_MM_VERIFY_STATUS_SUCCEEDED;
+		break;
+	case 4:
+		state->verify_status = ETHTOOL_MM_VERIFY_STATUS_FAILED;
+		break;
+	case 5:
+	default:
+		state->verify_status = ETHTOOL_MM_VERIFY_STATUS_UNKNOWN;
+		break;
+	}
+
+	rafs = ENETC_MMCSR_GET_RAFS(val);
+	state->tx_min_frag_size = ethtool_mm_frag_size_add_to_min(rafs);
+	lafs = ENETC_MMCSR_GET_LAFS(val);
+	state->rx_min_frag_size = ethtool_mm_frag_size_add_to_min(lafs);
+	state->tx_enabled = !!(val & ENETC_MMCSR_LPE); /* mirror of MMCSR_ME */
+	state->tx_active = state->tx_enabled &&
+			   (state->verify_status == ETHTOOL_MM_VERIFY_STATUS_SUCCEEDED ||
+			    state->verify_status == ETHTOOL_MM_VERIFY_STATUS_DISABLED);
+	state->verify_enabled = !(val & ENETC_MMCSR_VDIS);
+	state->verify_time = ENETC_MMCSR_GET_VT(val);
+	/* A verifyTime of 128 ms would exceed the 7 bit width
+	 * of the ENETC_MMCSR_VT field
+	 */
+	state->max_verify_time = 127;
+
+	mutex_unlock(&priv->mm_lock);
+
+	return 0;
+}
+
+static int enetc_mm_wait_tx_active(struct enetc_hw *hw, int verify_time)
+{
+	int timeout = verify_time * USEC_PER_MSEC * ENETC_MM_VERIFY_RETRIES;
+	u32 val;
+
+	/* This will time out after the standard value of 3 verification
+	 * attempts. To not sleep forever, it relies on a non-zero verify_time,
+	 * guarantee which is provided by the ethtool nlattr policy.
+	 */
+	return read_poll_timeout(enetc_port_rd, val,
+				 ENETC_MMCSR_GET_VSTS(val) == 3,
+				 ENETC_MM_VERIFY_SLEEP_US, timeout,
+				 true, hw, ENETC_MMCSR);
+}
+
+static void enetc_set_ptcfpr(struct enetc_hw *hw, u8 preemptible_tcs)
+{
+	u32 val;
+	int tc;
+
+	for (tc = 0; tc < 8; tc++) {
+		val = enetc_port_rd(hw, ENETC_PTCFPR(tc));
+
+		if (preemptible_tcs & BIT(tc))
+			val |= ENETC_PTCFPR_FPE;
+		else
+			val &= ~ENETC_PTCFPR_FPE;
+
+		enetc_port_wr(hw, ENETC_PTCFPR(tc), val);
+	}
+}
+
+/* ENETC does not have an IRQ to notify changes to the MAC Merge TX status
+ * (active/inactive), but the preemptible traffic classes should only be
+ * committed to hardware once TX is active. Resort to polling.
+ */
+void enetc_mm_commit_preemptible_tcs(struct enetc_ndev_priv *priv)
+{
+	struct enetc_hw *hw = &priv->si->hw;
+	u8 preemptible_tcs = 0;
+	u32 val;
+	int err;
+
+	val = enetc_port_rd(hw, ENETC_MMCSR);
+	if (!(val & ENETC_MMCSR_ME))
+		goto out;
+
+	if (!(val & ENETC_MMCSR_VDIS)) {
+		err = enetc_mm_wait_tx_active(hw, ENETC_MMCSR_GET_VT(val));
+		if (err)
+			goto out;
+	}
+
+	preemptible_tcs = priv->preemptible_tcs;
+out:
+	enetc_set_ptcfpr(hw, preemptible_tcs);
+}
+
+/* FIXME: Workaround for the link partner's verification failing if ENETC
+ * priorly received too much express traffic. The documentation doesn't
+ * suggest this is needed.
+ */
+static void enetc_restart_emac_rx(struct enetc_si *si)
+{
+	u32 val = enetc_port_rd(&si->hw, ENETC_PM0_CMD_CFG);
+
+	enetc_port_wr(&si->hw, ENETC_PM0_CMD_CFG, val & ~ENETC_PM0_RX_EN);
+
+	if (val & ENETC_PM0_RX_EN)
+		enetc_port_wr(&si->hw, ENETC_PM0_CMD_CFG, val);
+}
+
+static int enetc_set_mm(struct net_device *ndev, struct ethtool_mm_cfg *cfg,
+			struct netlink_ext_ack *extack)
+{
+	struct enetc_ndev_priv *priv = netdev_priv(ndev);
+	struct enetc_hw *hw = &priv->si->hw;
+	struct enetc_si *si = priv->si;
+	u32 val, add_frag_size;
+	int err;
+
+	if (!(si->hw_features & ENETC_SI_F_QBU))
+		return -EOPNOTSUPP;
+
+	err = ethtool_mm_frag_size_min_to_add(cfg->tx_min_frag_size,
+					      &add_frag_size, extack);
+	if (err)
+		return err;
+
+	mutex_lock(&priv->mm_lock);
+
+	val = enetc_port_rd(hw, ENETC_PFPMR);
+	if (cfg->pmac_enabled)
+		val |= ENETC_PFPMR_PMACE;
+	else
+		val &= ~ENETC_PFPMR_PMACE;
+	enetc_port_wr(hw, ENETC_PFPMR, val);
+
+	val = enetc_port_rd(hw, ENETC_MMCSR);
+
+	if (cfg->verify_enabled)
+		val &= ~ENETC_MMCSR_VDIS;
+	else
+		val |= ENETC_MMCSR_VDIS;
+
+	if (cfg->tx_enabled)
+		priv->active_offloads |= ENETC_F_QBU;
+	else
+		priv->active_offloads &= ~ENETC_F_QBU;
+
+	/* If link is up, enable/disable MAC Merge right away */
+	if (!(val & ENETC_MMCSR_LINK_FAIL)) {
+		if (!!(priv->active_offloads & ENETC_F_QBU))
+			val |= ENETC_MMCSR_ME;
+		else
+			val &= ~ENETC_MMCSR_ME;
+	}
+
+	val &= ~ENETC_MMCSR_VT_MASK;
+	val |= ENETC_MMCSR_VT(cfg->verify_time);
+
+	val &= ~ENETC_MMCSR_RAFS_MASK;
+	val |= ENETC_MMCSR_RAFS(add_frag_size);
+
+	enetc_port_wr(hw, ENETC_MMCSR, val);
+
+	enetc_restart_emac_rx(priv->si);
+
+	enetc_mm_commit_preemptible_tcs(priv);
+
+	mutex_unlock(&priv->mm_lock);
+
+	return 0;
+}
+
+/* When the link is lost, the verification state machine goes to the FAILED
+ * state and doesn't restart on its own after a new link up event.
+ * According to 802.3 Figure 99-8 - Verify state diagram, the LINK_FAIL bit
+ * should have been sufficient to re-trigger verification, but for ENETC it
+ * doesn't. As a workaround, we need to toggle the Merge Enable bit to
+ * re-trigger verification when link comes up.
+ */
+void enetc_mm_link_state_update(struct enetc_ndev_priv *priv, bool link)
+{
+	struct enetc_hw *hw = &priv->si->hw;
+	u32 val;
+
+	mutex_lock(&priv->mm_lock);
+
+	val = enetc_port_rd(hw, ENETC_MMCSR);
+
+	if (link) {
+		val &= ~ENETC_MMCSR_LINK_FAIL;
+		if (priv->active_offloads & ENETC_F_QBU)
+			val |= ENETC_MMCSR_ME;
+	} else {
+		val |= ENETC_MMCSR_LINK_FAIL;
+		if (priv->active_offloads & ENETC_F_QBU)
+			val &= ~ENETC_MMCSR_ME;
+	}
+
+	enetc_port_wr(hw, ENETC_MMCSR, val);
+
+	enetc_mm_commit_preemptible_tcs(priv);
+
+	mutex_unlock(&priv->mm_lock);
+}
+EXPORT_SYMBOL_GPL(enetc_mm_link_state_update);
+
+const struct ethtool_ops enetc_pf_ethtool_ops = {
+	.supported_coalesce_params = ETHTOOL_COALESCE_USECS |
+				     ETHTOOL_COALESCE_MAX_FRAMES |
+				     ETHTOOL_COALESCE_USE_ADAPTIVE_RX,
+	.get_regs_len = enetc_get_reglen,
+	.get_regs = enetc_get_regs,
+	.get_sset_count = enetc_get_sset_count,
+	.get_strings = enetc_get_strings,
+	.get_ethtool_stats = enetc_get_ethtool_stats,
+	.get_pause_stats = enetc_get_pause_stats,
+	.get_rmon_stats = enetc_get_rmon_stats,
+	.get_eth_ctrl_stats = enetc_get_eth_ctrl_stats,
+	.get_eth_mac_stats = enetc_get_eth_mac_stats,
+	.get_rxnfc = enetc_get_rxnfc,
+	.set_rxnfc = enetc_set_rxnfc,
+	.get_rxfh_key_size = enetc_get_rxfh_key_size,
+	.get_rxfh_indir_size = enetc_get_rxfh_indir_size,
+	.get_rxfh = enetc_get_rxfh,
+	.set_rxfh = enetc_set_rxfh,
+	.get_rxfh_fields = enetc_get_rxfh_fields,
+	.get_ringparam = enetc_get_ringparam,
+	.get_coalesce = enetc_get_coalesce,
+	.set_coalesce = enetc_set_coalesce,
+	.get_link_ksettings = enetc_get_link_ksettings,
+	.set_link_ksettings = enetc_set_link_ksettings,
+	.get_link = ethtool_op_get_link,
+	.get_ts_info = enetc_get_ts_info,
+	.get_wol = enetc_get_wol,
+	.set_wol = enetc_set_wol,
+	.get_pauseparam = enetc_get_pauseparam,
+	.set_pauseparam = enetc_set_pauseparam,
+	.get_mm = enetc_get_mm,
+	.set_mm = enetc_set_mm,
+	.get_mm_stats = enetc_get_mm_stats,
+};
+
+const struct ethtool_ops enetc_vf_ethtool_ops = {
+	.supported_coalesce_params = ETHTOOL_COALESCE_USECS |
+				     ETHTOOL_COALESCE_MAX_FRAMES |
+				     ETHTOOL_COALESCE_USE_ADAPTIVE_RX,
+	.get_regs_len = enetc_get_reglen,
+	.get_regs = enetc_get_regs,
+	.get_sset_count = enetc_get_sset_count,
+	.get_strings = enetc_get_strings,
+	.get_ethtool_stats = enetc_get_ethtool_stats,
+	.get_rxnfc = enetc_get_rxnfc,
+	.set_rxnfc = enetc_set_rxnfc,
+	.get_rxfh_indir_size = enetc_get_rxfh_indir_size,
+	.get_rxfh = enetc_get_rxfh,
+	.set_rxfh = enetc_set_rxfh,
+	.get_rxfh_fields = enetc_get_rxfh_fields,
+	.get_ringparam = enetc_get_ringparam,
+	.get_coalesce = enetc_get_coalesce,
+	.set_coalesce = enetc_set_coalesce,
+	.get_link = ethtool_op_get_link,
+	.get_ts_info = enetc_get_ts_info,
+};
+
+const struct ethtool_ops enetc4_pf_ethtool_ops = {
+	.supported_coalesce_params = ETHTOOL_COALESCE_USECS |
+				     ETHTOOL_COALESCE_MAX_FRAMES |
+				     ETHTOOL_COALESCE_USE_ADAPTIVE_RX,
+	.get_ringparam = enetc_get_ringparam,
+	.get_coalesce = enetc_get_coalesce,
+	.set_coalesce = enetc_set_coalesce,
+	.get_link_ksettings = enetc_get_link_ksettings,
+	.set_link_ksettings = enetc_set_link_ksettings,
+	.get_link = ethtool_op_get_link,
+	.get_wol = enetc_get_wol,
+	.set_wol = enetc_set_wol,
+	.get_pauseparam = enetc_get_pauseparam,
+	.set_pauseparam = enetc_set_pauseparam,
+	.get_rxnfc = enetc4_get_rxnfc,
+	.get_rxfh_key_size = enetc_get_rxfh_key_size,
+	.get_rxfh_indir_size = enetc_get_rxfh_indir_size,
+	.get_rxfh = enetc_get_rxfh,
+	.set_rxfh = enetc_set_rxfh,
+	.get_rxfh_fields = enetc_get_rxfh_fields,
+	.get_ts_info = enetc_get_ts_info,
+};
+
+void enetc_set_ethtool_ops(struct net_device *ndev)
+{
+	struct enetc_ndev_priv *priv = netdev_priv(ndev);
+
+	ndev->ethtool_ops = priv->si->drvdata->eth_ops;
+}
+EXPORT_SYMBOL_GPL(enetc_set_ethtool_ops);
diff --git a/drivers/net/ethernet/freescale/enetc/enetc_hw.h b/drivers/net/ethernet/freescale/enetc/enetc_hw.h
new file mode 100644
index 000000000000..377c96325814
--- /dev/null
+++ b/drivers/net/ethernet/freescale/enetc/enetc_hw.h
@@ -0,0 +1,1040 @@
+/* SPDX-License-Identifier: (GPL-2.0+ OR BSD-3-Clause) */
+/* Copyright 2017-2019 NXP */
+
+#include <linux/bitops.h>
+
+#define ENETC_MM_VERIFY_SLEEP_US	USEC_PER_MSEC
+#define ENETC_MM_VERIFY_RETRIES		3
+
+#define ENETC_NUM_TC			8
+
+/* ENETC device IDs */
+#define ENETC_DEV_ID_PF		0xe100
+#define ENETC_DEV_ID_VF		0xef00
+#define ENETC_DEV_ID_PTP	0xee02
+
+/* ENETC register block BAR */
+#define ENETC_BAR_REGS	0
+
+/** SI regs, offset: 0h */
+#define ENETC_SIMR	0
+#define ENETC_SIMR_EN	BIT(31)
+#define ENETC_SIMR_RSSE	BIT(0)
+#define ENETC_SICTR0	0x18
+#define ENETC_SICTR1	0x1c
+#define ENETC_SIPCAPR0	0x20
+#define ENETC_SIPCAPR0_RSS	BIT(8)
+#define ENETC_SIPCAPR0_RFS	BIT(2)
+#define ENETC_SIPCAPR0_LSO	BIT(1)
+#define ENETC_SIPCAPR1	0x24
+#define ENETC_SITGTGR	0x30
+#define ENETC_SIRBGCR	0x38
+/* cache attribute registers for transactions initiated by ENETC */
+#define ENETC_SICAR0	0x40
+#define ENETC_SICAR1	0x44
+#define ENETC_SICAR2	0x48
+/* rd snoop, no alloc
+ * wr snoop, no alloc, partial cache line update for BDs and full cache line
+ * update for data
+ */
+#define ENETC_SICAR_RD_COHERENT	0x2b2b0000
+#define ENETC_SICAR_WR_COHERENT	0x00006727
+#define ENETC_SICAR_MSI	0x00300030 /* rd/wr device, no snoop, no alloc */
+
+#define ENETC_SIPMAR0	0x80
+#define ENETC_SIPMAR1	0x84
+#define ENETC_SICVLANR1	0x90
+#define ENETC_SICVLANR2	0x94
+#define  SICVLANR_ETYPE	GENMASK(15, 0)
+
+/* VF-PF Message passing */
+#define ENETC_DEFAULT_MSG_SIZE	1024	/* and max size */
+/* msg size encoding: default and max msg value of 1024B encoded as 0 */
+static inline u32 enetc_vsi_set_msize(u32 size)
+{
+	return size < ENETC_DEFAULT_MSG_SIZE ? size >> 5 : 0;
+}
+
+#define ENETC_PSIMSGRR	0x204
+#define ENETC_PSIMSGRR_MR_MASK	GENMASK(2, 1)
+#define ENETC_PSIMSGRR_MR(n) BIT((n) + 1) /* n = VSI index */
+#define ENETC_PSIVMSGRCVAR0(n)	(0x210 + (n) * 0x8) /* n = VSI index */
+#define ENETC_PSIVMSGRCVAR1(n)	(0x214 + (n) * 0x8)
+
+#define ENETC_VSIMSGSR	0x204	/* RO */
+#define ENETC_VSIMSGSR_MB	BIT(0)
+#define ENETC_VSIMSGSR_MS	BIT(1)
+#define ENETC_VSIMSGSNDAR0	0x210
+#define ENETC_VSIMSGSNDAR1	0x214
+
+#define ENETC_SIMSGSR_SET_MC(val) ((val) << 16)
+#define ENETC_SIMSGSR_GET_MC(val) ((val) >> 16)
+
+/* SI statistics */
+#define ENETC_SIROCT	0x300
+#define ENETC_SIRFRM	0x308
+#define ENETC_SIRUCA	0x310
+#define ENETC_SIRMCA	0x318
+#define ENETC_SITOCT	0x320
+#define ENETC_SITFRM	0x328
+#define ENETC_SITUCA	0x330
+#define ENETC_SITMCA	0x338
+#define ENETC_RBDCR(n)	(0x8180 + (n) * 0x200)
+
+/* Control BDR regs */
+#define ENETC_SICBDRMR		0x800
+#define ENETC_SICBDRSR		0x804	/* RO */
+#define ENETC_SICBDRBAR0	0x810
+#define ENETC_SICBDRBAR1	0x814
+#define ENETC_SICBDRPIR		0x818
+#define ENETC_SICBDRCIR		0x81c
+#define ENETC_SICBDRLENR	0x820
+
+#define ENETC_SICAPR0	0x900
+#define ENETC_SICAPR1	0x904
+
+#define ENETC_PSIIER	0xa00
+#define ENETC_PSIIER_MR_MASK	GENMASK(2, 1)
+#define ENETC_PSIIDR	0xa08
+#define ENETC_SITXIDR	0xa18
+#define ENETC_SIRXIDR	0xa28
+#define ENETC_SIMSIVR	0xa30
+
+#define ENETC_SIMSITRV(n) (0xB00 + (n) * 0x4)
+#define ENETC_SIMSIRRV(n) (0xB80 + (n) * 0x4)
+
+#define ENETC_SIUEFDCR	0xe28
+
+#define ENETC_SIRFSCAPR	0x1200
+#define ENETC_SIRFSCAPR_GET_NUM_RFS(val) ((val) & 0x7f)
+#define ENETC_SIRSSCAPR	0x1600
+#define ENETC_SIRSSCAPR_GET_NUM_RSS(val) (BIT((val) & 0xf) * 32)
+
+/** SI BDR sub-blocks, n = 0..7 */
+enum enetc_bdr_type {TX, RX};
+#define ENETC_BDR_OFF(i)	((i) * 0x200)
+#define ENETC_BDR(t, i, r)	(0x8000 + (t) * 0x100 + ENETC_BDR_OFF(i) + (r))
+/* RX BDR reg offsets */
+#define ENETC_RBMR	0
+#define ENETC_RBMR_BDS	BIT(2)
+#define ENETC_RBMR_CM	BIT(4)
+#define ENETC_RBMR_VTE	BIT(5)
+#define ENETC_RBMR_EN	BIT(31)
+#define ENETC_RBSR	0x4
+#define ENETC_RBBSR	0x8
+#define ENETC_RBCIR	0xc
+#define ENETC_RBBAR0	0x10
+#define ENETC_RBBAR1	0x14
+#define ENETC_RBPIR	0x18
+#define ENETC_RBLENR	0x20
+#define ENETC_RBIER	0xa0
+#define ENETC_RBIER_RXTIE	BIT(0)
+#define ENETC_RBIDR	0xa4
+#define ENETC_RBICR0	0xa8
+#define ENETC_RBICR0_ICEN		BIT(31)
+#define ENETC_RBICR0_ICPT_MASK		0x1ff
+#define ENETC_RBICR0_SET_ICPT(n)	((n) & ENETC_RBICR0_ICPT_MASK)
+#define ENETC_RBICR1	0xac
+
+/* TX BDR reg offsets */
+#define ENETC_TBMR	0
+#define ENETC_TBSR_BUSY	BIT(0)
+#define ENETC_TBMR_VIH	BIT(9)
+#define ENETC_TBMR_PRIO_MASK		GENMASK(2, 0)
+#define ENETC_TBMR_SET_PRIO(val)	((val) & ENETC_TBMR_PRIO_MASK)
+#define ENETC_TBMR_EN	BIT(31)
+#define ENETC_TBSR	0x4
+#define ENETC_TBBAR0	0x10
+#define ENETC_TBBAR1	0x14
+#define ENETC_TBPIR	0x18
+#define ENETC_TBCIR	0x1c
+#define ENETC_TBCIR_IDX_MASK	0xffff
+#define ENETC_TBLENR	0x20
+#define ENETC_TBIER	0xa0
+#define ENETC_TBIER_TXTIE	BIT(0)
+#define ENETC_TBIDR	0xa4
+#define ENETC_TBICR0	0xa8
+#define ENETC_TBICR0_ICEN		BIT(31)
+#define ENETC_TBICR0_ICPT_MASK		0xf
+#define ENETC_TBICR0_SET_ICPT(n) ((ilog2(n) + 1) & ENETC_TBICR0_ICPT_MASK)
+#define ENETC_TBICR1	0xac
+
+#define ENETC_RTBLENR_LEN(n)	((n) & ~0x7)
+
+/* Port regs, offset: 1_0000h */
+#define ENETC_PORT_BASE		0x10000
+#define ENETC_PMR		0x0000
+#define ENETC_PMR_EN	GENMASK(18, 16)
+#define ENETC_PMR_PSPEED_MASK GENMASK(11, 8)
+#define ENETC_PMR_PSPEED_10M	0
+#define ENETC_PMR_PSPEED_100M	BIT(8)
+#define ENETC_PMR_PSPEED_1000M	BIT(9)
+#define ENETC_PMR_PSPEED_2500M	BIT(10)
+#define ENETC_PSR		0x0004 /* RO */
+#define ENETC_PSIPMR		0x0018
+#define ENETC_PSIPMR_SET_UP(n)	BIT(n) /* n = SI index */
+#define ENETC_PSIPMR_SET_MP(n)	BIT((n) + 16)
+#define ENETC_PSIPVMR		0x001c
+#define ENETC_VLAN_PROMISC_MAP_ALL	0x7
+#define ENETC_PSIPVMR_SET_VP(simap)	((simap) & 0x7)
+#define ENETC_PSIPVMR_SET_VUTA(simap)	(((simap) & 0x7) << 16)
+#define ENETC_PSIPMAR0(n)	(0x0100 + (n) * 0x8) /* n = SI index */
+#define ENETC_PSIPMAR1(n)	(0x0104 + (n) * 0x8)
+#define ENETC_PVCLCTR		0x0208
+#define ENETC_PCVLANR1		0x0210
+#define ENETC_PCVLANR2		0x0214
+#define ENETC_VLAN_TYPE_C	BIT(0)
+#define ENETC_VLAN_TYPE_S	BIT(1)
+#define ENETC_PVCLCTR_OVTPIDL(bmp)	((bmp) & 0xff) /* VLAN_TYPE */
+#define ENETC_PSIVLANR(n)	(0x0240 + (n) * 4) /* n = SI index */
+#define ENETC_PSIVLAN_EN	BIT(31)
+#define ENETC_PSIVLAN_SET_QOS(val)	((u32)(val) << 12)
+#define ENETC_PPAUONTR		0x0410
+#define ENETC_PPAUOFFTR		0x0414
+#define ENETC_PTXMBAR		0x0608
+#define ENETC_PCAPR0		0x0900
+#define ENETC_PCAPR0_RXBDR(val)	((val) >> 24)
+#define ENETC_PCAPR0_TXBDR(val)	(((val) >> 16) & 0xff)
+#define ENETC_PCAPR0_PSFP	BIT(9)
+#define ENETC_PCAPR0_QBV	BIT(4)
+#define ENETC_PCAPR0_QBU	BIT(3)
+#define ENETC_PCAPR1		0x0904
+#define ENETC_PSICFGR0(n)	(0x0940 + (n) * 0xc)  /* n = SI index */
+#define ENETC_PSICFGR0_SET_TXBDR(val)	((val) & 0xff)
+#define ENETC_PSICFGR0_SET_RXBDR(val)	(((val) & 0xff) << 16)
+#define ENETC_PSICFGR0_VTE	BIT(12)
+#define ENETC_PSICFGR0_SIVIE	BIT(14)
+#define ENETC_PSICFGR0_ASE	BIT(15)
+#define ENETC_PSICFGR0_SIVC(bmp)	(((bmp) & 0xff) << 24) /* VLAN_TYPE */
+
+#define ENETC_PTCCBSR0(n)	(0x1110 + (n) * 8) /* n = 0 to 7*/
+#define ENETC_CBSE		BIT(31)
+#define ENETC_CBS_BW_MASK	GENMASK(6, 0)
+#define ENETC_PTCCBSR1(n)	(0x1114 + (n) * 8) /* n = 0 to 7*/
+#define ENETC_RSSHASH_KEY_SIZE	40
+#define ENETC_PRSSCAPR		0x1404
+#define ENETC_PRSSCAPR_GET_NUM_RSS(val)	(BIT((val) & 0xf) * 32)
+#define ENETC_PRSSK(n)		(0x1410 + (n) * 4) /* n = [0..9] */
+#define ENETC_PSIVLANFMR	0x1700
+#define ENETC_PSIVLANFMR_VS	BIT(0)
+#define ENETC_PRFSMR		0x1800
+#define ENETC_PRFSMR_RFSE	BIT(31)
+#define ENETC_PRFSCAPR		0x1804
+#define ENETC_PRFSCAPR_GET_NUM_RFS(val)	((((val) & 0xf) + 1) * 16)
+#define ENETC_PSIRFSCFGR(n)	(0x1814 + (n) * 4) /* n = SI index */
+#define ENETC_PFPMR		0x1900
+#define ENETC_PFPMR_PMACE	BIT(1)
+#define ENETC_EMDIO_BASE	0x1c00
+#define ENETC_PSIUMHFR0(n, err)	(((err) ? 0x1d08 : 0x1d00) + (n) * 0x10)
+#define ENETC_PSIUMHFR1(n)	(0x1d04 + (n) * 0x10)
+#define ENETC_PSIMMHFR0(n, err)	(((err) ? 0x1d00 : 0x1d08) + (n) * 0x10)
+#define ENETC_PSIMMHFR1(n)	(0x1d0c + (n) * 0x10)
+#define ENETC_PSIVHFR0(n)	(0x1e00 + (n) * 8) /* n = SI index */
+#define ENETC_PSIVHFR1(n)	(0x1e04 + (n) * 8) /* n = SI index */
+#define ENETC_MMCSR		0x1f00
+#define ENETC_MMCSR_LINK_FAIL	BIT(31)
+#define ENETC_MMCSR_VT_MASK	GENMASK(29, 23) /* Verify Time */
+#define ENETC_MMCSR_VT(x)	(((x) << 23) & ENETC_MMCSR_VT_MASK)
+#define ENETC_MMCSR_GET_VT(x)	(((x) & ENETC_MMCSR_VT_MASK) >> 23)
+#define ENETC_MMCSR_TXSTS_MASK	GENMASK(22, 21) /* Merge Status */
+#define ENETC_MMCSR_GET_TXSTS(x) (((x) & ENETC_MMCSR_TXSTS_MASK) >> 21)
+#define ENETC_MMCSR_VSTS_MASK	GENMASK(20, 18) /* Verify Status */
+#define ENETC_MMCSR_GET_VSTS(x) (((x) & ENETC_MMCSR_VSTS_MASK) >> 18)
+#define ENETC_MMCSR_VDIS	BIT(17) /* Verify Disabled */
+#define ENETC_MMCSR_ME		BIT(16) /* Merge Enabled */
+#define ENETC_MMCSR_RAFS_MASK	GENMASK(9, 8) /* Remote Additional Fragment Size */
+#define ENETC_MMCSR_RAFS(x)	(((x) << 8) & ENETC_MMCSR_RAFS_MASK)
+#define ENETC_MMCSR_GET_RAFS(x)	(((x) & ENETC_MMCSR_RAFS_MASK) >> 8)
+#define ENETC_MMCSR_LAFS_MASK	GENMASK(4, 3) /* Local Additional Fragment Size */
+#define ENETC_MMCSR_GET_LAFS(x)	(((x) & ENETC_MMCSR_LAFS_MASK) >> 3)
+#define ENETC_MMCSR_LPA		BIT(2) /* Local Preemption Active */
+#define ENETC_MMCSR_LPE		BIT(1) /* Local Preemption Enabled */
+#define ENETC_MMCSR_LPS		BIT(0) /* Local Preemption Supported */
+#define ENETC_MMFAECR		0x1f08
+#define ENETC_MMFSECR		0x1f0c
+#define ENETC_MMFAOCR		0x1f10
+#define ENETC_MMFCRXR		0x1f14
+#define ENETC_MMFCTXR		0x1f18
+#define ENETC_MMHCR		0x1f1c
+#define ENETC_PTCMSDUR(n)	(0x2020 + (n) * 4) /* n = TC index [0..7] */
+
+#define ENETC_PMAC_OFFSET	0x1000
+
+#define ENETC_PM0_CMD_CFG	0x8008
+#define ENETC_PM0_TX_EN		BIT(0)
+#define ENETC_PM0_RX_EN		BIT(1)
+#define ENETC_PM0_PROMISC	BIT(4)
+#define ENETC_PM0_PAUSE_IGN	BIT(8)
+#define ENETC_PM0_CMD_XGLP	BIT(10)
+#define ENETC_PM0_CMD_TXP	BIT(11)
+#define ENETC_PM0_CMD_PHY_TX_EN	BIT(15)
+#define ENETC_PM0_CMD_SFD	BIT(21)
+#define ENETC_PM0_MAXFRM	0x8014
+#define ENETC_SET_TX_MTU(val)	((val) << 16)
+#define ENETC_SET_MAXFRM(val)	((val) & 0xffff)
+#define ENETC_PM0_RX_FIFO	0x801c
+#define ENETC_PM0_RX_FIFO_VAL	1
+
+#define ENETC_PM_IMDIO_BASE	0x8030
+
+#define ENETC_PM0_PAUSE_QUANTA	0x8054
+#define ENETC_PM0_PAUSE_THRESH	0x8064
+
+#define ENETC_PM0_SINGLE_STEP		0x80c0
+#define ENETC_PM0_SINGLE_STEP_CH	BIT(7)
+#define ENETC_PM0_SINGLE_STEP_EN	BIT(31)
+#define ENETC_SET_SINGLE_STEP_OFFSET(v)	(((v) & 0xff) << 8)
+
+#define ENETC_PM0_IF_MODE	0x8300
+#define ENETC_PM0_IFM_RG	BIT(2)
+#define ENETC_PM0_IFM_RLP	(BIT(5) | BIT(11))
+#define ENETC_PM0_IFM_EN_AUTO	BIT(15)
+#define ENETC_PM0_IFM_SSP_MASK	GENMASK(14, 13)
+#define ENETC_PM0_IFM_SSP_1000	(2 << 13)
+#define ENETC_PM0_IFM_SSP_100	(0 << 13)
+#define ENETC_PM0_IFM_SSP_10	(1 << 13)
+#define ENETC_PM0_IFM_FULL_DPX	BIT(12)
+#define ENETC_PM0_IFM_IFMODE_MASK GENMASK(1, 0)
+#define ENETC_PM0_IFM_IFMODE_XGMII 0
+#define ENETC_PM0_IFM_IFMODE_GMII 2
+#define ENETC_PSIDCAPR		0x1b08
+#define ENETC_PSIDCAPR_MSK	GENMASK(15, 0)
+#define ENETC_PSFCAPR		0x1b18
+#define ENETC_PSFCAPR_MSK	GENMASK(15, 0)
+#define ENETC_PSGCAPR		0x1b28
+#define ENETC_PSGCAPR_GCL_MSK	GENMASK(18, 16)
+#define ENETC_PSGCAPR_SGIT_MSK	GENMASK(15, 0)
+#define ENETC_PFMCAPR		0x1b38
+#define ENETC_PFMCAPR_MSK	GENMASK(15, 0)
+
+/* Port MAC counters: Port MAC 0 corresponds to the eMAC and
+ * Port MAC 1 to the pMAC.
+ */
+#define ENETC_PM_REOCT(mac)	(0x8100 + ENETC_PMAC_OFFSET * (mac))
+#define ENETC_PM_RALN(mac)	(0x8110 + ENETC_PMAC_OFFSET * (mac))
+#define ENETC_PM_RXPF(mac)	(0x8118 + ENETC_PMAC_OFFSET * (mac))
+#define ENETC_PM_RFRM(mac)	(0x8120 + ENETC_PMAC_OFFSET * (mac))
+#define ENETC_PM_RFCS(mac)	(0x8128 + ENETC_PMAC_OFFSET * (mac))
+#define ENETC_PM_RVLAN(mac)	(0x8130 + ENETC_PMAC_OFFSET * (mac))
+#define ENETC_PM_RERR(mac)	(0x8138 + ENETC_PMAC_OFFSET * (mac))
+#define ENETC_PM_RUCA(mac)	(0x8140 + ENETC_PMAC_OFFSET * (mac))
+#define ENETC_PM_RMCA(mac)	(0x8148 + ENETC_PMAC_OFFSET * (mac))
+#define ENETC_PM_RBCA(mac)	(0x8150 + ENETC_PMAC_OFFSET * (mac))
+#define ENETC_PM_RDRP(mac)	(0x8158 + ENETC_PMAC_OFFSET * (mac))
+#define ENETC_PM_RPKT(mac)	(0x8160 + ENETC_PMAC_OFFSET * (mac))
+#define ENETC_PM_RUND(mac)	(0x8168 + ENETC_PMAC_OFFSET * (mac))
+#define ENETC_PM_R64(mac)	(0x8170 + ENETC_PMAC_OFFSET * (mac))
+#define ENETC_PM_R127(mac)	(0x8178 + ENETC_PMAC_OFFSET * (mac))
+#define ENETC_PM_R255(mac)	(0x8180 + ENETC_PMAC_OFFSET * (mac))
+#define ENETC_PM_R511(mac)	(0x8188 + ENETC_PMAC_OFFSET * (mac))
+#define ENETC_PM_R1023(mac)	(0x8190 + ENETC_PMAC_OFFSET * (mac))
+#define ENETC_PM_R1522(mac)	(0x8198 + ENETC_PMAC_OFFSET * (mac))
+#define ENETC_PM_R1523X(mac)	(0x81A0 + ENETC_PMAC_OFFSET * (mac))
+#define ENETC_PM_ROVR(mac)	(0x81A8 + ENETC_PMAC_OFFSET * (mac))
+#define ENETC_PM_RJBR(mac)	(0x81B0 + ENETC_PMAC_OFFSET * (mac))
+#define ENETC_PM_RFRG(mac)	(0x81B8 + ENETC_PMAC_OFFSET * (mac))
+#define ENETC_PM_RCNP(mac)	(0x81C0 + ENETC_PMAC_OFFSET * (mac))
+#define ENETC_PM_RDRNTP(mac)	(0x81C8 + ENETC_PMAC_OFFSET * (mac))
+#define ENETC_PM_TEOCT(mac)	(0x8200 + ENETC_PMAC_OFFSET * (mac))
+#define ENETC_PM_TOCT(mac)	(0x8208 + ENETC_PMAC_OFFSET * (mac))
+#define ENETC_PM_TCRSE(mac)	(0x8210 + ENETC_PMAC_OFFSET * (mac))
+#define ENETC_PM_TXPF(mac)	(0x8218 + ENETC_PMAC_OFFSET * (mac))
+#define ENETC_PM_TFRM(mac)	(0x8220 + ENETC_PMAC_OFFSET * (mac))
+#define ENETC_PM_TFCS(mac)	(0x8228 + ENETC_PMAC_OFFSET * (mac))
+#define ENETC_PM_TVLAN(mac)	(0x8230 + ENETC_PMAC_OFFSET * (mac))
+#define ENETC_PM_TERR(mac)	(0x8238 + ENETC_PMAC_OFFSET * (mac))
+#define ENETC_PM_TUCA(mac)	(0x8240 + ENETC_PMAC_OFFSET * (mac))
+#define ENETC_PM_TMCA(mac)	(0x8248 + ENETC_PMAC_OFFSET * (mac))
+#define ENETC_PM_TBCA(mac)	(0x8250 + ENETC_PMAC_OFFSET * (mac))
+#define ENETC_PM_TPKT(mac)	(0x8260 + ENETC_PMAC_OFFSET * (mac))
+#define ENETC_PM_TUND(mac)	(0x8268 + ENETC_PMAC_OFFSET * (mac))
+#define ENETC_PM_T64(mac)	(0x8270 + ENETC_PMAC_OFFSET * (mac))
+#define ENETC_PM_T127(mac)	(0x8278 + ENETC_PMAC_OFFSET * (mac))
+#define ENETC_PM_T255(mac)	(0x8280 + ENETC_PMAC_OFFSET * (mac))
+#define ENETC_PM_T511(mac)	(0x8288 + ENETC_PMAC_OFFSET * (mac))
+#define ENETC_PM_T1023(mac)	(0x8290 + ENETC_PMAC_OFFSET * (mac))
+#define ENETC_PM_T1522(mac)	(0x8298 + ENETC_PMAC_OFFSET * (mac))
+#define ENETC_PM_T1523X(mac)	(0x82A0 + ENETC_PMAC_OFFSET * (mac))
+#define ENETC_PM_TCNP(mac)	(0x82C0 + ENETC_PMAC_OFFSET * (mac))
+#define ENETC_PM_TDFR(mac)	(0x82D0 + ENETC_PMAC_OFFSET * (mac))
+#define ENETC_PM_TMCOL(mac)	(0x82D8 + ENETC_PMAC_OFFSET * (mac))
+#define ENETC_PM_TSCOL(mac)	(0x82E0 + ENETC_PMAC_OFFSET * (mac))
+#define ENETC_PM_TLCOL(mac)	(0x82E8 + ENETC_PMAC_OFFSET * (mac))
+#define ENETC_PM_TECOL(mac)	(0x82F0 + ENETC_PMAC_OFFSET * (mac))
+
+/* Port counters */
+#define ENETC_PICDR(n)		(0x0700 + (n) * 8) /* n = [0..3] */
+#define ENETC_PBFDSIR		0x0810
+#define ENETC_PFDMSAPR		0x0814
+#define ENETC_UFDMF		0x1680
+#define ENETC_MFDMF		0x1684
+#define ENETC_PUFDVFR		0x1780
+#define ENETC_PMFDVFR		0x1784
+#define ENETC_PBFDVFR		0x1788
+
+/** Global regs, offset: 2_0000h */
+#define ENETC_GLOBAL_BASE	0x20000
+#define ENETC_G_EIPBRR0		0x0bf8
+#define EIPBRR0_REVISION	GENMASK(15, 0)
+#define ENETC_REV_1_0		0x0100
+#define ENETC_REV_4_1		0X0401
+
+#define ENETC_G_EIPBRR1		0x0bfc
+#define ENETC_G_EPFBLPR(n)	(0xd00 + 4 * (n))
+#define ENETC_G_EPFBLPR1_XGMII	0x80000000
+
+/* PCI device info */
+struct enetc_hw {
+	/* SI registers, used by all PCI functions */
+	void __iomem *reg;
+	/* Port registers, PF only */
+	void __iomem *port;
+	/* IP global registers, PF only */
+	void __iomem *global;
+};
+
+/* ENETC register accessors */
+
+/* MDIO issue workaround (on LS1028A) -
+ * Due to a hardware issue, an access to MDIO registers
+ * that is concurrent with other ENETC register accesses
+ * may lead to the MDIO access being dropped or corrupted.
+ * To protect the MDIO accesses a readers-writers locking
+ * scheme is used, where the MDIO register accesses are
+ * protected by write locks to insure exclusivity, while
+ * the remaining ENETC registers are accessed under read
+ * locks since they only compete with MDIO accesses.
+ */
+extern rwlock_t enetc_mdio_lock;
+
+DECLARE_STATIC_KEY_FALSE(enetc_has_err050089);
+
+/* use this locking primitive only on the fast datapath to
+ * group together multiple non-MDIO register accesses to
+ * minimize the overhead of the lock
+ */
+static inline void enetc_lock_mdio(void)
+{
+	if (static_branch_unlikely(&enetc_has_err050089))
+		read_lock(&enetc_mdio_lock);
+}
+
+static inline void enetc_unlock_mdio(void)
+{
+	if (static_branch_unlikely(&enetc_has_err050089))
+		read_unlock(&enetc_mdio_lock);
+}
+
+/* use these accessors only on the fast datapath under
+ * the enetc_lock_mdio() locking primitive to minimize
+ * the overhead of the lock
+ */
+static inline u32 enetc_rd_reg_hot(void __iomem *reg)
+{
+	if (static_branch_unlikely(&enetc_has_err050089))
+		lockdep_assert_held(&enetc_mdio_lock);
+
+	return ioread32(reg);
+}
+
+static inline void enetc_wr_reg_hot(void __iomem *reg, u32 val)
+{
+	if (static_branch_unlikely(&enetc_has_err050089))
+		lockdep_assert_held(&enetc_mdio_lock);
+
+	iowrite32(val, reg);
+}
+
+/* internal helpers for the MDIO w/a */
+static inline u32 _enetc_rd_reg_wa(void __iomem *reg)
+{
+	u32 val;
+
+	enetc_lock_mdio();
+	val = ioread32(reg);
+	enetc_unlock_mdio();
+
+	return val;
+}
+
+static inline void _enetc_wr_reg_wa(void __iomem *reg, u32 val)
+{
+	enetc_lock_mdio();
+	iowrite32(val, reg);
+	enetc_unlock_mdio();
+}
+
+static inline u32 _enetc_rd_mdio_reg_wa(void __iomem *reg)
+{
+	unsigned long flags;
+	u32 val;
+
+	if (static_branch_unlikely(&enetc_has_err050089)) {
+		write_lock_irqsave(&enetc_mdio_lock, flags);
+		val = ioread32(reg);
+		write_unlock_irqrestore(&enetc_mdio_lock, flags);
+	} else {
+		val = ioread32(reg);
+	}
+
+	return val;
+}
+
+static inline void _enetc_wr_mdio_reg_wa(void __iomem *reg, u32 val)
+{
+	unsigned long flags;
+
+	if (static_branch_unlikely(&enetc_has_err050089)) {
+		write_lock_irqsave(&enetc_mdio_lock, flags);
+		iowrite32(val, reg);
+		write_unlock_irqrestore(&enetc_mdio_lock, flags);
+	} else {
+		iowrite32(val, reg);
+	}
+}
+
+#ifdef ioread64
+static inline u64 _enetc_rd_reg64(void __iomem *reg)
+{
+	return ioread64(reg);
+}
+#else
+/* using this to read out stats on 32b systems */
+static inline u64 _enetc_rd_reg64(void __iomem *reg)
+{
+	u32 low, high, tmp;
+
+	do {
+		high = ioread32(reg + 4);
+		low = ioread32(reg);
+		tmp = ioread32(reg + 4);
+	} while (high != tmp);
+
+	return (u64)high << 32 | low;
+}
+#endif
+
+static inline u64 _enetc_rd_reg64_wa(void __iomem *reg)
+{
+	u64 val;
+
+	enetc_lock_mdio();
+	val = _enetc_rd_reg64(reg);
+	enetc_unlock_mdio();
+
+	return val;
+}
+
+/* general register accessors */
+#define enetc_rd_reg(reg)		_enetc_rd_reg_wa((reg))
+#define enetc_wr_reg(reg, val)		_enetc_wr_reg_wa((reg), (val))
+#define enetc_rd(hw, off)		enetc_rd_reg((hw)->reg + (off))
+#define enetc_wr(hw, off, val)		enetc_wr_reg((hw)->reg + (off), val)
+#define enetc_rd_hot(hw, off)		enetc_rd_reg_hot((hw)->reg + (off))
+#define enetc_wr_hot(hw, off, val)	enetc_wr_reg_hot((hw)->reg + (off), val)
+#define enetc_rd64(hw, off)		_enetc_rd_reg64_wa((hw)->reg + (off))
+/* port register accessors - PF only */
+#define enetc_port_rd(hw, off)		enetc_rd_reg((hw)->port + (off))
+#define enetc_port_wr(hw, off, val)	enetc_wr_reg((hw)->port + (off), val)
+#define enetc_port_rd64(hw, off)	_enetc_rd_reg64_wa((hw)->port + (off))
+#define enetc_port_rd_mdio(hw, off)	_enetc_rd_mdio_reg_wa((hw)->port + (off))
+#define enetc_port_wr_mdio(hw, off, val)	_enetc_wr_mdio_reg_wa(\
+							(hw)->port + (off), val)
+/* global register accessors - PF only */
+#define enetc_global_rd(hw, off)	enetc_rd_reg((hw)->global + (off))
+#define enetc_global_wr(hw, off, val)	enetc_wr_reg((hw)->global + (off), val)
+/* BDR register accessors, see ENETC_BDR() */
+#define enetc_bdr_rd(hw, t, n, off) \
+				enetc_rd(hw, ENETC_BDR(t, n, off))
+#define enetc_bdr_wr(hw, t, n, off, val) \
+				enetc_wr(hw, ENETC_BDR(t, n, off), val)
+#define enetc_txbdr_rd(hw, n, off) enetc_bdr_rd(hw, TX, n, off)
+#define enetc_rxbdr_rd(hw, n, off) enetc_bdr_rd(hw, RX, n, off)
+#define enetc_txbdr_wr(hw, n, off, val) \
+				enetc_bdr_wr(hw, TX, n, off, val)
+#define enetc_rxbdr_wr(hw, n, off, val) \
+				enetc_bdr_wr(hw, RX, n, off, val)
+
+/* Buffer Descriptors (BD) */
+union enetc_tx_bd {
+	struct {
+		__le64 addr;
+		union {
+			__le16 buf_len;
+			__le16 hdr_len;	/* For LSO, ENETC 4.1 and later */
+		};
+		__le16 frm_len;
+		union {
+			struct {
+				u8 l3_aux0;
+#define ENETC_TX_BD_L3_START	GENMASK(6, 0)
+#define ENETC_TX_BD_IPCS	BIT(7)
+				u8 l3_aux1;
+#define ENETC_TX_BD_L3_HDR_LEN	GENMASK(6, 0)
+#define ENETC_TX_BD_L3T		BIT(7)
+				u8 l4_aux;
+#define ENETC_TX_BD_L4T		GENMASK(7, 5)
+#define ENETC_TXBD_L4T_UDP	1
+#define ENETC_TXBD_L4T_TCP	2
+				u8 flags;
+			}; /* default layout */
+			__le32 txstart;
+			__le32 lstatus;
+		};
+	};
+	struct {
+		__le32 tstamp;
+		__le16 tpid;
+		__le16 vid;
+		__le16 lso_sg_size; /* For ENETC 4.1 and later */
+		__le16 frm_len_ext; /* For ENETC 4.1 and later */
+		u8 reserved[2];
+		u8 e_flags;
+		u8 flags;
+	} ext; /* Tx BD extension */
+	struct {
+		__le32 tstamp;
+		u8 reserved[8];
+		__le16 lso_err_count; /* For ENETC 4.1 and later */
+		u8 status;
+		u8 flags;
+	} wb; /* writeback descriptor */
+};
+
+enum enetc_txbd_flags {
+	ENETC_TXBD_FLAGS_L4CS = BIT(0), /* For ENETC 4.1 and later */
+	ENETC_TXBD_FLAGS_TSE = BIT(1),
+	ENETC_TXBD_FLAGS_LSO = BIT(1), /* For ENETC 4.1 and later */
+	ENETC_TXBD_FLAGS_W = BIT(2),
+	ENETC_TXBD_FLAGS_CSUM_LSO = BIT(3), /* For ENETC 4.1 and later */
+	ENETC_TXBD_FLAGS_TXSTART = BIT(4),
+	ENETC_TXBD_FLAGS_EX = BIT(6),
+	ENETC_TXBD_FLAGS_F = BIT(7)
+};
+#define ENETC_TXBD_STATS_WIN	BIT(7)
+#define ENETC_TXBD_TXSTART_MASK GENMASK(24, 0)
+#define ENETC_TXBD_FLAGS_OFFSET 24
+#define ENETC_TXBD_TSTAMP	GENMASK(29, 0)
+
+static inline __le32 enetc_txbd_set_tx_start(u64 tx_start, u8 flags)
+{
+	u32 temp;
+
+	temp = (tx_start >> 5 & ENETC_TXBD_TXSTART_MASK) |
+	       (flags << ENETC_TXBD_FLAGS_OFFSET);
+
+	return cpu_to_le32(temp);
+}
+
+static inline void enetc_clear_tx_bd(union enetc_tx_bd *txbd)
+{
+	memset(txbd, 0, sizeof(*txbd));
+}
+
+/* Extension flags */
+#define ENETC_TXBD_E_FLAGS_VLAN_INS	BIT(0)
+#define ENETC_TXBD_E_FLAGS_ONE_STEP_PTP	BIT(1)
+#define ENETC_TXBD_E_FLAGS_TWO_STEP_PTP	BIT(2)
+
+union enetc_rx_bd {
+	struct {
+		__le64 addr;
+		u8 reserved[8];
+	} w;
+	struct {
+		__le16 inet_csum;
+		__le16 parse_summary;
+		__le32 rss_hash;
+		__le16 buf_len;
+		__le16 vlan_opt;
+		union {
+			struct {
+				__le16 flags;
+				__le16 error;
+			};
+			__le32 lstatus;
+		};
+	} r;
+	struct {
+		__le32 tstamp;
+		u8 reserved[12];
+	} ext;
+};
+
+#define ENETC_RXBD_LSTATUS_R	BIT(30)
+#define ENETC_RXBD_LSTATUS_F	BIT(31)
+#define ENETC_RXBD_ERR_MASK	0xff
+#define ENETC_RXBD_LSTATUS(flags)	((flags) << 16)
+#define ENETC_RXBD_FLAG_VLAN	BIT(9)
+#define ENETC_RXBD_FLAG_TSTMP	BIT(10)
+#define ENETC_RXBD_FLAG_TPID	GENMASK(1, 0)
+
+#define ENETC_MAC_ADDR_FILT_CNT	8 /* # of supported entries per port */
+#define EMETC_MAC_ADDR_FILT_RES	3 /* # of reserved entries at the beginning */
+#define ENETC_MAX_NUM_VFS	2
+
+#define ENETC_CBD_FLAGS_SF	BIT(7) /* short format */
+#define ENETC_CBD_STATUS_MASK	0xf
+
+#define ENETC_TPID_8021Q	0
+
+struct enetc_cmd_rfse {
+	u8 smac_h[6];
+	u8 smac_m[6];
+	u8 dmac_h[6];
+	u8 dmac_m[6];
+	__be32 sip_h[4];
+	__be32 sip_m[4];
+	__be32 dip_h[4];
+	__be32 dip_m[4];
+	u16 ethtype_h;
+	u16 ethtype_m;
+	u16 ethtype4_h;
+	u16 ethtype4_m;
+	u16 sport_h;
+	u16 sport_m;
+	u16 dport_h;
+	u16 dport_m;
+	u16 vlan_h;
+	u16 vlan_m;
+	u8 proto_h;
+	u8 proto_m;
+	u16 flags;
+	u16 result;
+	u16 mode;
+};
+
+#define ENETC_RFSE_EN	BIT(15)
+#define ENETC_RFSE_MODE_BD	2
+
+static inline void enetc_load_primary_mac_addr(struct enetc_hw *hw,
+					       struct net_device *ndev)
+{
+	u8 addr[ETH_ALEN] __aligned(4);
+
+	*(u32 *)addr = __raw_readl(hw->reg + ENETC_SIPMAR0);
+	*(u16 *)(addr + 4) = __raw_readw(hw->reg + ENETC_SIPMAR1);
+	eth_hw_addr_set(ndev, addr);
+}
+
+#define ENETC_SI_INT_IDX	0
+/* base index for Rx/Tx interrupts */
+#define ENETC_BDR_INT_BASE_IDX	1
+
+/* Messaging */
+
+/* Command completion status */
+enum enetc_msg_cmd_status {
+	ENETC_MSG_CMD_STATUS_OK,
+	ENETC_MSG_CMD_STATUS_FAIL
+};
+
+/* VSI-PSI command message types */
+enum enetc_msg_cmd_type {
+	ENETC_MSG_CMD_MNG_MAC = 1, /* manage MAC address */
+	ENETC_MSG_CMD_MNG_RX_MAC_FILTER,/* manage RX MAC table */
+	ENETC_MSG_CMD_MNG_RX_VLAN_FILTER /* manage RX VLAN table */
+};
+
+/* VSI-PSI command action types */
+enum enetc_msg_cmd_action_type {
+	ENETC_MSG_CMD_MNG_ADD = 1,
+	ENETC_MSG_CMD_MNG_REMOVE
+};
+
+/* PSI-VSI command header format */
+struct enetc_msg_cmd_header {
+	u16 type;	/* command class type */
+	u16 id;		/* denotes the specific required action */
+};
+
+/* Common H/W utility functions */
+
+static inline void enetc_bdr_enable_rxvlan(struct enetc_hw *hw, int idx,
+					   bool en)
+{
+	u32 val = enetc_rxbdr_rd(hw, idx, ENETC_RBMR);
+
+	val = (val & ~ENETC_RBMR_VTE) | (en ? ENETC_RBMR_VTE : 0);
+	enetc_rxbdr_wr(hw, idx, ENETC_RBMR, val);
+}
+
+static inline void enetc_bdr_enable_txvlan(struct enetc_hw *hw, int idx,
+					   bool en)
+{
+	u32 val = enetc_txbdr_rd(hw, idx, ENETC_TBMR);
+
+	val = (val & ~ENETC_TBMR_VIH) | (en ? ENETC_TBMR_VIH : 0);
+	enetc_txbdr_wr(hw, idx, ENETC_TBMR, val);
+}
+
+static inline void enetc_set_bdr_prio(struct enetc_hw *hw, int bdr_idx,
+				      int prio)
+{
+	u32 val = enetc_txbdr_rd(hw, bdr_idx, ENETC_TBMR);
+
+	val &= ~ENETC_TBMR_PRIO_MASK;
+	val |= ENETC_TBMR_SET_PRIO(prio);
+	enetc_txbdr_wr(hw, bdr_idx, ENETC_TBMR, val);
+}
+
+enum bdcr_cmd_class {
+	BDCR_CMD_UNSPEC = 0,
+	BDCR_CMD_MAC_FILTER,
+	BDCR_CMD_VLAN_FILTER,
+	BDCR_CMD_RSS,
+	BDCR_CMD_RFS,
+	BDCR_CMD_PORT_GCL,
+	BDCR_CMD_RECV_CLASSIFIER,
+	BDCR_CMD_STREAM_IDENTIFY,
+	BDCR_CMD_STREAM_FILTER,
+	BDCR_CMD_STREAM_GCL,
+	BDCR_CMD_FLOW_METER,
+	__BDCR_CMD_MAX_LEN,
+	BDCR_CMD_MAX_LEN = __BDCR_CMD_MAX_LEN - 1,
+};
+
+/* class 5, command 0 */
+struct tgs_gcl_conf {
+	u8	atc;	/* init gate value */
+	u8	res[7];
+	struct {
+		u8	res1[4];
+		__le16	acl_len;
+		u8	res2[2];
+	};
+};
+
+/* gate control list entry */
+struct gce {
+	__le32	period;
+	u8	gate;
+	u8	res[3];
+};
+
+/* tgs_gcl_conf address point to this data space */
+struct tgs_gcl_data {
+	__le32		btl;
+	__le32		bth;
+	__le32		ct;
+	__le32		cte;
+	struct gce	entry[];
+};
+
+/* class 7, command 0, Stream Identity Entry Configuration */
+struct streamid_conf {
+	__le32	stream_handle;	/* init gate value */
+	__le32	iports;
+		u8	id_type;
+		u8	oui[3];
+		u8	res[3];
+		u8	en;
+};
+
+#define ENETC_CBDR_SID_VID_MASK 0xfff
+#define ENETC_CBDR_SID_VIDM BIT(12)
+#define ENETC_CBDR_SID_TG_MASK 0xc000
+/* streamid_conf address point to this data space */
+struct streamid_data {
+	union {
+		u8 dmac[6];
+		u8 smac[6];
+	};
+	u16     vid_vidm_tg;
+};
+
+#define ENETC_CBDR_SFI_PRI_MASK 0x7
+#define ENETC_CBDR_SFI_PRIM		BIT(3)
+#define ENETC_CBDR_SFI_BLOV		BIT(4)
+#define ENETC_CBDR_SFI_BLEN		BIT(5)
+#define ENETC_CBDR_SFI_MSDUEN	BIT(6)
+#define ENETC_CBDR_SFI_FMITEN	BIT(7)
+#define ENETC_CBDR_SFI_ENABLE	BIT(7)
+/* class 8, command 0, Stream Filter Instance, Short Format */
+struct sfi_conf {
+	__le32	stream_handle;
+		u8	multi;
+		u8	res[2];
+		u8	sthm;
+	/* Max Service Data Unit or Flow Meter Instance Table index.
+	 * Depending on the value of FLT this represents either Max
+	 * Service Data Unit (max frame size) allowed by the filter
+	 * entry or is an index into the Flow Meter Instance table
+	 * index identifying the policer which will be used to police
+	 * it.
+	 */
+	__le16	fm_inst_table_index;
+	__le16	msdu;
+	__le16	sg_inst_table_index;
+		u8	res1[2];
+	__le32	input_ports;
+		u8	res2[3];
+		u8	en;
+};
+
+/* class 8, command 2 stream Filter Instance status query short format
+ * command no need structure define
+ * Stream Filter Instance Query Statistics Response data
+ */
+struct sfi_counter_data {
+	u32 matchl;
+	u32 matchh;
+	u32 msdu_dropl;
+	u32 msdu_droph;
+	u32 stream_gate_dropl;
+	u32 stream_gate_droph;
+	u32 flow_meter_dropl;
+	u32 flow_meter_droph;
+};
+
+#define ENETC_CBDR_SGI_OIPV_MASK 0x7
+#define ENETC_CBDR_SGI_OIPV_EN	BIT(3)
+#define ENETC_CBDR_SGI_CGTST	BIT(6)
+#define ENETC_CBDR_SGI_OGTST	BIT(7)
+#define ENETC_CBDR_SGI_CFG_CHG  BIT(1)
+#define ENETC_CBDR_SGI_CFG_PND  BIT(2)
+#define ENETC_CBDR_SGI_OEX		BIT(4)
+#define ENETC_CBDR_SGI_OEXEN	BIT(5)
+#define ENETC_CBDR_SGI_IRX		BIT(6)
+#define ENETC_CBDR_SGI_IRXEN	BIT(7)
+#define ENETC_CBDR_SGI_ACLLEN_MASK 0x3
+#define ENETC_CBDR_SGI_OCLLEN_MASK 0xc
+#define	ENETC_CBDR_SGI_EN		BIT(7)
+/* class 9, command 0, Stream Gate Instance Table, Short Format
+ * class 9, command 2, Stream Gate Instance Table entry query write back
+ * Short Format
+ */
+struct sgi_table {
+	u8	res[8];
+	u8	oipv;
+	u8	res0[2];
+	u8	ocgtst;
+	u8	res1[7];
+	u8	gset;
+	u8	oacl_len;
+	u8	res2[2];
+	u8	en;
+};
+
+#define ENETC_CBDR_SGI_AIPV_MASK 0x7
+#define ENETC_CBDR_SGI_AIPV_EN	BIT(3)
+#define ENETC_CBDR_SGI_AGTST	BIT(7)
+
+/* class 9, command 1, Stream Gate Control List, Long Format */
+struct sgcl_conf {
+	u8	aipv;
+	u8	res[2];
+	u8	agtst;
+	u8	res1[4];
+	union {
+		struct {
+			u8 res2[4];
+			u8 acl_len;
+			u8 res3[3];
+		};
+		u8 cct[8]; /* Config change time */
+	};
+};
+
+#define ENETC_CBDR_SGL_IOMEN	BIT(0)
+#define ENETC_CBDR_SGL_IPVEN	BIT(3)
+#define ENETC_CBDR_SGL_GTST		BIT(4)
+#define ENETC_CBDR_SGL_IPV_MASK 0xe
+/* Stream Gate Control List Entry */
+struct sgce {
+	u32	interval;
+	u8	msdu[3];
+	u8	multi;
+};
+
+/* stream control list class 9 , cmd 1 data buffer */
+struct sgcl_data {
+	u32		btl;
+	u32		bth;
+	u32		ct;
+	u32		cte;
+	struct sgce	sgcl[];
+};
+
+#define ENETC_CBDR_FMI_MR	BIT(0)
+#define ENETC_CBDR_FMI_MREN	BIT(1)
+#define ENETC_CBDR_FMI_DOY	BIT(2)
+#define	ENETC_CBDR_FMI_CM	BIT(3)
+#define ENETC_CBDR_FMI_CF	BIT(4)
+#define ENETC_CBDR_FMI_NDOR	BIT(5)
+#define ENETC_CBDR_FMI_OALEN	BIT(6)
+#define ENETC_CBDR_FMI_IRFPP_MASK GENMASK(4, 0)
+
+/* class 10: command 0/1, Flow Meter Instance Set, short Format */
+struct fmi_conf {
+	__le32	cir;
+	__le32	cbs;
+	__le32	eir;
+	__le32	ebs;
+		u8	conf;
+		u8	res1;
+		u8	ir_fpp;
+		u8	res2[4];
+		u8	en;
+};
+
+struct enetc_cbd {
+	union{
+		struct sfi_conf sfi_conf;
+		struct sgi_table sgi_table;
+		struct fmi_conf fmi_conf;
+		struct {
+			__le32	addr[2];
+			union {
+				__le32	opt[4];
+				struct tgs_gcl_conf	gcl_conf;
+				struct streamid_conf	sid_set;
+				struct sgcl_conf	sgcl_conf;
+			};
+		};	/* Long format */
+		__le32 data[6];
+	};
+	__le16 index;
+	__le16 length;
+	u8 cmd;
+	u8 cls;
+	u8 _res;
+	u8 status_flags;
+};
+
+#define ENETC_CLK_400M		400000000ULL
+#define ENETC_CLK_333M		333000000ULL
+
+static inline u32 enetc_cycles_to_usecs(u32 cycles, u64 clk_freq)
+{
+	return (u32)div_u64(cycles * 1000000ULL, clk_freq);
+}
+
+static inline u32 enetc_usecs_to_cycles(u32 usecs, u64 clk_freq)
+{
+	return (u32)div_u64(usecs * clk_freq, 1000000ULL);
+}
+
+/* Port traffic class frame preemption register */
+#define ENETC_PTCFPR(n)			(0x1910 + (n) * 4) /* n = [0 ..7] */
+#define ENETC_PTCFPR_FPE		BIT(31)
+
+/* port time gating control register */
+#define ENETC_PTGCR			0x11a00
+#define ENETC_PTGCR_TGE			BIT(31)
+#define ENETC_PTGCR_TGPE		BIT(30)
+
+/* Port time gating capability register */
+#define ENETC_PTGCAPR			0x11a08
+#define ENETC_PTGCAPR_MAX_GCL_LEN_MASK	GENMASK(15, 0)
+
+/* Port time specific departure */
+#define ENETC_PTCTSDR(n)	(0x1210 + 4 * (n))
+#define ENETC_TSDE		BIT(31)
+
+/* PSFP setting */
+#define ENETC_PPSFPMR 0x11b00
+#define ENETC_PPSFPMR_PSFPEN BIT(0)
+#define ENETC_PPSFPMR_VS BIT(1)
+#define ENETC_PPSFPMR_PVC BIT(2)
+#define ENETC_PPSFPMR_PVZC BIT(3)
diff --git a/drivers/net/ethernet/freescale/enetc/enetc_ierb.c b/drivers/net/ethernet/freescale/enetc/enetc_ierb.c
new file mode 100644
index 000000000000..d39617ab9306
--- /dev/null
+++ b/drivers/net/ethernet/freescale/enetc/enetc_ierb.c
@@ -0,0 +1,147 @@
+// SPDX-License-Identifier: (GPL-2.0+ OR BSD-3-Clause)
+/* Copyright 2021 NXP
+ *
+ * The Integrated Endpoint Register Block (IERB) is configured by pre-boot
+ * software and is supposed to be to ENETC what a NVRAM is to a 'real' PCIe
+ * card. Upon FLR, values from the IERB are transferred to the ENETC PFs, and
+ * are read-only in the PF memory space.
+ *
+ * This driver fixes up the power-on reset values for the ENETC shared FIFO,
+ * such that the TX and RX allocations are sufficient for jumbo frames, and
+ * that intelligent FIFO dropping is enabled before the internal data
+ * structures are corrupted.
+ *
+ * Even though not all ports might be used on a given board, we are not
+ * concerned with partitioning the FIFO, because the default values configure
+ * no strict reservations, so the entire FIFO can be used by the RX of a single
+ * port, or the TX of a single port.
+ */
+
+#include <linux/io.h>
+#include <linux/mod_devicetable.h>
+#include <linux/module.h>
+#include <linux/pci.h>
+#include <linux/platform_device.h>
+#include "enetc.h"
+#include "enetc_ierb.h"
+
+/* IERB registers */
+#define ENETC_IERB_TXMBAR(port)			(((port) * 0x100) + 0x8080)
+#define ENETC_IERB_RXMBER(port)			(((port) * 0x100) + 0x8090)
+#define ENETC_IERB_RXMBLR(port)			(((port) * 0x100) + 0x8094)
+#define ENETC_IERB_RXBCR(port)			(((port) * 0x100) + 0x80a0)
+#define ENETC_IERB_TXBCR(port)			(((port) * 0x100) + 0x80a8)
+#define ENETC_IERB_FMBDTR			0xa000
+
+#define ENETC_RESERVED_FOR_ICM			1024
+
+struct enetc_ierb {
+	void __iomem *regs;
+};
+
+static void enetc_ierb_write(struct enetc_ierb *ierb, u32 offset, u32 val)
+{
+	iowrite32(val, ierb->regs + offset);
+}
+
+int enetc_ierb_register_pf(struct platform_device *pdev,
+			   struct pci_dev *pf_pdev)
+{
+	struct enetc_ierb *ierb = platform_get_drvdata(pdev);
+	int port = enetc_pf_to_port(pf_pdev);
+	u16 tx_credit, rx_credit, tx_alloc;
+
+	if (port < 0)
+		return -ENODEV;
+
+	if (!ierb)
+		return -EPROBE_DEFER;
+
+	/* By default, it is recommended to set the Host Transfer Agent
+	 * per port transmit byte credit to "1000 + max_frame_size/2".
+	 * The power-on reset value (1800 bytes) is rounded up to the nearest
+	 * 100 assuming a maximum frame size of 1536 bytes.
+	 */
+	tx_credit = roundup(1000 + ENETC_MAC_MAXFRM_SIZE / 2, 100);
+
+	/* Internal memory allocated for transmit buffering is guaranteed but
+	 * not reserved; i.e. if the total transmit allocation is not used,
+	 * then the unused portion is not left idle, it can be used for receive
+	 * buffering but it will be reclaimed, if required, from receive by
+	 * intelligently dropping already stored receive frames in the internal
+	 * memory to ensure that the transmit allocation is respected.
+	 *
+	 * PaTXMBAR must be set to a value larger than
+	 *     PaTXBCR + 2 * max_frame_size + 32
+	 * if frame preemption is not enabled, or to
+	 *     2 * PaTXBCR + 2 * p_max_frame_size (pMAC maximum frame size) +
+	 *     2 * np_max_frame_size (eMAC maximum frame size) + 64
+	 * if frame preemption is enabled.
+	 */
+	tx_alloc = roundup(2 * tx_credit + 4 * ENETC_MAC_MAXFRM_SIZE + 64, 16);
+
+	/* Initial credits, in units of 8 bytes, to the Ingress Congestion
+	 * Manager for the maximum amount of bytes the port is allocated for
+	 * pending traffic.
+	 * It is recommended to set the initial credits to 2 times the maximum
+	 * frame size (2 frames of maximum size).
+	 */
+	rx_credit = DIV_ROUND_UP(ENETC_MAC_MAXFRM_SIZE * 2, 8);
+
+	enetc_ierb_write(ierb, ENETC_IERB_TXBCR(port), tx_credit);
+	enetc_ierb_write(ierb, ENETC_IERB_TXMBAR(port), tx_alloc);
+	enetc_ierb_write(ierb, ENETC_IERB_RXBCR(port), rx_credit);
+
+	return 0;
+}
+EXPORT_SYMBOL(enetc_ierb_register_pf);
+
+static int enetc_ierb_probe(struct platform_device *pdev)
+{
+	struct enetc_ierb *ierb;
+	void __iomem *regs;
+
+	ierb = devm_kzalloc(&pdev->dev, sizeof(*ierb), GFP_KERNEL);
+	if (!ierb)
+		return -ENOMEM;
+
+	regs = devm_platform_get_and_ioremap_resource(pdev, 0, NULL);
+	if (IS_ERR(regs))
+		return PTR_ERR(regs);
+
+	ierb->regs = regs;
+
+	/* Free buffer depletion threshold in bytes.
+	 * This sets the minimum amount of free buffer memory that should be
+	 * maintained in the datapath sub system, and when the amount of free
+	 * buffer memory falls below this threshold, a depletion indication is
+	 * asserted, which may trigger "intelligent drop" frame releases from
+	 * the ingress queues in the ICM.
+	 * It is recommended to set the free buffer depletion threshold to 1024
+	 * bytes, since the ICM needs some FIFO memory for its own use.
+	 */
+	enetc_ierb_write(ierb, ENETC_IERB_FMBDTR, ENETC_RESERVED_FOR_ICM);
+
+	platform_set_drvdata(pdev, ierb);
+
+	return 0;
+}
+
+static const struct of_device_id enetc_ierb_match[] = {
+	{ .compatible = "fsl,ls1028a-enetc-ierb", },
+	{},
+};
+MODULE_DEVICE_TABLE(of, enetc_ierb_match);
+
+static struct platform_driver enetc_ierb_driver = {
+	.driver = {
+		.name = "fsl-enetc-ierb",
+		.of_match_table = enetc_ierb_match,
+	},
+	.probe = enetc_ierb_probe,
+};
+
+module_platform_driver(enetc_ierb_driver);
+
+MODULE_DESCRIPTION("NXP ENETC IERB");
+MODULE_LICENSE("Dual BSD/GPL");
diff --git a/drivers/net/ethernet/freescale/enetc/enetc_ierb.h b/drivers/net/ethernet/freescale/enetc/enetc_ierb.h
new file mode 100644
index 000000000000..c2ce47c4be9f
--- /dev/null
+++ b/drivers/net/ethernet/freescale/enetc/enetc_ierb.h
@@ -0,0 +1,20 @@
+/* SPDX-License-Identifier: (GPL-2.0+ OR BSD-3-Clause) */
+/* Copyright 2021 NXP */
+
+#include <linux/pci.h>
+#include <linux/platform_device.h>
+
+#if IS_ENABLED(CONFIG_FSL_ENETC_IERB)
+
+int enetc_ierb_register_pf(struct platform_device *pdev,
+			   struct pci_dev *pf_pdev);
+
+#else
+
+static inline int enetc_ierb_register_pf(struct platform_device *pdev,
+					 struct pci_dev *pf_pdev)
+{
+	return -EOPNOTSUPP;
+}
+
+#endif
diff --git a/drivers/net/ethernet/freescale/enetc/enetc_mdio.c b/drivers/net/ethernet/freescale/enetc/enetc_mdio.c
new file mode 100644
index 000000000000..998aaa394e9c
--- /dev/null
+++ b/drivers/net/ethernet/freescale/enetc/enetc_mdio.c
@@ -0,0 +1,236 @@
+// SPDX-License-Identifier: (GPL-2.0+ OR BSD-3-Clause)
+/* Copyright 2019 NXP */
+
+#include <linux/fsl/enetc_mdio.h>
+#include <linux/mdio.h>
+#include <linux/of_mdio.h>
+#include <linux/iopoll.h>
+#include <linux/of.h>
+
+#include "enetc_pf.h"
+
+#define	ENETC_MDIO_CFG	0x0	/* MDIO configuration and status */
+#define	ENETC_MDIO_CTL	0x4	/* MDIO control */
+#define	ENETC_MDIO_DATA	0x8	/* MDIO data */
+#define	ENETC_MDIO_ADDR	0xc	/* MDIO address */
+
+#define MDIO_CFG_CLKDIV(x)	((((x) >> 1) & 0xff) << 8)
+#define MDIO_CFG_BSY		BIT(0)
+#define MDIO_CFG_RD_ER		BIT(1)
+#define MDIO_CFG_HOLD(x)	(((x) << 2) & GENMASK(4, 2))
+#define MDIO_CFG_ENC45		BIT(6)
+ /* external MDIO only - driven on neg MDC edge */
+#define MDIO_CFG_NEG		BIT(23)
+
+#define ENETC_EMDIO_CFG \
+	(MDIO_CFG_HOLD(2) | \
+	 MDIO_CFG_CLKDIV(258) | \
+	 MDIO_CFG_NEG)
+
+#define MDIO_CTL_DEV_ADDR(x)	((x) & 0x1f)
+#define MDIO_CTL_PORT_ADDR(x)	(((x) & 0x1f) << 5)
+#define MDIO_CTL_READ		BIT(15)
+
+static inline u32 enetc_mdio_rd(struct enetc_mdio_priv *mdio_priv, int off)
+{
+	return enetc_port_rd_mdio(mdio_priv->hw, mdio_priv->mdio_base + off);
+}
+
+static inline void enetc_mdio_wr(struct enetc_mdio_priv *mdio_priv, int off,
+				 u32 val)
+{
+	enetc_port_wr_mdio(mdio_priv->hw, mdio_priv->mdio_base + off, val);
+}
+
+static bool enetc_mdio_is_busy(struct enetc_mdio_priv *mdio_priv)
+{
+	return enetc_mdio_rd(mdio_priv, ENETC_MDIO_CFG) & MDIO_CFG_BSY;
+}
+
+static int enetc_mdio_wait_complete(struct enetc_mdio_priv *mdio_priv)
+{
+	bool is_busy;
+
+	return readx_poll_timeout(enetc_mdio_is_busy, mdio_priv,
+				  is_busy, !is_busy, 10, 10 * 1000);
+}
+
+int enetc_mdio_write_c22(struct mii_bus *bus, int phy_id, int regnum,
+			 u16 value)
+{
+	struct enetc_mdio_priv *mdio_priv = bus->priv;
+	u32 mdio_ctl, mdio_cfg;
+	u16 dev_addr;
+	int ret;
+
+	mdio_cfg = ENETC_EMDIO_CFG;
+	dev_addr = regnum & 0x1f;
+	mdio_cfg &= ~MDIO_CFG_ENC45;
+
+	enetc_mdio_wr(mdio_priv, ENETC_MDIO_CFG, mdio_cfg);
+
+	ret = enetc_mdio_wait_complete(mdio_priv);
+	if (ret)
+		return ret;
+
+	/* set port and dev addr */
+	mdio_ctl = MDIO_CTL_PORT_ADDR(phy_id) | MDIO_CTL_DEV_ADDR(dev_addr);
+	enetc_mdio_wr(mdio_priv, ENETC_MDIO_CTL, mdio_ctl);
+
+	/* write the value */
+	enetc_mdio_wr(mdio_priv, ENETC_MDIO_DATA, value);
+
+	ret = enetc_mdio_wait_complete(mdio_priv);
+	if (ret)
+		return ret;
+
+	return 0;
+}
+EXPORT_SYMBOL_GPL(enetc_mdio_write_c22);
+
+int enetc_mdio_write_c45(struct mii_bus *bus, int phy_id, int dev_addr,
+			 int regnum, u16 value)
+{
+	struct enetc_mdio_priv *mdio_priv = bus->priv;
+	u32 mdio_ctl, mdio_cfg;
+	int ret;
+
+	mdio_cfg = ENETC_EMDIO_CFG;
+	mdio_cfg |= MDIO_CFG_ENC45;
+
+	enetc_mdio_wr(mdio_priv, ENETC_MDIO_CFG, mdio_cfg);
+
+	ret = enetc_mdio_wait_complete(mdio_priv);
+	if (ret)
+		return ret;
+
+	/* set port and dev addr */
+	mdio_ctl = MDIO_CTL_PORT_ADDR(phy_id) | MDIO_CTL_DEV_ADDR(dev_addr);
+	enetc_mdio_wr(mdio_priv, ENETC_MDIO_CTL, mdio_ctl);
+
+	/* set the register address */
+	enetc_mdio_wr(mdio_priv, ENETC_MDIO_ADDR, regnum & 0xffff);
+
+	ret = enetc_mdio_wait_complete(mdio_priv);
+	if (ret)
+		return ret;
+
+	/* write the value */
+	enetc_mdio_wr(mdio_priv, ENETC_MDIO_DATA, value);
+
+	ret = enetc_mdio_wait_complete(mdio_priv);
+	if (ret)
+		return ret;
+
+	return 0;
+}
+EXPORT_SYMBOL_GPL(enetc_mdio_write_c45);
+
+int enetc_mdio_read_c22(struct mii_bus *bus, int phy_id, int regnum)
+{
+	struct enetc_mdio_priv *mdio_priv = bus->priv;
+	u32 mdio_ctl, mdio_cfg;
+	u16 dev_addr, value;
+	int ret;
+
+	mdio_cfg = ENETC_EMDIO_CFG;
+	dev_addr = regnum & 0x1f;
+	mdio_cfg &= ~MDIO_CFG_ENC45;
+
+	enetc_mdio_wr(mdio_priv, ENETC_MDIO_CFG, mdio_cfg);
+
+	ret = enetc_mdio_wait_complete(mdio_priv);
+	if (ret)
+		return ret;
+
+	/* set port and device addr */
+	mdio_ctl = MDIO_CTL_PORT_ADDR(phy_id) | MDIO_CTL_DEV_ADDR(dev_addr);
+	enetc_mdio_wr(mdio_priv, ENETC_MDIO_CTL, mdio_ctl);
+
+	/* initiate the read */
+	enetc_mdio_wr(mdio_priv, ENETC_MDIO_CTL, mdio_ctl | MDIO_CTL_READ);
+
+	ret = enetc_mdio_wait_complete(mdio_priv);
+	if (ret)
+		return ret;
+
+	/* return all Fs if nothing was there */
+	if (enetc_mdio_rd(mdio_priv, ENETC_MDIO_CFG) & MDIO_CFG_RD_ER) {
+		dev_dbg(&bus->dev,
+			"Error while reading PHY%d reg at %d.%d\n",
+			phy_id, dev_addr, regnum);
+		return 0xffff;
+	}
+
+	value = enetc_mdio_rd(mdio_priv, ENETC_MDIO_DATA) & 0xffff;
+
+	return value;
+}
+EXPORT_SYMBOL_GPL(enetc_mdio_read_c22);
+
+int enetc_mdio_read_c45(struct mii_bus *bus, int phy_id, int dev_addr,
+			int regnum)
+{
+	struct enetc_mdio_priv *mdio_priv = bus->priv;
+	u32 mdio_ctl, mdio_cfg;
+	u16 value;
+	int ret;
+
+	mdio_cfg = ENETC_EMDIO_CFG;
+	mdio_cfg |= MDIO_CFG_ENC45;
+
+	enetc_mdio_wr(mdio_priv, ENETC_MDIO_CFG, mdio_cfg);
+
+	ret = enetc_mdio_wait_complete(mdio_priv);
+	if (ret)
+		return ret;
+
+	/* set port and device addr */
+	mdio_ctl = MDIO_CTL_PORT_ADDR(phy_id) | MDIO_CTL_DEV_ADDR(dev_addr);
+	enetc_mdio_wr(mdio_priv, ENETC_MDIO_CTL, mdio_ctl);
+
+	/* set the register address */
+	enetc_mdio_wr(mdio_priv, ENETC_MDIO_ADDR, regnum & 0xffff);
+
+	ret = enetc_mdio_wait_complete(mdio_priv);
+	if (ret)
+		return ret;
+
+	/* initiate the read */
+	enetc_mdio_wr(mdio_priv, ENETC_MDIO_CTL, mdio_ctl | MDIO_CTL_READ);
+
+	ret = enetc_mdio_wait_complete(mdio_priv);
+	if (ret)
+		return ret;
+
+	/* return all Fs if nothing was there */
+	if (enetc_mdio_rd(mdio_priv, ENETC_MDIO_CFG) & MDIO_CFG_RD_ER) {
+		dev_dbg(&bus->dev,
+			"Error while reading PHY%d reg at %d.%d\n",
+			phy_id, dev_addr, regnum);
+		return 0xffff;
+	}
+
+	value = enetc_mdio_rd(mdio_priv, ENETC_MDIO_DATA) & 0xffff;
+
+	return value;
+}
+EXPORT_SYMBOL_GPL(enetc_mdio_read_c45);
+
+struct enetc_hw *enetc_hw_alloc(struct device *dev, void __iomem *port_regs)
+{
+	struct enetc_hw *hw;
+
+	hw = devm_kzalloc(dev, sizeof(*hw), GFP_KERNEL);
+	if (!hw)
+		return ERR_PTR(-ENOMEM);
+
+	hw->port = port_regs;
+
+	return hw;
+}
+EXPORT_SYMBOL_GPL(enetc_hw_alloc);
+
+/* Lock for MDIO access errata on LS1028A */
+DEFINE_RWLOCK(enetc_mdio_lock);
+EXPORT_SYMBOL_GPL(enetc_mdio_lock);
diff --git a/drivers/net/ethernet/freescale/enetc/enetc_msg.c b/drivers/net/ethernet/freescale/enetc/enetc_msg.c
new file mode 100644
index 000000000000..40d22ebe9224
--- /dev/null
+++ b/drivers/net/ethernet/freescale/enetc/enetc_msg.c
@@ -0,0 +1,164 @@
+// SPDX-License-Identifier: (GPL-2.0+ OR BSD-3-Clause)
+/* Copyright 2017-2019 NXP */
+
+#include "enetc_pf.h"
+
+static void enetc_msg_disable_mr_int(struct enetc_hw *hw)
+{
+	u32 psiier = enetc_rd(hw, ENETC_PSIIER);
+	/* disable MR int source(s) */
+	enetc_wr(hw, ENETC_PSIIER, psiier & ~ENETC_PSIIER_MR_MASK);
+}
+
+static void enetc_msg_enable_mr_int(struct enetc_hw *hw)
+{
+	u32 psiier = enetc_rd(hw, ENETC_PSIIER);
+
+	enetc_wr(hw, ENETC_PSIIER, psiier | ENETC_PSIIER_MR_MASK);
+}
+
+static irqreturn_t enetc_msg_psi_msix(int irq, void *data)
+{
+	struct enetc_si *si = (struct enetc_si *)data;
+	struct enetc_pf *pf = enetc_si_priv(si);
+
+	enetc_msg_disable_mr_int(&si->hw);
+	schedule_work(&pf->msg_task);
+
+	return IRQ_HANDLED;
+}
+
+static void enetc_msg_task(struct work_struct *work)
+{
+	struct enetc_pf *pf = container_of(work, struct enetc_pf, msg_task);
+	struct enetc_hw *hw = &pf->si->hw;
+	unsigned long mr_mask;
+	int i;
+
+	for (;;) {
+		mr_mask = enetc_rd(hw, ENETC_PSIMSGRR) & ENETC_PSIMSGRR_MR_MASK;
+		if (!mr_mask) {
+			/* re-arm MR interrupts, w1c the IDR reg */
+			enetc_wr(hw, ENETC_PSIIDR, ENETC_PSIIER_MR_MASK);
+			enetc_msg_enable_mr_int(hw);
+			return;
+		}
+
+		for (i = 0; i < pf->num_vfs; i++) {
+			u32 psimsgrr;
+			u16 msg_code;
+
+			if (!(ENETC_PSIMSGRR_MR(i) & mr_mask))
+				continue;
+
+			enetc_msg_handle_rxmsg(pf, i, &msg_code);
+
+			psimsgrr = ENETC_SIMSGSR_SET_MC(msg_code);
+			psimsgrr |= ENETC_PSIMSGRR_MR(i); /* w1c */
+			enetc_wr(hw, ENETC_PSIMSGRR, psimsgrr);
+		}
+	}
+}
+
+/* Init */
+static int enetc_msg_alloc_mbx(struct enetc_si *si, int idx)
+{
+	struct enetc_pf *pf = enetc_si_priv(si);
+	struct device *dev = &si->pdev->dev;
+	struct enetc_hw *hw = &si->hw;
+	struct enetc_msg_swbd *msg;
+	u32 val;
+
+	msg = &pf->rxmsg[idx];
+	/* allocate and set receive buffer */
+	msg->size = ENETC_DEFAULT_MSG_SIZE;
+
+	msg->vaddr = dma_alloc_coherent(dev, msg->size, &msg->dma,
+					GFP_KERNEL);
+	if (!msg->vaddr) {
+		dev_err(dev, "msg: fail to alloc dma buffer of size: %d\n",
+			msg->size);
+		return -ENOMEM;
+	}
+
+	/* set multiple of 32 bytes */
+	val = lower_32_bits(msg->dma);
+	enetc_wr(hw, ENETC_PSIVMSGRCVAR0(idx), val);
+	val = upper_32_bits(msg->dma);
+	enetc_wr(hw, ENETC_PSIVMSGRCVAR1(idx), val);
+
+	return 0;
+}
+
+static void enetc_msg_free_mbx(struct enetc_si *si, int idx)
+{
+	struct enetc_pf *pf = enetc_si_priv(si);
+	struct enetc_hw *hw = &si->hw;
+	struct enetc_msg_swbd *msg;
+
+	msg = &pf->rxmsg[idx];
+	dma_free_coherent(&si->pdev->dev, msg->size, msg->vaddr, msg->dma);
+	memset(msg, 0, sizeof(*msg));
+
+	enetc_wr(hw, ENETC_PSIVMSGRCVAR0(idx), 0);
+	enetc_wr(hw, ENETC_PSIVMSGRCVAR1(idx), 0);
+}
+
+int enetc_msg_psi_init(struct enetc_pf *pf)
+{
+	struct enetc_si *si = pf->si;
+	int vector, i, err;
+
+	/* register message passing interrupt handler */
+	snprintf(pf->msg_int_name, sizeof(pf->msg_int_name), "%s-vfmsg",
+		 si->ndev->name);
+	vector = pci_irq_vector(si->pdev, ENETC_SI_INT_IDX);
+	err = request_irq(vector, enetc_msg_psi_msix, 0, pf->msg_int_name, si);
+	if (err) {
+		dev_err(&si->pdev->dev,
+			"PSI messaging: request_irq() failed!\n");
+		return err;
+	}
+
+	/* set one IRQ entry for PSI message receive notification (SI int) */
+	enetc_wr(&si->hw, ENETC_SIMSIVR, ENETC_SI_INT_IDX);
+
+	/* initialize PSI mailbox */
+	INIT_WORK(&pf->msg_task, enetc_msg_task);
+
+	for (i = 0; i < pf->num_vfs; i++) {
+		err = enetc_msg_alloc_mbx(si, i);
+		if (err)
+			goto err_init_mbx;
+	}
+
+	/* enable MR interrupts */
+	enetc_msg_enable_mr_int(&si->hw);
+
+	return 0;
+
+err_init_mbx:
+	for (i--; i >= 0; i--)
+		enetc_msg_free_mbx(si, i);
+
+	free_irq(vector, si);
+
+	return err;
+}
+
+void enetc_msg_psi_free(struct enetc_pf *pf)
+{
+	struct enetc_si *si = pf->si;
+	int i;
+
+	cancel_work_sync(&pf->msg_task);
+
+	/* disable MR interrupts */
+	enetc_msg_disable_mr_int(&si->hw);
+
+	for (i = 0; i < pf->num_vfs; i++)
+		enetc_msg_free_mbx(si, i);
+
+	/* de-register message passing interrupt handler */
+	free_irq(pci_irq_vector(si->pdev, ENETC_SI_INT_IDX), si);
+}
diff --git a/drivers/net/ethernet/freescale/enetc/enetc_pci_mdio.c b/drivers/net/ethernet/freescale/enetc/enetc_pci_mdio.c
new file mode 100644
index 000000000000..e108cac8288d
--- /dev/null
+++ b/drivers/net/ethernet/freescale/enetc/enetc_pci_mdio.c
@@ -0,0 +1,143 @@
+// SPDX-License-Identifier: (GPL-2.0+ OR BSD-3-Clause)
+/* Copyright 2019 NXP */
+#include <linux/fsl/enetc_mdio.h>
+#include <linux/of_mdio.h>
+#include "enetc_pf.h"
+
+#define NETC_EMDIO_VEN_ID	0x1131
+#define NETC_EMDIO_DEV_ID	0xee00
+#define ENETC_MDIO_DEV_ID	0xee01
+#define ENETC_MDIO_DEV_NAME	"FSL PCIe IE Central MDIO"
+#define ENETC_MDIO_BUS_NAME	ENETC_MDIO_DEV_NAME " Bus"
+#define ENETC_MDIO_DRV_NAME	ENETC_MDIO_DEV_NAME " driver"
+
+DEFINE_STATIC_KEY_FALSE(enetc_has_err050089);
+EXPORT_SYMBOL_GPL(enetc_has_err050089);
+
+static void enetc_emdio_enable_err050089(struct pci_dev *pdev)
+{
+	if (pdev->vendor == PCI_VENDOR_ID_FREESCALE &&
+	    pdev->device == ENETC_MDIO_DEV_ID) {
+		static_branch_inc(&enetc_has_err050089);
+		dev_info(&pdev->dev, "Enabled ERR050089 workaround\n");
+	}
+}
+
+static void enetc_emdio_disable_err050089(struct pci_dev *pdev)
+{
+	if (pdev->vendor == PCI_VENDOR_ID_FREESCALE &&
+	    pdev->device == ENETC_MDIO_DEV_ID) {
+		static_branch_dec(&enetc_has_err050089);
+		if (!static_key_enabled(&enetc_has_err050089.key))
+			dev_info(&pdev->dev, "Disabled ERR050089 workaround\n");
+	}
+}
+
+static int enetc_pci_mdio_probe(struct pci_dev *pdev,
+				const struct pci_device_id *ent)
+{
+	struct enetc_mdio_priv *mdio_priv;
+	struct device *dev = &pdev->dev;
+	void __iomem *port_regs;
+	struct enetc_hw *hw;
+	struct mii_bus *bus;
+	int err;
+
+	port_regs = pci_iomap(pdev, 0, 0);
+	if (!port_regs) {
+		dev_err(dev, "iomap failed\n");
+		err = -ENXIO;
+		goto err_ioremap;
+	}
+
+	hw = enetc_hw_alloc(dev, port_regs);
+	if (IS_ERR(hw)) {
+		err = PTR_ERR(hw);
+		goto err_hw_alloc;
+	}
+
+	bus = devm_mdiobus_alloc_size(dev, sizeof(*mdio_priv));
+	if (!bus) {
+		err = -ENOMEM;
+		goto err_mdiobus_alloc;
+	}
+
+	bus->name = ENETC_MDIO_BUS_NAME;
+	bus->read = enetc_mdio_read_c22;
+	bus->write = enetc_mdio_write_c22;
+	bus->read_c45 = enetc_mdio_read_c45;
+	bus->write_c45 = enetc_mdio_write_c45;
+	bus->parent = dev;
+	mdio_priv = bus->priv;
+	mdio_priv->hw = hw;
+	mdio_priv->mdio_base = ENETC_EMDIO_BASE;
+	snprintf(bus->id, MII_BUS_ID_SIZE, "%s", dev_name(dev));
+
+	pcie_flr(pdev);
+	err = pci_enable_device_mem(pdev);
+	if (err) {
+		dev_err(dev, "device enable failed\n");
+		goto err_pci_enable;
+	}
+
+	err = pci_request_region(pdev, 0, KBUILD_MODNAME);
+	if (err) {
+		dev_err(dev, "pci_request_region failed\n");
+		goto err_pci_mem_reg;
+	}
+
+	enetc_emdio_enable_err050089(pdev);
+
+	err = of_mdiobus_register(bus, dev->of_node);
+	if (err)
+		goto err_mdiobus_reg;
+
+	pci_set_drvdata(pdev, bus);
+
+	return 0;
+
+err_mdiobus_reg:
+	enetc_emdio_disable_err050089(pdev);
+	pci_release_region(pdev, 0);
+err_pci_mem_reg:
+	pci_disable_device(pdev);
+err_pci_enable:
+err_mdiobus_alloc:
+err_hw_alloc:
+	iounmap(port_regs);
+err_ioremap:
+	return err;
+}
+
+static void enetc_pci_mdio_remove(struct pci_dev *pdev)
+{
+	struct mii_bus *bus = pci_get_drvdata(pdev);
+	struct enetc_mdio_priv *mdio_priv;
+
+	mdiobus_unregister(bus);
+
+	enetc_emdio_disable_err050089(pdev);
+
+	mdio_priv = bus->priv;
+	iounmap(mdio_priv->hw->port);
+	pci_release_region(pdev, 0);
+	pci_disable_device(pdev);
+}
+
+static const struct pci_device_id enetc_pci_mdio_id_table[] = {
+	{ PCI_DEVICE(PCI_VENDOR_ID_FREESCALE, ENETC_MDIO_DEV_ID) },
+	{ PCI_DEVICE(NETC_EMDIO_VEN_ID, NETC_EMDIO_DEV_ID) },
+	{ 0, } /* End of table. */
+};
+MODULE_DEVICE_TABLE(pci, enetc_pci_mdio_id_table);
+
+static struct pci_driver enetc_pci_mdio_driver = {
+	.name = KBUILD_MODNAME,
+	.id_table = enetc_pci_mdio_id_table,
+	.probe = enetc_pci_mdio_probe,
+	.remove = enetc_pci_mdio_remove,
+};
+module_pci_driver(enetc_pci_mdio_driver);
+
+MODULE_DESCRIPTION(ENETC_MDIO_DRV_NAME);
+MODULE_LICENSE("Dual BSD/GPL");
diff --git a/drivers/net/ethernet/freescale/enetc/enetc_pf.c b/drivers/net/ethernet/freescale/enetc/enetc_pf.c
new file mode 100644
index 000000000000..de0fb272c847
--- /dev/null
+++ b/drivers/net/ethernet/freescale/enetc/enetc_pf.c
@@ -0,0 +1,1112 @@
+// SPDX-License-Identifier: (GPL-2.0+ OR BSD-3-Clause)
+/* Copyright 2017-2019 NXP */
+
+#include <linux/unaligned.h>
+#include <linux/module.h>
+#include <linux/of.h>
+#include <linux/of_platform.h>
+#include <linux/of_net.h>
+#include <linux/pcs-lynx.h>
+#include "enetc_ierb.h"
+#include "enetc_pf_common.h"
+
+#define ENETC_DRV_NAME_STR "ENETC PF driver"
+
+static void enetc_pf_get_primary_mac_addr(struct enetc_hw *hw, int si, u8 *addr)
+{
+	u32 upper = __raw_readl(hw->port + ENETC_PSIPMAR0(si));
+	u16 lower = __raw_readw(hw->port + ENETC_PSIPMAR1(si));
+
+	put_unaligned_le32(upper, addr);
+	put_unaligned_le16(lower, addr + 4);
+}
+
+static void enetc_pf_set_primary_mac_addr(struct enetc_hw *hw, int si,
+					  const u8 *addr)
+{
+	u32 upper = get_unaligned_le32(addr);
+	u16 lower = get_unaligned_le16(addr + 4);
+
+	__raw_writel(upper, hw->port + ENETC_PSIPMAR0(si));
+	__raw_writew(lower, hw->port + ENETC_PSIPMAR1(si));
+}
+
+static struct phylink_pcs *enetc_pf_create_pcs(struct enetc_pf *pf,
+					       struct mii_bus *bus)
+{
+	return lynx_pcs_create_mdiodev(bus, 0);
+}
+
+static void enetc_pf_destroy_pcs(struct phylink_pcs *pcs)
+{
+	lynx_pcs_destroy(pcs);
+}
+
+static void enetc_set_vlan_promisc(struct enetc_hw *hw, char si_map)
+{
+	u32 val = enetc_port_rd(hw, ENETC_PSIPVMR);
+
+	val &= ~ENETC_PSIPVMR_SET_VP(ENETC_VLAN_PROMISC_MAP_ALL);
+	enetc_port_wr(hw, ENETC_PSIPVMR, ENETC_PSIPVMR_SET_VP(si_map) | val);
+}
+
+static void enetc_enable_si_vlan_promisc(struct enetc_pf *pf, int si_idx)
+{
+	pf->vlan_promisc_simap |= BIT(si_idx);
+	enetc_set_vlan_promisc(&pf->si->hw, pf->vlan_promisc_simap);
+}
+
+static void enetc_disable_si_vlan_promisc(struct enetc_pf *pf, int si_idx)
+{
+	pf->vlan_promisc_simap &= ~BIT(si_idx);
+	enetc_set_vlan_promisc(&pf->si->hw, pf->vlan_promisc_simap);
+}
+
+static void enetc_set_isol_vlan(struct enetc_hw *hw, int si, u16 vlan, u8 qos)
+{
+	u32 val = 0;
+
+	if (vlan)
+		val = ENETC_PSIVLAN_EN | ENETC_PSIVLAN_SET_QOS(qos) | vlan;
+
+	enetc_port_wr(hw, ENETC_PSIVLANR(si), val);
+}
+
+static void enetc_add_mac_addr_em_filter(struct enetc_mac_filter *filter,
+					 const unsigned char *addr)
+{
+	/* add exact match addr */
+	ether_addr_copy(filter->mac_addr, addr);
+	filter->mac_addr_cnt++;
+}
+
+static void enetc_clear_mac_ht_flt(struct enetc_si *si, int si_idx, int type)
+{
+	bool err = si->errata & ENETC_ERR_UCMCSWP;
+
+	if (type == UC) {
+		enetc_port_wr(&si->hw, ENETC_PSIUMHFR0(si_idx, err), 0);
+		enetc_port_wr(&si->hw, ENETC_PSIUMHFR1(si_idx), 0);
+	} else { /* MC */
+		enetc_port_wr(&si->hw, ENETC_PSIMMHFR0(si_idx, err), 0);
+		enetc_port_wr(&si->hw, ENETC_PSIMMHFR1(si_idx), 0);
+	}
+}
+
+static void enetc_set_mac_ht_flt(struct enetc_si *si, int si_idx, int type,
+				 unsigned long hash)
+{
+	bool err = si->errata & ENETC_ERR_UCMCSWP;
+
+	if (type == UC) {
+		enetc_port_wr(&si->hw, ENETC_PSIUMHFR0(si_idx, err),
+			      lower_32_bits(hash));
+		enetc_port_wr(&si->hw, ENETC_PSIUMHFR1(si_idx),
+			      upper_32_bits(hash));
+	} else { /* MC */
+		enetc_port_wr(&si->hw, ENETC_PSIMMHFR0(si_idx, err),
+			      lower_32_bits(hash));
+		enetc_port_wr(&si->hw, ENETC_PSIMMHFR1(si_idx),
+			      upper_32_bits(hash));
+	}
+}
+
+static void enetc_sync_mac_filters(struct enetc_pf *pf)
+{
+	struct enetc_mac_filter *f = pf->mac_filter;
+	struct enetc_si *si = pf->si;
+	int i, pos;
+
+	pos = EMETC_MAC_ADDR_FILT_RES;
+
+	for (i = 0; i < MADDR_TYPE; i++, f++) {
+		bool em = (f->mac_addr_cnt == 1) && (i == UC);
+		bool clear = !f->mac_addr_cnt;
+
+		if (clear) {
+			if (i == UC)
+				enetc_clear_mac_flt_entry(si, pos);
+
+			enetc_clear_mac_ht_flt(si, 0, i);
+			continue;
+		}
+
+		/* exact match filter */
+		if (em) {
+			int err;
+
+			enetc_clear_mac_ht_flt(si, 0, UC);
+
+			err = enetc_set_mac_flt_entry(si, pos, f->mac_addr,
+						      BIT(0));
+			if (!err)
+				continue;
+
+			/* fallback to HT filtering */
+			dev_warn(&si->pdev->dev, "fallback to HT filt (%d)\n",
+				 err);
+		}
+
+		/* hash table filter, clear EM filter for UC entries */
+		if (i == UC)
+			enetc_clear_mac_flt_entry(si, pos);
+
+		enetc_set_mac_ht_flt(si, 0, i, *f->mac_hash_table);
+	}
+}
+
+static void enetc_pf_set_rx_mode(struct net_device *ndev)
+{
+	struct enetc_ndev_priv *priv = netdev_priv(ndev);
+	struct enetc_pf *pf = enetc_si_priv(priv->si);
+	struct enetc_hw *hw = &priv->si->hw;
+	bool uprom = false, mprom = false;
+	struct enetc_mac_filter *filter;
+	struct netdev_hw_addr *ha;
+	u32 psipmr = 0;
+	bool em;
+
+	if (ndev->flags & IFF_PROMISC) {
+		/* enable promisc mode for SI0 (PF) */
+		psipmr = ENETC_PSIPMR_SET_UP(0) | ENETC_PSIPMR_SET_MP(0);
+		uprom = true;
+		mprom = true;
+	} else if (ndev->flags & IFF_ALLMULTI) {
+		/* enable multi cast promisc mode for SI0 (PF) */
+		psipmr = ENETC_PSIPMR_SET_MP(0);
+		mprom = true;
+	}
+
+	/* first 2 filter entries belong to PF */
+	if (!uprom) {
+		/* Update unicast filters */
+		filter = &pf->mac_filter[UC];
+		enetc_reset_mac_addr_filter(filter);
+
+		em = (netdev_uc_count(ndev) == 1);
+		netdev_for_each_uc_addr(ha, ndev) {
+			if (em) {
+				enetc_add_mac_addr_em_filter(filter, ha->addr);
+				break;
+			}
+
+			enetc_add_mac_addr_ht_filter(filter, ha->addr);
+		}
+	}
+
+	if (!mprom) {
+		/* Update multicast filters */
+		filter = &pf->mac_filter[MC];
+		enetc_reset_mac_addr_filter(filter);
+
+		netdev_for_each_mc_addr(ha, ndev) {
+			if (!is_multicast_ether_addr(ha->addr))
+				continue;
+
+			enetc_add_mac_addr_ht_filter(filter, ha->addr);
+		}
+	}
+
+	if (!uprom || !mprom)
+		/* update PF entries */
+		enetc_sync_mac_filters(pf);
+
+	psipmr |= enetc_port_rd(hw, ENETC_PSIPMR) &
+		  ~(ENETC_PSIPMR_SET_UP(0) | ENETC_PSIPMR_SET_MP(0));
+	enetc_port_wr(hw, ENETC_PSIPMR, psipmr);
+}
+
+static void enetc_set_loopback(struct net_device *ndev, bool en)
+{
+	struct enetc_ndev_priv *priv = netdev_priv(ndev);
+	struct enetc_si *si = priv->si;
+	u32 reg;
+
+	reg = enetc_port_mac_rd(si, ENETC_PM0_IF_MODE);
+	if (reg & ENETC_PM0_IFM_RG) {
+		/* RGMII mode */
+		reg = (reg & ~ENETC_PM0_IFM_RLP) |
+		      (en ? ENETC_PM0_IFM_RLP : 0);
+		enetc_port_mac_wr(si, ENETC_PM0_IF_MODE, reg);
+	} else {
+		/* assume SGMII mode */
+		reg = enetc_port_mac_rd(si, ENETC_PM0_CMD_CFG);
+		reg = (reg & ~ENETC_PM0_CMD_XGLP) |
+		      (en ? ENETC_PM0_CMD_XGLP : 0);
+		reg = (reg & ~ENETC_PM0_CMD_PHY_TX_EN) |
+		      (en ? ENETC_PM0_CMD_PHY_TX_EN : 0);
+		enetc_port_mac_wr(si, ENETC_PM0_CMD_CFG, reg);
+	}
+}
+
+static int enetc_pf_set_vf_mac(struct net_device *ndev, int vf, u8 *mac)
+{
+	struct enetc_ndev_priv *priv = netdev_priv(ndev);
+	struct enetc_pf *pf = enetc_si_priv(priv->si);
+	struct enetc_vf_state *vf_state;
+
+	if (vf >= pf->total_vfs)
+		return -EINVAL;
+
+	if (!is_valid_ether_addr(mac))
+		return -EADDRNOTAVAIL;
+
+	vf_state = &pf->vf_state[vf];
+	vf_state->flags |= ENETC_VF_FLAG_PF_SET_MAC;
+	enetc_pf_set_primary_mac_addr(&priv->si->hw, vf + 1, mac);
+	return 0;
+}
+
+static int enetc_pf_set_vf_vlan(struct net_device *ndev, int vf, u16 vlan,
+				u8 qos, __be16 proto)
+{
+	struct enetc_ndev_priv *priv = netdev_priv(ndev);
+	struct enetc_pf *pf = enetc_si_priv(priv->si);
+
+	if (priv->si->errata & ENETC_ERR_VLAN_ISOL)
+		return -EOPNOTSUPP;
+
+	if (vf >= pf->total_vfs)
+		return -EINVAL;
+
+	if (proto != htons(ETH_P_8021Q))
+		/* only C-tags supported for now */
+		return -EPROTONOSUPPORT;
+
+	enetc_set_isol_vlan(&priv->si->hw, vf + 1, vlan, qos);
+	return 0;
+}
+
+static int enetc_pf_set_vf_spoofchk(struct net_device *ndev, int vf, bool en)
+{
+	struct enetc_ndev_priv *priv = netdev_priv(ndev);
+	struct enetc_pf *pf = enetc_si_priv(priv->si);
+	u32 cfgr;
+
+	if (vf >= pf->total_vfs)
+		return -EINVAL;
+
+	cfgr = enetc_port_rd(&priv->si->hw, ENETC_PSICFGR0(vf + 1));
+	cfgr = (cfgr & ~ENETC_PSICFGR0_ASE) | (en ? ENETC_PSICFGR0_ASE : 0);
+	enetc_port_wr(&priv->si->hw, ENETC_PSICFGR0(vf + 1), cfgr);
+
+	return 0;
+}
+
+static void enetc_port_assign_rfs_entries(struct enetc_si *si)
+{
+	struct enetc_pf *pf = enetc_si_priv(si);
+	struct enetc_hw *hw = &si->hw;
+	int num_entries, vf_entries, i;
+	u32 val;
+
+	/* split RFS entries between functions */
+	val = enetc_port_rd(hw, ENETC_PRFSCAPR);
+	num_entries = ENETC_PRFSCAPR_GET_NUM_RFS(val);
+	vf_entries = num_entries / (pf->total_vfs + 1);
+
+	for (i = 0; i < pf->total_vfs; i++)
+		enetc_port_wr(hw, ENETC_PSIRFSCFGR(i + 1), vf_entries);
+	enetc_port_wr(hw, ENETC_PSIRFSCFGR(0),
+		      num_entries - vf_entries * pf->total_vfs);
+
+	/* enable RFS on port */
+	enetc_port_wr(hw, ENETC_PRFSMR, ENETC_PRFSMR_RFSE);
+}
+
+static void enetc_port_get_caps(struct enetc_si *si)
+{
+	struct enetc_hw *hw = &si->hw;
+	u32 val;
+
+	val = enetc_port_rd(hw, ENETC_PCAPR0);
+
+	if (val & ENETC_PCAPR0_QBV)
+		si->hw_features |= ENETC_SI_F_QBV;
+
+	if (val & ENETC_PCAPR0_QBU)
+		si->hw_features |= ENETC_SI_F_QBU;
+
+	if (val & ENETC_PCAPR0_PSFP)
+		si->hw_features |= ENETC_SI_F_PSFP;
+}
+
+static void enetc_port_si_configure(struct enetc_si *si)
+{
+	struct enetc_pf *pf = enetc_si_priv(si);
+	struct enetc_hw *hw = &si->hw;
+	int num_rings, i;
+	u32 val;
+
+	enetc_port_get_caps(si);
+
+	val = enetc_port_rd(hw, ENETC_PCAPR0);
+	num_rings = min(ENETC_PCAPR0_RXBDR(val), ENETC_PCAPR0_TXBDR(val));
+
+	val = ENETC_PSICFGR0_SET_TXBDR(ENETC_PF_NUM_RINGS);
+	val |= ENETC_PSICFGR0_SET_RXBDR(ENETC_PF_NUM_RINGS);
+
+	if (unlikely(num_rings < ENETC_PF_NUM_RINGS)) {
+		val = ENETC_PSICFGR0_SET_TXBDR(num_rings);
+		val |= ENETC_PSICFGR0_SET_RXBDR(num_rings);
+
+		dev_warn(&si->pdev->dev, "Found %d rings, expected %d!\n",
+			 num_rings, ENETC_PF_NUM_RINGS);
+
+		num_rings = 0;
+	}
+
+	/* Add default one-time settings for SI0 (PF) */
+	val |= ENETC_PSICFGR0_SIVC(ENETC_VLAN_TYPE_C | ENETC_VLAN_TYPE_S);
+
+	enetc_port_wr(hw, ENETC_PSICFGR0(0), val);
+
+	if (num_rings)
+		num_rings -= ENETC_PF_NUM_RINGS;
+
+	/* Configure the SIs for each available VF */
+	val = ENETC_PSICFGR0_SIVC(ENETC_VLAN_TYPE_C | ENETC_VLAN_TYPE_S);
+	val |= ENETC_PSICFGR0_VTE | ENETC_PSICFGR0_SIVIE;
+
+	if (num_rings) {
+		num_rings /= pf->total_vfs;
+		val |= ENETC_PSICFGR0_SET_TXBDR(num_rings);
+		val |= ENETC_PSICFGR0_SET_RXBDR(num_rings);
+	}
+
+	for (i = 0; i < pf->total_vfs; i++)
+		enetc_port_wr(hw, ENETC_PSICFGR0(i + 1), val);
+
+	/* Port level VLAN settings */
+	val = ENETC_PVCLCTR_OVTPIDL(ENETC_VLAN_TYPE_C | ENETC_VLAN_TYPE_S);
+	enetc_port_wr(hw, ENETC_PVCLCTR, val);
+	/* use outer tag for VLAN filtering */
+	enetc_port_wr(hw, ENETC_PSIVLANFMR, ENETC_PSIVLANFMR_VS);
+}
+
+void enetc_set_ptcmsdur(struct enetc_hw *hw, u32 *max_sdu)
+{
+	int tc;
+
+	for (tc = 0; tc < 8; tc++) {
+		u32 val = ENETC_MAC_MAXFRM_SIZE;
+
+		if (max_sdu[tc])
+			val = max_sdu[tc] + VLAN_ETH_HLEN;
+
+		enetc_port_wr(hw, ENETC_PTCMSDUR(tc), val);
+	}
+}
+
+void enetc_reset_ptcmsdur(struct enetc_hw *hw)
+{
+	int tc;
+
+	for (tc = 0; tc < 8; tc++)
+		enetc_port_wr(hw, ENETC_PTCMSDUR(tc), ENETC_MAC_MAXFRM_SIZE);
+}
+
+static void enetc_configure_port_mac(struct enetc_si *si)
+{
+	struct enetc_hw *hw = &si->hw;
+
+	enetc_port_mac_wr(si, ENETC_PM0_MAXFRM,
+			  ENETC_SET_MAXFRM(ENETC_RX_MAXFRM_SIZE));
+
+	enetc_reset_ptcmsdur(hw);
+
+	enetc_port_mac_wr(si, ENETC_PM0_CMD_CFG, ENETC_PM0_CMD_PHY_TX_EN |
+			  ENETC_PM0_CMD_TXP | ENETC_PM0_PROMISC);
+
+	/* On LS1028A, the MAC RX FIFO defaults to 2, which is too high
+	 * and may lead to RX lock-up under traffic. Set it to 1 instead,
+	 * as recommended by the hardware team.
+	 */
+	enetc_port_mac_wr(si, ENETC_PM0_RX_FIFO, ENETC_PM0_RX_FIFO_VAL);
+}
+
+static void enetc_mac_config(struct enetc_si *si, phy_interface_t phy_mode)
+{
+	u32 val;
+
+	if (phy_interface_mode_is_rgmii(phy_mode)) {
+		val = enetc_port_mac_rd(si, ENETC_PM0_IF_MODE);
+		val &= ~(ENETC_PM0_IFM_EN_AUTO | ENETC_PM0_IFM_IFMODE_MASK);
+		val |= ENETC_PM0_IFM_IFMODE_GMII | ENETC_PM0_IFM_RG;
+		enetc_port_mac_wr(si, ENETC_PM0_IF_MODE, val);
+	}
+
+	if (phy_mode == PHY_INTERFACE_MODE_USXGMII) {
+		val = ENETC_PM0_IFM_FULL_DPX | ENETC_PM0_IFM_IFMODE_XGMII;
+		enetc_port_mac_wr(si, ENETC_PM0_IF_MODE, val);
+	}
+}
+
+static void enetc_mac_enable(struct enetc_si *si, bool en)
+{
+	u32 val = enetc_port_mac_rd(si, ENETC_PM0_CMD_CFG);
+
+	val &= ~(ENETC_PM0_TX_EN | ENETC_PM0_RX_EN);
+	val |= en ? (ENETC_PM0_TX_EN | ENETC_PM0_RX_EN) : 0;
+
+	enetc_port_mac_wr(si, ENETC_PM0_CMD_CFG, val);
+}
+
+static void enetc_configure_port(struct enetc_pf *pf)
+{
+	struct enetc_hw *hw = &pf->si->hw;
+
+	enetc_configure_port_mac(pf->si);
+
+	enetc_port_si_configure(pf->si);
+
+	/* set up hash key */
+	enetc_set_default_rss_key(pf);
+
+	/* split up RFS entries */
+	enetc_port_assign_rfs_entries(pf->si);
+
+	/* enforce VLAN promisc mode for all SIs */
+	pf->vlan_promisc_simap = ENETC_VLAN_PROMISC_MAP_ALL;
+	enetc_set_vlan_promisc(hw, pf->vlan_promisc_simap);
+
+	enetc_port_wr(hw, ENETC_PSIPMR, 0);
+
+	/* enable port */
+	enetc_port_wr(hw, ENETC_PMR, ENETC_PMR_EN);
+}
+
+/* Messaging */
+static u16 enetc_msg_pf_set_vf_primary_mac_addr(struct enetc_pf *pf,
+						int vf_id)
+{
+	struct enetc_vf_state *vf_state = &pf->vf_state[vf_id];
+	struct enetc_msg_swbd *msg = &pf->rxmsg[vf_id];
+	struct enetc_msg_cmd_set_primary_mac *cmd;
+	struct device *dev = &pf->si->pdev->dev;
+	u16 cmd_id;
+	char *addr;
+
+	cmd = (struct enetc_msg_cmd_set_primary_mac *)msg->vaddr;
+	cmd_id = cmd->header.id;
+	if (cmd_id != ENETC_MSG_CMD_MNG_ADD)
+		return ENETC_MSG_CMD_STATUS_FAIL;
+
+	addr = cmd->mac.sa_data;
+	if (vf_state->flags & ENETC_VF_FLAG_PF_SET_MAC)
+		dev_warn(dev, "Attempt to override PF set mac addr for VF%d\n",
+			 vf_id);
+	else
+		enetc_pf_set_primary_mac_addr(&pf->si->hw, vf_id + 1, addr);
+
+	return ENETC_MSG_CMD_STATUS_OK;
+}
+
+void enetc_msg_handle_rxmsg(struct enetc_pf *pf, int vf_id, u16 *status)
+{
+	struct enetc_msg_swbd *msg = &pf->rxmsg[vf_id];
+	struct device *dev = &pf->si->pdev->dev;
+	struct enetc_msg_cmd_header *cmd_hdr;
+	u16 cmd_type;
+
+	*status = ENETC_MSG_CMD_STATUS_OK;
+	cmd_hdr = (struct enetc_msg_cmd_header *)msg->vaddr;
+	cmd_type = cmd_hdr->type;
+
+	switch (cmd_type) {
+	case ENETC_MSG_CMD_MNG_MAC:
+		*status = enetc_msg_pf_set_vf_primary_mac_addr(pf, vf_id);
+		break;
+	default:
+		dev_err(dev, "command not supported (cmd_type: 0x%x)\n",
+			cmd_type);
+	}
+}
+
+#ifdef CONFIG_PCI_IOV
+static int enetc_sriov_configure(struct pci_dev *pdev, int num_vfs)
+{
+	struct enetc_si *si = pci_get_drvdata(pdev);
+	struct enetc_pf *pf = enetc_si_priv(si);
+	int err;
+
+	if (!num_vfs) {
+		enetc_msg_psi_free(pf);
+		pf->num_vfs = 0;
+		pci_disable_sriov(pdev);
+	} else {
+		pf->num_vfs = num_vfs;
+
+		err = enetc_msg_psi_init(pf);
+		if (err) {
+			dev_err(&pdev->dev, "enetc_msg_psi_init (%d)\n", err);
+			goto err_msg_psi;
+		}
+
+		err = pci_enable_sriov(pdev, num_vfs);
+		if (err) {
+			dev_err(&pdev->dev, "pci_enable_sriov err %d\n", err);
+			goto err_en_sriov;
+		}
+	}
+
+	return num_vfs;
+
+err_en_sriov:
+	enetc_msg_psi_free(pf);
+err_msg_psi:
+	pf->num_vfs = 0;
+
+	return err;
+}
+#else
+#define enetc_sriov_configure(pdev, num_vfs)	(void)0
+#endif
+
+static int enetc_pf_set_features(struct net_device *ndev,
+				 netdev_features_t features)
+{
+	netdev_features_t changed = ndev->features ^ features;
+	struct enetc_ndev_priv *priv = netdev_priv(ndev);
+	int err;
+
+	if (changed & NETIF_F_HW_TC) {
+		err = enetc_set_psfp(ndev, !!(features & NETIF_F_HW_TC));
+		if (err)
+			return err;
+	}
+
+	if (changed & NETIF_F_HW_VLAN_CTAG_FILTER) {
+		struct enetc_pf *pf = enetc_si_priv(priv->si);
+
+		if (!!(features & NETIF_F_HW_VLAN_CTAG_FILTER))
+			enetc_disable_si_vlan_promisc(pf, 0);
+		else
+			enetc_enable_si_vlan_promisc(pf, 0);
+	}
+
+	if (changed & NETIF_F_LOOPBACK)
+		enetc_set_loopback(ndev, !!(features & NETIF_F_LOOPBACK));
+
+	enetc_set_features(ndev, features);
+
+	return 0;
+}
+
+static int enetc_pf_setup_tc(struct net_device *ndev, enum tc_setup_type type,
+			     void *type_data)
+{
+	switch (type) {
+	case TC_QUERY_CAPS:
+		return enetc_qos_query_caps(ndev, type_data);
+	case TC_SETUP_QDISC_MQPRIO:
+		return enetc_setup_tc_mqprio(ndev, type_data);
+	case TC_SETUP_QDISC_TAPRIO:
+		return enetc_setup_tc_taprio(ndev, type_data);
+	case TC_SETUP_QDISC_CBS:
+		return enetc_setup_tc_cbs(ndev, type_data);
+	case TC_SETUP_QDISC_ETF:
+		return enetc_setup_tc_txtime(ndev, type_data);
+	case TC_SETUP_BLOCK:
+		return enetc_setup_tc_psfp(ndev, type_data);
+	default:
+		return -EOPNOTSUPP;
+	}
+}
+
+static const struct net_device_ops enetc_ndev_ops = {
+	.ndo_open		= enetc_open,
+	.ndo_stop		= enetc_close,
+	.ndo_start_xmit		= enetc_xmit,
+	.ndo_get_stats		= enetc_get_stats,
+	.ndo_set_mac_address	= enetc_pf_set_mac_addr,
+	.ndo_set_rx_mode	= enetc_pf_set_rx_mode,
+	.ndo_vlan_rx_add_vid	= enetc_vlan_rx_add_vid,
+	.ndo_vlan_rx_kill_vid	= enetc_vlan_rx_del_vid,
+	.ndo_set_vf_mac		= enetc_pf_set_vf_mac,
+	.ndo_set_vf_vlan	= enetc_pf_set_vf_vlan,
+	.ndo_set_vf_spoofchk	= enetc_pf_set_vf_spoofchk,
+	.ndo_set_features	= enetc_pf_set_features,
+	.ndo_eth_ioctl		= enetc_ioctl,
+	.ndo_setup_tc		= enetc_pf_setup_tc,
+	.ndo_bpf		= enetc_setup_bpf,
+	.ndo_xdp_xmit		= enetc_xdp_xmit,
+	.ndo_hwtstamp_get	= enetc_hwtstamp_get,
+	.ndo_hwtstamp_set	= enetc_hwtstamp_set,
+};
+
+static struct phylink_pcs *
+enetc_pl_mac_select_pcs(struct phylink_config *config, phy_interface_t iface)
+{
+	struct enetc_pf *pf = phylink_to_enetc_pf(config);
+
+	return pf->pcs;
+}
+
+static void enetc_pl_mac_config(struct phylink_config *config,
+				unsigned int mode,
+				const struct phylink_link_state *state)
+{
+	struct enetc_pf *pf = phylink_to_enetc_pf(config);
+
+	enetc_mac_config(pf->si, state->interface);
+}
+
+static void enetc_force_rgmii_mac(struct enetc_si *si, int speed, int duplex)
+{
+	u32 old_val, val;
+
+	old_val = val = enetc_port_mac_rd(si, ENETC_PM0_IF_MODE);
+
+	if (speed == SPEED_1000) {
+		val &= ~ENETC_PM0_IFM_SSP_MASK;
+		val |= ENETC_PM0_IFM_SSP_1000;
+	} else if (speed == SPEED_100) {
+		val &= ~ENETC_PM0_IFM_SSP_MASK;
+		val |= ENETC_PM0_IFM_SSP_100;
+	} else if (speed == SPEED_10) {
+		val &= ~ENETC_PM0_IFM_SSP_MASK;
+		val |= ENETC_PM0_IFM_SSP_10;
+	}
+
+	if (duplex == DUPLEX_FULL)
+		val |= ENETC_PM0_IFM_FULL_DPX;
+	else
+		val &= ~ENETC_PM0_IFM_FULL_DPX;
+
+	if (val == old_val)
+		return;
+
+	enetc_port_mac_wr(si, ENETC_PM0_IF_MODE, val);
+}
+
+static void enetc_pl_mac_link_up(struct phylink_config *config,
+				 struct phy_device *phy, unsigned int mode,
+				 phy_interface_t interface, int speed,
+				 int duplex, bool tx_pause, bool rx_pause)
+{
+	struct enetc_pf *pf = phylink_to_enetc_pf(config);
+	u32 pause_off_thresh = 0, pause_on_thresh = 0;
+	u32 init_quanta = 0, refresh_quanta = 0;
+	struct enetc_hw *hw = &pf->si->hw;
+	struct enetc_si *si = pf->si;
+	struct enetc_ndev_priv *priv;
+	u32 rbmr, cmd_cfg;
+	int idx;
+
+	priv = netdev_priv(pf->si->ndev);
+
+	if (pf->si->hw_features & ENETC_SI_F_QBV)
+		enetc_sched_speed_set(priv, speed);
+
+	if (!phylink_autoneg_inband(mode) &&
+	    phy_interface_mode_is_rgmii(interface))
+		enetc_force_rgmii_mac(si, speed, duplex);
+
+	/* Flow control */
+	for (idx = 0; idx < priv->num_rx_rings; idx++) {
+		rbmr = enetc_rxbdr_rd(hw, idx, ENETC_RBMR);
+
+		if (tx_pause)
+			rbmr |= ENETC_RBMR_CM;
+		else
+			rbmr &= ~ENETC_RBMR_CM;
+
+		enetc_rxbdr_wr(hw, idx, ENETC_RBMR, rbmr);
+	}
+
+	if (tx_pause) {
+		/* When the port first enters congestion, send a PAUSE request
+		 * with the maximum number of quanta. When the port exits
+		 * congestion, it will automatically send a PAUSE frame with
+		 * zero quanta.
+		 */
+		init_quanta = 0xffff;
+
+		/* Also, set up the refresh timer to send follow-up PAUSE
+		 * frames at half the quanta value, in case the congestion
+		 * condition persists.
+		 */
+		refresh_quanta = 0xffff / 2;
+
+		/* Start emitting PAUSE frames when 3 large frames (or more
+		 * smaller frames) have accumulated in the FIFO waiting to be
+		 * DMAed to the RX ring.
+		 */
+		pause_on_thresh = 3 * ENETC_MAC_MAXFRM_SIZE;
+		pause_off_thresh = 1 * ENETC_MAC_MAXFRM_SIZE;
+	}
+
+	enetc_port_mac_wr(si, ENETC_PM0_PAUSE_QUANTA, init_quanta);
+	enetc_port_mac_wr(si, ENETC_PM0_PAUSE_THRESH, refresh_quanta);
+	enetc_port_wr(hw, ENETC_PPAUONTR, pause_on_thresh);
+	enetc_port_wr(hw, ENETC_PPAUOFFTR, pause_off_thresh);
+
+	cmd_cfg = enetc_port_mac_rd(si, ENETC_PM0_CMD_CFG);
+
+	if (rx_pause)
+		cmd_cfg &= ~ENETC_PM0_PAUSE_IGN;
+	else
+		cmd_cfg |= ENETC_PM0_PAUSE_IGN;
+
+	enetc_port_mac_wr(si, ENETC_PM0_CMD_CFG, cmd_cfg);
+
+	enetc_mac_enable(si, true);
+
+	if (si->hw_features & ENETC_SI_F_QBU)
+		enetc_mm_link_state_update(priv, true);
+}
+
+static void enetc_pl_mac_link_down(struct phylink_config *config,
+				   unsigned int mode,
+				   phy_interface_t interface)
+{
+	struct enetc_pf *pf = phylink_to_enetc_pf(config);
+	struct enetc_si *si = pf->si;
+	struct enetc_ndev_priv *priv;
+
+	priv = netdev_priv(si->ndev);
+
+	if (si->hw_features & ENETC_SI_F_QBU)
+		enetc_mm_link_state_update(priv, false);
+
+	enetc_mac_enable(si, false);
+}
+
+static const struct phylink_mac_ops enetc_mac_phylink_ops = {
+	.mac_select_pcs = enetc_pl_mac_select_pcs,
+	.mac_config = enetc_pl_mac_config,
+	.mac_link_up = enetc_pl_mac_link_up,
+	.mac_link_down = enetc_pl_mac_link_down,
+};
+
+/* Initialize the entire shared memory for the flow steering entries
+ * of this port (PF + VFs)
+ */
+static int enetc_init_port_rfs_memory(struct enetc_si *si)
+{
+	struct enetc_cmd_rfse rfse = {0};
+	struct enetc_hw *hw = &si->hw;
+	int num_rfs, i, err = 0;
+	u32 val;
+
+	val = enetc_port_rd(hw, ENETC_PRFSCAPR);
+	num_rfs = ENETC_PRFSCAPR_GET_NUM_RFS(val);
+
+	for (i = 0; i < num_rfs; i++) {
+		err = enetc_set_fs_entry(si, &rfse, i);
+		if (err)
+			break;
+	}
+
+	return err;
+}
+
+static int enetc_init_port_rss_memory(struct enetc_si *si)
+{
+	struct enetc_hw *hw = &si->hw;
+	int num_rss, err;
+	int *rss_table;
+	u32 val;
+
+	val = enetc_port_rd(hw, ENETC_PRSSCAPR);
+	num_rss = ENETC_PRSSCAPR_GET_NUM_RSS(val);
+	if (!num_rss)
+		return 0;
+
+	rss_table = kcalloc(num_rss, sizeof(*rss_table), GFP_KERNEL);
+	if (!rss_table)
+		return -ENOMEM;
+
+	err = enetc_set_rss_table(si, rss_table, num_rss);
+
+	kfree(rss_table);
+
+	return err;
+}
+
+static int enetc_pf_register_with_ierb(struct pci_dev *pdev)
+{
+	struct platform_device *ierb_pdev;
+	struct device_node *ierb_node;
+	int ret;
+
+	ierb_node = of_find_compatible_node(NULL, NULL,
+					    "fsl,ls1028a-enetc-ierb");
+	if (!ierb_node)
+		return -ENODEV;
+
+	if (!of_device_is_available(ierb_node)) {
+		of_node_put(ierb_node);
+		return -ENODEV;
+	}
+
+	ierb_pdev = of_find_device_by_node(ierb_node);
+	of_node_put(ierb_node);
+
+	if (!ierb_pdev)
+		return -EPROBE_DEFER;
+
+	ret = enetc_ierb_register_pf(ierb_pdev, pdev);
+
+	put_device(&ierb_pdev->dev);
+
+	return ret;
+}
+
+static const struct enetc_si_ops enetc_psi_ops = {
+	.get_rss_table = enetc_get_rss_table,
+	.set_rss_table = enetc_set_rss_table,
+};
+
+static struct enetc_si *enetc_psi_create(struct pci_dev *pdev)
+{
+	struct enetc_si *si;
+	int err;
+
+	err = enetc_pci_probe(pdev, KBUILD_MODNAME, sizeof(struct enetc_pf));
+	if (err) {
+		dev_err_probe(&pdev->dev, err, "PCI probing failed\n");
+		goto out;
+	}
+
+	si = pci_get_drvdata(pdev);
+	if (!si->hw.port || !si->hw.global) {
+		err = -ENODEV;
+		dev_err(&pdev->dev, "could not map PF space, probing a VF?\n");
+		goto out_pci_remove;
+	}
+
+	si->revision = enetc_get_ip_revision(&si->hw);
+	si->ops = &enetc_psi_ops;
+	err = enetc_get_driver_data(si);
+	if (err) {
+		dev_err(&pdev->dev, "Could not get PF driver data\n");
+		goto out_pci_remove;
+	}
+
+	err = enetc_setup_cbdr(&pdev->dev, &si->hw, ENETC_CBDR_DEFAULT_SIZE,
+			       &si->cbd_ring);
+	if (err)
+		goto out_pci_remove;
+
+	err = enetc_init_port_rfs_memory(si);
+	if (err) {
+		dev_err(&pdev->dev, "Failed to initialize RFS memory\n");
+		goto out_teardown_cbdr;
+	}
+
+	err = enetc_init_port_rss_memory(si);
+	if (err) {
+		dev_err(&pdev->dev, "Failed to initialize RSS memory\n");
+		goto out_teardown_cbdr;
+	}
+
+	return si;
+
+out_teardown_cbdr:
+	enetc_teardown_cbdr(&si->cbd_ring);
+out_pci_remove:
+	enetc_pci_remove(pdev);
+out:
+	return ERR_PTR(err);
+}
+
+static void enetc_psi_destroy(struct pci_dev *pdev)
+{
+	struct enetc_si *si = pci_get_drvdata(pdev);
+
+	enetc_teardown_cbdr(&si->cbd_ring);
+	enetc_pci_remove(pdev);
+}
+
+static const struct enetc_pf_ops enetc_pf_ops = {
+	.set_si_primary_mac = enetc_pf_set_primary_mac_addr,
+	.get_si_primary_mac = enetc_pf_get_primary_mac_addr,
+	.create_pcs = enetc_pf_create_pcs,
+	.destroy_pcs = enetc_pf_destroy_pcs,
+	.enable_psfp = enetc_psfp_enable,
+};
+
+static int enetc_pf_probe(struct pci_dev *pdev,
+			  const struct pci_device_id *ent)
+{
+	struct device_node *node = pdev->dev.of_node;
+	struct enetc_ndev_priv *priv;
+	struct net_device *ndev;
+	struct enetc_si *si;
+	struct enetc_pf *pf;
+	int err;
+
+	err = enetc_pf_register_with_ierb(pdev);
+	if (err == -EPROBE_DEFER)
+		return err;
+	if (err)
+		dev_warn(&pdev->dev,
+			 "Could not register with IERB driver: %pe, please update the device tree\n",
+			 ERR_PTR(err));
+
+	si = enetc_psi_create(pdev);
+	if (IS_ERR(si)) {
+		err = PTR_ERR(si);
+		goto err_psi_create;
+	}
+
+	pf = enetc_si_priv(si);
+	pf->si = si;
+	pf->ops = &enetc_pf_ops;
+
+	pf->total_vfs = pci_sriov_get_totalvfs(pdev);
+	if (pf->total_vfs) {
+		pf->vf_state = kcalloc(pf->total_vfs, sizeof(struct enetc_vf_state),
+				       GFP_KERNEL);
+		if (!pf->vf_state)
+			goto err_alloc_vf_state;
+	}
+
+	err = enetc_setup_mac_addresses(node, pf);
+	if (err)
+		goto err_setup_mac_addresses;
+
+	enetc_configure_port(pf);
+
+	enetc_get_si_caps(si);
+
+	ndev = alloc_etherdev_mq(sizeof(*priv), ENETC_MAX_NUM_TXQS);
+	if (!ndev) {
+		err = -ENOMEM;
+		dev_err(&pdev->dev, "netdev creation failed\n");
+		goto err_alloc_netdev;
+	}
+
+	enetc_pf_netdev_setup(si, ndev, &enetc_ndev_ops);
+
+	priv = netdev_priv(ndev);
+
+	mutex_init(&priv->mm_lock);
+
+	enetc_init_si_rings_params(priv);
+
+	err = enetc_alloc_si_resources(priv);
+	if (err) {
+		dev_err(&pdev->dev, "SI resource alloc failed\n");
+		goto err_alloc_si_res;
+	}
+
+	err = enetc_configure_si(priv);
+	if (err) {
+		dev_err(&pdev->dev, "Failed to configure SI\n");
+		goto err_config_si;
+	}
+
+	err = enetc_alloc_msix(priv);
+	if (err) {
+		dev_err(&pdev->dev, "MSIX alloc failed\n");
+		goto err_alloc_msix;
+	}
+
+	err = of_get_phy_mode(node, &pf->if_mode);
+	if (err) {
+		dev_err(&pdev->dev, "Failed to read PHY mode\n");
+		goto err_phy_mode;
+	}
+
+	err = enetc_mdiobus_create(pf, node);
+	if (err)
+		goto err_mdiobus_create;
+
+	err = enetc_phylink_create(priv, node, &enetc_mac_phylink_ops);
+	if (err)
+		goto err_phylink_create;
+
+	err = register_netdev(ndev);
+	if (err)
+		goto err_reg_netdev;
+
+	return 0;
+
+err_reg_netdev:
+	enetc_phylink_destroy(priv);
+err_phylink_create:
+	enetc_mdiobus_destroy(pf);
+err_mdiobus_create:
+err_phy_mode:
+	enetc_free_msix(priv);
+err_config_si:
+err_alloc_msix:
+	enetc_free_si_resources(priv);
+err_alloc_si_res:
+	si->ndev = NULL;
+	free_netdev(ndev);
+err_alloc_netdev:
+err_setup_mac_addresses:
+	kfree(pf->vf_state);
+err_alloc_vf_state:
+	enetc_psi_destroy(pdev);
+err_psi_create:
+	return err;
+}
+
+static void enetc_pf_remove(struct pci_dev *pdev)
+{
+	struct enetc_si *si = pci_get_drvdata(pdev);
+	struct enetc_pf *pf = enetc_si_priv(si);
+	struct enetc_ndev_priv *priv;
+
+	priv = netdev_priv(si->ndev);
+
+	if (pf->num_vfs)
+		enetc_sriov_configure(pdev, 0);
+
+	unregister_netdev(si->ndev);
+
+	enetc_phylink_destroy(priv);
+	enetc_mdiobus_destroy(pf);
+
+	enetc_free_msix(priv);
+
+	enetc_free_si_resources(priv);
+
+	free_netdev(si->ndev);
+	kfree(pf->vf_state);
+
+	enetc_psi_destroy(pdev);
+}
+
+static void enetc_fixup_clear_rss_rfs(struct pci_dev *pdev)
+{
+	struct device_node *node = pdev->dev.of_node;
+	struct enetc_si *si;
+
+	/* Only apply quirk for disabled functions. For the ones
+	 * that are enabled, enetc_pf_probe() will apply it.
+	 */
+	if (node && of_device_is_available(node))
+		return;
+
+	si = enetc_psi_create(pdev);
+	if (!IS_ERR(si))
+		enetc_psi_destroy(pdev);
+}
+DECLARE_PCI_FIXUP_FINAL(PCI_VENDOR_ID_FREESCALE, ENETC_DEV_ID_PF,
+			enetc_fixup_clear_rss_rfs);
+
+static const struct pci_device_id enetc_pf_id_table[] = {
+	{ PCI_DEVICE(PCI_VENDOR_ID_FREESCALE, ENETC_DEV_ID_PF) },
+	{ 0, } /* End of table. */
+};
+MODULE_DEVICE_TABLE(pci, enetc_pf_id_table);
+
+static struct pci_driver enetc_pf_driver = {
+	.name = KBUILD_MODNAME,
+	.id_table = enetc_pf_id_table,
+	.probe = enetc_pf_probe,
+	.remove = enetc_pf_remove,
+#ifdef CONFIG_PCI_IOV
+	.sriov_configure = enetc_sriov_configure,
+#endif
+};
+module_pci_driver(enetc_pf_driver);
+
+MODULE_DESCRIPTION(ENETC_DRV_NAME_STR);
+MODULE_LICENSE("Dual BSD/GPL");
diff --git a/drivers/net/ethernet/freescale/enetc/enetc_pf.h b/drivers/net/ethernet/freescale/enetc/enetc_pf.h
new file mode 100644
index 000000000000..ae407e9e9ee7
--- /dev/null
+++ b/drivers/net/ethernet/freescale/enetc/enetc_pf.h
@@ -0,0 +1,73 @@
+/* SPDX-License-Identifier: (GPL-2.0+ OR BSD-3-Clause) */
+/* Copyright 2017-2019 NXP */
+
+#include "enetc.h"
+#include <linux/phylink.h>
+
+#define ENETC_PF_NUM_RINGS	8
+#define ENETC_MAX_NUM_MAC_FLT	((ENETC_MAX_NUM_VFS + 1) * MADDR_TYPE)
+
+#define ENETC_VLAN_HT_SIZE	64
+
+enum enetc_vf_flags {
+	ENETC_VF_FLAG_PF_SET_MAC	= BIT(0),
+};
+
+struct enetc_vf_state {
+	enum enetc_vf_flags flags;
+};
+
+struct enetc_port_caps {
+	u32 half_duplex:1;
+	int num_vsi;
+	int num_msix;
+	int num_rx_bdr;
+	int num_tx_bdr;
+	int mac_filter_num;
+};
+
+struct enetc_pf;
+
+struct enetc_pf_ops {
+	void (*set_si_primary_mac)(struct enetc_hw *hw, int si, const u8 *addr);
+	void (*get_si_primary_mac)(struct enetc_hw *hw, int si, u8 *addr);
+	struct phylink_pcs *(*create_pcs)(struct enetc_pf *pf, struct mii_bus *bus);
+	void (*destroy_pcs)(struct phylink_pcs *pcs);
+	int (*enable_psfp)(struct enetc_ndev_priv *priv);
+};
+
+struct enetc_pf {
+	struct enetc_si *si;
+	int num_vfs; /* number of active VFs, after sriov_init */
+	int total_vfs; /* max number of VFs, set for PF at probe */
+	struct enetc_vf_state *vf_state;
+
+	struct enetc_mac_filter mac_filter[ENETC_MAX_NUM_MAC_FLT];
+
+	struct enetc_msg_swbd rxmsg[ENETC_MAX_NUM_VFS];
+	struct work_struct msg_task;
+	char msg_int_name[ENETC_INT_NAME_MAX];
+
+	char vlan_promisc_simap; /* bitmap of SIs in VLAN promisc mode */
+	DECLARE_BITMAP(vlan_ht_filter, ENETC_VLAN_HT_SIZE);
+	DECLARE_BITMAP(active_vlans, VLAN_N_VID);
+
+	struct mii_bus *mdio; /* saved for cleanup */
+	struct mii_bus *imdio;
+	struct phylink_pcs *pcs;
+
+	phy_interface_t if_mode;
+	struct phylink_config phylink_config;
+
+	struct enetc_port_caps caps;
+	const struct enetc_pf_ops *ops;
+
+	int num_mfe;	/* number of mac address filter table entries */
+};
+
+#define phylink_to_enetc_pf(config) \
+	container_of((config), struct enetc_pf, phylink_config)
+
+int enetc_msg_psi_init(struct enetc_pf *pf);
+void enetc_msg_psi_free(struct enetc_pf *pf);
+void enetc_msg_handle_rxmsg(struct enetc_pf *pf, int mbox_id, u16 *status);
diff --git a/drivers/net/ethernet/freescale/enetc/enetc_pf_common.c b/drivers/net/ethernet/freescale/enetc/enetc_pf_common.c
new file mode 100644
index 000000000000..edf14a95cab7
--- /dev/null
+++ b/drivers/net/ethernet/freescale/enetc/enetc_pf_common.c
@@ -0,0 +1,426 @@
+// SPDX-License-Identifier: (GPL-2.0+ OR BSD-3-Clause)
+/* Copyright 2024 NXP */
+
+#include <linux/fsl/enetc_mdio.h>
+#include <linux/of_mdio.h>
+#include <linux/of_net.h>
+
+#include "enetc_pf_common.h"
+
+static void enetc_set_si_hw_addr(struct enetc_pf *pf, int si,
+				 const u8 *mac_addr)
+{
+	struct enetc_hw *hw = &pf->si->hw;
+
+	pf->ops->set_si_primary_mac(hw, si, mac_addr);
+}
+
+static void enetc_get_si_hw_addr(struct enetc_pf *pf, int si, u8 *mac_addr)
+{
+	struct enetc_hw *hw = &pf->si->hw;
+
+	pf->ops->get_si_primary_mac(hw, si, mac_addr);
+}
+
+int enetc_pf_set_mac_addr(struct net_device *ndev, void *addr)
+{
+	struct enetc_ndev_priv *priv = netdev_priv(ndev);
+	struct enetc_pf *pf = enetc_si_priv(priv->si);
+	struct sockaddr *saddr = addr;
+
+	if (!is_valid_ether_addr(saddr->sa_data))
+		return -EADDRNOTAVAIL;
+
+	eth_hw_addr_set(ndev, saddr->sa_data);
+	enetc_set_si_hw_addr(pf, 0, saddr->sa_data);
+
+	return 0;
+}
+EXPORT_SYMBOL_GPL(enetc_pf_set_mac_addr);
+
+static int enetc_setup_mac_address(struct device_node *np, struct enetc_pf *pf,
+				   int si)
+{
+	struct device *dev = &pf->si->pdev->dev;
+	u8 mac_addr[ETH_ALEN] = { 0 };
+	int err;
+
+	/* (1) try to get the MAC address from the device tree */
+	if (np) {
+		err = of_get_mac_address(np, mac_addr);
+		if (err == -EPROBE_DEFER)
+			return err;
+	}
+
+	/* (2) bootloader supplied MAC address */
+	if (is_zero_ether_addr(mac_addr))
+		enetc_get_si_hw_addr(pf, si, mac_addr);
+
+	/* (3) choose a random one */
+	if (is_zero_ether_addr(mac_addr)) {
+		eth_random_addr(mac_addr);
+		dev_info(dev, "no MAC address specified for SI%d, using %pM\n",
+			 si, mac_addr);
+	}
+
+	enetc_set_si_hw_addr(pf, si, mac_addr);
+
+	return 0;
+}
+
+int enetc_setup_mac_addresses(struct device_node *np, struct enetc_pf *pf)
+{
+	int err, i;
+
+	/* The PF might take its MAC from the device tree */
+	err = enetc_setup_mac_address(np, pf, 0);
+	if (err)
+		return err;
+
+	for (i = 0; i < pf->total_vfs; i++) {
+		err = enetc_setup_mac_address(NULL, pf, i + 1);
+		if (err)
+			return err;
+	}
+
+	return 0;
+}
+EXPORT_SYMBOL_GPL(enetc_setup_mac_addresses);
+
+void enetc_pf_netdev_setup(struct enetc_si *si, struct net_device *ndev,
+			   const struct net_device_ops *ndev_ops)
+{
+	struct enetc_ndev_priv *priv = netdev_priv(ndev);
+	struct enetc_pf *pf = enetc_si_priv(si);
+
+	SET_NETDEV_DEV(ndev, &si->pdev->dev);
+	priv->ndev = ndev;
+	priv->si = si;
+	priv->dev = &si->pdev->dev;
+	si->ndev = ndev;
+
+	priv->msg_enable = (NETIF_MSG_WOL << 1) - 1;
+	priv->sysclk_freq = si->drvdata->sysclk_freq;
+	priv->max_frags = si->drvdata->max_frags;
+	ndev->netdev_ops = ndev_ops;
+	enetc_set_ethtool_ops(ndev);
+	ndev->watchdog_timeo = 5 * HZ;
+	ndev->max_mtu = ENETC_MAX_MTU;
+
+	ndev->hw_features = NETIF_F_SG | NETIF_F_RXCSUM |
+			    NETIF_F_HW_VLAN_CTAG_TX | NETIF_F_HW_VLAN_CTAG_RX |
+			    NETIF_F_HW_VLAN_CTAG_FILTER | NETIF_F_LOOPBACK |
+			    NETIF_F_HW_CSUM | NETIF_F_TSO | NETIF_F_TSO6 |
+			    NETIF_F_GSO_UDP_L4;
+	ndev->features = NETIF_F_HIGHDMA | NETIF_F_SG | NETIF_F_RXCSUM |
+			 NETIF_F_HW_VLAN_CTAG_TX |
+			 NETIF_F_HW_VLAN_CTAG_RX |
+			 NETIF_F_HW_CSUM | NETIF_F_TSO | NETIF_F_TSO6 |
+			 NETIF_F_GSO_UDP_L4;
+	ndev->vlan_features = NETIF_F_SG | NETIF_F_HW_CSUM |
+			      NETIF_F_TSO | NETIF_F_TSO6;
+
+	ndev->priv_flags |= IFF_UNICAST_FLT;
+
+	if (si->drvdata->tx_csum)
+		priv->active_offloads |= ENETC_F_TXCSUM;
+
+	if (si->hw_features & ENETC_SI_F_LSO)
+		priv->active_offloads |= ENETC_F_LSO;
+
+	if (si->num_rss) {
+		ndev->hw_features |= NETIF_F_RXHASH;
+		ndev->features |= NETIF_F_RXHASH;
+	}
+
+	/* TODO: currently, i.MX95 ENETC driver does not support advanced features */
+	if (!is_enetc_rev1(si))
+		goto end;
+
+	ndev->xdp_features = NETDEV_XDP_ACT_BASIC | NETDEV_XDP_ACT_REDIRECT |
+			     NETDEV_XDP_ACT_NDO_XMIT | NETDEV_XDP_ACT_RX_SG |
+			     NETDEV_XDP_ACT_NDO_XMIT_SG;
+
+	if (si->hw_features & ENETC_SI_F_PSFP && pf->ops->enable_psfp &&
+	    !pf->ops->enable_psfp(priv)) {
+		priv->active_offloads |= ENETC_F_QCI;
+		ndev->features |= NETIF_F_HW_TC;
+		ndev->hw_features |= NETIF_F_HW_TC;
+	}
+
+end:
+	/* pick up primary MAC address from SI */
+	enetc_load_primary_mac_addr(&si->hw, ndev);
+}
+EXPORT_SYMBOL_GPL(enetc_pf_netdev_setup);
+
+static int enetc_mdio_probe(struct enetc_pf *pf, struct device_node *np)
+{
+	struct device *dev = &pf->si->pdev->dev;
+	struct enetc_mdio_priv *mdio_priv;
+	struct mii_bus *bus;
+	int err;
+
+	bus = devm_mdiobus_alloc_size(dev, sizeof(*mdio_priv));
+	if (!bus)
+		return -ENOMEM;
+
+	bus->name = "Freescale ENETC MDIO Bus";
+	bus->read = enetc_mdio_read_c22;
+	bus->write = enetc_mdio_write_c22;
+	bus->read_c45 = enetc_mdio_read_c45;
+	bus->write_c45 = enetc_mdio_write_c45;
+	bus->parent = dev;
+	mdio_priv = bus->priv;
+	mdio_priv->hw = &pf->si->hw;
+	mdio_priv->mdio_base = ENETC_EMDIO_BASE;
+	snprintf(bus->id, MII_BUS_ID_SIZE, "%s", dev_name(dev));
+
+	err = of_mdiobus_register(bus, np);
+	if (err)
+		return dev_err_probe(dev, err, "cannot register MDIO bus\n");
+
+	pf->mdio = bus;
+
+	return 0;
+}
+
+static void enetc_mdio_remove(struct enetc_pf *pf)
+{
+	if (pf->mdio)
+		mdiobus_unregister(pf->mdio);
+}
+
+static int enetc_imdio_create(struct enetc_pf *pf)
+{
+	struct device *dev = &pf->si->pdev->dev;
+	struct enetc_mdio_priv *mdio_priv;
+	struct phylink_pcs *phylink_pcs;
+	struct mii_bus *bus;
+	int err;
+
+	if (!pf->ops->create_pcs) {
+		dev_err(dev, "Creating PCS is not supported\n");
+
+		return -EOPNOTSUPP;
+	}
+
+	bus = mdiobus_alloc_size(sizeof(*mdio_priv));
+	if (!bus)
+		return -ENOMEM;
+
+	bus->name = "Freescale ENETC internal MDIO Bus";
+	bus->read = enetc_mdio_read_c22;
+	bus->write = enetc_mdio_write_c22;
+	bus->read_c45 = enetc_mdio_read_c45;
+	bus->write_c45 = enetc_mdio_write_c45;
+	bus->parent = dev;
+	bus->phy_mask = ~0;
+	mdio_priv = bus->priv;
+	mdio_priv->hw = &pf->si->hw;
+	mdio_priv->mdio_base = ENETC_PM_IMDIO_BASE;
+	snprintf(bus->id, MII_BUS_ID_SIZE, "%s-imdio", dev_name(dev));
+
+	err = mdiobus_register(bus);
+	if (err) {
+		dev_err(dev, "cannot register internal MDIO bus (%d)\n", err);
+		goto free_mdio_bus;
+	}
+
+	phylink_pcs = pf->ops->create_pcs(pf, bus);
+	if (IS_ERR(phylink_pcs)) {
+		err = PTR_ERR(phylink_pcs);
+		dev_err(dev, "cannot create lynx pcs (%d)\n", err);
+		goto unregister_mdiobus;
+	}
+
+	pf->imdio = bus;
+	pf->pcs = phylink_pcs;
+
+	return 0;
+
+unregister_mdiobus:
+	mdiobus_unregister(bus);
+free_mdio_bus:
+	mdiobus_free(bus);
+	return err;
+}
+
+static void enetc_imdio_remove(struct enetc_pf *pf)
+{
+	if (pf->pcs && pf->ops->destroy_pcs)
+		pf->ops->destroy_pcs(pf->pcs);
+
+	if (pf->imdio) {
+		mdiobus_unregister(pf->imdio);
+		mdiobus_free(pf->imdio);
+	}
+}
+
+static bool enetc_port_has_pcs(struct enetc_pf *pf)
+{
+	return (pf->if_mode == PHY_INTERFACE_MODE_SGMII ||
+		pf->if_mode == PHY_INTERFACE_MODE_1000BASEX ||
+		pf->if_mode == PHY_INTERFACE_MODE_2500BASEX ||
+		pf->if_mode == PHY_INTERFACE_MODE_USXGMII);
+}
+
+int enetc_mdiobus_create(struct enetc_pf *pf, struct device_node *node)
+{
+	struct device_node *mdio_np;
+	int err;
+
+	mdio_np = of_get_child_by_name(node, "mdio");
+	if (mdio_np) {
+		err = enetc_mdio_probe(pf, mdio_np);
+
+		of_node_put(mdio_np);
+		if (err)
+			return err;
+	}
+
+	if (enetc_port_has_pcs(pf)) {
+		err = enetc_imdio_create(pf);
+		if (err) {
+			enetc_mdio_remove(pf);
+			return err;
+		}
+	}
+
+	return 0;
+}
+EXPORT_SYMBOL_GPL(enetc_mdiobus_create);
+
+void enetc_mdiobus_destroy(struct enetc_pf *pf)
+{
+	enetc_mdio_remove(pf);
+	enetc_imdio_remove(pf);
+}
+EXPORT_SYMBOL_GPL(enetc_mdiobus_destroy);
+
+int enetc_phylink_create(struct enetc_ndev_priv *priv, struct device_node *node,
+			 const struct phylink_mac_ops *ops)
+{
+	struct enetc_pf *pf = enetc_si_priv(priv->si);
+	struct phylink *phylink;
+	int err;
+
+	pf->phylink_config.dev = &priv->ndev->dev;
+	pf->phylink_config.type = PHYLINK_NETDEV;
+	pf->phylink_config.mac_capabilities = MAC_ASYM_PAUSE | MAC_SYM_PAUSE |
+		MAC_10 | MAC_100 | MAC_1000 | MAC_2500FD;
+
+	__set_bit(PHY_INTERFACE_MODE_INTERNAL,
+		  pf->phylink_config.supported_interfaces);
+	__set_bit(PHY_INTERFACE_MODE_SGMII,
+		  pf->phylink_config.supported_interfaces);
+	__set_bit(PHY_INTERFACE_MODE_1000BASEX,
+		  pf->phylink_config.supported_interfaces);
+	__set_bit(PHY_INTERFACE_MODE_2500BASEX,
+		  pf->phylink_config.supported_interfaces);
+	__set_bit(PHY_INTERFACE_MODE_USXGMII,
+		  pf->phylink_config.supported_interfaces);
+	phy_interface_set_rgmii(pf->phylink_config.supported_interfaces);
+
+	phylink = phylink_create(&pf->phylink_config, of_fwnode_handle(node),
+				 pf->if_mode, ops);
+	if (IS_ERR(phylink)) {
+		err = PTR_ERR(phylink);
+		return err;
+	}
+
+	priv->phylink = phylink;
+
+	return 0;
+}
+EXPORT_SYMBOL_GPL(enetc_phylink_create);
+
+void enetc_phylink_destroy(struct enetc_ndev_priv *priv)
+{
+	phylink_destroy(priv->phylink);
+}
+EXPORT_SYMBOL_GPL(enetc_phylink_destroy);
+
+void enetc_set_default_rss_key(struct enetc_pf *pf)
+{
+	u8 hash_key[ENETC_RSSHASH_KEY_SIZE] = {0};
+
+	/* set up hash key */
+	get_random_bytes(hash_key, ENETC_RSSHASH_KEY_SIZE);
+	enetc_set_rss_key(pf->si, hash_key);
+}
+EXPORT_SYMBOL_GPL(enetc_set_default_rss_key);
+
+static int enetc_vid_hash_idx(unsigned int vid)
+{
+	int res = 0;
+	int i;
+
+	for (i = 0; i < 6; i++)
+		res |= (hweight8(vid & (BIT(i) | BIT(i + 6))) & 0x1) << i;
+
+	return res;
+}
+
+static void enetc_refresh_vlan_ht_filter(struct enetc_pf *pf)
+{
+	int i;
+
+	bitmap_zero(pf->vlan_ht_filter, ENETC_VLAN_HT_SIZE);
+	for_each_set_bit(i, pf->active_vlans, VLAN_N_VID) {
+		int hidx = enetc_vid_hash_idx(i);
+
+		__set_bit(hidx, pf->vlan_ht_filter);
+	}
+}
+
+static void enetc_set_si_vlan_ht_filter(struct enetc_si *si,
+					int si_id, u64 hash)
+{
+	struct enetc_hw *hw = &si->hw;
+	int high_reg_off, low_reg_off;
+
+	if (is_enetc_rev1(si)) {
+		low_reg_off = ENETC_PSIVHFR0(si_id);
+		high_reg_off = ENETC_PSIVHFR1(si_id);
+	} else {
+		low_reg_off = ENETC4_PSIVHFR0(si_id);
+		high_reg_off = ENETC4_PSIVHFR1(si_id);
+	}
+
+	enetc_port_wr(hw, low_reg_off, lower_32_bits(hash));
+	enetc_port_wr(hw, high_reg_off, upper_32_bits(hash));
+}
+
+int enetc_vlan_rx_add_vid(struct net_device *ndev, __be16 prot, u16 vid)
+{
+	struct enetc_ndev_priv *priv = netdev_priv(ndev);
+	struct enetc_pf *pf = enetc_si_priv(priv->si);
+	int idx;
+
+	__set_bit(vid, pf->active_vlans);
+
+	idx = enetc_vid_hash_idx(vid);
+	if (!__test_and_set_bit(idx, pf->vlan_ht_filter))
+		enetc_set_si_vlan_ht_filter(pf->si, 0, *pf->vlan_ht_filter);
+
+	return 0;
+}
+EXPORT_SYMBOL_GPL(enetc_vlan_rx_add_vid);
+
+int enetc_vlan_rx_del_vid(struct net_device *ndev, __be16 prot, u16 vid)
+{
+	struct enetc_ndev_priv *priv = netdev_priv(ndev);
+	struct enetc_pf *pf = enetc_si_priv(priv->si);
+
+	if (__test_and_clear_bit(vid, pf->active_vlans)) {
+		enetc_refresh_vlan_ht_filter(pf);
+		enetc_set_si_vlan_ht_filter(pf->si, 0, *pf->vlan_ht_filter);
+	}
+
+	return 0;
+}
+EXPORT_SYMBOL_GPL(enetc_vlan_rx_del_vid);
+
+MODULE_DESCRIPTION("NXP ENETC PF common functionality driver");
+MODULE_LICENSE("Dual BSD/GPL");
diff --git a/drivers/net/ethernet/freescale/enetc/enetc_pf_common.h b/drivers/net/ethernet/freescale/enetc/enetc_pf_common.h
new file mode 100644
index 000000000000..96d4840a3107
--- /dev/null
+++ b/drivers/net/ethernet/freescale/enetc/enetc_pf_common.h
@@ -0,0 +1,22 @@
+/* SPDX-License-Identifier: (GPL-2.0+ OR BSD-3-Clause) */
+/* Copyright 2024 NXP */
+
+#include "enetc_pf.h"
+
+int enetc_pf_set_mac_addr(struct net_device *ndev, void *addr);
+int enetc_setup_mac_addresses(struct device_node *np, struct enetc_pf *pf);
+void enetc_pf_netdev_setup(struct enetc_si *si, struct net_device *ndev,
+			   const struct net_device_ops *ndev_ops);
+int enetc_mdiobus_create(struct enetc_pf *pf, struct device_node *node);
+void enetc_mdiobus_destroy(struct enetc_pf *pf);
+int enetc_phylink_create(struct enetc_ndev_priv *priv, struct device_node *node,
+			 const struct phylink_mac_ops *ops);
+void enetc_phylink_destroy(struct enetc_ndev_priv *priv);
+void enetc_set_default_rss_key(struct enetc_pf *pf);
+int enetc_vlan_rx_add_vid(struct net_device *ndev, __be16 prot, u16 vid);
+int enetc_vlan_rx_del_vid(struct net_device *ndev, __be16 prot, u16 vid);
+
+static inline u16 enetc_get_ip_revision(struct enetc_hw *hw)
+{
+	return enetc_global_rd(hw, ENETC_G_EIPBRR0) & EIPBRR0_REVISION;
+}
diff --git a/drivers/net/ethernet/freescale/enetc/enetc_ptp.c b/drivers/net/ethernet/freescale/enetc/enetc_ptp.c
new file mode 100644
index 000000000000..b8413d3b4f16
--- /dev/null
+++ b/drivers/net/ethernet/freescale/enetc/enetc_ptp.c
@@ -0,0 +1,140 @@
+// SPDX-License-Identifier: (GPL-2.0+ OR BSD-3-Clause)
+/* Copyright 2019 NXP */
+
+#include <linux/module.h>
+#include <linux/of.h>
+#include <linux/fsl/ptp_qoriq.h>
+
+#include "enetc.h"
+
+static struct ptp_clock_info enetc_ptp_caps = {
+	.owner		= THIS_MODULE,
+	.name		= "ENETC PTP clock",
+	.max_adj	= 512000,
+	.n_alarm	= 0,
+	.n_ext_ts	= 2,
+	.n_per_out	= 0,
+	.n_pins		= 0,
+	.pps		= 1,
+	.adjfine	= ptp_qoriq_adjfine,
+	.adjtime	= ptp_qoriq_adjtime,
+	.gettime64	= ptp_qoriq_gettime,
+	.settime64	= ptp_qoriq_settime,
+	.enable		= ptp_qoriq_enable,
+};
+
+static int enetc_ptp_probe(struct pci_dev *pdev,
+			   const struct pci_device_id *ent)
+{
+	struct ptp_qoriq *ptp_qoriq;
+	void __iomem *base;
+	int err, len, n;
+
+	if (pdev->dev.of_node && !of_device_is_available(pdev->dev.of_node)) {
+		dev_info(&pdev->dev, "device is disabled, skipping\n");
+		return -ENODEV;
+	}
+
+	err = pci_enable_device_mem(pdev);
+	if (err)
+		return dev_err_probe(&pdev->dev, err, "device enable failed\n");
+
+	err = dma_set_mask_and_coherent(&pdev->dev, DMA_BIT_MASK(64));
+	if (err) {
+		dev_err(&pdev->dev, "DMA configuration failed: 0x%x\n", err);
+		goto err_dma;
+	}
+
+	err = pci_request_mem_regions(pdev, KBUILD_MODNAME);
+	if (err) {
+		dev_err(&pdev->dev, "pci_request_regions failed err=%d\n", err);
+		goto err_pci_mem_reg;
+	}
+
+	pci_set_master(pdev);
+
+	ptp_qoriq = kzalloc(sizeof(*ptp_qoriq), GFP_KERNEL);
+	if (!ptp_qoriq) {
+		err = -ENOMEM;
+		goto err_alloc_ptp;
+	}
+
+	len = pci_resource_len(pdev, ENETC_BAR_REGS);
+
+	base = ioremap(pci_resource_start(pdev, ENETC_BAR_REGS), len);
+	if (!base) {
+		err = -ENXIO;
+		dev_err(&pdev->dev, "ioremap() failed\n");
+		goto err_ioremap;
+	}
+
+	/* Allocate 1 interrupt */
+	n = pci_alloc_irq_vectors(pdev, 1, 1, PCI_IRQ_MSIX);
+	if (n != 1) {
+		err = -EPERM;
+		goto err_irq_vectors;
+	}
+
+	ptp_qoriq->irq = pci_irq_vector(pdev, 0);
+
+	err = request_irq(ptp_qoriq->irq, ptp_qoriq_isr, 0, DRIVER, ptp_qoriq);
+	if (err) {
+		dev_err(&pdev->dev, "request_irq() failed!\n");
+		goto err_irq;
+	}
+
+	ptp_qoriq->dev = &pdev->dev;
+
+	err = ptp_qoriq_init(ptp_qoriq, base, &enetc_ptp_caps);
+	if (err)
+		goto err_no_clock;
+
+	pci_set_drvdata(pdev, ptp_qoriq);
+
+	return 0;
+
+err_no_clock:
+	free_irq(ptp_qoriq->irq, ptp_qoriq);
+err_irq:
+	pci_free_irq_vectors(pdev);
+err_irq_vectors:
+	iounmap(base);
+err_ioremap:
+	kfree(ptp_qoriq);
+err_alloc_ptp:
+	pci_release_mem_regions(pdev);
+err_pci_mem_reg:
+err_dma:
+	pci_disable_device(pdev);
+
+	return err;
+}
+
+static void enetc_ptp_remove(struct pci_dev *pdev)
+{
+	struct ptp_qoriq *ptp_qoriq = pci_get_drvdata(pdev);
+
+	ptp_qoriq_free(ptp_qoriq);
+	pci_free_irq_vectors(pdev);
+	kfree(ptp_qoriq);
+
+	pci_release_mem_regions(pdev);
+	pci_disable_device(pdev);
+}
+
+static const struct pci_device_id enetc_ptp_id_table[] = {
+	{ PCI_DEVICE(PCI_VENDOR_ID_FREESCALE, ENETC_DEV_ID_PTP) },
+	{ 0, } /* End of table. */
+};
+MODULE_DEVICE_TABLE(pci, enetc_ptp_id_table);
+
+static struct pci_driver enetc_ptp_driver = {
+	.name = KBUILD_MODNAME,
+	.id_table = enetc_ptp_id_table,
+	.probe = enetc_ptp_probe,
+	.remove = enetc_ptp_remove,
+};
+module_pci_driver(enetc_ptp_driver);
+
+MODULE_DESCRIPTION("ENETC PTP clock driver");
+MODULE_LICENSE("Dual BSD/GPL");
diff --git a/drivers/net/ethernet/freescale/enetc/enetc_qos.c b/drivers/net/ethernet/freescale/enetc/enetc_qos.c
new file mode 100644
index 000000000000..ccf86651455c
--- /dev/null
+++ b/drivers/net/ethernet/freescale/enetc/enetc_qos.c
@@ -0,0 +1,1689 @@
+// SPDX-License-Identifier: (GPL-2.0+ OR BSD-3-Clause)
+/* Copyright 2019 NXP */
+
+#include "enetc.h"
+
+#include <net/pkt_sched.h>
+#include <linux/math64.h>
+#include <linux/refcount.h>
+#include <net/pkt_cls.h>
+#include <net/tc_act/tc_gate.h>
+
+static u16 enetc_get_max_gcl_len(struct enetc_hw *hw)
+{
+	return enetc_rd(hw, ENETC_PTGCAPR) & ENETC_PTGCAPR_MAX_GCL_LEN_MASK;
+}
+
+void enetc_sched_speed_set(struct enetc_ndev_priv *priv, int speed)
+{
+	struct enetc_hw *hw = &priv->si->hw;
+	u32 old_speed = priv->speed;
+	u32 pspeed, tmp;
+
+	if (speed == old_speed)
+		return;
+
+	switch (speed) {
+	case SPEED_1000:
+		pspeed = ENETC_PMR_PSPEED_1000M;
+		break;
+	case SPEED_2500:
+		pspeed = ENETC_PMR_PSPEED_2500M;
+		break;
+	case SPEED_100:
+		pspeed = ENETC_PMR_PSPEED_100M;
+		break;
+	case SPEED_10:
+	default:
+		pspeed = ENETC_PMR_PSPEED_10M;
+	}
+
+	priv->speed = speed;
+	tmp = enetc_port_rd(hw, ENETC_PMR);
+	enetc_port_wr(hw, ENETC_PMR, (tmp & ~ENETC_PMR_PSPEED_MASK) | pspeed);
+}
+
+static int enetc_setup_taprio(struct enetc_ndev_priv *priv,
+			      struct tc_taprio_qopt_offload *admin_conf)
+{
+	struct enetc_hw *hw = &priv->si->hw;
+	struct enetc_cbd cbd = {.cmd = 0};
+	struct tgs_gcl_conf *gcl_config;
+	struct tgs_gcl_data *gcl_data;
+	dma_addr_t dma;
+	struct gce *gce;
+	u16 data_size;
+	u16 gcl_len;
+	void *tmp;
+	u32 tge;
+	int err;
+	int i;
+
+	/* TSD and Qbv are mutually exclusive in hardware */
+	for (i = 0; i < priv->num_tx_rings; i++)
+		if (priv->tx_ring[i]->tsd_enable)
+			return -EBUSY;
+
+	if (admin_conf->num_entries > enetc_get_max_gcl_len(hw))
+		return -EINVAL;
+
+	if (admin_conf->cycle_time > U32_MAX ||
+	    admin_conf->cycle_time_extension > U32_MAX)
+		return -EINVAL;
+
+	/* Configure the (administrative) gate control list using the
+	 * control BD descriptor.
+	 */
+	gcl_config = &cbd.gcl_conf;
+	gcl_len = admin_conf->num_entries;
+
+	data_size = struct_size(gcl_data, entry, gcl_len);
+	tmp = enetc_cbd_alloc_data_mem(priv->si, &cbd, data_size,
+				       &dma, (void *)&gcl_data);
+	if (!tmp)
+		return -ENOMEM;
+
+	gce = (struct gce *)(gcl_data + 1);
+
+	/* Set all gates open as default */
+	gcl_config->atc = 0xff;
+	gcl_config->acl_len = cpu_to_le16(gcl_len);
+
+	gcl_data->btl = cpu_to_le32(lower_32_bits(admin_conf->base_time));
+	gcl_data->bth = cpu_to_le32(upper_32_bits(admin_conf->base_time));
+	gcl_data->ct = cpu_to_le32(admin_conf->cycle_time);
+	gcl_data->cte = cpu_to_le32(admin_conf->cycle_time_extension);
+
+	for (i = 0; i < gcl_len; i++) {
+		struct tc_taprio_sched_entry *temp_entry;
+		struct gce *temp_gce = gce + i;
+
+		temp_entry = &admin_conf->entries[i];
+
+		temp_gce->gate = (u8)temp_entry->gate_mask;
+		temp_gce->period = cpu_to_le32(temp_entry->interval);
+	}
+
+	cbd.status_flags = 0;
+
+	cbd.cls = BDCR_CMD_PORT_GCL;
+	cbd.status_flags = 0;
+
+	tge = enetc_rd(hw, ENETC_PTGCR);
+	enetc_wr(hw, ENETC_PTGCR, tge | ENETC_PTGCR_TGE);
+
+	err = enetc_send_cmd(priv->si, &cbd);
+	if (err)
+		enetc_wr(hw, ENETC_PTGCR, tge & ~ENETC_PTGCR_TGE);
+
+	enetc_cbd_free_data_mem(priv->si, data_size, tmp, &dma);
+
+	if (err)
+		return err;
+
+	enetc_set_ptcmsdur(hw, admin_conf->max_sdu);
+	priv->active_offloads |= ENETC_F_QBV;
+
+	return 0;
+}
+
+static void enetc_reset_taprio_stats(struct enetc_ndev_priv *priv)
+{
+	int i;
+
+	for (i = 0; i < priv->num_tx_rings; i++)
+		priv->tx_ring[i]->stats.win_drop = 0;
+}
+
+static void enetc_reset_taprio(struct enetc_ndev_priv *priv)
+{
+	struct enetc_hw *hw = &priv->si->hw;
+	u32 val;
+
+	val = enetc_rd(hw, ENETC_PTGCR);
+	enetc_wr(hw, ENETC_PTGCR, val & ~ENETC_PTGCR_TGE);
+	enetc_reset_ptcmsdur(hw);
+
+	priv->active_offloads &= ~ENETC_F_QBV;
+}
+
+static void enetc_taprio_destroy(struct net_device *ndev)
+{
+	struct enetc_ndev_priv *priv = netdev_priv(ndev);
+
+	enetc_reset_taprio(priv);
+	enetc_reset_tc_mqprio(ndev);
+	enetc_reset_taprio_stats(priv);
+}
+
+static void enetc_taprio_stats(struct net_device *ndev,
+			       struct tc_taprio_qopt_stats *stats)
+{
+	struct enetc_ndev_priv *priv = netdev_priv(ndev);
+	u64 window_drops = 0;
+	int i;
+
+	for (i = 0; i < priv->num_tx_rings; i++)
+		window_drops += priv->tx_ring[i]->stats.win_drop;
+
+	stats->window_drops = window_drops;
+}
+
+static void enetc_taprio_queue_stats(struct net_device *ndev,
+				     struct tc_taprio_qopt_queue_stats *queue_stats)
+{
+	struct tc_taprio_qopt_stats *stats = &queue_stats->stats;
+	struct enetc_ndev_priv *priv = netdev_priv(ndev);
+	int queue = queue_stats->queue;
+
+	stats->window_drops = priv->tx_ring[queue]->stats.win_drop;
+}
+
+static int enetc_taprio_replace(struct net_device *ndev,
+				struct tc_taprio_qopt_offload *offload)
+{
+	struct enetc_ndev_priv *priv = netdev_priv(ndev);
+	int err;
+
+	err = enetc_setup_tc_mqprio(ndev, &offload->mqprio);
+	if (err)
+		return err;
+
+	err = enetc_setup_taprio(priv, offload);
+	if (err)
+		enetc_reset_tc_mqprio(ndev);
+
+	return err;
+}
+
+int enetc_setup_tc_taprio(struct net_device *ndev, void *type_data)
+{
+	struct tc_taprio_qopt_offload *offload = type_data;
+	int err = 0;
+
+	switch (offload->cmd) {
+	case TAPRIO_CMD_REPLACE:
+		err = enetc_taprio_replace(ndev, offload);
+		break;
+	case TAPRIO_CMD_DESTROY:
+		enetc_taprio_destroy(ndev);
+		break;
+	case TAPRIO_CMD_STATS:
+		enetc_taprio_stats(ndev, &offload->stats);
+		break;
+	case TAPRIO_CMD_QUEUE_STATS:
+		enetc_taprio_queue_stats(ndev, &offload->queue_stats);
+		break;
+	default:
+		err = -EOPNOTSUPP;
+	}
+
+	return err;
+}
+
+static u32 enetc_get_cbs_enable(struct enetc_hw *hw, u8 tc)
+{
+	return enetc_port_rd(hw, ENETC_PTCCBSR0(tc)) & ENETC_CBSE;
+}
+
+static u8 enetc_get_cbs_bw(struct enetc_hw *hw, u8 tc)
+{
+	return enetc_port_rd(hw, ENETC_PTCCBSR0(tc)) & ENETC_CBS_BW_MASK;
+}
+
+int enetc_setup_tc_cbs(struct net_device *ndev, void *type_data)
+{
+	struct enetc_ndev_priv *priv = netdev_priv(ndev);
+	struct tc_cbs_qopt_offload *cbs = type_data;
+	u32 port_transmit_rate = priv->speed;
+	u8 tc_nums = netdev_get_num_tc(ndev);
+	struct enetc_hw *hw = &priv->si->hw;
+	u32 hi_credit_bit, hi_credit_reg;
+	u32 max_interference_size;
+	u32 port_frame_max_size;
+	u8 tc = cbs->queue;
+	u8 prio_top, prio_next;
+	int bw_sum = 0;
+	u8 bw;
+
+	prio_top = tc_nums - 1;
+	prio_next = tc_nums - 2;
+
+	/* Support highest prio and second prio tc in cbs mode */
+	if (tc != prio_top && tc != prio_next)
+		return -EOPNOTSUPP;
+
+	if (!cbs->enable) {
+		/* Make sure the other TC that are numerically
+		 * lower than this TC have been disabled.
+		 */
+		if (tc == prio_top &&
+		    enetc_get_cbs_enable(hw, prio_next)) {
+			dev_err(&ndev->dev,
+				"Disable TC%d before disable TC%d\n",
+				prio_next, tc);
+			return -EINVAL;
+		}
+
+		enetc_port_wr(hw, ENETC_PTCCBSR1(tc), 0);
+		enetc_port_wr(hw, ENETC_PTCCBSR0(tc), 0);
+
+		return 0;
+	}
+
+	if (cbs->idleslope - cbs->sendslope != port_transmit_rate * 1000L ||
+	    cbs->idleslope < 0 || cbs->sendslope > 0)
+		return -EOPNOTSUPP;
+
+	port_frame_max_size = ndev->mtu + VLAN_ETH_HLEN + ETH_FCS_LEN;
+
+	bw = cbs->idleslope / (port_transmit_rate * 10UL);
+
+	/* Make sure the other TC that are numerically
+	 * higher than this TC have been enabled.
+	 */
+	if (tc == prio_next) {
+		if (!enetc_get_cbs_enable(hw, prio_top)) {
+			dev_err(&ndev->dev,
+				"Enable TC%d first before enable TC%d\n",
+				prio_top, prio_next);
+			return -EINVAL;
+		}
+		bw_sum += enetc_get_cbs_bw(hw, prio_top);
+	}
+
+	if (bw_sum + bw >= 100) {
+		dev_err(&ndev->dev,
+			"The sum of all CBS Bandwidth can't exceed 100\n");
+		return -EINVAL;
+	}
+
+	enetc_port_rd(hw, ENETC_PTCMSDUR(tc));
+
+	/* For top prio TC, the max_interfrence_size is maxSizedFrame.
+	 *
+	 * For next prio TC, the max_interfrence_size is calculated as below:
+	 *
+	 *      max_interference_size = M0 + Ma + Ra * M0 / (R0 - Ra)
+	 *
+	 *	- RA: idleSlope for AVB Class A
+	 *	- R0: port transmit rate
+	 *	- M0: maximum sized frame for the port
+	 *	- MA: maximum sized frame for AVB Class A
+	 */
+
+	if (tc == prio_top) {
+		max_interference_size = port_frame_max_size * 8;
+	} else {
+		u32 m0, ma, r0, ra;
+
+		m0 = port_frame_max_size * 8;
+		ma = enetc_port_rd(hw, ENETC_PTCMSDUR(prio_top)) * 8;
+		ra = enetc_get_cbs_bw(hw, prio_top) *
+			port_transmit_rate * 10000ULL;
+		r0 = port_transmit_rate * 1000000ULL;
+		max_interference_size = m0 + ma +
+			(u32)div_u64((u64)ra * m0, r0 - ra);
+	}
+
+	/* hiCredit bits calculate by:
+	 *
+	 * maxSizedFrame * (idleSlope/portTxRate)
+	 */
+	hi_credit_bit = max_interference_size * bw / 100;
+
+	/* hiCredit bits to hiCredit register need to calculated as:
+	 *
+	 * (enetClockFrequency / portTransmitRate) * 100
+	 */
+	hi_credit_reg = (u32)div_u64((priv->sysclk_freq * 100ULL) * hi_credit_bit,
+				     port_transmit_rate * 1000000ULL);
+
+	enetc_port_wr(hw, ENETC_PTCCBSR1(tc), hi_credit_reg);
+
+	/* Set bw register and enable this traffic class */
+	enetc_port_wr(hw, ENETC_PTCCBSR0(tc), bw | ENETC_CBSE);
+
+	return 0;
+}
+
+int enetc_setup_tc_txtime(struct net_device *ndev, void *type_data)
+{
+	struct enetc_ndev_priv *priv = netdev_priv(ndev);
+	struct tc_etf_qopt_offload *qopt = type_data;
+	u8 tc_nums = netdev_get_num_tc(ndev);
+	struct enetc_hw *hw = &priv->si->hw;
+	int tc;
+
+	if (!tc_nums)
+		return -EOPNOTSUPP;
+
+	tc = qopt->queue;
+
+	if (tc < 0 || tc >= priv->num_tx_rings)
+		return -EINVAL;
+
+	/* TSD and Qbv are mutually exclusive in hardware */
+	if (enetc_rd(hw, ENETC_PTGCR) & ENETC_PTGCR_TGE)
+		return -EBUSY;
+
+	priv->tx_ring[tc]->tsd_enable = qopt->enable;
+	enetc_port_wr(hw, ENETC_PTCTSDR(tc), qopt->enable ? ENETC_TSDE : 0);
+
+	return 0;
+}
+
+enum streamid_type {
+	STREAMID_TYPE_RESERVED = 0,
+	STREAMID_TYPE_NULL,
+	STREAMID_TYPE_SMAC,
+};
+
+enum streamid_vlan_tagged {
+	STREAMID_VLAN_RESERVED = 0,
+	STREAMID_VLAN_TAGGED,
+	STREAMID_VLAN_UNTAGGED,
+	STREAMID_VLAN_ALL,
+};
+
+#define ENETC_PSFP_WILDCARD -1
+#define HANDLE_OFFSET 100
+
+enum forward_type {
+	FILTER_ACTION_TYPE_PSFP = BIT(0),
+	FILTER_ACTION_TYPE_ACL = BIT(1),
+	FILTER_ACTION_TYPE_BOTH = GENMASK(1, 0),
+};
+
+/* This is for limit output type for input actions */
+struct actions_fwd {
+	u64 actions;
+	u64 keys;	/* include the must needed keys */
+	enum forward_type output;
+};
+
+struct psfp_streamfilter_counters {
+	u64 matching_frames_count;
+	u64 passing_frames_count;
+	u64 not_passing_frames_count;
+	u64 passing_sdu_count;
+	u64 not_passing_sdu_count;
+	u64 red_frames_count;
+};
+
+struct enetc_streamid {
+	u32 index;
+	union {
+		u8 src_mac[6];
+		u8 dst_mac[6];
+	};
+	u8 filtertype;
+	u16 vid;
+	u8 tagged;
+	s32 handle;
+};
+
+struct enetc_psfp_filter {
+	u32 index;
+	s32 handle;
+	s8 prio;
+	u32 maxsdu;
+	u32 gate_id;
+	s32 meter_id;
+	refcount_t refcount;
+	struct hlist_node node;
+};
+
+struct enetc_psfp_gate {
+	u32 index;
+	s8 init_ipv;
+	u64 basetime;
+	u64 cycletime;
+	u64 cycletimext;
+	u32 num_entries;
+	refcount_t refcount;
+	struct hlist_node node;
+	struct action_gate_entry entries[] __counted_by(num_entries);
+};
+
+/* Only enable the green color frame now
+ * Will add eir and ebs color blind, couple flag etc when
+ * policing action add more offloading parameters
+ */
+struct enetc_psfp_meter {
+	u32 index;
+	u32 cir;
+	u32 cbs;
+	refcount_t refcount;
+	struct hlist_node node;
+};
+
+#define ENETC_PSFP_FLAGS_FMI BIT(0)
+
+struct enetc_stream_filter {
+	struct enetc_streamid sid;
+	u32 sfi_index;
+	u32 sgi_index;
+	u32 flags;
+	u32 fmi_index;
+	struct flow_stats stats;
+	struct hlist_node node;
+};
+
+struct enetc_psfp {
+	unsigned long dev_bitmap;
+	unsigned long *psfp_sfi_bitmap;
+	struct hlist_head stream_list;
+	struct hlist_head psfp_filter_list;
+	struct hlist_head psfp_gate_list;
+	struct hlist_head psfp_meter_list;
+	spinlock_t psfp_lock; /* spinlock for the struct enetc_psfp r/w */
+};
+
+static struct actions_fwd enetc_act_fwd[] = {
+	{
+		BIT(FLOW_ACTION_GATE),
+		BIT_ULL(FLOW_DISSECTOR_KEY_ETH_ADDRS),
+		FILTER_ACTION_TYPE_PSFP
+	},
+	{
+		BIT(FLOW_ACTION_POLICE) |
+		BIT(FLOW_ACTION_GATE),
+		BIT_ULL(FLOW_DISSECTOR_KEY_ETH_ADDRS),
+		FILTER_ACTION_TYPE_PSFP
+	},
+	/* example for ACL actions */
+	{
+		BIT(FLOW_ACTION_DROP),
+		0,
+		FILTER_ACTION_TYPE_ACL
+	}
+};
+
+static struct enetc_psfp epsfp = {
+	.dev_bitmap = 0,
+	.psfp_sfi_bitmap = NULL,
+};
+
+static LIST_HEAD(enetc_block_cb_list);
+
+/* Stream Identity Entry Set Descriptor */
+static int enetc_streamid_hw_set(struct enetc_ndev_priv *priv,
+				 struct enetc_streamid *sid,
+				 u8 enable)
+{
+	struct enetc_cbd cbd = {.cmd = 0};
+	struct streamid_data *si_data;
+	struct streamid_conf *si_conf;
+	dma_addr_t dma;
+	u16 data_size;
+	void *tmp;
+	int port;
+	int err;
+
+	port = enetc_pf_to_port(priv->si->pdev);
+	if (port < 0)
+		return -EINVAL;
+
+	if (sid->index >= priv->psfp_cap.max_streamid)
+		return -EINVAL;
+
+	if (sid->filtertype != STREAMID_TYPE_NULL &&
+	    sid->filtertype != STREAMID_TYPE_SMAC)
+		return -EOPNOTSUPP;
+
+	/* Disable operation before enable */
+	cbd.index = cpu_to_le16((u16)sid->index);
+	cbd.cls = BDCR_CMD_STREAM_IDENTIFY;
+	cbd.status_flags = 0;
+
+	data_size = sizeof(struct streamid_data);
+	tmp = enetc_cbd_alloc_data_mem(priv->si, &cbd, data_size,
+				       &dma, (void *)&si_data);
+	if (!tmp)
+		return -ENOMEM;
+
+	eth_broadcast_addr(si_data->dmac);
+	si_data->vid_vidm_tg = (ENETC_CBDR_SID_VID_MASK
+			       + ((0x3 << 14) | ENETC_CBDR_SID_VIDM));
+
+	si_conf = &cbd.sid_set;
+	/* Only one port supported for one entry, set itself */
+	si_conf->iports = cpu_to_le32(1 << port);
+	si_conf->id_type = 1;
+	si_conf->oui[2] = 0x0;
+	si_conf->oui[1] = 0x80;
+	si_conf->oui[0] = 0xC2;
+
+	err = enetc_send_cmd(priv->si, &cbd);
+	if (err)
+		goto out;
+
+	if (!enable)
+		goto out;
+
+	/* Enable the entry overwrite again incase space flushed by hardware */
+	cbd.status_flags = 0;
+
+	si_conf->en = 0x80;
+	si_conf->stream_handle = cpu_to_le32(sid->handle);
+	si_conf->iports = cpu_to_le32(1 << port);
+	si_conf->id_type = sid->filtertype;
+	si_conf->oui[2] = 0x0;
+	si_conf->oui[1] = 0x80;
+	si_conf->oui[0] = 0xC2;
+
+	memset(si_data, 0, data_size);
+
+	/* VIDM default to be 1.
+	 * VID Match. If set (b1) then the VID must match, otherwise
+	 * any VID is considered a match. VIDM setting is only used
+	 * when TG is set to b01.
+	 */
+	if (si_conf->id_type == STREAMID_TYPE_NULL) {
+		ether_addr_copy(si_data->dmac, sid->dst_mac);
+		si_data->vid_vidm_tg = (sid->vid & ENETC_CBDR_SID_VID_MASK) +
+				       ((((u16)(sid->tagged) & 0x3) << 14)
+				       | ENETC_CBDR_SID_VIDM);
+	} else if (si_conf->id_type == STREAMID_TYPE_SMAC) {
+		ether_addr_copy(si_data->smac, sid->src_mac);
+		si_data->vid_vidm_tg = (sid->vid & ENETC_CBDR_SID_VID_MASK) +
+				       ((((u16)(sid->tagged) & 0x3) << 14)
+				       | ENETC_CBDR_SID_VIDM);
+	}
+
+	err = enetc_send_cmd(priv->si, &cbd);
+out:
+	enetc_cbd_free_data_mem(priv->si, data_size, tmp, &dma);
+
+	return err;
+}
+
+/* Stream Filter Instance Set Descriptor */
+static int enetc_streamfilter_hw_set(struct enetc_ndev_priv *priv,
+				     struct enetc_psfp_filter *sfi,
+				     u8 enable)
+{
+	struct enetc_cbd cbd = {.cmd = 0};
+	struct sfi_conf *sfi_config;
+	int port;
+
+	port = enetc_pf_to_port(priv->si->pdev);
+	if (port < 0)
+		return -EINVAL;
+
+	cbd.index = cpu_to_le16(sfi->index);
+	cbd.cls = BDCR_CMD_STREAM_FILTER;
+	cbd.status_flags = 0x80;
+	cbd.length = cpu_to_le16(1);
+
+	sfi_config = &cbd.sfi_conf;
+	if (!enable)
+		goto exit;
+
+	sfi_config->en = 0x80;
+
+	if (sfi->handle >= 0) {
+		sfi_config->stream_handle =
+			cpu_to_le32(sfi->handle);
+		sfi_config->sthm |= 0x80;
+	}
+
+	sfi_config->sg_inst_table_index = cpu_to_le16(sfi->gate_id);
+	sfi_config->input_ports = cpu_to_le32(1 << port);
+
+	/* The priority value which may be matched against the
+	 * frame’s priority value to determine a match for this entry.
+	 */
+	if (sfi->prio >= 0)
+		sfi_config->multi |= (sfi->prio & 0x7) | 0x8;
+
+	/* Filter Type. Identifies the contents of the MSDU/FM_INST_INDEX
+	 * field as being either an MSDU value or an index into the Flow
+	 * Meter Instance table.
+	 */
+	if (sfi->maxsdu) {
+		sfi_config->msdu =
+		cpu_to_le16(sfi->maxsdu);
+		sfi_config->multi |= 0x40;
+	}
+
+	if (sfi->meter_id >= 0) {
+		sfi_config->fm_inst_table_index = cpu_to_le16(sfi->meter_id);
+		sfi_config->multi |= 0x80;
+	}
+
+exit:
+	return enetc_send_cmd(priv->si, &cbd);
+}
+
+static int enetc_streamcounter_hw_get(struct enetc_ndev_priv *priv,
+				      u32 index,
+				      struct psfp_streamfilter_counters *cnt)
+{
+	struct enetc_cbd cbd = { .cmd = 2 };
+	struct sfi_counter_data *data_buf;
+	dma_addr_t dma;
+	u16 data_size;
+	void *tmp;
+	int err;
+
+	cbd.index = cpu_to_le16((u16)index);
+	cbd.cmd = 2;
+	cbd.cls = BDCR_CMD_STREAM_FILTER;
+	cbd.status_flags = 0;
+
+	data_size = sizeof(struct sfi_counter_data);
+
+	tmp = enetc_cbd_alloc_data_mem(priv->si, &cbd, data_size,
+				       &dma, (void *)&data_buf);
+	if (!tmp)
+		return -ENOMEM;
+
+	err = enetc_send_cmd(priv->si, &cbd);
+	if (err)
+		goto exit;
+
+	cnt->matching_frames_count = ((u64)data_buf->matchh << 32) +
+				     data_buf->matchl;
+
+	cnt->not_passing_sdu_count = ((u64)data_buf->msdu_droph << 32) +
+				     data_buf->msdu_dropl;
+
+	cnt->passing_sdu_count = cnt->matching_frames_count
+				- cnt->not_passing_sdu_count;
+
+	cnt->not_passing_frames_count =
+				((u64)data_buf->stream_gate_droph << 32) +
+				data_buf->stream_gate_dropl;
+
+	cnt->passing_frames_count = cnt->matching_frames_count -
+				    cnt->not_passing_sdu_count -
+				    cnt->not_passing_frames_count;
+
+	cnt->red_frames_count =	((u64)data_buf->flow_meter_droph << 32)	+
+				data_buf->flow_meter_dropl;
+
+exit:
+	enetc_cbd_free_data_mem(priv->si, data_size, tmp, &dma);
+
+	return err;
+}
+
+static u64 get_ptp_now(struct enetc_hw *hw)
+{
+	u64 now_lo, now_hi, now;
+
+	now_lo = enetc_rd(hw, ENETC_SICTR0);
+	now_hi = enetc_rd(hw, ENETC_SICTR1);
+	now = now_lo | now_hi << 32;
+
+	return now;
+}
+
+static int get_start_ns(u64 now, u64 cycle, u64 *start)
+{
+	u64 n;
+
+	if (!cycle)
+		return -EFAULT;
+
+	n = div64_u64(now, cycle);
+
+	*start = (n + 1) * cycle;
+
+	return 0;
+}
+
+/* Stream Gate Instance Set Descriptor */
+static int enetc_streamgate_hw_set(struct enetc_ndev_priv *priv,
+				   struct enetc_psfp_gate *sgi,
+				   u8 enable)
+{
+	struct enetc_cbd cbd = { .cmd = 0 };
+	struct sgi_table *sgi_config;
+	struct sgcl_conf *sgcl_config;
+	struct sgcl_data *sgcl_data;
+	struct sgce *sgce;
+	dma_addr_t dma;
+	u16 data_size;
+	int err, i;
+	void *tmp;
+	u64 now;
+
+	cbd.index = cpu_to_le16(sgi->index);
+	cbd.cmd = 0;
+	cbd.cls = BDCR_CMD_STREAM_GCL;
+	cbd.status_flags = 0x80;
+
+	/* disable */
+	if (!enable)
+		return enetc_send_cmd(priv->si, &cbd);
+
+	if (!sgi->num_entries)
+		return 0;
+
+	if (sgi->num_entries > priv->psfp_cap.max_psfp_gatelist ||
+	    !sgi->cycletime)
+		return -EINVAL;
+
+	/* enable */
+	sgi_config = &cbd.sgi_table;
+
+	/* Keep open before gate list start */
+	sgi_config->ocgtst = 0x80;
+
+	sgi_config->oipv = (sgi->init_ipv < 0) ?
+				0x0 : ((sgi->init_ipv & 0x7) | 0x8);
+
+	sgi_config->en = 0x80;
+
+	/* Basic config */
+	err = enetc_send_cmd(priv->si, &cbd);
+	if (err)
+		return -EINVAL;
+
+	memset(&cbd, 0, sizeof(cbd));
+
+	cbd.index = cpu_to_le16(sgi->index);
+	cbd.cmd = 1;
+	cbd.cls = BDCR_CMD_STREAM_GCL;
+	cbd.status_flags = 0;
+
+	sgcl_config = &cbd.sgcl_conf;
+
+	sgcl_config->acl_len = (sgi->num_entries - 1) & 0x3;
+
+	data_size = struct_size(sgcl_data, sgcl, sgi->num_entries);
+	tmp = enetc_cbd_alloc_data_mem(priv->si, &cbd, data_size,
+				       &dma, (void *)&sgcl_data);
+	if (!tmp)
+		return -ENOMEM;
+
+	sgce = &sgcl_data->sgcl[0];
+
+	sgcl_config->agtst = 0x80;
+
+	sgcl_data->ct = sgi->cycletime;
+	sgcl_data->cte = sgi->cycletimext;
+
+	if (sgi->init_ipv >= 0)
+		sgcl_config->aipv = (sgi->init_ipv & 0x7) | 0x8;
+
+	for (i = 0; i < sgi->num_entries; i++) {
+		struct action_gate_entry *from = &sgi->entries[i];
+		struct sgce *to = &sgce[i];
+
+		if (from->gate_state)
+			to->multi |= 0x10;
+
+		if (from->ipv >= 0)
+			to->multi |= ((from->ipv & 0x7) << 5) | 0x08;
+
+		if (from->maxoctets >= 0) {
+			to->multi |= 0x01;
+			to->msdu[0] = from->maxoctets & 0xFF;
+			to->msdu[1] = (from->maxoctets >> 8) & 0xFF;
+			to->msdu[2] = (from->maxoctets >> 16) & 0xFF;
+		}
+
+		to->interval = from->interval;
+	}
+
+	/* If basetime is less than now, calculate start time */
+	now = get_ptp_now(&priv->si->hw);
+
+	if (sgi->basetime < now) {
+		u64 start;
+
+		err = get_start_ns(now, sgi->cycletime, &start);
+		if (err)
+			goto exit;
+		sgcl_data->btl = lower_32_bits(start);
+		sgcl_data->bth = upper_32_bits(start);
+	} else {
+		u32 hi, lo;
+
+		hi = upper_32_bits(sgi->basetime);
+		lo = lower_32_bits(sgi->basetime);
+		sgcl_data->bth = hi;
+		sgcl_data->btl = lo;
+	}
+
+	err = enetc_send_cmd(priv->si, &cbd);
+
+exit:
+	enetc_cbd_free_data_mem(priv->si, data_size, tmp, &dma);
+	return err;
+}
+
+static int enetc_flowmeter_hw_set(struct enetc_ndev_priv *priv,
+				  struct enetc_psfp_meter *fmi,
+				  u8 enable)
+{
+	struct enetc_cbd cbd = { .cmd = 0 };
+	struct fmi_conf *fmi_config;
+	u64 temp = 0;
+
+	cbd.index = cpu_to_le16((u16)fmi->index);
+	cbd.cls = BDCR_CMD_FLOW_METER;
+	cbd.status_flags = 0x80;
+
+	if (!enable)
+		return enetc_send_cmd(priv->si, &cbd);
+
+	fmi_config = &cbd.fmi_conf;
+	fmi_config->en = 0x80;
+
+	if (fmi->cir) {
+		temp = (u64)8000 * fmi->cir;
+		temp = div_u64(temp, 3725);
+	}
+
+	fmi_config->cir = cpu_to_le32((u32)temp);
+	fmi_config->cbs = cpu_to_le32(fmi->cbs);
+
+	/* Default for eir ebs disable */
+	fmi_config->eir = 0;
+	fmi_config->ebs = 0;
+
+	/* Default:
+	 * mark red disable
+	 * drop on yellow disable
+	 * color mode disable
+	 * couple flag disable
+	 */
+	fmi_config->conf = 0;
+
+	return enetc_send_cmd(priv->si, &cbd);
+}
+
+static struct enetc_stream_filter *enetc_get_stream_by_index(u32 index)
+{
+	struct enetc_stream_filter *f;
+
+	hlist_for_each_entry(f, &epsfp.stream_list, node)
+		if (f->sid.index == index)
+			return f;
+
+	return NULL;
+}
+
+static struct enetc_psfp_gate *enetc_get_gate_by_index(u32 index)
+{
+	struct enetc_psfp_gate *g;
+
+	hlist_for_each_entry(g, &epsfp.psfp_gate_list, node)
+		if (g->index == index)
+			return g;
+
+	return NULL;
+}
+
+static struct enetc_psfp_filter *enetc_get_filter_by_index(u32 index)
+{
+	struct enetc_psfp_filter *s;
+
+	hlist_for_each_entry(s, &epsfp.psfp_filter_list, node)
+		if (s->index == index)
+			return s;
+
+	return NULL;
+}
+
+static struct enetc_psfp_meter *enetc_get_meter_by_index(u32 index)
+{
+	struct enetc_psfp_meter *m;
+
+	hlist_for_each_entry(m, &epsfp.psfp_meter_list, node)
+		if (m->index == index)
+			return m;
+
+	return NULL;
+}
+
+static struct enetc_psfp_filter
+	*enetc_psfp_check_sfi(struct enetc_psfp_filter *sfi)
+{
+	struct enetc_psfp_filter *s;
+
+	hlist_for_each_entry(s, &epsfp.psfp_filter_list, node)
+		if (s->gate_id == sfi->gate_id &&
+		    s->prio == sfi->prio &&
+		    s->maxsdu == sfi->maxsdu &&
+		    s->meter_id == sfi->meter_id)
+			return s;
+
+	return NULL;
+}
+
+static int enetc_get_free_index(struct enetc_ndev_priv *priv)
+{
+	u32 max_size = priv->psfp_cap.max_psfp_filter;
+	unsigned long index;
+
+	index = find_first_zero_bit(epsfp.psfp_sfi_bitmap, max_size);
+	if (index == max_size)
+		return -1;
+
+	return index;
+}
+
+static void stream_filter_unref(struct enetc_ndev_priv *priv, u32 index)
+{
+	struct enetc_psfp_filter *sfi;
+	u8 z;
+
+	sfi = enetc_get_filter_by_index(index);
+	WARN_ON(!sfi);
+	z = refcount_dec_and_test(&sfi->refcount);
+
+	if (z) {
+		enetc_streamfilter_hw_set(priv, sfi, false);
+		hlist_del(&sfi->node);
+		kfree(sfi);
+		clear_bit(index, epsfp.psfp_sfi_bitmap);
+	}
+}
+
+static void stream_gate_unref(struct enetc_ndev_priv *priv, u32 index)
+{
+	struct enetc_psfp_gate *sgi;
+	u8 z;
+
+	sgi = enetc_get_gate_by_index(index);
+	WARN_ON(!sgi);
+	z = refcount_dec_and_test(&sgi->refcount);
+	if (z) {
+		enetc_streamgate_hw_set(priv, sgi, false);
+		hlist_del(&sgi->node);
+		kfree(sgi);
+	}
+}
+
+static void flow_meter_unref(struct enetc_ndev_priv *priv, u32 index)
+{
+	struct enetc_psfp_meter *fmi;
+	u8 z;
+
+	fmi = enetc_get_meter_by_index(index);
+	WARN_ON(!fmi);
+	z = refcount_dec_and_test(&fmi->refcount);
+	if (z) {
+		enetc_flowmeter_hw_set(priv, fmi, false);
+		hlist_del(&fmi->node);
+		kfree(fmi);
+	}
+}
+
+static void remove_one_chain(struct enetc_ndev_priv *priv,
+			     struct enetc_stream_filter *filter)
+{
+	if (filter->flags & ENETC_PSFP_FLAGS_FMI)
+		flow_meter_unref(priv, filter->fmi_index);
+
+	stream_gate_unref(priv, filter->sgi_index);
+	stream_filter_unref(priv, filter->sfi_index);
+
+	hlist_del(&filter->node);
+	kfree(filter);
+}
+
+static int enetc_psfp_hw_set(struct enetc_ndev_priv *priv,
+			     struct enetc_streamid *sid,
+			     struct enetc_psfp_filter *sfi,
+			     struct enetc_psfp_gate *sgi,
+			     struct enetc_psfp_meter *fmi)
+{
+	int err;
+
+	err = enetc_streamid_hw_set(priv, sid, true);
+	if (err)
+		return err;
+
+	if (sfi) {
+		err = enetc_streamfilter_hw_set(priv, sfi, true);
+		if (err)
+			goto revert_sid;
+	}
+
+	err = enetc_streamgate_hw_set(priv, sgi, true);
+	if (err)
+		goto revert_sfi;
+
+	if (fmi) {
+		err = enetc_flowmeter_hw_set(priv, fmi, true);
+		if (err)
+			goto revert_sgi;
+	}
+
+	return 0;
+
+revert_sgi:
+	enetc_streamgate_hw_set(priv, sgi, false);
+revert_sfi:
+	if (sfi)
+		enetc_streamfilter_hw_set(priv, sfi, false);
+revert_sid:
+	enetc_streamid_hw_set(priv, sid, false);
+	return err;
+}
+
+static struct actions_fwd *
+enetc_check_flow_actions(u64 acts, unsigned long long inputkeys)
+{
+	int i;
+
+	for (i = 0; i < ARRAY_SIZE(enetc_act_fwd); i++)
+		if (acts == enetc_act_fwd[i].actions &&
+		    inputkeys & enetc_act_fwd[i].keys)
+			return &enetc_act_fwd[i];
+
+	return NULL;
+}
+
+static int enetc_psfp_policer_validate(const struct flow_action *action,
+				       const struct flow_action_entry *act,
+				       struct netlink_ext_ack *extack)
+{
+	if (act->police.exceed.act_id != FLOW_ACTION_DROP) {
+		NL_SET_ERR_MSG_MOD(extack,
+				   "Offload not supported when exceed action is not drop");
+		return -EOPNOTSUPP;
+	}
+
+	if (act->police.notexceed.act_id != FLOW_ACTION_PIPE &&
+	    act->police.notexceed.act_id != FLOW_ACTION_ACCEPT) {
+		NL_SET_ERR_MSG_MOD(extack,
+				   "Offload not supported when conform action is not pipe or ok");
+		return -EOPNOTSUPP;
+	}
+
+	if (act->police.notexceed.act_id == FLOW_ACTION_ACCEPT &&
+	    !flow_action_is_last_entry(action, act)) {
+		NL_SET_ERR_MSG_MOD(extack,
+				   "Offload not supported when conform action is ok, but action is not last");
+		return -EOPNOTSUPP;
+	}
+
+	if (act->police.peakrate_bytes_ps ||
+	    act->police.avrate || act->police.overhead) {
+		NL_SET_ERR_MSG_MOD(extack,
+				   "Offload not supported when peakrate/avrate/overhead is configured");
+		return -EOPNOTSUPP;
+	}
+
+	if (act->police.rate_pkt_ps) {
+		NL_SET_ERR_MSG_MOD(extack,
+				   "QoS offload not support packets per second");
+		return -EOPNOTSUPP;
+	}
+
+	return 0;
+}
+
+static int enetc_psfp_parse_clsflower(struct enetc_ndev_priv *priv,
+				      struct flow_cls_offload *f)
+{
+	struct flow_action_entry *entryg = NULL, *entryp = NULL;
+	struct flow_rule *rule = flow_cls_offload_flow_rule(f);
+	struct netlink_ext_ack *extack = f->common.extack;
+	struct enetc_stream_filter *filter, *old_filter;
+	struct enetc_psfp_meter *fmi = NULL, *old_fmi;
+	struct enetc_psfp_filter *sfi, *old_sfi;
+	struct enetc_psfp_gate *sgi, *old_sgi;
+	struct flow_action_entry *entry;
+	struct action_gate_entry *e;
+	u8 sfi_overwrite = 0;
+	int entries_size;
+	int i, err;
+
+	if (f->common.chain_index >= priv->psfp_cap.max_streamid) {
+		NL_SET_ERR_MSG_MOD(extack, "No Stream identify resource!");
+		return -ENOSPC;
+	}
+
+	flow_action_for_each(i, entry, &rule->action)
+		if (entry->id == FLOW_ACTION_GATE)
+			entryg = entry;
+		else if (entry->id == FLOW_ACTION_POLICE)
+			entryp = entry;
+
+	/* Not support without gate action */
+	if (!entryg)
+		return -EINVAL;
+
+	filter = kzalloc(sizeof(*filter), GFP_KERNEL);
+	if (!filter)
+		return -ENOMEM;
+
+	filter->sid.index = f->common.chain_index;
+
+	if (flow_rule_match_key(rule, FLOW_DISSECTOR_KEY_ETH_ADDRS)) {
+		struct flow_match_eth_addrs match;
+
+		flow_rule_match_eth_addrs(rule, &match);
+
+		if (!is_zero_ether_addr(match.mask->dst) &&
+		    !is_zero_ether_addr(match.mask->src)) {
+			NL_SET_ERR_MSG_MOD(extack,
+					   "Cannot match on both source and destination MAC");
+			err = -EINVAL;
+			goto free_filter;
+		}
+
+		if (!is_zero_ether_addr(match.mask->dst)) {
+			if (!is_broadcast_ether_addr(match.mask->dst)) {
+				NL_SET_ERR_MSG_MOD(extack,
+						   "Masked matching on destination MAC not supported");
+				err = -EINVAL;
+				goto free_filter;
+			}
+			ether_addr_copy(filter->sid.dst_mac, match.key->dst);
+			filter->sid.filtertype = STREAMID_TYPE_NULL;
+		}
+
+		if (!is_zero_ether_addr(match.mask->src)) {
+			if (!is_broadcast_ether_addr(match.mask->src)) {
+				NL_SET_ERR_MSG_MOD(extack,
+						   "Masked matching on source MAC not supported");
+				err = -EINVAL;
+				goto free_filter;
+			}
+			ether_addr_copy(filter->sid.src_mac, match.key->src);
+			filter->sid.filtertype = STREAMID_TYPE_SMAC;
+		}
+	} else {
+		NL_SET_ERR_MSG_MOD(extack, "Unsupported, must include ETH_ADDRS");
+		err = -EINVAL;
+		goto free_filter;
+	}
+
+	if (flow_rule_match_key(rule, FLOW_DISSECTOR_KEY_VLAN)) {
+		struct flow_match_vlan match;
+
+		flow_rule_match_vlan(rule, &match);
+		if (match.mask->vlan_priority) {
+			if (match.mask->vlan_priority !=
+			    (VLAN_PRIO_MASK >> VLAN_PRIO_SHIFT)) {
+				NL_SET_ERR_MSG_MOD(extack, "Only full mask is supported for VLAN priority");
+				err = -EINVAL;
+				goto free_filter;
+			}
+		}
+
+		if (match.mask->vlan_id) {
+			if (match.mask->vlan_id != VLAN_VID_MASK) {
+				NL_SET_ERR_MSG_MOD(extack, "Only full mask is supported for VLAN id");
+				err = -EINVAL;
+				goto free_filter;
+			}
+
+			filter->sid.vid = match.key->vlan_id;
+			if (!filter->sid.vid)
+				filter->sid.tagged = STREAMID_VLAN_UNTAGGED;
+			else
+				filter->sid.tagged = STREAMID_VLAN_TAGGED;
+		}
+	} else {
+		filter->sid.tagged = STREAMID_VLAN_ALL;
+	}
+
+	/* parsing gate action */
+	if (entryg->hw_index >= priv->psfp_cap.max_psfp_gate) {
+		NL_SET_ERR_MSG_MOD(extack, "No Stream Gate resource!");
+		err = -ENOSPC;
+		goto free_filter;
+	}
+
+	if (entryg->gate.num_entries >= priv->psfp_cap.max_psfp_gatelist) {
+		NL_SET_ERR_MSG_MOD(extack, "No Stream Gate resource!");
+		err = -ENOSPC;
+		goto free_filter;
+	}
+
+	entries_size = struct_size(sgi, entries, entryg->gate.num_entries);
+	sgi = kzalloc(entries_size, GFP_KERNEL);
+	if (!sgi) {
+		err = -ENOMEM;
+		goto free_filter;
+	}
+
+	refcount_set(&sgi->refcount, 1);
+	sgi->index = entryg->hw_index;
+	sgi->init_ipv = entryg->gate.prio;
+	sgi->basetime = entryg->gate.basetime;
+	sgi->cycletime = entryg->gate.cycletime;
+	sgi->num_entries = entryg->gate.num_entries;
+
+	e = sgi->entries;
+	for (i = 0; i < entryg->gate.num_entries; i++) {
+		e[i].gate_state = entryg->gate.entries[i].gate_state;
+		e[i].interval = entryg->gate.entries[i].interval;
+		e[i].ipv = entryg->gate.entries[i].ipv;
+		e[i].maxoctets = entryg->gate.entries[i].maxoctets;
+	}
+
+	filter->sgi_index = sgi->index;
+
+	sfi = kzalloc(sizeof(*sfi), GFP_KERNEL);
+	if (!sfi) {
+		err = -ENOMEM;
+		goto free_gate;
+	}
+
+	refcount_set(&sfi->refcount, 1);
+	sfi->gate_id = sgi->index;
+	sfi->meter_id = ENETC_PSFP_WILDCARD;
+
+	/* Flow meter and max frame size */
+	if (entryp) {
+		err = enetc_psfp_policer_validate(&rule->action, entryp, extack);
+		if (err)
+			goto free_sfi;
+
+		if (entryp->police.burst) {
+			fmi = kzalloc(sizeof(*fmi), GFP_KERNEL);
+			if (!fmi) {
+				err = -ENOMEM;
+				goto free_sfi;
+			}
+			refcount_set(&fmi->refcount, 1);
+			fmi->cir = entryp->police.rate_bytes_ps;
+			fmi->cbs = entryp->police.burst;
+			fmi->index = entryp->hw_index;
+			filter->flags |= ENETC_PSFP_FLAGS_FMI;
+			filter->fmi_index = fmi->index;
+			sfi->meter_id = fmi->index;
+		}
+
+		if (entryp->police.mtu)
+			sfi->maxsdu = entryp->police.mtu;
+	}
+
+	/* prio ref the filter prio */
+	if (f->common.prio && f->common.prio <= BIT(3))
+		sfi->prio = f->common.prio - 1;
+	else
+		sfi->prio = ENETC_PSFP_WILDCARD;
+
+	old_sfi = enetc_psfp_check_sfi(sfi);
+	if (!old_sfi) {
+		int index;
+
+		index = enetc_get_free_index(priv);
+		if (index < 0) {
+			NL_SET_ERR_MSG_MOD(extack, "No Stream Filter resource!");
+			err = -ENOSPC;
+			goto free_fmi;
+		}
+
+		sfi->index = index;
+		sfi->handle = index + HANDLE_OFFSET;
+		/* Update the stream filter handle also */
+		filter->sid.handle = sfi->handle;
+		filter->sfi_index = sfi->index;
+		sfi_overwrite = 0;
+	} else {
+		filter->sfi_index = old_sfi->index;
+		filter->sid.handle = old_sfi->handle;
+		sfi_overwrite = 1;
+	}
+
+	err = enetc_psfp_hw_set(priv, &filter->sid,
+				sfi_overwrite ? NULL : sfi, sgi, fmi);
+	if (err)
+		goto free_fmi;
+
+	spin_lock(&epsfp.psfp_lock);
+	if (filter->flags & ENETC_PSFP_FLAGS_FMI) {
+		old_fmi = enetc_get_meter_by_index(filter->fmi_index);
+		if (old_fmi) {
+			fmi->refcount = old_fmi->refcount;
+			refcount_set(&fmi->refcount,
+				     refcount_read(&old_fmi->refcount) + 1);
+			hlist_del(&old_fmi->node);
+			kfree(old_fmi);
+		}
+		hlist_add_head(&fmi->node, &epsfp.psfp_meter_list);
+	}
+
+	/* Remove the old node if exist and update with a new node */
+	old_sgi = enetc_get_gate_by_index(filter->sgi_index);
+	if (old_sgi) {
+		refcount_set(&sgi->refcount,
+			     refcount_read(&old_sgi->refcount) + 1);
+		hlist_del(&old_sgi->node);
+		kfree(old_sgi);
+	}
+
+	hlist_add_head(&sgi->node, &epsfp.psfp_gate_list);
+
+	if (!old_sfi) {
+		hlist_add_head(&sfi->node, &epsfp.psfp_filter_list);
+		set_bit(sfi->index, epsfp.psfp_sfi_bitmap);
+	} else {
+		kfree(sfi);
+		refcount_inc(&old_sfi->refcount);
+	}
+
+	old_filter = enetc_get_stream_by_index(filter->sid.index);
+	if (old_filter)
+		remove_one_chain(priv, old_filter);
+
+	filter->stats.lastused = jiffies;
+	hlist_add_head(&filter->node, &epsfp.stream_list);
+
+	spin_unlock(&epsfp.psfp_lock);
+
+	return 0;
+
+free_fmi:
+	kfree(fmi);
+free_sfi:
+	kfree(sfi);
+free_gate:
+	kfree(sgi);
+free_filter:
+	kfree(filter);
+
+	return err;
+}
+
+static int enetc_config_clsflower(struct enetc_ndev_priv *priv,
+				  struct flow_cls_offload *cls_flower)
+{
+	struct flow_rule *rule = flow_cls_offload_flow_rule(cls_flower);
+	struct netlink_ext_ack *extack = cls_flower->common.extack;
+	struct flow_dissector *dissector = rule->match.dissector;
+	struct flow_action *action = &rule->action;
+	struct flow_action_entry *entry;
+	struct actions_fwd *fwd;
+	u64 actions = 0;
+	int i, err;
+
+	if (!flow_action_has_entries(action)) {
+		NL_SET_ERR_MSG_MOD(extack, "At least one action is needed");
+		return -EINVAL;
+	}
+
+	flow_action_for_each(i, entry, action)
+		actions |= BIT(entry->id);
+
+	fwd = enetc_check_flow_actions(actions, dissector->used_keys);
+	if (!fwd) {
+		NL_SET_ERR_MSG_MOD(extack, "Unsupported filter type!");
+		return -EOPNOTSUPP;
+	}
+
+	if (fwd->output & FILTER_ACTION_TYPE_PSFP) {
+		err = enetc_psfp_parse_clsflower(priv, cls_flower);
+		if (err) {
+			NL_SET_ERR_MSG_MOD(extack, "Invalid PSFP inputs");
+			return err;
+		}
+	} else {
+		NL_SET_ERR_MSG_MOD(extack, "Unsupported actions");
+		return -EOPNOTSUPP;
+	}
+
+	return 0;
+}
+
+static int enetc_psfp_destroy_clsflower(struct enetc_ndev_priv *priv,
+					struct flow_cls_offload *f)
+{
+	struct enetc_stream_filter *filter;
+	struct netlink_ext_ack *extack = f->common.extack;
+	int err;
+
+	if (f->common.chain_index >= priv->psfp_cap.max_streamid) {
+		NL_SET_ERR_MSG_MOD(extack, "No Stream identify resource!");
+		return -ENOSPC;
+	}
+
+	filter = enetc_get_stream_by_index(f->common.chain_index);
+	if (!filter)
+		return -EINVAL;
+
+	err = enetc_streamid_hw_set(priv, &filter->sid, false);
+	if (err)
+		return err;
+
+	remove_one_chain(priv, filter);
+
+	return 0;
+}
+
+static int enetc_destroy_clsflower(struct enetc_ndev_priv *priv,
+				   struct flow_cls_offload *f)
+{
+	return enetc_psfp_destroy_clsflower(priv, f);
+}
+
+static int enetc_psfp_get_stats(struct enetc_ndev_priv *priv,
+				struct flow_cls_offload *f)
+{
+	struct psfp_streamfilter_counters counters = {};
+	struct enetc_stream_filter *filter;
+	struct flow_stats stats = {};
+	int err;
+
+	filter = enetc_get_stream_by_index(f->common.chain_index);
+	if (!filter)
+		return -EINVAL;
+
+	err = enetc_streamcounter_hw_get(priv, filter->sfi_index, &counters);
+	if (err)
+		return -EINVAL;
+
+	spin_lock(&epsfp.psfp_lock);
+	stats.pkts = counters.matching_frames_count +
+		     counters.not_passing_sdu_count -
+		     filter->stats.pkts;
+	stats.drops = counters.not_passing_frames_count +
+		      counters.not_passing_sdu_count +
+		      counters.red_frames_count -
+		      filter->stats.drops;
+	stats.lastused = filter->stats.lastused;
+	filter->stats.pkts += stats.pkts;
+	filter->stats.drops += stats.drops;
+	spin_unlock(&epsfp.psfp_lock);
+
+	flow_stats_update(&f->stats, 0x0, stats.pkts, stats.drops,
+			  stats.lastused, FLOW_ACTION_HW_STATS_DELAYED);
+
+	return 0;
+}
+
+static int enetc_setup_tc_cls_flower(struct enetc_ndev_priv *priv,
+				     struct flow_cls_offload *cls_flower)
+{
+	switch (cls_flower->command) {
+	case FLOW_CLS_REPLACE:
+		return enetc_config_clsflower(priv, cls_flower);
+	case FLOW_CLS_DESTROY:
+		return enetc_destroy_clsflower(priv, cls_flower);
+	case FLOW_CLS_STATS:
+		return enetc_psfp_get_stats(priv, cls_flower);
+	default:
+		return -EOPNOTSUPP;
+	}
+}
+
+static inline void clean_psfp_sfi_bitmap(void)
+{
+	bitmap_free(epsfp.psfp_sfi_bitmap);
+	epsfp.psfp_sfi_bitmap = NULL;
+}
+
+static void clean_stream_list(void)
+{
+	struct enetc_stream_filter *s;
+	struct hlist_node *tmp;
+
+	hlist_for_each_entry_safe(s, tmp, &epsfp.stream_list, node) {
+		hlist_del(&s->node);
+		kfree(s);
+	}
+}
+
+static void clean_sfi_list(void)
+{
+	struct enetc_psfp_filter *sfi;
+	struct hlist_node *tmp;
+
+	hlist_for_each_entry_safe(sfi, tmp, &epsfp.psfp_filter_list, node) {
+		hlist_del(&sfi->node);
+		kfree(sfi);
+	}
+}
+
+static void clean_sgi_list(void)
+{
+	struct enetc_psfp_gate *sgi;
+	struct hlist_node *tmp;
+
+	hlist_for_each_entry_safe(sgi, tmp, &epsfp.psfp_gate_list, node) {
+		hlist_del(&sgi->node);
+		kfree(sgi);
+	}
+}
+
+static void clean_psfp_all(void)
+{
+	/* Disable all list nodes and free all memory */
+	clean_sfi_list();
+	clean_sgi_list();
+	clean_stream_list();
+	epsfp.dev_bitmap = 0;
+	clean_psfp_sfi_bitmap();
+}
+
+int enetc_setup_tc_block_cb(enum tc_setup_type type, void *type_data,
+			    void *cb_priv)
+{
+	struct net_device *ndev = cb_priv;
+
+	if (!tc_can_offload(ndev))
+		return -EOPNOTSUPP;
+
+	switch (type) {
+	case TC_SETUP_CLSFLOWER:
+		return enetc_setup_tc_cls_flower(netdev_priv(ndev), type_data);
+	default:
+		return -EOPNOTSUPP;
+	}
+}
+
+int enetc_set_psfp(struct net_device *ndev, bool en)
+{
+	struct enetc_ndev_priv *priv = netdev_priv(ndev);
+	int err;
+
+	if (en) {
+		err = enetc_psfp_enable(priv);
+		if (err)
+			return err;
+
+		priv->active_offloads |= ENETC_F_QCI;
+		return 0;
+	}
+
+	err = enetc_psfp_disable(priv);
+	if (err)
+		return err;
+
+	priv->active_offloads &= ~ENETC_F_QCI;
+
+	return 0;
+}
+
+int enetc_psfp_init(struct enetc_ndev_priv *priv)
+{
+	if (epsfp.psfp_sfi_bitmap)
+		return 0;
+
+	epsfp.psfp_sfi_bitmap = bitmap_zalloc(priv->psfp_cap.max_psfp_filter,
+					      GFP_KERNEL);
+	if (!epsfp.psfp_sfi_bitmap)
+		return -ENOMEM;
+
+	spin_lock_init(&epsfp.psfp_lock);
+
+	if (list_empty(&enetc_block_cb_list))
+		epsfp.dev_bitmap = 0;
+
+	return 0;
+}
+
+int enetc_psfp_clean(struct enetc_ndev_priv *priv)
+{
+	if (!list_empty(&enetc_block_cb_list))
+		return -EBUSY;
+
+	clean_psfp_all();
+
+	return 0;
+}
+
+int enetc_setup_tc_psfp(struct net_device *ndev, void *type_data)
+{
+	struct enetc_ndev_priv *priv = netdev_priv(ndev);
+	struct flow_block_offload *f = type_data;
+	int port, err;
+
+	err = flow_block_cb_setup_simple(f, &enetc_block_cb_list,
+					 enetc_setup_tc_block_cb,
+					 ndev, ndev, true);
+	if (err)
+		return err;
+
+	switch (f->command) {
+	case FLOW_BLOCK_BIND:
+		port = enetc_pf_to_port(priv->si->pdev);
+		if (port < 0)
+			return -EINVAL;
+
+		set_bit(port, &epsfp.dev_bitmap);
+		break;
+	case FLOW_BLOCK_UNBIND:
+		port = enetc_pf_to_port(priv->si->pdev);
+		if (port < 0)
+			return -EINVAL;
+
+		clear_bit(port, &epsfp.dev_bitmap);
+		if (!epsfp.dev_bitmap)
+			clean_psfp_all();
+		break;
+	}
+
+	return 0;
+}
+
+int enetc_qos_query_caps(struct net_device *ndev, void *type_data)
+{
+	struct enetc_ndev_priv *priv = netdev_priv(ndev);
+	struct tc_query_caps_base *base = type_data;
+	struct enetc_si *si = priv->si;
+
+	switch (base->type) {
+	case TC_SETUP_QDISC_MQPRIO: {
+		struct tc_mqprio_caps *caps = base->caps;
+
+		caps->validate_queue_counts = true;
+
+		return 0;
+	}
+	case TC_SETUP_QDISC_TAPRIO: {
+		struct tc_taprio_caps *caps = base->caps;
+
+		if (si->hw_features & ENETC_SI_F_QBV)
+			caps->supports_queue_max_sdu = true;
+
+		return 0;
+	}
+	default:
+		return -EOPNOTSUPP;
+	}
+}
diff --git a/drivers/net/ethernet/freescale/enetc/enetc_vf.c b/drivers/net/ethernet/freescale/enetc/enetc_vf.c
new file mode 100644
index 000000000000..6c4b374bcb0e
--- /dev/null
+++ b/drivers/net/ethernet/freescale/enetc/enetc_vf.c
@@ -0,0 +1,291 @@
+// SPDX-License-Identifier: (GPL-2.0+ OR BSD-3-Clause)
+/* Copyright 2017-2019 NXP */
+
+#include <linux/module.h>
+#include "enetc.h"
+
+#define ENETC_DRV_NAME_STR "ENETC VF driver"
+
+/* Messaging */
+static void enetc_msg_vsi_write_msg(struct enetc_hw *hw,
+				    struct enetc_msg_swbd *msg)
+{
+	u32 val;
+
+	val = enetc_vsi_set_msize(msg->size) | lower_32_bits(msg->dma);
+	enetc_wr(hw, ENETC_VSIMSGSNDAR1, upper_32_bits(msg->dma));
+	enetc_wr(hw, ENETC_VSIMSGSNDAR0, val);
+}
+
+static int enetc_msg_vsi_send(struct enetc_si *si, struct enetc_msg_swbd *msg)
+{
+	int timeout = 100;
+	u32 vsimsgsr;
+
+	enetc_msg_vsi_write_msg(&si->hw, msg);
+
+	do {
+		vsimsgsr = enetc_rd(&si->hw, ENETC_VSIMSGSR);
+		if (!(vsimsgsr & ENETC_VSIMSGSR_MB))
+			break;
+
+		usleep_range(1000, 2000);
+	} while (--timeout);
+
+	if (!timeout)
+		return -ETIMEDOUT;
+
+	/* check for message delivery error */
+	if (vsimsgsr & ENETC_VSIMSGSR_MS) {
+		dev_err(&si->pdev->dev, "VSI command execute error: %d\n",
+			ENETC_SIMSGSR_GET_MC(vsimsgsr));
+		return -EIO;
+	}
+
+	return 0;
+}
+
+static int enetc_msg_vsi_set_primary_mac_addr(struct enetc_ndev_priv *priv,
+					      struct sockaddr *saddr)
+{
+	struct enetc_msg_cmd_set_primary_mac *cmd;
+	struct enetc_msg_swbd msg;
+	int err;
+
+	msg.size = ALIGN(sizeof(struct enetc_msg_cmd_set_primary_mac), 64);
+	msg.vaddr = dma_alloc_coherent(priv->dev, msg.size, &msg.dma,
+				       GFP_KERNEL);
+	if (!msg.vaddr) {
+		dev_err(priv->dev, "Failed to alloc Tx msg (size: %d)\n",
+			msg.size);
+		return -ENOMEM;
+	}
+
+	cmd = (struct enetc_msg_cmd_set_primary_mac *)msg.vaddr;
+	cmd->header.type = ENETC_MSG_CMD_MNG_MAC;
+	cmd->header.id = ENETC_MSG_CMD_MNG_ADD;
+	memcpy(&cmd->mac, saddr, sizeof(struct sockaddr));
+
+	/* send the command and wait */
+	err = enetc_msg_vsi_send(priv->si, &msg);
+
+	dma_free_coherent(priv->dev, msg.size, msg.vaddr, msg.dma);
+
+	return err;
+}
+
+static int enetc_vf_set_mac_addr(struct net_device *ndev, void *addr)
+{
+	struct enetc_ndev_priv *priv = netdev_priv(ndev);
+	struct sockaddr *saddr = addr;
+	int err;
+
+	if (!is_valid_ether_addr(saddr->sa_data))
+		return -EADDRNOTAVAIL;
+
+	err = enetc_msg_vsi_set_primary_mac_addr(priv, saddr);
+	if (err)
+		return err;
+
+	eth_hw_addr_set(ndev, saddr->sa_data);
+
+	return 0;
+}
+
+static int enetc_vf_set_features(struct net_device *ndev,
+				 netdev_features_t features)
+{
+	enetc_set_features(ndev, features);
+
+	return 0;
+}
+
+static int enetc_vf_setup_tc(struct net_device *ndev, enum tc_setup_type type,
+			     void *type_data)
+{
+	switch (type) {
+	case TC_SETUP_QDISC_MQPRIO:
+		return enetc_setup_tc_mqprio(ndev, type_data);
+	default:
+		return -EOPNOTSUPP;
+	}
+}
+
+/* Probing/ Init */
+static const struct net_device_ops enetc_ndev_ops = {
+	.ndo_open		= enetc_open,
+	.ndo_stop		= enetc_close,
+	.ndo_start_xmit		= enetc_xmit,
+	.ndo_get_stats		= enetc_get_stats,
+	.ndo_set_mac_address	= enetc_vf_set_mac_addr,
+	.ndo_set_features	= enetc_vf_set_features,
+	.ndo_eth_ioctl		= enetc_ioctl,
+	.ndo_setup_tc		= enetc_vf_setup_tc,
+	.ndo_hwtstamp_get	= enetc_hwtstamp_get,
+	.ndo_hwtstamp_set	= enetc_hwtstamp_set,
+};
+
+static void enetc_vf_netdev_setup(struct enetc_si *si, struct net_device *ndev,
+				  const struct net_device_ops *ndev_ops)
+{
+	struct enetc_ndev_priv *priv = netdev_priv(ndev);
+
+	SET_NETDEV_DEV(ndev, &si->pdev->dev);
+	priv->ndev = ndev;
+	priv->si = si;
+	priv->dev = &si->pdev->dev;
+	si->ndev = ndev;
+
+	priv->msg_enable = (NETIF_MSG_IFUP << 1) - 1;
+	priv->sysclk_freq = si->drvdata->sysclk_freq;
+	priv->max_frags = si->drvdata->max_frags;
+	ndev->netdev_ops = ndev_ops;
+	enetc_set_ethtool_ops(ndev);
+	ndev->watchdog_timeo = 5 * HZ;
+	ndev->max_mtu = ENETC_MAX_MTU;
+
+	ndev->hw_features = NETIF_F_SG | NETIF_F_RXCSUM |
+			    NETIF_F_HW_VLAN_CTAG_TX |
+			    NETIF_F_HW_VLAN_CTAG_RX |
+			    NETIF_F_HW_CSUM | NETIF_F_TSO | NETIF_F_TSO6 |
+			    NETIF_F_GSO_UDP_L4;
+	ndev->features = NETIF_F_HIGHDMA | NETIF_F_SG | NETIF_F_RXCSUM |
+			 NETIF_F_HW_VLAN_CTAG_TX |
+			 NETIF_F_HW_VLAN_CTAG_RX |
+			 NETIF_F_HW_CSUM | NETIF_F_TSO | NETIF_F_TSO6 |
+			 NETIF_F_GSO_UDP_L4;
+	ndev->vlan_features = NETIF_F_SG | NETIF_F_HW_CSUM |
+			      NETIF_F_TSO | NETIF_F_TSO6;
+
+	if (si->num_rss) {
+		ndev->hw_features |= NETIF_F_RXHASH;
+		ndev->features |= NETIF_F_RXHASH;
+	}
+
+	/* pick up primary MAC address from SI */
+	enetc_load_primary_mac_addr(&si->hw, ndev);
+}
+
+static const struct enetc_si_ops enetc_vsi_ops = {
+	.get_rss_table = enetc_get_rss_table,
+	.set_rss_table = enetc_set_rss_table,
+};
+
+static int enetc_vf_probe(struct pci_dev *pdev,
+			  const struct pci_device_id *ent)
+{
+	struct enetc_ndev_priv *priv;
+	struct net_device *ndev;
+	struct enetc_si *si;
+	int err;
+
+	err = enetc_pci_probe(pdev, KBUILD_MODNAME, 0);
+	if (err)
+		return dev_err_probe(&pdev->dev, err, "PCI probing failed\n");
+
+	si = pci_get_drvdata(pdev);
+	si->revision = ENETC_REV_1_0;
+	si->ops = &enetc_vsi_ops;
+	err = enetc_get_driver_data(si);
+	if (err) {
+		dev_err_probe(&pdev->dev, err,
+			      "Could not get VF driver data\n");
+		goto err_alloc_netdev;
+	}
+
+	enetc_get_si_caps(si);
+
+	ndev = alloc_etherdev_mq(sizeof(*priv), ENETC_MAX_NUM_TXQS);
+	if (!ndev) {
+		err = -ENOMEM;
+		dev_err(&pdev->dev, "netdev creation failed\n");
+		goto err_alloc_netdev;
+	}
+
+	enetc_vf_netdev_setup(si, ndev, &enetc_ndev_ops);
+
+	priv = netdev_priv(ndev);
+
+	enetc_init_si_rings_params(priv);
+
+	err = enetc_setup_cbdr(priv->dev, &si->hw, ENETC_CBDR_DEFAULT_SIZE,
+			       &si->cbd_ring);
+	if (err)
+		goto err_setup_cbdr;
+
+	err = enetc_alloc_si_resources(priv);
+	if (err) {
+		dev_err(&pdev->dev, "SI resource alloc failed\n");
+		goto err_alloc_si_res;
+	}
+
+	err = enetc_configure_si(priv);
+	if (err) {
+		dev_err(&pdev->dev, "Failed to configure SI\n");
+		goto err_config_si;
+	}
+
+	err = enetc_alloc_msix(priv);
+	if (err) {
+		dev_err(&pdev->dev, "MSIX alloc failed\n");
+		goto err_alloc_msix;
+	}
+
+	err = register_netdev(ndev);
+	if (err)
+		goto err_reg_netdev;
+
+	netif_carrier_off(ndev);
+
+	return 0;
+
+err_reg_netdev:
+	enetc_free_msix(priv);
+err_config_si:
+err_alloc_msix:
+	enetc_free_si_resources(priv);
+err_alloc_si_res:
+	enetc_teardown_cbdr(&si->cbd_ring);
+err_setup_cbdr:
+	si->ndev = NULL;
+	free_netdev(ndev);
+err_alloc_netdev:
+	enetc_pci_remove(pdev);
+
+	return err;
+}
+
+static void enetc_vf_remove(struct pci_dev *pdev)
+{
+	struct enetc_si *si = pci_get_drvdata(pdev);
+	struct enetc_ndev_priv *priv;
+
+	priv = netdev_priv(si->ndev);
+	unregister_netdev(si->ndev);
+
+	enetc_free_msix(priv);
+
+	enetc_free_si_resources(priv);
+	enetc_teardown_cbdr(&si->cbd_ring);
+
+	free_netdev(si->ndev);
+
+	enetc_pci_remove(pdev);
+}
+
+static const struct pci_device_id enetc_vf_id_table[] = {
+	{ PCI_DEVICE(PCI_VENDOR_ID_FREESCALE, ENETC_DEV_ID_VF) },
+	{ 0, } /* End of table. */
+};
+MODULE_DEVICE_TABLE(pci, enetc_vf_id_table);
+
+static struct pci_driver enetc_vf_driver = {
+	.name = KBUILD_MODNAME,
+	.id_table = enetc_vf_id_table,
+	.probe = enetc_vf_probe,
+	.remove = enetc_vf_remove,
+};
+module_pci_driver(enetc_vf_driver);
+
+MODULE_DESCRIPTION(ENETC_DRV_NAME_STR);
+MODULE_LICENSE("Dual BSD/GPL");
diff --git a/drivers/net/ethernet/freescale/enetc/netc_blk_ctrl.c b/drivers/net/ethernet/freescale/enetc/netc_blk_ctrl.c
new file mode 100644
index 000000000000..bcb8eefeb93c
--- /dev/null
+++ b/drivers/net/ethernet/freescale/enetc/netc_blk_ctrl.c
@@ -0,0 +1,445 @@
+// SPDX-License-Identifier: (GPL-2.0+ OR BSD-3-Clause)
+/*
+ * NXP NETC Blocks Control Driver
+ *
+ * Copyright 2024 NXP
+ *
+ * This driver is used for pre-initialization of NETC, such as PCS and MII
+ * protocols, LDID, warm reset, etc. Therefore, all NETC device drivers can
+ * only be probed after the netc-blk-crtl driver has completed initialization.
+ * In addition, when the system enters suspend mode, IERB, PRB, and NETCMIX
+ * will be powered off, except for WOL. Therefore, when the system resumes,
+ * these blocks need to be reinitialized.
+ */
+
+#include <linux/bits.h>
+#include <linux/clk.h>
+#include <linux/debugfs.h>
+#include <linux/delay.h>
+#include <linux/fsl/netc_global.h>
+#include <linux/module.h>
+#include <linux/of.h>
+#include <linux/of_device.h>
+#include <linux/of_net.h>
+#include <linux/of_platform.h>
+#include <linux/phy.h>
+#include <linux/platform_device.h>
+#include <linux/seq_file.h>
+
+/* NETCMIX registers */
+#define IMX95_CFG_LINK_IO_VAR		0x0
+#define  IO_VAR_16FF_16G_SERDES		0x1
+#define  IO_VAR(port, var)		(((var) & 0xf) << ((port) << 2))
+
+#define IMX95_CFG_LINK_MII_PROT		0x4
+#define CFG_LINK_MII_PORT_0		GENMASK(3, 0)
+#define CFG_LINK_MII_PORT_1		GENMASK(7, 4)
+#define  MII_PROT_MII			0x0
+#define  MII_PROT_RMII			0x1
+#define  MII_PROT_RGMII			0x2
+#define  MII_PROT_SERIAL		0x3
+#define  MII_PROT(port, prot)		(((prot) & 0xf) << ((port) << 2))
+
+#define IMX95_CFG_LINK_PCS_PROT(a)	(0x8 + (a) * 4)
+#define PCS_PROT_1G_SGMII		BIT(0)
+#define PCS_PROT_2500M_SGMII		BIT(1)
+#define PCS_PROT_XFI			BIT(3)
+#define PCS_PROT_SFI			BIT(4)
+#define PCS_PROT_10G_SXGMII		BIT(6)
+
+/* NETC privileged register block register */
+#define PRB_NETCRR			0x100
+#define  NETCRR_SR			BIT(0)
+#define  NETCRR_LOCK			BIT(1)
+
+#define PRB_NETCSR			0x104
+#define  NETCSR_ERROR			BIT(0)
+#define  NETCSR_STATE			BIT(1)
+
+/* NETC integrated endpoint register block register */
+#define IERB_EMDIOFAUXR			0x344
+#define IERB_T0FAUXR			0x444
+#define IERB_EFAUXR(a)			(0x3044 + 0x100 * (a))
+#define IERB_VFAUXR(a)			(0x4004 + 0x40 * (a))
+#define FAUXR_LDID			GENMASK(3, 0)
+
+/* Platform information */
+#define IMX95_ENETC0_BUS_DEVFN		0x0
+#define IMX95_ENETC1_BUS_DEVFN		0x40
+#define IMX95_ENETC2_BUS_DEVFN		0x80
+
+/* Flags for different platforms */
+#define NETC_HAS_NETCMIX		BIT(0)
+
+struct netc_devinfo {
+	u32 flags;
+	int (*netcmix_init)(struct platform_device *pdev);
+	int (*ierb_init)(struct platform_device *pdev);
+};
+
+struct netc_blk_ctrl {
+	void __iomem *prb;
+	void __iomem *ierb;
+	void __iomem *netcmix;
+
+	const struct netc_devinfo *devinfo;
+	struct platform_device *pdev;
+	struct dentry *debugfs_root;
+};
+
+static void netc_reg_write(void __iomem *base, u32 offset, u32 val)
+{
+	netc_write(base + offset, val);
+}
+
+static u32 netc_reg_read(void __iomem *base, u32 offset)
+{
+	return netc_read(base + offset);
+}
+
+static int netc_of_pci_get_bus_devfn(struct device_node *np)
+{
+	u32 reg[5];
+	int error;
+
+	error = of_property_read_u32_array(np, "reg", reg, ARRAY_SIZE(reg));
+	if (error)
+		return error;
+
+	return (reg[0] >> 8) & 0xffff;
+}
+
+static int netc_get_link_mii_protocol(phy_interface_t interface)
+{
+	switch (interface) {
+	case PHY_INTERFACE_MODE_MII:
+		return MII_PROT_MII;
+	case PHY_INTERFACE_MODE_RMII:
+		return MII_PROT_RMII;
+	case PHY_INTERFACE_MODE_RGMII:
+	case PHY_INTERFACE_MODE_RGMII_ID:
+	case PHY_INTERFACE_MODE_RGMII_RXID:
+	case PHY_INTERFACE_MODE_RGMII_TXID:
+		return MII_PROT_RGMII;
+	case PHY_INTERFACE_MODE_SGMII:
+	case PHY_INTERFACE_MODE_2500BASEX:
+	case PHY_INTERFACE_MODE_10GBASER:
+	case PHY_INTERFACE_MODE_XGMII:
+	case PHY_INTERFACE_MODE_USXGMII:
+		return MII_PROT_SERIAL;
+	default:
+		return -EINVAL;
+	}
+}
+
+static int imx95_netcmix_init(struct platform_device *pdev)
+{
+	struct netc_blk_ctrl *priv = platform_get_drvdata(pdev);
+	struct device_node *np = pdev->dev.of_node;
+	phy_interface_t interface;
+	int bus_devfn, mii_proto;
+	u32 val;
+	int err;
+
+	/* Default setting of MII protocol */
+	val = MII_PROT(0, MII_PROT_RGMII) | MII_PROT(1, MII_PROT_RGMII) |
+	      MII_PROT(2, MII_PROT_SERIAL);
+
+	/* Update the link MII protocol through parsing phy-mode */
+	for_each_available_child_of_node_scoped(np, child) {
+		for_each_available_child_of_node_scoped(child, gchild) {
+			if (!of_device_is_compatible(gchild, "pci1131,e101"))
+				continue;
+
+			bus_devfn = netc_of_pci_get_bus_devfn(gchild);
+			if (bus_devfn < 0)
+				return -EINVAL;
+
+			if (bus_devfn == IMX95_ENETC2_BUS_DEVFN)
+				continue;
+
+			err = of_get_phy_mode(gchild, &interface);
+			if (err)
+				continue;
+
+			mii_proto = netc_get_link_mii_protocol(interface);
+			if (mii_proto < 0)
+				return -EINVAL;
+
+			switch (bus_devfn) {
+			case IMX95_ENETC0_BUS_DEVFN:
+				val = u32_replace_bits(val, mii_proto,
+						       CFG_LINK_MII_PORT_0);
+				break;
+			case IMX95_ENETC1_BUS_DEVFN:
+				val = u32_replace_bits(val, mii_proto,
+						       CFG_LINK_MII_PORT_1);
+				break;
+			default:
+				return -EINVAL;
+			}
+		}
+	}
+
+	/* Configure Link I/O variant */
+	netc_reg_write(priv->netcmix, IMX95_CFG_LINK_IO_VAR,
+		       IO_VAR(2, IO_VAR_16FF_16G_SERDES));
+	/* Configure Link 2 PCS protocol */
+	netc_reg_write(priv->netcmix, IMX95_CFG_LINK_PCS_PROT(2),
+		       PCS_PROT_10G_SXGMII);
+	netc_reg_write(priv->netcmix, IMX95_CFG_LINK_MII_PROT, val);
+
+	return 0;
+}
+
+static bool netc_ierb_is_locked(struct netc_blk_ctrl *priv)
+{
+	return !!(netc_reg_read(priv->prb, PRB_NETCRR) & NETCRR_LOCK);
+}
+
+static int netc_lock_ierb(struct netc_blk_ctrl *priv)
+{
+	u32 val;
+
+	netc_reg_write(priv->prb, PRB_NETCRR, NETCRR_LOCK);
+
+	return read_poll_timeout(netc_reg_read, val, !(val & NETCSR_STATE),
+				 100, 2000, false, priv->prb, PRB_NETCSR);
+}
+
+static int netc_unlock_ierb_with_warm_reset(struct netc_blk_ctrl *priv)
+{
+	u32 val;
+
+	netc_reg_write(priv->prb, PRB_NETCRR, 0);
+
+	return read_poll_timeout(netc_reg_read, val, !(val & NETCRR_LOCK),
+				 1000, 100000, true, priv->prb, PRB_NETCRR);
+}
+
+static int imx95_ierb_init(struct platform_device *pdev)
+{
+	struct netc_blk_ctrl *priv = platform_get_drvdata(pdev);
+
+	/* EMDIO : No MSI-X intterupt */
+	netc_reg_write(priv->ierb, IERB_EMDIOFAUXR, 0);
+	/* ENETC0 PF */
+	netc_reg_write(priv->ierb, IERB_EFAUXR(0), 0);
+	/* ENETC0 VF0 */
+	netc_reg_write(priv->ierb, IERB_VFAUXR(0), 1);
+	/* ENETC0 VF1 */
+	netc_reg_write(priv->ierb, IERB_VFAUXR(1), 2);
+	/* ENETC1 PF */
+	netc_reg_write(priv->ierb, IERB_EFAUXR(1), 3);
+	/* ENETC1 VF0 */
+	netc_reg_write(priv->ierb, IERB_VFAUXR(2), 5);
+	/* ENETC1 VF1 */
+	netc_reg_write(priv->ierb, IERB_VFAUXR(3), 6);
+	/* ENETC2 PF */
+	netc_reg_write(priv->ierb, IERB_EFAUXR(2), 4);
+	/* ENETC2 VF0 */
+	netc_reg_write(priv->ierb, IERB_VFAUXR(4), 5);
+	/* ENETC2 VF1 */
+	netc_reg_write(priv->ierb, IERB_VFAUXR(5), 6);
+	/* NETC TIMER */
+	netc_reg_write(priv->ierb, IERB_T0FAUXR, 7);
+
+	return 0;
+}
+
+static int netc_ierb_init(struct platform_device *pdev)
+{
+	struct netc_blk_ctrl *priv = platform_get_drvdata(pdev);
+	const struct netc_devinfo *devinfo = priv->devinfo;
+	int err;
+
+	if (netc_ierb_is_locked(priv)) {
+		err = netc_unlock_ierb_with_warm_reset(priv);
+		if (err) {
+			dev_err(&pdev->dev, "Unlock IERB failed.\n");
+			return err;
+		}
+	}
+
+	if (devinfo->ierb_init) {
+		err = devinfo->ierb_init(pdev);
+		if (err)
+			return err;
+	}
+
+	err = netc_lock_ierb(priv);
+	if (err) {
+		dev_err(&pdev->dev, "Lock IERB failed.\n");
+		return err;
+	}
+
+	return 0;
+}
+
+#if IS_ENABLED(CONFIG_DEBUG_FS)
+static int netc_prb_show(struct seq_file *s, void *data)
+{
+	struct netc_blk_ctrl *priv = s->private;
+	u32 val;
+
+	val = netc_reg_read(priv->prb, PRB_NETCRR);
+	seq_printf(s, "[PRB NETCRR] Lock:%d SR:%d\n",
+		   (val & NETCRR_LOCK) ? 1 : 0,
+		   (val & NETCRR_SR) ? 1 : 0);
+
+	val = netc_reg_read(priv->prb, PRB_NETCSR);
+	seq_printf(s, "[PRB NETCSR] State:%d Error:%d\n",
+		   (val & NETCSR_STATE) ? 1 : 0,
+		   (val & NETCSR_ERROR) ? 1 : 0);
+
+	return 0;
+}
+DEFINE_SHOW_ATTRIBUTE(netc_prb);
+
+static void netc_blk_ctrl_create_debugfs(struct netc_blk_ctrl *priv)
+{
+	struct dentry *root;
+
+	root = debugfs_create_dir("netc_blk_ctrl", NULL);
+	if (IS_ERR(root))
+		return;
+
+	priv->debugfs_root = root;
+
+	debugfs_create_file("prb", 0444, root, priv, &netc_prb_fops);
+}
+
+static void netc_blk_ctrl_remove_debugfs(struct netc_blk_ctrl *priv)
+{
+	debugfs_remove_recursive(priv->debugfs_root);
+	priv->debugfs_root = NULL;
+}
+
+#else
+
+static void netc_blk_ctrl_create_debugfs(struct netc_blk_ctrl *priv)
+{
+}
+
+static void netc_blk_ctrl_remove_debugfs(struct netc_blk_ctrl *priv)
+{
+}
+#endif
+
+static int netc_prb_check_error(struct netc_blk_ctrl *priv)
+{
+	if (netc_reg_read(priv->prb, PRB_NETCSR) & NETCSR_ERROR)
+		return -1;
+
+	return 0;
+}
+
+static const struct netc_devinfo imx95_devinfo = {
+	.flags = NETC_HAS_NETCMIX,
+	.netcmix_init = imx95_netcmix_init,
+	.ierb_init = imx95_ierb_init,
+};
+
+static const struct of_device_id netc_blk_ctrl_match[] = {
+	{ .compatible = "nxp,imx95-netc-blk-ctrl", .data = &imx95_devinfo },
+	{},
+};
+MODULE_DEVICE_TABLE(of, netc_blk_ctrl_match);
+
+static int netc_blk_ctrl_probe(struct platform_device *pdev)
+{
+	struct device_node *node = pdev->dev.of_node;
+	const struct netc_devinfo *devinfo;
+	struct device *dev = &pdev->dev;
+	const struct of_device_id *id;
+	struct netc_blk_ctrl *priv;
+	struct clk *ipg_clk;
+	void __iomem *regs;
+	int err;
+
+	priv = devm_kzalloc(dev, sizeof(*priv), GFP_KERNEL);
+	if (!priv)
+		return -ENOMEM;
+
+	priv->pdev = pdev;
+	ipg_clk = devm_clk_get_optional_enabled(dev, "ipg");
+	if (IS_ERR(ipg_clk))
+		return dev_err_probe(dev, PTR_ERR(ipg_clk),
+				     "Set ipg clock failed\n");
+
+	id = of_match_device(netc_blk_ctrl_match, dev);
+	if (!id)
+		return dev_err_probe(dev, -EINVAL, "Cannot match device\n");
+
+	devinfo = (struct netc_devinfo *)id->data;
+	if (!devinfo)
+		return dev_err_probe(dev, -EINVAL, "No device information\n");
+
+	priv->devinfo = devinfo;
+	regs = devm_platform_ioremap_resource_byname(pdev, "ierb");
+	if (IS_ERR(regs))
+		return dev_err_probe(dev, PTR_ERR(regs),
+				     "Missing IERB resource\n");
+
+	priv->ierb = regs;
+	regs = devm_platform_ioremap_resource_byname(pdev, "prb");
+	if (IS_ERR(regs))
+		return dev_err_probe(dev, PTR_ERR(regs),
+				     "Missing PRB resource\n");
+
+	priv->prb = regs;
+	if (devinfo->flags & NETC_HAS_NETCMIX) {
+		regs = devm_platform_ioremap_resource_byname(pdev, "netcmix");
+		if (IS_ERR(regs))
+			return dev_err_probe(dev, PTR_ERR(regs),
+					     "Missing NETCMIX resource\n");
+		priv->netcmix = regs;
+	}
+
+	platform_set_drvdata(pdev, priv);
+	if (devinfo->netcmix_init) {
+		err = devinfo->netcmix_init(pdev);
+		if (err)
+			return dev_err_probe(dev, err,
+					     "Initializing NETCMIX failed\n");
+	}
+
+	err = netc_ierb_init(pdev);
+	if (err)
+		return dev_err_probe(dev, err, "Initializing IERB failed\n");
+
+	if (netc_prb_check_error(priv) < 0)
+		dev_warn(dev, "The current IERB configuration is invalid\n");
+
+	netc_blk_ctrl_create_debugfs(priv);
+
+	err = of_platform_populate(node, NULL, NULL, dev);
+	if (err) {
+		netc_blk_ctrl_remove_debugfs(priv);
+		return dev_err_probe(dev, err, "of_platform_populate failed\n");
+	}
+
+	return 0;
+}
+
+static void netc_blk_ctrl_remove(struct platform_device *pdev)
+{
+	struct netc_blk_ctrl *priv = platform_get_drvdata(pdev);
+
+	of_platform_depopulate(&pdev->dev);
+	netc_blk_ctrl_remove_debugfs(priv);
+}
+
+static struct platform_driver netc_blk_ctrl_driver = {
+	.driver = {
+		.name = "nxp-netc-blk-ctrl",
+		.of_match_table = netc_blk_ctrl_match,
+	},
+	.probe = netc_blk_ctrl_probe,
+	.remove = netc_blk_ctrl_remove,
+};
+
+module_platform_driver(netc_blk_ctrl_driver);
+
+MODULE_DESCRIPTION("NXP NETC Blocks Control Driver");
+MODULE_LICENSE("Dual BSD/GPL");
diff --git a/drivers/net/ethernet/freescale/enetc/ntmp.c b/drivers/net/ethernet/freescale/enetc/ntmp.c
new file mode 100644
index 000000000000..0c1d343253bf
--- /dev/null
+++ b/drivers/net/ethernet/freescale/enetc/ntmp.c
@@ -0,0 +1,457 @@
+// SPDX-License-Identifier: (GPL-2.0+ OR BSD-3-Clause)
+/*
+ * NETC NTMP (NETC Table Management Protocol) 2.0 Library
+ * Copyright 2025 NXP
+ */
+
+#include <linux/dma-mapping.h>
+#include <linux/fsl/netc_global.h>
+#include <linux/iopoll.h>
+
+#include "ntmp_private.h"
+
+#define NETC_CBDR_TIMEOUT		1000 /* us */
+#define NETC_CBDR_DELAY_US		10
+#define NETC_CBDR_MR_EN			BIT(31)
+
+#define NTMP_BASE_ADDR_ALIGN		128
+#define NTMP_DATA_ADDR_ALIGN		32
+
+/* Define NTMP Table ID */
+#define NTMP_MAFT_ID			1
+#define NTMP_RSST_ID			3
+
+/* Generic Update Actions for most tables */
+#define NTMP_GEN_UA_CFGEU		BIT(0)
+#define NTMP_GEN_UA_STSEU		BIT(1)
+
+#define NTMP_ENTRY_ID_SIZE		4
+#define RSST_ENTRY_NUM			64
+#define RSST_STSE_DATA_SIZE(n)		((n) * 8)
+#define RSST_CFGE_DATA_SIZE(n)		(n)
+
+int ntmp_init_cbdr(struct netc_cbdr *cbdr, struct device *dev,
+		   const struct netc_cbdr_regs *regs)
+{
+	int cbd_num = NETC_CBDR_BD_NUM;
+	size_t size;
+
+	size = cbd_num * sizeof(union netc_cbd) + NTMP_BASE_ADDR_ALIGN;
+	cbdr->addr_base = dma_alloc_coherent(dev, size, &cbdr->dma_base,
+					     GFP_KERNEL);
+	if (!cbdr->addr_base)
+		return -ENOMEM;
+
+	cbdr->dma_size = size;
+	cbdr->bd_num = cbd_num;
+	cbdr->regs = *regs;
+	cbdr->dev = dev;
+
+	/* The base address of the Control BD Ring must be 128 bytes aligned */
+	cbdr->dma_base_align =  ALIGN(cbdr->dma_base,  NTMP_BASE_ADDR_ALIGN);
+	cbdr->addr_base_align = PTR_ALIGN(cbdr->addr_base,
+					  NTMP_BASE_ADDR_ALIGN);
+
+	spin_lock_init(&cbdr->ring_lock);
+
+	cbdr->next_to_use = netc_read(cbdr->regs.pir);
+	cbdr->next_to_clean = netc_read(cbdr->regs.cir);
+
+	/* Step 1: Configure the base address of the Control BD Ring */
+	netc_write(cbdr->regs.bar0, lower_32_bits(cbdr->dma_base_align));
+	netc_write(cbdr->regs.bar1, upper_32_bits(cbdr->dma_base_align));
+
+	/* Step 2: Configure the number of BDs of the Control BD Ring */
+	netc_write(cbdr->regs.lenr, cbdr->bd_num);
+
+	/* Step 3: Enable the Control BD Ring */
+	netc_write(cbdr->regs.mr, NETC_CBDR_MR_EN);
+
+	return 0;
+}
+EXPORT_SYMBOL_GPL(ntmp_init_cbdr);
+
+void ntmp_free_cbdr(struct netc_cbdr *cbdr)
+{
+	/* Disable the Control BD Ring */
+	netc_write(cbdr->regs.mr, 0);
+	dma_free_coherent(cbdr->dev, cbdr->dma_size, cbdr->addr_base,
+			  cbdr->dma_base);
+	memset(cbdr, 0, sizeof(*cbdr));
+}
+EXPORT_SYMBOL_GPL(ntmp_free_cbdr);
+
+static int ntmp_get_free_cbd_num(struct netc_cbdr *cbdr)
+{
+	return (cbdr->next_to_clean - cbdr->next_to_use - 1 +
+		cbdr->bd_num) % cbdr->bd_num;
+}
+
+static union netc_cbd *ntmp_get_cbd(struct netc_cbdr *cbdr, int index)
+{
+	return &((union netc_cbd *)(cbdr->addr_base_align))[index];
+}
+
+static void ntmp_clean_cbdr(struct netc_cbdr *cbdr)
+{
+	union netc_cbd *cbd;
+	int i;
+
+	i = cbdr->next_to_clean;
+	while (netc_read(cbdr->regs.cir) != i) {
+		cbd = ntmp_get_cbd(cbdr, i);
+		memset(cbd, 0, sizeof(*cbd));
+		i = (i + 1) % cbdr->bd_num;
+	}
+
+	cbdr->next_to_clean = i;
+}
+
+static int netc_xmit_ntmp_cmd(struct ntmp_user *user, union netc_cbd *cbd)
+{
+	union netc_cbd *cur_cbd;
+	struct netc_cbdr *cbdr;
+	int i, err;
+	u16 status;
+	u32 val;
+
+	/* Currently only i.MX95 ENETC is supported, and it only has one
+	 * command BD ring
+	 */
+	cbdr = &user->ring[0];
+
+	spin_lock_bh(&cbdr->ring_lock);
+
+	if (unlikely(!ntmp_get_free_cbd_num(cbdr)))
+		ntmp_clean_cbdr(cbdr);
+
+	i = cbdr->next_to_use;
+	cur_cbd = ntmp_get_cbd(cbdr, i);
+	*cur_cbd = *cbd;
+	dma_wmb();
+
+	/* Update producer index of both software and hardware */
+	i = (i + 1) % cbdr->bd_num;
+	cbdr->next_to_use = i;
+	netc_write(cbdr->regs.pir, i);
+
+	err = read_poll_timeout_atomic(netc_read, val, val == i,
+				       NETC_CBDR_DELAY_US, NETC_CBDR_TIMEOUT,
+				       true, cbdr->regs.cir);
+	if (unlikely(err))
+		goto cbdr_unlock;
+
+	dma_rmb();
+	/* Get the writeback command BD, because the caller may need
+	 * to check some other fields of the response header.
+	 */
+	*cbd = *cur_cbd;
+
+	/* Check the writeback error status */
+	status = le16_to_cpu(cbd->resp_hdr.error_rr) & NTMP_RESP_ERROR;
+	if (unlikely(status)) {
+		err = -EIO;
+		dev_err(user->dev, "Command BD error: 0x%04x\n", status);
+	}
+
+	ntmp_clean_cbdr(cbdr);
+	dma_wmb();
+
+cbdr_unlock:
+	spin_unlock_bh(&cbdr->ring_lock);
+
+	return err;
+}
+
+static int ntmp_alloc_data_mem(struct ntmp_dma_buf *data, void **buf_align)
+{
+	void *buf;
+
+	buf = dma_alloc_coherent(data->dev, data->size + NTMP_DATA_ADDR_ALIGN,
+				 &data->dma, GFP_KERNEL);
+	if (!buf)
+		return -ENOMEM;
+
+	data->buf = buf;
+	*buf_align = PTR_ALIGN(buf, NTMP_DATA_ADDR_ALIGN);
+
+	return 0;
+}
+
+static void ntmp_free_data_mem(struct ntmp_dma_buf *data)
+{
+	dma_free_coherent(data->dev, data->size + NTMP_DATA_ADDR_ALIGN,
+			  data->buf, data->dma);
+}
+
+static void ntmp_fill_request_hdr(union netc_cbd *cbd, dma_addr_t dma,
+				  int len, int table_id, int cmd,
+				  int access_method)
+{
+	dma_addr_t dma_align;
+
+	memset(cbd, 0, sizeof(*cbd));
+	dma_align = ALIGN(dma, NTMP_DATA_ADDR_ALIGN);
+	cbd->req_hdr.addr = cpu_to_le64(dma_align);
+	cbd->req_hdr.len = cpu_to_le32(len);
+	cbd->req_hdr.cmd = cmd;
+	cbd->req_hdr.access_method = FIELD_PREP(NTMP_ACCESS_METHOD,
+						access_method);
+	cbd->req_hdr.table_id = table_id;
+	cbd->req_hdr.ver_cci_rr = FIELD_PREP(NTMP_HDR_VERSION,
+					     NTMP_HDR_VER2);
+	/* For NTMP version 2.0 or later version */
+	cbd->req_hdr.npf = cpu_to_le32(NTMP_NPF);
+}
+
+static void ntmp_fill_crd(struct ntmp_cmn_req_data *crd, u8 tblv,
+			  u8 qa, u16 ua)
+{
+	crd->update_act = cpu_to_le16(ua);
+	crd->tblv_qact = NTMP_TBLV_QACT(tblv, qa);
+}
+
+static void ntmp_fill_crd_eid(struct ntmp_req_by_eid *rbe, u8 tblv,
+			      u8 qa, u16 ua, u32 entry_id)
+{
+	ntmp_fill_crd(&rbe->crd, tblv, qa, ua);
+	rbe->entry_id = cpu_to_le32(entry_id);
+}
+
+static const char *ntmp_table_name(int tbl_id)
+{
+	switch (tbl_id) {
+	case NTMP_MAFT_ID:
+		return "MAC Address Filter Table";
+	case NTMP_RSST_ID:
+		return "RSS Table";
+	default:
+		return "Unknown Table";
+	};
+}
+
+static int ntmp_delete_entry_by_id(struct ntmp_user *user, int tbl_id,
+				   u8 tbl_ver, u32 entry_id, u32 req_len,
+				   u32 resp_len)
+{
+	struct ntmp_dma_buf data = {
+		.dev = user->dev,
+		.size = max(req_len, resp_len),
+	};
+	struct ntmp_req_by_eid *req;
+	union netc_cbd cbd;
+	int err;
+
+	err = ntmp_alloc_data_mem(&data, (void **)&req);
+	if (err)
+		return err;
+
+	ntmp_fill_crd_eid(req, tbl_ver, 0, 0, entry_id);
+	ntmp_fill_request_hdr(&cbd, data.dma, NTMP_LEN(req_len, resp_len),
+			      tbl_id, NTMP_CMD_DELETE, NTMP_AM_ENTRY_ID);
+
+	err = netc_xmit_ntmp_cmd(user, &cbd);
+	if (err)
+		dev_err(user->dev,
+			"Failed to delete entry 0x%x of %s, err: %pe",
+			entry_id, ntmp_table_name(tbl_id), ERR_PTR(err));
+
+	ntmp_free_data_mem(&data);
+
+	return err;
+}
+
+static int ntmp_query_entry_by_id(struct ntmp_user *user, int tbl_id,
+				  u32 len, struct ntmp_req_by_eid *req,
+				  dma_addr_t dma, bool compare_eid)
+{
+	struct ntmp_cmn_resp_query *resp;
+	int cmd = NTMP_CMD_QUERY;
+	union netc_cbd cbd;
+	u32 entry_id;
+	int err;
+
+	entry_id = le32_to_cpu(req->entry_id);
+	if (le16_to_cpu(req->crd.update_act))
+		cmd = NTMP_CMD_QU;
+
+	/* Request header */
+	ntmp_fill_request_hdr(&cbd, dma, len, tbl_id, cmd, NTMP_AM_ENTRY_ID);
+	err = netc_xmit_ntmp_cmd(user, &cbd);
+	if (err) {
+		dev_err(user->dev,
+			"Failed to query entry 0x%x of %s, err: %pe\n",
+			entry_id, ntmp_table_name(tbl_id), ERR_PTR(err));
+		return err;
+	}
+
+	/* For a few tables, the first field of their response data is not
+	 * entry_id, so directly return success.
+	 */
+	if (!compare_eid)
+		return 0;
+
+	resp = (struct ntmp_cmn_resp_query *)req;
+	if (unlikely(le32_to_cpu(resp->entry_id) != entry_id)) {
+		dev_err(user->dev,
+			"%s: query EID 0x%x doesn't match response EID 0x%x\n",
+			ntmp_table_name(tbl_id), entry_id, le32_to_cpu(resp->entry_id));
+		return -EIO;
+	}
+
+	return 0;
+}
+
+int ntmp_maft_add_entry(struct ntmp_user *user, u32 entry_id,
+			struct maft_entry_data *maft)
+{
+	struct ntmp_dma_buf data = {
+		.dev = user->dev,
+		.size = sizeof(struct maft_req_add),
+	};
+	struct maft_req_add *req;
+	union netc_cbd cbd;
+	int err;
+
+	err = ntmp_alloc_data_mem(&data, (void **)&req);
+	if (err)
+		return err;
+
+	/* Set mac address filter table request data buffer */
+	ntmp_fill_crd_eid(&req->rbe, user->tbl.maft_ver, 0, 0, entry_id);
+	req->keye = maft->keye;
+	req->cfge = maft->cfge;
+
+	ntmp_fill_request_hdr(&cbd, data.dma, NTMP_LEN(data.size, 0),
+			      NTMP_MAFT_ID, NTMP_CMD_ADD, NTMP_AM_ENTRY_ID);
+	err = netc_xmit_ntmp_cmd(user, &cbd);
+	if (err)
+		dev_err(user->dev, "Failed to add MAFT entry 0x%x, err: %pe\n",
+			entry_id, ERR_PTR(err));
+
+	ntmp_free_data_mem(&data);
+
+	return err;
+}
+EXPORT_SYMBOL_GPL(ntmp_maft_add_entry);
+
+int ntmp_maft_query_entry(struct ntmp_user *user, u32 entry_id,
+			  struct maft_entry_data *maft)
+{
+	struct ntmp_dma_buf data = {
+		.dev = user->dev,
+		.size = sizeof(struct maft_resp_query),
+	};
+	struct maft_resp_query *resp;
+	struct ntmp_req_by_eid *req;
+	int err;
+
+	err = ntmp_alloc_data_mem(&data, (void **)&req);
+	if (err)
+		return err;
+
+	ntmp_fill_crd_eid(req, user->tbl.maft_ver, 0, 0, entry_id);
+	err = ntmp_query_entry_by_id(user, NTMP_MAFT_ID,
+				     NTMP_LEN(sizeof(*req), data.size),
+				     req, data.dma, true);
+	if (err)
+		goto end;
+
+	resp = (struct maft_resp_query *)req;
+	maft->keye = resp->keye;
+	maft->cfge = resp->cfge;
+
+end:
+	ntmp_free_data_mem(&data);
+
+	return err;
+}
+EXPORT_SYMBOL_GPL(ntmp_maft_query_entry);
+
+int ntmp_maft_delete_entry(struct ntmp_user *user, u32 entry_id)
+{
+	return ntmp_delete_entry_by_id(user, NTMP_MAFT_ID, user->tbl.maft_ver,
+				       entry_id, NTMP_EID_REQ_LEN, 0);
+}
+EXPORT_SYMBOL_GPL(ntmp_maft_delete_entry);
+
+int ntmp_rsst_update_entry(struct ntmp_user *user, const u32 *table,
+			   int count)
+{
+	struct ntmp_dma_buf data = {.dev = user->dev};
+	struct rsst_req_update *req;
+	union netc_cbd cbd;
+	int err, i;
+
+	if (count != RSST_ENTRY_NUM)
+		/* HW only takes in a full 64 entry table */
+		return -EINVAL;
+
+	data.size = struct_size(req, groups, count);
+	err = ntmp_alloc_data_mem(&data, (void **)&req);
+	if (err)
+		return err;
+
+	/* Set the request data buffer */
+	ntmp_fill_crd_eid(&req->rbe, user->tbl.rsst_ver, 0,
+			  NTMP_GEN_UA_CFGEU | NTMP_GEN_UA_STSEU, 0);
+	for (i = 0; i < count; i++)
+		req->groups[i] = (u8)(table[i]);
+
+	ntmp_fill_request_hdr(&cbd, data.dma, NTMP_LEN(data.size, 0),
+			      NTMP_RSST_ID, NTMP_CMD_UPDATE, NTMP_AM_ENTRY_ID);
+
+	err = netc_xmit_ntmp_cmd(user, &cbd);
+	if (err)
+		dev_err(user->dev, "Failed to update RSST entry, err: %pe\n",
+			ERR_PTR(err));
+
+	ntmp_free_data_mem(&data);
+
+	return err;
+}
+EXPORT_SYMBOL_GPL(ntmp_rsst_update_entry);
+
+int ntmp_rsst_query_entry(struct ntmp_user *user, u32 *table, int count)
+{
+	struct ntmp_dma_buf data = {.dev = user->dev};
+	struct ntmp_req_by_eid *req;
+	union netc_cbd cbd;
+	int err, i;
+	u8 *group;
+
+	if (count != RSST_ENTRY_NUM)
+		/* HW only takes in a full 64 entry table */
+		return -EINVAL;
+
+	data.size = NTMP_ENTRY_ID_SIZE + RSST_STSE_DATA_SIZE(count) +
+		    RSST_CFGE_DATA_SIZE(count);
+	err = ntmp_alloc_data_mem(&data, (void **)&req);
+	if (err)
+		return err;
+
+	/* Set the request data buffer */
+	ntmp_fill_crd_eid(req, user->tbl.rsst_ver, 0, 0, 0);
+	ntmp_fill_request_hdr(&cbd, data.dma, NTMP_LEN(sizeof(*req), data.size),
+			      NTMP_RSST_ID, NTMP_CMD_QUERY, NTMP_AM_ENTRY_ID);
+	err = netc_xmit_ntmp_cmd(user, &cbd);
+	if (err) {
+		dev_err(user->dev, "Failed to query RSST entry, err: %pe\n",
+			ERR_PTR(err));
+		goto end;
+	}
+
+	group = (u8 *)req;
+	group += NTMP_ENTRY_ID_SIZE + RSST_STSE_DATA_SIZE(count);
+	for (i = 0; i < count; i++)
+		table[i] = group[i];
+
+end:
+	ntmp_free_data_mem(&data);
+
+	return err;
+}
+EXPORT_SYMBOL_GPL(ntmp_rsst_query_entry);
+
+MODULE_DESCRIPTION("NXP NETC Library");
+MODULE_LICENSE("Dual BSD/GPL");
diff --git a/drivers/net/ethernet/freescale/enetc/ntmp_private.h b/drivers/net/ethernet/freescale/enetc/ntmp_private.h
new file mode 100644
index 000000000000..34394e40fddd
--- /dev/null
+++ b/drivers/net/ethernet/freescale/enetc/ntmp_private.h
@@ -0,0 +1,104 @@
+/* SPDX-License-Identifier: (GPL-2.0+ OR BSD-3-Clause) */
+/*
+ * NTMP table request and response data buffer formats
+ * Copyright 2025 NXP
+ */
+
+#ifndef __NTMP_PRIVATE_H
+#define __NTMP_PRIVATE_H
+
+#include <linux/bitfield.h>
+#include <linux/fsl/ntmp.h>
+
+#define NTMP_EID_REQ_LEN	8
+#define NETC_CBDR_BD_NUM	256
+
+union netc_cbd {
+	struct {
+		__le64 addr;
+		__le32 len;
+#define NTMP_RESP_LEN		GENMASK(19, 0)
+#define NTMP_REQ_LEN		GENMASK(31, 20)
+#define NTMP_LEN(req, resp)	(FIELD_PREP(NTMP_REQ_LEN, (req)) | \
+				((resp) & NTMP_RESP_LEN))
+		u8 cmd;
+#define NTMP_CMD_DELETE		BIT(0)
+#define NTMP_CMD_UPDATE		BIT(1)
+#define NTMP_CMD_QUERY		BIT(2)
+#define NTMP_CMD_ADD		BIT(3)
+#define NTMP_CMD_QU		(NTMP_CMD_QUERY | NTMP_CMD_UPDATE)
+		u8 access_method;
+#define NTMP_ACCESS_METHOD	GENMASK(7, 4)
+#define NTMP_AM_ENTRY_ID	0
+#define NTMP_AM_EXACT_KEY	1
+#define NTMP_AM_SEARCH		2
+#define NTMP_AM_TERNARY_KEY	3
+		u8 table_id;
+		u8 ver_cci_rr;
+#define NTMP_HDR_VERSION	GENMASK(5, 0)
+#define NTMP_HDR_VER2		2
+#define NTMP_CCI		BIT(6)
+#define NTMP_RR			BIT(7)
+		__le32 resv[3];
+		__le32 npf;
+#define NTMP_NPF		BIT(15)
+	} req_hdr;	/* NTMP Request Message Header Format */
+
+	struct {
+		__le32 resv0[3];
+		__le16 num_matched;
+		__le16 error_rr;
+#define NTMP_RESP_ERROR		GENMASK(11, 0)
+#define NTMP_RESP_RR		BIT(15)
+		__le32 resv1[4];
+	} resp_hdr; /* NTMP Response Message Header Format */
+};
+
+struct ntmp_dma_buf {
+	struct device *dev;
+	size_t size;
+	void *buf;
+	dma_addr_t dma;
+};
+
+struct ntmp_cmn_req_data {
+	__le16 update_act;
+	u8 dbg_opt;
+	u8 tblv_qact;
+#define NTMP_QUERY_ACT		GENMASK(3, 0)
+#define NTMP_TBL_VER		GENMASK(7, 4)
+#define NTMP_TBLV_QACT(v, a)	(FIELD_PREP(NTMP_TBL_VER, (v)) | \
+				 ((a) & NTMP_QUERY_ACT))
+};
+
+struct ntmp_cmn_resp_query {
+	__le32 entry_id;
+};
+
+/* Generic structure for request data by entry ID  */
+struct ntmp_req_by_eid {
+	struct ntmp_cmn_req_data crd;
+	__le32 entry_id;
+};
+
+/* MAC Address Filter Table Request Data Buffer Format of Add action */
+struct maft_req_add {
+	struct ntmp_req_by_eid rbe;
+	struct maft_keye_data keye;
+	struct maft_cfge_data cfge;
+};
+
+/* MAC Address Filter Table Response Data Buffer Format of Query action */
+struct maft_resp_query {
+	__le32 entry_id;
+	struct maft_keye_data keye;
+	struct maft_cfge_data cfge;
+};
+
+/* RSS Table Request Data Buffer Format of Update action */
+struct rsst_req_update {
+	struct ntmp_req_by_eid rbe;
+	u8 groups[];
+};
+
+#endif
diff --git a/drivers/net/ethernet/freescale/fec.h b/drivers/net/ethernet/freescale/fec.h
index 5ea740b4cf14..41e0d85d15da 100644
--- a/drivers/net/ethernet/freescale/fec.h
+++ b/drivers/net/ethernet/freescale/fec.h
@@ -1,3 +1,4 @@
+/* SPDX-License-Identifier: GPL-2.0 */
 /****************************************************************************/
 
 /*
@@ -13,13 +14,19 @@
 #define	FEC_H
 /****************************************************************************/
 
+#include <dt-bindings/firmware/imx/rsrc.h>
+#include <linux/bpf.h>
 #include <linux/clocksource.h>
+#include <linux/firmware/imx/sci.h>
 #include <linux/net_tstamp.h>
+#include <linux/pm_qos.h>
 #include <linux/ptp_clock_kernel.h>
 #include <linux/timecounter.h>
+#include <net/xdp.h>
 
 #if defined(CONFIG_M523x) || defined(CONFIG_M527x) || defined(CONFIG_M528x) || \
-    defined(CONFIG_M520x) || defined(CONFIG_M532x) || defined(CONFIG_ARM)
+    defined(CONFIG_M520x) || defined(CONFIG_M532x) || defined(CONFIG_ARM) || \
+    defined(CONFIG_ARM64) || defined(CONFIG_COMPILE_TEST)
 /*
  *	Just figures, Motorola would have to change the offsets for
  *	registers in the same peripheral device on different models
@@ -75,6 +82,8 @@
 #define FEC_R_DES_ACTIVE_2	0x1e8 /* Rx descriptor active for ring 2 */
 #define FEC_X_DES_ACTIVE_2	0x1ec /* Tx descriptor active for ring 2 */
 #define FEC_QOS_SCHEME		0x1f0 /* Set multi queues Qos scheme */
+#define FEC_LPI_SLEEP		0x1f4 /* Set IEEE802.3az LPI Sleep Ts time */
+#define FEC_LPI_WAKE		0x1f8 /* Set IEEE802.3az LPI Wake Tw time */
 #define FEC_MIIGSK_CFGR		0x300 /* MIIGSK Configuration reg */
 #define FEC_MIIGSK_ENR		0x308 /* MIIGSK Enable reg */
 
@@ -106,7 +115,7 @@
 #define IEEE_T_MCOL		0x254 /* Frames tx'd with multiple collision */
 #define IEEE_T_DEF		0x258 /* Frames tx'd after deferral delay */
 #define IEEE_T_LCOL		0x25c /* Frames tx'd with late collision */
-#define IEEE_T_EXCOL		0x260 /* Frames tx'd with excesv collisions */
+#define IEEE_T_EXCOL		0x260 /* Frames tx'd with excessive collisions */
 #define IEEE_T_MACERR		0x264 /* Frames tx'd with TX FIFO underrun */
 #define IEEE_T_CSERR		0x268 /* Frames tx'd with carrier sense err */
 #define IEEE_T_SQE		0x26c /* Frames tx'd with SQE err */
@@ -185,6 +194,8 @@
 #define FEC_RXIC0		0xfff
 #define FEC_RXIC1		0xfff
 #define FEC_RXIC2		0xfff
+#define FEC_LPI_SLEEP		0xfff
+#define FEC_LPI_WAKE		0xfff
 #endif /* CONFIG_M5272 */
 
 
@@ -194,7 +205,7 @@
  *	Evidently, ARM SoCs have the FEC block generated in a
  *	little endian mode so adjust endianness accordingly.
  */
-#if defined(CONFIG_ARM)
+#if defined(CONFIG_ARM) || defined(CONFIG_ARM64)
 #define fec32_to_cpu le32_to_cpu
 #define fec16_to_cpu le16_to_cpu
 #define cpu_to_fec32 cpu_to_le32
@@ -331,19 +342,22 @@ struct bufdesc_ex {
 #define FEC_TX_BD_FTYPE(X)	(((X) & 0xf) << 20)
 
 /* The number of Tx and Rx buffers.  These are allocated from the page
- * pool.  The code may assume these are power of two, so it it best
+ * pool.  The code may assume these are power of two, so it is best
  * to keep them that size.
  * We don't need to allocate pages for the transmitter.  We just use
  * the skbuffer directly.
  */
 
+#define FEC_DRV_RESERVE_SPACE (XDP_PACKET_HEADROOM + \
+		SKB_DATA_ALIGN(sizeof(struct skb_shared_info)))
+#define FEC_ENET_XDP_HEADROOM	(XDP_PACKET_HEADROOM)
 #define FEC_ENET_RX_PAGES	256
-#define FEC_ENET_RX_FRSIZE	2048
+#define FEC_ENET_RX_FRSIZE	(PAGE_SIZE - FEC_DRV_RESERVE_SPACE)
 #define FEC_ENET_RX_FRPPG	(PAGE_SIZE / FEC_ENET_RX_FRSIZE)
 #define RX_RING_SIZE		(FEC_ENET_RX_FRPPG * FEC_ENET_RX_PAGES)
 #define FEC_ENET_TX_FRSIZE	2048
 #define FEC_ENET_TX_FRPPG	(PAGE_SIZE / FEC_ENET_TX_FRSIZE)
-#define TX_RING_SIZE		512	/* Must be power of two */
+#define TX_RING_SIZE		1024	/* Must be power of two */
 #define TX_RING_MOD_MASK	511	/*   for this to work */
 
 #define BD_ENET_RX_INT		0x00800000
@@ -371,13 +385,18 @@ struct bufdesc_ex {
 #define FEC_ENET_WAKEUP	((uint)0x00020000)	/* Wakeup request */
 #define FEC_ENET_TXF	(FEC_ENET_TXF_0 | FEC_ENET_TXF_1 | FEC_ENET_TXF_2)
 #define FEC_ENET_RXF	(FEC_ENET_RXF_0 | FEC_ENET_RXF_1 | FEC_ENET_RXF_2)
+#define FEC_ENET_RXF_GET(X)	(((X) == 0) ? FEC_ENET_RXF_0 :	\
+				(((X) == 1) ? FEC_ENET_RXF_1 :	\
+				FEC_ENET_RXF_2))
 #define FEC_ENET_TS_AVAIL       ((uint)0x00010000)
 #define FEC_ENET_TS_TIMER       ((uint)0x00008000)
 
-#define FEC_DEFAULT_IMASK (FEC_ENET_TXF | FEC_ENET_RXF | FEC_ENET_MII | FEC_ENET_TS_TIMER)
-#define FEC_NAPI_IMASK	(FEC_ENET_MII | FEC_ENET_TS_TIMER)
+#define FEC_DEFAULT_IMASK (FEC_ENET_TXF | FEC_ENET_RXF)
 #define FEC_RX_DISABLED_IMASK (FEC_DEFAULT_IMASK & (~FEC_ENET_RXF))
 
+#define FEC_ENET_TXC_DLY	((uint)0x00010000)
+#define FEC_ENET_RXC_DLY	((uint)0x00020000)
+
 /* ENET interrupt coalescing macro define */
 #define FEC_ITR_CLK_SEL		(0x1 << 30)
 #define FEC_ITR_EN		(0x1 << 31)
@@ -442,10 +461,61 @@ struct bufdesc_ex {
 #define FEC_QUIRK_SINGLE_MDIO		(1 << 11)
 /* Controller supports RACC register */
 #define FEC_QUIRK_HAS_RACC		(1 << 12)
-/* Controller supports interrupt coalesc */
+/* Controller supports interrupt coalesce */
 #define FEC_QUIRK_HAS_COALESCE		(1 << 13)
 /* Interrupt doesn't wake CPU from deep idle */
 #define FEC_QUIRK_ERR006687		(1 << 14)
+/* The MIB counters should be cleared and enabled during
+ * initialisation.
+ */
+#define FEC_QUIRK_MIB_CLEAR		(1 << 15)
+/* Only i.MX25/i.MX27/i.MX28 controller supports FRBR,FRSR registers,
+ * those FIFO receive registers are resolved in other platforms.
+ */
+#define FEC_QUIRK_HAS_FRREG		(1 << 16)
+
+/* Some FEC hardware blocks need the MMFR cleared at setup time to avoid
+ * the generation of an MII event. This must be avoided in the older
+ * FEC blocks where it will stop MII events being generated.
+ */
+#define FEC_QUIRK_CLEAR_SETUP_MII	(1 << 17)
+
+/* Some link partners do not tolerate the momentary reset of the REF_CLK
+ * frequency when the RNCTL register is cleared by hardware reset.
+ */
+#define FEC_QUIRK_NO_HARD_RESET		(1 << 18)
+
+/* i.MX6SX ENET IP supports multiple queues (3 queues), use this quirk to
+ * represents this ENET IP.
+ */
+#define FEC_QUIRK_HAS_MULTI_QUEUES	(1 << 19)
+
+/* i.MX8MQ ENET IP version add new feature to support IEEE 802.3az EEE
+ * standard. For the transmission, MAC supply two user registers to set
+ * Sleep (TS) and Wake (TW) time.
+ */
+#define FEC_QUIRK_HAS_EEE		(1 << 20)
+
+/* i.MX8QM ENET IP version add new feature to generate delayed TXC/RXC
+ * as an alternative option to make sure it works well with various PHYs.
+ * For the implementation of delayed clock, ENET takes synchronized 250MHz
+ * clocks to generate 2ns delay.
+ */
+#define FEC_QUIRK_DELAYED_CLKS_SUPPORT	(1 << 21)
+
+/* i.MX8MQ SoC integration mix wakeup interrupt signal into "int2" interrupt line. */
+#define FEC_QUIRK_WAKEUP_FROM_INT2	(1 << 22)
+
+/* i.MX6Q adds pm_qos support */
+#define FEC_QUIRK_HAS_PMQOS			BIT(23)
+
+/* Not all FEC hardware block MDIOs support accesses in C45 mode.
+ * Older blocks in the ColdFire parts do not support it.
+ */
+#define FEC_QUIRK_HAS_MDIO_C45		BIT(24)
+
+/* Jumbo Frame support */
+#define FEC_QUIRK_JUMBO_FRAME		BIT(25)
 
 struct bufdesc_prop {
 	int qid;
@@ -460,10 +530,40 @@ struct bufdesc_prop {
 	unsigned char dsize_log2;
 };
 
+struct fec_enet_priv_txrx_info {
+	int	offset;
+	struct	page *page;
+	struct  sk_buff *skb;
+};
+
+enum {
+	RX_XDP_REDIRECT = 0,
+	RX_XDP_PASS,
+	RX_XDP_DROP,
+	RX_XDP_TX,
+	RX_XDP_TX_ERRORS,
+	TX_XDP_XMIT,
+	TX_XDP_XMIT_ERRORS,
+
+	/* The following must be the last one */
+	XDP_STATS_TOTAL,
+};
+
+enum fec_txbuf_type {
+	FEC_TXBUF_T_SKB,
+	FEC_TXBUF_T_XDP_NDO,
+	FEC_TXBUF_T_XDP_TX,
+};
+
+struct fec_tx_buffer {
+	void *buf_p;
+	enum fec_txbuf_type type;
+};
+
 struct fec_enet_priv_tx_q {
 	struct bufdesc_prop bd;
 	unsigned char *tx_bounce[TX_RING_SIZE];
-	struct  sk_buff *tx_skbuff[TX_RING_SIZE];
+	struct fec_tx_buffer tx_buf[TX_RING_SIZE];
 
 	unsigned short tx_stop_threshold;
 	unsigned short tx_wake_threshold;
@@ -475,7 +575,21 @@ struct fec_enet_priv_tx_q {
 
 struct fec_enet_priv_rx_q {
 	struct bufdesc_prop bd;
-	struct  sk_buff *rx_skbuff[RX_RING_SIZE];
+	struct  fec_enet_priv_txrx_info rx_skb_info[RX_RING_SIZE];
+
+	/* page_pool */
+	struct page_pool *page_pool;
+	struct xdp_rxq_info xdp_rxq;
+	u32 stats[XDP_STATS_TOTAL];
+
+	/* rx queue number, in the range 0-7 */
+	u8 id;
+};
+
+struct fec_stop_mode_gpr {
+	struct regmap *gpr;
+	u8 reg;
+	u8 bit;
 };
 
 /* The FEC buffer descriptors track the ring buffers.  The rx_bd_base and
@@ -497,23 +611,21 @@ struct fec_enet_private {
 	struct clk *clk_ref;
 	struct clk *clk_enet_out;
 	struct clk *clk_ptp;
+	struct clk *clk_2x_txclk;
 
 	bool ptp_clk_on;
 	struct mutex ptp_clk_mutex;
 	unsigned int num_tx_queues;
 	unsigned int num_rx_queues;
 
-	/* The saved address of a sent-in-place packet/buffer, for skfree(). */
 	struct fec_enet_priv_tx_q *tx_queue[FEC_ENET_MAX_TX_QS];
 	struct fec_enet_priv_rx_q *rx_queue[FEC_ENET_MAX_RX_QS];
 
 	unsigned int total_tx_ring_size;
 	unsigned int total_rx_ring_size;
-
-	unsigned long work_tx;
-	unsigned long work_rx;
-	unsigned long work_ts;
-	unsigned long work_mdio;
+	unsigned int max_buf_size;
+	unsigned int pagepool_order;
+	unsigned int rx_frame_size;
 
 	struct	platform_device *pdev;
 
@@ -521,18 +633,20 @@ struct fec_enet_private {
 
 	/* Phylib and MDIO interface */
 	struct	mii_bus *mii_bus;
-	int	mii_timeout;
 	uint	phy_speed;
 	phy_interface_t	phy_interface;
 	struct device_node *phy_node;
+	bool	rgmii_txc_dly;
+	bool	rgmii_rxc_dly;
+	bool	rpm_active;
 	int	link;
 	int	full_duplex;
 	int	speed;
-	struct	completion mdio_done;
 	int	irq[FEC_IRQ_NUM];
 	bool	bufdesc_ex;
 	int	pause_flag;
 	int	wol_flag;
+	int	wake_irq;
 	u32	quirks;
 
 	struct	napi_struct napi;
@@ -542,17 +656,16 @@ struct fec_enet_private {
 
 	struct ptp_clock *ptp_clock;
 	struct ptp_clock_info ptp_caps;
-	unsigned long last_overflow_check;
 	spinlock_t tmreg_lock;
 	struct cyclecounter cc;
 	struct timecounter tc;
-	int rx_hwtstamp_filter;
-	u32 base_incval;
 	u32 cycle_speed;
 	int hwts_rx_en;
 	int hwts_tx_en;
 	struct delayed_work time_keep;
 	struct regulator *reg_phy;
+	struct fec_stop_mode_gpr stop_gpr;
+	struct pm_qos_request pm_qos_req;
 
 	unsigned int tx_align;
 	unsigned int rx_align;
@@ -564,7 +677,7 @@ struct fec_enet_private {
 	unsigned int tx_time_itr;
 	unsigned int itr_clk_rate;
 
-	u32 rx_copybreak;
+	unsigned int clk_ref_rate;
 
 	/* ptp clock period in ns*/
 	unsigned int ptp_inc;
@@ -574,16 +687,32 @@ struct fec_enet_private {
 	unsigned int reload_period;
 	int pps_enable;
 	unsigned int next_counter;
+	struct hrtimer perout_timer;
+	u64 perout_stime;
+
+	struct imx_sc_ipc *ipc_handle;
+
+	/* XDP BPF Program */
+	struct bpf_prog *xdp_prog;
+
+	struct {
+		int pps_enable;
+		u64 ns_sys, ns_phc;
+		u32 at_corr;
+		u8 at_inc_corr;
+	} ptp_saved_state;
 
-	u64 ethtool_stats[0];
+	u64 ethtool_stats[];
 };
 
-void fec_ptp_init(struct platform_device *pdev);
+void fec_ptp_init(struct platform_device *pdev, int irq_idx);
+void fec_ptp_restore_state(struct fec_enet_private *fep);
+void fec_ptp_save_state(struct fec_enet_private *fep);
 void fec_ptp_stop(struct platform_device *pdev);
 void fec_ptp_start_cyclecounter(struct net_device *ndev);
-int fec_ptp_set(struct net_device *ndev, struct ifreq *ifr);
-int fec_ptp_get(struct net_device *ndev, struct ifreq *ifr);
-uint fec_ptp_check_pps_event(struct fec_enet_private *fep);
+int fec_ptp_set(struct net_device *ndev, struct kernel_hwtstamp_config *config,
+		struct netlink_ext_ack *extack);
+void fec_ptp_get(struct net_device *ndev, struct kernel_hwtstamp_config *config);
 
 /****************************************************************************/
 #endif /* FEC_H */
diff --git a/drivers/net/ethernet/freescale/fec_main.c b/drivers/net/ethernet/freescale/fec_main.c
index f7c8649fd28f..1edcfaee6819 100644
--- a/drivers/net/ethernet/freescale/fec_main.c
+++ b/drivers/net/ethernet/freescale/fec_main.c
@@ -1,3 +1,4 @@
+// SPDX-License-Identifier: GPL-2.0+
 /*
  * Fast Ethernet Controller (FEC) driver for Motorola MPC8xx.
  * Copyright (c) 1997 Dan Malek (dmalek@jlc.net)
@@ -21,60 +22,68 @@
  * Copyright (C) 2010-2011 Freescale Semiconductor, Inc.
  */
 
-#include <linux/module.h>
-#include <linux/kernel.h>
-#include <linux/string.h>
-#include <linux/pm_runtime.h>
-#include <linux/ptrace.h>
-#include <linux/errno.h>
-#include <linux/ioport.h>
-#include <linux/slab.h>
-#include <linux/interrupt.h>
+#include <linux/bitops.h>
+#include <linux/bpf.h>
+#include <linux/bpf_trace.h>
+#include <linux/cacheflush.h>
+#include <linux/clk.h>
+#include <linux/crc32.h>
 #include <linux/delay.h>
-#include <linux/netdevice.h>
+#include <linux/errno.h>
 #include <linux/etherdevice.h>
-#include <linux/skbuff.h>
-#include <linux/in.h>
-#include <linux/ip.h>
-#include <net/ip.h>
-#include <net/tso.h>
-#include <linux/tcp.h>
-#include <linux/udp.h>
+#include <linux/fec.h>
+#include <linux/filter.h>
+#include <linux/gpio/consumer.h>
 #include <linux/icmp.h>
-#include <linux/spinlock.h>
-#include <linux/workqueue.h>
-#include <linux/bitops.h>
+#include <linux/if_vlan.h>
+#include <linux/in.h>
+#include <linux/interrupt.h>
 #include <linux/io.h>
+#include <linux/ioport.h>
+#include <linux/ip.h>
 #include <linux/irq.h>
-#include <linux/clk.h>
-#include <linux/platform_device.h>
+#include <linux/kernel.h>
 #include <linux/mdio.h>
-#include <linux/phy.h>
-#include <linux/fec.h>
+#include <linux/mfd/syscon.h>
+#include <linux/module.h>
+#include <linux/netdevice.h>
 #include <linux/of.h>
-#include <linux/of_device.h>
-#include <linux/of_gpio.h>
 #include <linux/of_mdio.h>
 #include <linux/of_net.h>
-#include <linux/regulator/consumer.h>
-#include <linux/if_vlan.h>
+#include <linux/phy.h>
 #include <linux/pinctrl/consumer.h>
+#include <linux/platform_device.h>
+#include <linux/pm_runtime.h>
 #include <linux/prefetch.h>
+#include <linux/property.h>
+#include <linux/ptrace.h>
+#include <linux/regmap.h>
+#include <linux/regulator/consumer.h>
+#include <linux/skbuff.h>
+#include <linux/slab.h>
+#include <linux/spinlock.h>
+#include <linux/string.h>
+#include <linux/tcp.h>
+#include <linux/udp.h>
+#include <linux/workqueue.h>
+#include <net/ip.h>
+#include <net/page_pool/helpers.h>
+#include <net/selftests.h>
+#include <net/tso.h>
 #include <soc/imx/cpuidle.h>
 
-#include <asm/cacheflush.h>
-
 #include "fec.h"
 
 static void set_multicast_list(struct net_device *ndev);
-static void fec_enet_itr_coal_init(struct net_device *ndev);
+static void fec_enet_itr_coal_set(struct net_device *ndev);
+static int fec_enet_xdp_tx_xmit(struct fec_enet_private *fep,
+				int cpu, struct xdp_buff *xdp,
+				u32 dma_sync_len);
 
 #define DRIVER_NAME	"fec"
 
-#define FEC_ENET_GET_QUQUE(_x) ((_x == 0) ? 1 : ((_x == 1) ? 2 : 0))
+static const u16 fec_enet_vlan_pri_to_queue[8] = {0, 0, 1, 1, 1, 2, 2, 2};
 
-/* Pause frame feild and FIFO threshold */
-#define FEC_ENET_FCE	(1 << 5)
 #define FEC_ENET_RSEM_V	0x84
 #define FEC_ENET_RSFL_V	16
 #define FEC_ENET_RAEM_V	0x8
@@ -82,68 +91,116 @@ static void fec_enet_itr_coal_init(struct net_device *ndev);
 #define FEC_ENET_OPD_V	0xFFF0
 #define FEC_MDIO_PM_TIMEOUT  100 /* ms */
 
+#define FEC_ENET_XDP_PASS          0
+#define FEC_ENET_XDP_CONSUMED      BIT(0)
+#define FEC_ENET_XDP_TX            BIT(1)
+#define FEC_ENET_XDP_REDIR         BIT(2)
+
+struct fec_devinfo {
+	u32 quirks;
+};
+
+static const struct fec_devinfo fec_imx25_info = {
+	.quirks = FEC_QUIRK_USE_GASKET | FEC_QUIRK_MIB_CLEAR |
+		  FEC_QUIRK_HAS_FRREG | FEC_QUIRK_HAS_MDIO_C45,
+};
+
+static const struct fec_devinfo fec_imx27_info = {
+	.quirks = FEC_QUIRK_MIB_CLEAR | FEC_QUIRK_HAS_FRREG |
+		  FEC_QUIRK_HAS_MDIO_C45,
+};
+
+static const struct fec_devinfo fec_imx28_info = {
+	.quirks = FEC_QUIRK_ENET_MAC | FEC_QUIRK_SWAP_FRAME |
+		  FEC_QUIRK_SINGLE_MDIO | FEC_QUIRK_HAS_RACC |
+		  FEC_QUIRK_HAS_FRREG | FEC_QUIRK_CLEAR_SETUP_MII |
+		  FEC_QUIRK_NO_HARD_RESET | FEC_QUIRK_HAS_MDIO_C45,
+};
+
+static const struct fec_devinfo fec_imx6q_info = {
+	.quirks = FEC_QUIRK_ENET_MAC | FEC_QUIRK_HAS_GBIT |
+		  FEC_QUIRK_HAS_BUFDESC_EX | FEC_QUIRK_HAS_CSUM |
+		  FEC_QUIRK_HAS_VLAN | FEC_QUIRK_ERR006358 |
+		  FEC_QUIRK_HAS_RACC | FEC_QUIRK_CLEAR_SETUP_MII |
+		  FEC_QUIRK_HAS_PMQOS | FEC_QUIRK_HAS_MDIO_C45,
+};
+
+static const struct fec_devinfo fec_mvf600_info = {
+	.quirks = FEC_QUIRK_ENET_MAC | FEC_QUIRK_HAS_RACC |
+		  FEC_QUIRK_HAS_MDIO_C45,
+};
+
+static const struct fec_devinfo fec_imx6sx_info = {
+	.quirks = FEC_QUIRK_ENET_MAC | FEC_QUIRK_HAS_GBIT |
+		  FEC_QUIRK_HAS_BUFDESC_EX | FEC_QUIRK_HAS_CSUM |
+		  FEC_QUIRK_HAS_VLAN | FEC_QUIRK_HAS_AVB |
+		  FEC_QUIRK_ERR007885 | FEC_QUIRK_BUG_CAPTURE |
+		  FEC_QUIRK_HAS_RACC | FEC_QUIRK_HAS_COALESCE |
+		  FEC_QUIRK_CLEAR_SETUP_MII | FEC_QUIRK_HAS_MULTI_QUEUES |
+		  FEC_QUIRK_HAS_MDIO_C45,
+};
+
+static const struct fec_devinfo fec_imx6ul_info = {
+	.quirks = FEC_QUIRK_ENET_MAC | FEC_QUIRK_HAS_GBIT |
+		  FEC_QUIRK_HAS_BUFDESC_EX | FEC_QUIRK_HAS_CSUM |
+		  FEC_QUIRK_HAS_VLAN | FEC_QUIRK_ERR007885 |
+		  FEC_QUIRK_BUG_CAPTURE | FEC_QUIRK_HAS_RACC |
+		  FEC_QUIRK_HAS_COALESCE | FEC_QUIRK_CLEAR_SETUP_MII |
+		  FEC_QUIRK_HAS_MDIO_C45,
+};
+
+static const struct fec_devinfo fec_imx8mq_info = {
+	.quirks = FEC_QUIRK_ENET_MAC | FEC_QUIRK_HAS_GBIT |
+		  FEC_QUIRK_HAS_BUFDESC_EX | FEC_QUIRK_HAS_CSUM |
+		  FEC_QUIRK_HAS_VLAN | FEC_QUIRK_HAS_AVB |
+		  FEC_QUIRK_ERR007885 | FEC_QUIRK_BUG_CAPTURE |
+		  FEC_QUIRK_HAS_RACC | FEC_QUIRK_HAS_COALESCE |
+		  FEC_QUIRK_CLEAR_SETUP_MII | FEC_QUIRK_HAS_MULTI_QUEUES |
+		  FEC_QUIRK_HAS_EEE | FEC_QUIRK_WAKEUP_FROM_INT2 |
+		  FEC_QUIRK_HAS_MDIO_C45,
+};
+
+static const struct fec_devinfo fec_imx8qm_info = {
+	.quirks = FEC_QUIRK_ENET_MAC | FEC_QUIRK_HAS_GBIT |
+		  FEC_QUIRK_HAS_BUFDESC_EX | FEC_QUIRK_HAS_CSUM |
+		  FEC_QUIRK_HAS_VLAN | FEC_QUIRK_HAS_AVB |
+		  FEC_QUIRK_ERR007885 | FEC_QUIRK_BUG_CAPTURE |
+		  FEC_QUIRK_HAS_RACC | FEC_QUIRK_HAS_COALESCE |
+		  FEC_QUIRK_CLEAR_SETUP_MII | FEC_QUIRK_HAS_MULTI_QUEUES |
+		  FEC_QUIRK_DELAYED_CLKS_SUPPORT | FEC_QUIRK_HAS_MDIO_C45 |
+		  FEC_QUIRK_JUMBO_FRAME,
+};
+
+static const struct fec_devinfo fec_s32v234_info = {
+	.quirks = FEC_QUIRK_ENET_MAC | FEC_QUIRK_HAS_GBIT |
+		  FEC_QUIRK_HAS_BUFDESC_EX | FEC_QUIRK_HAS_CSUM |
+		  FEC_QUIRK_HAS_VLAN | FEC_QUIRK_HAS_AVB |
+		  FEC_QUIRK_ERR007885 | FEC_QUIRK_BUG_CAPTURE |
+		  FEC_QUIRK_HAS_MDIO_C45,
+};
+
 static struct platform_device_id fec_devtype[] = {
 	{
 		/* keep it for coldfire */
 		.name = DRIVER_NAME,
 		.driver_data = 0,
 	}, {
-		.name = "imx25-fec",
-		.driver_data = FEC_QUIRK_USE_GASKET,
-	}, {
-		.name = "imx27-fec",
-		.driver_data = 0,
-	}, {
-		.name = "imx28-fec",
-		.driver_data = FEC_QUIRK_ENET_MAC | FEC_QUIRK_SWAP_FRAME |
-				FEC_QUIRK_SINGLE_MDIO | FEC_QUIRK_HAS_RACC,
-	}, {
-		.name = "imx6q-fec",
-		.driver_data = FEC_QUIRK_ENET_MAC | FEC_QUIRK_HAS_GBIT |
-				FEC_QUIRK_HAS_BUFDESC_EX | FEC_QUIRK_HAS_CSUM |
-				FEC_QUIRK_HAS_VLAN | FEC_QUIRK_ERR006358 |
-				FEC_QUIRK_HAS_RACC,
-	}, {
-		.name = "mvf600-fec",
-		.driver_data = FEC_QUIRK_ENET_MAC | FEC_QUIRK_HAS_RACC,
-	}, {
-		.name = "imx6sx-fec",
-		.driver_data = FEC_QUIRK_ENET_MAC | FEC_QUIRK_HAS_GBIT |
-				FEC_QUIRK_HAS_BUFDESC_EX | FEC_QUIRK_HAS_CSUM |
-				FEC_QUIRK_HAS_VLAN | FEC_QUIRK_HAS_AVB |
-				FEC_QUIRK_ERR007885 | FEC_QUIRK_BUG_CAPTURE |
-				FEC_QUIRK_HAS_RACC | FEC_QUIRK_HAS_COALESCE,
-	}, {
-		.name = "imx6ul-fec",
-		.driver_data = FEC_QUIRK_ENET_MAC | FEC_QUIRK_HAS_GBIT |
-				FEC_QUIRK_HAS_BUFDESC_EX | FEC_QUIRK_HAS_CSUM |
-				FEC_QUIRK_HAS_VLAN | FEC_QUIRK_ERR007885 |
-				FEC_QUIRK_BUG_CAPTURE | FEC_QUIRK_HAS_RACC |
-				FEC_QUIRK_HAS_COALESCE,
-	}, {
 		/* sentinel */
 	}
 };
 MODULE_DEVICE_TABLE(platform, fec_devtype);
 
-enum imx_fec_type {
-	IMX25_FEC = 1,	/* runs on i.mx25/50/53 */
-	IMX27_FEC,	/* runs on i.mx27/35/51 */
-	IMX28_FEC,
-	IMX6Q_FEC,
-	MVF600_FEC,
-	IMX6SX_FEC,
-	IMX6UL_FEC,
-};
-
 static const struct of_device_id fec_dt_ids[] = {
-	{ .compatible = "fsl,imx25-fec", .data = &fec_devtype[IMX25_FEC], },
-	{ .compatible = "fsl,imx27-fec", .data = &fec_devtype[IMX27_FEC], },
-	{ .compatible = "fsl,imx28-fec", .data = &fec_devtype[IMX28_FEC], },
-	{ .compatible = "fsl,imx6q-fec", .data = &fec_devtype[IMX6Q_FEC], },
-	{ .compatible = "fsl,mvf600-fec", .data = &fec_devtype[MVF600_FEC], },
-	{ .compatible = "fsl,imx6sx-fec", .data = &fec_devtype[IMX6SX_FEC], },
-	{ .compatible = "fsl,imx6ul-fec", .data = &fec_devtype[IMX6UL_FEC], },
+	{ .compatible = "fsl,imx25-fec", .data = &fec_imx25_info, },
+	{ .compatible = "fsl,imx27-fec", .data = &fec_imx27_info, },
+	{ .compatible = "fsl,imx28-fec", .data = &fec_imx28_info, },
+	{ .compatible = "fsl,imx6q-fec", .data = &fec_imx6q_info, },
+	{ .compatible = "fsl,mvf600-fec", .data = &fec_mvf600_info, },
+	{ .compatible = "fsl,imx6sx-fec", .data = &fec_imx6sx_info, },
+	{ .compatible = "fsl,imx6ul-fec", .data = &fec_imx6ul_info, },
+	{ .compatible = "fsl,imx8mq-fec", .data = &fec_imx8mq_info, },
+	{ .compatible = "fsl,imx8qm-fec", .data = &fec_imx8qm_info, },
+	{ .compatible = "fsl,s32v234-fec", .data = &fec_s32v234_info, },
 	{ /* sentinel */ }
 };
 MODULE_DEVICE_TABLE(of, fec_dt_ids);
@@ -173,40 +230,68 @@ MODULE_PARM_DESC(macaddr, "FEC Ethernet MAC address");
 #endif /* CONFIG_M5272 */
 
 /* The FEC stores dest/src/type/vlan, data, and checksum for receive packets.
+ *
+ * 2048 byte skbufs are allocated. However, alignment requirements
+ * varies between FEC variants. Worst case is 64, so round down by 64.
  */
-#define PKT_MAXBUF_SIZE		1522
+#define MAX_JUMBO_BUF_SIZE	(round_down(16384 - FEC_DRV_RESERVE_SPACE - 64, 64))
+#define PKT_MAXBUF_SIZE		(round_down(2048 - 64, 64))
 #define PKT_MINBUF_SIZE		64
-#define PKT_MAXBLR_SIZE		1536
 
 /* FEC receive acceleration */
-#define FEC_RACC_IPDIS		(1 << 1)
-#define FEC_RACC_PRODIS		(1 << 2)
+#define FEC_RACC_IPDIS		BIT(1)
+#define FEC_RACC_PRODIS		BIT(2)
 #define FEC_RACC_SHIFT16	BIT(7)
 #define FEC_RACC_OPTIONS	(FEC_RACC_IPDIS | FEC_RACC_PRODIS)
 
+/* MIB Control Register */
+#define FEC_MIB_CTRLSTAT_DISABLE	BIT(31)
+
 /*
  * The 5270/5271/5280/5282/532x RX control register also contains maximum frame
  * size bits. Other FEC hardware does not, so we need to take that into
  * account when setting it.
  */
 #if defined(CONFIG_M523x) || defined(CONFIG_M527x) || defined(CONFIG_M528x) || \
-    defined(CONFIG_M520x) || defined(CONFIG_M532x) || defined(CONFIG_ARM)
-#define	OPT_FRAME_SIZE	(PKT_MAXBUF_SIZE << 16)
+    defined(CONFIG_M520x) || defined(CONFIG_M532x) || defined(CONFIG_ARM) || \
+    defined(CONFIG_ARM64)
+#define	OPT_ARCH_HAS_MAX_FL	1
 #else
-#define	OPT_FRAME_SIZE	0
+#define	OPT_ARCH_HAS_MAX_FL	0
 #endif
 
 /* FEC MII MMFR bits definition */
 #define FEC_MMFR_ST		(1 << 30)
+#define FEC_MMFR_ST_C45		(0)
 #define FEC_MMFR_OP_READ	(2 << 28)
+#define FEC_MMFR_OP_READ_C45	(3 << 28)
 #define FEC_MMFR_OP_WRITE	(1 << 28)
+#define FEC_MMFR_OP_ADDR_WRITE	(0)
 #define FEC_MMFR_PA(v)		((v & 0x1f) << 23)
 #define FEC_MMFR_RA(v)		((v & 0x1f) << 18)
 #define FEC_MMFR_TA		(2 << 16)
 #define FEC_MMFR_DATA(v)	(v & 0xffff)
 /* FEC ECR bits definition */
-#define FEC_ECR_MAGICEN		(1 << 2)
-#define FEC_ECR_SLEEP		(1 << 3)
+#define FEC_ECR_RESET           BIT(0)
+#define FEC_ECR_ETHEREN         BIT(1)
+#define FEC_ECR_MAGICEN         BIT(2)
+#define FEC_ECR_SLEEP           BIT(3)
+#define FEC_ECR_EN1588          BIT(4)
+#define FEC_ECR_SPEED           BIT(5)
+#define FEC_ECR_BYTESWP         BIT(8)
+/* FEC RCR bits definition */
+#define FEC_RCR_LOOP            BIT(0)
+#define FEC_RCR_DRT		BIT(1)
+#define FEC_RCR_MII             BIT(2)
+#define FEC_RCR_PROMISC         BIT(3)
+#define FEC_RCR_BC_REJ          BIT(4)
+#define FEC_RCR_FLOWCTL         BIT(5)
+#define FEC_RCR_RGMII		BIT(6)
+#define FEC_RCR_RMII            BIT(8)
+#define FEC_RCR_10BASET         BIT(9)
+#define FEC_RCR_NLC		BIT(30)
+/* TX WMARK bits */
+#define FEC_TXWMRK_STRFWD       BIT(8)
 
 #define FEC_MII_TIMEOUT		30000 /* us */
 
@@ -219,9 +304,6 @@ MODULE_PARM_DESC(macaddr, "FEC Ethernet MAC address");
 #define FEC_WOL_FLAG_ENABLE		(0x1 << 1)
 #define FEC_WOL_FLAG_SLEEP_ON		(0x1 << 2)
 
-#define COPYBREAK_DEFAULT	256
-
-#define TSO_HEADER_SIZE		128
 /* Max number of allowed TCP segments for software TSO */
 #define FEC_MAX_TSO_SEGS	100
 #define FEC_MAX_SKB_DESCS	(FEC_MAX_TSO_SEGS * 2 + MAX_SKB_FRAGS)
@@ -271,16 +353,6 @@ static void swap_buffer(void *bufaddr, int len)
 		swab32s(buf);
 }
 
-static void swap_buffer2(void *dst_buf, void *src_buf, int len)
-{
-	int i;
-	unsigned int *src = src_buf;
-	unsigned int *dst = dst_buf;
-
-	for (i = 0; i < len; i += 4, src++, dst++)
-		*dst = swab32p(src);
-}
-
 static void fec_dump(struct net_device *ndev)
 {
 	struct fec_enet_private *fep = netdev_priv(ndev);
@@ -302,12 +374,76 @@ static void fec_dump(struct net_device *ndev)
 			fec16_to_cpu(bdp->cbd_sc),
 			fec32_to_cpu(bdp->cbd_bufaddr),
 			fec16_to_cpu(bdp->cbd_datlen),
-			txq->tx_skbuff[index]);
+			txq->tx_buf[index].buf_p);
 		bdp = fec_enet_get_nextdesc(bdp, &txq->bd);
 		index++;
 	} while (bdp != txq->bd.base);
 }
 
+/*
+ * Coldfire does not support DMA coherent allocations, and has historically used
+ * a band-aid with a manual flush in fec_enet_rx_queue.
+ */
+#if defined(CONFIG_COLDFIRE) && !defined(CONFIG_COLDFIRE_COHERENT_DMA)
+static void *fec_dma_alloc(struct device *dev, size_t size, dma_addr_t *handle,
+		gfp_t gfp)
+{
+	return dma_alloc_noncoherent(dev, size, handle, DMA_BIDIRECTIONAL, gfp);
+}
+
+static void fec_dma_free(struct device *dev, size_t size, void *cpu_addr,
+		dma_addr_t handle)
+{
+	dma_free_noncoherent(dev, size, cpu_addr, handle, DMA_BIDIRECTIONAL);
+}
+#else /* !CONFIG_COLDFIRE || CONFIG_COLDFIRE_COHERENT_DMA */
+static void *fec_dma_alloc(struct device *dev, size_t size, dma_addr_t *handle,
+		gfp_t gfp)
+{
+	return dma_alloc_coherent(dev, size, handle, gfp);
+}
+
+static void fec_dma_free(struct device *dev, size_t size, void *cpu_addr,
+		dma_addr_t handle)
+{
+	dma_free_coherent(dev, size, cpu_addr, handle);
+}
+#endif /* !CONFIG_COLDFIRE || CONFIG_COLDFIRE_COHERENT_DMA */
+
+struct fec_dma_devres {
+	size_t		size;
+	void		*vaddr;
+	dma_addr_t	dma_handle;
+};
+
+static void fec_dmam_release(struct device *dev, void *res)
+{
+	struct fec_dma_devres *this = res;
+
+	fec_dma_free(dev, this->size, this->vaddr, this->dma_handle);
+}
+
+static void *fec_dmam_alloc(struct device *dev, size_t size, dma_addr_t *handle,
+		gfp_t gfp)
+{
+	struct fec_dma_devres *dr;
+	void *vaddr;
+
+	dr = devres_alloc(fec_dmam_release, sizeof(*dr), gfp);
+	if (!dr)
+		return NULL;
+	vaddr = fec_dma_alloc(dev, size, handle, gfp);
+	if (!vaddr) {
+		devres_free(dr);
+		return NULL;
+	}
+	dr->vaddr = vaddr;
+	dr->dma_handle = *handle;
+	dr->size = size;
+	devres_add(dev, dr);
+	return vaddr;
+}
+
 static inline bool is_ipv4_pkt(struct sk_buff *skb)
 {
 	return skb->protocol == htons(ETH_P_IP) && ip_hdr(skb)->version == 4;
@@ -330,6 +466,49 @@ fec_enet_clear_csum(struct sk_buff *skb, struct net_device *ndev)
 	return 0;
 }
 
+static int
+fec_enet_create_page_pool(struct fec_enet_private *fep,
+			  struct fec_enet_priv_rx_q *rxq, int size)
+{
+	struct bpf_prog *xdp_prog = READ_ONCE(fep->xdp_prog);
+	struct page_pool_params pp_params = {
+		.order = fep->pagepool_order,
+		.flags = PP_FLAG_DMA_MAP | PP_FLAG_DMA_SYNC_DEV,
+		.pool_size = size,
+		.nid = dev_to_node(&fep->pdev->dev),
+		.dev = &fep->pdev->dev,
+		.dma_dir = xdp_prog ? DMA_BIDIRECTIONAL : DMA_FROM_DEVICE,
+		.offset = FEC_ENET_XDP_HEADROOM,
+		.max_len = fep->rx_frame_size,
+	};
+	int err;
+
+	rxq->page_pool = page_pool_create(&pp_params);
+	if (IS_ERR(rxq->page_pool)) {
+		err = PTR_ERR(rxq->page_pool);
+		rxq->page_pool = NULL;
+		return err;
+	}
+
+	err = xdp_rxq_info_reg(&rxq->xdp_rxq, fep->netdev, rxq->id, 0);
+	if (err < 0)
+		goto err_free_pp;
+
+	err = xdp_rxq_info_reg_mem_model(&rxq->xdp_rxq, MEM_TYPE_PAGE_POOL,
+					 rxq->page_pool);
+	if (err)
+		goto err_unregister_rxq;
+
+	return 0;
+
+err_unregister_rxq:
+	xdp_rxq_info_unreg(&rxq->xdp_rxq);
+err_free_pp:
+	page_pool_destroy(rxq->page_pool);
+	rxq->page_pool = NULL;
+	return err;
+}
+
 static struct bufdesc *
 fec_enet_txq_submit_frag_skb(struct fec_enet_priv_tx_q *txq,
 			     struct sk_buff *skb,
@@ -356,7 +535,7 @@ fec_enet_txq_submit_frag_skb(struct fec_enet_priv_tx_q *txq,
 		status = fec16_to_cpu(bdp->cbd_sc);
 		status &= ~BD_ENET_TX_STATS;
 		status |= (BD_ENET_TX_TC | BD_ENET_TX_READY);
-		frag_len = skb_shinfo(skb)->frags[frag].size;
+		frag_len = skb_frag_size(&skb_shinfo(skb)->frags[frag]);
 
 		/* Handle the last BD specially */
 		if (frag == nr_frags - 1) {
@@ -374,11 +553,12 @@ fec_enet_txq_submit_frag_skb(struct fec_enet_priv_tx_q *txq,
 				estatus |= FEC_TX_BD_FTYPE(txq->bd.qid);
 			if (skb->ip_summed == CHECKSUM_PARTIAL)
 				estatus |= BD_ENET_TX_PINS | BD_ENET_TX_IINS;
+
 			ebdp->cbd_bdu = 0;
 			ebdp->cbd_esc = cpu_to_fec32(estatus);
 		}
 
-		bufaddr = page_address(this_frag->page.p) + this_frag->page_offset;
+		bufaddr = skb_frag_address(this_frag);
 
 		index = fec_enet_get_bd_index(bdp, &txq->bd);
 		if (((unsigned long) bufaddr) & fep->tx_align ||
@@ -515,7 +695,7 @@ static int fec_enet_txq_submit_skb(struct fec_enet_priv_tx_q *txq,
 
 	index = fec_enet_get_bd_index(last_bdp, &txq->bd);
 	/* Save skb pointer */
-	txq->tx_skbuff[index] = skb;
+	txq->tx_buf[index].buf_p = skb;
 
 	/* Make sure the updates to rest of the descriptor are performed before
 	 * transferring ownership.
@@ -533,14 +713,17 @@ static int fec_enet_txq_submit_skb(struct fec_enet_priv_tx_q *txq,
 
 	skb_tx_timestamp(skb);
 
-	/* Make sure the update to bdp and tx_skbuff are performed before
-	 * txq->bd.cur.
-	 */
+	/* Make sure the update to bdp is performed before txq->bd.cur. */
 	wmb();
 	txq->bd.cur = bdp;
 
 	/* Trigger transmission start */
-	writel(0, txq->bd.reg_desc_active);
+	if (!(fep->quirks & FEC_QUIRK_ERR007885) ||
+	    !readl(txq->bd.reg_desc_active) ||
+	    !readl(txq->bd.reg_desc_active) ||
+	    !readl(txq->bd.reg_desc_active) ||
+	    !readl(txq->bd.reg_desc_active))
+		writel(0, txq->bd.reg_desc_active);
 
 	return 0;
 }
@@ -576,7 +759,7 @@ fec_enet_txq_put_data_tso(struct fec_enet_priv_tx_q *txq, struct sk_buff *skb,
 		dev_kfree_skb_any(skb);
 		if (net_ratelimit())
 			netdev_err(ndev, "Tx DMA memory map failed\n");
-		return NETDEV_TX_BUSY;
+		return NETDEV_TX_OK;
 	}
 
 	bdp->cbd_datlen = cpu_to_fec16(size);
@@ -611,7 +794,7 @@ fec_enet_txq_put_hdr_tso(struct fec_enet_priv_tx_q *txq,
 			 struct bufdesc *bdp, int index)
 {
 	struct fec_enet_private *fep = netdev_priv(ndev);
-	int hdr_len = skb_transport_offset(skb) + tcp_hdrlen(skb);
+	int hdr_len = skb_tcp_all_headers(skb);
 	struct bufdesc_ex *ebdp = container_of(bdp, struct bufdesc_ex, desc);
 	void *bufaddr;
 	unsigned long dmabuf;
@@ -638,7 +821,7 @@ fec_enet_txq_put_hdr_tso(struct fec_enet_priv_tx_q *txq,
 			dev_kfree_skb_any(skb);
 			if (net_ratelimit())
 				netdev_err(ndev, "Tx DMA memory map failed\n");
-			return NETDEV_TX_BUSY;
+			return NETDEV_TX_OK;
 		}
 	}
 
@@ -664,9 +847,10 @@ static int fec_enet_txq_submit_tso(struct fec_enet_priv_tx_q *txq,
 				   struct net_device *ndev)
 {
 	struct fec_enet_private *fep = netdev_priv(ndev);
-	int hdr_len = skb_transport_offset(skb) + tcp_hdrlen(skb);
-	int total_len, data_left;
+	int hdr_len, total_len, data_left;
 	struct bufdesc *bdp = txq->bd.cur;
+	struct bufdesc *tmp_bdp;
+	struct bufdesc_ex *ebdp;
 	struct tso_t tso;
 	unsigned int index = 0;
 	int ret;
@@ -685,7 +869,7 @@ static int fec_enet_txq_submit_tso(struct fec_enet_priv_tx_q *txq,
 	}
 
 	/* Initialize the TSO handler, and prepare the first payload */
-	tso_start(skb, &tso);
+	hdr_len = tso_start(skb, &tso);
 
 	total_len = skb->len - hdr_len;
 	while (total_len > 0) {
@@ -724,7 +908,7 @@ static int fec_enet_txq_submit_tso(struct fec_enet_priv_tx_q *txq,
 	}
 
 	/* Save skb pointer */
-	txq->tx_skbuff[index] = skb;
+	txq->tx_buf[index].buf_p = skb;
 
 	skb_tx_timestamp(skb);
 	txq->bd.cur = bdp;
@@ -740,7 +924,34 @@ static int fec_enet_txq_submit_tso(struct fec_enet_priv_tx_q *txq,
 	return 0;
 
 err_release:
-	/* TODO: Release all used data descriptors for TSO */
+	/* Release all used data descriptors for TSO */
+	tmp_bdp = txq->bd.cur;
+
+	while (tmp_bdp != bdp) {
+		/* Unmap data buffers */
+		if (tmp_bdp->cbd_bufaddr &&
+		    !IS_TSO_HEADER(txq, fec32_to_cpu(tmp_bdp->cbd_bufaddr)))
+			dma_unmap_single(&fep->pdev->dev,
+					 fec32_to_cpu(tmp_bdp->cbd_bufaddr),
+					 fec16_to_cpu(tmp_bdp->cbd_datlen),
+					 DMA_TO_DEVICE);
+
+		/* Clear standard buffer descriptor fields */
+		tmp_bdp->cbd_sc = 0;
+		tmp_bdp->cbd_datlen = 0;
+		tmp_bdp->cbd_bufaddr = 0;
+
+		/* Handle extended descriptor if enabled */
+		if (fep->bufdesc_ex) {
+			ebdp = (struct bufdesc_ex *)tmp_bdp;
+			ebdp->cbd_esc = 0;
+		}
+
+		tmp_bdp = fec_enet_get_nextdesc(tmp_bdp, &txq->bd);
+	}
+
+	dev_kfree_skb_any(skb);
+
 	return ret;
 }
 
@@ -814,10 +1025,36 @@ static void fec_enet_bd_init(struct net_device *dev)
 		for (i = 0; i < txq->bd.ring_size; i++) {
 			/* Initialize the BD for every fragment in the page. */
 			bdp->cbd_sc = cpu_to_fec16(0);
-			if (txq->tx_skbuff[i]) {
-				dev_kfree_skb_any(txq->tx_skbuff[i]);
-				txq->tx_skbuff[i] = NULL;
+			if (txq->tx_buf[i].type == FEC_TXBUF_T_SKB) {
+				if (bdp->cbd_bufaddr &&
+				    !IS_TSO_HEADER(txq, fec32_to_cpu(bdp->cbd_bufaddr)))
+					dma_unmap_single(&fep->pdev->dev,
+							 fec32_to_cpu(bdp->cbd_bufaddr),
+							 fec16_to_cpu(bdp->cbd_datlen),
+							 DMA_TO_DEVICE);
+				if (txq->tx_buf[i].buf_p)
+					dev_kfree_skb_any(txq->tx_buf[i].buf_p);
+			} else if (txq->tx_buf[i].type == FEC_TXBUF_T_XDP_NDO) {
+				if (bdp->cbd_bufaddr)
+					dma_unmap_single(&fep->pdev->dev,
+							 fec32_to_cpu(bdp->cbd_bufaddr),
+							 fec16_to_cpu(bdp->cbd_datlen),
+							 DMA_TO_DEVICE);
+
+				if (txq->tx_buf[i].buf_p)
+					xdp_return_frame(txq->tx_buf[i].buf_p);
+			} else {
+				struct page *page = txq->tx_buf[i].buf_p;
+
+				if (page)
+					page_pool_put_page(pp_page_to_nmdesc(page)->pp,
+							   page, 0,
+							   false);
 			}
+
+			txq->tx_buf[i].buf_p = NULL;
+			/* restore default tx buffer type: FEC_TXBUF_T_SKB */
+			txq->tx_buf[i].type = FEC_TXBUF_T_SKB;
 			bdp->cbd_bufaddr = cpu_to_fec32(0);
 			bdp = fec_enet_get_nextdesc(bdp, &txq->bd);
 		}
@@ -848,7 +1085,7 @@ static void fec_enet_enable_ring(struct net_device *ndev)
 	for (i = 0; i < fep->num_rx_queues; i++) {
 		rxq = fep->rx_queue[i];
 		writel(rxq->bd.dma, fep->hwp + FEC_R_DES_START(i));
-		writel(PKT_MAXBLR_SIZE, fep->hwp + FEC_R_BUFF_SIZE(i));
+		writel(fep->max_buf_size, fep->hwp + FEC_R_BUFF_SIZE(i));
 
 		/* enable DMA1/2 */
 		if (i)
@@ -867,24 +1104,40 @@ static void fec_enet_enable_ring(struct net_device *ndev)
 	}
 }
 
-static void fec_enet_reset_skb(struct net_device *ndev)
+/* Whack a reset.  We should wait for this.
+ * For i.MX6SX SOC, enet use AXI bus, we use disable MAC
+ * instead of reset MAC itself.
+ */
+static void fec_ctrl_reset(struct fec_enet_private *fep, bool allow_wol)
 {
-	struct fec_enet_private *fep = netdev_priv(ndev);
-	struct fec_enet_priv_tx_q *txq;
-	int i, j;
-
-	for (i = 0; i < fep->num_tx_queues; i++) {
-		txq = fep->tx_queue[i];
+	u32 val;
 
-		for (j = 0; j < txq->bd.ring_size; j++) {
-			if (txq->tx_skbuff[j]) {
-				dev_kfree_skb_any(txq->tx_skbuff[j]);
-				txq->tx_skbuff[j] = NULL;
-			}
+	if (!allow_wol || !(fep->wol_flag & FEC_WOL_FLAG_SLEEP_ON)) {
+		if (fep->quirks & FEC_QUIRK_HAS_MULTI_QUEUES ||
+		    ((fep->quirks & FEC_QUIRK_NO_HARD_RESET) && fep->link)) {
+			writel(0, fep->hwp + FEC_ECNTRL);
+		} else {
+			writel(FEC_ECR_RESET, fep->hwp + FEC_ECNTRL);
+			udelay(10);
 		}
+	} else {
+		val = readl(fep->hwp + FEC_ECNTRL);
+		val |= (FEC_ECR_MAGICEN | FEC_ECR_SLEEP);
+		writel(val, fep->hwp + FEC_ECNTRL);
 	}
 }
 
+static void fec_set_hw_mac_addr(struct net_device *ndev)
+{
+	struct fec_enet_private *fep = netdev_priv(ndev);
+
+	writel(ndev->dev_addr[3] | (ndev->dev_addr[2] << 8) |
+	       (ndev->dev_addr[1] << 16) | (ndev->dev_addr[0] << 24),
+	       fep->hwp + FEC_ADDR_LOW);
+	writel((ndev->dev_addr[5] << 16) | (ndev->dev_addr[4] << 24),
+	       fep->hwp + FEC_ADDR_HIGH);
+}
+
 /*
  * This function is called to start or restart the FEC during a link
  * change, transmit timeout, or to reconfigure the FEC.  The network
@@ -894,49 +1147,37 @@ static void
 fec_restart(struct net_device *ndev)
 {
 	struct fec_enet_private *fep = netdev_priv(ndev);
-	u32 val;
-	u32 temp_mac[2];
-	u32 rcntl = OPT_FRAME_SIZE | 0x04;
-	u32 ecntl = 0x2; /* ETHEREN */
+	u32 ecntl = FEC_ECR_ETHEREN;
+	u32 rcntl = FEC_RCR_MII;
 
-	/* Whack a reset.  We should wait for this.
-	 * For i.MX6SX SOC, enet use AXI bus, we use disable MAC
-	 * instead of reset MAC itself.
-	 */
-	if (fep->quirks & FEC_QUIRK_HAS_AVB) {
-		writel(0, fep->hwp + FEC_ECNTRL);
-	} else {
-		writel(1, fep->hwp + FEC_ECNTRL);
-		udelay(10);
-	}
+	if (OPT_ARCH_HAS_MAX_FL)
+		rcntl |= (fep->netdev->mtu + ETH_HLEN + ETH_FCS_LEN) << 16;
+
+	if (fep->bufdesc_ex)
+		fec_ptp_save_state(fep);
+
+	fec_ctrl_reset(fep, false);
 
 	/*
 	 * enet-mac reset will reset mac address registers too,
 	 * so need to reconfigure it.
 	 */
-	memcpy(&temp_mac, ndev->dev_addr, ETH_ALEN);
-	writel((__force u32)cpu_to_be32(temp_mac[0]),
-	       fep->hwp + FEC_ADDR_LOW);
-	writel((__force u32)cpu_to_be32(temp_mac[1]),
-	       fep->hwp + FEC_ADDR_HIGH);
+	fec_set_hw_mac_addr(ndev);
 
-	/* Clear any outstanding interrupt. */
-	writel(0xffffffff, fep->hwp + FEC_IEVENT);
+	/* Clear any outstanding interrupt, except MDIO. */
+	writel((0xffffffff & ~FEC_ENET_MII), fep->hwp + FEC_IEVENT);
 
 	fec_enet_bd_init(ndev);
 
 	fec_enet_enable_ring(ndev);
 
-	/* Reset tx SKB buffers. */
-	fec_enet_reset_skb(ndev);
-
 	/* Enable MII mode */
 	if (fep->full_duplex == DUPLEX_FULL) {
 		/* FD enable */
 		writel(0x04, fep->hwp + FEC_X_CNTRL);
 	} else {
 		/* No Rcv on Xmit */
-		rcntl |= 0x02;
+		rcntl |= FEC_RCR_DRT;
 		writel(0x0, fep->hwp + FEC_X_CNTRL);
 	}
 
@@ -945,7 +1186,8 @@ fec_restart(struct net_device *ndev)
 
 #if !defined(CONFIG_M5272)
 	if (fep->quirks & FEC_QUIRK_HAS_RACC) {
-		val = readl(fep->hwp + FEC_RACC);
+		u32 val = readl(fep->hwp + FEC_RACC);
+
 		/* align IP header */
 		val |= FEC_RACC_SHIFT16;
 		if (fep->csum_flags & FLAG_RX_CSUM_ENABLED)
@@ -954,7 +1196,7 @@ fec_restart(struct net_device *ndev)
 		else
 			val &= ~FEC_RACC_OPTIONS;
 		writel(val, fep->hwp + FEC_RACC);
-		writel(PKT_MAXBUF_SIZE, fep->hwp + FEC_FTRL);
+		writel(min(fep->rx_frame_size, fep->max_buf_size), fep->hwp + FEC_FTRL);
 	}
 #endif
 
@@ -964,27 +1206,24 @@ fec_restart(struct net_device *ndev)
 	 */
 	if (fep->quirks & FEC_QUIRK_ENET_MAC) {
 		/* Enable flow control and length check */
-		rcntl |= 0x40000000 | 0x00000020;
+		rcntl |= FEC_RCR_NLC | FEC_RCR_FLOWCTL;
 
 		/* RGMII, RMII or MII */
-		if (fep->phy_interface == PHY_INTERFACE_MODE_RGMII ||
-		    fep->phy_interface == PHY_INTERFACE_MODE_RGMII_ID ||
-		    fep->phy_interface == PHY_INTERFACE_MODE_RGMII_RXID ||
-		    fep->phy_interface == PHY_INTERFACE_MODE_RGMII_TXID)
-			rcntl |= (1 << 6);
+		if (phy_interface_mode_is_rgmii(fep->phy_interface))
+			rcntl |= FEC_RCR_RGMII;
 		else if (fep->phy_interface == PHY_INTERFACE_MODE_RMII)
-			rcntl |= (1 << 8);
+			rcntl |= FEC_RCR_RMII;
 		else
-			rcntl &= ~(1 << 8);
+			rcntl &= ~FEC_RCR_RMII;
 
 		/* 1G, 100M or 10M */
 		if (ndev->phydev) {
 			if (ndev->phydev->speed == SPEED_1000)
-				ecntl |= (1 << 5);
+				ecntl |= FEC_ECR_SPEED;
 			else if (ndev->phydev->speed == SPEED_100)
-				rcntl &= ~(1 << 9);
+				rcntl &= ~FEC_RCR_10BASET;
 			else
-				rcntl |= (1 << 9);
+				rcntl |= FEC_RCR_10BASET;
 		}
 	} else {
 #ifdef FEC_MIIGSK_ENR
@@ -1017,7 +1256,7 @@ fec_restart(struct net_device *ndev)
 	if ((fep->pause_flag & FEC_PAUSE_FLAG_ENABLE) ||
 	    ((fep->pause_flag & FEC_PAUSE_FLAG_AUTONEG) &&
 	     ndev->phydev && ndev->phydev->pause)) {
-		rcntl |= FEC_ENET_FCE;
+		rcntl |= FEC_RCR_FLOWCTL;
 
 		/* set FIFO threshold parameter to reduce overrun */
 		writel(FEC_ENET_RSEM_V, fep->hwp + FEC_R_FIFO_RSEM);
@@ -1028,7 +1267,7 @@ fec_restart(struct net_device *ndev)
 		/* OPD */
 		writel(FEC_ENET_OPD_V, fep->hwp + FEC_OPD);
 	} else {
-		rcntl &= ~FEC_ENET_FCE;
+		rcntl &= ~FEC_RCR_FLOWCTL;
 	}
 #endif /* !defined(CONFIG_M5272) */
 
@@ -1043,13 +1282,30 @@ fec_restart(struct net_device *ndev)
 
 	if (fep->quirks & FEC_QUIRK_ENET_MAC) {
 		/* enable ENET endian swap */
-		ecntl |= (1 << 8);
-		/* enable ENET store and forward mode */
-		writel(1 << 8, fep->hwp + FEC_X_WMRK);
+		ecntl |= FEC_ECR_BYTESWP;
+
+		/* When Jumbo Frame is enabled, the FIFO may not be large enough
+		 * to hold an entire frame. In such cases, if the MTU exceeds
+		 * (PKT_MAXBUF_SIZE - ETH_HLEN - ETH_FCS_LEN), configure the interface
+		 * to operate in cut-through mode, triggered by the FIFO threshold.
+		 * Otherwise, enable the ENET store-and-forward mode.
+		 */
+		if ((fep->quirks & FEC_QUIRK_JUMBO_FRAME) &&
+		    (ndev->mtu > (PKT_MAXBUF_SIZE - ETH_HLEN - ETH_FCS_LEN)))
+			writel(0xF, fep->hwp + FEC_X_WMRK);
+		else
+			writel(FEC_TXWMRK_STRFWD, fep->hwp + FEC_X_WMRK);
 	}
 
 	if (fep->bufdesc_ex)
-		ecntl |= (1 << 4);
+		ecntl |= FEC_ECR_EN1588;
+
+	if (fep->quirks & FEC_QUIRK_DELAYED_CLKS_SUPPORT &&
+	    fep->rgmii_txc_dly)
+		ecntl |= FEC_ENET_TXC_DLY;
+	if (fep->quirks & FEC_QUIRK_DELAYED_CLKS_SUPPORT &&
+	    fep->rgmii_rxc_dly)
+		ecntl |= FEC_ENET_RXC_DLY;
 
 #ifndef CONFIG_M5272
 	/* Enable the MIB statistic event counters */
@@ -1060,26 +1316,90 @@ fec_restart(struct net_device *ndev)
 	writel(ecntl, fep->hwp + FEC_ECNTRL);
 	fec_enet_active_rxring(ndev);
 
-	if (fep->bufdesc_ex)
+	if (fep->bufdesc_ex) {
 		fec_ptp_start_cyclecounter(ndev);
+		fec_ptp_restore_state(fep);
+	}
 
 	/* Enable interrupts we wish to service */
 	if (fep->link)
 		writel(FEC_DEFAULT_IMASK, fep->hwp + FEC_IMASK);
 	else
-		writel(FEC_ENET_MII, fep->hwp + FEC_IMASK);
+		writel(0, fep->hwp + FEC_IMASK);
 
 	/* Init the interrupt coalescing */
-	fec_enet_itr_coal_init(ndev);
+	if (fep->quirks & FEC_QUIRK_HAS_COALESCE)
+		fec_enet_itr_coal_set(ndev);
+}
+
+static int fec_enet_ipc_handle_init(struct fec_enet_private *fep)
+{
+	if (!(of_machine_is_compatible("fsl,imx8qm") ||
+	      of_machine_is_compatible("fsl,imx8qxp") ||
+	      of_machine_is_compatible("fsl,imx8dxl")))
+		return 0;
+
+	return imx_scu_get_handle(&fep->ipc_handle);
+}
+
+static void fec_enet_ipg_stop_set(struct fec_enet_private *fep, bool enabled)
+{
+	struct device_node *np = fep->pdev->dev.of_node;
+	u32 rsrc_id, val;
+	int idx;
+
+	if (!np || !fep->ipc_handle)
+		return;
+
+	idx = of_alias_get_id(np, "ethernet");
+	if (idx < 0)
+		idx = 0;
+	rsrc_id = idx ? IMX_SC_R_ENET_1 : IMX_SC_R_ENET_0;
+
+	val = enabled ? 1 : 0;
+	imx_sc_misc_set_control(fep->ipc_handle, rsrc_id, IMX_SC_C_IPG_STOP, val);
+}
 
+static void fec_enet_stop_mode(struct fec_enet_private *fep, bool enabled)
+{
+	struct fec_platform_data *pdata = fep->pdev->dev.platform_data;
+	struct fec_stop_mode_gpr *stop_gpr = &fep->stop_gpr;
+
+	if (stop_gpr->gpr) {
+		if (enabled)
+			regmap_update_bits(stop_gpr->gpr, stop_gpr->reg,
+					   BIT(stop_gpr->bit),
+					   BIT(stop_gpr->bit));
+		else
+			regmap_update_bits(stop_gpr->gpr, stop_gpr->reg,
+					   BIT(stop_gpr->bit), 0);
+	} else if (pdata && pdata->sleep_mode_enable) {
+		pdata->sleep_mode_enable(enabled);
+	} else {
+		fec_enet_ipg_stop_set(fep, enabled);
+	}
+}
+
+static void fec_irqs_disable(struct net_device *ndev)
+{
+	struct fec_enet_private *fep = netdev_priv(ndev);
+
+	writel(0, fep->hwp + FEC_IMASK);
+}
+
+static void fec_irqs_disable_except_wakeup(struct net_device *ndev)
+{
+	struct fec_enet_private *fep = netdev_priv(ndev);
+
+	writel(0, fep->hwp + FEC_IMASK);
+	writel(FEC_ENET_WAKEUP, fep->hwp + FEC_IMASK);
 }
 
 static void
 fec_stop(struct net_device *ndev)
 {
 	struct fec_enet_private *fep = netdev_priv(ndev);
-	struct fec_platform_data *pdata = fep->pdev->dev.platform_data;
-	u32 rmii_mode = readl(fep->hwp + FEC_R_CNTRL) & (1 << 8);
+	u32 rmii_mode = readl(fep->hwp + FEC_R_CNTRL) & FEC_RCR_RMII;
 	u32 val;
 
 	/* We cannot expect a graceful transmit stop without link !!! */
@@ -1090,40 +1410,32 @@ fec_stop(struct net_device *ndev)
 			netdev_err(ndev, "Graceful transmit stop did not complete!\n");
 	}
 
-	/* Whack a reset.  We should wait for this.
-	 * For i.MX6SX SOC, enet use AXI bus, we use disable MAC
-	 * instead of reset MAC itself.
-	 */
-	if (!(fep->wol_flag & FEC_WOL_FLAG_SLEEP_ON)) {
-		if (fep->quirks & FEC_QUIRK_HAS_AVB) {
-			writel(0, fep->hwp + FEC_ECNTRL);
-		} else {
-			writel(1, fep->hwp + FEC_ECNTRL);
-			udelay(10);
-		}
-		writel(FEC_DEFAULT_IMASK, fep->hwp + FEC_IMASK);
-	} else {
-		writel(FEC_DEFAULT_IMASK | FEC_ENET_WAKEUP, fep->hwp + FEC_IMASK);
-		val = readl(fep->hwp + FEC_ECNTRL);
-		val |= (FEC_ECR_MAGICEN | FEC_ECR_SLEEP);
-		writel(val, fep->hwp + FEC_ECNTRL);
+	if (fep->bufdesc_ex)
+		fec_ptp_save_state(fep);
 
-		if (pdata && pdata->sleep_mode_enable)
-			pdata->sleep_mode_enable(true);
-	}
+	fec_ctrl_reset(fep, true);
 	writel(fep->phy_speed, fep->hwp + FEC_MII_SPEED);
+	writel(FEC_DEFAULT_IMASK, fep->hwp + FEC_IMASK);
 
 	/* We have to keep ENET enabled to have MII interrupt stay working */
 	if (fep->quirks & FEC_QUIRK_ENET_MAC &&
 		!(fep->wol_flag & FEC_WOL_FLAG_SLEEP_ON)) {
-		writel(2, fep->hwp + FEC_ECNTRL);
+		writel(FEC_ECR_ETHEREN, fep->hwp + FEC_ECNTRL);
 		writel(rmii_mode, fep->hwp + FEC_R_CNTRL);
 	}
-}
 
+	if (fep->bufdesc_ex) {
+		val = readl(fep->hwp + FEC_ECNTRL);
+		val |= FEC_ECR_EN1588;
+		writel(val, fep->hwp + FEC_ECNTRL);
+
+		fec_ptp_start_cyclecounter(ndev);
+		fec_ptp_restore_state(fep);
+	}
+}
 
 static void
-fec_timeout(struct net_device *ndev)
+fec_timeout(struct net_device *ndev, unsigned int txqueue)
 {
 	struct fec_enet_private *fep = netdev_priv(ndev);
 
@@ -1145,7 +1457,7 @@ static void fec_enet_timeout_work(struct work_struct *work)
 		napi_disable(&fep->napi);
 		netif_tx_lock_bh(ndev);
 		fec_restart(ndev);
-		netif_wake_queue(ndev);
+		netif_tx_wake_all_queues(ndev);
 		netif_tx_unlock_bh(ndev);
 		napi_enable(&fep->napi);
 	}
@@ -1168,9 +1480,10 @@ fec_enet_hwtstamp(struct fec_enet_private *fep, unsigned ts,
 }
 
 static void
-fec_enet_tx_queue(struct net_device *ndev, u16 queue_id)
+fec_enet_tx_queue(struct net_device *ndev, u16 queue_id, int budget)
 {
 	struct	fec_enet_private *fep;
+	struct xdp_frame *xdpf;
 	struct bufdesc *bdp;
 	unsigned short status;
 	struct	sk_buff	*skb;
@@ -1178,11 +1491,11 @@ fec_enet_tx_queue(struct net_device *ndev, u16 queue_id)
 	struct netdev_queue *nq;
 	int	index = 0;
 	int	entries_free;
+	struct page *page;
+	int frame_len;
 
 	fep = netdev_priv(ndev);
 
-	queue_id = FEC_ENET_GET_QUQUE(queue_id);
-
 	txq = fep->tx_queue[queue_id];
 	/* get next bdp of dirty_tx */
 	nq = netdev_get_tx_queue(ndev, queue_id);
@@ -1200,16 +1513,45 @@ fec_enet_tx_queue(struct net_device *ndev, u16 queue_id)
 
 		index = fec_enet_get_bd_index(bdp, &txq->bd);
 
-		skb = txq->tx_skbuff[index];
-		txq->tx_skbuff[index] = NULL;
-		if (!IS_TSO_HEADER(txq, fec32_to_cpu(bdp->cbd_bufaddr)))
-			dma_unmap_single(&fep->pdev->dev,
-					 fec32_to_cpu(bdp->cbd_bufaddr),
-					 fec16_to_cpu(bdp->cbd_datlen),
-					 DMA_TO_DEVICE);
-		bdp->cbd_bufaddr = cpu_to_fec32(0);
-		if (!skb)
-			goto skb_done;
+		if (txq->tx_buf[index].type == FEC_TXBUF_T_SKB) {
+			skb = txq->tx_buf[index].buf_p;
+			if (bdp->cbd_bufaddr &&
+			    !IS_TSO_HEADER(txq, fec32_to_cpu(bdp->cbd_bufaddr)))
+				dma_unmap_single(&fep->pdev->dev,
+						 fec32_to_cpu(bdp->cbd_bufaddr),
+						 fec16_to_cpu(bdp->cbd_datlen),
+						 DMA_TO_DEVICE);
+			bdp->cbd_bufaddr = cpu_to_fec32(0);
+			if (!skb)
+				goto tx_buf_done;
+		} else {
+			/* Tx processing cannot call any XDP (or page pool) APIs if
+			 * the "budget" is 0. Because NAPI is called with budget of
+			 * 0 (such as netpoll) indicates we may be in an IRQ context,
+			 * however, we can't use the page pool from IRQ context.
+			 */
+			if (unlikely(!budget))
+				break;
+
+			if (txq->tx_buf[index].type == FEC_TXBUF_T_XDP_NDO) {
+				xdpf = txq->tx_buf[index].buf_p;
+				if (bdp->cbd_bufaddr)
+					dma_unmap_single(&fep->pdev->dev,
+							 fec32_to_cpu(bdp->cbd_bufaddr),
+							 fec16_to_cpu(bdp->cbd_datlen),
+							 DMA_TO_DEVICE);
+			} else {
+				page = txq->tx_buf[index].buf_p;
+			}
+
+			bdp->cbd_bufaddr = cpu_to_fec32(0);
+			if (unlikely(!txq->tx_buf[index].buf_p)) {
+				txq->tx_buf[index].type = FEC_TXBUF_T_SKB;
+				goto tx_buf_done;
+			}
+
+			frame_len = fec16_to_cpu(bdp->cbd_datlen);
+		}
 
 		/* Check for errors. */
 		if (status & (BD_ENET_TX_HB | BD_ENET_TX_LC |
@@ -1228,16 +1570,11 @@ fec_enet_tx_queue(struct net_device *ndev, u16 queue_id)
 				ndev->stats.tx_carrier_errors++;
 		} else {
 			ndev->stats.tx_packets++;
-			ndev->stats.tx_bytes += skb->len;
-		}
-
-		if (unlikely(skb_shinfo(skb)->tx_flags & SKBTX_IN_PROGRESS) &&
-			fep->bufdesc_ex) {
-			struct skb_shared_hwtstamps shhwtstamps;
-			struct bufdesc_ex *ebdp = (struct bufdesc_ex *)bdp;
 
-			fec_enet_hwtstamp(fep, fec32_to_cpu(ebdp->ts), &shhwtstamps);
-			skb_tstamp_tx(skb, &shhwtstamps);
+			if (txq->tx_buf[index].type == FEC_TXBUF_T_SKB)
+				ndev->stats.tx_bytes += skb->len;
+			else
+				ndev->stats.tx_bytes += frame_len;
 		}
 
 		/* Deferred means some collisions occurred during transmit,
@@ -1246,10 +1583,36 @@ fec_enet_tx_queue(struct net_device *ndev, u16 queue_id)
 		if (status & BD_ENET_TX_DEF)
 			ndev->stats.collisions++;
 
-		/* Free the sk buffer associated with this last transmit */
-		dev_kfree_skb_any(skb);
-skb_done:
-		/* Make sure the update to bdp and tx_skbuff are performed
+		if (txq->tx_buf[index].type == FEC_TXBUF_T_SKB) {
+			/* NOTE: SKBTX_IN_PROGRESS being set does not imply it's we who
+			 * are to time stamp the packet, so we still need to check time
+			 * stamping enabled flag.
+			 */
+			if (unlikely(skb_shinfo(skb)->tx_flags & SKBTX_IN_PROGRESS &&
+				     fep->hwts_tx_en) && fep->bufdesc_ex) {
+				struct skb_shared_hwtstamps shhwtstamps;
+				struct bufdesc_ex *ebdp = (struct bufdesc_ex *)bdp;
+
+				fec_enet_hwtstamp(fep, fec32_to_cpu(ebdp->ts), &shhwtstamps);
+				skb_tstamp_tx(skb, &shhwtstamps);
+			}
+
+			/* Free the sk buffer associated with this last transmit */
+			napi_consume_skb(skb, budget);
+		} else if (txq->tx_buf[index].type == FEC_TXBUF_T_XDP_NDO) {
+			xdp_return_frame_rx_napi(xdpf);
+		} else { /* recycle pages of XDP_TX frames */
+			/* The dma_sync_size = 0 as XDP_TX has already synced DMA for_device */
+			page_pool_put_page(pp_page_to_nmdesc(page)->pp, page,
+					   0, true);
+		}
+
+		txq->tx_buf[index].buf_p = NULL;
+		/* restore default tx buffer type: FEC_TXBUF_T_SKB */
+		txq->tx_buf[index].type = FEC_TXBUF_T_SKB;
+
+tx_buf_done:
+		/* Make sure the update to bdp and tx_buf are performed
 		 * before dirty_tx
 		 */
 		wmb();
@@ -1260,7 +1623,7 @@ skb_done:
 
 		/* Since we have freed up a buffer, the ring is no longer full
 		 */
-		if (netif_queue_stopped(ndev)) {
+		if (netif_tx_queue_stopped(nq)) {
 			entries_free = fec_enet_get_free_txdesc_num(txq);
 			if (entries_free >= txq->tx_wake_threshold)
 				netif_tx_wake_queue(nq);
@@ -1273,63 +1636,113 @@ skb_done:
 		writel(0, txq->bd.reg_desc_active);
 }
 
-static void
-fec_enet_tx(struct net_device *ndev)
+static void fec_enet_tx(struct net_device *ndev, int budget)
 {
 	struct fec_enet_private *fep = netdev_priv(ndev);
-	u16 queue_id;
-	/* First process class A queue, then Class B and Best Effort queue */
-	for_each_set_bit(queue_id, &fep->work_tx, FEC_ENET_MAX_TX_QS) {
-		clear_bit(queue_id, &fep->work_tx);
-		fec_enet_tx_queue(ndev, queue_id);
-	}
-	return;
+	int i;
+
+	/* Make sure that AVB queues are processed first. */
+	for (i = fep->num_tx_queues - 1; i >= 0; i--)
+		fec_enet_tx_queue(ndev, i, budget);
 }
 
-static int
-fec_enet_new_rxbdp(struct net_device *ndev, struct bufdesc *bdp, struct sk_buff *skb)
+static int fec_enet_update_cbd(struct fec_enet_priv_rx_q *rxq,
+				struct bufdesc *bdp, int index)
 {
-	struct  fec_enet_private *fep = netdev_priv(ndev);
-	int off;
-
-	off = ((unsigned long)skb->data) & fep->rx_align;
-	if (off)
-		skb_reserve(skb, fep->rx_align + 1 - off);
+	struct page *new_page;
+	dma_addr_t phys_addr;
 
-	bdp->cbd_bufaddr = cpu_to_fec32(dma_map_single(&fep->pdev->dev, skb->data, FEC_ENET_RX_FRSIZE - fep->rx_align, DMA_FROM_DEVICE));
-	if (dma_mapping_error(&fep->pdev->dev, fec32_to_cpu(bdp->cbd_bufaddr))) {
-		if (net_ratelimit())
-			netdev_err(ndev, "Rx DMA memory map failed\n");
+	new_page = page_pool_dev_alloc_pages(rxq->page_pool);
+	if (unlikely(!new_page))
 		return -ENOMEM;
-	}
+
+	rxq->rx_skb_info[index].page = new_page;
+	rxq->rx_skb_info[index].offset = FEC_ENET_XDP_HEADROOM;
+	phys_addr = page_pool_get_dma_addr(new_page) + FEC_ENET_XDP_HEADROOM;
+	bdp->cbd_bufaddr = cpu_to_fec32(phys_addr);
 
 	return 0;
 }
 
-static bool fec_enet_copybreak(struct net_device *ndev, struct sk_buff **skb,
-			       struct bufdesc *bdp, u32 length, bool swap)
+static u32
+fec_enet_run_xdp(struct fec_enet_private *fep, struct bpf_prog *prog,
+		 struct xdp_buff *xdp, struct fec_enet_priv_rx_q *rxq, int cpu)
 {
-	struct  fec_enet_private *fep = netdev_priv(ndev);
-	struct sk_buff *new_skb;
+	unsigned int sync, len = xdp->data_end - xdp->data;
+	u32 ret = FEC_ENET_XDP_PASS;
+	struct page *page;
+	int err;
+	u32 act;
 
-	if (length > fep->rx_copybreak)
-		return false;
+	act = bpf_prog_run_xdp(prog, xdp);
 
-	new_skb = netdev_alloc_skb(ndev, length);
-	if (!new_skb)
-		return false;
+	/* Due xdp_adjust_tail and xdp_adjust_head: DMA sync for_device cover
+	 * max len CPU touch
+	 */
+	sync = xdp->data_end - xdp->data;
+	sync = max(sync, len);
 
-	dma_sync_single_for_cpu(&fep->pdev->dev,
-				fec32_to_cpu(bdp->cbd_bufaddr),
-				FEC_ENET_RX_FRSIZE - fep->rx_align,
-				DMA_FROM_DEVICE);
-	if (!swap)
-		memcpy(new_skb->data, (*skb)->data, length);
-	else
-		swap_buffer2(new_skb->data, (*skb)->data, length);
-	*skb = new_skb;
+	switch (act) {
+	case XDP_PASS:
+		rxq->stats[RX_XDP_PASS]++;
+		ret = FEC_ENET_XDP_PASS;
+		break;
+
+	case XDP_REDIRECT:
+		rxq->stats[RX_XDP_REDIRECT]++;
+		err = xdp_do_redirect(fep->netdev, xdp, prog);
+		if (unlikely(err))
+			goto xdp_err;
+
+		ret = FEC_ENET_XDP_REDIR;
+		break;
+
+	case XDP_TX:
+		rxq->stats[RX_XDP_TX]++;
+		err = fec_enet_xdp_tx_xmit(fep, cpu, xdp, sync);
+		if (unlikely(err)) {
+			rxq->stats[RX_XDP_TX_ERRORS]++;
+			goto xdp_err;
+		}
 
-	return true;
+		ret = FEC_ENET_XDP_TX;
+		break;
+
+	default:
+		bpf_warn_invalid_xdp_action(fep->netdev, prog, act);
+		fallthrough;
+
+	case XDP_ABORTED:
+		fallthrough;    /* handle aborts by dropping packet */
+
+	case XDP_DROP:
+		rxq->stats[RX_XDP_DROP]++;
+xdp_err:
+		ret = FEC_ENET_XDP_CONSUMED;
+		page = virt_to_head_page(xdp->data);
+		page_pool_put_page(rxq->page_pool, page, sync, true);
+		if (act != XDP_DROP)
+			trace_xdp_exception(fep->netdev, prog, act);
+		break;
+	}
+
+	return ret;
+}
+
+static void fec_enet_rx_vlan(const struct net_device *ndev, struct sk_buff *skb)
+{
+	if (ndev->features & NETIF_F_HW_VLAN_CTAG_RX) {
+		const struct vlan_ethhdr *vlan_header = skb_vlan_eth_hdr(skb);
+		const u16 vlan_tag = ntohs(vlan_header->h_vlan_TCI);
+
+		/* Push and remove the vlan tag */
+
+		memmove(skb->data + VLAN_HLEN, skb->data, ETH_ALEN * 2);
+		skb_pull(skb, VLAN_HLEN);
+		__vlan_hwaccel_put_tag(skb,
+				       htons(ETH_P_8021Q),
+				       vlan_tag);
+	}
 }
 
 /* During a receive, the bd_rx.cur points to the current incoming buffer.
@@ -1338,34 +1751,51 @@ static bool fec_enet_copybreak(struct net_device *ndev, struct sk_buff **skb,
  * effectively tossing the packet.
  */
 static int
-fec_enet_rx_queue(struct net_device *ndev, int budget, u16 queue_id)
+fec_enet_rx_queue(struct net_device *ndev, u16 queue_id, int budget)
 {
 	struct fec_enet_private *fep = netdev_priv(ndev);
 	struct fec_enet_priv_rx_q *rxq;
 	struct bufdesc *bdp;
 	unsigned short status;
-	struct  sk_buff *skb_new = NULL;
 	struct  sk_buff *skb;
 	ushort	pkt_len;
-	__u8 *data;
 	int	pkt_received = 0;
 	struct	bufdesc_ex *ebdp = NULL;
-	bool	vlan_packet_rcvd = false;
-	u16	vlan_tag;
 	int	index = 0;
-	bool	is_copybreak;
 	bool	need_swap = fep->quirks & FEC_QUIRK_SWAP_FRAME;
+	struct bpf_prog *xdp_prog = READ_ONCE(fep->xdp_prog);
+	u32 ret, xdp_result = FEC_ENET_XDP_PASS;
+	u32 data_start = FEC_ENET_XDP_HEADROOM;
+	int cpu = smp_processor_id();
+	struct xdp_buff xdp;
+	struct page *page;
+	__fec32 cbd_bufaddr;
+	u32 sub_len = 4;
 
-#ifdef CONFIG_M532x
+#if !defined(CONFIG_M5272)
+	/*If it has the FEC_QUIRK_HAS_RACC quirk property, the bit of
+	 * FEC_RACC_SHIFT16 is set by default in the probe function.
+	 */
+	if (fep->quirks & FEC_QUIRK_HAS_RACC) {
+		data_start += 2;
+		sub_len += 2;
+	}
+#endif
+
+#if defined(CONFIG_COLDFIRE) && !defined(CONFIG_COLDFIRE_COHERENT_DMA)
+	/*
+	 * Hacky flush of all caches instead of using the DMA API for the TSO
+	 * headers.
+	 */
 	flush_cache_all();
 #endif
-	queue_id = FEC_ENET_GET_QUQUE(queue_id);
 	rxq = fep->rx_queue[queue_id];
 
 	/* First, grab all of the stats for the incoming packet.
 	 * These get messed up if we get called due to a busy condition.
 	 */
 	bdp = rxq->bd.cur;
+	xdp_init_buff(&xdp, PAGE_SIZE << fep->pagepool_order, &rxq->xdp_rxq);
 
 	while (!((status = fec16_to_cpu(bdp->cbd_sc)) & BD_ENET_RX_EMPTY)) {
 
@@ -1373,7 +1803,7 @@ fec_enet_rx_queue(struct net_device *ndev, int budget, u16 queue_id)
 			break;
 		pkt_received++;
 
-		writel(FEC_ENET_RXF, fep->hwp + FEC_IEVENT);
+		writel(FEC_ENET_RXF_GET(queue_id), fep->hwp + FEC_IEVENT);
 
 		/* Check for errors. */
 		status ^= BD_ENET_RX_LAST;
@@ -1407,37 +1837,54 @@ fec_enet_rx_queue(struct net_device *ndev, int budget, u16 queue_id)
 		ndev->stats.rx_bytes += pkt_len;
 
 		index = fec_enet_get_bd_index(bdp, &rxq->bd);
-		skb = rxq->rx_skbuff[index];
+		page = rxq->rx_skb_info[index].page;
+		cbd_bufaddr = bdp->cbd_bufaddr;
+		if (fec_enet_update_cbd(rxq, bdp, index)) {
+			ndev->stats.rx_dropped++;
+			goto rx_processing_done;
+		}
+
+		dma_sync_single_for_cpu(&fep->pdev->dev,
+					fec32_to_cpu(cbd_bufaddr),
+					pkt_len,
+					DMA_FROM_DEVICE);
+		prefetch(page_address(page));
+
+		if (xdp_prog) {
+			xdp_buff_clear_frags_flag(&xdp);
+			/* subtract 16bit shift and FCS */
+			xdp_prepare_buff(&xdp, page_address(page),
+					 data_start, pkt_len - sub_len, false);
+			ret = fec_enet_run_xdp(fep, xdp_prog, &xdp, rxq, cpu);
+			xdp_result |= ret;
+			if (ret != FEC_ENET_XDP_PASS)
+				goto rx_processing_done;
+		}
 
 		/* The packet length includes FCS, but we don't want to
 		 * include that when passing upstream as it messes up
 		 * bridging applications.
 		 */
-		is_copybreak = fec_enet_copybreak(ndev, &skb, bdp, pkt_len - 4,
-						  need_swap);
-		if (!is_copybreak) {
-			skb_new = netdev_alloc_skb(ndev, FEC_ENET_RX_FRSIZE);
-			if (unlikely(!skb_new)) {
-				ndev->stats.rx_dropped++;
-				goto rx_processing_done;
-			}
-			dma_unmap_single(&fep->pdev->dev,
-					 fec32_to_cpu(bdp->cbd_bufaddr),
-					 FEC_ENET_RX_FRSIZE - fep->rx_align,
-					 DMA_FROM_DEVICE);
+		skb = build_skb(page_address(page),
+				PAGE_SIZE << fep->pagepool_order);
+		if (unlikely(!skb)) {
+			page_pool_recycle_direct(rxq->page_pool, page);
+			ndev->stats.rx_dropped++;
+
+			netdev_err_once(ndev, "build_skb failed!\n");
+			goto rx_processing_done;
 		}
 
-		prefetch(skb->data - NET_IP_ALIGN);
-		skb_put(skb, pkt_len - 4);
-		data = skb->data;
+		skb_reserve(skb, data_start);
+		skb_put(skb, pkt_len - sub_len);
+		skb_mark_for_recycle(skb);
 
-		if (!is_copybreak && need_swap)
-			swap_buffer(data, pkt_len);
+		if (unlikely(need_swap)) {
+			u8 *data;
 
-#if !defined(CONFIG_M5272)
-		if (fep->quirks & FEC_QUIRK_HAS_RACC)
-			data = skb_pull_inline(skb, 2);
-#endif
+			data = page_address(page) + FEC_ENET_XDP_HEADROOM;
+			swap_buffer(data, pkt_len);
+		}
 
 		/* Extract the enhanced buffer descriptor */
 		ebdp = NULL;
@@ -1445,20 +1892,9 @@ fec_enet_rx_queue(struct net_device *ndev, int budget, u16 queue_id)
 			ebdp = (struct bufdesc_ex *)bdp;
 
 		/* If this is a VLAN packet remove the VLAN Tag */
-		vlan_packet_rcvd = false;
-		if ((ndev->features & NETIF_F_HW_VLAN_CTAG_RX) &&
-		    fep->bufdesc_ex &&
-		    (ebdp->cbd_esc & cpu_to_fec32(BD_ENET_RX_VLAN))) {
-			/* Push and remove the vlan tag */
-			struct vlan_hdr *vlan_header =
-					(struct vlan_hdr *) (data + ETH_HLEN);
-			vlan_tag = ntohs(vlan_header->h_vlan_TCI);
-
-			vlan_packet_rcvd = true;
-
-			memmove(skb->data + VLAN_HLEN, data, ETH_ALEN * 2);
-			skb_pull(skb, VLAN_HLEN);
-		}
+		if (fep->bufdesc_ex &&
+		    (ebdp->cbd_esc & cpu_to_fec32(BD_ENET_RX_VLAN)))
+			fec_enet_rx_vlan(ndev, skb);
 
 		skb->protocol = eth_type_trans(skb, ndev);
 
@@ -1477,24 +1913,9 @@ fec_enet_rx_queue(struct net_device *ndev, int budget, u16 queue_id)
 			}
 		}
 
-		/* Handle received VLAN packets */
-		if (vlan_packet_rcvd)
-			__vlan_hwaccel_put_tag(skb,
-					       htons(ETH_P_8021Q),
-					       vlan_tag);
-
+		skb_record_rx_queue(skb, queue_id);
 		napi_gro_receive(&fep->napi, skb);
 
-		if (is_copybreak) {
-			dma_sync_single_for_device(&fep->pdev->dev,
-						   fec32_to_cpu(bdp->cbd_bufaddr),
-						   FEC_ENET_RX_FRSIZE - fep->rx_align,
-						   DMA_FROM_DEVICE);
-		} else {
-			rxq->rx_skbuff[index] = skb_new;
-			fec_enet_new_rxbdp(ndev, bdp, skb_new);
-		}
-
 rx_processing_done:
 		/* Clear the status flags for this buffer */
 		status &= ~BD_ENET_RX_STATS;
@@ -1525,51 +1946,37 @@ rx_processing_done:
 		writel(0, rxq->bd.reg_desc_active);
 	}
 	rxq->bd.cur = bdp;
+
+	if (xdp_result & FEC_ENET_XDP_REDIR)
+		xdp_do_flush();
+
 	return pkt_received;
 }
 
-static int
-fec_enet_rx(struct net_device *ndev, int budget)
+static int fec_enet_rx(struct net_device *ndev, int budget)
 {
-	int     pkt_received = 0;
-	u16	queue_id;
 	struct fec_enet_private *fep = netdev_priv(ndev);
+	int i, done = 0;
 
-	for_each_set_bit(queue_id, &fep->work_rx, FEC_ENET_MAX_RX_QS) {
-		int ret;
+	/* Make sure that AVB queues are processed first. */
+	for (i = fep->num_rx_queues - 1; i >= 0; i--)
+		done += fec_enet_rx_queue(ndev, i, budget - done);
 
-		ret = fec_enet_rx_queue(ndev,
-					budget - pkt_received, queue_id);
-
-		if (ret < budget - pkt_received)
-			clear_bit(queue_id, &fep->work_rx);
-
-		pkt_received += ret;
-	}
-	return pkt_received;
+	return done;
 }
 
-static bool
-fec_enet_collect_events(struct fec_enet_private *fep, uint int_events)
+static bool fec_enet_collect_events(struct fec_enet_private *fep)
 {
-	if (int_events == 0)
-		return false;
+	uint int_events;
+
+	int_events = readl(fep->hwp + FEC_IEVENT);
 
-	if (int_events & FEC_ENET_RXF)
-		fep->work_rx |= (1 << 2);
-	if (int_events & FEC_ENET_RXF_1)
-		fep->work_rx |= (1 << 0);
-	if (int_events & FEC_ENET_RXF_2)
-		fep->work_rx |= (1 << 1);
+	/* Don't clear MDIO events, we poll for those */
+	int_events &= ~FEC_ENET_MII;
 
-	if (int_events & FEC_ENET_TXF)
-		fep->work_tx |= (1 << 2);
-	if (int_events & FEC_ENET_TXF_1)
-		fep->work_tx |= (1 << 0);
-	if (int_events & FEC_ENET_TXF_2)
-		fep->work_tx |= (1 << 1);
+	writel(int_events, fep->hwp + FEC_IEVENT);
 
-	return true;
+	return int_events != 0;
 }
 
 static irqreturn_t
@@ -1577,31 +1984,18 @@ fec_enet_interrupt(int irq, void *dev_id)
 {
 	struct net_device *ndev = dev_id;
 	struct fec_enet_private *fep = netdev_priv(ndev);
-	uint int_events;
 	irqreturn_t ret = IRQ_NONE;
 
-	int_events = readl(fep->hwp + FEC_IEVENT);
-	writel(int_events, fep->hwp + FEC_IEVENT);
-	fec_enet_collect_events(fep, int_events);
-
-	if ((fep->work_tx || fep->work_rx) && fep->link) {
+	if (fec_enet_collect_events(fep) && fep->link) {
 		ret = IRQ_HANDLED;
 
 		if (napi_schedule_prep(&fep->napi)) {
-			/* Disable the NAPI interrupts */
-			writel(FEC_NAPI_IMASK, fep->hwp + FEC_IMASK);
+			/* Disable interrupts */
+			writel(0, fep->hwp + FEC_IMASK);
 			__napi_schedule(&fep->napi);
 		}
 	}
 
-	if (int_events & FEC_ENET_MII) {
-		ret = IRQ_HANDLED;
-		complete(&fep->mdio_done);
-	}
-
-	if (fep->ptp_clock)
-		fec_ptp_check_pps_event(fep);
-
 	return ret;
 }
 
@@ -1609,25 +2003,27 @@ static int fec_enet_rx_napi(struct napi_struct *napi, int budget)
 {
 	struct net_device *ndev = napi->dev;
 	struct fec_enet_private *fep = netdev_priv(ndev);
-	int pkts;
-
-	pkts = fec_enet_rx(ndev, budget);
+	int done = 0;
 
-	fec_enet_tx(ndev);
+	do {
+		done += fec_enet_rx(ndev, budget - done);
+		fec_enet_tx(ndev, budget);
+	} while ((done < budget) && fec_enet_collect_events(fep));
 
-	if (pkts < budget) {
-		napi_complete_done(napi, pkts);
+	if (done < budget) {
+		napi_complete_done(napi, done);
 		writel(FEC_DEFAULT_IMASK, fep->hwp + FEC_IMASK);
 	}
-	return pkts;
+
+	return done;
 }
 
 /* ------------------------------------------------------------------------- */
-static void fec_get_mac(struct net_device *ndev)
+static int fec_get_mac(struct net_device *ndev)
 {
 	struct fec_enet_private *fep = netdev_priv(ndev);
-	struct fec_platform_data *pdata = dev_get_platdata(&fep->pdev->dev);
 	unsigned char *iap, tmpaddr[ETH_ALEN];
+	int ret;
 
 	/*
 	 * try to get mac address in following order:
@@ -1643,9 +2039,11 @@ static void fec_get_mac(struct net_device *ndev)
 	if (!is_valid_ether_addr(iap)) {
 		struct device_node *np = fep->pdev->dev.of_node;
 		if (np) {
-			const char *mac = of_get_mac_address(np);
-			if (mac)
-				iap = (unsigned char *) mac;
+			ret = of_get_mac_address(np, tmpaddr);
+			if (!ret)
+				iap = tmpaddr;
+			else if (ret == -EPROBE_DEFER)
+				return ret;
 		}
 	}
 
@@ -1657,6 +2055,8 @@ static void fec_get_mac(struct net_device *ndev)
 		if (FEC_FLASHMAC)
 			iap = (unsigned char *)FEC_FLASHMAC;
 #else
+		struct fec_platform_data *pdata = dev_get_platdata(&fep->pdev->dev);
+
 		if (pdata)
 			iap = (unsigned char *)&pdata->mac;
 #endif
@@ -1678,18 +2078,17 @@ static void fec_get_mac(struct net_device *ndev)
 	 */
 	if (!is_valid_ether_addr(iap)) {
 		/* Report it and use a random ethernet address instead */
-		netdev_err(ndev, "Invalid MAC address: %pM\n", iap);
+		dev_err(&fep->pdev->dev, "Invalid MAC address: %pM\n", iap);
 		eth_hw_addr_random(ndev);
-		netdev_info(ndev, "Using random MAC address: %pM\n",
-			    ndev->dev_addr);
-		return;
+		dev_info(&fep->pdev->dev, "Using random MAC address: %pM\n",
+			 ndev->dev_addr);
+		return 0;
 	}
 
-	memcpy(ndev->dev_addr, iap, ETH_ALEN);
-
 	/* Adjust MAC if using macaddr */
-	if (iap == macaddr)
-		 ndev->dev_addr[ETH_ALEN-1] = macaddr[ETH_ALEN-1] + fep->dev_id;
+	eth_hw_addr_gen(ndev, iap, iap == macaddr ? fep->dev_id : 0);
+
+	return 0;
 }
 
 /* ------------------------------------------------------------------------- */
@@ -1697,18 +2096,43 @@ static void fec_get_mac(struct net_device *ndev)
 /*
  * Phy section
  */
+
+/* LPI Sleep Ts count base on tx clk (clk_ref).
+ * The lpi sleep cnt value = X us / (cycle_ns).
+ */
+static int fec_enet_us_to_tx_cycle(struct net_device *ndev, int us)
+{
+	struct fec_enet_private *fep = netdev_priv(ndev);
+
+	return us * (fep->clk_ref_rate / 1000) / 1000;
+}
+
+static int fec_enet_eee_mode_set(struct net_device *ndev, u32 lpi_timer,
+				 bool enable)
+{
+	struct fec_enet_private *fep = netdev_priv(ndev);
+	unsigned int sleep_cycle, wake_cycle;
+
+	if (enable) {
+		sleep_cycle = fec_enet_us_to_tx_cycle(ndev, lpi_timer);
+		wake_cycle = sleep_cycle;
+	} else {
+		sleep_cycle = 0;
+		wake_cycle = 0;
+	}
+
+	writel(sleep_cycle, fep->hwp + FEC_LPI_SLEEP);
+	writel(wake_cycle, fep->hwp + FEC_LPI_WAKE);
+
+	return 0;
+}
+
 static void fec_enet_adjust_link(struct net_device *ndev)
 {
 	struct fec_enet_private *fep = netdev_priv(ndev);
 	struct phy_device *phy_dev = ndev->phydev;
 	int status_change = 0;
 
-	/* Prevent a state halted on mii error */
-	if (fep->mii_timeout && phy_dev->state == PHY_HALTED) {
-		phy_dev->state = PHY_RESUMING;
-		return;
-	}
-
 	/*
 	 * If the netdev is down, or is going down, we're not interested
 	 * in link state events, so just mark our idea of the link as down
@@ -1734,15 +2158,21 @@ static void fec_enet_adjust_link(struct net_device *ndev)
 
 		/* if any of the above changed restart the FEC */
 		if (status_change) {
+			netif_stop_queue(ndev);
 			napi_disable(&fep->napi);
 			netif_tx_lock_bh(ndev);
 			fec_restart(ndev);
-			netif_wake_queue(ndev);
+			netif_tx_wake_all_queues(ndev);
 			netif_tx_unlock_bh(ndev);
 			napi_enable(&fep->napi);
 		}
+		if (fep->quirks & FEC_QUIRK_HAS_EEE)
+			fec_enet_eee_mode_set(ndev,
+					      phy_dev->eee_cfg.tx_lpi_timer,
+					      phy_dev->enable_tx_lpi);
 	} else {
 		if (fep->link) {
+			netif_stop_queue(ndev);
 			napi_disable(&fep->napi);
 			netif_tx_lock_bh(ndev);
 			fec_stop(ndev);
@@ -1757,32 +2187,44 @@ static void fec_enet_adjust_link(struct net_device *ndev)
 		phy_print_status(phy_dev);
 }
 
-static int fec_enet_mdio_read(struct mii_bus *bus, int mii_id, int regnum)
+static int fec_enet_mdio_wait(struct fec_enet_private *fep)
+{
+	uint ievent;
+	int ret;
+
+	ret = readl_poll_timeout_atomic(fep->hwp + FEC_IEVENT, ievent,
+					ievent & FEC_ENET_MII, 2, 30000);
+
+	if (!ret)
+		writel(FEC_ENET_MII, fep->hwp + FEC_IEVENT);
+
+	return ret;
+}
+
+static int fec_enet_mdio_read_c22(struct mii_bus *bus, int mii_id, int regnum)
 {
 	struct fec_enet_private *fep = bus->priv;
 	struct device *dev = &fep->pdev->dev;
-	unsigned long time_left;
-	int ret = 0;
+	int ret = 0, frame_start, frame_addr, frame_op;
 
-	ret = pm_runtime_get_sync(dev);
+	ret = pm_runtime_resume_and_get(dev);
 	if (ret < 0)
 		return ret;
 
-	fep->mii_timeout = 0;
-	reinit_completion(&fep->mdio_done);
+	/* C22 read */
+	frame_op = FEC_MMFR_OP_READ;
+	frame_start = FEC_MMFR_ST;
+	frame_addr = regnum;
 
 	/* start a read op */
-	writel(FEC_MMFR_ST | FEC_MMFR_OP_READ |
-		FEC_MMFR_PA(mii_id) | FEC_MMFR_RA(regnum) |
-		FEC_MMFR_TA, fep->hwp + FEC_MII_DATA);
+	writel(frame_start | frame_op |
+	       FEC_MMFR_PA(mii_id) | FEC_MMFR_RA(frame_addr) |
+	       FEC_MMFR_TA, fep->hwp + FEC_MII_DATA);
 
 	/* wait for end of transfer */
-	time_left = wait_for_completion_timeout(&fep->mdio_done,
-			usecs_to_jiffies(FEC_MII_TIMEOUT));
-	if (time_left == 0) {
-		fep->mii_timeout = 1;
+	ret = fec_enet_mdio_wait(fep);
+	if (ret) {
 		netdev_err(fep->netdev, "MDIO read timeout\n");
-		ret = -ETIMEDOUT;
 		goto out;
 	}
 
@@ -1795,57 +2237,162 @@ out:
 	return ret;
 }
 
-static int fec_enet_mdio_write(struct mii_bus *bus, int mii_id, int regnum,
-			   u16 value)
+static int fec_enet_mdio_read_c45(struct mii_bus *bus, int mii_id,
+				  int devad, int regnum)
 {
 	struct fec_enet_private *fep = bus->priv;
 	struct device *dev = &fep->pdev->dev;
-	unsigned long time_left;
-	int ret;
+	int ret = 0, frame_start, frame_op;
 
-	ret = pm_runtime_get_sync(dev);
+	ret = pm_runtime_resume_and_get(dev);
 	if (ret < 0)
 		return ret;
-	else
-		ret = 0;
 
-	fep->mii_timeout = 0;
-	reinit_completion(&fep->mdio_done);
+	frame_start = FEC_MMFR_ST_C45;
+
+	/* write address */
+	writel(frame_start | FEC_MMFR_OP_ADDR_WRITE |
+	       FEC_MMFR_PA(mii_id) | FEC_MMFR_RA(devad) |
+	       FEC_MMFR_TA | (regnum & 0xFFFF),
+	       fep->hwp + FEC_MII_DATA);
+
+	/* wait for end of transfer */
+	ret = fec_enet_mdio_wait(fep);
+	if (ret) {
+		netdev_err(fep->netdev, "MDIO address write timeout\n");
+		goto out;
+	}
+
+	frame_op = FEC_MMFR_OP_READ_C45;
+
+	/* start a read op */
+	writel(frame_start | frame_op |
+	       FEC_MMFR_PA(mii_id) | FEC_MMFR_RA(devad) |
+	       FEC_MMFR_TA, fep->hwp + FEC_MII_DATA);
+
+	/* wait for end of transfer */
+	ret = fec_enet_mdio_wait(fep);
+	if (ret) {
+		netdev_err(fep->netdev, "MDIO read timeout\n");
+		goto out;
+	}
+
+	ret = FEC_MMFR_DATA(readl(fep->hwp + FEC_MII_DATA));
+
+out:
+	pm_runtime_mark_last_busy(dev);
+	pm_runtime_put_autosuspend(dev);
+
+	return ret;
+}
+
+static int fec_enet_mdio_write_c22(struct mii_bus *bus, int mii_id, int regnum,
+				   u16 value)
+{
+	struct fec_enet_private *fep = bus->priv;
+	struct device *dev = &fep->pdev->dev;
+	int ret, frame_start, frame_addr;
+
+	ret = pm_runtime_resume_and_get(dev);
+	if (ret < 0)
+		return ret;
+
+	/* C22 write */
+	frame_start = FEC_MMFR_ST;
+	frame_addr = regnum;
 
 	/* start a write op */
-	writel(FEC_MMFR_ST | FEC_MMFR_OP_WRITE |
-		FEC_MMFR_PA(mii_id) | FEC_MMFR_RA(regnum) |
-		FEC_MMFR_TA | FEC_MMFR_DATA(value),
-		fep->hwp + FEC_MII_DATA);
+	writel(frame_start | FEC_MMFR_OP_WRITE |
+	       FEC_MMFR_PA(mii_id) | FEC_MMFR_RA(frame_addr) |
+	       FEC_MMFR_TA | FEC_MMFR_DATA(value),
+	       fep->hwp + FEC_MII_DATA);
 
 	/* wait for end of transfer */
-	time_left = wait_for_completion_timeout(&fep->mdio_done,
-			usecs_to_jiffies(FEC_MII_TIMEOUT));
-	if (time_left == 0) {
-		fep->mii_timeout = 1;
+	ret = fec_enet_mdio_wait(fep);
+	if (ret)
 		netdev_err(fep->netdev, "MDIO write timeout\n");
-		ret  = -ETIMEDOUT;
+
+	pm_runtime_mark_last_busy(dev);
+	pm_runtime_put_autosuspend(dev);
+
+	return ret;
+}
+
+static int fec_enet_mdio_write_c45(struct mii_bus *bus, int mii_id,
+				   int devad, int regnum, u16 value)
+{
+	struct fec_enet_private *fep = bus->priv;
+	struct device *dev = &fep->pdev->dev;
+	int ret, frame_start;
+
+	ret = pm_runtime_resume_and_get(dev);
+	if (ret < 0)
+		return ret;
+
+	frame_start = FEC_MMFR_ST_C45;
+
+	/* write address */
+	writel(frame_start | FEC_MMFR_OP_ADDR_WRITE |
+	       FEC_MMFR_PA(mii_id) | FEC_MMFR_RA(devad) |
+	       FEC_MMFR_TA | (regnum & 0xFFFF),
+	       fep->hwp + FEC_MII_DATA);
+
+	/* wait for end of transfer */
+	ret = fec_enet_mdio_wait(fep);
+	if (ret) {
+		netdev_err(fep->netdev, "MDIO address write timeout\n");
+		goto out;
 	}
 
+	/* start a write op */
+	writel(frame_start | FEC_MMFR_OP_WRITE |
+	       FEC_MMFR_PA(mii_id) | FEC_MMFR_RA(devad) |
+	       FEC_MMFR_TA | FEC_MMFR_DATA(value),
+	       fep->hwp + FEC_MII_DATA);
+
+	/* wait for end of transfer */
+	ret = fec_enet_mdio_wait(fep);
+	if (ret)
+		netdev_err(fep->netdev, "MDIO write timeout\n");
+
+out:
 	pm_runtime_mark_last_busy(dev);
 	pm_runtime_put_autosuspend(dev);
 
 	return ret;
 }
 
+static void fec_enet_phy_reset_after_clk_enable(struct net_device *ndev)
+{
+	struct fec_enet_private *fep = netdev_priv(ndev);
+	struct phy_device *phy_dev = ndev->phydev;
+
+	if (phy_dev) {
+		phy_reset_after_clk_enable(phy_dev);
+	} else if (fep->phy_node) {
+		/*
+		 * If the PHY still is not bound to the MAC, but there is
+		 * OF PHY node and a matching PHY device instance already,
+		 * use the OF PHY node to obtain the PHY device instance,
+		 * and then use that PHY device instance when triggering
+		 * the PHY reset.
+		 */
+		phy_dev = of_phy_find_device(fep->phy_node);
+		phy_reset_after_clk_enable(phy_dev);
+		if (phy_dev)
+			put_device(&phy_dev->mdio.dev);
+	}
+}
+
 static int fec_enet_clk_enable(struct net_device *ndev, bool enable)
 {
 	struct fec_enet_private *fep = netdev_priv(ndev);
 	int ret;
 
 	if (enable) {
-		ret = clk_prepare_enable(fep->clk_ahb);
-		if (ret)
-			return ret;
-
 		ret = clk_prepare_enable(fep->clk_enet_out);
 		if (ret)
-			goto failed_clk_enet_out;
+			return ret;
 
 		if (fep->clk_ptp) {
 			mutex_lock(&fep->ptp_clk_mutex);
@@ -1862,8 +2409,13 @@ static int fec_enet_clk_enable(struct net_device *ndev, bool enable)
 		ret = clk_prepare_enable(fep->clk_ref);
 		if (ret)
 			goto failed_clk_ref;
+
+		ret = clk_prepare_enable(fep->clk_2x_txclk);
+		if (ret)
+			goto failed_clk_2x_txclk;
+
+		fec_enet_phy_reset_after_clk_enable(ndev);
 	} else {
-		clk_disable_unprepare(fep->clk_ahb);
 		clk_disable_unprepare(fep->clk_enet_out);
 		if (fep->clk_ptp) {
 			mutex_lock(&fep->ptp_clk_mutex);
@@ -1872,22 +2424,55 @@ static int fec_enet_clk_enable(struct net_device *ndev, bool enable)
 			mutex_unlock(&fep->ptp_clk_mutex);
 		}
 		clk_disable_unprepare(fep->clk_ref);
+		clk_disable_unprepare(fep->clk_2x_txclk);
 	}
 
 	return 0;
 
-failed_clk_ref:
+failed_clk_2x_txclk:
 	if (fep->clk_ref)
 		clk_disable_unprepare(fep->clk_ref);
+failed_clk_ref:
+	if (fep->clk_ptp) {
+		mutex_lock(&fep->ptp_clk_mutex);
+		clk_disable_unprepare(fep->clk_ptp);
+		fep->ptp_clk_on = false;
+		mutex_unlock(&fep->ptp_clk_mutex);
+	}
 failed_clk_ptp:
-	if (fep->clk_enet_out)
-		clk_disable_unprepare(fep->clk_enet_out);
-failed_clk_enet_out:
-		clk_disable_unprepare(fep->clk_ahb);
+	clk_disable_unprepare(fep->clk_enet_out);
 
 	return ret;
 }
 
+static int fec_enet_parse_rgmii_delay(struct fec_enet_private *fep,
+				      struct device_node *np)
+{
+	u32 rgmii_tx_delay, rgmii_rx_delay;
+
+	/* For rgmii tx internal delay, valid values are 0ps and 2000ps */
+	if (!of_property_read_u32(np, "tx-internal-delay-ps", &rgmii_tx_delay)) {
+		if (rgmii_tx_delay != 0 && rgmii_tx_delay != 2000) {
+			dev_err(&fep->pdev->dev, "The only allowed RGMII TX delay values are: 0ps, 2000ps");
+			return -EINVAL;
+		} else if (rgmii_tx_delay == 2000) {
+			fep->rgmii_txc_dly = true;
+		}
+	}
+
+	/* For rgmii rx internal delay, valid values are 0ps and 2000ps */
+	if (!of_property_read_u32(np, "rx-internal-delay-ps", &rgmii_rx_delay)) {
+		if (rgmii_rx_delay != 0 && rgmii_rx_delay != 2000) {
+			dev_err(&fep->pdev->dev, "The only allowed RGMII RX delay values are: 0ps, 2000ps");
+			return -EINVAL;
+		} else if (rgmii_rx_delay == 2000) {
+			fep->rgmii_rxc_dly = true;
+		}
+	}
+
+	return 0;
+}
+
 static int fec_enet_mii_probe(struct net_device *ndev)
 {
 	struct fec_enet_private *fep = netdev_priv(ndev);
@@ -1901,8 +2486,10 @@ static int fec_enet_mii_probe(struct net_device *ndev)
 		phy_dev = of_phy_connect(ndev, fep->phy_node,
 					 &fec_enet_adjust_link, 0,
 					 fep->phy_interface);
-		if (!phy_dev)
+		if (!phy_dev) {
+			netdev_err(ndev, "Unable to connect to phy\n");
 			return -ENODEV;
+		}
 	} else {
 		/* check for attached phy */
 		for (phy_id = 0; (phy_id < PHY_MAX_ADDR); phy_id++) {
@@ -1910,13 +2497,13 @@ static int fec_enet_mii_probe(struct net_device *ndev)
 				continue;
 			if (dev_id--)
 				continue;
-			strlcpy(mdio_bus_id, fep->mii_bus->id, MII_BUS_ID_SIZE);
+			strscpy(mdio_bus_id, fep->mii_bus->id, MII_BUS_ID_SIZE);
 			break;
 		}
 
 		if (phy_id >= PHY_MAX_ADDR) {
 			netdev_info(ndev, "no PHY, assuming direct connection to switch\n");
-			strlcpy(mdio_bus_id, "fixed-0", MII_BUS_ID_SIZE);
+			strscpy(mdio_bus_id, "fixed-0", MII_BUS_ID_SIZE);
 			phy_id = 0;
 		}
 
@@ -1933,16 +2520,18 @@ static int fec_enet_mii_probe(struct net_device *ndev)
 
 	/* mask with MAC supported features */
 	if (fep->quirks & FEC_QUIRK_HAS_GBIT) {
-		phy_dev->supported &= PHY_GBIT_FEATURES;
-		phy_dev->supported &= ~SUPPORTED_1000baseT_Half;
+		phy_set_max_speed(phy_dev, 1000);
+		phy_remove_link_mode(phy_dev,
+				     ETHTOOL_LINK_MODE_1000baseT_Half_BIT);
 #if !defined(CONFIG_M5272)
-		phy_dev->supported |= SUPPORTED_Pause;
+		phy_support_sym_pause(phy_dev);
 #endif
 	}
 	else
-		phy_dev->supported &= PHY_BASIC_FEATURES;
+		phy_set_max_speed(phy_dev, 100);
 
-	phy_dev->advertising = phy_dev->supported;
+	if (fep->quirks & FEC_QUIRK_HAS_EEE)
+		phy_support_eee(phy_dev);
 
 	fep->link = 0;
 	fep->full_duplex = 0;
@@ -1957,9 +2546,13 @@ static int fec_enet_mii_init(struct platform_device *pdev)
 	static struct mii_bus *fec0_mii_bus;
 	struct net_device *ndev = platform_get_drvdata(pdev);
 	struct fec_enet_private *fep = netdev_priv(ndev);
+	bool suppress_preamble = false;
+	struct phy_device *phydev;
 	struct device_node *node;
 	int err = -ENXIO;
 	u32 mii_speed, holdtime;
+	u32 bus_freq;
+	int addr;
 
 	/*
 	 * The i.MX28 dual fec interfaces are not equal.
@@ -1987,17 +2580,23 @@ static int fec_enet_mii_init(struct platform_device *pdev)
 		return -ENOENT;
 	}
 
-	fep->mii_timeout = 0;
+	bus_freq = 2500000; /* 2.5MHz by default */
+	node = of_get_child_by_name(pdev->dev.of_node, "mdio");
+	if (node) {
+		of_property_read_u32(node, "clock-frequency", &bus_freq);
+		suppress_preamble = of_property_read_bool(node,
+							  "suppress-preamble");
+	}
 
 	/*
-	 * Set MII speed to 2.5 MHz (= clk_get_rate() / 2 * phy_speed)
+	 * Set MII speed (= clk_get_rate() / 2 * phy_speed)
 	 *
 	 * The formula for FEC MDC is 'ref_freq / (MII_SPEED x 2)' while
 	 * for ENET-MAC is 'ref_freq / ((MII_SPEED + 1) x 2)'.  The i.MX28
 	 * Reference Manual has an error on this, and gets fixed on i.MX6Q
 	 * document.
 	 */
-	mii_speed = DIV_ROUND_UP(clk_get_rate(fep->clk_ipg), 5000000);
+	mii_speed = DIV_ROUND_UP(clk_get_rate(fep->clk_ipg), bus_freq * 2);
 	if (fep->quirks & FEC_QUIRK_ENET_MAC)
 		mii_speed--;
 	if (mii_speed > 63) {
@@ -2024,8 +2623,26 @@ static int fec_enet_mii_init(struct platform_device *pdev)
 
 	fep->phy_speed = mii_speed << 1 | holdtime << 8;
 
+	if (suppress_preamble)
+		fep->phy_speed |= BIT(7);
+
+	if (fep->quirks & FEC_QUIRK_CLEAR_SETUP_MII) {
+		/* Clear MMFR to avoid to generate MII event by writing MSCR.
+		 * MII event generation condition:
+		 * - writing MSCR:
+		 *	- mmfr[31:0]_not_zero & mscr[7:0]_is_zero &
+		 *	  mscr_reg_data_in[7:0] != 0
+		 * - writing MMFR:
+		 *	- mscr[7:0]_not_zero
+		 */
+		writel(0, fep->hwp + FEC_MII_DATA);
+	}
+
 	writel(fep->phy_speed, fep->hwp + FEC_MII_SPEED);
 
+	/* Clear any pending transaction complete indication */
+	writel(FEC_ENET_MII, fep->hwp + FEC_IEVENT);
+
 	fep->mii_bus = mdiobus_alloc();
 	if (fep->mii_bus == NULL) {
 		err = -ENOMEM;
@@ -2033,23 +2650,28 @@ static int fec_enet_mii_init(struct platform_device *pdev)
 	}
 
 	fep->mii_bus->name = "fec_enet_mii_bus";
-	fep->mii_bus->read = fec_enet_mdio_read;
-	fep->mii_bus->write = fec_enet_mdio_write;
+	fep->mii_bus->read = fec_enet_mdio_read_c22;
+	fep->mii_bus->write = fec_enet_mdio_write_c22;
+	if (fep->quirks & FEC_QUIRK_HAS_MDIO_C45) {
+		fep->mii_bus->read_c45 = fec_enet_mdio_read_c45;
+		fep->mii_bus->write_c45 = fec_enet_mdio_write_c45;
+	}
 	snprintf(fep->mii_bus->id, MII_BUS_ID_SIZE, "%s-%x",
 		pdev->name, fep->dev_id + 1);
 	fep->mii_bus->priv = fep;
 	fep->mii_bus->parent = &pdev->dev;
 
-	node = of_get_child_by_name(pdev->dev.of_node, "mdio");
-	if (node) {
-		err = of_mdiobus_register(fep->mii_bus, node);
-		of_node_put(node);
-	} else {
-		err = mdiobus_register(fep->mii_bus);
-	}
-
+	err = of_mdiobus_register(fep->mii_bus, node);
 	if (err)
 		goto err_out_free_mdiobus;
+	of_node_put(node);
+
+	/* find all the PHY devices on the bus and set mac_managed_pm to true */
+	for (addr = 0; addr < PHY_MAX_ADDR; addr++) {
+		phydev = mdiobus_get_phy(fep->mii_bus, addr);
+		if (phydev)
+			phydev->mac_managed_pm = true;
+	}
 
 	mii_cnt++;
 
@@ -2062,6 +2684,7 @@ static int fec_enet_mii_init(struct platform_device *pdev)
 err_out_free_mdiobus:
 	mdiobus_free(fep->mii_bus);
 err_out:
+	of_node_put(node);
 	return err;
 }
 
@@ -2078,10 +2701,9 @@ static void fec_enet_get_drvinfo(struct net_device *ndev,
 {
 	struct fec_enet_private *fep = netdev_priv(ndev);
 
-	strlcpy(info->driver, fep->pdev->dev.driver->name,
+	strscpy(info->driver, fep->pdev->dev.driver->name,
 		sizeof(info->driver));
-	strlcpy(info->version, "Revision: 1.0", sizeof(info->version));
-	strlcpy(info->bus_info, dev_name(&ndev->dev), sizeof(info->bus_info));
+	strscpy(info->bus_info, dev_name(&ndev->dev), sizeof(info->bus_info));
 }
 
 static int fec_enet_get_regs_len(struct net_device *ndev)
@@ -2099,7 +2721,9 @@ static int fec_enet_get_regs_len(struct net_device *ndev)
 
 /* List of registers that can be safety be read to dump them with ethtool */
 #if defined(CONFIG_M523x) || defined(CONFIG_M527x) || defined(CONFIG_M528x) || \
-	defined(CONFIG_M520x) || defined(CONFIG_M532x) || defined(CONFIG_ARM)
+	defined(CONFIG_M520x) || defined(CONFIG_M532x) || defined(CONFIG_ARM) || \
+	defined(CONFIG_ARM64) || defined(CONFIG_COMPILE_TEST)
+static __u32 fec_enet_register_version = 2;
 static u32 fec_enet_register_offset[] = {
 	FEC_IEVENT, FEC_IMASK, FEC_R_DES_ACTIVE_0, FEC_X_DES_ACTIVE_0,
 	FEC_ECNTRL, FEC_MII_DATA, FEC_MII_SPEED, FEC_MIB_CTRLSTAT, FEC_R_CNTRL,
@@ -2129,7 +2753,33 @@ static u32 fec_enet_register_offset[] = {
 	IEEE_R_DROP, IEEE_R_FRAME_OK, IEEE_R_CRC, IEEE_R_ALIGN, IEEE_R_MACERR,
 	IEEE_R_FDXFC, IEEE_R_OCTETS_OK
 };
+/* for i.MX6ul */
+static u32 fec_enet_register_offset_6ul[] = {
+	FEC_IEVENT, FEC_IMASK, FEC_R_DES_ACTIVE_0, FEC_X_DES_ACTIVE_0,
+	FEC_ECNTRL, FEC_MII_DATA, FEC_MII_SPEED, FEC_MIB_CTRLSTAT, FEC_R_CNTRL,
+	FEC_X_CNTRL, FEC_ADDR_LOW, FEC_ADDR_HIGH, FEC_OPD, FEC_TXIC0, FEC_RXIC0,
+	FEC_HASH_TABLE_HIGH, FEC_HASH_TABLE_LOW, FEC_GRP_HASH_TABLE_HIGH,
+	FEC_GRP_HASH_TABLE_LOW, FEC_X_WMRK, FEC_R_DES_START_0,
+	FEC_X_DES_START_0, FEC_R_BUFF_SIZE_0, FEC_R_FIFO_RSFL, FEC_R_FIFO_RSEM,
+	FEC_R_FIFO_RAEM, FEC_R_FIFO_RAFL, FEC_RACC,
+	RMON_T_DROP, RMON_T_PACKETS, RMON_T_BC_PKT, RMON_T_MC_PKT,
+	RMON_T_CRC_ALIGN, RMON_T_UNDERSIZE, RMON_T_OVERSIZE, RMON_T_FRAG,
+	RMON_T_JAB, RMON_T_COL, RMON_T_P64, RMON_T_P65TO127, RMON_T_P128TO255,
+	RMON_T_P256TO511, RMON_T_P512TO1023, RMON_T_P1024TO2047,
+	RMON_T_P_GTE2048, RMON_T_OCTETS,
+	IEEE_T_DROP, IEEE_T_FRAME_OK, IEEE_T_1COL, IEEE_T_MCOL, IEEE_T_DEF,
+	IEEE_T_LCOL, IEEE_T_EXCOL, IEEE_T_MACERR, IEEE_T_CSERR, IEEE_T_SQE,
+	IEEE_T_FDXFC, IEEE_T_OCTETS_OK,
+	RMON_R_PACKETS, RMON_R_BC_PKT, RMON_R_MC_PKT, RMON_R_CRC_ALIGN,
+	RMON_R_UNDERSIZE, RMON_R_OVERSIZE, RMON_R_FRAG, RMON_R_JAB,
+	RMON_R_RESVD_O, RMON_R_P64, RMON_R_P65TO127, RMON_R_P128TO255,
+	RMON_R_P256TO511, RMON_R_P512TO1023, RMON_R_P1024TO2047,
+	RMON_R_P_GTE2048, RMON_R_OCTETS,
+	IEEE_R_DROP, IEEE_R_FRAME_OK, IEEE_R_CRC, IEEE_R_ALIGN, IEEE_R_MACERR,
+	IEEE_R_FDXFC, IEEE_R_OCTETS_OK
+};
 #else
+static __u32 fec_enet_register_version = 1;
 static u32 fec_enet_register_offset[] = {
 	FEC_ECNTRL, FEC_IEVENT, FEC_IMASK, FEC_IVEC, FEC_R_DES_ACTIVE_0,
 	FEC_R_DES_ACTIVE_1, FEC_R_DES_ACTIVE_2, FEC_X_DES_ACTIVE_0,
@@ -2148,34 +2798,64 @@ static void fec_enet_get_regs(struct net_device *ndev,
 {
 	struct fec_enet_private *fep = netdev_priv(ndev);
 	u32 __iomem *theregs = (u32 __iomem *)fep->hwp;
+	struct device *dev = &fep->pdev->dev;
 	u32 *buf = (u32 *)regbuf;
 	u32 i, off;
+	int ret;
+#if defined(CONFIG_M523x) || defined(CONFIG_M527x) || defined(CONFIG_M528x) || \
+	defined(CONFIG_M520x) || defined(CONFIG_M532x) || defined(CONFIG_ARM) || \
+	defined(CONFIG_ARM64) || defined(CONFIG_COMPILE_TEST)
+	u32 *reg_list;
+	u32 reg_cnt;
+
+	if (!of_machine_is_compatible("fsl,imx6ul")) {
+		reg_list = fec_enet_register_offset;
+		reg_cnt = ARRAY_SIZE(fec_enet_register_offset);
+	} else {
+		reg_list = fec_enet_register_offset_6ul;
+		reg_cnt = ARRAY_SIZE(fec_enet_register_offset_6ul);
+	}
+#else
+	/* coldfire */
+	static u32 *reg_list = fec_enet_register_offset;
+	static const u32 reg_cnt = ARRAY_SIZE(fec_enet_register_offset);
+#endif
+	ret = pm_runtime_resume_and_get(dev);
+	if (ret < 0)
+		return;
+
+	regs->version = fec_enet_register_version;
 
 	memset(buf, 0, regs->len);
 
-	for (i = 0; i < ARRAY_SIZE(fec_enet_register_offset); i++) {
-		off = fec_enet_register_offset[i] / 4;
+	for (i = 0; i < reg_cnt; i++) {
+		off = reg_list[i];
+
+		if ((off == FEC_R_BOUND || off == FEC_R_FSTART) &&
+		    !(fep->quirks & FEC_QUIRK_HAS_FRREG))
+			continue;
+
+		off >>= 2;
 		buf[off] = readl(&theregs[off]);
 	}
+
+	pm_runtime_mark_last_busy(dev);
+	pm_runtime_put_autosuspend(dev);
 }
 
 static int fec_enet_get_ts_info(struct net_device *ndev,
-				struct ethtool_ts_info *info)
+				struct kernel_ethtool_ts_info *info)
 {
 	struct fec_enet_private *fep = netdev_priv(ndev);
 
 	if (fep->bufdesc_ex) {
 
 		info->so_timestamping = SOF_TIMESTAMPING_TX_SOFTWARE |
-					SOF_TIMESTAMPING_RX_SOFTWARE |
-					SOF_TIMESTAMPING_SOFTWARE |
 					SOF_TIMESTAMPING_TX_HARDWARE |
 					SOF_TIMESTAMPING_RX_HARDWARE |
 					SOF_TIMESTAMPING_RAW_HARDWARE;
 		if (fep->ptp_clock)
 			info->phc_index = ptp_clock_index(fep->ptp_clock);
-		else
-			info->phc_index = -1;
 
 		info->tx_types = (1 << HWTSTAMP_TX_OFF) |
 				 (1 << HWTSTAMP_TX_ON);
@@ -2220,13 +2900,8 @@ static int fec_enet_set_pauseparam(struct net_device *ndev,
 	fep->pause_flag |= pause->rx_pause ? FEC_PAUSE_FLAG_ENABLE : 0;
 	fep->pause_flag |= pause->autoneg ? FEC_PAUSE_FLAG_AUTONEG : 0;
 
-	if (pause->rx_pause || pause->autoneg) {
-		ndev->phydev->supported |= ADVERTISED_Pause;
-		ndev->phydev->advertising |= ADVERTISED_Pause;
-	} else {
-		ndev->phydev->supported &= ~ADVERTISED_Pause;
-		ndev->phydev->advertising &= ~ADVERTISED_Pause;
-	}
+	phy_set_sym_pause(ndev->phydev, pause->rx_pause, pause->tx_pause,
+			  pause->autoneg);
 
 	if (pause->autoneg) {
 		if (netif_running(ndev))
@@ -2237,7 +2912,7 @@ static int fec_enet_set_pauseparam(struct net_device *ndev,
 		napi_disable(&fep->napi);
 		netif_tx_lock_bh(ndev);
 		fec_restart(ndev);
-		netif_wake_queue(ndev);
+		netif_tx_wake_all_queues(ndev);
 		netif_tx_unlock_bh(ndev);
 		napi_enable(&fep->napi);
 	}
@@ -2313,6 +2988,16 @@ static const struct fec_stat {
 
 #define FEC_STATS_SIZE		(ARRAY_SIZE(fec_stats) * sizeof(u64))
 
+static const char *fec_xdp_stat_strs[XDP_STATS_TOTAL] = {
+	"rx_xdp_redirect",           /* RX_XDP_REDIRECT = 0, */
+	"rx_xdp_pass",               /* RX_XDP_PASS, */
+	"rx_xdp_drop",               /* RX_XDP_DROP, */
+	"rx_xdp_tx",                 /* RX_XDP_TX, */
+	"rx_xdp_tx_errors",          /* RX_XDP_TX_ERRORS, */
+	"tx_xdp_xmit",               /* TX_XDP_XMIT, */
+	"tx_xdp_xmit_errors",        /* TX_XDP_XMIT_ERRORS, */
+};
+
 static void fec_enet_update_ethtool_stats(struct net_device *dev)
 {
 	struct fec_enet_private *fep = netdev_priv(dev);
@@ -2322,6 +3007,42 @@ static void fec_enet_update_ethtool_stats(struct net_device *dev)
 		fep->ethtool_stats[i] = readl(fep->hwp + fec_stats[i].offset);
 }
 
+static void fec_enet_get_xdp_stats(struct fec_enet_private *fep, u64 *data)
+{
+	u64 xdp_stats[XDP_STATS_TOTAL] = { 0 };
+	struct fec_enet_priv_rx_q *rxq;
+	int i, j;
+
+	for (i = fep->num_rx_queues - 1; i >= 0; i--) {
+		rxq = fep->rx_queue[i];
+
+		for (j = 0; j < XDP_STATS_TOTAL; j++)
+			xdp_stats[j] += rxq->stats[j];
+	}
+
+	memcpy(data, xdp_stats, sizeof(xdp_stats));
+}
+
+static void fec_enet_page_pool_stats(struct fec_enet_private *fep, u64 *data)
+{
+#ifdef CONFIG_PAGE_POOL_STATS
+	struct page_pool_stats stats = {};
+	struct fec_enet_priv_rx_q *rxq;
+	int i;
+
+	for (i = fep->num_rx_queues - 1; i >= 0; i--) {
+		rxq = fep->rx_queue[i];
+
+		if (!rxq->page_pool)
+			continue;
+
+		page_pool_get_stats(rxq->page_pool, &stats);
+	}
+
+	page_pool_ethtool_stats_get(data, &stats);
+#endif
+}
+
 static void fec_enet_get_ethtool_stats(struct net_device *dev,
 				       struct ethtool_stats *stats, u64 *data)
 {
@@ -2331,6 +3052,12 @@ static void fec_enet_get_ethtool_stats(struct net_device *dev,
 		fec_enet_update_ethtool_stats(dev);
 
 	memcpy(data, fep->ethtool_stats, FEC_STATS_SIZE);
+	data += FEC_STATS_SIZE / sizeof(u64);
+
+	fec_enet_get_xdp_stats(fep, data);
+	data += XDP_STATS_TOTAL;
+
+	fec_enet_page_pool_stats(fep, data);
 }
 
 static void fec_enet_get_strings(struct net_device *netdev,
@@ -2339,28 +3066,69 @@ static void fec_enet_get_strings(struct net_device *netdev,
 	int i;
 	switch (stringset) {
 	case ETH_SS_STATS:
-		for (i = 0; i < ARRAY_SIZE(fec_stats); i++)
-			memcpy(data + i * ETH_GSTRING_LEN,
-				fec_stats[i].name, ETH_GSTRING_LEN);
+		for (i = 0; i < ARRAY_SIZE(fec_stats); i++) {
+			ethtool_puts(&data, fec_stats[i].name);
+		}
+		for (i = 0; i < ARRAY_SIZE(fec_xdp_stat_strs); i++) {
+			ethtool_puts(&data, fec_xdp_stat_strs[i]);
+		}
+		page_pool_ethtool_stats_get_strings(data);
+
+		break;
+	case ETH_SS_TEST:
+		net_selftest_get_strings(data);
 		break;
 	}
 }
 
 static int fec_enet_get_sset_count(struct net_device *dev, int sset)
 {
+	int count;
+
 	switch (sset) {
 	case ETH_SS_STATS:
-		return ARRAY_SIZE(fec_stats);
+		count = ARRAY_SIZE(fec_stats) + XDP_STATS_TOTAL;
+		count += page_pool_ethtool_stats_get_count();
+		return count;
+
+	case ETH_SS_TEST:
+		return net_selftest_get_count();
 	default:
 		return -EOPNOTSUPP;
 	}
 }
 
+static void fec_enet_clear_ethtool_stats(struct net_device *dev)
+{
+	struct fec_enet_private *fep = netdev_priv(dev);
+	struct fec_enet_priv_rx_q *rxq;
+	int i, j;
+
+	/* Disable MIB statistics counters */
+	writel(FEC_MIB_CTRLSTAT_DISABLE, fep->hwp + FEC_MIB_CTRLSTAT);
+
+	for (i = 0; i < ARRAY_SIZE(fec_stats); i++)
+		writel(0, fep->hwp + fec_stats[i].offset);
+
+	for (i = fep->num_rx_queues - 1; i >= 0; i--) {
+		rxq = fep->rx_queue[i];
+		for (j = 0; j < XDP_STATS_TOTAL; j++)
+			rxq->stats[j] = 0;
+	}
+
+	/* Don't disable MIB statistics counters */
+	writel(0, fep->hwp + FEC_MIB_CTRLSTAT);
+}
+
 #else	/* !defined(CONFIG_M5272) */
 #define FEC_STATS_SIZE	0
 static inline void fec_enet_update_ethtool_stats(struct net_device *dev)
 {
 }
+
+static inline void fec_enet_clear_ethtool_stats(struct net_device *dev)
+{
+}
 #endif /* !defined(CONFIG_M5272) */
 
 /* ITR clock source is enet system clock (clk_ahb).
@@ -2378,31 +3146,29 @@ static int fec_enet_us_to_itr_clock(struct net_device *ndev, int us)
 static void fec_enet_itr_coal_set(struct net_device *ndev)
 {
 	struct fec_enet_private *fep = netdev_priv(ndev);
-	int rx_itr, tx_itr;
+	u32 rx_itr = 0, tx_itr = 0;
+	int rx_ictt, tx_ictt;
 
-	/* Must be greater than zero to avoid unpredictable behavior */
-	if (!fep->rx_time_itr || !fep->rx_pkts_itr ||
-	    !fep->tx_time_itr || !fep->tx_pkts_itr)
-		return;
-
-	/* Select enet system clock as Interrupt Coalescing
-	 * timer Clock Source
-	 */
-	rx_itr = FEC_ITR_CLK_SEL;
-	tx_itr = FEC_ITR_CLK_SEL;
+	rx_ictt = fec_enet_us_to_itr_clock(ndev, fep->rx_time_itr);
+	tx_ictt = fec_enet_us_to_itr_clock(ndev, fep->tx_time_itr);
 
-	/* set ICFT and ICTT */
-	rx_itr |= FEC_ITR_ICFT(fep->rx_pkts_itr);
-	rx_itr |= FEC_ITR_ICTT(fec_enet_us_to_itr_clock(ndev, fep->rx_time_itr));
-	tx_itr |= FEC_ITR_ICFT(fep->tx_pkts_itr);
-	tx_itr |= FEC_ITR_ICTT(fec_enet_us_to_itr_clock(ndev, fep->tx_time_itr));
+	if (rx_ictt > 0 && fep->rx_pkts_itr > 1) {
+		/* Enable with enet system clock as Interrupt Coalescing timer Clock Source */
+		rx_itr = FEC_ITR_EN | FEC_ITR_CLK_SEL;
+		rx_itr |= FEC_ITR_ICFT(fep->rx_pkts_itr);
+		rx_itr |= FEC_ITR_ICTT(rx_ictt);
+	}
 
-	rx_itr |= FEC_ITR_EN;
-	tx_itr |= FEC_ITR_EN;
+	if (tx_ictt > 0 && fep->tx_pkts_itr > 1) {
+		/* Enable with enet system clock as Interrupt Coalescing timer Clock Source */
+		tx_itr = FEC_ITR_EN | FEC_ITR_CLK_SEL;
+		tx_itr |= FEC_ITR_ICFT(fep->tx_pkts_itr);
+		tx_itr |= FEC_ITR_ICTT(tx_ictt);
+	}
 
 	writel(tx_itr, fep->hwp + FEC_TXIC0);
 	writel(rx_itr, fep->hwp + FEC_RXIC0);
-	if (fep->quirks & FEC_QUIRK_HAS_AVB) {
+	if (fep->quirks & FEC_QUIRK_HAS_MULTI_QUEUES) {
 		writel(tx_itr, fep->hwp + FEC_TXIC1);
 		writel(rx_itr, fep->hwp + FEC_RXIC1);
 		writel(tx_itr, fep->hwp + FEC_TXIC2);
@@ -2410,8 +3176,10 @@ static void fec_enet_itr_coal_set(struct net_device *ndev)
 	}
 }
 
-static int
-fec_enet_get_coalesce(struct net_device *ndev, struct ethtool_coalesce *ec)
+static int fec_enet_get_coalesce(struct net_device *ndev,
+				 struct ethtool_coalesce *ec,
+				 struct kernel_ethtool_coalesce *kernel_coal,
+				 struct netlink_ext_ack *extack)
 {
 	struct fec_enet_private *fep = netdev_priv(ndev);
 
@@ -2427,34 +3195,37 @@ fec_enet_get_coalesce(struct net_device *ndev, struct ethtool_coalesce *ec)
 	return 0;
 }
 
-static int
-fec_enet_set_coalesce(struct net_device *ndev, struct ethtool_coalesce *ec)
+static int fec_enet_set_coalesce(struct net_device *ndev,
+				 struct ethtool_coalesce *ec,
+				 struct kernel_ethtool_coalesce *kernel_coal,
+				 struct netlink_ext_ack *extack)
 {
 	struct fec_enet_private *fep = netdev_priv(ndev);
+	struct device *dev = &fep->pdev->dev;
 	unsigned int cycle;
 
 	if (!(fep->quirks & FEC_QUIRK_HAS_COALESCE))
 		return -EOPNOTSUPP;
 
 	if (ec->rx_max_coalesced_frames > 255) {
-		pr_err("Rx coalesced frames exceed hardware limitation\n");
+		dev_err(dev, "Rx coalesced frames exceed hardware limitation\n");
 		return -EINVAL;
 	}
 
 	if (ec->tx_max_coalesced_frames > 255) {
-		pr_err("Tx coalesced frame exceed hardware limitation\n");
+		dev_err(dev, "Tx coalesced frame exceed hardware limitation\n");
 		return -EINVAL;
 	}
 
-	cycle = fec_enet_us_to_itr_clock(ndev, fep->rx_time_itr);
+	cycle = fec_enet_us_to_itr_clock(ndev, ec->rx_coalesce_usecs);
 	if (cycle > 0xFFFF) {
-		pr_err("Rx coalesced usec exceed hardware limitation\n");
+		dev_err(dev, "Rx coalesced usec exceed hardware limitation\n");
 		return -EINVAL;
 	}
 
-	cycle = fec_enet_us_to_itr_clock(ndev, fep->tx_time_itr);
+	cycle = fec_enet_us_to_itr_clock(ndev, ec->tx_coalesce_usecs);
 	if (cycle > 0xFFFF) {
-		pr_err("Rx coalesced usec exceed hardware limitation\n");
+		dev_err(dev, "Tx coalesced usec exceed hardware limitation\n");
 		return -EINVAL;
 	}
 
@@ -2469,55 +3240,32 @@ fec_enet_set_coalesce(struct net_device *ndev, struct ethtool_coalesce *ec)
 	return 0;
 }
 
-static void fec_enet_itr_coal_init(struct net_device *ndev)
+static int
+fec_enet_get_eee(struct net_device *ndev, struct ethtool_keee *edata)
 {
-	struct ethtool_coalesce ec;
+	struct fec_enet_private *fep = netdev_priv(ndev);
 
-	ec.rx_coalesce_usecs = FEC_ITR_ICTT_DEFAULT;
-	ec.rx_max_coalesced_frames = FEC_ITR_ICFT_DEFAULT;
+	if (!(fep->quirks & FEC_QUIRK_HAS_EEE))
+		return -EOPNOTSUPP;
 
-	ec.tx_coalesce_usecs = FEC_ITR_ICTT_DEFAULT;
-	ec.tx_max_coalesced_frames = FEC_ITR_ICFT_DEFAULT;
+	if (!netif_running(ndev))
+		return -ENETDOWN;
 
-	fec_enet_set_coalesce(ndev, &ec);
+	return phy_ethtool_get_eee(ndev->phydev, edata);
 }
 
-static int fec_enet_get_tunable(struct net_device *netdev,
-				const struct ethtool_tunable *tuna,
-				void *data)
+static int
+fec_enet_set_eee(struct net_device *ndev, struct ethtool_keee *edata)
 {
-	struct fec_enet_private *fep = netdev_priv(netdev);
-	int ret = 0;
-
-	switch (tuna->id) {
-	case ETHTOOL_RX_COPYBREAK:
-		*(u32 *)data = fep->rx_copybreak;
-		break;
-	default:
-		ret = -EINVAL;
-		break;
-	}
-
-	return ret;
-}
+	struct fec_enet_private *fep = netdev_priv(ndev);
 
-static int fec_enet_set_tunable(struct net_device *netdev,
-				const struct ethtool_tunable *tuna,
-				const void *data)
-{
-	struct fec_enet_private *fep = netdev_priv(netdev);
-	int ret = 0;
+	if (!(fep->quirks & FEC_QUIRK_HAS_EEE))
+		return -EOPNOTSUPP;
 
-	switch (tuna->id) {
-	case ETHTOOL_RX_COPYBREAK:
-		fep->rx_copybreak = *(u32 *)data;
-		break;
-	default:
-		ret = -EINVAL;
-		break;
-	}
+	if (!netif_running(ndev))
+		return -ENETDOWN;
 
-	return ret;
+	return phy_ethtool_set_eee(ndev->phydev, edata);
 }
 
 static void
@@ -2545,20 +3293,17 @@ fec_enet_set_wol(struct net_device *ndev, struct ethtool_wolinfo *wol)
 		return -EINVAL;
 
 	device_set_wakeup_enable(&ndev->dev, wol->wolopts & WAKE_MAGIC);
-	if (device_may_wakeup(&ndev->dev)) {
+	if (device_may_wakeup(&ndev->dev))
 		fep->wol_flag |= FEC_WOL_FLAG_ENABLE;
-		if (fep->irq[0] > 0)
-			enable_irq_wake(fep->irq[0]);
-	} else {
+	else
 		fep->wol_flag &= (~FEC_WOL_FLAG_ENABLE);
-		if (fep->irq[0] > 0)
-			disable_irq_wake(fep->irq[0]);
-	}
 
 	return 0;
 }
 
 static const struct ethtool_ops fec_enet_ethtool_ops = {
+	.supported_coalesce_params = ETHTOOL_COALESCE_USECS |
+				     ETHTOOL_COALESCE_MAX_FRAMES,
 	.get_drvinfo		= fec_enet_get_drvinfo,
 	.get_regs_len		= fec_enet_get_regs_len,
 	.get_regs		= fec_enet_get_regs,
@@ -2574,71 +3319,61 @@ static const struct ethtool_ops fec_enet_ethtool_ops = {
 	.get_sset_count		= fec_enet_get_sset_count,
 #endif
 	.get_ts_info		= fec_enet_get_ts_info,
-	.get_tunable		= fec_enet_get_tunable,
-	.set_tunable		= fec_enet_set_tunable,
 	.get_wol		= fec_enet_get_wol,
 	.set_wol		= fec_enet_set_wol,
+	.get_eee		= fec_enet_get_eee,
+	.set_eee		= fec_enet_set_eee,
 	.get_link_ksettings	= phy_ethtool_get_link_ksettings,
 	.set_link_ksettings	= phy_ethtool_set_link_ksettings,
+	.self_test		= net_selftest,
 };
 
-static int fec_enet_ioctl(struct net_device *ndev, struct ifreq *rq, int cmd)
-{
-	struct fec_enet_private *fep = netdev_priv(ndev);
-	struct phy_device *phydev = ndev->phydev;
-
-	if (!netif_running(ndev))
-		return -EINVAL;
-
-	if (!phydev)
-		return -ENODEV;
-
-	if (fep->bufdesc_ex) {
-		if (cmd == SIOCSHWTSTAMP)
-			return fec_ptp_set(ndev, rq);
-		if (cmd == SIOCGHWTSTAMP)
-			return fec_ptp_get(ndev, rq);
-	}
-
-	return phy_mii_ioctl(phydev, rq, cmd);
-}
-
 static void fec_enet_free_buffers(struct net_device *ndev)
 {
 	struct fec_enet_private *fep = netdev_priv(ndev);
 	unsigned int i;
-	struct sk_buff *skb;
-	struct bufdesc	*bdp;
 	struct fec_enet_priv_tx_q *txq;
 	struct fec_enet_priv_rx_q *rxq;
 	unsigned int q;
 
 	for (q = 0; q < fep->num_rx_queues; q++) {
 		rxq = fep->rx_queue[q];
-		bdp = rxq->bd.base;
-		for (i = 0; i < rxq->bd.ring_size; i++) {
-			skb = rxq->rx_skbuff[i];
-			rxq->rx_skbuff[i] = NULL;
-			if (skb) {
-				dma_unmap_single(&fep->pdev->dev,
-						 fec32_to_cpu(bdp->cbd_bufaddr),
-						 FEC_ENET_RX_FRSIZE - fep->rx_align,
-						 DMA_FROM_DEVICE);
-				dev_kfree_skb(skb);
-			}
-			bdp = fec_enet_get_nextdesc(bdp, &rxq->bd);
-		}
+		for (i = 0; i < rxq->bd.ring_size; i++)
+			page_pool_put_full_page(rxq->page_pool, rxq->rx_skb_info[i].page, false);
+
+		for (i = 0; i < XDP_STATS_TOTAL; i++)
+			rxq->stats[i] = 0;
+
+		if (xdp_rxq_info_is_reg(&rxq->xdp_rxq))
+			xdp_rxq_info_unreg(&rxq->xdp_rxq);
+		page_pool_destroy(rxq->page_pool);
+		rxq->page_pool = NULL;
 	}
 
 	for (q = 0; q < fep->num_tx_queues; q++) {
 		txq = fep->tx_queue[q];
-		bdp = txq->bd.base;
 		for (i = 0; i < txq->bd.ring_size; i++) {
 			kfree(txq->tx_bounce[i]);
 			txq->tx_bounce[i] = NULL;
-			skb = txq->tx_skbuff[i];
-			txq->tx_skbuff[i] = NULL;
-			dev_kfree_skb(skb);
+
+			if (!txq->tx_buf[i].buf_p) {
+				txq->tx_buf[i].type = FEC_TXBUF_T_SKB;
+				continue;
+			}
+
+			if (txq->tx_buf[i].type == FEC_TXBUF_T_SKB) {
+				dev_kfree_skb(txq->tx_buf[i].buf_p);
+			} else if (txq->tx_buf[i].type == FEC_TXBUF_T_XDP_NDO) {
+				xdp_return_frame(txq->tx_buf[i].buf_p);
+			} else {
+				struct page *page = txq->tx_buf[i].buf_p;
+
+				page_pool_put_page(pp_page_to_nmdesc(page)->pp,
+						   page, 0, false);
+			}
+
+			txq->tx_buf[i].buf_p = NULL;
+			txq->tx_buf[i].type = FEC_TXBUF_T_SKB;
 		}
 	}
 }
@@ -2652,10 +3387,9 @@ static void fec_enet_free_queue(struct net_device *ndev)
 	for (i = 0; i < fep->num_tx_queues; i++)
 		if (fep->tx_queue[i] && fep->tx_queue[i]->tso_hdrs) {
 			txq = fep->tx_queue[i];
-			dma_free_coherent(&fep->pdev->dev,
-					  txq->bd.ring_size * TSO_HEADER_SIZE,
-					  txq->tso_hdrs,
-					  txq->tso_hdrs_dma);
+			fec_dma_free(&fep->pdev->dev,
+				     txq->bd.ring_size * TSO_HEADER_SIZE,
+				     txq->tso_hdrs, txq->tso_hdrs_dma);
 		}
 
 	for (i = 0; i < fep->num_rx_queues; i++)
@@ -2683,13 +3417,11 @@ static int fec_enet_alloc_queue(struct net_device *ndev)
 		fep->total_tx_ring_size += fep->tx_queue[i]->bd.ring_size;
 
 		txq->tx_stop_threshold = FEC_MAX_SKB_DESCS;
-		txq->tx_wake_threshold =
-			(txq->bd.ring_size - txq->tx_stop_threshold) / 2;
+		txq->tx_wake_threshold = FEC_MAX_SKB_DESCS + 2 * MAX_SKB_FRAGS;
 
-		txq->tso_hdrs = dma_alloc_coherent(&fep->pdev->dev,
+		txq->tso_hdrs = fec_dma_alloc(&fep->pdev->dev,
 					txq->bd.ring_size * TSO_HEADER_SIZE,
-					&txq->tso_hdrs_dma,
-					GFP_KERNEL);
+					&txq->tso_hdrs_dma, GFP_KERNEL);
 		if (!txq->tso_hdrs) {
 			ret = -ENOMEM;
 			goto alloc_failed;
@@ -2718,24 +3450,31 @@ static int
 fec_enet_alloc_rxq_buffers(struct net_device *ndev, unsigned int queue)
 {
 	struct fec_enet_private *fep = netdev_priv(ndev);
-	unsigned int i;
-	struct sk_buff *skb;
-	struct bufdesc	*bdp;
 	struct fec_enet_priv_rx_q *rxq;
+	dma_addr_t phys_addr;
+	struct bufdesc	*bdp;
+	struct page *page;
+	int i, err;
 
 	rxq = fep->rx_queue[queue];
 	bdp = rxq->bd.base;
+
+	err = fec_enet_create_page_pool(fep, rxq, rxq->bd.ring_size);
+	if (err < 0) {
+		netdev_err(ndev, "%s failed queue %d (%d)\n", __func__, queue, err);
+		return err;
+	}
+
 	for (i = 0; i < rxq->bd.ring_size; i++) {
-		skb = netdev_alloc_skb(ndev, FEC_ENET_RX_FRSIZE);
-		if (!skb)
+		page = page_pool_dev_alloc_pages(rxq->page_pool);
+		if (!page)
 			goto err_alloc;
 
-		if (fec_enet_new_rxbdp(ndev, bdp, skb)) {
-			dev_kfree_skb(skb);
-			goto err_alloc;
-		}
+		phys_addr = page_pool_get_dma_addr(page) + FEC_ENET_XDP_HEADROOM;
+		bdp->cbd_bufaddr = cpu_to_fec32(phys_addr);
 
-		rxq->rx_skbuff[i] = skb;
+		rxq->rx_skb_info[i].page = page;
+		rxq->rx_skb_info[i].offset = FEC_ENET_XDP_HEADROOM;
 		bdp->cbd_sc = cpu_to_fec16(BD_ENET_RX_EMPTY);
 
 		if (fep->bufdesc_ex) {
@@ -2813,8 +3552,9 @@ fec_enet_open(struct net_device *ndev)
 {
 	struct fec_enet_private *fep = netdev_priv(ndev);
 	int ret;
+	bool reset_again;
 
-	ret = pm_runtime_get_sync(&fep->pdev->dev);
+	ret = pm_runtime_resume_and_get(&fep->pdev->dev);
 	if (ret < 0)
 		return ret;
 
@@ -2823,6 +3563,17 @@ fec_enet_open(struct net_device *ndev)
 	if (ret)
 		goto clk_enable;
 
+	/* During the first fec_enet_open call the PHY isn't probed at this
+	 * point. Therefore the phy_reset_after_clk_enable() call within
+	 * fec_enet_clk_enable() fails. As we need this reset in order to be
+	 * sure the PHY is working correctly we check if we need to reset again
+	 * later when the PHY is probed
+	 */
+	if (ndev->phydev && ndev->phydev->drv)
+		reset_again = false;
+	else
+		reset_again = true;
+
 	/* I should reset the ring buffers here, but I don't yet know
 	 * a simple way to do that.
 	 */
@@ -2834,6 +3585,12 @@ fec_enet_open(struct net_device *ndev)
 	/* Init MAC prior to mii bus probe */
 	fec_restart(ndev);
 
+	/* Call phy_reset_after_clk_enable() again if it failed during
+	 * phy_reset_after_clk_enable() before because the PHY wasn't probed.
+	 */
+	if (reset_again)
+		fec_enet_phy_reset_after_clk_enable(ndev);
+
 	/* Probe and connect to PHY when open the interface */
 	ret = fec_enet_mii_probe(ndev);
 	if (ret)
@@ -2842,6 +3599,9 @@ fec_enet_open(struct net_device *ndev)
 	if (fep->quirks & FEC_QUIRK_ERR006687)
 		imx6q_cpuidle_fec_irqs_used();
 
+	if (fep->quirks & FEC_QUIRK_HAS_PMQOS)
+		cpu_latency_qos_add_request(&fep->pm_qos_req, 0);
+
 	napi_enable(&fep->napi);
 	phy_start(ndev->phydev);
 	netif_tx_start_all_queues(ndev);
@@ -2883,6 +3643,9 @@ fec_enet_close(struct net_device *ndev)
 	fec_enet_update_ethtool_stats(ndev);
 
 	fec_enet_clk_enable(ndev, false);
+	if (fep->quirks & FEC_QUIRK_HAS_PMQOS)
+		cpu_latency_qos_remove_request(&fep->pm_qos_req);
+
 	pinctrl_pm_select_sleep_state(&fep->pdev->dev);
 	pm_runtime_mark_last_busy(&fep->pdev->dev);
 	pm_runtime_put_autosuspend(&fep->pdev->dev);
@@ -2903,13 +3666,12 @@ fec_enet_close(struct net_device *ndev)
  */
 
 #define FEC_HASH_BITS	6		/* #bits in hash */
-#define CRC32_POLY	0xEDB88320
 
 static void set_multicast_list(struct net_device *ndev)
 {
 	struct fec_enet_private *fep = netdev_priv(ndev);
 	struct netdev_hw_addr *ha;
-	unsigned int i, bit, data, crc, tmp;
+	unsigned int crc, tmp;
 	unsigned char hash;
 	unsigned int hash_high = 0, hash_low = 0;
 
@@ -2937,15 +3699,7 @@ static void set_multicast_list(struct net_device *ndev)
 	/* Add the addresses in hash register */
 	netdev_for_each_mc_addr(ha, ndev) {
 		/* calculate crc32 value of mac address */
-		crc = 0xffffffff;
-
-		for (i = 0; i < ndev->addr_len; i++) {
-			data = ha->addr[i];
-			for (bit = 0; bit < 8; bit++, data >>= 1) {
-				crc = (crc >> 1) ^
-				(((crc ^ data) & 1) ? CRC32_POLY : 0);
-			}
-		}
+		crc = ether_crc_le(ndev->addr_len, ha->addr);
 
 		/* only upper 6 bits (FEC_HASH_BITS) are used
 		 * which point to specific bit in the hash registers
@@ -2966,13 +3720,12 @@ static void set_multicast_list(struct net_device *ndev)
 static int
 fec_set_mac_address(struct net_device *ndev, void *p)
 {
-	struct fec_enet_private *fep = netdev_priv(ndev);
 	struct sockaddr *addr = p;
 
 	if (addr) {
 		if (!is_valid_ether_addr(addr->sa_data))
 			return -EADDRNOTAVAIL;
-		memcpy(ndev->dev_addr, addr->sa_data, ndev->addr_len);
+		eth_hw_addr_set(ndev, addr->sa_data);
 	}
 
 	/* Add netif status check here to avoid system hang in below case:
@@ -2983,36 +3736,10 @@ fec_set_mac_address(struct net_device *ndev, void *p)
 	if (!netif_running(ndev))
 		return 0;
 
-	writel(ndev->dev_addr[3] | (ndev->dev_addr[2] << 8) |
-		(ndev->dev_addr[1] << 16) | (ndev->dev_addr[0] << 24),
-		fep->hwp + FEC_ADDR_LOW);
-	writel((ndev->dev_addr[5] << 16) | (ndev->dev_addr[4] << 24),
-		fep->hwp + FEC_ADDR_HIGH);
-	return 0;
-}
-
-#ifdef CONFIG_NET_POLL_CONTROLLER
-/**
- * fec_poll_controller - FEC Poll controller function
- * @dev: The FEC network adapter
- *
- * Polled functionality used by netconsole and others in non interrupt mode
- *
- */
-static void fec_poll_controller(struct net_device *dev)
-{
-	int i;
-	struct fec_enet_private *fep = netdev_priv(dev);
+	fec_set_hw_mac_addr(ndev);
 
-	for (i = 0; i < FEC_IRQ_NUM; i++) {
-		if (fep->irq[i] > 0) {
-			disable_irq(fep->irq[i]);
-			fec_enet_interrupt(fep->irq[i], dev);
-			enable_irq(fep->irq[i]);
-		}
-	}
+	return 0;
 }
-#endif
 
 static inline void fec_enet_set_netdev_features(struct net_device *netdev,
 	netdev_features_t features)
@@ -3053,19 +3780,296 @@ static int fec_set_features(struct net_device *netdev,
 	return 0;
 }
 
+static u16 fec_enet_select_queue(struct net_device *ndev, struct sk_buff *skb,
+				 struct net_device *sb_dev)
+{
+	struct fec_enet_private *fep = netdev_priv(ndev);
+	u16 vlan_tag = 0;
+
+	if (!(fep->quirks & FEC_QUIRK_HAS_AVB))
+		return netdev_pick_tx(ndev, skb, NULL);
+
+	/* VLAN is present in the payload.*/
+	if (eth_type_vlan(skb->protocol)) {
+		struct vlan_ethhdr *vhdr = skb_vlan_eth_hdr(skb);
+
+		vlan_tag = ntohs(vhdr->h_vlan_TCI);
+	/*  VLAN is present in the skb but not yet pushed in the payload.*/
+	} else if (skb_vlan_tag_present(skb)) {
+		vlan_tag = skb->vlan_tci;
+	} else {
+		return vlan_tag;
+	}
+
+	return fec_enet_vlan_pri_to_queue[vlan_tag >> 13];
+}
+
+static int fec_enet_bpf(struct net_device *dev, struct netdev_bpf *bpf)
+{
+	struct fec_enet_private *fep = netdev_priv(dev);
+	bool is_run = netif_running(dev);
+	struct bpf_prog *old_prog;
+
+	switch (bpf->command) {
+	case XDP_SETUP_PROG:
+		/* No need to support the SoCs that require to
+		 * do the frame swap because the performance wouldn't be
+		 * better than the skb mode.
+		 */
+		if (fep->quirks & FEC_QUIRK_SWAP_FRAME)
+			return -EOPNOTSUPP;
+
+		if (!bpf->prog)
+			xdp_features_clear_redirect_target(dev);
+
+		if (is_run) {
+			napi_disable(&fep->napi);
+			netif_tx_disable(dev);
+		}
+
+		old_prog = xchg(&fep->xdp_prog, bpf->prog);
+		if (old_prog)
+			bpf_prog_put(old_prog);
+
+		fec_restart(dev);
+
+		if (is_run) {
+			napi_enable(&fep->napi);
+			netif_tx_start_all_queues(dev);
+		}
+
+		if (bpf->prog)
+			xdp_features_set_redirect_target(dev, false);
+
+		return 0;
+
+	case XDP_SETUP_XSK_POOL:
+		return -EOPNOTSUPP;
+
+	default:
+		return -EOPNOTSUPP;
+	}
+}
+
+static int
+fec_enet_xdp_get_tx_queue(struct fec_enet_private *fep, int index)
+{
+	if (unlikely(index < 0))
+		return 0;
+
+	return (index % fep->num_tx_queues);
+}
+
+static int fec_enet_txq_xmit_frame(struct fec_enet_private *fep,
+				   struct fec_enet_priv_tx_q *txq,
+				   void *frame, u32 dma_sync_len,
+				   bool ndo_xmit)
+{
+	unsigned int index, status, estatus;
+	struct bufdesc *bdp;
+	dma_addr_t dma_addr;
+	int entries_free;
+	u16 frame_len;
+
+	entries_free = fec_enet_get_free_txdesc_num(txq);
+	if (entries_free < MAX_SKB_FRAGS + 1) {
+		netdev_err_once(fep->netdev, "NOT enough BD for SG!\n");
+		return -EBUSY;
+	}
+
+	/* Fill in a Tx ring entry */
+	bdp = txq->bd.cur;
+	status = fec16_to_cpu(bdp->cbd_sc);
+	status &= ~BD_ENET_TX_STATS;
+
+	index = fec_enet_get_bd_index(bdp, &txq->bd);
+
+	if (ndo_xmit) {
+		struct xdp_frame *xdpf = frame;
+
+		dma_addr = dma_map_single(&fep->pdev->dev, xdpf->data,
+					  xdpf->len, DMA_TO_DEVICE);
+		if (dma_mapping_error(&fep->pdev->dev, dma_addr))
+			return -ENOMEM;
+
+		frame_len = xdpf->len;
+		txq->tx_buf[index].buf_p = xdpf;
+		txq->tx_buf[index].type = FEC_TXBUF_T_XDP_NDO;
+	} else {
+		struct xdp_buff *xdpb = frame;
+		struct page *page;
+
+		page = virt_to_page(xdpb->data);
+		dma_addr = page_pool_get_dma_addr(page) +
+			   (xdpb->data - xdpb->data_hard_start);
+		dma_sync_single_for_device(&fep->pdev->dev, dma_addr,
+					   dma_sync_len, DMA_BIDIRECTIONAL);
+		frame_len = xdpb->data_end - xdpb->data;
+		txq->tx_buf[index].buf_p = page;
+		txq->tx_buf[index].type = FEC_TXBUF_T_XDP_TX;
+	}
+
+	status |= (BD_ENET_TX_INTR | BD_ENET_TX_LAST);
+	if (fep->bufdesc_ex)
+		estatus = BD_ENET_TX_INT;
+
+	bdp->cbd_bufaddr = cpu_to_fec32(dma_addr);
+	bdp->cbd_datlen = cpu_to_fec16(frame_len);
+
+	if (fep->bufdesc_ex) {
+		struct bufdesc_ex *ebdp = (struct bufdesc_ex *)bdp;
+
+		if (fep->quirks & FEC_QUIRK_HAS_AVB)
+			estatus |= FEC_TX_BD_FTYPE(txq->bd.qid);
+
+		ebdp->cbd_bdu = 0;
+		ebdp->cbd_esc = cpu_to_fec32(estatus);
+	}
+
+	/* Make sure the updates to rest of the descriptor are performed before
+	 * transferring ownership.
+	 */
+	dma_wmb();
+
+	/* Send it on its way.  Tell FEC it's ready, interrupt when done,
+	 * it's the last BD of the frame, and to put the CRC on the end.
+	 */
+	status |= (BD_ENET_TX_READY | BD_ENET_TX_TC);
+	bdp->cbd_sc = cpu_to_fec16(status);
+
+	/* If this was the last BD in the ring, start at the beginning again. */
+	bdp = fec_enet_get_nextdesc(bdp, &txq->bd);
+
+	/* Make sure the update to bdp are performed before txq->bd.cur. */
+	dma_wmb();
+
+	txq->bd.cur = bdp;
+
+	/* Trigger transmission start */
+	writel(0, txq->bd.reg_desc_active);
+
+	return 0;
+}
+
+static int fec_enet_xdp_tx_xmit(struct fec_enet_private *fep,
+				int cpu, struct xdp_buff *xdp,
+				u32 dma_sync_len)
+{
+	struct fec_enet_priv_tx_q *txq;
+	struct netdev_queue *nq;
+	int queue, ret;
+
+	queue = fec_enet_xdp_get_tx_queue(fep, cpu);
+	txq = fep->tx_queue[queue];
+	nq = netdev_get_tx_queue(fep->netdev, queue);
+
+	__netif_tx_lock(nq, cpu);
+
+	/* Avoid tx timeout as XDP shares the queue with kernel stack */
+	txq_trans_cond_update(nq);
+	ret = fec_enet_txq_xmit_frame(fep, txq, xdp, dma_sync_len, false);
+
+	__netif_tx_unlock(nq);
+
+	return ret;
+}
+
+static int fec_enet_xdp_xmit(struct net_device *dev,
+			     int num_frames,
+			     struct xdp_frame **frames,
+			     u32 flags)
+{
+	struct fec_enet_private *fep = netdev_priv(dev);
+	struct fec_enet_priv_tx_q *txq;
+	int cpu = smp_processor_id();
+	unsigned int sent_frames = 0;
+	struct netdev_queue *nq;
+	unsigned int queue;
+	int i;
+
+	queue = fec_enet_xdp_get_tx_queue(fep, cpu);
+	txq = fep->tx_queue[queue];
+	nq = netdev_get_tx_queue(fep->netdev, queue);
+
+	__netif_tx_lock(nq, cpu);
+
+	/* Avoid tx timeout as XDP shares the queue with kernel stack */
+	txq_trans_cond_update(nq);
+	for (i = 0; i < num_frames; i++) {
+		if (fec_enet_txq_xmit_frame(fep, txq, frames[i], 0, true) < 0)
+			break;
+		sent_frames++;
+	}
+
+	__netif_tx_unlock(nq);
+
+	return sent_frames;
+}
+
+static int fec_hwtstamp_get(struct net_device *ndev,
+			    struct kernel_hwtstamp_config *config)
+{
+	struct fec_enet_private *fep = netdev_priv(ndev);
+
+	if (!netif_running(ndev))
+		return -EINVAL;
+
+	if (!fep->bufdesc_ex)
+		return -EOPNOTSUPP;
+
+	fec_ptp_get(ndev, config);
+
+	return 0;
+}
+
+static int fec_hwtstamp_set(struct net_device *ndev,
+			    struct kernel_hwtstamp_config *config,
+			    struct netlink_ext_ack *extack)
+{
+	struct fec_enet_private *fep = netdev_priv(ndev);
+
+	if (!netif_running(ndev))
+		return -EINVAL;
+
+	if (!fep->bufdesc_ex)
+		return -EOPNOTSUPP;
+
+	return fec_ptp_set(ndev, config, extack);
+}
+
+static int fec_change_mtu(struct net_device *ndev, int new_mtu)
+{
+	struct fec_enet_private *fep = netdev_priv(ndev);
+	int order;
+
+	if (netif_running(ndev))
+		return -EBUSY;
+
+	order = get_order(new_mtu + ETH_HLEN + ETH_FCS_LEN
+			  + FEC_DRV_RESERVE_SPACE);
+	fep->rx_frame_size = (PAGE_SIZE << order) - FEC_DRV_RESERVE_SPACE;
+	fep->pagepool_order = order;
+	WRITE_ONCE(ndev->mtu, new_mtu);
+
+	return 0;
+}
+
 static const struct net_device_ops fec_netdev_ops = {
 	.ndo_open		= fec_enet_open,
 	.ndo_stop		= fec_enet_close,
 	.ndo_start_xmit		= fec_enet_start_xmit,
+	.ndo_select_queue       = fec_enet_select_queue,
 	.ndo_set_rx_mode	= set_multicast_list,
 	.ndo_validate_addr	= eth_validate_addr,
 	.ndo_tx_timeout		= fec_timeout,
 	.ndo_set_mac_address	= fec_set_mac_address,
-	.ndo_do_ioctl		= fec_enet_ioctl,
-#ifdef CONFIG_NET_POLL_CONTROLLER
-	.ndo_poll_controller	= fec_poll_controller,
-#endif
+	.ndo_change_mtu		= fec_change_mtu,
+	.ndo_eth_ioctl		= phy_do_ioctl_running,
 	.ndo_set_features	= fec_set_features,
+	.ndo_bpf		= fec_enet_bpf,
+	.ndo_xdp_xmit		= fec_enet_xdp_xmit,
+	.ndo_hwtstamp_get	= fec_hwtstamp_get,
+	.ndo_hwtstamp_set	= fec_hwtstamp_set,
 };
 
 static const unsigned short offset_des_active_rxq[] = {
@@ -3090,33 +4094,46 @@ static int fec_enet_init(struct net_device *ndev)
 	unsigned dsize = fep->bufdesc_ex ? sizeof(struct bufdesc_ex) :
 			sizeof(struct bufdesc);
 	unsigned dsize_log2 = __fls(dsize);
+	int ret;
 
 	WARN_ON(dsize != (1 << dsize_log2));
-#if defined(CONFIG_ARM)
+#if defined(CONFIG_ARM) || defined(CONFIG_ARM64)
 	fep->rx_align = 0xf;
 	fep->tx_align = 0xf;
 #else
 	fep->rx_align = 0x3;
 	fep->tx_align = 0x3;
 #endif
+	fep->rx_pkts_itr = FEC_ITR_ICFT_DEFAULT;
+	fep->tx_pkts_itr = FEC_ITR_ICFT_DEFAULT;
+	fep->rx_time_itr = FEC_ITR_ICTT_DEFAULT;
+	fep->tx_time_itr = FEC_ITR_ICTT_DEFAULT;
+
+	/* Check mask of the streaming and coherent API */
+	ret = dma_set_mask_and_coherent(&fep->pdev->dev, DMA_BIT_MASK(32));
+	if (ret < 0) {
+		dev_warn(&fep->pdev->dev, "No suitable DMA available\n");
+		return ret;
+	}
 
-	fec_enet_alloc_queue(ndev);
+	ret = fec_enet_alloc_queue(ndev);
+	if (ret)
+		return ret;
 
 	bd_size = (fep->total_tx_ring_size + fep->total_rx_ring_size) * dsize;
 
 	/* Allocate memory for buffer descriptors. */
-	cbd_base = dmam_alloc_coherent(&fep->pdev->dev, bd_size, &bd_dma,
-				       GFP_KERNEL);
+	cbd_base = fec_dmam_alloc(&fep->pdev->dev, bd_size, &bd_dma,
+				  GFP_KERNEL);
 	if (!cbd_base) {
-		return -ENOMEM;
+		ret = -ENOMEM;
+		goto free_queue_mem;
 	}
 
-	memset(cbd_base, 0, bd_size);
-
 	/* Get the Ethernet address */
-	fec_get_mac(ndev);
-	/* make sure MAC we just acquired is programmed into the hw */
-	fec_set_mac_address(ndev, NULL);
+	ret = fec_get_mac(ndev);
+	if (ret)
+		goto free_queue_mem;
 
 	/* Set receive and transmit descriptor base. */
 	for (i = 0; i < fep->num_rx_queues; i++) {
@@ -3158,14 +4175,14 @@ static int fec_enet_init(struct net_device *ndev)
 	ndev->ethtool_ops = &fec_enet_ethtool_ops;
 
 	writel(FEC_RX_DISABLED_IMASK, fep->hwp + FEC_IMASK);
-	netif_napi_add(ndev, &fep->napi, fec_enet_rx_napi, NAPI_POLL_WEIGHT);
+	netif_napi_add(ndev, &fep->napi, fec_enet_rx_napi);
 
 	if (fep->quirks & FEC_QUIRK_HAS_VLAN)
 		/* enable hw VLAN support */
 		ndev->features |= NETIF_F_HW_VLAN_CTAG_RX;
 
 	if (fep->quirks & FEC_QUIRK_HAS_CSUM) {
-		ndev->gso_max_segs = FEC_MAX_TSO_SEGS;
+		netif_set_tso_max_segs(ndev, FEC_MAX_TSO_SEGS);
 
 		/* enable hw accelerator */
 		ndev->features |= (NETIF_F_IP_CSUM | NETIF_F_IPV6_CSUM
@@ -3173,27 +4190,46 @@ static int fec_enet_init(struct net_device *ndev)
 		fep->csum_flags |= FLAG_RX_CSUM_ENABLED;
 	}
 
-	if (fep->quirks & FEC_QUIRK_HAS_AVB) {
+	if (fep->quirks & FEC_QUIRK_HAS_MULTI_QUEUES) {
 		fep->tx_align = 0;
 		fep->rx_align = 0x3f;
 	}
 
 	ndev->hw_features = ndev->features;
 
+	if (!(fep->quirks & FEC_QUIRK_SWAP_FRAME))
+		ndev->xdp_features = NETDEV_XDP_ACT_BASIC |
+				     NETDEV_XDP_ACT_REDIRECT;
+
 	fec_restart(ndev);
 
-	fec_enet_update_ethtool_stats(ndev);
+	if (fep->quirks & FEC_QUIRK_MIB_CLEAR)
+		fec_enet_clear_ethtool_stats(ndev);
+	else
+		fec_enet_update_ethtool_stats(ndev);
 
 	return 0;
+
+free_queue_mem:
+	fec_enet_free_queue(ndev);
+	return ret;
+}
+
+static void fec_enet_deinit(struct net_device *ndev)
+{
+	struct fec_enet_private *fep = netdev_priv(ndev);
+
+	netif_napi_del(&fep->napi);
+	fec_enet_free_queue(ndev);
 }
 
 #ifdef CONFIG_OF
 static int fec_reset_phy(struct platform_device *pdev)
 {
-	int err, phy_reset;
-	bool active_high = false;
+	struct gpio_desc *phy_reset;
 	int msec = 1, phy_post_delay = 0;
 	struct device_node *np = pdev->dev.of_node;
+	int err;
 
 	if (!np)
 		return 0;
@@ -3203,33 +4239,26 @@ static int fec_reset_phy(struct platform_device *pdev)
 	if (!err && msec > 1000)
 		msec = 1;
 
-	phy_reset = of_get_named_gpio(np, "phy-reset-gpios", 0);
-	if (phy_reset == -EPROBE_DEFER)
-		return phy_reset;
-	else if (!gpio_is_valid(phy_reset))
-		return 0;
-
 	err = of_property_read_u32(np, "phy-reset-post-delay", &phy_post_delay);
 	/* valid reset duration should be less than 1s */
 	if (!err && phy_post_delay > 1000)
 		return -EINVAL;
 
-	active_high = of_property_read_bool(np, "phy-reset-active-high");
+	phy_reset = devm_gpiod_get_optional(&pdev->dev, "phy-reset",
+					    GPIOD_OUT_HIGH);
+	if (IS_ERR(phy_reset))
+		return dev_err_probe(&pdev->dev, PTR_ERR(phy_reset),
+				     "failed to get phy-reset-gpios\n");
 
-	err = devm_gpio_request_one(&pdev->dev, phy_reset,
-			active_high ? GPIOF_OUT_INIT_HIGH : GPIOF_OUT_INIT_LOW,
-			"phy-reset");
-	if (err) {
-		dev_err(&pdev->dev, "failed to get phy-reset-gpios: %d\n", err);
-		return err;
-	}
+	if (!phy_reset)
+		return 0;
 
 	if (msec > 20)
 		msleep(msec);
 	else
 		usleep_range(msec * 1000, msec * 1000 + 1000);
 
-	gpio_set_value_cansleep(phy_reset, !active_high);
+	gpiod_set_value_cansleep(phy_reset, 0);
 
 	if (!phy_post_delay)
 		return 0;
@@ -3284,19 +4313,80 @@ fec_enet_get_queue_num(struct platform_device *pdev, int *num_tx, int *num_rx)
 
 }
 
+static int fec_enet_get_irq_cnt(struct platform_device *pdev)
+{
+	int irq_cnt = platform_irq_count(pdev);
+
+	if (irq_cnt > FEC_IRQ_NUM)
+		irq_cnt = FEC_IRQ_NUM;	/* last for pps */
+	else if (irq_cnt == 2)
+		irq_cnt = 1;	/* last for pps */
+	else if (irq_cnt <= 0)
+		irq_cnt = 1;	/* At least 1 irq is needed */
+	return irq_cnt;
+}
+
+static void fec_enet_get_wakeup_irq(struct platform_device *pdev)
+{
+	struct net_device *ndev = platform_get_drvdata(pdev);
+	struct fec_enet_private *fep = netdev_priv(ndev);
+
+	if (fep->quirks & FEC_QUIRK_WAKEUP_FROM_INT2)
+		fep->wake_irq = fep->irq[2];
+	else
+		fep->wake_irq = fep->irq[0];
+}
+
+static int fec_enet_init_stop_mode(struct fec_enet_private *fep,
+				   struct device_node *np)
+{
+	struct device_node *gpr_np;
+	u32 out_val[3];
+	int ret = 0;
+
+	gpr_np = of_parse_phandle(np, "fsl,stop-mode", 0);
+	if (!gpr_np)
+		return 0;
+
+	ret = of_property_read_u32_array(np, "fsl,stop-mode", out_val,
+					 ARRAY_SIZE(out_val));
+	if (ret) {
+		dev_dbg(&fep->pdev->dev, "no stop mode property\n");
+		goto out;
+	}
+
+	fep->stop_gpr.gpr = syscon_node_to_regmap(gpr_np);
+	if (IS_ERR(fep->stop_gpr.gpr)) {
+		dev_err(&fep->pdev->dev, "could not find gpr regmap\n");
+		ret = PTR_ERR(fep->stop_gpr.gpr);
+		fep->stop_gpr.gpr = NULL;
+		goto out;
+	}
+
+	fep->stop_gpr.reg = out_val[1];
+	fep->stop_gpr.bit = out_val[2];
+
+out:
+	of_node_put(gpr_np);
+
+	return ret;
+}
+
 static int
 fec_probe(struct platform_device *pdev)
 {
 	struct fec_enet_private *fep;
 	struct fec_platform_data *pdata;
+	phy_interface_t interface;
 	struct net_device *ndev;
 	int i, irq, ret = 0;
-	struct resource *r;
-	const struct of_device_id *of_id;
 	static int dev_id;
 	struct device_node *np = pdev->dev.of_node, *phy_node;
 	int num_tx_qs;
 	int num_rx_qs;
+	char irq_name[8];
+	int irq_cnt;
+	const struct fec_devinfo *dev_info;
 
 	fec_enet_get_queue_num(pdev, &num_tx_qs, &num_rx_qs);
 
@@ -3311,10 +4401,11 @@ fec_probe(struct platform_device *pdev)
 	/* setup board info structure */
 	fep = netdev_priv(ndev);
 
-	of_id = of_match_device(fec_dt_ids, &pdev->dev);
-	if (of_id)
-		pdev->id_entry = of_id->data;
-	fep->quirks = pdev->id_entry->driver_data;
+	dev_info = device_get_match_data(&pdev->dev);
+	if (!dev_info)
+		dev_info = (const struct fec_devinfo *)pdev->id_entry->driver_data;
+	if (dev_info)
+		fep->quirks = dev_info->quirks;
 
 	fep->netdev = ndev;
 	fep->num_rx_queues = num_rx_qs;
@@ -3329,8 +4420,7 @@ fec_probe(struct platform_device *pdev)
 	/* Select default pin state */
 	pinctrl_pm_select_default_state(&pdev->dev);
 
-	r = platform_get_resource(pdev, IORESOURCE_MEM, 0);
-	fep->hwp = devm_ioremap_resource(&pdev->dev, r);
+	fep->hwp = devm_platform_ioremap_resource(pdev, 0);
 	if (IS_ERR(fep->hwp)) {
 		ret = PTR_ERR(fep->hwp);
 		goto failed_ioremap;
@@ -3346,9 +4436,17 @@ fec_probe(struct platform_device *pdev)
 	    !of_property_read_bool(np, "fsl,err006687-workaround-present"))
 		fep->quirks |= FEC_QUIRK_ERR006687;
 
-	if (of_get_property(np, "fsl,magic-packet", NULL))
+	ret = fec_enet_ipc_handle_init(fep);
+	if (ret)
+		goto failed_ipc_init;
+
+	if (of_property_read_bool(np, "fsl,magic-packet"))
 		fep->wol_flag |= FEC_WOL_HAS_MAGIC_PACKET;
 
+	ret = fec_enet_init_stop_mode(fep, np);
+	if (ret)
+		goto failed_stop_mode;
+
 	phy_node = of_parse_phandle(np, "phy-handle", 0);
 	if (!phy_node && of_phy_is_fixed_link(np)) {
 		ret = of_phy_register_fixed_link(np);
@@ -3361,17 +4459,21 @@ fec_probe(struct platform_device *pdev)
 	}
 	fep->phy_node = phy_node;
 
-	ret = of_get_phy_mode(pdev->dev.of_node);
-	if (ret < 0) {
+	ret = of_get_phy_mode(pdev->dev.of_node, &interface);
+	if (ret) {
 		pdata = dev_get_platdata(&pdev->dev);
 		if (pdata)
 			fep->phy_interface = pdata->phy;
 		else
 			fep->phy_interface = PHY_INTERFACE_MODE_MII;
 	} else {
-		fep->phy_interface = ret;
+		fep->phy_interface = interface;
 	}
 
+	ret = fec_enet_parse_rgmii_delay(fep, np);
+	if (ret)
+		goto failed_rgmii_delay;
+
 	fep->clk_ipg = devm_clk_get(&pdev->dev, "ipg");
 	if (IS_ERR(fep->clk_ipg)) {
 		ret = PTR_ERR(fep->clk_ipg);
@@ -3387,17 +4489,29 @@ fec_probe(struct platform_device *pdev)
 	fep->itr_clk_rate = clk_get_rate(fep->clk_ahb);
 
 	/* enet_out is optional, depends on board */
-	fep->clk_enet_out = devm_clk_get(&pdev->dev, "enet_out");
-	if (IS_ERR(fep->clk_enet_out))
-		fep->clk_enet_out = NULL;
+	fep->clk_enet_out = devm_clk_get_optional(&pdev->dev, "enet_out");
+	if (IS_ERR(fep->clk_enet_out)) {
+		ret = PTR_ERR(fep->clk_enet_out);
+		goto failed_clk;
+	}
 
 	fep->ptp_clk_on = false;
 	mutex_init(&fep->ptp_clk_mutex);
 
 	/* clk_ref is optional, depends on board */
-	fep->clk_ref = devm_clk_get(&pdev->dev, "enet_clk_ref");
-	if (IS_ERR(fep->clk_ref))
-		fep->clk_ref = NULL;
+	fep->clk_ref = devm_clk_get_optional(&pdev->dev, "enet_clk_ref");
+	if (IS_ERR(fep->clk_ref)) {
+		ret = PTR_ERR(fep->clk_ref);
+		goto failed_clk;
+	}
+	fep->clk_ref_rate = clk_get_rate(fep->clk_ref);
+
+	/* clk_2x_txclk is optional, depends on board */
+	if (fep->rgmii_txc_dly || fep->rgmii_rxc_dly) {
+		fep->clk_2x_txclk = devm_clk_get(&pdev->dev, "enet_2x_txclk");
+		if (IS_ERR(fep->clk_2x_txclk))
+			fep->clk_2x_txclk = NULL;
+	}
 
 	fep->bufdesc_ex = fep->quirks & FEC_QUIRK_HAS_BUFDESC_EX;
 	fep->clk_ptp = devm_clk_get(&pdev->dev, "ptp");
@@ -3413,17 +4527,23 @@ fec_probe(struct platform_device *pdev)
 	ret = clk_prepare_enable(fep->clk_ipg);
 	if (ret)
 		goto failed_clk_ipg;
+	ret = clk_prepare_enable(fep->clk_ahb);
+	if (ret)
+		goto failed_clk_ahb;
 
-	fep->reg_phy = devm_regulator_get(&pdev->dev, "phy");
+	fep->reg_phy = devm_regulator_get_optional(&pdev->dev, "phy");
 	if (!IS_ERR(fep->reg_phy)) {
 		ret = regulator_enable(fep->reg_phy);
 		if (ret) {
 			dev_err(&pdev->dev,
 				"Failed to enable phy regulator: %d\n", ret);
-			clk_disable_unprepare(fep->clk_ipg);
 			goto failed_regulator;
 		}
 	} else {
+		if (PTR_ERR(fep->reg_phy) == -EPROBE_DEFER) {
+			ret = -EPROBE_DEFER;
+			goto failed_regulator;
+		}
 		fep->reg_phy = NULL;
 	}
 
@@ -3437,18 +4557,20 @@ fec_probe(struct platform_device *pdev)
 	if (ret)
 		goto failed_reset;
 
+	irq_cnt = fec_enet_get_irq_cnt(pdev);
 	if (fep->bufdesc_ex)
-		fec_ptp_init(pdev);
+		fec_ptp_init(pdev, irq_cnt);
 
 	ret = fec_enet_init(ndev);
 	if (ret)
 		goto failed_init;
 
-	for (i = 0; i < FEC_IRQ_NUM; i++) {
-		irq = platform_get_irq(pdev, i);
+	for (i = 0; i < irq_cnt; i++) {
+		snprintf(irq_name, sizeof(irq_name), "int%d", i);
+		irq = platform_get_irq_byname_optional(pdev, irq_name);
+		if (irq < 0)
+			irq = platform_get_irq(pdev, i);
 		if (irq < 0) {
-			if (i)
-				break;
 			ret = irq;
 			goto failed_irq;
 		}
@@ -3460,7 +4582,9 @@ fec_probe(struct platform_device *pdev)
 		fep->irq[i] = irq;
 	}
 
-	init_completion(&fep->mdio_done);
+	/* Decide which interrupt line is wakeup capable */
+	fec_enet_get_wakeup_irq(pdev);
+
 	ret = fec_enet_mii_init(pdev);
 	if (ret)
 		goto failed_mii_init;
@@ -3470,6 +4594,16 @@ fec_probe(struct platform_device *pdev)
 	fec_enet_clk_enable(ndev, false);
 	pinctrl_pm_select_sleep_state(&pdev->dev);
 
+	fep->pagepool_order = 0;
+	fep->rx_frame_size = FEC_ENET_RX_FRSIZE;
+
+	if (fep->quirks & FEC_QUIRK_JUMBO_FRAME)
+		fep->max_buf_size = MAX_JUMBO_BUF_SIZE;
+	else
+		fep->max_buf_size = PKT_MAXBUF_SIZE;
+
+	ndev->max_mtu = fep->max_buf_size - ETH_HLEN - ETH_FCS_LEN;
+
 	ret = register_netdev(ndev);
 	if (ret)
 		goto failed_register;
@@ -3480,7 +4614,6 @@ fec_probe(struct platform_device *pdev)
 	if (fep->bufdesc_ex && fep->ptp_clock)
 		netdev_info(ndev, "registered PHC device %d\n", fep->dev_id);
 
-	fep->rx_copybreak = COPYBREAK_DEFAULT;
 	INIT_WORK(&fep->tx_timeout_work, fec_enet_timeout_work);
 
 	pm_runtime_mark_last_busy(&pdev->dev);
@@ -3492,33 +4625,48 @@ failed_register:
 	fec_enet_mii_remove(fep);
 failed_mii_init:
 failed_irq:
+	fec_enet_deinit(ndev);
 failed_init:
 	fec_ptp_stop(pdev);
-	if (fep->reg_phy)
-		regulator_disable(fep->reg_phy);
 failed_reset:
-	pm_runtime_put(&pdev->dev);
+	pm_runtime_put_noidle(&pdev->dev);
 	pm_runtime_disable(&pdev->dev);
+	if (fep->reg_phy)
+		regulator_disable(fep->reg_phy);
 failed_regulator:
+	clk_disable_unprepare(fep->clk_ahb);
+failed_clk_ahb:
+	clk_disable_unprepare(fep->clk_ipg);
 failed_clk_ipg:
 	fec_enet_clk_enable(ndev, false);
 failed_clk:
+failed_rgmii_delay:
 	if (of_phy_is_fixed_link(np))
 		of_phy_deregister_fixed_link(np);
-failed_phy:
 	of_node_put(phy_node);
+failed_stop_mode:
+failed_ipc_init:
+failed_phy:
+	dev_id--;
 failed_ioremap:
 	free_netdev(ndev);
 
 	return ret;
 }
 
-static int
+static void
 fec_drv_remove(struct platform_device *pdev)
 {
 	struct net_device *ndev = platform_get_drvdata(pdev);
 	struct fec_enet_private *fep = netdev_priv(ndev);
 	struct device_node *np = pdev->dev.of_node;
+	int ret;
+
+	ret = pm_runtime_get_sync(&pdev->dev);
+	if (ret < 0)
+		dev_err(&pdev->dev,
+			"Failed to resume device in remove callback (%pe)\n",
+			ERR_PTR(ret));
 
 	cancel_work_sync(&fep->tx_timeout_work);
 	fec_ptp_stop(pdev);
@@ -3526,18 +4674,30 @@ fec_drv_remove(struct platform_device *pdev)
 	fec_enet_mii_remove(fep);
 	if (fep->reg_phy)
 		regulator_disable(fep->reg_phy);
+
 	if (of_phy_is_fixed_link(np))
 		of_phy_deregister_fixed_link(np);
 	of_node_put(fep->phy_node);
-	free_netdev(ndev);
 
-	return 0;
+	/* After pm_runtime_get_sync() failed, the clks are still off, so skip
+	 * disabling them again.
+	 */
+	if (ret >= 0) {
+		clk_disable_unprepare(fep->clk_ahb);
+		clk_disable_unprepare(fep->clk_ipg);
+	}
+	pm_runtime_put_noidle(&pdev->dev);
+	pm_runtime_disable(&pdev->dev);
+
+	fec_enet_deinit(ndev);
+	free_netdev(ndev);
 }
 
-static int __maybe_unused fec_suspend(struct device *dev)
+static int fec_suspend(struct device *dev)
 {
 	struct net_device *ndev = dev_get_drvdata(dev);
 	struct fec_enet_private *fep = netdev_priv(ndev);
+	int ret;
 
 	rtnl_lock();
 	if (netif_running(ndev)) {
@@ -3549,9 +4709,28 @@ static int __maybe_unused fec_suspend(struct device *dev)
 		netif_device_detach(ndev);
 		netif_tx_unlock_bh(ndev);
 		fec_stop(ndev);
-		fec_enet_clk_enable(ndev, false);
-		if (!(fep->wol_flag & FEC_WOL_FLAG_ENABLE))
+		if (!(fep->wol_flag & FEC_WOL_FLAG_ENABLE)) {
+			fec_irqs_disable(ndev);
 			pinctrl_pm_select_sleep_state(&fep->pdev->dev);
+		} else {
+			fec_irqs_disable_except_wakeup(ndev);
+			if (fep->wake_irq > 0) {
+				disable_irq(fep->wake_irq);
+				enable_irq_wake(fep->wake_irq);
+			}
+			fec_enet_stop_mode(fep, true);
+		}
+		/* It's safe to disable clocks since interrupts are masked */
+		fec_enet_clk_enable(ndev, false);
+
+		fep->rpm_active = !pm_runtime_status_suspended(dev);
+		if (fep->rpm_active) {
+			ret = pm_runtime_force_suspend(dev);
+			if (ret < 0) {
+				rtnl_unlock();
+				return ret;
+			}
+		}
 	}
 	rtnl_unlock();
 
@@ -3567,11 +4746,10 @@ static int __maybe_unused fec_suspend(struct device *dev)
 	return 0;
 }
 
-static int __maybe_unused fec_resume(struct device *dev)
+static int fec_resume(struct device *dev)
 {
 	struct net_device *ndev = dev_get_drvdata(dev);
 	struct fec_enet_private *fep = netdev_priv(ndev);
-	struct fec_platform_data *pdata = fep->pdev->dev.platform_data;
 	int ret;
 	int val;
 
@@ -3583,14 +4761,21 @@ static int __maybe_unused fec_resume(struct device *dev)
 
 	rtnl_lock();
 	if (netif_running(ndev)) {
+		if (fep->rpm_active)
+			pm_runtime_force_resume(dev);
+
 		ret = fec_enet_clk_enable(ndev, true);
 		if (ret) {
 			rtnl_unlock();
 			goto failed_clk;
 		}
 		if (fep->wol_flag & FEC_WOL_FLAG_ENABLE) {
-			if (pdata && pdata->sleep_mode_enable)
-				pdata->sleep_mode_enable(false);
+			fec_enet_stop_mode(fep, false);
+			if (fep->wake_irq) {
+				disable_irq_wake(fep->wake_irq);
+				enable_irq(fep->wake_irq);
+			}
+
 			val = readl(fep->hwp + FEC_ECNTRL);
 			val &= ~(FEC_ECR_MAGICEN | FEC_ECR_SLEEP);
 			writel(val, fep->hwp + FEC_ECNTRL);
@@ -3603,6 +4788,7 @@ static int __maybe_unused fec_resume(struct device *dev)
 		netif_device_attach(ndev);
 		netif_tx_unlock_bh(ndev);
 		napi_enable(&fep->napi);
+		phy_init_hw(ndev->phydev);
 		phy_start(ndev->phydev);
 	}
 	rtnl_unlock();
@@ -3615,41 +4801,55 @@ failed_clk:
 	return ret;
 }
 
-static int __maybe_unused fec_runtime_suspend(struct device *dev)
+static int fec_runtime_suspend(struct device *dev)
 {
 	struct net_device *ndev = dev_get_drvdata(dev);
 	struct fec_enet_private *fep = netdev_priv(ndev);
 
+	clk_disable_unprepare(fep->clk_ahb);
 	clk_disable_unprepare(fep->clk_ipg);
 
 	return 0;
 }
 
-static int __maybe_unused fec_runtime_resume(struct device *dev)
+static int fec_runtime_resume(struct device *dev)
 {
 	struct net_device *ndev = dev_get_drvdata(dev);
 	struct fec_enet_private *fep = netdev_priv(ndev);
+	int ret;
 
-	return clk_prepare_enable(fep->clk_ipg);
+	ret = clk_prepare_enable(fep->clk_ahb);
+	if (ret)
+		return ret;
+	ret = clk_prepare_enable(fep->clk_ipg);
+	if (ret)
+		goto failed_clk_ipg;
+
+	return 0;
+
+failed_clk_ipg:
+	clk_disable_unprepare(fep->clk_ahb);
+	return ret;
 }
 
 static const struct dev_pm_ops fec_pm_ops = {
-	SET_SYSTEM_SLEEP_PM_OPS(fec_suspend, fec_resume)
-	SET_RUNTIME_PM_OPS(fec_runtime_suspend, fec_runtime_resume, NULL)
+	SYSTEM_SLEEP_PM_OPS(fec_suspend, fec_resume)
+	RUNTIME_PM_OPS(fec_runtime_suspend, fec_runtime_resume, NULL)
 };
 
 static struct platform_driver fec_driver = {
 	.driver	= {
 		.name	= DRIVER_NAME,
-		.pm	= &fec_pm_ops,
+		.pm	= pm_ptr(&fec_pm_ops),
 		.of_match_table = fec_dt_ids,
+		.suppress_bind_attrs = true,
 	},
 	.id_table = fec_devtype,
 	.probe	= fec_probe,
-	.remove	= fec_drv_remove,
+	.remove = fec_drv_remove,
 };
 
 module_platform_driver(fec_driver);
 
-MODULE_ALIAS("platform:"DRIVER_NAME);
+MODULE_DESCRIPTION("NXP Fast Ethernet Controller (FEC) driver");
 MODULE_LICENSE("GPL");
diff --git a/drivers/net/ethernet/freescale/fec_mpc52xx.c b/drivers/net/ethernet/freescale/fec_mpc52xx.c
index aa8cf5d2a53c..3fc29afc9854 100644
--- a/drivers/net/ethernet/freescale/fec_mpc52xx.c
+++ b/drivers/net/ethernet/freescale/fec_mpc52xx.c
@@ -29,10 +29,12 @@
 #include <linux/crc32.h>
 #include <linux/hardirq.h>
 #include <linux/delay.h>
-#include <linux/of_device.h>
+#include <linux/of.h>
+#include <linux/of_address.h>
+#include <linux/of_irq.h>
 #include <linux/of_mdio.h>
 #include <linux/of_net.h>
-#include <linux/of_platform.h>
+#include <linux/platform_device.h>
 
 #include <linux/netdevice.h>
 #include <linux/etherdevice.h>
@@ -74,7 +76,7 @@ struct mpc52xx_fec_priv {
 static irqreturn_t mpc52xx_fec_interrupt(int, void *);
 static irqreturn_t mpc52xx_fec_rx_interrupt(int, void *);
 static irqreturn_t mpc52xx_fec_tx_interrupt(int, void *);
-static void mpc52xx_fec_stop(struct net_device *dev);
+static void mpc52xx_fec_stop(struct net_device *dev, bool may_sleep);
 static void mpc52xx_fec_start(struct net_device *dev);
 static void mpc52xx_fec_reset(struct net_device *dev);
 
@@ -84,7 +86,7 @@ static int debug = -1;	/* the above default */
 module_param(debug, int, 0);
 MODULE_PARM_DESC(debug, "debugging messages level");
 
-static void mpc52xx_fec_tx_timeout(struct net_device *dev)
+static void mpc52xx_fec_tx_timeout(struct net_device *dev, unsigned int txqueue)
 {
 	struct mpc52xx_fec_priv *priv = netdev_priv(dev);
 	unsigned long flags;
@@ -99,20 +101,20 @@ static void mpc52xx_fec_tx_timeout(struct net_device *dev)
 	netif_wake_queue(dev);
 }
 
-static void mpc52xx_fec_set_paddr(struct net_device *dev, u8 *mac)
+static void mpc52xx_fec_set_paddr(struct net_device *dev, const u8 *mac)
 {
 	struct mpc52xx_fec_priv *priv = netdev_priv(dev);
 	struct mpc52xx_fec __iomem *fec = priv->fec;
 
-	out_be32(&fec->paddr1, *(u32 *)(&mac[0]));
-	out_be32(&fec->paddr2, (*(u16 *)(&mac[4]) << 16) | FEC_PADDR2_TYPE);
+	out_be32(&fec->paddr1, *(const u32 *)(&mac[0]));
+	out_be32(&fec->paddr2, (*(const u16 *)(&mac[4]) << 16) | FEC_PADDR2_TYPE);
 }
 
 static int mpc52xx_fec_set_mac_address(struct net_device *dev, void *addr)
 {
 	struct sockaddr *sock = addr;
 
-	memcpy(dev->dev_addr, sock->sa_data, dev->addr_len);
+	eth_hw_addr_set(dev, sock->sa_data);
 
 	mpc52xx_fec_set_paddr(dev, sock->sa_data);
 	return 0;
@@ -283,7 +285,7 @@ static int mpc52xx_fec_close(struct net_device *dev)
 
 	netif_stop_queue(dev);
 
-	mpc52xx_fec_stop(dev);
+	mpc52xx_fec_stop(dev, true);
 
 	mpc52xx_fec_free_rx_buffers(dev, priv->rx_dmatsk);
 
@@ -305,7 +307,8 @@ static int mpc52xx_fec_close(struct net_device *dev)
  * invariant will hold if you make sure that the netif_*_queue()
  * calls are done at the proper times.
  */
-static int mpc52xx_fec_start_xmit(struct sk_buff *skb, struct net_device *dev)
+static netdev_tx_t
+mpc52xx_fec_start_xmit(struct sk_buff *skb, struct net_device *dev)
 {
 	struct mpc52xx_fec_priv *priv = netdev_priv(dev);
 	struct bcom_fec_bd *bd;
@@ -368,7 +371,7 @@ static irqreturn_t mpc52xx_fec_tx_interrupt(int irq, void *dev_id)
 		dma_unmap_single(dev->dev.parent, bd->skb_pa, skb->len,
 				 DMA_TO_DEVICE);
 
-		dev_kfree_skb_irq(skb);
+		dev_consume_skb_irq(skb);
 	}
 	spin_unlock(&priv->lock);
 
@@ -616,7 +619,7 @@ static void mpc52xx_fec_hw_init(struct net_device *dev)
 	out_be32(&fec->rfifo_alarm, 0x0000030c);
 	out_be32(&fec->tfifo_alarm, 0x00000100);
 
-	/* begin transmittion when 256 bytes are in FIFO (or EOF or FIFO full) */
+	/* begin transmission when 256 bytes are in FIFO (or EOF or FIFO full) */
 	out_be32(&fec->x_wmrk, FEC_FIFO_WMRK_256B);
 
 	/* enable crc generation */
@@ -692,7 +695,7 @@ static void mpc52xx_fec_start(struct net_device *dev)
  *
  * stop all activity on fec and empty dma buffers
  */
-static void mpc52xx_fec_stop(struct net_device *dev)
+static void mpc52xx_fec_stop(struct net_device *dev, bool may_sleep)
 {
 	struct mpc52xx_fec_priv *priv = netdev_priv(dev);
 	struct mpc52xx_fec __iomem *fec = priv->fec;
@@ -705,7 +708,7 @@ static void mpc52xx_fec_stop(struct net_device *dev)
 	bcom_disable(priv->rx_dmatsk);
 
 	/* Wait for tx queue to drain, but only if we're in process context */
-	if (!in_interrupt()) {
+	if (may_sleep) {
 		timeout = jiffies + msecs_to_jiffies(2000);
 		while (time_before(jiffies, timeout) &&
 				!bcom_queue_empty(priv->tx_dmatsk))
@@ -737,7 +740,7 @@ static void mpc52xx_fec_reset(struct net_device *dev)
 	struct mpc52xx_fec_priv *priv = netdev_priv(dev);
 	struct mpc52xx_fec __iomem *fec = priv->fec;
 
-	mpc52xx_fec_stop(dev);
+	mpc52xx_fec_stop(dev, false);
 
 	out_be32(&fec->rfifo_status, in_be32(&fec->rfifo_status));
 	out_be32(&fec->reset_cntrl, FEC_RESET_CNTRL_RESET_FIFO);
@@ -784,16 +787,6 @@ static const struct ethtool_ops mpc52xx_fec_ethtool_ops = {
 };
 
 
-static int mpc52xx_fec_ioctl(struct net_device *dev, struct ifreq *rq, int cmd)
-{
-	struct phy_device *phydev = dev->phydev;
-
-	if (!phydev)
-		return -ENOTSUPP;
-
-	return phy_mii_ioctl(phydev, rq, cmd);
-}
-
 static const struct net_device_ops mpc52xx_fec_netdev_ops = {
 	.ndo_open = mpc52xx_fec_open,
 	.ndo_stop = mpc52xx_fec_close,
@@ -801,7 +794,7 @@ static const struct net_device_ops mpc52xx_fec_netdev_ops = {
 	.ndo_set_rx_mode = mpc52xx_fec_set_multicast_list,
 	.ndo_set_mac_address = mpc52xx_fec_set_mac_address,
 	.ndo_validate_addr = eth_validate_addr,
-	.ndo_do_ioctl = mpc52xx_fec_ioctl,
+	.ndo_eth_ioctl = phy_do_ioctl,
 	.ndo_tx_timeout = mpc52xx_fec_tx_timeout,
 	.ndo_get_stats = mpc52xx_fec_get_stats,
 #ifdef CONFIG_NET_POLL_CONTROLLER
@@ -822,7 +815,6 @@ static int mpc52xx_fec_probe(struct platform_device *op)
 	const u32 *prop;
 	int prop_size;
 	struct device_node *np = op->dev.of_node;
-	const char *mac_addr;
 
 	phys_addr_t rx_fifo;
 	phys_addr_t tx_fifo;
@@ -900,18 +892,18 @@ static int mpc52xx_fec_probe(struct platform_device *op)
 	 *
 	 * First try to read MAC address from DT
 	 */
-	mac_addr = of_get_mac_address(np);
-	if (mac_addr) {
-		memcpy(ndev->dev_addr, mac_addr, ETH_ALEN);
-	} else {
+	rv = of_get_ethdev_address(np, ndev);
+	if (rv) {
 		struct mpc52xx_fec __iomem *fec = priv->fec;
+		u8 addr[ETH_ALEN] __aligned(4);
 
 		/*
 		 * If the MAC addresse is not provided via DT then read
 		 * it back from the controller regs
 		 */
-		*(u32 *)(&ndev->dev_addr[0]) = in_be32(&fec->paddr1);
-		*(u16 *)(&ndev->dev_addr[4]) = in_be32(&fec->paddr2) >> 16;
+		*(u32 *)(&addr[0]) = in_be32(&fec->paddr1);
+		*(u16 *)(&addr[4]) = in_be32(&fec->paddr2) >> 16;
+		eth_hw_addr_set(ndev, addr);
 	}
 
 	/*
@@ -932,7 +924,7 @@ static int mpc52xx_fec_probe(struct platform_device *op)
 	/* Start with safe defaults for link connection */
 	priv->speed = 100;
 	priv->duplex = DUPLEX_HALF;
-	priv->mdio_speed = ((mpc5xxx_get_bus_frequency(np) >> 20) / 5) << 1;
+	priv->mdio_speed = ((mpc5xxx_get_bus_frequency(&op->dev) >> 20) / 5) << 1;
 
 	/* The current speed preconfigures the speed of the MII link */
 	prop = of_get_property(np, "current-speed", &prop_size);
@@ -945,7 +937,7 @@ static int mpc52xx_fec_probe(struct platform_device *op)
 	priv->phy_node = of_parse_phandle(np, "phy-handle", 0);
 
 	/* the 7-wire property means don't use MII mode */
-	if (of_find_property(np, "fsl,7-wire-mode", NULL)) {
+	if (of_property_read_bool(np, "fsl,7-wire-mode")) {
 		priv->seven_wire_mode = 1;
 		dev_info(&ndev->dev, "using 7-wire PHY mode\n");
 	}
@@ -960,8 +952,8 @@ static int mpc52xx_fec_probe(struct platform_device *op)
 
 	/* We're done ! */
 	platform_set_drvdata(op, ndev);
-	netdev_info(ndev, "%s MAC %pM\n",
-		    op->dev.of_node->full_name, ndev->dev_addr);
+	netdev_info(ndev, "%pOF MAC %pM\n",
+		    op->dev.of_node, ndev->dev_addr);
 
 	return 0;
 
@@ -982,7 +974,7 @@ err_netdev:
 	return rv;
 }
 
-static int
+static void
 mpc52xx_fec_remove(struct platform_device *op)
 {
 	struct net_device *ndev;
@@ -1006,8 +998,6 @@ mpc52xx_fec_remove(struct platform_device *op)
 	release_mem_region(ndev->base_addr, sizeof(struct mpc52xx_fec));
 
 	free_netdev(ndev);
-
-	return 0;
 }
 
 #ifdef CONFIG_PM
diff --git a/drivers/net/ethernet/freescale/fec_mpc52xx_phy.c b/drivers/net/ethernet/freescale/fec_mpc52xx_phy.c
index b5497e308302..3d073f0fae63 100644
--- a/drivers/net/ethernet/freescale/fec_mpc52xx_phy.c
+++ b/drivers/net/ethernet/freescale/fec_mpc52xx_phy.c
@@ -13,9 +13,11 @@
 #include <linux/module.h>
 #include <linux/netdevice.h>
 #include <linux/phy.h>
-#include <linux/of_platform.h>
 #include <linux/slab.h>
+#include <linux/of.h>
+#include <linux/of_address.h>
 #include <linux/of_mdio.h>
+#include <linux/platform_device.h>
 #include <asm/io.h>
 #include <asm/mpc52xx.h>
 #include "fec_mpc52xx.h"
@@ -92,15 +94,14 @@ static int mpc52xx_fec_mdio_probe(struct platform_device *of)
 		goto out_free;
 	}
 
-	snprintf(bus->id, MII_BUS_ID_SIZE, "%x", res.start);
+	snprintf(bus->id, MII_BUS_ID_SIZE, "%pa", &res.start);
 	bus->priv = priv;
 
 	bus->parent = dev;
 	dev_set_drvdata(dev, bus);
 
 	/* set MII speed */
-	out_be32(&priv->regs->mii_speed,
-		((mpc5xxx_get_bus_frequency(of->dev.of_node) >> 20) / 5) << 1);
+	out_be32(&priv->regs->mii_speed, ((mpc5xxx_get_bus_frequency(dev) >> 20) / 5) << 1);
 
 	err = of_mdiobus_register(bus, np);
 	if (err)
@@ -117,7 +118,7 @@ static int mpc52xx_fec_mdio_probe(struct platform_device *of)
 	return err;
 }
 
-static int mpc52xx_fec_mdio_remove(struct platform_device *of)
+static void mpc52xx_fec_mdio_remove(struct platform_device *of)
 {
 	struct mii_bus *bus = platform_get_drvdata(of);
 	struct mpc52xx_fec_mdio_priv *priv = bus->priv;
@@ -126,8 +127,6 @@ static int mpc52xx_fec_mdio_remove(struct platform_device *of)
 	iounmap(priv->regs);
 	kfree(priv);
 	mdiobus_free(bus);
-
-	return 0;
 }
 
 static const struct of_device_id mpc52xx_fec_mdio_match[] = {
diff --git a/drivers/net/ethernet/freescale/fec_ptp.c b/drivers/net/ethernet/freescale/fec_ptp.c
index 6ebad3fac81d..fa88b47d526c 100644
--- a/drivers/net/ethernet/freescale/fec_ptp.c
+++ b/drivers/net/ethernet/freescale/fec_ptp.c
@@ -1,50 +1,36 @@
+// SPDX-License-Identifier: GPL-2.0
 /*
  * Fast Ethernet Controller (ENET) PTP driver for MX6x.
  *
  * Copyright (C) 2012 Freescale Semiconductor, Inc.
- *
- * This program is free software; you can redistribute it and/or modify it
- * under the terms and conditions of the GNU General Public License,
- * version 2, as published by the Free Software Foundation.
- *
- * This program is distributed in the hope it will be useful, but WITHOUT
- * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
- * FITNESS FOR A PARTICULAR PURPOSE.  See the GNU General Public License for
- * more details.
- *
- * You should have received a copy of the GNU General Public License along with
- * this program; if not, write to the Free Software Foundation, Inc.,
- * 51 Franklin St - Fifth Floor, Boston, MA 02110-1301 USA.
  */
 
 #define pr_fmt(fmt) KBUILD_MODNAME ": " fmt
 
-#include <linux/module.h>
-#include <linux/kernel.h>
-#include <linux/string.h>
-#include <linux/ptrace.h>
-#include <linux/errno.h>
-#include <linux/ioport.h>
-#include <linux/slab.h>
-#include <linux/interrupt.h>
-#include <linux/pci.h>
+#include <linux/bitops.h>
+#include <linux/clk.h>
 #include <linux/delay.h>
-#include <linux/netdevice.h>
+#include <linux/errno.h>
 #include <linux/etherdevice.h>
-#include <linux/skbuff.h>
-#include <linux/spinlock.h>
-#include <linux/workqueue.h>
-#include <linux/bitops.h>
+#include <linux/fec.h>
+#include <linux/interrupt.h>
 #include <linux/io.h>
+#include <linux/ioport.h>
 #include <linux/irq.h>
-#include <linux/clk.h>
-#include <linux/platform_device.h>
-#include <linux/phy.h>
-#include <linux/fec.h>
+#include <linux/kernel.h>
+#include <linux/module.h>
+#include <linux/netdevice.h>
 #include <linux/of.h>
-#include <linux/of_device.h>
-#include <linux/of_gpio.h>
 #include <linux/of_net.h>
+#include <linux/pci.h>
+#include <linux/phy.h>
+#include <linux/platform_device.h>
+#include <linux/ptrace.h>
+#include <linux/skbuff.h>
+#include <linux/slab.h>
+#include <linux/spinlock.h>
+#include <linux/string.h>
+#include <linux/workqueue.h>
 
 #include "fec.h"
 
@@ -97,38 +83,55 @@
 #define FEC_CC_MULT	(1 << 31)
 #define FEC_COUNTER_PERIOD	(1 << 31)
 #define PPS_OUPUT_RELOAD_PERIOD	NSEC_PER_SEC
-#define FEC_CHANNLE_0		0
-#define DEFAULT_PPS_CHANNEL	FEC_CHANNLE_0
+#define DEFAULT_PPS_CHANNEL	0
+
+#define FEC_PTP_MAX_NSEC_PERIOD		4000000000ULL
+#define FEC_PTP_MAX_NSEC_COUNTER	0x80000000ULL
+
+/**
+ * fec_ptp_read - read raw cycle counter (to be used by time counter)
+ * @cc: the cyclecounter structure
+ *
+ * this function reads the cyclecounter registers and is called by the
+ * cyclecounter structure used to construct a ns counter from the
+ * arbitrary fixed point registers
+ */
+static u64 fec_ptp_read(struct cyclecounter *cc)
+{
+	struct fec_enet_private *fep =
+		container_of(cc, struct fec_enet_private, cc);
+	u32 tempval;
+
+	tempval = readl(fep->hwp + FEC_ATIME_CTRL);
+	tempval |= FEC_T_CTRL_CAPTURE;
+	writel(tempval, fep->hwp + FEC_ATIME_CTRL);
+
+	if (fep->quirks & FEC_QUIRK_BUG_CAPTURE)
+		udelay(1);
+
+	return readl(fep->hwp + FEC_ATIME);
+}
 
 /**
  * fec_ptp_enable_pps
  * @fep: the fec_enet_private structure handle
  * @enable: enable the channel pps output
  *
- * This function enble the PPS ouput on the timer channel.
+ * This function enables the PPS output on the timer channel.
  */
 static int fec_ptp_enable_pps(struct fec_enet_private *fep, uint enable)
 {
 	unsigned long flags;
 	u32 val, tempval;
-	int inc;
 	struct timespec64 ts;
 	u64 ns;
-	val = 0;
 
-	if (!(fep->hwts_tx_en || fep->hwts_rx_en)) {
-		dev_err(&fep->pdev->dev, "No ptp stack is running\n");
-		return -EINVAL;
-	}
+	spin_lock_irqsave(&fep->tmreg_lock, flags);
 
-	if (fep->pps_enable == enable)
+	if (fep->pps_enable == enable) {
+		spin_unlock_irqrestore(&fep->tmreg_lock, flags);
 		return 0;
-
-	fep->pps_channel = DEFAULT_PPS_CHANNEL;
-	fep->reload_period = PPS_OUPUT_RELOAD_PERIOD;
-	inc = fep->ptp_inc;
-
-	spin_lock_irqsave(&fep->tmreg_lock, flags);
+	}
 
 	if (enable) {
 		/* clear capture or output compare interrupt status if have.
@@ -155,11 +158,7 @@ static int fec_ptp_enable_pps(struct fec_enet_private *fep, uint enable)
 		 * NSEC_PER_SEC - ts.tv_nsec. Add the remaining nanoseconds
 		 * to current timer would be next second.
 		 */
-		tempval = readl(fep->hwp + FEC_ATIME_CTRL);
-		tempval |= FEC_T_CTRL_CAPTURE;
-		writel(tempval, fep->hwp + FEC_ATIME_CTRL);
-
-		tempval = readl(fep->hwp + FEC_ATIME);
+		tempval = fec_ptp_read(&fep->cc);
 		/* Convert the ptp local counter to 1588 timestamp */
 		ns = timecounter_cyc2time(&fep->tc, tempval);
 		ts = ns_to_timespec64(ns);
@@ -173,7 +172,7 @@ static int fec_ptp_enable_pps(struct fec_enet_private *fep, uint enable)
 		 * very close to the second point, which means NSEC_PER_SEC
 		 * - ts.tv_nsec is close to be zero(For example 20ns); Since the timer
 		 * is still running when we calculate the first compare event, it is
-		 * possible that the remaining nanoseonds run out before the compare
+		 * possible that the remaining nanoseconds run out before the compare
 		 * counter is calculated and written into TCCR register. To avoid
 		 * this possibility, we will set the compare event to be the next
 		 * of next second. The current setting is 31-bit timer and wrap
@@ -222,30 +221,70 @@ static int fec_ptp_enable_pps(struct fec_enet_private *fep, uint enable)
 	return 0;
 }
 
-/**
- * fec_ptp_read - read raw cycle counter (to be used by time counter)
- * @cc: the cyclecounter structure
- *
- * this function reads the cyclecounter registers and is called by the
- * cyclecounter structure used to construct a ns counter from the
- * arbitrary fixed point registers
- */
-static u64 fec_ptp_read(const struct cyclecounter *cc)
+static int fec_ptp_pps_perout(struct fec_enet_private *fep)
 {
-	struct fec_enet_private *fep =
-		container_of(cc, struct fec_enet_private, cc);
-	const struct platform_device_id *id_entry =
-		platform_get_device_id(fep->pdev);
-	u32 tempval;
+	u32 compare_val, ptp_hc, temp_val;
+	u64 curr_time;
+	unsigned long flags;
 
-	tempval = readl(fep->hwp + FEC_ATIME_CTRL);
-	tempval |= FEC_T_CTRL_CAPTURE;
-	writel(tempval, fep->hwp + FEC_ATIME_CTRL);
+	spin_lock_irqsave(&fep->tmreg_lock, flags);
 
-	if (id_entry->driver_data & FEC_QUIRK_BUG_CAPTURE)
-		udelay(1);
+	/* Update time counter */
+	timecounter_read(&fep->tc);
 
-	return readl(fep->hwp + FEC_ATIME);
+	/* Get the current ptp hardware time counter */
+	ptp_hc = fec_ptp_read(&fep->cc);
+
+	/* Convert the ptp local counter to 1588 timestamp */
+	curr_time = timecounter_cyc2time(&fep->tc, ptp_hc);
+
+	/* If the pps start time less than current time add 100ms, just return.
+	 * Because the software might not able to set the comparison time into
+	 * the FEC_TCCR register in time and missed the start time.
+	 */
+	if (fep->perout_stime < curr_time + 100 * NSEC_PER_MSEC) {
+		dev_err(&fep->pdev->dev, "Current time is too close to the start time!\n");
+		spin_unlock_irqrestore(&fep->tmreg_lock, flags);
+		return -1;
+	}
+
+	compare_val = fep->perout_stime - curr_time + ptp_hc;
+	compare_val &= fep->cc.mask;
+
+	writel(compare_val, fep->hwp + FEC_TCCR(fep->pps_channel));
+	fep->next_counter = (compare_val + fep->reload_period) & fep->cc.mask;
+
+	/* Enable compare event when overflow */
+	temp_val = readl(fep->hwp + FEC_ATIME_CTRL);
+	temp_val |= FEC_T_CTRL_PINPER;
+	writel(temp_val, fep->hwp + FEC_ATIME_CTRL);
+
+	/* Compare channel setting. */
+	temp_val = readl(fep->hwp + FEC_TCSR(fep->pps_channel));
+	temp_val |= (1 << FEC_T_TF_OFFSET | 1 << FEC_T_TIE_OFFSET);
+	temp_val &= ~(1 << FEC_T_TDRE_OFFSET);
+	temp_val &= ~(FEC_T_TMODE_MASK);
+	temp_val |= (FEC_TMODE_TOGGLE << FEC_T_TMODE_OFFSET);
+	writel(temp_val, fep->hwp + FEC_TCSR(fep->pps_channel));
+
+	/* Write the second compare event timestamp and calculate
+	 * the third timestamp. Refer the TCCR register detail in the spec.
+	 */
+	writel(fep->next_counter, fep->hwp + FEC_TCCR(fep->pps_channel));
+	fep->next_counter = (fep->next_counter + fep->reload_period) & fep->cc.mask;
+	spin_unlock_irqrestore(&fep->tmreg_lock, flags);
+
+	return 0;
+}
+
+static enum hrtimer_restart fec_ptp_pps_perout_handler(struct hrtimer *timer)
+{
+	struct fec_enet_private *fep = container_of(timer,
+					struct fec_enet_private, perout_timer);
+
+	fec_ptp_pps_perout(fep);
+
+	return HRTIMER_NORESTART;
 }
 
 /**
@@ -283,24 +322,27 @@ void fec_ptp_start_cyclecounter(struct net_device *ndev)
 	fep->cc.mult = FEC_CC_MULT;
 
 	/* reset the ns time counter */
-	timecounter_init(&fep->tc, &fep->cc, ktime_to_ns(ktime_get_real()));
+	timecounter_init(&fep->tc, &fep->cc, 0);
 
 	spin_unlock_irqrestore(&fep->tmreg_lock, flags);
 }
 
 /**
- * fec_ptp_adjfreq - adjust ptp cycle frequency
+ * fec_ptp_adjfine - adjust ptp cycle frequency
  * @ptp: the ptp clock structure
- * @ppb: parts per billion adjustment from base
+ * @scaled_ppm: scaled parts per million adjustment from base
  *
  * Adjust the frequency of the ptp cycle counter by the
- * indicated ppb from the base frequency.
+ * indicated amount from the base frequency.
+ *
+ * Scaled parts per million is ppm with a 16-bit binary fractional field.
  *
  * Because ENET hardware frequency adjust is complex,
  * using software method to do that.
  */
-static int fec_ptp_adjfreq(struct ptp_clock_info *ptp, s32 ppb)
+static int fec_ptp_adjfine(struct ptp_clock_info *ptp, long scaled_ppm)
 {
+	s32 ppb = scaled_ppm_to_ppb(scaled_ppm);
 	unsigned long flags;
 	int neg_adj = 0;
 	u32 i, tmp;
@@ -391,14 +433,21 @@ static int fec_ptp_adjtime(struct ptp_clock_info *ptp, s64 delta)
  */
 static int fec_ptp_gettime(struct ptp_clock_info *ptp, struct timespec64 *ts)
 {
-	struct fec_enet_private *adapter =
+	struct fec_enet_private *fep =
 	    container_of(ptp, struct fec_enet_private, ptp_caps);
 	u64 ns;
 	unsigned long flags;
 
-	spin_lock_irqsave(&adapter->tmreg_lock, flags);
-	ns = timecounter_read(&adapter->tc);
-	spin_unlock_irqrestore(&adapter->tmreg_lock, flags);
+	mutex_lock(&fep->ptp_clk_mutex);
+	/* Check the ptp clock */
+	if (!fep->ptp_clk_on) {
+		mutex_unlock(&fep->ptp_clk_mutex);
+		return -EINVAL;
+	}
+	spin_lock_irqsave(&fep->tmreg_lock, flags);
+	ns = timecounter_read(&fep->tc);
+	spin_unlock_irqrestore(&fep->tmreg_lock, flags);
+	mutex_unlock(&fep->ptp_clk_mutex);
 
 	*ts = ns_to_timespec64(ns);
 
@@ -444,6 +493,17 @@ static int fec_ptp_settime(struct ptp_clock_info *ptp,
 	return 0;
 }
 
+static int fec_ptp_pps_disable(struct fec_enet_private *fep, uint channel)
+{
+	unsigned long flags;
+
+	spin_lock_irqsave(&fep->tmreg_lock, flags);
+	writel(0, fep->hwp + FEC_TCSR(channel));
+	spin_unlock_irqrestore(&fep->tmreg_lock, flags);
+
+	return 0;
+}
+
 /**
  * fec_ptp_enable
  * @ptp: the ptp clock structure
@@ -456,36 +516,93 @@ static int fec_ptp_enable(struct ptp_clock_info *ptp,
 {
 	struct fec_enet_private *fep =
 	    container_of(ptp, struct fec_enet_private, ptp_caps);
+	ktime_t timeout;
+	struct timespec64 start_time, period;
+	u64 curr_time, delta, period_ns;
+	unsigned long flags;
 	int ret = 0;
 
 	if (rq->type == PTP_CLK_REQ_PPS) {
+		fep->reload_period = PPS_OUPUT_RELOAD_PERIOD;
+
 		ret = fec_ptp_enable_pps(fep, on);
 
 		return ret;
+	} else if (rq->type == PTP_CLK_REQ_PEROUT) {
+		/* Reject requests with unsupported flags */
+		if (rq->perout.flags)
+			return -EOPNOTSUPP;
+
+		if (rq->perout.index != fep->pps_channel)
+			return -EOPNOTSUPP;
+
+		period.tv_sec = rq->perout.period.sec;
+		period.tv_nsec = rq->perout.period.nsec;
+		period_ns = timespec64_to_ns(&period);
+
+		/* FEC PTP timer only has 31 bits, so if the period exceed
+		 * 4s is not supported.
+		 */
+		if (period_ns > FEC_PTP_MAX_NSEC_PERIOD) {
+			dev_err(&fep->pdev->dev, "The period must equal to or less than 4s!\n");
+			return -EOPNOTSUPP;
+		}
+
+		fep->reload_period = div_u64(period_ns, 2);
+		if (on && fep->reload_period) {
+			/* Convert 1588 timestamp to ns*/
+			start_time.tv_sec = rq->perout.start.sec;
+			start_time.tv_nsec = rq->perout.start.nsec;
+			fep->perout_stime = timespec64_to_ns(&start_time);
+
+			mutex_lock(&fep->ptp_clk_mutex);
+			if (!fep->ptp_clk_on) {
+				dev_err(&fep->pdev->dev, "Error: PTP clock is closed!\n");
+				mutex_unlock(&fep->ptp_clk_mutex);
+				return -EOPNOTSUPP;
+			}
+			spin_lock_irqsave(&fep->tmreg_lock, flags);
+			/* Read current timestamp */
+			curr_time = timecounter_read(&fep->tc);
+			spin_unlock_irqrestore(&fep->tmreg_lock, flags);
+			mutex_unlock(&fep->ptp_clk_mutex);
+
+			/* Calculate time difference */
+			delta = fep->perout_stime - curr_time;
+
+			if (fep->perout_stime <= curr_time) {
+				dev_err(&fep->pdev->dev, "Start time must larger than current time!\n");
+				return -EINVAL;
+			}
+
+			/* Because the timer counter of FEC only has 31-bits, correspondingly,
+			 * the time comparison register FEC_TCCR also only low 31 bits can be
+			 * set. If the start time of pps signal exceeds current time more than
+			 * 0x80000000 ns, a software timer is used and the timer expires about
+			 * 1 second before the start time to be able to set FEC_TCCR.
+			 */
+			if (delta > FEC_PTP_MAX_NSEC_COUNTER) {
+				timeout = ns_to_ktime(delta - NSEC_PER_SEC);
+				hrtimer_start(&fep->perout_timer, timeout, HRTIMER_MODE_REL);
+			} else {
+				return fec_ptp_pps_perout(fep);
+			}
+		} else {
+			fec_ptp_pps_disable(fep, fep->pps_channel);
+		}
+
+		return 0;
+	} else {
+		return -EOPNOTSUPP;
 	}
-	return -EOPNOTSUPP;
 }
 
-/**
- * fec_ptp_hwtstamp_ioctl - control hardware time stamping
- * @ndev: pointer to net_device
- * @ifreq: ioctl data
- * @cmd: particular ioctl requested
- */
-int fec_ptp_set(struct net_device *ndev, struct ifreq *ifr)
+int fec_ptp_set(struct net_device *ndev, struct kernel_hwtstamp_config *config,
+		struct netlink_ext_ack *extack)
 {
 	struct fec_enet_private *fep = netdev_priv(ndev);
 
-	struct hwtstamp_config config;
-
-	if (copy_from_user(&config, ifr->ifr_data, sizeof(config)))
-		return -EFAULT;
-
-	/* reserved for future extensions */
-	if (config.flags)
-		return -EINVAL;
-
-	switch (config.tx_type) {
+	switch (config->tx_type) {
 	case HWTSTAMP_TX_OFF:
 		fep->hwts_tx_en = 0;
 		break;
@@ -496,38 +613,31 @@ int fec_ptp_set(struct net_device *ndev, struct ifreq *ifr)
 		return -ERANGE;
 	}
 
-	switch (config.rx_filter) {
+	switch (config->rx_filter) {
 	case HWTSTAMP_FILTER_NONE:
-		if (fep->hwts_rx_en)
-			fep->hwts_rx_en = 0;
-		config.rx_filter = HWTSTAMP_FILTER_NONE;
+		fep->hwts_rx_en = 0;
 		break;
 
 	default:
 		fep->hwts_rx_en = 1;
-		config.rx_filter = HWTSTAMP_FILTER_ALL;
+		config->rx_filter = HWTSTAMP_FILTER_ALL;
 		break;
 	}
 
-	return copy_to_user(ifr->ifr_data, &config, sizeof(config)) ?
-	    -EFAULT : 0;
+	return 0;
 }
 
-int fec_ptp_get(struct net_device *ndev, struct ifreq *ifr)
+void fec_ptp_get(struct net_device *ndev, struct kernel_hwtstamp_config *config)
 {
 	struct fec_enet_private *fep = netdev_priv(ndev);
-	struct hwtstamp_config config;
-
-	config.flags = 0;
-	config.tx_type = fep->hwts_tx_en ? HWTSTAMP_TX_ON : HWTSTAMP_TX_OFF;
-	config.rx_filter = (fep->hwts_rx_en ?
-			    HWTSTAMP_FILTER_ALL : HWTSTAMP_FILTER_NONE);
 
-	return copy_to_user(ifr->ifr_data, &config, sizeof(config)) ?
-		-EFAULT : 0;
+	config->flags = 0;
+	config->tx_type = fep->hwts_tx_en ? HWTSTAMP_TX_ON : HWTSTAMP_TX_OFF;
+	config->rx_filter = (fep->hwts_rx_en ?
+			     HWTSTAMP_FILTER_ALL : HWTSTAMP_FILTER_NONE);
 }
 
-/**
+/*
  * fec_time_keep - call timecounter_read every second to avoid timer overrun
  *                 because ENET just support 32bit counter, will timeout in 4s
  */
@@ -535,13 +645,12 @@ static void fec_time_keep(struct work_struct *work)
 {
 	struct delayed_work *dwork = to_delayed_work(work);
 	struct fec_enet_private *fep = container_of(dwork, struct fec_enet_private, time_keep);
-	u64 ns;
 	unsigned long flags;
 
 	mutex_lock(&fep->ptp_clk_mutex);
 	if (fep->ptp_clk_on) {
 		spin_lock_irqsave(&fep->tmreg_lock, flags);
-		ns = timecounter_read(&fep->tc);
+		timecounter_read(&fep->tc);
 		spin_unlock_irqrestore(&fep->tmreg_lock, flags);
 	}
 	mutex_unlock(&fep->ptp_clk_mutex);
@@ -549,36 +658,78 @@ static void fec_time_keep(struct work_struct *work)
 	schedule_delayed_work(&fep->time_keep, HZ);
 }
 
+/* This function checks the pps event and reloads the timer compare counter. */
+static irqreturn_t fec_pps_interrupt(int irq, void *dev_id)
+{
+	struct net_device *ndev = dev_id;
+	struct fec_enet_private *fep = netdev_priv(ndev);
+	u32 val;
+	u8 channel = fep->pps_channel;
+	struct ptp_clock_event event;
+
+	val = readl(fep->hwp + FEC_TCSR(channel));
+	if (val & FEC_T_TF_MASK) {
+		/* Write the next next compare(not the next according the spec)
+		 * value to the register
+		 */
+		writel(fep->next_counter, fep->hwp + FEC_TCCR(channel));
+		do {
+			writel(val, fep->hwp + FEC_TCSR(channel));
+		} while (readl(fep->hwp + FEC_TCSR(channel)) & FEC_T_TF_MASK);
+
+		/* Update the counter; */
+		fep->next_counter = (fep->next_counter + fep->reload_period) &
+				fep->cc.mask;
+
+		event.type = PTP_CLOCK_PPS;
+		ptp_clock_event(fep->ptp_clock, &event);
+		return IRQ_HANDLED;
+	}
+
+	return IRQ_NONE;
+}
+
 /**
  * fec_ptp_init
- * @ndev: The FEC network adapter
+ * @pdev: The FEC network adapter
+ * @irq_idx: the interrupt index
  *
  * This function performs the required steps for enabling ptp
  * support. If ptp support has already been loaded it simply calls the
  * cyclecounter init routine and exits.
  */
 
-void fec_ptp_init(struct platform_device *pdev)
+void fec_ptp_init(struct platform_device *pdev, int irq_idx)
 {
 	struct net_device *ndev = platform_get_drvdata(pdev);
 	struct fec_enet_private *fep = netdev_priv(ndev);
+	struct device_node *np = fep->pdev->dev.of_node;
+	int irq;
+	int ret;
 
 	fep->ptp_caps.owner = THIS_MODULE;
-	snprintf(fep->ptp_caps.name, 16, "fec ptp");
+	strscpy(fep->ptp_caps.name, "fec ptp", sizeof(fep->ptp_caps.name));
+
+	fep->pps_channel = DEFAULT_PPS_CHANNEL;
+	of_property_read_u32(np, "fsl,pps-channel", &fep->pps_channel);
 
 	fep->ptp_caps.max_adj = 250000000;
 	fep->ptp_caps.n_alarm = 0;
 	fep->ptp_caps.n_ext_ts = 0;
-	fep->ptp_caps.n_per_out = 0;
+	fep->ptp_caps.n_per_out = 1;
 	fep->ptp_caps.n_pins = 0;
 	fep->ptp_caps.pps = 1;
-	fep->ptp_caps.adjfreq = fec_ptp_adjfreq;
+	fep->ptp_caps.adjfine = fec_ptp_adjfine;
 	fep->ptp_caps.adjtime = fec_ptp_adjtime;
 	fep->ptp_caps.gettime64 = fec_ptp_gettime;
 	fep->ptp_caps.settime64 = fec_ptp_settime;
 	fep->ptp_caps.enable = fec_ptp_enable;
 
 	fep->cycle_speed = clk_get_rate(fep->clk_ptp);
+	if (!fep->cycle_speed) {
+		fep->cycle_speed = NSEC_PER_SEC;
+		dev_err(&fep->pdev->dev, "clk_ptp clock rate is zero\n");
+	}
 	fep->ptp_inc = NSEC_PER_SEC / fep->cycle_speed;
 
 	spin_lock_init(&fep->tmreg_lock);
@@ -587,54 +738,92 @@ void fec_ptp_init(struct platform_device *pdev)
 
 	INIT_DELAYED_WORK(&fep->time_keep, fec_time_keep);
 
+	hrtimer_setup(&fep->perout_timer, fec_ptp_pps_perout_handler, CLOCK_REALTIME,
+		      HRTIMER_MODE_REL);
+
+	irq = platform_get_irq_byname_optional(pdev, "pps");
+	if (irq < 0)
+		irq = platform_get_irq_optional(pdev, irq_idx);
+	/* Failure to get an irq is not fatal,
+	 * only the PTP_CLOCK_PPS clock events should stop
+	 */
+	if (irq >= 0) {
+		ret = devm_request_irq(&pdev->dev, irq, fec_pps_interrupt,
+				       0, pdev->name, ndev);
+		if (ret < 0)
+			dev_warn(&pdev->dev, "request for pps irq failed(%d)\n",
+				 ret);
+	}
+
 	fep->ptp_clock = ptp_clock_register(&fep->ptp_caps, &pdev->dev);
 	if (IS_ERR(fep->ptp_clock)) {
 		fep->ptp_clock = NULL;
-		pr_err("ptp_clock_register failed\n");
+		dev_err(&pdev->dev, "ptp_clock_register failed\n");
 	}
 
 	schedule_delayed_work(&fep->time_keep, HZ);
 }
 
-void fec_ptp_stop(struct platform_device *pdev)
+void fec_ptp_save_state(struct fec_enet_private *fep)
 {
-	struct net_device *ndev = platform_get_drvdata(pdev);
-	struct fec_enet_private *fep = netdev_priv(ndev);
+	unsigned long flags;
+	u32 atime_inc_corr;
 
-	cancel_delayed_work_sync(&fep->time_keep);
-	if (fep->ptp_clock)
-		ptp_clock_unregister(fep->ptp_clock);
+	spin_lock_irqsave(&fep->tmreg_lock, flags);
+
+	fep->ptp_saved_state.pps_enable = fep->pps_enable;
+
+	fep->ptp_saved_state.ns_phc = timecounter_read(&fep->tc);
+	fep->ptp_saved_state.ns_sys = ktime_get_ns();
+
+	fep->ptp_saved_state.at_corr = readl(fep->hwp + FEC_ATIME_CORR);
+	atime_inc_corr = readl(fep->hwp + FEC_ATIME_INC) & FEC_T_INC_CORR_MASK;
+	fep->ptp_saved_state.at_inc_corr = (u8)(atime_inc_corr >> FEC_T_INC_CORR_OFFSET);
+
+	spin_unlock_irqrestore(&fep->tmreg_lock, flags);
 }
 
-/**
- * fec_ptp_check_pps_event
- * @fep: the fec_enet_private structure handle
- *
- * This function check the pps event and reload the timer compare counter.
- */
-uint fec_ptp_check_pps_event(struct fec_enet_private *fep)
+/* Restore PTP functionality after a reset */
+void fec_ptp_restore_state(struct fec_enet_private *fep)
 {
-	u32 val;
-	u8 channel = fep->pps_channel;
-	struct ptp_clock_event event;
+	u32 atime_inc = readl(fep->hwp + FEC_ATIME_INC) & FEC_T_INC_MASK;
+	unsigned long flags;
+	u32 counter;
+	u64 ns;
 
-	val = readl(fep->hwp + FEC_TCSR(channel));
-	if (val & FEC_T_TF_MASK) {
-		/* Write the next next compare(not the next according the spec)
-		 * value to the register
-		 */
-		writel(fep->next_counter, fep->hwp + FEC_TCCR(channel));
-		do {
-			writel(val, fep->hwp + FEC_TCSR(channel));
-		} while (readl(fep->hwp + FEC_TCSR(channel)) & FEC_T_TF_MASK);
+	spin_lock_irqsave(&fep->tmreg_lock, flags);
 
-		/* Update the counter; */
-		fep->next_counter = (fep->next_counter + fep->reload_period) & fep->cc.mask;
+	/* Reset turned it off, so adjust our status flag */
+	fep->pps_enable = 0;
 
-		event.type = PTP_CLOCK_PPS;
-		ptp_clock_event(fep->ptp_clock, &event);
-		return 1;
+	writel(fep->ptp_saved_state.at_corr, fep->hwp + FEC_ATIME_CORR);
+	atime_inc |= ((u32)fep->ptp_saved_state.at_inc_corr) << FEC_T_INC_CORR_OFFSET;
+	writel(atime_inc, fep->hwp + FEC_ATIME_INC);
+
+	ns = ktime_get_ns() - fep->ptp_saved_state.ns_sys + fep->ptp_saved_state.ns_phc;
+	counter = ns & fep->cc.mask;
+	writel(counter, fep->hwp + FEC_ATIME);
+	timecounter_init(&fep->tc, &fep->cc, ns);
+
+	spin_unlock_irqrestore(&fep->tmreg_lock, flags);
+
+	/* Restart PPS if needed */
+	if (fep->ptp_saved_state.pps_enable) {
+		/* Re-enable PPS */
+		fec_ptp_enable_pps(fep, 1);
 	}
+}
 
-	return 0;
+void fec_ptp_stop(struct platform_device *pdev)
+{
+	struct net_device *ndev = platform_get_drvdata(pdev);
+	struct fec_enet_private *fep = netdev_priv(ndev);
+
+	if (fep->pps_enable)
+		fec_ptp_enable_pps(fep, 0);
+
+	cancel_delayed_work_sync(&fep->time_keep);
+	hrtimer_cancel(&fep->perout_timer);
+	if (fep->ptp_clock)
+		ptp_clock_unregister(fep->ptp_clock);
 }
diff --git a/drivers/net/ethernet/freescale/fman/Kconfig b/drivers/net/ethernet/freescale/fman/Kconfig
index dc0850b3b517..a55542c1ad65 100644
--- a/drivers/net/ethernet/freescale/fman/Kconfig
+++ b/drivers/net/ethernet/freescale/fman/Kconfig
@@ -1,9 +1,40 @@
+# SPDX-License-Identifier: GPL-2.0-only
 config FSL_FMAN
 	tristate "FMan support"
 	depends on FSL_SOC || ARCH_LAYERSCAPE || COMPILE_TEST
 	select GENERIC_ALLOCATOR
-	select PHYLIB
+	select PHYLINK
+	select PCS_LYNX
+	select CRC32
 	default n
 	help
 		Freescale Data-Path Acceleration Architecture Frame Manager
 		(FMan) support
+
+config DPAA_ERRATUM_A050385
+	bool
+	depends on ARM64 && FSL_DPAA
+	default y
+	help
+		DPAA FMan erratum A050385 software workaround implementation:
+		align buffers, data start, SG fragment length to avoid FMan DMA
+		splits.
+		FMAN DMA read or writes under heavy traffic load may cause FMAN
+		internal resource leak thus stopping further packet processing.
+		The FMAN internal queue can overflow when FMAN splits single
+		read or write transactions into multiple smaller transactions
+		such that more than 17 AXI transactions are in flight from FMAN
+		to interconnect. When the FMAN internal queue overflows, it can
+		stall further packet processing. The issue can occur with any
+		one of the following three conditions:
+		1. FMAN AXI transaction crosses 4K address boundary (Errata
+		A010022)
+		2. FMAN DMA address for an AXI transaction is not 16 byte
+		aligned, i.e. the last 4 bits of an address are non-zero
+		3. Scatter Gather (SG) frames have more than one SG buffer in
+		the SG list and any one of the buffers, except the last
+		buffer in the SG list has data size that is not a multiple
+		of 16 bytes, i.e., other than 16, 32, 48, 64, etc.
+		With any one of the above three conditions present, there is
+		likelihood of stalled FMAN packet processing, especially under
+		stress with multiple ports injecting line-rate traffic.
diff --git a/drivers/net/ethernet/freescale/fman/Makefile b/drivers/net/ethernet/freescale/fman/Makefile
index 60491779e49f..b618091db091 100644
--- a/drivers/net/ethernet/freescale/fman/Makefile
+++ b/drivers/net/ethernet/freescale/fman/Makefile
@@ -1,9 +1,10 @@
+# SPDX-License-Identifier: GPL-2.0
 subdir-ccflags-y +=  -I$(srctree)/drivers/net/ethernet/freescale/fman
 
-obj-$(CONFIG_FSL_FMAN) += fsl_fman.o
-obj-$(CONFIG_FSL_FMAN) += fsl_fman_port.o
-obj-$(CONFIG_FSL_FMAN) += fsl_mac.o
+obj-$(CONFIG_FSL_FMAN) += fsl_dpaa_fman.o
+obj-$(CONFIG_FSL_FMAN) += fsl_dpaa_fman_port.o
+obj-$(CONFIG_FSL_FMAN) += fsl_dpaa_mac.o
 
-fsl_fman-objs	:= fman_muram.o fman.o fman_sp.o
-fsl_fman_port-objs := fman_port.o
-fsl_mac-objs:= mac.o fman_dtsec.o fman_memac.o fman_tgec.o
+fsl_dpaa_fman-objs	:= fman_muram.o fman.o fman_sp.o fman_keygen.o
+fsl_dpaa_fman_port-objs := fman_port.o
+fsl_dpaa_mac-objs:= mac.o fman_dtsec.o fman_memac.o fman_tgec.o
diff --git a/drivers/net/ethernet/freescale/fman/fman.c b/drivers/net/ethernet/freescale/fman/fman.c
index 4aefe2438969..11887458f050 100644
--- a/drivers/net/ethernet/freescale/fman/fman.c
+++ b/drivers/net/ethernet/freescale/fman/fman.c
@@ -1,41 +1,13 @@
+// SPDX-License-Identifier: BSD-3-Clause OR GPL-2.0-or-later
 /*
- * Copyright 2008-2015 Freescale Semiconductor Inc.
- *
- * Redistribution and use in source and binary forms, with or without
- * modification, are permitted provided that the following conditions are met:
- *     * Redistributions of source code must retain the above copyright
- *       notice, this list of conditions and the following disclaimer.
- *     * Redistributions in binary form must reproduce the above copyright
- *       notice, this list of conditions and the following disclaimer in the
- *       documentation and/or other materials provided with the distribution.
- *     * Neither the name of Freescale Semiconductor nor the
- *       names of its contributors may be used to endorse or promote products
- *       derived from this software without specific prior written permission.
- *
- *
- * ALTERNATIVELY, this software may be distributed under the terms of the
- * GNU General Public License ("GPL") as published by the Free Software
- * Foundation, either version 2 of that License or (at your option) any
- * later version.
- *
- * THIS SOFTWARE IS PROVIDED BY Freescale Semiconductor ``AS IS'' AND ANY
- * EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED
- * WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE
- * DISCLAIMED. IN NO EVENT SHALL Freescale Semiconductor BE LIABLE FOR ANY
- * DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES
- * (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES;
- * LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND
- * ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
- * (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS
- * SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
+ * Copyright 2008 - 2015 Freescale Semiconductor Inc.
+ * Copyright 2020 NXP
  */
 
 #define pr_fmt(fmt) KBUILD_MODNAME ": " fmt
 
-#include "fman.h"
-#include "fman_muram.h"
-
 #include <linux/fsl/guts.h>
+#include <linux/platform_device.h>
 #include <linux/slab.h>
 #include <linux/delay.h>
 #include <linux/module.h>
@@ -46,9 +18,12 @@
 #include <linux/interrupt.h>
 #include <linux/libfdt_env.h>
 
+#include "fman.h"
+#include "fman_muram.h"
+#include "fman_keygen.h"
+
 /* General defines */
 #define FMAN_LIODN_TBL			64	/* size of LIODN table */
-#define MAX_NUM_OF_MACS			10
 #define FM_NUM_OF_FMAN_CTRL_EVENT_REGS	4
 #define BASE_RX_PORTID			0x08
 #define BASE_TX_PORTID			0x28
@@ -56,6 +31,7 @@
 /* Modules registers offsets */
 #define BMI_OFFSET		0x00080000
 #define QMI_OFFSET		0x00080400
+#define KG_OFFSET		0x000C1000
 #define DMA_OFFSET		0x000C2000
 #define FPM_OFFSET		0x000C3000
 #define IMEM_OFFSET		0x000C4000
@@ -564,79 +540,9 @@ struct fman_cfg {
 	u32 qmi_def_tnums_thresh;
 };
 
-/* Structure that holds information received from device tree */
-struct fman_dts_params {
-	void __iomem *base_addr;		/* FMan virtual address */
-	struct resource *res;			/* FMan memory resource */
-	u8 id;					/* FMan ID */
-
-	int err_irq;				/* FMan Error IRQ */
-
-	u16 clk_freq;				/* FMan clock freq (In Mhz) */
-
-	u32 qman_channel_base;			/* QMan channels base */
-	u32 num_of_qman_channels;		/* Number of QMan channels */
-
-	struct resource muram_res;		/* MURAM resource */
-};
-
-/** fman_exceptions_cb
- * fman		- Pointer to FMan
- * exception	- The exception.
- *
- * Exceptions user callback routine, will be called upon an exception
- * passing the exception identification.
- *
- * Return: irq status
- */
-typedef irqreturn_t (fman_exceptions_cb)(struct fman *fman,
-					 enum fman_exceptions exception);
-
-/** fman_bus_error_cb
- * fman		- Pointer to FMan
- * port_id	- Port id
- * addr		- Address that caused the error
- * tnum		- Owner of error
- * liodn	- Logical IO device number
- *
- * Bus error user callback routine, will be called upon bus error,
- * passing parameters describing the errors and the owner.
- *
- * Return: IRQ status
- */
-typedef irqreturn_t (fman_bus_error_cb)(struct fman *fman, u8 port_id,
-					u64 addr, u8 tnum, u16 liodn);
-
-struct fman {
-	struct device *dev;
-	void __iomem *base_addr;
-	struct fman_intr_src intr_mng[FMAN_EV_CNT];
-
-	struct fman_fpm_regs __iomem *fpm_regs;
-	struct fman_bmi_regs __iomem *bmi_regs;
-	struct fman_qmi_regs __iomem *qmi_regs;
-	struct fman_dma_regs __iomem *dma_regs;
-	struct fman_hwp_regs __iomem *hwp_regs;
-	fman_exceptions_cb *exception_cb;
-	fman_bus_error_cb *bus_error_cb;
-	/* Spinlock for FMan use */
-	spinlock_t spinlock;
-	struct fman_state_struct *state;
-
-	struct fman_cfg *cfg;
-	struct muram_info *muram;
-	/* cam section in muram */
-	unsigned long cam_offset;
-	size_t cam_size;
-	/* Fifo in MURAM */
-	unsigned long fifo_offset;
-	size_t fifo_size;
-
-	u32 liodn_base[64];
-	u32 liodn_offset[64];
-
-	struct fman_dts_params dts_params;
-};
+#ifdef CONFIG_DPAA_ERRATUM_A050385
+static bool fman_has_err_a050385;
+#endif
 
 static irqreturn_t fman_exceptions(struct fman *fman,
 				   enum fman_exceptions exception)
@@ -706,6 +612,9 @@ static void set_port_liodn(struct fman *fman, u8 port_id,
 {
 	u32 tmp;
 
+	iowrite32be(liodn_ofst, &fman->bmi_regs->fmbm_spliodn[port_id - 1]);
+	if (!IS_ENABLED(CONFIG_FSL_PAMU))
+		return;
 	/* set LIODN base for this port */
 	tmp = ioread32be(&fman->dma_regs->fmdmplr[port_id / 2]);
 	if (port_id % 2) {
@@ -716,7 +625,6 @@ static void set_port_liodn(struct fman *fman, u8 port_id,
 		tmp |= liodn_base << DMA_LIODN_SHIFT;
 	}
 	iowrite32be(tmp, &fman->dma_regs->fmdmplr[port_id / 2]);
-	iowrite32be(liodn_ofst, &fman->bmi_regs->fmbm_spliodn[port_id - 1]);
 }
 
 static void enable_rams_ecc(struct fman_fpm_regs __iomem *fpm_rg)
@@ -1463,8 +1371,7 @@ static void enable_time_stamp(struct fman *fman)
 {
 	struct fman_fpm_regs __iomem *fpm_rg = fman->fpm_regs;
 	u16 fm_clk_freq = fman->state->fm_clk_freq;
-	u32 tmp, intgr, ts_freq;
-	u64 frac;
+	u32 tmp, intgr, ts_freq, frac;
 
 	ts_freq = (u32)(1 << fman->state->count1_micro_bit);
 	/* configure timestamp so that bit 8 will count 1 microsecond
@@ -1811,6 +1718,7 @@ static int fman_config(struct fman *fman)
 	fman->qmi_regs = base_addr + QMI_OFFSET;
 	fman->dma_regs = base_addr + DMA_OFFSET;
 	fman->hwp_regs = base_addr + HWP_OFFSET;
+	fman->kg_regs = base_addr + KG_OFFSET;
 	fman->base_addr = base_addr;
 
 	spin_lock_init(&fman->spinlock);
@@ -1925,8 +1833,8 @@ static int fman_reset(struct fman *fman)
 
 		guts_regs = of_iomap(guts_node, 0);
 		if (!guts_regs) {
-			dev_err(fman->dev, "%s: Couldn't map %s regs\n",
-				__func__, guts_node->full_name);
+			dev_err(fman->dev, "%s: Couldn't map %pOF regs\n",
+				__func__, guts_node);
 			goto guts_regs;
 		}
 #define FMAN1_ALL_MACS_MASK	0xFCC00000
@@ -2013,6 +1921,8 @@ static int fman_init(struct fman *fman)
 
 		fman->liodn_offset[i] =
 			ioread32be(&fman->bmi_regs->fmbm_spliodn[i - 1]);
+		if (!IS_ENABLED(CONFIG_FSL_PAMU))
+			continue;
 		liodn_base = ioread32be(&fman->dma_regs->fmdmplr[i / 2]);
 		if (i % 2) {
 			/* FMDM_PLR LSB holds LIODN base for odd ports */
@@ -2083,6 +1993,11 @@ static int fman_init(struct fman *fman)
 	/* Init HW Parser */
 	hwp_init(fman->hwp_regs);
 
+	/* Init KeyGen */
+	fman->keygen = keygen_init(fman->kg_regs);
+	if (!fman->keygen)
+		return -EINVAL;
+
 	err = enable(fman, cfg);
 	if (err != 0)
 		return err;
@@ -2121,11 +2036,11 @@ static int fman_set_exception(struct fman *fman,
 /**
  * fman_register_intr
  * @fman:	A Pointer to FMan device
- * @mod:	Calling module
+ * @module:	Calling module
  * @mod_id:	Module id (if more than 1 exists, '0' if not)
  * @intr_type:	Interrupt type (error/normal) selection.
- * @f_isr:	The interrupt service routine.
- * @h_src_arg:	Argument to be passed to f_isr.
+ * @isr_cb:	The interrupt service routine.
+ * @src_arg:	Argument to be passed to isr_cb.
  *
  * Used to register an event handler to be processed by FMan
  *
@@ -2149,7 +2064,7 @@ EXPORT_SYMBOL(fman_register_intr);
 /**
  * fman_unregister_intr
  * @fman:	A Pointer to FMan device
- * @mod:	Calling module
+ * @module:	Calling module
  * @mod_id:	Module id (if more than 1 exists, '0' if not)
  * @intr_type:	Interrupt type (error/normal) selection.
  *
@@ -2400,8 +2315,8 @@ EXPORT_SYMBOL(fman_get_bmi_max_fifo_size);
 
 /**
  * fman_get_revision
- * @fman		- Pointer to the FMan module
- * @rev_info		- A structure of revision information parameters.
+ * @fman:		- Pointer to the FMan module
+ * @rev_info:		- A structure of revision information parameters.
  *
  * Returns the FM revision
  *
@@ -2434,15 +2349,21 @@ u32 fman_get_qman_channel_id(struct fman *fman, u32 port_id)
 	int i;
 
 	if (fman->state->rev_info.major >= 6) {
-		u32 port_ids[] = {0x30, 0x31, 0x28, 0x29, 0x2a, 0x2b,
-				  0x2c, 0x2d, 0x2, 0x3, 0x4, 0x5, 0x7, 0x7};
+		static const u32 port_ids[] = {
+			0x30, 0x31, 0x28, 0x29, 0x2a, 0x2b,
+			0x2c, 0x2d, 0x2, 0x3, 0x4, 0x5, 0x7, 0x7
+		};
+
 		for (i = 0; i < fman->state->num_of_qman_channels; i++) {
 			if (port_ids[i] == port_id)
 				break;
 		}
 	} else {
-		u32 port_ids[] = {0x30, 0x28, 0x29, 0x2a, 0x2b, 0x2c, 0x1,
-				  0x2, 0x3, 0x4, 0x5, 0x7, 0x7};
+		static const u32 port_ids[] = {
+			0x30, 0x28, 0x29, 0x2a, 0x2b, 0x2c, 0x1,
+			0x2, 0x3, 0x4, 0x5, 0x7, 0x7
+		};
+
 		for (i = 0; i < fman->state->num_of_qman_channels; i++) {
 			if (port_ids[i] == port_id)
 				break;
@@ -2499,9 +2420,6 @@ MODULE_PARM_DESC(fsl_fm_rx_extra_headroom, "Extra headroom for Rx buffers");
  * buffers when not using jumbo frames.
  * Must be large enough to accommodate the network MTU, but small enough
  * to avoid wasting skb memory.
- *
- * Could be overridden once, at boot-time, via the
- * fm_set_max_frm() callback.
  */
 static int fsl_fm_max_frm = FSL_FM_MAX_FRAME_SIZE;
 module_param(fsl_fm_max_frm, int, 0);
@@ -2563,7 +2481,7 @@ EXPORT_SYMBOL(fman_get_rx_extra_headroom);
 
 /**
  * fman_bind
- * @dev:	FMan OF device pointer
+ * @fm_dev:	FMan OF device pointer
  *
  * Bind to a specific FMan device.
  *
@@ -2577,6 +2495,14 @@ struct fman *fman_bind(struct device *fm_dev)
 }
 EXPORT_SYMBOL(fman_bind);
 
+#ifdef CONFIG_DPAA_ERRATUM_A050385
+bool fman_has_errata_a050385(void)
+{
+	return fman_has_err_a050385;
+}
+EXPORT_SYMBOL(fman_has_errata_a050385);
+#endif
+
 static irqreturn_t fman_err_irq(int irq, void *handle)
 {
 	struct fman *fman = (struct fman *)handle;
@@ -2764,59 +2690,42 @@ static struct fman *read_dts_node(struct platform_device *of_dev)
 {
 	struct fman *fman;
 	struct device_node *fm_node, *muram_node;
+	void __iomem *base_addr;
 	struct resource *res;
 	u32 val, range[2];
 	int err, irq;
 	struct clk *clk;
 	u32 clk_rate;
-	phys_addr_t phys_base_addr;
-	resource_size_t mem_size;
 
 	fman = kzalloc(sizeof(*fman), GFP_KERNEL);
 	if (!fman)
-		return NULL;
+		return ERR_PTR(-ENOMEM);
 
 	fm_node = of_node_get(of_dev->dev.of_node);
 
 	err = of_property_read_u32(fm_node, "cell-index", &val);
 	if (err) {
-		dev_err(&of_dev->dev, "%s: failed to read cell-index for %s\n",
-			__func__, fm_node->full_name);
+		dev_err(&of_dev->dev, "%s: failed to read cell-index for %pOF\n",
+			__func__, fm_node);
 		goto fman_node_put;
 	}
 	fman->dts_params.id = (u8)val;
 
 	/* Get the FM interrupt */
-	res = platform_get_resource(of_dev, IORESOURCE_IRQ, 0);
-	if (!res) {
-		dev_err(&of_dev->dev, "%s: Can't get FMan IRQ resource\n",
-			__func__);
+	err = platform_get_irq(of_dev, 0);
+	if (err < 0)
 		goto fman_node_put;
-	}
-	irq = res->start;
+	irq = err;
 
 	/* Get the FM error interrupt */
-	res = platform_get_resource(of_dev, IORESOURCE_IRQ, 1);
-	if (!res) {
-		dev_err(&of_dev->dev, "%s: Can't get FMan Error IRQ resource\n",
-			__func__);
-		goto fman_node_put;
-	}
-	fman->dts_params.err_irq = res->start;
-
-	/* Get the FM address */
-	res = platform_get_resource(of_dev, IORESOURCE_MEM, 0);
-	if (!res) {
-		dev_err(&of_dev->dev, "%s: Can't get FMan memory resource\n",
-			__func__);
+	err = platform_get_irq(of_dev, 1);
+	if (err < 0)
 		goto fman_node_put;
-	}
-
-	phys_base_addr = res->start;
-	mem_size = resource_size(res);
+	fman->dts_params.err_irq = err;
 
 	clk = of_clk_get(fm_node, 0);
 	if (IS_ERR(clk)) {
+		err = PTR_ERR(clk);
 		dev_err(&of_dev->dev, "%s: Failed to get FM%d clock structure\n",
 			__func__, fman->dts_params.id);
 		goto fman_node_put;
@@ -2824,6 +2733,7 @@ static struct fman *read_dts_node(struct platform_device *of_dev)
 
 	clk_rate = clk_get_rate(clk);
 	if (!clk_rate) {
+		err = -EINVAL;
 		dev_err(&of_dev->dev, "%s: Failed to determine FM%d clock rate\n",
 			__func__, fman->dts_params.id);
 		goto fman_node_put;
@@ -2834,8 +2744,8 @@ static struct fman *read_dts_node(struct platform_device *of_dev)
 	err = of_property_read_u32_array(fm_node, "fsl,qman-channel-range",
 					 &range[0], 2);
 	if (err) {
-		dev_err(&of_dev->dev, "%s: failed to read fsl,qman-channel-range for %s\n",
-			__func__, fm_node->full_name);
+		dev_err(&of_dev->dev, "%s: failed to read fsl,qman-channel-range for %pOF\n",
+			__func__, fm_node);
 		goto fman_node_put;
 	}
 	fman->dts_params.qman_channel_base = range[0];
@@ -2844,9 +2754,10 @@ static struct fman *read_dts_node(struct platform_device *of_dev)
 	/* Get the MURAM base address and size */
 	muram_node = of_find_matching_node(fm_node, fman_muram_match);
 	if (!muram_node) {
+		err = -EINVAL;
 		dev_err(&of_dev->dev, "%s: could not find MURAM node\n",
 			__func__);
-		goto fman_node_put;
+		goto fman_free;
 	}
 
 	err = of_address_to_resource(muram_node, 0,
@@ -2855,13 +2766,13 @@ static struct fman *read_dts_node(struct platform_device *of_dev)
 		of_node_put(muram_node);
 		dev_err(&of_dev->dev, "%s: of_address_to_resource() = %d\n",
 			__func__, err);
-		goto fman_node_put;
+		goto fman_free;
 	}
 
 	of_node_put(muram_node);
-	of_node_put(fm_node);
 
-	err = devm_request_irq(&of_dev->dev, irq, fman_irq, 0, "fman", fman);
+	err = devm_request_irq(&of_dev->dev, irq, fman_irq, IRQF_SHARED,
+			       "fman", fman);
 	if (err < 0) {
 		dev_err(&of_dev->dev, "%s: irq %d allocation failed (error = %d)\n",
 			__func__, irq, err);
@@ -2879,22 +2790,16 @@ static struct fman *read_dts_node(struct platform_device *of_dev)
 		}
 	}
 
-	fman->dts_params.res =
-		devm_request_mem_region(&of_dev->dev, phys_base_addr,
-					mem_size, "fman");
-	if (!fman->dts_params.res) {
-		dev_err(&of_dev->dev, "%s: request_mem_region() failed\n",
-			__func__);
-		goto fman_free;
-	}
-
-	fman->dts_params.base_addr =
-		devm_ioremap(&of_dev->dev, phys_base_addr, mem_size);
-	if (!fman->dts_params.base_addr) {
+	base_addr = devm_platform_get_and_ioremap_resource(of_dev, 0, &res);
+	if (IS_ERR(base_addr)) {
+		err = PTR_ERR(base_addr);
 		dev_err(&of_dev->dev, "%s: devm_ioremap() failed\n", __func__);
 		goto fman_free;
 	}
 
+	fman->dts_params.base_addr = base_addr;
+	fman->dts_params.res = res;
+
 	fman->dev = &of_dev->dev;
 
 	err = of_platform_populate(fm_node, NULL, NULL, &of_dev->dev);
@@ -2904,13 +2809,18 @@ static struct fman *read_dts_node(struct platform_device *of_dev)
 		goto fman_free;
 	}
 
+#ifdef CONFIG_DPAA_ERRATUM_A050385
+	fman_has_err_a050385 =
+		of_property_read_bool(fm_node, "fsl,erratum-a050385");
+#endif
+
 	return fman;
 
 fman_node_put:
 	of_node_put(fm_node);
 fman_free:
 	kfree(fman);
-	return NULL;
+	return ERR_PTR(err);
 }
 
 static int fman_probe(struct platform_device *of_dev)
@@ -2922,8 +2832,8 @@ static int fman_probe(struct platform_device *of_dev)
 	dev = &of_dev->dev;
 
 	fman = read_dts_node(of_dev);
-	if (!fman)
-		return -EIO;
+	if (IS_ERR(fman))
+		return PTR_ERR(fman);
 
 	err = fman_config(fman);
 	if (err) {
diff --git a/drivers/net/ethernet/freescale/fman/fman.h b/drivers/net/ethernet/freescale/fman/fman.h
index f53e1473dbcc..74eb62eba0d7 100644
--- a/drivers/net/ethernet/freescale/fman/fman.h
+++ b/drivers/net/ethernet/freescale/fman/fman.h
@@ -1,44 +1,21 @@
+/* SPDX-License-Identifier: BSD-3-Clause OR GPL-2.0-or-later */
 /*
- * Copyright 2008-2015 Freescale Semiconductor Inc.
- *
- * Redistribution and use in source and binary forms, with or without
- * modification, are permitted provided that the following conditions are met:
- *     * Redistributions of source code must retain the above copyright
- *       notice, this list of conditions and the following disclaimer.
- *     * Redistributions in binary form must reproduce the above copyright
- *       notice, this list of conditions and the following disclaimer in the
- *       documentation and/or other materials provided with the distribution.
- *     * Neither the name of Freescale Semiconductor nor the
- *       names of its contributors may be used to endorse or promote products
- *       derived from this software without specific prior written permission.
- *
- *
- * ALTERNATIVELY, this software may be distributed under the terms of the
- * GNU General Public License ("GPL") as published by the Free Software
- * Foundation, either version 2 of that License or (at your option) any
- * later version.
- *
- * THIS SOFTWARE IS PROVIDED BY Freescale Semiconductor ``AS IS'' AND ANY
- * EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED
- * WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE
- * DISCLAIMED. IN NO EVENT SHALL Freescale Semiconductor BE LIABLE FOR ANY
- * DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES
- * (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES;
- * LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND
- * ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
- * (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS
- * SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
+ * Copyright 2008 - 2015 Freescale Semiconductor Inc.
+ * Copyright 2020 NXP
  */
 
 #ifndef __FM_H
 #define __FM_H
 
 #include <linux/io.h>
+#include <linux/interrupt.h>
+#include <linux/of_irq.h>
 
 /* FM Frame descriptor macros  */
 /* Frame queue Context Override */
 #define FM_FD_CMD_FCO                   0x80000000
 #define FM_FD_CMD_RPD                   0x40000000  /* Read Prepended Data */
+#define FM_FD_CMD_UPD			0x20000000  /* Update Prepended Data */
 #define FM_FD_CMD_DTC                   0x10000000  /* Do L4 Checksum */
 
 /* TX-Port: Unsupported Format */
@@ -97,6 +74,9 @@
 #define BM_MAX_NUM_OF_POOLS		64 /* Buffers pools */
 #define FMAN_PORT_MAX_EXT_POOLS_NUM	8  /* External BM pools per Rx port */
 
+/* General defines */
+#define MAX_NUM_OF_MACS			10
+
 struct fman; /* FMan data */
 
 /* Enum for defining port types */
@@ -274,6 +254,81 @@ struct fman_intr_src {
 	void *src_handle;
 };
 
+/** fman_exceptions_cb
+ * fman         - Pointer to FMan
+ * exception    - The exception.
+ *
+ * Exceptions user callback routine, will be called upon an exception
+ * passing the exception identification.
+ *
+ * Return: irq status
+ */
+typedef irqreturn_t (fman_exceptions_cb)(struct fman *fman,
+					 enum fman_exceptions exception);
+/** fman_bus_error_cb
+ * fman         - Pointer to FMan
+ * port_id      - Port id
+ * addr         - Address that caused the error
+ * tnum         - Owner of error
+ * liodn        - Logical IO device number
+ *
+ * Bus error user callback routine, will be called upon bus error,
+ * passing parameters describing the errors and the owner.
+ *
+ * Return: IRQ status
+ */
+typedef irqreturn_t (fman_bus_error_cb)(struct fman *fman, u8 port_id,
+					u64 addr, u8 tnum, u16 liodn);
+
+/* Structure that holds information received from device tree */
+struct fman_dts_params {
+	void __iomem *base_addr;                /* FMan virtual address */
+	struct resource *res;                   /* FMan memory resource */
+	u8 id;                                  /* FMan ID */
+
+	int err_irq;                            /* FMan Error IRQ */
+
+	u16 clk_freq;                           /* FMan clock freq (In Mhz) */
+
+	u32 qman_channel_base;                  /* QMan channels base */
+	u32 num_of_qman_channels;               /* Number of QMan channels */
+
+	struct resource muram_res;              /* MURAM resource */
+};
+
+struct fman {
+	struct device *dev;
+	void __iomem *base_addr;
+	struct fman_intr_src intr_mng[FMAN_EV_CNT];
+
+	struct fman_fpm_regs __iomem *fpm_regs;
+	struct fman_bmi_regs __iomem *bmi_regs;
+	struct fman_qmi_regs __iomem *qmi_regs;
+	struct fman_dma_regs __iomem *dma_regs;
+	struct fman_hwp_regs __iomem *hwp_regs;
+	struct fman_kg_regs __iomem *kg_regs;
+	fman_exceptions_cb *exception_cb;
+	fman_bus_error_cb *bus_error_cb;
+	/* Spinlock for FMan use */
+	spinlock_t spinlock;
+	struct fman_state_struct *state;
+
+	struct fman_cfg *cfg;
+	struct muram_info *muram;
+	struct fman_keygen *keygen;
+	/* cam section in muram */
+	unsigned long cam_offset;
+	size_t cam_size;
+	/* Fifo in MURAM */
+	unsigned long fifo_offset;
+	size_t fifo_size;
+
+	u32 liodn_base[64];
+	u32 liodn_offset[64];
+
+	struct fman_dts_params dts_params;
+};
+
 /* Structure for port-FM communication during fman_port_init. */
 struct fman_port_init_params {
 	u8 port_id;			/* port Id */
@@ -320,6 +375,10 @@ u16 fman_get_max_frm(void);
 
 int fman_get_rx_extra_headroom(void);
 
+#ifdef CONFIG_DPAA_ERRATUM_A050385
+bool fman_has_errata_a050385(void);
+#endif
+
 struct fman *fman_bind(struct device *dev);
 
 #endif /* __FM_H */
diff --git a/drivers/net/ethernet/freescale/fman/fman_dtsec.c b/drivers/net/ethernet/freescale/fman/fman_dtsec.c
index 98bba10fc38c..51402dff72c5 100644
--- a/drivers/net/ethernet/freescale/fman/fman_dtsec.c
+++ b/drivers/net/ethernet/freescale/fman/fman_dtsec.c
@@ -1,39 +1,13 @@
+// SPDX-License-Identifier: BSD-3-Clause OR GPL-2.0-or-later
 /*
- * Copyright 2008-2015 Freescale Semiconductor Inc.
- *
- * Redistribution and use in source and binary forms, with or without
- * modification, are permitted provided that the following conditions are met:
- *     * Redistributions of source code must retain the above copyright
- *       notice, this list of conditions and the following disclaimer.
- *     * Redistributions in binary form must reproduce the above copyright
- *       notice, this list of conditions and the following disclaimer in the
- *       documentation and/or other materials provided with the distribution.
- *     * Neither the name of Freescale Semiconductor nor the
- *       names of its contributors may be used to endorse or promote products
- *       derived from this software without specific prior written permission.
- *
- *
- * ALTERNATIVELY, this software may be distributed under the terms of the
- * GNU General Public License ("GPL") as published by the Free Software
- * Foundation, either version 2 of that License or (at your option) any
- * later version.
- *
- * THIS SOFTWARE IS PROVIDED BY Freescale Semiconductor ``AS IS'' AND ANY
- * EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED
- * WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE
- * DISCLAIMED. IN NO EVENT SHALL Freescale Semiconductor BE LIABLE FOR ANY
- * DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES
- * (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES;
- * LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND
- * ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
- * (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS
- * SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
+ * Copyright 2008 - 2015 Freescale Semiconductor Inc.
  */
 
 #define pr_fmt(fmt) KBUILD_MODNAME ": " fmt
 
 #include "fman_dtsec.h"
 #include "fman.h"
+#include "mac.h"
 
 #include <linux/slab.h>
 #include <linux/bitrev.h>
@@ -43,6 +17,7 @@
 #include <linux/crc32.h>
 #include <linux/of_mdio.h>
 #include <linux/mii.h>
+#include <linux/netdevice.h>
 
 /* TBI register addresses */
 #define MII_TBICON		0x11
@@ -55,9 +30,6 @@
 #define TBICON_CLK_SELECT	0x0020	/* Clock select */
 #define TBICON_MI_MODE		0x0010	/* GMII mode (TBI if not set) */
 
-#define TBIANA_SGMII		0x4001
-#define TBIANA_1000X		0x01a0
-
 /* Interrupt Mask Register (IMASK) */
 #define DTSEC_IMASK_BREN	0x80000000
 #define DTSEC_IMASK_RXCEN	0x40000000
@@ -118,16 +90,19 @@
 
 #define DTSEC_ECNTRL_GMIIM		0x00000040
 #define DTSEC_ECNTRL_TBIM		0x00000020
-#define DTSEC_ECNTRL_SGMIIM		0x00000002
 #define DTSEC_ECNTRL_RPM		0x00000010
 #define DTSEC_ECNTRL_R100M		0x00000008
+#define DTSEC_ECNTRL_RMM		0x00000004
+#define DTSEC_ECNTRL_SGMIIM		0x00000002
 #define DTSEC_ECNTRL_QSGMIIM		0x00000001
 
-#define DTSEC_TCTRL_GTS			0x00000020
+#define TCTRL_TTSE			0x00000040
+#define TCTRL_GTS			0x00000020
 
 #define RCTRL_PAL_MASK			0x001f0000
 #define RCTRL_PAL_SHIFT			16
 #define RCTRL_GHTX			0x00000400
+#define RCTRL_RTSE			0x00000040
 #define RCTRL_GRS			0x00000020
 #define RCTRL_MPROM			0x00000008
 #define RCTRL_RSF			0x00000004
@@ -325,7 +300,7 @@ struct fman_mac {
 	/* Ethernet physical interface */
 	phy_interface_t phy_if;
 	u16 max_speed;
-	void *dev_id; /* device cookie used by the exception cbs */
+	struct mac_device *dev_id; /* device cookie used by the exception cbs */
 	fman_mac_exception_cb *exception_cb;
 	fman_mac_exception_cb *event_cb;
 	/* Number of individual addresses in registers for this station */
@@ -342,7 +317,8 @@ struct fman_mac {
 	void *fm;
 	struct fman_rev_info fm_rev_info;
 	bool basex_if;
-	struct phy_device *tbiphy;
+	struct mdio_device *tbidev;
+	struct phylink_pcs pcs;
 };
 
 static void set_dflts(struct dtsec_cfg *cfg)
@@ -364,59 +340,30 @@ static void set_dflts(struct dtsec_cfg *cfg)
 	cfg->maximum_frame = DEFAULT_MAXIMUM_FRAME;
 }
 
+static void set_mac_address(struct dtsec_regs __iomem *regs, const u8 *adr)
+{
+	u32 tmp;
+
+	tmp = (u32)((adr[5] << 24) |
+		    (adr[4] << 16) | (adr[3] << 8) | adr[2]);
+	iowrite32be(tmp, &regs->macstnaddr1);
+
+	tmp = (u32)((adr[1] << 24) | (adr[0] << 16));
+	iowrite32be(tmp, &regs->macstnaddr2);
+}
+
 static int init(struct dtsec_regs __iomem *regs, struct dtsec_cfg *cfg,
-		phy_interface_t iface, u16 iface_speed, u8 *macaddr,
+		phy_interface_t iface, u16 iface_speed, u64 addr,
 		u32 exception_mask, u8 tbi_addr)
 {
-	bool is_rgmii, is_sgmii, is_qsgmii;
+	enet_addr_t eth_addr;
+	u32 tmp = 0;
 	int i;
-	u32 tmp;
 
 	/* Soft reset */
 	iowrite32be(MACCFG1_SOFT_RESET, &regs->maccfg1);
 	iowrite32be(0, &regs->maccfg1);
 
-	/* dtsec_id2 */
-	tmp = ioread32be(&regs->tsec_id2);
-
-	/* check RGMII support */
-	if (iface == PHY_INTERFACE_MODE_RGMII ||
-	    iface == PHY_INTERFACE_MODE_RGMII_ID ||
-	    iface == PHY_INTERFACE_MODE_RGMII_RXID ||
-	    iface == PHY_INTERFACE_MODE_RGMII_TXID ||
-	    iface == PHY_INTERFACE_MODE_RMII)
-		if (tmp & DTSEC_ID2_INT_REDUCED_OFF)
-			return -EINVAL;
-
-	if (iface == PHY_INTERFACE_MODE_SGMII ||
-	    iface == PHY_INTERFACE_MODE_MII)
-		if (tmp & DTSEC_ID2_INT_REDUCED_OFF)
-			return -EINVAL;
-
-	is_rgmii = iface == PHY_INTERFACE_MODE_RGMII ||
-		   iface == PHY_INTERFACE_MODE_RGMII_ID ||
-		   iface == PHY_INTERFACE_MODE_RGMII_RXID ||
-		   iface == PHY_INTERFACE_MODE_RGMII_TXID;
-	is_sgmii = iface == PHY_INTERFACE_MODE_SGMII;
-	is_qsgmii = iface == PHY_INTERFACE_MODE_QSGMII;
-
-	tmp = 0;
-	if (is_rgmii || iface == PHY_INTERFACE_MODE_GMII)
-		tmp |= DTSEC_ECNTRL_GMIIM;
-	if (is_sgmii)
-		tmp |= (DTSEC_ECNTRL_SGMIIM | DTSEC_ECNTRL_TBIM);
-	if (is_qsgmii)
-		tmp |= (DTSEC_ECNTRL_SGMIIM | DTSEC_ECNTRL_TBIM |
-			DTSEC_ECNTRL_QSGMIIM);
-	if (is_rgmii)
-		tmp |= DTSEC_ECNTRL_RPM;
-	if (iface_speed == SPEED_100)
-		tmp |= DTSEC_ECNTRL_R100M;
-
-	iowrite32be(tmp, &regs->ecntrl);
-
-	tmp = 0;
-
 	if (cfg->tx_pause_time)
 		tmp |= cfg->tx_pause_time;
 	if (cfg->tx_pause_time_extd)
@@ -457,17 +404,10 @@ static int init(struct dtsec_regs __iomem *regs, struct dtsec_cfg *cfg,
 
 	tmp = 0;
 
-	if (iface_speed < SPEED_1000)
-		tmp |= MACCFG2_NIBBLE_MODE;
-	else if (iface_speed == SPEED_1000)
-		tmp |= MACCFG2_BYTE_MODE;
-
 	tmp |= (cfg->preamble_len << MACCFG2_PREAMBLE_LENGTH_SHIFT) &
 		MACCFG2_PREAMBLE_LENGTH_MASK;
 	if (cfg->tx_pad_crc)
 		tmp |= MACCFG2_PAD_CRC_EN;
-	/* Full Duplex */
-	tmp |= MACCFG2_FULL_DUPLEX;
 	iowrite32be(tmp, &regs->maccfg2);
 
 	tmp = (((cfg->non_back_to_back_ipg1 <<
@@ -499,12 +439,10 @@ static int init(struct dtsec_regs __iomem *regs, struct dtsec_cfg *cfg,
 
 	iowrite32be(0xffffffff, &regs->ievent);
 
-	tmp = (u32)((macaddr[5] << 24) |
-		    (macaddr[4] << 16) | (macaddr[3] << 8) | macaddr[2]);
-	iowrite32be(tmp, &regs->macstnaddr1);
-
-	tmp = (u32)((macaddr[1] << 24) | (macaddr[0] << 16));
-	iowrite32be(tmp, &regs->macstnaddr2);
+	if (addr) {
+		MAKE_ENET_ADDR_FROM_UINT64(addr, eth_addr);
+		set_mac_address(regs, (const u8 *)eth_addr);
+	}
 
 	/* HASH */
 	for (i = 0; i < NUM_OF_HASH_REGS; i++) {
@@ -517,18 +455,6 @@ static int init(struct dtsec_regs __iomem *regs, struct dtsec_cfg *cfg,
 	return 0;
 }
 
-static void set_mac_address(struct dtsec_regs __iomem *regs, u8 *adr)
-{
-	u32 tmp;
-
-	tmp = (u32)((adr[5] << 24) |
-		    (adr[4] << 16) | (adr[3] << 8) | adr[2]);
-	iowrite32be(tmp, &regs->macstnaddr1);
-
-	tmp = (u32)((adr[1] << 24) | (adr[0] << 16));
-	iowrite32be(tmp, &regs->macstnaddr2);
-}
-
 static void set_bucket(struct dtsec_regs __iomem *regs, int bucket,
 		       bool enable)
 {
@@ -550,14 +476,6 @@ static void set_bucket(struct dtsec_regs __iomem *regs, int bucket,
 
 static int check_init_parameters(struct fman_mac *dtsec)
 {
-	if (dtsec->max_speed >= SPEED_10000) {
-		pr_err("1G MAC driver supports 1G or lower speeds\n");
-		return -EINVAL;
-	}
-	if (dtsec->addr == 0) {
-		pr_err("Ethernet MAC Must have a valid MAC Address\n");
-		return -EINVAL;
-	}
 	if ((dtsec->dtsec_drv_param)->rx_prepend >
 	    MAX_PACKET_ALIGNMENT) {
 		pr_err("packetAlignmentPadding can't be > than %d\n",
@@ -659,22 +577,10 @@ static int get_exception_flag(enum fman_mac_exceptions exception)
 	return bit_mask;
 }
 
-static bool is_init_done(struct dtsec_cfg *dtsec_drv_params)
-{
-	/* Checks if dTSEC driver parameters were initialized */
-	if (!dtsec_drv_params)
-		return true;
-
-	return false;
-}
-
 static u16 dtsec_get_max_frame_length(struct fman_mac *dtsec)
 {
 	struct dtsec_regs __iomem *regs = dtsec->regs;
 
-	if (is_init_done(dtsec->dtsec_drv_param))
-		return 0;
-
 	return (u16)ioread32be(&regs->maxfrm);
 }
 
@@ -711,6 +617,7 @@ static void dtsec_isr(void *handle)
 		dtsec->exception_cb(dtsec->dev_id, FM_MAC_EX_1G_COL_RET_LMT);
 	if (event & DTSEC_IMASK_XFUNEN) {
 		/* FM_TX_LOCKUP_ERRATA_DTSEC6 Errata workaround */
+		/* FIXME: This races with the rest of the driver! */
 		if (dtsec->fm_rev_info.major == 2) {
 			u32 tpkt1, tmp_reg1, tpkt2, tmp_reg2, i;
 			/* a. Write 0x00E0_0C00 to DTSEC_ID
@@ -843,101 +750,98 @@ static void free_init_resources(struct fman_mac *dtsec)
 	dtsec->unicast_addr_hash = NULL;
 }
 
-int dtsec_cfg_max_frame_len(struct fman_mac *dtsec, u16 new_val)
+static struct fman_mac *pcs_to_dtsec(struct phylink_pcs *pcs)
 {
-	if (is_init_done(dtsec->dtsec_drv_param))
-		return -EINVAL;
+	return container_of(pcs, struct fman_mac, pcs);
+}
 
-	dtsec->dtsec_drv_param->maximum_frame = new_val;
+static void dtsec_pcs_get_state(struct phylink_pcs *pcs, unsigned int neg_mode,
+				struct phylink_link_state *state)
+{
+	struct fman_mac *dtsec = pcs_to_dtsec(pcs);
 
-	return 0;
+	phylink_mii_c22_pcs_get_state(dtsec->tbidev, neg_mode, state);
 }
 
-int dtsec_cfg_pad_and_crc(struct fman_mac *dtsec, bool new_val)
+static int dtsec_pcs_config(struct phylink_pcs *pcs, unsigned int neg_mode,
+			    phy_interface_t interface,
+			    const unsigned long *advertising,
+			    bool permit_pause_to_mac)
 {
-	if (is_init_done(dtsec->dtsec_drv_param))
-		return -EINVAL;
+	struct fman_mac *dtsec = pcs_to_dtsec(pcs);
 
-	dtsec->dtsec_drv_param->tx_pad_crc = new_val;
-
-	return 0;
+	return phylink_mii_c22_pcs_config(dtsec->tbidev, interface,
+					  advertising, neg_mode);
 }
 
-int dtsec_enable(struct fman_mac *dtsec, enum comm_mode mode)
+static void dtsec_pcs_an_restart(struct phylink_pcs *pcs)
 {
-	struct dtsec_regs __iomem *regs = dtsec->regs;
-	u32 tmp;
+	struct fman_mac *dtsec = pcs_to_dtsec(pcs);
 
-	if (!is_init_done(dtsec->dtsec_drv_param))
-		return -EINVAL;
-
-	/* Enable */
-	tmp = ioread32be(&regs->maccfg1);
-	if (mode & COMM_MODE_RX)
-		tmp |= MACCFG1_RX_EN;
-	if (mode & COMM_MODE_TX)
-		tmp |= MACCFG1_TX_EN;
+	phylink_mii_c22_pcs_an_restart(dtsec->tbidev);
+}
 
-	iowrite32be(tmp, &regs->maccfg1);
+static const struct phylink_pcs_ops dtsec_pcs_ops = {
+	.pcs_get_state = dtsec_pcs_get_state,
+	.pcs_config = dtsec_pcs_config,
+	.pcs_an_restart = dtsec_pcs_an_restart,
+};
 
-	/* Graceful start - clear the graceful receive stop bit */
-	if (mode & COMM_MODE_TX)
-		iowrite32be(ioread32be(&regs->tctrl) & ~DTSEC_TCTRL_GTS,
-			    &regs->tctrl);
-	if (mode & COMM_MODE_RX)
-		iowrite32be(ioread32be(&regs->rctrl) & ~RCTRL_GRS,
-			    &regs->rctrl);
+static void graceful_start(struct fman_mac *dtsec)
+{
+	struct dtsec_regs __iomem *regs = dtsec->regs;
 
-	return 0;
+	iowrite32be(ioread32be(&regs->tctrl) & ~TCTRL_GTS, &regs->tctrl);
+	iowrite32be(ioread32be(&regs->rctrl) & ~RCTRL_GRS, &regs->rctrl);
 }
 
-int dtsec_disable(struct fman_mac *dtsec, enum comm_mode mode)
+static void graceful_stop(struct fman_mac *dtsec)
 {
 	struct dtsec_regs __iomem *regs = dtsec->regs;
 	u32 tmp;
 
-	if (!is_init_done(dtsec->dtsec_drv_param))
-		return -EINVAL;
-
-	/* Gracefull stop - Assert the graceful transmit stop bit */
-	if (mode & COMM_MODE_RX) {
-		tmp = ioread32be(&regs->rctrl) | RCTRL_GRS;
-		iowrite32be(tmp, &regs->rctrl);
-
-		if (dtsec->fm_rev_info.major == 2)
-			usleep_range(100, 200);
-		else
-			udelay(10);
-	}
+	/* Graceful stop - Assert the graceful Rx stop bit */
+	tmp = ioread32be(&regs->rctrl) | RCTRL_GRS;
+	iowrite32be(tmp, &regs->rctrl);
 
-	if (mode & COMM_MODE_TX) {
-		if (dtsec->fm_rev_info.major == 2)
-			pr_debug("GTS not supported due to DTSEC_A004 errata.\n");
-		else
-			pr_debug("GTS not supported due to DTSEC_A0014 errata.\n");
+	if (dtsec->fm_rev_info.major == 2) {
+		/* Workaround for dTSEC Errata A002 */
+		usleep_range(100, 200);
+	} else {
+		/* Workaround for dTSEC Errata A004839 */
+		usleep_range(10, 50);
 	}
 
-	tmp = ioread32be(&regs->maccfg1);
-	if (mode & COMM_MODE_RX)
-		tmp &= ~MACCFG1_RX_EN;
-	if (mode & COMM_MODE_TX)
-		tmp &= ~MACCFG1_TX_EN;
+	/* Graceful stop - Assert the graceful Tx stop bit */
+	if (dtsec->fm_rev_info.major == 2) {
+		/* dTSEC Errata A004: Do not use TCTRL[GTS]=1 */
+		pr_debug("GTS not supported due to DTSEC_A004 Errata.\n");
+	} else {
+		tmp = ioread32be(&regs->tctrl) | TCTRL_GTS;
+		iowrite32be(tmp, &regs->tctrl);
 
-	iowrite32be(tmp, &regs->maccfg1);
+		/* Workaround for dTSEC Errata A0012, A0014 */
+		usleep_range(10, 50);
+	}
+}
 
+static int dtsec_enable(struct fman_mac *dtsec)
+{
 	return 0;
 }
 
-int dtsec_set_tx_pause_frames(struct fman_mac *dtsec,
-			      u8 __maybe_unused priority,
-			      u16 pause_time, u16 __maybe_unused thresh_time)
+static void dtsec_disable(struct fman_mac *dtsec)
+{
+}
+
+static int dtsec_set_tx_pause_frames(struct fman_mac *dtsec,
+				     u8 __maybe_unused priority,
+				     u16 pause_time,
+				     u16 __maybe_unused thresh_time)
 {
 	struct dtsec_regs __iomem *regs = dtsec->regs;
 	u32 ptv = 0;
 
-	if (!is_init_done(dtsec->dtsec_drv_param))
-		return -EINVAL;
-
 	if (pause_time) {
 		/* FM_BAD_TX_TS_IN_B_2_B_ERRATA_DTSEC_A003 Errata workaround */
 		if (dtsec->fm_rev_info.major == 2 && pause_time <= 320) {
@@ -961,14 +865,11 @@ int dtsec_set_tx_pause_frames(struct fman_mac *dtsec,
 	return 0;
 }
 
-int dtsec_accept_rx_pause_frames(struct fman_mac *dtsec, bool en)
+static int dtsec_accept_rx_pause_frames(struct fman_mac *dtsec, bool en)
 {
 	struct dtsec_regs __iomem *regs = dtsec->regs;
 	u32 tmp;
 
-	if (!is_init_done(dtsec->dtsec_drv_param))
-		return -EINVAL;
-
 	tmp = ioread32be(&regs->maccfg1);
 	if (en)
 		tmp |= MACCFG1_RX_FLOW;
@@ -979,21 +880,136 @@ int dtsec_accept_rx_pause_frames(struct fman_mac *dtsec, bool en)
 	return 0;
 }
 
-int dtsec_modify_mac_address(struct fman_mac *dtsec, enet_addr_t *enet_addr)
+static struct phylink_pcs *dtsec_select_pcs(struct phylink_config *config,
+					    phy_interface_t iface)
 {
-	if (!is_init_done(dtsec->dtsec_drv_param))
-		return -EINVAL;
+	struct fman_mac *dtsec = fman_config_to_mac(config)->fman_mac;
+
+	switch (iface) {
+	case PHY_INTERFACE_MODE_SGMII:
+	case PHY_INTERFACE_MODE_1000BASEX:
+	case PHY_INTERFACE_MODE_2500BASEX:
+		return &dtsec->pcs;
+	default:
+		return NULL;
+	}
+}
+
+static void dtsec_mac_config(struct phylink_config *config, unsigned int mode,
+			     const struct phylink_link_state *state)
+{
+	struct mac_device *mac_dev = fman_config_to_mac(config);
+	struct dtsec_regs __iomem *regs = mac_dev->fman_mac->regs;
+	u32 tmp;
+
+	switch (state->interface) {
+	case PHY_INTERFACE_MODE_RMII:
+		tmp = DTSEC_ECNTRL_RMM;
+		break;
+	case PHY_INTERFACE_MODE_RGMII:
+	case PHY_INTERFACE_MODE_RGMII_ID:
+	case PHY_INTERFACE_MODE_RGMII_RXID:
+	case PHY_INTERFACE_MODE_RGMII_TXID:
+		tmp = DTSEC_ECNTRL_GMIIM | DTSEC_ECNTRL_RPM;
+		break;
+	case PHY_INTERFACE_MODE_SGMII:
+	case PHY_INTERFACE_MODE_1000BASEX:
+	case PHY_INTERFACE_MODE_2500BASEX:
+		tmp = DTSEC_ECNTRL_TBIM | DTSEC_ECNTRL_SGMIIM;
+		break;
+	default:
+		dev_warn(mac_dev->dev, "cannot configure dTSEC for %s\n",
+			 phy_modes(state->interface));
+		return;
+	}
+
+	iowrite32be(tmp, &regs->ecntrl);
+}
+
+static void dtsec_link_up(struct phylink_config *config, struct phy_device *phy,
+			  unsigned int mode, phy_interface_t interface,
+			  int speed, int duplex, bool tx_pause, bool rx_pause)
+{
+	struct mac_device *mac_dev = fman_config_to_mac(config);
+	struct fman_mac *dtsec = mac_dev->fman_mac;
+	struct dtsec_regs __iomem *regs = dtsec->regs;
+	u16 pause_time = tx_pause ? FSL_FM_PAUSE_TIME_ENABLE :
+			 FSL_FM_PAUSE_TIME_DISABLE;
+	u32 tmp;
+
+	dtsec_set_tx_pause_frames(dtsec, 0, pause_time, 0);
+	dtsec_accept_rx_pause_frames(dtsec, rx_pause);
+
+	tmp = ioread32be(&regs->ecntrl);
+	if (speed == SPEED_100)
+		tmp |= DTSEC_ECNTRL_R100M;
+	else
+		tmp &= ~DTSEC_ECNTRL_R100M;
+	iowrite32be(tmp, &regs->ecntrl);
+
+	tmp = ioread32be(&regs->maccfg2);
+	tmp &= ~(MACCFG2_NIBBLE_MODE | MACCFG2_BYTE_MODE | MACCFG2_FULL_DUPLEX);
+	if (speed >= SPEED_1000)
+		tmp |= MACCFG2_BYTE_MODE;
+	else
+		tmp |= MACCFG2_NIBBLE_MODE;
+
+	if (duplex == DUPLEX_FULL)
+		tmp |= MACCFG2_FULL_DUPLEX;
+
+	iowrite32be(tmp, &regs->maccfg2);
+
+	mac_dev->update_speed(mac_dev, speed);
+
+	/* Enable */
+	tmp = ioread32be(&regs->maccfg1);
+	tmp |= MACCFG1_RX_EN | MACCFG1_TX_EN;
+	iowrite32be(tmp, &regs->maccfg1);
+
+	/* Graceful start - clear the graceful Rx/Tx stop bit */
+	graceful_start(dtsec);
+}
+
+static void dtsec_link_down(struct phylink_config *config, unsigned int mode,
+			    phy_interface_t interface)
+{
+	struct fman_mac *dtsec = fman_config_to_mac(config)->fman_mac;
+	struct dtsec_regs __iomem *regs = dtsec->regs;
+	u32 tmp;
+
+	/* Graceful stop - Assert the graceful Rx/Tx stop bit */
+	graceful_stop(dtsec);
+
+	tmp = ioread32be(&regs->maccfg1);
+	tmp &= ~(MACCFG1_RX_EN | MACCFG1_TX_EN);
+	iowrite32be(tmp, &regs->maccfg1);
+}
+
+static const struct phylink_mac_ops dtsec_mac_ops = {
+	.mac_select_pcs = dtsec_select_pcs,
+	.mac_config = dtsec_mac_config,
+	.mac_link_up = dtsec_link_up,
+	.mac_link_down = dtsec_link_down,
+};
+
+static int dtsec_modify_mac_address(struct fman_mac *dtsec,
+				    const enet_addr_t *enet_addr)
+{
+	graceful_stop(dtsec);
 
 	/* Initialize MAC Station Address registers (1 & 2)
 	 * Station address have to be swapped (big endian to little endian
 	 */
 	dtsec->addr = ENET_ADDR_TO_UINT64(*enet_addr);
-	set_mac_address(dtsec->regs, (u8 *)(*enet_addr));
+	set_mac_address(dtsec->regs, (const u8 *)(*enet_addr));
+
+	graceful_start(dtsec);
 
 	return 0;
 }
 
-int dtsec_add_hash_mac_address(struct fman_mac *dtsec, enet_addr_t *eth_addr)
+static int dtsec_add_hash_mac_address(struct fman_mac *dtsec,
+				      enet_addr_t *eth_addr)
 {
 	struct dtsec_regs __iomem *regs = dtsec->regs;
 	struct eth_hash_entry *hash_entry;
@@ -1002,9 +1018,6 @@ int dtsec_add_hash_mac_address(struct fman_mac *dtsec, enet_addr_t *eth_addr)
 	u32 crc = 0xFFFFFFFF;
 	bool mcast, ghtx;
 
-	if (!is_init_done(dtsec->dtsec_drv_param))
-		return -EINVAL;
-
 	addr = ENET_ADDR_TO_UINT64(*eth_addr);
 
 	ghtx = (bool)((ioread32be(&regs->rctrl) & RCTRL_GHTX) ? true : false);
@@ -1042,7 +1055,7 @@ int dtsec_add_hash_mac_address(struct fman_mac *dtsec, enet_addr_t *eth_addr)
 	set_bucket(dtsec->regs, bucket, true);
 
 	/* Create element to be added to the driver hash table */
-	hash_entry = kmalloc(sizeof(*hash_entry), GFP_KERNEL);
+	hash_entry = kmalloc(sizeof(*hash_entry), GFP_ATOMIC);
 	if (!hash_entry)
 		return -ENOMEM;
 	hash_entry->addr = addr;
@@ -1059,7 +1072,46 @@ int dtsec_add_hash_mac_address(struct fman_mac *dtsec, enet_addr_t *eth_addr)
 	return 0;
 }
 
-int dtsec_del_hash_mac_address(struct fman_mac *dtsec, enet_addr_t *eth_addr)
+static int dtsec_set_allmulti(struct fman_mac *dtsec, bool enable)
+{
+	u32 tmp;
+	struct dtsec_regs __iomem *regs = dtsec->regs;
+
+	tmp = ioread32be(&regs->rctrl);
+	if (enable)
+		tmp |= RCTRL_MPROM;
+	else
+		tmp &= ~RCTRL_MPROM;
+
+	iowrite32be(tmp, &regs->rctrl);
+
+	return 0;
+}
+
+static int dtsec_set_tstamp(struct fman_mac *dtsec, bool enable)
+{
+	struct dtsec_regs __iomem *regs = dtsec->regs;
+	u32 rctrl, tctrl;
+
+	rctrl = ioread32be(&regs->rctrl);
+	tctrl = ioread32be(&regs->tctrl);
+
+	if (enable) {
+		rctrl |= RCTRL_RTSE;
+		tctrl |= TCTRL_TTSE;
+	} else {
+		rctrl &= ~RCTRL_RTSE;
+		tctrl &= ~TCTRL_TTSE;
+	}
+
+	iowrite32be(rctrl, &regs->rctrl);
+	iowrite32be(tctrl, &regs->tctrl);
+
+	return 0;
+}
+
+static int dtsec_del_hash_mac_address(struct fman_mac *dtsec,
+				      enet_addr_t *eth_addr)
 {
 	struct dtsec_regs __iomem *regs = dtsec->regs;
 	struct list_head *pos;
@@ -1069,9 +1121,6 @@ int dtsec_del_hash_mac_address(struct fman_mac *dtsec, enet_addr_t *eth_addr)
 	u32 crc = 0xFFFFFFFF;
 	bool mcast, ghtx;
 
-	if (!is_init_done(dtsec->dtsec_drv_param))
-		return -EINVAL;
-
 	addr = ENET_ADDR_TO_UINT64(*eth_addr);
 
 	ghtx = (bool)((ioread32be(&regs->rctrl) & RCTRL_GHTX) ? true : false);
@@ -1101,7 +1150,7 @@ int dtsec_del_hash_mac_address(struct fman_mac *dtsec, enet_addr_t *eth_addr)
 		list_for_each(pos,
 			      &dtsec->multicast_addr_hash->lsts[bucket]) {
 			hash_entry = ETH_HASH_ENTRY_OBJ(pos);
-			if (hash_entry->addr == addr) {
+			if (hash_entry && hash_entry->addr == addr) {
 				list_del_init(&hash_entry->node);
 				kfree(hash_entry);
 				break;
@@ -1114,7 +1163,7 @@ int dtsec_del_hash_mac_address(struct fman_mac *dtsec, enet_addr_t *eth_addr)
 		list_for_each(pos,
 			      &dtsec->unicast_addr_hash->lsts[bucket]) {
 			hash_entry = ETH_HASH_ENTRY_OBJ(pos);
-			if (hash_entry->addr == addr) {
+			if (hash_entry && hash_entry->addr == addr) {
 				list_del_init(&hash_entry->node);
 				kfree(hash_entry);
 				break;
@@ -1130,14 +1179,11 @@ int dtsec_del_hash_mac_address(struct fman_mac *dtsec, enet_addr_t *eth_addr)
 	return 0;
 }
 
-int dtsec_set_promiscuous(struct fman_mac *dtsec, bool new_val)
+static int dtsec_set_promiscuous(struct fman_mac *dtsec, bool new_val)
 {
 	struct dtsec_regs __iomem *regs = dtsec->regs;
 	u32 tmp;
 
-	if (!is_init_done(dtsec->dtsec_drv_param))
-		return -EINVAL;
-
 	/* Set unicast promiscuous */
 	tmp = ioread32be(&regs->rctrl);
 	if (new_val)
@@ -1159,75 +1205,12 @@ int dtsec_set_promiscuous(struct fman_mac *dtsec, bool new_val)
 	return 0;
 }
 
-int dtsec_adjust_link(struct fman_mac *dtsec, u16 speed)
-{
-	struct dtsec_regs __iomem *regs = dtsec->regs;
-	u32 tmp;
-
-	if (!is_init_done(dtsec->dtsec_drv_param))
-		return -EINVAL;
-
-	tmp = ioread32be(&regs->maccfg2);
-
-	/* Full Duplex */
-	tmp |= MACCFG2_FULL_DUPLEX;
-
-	tmp &= ~(MACCFG2_NIBBLE_MODE | MACCFG2_BYTE_MODE);
-	if (speed < SPEED_1000)
-		tmp |= MACCFG2_NIBBLE_MODE;
-	else if (speed == SPEED_1000)
-		tmp |= MACCFG2_BYTE_MODE;
-	iowrite32be(tmp, &regs->maccfg2);
-
-	tmp = ioread32be(&regs->ecntrl);
-	if (speed == SPEED_100)
-		tmp |= DTSEC_ECNTRL_R100M;
-	else
-		tmp &= ~DTSEC_ECNTRL_R100M;
-	iowrite32be(tmp, &regs->ecntrl);
-
-	return 0;
-}
-
-int dtsec_restart_autoneg(struct fman_mac *dtsec)
-{
-	u16 tmp_reg16;
-
-	if (!is_init_done(dtsec->dtsec_drv_param))
-		return -EINVAL;
-
-	tmp_reg16 = phy_read(dtsec->tbiphy, MII_BMCR);
-
-	tmp_reg16 &= ~(BMCR_SPEED100 | BMCR_SPEED1000);
-	tmp_reg16 |= (BMCR_ANENABLE | BMCR_ANRESTART |
-		      BMCR_FULLDPLX | BMCR_SPEED1000);
-
-	phy_write(dtsec->tbiphy, MII_BMCR, tmp_reg16);
-
-	return 0;
-}
-
-int dtsec_get_version(struct fman_mac *dtsec, u32 *mac_version)
-{
-	struct dtsec_regs __iomem *regs = dtsec->regs;
-
-	if (!is_init_done(dtsec->dtsec_drv_param))
-		return -EINVAL;
-
-	*mac_version = ioread32be(&regs->tsec_id);
-
-	return 0;
-}
-
-int dtsec_set_exception(struct fman_mac *dtsec,
-			enum fman_mac_exceptions exception, bool enable)
+static int dtsec_set_exception(struct fman_mac *dtsec,
+			       enum fman_mac_exceptions exception, bool enable)
 {
 	struct dtsec_regs __iomem *regs = dtsec->regs;
 	u32 bit_mask = 0;
 
-	if (!is_init_done(dtsec->dtsec_drv_param))
-		return -EINVAL;
-
 	if (exception != FM_MAC_EX_1G_1588_TS_RX_ERR) {
 		bit_mask = get_exception_flag(exception);
 		if (bit_mask) {
@@ -1273,16 +1256,12 @@ int dtsec_set_exception(struct fman_mac *dtsec,
 	return 0;
 }
 
-int dtsec_init(struct fman_mac *dtsec)
+static int dtsec_init(struct fman_mac *dtsec)
 {
 	struct dtsec_regs __iomem *regs = dtsec->regs;
 	struct dtsec_cfg *dtsec_drv_param;
+	u16 max_frm_ln, tbicon;
 	int err;
-	u16 max_frm_ln;
-	enet_addr_t eth_addr;
-
-	if (is_init_done(dtsec->dtsec_drv_param))
-		return -EINVAL;
 
 	if (DEFAULT_RESET_ON_INIT &&
 	    (fman_reset_mac(dtsec->fm, dtsec->mac_id) != 0)) {
@@ -1296,42 +1275,21 @@ int dtsec_init(struct fman_mac *dtsec)
 
 	dtsec_drv_param = dtsec->dtsec_drv_param;
 
-	MAKE_ENET_ADDR_FROM_UINT64(dtsec->addr, eth_addr);
-
 	err = init(dtsec->regs, dtsec_drv_param, dtsec->phy_if,
-		   dtsec->max_speed, (u8 *)eth_addr, dtsec->exceptions,
-		   dtsec->tbiphy->mdio.addr);
+		   dtsec->max_speed, dtsec->addr, dtsec->exceptions,
+		   dtsec->tbidev->addr);
 	if (err) {
 		free_init_resources(dtsec);
 		pr_err("DTSEC version doesn't support this i/f mode\n");
 		return err;
 	}
 
-	if (dtsec->phy_if == PHY_INTERFACE_MODE_SGMII) {
-		u16 tmp_reg16;
-
-		/* Configure the TBI PHY Control Register */
-		tmp_reg16 = TBICON_CLK_SELECT | TBICON_SOFT_RESET;
-		phy_write(dtsec->tbiphy, MII_TBICON, tmp_reg16);
+	/* Configure the TBI PHY Control Register */
+	tbicon = TBICON_CLK_SELECT | TBICON_SOFT_RESET;
+	mdiodev_write(dtsec->tbidev, MII_TBICON, tbicon);
 
-		tmp_reg16 = TBICON_CLK_SELECT;
-		phy_write(dtsec->tbiphy, MII_TBICON, tmp_reg16);
-
-		tmp_reg16 = (BMCR_RESET | BMCR_ANENABLE |
-			     BMCR_FULLDPLX | BMCR_SPEED1000);
-		phy_write(dtsec->tbiphy, MII_BMCR, tmp_reg16);
-
-		if (dtsec->basex_if)
-			tmp_reg16 = TBIANA_1000X;
-		else
-			tmp_reg16 = TBIANA_SGMII;
-		phy_write(dtsec->tbiphy, MII_ADVERTISE, tmp_reg16);
-
-		tmp_reg16 = (BMCR_ANENABLE | BMCR_ANRESTART |
-			     BMCR_FULLDPLX | BMCR_SPEED1000);
-
-		phy_write(dtsec->tbiphy, MII_BMCR, tmp_reg16);
-	}
+	tbicon = TBICON_CLK_SELECT;
+	mdiodev_write(dtsec->tbidev, MII_TBICON, tbicon);
 
 	/* Max Frame Length */
 	max_frm_ln = (u16)ioread32be(&regs->maxfrm);
@@ -1370,24 +1328,24 @@ int dtsec_init(struct fman_mac *dtsec)
 	return 0;
 }
 
-int dtsec_free(struct fman_mac *dtsec)
+static int dtsec_free(struct fman_mac *dtsec)
 {
 	free_init_resources(dtsec);
 
 	kfree(dtsec->dtsec_drv_param);
 	dtsec->dtsec_drv_param = NULL;
+	if (!IS_ERR_OR_NULL(dtsec->tbidev))
+		put_device(&dtsec->tbidev->dev);
 	kfree(dtsec);
 
 	return 0;
 }
 
-struct fman_mac *dtsec_config(struct fman_mac_params *params)
+static struct fman_mac *dtsec_config(struct mac_device *mac_dev,
+				     struct fman_mac_params *params)
 {
 	struct fman_mac *dtsec;
 	struct dtsec_cfg *dtsec_drv_param;
-	void __iomem *base_addr;
-
-	base_addr = params->base_addr;
 
 	/* allocate memory for the UCC GETH data structure. */
 	dtsec = kzalloc(sizeof(*dtsec), GFP_KERNEL);
@@ -1404,10 +1362,9 @@ struct fman_mac *dtsec_config(struct fman_mac_params *params)
 
 	set_dflts(dtsec_drv_param);
 
-	dtsec->regs = base_addr;
-	dtsec->addr = ENET_ADDR_TO_UINT64(params->addr);
-	dtsec->max_speed = params->max_speed;
-	dtsec->phy_if = params->phy_if;
+	dtsec->regs = mac_dev->vaddr;
+	dtsec->addr = ENET_ADDR_TO_UINT64(mac_dev->addr);
+	dtsec->phy_if = mac_dev->phy_if;
 	dtsec->mac_id = params->mac_id;
 	dtsec->exceptions = (DTSEC_IMASK_BREN	|
 			     DTSEC_IMASK_RXCEN	|
@@ -1424,34 +1381,122 @@ struct fman_mac *dtsec_config(struct fman_mac_params *params)
 			     DTSEC_IMASK_RDPEEN);
 	dtsec->exception_cb = params->exception_cb;
 	dtsec->event_cb = params->event_cb;
-	dtsec->dev_id = params->dev_id;
+	dtsec->dev_id = mac_dev;
 	dtsec->ptp_tsu_enabled = dtsec->dtsec_drv_param->ptp_tsu_en;
 	dtsec->en_tsu_err_exception = dtsec->dtsec_drv_param->ptp_exception_en;
 
 	dtsec->fm = params->fm;
-	dtsec->basex_if = params->basex_if;
-
-	if (!params->internal_phy_node) {
-		pr_err("TBI PHY node is not available\n");
-		goto err_dtsec_drv_param;
-	}
-
-	dtsec->tbiphy = of_phy_find_device(params->internal_phy_node);
-	if (!dtsec->tbiphy) {
-		pr_err("of_phy_find_device (TBI PHY) failed\n");
-		goto err_dtsec_drv_param;
-	}
-
-	put_device(&dtsec->tbiphy->mdio.dev);
 
 	/* Save FMan revision */
 	fman_get_revision(dtsec->fm, &dtsec->fm_rev_info);
 
 	return dtsec;
 
-err_dtsec_drv_param:
-	kfree(dtsec_drv_param);
 err_dtsec:
 	kfree(dtsec);
 	return NULL;
 }
+
+int dtsec_initialization(struct mac_device *mac_dev,
+			 struct device_node *mac_node,
+			 struct fman_mac_params *params)
+{
+	int			err;
+	struct fman_mac		*dtsec;
+	struct device_node	*phy_node;
+	unsigned long		 capabilities;
+	unsigned long		*supported;
+
+	mac_dev->phylink_ops		= &dtsec_mac_ops;
+	mac_dev->set_promisc		= dtsec_set_promiscuous;
+	mac_dev->change_addr		= dtsec_modify_mac_address;
+	mac_dev->add_hash_mac_addr	= dtsec_add_hash_mac_address;
+	mac_dev->remove_hash_mac_addr	= dtsec_del_hash_mac_address;
+	mac_dev->set_exception		= dtsec_set_exception;
+	mac_dev->set_allmulti		= dtsec_set_allmulti;
+	mac_dev->set_tstamp		= dtsec_set_tstamp;
+	mac_dev->enable			= dtsec_enable;
+	mac_dev->disable		= dtsec_disable;
+
+	mac_dev->fman_mac = dtsec_config(mac_dev, params);
+	if (!mac_dev->fman_mac) {
+		err = -EINVAL;
+		goto _return;
+	}
+
+	dtsec = mac_dev->fman_mac;
+	dtsec->dtsec_drv_param->maximum_frame = fman_get_max_frm();
+	dtsec->dtsec_drv_param->tx_pad_crc = true;
+
+	phy_node = of_parse_phandle(mac_node, "tbi-handle", 0);
+	if (!phy_node || !of_device_is_available(phy_node)) {
+		of_node_put(phy_node);
+		err = -EINVAL;
+		dev_err_probe(mac_dev->dev, err,
+			      "TBI PCS node is not available\n");
+		goto _return_fm_mac_free;
+	}
+
+	dtsec->tbidev = of_mdio_find_device(phy_node);
+	of_node_put(phy_node);
+	if (!dtsec->tbidev) {
+		err = -EPROBE_DEFER;
+		dev_err_probe(mac_dev->dev, err,
+			      "could not find mdiodev for PCS\n");
+		goto _return_fm_mac_free;
+	}
+	dtsec->pcs.ops = &dtsec_pcs_ops;
+	dtsec->pcs.poll = true;
+
+	supported = mac_dev->phylink_config.supported_interfaces;
+
+	/* FIXME: Can we use DTSEC_ID2_INT_FULL_OFF to determine if these are
+	 * supported? If not, we can determine support via the phy if SerDes
+	 * support is added.
+	 */
+	if (mac_dev->phy_if == PHY_INTERFACE_MODE_SGMII ||
+	    mac_dev->phy_if == PHY_INTERFACE_MODE_1000BASEX) {
+		__set_bit(PHY_INTERFACE_MODE_SGMII, supported);
+		__set_bit(PHY_INTERFACE_MODE_1000BASEX, supported);
+	} else if (mac_dev->phy_if == PHY_INTERFACE_MODE_2500BASEX) {
+		__set_bit(PHY_INTERFACE_MODE_2500BASEX, supported);
+	}
+
+	if (!(ioread32be(&dtsec->regs->tsec_id2) & DTSEC_ID2_INT_REDUCED_OFF)) {
+		phy_interface_set_rgmii(supported);
+
+		/* DTSEC_ID2_INT_REDUCED_OFF indicates that the dTSEC supports
+		 * RMII and RGMII. However, the only SoCs which support RMII
+		 * are the P1017 and P1023. Avoid advertising this mode on
+		 * other SoCs. This is a bit of a moot point, since there's no
+		 * in-tree support for ethernet on these platforms...
+		 */
+		if (of_machine_is_compatible("fsl,P1023") ||
+		    of_machine_is_compatible("fsl,P1023RDB"))
+			__set_bit(PHY_INTERFACE_MODE_RMII, supported);
+	}
+
+	capabilities = MAC_SYM_PAUSE | MAC_ASYM_PAUSE;
+	capabilities |= MAC_10 | MAC_100 | MAC_1000FD | MAC_2500FD;
+	mac_dev->phylink_config.mac_capabilities = capabilities;
+
+	err = dtsec_init(dtsec);
+	if (err < 0)
+		goto _return_fm_mac_free;
+
+	/* For 1G MAC, disable by default the MIB counters overflow interrupt */
+	err = dtsec_set_exception(dtsec, FM_MAC_EX_1G_RX_MIB_CNT_OVFL, false);
+	if (err < 0)
+		goto _return_fm_mac_free;
+
+	dev_info(mac_dev->dev, "FMan dTSEC version: 0x%08x\n",
+		 ioread32be(&dtsec->regs->tsec_id));
+
+	goto _return;
+
+_return_fm_mac_free:
+	dtsec_free(dtsec);
+
+_return:
+	return err;
+}
diff --git a/drivers/net/ethernet/freescale/fman/fman_dtsec.h b/drivers/net/ethernet/freescale/fman/fman_dtsec.h
index c4467c072058..8c72d280c51a 100644
--- a/drivers/net/ethernet/freescale/fman/fman_dtsec.h
+++ b/drivers/net/ethernet/freescale/fman/fman_dtsec.h
@@ -1,33 +1,6 @@
+/* SPDX-License-Identifier: BSD-3-Clause OR GPL-2.0-or-later */
 /*
- * Copyright 2008-2015 Freescale Semiconductor Inc.
- *
- * Redistribution and use in source and binary forms, with or without
- * modification, are permitted provided that the following conditions are met:
- *     * Redistributions of source code must retain the above copyright
- *       notice, this list of conditions and the following disclaimer.
- *     * Redistributions in binary form must reproduce the above copyright
- *       notice, this list of conditions and the following disclaimer in the
- *       documentation and/or other materials provided with the distribution.
- *     * Neither the name of Freescale Semiconductor nor the
- *       names of its contributors may be used to endorse or promote products
- *       derived from this software without specific prior written permission.
- *
- *
- * ALTERNATIVELY, this software may be distributed under the terms of the
- * GNU General Public License ("GPL") as published by the Free Software
- * Foundation, either version 2 of that License or (at your option) any
- * later version.
- *
- * THIS SOFTWARE IS PROVIDED BY Freescale Semiconductor ``AS IS'' AND ANY
- * EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED
- * WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE
- * DISCLAIMED. IN NO EVENT SHALL Freescale Semiconductor BE LIABLE FOR ANY
- * DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES
- * (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES;
- * LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND
- * ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
- * (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS
- * SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
+ * Copyright 2008 - 2015 Freescale Semiconductor Inc.
  */
 
 #ifndef __DTSEC_H
@@ -35,25 +8,10 @@
 
 #include "fman_mac.h"
 
-struct fman_mac *dtsec_config(struct fman_mac_params *params);
-int dtsec_set_promiscuous(struct fman_mac *dtsec, bool new_val);
-int dtsec_modify_mac_address(struct fman_mac *dtsec, enet_addr_t *enet_addr);
-int dtsec_adjust_link(struct fman_mac *dtsec,
-		      u16 speed);
-int dtsec_restart_autoneg(struct fman_mac *dtsec);
-int dtsec_cfg_max_frame_len(struct fman_mac *dtsec, u16 new_val);
-int dtsec_cfg_pad_and_crc(struct fman_mac *dtsec, bool new_val);
-int dtsec_enable(struct fman_mac *dtsec, enum comm_mode mode);
-int dtsec_disable(struct fman_mac *dtsec, enum comm_mode mode);
-int dtsec_init(struct fman_mac *dtsec);
-int dtsec_free(struct fman_mac *dtsec);
-int dtsec_accept_rx_pause_frames(struct fman_mac *dtsec, bool en);
-int dtsec_set_tx_pause_frames(struct fman_mac *dtsec, u8 priority,
-			      u16 pause_time, u16 thresh_time);
-int dtsec_set_exception(struct fman_mac *dtsec,
-			enum fman_mac_exceptions exception, bool enable);
-int dtsec_add_hash_mac_address(struct fman_mac *dtsec, enet_addr_t *eth_addr);
-int dtsec_del_hash_mac_address(struct fman_mac *dtsec, enet_addr_t *eth_addr);
-int dtsec_get_version(struct fman_mac *dtsec, u32 *mac_version);
+struct mac_device;
+
+int dtsec_initialization(struct mac_device *mac_dev,
+			 struct device_node *mac_node,
+			 struct fman_mac_params *params);
 
 #endif /* __DTSEC_H */
diff --git a/drivers/net/ethernet/freescale/fman/fman_keygen.c b/drivers/net/ethernet/freescale/fman/fman_keygen.c
new file mode 100644
index 000000000000..e73f6ef3c6ee
--- /dev/null
+++ b/drivers/net/ethernet/freescale/fman/fman_keygen.c
@@ -0,0 +1,757 @@
+// SPDX-License-Identifier: BSD-3-Clause OR GPL-2.0-or-later
+/*
+ * Copyright 2017 NXP
+ */
+
+#define pr_fmt(fmt) KBUILD_MODNAME ": " fmt
+
+#include <linux/slab.h>
+
+#include "fman_keygen.h"
+
+/* Maximum number of HW Ports */
+#define FMAN_MAX_NUM_OF_HW_PORTS		64
+
+/* Maximum number of KeyGen Schemes */
+#define FM_KG_MAX_NUM_OF_SCHEMES		32
+
+/* Number of generic KeyGen Generic Extract Command Registers */
+#define FM_KG_NUM_OF_GENERIC_REGS		8
+
+/* Dummy port ID */
+#define DUMMY_PORT_ID				0
+
+/* Select Scheme Value Register */
+#define KG_SCH_DEF_USE_KGSE_DV_0		2
+#define KG_SCH_DEF_USE_KGSE_DV_1		3
+
+/* Registers Shifting values */
+#define FM_KG_KGAR_NUM_SHIFT			16
+#define KG_SCH_DEF_L4_PORT_SHIFT		8
+#define KG_SCH_DEF_IP_ADDR_SHIFT		18
+#define KG_SCH_HASH_CONFIG_SHIFT_SHIFT		24
+
+/* KeyGen Registers bit field masks: */
+
+/* Enable bit field mask for KeyGen General Configuration Register */
+#define FM_KG_KGGCR_EN				0x80000000
+
+/* KeyGen Global Registers bit field masks */
+#define FM_KG_KGAR_GO				0x80000000
+#define FM_KG_KGAR_READ				0x40000000
+#define FM_KG_KGAR_WRITE			0x00000000
+#define FM_KG_KGAR_SEL_SCHEME_ENTRY		0x00000000
+#define FM_KG_KGAR_SCM_WSEL_UPDATE_CNT		0x00008000
+
+#define FM_KG_KGAR_ERR				0x20000000
+#define FM_KG_KGAR_SEL_CLS_PLAN_ENTRY		0x01000000
+#define FM_KG_KGAR_SEL_PORT_ENTRY		0x02000000
+#define FM_KG_KGAR_SEL_PORT_WSEL_SP		0x00008000
+#define FM_KG_KGAR_SEL_PORT_WSEL_CPP		0x00004000
+
+/* Error events exceptions */
+#define FM_EX_KG_DOUBLE_ECC			0x80000000
+#define FM_EX_KG_KEYSIZE_OVERFLOW		0x40000000
+
+/* Scheme Registers bit field masks */
+#define KG_SCH_MODE_EN				0x80000000
+#define KG_SCH_VSP_NO_KSP_EN			0x80000000
+#define KG_SCH_HASH_CONFIG_SYM			0x40000000
+
+/* Known Protocol field codes */
+#define KG_SCH_KN_PORT_ID		0x80000000
+#define KG_SCH_KN_MACDST		0x40000000
+#define KG_SCH_KN_MACSRC		0x20000000
+#define KG_SCH_KN_TCI1			0x10000000
+#define KG_SCH_KN_TCI2			0x08000000
+#define KG_SCH_KN_ETYPE			0x04000000
+#define KG_SCH_KN_PPPSID		0x02000000
+#define KG_SCH_KN_PPPID			0x01000000
+#define KG_SCH_KN_MPLS1			0x00800000
+#define KG_SCH_KN_MPLS2			0x00400000
+#define KG_SCH_KN_MPLS_LAST		0x00200000
+#define KG_SCH_KN_IPSRC1		0x00100000
+#define KG_SCH_KN_IPDST1		0x00080000
+#define KG_SCH_KN_PTYPE1		0x00040000
+#define KG_SCH_KN_IPTOS_TC1		0x00020000
+#define KG_SCH_KN_IPV6FL1		0x00010000
+#define KG_SCH_KN_IPSRC2		0x00008000
+#define KG_SCH_KN_IPDST2		0x00004000
+#define KG_SCH_KN_PTYPE2		0x00002000
+#define KG_SCH_KN_IPTOS_TC2		0x00001000
+#define KG_SCH_KN_IPV6FL2		0x00000800
+#define KG_SCH_KN_GREPTYPE		0x00000400
+#define KG_SCH_KN_IPSEC_SPI		0x00000200
+#define KG_SCH_KN_IPSEC_NH		0x00000100
+#define KG_SCH_KN_IPPID			0x00000080
+#define KG_SCH_KN_L4PSRC		0x00000004
+#define KG_SCH_KN_L4PDST		0x00000002
+#define KG_SCH_KN_TFLG			0x00000001
+
+/* NIA values */
+#define NIA_ENG_BMI			0x00500000
+#define NIA_BMI_AC_ENQ_FRAME		0x00000002
+#define ENQUEUE_KG_DFLT_NIA		(NIA_ENG_BMI | NIA_BMI_AC_ENQ_FRAME)
+
+/* Hard-coded configuration:
+ * These values are used as hard-coded values for KeyGen configuration
+ * and they replace user selections for this hard-coded version
+ */
+
+/* Hash distribution shift */
+#define DEFAULT_HASH_DIST_FQID_SHIFT		0
+
+/* Hash shift */
+#define DEFAULT_HASH_SHIFT			0
+
+/* Symmetric hash usage:
+ * Warning:
+ * - the value for symmetric hash usage must be in accordance with hash
+ *	key defined below
+ * - according to tests performed, spreading is not working if symmetric
+ *	hash is set on true
+ * So ultimately symmetric hash functionality should be always disabled:
+ */
+#define DEFAULT_SYMMETRIC_HASH			false
+
+/* Hash Key extraction fields: */
+#define DEFAULT_HASH_KEY_EXTRACT_FIELDS		\
+	(KG_SCH_KN_IPSRC1 | KG_SCH_KN_IPDST1 | \
+	 KG_SCH_KN_L4PSRC | KG_SCH_KN_L4PDST | \
+	 KG_SCH_KN_IPSEC_SPI)
+
+/* Default values to be used as hash key in case IPv4 or L4 (TCP, UDP)
+ * don't exist in the frame
+ */
+/* Default IPv4 address */
+#define DEFAULT_HASH_KEY_IPv4_ADDR		0x0A0A0A0A
+/* Default L4 port */
+#define DEFAULT_HASH_KEY_L4_PORT		0x0B0B0B0B
+
+/* KeyGen Memory Mapped Registers: */
+
+/* Scheme Configuration RAM Registers */
+struct fman_kg_scheme_regs {
+	u32 kgse_mode;		/* 0x100: MODE */
+	u32 kgse_ekfc;		/* 0x104: Extract Known Fields Command */
+	u32 kgse_ekdv;		/* 0x108: Extract Known Default Value */
+	u32 kgse_bmch;		/* 0x10C: Bit Mask Command High */
+	u32 kgse_bmcl;		/* 0x110: Bit Mask Command Low */
+	u32 kgse_fqb;		/* 0x114: Frame Queue Base */
+	u32 kgse_hc;		/* 0x118: Hash Command */
+	u32 kgse_ppc;		/* 0x11C: Policer Profile Command */
+	u32 kgse_gec[FM_KG_NUM_OF_GENERIC_REGS];
+			/* 0x120: Generic Extract Command */
+	u32 kgse_spc;
+		/* 0x140: KeyGen Scheme Entry Statistic Packet Counter */
+	u32 kgse_dv0;	/* 0x144: KeyGen Scheme Entry Default Value 0 */
+	u32 kgse_dv1;	/* 0x148: KeyGen Scheme Entry Default Value 1 */
+	u32 kgse_ccbs;
+		/* 0x14C: KeyGen Scheme Entry Coarse Classification Bit*/
+	u32 kgse_mv;	/* 0x150: KeyGen Scheme Entry Match vector */
+	u32 kgse_om;	/* 0x154: KeyGen Scheme Entry Operation Mode bits */
+	u32 kgse_vsp;
+		/* 0x158: KeyGen Scheme Entry Virtual Storage Profile */
+};
+
+/* Port Partition Configuration Registers */
+struct fman_kg_pe_regs {
+	u32 fmkg_pe_sp;		/* 0x100: KeyGen Port entry Scheme Partition */
+	u32 fmkg_pe_cpp;
+		/* 0x104: KeyGen Port Entry Classification Plan Partition */
+};
+
+/* General Configuration and Status Registers
+ * Global Statistic Counters
+ * KeyGen Global Registers
+ */
+struct fman_kg_regs {
+	u32 fmkg_gcr;	/* 0x000: KeyGen General Configuration Register */
+	u32 res004;	/* 0x004: Reserved */
+	u32 res008;	/* 0x008: Reserved */
+	u32 fmkg_eer;	/* 0x00C: KeyGen Error Event Register */
+	u32 fmkg_eeer;	/* 0x010: KeyGen Error Event Enable Register */
+	u32 res014;	/* 0x014: Reserved */
+	u32 res018;	/* 0x018: Reserved */
+	u32 fmkg_seer;	/* 0x01C: KeyGen Scheme Error Event Register */
+	u32 fmkg_seeer;	/* 0x020: KeyGen Scheme Error Event Enable Register */
+	u32 fmkg_gsr;	/* 0x024: KeyGen Global Status Register */
+	u32 fmkg_tpc;	/* 0x028: Total Packet Counter Register */
+	u32 fmkg_serc;	/* 0x02C: Soft Error Capture Register */
+	u32 res030[4];	/* 0x030: Reserved */
+	u32 fmkg_fdor;	/* 0x034: Frame Data Offset Register */
+	u32 fmkg_gdv0r;	/* 0x038: Global Default Value Register 0 */
+	u32 fmkg_gdv1r;	/* 0x03C: Global Default Value Register 1 */
+	u32 res04c[6];	/* 0x040: Reserved */
+	u32 fmkg_feer;	/* 0x044: Force Error Event Register */
+	u32 res068[38];	/* 0x048: Reserved */
+	union {
+		u32 fmkg_indirect[63];	/* 0x100: Indirect Access Registers */
+		struct fman_kg_scheme_regs fmkg_sch; /* Scheme Registers */
+		struct fman_kg_pe_regs fmkg_pe; /* Port Partition Registers */
+	};
+	u32 fmkg_ar;	/* 0x1FC: KeyGen Action Register */
+};
+
+/* KeyGen Scheme data */
+struct keygen_scheme {
+	bool used;	/* Specifies if this scheme is used */
+	u8 hw_port_id;
+		/* Hardware port ID
+		 * schemes sharing between multiple ports is not
+		 * currently supported
+		 * so we have only one port id bound to a scheme
+		 */
+	u32 base_fqid;
+		/* Base FQID:
+		 * Must be between 1 and 2^24-1
+		 * If hash is used and an even distribution is
+		 * expected according to hash_fqid_count,
+		 * base_fqid must be aligned to hash_fqid_count
+		 */
+	u32 hash_fqid_count;
+		/* FQ range for hash distribution:
+		 * Must be a power of 2
+		 * Represents the range of queues for spreading
+		 */
+	bool use_hashing;	/* Usage of Hashing and spreading over FQ */
+	bool symmetric_hash;	/* Symmetric Hash option usage */
+	u8 hashShift;
+		/* Hash result right shift.
+		 * Select the 24 bits out of the 64 hash result.
+		 * 0 means using the 24 LSB's, otherwise
+		 * use the 24 LSB's after shifting right
+		 */
+	u32 match_vector;	/* Match Vector */
+};
+
+/* KeyGen driver data */
+struct fman_keygen {
+	struct keygen_scheme schemes[FM_KG_MAX_NUM_OF_SCHEMES];
+				/* Array of schemes */
+	struct fman_kg_regs __iomem *keygen_regs;	/* KeyGen registers */
+};
+
+/* keygen_write_ar_wait
+ *
+ * Write Action Register with specified value, wait for GO bit field to be
+ * idle and then read the error
+ *
+ * regs: KeyGen registers
+ * fmkg_ar: Action Register value
+ *
+ * Return: Zero for success or error code in case of failure
+ */
+static int keygen_write_ar_wait(struct fman_kg_regs __iomem *regs, u32 fmkg_ar)
+{
+	iowrite32be(fmkg_ar, &regs->fmkg_ar);
+
+	/* Wait for GO bit field to be idle */
+	while (fmkg_ar & FM_KG_KGAR_GO)
+		fmkg_ar = ioread32be(&regs->fmkg_ar);
+
+	if (fmkg_ar & FM_KG_KGAR_ERR)
+		return -EINVAL;
+
+	return 0;
+}
+
+/* build_ar_scheme
+ *
+ * Build Action Register value for scheme settings
+ *
+ * scheme_id: Scheme ID
+ * update_counter: update scheme counter
+ * write: true for action to write the scheme or false for read action
+ *
+ * Return: AR value
+ */
+static u32 build_ar_scheme(u8 scheme_id, bool update_counter, bool write)
+{
+	u32 rw = (u32)(write ? FM_KG_KGAR_WRITE : FM_KG_KGAR_READ);
+
+	return (u32)(FM_KG_KGAR_GO |
+			rw |
+			FM_KG_KGAR_SEL_SCHEME_ENTRY |
+			DUMMY_PORT_ID |
+			((u32)scheme_id << FM_KG_KGAR_NUM_SHIFT) |
+			(update_counter ? FM_KG_KGAR_SCM_WSEL_UPDATE_CNT : 0));
+}
+
+/* build_ar_bind_scheme
+ *
+ * Build Action Register value for port binding to schemes
+ *
+ * hwport_id: HW Port ID
+ * write: true for action to write the bind or false for read action
+ *
+ * Return: AR value
+ */
+static u32 build_ar_bind_scheme(u8 hwport_id, bool write)
+{
+	u32 rw = write ? (u32)FM_KG_KGAR_WRITE : (u32)FM_KG_KGAR_READ;
+
+	return (u32)(FM_KG_KGAR_GO |
+			rw |
+			FM_KG_KGAR_SEL_PORT_ENTRY |
+			hwport_id |
+			FM_KG_KGAR_SEL_PORT_WSEL_SP);
+}
+
+/* keygen_write_sp
+ *
+ * Write Scheme Partition Register with specified value
+ *
+ * regs: KeyGen Registers
+ * sp: Scheme Partition register value
+ * add: true to add a scheme partition or false to clear
+ *
+ * Return: none
+ */
+static void keygen_write_sp(struct fman_kg_regs __iomem *regs, u32 sp, bool add)
+{
+	u32 tmp;
+
+	tmp = ioread32be(&regs->fmkg_pe.fmkg_pe_sp);
+
+	if (add)
+		tmp |= sp;
+	else
+		tmp &= ~sp;
+
+	iowrite32be(tmp, &regs->fmkg_pe.fmkg_pe_sp);
+}
+
+/* build_ar_bind_cls_plan
+ *
+ * Build Action Register value for Classification Plan
+ *
+ * hwport_id: HW Port ID
+ * write: true for action to write the CP or false for read action
+ *
+ * Return: AR value
+ */
+static u32 build_ar_bind_cls_plan(u8 hwport_id, bool write)
+{
+	u32 rw = write ? (u32)FM_KG_KGAR_WRITE : (u32)FM_KG_KGAR_READ;
+
+	return (u32)(FM_KG_KGAR_GO |
+			rw |
+			FM_KG_KGAR_SEL_PORT_ENTRY |
+			hwport_id |
+			FM_KG_KGAR_SEL_PORT_WSEL_CPP);
+}
+
+/* keygen_write_cpp
+ *
+ * Write Classification Plan Partition Register with specified value
+ *
+ * regs: KeyGen Registers
+ * cpp: CPP register value
+ *
+ * Return: none
+ */
+static void keygen_write_cpp(struct fman_kg_regs __iomem *regs, u32 cpp)
+{
+	iowrite32be(cpp, &regs->fmkg_pe.fmkg_pe_cpp);
+}
+
+/* keygen_write_scheme
+ *
+ * Write all Schemes Registers with specified values
+ *
+ * regs: KeyGen Registers
+ * scheme_id: Scheme ID
+ * scheme_regs: Scheme registers values desired to be written
+ * update_counter: update scheme counter
+ *
+ * Return: Zero for success or error code in case of failure
+ */
+static int keygen_write_scheme(struct fman_kg_regs __iomem *regs, u8 scheme_id,
+			       struct fman_kg_scheme_regs *scheme_regs,
+				bool update_counter)
+{
+	u32 ar_reg;
+	int err, i;
+
+	/* Write indirect scheme registers */
+	iowrite32be(scheme_regs->kgse_mode, &regs->fmkg_sch.kgse_mode);
+	iowrite32be(scheme_regs->kgse_ekfc, &regs->fmkg_sch.kgse_ekfc);
+	iowrite32be(scheme_regs->kgse_ekdv, &regs->fmkg_sch.kgse_ekdv);
+	iowrite32be(scheme_regs->kgse_bmch, &regs->fmkg_sch.kgse_bmch);
+	iowrite32be(scheme_regs->kgse_bmcl, &regs->fmkg_sch.kgse_bmcl);
+	iowrite32be(scheme_regs->kgse_fqb, &regs->fmkg_sch.kgse_fqb);
+	iowrite32be(scheme_regs->kgse_hc, &regs->fmkg_sch.kgse_hc);
+	iowrite32be(scheme_regs->kgse_ppc, &regs->fmkg_sch.kgse_ppc);
+	iowrite32be(scheme_regs->kgse_spc, &regs->fmkg_sch.kgse_spc);
+	iowrite32be(scheme_regs->kgse_dv0, &regs->fmkg_sch.kgse_dv0);
+	iowrite32be(scheme_regs->kgse_dv1, &regs->fmkg_sch.kgse_dv1);
+	iowrite32be(scheme_regs->kgse_ccbs, &regs->fmkg_sch.kgse_ccbs);
+	iowrite32be(scheme_regs->kgse_mv, &regs->fmkg_sch.kgse_mv);
+	iowrite32be(scheme_regs->kgse_om, &regs->fmkg_sch.kgse_om);
+	iowrite32be(scheme_regs->kgse_vsp, &regs->fmkg_sch.kgse_vsp);
+
+	for (i = 0 ; i < FM_KG_NUM_OF_GENERIC_REGS ; i++)
+		iowrite32be(scheme_regs->kgse_gec[i],
+			    &regs->fmkg_sch.kgse_gec[i]);
+
+	/* Write AR (Action register) */
+	ar_reg = build_ar_scheme(scheme_id, update_counter, true);
+	err = keygen_write_ar_wait(regs, ar_reg);
+	if (err != 0) {
+		pr_err("Writing Action Register failed\n");
+		return err;
+	}
+
+	return err;
+}
+
+/* get_free_scheme_id
+ *
+ * Find the first free scheme available to be used
+ *
+ * keygen: KeyGen handle
+ * scheme_id: pointer to scheme id
+ *
+ * Return: 0 on success, -EINVAL when the are no available free schemes
+ */
+static int get_free_scheme_id(struct fman_keygen *keygen, u8 *scheme_id)
+{
+	u8 i;
+
+	for (i = 0; i < FM_KG_MAX_NUM_OF_SCHEMES; i++)
+		if (!keygen->schemes[i].used) {
+			*scheme_id = i;
+			return 0;
+		}
+
+	return -EINVAL;
+}
+
+/* get_scheme
+ *
+ * Provides the scheme for specified ID
+ *
+ * keygen: KeyGen handle
+ * scheme_id: Scheme ID
+ *
+ * Return: handle to required scheme
+ */
+static struct keygen_scheme *get_scheme(struct fman_keygen *keygen,
+					u8 scheme_id)
+{
+	if (scheme_id >= FM_KG_MAX_NUM_OF_SCHEMES)
+		return NULL;
+	return &keygen->schemes[scheme_id];
+}
+
+/* keygen_bind_port_to_schemes
+ *
+ * Bind the port to schemes
+ *
+ * keygen: KeyGen handle
+ * scheme_id: id of the scheme to bind to
+ * bind: true to bind the port or false to unbind it
+ *
+ * Return: Zero for success or error code in case of failure
+ */
+static int keygen_bind_port_to_schemes(struct fman_keygen *keygen,
+				       u8 scheme_id,
+					bool bind)
+{
+	struct fman_kg_regs __iomem *keygen_regs = keygen->keygen_regs;
+	struct keygen_scheme *scheme;
+	u32 ar_reg;
+	u32 schemes_vector = 0;
+	int err;
+
+	scheme = get_scheme(keygen, scheme_id);
+	if (!scheme) {
+		pr_err("Requested Scheme does not exist\n");
+		return -EINVAL;
+	}
+	if (!scheme->used) {
+		pr_err("Cannot bind port to an invalid scheme\n");
+		return -EINVAL;
+	}
+
+	schemes_vector |= 1 << (31 - scheme_id);
+
+	ar_reg = build_ar_bind_scheme(scheme->hw_port_id, false);
+	err = keygen_write_ar_wait(keygen_regs, ar_reg);
+	if (err != 0) {
+		pr_err("Reading Action Register failed\n");
+		return err;
+	}
+
+	keygen_write_sp(keygen_regs, schemes_vector, bind);
+
+	ar_reg = build_ar_bind_scheme(scheme->hw_port_id, true);
+	err = keygen_write_ar_wait(keygen_regs, ar_reg);
+	if (err != 0) {
+		pr_err("Writing Action Register failed\n");
+		return err;
+	}
+
+	return 0;
+}
+
+/* keygen_scheme_setup
+ *
+ * Setup the scheme according to required configuration
+ *
+ * keygen: KeyGen handle
+ * scheme_id: scheme ID
+ * enable: true to enable scheme or false to disable it
+ *
+ * Return: Zero for success or error code in case of failure
+ */
+static int keygen_scheme_setup(struct fman_keygen *keygen, u8 scheme_id,
+			       bool enable)
+{
+	struct fman_kg_regs __iomem *keygen_regs = keygen->keygen_regs;
+	struct fman_kg_scheme_regs scheme_regs;
+	struct keygen_scheme *scheme;
+	u32 tmp_reg;
+	int err;
+
+	scheme = get_scheme(keygen, scheme_id);
+	if (!scheme) {
+		pr_err("Requested Scheme does not exist\n");
+		return -EINVAL;
+	}
+	if (enable && scheme->used) {
+		pr_err("The requested Scheme is already used\n");
+		return -EINVAL;
+	}
+
+	/* Clear scheme registers */
+	memset(&scheme_regs, 0, sizeof(struct fman_kg_scheme_regs));
+
+	/* Setup all scheme registers: */
+	tmp_reg = 0;
+
+	if (enable) {
+		/* Enable Scheme */
+		tmp_reg |= KG_SCH_MODE_EN;
+		/* Enqueue frame NIA */
+		tmp_reg |= ENQUEUE_KG_DFLT_NIA;
+	}
+
+	scheme_regs.kgse_mode = tmp_reg;
+
+	scheme_regs.kgse_mv = scheme->match_vector;
+
+	/* Scheme don't override StorageProfile:
+	 * valid only for DPAA_VERSION >= 11
+	 */
+	scheme_regs.kgse_vsp = KG_SCH_VSP_NO_KSP_EN;
+
+	/* Configure Hard-Coded Rx Hashing: */
+
+	if (scheme->use_hashing) {
+		/* configure kgse_ekfc */
+		scheme_regs.kgse_ekfc = DEFAULT_HASH_KEY_EXTRACT_FIELDS;
+
+		/* configure kgse_ekdv */
+		tmp_reg = 0;
+		tmp_reg |= (KG_SCH_DEF_USE_KGSE_DV_0 <<
+				KG_SCH_DEF_IP_ADDR_SHIFT);
+		tmp_reg |= (KG_SCH_DEF_USE_KGSE_DV_1 <<
+				KG_SCH_DEF_L4_PORT_SHIFT);
+		scheme_regs.kgse_ekdv = tmp_reg;
+
+		/* configure kgse_dv0 */
+		scheme_regs.kgse_dv0 = DEFAULT_HASH_KEY_IPv4_ADDR;
+		/* configure kgse_dv1 */
+		scheme_regs.kgse_dv1 = DEFAULT_HASH_KEY_L4_PORT;
+
+		/* configure kgse_hc  */
+		tmp_reg = 0;
+		tmp_reg |= ((scheme->hash_fqid_count - 1) <<
+				DEFAULT_HASH_DIST_FQID_SHIFT);
+		tmp_reg |= scheme->hashShift << KG_SCH_HASH_CONFIG_SHIFT_SHIFT;
+
+		if (scheme->symmetric_hash) {
+			/* Normally extraction key should be verified if
+			 * complies with symmetric hash
+			 * But because extraction is hard-coded, we are sure
+			 * the key is symmetric
+			 */
+			tmp_reg |= KG_SCH_HASH_CONFIG_SYM;
+		}
+		scheme_regs.kgse_hc = tmp_reg;
+	} else {
+		scheme_regs.kgse_ekfc = 0;
+		scheme_regs.kgse_hc = 0;
+		scheme_regs.kgse_ekdv = 0;
+		scheme_regs.kgse_dv0 = 0;
+		scheme_regs.kgse_dv1 = 0;
+	}
+
+	/* configure kgse_fqb: Scheme FQID base */
+	tmp_reg = 0;
+	tmp_reg |= scheme->base_fqid;
+	scheme_regs.kgse_fqb = tmp_reg;
+
+	/* features not used by hard-coded configuration */
+	scheme_regs.kgse_bmch = 0;
+	scheme_regs.kgse_bmcl = 0;
+	scheme_regs.kgse_spc = 0;
+
+	/* Write scheme registers */
+	err = keygen_write_scheme(keygen_regs, scheme_id, &scheme_regs, true);
+	if (err != 0) {
+		pr_err("Writing scheme registers failed\n");
+		return err;
+	}
+
+	/* Update used field for Scheme */
+	scheme->used = enable;
+
+	return 0;
+}
+
+/* keygen_init
+ *
+ * KeyGen initialization:
+ * Initializes and enables KeyGen, allocate driver memory, setup registers,
+ * clear port bindings, invalidate all schemes
+ *
+ * keygen_regs: KeyGen registers base address
+ *
+ * Return: Handle to KeyGen driver
+ */
+struct fman_keygen *keygen_init(struct fman_kg_regs __iomem *keygen_regs)
+{
+	struct fman_keygen *keygen;
+	u32 ar;
+	int i;
+
+	/* Allocate memory for KeyGen driver */
+	keygen = kzalloc(sizeof(*keygen), GFP_KERNEL);
+	if (!keygen)
+		return NULL;
+
+	keygen->keygen_regs = keygen_regs;
+
+	/* KeyGen initialization (for Master partition):
+	 * Setup KeyGen registers
+	 */
+	iowrite32be(ENQUEUE_KG_DFLT_NIA, &keygen_regs->fmkg_gcr);
+
+	iowrite32be(FM_EX_KG_DOUBLE_ECC | FM_EX_KG_KEYSIZE_OVERFLOW,
+		    &keygen_regs->fmkg_eer);
+
+	iowrite32be(0, &keygen_regs->fmkg_fdor);
+	iowrite32be(0, &keygen_regs->fmkg_gdv0r);
+	iowrite32be(0, &keygen_regs->fmkg_gdv1r);
+
+	/* Clear binding between ports to schemes and classification plans
+	 * so that all ports are not bound to any scheme/classification plan
+	 */
+	for (i = 0; i < FMAN_MAX_NUM_OF_HW_PORTS; i++) {
+		/* Clear all pe sp schemes registers */
+		keygen_write_sp(keygen_regs, 0xffffffff, false);
+		ar = build_ar_bind_scheme(i, true);
+		keygen_write_ar_wait(keygen_regs, ar);
+
+		/* Clear all pe cpp classification plans registers */
+		keygen_write_cpp(keygen_regs, 0);
+		ar = build_ar_bind_cls_plan(i, true);
+		keygen_write_ar_wait(keygen_regs, ar);
+	}
+
+	/* Enable all scheme interrupts */
+	iowrite32be(0xFFFFFFFF, &keygen_regs->fmkg_seer);
+	iowrite32be(0xFFFFFFFF, &keygen_regs->fmkg_seeer);
+
+	/* Enable KyeGen */
+	iowrite32be(ioread32be(&keygen_regs->fmkg_gcr) | FM_KG_KGGCR_EN,
+		    &keygen_regs->fmkg_gcr);
+
+	return keygen;
+}
+EXPORT_SYMBOL(keygen_init);
+
+/* keygen_port_hashing_init
+ *
+ * Initializes a port for Rx Hashing with specified configuration parameters
+ *
+ * keygen: KeyGen handle
+ * hw_port_id: HW Port ID
+ * hash_base_fqid: Hashing Base FQID used for spreading
+ * hash_size: Hashing size
+ *
+ * Return: Zero for success or error code in case of failure
+ */
+int keygen_port_hashing_init(struct fman_keygen *keygen, u8 hw_port_id,
+			     u32 hash_base_fqid, u32 hash_size)
+{
+	struct keygen_scheme *scheme;
+	u8 scheme_id;
+	int err;
+
+	/* Validate Scheme configuration parameters */
+	if (hash_base_fqid == 0 || (hash_base_fqid & ~0x00FFFFFF)) {
+		pr_err("Base FQID must be between 1 and 2^24-1\n");
+		return -EINVAL;
+	}
+	if (hash_size == 0 || (hash_size & (hash_size - 1)) != 0) {
+		pr_err("Hash size must be power of two\n");
+		return -EINVAL;
+	}
+
+	/* Find a free scheme */
+	err = get_free_scheme_id(keygen, &scheme_id);
+	if (err) {
+		pr_err("The maximum number of available Schemes has been exceeded\n");
+		return -EINVAL;
+	}
+
+	/* Create and configure Hard-Coded Scheme: */
+
+	scheme = get_scheme(keygen, scheme_id);
+	if (!scheme) {
+		pr_err("Requested Scheme does not exist\n");
+		return -EINVAL;
+	}
+	if (scheme->used) {
+		pr_err("The requested Scheme is already used\n");
+		return -EINVAL;
+	}
+
+	/* Clear all scheme fields because the scheme may have been
+	 * previously used
+	 */
+	memset(scheme, 0, sizeof(struct keygen_scheme));
+
+	/* Setup scheme: */
+	scheme->hw_port_id = hw_port_id;
+	scheme->use_hashing = true;
+	scheme->base_fqid = hash_base_fqid;
+	scheme->hash_fqid_count = hash_size;
+	scheme->symmetric_hash = DEFAULT_SYMMETRIC_HASH;
+	scheme->hashShift = DEFAULT_HASH_SHIFT;
+
+	/* All Schemes in hard-coded configuration
+	 * are Indirect Schemes
+	 */
+	scheme->match_vector = 0;
+
+	err = keygen_scheme_setup(keygen, scheme_id, true);
+	if (err != 0) {
+		pr_err("Scheme setup failed\n");
+		return err;
+	}
+
+	/* Bind Rx port to Scheme */
+	err = keygen_bind_port_to_schemes(keygen, scheme_id, true);
+	if (err != 0) {
+		pr_err("Binding port to schemes failed\n");
+		return err;
+	}
+
+	return 0;
+}
+EXPORT_SYMBOL(keygen_port_hashing_init);
diff --git a/drivers/net/ethernet/freescale/fman/fman_keygen.h b/drivers/net/ethernet/freescale/fman/fman_keygen.h
new file mode 100644
index 000000000000..2cb0df453074
--- /dev/null
+++ b/drivers/net/ethernet/freescale/fman/fman_keygen.h
@@ -0,0 +1,19 @@
+/* SPDX-License-Identifier: BSD-3-Clause OR GPL-2.0-or-later */
+/*
+ * Copyright 2017 NXP
+ */
+
+#ifndef __KEYGEN_H
+#define __KEYGEN_H
+
+#include <linux/io.h>
+
+struct fman_keygen;
+struct fman_kg_regs;
+
+struct fman_keygen *keygen_init(struct fman_kg_regs __iomem *keygen_regs);
+
+int keygen_port_hashing_init(struct fman_keygen *keygen, u8 hw_port_id,
+			     u32 hash_base_fqid, u32 hash_size);
+
+#endif /* __KEYGEN_H */
diff --git a/drivers/net/ethernet/freescale/fman/fman_mac.h b/drivers/net/ethernet/freescale/fman/fman_mac.h
index dd6d0526f6c1..e5d6cddea731 100644
--- a/drivers/net/ethernet/freescale/fman/fman_mac.h
+++ b/drivers/net/ethernet/freescale/fman/fman_mac.h
@@ -41,6 +41,7 @@
 #include <linux/if_ether.h>
 
 struct fman_mac;
+struct mac_device;
 
 /* Ethernet Address */
 typedef u8 enet_addr_t[ETH_ALEN];
@@ -75,16 +76,6 @@ typedef u8 enet_addr_t[ETH_ALEN];
 #define ETH_HASH_ENTRY_OBJ(ptr)	\
 	hlist_entry_safe(ptr, struct eth_hash_entry, node)
 
-/* Enumeration (bit flags) of communication modes (Transmit,
- * receive or both).
- */
-enum comm_mode {
-	COMM_MODE_NONE = 0,	/* No transmit/receive communication */
-	COMM_MODE_RX = 1,	/* Only receive communication */
-	COMM_MODE_TX = 2,	/* Only transmit communication */
-	COMM_MODE_RX_AND_TX = 3	/* Both transmit and receive communication */
-};
-
 /* FM MAC Exceptions */
 enum fman_mac_exceptions {
 	FM_MAC_EX_10G_MDIO_SCAN_EVENT = 0
@@ -168,40 +159,21 @@ struct eth_hash_entry {
 	struct list_head node;
 };
 
-typedef void (fman_mac_exception_cb)(void *dev_id,
-				    enum fman_mac_exceptions exceptions);
+typedef void (fman_mac_exception_cb)(struct mac_device *dev_id,
+				     enum fman_mac_exceptions exceptions);
 
 /* FMan MAC config input */
 struct fman_mac_params {
-	/* Base of memory mapped FM MAC registers */
-	void __iomem *base_addr;
-	/* MAC address of device; First octet is sent first */
-	enet_addr_t addr;
 	/* MAC ID; numbering of dTSEC and 1G-mEMAC:
 	 * 0 - FM_MAX_NUM_OF_1G_MACS;
 	 * numbering of 10G-MAC (TGEC) and 10G-mEMAC:
 	 * 0 - FM_MAX_NUM_OF_10G_MACS
 	 */
 	u8 mac_id;
-	/* PHY interface */
-	phy_interface_t	 phy_if;
-	/* Note that the speed should indicate the maximum rate that
-	 * this MAC should support rather than the actual speed;
-	 */
-	u16 max_speed;
 	/* A handle to the FM object this port related to */
 	void *fm;
-	void *dev_id; /* device cookie used by the exception cbs */
 	fman_mac_exception_cb *event_cb;    /* MDIO Events Callback Routine */
 	fman_mac_exception_cb *exception_cb;/* Exception Callback Routine */
-	/* SGMII/QSGII interface with 1000BaseX auto-negotiation between MAC
-	 * and phy or backplane; Note: 1000BaseX auto-negotiation relates only
-	 * to interface between MAC and phy/backplane, SGMII phy can still
-	 * synchronize with far-end phy at 10Mbps, 100Mbps or 1000Mbps
-	*/
-	bool basex_if;
-	/* Pointer to TBI/PCS PHY node, used for TBI/PCS PHY access */
-	struct device_node *internal_phy_node;
 };
 
 struct eth_hash_t {
@@ -252,7 +224,7 @@ static inline struct eth_hash_t *alloc_hash_table(u16 size)
 	struct eth_hash_t *hash;
 
 	/* Allocate address hash table */
-	hash = kmalloc_array(size, sizeof(struct eth_hash_t *), GFP_KERNEL);
+	hash = kmalloc(sizeof(*hash), GFP_KERNEL);
 	if (!hash)
 		return NULL;
 
diff --git a/drivers/net/ethernet/freescale/fman/fman_memac.c b/drivers/net/ethernet/freescale/fman/fman_memac.c
index c0296880feba..0291093f2e4e 100644
--- a/drivers/net/ethernet/freescale/fman/fman_memac.c
+++ b/drivers/net/ethernet/freescale/fman/fman_memac.c
@@ -1,78 +1,22 @@
+// SPDX-License-Identifier: BSD-3-Clause OR GPL-2.0-or-later
 /*
- * Copyright 2008-2015 Freescale Semiconductor Inc.
- *
- * Redistribution and use in source and binary forms, with or without
- * modification, are permitted provided that the following conditions are met:
- *     * Redistributions of source code must retain the above copyright
- *       notice, this list of conditions and the following disclaimer.
- *     * Redistributions in binary form must reproduce the above copyright
- *       notice, this list of conditions and the following disclaimer in the
- *       documentation and/or other materials provided with the distribution.
- *     * Neither the name of Freescale Semiconductor nor the
- *       names of its contributors may be used to endorse or promote products
- *       derived from this software without specific prior written permission.
- *
- *
- * ALTERNATIVELY, this software may be distributed under the terms of the
- * GNU General Public License ("GPL") as published by the Free Software
- * Foundation, either version 2 of that License or (at your option) any
- * later version.
- *
- * THIS SOFTWARE IS PROVIDED BY Freescale Semiconductor ``AS IS'' AND ANY
- * EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED
- * WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE
- * DISCLAIMED. IN NO EVENT SHALL Freescale Semiconductor BE LIABLE FOR ANY
- * DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES
- * (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES;
- * LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND
- * ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
- * (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS
- * SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
+ * Copyright 2008 - 2015 Freescale Semiconductor Inc.
  */
 
 #define pr_fmt(fmt) KBUILD_MODNAME ": " fmt
 
 #include "fman_memac.h"
 #include "fman.h"
+#include "mac.h"
 
 #include <linux/slab.h>
 #include <linux/io.h>
+#include <linux/pcs-lynx.h>
 #include <linux/phy.h>
 #include <linux/phy_fixed.h>
+#include <linux/phy/phy.h>
 #include <linux/of_mdio.h>
 
-/* PCS registers */
-#define MDIO_SGMII_CR			0x00
-#define MDIO_SGMII_DEV_ABIL_SGMII	0x04
-#define MDIO_SGMII_LINK_TMR_L		0x12
-#define MDIO_SGMII_LINK_TMR_H		0x13
-#define MDIO_SGMII_IF_MODE		0x14
-
-/* SGMII Control defines */
-#define SGMII_CR_AN_EN			0x1000
-#define SGMII_CR_RESTART_AN		0x0200
-#define SGMII_CR_FD			0x0100
-#define SGMII_CR_SPEED_SEL1_1G		0x0040
-#define SGMII_CR_DEF_VAL		(SGMII_CR_AN_EN | SGMII_CR_FD | \
-					 SGMII_CR_SPEED_SEL1_1G)
-
-/* SGMII Device Ability for SGMII defines */
-#define MDIO_SGMII_DEV_ABIL_SGMII_MODE	0x4001
-#define MDIO_SGMII_DEV_ABIL_BASEX_MODE	0x01A0
-
-/* Link timer define */
-#define LINK_TMR_L			0xa120
-#define LINK_TMR_H			0x0007
-#define LINK_TMR_L_BASEX		0xaf08
-#define LINK_TMR_H_BASEX		0x002f
-
-/* SGMII IF Mode defines */
-#define IF_MODE_USE_SGMII_AN		0x0002
-#define IF_MODE_SGMII_EN		0x0001
-#define IF_MODE_SGMII_SPEED_100M	0x0004
-#define IF_MODE_SGMII_SPEED_1G		0x0008
-#define IF_MODE_SGMII_DUPLEX_HALF	0x0010
-
 /* Num of additional exact match MAC adr regs */
 #define MEMAC_NUM_OF_PADDRS 7
 
@@ -110,7 +54,8 @@ do {									\
 /* Interface Mode Register (IF_MODE) */
 
 #define IF_MODE_MASK		0x00000003 /* 30-31 Mask on i/f mode bits */
-#define IF_MODE_XGMII		0x00000000 /* 30-31 XGMII (10G) interface */
+#define IF_MODE_10G		0x00000000 /* 30-31 10G interface */
+#define IF_MODE_MII		0x00000001 /* 30-31 MII interface */
 #define IF_MODE_GMII		0x00000002 /* 30-31 GMII (1G) interface */
 #define IF_MODE_RGMII		0x00000004
 #define IF_MODE_RGMII_AUTO	0x00008000
@@ -322,7 +267,6 @@ struct memac_cfg {
 	bool reset_on_init;
 	bool pause_ignore;
 	bool promiscuous_mode_enable;
-	struct fixed_phy_status *fixed_link;
 	u16 max_frame_length;
 	u16 pause_quanta;
 	u32 tx_ipg_length;
@@ -333,10 +277,7 @@ struct fman_mac {
 	struct memac_regs __iomem *regs;
 	/* MAC address of device */
 	u64 addr;
-	/* Ethernet physical interface */
-	phy_interface_t phy_if;
-	u16 max_speed;
-	void *dev_id; /* device cookie used by the exception cbs */
+	struct mac_device *dev_id; /* device cookie used by the exception cbs */
 	fman_mac_exception_cb *exception_cb;
 	fman_mac_exception_cb *event_cb;
 	/* Pointer to driver's global address hash table  */
@@ -348,11 +289,15 @@ struct fman_mac {
 	struct memac_cfg *memac_drv_param;
 	void *fm;
 	struct fman_rev_info fm_rev_info;
-	bool basex_if;
-	struct phy_device *pcsphy;
+	struct phy *serdes;
+	struct phylink_pcs *sgmii_pcs;
+	struct phylink_pcs *qsgmii_pcs;
+	struct phylink_pcs *xfi_pcs;
+	bool allmulti_enabled;
+	bool rgmii_no_half_duplex;
 };
 
-static void add_addr_in_paddr(struct memac_regs __iomem *regs, u8 *adr,
+static void add_addr_in_paddr(struct memac_regs __iomem *regs, const u8 *adr,
 			      u8 paddr_num)
 {
 	u32 tmp0, tmp1;
@@ -407,7 +352,6 @@ static void set_exception(struct memac_regs __iomem *regs, u32 val,
 }
 
 static int init(struct memac_regs __iomem *regs, struct memac_cfg *cfg,
-		phy_interface_t phy_if, u16 speed, bool slow_10g_if,
 		u32 exceptions)
 {
 	u32 tmp;
@@ -435,38 +379,6 @@ static int init(struct memac_regs __iomem *regs, struct memac_cfg *cfg,
 	iowrite32be((u32)cfg->pause_quanta, &regs->pause_quanta[0]);
 	iowrite32be((u32)0, &regs->pause_thresh[0]);
 
-	/* IF_MODE */
-	tmp = 0;
-	switch (phy_if) {
-	case PHY_INTERFACE_MODE_XGMII:
-		tmp |= IF_MODE_XGMII;
-		break;
-	default:
-		tmp |= IF_MODE_GMII;
-		if (phy_if == PHY_INTERFACE_MODE_RGMII ||
-		    phy_if == PHY_INTERFACE_MODE_RGMII_ID ||
-		    phy_if == PHY_INTERFACE_MODE_RGMII_RXID ||
-		    phy_if == PHY_INTERFACE_MODE_RGMII_TXID)
-			tmp |= IF_MODE_RGMII | IF_MODE_RGMII_AUTO;
-	}
-	iowrite32be(tmp, &regs->if_mode);
-
-	/* TX_FIFO_SECTIONS */
-	tmp = 0;
-	if (phy_if == PHY_INTERFACE_MODE_XGMII) {
-		if (slow_10g_if) {
-			tmp |= (TX_FIFO_SECTIONS_TX_AVAIL_SLOW_10G |
-				TX_FIFO_SECTIONS_TX_EMPTY_DEFAULT_10G);
-		} else {
-			tmp |= (TX_FIFO_SECTIONS_TX_AVAIL_10G |
-				TX_FIFO_SECTIONS_TX_EMPTY_DEFAULT_10G);
-		}
-	} else {
-		tmp |= (TX_FIFO_SECTIONS_TX_AVAIL_1G |
-			TX_FIFO_SECTIONS_TX_EMPTY_DEFAULT_1G);
-	}
-	iowrite32be(tmp, &regs->tx_fifo_sections);
-
 	/* clear all pending events and set-up interrupts */
 	iowrite32be(0xffffffff, &regs->ievent);
 	set_exception(regs, exceptions, true);
@@ -506,99 +418,8 @@ static u32 get_mac_addr_hash_code(u64 eth_addr)
 	return xor_val;
 }
 
-static void setup_sgmii_internal_phy(struct fman_mac *memac,
-				     struct fixed_phy_status *fixed_link)
-{
-	u16 tmp_reg16;
-
-	if (WARN_ON(!memac->pcsphy))
-		return;
-
-	/* SGMII mode */
-	tmp_reg16 = IF_MODE_SGMII_EN;
-	if (!fixed_link)
-		/* AN enable */
-		tmp_reg16 |= IF_MODE_USE_SGMII_AN;
-	else {
-		switch (fixed_link->speed) {
-		case 10:
-			/* For 10M: IF_MODE[SPEED_10M] = 0 */
-		break;
-		case 100:
-			tmp_reg16 |= IF_MODE_SGMII_SPEED_100M;
-		break;
-		case 1000: /* fallthrough */
-		default:
-			tmp_reg16 |= IF_MODE_SGMII_SPEED_1G;
-		break;
-		}
-		if (!fixed_link->duplex)
-			tmp_reg16 |= IF_MODE_SGMII_DUPLEX_HALF;
-	}
-	phy_write(memac->pcsphy, MDIO_SGMII_IF_MODE, tmp_reg16);
-
-	/* Device ability according to SGMII specification */
-	tmp_reg16 = MDIO_SGMII_DEV_ABIL_SGMII_MODE;
-	phy_write(memac->pcsphy, MDIO_SGMII_DEV_ABIL_SGMII, tmp_reg16);
-
-	/* Adjust link timer for SGMII  -
-	 * According to Cisco SGMII specification the timer should be 1.6 ms.
-	 * The link_timer register is configured in units of the clock.
-	 * - When running as 1G SGMII, Serdes clock is 125 MHz, so
-	 * unit = 1 / (125*10^6 Hz) = 8 ns.
-	 * 1.6 ms in units of 8 ns = 1.6ms / 8ns = 2*10^5 = 0x30d40
-	 * - When running as 2.5G SGMII, Serdes clock is 312.5 MHz, so
-	 * unit = 1 / (312.5*10^6 Hz) = 3.2 ns.
-	 * 1.6 ms in units of 3.2 ns = 1.6ms / 3.2ns = 5*10^5 = 0x7a120.
-	 * Since link_timer value of 1G SGMII will be too short for 2.5 SGMII,
-	 * we always set up here a value of 2.5 SGMII.
-	 */
-	phy_write(memac->pcsphy, MDIO_SGMII_LINK_TMR_H, LINK_TMR_H);
-	phy_write(memac->pcsphy, MDIO_SGMII_LINK_TMR_L, LINK_TMR_L);
-
-	if (!fixed_link)
-		/* Restart AN */
-		tmp_reg16 = SGMII_CR_DEF_VAL | SGMII_CR_RESTART_AN;
-	else
-		/* AN disabled */
-		tmp_reg16 = SGMII_CR_DEF_VAL & ~SGMII_CR_AN_EN;
-	phy_write(memac->pcsphy, 0x0, tmp_reg16);
-}
-
-static void setup_sgmii_internal_phy_base_x(struct fman_mac *memac)
-{
-	u16 tmp_reg16;
-
-	/* AN Device capability  */
-	tmp_reg16 = MDIO_SGMII_DEV_ABIL_BASEX_MODE;
-	phy_write(memac->pcsphy, MDIO_SGMII_DEV_ABIL_SGMII, tmp_reg16);
-
-	/* Adjust link timer for SGMII  -
-	 * For Serdes 1000BaseX auto-negotiation the timer should be 10 ms.
-	 * The link_timer register is configured in units of the clock.
-	 * - When running as 1G SGMII, Serdes clock is 125 MHz, so
-	 * unit = 1 / (125*10^6 Hz) = 8 ns.
-	 * 10 ms in units of 8 ns = 10ms / 8ns = 1250000 = 0x1312d0
-	 * - When running as 2.5G SGMII, Serdes clock is 312.5 MHz, so
-	 * unit = 1 / (312.5*10^6 Hz) = 3.2 ns.
-	 * 10 ms in units of 3.2 ns = 10ms / 3.2ns = 3125000 = 0x2faf08.
-	 * Since link_timer value of 1G SGMII will be too short for 2.5 SGMII,
-	 * we always set up here a value of 2.5 SGMII.
-	 */
-	phy_write(memac->pcsphy, MDIO_SGMII_LINK_TMR_H, LINK_TMR_H_BASEX);
-	phy_write(memac->pcsphy, MDIO_SGMII_LINK_TMR_L, LINK_TMR_L_BASEX);
-
-	/* Restart AN */
-	tmp_reg16 = SGMII_CR_DEF_VAL | SGMII_CR_RESTART_AN;
-	phy_write(memac->pcsphy, 0x0, tmp_reg16);
-}
-
 static int check_init_parameters(struct fman_mac *memac)
 {
-	if (memac->addr == 0) {
-		pr_err("Ethernet MAC must have a valid MAC address\n");
-		return -EINVAL;
-	}
 	if (!memac->exception_cb) {
 		pr_err("Uninitialized exception handler\n");
 		return -EINVAL;
@@ -702,221 +523,281 @@ static void free_init_resources(struct fman_mac *memac)
 	memac->unicast_addr_hash = NULL;
 }
 
-static bool is_init_done(struct memac_cfg *memac_drv_params)
+static int memac_enable(struct fman_mac *memac)
 {
-	/* Checks if mEMAC driver parameters were initialized */
-	if (!memac_drv_params)
-		return true;
+	int ret;
+
+	ret = phy_init(memac->serdes);
+	if (ret) {
+		dev_err(memac->dev_id->dev,
+			"could not initialize serdes: %pe\n", ERR_PTR(ret));
+		return ret;
+	}
 
-	return false;
+	ret = phy_power_on(memac->serdes);
+	if (ret) {
+		dev_err(memac->dev_id->dev,
+			"could not power on serdes: %pe\n", ERR_PTR(ret));
+		phy_exit(memac->serdes);
+	}
+
+	return ret;
+}
+
+static void memac_disable(struct fman_mac *memac)
+{
+	phy_power_off(memac->serdes);
+	phy_exit(memac->serdes);
 }
 
-int memac_enable(struct fman_mac *memac, enum comm_mode mode)
+static int memac_set_promiscuous(struct fman_mac *memac, bool new_val)
 {
 	struct memac_regs __iomem *regs = memac->regs;
 	u32 tmp;
 
-	if (!is_init_done(memac->memac_drv_param))
-		return -EINVAL;
-
 	tmp = ioread32be(&regs->command_config);
-	if (mode & COMM_MODE_RX)
-		tmp |= CMD_CFG_RX_EN;
-	if (mode & COMM_MODE_TX)
-		tmp |= CMD_CFG_TX_EN;
+	if (new_val)
+		tmp |= CMD_CFG_PROMIS_EN;
+	else
+		tmp &= ~CMD_CFG_PROMIS_EN;
 
 	iowrite32be(tmp, &regs->command_config);
 
 	return 0;
 }
 
-int memac_disable(struct fman_mac *memac, enum comm_mode mode)
+static int memac_set_tx_pause_frames(struct fman_mac *memac, u8 priority,
+				     u16 pause_time, u16 thresh_time)
 {
 	struct memac_regs __iomem *regs = memac->regs;
 	u32 tmp;
 
-	if (!is_init_done(memac->memac_drv_param))
-		return -EINVAL;
+	tmp = ioread32be(&regs->tx_fifo_sections);
+
+	GET_TX_EMPTY_DEFAULT_VALUE(tmp);
+	iowrite32be(tmp, &regs->tx_fifo_sections);
 
 	tmp = ioread32be(&regs->command_config);
-	if (mode & COMM_MODE_RX)
-		tmp &= ~CMD_CFG_RX_EN;
-	if (mode & COMM_MODE_TX)
-		tmp &= ~CMD_CFG_TX_EN;
+	tmp &= ~CMD_CFG_PFC_MODE;
 
 	iowrite32be(tmp, &regs->command_config);
 
+	tmp = ioread32be(&regs->pause_quanta[priority / 2]);
+	if (priority % 2)
+		tmp &= CLXY_PAUSE_QUANTA_CLX_PQNT;
+	else
+		tmp &= CLXY_PAUSE_QUANTA_CLY_PQNT;
+	tmp |= ((u32)pause_time << (16 * (priority % 2)));
+	iowrite32be(tmp, &regs->pause_quanta[priority / 2]);
+
+	tmp = ioread32be(&regs->pause_thresh[priority / 2]);
+	if (priority % 2)
+		tmp &= CLXY_PAUSE_THRESH_CLX_QTH;
+	else
+		tmp &= CLXY_PAUSE_THRESH_CLY_QTH;
+	tmp |= ((u32)thresh_time << (16 * (priority % 2)));
+	iowrite32be(tmp, &regs->pause_thresh[priority / 2]);
+
 	return 0;
 }
 
-int memac_set_promiscuous(struct fman_mac *memac, bool new_val)
+static int memac_accept_rx_pause_frames(struct fman_mac *memac, bool en)
 {
 	struct memac_regs __iomem *regs = memac->regs;
 	u32 tmp;
 
-	if (!is_init_done(memac->memac_drv_param))
-		return -EINVAL;
-
 	tmp = ioread32be(&regs->command_config);
-	if (new_val)
-		tmp |= CMD_CFG_PROMIS_EN;
+	if (en)
+		tmp &= ~CMD_CFG_PAUSE_IGNORE;
 	else
-		tmp &= ~CMD_CFG_PROMIS_EN;
+		tmp |= CMD_CFG_PAUSE_IGNORE;
 
 	iowrite32be(tmp, &regs->command_config);
 
 	return 0;
 }
 
-int memac_adjust_link(struct fman_mac *memac, u16 speed)
+static unsigned long memac_get_caps(struct phylink_config *config,
+				    phy_interface_t interface)
 {
-	struct memac_regs __iomem *regs = memac->regs;
-	u32 tmp;
+	struct fman_mac *memac = fman_config_to_mac(config)->fman_mac;
+	unsigned long caps = config->mac_capabilities;
 
-	if (!is_init_done(memac->memac_drv_param))
-		return -EINVAL;
-
-	tmp = ioread32be(&regs->if_mode);
+	if (phy_interface_mode_is_rgmii(interface) &&
+	    memac->rgmii_no_half_duplex)
+		caps &= ~(MAC_10HD | MAC_100HD);
 
-	/* Set full duplex */
-	tmp &= ~IF_MODE_HD;
-
-	if (memac->phy_if == PHY_INTERFACE_MODE_RGMII) {
-		/* Configure RGMII in manual mode */
-		tmp &= ~IF_MODE_RGMII_AUTO;
-		tmp &= ~IF_MODE_RGMII_SP_MASK;
-		/* Full duplex */
-		tmp |= IF_MODE_RGMII_FD;
+	return caps;
+}
 
-		switch (speed) {
-		case SPEED_1000:
-			tmp |= IF_MODE_RGMII_1000;
-			break;
-		case SPEED_100:
-			tmp |= IF_MODE_RGMII_100;
-			break;
-		case SPEED_10:
-			tmp |= IF_MODE_RGMII_10;
-			break;
-		default:
-			break;
-		}
+/**
+ * memac_if_mode() - Convert an interface mode into an IF_MODE config
+ * @interface: A phy interface mode
+ *
+ * Return: A configuration word, suitable for programming into the lower bits
+ *         of %IF_MODE.
+ */
+static u32 memac_if_mode(phy_interface_t interface)
+{
+	switch (interface) {
+	case PHY_INTERFACE_MODE_MII:
+		return IF_MODE_MII;
+	case PHY_INTERFACE_MODE_RGMII:
+	case PHY_INTERFACE_MODE_RGMII_ID:
+	case PHY_INTERFACE_MODE_RGMII_RXID:
+	case PHY_INTERFACE_MODE_RGMII_TXID:
+		return IF_MODE_GMII | IF_MODE_RGMII;
+	case PHY_INTERFACE_MODE_SGMII:
+	case PHY_INTERFACE_MODE_1000BASEX:
+	case PHY_INTERFACE_MODE_QSGMII:
+		return IF_MODE_GMII;
+	case PHY_INTERFACE_MODE_10GBASER:
+		return IF_MODE_10G;
+	default:
+		WARN_ON_ONCE(1);
+		return 0;
 	}
-
-	iowrite32be(tmp, &regs->if_mode);
-
-	return 0;
 }
 
-int memac_cfg_max_frame_len(struct fman_mac *memac, u16 new_val)
+static struct phylink_pcs *memac_select_pcs(struct phylink_config *config,
+					    phy_interface_t iface)
 {
-	if (is_init_done(memac->memac_drv_param))
-		return -EINVAL;
-
-	memac->memac_drv_param->max_frame_length = new_val;
-
-	return 0;
+	struct fman_mac *memac = fman_config_to_mac(config)->fman_mac;
+
+	switch (iface) {
+	case PHY_INTERFACE_MODE_SGMII:
+	case PHY_INTERFACE_MODE_1000BASEX:
+		return memac->sgmii_pcs;
+	case PHY_INTERFACE_MODE_QSGMII:
+		return memac->qsgmii_pcs;
+	case PHY_INTERFACE_MODE_10GBASER:
+		return memac->xfi_pcs;
+	default:
+		return NULL;
+	}
 }
 
-int memac_cfg_reset_on_init(struct fman_mac *memac, bool enable)
+static int memac_prepare(struct phylink_config *config, unsigned int mode,
+			 phy_interface_t iface)
 {
-	if (is_init_done(memac->memac_drv_param))
-		return -EINVAL;
-
-	memac->memac_drv_param->reset_on_init = enable;
-
-	return 0;
+	struct fman_mac *memac = fman_config_to_mac(config)->fman_mac;
+
+	switch (iface) {
+	case PHY_INTERFACE_MODE_SGMII:
+	case PHY_INTERFACE_MODE_1000BASEX:
+	case PHY_INTERFACE_MODE_QSGMII:
+	case PHY_INTERFACE_MODE_10GBASER:
+		return phy_set_mode_ext(memac->serdes, PHY_MODE_ETHERNET,
+					iface);
+	default:
+		return 0;
+	}
 }
 
-int memac_cfg_fixed_link(struct fman_mac *memac,
-			 struct fixed_phy_status *fixed_link)
+static void memac_mac_config(struct phylink_config *config, unsigned int mode,
+			     const struct phylink_link_state *state)
 {
-	if (is_init_done(memac->memac_drv_param))
-		return -EINVAL;
-
-	memac->memac_drv_param->fixed_link = fixed_link;
-
-	return 0;
+	struct mac_device *mac_dev = fman_config_to_mac(config);
+	struct memac_regs __iomem *regs = mac_dev->fman_mac->regs;
+	u32 tmp = ioread32be(&regs->if_mode);
+
+	tmp &= ~(IF_MODE_MASK | IF_MODE_RGMII);
+	tmp |= memac_if_mode(state->interface);
+	if (phylink_autoneg_inband(mode))
+		tmp |= IF_MODE_RGMII_AUTO;
+	iowrite32be(tmp, &regs->if_mode);
 }
 
-int memac_set_tx_pause_frames(struct fman_mac *memac, u8 priority,
-			      u16 pause_time, u16 thresh_time)
+static void memac_link_up(struct phylink_config *config, struct phy_device *phy,
+			  unsigned int mode, phy_interface_t interface,
+			  int speed, int duplex, bool tx_pause, bool rx_pause)
 {
+	struct mac_device *mac_dev = fman_config_to_mac(config);
+	struct fman_mac *memac = mac_dev->fman_mac;
 	struct memac_regs __iomem *regs = memac->regs;
-	u32 tmp;
+	u32 tmp = memac_if_mode(interface);
+	u16 pause_time = tx_pause ? FSL_FM_PAUSE_TIME_ENABLE :
+			 FSL_FM_PAUSE_TIME_DISABLE;
 
-	if (!is_init_done(memac->memac_drv_param))
-		return -EINVAL;
+	memac_set_tx_pause_frames(memac, 0, pause_time, 0);
+	memac_accept_rx_pause_frames(memac, rx_pause);
 
-	tmp = ioread32be(&regs->tx_fifo_sections);
+	if (duplex == DUPLEX_HALF)
+		tmp |= IF_MODE_HD;
 
-	GET_TX_EMPTY_DEFAULT_VALUE(tmp);
-	iowrite32be(tmp, &regs->tx_fifo_sections);
-
-	tmp = ioread32be(&regs->command_config);
-	tmp &= ~CMD_CFG_PFC_MODE;
-	priority = 0;
+	switch (speed) {
+	case SPEED_1000:
+		tmp |= IF_MODE_RGMII_1000;
+		break;
+	case SPEED_100:
+		tmp |= IF_MODE_RGMII_100;
+		break;
+	case SPEED_10:
+		tmp |= IF_MODE_RGMII_10;
+		break;
+	}
+	iowrite32be(tmp, &regs->if_mode);
 
-	iowrite32be(tmp, &regs->command_config);
+	/* TODO: EEE? */
 
-	tmp = ioread32be(&regs->pause_quanta[priority / 2]);
-	if (priority % 2)
-		tmp &= CLXY_PAUSE_QUANTA_CLX_PQNT;
-	else
-		tmp &= CLXY_PAUSE_QUANTA_CLY_PQNT;
-	tmp |= ((u32)pause_time << (16 * (priority % 2)));
-	iowrite32be(tmp, &regs->pause_quanta[priority / 2]);
+	if (speed == SPEED_10000) {
+		if (memac->fm_rev_info.major == 6 &&
+		    memac->fm_rev_info.minor == 4)
+			tmp = TX_FIFO_SECTIONS_TX_AVAIL_SLOW_10G;
+		else
+			tmp = TX_FIFO_SECTIONS_TX_AVAIL_10G;
+		tmp |= TX_FIFO_SECTIONS_TX_EMPTY_DEFAULT_10G;
+	} else {
+		tmp = TX_FIFO_SECTIONS_TX_AVAIL_1G |
+		      TX_FIFO_SECTIONS_TX_EMPTY_DEFAULT_1G;
+	}
+	iowrite32be(tmp, &regs->tx_fifo_sections);
 
-	tmp = ioread32be(&regs->pause_thresh[priority / 2]);
-	if (priority % 2)
-		tmp &= CLXY_PAUSE_THRESH_CLX_QTH;
-	else
-		tmp &= CLXY_PAUSE_THRESH_CLY_QTH;
-	tmp |= ((u32)thresh_time << (16 * (priority % 2)));
-	iowrite32be(tmp, &regs->pause_thresh[priority / 2]);
+	mac_dev->update_speed(mac_dev, speed);
 
-	return 0;
+	tmp = ioread32be(&regs->command_config);
+	tmp |= CMD_CFG_RX_EN | CMD_CFG_TX_EN;
+	iowrite32be(tmp, &regs->command_config);
 }
 
-int memac_accept_rx_pause_frames(struct fman_mac *memac, bool en)
+static void memac_link_down(struct phylink_config *config, unsigned int mode,
+			    phy_interface_t interface)
 {
+	struct fman_mac *memac = fman_config_to_mac(config)->fman_mac;
 	struct memac_regs __iomem *regs = memac->regs;
 	u32 tmp;
 
-	if (!is_init_done(memac->memac_drv_param))
-		return -EINVAL;
-
+	/* TODO: graceful */
 	tmp = ioread32be(&regs->command_config);
-	if (en)
-		tmp &= ~CMD_CFG_PAUSE_IGNORE;
-	else
-		tmp |= CMD_CFG_PAUSE_IGNORE;
-
+	tmp &= ~(CMD_CFG_RX_EN | CMD_CFG_TX_EN);
 	iowrite32be(tmp, &regs->command_config);
-
-	return 0;
 }
 
-int memac_modify_mac_address(struct fman_mac *memac, enet_addr_t *enet_addr)
-{
-	if (!is_init_done(memac->memac_drv_param))
-		return -EINVAL;
+static const struct phylink_mac_ops memac_mac_ops = {
+	.mac_get_caps = memac_get_caps,
+	.mac_select_pcs = memac_select_pcs,
+	.mac_prepare = memac_prepare,
+	.mac_config = memac_mac_config,
+	.mac_link_up = memac_link_up,
+	.mac_link_down = memac_link_down,
+};
 
-	add_addr_in_paddr(memac->regs, (u8 *)(*enet_addr), 0);
+static int memac_modify_mac_address(struct fman_mac *memac,
+				    const enet_addr_t *enet_addr)
+{
+	add_addr_in_paddr(memac->regs, (const u8 *)(*enet_addr), 0);
 
 	return 0;
 }
 
-int memac_add_hash_mac_address(struct fman_mac *memac, enet_addr_t *eth_addr)
+static int memac_add_hash_mac_address(struct fman_mac *memac,
+				      enet_addr_t *eth_addr)
 {
 	struct memac_regs __iomem *regs = memac->regs;
 	struct eth_hash_entry *hash_entry;
 	u32 hash;
 	u64 addr;
 
-	if (!is_init_done(memac->memac_drv_param))
-		return -EINVAL;
-
 	addr = ENET_ADDR_TO_UINT64(*eth_addr);
 
 	if (!(addr & GROUP_ADDRESS)) {
@@ -927,7 +808,7 @@ int memac_add_hash_mac_address(struct fman_mac *memac, enet_addr_t *eth_addr)
 	hash = get_mac_addr_hash_code(addr) & HASH_CTRL_ADDR_MASK;
 
 	/* Create element to be added to the driver hash table */
-	hash_entry = kmalloc(sizeof(*hash_entry), GFP_KERNEL);
+	hash_entry = kmalloc(sizeof(*hash_entry), GFP_ATOMIC);
 	if (!hash_entry)
 		return -ENOMEM;
 	hash_entry->addr = addr;
@@ -940,7 +821,33 @@ int memac_add_hash_mac_address(struct fman_mac *memac, enet_addr_t *eth_addr)
 	return 0;
 }
 
-int memac_del_hash_mac_address(struct fman_mac *memac, enet_addr_t *eth_addr)
+static int memac_set_allmulti(struct fman_mac *memac, bool enable)
+{
+	u32 entry;
+	struct memac_regs __iomem *regs = memac->regs;
+
+	if (enable) {
+		for (entry = 0; entry < HASH_TABLE_SIZE; entry++)
+			iowrite32be(entry | HASH_CTRL_MCAST_EN,
+				    &regs->hashtable_ctrl);
+	} else {
+		for (entry = 0; entry < HASH_TABLE_SIZE; entry++)
+			iowrite32be(entry & ~HASH_CTRL_MCAST_EN,
+				    &regs->hashtable_ctrl);
+	}
+
+	memac->allmulti_enabled = enable;
+
+	return 0;
+}
+
+static int memac_set_tstamp(struct fman_mac *memac, bool enable)
+{
+	return 0; /* Always enabled. */
+}
+
+static int memac_del_hash_mac_address(struct fman_mac *memac,
+				      enet_addr_t *eth_addr)
 {
 	struct memac_regs __iomem *regs = memac->regs;
 	struct eth_hash_entry *hash_entry = NULL;
@@ -948,35 +855,33 @@ int memac_del_hash_mac_address(struct fman_mac *memac, enet_addr_t *eth_addr)
 	u32 hash;
 	u64 addr;
 
-	if (!is_init_done(memac->memac_drv_param))
-		return -EINVAL;
-
 	addr = ENET_ADDR_TO_UINT64(*eth_addr);
 
 	hash = get_mac_addr_hash_code(addr) & HASH_CTRL_ADDR_MASK;
 
 	list_for_each(pos, &memac->multicast_addr_hash->lsts[hash]) {
 		hash_entry = ETH_HASH_ENTRY_OBJ(pos);
-		if (hash_entry->addr == addr) {
+		if (hash_entry && hash_entry->addr == addr) {
 			list_del_init(&hash_entry->node);
 			kfree(hash_entry);
 			break;
 		}
 	}
-	if (list_empty(&memac->multicast_addr_hash->lsts[hash]))
-		iowrite32be(hash & ~HASH_CTRL_MCAST_EN, &regs->hashtable_ctrl);
+
+	if (!memac->allmulti_enabled) {
+		if (list_empty(&memac->multicast_addr_hash->lsts[hash]))
+			iowrite32be(hash & ~HASH_CTRL_MCAST_EN,
+				    &regs->hashtable_ctrl);
+	}
 
 	return 0;
 }
 
-int memac_set_exception(struct fman_mac *memac,
-			enum fman_mac_exceptions exception, bool enable)
+static int memac_set_exception(struct fman_mac *memac,
+			       enum fman_mac_exceptions exception, bool enable)
 {
 	u32 bit_mask = 0;
 
-	if (!is_init_done(memac->memac_drv_param))
-		return -EINVAL;
-
 	bit_mask = get_exception_flag(exception);
 	if (bit_mask) {
 		if (enable)
@@ -992,28 +897,19 @@ int memac_set_exception(struct fman_mac *memac,
 	return 0;
 }
 
-int memac_init(struct fman_mac *memac)
+static int memac_init(struct fman_mac *memac)
 {
 	struct memac_cfg *memac_drv_param;
-	u8 i;
 	enet_addr_t eth_addr;
-	bool slow_10g_if = false;
-	struct fixed_phy_status *fixed_link;
 	int err;
 	u32 reg32 = 0;
 
-	if (is_init_done(memac->memac_drv_param))
-		return -EINVAL;
-
 	err = check_init_parameters(memac);
 	if (err)
 		return err;
 
 	memac_drv_param = memac->memac_drv_param;
 
-	if (memac->fm_rev_info.major == 6 && memac->fm_rev_info.minor == 4)
-		slow_10g_if = true;
-
 	/* First, reset the MAC if desired. */
 	if (memac_drv_param->reset_on_init) {
 		err = reset(memac->regs);
@@ -1024,13 +920,12 @@ int memac_init(struct fman_mac *memac)
 	}
 
 	/* MAC Address */
-	MAKE_ENET_ADDR_FROM_UINT64(memac->addr, eth_addr);
-	add_addr_in_paddr(memac->regs, (u8 *)eth_addr, 0);
-
-	fixed_link = memac_drv_param->fixed_link;
+	if (memac->addr != 0) {
+		MAKE_ENET_ADDR_FROM_UINT64(memac->addr, eth_addr);
+		add_addr_in_paddr(memac->regs, (const u8 *)eth_addr, 0);
+	}
 
-	init(memac->regs, memac->memac_drv_param, memac->phy_if,
-	     memac->max_speed, slow_10g_if, memac->exceptions);
+	init(memac->regs, memac->memac_drv_param, memac->exceptions);
 
 	/* FM_RX_FIFO_CORRUPT_ERRATA_10GMAC_A006320 errata workaround
 	 * Exists only in FMan 6.0 and 6.3.
@@ -1046,33 +941,6 @@ int memac_init(struct fman_mac *memac)
 		iowrite32be(reg32, &memac->regs->command_config);
 	}
 
-	if (memac->phy_if == PHY_INTERFACE_MODE_SGMII) {
-		/* Configure internal SGMII PHY */
-		if (memac->basex_if)
-			setup_sgmii_internal_phy_base_x(memac);
-		else
-			setup_sgmii_internal_phy(memac, fixed_link);
-	} else if (memac->phy_if == PHY_INTERFACE_MODE_QSGMII) {
-		/* Configure 4 internal SGMII PHYs */
-		for (i = 0; i < 4; i++) {
-			u8 qsmgii_phy_addr, phy_addr;
-			/* QSGMII PHY address occupies 3 upper bits of 5-bit
-			 * phy_address; the lower 2 bits are used to extend
-			 * register address space and access each one of 4
-			 * ports inside QSGMII.
-			 */
-			phy_addr = memac->pcsphy->mdio.addr;
-			qsmgii_phy_addr = (u8)((phy_addr << 2) | i);
-			memac->pcsphy->mdio.addr = qsmgii_phy_addr;
-			if (memac->basex_if)
-				setup_sgmii_internal_phy_base_x(memac);
-			else
-				setup_sgmii_internal_phy(memac, fixed_link);
-
-			memac->pcsphy->mdio.addr = phy_addr;
-		}
-	}
-
 	/* Max Frame Length */
 	err = fman_set_mac_max_frame(memac->fm, memac->mac_id,
 				     memac_drv_param->max_frame_length);
@@ -1101,32 +969,36 @@ int memac_init(struct fman_mac *memac)
 	fman_register_intr(memac->fm, FMAN_MOD_MAC, memac->mac_id,
 			   FMAN_INTR_TYPE_NORMAL, memac_exception, memac);
 
-	kfree(memac_drv_param);
-	memac->memac_drv_param = NULL;
-
 	return 0;
 }
 
-int memac_free(struct fman_mac *memac)
+static void pcs_put(struct phylink_pcs *pcs)
 {
-	free_init_resources(memac);
+	if (IS_ERR_OR_NULL(pcs))
+		return;
+
+	lynx_pcs_destroy(pcs);
+}
 
-	if (memac->pcsphy)
-		put_device(&memac->pcsphy->mdio.dev);
+static int memac_free(struct fman_mac *memac)
+{
+	free_init_resources(memac);
 
+	pcs_put(memac->sgmii_pcs);
+	pcs_put(memac->qsgmii_pcs);
+	pcs_put(memac->xfi_pcs);
 	kfree(memac->memac_drv_param);
 	kfree(memac);
 
 	return 0;
 }
 
-struct fman_mac *memac_config(struct fman_mac_params *params)
+static struct fman_mac *memac_config(struct mac_device *mac_dev,
+				     struct fman_mac_params *params)
 {
 	struct fman_mac *memac;
 	struct memac_cfg *memac_drv_param;
-	void __iomem *base_addr;
 
-	base_addr = params->base_addr;
 	/* allocate memory for the m_emac data structure */
 	memac = kzalloc(sizeof(*memac), GFP_KERNEL);
 	if (!memac)
@@ -1144,38 +1016,229 @@ struct fman_mac *memac_config(struct fman_mac_params *params)
 
 	set_dflts(memac_drv_param);
 
-	memac->addr = ENET_ADDR_TO_UINT64(params->addr);
+	memac->addr = ENET_ADDR_TO_UINT64(mac_dev->addr);
 
-	memac->regs = base_addr;
-	memac->max_speed = params->max_speed;
-	memac->phy_if = params->phy_if;
+	memac->regs = mac_dev->vaddr;
 	memac->mac_id = params->mac_id;
 	memac->exceptions = (MEMAC_IMASK_TSECC_ER | MEMAC_IMASK_TECC_ER |
 			     MEMAC_IMASK_RECC_ER | MEMAC_IMASK_MGI);
 	memac->exception_cb = params->exception_cb;
 	memac->event_cb = params->event_cb;
-	memac->dev_id = params->dev_id;
+	memac->dev_id = mac_dev;
 	memac->fm = params->fm;
-	memac->basex_if = params->basex_if;
 
 	/* Save FMan revision */
 	fman_get_revision(memac->fm, &memac->fm_rev_info);
 
-	if (memac->phy_if == PHY_INTERFACE_MODE_SGMII ||
-	    memac->phy_if == PHY_INTERFACE_MODE_QSGMII) {
-		if (!params->internal_phy_node) {
-			pr_err("PCS PHY node is not available\n");
-			memac_free(memac);
-			return NULL;
+	return memac;
+}
+
+static struct phylink_pcs *memac_pcs_create(struct device_node *mac_node,
+					    int index)
+{
+	struct device_node *node;
+	struct phylink_pcs *pcs;
+
+	node = of_parse_phandle(mac_node, "pcsphy-handle", index);
+	if (!node)
+		return ERR_PTR(-ENODEV);
+
+	pcs = lynx_pcs_create_fwnode(of_fwnode_handle(node));
+	of_node_put(node);
+
+	return pcs;
+}
+
+static bool memac_supports(struct mac_device *mac_dev, phy_interface_t iface)
+{
+	/* If there's no serdes device, assume that it's been configured for
+	 * whatever the default interface mode is.
+	 */
+	if (!mac_dev->fman_mac->serdes)
+		return mac_dev->phy_if == iface;
+	/* Otherwise, ask the serdes */
+	return !phy_validate(mac_dev->fman_mac->serdes, PHY_MODE_ETHERNET,
+			     iface, NULL);
+}
+
+int memac_initialization(struct mac_device *mac_dev,
+			 struct device_node *mac_node,
+			 struct fman_mac_params *params)
+{
+	int			 err;
+	struct phylink_pcs	*pcs;
+	struct fman_mac		*memac;
+	unsigned long		 capabilities;
+	unsigned long		*supported;
+
+	/* The internal connection to the serdes is XGMII, but this isn't
+	 * really correct for the phy mode (which is the external connection).
+	 * However, this is how all older device trees say that they want
+	 * 10GBASE-R (aka XFI), so just convert it for them.
+	 */
+	if (mac_dev->phy_if == PHY_INTERFACE_MODE_XGMII)
+		mac_dev->phy_if = PHY_INTERFACE_MODE_10GBASER;
+
+	mac_dev->phylink_ops		= &memac_mac_ops;
+	mac_dev->set_promisc		= memac_set_promiscuous;
+	mac_dev->change_addr		= memac_modify_mac_address;
+	mac_dev->add_hash_mac_addr	= memac_add_hash_mac_address;
+	mac_dev->remove_hash_mac_addr	= memac_del_hash_mac_address;
+	mac_dev->set_exception		= memac_set_exception;
+	mac_dev->set_allmulti		= memac_set_allmulti;
+	mac_dev->set_tstamp		= memac_set_tstamp;
+	mac_dev->enable			= memac_enable;
+	mac_dev->disable		= memac_disable;
+
+	mac_dev->fman_mac = memac_config(mac_dev, params);
+	if (!mac_dev->fman_mac)
+		return -EINVAL;
+
+	memac = mac_dev->fman_mac;
+	memac->memac_drv_param->max_frame_length = fman_get_max_frm();
+	memac->memac_drv_param->reset_on_init = true;
+
+	err = of_property_match_string(mac_node, "pcs-handle-names", "xfi");
+	if (err >= 0) {
+		memac->xfi_pcs = memac_pcs_create(mac_node, err);
+		if (IS_ERR(memac->xfi_pcs)) {
+			err = PTR_ERR(memac->xfi_pcs);
+			dev_err_probe(mac_dev->dev, err, "missing xfi pcs\n");
+			goto _return_fm_mac_free;
 		}
+	} else if (err != -EINVAL && err != -ENODATA) {
+		goto _return_fm_mac_free;
+	}
 
-		memac->pcsphy = of_phy_find_device(params->internal_phy_node);
-		if (!memac->pcsphy) {
-			pr_err("of_phy_find_device (PCS PHY) failed\n");
-			memac_free(memac);
-			return NULL;
+	err = of_property_match_string(mac_node, "pcs-handle-names", "qsgmii");
+	if (err >= 0) {
+		memac->qsgmii_pcs = memac_pcs_create(mac_node, err);
+		if (IS_ERR(memac->qsgmii_pcs)) {
+			err = PTR_ERR(memac->qsgmii_pcs);
+			dev_err_probe(mac_dev->dev, err,
+				      "missing qsgmii pcs\n");
+			goto _return_fm_mac_free;
 		}
+	} else if (err != -EINVAL && err != -ENODATA) {
+		goto _return_fm_mac_free;
 	}
 
-	return memac;
+	/* For compatibility, if pcs-handle-names is missing, we assume this
+	 * phy is the first one in pcsphy-handle
+	 */
+	err = of_property_match_string(mac_node, "pcs-handle-names", "sgmii");
+	if (err == -EINVAL || err == -ENODATA)
+		pcs = memac_pcs_create(mac_node, 0);
+	else if (err < 0)
+		goto _return_fm_mac_free;
+	else
+		pcs = memac_pcs_create(mac_node, err);
+
+	if (IS_ERR(pcs)) {
+		err = PTR_ERR(pcs);
+		dev_err_probe(mac_dev->dev, err, "missing pcs\n");
+		goto _return_fm_mac_free;
+	}
+
+	/* If err is set here, it means that pcs-handle-names was missing above
+	 * (and therefore that xfi_pcs cannot be set). If we are defaulting to
+	 * XGMII, assume this is for XFI. Otherwise, assume it is for SGMII.
+	 */
+	if (err && mac_dev->phy_if == PHY_INTERFACE_MODE_10GBASER)
+		memac->xfi_pcs = pcs;
+	else
+		memac->sgmii_pcs = pcs;
+
+	memac->serdes = devm_of_phy_optional_get(mac_dev->dev, mac_node,
+						 "serdes");
+	if (!memac->serdes) {
+		dev_dbg(mac_dev->dev, "could not get (optional) serdes\n");
+	} else if (IS_ERR(memac->serdes)) {
+		err = PTR_ERR(memac->serdes);
+		goto _return_fm_mac_free;
+	}
+
+	/* TODO: The following interface modes are supported by (some) hardware
+	 * but not by this driver:
+	 * - 1000BASE-KX
+	 * - 10GBASE-KR
+	 * - XAUI/HiGig
+	 */
+	supported = mac_dev->phylink_config.supported_interfaces;
+
+	/* Note that half duplex is only supported on 10/100M interfaces. */
+
+	if (memac->sgmii_pcs &&
+	    (memac_supports(mac_dev, PHY_INTERFACE_MODE_SGMII) ||
+	     memac_supports(mac_dev, PHY_INTERFACE_MODE_1000BASEX))) {
+		__set_bit(PHY_INTERFACE_MODE_SGMII, supported);
+		__set_bit(PHY_INTERFACE_MODE_1000BASEX, supported);
+	}
+
+	if (memac->sgmii_pcs &&
+	    memac_supports(mac_dev, PHY_INTERFACE_MODE_2500BASEX))
+		__set_bit(PHY_INTERFACE_MODE_2500BASEX, supported);
+
+	if (memac->qsgmii_pcs &&
+	    memac_supports(mac_dev, PHY_INTERFACE_MODE_QSGMII))
+		__set_bit(PHY_INTERFACE_MODE_QSGMII, supported);
+	else if (mac_dev->phy_if == PHY_INTERFACE_MODE_QSGMII)
+		dev_warn(mac_dev->dev, "no QSGMII pcs specified\n");
+
+	if (memac->xfi_pcs &&
+	    memac_supports(mac_dev, PHY_INTERFACE_MODE_10GBASER)) {
+		__set_bit(PHY_INTERFACE_MODE_10GBASER, supported);
+	} else {
+		/* From what I can tell, no 10g macs support RGMII. */
+		phy_interface_set_rgmii(supported);
+		__set_bit(PHY_INTERFACE_MODE_MII, supported);
+	}
+
+	capabilities = MAC_SYM_PAUSE | MAC_ASYM_PAUSE | MAC_10 | MAC_100;
+	capabilities |= MAC_1000FD | MAC_2500FD | MAC_10000FD;
+
+	/* These SoCs don't support half duplex at all; there's no different
+	 * FMan version or compatible, so we just have to check the machine
+	 * compatible instead
+	 */
+	if (of_machine_is_compatible("fsl,ls1043a") ||
+	    of_machine_is_compatible("fsl,ls1046a") ||
+	    of_machine_is_compatible("fsl,B4QDS"))
+		capabilities &= ~(MAC_10HD | MAC_100HD);
+
+	mac_dev->phylink_config.mac_capabilities = capabilities;
+
+	/* The T2080 and T4240 don't support half duplex RGMII. There is no
+	 * other way to identify these SoCs, so just use the machine
+	 * compatible.
+	 */
+	if (of_machine_is_compatible("fsl,T2080QDS") ||
+	    of_machine_is_compatible("fsl,T2080RDB") ||
+	    of_machine_is_compatible("fsl,T2081QDS") ||
+	    of_machine_is_compatible("fsl,T4240QDS") ||
+	    of_machine_is_compatible("fsl,T4240RDB"))
+		memac->rgmii_no_half_duplex = true;
+
+	/* Most boards should use MLO_AN_INBAND, but existing boards don't have
+	 * a managed property. Default to MLO_AN_INBAND rather than MLO_AN_PHY.
+	 * Phylink will allow this to be overriden by a fixed link. We need to
+	 * be careful and not enable this if we are using MII or RGMII, since
+	 * those configurations modes don't use in-band autonegotiation.
+	 */
+	if (!of_property_present(mac_node, "managed") &&
+	    mac_dev->phy_if != PHY_INTERFACE_MODE_MII &&
+	    !phy_interface_mode_is_rgmii(mac_dev->phy_if))
+		mac_dev->phylink_config.default_an_inband = true;
+
+	err = memac_init(mac_dev->fman_mac);
+	if (err < 0)
+		goto _return_fm_mac_free;
+
+	dev_info(mac_dev->dev, "FMan MEMAC\n");
+
+	return 0;
+
+_return_fm_mac_free:
+	memac_free(mac_dev->fman_mac);
+	return err;
 }
diff --git a/drivers/net/ethernet/freescale/fman/fman_memac.h b/drivers/net/ethernet/freescale/fman/fman_memac.h
index c4a66469a907..5a3a14f9684f 100644
--- a/drivers/net/ethernet/freescale/fman/fman_memac.h
+++ b/drivers/net/ethernet/freescale/fman/fman_memac.h
@@ -1,33 +1,6 @@
+/* SPDX-License-Identifier: BSD-3-Clause OR GPL-2.0-or-later */
 /*
- * Copyright 2008-2015 Freescale Semiconductor Inc.
- *
- * Redistribution and use in source and binary forms, with or without
- * modification, are permitted provided that the following conditions are met:
- *     * Redistributions of source code must retain the above copyright
- *       notice, this list of conditions and the following disclaimer.
- *     * Redistributions in binary form must reproduce the above copyright
- *       notice, this list of conditions and the following disclaimer in the
- *       documentation and/or other materials provided with the distribution.
- *     * Neither the name of Freescale Semiconductor nor the
- *       names of its contributors may be used to endorse or promote products
- *       derived from this software without specific prior written permission.
- *
- *
- * ALTERNATIVELY, this software may be distributed under the terms of the
- * GNU General Public License ("GPL") as published by the Free Software
- * Foundation, either version 2 of that License or (at your option) any
- * later version.
- *
- * THIS SOFTWARE IS PROVIDED BY Freescale Semiconductor ``AS IS'' AND ANY
- * EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED
- * WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE
- * DISCLAIMED. IN NO EVENT SHALL Freescale Semiconductor BE LIABLE FOR ANY
- * DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES
- * (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES;
- * LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND
- * ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
- * (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS
- * SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
+ * Copyright 2008 - 2015 Freescale Semiconductor Inc.
  */
 
 #ifndef __MEMAC_H
@@ -38,24 +11,10 @@
 #include <linux/netdevice.h>
 #include <linux/phy_fixed.h>
 
-struct fman_mac *memac_config(struct fman_mac_params *params);
-int memac_set_promiscuous(struct fman_mac *memac, bool new_val);
-int memac_modify_mac_address(struct fman_mac *memac, enet_addr_t *enet_addr);
-int memac_adjust_link(struct fman_mac *memac, u16 speed);
-int memac_cfg_max_frame_len(struct fman_mac *memac, u16 new_val);
-int memac_cfg_reset_on_init(struct fman_mac *memac, bool enable);
-int memac_cfg_fixed_link(struct fman_mac *memac,
-			 struct fixed_phy_status *fixed_link);
-int memac_enable(struct fman_mac *memac, enum comm_mode mode);
-int memac_disable(struct fman_mac *memac, enum comm_mode mode);
-int memac_init(struct fman_mac *memac);
-int memac_free(struct fman_mac *memac);
-int memac_accept_rx_pause_frames(struct fman_mac *memac, bool en);
-int memac_set_tx_pause_frames(struct fman_mac *memac, u8 priority,
-			      u16 pause_time, u16 thresh_time);
-int memac_set_exception(struct fman_mac *memac,
-			enum fman_mac_exceptions exception, bool enable);
-int memac_add_hash_mac_address(struct fman_mac *memac, enet_addr_t *eth_addr);
-int memac_del_hash_mac_address(struct fman_mac *memac, enet_addr_t *eth_addr);
+struct mac_device;
+
+int memac_initialization(struct mac_device *mac_dev,
+			 struct device_node *mac_node,
+			 struct fman_mac_params *params);
 
 #endif /* __MEMAC_H */
diff --git a/drivers/net/ethernet/freescale/fman/fman_muram.c b/drivers/net/ethernet/freescale/fman/fman_muram.c
index 5ec94d243da0..1ed245a2ee01 100644
--- a/drivers/net/ethernet/freescale/fman/fman_muram.c
+++ b/drivers/net/ethernet/freescale/fman/fman_muram.c
@@ -1,33 +1,6 @@
+// SPDX-License-Identifier: BSD-3-Clause OR GPL-2.0-or-later
 /*
- * Copyright 2008-2015 Freescale Semiconductor Inc.
- *
- * Redistribution and use in source and binary forms, with or without
- * modification, are permitted provided that the following conditions are met:
- *     * Redistributions of source code must retain the above copyright
- *       notice, this list of conditions and the following disclaimer.
- *     * Redistributions in binary form must reproduce the above copyright
- *       notice, this list of conditions and the following disclaimer in the
- *       documentation and/or other materials provided with the distribution.
- *     * Neither the name of Freescale Semiconductor nor the
- *       names of its contributors may be used to endorse or promote products
- *       derived from this software without specific prior written permission.
- *
- *
- * ALTERNATIVELY, this software may be distributed under the terms of the
- * GNU General Public License ("GPL") as published by the Free Software
- * Foundation, either version 2 of that License or (at your option) any
- * later version.
- *
- * THIS SOFTWARE IS PROVIDED BY Freescale Semiconductor ``AS IS'' AND ANY
- * EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED
- * WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE
- * DISCLAIMED. IN NO EVENT SHALL Freescale Semiconductor BE LIABLE FOR ANY
- * DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES
- * (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES;
- * LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND
- * ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
- * (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS
- * SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
+ * Copyright 2008 - 2015 Freescale Semiconductor Inc.
  */
 
 #include "fman_muram.h"
@@ -39,7 +12,6 @@
 struct muram_info {
 	struct gen_pool *pool;
 	void __iomem *vbase;
-	size_t size;
 	phys_addr_t pbase;
 };
 
@@ -144,9 +116,9 @@ unsigned long fman_muram_alloc(struct muram_info *muram, size_t size)
 
 /**
  * fman_muram_free_mem
- * muram:	FM-MURAM module pointer.
- * offset:	offset of the memory region to be freed.
- * size:	size of the memory to be freed.
+ * @muram:	FM-MURAM module pointer.
+ * @offset:	offset of the memory region to be freed.
+ * @size:	size of the memory to be freed.
  *
  * Free an allocated memory from FM-MURAM partition.
  */
diff --git a/drivers/net/ethernet/freescale/fman/fman_muram.h b/drivers/net/ethernet/freescale/fman/fman_muram.h
index 453bf849eee1..3643af61bae2 100644
--- a/drivers/net/ethernet/freescale/fman/fman_muram.h
+++ b/drivers/net/ethernet/freescale/fman/fman_muram.h
@@ -1,34 +1,8 @@
+/* SPDX-License-Identifier: BSD-3-Clause OR GPL-2.0-or-later */
 /*
- * Copyright 2008-2015 Freescale Semiconductor Inc.
- *
- * Redistribution and use in source and binary forms, with or without
- * modification, are permitted provided that the following conditions are met:
- *     * Redistributions of source code must retain the above copyright
- *       notice, this list of conditions and the following disclaimer.
- *     * Redistributions in binary form must reproduce the above copyright
- *       notice, this list of conditions and the following disclaimer in the
- *       documentation and/or other materials provided with the distribution.
- *     * Neither the name of Freescale Semiconductor nor the
- *       names of its contributors may be used to endorse or promote products
- *       derived from this software without specific prior written permission.
- *
- *
- * ALTERNATIVELY, this software may be distributed under the terms of the
- * GNU General Public License ("GPL") as published by the Free Software
- * Foundation, either version 2 of that License or (at your option) any
- * later version.
- *
- * THIS SOFTWARE IS PROVIDED BY Freescale Semiconductor ``AS IS'' AND ANY
- * EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED
- * WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE
- * DISCLAIMED. IN NO EVENT SHALL Freescale Semiconductor BE LIABLE FOR ANY
- * DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES
- * (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES;
- * LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND
- * ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
- * (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS
- * SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
+ * Copyright 2008 - 2015 Freescale Semiconductor Inc.
  */
+
 #ifndef __FM_MURAM_EXT
 #define __FM_MURAM_EXT
 
diff --git a/drivers/net/ethernet/freescale/fman/fman_port.c b/drivers/net/ethernet/freescale/fman/fman_port.c
index 57bf44fa16a1..e977389f7088 100644
--- a/drivers/net/ethernet/freescale/fman/fman_port.c
+++ b/drivers/net/ethernet/freescale/fman/fman_port.c
@@ -1,42 +1,12 @@
+// SPDX-License-Identifier: BSD-3-Clause OR GPL-2.0-or-later
 /*
  * Copyright 2008 - 2015 Freescale Semiconductor Inc.
- *
- * Redistribution and use in source and binary forms, with or without
- * modification, are permitted provided that the following conditions are met:
- *     * Redistributions of source code must retain the above copyright
- *       notice, this list of conditions and the following disclaimer.
- *     * Redistributions in binary form must reproduce the above copyright
- *       notice, this list of conditions and the following disclaimer in the
- *       documentation and/or other materials provided with the distribution.
- *     * Neither the name of Freescale Semiconductor nor the
- *       names of its contributors may be used to endorse or promote products
- *       derived from this software without specific prior written permission.
- *
- *
- * ALTERNATIVELY, this software may be distributed under the terms of the
- * GNU General Public License ("GPL") as published by the Free Software
- * Foundation, either version 2 of that License or (at your option) any
- * later version.
- *
- * THIS SOFTWARE IS PROVIDED BY Freescale Semiconductor ``AS IS'' AND ANY
- * EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED
- * WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE
- * DISCLAIMED. IN NO EVENT SHALL Freescale Semiconductor BE LIABLE FOR ANY
- * DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES
- * (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES;
- * LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND
- * ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
- * (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS
- * SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
  */
 
 #define pr_fmt(fmt) KBUILD_MODNAME ": " fmt
 
-#include "fman_port.h"
-#include "fman.h"
-#include "fman_sp.h"
-
 #include <linux/io.h>
+#include <linux/platform_device.h>
 #include <linux/slab.h>
 #include <linux/module.h>
 #include <linux/interrupt.h>
@@ -45,6 +15,11 @@
 #include <linux/delay.h>
 #include <linux/libfdt_env.h>
 
+#include "fman.h"
+#include "fman_port.h"
+#include "fman_sp.h"
+#include "fman_keygen.h"
+
 /* Queue ID */
 #define DFLT_FQ_ID		0x00FFFFFF
 
@@ -184,6 +159,7 @@
 #define NIA_ENG_QMI_ENQ					0x00540000
 #define NIA_ENG_QMI_DEQ					0x00580000
 #define NIA_ENG_HWP					0x00440000
+#define NIA_ENG_HWK					0x00480000
 #define NIA_BMI_AC_ENQ_FRAME				0x00000002
 #define NIA_BMI_AC_TX_RELEASE				0x000002C0
 #define NIA_BMI_AC_RELEASE				0x000000C0
@@ -322,6 +298,10 @@ struct fman_port_qmi_regs {
 #define HWP_HXS_PHE_REPORT 0x00000800
 #define HWP_HXS_PCAC_PSTAT 0x00000100
 #define HWP_HXS_PCAC_PSTOP 0x00000001
+#define HWP_HXS_TCP_OFFSET 0xA
+#define HWP_HXS_UDP_OFFSET 0xB
+#define HWP_HXS_SH_PAD_REM 0x80000000
+
 struct fman_port_hwp_regs {
 	struct {
 		u32 ssa; /* Soft Sequence Attachment */
@@ -394,6 +374,8 @@ struct fman_port_bpools {
 struct fman_port_cfg {
 	u32 dflt_fqid;
 	u32 err_fqid;
+	u32 pcd_base_fqid;
+	u32 pcd_fqs_count;
 	u8 deq_sp;
 	bool deq_high_priority;
 	enum fman_port_deq_type deq_type;
@@ -427,7 +409,6 @@ struct fman_port_cfg {
 
 struct fman_port_rx_pools_params {
 	u8 num_of_pools;
-	u16 second_largest_buf_size;
 	u16 largest_buf_size;
 };
 
@@ -724,6 +705,10 @@ static void init_hwp(struct fman_port *port)
 		iowrite32be(0xffffffff, &regs->pmda[i].lcv);
 	}
 
+	/* Short packet padding removal from checksum calculation */
+	iowrite32be(HWP_HXS_SH_PAD_REM, &regs->pmda[HWP_HXS_TCP_OFFSET].ssa);
+	iowrite32be(HWP_HXS_SH_PAD_REM, &regs->pmda[HWP_HXS_UDP_OFFSET].ssa);
+
 	start_port_hwp(port);
 }
 
@@ -934,8 +919,6 @@ static int set_ext_buffer_pools(struct fman_port *port)
 	port->rx_pools_params.num_of_pools = ext_buf_pools->num_of_pools_used;
 	port->rx_pools_params.largest_buf_size =
 	    sizes_array[ordered_array[ext_buf_pools->num_of_pools_used - 1]];
-	port->rx_pools_params.second_largest_buf_size =
-	    sizes_array[ordered_array[ext_buf_pools->num_of_pools_used - 2]];
 
 	/* FMBM_RMPD reg. - pool depletion */
 	if (buf_pool_depletion->pools_grp_mode_enable) {
@@ -1004,7 +987,7 @@ static int init_low_level_driver(struct fman_port *port)
 		return -ENODEV;
 	}
 
-	/* The code bellow is a trick so the FM will not release the buffer
+	/* The code below is a trick so the FM will not release the buffer
 	 * to BM nor will try to enqueue the frame to QM
 	 */
 	if (port->port_type == FMAN_PORT_TYPE_TX) {
@@ -1271,6 +1254,10 @@ static void set_rx_dflt_cfg(struct fman_port *port,
 		port_params->specific_params.rx_params.err_fqid;
 	port->cfg->dflt_fqid =
 		port_params->specific_params.rx_params.dflt_fqid;
+	port->cfg->pcd_base_fqid =
+		port_params->specific_params.rx_params.pcd_base_fqid;
+	port->cfg->pcd_fqs_count =
+		port_params->specific_params.rx_params.pcd_fqs_count;
 }
 
 static void set_tx_dflt_cfg(struct fman_port *port,
@@ -1331,8 +1318,10 @@ int fman_port_config(struct fman_port *port, struct fman_port_params *params)
 	switch (port->port_type) {
 	case FMAN_PORT_TYPE_RX:
 		set_rx_dflt_cfg(port, params);
+		fallthrough;
 	case FMAN_PORT_TYPE_TX:
 		set_tx_dflt_cfg(port, params, &port->dts_params);
+		fallthrough;
 	default:
 		set_dflt_cfg(port, params);
 	}
@@ -1381,12 +1370,10 @@ int fman_port_config(struct fman_port *port, struct fman_port_params *params)
 		/* FM_WRONG_RESET_VALUES_ERRATA_FMAN_A005127 Errata
 		 * workaround
 		 */
-		if (port->rev_info.major >= 6) {
-			u32 reg;
+		u32 reg;
 
-			reg = 0x00001013;
-			iowrite32be(reg, &port->bmi_regs->tx.fmbm_tfp);
-		}
+		reg = 0x00001013;
+		iowrite32be(reg, &port->bmi_regs->tx.fmbm_tfp);
 	}
 
 	return 0;
@@ -1397,9 +1384,30 @@ err_port_cfg:
 }
 EXPORT_SYMBOL(fman_port_config);
 
+/*
+ * fman_port_use_kg_hash
+ * @port: A pointer to a FM Port module.
+ * @enable: enable or disable
+ *
+ * Sets the HW KeyGen or the BMI as HW Parser next engine, enabling
+ * or bypassing the KeyGen hashing of Rx traffic
+ */
+void fman_port_use_kg_hash(struct fman_port *port, bool enable)
+{
+	if (enable)
+		/* After the Parser frames go to KeyGen */
+		iowrite32be(NIA_ENG_HWK, &port->bmi_regs->rx.fmbm_rfpne);
+	else
+		/* After the Parser frames go to BMI */
+		iowrite32be(NIA_ENG_BMI | NIA_BMI_AC_ENQ_FRAME,
+			    &port->bmi_regs->rx.fmbm_rfpne);
+}
+EXPORT_SYMBOL(fman_port_use_kg_hash);
+
 /**
  * fman_port_init
- * port:	A pointer to a FM Port module.
+ * @port:	A pointer to a FM Port module.
+ *
  * Initializes the FM PORT module by defining the software structure and
  * configuring the hardware registers.
  *
@@ -1407,9 +1415,10 @@ EXPORT_SYMBOL(fman_port_config);
  */
 int fman_port_init(struct fman_port *port)
 {
+	struct fman_port_init_params params;
+	struct fman_keygen *keygen;
 	struct fman_port_cfg *cfg;
 	int err;
-	struct fman_port_init_params params;
 
 	if (is_init_done(port->cfg))
 		return -EINVAL;
@@ -1472,6 +1481,17 @@ int fman_port_init(struct fman_port *port)
 	if (err)
 		return err;
 
+	if (port->cfg->pcd_fqs_count) {
+		keygen = port->dts_params.fman->keygen;
+		err = keygen_port_hashing_init(keygen, port->port_id,
+					       port->cfg->pcd_base_fqid,
+					       port->cfg->pcd_fqs_count);
+		if (err)
+			return err;
+
+		fman_port_use_kg_hash(port, true);
+	}
+
 	kfree(port->cfg);
 	port->cfg = NULL;
 
@@ -1481,8 +1501,8 @@ EXPORT_SYMBOL(fman_port_init);
 
 /**
  * fman_port_cfg_buf_prefix_content
- * @port			A pointer to a FM Port module.
- * @buffer_prefix_content	A structure of parameters describing
+ * @port:			A pointer to a FM Port module.
+ * @buffer_prefix_content:	A structure of parameters describing
  *				the structure of the buffer.
  *				Out parameter:
  *				Start margin - offset of data from
@@ -1527,7 +1547,7 @@ EXPORT_SYMBOL(fman_port_cfg_buf_prefix_content);
 
 /**
  * fman_port_disable
- * port:	A pointer to a FM Port module.
+ * @port:	A pointer to a FM Port module.
  *
  * Gracefully disable an FM port. The port will not start new	tasks after all
  * tasks associated with the port are terminated.
@@ -1608,7 +1628,7 @@ EXPORT_SYMBOL(fman_port_disable);
 
 /**
  * fman_port_enable
- * port:	A pointer to a FM Port module.
+ * @port:	A pointer to a FM Port module.
  *
  * A runtime routine provided to allow disable/enable of port.
  *
@@ -1654,7 +1674,7 @@ EXPORT_SYMBOL(fman_port_enable);
 
 /**
  * fman_port_bind
- * dev:		FMan Port OF device pointer
+ * @dev:		FMan Port OF device pointer
  *
  * Bind to a specific FMan Port.
  *
@@ -1670,7 +1690,7 @@ EXPORT_SYMBOL(fman_port_bind);
 
 /**
  * fman_port_get_qman_channel_id
- * port:	Pointer to the FMan port devuce
+ * @port:	Pointer to the FMan port devuce
  *
  * Get the QMan channel ID for the specific port
  *
@@ -1682,15 +1702,53 @@ u32 fman_port_get_qman_channel_id(struct fman_port *port)
 }
 EXPORT_SYMBOL(fman_port_get_qman_channel_id);
 
+/**
+ * fman_port_get_device
+ * @port:	Pointer to the FMan port device
+ *
+ * Get the 'struct device' associated to the specified FMan port device
+ *
+ * Return: pointer to associated 'struct device'
+ */
+struct device *fman_port_get_device(struct fman_port *port)
+{
+	return port->dev;
+}
+EXPORT_SYMBOL(fman_port_get_device);
+
+int fman_port_get_hash_result_offset(struct fman_port *port, u32 *offset)
+{
+	if (port->buffer_offsets.hash_result_offset == ILLEGAL_BASE)
+		return -EINVAL;
+
+	*offset = port->buffer_offsets.hash_result_offset;
+
+	return 0;
+}
+EXPORT_SYMBOL(fman_port_get_hash_result_offset);
+
+int fman_port_get_tstamp(struct fman_port *port, const void *data, u64 *tstamp)
+{
+	if (port->buffer_offsets.time_stamp_offset == ILLEGAL_BASE)
+		return -EINVAL;
+
+	*tstamp = be64_to_cpu(*(__be64 *)(data +
+			port->buffer_offsets.time_stamp_offset));
+
+	return 0;
+}
+EXPORT_SYMBOL(fman_port_get_tstamp);
+
 static int fman_port_probe(struct platform_device *of_dev)
 {
 	struct fman_port *port;
 	struct fman *fman;
 	struct device_node *fm_node, *port_node;
+	struct platform_device *fm_pdev;
 	struct resource res;
 	struct resource *dev_res;
 	u32 val;
-	int err = 0, lenp;
+	int err = 0;
 	enum fman_port_type port_type;
 	u16 port_speed;
 	u8 port_id;
@@ -1711,19 +1769,25 @@ static int fman_port_probe(struct platform_device *of_dev)
 		goto return_err;
 	}
 
-	fman = dev_get_drvdata(&of_find_device_by_node(fm_node)->dev);
+	fm_pdev = of_find_device_by_node(fm_node);
 	of_node_put(fm_node);
-	if (!fman) {
+	if (!fm_pdev) {
 		err = -EINVAL;
 		goto return_err;
 	}
 
+	fman = dev_get_drvdata(&fm_pdev->dev);
+	if (!fman) {
+		err = -EINVAL;
+		goto put_device;
+	}
+
 	err = of_property_read_u32(port_node, "cell-index", &val);
 	if (err) {
-		dev_err(port->dev, "%s: reading cell-index for %s failed\n",
-			__func__, port_node->full_name);
+		dev_err(port->dev, "%s: reading cell-index for %pOF failed\n",
+			__func__, port_node);
 		err = -EINVAL;
-		goto return_err;
+		goto put_device;
 	}
 	port_id = (u8)val;
 	port->dts_params.id = port_id;
@@ -1731,7 +1795,7 @@ static int fman_port_probe(struct platform_device *of_dev)
 	if (of_device_is_compatible(port_node, "fsl,fman-v3-port-tx")) {
 		port_type = FMAN_PORT_TYPE_TX;
 		port_speed = 1000;
-		if (of_find_property(port_node, "fsl,fman-10g-port", &lenp))
+		if (of_property_read_bool(port_node, "fsl,fman-10g-port"))
 			port_speed = 10000;
 
 	} else if (of_device_is_compatible(port_node, "fsl,fman-v2-port-tx")) {
@@ -1744,7 +1808,7 @@ static int fman_port_probe(struct platform_device *of_dev)
 	} else if (of_device_is_compatible(port_node, "fsl,fman-v3-port-rx")) {
 		port_type = FMAN_PORT_TYPE_RX;
 		port_speed = 1000;
-		if (of_find_property(port_node, "fsl,fman-10g-port", &lenp))
+		if (of_property_read_bool(port_node, "fsl,fman-10g-port"))
 			port_speed = 10000;
 
 	} else if (of_device_is_compatible(port_node, "fsl,fman-v2-port-rx")) {
@@ -1757,7 +1821,7 @@ static int fman_port_probe(struct platform_device *of_dev)
 	}  else {
 		dev_err(port->dev, "%s: Illegal port type\n", __func__);
 		err = -EINVAL;
-		goto return_err;
+		goto put_device;
 	}
 
 	port->dts_params.type = port_type;
@@ -1771,7 +1835,7 @@ static int fman_port_probe(struct platform_device *of_dev)
 			dev_err(port->dev, "%s: incorrect qman-channel-id\n",
 				__func__);
 			err = -EINVAL;
-			goto return_err;
+			goto put_device;
 		}
 		port->dts_params.qman_channel_id = qman_channel_id;
 	}
@@ -1781,7 +1845,7 @@ static int fman_port_probe(struct platform_device *of_dev)
 		dev_err(port->dev, "%s: of_address_to_resource() failed\n",
 			__func__);
 		err = -ENOMEM;
-		goto return_err;
+		goto put_device;
 	}
 
 	port->dts_params.fman = fman;
@@ -1806,6 +1870,8 @@ static int fman_port_probe(struct platform_device *of_dev)
 
 	return 0;
 
+put_device:
+	put_device(&fm_pdev->dev);
 return_err:
 	of_node_put(port_node);
 free_port:
diff --git a/drivers/net/ethernet/freescale/fman/fman_port.h b/drivers/net/ethernet/freescale/fman/fman_port.h
index 8ba901737048..4917fe8f0617 100644
--- a/drivers/net/ethernet/freescale/fman/fman_port.h
+++ b/drivers/net/ethernet/freescale/fman/fman_port.h
@@ -1,33 +1,6 @@
+/* SPDX-License-Identifier: BSD-3-Clause OR GPL-2.0-or-later */
 /*
  * Copyright 2008 - 2015 Freescale Semiconductor Inc.
- *
- * Redistribution and use in source and binary forms, with or without
- * modification, are permitted provided that the following conditions are met:
- *     * Redistributions of source code must retain the above copyright
- *       notice, this list of conditions and the following disclaimer.
- *     * Redistributions in binary form must reproduce the above copyright
- *       notice, this list of conditions and the following disclaimer in the
- *       documentation and/or other materials provided with the distribution.
- *     * Neither the name of Freescale Semiconductor nor the
- *       names of its contributors may be used to endorse or promote products
- *       derived from this software without specific prior written permission.
- *
- *
- * ALTERNATIVELY, this software may be distributed under the terms of the
- * GNU General Public License ("GPL") as published by the Free Software
- * Foundation, either version 2 of that License or (at your option) any
- * later version.
- *
- * THIS SOFTWARE IS PROVIDED BY Freescale Semiconductor ``AS IS'' AND ANY
- * EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED
- * WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE
- * DISCLAIMED. IN NO EVENT SHALL Freescale Semiconductor BE LIABLE FOR ANY
- * DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES
- * (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES;
- * LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND
- * ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
- * (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS
- * SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
  */
 
 #ifndef __FMAN_PORT_H
@@ -100,6 +73,9 @@ struct fman_port;
 struct fman_port_rx_params {
 	u32 err_fqid;			/* Error Queue Id. */
 	u32 dflt_fqid;			/* Default Queue Id. */
+	u32 pcd_base_fqid;		/* PCD base Queue Id. */
+	u32 pcd_fqs_count;		/* Number of PCD FQs. */
+
 	/* Which external buffer pools are used
 	 * (up to FMAN_PORT_MAX_EXT_POOLS_NUM), and their sizes.
 	 */
@@ -134,6 +110,8 @@ struct fman_port_params {
 
 int fman_port_config(struct fman_port *port, struct fman_port_params *params);
 
+void fman_port_use_kg_hash(struct fman_port *port, bool enable);
+
 int fman_port_init(struct fman_port *port);
 
 int fman_port_cfg_buf_prefix_content(struct fman_port *port,
@@ -146,6 +124,12 @@ int fman_port_enable(struct fman_port *port);
 
 u32 fman_port_get_qman_channel_id(struct fman_port *port);
 
+int fman_port_get_hash_result_offset(struct fman_port *port, u32 *offset);
+
+int fman_port_get_tstamp(struct fman_port *port, const void *data, u64 *tstamp);
+
 struct fman_port *fman_port_bind(struct device *dev);
 
+struct device *fman_port_get_device(struct fman_port *port);
+
 #endif /* __FMAN_PORT_H */
diff --git a/drivers/net/ethernet/freescale/fman/fman_sp.c b/drivers/net/ethernet/freescale/fman/fman_sp.c
index 248f5bcca468..0fac60aa5283 100644
--- a/drivers/net/ethernet/freescale/fman/fman_sp.c
+++ b/drivers/net/ethernet/freescale/fman/fman_sp.c
@@ -1,33 +1,6 @@
+// SPDX-License-Identifier: BSD-3-Clause OR GPL-2.0-or-later
 /*
  * Copyright 2008 - 2015 Freescale Semiconductor Inc.
- *
- * Redistribution and use in source and binary forms, with or without
- * modification, are permitted provided that the following conditions are met:
- *     * Redistributions of source code must retain the above copyright
- *       notice, this list of conditions and the following disclaimer.
- *     * Redistributions in binary form must reproduce the above copyright
- *       notice, this list of conditions and the following disclaimer in the
- *       documentation and/or other materials provided with the distribution.
- *     * Neither the name of Freescale Semiconductor nor the
- *       names of its contributors may be used to endorse or promote products
- *       derived from this software without specific prior written permission.
- *
- *
- * ALTERNATIVELY, this software may be distributed under the terms of the
- * GNU General Public License ("GPL") as published by the Free Software
- * Foundation, either version 2 of that License or (at your option) any
- * later version.
- *
- * THIS SOFTWARE IS PROVIDED BY Freescale Semiconductor ``AS IS'' AND ANY
- * EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED
- * WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE
- * DISCLAIMED. IN NO EVENT SHALL Freescale Semiconductor BE LIABLE FOR ANY
- * DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES
- * (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES;
- * LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND
- * ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
- * (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS
- * SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
  */
 
 #include "fman_sp.h"
diff --git a/drivers/net/ethernet/freescale/fman/fman_sp.h b/drivers/net/ethernet/freescale/fman/fman_sp.h
index 820b7f63088f..a62dd21c81f1 100644
--- a/drivers/net/ethernet/freescale/fman/fman_sp.h
+++ b/drivers/net/ethernet/freescale/fman/fman_sp.h
@@ -1,32 +1,6 @@
+/* SPDX-License-Identifier: BSD-3-Clause OR GPL-2.0-or-later */
 /*
  * Copyright 2008 - 2015 Freescale Semiconductor Inc.
- *
- * Redistribution and use in source and binary forms, with or without
- * modification, are permitted provided that the following conditions are met:
- *     * Redistributions of source code must retain the above copyright
- *	 notice, this list of conditions and the following disclaimer.
- *     * Redistributions in binary form must reproduce the above copyright
- *	 notice, this list of conditions and the following disclaimer in the
- *	 documentation and/or other materials provided with the distribution.
- *     * Neither the name of Freescale Semiconductor nor the
- *	 names of its contributors may be used to endorse or promote products
- *	 derived from this software without specific prior written permission.
- *
- * ALTERNATIVELY, this software may be distributed under the terms of the
- * GNU General Public License ("GPL") as published by the Free Software
- * Foundation, either version 2 of that License or (at your option) any
- * later version.
- *
- * THIS SOFTWARE IS PROVIDED BY Freescale Semiconductor ``AS IS'' AND ANY
- * EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED
- * WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE
- * DISCLAIMED. IN NO EVENT SHALL Freescale Semiconductor BE LIABLE FOR ANY
- * DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES
- * (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES;
- * LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND
- * ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
- * (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS
- * SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
  */
 
 #ifndef __FM_SP_H
diff --git a/drivers/net/ethernet/freescale/fman/fman_tgec.c b/drivers/net/ethernet/freescale/fman/fman_tgec.c
index 4b0f3a50b293..fecfca6eba03 100644
--- a/drivers/net/ethernet/freescale/fman/fman_tgec.c
+++ b/drivers/net/ethernet/freescale/fman/fman_tgec.c
@@ -1,49 +1,25 @@
+// SPDX-License-Identifier: BSD-3-Clause OR GPL-2.0-or-later
 /*
- * Copyright 2008-2015 Freescale Semiconductor Inc.
- *
- * Redistribution and use in source and binary forms, with or without
- * modification, are permitted provided that the following conditions are met:
- *     * Redistributions of source code must retain the above copyright
- *       notice, this list of conditions and the following disclaimer.
- *     * Redistributions in binary form must reproduce the above copyright
- *       notice, this list of conditions and the following disclaimer in the
- *       documentation and/or other materials provided with the distribution.
- *     * Neither the name of Freescale Semiconductor nor the
- *       names of its contributors may be used to endorse or promote products
- *       derived from this software without specific prior written permission.
- *
- *
- * ALTERNATIVELY, this software may be distributed under the terms of the
- * GNU General Public License ("GPL") as published by the Free Software
- * Foundation, either version 2 of that License or (at your option) any
- * later version.
- *
- * THIS SOFTWARE IS PROVIDED BY Freescale Semiconductor ``AS IS'' AND ANY
- * EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED
- * WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE
- * DISCLAIMED. IN NO EVENT SHALL Freescale Semiconductor BE LIABLE FOR ANY
- * DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES
- * (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES;
- * LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND
- * ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
- * (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS
- * SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
+ * Copyright 2008 - 2015 Freescale Semiconductor Inc.
  */
 
 #define pr_fmt(fmt) KBUILD_MODNAME ": " fmt
 
 #include "fman_tgec.h"
 #include "fman.h"
+#include "mac.h"
 
 #include <linux/slab.h>
 #include <linux/bitrev.h>
 #include <linux/io.h>
 #include <linux/crc32.h>
+#include <linux/netdevice.h>
 
 /* Transmit Inter-Packet Gap Length Register (TX_IPG_LENGTH) */
 #define TGEC_TX_IPG_LENGTH_MASK	0x000003ff
 
 /* Command and Configuration Register (COMMAND_CONFIG) */
+#define CMD_CFG_EN_TIMESTAMP		0x00100000
 #define CMD_CFG_NO_LEN_CHK		0x00020000
 #define CMD_CFG_PAUSE_IGNORE		0x00000100
 #define CMF_CFG_CRC_FWD			0x00000040
@@ -205,7 +181,7 @@ struct fman_mac {
 	/* MAC address of device; */
 	u64 addr;
 	u16 max_speed;
-	void *dev_id; /* device cookie used by the exception cbs */
+	struct mac_device *dev_id; /* device cookie used by the exception cbs */
 	fman_mac_exception_cb *exception_cb;
 	fman_mac_exception_cb *event_cb;
 	/* pointer to driver's global address hash table  */
@@ -217,9 +193,10 @@ struct fman_mac {
 	struct tgec_cfg *cfg;
 	void *fm;
 	struct fman_rev_info fm_rev_info;
+	bool allmulti_enabled;
 };
 
-static void set_mac_address(struct tgec_regs __iomem *regs, u8 *adr)
+static void set_mac_address(struct tgec_regs __iomem *regs, const u8 *adr)
 {
 	u32 tmp0, tmp1;
 
@@ -267,14 +244,6 @@ static int init(struct tgec_regs __iomem *regs, struct tgec_cfg *cfg,
 
 static int check_init_parameters(struct fman_mac *tgec)
 {
-	if (tgec->max_speed < SPEED_10000) {
-		pr_err("10G MAC driver only support 10G speed\n");
-		return -EINVAL;
-	}
-	if (tgec->addr == 0) {
-		pr_err("Ethernet 10G MAC Must have valid MAC Address\n");
-		return -EINVAL;
-	}
 	if (!tgec->exception_cb) {
 		pr_err("uninitialized exception_cb\n");
 		return -EINVAL;
@@ -412,131 +381,116 @@ static void free_init_resources(struct fman_mac *tgec)
 	tgec->unicast_addr_hash = NULL;
 }
 
-static bool is_init_done(struct tgec_cfg *cfg)
+static int tgec_enable(struct fman_mac *tgec)
 {
-	/* Checks if tGEC driver parameters were initialized */
-	if (!cfg)
-		return true;
+	return 0;
+}
 
-	return false;
+static void tgec_disable(struct fman_mac *tgec)
+{
 }
 
-int tgec_enable(struct fman_mac *tgec, enum comm_mode mode)
+static int tgec_set_promiscuous(struct fman_mac *tgec, bool new_val)
 {
 	struct tgec_regs __iomem *regs = tgec->regs;
 	u32 tmp;
 
-	if (!is_init_done(tgec->cfg))
-		return -EINVAL;
-
 	tmp = ioread32be(&regs->command_config);
-	if (mode & COMM_MODE_RX)
-		tmp |= CMD_CFG_RX_EN;
-	if (mode & COMM_MODE_TX)
-		tmp |= CMD_CFG_TX_EN;
+	if (new_val)
+		tmp |= CMD_CFG_PROMIS_EN;
+	else
+		tmp &= ~CMD_CFG_PROMIS_EN;
 	iowrite32be(tmp, &regs->command_config);
 
 	return 0;
 }
 
-int tgec_disable(struct fman_mac *tgec, enum comm_mode mode)
+static int tgec_set_tx_pause_frames(struct fman_mac *tgec,
+				    u8 __maybe_unused priority, u16 pause_time,
+				    u16 __maybe_unused thresh_time)
 {
 	struct tgec_regs __iomem *regs = tgec->regs;
-	u32 tmp;
-
-	if (!is_init_done(tgec->cfg))
-		return -EINVAL;
 
-	tmp = ioread32be(&regs->command_config);
-	if (mode & COMM_MODE_RX)
-		tmp &= ~CMD_CFG_RX_EN;
-	if (mode & COMM_MODE_TX)
-		tmp &= ~CMD_CFG_TX_EN;
-	iowrite32be(tmp, &regs->command_config);
+	iowrite32be((u32)pause_time, &regs->pause_quant);
 
 	return 0;
 }
 
-int tgec_set_promiscuous(struct fman_mac *tgec, bool new_val)
+static int tgec_accept_rx_pause_frames(struct fman_mac *tgec, bool en)
 {
 	struct tgec_regs __iomem *regs = tgec->regs;
 	u32 tmp;
 
-	if (!is_init_done(tgec->cfg))
-		return -EINVAL;
-
 	tmp = ioread32be(&regs->command_config);
-	if (new_val)
-		tmp |= CMD_CFG_PROMIS_EN;
+	if (!en)
+		tmp |= CMD_CFG_PAUSE_IGNORE;
 	else
-		tmp &= ~CMD_CFG_PROMIS_EN;
+		tmp &= ~CMD_CFG_PAUSE_IGNORE;
 	iowrite32be(tmp, &regs->command_config);
 
 	return 0;
 }
 
-int tgec_cfg_max_frame_len(struct fman_mac *tgec, u16 new_val)
+static void tgec_mac_config(struct phylink_config *config, unsigned int mode,
+			    const struct phylink_link_state *state)
 {
-	if (is_init_done(tgec->cfg))
-		return -EINVAL;
-
-	tgec->cfg->max_frame_length = new_val;
-
-	return 0;
 }
 
-int tgec_set_tx_pause_frames(struct fman_mac *tgec, u8 __maybe_unused priority,
-			     u16 pause_time, u16 __maybe_unused thresh_time)
+static void tgec_link_up(struct phylink_config *config, struct phy_device *phy,
+			 unsigned int mode, phy_interface_t interface,
+			 int speed, int duplex, bool tx_pause, bool rx_pause)
 {
+	struct mac_device *mac_dev = fman_config_to_mac(config);
+	struct fman_mac *tgec = mac_dev->fman_mac;
 	struct tgec_regs __iomem *regs = tgec->regs;
+	u16 pause_time = tx_pause ? FSL_FM_PAUSE_TIME_ENABLE :
+			 FSL_FM_PAUSE_TIME_DISABLE;
+	u32 tmp;
 
-	if (!is_init_done(tgec->cfg))
-		return -EINVAL;
-
-	iowrite32be((u32)pause_time, &regs->pause_quant);
+	tgec_set_tx_pause_frames(tgec, 0, pause_time, 0);
+	tgec_accept_rx_pause_frames(tgec, rx_pause);
+	mac_dev->update_speed(mac_dev, speed);
 
-	return 0;
+	tmp = ioread32be(&regs->command_config);
+	tmp |= CMD_CFG_RX_EN | CMD_CFG_TX_EN;
+	iowrite32be(tmp, &regs->command_config);
 }
 
-int tgec_accept_rx_pause_frames(struct fman_mac *tgec, bool en)
+static void tgec_link_down(struct phylink_config *config, unsigned int mode,
+			   phy_interface_t interface)
 {
+	struct fman_mac *tgec = fman_config_to_mac(config)->fman_mac;
 	struct tgec_regs __iomem *regs = tgec->regs;
 	u32 tmp;
 
-	if (!is_init_done(tgec->cfg))
-		return -EINVAL;
-
 	tmp = ioread32be(&regs->command_config);
-	if (!en)
-		tmp |= CMD_CFG_PAUSE_IGNORE;
-	else
-		tmp &= ~CMD_CFG_PAUSE_IGNORE;
+	tmp &= ~(CMD_CFG_RX_EN | CMD_CFG_TX_EN);
 	iowrite32be(tmp, &regs->command_config);
-
-	return 0;
 }
 
-int tgec_modify_mac_address(struct fman_mac *tgec, enet_addr_t *p_enet_addr)
-{
-	if (!is_init_done(tgec->cfg))
-		return -EINVAL;
+static const struct phylink_mac_ops tgec_mac_ops = {
+	.mac_config = tgec_mac_config,
+	.mac_link_up = tgec_link_up,
+	.mac_link_down = tgec_link_down,
+};
 
+static int tgec_modify_mac_address(struct fman_mac *tgec,
+				   const enet_addr_t *p_enet_addr)
+{
 	tgec->addr = ENET_ADDR_TO_UINT64(*p_enet_addr);
-	set_mac_address(tgec->regs, (u8 *)(*p_enet_addr));
+	set_mac_address(tgec->regs, (const u8 *)(*p_enet_addr));
 
 	return 0;
 }
 
-int tgec_add_hash_mac_address(struct fman_mac *tgec, enet_addr_t *eth_addr)
+static int tgec_add_hash_mac_address(struct fman_mac *tgec,
+				     enet_addr_t *eth_addr)
 {
 	struct tgec_regs __iomem *regs = tgec->regs;
 	struct eth_hash_entry *hash_entry;
 	u32 crc = 0xFFFFFFFF, hash;
 	u64 addr;
 
-	if (!is_init_done(tgec->cfg))
-		return -EINVAL;
-
 	addr = ENET_ADDR_TO_UINT64(*eth_addr);
 
 	if (!(addr & GROUP_ADDRESS)) {
@@ -551,7 +505,7 @@ int tgec_add_hash_mac_address(struct fman_mac *tgec, enet_addr_t *eth_addr)
 	hash = (crc >> TGEC_HASH_MCAST_SHIFT) & TGEC_HASH_ADR_MSK;
 
 	/* Create element to be added to the driver hash table */
-	hash_entry = kmalloc(sizeof(*hash_entry), GFP_KERNEL);
+	hash_entry = kmalloc(sizeof(*hash_entry), GFP_ATOMIC);
 	if (!hash_entry)
 		return -ENOMEM;
 	hash_entry->addr = addr;
@@ -564,7 +518,45 @@ int tgec_add_hash_mac_address(struct fman_mac *tgec, enet_addr_t *eth_addr)
 	return 0;
 }
 
-int tgec_del_hash_mac_address(struct fman_mac *tgec, enet_addr_t *eth_addr)
+static int tgec_set_allmulti(struct fman_mac *tgec, bool enable)
+{
+	u32 entry;
+	struct tgec_regs __iomem *regs = tgec->regs;
+
+	if (enable) {
+		for (entry = 0; entry < TGEC_HASH_TABLE_SIZE; entry++)
+			iowrite32be(entry | TGEC_HASH_MCAST_EN,
+				    &regs->hashtable_ctrl);
+	} else {
+		for (entry = 0; entry < TGEC_HASH_TABLE_SIZE; entry++)
+			iowrite32be(entry & ~TGEC_HASH_MCAST_EN,
+				    &regs->hashtable_ctrl);
+	}
+
+	tgec->allmulti_enabled = enable;
+
+	return 0;
+}
+
+static int tgec_set_tstamp(struct fman_mac *tgec, bool enable)
+{
+	struct tgec_regs __iomem *regs = tgec->regs;
+	u32 tmp;
+
+	tmp = ioread32be(&regs->command_config);
+
+	if (enable)
+		tmp |= CMD_CFG_EN_TIMESTAMP;
+	else
+		tmp &= ~CMD_CFG_EN_TIMESTAMP;
+
+	iowrite32be(tmp, &regs->command_config);
+
+	return 0;
+}
+
+static int tgec_del_hash_mac_address(struct fman_mac *tgec,
+				     enet_addr_t *eth_addr)
 {
 	struct tgec_regs __iomem *regs = tgec->regs;
 	struct eth_hash_entry *hash_entry = NULL;
@@ -572,9 +564,6 @@ int tgec_del_hash_mac_address(struct fman_mac *tgec, enet_addr_t *eth_addr)
 	u32 crc = 0xFFFFFFFF, hash;
 	u64 addr;
 
-	if (!is_init_done(tgec->cfg))
-		return -EINVAL;
-
 	addr = ((*(u64 *)eth_addr) >> 16);
 
 	/* CRC calculation */
@@ -585,40 +574,28 @@ int tgec_del_hash_mac_address(struct fman_mac *tgec, enet_addr_t *eth_addr)
 
 	list_for_each(pos, &tgec->multicast_addr_hash->lsts[hash]) {
 		hash_entry = ETH_HASH_ENTRY_OBJ(pos);
-		if (hash_entry->addr == addr) {
+		if (hash_entry && hash_entry->addr == addr) {
 			list_del_init(&hash_entry->node);
 			kfree(hash_entry);
 			break;
 		}
 	}
-	if (list_empty(&tgec->multicast_addr_hash->lsts[hash]))
-		iowrite32be((hash & ~TGEC_HASH_MCAST_EN),
-			    &regs->hashtable_ctrl);
-
-	return 0;
-}
-
-int tgec_get_version(struct fman_mac *tgec, u32 *mac_version)
-{
-	struct tgec_regs __iomem *regs = tgec->regs;
-
-	if (!is_init_done(tgec->cfg))
-		return -EINVAL;
 
-	*mac_version = ioread32be(&regs->tgec_id);
+	if (!tgec->allmulti_enabled) {
+		if (list_empty(&tgec->multicast_addr_hash->lsts[hash]))
+			iowrite32be((hash & ~TGEC_HASH_MCAST_EN),
+				    &regs->hashtable_ctrl);
+	}
 
 	return 0;
 }
 
-int tgec_set_exception(struct fman_mac *tgec,
-		       enum fman_mac_exceptions exception, bool enable)
+static int tgec_set_exception(struct fman_mac *tgec,
+			      enum fman_mac_exceptions exception, bool enable)
 {
 	struct tgec_regs __iomem *regs = tgec->regs;
 	u32 bit_mask = 0;
 
-	if (!is_init_done(tgec->cfg))
-		return -EINVAL;
-
 	bit_mask = get_exception_flag(exception);
 	if (bit_mask) {
 		if (enable)
@@ -637,15 +614,12 @@ int tgec_set_exception(struct fman_mac *tgec,
 	return 0;
 }
 
-int tgec_init(struct fman_mac *tgec)
+static int tgec_init(struct fman_mac *tgec)
 {
 	struct tgec_cfg *cfg;
 	enet_addr_t eth_addr;
 	int err;
 
-	if (is_init_done(tgec->cfg))
-		return -EINVAL;
-
 	if (DEFAULT_RESET_ON_INIT &&
 	    (fman_reset_mac(tgec->fm, tgec->mac_id) != 0)) {
 		pr_err("Can't reset MAC!\n");
@@ -658,8 +632,10 @@ int tgec_init(struct fman_mac *tgec)
 
 	cfg = tgec->cfg;
 
-	MAKE_ENET_ADDR_FROM_UINT64(tgec->addr, eth_addr);
-	set_mac_address(tgec->regs, (u8 *)eth_addr);
+	if (tgec->addr) {
+		MAKE_ENET_ADDR_FROM_UINT64(tgec->addr, eth_addr);
+		set_mac_address(tgec->regs, (const u8 *)eth_addr);
+	}
 
 	/* interrupts */
 	/* FM_10G_REM_N_LCL_FLT_EX_10GMAC_ERRATA_SW005 Errata workaround */
@@ -718,7 +694,7 @@ int tgec_init(struct fman_mac *tgec)
 	return 0;
 }
 
-int tgec_free(struct fman_mac *tgec)
+static int tgec_free(struct fman_mac *tgec)
 {
 	free_init_resources(tgec);
 
@@ -728,13 +704,12 @@ int tgec_free(struct fman_mac *tgec)
 	return 0;
 }
 
-struct fman_mac *tgec_config(struct fman_mac_params *params)
+static struct fman_mac *tgec_config(struct mac_device *mac_dev,
+				    struct fman_mac_params *params)
 {
 	struct fman_mac *tgec;
 	struct tgec_cfg *cfg;
-	void __iomem *base_addr;
 
-	base_addr = params->base_addr;
 	/* allocate memory for the UCC GETH data structure. */
 	tgec = kzalloc(sizeof(*tgec), GFP_KERNEL);
 	if (!tgec)
@@ -752,9 +727,8 @@ struct fman_mac *tgec_config(struct fman_mac_params *params)
 
 	set_dflts(cfg);
 
-	tgec->regs = base_addr;
-	tgec->addr = ENET_ADDR_TO_UINT64(params->addr);
-	tgec->max_speed = params->max_speed;
+	tgec->regs = mac_dev->vaddr;
+	tgec->addr = ENET_ADDR_TO_UINT64(mac_dev->addr);
 	tgec->mac_id = params->mac_id;
 	tgec->exceptions = (TGEC_IMASK_MDIO_SCAN_EVENT	|
 			    TGEC_IMASK_REM_FAULT	|
@@ -773,7 +747,7 @@ struct fman_mac *tgec_config(struct fman_mac_params *params)
 			    TGEC_IMASK_RX_ALIGN_ER);
 	tgec->exception_cb = params->exception_cb;
 	tgec->event_cb = params->event_cb;
-	tgec->dev_id = params->dev_id;
+	tgec->dev_id = mac_dev;
 	tgec->fm = params->fm;
 
 	/* Save FMan revision */
@@ -781,3 +755,62 @@ struct fman_mac *tgec_config(struct fman_mac_params *params)
 
 	return tgec;
 }
+
+int tgec_initialization(struct mac_device *mac_dev,
+			struct device_node *mac_node,
+			struct fman_mac_params *params)
+{
+	int err;
+	struct fman_mac		*tgec;
+
+	mac_dev->phylink_ops		= &tgec_mac_ops;
+	mac_dev->set_promisc		= tgec_set_promiscuous;
+	mac_dev->change_addr		= tgec_modify_mac_address;
+	mac_dev->add_hash_mac_addr	= tgec_add_hash_mac_address;
+	mac_dev->remove_hash_mac_addr	= tgec_del_hash_mac_address;
+	mac_dev->set_exception		= tgec_set_exception;
+	mac_dev->set_allmulti		= tgec_set_allmulti;
+	mac_dev->set_tstamp		= tgec_set_tstamp;
+	mac_dev->enable			= tgec_enable;
+	mac_dev->disable		= tgec_disable;
+
+	mac_dev->fman_mac = tgec_config(mac_dev, params);
+	if (!mac_dev->fman_mac) {
+		err = -EINVAL;
+		goto _return;
+	}
+
+	/* The internal connection to the serdes is XGMII, but this isn't
+	 * really correct for the phy mode (which is the external connection).
+	 * However, this is how all older device trees say that they want
+	 * XAUI, so just convert it for them.
+	 */
+	if (mac_dev->phy_if == PHY_INTERFACE_MODE_XGMII)
+		mac_dev->phy_if = PHY_INTERFACE_MODE_XAUI;
+
+	__set_bit(PHY_INTERFACE_MODE_XAUI,
+		  mac_dev->phylink_config.supported_interfaces);
+	mac_dev->phylink_config.mac_capabilities =
+		MAC_SYM_PAUSE | MAC_ASYM_PAUSE | MAC_10000FD;
+
+	tgec = mac_dev->fman_mac;
+	tgec->cfg->max_frame_length = fman_get_max_frm();
+	err = tgec_init(tgec);
+	if (err < 0)
+		goto _return_fm_mac_free;
+
+	/* For 10G MAC, disable Tx ECC exception */
+	err = tgec_set_exception(tgec, FM_MAC_EX_10G_TX_ECC_ER, false);
+	if (err < 0)
+		goto _return_fm_mac_free;
+
+	pr_info("FMan XGEC version: 0x%08x\n",
+		ioread32be(&tgec->regs->tgec_id));
+	goto _return;
+
+_return_fm_mac_free:
+	tgec_free(mac_dev->fman_mac);
+
+_return:
+	return err;
+}
diff --git a/drivers/net/ethernet/freescale/fman/fman_tgec.h b/drivers/net/ethernet/freescale/fman/fman_tgec.h
index 514bba9f47ce..768b8d165e05 100644
--- a/drivers/net/ethernet/freescale/fman/fman_tgec.h
+++ b/drivers/net/ethernet/freescale/fman/fman_tgec.h
@@ -1,33 +1,6 @@
+/* SPDX-License-Identifier: BSD-3-Clause OR GPL-2.0-or-later */
 /*
- * Copyright 2008-2015 Freescale Semiconductor Inc.
- *
- * Redistribution and use in source and binary forms, with or without
- * modification, are permitted provided that the following conditions are met:
- *     * Redistributions of source code must retain the above copyright
- *       notice, this list of conditions and the following disclaimer.
- *     * Redistributions in binary form must reproduce the above copyright
- *       notice, this list of conditions and the following disclaimer in the
- *       documentation and/or other materials provided with the distribution.
- *     * Neither the name of Freescale Semiconductor nor the
- *       names of its contributors may be used to endorse or promote products
- *       derived from this software without specific prior written permission.
- *
- *
- * ALTERNATIVELY, this software may be distributed under the terms of the
- * GNU General Public License ("GPL") as published by the Free Software
- * Foundation, either version 2 of that License or (at your option) any
- * later version.
- *
- * THIS SOFTWARE IS PROVIDED BY Freescale Semiconductor ``AS IS'' AND ANY
- * EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED
- * WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE
- * DISCLAIMED. IN NO EVENT SHALL Freescale Semiconductor BE LIABLE FOR ANY
- * DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES
- * (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES;
- * LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND
- * ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
- * (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS
- * SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
+ * Copyright 2008 - 2015 Freescale Semiconductor Inc.
  */
 
 #ifndef __TGEC_H
@@ -35,21 +8,10 @@
 
 #include "fman_mac.h"
 
-struct fman_mac *tgec_config(struct fman_mac_params *params);
-int tgec_set_promiscuous(struct fman_mac *tgec, bool new_val);
-int tgec_modify_mac_address(struct fman_mac *tgec, enet_addr_t *enet_addr);
-int tgec_cfg_max_frame_len(struct fman_mac *tgec, u16 new_val);
-int tgec_enable(struct fman_mac *tgec, enum comm_mode mode);
-int tgec_disable(struct fman_mac *tgec, enum comm_mode mode);
-int tgec_init(struct fman_mac *tgec);
-int tgec_free(struct fman_mac *tgec);
-int tgec_accept_rx_pause_frames(struct fman_mac *tgec, bool en);
-int tgec_set_tx_pause_frames(struct fman_mac *tgec, u8 priority,
-			     u16 pause_time, u16 thresh_time);
-int tgec_set_exception(struct fman_mac *tgec,
-		       enum fman_mac_exceptions exception, bool enable);
-int tgec_add_hash_mac_address(struct fman_mac *tgec, enet_addr_t *eth_addr);
-int tgec_del_hash_mac_address(struct fman_mac *tgec, enet_addr_t *eth_addr);
-int tgec_get_version(struct fman_mac *tgec, u32 *mac_version);
+struct mac_device;
+
+int tgec_initialization(struct mac_device *mac_dev,
+			struct device_node *mac_node,
+			struct fman_mac_params *params);
 
 #endif /* __TGEC_H */
diff --git a/drivers/net/ethernet/freescale/fman/mac.c b/drivers/net/ethernet/freescale/fman/mac.c
index 0b31f8502ada..f27ff625fe29 100644
--- a/drivers/net/ethernet/freescale/fman/mac.c
+++ b/drivers/net/ethernet/freescale/fman/mac.c
@@ -1,32 +1,6 @@
-/* Copyright 2008-2015 Freescale Semiconductor, Inc.
- *
- * Redistribution and use in source and binary forms, with or without
- * modification, are permitted provided that the following conditions are met:
- *     * Redistributions of source code must retain the above copyright
- *	 notice, this list of conditions and the following disclaimer.
- *     * Redistributions in binary form must reproduce the above copyright
- *	 notice, this list of conditions and the following disclaimer in the
- *	 documentation and/or other materials provided with the distribution.
- *     * Neither the name of Freescale Semiconductor nor the
- *	 names of its contributors may be used to endorse or promote products
- *	 derived from this software without specific prior written permission.
- *
- *
- * ALTERNATIVELY, this software may be distributed under the terms of the
- * GNU General Public License ("GPL") as published by the Free Software
- * Foundation, either version 2 of that License or (at your option) any
- * later version.
- *
- * THIS SOFTWARE IS PROVIDED BY Freescale Semiconductor ``AS IS'' AND ANY
- * EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED
- * WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE
- * DISCLAIMED. IN NO EVENT SHALL Freescale Semiconductor BE LIABLE FOR ANY
- * DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES
- * (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES;
- * LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND
- * ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
- * (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS
- * SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
+// SPDX-License-Identifier: BSD-3-Clause OR GPL-2.0-or-later
+/*
+ * Copyright 2008 - 2015 Freescale Semiconductor Inc.
  */
 
 #define pr_fmt(fmt) KBUILD_MODNAME ": " fmt
@@ -40,9 +14,9 @@
 #include <linux/device.h>
 #include <linux/phy.h>
 #include <linux/netdevice.h>
-#include <linux/phy_fixed.h>
 #include <linux/etherdevice.h>
 #include <linux/libfdt_env.h>
+#include <linux/platform_device.h>
 
 #include "mac.h"
 #include "fman_mac.h"
@@ -54,22 +28,10 @@ MODULE_LICENSE("Dual BSD/GPL");
 MODULE_DESCRIPTION("FSL FMan MAC API based driver");
 
 struct mac_priv_s {
-	struct device			*dev;
-	void __iomem			*vaddr;
 	u8				cell_index;
-	phy_interface_t			phy_if;
 	struct fman			*fman;
-	struct device_node		*phy_node;
-	struct device_node		*internal_phy_node;
-	/* List of multicast addresses */
-	struct list_head		mc_addr_list;
 	struct platform_device		*eth_dev;
-	struct fixed_phy_status		*fixed_link;
 	u16				speed;
-	u16				max_speed;
-
-	int (*enable)(struct fman_mac *mac_dev, enum comm_mode mode);
-	int (*disable)(struct fman_mac *mac_dev, enum comm_mode mode);
 };
 
 struct mac_address {
@@ -77,530 +39,24 @@ struct mac_address {
 	struct list_head list;
 };
 
-static void mac_exception(void *handle, enum fman_mac_exceptions ex)
+static void mac_exception(struct mac_device *mac_dev,
+			  enum fman_mac_exceptions ex)
 {
-	struct mac_device	*mac_dev;
-	struct mac_priv_s	*priv;
-
-	mac_dev = handle;
-	priv = mac_dev->priv;
-
 	if (ex == FM_MAC_EX_10G_RX_FIFO_OVFL) {
 		/* don't flag RX FIFO after the first */
 		mac_dev->set_exception(mac_dev->fman_mac,
 				       FM_MAC_EX_10G_RX_FIFO_OVFL, false);
-		dev_err(priv->dev, "10G MAC got RX FIFO Error = %x\n", ex);
+		dev_err(mac_dev->dev, "10G MAC got RX FIFO Error = %x\n", ex);
 	}
 
-	dev_dbg(priv->dev, "%s:%s() -> %d\n", KBUILD_BASENAME ".c",
+	dev_dbg(mac_dev->dev, "%s:%s() -> %d\n", KBUILD_BASENAME ".c",
 		__func__, ex);
 }
 
-static void set_fman_mac_params(struct mac_device *mac_dev,
-				struct fman_mac_params *params)
-{
-	struct mac_priv_s *priv = mac_dev->priv;
-
-	params->base_addr = (typeof(params->base_addr))
-		devm_ioremap(priv->dev, mac_dev->res->start,
-			     resource_size(mac_dev->res));
-	memcpy(&params->addr, mac_dev->addr, sizeof(mac_dev->addr));
-	params->max_speed	= priv->max_speed;
-	params->phy_if		= priv->phy_if;
-	params->basex_if	= false;
-	params->mac_id		= priv->cell_index;
-	params->fm		= (void *)priv->fman;
-	params->exception_cb	= mac_exception;
-	params->event_cb	= mac_exception;
-	params->dev_id		= mac_dev;
-	params->internal_phy_node = priv->internal_phy_node;
-}
-
-static int tgec_initialization(struct mac_device *mac_dev)
-{
-	int err;
-	struct mac_priv_s	*priv;
-	struct fman_mac_params	params;
-	u32			version;
-
-	priv = mac_dev->priv;
-
-	set_fman_mac_params(mac_dev, &params);
-
-	mac_dev->fman_mac = tgec_config(&params);
-	if (!mac_dev->fman_mac) {
-		err = -EINVAL;
-		goto _return;
-	}
-
-	err = tgec_cfg_max_frame_len(mac_dev->fman_mac, fman_get_max_frm());
-	if (err < 0)
-		goto _return_fm_mac_free;
-
-	err = tgec_init(mac_dev->fman_mac);
-	if (err < 0)
-		goto _return_fm_mac_free;
-
-	/* For 10G MAC, disable Tx ECC exception */
-	err = mac_dev->set_exception(mac_dev->fman_mac,
-				     FM_MAC_EX_10G_TX_ECC_ER, false);
-	if (err < 0)
-		goto _return_fm_mac_free;
-
-	err = tgec_get_version(mac_dev->fman_mac, &version);
-	if (err < 0)
-		goto _return_fm_mac_free;
-
-	dev_info(priv->dev, "FMan XGEC version: 0x%08x\n", version);
-
-	goto _return;
-
-_return_fm_mac_free:
-	tgec_free(mac_dev->fman_mac);
-
-_return:
-	return err;
-}
-
-static int dtsec_initialization(struct mac_device *mac_dev)
-{
-	int			err;
-	struct mac_priv_s	*priv;
-	struct fman_mac_params	params;
-	u32			version;
-
-	priv = mac_dev->priv;
-
-	set_fman_mac_params(mac_dev, &params);
-
-	mac_dev->fman_mac = dtsec_config(&params);
-	if (!mac_dev->fman_mac) {
-		err = -EINVAL;
-		goto _return;
-	}
-
-	err = dtsec_cfg_max_frame_len(mac_dev->fman_mac, fman_get_max_frm());
-	if (err < 0)
-		goto _return_fm_mac_free;
-
-	err = dtsec_cfg_pad_and_crc(mac_dev->fman_mac, true);
-	if (err < 0)
-		goto _return_fm_mac_free;
-
-	err = dtsec_init(mac_dev->fman_mac);
-	if (err < 0)
-		goto _return_fm_mac_free;
-
-	/* For 1G MAC, disable by default the MIB counters overflow interrupt */
-	err = mac_dev->set_exception(mac_dev->fman_mac,
-				     FM_MAC_EX_1G_RX_MIB_CNT_OVFL, false);
-	if (err < 0)
-		goto _return_fm_mac_free;
-
-	err = dtsec_get_version(mac_dev->fman_mac, &version);
-	if (err < 0)
-		goto _return_fm_mac_free;
-
-	dev_info(priv->dev, "FMan dTSEC version: 0x%08x\n", version);
-
-	goto _return;
-
-_return_fm_mac_free:
-	dtsec_free(mac_dev->fman_mac);
-
-_return:
-	return err;
-}
-
-static int memac_initialization(struct mac_device *mac_dev)
-{
-	int			 err;
-	struct mac_priv_s	*priv;
-	struct fman_mac_params	 params;
-
-	priv = mac_dev->priv;
-
-	set_fman_mac_params(mac_dev, &params);
-
-	if (priv->max_speed == SPEED_10000)
-		params.phy_if = PHY_INTERFACE_MODE_XGMII;
-
-	mac_dev->fman_mac = memac_config(&params);
-	if (!mac_dev->fman_mac) {
-		err = -EINVAL;
-		goto _return;
-	}
-
-	err = memac_cfg_max_frame_len(mac_dev->fman_mac, fman_get_max_frm());
-	if (err < 0)
-		goto _return_fm_mac_free;
-
-	err = memac_cfg_reset_on_init(mac_dev->fman_mac, true);
-	if (err < 0)
-		goto _return_fm_mac_free;
-
-	err = memac_cfg_fixed_link(mac_dev->fman_mac, priv->fixed_link);
-	if (err < 0)
-		goto _return_fm_mac_free;
-
-	err = memac_init(mac_dev->fman_mac);
-	if (err < 0)
-		goto _return_fm_mac_free;
-
-	dev_info(priv->dev, "FMan MEMAC\n");
-
-	goto _return;
-
-_return_fm_mac_free:
-	memac_free(mac_dev->fman_mac);
-
-_return:
-	return err;
-}
-
-static int start(struct mac_device *mac_dev)
-{
-	int	 err;
-	struct phy_device *phy_dev = mac_dev->phy_dev;
-	struct mac_priv_s *priv = mac_dev->priv;
-
-	err = priv->enable(mac_dev->fman_mac, COMM_MODE_RX_AND_TX);
-	if (!err && phy_dev)
-		phy_start(phy_dev);
-
-	return err;
-}
-
-static int stop(struct mac_device *mac_dev)
-{
-	struct mac_priv_s *priv = mac_dev->priv;
-
-	if (mac_dev->phy_dev)
-		phy_stop(mac_dev->phy_dev);
-
-	return priv->disable(mac_dev->fman_mac, COMM_MODE_RX_AND_TX);
-}
-
-static int set_multi(struct net_device *net_dev, struct mac_device *mac_dev)
-{
-	struct mac_priv_s	*priv;
-	struct mac_address	*old_addr, *tmp;
-	struct netdev_hw_addr	*ha;
-	int			err;
-	enet_addr_t		*addr;
-
-	priv = mac_dev->priv;
-
-	/* Clear previous address list */
-	list_for_each_entry_safe(old_addr, tmp, &priv->mc_addr_list, list) {
-		addr = (enet_addr_t *)old_addr->addr;
-		err = mac_dev->remove_hash_mac_addr(mac_dev->fman_mac, addr);
-		if (err < 0)
-			return err;
-
-		list_del(&old_addr->list);
-		kfree(old_addr);
-	}
-
-	/* Add all the addresses from the new list */
-	netdev_for_each_mc_addr(ha, net_dev) {
-		addr = (enet_addr_t *)ha->addr;
-		err = mac_dev->add_hash_mac_addr(mac_dev->fman_mac, addr);
-		if (err < 0)
-			return err;
-
-		tmp = kmalloc(sizeof(*tmp), GFP_ATOMIC);
-		if (!tmp)
-			return -ENOMEM;
-
-		ether_addr_copy(tmp->addr, ha->addr);
-		list_add(&tmp->list, &priv->mc_addr_list);
-	}
-	return 0;
-}
-
-/**
- * fman_set_mac_active_pause
- * @mac_dev:	A pointer to the MAC device
- * @rx:		Pause frame setting for RX
- * @tx:		Pause frame setting for TX
- *
- * Set the MAC RX/TX PAUSE frames settings
- *
- * Avoid redundant calls to FMD, if the MAC driver already contains the desired
- * active PAUSE settings. Otherwise, the new active settings should be reflected
- * in FMan.
- *
- * Return: 0 on success; Error code otherwise.
- */
-int fman_set_mac_active_pause(struct mac_device *mac_dev, bool rx, bool tx)
-{
-	struct fman_mac *fman_mac = mac_dev->fman_mac;
-	int err = 0;
-
-	if (rx != mac_dev->rx_pause_active) {
-		err = mac_dev->set_rx_pause(fman_mac, rx);
-		if (likely(err == 0))
-			mac_dev->rx_pause_active = rx;
-	}
-
-	if (tx != mac_dev->tx_pause_active) {
-		u16 pause_time = (tx ? FSL_FM_PAUSE_TIME_ENABLE :
-					 FSL_FM_PAUSE_TIME_DISABLE);
-
-		err = mac_dev->set_tx_pause(fman_mac, 0, pause_time, 0);
-
-		if (likely(err == 0))
-			mac_dev->tx_pause_active = tx;
-	}
-
-	return err;
-}
-EXPORT_SYMBOL(fman_set_mac_active_pause);
-
-/**
- * fman_get_pause_cfg
- * @mac_dev:	A pointer to the MAC device
- * @rx:		Return value for RX setting
- * @tx:		Return value for TX setting
- *
- * Determine the MAC RX/TX PAUSE frames settings based on PHY
- * autonegotiation or values set by eththool.
- *
- * Return: Pointer to FMan device.
- */
-void fman_get_pause_cfg(struct mac_device *mac_dev, bool *rx_pause,
-			bool *tx_pause)
-{
-	struct phy_device *phy_dev = mac_dev->phy_dev;
-	u16 lcl_adv, rmt_adv;
-	u8 flowctrl;
-
-	*rx_pause = *tx_pause = false;
-
-	if (!phy_dev->duplex)
-		return;
-
-	/* If PAUSE autonegotiation is disabled, the TX/RX PAUSE settings
-	 * are those set by ethtool.
-	 */
-	if (!mac_dev->autoneg_pause) {
-		*rx_pause = mac_dev->rx_pause_req;
-		*tx_pause = mac_dev->tx_pause_req;
-		return;
-	}
-
-	/* Else if PAUSE autonegotiation is enabled, the TX/RX PAUSE
-	 * settings depend on the result of the link negotiation.
-	 */
-
-	/* get local capabilities */
-	lcl_adv = 0;
-	if (phy_dev->advertising & ADVERTISED_Pause)
-		lcl_adv |= ADVERTISE_PAUSE_CAP;
-	if (phy_dev->advertising & ADVERTISED_Asym_Pause)
-		lcl_adv |= ADVERTISE_PAUSE_ASYM;
-
-	/* get link partner capabilities */
-	rmt_adv = 0;
-	if (phy_dev->pause)
-		rmt_adv |= LPA_PAUSE_CAP;
-	if (phy_dev->asym_pause)
-		rmt_adv |= LPA_PAUSE_ASYM;
-
-	/* Calculate TX/RX settings based on local and peer advertised
-	 * symmetric/asymmetric PAUSE capabilities.
-	 */
-	flowctrl = mii_resolve_flowctrl_fdx(lcl_adv, rmt_adv);
-	if (flowctrl & FLOW_CTRL_RX)
-		*rx_pause = true;
-	if (flowctrl & FLOW_CTRL_TX)
-		*tx_pause = true;
-}
-EXPORT_SYMBOL(fman_get_pause_cfg);
-
-static void adjust_link_void(struct net_device *net_dev)
-{
-}
-
-static void adjust_link_dtsec(struct net_device *net_dev)
-{
-	struct device *dev = net_dev->dev.parent;
-	struct dpaa_eth_data *eth_data = dev->platform_data;
-	struct mac_device *mac_dev = eth_data->mac_dev;
-	struct phy_device *phy_dev = mac_dev->phy_dev;
-	struct fman_mac *fman_mac;
-	bool rx_pause, tx_pause;
-	int err;
-
-	fman_mac = mac_dev->fman_mac;
-	if (!phy_dev->link) {
-		dtsec_restart_autoneg(fman_mac);
-
-		return;
-	}
-
-	dtsec_adjust_link(fman_mac, phy_dev->speed);
-	fman_get_pause_cfg(mac_dev, &rx_pause, &tx_pause);
-	err = fman_set_mac_active_pause(mac_dev, rx_pause, tx_pause);
-	if (err < 0)
-		netdev_err(net_dev, "fman_set_mac_active_pause() = %d\n", err);
-}
-
-static void adjust_link_memac(struct net_device *net_dev)
-{
-	struct device *dev = net_dev->dev.parent;
-	struct dpaa_eth_data *eth_data = dev->platform_data;
-	struct mac_device *mac_dev = eth_data->mac_dev;
-	struct phy_device *phy_dev = mac_dev->phy_dev;
-	struct fman_mac *fman_mac;
-	bool rx_pause, tx_pause;
-	int err;
-
-	fman_mac = mac_dev->fman_mac;
-	memac_adjust_link(fman_mac, phy_dev->speed);
-
-	fman_get_pause_cfg(mac_dev, &rx_pause, &tx_pause);
-	err = fman_set_mac_active_pause(mac_dev, rx_pause, tx_pause);
-	if (err < 0)
-		netdev_err(net_dev, "fman_set_mac_active_pause() = %d\n", err);
-}
-
-/* Initializes driver's PHY state, and attaches to the PHY.
- * Returns 0 on success.
- */
-static struct phy_device *init_phy(struct net_device *net_dev,
-				   struct mac_device *mac_dev,
-				   void (*adj_lnk)(struct net_device *))
-{
-	struct phy_device	*phy_dev;
-	struct mac_priv_s	*priv = mac_dev->priv;
-
-	phy_dev = of_phy_connect(net_dev, priv->phy_node, adj_lnk, 0,
-				 priv->phy_if);
-	if (!phy_dev) {
-		netdev_err(net_dev, "Could not connect to PHY\n");
-		return NULL;
-	}
-
-	/* Remove any features not supported by the controller */
-	phy_dev->supported &= mac_dev->if_support;
-	/* Enable the symmetric and asymmetric PAUSE frame advertisements,
-	 * as most of the PHY drivers do not enable them by default.
-	 */
-	phy_dev->supported |= (SUPPORTED_Pause | SUPPORTED_Asym_Pause);
-	phy_dev->advertising = phy_dev->supported;
-
-	mac_dev->phy_dev = phy_dev;
-
-	return phy_dev;
-}
-
-static struct phy_device *dtsec_init_phy(struct net_device *net_dev,
-					 struct mac_device *mac_dev)
-{
-	return init_phy(net_dev, mac_dev, &adjust_link_dtsec);
-}
-
-static struct phy_device *tgec_init_phy(struct net_device *net_dev,
-					struct mac_device *mac_dev)
-{
-	return init_phy(net_dev, mac_dev, adjust_link_void);
-}
-
-static struct phy_device *memac_init_phy(struct net_device *net_dev,
-					 struct mac_device *mac_dev)
-{
-	return init_phy(net_dev, mac_dev, &adjust_link_memac);
-}
-
-static void setup_dtsec(struct mac_device *mac_dev)
-{
-	mac_dev->init_phy		= dtsec_init_phy;
-	mac_dev->init			= dtsec_initialization;
-	mac_dev->set_promisc		= dtsec_set_promiscuous;
-	mac_dev->change_addr		= dtsec_modify_mac_address;
-	mac_dev->add_hash_mac_addr	= dtsec_add_hash_mac_address;
-	mac_dev->remove_hash_mac_addr	= dtsec_del_hash_mac_address;
-	mac_dev->set_tx_pause		= dtsec_set_tx_pause_frames;
-	mac_dev->set_rx_pause		= dtsec_accept_rx_pause_frames;
-	mac_dev->set_exception		= dtsec_set_exception;
-	mac_dev->set_multi		= set_multi;
-	mac_dev->start			= start;
-	mac_dev->stop			= stop;
-
-	mac_dev->priv->enable		= dtsec_enable;
-	mac_dev->priv->disable		= dtsec_disable;
-}
-
-static void setup_tgec(struct mac_device *mac_dev)
-{
-	mac_dev->init_phy		= tgec_init_phy;
-	mac_dev->init			= tgec_initialization;
-	mac_dev->set_promisc		= tgec_set_promiscuous;
-	mac_dev->change_addr		= tgec_modify_mac_address;
-	mac_dev->add_hash_mac_addr	= tgec_add_hash_mac_address;
-	mac_dev->remove_hash_mac_addr	= tgec_del_hash_mac_address;
-	mac_dev->set_tx_pause		= tgec_set_tx_pause_frames;
-	mac_dev->set_rx_pause		= tgec_accept_rx_pause_frames;
-	mac_dev->set_exception		= tgec_set_exception;
-	mac_dev->set_multi		= set_multi;
-	mac_dev->start			= start;
-	mac_dev->stop			= stop;
-
-	mac_dev->priv->enable		= tgec_enable;
-	mac_dev->priv->disable		= tgec_disable;
-}
-
-static void setup_memac(struct mac_device *mac_dev)
-{
-	mac_dev->init_phy		= memac_init_phy;
-	mac_dev->init			= memac_initialization;
-	mac_dev->set_promisc		= memac_set_promiscuous;
-	mac_dev->change_addr		= memac_modify_mac_address;
-	mac_dev->add_hash_mac_addr	= memac_add_hash_mac_address;
-	mac_dev->remove_hash_mac_addr	= memac_del_hash_mac_address;
-	mac_dev->set_tx_pause		= memac_set_tx_pause_frames;
-	mac_dev->set_rx_pause		= memac_accept_rx_pause_frames;
-	mac_dev->set_exception		= memac_set_exception;
-	mac_dev->set_multi		= set_multi;
-	mac_dev->start			= start;
-	mac_dev->stop			= stop;
-
-	mac_dev->priv->enable		= memac_enable;
-	mac_dev->priv->disable		= memac_disable;
-}
-
-#define DTSEC_SUPPORTED \
-	(SUPPORTED_10baseT_Half \
-	| SUPPORTED_10baseT_Full \
-	| SUPPORTED_100baseT_Half \
-	| SUPPORTED_100baseT_Full \
-	| SUPPORTED_Autoneg \
-	| SUPPORTED_Pause \
-	| SUPPORTED_Asym_Pause \
-	| SUPPORTED_MII)
-
 static DEFINE_MUTEX(eth_lock);
 
-static const u16 phy2speed[] = {
-	[PHY_INTERFACE_MODE_MII]		= SPEED_100,
-	[PHY_INTERFACE_MODE_GMII]		= SPEED_1000,
-	[PHY_INTERFACE_MODE_SGMII]		= SPEED_1000,
-	[PHY_INTERFACE_MODE_TBI]		= SPEED_1000,
-	[PHY_INTERFACE_MODE_RMII]		= SPEED_100,
-	[PHY_INTERFACE_MODE_RGMII]		= SPEED_1000,
-	[PHY_INTERFACE_MODE_RGMII_ID]		= SPEED_1000,
-	[PHY_INTERFACE_MODE_RGMII_RXID]	= SPEED_1000,
-	[PHY_INTERFACE_MODE_RGMII_TXID]	= SPEED_1000,
-	[PHY_INTERFACE_MODE_RTBI]		= SPEED_1000,
-	[PHY_INTERFACE_MODE_QSGMII]		= SPEED_1000,
-	[PHY_INTERFACE_MODE_XGMII]		= SPEED_10000
-};
-
 static struct platform_device *dpaa_eth_add_device(int fman_id,
-						   struct mac_device *mac_dev,
-						   struct device_node *node)
+						   struct mac_device *mac_dev)
 {
 	struct platform_device *pdev;
 	struct dpaa_eth_data data;
@@ -613,16 +69,16 @@ static struct platform_device *dpaa_eth_add_device(int fman_id,
 	data.mac_dev = mac_dev;
 	data.mac_hw_id = priv->cell_index;
 	data.fman_hw_id = fman_id;
-	data.mac_node = node;
 
 	mutex_lock(&eth_lock);
-
 	pdev = platform_device_alloc("dpaa-ethernet", dpaa_eth_dev_cnt);
 	if (!pdev) {
 		ret = -ENOMEM;
 		goto no_mem;
 	}
 
+	pdev->dev.parent = mac_dev->dev;
+
 	ret = platform_device_add_data(pdev, &data, sizeof(data));
 	if (ret)
 		goto err;
@@ -645,9 +101,9 @@ no_mem:
 }
 
 static const struct of_device_id mac_match[] = {
-	{ .compatible	= "fsl,fman-dtsec" },
-	{ .compatible	= "fsl,fman-xgec" },
-	{ .compatible	= "fsl,fman-memac" },
+	{ .compatible   = "fsl,fman-dtsec", .data = dtsec_initialization },
+	{ .compatible   = "fsl,fman-xgec", .data = tgec_initialization },
+	{ .compatible	= "fsl,fman-memac", .data = memac_initialization },
 	{}
 };
 MODULE_DEVICE_TABLE(of, mac_match);
@@ -655,301 +111,236 @@ MODULE_DEVICE_TABLE(of, mac_match);
 static int mac_probe(struct platform_device *_of_dev)
 {
 	int			 err, i, nph;
+	int (*init)(struct mac_device *mac_dev, struct device_node *mac_node,
+		    struct fman_mac_params *params);
 	struct device		*dev;
 	struct device_node	*mac_node, *dev_node;
 	struct mac_device	*mac_dev;
 	struct platform_device	*of_dev;
-	struct resource		 res;
 	struct mac_priv_s	*priv;
-	const u8		*mac_addr;
+	struct fman_mac_params	 params;
 	u32			 val;
 	u8			fman_id;
-	int			phy_if;
+	phy_interface_t          phy_if;
 
 	dev = &_of_dev->dev;
 	mac_node = dev->of_node;
+	init = of_device_get_match_data(dev);
 
 	mac_dev = devm_kzalloc(dev, sizeof(*mac_dev), GFP_KERNEL);
-	if (!mac_dev) {
-		err = -ENOMEM;
-		dev_err(dev, "devm_kzalloc() = %d\n", err);
-		goto _return;
-	}
+	if (!mac_dev)
+		return -ENOMEM;
 	priv = devm_kzalloc(dev, sizeof(*priv), GFP_KERNEL);
-	if (!priv) {
-		err = -ENOMEM;
-		goto _return;
-	}
+	if (!priv)
+		return -ENOMEM;
+	platform_set_drvdata(_of_dev, mac_dev);
 
 	/* Save private information */
 	mac_dev->priv = priv;
-	priv->dev = dev;
-
-	if (of_device_is_compatible(mac_node, "fsl,fman-dtsec")) {
-		setup_dtsec(mac_dev);
-		priv->internal_phy_node = of_parse_phandle(mac_node,
-							  "tbi-handle", 0);
-	} else if (of_device_is_compatible(mac_node, "fsl,fman-xgec")) {
-		setup_tgec(mac_dev);
-	} else if (of_device_is_compatible(mac_node, "fsl,fman-memac")) {
-		setup_memac(mac_dev);
-		priv->internal_phy_node = of_parse_phandle(mac_node,
-							  "pcsphy-handle", 0);
-	} else {
-		dev_err(dev, "MAC node (%s) contains unsupported MAC\n",
-			mac_node->full_name);
-		err = -EINVAL;
-		goto _return;
-	}
-
-	/* Register mac_dev */
-	dev_set_drvdata(dev, mac_dev);
-
-	INIT_LIST_HEAD(&priv->mc_addr_list);
+	mac_dev->dev = dev;
 
 	/* Get the FM node */
 	dev_node = of_get_parent(mac_node);
 	if (!dev_node) {
-		dev_err(dev, "of_get_parent(%s) failed\n",
-			mac_node->full_name);
-		err = -EINVAL;
-		goto _return_dev_set_drvdata;
+		dev_err(dev, "of_get_parent(%pOF) failed\n",
+			mac_node);
+		return -EINVAL;
 	}
 
 	of_dev = of_find_device_by_node(dev_node);
 	if (!of_dev) {
-		dev_err(dev, "of_find_device_by_node(%s) failed\n",
-			dev_node->full_name);
+		dev_err(dev, "of_find_device_by_node(%pOF) failed\n", dev_node);
 		err = -EINVAL;
 		goto _return_of_node_put;
 	}
+	mac_dev->fman_dev = &of_dev->dev;
 
 	/* Get the FMan cell-index */
 	err = of_property_read_u32(dev_node, "cell-index", &val);
 	if (err) {
-		dev_err(dev, "failed to read cell-index for %s\n",
-			dev_node->full_name);
+		dev_err(dev, "failed to read cell-index for %pOF\n", dev_node);
 		err = -EINVAL;
-		goto _return_of_node_put;
+		goto _return_dev_put;
 	}
 	/* cell-index 0 => FMan id 1 */
 	fman_id = (u8)(val + 1);
 
-	priv->fman = fman_bind(&of_dev->dev);
+	priv->fman = fman_bind(mac_dev->fman_dev);
 	if (!priv->fman) {
-		dev_err(dev, "fman_bind(%s) failed\n", dev_node->full_name);
+		dev_err(dev, "fman_bind(%pOF) failed\n", dev_node);
 		err = -ENODEV;
-		goto _return_of_node_put;
+		goto _return_dev_put;
 	}
 
+	/* Two references have been taken in of_find_device_by_node()
+	 * and fman_bind(). Release one of them here. The second one
+	 * will be released in mac_remove().
+	 */
+	put_device(mac_dev->fman_dev);
 	of_node_put(dev_node);
+	dev_node = NULL;
 
 	/* Get the address of the memory mapped registers */
-	err = of_address_to_resource(mac_node, 0, &res);
-	if (err < 0) {
-		dev_err(dev, "of_address_to_resource(%s) = %d\n",
-			mac_node->full_name, err);
-		goto _return_dev_set_drvdata;
+	mac_dev->res = platform_get_mem_or_io(_of_dev, 0);
+	if (!mac_dev->res) {
+		dev_err(dev, "could not get registers\n");
+		err = -EINVAL;
+		goto _return_dev_put;
 	}
 
-	mac_dev->res = __devm_request_region(dev,
-					     fman_get_mem_region(priv->fman),
-					     res.start, res.end + 1 - res.start,
-					     "mac");
-	if (!mac_dev->res) {
-		dev_err(dev, "__devm_request_mem_region(mac) failed\n");
-		err = -EBUSY;
-		goto _return_dev_set_drvdata;
+	err = devm_request_resource(dev, fman_get_mem_region(priv->fman),
+				    mac_dev->res);
+	if (err) {
+		dev_err_probe(dev, err, "could not request resource\n");
+		goto _return_dev_put;
 	}
 
-	priv->vaddr = devm_ioremap(dev, mac_dev->res->start,
-				   mac_dev->res->end + 1 - mac_dev->res->start);
-	if (!priv->vaddr) {
+	mac_dev->vaddr = devm_ioremap(dev, mac_dev->res->start,
+				      resource_size(mac_dev->res));
+	if (!mac_dev->vaddr) {
 		dev_err(dev, "devm_ioremap() failed\n");
 		err = -EIO;
-		goto _return_dev_set_drvdata;
+		goto _return_dev_put;
 	}
 
 	if (!of_device_is_available(mac_node)) {
-		devm_iounmap(dev, priv->vaddr);
-		__devm_release_region(dev, fman_get_mem_region(priv->fman),
-				      res.start, res.end + 1 - res.start);
-		devm_kfree(dev, mac_dev);
-		dev_set_drvdata(dev, NULL);
-		return -ENODEV;
+		err = -ENODEV;
+		goto _return_dev_put;
 	}
 
 	/* Get the cell-index */
 	err = of_property_read_u32(mac_node, "cell-index", &val);
 	if (err) {
-		dev_err(dev, "failed to read cell-index for %s\n",
-			mac_node->full_name);
+		dev_err(dev, "failed to read cell-index for %pOF\n", mac_node);
+		err = -EINVAL;
+		goto _return_dev_put;
+	}
+	if (val >= MAX_NUM_OF_MACS) {
+		dev_err(dev, "cell-index value is too big for %pOF\n", mac_node);
 		err = -EINVAL;
-		goto _return_dev_set_drvdata;
+		goto _return_dev_put;
 	}
 	priv->cell_index = (u8)val;
 
 	/* Get the MAC address */
-	mac_addr = of_get_mac_address(mac_node);
-	if (!mac_addr) {
-		dev_err(dev, "of_get_mac_address(%s) failed\n",
-			mac_node->full_name);
-		err = -EINVAL;
-		goto _return_dev_set_drvdata;
-	}
-	memcpy(mac_dev->addr, mac_addr, sizeof(mac_dev->addr));
+	err = of_get_mac_address(mac_node, mac_dev->addr);
+	if (err)
+		dev_warn(dev, "of_get_mac_address(%pOF) failed\n", mac_node);
 
 	/* Get the port handles */
 	nph = of_count_phandle_with_args(mac_node, "fsl,fman-ports", NULL);
 	if (unlikely(nph < 0)) {
-		dev_err(dev, "of_count_phandle_with_args(%s, fsl,fman-ports) failed\n",
-			mac_node->full_name);
+		dev_err(dev, "of_count_phandle_with_args(%pOF, fsl,fman-ports) failed\n",
+			mac_node);
 		err = nph;
-		goto _return_dev_set_drvdata;
+		goto _return_dev_put;
 	}
 
 	if (nph != ARRAY_SIZE(mac_dev->port)) {
-		dev_err(dev, "Not supported number of fman-ports handles of mac node %s from device tree\n",
-			mac_node->full_name);
+		dev_err(dev, "Not supported number of fman-ports handles of mac node %pOF from device tree\n",
+			mac_node);
 		err = -EINVAL;
-		goto _return_dev_set_drvdata;
+		goto _return_dev_put;
 	}
 
-	for (i = 0; i < ARRAY_SIZE(mac_dev->port); i++) {
+	/* PORT_NUM determines the size of the port array */
+	for (i = 0; i < PORT_NUM; i++) {
 		/* Find the port node */
 		dev_node = of_parse_phandle(mac_node, "fsl,fman-ports", i);
 		if (!dev_node) {
-			dev_err(dev, "of_parse_phandle(%s, fsl,fman-ports) failed\n",
-				mac_node->full_name);
+			dev_err(dev, "of_parse_phandle(%pOF, fsl,fman-ports) failed\n",
+				mac_node);
 			err = -EINVAL;
-			goto _return_of_node_put;
+			goto _return_dev_arr_put;
 		}
 
 		of_dev = of_find_device_by_node(dev_node);
 		if (!of_dev) {
-			dev_err(dev, "of_find_device_by_node(%s) failed\n",
-				dev_node->full_name);
+			dev_err(dev, "of_find_device_by_node(%pOF) failed\n",
+				dev_node);
 			err = -EINVAL;
-			goto _return_of_node_put;
+			goto _return_dev_arr_put;
 		}
+		mac_dev->fman_port_devs[i] = &of_dev->dev;
 
-		mac_dev->port[i] = fman_port_bind(&of_dev->dev);
+		mac_dev->port[i] = fman_port_bind(mac_dev->fman_port_devs[i]);
 		if (!mac_dev->port[i]) {
-			dev_err(dev, "dev_get_drvdata(%s) failed\n",
-				dev_node->full_name);
+			dev_err(dev, "dev_get_drvdata(%pOF) failed\n",
+				dev_node);
 			err = -EINVAL;
-			goto _return_of_node_put;
+			goto _return_dev_arr_put;
 		}
+		/* Two references have been taken in of_find_device_by_node()
+		 * and fman_port_bind(). Release one of them here. The second
+		 * one will be released in mac_remove().
+		 */
+		put_device(mac_dev->fman_port_devs[i]);
 		of_node_put(dev_node);
+		dev_node = NULL;
 	}
 
 	/* Get the PHY connection type */
-	phy_if = of_get_phy_mode(mac_node);
-	if (phy_if < 0) {
+	err = of_get_phy_mode(mac_node, &phy_if);
+	if (err) {
 		dev_warn(dev,
-			 "of_get_phy_mode() for %s failed. Defaulting to SGMII\n",
-			 mac_node->full_name);
+			 "of_get_phy_mode() for %pOF failed. Defaulting to SGMII\n",
+			 mac_node);
 		phy_if = PHY_INTERFACE_MODE_SGMII;
 	}
-	priv->phy_if = phy_if;
-
-	priv->speed		= phy2speed[priv->phy_if];
-	priv->max_speed		= priv->speed;
-	mac_dev->if_support	= DTSEC_SUPPORTED;
-	/* We don't support half-duplex in SGMII mode */
-	if (priv->phy_if == PHY_INTERFACE_MODE_SGMII)
-		mac_dev->if_support &= ~(SUPPORTED_10baseT_Half |
-					SUPPORTED_100baseT_Half);
-
-	/* Gigabit support (no half-duplex) */
-	if (priv->max_speed == 1000)
-		mac_dev->if_support |= SUPPORTED_1000baseT_Full;
-
-	/* The 10G interface only supports one mode */
-	if (priv->phy_if == PHY_INTERFACE_MODE_XGMII)
-		mac_dev->if_support = SUPPORTED_10000baseT_Full;
-
-	/* Get the rest of the PHY information */
-	priv->phy_node = of_parse_phandle(mac_node, "phy-handle", 0);
-	if (!priv->phy_node && of_phy_is_fixed_link(mac_node)) {
-		struct phy_device *phy;
-
-		err = of_phy_register_fixed_link(mac_node);
-		if (err)
-			goto _return_dev_set_drvdata;
-
-		priv->fixed_link = kzalloc(sizeof(*priv->fixed_link),
-					   GFP_KERNEL);
-		if (!priv->fixed_link) {
-			err = -ENOMEM;
-			goto _return_dev_set_drvdata;
-		}
-
-		priv->phy_node = of_node_get(mac_node);
-		phy = of_phy_find_device(priv->phy_node);
-		if (!phy) {
-			err = -EINVAL;
-			goto _return_dev_set_drvdata;
-		}
-
-		priv->fixed_link->link = phy->link;
-		priv->fixed_link->speed = phy->speed;
-		priv->fixed_link->duplex = phy->duplex;
-		priv->fixed_link->pause = phy->pause;
-		priv->fixed_link->asym_pause = phy->asym_pause;
-
-		put_device(&phy->mdio.dev);
-	}
+	mac_dev->phy_if = phy_if;
 
-	err = mac_dev->init(mac_dev);
-	if (err < 0) {
-		dev_err(dev, "mac_dev->init() = %d\n", err);
-		of_node_put(priv->phy_node);
-		goto _return_dev_set_drvdata;
-	}
-
-	/* pause frame autonegotiation enabled */
-	mac_dev->autoneg_pause = true;
+	params.mac_id		= priv->cell_index;
+	params.fm		= (void *)priv->fman;
+	params.exception_cb	= mac_exception;
+	params.event_cb		= mac_exception;
 
-	/* By intializing the values to false, force FMD to enable PAUSE frames
-	 * on RX and TX
-	 */
-	mac_dev->rx_pause_req = true;
-	mac_dev->tx_pause_req = true;
-	mac_dev->rx_pause_active = false;
-	mac_dev->tx_pause_active = false;
-	err = fman_set_mac_active_pause(mac_dev, true, true);
+	err = init(mac_dev, mac_node, &params);
 	if (err < 0)
-		dev_err(dev, "fman_set_mac_active_pause() = %d\n", err);
+		goto _return_dev_arr_put;
 
-	dev_info(dev, "FMan MAC address: %02hx:%02hx:%02hx:%02hx:%02hx:%02hx\n",
-		 mac_dev->addr[0], mac_dev->addr[1], mac_dev->addr[2],
-		 mac_dev->addr[3], mac_dev->addr[4], mac_dev->addr[5]);
+	if (!is_zero_ether_addr(mac_dev->addr))
+		dev_info(dev, "FMan MAC address: %pM\n", mac_dev->addr);
 
-	priv->eth_dev = dpaa_eth_add_device(fman_id, mac_dev, mac_node);
+	priv->eth_dev = dpaa_eth_add_device(fman_id, mac_dev);
 	if (IS_ERR(priv->eth_dev)) {
+		err = PTR_ERR(priv->eth_dev);
 		dev_err(dev, "failed to add Ethernet platform device for MAC %d\n",
 			priv->cell_index);
 		priv->eth_dev = NULL;
 	}
 
-	goto _return;
+	return err;
 
+_return_dev_arr_put:
+	/* mac_dev is kzalloc'ed */
+	for (i = 0; i < PORT_NUM; i++)
+		put_device(mac_dev->fman_port_devs[i]);
+_return_dev_put:
+	put_device(mac_dev->fman_dev);
 _return_of_node_put:
 	of_node_put(dev_node);
-_return_dev_set_drvdata:
-	kfree(priv->fixed_link);
-	dev_set_drvdata(dev, NULL);
-_return:
 	return err;
 }
 
+static void mac_remove(struct platform_device *pdev)
+{
+	struct mac_device *mac_dev = platform_get_drvdata(pdev);
+	int		   i;
+
+	for (i = 0; i < PORT_NUM; i++)
+		put_device(mac_dev->fman_port_devs[i]);
+	put_device(mac_dev->fman_dev);
+
+	platform_device_unregister(mac_dev->priv->eth_dev);
+}
+
 static struct platform_driver mac_driver = {
 	.driver = {
 		.name		= KBUILD_MODNAME,
 		.of_match_table	= mac_match,
 	},
 	.probe		= mac_probe,
+	.remove		= mac_remove,
 };
 
 builtin_platform_driver(mac_driver);
diff --git a/drivers/net/ethernet/freescale/fman/mac.h b/drivers/net/ethernet/freescale/fman/mac.h
index d7313f0c5135..955ace338965 100644
--- a/drivers/net/ethernet/freescale/fman/mac.h
+++ b/drivers/net/ethernet/freescale/fman/mac.h
@@ -1,32 +1,6 @@
-/* Copyright 2008-2015 Freescale Semiconductor, Inc.
- *
- * Redistribution and use in source and binary forms, with or without
- * modification, are permitted provided that the following conditions are met:
- *     * Redistributions of source code must retain the above copyright
- *	 notice, this list of conditions and the following disclaimer.
- *     * Redistributions in binary form must reproduce the above copyright
- *	 notice, this list of conditions and the following disclaimer in the
- *	 documentation and/or other materials provided with the distribution.
- *     * Neither the name of Freescale Semiconductor nor the
- *	 names of its contributors may be used to endorse or promote products
- *	 derived from this software without specific prior written permission.
- *
- *
- * ALTERNATIVELY, this software may be distributed under the terms of the
- * GNU General Public License ("GPL") as published by the Free Software
- * Foundation, either version 2 of that License or (at your option) any
- * later version.
- *
- * THIS SOFTWARE IS PROVIDED BY Freescale Semiconductor ``AS IS'' AND ANY
- * EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED
- * WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE
- * DISCLAIMED. IN NO EVENT SHALL Freescale Semiconductor BE LIABLE FOR ANY
- * DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES
- * (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES;
- * LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND
- * ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
- * (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS
- * SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
+/* SPDX-License-Identifier: BSD-3-Clause OR GPL-2.0-or-later */
+/*
+ * Copyright 2008 - 2015 Freescale Semiconductor Inc.
  */
 
 #ifndef __MAC_H
@@ -35,6 +9,7 @@
 #include <linux/device.h>
 #include <linux/if_ether.h>
 #include <linux/phy.h>
+#include <linux/phylink.h>
 #include <linux/list.h>
 
 #include "fman_port.h"
@@ -44,32 +19,27 @@
 struct fman_mac;
 struct mac_priv_s;
 
+#define PORT_NUM 2
 struct mac_device {
+	void __iomem		*vaddr;
+	struct device		*dev;
 	struct resource		*res;
 	u8			 addr[ETH_ALEN];
-	struct fman_port	*port[2];
-	u32			 if_support;
-	struct phy_device	*phy_dev;
+	struct fman_port	*port[PORT_NUM];
+	struct phylink		*phylink;
+	struct phylink_config	phylink_config;
+	phy_interface_t		phy_if;
 
-	bool autoneg_pause;
-	bool rx_pause_req;
-	bool tx_pause_req;
-	bool rx_pause_active;
-	bool tx_pause_active;
 	bool promisc;
+	bool allmulti;
 
-	struct phy_device *(*init_phy)(struct net_device *net_dev,
-				       struct mac_device *mac_dev);
-	int (*init)(struct mac_device *mac_dev);
-	int (*start)(struct mac_device *mac_dev);
-	int (*stop)(struct mac_device *mac_dev);
+	const struct phylink_mac_ops *phylink_ops;
+	int (*enable)(struct fman_mac *mac_dev);
+	void (*disable)(struct fman_mac *mac_dev);
 	int (*set_promisc)(struct fman_mac *mac_dev, bool enable);
-	int (*change_addr)(struct fman_mac *mac_dev, enet_addr_t *enet_addr);
-	int (*set_multi)(struct net_device *net_dev,
-			 struct mac_device *mac_dev);
-	int (*set_rx_pause)(struct fman_mac *mac_dev, bool en);
-	int (*set_tx_pause)(struct fman_mac *mac_dev, u8 priority,
-			    u16 pause_time, u16 thresh_time);
+	int (*change_addr)(struct fman_mac *mac_dev, const enet_addr_t *enet_addr);
+	int (*set_allmulti)(struct fman_mac *mac_dev, bool enable);
+	int (*set_tstamp)(struct fman_mac *mac_dev, bool enable);
 	int (*set_exception)(struct fman_mac *mac_dev,
 			     enum fman_mac_exceptions exception, bool enable);
 	int (*add_hash_mac_addr)(struct fman_mac *mac_dev,
@@ -77,12 +47,22 @@ struct mac_device {
 	int (*remove_hash_mac_addr)(struct fman_mac *mac_dev,
 				    enet_addr_t *eth_addr);
 
+	void (*update_speed)(struct mac_device *mac_dev, int speed);
+
 	struct fman_mac		*fman_mac;
 	struct mac_priv_s	*priv;
+
+	struct device		*fman_dev;
+	struct device		*fman_port_devs[PORT_NUM];
 };
 
+static inline struct mac_device
+*fman_config_to_mac(struct phylink_config *config)
+{
+	return container_of(config, struct mac_device, phylink_config);
+}
+
 struct dpaa_eth_data {
-	struct device_node *mac_node;
 	struct mac_device *mac_dev;
 	int mac_hw_id;
 	int fman_hw_id;
@@ -90,9 +70,6 @@ struct dpaa_eth_data {
 
 extern const char	*mac_driver_description;
 
-int fman_set_mac_active_pause(struct mac_device *mac_dev, bool rx, bool tx);
-
-void fman_get_pause_cfg(struct mac_device *mac_dev, bool *rx_pause,
-			bool *tx_pause);
+int fman_set_multi(struct net_device *net_dev, struct mac_device *mac_dev);
 
 #endif	/* __MAC_H */
diff --git a/drivers/net/ethernet/freescale/fs_enet/Kconfig b/drivers/net/ethernet/freescale/fs_enet/Kconfig
index be92229f2c2a..57013bf14d7c 100644
--- a/drivers/net/ethernet/freescale/fs_enet/Kconfig
+++ b/drivers/net/ethernet/freescale/fs_enet/Kconfig
@@ -1,8 +1,9 @@
+# SPDX-License-Identifier: GPL-2.0-only
 config FS_ENET
-       tristate "Freescale Ethernet Driver"
-       depends on NET_VENDOR_FREESCALE && (CPM1 || CPM2 || PPC_MPC512x)
-       select MII
-       select PHYLIB
+	tristate "Freescale Ethernet Driver"
+	depends on NET_VENDOR_FREESCALE && (CPM1 || CPM2 || PPC_MPC512x)
+	select MII
+	select PHYLINK
 
 config FS_ENET_MPC5121_FEC
 	def_bool y if (FS_ENET && PPC_MPC512x)
diff --git a/drivers/net/ethernet/freescale/fs_enet/Makefile b/drivers/net/ethernet/freescale/fs_enet/Makefile
index d4a305ee3455..1821f94ef619 100644
--- a/drivers/net/ethernet/freescale/fs_enet/Makefile
+++ b/drivers/net/ethernet/freescale/fs_enet/Makefile
@@ -1,3 +1,4 @@
+# SPDX-License-Identifier: GPL-2.0
 #
 # Makefile for the Freescale Ethernet controllers
 #
diff --git a/drivers/net/ethernet/freescale/fs_enet/fec.h b/drivers/net/ethernet/freescale/fs_enet/fec.h
index b9fe5bde432a..1dbee5d898b3 100644
--- a/drivers/net/ethernet/freescale/fs_enet/fec.h
+++ b/drivers/net/ethernet/freescale/fs_enet/fec.h
@@ -1,9 +1,7 @@
+/* SPDX-License-Identifier: GPL-2.0 */
 #ifndef FS_ENET_FEC_H
 #define FS_ENET_FEC_H
 
-/* CRC polynomium used by the FEC for the multicast group filtering */
-#define FEC_CRC_POLY   0x04C11DB7
-
 #define FEC_MAX_MULTICAST_ADDRS	64
 
 /* Interrupt events/masks.
diff --git a/drivers/net/ethernet/freescale/fs_enet/fs_enet-main.c b/drivers/net/ethernet/freescale/fs_enet/fs_enet-main.c
index 753259091b22..f563692a4a00 100644
--- a/drivers/net/ethernet/freescale/fs_enet/fs_enet-main.c
+++ b/drivers/net/ethernet/freescale/fs_enet/fs_enet-main.c
@@ -1,3 +1,4 @@
+// SPDX-License-Identifier: GPL-2.0-only
 /*
  * Combined Ethernet driver for Motorola MPC8xx and MPC82xx.
  *
@@ -9,10 +10,6 @@
  *
  * Heavily based on original FEC driver by Dan Malek <dan@embeddededge.com>
  * and modifications by Joakim Tjernlund <joakim.tjernlund@lumentis.se>
- *
- * This file is licensed under the terms of the GNU General Public License
- * version 2. This program is licensed "as is" without any warranty of any
- * kind, whether express or implied.
  */
 
 #include <linux/module.h>
@@ -29,20 +26,20 @@
 #include <linux/etherdevice.h>
 #include <linux/skbuff.h>
 #include <linux/spinlock.h>
-#include <linux/mii.h>
 #include <linux/ethtool.h>
 #include <linux/bitops.h>
 #include <linux/fs.h>
 #include <linux/platform_device.h>
 #include <linux/phy.h>
+#include <linux/phylink.h>
+#include <linux/property.h>
 #include <linux/of.h>
 #include <linux/of_mdio.h>
-#include <linux/of_platform.h>
-#include <linux/of_gpio.h>
 #include <linux/of_net.h>
+#include <linux/pgtable.h>
+#include <linux/rtnetlink.h>
 
 #include <linux/vmalloc.h>
-#include <asm/pgtable.h>
 #include <asm/irq.h>
 #include <linux/uaccess.h>
 
@@ -53,7 +50,6 @@
 MODULE_AUTHOR("Pantelis Antoniou <panto@intracom.gr>");
 MODULE_DESCRIPTION("Freescale Ethernet Driver");
 MODULE_LICENSE("GPL");
-MODULE_VERSION(DRV_MODULE_VERSION);
 
 static int fs_enet_debug = -1; /* -1 == use FS_ENET_DEF_MSG_ENABLE as value */
 module_param(fs_enet_debug, int, 0);
@@ -74,6 +70,13 @@ static void fs_set_multicast_list(struct net_device *dev)
 	(*fep->ops->set_multicast_list)(dev);
 }
 
+static int fs_eth_ioctl(struct net_device *dev, struct ifreq *ifr, int cmd)
+{
+	struct fs_enet_private *fep = netdev_priv(dev);
+
+	return phylink_mii_ioctl(fep->phylink, ifr, cmd);
+}
+
 static void skb_align(struct sk_buff *skb, int align)
 {
 	int off = ((unsigned long)skb->data) & (align - 1);
@@ -86,15 +89,14 @@ static void skb_align(struct sk_buff *skb, int align)
 static int fs_enet_napi(struct napi_struct *napi, int budget)
 {
 	struct fs_enet_private *fep = container_of(napi, struct fs_enet_private, napi);
-	struct net_device *dev = fep->ndev;
 	const struct fs_platform_info *fpi = fep->fpi;
-	cbd_t __iomem *bdp;
+	struct net_device *dev = fep->ndev;
+	int curidx, dirtyidx, received = 0;
+	int do_wake = 0, do_restart = 0;
+	int tx_left = TX_RING_SIZE;
 	struct sk_buff *skb, *skbn;
-	int received = 0;
+	cbd_t __iomem *bdp;
 	u16 pkt_len, sc;
-	int curidx;
-	int dirtyidx, do_wake, do_restart;
-	int tx_left = TX_RING_SIZE;
 
 	spin_lock(&fep->tx_lock);
 	bdp = fep->dirty_tx;
@@ -102,7 +104,6 @@ static int fs_enet_napi(struct napi_struct *napi, int budget)
 	/* clear status bits for napi*/
 	(*fep->ops->napi_clear_event)(dev);
 
-	do_wake = do_restart = 0;
 	while (((sc = CBDR_SC(bdp)) & BD_ENET_TX_READY) == 0 && tx_left) {
 		dirtyidx = bdp - fep->tx_bd_base;
 
@@ -111,12 +112,9 @@ static int fs_enet_napi(struct napi_struct *napi, int budget)
 
 		skb = fep->tx_skbuff[dirtyidx];
 
-		/*
-		 * Check for errors.
-		 */
+		 /* Check for errors. */
 		if (sc & (BD_ENET_TX_HB | BD_ENET_TX_LC |
 			  BD_ENET_TX_RL | BD_ENET_TX_UN | BD_ENET_TX_CSL)) {
-
 			if (sc & BD_ENET_TX_HB)	/* No heartbeat */
 				dev->stats.tx_heartbeat_errors++;
 			if (sc & BD_ENET_TX_LC)	/* Late collision */
@@ -132,16 +130,16 @@ static int fs_enet_napi(struct napi_struct *napi, int budget)
 				dev->stats.tx_errors++;
 				do_restart = 1;
 			}
-		} else
+		} else {
 			dev->stats.tx_packets++;
+		}
 
 		if (sc & BD_ENET_TX_READY) {
 			dev_warn(fep->dev,
 				 "HEY! Enet xmit interrupt and TX_READY.\n");
 		}
 
-		/*
-		 * Deferred means some collisions occurred during transmit,
+		/* Deferred means some collisions occurred during transmit,
 		 * but we eventually sent the packet OK.
 		 */
 		if (sc & BD_ENET_TX_DEF)
@@ -155,25 +153,20 @@ static int fs_enet_napi(struct napi_struct *napi, int budget)
 			dma_unmap_single(fep->dev, CBDR_BUFADDR(bdp),
 					 CBDR_DATLEN(bdp), DMA_TO_DEVICE);
 
-		/*
-		 * Free the sk buffer associated with this last transmit.
-		 */
+		/* Free the sk buffer associated with this last transmit. */
 		if (skb) {
 			dev_kfree_skb(skb);
 			fep->tx_skbuff[dirtyidx] = NULL;
 		}
 
-		/*
-		 * Update pointer to next buffer descriptor to be transmitted.
+		/* Update pointer to next buffer descriptor to be transmitted.
 		 */
 		if ((sc & BD_ENET_TX_WRAP) == 0)
 			bdp++;
 		else
 			bdp = fep->tx_bd_base;
 
-		/*
-		 * Since we have freed up a buffer, the ring is no longer
-		 * full.
+		/* Since we have freed up a buffer, the ring is no longer full.
 		 */
 		if (++fep->tx_free == MAX_SKB_FRAGS)
 			do_wake = 1;
@@ -190,8 +183,7 @@ static int fs_enet_napi(struct napi_struct *napi, int budget)
 	if (do_wake)
 		netif_wake_queue(dev);
 
-	/*
-	 * First, grab all of the stats for the incoming packet.
+	/* First, grab all of the stats for the incoming packet.
 	 * These get messed up if we get called due to a busy condition.
 	 */
 	bdp = fep->cur_rx;
@@ -200,16 +192,13 @@ static int fs_enet_napi(struct napi_struct *napi, int budget)
 	       received < budget) {
 		curidx = bdp - fep->rx_bd_base;
 
-		/*
-		 * Since we have allocated space to hold a complete frame,
+		/* Since we have allocated space to hold a complete frame,
 		 * the last indicator should be set.
 		 */
 		if ((sc & BD_ENET_RX_LAST) == 0)
 			dev_warn(fep->dev, "rcv is not +last\n");
 
-		/*
-		 * Check for errors.
-		 */
+		/* Check for errors. */
 		if (sc & (BD_ENET_RX_LG | BD_ENET_RX_SH | BD_ENET_RX_CL |
 			  BD_ENET_RX_NO | BD_ENET_RX_CR | BD_ENET_RX_OV)) {
 			dev->stats.rx_errors++;
@@ -230,9 +219,7 @@ static int fs_enet_napi(struct napi_struct *napi, int budget)
 		} else {
 			skb = fep->rx_skbuff[curidx];
 
-			/*
-			 * Process the incoming frame.
-			 */
+			/* Process the incoming frame */
 			dev->stats.rx_packets++;
 			pkt_len = CBDR_DATLEN(bdp) - 4;	/* remove CRC */
 			dev->stats.rx_bytes += pkt_len + 4;
@@ -240,15 +227,15 @@ static int fs_enet_napi(struct napi_struct *napi, int budget)
 			if (pkt_len <= fpi->rx_copybreak) {
 				/* +2 to make IP header L1 cache aligned */
 				skbn = netdev_alloc_skb(dev, pkt_len + 2);
-				if (skbn != NULL) {
+				if (skbn) {
 					skb_reserve(skbn, 2);	/* align IP header */
-					skb_copy_from_linear_data(skb,
-						      skbn->data, pkt_len);
+					skb_copy_from_linear_data(skb, skbn->data,
+								  pkt_len);
 					swap(skb, skbn);
 					dma_sync_single_for_cpu(fep->dev,
-						CBDR_BUFADDR(bdp),
-						L1_CACHE_ALIGN(pkt_len),
-						DMA_FROM_DEVICE);
+								CBDR_BUFADDR(bdp),
+								L1_CACHE_ALIGN(pkt_len),
+								DMA_FROM_DEVICE);
 				}
 			} else {
 				skbn = netdev_alloc_skb(dev, ENET_RX_FRSIZE);
@@ -258,20 +245,18 @@ static int fs_enet_napi(struct napi_struct *napi, int budget)
 
 					skb_align(skbn, ENET_RX_ALIGN);
 
-					dma_unmap_single(fep->dev,
-						CBDR_BUFADDR(bdp),
-						L1_CACHE_ALIGN(PKT_MAXBUF_SIZE),
-						DMA_FROM_DEVICE);
+					dma_unmap_single(fep->dev, CBDR_BUFADDR(bdp),
+							 L1_CACHE_ALIGN(PKT_MAXBUF_SIZE),
+							 DMA_FROM_DEVICE);
 
-					dma = dma_map_single(fep->dev,
-						skbn->data,
-						L1_CACHE_ALIGN(PKT_MAXBUF_SIZE),
-						DMA_FROM_DEVICE);
+					dma = dma_map_single(fep->dev, skbn->data,
+							     L1_CACHE_ALIGN(PKT_MAXBUF_SIZE),
+							     DMA_FROM_DEVICE);
 					CBDW_BUFADDR(bdp, dma);
 				}
 			}
 
-			if (skbn != NULL) {
+			if (skbn) {
 				skb_put(skb, pkt_len);	/* Make room */
 				skb->protocol = eth_type_trans(skb, dev);
 				received++;
@@ -286,9 +271,7 @@ static int fs_enet_napi(struct napi_struct *napi, int budget)
 		CBDW_DATLEN(bdp, 0);
 		CBDW_SC(bdp, (sc & ~BD_ENET_RX_STATS) | BD_ENET_RX_EMPTY);
 
-		/*
-		 * Update BD pointer to next entry.
-		 */
+		/* Update BD pointer to next entry */
 		if ((sc & BD_ENET_RX_WRAP) == 0)
 			bdp++;
 		else
@@ -310,23 +293,18 @@ static int fs_enet_napi(struct napi_struct *napi, int budget)
 	return budget;
 }
 
-/*
- * The interrupt handler.
+/* The interrupt handler.
  * This is called from the MPC core interrupt.
  */
 static irqreturn_t
 fs_enet_interrupt(int irq, void *dev_id)
 {
 	struct net_device *dev = dev_id;
+	u32 int_events, int_clr_events;
 	struct fs_enet_private *fep;
-	const struct fs_platform_info *fpi;
-	u32 int_events;
-	u32 int_clr_events;
-	int nr, napi_ok;
-	int handled;
+	int nr, napi_ok, handled;
 
 	fep = netdev_priv(dev);
-	fpi = fep->fpi;
 
 	nr = 0;
 	while ((int_events = (*fep->ops->get_int_events)(dev)) != 0) {
@@ -346,12 +324,12 @@ fs_enet_interrupt(int irq, void *dev_id)
 			(*fep->ops->napi_disable)(dev);
 			(*fep->ops->clear_int_events)(dev, fep->ev_napi);
 
-			/* NOTE: it is possible for FCCs in NAPI mode    */
-			/* to submit a spurious interrupt while in poll  */
+			/* NOTE: it is possible for FCCs in NAPI mode
+			 * to submit a spurious interrupt while in poll
+			 */
 			if (napi_ok)
 				__napi_schedule(&fep->napi);
 		}
-
 	}
 
 	handled = nr > 0;
@@ -361,45 +339,40 @@ fs_enet_interrupt(int irq, void *dev_id)
 void fs_init_bds(struct net_device *dev)
 {
 	struct fs_enet_private *fep = netdev_priv(dev);
-	cbd_t __iomem *bdp;
 	struct sk_buff *skb;
+	cbd_t __iomem *bdp;
 	int i;
 
 	fs_cleanup_bds(dev);
 
-	fep->dirty_tx = fep->cur_tx = fep->tx_bd_base;
+	fep->dirty_tx = fep->tx_bd_base;
+	fep->cur_tx = fep->tx_bd_base;
 	fep->tx_free = fep->tx_ring;
 	fep->cur_rx = fep->rx_bd_base;
 
-	/*
-	 * Initialize the receive buffer descriptors.
-	 */
+	/* Initialize the receive buffer descriptors */
 	for (i = 0, bdp = fep->rx_bd_base; i < fep->rx_ring; i++, bdp++) {
 		skb = netdev_alloc_skb(dev, ENET_RX_FRSIZE);
-		if (skb == NULL)
+		if (!skb)
 			break;
 
 		skb_align(skb, ENET_RX_ALIGN);
 		fep->rx_skbuff[i] = skb;
-		CBDW_BUFADDR(bdp,
-			dma_map_single(fep->dev, skb->data,
-				L1_CACHE_ALIGN(PKT_MAXBUF_SIZE),
-				DMA_FROM_DEVICE));
+		CBDW_BUFADDR(bdp, dma_map_single(fep->dev, skb->data,
+						 L1_CACHE_ALIGN(PKT_MAXBUF_SIZE),
+						 DMA_FROM_DEVICE));
 		CBDW_DATLEN(bdp, 0);	/* zero */
 		CBDW_SC(bdp, BD_ENET_RX_EMPTY |
 			((i < fep->rx_ring - 1) ? 0 : BD_SC_WRAP));
 	}
-	/*
-	 * if we failed, fillup remainder
-	 */
+
+	/* if we failed, fillup remainder */
 	for (; i < fep->rx_ring; i++, bdp++) {
 		fep->rx_skbuff[i] = NULL;
 		CBDW_SC(bdp, (i < fep->rx_ring - 1) ? 0 : BD_SC_WRAP);
 	}
 
-	/*
-	 * ...and the same for transmit.
-	 */
+	/* ...and the same for transmit. */
 	for (i = 0, bdp = fep->tx_bd_base; i < fep->tx_ring; i++, bdp++) {
 		fep->tx_skbuff[i] = NULL;
 		CBDW_BUFADDR(bdp, 0);
@@ -415,32 +388,30 @@ void fs_cleanup_bds(struct net_device *dev)
 	cbd_t __iomem *bdp;
 	int i;
 
-	/*
-	 * Reset SKB transmit buffers.
-	 */
+	/* Reset SKB transmit buffers. */
 	for (i = 0, bdp = fep->tx_bd_base; i < fep->tx_ring; i++, bdp++) {
-		if ((skb = fep->tx_skbuff[i]) == NULL)
+		skb = fep->tx_skbuff[i];
+		if (!skb)
 			continue;
 
 		/* unmap */
 		dma_unmap_single(fep->dev, CBDR_BUFADDR(bdp),
-				skb->len, DMA_TO_DEVICE);
+				 skb->len, DMA_TO_DEVICE);
 
 		fep->tx_skbuff[i] = NULL;
 		dev_kfree_skb(skb);
 	}
 
-	/*
-	 * Reset SKB receive buffers
-	 */
+	/* Reset SKB receive buffers */
 	for (i = 0, bdp = fep->rx_bd_base; i < fep->rx_ring; i++, bdp++) {
-		if ((skb = fep->rx_skbuff[i]) == NULL)
+		skb = fep->rx_skbuff[i];
+		if (!skb)
 			continue;
 
 		/* unmap */
 		dma_unmap_single(fep->dev, CBDR_BUFADDR(bdp),
-			L1_CACHE_ALIGN(PKT_MAXBUF_SIZE),
-			DMA_FROM_DEVICE);
+				 L1_CACHE_ALIGN(PKT_MAXBUF_SIZE),
+				 DMA_FROM_DEVICE);
 
 		fep->rx_skbuff[i] = NULL;
 
@@ -448,12 +419,8 @@ void fs_cleanup_bds(struct net_device *dev)
 	}
 }
 
-/**********************************************************************************/
-
 #ifdef CONFIG_FS_ENET_MPC5121_FEC
-/*
- * MPC5121 FEC requeries 4-byte alignment for TX data buffer!
- */
+/* MPC5121 FEC requires 4-byte alignment for TX data buffer! */
 static struct sk_buff *tx_skb_align_workaround(struct net_device *dev,
 					       struct sk_buff *skb)
 {
@@ -481,18 +448,16 @@ static struct sk_buff *tx_skb_align_workaround(struct net_device *dev,
 }
 #endif
 
-static int fs_enet_start_xmit(struct sk_buff *skb, struct net_device *dev)
+static netdev_tx_t
+fs_enet_start_xmit(struct sk_buff *skb, struct net_device *dev)
 {
 	struct fs_enet_private *fep = netdev_priv(dev);
+	int curidx, nr_frags, len;
 	cbd_t __iomem *bdp;
-	int curidx;
-	u16 sc;
-	int nr_frags;
 	skb_frag_t *frag;
-	int len;
+	u16 sc;
 #ifdef CONFIG_FS_ENET_MPC5121_FEC
-	int is_aligned = 1;
-	int i;
+	int i, is_aligned = 1;
 
 	if (!IS_ALIGNED((unsigned long)skb->data, 4)) {
 		is_aligned = 0;
@@ -500,7 +465,7 @@ static int fs_enet_start_xmit(struct sk_buff *skb, struct net_device *dev)
 		nr_frags = skb_shinfo(skb)->nr_frags;
 		frag = skb_shinfo(skb)->frags;
 		for (i = 0; i < nr_frags; i++, frag++) {
-			if (!IS_ALIGNED(frag->page_offset, 4)) {
+			if (!IS_ALIGNED(skb_frag_off(frag), 4)) {
 				is_aligned = 0;
 				break;
 			}
@@ -510,8 +475,7 @@ static int fs_enet_start_xmit(struct sk_buff *skb, struct net_device *dev)
 	if (!is_aligned) {
 		skb = tx_skb_align_workaround(dev, skb);
 		if (!skb) {
-			/*
-			 * We have lost packet due to memory allocation error
+			/* We have lost packet due to memory allocation error
 			 * in tx_skb_align_workaround(). Hopefully original
 			 * skb is still valid, so try transmit it later.
 			 */
@@ -522,9 +486,7 @@ static int fs_enet_start_xmit(struct sk_buff *skb, struct net_device *dev)
 
 	spin_lock(&fep->tx_lock);
 
-	/*
-	 * Fill in a Tx ring entry
-	 */
+	/* Fill in a Tx ring entry */
 	bdp = fep->cur_tx;
 
 	nr_frags = skb_shinfo(skb)->nr_frags;
@@ -532,8 +494,7 @@ static int fs_enet_start_xmit(struct sk_buff *skb, struct net_device *dev)
 		netif_stop_queue(dev);
 		spin_unlock(&fep->tx_lock);
 
-		/*
-		 * Ooops.  All transmit buffers are full.  Bail out.
+		/* Ooops.  All transmit buffers are full.  Bail out.
 		 * This should not happen, since the tx queue should be stopped.
 		 */
 		dev_warn(fep->dev, "tx queue full!.\n");
@@ -546,12 +507,12 @@ static int fs_enet_start_xmit(struct sk_buff *skb, struct net_device *dev)
 	dev->stats.tx_bytes += len;
 	if (nr_frags)
 		len -= skb->data_len;
+
 	fep->tx_free -= nr_frags + 1;
-	/*
-	 * Push the data cache so the CPM does not get stale memory data.
+	/* Push the data cache so the CPM does not get stale memory data.
 	 */
 	CBDW_BUFADDR(bdp, dma_map_single(fep->dev,
-				skb->data, len, DMA_TO_DEVICE));
+					 skb->data, len, DMA_TO_DEVICE));
 	CBDW_DATLEN(bdp, len);
 
 	fep->mapped_as_page[curidx] = 0;
@@ -562,10 +523,13 @@ static int fs_enet_start_xmit(struct sk_buff *skb, struct net_device *dev)
 			BD_ENET_TX_TC);
 		CBDS_SC(bdp, BD_ENET_TX_READY);
 
-		if ((CBDR_SC(bdp) & BD_ENET_TX_WRAP) == 0)
-			bdp++, curidx++;
-		else
-			bdp = fep->tx_bd_base, curidx = 0;
+		if ((CBDR_SC(bdp) & BD_ENET_TX_WRAP) == 0) {
+			bdp++;
+			curidx++;
+		} else {
+			bdp = fep->tx_bd_base;
+			curidx = 0;
+		}
 
 		len = skb_frag_size(frag);
 		CBDW_BUFADDR(bdp, skb_frag_dma_map(fep->dev, frag, 0, len,
@@ -585,9 +549,11 @@ static int fs_enet_start_xmit(struct sk_buff *skb, struct net_device *dev)
 
 	/* note that while FEC does not have this bit
 	 * it marks it as available for software use
-	 * yay for hw reuse :) */
+	 * yay for hw reuse :)
+	 */
 	if (skb->len <= 60)
 		sc |= BD_ENET_TX_PAD;
+
 	CBDC_SC(bdp, BD_ENET_TX_STATS);
 	CBDS_SC(bdp, sc);
 
@@ -599,6 +565,7 @@ static int fs_enet_start_xmit(struct sk_buff *skb, struct net_device *dev)
 		bdp++;
 	else
 		bdp = fep->tx_bd_base;
+
 	fep->cur_tx = bdp;
 
 	if (fep->tx_free < MAX_SKB_FRAGS)
@@ -613,24 +580,31 @@ static int fs_enet_start_xmit(struct sk_buff *skb, struct net_device *dev)
 	return NETDEV_TX_OK;
 }
 
-static void fs_timeout(struct net_device *dev)
+static void fs_timeout_work(struct work_struct *work)
 {
-	struct fs_enet_private *fep = netdev_priv(dev);
+	struct fs_enet_private *fep = container_of(work, struct fs_enet_private,
+						   timeout_work);
+	struct net_device *dev = fep->ndev;
 	unsigned long flags;
 	int wake = 0;
 
 	dev->stats.tx_errors++;
 
-	spin_lock_irqsave(&fep->lock, flags);
+	/* In the event a timeout was detected, but the netdev is brought down
+	 * shortly after, it no longer makes sense to try to recover from the
+	 * timeout. netif_running() will return false when called from the
+	 * .ndo_close() callback. Calling the following recovery code while
+	 * called from .ndo_close() could deadlock on rtnl.
+	 */
+	if (!netif_running(dev))
+		return;
 
-	if (dev->flags & IFF_UP) {
-		phy_stop(dev->phydev);
-		(*fep->ops->stop)(dev);
-		(*fep->ops->restart)(dev);
-		phy_start(dev->phydev);
-	}
+	rtnl_lock();
+	phylink_stop(fep->phylink);
+	phylink_start(fep->phylink);
+	rtnl_unlock();
 
-	phy_start(dev->phydev);
+	spin_lock_irqsave(&fep->lock, flags);
 	wake = fep->tx_free >= MAX_SKB_FRAGS &&
 	       !(CBDR_SC(fep->cur_tx) & BD_ENET_TX_READY);
 	spin_unlock_irqrestore(&fep->lock, flags);
@@ -639,82 +613,44 @@ static void fs_timeout(struct net_device *dev)
 		netif_wake_queue(dev);
 }
 
-/*-----------------------------------------------------------------------------
- *  generic link-change handler - should be sufficient for most cases
- *-----------------------------------------------------------------------------*/
-static void generic_adjust_link(struct  net_device *dev)
+static void fs_timeout(struct net_device *dev, unsigned int txqueue)
 {
 	struct fs_enet_private *fep = netdev_priv(dev);
-	struct phy_device *phydev = dev->phydev;
-	int new_state = 0;
-
-	if (phydev->link) {
-		/* adjust to duplex mode */
-		if (phydev->duplex != fep->oldduplex) {
-			new_state = 1;
-			fep->oldduplex = phydev->duplex;
-		}
-
-		if (phydev->speed != fep->oldspeed) {
-			new_state = 1;
-			fep->oldspeed = phydev->speed;
-		}
 
-		if (!fep->oldlink) {
-			new_state = 1;
-			fep->oldlink = 1;
-		}
-
-		if (new_state)
-			fep->ops->restart(dev);
-	} else if (fep->oldlink) {
-		new_state = 1;
-		fep->oldlink = 0;
-		fep->oldspeed = 0;
-		fep->oldduplex = -1;
-	}
-
-	if (new_state && netif_msg_link(fep))
-		phy_print_status(phydev);
+	schedule_work(&fep->timeout_work);
 }
 
-
-static void fs_adjust_link(struct net_device *dev)
+static void fs_mac_link_up(struct phylink_config *config,
+			   struct phy_device *phy,
+			   unsigned int mode, phy_interface_t interface,
+			   int speed, int duplex,
+			   bool tx_pause, bool rx_pause)
 {
-	struct fs_enet_private *fep = netdev_priv(dev);
+	struct net_device *ndev = to_net_dev(config->dev);
+	struct fs_enet_private *fep = netdev_priv(ndev);
 	unsigned long flags;
 
 	spin_lock_irqsave(&fep->lock, flags);
-
-	if(fep->ops->adjust_link)
-		fep->ops->adjust_link(dev);
-	else
-		generic_adjust_link(dev);
-
+	fep->ops->restart(ndev, interface, speed, duplex);
 	spin_unlock_irqrestore(&fep->lock, flags);
 }
 
-static int fs_init_phy(struct net_device *dev)
+static void fs_mac_link_down(struct phylink_config *config,
+			     unsigned int mode, phy_interface_t interface)
 {
-	struct fs_enet_private *fep = netdev_priv(dev);
-	struct phy_device *phydev;
-	phy_interface_t iface;
-
-	fep->oldlink = 0;
-	fep->oldspeed = 0;
-	fep->oldduplex = -1;
-
-	iface = fep->fpi->use_rmii ?
-		PHY_INTERFACE_MODE_RMII : PHY_INTERFACE_MODE_MII;
+	struct net_device *ndev = to_net_dev(config->dev);
+	struct fs_enet_private *fep = netdev_priv(ndev);
+	unsigned long flags;
 
-	phydev = of_phy_connect(dev, fep->fpi->phy_node, &fs_adjust_link, 0,
-				iface);
-	if (!phydev) {
-		dev_err(&dev->dev, "Could not attach to PHY\n");
-		return -ENODEV;
-	}
+	spin_lock_irqsave(&fep->lock, flags);
+	fep->ops->stop(ndev);
+	spin_unlock_irqrestore(&fep->lock, flags);
+}
 
-	return 0;
+static void fs_mac_config(struct phylink_config *config, unsigned int mode,
+			  const struct phylink_link_state *state)
+{
+	/* Nothing to do */
 }
 
 static int fs_enet_open(struct net_device *dev)
@@ -723,8 +659,9 @@ static int fs_enet_open(struct net_device *dev)
 	int r;
 	int err;
 
-	/* to initialize the fep->cur_rx,... */
-	/* not doing this, will cause a crash in fs_enet_napi */
+	/* to initialize the fep->cur_rx,...
+	 * not doing this, will cause a crash in fs_enet_napi
+	 */
 	fs_init_bds(fep->ndev);
 
 	napi_enable(&fep->napi);
@@ -738,13 +675,13 @@ static int fs_enet_open(struct net_device *dev)
 		return -EINVAL;
 	}
 
-	err = fs_init_phy(dev);
+	err = phylink_of_phy_connect(fep->phylink, fep->dev->of_node, 0);
 	if (err) {
 		free_irq(fep->interrupt, dev);
 		napi_disable(&fep->napi);
 		return err;
 	}
-	phy_start(dev->phydev);
+	phylink_start(fep->phylink);
 
 	netif_start_queue(dev);
 
@@ -757,30 +694,27 @@ static int fs_enet_close(struct net_device *dev)
 	unsigned long flags;
 
 	netif_stop_queue(dev);
-	netif_carrier_off(dev);
 	napi_disable(&fep->napi);
-	phy_stop(dev->phydev);
+	cancel_work(&fep->timeout_work);
+	phylink_stop(fep->phylink);
 
 	spin_lock_irqsave(&fep->lock, flags);
 	spin_lock(&fep->tx_lock);
 	(*fep->ops->stop)(dev);
 	spin_unlock(&fep->tx_lock);
 	spin_unlock_irqrestore(&fep->lock, flags);
+	phylink_disconnect_phy(fep->phylink);
 
 	/* release any irqs */
-	phy_disconnect(dev->phydev);
 	free_irq(fep->interrupt, dev);
 
 	return 0;
 }
 
-/*************************************************************************/
-
 static void fs_get_drvinfo(struct net_device *dev,
-			    struct ethtool_drvinfo *info)
+			   struct ethtool_drvinfo *info)
 {
-	strlcpy(info->driver, DRV_MODULE_NAME, sizeof(info->driver));
-	strlcpy(info->version, DRV_MODULE_VERSION, sizeof(info->version));
+	strscpy(info->driver, DRV_MODULE_NAME, sizeof(info->driver));
 }
 
 static int fs_get_regs_len(struct net_device *dev)
@@ -791,7 +725,7 @@ static int fs_get_regs_len(struct net_device *dev)
 }
 
 static void fs_get_regs(struct net_device *dev, struct ethtool_regs *regs,
-			 void *p)
+			void *p)
 {
 	struct fs_enet_private *fep = netdev_priv(dev);
 	unsigned long flags;
@@ -810,12 +744,14 @@ static void fs_get_regs(struct net_device *dev, struct ethtool_regs *regs,
 static u32 fs_get_msglevel(struct net_device *dev)
 {
 	struct fs_enet_private *fep = netdev_priv(dev);
+
 	return fep->msg_enable;
 }
 
 static void fs_set_msglevel(struct net_device *dev, u32 value)
 {
 	struct fs_enet_private *fep = netdev_priv(dev);
+
 	fep->msg_enable = value;
 }
 
@@ -857,6 +793,22 @@ static int fs_set_tunable(struct net_device *dev,
 	return ret;
 }
 
+static int fs_ethtool_set_link_ksettings(struct net_device *dev,
+					 const struct ethtool_link_ksettings *cmd)
+{
+	struct fs_enet_private *fep = netdev_priv(dev);
+
+	return phylink_ethtool_ksettings_set(fep->phylink, cmd);
+}
+
+static int fs_ethtool_get_link_ksettings(struct net_device *dev,
+					 struct ethtool_link_ksettings *cmd)
+{
+	struct fs_enet_private *fep = netdev_priv(dev);
+
+	return phylink_ethtool_ksettings_get(fep->phylink, cmd);
+}
+
 static const struct ethtool_ops fs_ethtool_ops = {
 	.get_drvinfo = fs_get_drvinfo,
 	.get_regs_len = fs_get_regs_len,
@@ -866,29 +818,16 @@ static const struct ethtool_ops fs_ethtool_ops = {
 	.set_msglevel = fs_set_msglevel,
 	.get_regs = fs_get_regs,
 	.get_ts_info = ethtool_op_get_ts_info,
-	.get_link_ksettings = phy_ethtool_get_link_ksettings,
-	.set_link_ksettings = phy_ethtool_set_link_ksettings,
+	.get_link_ksettings = fs_ethtool_get_link_ksettings,
+	.set_link_ksettings = fs_ethtool_set_link_ksettings,
 	.get_tunable = fs_get_tunable,
 	.set_tunable = fs_set_tunable,
 };
 
-static int fs_ioctl(struct net_device *dev, struct ifreq *rq, int cmd)
-{
-	if (!netif_running(dev))
-		return -EINVAL;
-
-	return phy_mii_ioctl(dev->phydev, rq, cmd);
-}
-
-extern int fs_mii_connect(struct net_device *dev);
-extern void fs_mii_disconnect(struct net_device *dev);
-
-/**************************************************************************************/
-
 #ifdef CONFIG_FS_ENET_HAS_FEC
-#define IS_FEC(match) ((match)->data == &fs_fec_ops)
+#define IS_FEC(ops) ((ops) == &fs_fec_ops)
 #else
-#define IS_FEC(match) 0
+#define IS_FEC(ops) 0
 #endif
 
 static const struct net_device_ops fs_enet_netdev_ops = {
@@ -897,7 +836,7 @@ static const struct net_device_ops fs_enet_netdev_ops = {
 	.ndo_start_xmit		= fs_enet_start_xmit,
 	.ndo_tx_timeout		= fs_timeout,
 	.ndo_set_rx_mode	= fs_set_multicast_list,
-	.ndo_do_ioctl		= fs_ioctl,
+	.ndo_eth_ioctl		= fs_eth_ioctl,
 	.ndo_validate_addr	= eth_validate_addr,
 	.ndo_set_mac_address	= eth_mac_addr,
 #ifdef CONFIG_NET_POLL_CONTROLLER
@@ -905,29 +844,33 @@ static const struct net_device_ops fs_enet_netdev_ops = {
 #endif
 };
 
-static const struct of_device_id fs_enet_match[];
+static const struct phylink_mac_ops fs_enet_phylink_mac_ops = {
+	.mac_config = fs_mac_config,
+	.mac_link_down = fs_mac_link_down,
+	.mac_link_up = fs_mac_link_up,
+};
+
 static int fs_enet_probe(struct platform_device *ofdev)
 {
-	const struct of_device_id *match;
-	struct net_device *ndev;
-	struct fs_enet_private *fep;
+	int privsize, len, ret = -ENODEV;
 	struct fs_platform_info *fpi;
+	struct fs_enet_private *fep;
+	phy_interface_t phy_mode;
+	const struct fs_ops *ops;
+	struct net_device *ndev;
+	struct phylink *phylink;
 	const u32 *data;
 	struct clk *clk;
-	int err;
-	const u8 *mac_addr;
-	const char *phy_connection_type;
-	int privsize, len, ret = -ENODEV;
 
-	match = of_match_device(fs_enet_match, &ofdev->dev);
-	if (!match)
+	ops = device_get_match_data(&ofdev->dev);
+	if (!ops)
 		return -EINVAL;
 
 	fpi = kzalloc(sizeof(*fpi), GFP_KERNEL);
 	if (!fpi)
 		return -ENOMEM;
 
-	if (!IS_FEC(match)) {
+	if (!IS_FEC(ops)) {
 		data = of_get_property(ofdev->dev.of_node, "fsl,cpm-command", &len);
 		if (!data || len != 4)
 			goto out_free_fpi;
@@ -935,51 +878,36 @@ static int fs_enet_probe(struct platform_device *ofdev)
 		fpi->cp_command = *data;
 	}
 
+	ret = of_get_phy_mode(ofdev->dev.of_node, &phy_mode);
+	if (ret) {
+		/* For compatibility, if the mode isn't specified in DT,
+		 * assume MII
+		 */
+		phy_mode = PHY_INTERFACE_MODE_MII;
+	}
+
 	fpi->rx_ring = RX_RING_SIZE;
 	fpi->tx_ring = TX_RING_SIZE;
 	fpi->rx_copybreak = 240;
 	fpi->napi_weight = 17;
-	fpi->phy_node = of_parse_phandle(ofdev->dev.of_node, "phy-handle", 0);
-	if (!fpi->phy_node && of_phy_is_fixed_link(ofdev->dev.of_node)) {
-		err = of_phy_register_fixed_link(ofdev->dev.of_node);
-		if (err)
-			goto out_free_fpi;
-
-		/* In the case of a fixed PHY, the DT node associated
-		 * to the PHY is the Ethernet MAC DT node.
-		 */
-		fpi->phy_node = of_node_get(ofdev->dev.of_node);
-	}
-
-	if (of_device_is_compatible(ofdev->dev.of_node, "fsl,mpc5125-fec")) {
-		phy_connection_type = of_get_property(ofdev->dev.of_node,
-						"phy-connection-type", NULL);
-		if (phy_connection_type && !strcmp("rmii", phy_connection_type))
-			fpi->use_rmii = 1;
-	}
 
 	/* make clock lookup non-fatal (the driver is shared among platforms),
 	 * but require enable to succeed when a clock was specified/found,
 	 * keep a reference to the clock upon successful acquisition
 	 */
-	clk = devm_clk_get(&ofdev->dev, "per");
-	if (!IS_ERR(clk)) {
-		ret = clk_prepare_enable(clk);
-		if (ret)
-			goto out_deregister_fixed_link;
-
-		fpi->clk_per = clk;
-	}
+	clk = devm_clk_get_optional_enabled(&ofdev->dev, "per");
+	if (IS_ERR(clk))
+		goto out_free_fpi;
 
 	privsize = sizeof(*fep) +
-	           sizeof(struct sk_buff **) *
+		   sizeof(struct sk_buff **) *
 		     (fpi->rx_ring + fpi->tx_ring) +
 		   sizeof(char) * fpi->tx_ring;
 
 	ndev = alloc_etherdev(privsize);
 	if (!ndev) {
 		ret = -ENOMEM;
-		goto out_put;
+		goto out_free_fpi;
 	}
 
 	SET_NETDEV_DEV(ndev, &ofdev->dev);
@@ -989,11 +917,31 @@ static int fs_enet_probe(struct platform_device *ofdev)
 	fep->dev = &ofdev->dev;
 	fep->ndev = ndev;
 	fep->fpi = fpi;
-	fep->ops = match->data;
+	fep->ops = ops;
+
+	fep->phylink_config.dev = &ndev->dev;
+	fep->phylink_config.type = PHYLINK_NETDEV;
+	fep->phylink_config.mac_capabilities = MAC_10 | MAC_100;
+
+	__set_bit(PHY_INTERFACE_MODE_MII,
+		  fep->phylink_config.supported_interfaces);
+
+	if (of_device_is_compatible(ofdev->dev.of_node, "fsl,mpc5125-fec"))
+		__set_bit(PHY_INTERFACE_MODE_RMII,
+			  fep->phylink_config.supported_interfaces);
+
+	phylink = phylink_create(&fep->phylink_config, dev_fwnode(fep->dev),
+				 phy_mode, &fs_enet_phylink_mac_ops);
+	if (IS_ERR(phylink)) {
+		ret = PTR_ERR(phylink);
+		goto out_free_dev;
+	}
+
+	fep->phylink = phylink;
 
 	ret = fep->ops->setup_data(ndev);
 	if (ret)
-		goto out_free_dev;
+		goto out_phylink;
 
 	fep->rx_skbuff = (struct sk_buff **)&fep[1];
 	fep->tx_skbuff = fep->rx_skbuff + fpi->rx_ring;
@@ -1003,9 +951,7 @@ static int fs_enet_probe(struct platform_device *ofdev)
 	spin_lock_init(&fep->lock);
 	spin_lock_init(&fep->tx_lock);
 
-	mac_addr = of_get_mac_address(ofdev->dev.of_node);
-	if (mac_addr)
-		memcpy(ndev->dev_addr, mac_addr, ETH_ALEN);
+	of_get_ethdev_address(ofdev->dev.of_node, ndev);
 
 	ret = fep->ops->allocate_bd(ndev);
 	if (ret)
@@ -1019,14 +965,12 @@ static int fs_enet_probe(struct platform_device *ofdev)
 
 	ndev->netdev_ops = &fs_enet_netdev_ops;
 	ndev->watchdog_timeo = 2 * HZ;
-	netif_napi_add(ndev, &fep->napi, fs_enet_napi, fpi->napi_weight);
+	INIT_WORK(&fep->timeout_work, fs_timeout_work);
+	netif_napi_add_weight(ndev, &fep->napi, fs_enet_napi,
+			      fpi->napi_weight);
 
 	ndev->ethtool_ops = &fs_ethtool_ops;
 
-	init_timer(&fep->phy_timer_list);
-
-	netif_carrier_off(ndev);
-
 	ndev->features |= NETIF_F_SG;
 
 	ret = register_netdev(ndev);
@@ -1041,21 +985,16 @@ out_free_bd:
 	fep->ops->free_bd(ndev);
 out_cleanup_data:
 	fep->ops->cleanup_data(ndev);
+out_phylink:
+	phylink_destroy(fep->phylink);
 out_free_dev:
 	free_netdev(ndev);
-out_put:
-	if (fpi->clk_per)
-		clk_disable_unprepare(fpi->clk_per);
-out_deregister_fixed_link:
-	of_node_put(fpi->phy_node);
-	if (of_phy_is_fixed_link(ofdev->dev.of_node))
-		of_phy_deregister_fixed_link(ofdev->dev.of_node);
 out_free_fpi:
 	kfree(fpi);
 	return ret;
 }
 
-static int fs_enet_remove(struct platform_device *ofdev)
+static void fs_enet_remove(struct platform_device *ofdev)
 {
 	struct net_device *ndev = platform_get_drvdata(ofdev);
 	struct fs_enet_private *fep = netdev_priv(ndev);
@@ -1065,13 +1004,8 @@ static int fs_enet_remove(struct platform_device *ofdev)
 	fep->ops->free_bd(ndev);
 	fep->ops->cleanup_data(ndev);
 	dev_set_drvdata(fep->dev, NULL);
-	of_node_put(fep->fpi->phy_node);
-	if (fep->fpi->clk_per)
-		clk_disable_unprepare(fep->fpi->clk_per);
-	if (of_phy_is_fixed_link(ofdev->dev.of_node))
-		of_phy_deregister_fixed_link(ofdev->dev.of_node);
+	phylink_destroy(fep->phylink);
 	free_netdev(ndev);
-	return 0;
 }
 
 static const struct of_device_id fs_enet_match[] = {
@@ -1124,9 +1058,9 @@ static struct platform_driver fs_enet_driver = {
 #ifdef CONFIG_NET_POLL_CONTROLLER
 static void fs_enet_netpoll(struct net_device *dev)
 {
-       disable_irq(dev->irq);
-       fs_enet_interrupt(dev->irq, dev);
-       enable_irq(dev->irq);
+	disable_irq(dev->irq);
+	fs_enet_interrupt(dev->irq, dev);
+	enable_irq(dev->irq);
 }
 #endif
 
diff --git a/drivers/net/ethernet/freescale/fs_enet/fs_enet.h b/drivers/net/ethernet/freescale/fs_enet/fs_enet.h
index 5ce516c8a62a..36e4fcc29e36 100644
--- a/drivers/net/ethernet/freescale/fs_enet/fs_enet.h
+++ b/drivers/net/ethernet/freescale/fs_enet/fs_enet.h
@@ -1,16 +1,15 @@
+/* SPDX-License-Identifier: GPL-2.0 */
 #ifndef FS_ENET_H
 #define FS_ENET_H
 
-#include <linux/mii.h>
+#include <linux/clk.h>
 #include <linux/netdevice.h>
 #include <linux/types.h>
 #include <linux/list.h>
 #include <linux/phy.h>
+#include <linux/phylink.h>
 #include <linux/dma-mapping.h>
 
-#include <linux/fs_enet_pd.h>
-#include <asm/fs_pd.h>
-
 #ifdef CONFIG_CPM1
 #include <asm/cpm1.h>
 #endif
@@ -78,8 +77,8 @@ struct fs_ops {
 	void (*free_bd)(struct net_device *dev);
 	void (*cleanup_data)(struct net_device *dev);
 	void (*set_multicast_list)(struct net_device *dev);
-	void (*adjust_link)(struct net_device *dev);
-	void (*restart)(struct net_device *dev);
+	void (*restart)(struct net_device *dev, phy_interface_t interface,
+			int speed, int duplex);
 	void (*stop)(struct net_device *dev);
 	void (*napi_clear_event)(struct net_device *dev);
 	void (*napi_enable)(struct net_device *dev);
@@ -94,14 +93,6 @@ struct fs_ops {
 	void (*tx_restart)(struct net_device *dev);
 };
 
-struct phy_info {
-	unsigned int id;
-	const char *name;
-	void (*startup) (struct net_device * dev);
-	void (*shutdown) (struct net_device * dev);
-	void (*ack_int) (struct net_device * dev);
-};
-
 /* The FEC stores dest/src/type, data, and checksum for receive packets.
  */
 #define MAX_MTU 1508		/* Allow fullsized pppoe packets over VLAN */
@@ -117,6 +108,17 @@ struct phy_info {
 #define ENET_RX_ALIGN  16
 #define ENET_RX_FRSIZE L1_CACHE_ALIGN(PKT_MAXBUF_SIZE + ENET_RX_ALIGN - 1)
 
+struct fs_platform_info {
+	/* device specific information */
+	u32 cp_command;		/* CPM page/sblock/mcn */
+
+	u32 dpram_offset;
+
+	int rx_ring, tx_ring;	/* number of buffers on rx	*/
+	int rx_copybreak;	/* limit we copy small frames	*/
+	int napi_weight;	/* NAPI weight			*/
+};
+
 struct fs_enet_private {
 	struct napi_struct napi;
 	struct device *dev;	/* pointer back to the device (must be initialized first) */
@@ -124,6 +126,7 @@ struct fs_enet_private {
 	spinlock_t lock;	/* during all ops except TX pckt processing */
 	spinlock_t tx_lock;	/* during fs_start_xmit and fs_tx         */
 	struct fs_platform_info *fpi;
+	struct work_struct timeout_work;
 	const struct fs_ops *ops;
 	int rx_ring, tx_ring;
 	dma_addr_t ring_mem_addr;
@@ -137,15 +140,11 @@ struct fs_enet_private {
 	cbd_t __iomem *cur_rx;
 	cbd_t __iomem *cur_tx;
 	int tx_free;
-	struct timer_list phy_timer_list;
-	const struct phy_info *phy;
 	u32 msg_enable;
-	struct mii_if_info mii_if;
-	unsigned int last_mii_status;
+	struct phylink *phylink;
+	struct phylink_config phylink_config;
 	int interrupt;
 
-	int oldduplex, oldspeed, oldlink;	/* current settings */
-
 	/* event masks */
 	u32 ev_napi;		/* mask of NAPI events */
 	u32 ev;			/* event mask          */
@@ -189,20 +188,13 @@ void fs_cleanup_bds(struct net_device *dev);
 
 #define DRV_MODULE_NAME		"fs_enet"
 #define PFX DRV_MODULE_NAME	": "
-#define DRV_MODULE_VERSION	"1.1"
-#define DRV_MODULE_RELDATE	"Sep 22, 2014"
-
-/***************************************************************************/
-
-int fs_enet_platform_init(void);
-void fs_enet_platform_cleanup(void);
 
 /***************************************************************************/
 /* buffer descriptor access macros */
 
 /* access macros */
 #if defined(CONFIG_CPM1)
-/* for a a CPM1 __raw_xxx's are sufficient */
+/* for a CPM1 __raw_xxx's are sufficient */
 #define __cbd_out32(addr, x)	__raw_writel(x, addr)
 #define __cbd_out16(addr, x)	__raw_writew(x, addr)
 #define __cbd_in32(addr)	__raw_readl(addr)
diff --git a/drivers/net/ethernet/freescale/fs_enet/mac-fcc.c b/drivers/net/ethernet/freescale/fs_enet/mac-fcc.c
index 6e64989f8478..be63293511d9 100644
--- a/drivers/net/ethernet/freescale/fs_enet/mac-fcc.c
+++ b/drivers/net/ethernet/freescale/fs_enet/mac-fcc.c
@@ -1,3 +1,4 @@
+// SPDX-License-Identifier: GPL-2.0-only
 /*
  * FCC driver for Motorola MPC82xx (PQ2).
  *
@@ -6,10 +7,6 @@
  *
  * 2005 (c) MontaVista Software, Inc.
  * Vitaly Bordug <vbordug@ru.mvista.com>
- *
- * This file is licensed under the terms of the GNU General Public License
- * version 2. This program is licensed "as is" without any warranty of any
- * kind, whether express or implied.
  */
 
 #include <linux/module.h>
@@ -25,22 +22,19 @@
 #include <linux/etherdevice.h>
 #include <linux/skbuff.h>
 #include <linux/spinlock.h>
-#include <linux/mii.h>
 #include <linux/ethtool.h>
 #include <linux/bitops.h>
 #include <linux/fs.h>
 #include <linux/platform_device.h>
 #include <linux/phy.h>
 #include <linux/of_address.h>
-#include <linux/of_device.h>
 #include <linux/of_irq.h>
 #include <linux/gfp.h>
+#include <linux/pgtable.h>
 
 #include <asm/immap_cpm2.h>
-#include <asm/mpc8260.h>
 #include <asm/cpm2.h>
 
-#include <asm/pgtable.h>
 #include <asm/irq.h>
 #include <linux/uaccess.h>
 
@@ -106,7 +100,7 @@ static int do_pd_setup(struct fs_enet_private *fep)
 		goto out_ep;
 
 	fep->fcc.mem = (void __iomem *)cpm2_immr;
-	fpi->dpram_offset = cpm_dpalloc(128, 32);
+	fpi->dpram_offset = cpm_muram_alloc(128, 32);
 	if (IS_ERR_VALUE(fpi->dpram_offset)) {
 		ret = fpi->dpram_offset;
 		goto out_fcccp;
@@ -241,7 +235,8 @@ static void set_multicast_list(struct net_device *dev)
 		set_promiscuous_mode(dev);
 }
 
-static void restart(struct net_device *dev)
+static void restart(struct net_device *dev, phy_interface_t interface,
+		    int speed, int duplex)
 {
 	struct fs_enet_private *fep = netdev_priv(dev);
 	const struct fs_platform_info *fpi = fep->fpi;
@@ -365,8 +360,8 @@ static void restart(struct net_device *dev)
 	fs_init_bds(dev);
 
 	/* adjust to speed (for RMII mode) */
-	if (fpi->use_rmii) {
-		if (dev->phydev->speed == 100)
+	if (interface == PHY_INTERFACE_MODE_RMII) {
+		if (speed == SPEED_100)
 			C8(fcccp, fcc_gfemr, 0x20);
 		else
 			S8(fcccp, fcc_gfemr, 0x20);
@@ -388,11 +383,11 @@ static void restart(struct net_device *dev)
 
 	W32(fccp, fcc_fpsmr, FCC_PSMR_ENCRC);
 
-	if (fpi->use_rmii)
+	if (interface == PHY_INTERFACE_MODE_RMII)
 		S32(fccp, fcc_fpsmr, FCC_PSMR_RMII);
 
 	/* adjust to duplex mode */
-	if (dev->phydev->duplex)
+	if (duplex == DUPLEX_FULL)
 		S32(fccp, fcc_fpsmr, FCC_PSMR_FDE | FCC_PSMR_LPB);
 	else
 		C32(fccp, fcc_fpsmr, FCC_PSMR_FDE | FCC_PSMR_LPB);
@@ -548,7 +543,7 @@ static void tx_restart(struct net_device *dev)
 	}
 	/* Now update the TBPTR and dirty flag to the current buffer */
 	W32(ep, fen_genfcc.fcc_tbptr,
-		(uint) (((void *)recheck_bd - fep->ring_base) +
+		(uint)(((void __iomem *)recheck_bd - fep->ring_base) +
 		fep->ring_mem_addr));
 	fep->dirty_tx = recheck_bd;
 
diff --git a/drivers/net/ethernet/freescale/fs_enet/mac-fec.c b/drivers/net/ethernet/freescale/fs_enet/mac-fec.c
index 1fc27c97e3b2..f2ecd20027cf 100644
--- a/drivers/net/ethernet/freescale/fs_enet/mac-fec.c
+++ b/drivers/net/ethernet/freescale/fs_enet/mac-fec.c
@@ -1,3 +1,4 @@
+// SPDX-License-Identifier: GPL-2.0-only
 /*
  * Freescale Ethernet controllers
  *
@@ -6,10 +7,6 @@
  *
  * 2005 (c) MontaVista Software, Inc.
  * Vitaly Bordug <vbordug@ru.mvista.com>
- *
- * This file is licensed under the terms of the GNU General Public License
- * version 2. This program is licensed "as is" without any warranty of any
- * kind, whether express or implied.
  */
 
 #include <linux/module.h>
@@ -18,6 +15,7 @@
 #include <linux/string.h>
 #include <linux/ptrace.h>
 #include <linux/errno.h>
+#include <linux/crc32.h>
 #include <linux/ioport.h>
 #include <linux/interrupt.h>
 #include <linux/delay.h>
@@ -25,13 +23,11 @@
 #include <linux/etherdevice.h>
 #include <linux/skbuff.h>
 #include <linux/spinlock.h>
-#include <linux/mii.h>
 #include <linux/ethtool.h>
 #include <linux/bitops.h>
 #include <linux/fs.h>
 #include <linux/platform_device.h>
 #include <linux/of_address.h>
-#include <linux/of_device.h>
 #include <linux/of_irq.h>
 #include <linux/gfp.h>
 
@@ -97,7 +93,7 @@ static int do_pd_setup(struct fs_enet_private *fep)
 		return -EINVAL;
 
 	fep->fec.fecp = of_iomap(ofdev->dev.of_node, 0);
-	if (!fep->fcc.fccp)
+	if (!fep->fec.fecp)
 		return -EINVAL;
 
 	return 0;
@@ -176,21 +172,10 @@ static void set_multicast_start(struct net_device *dev)
 static void set_multicast_one(struct net_device *dev, const u8 *mac)
 {
 	struct fs_enet_private *fep = netdev_priv(dev);
-	int temp, hash_index, i, j;
+	int temp, hash_index;
 	u32 crc, csrVal;
-	u8 byte, msb;
-
-	crc = 0xffffffff;
-	for (i = 0; i < 6; i++) {
-		byte = mac[i];
-		for (j = 0; j < 8; j++) {
-			msb = crc >> 31;
-			crc <<= 1;
-			if (msb ^ (byte & 0x1))
-				crc ^= FEC_CRC_POLY;
-			byte >>= 1;
-		}
-	}
+
+	crc = ether_crc(6, mac);
 
 	temp = (crc & 0x3f) >> 1;
 	hash_index = ((temp & 0x01) << 4) |
@@ -235,7 +220,8 @@ static void set_multicast_list(struct net_device *dev)
 		set_promiscuous_mode(dev);
 }
 
-static void restart(struct net_device *dev)
+static void restart(struct net_device *dev, phy_interface_t interface,
+		    int speed, int duplex)
 {
 	struct fs_enet_private *fep = netdev_priv(dev);
 	struct fec __iomem *fecp = fep->fec.fecp;
@@ -317,13 +303,13 @@ static void restart(struct net_device *dev)
 	 * Only set MII/RMII mode - do not touch maximum frame length
 	 * configured before.
 	 */
-	FS(fecp, r_cntrl, fpi->use_rmii ?
-			FEC_RCNTRL_RMII_MODE : FEC_RCNTRL_MII_MODE);
+	FS(fecp, r_cntrl, interface == PHY_INTERFACE_MODE_RMII ?
+			  FEC_RCNTRL_RMII_MODE : FEC_RCNTRL_MII_MODE);
 #endif
 	/*
 	 * adjust to duplex mode
 	 */
-	if (dev->phydev->duplex) {
+	if (duplex == DUPLEX_FULL) {
 		FC(fecp, r_cntrl, FEC_RCNTRL_DRT);
 		FS(fecp, x_cntrl, FEC_TCNTRL_FDEN);	/* FD enable */
 	} else {
@@ -350,11 +336,7 @@ static void restart(struct net_device *dev)
 static void stop(struct net_device *dev)
 {
 	struct fs_enet_private *fep = netdev_priv(dev);
-	const struct fs_platform_info *fpi = fep->fpi;
 	struct fec __iomem *fecp = fep->fec.fecp;
-
-	struct fec_info *feci = dev->phydev->mdio.bus->priv;
-
 	int i;
 
 	if ((FR(fecp, ecntrl) & FEC_ECNTRL_ETHER_EN) == 0)
@@ -374,16 +356,6 @@ static void stop(struct net_device *dev)
 	FC(fecp, ecntrl, FEC_ECNTRL_ETHER_EN);
 
 	fs_cleanup_bds(dev);
-
-	/* shut down FEC1? that's where the mii bus is */
-	if (fpi->has_phy) {
-		FS(fecp, r_cntrl, fpi->use_rmii ?
-				FEC_RCNTRL_RMII_MODE :
-				FEC_RCNTRL_MII_MODE);	/* MII/RMII enable */
-		FS(fecp, ecntrl, FEC_ECNTRL_PINMUX | FEC_ECNTRL_ETHER_EN);
-		FW(fecp, ievent, FEC_ENET_MII);
-		FW(fecp, mii_speed, feci->mii_speed);
-	}
 }
 
 static void napi_clear_event_fs(struct net_device *dev)
diff --git a/drivers/net/ethernet/freescale/fs_enet/mac-scc.c b/drivers/net/ethernet/freescale/fs_enet/mac-scc.c
index 64300ac13e02..6c97191649de 100644
--- a/drivers/net/ethernet/freescale/fs_enet/mac-scc.c
+++ b/drivers/net/ethernet/freescale/fs_enet/mac-scc.c
@@ -1,3 +1,4 @@
+// SPDX-License-Identifier: GPL-2.0-only
 /*
  * Ethernet on Serial Communications Controller (SCC) driver for Motorola MPC8xx and MPC82xx.
  *
@@ -6,10 +7,6 @@
  *
  * 2005 (c) MontaVista Software, Inc.
  * Vitaly Bordug <vbordug@ru.mvista.com>
- *
- * This file is licensed under the terms of the GNU General Public License
- * version 2. This program is licensed "as is" without any warranty of any
- * kind, whether express or implied.
  */
 
 #include <linux/module.h>
@@ -25,14 +22,12 @@
 #include <linux/etherdevice.h>
 #include <linux/skbuff.h>
 #include <linux/spinlock.h>
-#include <linux/mii.h>
 #include <linux/ethtool.h>
 #include <linux/bitops.h>
 #include <linux/fs.h>
 #include <linux/platform_device.h>
 #include <linux/of_address.h>
 #include <linux/of_irq.h>
-#include <linux/of_platform.h>
 
 #include <asm/irq.h>
 #include <linux/uaccess.h>
@@ -132,15 +127,14 @@ static int setup_data(struct net_device *dev)
 static int allocate_bd(struct net_device *dev)
 {
 	struct fs_enet_private *fep = netdev_priv(dev);
-	const struct fs_platform_info *fpi = fep->fpi;
+	struct fs_platform_info *fpi = fep->fpi;
 
-	fep->ring_mem_addr = cpm_dpalloc((fpi->tx_ring + fpi->rx_ring) *
-					 sizeof(cbd_t), 8);
-	if (IS_ERR_VALUE(fep->ring_mem_addr))
+	fpi->dpram_offset = cpm_muram_alloc((fpi->tx_ring + fpi->rx_ring) *
+					    sizeof(cbd_t), 8);
+	if (IS_ERR_VALUE(fpi->dpram_offset))
 		return -ENOMEM;
 
-	fep->ring_base = (void __iomem __force*)
-		cpm_dpram_addr(fep->ring_mem_addr);
+	fep->ring_base = cpm_muram_addr(fpi->dpram_offset);
 
 	return 0;
 }
@@ -148,9 +142,10 @@ static int allocate_bd(struct net_device *dev)
 static void free_bd(struct net_device *dev)
 {
 	struct fs_enet_private *fep = netdev_priv(dev);
+	const struct fs_platform_info *fpi = fep->fpi;
 
 	if (fep->ring_base)
-		cpm_dpfree(fep->ring_mem_addr);
+		cpm_muram_free(fpi->dpram_offset);
 }
 
 static void cleanup_data(struct net_device *dev)
@@ -231,7 +226,8 @@ static void set_multicast_list(struct net_device *dev)
  * change.  This only happens when switching between half and full
  * duplex.
  */
-static void restart(struct net_device *dev)
+static void restart(struct net_device *dev, phy_interface_t interface,
+		    int speed, int duplex)
 {
 	struct fs_enet_private *fep = netdev_priv(dev);
 	scc_t __iomem *sccp = fep->scc.sccp;
@@ -248,9 +244,9 @@ static void restart(struct net_device *dev)
 		__fs_out8((u8 __iomem *)ep + i, 0);
 
 	/* point to bds */
-	W16(ep, sen_genscc.scc_rbase, fep->ring_mem_addr);
+	W16(ep, sen_genscc.scc_rbase, fpi->dpram_offset);
 	W16(ep, sen_genscc.scc_tbase,
-	    fep->ring_mem_addr + sizeof(cbd_t) * fpi->rx_ring);
+	    fpi->dpram_offset + sizeof(cbd_t) * fpi->rx_ring);
 
 	/* Initialize function code registers for big-endian.
 	 */
@@ -342,7 +338,7 @@ static void restart(struct net_device *dev)
 	W16(sccp, scc_psmr, SCC_PSMR_ENCRC | SCC_PSMR_NIB22);
 
 	/* Set full duplex mode if needed */
-	if (dev->phydev->duplex)
+	if (duplex == DUPLEX_FULL)
 		S16(sccp, scc_psmr, SCC_PSMR_LPB | SCC_PSMR_FDE);
 
 	/* Restore multicast and promiscuous settings */
diff --git a/drivers/net/ethernet/freescale/fs_enet/mii-bitbang.c b/drivers/net/ethernet/freescale/fs_enet/mii-bitbang.c
index 1f015edcca22..66038e2a4ae3 100644
--- a/drivers/net/ethernet/freescale/fs_enet/mii-bitbang.c
+++ b/drivers/net/ethernet/freescale/fs_enet/mii-bitbang.c
@@ -1,3 +1,4 @@
+// SPDX-License-Identifier: GPL-2.0-only
 /*
  * Combined Ethernet driver for Motorola MPC8xx and MPC82xx.
  *
@@ -6,10 +7,6 @@
  *
  * 2005 (c) MontaVista Software, Inc.
  * Vitaly Bordug <vbordug@ru.mvista.com>
- *
- * This file is licensed under the terms of the GNU General Public License
- * version 2. This program is licensed "as is" without any warranty of any
- * kind, whether express or implied.
  */
 
 #include <linux/module.h>
@@ -29,8 +26,8 @@
 
 struct bb_info {
 	struct mdiobb_ctrl ctrl;
-	__be32 __iomem *dir;
-	__be32 __iomem *dat;
+	u32 __iomem *dir;
+	u32 __iomem *dat;
 	u32 mdio_msk;
 	u32 mdc_msk;
 };
@@ -100,7 +97,7 @@ static inline void mdc(struct mdiobb_ctrl *ctrl, int what)
 	in_be32(bitbang->dat);
 }
 
-static struct mdiobb_ops bb_ops = {
+static const struct mdiobb_ops bb_ops = {
 	.owner = THIS_MODULE,
 	.set_mdc = mdc,
 	.set_mdio_dir = mdio_dir,
@@ -126,7 +123,7 @@ static int fs_mii_bitbang_init(struct mii_bus *bus, struct device_node *np)
 	 * we get is an int, and the odds of multiple bitbang mdio buses
 	 * is low enough that it's not worth going too crazy.
 	 */
-	snprintf(bus->id, MII_BUS_ID_SIZE, "%x", res.start);
+	snprintf(bus->id, MII_BUS_ID_SIZE, "%pa", &res.start);
 
 	data = of_get_property(np, "fsl,mdio-pin", &len);
 	if (!data || len != 4)
@@ -192,7 +189,7 @@ out:
 	return ret;
 }
 
-static int fs_enet_mdio_remove(struct platform_device *ofdev)
+static void fs_enet_mdio_remove(struct platform_device *ofdev)
 {
 	struct mii_bus *bus = platform_get_drvdata(ofdev);
 	struct bb_info *bitbang = bus->priv;
@@ -201,8 +198,6 @@ static int fs_enet_mdio_remove(struct platform_device *ofdev)
 	free_mdio_bitbang(bus);
 	iounmap(bitbang->dir);
 	kfree(bitbang);
-
-	return 0;
 }
 
 static const struct of_device_id fs_enet_mdio_bb_match[] = {
@@ -223,3 +218,4 @@ static struct platform_driver fs_enet_bb_mdio_driver = {
 };
 
 module_platform_driver(fs_enet_bb_mdio_driver);
+MODULE_LICENSE("GPL");
diff --git a/drivers/net/ethernet/freescale/fs_enet/mii-fec.c b/drivers/net/ethernet/freescale/fs_enet/mii-fec.c
index 1582d82483ec..dec31b638941 100644
--- a/drivers/net/ethernet/freescale/fs_enet/mii-fec.c
+++ b/drivers/net/ethernet/freescale/fs_enet/mii-fec.c
@@ -1,3 +1,4 @@
+// SPDX-License-Identifier: GPL-2.0-only
 /*
  * Combined Ethernet driver for Motorola MPC8xx and MPC82xx.
  *
@@ -6,10 +7,6 @@
  *
  * 2005 (c) MontaVista Software, Inc.
  * Vitaly Bordug <vbordug@ru.mvista.com>
- *
- * This file is licensed under the terms of the GNU General Public License
- * version 2. This program is licensed "as is" without any warranty of any
- * kind, whether express or implied.
  */
 
 #include <linux/module.h>
@@ -30,10 +27,12 @@
 #include <linux/ethtool.h>
 #include <linux/bitops.h>
 #include <linux/platform_device.h>
+#include <linux/property.h>
+#include <linux/of.h>
 #include <linux/of_address.h>
-#include <linux/of_platform.h>
+#include <linux/of_mdio.h>
+#include <linux/pgtable.h>
 
-#include <asm/pgtable.h>
 #include <asm/irq.h>
 #include <linux/uaccess.h>
 #include <asm/mpc5xxx.h>
@@ -95,20 +94,15 @@ static int fs_enet_fec_mii_write(struct mii_bus *bus, int phy_id, int location,
 
 }
 
-static const struct of_device_id fs_enet_mdio_fec_match[];
 static int fs_enet_mdio_probe(struct platform_device *ofdev)
 {
-	const struct of_device_id *match;
 	struct resource res;
 	struct mii_bus *new_bus;
 	struct fec_info *fec;
-	int (*get_bus_freq)(struct device_node *);
+	int (*get_bus_freq)(struct device *);
 	int ret = -ENOMEM, clock, speed;
 
-	match = of_match_device(fs_enet_mdio_fec_match, &ofdev->dev);
-	if (!match)
-		return -EINVAL;
-	get_bus_freq = match->data;
+	get_bus_freq = device_get_match_data(&ofdev->dev);
 
 	new_bus = mdiobus_alloc();
 	if (!new_bus)
@@ -127,7 +121,7 @@ static int fs_enet_mdio_probe(struct platform_device *ofdev)
 	if (ret)
 		goto out_res;
 
-	snprintf(new_bus->id, MII_BUS_ID_SIZE, "%x", res.start);
+	snprintf(new_bus->id, MII_BUS_ID_SIZE, "%pap", &res.start);
 
 	fec->fecp = ioremap(res.start, resource_size(&res));
 	if (!fec->fecp) {
@@ -136,7 +130,7 @@ static int fs_enet_mdio_probe(struct platform_device *ofdev)
 	}
 
 	if (get_bus_freq) {
-		clock = get_bus_freq(ofdev->dev.of_node);
+		clock = get_bus_freq(&ofdev->dev);
 		if (!clock) {
 			/* Use maximum divider if clock is unknown */
 			dev_warn(&ofdev->dev, "could not determine IPS clock\n");
@@ -187,7 +181,7 @@ out:
 	return ret;
 }
 
-static int fs_enet_mdio_remove(struct platform_device *ofdev)
+static void fs_enet_mdio_remove(struct platform_device *ofdev)
 {
 	struct mii_bus *bus = platform_get_drvdata(ofdev);
 	struct fec_info *fec = bus->priv;
@@ -196,8 +190,6 @@ static int fs_enet_mdio_remove(struct platform_device *ofdev)
 	iounmap(fec->fecp);
 	kfree(fec);
 	mdiobus_free(bus);
-
-	return 0;
 }
 
 static const struct of_device_id fs_enet_mdio_fec_match[] = {
@@ -224,3 +216,4 @@ static struct platform_driver fs_enet_fec_mdio_driver = {
 };
 
 module_platform_driver(fs_enet_fec_mdio_driver);
+MODULE_LICENSE("GPL");
diff --git a/drivers/net/ethernet/freescale/fsl_pq_mdio.c b/drivers/net/ethernet/freescale/fsl_pq_mdio.c
index 446c7b374ff5..de88776dd2a2 100644
--- a/drivers/net/ethernet/freescale/fsl_pq_mdio.c
+++ b/drivers/net/ethernet/freescale/fsl_pq_mdio.c
@@ -1,3 +1,4 @@
+// SPDX-License-Identifier: GPL-2.0-or-later
 /*
  * Freescale PowerQUICC Ethernet Driver -- MIIM bus implementation
  * Provides Bus interface for MIIM regs
@@ -8,24 +9,20 @@
  * Copyright 2002-2004, 2008-2009 Freescale Semiconductor, Inc.
  *
  * Based on gianfar_mii.c and ucc_geth_mii.c (Li Yang, Kim Phillips)
- *
- * This program is free software; you can redistribute  it and/or modify it
- * under  the terms of  the GNU General  Public License as published by the
- * Free Software Foundation;  either version 2 of the  License, or (at your
- * option) any later version.
- *
  */
 
 #include <linux/kernel.h>
+#include <linux/platform_device.h>
 #include <linux/string.h>
 #include <linux/errno.h>
 #include <linux/slab.h>
 #include <linux/delay.h>
 #include <linux/module.h>
 #include <linux/mii.h>
+#include <linux/of.h>
 #include <linux/of_address.h>
 #include <linux/of_mdio.h>
-#include <linux/of_device.h>
+#include <linux/property.h>
 
 #include <asm/io.h>
 #if IS_ENABLED(CONFIG_UCC_GETH)
@@ -267,8 +264,8 @@ static void ucc_configure(phys_addr_t start, phys_addr_t end)
 
 		ret = of_address_to_resource(np, 0, &res);
 		if (ret < 0) {
-			pr_debug("fsl-pq-mdio: no address range in node %s\n",
-				 np->full_name);
+			pr_debug("fsl-pq-mdio: no address range in node %pOF\n",
+				 np);
 			continue;
 		}
 
@@ -280,8 +277,8 @@ static void ucc_configure(phys_addr_t start, phys_addr_t end)
 		if (!iprop) {
 			iprop = of_get_property(np, "device-id", NULL);
 			if (!iprop) {
-				pr_debug("fsl-pq-mdio: no UCC ID in node %s\n",
-					 np->full_name);
+				pr_debug("fsl-pq-mdio: no UCC ID in node %pOF\n",
+					 np);
 				continue;
 			}
 		}
@@ -293,8 +290,8 @@ static void ucc_configure(phys_addr_t start, phys_addr_t end)
 		 * numbered from 1, not 0.
 		 */
 		if (ucc_set_qe_mux_mii_mng(id - 1) < 0) {
-			pr_debug("fsl-pq-mdio: invalid UCC ID in node %s\n",
-				 np->full_name);
+			pr_debug("fsl-pq-mdio: invalid UCC ID in node %pOF\n",
+				 np);
 			continue;
 		}
 
@@ -377,11 +374,41 @@ static const struct of_device_id fsl_pq_mdio_match[] = {
 };
 MODULE_DEVICE_TABLE(of, fsl_pq_mdio_match);
 
+static void set_tbipa(const u32 tbipa_val, struct platform_device *pdev,
+		      uint32_t __iomem * (*get_tbipa)(void __iomem *),
+		      void __iomem *reg_map, struct resource *reg_res)
+{
+	struct device_node *np = pdev->dev.of_node;
+	uint32_t __iomem *tbipa;
+	bool tbipa_mapped;
+
+	tbipa = of_iomap(np, 1);
+	if (tbipa) {
+		tbipa_mapped = true;
+	} else {
+		tbipa_mapped = false;
+		tbipa = (*get_tbipa)(reg_map);
+
+		/*
+		 * Add consistency check to make sure TBI is contained within
+		 * the mapped range (not because we would get a segfault,
+		 * rather to catch bugs in computing TBI address). Print error
+		 * message but continue anyway.
+		 */
+		if ((void *)tbipa > reg_map + resource_size(reg_res) - 4)
+			dev_err(&pdev->dev, "invalid register map (should be at least 0x%04zx to contain TBI address)\n",
+				((void *)tbipa - reg_map) + 4);
+	}
+
+	iowrite32be(be32_to_cpu(tbipa_val), tbipa);
+
+	if (tbipa_mapped)
+		iounmap(tbipa);
+}
+
 static int fsl_pq_mdio_probe(struct platform_device *pdev)
 {
-	const struct of_device_id *id =
-		of_match_device(fsl_pq_mdio_match, &pdev->dev);
-	const struct fsl_pq_mdio_data *data = id->data;
+	const struct fsl_pq_mdio_data *data;
 	struct device_node *np = pdev->dev.of_node;
 	struct resource res;
 	struct device_node *tbi;
@@ -389,14 +416,18 @@ static int fsl_pq_mdio_probe(struct platform_device *pdev)
 	struct mii_bus *new_bus;
 	int err;
 
-	dev_dbg(&pdev->dev, "found %s compatible node\n", id->compatible);
+	data = device_get_match_data(&pdev->dev);
+	if (!data) {
+		dev_err(&pdev->dev, "Failed to match device\n");
+		return -ENODEV;
+	}
 
 	new_bus = mdiobus_alloc_size(sizeof(*priv));
 	if (!new_bus)
 		return -ENOMEM;
 
 	priv = new_bus->priv;
-	new_bus->name = "Freescale PowerQUICC MII Bus",
+	new_bus->name = "Freescale PowerQUICC MII Bus";
 	new_bus->read = &fsl_pq_mdio_read;
 	new_bus->write = &fsl_pq_mdio_write;
 	new_bus->reset = &fsl_pq_mdio_reset;
@@ -407,8 +438,8 @@ static int fsl_pq_mdio_probe(struct platform_device *pdev)
 		goto error;
 	}
 
-	snprintf(new_bus->id, MII_BUS_ID_SIZE, "%s@%llx", np->name,
-		(unsigned long long)res.start);
+	snprintf(new_bus->id, MII_BUS_ID_SIZE, "%pOFn@%llx", np,
+		 (unsigned long long)res.start);
 
 	priv->map = of_iomap(np, 0);
 	if (!priv->map) {
@@ -434,38 +465,26 @@ static int fsl_pq_mdio_probe(struct platform_device *pdev)
 
 	if (data->get_tbipa) {
 		for_each_child_of_node(np, tbi) {
-			if (strcmp(tbi->type, "tbi-phy") == 0) {
-				dev_dbg(&pdev->dev, "found TBI PHY node %s\n",
-					strrchr(tbi->full_name, '/') + 1);
+			if (of_node_is_type(tbi, "tbi-phy")) {
+				dev_dbg(&pdev->dev, "found TBI PHY node %pOFP\n",
+					tbi);
 				break;
 			}
 		}
 
 		if (tbi) {
 			const u32 *prop = of_get_property(tbi, "reg", NULL);
-			uint32_t __iomem *tbipa;
-
 			if (!prop) {
 				dev_err(&pdev->dev,
-					"missing 'reg' property in node %s\n",
-					tbi->full_name);
+					"missing 'reg' property in node %pOF\n",
+					tbi);
 				err = -EBUSY;
+				of_node_put(tbi);
 				goto error;
 			}
-
-			tbipa = data->get_tbipa(priv->map);
-
-			/*
-			 * Add consistency check to make sure TBI is contained
-			 * within the mapped range (not because we would get a
-			 * segfault, rather to catch bugs in computing TBI
-			 * address). Print error message but continue anyway.
-			 */
-			if ((void *)tbipa > priv->map + resource_size(&res) - 4)
-				dev_err(&pdev->dev, "invalid register map (should be at least 0x%04zx to contain TBI address)\n",
-					((void *)tbipa - priv->map) + 4);
-
-			iowrite32be(be32_to_cpup(prop), tbipa);
+			set_tbipa(*prop, pdev,
+				  data->get_tbipa, priv->map, &res);
+			of_node_put(tbi);
 		}
 	}
 
@@ -474,8 +493,8 @@ static int fsl_pq_mdio_probe(struct platform_device *pdev)
 
 	err = of_mdiobus_register(new_bus, np);
 	if (err) {
-		dev_err(&pdev->dev, "cannot register %s as MDIO bus\n",
-			new_bus->name);
+		dev_err_probe(&pdev->dev, err, "cannot register %s as MDIO bus\n",
+			      new_bus->name);
 		goto error;
 	}
 
@@ -491,7 +510,7 @@ error:
 }
 
 
-static int fsl_pq_mdio_remove(struct platform_device *pdev)
+static void fsl_pq_mdio_remove(struct platform_device *pdev)
 {
 	struct device *device = &pdev->dev;
 	struct mii_bus *bus = dev_get_drvdata(device);
@@ -501,8 +520,6 @@ static int fsl_pq_mdio_remove(struct platform_device *pdev)
 
 	iounmap(priv->map);
 	mdiobus_free(bus);
-
-	return 0;
 }
 
 static struct platform_driver fsl_pq_mdio_driver = {
@@ -516,4 +533,5 @@ static struct platform_driver fsl_pq_mdio_driver = {
 
 module_platform_driver(fsl_pq_mdio_driver);
 
+MODULE_DESCRIPTION("Freescale PQ MDIO helpers");
 MODULE_LICENSE("GPL");
diff --git a/drivers/net/ethernet/freescale/gianfar.c b/drivers/net/ethernet/freescale/gianfar.c
index 0ff166ec3e7e..7c0f049f0938 100644
--- a/drivers/net/ethernet/freescale/gianfar.c
+++ b/drivers/net/ethernet/freescale/gianfar.c
@@ -1,3 +1,4 @@
+// SPDX-License-Identifier: GPL-2.0-or-later
 /* drivers/net/ethernet/freescale/gianfar.c
  *
  * Gianfar Ethernet Driver
@@ -12,11 +13,6 @@
  * Copyright 2002-2009, 2011-2013 Freescale Semiconductor, Inc.
  * Copyright 2007 MontaVista Software, Inc.
  *
- * This program is free software; you can redistribute  it and/or modify it
- * under  the terms of  the GNU General  Public License as published by the
- * Free Software Foundation;  either version 2 of the  License, or (at your
- * option) any later version.
- *
  *  Gianfar:  AKA Lambda Draconis, "Dragon"
  *  RA 11 31 24.2
  *  Dec +69 19 52
@@ -62,9 +58,9 @@
  */
 
 #define pr_fmt(fmt) KBUILD_MODNAME ": " fmt
-#define DEBUG
 
 #include <linux/kernel.h>
+#include <linux/platform_device.h>
 #include <linux/string.h>
 #include <linux/errno.h>
 #include <linux/unistd.h>
@@ -80,7 +76,6 @@
 #include <linux/of_address.h>
 #include <linux/of_irq.h>
 #include <linux/of_mdio.h>
-#include <linux/of_platform.h>
 #include <linux/ip.h>
 #include <linux/tcp.h>
 #include <linux/udp.h>
@@ -102,52 +97,12 @@
 #include <linux/phy_fixed.h>
 #include <linux/of.h>
 #include <linux/of_net.h>
-#include <linux/of_address.h>
-#include <linux/of_irq.h>
+#include <linux/property.h>
 
 #include "gianfar.h"
 
 #define TX_TIMEOUT      (5*HZ)
 
-const char gfar_driver_version[] = "2.0";
-
-static int gfar_enet_open(struct net_device *dev);
-static int gfar_start_xmit(struct sk_buff *skb, struct net_device *dev);
-static void gfar_reset_task(struct work_struct *work);
-static void gfar_timeout(struct net_device *dev);
-static int gfar_close(struct net_device *dev);
-static void gfar_alloc_rx_buffs(struct gfar_priv_rx_q *rx_queue,
-				int alloc_cnt);
-static int gfar_set_mac_address(struct net_device *dev);
-static int gfar_change_mtu(struct net_device *dev, int new_mtu);
-static irqreturn_t gfar_error(int irq, void *dev_id);
-static irqreturn_t gfar_transmit(int irq, void *dev_id);
-static irqreturn_t gfar_interrupt(int irq, void *dev_id);
-static void adjust_link(struct net_device *dev);
-static noinline void gfar_update_link_state(struct gfar_private *priv);
-static int init_phy(struct net_device *dev);
-static int gfar_probe(struct platform_device *ofdev);
-static int gfar_remove(struct platform_device *ofdev);
-static void free_skb_resources(struct gfar_private *priv);
-static void gfar_set_multi(struct net_device *dev);
-static void gfar_set_hash_for_addr(struct net_device *dev, u8 *addr);
-static void gfar_configure_serdes(struct net_device *dev);
-static int gfar_poll_rx(struct napi_struct *napi, int budget);
-static int gfar_poll_tx(struct napi_struct *napi, int budget);
-static int gfar_poll_rx_sq(struct napi_struct *napi, int budget);
-static int gfar_poll_tx_sq(struct napi_struct *napi, int budget);
-#ifdef CONFIG_NET_POLL_CONTROLLER
-static void gfar_netpoll(struct net_device *dev);
-#endif
-int gfar_clean_rx_ring(struct gfar_priv_rx_q *rx_queue, int rx_work_limit);
-static void gfar_clean_tx_ring(struct gfar_priv_tx_q *tx_queue);
-static void gfar_process_frame(struct net_device *ndev, struct sk_buff *skb);
-static void gfar_halt_nodisable(struct gfar_private *priv);
-static void gfar_clear_exact_match(struct net_device *dev);
-static void gfar_set_mac_for_addr(struct net_device *dev, int num,
-				  const u8 *addr);
-static int gfar_ioctl(struct net_device *dev, struct ifreq *rq, int cmd);
-
 MODULE_AUTHOR("Freescale Semiconductor, Inc");
 MODULE_DESCRIPTION("Gianfar Ethernet Driver");
 MODULE_LICENSE("GPL");
@@ -168,138 +123,6 @@ static void gfar_init_rxbdp(struct gfar_priv_rx_q *rx_queue, struct rxbd8 *bdp,
 	bdp->lstatus = cpu_to_be32(lstatus);
 }
 
-static void gfar_init_bds(struct net_device *ndev)
-{
-	struct gfar_private *priv = netdev_priv(ndev);
-	struct gfar __iomem *regs = priv->gfargrp[0].regs;
-	struct gfar_priv_tx_q *tx_queue = NULL;
-	struct gfar_priv_rx_q *rx_queue = NULL;
-	struct txbd8 *txbdp;
-	u32 __iomem *rfbptr;
-	int i, j;
-
-	for (i = 0; i < priv->num_tx_queues; i++) {
-		tx_queue = priv->tx_queue[i];
-		/* Initialize some variables in our dev structure */
-		tx_queue->num_txbdfree = tx_queue->tx_ring_size;
-		tx_queue->dirty_tx = tx_queue->tx_bd_base;
-		tx_queue->cur_tx = tx_queue->tx_bd_base;
-		tx_queue->skb_curtx = 0;
-		tx_queue->skb_dirtytx = 0;
-
-		/* Initialize Transmit Descriptor Ring */
-		txbdp = tx_queue->tx_bd_base;
-		for (j = 0; j < tx_queue->tx_ring_size; j++) {
-			txbdp->lstatus = 0;
-			txbdp->bufPtr = 0;
-			txbdp++;
-		}
-
-		/* Set the last descriptor in the ring to indicate wrap */
-		txbdp--;
-		txbdp->status = cpu_to_be16(be16_to_cpu(txbdp->status) |
-					    TXBD_WRAP);
-	}
-
-	rfbptr = &regs->rfbptr0;
-	for (i = 0; i < priv->num_rx_queues; i++) {
-		rx_queue = priv->rx_queue[i];
-
-		rx_queue->next_to_clean = 0;
-		rx_queue->next_to_use = 0;
-		rx_queue->next_to_alloc = 0;
-
-		/* make sure next_to_clean != next_to_use after this
-		 * by leaving at least 1 unused descriptor
-		 */
-		gfar_alloc_rx_buffs(rx_queue, gfar_rxbd_unused(rx_queue));
-
-		rx_queue->rfbptr = rfbptr;
-		rfbptr += 2;
-	}
-}
-
-static int gfar_alloc_skb_resources(struct net_device *ndev)
-{
-	void *vaddr;
-	dma_addr_t addr;
-	int i, j;
-	struct gfar_private *priv = netdev_priv(ndev);
-	struct device *dev = priv->dev;
-	struct gfar_priv_tx_q *tx_queue = NULL;
-	struct gfar_priv_rx_q *rx_queue = NULL;
-
-	priv->total_tx_ring_size = 0;
-	for (i = 0; i < priv->num_tx_queues; i++)
-		priv->total_tx_ring_size += priv->tx_queue[i]->tx_ring_size;
-
-	priv->total_rx_ring_size = 0;
-	for (i = 0; i < priv->num_rx_queues; i++)
-		priv->total_rx_ring_size += priv->rx_queue[i]->rx_ring_size;
-
-	/* Allocate memory for the buffer descriptors */
-	vaddr = dma_alloc_coherent(dev,
-				   (priv->total_tx_ring_size *
-				    sizeof(struct txbd8)) +
-				   (priv->total_rx_ring_size *
-				    sizeof(struct rxbd8)),
-				   &addr, GFP_KERNEL);
-	if (!vaddr)
-		return -ENOMEM;
-
-	for (i = 0; i < priv->num_tx_queues; i++) {
-		tx_queue = priv->tx_queue[i];
-		tx_queue->tx_bd_base = vaddr;
-		tx_queue->tx_bd_dma_base = addr;
-		tx_queue->dev = ndev;
-		/* enet DMA only understands physical addresses */
-		addr  += sizeof(struct txbd8) * tx_queue->tx_ring_size;
-		vaddr += sizeof(struct txbd8) * tx_queue->tx_ring_size;
-	}
-
-	/* Start the rx descriptor ring where the tx ring leaves off */
-	for (i = 0; i < priv->num_rx_queues; i++) {
-		rx_queue = priv->rx_queue[i];
-		rx_queue->rx_bd_base = vaddr;
-		rx_queue->rx_bd_dma_base = addr;
-		rx_queue->ndev = ndev;
-		rx_queue->dev = dev;
-		addr  += sizeof(struct rxbd8) * rx_queue->rx_ring_size;
-		vaddr += sizeof(struct rxbd8) * rx_queue->rx_ring_size;
-	}
-
-	/* Setup the skbuff rings */
-	for (i = 0; i < priv->num_tx_queues; i++) {
-		tx_queue = priv->tx_queue[i];
-		tx_queue->tx_skbuff =
-			kmalloc_array(tx_queue->tx_ring_size,
-				      sizeof(*tx_queue->tx_skbuff),
-				      GFP_KERNEL);
-		if (!tx_queue->tx_skbuff)
-			goto cleanup;
-
-		for (j = 0; j < tx_queue->tx_ring_size; j++)
-			tx_queue->tx_skbuff[j] = NULL;
-	}
-
-	for (i = 0; i < priv->num_rx_queues; i++) {
-		rx_queue = priv->rx_queue[i];
-		rx_queue->rx_buff = kcalloc(rx_queue->rx_ring_size,
-					    sizeof(*rx_queue->rx_buff),
-					    GFP_KERNEL);
-		if (!rx_queue->rx_buff)
-			goto cleanup;
-	}
-
-	gfar_init_bds(ndev);
-
-	return 0;
-
-cleanup:
-	free_skb_resources(priv);
-	return -ENOMEM;
-}
-
 static void gfar_init_tx_rx_base(struct gfar_private *priv)
 {
 	struct gfar __iomem *regs = priv->gfargrp[0].regs;
@@ -353,10 +176,7 @@ static void gfar_mac_rx_config(struct gfar_private *priv)
 	if (priv->rx_filer_enable) {
 		rctrl |= RCTRL_FILREN | RCTRL_PRSDEP_INIT;
 		/* Program the RIR0 reg with the required distribution */
-		if (priv->poll_mode == GFAR_SQ_POLLING)
-			gfar_write(&regs->rir0, DEFAULT_2RXQ_RIR0);
-		else /* GFAR_MQ_POLLING */
-			gfar_write(&regs->rir0, DEFAULT_8RXQ_RIR0);
+		gfar_write(&regs->rir0, DEFAULT_2RXQ_RIR0);
 	}
 
 	/* Restore PROMISC mode */
@@ -450,65 +270,120 @@ static void gfar_configure_coalescing(struct gfar_private *priv,
 	}
 }
 
-void gfar_configure_coalescing_all(struct gfar_private *priv)
+static void gfar_configure_coalescing_all(struct gfar_private *priv)
 {
 	gfar_configure_coalescing(priv, 0xFF, 0xFF);
 }
 
-static struct net_device_stats *gfar_get_stats(struct net_device *dev)
+static void gfar_get_stats64(struct net_device *dev, struct rtnl_link_stats64 *stats)
 {
 	struct gfar_private *priv = netdev_priv(dev);
-	unsigned long rx_packets = 0, rx_bytes = 0, rx_dropped = 0;
-	unsigned long tx_packets = 0, tx_bytes = 0;
 	int i;
 
 	for (i = 0; i < priv->num_rx_queues; i++) {
-		rx_packets += priv->rx_queue[i]->stats.rx_packets;
-		rx_bytes   += priv->rx_queue[i]->stats.rx_bytes;
-		rx_dropped += priv->rx_queue[i]->stats.rx_dropped;
+		stats->rx_packets += priv->rx_queue[i]->stats.rx_packets;
+		stats->rx_bytes   += priv->rx_queue[i]->stats.rx_bytes;
+		stats->rx_dropped += priv->rx_queue[i]->stats.rx_dropped;
 	}
 
-	dev->stats.rx_packets = rx_packets;
-	dev->stats.rx_bytes   = rx_bytes;
-	dev->stats.rx_dropped = rx_dropped;
-
 	for (i = 0; i < priv->num_tx_queues; i++) {
-		tx_bytes += priv->tx_queue[i]->stats.tx_bytes;
-		tx_packets += priv->tx_queue[i]->stats.tx_packets;
+		stats->tx_bytes += priv->tx_queue[i]->stats.tx_bytes;
+		stats->tx_packets += priv->tx_queue[i]->stats.tx_packets;
 	}
 
-	dev->stats.tx_bytes   = tx_bytes;
-	dev->stats.tx_packets = tx_packets;
+	if (priv->device_flags & FSL_GIANFAR_DEV_HAS_RMON) {
+		struct rmon_mib __iomem *rmon = &priv->gfargrp[0].regs->rmon;
+		unsigned long flags;
+		u32 rdrp, car, car_before;
+		u64 rdrp_offset;
+
+		spin_lock_irqsave(&priv->rmon_overflow.lock, flags);
+		car = gfar_read(&rmon->car1) & CAR1_C1RDR;
+		do {
+			car_before = car;
+			rdrp = gfar_read(&rmon->rdrp);
+			car = gfar_read(&rmon->car1) & CAR1_C1RDR;
+		} while (car != car_before);
+		if (car) {
+			priv->rmon_overflow.rdrp++;
+			gfar_write(&rmon->car1, car);
+		}
+		rdrp_offset = priv->rmon_overflow.rdrp;
+		spin_unlock_irqrestore(&priv->rmon_overflow.lock, flags);
 
-	return &dev->stats;
+		stats->rx_missed_errors = rdrp + (rdrp_offset << 16);
+	}
+}
+
+/* Set the appropriate hash bit for the given addr */
+/* The algorithm works like so:
+ * 1) Take the Destination Address (ie the multicast address), and
+ * do a CRC on it (little endian), and reverse the bits of the
+ * result.
+ * 2) Use the 8 most significant bits as a hash into a 256-entry
+ * table.  The table is controlled through 8 32-bit registers:
+ * gaddr0-7.  gaddr0's MSB is entry 0, and gaddr7's LSB is
+ * gaddr7.  This means that the 3 most significant bits in the
+ * hash index which gaddr register to use, and the 5 other bits
+ * indicate which bit (assuming an IBM numbering scheme, which
+ * for PowerPC (tm) is usually the case) in the register holds
+ * the entry.
+ */
+static void gfar_set_hash_for_addr(struct net_device *dev, u8 *addr)
+{
+	u32 tempval;
+	struct gfar_private *priv = netdev_priv(dev);
+	u32 result = ether_crc(ETH_ALEN, addr);
+	int width = priv->hash_width;
+	u8 whichbit = (result >> (32 - width)) & 0x1f;
+	u8 whichreg = result >> (32 - width + 5);
+	u32 value = (1 << (31-whichbit));
+
+	tempval = gfar_read(priv->hash_regs[whichreg]);
+	tempval |= value;
+	gfar_write(priv->hash_regs[whichreg], tempval);
+}
+
+/* There are multiple MAC Address register pairs on some controllers
+ * This function sets the numth pair to a given address
+ */
+static void gfar_set_mac_for_addr(struct net_device *dev, int num,
+				  const u8 *addr)
+{
+	struct gfar_private *priv = netdev_priv(dev);
+	struct gfar __iomem *regs = priv->gfargrp[0].regs;
+	u32 tempval;
+	u32 __iomem *macptr = &regs->macstnaddr1;
+
+	macptr += num*2;
+
+	/* For a station address of 0x12345678ABCD in transmission
+	 * order (BE), MACnADDR1 is set to 0xCDAB7856 and
+	 * MACnADDR2 is set to 0x34120000.
+	 */
+	tempval = (addr[5] << 24) | (addr[4] << 16) |
+		  (addr[3] << 8)  |  addr[2];
+
+	gfar_write(macptr, tempval);
+
+	tempval = (addr[1] << 24) | (addr[0] << 16);
+
+	gfar_write(macptr+1, tempval);
 }
 
 static int gfar_set_mac_addr(struct net_device *dev, void *p)
 {
-	eth_mac_addr(dev, p);
+	int ret;
+
+	ret = eth_mac_addr(dev, p);
+	if (ret)
+		return ret;
 
 	gfar_set_mac_for_addr(dev, 0, dev->dev_addr);
 
 	return 0;
 }
 
-static const struct net_device_ops gfar_netdev_ops = {
-	.ndo_open = gfar_enet_open,
-	.ndo_start_xmit = gfar_start_xmit,
-	.ndo_stop = gfar_close,
-	.ndo_change_mtu = gfar_change_mtu,
-	.ndo_set_features = gfar_set_features,
-	.ndo_set_rx_mode = gfar_set_multi,
-	.ndo_tx_timeout = gfar_timeout,
-	.ndo_do_ioctl = gfar_ioctl,
-	.ndo_get_stats = gfar_get_stats,
-	.ndo_set_mac_address = gfar_set_mac_addr,
-	.ndo_validate_addr = eth_validate_addr,
-#ifdef CONFIG_NET_POLL_CONTROLLER
-	.ndo_poll_controller = gfar_netpoll,
-#endif
-};
-
 static void gfar_ints_disable(struct gfar_private *priv)
 {
 	int i;
@@ -528,7 +403,8 @@ static void gfar_ints_enable(struct gfar_private *priv)
 	for (i = 0; i < priv->num_grps; i++) {
 		struct gfar __iomem *regs = priv->gfargrp[i].regs;
 		/* Unmask the interrupts we look for */
-		gfar_write(&regs->imask, IMASK_DEFAULT);
+		gfar_write(&regs->imask,
+			   IMASK_DEFAULT | priv->rmon_overflow.imask);
 	}
 }
 
@@ -656,29 +532,9 @@ static int gfar_parse_group(struct device_node *np,
 	grp->priv = priv;
 	spin_lock_init(&grp->grplock);
 	if (priv->mode == MQ_MG_MODE) {
-		u32 rxq_mask, txq_mask;
-		int ret;
-
+		/* One Q per interrupt group: Q0 to G0, Q1 to G1 */
 		grp->rx_bit_map = (DEFAULT_MAPPING >> priv->num_grps);
 		grp->tx_bit_map = (DEFAULT_MAPPING >> priv->num_grps);
-
-		ret = of_property_read_u32(np, "fsl,rx-bit-map", &rxq_mask);
-		if (!ret) {
-			grp->rx_bit_map = rxq_mask ?
-			rxq_mask : (DEFAULT_MAPPING >> priv->num_grps);
-		}
-
-		ret = of_property_read_u32(np, "fsl,tx-bit-map", &txq_mask);
-		if (!ret) {
-			grp->tx_bit_map = txq_mask ?
-			txq_mask : (DEFAULT_MAPPING >> priv->num_grps);
-		}
-
-		if (priv->poll_mode == GFAR_SQ_POLLING) {
-			/* One Q per interrupt group: Q0 to G0, Q1 to G1 */
-			grp->rx_bit_map = (DEFAULT_MAPPING >> priv->num_grps);
-			grp->tx_bit_map = (DEFAULT_MAPPING >> priv->num_grps);
-		}
 	} else {
 		grp->rx_bit_map = 0xFF;
 		grp->tx_bit_map = 0xFF;
@@ -716,24 +572,55 @@ static int gfar_parse_group(struct device_node *np,
 	return 0;
 }
 
-static int gfar_of_group_count(struct device_node *np)
+/* Reads the controller's registers to determine what interface
+ * connects it to the PHY.
+ */
+static phy_interface_t gfar_get_interface(struct net_device *dev)
 {
-	struct device_node *child;
-	int num = 0;
+	struct gfar_private *priv = netdev_priv(dev);
+	struct gfar __iomem *regs = priv->gfargrp[0].regs;
+	u32 ecntrl;
+
+	ecntrl = gfar_read(&regs->ecntrl);
+
+	if (ecntrl & ECNTRL_SGMII_MODE)
+		return PHY_INTERFACE_MODE_SGMII;
 
-	for_each_available_child_of_node(np, child)
-		if (!of_node_cmp(child->name, "queue-group"))
-			num++;
+	if (ecntrl & ECNTRL_TBI_MODE) {
+		if (ecntrl & ECNTRL_REDUCED_MODE)
+			return PHY_INTERFACE_MODE_RTBI;
+		else
+			return PHY_INTERFACE_MODE_TBI;
+	}
 
-	return num;
+	if (ecntrl & ECNTRL_REDUCED_MODE) {
+		if (ecntrl & ECNTRL_REDUCED_MII_MODE) {
+			return PHY_INTERFACE_MODE_RMII;
+		}
+		else {
+			phy_interface_t interface = priv->interface;
+
+			/* This isn't autodetected right now, so it must
+			 * be set by the device tree or platform code.
+			 */
+			if (interface == PHY_INTERFACE_MODE_RGMII_ID)
+				return PHY_INTERFACE_MODE_RGMII_ID;
+
+			return PHY_INTERFACE_MODE_RGMII;
+		}
+	}
+
+	if (priv->device_flags & FSL_GIANFAR_DEV_HAS_GIGABIT)
+		return PHY_INTERFACE_MODE_GMII;
+
+	return PHY_INTERFACE_MODE_MII;
 }
 
 static int gfar_of_init(struct platform_device *ofdev, struct net_device **pdev)
 {
 	const char *model;
-	const char *ctype;
-	const void *mac_addr;
 	int err = 0, i;
+	phy_interface_t interface;
 	struct net_device *dev = NULL;
 	struct gfar_private *priv = NULL;
 	struct device_node *np = ofdev->dev.of_node;
@@ -741,26 +628,25 @@ static int gfar_of_init(struct platform_device *ofdev, struct net_device **pdev)
 	u32 stash_len = 0;
 	u32 stash_idx = 0;
 	unsigned int num_tx_qs, num_rx_qs;
-	unsigned short mode, poll_mode;
+	unsigned short mode;
 
 	if (!np)
 		return -ENODEV;
 
-	if (of_device_is_compatible(np, "fsl,etsec2")) {
+	if (of_device_is_compatible(np, "fsl,etsec2"))
 		mode = MQ_MG_MODE;
-		poll_mode = GFAR_SQ_POLLING;
-	} else {
+	else
 		mode = SQ_SG_MODE;
-		poll_mode = GFAR_SQ_POLLING;
-	}
 
 	if (mode == SQ_SG_MODE) {
 		num_tx_qs = 1;
 		num_rx_qs = 1;
 	} else { /* MQ_MG_MODE */
 		/* get the actual number of supported groups */
-		unsigned int num_grps = gfar_of_group_count(np);
+		unsigned int num_grps;
 
+		num_grps = device_get_named_child_node_count(&ofdev->dev,
+							     "queue-group");
 		if (num_grps == 0 || num_grps > MAXGROUPS) {
 			dev_err(&ofdev->dev, "Invalid # of int groups(%d)\n",
 				num_grps);
@@ -768,22 +654,8 @@ static int gfar_of_init(struct platform_device *ofdev, struct net_device **pdev)
 			return -EINVAL;
 		}
 
-		if (poll_mode == GFAR_SQ_POLLING) {
-			num_tx_qs = num_grps; /* one txq per int group */
-			num_rx_qs = num_grps; /* one rxq per int group */
-		} else { /* GFAR_MQ_POLLING */
-			u32 tx_queues, rx_queues;
-			int ret;
-
-			/* parse the num of HW tx and rx queues */
-			ret = of_property_read_u32(np, "fsl,num_tx_queues",
-						   &tx_queues);
-			num_tx_qs = ret ? 1 : tx_queues;
-
-			ret = of_property_read_u32(np, "fsl,num_rx_queues",
-						   &rx_queues);
-			num_rx_qs = ret ? 1 : rx_queues;
-		}
+		num_tx_qs = num_grps; /* one txq per int group */
+		num_rx_qs = num_grps; /* one rxq per int group */
 	}
 
 	if (num_tx_qs > MAX_TX_QS) {
@@ -809,7 +681,6 @@ static int gfar_of_init(struct platform_device *ofdev, struct net_device **pdev)
 	priv->ndev = dev;
 
 	priv->mode = mode;
-	priv->poll_mode = poll_mode;
 
 	priv->num_tx_queues = num_tx_qs;
 	netif_set_real_num_rx_queues(dev, num_rx_qs);
@@ -840,12 +711,14 @@ static int gfar_of_init(struct platform_device *ofdev, struct net_device **pdev)
 	/* Parse and initialize group specific information */
 	if (priv->mode == MQ_MG_MODE) {
 		for_each_available_child_of_node(np, child) {
-			if (of_node_cmp(child->name, "queue-group"))
+			if (!of_node_name_eq(child, "queue-group"))
 				continue;
 
 			err = gfar_parse_group(child, priv, model);
-			if (err)
+			if (err) {
+				of_node_put(child);
 				goto err_grp_init;
+			}
 		}
 	} else { /* SQ_SG_MODE */
 		err = gfar_parse_group(np, priv, model);
@@ -871,10 +744,13 @@ static int gfar_of_init(struct platform_device *ofdev, struct net_device **pdev)
 	if (stash_len || stash_idx)
 		priv->device_flags |= FSL_GIANFAR_DEV_HAS_BUF_STASHING;
 
-	mac_addr = of_get_mac_address(np);
-
-	if (mac_addr)
-		memcpy(dev->dev_addr, mac_addr, ETH_ALEN);
+	err = of_get_ethdev_address(np, dev);
+	if (err == -EPROBE_DEFER)
+		goto err_grp_init;
+	if (err) {
+		eth_hw_addr_random(dev);
+		dev_info(&ofdev->dev, "Using random MAC address: %pM\n", dev->dev_addr);
+	}
 
 	if (model && !strcasecmp(model, "TSEC"))
 		priv->device_flags |= FSL_GIANFAR_DEV_HAS_GIGABIT |
@@ -894,18 +770,20 @@ static int gfar_of_init(struct platform_device *ofdev, struct net_device **pdev)
 				     FSL_GIANFAR_DEV_HAS_TIMER |
 				     FSL_GIANFAR_DEV_HAS_RX_FILER;
 
-	err = of_property_read_string(np, "phy-connection-type", &ctype);
-
-	/* We only care about rgmii-id.  The rest are autodetected */
-	if (err == 0 && !strcmp(ctype, "rgmii-id"))
-		priv->interface = PHY_INTERFACE_MODE_RGMII_ID;
+	/* Use PHY connection type from the DT node if one is specified there.
+	 * rgmii-id really needs to be specified. Other types can be
+	 * detected by hardware
+	 */
+	err = of_get_phy_mode(np, &interface);
+	if (!err)
+		priv->interface = interface;
 	else
-		priv->interface = PHY_INTERFACE_MODE_MII;
+		priv->interface = gfar_get_interface(dev);
 
-	if (of_find_property(np, "fsl,magic-packet", NULL))
+	if (of_property_read_bool(np, "fsl,magic-packet"))
 		priv->device_flags |= FSL_GIANFAR_DEV_HAS_MAGIC_PACKET;
 
-	if (of_get_property(np, "fsl,wake-on-filer", NULL))
+	if (of_property_read_bool(np, "fsl,wake-on-filer"))
 		priv->device_flags |= FSL_GIANFAR_DEV_HAS_WAKE_ON_FILER;
 
 	priv->phy_node = of_parse_phandle(np, "phy-handle", 0);
@@ -936,85 +814,6 @@ tx_alloc_failed:
 	return err;
 }
 
-static int gfar_hwtstamp_set(struct net_device *netdev, struct ifreq *ifr)
-{
-	struct hwtstamp_config config;
-	struct gfar_private *priv = netdev_priv(netdev);
-
-	if (copy_from_user(&config, ifr->ifr_data, sizeof(config)))
-		return -EFAULT;
-
-	/* reserved for future extensions */
-	if (config.flags)
-		return -EINVAL;
-
-	switch (config.tx_type) {
-	case HWTSTAMP_TX_OFF:
-		priv->hwts_tx_en = 0;
-		break;
-	case HWTSTAMP_TX_ON:
-		if (!(priv->device_flags & FSL_GIANFAR_DEV_HAS_TIMER))
-			return -ERANGE;
-		priv->hwts_tx_en = 1;
-		break;
-	default:
-		return -ERANGE;
-	}
-
-	switch (config.rx_filter) {
-	case HWTSTAMP_FILTER_NONE:
-		if (priv->hwts_rx_en) {
-			priv->hwts_rx_en = 0;
-			reset_gfar(netdev);
-		}
-		break;
-	default:
-		if (!(priv->device_flags & FSL_GIANFAR_DEV_HAS_TIMER))
-			return -ERANGE;
-		if (!priv->hwts_rx_en) {
-			priv->hwts_rx_en = 1;
-			reset_gfar(netdev);
-		}
-		config.rx_filter = HWTSTAMP_FILTER_ALL;
-		break;
-	}
-
-	return copy_to_user(ifr->ifr_data, &config, sizeof(config)) ?
-		-EFAULT : 0;
-}
-
-static int gfar_hwtstamp_get(struct net_device *netdev, struct ifreq *ifr)
-{
-	struct hwtstamp_config config;
-	struct gfar_private *priv = netdev_priv(netdev);
-
-	config.flags = 0;
-	config.tx_type = priv->hwts_tx_en ? HWTSTAMP_TX_ON : HWTSTAMP_TX_OFF;
-	config.rx_filter = (priv->hwts_rx_en ?
-			    HWTSTAMP_FILTER_ALL : HWTSTAMP_FILTER_NONE);
-
-	return copy_to_user(ifr->ifr_data, &config, sizeof(config)) ?
-		-EFAULT : 0;
-}
-
-static int gfar_ioctl(struct net_device *dev, struct ifreq *rq, int cmd)
-{
-	struct phy_device *phydev = dev->phydev;
-
-	if (!netif_running(dev))
-		return -EINVAL;
-
-	if (cmd == SIOCSHWTSTAMP)
-		return gfar_hwtstamp_set(dev, rq);
-	if (cmd == SIOCGHWTSTAMP)
-		return gfar_hwtstamp_get(dev, rq);
-
-	if (!phydev)
-		return -ENODEV;
-
-	return phy_mii_ioctl(phydev, rq, cmd);
-}
-
 static u32 cluster_entry_per_class(struct gfar_private *priv, u32 rqfar,
 				   u32 class)
 {
@@ -1138,135 +937,6 @@ static void gfar_detect_errata(struct gfar_private *priv)
 			 priv->errata);
 }
 
-void gfar_mac_reset(struct gfar_private *priv)
-{
-	struct gfar __iomem *regs = priv->gfargrp[0].regs;
-	u32 tempval;
-
-	/* Reset MAC layer */
-	gfar_write(&regs->maccfg1, MACCFG1_SOFT_RESET);
-
-	/* We need to delay at least 3 TX clocks */
-	udelay(3);
-
-	/* the soft reset bit is not self-resetting, so we need to
-	 * clear it before resuming normal operation
-	 */
-	gfar_write(&regs->maccfg1, 0);
-
-	udelay(3);
-
-	gfar_rx_offload_en(priv);
-
-	/* Initialize the max receive frame/buffer lengths */
-	gfar_write(&regs->maxfrm, GFAR_JUMBO_FRAME_SIZE);
-	gfar_write(&regs->mrblr, GFAR_RXB_SIZE);
-
-	/* Initialize the Minimum Frame Length Register */
-	gfar_write(&regs->minflr, MINFLR_INIT_SETTINGS);
-
-	/* Initialize MACCFG2. */
-	tempval = MACCFG2_INIT_SETTINGS;
-
-	/* eTSEC74 erratum: Rx frames of length MAXFRM or MAXFRM-1
-	 * are marked as truncated.  Avoid this by MACCFG2[Huge Frame]=1,
-	 * and by checking RxBD[LG] and discarding larger than MAXFRM.
-	 */
-	if (gfar_has_errata(priv, GFAR_ERRATA_74))
-		tempval |= MACCFG2_HUGEFRAME | MACCFG2_LENGTHCHECK;
-
-	gfar_write(&regs->maccfg2, tempval);
-
-	/* Clear mac addr hash registers */
-	gfar_write(&regs->igaddr0, 0);
-	gfar_write(&regs->igaddr1, 0);
-	gfar_write(&regs->igaddr2, 0);
-	gfar_write(&regs->igaddr3, 0);
-	gfar_write(&regs->igaddr4, 0);
-	gfar_write(&regs->igaddr5, 0);
-	gfar_write(&regs->igaddr6, 0);
-	gfar_write(&regs->igaddr7, 0);
-
-	gfar_write(&regs->gaddr0, 0);
-	gfar_write(&regs->gaddr1, 0);
-	gfar_write(&regs->gaddr2, 0);
-	gfar_write(&regs->gaddr3, 0);
-	gfar_write(&regs->gaddr4, 0);
-	gfar_write(&regs->gaddr5, 0);
-	gfar_write(&regs->gaddr6, 0);
-	gfar_write(&regs->gaddr7, 0);
-
-	if (priv->extended_hash)
-		gfar_clear_exact_match(priv->ndev);
-
-	gfar_mac_rx_config(priv);
-
-	gfar_mac_tx_config(priv);
-
-	gfar_set_mac_address(priv->ndev);
-
-	gfar_set_multi(priv->ndev);
-
-	/* clear ievent and imask before configuring coalescing */
-	gfar_ints_disable(priv);
-
-	/* Configure the coalescing support */
-	gfar_configure_coalescing_all(priv);
-}
-
-static void gfar_hw_init(struct gfar_private *priv)
-{
-	struct gfar __iomem *regs = priv->gfargrp[0].regs;
-	u32 attrs;
-
-	/* Stop the DMA engine now, in case it was running before
-	 * (The firmware could have used it, and left it running).
-	 */
-	gfar_halt(priv);
-
-	gfar_mac_reset(priv);
-
-	/* Zero out the rmon mib registers if it has them */
-	if (priv->device_flags & FSL_GIANFAR_DEV_HAS_RMON) {
-		memset_io(&(regs->rmon), 0, sizeof(struct rmon_mib));
-
-		/* Mask off the CAM interrupts */
-		gfar_write(&regs->rmon.cam1, 0xffffffff);
-		gfar_write(&regs->rmon.cam2, 0xffffffff);
-	}
-
-	/* Initialize ECNTRL */
-	gfar_write(&regs->ecntrl, ECNTRL_INIT_SETTINGS);
-
-	/* Set the extraction length and index */
-	attrs = ATTRELI_EL(priv->rx_stash_size) |
-		ATTRELI_EI(priv->rx_stash_index);
-
-	gfar_write(&regs->attreli, attrs);
-
-	/* Start with defaults, and add stashing
-	 * depending on driver parameters
-	 */
-	attrs = ATTR_INIT_SETTINGS;
-
-	if (priv->bd_stash_en)
-		attrs |= ATTR_BDSTASH;
-
-	if (priv->rx_stash_size != 0)
-		attrs |= ATTR_BUFSTASH;
-
-	gfar_write(&regs->attr, attrs);
-
-	/* FIFO configs */
-	gfar_write(&regs->fifo_tx_thr, DEFAULT_FIFO_TX_THR);
-	gfar_write(&regs->fifo_tx_starve, DEFAULT_FIFO_TX_STARVE);
-	gfar_write(&regs->fifo_tx_starve_shutoff, DEFAULT_FIFO_TX_STARVE_OFF);
-
-	/* Program the interrupt steering regs, only for MG devices */
-	if (priv->num_grps > 1)
-		gfar_write_isrg(priv);
-}
-
 static void gfar_init_addr_hash_table(struct gfar_private *priv)
 {
 	struct gfar __iomem *regs = priv->gfargrp[0].regs;
@@ -1307,565 +977,6 @@ static void gfar_init_addr_hash_table(struct gfar_private *priv)
 	}
 }
 
-/* Set up the ethernet device structure, private data,
- * and anything else we need before we start
- */
-static int gfar_probe(struct platform_device *ofdev)
-{
-	struct device_node *np = ofdev->dev.of_node;
-	struct net_device *dev = NULL;
-	struct gfar_private *priv = NULL;
-	int err = 0, i;
-
-	err = gfar_of_init(ofdev, &dev);
-
-	if (err)
-		return err;
-
-	priv = netdev_priv(dev);
-	priv->ndev = dev;
-	priv->ofdev = ofdev;
-	priv->dev = &ofdev->dev;
-	SET_NETDEV_DEV(dev, &ofdev->dev);
-
-	INIT_WORK(&priv->reset_task, gfar_reset_task);
-
-	platform_set_drvdata(ofdev, priv);
-
-	gfar_detect_errata(priv);
-
-	/* Set the dev->base_addr to the gfar reg region */
-	dev->base_addr = (unsigned long) priv->gfargrp[0].regs;
-
-	/* Fill in the dev structure */
-	dev->watchdog_timeo = TX_TIMEOUT;
-	/* MTU range: 50 - 9586 */
-	dev->mtu = 1500;
-	dev->min_mtu = 50;
-	dev->max_mtu = GFAR_JUMBO_FRAME_SIZE - ETH_HLEN;
-	dev->netdev_ops = &gfar_netdev_ops;
-	dev->ethtool_ops = &gfar_ethtool_ops;
-
-	/* Register for napi ...We are registering NAPI for each grp */
-	for (i = 0; i < priv->num_grps; i++) {
-		if (priv->poll_mode == GFAR_SQ_POLLING) {
-			netif_napi_add(dev, &priv->gfargrp[i].napi_rx,
-				       gfar_poll_rx_sq, GFAR_DEV_WEIGHT);
-			netif_tx_napi_add(dev, &priv->gfargrp[i].napi_tx,
-				       gfar_poll_tx_sq, 2);
-		} else {
-			netif_napi_add(dev, &priv->gfargrp[i].napi_rx,
-				       gfar_poll_rx, GFAR_DEV_WEIGHT);
-			netif_tx_napi_add(dev, &priv->gfargrp[i].napi_tx,
-				       gfar_poll_tx, 2);
-		}
-	}
-
-	if (priv->device_flags & FSL_GIANFAR_DEV_HAS_CSUM) {
-		dev->hw_features = NETIF_F_IP_CSUM | NETIF_F_SG |
-				   NETIF_F_RXCSUM;
-		dev->features |= NETIF_F_IP_CSUM | NETIF_F_SG |
-				 NETIF_F_RXCSUM | NETIF_F_HIGHDMA;
-	}
-
-	if (priv->device_flags & FSL_GIANFAR_DEV_HAS_VLAN) {
-		dev->hw_features |= NETIF_F_HW_VLAN_CTAG_TX |
-				    NETIF_F_HW_VLAN_CTAG_RX;
-		dev->features |= NETIF_F_HW_VLAN_CTAG_RX;
-	}
-
-	dev->priv_flags |= IFF_LIVE_ADDR_CHANGE;
-
-	gfar_init_addr_hash_table(priv);
-
-	/* Insert receive time stamps into padding alignment bytes */
-	if (priv->device_flags & FSL_GIANFAR_DEV_HAS_TIMER)
-		priv->padding = 8;
-
-	if (dev->features & NETIF_F_IP_CSUM ||
-	    priv->device_flags & FSL_GIANFAR_DEV_HAS_TIMER)
-		dev->needed_headroom = GMAC_FCB_LEN;
-
-	/* Initializing some of the rx/tx queue level parameters */
-	for (i = 0; i < priv->num_tx_queues; i++) {
-		priv->tx_queue[i]->tx_ring_size = DEFAULT_TX_RING_SIZE;
-		priv->tx_queue[i]->num_txbdfree = DEFAULT_TX_RING_SIZE;
-		priv->tx_queue[i]->txcoalescing = DEFAULT_TX_COALESCE;
-		priv->tx_queue[i]->txic = DEFAULT_TXIC;
-	}
-
-	for (i = 0; i < priv->num_rx_queues; i++) {
-		priv->rx_queue[i]->rx_ring_size = DEFAULT_RX_RING_SIZE;
-		priv->rx_queue[i]->rxcoalescing = DEFAULT_RX_COALESCE;
-		priv->rx_queue[i]->rxic = DEFAULT_RXIC;
-	}
-
-	/* Always enable rx filer if available */
-	priv->rx_filer_enable =
-	    (priv->device_flags & FSL_GIANFAR_DEV_HAS_RX_FILER) ? 1 : 0;
-	/* Enable most messages by default */
-	priv->msg_enable = (NETIF_MSG_IFUP << 1 ) - 1;
-	/* use pritority h/w tx queue scheduling for single queue devices */
-	if (priv->num_tx_queues == 1)
-		priv->prio_sched_en = 1;
-
-	set_bit(GFAR_DOWN, &priv->state);
-
-	gfar_hw_init(priv);
-
-	/* Carrier starts down, phylib will bring it up */
-	netif_carrier_off(dev);
-
-	err = register_netdev(dev);
-
-	if (err) {
-		pr_err("%s: Cannot register net device, aborting\n", dev->name);
-		goto register_fail;
-	}
-
-	if (priv->device_flags & FSL_GIANFAR_DEV_HAS_MAGIC_PACKET)
-		priv->wol_supported |= GFAR_WOL_MAGIC;
-
-	if ((priv->device_flags & FSL_GIANFAR_DEV_HAS_WAKE_ON_FILER) &&
-	    priv->rx_filer_enable)
-		priv->wol_supported |= GFAR_WOL_FILER_UCAST;
-
-	device_set_wakeup_capable(&ofdev->dev, priv->wol_supported);
-
-	/* fill out IRQ number and name fields */
-	for (i = 0; i < priv->num_grps; i++) {
-		struct gfar_priv_grp *grp = &priv->gfargrp[i];
-		if (priv->device_flags & FSL_GIANFAR_DEV_HAS_MULTI_INTR) {
-			sprintf(gfar_irq(grp, TX)->name, "%s%s%c%s",
-				dev->name, "_g", '0' + i, "_tx");
-			sprintf(gfar_irq(grp, RX)->name, "%s%s%c%s",
-				dev->name, "_g", '0' + i, "_rx");
-			sprintf(gfar_irq(grp, ER)->name, "%s%s%c%s",
-				dev->name, "_g", '0' + i, "_er");
-		} else
-			strcpy(gfar_irq(grp, TX)->name, dev->name);
-	}
-
-	/* Initialize the filer table */
-	gfar_init_filer_table(priv);
-
-	/* Print out the device info */
-	netdev_info(dev, "mac: %pM\n", dev->dev_addr);
-
-	/* Even more device info helps when determining which kernel
-	 * provided which set of benchmarks.
-	 */
-	netdev_info(dev, "Running with NAPI enabled\n");
-	for (i = 0; i < priv->num_rx_queues; i++)
-		netdev_info(dev, "RX BD ring size for Q[%d]: %d\n",
-			    i, priv->rx_queue[i]->rx_ring_size);
-	for (i = 0; i < priv->num_tx_queues; i++)
-		netdev_info(dev, "TX BD ring size for Q[%d]: %d\n",
-			    i, priv->tx_queue[i]->tx_ring_size);
-
-	return 0;
-
-register_fail:
-	if (of_phy_is_fixed_link(np))
-		of_phy_deregister_fixed_link(np);
-	unmap_group_regs(priv);
-	gfar_free_rx_queues(priv);
-	gfar_free_tx_queues(priv);
-	of_node_put(priv->phy_node);
-	of_node_put(priv->tbi_node);
-	free_gfar_dev(priv);
-	return err;
-}
-
-static int gfar_remove(struct platform_device *ofdev)
-{
-	struct gfar_private *priv = platform_get_drvdata(ofdev);
-	struct device_node *np = ofdev->dev.of_node;
-
-	of_node_put(priv->phy_node);
-	of_node_put(priv->tbi_node);
-
-	unregister_netdev(priv->ndev);
-
-	if (of_phy_is_fixed_link(np))
-		of_phy_deregister_fixed_link(np);
-
-	unmap_group_regs(priv);
-	gfar_free_rx_queues(priv);
-	gfar_free_tx_queues(priv);
-	free_gfar_dev(priv);
-
-	return 0;
-}
-
-#ifdef CONFIG_PM
-
-static void __gfar_filer_disable(struct gfar_private *priv)
-{
-	struct gfar __iomem *regs = priv->gfargrp[0].regs;
-	u32 temp;
-
-	temp = gfar_read(&regs->rctrl);
-	temp &= ~(RCTRL_FILREN | RCTRL_PRSDEP_INIT);
-	gfar_write(&regs->rctrl, temp);
-}
-
-static void __gfar_filer_enable(struct gfar_private *priv)
-{
-	struct gfar __iomem *regs = priv->gfargrp[0].regs;
-	u32 temp;
-
-	temp = gfar_read(&regs->rctrl);
-	temp |= RCTRL_FILREN | RCTRL_PRSDEP_INIT;
-	gfar_write(&regs->rctrl, temp);
-}
-
-/* Filer rules implementing wol capabilities */
-static void gfar_filer_config_wol(struct gfar_private *priv)
-{
-	unsigned int i;
-	u32 rqfcr;
-
-	__gfar_filer_disable(priv);
-
-	/* clear the filer table, reject any packet by default */
-	rqfcr = RQFCR_RJE | RQFCR_CMP_MATCH;
-	for (i = 0; i <= MAX_FILER_IDX; i++)
-		gfar_write_filer(priv, i, rqfcr, 0);
-
-	i = 0;
-	if (priv->wol_opts & GFAR_WOL_FILER_UCAST) {
-		/* unicast packet, accept it */
-		struct net_device *ndev = priv->ndev;
-		/* get the default rx queue index */
-		u8 qindex = (u8)priv->gfargrp[0].rx_queue->qindex;
-		u32 dest_mac_addr = (ndev->dev_addr[0] << 16) |
-				    (ndev->dev_addr[1] << 8) |
-				     ndev->dev_addr[2];
-
-		rqfcr = (qindex << 10) | RQFCR_AND |
-			RQFCR_CMP_EXACT | RQFCR_PID_DAH;
-
-		gfar_write_filer(priv, i++, rqfcr, dest_mac_addr);
-
-		dest_mac_addr = (ndev->dev_addr[3] << 16) |
-				(ndev->dev_addr[4] << 8) |
-				 ndev->dev_addr[5];
-		rqfcr = (qindex << 10) | RQFCR_GPI |
-			RQFCR_CMP_EXACT | RQFCR_PID_DAL;
-		gfar_write_filer(priv, i++, rqfcr, dest_mac_addr);
-	}
-
-	__gfar_filer_enable(priv);
-}
-
-static void gfar_filer_restore_table(struct gfar_private *priv)
-{
-	u32 rqfcr, rqfpr;
-	unsigned int i;
-
-	__gfar_filer_disable(priv);
-
-	for (i = 0; i <= MAX_FILER_IDX; i++) {
-		rqfcr = priv->ftp_rqfcr[i];
-		rqfpr = priv->ftp_rqfpr[i];
-		gfar_write_filer(priv, i, rqfcr, rqfpr);
-	}
-
-	__gfar_filer_enable(priv);
-}
-
-/* gfar_start() for Rx only and with the FGPI filer interrupt enabled */
-static void gfar_start_wol_filer(struct gfar_private *priv)
-{
-	struct gfar __iomem *regs = priv->gfargrp[0].regs;
-	u32 tempval;
-	int i = 0;
-
-	/* Enable Rx hw queues */
-	gfar_write(&regs->rqueue, priv->rqueue);
-
-	/* Initialize DMACTRL to have WWR and WOP */
-	tempval = gfar_read(&regs->dmactrl);
-	tempval |= DMACTRL_INIT_SETTINGS;
-	gfar_write(&regs->dmactrl, tempval);
-
-	/* Make sure we aren't stopped */
-	tempval = gfar_read(&regs->dmactrl);
-	tempval &= ~DMACTRL_GRS;
-	gfar_write(&regs->dmactrl, tempval);
-
-	for (i = 0; i < priv->num_grps; i++) {
-		regs = priv->gfargrp[i].regs;
-		/* Clear RHLT, so that the DMA starts polling now */
-		gfar_write(&regs->rstat, priv->gfargrp[i].rstat);
-		/* enable the Filer General Purpose Interrupt */
-		gfar_write(&regs->imask, IMASK_FGPI);
-	}
-
-	/* Enable Rx DMA */
-	tempval = gfar_read(&regs->maccfg1);
-	tempval |= MACCFG1_RX_EN;
-	gfar_write(&regs->maccfg1, tempval);
-}
-
-static int gfar_suspend(struct device *dev)
-{
-	struct gfar_private *priv = dev_get_drvdata(dev);
-	struct net_device *ndev = priv->ndev;
-	struct gfar __iomem *regs = priv->gfargrp[0].regs;
-	u32 tempval;
-	u16 wol = priv->wol_opts;
-
-	if (!netif_running(ndev))
-		return 0;
-
-	disable_napi(priv);
-	netif_tx_lock(ndev);
-	netif_device_detach(ndev);
-	netif_tx_unlock(ndev);
-
-	gfar_halt(priv);
-
-	if (wol & GFAR_WOL_MAGIC) {
-		/* Enable interrupt on Magic Packet */
-		gfar_write(&regs->imask, IMASK_MAG);
-
-		/* Enable Magic Packet mode */
-		tempval = gfar_read(&regs->maccfg2);
-		tempval |= MACCFG2_MPEN;
-		gfar_write(&regs->maccfg2, tempval);
-
-		/* re-enable the Rx block */
-		tempval = gfar_read(&regs->maccfg1);
-		tempval |= MACCFG1_RX_EN;
-		gfar_write(&regs->maccfg1, tempval);
-
-	} else if (wol & GFAR_WOL_FILER_UCAST) {
-		gfar_filer_config_wol(priv);
-		gfar_start_wol_filer(priv);
-
-	} else {
-		phy_stop(ndev->phydev);
-	}
-
-	return 0;
-}
-
-static int gfar_resume(struct device *dev)
-{
-	struct gfar_private *priv = dev_get_drvdata(dev);
-	struct net_device *ndev = priv->ndev;
-	struct gfar __iomem *regs = priv->gfargrp[0].regs;
-	u32 tempval;
-	u16 wol = priv->wol_opts;
-
-	if (!netif_running(ndev))
-		return 0;
-
-	if (wol & GFAR_WOL_MAGIC) {
-		/* Disable Magic Packet mode */
-		tempval = gfar_read(&regs->maccfg2);
-		tempval &= ~MACCFG2_MPEN;
-		gfar_write(&regs->maccfg2, tempval);
-
-	} else if (wol & GFAR_WOL_FILER_UCAST) {
-		/* need to stop rx only, tx is already down */
-		gfar_halt(priv);
-		gfar_filer_restore_table(priv);
-
-	} else {
-		phy_start(ndev->phydev);
-	}
-
-	gfar_start(priv);
-
-	netif_device_attach(ndev);
-	enable_napi(priv);
-
-	return 0;
-}
-
-static int gfar_restore(struct device *dev)
-{
-	struct gfar_private *priv = dev_get_drvdata(dev);
-	struct net_device *ndev = priv->ndev;
-
-	if (!netif_running(ndev)) {
-		netif_device_attach(ndev);
-
-		return 0;
-	}
-
-	gfar_init_bds(ndev);
-
-	gfar_mac_reset(priv);
-
-	gfar_init_tx_rx_base(priv);
-
-	gfar_start(priv);
-
-	priv->oldlink = 0;
-	priv->oldspeed = 0;
-	priv->oldduplex = -1;
-
-	if (ndev->phydev)
-		phy_start(ndev->phydev);
-
-	netif_device_attach(ndev);
-	enable_napi(priv);
-
-	return 0;
-}
-
-static struct dev_pm_ops gfar_pm_ops = {
-	.suspend = gfar_suspend,
-	.resume = gfar_resume,
-	.freeze = gfar_suspend,
-	.thaw = gfar_resume,
-	.restore = gfar_restore,
-};
-
-#define GFAR_PM_OPS (&gfar_pm_ops)
-
-#else
-
-#define GFAR_PM_OPS NULL
-
-#endif
-
-/* Reads the controller's registers to determine what interface
- * connects it to the PHY.
- */
-static phy_interface_t gfar_get_interface(struct net_device *dev)
-{
-	struct gfar_private *priv = netdev_priv(dev);
-	struct gfar __iomem *regs = priv->gfargrp[0].regs;
-	u32 ecntrl;
-
-	ecntrl = gfar_read(&regs->ecntrl);
-
-	if (ecntrl & ECNTRL_SGMII_MODE)
-		return PHY_INTERFACE_MODE_SGMII;
-
-	if (ecntrl & ECNTRL_TBI_MODE) {
-		if (ecntrl & ECNTRL_REDUCED_MODE)
-			return PHY_INTERFACE_MODE_RTBI;
-		else
-			return PHY_INTERFACE_MODE_TBI;
-	}
-
-	if (ecntrl & ECNTRL_REDUCED_MODE) {
-		if (ecntrl & ECNTRL_REDUCED_MII_MODE) {
-			return PHY_INTERFACE_MODE_RMII;
-		}
-		else {
-			phy_interface_t interface = priv->interface;
-
-			/* This isn't autodetected right now, so it must
-			 * be set by the device tree or platform code.
-			 */
-			if (interface == PHY_INTERFACE_MODE_RGMII_ID)
-				return PHY_INTERFACE_MODE_RGMII_ID;
-
-			return PHY_INTERFACE_MODE_RGMII;
-		}
-	}
-
-	if (priv->device_flags & FSL_GIANFAR_DEV_HAS_GIGABIT)
-		return PHY_INTERFACE_MODE_GMII;
-
-	return PHY_INTERFACE_MODE_MII;
-}
-
-
-/* Initializes driver's PHY state, and attaches to the PHY.
- * Returns 0 on success.
- */
-static int init_phy(struct net_device *dev)
-{
-	struct gfar_private *priv = netdev_priv(dev);
-	uint gigabit_support =
-		priv->device_flags & FSL_GIANFAR_DEV_HAS_GIGABIT ?
-		GFAR_SUPPORTED_GBIT : 0;
-	phy_interface_t interface;
-	struct phy_device *phydev;
-
-	priv->oldlink = 0;
-	priv->oldspeed = 0;
-	priv->oldduplex = -1;
-
-	interface = gfar_get_interface(dev);
-
-	phydev = of_phy_connect(dev, priv->phy_node, &adjust_link, 0,
-				interface);
-	if (!phydev) {
-		dev_err(&dev->dev, "could not attach to PHY\n");
-		return -ENODEV;
-	}
-
-	if (interface == PHY_INTERFACE_MODE_SGMII)
-		gfar_configure_serdes(dev);
-
-	/* Remove any features not supported by the controller */
-	phydev->supported &= (GFAR_SUPPORTED | gigabit_support);
-	phydev->advertising = phydev->supported;
-
-	/* Add support for flow control, but don't advertise it by default */
-	phydev->supported |= (SUPPORTED_Pause | SUPPORTED_Asym_Pause);
-
-	return 0;
-}
-
-/* Initialize TBI PHY interface for communicating with the
- * SERDES lynx PHY on the chip.  We communicate with this PHY
- * through the MDIO bus on each controller, treating it as a
- * "normal" PHY at the address found in the TBIPA register.  We assume
- * that the TBIPA register is valid.  Either the MDIO bus code will set
- * it to a value that doesn't conflict with other PHYs on the bus, or the
- * value doesn't matter, as there are no other PHYs on the bus.
- */
-static void gfar_configure_serdes(struct net_device *dev)
-{
-	struct gfar_private *priv = netdev_priv(dev);
-	struct phy_device *tbiphy;
-
-	if (!priv->tbi_node) {
-		dev_warn(&dev->dev, "error: SGMII mode requires that the "
-				    "device tree specify a tbi-handle\n");
-		return;
-	}
-
-	tbiphy = of_phy_find_device(priv->tbi_node);
-	if (!tbiphy) {
-		dev_err(&dev->dev, "error: Could not get TBI device\n");
-		return;
-	}
-
-	/* If the link is already up, we must already be ok, and don't need to
-	 * configure and reset the TBI<->SerDes link.  Maybe U-Boot configured
-	 * everything for us?  Resetting it takes the link down and requires
-	 * several seconds for it to come back.
-	 */
-	if (phy_read(tbiphy, MII_BMSR) & BMSR_LSTATUS) {
-		put_device(&tbiphy->mdio.dev);
-		return;
-	}
-
-	/* Single clk mode, mii mode off(for serdes communication) */
-	phy_write(tbiphy, MII_TBICON, TBICON_CLK_SELECT);
-
-	phy_write(tbiphy, MII_ADVERTISE,
-		  ADVERTISE_1000XFULL | ADVERTISE_1000XPAUSE |
-		  ADVERTISE_1000XPSE_ASYM);
-
-	phy_write(tbiphy, MII_BMCR,
-		  BMCR_ANENABLE | BMCR_ANRESTART | BMCR_FULLDPLX |
-		  BMCR_SPEED1000);
-
-	put_device(&tbiphy->mdio.dev);
-}
-
 static int __gfar_is_rx_idle(struct gfar_private *priv)
 {
 	u32 res;
@@ -1922,7 +1033,7 @@ retry:
 }
 
 /* Halt the receive and transmit queues */
-void gfar_halt(struct gfar_private *priv)
+static void gfar_halt(struct gfar_private *priv)
 {
 	struct gfar __iomem *regs = priv->gfargrp[0].regs;
 	u32 tempval;
@@ -1941,26 +1052,6 @@ void gfar_halt(struct gfar_private *priv)
 	gfar_write(&regs->maccfg1, tempval);
 }
 
-void stop_gfar(struct net_device *dev)
-{
-	struct gfar_private *priv = netdev_priv(dev);
-
-	netif_tx_stop_all_queues(dev);
-
-	smp_mb__before_atomic();
-	set_bit(GFAR_DOWN, &priv->state);
-	smp_mb__after_atomic();
-
-	disable_napi(priv);
-
-	/* disable ints and gracefully shut down Rx/Tx DMA */
-	gfar_halt(priv);
-
-	phy_stop(dev->phydev);
-
-	free_skb_resources(priv);
-}
-
 static void free_skb_tx_queue(struct gfar_priv_tx_q *tx_queue)
 {
 	struct txbd8 *txbdp;
@@ -1997,8 +1088,7 @@ static void free_skb_rx_queue(struct gfar_priv_rx_q *rx_queue)
 
 	struct rxbd8 *rxbdp = rx_queue->rx_bd_base;
 
-	if (rx_queue->skb)
-		dev_kfree_skb(rx_queue->skb);
+	dev_kfree_skb(rx_queue->skb);
 
 	for (i = 0; i < rx_queue->rx_ring_size; i++) {
 		struct	gfar_rx_buff *rxb = &rx_queue->rx_buff[i];
@@ -2054,7 +1144,27 @@ static void free_skb_resources(struct gfar_private *priv)
 			  priv->tx_queue[0]->tx_bd_dma_base);
 }
 
-void gfar_start(struct gfar_private *priv)
+void stop_gfar(struct net_device *dev)
+{
+	struct gfar_private *priv = netdev_priv(dev);
+
+	netif_tx_stop_all_queues(dev);
+
+	smp_mb__before_atomic();
+	set_bit(GFAR_DOWN, &priv->state);
+	smp_mb__after_atomic();
+
+	disable_napi(priv);
+
+	/* disable ints and gracefully shut down Rx/Tx DMA */
+	gfar_halt(priv);
+
+	phy_stop(dev->phydev);
+
+	free_skb_resources(priv);
+}
+
+static void gfar_start(struct gfar_private *priv)
 {
 	struct gfar __iomem *regs = priv->gfargrp[0].regs;
 	u32 tempval;
@@ -2091,103 +1201,207 @@ void gfar_start(struct gfar_private *priv)
 	netif_trans_update(priv->ndev); /* prevent tx timeout */
 }
 
-static void free_grp_irqs(struct gfar_priv_grp *grp)
+static bool gfar_new_page(struct gfar_priv_rx_q *rxq, struct gfar_rx_buff *rxb)
 {
-	free_irq(gfar_irq(grp, TX)->irq, grp);
-	free_irq(gfar_irq(grp, RX)->irq, grp);
-	free_irq(gfar_irq(grp, ER)->irq, grp);
+	struct page *page;
+	dma_addr_t addr;
+
+	page = dev_alloc_page();
+	if (unlikely(!page))
+		return false;
+
+	addr = dma_map_page(rxq->dev, page, 0, PAGE_SIZE, DMA_FROM_DEVICE);
+	if (unlikely(dma_mapping_error(rxq->dev, addr))) {
+		__free_page(page);
+
+		return false;
+	}
+
+	rxb->dma = addr;
+	rxb->page = page;
+	rxb->page_offset = 0;
+
+	return true;
 }
 
-static int register_grp_irqs(struct gfar_priv_grp *grp)
+static void gfar_rx_alloc_err(struct gfar_priv_rx_q *rx_queue)
 {
-	struct gfar_private *priv = grp->priv;
-	struct net_device *dev = priv->ndev;
-	int err;
+	struct gfar_private *priv = netdev_priv(rx_queue->ndev);
+	struct gfar_extra_stats *estats = &priv->extra_stats;
 
-	/* If the device has multiple interrupts, register for
-	 * them.  Otherwise, only register for the one
-	 */
-	if (priv->device_flags & FSL_GIANFAR_DEV_HAS_MULTI_INTR) {
-		/* Install our interrupt handlers for Error,
-		 * Transmit, and Receive
-		 */
-		err = request_irq(gfar_irq(grp, ER)->irq, gfar_error, 0,
-				  gfar_irq(grp, ER)->name, grp);
-		if (err < 0) {
-			netif_err(priv, intr, dev, "Can't get IRQ %d\n",
-				  gfar_irq(grp, ER)->irq);
+	netdev_err(rx_queue->ndev, "Can't alloc RX buffers\n");
+	atomic64_inc(&estats->rx_alloc_err);
+}
 
-			goto err_irq_fail;
-		}
-		enable_irq_wake(gfar_irq(grp, ER)->irq);
+static void gfar_alloc_rx_buffs(struct gfar_priv_rx_q *rx_queue,
+				int alloc_cnt)
+{
+	struct rxbd8 *bdp;
+	struct gfar_rx_buff *rxb;
+	int i;
 
-		err = request_irq(gfar_irq(grp, TX)->irq, gfar_transmit, 0,
-				  gfar_irq(grp, TX)->name, grp);
-		if (err < 0) {
-			netif_err(priv, intr, dev, "Can't get IRQ %d\n",
-				  gfar_irq(grp, TX)->irq);
-			goto tx_irq_fail;
+	i = rx_queue->next_to_use;
+	bdp = &rx_queue->rx_bd_base[i];
+	rxb = &rx_queue->rx_buff[i];
+
+	while (alloc_cnt--) {
+		/* try reuse page */
+		if (unlikely(!rxb->page)) {
+			if (unlikely(!gfar_new_page(rx_queue, rxb))) {
+				gfar_rx_alloc_err(rx_queue);
+				break;
+			}
 		}
-		err = request_irq(gfar_irq(grp, RX)->irq, gfar_receive, 0,
-				  gfar_irq(grp, RX)->name, grp);
-		if (err < 0) {
-			netif_err(priv, intr, dev, "Can't get IRQ %d\n",
-				  gfar_irq(grp, RX)->irq);
-			goto rx_irq_fail;
+
+		/* Setup the new RxBD */
+		gfar_init_rxbdp(rx_queue, bdp,
+				rxb->dma + rxb->page_offset + RXBUF_ALIGNMENT);
+
+		/* Update to the next pointer */
+		bdp++;
+		rxb++;
+
+		if (unlikely(++i == rx_queue->rx_ring_size)) {
+			i = 0;
+			bdp = rx_queue->rx_bd_base;
+			rxb = rx_queue->rx_buff;
 		}
-		enable_irq_wake(gfar_irq(grp, RX)->irq);
+	}
 
-	} else {
-		err = request_irq(gfar_irq(grp, TX)->irq, gfar_interrupt, 0,
-				  gfar_irq(grp, TX)->name, grp);
-		if (err < 0) {
-			netif_err(priv, intr, dev, "Can't get IRQ %d\n",
-				  gfar_irq(grp, TX)->irq);
-			goto err_irq_fail;
+	rx_queue->next_to_use = i;
+	rx_queue->next_to_alloc = i;
+}
+
+static void gfar_init_bds(struct net_device *ndev)
+{
+	struct gfar_private *priv = netdev_priv(ndev);
+	struct gfar __iomem *regs = priv->gfargrp[0].regs;
+	struct gfar_priv_tx_q *tx_queue = NULL;
+	struct gfar_priv_rx_q *rx_queue = NULL;
+	struct txbd8 *txbdp;
+	u32 __iomem *rfbptr;
+	int i, j;
+
+	for (i = 0; i < priv->num_tx_queues; i++) {
+		tx_queue = priv->tx_queue[i];
+		/* Initialize some variables in our dev structure */
+		tx_queue->num_txbdfree = tx_queue->tx_ring_size;
+		tx_queue->dirty_tx = tx_queue->tx_bd_base;
+		tx_queue->cur_tx = tx_queue->tx_bd_base;
+		tx_queue->skb_curtx = 0;
+		tx_queue->skb_dirtytx = 0;
+
+		/* Initialize Transmit Descriptor Ring */
+		txbdp = tx_queue->tx_bd_base;
+		for (j = 0; j < tx_queue->tx_ring_size; j++) {
+			txbdp->lstatus = 0;
+			txbdp->bufPtr = 0;
+			txbdp++;
 		}
-		enable_irq_wake(gfar_irq(grp, TX)->irq);
+
+		/* Set the last descriptor in the ring to indicate wrap */
+		txbdp--;
+		txbdp->status = cpu_to_be16(be16_to_cpu(txbdp->status) |
+					    TXBD_WRAP);
 	}
 
-	return 0;
+	rfbptr = &regs->rfbptr0;
+	for (i = 0; i < priv->num_rx_queues; i++) {
+		rx_queue = priv->rx_queue[i];
 
-rx_irq_fail:
-	free_irq(gfar_irq(grp, TX)->irq, grp);
-tx_irq_fail:
-	free_irq(gfar_irq(grp, ER)->irq, grp);
-err_irq_fail:
-	return err;
+		rx_queue->next_to_clean = 0;
+		rx_queue->next_to_use = 0;
+		rx_queue->next_to_alloc = 0;
+
+		/* make sure next_to_clean != next_to_use after this
+		 * by leaving at least 1 unused descriptor
+		 */
+		gfar_alloc_rx_buffs(rx_queue, gfar_rxbd_unused(rx_queue));
 
+		rx_queue->rfbptr = rfbptr;
+		rfbptr += 2;
+	}
 }
 
-static void gfar_free_irq(struct gfar_private *priv)
+static int gfar_alloc_skb_resources(struct net_device *ndev)
 {
-	int i;
+	void *vaddr;
+	dma_addr_t addr;
+	int i, j;
+	struct gfar_private *priv = netdev_priv(ndev);
+	struct device *dev = priv->dev;
+	struct gfar_priv_tx_q *tx_queue = NULL;
+	struct gfar_priv_rx_q *rx_queue = NULL;
 
-	/* Free the IRQs */
-	if (priv->device_flags & FSL_GIANFAR_DEV_HAS_MULTI_INTR) {
-		for (i = 0; i < priv->num_grps; i++)
-			free_grp_irqs(&priv->gfargrp[i]);
-	} else {
-		for (i = 0; i < priv->num_grps; i++)
-			free_irq(gfar_irq(&priv->gfargrp[i], TX)->irq,
-				 &priv->gfargrp[i]);
+	priv->total_tx_ring_size = 0;
+	for (i = 0; i < priv->num_tx_queues; i++)
+		priv->total_tx_ring_size += priv->tx_queue[i]->tx_ring_size;
+
+	priv->total_rx_ring_size = 0;
+	for (i = 0; i < priv->num_rx_queues; i++)
+		priv->total_rx_ring_size += priv->rx_queue[i]->rx_ring_size;
+
+	/* Allocate memory for the buffer descriptors */
+	vaddr = dma_alloc_coherent(dev,
+				   (priv->total_tx_ring_size *
+				    sizeof(struct txbd8)) +
+				   (priv->total_rx_ring_size *
+				    sizeof(struct rxbd8)),
+				   &addr, GFP_KERNEL);
+	if (!vaddr)
+		return -ENOMEM;
+
+	for (i = 0; i < priv->num_tx_queues; i++) {
+		tx_queue = priv->tx_queue[i];
+		tx_queue->tx_bd_base = vaddr;
+		tx_queue->tx_bd_dma_base = addr;
+		tx_queue->dev = ndev;
+		/* enet DMA only understands physical addresses */
+		addr  += sizeof(struct txbd8) * tx_queue->tx_ring_size;
+		vaddr += sizeof(struct txbd8) * tx_queue->tx_ring_size;
 	}
-}
 
-static int gfar_request_irq(struct gfar_private *priv)
-{
-	int err, i, j;
+	/* Start the rx descriptor ring where the tx ring leaves off */
+	for (i = 0; i < priv->num_rx_queues; i++) {
+		rx_queue = priv->rx_queue[i];
+		rx_queue->rx_bd_base = vaddr;
+		rx_queue->rx_bd_dma_base = addr;
+		rx_queue->ndev = ndev;
+		rx_queue->dev = dev;
+		addr  += sizeof(struct rxbd8) * rx_queue->rx_ring_size;
+		vaddr += sizeof(struct rxbd8) * rx_queue->rx_ring_size;
+	}
 
-	for (i = 0; i < priv->num_grps; i++) {
-		err = register_grp_irqs(&priv->gfargrp[i]);
-		if (err) {
-			for (j = 0; j < i; j++)
-				free_grp_irqs(&priv->gfargrp[j]);
-			return err;
-		}
+	/* Setup the skbuff rings */
+	for (i = 0; i < priv->num_tx_queues; i++) {
+		tx_queue = priv->tx_queue[i];
+		tx_queue->tx_skbuff =
+			kmalloc_array(tx_queue->tx_ring_size,
+				      sizeof(*tx_queue->tx_skbuff),
+				      GFP_KERNEL);
+		if (!tx_queue->tx_skbuff)
+			goto cleanup;
+
+		for (j = 0; j < tx_queue->tx_ring_size; j++)
+			tx_queue->tx_skbuff[j] = NULL;
 	}
 
+	for (i = 0; i < priv->num_rx_queues; i++) {
+		rx_queue = priv->rx_queue[i];
+		rx_queue->rx_buff = kcalloc(rx_queue->rx_ring_size,
+					    sizeof(*rx_queue->rx_buff),
+					    GFP_KERNEL);
+		if (!rx_queue->rx_buff)
+			goto cleanup;
+	}
+
+	gfar_init_bds(ndev);
+
 	return 0;
+
+cleanup:
+	free_skb_resources(priv);
+	return -ENOMEM;
 }
 
 /* Bring the controller up and running */
@@ -2225,32 +1439,240 @@ int startup_gfar(struct net_device *ndev)
 	return 0;
 }
 
-/* Called when something needs to use the ethernet device
- * Returns 0 for success.
+static u32 gfar_get_flowctrl_cfg(struct gfar_private *priv)
+{
+	struct net_device *ndev = priv->ndev;
+	struct phy_device *phydev = ndev->phydev;
+	u32 val = 0;
+
+	if (!phydev->duplex)
+		return val;
+
+	if (!priv->pause_aneg_en) {
+		if (priv->tx_pause_en)
+			val |= MACCFG1_TX_FLOW;
+		if (priv->rx_pause_en)
+			val |= MACCFG1_RX_FLOW;
+	} else {
+		u16 lcl_adv, rmt_adv;
+		u8 flowctrl;
+		/* get link partner capabilities */
+		rmt_adv = 0;
+		if (phydev->pause)
+			rmt_adv = LPA_PAUSE_CAP;
+		if (phydev->asym_pause)
+			rmt_adv |= LPA_PAUSE_ASYM;
+
+		lcl_adv = linkmode_adv_to_lcl_adv_t(phydev->advertising);
+		flowctrl = mii_resolve_flowctrl_fdx(lcl_adv, rmt_adv);
+		if (flowctrl & FLOW_CTRL_TX)
+			val |= MACCFG1_TX_FLOW;
+		if (flowctrl & FLOW_CTRL_RX)
+			val |= MACCFG1_RX_FLOW;
+	}
+
+	return val;
+}
+
+static noinline void gfar_update_link_state(struct gfar_private *priv)
+{
+	struct gfar __iomem *regs = priv->gfargrp[0].regs;
+	struct net_device *ndev = priv->ndev;
+	struct phy_device *phydev = ndev->phydev;
+	struct gfar_priv_rx_q *rx_queue = NULL;
+	int i;
+
+	if (unlikely(test_bit(GFAR_RESETTING, &priv->state)))
+		return;
+
+	if (phydev->link) {
+		u32 tempval1 = gfar_read(&regs->maccfg1);
+		u32 tempval = gfar_read(&regs->maccfg2);
+		u32 ecntrl = gfar_read(&regs->ecntrl);
+		u32 tx_flow_oldval = (tempval1 & MACCFG1_TX_FLOW);
+
+		if (phydev->duplex != priv->oldduplex) {
+			if (!(phydev->duplex))
+				tempval &= ~(MACCFG2_FULL_DUPLEX);
+			else
+				tempval |= MACCFG2_FULL_DUPLEX;
+
+			priv->oldduplex = phydev->duplex;
+		}
+
+		if (phydev->speed != priv->oldspeed) {
+			switch (phydev->speed) {
+			case 1000:
+				tempval =
+				    ((tempval & ~(MACCFG2_IF)) | MACCFG2_GMII);
+
+				ecntrl &= ~(ECNTRL_R100);
+				break;
+			case 100:
+			case 10:
+				tempval =
+				    ((tempval & ~(MACCFG2_IF)) | MACCFG2_MII);
+
+				/* Reduced mode distinguishes
+				 * between 10 and 100
+				 */
+				if (phydev->speed == SPEED_100)
+					ecntrl |= ECNTRL_R100;
+				else
+					ecntrl &= ~(ECNTRL_R100);
+				break;
+			default:
+				netif_warn(priv, link, priv->ndev,
+					   "Ack!  Speed (%d) is not 10/100/1000!\n",
+					   phydev->speed);
+				break;
+			}
+
+			priv->oldspeed = phydev->speed;
+		}
+
+		tempval1 &= ~(MACCFG1_TX_FLOW | MACCFG1_RX_FLOW);
+		tempval1 |= gfar_get_flowctrl_cfg(priv);
+
+		/* Turn last free buffer recording on */
+		if ((tempval1 & MACCFG1_TX_FLOW) && !tx_flow_oldval) {
+			for (i = 0; i < priv->num_rx_queues; i++) {
+				u32 bdp_dma;
+
+				rx_queue = priv->rx_queue[i];
+				bdp_dma = gfar_rxbd_dma_lastfree(rx_queue);
+				gfar_write(rx_queue->rfbptr, bdp_dma);
+			}
+
+			priv->tx_actual_en = 1;
+		}
+
+		if (unlikely(!(tempval1 & MACCFG1_TX_FLOW) && tx_flow_oldval))
+			priv->tx_actual_en = 0;
+
+		gfar_write(&regs->maccfg1, tempval1);
+		gfar_write(&regs->maccfg2, tempval);
+		gfar_write(&regs->ecntrl, ecntrl);
+
+		if (!priv->oldlink)
+			priv->oldlink = 1;
+
+	} else if (priv->oldlink) {
+		priv->oldlink = 0;
+		priv->oldspeed = 0;
+		priv->oldduplex = -1;
+	}
+
+	if (netif_msg_link(priv))
+		phy_print_status(phydev);
+}
+
+/* Called every time the controller might need to be made
+ * aware of new link state.  The PHY code conveys this
+ * information through variables in the phydev structure, and this
+ * function converts those variables into the appropriate
+ * register values, and can bring down the device if needed.
  */
-static int gfar_enet_open(struct net_device *dev)
+static void adjust_link(struct net_device *dev)
 {
 	struct gfar_private *priv = netdev_priv(dev);
-	int err;
+	struct phy_device *phydev = dev->phydev;
 
-	err = init_phy(dev);
-	if (err)
-		return err;
+	if (unlikely(phydev->link != priv->oldlink ||
+		     (phydev->link && (phydev->duplex != priv->oldduplex ||
+				       phydev->speed != priv->oldspeed))))
+		gfar_update_link_state(priv);
+}
 
-	err = gfar_request_irq(priv);
-	if (err)
-		return err;
+/* Initialize TBI PHY interface for communicating with the
+ * SERDES lynx PHY on the chip.  We communicate with this PHY
+ * through the MDIO bus on each controller, treating it as a
+ * "normal" PHY at the address found in the TBIPA register.  We assume
+ * that the TBIPA register is valid.  Either the MDIO bus code will set
+ * it to a value that doesn't conflict with other PHYs on the bus, or the
+ * value doesn't matter, as there are no other PHYs on the bus.
+ */
+static void gfar_configure_serdes(struct net_device *dev)
+{
+	struct gfar_private *priv = netdev_priv(dev);
+	struct phy_device *tbiphy;
 
-	err = startup_gfar(dev);
-	if (err)
-		return err;
+	if (!priv->tbi_node) {
+		dev_warn(&dev->dev, "error: SGMII mode requires that the "
+				    "device tree specify a tbi-handle\n");
+		return;
+	}
 
-	return err;
+	tbiphy = of_phy_find_device(priv->tbi_node);
+	if (!tbiphy) {
+		dev_err(&dev->dev, "error: Could not get TBI device\n");
+		return;
+	}
+
+	/* If the link is already up, we must already be ok, and don't need to
+	 * configure and reset the TBI<->SerDes link.  Maybe U-Boot configured
+	 * everything for us?  Resetting it takes the link down and requires
+	 * several seconds for it to come back.
+	 */
+	if (phy_read(tbiphy, MII_BMSR) & BMSR_LSTATUS) {
+		put_device(&tbiphy->mdio.dev);
+		return;
+	}
+
+	/* Single clk mode, mii mode off(for serdes communication) */
+	phy_write(tbiphy, MII_TBICON, TBICON_CLK_SELECT);
+
+	phy_write(tbiphy, MII_ADVERTISE,
+		  ADVERTISE_1000XFULL | ADVERTISE_1000XPAUSE |
+		  ADVERTISE_1000XPSE_ASYM);
+
+	phy_write(tbiphy, MII_BMCR,
+		  BMCR_ANENABLE | BMCR_ANRESTART | BMCR_FULLDPLX |
+		  BMCR_SPEED1000);
+
+	put_device(&tbiphy->mdio.dev);
+}
+
+/* Initializes driver's PHY state, and attaches to the PHY.
+ * Returns 0 on success.
+ */
+static int init_phy(struct net_device *dev)
+{
+	struct gfar_private *priv = netdev_priv(dev);
+	phy_interface_t interface = priv->interface;
+	struct phy_device *phydev;
+	struct ethtool_keee edata;
+
+	priv->oldlink = 0;
+	priv->oldspeed = 0;
+	priv->oldduplex = -1;
+
+	phydev = of_phy_connect(dev, priv->phy_node, &adjust_link, 0,
+				interface);
+	if (!phydev) {
+		dev_err(&dev->dev, "could not attach to PHY\n");
+		return -ENODEV;
+	}
+
+	if (interface == PHY_INTERFACE_MODE_SGMII)
+		gfar_configure_serdes(dev);
+
+	if (!(priv->device_flags & FSL_GIANFAR_DEV_HAS_GIGABIT))
+		phy_set_max_speed(phydev, SPEED_100);
+
+	/* Add support for flow control */
+	phy_support_asym_pause(phydev);
+
+	/* disable EEE autoneg, EEE not supported by eTSEC */
+	memset(&edata, 0, sizeof(struct ethtool_keee));
+	phy_ethtool_set_eee(phydev, &edata);
+
+	return 0;
 }
 
 static inline struct txfcb *gfar_add_fcb(struct sk_buff *skb)
 {
-	struct txfcb *fcb = (struct txfcb *)skb_push(skb, GMAC_FCB_LEN);
+	struct txfcb *fcb = skb_push(skb, GMAC_FCB_LEN);
 
 	memset(fcb, 0, GMAC_FCB_LEN);
 
@@ -2327,7 +1749,7 @@ static inline bool gfar_csum_errata_76(struct gfar_private *priv,
 /* This is called by the kernel when a frame is ready for transmission.
  * It is pointed to by the dev->hard_start_xmit function pointer
  */
-static int gfar_start_xmit(struct sk_buff *skb, struct net_device *dev)
+static netdev_tx_t gfar_start_xmit(struct sk_buff *skb, struct net_device *dev)
 {
 	struct gfar_private *priv = netdev_priv(dev);
 	struct gfar_priv_tx_q *tx_queue = NULL;
@@ -2361,20 +1783,12 @@ static int gfar_start_xmit(struct sk_buff *skb, struct net_device *dev)
 		fcb_len = GMAC_FCB_LEN + GMAC_TXPAL_LEN;
 
 	/* make space for additional header when fcb is needed */
-	if (fcb_len && unlikely(skb_headroom(skb) < fcb_len)) {
-		struct sk_buff *skb_new;
-
-		skb_new = skb_realloc_headroom(skb, fcb_len);
-		if (!skb_new) {
+	if (fcb_len) {
+		if (unlikely(skb_cow_head(skb, fcb_len))) {
 			dev->stats.tx_errors++;
 			dev_kfree_skb_any(skb);
 			return NETDEV_TX_OK;
 		}
-
-		if (skb->sk)
-			skb_set_owner_w(skb_new, skb->sk);
-		dev_consume_skb_any(skb);
-		skb = skb_new;
 	}
 
 	/* total number of fragments in the SKB */
@@ -2513,6 +1927,7 @@ static int gfar_start_xmit(struct sk_buff *skb, struct net_device *dev)
 		lstatus |= BD_LFLAG(TXBD_CRC | TXBD_READY) | skb_headlen(skb);
 	}
 
+	skb_tx_timestamp(skb);
 	netdev_tx_sent_queue(txq, bytes_sent);
 
 	gfar_wmb();
@@ -2576,22 +1991,6 @@ dma_map_err:
 	return NETDEV_TX_OK;
 }
 
-/* Stops the kernel queue, and halts the controller */
-static int gfar_close(struct net_device *dev)
-{
-	struct gfar_private *priv = netdev_priv(dev);
-
-	cancel_work_sync(&priv->reset_task);
-	stop_gfar(dev);
-
-	/* Disconnect from the PHY */
-	phy_disconnect(dev->phydev);
-
-	gfar_free_irq(priv);
-
-	return 0;
-}
-
 /* Changes the mac address if the controller is not running. */
 static int gfar_set_mac_address(struct net_device *dev)
 {
@@ -2610,7 +2009,7 @@ static int gfar_change_mtu(struct net_device *dev, int new_mtu)
 	if (dev->flags & IFF_UP)
 		stop_gfar(dev);
 
-	dev->mtu = new_mtu;
+	WRITE_ONCE(dev->mtu, new_mtu);
 
 	if (dev->flags & IFF_UP)
 		startup_gfar(dev);
@@ -2620,7 +2019,7 @@ static int gfar_change_mtu(struct net_device *dev, int new_mtu)
 	return 0;
 }
 
-void reset_gfar(struct net_device *ndev)
+static void reset_gfar(struct net_device *ndev)
 {
 	struct gfar_private *priv = netdev_priv(ndev);
 
@@ -2645,7 +2044,7 @@ static void gfar_reset_task(struct work_struct *work)
 	reset_gfar(priv->ndev);
 }
 
-static void gfar_timeout(struct net_device *dev)
+static void gfar_timeout(struct net_device *dev, unsigned int txqueue)
 {
 	struct gfar_private *priv = netdev_priv(dev);
 
@@ -2653,6 +2052,58 @@ static void gfar_timeout(struct net_device *dev)
 	schedule_work(&priv->reset_task);
 }
 
+static int gfar_hwtstamp_set(struct net_device *netdev,
+			     struct kernel_hwtstamp_config *config,
+			     struct netlink_ext_ack *extack)
+{
+	struct gfar_private *priv = netdev_priv(netdev);
+
+	switch (config->tx_type) {
+	case HWTSTAMP_TX_OFF:
+		priv->hwts_tx_en = 0;
+		break;
+	case HWTSTAMP_TX_ON:
+		if (!(priv->device_flags & FSL_GIANFAR_DEV_HAS_TIMER))
+			return -ERANGE;
+		priv->hwts_tx_en = 1;
+		break;
+	default:
+		return -ERANGE;
+	}
+
+	switch (config->rx_filter) {
+	case HWTSTAMP_FILTER_NONE:
+		if (priv->hwts_rx_en) {
+			priv->hwts_rx_en = 0;
+			reset_gfar(netdev);
+		}
+		break;
+	default:
+		if (!(priv->device_flags & FSL_GIANFAR_DEV_HAS_TIMER))
+			return -ERANGE;
+		if (!priv->hwts_rx_en) {
+			priv->hwts_rx_en = 1;
+			reset_gfar(netdev);
+		}
+		config->rx_filter = HWTSTAMP_FILTER_ALL;
+		break;
+	}
+
+	return 0;
+}
+
+static int gfar_hwtstamp_get(struct net_device *netdev,
+			     struct kernel_hwtstamp_config *config)
+{
+	struct gfar_private *priv = netdev_priv(netdev);
+
+	config->tx_type = priv->hwts_tx_en ? HWTSTAMP_TX_ON : HWTSTAMP_TX_OFF;
+	config->rx_filter = priv->hwts_rx_en ? HWTSTAMP_FILTER_ALL :
+			    HWTSTAMP_FILTER_NONE;
+
+	return 0;
+}
+
 /* Interrupt Handler for Transmit complete */
 static void gfar_clean_tx_ring(struct gfar_priv_tx_q *tx_queue)
 {
@@ -2678,13 +2129,17 @@ static void gfar_clean_tx_ring(struct gfar_priv_tx_q *tx_queue)
 	skb_dirtytx = tx_queue->skb_dirtytx;
 
 	while ((skb = tx_queue->tx_skbuff[skb_dirtytx])) {
+		bool do_tstamp;
+
+		do_tstamp = (skb_shinfo(skb)->tx_flags & SKBTX_HW_TSTAMP) &&
+			    priv->hwts_tx_en;
 
 		frags = skb_shinfo(skb)->nr_frags;
 
 		/* When time stamping, one additional TxBD must be freed.
 		 * Also, we need to dma_unmap_single() the TxPAL.
 		 */
-		if (unlikely(skb_shinfo(skb)->tx_flags & SKBTX_IN_PROGRESS))
+		if (unlikely(do_tstamp))
 			nr_txbds = frags + 2;
 		else
 			nr_txbds = frags + 1;
@@ -2698,7 +2153,7 @@ static void gfar_clean_tx_ring(struct gfar_priv_tx_q *tx_queue)
 		    (lstatus & BD_LENGTH_MASK))
 			break;
 
-		if (unlikely(skb_shinfo(skb)->tx_flags & SKBTX_IN_PROGRESS)) {
+		if (unlikely(do_tstamp)) {
 			next = next_txbd(bdp, base, tx_ring_size);
 			buflen = be16_to_cpu(next->length) +
 				 GMAC_FCB_LEN + GMAC_TXPAL_LEN;
@@ -2708,10 +2163,11 @@ static void gfar_clean_tx_ring(struct gfar_priv_tx_q *tx_queue)
 		dma_unmap_single(priv->dev, be32_to_cpu(bdp->bufPtr),
 				 buflen, DMA_TO_DEVICE);
 
-		if (unlikely(skb_shinfo(skb)->tx_flags & SKBTX_IN_PROGRESS)) {
+		if (unlikely(do_tstamp)) {
 			struct skb_shared_hwtstamps shhwtstamps;
-			u64 *ns = (u64 *)(((uintptr_t)skb->data + 0x10) &
-					  ~0x7UL);
+			__be64 *ns;
+
+			ns = (__be64 *)(((uintptr_t)skb->data + 0x10) & ~0x7UL);
 
 			memset(&shhwtstamps, 0, sizeof(shhwtstamps));
 			shhwtstamps.hwtstamp = ns_to_ktime(be64_to_cpu(*ns));
@@ -2760,77 +2216,6 @@ static void gfar_clean_tx_ring(struct gfar_priv_tx_q *tx_queue)
 	netdev_tx_completed_queue(txq, howmany, bytes_sent);
 }
 
-static bool gfar_new_page(struct gfar_priv_rx_q *rxq, struct gfar_rx_buff *rxb)
-{
-	struct page *page;
-	dma_addr_t addr;
-
-	page = dev_alloc_page();
-	if (unlikely(!page))
-		return false;
-
-	addr = dma_map_page(rxq->dev, page, 0, PAGE_SIZE, DMA_FROM_DEVICE);
-	if (unlikely(dma_mapping_error(rxq->dev, addr))) {
-		__free_page(page);
-
-		return false;
-	}
-
-	rxb->dma = addr;
-	rxb->page = page;
-	rxb->page_offset = 0;
-
-	return true;
-}
-
-static void gfar_rx_alloc_err(struct gfar_priv_rx_q *rx_queue)
-{
-	struct gfar_private *priv = netdev_priv(rx_queue->ndev);
-	struct gfar_extra_stats *estats = &priv->extra_stats;
-
-	netdev_err(rx_queue->ndev, "Can't alloc RX buffers\n");
-	atomic64_inc(&estats->rx_alloc_err);
-}
-
-static void gfar_alloc_rx_buffs(struct gfar_priv_rx_q *rx_queue,
-				int alloc_cnt)
-{
-	struct rxbd8 *bdp;
-	struct gfar_rx_buff *rxb;
-	int i;
-
-	i = rx_queue->next_to_use;
-	bdp = &rx_queue->rx_bd_base[i];
-	rxb = &rx_queue->rx_buff[i];
-
-	while (alloc_cnt--) {
-		/* try reuse page */
-		if (unlikely(!rxb->page)) {
-			if (unlikely(!gfar_new_page(rx_queue, rxb))) {
-				gfar_rx_alloc_err(rx_queue);
-				break;
-			}
-		}
-
-		/* Setup the new RxBD */
-		gfar_init_rxbdp(rx_queue, bdp,
-				rxb->dma + rxb->page_offset + RXBUF_ALIGNMENT);
-
-		/* Update to the next pointer */
-		bdp++;
-		rxb++;
-
-		if (unlikely(++i == rx_queue->rx_ring_size)) {
-			i = 0;
-			bdp = rx_queue->rx_bd_base;
-			rxb = rx_queue->rx_buff;
-		}
-	}
-
-	rx_queue->next_to_use = i;
-	rx_queue->next_to_alloc = i;
-}
-
 static void count_errors(u32 lstatus, struct net_device *ndev)
 {
 	struct gfar_private *priv = netdev_priv(ndev);
@@ -2868,7 +2253,7 @@ static void count_errors(u32 lstatus, struct net_device *ndev)
 	}
 }
 
-irqreturn_t gfar_receive(int irq, void *grp_id)
+static irqreturn_t gfar_receive(int irq, void *grp_id)
 {
 	struct gfar_priv_grp *grp = (struct gfar_priv_grp *)grp_id;
 	unsigned long flags;
@@ -2884,7 +2269,7 @@ irqreturn_t gfar_receive(int irq, void *grp_id)
 	if (likely(napi_schedule_prep(&grp->napi_rx))) {
 		spin_lock_irqsave(&grp->grplock, flags);
 		imask = gfar_read(&grp->regs->imask);
-		imask &= IMASK_RX_DISABLED;
+		imask &= IMASK_RX_DISABLED | grp->priv->rmon_overflow.imask;
 		gfar_write(&grp->regs->imask, imask);
 		spin_unlock_irqrestore(&grp->grplock, flags);
 		__napi_schedule(&grp->napi_rx);
@@ -2908,7 +2293,7 @@ static irqreturn_t gfar_transmit(int irq, void *grp_id)
 	if (likely(napi_schedule_prep(&grp->napi_tx))) {
 		spin_lock_irqsave(&grp->grplock, flags);
 		imask = gfar_read(&grp->regs->imask);
-		imask &= IMASK_TX_DISABLED;
+		imask &= IMASK_TX_DISABLED | grp->priv->rmon_overflow.imask;
 		gfar_write(&grp->regs->imask, imask);
 		spin_unlock_irqrestore(&grp->grplock, flags);
 		__napi_schedule(&grp->napi_tx);
@@ -2925,26 +2310,23 @@ static irqreturn_t gfar_transmit(int irq, void *grp_id)
 static bool gfar_add_rx_frag(struct gfar_rx_buff *rxb, u32 lstatus,
 			     struct sk_buff *skb, bool first)
 {
-	unsigned int size = lstatus & BD_LENGTH_MASK;
+	int size = lstatus & BD_LENGTH_MASK;
 	struct page *page = rxb->page;
-	bool last = !!(lstatus & BD_LFLAG(RXBD_LAST));
-
-	/* Remove the FCS from the packet length */
-	if (last)
-		size -= ETH_FCS_LEN;
 
 	if (likely(first)) {
 		skb_put(skb, size);
 	} else {
 		/* the last fragments' length contains the full frame length */
-		if (last)
+		if (lstatus & BD_LFLAG(RXBD_LAST))
 			size -= skb->len;
 
-		/* in case the last fragment consisted only of the FCS */
-		if (size > 0)
-			skb_add_rx_frag(skb, skb_shinfo(skb)->nr_frags, page,
-					rxb->page_offset + RXBUF_ALIGNMENT,
-					size, GFAR_RXB_TRUESIZE);
+		WARN(size < 0, "gianfar: rx fragment size underflow");
+		if (size < 0)
+			return false;
+
+		skb_add_rx_frag(skb, skb_shinfo(skb)->nr_frags, page,
+				rxb->page_offset + RXBUF_ALIGNMENT,
+				size, GFAR_RXB_TRUESIZE);
 	}
 
 	/* try reuse page */
@@ -3048,7 +2430,7 @@ static void gfar_process_frame(struct net_device *ndev, struct sk_buff *skb)
 	/* Get receive timestamp from the skb */
 	if (priv->hwts_rx_en) {
 		struct skb_shared_hwtstamps *shhwtstamps = skb_hwtstamps(skb);
-		u64 *ns = (u64 *) skb->data;
+		__be64 *ns = (__be64 *)skb->data;
 
 		memset(shhwtstamps, 0, sizeof(*shhwtstamps));
 		shhwtstamps->hwtstamp = ns_to_ktime(be64_to_cpu(*ns));
@@ -3057,12 +2439,12 @@ static void gfar_process_frame(struct net_device *ndev, struct sk_buff *skb)
 	if (priv->padding)
 		skb_pull(skb, priv->padding);
 
+	/* Trim off the FCS */
+	pskb_trim(skb, skb->len - ETH_FCS_LEN);
+
 	if (ndev->features & NETIF_F_RXCSUM)
 		gfar_rx_checksum(skb, fcb);
 
-	/* Tell the skb what kind of packet this is */
-	skb->protocol = eth_type_trans(skb, ndev);
-
 	/* There's need to check for NETIF_F_HW_VLAN_CTAG_RX here.
 	 * Even if vlan rx accel is disabled, on some chips
 	 * RXFCB_VLN is pseudo randomly set.
@@ -3077,7 +2459,8 @@ static void gfar_process_frame(struct net_device *ndev, struct sk_buff *skb)
  * until the budget/quota has been reached. Returns the number
  * of frames handled
  */
-int gfar_clean_rx_ring(struct gfar_priv_rx_q *rx_queue, int rx_work_limit)
+static int gfar_clean_rx_ring(struct gfar_priv_rx_q *rx_queue,
+			      int rx_work_limit)
 {
 	struct net_device *ndev = rx_queue->ndev;
 	struct gfar_private *priv = netdev_priv(ndev);
@@ -3103,6 +2486,17 @@ int gfar_clean_rx_ring(struct gfar_priv_rx_q *rx_queue, int rx_work_limit)
 		if (lstatus & BD_LFLAG(RXBD_EMPTY))
 			break;
 
+		/* lost RXBD_LAST descriptor due to overrun */
+		if (skb &&
+		    (lstatus & BD_LFLAG(RXBD_FIRST))) {
+			/* discard faulty buffer */
+			dev_kfree_skb(skb);
+			skb = NULL;
+			rx_queue->stats.rx_dropped++;
+
+			/* can continue normally */
+		}
+
 		/* order rx buffer descriptor reads */
 		rmb();
 
@@ -3133,13 +2527,15 @@ int gfar_clean_rx_ring(struct gfar_priv_rx_q *rx_queue, int rx_work_limit)
 			continue;
 		}
 
+		gfar_process_frame(ndev, skb);
+
 		/* Increment the number of packets */
 		total_pkts++;
 		total_bytes += skb->len;
 
 		skb_record_rx_queue(skb, rx_queue->qindex);
 
-		gfar_process_frame(ndev, skb);
+		skb->protocol = eth_type_trans(skb, ndev);
 
 		/* Send the packet up the stack */
 		napi_gro_receive(&rx_queue->grp->napi_rx, skb);
@@ -3225,107 +2621,111 @@ static int gfar_poll_tx_sq(struct napi_struct *napi, int budget)
 	return 0;
 }
 
-static int gfar_poll_rx(struct napi_struct *napi, int budget)
+/* GFAR error interrupt handler */
+static irqreturn_t gfar_error(int irq, void *grp_id)
 {
-	struct gfar_priv_grp *gfargrp =
-		container_of(napi, struct gfar_priv_grp, napi_rx);
-	struct gfar_private *priv = gfargrp->priv;
+	struct gfar_priv_grp *gfargrp = grp_id;
 	struct gfar __iomem *regs = gfargrp->regs;
-	struct gfar_priv_rx_q *rx_queue = NULL;
-	int work_done = 0, work_done_per_q = 0;
-	int i, budget_per_q = 0;
-	unsigned long rstat_rxf;
-	int num_act_queues;
+	struct gfar_private *priv= gfargrp->priv;
+	struct net_device *dev = priv->ndev;
 
-	/* Clear IEVENT, so interrupts aren't called again
-	 * because of the packets that have already arrived
-	 */
-	gfar_write(&regs->ievent, IEVENT_RX_MASK);
+	/* Save ievent for future reference */
+	u32 events = gfar_read(&regs->ievent);
 
-	rstat_rxf = gfar_read(&regs->rstat) & RSTAT_RXF_MASK;
+	/* Clear IEVENT */
+	gfar_write(&regs->ievent, events & IEVENT_ERR_MASK);
 
-	num_act_queues = bitmap_weight(&rstat_rxf, MAX_RX_QS);
-	if (num_act_queues)
-		budget_per_q = budget/num_act_queues;
+	/* Magic Packet is not an error. */
+	if ((priv->device_flags & FSL_GIANFAR_DEV_HAS_MAGIC_PACKET) &&
+	    (events & IEVENT_MAG))
+		events &= ~IEVENT_MAG;
 
-	for_each_set_bit(i, &gfargrp->rx_bit_map, priv->num_rx_queues) {
-		/* skip queue if not active */
-		if (!(rstat_rxf & (RSTAT_CLEAR_RXF0 >> i)))
-			continue;
+	/* Hmm... */
+	if (netif_msg_rx_err(priv) || netif_msg_tx_err(priv))
+		netdev_dbg(dev,
+			   "error interrupt (ievent=0x%08x imask=0x%08x)\n",
+			   events, gfar_read(&regs->imask));
 
-		rx_queue = priv->rx_queue[i];
-		work_done_per_q =
-			gfar_clean_rx_ring(rx_queue, budget_per_q);
-		work_done += work_done_per_q;
-
-		/* finished processing this queue */
-		if (work_done_per_q < budget_per_q) {
-			/* clear active queue hw indication */
-			gfar_write(&regs->rstat,
-				   RSTAT_CLEAR_RXF0 >> i);
-			num_act_queues--;
-
-			if (!num_act_queues)
-				break;
+	/* Update the error counters */
+	if (events & IEVENT_TXE) {
+		dev->stats.tx_errors++;
+
+		if (events & IEVENT_LC)
+			dev->stats.tx_window_errors++;
+		if (events & IEVENT_CRL)
+			dev->stats.tx_aborted_errors++;
+		if (events & IEVENT_XFUN) {
+			netif_dbg(priv, tx_err, dev,
+				  "TX FIFO underrun, packet dropped\n");
+			dev->stats.tx_dropped++;
+			atomic64_inc(&priv->extra_stats.tx_underrun);
+
+			schedule_work(&priv->reset_task);
 		}
+		netif_dbg(priv, tx_err, dev, "Transmit Error\n");
 	}
+	if (events & IEVENT_MSRO) {
+		struct rmon_mib __iomem *rmon = &regs->rmon;
+		u32 car;
 
-	if (!num_act_queues) {
-		u32 imask;
-		napi_complete_done(napi, work_done);
+		spin_lock(&priv->rmon_overflow.lock);
+		car = gfar_read(&rmon->car1) & CAR1_C1RDR;
+		if (car) {
+			priv->rmon_overflow.rdrp++;
+			gfar_write(&rmon->car1, car);
+		}
+		spin_unlock(&priv->rmon_overflow.lock);
+	}
+	if (events & IEVENT_BSY) {
+		dev->stats.rx_over_errors++;
+		atomic64_inc(&priv->extra_stats.rx_bsy);
 
-		/* Clear the halt bit in RSTAT */
-		gfar_write(&regs->rstat, gfargrp->rstat);
+		netif_dbg(priv, rx_err, dev, "busy error (rstat: %x)\n",
+			  gfar_read(&regs->rstat));
+	}
+	if (events & IEVENT_BABR) {
+		dev->stats.rx_errors++;
+		atomic64_inc(&priv->extra_stats.rx_babr);
 
-		spin_lock_irq(&gfargrp->grplock);
-		imask = gfar_read(&regs->imask);
-		imask |= IMASK_RX_DEFAULT;
-		gfar_write(&regs->imask, imask);
-		spin_unlock_irq(&gfargrp->grplock);
+		netif_dbg(priv, rx_err, dev, "babbling RX error\n");
+	}
+	if (events & IEVENT_EBERR) {
+		atomic64_inc(&priv->extra_stats.eberr);
+		netif_dbg(priv, rx_err, dev, "bus error\n");
 	}
+	if (events & IEVENT_RXC)
+		netif_dbg(priv, rx_status, dev, "control frame\n");
 
-	return work_done;
+	if (events & IEVENT_BABT) {
+		atomic64_inc(&priv->extra_stats.tx_babt);
+		netif_dbg(priv, tx_err, dev, "babbling TX error\n");
+	}
+	return IRQ_HANDLED;
 }
 
-static int gfar_poll_tx(struct napi_struct *napi, int budget)
+/* The interrupt handler for devices with one interrupt */
+static irqreturn_t gfar_interrupt(int irq, void *grp_id)
 {
-	struct gfar_priv_grp *gfargrp =
-		container_of(napi, struct gfar_priv_grp, napi_tx);
-	struct gfar_private *priv = gfargrp->priv;
-	struct gfar __iomem *regs = gfargrp->regs;
-	struct gfar_priv_tx_q *tx_queue = NULL;
-	int has_tx_work = 0;
-	int i;
+	struct gfar_priv_grp *gfargrp = grp_id;
 
-	/* Clear IEVENT, so interrupts aren't called again
-	 * because of the packets that have already arrived
-	 */
-	gfar_write(&regs->ievent, IEVENT_TX_MASK);
+	/* Save ievent for future reference */
+	u32 events = gfar_read(&gfargrp->regs->ievent);
 
-	for_each_set_bit(i, &gfargrp->tx_bit_map, priv->num_tx_queues) {
-		tx_queue = priv->tx_queue[i];
-		/* run Tx cleanup to completion */
-		if (tx_queue->tx_skbuff[tx_queue->skb_dirtytx]) {
-			gfar_clean_tx_ring(tx_queue);
-			has_tx_work = 1;
-		}
-	}
+	/* Check for reception */
+	if (events & IEVENT_RX_MASK)
+		gfar_receive(irq, grp_id);
 
-	if (!has_tx_work) {
-		u32 imask;
-		napi_complete(napi);
+	/* Check for transmit completion */
+	if (events & IEVENT_TX_MASK)
+		gfar_transmit(irq, grp_id);
 
-		spin_lock_irq(&gfargrp->grplock);
-		imask = gfar_read(&regs->imask);
-		imask |= IMASK_TX_DEFAULT;
-		gfar_write(&regs->imask, imask);
-		spin_unlock_irq(&gfargrp->grplock);
-	}
+	/* Check for errors */
+	if (events & IEVENT_ERR_MASK)
+		gfar_error(irq, grp_id);
 
-	return 0;
+	return IRQ_HANDLED;
 }
 
-
 #ifdef CONFIG_NET_POLL_CONTROLLER
 /* Polling 'interrupt' - used by things like netconsole to send skbs
  * without having to re-enable interrupts. It's not called while
@@ -3361,44 +2761,154 @@ static void gfar_netpoll(struct net_device *dev)
 }
 #endif
 
-/* The interrupt handler for devices with one interrupt */
-static irqreturn_t gfar_interrupt(int irq, void *grp_id)
+static void free_grp_irqs(struct gfar_priv_grp *grp)
 {
-	struct gfar_priv_grp *gfargrp = grp_id;
+	free_irq(gfar_irq(grp, TX)->irq, grp);
+	free_irq(gfar_irq(grp, RX)->irq, grp);
+	free_irq(gfar_irq(grp, ER)->irq, grp);
+}
 
-	/* Save ievent for future reference */
-	u32 events = gfar_read(&gfargrp->regs->ievent);
+static int register_grp_irqs(struct gfar_priv_grp *grp)
+{
+	struct gfar_private *priv = grp->priv;
+	struct net_device *dev = priv->ndev;
+	int err;
 
-	/* Check for reception */
-	if (events & IEVENT_RX_MASK)
-		gfar_receive(irq, grp_id);
+	/* If the device has multiple interrupts, register for
+	 * them.  Otherwise, only register for the one
+	 */
+	if (priv->device_flags & FSL_GIANFAR_DEV_HAS_MULTI_INTR) {
+		/* Install our interrupt handlers for Error,
+		 * Transmit, and Receive
+		 */
+		err = request_irq(gfar_irq(grp, ER)->irq, gfar_error, 0,
+				  gfar_irq(grp, ER)->name, grp);
+		if (err < 0) {
+			netif_err(priv, intr, dev, "Can't get IRQ %d\n",
+				  gfar_irq(grp, ER)->irq);
 
-	/* Check for transmit completion */
-	if (events & IEVENT_TX_MASK)
-		gfar_transmit(irq, grp_id);
+			goto err_irq_fail;
+		}
+		enable_irq_wake(gfar_irq(grp, ER)->irq);
 
-	/* Check for errors */
-	if (events & IEVENT_ERR_MASK)
-		gfar_error(irq, grp_id);
+		err = request_irq(gfar_irq(grp, TX)->irq, gfar_transmit, 0,
+				  gfar_irq(grp, TX)->name, grp);
+		if (err < 0) {
+			netif_err(priv, intr, dev, "Can't get IRQ %d\n",
+				  gfar_irq(grp, TX)->irq);
+			goto tx_irq_fail;
+		}
+		err = request_irq(gfar_irq(grp, RX)->irq, gfar_receive, 0,
+				  gfar_irq(grp, RX)->name, grp);
+		if (err < 0) {
+			netif_err(priv, intr, dev, "Can't get IRQ %d\n",
+				  gfar_irq(grp, RX)->irq);
+			goto rx_irq_fail;
+		}
+		enable_irq_wake(gfar_irq(grp, RX)->irq);
+
+	} else {
+		err = request_irq(gfar_irq(grp, TX)->irq, gfar_interrupt, 0,
+				  gfar_irq(grp, TX)->name, grp);
+		if (err < 0) {
+			netif_err(priv, intr, dev, "Can't get IRQ %d\n",
+				  gfar_irq(grp, TX)->irq);
+			goto err_irq_fail;
+		}
+		enable_irq_wake(gfar_irq(grp, TX)->irq);
+	}
+
+	return 0;
+
+rx_irq_fail:
+	free_irq(gfar_irq(grp, TX)->irq, grp);
+tx_irq_fail:
+	free_irq(gfar_irq(grp, ER)->irq, grp);
+err_irq_fail:
+	return err;
 
-	return IRQ_HANDLED;
 }
 
-/* Called every time the controller might need to be made
- * aware of new link state.  The PHY code conveys this
- * information through variables in the phydev structure, and this
- * function converts those variables into the appropriate
- * register values, and can bring down the device if needed.
+static void gfar_free_irq(struct gfar_private *priv)
+{
+	int i;
+
+	/* Free the IRQs */
+	if (priv->device_flags & FSL_GIANFAR_DEV_HAS_MULTI_INTR) {
+		for (i = 0; i < priv->num_grps; i++)
+			free_grp_irqs(&priv->gfargrp[i]);
+	} else {
+		for (i = 0; i < priv->num_grps; i++)
+			free_irq(gfar_irq(&priv->gfargrp[i], TX)->irq,
+				 &priv->gfargrp[i]);
+	}
+}
+
+static int gfar_request_irq(struct gfar_private *priv)
+{
+	int err, i, j;
+
+	for (i = 0; i < priv->num_grps; i++) {
+		err = register_grp_irqs(&priv->gfargrp[i]);
+		if (err) {
+			for (j = 0; j < i; j++)
+				free_grp_irqs(&priv->gfargrp[j]);
+			return err;
+		}
+	}
+
+	return 0;
+}
+
+/* Called when something needs to use the ethernet device
+ * Returns 0 for success.
  */
-static void adjust_link(struct net_device *dev)
+static int gfar_enet_open(struct net_device *dev)
 {
 	struct gfar_private *priv = netdev_priv(dev);
-	struct phy_device *phydev = dev->phydev;
+	int err;
 
-	if (unlikely(phydev->link != priv->oldlink ||
-		     (phydev->link && (phydev->duplex != priv->oldduplex ||
-				       phydev->speed != priv->oldspeed))))
-		gfar_update_link_state(priv);
+	err = init_phy(dev);
+	if (err)
+		return err;
+
+	err = gfar_request_irq(priv);
+	if (err)
+		return err;
+
+	err = startup_gfar(dev);
+	if (err)
+		return err;
+
+	return err;
+}
+
+/* Stops the kernel queue, and halts the controller */
+static int gfar_close(struct net_device *dev)
+{
+	struct gfar_private *priv = netdev_priv(dev);
+
+	cancel_work_sync(&priv->reset_task);
+	stop_gfar(dev);
+
+	/* Disconnect from the PHY */
+	phy_disconnect(dev->phydev);
+
+	gfar_free_irq(priv);
+
+	return 0;
+}
+
+/* Clears each of the exact match registers to zero, so they
+ * don't interfere with normal reception
+ */
+static void gfar_clear_exact_match(struct net_device *dev)
+{
+	int idx;
+	static const u8 zero_arr[ETH_ALEN] = {0, 0, 0, 0, 0, 0};
+
+	for (idx = 1; idx < GFAR_EM_NUM + 1; idx++)
+		gfar_set_mac_for_addr(dev, idx, zero_arr);
 }
 
 /* Update the hash table based on the current list of multicast
@@ -3492,279 +3002,586 @@ static void gfar_set_multi(struct net_device *dev)
 	}
 }
 
-
-/* Clears each of the exact match registers to zero, so they
- * don't interfere with normal reception
- */
-static void gfar_clear_exact_match(struct net_device *dev)
+void gfar_mac_reset(struct gfar_private *priv)
 {
-	int idx;
-	static const u8 zero_arr[ETH_ALEN] = {0, 0, 0, 0, 0, 0};
+	struct gfar __iomem *regs = priv->gfargrp[0].regs;
+	u32 tempval;
 
-	for (idx = 1; idx < GFAR_EM_NUM + 1; idx++)
-		gfar_set_mac_for_addr(dev, idx, zero_arr);
-}
+	/* Reset MAC layer */
+	gfar_write(&regs->maccfg1, MACCFG1_SOFT_RESET);
 
-/* Set the appropriate hash bit for the given addr */
-/* The algorithm works like so:
- * 1) Take the Destination Address (ie the multicast address), and
- * do a CRC on it (little endian), and reverse the bits of the
- * result.
- * 2) Use the 8 most significant bits as a hash into a 256-entry
- * table.  The table is controlled through 8 32-bit registers:
- * gaddr0-7.  gaddr0's MSB is entry 0, and gaddr7's LSB is
- * gaddr7.  This means that the 3 most significant bits in the
- * hash index which gaddr register to use, and the 5 other bits
- * indicate which bit (assuming an IBM numbering scheme, which
- * for PowerPC (tm) is usually the case) in the register holds
- * the entry.
- */
-static void gfar_set_hash_for_addr(struct net_device *dev, u8 *addr)
-{
-	u32 tempval;
-	struct gfar_private *priv = netdev_priv(dev);
-	u32 result = ether_crc(ETH_ALEN, addr);
-	int width = priv->hash_width;
-	u8 whichbit = (result >> (32 - width)) & 0x1f;
-	u8 whichreg = result >> (32 - width + 5);
-	u32 value = (1 << (31-whichbit));
+	/* We need to delay at least 3 TX clocks */
+	udelay(3);
 
-	tempval = gfar_read(priv->hash_regs[whichreg]);
-	tempval |= value;
-	gfar_write(priv->hash_regs[whichreg], tempval);
-}
+	/* the soft reset bit is not self-resetting, so we need to
+	 * clear it before resuming normal operation
+	 */
+	gfar_write(&regs->maccfg1, 0);
 
+	udelay(3);
 
-/* There are multiple MAC Address register pairs on some controllers
- * This function sets the numth pair to a given address
- */
-static void gfar_set_mac_for_addr(struct net_device *dev, int num,
-				  const u8 *addr)
+	gfar_rx_offload_en(priv);
+
+	/* Initialize the max receive frame/buffer lengths */
+	gfar_write(&regs->maxfrm, GFAR_JUMBO_FRAME_SIZE);
+	gfar_write(&regs->mrblr, GFAR_RXB_SIZE);
+
+	/* Initialize the Minimum Frame Length Register */
+	gfar_write(&regs->minflr, MINFLR_INIT_SETTINGS);
+
+	/* Initialize MACCFG2. */
+	tempval = MACCFG2_INIT_SETTINGS;
+
+	/* eTSEC74 erratum: Rx frames of length MAXFRM or MAXFRM-1
+	 * are marked as truncated.  Avoid this by MACCFG2[Huge Frame]=1,
+	 * and by checking RxBD[LG] and discarding larger than MAXFRM.
+	 */
+	if (gfar_has_errata(priv, GFAR_ERRATA_74))
+		tempval |= MACCFG2_HUGEFRAME | MACCFG2_LENGTHCHECK;
+
+	gfar_write(&regs->maccfg2, tempval);
+
+	/* Clear mac addr hash registers */
+	gfar_write(&regs->igaddr0, 0);
+	gfar_write(&regs->igaddr1, 0);
+	gfar_write(&regs->igaddr2, 0);
+	gfar_write(&regs->igaddr3, 0);
+	gfar_write(&regs->igaddr4, 0);
+	gfar_write(&regs->igaddr5, 0);
+	gfar_write(&regs->igaddr6, 0);
+	gfar_write(&regs->igaddr7, 0);
+
+	gfar_write(&regs->gaddr0, 0);
+	gfar_write(&regs->gaddr1, 0);
+	gfar_write(&regs->gaddr2, 0);
+	gfar_write(&regs->gaddr3, 0);
+	gfar_write(&regs->gaddr4, 0);
+	gfar_write(&regs->gaddr5, 0);
+	gfar_write(&regs->gaddr6, 0);
+	gfar_write(&regs->gaddr7, 0);
+
+	if (priv->extended_hash)
+		gfar_clear_exact_match(priv->ndev);
+
+	gfar_mac_rx_config(priv);
+
+	gfar_mac_tx_config(priv);
+
+	gfar_set_mac_address(priv->ndev);
+
+	gfar_set_multi(priv->ndev);
+
+	/* clear ievent and imask before configuring coalescing */
+	gfar_ints_disable(priv);
+
+	/* Configure the coalescing support */
+	gfar_configure_coalescing_all(priv);
+}
+
+static void gfar_hw_init(struct gfar_private *priv)
 {
-	struct gfar_private *priv = netdev_priv(dev);
 	struct gfar __iomem *regs = priv->gfargrp[0].regs;
-	u32 tempval;
-	u32 __iomem *macptr = &regs->macstnaddr1;
+	u32 attrs;
 
-	macptr += num*2;
+	/* Stop the DMA engine now, in case it was running before
+	 * (The firmware could have used it, and left it running).
+	 */
+	gfar_halt(priv);
 
-	/* For a station address of 0x12345678ABCD in transmission
-	 * order (BE), MACnADDR1 is set to 0xCDAB7856 and
-	 * MACnADDR2 is set to 0x34120000.
+	gfar_mac_reset(priv);
+
+	/* Zero out the rmon mib registers if it has them */
+	if (priv->device_flags & FSL_GIANFAR_DEV_HAS_RMON) {
+		memset_io(&regs->rmon, 0, offsetof(struct rmon_mib, car1));
+
+		/* Mask off the CAM interrupts */
+		gfar_write(&regs->rmon.cam1, 0xffffffff);
+		gfar_write(&regs->rmon.cam2, 0xffffffff);
+		/* Clear the CAR registers (w1c style) */
+		gfar_write(&regs->rmon.car1, 0xffffffff);
+		gfar_write(&regs->rmon.car2, 0xffffffff);
+	}
+
+	/* Initialize ECNTRL */
+	gfar_write(&regs->ecntrl, ECNTRL_INIT_SETTINGS);
+
+	/* Set the extraction length and index */
+	attrs = ATTRELI_EL(priv->rx_stash_size) |
+		ATTRELI_EI(priv->rx_stash_index);
+
+	gfar_write(&regs->attreli, attrs);
+
+	/* Start with defaults, and add stashing
+	 * depending on driver parameters
 	 */
-	tempval = (addr[5] << 24) | (addr[4] << 16) |
-		  (addr[3] << 8)  |  addr[2];
+	attrs = ATTR_INIT_SETTINGS;
 
-	gfar_write(macptr, tempval);
+	if (priv->bd_stash_en)
+		attrs |= ATTR_BDSTASH;
 
-	tempval = (addr[1] << 24) | (addr[0] << 16);
+	if (priv->rx_stash_size != 0)
+		attrs |= ATTR_BUFSTASH;
 
-	gfar_write(macptr+1, tempval);
+	gfar_write(&regs->attr, attrs);
+
+	/* FIFO configs */
+	gfar_write(&regs->fifo_tx_thr, DEFAULT_FIFO_TX_THR);
+	gfar_write(&regs->fifo_tx_starve, DEFAULT_FIFO_TX_STARVE);
+	gfar_write(&regs->fifo_tx_starve_shutoff, DEFAULT_FIFO_TX_STARVE_OFF);
+
+	/* Program the interrupt steering regs, only for MG devices */
+	if (priv->num_grps > 1)
+		gfar_write_isrg(priv);
 }
 
-/* GFAR error interrupt handler */
-static irqreturn_t gfar_error(int irq, void *grp_id)
+static const struct net_device_ops gfar_netdev_ops = {
+	.ndo_open = gfar_enet_open,
+	.ndo_start_xmit = gfar_start_xmit,
+	.ndo_stop = gfar_close,
+	.ndo_change_mtu = gfar_change_mtu,
+	.ndo_set_features = gfar_set_features,
+	.ndo_set_rx_mode = gfar_set_multi,
+	.ndo_tx_timeout = gfar_timeout,
+	.ndo_eth_ioctl = phy_do_ioctl_running,
+	.ndo_get_stats64 = gfar_get_stats64,
+	.ndo_change_carrier = fixed_phy_change_carrier,
+	.ndo_set_mac_address = gfar_set_mac_addr,
+	.ndo_validate_addr = eth_validate_addr,
+#ifdef CONFIG_NET_POLL_CONTROLLER
+	.ndo_poll_controller = gfar_netpoll,
+#endif
+	.ndo_hwtstamp_get = gfar_hwtstamp_get,
+	.ndo_hwtstamp_set = gfar_hwtstamp_set,
+};
+
+/* Set up the ethernet device structure, private data,
+ * and anything else we need before we start
+ */
+static int gfar_probe(struct platform_device *ofdev)
 {
-	struct gfar_priv_grp *gfargrp = grp_id;
-	struct gfar __iomem *regs = gfargrp->regs;
-	struct gfar_private *priv= gfargrp->priv;
-	struct net_device *dev = priv->ndev;
+	struct device_node *np = ofdev->dev.of_node;
+	struct net_device *dev = NULL;
+	struct gfar_private *priv = NULL;
+	int err = 0, i;
 
-	/* Save ievent for future reference */
-	u32 events = gfar_read(&regs->ievent);
+	err = gfar_of_init(ofdev, &dev);
 
-	/* Clear IEVENT */
-	gfar_write(&regs->ievent, events & IEVENT_ERR_MASK);
+	if (err)
+		return err;
 
-	/* Magic Packet is not an error. */
-	if ((priv->device_flags & FSL_GIANFAR_DEV_HAS_MAGIC_PACKET) &&
-	    (events & IEVENT_MAG))
-		events &= ~IEVENT_MAG;
+	priv = netdev_priv(dev);
+	priv->ndev = dev;
+	priv->ofdev = ofdev;
+	priv->dev = &ofdev->dev;
+	SET_NETDEV_DEV(dev, &ofdev->dev);
 
-	/* Hmm... */
-	if (netif_msg_rx_err(priv) || netif_msg_tx_err(priv))
-		netdev_dbg(dev,
-			   "error interrupt (ievent=0x%08x imask=0x%08x)\n",
-			   events, gfar_read(&regs->imask));
+	INIT_WORK(&priv->reset_task, gfar_reset_task);
 
-	/* Update the error counters */
-	if (events & IEVENT_TXE) {
-		dev->stats.tx_errors++;
+	platform_set_drvdata(ofdev, priv);
 
-		if (events & IEVENT_LC)
-			dev->stats.tx_window_errors++;
-		if (events & IEVENT_CRL)
-			dev->stats.tx_aborted_errors++;
-		if (events & IEVENT_XFUN) {
-			netif_dbg(priv, tx_err, dev,
-				  "TX FIFO underrun, packet dropped\n");
-			dev->stats.tx_dropped++;
-			atomic64_inc(&priv->extra_stats.tx_underrun);
+	gfar_detect_errata(priv);
 
-			schedule_work(&priv->reset_task);
-		}
-		netif_dbg(priv, tx_err, dev, "Transmit Error\n");
+	/* Set the dev->base_addr to the gfar reg region */
+	dev->base_addr = (unsigned long) priv->gfargrp[0].regs;
+
+	/* Fill in the dev structure */
+	dev->watchdog_timeo = TX_TIMEOUT;
+	/* MTU range: 50 - 9586 */
+	dev->mtu = 1500;
+	dev->min_mtu = 50;
+	dev->max_mtu = GFAR_JUMBO_FRAME_SIZE - ETH_HLEN;
+	dev->netdev_ops = &gfar_netdev_ops;
+	dev->ethtool_ops = &gfar_ethtool_ops;
+
+	/* Register for napi ...We are registering NAPI for each grp */
+	for (i = 0; i < priv->num_grps; i++) {
+		netif_napi_add(dev, &priv->gfargrp[i].napi_rx,
+			       gfar_poll_rx_sq);
+		netif_napi_add_tx_weight(dev, &priv->gfargrp[i].napi_tx,
+					 gfar_poll_tx_sq, 2);
 	}
-	if (events & IEVENT_BSY) {
-		dev->stats.rx_over_errors++;
-		atomic64_inc(&priv->extra_stats.rx_bsy);
 
-		netif_dbg(priv, rx_err, dev, "busy error (rstat: %x)\n",
-			  gfar_read(&regs->rstat));
+	if (priv->device_flags & FSL_GIANFAR_DEV_HAS_CSUM) {
+		dev->hw_features = NETIF_F_IP_CSUM | NETIF_F_SG |
+				   NETIF_F_RXCSUM;
+		dev->features |= NETIF_F_IP_CSUM | NETIF_F_SG |
+				 NETIF_F_RXCSUM | NETIF_F_HIGHDMA;
 	}
-	if (events & IEVENT_BABR) {
-		dev->stats.rx_errors++;
-		atomic64_inc(&priv->extra_stats.rx_babr);
 
-		netif_dbg(priv, rx_err, dev, "babbling RX error\n");
+	if (priv->device_flags & FSL_GIANFAR_DEV_HAS_VLAN) {
+		dev->hw_features |= NETIF_F_HW_VLAN_CTAG_TX |
+				    NETIF_F_HW_VLAN_CTAG_RX;
+		dev->features |= NETIF_F_HW_VLAN_CTAG_RX;
 	}
-	if (events & IEVENT_EBERR) {
-		atomic64_inc(&priv->extra_stats.eberr);
-		netif_dbg(priv, rx_err, dev, "bus error\n");
+
+	dev->priv_flags |= IFF_LIVE_ADDR_CHANGE;
+
+	gfar_init_addr_hash_table(priv);
+
+	/* Insert receive time stamps into padding alignment bytes, and
+	 * plus 2 bytes padding to ensure the cpu alignment.
+	 */
+	if (priv->device_flags & FSL_GIANFAR_DEV_HAS_TIMER)
+		priv->padding = 8 + DEFAULT_PADDING;
+
+	if (dev->features & NETIF_F_IP_CSUM ||
+	    priv->device_flags & FSL_GIANFAR_DEV_HAS_TIMER)
+		dev->needed_headroom = GMAC_FCB_LEN + GMAC_TXPAL_LEN;
+
+	/* Initializing some of the rx/tx queue level parameters */
+	for (i = 0; i < priv->num_tx_queues; i++) {
+		priv->tx_queue[i]->tx_ring_size = DEFAULT_TX_RING_SIZE;
+		priv->tx_queue[i]->num_txbdfree = DEFAULT_TX_RING_SIZE;
+		priv->tx_queue[i]->txcoalescing = DEFAULT_TX_COALESCE;
+		priv->tx_queue[i]->txic = DEFAULT_TXIC;
 	}
-	if (events & IEVENT_RXC)
-		netif_dbg(priv, rx_status, dev, "control frame\n");
 
-	if (events & IEVENT_BABT) {
-		atomic64_inc(&priv->extra_stats.tx_babt);
-		netif_dbg(priv, tx_err, dev, "babbling TX error\n");
+	for (i = 0; i < priv->num_rx_queues; i++) {
+		priv->rx_queue[i]->rx_ring_size = DEFAULT_RX_RING_SIZE;
+		priv->rx_queue[i]->rxcoalescing = DEFAULT_RX_COALESCE;
+		priv->rx_queue[i]->rxic = DEFAULT_RXIC;
 	}
-	return IRQ_HANDLED;
+
+	/* Always enable rx filer if available */
+	priv->rx_filer_enable =
+	    (priv->device_flags & FSL_GIANFAR_DEV_HAS_RX_FILER) ? 1 : 0;
+	/* Enable most messages by default */
+	priv->msg_enable = (NETIF_MSG_IFUP << 1 ) - 1;
+	/* use pritority h/w tx queue scheduling for single queue devices */
+	if (priv->num_tx_queues == 1)
+		priv->prio_sched_en = 1;
+
+	set_bit(GFAR_DOWN, &priv->state);
+
+	gfar_hw_init(priv);
+
+	if (priv->device_flags & FSL_GIANFAR_DEV_HAS_RMON) {
+		struct rmon_mib __iomem *rmon = &priv->gfargrp[0].regs->rmon;
+
+		spin_lock_init(&priv->rmon_overflow.lock);
+		priv->rmon_overflow.imask = IMASK_MSRO;
+		gfar_write(&rmon->cam1, gfar_read(&rmon->cam1) & ~CAM1_M1RDR);
+	}
+
+	/* Carrier starts down, phylib will bring it up */
+	netif_carrier_off(dev);
+
+	err = register_netdev(dev);
+
+	if (err) {
+		pr_err("%s: Cannot register net device, aborting\n", dev->name);
+		goto register_fail;
+	}
+
+	if (priv->device_flags & FSL_GIANFAR_DEV_HAS_MAGIC_PACKET)
+		priv->wol_supported |= GFAR_WOL_MAGIC;
+
+	if ((priv->device_flags & FSL_GIANFAR_DEV_HAS_WAKE_ON_FILER) &&
+	    priv->rx_filer_enable)
+		priv->wol_supported |= GFAR_WOL_FILER_UCAST;
+
+	device_set_wakeup_capable(&ofdev->dev, priv->wol_supported);
+
+	/* fill out IRQ number and name fields */
+	for (i = 0; i < priv->num_grps; i++) {
+		struct gfar_priv_grp *grp = &priv->gfargrp[i];
+		if (priv->device_flags & FSL_GIANFAR_DEV_HAS_MULTI_INTR) {
+			sprintf(gfar_irq(grp, TX)->name, "%s%s%c%s",
+				dev->name, "_g", '0' + i, "_tx");
+			sprintf(gfar_irq(grp, RX)->name, "%s%s%c%s",
+				dev->name, "_g", '0' + i, "_rx");
+			sprintf(gfar_irq(grp, ER)->name, "%s%s%c%s",
+				dev->name, "_g", '0' + i, "_er");
+		} else
+			strcpy(gfar_irq(grp, TX)->name, dev->name);
+	}
+
+	/* Initialize the filer table */
+	gfar_init_filer_table(priv);
+
+	/* Print out the device info */
+	netdev_info(dev, "mac: %pM\n", dev->dev_addr);
+
+	/* Even more device info helps when determining which kernel
+	 * provided which set of benchmarks.
+	 */
+	netdev_info(dev, "Running with NAPI enabled\n");
+	for (i = 0; i < priv->num_rx_queues; i++)
+		netdev_info(dev, "RX BD ring size for Q[%d]: %d\n",
+			    i, priv->rx_queue[i]->rx_ring_size);
+	for (i = 0; i < priv->num_tx_queues; i++)
+		netdev_info(dev, "TX BD ring size for Q[%d]: %d\n",
+			    i, priv->tx_queue[i]->tx_ring_size);
+
+	return 0;
+
+register_fail:
+	if (of_phy_is_fixed_link(np))
+		of_phy_deregister_fixed_link(np);
+	unmap_group_regs(priv);
+	gfar_free_rx_queues(priv);
+	gfar_free_tx_queues(priv);
+	of_node_put(priv->phy_node);
+	of_node_put(priv->tbi_node);
+	free_gfar_dev(priv);
+	return err;
 }
 
-static u32 gfar_get_flowctrl_cfg(struct gfar_private *priv)
+static void gfar_remove(struct platform_device *ofdev)
 {
-	struct net_device *ndev = priv->ndev;
-	struct phy_device *phydev = ndev->phydev;
-	u32 val = 0;
+	struct gfar_private *priv = platform_get_drvdata(ofdev);
+	struct device_node *np = ofdev->dev.of_node;
 
-	if (!phydev->duplex)
-		return val;
+	of_node_put(priv->phy_node);
+	of_node_put(priv->tbi_node);
 
-	if (!priv->pause_aneg_en) {
-		if (priv->tx_pause_en)
-			val |= MACCFG1_TX_FLOW;
-		if (priv->rx_pause_en)
-			val |= MACCFG1_RX_FLOW;
-	} else {
-		u16 lcl_adv, rmt_adv;
-		u8 flowctrl;
-		/* get link partner capabilities */
-		rmt_adv = 0;
-		if (phydev->pause)
-			rmt_adv = LPA_PAUSE_CAP;
-		if (phydev->asym_pause)
-			rmt_adv |= LPA_PAUSE_ASYM;
+	unregister_netdev(priv->ndev);
 
-		lcl_adv = 0;
-		if (phydev->advertising & ADVERTISED_Pause)
-			lcl_adv |= ADVERTISE_PAUSE_CAP;
-		if (phydev->advertising & ADVERTISED_Asym_Pause)
-			lcl_adv |= ADVERTISE_PAUSE_ASYM;
+	if (of_phy_is_fixed_link(np))
+		of_phy_deregister_fixed_link(np);
 
-		flowctrl = mii_resolve_flowctrl_fdx(lcl_adv, rmt_adv);
-		if (flowctrl & FLOW_CTRL_TX)
-			val |= MACCFG1_TX_FLOW;
-		if (flowctrl & FLOW_CTRL_RX)
-			val |= MACCFG1_RX_FLOW;
+	unmap_group_regs(priv);
+	gfar_free_rx_queues(priv);
+	gfar_free_tx_queues(priv);
+	free_gfar_dev(priv);
+}
+
+#ifdef CONFIG_PM
+
+static void __gfar_filer_disable(struct gfar_private *priv)
+{
+	struct gfar __iomem *regs = priv->gfargrp[0].regs;
+	u32 temp;
+
+	temp = gfar_read(&regs->rctrl);
+	temp &= ~(RCTRL_FILREN | RCTRL_PRSDEP_INIT);
+	gfar_write(&regs->rctrl, temp);
+}
+
+static void __gfar_filer_enable(struct gfar_private *priv)
+{
+	struct gfar __iomem *regs = priv->gfargrp[0].regs;
+	u32 temp;
+
+	temp = gfar_read(&regs->rctrl);
+	temp |= RCTRL_FILREN | RCTRL_PRSDEP_INIT;
+	gfar_write(&regs->rctrl, temp);
+}
+
+/* Filer rules implementing wol capabilities */
+static void gfar_filer_config_wol(struct gfar_private *priv)
+{
+	unsigned int i;
+	u32 rqfcr;
+
+	__gfar_filer_disable(priv);
+
+	/* clear the filer table, reject any packet by default */
+	rqfcr = RQFCR_RJE | RQFCR_CMP_MATCH;
+	for (i = 0; i <= MAX_FILER_IDX; i++)
+		gfar_write_filer(priv, i, rqfcr, 0);
+
+	i = 0;
+	if (priv->wol_opts & GFAR_WOL_FILER_UCAST) {
+		/* unicast packet, accept it */
+		struct net_device *ndev = priv->ndev;
+		/* get the default rx queue index */
+		u8 qindex = (u8)priv->gfargrp[0].rx_queue->qindex;
+		u32 dest_mac_addr = (ndev->dev_addr[0] << 16) |
+				    (ndev->dev_addr[1] << 8) |
+				     ndev->dev_addr[2];
+
+		rqfcr = (qindex << 10) | RQFCR_AND |
+			RQFCR_CMP_EXACT | RQFCR_PID_DAH;
+
+		gfar_write_filer(priv, i++, rqfcr, dest_mac_addr);
+
+		dest_mac_addr = (ndev->dev_addr[3] << 16) |
+				(ndev->dev_addr[4] << 8) |
+				 ndev->dev_addr[5];
+		rqfcr = (qindex << 10) | RQFCR_GPI |
+			RQFCR_CMP_EXACT | RQFCR_PID_DAL;
+		gfar_write_filer(priv, i++, rqfcr, dest_mac_addr);
 	}
 
-	return val;
+	__gfar_filer_enable(priv);
 }
 
-static noinline void gfar_update_link_state(struct gfar_private *priv)
+static void gfar_filer_restore_table(struct gfar_private *priv)
+{
+	u32 rqfcr, rqfpr;
+	unsigned int i;
+
+	__gfar_filer_disable(priv);
+
+	for (i = 0; i <= MAX_FILER_IDX; i++) {
+		rqfcr = priv->ftp_rqfcr[i];
+		rqfpr = priv->ftp_rqfpr[i];
+		gfar_write_filer(priv, i, rqfcr, rqfpr);
+	}
+
+	__gfar_filer_enable(priv);
+}
+
+/* gfar_start() for Rx only and with the FGPI filer interrupt enabled */
+static void gfar_start_wol_filer(struct gfar_private *priv)
 {
 	struct gfar __iomem *regs = priv->gfargrp[0].regs;
-	struct net_device *ndev = priv->ndev;
-	struct phy_device *phydev = ndev->phydev;
-	struct gfar_priv_rx_q *rx_queue = NULL;
-	int i;
+	u32 tempval;
+	int i = 0;
 
-	if (unlikely(test_bit(GFAR_RESETTING, &priv->state)))
-		return;
+	/* Enable Rx hw queues */
+	gfar_write(&regs->rqueue, priv->rqueue);
 
-	if (phydev->link) {
-		u32 tempval1 = gfar_read(&regs->maccfg1);
-		u32 tempval = gfar_read(&regs->maccfg2);
-		u32 ecntrl = gfar_read(&regs->ecntrl);
-		u32 tx_flow_oldval = (tempval & MACCFG1_TX_FLOW);
+	/* Initialize DMACTRL to have WWR and WOP */
+	tempval = gfar_read(&regs->dmactrl);
+	tempval |= DMACTRL_INIT_SETTINGS;
+	gfar_write(&regs->dmactrl, tempval);
 
-		if (phydev->duplex != priv->oldduplex) {
-			if (!(phydev->duplex))
-				tempval &= ~(MACCFG2_FULL_DUPLEX);
-			else
-				tempval |= MACCFG2_FULL_DUPLEX;
+	/* Make sure we aren't stopped */
+	tempval = gfar_read(&regs->dmactrl);
+	tempval &= ~DMACTRL_GRS;
+	gfar_write(&regs->dmactrl, tempval);
 
-			priv->oldduplex = phydev->duplex;
-		}
+	for (i = 0; i < priv->num_grps; i++) {
+		regs = priv->gfargrp[i].regs;
+		/* Clear RHLT, so that the DMA starts polling now */
+		gfar_write(&regs->rstat, priv->gfargrp[i].rstat);
+		/* enable the Filer General Purpose Interrupt */
+		gfar_write(&regs->imask, IMASK_FGPI);
+	}
 
-		if (phydev->speed != priv->oldspeed) {
-			switch (phydev->speed) {
-			case 1000:
-				tempval =
-				    ((tempval & ~(MACCFG2_IF)) | MACCFG2_GMII);
+	/* Enable Rx DMA */
+	tempval = gfar_read(&regs->maccfg1);
+	tempval |= MACCFG1_RX_EN;
+	gfar_write(&regs->maccfg1, tempval);
+}
 
-				ecntrl &= ~(ECNTRL_R100);
-				break;
-			case 100:
-			case 10:
-				tempval =
-				    ((tempval & ~(MACCFG2_IF)) | MACCFG2_MII);
+static int gfar_suspend(struct device *dev)
+{
+	struct gfar_private *priv = dev_get_drvdata(dev);
+	struct net_device *ndev = priv->ndev;
+	struct gfar __iomem *regs = priv->gfargrp[0].regs;
+	u32 tempval;
+	u16 wol = priv->wol_opts;
 
-				/* Reduced mode distinguishes
-				 * between 10 and 100
-				 */
-				if (phydev->speed == SPEED_100)
-					ecntrl |= ECNTRL_R100;
-				else
-					ecntrl &= ~(ECNTRL_R100);
-				break;
-			default:
-				netif_warn(priv, link, priv->ndev,
-					   "Ack!  Speed (%d) is not 10/100/1000!\n",
-					   phydev->speed);
-				break;
-			}
+	if (!netif_running(ndev))
+		return 0;
 
-			priv->oldspeed = phydev->speed;
-		}
+	disable_napi(priv);
+	netif_tx_lock(ndev);
+	netif_device_detach(ndev);
+	netif_tx_unlock(ndev);
 
-		tempval1 &= ~(MACCFG1_TX_FLOW | MACCFG1_RX_FLOW);
-		tempval1 |= gfar_get_flowctrl_cfg(priv);
+	gfar_halt(priv);
 
-		/* Turn last free buffer recording on */
-		if ((tempval1 & MACCFG1_TX_FLOW) && !tx_flow_oldval) {
-			for (i = 0; i < priv->num_rx_queues; i++) {
-				u32 bdp_dma;
+	if (wol & GFAR_WOL_MAGIC) {
+		/* Enable interrupt on Magic Packet */
+		gfar_write(&regs->imask, IMASK_MAG);
 
-				rx_queue = priv->rx_queue[i];
-				bdp_dma = gfar_rxbd_dma_lastfree(rx_queue);
-				gfar_write(rx_queue->rfbptr, bdp_dma);
-			}
+		/* Enable Magic Packet mode */
+		tempval = gfar_read(&regs->maccfg2);
+		tempval |= MACCFG2_MPEN;
+		gfar_write(&regs->maccfg2, tempval);
 
-			priv->tx_actual_en = 1;
-		}
+		/* re-enable the Rx block */
+		tempval = gfar_read(&regs->maccfg1);
+		tempval |= MACCFG1_RX_EN;
+		gfar_write(&regs->maccfg1, tempval);
 
-		if (unlikely(!(tempval1 & MACCFG1_TX_FLOW) && tx_flow_oldval))
-			priv->tx_actual_en = 0;
+	} else if (wol & GFAR_WOL_FILER_UCAST) {
+		gfar_filer_config_wol(priv);
+		gfar_start_wol_filer(priv);
 
-		gfar_write(&regs->maccfg1, tempval1);
+	} else {
+		phy_stop(ndev->phydev);
+	}
+
+	return 0;
+}
+
+static int gfar_resume(struct device *dev)
+{
+	struct gfar_private *priv = dev_get_drvdata(dev);
+	struct net_device *ndev = priv->ndev;
+	struct gfar __iomem *regs = priv->gfargrp[0].regs;
+	u32 tempval;
+	u16 wol = priv->wol_opts;
+
+	if (!netif_running(ndev))
+		return 0;
+
+	if (wol & GFAR_WOL_MAGIC) {
+		/* Disable Magic Packet mode */
+		tempval = gfar_read(&regs->maccfg2);
+		tempval &= ~MACCFG2_MPEN;
 		gfar_write(&regs->maccfg2, tempval);
-		gfar_write(&regs->ecntrl, ecntrl);
 
-		if (!priv->oldlink)
-			priv->oldlink = 1;
+	} else if (wol & GFAR_WOL_FILER_UCAST) {
+		/* need to stop rx only, tx is already down */
+		gfar_halt(priv);
+		gfar_filer_restore_table(priv);
 
-	} else if (priv->oldlink) {
-		priv->oldlink = 0;
-		priv->oldspeed = 0;
-		priv->oldduplex = -1;
+	} else {
+		phy_start(ndev->phydev);
 	}
 
-	if (netif_msg_link(priv))
-		phy_print_status(phydev);
+	gfar_start(priv);
+
+	netif_device_attach(ndev);
+	enable_napi(priv);
+
+	return 0;
 }
 
+static int gfar_restore(struct device *dev)
+{
+	struct gfar_private *priv = dev_get_drvdata(dev);
+	struct net_device *ndev = priv->ndev;
+
+	if (!netif_running(ndev)) {
+		netif_device_attach(ndev);
+
+		return 0;
+	}
+
+	gfar_init_bds(ndev);
+
+	gfar_mac_reset(priv);
+
+	gfar_init_tx_rx_base(priv);
+
+	gfar_start(priv);
+
+	priv->oldlink = 0;
+	priv->oldspeed = 0;
+	priv->oldduplex = -1;
+
+	if (ndev->phydev)
+		phy_start(ndev->phydev);
+
+	netif_device_attach(ndev);
+	enable_napi(priv);
+
+	return 0;
+}
+
+static const struct dev_pm_ops gfar_pm_ops = {
+	.suspend = gfar_suspend,
+	.resume = gfar_resume,
+	.freeze = gfar_suspend,
+	.thaw = gfar_resume,
+	.restore = gfar_restore,
+};
+
+#define GFAR_PM_OPS (&gfar_pm_ops)
+
+#else
+
+#define GFAR_PM_OPS NULL
+
+#endif
+
 static const struct of_device_id gfar_match[] =
 {
 	{
diff --git a/drivers/net/ethernet/freescale/gianfar.h b/drivers/net/ethernet/freescale/gianfar.h
index 5aa814799d70..68b59d3202e3 100644
--- a/drivers/net/ethernet/freescale/gianfar.h
+++ b/drivers/net/ethernet/freescale/gianfar.h
@@ -1,3 +1,4 @@
+/* SPDX-License-Identifier: GPL-2.0-or-later */
 /*
  * drivers/net/ethernet/freescale/gianfar.h
  *
@@ -11,11 +12,6 @@
  *
  * Copyright 2002-2009, 2011-2013 Freescale Semiconductor, Inc.
  *
- * This program is free software; you can redistribute  it and/or modify it
- * under  the terms of  the GNU General  Public License as published by the
- * Free Software Foundation;  either version 2 of the  License, or (at your
- * option) any later version.
- *
  *  Still left to do:
  *      -Add support for module parameters
  *	-Add patch for ethtool phys id
@@ -56,9 +52,6 @@ struct ethtool_rx_list {
 	unsigned int count;
 };
 
-/* The maximum number of packets to be handled in one call of gfar_poll */
-#define GFAR_DEV_WEIGHT 64
-
 /* Length for FCB */
 #define GMAC_FCB_LEN 8
 
@@ -71,10 +64,7 @@ struct ethtool_rx_list {
 /* Number of bytes to align the rx bufs to */
 #define RXBUF_ALIGNMENT 64
 
-#define PHY_INIT_TIMEOUT 100000
-
 #define DRV_NAME "gfar-enet"
-extern const char gfar_driver_version[];
 
 /* MAXIMUM NUMBER OF QUEUES SUPPORTED */
 #define MAX_TX_QS	0x8
@@ -92,19 +82,15 @@ extern const char gfar_driver_version[];
 #define GFAR_RX_MAX_RING_SIZE   256
 #define GFAR_TX_MAX_RING_SIZE   256
 
-#define GFAR_MAX_FIFO_THRESHOLD 511
-#define GFAR_MAX_FIFO_STARVE	511
-#define GFAR_MAX_FIFO_STARVE_OFF 511
-
 #define FBTHR_SHIFT        24
 #define DEFAULT_RX_LFC_THR  16
 #define DEFAULT_LFC_PTVVAL  4
 
-/* prevent fragmenation by HW in DSA environments */
-#define GFAR_RXB_SIZE roundup(1536 + 8, 64)
-#define GFAR_SKBFRAG_SIZE (RXBUF_ALIGNMENT + GFAR_RXB_SIZE \
-			  + SKB_DATA_ALIGN(sizeof(struct skb_shared_info)))
 #define GFAR_RXB_TRUESIZE 2048
+#define GFAR_SKBFRAG_OVR (RXBUF_ALIGNMENT \
+			  + SKB_DATA_ALIGN(sizeof(struct skb_shared_info)))
+#define GFAR_RXB_SIZE rounddown(GFAR_RXB_TRUESIZE - GFAR_SKBFRAG_OVR, 64)
+#define GFAR_SKBFRAG_SIZE (GFAR_RXB_SIZE + GFAR_SKBFRAG_OVR)
 
 #define TX_RING_MOD_MASK(size) (size-1)
 #define RX_RING_MOD_MASK(size) (size-1)
@@ -113,9 +99,6 @@ extern const char gfar_driver_version[];
 #define DEFAULT_FIFO_TX_THR 0x100
 #define DEFAULT_FIFO_TX_STARVE 0x40
 #define DEFAULT_FIFO_TX_STARVE_OFF 0x80
-#define DEFAULT_BD_STASH 1
-#define DEFAULT_STASH_LENGTH	96
-#define DEFAULT_STASH_INDEX	0
 
 /* The number of Exact Match registers */
 #define GFAR_EM_NUM	15
@@ -143,15 +126,6 @@ extern const char gfar_driver_version[];
 #define DEFAULT_RX_COALESCE 0
 #define DEFAULT_RXCOUNT	0
 
-#define GFAR_SUPPORTED (SUPPORTED_10baseT_Half \
-		| SUPPORTED_10baseT_Full \
-		| SUPPORTED_100baseT_Half \
-		| SUPPORTED_100baseT_Full \
-		| SUPPORTED_Autoneg \
-		| SUPPORTED_MII)
-
-#define GFAR_SUPPORTED_GBIT SUPPORTED_1000baseT_Full
-
 /* TBI register addresses */
 #define MII_TBICON		0x11
 
@@ -189,8 +163,6 @@ extern const char gfar_driver_version[];
 #define ECNTRL_REDUCED_MII_MODE	0x00000004
 #define ECNTRL_SGMII_MODE	0x00000002
 
-#define MRBLR_INIT_SETTINGS	DEFAULT_RX_BUFFER_SIZE
-
 #define MINFLR_INIT_SETTINGS	0x00000040
 
 /* Tqueue control */
@@ -270,12 +242,6 @@ extern const char gfar_driver_version[];
 #define DEFAULT_TXIC mk_ic_value(DEFAULT_TXCOUNT, DEFAULT_TXTIME)
 #define DEFAULT_RXIC mk_ic_value(DEFAULT_RXCOUNT, DEFAULT_RXTIME)
 
-#define skip_bd(bdp, stride, base, ring_size) ({ \
-	typeof(bdp) new_bd = (bdp) + (stride); \
-	(new_bd >= (base) + (ring_size)) ? (new_bd - (ring_size)) : new_bd; })
-
-#define next_bd(bdp, base, ring_size) skip_bd(bdp, 1, base, ring_size)
-
 #define RCTRL_TS_ENABLE 	0x01000000
 #define RCTRL_PAL_MASK		0x001f0000
 #define RCTRL_LFC		0x00004000
@@ -389,11 +355,6 @@ extern const char gfar_driver_version[];
 #define IMASK_RX_DISABLED ((~(IMASK_RX_DEFAULT)) & IMASK_DEFAULT)
 #define IMASK_TX_DISABLED ((~(IMASK_TX_DEFAULT)) & IMASK_DEFAULT)
 
-/* Fifo management */
-#define FIFO_TX_THR_MASK	0x01ff
-#define FIFO_TX_STARVE_MASK	0x01ff
-#define FIFO_TX_STARVE_OFF_MASK	0x01ff
-
 /* Attribute fields */
 
 /* This enables rx snooping for buffers and descriptors */
@@ -481,6 +442,60 @@ extern const char gfar_driver_version[];
 #define RQFPR_PER		0x00000002
 #define RQFPR_EER		0x00000001
 
+/* CAR1 bits */
+#define CAR1_C164		0x80000000
+#define CAR1_C1127		0x40000000
+#define CAR1_C1255		0x20000000
+#define CAR1_C1511		0x10000000
+#define CAR1_C11K		0x08000000
+#define CAR1_C1MAX		0x04000000
+#define CAR1_C1MGV		0x02000000
+#define CAR1_C1REJ		0x00020000
+#define CAR1_C1RBY		0x00010000
+#define CAR1_C1RPK		0x00008000
+#define CAR1_C1RFC		0x00004000
+#define CAR1_C1RMC		0x00002000
+#define CAR1_C1RBC		0x00001000
+#define CAR1_C1RXC		0x00000800
+#define CAR1_C1RXP		0x00000400
+#define CAR1_C1RXU		0x00000200
+#define CAR1_C1RAL		0x00000100
+#define CAR1_C1RFL		0x00000080
+#define CAR1_C1RCD		0x00000040
+#define CAR1_C1RCS		0x00000020
+#define CAR1_C1RUN		0x00000010
+#define CAR1_C1ROV		0x00000008
+#define CAR1_C1RFR		0x00000004
+#define CAR1_C1RJB		0x00000002
+#define CAR1_C1RDR		0x00000001
+
+/* CAM1 bits */
+#define CAM1_M164		0x80000000
+#define CAM1_M1127		0x40000000
+#define CAM1_M1255		0x20000000
+#define CAM1_M1511		0x10000000
+#define CAM1_M11K		0x08000000
+#define CAM1_M1MAX		0x04000000
+#define CAM1_M1MGV		0x02000000
+#define CAM1_M1REJ		0x00020000
+#define CAM1_M1RBY		0x00010000
+#define CAM1_M1RPK		0x00008000
+#define CAM1_M1RFC		0x00004000
+#define CAM1_M1RMC		0x00002000
+#define CAM1_M1RBC		0x00001000
+#define CAM1_M1RXC		0x00000800
+#define CAM1_M1RXP		0x00000400
+#define CAM1_M1RXU		0x00000200
+#define CAM1_M1RAL		0x00000100
+#define CAM1_M1RFL		0x00000080
+#define CAM1_M1RCD		0x00000040
+#define CAM1_M1RCS		0x00000020
+#define CAM1_M1RUN		0x00000010
+#define CAM1_M1ROV		0x00000008
+#define CAM1_M1RFR		0x00000004
+#define CAM1_M1RJB		0x00000002
+#define CAM1_M1RDR		0x00000001
+
 /* TxBD status field bits */
 #define TXBD_READY		0x8000
 #define TXBD_PADCRC		0x4000
@@ -645,6 +660,15 @@ struct rmon_mib
 	u32	cam2;	/* 0x.73c - Carry Mask Register Two */
 };
 
+struct rmon_overflow {
+	/* lock for synchronization of the rdrp field of this struct, and
+	 * CAR1/CAR2 registers
+	 */
+	spinlock_t lock;
+	u32	imask;
+	u64	rdrp;
+};
+
 struct gfar_extra_stats {
 	atomic64_t rx_alloc_err;
 	atomic64_t rx_large;
@@ -945,28 +969,12 @@ enum {
 	MQ_MG_MODE
 };
 
-/* GFAR_SQ_POLLING: Single Queue NAPI polling mode
- *	The driver supports a single pair of RX/Tx queues
- *	per interrupt group (Rx/Tx int line). MQ_MG mode
- *	devices have 2 interrupt groups, so the device will
- *	have a total of 2 Tx and 2 Rx queues in this case.
- * GFAR_MQ_POLLING: Multi Queue NAPI polling mode
- *	The driver supports all the 8 Rx and Tx HW queues
- *	each queue mapped by the Device Tree to one of
- *	the 2 interrupt groups. This mode implies significant
- *	processing overhead (CPU and controller level).
- */
-enum gfar_poll_mode {
-	GFAR_SQ_POLLING = 0,
-	GFAR_MQ_POLLING
-};
-
 /*
  * Per TX queue stats
  */
 struct tx_q_stats {
-	unsigned long tx_packets;
-	unsigned long tx_bytes;
+	u64 tx_packets;
+	u64 tx_bytes;
 };
 
 /**
@@ -1015,9 +1023,9 @@ struct gfar_priv_tx_q {
  * Per RX queue stats
  */
 struct rx_q_stats {
-	unsigned long rx_packets;
-	unsigned long rx_bytes;
-	unsigned long rx_dropped;
+	u64 rx_packets;
+	u64 rx_bytes;
+	u64 rx_dropped;
 };
 
 struct gfar_rx_buff {
@@ -1141,7 +1149,6 @@ struct gfar_private {
 	unsigned long state;
 
 	unsigned short mode;
-	unsigned short poll_mode;
 	unsigned int num_tx_queues;
 	unsigned int num_rx_queues;
 	unsigned int num_grps;
@@ -1149,6 +1156,7 @@ struct gfar_private {
 
 	/* Network Statistics */
 	struct gfar_extra_stats extra_stats;
+	struct rmon_overflow rmon_overflow;
 
 	/* PHY stuff */
 	phy_interface_t interface;
@@ -1330,16 +1338,9 @@ static inline u32 gfar_rxbd_dma_lastfree(struct gfar_priv_rx_q *rxq)
 	return bdp_dma;
 }
 
-irqreturn_t gfar_receive(int irq, void *dev_id);
 int startup_gfar(struct net_device *dev);
 void stop_gfar(struct net_device *dev);
-void reset_gfar(struct net_device *dev);
 void gfar_mac_reset(struct gfar_private *priv);
-void gfar_halt(struct gfar_private *priv);
-void gfar_start(struct gfar_private *priv);
-void gfar_phy_test(struct mii_bus *bus, struct phy_device *phydev, int enable,
-		   u32 regnum, u32 read);
-void gfar_configure_coalescing_all(struct gfar_private *priv);
 int gfar_set_features(struct net_device *dev, netdev_features_t features);
 
 extern const struct ethtool_ops gfar_ethtool_ops;
@@ -1352,13 +1353,6 @@ extern const struct ethtool_ops gfar_ethtool_ops;
 #define RQFCR_PID_PORT_MASK 0xFFFF0000
 #define RQFCR_PID_MAC_MASK 0xFF000000
 
-struct gfar_mask_entry {
-	unsigned int mask; /* The mask value which is valid form start to end */
-	unsigned int start;
-	unsigned int end;
-	unsigned int block; /* Same block values indicate depended entries */
-};
-
 /* Represents a receive filer table entry */
 struct gfar_filer_entry {
 	u32 ctrl;
@@ -1372,7 +1366,4 @@ struct filer_table {
 	struct gfar_filer_entry fe[MAX_FILER_CACHE_IDX + 20];
 };
 
-/* The gianfar_ptp module will set this variable */
-extern int gfar_phc_index;
-
 #endif /* __GIANFAR_H */
diff --git a/drivers/net/ethernet/freescale/gianfar_ethtool.c b/drivers/net/ethernet/freescale/gianfar_ethtool.c
index a93e0199c369..5fd1f7327680 100644
--- a/drivers/net/ethernet/freescale/gianfar_ethtool.c
+++ b/drivers/net/ethernet/freescale/gianfar_ethtool.c
@@ -1,3 +1,4 @@
+// SPDX-License-Identifier: GPL-2.0-only
 /*
  *  drivers/net/ethernet/freescale/gianfar_ethtool.c
  *
@@ -10,10 +11,6 @@
  *  Modifier: Sandeep Gopalpet <sandeep.kumar@freescale.com>
  *
  *  Copyright 2003-2006, 2008-2009, 2011 Freescale Semiconductor, Inc.
- *
- *  This software may be used and distributed according to
- *  the terms of the GNU Public License, Version 2, incorporated herein
- *  by reference.
  */
 
 #define pr_fmt(fmt) KBUILD_MODNAME ": " fmt
@@ -41,24 +38,15 @@
 #include <linux/phy.h>
 #include <linux/sort.h>
 #include <linux/if_vlan.h>
+#include <linux/of.h>
+#include <linux/of_platform.h>
+#include <linux/platform_device.h>
+#include <linux/fsl/ptp_qoriq.h>
 
 #include "gianfar.h"
 
 #define GFAR_MAX_COAL_USECS 0xffff
 #define GFAR_MAX_COAL_FRAMES 0xff
-static void gfar_fill_stats(struct net_device *dev, struct ethtool_stats *dummy,
-			    u64 *buf);
-static void gfar_gstrings(struct net_device *dev, u32 stringset, u8 * buf);
-static int gfar_gcoalesce(struct net_device *dev,
-			  struct ethtool_coalesce *cvals);
-static int gfar_scoalesce(struct net_device *dev,
-			  struct ethtool_coalesce *cvals);
-static void gfar_gringparam(struct net_device *dev,
-			    struct ethtool_ringparam *rvals);
-static int gfar_sringparam(struct net_device *dev,
-			   struct ethtool_ringparam *rvals);
-static void gfar_gdrvinfo(struct net_device *dev,
-			  struct ethtool_drvinfo *drvinfo);
 
 static const char stat_gstrings[][ETH_GSTRING_LEN] = {
 	/* extra stats */
@@ -127,12 +115,14 @@ static const char stat_gstrings[][ETH_GSTRING_LEN] = {
 static void gfar_gstrings(struct net_device *dev, u32 stringset, u8 * buf)
 {
 	struct gfar_private *priv = netdev_priv(dev);
+	int i;
 
 	if (priv->device_flags & FSL_GIANFAR_DEV_HAS_RMON)
-		memcpy(buf, stat_gstrings, GFAR_STATS_LEN * ETH_GSTRING_LEN);
+		for (i = 0; i < GFAR_STATS_LEN; i++)
+			ethtool_puts(&buf, stat_gstrings[i]);
 	else
-		memcpy(buf, stat_gstrings,
-		       GFAR_EXTRA_STATS_LEN * ETH_GSTRING_LEN);
+		for (i = 0; i < GFAR_EXTRA_STATS_LEN; i++)
+			ethtool_puts(&buf, stat_gstrings[i]);
 }
 
 /* Fill in an array of 64-bit statistics from various sources.
@@ -177,11 +167,7 @@ static int gfar_sset_count(struct net_device *dev, int sset)
 static void gfar_gdrvinfo(struct net_device *dev,
 			  struct ethtool_drvinfo *drvinfo)
 {
-	strlcpy(drvinfo->driver, DRV_NAME, sizeof(drvinfo->driver));
-	strlcpy(drvinfo->version, gfar_driver_version,
-		sizeof(drvinfo->version));
-	strlcpy(drvinfo->fw_version, "N/A", sizeof(drvinfo->fw_version));
-	strlcpy(drvinfo->bus_info, "N/A", sizeof(drvinfo->bus_info));
+	strscpy(drvinfo->driver, DRV_NAME, sizeof(drvinfo->driver));
 }
 
 /* Return the length of the register structure */
@@ -228,7 +214,7 @@ static unsigned int gfar_usecs2ticks(struct gfar_private *priv,
 
 	/* Make sure we return a number greater than 0
 	 * if usecs > 0 */
-	return (usecs * 1000 + count - 1) / count;
+	return DIV_ROUND_UP(usecs * 1000, count);
 }
 
 /* Convert ethernet clock ticks to microseconds */
@@ -261,7 +247,9 @@ static unsigned int gfar_ticks2usecs(struct gfar_private *priv,
 /* Get the coalescing parameters, and put them in the cvals
  * structure.  */
 static int gfar_gcoalesce(struct net_device *dev,
-			  struct ethtool_coalesce *cvals)
+			  struct ethtool_coalesce *cvals,
+			  struct kernel_ethtool_coalesce *kernel_coal,
+			  struct netlink_ext_ack *extack)
 {
 	struct gfar_private *priv = netdev_priv(dev);
 	struct gfar_priv_rx_q *rx_queue = NULL;
@@ -290,35 +278,6 @@ static int gfar_gcoalesce(struct net_device *dev,
 	cvals->tx_coalesce_usecs = gfar_ticks2usecs(priv, txtime);
 	cvals->tx_max_coalesced_frames = txcount;
 
-	cvals->use_adaptive_rx_coalesce = 0;
-	cvals->use_adaptive_tx_coalesce = 0;
-
-	cvals->pkt_rate_low = 0;
-	cvals->rx_coalesce_usecs_low = 0;
-	cvals->rx_max_coalesced_frames_low = 0;
-	cvals->tx_coalesce_usecs_low = 0;
-	cvals->tx_max_coalesced_frames_low = 0;
-
-	/* When the packet rate is below pkt_rate_high but above
-	 * pkt_rate_low (both measured in packets per second) the
-	 * normal {rx,tx}_* coalescing parameters are used.
-	 */
-
-	/* When the packet rate is (measured in packets per second)
-	 * is above pkt_rate_high, the {rx,tx}_*_high parameters are
-	 * used.
-	 */
-	cvals->pkt_rate_high = 0;
-	cvals->rx_coalesce_usecs_high = 0;
-	cvals->rx_max_coalesced_frames_high = 0;
-	cvals->tx_coalesce_usecs_high = 0;
-	cvals->tx_max_coalesced_frames_high = 0;
-
-	/* How often to do adaptive coalescing packet rate sampling,
-	 * measured in seconds.  Must not be zero.
-	 */
-	cvals->rate_sample_interval = 0;
-
 	return 0;
 }
 
@@ -327,7 +286,9 @@ static int gfar_gcoalesce(struct net_device *dev,
  * in order for coalescing to be active
  */
 static int gfar_scoalesce(struct net_device *dev,
-			  struct ethtool_coalesce *cvals)
+			  struct ethtool_coalesce *cvals,
+			  struct kernel_ethtool_coalesce *kernel_coal,
+			  struct netlink_ext_ack *extack)
 {
 	struct gfar_private *priv = netdev_priv(dev);
 	int i, err = 0;
@@ -415,7 +376,9 @@ static int gfar_scoalesce(struct net_device *dev,
  * rx, rx_mini, and rx_jumbo rings are the same size, as mini and
  * jumbo are ignored by the driver */
 static void gfar_gringparam(struct net_device *dev,
-			    struct ethtool_ringparam *rvals)
+			    struct ethtool_ringparam *rvals,
+			    struct kernel_ethtool_ringparam *kernel_rvals,
+			    struct netlink_ext_ack *extack)
 {
 	struct gfar_private *priv = netdev_priv(dev);
 	struct gfar_priv_tx_q *tx_queue = NULL;
@@ -442,7 +405,9 @@ static void gfar_gringparam(struct net_device *dev,
  * necessary so that we don't mess things up while we're in motion.
  */
 static int gfar_sringparam(struct net_device *dev,
-			   struct ethtool_ringparam *rvals)
+			   struct ethtool_ringparam *rvals,
+			   struct kernel_ethtool_ringparam *kernel_rvals,
+			   struct netlink_ext_ack *extack)
 {
 	struct gfar_private *priv = netdev_priv(dev);
 	int err = 0, i;
@@ -501,65 +466,44 @@ static int gfar_spauseparam(struct net_device *dev,
 	struct gfar_private *priv = netdev_priv(dev);
 	struct phy_device *phydev = dev->phydev;
 	struct gfar __iomem *regs = priv->gfargrp[0].regs;
-	u32 oldadv, newadv;
 
 	if (!phydev)
 		return -ENODEV;
 
-	if (!(phydev->supported & SUPPORTED_Pause) ||
-	    (!(phydev->supported & SUPPORTED_Asym_Pause) &&
-	     (epause->rx_pause != epause->tx_pause)))
+	if (!phy_validate_pause(phydev, epause))
 		return -EINVAL;
 
 	priv->rx_pause_en = priv->tx_pause_en = 0;
+	phy_set_asym_pause(phydev, epause->rx_pause, epause->tx_pause);
 	if (epause->rx_pause) {
 		priv->rx_pause_en = 1;
 
 		if (epause->tx_pause) {
 			priv->tx_pause_en = 1;
-			/* FLOW_CTRL_RX & TX */
-			newadv = ADVERTISED_Pause;
-		} else  /* FLOW_CTLR_RX */
-			newadv = ADVERTISED_Pause | ADVERTISED_Asym_Pause;
+		}
 	} else if (epause->tx_pause) {
 		priv->tx_pause_en = 1;
-		/* FLOW_CTLR_TX */
-		newadv = ADVERTISED_Asym_Pause;
-	} else
-		newadv = 0;
+	}
 
 	if (epause->autoneg)
 		priv->pause_aneg_en = 1;
 	else
 		priv->pause_aneg_en = 0;
 
-	oldadv = phydev->advertising &
-		(ADVERTISED_Pause | ADVERTISED_Asym_Pause);
-	if (oldadv != newadv) {
-		phydev->advertising &=
-			~(ADVERTISED_Pause | ADVERTISED_Asym_Pause);
-		phydev->advertising |= newadv;
-		if (phydev->autoneg)
-			/* inform link partner of our
-			 * new flow ctrl settings
-			 */
-			return phy_start_aneg(phydev);
-
-		if (!epause->autoneg) {
-			u32 tempval;
-			tempval = gfar_read(&regs->maccfg1);
-			tempval &= ~(MACCFG1_TX_FLOW | MACCFG1_RX_FLOW);
-
-			priv->tx_actual_en = 0;
-			if (priv->tx_pause_en) {
-				priv->tx_actual_en = 1;
-				tempval |= MACCFG1_TX_FLOW;
-			}
+	if (!epause->autoneg) {
+		u32 tempval = gfar_read(&regs->maccfg1);
+
+		tempval &= ~(MACCFG1_TX_FLOW | MACCFG1_RX_FLOW);
 
-			if (priv->rx_pause_en)
-				tempval |= MACCFG1_RX_FLOW;
-			gfar_write(&regs->maccfg1, tempval);
+		priv->tx_actual_en = 0;
+		if (priv->tx_pause_en) {
+			priv->tx_actual_en = 1;
+			tempval |= MACCFG1_TX_FLOW;
 		}
+
+		if (priv->rx_pause_en)
+			tempval |= MACCFG1_RX_FLOW;
+		gfar_write(&regs->maccfg1, tempval);
 	}
 
 	return 0;
@@ -738,7 +682,6 @@ static void ethflow_to_filer_rules (struct gfar_private *priv, u64 ethflow)
 static int gfar_ethflow_to_filer_table(struct gfar_private *priv, u64 ethflow,
 				       u64 class)
 {
-	unsigned int last_rule_idx = priv->cur_filer_idx;
 	unsigned int cmp_rqfpr;
 	unsigned int *local_rqfpr;
 	unsigned int *local_rqfcr;
@@ -817,7 +760,6 @@ static int gfar_ethflow_to_filer_table(struct gfar_private *priv, u64 ethflow,
 	}
 
 	priv->cur_filer_idx = l - 1;
-	last_rule_idx = l;
 
 	/* hash rules */
 	ethflow_to_filer_rules(priv, ethflow);
@@ -839,14 +781,26 @@ err:
 	return ret;
 }
 
-static int gfar_set_hash_opts(struct gfar_private *priv,
-			      struct ethtool_rxnfc *cmd)
+static int gfar_set_rxfh_fields(struct net_device *dev,
+				const struct ethtool_rxfh_fields *cmd,
+				struct netlink_ext_ack *extack)
 {
+	struct gfar_private *priv = netdev_priv(dev);
+	int ret;
+
+	if (test_bit(GFAR_RESETTING, &priv->state))
+		return -EBUSY;
+
+	mutex_lock(&priv->rx_queue_access);
+
+	ret = 0;
 	/* write the filer rules here */
 	if (!gfar_ethflow_to_filer_table(priv, cmd->data, cmd->flow_type))
-		return -EINVAL;
+		ret = -EINVAL;
 
-	return 0;
+	mutex_unlock(&priv->rx_queue_access);
+
+	return ret;
 }
 
 static int gfar_check_filer_hardware(struct gfar_private *priv)
@@ -1155,11 +1109,9 @@ static int gfar_convert_to_filer(struct ethtool_rx_flow_spec *rule,
 		prio = vlan_tci_prio(rule);
 		prio_mask = vlan_tci_priom(rule);
 
-		if (cfi == VLAN_TAG_PRESENT && cfi_mask == VLAN_TAG_PRESENT) {
-			vlan |= RQFPR_CFI;
-			vlan_mask |= RQFPR_CFI;
-		} else if (cfi != VLAN_TAG_PRESENT &&
-			   cfi_mask == VLAN_TAG_PRESENT) {
+		if (cfi_mask) {
+			if (cfi)
+				vlan |= RQFPR_CFI;
 			vlan_mask |= RQFPR_CFI;
 		}
 	}
@@ -1458,9 +1410,6 @@ static int gfar_set_nfc(struct net_device *dev, struct ethtool_rxnfc *cmd)
 	mutex_lock(&priv->rx_queue_access);
 
 	switch (cmd->cmd) {
-	case ETHTOOL_SRXFH:
-		ret = gfar_set_hash_opts(priv, cmd);
-		break;
 	case ETHTOOL_SRXCLSRLINS:
 		if ((cmd->fs.ring_cookie != RX_CLS_FLOW_DISC &&
 		     cmd->fs.ring_cookie >= priv->num_rx_queues) ||
@@ -1509,24 +1458,36 @@ static int gfar_get_nfc(struct net_device *dev, struct ethtool_rxnfc *cmd,
 	return ret;
 }
 
-int gfar_phc_index = -1;
-EXPORT_SYMBOL(gfar_phc_index);
-
 static int gfar_get_ts_info(struct net_device *dev,
-			    struct ethtool_ts_info *info)
+			    struct kernel_ethtool_ts_info *info)
 {
 	struct gfar_private *priv = netdev_priv(dev);
+	struct platform_device *ptp_dev;
+	struct device_node *ptp_node;
+	struct ptp_qoriq *ptp = NULL;
 
 	if (!(priv->device_flags & FSL_GIANFAR_DEV_HAS_TIMER)) {
-		info->so_timestamping = SOF_TIMESTAMPING_RX_SOFTWARE |
-					SOF_TIMESTAMPING_SOFTWARE;
-		info->phc_index = -1;
+		info->so_timestamping = SOF_TIMESTAMPING_TX_SOFTWARE;
 		return 0;
 	}
+
+	ptp_node = of_find_compatible_node(NULL, NULL, "fsl,etsec-ptp");
+	if (ptp_node) {
+		ptp_dev = of_find_device_by_node(ptp_node);
+		of_node_put(ptp_node);
+		if (ptp_dev) {
+			ptp = platform_get_drvdata(ptp_dev);
+			put_device(&ptp_dev->dev);
+		}
+	}
+
+	if (ptp)
+		info->phc_index = ptp->phc_index;
+
 	info->so_timestamping = SOF_TIMESTAMPING_TX_HARDWARE |
 				SOF_TIMESTAMPING_RX_HARDWARE |
-				SOF_TIMESTAMPING_RAW_HARDWARE;
-	info->phc_index = gfar_phc_index;
+				SOF_TIMESTAMPING_RAW_HARDWARE |
+				SOF_TIMESTAMPING_TX_SOFTWARE;
 	info->tx_types = (1 << HWTSTAMP_TX_OFF) |
 			 (1 << HWTSTAMP_TX_ON);
 	info->rx_filters = (1 << HWTSTAMP_FILTER_NONE) |
@@ -1535,6 +1496,8 @@ static int gfar_get_ts_info(struct net_device *dev,
 }
 
 const struct ethtool_ops gfar_ethtool_ops = {
+	.supported_coalesce_params = ETHTOOL_COALESCE_USECS |
+				     ETHTOOL_COALESCE_MAX_FRAMES,
 	.get_drvinfo = gfar_gdrvinfo,
 	.get_regs_len = gfar_reglen,
 	.get_regs = gfar_get_regs,
@@ -1556,6 +1519,7 @@ const struct ethtool_ops gfar_ethtool_ops = {
 #endif
 	.set_rxnfc = gfar_set_nfc,
 	.get_rxnfc = gfar_get_nfc,
+	.set_rxfh_fields = gfar_set_rxfh_fields,
 	.get_ts_info = gfar_get_ts_info,
 	.get_link_ksettings = phy_ethtool_get_link_ksettings,
 	.set_link_ksettings = phy_ethtool_set_link_ksettings,
diff --git a/drivers/net/ethernet/freescale/gianfar_ptp.c b/drivers/net/ethernet/freescale/gianfar_ptp.c
deleted file mode 100644
index 721be13081f9..000000000000
--- a/drivers/net/ethernet/freescale/gianfar_ptp.c
+++ /dev/null
@@ -1,573 +0,0 @@
-/*
- * PTP 1588 clock using the eTSEC
- *
- * Copyright (C) 2010 OMICRON electronics GmbH
- *
- *  This program is free software; you can redistribute it and/or modify
- *  it under the terms of the GNU General Public License as published by
- *  the Free Software Foundation; either version 2 of the License, or
- *  (at your option) any later version.
- *
- *  This program is distributed in the hope that it will be useful,
- *  but WITHOUT ANY WARRANTY; without even the implied warranty of
- *  MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
- *  GNU General Public License for more details.
- *
- *  You should have received a copy of the GNU General Public License
- *  along with this program; if not, write to the Free Software
- *  Foundation, Inc., 675 Mass Ave, Cambridge, MA 02139, USA.
- */
-
-#define pr_fmt(fmt) KBUILD_MODNAME ": " fmt
-
-#include <linux/device.h>
-#include <linux/hrtimer.h>
-#include <linux/interrupt.h>
-#include <linux/kernel.h>
-#include <linux/module.h>
-#include <linux/of.h>
-#include <linux/of_platform.h>
-#include <linux/timex.h>
-#include <linux/io.h>
-
-#include <linux/ptp_clock_kernel.h>
-
-#include "gianfar.h"
-
-/*
- * gianfar ptp registers
- * Generated by regen.tcl on Thu May 13 01:38:57 PM CEST 2010
- */
-struct gianfar_ptp_registers {
-	u32 tmr_ctrl;     /* Timer control register */
-	u32 tmr_tevent;   /* Timestamp event register */
-	u32 tmr_temask;   /* Timer event mask register */
-	u32 tmr_pevent;   /* Timestamp event register */
-	u32 tmr_pemask;   /* Timer event mask register */
-	u32 tmr_stat;     /* Timestamp status register */
-	u32 tmr_cnt_h;    /* Timer counter high register */
-	u32 tmr_cnt_l;    /* Timer counter low register */
-	u32 tmr_add;      /* Timer drift compensation addend register */
-	u32 tmr_acc;      /* Timer accumulator register */
-	u32 tmr_prsc;     /* Timer prescale */
-	u8  res1[4];
-	u32 tmroff_h;     /* Timer offset high */
-	u32 tmroff_l;     /* Timer offset low */
-	u8  res2[8];
-	u32 tmr_alarm1_h; /* Timer alarm 1 high register */
-	u32 tmr_alarm1_l; /* Timer alarm 1 high register */
-	u32 tmr_alarm2_h; /* Timer alarm 2 high register */
-	u32 tmr_alarm2_l; /* Timer alarm 2 high register */
-	u8  res3[48];
-	u32 tmr_fiper1;   /* Timer fixed period interval */
-	u32 tmr_fiper2;   /* Timer fixed period interval */
-	u32 tmr_fiper3;   /* Timer fixed period interval */
-	u8  res4[20];
-	u32 tmr_etts1_h;  /* Timestamp of general purpose external trigger */
-	u32 tmr_etts1_l;  /* Timestamp of general purpose external trigger */
-	u32 tmr_etts2_h;  /* Timestamp of general purpose external trigger */
-	u32 tmr_etts2_l;  /* Timestamp of general purpose external trigger */
-};
-
-/* Bit definitions for the TMR_CTRL register */
-#define ALM1P                 (1<<31) /* Alarm1 output polarity */
-#define ALM2P                 (1<<30) /* Alarm2 output polarity */
-#define FIPERST               (1<<28) /* FIPER start indication */
-#define PP1L                  (1<<27) /* Fiper1 pulse loopback mode enabled. */
-#define PP2L                  (1<<26) /* Fiper2 pulse loopback mode enabled. */
-#define TCLK_PERIOD_SHIFT     (16) /* 1588 timer reference clock period. */
-#define TCLK_PERIOD_MASK      (0x3ff)
-#define RTPE                  (1<<15) /* Record Tx Timestamp to PAL Enable. */
-#define FRD                   (1<<14) /* FIPER Realignment Disable */
-#define ESFDP                 (1<<11) /* External Tx/Rx SFD Polarity. */
-#define ESFDE                 (1<<10) /* External Tx/Rx SFD Enable. */
-#define ETEP2                 (1<<9) /* External trigger 2 edge polarity */
-#define ETEP1                 (1<<8) /* External trigger 1 edge polarity */
-#define COPH                  (1<<7) /* Generated clock output phase. */
-#define CIPH                  (1<<6) /* External oscillator input clock phase */
-#define TMSR                  (1<<5) /* Timer soft reset. */
-#define BYP                   (1<<3) /* Bypass drift compensated clock */
-#define TE                    (1<<2) /* 1588 timer enable. */
-#define CKSEL_SHIFT           (0)    /* 1588 Timer reference clock source */
-#define CKSEL_MASK            (0x3)
-
-/* Bit definitions for the TMR_TEVENT register */
-#define ETS2                  (1<<25) /* External trigger 2 timestamp sampled */
-#define ETS1                  (1<<24) /* External trigger 1 timestamp sampled */
-#define ALM2                  (1<<17) /* Current time = alarm time register 2 */
-#define ALM1                  (1<<16) /* Current time = alarm time register 1 */
-#define PP1                   (1<<7)  /* periodic pulse generated on FIPER1 */
-#define PP2                   (1<<6)  /* periodic pulse generated on FIPER2 */
-#define PP3                   (1<<5)  /* periodic pulse generated on FIPER3 */
-
-/* Bit definitions for the TMR_TEMASK register */
-#define ETS2EN                (1<<25) /* External trigger 2 timestamp enable */
-#define ETS1EN                (1<<24) /* External trigger 1 timestamp enable */
-#define ALM2EN                (1<<17) /* Timer ALM2 event enable */
-#define ALM1EN                (1<<16) /* Timer ALM1 event enable */
-#define PP1EN                 (1<<7) /* Periodic pulse event 1 enable */
-#define PP2EN                 (1<<6) /* Periodic pulse event 2 enable */
-
-/* Bit definitions for the TMR_PEVENT register */
-#define TXP2                  (1<<9) /* PTP transmitted timestamp im TXTS2 */
-#define TXP1                  (1<<8) /* PTP transmitted timestamp in TXTS1 */
-#define RXP                   (1<<0) /* PTP frame has been received */
-
-/* Bit definitions for the TMR_PEMASK register */
-#define TXP2EN                (1<<9) /* Transmit PTP packet event 2 enable */
-#define TXP1EN                (1<<8) /* Transmit PTP packet event 1 enable */
-#define RXPEN                 (1<<0) /* Receive PTP packet event enable */
-
-/* Bit definitions for the TMR_STAT register */
-#define STAT_VEC_SHIFT        (0) /* Timer general purpose status vector */
-#define STAT_VEC_MASK         (0x3f)
-
-/* Bit definitions for the TMR_PRSC register */
-#define PRSC_OCK_SHIFT        (0) /* Output clock division/prescale factor. */
-#define PRSC_OCK_MASK         (0xffff)
-
-
-#define DRIVER		"gianfar_ptp"
-#define DEFAULT_CKSEL	1
-#define N_EXT_TS	2
-#define REG_SIZE	sizeof(struct gianfar_ptp_registers)
-
-struct etsects {
-	struct gianfar_ptp_registers __iomem *regs;
-	spinlock_t lock; /* protects regs */
-	struct ptp_clock *clock;
-	struct ptp_clock_info caps;
-	struct resource *rsrc;
-	int irq;
-	u64 alarm_interval; /* for periodic alarm */
-	u64 alarm_value;
-	u32 tclk_period;  /* nanoseconds */
-	u32 tmr_prsc;
-	u32 tmr_add;
-	u32 cksel;
-	u32 tmr_fiper1;
-	u32 tmr_fiper2;
-};
-
-/*
- * Register access functions
- */
-
-/* Caller must hold etsects->lock. */
-static u64 tmr_cnt_read(struct etsects *etsects)
-{
-	u64 ns;
-	u32 lo, hi;
-
-	lo = gfar_read(&etsects->regs->tmr_cnt_l);
-	hi = gfar_read(&etsects->regs->tmr_cnt_h);
-	ns = ((u64) hi) << 32;
-	ns |= lo;
-	return ns;
-}
-
-/* Caller must hold etsects->lock. */
-static void tmr_cnt_write(struct etsects *etsects, u64 ns)
-{
-	u32 hi = ns >> 32;
-	u32 lo = ns & 0xffffffff;
-
-	gfar_write(&etsects->regs->tmr_cnt_l, lo);
-	gfar_write(&etsects->regs->tmr_cnt_h, hi);
-}
-
-/* Caller must hold etsects->lock. */
-static void set_alarm(struct etsects *etsects)
-{
-	u64 ns;
-	u32 lo, hi;
-
-	ns = tmr_cnt_read(etsects) + 1500000000ULL;
-	ns = div_u64(ns, 1000000000UL) * 1000000000ULL;
-	ns -= etsects->tclk_period;
-	hi = ns >> 32;
-	lo = ns & 0xffffffff;
-	gfar_write(&etsects->regs->tmr_alarm1_l, lo);
-	gfar_write(&etsects->regs->tmr_alarm1_h, hi);
-}
-
-/* Caller must hold etsects->lock. */
-static void set_fipers(struct etsects *etsects)
-{
-	set_alarm(etsects);
-	gfar_write(&etsects->regs->tmr_fiper1, etsects->tmr_fiper1);
-	gfar_write(&etsects->regs->tmr_fiper2, etsects->tmr_fiper2);
-}
-
-/*
- * Interrupt service routine
- */
-
-static irqreturn_t isr(int irq, void *priv)
-{
-	struct etsects *etsects = priv;
-	struct ptp_clock_event event;
-	u64 ns;
-	u32 ack = 0, lo, hi, mask, val;
-
-	val = gfar_read(&etsects->regs->tmr_tevent);
-
-	if (val & ETS1) {
-		ack |= ETS1;
-		hi = gfar_read(&etsects->regs->tmr_etts1_h);
-		lo = gfar_read(&etsects->regs->tmr_etts1_l);
-		event.type = PTP_CLOCK_EXTTS;
-		event.index = 0;
-		event.timestamp = ((u64) hi) << 32;
-		event.timestamp |= lo;
-		ptp_clock_event(etsects->clock, &event);
-	}
-
-	if (val & ETS2) {
-		ack |= ETS2;
-		hi = gfar_read(&etsects->regs->tmr_etts2_h);
-		lo = gfar_read(&etsects->regs->tmr_etts2_l);
-		event.type = PTP_CLOCK_EXTTS;
-		event.index = 1;
-		event.timestamp = ((u64) hi) << 32;
-		event.timestamp |= lo;
-		ptp_clock_event(etsects->clock, &event);
-	}
-
-	if (val & ALM2) {
-		ack |= ALM2;
-		if (etsects->alarm_value) {
-			event.type = PTP_CLOCK_ALARM;
-			event.index = 0;
-			event.timestamp = etsects->alarm_value;
-			ptp_clock_event(etsects->clock, &event);
-		}
-		if (etsects->alarm_interval) {
-			ns = etsects->alarm_value + etsects->alarm_interval;
-			hi = ns >> 32;
-			lo = ns & 0xffffffff;
-			spin_lock(&etsects->lock);
-			gfar_write(&etsects->regs->tmr_alarm2_l, lo);
-			gfar_write(&etsects->regs->tmr_alarm2_h, hi);
-			spin_unlock(&etsects->lock);
-			etsects->alarm_value = ns;
-		} else {
-			gfar_write(&etsects->regs->tmr_tevent, ALM2);
-			spin_lock(&etsects->lock);
-			mask = gfar_read(&etsects->regs->tmr_temask);
-			mask &= ~ALM2EN;
-			gfar_write(&etsects->regs->tmr_temask, mask);
-			spin_unlock(&etsects->lock);
-			etsects->alarm_value = 0;
-			etsects->alarm_interval = 0;
-		}
-	}
-
-	if (val & PP1) {
-		ack |= PP1;
-		event.type = PTP_CLOCK_PPS;
-		ptp_clock_event(etsects->clock, &event);
-	}
-
-	if (ack) {
-		gfar_write(&etsects->regs->tmr_tevent, ack);
-		return IRQ_HANDLED;
-	} else
-		return IRQ_NONE;
-}
-
-/*
- * PTP clock operations
- */
-
-static int ptp_gianfar_adjfine(struct ptp_clock_info *ptp, long scaled_ppm)
-{
-	u64 adj, diff;
-	u32 tmr_add;
-	int neg_adj = 0;
-	struct etsects *etsects = container_of(ptp, struct etsects, caps);
-
-	if (scaled_ppm < 0) {
-		neg_adj = 1;
-		scaled_ppm = -scaled_ppm;
-	}
-	tmr_add = etsects->tmr_add;
-	adj = tmr_add;
-
-	/* calculate diff as adj*(scaled_ppm/65536)/1000000
-	 * and round() to the nearest integer
-	 */
-	adj *= scaled_ppm;
-	diff = div_u64(adj, 8000000);
-	diff = (diff >> 13) + ((diff >> 12) & 1);
-
-	tmr_add = neg_adj ? tmr_add - diff : tmr_add + diff;
-
-	gfar_write(&etsects->regs->tmr_add, tmr_add);
-
-	return 0;
-}
-
-static int ptp_gianfar_adjtime(struct ptp_clock_info *ptp, s64 delta)
-{
-	s64 now;
-	unsigned long flags;
-	struct etsects *etsects = container_of(ptp, struct etsects, caps);
-
-	spin_lock_irqsave(&etsects->lock, flags);
-
-	now = tmr_cnt_read(etsects);
-	now += delta;
-	tmr_cnt_write(etsects, now);
-
-	spin_unlock_irqrestore(&etsects->lock, flags);
-
-	set_fipers(etsects);
-
-	return 0;
-}
-
-static int ptp_gianfar_gettime(struct ptp_clock_info *ptp,
-			       struct timespec64 *ts)
-{
-	u64 ns;
-	unsigned long flags;
-	struct etsects *etsects = container_of(ptp, struct etsects, caps);
-
-	spin_lock_irqsave(&etsects->lock, flags);
-
-	ns = tmr_cnt_read(etsects);
-
-	spin_unlock_irqrestore(&etsects->lock, flags);
-
-	*ts = ns_to_timespec64(ns);
-
-	return 0;
-}
-
-static int ptp_gianfar_settime(struct ptp_clock_info *ptp,
-			       const struct timespec64 *ts)
-{
-	u64 ns;
-	unsigned long flags;
-	struct etsects *etsects = container_of(ptp, struct etsects, caps);
-
-	ns = timespec64_to_ns(ts);
-
-	spin_lock_irqsave(&etsects->lock, flags);
-
-	tmr_cnt_write(etsects, ns);
-	set_fipers(etsects);
-
-	spin_unlock_irqrestore(&etsects->lock, flags);
-
-	return 0;
-}
-
-static int ptp_gianfar_enable(struct ptp_clock_info *ptp,
-			      struct ptp_clock_request *rq, int on)
-{
-	struct etsects *etsects = container_of(ptp, struct etsects, caps);
-	unsigned long flags;
-	u32 bit, mask;
-
-	switch (rq->type) {
-	case PTP_CLK_REQ_EXTTS:
-		switch (rq->extts.index) {
-		case 0:
-			bit = ETS1EN;
-			break;
-		case 1:
-			bit = ETS2EN;
-			break;
-		default:
-			return -EINVAL;
-		}
-		spin_lock_irqsave(&etsects->lock, flags);
-		mask = gfar_read(&etsects->regs->tmr_temask);
-		if (on)
-			mask |= bit;
-		else
-			mask &= ~bit;
-		gfar_write(&etsects->regs->tmr_temask, mask);
-		spin_unlock_irqrestore(&etsects->lock, flags);
-		return 0;
-
-	case PTP_CLK_REQ_PPS:
-		spin_lock_irqsave(&etsects->lock, flags);
-		mask = gfar_read(&etsects->regs->tmr_temask);
-		if (on)
-			mask |= PP1EN;
-		else
-			mask &= ~PP1EN;
-		gfar_write(&etsects->regs->tmr_temask, mask);
-		spin_unlock_irqrestore(&etsects->lock, flags);
-		return 0;
-
-	default:
-		break;
-	}
-
-	return -EOPNOTSUPP;
-}
-
-static struct ptp_clock_info ptp_gianfar_caps = {
-	.owner		= THIS_MODULE,
-	.name		= "gianfar clock",
-	.max_adj	= 512000,
-	.n_alarm	= 0,
-	.n_ext_ts	= N_EXT_TS,
-	.n_per_out	= 0,
-	.n_pins		= 0,
-	.pps		= 1,
-	.adjfine	= ptp_gianfar_adjfine,
-	.adjtime	= ptp_gianfar_adjtime,
-	.gettime64	= ptp_gianfar_gettime,
-	.settime64	= ptp_gianfar_settime,
-	.enable		= ptp_gianfar_enable,
-};
-
-static int gianfar_ptp_probe(struct platform_device *dev)
-{
-	struct device_node *node = dev->dev.of_node;
-	struct etsects *etsects;
-	struct timespec64 now;
-	int err = -ENOMEM;
-	u32 tmr_ctrl;
-	unsigned long flags;
-
-	etsects = kzalloc(sizeof(*etsects), GFP_KERNEL);
-	if (!etsects)
-		goto no_memory;
-
-	err = -ENODEV;
-
-	etsects->caps = ptp_gianfar_caps;
-
-	if (of_property_read_u32(node, "fsl,cksel", &etsects->cksel))
-		etsects->cksel = DEFAULT_CKSEL;
-
-	if (of_property_read_u32(node,
-				 "fsl,tclk-period", &etsects->tclk_period) ||
-	    of_property_read_u32(node,
-				 "fsl,tmr-prsc", &etsects->tmr_prsc) ||
-	    of_property_read_u32(node,
-				 "fsl,tmr-add", &etsects->tmr_add) ||
-	    of_property_read_u32(node,
-				 "fsl,tmr-fiper1", &etsects->tmr_fiper1) ||
-	    of_property_read_u32(node,
-				 "fsl,tmr-fiper2", &etsects->tmr_fiper2) ||
-	    of_property_read_u32(node,
-				 "fsl,max-adj", &etsects->caps.max_adj)) {
-		pr_err("device tree node missing required elements\n");
-		goto no_node;
-	}
-
-	etsects->irq = platform_get_irq(dev, 0);
-
-	if (etsects->irq < 0) {
-		pr_err("irq not in device tree\n");
-		goto no_node;
-	}
-	if (request_irq(etsects->irq, isr, 0, DRIVER, etsects)) {
-		pr_err("request_irq failed\n");
-		goto no_node;
-	}
-
-	etsects->rsrc = platform_get_resource(dev, IORESOURCE_MEM, 0);
-	if (!etsects->rsrc) {
-		pr_err("no resource\n");
-		goto no_resource;
-	}
-	if (request_resource(&iomem_resource, etsects->rsrc)) {
-		pr_err("resource busy\n");
-		goto no_resource;
-	}
-
-	spin_lock_init(&etsects->lock);
-
-	etsects->regs = ioremap(etsects->rsrc->start,
-				resource_size(etsects->rsrc));
-	if (!etsects->regs) {
-		pr_err("ioremap ptp registers failed\n");
-		goto no_ioremap;
-	}
-	getnstimeofday64(&now);
-	ptp_gianfar_settime(&etsects->caps, &now);
-
-	tmr_ctrl =
-	  (etsects->tclk_period & TCLK_PERIOD_MASK) << TCLK_PERIOD_SHIFT |
-	  (etsects->cksel & CKSEL_MASK) << CKSEL_SHIFT;
-
-	spin_lock_irqsave(&etsects->lock, flags);
-
-	gfar_write(&etsects->regs->tmr_ctrl,   tmr_ctrl);
-	gfar_write(&etsects->regs->tmr_add,    etsects->tmr_add);
-	gfar_write(&etsects->regs->tmr_prsc,   etsects->tmr_prsc);
-	gfar_write(&etsects->regs->tmr_fiper1, etsects->tmr_fiper1);
-	gfar_write(&etsects->regs->tmr_fiper2, etsects->tmr_fiper2);
-	set_alarm(etsects);
-	gfar_write(&etsects->regs->tmr_ctrl,   tmr_ctrl|FIPERST|RTPE|TE|FRD);
-
-	spin_unlock_irqrestore(&etsects->lock, flags);
-
-	etsects->clock = ptp_clock_register(&etsects->caps, &dev->dev);
-	if (IS_ERR(etsects->clock)) {
-		err = PTR_ERR(etsects->clock);
-		goto no_clock;
-	}
-	gfar_phc_index = ptp_clock_index(etsects->clock);
-
-	platform_set_drvdata(dev, etsects);
-
-	return 0;
-
-no_clock:
-	iounmap(etsects->regs);
-no_ioremap:
-	release_resource(etsects->rsrc);
-no_resource:
-	free_irq(etsects->irq, etsects);
-no_node:
-	kfree(etsects);
-no_memory:
-	return err;
-}
-
-static int gianfar_ptp_remove(struct platform_device *dev)
-{
-	struct etsects *etsects = platform_get_drvdata(dev);
-
-	gfar_write(&etsects->regs->tmr_temask, 0);
-	gfar_write(&etsects->regs->tmr_ctrl,   0);
-
-	gfar_phc_index = -1;
-	ptp_clock_unregister(etsects->clock);
-	iounmap(etsects->regs);
-	release_resource(etsects->rsrc);
-	free_irq(etsects->irq, etsects);
-	kfree(etsects);
-
-	return 0;
-}
-
-static const struct of_device_id match_table[] = {
-	{ .compatible = "fsl,etsec-ptp" },
-	{},
-};
-MODULE_DEVICE_TABLE(of, match_table);
-
-static struct platform_driver gianfar_ptp_driver = {
-	.driver = {
-		.name		= "gianfar_ptp",
-		.of_match_table	= match_table,
-	},
-	.probe       = gianfar_ptp_probe,
-	.remove      = gianfar_ptp_remove,
-};
-
-module_platform_driver(gianfar_ptp_driver);
-
-MODULE_AUTHOR("Richard Cochran <richardcochran@gmail.com>");
-MODULE_DESCRIPTION("PTP clock using the eTSEC");
-MODULE_LICENSE("GPL");
diff --git a/drivers/net/ethernet/freescale/ucc_geth.c b/drivers/net/ethernet/freescale/ucc_geth.c
index f77ba9fa257b..affd5a6c44e7 100644
--- a/drivers/net/ethernet/freescale/ucc_geth.c
+++ b/drivers/net/ethernet/freescale/ucc_geth.c
@@ -1,3 +1,4 @@
+// SPDX-License-Identifier: GPL-2.0-or-later
 /*
  * Copyright (C) 2006-2009 Freescale Semicondutor, Inc. All rights reserved.
  *
@@ -6,11 +7,6 @@
  *
  * Description:
  * QE UCC Gigabit Ethernet Driver
- *
- * This program is free software; you can redistribute  it and/or modify it
- * under  the terms of  the GNU General  Public License as published by the
- * Free Software Foundation;  either version 2 of the  License, or (at your
- * option) any later version.
  */
 
 #define pr_fmt(fmt) KBUILD_MODNAME ": " fmt
@@ -30,12 +26,15 @@
 #include <linux/dma-mapping.h>
 #include <linux/mii.h>
 #include <linux/phy.h>
+#include <linux/phylink.h>
 #include <linux/workqueue.h>
+#include <linux/of.h>
 #include <linux/of_address.h>
 #include <linux/of_irq.h>
 #include <linux/of_mdio.h>
 #include <linux/of_net.h>
-#include <linux/of_platform.h>
+#include <linux/platform_device.h>
+#include <linux/rtnetlink.h>
 
 #include <linux/uaccess.h>
 #include <asm/irq.h>
@@ -73,9 +72,32 @@ static struct {
 module_param_named(debug, debug.msg_enable, int, 0);
 MODULE_PARM_DESC(debug, "Debug verbosity level (0=none, ..., 0xffff=all)");
 
-static struct ucc_geth_info ugeth_primary_info = {
+static int ucc_geth_thread_count(enum ucc_geth_num_of_threads idx)
+{
+	static const u8 count[] = {
+		[UCC_GETH_NUM_OF_THREADS_1] = 1,
+		[UCC_GETH_NUM_OF_THREADS_2] = 2,
+		[UCC_GETH_NUM_OF_THREADS_4] = 4,
+		[UCC_GETH_NUM_OF_THREADS_6] = 6,
+		[UCC_GETH_NUM_OF_THREADS_8] = 8,
+	};
+	if (idx >= ARRAY_SIZE(count))
+		return 0;
+	return count[idx];
+}
+
+static inline int ucc_geth_tx_queues(const struct ucc_geth_info *info)
+{
+	return 1;
+}
+
+static inline int ucc_geth_rx_queues(const struct ucc_geth_info *info)
+{
+	return 1;
+}
+
+static const struct ucc_geth_info ugeth_primary_info = {
 	.uf_info = {
-		    .bd_mem_part = MEM_PART_SYSTEM,
 		    .rtsm = UCC_FAST_SEND_IDLES_BETWEEN_FRAMES,
 		    .max_rx_buf_length = 1536,
 		    /* adjusted at startup if max-speed 1000 */
@@ -93,8 +115,6 @@ static struct ucc_geth_info ugeth_primary_info = {
 		    .tcrc = UCC_FAST_16_BIT_CRC,
 		    .synl = UCC_FAST_SYNC_LEN_NOT_USED,
 		    },
-	.numQueuesTx = 1,
-	.numQueuesRx = 1,
 	.extendedFilteringChainPointer = ((uint32_t) NULL),
 	.typeorlen = 3072 /*1536 */ ,
 	.nonBackToBackIfgPart1 = 0x40,
@@ -113,7 +133,6 @@ static struct ucc_geth_info ugeth_primary_info = {
 	.transmitFlowControl = 1,
 	.maxGroupAddrInHash = 4,
 	.maxIndAddrInHash = 4,
-	.prel = 7,
 	.maxFrameLength = 1518+16, /* Add extra bytes for VLANs etc. */
 	.minFrameLength = 64,
 	.maxD1Length = 1520+16, /* Add extra bytes for VLANs etc. */
@@ -160,8 +179,6 @@ static struct ucc_geth_info ugeth_primary_info = {
 	.riscRx = QE_RISC_ALLOCATION_RISC1_AND_RISC2,
 };
 
-static struct ucc_geth_info ugeth_info[8];
-
 #ifdef DEBUG
 static void mem_disp(u8 *addr, int size)
 {
@@ -561,7 +578,7 @@ static void dump_bds(struct ucc_geth_private *ugeth)
 	int i;
 	int length;
 
-	for (i = 0; i < ugeth->ug_info->numQueuesTx; i++) {
+	for (i = 0; i < ucc_geth_tx_queues(ugeth->ug_info); i++) {
 		if (ugeth->p_tx_bd_ring[i]) {
 			length =
 			    (ugeth->ug_info->bdRingLenTx[i] *
@@ -570,7 +587,7 @@ static void dump_bds(struct ucc_geth_private *ugeth)
 			mem_disp(ugeth->p_tx_bd_ring[i], length);
 		}
 	}
-	for (i = 0; i < ugeth->ug_info->numQueuesRx; i++) {
+	for (i = 0; i < ucc_geth_rx_queues(ugeth->ug_info); i++) {
 		if (ugeth->p_rx_bd_ring[i]) {
 			length =
 			    (ugeth->ug_info->bdRingLenRx[i] *
@@ -674,32 +691,12 @@ static void dump_regs(struct ucc_geth_private *ugeth)
 		in_be32(&ugeth->ug_regs->scam));
 
 	if (ugeth->p_thread_data_tx) {
-		int numThreadsTxNumerical;
-		switch (ugeth->ug_info->numThreadsTx) {
-		case UCC_GETH_NUM_OF_THREADS_1:
-			numThreadsTxNumerical = 1;
-			break;
-		case UCC_GETH_NUM_OF_THREADS_2:
-			numThreadsTxNumerical = 2;
-			break;
-		case UCC_GETH_NUM_OF_THREADS_4:
-			numThreadsTxNumerical = 4;
-			break;
-		case UCC_GETH_NUM_OF_THREADS_6:
-			numThreadsTxNumerical = 6;
-			break;
-		case UCC_GETH_NUM_OF_THREADS_8:
-			numThreadsTxNumerical = 8;
-			break;
-		default:
-			numThreadsTxNumerical = 0;
-			break;
-		}
+		int count = ucc_geth_thread_count(ugeth->ug_info->numThreadsTx);
 
 		pr_info("Thread data TXs:\n");
 		pr_info("Base address: 0x%08x\n",
 			(u32)ugeth->p_thread_data_tx);
-		for (i = 0; i < numThreadsTxNumerical; i++) {
+		for (i = 0; i < count; i++) {
 			pr_info("Thread data TX[%d]:\n", i);
 			pr_info("Base address: 0x%08x\n",
 				(u32)&ugeth->p_thread_data_tx[i]);
@@ -708,32 +705,12 @@ static void dump_regs(struct ucc_geth_private *ugeth)
 		}
 	}
 	if (ugeth->p_thread_data_rx) {
-		int numThreadsRxNumerical;
-		switch (ugeth->ug_info->numThreadsRx) {
-		case UCC_GETH_NUM_OF_THREADS_1:
-			numThreadsRxNumerical = 1;
-			break;
-		case UCC_GETH_NUM_OF_THREADS_2:
-			numThreadsRxNumerical = 2;
-			break;
-		case UCC_GETH_NUM_OF_THREADS_4:
-			numThreadsRxNumerical = 4;
-			break;
-		case UCC_GETH_NUM_OF_THREADS_6:
-			numThreadsRxNumerical = 6;
-			break;
-		case UCC_GETH_NUM_OF_THREADS_8:
-			numThreadsRxNumerical = 8;
-			break;
-		default:
-			numThreadsRxNumerical = 0;
-			break;
-		}
+		int count = ucc_geth_thread_count(ugeth->ug_info->numThreadsRx);
 
 		pr_info("Thread data RX:\n");
 		pr_info("Base address: 0x%08x\n",
 			(u32)ugeth->p_thread_data_rx);
-		for (i = 0; i < numThreadsRxNumerical; i++) {
+		for (i = 0; i < count; i++) {
 			pr_info("Thread data RX[%d]:\n", i);
 			pr_info("Base address: 0x%08x\n",
 				(u32)&ugeth->p_thread_data_rx[i]);
@@ -908,7 +885,7 @@ static void dump_regs(struct ucc_geth_private *ugeth)
 	if (ugeth->p_send_q_mem_reg) {
 		pr_info("Send Q memory registers:\n");
 		pr_info("Base address: 0x%08x\n", (u32)ugeth->p_send_q_mem_reg);
-		for (i = 0; i < ugeth->ug_info->numQueuesTx; i++) {
+		for (i = 0; i < ucc_geth_tx_queues(ugeth->ug_info); i++) {
 			pr_info("SQQD[%d]:\n", i);
 			pr_info("Base address: 0x%08x\n",
 				(u32)&ugeth->p_send_q_mem_reg->sqqd[i]);
@@ -940,7 +917,7 @@ static void dump_regs(struct ucc_geth_private *ugeth)
 		pr_info("RX IRQ coalescing tables:\n");
 		pr_info("Base address: 0x%08x\n",
 			(u32)ugeth->p_rx_irq_coalescing_tbl);
-		for (i = 0; i < ugeth->ug_info->numQueuesRx; i++) {
+		for (i = 0; i < ucc_geth_rx_queues(ugeth->ug_info); i++) {
 			pr_info("RX IRQ coalescing table entry[%d]:\n", i);
 			pr_info("Base address: 0x%08x\n",
 				(u32)&ugeth->p_rx_irq_coalescing_tbl->
@@ -962,7 +939,7 @@ static void dump_regs(struct ucc_geth_private *ugeth)
 	if (ugeth->p_rx_bd_qs_tbl) {
 		pr_info("RX BD QS tables:\n");
 		pr_info("Base address: 0x%08x\n", (u32)ugeth->p_rx_bd_qs_tbl);
-		for (i = 0; i < ugeth->ug_info->numQueuesRx; i++) {
+		for (i = 0; i < ucc_geth_rx_queues(ugeth->ug_info); i++) {
 			pr_info("RX BD QS table[%d]:\n", i);
 			pr_info("Base address: 0x%08x\n",
 				(u32)&ugeth->p_rx_bd_qs_tbl[i]);
@@ -1228,34 +1205,6 @@ static int init_mac_station_addr_regs(u8 address_byte_0,
 	return 0;
 }
 
-static int init_check_frame_length_mode(int length_check,
-					u32 __iomem *maccfg2_register)
-{
-	u32 value = 0;
-
-	value = in_be32(maccfg2_register);
-
-	if (length_check)
-		value |= MACCFG2_LC;
-	else
-		value &= ~MACCFG2_LC;
-
-	out_be32(maccfg2_register, value);
-	return 0;
-}
-
-static int init_preamble_length(u8 preamble_length,
-				u32 __iomem *maccfg2_register)
-{
-	if ((preamble_length < 3) || (preamble_length > 7))
-		return -EINVAL;
-
-	clrsetbits_be32(maccfg2_register, MACCFG2_PREL_MASK,
-			preamble_length << MACCFG2_PREL_SHIFT);
-
-	return 0;
-}
-
 static int init_rx_parameters(int reject_broadcast,
 			      int receive_short_frames,
 			      int promiscuous, u32 __iomem *upsmr_register)
@@ -1310,94 +1259,11 @@ static int init_min_frame_len(u16 min_frame_length,
 	return 0;
 }
 
-static int adjust_enet_interface(struct ucc_geth_private *ugeth)
+static bool phy_interface_mode_is_reduced(phy_interface_t interface)
 {
-	struct ucc_geth_info *ug_info;
-	struct ucc_geth __iomem *ug_regs;
-	struct ucc_fast __iomem *uf_regs;
-	int ret_val;
-	u32 upsmr, maccfg2;
-	u16 value;
-
-	ugeth_vdbg("%s: IN", __func__);
-
-	ug_info = ugeth->ug_info;
-	ug_regs = ugeth->ug_regs;
-	uf_regs = ugeth->uccf->uf_regs;
-
-	/*                    Set MACCFG2                    */
-	maccfg2 = in_be32(&ug_regs->maccfg2);
-	maccfg2 &= ~MACCFG2_INTERFACE_MODE_MASK;
-	if ((ugeth->max_speed == SPEED_10) ||
-	    (ugeth->max_speed == SPEED_100))
-		maccfg2 |= MACCFG2_INTERFACE_MODE_NIBBLE;
-	else if (ugeth->max_speed == SPEED_1000)
-		maccfg2 |= MACCFG2_INTERFACE_MODE_BYTE;
-	maccfg2 |= ug_info->padAndCrc;
-	out_be32(&ug_regs->maccfg2, maccfg2);
-
-	/*                    Set UPSMR                      */
-	upsmr = in_be32(&uf_regs->upsmr);
-	upsmr &= ~(UCC_GETH_UPSMR_RPM | UCC_GETH_UPSMR_R10M |
-		   UCC_GETH_UPSMR_TBIM | UCC_GETH_UPSMR_RMM);
-	if ((ugeth->phy_interface == PHY_INTERFACE_MODE_RMII) ||
-	    (ugeth->phy_interface == PHY_INTERFACE_MODE_RGMII) ||
-	    (ugeth->phy_interface == PHY_INTERFACE_MODE_RGMII_ID) ||
-	    (ugeth->phy_interface == PHY_INTERFACE_MODE_RGMII_RXID) ||
-	    (ugeth->phy_interface == PHY_INTERFACE_MODE_RGMII_TXID) ||
-	    (ugeth->phy_interface == PHY_INTERFACE_MODE_RTBI)) {
-		if (ugeth->phy_interface != PHY_INTERFACE_MODE_RMII)
-			upsmr |= UCC_GETH_UPSMR_RPM;
-		switch (ugeth->max_speed) {
-		case SPEED_10:
-			upsmr |= UCC_GETH_UPSMR_R10M;
-			/* FALLTHROUGH */
-		case SPEED_100:
-			if (ugeth->phy_interface != PHY_INTERFACE_MODE_RTBI)
-				upsmr |= UCC_GETH_UPSMR_RMM;
-		}
-	}
-	if ((ugeth->phy_interface == PHY_INTERFACE_MODE_TBI) ||
-	    (ugeth->phy_interface == PHY_INTERFACE_MODE_RTBI)) {
-		upsmr |= UCC_GETH_UPSMR_TBIM;
-	}
-	if ((ugeth->phy_interface == PHY_INTERFACE_MODE_SGMII))
-		upsmr |= UCC_GETH_UPSMR_SGMM;
-
-	out_be32(&uf_regs->upsmr, upsmr);
-
-	/* Disable autonegotiation in tbi mode, because by default it
-	comes up in autonegotiation mode. */
-	/* Note that this depends on proper setting in utbipar register. */
-	if ((ugeth->phy_interface == PHY_INTERFACE_MODE_TBI) ||
-	    (ugeth->phy_interface == PHY_INTERFACE_MODE_RTBI)) {
-		struct ucc_geth_info *ug_info = ugeth->ug_info;
-		struct phy_device *tbiphy;
-
-		if (!ug_info->tbi_node)
-			pr_warn("TBI mode requires that the device tree specify a tbi-handle\n");
-
-		tbiphy = of_phy_find_device(ug_info->tbi_node);
-		if (!tbiphy)
-			pr_warn("Could not get TBI device\n");
-
-		value = phy_read(tbiphy, ENET_TBI_MII_CR);
-		value &= ~0x1000;	/* Turn off autonegotiation */
-		phy_write(tbiphy, ENET_TBI_MII_CR, value);
-
-		put_device(&tbiphy->mdio.dev);
-	}
-
-	init_check_frame_length_mode(ug_info->lengthCheckRx, &ug_regs->maccfg2);
-
-	ret_val = init_preamble_length(ug_info->prel, &ug_regs->maccfg2);
-	if (ret_val != 0) {
-		if (netif_msg_probe(ugeth))
-			pr_err("Preamble length must be between 3 and 7 inclusive\n");
-		return ret_val;
-	}
-
-	return 0;
+	return phy_interface_mode_is_rgmii(interface) ||
+	       interface == PHY_INTERFACE_MODE_RMII ||
+	       interface == PHY_INTERFACE_MODE_RTBI;
 }
 
 static int ugeth_graceful_stop_tx(struct ucc_geth_private *ugeth)
@@ -1551,11 +1417,8 @@ static int ugeth_disable(struct ucc_geth_private *ugeth, enum comm_dir mode)
 
 static void ugeth_quiesce(struct ucc_geth_private *ugeth)
 {
-	/* Prevent any further xmits, plus detach the device. */
-	netif_device_detach(ugeth->ndev);
-
-	/* Wait for any current xmits to finish. */
-	netif_tx_disable(ugeth->ndev);
+	/* Prevent any further xmits */
+	netif_tx_stop_all_queues(ugeth->ndev);
 
 	/* Disable the interrupt to avoid NAPI rescheduling. */
 	disable_irq(ugeth->ug_info->uf_info.irq);
@@ -1568,108 +1431,10 @@ static void ugeth_activate(struct ucc_geth_private *ugeth)
 {
 	napi_enable(&ugeth->napi);
 	enable_irq(ugeth->ug_info->uf_info.irq);
-	netif_device_attach(ugeth->ndev);
-}
-
-/* Called every time the controller might need to be made
- * aware of new link state.  The PHY code conveys this
- * information through variables in the ugeth structure, and this
- * function converts those variables into the appropriate
- * register values, and can bring down the device if needed.
- */
-
-static void adjust_link(struct net_device *dev)
-{
-	struct ucc_geth_private *ugeth = netdev_priv(dev);
-	struct ucc_geth __iomem *ug_regs;
-	struct ucc_fast __iomem *uf_regs;
-	struct phy_device *phydev = ugeth->phydev;
-	int new_state = 0;
 
-	ug_regs = ugeth->ug_regs;
-	uf_regs = ugeth->uccf->uf_regs;
-
-	if (phydev->link) {
-		u32 tempval = in_be32(&ug_regs->maccfg2);
-		u32 upsmr = in_be32(&uf_regs->upsmr);
-		/* Now we make sure that we can be in full duplex mode.
-		 * If not, we operate in half-duplex mode. */
-		if (phydev->duplex != ugeth->oldduplex) {
-			new_state = 1;
-			if (!(phydev->duplex))
-				tempval &= ~(MACCFG2_FDX);
-			else
-				tempval |= MACCFG2_FDX;
-			ugeth->oldduplex = phydev->duplex;
-		}
-
-		if (phydev->speed != ugeth->oldspeed) {
-			new_state = 1;
-			switch (phydev->speed) {
-			case SPEED_1000:
-				tempval = ((tempval &
-					    ~(MACCFG2_INTERFACE_MODE_MASK)) |
-					    MACCFG2_INTERFACE_MODE_BYTE);
-				break;
-			case SPEED_100:
-			case SPEED_10:
-				tempval = ((tempval &
-					    ~(MACCFG2_INTERFACE_MODE_MASK)) |
-					    MACCFG2_INTERFACE_MODE_NIBBLE);
-				/* if reduced mode, re-set UPSMR.R10M */
-				if ((ugeth->phy_interface == PHY_INTERFACE_MODE_RMII) ||
-				    (ugeth->phy_interface == PHY_INTERFACE_MODE_RGMII) ||
-				    (ugeth->phy_interface == PHY_INTERFACE_MODE_RGMII_ID) ||
-				    (ugeth->phy_interface == PHY_INTERFACE_MODE_RGMII_RXID) ||
-				    (ugeth->phy_interface == PHY_INTERFACE_MODE_RGMII_TXID) ||
-				    (ugeth->phy_interface == PHY_INTERFACE_MODE_RTBI)) {
-					if (phydev->speed == SPEED_10)
-						upsmr |= UCC_GETH_UPSMR_R10M;
-					else
-						upsmr &= ~UCC_GETH_UPSMR_R10M;
-				}
-				break;
-			default:
-				if (netif_msg_link(ugeth))
-					pr_warn(
-						"%s: Ack!  Speed (%d) is not 10/100/1000!",
-						dev->name, phydev->speed);
-				break;
-			}
-			ugeth->oldspeed = phydev->speed;
-		}
-
-		if (!ugeth->oldlink) {
-			new_state = 1;
-			ugeth->oldlink = 1;
-		}
-
-		if (new_state) {
-			/*
-			 * To change the MAC configuration we need to disable
-			 * the controller. To do so, we have to either grab
-			 * ugeth->lock, which is a bad idea since 'graceful
-			 * stop' commands might take quite a while, or we can
-			 * quiesce driver's activity.
-			 */
-			ugeth_quiesce(ugeth);
-			ugeth_disable(ugeth, COMM_DIR_RX_AND_TX);
-
-			out_be32(&ug_regs->maccfg2, tempval);
-			out_be32(&uf_regs->upsmr, upsmr);
-
-			ugeth_enable(ugeth, COMM_DIR_RX_AND_TX);
-			ugeth_activate(ugeth);
-		}
-	} else if (ugeth->oldlink) {
-			new_state = 1;
-			ugeth->oldlink = 0;
-			ugeth->oldspeed = 0;
-			ugeth->oldduplex = -1;
-	}
-
-	if (new_state && netif_msg_link(ugeth))
-		phy_print_status(phydev);
+	/* allow to xmit again  */
+	netif_tx_wake_all_queues(ugeth->ndev);
+	netdev_watchdog_up(ugeth->ndev);
 }
 
 /* Initialize TBI PHY interface for communicating with the
@@ -1687,8 +1452,7 @@ static void uec_configure_serdes(struct net_device *dev)
 	struct phy_device *tbiphy;
 
 	if (!ug_info->tbi_node) {
-		dev_warn(&dev->dev, "SGMII mode requires that the device "
-			"tree specify a tbi-handle\n");
+		dev_warn(&dev->dev, "SGMII mode requires that the device tree specify a tbi-handle\n");
 		return;
 	}
 
@@ -1719,44 +1483,145 @@ static void uec_configure_serdes(struct net_device *dev)
 	put_device(&tbiphy->mdio.dev);
 }
 
-/* Configure the PHY for dev.
- * returns 0 if success.  -1 if failure
- */
-static int init_phy(struct net_device *dev)
+static void ugeth_mac_link_up(struct phylink_config *config, struct phy_device *phy,
+			      unsigned int mode, phy_interface_t interface,
+			      int speed, int duplex, bool tx_pause, bool rx_pause)
 {
-	struct ucc_geth_private *priv = netdev_priv(dev);
-	struct ucc_geth_info *ug_info = priv->ug_info;
-	struct phy_device *phydev;
+	struct net_device *ndev = to_net_dev(config->dev);
+	struct ucc_geth_private *ugeth = netdev_priv(ndev);
+	struct ucc_geth_info *ug_info = ugeth->ug_info;
+	struct ucc_geth __iomem *ug_regs = ugeth->ug_regs;
+	struct ucc_fast __iomem *uf_regs = ugeth->uccf->uf_regs;
+	u32 old_maccfg2, maccfg2 = in_be32(&ug_regs->maccfg2);
+	u32 old_upsmr, upsmr = in_be32(&uf_regs->upsmr);
 
-	priv->oldlink = 0;
-	priv->oldspeed = 0;
-	priv->oldduplex = -1;
+	old_maccfg2 = maccfg2;
+	old_upsmr = upsmr;
 
-	phydev = of_phy_connect(dev, ug_info->phy_node, &adjust_link, 0,
-				priv->phy_interface);
-	if (!phydev) {
-		dev_err(&dev->dev, "Could not attach to PHY\n");
-		return -ENODEV;
+	/* No length check */
+	maccfg2 &= ~MACCFG2_LC;
+	maccfg2 &= ~MACCFG2_INTERFACE_MODE_MASK;
+	upsmr &= ~(UCC_GETH_UPSMR_RPM | UCC_GETH_UPSMR_R10M |
+		   UCC_GETH_UPSMR_TBIM | UCC_GETH_UPSMR_RMM);
+
+	if (speed == SPEED_10 || speed == SPEED_100)
+		maccfg2 |= MACCFG2_INTERFACE_MODE_NIBBLE;
+	else if (speed == SPEED_1000)
+		maccfg2 |= MACCFG2_INTERFACE_MODE_BYTE;
+
+	maccfg2 |= ug_info->padAndCrc;
+
+	if (phy_interface_mode_is_reduced(interface)) {
+
+		if (interface != PHY_INTERFACE_MODE_RMII)
+			upsmr |= UCC_GETH_UPSMR_RPM;
+
+		switch (speed) {
+		case SPEED_10:
+			upsmr |= UCC_GETH_UPSMR_R10M;
+			fallthrough;
+		case SPEED_100:
+			if (interface != PHY_INTERFACE_MODE_RTBI)
+				upsmr |= UCC_GETH_UPSMR_RMM;
+		}
+	}
+
+	if (interface == PHY_INTERFACE_MODE_TBI ||
+	    interface == PHY_INTERFACE_MODE_RTBI)
+		upsmr |= UCC_GETH_UPSMR_TBIM;
+
+	if (interface == PHY_INTERFACE_MODE_SGMII)
+		upsmr |= UCC_GETH_UPSMR_SGMM;
+
+	if (duplex == DUPLEX_HALF)
+		maccfg2 &= ~(MACCFG2_FDX);
+	else
+		maccfg2 |= MACCFG2_FDX;
+
+	if (maccfg2 != old_maccfg2 || upsmr != old_upsmr) {
+		/*
+		 * To change the MAC configuration we need to disable
+		 * the controller. To do so, we have to either grab
+		 * ugeth->lock, which is a bad idea since 'graceful
+		 * stop' commands might take quite a while, or we can
+		 * quiesce driver's activity.
+		 */
+		ugeth_quiesce(ugeth);
+		ugeth_disable(ugeth, COMM_DIR_RX_AND_TX);
+
+		out_be32(&ug_regs->maccfg2, maccfg2);
+		out_be32(&uf_regs->upsmr, upsmr);
+
+		ugeth_enable(ugeth, COMM_DIR_RX_AND_TX);
+		ugeth_activate(ugeth);
 	}
 
-	if (priv->phy_interface == PHY_INTERFACE_MODE_SGMII)
-		uec_configure_serdes(dev);
+	if (interface == PHY_INTERFACE_MODE_SGMII)
+		uec_configure_serdes(ndev);
 
-	phydev->supported &= (SUPPORTED_MII |
-			      SUPPORTED_Autoneg |
-			      ADVERTISED_10baseT_Half |
-			      ADVERTISED_10baseT_Full |
-			      ADVERTISED_100baseT_Half |
-			      ADVERTISED_100baseT_Full);
+	if (!phylink_autoneg_inband(mode)) {
+		ug_info->aufc = 0;
+		ug_info->receiveFlowControl = rx_pause;
+		ug_info->transmitFlowControl = tx_pause;
 
-	if (priv->max_speed == SPEED_1000)
-		phydev->supported |= ADVERTISED_1000baseT_Full;
+		init_flow_control_params(ug_info->aufc,
+					 ug_info->receiveFlowControl,
+					 ug_info->transmitFlowControl,
+					 ug_info->pausePeriod,
+					 ug_info->extensionField,
+					 &ugeth->uccf->uf_regs->upsmr,
+					 &ugeth->ug_regs->uempr,
+					 &ugeth->ug_regs->maccfg1);
+	}
 
-	phydev->advertising = phydev->supported;
+	ugeth_enable(ugeth, COMM_DIR_RX_AND_TX);
+}
 
-	priv->phydev = phydev;
+static void ugeth_mac_link_down(struct phylink_config *config,
+				unsigned int mode, phy_interface_t interface)
+{
+	struct net_device *ndev = to_net_dev(config->dev);
+	struct ucc_geth_private *ugeth = netdev_priv(ndev);
 
-	return 0;
+	ugeth_disable(ugeth, COMM_DIR_RX_AND_TX);
+}
+
+static void ugeth_mac_config(struct phylink_config *config, unsigned int mode,
+			     const struct phylink_link_state *state)
+{
+	struct net_device *ndev = to_net_dev(config->dev);
+	struct ucc_geth_private *ugeth = netdev_priv(ndev);
+	struct ucc_geth_info *ug_info = ugeth->ug_info;
+	u16 value;
+
+	if (state->interface == PHY_INTERFACE_MODE_TBI ||
+	    state->interface == PHY_INTERFACE_MODE_RTBI) {
+		struct phy_device *tbiphy;
+
+		if (!ug_info->tbi_node)
+			pr_warn("TBI mode requires that the device tree specify a tbi-handle\n");
+
+		tbiphy = of_phy_find_device(ug_info->tbi_node);
+		if (!tbiphy)
+			pr_warn("Could not get TBI device\n");
+
+		value = phy_read(tbiphy, ENET_TBI_MII_CR);
+		value &= ~0x1000;	/* Turn off autonegotiation */
+		phy_write(tbiphy, ENET_TBI_MII_CR, value);
+
+		put_device(&tbiphy->mdio.dev);
+	}
+
+	if (phylink_autoneg_inband(mode)) {
+		ug_info->aufc = 1;
+
+		init_flow_control_params(ug_info->aufc, 1, 1,
+					 ug_info->pausePeriod,
+					 ug_info->extensionField,
+					 &ugeth->uccf->uf_regs->upsmr,
+					 &ugeth->ug_regs->uempr,
+					 &ugeth->ug_regs->maccfg1);
+	}
 }
 
 static void ugeth_dump_regs(struct ucc_geth_private *ugeth)
@@ -1848,7 +1713,7 @@ static void ucc_geth_free_rx(struct ucc_geth_private *ugeth)
 	ug_info = ugeth->ug_info;
 	uf_info = &ug_info->uf_info;
 
-	for (i = 0; i < ugeth->ug_info->numQueuesRx; i++) {
+	for (i = 0; i < ucc_geth_rx_queues(ugeth->ug_info); i++) {
 		if (ugeth->p_rx_bd_ring[i]) {
 			/* Return existing data buffers in ring */
 			bd = ugeth->p_rx_bd_ring[i];
@@ -1869,12 +1734,7 @@ static void ucc_geth_free_rx(struct ucc_geth_private *ugeth)
 
 			kfree(ugeth->rx_skbuff[i]);
 
-			if (ugeth->ug_info->uf_info.bd_mem_part ==
-			    MEM_PART_SYSTEM)
-				kfree((void *)ugeth->rx_bd_ring_offset[i]);
-			else if (ugeth->ug_info->uf_info.bd_mem_part ==
-				 MEM_PART_MURAM)
-				qe_muram_free(ugeth->rx_bd_ring_offset[i]);
+			kfree(ugeth->p_rx_bd_ring[i]);
 			ugeth->p_rx_bd_ring[i] = NULL;
 		}
 	}
@@ -1888,10 +1748,12 @@ static void ucc_geth_free_tx(struct ucc_geth_private *ugeth)
 	u16 i, j;
 	u8 __iomem *bd;
 
+	netdev_reset_queue(ugeth->ndev);
+
 	ug_info = ugeth->ug_info;
 	uf_info = &ug_info->uf_info;
 
-	for (i = 0; i < ugeth->ug_info->numQueuesTx; i++) {
+	for (i = 0; i < ucc_geth_tx_queues(ugeth->ug_info); i++) {
 		bd = ugeth->p_tx_bd_ring[i];
 		if (!bd)
 			continue;
@@ -1909,15 +1771,8 @@ static void ucc_geth_free_tx(struct ucc_geth_private *ugeth)
 
 		kfree(ugeth->tx_skbuff[i]);
 
-		if (ugeth->p_tx_bd_ring[i]) {
-			if (ugeth->ug_info->uf_info.bd_mem_part ==
-			    MEM_PART_SYSTEM)
-				kfree((void *)ugeth->tx_bd_ring_offset[i]);
-			else if (ugeth->ug_info->uf_info.bd_mem_part ==
-				 MEM_PART_MURAM)
-				qe_muram_free(ugeth->tx_bd_ring_offset[i]);
-			ugeth->p_tx_bd_ring[i] = NULL;
-		}
+		kfree(ugeth->p_tx_bd_ring[i]);
+		ugeth->p_tx_bd_ring[i] = NULL;
 	}
 
 }
@@ -1932,50 +1787,39 @@ static void ucc_geth_memclean(struct ucc_geth_private *ugeth)
 		ugeth->uccf = NULL;
 	}
 
-	if (ugeth->p_thread_data_tx) {
-		qe_muram_free(ugeth->thread_dat_tx_offset);
-		ugeth->p_thread_data_tx = NULL;
-	}
-	if (ugeth->p_thread_data_rx) {
-		qe_muram_free(ugeth->thread_dat_rx_offset);
-		ugeth->p_thread_data_rx = NULL;
-	}
-	if (ugeth->p_exf_glbl_param) {
-		qe_muram_free(ugeth->exf_glbl_param_offset);
-		ugeth->p_exf_glbl_param = NULL;
-	}
-	if (ugeth->p_rx_glbl_pram) {
-		qe_muram_free(ugeth->rx_glbl_pram_offset);
-		ugeth->p_rx_glbl_pram = NULL;
-	}
-	if (ugeth->p_tx_glbl_pram) {
-		qe_muram_free(ugeth->tx_glbl_pram_offset);
-		ugeth->p_tx_glbl_pram = NULL;
-	}
-	if (ugeth->p_send_q_mem_reg) {
-		qe_muram_free(ugeth->send_q_mem_reg_offset);
-		ugeth->p_send_q_mem_reg = NULL;
-	}
-	if (ugeth->p_scheduler) {
-		qe_muram_free(ugeth->scheduler_offset);
-		ugeth->p_scheduler = NULL;
-	}
-	if (ugeth->p_tx_fw_statistics_pram) {
-		qe_muram_free(ugeth->tx_fw_statistics_pram_offset);
-		ugeth->p_tx_fw_statistics_pram = NULL;
-	}
-	if (ugeth->p_rx_fw_statistics_pram) {
-		qe_muram_free(ugeth->rx_fw_statistics_pram_offset);
-		ugeth->p_rx_fw_statistics_pram = NULL;
-	}
-	if (ugeth->p_rx_irq_coalescing_tbl) {
-		qe_muram_free(ugeth->rx_irq_coalescing_tbl_offset);
-		ugeth->p_rx_irq_coalescing_tbl = NULL;
-	}
-	if (ugeth->p_rx_bd_qs_tbl) {
-		qe_muram_free(ugeth->rx_bd_qs_tbl_offset);
-		ugeth->p_rx_bd_qs_tbl = NULL;
-	}
+	qe_muram_free_addr(ugeth->p_thread_data_tx);
+	ugeth->p_thread_data_tx = NULL;
+
+	qe_muram_free_addr(ugeth->p_thread_data_rx);
+	ugeth->p_thread_data_rx = NULL;
+
+	qe_muram_free_addr(ugeth->p_exf_glbl_param);
+	ugeth->p_exf_glbl_param = NULL;
+
+	qe_muram_free_addr(ugeth->p_rx_glbl_pram);
+	ugeth->p_rx_glbl_pram = NULL;
+
+	qe_muram_free_addr(ugeth->p_tx_glbl_pram);
+	ugeth->p_tx_glbl_pram = NULL;
+
+	qe_muram_free_addr(ugeth->p_send_q_mem_reg);
+	ugeth->p_send_q_mem_reg = NULL;
+
+	qe_muram_free_addr(ugeth->p_scheduler);
+	ugeth->p_scheduler = NULL;
+
+	qe_muram_free_addr(ugeth->p_tx_fw_statistics_pram);
+	ugeth->p_tx_fw_statistics_pram = NULL;
+
+	qe_muram_free_addr(ugeth->p_rx_fw_statistics_pram);
+	ugeth->p_rx_fw_statistics_pram = NULL;
+
+	qe_muram_free_addr(ugeth->p_rx_irq_coalescing_tbl);
+	ugeth->p_rx_irq_coalescing_tbl = NULL;
+
+	qe_muram_free_addr(ugeth->p_rx_bd_qs_tbl);
+	ugeth->p_rx_bd_qs_tbl = NULL;
+
 	if (ugeth->p_init_enet_param_shadow) {
 		return_init_enet_entries(ugeth,
 					 &(ugeth->p_init_enet_param_shadow->
@@ -2049,7 +1893,6 @@ static void ucc_geth_set_multi(struct net_device *dev)
 static void ucc_geth_stop(struct ucc_geth_private *ugeth)
 {
 	struct ucc_geth __iomem *ug_regs = ugeth->ug_regs;
-	struct phy_device *phydev = ugeth->phydev;
 
 	ugeth_vdbg("%s: IN", __func__);
 
@@ -2058,7 +1901,7 @@ static void ucc_geth_stop(struct ucc_geth_private *ugeth)
 	 * Must be done before disabling the controller
 	 * or deadlock may happen.
 	 */
-	phy_stop(phydev);
+	phylink_stop(ugeth->phylink);
 
 	/* Disable the controller */
 	ugeth_disable(ugeth, COMM_DIR_RX_AND_TX);
@@ -2084,15 +1927,8 @@ static int ucc_struct_init(struct ucc_geth_private *ugeth)
 	ug_info = ugeth->ug_info;
 	uf_info = &ug_info->uf_info;
 
-	if (!((uf_info->bd_mem_part == MEM_PART_SYSTEM) ||
-	      (uf_info->bd_mem_part == MEM_PART_MURAM))) {
-		if (netif_msg_probe(ugeth))
-			pr_err("Bad memory partition value\n");
-		return -EINVAL;
-	}
-
 	/* Rx BD lengths */
-	for (i = 0; i < ug_info->numQueuesRx; i++) {
+	for (i = 0; i < ucc_geth_rx_queues(ug_info); i++) {
 		if ((ug_info->bdRingLenRx[i] < UCC_GETH_RX_BD_RING_SIZE_MIN) ||
 		    (ug_info->bdRingLenRx[i] %
 		     UCC_GETH_RX_BD_RING_SIZE_ALIGNMENT)) {
@@ -2103,7 +1939,7 @@ static int ucc_struct_init(struct ucc_geth_private *ugeth)
 	}
 
 	/* Tx BD lengths */
-	for (i = 0; i < ug_info->numQueuesTx; i++) {
+	for (i = 0; i < ucc_geth_tx_queues(ug_info); i++) {
 		if (ug_info->bdRingLenTx[i] < UCC_GETH_TX_BD_RING_SIZE_MIN) {
 			if (netif_msg_probe(ugeth))
 				pr_err("Tx BD ring length must be no smaller than 2\n");
@@ -2120,14 +1956,14 @@ static int ucc_struct_init(struct ucc_geth_private *ugeth)
 	}
 
 	/* num Tx queues */
-	if (ug_info->numQueuesTx > NUM_TX_QUEUES) {
+	if (ucc_geth_tx_queues(ug_info) > NUM_TX_QUEUES) {
 		if (netif_msg_probe(ugeth))
 			pr_err("number of tx queues too large\n");
 		return -EINVAL;
 	}
 
 	/* num Rx queues */
-	if (ug_info->numQueuesRx > NUM_RX_QUEUES) {
+	if (ucc_geth_rx_queues(ug_info) > NUM_RX_QUEUES) {
 		if (netif_msg_probe(ugeth))
 			pr_err("number of rx queues too large\n");
 		return -EINVAL;
@@ -2135,7 +1971,7 @@ static int ucc_struct_init(struct ucc_geth_private *ugeth)
 
 	/* l2qt */
 	for (i = 0; i < UCC_GETH_VLAN_PRIORITY_MAX; i++) {
-		if (ug_info->l2qt[i] >= ug_info->numQueuesRx) {
+		if (ug_info->l2qt[i] >= ucc_geth_rx_queues(ug_info)) {
 			if (netif_msg_probe(ugeth))
 				pr_err("VLAN priority table entry must not be larger than number of Rx queues\n");
 			return -EINVAL;
@@ -2144,7 +1980,7 @@ static int ucc_struct_init(struct ucc_geth_private *ugeth)
 
 	/* l3qt */
 	for (i = 0; i < UCC_GETH_IP_PRIORITY_MAX; i++) {
-		if (ug_info->l3qt[i] >= ug_info->numQueuesRx) {
+		if (ug_info->l3qt[i] >= ucc_geth_rx_queues(ug_info)) {
 			if (netif_msg_probe(ugeth))
 				pr_err("IP priority table entry must not be larger than number of Rx queues\n");
 			return -EINVAL;
@@ -2167,10 +2003,10 @@ static int ucc_struct_init(struct ucc_geth_private *ugeth)
 
 	/* Generate uccm_mask for receive */
 	uf_info->uccm_mask = ug_info->eventRegMask & UCCE_OTHER;/* Errors */
-	for (i = 0; i < ug_info->numQueuesRx; i++)
+	for (i = 0; i < ucc_geth_rx_queues(ug_info); i++)
 		uf_info->uccm_mask |= (UCC_GETH_UCCE_RXF0 << i);
 
-	for (i = 0; i < ug_info->numQueuesTx; i++)
+	for (i = 0; i < ucc_geth_tx_queues(ug_info); i++)
 		uf_info->uccm_mask |= (UCC_GETH_UCCE_TXB0 << i);
 	/* Initialize the general fast UCC block. */
 	if (ucc_fast_init(uf_info, &ugeth->uccf)) {
@@ -2209,53 +2045,32 @@ static int ucc_geth_alloc_tx(struct ucc_geth_private *ugeth)
 	uf_info = &ug_info->uf_info;
 
 	/* Allocate Tx bds */
-	for (j = 0; j < ug_info->numQueuesTx; j++) {
-		/* Allocate in multiple of
-		   UCC_GETH_TX_BD_RING_SIZE_MEMORY_ALIGNMENT,
-		   according to spec */
-		length = ((ug_info->bdRingLenTx[j] * sizeof(struct qe_bd))
-			  / UCC_GETH_TX_BD_RING_SIZE_MEMORY_ALIGNMENT)
-		    * UCC_GETH_TX_BD_RING_SIZE_MEMORY_ALIGNMENT;
-		if ((ug_info->bdRingLenTx[j] * sizeof(struct qe_bd)) %
-		    UCC_GETH_TX_BD_RING_SIZE_MEMORY_ALIGNMENT)
-			length += UCC_GETH_TX_BD_RING_SIZE_MEMORY_ALIGNMENT;
-		if (uf_info->bd_mem_part == MEM_PART_SYSTEM) {
-			u32 align = 4;
-			if (UCC_GETH_TX_BD_RING_ALIGNMENT > 4)
-				align = UCC_GETH_TX_BD_RING_ALIGNMENT;
-			ugeth->tx_bd_ring_offset[j] =
-				(u32) kmalloc((u32) (length + align), GFP_KERNEL);
-
-			if (ugeth->tx_bd_ring_offset[j] != 0)
-				ugeth->p_tx_bd_ring[j] =
-					(u8 __iomem *)((ugeth->tx_bd_ring_offset[j] +
-					align) & ~(align - 1));
-		} else if (uf_info->bd_mem_part == MEM_PART_MURAM) {
-			ugeth->tx_bd_ring_offset[j] =
-			    qe_muram_alloc(length,
-					   UCC_GETH_TX_BD_RING_ALIGNMENT);
-			if (!IS_ERR_VALUE(ugeth->tx_bd_ring_offset[j]))
-				ugeth->p_tx_bd_ring[j] =
-				    (u8 __iomem *) qe_muram_addr(ugeth->
-							 tx_bd_ring_offset[j]);
-		}
+	for (j = 0; j < ucc_geth_tx_queues(ug_info); j++) {
+		u32 align = max(UCC_GETH_TX_BD_RING_ALIGNMENT,
+				UCC_GETH_TX_BD_RING_SIZE_MEMORY_ALIGNMENT);
+		u32 alloc;
+
+		length = ug_info->bdRingLenTx[j] * sizeof(struct qe_bd);
+		alloc = round_up(length, align);
+		alloc = roundup_pow_of_two(alloc);
+
+		ugeth->p_tx_bd_ring[j] = kmalloc(alloc, GFP_KERNEL);
+
 		if (!ugeth->p_tx_bd_ring[j]) {
 			if (netif_msg_ifup(ugeth))
 				pr_err("Can not allocate memory for Tx bd rings\n");
 			return -ENOMEM;
 		}
 		/* Zero unused end of bd ring, according to spec */
-		memset_io((void __iomem *)(ugeth->p_tx_bd_ring[j] +
-		       ug_info->bdRingLenTx[j] * sizeof(struct qe_bd)), 0,
-		       length - ug_info->bdRingLenTx[j] * sizeof(struct qe_bd));
+		memset(ugeth->p_tx_bd_ring[j] + length, 0, alloc - length);
 	}
 
 	/* Init Tx bds */
-	for (j = 0; j < ug_info->numQueuesTx; j++) {
+	for (j = 0; j < ucc_geth_tx_queues(ug_info); j++) {
 		/* Setup the skbuff rings */
-		ugeth->tx_skbuff[j] = kmalloc(sizeof(struct sk_buff *) *
-					      ugeth->ug_info->bdRingLenTx[j],
-					      GFP_KERNEL);
+		ugeth->tx_skbuff[j] =
+			kcalloc(ugeth->ug_info->bdRingLenTx[j],
+				sizeof(struct sk_buff *), GFP_KERNEL);
 
 		if (ugeth->tx_skbuff[j] == NULL) {
 			if (netif_msg_ifup(ugeth))
@@ -2263,9 +2078,6 @@ static int ucc_geth_alloc_tx(struct ucc_geth_private *ugeth)
 			return -ENOMEM;
 		}
 
-		for (i = 0; i < ugeth->ug_info->bdRingLenTx[j]; i++)
-			ugeth->tx_skbuff[j][i] = NULL;
-
 		ugeth->skb_curtx[j] = ugeth->skb_dirtytx[j] = 0;
 		bd = ugeth->confBd[j] = ugeth->txBd[j] = ugeth->p_tx_bd_ring[j];
 		for (i = 0; i < ug_info->bdRingLenTx[j]; i++) {
@@ -2295,27 +2107,15 @@ static int ucc_geth_alloc_rx(struct ucc_geth_private *ugeth)
 	uf_info = &ug_info->uf_info;
 
 	/* Allocate Rx bds */
-	for (j = 0; j < ug_info->numQueuesRx; j++) {
+	for (j = 0; j < ucc_geth_rx_queues(ug_info); j++) {
+		u32 align = UCC_GETH_RX_BD_RING_ALIGNMENT;
+		u32 alloc;
+
 		length = ug_info->bdRingLenRx[j] * sizeof(struct qe_bd);
-		if (uf_info->bd_mem_part == MEM_PART_SYSTEM) {
-			u32 align = 4;
-			if (UCC_GETH_RX_BD_RING_ALIGNMENT > 4)
-				align = UCC_GETH_RX_BD_RING_ALIGNMENT;
-			ugeth->rx_bd_ring_offset[j] =
-				(u32) kmalloc((u32) (length + align), GFP_KERNEL);
-			if (ugeth->rx_bd_ring_offset[j] != 0)
-				ugeth->p_rx_bd_ring[j] =
-					(u8 __iomem *)((ugeth->rx_bd_ring_offset[j] +
-					align) & ~(align - 1));
-		} else if (uf_info->bd_mem_part == MEM_PART_MURAM) {
-			ugeth->rx_bd_ring_offset[j] =
-			    qe_muram_alloc(length,
-					   UCC_GETH_RX_BD_RING_ALIGNMENT);
-			if (!IS_ERR_VALUE(ugeth->rx_bd_ring_offset[j]))
-				ugeth->p_rx_bd_ring[j] =
-				    (u8 __iomem *) qe_muram_addr(ugeth->
-							 rx_bd_ring_offset[j]);
-		}
+		alloc = round_up(length, align);
+		alloc = roundup_pow_of_two(alloc);
+
+		ugeth->p_rx_bd_ring[j] = kmalloc(alloc, GFP_KERNEL);
 		if (!ugeth->p_rx_bd_ring[j]) {
 			if (netif_msg_ifup(ugeth))
 				pr_err("Can not allocate memory for Rx bd rings\n");
@@ -2324,11 +2124,11 @@ static int ucc_geth_alloc_rx(struct ucc_geth_private *ugeth)
 	}
 
 	/* Init Rx bds */
-	for (j = 0; j < ug_info->numQueuesRx; j++) {
+	for (j = 0; j < ucc_geth_rx_queues(ug_info); j++) {
 		/* Setup the skbuff rings */
-		ugeth->rx_skbuff[j] = kmalloc(sizeof(struct sk_buff *) *
-					      ugeth->ug_info->bdRingLenRx[j],
-					      GFP_KERNEL);
+		ugeth->rx_skbuff[j] =
+			kcalloc(ugeth->ug_info->bdRingLenRx[j],
+				sizeof(struct sk_buff *), GFP_KERNEL);
 
 		if (ugeth->rx_skbuff[j] == NULL) {
 			if (netif_msg_ifup(ugeth))
@@ -2336,9 +2136,6 @@ static int ucc_geth_alloc_rx(struct ucc_geth_private *ugeth)
 			return -ENOMEM;
 		}
 
-		for (i = 0; i < ugeth->ug_info->bdRingLenRx[j]; i++)
-			ugeth->rx_skbuff[j][i] = NULL;
-
 		ugeth->skb_currx[j] = 0;
 		bd = ugeth->rxBd[j] = ugeth->p_rx_bd_ring[j];
 		for (i = 0; i < ug_info->bdRingLenRx[j]; i++) {
@@ -2370,10 +2167,10 @@ static int ucc_geth_startup(struct ucc_geth_private *ugeth)
 	u32 init_enet_pram_offset, cecr_subblock, command;
 	u32 ifstat, i, j, size, l2qt, l3qt;
 	u16 temoder = UCC_GETH_TEMODER_INIT;
-	u16 test;
 	u8 function_code = 0;
 	u8 __iomem *endOfRing;
 	u8 numThreadsRxNumerical, numThreadsTxNumerical;
+	s32 rx_glbl_pram_offset, tx_glbl_pram_offset;
 
 	ugeth_vdbg("%s: IN", __func__);
 	uccf = ugeth->uccf;
@@ -2382,45 +2179,15 @@ static int ucc_geth_startup(struct ucc_geth_private *ugeth)
 	uf_regs = uccf->uf_regs;
 	ug_regs = ugeth->ug_regs;
 
-	switch (ug_info->numThreadsRx) {
-	case UCC_GETH_NUM_OF_THREADS_1:
-		numThreadsRxNumerical = 1;
-		break;
-	case UCC_GETH_NUM_OF_THREADS_2:
-		numThreadsRxNumerical = 2;
-		break;
-	case UCC_GETH_NUM_OF_THREADS_4:
-		numThreadsRxNumerical = 4;
-		break;
-	case UCC_GETH_NUM_OF_THREADS_6:
-		numThreadsRxNumerical = 6;
-		break;
-	case UCC_GETH_NUM_OF_THREADS_8:
-		numThreadsRxNumerical = 8;
-		break;
-	default:
+	numThreadsRxNumerical = ucc_geth_thread_count(ug_info->numThreadsRx);
+	if (!numThreadsRxNumerical) {
 		if (netif_msg_ifup(ugeth))
 			pr_err("Bad number of Rx threads value\n");
 		return -EINVAL;
 	}
 
-	switch (ug_info->numThreadsTx) {
-	case UCC_GETH_NUM_OF_THREADS_1:
-		numThreadsTxNumerical = 1;
-		break;
-	case UCC_GETH_NUM_OF_THREADS_2:
-		numThreadsTxNumerical = 2;
-		break;
-	case UCC_GETH_NUM_OF_THREADS_4:
-		numThreadsTxNumerical = 4;
-		break;
-	case UCC_GETH_NUM_OF_THREADS_6:
-		numThreadsTxNumerical = 6;
-		break;
-	case UCC_GETH_NUM_OF_THREADS_8:
-		numThreadsTxNumerical = 8;
-		break;
-	default:
+	numThreadsTxNumerical = ucc_geth_thread_count(ug_info->numThreadsTx);
+	if (!numThreadsTxNumerical) {
 		if (netif_msg_ifup(ugeth))
 			pr_err("Bad number of Tx threads value\n");
 		return -EINVAL;
@@ -2518,20 +2285,15 @@ static int ucc_geth_startup(struct ucc_geth_private *ugeth)
 	 */
 	/* Tx global PRAM */
 	/* Allocate global tx parameter RAM page */
-	ugeth->tx_glbl_pram_offset =
+	tx_glbl_pram_offset =
 	    qe_muram_alloc(sizeof(struct ucc_geth_tx_global_pram),
 			   UCC_GETH_TX_GLOBAL_PRAM_ALIGNMENT);
-	if (IS_ERR_VALUE(ugeth->tx_glbl_pram_offset)) {
+	if (tx_glbl_pram_offset < 0) {
 		if (netif_msg_ifup(ugeth))
 			pr_err("Can not allocate DPRAM memory for p_tx_glbl_pram\n");
 		return -ENOMEM;
 	}
-	ugeth->p_tx_glbl_pram =
-	    (struct ucc_geth_tx_global_pram __iomem *) qe_muram_addr(ugeth->
-							tx_glbl_pram_offset);
-	/* Zero out p_tx_glbl_pram */
-	memset_io((void __iomem *)ugeth->p_tx_glbl_pram, 0, sizeof(struct ucc_geth_tx_global_pram));
-
+	ugeth->p_tx_glbl_pram = qe_muram_addr(tx_glbl_pram_offset);
 	/* Fill global PRAM */
 
 	/* TQPTR */
@@ -2565,7 +2327,7 @@ static int ucc_geth_startup(struct ucc_geth_private *ugeth)
 	/* SQPTR */
 	/* Size varies with number of Tx queues */
 	ugeth->send_q_mem_reg_offset =
-	    qe_muram_alloc(ug_info->numQueuesTx *
+	    qe_muram_alloc(ucc_geth_tx_queues(ug_info) *
 			   sizeof(struct ucc_geth_send_queue_qd),
 			   UCC_GETH_SEND_QUEUE_QUEUE_DESCRIPTOR_ALIGNMENT);
 	if (IS_ERR_VALUE(ugeth->send_q_mem_reg_offset)) {
@@ -2581,29 +2343,20 @@ static int ucc_geth_startup(struct ucc_geth_private *ugeth)
 
 	/* Setup the table */
 	/* Assume BD rings are already established */
-	for (i = 0; i < ug_info->numQueuesTx; i++) {
+	for (i = 0; i < ucc_geth_tx_queues(ug_info); i++) {
 		endOfRing =
 		    ugeth->p_tx_bd_ring[i] + (ug_info->bdRingLenTx[i] -
 					      1) * sizeof(struct qe_bd);
-		if (ugeth->ug_info->uf_info.bd_mem_part == MEM_PART_SYSTEM) {
-			out_be32(&ugeth->p_send_q_mem_reg->sqqd[i].bd_ring_base,
-				 (u32) virt_to_phys(ugeth->p_tx_bd_ring[i]));
-			out_be32(&ugeth->p_send_q_mem_reg->sqqd[i].
-				 last_bd_completed_address,
-				 (u32) virt_to_phys(endOfRing));
-		} else if (ugeth->ug_info->uf_info.bd_mem_part ==
-			   MEM_PART_MURAM) {
-			out_be32(&ugeth->p_send_q_mem_reg->sqqd[i].bd_ring_base,
-				 (u32)qe_muram_dma(ugeth->p_tx_bd_ring[i]));
-			out_be32(&ugeth->p_send_q_mem_reg->sqqd[i].
-				 last_bd_completed_address,
-				 (u32)qe_muram_dma(endOfRing));
-		}
+		out_be32(&ugeth->p_send_q_mem_reg->sqqd[i].bd_ring_base,
+			 (u32) virt_to_phys(ugeth->p_tx_bd_ring[i]));
+		out_be32(&ugeth->p_send_q_mem_reg->sqqd[i].
+			 last_bd_completed_address,
+			 (u32) virt_to_phys(endOfRing));
 	}
 
 	/* schedulerbasepointer */
 
-	if (ug_info->numQueuesTx > 1) {
+	if (ucc_geth_tx_queues(ug_info) > 1) {
 	/* scheduler exists only if more than 1 tx queue */
 		ugeth->scheduler_offset =
 		    qe_muram_alloc(sizeof(struct ucc_geth_scheduler),
@@ -2619,8 +2372,6 @@ static int ucc_geth_startup(struct ucc_geth_private *ugeth)
 							   scheduler_offset);
 		out_be32(&ugeth->p_tx_glbl_pram->schedulerbasepointer,
 			 ugeth->scheduler_offset);
-		/* Zero out p_scheduler */
-		memset_io((void __iomem *)ugeth->p_scheduler, 0, sizeof(struct ucc_geth_scheduler));
 
 		/* Set values in scheduler */
 		out_be32(&ugeth->p_scheduler->mblinterval,
@@ -2663,23 +2414,18 @@ static int ucc_geth_startup(struct ucc_geth_private *ugeth)
 		ugeth->p_tx_fw_statistics_pram =
 		    (struct ucc_geth_tx_firmware_statistics_pram __iomem *)
 		    qe_muram_addr(ugeth->tx_fw_statistics_pram_offset);
-		/* Zero out p_tx_fw_statistics_pram */
-		memset_io((void __iomem *)ugeth->p_tx_fw_statistics_pram,
-		       0, sizeof(struct ucc_geth_tx_firmware_statistics_pram));
 	}
 
 	/* temoder */
 	/* Already has speed set */
 
-	if (ug_info->numQueuesTx > 1)
+	if (ucc_geth_tx_queues(ug_info) > 1)
 		temoder |= TEMODER_SCHEDULER_ENABLE;
 	if (ug_info->ipCheckSumGenerate)
 		temoder |= TEMODER_IP_CHECKSUM_GENERATE;
-	temoder |= ((ug_info->numQueuesTx - 1) << TEMODER_NUM_OF_QUEUES_SHIFT);
+	temoder |= ((ucc_geth_tx_queues(ug_info) - 1) << TEMODER_NUM_OF_QUEUES_SHIFT);
 	out_be16(&ugeth->p_tx_glbl_pram->temoder, temoder);
 
-	test = in_be16(&ugeth->p_tx_glbl_pram->temoder);
-
 	/* Function code register value to be used later */
 	function_code = UCC_BMR_BO_BE | UCC_BMR_GBL;
 	/* Required for QE */
@@ -2689,20 +2435,15 @@ static int ucc_geth_startup(struct ucc_geth_private *ugeth)
 
 	/* Rx global PRAM */
 	/* Allocate global rx parameter RAM page */
-	ugeth->rx_glbl_pram_offset =
+	rx_glbl_pram_offset =
 	    qe_muram_alloc(sizeof(struct ucc_geth_rx_global_pram),
 			   UCC_GETH_RX_GLOBAL_PRAM_ALIGNMENT);
-	if (IS_ERR_VALUE(ugeth->rx_glbl_pram_offset)) {
+	if (rx_glbl_pram_offset < 0) {
 		if (netif_msg_ifup(ugeth))
 			pr_err("Can not allocate DPRAM memory for p_rx_glbl_pram\n");
 		return -ENOMEM;
 	}
-	ugeth->p_rx_glbl_pram =
-	    (struct ucc_geth_rx_global_pram __iomem *) qe_muram_addr(ugeth->
-							rx_glbl_pram_offset);
-	/* Zero out p_rx_glbl_pram */
-	memset_io((void __iomem *)ugeth->p_rx_glbl_pram, 0, sizeof(struct ucc_geth_rx_global_pram));
-
+	ugeth->p_rx_glbl_pram = qe_muram_addr(rx_glbl_pram_offset);
 	/* Fill global PRAM */
 
 	/* RQPTR */
@@ -2740,16 +2481,13 @@ static int ucc_geth_startup(struct ucc_geth_private *ugeth)
 		ugeth->p_rx_fw_statistics_pram =
 		    (struct ucc_geth_rx_firmware_statistics_pram __iomem *)
 		    qe_muram_addr(ugeth->rx_fw_statistics_pram_offset);
-		/* Zero out p_rx_fw_statistics_pram */
-		memset_io((void __iomem *)ugeth->p_rx_fw_statistics_pram, 0,
-		       sizeof(struct ucc_geth_rx_firmware_statistics_pram));
 	}
 
 	/* intCoalescingPtr */
 
 	/* Size varies with number of Rx queues */
 	ugeth->rx_irq_coalescing_tbl_offset =
-	    qe_muram_alloc(ug_info->numQueuesRx *
+	    qe_muram_alloc(ucc_geth_rx_queues(ug_info) *
 			   sizeof(struct ucc_geth_rx_interrupt_coalescing_entry)
 			   + 4, UCC_GETH_RX_INTERRUPT_COALESCING_ALIGNMENT);
 	if (IS_ERR_VALUE(ugeth->rx_irq_coalescing_tbl_offset)) {
@@ -2765,7 +2503,7 @@ static int ucc_geth_startup(struct ucc_geth_private *ugeth)
 		 ugeth->rx_irq_coalescing_tbl_offset);
 
 	/* Fill interrupt coalescing table */
-	for (i = 0; i < ug_info->numQueuesRx; i++) {
+	for (i = 0; i < ucc_geth_rx_queues(ug_info); i++) {
 		out_be32(&ugeth->p_rx_irq_coalescing_tbl->coalescingentry[i].
 			 interruptcoalescingmaxvalue,
 			 ug_info->interruptcoalescingmaxvalue[i]);
@@ -2814,7 +2552,7 @@ static int ucc_geth_startup(struct ucc_geth_private *ugeth)
 	/* RBDQPTR */
 	/* Size varies with number of Rx queues */
 	ugeth->rx_bd_qs_tbl_offset =
-	    qe_muram_alloc(ug_info->numQueuesRx *
+	    qe_muram_alloc(ucc_geth_rx_queues(ug_info) *
 			   (sizeof(struct ucc_geth_rx_bd_queues_entry) +
 			    sizeof(struct ucc_geth_rx_prefetched_bds)),
 			   UCC_GETH_RX_BD_QUEUES_ALIGNMENT);
@@ -2828,23 +2566,12 @@ static int ucc_geth_startup(struct ucc_geth_private *ugeth)
 	    (struct ucc_geth_rx_bd_queues_entry __iomem *) qe_muram_addr(ugeth->
 				    rx_bd_qs_tbl_offset);
 	out_be32(&ugeth->p_rx_glbl_pram->rbdqptr, ugeth->rx_bd_qs_tbl_offset);
-	/* Zero out p_rx_bd_qs_tbl */
-	memset_io((void __iomem *)ugeth->p_rx_bd_qs_tbl,
-	       0,
-	       ug_info->numQueuesRx * (sizeof(struct ucc_geth_rx_bd_queues_entry) +
-				       sizeof(struct ucc_geth_rx_prefetched_bds)));
 
 	/* Setup the table */
 	/* Assume BD rings are already established */
-	for (i = 0; i < ug_info->numQueuesRx; i++) {
-		if (ugeth->ug_info->uf_info.bd_mem_part == MEM_PART_SYSTEM) {
-			out_be32(&ugeth->p_rx_bd_qs_tbl[i].externalbdbaseptr,
-				 (u32) virt_to_phys(ugeth->p_rx_bd_ring[i]));
-		} else if (ugeth->ug_info->uf_info.bd_mem_part ==
-			   MEM_PART_MURAM) {
-			out_be32(&ugeth->p_rx_bd_qs_tbl[i].externalbdbaseptr,
-				 (u32)qe_muram_dma(ugeth->p_rx_bd_ring[i]));
-		}
+	for (i = 0; i < ucc_geth_rx_queues(ug_info); i++) {
+		out_be32(&ugeth->p_rx_bd_qs_tbl[i].externalbdbaseptr,
+			 (u32) virt_to_phys(ugeth->p_rx_bd_ring[i]));
 		/* rest of fields handled by QE */
 	}
 
@@ -2865,7 +2592,7 @@ static int ucc_geth_startup(struct ucc_geth_private *ugeth)
 	    ug_info->
 	    vlanOperationNonTagged << REMODER_VLAN_OPERATION_NON_TAGGED_SHIFT;
 	remoder |= ug_info->rxQoSMode << REMODER_RX_QOS_MODE_SHIFT;
-	remoder |= ((ug_info->numQueuesRx - 1) << REMODER_NUM_OF_QUEUES_SHIFT);
+	remoder |= ((ucc_geth_rx_queues(ug_info) - 1) << REMODER_NUM_OF_QUEUES_SHIFT);
 	if (ug_info->ipCheckSumCheck)
 		remoder |= REMODER_IP_CHECKSUM_CHECK;
 	if (ug_info->ipAddressAlignment)
@@ -2948,14 +2675,11 @@ static int ucc_geth_startup(struct ucc_geth_private *ugeth)
 	 * allocated resources can be released when the channel is freed.
 	 */
 	if (!(ugeth->p_init_enet_param_shadow =
-	      kmalloc(sizeof(struct ucc_geth_init_pram), GFP_KERNEL))) {
+	      kzalloc(sizeof(struct ucc_geth_init_pram), GFP_KERNEL))) {
 		if (netif_msg_ifup(ugeth))
 			pr_err("Can not allocate memory for p_UccInitEnetParamShadows\n");
 		return -ENOMEM;
 	}
-	/* Zero out *p_init_enet_param_shadow */
-	memset((char *)ugeth->p_init_enet_param_shadow,
-	       0, sizeof(struct ucc_geth_init_pram));
 
 	/* Fill shadow InitEnet command parameter structure */
 
@@ -2975,7 +2699,7 @@ static int ucc_geth_startup(struct ucc_geth_private *ugeth)
 	    ((u32) ug_info->numThreadsTx) << ENET_INIT_PARAM_TGF_SHIFT;
 
 	ugeth->p_init_enet_param_shadow->rgftgfrxglobal |=
-	    ugeth->rx_glbl_pram_offset | ug_info->riscRx;
+	    rx_glbl_pram_offset | ug_info->riscRx;
 	if ((ug_info->largestexternallookupkeysize !=
 	     QE_FLTR_LARGEST_EXTERNAL_TABLE_LOOKUP_KEY_SIZE_NONE) &&
 	    (ug_info->largestexternallookupkeysize !=
@@ -3013,7 +2737,7 @@ static int ucc_geth_startup(struct ucc_geth_private *ugeth)
 	}
 
 	ugeth->p_init_enet_param_shadow->txglobal =
-	    ugeth->tx_glbl_pram_offset | ug_info->riscTx;
+	    tx_glbl_pram_offset | ug_info->riscTx;
 	if ((ret_val =
 	     fill_init_enet_entries(ugeth,
 				    &(ugeth->p_init_enet_param_shadow->
@@ -3027,7 +2751,7 @@ static int ucc_geth_startup(struct ucc_geth_private *ugeth)
 	}
 
 	/* Load Rx bds with buffers */
-	for (i = 0; i < ug_info->numQueuesRx; i++) {
+	for (i = 0; i < ucc_geth_rx_queues(ug_info); i++) {
 		if ((ret_val = rx_bd_buffer_set(ugeth, (u8) i)) != 0) {
 			if (netif_msg_ifup(ugeth))
 				pr_err("Can not fill Rx bds with buffers\n");
@@ -3083,7 +2807,8 @@ static int ucc_geth_startup(struct ucc_geth_private *ugeth)
 
 /* This is called by the kernel when a frame is ready for transmission. */
 /* It is pointed to by the dev->hard_start_xmit function pointer */
-static int ucc_geth_start_xmit(struct sk_buff *skb, struct net_device *dev)
+static netdev_tx_t
+ucc_geth_start_xmit(struct sk_buff *skb, struct net_device *dev)
 {
 	struct ucc_geth_private *ugeth = netdev_priv(dev);
 #ifdef CONFIG_UGETH_TX_ON_DEMAND
@@ -3096,6 +2821,7 @@ static int ucc_geth_start_xmit(struct sk_buff *skb, struct net_device *dev)
 
 	ugeth_vdbg("%s: IN", __func__);
 
+	netdev_sent_queue(dev, skb->len);
 	spin_lock_irqsave(&ugeth->lock, flags);
 
 	dev->stats.tx_bytes += skb->len;
@@ -3240,6 +2966,8 @@ static int ucc_geth_tx(struct net_device *dev, u8 txQ)
 {
 	/* Start from the next BD that should be filled */
 	struct ucc_geth_private *ugeth = netdev_priv(dev);
+	unsigned int bytes_sent = 0;
+	int howmany = 0;
 	u8 __iomem *bd;		/* BD pointer */
 	u32 bd_status;
 
@@ -3257,7 +2985,8 @@ static int ucc_geth_tx(struct net_device *dev, u8 txQ)
 		skb = ugeth->tx_skbuff[txQ][ugeth->skb_dirtytx[txQ]];
 		if (!skb)
 			break;
-
+		howmany++;
+		bytes_sent += skb->len;
 		dev->stats.tx_packets++;
 
 		dev_consume_skb_any(skb);
@@ -3279,6 +3008,7 @@ static int ucc_geth_tx(struct net_device *dev, u8 txQ)
 		bd_status = in_be32((u32 __iomem *)bd);
 	}
 	ugeth->confBd[txQ] = bd;
+	netdev_completed_queue(dev, howmany, bytes_sent);
 	return 0;
 }
 
@@ -3292,12 +3022,12 @@ static int ucc_geth_poll(struct napi_struct *napi, int budget)
 
 	/* Tx event processing */
 	spin_lock(&ugeth->lock);
-	for (i = 0; i < ug_info->numQueuesTx; i++)
+	for (i = 0; i < ucc_geth_tx_queues(ug_info); i++)
 		ucc_geth_tx(ugeth->ndev, i);
 	spin_unlock(&ugeth->lock);
 
 	howmany = 0;
-	for (i = 0; i < ug_info->numQueuesRx; i++)
+	for (i = 0; i < ucc_geth_rx_queues(ug_info); i++)
 		howmany += ucc_geth_rx(ugeth, i, budget - howmany);
 
 	if (howmany < budget) {
@@ -3373,7 +3103,7 @@ static int ucc_geth_set_mac_addr(struct net_device *dev, void *p)
 	if (!is_valid_ether_addr(addr->sa_data))
 		return -EADDRNOTAVAIL;
 
-	memcpy(dev->dev_addr, addr->sa_data, dev->addr_len);
+	eth_hw_addr_set(dev, addr->sa_data);
 
 	/*
 	 * If device is not running, we will set mac addr register
@@ -3413,12 +3143,6 @@ static int ucc_geth_init_mac(struct ucc_geth_private *ugeth)
 		goto err;
 	}
 
-	err = adjust_enet_interface(ugeth);
-	if (err) {
-		netif_err(ugeth, ifup, dev, "Cannot configure net device, aborting\n");
-		goto err;
-	}
-
 	/*       Set MACSTNADDR1, MACSTNADDR2                */
 	/* For more details see the hardware spec.           */
 	init_mac_station_addr_regs(dev->dev_addr[0],
@@ -3430,12 +3154,6 @@ static int ucc_geth_init_mac(struct ucc_geth_private *ugeth)
 				   &ugeth->ug_regs->macstnaddr1,
 				   &ugeth->ug_regs->macstnaddr2);
 
-	err = ugeth_enable(ugeth, COMM_DIR_RX_AND_TX);
-	if (err) {
-		netif_err(ugeth, ifup, dev, "Cannot enable net device, aborting\n");
-		goto err;
-	}
-
 	return 0;
 err:
 	ucc_geth_stop(ugeth);
@@ -3458,10 +3176,10 @@ static int ucc_geth_open(struct net_device *dev)
 		return -EINVAL;
 	}
 
-	err = init_phy(dev);
+	err = phylink_of_phy_connect(ugeth->phylink, ugeth->dev->of_node, 0);
 	if (err) {
-		netif_err(ugeth, ifup, dev, "Cannot initialize PHY, aborting\n");
-		return err;
+		dev_err(&dev->dev, "Could not attach to PHY\n");
+		return -ENODEV;
 	}
 
 	err = ucc_geth_init_mac(ugeth);
@@ -3477,12 +3195,13 @@ static int ucc_geth_open(struct net_device *dev)
 		goto err;
 	}
 
-	phy_start(ugeth->phydev);
+	phylink_start(ugeth->phylink);
 	napi_enable(&ugeth->napi);
+	netdev_reset_queue(dev);
 	netif_start_queue(dev);
 
 	device_set_wakeup_capable(&dev->dev,
-			qe_alive_during_sleep() || ugeth->phydev->irq);
+			qe_alive_during_sleep() || dev->phydev->irq);
 	device_set_wakeup_enable(&dev->dev, ugeth->wol_en);
 
 	return err;
@@ -3503,12 +3222,12 @@ static int ucc_geth_close(struct net_device *dev)
 
 	cancel_work_sync(&ugeth->timeout_work);
 	ucc_geth_stop(ugeth);
-	phy_disconnect(ugeth->phydev);
-	ugeth->phydev = NULL;
+	phylink_disconnect_phy(ugeth->phylink);
 
 	free_irq(ugeth->ug_info->uf_info.irq, ugeth->ndev);
 
 	netif_stop_queue(dev);
+	netdev_reset_queue(dev);
 
 	return 0;
 }
@@ -3537,7 +3256,7 @@ static void ucc_geth_timeout_work(struct work_struct *work)
 		ucc_geth_stop(ugeth);
 		ucc_geth_init_mac(ugeth);
 		/* Must start PHY here */
-		phy_start(ugeth->phydev);
+		phylink_start(ugeth->phylink);
 		netif_tx_start_all_queues(dev);
 	}
 
@@ -3548,7 +3267,7 @@ static void ucc_geth_timeout_work(struct work_struct *work)
  * ucc_geth_timeout gets called when a packet has not been
  * transmitted after a set amount of time.
  */
-static void ucc_geth_timeout(struct net_device *dev)
+static void ucc_geth_timeout(struct net_device *dev, unsigned int txqueue)
 {
 	struct ucc_geth_private *ugeth = netdev_priv(dev);
 
@@ -3562,6 +3281,7 @@ static int ucc_geth_suspend(struct platform_device *ofdev, pm_message_t state)
 {
 	struct net_device *ndev = platform_get_drvdata(ofdev);
 	struct ucc_geth_private *ugeth = netdev_priv(ndev);
+	bool mac_wol = false;
 
 	if (!netif_running(ndev))
 		return 0;
@@ -3575,14 +3295,17 @@ static int ucc_geth_suspend(struct platform_device *ofdev, pm_message_t state)
 	 */
 	ugeth_disable(ugeth, COMM_DIR_RX_AND_TX);
 
-	if (ugeth->wol_en & WAKE_MAGIC) {
+	if (ugeth->wol_en & WAKE_MAGIC && !ugeth->phy_wol_en) {
 		setbits32(ugeth->uccf->p_uccm, UCC_GETH_UCCE_MPD);
 		setbits32(&ugeth->ug_regs->maccfg2, MACCFG2_MPE);
 		ucc_fast_enable(ugeth->uccf, COMM_DIR_RX_AND_TX);
-	} else if (!(ugeth->wol_en & WAKE_PHY)) {
-		phy_stop(ugeth->phydev);
+		mac_wol = true;
 	}
 
+	rtnl_lock();
+	phylink_suspend(ugeth->phylink, mac_wol);
+	rtnl_unlock();
+
 	return 0;
 }
 
@@ -3616,12 +3339,9 @@ static int ucc_geth_resume(struct platform_device *ofdev)
 		}
 	}
 
-	ugeth->oldlink = 0;
-	ugeth->oldspeed = 0;
-	ugeth->oldduplex = -1;
-
-	phy_stop(ugeth->phydev);
-	phy_start(ugeth->phydev);
+	rtnl_lock();
+	phylink_resume(ugeth->phylink);
+	rtnl_unlock();
 
 	napi_enable(&ugeth->napi);
 	netif_device_attach(ndev);
@@ -3634,32 +3354,6 @@ static int ucc_geth_resume(struct platform_device *ofdev)
 #define ucc_geth_resume NULL
 #endif
 
-static phy_interface_t to_phy_interface(const char *phy_connection_type)
-{
-	if (strcasecmp(phy_connection_type, "mii") == 0)
-		return PHY_INTERFACE_MODE_MII;
-	if (strcasecmp(phy_connection_type, "gmii") == 0)
-		return PHY_INTERFACE_MODE_GMII;
-	if (strcasecmp(phy_connection_type, "tbi") == 0)
-		return PHY_INTERFACE_MODE_TBI;
-	if (strcasecmp(phy_connection_type, "rmii") == 0)
-		return PHY_INTERFACE_MODE_RMII;
-	if (strcasecmp(phy_connection_type, "rgmii") == 0)
-		return PHY_INTERFACE_MODE_RGMII;
-	if (strcasecmp(phy_connection_type, "rgmii-id") == 0)
-		return PHY_INTERFACE_MODE_RGMII_ID;
-	if (strcasecmp(phy_connection_type, "rgmii-txid") == 0)
-		return PHY_INTERFACE_MODE_RGMII_TXID;
-	if (strcasecmp(phy_connection_type, "rgmii-rxid") == 0)
-		return PHY_INTERFACE_MODE_RGMII_RXID;
-	if (strcasecmp(phy_connection_type, "rtbi") == 0)
-		return PHY_INTERFACE_MODE_RTBI;
-	if (strcasecmp(phy_connection_type, "sgmii") == 0)
-		return PHY_INTERFACE_MODE_SGMII;
-
-	return PHY_INTERFACE_MODE_MII;
-}
-
 static int ucc_geth_ioctl(struct net_device *dev, struct ifreq *rq, int cmd)
 {
 	struct ucc_geth_private *ugeth = netdev_priv(dev);
@@ -3667,10 +3361,7 @@ static int ucc_geth_ioctl(struct net_device *dev, struct ifreq *rq, int cmd)
 	if (!netif_running(dev))
 		return -EINVAL;
 
-	if (!ugeth->phydev)
-		return -ENODEV;
-
-	return phy_mii_ioctl(ugeth->phydev, rq, cmd);
+	return phylink_mii_ioctl(ugeth->phylink, rq, cmd);
 }
 
 static const struct net_device_ops ucc_geth_netdev_ops = {
@@ -3681,12 +3372,48 @@ static const struct net_device_ops ucc_geth_netdev_ops = {
 	.ndo_set_mac_address	= ucc_geth_set_mac_addr,
 	.ndo_set_rx_mode	= ucc_geth_set_multi,
 	.ndo_tx_timeout		= ucc_geth_timeout,
-	.ndo_do_ioctl		= ucc_geth_ioctl,
+	.ndo_eth_ioctl		= ucc_geth_ioctl,
 #ifdef CONFIG_NET_POLL_CONTROLLER
 	.ndo_poll_controller	= ucc_netpoll,
 #endif
 };
 
+static int ucc_geth_parse_clock(struct device_node *np, const char *which,
+				enum qe_clock *out)
+{
+	const char *sprop;
+	char buf[24];
+
+	snprintf(buf, sizeof(buf), "%s-clock-name", which);
+	sprop = of_get_property(np, buf, NULL);
+	if (sprop) {
+		*out = qe_clock_source(sprop);
+	} else {
+		u32 val;
+
+		snprintf(buf, sizeof(buf), "%s-clock", which);
+		if (of_property_read_u32(np, buf, &val)) {
+			/* If both *-clock-name and *-clock are missing,
+			 * we want to tell people to use *-clock-name.
+			 */
+			pr_err("missing %s-clock-name property\n", buf);
+			return -EINVAL;
+		}
+		*out = val;
+	}
+	if (*out < QE_CLK_NONE || *out > QE_CLK24) {
+		pr_err("invalid %s property\n", buf);
+		return -EINVAL;
+	}
+	return 0;
+}
+
+static const struct phylink_mac_ops ugeth_mac_ops = {
+	.mac_link_up = ugeth_mac_link_up,
+	.mac_link_down = ugeth_mac_link_down,
+	.mac_config = ugeth_mac_config,
+};
+
 static int ucc_geth_probe(struct platform_device* ofdev)
 {
 	struct device *device = &ofdev->dev;
@@ -3694,25 +3421,12 @@ static int ucc_geth_probe(struct platform_device* ofdev)
 	struct net_device *dev = NULL;
 	struct ucc_geth_private *ugeth = NULL;
 	struct ucc_geth_info *ug_info;
+	struct device_node *phy_node;
+	struct phylink *phylink;
 	struct resource res;
-	int err, ucc_num, max_speed = 0;
+	int err, ucc_num;
 	const unsigned int *prop;
-	const char *sprop;
-	const void *mac_addr;
 	phy_interface_t phy_interface;
-	static const int enet_to_speed[] = {
-		SPEED_10, SPEED_10, SPEED_10,
-		SPEED_100, SPEED_100, SPEED_100,
-		SPEED_1000, SPEED_1000, SPEED_1000, SPEED_1000,
-	};
-	static const phy_interface_t enet_to_phy_interface[] = {
-		PHY_INTERFACE_MODE_MII, PHY_INTERFACE_MODE_RMII,
-		PHY_INTERFACE_MODE_RGMII, PHY_INTERFACE_MODE_MII,
-		PHY_INTERFACE_MODE_RMII, PHY_INTERFACE_MODE_RGMII,
-		PHY_INTERFACE_MODE_GMII, PHY_INTERFACE_MODE_RGMII,
-		PHY_INTERFACE_MODE_TBI, PHY_INTERFACE_MODE_RTBI,
-		PHY_INTERFACE_MODE_SGMII,
-	};
 
 	ugeth_vdbg("%s: IN", __func__);
 
@@ -3727,117 +3441,56 @@ static int ucc_geth_probe(struct platform_device* ofdev)
 	if ((ucc_num < 0) || (ucc_num > 7))
 		return -ENODEV;
 
-	ug_info = &ugeth_info[ucc_num];
-	if (ug_info == NULL) {
-		if (netif_msg_probe(&debug))
-			pr_err("[%d] Missing additional data!\n", ucc_num);
-		return -ENODEV;
-	}
+	ug_info = devm_kmemdup(&ofdev->dev, &ugeth_primary_info,
+			       sizeof(*ug_info), GFP_KERNEL);
+	if (!ug_info)
+		return -ENOMEM;
 
 	ug_info->uf_info.ucc_num = ucc_num;
 
-	sprop = of_get_property(np, "rx-clock-name", NULL);
-	if (sprop) {
-		ug_info->uf_info.rx_clock = qe_clock_source(sprop);
-		if ((ug_info->uf_info.rx_clock < QE_CLK_NONE) ||
-		    (ug_info->uf_info.rx_clock > QE_CLK24)) {
-			pr_err("invalid rx-clock-name property\n");
-			return -EINVAL;
-		}
-	} else {
-		prop = of_get_property(np, "rx-clock", NULL);
-		if (!prop) {
-			/* If both rx-clock-name and rx-clock are missing,
-			   we want to tell people to use rx-clock-name. */
-			pr_err("missing rx-clock-name property\n");
-			return -EINVAL;
-		}
-		if ((*prop < QE_CLK_NONE) || (*prop > QE_CLK24)) {
-			pr_err("invalid rx-clock property\n");
-			return -EINVAL;
-		}
-		ug_info->uf_info.rx_clock = *prop;
-	}
-
-	sprop = of_get_property(np, "tx-clock-name", NULL);
-	if (sprop) {
-		ug_info->uf_info.tx_clock = qe_clock_source(sprop);
-		if ((ug_info->uf_info.tx_clock < QE_CLK_NONE) ||
-		    (ug_info->uf_info.tx_clock > QE_CLK24)) {
-			pr_err("invalid tx-clock-name property\n");
-			return -EINVAL;
-		}
-	} else {
-		prop = of_get_property(np, "tx-clock", NULL);
-		if (!prop) {
-			pr_err("missing tx-clock-name property\n");
-			return -EINVAL;
-		}
-		if ((*prop < QE_CLK_NONE) || (*prop > QE_CLK24)) {
-			pr_err("invalid tx-clock property\n");
-			return -EINVAL;
-		}
-		ug_info->uf_info.tx_clock = *prop;
-	}
+	err = ucc_geth_parse_clock(np, "rx", &ug_info->uf_info.rx_clock);
+	if (err)
+		return err;
+	err = ucc_geth_parse_clock(np, "tx", &ug_info->uf_info.tx_clock);
+	if (err)
+		return err;
 
 	err = of_address_to_resource(np, 0, &res);
 	if (err)
-		return -EINVAL;
+		return err;
 
 	ug_info->uf_info.regs = res.start;
 	ug_info->uf_info.irq = irq_of_parse_and_map(np, 0);
 
-	ug_info->phy_node = of_parse_phandle(np, "phy-handle", 0);
-	if (!ug_info->phy_node && of_phy_is_fixed_link(np)) {
-		/*
-		 * In the case of a fixed PHY, the DT node associated
-		 * to the PHY is the Ethernet MAC DT node.
-		 */
-		err = of_phy_register_fixed_link(np);
-		if (err)
-			return err;
-		ug_info->phy_node = of_node_get(np);
-	}
-
 	/* Find the TBI PHY node.  If it's not there, we don't support SGMII */
 	ug_info->tbi_node = of_parse_phandle(np, "tbi-handle", 0);
 
-	/* get the phy interface type, or default to MII */
-	prop = of_get_property(np, "phy-connection-type", NULL);
-	if (!prop) {
-		/* handle interface property present in old trees */
-		prop = of_get_property(ug_info->phy_node, "interface", NULL);
-		if (prop != NULL) {
-			phy_interface = enet_to_phy_interface[*prop];
-			max_speed = enet_to_speed[*prop];
-		} else
-			phy_interface = PHY_INTERFACE_MODE_MII;
-	} else {
-		phy_interface = to_phy_interface((const char *)prop);
-	}
-
-	/* get speed, or derive from PHY interface */
-	if (max_speed == 0)
-		switch (phy_interface) {
-		case PHY_INTERFACE_MODE_GMII:
-		case PHY_INTERFACE_MODE_RGMII:
-		case PHY_INTERFACE_MODE_RGMII_ID:
-		case PHY_INTERFACE_MODE_RGMII_RXID:
-		case PHY_INTERFACE_MODE_RGMII_TXID:
-		case PHY_INTERFACE_MODE_TBI:
-		case PHY_INTERFACE_MODE_RTBI:
-		case PHY_INTERFACE_MODE_SGMII:
-			max_speed = SPEED_1000;
-			break;
-		default:
-			max_speed = SPEED_100;
-			break;
+	phy_node = of_parse_phandle(np, "phy-handle", 0);
+	if (phy_node) {
+		prop = of_get_property(phy_node, "interface", NULL);
+		if (prop) {
+			dev_err(&ofdev->dev,
+				"Device-tree property 'interface' is no longer supported. Please use 'phy-connection-type' instead.");
+			of_node_put(phy_node);
+			err = -EINVAL;
+			goto err_put_tbi;
 		}
+		of_node_put(phy_node);
+	}
 
-	if (max_speed == SPEED_1000) {
+	err = of_get_phy_mode(np, &phy_interface);
+	if (err) {
+		dev_err(&ofdev->dev, "Invalid phy-connection-type");
+		goto err_put_tbi;
+	}
+
+	if (phy_interface == PHY_INTERFACE_MODE_GMII ||
+	    phy_interface_mode_is_rgmii(phy_interface) ||
+	    phy_interface == PHY_INTERFACE_MODE_TBI ||
+	    phy_interface == PHY_INTERFACE_MODE_RTBI ||
+	    phy_interface == PHY_INTERFACE_MODE_SGMII) {
 		unsigned int snums = qe_get_num_of_snums();
 
-		/* configure muram FIFOs for gigabit operation */
 		ug_info->uf_info.urfs = UCC_GETH_URFS_GIGA_INIT;
 		ug_info->uf_info.urfet = UCC_GETH_URFET_GIGA_INIT;
 		ug_info->uf_info.urfset = UCC_GETH_URFSET_GIGA_INIT;
@@ -3857,16 +3510,16 @@ static int ucc_geth_probe(struct platform_device* ofdev)
 	}
 
 	if (netif_msg_probe(&debug))
-		pr_info("UCC%1d at 0x%8x (irq = %d)\n",
-			ug_info->uf_info.ucc_num + 1, ug_info->uf_info.regs,
+		pr_info("UCC%1d at 0x%8llx (irq = %d)\n",
+			ug_info->uf_info.ucc_num + 1,
+			(u64)ug_info->uf_info.regs,
 			ug_info->uf_info.irq);
 
 	/* Create an ethernet device instance */
-	dev = alloc_etherdev(sizeof(*ugeth));
-
-	if (dev == NULL) {
+	dev = devm_alloc_etherdev(&ofdev->dev, sizeof(*ugeth));
+	if (!dev) {
 		err = -ENOMEM;
-		goto err_deregister_fixed_link;
+		goto err_put_tbi;
 	}
 
 	ugeth = netdev_priv(dev);
@@ -3888,27 +3541,55 @@ static int ucc_geth_probe(struct platform_device* ofdev)
 	dev->netdev_ops = &ucc_geth_netdev_ops;
 	dev->watchdog_timeo = TX_TIMEOUT;
 	INIT_WORK(&ugeth->timeout_work, ucc_geth_timeout_work);
-	netif_napi_add(dev, &ugeth->napi, ucc_geth_poll, 64);
+	netif_napi_add(dev, &ugeth->napi, ucc_geth_poll);
 	dev->mtu = 1500;
+	dev->max_mtu = 1518;
 
 	ugeth->msg_enable = netif_msg_init(debug.msg_enable, UGETH_MSG_DEFAULT);
-	ugeth->phy_interface = phy_interface;
-	ugeth->max_speed = max_speed;
 
-	/* Carrier starts down, phylib will bring it up */
-	netif_carrier_off(dev);
+	ugeth->phylink_config.dev = &dev->dev;
+	ugeth->phylink_config.type = PHYLINK_NETDEV;
+
+	ugeth->phylink_config.mac_capabilities =
+		MAC_SYM_PAUSE | MAC_10 | MAC_100 | MAC_1000FD;
+
+	__set_bit(PHY_INTERFACE_MODE_MII,
+		  ugeth->phylink_config.supported_interfaces);
+	__set_bit(PHY_INTERFACE_MODE_RMII,
+		  ugeth->phylink_config.supported_interfaces);
+	__set_bit(PHY_INTERFACE_MODE_GMII,
+		  ugeth->phylink_config.supported_interfaces);
+	phy_interface_set_rgmii(ugeth->phylink_config.supported_interfaces);
 
-	err = register_netdev(dev);
+	if (ug_info->tbi_node) {
+		__set_bit(PHY_INTERFACE_MODE_SGMII,
+			  ugeth->phylink_config.supported_interfaces);
+		__set_bit(PHY_INTERFACE_MODE_TBI,
+			  ugeth->phylink_config.supported_interfaces);
+		__set_bit(PHY_INTERFACE_MODE_RTBI,
+			  ugeth->phylink_config.supported_interfaces);
+	}
+
+	phylink = phylink_create(&ugeth->phylink_config, dev_fwnode(&dev->dev),
+				 phy_interface, &ugeth_mac_ops);
+	if (IS_ERR(phylink)) {
+		err = PTR_ERR(phylink);
+		goto err_put_tbi;
+	}
+
+	ugeth->phylink = phylink;
+
+	err = devm_register_netdev(&ofdev->dev, dev);
 	if (err) {
 		if (netif_msg_probe(ugeth))
 			pr_err("%s: Cannot register net device, aborting\n",
 			       dev->name);
-		goto err_free_netdev;
+		goto err_destroy_phylink;
 	}
 
-	mac_addr = of_get_mac_address(np);
-	if (mac_addr)
-		memcpy(dev->dev_addr, mac_addr, ETH_ALEN);
+	err = of_get_ethdev_address(np, dev);
+	if (err == -EPROBE_DEFER)
+		goto err_destroy_phylink;
 
 	ugeth->ug_info = ug_info;
 	ugeth->dev = device;
@@ -3917,32 +3598,22 @@ static int ucc_geth_probe(struct platform_device* ofdev)
 
 	return 0;
 
-err_free_netdev:
-	free_netdev(dev);
-err_deregister_fixed_link:
-	if (of_phy_is_fixed_link(np))
-		of_phy_deregister_fixed_link(np);
+err_destroy_phylink:
+	phylink_destroy(phylink);
+err_put_tbi:
 	of_node_put(ug_info->tbi_node);
-	of_node_put(ug_info->phy_node);
 
 	return err;
 }
 
-static int ucc_geth_remove(struct platform_device* ofdev)
+static void ucc_geth_remove(struct platform_device* ofdev)
 {
 	struct net_device *dev = platform_get_drvdata(ofdev);
 	struct ucc_geth_private *ugeth = netdev_priv(dev);
-	struct device_node *np = ofdev->dev.of_node;
 
-	unregister_netdev(dev);
-	free_netdev(dev);
 	ucc_geth_memclean(ugeth);
-	if (of_phy_is_fixed_link(np))
-		of_phy_deregister_fixed_link(np);
+	phylink_destroy(ugeth->phylink);
 	of_node_put(ugeth->ug_info->tbi_node);
-	of_node_put(ugeth->ug_info->phy_node);
-
-	return 0;
 }
 
 static const struct of_device_id ucc_geth_match[] = {
@@ -3968,17 +3639,10 @@ static struct platform_driver ucc_geth_driver = {
 
 static int __init ucc_geth_init(void)
 {
-	int i, ret;
-
 	if (netif_msg_drv(&debug))
 		pr_info(DRV_DESC "\n");
-	for (i = 0; i < 8; i++)
-		memcpy(&(ugeth_info[i]), &ugeth_primary_info,
-		       sizeof(ugeth_primary_info));
-
-	ret = platform_driver_register(&ucc_geth_driver);
 
-	return ret;
+	return platform_driver_register(&ucc_geth_driver);
 }
 
 static void __exit ucc_geth_exit(void)
@@ -3991,5 +3655,4 @@ module_exit(ucc_geth_exit);
 
 MODULE_AUTHOR("Freescale Semiconductor, Inc");
 MODULE_DESCRIPTION(DRV_DESC);
-MODULE_VERSION(DRV_VERSION);
 MODULE_LICENSE("GPL");
diff --git a/drivers/net/ethernet/freescale/ucc_geth.h b/drivers/net/ethernet/freescale/ucc_geth.h
index 5da19b440a6a..84f92f6384e7 100644
--- a/drivers/net/ethernet/freescale/ucc_geth.h
+++ b/drivers/net/ethernet/freescale/ucc_geth.h
@@ -1,3 +1,4 @@
+/* SPDX-License-Identifier: GPL-2.0-or-later */
 /*
  * Copyright (C) Freescale Semicondutor, Inc. 2006-2009. All rights reserved.
  *
@@ -9,17 +10,13 @@
  * Changelog:
  * Jun 28, 2006 Li Yang <LeoLi@freescale.com>
  * - Rearrange code and style fixes
- *
- * This program is free software; you can redistribute  it and/or modify it
- * under  the terms of  the GNU General  Public License as published by the
- * Free Software Foundation;  either version 2 of the  License, or (at your
- * option) any later version.
  */
 #ifndef __UCC_GETH_H__
 #define __UCC_GETH_H__
 
 #include <linux/kernel.h>
 #include <linux/list.h>
+#include <linux/phylink.h>
 #include <linux/if_ether.h>
 
 #include <soc/fsl/qe/immap_qe.h>
@@ -30,7 +27,6 @@
 
 #define DRV_DESC "QE UCC Gigabit Ethernet Controller"
 #define DRV_NAME "ucc_geth"
-#define DRV_VERSION "1.1"
 
 #define NUM_TX_QUEUES                   8
 #define NUM_RX_QUEUES                   8
@@ -580,7 +576,14 @@ struct ucc_geth_tx_global_pram {
 	u32 vtagtable[0x8];	/* 8 4-byte VLAN tags */
 	u32 tqptr;		/* a base pointer to the Tx Queues Memory
 				   Region */
-	u8 res2[0x80 - 0x74];
+	u8 res2[0x78 - 0x74];
+	u64 snums_en;
+	u32 l2l3baseptr;	/* top byte consists of a few other bit fields */
+
+	u16 mtu[8];
+	u8 res3[0xa8 - 0x94];
+	u32 wrrtablebase;	/* top byte is reserved */
+	u8 res4[0xc0 - 0xac];
 } __packed;
 
 /* structure representing Extended Filtering Global Parameters in PRAM */
@@ -887,9 +890,6 @@ struct ucc_geth_hardware_statistics {
 							   addresses */
 
 #define TX_TIMEOUT                              (1*HZ)
-#define SKB_ALLOC_TIMEOUT                       100000
-#define PHY_INIT_TIMEOUT                        100000
-#define PHY_CHANGE_TIME                         2
 
 /* Fast Ethernet (10/100 Mbps) */
 #define UCC_GETH_URFS_INIT                      512	/* Rx virtual FIFO size
@@ -920,7 +920,8 @@ struct ucc_geth_hardware_statistics {
 #define UCC_GETH_UPSMR_INIT                     UCC_GETH_UPSMR_RES1
 
 #define UCC_GETH_MACCFG1_INIT                   0
-#define UCC_GETH_MACCFG2_INIT                   (MACCFG2_RESERVED_1)
+#define UCC_GETH_MACCFG2_INIT                   (MACCFG2_RESERVED_1 | \
+						 (7 << MACCFG2_PREL_SHIFT))
 
 /* Ethernet Address Type. */
 enum enet_addr_type {
@@ -1072,11 +1073,12 @@ struct ucc_geth_tad_params {
 	u16 vid;
 };
 
+struct phylink;
+struct phylink_config;
+
 /* GETH protocol initialization structure */
 struct ucc_geth_info {
 	struct ucc_fast_info uf_info;
-	u8 numQueuesTx;
-	u8 numQueuesRx;
 	int ipCheckSumCheck;
 	int ipCheckSumGenerate;
 	int rxExtendedFiltering;
@@ -1089,7 +1091,6 @@ struct ucc_geth_info {
 	u8 miminumInterFrameGapEnforcement;
 	u8 backToBackInterFrameGap;
 	int ipAddressAlignment;
-	int lengthCheckRx;
 	u32 mblinterval;
 	u16 nortsrbytetime;
 	u8 fracsiz;
@@ -1115,7 +1116,6 @@ struct ucc_geth_info {
 	int transmitFlowControl;
 	u8 maxGroupAddrInHash;
 	u8 maxIndAddrInHash;
-	u8 prel;
 	u16 maxFrameLength;
 	u16 minFrameLength;
 	u16 maxD1Length;
@@ -1126,7 +1126,6 @@ struct ucc_geth_info {
 	u32 eventRegMask;
 	u16 pausePeriod;
 	u16 extensionField;
-	struct device_node *phy_node;
 	struct device_node *tbi_node;
 	u8 weightfactor[NUM_TX_QUEUES];
 	u8 interruptcoalescingmaxvalue[NUM_RX_QUEUES];
@@ -1164,9 +1163,7 @@ struct ucc_geth_private {
 	struct ucc_geth_exf_global_pram __iomem *p_exf_glbl_param;
 	u32 exf_glbl_param_offset;
 	struct ucc_geth_rx_global_pram __iomem *p_rx_glbl_pram;
-	u32 rx_glbl_pram_offset;
 	struct ucc_geth_tx_global_pram __iomem *p_tx_glbl_pram;
-	u32 tx_glbl_pram_offset;
 	struct ucc_geth_send_queue_mem_region __iomem *p_send_q_mem_reg;
 	u32 send_q_mem_reg_offset;
 	struct ucc_geth_thread_data_tx __iomem *p_thread_data_tx;
@@ -1184,9 +1181,7 @@ struct ucc_geth_private {
 	struct ucc_geth_rx_bd_queues_entry __iomem *p_rx_bd_qs_tbl;
 	u32 rx_bd_qs_tbl_offset;
 	u8 __iomem *p_tx_bd_ring[NUM_TX_QUEUES];
-	u32 tx_bd_ring_offset[NUM_TX_QUEUES];
 	u8 __iomem *p_rx_bd_ring[NUM_RX_QUEUES];
-	u32 rx_bd_ring_offset[NUM_RX_QUEUES];
 	u8 __iomem *confBd[NUM_TX_QUEUES];
 	u8 __iomem *txBd[NUM_TX_QUEUES];
 	u8 __iomem *rxBd[NUM_RX_QUEUES];
@@ -1215,14 +1210,12 @@ struct ucc_geth_private {
 	u16 skb_dirtytx[NUM_TX_QUEUES];
 
 	struct ugeth_mii_info *mii_info;
-	struct phy_device *phydev;
-	phy_interface_t phy_interface;
-	int max_speed;
 	uint32_t msg_enable;
-	int oldspeed;
-	int oldduplex;
-	int oldlink;
-	int wol_en;
+	u32 wol_en;
+	u32 phy_wol_en;
+
+	struct phylink *phylink;
+	struct phylink_config phylink_config;
 
 	struct device_node *node;
 };
diff --git a/drivers/net/ethernet/freescale/ucc_geth_ethtool.c b/drivers/net/ethernet/freescale/ucc_geth_ethtool.c
index b642990b549c..1fb49e5a414a 100644
--- a/drivers/net/ethernet/freescale/ucc_geth_ethtool.c
+++ b/drivers/net/ethernet/freescale/ucc_geth_ethtool.c
@@ -1,3 +1,4 @@
+// SPDX-License-Identifier: GPL-2.0-or-later
 /*
  * Copyright (c) 2007 Freescale Semiconductor, Inc. All rights reserved.
  *
@@ -8,11 +9,6 @@
  * Limitation:
  * Can only get/set settings of the first queue.
  * Need to re-open the interface manually after changing some parameters.
- *
- * This program is free software; you can redistribute  it and/or modify it
- * under  the terms of  the GNU General  Public License as published by the
- * Free Software Foundation;  either version 2 of the  License, or (at your
- * option) any later version.
  */
 
 #include <linux/kernel.h>
@@ -61,7 +57,7 @@ static const char hw_stat_gstrings[][ETH_GSTRING_LEN] = {
 static const char tx_fw_stat_gstrings[][ETH_GSTRING_LEN] = {
 	"tx-single-collision",
 	"tx-multiple-collision",
-	"tx-late-collsion",
+	"tx-late-collision",
 	"tx-aborted-frames",
 	"tx-lost-frames",
 	"tx-carrier-sense-errors",
@@ -108,12 +104,8 @@ static int
 uec_get_ksettings(struct net_device *netdev, struct ethtool_link_ksettings *cmd)
 {
 	struct ucc_geth_private *ugeth = netdev_priv(netdev);
-	struct phy_device *phydev = ugeth->phydev;
-
-	if (!phydev)
-		return -ENODEV;
 
-	return phy_ethtool_ksettings_get(phydev, cmd);
+	return phylink_ethtool_ksettings_get(ugeth->phylink, cmd);
 }
 
 static int
@@ -121,12 +113,8 @@ uec_set_ksettings(struct net_device *netdev,
 		  const struct ethtool_link_ksettings *cmd)
 {
 	struct ucc_geth_private *ugeth = netdev_priv(netdev);
-	struct phy_device *phydev = ugeth->phydev;
 
-	if (!phydev)
-		return -ENODEV;
-
-	return phy_ethtool_ksettings_set(phydev, cmd);
+	return phylink_ethtool_ksettings_set(ugeth->phylink, cmd);
 }
 
 static void
@@ -135,12 +123,7 @@ uec_get_pauseparam(struct net_device *netdev,
 {
 	struct ucc_geth_private *ugeth = netdev_priv(netdev);
 
-	pause->autoneg = ugeth->phydev->autoneg;
-
-	if (ugeth->ug_info->receiveFlowControl)
-		pause->rx_pause = 1;
-	if (ugeth->ug_info->transmitFlowControl)
-		pause->tx_pause = 1;
+	return phylink_ethtool_get_pauseparam(ugeth->phylink, pause);
 }
 
 static int
@@ -148,30 +131,11 @@ uec_set_pauseparam(struct net_device *netdev,
                      struct ethtool_pauseparam *pause)
 {
 	struct ucc_geth_private *ugeth = netdev_priv(netdev);
-	int ret = 0;
 
 	ugeth->ug_info->receiveFlowControl = pause->rx_pause;
 	ugeth->ug_info->transmitFlowControl = pause->tx_pause;
 
-	if (ugeth->phydev->autoneg) {
-		if (netif_running(netdev)) {
-			/* FIXME: automatically restart */
-			netdev_info(netdev, "Please re-open the interface\n");
-		}
-	} else {
-		struct ucc_geth_info *ug_info = ugeth->ug_info;
-
-		ret = init_flow_control_params(ug_info->aufc,
-					ug_info->receiveFlowControl,
-					ug_info->transmitFlowControl,
-					ug_info->pausePeriod,
-					ug_info->extensionField,
-					&ugeth->uccf->uf_regs->upsmr,
-					&ugeth->ug_regs->uempr,
-					&ugeth->ug_regs->maccfg1);
-	}
-
-	return ret;
+	return phylink_ethtool_set_pauseparam(ugeth->phylink, pause);
 }
 
 static uint32_t
@@ -209,7 +173,9 @@ uec_get_regs(struct net_device *netdev,
 
 static void
 uec_get_ringparam(struct net_device *netdev,
-                    struct ethtool_ringparam *ring)
+		  struct ethtool_ringparam *ring,
+		  struct kernel_ethtool_ringparam *kernel_ring,
+		  struct netlink_ext_ack *extack)
 {
 	struct ucc_geth_private *ugeth = netdev_priv(netdev);
 	struct ucc_geth_info *ug_info = ugeth->ug_info;
@@ -228,7 +194,9 @@ uec_get_ringparam(struct net_device *netdev,
 
 static int
 uec_set_ringparam(struct net_device *netdev,
-                    struct ethtool_ringparam *ring)
+		  struct ethtool_ringparam *ring,
+		  struct kernel_ethtool_ringparam *kernel_ring,
+		  struct netlink_ext_ack *extack)
 {
 	struct ucc_geth_private *ugeth = netdev_priv(netdev);
 	struct ucc_geth_info *ug_info = ugeth->ug_info;
@@ -250,14 +218,12 @@ uec_set_ringparam(struct net_device *netdev,
 		return -EINVAL;
 	}
 
+	if (netif_running(netdev))
+		return -EBUSY;
+
 	ug_info->bdRingLenRx[queue] = ring->rx_pending;
 	ug_info->bdRingLenTx[queue] = ring->tx_pending;
 
-	if (netif_running(netdev)) {
-		/* FIXME: restart automatically */
-		netdev_info(netdev, "Please re-open the interface\n");
-	}
-
 	return ret;
 }
 
@@ -287,20 +253,17 @@ static void uec_get_strings(struct net_device *netdev, u32 stringset, u8 *buf)
 {
 	struct ucc_geth_private *ugeth = netdev_priv(netdev);
 	u32 stats_mode = ugeth->ug_info->statisticsMode;
+	int i;
 
-	if (stats_mode & UCC_GETH_STATISTICS_GATHERING_MODE_HARDWARE) {
-		memcpy(buf, hw_stat_gstrings, UEC_HW_STATS_LEN *
-			       	ETH_GSTRING_LEN);
-		buf += UEC_HW_STATS_LEN * ETH_GSTRING_LEN;
-	}
-	if (stats_mode & UCC_GETH_STATISTICS_GATHERING_MODE_FIRMWARE_TX) {
-		memcpy(buf, tx_fw_stat_gstrings, UEC_TX_FW_STATS_LEN *
-			       	ETH_GSTRING_LEN);
-		buf += UEC_TX_FW_STATS_LEN * ETH_GSTRING_LEN;
-	}
+	if (stats_mode & UCC_GETH_STATISTICS_GATHERING_MODE_HARDWARE)
+		for (i = 0; i < UEC_HW_STATS_LEN; i++)
+			ethtool_puts(&buf, hw_stat_gstrings[i]);
+	if (stats_mode & UCC_GETH_STATISTICS_GATHERING_MODE_FIRMWARE_TX)
+		for (i = 0; i < UEC_TX_FW_STATS_LEN; i++)
+			ethtool_puts(&buf, tx_fw_stat_gstrings[i]);
 	if (stats_mode & UCC_GETH_STATISTICS_GATHERING_MODE_FIRMWARE_RX)
-		memcpy(buf, rx_fw_stat_gstrings, UEC_RX_FW_STATS_LEN *
-			       	ETH_GSTRING_LEN);
+		for (i = 0; i < UEC_RX_FW_STATS_LEN; i++)
+			ethtool_puts(&buf, rx_fw_stat_gstrings[i]);
 }
 
 static void uec_get_ethtool_stats(struct net_device *netdev,
@@ -337,10 +300,8 @@ static void
 uec_get_drvinfo(struct net_device *netdev,
                        struct ethtool_drvinfo *drvinfo)
 {
-	strlcpy(drvinfo->driver, DRV_NAME, sizeof(drvinfo->driver));
-	strlcpy(drvinfo->version, DRV_VERSION, sizeof(drvinfo->version));
-	strlcpy(drvinfo->fw_version, "N/A", sizeof(drvinfo->fw_version));
-	strlcpy(drvinfo->bus_info, "QUICC ENGINE", sizeof(drvinfo->bus_info));
+	strscpy(drvinfo->driver, DRV_NAME, sizeof(drvinfo->driver));
+	strscpy(drvinfo->bus_info, "QUICC ENGINE", sizeof(drvinfo->bus_info));
 }
 
 #ifdef CONFIG_PM
@@ -348,28 +309,42 @@ uec_get_drvinfo(struct net_device *netdev,
 static void uec_get_wol(struct net_device *netdev, struct ethtool_wolinfo *wol)
 {
 	struct ucc_geth_private *ugeth = netdev_priv(netdev);
-	struct phy_device *phydev = ugeth->phydev;
 
-	if (phydev && phydev->irq)
-		wol->supported |= WAKE_PHY;
+	phylink_ethtool_get_wol(ugeth->phylink, wol);
+
 	if (qe_alive_during_sleep())
 		wol->supported |= WAKE_MAGIC;
 
-	wol->wolopts = ugeth->wol_en;
+	wol->wolopts |= ugeth->wol_en;
 }
 
 static int uec_set_wol(struct net_device *netdev, struct ethtool_wolinfo *wol)
 {
 	struct ucc_geth_private *ugeth = netdev_priv(netdev);
-	struct phy_device *phydev = ugeth->phydev;
+	int ret = 0;
 
-	if (wol->wolopts & ~(WAKE_PHY | WAKE_MAGIC))
-		return -EINVAL;
-	else if (wol->wolopts & WAKE_PHY && (!phydev || !phydev->irq))
+	ret = phylink_ethtool_set_wol(ugeth->phylink, wol);
+	if (ret == -EOPNOTSUPP) {
+		ugeth->phy_wol_en = 0;
+	} else if (ret) {
+		return ret;
+	} else {
+		ugeth->phy_wol_en = wol->wolopts;
+		goto out;
+	}
+
+	/* If the PHY isn't handling the WoL and the MAC is asked to more than
+	 * WAKE_MAGIC, error-out
+	 */
+	if (!ugeth->phy_wol_en &&
+	    wol->wolopts & ~WAKE_MAGIC)
 		return -EINVAL;
-	else if (wol->wolopts & WAKE_MAGIC && !qe_alive_during_sleep())
+
+	if (wol->wolopts & WAKE_MAGIC &&
+	    !qe_alive_during_sleep())
 		return -EINVAL;
 
+out:
 	ugeth->wol_en = wol->wolopts;
 	device_set_wakeup_enable(&netdev->dev, ugeth->wol_en);
 
diff --git a/drivers/net/ethernet/freescale/xgmac_mdio.c b/drivers/net/ethernet/freescale/xgmac_mdio.c
index e03b30c60dcf..65dc07d0df0f 100644
--- a/drivers/net/ethernet/freescale/xgmac_mdio.c
+++ b/drivers/net/ethernet/freescale/xgmac_mdio.c
@@ -2,6 +2,7 @@
  * QorIQ 10G MDIO Controller
  *
  * Copyright 2012 Freescale Semiconductor, Inc.
+ * Copyright 2021 NXP
  *
  * Authors: Andy Fleming <afleming@freescale.com>
  *          Timur Tabi <timur@freescale.com>
@@ -11,15 +12,18 @@
  * kind, whether express or implied.
  */
 
-#include <linux/kernel.h>
-#include <linux/slab.h>
+#include <linux/acpi.h>
+#include <linux/acpi_mdio.h>
+#include <linux/clk.h>
 #include <linux/interrupt.h>
-#include <linux/module.h>
-#include <linux/phy.h>
+#include <linux/kernel.h>
 #include <linux/mdio.h>
-#include <linux/of_address.h>
-#include <linux/of_platform.h>
+#include <linux/module.h>
+#include <linux/of.h>
 #include <linux/of_mdio.h>
+#include <linux/phy.h>
+#include <linux/platform_device.h>
+#include <linux/slab.h>
 
 /* Number of microseconds to wait for a register to respond */
 #define TIMEOUT	1000
@@ -33,9 +37,10 @@ struct tgec_mdio_controller {
 } __packed;
 
 #define MDIO_STAT_ENC		BIT(6)
-#define MDIO_STAT_CLKDIV(x)	(((x>>1) & 0xff) << 8)
+#define MDIO_STAT_CLKDIV(x)	(((x) & 0x1ff) << 7)
 #define MDIO_STAT_BSY		BIT(0)
 #define MDIO_STAT_RD_ER		BIT(1)
+#define MDIO_STAT_PRE_DIS	BIT(5)
 #define MDIO_CTL_DEV_ADDR(x) 	(x & 0x1f)
 #define MDIO_CTL_PORT_ADDR(x)	((x & 0x1f) << 5)
 #define MDIO_CTL_PRE_DIS	BIT(10)
@@ -44,11 +49,14 @@ struct tgec_mdio_controller {
 #define MDIO_CTL_READ		BIT(15)
 
 #define MDIO_DATA(x)		(x & 0xffff)
-#define MDIO_DATA_BSY		BIT(31)
 
 struct mdio_fsl_priv {
 	struct	tgec_mdio_controller __iomem *mdio_base;
+	struct	clk *enet_clk;
+	u32	mdc_freq;
 	bool	is_little_endian;
+	bool	has_a009885;
+	bool	has_a011043;
 };
 
 static u32 xgmac_read32(void __iomem *regs,
@@ -120,30 +128,49 @@ static int xgmac_wait_until_done(struct device *dev,
 	return 0;
 }
 
-/*
- * Write value to the PHY for this device to the register at regnum,waiting
- * until the write is done before it returns.  All PHY configuration has to be
- * done through the TSEC1 MIIM regs.
- */
-static int xgmac_mdio_write(struct mii_bus *bus, int phy_id, int regnum, u16 value)
+static int xgmac_mdio_write_c22(struct mii_bus *bus, int phy_id, int regnum,
+				u16 value)
 {
 	struct mdio_fsl_priv *priv = (struct mdio_fsl_priv *)bus->priv;
 	struct tgec_mdio_controller __iomem *regs = priv->mdio_base;
-	uint16_t dev_addr;
+	bool endian = priv->is_little_endian;
+	u16 dev_addr = regnum & 0x1f;
 	u32 mdio_ctl, mdio_stat;
 	int ret;
+
+	mdio_stat = xgmac_read32(&regs->mdio_stat, endian);
+	mdio_stat &= ~MDIO_STAT_ENC;
+	xgmac_write32(mdio_stat, &regs->mdio_stat, endian);
+
+	ret = xgmac_wait_until_free(&bus->dev, regs, endian);
+	if (ret)
+		return ret;
+
+	/* Set the port and dev addr */
+	mdio_ctl = MDIO_CTL_PORT_ADDR(phy_id) | MDIO_CTL_DEV_ADDR(dev_addr);
+	xgmac_write32(mdio_ctl, &regs->mdio_ctl, endian);
+
+	/* Write the value to the register */
+	xgmac_write32(MDIO_DATA(value), &regs->mdio_data, endian);
+
+	ret = xgmac_wait_until_done(&bus->dev, regs, endian);
+	if (ret)
+		return ret;
+
+	return 0;
+}
+
+static int xgmac_mdio_write_c45(struct mii_bus *bus, int phy_id, int dev_addr,
+				int regnum, u16 value)
+{
+	struct mdio_fsl_priv *priv = (struct mdio_fsl_priv *)bus->priv;
+	struct tgec_mdio_controller __iomem *regs = priv->mdio_base;
 	bool endian = priv->is_little_endian;
+	u32 mdio_ctl, mdio_stat;
+	int ret;
 
 	mdio_stat = xgmac_read32(&regs->mdio_stat, endian);
-	if (regnum & MII_ADDR_C45) {
-		/* Clause 45 (ie 10G) */
-		dev_addr = (regnum >> 16) & 0x1f;
-		mdio_stat |= MDIO_STAT_ENC;
-	} else {
-		/* Clause 22 (ie 1G) */
-		dev_addr = regnum & 0x1f;
-		mdio_stat &= ~MDIO_STAT_ENC;
-	}
+	mdio_stat |= MDIO_STAT_ENC;
 
 	xgmac_write32(mdio_stat, &regs->mdio_stat, endian);
 
@@ -156,13 +183,11 @@ static int xgmac_mdio_write(struct mii_bus *bus, int phy_id, int regnum, u16 val
 	xgmac_write32(mdio_ctl, &regs->mdio_ctl, endian);
 
 	/* Set the register address */
-	if (regnum & MII_ADDR_C45) {
-		xgmac_write32(regnum & 0xffff, &regs->mdio_addr, endian);
+	xgmac_write32(regnum & 0xffff, &regs->mdio_addr, endian);
 
-		ret = xgmac_wait_until_free(&bus->dev, regs, endian);
-		if (ret)
-			return ret;
-	}
+	ret = xgmac_wait_until_free(&bus->dev, regs, endian);
+	if (ret)
+		return ret;
 
 	/* Write the value to the register */
 	xgmac_write32(MDIO_DATA(value), &regs->mdio_data, endian);
@@ -174,31 +199,82 @@ static int xgmac_mdio_write(struct mii_bus *bus, int phy_id, int regnum, u16 val
 	return 0;
 }
 
-/*
- * Reads from register regnum in the PHY for device dev, returning the value.
+/* Reads from register regnum in the PHY for device dev, returning the value.
  * Clears miimcom first.  All PHY configuration has to be done through the
  * TSEC1 MIIM regs.
  */
-static int xgmac_mdio_read(struct mii_bus *bus, int phy_id, int regnum)
+static int xgmac_mdio_read_c22(struct mii_bus *bus, int phy_id, int regnum)
 {
 	struct mdio_fsl_priv *priv = (struct mdio_fsl_priv *)bus->priv;
 	struct tgec_mdio_controller __iomem *regs = priv->mdio_base;
-	uint16_t dev_addr;
+	bool endian = priv->is_little_endian;
+	u16 dev_addr = regnum & 0x1f;
+	unsigned long flags;
 	uint32_t mdio_stat;
 	uint32_t mdio_ctl;
-	uint16_t value;
 	int ret;
-	bool endian = priv->is_little_endian;
 
 	mdio_stat = xgmac_read32(&regs->mdio_stat, endian);
-	if (regnum & MII_ADDR_C45) {
-		dev_addr = (regnum >> 16) & 0x1f;
-		mdio_stat |= MDIO_STAT_ENC;
+	mdio_stat &= ~MDIO_STAT_ENC;
+	xgmac_write32(mdio_stat, &regs->mdio_stat, endian);
+
+	ret = xgmac_wait_until_free(&bus->dev, regs, endian);
+	if (ret)
+		return ret;
+
+	/* Set the Port and Device Addrs */
+	mdio_ctl = MDIO_CTL_PORT_ADDR(phy_id) | MDIO_CTL_DEV_ADDR(dev_addr);
+	xgmac_write32(mdio_ctl, &regs->mdio_ctl, endian);
+
+	if (priv->has_a009885)
+		/* Once the operation completes, i.e. MDIO_STAT_BSY clears, we
+		 * must read back the data register within 16 MDC cycles.
+		 */
+		local_irq_save(flags);
+
+	/* Initiate the read */
+	xgmac_write32(mdio_ctl | MDIO_CTL_READ, &regs->mdio_ctl, endian);
+
+	ret = xgmac_wait_until_done(&bus->dev, regs, endian);
+	if (ret)
+		goto irq_restore;
+
+	/* Return all Fs if nothing was there */
+	if ((xgmac_read32(&regs->mdio_stat, endian) & MDIO_STAT_RD_ER) &&
+	    !priv->has_a011043) {
+		dev_dbg(&bus->dev,
+			"Error while reading PHY%d reg at %d.%d\n",
+			phy_id, dev_addr, regnum);
+		ret = 0xffff;
 	} else {
-		dev_addr = regnum & 0x1f;
-		mdio_stat &= ~MDIO_STAT_ENC;
+		ret = xgmac_read32(&regs->mdio_data, endian) & 0xffff;
+		dev_dbg(&bus->dev, "read %04x\n", ret);
 	}
 
+irq_restore:
+	if (priv->has_a009885)
+		local_irq_restore(flags);
+
+	return ret;
+}
+
+/* Reads from register regnum in the PHY for device dev, returning the value.
+ * Clears miimcom first.  All PHY configuration has to be done through the
+ * TSEC1 MIIM regs.
+ */
+static int xgmac_mdio_read_c45(struct mii_bus *bus, int phy_id, int dev_addr,
+			       int regnum)
+{
+	struct mdio_fsl_priv *priv = (struct mdio_fsl_priv *)bus->priv;
+	struct tgec_mdio_controller __iomem *regs = priv->mdio_base;
+	bool endian = priv->is_little_endian;
+	u32 mdio_stat, mdio_ctl;
+	unsigned long flags;
+	int ret;
+
+	mdio_stat = xgmac_read32(&regs->mdio_stat, endian);
+	mdio_stat |= MDIO_STAT_ENC;
+
 	xgmac_write32(mdio_stat, &regs->mdio_stat, endian);
 
 	ret = xgmac_wait_until_free(&bus->dev, regs, endian);
@@ -210,98 +286,157 @@ static int xgmac_mdio_read(struct mii_bus *bus, int phy_id, int regnum)
 	xgmac_write32(mdio_ctl, &regs->mdio_ctl, endian);
 
 	/* Set the register address */
-	if (regnum & MII_ADDR_C45) {
-		xgmac_write32(regnum & 0xffff, &regs->mdio_addr, endian);
+	xgmac_write32(regnum & 0xffff, &regs->mdio_addr, endian);
 
-		ret = xgmac_wait_until_free(&bus->dev, regs, endian);
-		if (ret)
-			return ret;
-	}
+	ret = xgmac_wait_until_free(&bus->dev, regs, endian);
+	if (ret)
+		return ret;
+
+	if (priv->has_a009885)
+		/* Once the operation completes, i.e. MDIO_STAT_BSY clears, we
+		 * must read back the data register within 16 MDC cycles.
+		 */
+		local_irq_save(flags);
 
 	/* Initiate the read */
 	xgmac_write32(mdio_ctl | MDIO_CTL_READ, &regs->mdio_ctl, endian);
 
 	ret = xgmac_wait_until_done(&bus->dev, regs, endian);
 	if (ret)
-		return ret;
+		goto irq_restore;
 
 	/* Return all Fs if nothing was there */
-	if (xgmac_read32(&regs->mdio_stat, endian) & MDIO_STAT_RD_ER) {
-		dev_err(&bus->dev,
-			"Error while reading PHY%d reg at %d.%hhu\n",
+	if ((xgmac_read32(&regs->mdio_stat, endian) & MDIO_STAT_RD_ER) &&
+	    !priv->has_a011043) {
+		dev_dbg(&bus->dev,
+			"Error while reading PHY%d reg at %d.%d\n",
 			phy_id, dev_addr, regnum);
-		return 0xffff;
+		ret = 0xffff;
+	} else {
+		ret = xgmac_read32(&regs->mdio_data, endian) & 0xffff;
+		dev_dbg(&bus->dev, "read %04x\n", ret);
+	}
+
+irq_restore:
+	if (priv->has_a009885)
+		local_irq_restore(flags);
+
+	return ret;
+}
+
+static int xgmac_mdio_set_mdc_freq(struct mii_bus *bus)
+{
+	struct mdio_fsl_priv *priv = (struct mdio_fsl_priv *)bus->priv;
+	struct tgec_mdio_controller __iomem *regs = priv->mdio_base;
+	struct device *dev = bus->parent;
+	u32 mdio_stat, div;
+
+	if (device_property_read_u32(dev, "clock-frequency", &priv->mdc_freq))
+		return 0;
+
+	priv->enet_clk = devm_clk_get(dev, NULL);
+	if (IS_ERR(priv->enet_clk)) {
+		dev_err(dev, "Input clock unknown, not changing MDC frequency");
+		return PTR_ERR(priv->enet_clk);
+	}
+
+	div = ((clk_get_rate(priv->enet_clk) / priv->mdc_freq) - 1) / 2;
+	if (div < 5 || div > 0x1ff) {
+		dev_err(dev, "Requested MDC frequency is out of range, ignoring");
+		return -EINVAL;
 	}
 
-	value = xgmac_read32(&regs->mdio_data, endian) & 0xffff;
-	dev_dbg(&bus->dev, "read %04x\n", value);
+	mdio_stat = xgmac_read32(&regs->mdio_stat, priv->is_little_endian);
+	mdio_stat &= ~MDIO_STAT_CLKDIV(0x1ff);
+	mdio_stat |= MDIO_STAT_CLKDIV(div);
+	xgmac_write32(mdio_stat, &regs->mdio_stat, priv->is_little_endian);
+	return 0;
+}
 
-	return value;
+static void xgmac_mdio_set_suppress_preamble(struct mii_bus *bus)
+{
+	struct mdio_fsl_priv *priv = (struct mdio_fsl_priv *)bus->priv;
+	struct tgec_mdio_controller __iomem *regs = priv->mdio_base;
+	struct device *dev = bus->parent;
+	u32 mdio_stat;
+
+	if (!device_property_read_bool(dev, "suppress-preamble"))
+		return;
+
+	mdio_stat = xgmac_read32(&regs->mdio_stat, priv->is_little_endian);
+	mdio_stat |= MDIO_STAT_PRE_DIS;
+	xgmac_write32(mdio_stat, &regs->mdio_stat, priv->is_little_endian);
 }
 
 static int xgmac_mdio_probe(struct platform_device *pdev)
 {
-	struct device_node *np = pdev->dev.of_node;
-	struct mii_bus *bus;
-	struct resource res;
+	struct fwnode_handle *fwnode;
 	struct mdio_fsl_priv *priv;
+	struct resource *res;
+	struct mii_bus *bus;
 	int ret;
 
-	ret = of_address_to_resource(np, 0, &res);
-	if (ret) {
+	/* In DPAA-1, MDIO is one of the many FMan sub-devices. The FMan
+	 * defines a register space that spans a large area, covering all the
+	 * subdevice areas. Therefore, MDIO cannot claim exclusive access to
+	 * this register area.
+	 */
+	res = platform_get_resource(pdev, IORESOURCE_MEM, 0);
+	if (!res) {
 		dev_err(&pdev->dev, "could not obtain address\n");
-		return ret;
+		return -EINVAL;
 	}
 
-	bus = mdiobus_alloc_size(sizeof(struct mdio_fsl_priv));
+	bus = devm_mdiobus_alloc_size(&pdev->dev, sizeof(struct mdio_fsl_priv));
 	if (!bus)
 		return -ENOMEM;
 
 	bus->name = "Freescale XGMAC MDIO Bus";
-	bus->read = xgmac_mdio_read;
-	bus->write = xgmac_mdio_write;
+	bus->read = xgmac_mdio_read_c22;
+	bus->write = xgmac_mdio_write_c22;
+	bus->read_c45 = xgmac_mdio_read_c45;
+	bus->write_c45 = xgmac_mdio_write_c45;
 	bus->parent = &pdev->dev;
-	snprintf(bus->id, MII_BUS_ID_SIZE, "%llx", (unsigned long long)res.start);
+	snprintf(bus->id, MII_BUS_ID_SIZE, "%pa", &res->start);
 
-	/* Set the PHY base address */
 	priv = bus->priv;
-	priv->mdio_base = of_iomap(np, 0);
-	if (!priv->mdio_base) {
-		ret = -ENOMEM;
-		goto err_ioremap;
-	}
+	priv->mdio_base = devm_ioremap(&pdev->dev, res->start,
+				       resource_size(res));
+	if (!priv->mdio_base)
+		return -ENOMEM;
+
+	/* For both ACPI and DT cases, endianness of MDIO controller
+	 * needs to be specified using "little-endian" property.
+	 */
+	priv->is_little_endian = device_property_read_bool(&pdev->dev,
+							   "little-endian");
 
-	priv->is_little_endian = of_property_read_bool(pdev->dev.of_node,
-						       "little-endian");
+	priv->has_a009885 = device_property_read_bool(&pdev->dev,
+						      "fsl,erratum-a009885");
+	priv->has_a011043 = device_property_read_bool(&pdev->dev,
+						      "fsl,erratum-a011043");
 
-	ret = of_mdiobus_register(bus, np);
+	xgmac_mdio_set_suppress_preamble(bus);
+
+	ret = xgmac_mdio_set_mdc_freq(bus);
+	if (ret)
+		return ret;
+
+	fwnode = dev_fwnode(&pdev->dev);
+	if (is_of_node(fwnode))
+		ret = of_mdiobus_register(bus, to_of_node(fwnode));
+	else if (is_acpi_node(fwnode))
+		ret = acpi_mdiobus_register(bus, fwnode);
+	else
+		ret = -EINVAL;
 	if (ret) {
 		dev_err(&pdev->dev, "cannot register MDIO bus\n");
-		goto err_registration;
+		return ret;
 	}
 
 	platform_set_drvdata(pdev, bus);
 
 	return 0;
-
-err_registration:
-	iounmap(priv->mdio_base);
-
-err_ioremap:
-	mdiobus_free(bus);
-
-	return ret;
-}
-
-static int xgmac_mdio_remove(struct platform_device *pdev)
-{
-	struct mii_bus *bus = platform_get_drvdata(pdev);
-
-	mdiobus_unregister(bus);
-	iounmap(bus->priv);
-	mdiobus_free(bus);
-
-	return 0;
 }
 
 static const struct of_device_id xgmac_mdio_match[] = {
@@ -315,13 +450,19 @@ static const struct of_device_id xgmac_mdio_match[] = {
 };
 MODULE_DEVICE_TABLE(of, xgmac_mdio_match);
 
+static const struct acpi_device_id xgmac_acpi_match[] = {
+	{ "NXP0006" },
+	{ }
+};
+MODULE_DEVICE_TABLE(acpi, xgmac_acpi_match);
+
 static struct platform_driver xgmac_mdio_driver = {
 	.driver = {
 		.name = "fsl-fman_xmdio",
 		.of_match_table = xgmac_mdio_match,
+		.acpi_match_table = xgmac_acpi_match,
 	},
 	.probe = xgmac_mdio_probe,
-	.remove = xgmac_mdio_remove,
 };
 
 module_platform_driver(xgmac_mdio_driver);
diff --git a/drivers/net/ethernet/fujitsu/Kconfig b/drivers/net/ethernet/fujitsu/Kconfig
index faee34e44a35..06a28bce5d27 100644
--- a/drivers/net/ethernet/fujitsu/Kconfig
+++ b/drivers/net/ethernet/fujitsu/Kconfig
@@ -1,3 +1,4 @@
+# SPDX-License-Identifier: GPL-2.0-only
 #
 # Fujitsu Network device configuration
 #
@@ -6,7 +7,7 @@ config NET_VENDOR_FUJITSU
 	bool "Fujitsu devices"
 	default y
 	depends on PCMCIA
-	---help---
+	help
 	  If you have a network (Ethernet) card belonging to this class, say Y.
 
 	  Note that the answer to this question doesn't directly affect the
@@ -17,9 +18,9 @@ if NET_VENDOR_FUJITSU
 
 config PCMCIA_FMVJ18X
 	tristate "Fujitsu FMV-J18x PCMCIA support"
-	depends on PCMCIA
+	depends on PCMCIA && HAS_IOPORT
 	select CRC32
-	---help---
+	help
 	  Say Y here if you intend to attach a Fujitsu FMV-J18x or compatible
 	  PCMCIA (PC-card) Ethernet card to your computer.
 
diff --git a/drivers/net/ethernet/fujitsu/Makefile b/drivers/net/ethernet/fujitsu/Makefile
index 21561fdcc69f..74feebbf4572 100644
--- a/drivers/net/ethernet/fujitsu/Makefile
+++ b/drivers/net/ethernet/fujitsu/Makefile
@@ -1,3 +1,4 @@
+# SPDX-License-Identifier: GPL-2.0-only
 #
 # Makefile for the Fujitsu network device drivers.
 #
diff --git a/drivers/net/ethernet/fujitsu/fmvj18x_cs.c b/drivers/net/ethernet/fujitsu/fmvj18x_cs.c
index a69cd19a55ae..4859493471db 100644
--- a/drivers/net/ethernet/fujitsu/fmvj18x_cs.c
+++ b/drivers/net/ethernet/fujitsu/fmvj18x_cs.c
@@ -93,7 +93,7 @@ static irqreturn_t fjn_interrupt(int irq, void *dev_id);
 static void fjn_rx(struct net_device *dev);
 static void fjn_reset(struct net_device *dev);
 static void set_rx_mode(struct net_device *dev);
-static void fjn_tx_timeout(struct net_device *dev);
+static void fjn_tx_timeout(struct net_device *dev, unsigned int txqueue);
 static const struct ethtool_ops netdev_ethtool_ops;
 
 /*
@@ -334,6 +334,7 @@ static int fmvj18x_config(struct pcmcia_device *link)
     u8 *buf;
     size_t len;
     u_char buggybuf[32];
+    u8 addr[ETH_ALEN];
 
     dev_dbg(&link->dev, "fmvj18x_config\n");
 
@@ -468,8 +469,7 @@ static int fmvj18x_config(struct pcmcia_device *link)
 		    goto failed;
 	    }
 	    /* Read MACID from CIS */
-	    for (i = 0; i < 6; i++)
-		    dev->dev_addr[i] = buf[i + 5];
+	    eth_hw_addr_set(dev, &buf[5]);
 	    kfree(buf);
 	} else {
 	    if (pcmcia_get_mac_from_cis(link, dev))
@@ -490,7 +490,8 @@ static int fmvj18x_config(struct pcmcia_device *link)
     case UNGERMANN:
 	/* Read MACID from register */
 	for (i = 0; i < 6; i++) 
-	    dev->dev_addr[i] = inb(ioaddr + UNGERMANN_MAC_ID + i);
+	    addr[i] = inb(ioaddr + UNGERMANN_MAC_ID + i);
+	eth_hw_addr_set(dev, addr);
 	card_name = "Access/CARD";
 	break;
     case XXX10304:
@@ -499,16 +500,15 @@ static int fmvj18x_config(struct pcmcia_device *link)
 	    pr_notice("unable to read hardware net address\n");
 	    goto failed;
 	}
-	for (i = 0 ; i < 6; i++) {
-	    dev->dev_addr[i] = buggybuf[i];
-	}
+	eth_hw_addr_set(dev, buggybuf);
 	card_name = "FMV-J182";
 	break;
     case MBH10302:
     default:
 	/* Read MACID from register */
 	for (i = 0; i < 6; i++) 
-	    dev->dev_addr[i] = inb(ioaddr + MAC_ID + i);
+	    addr[i] = inb(ioaddr + MAC_ID + i);
+	eth_hw_addr_set(dev, addr);
 	card_name = "FMV-J181";
 	break;
     }
@@ -547,6 +547,11 @@ static int fmvj18x_get_hwinfo(struct pcmcia_device *link, u_char *node_id)
 	return -1;
 
     base = ioremap(link->resource[2]->start, resource_size(link->resource[2]));
+    if (!base) {
+	pcmcia_release_window(link, link->resource[2]);
+	return -1;
+    }
+
     pcmcia_map_mem_page(link, link->resource[2], 0);
 
     /*
@@ -769,7 +774,7 @@ static irqreturn_t fjn_interrupt(int dummy, void *dev_id)
 
 /*====================================================================*/
 
-static void fjn_tx_timeout(struct net_device *dev)
+static void fjn_tx_timeout(struct net_device *dev, unsigned int txqueue)
 {
     struct local_info *lp = netdev_priv(dev);
     unsigned int ioaddr = dev->base_addr;
@@ -807,9 +812,9 @@ static netdev_tx_t fjn_start_xmit(struct sk_buff *skb,
     
     if (length < ETH_ZLEN)
     {
-    	if (skb_padto(skb, ETH_ZLEN))
-    		return NETDEV_TX_OK;
-    	length = ETH_ZLEN;
+	if (skb_padto(skb, ETH_ZLEN))
+		return NETDEV_TX_OK;
+	length = ETH_ZLEN;
     }
 
     netif_stop_queue(dev);
@@ -1041,8 +1046,8 @@ static void fjn_rx(struct net_device *dev)
 static void netdev_get_drvinfo(struct net_device *dev,
 			       struct ethtool_drvinfo *info)
 {
-	strlcpy(info->driver, DRV_NAME, sizeof(info->driver));
-	strlcpy(info->version, DRV_VERSION, sizeof(info->version));
+	strscpy(info->driver, DRV_NAME, sizeof(info->driver));
+	strscpy(info->version, DRV_VERSION, sizeof(info->version));
 	snprintf(info->bus_info, sizeof(info->bus_info),
 		"PCMCIA 0x%lx", dev->base_addr);
 }
diff --git a/drivers/net/ethernet/fungible/Kconfig b/drivers/net/ethernet/fungible/Kconfig
new file mode 100644
index 000000000000..1ecedecc0f6c
--- /dev/null
+++ b/drivers/net/ethernet/fungible/Kconfig
@@ -0,0 +1,28 @@
+# SPDX-License-Identifier: GPL-2.0-only
+#
+# Fungible network driver configuration
+#
+
+config NET_VENDOR_FUNGIBLE
+	bool "Fungible devices"
+	default y
+	help
+	  If you have a Fungible network device, say Y.
+
+	  Note that the answer to this question doesn't directly affect the
+	  kernel: saying N will just cause the configurator to skip all
+	  the questions about Fungible cards. If you say Y, you will be asked
+	  for your specific card in the following questions.
+
+if NET_VENDOR_FUNGIBLE
+
+config FUN_CORE
+	tristate
+	select SBITMAP
+	help
+	  A service module offering basic common services to Fungible
+	  device drivers.
+
+source "drivers/net/ethernet/fungible/funeth/Kconfig"
+
+endif # NET_VENDOR_FUNGIBLE
diff --git a/drivers/net/ethernet/fungible/Makefile b/drivers/net/ethernet/fungible/Makefile
new file mode 100644
index 000000000000..df759f1585a1
--- /dev/null
+++ b/drivers/net/ethernet/fungible/Makefile
@@ -0,0 +1,7 @@
+# SPDX-License-Identifier: (GPL-2.0-only OR BSD-3-Clause)
+#
+# Makefile for the Fungible network device drivers.
+#
+
+obj-$(CONFIG_FUN_CORE) += funcore/
+obj-$(CONFIG_FUN_ETH) += funeth/
diff --git a/drivers/net/ethernet/fungible/funcore/Makefile b/drivers/net/ethernet/fungible/funcore/Makefile
new file mode 100644
index 000000000000..bc16b264b53e
--- /dev/null
+++ b/drivers/net/ethernet/fungible/funcore/Makefile
@@ -0,0 +1,5 @@
+# SPDX-License-Identifier: (GPL-2.0-only OR BSD-3-Clause)
+
+obj-$(CONFIG_FUN_CORE) += funcore.o
+
+funcore-y := fun_dev.o fun_queue.o
diff --git a/drivers/net/ethernet/fungible/funcore/fun_dev.c b/drivers/net/ethernet/fungible/funcore/fun_dev.c
new file mode 100644
index 000000000000..ce97b76f9ae0
--- /dev/null
+++ b/drivers/net/ethernet/fungible/funcore/fun_dev.c
@@ -0,0 +1,833 @@
+// SPDX-License-Identifier: (GPL-2.0-only OR BSD-3-Clause)
+
+#include <linux/bitmap.h>
+#include <linux/delay.h>
+#include <linux/interrupt.h>
+#include <linux/io.h>
+#include <linux/io-64-nonatomic-lo-hi.h>
+#include <linux/mm.h>
+#include <linux/module.h>
+#include <linux/nvme.h>
+#include <linux/pci.h>
+#include <linux/wait.h>
+#include <linux/sched/signal.h>
+
+#include "fun_queue.h"
+#include "fun_dev.h"
+
+#define FUN_ADMIN_CMD_TO_MS 3000
+
+enum {
+	AQA_ASQS_SHIFT = 0,
+	AQA_ACQS_SHIFT = 16,
+	AQA_MIN_QUEUE_SIZE = 2,
+	AQA_MAX_QUEUE_SIZE = 4096
+};
+
+/* context for admin commands */
+struct fun_cmd_ctx {
+	fun_admin_callback_t cb;  /* callback to invoke on completion */
+	void *cb_data;            /* user data provided to callback */
+	int cpu;                  /* CPU where the cmd's tag was allocated */
+};
+
+/* Context for synchronous admin commands. */
+struct fun_sync_cmd_ctx {
+	struct completion compl;
+	u8 *rsp_buf;              /* caller provided response buffer */
+	unsigned int rsp_len;     /* response buffer size */
+	u8 rsp_status;            /* command response status */
+};
+
+/* Wait for the CSTS.RDY bit to match @enabled. */
+static int fun_wait_ready(struct fun_dev *fdev, bool enabled)
+{
+	unsigned int cap_to = NVME_CAP_TIMEOUT(fdev->cap_reg);
+	u32 bit = enabled ? NVME_CSTS_RDY : 0;
+	unsigned long deadline;
+
+	deadline = ((cap_to + 1) * HZ / 2) + jiffies; /* CAP.TO is in 500ms */
+
+	for (;;) {
+		u32 csts = readl(fdev->bar + NVME_REG_CSTS);
+
+		if (csts == ~0) {
+			dev_err(fdev->dev, "CSTS register read %#x\n", csts);
+			return -EIO;
+		}
+
+		if ((csts & NVME_CSTS_RDY) == bit)
+			return 0;
+
+		if (time_is_before_jiffies(deadline))
+			break;
+
+		msleep(100);
+	}
+
+	dev_err(fdev->dev,
+		"Timed out waiting for device to indicate RDY %u; aborting %s\n",
+		enabled, enabled ? "initialization" : "reset");
+	return -ETIMEDOUT;
+}
+
+/* Check CSTS and return an error if it is unreadable or has unexpected
+ * RDY value.
+ */
+static int fun_check_csts_rdy(struct fun_dev *fdev, unsigned int expected_rdy)
+{
+	u32 csts = readl(fdev->bar + NVME_REG_CSTS);
+	u32 actual_rdy = csts & NVME_CSTS_RDY;
+
+	if (csts == ~0) {
+		dev_err(fdev->dev, "CSTS register read %#x\n", csts);
+		return -EIO;
+	}
+	if (actual_rdy != expected_rdy) {
+		dev_err(fdev->dev, "Unexpected CSTS RDY %u\n", actual_rdy);
+		return -EINVAL;
+	}
+	return 0;
+}
+
+/* Check that CSTS RDY has the expected value. Then write a new value to the CC
+ * register and wait for CSTS RDY to match the new CC ENABLE state.
+ */
+static int fun_update_cc_enable(struct fun_dev *fdev, unsigned int initial_rdy)
+{
+	int rc = fun_check_csts_rdy(fdev, initial_rdy);
+
+	if (rc)
+		return rc;
+	writel(fdev->cc_reg, fdev->bar + NVME_REG_CC);
+	return fun_wait_ready(fdev, !!(fdev->cc_reg & NVME_CC_ENABLE));
+}
+
+static int fun_disable_ctrl(struct fun_dev *fdev)
+{
+	fdev->cc_reg &= ~(NVME_CC_SHN_MASK | NVME_CC_ENABLE);
+	return fun_update_cc_enable(fdev, 1);
+}
+
+static int fun_enable_ctrl(struct fun_dev *fdev, u32 admin_cqesz_log2,
+			   u32 admin_sqesz_log2)
+{
+	fdev->cc_reg = (admin_cqesz_log2 << NVME_CC_IOCQES_SHIFT) |
+		       (admin_sqesz_log2 << NVME_CC_IOSQES_SHIFT) |
+		       ((PAGE_SHIFT - 12) << NVME_CC_MPS_SHIFT) |
+		       NVME_CC_ENABLE;
+
+	return fun_update_cc_enable(fdev, 0);
+}
+
+static int fun_map_bars(struct fun_dev *fdev, const char *name)
+{
+	struct pci_dev *pdev = to_pci_dev(fdev->dev);
+	int err;
+
+	err = pci_request_mem_regions(pdev, name);
+	if (err) {
+		dev_err(&pdev->dev,
+			"Couldn't get PCI memory resources, err %d\n", err);
+		return err;
+	}
+
+	fdev->bar = pci_ioremap_bar(pdev, 0);
+	if (!fdev->bar) {
+		dev_err(&pdev->dev, "Couldn't map BAR 0\n");
+		pci_release_mem_regions(pdev);
+		return -ENOMEM;
+	}
+
+	return 0;
+}
+
+static void fun_unmap_bars(struct fun_dev *fdev)
+{
+	struct pci_dev *pdev = to_pci_dev(fdev->dev);
+
+	if (fdev->bar) {
+		iounmap(fdev->bar);
+		fdev->bar = NULL;
+		pci_release_mem_regions(pdev);
+	}
+}
+
+static int fun_set_dma_masks(struct device *dev)
+{
+	int err;
+
+	err = dma_set_mask_and_coherent(dev, DMA_BIT_MASK(64));
+	if (err)
+		dev_err(dev, "DMA mask configuration failed, err %d\n", err);
+	return err;
+}
+
+static irqreturn_t fun_admin_irq(int irq, void *data)
+{
+	struct fun_queue *funq = data;
+
+	return fun_process_cq(funq, 0) ? IRQ_HANDLED : IRQ_NONE;
+}
+
+static void fun_complete_admin_cmd(struct fun_queue *funq, void *data,
+				   void *entry, const struct fun_cqe_info *info)
+{
+	const struct fun_admin_rsp_common *rsp_common = entry;
+	struct fun_dev *fdev = funq->fdev;
+	struct fun_cmd_ctx *cmd_ctx;
+	int cpu;
+	u16 cid;
+
+	if (info->sqhd == cpu_to_be16(0xffff)) {
+		dev_dbg(fdev->dev, "adminq event");
+		if (fdev->adminq_cb)
+			fdev->adminq_cb(fdev, entry);
+		return;
+	}
+
+	cid = be16_to_cpu(rsp_common->cid);
+	dev_dbg(fdev->dev, "admin CQE cid %u, op %u, ret %u\n", cid,
+		rsp_common->op, rsp_common->ret);
+
+	cmd_ctx = &fdev->cmd_ctx[cid];
+	if (cmd_ctx->cpu < 0) {
+		dev_err(fdev->dev,
+			"admin CQE with CID=%u, op=%u does not match a pending command\n",
+			cid, rsp_common->op);
+		return;
+	}
+
+	if (cmd_ctx->cb)
+		cmd_ctx->cb(fdev, entry, xchg(&cmd_ctx->cb_data, NULL));
+
+	cpu = cmd_ctx->cpu;
+	cmd_ctx->cpu = -1;
+	sbitmap_queue_clear(&fdev->admin_sbq, cid, cpu);
+}
+
+static int fun_init_cmd_ctx(struct fun_dev *fdev, unsigned int ntags)
+{
+	unsigned int i;
+
+	fdev->cmd_ctx = kvcalloc(ntags, sizeof(*fdev->cmd_ctx), GFP_KERNEL);
+	if (!fdev->cmd_ctx)
+		return -ENOMEM;
+
+	for (i = 0; i < ntags; i++)
+		fdev->cmd_ctx[i].cpu = -1;
+
+	return 0;
+}
+
+/* Allocate and enable an admin queue and assign it the first IRQ vector. */
+static int fun_enable_admin_queue(struct fun_dev *fdev,
+				  const struct fun_dev_params *areq)
+{
+	struct fun_queue_alloc_req qreq = {
+		.cqe_size_log2 = areq->cqe_size_log2,
+		.sqe_size_log2 = areq->sqe_size_log2,
+		.cq_depth = areq->cq_depth,
+		.sq_depth = areq->sq_depth,
+		.rq_depth = areq->rq_depth,
+	};
+	unsigned int ntags = areq->sq_depth - 1;
+	struct fun_queue *funq;
+	int rc;
+
+	if (fdev->admin_q)
+		return -EEXIST;
+
+	if (areq->sq_depth < AQA_MIN_QUEUE_SIZE ||
+	    areq->sq_depth > AQA_MAX_QUEUE_SIZE ||
+	    areq->cq_depth < AQA_MIN_QUEUE_SIZE ||
+	    areq->cq_depth > AQA_MAX_QUEUE_SIZE)
+		return -EINVAL;
+
+	fdev->admin_q = fun_alloc_queue(fdev, 0, &qreq);
+	if (!fdev->admin_q)
+		return -ENOMEM;
+
+	rc = fun_init_cmd_ctx(fdev, ntags);
+	if (rc)
+		goto free_q;
+
+	rc = sbitmap_queue_init_node(&fdev->admin_sbq, ntags, -1, false,
+				     GFP_KERNEL, dev_to_node(fdev->dev));
+	if (rc)
+		goto free_cmd_ctx;
+
+	funq = fdev->admin_q;
+	funq->cq_vector = 0;
+	rc = fun_request_irq(funq, dev_name(fdev->dev), fun_admin_irq, funq);
+	if (rc)
+		goto free_sbq;
+
+	fun_set_cq_callback(funq, fun_complete_admin_cmd, NULL);
+	fdev->adminq_cb = areq->event_cb;
+
+	writel((funq->sq_depth - 1) << AQA_ASQS_SHIFT |
+	       (funq->cq_depth - 1) << AQA_ACQS_SHIFT,
+	       fdev->bar + NVME_REG_AQA);
+
+	writeq(funq->sq_dma_addr, fdev->bar + NVME_REG_ASQ);
+	writeq(funq->cq_dma_addr, fdev->bar + NVME_REG_ACQ);
+
+	rc = fun_enable_ctrl(fdev, areq->cqe_size_log2, areq->sqe_size_log2);
+	if (rc)
+		goto free_irq;
+
+	if (areq->rq_depth) {
+		rc = fun_create_rq(funq);
+		if (rc)
+			goto disable_ctrl;
+
+		funq_rq_post(funq);
+	}
+
+	return 0;
+
+disable_ctrl:
+	fun_disable_ctrl(fdev);
+free_irq:
+	fun_free_irq(funq);
+free_sbq:
+	sbitmap_queue_free(&fdev->admin_sbq);
+free_cmd_ctx:
+	kvfree(fdev->cmd_ctx);
+	fdev->cmd_ctx = NULL;
+free_q:
+	fun_free_queue(fdev->admin_q);
+	fdev->admin_q = NULL;
+	return rc;
+}
+
+static void fun_disable_admin_queue(struct fun_dev *fdev)
+{
+	struct fun_queue *admq = fdev->admin_q;
+
+	if (!admq)
+		return;
+
+	fun_disable_ctrl(fdev);
+
+	fun_free_irq(admq);
+	__fun_process_cq(admq, 0);
+
+	sbitmap_queue_free(&fdev->admin_sbq);
+
+	kvfree(fdev->cmd_ctx);
+	fdev->cmd_ctx = NULL;
+
+	fun_free_queue(admq);
+	fdev->admin_q = NULL;
+}
+
+/* Return %true if the admin queue has stopped servicing commands as can be
+ * detected through registers. This isn't exhaustive and may provide false
+ * negatives.
+ */
+static bool fun_adminq_stopped(struct fun_dev *fdev)
+{
+	u32 csts = readl(fdev->bar + NVME_REG_CSTS);
+
+	return (csts & (NVME_CSTS_CFS | NVME_CSTS_RDY)) != NVME_CSTS_RDY;
+}
+
+static int fun_wait_for_tag(struct fun_dev *fdev, int *cpup)
+{
+	struct sbitmap_queue *sbq = &fdev->admin_sbq;
+	struct sbq_wait_state *ws = &sbq->ws[0];
+	DEFINE_SBQ_WAIT(wait);
+	int tag;
+
+	for (;;) {
+		sbitmap_prepare_to_wait(sbq, ws, &wait, TASK_UNINTERRUPTIBLE);
+		if (fdev->suppress_cmds) {
+			tag = -ESHUTDOWN;
+			break;
+		}
+		tag = sbitmap_queue_get(sbq, cpup);
+		if (tag >= 0)
+			break;
+		schedule();
+	}
+
+	sbitmap_finish_wait(sbq, ws, &wait);
+	return tag;
+}
+
+/* Submit an asynchronous admin command. Caller is responsible for implementing
+ * any waiting or timeout. Upon command completion the callback @cb is called.
+ */
+int fun_submit_admin_cmd(struct fun_dev *fdev, struct fun_admin_req_common *cmd,
+			 fun_admin_callback_t cb, void *cb_data, bool wait_ok)
+{
+	struct fun_queue *funq = fdev->admin_q;
+	unsigned int cmdsize = cmd->len8 * 8;
+	struct fun_cmd_ctx *cmd_ctx;
+	int tag, cpu, rc = 0;
+
+	if (WARN_ON(cmdsize > (1 << funq->sqe_size_log2)))
+		return -EMSGSIZE;
+
+	tag = sbitmap_queue_get(&fdev->admin_sbq, &cpu);
+	if (tag < 0) {
+		if (!wait_ok)
+			return -EAGAIN;
+		tag = fun_wait_for_tag(fdev, &cpu);
+		if (tag < 0)
+			return tag;
+	}
+
+	cmd->cid = cpu_to_be16(tag);
+
+	cmd_ctx = &fdev->cmd_ctx[tag];
+	cmd_ctx->cb = cb;
+	cmd_ctx->cb_data = cb_data;
+
+	spin_lock(&funq->sq_lock);
+
+	if (unlikely(fdev->suppress_cmds)) {
+		rc = -ESHUTDOWN;
+		sbitmap_queue_clear(&fdev->admin_sbq, tag, cpu);
+	} else {
+		cmd_ctx->cpu = cpu;
+		memcpy(fun_sqe_at(funq, funq->sq_tail), cmd, cmdsize);
+
+		dev_dbg(fdev->dev, "admin cmd @ %u: %8ph\n", funq->sq_tail,
+			cmd);
+
+		if (++funq->sq_tail == funq->sq_depth)
+			funq->sq_tail = 0;
+		writel(funq->sq_tail, funq->sq_db);
+	}
+	spin_unlock(&funq->sq_lock);
+	return rc;
+}
+
+/* Abandon a pending admin command by clearing the issuer's callback data.
+ * Failure indicates that the command either has already completed or its
+ * completion is racing with this call.
+ */
+static bool fun_abandon_admin_cmd(struct fun_dev *fd,
+				  const struct fun_admin_req_common *cmd,
+				  void *cb_data)
+{
+	u16 cid = be16_to_cpu(cmd->cid);
+	struct fun_cmd_ctx *cmd_ctx = &fd->cmd_ctx[cid];
+
+	return cmpxchg(&cmd_ctx->cb_data, cb_data, NULL) == cb_data;
+}
+
+/* Stop submission of new admin commands and wake up any processes waiting for
+ * tags. Already submitted commands are left to complete or time out.
+ */
+static void fun_admin_stop(struct fun_dev *fdev)
+{
+	spin_lock(&fdev->admin_q->sq_lock);
+	fdev->suppress_cmds = true;
+	spin_unlock(&fdev->admin_q->sq_lock);
+	sbitmap_queue_wake_all(&fdev->admin_sbq);
+}
+
+/* The callback for synchronous execution of admin commands. It copies the
+ * command response to the caller's buffer and signals completion.
+ */
+static void fun_admin_cmd_sync_cb(struct fun_dev *fd, void *rsp, void *cb_data)
+{
+	const struct fun_admin_rsp_common *rsp_common = rsp;
+	struct fun_sync_cmd_ctx *ctx = cb_data;
+
+	if (!ctx)
+		return;         /* command issuer timed out and left */
+	if (ctx->rsp_buf) {
+		unsigned int rsp_len = rsp_common->len8 * 8;
+
+		if (unlikely(rsp_len > ctx->rsp_len)) {
+			dev_err(fd->dev,
+				"response for op %u is %uB > response buffer %uB\n",
+				rsp_common->op, rsp_len, ctx->rsp_len);
+			rsp_len = ctx->rsp_len;
+		}
+		memcpy(ctx->rsp_buf, rsp, rsp_len);
+	}
+	ctx->rsp_status = rsp_common->ret;
+	complete(&ctx->compl);
+}
+
+/* Submit a synchronous admin command. */
+int fun_submit_admin_sync_cmd(struct fun_dev *fdev,
+			      struct fun_admin_req_common *cmd, void *rsp,
+			      size_t rspsize, unsigned int timeout)
+{
+	struct fun_sync_cmd_ctx ctx = {
+		.compl = COMPLETION_INITIALIZER_ONSTACK(ctx.compl),
+		.rsp_buf = rsp,
+		.rsp_len = rspsize,
+	};
+	unsigned int cmdlen = cmd->len8 * 8;
+	unsigned long jiffies_left;
+	int ret;
+
+	ret = fun_submit_admin_cmd(fdev, cmd, fun_admin_cmd_sync_cb, &ctx,
+				   true);
+	if (ret)
+		return ret;
+
+	if (!timeout)
+		timeout = FUN_ADMIN_CMD_TO_MS;
+
+	jiffies_left = wait_for_completion_timeout(&ctx.compl,
+						   msecs_to_jiffies(timeout));
+	if (!jiffies_left) {
+		/* The command timed out. Attempt to cancel it so we can return.
+		 * But if the command is in the process of completing we'll
+		 * wait for it.
+		 */
+		if (fun_abandon_admin_cmd(fdev, cmd, &ctx)) {
+			dev_err(fdev->dev, "admin command timed out: %*ph\n",
+				cmdlen, cmd);
+			fun_admin_stop(fdev);
+			/* see if the timeout was due to a queue failure */
+			if (fun_adminq_stopped(fdev))
+				dev_err(fdev->dev,
+					"device does not accept admin commands\n");
+
+			return -ETIMEDOUT;
+		}
+		wait_for_completion(&ctx.compl);
+	}
+
+	if (ctx.rsp_status) {
+		dev_err(fdev->dev, "admin command failed, err %d: %*ph\n",
+			ctx.rsp_status, cmdlen, cmd);
+	}
+
+	return -ctx.rsp_status;
+}
+EXPORT_SYMBOL_GPL(fun_submit_admin_sync_cmd);
+
+/* Return the number of device resources of the requested type. */
+int fun_get_res_count(struct fun_dev *fdev, enum fun_admin_op res)
+{
+	union {
+		struct fun_admin_res_count_req req;
+		struct fun_admin_res_count_rsp rsp;
+	} cmd;
+	int rc;
+
+	cmd.req.common = FUN_ADMIN_REQ_COMMON_INIT2(res, sizeof(cmd.req));
+	cmd.req.count = FUN_ADMIN_SIMPLE_SUBOP_INIT(FUN_ADMIN_SUBOP_RES_COUNT,
+						    0, 0);
+
+	rc = fun_submit_admin_sync_cmd(fdev, &cmd.req.common, &cmd.rsp,
+				       sizeof(cmd), 0);
+	return rc ? rc : be32_to_cpu(cmd.rsp.count.data);
+}
+EXPORT_SYMBOL_GPL(fun_get_res_count);
+
+/* Request that the instance of resource @res with the given id be deleted. */
+int fun_res_destroy(struct fun_dev *fdev, enum fun_admin_op res,
+		    unsigned int flags, u32 id)
+{
+	struct fun_admin_generic_destroy_req req = {
+		.common = FUN_ADMIN_REQ_COMMON_INIT2(res, sizeof(req)),
+		.destroy = FUN_ADMIN_SIMPLE_SUBOP_INIT(FUN_ADMIN_SUBOP_DESTROY,
+						       flags, id)
+	};
+
+	return fun_submit_admin_sync_cmd(fdev, &req.common, NULL, 0, 0);
+}
+EXPORT_SYMBOL_GPL(fun_res_destroy);
+
+/* Bind two entities of the given types and IDs. */
+int fun_bind(struct fun_dev *fdev, enum fun_admin_bind_type type0,
+	     unsigned int id0, enum fun_admin_bind_type type1,
+	     unsigned int id1)
+{
+	DEFINE_RAW_FLEX(struct fun_admin_bind_req, cmd, entry, 2);
+
+	cmd->common = FUN_ADMIN_REQ_COMMON_INIT2(FUN_ADMIN_OP_BIND,
+						 __struct_size(cmd));
+	cmd->entry[0] = FUN_ADMIN_BIND_ENTRY_INIT(type0, id0);
+	cmd->entry[1] = FUN_ADMIN_BIND_ENTRY_INIT(type1, id1);
+
+	return fun_submit_admin_sync_cmd(fdev, &cmd->common, NULL, 0, 0);
+}
+EXPORT_SYMBOL_GPL(fun_bind);
+
+static int fun_get_dev_limits(struct fun_dev *fdev)
+{
+	struct pci_dev *pdev = to_pci_dev(fdev->dev);
+	unsigned int cq_count, sq_count, num_dbs;
+	int rc;
+
+	rc = fun_get_res_count(fdev, FUN_ADMIN_OP_EPCQ);
+	if (rc < 0)
+		return rc;
+	cq_count = rc;
+
+	rc = fun_get_res_count(fdev, FUN_ADMIN_OP_EPSQ);
+	if (rc < 0)
+		return rc;
+	sq_count = rc;
+
+	/* The admin queue consumes 1 CQ and at least 1 SQ. To be usable the
+	 * device must provide additional queues.
+	 */
+	if (cq_count < 2 || sq_count < 2 + !!fdev->admin_q->rq_depth)
+		return -EINVAL;
+
+	/* Calculate the max QID based on SQ/CQ/doorbell counts.
+	 * SQ/CQ doorbells alternate.
+	 */
+	num_dbs = (pci_resource_len(pdev, 0) - NVME_REG_DBS) >>
+		  (2 + NVME_CAP_STRIDE(fdev->cap_reg));
+	fdev->max_qid = min3(cq_count, sq_count, num_dbs / 2) - 1;
+	fdev->kern_end_qid = fdev->max_qid + 1;
+	return 0;
+}
+
+/* Allocate all MSI-X vectors available on a function and at least @min_vecs. */
+static int fun_alloc_irqs(struct pci_dev *pdev, unsigned int min_vecs)
+{
+	int vecs, num_msix = pci_msix_vec_count(pdev);
+
+	if (num_msix < 0)
+		return num_msix;
+	if (min_vecs > num_msix)
+		return -ERANGE;
+
+	vecs = pci_alloc_irq_vectors(pdev, min_vecs, num_msix, PCI_IRQ_MSIX);
+	if (vecs > 0) {
+		dev_info(&pdev->dev,
+			 "Allocated %d IRQ vectors of %d requested\n",
+			 vecs, num_msix);
+	} else {
+		dev_err(&pdev->dev,
+			"Unable to allocate at least %u IRQ vectors\n",
+			min_vecs);
+	}
+	return vecs;
+}
+
+/* Allocate and initialize the IRQ manager state. */
+static int fun_alloc_irq_mgr(struct fun_dev *fdev)
+{
+	fdev->irq_map = bitmap_zalloc(fdev->num_irqs, GFP_KERNEL);
+	if (!fdev->irq_map)
+		return -ENOMEM;
+
+	spin_lock_init(&fdev->irqmgr_lock);
+	/* mark IRQ 0 allocated, it is used by the admin queue */
+	__set_bit(0, fdev->irq_map);
+	fdev->irqs_avail = fdev->num_irqs - 1;
+	return 0;
+}
+
+/* Reserve @nirqs of the currently available IRQs and return their indices. */
+int fun_reserve_irqs(struct fun_dev *fdev, unsigned int nirqs, u16 *irq_indices)
+{
+	unsigned int b, n = 0;
+	int err = -ENOSPC;
+
+	if (!nirqs)
+		return 0;
+
+	spin_lock(&fdev->irqmgr_lock);
+	if (nirqs > fdev->irqs_avail)
+		goto unlock;
+
+	for_each_clear_bit(b, fdev->irq_map, fdev->num_irqs) {
+		__set_bit(b, fdev->irq_map);
+		irq_indices[n++] = b;
+		if (n >= nirqs)
+			break;
+	}
+
+	WARN_ON(n < nirqs);
+	fdev->irqs_avail -= n;
+	err = n;
+unlock:
+	spin_unlock(&fdev->irqmgr_lock);
+	return err;
+}
+EXPORT_SYMBOL(fun_reserve_irqs);
+
+/* Release @nirqs previously allocated IRQS with the supplied indices. */
+void fun_release_irqs(struct fun_dev *fdev, unsigned int nirqs,
+		      u16 *irq_indices)
+{
+	unsigned int i;
+
+	spin_lock(&fdev->irqmgr_lock);
+	for (i = 0; i < nirqs; i++)
+		__clear_bit(irq_indices[i], fdev->irq_map);
+	fdev->irqs_avail += nirqs;
+	spin_unlock(&fdev->irqmgr_lock);
+}
+EXPORT_SYMBOL(fun_release_irqs);
+
+static void fun_serv_handler(struct work_struct *work)
+{
+	struct fun_dev *fd = container_of(work, struct fun_dev, service_task);
+
+	if (test_bit(FUN_SERV_DISABLED, &fd->service_flags))
+		return;
+	if (fd->serv_cb)
+		fd->serv_cb(fd);
+}
+
+void fun_serv_stop(struct fun_dev *fd)
+{
+	set_bit(FUN_SERV_DISABLED, &fd->service_flags);
+	cancel_work_sync(&fd->service_task);
+}
+EXPORT_SYMBOL_GPL(fun_serv_stop);
+
+void fun_serv_restart(struct fun_dev *fd)
+{
+	clear_bit(FUN_SERV_DISABLED, &fd->service_flags);
+	if (fd->service_flags)
+		schedule_work(&fd->service_task);
+}
+EXPORT_SYMBOL_GPL(fun_serv_restart);
+
+void fun_serv_sched(struct fun_dev *fd)
+{
+	if (!test_bit(FUN_SERV_DISABLED, &fd->service_flags))
+		schedule_work(&fd->service_task);
+}
+EXPORT_SYMBOL_GPL(fun_serv_sched);
+
+/* Check and try to get the device into a proper state for initialization,
+ * i.e., CSTS.RDY = CC.EN = 0.
+ */
+static int sanitize_dev(struct fun_dev *fdev)
+{
+	int rc;
+
+	fdev->cap_reg = readq(fdev->bar + NVME_REG_CAP);
+	fdev->cc_reg = readl(fdev->bar + NVME_REG_CC);
+
+	/* First get RDY to agree with the current EN. Give RDY the opportunity
+	 * to complete a potential recent EN change.
+	 */
+	rc = fun_wait_ready(fdev, fdev->cc_reg & NVME_CC_ENABLE);
+	if (rc)
+		return rc;
+
+	/* Next, reset the device if EN is currently 1. */
+	if (fdev->cc_reg & NVME_CC_ENABLE)
+		rc = fun_disable_ctrl(fdev);
+
+	return rc;
+}
+
+/* Undo the device initialization of fun_dev_enable(). */
+void fun_dev_disable(struct fun_dev *fdev)
+{
+	struct pci_dev *pdev = to_pci_dev(fdev->dev);
+
+	pci_set_drvdata(pdev, NULL);
+
+	if (fdev->fw_handle != FUN_HCI_ID_INVALID) {
+		fun_res_destroy(fdev, FUN_ADMIN_OP_SWUPGRADE, 0,
+				fdev->fw_handle);
+		fdev->fw_handle = FUN_HCI_ID_INVALID;
+	}
+
+	fun_disable_admin_queue(fdev);
+
+	bitmap_free(fdev->irq_map);
+	pci_free_irq_vectors(pdev);
+
+	pci_disable_device(pdev);
+
+	fun_unmap_bars(fdev);
+}
+EXPORT_SYMBOL(fun_dev_disable);
+
+/* Perform basic initialization of a device, including
+ * - PCI config space setup and BAR0 mapping
+ * - interrupt management initialization
+ * - 1 admin queue setup
+ * - determination of some device limits, such as number of queues.
+ */
+int fun_dev_enable(struct fun_dev *fdev, struct pci_dev *pdev,
+		   const struct fun_dev_params *areq, const char *name)
+{
+	int rc;
+
+	fdev->dev = &pdev->dev;
+	rc = fun_map_bars(fdev, name);
+	if (rc)
+		return rc;
+
+	rc = fun_set_dma_masks(fdev->dev);
+	if (rc)
+		goto unmap;
+
+	rc = pci_enable_device_mem(pdev);
+	if (rc) {
+		dev_err(&pdev->dev, "Couldn't enable device, err %d\n", rc);
+		goto unmap;
+	}
+
+	rc = sanitize_dev(fdev);
+	if (rc)
+		goto disable_dev;
+
+	fdev->fw_handle = FUN_HCI_ID_INVALID;
+	fdev->q_depth = NVME_CAP_MQES(fdev->cap_reg) + 1;
+	fdev->db_stride = 1 << NVME_CAP_STRIDE(fdev->cap_reg);
+	fdev->dbs = fdev->bar + NVME_REG_DBS;
+
+	INIT_WORK(&fdev->service_task, fun_serv_handler);
+	fdev->service_flags = FUN_SERV_DISABLED;
+	fdev->serv_cb = areq->serv_cb;
+
+	rc = fun_alloc_irqs(pdev, areq->min_msix + 1); /* +1 for admin CQ */
+	if (rc < 0)
+		goto disable_dev;
+	fdev->num_irqs = rc;
+
+	rc = fun_alloc_irq_mgr(fdev);
+	if (rc)
+		goto free_irqs;
+
+	pci_set_master(pdev);
+	rc = fun_enable_admin_queue(fdev, areq);
+	if (rc)
+		goto free_irq_mgr;
+
+	rc = fun_get_dev_limits(fdev);
+	if (rc < 0)
+		goto disable_admin;
+
+	pci_save_state(pdev);
+	pci_set_drvdata(pdev, fdev);
+	pcie_print_link_status(pdev);
+	dev_dbg(fdev->dev, "q_depth %u, db_stride %u, max qid %d kern_end_qid %d\n",
+		fdev->q_depth, fdev->db_stride, fdev->max_qid,
+		fdev->kern_end_qid);
+	return 0;
+
+disable_admin:
+	fun_disable_admin_queue(fdev);
+free_irq_mgr:
+	bitmap_free(fdev->irq_map);
+free_irqs:
+	pci_free_irq_vectors(pdev);
+disable_dev:
+	pci_disable_device(pdev);
+unmap:
+	fun_unmap_bars(fdev);
+	return rc;
+}
+EXPORT_SYMBOL(fun_dev_enable);
+
+MODULE_AUTHOR("Dimitris Michailidis <dmichail@fungible.com>");
+MODULE_DESCRIPTION("Core services driver for Fungible devices");
+MODULE_LICENSE("Dual BSD/GPL");
diff --git a/drivers/net/ethernet/fungible/funcore/fun_dev.h b/drivers/net/ethernet/fungible/funcore/fun_dev.h
new file mode 100644
index 000000000000..9e8c17ce8887
--- /dev/null
+++ b/drivers/net/ethernet/fungible/funcore/fun_dev.h
@@ -0,0 +1,150 @@
+/* SPDX-License-Identifier: (GPL-2.0-only OR BSD-3-Clause) */
+
+#ifndef _FUNDEV_H
+#define _FUNDEV_H
+
+#include <linux/sbitmap.h>
+#include <linux/spinlock_types.h>
+#include <linux/workqueue.h>
+#include "fun_hci.h"
+
+struct pci_dev;
+struct fun_dev;
+struct fun_queue;
+struct fun_cmd_ctx;
+struct fun_queue_alloc_req;
+
+/* doorbell fields */
+enum {
+	FUN_DB_QIDX_S = 0,
+	FUN_DB_INTCOAL_ENTRIES_S = 16,
+	FUN_DB_INTCOAL_ENTRIES_M = 0x7f,
+	FUN_DB_INTCOAL_USEC_S = 23,
+	FUN_DB_INTCOAL_USEC_M = 0x7f,
+	FUN_DB_IRQ_S = 30,
+	FUN_DB_IRQ_F = 1 << FUN_DB_IRQ_S,
+	FUN_DB_IRQ_ARM_S = 31,
+	FUN_DB_IRQ_ARM_F = 1U << FUN_DB_IRQ_ARM_S
+};
+
+/* Callback for asynchronous admin commands.
+ * Invoked on reception of command response.
+ */
+typedef void (*fun_admin_callback_t)(struct fun_dev *fdev, void *rsp,
+				     void *cb_data);
+
+/* Callback for events/notifications received by an admin queue. */
+typedef void (*fun_admin_event_cb)(struct fun_dev *fdev, void *cqe);
+
+/* Callback for pending work handled by the service task. */
+typedef void (*fun_serv_cb)(struct fun_dev *fd);
+
+/* service task flags */
+enum {
+	FUN_SERV_DISABLED,    /* service task is disabled */
+	FUN_SERV_FIRST_AVAIL
+};
+
+/* Driver state associated with a PCI function. */
+struct fun_dev {
+	struct device *dev;
+
+	void __iomem *bar;            /* start of BAR0 mapping */
+	u32 __iomem *dbs;             /* start of doorbells in BAR0 mapping */
+
+	/* admin queue */
+	struct fun_queue *admin_q;
+	struct sbitmap_queue admin_sbq;
+	struct fun_cmd_ctx *cmd_ctx;
+	fun_admin_event_cb adminq_cb;
+	bool suppress_cmds;           /* if set don't write commands to SQ */
+
+	/* address increment between consecutive doorbells, in 4B units */
+	unsigned int db_stride;
+
+	/* SW versions of device registers */
+	u32 cc_reg;         /* CC register */
+	u64 cap_reg;        /* CAPability register */
+
+	unsigned int q_depth;    /* max queue depth supported by device */
+	unsigned int max_qid;    /* = #queues - 1, separately for SQs and CQs */
+	unsigned int kern_end_qid; /* last qid in the kernel range + 1 */
+
+	unsigned int fw_handle;
+
+	/* IRQ manager */
+	unsigned int num_irqs;
+	unsigned int irqs_avail;
+	spinlock_t irqmgr_lock;
+	unsigned long *irq_map;
+
+	/* The service task handles work that needs a process context */
+	struct work_struct service_task;
+	unsigned long service_flags;
+	fun_serv_cb serv_cb;
+};
+
+struct fun_dev_params {
+	u8  cqe_size_log2; /* admin q CQE size */
+	u8  sqe_size_log2; /* admin q SQE size */
+
+	/* admin q depths */
+	u16 cq_depth;
+	u16 sq_depth;
+	u16 rq_depth;
+
+	u16 min_msix; /* min vectors needed by requesting driver */
+
+	fun_admin_event_cb event_cb;
+	fun_serv_cb serv_cb;
+};
+
+/* Return the BAR address of a doorbell. */
+static inline u32 __iomem *fun_db_addr(const struct fun_dev *fdev,
+				       unsigned int db_index)
+{
+	return &fdev->dbs[db_index * fdev->db_stride];
+}
+
+/* Return the BAR address of an SQ doorbell. SQ and CQ DBs alternate,
+ * SQs have even DB indices.
+ */
+static inline u32 __iomem *fun_sq_db_addr(const struct fun_dev *fdev,
+					  unsigned int sqid)
+{
+	return fun_db_addr(fdev, sqid * 2);
+}
+
+static inline u32 __iomem *fun_cq_db_addr(const struct fun_dev *fdev,
+					  unsigned int cqid)
+{
+	return fun_db_addr(fdev, cqid * 2 + 1);
+}
+
+int fun_get_res_count(struct fun_dev *fdev, enum fun_admin_op res);
+int fun_res_destroy(struct fun_dev *fdev, enum fun_admin_op res,
+		    unsigned int flags, u32 id);
+int fun_bind(struct fun_dev *fdev, enum fun_admin_bind_type type0,
+	     unsigned int id0, enum fun_admin_bind_type type1,
+	     unsigned int id1);
+
+int fun_submit_admin_cmd(struct fun_dev *fdev, struct fun_admin_req_common *cmd,
+			 fun_admin_callback_t cb, void *cb_data, bool wait_ok);
+int fun_submit_admin_sync_cmd(struct fun_dev *fdev,
+			      struct fun_admin_req_common *cmd, void *rsp,
+			      size_t rspsize, unsigned int timeout);
+
+int fun_dev_enable(struct fun_dev *fdev, struct pci_dev *pdev,
+		   const struct fun_dev_params *areq, const char *name);
+void fun_dev_disable(struct fun_dev *fdev);
+
+int fun_reserve_irqs(struct fun_dev *fdev, unsigned int nirqs,
+		     u16 *irq_indices);
+void fun_release_irqs(struct fun_dev *fdev, unsigned int nirqs,
+		      u16 *irq_indices);
+
+void fun_serv_stop(struct fun_dev *fd);
+void fun_serv_restart(struct fun_dev *fd);
+void fun_serv_sched(struct fun_dev *fd);
+
+#endif /* _FUNDEV_H */
diff --git a/drivers/net/ethernet/fungible/funcore/fun_hci.h b/drivers/net/ethernet/fungible/funcore/fun_hci.h
new file mode 100644
index 000000000000..f21819670106
--- /dev/null
+++ b/drivers/net/ethernet/fungible/funcore/fun_hci.h
@@ -0,0 +1,1242 @@
+/* SPDX-License-Identifier: (GPL-2.0-only OR BSD-3-Clause) */
+
+#ifndef __FUN_HCI_H
+#define __FUN_HCI_H
+
+enum {
+	FUN_HCI_ID_INVALID = 0xffffffff,
+};
+
+enum fun_admin_op {
+	FUN_ADMIN_OP_BIND = 0x1,
+	FUN_ADMIN_OP_EPCQ = 0x11,
+	FUN_ADMIN_OP_EPSQ = 0x12,
+	FUN_ADMIN_OP_PORT = 0x13,
+	FUN_ADMIN_OP_ETH = 0x14,
+	FUN_ADMIN_OP_VI = 0x15,
+	FUN_ADMIN_OP_SWUPGRADE = 0x1f,
+	FUN_ADMIN_OP_RSS = 0x21,
+	FUN_ADMIN_OP_ADI = 0x25,
+	FUN_ADMIN_OP_KTLS = 0x26,
+};
+
+enum {
+	FUN_REQ_COMMON_FLAG_RSP = 0x1,
+	FUN_REQ_COMMON_FLAG_HEAD_WB = 0x2,
+	FUN_REQ_COMMON_FLAG_INT = 0x4,
+	FUN_REQ_COMMON_FLAG_CQE_IN_RQBUF = 0x8,
+};
+
+struct fun_admin_req_common {
+	__u8 op;
+	__u8 len8;
+	__be16 flags;
+	__u8 suboff8;
+	__u8 rsvd0;
+	__be16 cid;
+};
+
+#define FUN_ADMIN_REQ_COMMON_INIT(_op, _len8, _flags, _suboff8, _cid)       \
+	(struct fun_admin_req_common) {                                     \
+		.op = (_op), .len8 = (_len8), .flags = cpu_to_be16(_flags), \
+		.suboff8 = (_suboff8), .cid = cpu_to_be16(_cid),            \
+	}
+
+#define FUN_ADMIN_REQ_COMMON_INIT2(_op, _len)    \
+	(struct fun_admin_req_common) {          \
+		.op = (_op), .len8 = (_len) / 8, \
+	}
+
+struct fun_admin_rsp_common {
+	__u8 op;
+	__u8 len8;
+	__be16 flags;
+	__u8 suboff8;
+	__u8 ret;
+	__be16 cid;
+};
+
+struct fun_admin_write48_req {
+	__be64 key_to_data;
+};
+
+#define FUN_ADMIN_WRITE48_REQ_KEY_S 56U
+#define FUN_ADMIN_WRITE48_REQ_KEY_M 0xff
+#define FUN_ADMIN_WRITE48_REQ_KEY_P_NOSWAP(x) \
+	(((__u64)x) << FUN_ADMIN_WRITE48_REQ_KEY_S)
+
+#define FUN_ADMIN_WRITE48_REQ_DATA_S 0U
+#define FUN_ADMIN_WRITE48_REQ_DATA_M 0xffffffffffff
+#define FUN_ADMIN_WRITE48_REQ_DATA_P_NOSWAP(x) \
+	(((__u64)x) << FUN_ADMIN_WRITE48_REQ_DATA_S)
+
+#define FUN_ADMIN_WRITE48_REQ_INIT(key, data)                       \
+	(struct fun_admin_write48_req) {                            \
+		.key_to_data = cpu_to_be64(                         \
+			FUN_ADMIN_WRITE48_REQ_KEY_P_NOSWAP(key) |   \
+			FUN_ADMIN_WRITE48_REQ_DATA_P_NOSWAP(data)), \
+	}
+
+struct fun_admin_write48_rsp {
+	__be64 key_to_data;
+};
+
+struct fun_admin_read48_req {
+	__be64 key_pack;
+};
+
+#define FUN_ADMIN_READ48_REQ_KEY_S 56U
+#define FUN_ADMIN_READ48_REQ_KEY_M 0xff
+#define FUN_ADMIN_READ48_REQ_KEY_P_NOSWAP(x) \
+	(((__u64)x) << FUN_ADMIN_READ48_REQ_KEY_S)
+
+#define FUN_ADMIN_READ48_REQ_INIT(key)                                       \
+	(struct fun_admin_read48_req) {                                      \
+		.key_pack =                                                  \
+			cpu_to_be64(FUN_ADMIN_READ48_REQ_KEY_P_NOSWAP(key)), \
+	}
+
+struct fun_admin_read48_rsp {
+	__be64 key_to_data;
+};
+
+#define FUN_ADMIN_READ48_RSP_KEY_S 56U
+#define FUN_ADMIN_READ48_RSP_KEY_M 0xff
+#define FUN_ADMIN_READ48_RSP_KEY_G(x)                     \
+	((be64_to_cpu(x) >> FUN_ADMIN_READ48_RSP_KEY_S) & \
+	 FUN_ADMIN_READ48_RSP_KEY_M)
+
+#define FUN_ADMIN_READ48_RSP_RET_S 48U
+#define FUN_ADMIN_READ48_RSP_RET_M 0xff
+#define FUN_ADMIN_READ48_RSP_RET_G(x)                     \
+	((be64_to_cpu(x) >> FUN_ADMIN_READ48_RSP_RET_S) & \
+	 FUN_ADMIN_READ48_RSP_RET_M)
+
+#define FUN_ADMIN_READ48_RSP_DATA_S 0U
+#define FUN_ADMIN_READ48_RSP_DATA_M 0xffffffffffff
+#define FUN_ADMIN_READ48_RSP_DATA_G(x)                     \
+	((be64_to_cpu(x) >> FUN_ADMIN_READ48_RSP_DATA_S) & \
+	 FUN_ADMIN_READ48_RSP_DATA_M)
+
+enum fun_admin_bind_type {
+	FUN_ADMIN_BIND_TYPE_EPCQ = 0x1,
+	FUN_ADMIN_BIND_TYPE_EPSQ = 0x2,
+	FUN_ADMIN_BIND_TYPE_PORT = 0x3,
+	FUN_ADMIN_BIND_TYPE_RSS = 0x4,
+	FUN_ADMIN_BIND_TYPE_VI = 0x5,
+	FUN_ADMIN_BIND_TYPE_ETH = 0x6,
+};
+
+struct fun_admin_bind_entry {
+	__u8 type;
+	__u8 rsvd0[3];
+	__be32 id;
+};
+
+#define FUN_ADMIN_BIND_ENTRY_INIT(_type, _id)            \
+	(struct fun_admin_bind_entry) {                  \
+		.type = (_type), .id = cpu_to_be32(_id), \
+	}
+
+struct fun_admin_bind_req {
+	struct fun_admin_req_common common;
+	struct fun_admin_bind_entry entry[];
+};
+
+struct fun_admin_bind_rsp {
+	struct fun_admin_rsp_common bind_rsp_common;
+};
+
+struct fun_admin_simple_subop {
+	__u8 subop;
+	__u8 rsvd0;
+	__be16 flags;
+	__be32 data;
+};
+
+#define FUN_ADMIN_SIMPLE_SUBOP_INIT(_subop, _flags, _data)       \
+	(struct fun_admin_simple_subop) {                        \
+		.subop = (_subop), .flags = cpu_to_be16(_flags), \
+		.data = cpu_to_be32(_data),                      \
+	}
+
+enum fun_admin_subop {
+	FUN_ADMIN_SUBOP_CREATE = 0x10,
+	FUN_ADMIN_SUBOP_DESTROY = 0x11,
+	FUN_ADMIN_SUBOP_MODIFY = 0x12,
+	FUN_ADMIN_SUBOP_RES_COUNT = 0x14,
+	FUN_ADMIN_SUBOP_READ = 0x15,
+	FUN_ADMIN_SUBOP_WRITE = 0x16,
+	FUN_ADMIN_SUBOP_NOTIFY = 0x17,
+};
+
+enum {
+	FUN_ADMIN_RES_CREATE_FLAG_ALLOCATOR = 0x1,
+};
+
+struct fun_admin_generic_destroy_req {
+	struct fun_admin_req_common common;
+	struct fun_admin_simple_subop destroy;
+};
+
+struct fun_admin_generic_create_rsp {
+	struct fun_admin_rsp_common common;
+
+	__u8 subop;
+	__u8 rsvd0;
+	__be16 flags;
+	__be32 id;
+};
+
+struct fun_admin_res_count_req {
+	struct fun_admin_req_common common;
+	struct fun_admin_simple_subop count;
+};
+
+struct fun_admin_res_count_rsp {
+	struct fun_admin_rsp_common common;
+	struct fun_admin_simple_subop count;
+};
+
+enum {
+	FUN_ADMIN_EPCQ_CREATE_FLAG_INT_EPCQ = 0x2,
+	FUN_ADMIN_EPCQ_CREATE_FLAG_ENTRY_WR_TPH = 0x4,
+	FUN_ADMIN_EPCQ_CREATE_FLAG_SL_WR_TPH = 0x8,
+	FUN_ADMIN_EPCQ_CREATE_FLAG_RQ = 0x80,
+	FUN_ADMIN_EPCQ_CREATE_FLAG_INT_IQ = 0x100,
+	FUN_ADMIN_EPCQ_CREATE_FLAG_INT_NOARM = 0x200,
+	FUN_ADMIN_EPCQ_CREATE_FLAG_DROP_ON_OVERFLOW = 0x400,
+};
+
+struct fun_admin_epcq_req {
+	struct fun_admin_req_common common;
+	union epcq_req_subop {
+		struct fun_admin_epcq_create_req {
+			__u8 subop;
+			__u8 rsvd0;
+			__be16 flags;
+			__be32 id;
+
+			__be32 epsqid;
+			__u8 rsvd1;
+			__u8 entry_size_log2;
+			__be16 nentries;
+
+			__be64 address;
+
+			__be16 tailroom; /* per packet tailroom in bytes */
+			__u8 headroom; /* per packet headroom in 2B units */
+			__u8 intcoal_kbytes;
+			__u8 intcoal_holdoff_nentries;
+			__u8 intcoal_holdoff_usecs;
+			__be16 intid;
+
+			__be32 scan_start_id;
+			__be32 scan_end_id;
+
+			__be16 tph_cpuid;
+			__u8 rsvd3[6];
+		} create;
+
+		struct fun_admin_epcq_modify_req {
+			__u8 subop;
+			__u8 rsvd0;
+			__be16 flags;
+			__be32 id;
+
+			__be16 headroom; /* headroom in bytes */
+			__u8 rsvd1[6];
+		} modify;
+	} u;
+};
+
+#define FUN_ADMIN_EPCQ_CREATE_REQ_INIT(                                      \
+	_subop, _flags, _id, _epsqid, _entry_size_log2, _nentries, _address, \
+	_tailroom, _headroom, _intcoal_kbytes, _intcoal_holdoff_nentries,    \
+	_intcoal_holdoff_usecs, _intid, _scan_start_id, _scan_end_id,        \
+	_tph_cpuid)                                                          \
+	(struct fun_admin_epcq_create_req) {                                 \
+		.subop = (_subop), .flags = cpu_to_be16(_flags),             \
+		.id = cpu_to_be32(_id), .epsqid = cpu_to_be32(_epsqid),      \
+		.entry_size_log2 = _entry_size_log2,                         \
+		.nentries = cpu_to_be16(_nentries),                          \
+		.address = cpu_to_be64(_address),                            \
+		.tailroom = cpu_to_be16(_tailroom), .headroom = _headroom,   \
+		.intcoal_kbytes = _intcoal_kbytes,                           \
+		.intcoal_holdoff_nentries = _intcoal_holdoff_nentries,       \
+		.intcoal_holdoff_usecs = _intcoal_holdoff_usecs,             \
+		.intid = cpu_to_be16(_intid),                                \
+		.scan_start_id = cpu_to_be32(_scan_start_id),                \
+		.scan_end_id = cpu_to_be32(_scan_end_id),                    \
+		.tph_cpuid = cpu_to_be16(_tph_cpuid),                        \
+	}
+
+#define FUN_ADMIN_EPCQ_MODIFY_REQ_INIT(_subop, _flags, _id, _headroom)      \
+	(struct fun_admin_epcq_modify_req) {                                \
+		.subop = (_subop), .flags = cpu_to_be16(_flags),            \
+		.id = cpu_to_be32(_id), .headroom = cpu_to_be16(_headroom), \
+	}
+
+enum {
+	FUN_ADMIN_EPSQ_CREATE_FLAG_INT_EPSQ = 0x2,
+	FUN_ADMIN_EPSQ_CREATE_FLAG_ENTRY_RD_TPH = 0x4,
+	FUN_ADMIN_EPSQ_CREATE_FLAG_GL_RD_TPH = 0x8,
+	FUN_ADMIN_EPSQ_CREATE_FLAG_HEAD_WB_ADDRESS = 0x10,
+	FUN_ADMIN_EPSQ_CREATE_FLAG_HEAD_WB_ADDRESS_TPH = 0x20,
+	FUN_ADMIN_EPSQ_CREATE_FLAG_HEAD_WB_EPCQ = 0x40,
+	FUN_ADMIN_EPSQ_CREATE_FLAG_RQ = 0x80,
+	FUN_ADMIN_EPSQ_CREATE_FLAG_INT_IQ = 0x100,
+	FUN_ADMIN_EPSQ_CREATE_FLAG_NO_CMPL = 0x200,
+};
+
+struct fun_admin_epsq_req {
+	struct fun_admin_req_common common;
+
+	union epsq_req_subop {
+		struct fun_admin_epsq_create_req {
+			__u8 subop;
+			__u8 rsvd0;
+			__be16 flags;
+			__be32 id;
+
+			__be32 epcqid;
+			__u8 rsvd1;
+			__u8 entry_size_log2;
+			__be16 nentries;
+
+			__be64 address; /* DMA address of epsq */
+
+			__u8 rsvd2[3];
+			__u8 intcoal_kbytes;
+			__u8 intcoal_holdoff_nentries;
+			__u8 intcoal_holdoff_usecs;
+			__be16 intid;
+
+			__be32 scan_start_id;
+			__be32 scan_end_id;
+
+			__u8 rsvd3[4];
+			__be16 tph_cpuid;
+			__u8 buf_size_log2; /* log2 of RQ buffer size */
+			__u8 head_wb_size_log2; /* log2 of head write back size */
+
+			__be64 head_wb_address; /* DMA address for head writeback */
+		} create;
+	} u;
+};
+
+#define FUN_ADMIN_EPSQ_CREATE_REQ_INIT(                                      \
+	_subop, _flags, _id, _epcqid, _entry_size_log2, _nentries, _address, \
+	_intcoal_kbytes, _intcoal_holdoff_nentries, _intcoal_holdoff_usecs,  \
+	_intid, _scan_start_id, _scan_end_id, _tph_cpuid, _buf_size_log2,    \
+	_head_wb_size_log2, _head_wb_address)                                \
+	(struct fun_admin_epsq_create_req) {                                 \
+		.subop = (_subop), .flags = cpu_to_be16(_flags),             \
+		.id = cpu_to_be32(_id), .epcqid = cpu_to_be32(_epcqid),      \
+		.entry_size_log2 = _entry_size_log2,                         \
+		.nentries = cpu_to_be16(_nentries),                          \
+		.address = cpu_to_be64(_address),                            \
+		.intcoal_kbytes = _intcoal_kbytes,                           \
+		.intcoal_holdoff_nentries = _intcoal_holdoff_nentries,       \
+		.intcoal_holdoff_usecs = _intcoal_holdoff_usecs,             \
+		.intid = cpu_to_be16(_intid),                                \
+		.scan_start_id = cpu_to_be32(_scan_start_id),                \
+		.scan_end_id = cpu_to_be32(_scan_end_id),                    \
+		.tph_cpuid = cpu_to_be16(_tph_cpuid),                        \
+		.buf_size_log2 = _buf_size_log2,                             \
+		.head_wb_size_log2 = _head_wb_size_log2,                     \
+		.head_wb_address = cpu_to_be64(_head_wb_address),            \
+	}
+
+enum {
+	FUN_PORT_CAP_OFFLOADS = 0x1,
+	FUN_PORT_CAP_STATS = 0x2,
+	FUN_PORT_CAP_LOOPBACK = 0x4,
+	FUN_PORT_CAP_VPORT = 0x8,
+	FUN_PORT_CAP_TX_PAUSE = 0x10,
+	FUN_PORT_CAP_RX_PAUSE = 0x20,
+	FUN_PORT_CAP_AUTONEG = 0x40,
+	FUN_PORT_CAP_RSS = 0x80,
+	FUN_PORT_CAP_VLAN_OFFLOADS = 0x100,
+	FUN_PORT_CAP_ENCAP_OFFLOADS = 0x200,
+	FUN_PORT_CAP_1000_X = 0x1000,
+	FUN_PORT_CAP_10G_R = 0x2000,
+	FUN_PORT_CAP_40G_R4 = 0x4000,
+	FUN_PORT_CAP_25G_R = 0x8000,
+	FUN_PORT_CAP_50G_R2 = 0x10000,
+	FUN_PORT_CAP_50G_R = 0x20000,
+	FUN_PORT_CAP_100G_R4 = 0x40000,
+	FUN_PORT_CAP_100G_R2 = 0x80000,
+	FUN_PORT_CAP_200G_R4 = 0x100000,
+	FUN_PORT_CAP_FEC_NONE = 0x10000000,
+	FUN_PORT_CAP_FEC_FC = 0x20000000,
+	FUN_PORT_CAP_FEC_RS = 0x40000000,
+};
+
+enum fun_port_brkout_mode {
+	FUN_PORT_BRKMODE_NA = 0x0,
+	FUN_PORT_BRKMODE_NONE = 0x1,
+	FUN_PORT_BRKMODE_2X = 0x2,
+	FUN_PORT_BRKMODE_4X = 0x3,
+};
+
+enum {
+	FUN_PORT_SPEED_AUTO = 0x0,
+	FUN_PORT_SPEED_10M = 0x1,
+	FUN_PORT_SPEED_100M = 0x2,
+	FUN_PORT_SPEED_1G = 0x4,
+	FUN_PORT_SPEED_10G = 0x8,
+	FUN_PORT_SPEED_25G = 0x10,
+	FUN_PORT_SPEED_40G = 0x20,
+	FUN_PORT_SPEED_50G = 0x40,
+	FUN_PORT_SPEED_100G = 0x80,
+	FUN_PORT_SPEED_200G = 0x100,
+};
+
+enum fun_port_duplex_mode {
+	FUN_PORT_FULL_DUPLEX = 0x0,
+	FUN_PORT_HALF_DUPLEX = 0x1,
+};
+
+enum {
+	FUN_PORT_FEC_NA = 0x0,
+	FUN_PORT_FEC_OFF = 0x1,
+	FUN_PORT_FEC_RS = 0x2,
+	FUN_PORT_FEC_FC = 0x4,
+	FUN_PORT_FEC_AUTO = 0x8,
+};
+
+enum fun_port_link_status {
+	FUN_PORT_LINK_UP = 0x0,
+	FUN_PORT_LINK_UP_WITH_ERR = 0x1,
+	FUN_PORT_LINK_DOWN = 0x2,
+};
+
+enum fun_port_led_type {
+	FUN_PORT_LED_OFF = 0x0,
+	FUN_PORT_LED_AMBER = 0x1,
+	FUN_PORT_LED_GREEN = 0x2,
+	FUN_PORT_LED_BEACON_ON = 0x3,
+	FUN_PORT_LED_BEACON_OFF = 0x4,
+};
+
+enum {
+	FUN_PORT_FLAG_MAC_DOWN = 0x1,
+	FUN_PORT_FLAG_MAC_UP = 0x2,
+	FUN_PORT_FLAG_NH_DOWN = 0x4,
+	FUN_PORT_FLAG_NH_UP = 0x8,
+};
+
+enum {
+	FUN_PORT_FLAG_ENABLE_NOTIFY = 0x1,
+};
+
+enum fun_port_lane_attr {
+	FUN_PORT_LANE_1 = 0x1,
+	FUN_PORT_LANE_2 = 0x2,
+	FUN_PORT_LANE_4 = 0x4,
+	FUN_PORT_LANE_SPEED_10G = 0x100,
+	FUN_PORT_LANE_SPEED_25G = 0x200,
+	FUN_PORT_LANE_SPEED_50G = 0x400,
+	FUN_PORT_LANE_SPLIT = 0x8000,
+};
+
+enum fun_admin_port_subop {
+	FUN_ADMIN_PORT_SUBOP_XCVR_READ = 0x23,
+	FUN_ADMIN_PORT_SUBOP_INETADDR_EVENT = 0x24,
+};
+
+enum fun_admin_port_key {
+	FUN_ADMIN_PORT_KEY_ILLEGAL = 0x0,
+	FUN_ADMIN_PORT_KEY_MTU = 0x1,
+	FUN_ADMIN_PORT_KEY_FEC = 0x2,
+	FUN_ADMIN_PORT_KEY_SPEED = 0x3,
+	FUN_ADMIN_PORT_KEY_DEBOUNCE = 0x4,
+	FUN_ADMIN_PORT_KEY_DUPLEX = 0x5,
+	FUN_ADMIN_PORT_KEY_MACADDR = 0x6,
+	FUN_ADMIN_PORT_KEY_LINKMODE = 0x7,
+	FUN_ADMIN_PORT_KEY_BREAKOUT = 0x8,
+	FUN_ADMIN_PORT_KEY_ENABLE = 0x9,
+	FUN_ADMIN_PORT_KEY_DISABLE = 0xa,
+	FUN_ADMIN_PORT_KEY_ERR_DISABLE = 0xb,
+	FUN_ADMIN_PORT_KEY_CAPABILITIES = 0xc,
+	FUN_ADMIN_PORT_KEY_LP_CAPABILITIES = 0xd,
+	FUN_ADMIN_PORT_KEY_STATS_DMA_LOW = 0xe,
+	FUN_ADMIN_PORT_KEY_STATS_DMA_HIGH = 0xf,
+	FUN_ADMIN_PORT_KEY_LANE_ATTRS = 0x10,
+	FUN_ADMIN_PORT_KEY_LED = 0x11,
+	FUN_ADMIN_PORT_KEY_ADVERT = 0x12,
+};
+
+struct fun_subop_imm {
+	__u8 subop; /* see fun_data_subop enum */
+	__u8 flags;
+	__u8 nsgl;
+	__u8 rsvd0;
+	__be32 len;
+
+	__u8 data[];
+};
+
+enum fun_subop_sgl_flags {
+	FUN_SUBOP_SGL_USE_OFF8 = 0x1,
+	FUN_SUBOP_FLAG_FREE_BUF = 0x2,
+	FUN_SUBOP_FLAG_IS_REFBUF = 0x4,
+	FUN_SUBOP_SGL_FLAG_LOCAL = 0x8,
+};
+
+enum fun_data_op {
+	FUN_DATAOP_INVALID = 0x0,
+	FUN_DATAOP_SL = 0x1, /* scatter */
+	FUN_DATAOP_GL = 0x2, /* gather */
+	FUN_DATAOP_SGL = 0x3, /* scatter-gather */
+	FUN_DATAOP_IMM = 0x4, /* immediate data */
+	FUN_DATAOP_RQBUF = 0x8, /* rq buffer */
+};
+
+struct fun_dataop_gl {
+	__u8 subop;
+	__u8 flags;
+	__be16 sgl_off;
+	__be32 sgl_len;
+
+	__be64 sgl_data;
+};
+
+static inline void fun_dataop_gl_init(struct fun_dataop_gl *s, u8 flags,
+				      u16 sgl_off, u32 sgl_len, u64 sgl_data)
+{
+	s->subop = FUN_DATAOP_GL;
+	s->flags = flags;
+	s->sgl_off = cpu_to_be16(sgl_off);
+	s->sgl_len = cpu_to_be32(sgl_len);
+	s->sgl_data = cpu_to_be64(sgl_data);
+}
+
+struct fun_dataop_imm {
+	__u8 subop;
+	__u8 flags;
+	__be16 rsvd0;
+	__be32 sgl_len;
+};
+
+struct fun_subop_sgl {
+	__u8 subop;
+	__u8 flags;
+	__u8 nsgl;
+	__u8 rsvd0;
+	__be32 sgl_len;
+
+	__be64 sgl_data;
+};
+
+#define FUN_SUBOP_SGL_INIT(_subop, _flags, _nsgl, _sgl_len, _sgl_data) \
+	(struct fun_subop_sgl) {                                       \
+		.subop = (_subop), .flags = (_flags), .nsgl = (_nsgl), \
+		.sgl_len = cpu_to_be32(_sgl_len),                      \
+		.sgl_data = cpu_to_be64(_sgl_data),                    \
+	}
+
+struct fun_dataop_rqbuf {
+	__u8 subop;
+	__u8 rsvd0;
+	__be16 cid;
+	__be32 bufoff;
+};
+
+struct fun_dataop_hdr {
+	__u8 nsgl;
+	__u8 flags;
+	__u8 ngather;
+	__u8 nscatter;
+	__be32 total_len;
+
+	struct fun_dataop_imm imm[];
+};
+
+#define FUN_DATAOP_HDR_INIT(_nsgl, _flags, _ngather, _nscatter, _total_len)  \
+	(struct fun_dataop_hdr) {                                            \
+		.nsgl = _nsgl, .flags = _flags, .ngather = _ngather,         \
+		.nscatter = _nscatter, .total_len = cpu_to_be32(_total_len), \
+	}
+
+enum fun_port_inetaddr_event_type {
+	FUN_PORT_INETADDR_ADD = 0x1,
+	FUN_PORT_INETADDR_DEL = 0x2,
+};
+
+enum fun_port_inetaddr_addr_family {
+	FUN_PORT_INETADDR_IPV4 = 0x1,
+	FUN_PORT_INETADDR_IPV6 = 0x2,
+};
+
+struct fun_admin_port_req {
+	struct fun_admin_req_common common;
+
+	union port_req_subop {
+		struct fun_admin_port_create_req {
+			__u8 subop;
+			__u8 rsvd0;
+			__be16 flags;
+			__be32 id;
+		} create;
+		struct fun_admin_port_write_req {
+			__u8 subop;
+			__u8 rsvd0;
+			__be16 flags;
+			__be32 id; /* portid */
+
+			struct fun_admin_write48_req write48[];
+		} write;
+		struct fun_admin_port_read_req {
+			__u8 subop;
+			__u8 rsvd0;
+			__be16 flags;
+			__be32 id; /* portid */
+
+			struct fun_admin_read48_req read48[];
+		} read;
+		struct fun_admin_port_xcvr_read_req {
+			u8 subop;
+			u8 rsvd0;
+			__be16 flags;
+			__be32 id;
+
+			u8 bank;
+			u8 page;
+			u8 offset;
+			u8 length;
+			u8 dev_addr;
+			u8 rsvd1[3];
+		} xcvr_read;
+		struct fun_admin_port_inetaddr_event_req {
+			__u8 subop;
+			__u8 rsvd0;
+			__u8 event_type;
+			__u8 addr_family;
+			__be32 id;
+
+			__u8 addr[];
+		} inetaddr_event;
+	} u;
+};
+
+#define FUN_ADMIN_PORT_CREATE_REQ_INIT(_subop, _flags, _id)      \
+	(struct fun_admin_port_create_req) {                     \
+		.subop = (_subop), .flags = cpu_to_be16(_flags), \
+		.id = cpu_to_be32(_id),                          \
+	}
+
+#define FUN_ADMIN_PORT_WRITE_REQ_INIT(_subop, _flags, _id)       \
+	(struct fun_admin_port_write_req) {                      \
+		.subop = (_subop), .flags = cpu_to_be16(_flags), \
+		.id = cpu_to_be32(_id),                          \
+	}
+
+#define FUN_ADMIN_PORT_READ_REQ_INIT(_subop, _flags, _id)        \
+	(struct fun_admin_port_read_req) {                       \
+		.subop = (_subop), .flags = cpu_to_be16(_flags), \
+		.id = cpu_to_be32(_id),                          \
+	}
+
+#define FUN_ADMIN_PORT_XCVR_READ_REQ_INIT(_flags, _id, _bank, _page,   \
+					  _offset, _length, _dev_addr) \
+	((struct fun_admin_port_xcvr_read_req) {                       \
+		.subop = FUN_ADMIN_PORT_SUBOP_XCVR_READ,               \
+		.flags = cpu_to_be16(_flags), .id = cpu_to_be32(_id),  \
+		.bank = (_bank), .page = (_page), .offset = (_offset), \
+		.length = (_length), .dev_addr = (_dev_addr),          \
+	})
+
+struct fun_admin_port_rsp {
+	struct fun_admin_rsp_common common;
+
+	union port_rsp_subop {
+		struct fun_admin_port_create_rsp {
+			__u8 subop;
+			__u8 rsvd0[3];
+			__be32 id;
+
+			__be16 lport;
+			__u8 rsvd1[6];
+		} create;
+		struct fun_admin_port_write_rsp {
+			__u8 subop;
+			__u8 rsvd0[3];
+			__be32 id; /* portid */
+
+			struct fun_admin_write48_rsp write48[];
+		} write;
+		struct fun_admin_port_read_rsp {
+			__u8 subop;
+			__u8 rsvd0[3];
+			__be32 id; /* portid */
+
+			struct fun_admin_read48_rsp read48[];
+		} read;
+		struct fun_admin_port_inetaddr_event_rsp {
+			__u8 subop;
+			__u8 rsvd0[3];
+			__be32 id; /* portid */
+		} inetaddr_event;
+	} u;
+};
+
+struct fun_admin_port_xcvr_read_rsp {
+	struct fun_admin_rsp_common common;
+
+	u8 subop;
+	u8 rsvd0[3];
+	__be32 id;
+
+	u8 bank;
+	u8 page;
+	u8 offset;
+	u8 length;
+	u8 dev_addr;
+	u8 rsvd1[3];
+
+	u8 data[128];
+};
+
+enum fun_xcvr_type {
+	FUN_XCVR_BASET = 0x0,
+	FUN_XCVR_CU = 0x1,
+	FUN_XCVR_SMF = 0x2,
+	FUN_XCVR_MMF = 0x3,
+	FUN_XCVR_AOC = 0x4,
+	FUN_XCVR_SFPP = 0x10, /* SFP+ or later */
+	FUN_XCVR_QSFPP = 0x11, /* QSFP+ or later */
+	FUN_XCVR_QSFPDD = 0x12, /* QSFP-DD */
+};
+
+struct fun_admin_port_notif {
+	struct fun_admin_rsp_common common;
+
+	__u8 subop;
+	__u8 rsvd0;
+	__be16 id;
+	__be32 speed; /* in 10 Mbps units */
+
+	__u8 link_state;
+	__u8 missed_events;
+	__u8 link_down_reason;
+	__u8 xcvr_type;
+	__u8 flow_ctrl;
+	__u8 fec;
+	__u8 active_lanes;
+	__u8 rsvd1;
+
+	__be64 advertising;
+
+	__be64 lp_advertising;
+};
+
+enum fun_eth_rss_const {
+	FUN_ETH_RSS_MAX_KEY_SIZE = 0x28,
+	FUN_ETH_RSS_MAX_INDIR_ENT = 0x40,
+};
+
+enum fun_eth_hash_alg {
+	FUN_ETH_RSS_ALG_INVALID = 0x0,
+	FUN_ETH_RSS_ALG_TOEPLITZ = 0x1,
+	FUN_ETH_RSS_ALG_CRC32 = 0x2,
+};
+
+struct fun_admin_rss_req {
+	struct fun_admin_req_common common;
+
+	union rss_req_subop {
+		struct fun_admin_rss_create_req {
+			__u8 subop;
+			__u8 rsvd0;
+			__be16 flags;
+			__be32 id;
+
+			__be32 rsvd1;
+			__be32 viid; /* VI flow id */
+
+			__be64 metadata[1];
+
+			__u8 alg;
+			__u8 keylen;
+			__u8 indir_nent;
+			__u8 rsvd2;
+			__be16 key_off;
+			__be16 indir_off;
+
+			struct fun_dataop_hdr dataop;
+		} create;
+	} u;
+};
+
+#define FUN_ADMIN_RSS_CREATE_REQ_INIT(_subop, _flags, _id, _viid, _alg,    \
+				      _keylen, _indir_nent, _key_off,      \
+				      _indir_off)                          \
+	(struct fun_admin_rss_create_req) {                                \
+		.subop = (_subop), .flags = cpu_to_be16(_flags),           \
+		.id = cpu_to_be32(_id), .viid = cpu_to_be32(_viid),        \
+		.alg = _alg, .keylen = _keylen, .indir_nent = _indir_nent, \
+		.key_off = cpu_to_be16(_key_off),                          \
+		.indir_off = cpu_to_be16(_indir_off),                      \
+	}
+
+struct fun_admin_vi_req {
+	struct fun_admin_req_common common;
+
+	union vi_req_subop {
+		struct fun_admin_vi_create_req {
+			__u8 subop;
+			__u8 rsvd0;
+			__be16 flags;
+			__be32 id;
+
+			__be32 rsvd1;
+			__be32 portid; /* port flow id */
+		} create;
+	} u;
+};
+
+#define FUN_ADMIN_VI_CREATE_REQ_INIT(_subop, _flags, _id, _portid)      \
+	(struct fun_admin_vi_create_req) {                              \
+		.subop = (_subop), .flags = cpu_to_be16(_flags),        \
+		.id = cpu_to_be32(_id), .portid = cpu_to_be32(_portid), \
+	}
+
+struct fun_admin_eth_req {
+	struct fun_admin_req_common common;
+
+	union eth_req_subop {
+		struct fun_admin_eth_create_req {
+			__u8 subop;
+			__u8 rsvd0;
+			__be16 flags;
+			__be32 id;
+
+			__be32 rsvd1;
+			__be32 portid; /* port flow id */
+		} create;
+	} u;
+};
+
+#define FUN_ADMIN_ETH_CREATE_REQ_INIT(_subop, _flags, _id, _portid)     \
+	(struct fun_admin_eth_create_req) {                             \
+		.subop = (_subop), .flags = cpu_to_be16(_flags),        \
+		.id = cpu_to_be32(_id), .portid = cpu_to_be32(_portid), \
+	}
+
+enum {
+	FUN_ADMIN_SWU_UPGRADE_FLAG_INIT = 0x10,
+	FUN_ADMIN_SWU_UPGRADE_FLAG_COMPLETE = 0x20,
+	FUN_ADMIN_SWU_UPGRADE_FLAG_DOWNGRADE = 0x40,
+	FUN_ADMIN_SWU_UPGRADE_FLAG_ACTIVE_IMAGE = 0x80,
+	FUN_ADMIN_SWU_UPGRADE_FLAG_ASYNC = 0x1,
+};
+
+enum fun_admin_swu_subop {
+	FUN_ADMIN_SWU_SUBOP_GET_VERSION = 0x20,
+	FUN_ADMIN_SWU_SUBOP_UPGRADE = 0x21,
+	FUN_ADMIN_SWU_SUBOP_UPGRADE_DATA = 0x22,
+	FUN_ADMIN_SWU_SUBOP_GET_ALL_VERSIONS = 0x23,
+};
+
+struct fun_admin_swu_req {
+	struct fun_admin_req_common common;
+
+	union swu_req_subop {
+		struct fun_admin_swu_create_req {
+			__u8 subop;
+			__u8 rsvd0;
+			__be16 flags;
+			__be32 id;
+		} create;
+		struct fun_admin_swu_upgrade_req {
+			__u8 subop;
+			__u8 rsvd0;
+			__be16 flags;
+			__be32 id;
+
+			__be32 fourcc;
+			__be32 rsvd1;
+
+			__be64 image_size; /* upgrade image length */
+		} upgrade;
+		struct fun_admin_swu_upgrade_data_req {
+			__u8 subop;
+			__u8 rsvd0;
+			__be16 flags;
+			__be32 id;
+
+			__be32 offset; /* offset of data in this command */
+			__be32 size; /* total size of data in this command */
+		} upgrade_data;
+	} u;
+
+	struct fun_subop_sgl sgl[]; /* in, out buffers through sgl */
+};
+
+#define FUN_ADMIN_SWU_CREATE_REQ_INIT(_subop, _flags, _id)       \
+	(struct fun_admin_swu_create_req) {                      \
+		.subop = (_subop), .flags = cpu_to_be16(_flags), \
+		.id = cpu_to_be32(_id),                          \
+	}
+
+#define FUN_ADMIN_SWU_UPGRADE_REQ_INIT(_subop, _flags, _id, _fourcc,    \
+				       _image_size)                     \
+	(struct fun_admin_swu_upgrade_req) {                            \
+		.subop = (_subop), .flags = cpu_to_be16(_flags),        \
+		.id = cpu_to_be32(_id), .fourcc = cpu_to_be32(_fourcc), \
+		.image_size = cpu_to_be64(_image_size),                 \
+	}
+
+#define FUN_ADMIN_SWU_UPGRADE_DATA_REQ_INIT(_subop, _flags, _id, _offset, \
+					    _size)                        \
+	(struct fun_admin_swu_upgrade_data_req) {                         \
+		.subop = (_subop), .flags = cpu_to_be16(_flags),          \
+		.id = cpu_to_be32(_id), .offset = cpu_to_be32(_offset),   \
+		.size = cpu_to_be32(_size),                               \
+	}
+
+struct fun_admin_swu_rsp {
+	struct fun_admin_rsp_common common;
+
+	union swu_rsp_subop {
+		struct fun_admin_swu_create_rsp {
+			__u8 subop;
+			__u8 rsvd0;
+			__be16 flags;
+			__be32 id;
+		} create;
+		struct fun_admin_swu_upgrade_rsp {
+			__u8 subop;
+			__u8 rsvd0[3];
+			__be32 id;
+
+			__be32 fourcc;
+			__be32 status;
+
+			__be32 progress;
+			__be32 unused;
+		} upgrade;
+		struct fun_admin_swu_upgrade_data_rsp {
+			__u8 subop;
+			__u8 rsvd0;
+			__be16 flags;
+			__be32 id;
+
+			__be32 offset;
+			__be32 size;
+		} upgrade_data;
+	} u;
+};
+
+enum fun_ktls_version {
+	FUN_KTLS_TLSV2 = 0x20,
+	FUN_KTLS_TLSV3 = 0x30,
+};
+
+enum fun_ktls_cipher {
+	FUN_KTLS_CIPHER_AES_GCM_128 = 0x33,
+	FUN_KTLS_CIPHER_AES_GCM_256 = 0x34,
+	FUN_KTLS_CIPHER_AES_CCM_128 = 0x35,
+	FUN_KTLS_CIPHER_CHACHA20_POLY1305 = 0x36,
+};
+
+enum fun_ktls_modify_flags {
+	FUN_KTLS_MODIFY_REMOVE = 0x1,
+};
+
+struct fun_admin_ktls_create_req {
+	struct fun_admin_req_common common;
+
+	__u8 subop;
+	__u8 rsvd0;
+	__be16 flags;
+	__be32 id;
+};
+
+#define FUN_ADMIN_KTLS_CREATE_REQ_INIT(_subop, _flags, _id)      \
+	(struct fun_admin_ktls_create_req) {                     \
+		.subop = (_subop), .flags = cpu_to_be16(_flags), \
+		.id = cpu_to_be32(_id),                          \
+	}
+
+struct fun_admin_ktls_create_rsp {
+	struct fun_admin_rsp_common common;
+
+	__u8 subop;
+	__u8 rsvd0[3];
+	__be32 id;
+};
+
+struct fun_admin_ktls_modify_req {
+	struct fun_admin_req_common common;
+
+	__u8 subop;
+	__u8 rsvd0;
+	__be16 flags;
+	__be32 id;
+
+	__be64 tlsid;
+
+	__be32 tcp_seq;
+	__u8 version;
+	__u8 cipher;
+	__u8 rsvd1[2];
+
+	__u8 record_seq[8];
+
+	__u8 key[32];
+
+	__u8 iv[16];
+
+	__u8 salt[8];
+};
+
+#define FUN_ADMIN_KTLS_MODIFY_REQ_INIT(_subop, _flags, _id, _tlsid, _tcp_seq, \
+				       _version, _cipher)                     \
+	(struct fun_admin_ktls_modify_req) {                                  \
+		.subop = (_subop), .flags = cpu_to_be16(_flags),              \
+		.id = cpu_to_be32(_id), .tlsid = cpu_to_be64(_tlsid),         \
+		.tcp_seq = cpu_to_be32(_tcp_seq), .version = _version,        \
+		.cipher = _cipher,                                            \
+	}
+
+struct fun_admin_ktls_modify_rsp {
+	struct fun_admin_rsp_common common;
+
+	__u8 subop;
+	__u8 rsvd0[3];
+	__be32 id;
+
+	__be64 tlsid;
+};
+
+struct fun_req_common {
+	__u8 op;
+	__u8 len8;
+	__be16 flags;
+	__u8 suboff8;
+	__u8 rsvd0;
+	__be16 cid;
+};
+
+struct fun_rsp_common {
+	__u8 op;
+	__u8 len8;
+	__be16 flags;
+	__u8 suboff8;
+	__u8 ret;
+	__be16 cid;
+};
+
+struct fun_cqe_info {
+	__be16 sqhd;
+	__be16 sqid;
+	__be16 cid;
+	__be16 sf_p;
+};
+
+enum fun_eprq_def {
+	FUN_EPRQ_PKT_ALIGN = 0x80,
+};
+
+struct fun_eprq_rqbuf {
+	__be64 bufaddr;
+};
+
+#define FUN_EPRQ_RQBUF_INIT(_bufaddr)             \
+	(struct fun_eprq_rqbuf) {                 \
+		.bufaddr = cpu_to_be64(_bufaddr), \
+	}
+
+enum fun_eth_op {
+	FUN_ETH_OP_TX = 0x1,
+	FUN_ETH_OP_RX = 0x2,
+};
+
+enum {
+	FUN_ETH_OFFLOAD_EN = 0x8000,
+	FUN_ETH_OUTER_EN = 0x4000,
+	FUN_ETH_INNER_LSO = 0x2000,
+	FUN_ETH_INNER_TSO = 0x1000,
+	FUN_ETH_OUTER_IPV6 = 0x800,
+	FUN_ETH_OUTER_UDP = 0x400,
+	FUN_ETH_INNER_IPV6 = 0x200,
+	FUN_ETH_INNER_UDP = 0x100,
+	FUN_ETH_UPDATE_OUTER_L3_LEN = 0x80,
+	FUN_ETH_UPDATE_OUTER_L3_CKSUM = 0x40,
+	FUN_ETH_UPDATE_OUTER_L4_LEN = 0x20,
+	FUN_ETH_UPDATE_OUTER_L4_CKSUM = 0x10,
+	FUN_ETH_UPDATE_INNER_L3_LEN = 0x8,
+	FUN_ETH_UPDATE_INNER_L3_CKSUM = 0x4,
+	FUN_ETH_UPDATE_INNER_L4_LEN = 0x2,
+	FUN_ETH_UPDATE_INNER_L4_CKSUM = 0x1,
+};
+
+struct fun_eth_offload {
+	__be16 flags; /* combination of above flags */
+	__be16 mss; /* TSO max seg size */
+	__be16 tcp_doff_flags; /* TCP data offset + flags 16b word */
+	__be16 vlan;
+
+	__be16 inner_l3_off; /* Inner L3 header offset */
+	__be16 inner_l4_off; /* Inner L4 header offset */
+	__be16 outer_l3_off; /* Outer L3 header offset */
+	__be16 outer_l4_off; /* Outer L4 header offset */
+};
+
+static inline void fun_eth_offload_init(struct fun_eth_offload *s, u16 flags,
+					u16 mss, __be16 tcp_doff_flags,
+					__be16 vlan, u16 inner_l3_off,
+					u16 inner_l4_off, u16 outer_l3_off,
+					u16 outer_l4_off)
+{
+	s->flags = cpu_to_be16(flags);
+	s->mss = cpu_to_be16(mss);
+	s->tcp_doff_flags = tcp_doff_flags;
+	s->vlan = vlan;
+	s->inner_l3_off = cpu_to_be16(inner_l3_off);
+	s->inner_l4_off = cpu_to_be16(inner_l4_off);
+	s->outer_l3_off = cpu_to_be16(outer_l3_off);
+	s->outer_l4_off = cpu_to_be16(outer_l4_off);
+}
+
+struct fun_eth_tls {
+	__be64 tlsid;
+};
+
+enum {
+	FUN_ETH_TX_TLS = 0x8000,
+};
+
+struct fun_eth_tx_req {
+	__u8 op;
+	__u8 len8;
+	__be16 flags;
+	__u8 suboff8;
+	__u8 repr_idn;
+	__be16 encap_proto;
+
+	struct fun_eth_offload offload;
+
+	struct fun_dataop_hdr dataop;
+};
+
+struct fun_eth_rx_cv {
+	__be16 il4_prot_to_l2_type;
+};
+
+#define FUN_ETH_RX_CV_IL4_PROT_S 13U
+#define FUN_ETH_RX_CV_IL4_PROT_M 0x3
+
+#define FUN_ETH_RX_CV_IL3_PROT_S 11U
+#define FUN_ETH_RX_CV_IL3_PROT_M 0x3
+
+#define FUN_ETH_RX_CV_OL4_PROT_S 8U
+#define FUN_ETH_RX_CV_OL4_PROT_M 0x7
+
+#define FUN_ETH_RX_CV_ENCAP_TYPE_S 6U
+#define FUN_ETH_RX_CV_ENCAP_TYPE_M 0x3
+
+#define FUN_ETH_RX_CV_OL3_PROT_S 4U
+#define FUN_ETH_RX_CV_OL3_PROT_M 0x3
+
+#define FUN_ETH_RX_CV_VLAN_TYPE_S 3U
+#define FUN_ETH_RX_CV_VLAN_TYPE_M 0x1
+
+#define FUN_ETH_RX_CV_L2_TYPE_S 2U
+#define FUN_ETH_RX_CV_L2_TYPE_M 0x1
+
+enum fun_rx_cv {
+	FUN_RX_CV_NONE = 0x0,
+	FUN_RX_CV_IP = 0x2,
+	FUN_RX_CV_IP6 = 0x3,
+	FUN_RX_CV_TCP = 0x2,
+	FUN_RX_CV_UDP = 0x3,
+	FUN_RX_CV_VXLAN = 0x2,
+	FUN_RX_CV_MPLS = 0x3,
+};
+
+struct fun_eth_cqe {
+	__u8 op;
+	__u8 len8;
+	__u8 nsgl;
+	__u8 repr_idn;
+	__be32 pkt_len;
+
+	__be64 timestamp;
+
+	__be16 pkt_cv;
+	__be16 rsvd0;
+	__be32 hash;
+
+	__be16 encap_proto;
+	__be16 vlan;
+	__be32 rsvd1;
+
+	__be32 buf_offset;
+	__be16 headroom;
+	__be16 csum;
+};
+
+enum fun_admin_adi_attr {
+	FUN_ADMIN_ADI_ATTR_MACADDR = 0x1,
+	FUN_ADMIN_ADI_ATTR_VLAN = 0x2,
+	FUN_ADMIN_ADI_ATTR_RATE = 0x3,
+};
+
+struct fun_adi_param {
+	union adi_param {
+		struct fun_adi_mac {
+			__be64 addr;
+		} mac;
+		struct fun_adi_vlan {
+			__be32 rsvd;
+			__be16 eth_type;
+			__be16 tci;
+		} vlan;
+		struct fun_adi_rate {
+			__be32 rsvd;
+			__be32 tx_mbps;
+		} rate;
+	} u;
+};
+
+#define FUN_ADI_MAC_INIT(_addr)             \
+	(struct fun_adi_mac) {              \
+		.addr = cpu_to_be64(_addr), \
+	}
+
+#define FUN_ADI_VLAN_INIT(_eth_type, _tci)                                    \
+	(struct fun_adi_vlan) {                                               \
+		.eth_type = cpu_to_be16(_eth_type), .tci = cpu_to_be16(_tci), \
+	}
+
+#define FUN_ADI_RATE_INIT(_tx_mbps)               \
+	(struct fun_adi_rate) {                   \
+		.tx_mbps = cpu_to_be32(_tx_mbps), \
+	}
+
+struct fun_admin_adi_req {
+	struct fun_admin_req_common common;
+
+	union adi_req_subop {
+		struct fun_admin_adi_write_req {
+			__u8 subop;
+			__u8 attribute;
+			__be16 rsvd;
+			__be32 id;
+
+			struct fun_adi_param param;
+		} write;
+	} u;
+};
+
+#define FUN_ADMIN_ADI_WRITE_REQ_INIT(_subop, _attribute, _id) \
+	(struct fun_admin_adi_write_req) {                    \
+		.subop = (_subop), .attribute = (_attribute), \
+		.id = cpu_to_be32(_id),                       \
+	}
+
+#endif /* __FUN_HCI_H */
diff --git a/drivers/net/ethernet/fungible/funcore/fun_queue.c b/drivers/net/ethernet/fungible/funcore/fun_queue.c
new file mode 100644
index 000000000000..d07ee3e4f52a
--- /dev/null
+++ b/drivers/net/ethernet/fungible/funcore/fun_queue.c
@@ -0,0 +1,536 @@
+// SPDX-License-Identifier: (GPL-2.0-only OR BSD-3-Clause)
+
+#include <linux/dma-mapping.h>
+#include <linux/interrupt.h>
+#include <linux/log2.h>
+#include <linux/mm.h>
+#include <linux/netdevice.h>
+#include <linux/pci.h>
+#include <linux/slab.h>
+
+#include "fun_dev.h"
+#include "fun_queue.h"
+
+/* Allocate memory for a queue. This includes the memory for the HW descriptor
+ * ring, an optional 64b HW write-back area, and an optional SW state ring.
+ * Returns the virtual and DMA addresses of the HW ring, the VA of the SW ring,
+ * and the VA of the write-back area.
+ */
+void *fun_alloc_ring_mem(struct device *dma_dev, size_t depth,
+			 size_t hw_desc_sz, size_t sw_desc_sz, bool wb,
+			 int numa_node, dma_addr_t *dma_addr, void **sw_va,
+			 volatile __be64 **wb_va)
+{
+	int dev_node = dev_to_node(dma_dev);
+	size_t dma_sz;
+	void *va;
+
+	if (numa_node == NUMA_NO_NODE)
+		numa_node = dev_node;
+
+	/* Place optional write-back area at end of descriptor ring. */
+	dma_sz = hw_desc_sz * depth;
+	if (wb)
+		dma_sz += sizeof(u64);
+
+	set_dev_node(dma_dev, numa_node);
+	va = dma_alloc_coherent(dma_dev, dma_sz, dma_addr, GFP_KERNEL);
+	set_dev_node(dma_dev, dev_node);
+	if (!va)
+		return NULL;
+
+	if (sw_desc_sz) {
+		*sw_va = kvzalloc_node(sw_desc_sz * depth, GFP_KERNEL,
+				       numa_node);
+		if (!*sw_va) {
+			dma_free_coherent(dma_dev, dma_sz, va, *dma_addr);
+			return NULL;
+		}
+	}
+
+	if (wb)
+		*wb_va = va + dma_sz - sizeof(u64);
+	return va;
+}
+EXPORT_SYMBOL_GPL(fun_alloc_ring_mem);
+
+void fun_free_ring_mem(struct device *dma_dev, size_t depth, size_t hw_desc_sz,
+		       bool wb, void *hw_va, dma_addr_t dma_addr, void *sw_va)
+{
+	if (hw_va) {
+		size_t sz = depth * hw_desc_sz;
+
+		if (wb)
+			sz += sizeof(u64);
+		dma_free_coherent(dma_dev, sz, hw_va, dma_addr);
+	}
+	kvfree(sw_va);
+}
+EXPORT_SYMBOL_GPL(fun_free_ring_mem);
+
+/* Prepare and issue an admin command to create an SQ on the device with the
+ * provided parameters. If the queue ID is auto-allocated by the device it is
+ * returned in *sqidp.
+ */
+int fun_sq_create(struct fun_dev *fdev, u16 flags, u32 sqid, u32 cqid,
+		  u8 sqe_size_log2, u32 sq_depth, dma_addr_t dma_addr,
+		  u8 coal_nentries, u8 coal_usec, u32 irq_num,
+		  u32 scan_start_id, u32 scan_end_id,
+		  u32 rq_buf_size_log2, u32 *sqidp, u32 __iomem **dbp)
+{
+	union {
+		struct fun_admin_epsq_req req;
+		struct fun_admin_generic_create_rsp rsp;
+	} cmd;
+	dma_addr_t wb_addr;
+	u32 hw_qid;
+	int rc;
+
+	if (sq_depth > fdev->q_depth)
+		return -EINVAL;
+	if (flags & FUN_ADMIN_EPSQ_CREATE_FLAG_RQ)
+		sqe_size_log2 = ilog2(sizeof(struct fun_eprq_rqbuf));
+
+	wb_addr = dma_addr + (sq_depth << sqe_size_log2);
+
+	cmd.req.common = FUN_ADMIN_REQ_COMMON_INIT2(FUN_ADMIN_OP_EPSQ,
+						    sizeof(cmd.req));
+	cmd.req.u.create =
+		FUN_ADMIN_EPSQ_CREATE_REQ_INIT(FUN_ADMIN_SUBOP_CREATE, flags,
+					       sqid, cqid, sqe_size_log2,
+					       sq_depth - 1, dma_addr, 0,
+					       coal_nentries, coal_usec,
+					       irq_num, scan_start_id,
+					       scan_end_id, 0,
+					       rq_buf_size_log2,
+					       ilog2(sizeof(u64)), wb_addr);
+
+	rc = fun_submit_admin_sync_cmd(fdev, &cmd.req.common,
+				       &cmd.rsp, sizeof(cmd.rsp), 0);
+	if (rc)
+		return rc;
+
+	hw_qid = be32_to_cpu(cmd.rsp.id);
+	*dbp = fun_sq_db_addr(fdev, hw_qid);
+	if (flags & FUN_ADMIN_RES_CREATE_FLAG_ALLOCATOR)
+		*sqidp = hw_qid;
+	return rc;
+}
+EXPORT_SYMBOL_GPL(fun_sq_create);
+
+/* Prepare and issue an admin command to create a CQ on the device with the
+ * provided parameters. If the queue ID is auto-allocated by the device it is
+ * returned in *cqidp.
+ */
+int fun_cq_create(struct fun_dev *fdev, u16 flags, u32 cqid, u32 rqid,
+		  u8 cqe_size_log2, u32 cq_depth, dma_addr_t dma_addr,
+		  u16 headroom, u16 tailroom, u8 coal_nentries, u8 coal_usec,
+		  u32 irq_num, u32 scan_start_id, u32 scan_end_id, u32 *cqidp,
+		  u32 __iomem **dbp)
+{
+	union {
+		struct fun_admin_epcq_req req;
+		struct fun_admin_generic_create_rsp rsp;
+	} cmd;
+	u32 hw_qid;
+	int rc;
+
+	if (cq_depth > fdev->q_depth)
+		return -EINVAL;
+
+	cmd.req.common = FUN_ADMIN_REQ_COMMON_INIT2(FUN_ADMIN_OP_EPCQ,
+						    sizeof(cmd.req));
+	cmd.req.u.create =
+		FUN_ADMIN_EPCQ_CREATE_REQ_INIT(FUN_ADMIN_SUBOP_CREATE, flags,
+					       cqid, rqid, cqe_size_log2,
+					       cq_depth - 1, dma_addr, tailroom,
+					       headroom / 2, 0, coal_nentries,
+					       coal_usec, irq_num,
+					       scan_start_id, scan_end_id, 0);
+
+	rc = fun_submit_admin_sync_cmd(fdev, &cmd.req.common,
+				       &cmd.rsp, sizeof(cmd.rsp), 0);
+	if (rc)
+		return rc;
+
+	hw_qid = be32_to_cpu(cmd.rsp.id);
+	*dbp = fun_cq_db_addr(fdev, hw_qid);
+	if (flags & FUN_ADMIN_RES_CREATE_FLAG_ALLOCATOR)
+		*cqidp = hw_qid;
+	return rc;
+}
+EXPORT_SYMBOL_GPL(fun_cq_create);
+
+static bool fun_sq_is_head_wb(const struct fun_queue *funq)
+{
+	return funq->sq_flags & FUN_ADMIN_EPSQ_CREATE_FLAG_HEAD_WB_ADDRESS;
+}
+
+static void fun_clean_rq(struct fun_queue *funq)
+{
+	struct fun_dev *fdev = funq->fdev;
+	struct fun_rq_info *rqinfo;
+	unsigned int i;
+
+	for (i = 0; i < funq->rq_depth; i++) {
+		rqinfo = &funq->rq_info[i];
+		if (rqinfo->page) {
+			dma_unmap_page(fdev->dev, rqinfo->dma, PAGE_SIZE,
+				       DMA_FROM_DEVICE);
+			put_page(rqinfo->page);
+			rqinfo->page = NULL;
+		}
+	}
+}
+
+static int fun_fill_rq(struct fun_queue *funq)
+{
+	struct device *dev = funq->fdev->dev;
+	int i, node = dev_to_node(dev);
+	struct fun_rq_info *rqinfo;
+
+	for (i = 0; i < funq->rq_depth; i++) {
+		rqinfo = &funq->rq_info[i];
+		rqinfo->page = alloc_pages_node(node, GFP_KERNEL, 0);
+		if (unlikely(!rqinfo->page))
+			return -ENOMEM;
+
+		rqinfo->dma = dma_map_page(dev, rqinfo->page, 0,
+					   PAGE_SIZE, DMA_FROM_DEVICE);
+		if (unlikely(dma_mapping_error(dev, rqinfo->dma))) {
+			put_page(rqinfo->page);
+			rqinfo->page = NULL;
+			return -ENOMEM;
+		}
+
+		funq->rqes[i] = FUN_EPRQ_RQBUF_INIT(rqinfo->dma);
+	}
+
+	funq->rq_tail = funq->rq_depth - 1;
+	return 0;
+}
+
+static void fun_rq_update_pos(struct fun_queue *funq, int buf_offset)
+{
+	if (buf_offset <= funq->rq_buf_offset) {
+		struct fun_rq_info *rqinfo = &funq->rq_info[funq->rq_buf_idx];
+		struct device *dev = funq->fdev->dev;
+
+		dma_sync_single_for_device(dev, rqinfo->dma, PAGE_SIZE,
+					   DMA_FROM_DEVICE);
+		funq->num_rqe_to_fill++;
+		if (++funq->rq_buf_idx == funq->rq_depth)
+			funq->rq_buf_idx = 0;
+	}
+	funq->rq_buf_offset = buf_offset;
+}
+
+/* Given a command response with data scattered across >= 1 RQ buffers return
+ * a pointer to a contiguous buffer containing all the data. If the data is in
+ * one RQ buffer the start address within that buffer is returned, otherwise a
+ * new buffer is allocated and the data is gathered into it.
+ */
+static void *fun_data_from_rq(struct fun_queue *funq,
+			      const struct fun_rsp_common *rsp, bool *need_free)
+{
+	u32 bufoff, total_len, remaining, fragsize, dataoff;
+	struct device *dma_dev = funq->fdev->dev;
+	const struct fun_dataop_rqbuf *databuf;
+	const struct fun_dataop_hdr *dataop;
+	const struct fun_rq_info *rqinfo;
+	void *data;
+
+	dataop = (void *)rsp + rsp->suboff8 * 8;
+	total_len = be32_to_cpu(dataop->total_len);
+
+	if (likely(dataop->nsgl == 1)) {
+		databuf = (struct fun_dataop_rqbuf *)dataop->imm;
+		bufoff = be32_to_cpu(databuf->bufoff);
+		fun_rq_update_pos(funq, bufoff);
+		rqinfo = &funq->rq_info[funq->rq_buf_idx];
+		dma_sync_single_for_cpu(dma_dev, rqinfo->dma + bufoff,
+					total_len, DMA_FROM_DEVICE);
+		*need_free = false;
+		return page_address(rqinfo->page) + bufoff;
+	}
+
+	/* For scattered completions gather the fragments into one buffer. */
+
+	data = kmalloc(total_len, GFP_ATOMIC);
+	/* NULL is OK here. In case of failure we still need to consume the data
+	 * for proper buffer accounting but indicate an error in the response.
+	 */
+	if (likely(data))
+		*need_free = true;
+
+	dataoff = 0;
+	for (remaining = total_len; remaining; remaining -= fragsize) {
+		fun_rq_update_pos(funq, 0);
+		fragsize = min_t(unsigned int, PAGE_SIZE, remaining);
+		if (data) {
+			rqinfo = &funq->rq_info[funq->rq_buf_idx];
+			dma_sync_single_for_cpu(dma_dev, rqinfo->dma, fragsize,
+						DMA_FROM_DEVICE);
+			memcpy(data + dataoff, page_address(rqinfo->page),
+			       fragsize);
+			dataoff += fragsize;
+		}
+	}
+	return data;
+}
+
+unsigned int __fun_process_cq(struct fun_queue *funq, unsigned int max)
+{
+	const struct fun_cqe_info *info;
+	struct fun_rsp_common *rsp;
+	unsigned int new_cqes;
+	u16 sf_p, flags;
+	bool need_free;
+	void *cqe;
+
+	if (!max)
+		max = funq->cq_depth - 1;
+
+	for (new_cqes = 0; new_cqes < max; new_cqes++) {
+		cqe = funq->cqes + (funq->cq_head << funq->cqe_size_log2);
+		info = funq_cqe_info(funq, cqe);
+		sf_p = be16_to_cpu(info->sf_p);
+
+		if ((sf_p & 1) != funq->cq_phase)
+			break;
+
+		/* ensure the phase tag is read before other CQE fields */
+		dma_rmb();
+
+		if (++funq->cq_head == funq->cq_depth) {
+			funq->cq_head = 0;
+			funq->cq_phase = !funq->cq_phase;
+		}
+
+		rsp = cqe;
+		flags = be16_to_cpu(rsp->flags);
+
+		need_free = false;
+		if (unlikely(flags & FUN_REQ_COMMON_FLAG_CQE_IN_RQBUF)) {
+			rsp = fun_data_from_rq(funq, rsp, &need_free);
+			if (!rsp) {
+				rsp = cqe;
+				rsp->len8 = 1;
+				if (rsp->ret == 0)
+					rsp->ret = ENOMEM;
+			}
+		}
+
+		if (funq->cq_cb)
+			funq->cq_cb(funq, funq->cb_data, rsp, info);
+		if (need_free)
+			kfree(rsp);
+	}
+
+	dev_dbg(funq->fdev->dev, "CQ %u, new CQEs %u/%u, head %u, phase %u\n",
+		funq->cqid, new_cqes, max, funq->cq_head, funq->cq_phase);
+	return new_cqes;
+}
+
+unsigned int fun_process_cq(struct fun_queue *funq, unsigned int max)
+{
+	unsigned int processed;
+	u32 db;
+
+	processed = __fun_process_cq(funq, max);
+
+	if (funq->num_rqe_to_fill) {
+		funq->rq_tail = (funq->rq_tail + funq->num_rqe_to_fill) %
+				funq->rq_depth;
+		funq->num_rqe_to_fill = 0;
+		writel(funq->rq_tail, funq->rq_db);
+	}
+
+	db = funq->cq_head | FUN_DB_IRQ_ARM_F;
+	writel(db, funq->cq_db);
+	return processed;
+}
+
+static int fun_alloc_sqes(struct fun_queue *funq)
+{
+	funq->sq_cmds = fun_alloc_ring_mem(funq->fdev->dev, funq->sq_depth,
+					   1 << funq->sqe_size_log2, 0,
+					   fun_sq_is_head_wb(funq),
+					   NUMA_NO_NODE, &funq->sq_dma_addr,
+					   NULL, &funq->sq_head);
+	return funq->sq_cmds ? 0 : -ENOMEM;
+}
+
+static int fun_alloc_cqes(struct fun_queue *funq)
+{
+	funq->cqes = fun_alloc_ring_mem(funq->fdev->dev, funq->cq_depth,
+					1 << funq->cqe_size_log2, 0, false,
+					NUMA_NO_NODE, &funq->cq_dma_addr, NULL,
+					NULL);
+	return funq->cqes ? 0 : -ENOMEM;
+}
+
+static int fun_alloc_rqes(struct fun_queue *funq)
+{
+	funq->rqes = fun_alloc_ring_mem(funq->fdev->dev, funq->rq_depth,
+					sizeof(*funq->rqes),
+					sizeof(*funq->rq_info), false,
+					NUMA_NO_NODE, &funq->rq_dma_addr,
+					(void **)&funq->rq_info, NULL);
+	return funq->rqes ? 0 : -ENOMEM;
+}
+
+/* Free a queue's structures. */
+void fun_free_queue(struct fun_queue *funq)
+{
+	struct device *dev = funq->fdev->dev;
+
+	fun_free_ring_mem(dev, funq->cq_depth, 1 << funq->cqe_size_log2, false,
+			  funq->cqes, funq->cq_dma_addr, NULL);
+	fun_free_ring_mem(dev, funq->sq_depth, 1 << funq->sqe_size_log2,
+			  fun_sq_is_head_wb(funq), funq->sq_cmds,
+			  funq->sq_dma_addr, NULL);
+
+	if (funq->rqes) {
+		fun_clean_rq(funq);
+		fun_free_ring_mem(dev, funq->rq_depth, sizeof(*funq->rqes),
+				  false, funq->rqes, funq->rq_dma_addr,
+				  funq->rq_info);
+	}
+
+	kfree(funq);
+}
+
+/* Allocate and initialize a funq's structures. */
+struct fun_queue *fun_alloc_queue(struct fun_dev *fdev, int qid,
+				  const struct fun_queue_alloc_req *req)
+{
+	struct fun_queue *funq = kzalloc(sizeof(*funq), GFP_KERNEL);
+
+	if (!funq)
+		return NULL;
+
+	funq->fdev = fdev;
+	spin_lock_init(&funq->sq_lock);
+
+	funq->qid = qid;
+
+	/* Initial CQ/SQ/RQ ids */
+	if (req->rq_depth) {
+		funq->cqid = 2 * qid;
+		if (funq->qid) {
+			/* I/O Q: use rqid = cqid, sqid = +1 */
+			funq->rqid = funq->cqid;
+			funq->sqid = funq->rqid + 1;
+		} else {
+			/* Admin Q: sqid is always 0, use ID 1 for RQ */
+			funq->sqid = 0;
+			funq->rqid = 1;
+		}
+	} else {
+		funq->cqid = qid;
+		funq->sqid = qid;
+	}
+
+	funq->cq_flags = req->cq_flags;
+	funq->sq_flags = req->sq_flags;
+
+	funq->cqe_size_log2 = req->cqe_size_log2;
+	funq->sqe_size_log2 = req->sqe_size_log2;
+
+	funq->cq_depth = req->cq_depth;
+	funq->sq_depth = req->sq_depth;
+
+	funq->cq_intcoal_nentries = req->cq_intcoal_nentries;
+	funq->cq_intcoal_usec = req->cq_intcoal_usec;
+
+	funq->sq_intcoal_nentries = req->sq_intcoal_nentries;
+	funq->sq_intcoal_usec = req->sq_intcoal_usec;
+
+	if (fun_alloc_cqes(funq))
+		goto free_funq;
+
+	funq->cq_phase = 1;
+
+	if (fun_alloc_sqes(funq))
+		goto free_funq;
+
+	if (req->rq_depth) {
+		funq->rq_flags = req->rq_flags | FUN_ADMIN_EPSQ_CREATE_FLAG_RQ;
+		funq->rq_depth = req->rq_depth;
+		funq->rq_buf_offset = -1;
+
+		if (fun_alloc_rqes(funq) || fun_fill_rq(funq))
+			goto free_funq;
+	}
+
+	funq->cq_vector = -1;
+	funq->cqe_info_offset = (1 << funq->cqe_size_log2) - sizeof(struct fun_cqe_info);
+
+	/* SQ/CQ 0 are implicitly created, assign their doorbells now.
+	 * Other queues are assigned doorbells at their explicit creation.
+	 */
+	if (funq->sqid == 0)
+		funq->sq_db = fun_sq_db_addr(fdev, 0);
+	if (funq->cqid == 0)
+		funq->cq_db = fun_cq_db_addr(fdev, 0);
+
+	return funq;
+
+free_funq:
+	fun_free_queue(funq);
+	return NULL;
+}
+
+/* Create a funq's RQ on the device. */
+int fun_create_rq(struct fun_queue *funq)
+{
+	struct fun_dev *fdev = funq->fdev;
+	int rc;
+
+	rc = fun_sq_create(fdev, funq->rq_flags, funq->rqid, funq->cqid, 0,
+			   funq->rq_depth, funq->rq_dma_addr, 0, 0,
+			   funq->cq_vector, 0, 0, PAGE_SHIFT, &funq->rqid,
+			   &funq->rq_db);
+	if (!rc)
+		dev_dbg(fdev->dev, "created RQ %u\n", funq->rqid);
+
+	return rc;
+}
+
+static unsigned int funq_irq(struct fun_queue *funq)
+{
+	return pci_irq_vector(to_pci_dev(funq->fdev->dev), funq->cq_vector);
+}
+
+int fun_request_irq(struct fun_queue *funq, const char *devname,
+		    irq_handler_t handler, void *data)
+{
+	int rc;
+
+	if (funq->cq_vector < 0)
+		return -EINVAL;
+
+	funq->irq_handler = handler;
+	funq->irq_data = data;
+
+	snprintf(funq->irqname, sizeof(funq->irqname),
+		 funq->qid ? "%s-q[%d]" : "%s-adminq", devname, funq->qid);
+
+	rc = request_irq(funq_irq(funq), handler, 0, funq->irqname, data);
+	if (rc)
+		funq->irq_handler = NULL;
+
+	return rc;
+}
+
+void fun_free_irq(struct fun_queue *funq)
+{
+	if (funq->irq_handler) {
+		unsigned int vector = funq_irq(funq);
+
+		free_irq(vector, funq->irq_data);
+		funq->irq_handler = NULL;
+		funq->irq_data = NULL;
+	}
+}
diff --git a/drivers/net/ethernet/fungible/funcore/fun_queue.h b/drivers/net/ethernet/fungible/funcore/fun_queue.h
new file mode 100644
index 000000000000..2d966afb187a
--- /dev/null
+++ b/drivers/net/ethernet/fungible/funcore/fun_queue.h
@@ -0,0 +1,174 @@
+/* SPDX-License-Identifier: (GPL-2.0-only OR BSD-3-Clause) */
+
+#ifndef _FUN_QEUEUE_H
+#define _FUN_QEUEUE_H
+
+#include <linux/interrupt.h>
+#include <linux/io.h>
+
+struct device;
+struct fun_dev;
+struct fun_queue;
+struct fun_cqe_info;
+struct fun_rsp_common;
+
+typedef void (*cq_callback_t)(struct fun_queue *funq, void *data, void *msg,
+			      const struct fun_cqe_info *info);
+
+struct fun_rq_info {
+	dma_addr_t dma;
+	struct page *page;
+};
+
+/* A queue group consisting of an SQ, a CQ, and an optional RQ. */
+struct fun_queue {
+	struct fun_dev *fdev;
+	spinlock_t sq_lock;
+
+	dma_addr_t cq_dma_addr;
+	dma_addr_t sq_dma_addr;
+	dma_addr_t rq_dma_addr;
+
+	u32 __iomem *cq_db;
+	u32 __iomem *sq_db;
+	u32 __iomem *rq_db;
+
+	void *cqes;
+	void *sq_cmds;
+	struct fun_eprq_rqbuf *rqes;
+	struct fun_rq_info *rq_info;
+
+	u32 cqid;
+	u32 sqid;
+	u32 rqid;
+
+	u32 cq_depth;
+	u32 sq_depth;
+	u32 rq_depth;
+
+	u16 cq_head;
+	u16 sq_tail;
+	u16 rq_tail;
+
+	u8 cqe_size_log2;
+	u8 sqe_size_log2;
+
+	u16 cqe_info_offset;
+
+	u16 rq_buf_idx;
+	int rq_buf_offset;
+	u16 num_rqe_to_fill;
+
+	u8 cq_intcoal_usec;
+	u8 cq_intcoal_nentries;
+	u8 sq_intcoal_usec;
+	u8 sq_intcoal_nentries;
+
+	u16 cq_flags;
+	u16 sq_flags;
+	u16 rq_flags;
+
+	/* SQ head writeback */
+	u16 sq_comp;
+
+	volatile __be64 *sq_head;
+
+	cq_callback_t cq_cb;
+	void *cb_data;
+
+	irq_handler_t irq_handler;
+	void *irq_data;
+	s16 cq_vector;
+	u8 cq_phase;
+
+	/* I/O q index */
+	u16 qid;
+
+	char irqname[24];
+};
+
+static inline void *fun_sqe_at(const struct fun_queue *funq, unsigned int pos)
+{
+	return funq->sq_cmds + (pos << funq->sqe_size_log2);
+}
+
+static inline void funq_sq_post_tail(struct fun_queue *funq, u16 tail)
+{
+	if (++tail == funq->sq_depth)
+		tail = 0;
+	funq->sq_tail = tail;
+	writel(tail, funq->sq_db);
+}
+
+static inline struct fun_cqe_info *funq_cqe_info(const struct fun_queue *funq,
+						 void *cqe)
+{
+	return cqe + funq->cqe_info_offset;
+}
+
+static inline void funq_rq_post(struct fun_queue *funq)
+{
+	writel(funq->rq_tail, funq->rq_db);
+}
+
+struct fun_queue_alloc_req {
+	u8  cqe_size_log2;
+	u8  sqe_size_log2;
+
+	u16 cq_flags;
+	u16 sq_flags;
+	u16 rq_flags;
+
+	u32 cq_depth;
+	u32 sq_depth;
+	u32 rq_depth;
+
+	u8 cq_intcoal_usec;
+	u8 cq_intcoal_nentries;
+	u8 sq_intcoal_usec;
+	u8 sq_intcoal_nentries;
+};
+
+int fun_sq_create(struct fun_dev *fdev, u16 flags, u32 sqid, u32 cqid,
+		  u8 sqe_size_log2, u32 sq_depth, dma_addr_t dma_addr,
+		  u8 coal_nentries, u8 coal_usec, u32 irq_num,
+		  u32 scan_start_id, u32 scan_end_id,
+		  u32 rq_buf_size_log2, u32 *sqidp, u32 __iomem **dbp);
+int fun_cq_create(struct fun_dev *fdev, u16 flags, u32 cqid, u32 rqid,
+		  u8 cqe_size_log2, u32 cq_depth, dma_addr_t dma_addr,
+		  u16 headroom, u16 tailroom, u8 coal_nentries, u8 coal_usec,
+		  u32 irq_num, u32 scan_start_id, u32 scan_end_id,
+		  u32 *cqidp, u32 __iomem **dbp);
+void *fun_alloc_ring_mem(struct device *dma_dev, size_t depth,
+			 size_t hw_desc_sz, size_t sw_desc_size, bool wb,
+			 int numa_node, dma_addr_t *dma_addr, void **sw_va,
+			 volatile __be64 **wb_va);
+void fun_free_ring_mem(struct device *dma_dev, size_t depth, size_t hw_desc_sz,
+		       bool wb, void *hw_va, dma_addr_t dma_addr, void *sw_va);
+
+#define fun_destroy_sq(fdev, sqid) \
+	fun_res_destroy((fdev), FUN_ADMIN_OP_EPSQ, 0, (sqid))
+#define fun_destroy_cq(fdev, cqid) \
+	fun_res_destroy((fdev), FUN_ADMIN_OP_EPCQ, 0, (cqid))
+
+struct fun_queue *fun_alloc_queue(struct fun_dev *fdev, int qid,
+				  const struct fun_queue_alloc_req *req);
+void fun_free_queue(struct fun_queue *funq);
+
+static inline void fun_set_cq_callback(struct fun_queue *funq, cq_callback_t cb,
+				       void *cb_data)
+{
+	funq->cq_cb = cb;
+	funq->cb_data = cb_data;
+}
+
+int fun_create_rq(struct fun_queue *funq);
+
+void fun_free_irq(struct fun_queue *funq);
+int fun_request_irq(struct fun_queue *funq, const char *devname,
+		    irq_handler_t handler, void *data);
+
+unsigned int __fun_process_cq(struct fun_queue *funq, unsigned int max);
+unsigned int fun_process_cq(struct fun_queue *funq, unsigned int max);
+
+#endif /* _FUN_QEUEUE_H */
diff --git a/drivers/net/ethernet/fungible/funeth/Kconfig b/drivers/net/ethernet/fungible/funeth/Kconfig
new file mode 100644
index 000000000000..e742e7663449
--- /dev/null
+++ b/drivers/net/ethernet/fungible/funeth/Kconfig
@@ -0,0 +1,17 @@
+# SPDX-License-Identifier: GPL-2.0-only
+#
+# Fungible Ethernet driver configuration
+#
+
+config FUN_ETH
+	tristate "Fungible Ethernet device driver"
+	depends on PCI_MSI
+	depends on TLS && TLS_DEVICE || TLS_DEVICE=n
+	select NET_DEVLINK
+	select FUN_CORE
+	help
+	  This driver supports the Ethernet functionality of Fungible adapters.
+	  It works with both physical and virtual functions.
+
+	  To compile this driver as a module, choose M here. The module
+          will be called funeth.
diff --git a/drivers/net/ethernet/fungible/funeth/Makefile b/drivers/net/ethernet/fungible/funeth/Makefile
new file mode 100644
index 000000000000..d51e4c2b4a1a
--- /dev/null
+++ b/drivers/net/ethernet/fungible/funeth/Makefile
@@ -0,0 +1,10 @@
+# SPDX-License-Identifier: (GPL-2.0-only OR BSD-3-Clause)
+
+ccflags-y += -I$(src)/../funcore -I$(src)
+
+obj-$(CONFIG_FUN_ETH) += funeth.o
+
+funeth-y := funeth_main.o funeth_rx.o funeth_tx.o funeth_devlink.o \
+	    funeth_ethtool.o
+
+funeth-$(CONFIG_TLS_DEVICE) += funeth_ktls.o
diff --git a/drivers/net/ethernet/fungible/funeth/fun_port.h b/drivers/net/ethernet/fungible/funeth/fun_port.h
new file mode 100644
index 000000000000..0f9da44e3786
--- /dev/null
+++ b/drivers/net/ethernet/fungible/funeth/fun_port.h
@@ -0,0 +1,97 @@
+/* SPDX-License-Identifier: (GPL-2.0-only OR BSD-3-Clause) */
+
+#ifndef _FUN_PORT_H
+#define _FUN_PORT_H
+
+enum port_mac_rx_stats {
+	PORT_MAC_RX_etherStatsOctets = 0x0,
+	PORT_MAC_RX_OctetsReceivedOK = 0x1,
+	PORT_MAC_RX_aAlignmentErrors = 0x2,
+	PORT_MAC_RX_aPAUSEMACCtrlFramesReceived = 0x3,
+	PORT_MAC_RX_aFrameTooLongErrors = 0x4,
+	PORT_MAC_RX_aInRangeLengthErrors = 0x5,
+	PORT_MAC_RX_aFramesReceivedOK = 0x6,
+	PORT_MAC_RX_aFrameCheckSequenceErrors = 0x7,
+	PORT_MAC_RX_VLANReceivedOK = 0x8,
+	PORT_MAC_RX_ifInErrors = 0x9,
+	PORT_MAC_RX_ifInUcastPkts = 0xa,
+	PORT_MAC_RX_ifInMulticastPkts = 0xb,
+	PORT_MAC_RX_ifInBroadcastPkts = 0xc,
+	PORT_MAC_RX_etherStatsDropEvents = 0xd,
+	PORT_MAC_RX_etherStatsPkts = 0xe,
+	PORT_MAC_RX_etherStatsUndersizePkts = 0xf,
+	PORT_MAC_RX_etherStatsPkts64Octets = 0x10,
+	PORT_MAC_RX_etherStatsPkts65to127Octets = 0x11,
+	PORT_MAC_RX_etherStatsPkts128to255Octets = 0x12,
+	PORT_MAC_RX_etherStatsPkts256to511Octets = 0x13,
+	PORT_MAC_RX_etherStatsPkts512to1023Octets = 0x14,
+	PORT_MAC_RX_etherStatsPkts1024to1518Octets = 0x15,
+	PORT_MAC_RX_etherStatsPkts1519toMaxOctets = 0x16,
+	PORT_MAC_RX_etherStatsOversizePkts = 0x17,
+	PORT_MAC_RX_etherStatsJabbers = 0x18,
+	PORT_MAC_RX_etherStatsFragments = 0x19,
+	PORT_MAC_RX_CBFCPAUSEFramesReceived_0 = 0x1a,
+	PORT_MAC_RX_CBFCPAUSEFramesReceived_1 = 0x1b,
+	PORT_MAC_RX_CBFCPAUSEFramesReceived_2 = 0x1c,
+	PORT_MAC_RX_CBFCPAUSEFramesReceived_3 = 0x1d,
+	PORT_MAC_RX_CBFCPAUSEFramesReceived_4 = 0x1e,
+	PORT_MAC_RX_CBFCPAUSEFramesReceived_5 = 0x1f,
+	PORT_MAC_RX_CBFCPAUSEFramesReceived_6 = 0x20,
+	PORT_MAC_RX_CBFCPAUSEFramesReceived_7 = 0x21,
+	PORT_MAC_RX_CBFCPAUSEFramesReceived_8 = 0x22,
+	PORT_MAC_RX_CBFCPAUSEFramesReceived_9 = 0x23,
+	PORT_MAC_RX_CBFCPAUSEFramesReceived_10 = 0x24,
+	PORT_MAC_RX_CBFCPAUSEFramesReceived_11 = 0x25,
+	PORT_MAC_RX_CBFCPAUSEFramesReceived_12 = 0x26,
+	PORT_MAC_RX_CBFCPAUSEFramesReceived_13 = 0x27,
+	PORT_MAC_RX_CBFCPAUSEFramesReceived_14 = 0x28,
+	PORT_MAC_RX_CBFCPAUSEFramesReceived_15 = 0x29,
+	PORT_MAC_RX_MACControlFramesReceived = 0x2a,
+	PORT_MAC_RX_STATS_MAX = 0x2b,
+};
+
+enum port_mac_tx_stats {
+	PORT_MAC_TX_etherStatsOctets = 0x0,
+	PORT_MAC_TX_OctetsTransmittedOK = 0x1,
+	PORT_MAC_TX_aPAUSEMACCtrlFramesTransmitted = 0x2,
+	PORT_MAC_TX_aFramesTransmittedOK = 0x3,
+	PORT_MAC_TX_VLANTransmittedOK = 0x4,
+	PORT_MAC_TX_ifOutErrors = 0x5,
+	PORT_MAC_TX_ifOutUcastPkts = 0x6,
+	PORT_MAC_TX_ifOutMulticastPkts = 0x7,
+	PORT_MAC_TX_ifOutBroadcastPkts = 0x8,
+	PORT_MAC_TX_etherStatsPkts64Octets = 0x9,
+	PORT_MAC_TX_etherStatsPkts65to127Octets = 0xa,
+	PORT_MAC_TX_etherStatsPkts128to255Octets = 0xb,
+	PORT_MAC_TX_etherStatsPkts256to511Octets = 0xc,
+	PORT_MAC_TX_etherStatsPkts512to1023Octets = 0xd,
+	PORT_MAC_TX_etherStatsPkts1024to1518Octets = 0xe,
+	PORT_MAC_TX_etherStatsPkts1519toMaxOctets = 0xf,
+	PORT_MAC_TX_CBFCPAUSEFramesTransmitted_0 = 0x10,
+	PORT_MAC_TX_CBFCPAUSEFramesTransmitted_1 = 0x11,
+	PORT_MAC_TX_CBFCPAUSEFramesTransmitted_2 = 0x12,
+	PORT_MAC_TX_CBFCPAUSEFramesTransmitted_3 = 0x13,
+	PORT_MAC_TX_CBFCPAUSEFramesTransmitted_4 = 0x14,
+	PORT_MAC_TX_CBFCPAUSEFramesTransmitted_5 = 0x15,
+	PORT_MAC_TX_CBFCPAUSEFramesTransmitted_6 = 0x16,
+	PORT_MAC_TX_CBFCPAUSEFramesTransmitted_7 = 0x17,
+	PORT_MAC_TX_CBFCPAUSEFramesTransmitted_8 = 0x18,
+	PORT_MAC_TX_CBFCPAUSEFramesTransmitted_9 = 0x19,
+	PORT_MAC_TX_CBFCPAUSEFramesTransmitted_10 = 0x1a,
+	PORT_MAC_TX_CBFCPAUSEFramesTransmitted_11 = 0x1b,
+	PORT_MAC_TX_CBFCPAUSEFramesTransmitted_12 = 0x1c,
+	PORT_MAC_TX_CBFCPAUSEFramesTransmitted_13 = 0x1d,
+	PORT_MAC_TX_CBFCPAUSEFramesTransmitted_14 = 0x1e,
+	PORT_MAC_TX_CBFCPAUSEFramesTransmitted_15 = 0x1f,
+	PORT_MAC_TX_MACControlFramesTransmitted = 0x20,
+	PORT_MAC_TX_etherStatsPkts = 0x21,
+	PORT_MAC_TX_STATS_MAX = 0x22,
+};
+
+enum port_mac_fec_stats {
+	PORT_MAC_FEC_Correctable = 0x0,
+	PORT_MAC_FEC_Uncorrectable = 0x1,
+	PORT_MAC_FEC_STATS_MAX = 0x2,
+};
+
+#endif /* _FUN_PORT_H */
diff --git a/drivers/net/ethernet/fungible/funeth/funeth.h b/drivers/net/ethernet/fungible/funeth/funeth.h
new file mode 100644
index 000000000000..1250e10d21db
--- /dev/null
+++ b/drivers/net/ethernet/fungible/funeth/funeth.h
@@ -0,0 +1,171 @@
+/* SPDX-License-Identifier: (GPL-2.0-only OR BSD-3-Clause) */
+
+#ifndef _FUNETH_H
+#define _FUNETH_H
+
+#include <uapi/linux/if_ether.h>
+#include <uapi/linux/net_tstamp.h>
+#include <linux/mutex.h>
+#include <linux/seqlock.h>
+#include <linux/xarray.h>
+#include <net/devlink.h>
+#include "fun_dev.h"
+
+#define ADMIN_SQE_SIZE SZ_128
+#define ADMIN_CQE_SIZE SZ_64
+#define ADMIN_RSP_MAX_LEN (ADMIN_CQE_SIZE - sizeof(struct fun_cqe_info))
+
+#define FUN_MAX_MTU 9024
+
+#define SQ_DEPTH 512U
+#define CQ_DEPTH 1024U
+#define RQ_DEPTH (512U / (PAGE_SIZE / 4096))
+
+#define CQ_INTCOAL_USEC 10
+#define CQ_INTCOAL_NPKT 16
+#define SQ_INTCOAL_USEC 10
+#define SQ_INTCOAL_NPKT 16
+
+#define INVALID_LPORT 0xffff
+
+#define FUN_PORT_CAP_PAUSE_MASK (FUN_PORT_CAP_TX_PAUSE | FUN_PORT_CAP_RX_PAUSE)
+
+struct fun_vport_info {
+	u8 mac[ETH_ALEN];
+	u16 vlan;
+	__be16 vlan_proto;
+	u8 qos;
+	u8 spoofchk:1;
+	u8 trusted:1;
+	unsigned int max_rate;
+};
+
+/* "subclass" of fun_dev for Ethernet functions */
+struct fun_ethdev {
+	struct fun_dev fdev;
+
+	/* the function's network ports */
+	struct net_device **netdevs;
+	unsigned int num_ports;
+
+	/* configuration for the function's virtual ports */
+	unsigned int num_vports;
+	struct fun_vport_info *vport_info;
+
+	struct mutex state_mutex; /* nests inside RTNL if both taken */
+
+	unsigned int nsqs_per_port;
+};
+
+static inline struct fun_ethdev *to_fun_ethdev(struct fun_dev *p)
+{
+	return container_of(p, struct fun_ethdev, fdev);
+}
+
+struct fun_qset {
+	struct funeth_rxq **rxqs;
+	struct funeth_txq **txqs;
+	struct funeth_txq **xdpqs;
+	unsigned int nrxqs;
+	unsigned int ntxqs;
+	unsigned int nxdpqs;
+	unsigned int rxq_start;
+	unsigned int txq_start;
+	unsigned int xdpq_start;
+	unsigned int cq_depth;
+	unsigned int rq_depth;
+	unsigned int sq_depth;
+	int state;
+};
+
+/* Per netdevice driver state, i.e., netdev_priv. */
+struct funeth_priv {
+	struct fun_dev *fdev;
+	struct pci_dev *pdev;
+	struct net_device *netdev;
+
+	struct funeth_rxq * __rcu *rxqs;
+	struct funeth_txq **txqs;
+	struct funeth_txq * __rcu *xdpqs;
+
+	struct xarray irqs;
+	unsigned int num_tx_irqs;
+	unsigned int num_rx_irqs;
+	unsigned int rx_irq_ofst;
+
+	unsigned int lane_attrs;
+	u16 lport;
+
+	/* link settings */
+	u64 port_caps;
+	u64 advertising;
+	u64 lp_advertising;
+	unsigned int link_speed;
+	u8 xcvr_type;
+	u8 active_fc;
+	u8 active_fec;
+	u8 link_down_reason;
+	seqcount_t link_seq;
+
+	u32 msg_enable;
+
+	unsigned int num_xdpqs;
+
+	/* ethtool, etc. config parameters */
+	unsigned int sq_depth;
+	unsigned int rq_depth;
+	unsigned int cq_depth;
+	unsigned int cq_irq_db;
+	u8 tx_coal_usec;
+	u8 tx_coal_count;
+	u8 rx_coal_usec;
+	u8 rx_coal_count;
+
+	struct hwtstamp_config hwtstamp_cfg;
+
+	/* cumulative queue stats from earlier queue instances */
+	u64 tx_packets;
+	u64 tx_bytes;
+	u64 tx_dropped;
+	u64 rx_packets;
+	u64 rx_bytes;
+	u64 rx_dropped;
+
+	/* RSS */
+	unsigned int rss_hw_id;
+	enum fun_eth_hash_alg hash_algo;
+	u8 rss_key[FUN_ETH_RSS_MAX_KEY_SIZE];
+	unsigned int indir_table_nentries;
+	u32 indir_table[FUN_ETH_RSS_MAX_INDIR_ENT];
+	dma_addr_t rss_dma_addr;
+	void *rss_cfg;
+
+	/* DMA area for port stats */
+	dma_addr_t stats_dma_addr;
+	__be64 *stats;
+
+	struct bpf_prog *xdp_prog;
+
+	struct devlink_port dl_port;
+
+	/* kTLS state */
+	unsigned int ktls_id;
+	atomic64_t tx_tls_add;
+	atomic64_t tx_tls_del;
+	atomic64_t tx_tls_resync;
+};
+
+void fun_set_ethtool_ops(struct net_device *netdev);
+int fun_port_write_cmd(struct funeth_priv *fp, int key, u64 data);
+int fun_port_read_cmd(struct funeth_priv *fp, int key, u64 *data);
+int fun_create_and_bind_tx(struct funeth_priv *fp, u32 sqid);
+int fun_replace_queues(struct net_device *dev, struct fun_qset *newqs,
+		       struct netlink_ext_ack *extack);
+int fun_change_num_queues(struct net_device *dev, unsigned int ntx,
+			  unsigned int nrx);
+void fun_set_ring_count(struct net_device *netdev, unsigned int ntx,
+			unsigned int nrx);
+int fun_config_rss(struct net_device *dev, int algo, const u8 *key,
+		   const u32 *qtable, u8 op);
+
+#endif /* _FUNETH_H */
diff --git a/drivers/net/ethernet/fungible/funeth/funeth_devlink.c b/drivers/net/ethernet/fungible/funeth/funeth_devlink.c
new file mode 100644
index 000000000000..4fbeb3fd71a8
--- /dev/null
+++ b/drivers/net/ethernet/fungible/funeth/funeth_devlink.c
@@ -0,0 +1,27 @@
+// SPDX-License-Identifier: (GPL-2.0-only OR BSD-3-Clause)
+
+#include "funeth.h"
+#include "funeth_devlink.h"
+
+static const struct devlink_ops fun_dl_ops = {
+};
+
+struct devlink *fun_devlink_alloc(struct device *dev)
+{
+	return devlink_alloc(&fun_dl_ops, sizeof(struct fun_ethdev), dev);
+}
+
+void fun_devlink_free(struct devlink *devlink)
+{
+	devlink_free(devlink);
+}
+
+void fun_devlink_register(struct devlink *devlink)
+{
+	devlink_register(devlink);
+}
+
+void fun_devlink_unregister(struct devlink *devlink)
+{
+	devlink_unregister(devlink);
+}
diff --git a/drivers/net/ethernet/fungible/funeth/funeth_devlink.h b/drivers/net/ethernet/fungible/funeth/funeth_devlink.h
new file mode 100644
index 000000000000..e40464d57ff4
--- /dev/null
+++ b/drivers/net/ethernet/fungible/funeth/funeth_devlink.h
@@ -0,0 +1,13 @@
+/* SPDX-License-Identifier: (GPL-2.0-only OR BSD-3-Clause) */
+
+#ifndef __FUNETH_DEVLINK_H
+#define __FUNETH_DEVLINK_H
+
+#include <net/devlink.h>
+
+struct devlink *fun_devlink_alloc(struct device *dev);
+void fun_devlink_free(struct devlink *devlink);
+void fun_devlink_register(struct devlink *devlink);
+void fun_devlink_unregister(struct devlink *devlink);
+
+#endif /* __FUNETH_DEVLINK_H */
diff --git a/drivers/net/ethernet/fungible/funeth/funeth_ethtool.c b/drivers/net/ethernet/fungible/funeth/funeth_ethtool.c
new file mode 100644
index 000000000000..1966dba512f8
--- /dev/null
+++ b/drivers/net/ethernet/fungible/funeth/funeth_ethtool.c
@@ -0,0 +1,1196 @@
+// SPDX-License-Identifier: (GPL-2.0-only OR BSD-3-Clause)
+
+#include <linux/ethtool.h>
+#include <linux/linkmode.h>
+#include <linux/netdevice.h>
+#include <linux/nvme.h>
+#include <linux/io.h>
+#include <linux/io-64-nonatomic-lo-hi.h>
+#include <linux/pci.h>
+#include <linux/rtnetlink.h>
+#include "funeth.h"
+#include "fun_port.h"
+#include "funeth_txrx.h"
+
+/* Min queue depth. The smallest power-of-2 supporting jumbo frames with 4K
+ * pages is 8. Require it for all types of queues though some could work with
+ * fewer entries.
+ */
+#define FUNETH_MIN_QDEPTH 8
+
+static const char mac_tx_stat_names[][ETH_GSTRING_LEN] = {
+	"mac_tx_octets_total",
+	"mac_tx_frames_total",
+	"mac_tx_vlan_frames_ok",
+	"mac_tx_unicast_frames",
+	"mac_tx_multicast_frames",
+	"mac_tx_broadcast_frames",
+	"mac_tx_errors",
+	"mac_tx_CBFCPAUSE0",
+	"mac_tx_CBFCPAUSE1",
+	"mac_tx_CBFCPAUSE2",
+	"mac_tx_CBFCPAUSE3",
+	"mac_tx_CBFCPAUSE4",
+	"mac_tx_CBFCPAUSE5",
+	"mac_tx_CBFCPAUSE6",
+	"mac_tx_CBFCPAUSE7",
+	"mac_tx_CBFCPAUSE8",
+	"mac_tx_CBFCPAUSE9",
+	"mac_tx_CBFCPAUSE10",
+	"mac_tx_CBFCPAUSE11",
+	"mac_tx_CBFCPAUSE12",
+	"mac_tx_CBFCPAUSE13",
+	"mac_tx_CBFCPAUSE14",
+	"mac_tx_CBFCPAUSE15",
+};
+
+static const char mac_rx_stat_names[][ETH_GSTRING_LEN] = {
+	"mac_rx_octets_total",
+	"mac_rx_frames_total",
+	"mac_rx_VLAN_frames_ok",
+	"mac_rx_unicast_frames",
+	"mac_rx_multicast_frames",
+	"mac_rx_broadcast_frames",
+	"mac_rx_drop_events",
+	"mac_rx_errors",
+	"mac_rx_alignment_errors",
+	"mac_rx_CBFCPAUSE0",
+	"mac_rx_CBFCPAUSE1",
+	"mac_rx_CBFCPAUSE2",
+	"mac_rx_CBFCPAUSE3",
+	"mac_rx_CBFCPAUSE4",
+	"mac_rx_CBFCPAUSE5",
+	"mac_rx_CBFCPAUSE6",
+	"mac_rx_CBFCPAUSE7",
+	"mac_rx_CBFCPAUSE8",
+	"mac_rx_CBFCPAUSE9",
+	"mac_rx_CBFCPAUSE10",
+	"mac_rx_CBFCPAUSE11",
+	"mac_rx_CBFCPAUSE12",
+	"mac_rx_CBFCPAUSE13",
+	"mac_rx_CBFCPAUSE14",
+	"mac_rx_CBFCPAUSE15",
+};
+
+static const char * const txq_stat_names[] = {
+	"tx_pkts",
+	"tx_bytes",
+	"tx_cso",
+	"tx_tso",
+	"tx_encapsulated_tso",
+	"tx_uso",
+	"tx_more",
+	"tx_queue_stops",
+	"tx_queue_restarts",
+	"tx_mapping_errors",
+	"tx_tls_encrypted_packets",
+	"tx_tls_encrypted_bytes",
+	"tx_tls_ooo",
+	"tx_tls_drop_no_sync_data",
+};
+
+static const char * const xdpq_stat_names[] = {
+	"tx_xdp_pkts",
+	"tx_xdp_bytes",
+	"tx_xdp_full",
+	"tx_xdp_mapping_errors",
+};
+
+static const char * const rxq_stat_names[] = {
+	"rx_pkts",
+	"rx_bytes",
+	"rx_cso",
+	"gro_pkts",
+	"gro_merged",
+	"rx_xdp_tx",
+	"rx_xdp_redir",
+	"rx_xdp_drops",
+	"rx_buffers",
+	"rx_page_allocs",
+	"rx_drops",
+	"rx_budget_exhausted",
+	"rx_mapping_errors",
+};
+
+static const char * const tls_stat_names[] = {
+	"tx_tls_ctx",
+	"tx_tls_del",
+	"tx_tls_resync",
+};
+
+static void fun_link_modes_to_ethtool(u64 modes,
+				      unsigned long *ethtool_modes_map)
+{
+#define ADD_LINK_MODE(mode) \
+	__set_bit(ETHTOOL_LINK_MODE_ ## mode ## _BIT, ethtool_modes_map)
+
+	if (modes & FUN_PORT_CAP_AUTONEG)
+		ADD_LINK_MODE(Autoneg);
+	if (modes & FUN_PORT_CAP_1000_X)
+		ADD_LINK_MODE(1000baseX_Full);
+	if (modes & FUN_PORT_CAP_10G_R) {
+		ADD_LINK_MODE(10000baseCR_Full);
+		ADD_LINK_MODE(10000baseSR_Full);
+		ADD_LINK_MODE(10000baseLR_Full);
+		ADD_LINK_MODE(10000baseER_Full);
+	}
+	if (modes & FUN_PORT_CAP_25G_R) {
+		ADD_LINK_MODE(25000baseCR_Full);
+		ADD_LINK_MODE(25000baseSR_Full);
+	}
+	if (modes & FUN_PORT_CAP_40G_R4) {
+		ADD_LINK_MODE(40000baseCR4_Full);
+		ADD_LINK_MODE(40000baseSR4_Full);
+		ADD_LINK_MODE(40000baseLR4_Full);
+	}
+	if (modes & FUN_PORT_CAP_50G_R2) {
+		ADD_LINK_MODE(50000baseCR2_Full);
+		ADD_LINK_MODE(50000baseSR2_Full);
+	}
+	if (modes & FUN_PORT_CAP_50G_R) {
+		ADD_LINK_MODE(50000baseCR_Full);
+		ADD_LINK_MODE(50000baseSR_Full);
+		ADD_LINK_MODE(50000baseLR_ER_FR_Full);
+	}
+	if (modes & FUN_PORT_CAP_100G_R4) {
+		ADD_LINK_MODE(100000baseCR4_Full);
+		ADD_LINK_MODE(100000baseSR4_Full);
+		ADD_LINK_MODE(100000baseLR4_ER4_Full);
+	}
+	if (modes & FUN_PORT_CAP_100G_R2) {
+		ADD_LINK_MODE(100000baseCR2_Full);
+		ADD_LINK_MODE(100000baseSR2_Full);
+		ADD_LINK_MODE(100000baseLR2_ER2_FR2_Full);
+	}
+	if (modes & FUN_PORT_CAP_FEC_NONE)
+		ADD_LINK_MODE(FEC_NONE);
+	if (modes & FUN_PORT_CAP_FEC_FC)
+		ADD_LINK_MODE(FEC_BASER);
+	if (modes & FUN_PORT_CAP_FEC_RS)
+		ADD_LINK_MODE(FEC_RS);
+	if (modes & FUN_PORT_CAP_RX_PAUSE)
+		ADD_LINK_MODE(Pause);
+
+#undef ADD_LINK_MODE
+}
+
+static void set_asym_pause(u64 advertising, struct ethtool_link_ksettings *ks)
+{
+	bool rx_pause, tx_pause;
+
+	rx_pause = advertising & FUN_PORT_CAP_RX_PAUSE;
+	tx_pause = advertising & FUN_PORT_CAP_TX_PAUSE;
+	if (tx_pause ^ rx_pause)
+		ethtool_link_ksettings_add_link_mode(ks, advertising,
+						     Asym_Pause);
+}
+
+static unsigned int fun_port_type(unsigned int xcvr)
+{
+	if (!xcvr)
+		return PORT_NONE;
+
+	switch (xcvr & 7) {
+	case FUN_XCVR_BASET:
+		return PORT_TP;
+	case FUN_XCVR_CU:
+		return PORT_DA;
+	default:
+		return PORT_FIBRE;
+	}
+}
+
+static int fun_get_link_ksettings(struct net_device *netdev,
+				  struct ethtool_link_ksettings *ks)
+{
+	const struct funeth_priv *fp = netdev_priv(netdev);
+	unsigned int seq, speed, xcvr;
+	u64 lp_advertising;
+	bool link_up;
+
+	ethtool_link_ksettings_zero_link_mode(ks, supported);
+	ethtool_link_ksettings_zero_link_mode(ks, advertising);
+	ethtool_link_ksettings_zero_link_mode(ks, lp_advertising);
+
+	/* Link settings change asynchronously, take a consistent snapshot */
+	do {
+		seq = read_seqcount_begin(&fp->link_seq);
+		link_up = netif_carrier_ok(netdev);
+		speed = fp->link_speed;
+		xcvr = fp->xcvr_type;
+		lp_advertising = fp->lp_advertising;
+	} while (read_seqcount_retry(&fp->link_seq, seq));
+
+	if (link_up) {
+		ks->base.speed = speed;
+		ks->base.duplex = DUPLEX_FULL;
+		fun_link_modes_to_ethtool(lp_advertising,
+					  ks->link_modes.lp_advertising);
+	} else {
+		ks->base.speed = SPEED_UNKNOWN;
+		ks->base.duplex = DUPLEX_UNKNOWN;
+	}
+
+	ks->base.autoneg = (fp->advertising & FUN_PORT_CAP_AUTONEG) ?
+			   AUTONEG_ENABLE : AUTONEG_DISABLE;
+	ks->base.port = fun_port_type(xcvr);
+
+	fun_link_modes_to_ethtool(fp->port_caps, ks->link_modes.supported);
+	if (fp->port_caps & (FUN_PORT_CAP_RX_PAUSE | FUN_PORT_CAP_TX_PAUSE))
+		ethtool_link_ksettings_add_link_mode(ks, supported, Asym_Pause);
+
+	fun_link_modes_to_ethtool(fp->advertising, ks->link_modes.advertising);
+	set_asym_pause(fp->advertising, ks);
+	return 0;
+}
+
+static u64 fun_advert_modes(const struct ethtool_link_ksettings *ks)
+{
+	u64 modes = 0;
+
+#define HAS_MODE(mode) \
+	ethtool_link_ksettings_test_link_mode(ks, advertising, mode)
+
+	if (HAS_MODE(1000baseX_Full))
+		modes |= FUN_PORT_CAP_1000_X;
+	if (HAS_MODE(10000baseCR_Full) || HAS_MODE(10000baseSR_Full) ||
+	    HAS_MODE(10000baseLR_Full) || HAS_MODE(10000baseER_Full))
+		modes |= FUN_PORT_CAP_10G_R;
+	if (HAS_MODE(25000baseCR_Full) || HAS_MODE(25000baseSR_Full))
+		modes |= FUN_PORT_CAP_25G_R;
+	if (HAS_MODE(40000baseCR4_Full) || HAS_MODE(40000baseSR4_Full) ||
+	    HAS_MODE(40000baseLR4_Full))
+		modes |= FUN_PORT_CAP_40G_R4;
+	if (HAS_MODE(50000baseCR2_Full) || HAS_MODE(50000baseSR2_Full))
+		modes |= FUN_PORT_CAP_50G_R2;
+	if (HAS_MODE(50000baseCR_Full) || HAS_MODE(50000baseSR_Full) ||
+	    HAS_MODE(50000baseLR_ER_FR_Full))
+		modes |= FUN_PORT_CAP_50G_R;
+	if (HAS_MODE(100000baseCR4_Full) || HAS_MODE(100000baseSR4_Full) ||
+	    HAS_MODE(100000baseLR4_ER4_Full))
+		modes |= FUN_PORT_CAP_100G_R4;
+	if (HAS_MODE(100000baseCR2_Full) || HAS_MODE(100000baseSR2_Full) ||
+	    HAS_MODE(100000baseLR2_ER2_FR2_Full))
+		modes |= FUN_PORT_CAP_100G_R2;
+
+	return modes;
+#undef HAS_MODE
+}
+
+static u64 fun_speed_to_link_mode(unsigned int speed)
+{
+	switch (speed) {
+	case SPEED_100000:
+		return FUN_PORT_CAP_100G_R4 | FUN_PORT_CAP_100G_R2;
+	case SPEED_50000:
+		return FUN_PORT_CAP_50G_R | FUN_PORT_CAP_50G_R2;
+	case SPEED_40000:
+		return FUN_PORT_CAP_40G_R4;
+	case SPEED_25000:
+		return FUN_PORT_CAP_25G_R;
+	case SPEED_10000:
+		return FUN_PORT_CAP_10G_R;
+	case SPEED_1000:
+		return FUN_PORT_CAP_1000_X;
+	default:
+		return 0;
+	}
+}
+
+static int fun_change_advert(struct funeth_priv *fp, u64 new_advert)
+{
+	int err;
+
+	if (new_advert == fp->advertising)
+		return 0;
+
+	err = fun_port_write_cmd(fp, FUN_ADMIN_PORT_KEY_ADVERT, new_advert);
+	if (!err)
+		fp->advertising = new_advert;
+	return err;
+}
+
+#define FUN_PORT_CAP_FEC_MASK \
+	(FUN_PORT_CAP_FEC_NONE | FUN_PORT_CAP_FEC_FC | FUN_PORT_CAP_FEC_RS)
+
+static int fun_set_link_ksettings(struct net_device *netdev,
+				  const struct ethtool_link_ksettings *ks)
+{
+	__ETHTOOL_DECLARE_LINK_MODE_MASK(supported) = {};
+	struct funeth_priv *fp = netdev_priv(netdev);
+	u64 new_advert;
+
+	/* eswitch ports don't support mode changes */
+	if (fp->port_caps & FUN_PORT_CAP_VPORT)
+		return -EOPNOTSUPP;
+
+	if (ks->base.duplex == DUPLEX_HALF)
+		return -EINVAL;
+	if (ks->base.autoneg == AUTONEG_ENABLE &&
+	    !(fp->port_caps & FUN_PORT_CAP_AUTONEG))
+		return -EINVAL;
+
+	if (ks->base.autoneg == AUTONEG_ENABLE) {
+		if (linkmode_empty(ks->link_modes.advertising))
+			return -EINVAL;
+
+		fun_link_modes_to_ethtool(fp->port_caps, supported);
+		if (!linkmode_subset(ks->link_modes.advertising, supported))
+			return -EINVAL;
+
+		new_advert = fun_advert_modes(ks) | FUN_PORT_CAP_AUTONEG;
+	} else {
+		new_advert = fun_speed_to_link_mode(ks->base.speed);
+		new_advert &= fp->port_caps;
+		if (!new_advert)
+			return -EINVAL;
+	}
+	new_advert |= fp->advertising &
+		      (FUN_PORT_CAP_PAUSE_MASK | FUN_PORT_CAP_FEC_MASK);
+
+	return fun_change_advert(fp, new_advert);
+}
+
+static void fun_get_pauseparam(struct net_device *netdev,
+			       struct ethtool_pauseparam *pause)
+{
+	const struct funeth_priv *fp = netdev_priv(netdev);
+	u8 active_pause = fp->active_fc;
+
+	pause->rx_pause = !!(active_pause & FUN_PORT_CAP_RX_PAUSE);
+	pause->tx_pause = !!(active_pause & FUN_PORT_CAP_TX_PAUSE);
+	pause->autoneg = !!(fp->advertising & FUN_PORT_CAP_AUTONEG);
+}
+
+static int fun_set_pauseparam(struct net_device *netdev,
+			      struct ethtool_pauseparam *pause)
+{
+	struct funeth_priv *fp = netdev_priv(netdev);
+	u64 new_advert;
+
+	if (fp->port_caps & FUN_PORT_CAP_VPORT)
+		return -EOPNOTSUPP;
+	/* Forcing PAUSE settings with AN enabled is unsupported. */
+	if (!pause->autoneg && (fp->advertising & FUN_PORT_CAP_AUTONEG))
+		return -EOPNOTSUPP;
+	if (pause->autoneg && !(fp->advertising & FUN_PORT_CAP_AUTONEG))
+		return -EINVAL;
+	if (pause->tx_pause && !(fp->port_caps & FUN_PORT_CAP_TX_PAUSE))
+		return -EINVAL;
+	if (pause->rx_pause && !(fp->port_caps & FUN_PORT_CAP_RX_PAUSE))
+		return -EINVAL;
+
+	new_advert = fp->advertising & ~FUN_PORT_CAP_PAUSE_MASK;
+	if (pause->tx_pause)
+		new_advert |= FUN_PORT_CAP_TX_PAUSE;
+	if (pause->rx_pause)
+		new_advert |= FUN_PORT_CAP_RX_PAUSE;
+
+	return fun_change_advert(fp, new_advert);
+}
+
+static int fun_restart_an(struct net_device *netdev)
+{
+	struct funeth_priv *fp = netdev_priv(netdev);
+
+	if (!(fp->advertising & FUN_PORT_CAP_AUTONEG))
+		return -EOPNOTSUPP;
+
+	return fun_port_write_cmd(fp, FUN_ADMIN_PORT_KEY_ADVERT,
+				  FUN_PORT_CAP_AUTONEG);
+}
+
+static int fun_set_phys_id(struct net_device *netdev,
+			   enum ethtool_phys_id_state state)
+{
+	struct funeth_priv *fp = netdev_priv(netdev);
+	unsigned int beacon;
+
+	if (fp->port_caps & FUN_PORT_CAP_VPORT)
+		return -EOPNOTSUPP;
+	if (state != ETHTOOL_ID_ACTIVE && state != ETHTOOL_ID_INACTIVE)
+		return -EOPNOTSUPP;
+
+	beacon = state == ETHTOOL_ID_ACTIVE ? FUN_PORT_LED_BEACON_ON :
+					      FUN_PORT_LED_BEACON_OFF;
+	return fun_port_write_cmd(fp, FUN_ADMIN_PORT_KEY_LED, beacon);
+}
+
+static void fun_get_drvinfo(struct net_device *netdev,
+			    struct ethtool_drvinfo *info)
+{
+	const struct funeth_priv *fp = netdev_priv(netdev);
+
+	strscpy(info->driver, KBUILD_MODNAME, sizeof(info->driver));
+	strscpy(info->bus_info, pci_name(fp->pdev), sizeof(info->bus_info));
+}
+
+static u32 fun_get_msglevel(struct net_device *netdev)
+{
+	const struct funeth_priv *fp = netdev_priv(netdev);
+
+	return fp->msg_enable;
+}
+
+static void fun_set_msglevel(struct net_device *netdev, u32 value)
+{
+	struct funeth_priv *fp = netdev_priv(netdev);
+
+	fp->msg_enable = value;
+}
+
+static int fun_get_regs_len(struct net_device *dev)
+{
+	return NVME_REG_ACQ + sizeof(u64);
+}
+
+static void fun_get_regs(struct net_device *dev, struct ethtool_regs *regs,
+			 void *buf)
+{
+	const struct funeth_priv *fp = netdev_priv(dev);
+	void __iomem *bar = fp->fdev->bar;
+
+	regs->version = 0;
+	*(u64 *)(buf + NVME_REG_CAP)   = readq(bar + NVME_REG_CAP);
+	*(u32 *)(buf + NVME_REG_VS)    = readl(bar + NVME_REG_VS);
+	*(u32 *)(buf + NVME_REG_INTMS) = readl(bar + NVME_REG_INTMS);
+	*(u32 *)(buf + NVME_REG_INTMC) = readl(bar + NVME_REG_INTMC);
+	*(u32 *)(buf + NVME_REG_CC)    = readl(bar + NVME_REG_CC);
+	*(u32 *)(buf + NVME_REG_CSTS)  = readl(bar + NVME_REG_CSTS);
+	*(u32 *)(buf + NVME_REG_AQA)   = readl(bar + NVME_REG_AQA);
+	*(u64 *)(buf + NVME_REG_ASQ)   = readq(bar + NVME_REG_ASQ);
+	*(u64 *)(buf + NVME_REG_ACQ)   = readq(bar + NVME_REG_ACQ);
+}
+
+static int fun_get_coalesce(struct net_device *netdev,
+			    struct ethtool_coalesce *coal,
+			    struct kernel_ethtool_coalesce *kcoal,
+			    struct netlink_ext_ack *ext_ack)
+{
+	const struct funeth_priv *fp = netdev_priv(netdev);
+
+	coal->rx_coalesce_usecs        = fp->rx_coal_usec;
+	coal->rx_max_coalesced_frames  = fp->rx_coal_count;
+	coal->use_adaptive_rx_coalesce = !fp->cq_irq_db;
+	coal->tx_coalesce_usecs        = fp->tx_coal_usec;
+	coal->tx_max_coalesced_frames  = fp->tx_coal_count;
+	return 0;
+}
+
+static int fun_set_coalesce(struct net_device *netdev,
+			    struct ethtool_coalesce *coal,
+			    struct kernel_ethtool_coalesce *kcoal,
+			    struct netlink_ext_ack *ext_ack)
+{
+	struct funeth_priv *fp = netdev_priv(netdev);
+	struct funeth_rxq **rxqs;
+	unsigned int i, db_val;
+
+	if (coal->rx_coalesce_usecs > FUN_DB_INTCOAL_USEC_M ||
+	    coal->rx_max_coalesced_frames > FUN_DB_INTCOAL_ENTRIES_M ||
+	    (coal->rx_coalesce_usecs | coal->rx_max_coalesced_frames) == 0 ||
+	    coal->tx_coalesce_usecs > FUN_DB_INTCOAL_USEC_M ||
+	    coal->tx_max_coalesced_frames > FUN_DB_INTCOAL_ENTRIES_M ||
+	    (coal->tx_coalesce_usecs | coal->tx_max_coalesced_frames) == 0)
+		return -EINVAL;
+
+	/* a timer is required if there's any coalescing */
+	if ((coal->rx_max_coalesced_frames > 1 && !coal->rx_coalesce_usecs) ||
+	    (coal->tx_max_coalesced_frames > 1 && !coal->tx_coalesce_usecs))
+		return -EINVAL;
+
+	fp->rx_coal_usec  = coal->rx_coalesce_usecs;
+	fp->rx_coal_count = coal->rx_max_coalesced_frames;
+	fp->tx_coal_usec  = coal->tx_coalesce_usecs;
+	fp->tx_coal_count = coal->tx_max_coalesced_frames;
+
+	db_val = FUN_IRQ_CQ_DB(fp->rx_coal_usec, fp->rx_coal_count);
+	WRITE_ONCE(fp->cq_irq_db, db_val);
+
+	rxqs = rtnl_dereference(fp->rxqs);
+	if (!rxqs)
+		return 0;
+
+	for (i = 0; i < netdev->real_num_rx_queues; i++)
+		WRITE_ONCE(rxqs[i]->irq_db_val, db_val);
+
+	db_val = FUN_IRQ_SQ_DB(fp->tx_coal_usec, fp->tx_coal_count);
+	for (i = 0; i < netdev->real_num_tx_queues; i++)
+		WRITE_ONCE(fp->txqs[i]->irq_db_val, db_val);
+
+	return 0;
+}
+
+static void fun_get_channels(struct net_device *netdev,
+			     struct ethtool_channels *chan)
+{
+	chan->max_rx   = netdev->num_rx_queues;
+	chan->rx_count = netdev->real_num_rx_queues;
+
+	chan->max_tx   = netdev->num_tx_queues;
+	chan->tx_count = netdev->real_num_tx_queues;
+}
+
+static int fun_set_channels(struct net_device *netdev,
+			    struct ethtool_channels *chan)
+{
+	if (!chan->tx_count || !chan->rx_count)
+		return -EINVAL;
+
+	if (chan->tx_count == netdev->real_num_tx_queues &&
+	    chan->rx_count == netdev->real_num_rx_queues)
+		return 0;
+
+	if (netif_running(netdev))
+		return fun_change_num_queues(netdev, chan->tx_count,
+					     chan->rx_count);
+
+	fun_set_ring_count(netdev, chan->tx_count, chan->rx_count);
+	return 0;
+}
+
+static void fun_get_ringparam(struct net_device *netdev,
+			      struct ethtool_ringparam *ring,
+			      struct kernel_ethtool_ringparam *kring,
+			      struct netlink_ext_ack *extack)
+{
+	const struct funeth_priv *fp = netdev_priv(netdev);
+	unsigned int max_depth = fp->fdev->q_depth;
+
+	/* We size CQs to be twice the RQ depth so max RQ depth is half the
+	 * max queue depth.
+	 */
+	ring->rx_max_pending = max_depth / 2;
+	ring->tx_max_pending = max_depth;
+
+	ring->rx_pending = fp->rq_depth;
+	ring->tx_pending = fp->sq_depth;
+
+	kring->rx_buf_len = PAGE_SIZE;
+	kring->cqe_size = FUNETH_CQE_SIZE;
+}
+
+static int fun_set_ringparam(struct net_device *netdev,
+			     struct ethtool_ringparam *ring,
+			     struct kernel_ethtool_ringparam *kring,
+			     struct netlink_ext_ack *extack)
+{
+	struct funeth_priv *fp = netdev_priv(netdev);
+	int rc;
+
+	if (ring->rx_mini_pending || ring->rx_jumbo_pending)
+		return -EINVAL;
+
+	/* queue depths must be powers-of-2 */
+	if (!is_power_of_2(ring->rx_pending) ||
+	    !is_power_of_2(ring->tx_pending))
+		return -EINVAL;
+
+	if (ring->rx_pending < FUNETH_MIN_QDEPTH ||
+	    ring->tx_pending < FUNETH_MIN_QDEPTH)
+		return -EINVAL;
+
+	if (fp->sq_depth == ring->tx_pending &&
+	    fp->rq_depth == ring->rx_pending)
+		return 0;
+
+	if (netif_running(netdev)) {
+		struct fun_qset req = {
+			.cq_depth = 2 * ring->rx_pending,
+			.rq_depth = ring->rx_pending,
+			.sq_depth = ring->tx_pending
+		};
+
+		rc = fun_replace_queues(netdev, &req, extack);
+		if (rc)
+			return rc;
+	}
+
+	fp->sq_depth = ring->tx_pending;
+	fp->rq_depth = ring->rx_pending;
+	fp->cq_depth = 2 * fp->rq_depth;
+	return 0;
+}
+
+static int fun_get_sset_count(struct net_device *dev, int sset)
+{
+	const struct funeth_priv *fp = netdev_priv(dev);
+	int n;
+
+	switch (sset) {
+	case ETH_SS_STATS:
+		n = (dev->real_num_tx_queues + 1) * ARRAY_SIZE(txq_stat_names) +
+		    (dev->real_num_rx_queues + 1) * ARRAY_SIZE(rxq_stat_names) +
+		    (fp->num_xdpqs + 1) * ARRAY_SIZE(xdpq_stat_names) +
+		    ARRAY_SIZE(tls_stat_names);
+		if (fp->port_caps & FUN_PORT_CAP_STATS) {
+			n += ARRAY_SIZE(mac_tx_stat_names) +
+			     ARRAY_SIZE(mac_rx_stat_names);
+		}
+		return n;
+	default:
+		break;
+	}
+	return 0;
+}
+
+static void fun_get_strings(struct net_device *netdev, u32 sset, u8 *data)
+{
+	const struct funeth_priv *fp = netdev_priv(netdev);
+	unsigned int i, j;
+	u8 *p = data;
+
+	switch (sset) {
+	case ETH_SS_STATS:
+		if (fp->port_caps & FUN_PORT_CAP_STATS) {
+			memcpy(p, mac_tx_stat_names, sizeof(mac_tx_stat_names));
+			p += sizeof(mac_tx_stat_names);
+			memcpy(p, mac_rx_stat_names, sizeof(mac_rx_stat_names));
+			p += sizeof(mac_rx_stat_names);
+		}
+
+		for (i = 0; i < netdev->real_num_tx_queues; i++) {
+			for (j = 0; j < ARRAY_SIZE(txq_stat_names); j++)
+				ethtool_sprintf(&p, "%s[%u]", txq_stat_names[j],
+						i);
+		}
+		for (j = 0; j < ARRAY_SIZE(txq_stat_names); j++)
+			ethtool_puts(&p, txq_stat_names[j]);
+
+		for (i = 0; i < fp->num_xdpqs; i++) {
+			for (j = 0; j < ARRAY_SIZE(xdpq_stat_names); j++)
+				ethtool_sprintf(&p, "%s[%u]",
+						xdpq_stat_names[j], i);
+		}
+		for (j = 0; j < ARRAY_SIZE(xdpq_stat_names); j++)
+			ethtool_puts(&p, xdpq_stat_names[j]);
+
+		for (i = 0; i < netdev->real_num_rx_queues; i++) {
+			for (j = 0; j < ARRAY_SIZE(rxq_stat_names); j++)
+				ethtool_sprintf(&p, "%s[%u]", rxq_stat_names[j],
+						i);
+		}
+		for (j = 0; j < ARRAY_SIZE(rxq_stat_names); j++)
+			ethtool_puts(&p, rxq_stat_names[j]);
+
+		for (j = 0; j < ARRAY_SIZE(tls_stat_names); j++)
+			ethtool_puts(&p, tls_stat_names[j]);
+		break;
+	default:
+		break;
+	}
+}
+
+static u64 *get_mac_stats(const struct funeth_priv *fp, u64 *data)
+{
+#define TX_STAT(s) \
+	*data++ = be64_to_cpu(fp->stats[PORT_MAC_RX_STATS_MAX + PORT_MAC_TX_##s])
+
+	TX_STAT(etherStatsOctets);
+	TX_STAT(etherStatsPkts);
+	TX_STAT(VLANTransmittedOK);
+	TX_STAT(ifOutUcastPkts);
+	TX_STAT(ifOutMulticastPkts);
+	TX_STAT(ifOutBroadcastPkts);
+	TX_STAT(ifOutErrors);
+	TX_STAT(CBFCPAUSEFramesTransmitted_0);
+	TX_STAT(CBFCPAUSEFramesTransmitted_1);
+	TX_STAT(CBFCPAUSEFramesTransmitted_2);
+	TX_STAT(CBFCPAUSEFramesTransmitted_3);
+	TX_STAT(CBFCPAUSEFramesTransmitted_4);
+	TX_STAT(CBFCPAUSEFramesTransmitted_5);
+	TX_STAT(CBFCPAUSEFramesTransmitted_6);
+	TX_STAT(CBFCPAUSEFramesTransmitted_7);
+	TX_STAT(CBFCPAUSEFramesTransmitted_8);
+	TX_STAT(CBFCPAUSEFramesTransmitted_9);
+	TX_STAT(CBFCPAUSEFramesTransmitted_10);
+	TX_STAT(CBFCPAUSEFramesTransmitted_11);
+	TX_STAT(CBFCPAUSEFramesTransmitted_12);
+	TX_STAT(CBFCPAUSEFramesTransmitted_13);
+	TX_STAT(CBFCPAUSEFramesTransmitted_14);
+	TX_STAT(CBFCPAUSEFramesTransmitted_15);
+
+#define RX_STAT(s) *data++ = be64_to_cpu(fp->stats[PORT_MAC_RX_##s])
+
+	RX_STAT(etherStatsOctets);
+	RX_STAT(etherStatsPkts);
+	RX_STAT(VLANReceivedOK);
+	RX_STAT(ifInUcastPkts);
+	RX_STAT(ifInMulticastPkts);
+	RX_STAT(ifInBroadcastPkts);
+	RX_STAT(etherStatsDropEvents);
+	RX_STAT(ifInErrors);
+	RX_STAT(aAlignmentErrors);
+	RX_STAT(CBFCPAUSEFramesReceived_0);
+	RX_STAT(CBFCPAUSEFramesReceived_1);
+	RX_STAT(CBFCPAUSEFramesReceived_2);
+	RX_STAT(CBFCPAUSEFramesReceived_3);
+	RX_STAT(CBFCPAUSEFramesReceived_4);
+	RX_STAT(CBFCPAUSEFramesReceived_5);
+	RX_STAT(CBFCPAUSEFramesReceived_6);
+	RX_STAT(CBFCPAUSEFramesReceived_7);
+	RX_STAT(CBFCPAUSEFramesReceived_8);
+	RX_STAT(CBFCPAUSEFramesReceived_9);
+	RX_STAT(CBFCPAUSEFramesReceived_10);
+	RX_STAT(CBFCPAUSEFramesReceived_11);
+	RX_STAT(CBFCPAUSEFramesReceived_12);
+	RX_STAT(CBFCPAUSEFramesReceived_13);
+	RX_STAT(CBFCPAUSEFramesReceived_14);
+	RX_STAT(CBFCPAUSEFramesReceived_15);
+
+	return data;
+
+#undef TX_STAT
+#undef RX_STAT
+}
+
+static void fun_get_ethtool_stats(struct net_device *netdev,
+				  struct ethtool_stats *stats, u64 *data)
+{
+	const struct funeth_priv *fp = netdev_priv(netdev);
+	struct funeth_txq_stats txs;
+	struct funeth_rxq_stats rxs;
+	struct funeth_txq **xdpqs;
+	struct funeth_rxq **rxqs;
+	unsigned int i, start;
+	u64 *totals, *tot;
+
+	if (fp->port_caps & FUN_PORT_CAP_STATS)
+		data = get_mac_stats(fp, data);
+
+	rxqs = rtnl_dereference(fp->rxqs);
+	if (!rxqs)
+		return;
+
+#define ADD_STAT(cnt) do { \
+	*data = (cnt); *tot++ += *data++; \
+} while (0)
+
+	/* Tx queues */
+	totals = data + netdev->real_num_tx_queues * ARRAY_SIZE(txq_stat_names);
+
+	for (i = 0; i < netdev->real_num_tx_queues; i++) {
+		tot = totals;
+
+		FUN_QSTAT_READ(fp->txqs[i], start, txs);
+
+		ADD_STAT(txs.tx_pkts);
+		ADD_STAT(txs.tx_bytes);
+		ADD_STAT(txs.tx_cso);
+		ADD_STAT(txs.tx_tso);
+		ADD_STAT(txs.tx_encap_tso);
+		ADD_STAT(txs.tx_uso);
+		ADD_STAT(txs.tx_more);
+		ADD_STAT(txs.tx_nstops);
+		ADD_STAT(txs.tx_nrestarts);
+		ADD_STAT(txs.tx_map_err);
+		ADD_STAT(txs.tx_tls_pkts);
+		ADD_STAT(txs.tx_tls_bytes);
+		ADD_STAT(txs.tx_tls_fallback);
+		ADD_STAT(txs.tx_tls_drops);
+	}
+	data += ARRAY_SIZE(txq_stat_names);
+
+	/* XDP Tx queues */
+	xdpqs = rtnl_dereference(fp->xdpqs);
+	totals = data + fp->num_xdpqs * ARRAY_SIZE(xdpq_stat_names);
+
+	for (i = 0; i < fp->num_xdpqs; i++) {
+		tot = totals;
+
+		FUN_QSTAT_READ(xdpqs[i], start, txs);
+
+		ADD_STAT(txs.tx_pkts);
+		ADD_STAT(txs.tx_bytes);
+		ADD_STAT(txs.tx_xdp_full);
+		ADD_STAT(txs.tx_map_err);
+	}
+	data += ARRAY_SIZE(xdpq_stat_names);
+
+	/* Rx queues */
+	totals = data + netdev->real_num_rx_queues * ARRAY_SIZE(rxq_stat_names);
+
+	for (i = 0; i < netdev->real_num_rx_queues; i++) {
+		tot = totals;
+
+		FUN_QSTAT_READ(rxqs[i], start, rxs);
+
+		ADD_STAT(rxs.rx_pkts);
+		ADD_STAT(rxs.rx_bytes);
+		ADD_STAT(rxs.rx_cso);
+		ADD_STAT(rxs.gro_pkts);
+		ADD_STAT(rxs.gro_merged);
+		ADD_STAT(rxs.xdp_tx);
+		ADD_STAT(rxs.xdp_redir);
+		ADD_STAT(rxs.xdp_drops);
+		ADD_STAT(rxs.rx_bufs);
+		ADD_STAT(rxs.rx_page_alloc);
+		ADD_STAT(rxs.rx_mem_drops + rxs.xdp_err);
+		ADD_STAT(rxs.rx_budget);
+		ADD_STAT(rxs.rx_map_err);
+	}
+	data += ARRAY_SIZE(rxq_stat_names);
+#undef ADD_STAT
+
+	*data++ = atomic64_read(&fp->tx_tls_add);
+	*data++ = atomic64_read(&fp->tx_tls_del);
+	*data++ = atomic64_read(&fp->tx_tls_resync);
+}
+
+#define RX_STAT(fp, s) be64_to_cpu((fp)->stats[PORT_MAC_RX_##s])
+#define TX_STAT(fp, s) \
+	be64_to_cpu((fp)->stats[PORT_MAC_RX_STATS_MAX + PORT_MAC_TX_##s])
+#define FEC_STAT(fp, s) \
+	be64_to_cpu((fp)->stats[PORT_MAC_RX_STATS_MAX + \
+				PORT_MAC_TX_STATS_MAX + PORT_MAC_FEC_##s])
+
+static void fun_get_pause_stats(struct net_device *netdev,
+				struct ethtool_pause_stats *stats)
+{
+	const struct funeth_priv *fp = netdev_priv(netdev);
+
+	if (!(fp->port_caps & FUN_PORT_CAP_STATS))
+		return;
+
+	stats->tx_pause_frames = TX_STAT(fp, aPAUSEMACCtrlFramesTransmitted);
+	stats->rx_pause_frames = RX_STAT(fp, aPAUSEMACCtrlFramesReceived);
+}
+
+static void fun_get_802_3_stats(struct net_device *netdev,
+				struct ethtool_eth_mac_stats *stats)
+{
+	const struct funeth_priv *fp = netdev_priv(netdev);
+
+	if (!(fp->port_caps & FUN_PORT_CAP_STATS))
+		return;
+
+	stats->FramesTransmittedOK = TX_STAT(fp, aFramesTransmittedOK);
+	stats->FramesReceivedOK = RX_STAT(fp, aFramesReceivedOK);
+	stats->FrameCheckSequenceErrors = RX_STAT(fp, aFrameCheckSequenceErrors);
+	stats->OctetsTransmittedOK = TX_STAT(fp, OctetsTransmittedOK);
+	stats->OctetsReceivedOK = RX_STAT(fp, OctetsReceivedOK);
+	stats->InRangeLengthErrors = RX_STAT(fp, aInRangeLengthErrors);
+	stats->FrameTooLongErrors = RX_STAT(fp, aFrameTooLongErrors);
+}
+
+static void fun_get_802_3_ctrl_stats(struct net_device *netdev,
+				     struct ethtool_eth_ctrl_stats *stats)
+{
+	const struct funeth_priv *fp = netdev_priv(netdev);
+
+	if (!(fp->port_caps & FUN_PORT_CAP_STATS))
+		return;
+
+	stats->MACControlFramesTransmitted = TX_STAT(fp, MACControlFramesTransmitted);
+	stats->MACControlFramesReceived = RX_STAT(fp, MACControlFramesReceived);
+}
+
+static void fun_get_rmon_stats(struct net_device *netdev,
+			       struct ethtool_rmon_stats *stats,
+			       const struct ethtool_rmon_hist_range **ranges)
+{
+	static const struct ethtool_rmon_hist_range rmon_ranges[] = {
+		{   64,    64 },
+		{   65,   127 },
+		{  128,   255 },
+		{  256,   511 },
+		{  512,  1023 },
+		{ 1024,  1518 },
+		{ 1519, 32767 },
+		{}
+	};
+
+	const struct funeth_priv *fp = netdev_priv(netdev);
+
+	if (!(fp->port_caps & FUN_PORT_CAP_STATS))
+		return;
+
+	stats->undersize_pkts = RX_STAT(fp, etherStatsUndersizePkts);
+	stats->oversize_pkts = RX_STAT(fp, etherStatsOversizePkts);
+	stats->fragments = RX_STAT(fp, etherStatsFragments);
+	stats->jabbers = RX_STAT(fp, etherStatsJabbers);
+
+	stats->hist[0] = RX_STAT(fp, etherStatsPkts64Octets);
+	stats->hist[1] = RX_STAT(fp, etherStatsPkts65to127Octets);
+	stats->hist[2] = RX_STAT(fp, etherStatsPkts128to255Octets);
+	stats->hist[3] = RX_STAT(fp, etherStatsPkts256to511Octets);
+	stats->hist[4] = RX_STAT(fp, etherStatsPkts512to1023Octets);
+	stats->hist[5] = RX_STAT(fp, etherStatsPkts1024to1518Octets);
+	stats->hist[6] = RX_STAT(fp, etherStatsPkts1519toMaxOctets);
+
+	stats->hist_tx[0] = TX_STAT(fp, etherStatsPkts64Octets);
+	stats->hist_tx[1] = TX_STAT(fp, etherStatsPkts65to127Octets);
+	stats->hist_tx[2] = TX_STAT(fp, etherStatsPkts128to255Octets);
+	stats->hist_tx[3] = TX_STAT(fp, etherStatsPkts256to511Octets);
+	stats->hist_tx[4] = TX_STAT(fp, etherStatsPkts512to1023Octets);
+	stats->hist_tx[5] = TX_STAT(fp, etherStatsPkts1024to1518Octets);
+	stats->hist_tx[6] = TX_STAT(fp, etherStatsPkts1519toMaxOctets);
+
+	*ranges = rmon_ranges;
+}
+
+static void fun_get_fec_stats(struct net_device *netdev,
+			      struct ethtool_fec_stats *stats,
+			      struct ethtool_fec_hist *hist)
+{
+	const struct funeth_priv *fp = netdev_priv(netdev);
+
+	if (!(fp->port_caps & FUN_PORT_CAP_STATS))
+		return;
+
+	stats->corrected_blocks.total = FEC_STAT(fp, Correctable);
+	stats->uncorrectable_blocks.total = FEC_STAT(fp, Uncorrectable);
+}
+
+#undef RX_STAT
+#undef TX_STAT
+#undef FEC_STAT
+
+static int fun_get_rxnfc(struct net_device *netdev, struct ethtool_rxnfc *cmd,
+			 u32 *rule_locs)
+{
+	switch (cmd->cmd) {
+	case ETHTOOL_GRXRINGS:
+		cmd->data = netdev->real_num_rx_queues;
+		return 0;
+	default:
+		break;
+	}
+	return -EOPNOTSUPP;
+}
+
+static int fun_set_rxnfc(struct net_device *netdev, struct ethtool_rxnfc *info)
+{
+	return 0;
+}
+
+static u32 fun_get_rxfh_indir_size(struct net_device *netdev)
+{
+	const struct funeth_priv *fp = netdev_priv(netdev);
+
+	return fp->indir_table_nentries;
+}
+
+static u32 fun_get_rxfh_key_size(struct net_device *netdev)
+{
+	const struct funeth_priv *fp = netdev_priv(netdev);
+
+	return sizeof(fp->rss_key);
+}
+
+static int fun_get_rxfh(struct net_device *netdev,
+			struct ethtool_rxfh_param *rxfh)
+{
+	const struct funeth_priv *fp = netdev_priv(netdev);
+
+	if (!fp->rss_cfg)
+		return -EOPNOTSUPP;
+
+	if (rxfh->indir)
+		memcpy(rxfh->indir, fp->indir_table,
+		       sizeof(u32) * fp->indir_table_nentries);
+
+	if (rxfh->key)
+		memcpy(rxfh->key, fp->rss_key, sizeof(fp->rss_key));
+
+	rxfh->hfunc = fp->hash_algo == FUN_ETH_RSS_ALG_TOEPLITZ ?
+			ETH_RSS_HASH_TOP : ETH_RSS_HASH_CRC32;
+
+	return 0;
+}
+
+static int fun_set_rxfh(struct net_device *netdev,
+			struct ethtool_rxfh_param *rxfh,
+			struct netlink_ext_ack *extack)
+{
+	struct funeth_priv *fp = netdev_priv(netdev);
+	const u32 *rss_indir = rxfh->indir ? rxfh->indir : fp->indir_table;
+	const u8 *rss_key = rxfh->key ? rxfh->key : fp->rss_key;
+	enum fun_eth_hash_alg algo;
+
+	if (!fp->rss_cfg)
+		return -EOPNOTSUPP;
+
+	if (rxfh->hfunc == ETH_RSS_HASH_NO_CHANGE)
+		algo = fp->hash_algo;
+	else if (rxfh->hfunc == ETH_RSS_HASH_CRC32)
+		algo = FUN_ETH_RSS_ALG_CRC32;
+	else if (rxfh->hfunc == ETH_RSS_HASH_TOP)
+		algo = FUN_ETH_RSS_ALG_TOEPLITZ;
+	else
+		return -EINVAL;
+
+	/* If the port is enabled try to reconfigure RSS and keep the new
+	 * settings if successful. If it is down we update the RSS settings
+	 * and apply them at the next UP time.
+	 */
+	if (netif_running(netdev)) {
+		int rc = fun_config_rss(netdev, algo, rss_key, rss_indir,
+					FUN_ADMIN_SUBOP_MODIFY);
+		if (rc)
+			return rc;
+	}
+
+	fp->hash_algo = algo;
+	if (rxfh->key)
+		memcpy(fp->rss_key, rxfh->key, sizeof(fp->rss_key));
+	if (rxfh->indir)
+		memcpy(fp->indir_table, rxfh->indir,
+		       sizeof(u32) * fp->indir_table_nentries);
+	return 0;
+}
+
+static int fun_get_ts_info(struct net_device *netdev,
+			   struct kernel_ethtool_ts_info *info)
+{
+	info->so_timestamping = SOF_TIMESTAMPING_RX_HARDWARE |
+				SOF_TIMESTAMPING_TX_SOFTWARE |
+				SOF_TIMESTAMPING_RAW_HARDWARE;
+	info->tx_types = BIT(HWTSTAMP_TX_OFF);
+	info->rx_filters = BIT(HWTSTAMP_FILTER_NONE) | BIT(HWTSTAMP_FILTER_ALL);
+	return 0;
+}
+
+static unsigned int to_ethtool_fec(unsigned int fun_fec)
+{
+	unsigned int fec = 0;
+
+	if (fun_fec == FUN_PORT_FEC_NA)
+		fec |= ETHTOOL_FEC_NONE;
+	if (fun_fec & FUN_PORT_FEC_OFF)
+		fec |= ETHTOOL_FEC_OFF;
+	if (fun_fec & FUN_PORT_FEC_RS)
+		fec |= ETHTOOL_FEC_RS;
+	if (fun_fec & FUN_PORT_FEC_FC)
+		fec |= ETHTOOL_FEC_BASER;
+	if (fun_fec & FUN_PORT_FEC_AUTO)
+		fec |= ETHTOOL_FEC_AUTO;
+	return fec;
+}
+
+static int fun_get_fecparam(struct net_device *netdev,
+			    struct ethtool_fecparam *fec)
+{
+	struct funeth_priv *fp = netdev_priv(netdev);
+	u64 fec_data;
+	int rc;
+
+	rc = fun_port_read_cmd(fp, FUN_ADMIN_PORT_KEY_FEC, &fec_data);
+	if (rc)
+		return rc;
+
+	fec->active_fec = to_ethtool_fec(fec_data & 0xff);
+	fec->fec = to_ethtool_fec(fec_data >> 8);
+	return 0;
+}
+
+static int fun_set_fecparam(struct net_device *netdev,
+			    struct ethtool_fecparam *fec)
+{
+	struct funeth_priv *fp = netdev_priv(netdev);
+	u64 fec_mode;
+
+	switch (fec->fec) {
+	case ETHTOOL_FEC_AUTO:
+		fec_mode = FUN_PORT_FEC_AUTO;
+		break;
+	case ETHTOOL_FEC_OFF:
+		if (!(fp->port_caps & FUN_PORT_CAP_FEC_NONE))
+			return -EINVAL;
+		fec_mode = FUN_PORT_FEC_OFF;
+		break;
+	case ETHTOOL_FEC_BASER:
+		if (!(fp->port_caps & FUN_PORT_CAP_FEC_FC))
+			return -EINVAL;
+		fec_mode = FUN_PORT_FEC_FC;
+		break;
+	case ETHTOOL_FEC_RS:
+		if (!(fp->port_caps & FUN_PORT_CAP_FEC_RS))
+			return -EINVAL;
+		fec_mode = FUN_PORT_FEC_RS;
+		break;
+	default:
+		return -EINVAL;
+	}
+
+	return fun_port_write_cmd(fp, FUN_ADMIN_PORT_KEY_FEC, fec_mode);
+}
+
+static int fun_get_port_module_page(struct net_device *netdev,
+				    const struct ethtool_module_eeprom *req,
+				    struct netlink_ext_ack *extack)
+{
+	union {
+		struct fun_admin_port_req req;
+		struct fun_admin_port_xcvr_read_rsp rsp;
+	} cmd;
+	struct funeth_priv *fp = netdev_priv(netdev);
+	int rc;
+
+	if (fp->port_caps & FUN_PORT_CAP_VPORT) {
+		NL_SET_ERR_MSG_MOD(extack,
+				   "Specified port is virtual, only physical ports have modules");
+		return -EOPNOTSUPP;
+	}
+
+	cmd.req.common = FUN_ADMIN_REQ_COMMON_INIT2(FUN_ADMIN_OP_PORT,
+						    sizeof(cmd.req));
+	cmd.req.u.xcvr_read =
+		FUN_ADMIN_PORT_XCVR_READ_REQ_INIT(0, netdev->dev_port,
+						  req->bank, req->page,
+						  req->offset, req->length,
+						  req->i2c_address);
+	rc = fun_submit_admin_sync_cmd(fp->fdev, &cmd.req.common, &cmd.rsp,
+				       sizeof(cmd.rsp), 0);
+	if (rc)
+		return rc;
+
+	memcpy(req->data, cmd.rsp.data, req->length);
+	return req->length;
+}
+
+static const struct ethtool_ops fun_ethtool_ops = {
+	.supported_coalesce_params = ETHTOOL_COALESCE_USECS |
+				     ETHTOOL_COALESCE_MAX_FRAMES,
+	.get_link_ksettings  = fun_get_link_ksettings,
+	.set_link_ksettings  = fun_set_link_ksettings,
+	.set_phys_id         = fun_set_phys_id,
+	.get_drvinfo         = fun_get_drvinfo,
+	.get_msglevel        = fun_get_msglevel,
+	.set_msglevel        = fun_set_msglevel,
+	.get_regs_len        = fun_get_regs_len,
+	.get_regs            = fun_get_regs,
+	.get_link	     = ethtool_op_get_link,
+	.get_coalesce        = fun_get_coalesce,
+	.set_coalesce        = fun_set_coalesce,
+	.get_ts_info         = fun_get_ts_info,
+	.get_ringparam       = fun_get_ringparam,
+	.set_ringparam       = fun_set_ringparam,
+	.get_sset_count      = fun_get_sset_count,
+	.get_strings         = fun_get_strings,
+	.get_ethtool_stats   = fun_get_ethtool_stats,
+	.get_rxnfc	     = fun_get_rxnfc,
+	.set_rxnfc           = fun_set_rxnfc,
+	.get_rxfh_indir_size = fun_get_rxfh_indir_size,
+	.get_rxfh_key_size   = fun_get_rxfh_key_size,
+	.get_rxfh            = fun_get_rxfh,
+	.set_rxfh            = fun_set_rxfh,
+	.get_channels        = fun_get_channels,
+	.set_channels        = fun_set_channels,
+	.get_fecparam	     = fun_get_fecparam,
+	.set_fecparam	     = fun_set_fecparam,
+	.get_pauseparam      = fun_get_pauseparam,
+	.set_pauseparam      = fun_set_pauseparam,
+	.nway_reset          = fun_restart_an,
+	.get_pause_stats     = fun_get_pause_stats,
+	.get_fec_stats       = fun_get_fec_stats,
+	.get_eth_mac_stats   = fun_get_802_3_stats,
+	.get_eth_ctrl_stats  = fun_get_802_3_ctrl_stats,
+	.get_rmon_stats      = fun_get_rmon_stats,
+	.get_module_eeprom_by_page = fun_get_port_module_page,
+};
+
+void fun_set_ethtool_ops(struct net_device *netdev)
+{
+	netdev->ethtool_ops = &fun_ethtool_ops;
+}
diff --git a/drivers/net/ethernet/fungible/funeth/funeth_ktls.c b/drivers/net/ethernet/fungible/funeth/funeth_ktls.c
new file mode 100644
index 000000000000..f871def70d70
--- /dev/null
+++ b/drivers/net/ethernet/fungible/funeth/funeth_ktls.c
@@ -0,0 +1,155 @@
+// SPDX-License-Identifier: (GPL-2.0-only OR BSD-3-Clause)
+
+#include "funeth.h"
+#include "funeth_ktls.h"
+
+static int fun_admin_ktls_create(struct funeth_priv *fp, unsigned int id)
+{
+	struct fun_admin_ktls_create_req req = {
+		.common = FUN_ADMIN_REQ_COMMON_INIT2(FUN_ADMIN_OP_KTLS,
+						     sizeof(req)),
+		.subop = FUN_ADMIN_SUBOP_CREATE,
+		.id = cpu_to_be32(id),
+	};
+
+	return fun_submit_admin_sync_cmd(fp->fdev, &req.common, NULL, 0, 0);
+}
+
+static int fun_ktls_add(struct net_device *netdev, struct sock *sk,
+			enum tls_offload_ctx_dir direction,
+			struct tls_crypto_info *crypto_info,
+			u32 start_offload_tcp_sn)
+{
+	struct funeth_priv *fp = netdev_priv(netdev);
+	struct fun_admin_ktls_modify_req req = {
+		.common = FUN_ADMIN_REQ_COMMON_INIT2(FUN_ADMIN_OP_KTLS,
+						     sizeof(req)),
+		.subop = FUN_ADMIN_SUBOP_MODIFY,
+		.id = cpu_to_be32(fp->ktls_id),
+		.tcp_seq = cpu_to_be32(start_offload_tcp_sn),
+	};
+	struct fun_admin_ktls_modify_rsp rsp;
+	struct fun_ktls_tx_ctx *tx_ctx;
+	int rc;
+
+	if (direction != TLS_OFFLOAD_CTX_DIR_TX)
+		return -EOPNOTSUPP;
+
+	if (crypto_info->version == TLS_1_2_VERSION)
+		req.version = FUN_KTLS_TLSV2;
+	else
+		return -EOPNOTSUPP;
+
+	switch (crypto_info->cipher_type) {
+	case TLS_CIPHER_AES_GCM_128: {
+		struct tls12_crypto_info_aes_gcm_128 *c = (void *)crypto_info;
+
+		req.cipher = FUN_KTLS_CIPHER_AES_GCM_128;
+		memcpy(req.key, c->key, sizeof(c->key));
+		memcpy(req.iv, c->iv, sizeof(c->iv));
+		memcpy(req.salt, c->salt, sizeof(c->salt));
+		memcpy(req.record_seq, c->rec_seq, sizeof(c->rec_seq));
+		break;
+	}
+	default:
+		return -EOPNOTSUPP;
+	}
+
+	rc = fun_submit_admin_sync_cmd(fp->fdev, &req.common, &rsp,
+				       sizeof(rsp), 0);
+	memzero_explicit(&req, sizeof(req));
+	if (rc)
+		return rc;
+
+	tx_ctx = tls_driver_ctx(sk, direction);
+	tx_ctx->tlsid = rsp.tlsid;
+	tx_ctx->next_seq = start_offload_tcp_sn;
+	atomic64_inc(&fp->tx_tls_add);
+	return 0;
+}
+
+static void fun_ktls_del(struct net_device *netdev,
+			 struct tls_context *tls_ctx,
+			 enum tls_offload_ctx_dir direction)
+{
+	struct funeth_priv *fp = netdev_priv(netdev);
+	struct fun_admin_ktls_modify_req req;
+	struct fun_ktls_tx_ctx *tx_ctx;
+
+	if (direction != TLS_OFFLOAD_CTX_DIR_TX)
+		return;
+
+	tx_ctx = __tls_driver_ctx(tls_ctx, direction);
+
+	req.common = FUN_ADMIN_REQ_COMMON_INIT2(FUN_ADMIN_OP_KTLS,
+			offsetof(struct fun_admin_ktls_modify_req, tcp_seq));
+	req.subop = FUN_ADMIN_SUBOP_MODIFY;
+	req.flags = cpu_to_be16(FUN_KTLS_MODIFY_REMOVE);
+	req.id = cpu_to_be32(fp->ktls_id);
+	req.tlsid = tx_ctx->tlsid;
+
+	fun_submit_admin_sync_cmd(fp->fdev, &req.common, NULL, 0, 0);
+	atomic64_inc(&fp->tx_tls_del);
+}
+
+static int fun_ktls_resync(struct net_device *netdev, struct sock *sk, u32 seq,
+			   u8 *rcd_sn, enum tls_offload_ctx_dir direction)
+{
+	struct funeth_priv *fp = netdev_priv(netdev);
+	struct fun_admin_ktls_modify_req req;
+	struct fun_ktls_tx_ctx *tx_ctx;
+	int rc;
+
+	if (direction != TLS_OFFLOAD_CTX_DIR_TX)
+		return -EOPNOTSUPP;
+
+	tx_ctx = tls_driver_ctx(sk, direction);
+
+	req.common = FUN_ADMIN_REQ_COMMON_INIT2(FUN_ADMIN_OP_KTLS,
+			offsetof(struct fun_admin_ktls_modify_req, key));
+	req.subop = FUN_ADMIN_SUBOP_MODIFY;
+	req.flags = 0;
+	req.id = cpu_to_be32(fp->ktls_id);
+	req.tlsid = tx_ctx->tlsid;
+	req.tcp_seq = cpu_to_be32(seq);
+	req.version = 0;
+	req.cipher = 0;
+	memcpy(req.record_seq, rcd_sn, sizeof(req.record_seq));
+
+	atomic64_inc(&fp->tx_tls_resync);
+	rc = fun_submit_admin_sync_cmd(fp->fdev, &req.common, NULL, 0, 0);
+	if (!rc)
+		tx_ctx->next_seq = seq;
+	return rc;
+}
+
+static const struct tlsdev_ops fun_ktls_ops = {
+	.tls_dev_add = fun_ktls_add,
+	.tls_dev_del = fun_ktls_del,
+	.tls_dev_resync = fun_ktls_resync,
+};
+
+int fun_ktls_init(struct net_device *netdev)
+{
+	struct funeth_priv *fp = netdev_priv(netdev);
+	int rc;
+
+	rc = fun_admin_ktls_create(fp, netdev->dev_port);
+	if (rc)
+		return rc;
+
+	fp->ktls_id = netdev->dev_port;
+	netdev->tlsdev_ops = &fun_ktls_ops;
+	netdev->hw_features |= NETIF_F_HW_TLS_TX;
+	netdev->features |= NETIF_F_HW_TLS_TX;
+	return 0;
+}
+
+void fun_ktls_cleanup(struct funeth_priv *fp)
+{
+	if (fp->ktls_id == FUN_HCI_ID_INVALID)
+		return;
+
+	fun_res_destroy(fp->fdev, FUN_ADMIN_OP_KTLS, 0, fp->ktls_id);
+	fp->ktls_id = FUN_HCI_ID_INVALID;
+}
diff --git a/drivers/net/ethernet/fungible/funeth/funeth_ktls.h b/drivers/net/ethernet/fungible/funeth/funeth_ktls.h
new file mode 100644
index 000000000000..9d6f2141a959
--- /dev/null
+++ b/drivers/net/ethernet/fungible/funeth/funeth_ktls.h
@@ -0,0 +1,30 @@
+/* SPDX-License-Identifier: (GPL-2.0-only OR BSD-3-Clause) */
+
+#ifndef _FUN_KTLS_H
+#define _FUN_KTLS_H
+
+#include <net/tls.h>
+
+struct funeth_priv;
+
+struct fun_ktls_tx_ctx {
+	__be64 tlsid;
+	u32 next_seq;
+};
+
+#if IS_ENABLED(CONFIG_TLS_DEVICE)
+int fun_ktls_init(struct net_device *netdev);
+void fun_ktls_cleanup(struct funeth_priv *fp);
+
+#else
+
+static inline void fun_ktls_init(struct net_device *netdev)
+{
+}
+
+static inline void fun_ktls_cleanup(struct funeth_priv *fp)
+{
+}
+#endif
+
+#endif /* _FUN_KTLS_H */
diff --git a/drivers/net/ethernet/fungible/funeth/funeth_main.c b/drivers/net/ethernet/fungible/funeth/funeth_main.c
new file mode 100644
index 000000000000..ac86179a0a81
--- /dev/null
+++ b/drivers/net/ethernet/fungible/funeth/funeth_main.c
@@ -0,0 +1,2081 @@
+// SPDX-License-Identifier: (GPL-2.0-only OR BSD-3-Clause)
+
+#include <linux/bpf.h>
+#include <linux/crash_dump.h>
+#include <linux/etherdevice.h>
+#include <linux/ethtool.h>
+#include <linux/filter.h>
+#include <linux/idr.h>
+#include <linux/if_vlan.h>
+#include <linux/module.h>
+#include <linux/netdevice.h>
+#include <linux/pci.h>
+#include <linux/rtnetlink.h>
+#include <linux/inetdevice.h>
+
+#include "funeth.h"
+#include "funeth_devlink.h"
+#include "funeth_ktls.h"
+#include "fun_port.h"
+#include "fun_queue.h"
+#include "funeth_txrx.h"
+
+#define ADMIN_SQ_DEPTH 32
+#define ADMIN_CQ_DEPTH 64
+#define ADMIN_RQ_DEPTH 16
+
+/* Default number of Tx/Rx queues. */
+#define FUN_DFLT_QUEUES 16U
+
+enum {
+	FUN_SERV_RES_CHANGE = FUN_SERV_FIRST_AVAIL,
+	FUN_SERV_DEL_PORTS,
+};
+
+static const struct pci_device_id funeth_id_table[] = {
+	{ PCI_VDEVICE(FUNGIBLE, 0x0101) },
+	{ PCI_VDEVICE(FUNGIBLE, 0x0181) },
+	{ 0, }
+};
+
+/* Issue a port write admin command with @n key/value pairs. */
+static int fun_port_write_cmds(struct funeth_priv *fp, unsigned int n,
+			       const int *keys, const u64 *data)
+{
+	unsigned int cmd_size, i;
+	union {
+		struct fun_admin_port_req req;
+		struct fun_admin_port_rsp rsp;
+		u8 v[ADMIN_SQE_SIZE];
+	} cmd;
+
+	cmd_size = offsetof(struct fun_admin_port_req, u.write.write48) +
+		n * sizeof(struct fun_admin_write48_req);
+	if (cmd_size > sizeof(cmd) || cmd_size > ADMIN_RSP_MAX_LEN)
+		return -EINVAL;
+
+	cmd.req.common = FUN_ADMIN_REQ_COMMON_INIT2(FUN_ADMIN_OP_PORT,
+						    cmd_size);
+	cmd.req.u.write =
+		FUN_ADMIN_PORT_WRITE_REQ_INIT(FUN_ADMIN_SUBOP_WRITE, 0,
+					      fp->netdev->dev_port);
+	for (i = 0; i < n; i++)
+		cmd.req.u.write.write48[i] =
+			FUN_ADMIN_WRITE48_REQ_INIT(keys[i], data[i]);
+
+	return fun_submit_admin_sync_cmd(fp->fdev, &cmd.req.common,
+					 &cmd.rsp, cmd_size, 0);
+}
+
+int fun_port_write_cmd(struct funeth_priv *fp, int key, u64 data)
+{
+	return fun_port_write_cmds(fp, 1, &key, &data);
+}
+
+/* Issue a port read admin command with @n key/value pairs. */
+static int fun_port_read_cmds(struct funeth_priv *fp, unsigned int n,
+			      const int *keys, u64 *data)
+{
+	const struct fun_admin_read48_rsp *r48rsp;
+	unsigned int cmd_size, i;
+	int rc;
+	union {
+		struct fun_admin_port_req req;
+		struct fun_admin_port_rsp rsp;
+		u8 v[ADMIN_SQE_SIZE];
+	} cmd;
+
+	cmd_size = offsetof(struct fun_admin_port_req, u.read.read48) +
+		n * sizeof(struct fun_admin_read48_req);
+	if (cmd_size > sizeof(cmd) || cmd_size > ADMIN_RSP_MAX_LEN)
+		return -EINVAL;
+
+	cmd.req.common = FUN_ADMIN_REQ_COMMON_INIT2(FUN_ADMIN_OP_PORT,
+						    cmd_size);
+	cmd.req.u.read =
+		FUN_ADMIN_PORT_READ_REQ_INIT(FUN_ADMIN_SUBOP_READ, 0,
+					     fp->netdev->dev_port);
+	for (i = 0; i < n; i++)
+		cmd.req.u.read.read48[i] = FUN_ADMIN_READ48_REQ_INIT(keys[i]);
+
+	rc = fun_submit_admin_sync_cmd(fp->fdev, &cmd.req.common,
+				       &cmd.rsp, cmd_size, 0);
+	if (rc)
+		return rc;
+
+	for (r48rsp = cmd.rsp.u.read.read48, i = 0; i < n; i++, r48rsp++) {
+		data[i] = FUN_ADMIN_READ48_RSP_DATA_G(r48rsp->key_to_data);
+		dev_dbg(fp->fdev->dev,
+			"port_read_rsp lport=%u (key_to_data=0x%llx) key=%d data:%lld retval:%lld",
+			fp->lport, r48rsp->key_to_data, keys[i], data[i],
+			FUN_ADMIN_READ48_RSP_RET_G(r48rsp->key_to_data));
+	}
+	return 0;
+}
+
+int fun_port_read_cmd(struct funeth_priv *fp, int key, u64 *data)
+{
+	return fun_port_read_cmds(fp, 1, &key, data);
+}
+
+static void fun_report_link(struct net_device *netdev)
+{
+	if (netif_carrier_ok(netdev)) {
+		const struct funeth_priv *fp = netdev_priv(netdev);
+		const char *fec = "", *pause = "";
+		int speed = fp->link_speed;
+		char unit = 'M';
+
+		if (fp->link_speed >= SPEED_1000) {
+			speed /= 1000;
+			unit = 'G';
+		}
+
+		if (fp->active_fec & FUN_PORT_FEC_RS)
+			fec = ", RS-FEC";
+		else if (fp->active_fec & FUN_PORT_FEC_FC)
+			fec = ", BASER-FEC";
+
+		if ((fp->active_fc & FUN_PORT_CAP_PAUSE_MASK) == FUN_PORT_CAP_PAUSE_MASK)
+			pause = ", Tx/Rx PAUSE";
+		else if (fp->active_fc & FUN_PORT_CAP_RX_PAUSE)
+			pause = ", Rx PAUSE";
+		else if (fp->active_fc & FUN_PORT_CAP_TX_PAUSE)
+			pause = ", Tx PAUSE";
+
+		netdev_info(netdev, "Link up at %d %cb/s full-duplex%s%s\n",
+			    speed, unit, pause, fec);
+	} else {
+		netdev_info(netdev, "Link down\n");
+	}
+}
+
+static int fun_adi_write(struct fun_dev *fdev, enum fun_admin_adi_attr attr,
+			 unsigned int adi_id, const struct fun_adi_param *param)
+{
+	struct fun_admin_adi_req req = {
+		.common = FUN_ADMIN_REQ_COMMON_INIT2(FUN_ADMIN_OP_ADI,
+						     sizeof(req)),
+		.u.write.subop = FUN_ADMIN_SUBOP_WRITE,
+		.u.write.attribute = attr,
+		.u.write.id = cpu_to_be32(adi_id),
+		.u.write.param = *param
+	};
+
+	return fun_submit_admin_sync_cmd(fdev, &req.common, NULL, 0, 0);
+}
+
+/* Configure RSS for the given port. @op determines whether a new RSS context
+ * is to be created or whether an existing one should be reconfigured. The
+ * remaining parameters specify the hashing algorithm, key, and indirection
+ * table.
+ *
+ * This initiates packet delivery to the Rx queues set in the indirection
+ * table.
+ */
+int fun_config_rss(struct net_device *dev, int algo, const u8 *key,
+		   const u32 *qtable, u8 op)
+{
+	struct funeth_priv *fp = netdev_priv(dev);
+	unsigned int table_len = fp->indir_table_nentries;
+	unsigned int len = FUN_ETH_RSS_MAX_KEY_SIZE + sizeof(u32) * table_len;
+	struct funeth_rxq **rxqs = rtnl_dereference(fp->rxqs);
+	union {
+		struct {
+			struct fun_admin_rss_req req;
+			struct fun_dataop_gl gl;
+		};
+		struct fun_admin_generic_create_rsp rsp;
+	} cmd;
+	__be32 *indir_tab;
+	u16 flags;
+	int rc;
+
+	if (op != FUN_ADMIN_SUBOP_CREATE && fp->rss_hw_id == FUN_HCI_ID_INVALID)
+		return -EINVAL;
+
+	flags = op == FUN_ADMIN_SUBOP_CREATE ?
+			FUN_ADMIN_RES_CREATE_FLAG_ALLOCATOR : 0;
+	cmd.req.common = FUN_ADMIN_REQ_COMMON_INIT2(FUN_ADMIN_OP_RSS,
+						    sizeof(cmd));
+	cmd.req.u.create =
+		FUN_ADMIN_RSS_CREATE_REQ_INIT(op, flags, fp->rss_hw_id,
+					      dev->dev_port, algo,
+					      FUN_ETH_RSS_MAX_KEY_SIZE,
+					      table_len, 0,
+					      FUN_ETH_RSS_MAX_KEY_SIZE);
+	cmd.req.u.create.dataop = FUN_DATAOP_HDR_INIT(1, 0, 1, 0, len);
+	fun_dataop_gl_init(&cmd.gl, 0, 0, len, fp->rss_dma_addr);
+
+	/* write the key and indirection table into the RSS DMA area */
+	memcpy(fp->rss_cfg, key, FUN_ETH_RSS_MAX_KEY_SIZE);
+	indir_tab = fp->rss_cfg + FUN_ETH_RSS_MAX_KEY_SIZE;
+	for (rc = 0; rc < table_len; rc++)
+		*indir_tab++ = cpu_to_be32(rxqs[*qtable++]->hw_cqid);
+
+	rc = fun_submit_admin_sync_cmd(fp->fdev, &cmd.req.common,
+				       &cmd.rsp, sizeof(cmd.rsp), 0);
+	if (!rc && op == FUN_ADMIN_SUBOP_CREATE)
+		fp->rss_hw_id = be32_to_cpu(cmd.rsp.id);
+	return rc;
+}
+
+/* Destroy the HW RSS conntext associated with the given port. This also stops
+ * all packet delivery to our Rx queues.
+ */
+static void fun_destroy_rss(struct funeth_priv *fp)
+{
+	if (fp->rss_hw_id != FUN_HCI_ID_INVALID) {
+		fun_res_destroy(fp->fdev, FUN_ADMIN_OP_RSS, 0, fp->rss_hw_id);
+		fp->rss_hw_id = FUN_HCI_ID_INVALID;
+	}
+}
+
+static void fun_irq_aff_notify(struct irq_affinity_notify *notify,
+			       const cpumask_t *mask)
+{
+	struct fun_irq *p = container_of(notify, struct fun_irq, aff_notify);
+
+	cpumask_copy(&p->affinity_mask, mask);
+}
+
+static void fun_irq_aff_release(struct kref __always_unused *ref)
+{
+}
+
+/* Allocate an IRQ structure, assign an MSI-X index and initial affinity to it,
+ * and add it to the IRQ XArray.
+ */
+static struct fun_irq *fun_alloc_qirq(struct funeth_priv *fp, unsigned int idx,
+				      int node, unsigned int xa_idx_offset)
+{
+	struct fun_irq *irq;
+	int cpu, res;
+
+	cpu = cpumask_local_spread(idx, node);
+	node = cpu_to_mem(cpu);
+
+	irq = kzalloc_node(sizeof(*irq), GFP_KERNEL, node);
+	if (!irq)
+		return ERR_PTR(-ENOMEM);
+
+	res = fun_reserve_irqs(fp->fdev, 1, &irq->irq_idx);
+	if (res != 1)
+		goto free_irq;
+
+	res = xa_insert(&fp->irqs, idx + xa_idx_offset, irq, GFP_KERNEL);
+	if (res)
+		goto release_irq;
+
+	irq->irq = pci_irq_vector(fp->pdev, irq->irq_idx);
+	cpumask_set_cpu(cpu, &irq->affinity_mask);
+	irq->aff_notify.notify = fun_irq_aff_notify;
+	irq->aff_notify.release = fun_irq_aff_release;
+	irq->state = FUN_IRQ_INIT;
+	return irq;
+
+release_irq:
+	fun_release_irqs(fp->fdev, 1, &irq->irq_idx);
+free_irq:
+	kfree(irq);
+	return ERR_PTR(res);
+}
+
+static void fun_free_qirq(struct funeth_priv *fp, struct fun_irq *irq)
+{
+	netif_napi_del(&irq->napi);
+	fun_release_irqs(fp->fdev, 1, &irq->irq_idx);
+	kfree(irq);
+}
+
+/* Release the IRQs reserved for Tx/Rx queues that aren't being used. */
+static void fun_prune_queue_irqs(struct net_device *dev)
+{
+	struct funeth_priv *fp = netdev_priv(dev);
+	unsigned int nreleased = 0;
+	struct fun_irq *irq;
+	unsigned long idx;
+
+	xa_for_each(&fp->irqs, idx, irq) {
+		if (irq->txq || irq->rxq)  /* skip those in use */
+			continue;
+
+		xa_erase(&fp->irqs, idx);
+		fun_free_qirq(fp, irq);
+		nreleased++;
+		if (idx < fp->rx_irq_ofst)
+			fp->num_tx_irqs--;
+		else
+			fp->num_rx_irqs--;
+	}
+	netif_info(fp, intr, dev, "Released %u queue IRQs\n", nreleased);
+}
+
+/* Reserve IRQs, one per queue, to acommodate the requested queue numbers @ntx
+ * and @nrx. IRQs are added incrementally to those we already have.
+ * We hold on to allocated IRQs until garbage collection of unused IRQs is
+ * separately requested.
+ */
+static int fun_alloc_queue_irqs(struct net_device *dev, unsigned int ntx,
+				unsigned int nrx)
+{
+	struct funeth_priv *fp = netdev_priv(dev);
+	int node = dev_to_node(&fp->pdev->dev);
+	struct fun_irq *irq;
+	unsigned int i;
+
+	for (i = fp->num_tx_irqs; i < ntx; i++) {
+		irq = fun_alloc_qirq(fp, i, node, 0);
+		if (IS_ERR(irq))
+			return PTR_ERR(irq);
+
+		fp->num_tx_irqs++;
+		netif_napi_add_tx(dev, &irq->napi, fun_txq_napi_poll);
+	}
+
+	for (i = fp->num_rx_irqs; i < nrx; i++) {
+		irq = fun_alloc_qirq(fp, i, node, fp->rx_irq_ofst);
+		if (IS_ERR(irq))
+			return PTR_ERR(irq);
+
+		fp->num_rx_irqs++;
+		netif_napi_add(dev, &irq->napi, fun_rxq_napi_poll);
+	}
+
+	netif_info(fp, intr, dev, "Reserved %u/%u IRQs for Tx/Rx queues\n",
+		   ntx, nrx);
+	return 0;
+}
+
+static void free_txqs(struct funeth_txq **txqs, unsigned int nqs,
+		      unsigned int start, int state)
+{
+	unsigned int i;
+
+	for (i = start; i < nqs && txqs[i]; i++)
+		txqs[i] = funeth_txq_free(txqs[i], state);
+}
+
+static int alloc_txqs(struct net_device *dev, struct funeth_txq **txqs,
+		      unsigned int nqs, unsigned int depth, unsigned int start,
+		      int state)
+{
+	struct funeth_priv *fp = netdev_priv(dev);
+	unsigned int i;
+	int err;
+
+	for (i = start; i < nqs; i++) {
+		err = funeth_txq_create(dev, i, depth, xa_load(&fp->irqs, i),
+					state, &txqs[i]);
+		if (err) {
+			free_txqs(txqs, nqs, start, FUN_QSTATE_DESTROYED);
+			return err;
+		}
+	}
+	return 0;
+}
+
+static void free_rxqs(struct funeth_rxq **rxqs, unsigned int nqs,
+		      unsigned int start, int state)
+{
+	unsigned int i;
+
+	for (i = start; i < nqs && rxqs[i]; i++)
+		rxqs[i] = funeth_rxq_free(rxqs[i], state);
+}
+
+static int alloc_rxqs(struct net_device *dev, struct funeth_rxq **rxqs,
+		      unsigned int nqs, unsigned int ncqe, unsigned int nrqe,
+		      unsigned int start, int state)
+{
+	struct funeth_priv *fp = netdev_priv(dev);
+	unsigned int i;
+	int err;
+
+	for (i = start; i < nqs; i++) {
+		err = funeth_rxq_create(dev, i, ncqe, nrqe,
+					xa_load(&fp->irqs, i + fp->rx_irq_ofst),
+					state, &rxqs[i]);
+		if (err) {
+			free_rxqs(rxqs, nqs, start, FUN_QSTATE_DESTROYED);
+			return err;
+		}
+	}
+	return 0;
+}
+
+static void free_xdpqs(struct funeth_txq **xdpqs, unsigned int nqs,
+		       unsigned int start, int state)
+{
+	unsigned int i;
+
+	for (i = start; i < nqs && xdpqs[i]; i++)
+		xdpqs[i] = funeth_txq_free(xdpqs[i], state);
+
+	if (state == FUN_QSTATE_DESTROYED)
+		kfree(xdpqs);
+}
+
+static struct funeth_txq **alloc_xdpqs(struct net_device *dev, unsigned int nqs,
+				       unsigned int depth, unsigned int start,
+				       int state)
+{
+	struct funeth_txq **xdpqs;
+	unsigned int i;
+	int err;
+
+	xdpqs = kcalloc(nqs, sizeof(*xdpqs), GFP_KERNEL);
+	if (!xdpqs)
+		return ERR_PTR(-ENOMEM);
+
+	for (i = start; i < nqs; i++) {
+		err = funeth_txq_create(dev, i, depth, NULL, state, &xdpqs[i]);
+		if (err) {
+			free_xdpqs(xdpqs, nqs, start, FUN_QSTATE_DESTROYED);
+			return ERR_PTR(err);
+		}
+	}
+	return xdpqs;
+}
+
+static void fun_free_rings(struct net_device *netdev, struct fun_qset *qset)
+{
+	struct funeth_priv *fp = netdev_priv(netdev);
+	struct funeth_txq **xdpqs = qset->xdpqs;
+	struct funeth_rxq **rxqs = qset->rxqs;
+
+	/* qset may not specify any queues to operate on. In that case the
+	 * currently installed queues are implied.
+	 */
+	if (!rxqs) {
+		rxqs = rtnl_dereference(fp->rxqs);
+		xdpqs = rtnl_dereference(fp->xdpqs);
+		qset->txqs = fp->txqs;
+		qset->nrxqs = netdev->real_num_rx_queues;
+		qset->ntxqs = netdev->real_num_tx_queues;
+		qset->nxdpqs = fp->num_xdpqs;
+	}
+	if (!rxqs)
+		return;
+
+	if (rxqs == rtnl_dereference(fp->rxqs)) {
+		rcu_assign_pointer(fp->rxqs, NULL);
+		rcu_assign_pointer(fp->xdpqs, NULL);
+		synchronize_net();
+		fp->txqs = NULL;
+	}
+
+	free_rxqs(rxqs, qset->nrxqs, qset->rxq_start, qset->state);
+	free_txqs(qset->txqs, qset->ntxqs, qset->txq_start, qset->state);
+	free_xdpqs(xdpqs, qset->nxdpqs, qset->xdpq_start, qset->state);
+	if (qset->state == FUN_QSTATE_DESTROYED)
+		kfree(rxqs);
+
+	/* Tell the caller which queues were operated on. */
+	qset->rxqs = rxqs;
+	qset->xdpqs = xdpqs;
+}
+
+static int fun_alloc_rings(struct net_device *netdev, struct fun_qset *qset)
+{
+	struct funeth_txq **xdpqs = NULL, **txqs;
+	struct funeth_rxq **rxqs;
+	int err;
+
+	err = fun_alloc_queue_irqs(netdev, qset->ntxqs, qset->nrxqs);
+	if (err)
+		return err;
+
+	rxqs = kcalloc(qset->ntxqs + qset->nrxqs, sizeof(*rxqs), GFP_KERNEL);
+	if (!rxqs)
+		return -ENOMEM;
+
+	if (qset->nxdpqs) {
+		xdpqs = alloc_xdpqs(netdev, qset->nxdpqs, qset->sq_depth,
+				    qset->xdpq_start, qset->state);
+		if (IS_ERR(xdpqs)) {
+			err = PTR_ERR(xdpqs);
+			goto free_qvec;
+		}
+	}
+
+	txqs = (struct funeth_txq **)&rxqs[qset->nrxqs];
+	err = alloc_txqs(netdev, txqs, qset->ntxqs, qset->sq_depth,
+			 qset->txq_start, qset->state);
+	if (err)
+		goto free_xdpqs;
+
+	err = alloc_rxqs(netdev, rxqs, qset->nrxqs, qset->cq_depth,
+			 qset->rq_depth, qset->rxq_start, qset->state);
+	if (err)
+		goto free_txqs;
+
+	qset->rxqs = rxqs;
+	qset->txqs = txqs;
+	qset->xdpqs = xdpqs;
+	return 0;
+
+free_txqs:
+	free_txqs(txqs, qset->ntxqs, qset->txq_start, FUN_QSTATE_DESTROYED);
+free_xdpqs:
+	free_xdpqs(xdpqs, qset->nxdpqs, qset->xdpq_start, FUN_QSTATE_DESTROYED);
+free_qvec:
+	kfree(rxqs);
+	return err;
+}
+
+/* Take queues to the next level. Presently this means creating them on the
+ * device.
+ */
+static int fun_advance_ring_state(struct net_device *dev, struct fun_qset *qset)
+{
+	struct funeth_priv *fp = netdev_priv(dev);
+	int i, err;
+
+	for (i = 0; i < qset->nrxqs; i++) {
+		err = fun_rxq_create_dev(qset->rxqs[i],
+					 xa_load(&fp->irqs,
+						 i + fp->rx_irq_ofst));
+		if (err)
+			goto out;
+	}
+
+	for (i = 0; i < qset->ntxqs; i++) {
+		err = fun_txq_create_dev(qset->txqs[i], xa_load(&fp->irqs, i));
+		if (err)
+			goto out;
+	}
+
+	for (i = 0; i < qset->nxdpqs; i++) {
+		err = fun_txq_create_dev(qset->xdpqs[i], NULL);
+		if (err)
+			goto out;
+	}
+
+	return 0;
+
+out:
+	fun_free_rings(dev, qset);
+	return err;
+}
+
+static int fun_port_create(struct net_device *netdev)
+{
+	struct funeth_priv *fp = netdev_priv(netdev);
+	union {
+		struct fun_admin_port_req req;
+		struct fun_admin_port_rsp rsp;
+	} cmd;
+	int rc;
+
+	if (fp->lport != INVALID_LPORT)
+		return 0;
+
+	cmd.req.common = FUN_ADMIN_REQ_COMMON_INIT2(FUN_ADMIN_OP_PORT,
+						    sizeof(cmd.req));
+	cmd.req.u.create =
+		FUN_ADMIN_PORT_CREATE_REQ_INIT(FUN_ADMIN_SUBOP_CREATE, 0,
+					       netdev->dev_port);
+
+	rc = fun_submit_admin_sync_cmd(fp->fdev, &cmd.req.common, &cmd.rsp,
+				       sizeof(cmd.rsp), 0);
+
+	if (!rc)
+		fp->lport = be16_to_cpu(cmd.rsp.u.create.lport);
+	return rc;
+}
+
+static int fun_port_destroy(struct net_device *netdev)
+{
+	struct funeth_priv *fp = netdev_priv(netdev);
+
+	if (fp->lport == INVALID_LPORT)
+		return 0;
+
+	fp->lport = INVALID_LPORT;
+	return fun_res_destroy(fp->fdev, FUN_ADMIN_OP_PORT, 0,
+			       netdev->dev_port);
+}
+
+static int fun_eth_create(struct funeth_priv *fp)
+{
+	union {
+		struct fun_admin_eth_req req;
+		struct fun_admin_generic_create_rsp rsp;
+	} cmd;
+	int rc;
+
+	cmd.req.common = FUN_ADMIN_REQ_COMMON_INIT2(FUN_ADMIN_OP_ETH,
+						    sizeof(cmd.req));
+	cmd.req.u.create = FUN_ADMIN_ETH_CREATE_REQ_INIT(
+				FUN_ADMIN_SUBOP_CREATE,
+				FUN_ADMIN_RES_CREATE_FLAG_ALLOCATOR,
+				0, fp->netdev->dev_port);
+
+	rc = fun_submit_admin_sync_cmd(fp->fdev, &cmd.req.common, &cmd.rsp,
+				       sizeof(cmd.rsp), 0);
+	return rc ? rc : be32_to_cpu(cmd.rsp.id);
+}
+
+static int fun_vi_create(struct funeth_priv *fp)
+{
+	struct fun_admin_vi_req req = {
+		.common = FUN_ADMIN_REQ_COMMON_INIT2(FUN_ADMIN_OP_VI,
+						     sizeof(req)),
+		.u.create = FUN_ADMIN_VI_CREATE_REQ_INIT(FUN_ADMIN_SUBOP_CREATE,
+							 0,
+							 fp->netdev->dev_port,
+							 fp->netdev->dev_port)
+	};
+
+	return fun_submit_admin_sync_cmd(fp->fdev, &req.common, NULL, 0, 0);
+}
+
+/* Helper to create an ETH flow and bind an SQ to it.
+ * Returns the ETH id (>= 0) on success or a negative error.
+ */
+int fun_create_and_bind_tx(struct funeth_priv *fp, u32 sqid)
+{
+	int rc, ethid;
+
+	ethid = fun_eth_create(fp);
+	if (ethid >= 0) {
+		rc = fun_bind(fp->fdev, FUN_ADMIN_BIND_TYPE_EPSQ, sqid,
+			      FUN_ADMIN_BIND_TYPE_ETH, ethid);
+		if (rc) {
+			fun_res_destroy(fp->fdev, FUN_ADMIN_OP_ETH, 0, ethid);
+			ethid = rc;
+		}
+	}
+	return ethid;
+}
+
+static irqreturn_t fun_queue_irq_handler(int irq, void *data)
+{
+	struct fun_irq *p = data;
+
+	if (p->rxq) {
+		prefetch(p->rxq->next_cqe_info);
+		p->rxq->irq_cnt++;
+	}
+	napi_schedule_irqoff(&p->napi);
+	return IRQ_HANDLED;
+}
+
+static int fun_enable_irqs(struct net_device *dev)
+{
+	struct funeth_priv *fp = netdev_priv(dev);
+	unsigned long idx, last;
+	unsigned int qidx;
+	struct fun_irq *p;
+	const char *qtype;
+	int err;
+
+	xa_for_each(&fp->irqs, idx, p) {
+		if (p->txq) {
+			qtype = "tx";
+			qidx = p->txq->qidx;
+		} else if (p->rxq) {
+			qtype = "rx";
+			qidx = p->rxq->qidx;
+		} else {
+			continue;
+		}
+
+		if (p->state != FUN_IRQ_INIT)
+			continue;
+
+		snprintf(p->name, sizeof(p->name) - 1, "%s-%s-%u", dev->name,
+			 qtype, qidx);
+		err = request_irq(p->irq, fun_queue_irq_handler, 0, p->name, p);
+		if (err) {
+			netdev_err(dev, "Failed to allocate IRQ %u, err %d\n",
+				   p->irq, err);
+			goto unroll;
+		}
+		p->state = FUN_IRQ_REQUESTED;
+	}
+
+	xa_for_each(&fp->irqs, idx, p) {
+		if (p->state != FUN_IRQ_REQUESTED)
+			continue;
+		irq_set_affinity_notifier(p->irq, &p->aff_notify);
+		irq_set_affinity_and_hint(p->irq, &p->affinity_mask);
+		napi_enable(&p->napi);
+		p->state = FUN_IRQ_ENABLED;
+	}
+
+	return 0;
+
+unroll:
+	last = idx - 1;
+	xa_for_each_range(&fp->irqs, idx, p, 0, last)
+		if (p->state == FUN_IRQ_REQUESTED) {
+			free_irq(p->irq, p);
+			p->state = FUN_IRQ_INIT;
+		}
+
+	return err;
+}
+
+static void fun_disable_one_irq(struct fun_irq *irq)
+{
+	napi_disable(&irq->napi);
+	irq_set_affinity_notifier(irq->irq, NULL);
+	irq_update_affinity_hint(irq->irq, NULL);
+	free_irq(irq->irq, irq);
+	irq->state = FUN_IRQ_INIT;
+}
+
+static void fun_disable_irqs(struct net_device *dev)
+{
+	struct funeth_priv *fp = netdev_priv(dev);
+	struct fun_irq *p;
+	unsigned long idx;
+
+	xa_for_each(&fp->irqs, idx, p)
+		if (p->state == FUN_IRQ_ENABLED)
+			fun_disable_one_irq(p);
+}
+
+static void fun_down(struct net_device *dev, struct fun_qset *qset)
+{
+	struct funeth_priv *fp = netdev_priv(dev);
+
+	/* If we don't have queues the data path is already down.
+	 * Note netif_running(dev) may be true.
+	 */
+	if (!rcu_access_pointer(fp->rxqs))
+		return;
+
+	/* It is also down if the queues aren't on the device. */
+	if (fp->txqs[0]->init_state >= FUN_QSTATE_INIT_FULL) {
+		netif_info(fp, ifdown, dev,
+			   "Tearing down data path on device\n");
+		fun_port_write_cmd(fp, FUN_ADMIN_PORT_KEY_DISABLE, 0);
+
+		netif_carrier_off(dev);
+		netif_tx_disable(dev);
+
+		fun_destroy_rss(fp);
+		fun_res_destroy(fp->fdev, FUN_ADMIN_OP_VI, 0, dev->dev_port);
+		fun_disable_irqs(dev);
+	}
+
+	fun_free_rings(dev, qset);
+}
+
+static int fun_up(struct net_device *dev, struct fun_qset *qset)
+{
+	static const int port_keys[] = {
+		FUN_ADMIN_PORT_KEY_STATS_DMA_LOW,
+		FUN_ADMIN_PORT_KEY_STATS_DMA_HIGH,
+		FUN_ADMIN_PORT_KEY_ENABLE
+	};
+
+	struct funeth_priv *fp = netdev_priv(dev);
+	u64 vals[] = {
+		lower_32_bits(fp->stats_dma_addr),
+		upper_32_bits(fp->stats_dma_addr),
+		FUN_PORT_FLAG_ENABLE_NOTIFY
+	};
+	int err;
+
+	netif_info(fp, ifup, dev, "Setting up data path on device\n");
+
+	if (qset->rxqs[0]->init_state < FUN_QSTATE_INIT_FULL) {
+		err = fun_advance_ring_state(dev, qset);
+		if (err)
+			return err;
+	}
+
+	err = fun_vi_create(fp);
+	if (err)
+		goto free_queues;
+
+	fp->txqs = qset->txqs;
+	rcu_assign_pointer(fp->rxqs, qset->rxqs);
+	rcu_assign_pointer(fp->xdpqs, qset->xdpqs);
+
+	err = fun_enable_irqs(dev);
+	if (err)
+		goto destroy_vi;
+
+	if (fp->rss_cfg) {
+		err = fun_config_rss(dev, fp->hash_algo, fp->rss_key,
+				     fp->indir_table, FUN_ADMIN_SUBOP_CREATE);
+	} else {
+		/* The non-RSS case has only 1 queue. */
+		err = fun_bind(fp->fdev, FUN_ADMIN_BIND_TYPE_VI, dev->dev_port,
+			       FUN_ADMIN_BIND_TYPE_EPCQ,
+			       qset->rxqs[0]->hw_cqid);
+	}
+	if (err)
+		goto disable_irqs;
+
+	err = fun_port_write_cmds(fp, 3, port_keys, vals);
+	if (err)
+		goto free_rss;
+
+	netif_tx_start_all_queues(dev);
+	return 0;
+
+free_rss:
+	fun_destroy_rss(fp);
+disable_irqs:
+	fun_disable_irqs(dev);
+destroy_vi:
+	fun_res_destroy(fp->fdev, FUN_ADMIN_OP_VI, 0, dev->dev_port);
+free_queues:
+	fun_free_rings(dev, qset);
+	return err;
+}
+
+static int funeth_open(struct net_device *netdev)
+{
+	struct funeth_priv *fp = netdev_priv(netdev);
+	struct fun_qset qset = {
+		.nrxqs = netdev->real_num_rx_queues,
+		.ntxqs = netdev->real_num_tx_queues,
+		.nxdpqs = fp->num_xdpqs,
+		.cq_depth = fp->cq_depth,
+		.rq_depth = fp->rq_depth,
+		.sq_depth = fp->sq_depth,
+		.state = FUN_QSTATE_INIT_FULL,
+	};
+	int rc;
+
+	rc = fun_alloc_rings(netdev, &qset);
+	if (rc)
+		return rc;
+
+	rc = fun_up(netdev, &qset);
+	if (rc) {
+		qset.state = FUN_QSTATE_DESTROYED;
+		fun_free_rings(netdev, &qset);
+	}
+
+	return rc;
+}
+
+static int funeth_close(struct net_device *netdev)
+{
+	struct fun_qset qset = { .state = FUN_QSTATE_DESTROYED };
+
+	fun_down(netdev, &qset);
+	return 0;
+}
+
+static void fun_get_stats64(struct net_device *netdev,
+			    struct rtnl_link_stats64 *stats)
+{
+	struct funeth_priv *fp = netdev_priv(netdev);
+	struct funeth_txq **xdpqs;
+	struct funeth_rxq **rxqs;
+	unsigned int i, start;
+
+	stats->tx_packets = fp->tx_packets;
+	stats->tx_bytes   = fp->tx_bytes;
+	stats->tx_dropped = fp->tx_dropped;
+
+	stats->rx_packets = fp->rx_packets;
+	stats->rx_bytes   = fp->rx_bytes;
+	stats->rx_dropped = fp->rx_dropped;
+
+	rcu_read_lock();
+	rxqs = rcu_dereference(fp->rxqs);
+	if (!rxqs)
+		goto unlock;
+
+	for (i = 0; i < netdev->real_num_tx_queues; i++) {
+		struct funeth_txq_stats txs;
+
+		FUN_QSTAT_READ(fp->txqs[i], start, txs);
+		stats->tx_packets += txs.tx_pkts;
+		stats->tx_bytes   += txs.tx_bytes;
+		stats->tx_dropped += txs.tx_map_err;
+	}
+
+	for (i = 0; i < netdev->real_num_rx_queues; i++) {
+		struct funeth_rxq_stats rxs;
+
+		FUN_QSTAT_READ(rxqs[i], start, rxs);
+		stats->rx_packets += rxs.rx_pkts;
+		stats->rx_bytes   += rxs.rx_bytes;
+		stats->rx_dropped += rxs.rx_map_err + rxs.rx_mem_drops;
+	}
+
+	xdpqs = rcu_dereference(fp->xdpqs);
+	if (!xdpqs)
+		goto unlock;
+
+	for (i = 0; i < fp->num_xdpqs; i++) {
+		struct funeth_txq_stats txs;
+
+		FUN_QSTAT_READ(xdpqs[i], start, txs);
+		stats->tx_packets += txs.tx_pkts;
+		stats->tx_bytes   += txs.tx_bytes;
+	}
+unlock:
+	rcu_read_unlock();
+}
+
+static int fun_change_mtu(struct net_device *netdev, int new_mtu)
+{
+	struct funeth_priv *fp = netdev_priv(netdev);
+	int rc;
+
+	rc = fun_port_write_cmd(fp, FUN_ADMIN_PORT_KEY_MTU, new_mtu);
+	if (!rc)
+		WRITE_ONCE(netdev->mtu, new_mtu);
+	return rc;
+}
+
+static int fun_set_macaddr(struct net_device *netdev, void *addr)
+{
+	struct funeth_priv *fp = netdev_priv(netdev);
+	struct sockaddr *saddr = addr;
+	int rc;
+
+	if (!is_valid_ether_addr(saddr->sa_data))
+		return -EADDRNOTAVAIL;
+
+	if (ether_addr_equal(netdev->dev_addr, saddr->sa_data))
+		return 0;
+
+	rc = fun_port_write_cmd(fp, FUN_ADMIN_PORT_KEY_MACADDR,
+				ether_addr_to_u64(saddr->sa_data));
+	if (!rc)
+		eth_hw_addr_set(netdev, saddr->sa_data);
+	return rc;
+}
+
+static int fun_get_port_attributes(struct net_device *netdev)
+{
+	static const int keys[] = {
+		FUN_ADMIN_PORT_KEY_MACADDR, FUN_ADMIN_PORT_KEY_CAPABILITIES,
+		FUN_ADMIN_PORT_KEY_ADVERT, FUN_ADMIN_PORT_KEY_MTU
+	};
+	static const int phys_keys[] = {
+		FUN_ADMIN_PORT_KEY_LANE_ATTRS,
+	};
+
+	struct funeth_priv *fp = netdev_priv(netdev);
+	u64 data[ARRAY_SIZE(keys)];
+	u8 mac[ETH_ALEN];
+	int i, rc;
+
+	rc = fun_port_read_cmds(fp, ARRAY_SIZE(keys), keys, data);
+	if (rc)
+		return rc;
+
+	for (i = 0; i < ARRAY_SIZE(keys); i++) {
+		switch (keys[i]) {
+		case FUN_ADMIN_PORT_KEY_MACADDR:
+			u64_to_ether_addr(data[i], mac);
+			if (is_zero_ether_addr(mac)) {
+				eth_hw_addr_random(netdev);
+			} else if (is_valid_ether_addr(mac)) {
+				eth_hw_addr_set(netdev, mac);
+			} else {
+				netdev_err(netdev,
+					   "device provided a bad MAC address %pM\n",
+					   mac);
+				return -EINVAL;
+			}
+			break;
+
+		case FUN_ADMIN_PORT_KEY_CAPABILITIES:
+			fp->port_caps = data[i];
+			break;
+
+		case FUN_ADMIN_PORT_KEY_ADVERT:
+			fp->advertising = data[i];
+			break;
+
+		case FUN_ADMIN_PORT_KEY_MTU:
+			netdev->mtu = data[i];
+			break;
+		}
+	}
+
+	if (!(fp->port_caps & FUN_PORT_CAP_VPORT)) {
+		rc = fun_port_read_cmds(fp, ARRAY_SIZE(phys_keys), phys_keys,
+					data);
+		if (rc)
+			return rc;
+
+		fp->lane_attrs = data[0];
+	}
+
+	if (netdev->addr_assign_type == NET_ADDR_RANDOM)
+		return fun_port_write_cmd(fp, FUN_ADMIN_PORT_KEY_MACADDR,
+					  ether_addr_to_u64(netdev->dev_addr));
+	return 0;
+}
+
+static int fun_hwtstamp_get(struct net_device *dev, struct ifreq *ifr)
+{
+	const struct funeth_priv *fp = netdev_priv(dev);
+
+	return copy_to_user(ifr->ifr_data, &fp->hwtstamp_cfg,
+			    sizeof(fp->hwtstamp_cfg)) ? -EFAULT : 0;
+}
+
+static int fun_hwtstamp_set(struct net_device *dev, struct ifreq *ifr)
+{
+	struct funeth_priv *fp = netdev_priv(dev);
+	struct hwtstamp_config cfg;
+
+	if (copy_from_user(&cfg, ifr->ifr_data, sizeof(cfg)))
+		return -EFAULT;
+
+	/* no TX HW timestamps */
+	cfg.tx_type = HWTSTAMP_TX_OFF;
+
+	switch (cfg.rx_filter) {
+	case HWTSTAMP_FILTER_NONE:
+		break;
+	case HWTSTAMP_FILTER_ALL:
+	case HWTSTAMP_FILTER_SOME:
+	case HWTSTAMP_FILTER_PTP_V1_L4_EVENT:
+	case HWTSTAMP_FILTER_PTP_V1_L4_SYNC:
+	case HWTSTAMP_FILTER_PTP_V1_L4_DELAY_REQ:
+	case HWTSTAMP_FILTER_PTP_V2_L4_EVENT:
+	case HWTSTAMP_FILTER_PTP_V2_L4_SYNC:
+	case HWTSTAMP_FILTER_PTP_V2_L4_DELAY_REQ:
+	case HWTSTAMP_FILTER_PTP_V2_L2_EVENT:
+	case HWTSTAMP_FILTER_PTP_V2_L2_SYNC:
+	case HWTSTAMP_FILTER_PTP_V2_L2_DELAY_REQ:
+	case HWTSTAMP_FILTER_PTP_V2_EVENT:
+	case HWTSTAMP_FILTER_PTP_V2_SYNC:
+	case HWTSTAMP_FILTER_PTP_V2_DELAY_REQ:
+	case HWTSTAMP_FILTER_NTP_ALL:
+		cfg.rx_filter = HWTSTAMP_FILTER_ALL;
+		break;
+	default:
+		return -ERANGE;
+	}
+
+	fp->hwtstamp_cfg = cfg;
+	return copy_to_user(ifr->ifr_data, &cfg, sizeof(cfg)) ? -EFAULT : 0;
+}
+
+static int fun_ioctl(struct net_device *dev, struct ifreq *ifr, int cmd)
+{
+	switch (cmd) {
+	case SIOCSHWTSTAMP:
+		return fun_hwtstamp_set(dev, ifr);
+	case SIOCGHWTSTAMP:
+		return fun_hwtstamp_get(dev, ifr);
+	default:
+		return -EOPNOTSUPP;
+	}
+}
+
+/* Prepare the queues for XDP. */
+static int fun_enter_xdp(struct net_device *dev, struct bpf_prog *prog)
+{
+	struct funeth_priv *fp = netdev_priv(dev);
+	unsigned int i, nqs = num_online_cpus();
+	struct funeth_txq **xdpqs;
+	struct funeth_rxq **rxqs;
+	int err;
+
+	xdpqs = alloc_xdpqs(dev, nqs, fp->sq_depth, 0, FUN_QSTATE_INIT_FULL);
+	if (IS_ERR(xdpqs))
+		return PTR_ERR(xdpqs);
+
+	rxqs = rtnl_dereference(fp->rxqs);
+	for (i = 0; i < dev->real_num_rx_queues; i++) {
+		err = fun_rxq_set_bpf(rxqs[i], prog);
+		if (err)
+			goto out;
+	}
+
+	fp->num_xdpqs = nqs;
+	rcu_assign_pointer(fp->xdpqs, xdpqs);
+	return 0;
+out:
+	while (i--)
+		fun_rxq_set_bpf(rxqs[i], NULL);
+
+	free_xdpqs(xdpqs, nqs, 0, FUN_QSTATE_DESTROYED);
+	return err;
+}
+
+/* Set the queues for non-XDP operation. */
+static void fun_end_xdp(struct net_device *dev)
+{
+	struct funeth_priv *fp = netdev_priv(dev);
+	struct funeth_txq **xdpqs;
+	struct funeth_rxq **rxqs;
+	unsigned int i;
+
+	xdpqs = rtnl_dereference(fp->xdpqs);
+	rcu_assign_pointer(fp->xdpqs, NULL);
+	synchronize_net();
+	/* at this point both Rx and Tx XDP processing has ended */
+
+	free_xdpqs(xdpqs, fp->num_xdpqs, 0, FUN_QSTATE_DESTROYED);
+	fp->num_xdpqs = 0;
+
+	rxqs = rtnl_dereference(fp->rxqs);
+	for (i = 0; i < dev->real_num_rx_queues; i++)
+		fun_rxq_set_bpf(rxqs[i], NULL);
+}
+
+#define XDP_MAX_MTU \
+	(PAGE_SIZE - FUN_XDP_HEADROOM - VLAN_ETH_HLEN - FUN_RX_TAILROOM)
+
+static int fun_xdp_setup(struct net_device *dev, struct netdev_bpf *xdp)
+{
+	struct bpf_prog *old_prog, *prog = xdp->prog;
+	struct funeth_priv *fp = netdev_priv(dev);
+	int i, err;
+
+	/* XDP uses at most one buffer */
+	if (prog && dev->mtu > XDP_MAX_MTU) {
+		netdev_err(dev, "device MTU %u too large for XDP\n", dev->mtu);
+		NL_SET_ERR_MSG_MOD(xdp->extack,
+				   "Device MTU too large for XDP");
+		return -EINVAL;
+	}
+
+	if (!netif_running(dev)) {
+		fp->num_xdpqs = prog ? num_online_cpus() : 0;
+	} else if (prog && !fp->xdp_prog) {
+		err = fun_enter_xdp(dev, prog);
+		if (err) {
+			NL_SET_ERR_MSG_MOD(xdp->extack,
+					   "Failed to set queues for XDP.");
+			return err;
+		}
+	} else if (!prog && fp->xdp_prog) {
+		fun_end_xdp(dev);
+	} else {
+		struct funeth_rxq **rxqs = rtnl_dereference(fp->rxqs);
+
+		for (i = 0; i < dev->real_num_rx_queues; i++)
+			WRITE_ONCE(rxqs[i]->xdp_prog, prog);
+	}
+
+	if (prog)
+		xdp_features_set_redirect_target(dev, true);
+	else
+		xdp_features_clear_redirect_target(dev);
+
+	dev->max_mtu = prog ? XDP_MAX_MTU : FUN_MAX_MTU;
+	old_prog = xchg(&fp->xdp_prog, prog);
+	if (old_prog)
+		bpf_prog_put(old_prog);
+
+	return 0;
+}
+
+static int fun_xdp(struct net_device *dev, struct netdev_bpf *xdp)
+{
+	switch (xdp->command) {
+	case XDP_SETUP_PROG:
+		return fun_xdp_setup(dev, xdp);
+	default:
+		return -EINVAL;
+	}
+}
+
+static int fun_init_vports(struct fun_ethdev *ed, unsigned int n)
+{
+	if (ed->num_vports)
+		return -EINVAL;
+
+	ed->vport_info = kvcalloc(n, sizeof(*ed->vport_info), GFP_KERNEL);
+	if (!ed->vport_info)
+		return -ENOMEM;
+	ed->num_vports = n;
+	return 0;
+}
+
+static void fun_free_vports(struct fun_ethdev *ed)
+{
+	kvfree(ed->vport_info);
+	ed->vport_info = NULL;
+	ed->num_vports = 0;
+}
+
+static struct fun_vport_info *fun_get_vport(struct fun_ethdev *ed,
+					    unsigned int vport)
+{
+	if (!ed->vport_info || vport >= ed->num_vports)
+		return NULL;
+
+	return ed->vport_info + vport;
+}
+
+static int fun_set_vf_mac(struct net_device *dev, int vf, u8 *mac)
+{
+	struct funeth_priv *fp = netdev_priv(dev);
+	struct fun_adi_param mac_param = {};
+	struct fun_dev *fdev = fp->fdev;
+	struct fun_ethdev *ed = to_fun_ethdev(fdev);
+	struct fun_vport_info *vi;
+	int rc = -EINVAL;
+
+	if (is_multicast_ether_addr(mac))
+		return -EINVAL;
+
+	mutex_lock(&ed->state_mutex);
+	vi = fun_get_vport(ed, vf);
+	if (!vi)
+		goto unlock;
+
+	mac_param.u.mac = FUN_ADI_MAC_INIT(ether_addr_to_u64(mac));
+	rc = fun_adi_write(fdev, FUN_ADMIN_ADI_ATTR_MACADDR, vf + 1,
+			   &mac_param);
+	if (!rc)
+		ether_addr_copy(vi->mac, mac);
+unlock:
+	mutex_unlock(&ed->state_mutex);
+	return rc;
+}
+
+static int fun_set_vf_vlan(struct net_device *dev, int vf, u16 vlan, u8 qos,
+			   __be16 vlan_proto)
+{
+	struct funeth_priv *fp = netdev_priv(dev);
+	struct fun_adi_param vlan_param = {};
+	struct fun_dev *fdev = fp->fdev;
+	struct fun_ethdev *ed = to_fun_ethdev(fdev);
+	struct fun_vport_info *vi;
+	int rc = -EINVAL;
+
+	if (vlan > 4095 || qos > 7)
+		return -EINVAL;
+	if (vlan_proto && vlan_proto != htons(ETH_P_8021Q) &&
+	    vlan_proto != htons(ETH_P_8021AD))
+		return -EINVAL;
+
+	mutex_lock(&ed->state_mutex);
+	vi = fun_get_vport(ed, vf);
+	if (!vi)
+		goto unlock;
+
+	vlan_param.u.vlan = FUN_ADI_VLAN_INIT(be16_to_cpu(vlan_proto),
+					      ((u16)qos << VLAN_PRIO_SHIFT) | vlan);
+	rc = fun_adi_write(fdev, FUN_ADMIN_ADI_ATTR_VLAN, vf + 1, &vlan_param);
+	if (!rc) {
+		vi->vlan = vlan;
+		vi->qos = qos;
+		vi->vlan_proto = vlan_proto;
+	}
+unlock:
+	mutex_unlock(&ed->state_mutex);
+	return rc;
+}
+
+static int fun_set_vf_rate(struct net_device *dev, int vf, int min_tx_rate,
+			   int max_tx_rate)
+{
+	struct funeth_priv *fp = netdev_priv(dev);
+	struct fun_adi_param rate_param = {};
+	struct fun_dev *fdev = fp->fdev;
+	struct fun_ethdev *ed = to_fun_ethdev(fdev);
+	struct fun_vport_info *vi;
+	int rc = -EINVAL;
+
+	if (min_tx_rate)
+		return -EINVAL;
+
+	mutex_lock(&ed->state_mutex);
+	vi = fun_get_vport(ed, vf);
+	if (!vi)
+		goto unlock;
+
+	rate_param.u.rate = FUN_ADI_RATE_INIT(max_tx_rate);
+	rc = fun_adi_write(fdev, FUN_ADMIN_ADI_ATTR_RATE, vf + 1, &rate_param);
+	if (!rc)
+		vi->max_rate = max_tx_rate;
+unlock:
+	mutex_unlock(&ed->state_mutex);
+	return rc;
+}
+
+static int fun_get_vf_config(struct net_device *dev, int vf,
+			     struct ifla_vf_info *ivi)
+{
+	struct funeth_priv *fp = netdev_priv(dev);
+	struct fun_ethdev *ed = to_fun_ethdev(fp->fdev);
+	const struct fun_vport_info *vi;
+
+	mutex_lock(&ed->state_mutex);
+	vi = fun_get_vport(ed, vf);
+	if (!vi)
+		goto unlock;
+
+	memset(ivi, 0, sizeof(*ivi));
+	ivi->vf = vf;
+	ether_addr_copy(ivi->mac, vi->mac);
+	ivi->vlan = vi->vlan;
+	ivi->qos = vi->qos;
+	ivi->vlan_proto = vi->vlan_proto;
+	ivi->max_tx_rate = vi->max_rate;
+	ivi->spoofchk = vi->spoofchk;
+unlock:
+	mutex_unlock(&ed->state_mutex);
+	return vi ? 0 : -EINVAL;
+}
+
+static void fun_uninit(struct net_device *dev)
+{
+	struct funeth_priv *fp = netdev_priv(dev);
+
+	fun_prune_queue_irqs(dev);
+	xa_destroy(&fp->irqs);
+}
+
+static const struct net_device_ops fun_netdev_ops = {
+	.ndo_open		= funeth_open,
+	.ndo_stop		= funeth_close,
+	.ndo_start_xmit		= fun_start_xmit,
+	.ndo_get_stats64	= fun_get_stats64,
+	.ndo_change_mtu		= fun_change_mtu,
+	.ndo_set_mac_address	= fun_set_macaddr,
+	.ndo_validate_addr	= eth_validate_addr,
+	.ndo_eth_ioctl		= fun_ioctl,
+	.ndo_uninit		= fun_uninit,
+	.ndo_bpf		= fun_xdp,
+	.ndo_xdp_xmit		= fun_xdp_xmit_frames,
+	.ndo_set_vf_mac		= fun_set_vf_mac,
+	.ndo_set_vf_vlan	= fun_set_vf_vlan,
+	.ndo_set_vf_rate	= fun_set_vf_rate,
+	.ndo_get_vf_config	= fun_get_vf_config,
+};
+
+#define GSO_ENCAP_FLAGS (NETIF_F_GSO_GRE | NETIF_F_GSO_IPXIP4 | \
+			 NETIF_F_GSO_IPXIP6 | NETIF_F_GSO_UDP_TUNNEL | \
+			 NETIF_F_GSO_UDP_TUNNEL_CSUM)
+#define TSO_FLAGS (NETIF_F_TSO | NETIF_F_TSO6 | NETIF_F_TSO_ECN | \
+		   NETIF_F_GSO_UDP_L4)
+#define VLAN_FEAT (NETIF_F_SG | NETIF_F_HW_CSUM | TSO_FLAGS | \
+		   GSO_ENCAP_FLAGS | NETIF_F_HIGHDMA)
+
+static void fun_dflt_rss_indir(struct funeth_priv *fp, unsigned int nrx)
+{
+	unsigned int i;
+
+	for (i = 0; i < fp->indir_table_nentries; i++)
+		fp->indir_table[i] = ethtool_rxfh_indir_default(i, nrx);
+}
+
+/* Reset the RSS indirection table to equal distribution across the current
+ * number of Rx queues. Called at init time and whenever the number of Rx
+ * queues changes subsequently. Note that this may also resize the indirection
+ * table.
+ */
+static void fun_reset_rss_indir(struct net_device *dev, unsigned int nrx)
+{
+	struct funeth_priv *fp = netdev_priv(dev);
+
+	if (!fp->rss_cfg)
+		return;
+
+	/* Set the table size to the max possible that allows an equal number
+	 * of occurrences of each CQ.
+	 */
+	fp->indir_table_nentries = rounddown(FUN_ETH_RSS_MAX_INDIR_ENT, nrx);
+	fun_dflt_rss_indir(fp, nrx);
+}
+
+/* Update the RSS LUT to contain only queues in [0, nrx). Normally this will
+ * update the LUT to an equal distribution among nrx queues, If @only_if_needed
+ * is set the LUT is left unchanged if it already does not reference any queues
+ * >= nrx.
+ */
+static int fun_rss_set_qnum(struct net_device *dev, unsigned int nrx,
+			    bool only_if_needed)
+{
+	struct funeth_priv *fp = netdev_priv(dev);
+	u32 old_lut[FUN_ETH_RSS_MAX_INDIR_ENT];
+	unsigned int i, oldsz;
+	int err;
+
+	if (!fp->rss_cfg)
+		return 0;
+
+	if (only_if_needed) {
+		for (i = 0; i < fp->indir_table_nentries; i++)
+			if (fp->indir_table[i] >= nrx)
+				break;
+
+		if (i >= fp->indir_table_nentries)
+			return 0;
+	}
+
+	memcpy(old_lut, fp->indir_table, sizeof(old_lut));
+	oldsz = fp->indir_table_nentries;
+	fun_reset_rss_indir(dev, nrx);
+
+	err = fun_config_rss(dev, fp->hash_algo, fp->rss_key,
+			     fp->indir_table, FUN_ADMIN_SUBOP_MODIFY);
+	if (!err)
+		return 0;
+
+	memcpy(fp->indir_table, old_lut, sizeof(old_lut));
+	fp->indir_table_nentries = oldsz;
+	return err;
+}
+
+/* Allocate the DMA area for the RSS configuration commands to the device, and
+ * initialize the hash, hash key, indirection table size and its entries to
+ * their defaults. The indirection table defaults to equal distribution across
+ * the Rx queues.
+ */
+static int fun_init_rss(struct net_device *dev)
+{
+	struct funeth_priv *fp = netdev_priv(dev);
+	size_t size = sizeof(fp->rss_key) + sizeof(fp->indir_table);
+
+	fp->rss_hw_id = FUN_HCI_ID_INVALID;
+	if (!(fp->port_caps & FUN_PORT_CAP_OFFLOADS))
+		return 0;
+
+	fp->rss_cfg = dma_alloc_coherent(&fp->pdev->dev, size,
+					 &fp->rss_dma_addr, GFP_KERNEL);
+	if (!fp->rss_cfg)
+		return -ENOMEM;
+
+	fp->hash_algo = FUN_ETH_RSS_ALG_TOEPLITZ;
+	netdev_rss_key_fill(fp->rss_key, sizeof(fp->rss_key));
+	fun_reset_rss_indir(dev, dev->real_num_rx_queues);
+	return 0;
+}
+
+static void fun_free_rss(struct funeth_priv *fp)
+{
+	if (fp->rss_cfg) {
+		dma_free_coherent(&fp->pdev->dev,
+				  sizeof(fp->rss_key) + sizeof(fp->indir_table),
+				  fp->rss_cfg, fp->rss_dma_addr);
+		fp->rss_cfg = NULL;
+	}
+}
+
+void fun_set_ring_count(struct net_device *netdev, unsigned int ntx,
+			unsigned int nrx)
+{
+	netif_set_real_num_tx_queues(netdev, ntx);
+	if (nrx != netdev->real_num_rx_queues) {
+		netif_set_real_num_rx_queues(netdev, nrx);
+		fun_reset_rss_indir(netdev, nrx);
+	}
+}
+
+static int fun_init_stats_area(struct funeth_priv *fp)
+{
+	unsigned int nstats;
+
+	if (!(fp->port_caps & FUN_PORT_CAP_STATS))
+		return 0;
+
+	nstats = PORT_MAC_RX_STATS_MAX + PORT_MAC_TX_STATS_MAX +
+		 PORT_MAC_FEC_STATS_MAX;
+
+	fp->stats = dma_alloc_coherent(&fp->pdev->dev, nstats * sizeof(u64),
+				       &fp->stats_dma_addr, GFP_KERNEL);
+	if (!fp->stats)
+		return -ENOMEM;
+	return 0;
+}
+
+static void fun_free_stats_area(struct funeth_priv *fp)
+{
+	unsigned int nstats;
+
+	if (fp->stats) {
+		nstats = PORT_MAC_RX_STATS_MAX + PORT_MAC_TX_STATS_MAX;
+		dma_free_coherent(&fp->pdev->dev, nstats * sizeof(u64),
+				  fp->stats, fp->stats_dma_addr);
+		fp->stats = NULL;
+	}
+}
+
+static int fun_dl_port_register(struct net_device *netdev)
+{
+	struct funeth_priv *fp = netdev_priv(netdev);
+	struct devlink *dl = priv_to_devlink(fp->fdev);
+	struct devlink_port_attrs attrs = {};
+	unsigned int idx;
+
+	if (fp->port_caps & FUN_PORT_CAP_VPORT) {
+		attrs.flavour = DEVLINK_PORT_FLAVOUR_VIRTUAL;
+		idx = fp->lport;
+	} else {
+		idx = netdev->dev_port;
+		attrs.flavour = DEVLINK_PORT_FLAVOUR_PHYSICAL;
+		attrs.lanes = fp->lane_attrs & 7;
+		if (fp->lane_attrs & FUN_PORT_LANE_SPLIT) {
+			attrs.split = 1;
+			attrs.phys.port_number = fp->lport & ~3;
+			attrs.phys.split_subport_number = fp->lport & 3;
+		} else {
+			attrs.phys.port_number = fp->lport;
+		}
+	}
+
+	devlink_port_attrs_set(&fp->dl_port, &attrs);
+
+	return devlink_port_register(dl, &fp->dl_port, idx);
+}
+
+/* Determine the max Tx/Rx queues for a port. */
+static int fun_max_qs(struct fun_ethdev *ed, unsigned int *ntx,
+		      unsigned int *nrx)
+{
+	int neth;
+
+	if (ed->num_ports > 1 || is_kdump_kernel()) {
+		*ntx = 1;
+		*nrx = 1;
+		return 0;
+	}
+
+	neth = fun_get_res_count(&ed->fdev, FUN_ADMIN_OP_ETH);
+	if (neth < 0)
+		return neth;
+
+	/* We determine the max number of queues based on the CPU
+	 * cores, device interrupts and queues, RSS size, and device Tx flows.
+	 *
+	 * - At least 1 Rx and 1 Tx queues.
+	 * - At most 1 Rx/Tx queue per core.
+	 * - Each Rx/Tx queue needs 1 SQ.
+	 */
+	*ntx = min(ed->nsqs_per_port - 1, num_online_cpus());
+	*nrx = *ntx;
+	if (*ntx > neth)
+		*ntx = neth;
+	if (*nrx > FUN_ETH_RSS_MAX_INDIR_ENT)
+		*nrx = FUN_ETH_RSS_MAX_INDIR_ENT;
+	return 0;
+}
+
+static void fun_queue_defaults(struct net_device *dev, unsigned int nsqs)
+{
+	unsigned int ntx, nrx;
+
+	ntx = min(dev->num_tx_queues, FUN_DFLT_QUEUES);
+	nrx = min(dev->num_rx_queues, FUN_DFLT_QUEUES);
+	if (ntx <= nrx) {
+		ntx = min(ntx, nsqs / 2);
+		nrx = min(nrx, nsqs - ntx);
+	} else {
+		nrx = min(nrx, nsqs / 2);
+		ntx = min(ntx, nsqs - nrx);
+	}
+
+	netif_set_real_num_tx_queues(dev, ntx);
+	netif_set_real_num_rx_queues(dev, nrx);
+}
+
+/* Replace the existing Rx/Tx/XDP queues with equal number of queues with
+ * different settings, e.g. depth. This is a disruptive replacement that
+ * temporarily shuts down the data path and should be limited to changes that
+ * can't be applied to live queues. The old queues are always discarded.
+ */
+int fun_replace_queues(struct net_device *dev, struct fun_qset *newqs,
+		       struct netlink_ext_ack *extack)
+{
+	struct fun_qset oldqs = { .state = FUN_QSTATE_DESTROYED };
+	struct funeth_priv *fp = netdev_priv(dev);
+	int err;
+
+	newqs->nrxqs = dev->real_num_rx_queues;
+	newqs->ntxqs = dev->real_num_tx_queues;
+	newqs->nxdpqs = fp->num_xdpqs;
+	newqs->state = FUN_QSTATE_INIT_SW;
+	err = fun_alloc_rings(dev, newqs);
+	if (err) {
+		NL_SET_ERR_MSG_MOD(extack,
+				   "Unable to allocate memory for new queues, keeping current settings");
+		return err;
+	}
+
+	fun_down(dev, &oldqs);
+
+	err = fun_up(dev, newqs);
+	if (!err)
+		return 0;
+
+	/* The new queues couldn't be installed. We do not retry the old queues
+	 * as they are the same to the device as the new queues and would
+	 * similarly fail.
+	 */
+	newqs->state = FUN_QSTATE_DESTROYED;
+	fun_free_rings(dev, newqs);
+	NL_SET_ERR_MSG_MOD(extack, "Unable to restore the data path with the new queues.");
+	return err;
+}
+
+/* Change the number of Rx/Tx queues of a device while it is up. This is done
+ * by incrementally adding/removing queues to meet the new requirements while
+ * handling ongoing traffic.
+ */
+int fun_change_num_queues(struct net_device *dev, unsigned int ntx,
+			  unsigned int nrx)
+{
+	unsigned int keep_tx = min(dev->real_num_tx_queues, ntx);
+	unsigned int keep_rx = min(dev->real_num_rx_queues, nrx);
+	struct funeth_priv *fp = netdev_priv(dev);
+	struct fun_qset oldqs = {
+		.rxqs = rtnl_dereference(fp->rxqs),
+		.txqs = fp->txqs,
+		.nrxqs = dev->real_num_rx_queues,
+		.ntxqs = dev->real_num_tx_queues,
+		.rxq_start = keep_rx,
+		.txq_start = keep_tx,
+		.state = FUN_QSTATE_DESTROYED
+	};
+	struct fun_qset newqs = {
+		.nrxqs = nrx,
+		.ntxqs = ntx,
+		.rxq_start = keep_rx,
+		.txq_start = keep_tx,
+		.cq_depth = fp->cq_depth,
+		.rq_depth = fp->rq_depth,
+		.sq_depth = fp->sq_depth,
+		.state = FUN_QSTATE_INIT_FULL
+	};
+	int i, err;
+
+	err = fun_alloc_rings(dev, &newqs);
+	if (err)
+		goto free_irqs;
+
+	err = fun_enable_irqs(dev); /* of any newly added queues */
+	if (err)
+		goto free_rings;
+
+	/* copy the queues we are keeping to the new set */
+	memcpy(newqs.rxqs, oldqs.rxqs, keep_rx * sizeof(*oldqs.rxqs));
+	memcpy(newqs.txqs, fp->txqs, keep_tx * sizeof(*fp->txqs));
+
+	if (nrx < dev->real_num_rx_queues) {
+		err = fun_rss_set_qnum(dev, nrx, true);
+		if (err)
+			goto disable_tx_irqs;
+
+		for (i = nrx; i < dev->real_num_rx_queues; i++)
+			fun_disable_one_irq(container_of(oldqs.rxqs[i]->napi,
+							 struct fun_irq, napi));
+
+		netif_set_real_num_rx_queues(dev, nrx);
+	}
+
+	if (ntx < dev->real_num_tx_queues)
+		netif_set_real_num_tx_queues(dev, ntx);
+
+	rcu_assign_pointer(fp->rxqs, newqs.rxqs);
+	fp->txqs = newqs.txqs;
+	synchronize_net();
+
+	if (ntx > dev->real_num_tx_queues)
+		netif_set_real_num_tx_queues(dev, ntx);
+
+	if (nrx > dev->real_num_rx_queues) {
+		netif_set_real_num_rx_queues(dev, nrx);
+		fun_rss_set_qnum(dev, nrx, false);
+	}
+
+	/* disable interrupts of any excess Tx queues */
+	for (i = keep_tx; i < oldqs.ntxqs; i++)
+		fun_disable_one_irq(oldqs.txqs[i]->irq);
+
+	fun_free_rings(dev, &oldqs);
+	fun_prune_queue_irqs(dev);
+	return 0;
+
+disable_tx_irqs:
+	for (i = oldqs.ntxqs; i < ntx; i++)
+		fun_disable_one_irq(newqs.txqs[i]->irq);
+free_rings:
+	newqs.state = FUN_QSTATE_DESTROYED;
+	fun_free_rings(dev, &newqs);
+free_irqs:
+	fun_prune_queue_irqs(dev);
+	return err;
+}
+
+static int fun_create_netdev(struct fun_ethdev *ed, unsigned int portid)
+{
+	struct fun_dev *fdev = &ed->fdev;
+	struct net_device *netdev;
+	struct funeth_priv *fp;
+	unsigned int ntx, nrx;
+	int rc;
+
+	rc = fun_max_qs(ed, &ntx, &nrx);
+	if (rc)
+		return rc;
+
+	netdev = alloc_etherdev_mqs(sizeof(*fp), ntx, nrx);
+	if (!netdev) {
+		rc = -ENOMEM;
+		goto done;
+	}
+
+	netdev->dev_port = portid;
+	fun_queue_defaults(netdev, ed->nsqs_per_port);
+
+	fp = netdev_priv(netdev);
+	fp->fdev = fdev;
+	fp->pdev = to_pci_dev(fdev->dev);
+	fp->netdev = netdev;
+	xa_init(&fp->irqs);
+	fp->rx_irq_ofst = ntx;
+	seqcount_init(&fp->link_seq);
+
+	fp->lport = INVALID_LPORT;
+	rc = fun_port_create(netdev);
+	if (rc)
+		goto free_netdev;
+
+	/* bind port to admin CQ for async events */
+	rc = fun_bind(fdev, FUN_ADMIN_BIND_TYPE_PORT, portid,
+		      FUN_ADMIN_BIND_TYPE_EPCQ, 0);
+	if (rc)
+		goto destroy_port;
+
+	rc = fun_get_port_attributes(netdev);
+	if (rc)
+		goto destroy_port;
+
+	rc = fun_init_rss(netdev);
+	if (rc)
+		goto destroy_port;
+
+	rc = fun_init_stats_area(fp);
+	if (rc)
+		goto free_rss;
+
+	SET_NETDEV_DEV(netdev, fdev->dev);
+	SET_NETDEV_DEVLINK_PORT(netdev, &fp->dl_port);
+	netdev->netdev_ops = &fun_netdev_ops;
+
+	netdev->hw_features = NETIF_F_SG | NETIF_F_RXHASH | NETIF_F_RXCSUM;
+	if (fp->port_caps & FUN_PORT_CAP_OFFLOADS)
+		netdev->hw_features |= NETIF_F_HW_CSUM | TSO_FLAGS;
+	if (fp->port_caps & FUN_PORT_CAP_ENCAP_OFFLOADS)
+		netdev->hw_features |= GSO_ENCAP_FLAGS;
+
+	netdev->features |= netdev->hw_features | NETIF_F_HIGHDMA;
+	netdev->vlan_features = netdev->features & VLAN_FEAT;
+	netdev->mpls_features = netdev->vlan_features;
+	netdev->hw_enc_features = netdev->hw_features;
+	netdev->xdp_features = NETDEV_XDP_ACT_BASIC | NETDEV_XDP_ACT_REDIRECT;
+
+	netdev->min_mtu = ETH_MIN_MTU;
+	netdev->max_mtu = FUN_MAX_MTU;
+
+	fun_set_ethtool_ops(netdev);
+
+	/* configurable parameters */
+	fp->sq_depth = min(SQ_DEPTH, fdev->q_depth);
+	fp->cq_depth = min(CQ_DEPTH, fdev->q_depth);
+	fp->rq_depth = min_t(unsigned int, RQ_DEPTH, fdev->q_depth);
+	fp->rx_coal_usec  = CQ_INTCOAL_USEC;
+	fp->rx_coal_count = CQ_INTCOAL_NPKT;
+	fp->tx_coal_usec  = SQ_INTCOAL_USEC;
+	fp->tx_coal_count = SQ_INTCOAL_NPKT;
+	fp->cq_irq_db = FUN_IRQ_CQ_DB(fp->rx_coal_usec, fp->rx_coal_count);
+
+	rc = fun_dl_port_register(netdev);
+	if (rc)
+		goto free_stats;
+
+	fp->ktls_id = FUN_HCI_ID_INVALID;
+	fun_ktls_init(netdev);            /* optional, failure OK */
+
+	netif_carrier_off(netdev);
+	ed->netdevs[portid] = netdev;
+	rc = register_netdev(netdev);
+	if (rc)
+		goto unreg_devlink;
+	return 0;
+
+unreg_devlink:
+	ed->netdevs[portid] = NULL;
+	fun_ktls_cleanup(fp);
+	devlink_port_unregister(&fp->dl_port);
+free_stats:
+	fun_free_stats_area(fp);
+free_rss:
+	fun_free_rss(fp);
+destroy_port:
+	fun_port_destroy(netdev);
+free_netdev:
+	free_netdev(netdev);
+done:
+	dev_err(fdev->dev, "couldn't allocate port %u, error %d", portid, rc);
+	return rc;
+}
+
+static void fun_destroy_netdev(struct net_device *netdev)
+{
+	struct funeth_priv *fp;
+
+	fp = netdev_priv(netdev);
+	unregister_netdev(netdev);
+	devlink_port_unregister(&fp->dl_port);
+	fun_ktls_cleanup(fp);
+	fun_free_stats_area(fp);
+	fun_free_rss(fp);
+	fun_port_destroy(netdev);
+	free_netdev(netdev);
+}
+
+static int fun_create_ports(struct fun_ethdev *ed, unsigned int nports)
+{
+	struct fun_dev *fd = &ed->fdev;
+	int i, rc;
+
+	/* The admin queue takes 1 IRQ and 2 SQs. */
+	ed->nsqs_per_port = min(fd->num_irqs - 1,
+				fd->kern_end_qid - 2) / nports;
+	if (ed->nsqs_per_port < 2) {
+		dev_err(fd->dev, "Too few SQs for %u ports", nports);
+		return -EINVAL;
+	}
+
+	ed->netdevs = kcalloc(nports, sizeof(*ed->netdevs), GFP_KERNEL);
+	if (!ed->netdevs)
+		return -ENOMEM;
+
+	ed->num_ports = nports;
+	for (i = 0; i < nports; i++) {
+		rc = fun_create_netdev(ed, i);
+		if (rc)
+			goto free_netdevs;
+	}
+
+	return 0;
+
+free_netdevs:
+	while (i)
+		fun_destroy_netdev(ed->netdevs[--i]);
+	kfree(ed->netdevs);
+	ed->netdevs = NULL;
+	ed->num_ports = 0;
+	return rc;
+}
+
+static void fun_destroy_ports(struct fun_ethdev *ed)
+{
+	unsigned int i;
+
+	for (i = 0; i < ed->num_ports; i++)
+		fun_destroy_netdev(ed->netdevs[i]);
+
+	kfree(ed->netdevs);
+	ed->netdevs = NULL;
+	ed->num_ports = 0;
+}
+
+static void fun_update_link_state(const struct fun_ethdev *ed,
+				  const struct fun_admin_port_notif *notif)
+{
+	unsigned int port_idx = be16_to_cpu(notif->id);
+	struct net_device *netdev;
+	struct funeth_priv *fp;
+
+	if (port_idx >= ed->num_ports)
+		return;
+
+	netdev = ed->netdevs[port_idx];
+	fp = netdev_priv(netdev);
+
+	write_seqcount_begin(&fp->link_seq);
+	fp->link_speed = be32_to_cpu(notif->speed) * 10;  /* 10 Mbps->Mbps */
+	fp->active_fc = notif->flow_ctrl;
+	fp->active_fec = notif->fec;
+	fp->xcvr_type = notif->xcvr_type;
+	fp->link_down_reason = notif->link_down_reason;
+	fp->lp_advertising = be64_to_cpu(notif->lp_advertising);
+
+	if ((notif->link_state | notif->missed_events) & FUN_PORT_FLAG_MAC_DOWN)
+		netif_carrier_off(netdev);
+	if (notif->link_state & FUN_PORT_FLAG_MAC_UP)
+		netif_carrier_on(netdev);
+
+	write_seqcount_end(&fp->link_seq);
+	fun_report_link(netdev);
+}
+
+/* handler for async events delivered through the admin CQ */
+static void fun_event_cb(struct fun_dev *fdev, void *entry)
+{
+	u8 op = ((struct fun_admin_rsp_common *)entry)->op;
+
+	if (op == FUN_ADMIN_OP_PORT) {
+		const struct fun_admin_port_notif *rsp = entry;
+
+		if (rsp->subop == FUN_ADMIN_SUBOP_NOTIFY) {
+			fun_update_link_state(to_fun_ethdev(fdev), rsp);
+		} else if (rsp->subop == FUN_ADMIN_SUBOP_RES_COUNT) {
+			const struct fun_admin_res_count_rsp *r = entry;
+
+			if (r->count.data)
+				set_bit(FUN_SERV_RES_CHANGE, &fdev->service_flags);
+			else
+				set_bit(FUN_SERV_DEL_PORTS, &fdev->service_flags);
+			fun_serv_sched(fdev);
+		} else {
+			dev_info(fdev->dev, "adminq event unexpected op %u subop %u",
+				 op, rsp->subop);
+		}
+	} else {
+		dev_info(fdev->dev, "adminq event unexpected op %u", op);
+	}
+}
+
+/* handler for pending work managed by the service task */
+static void fun_service_cb(struct fun_dev *fdev)
+{
+	struct fun_ethdev *ed = to_fun_ethdev(fdev);
+	int rc;
+
+	if (test_and_clear_bit(FUN_SERV_DEL_PORTS, &fdev->service_flags))
+		fun_destroy_ports(ed);
+
+	if (!test_and_clear_bit(FUN_SERV_RES_CHANGE, &fdev->service_flags))
+		return;
+
+	rc = fun_get_res_count(fdev, FUN_ADMIN_OP_PORT);
+	if (rc < 0 || rc == ed->num_ports)
+		return;
+
+	if (ed->num_ports)
+		fun_destroy_ports(ed);
+	if (rc)
+		fun_create_ports(ed, rc);
+}
+
+static int funeth_sriov_configure(struct pci_dev *pdev, int nvfs)
+{
+	struct fun_dev *fdev = pci_get_drvdata(pdev);
+	struct fun_ethdev *ed = to_fun_ethdev(fdev);
+	int rc;
+
+	if (nvfs == 0) {
+		if (pci_vfs_assigned(pdev)) {
+			dev_warn(&pdev->dev,
+				 "Cannot disable SR-IOV while VFs are assigned\n");
+			return -EPERM;
+		}
+
+		mutex_lock(&ed->state_mutex);
+		fun_free_vports(ed);
+		mutex_unlock(&ed->state_mutex);
+		pci_disable_sriov(pdev);
+		return 0;
+	}
+
+	rc = pci_enable_sriov(pdev, nvfs);
+	if (rc)
+		return rc;
+
+	mutex_lock(&ed->state_mutex);
+	rc = fun_init_vports(ed, nvfs);
+	mutex_unlock(&ed->state_mutex);
+	if (rc) {
+		pci_disable_sriov(pdev);
+		return rc;
+	}
+
+	return nvfs;
+}
+
+static int funeth_probe(struct pci_dev *pdev, const struct pci_device_id *id)
+{
+	struct fun_dev_params aqreq = {
+		.cqe_size_log2 = ilog2(ADMIN_CQE_SIZE),
+		.sqe_size_log2 = ilog2(ADMIN_SQE_SIZE),
+		.cq_depth      = ADMIN_CQ_DEPTH,
+		.sq_depth      = ADMIN_SQ_DEPTH,
+		.rq_depth      = ADMIN_RQ_DEPTH,
+		.min_msix      = 2,              /* 1 Rx + 1 Tx */
+		.event_cb      = fun_event_cb,
+		.serv_cb       = fun_service_cb,
+	};
+	struct devlink *devlink;
+	struct fun_ethdev *ed;
+	struct fun_dev *fdev;
+	int rc;
+
+	devlink = fun_devlink_alloc(&pdev->dev);
+	if (!devlink) {
+		dev_err(&pdev->dev, "devlink alloc failed\n");
+		return -ENOMEM;
+	}
+
+	ed = devlink_priv(devlink);
+	mutex_init(&ed->state_mutex);
+
+	fdev = &ed->fdev;
+	rc = fun_dev_enable(fdev, pdev, &aqreq, KBUILD_MODNAME);
+	if (rc)
+		goto free_devlink;
+
+	rc = fun_get_res_count(fdev, FUN_ADMIN_OP_PORT);
+	if (rc > 0)
+		rc = fun_create_ports(ed, rc);
+	if (rc < 0)
+		goto disable_dev;
+
+	fun_serv_restart(fdev);
+	fun_devlink_register(devlink);
+	return 0;
+
+disable_dev:
+	fun_dev_disable(fdev);
+free_devlink:
+	mutex_destroy(&ed->state_mutex);
+	fun_devlink_free(devlink);
+	return rc;
+}
+
+static void funeth_remove(struct pci_dev *pdev)
+{
+	struct fun_dev *fdev = pci_get_drvdata(pdev);
+	struct devlink *devlink;
+	struct fun_ethdev *ed;
+
+	ed = to_fun_ethdev(fdev);
+	devlink = priv_to_devlink(ed);
+	fun_devlink_unregister(devlink);
+
+#ifdef CONFIG_PCI_IOV
+	funeth_sriov_configure(pdev, 0);
+#endif
+
+	fun_serv_stop(fdev);
+	fun_destroy_ports(ed);
+	fun_dev_disable(fdev);
+	mutex_destroy(&ed->state_mutex);
+
+	fun_devlink_free(devlink);
+}
+
+static struct pci_driver funeth_driver = {
+	.name		 = KBUILD_MODNAME,
+	.id_table	 = funeth_id_table,
+	.probe		 = funeth_probe,
+	.remove		 = funeth_remove,
+	.shutdown	 = funeth_remove,
+	.sriov_configure = funeth_sriov_configure,
+};
+
+module_pci_driver(funeth_driver);
+
+MODULE_AUTHOR("Dimitris Michailidis <dmichail@fungible.com>");
+MODULE_DESCRIPTION("Fungible Ethernet Network Driver");
+MODULE_LICENSE("Dual BSD/GPL");
+MODULE_DEVICE_TABLE(pci, funeth_id_table);
diff --git a/drivers/net/ethernet/fungible/funeth/funeth_rx.c b/drivers/net/ethernet/fungible/funeth/funeth_rx.c
new file mode 100644
index 000000000000..7e2584895de3
--- /dev/null
+++ b/drivers/net/ethernet/fungible/funeth/funeth_rx.c
@@ -0,0 +1,828 @@
+// SPDX-License-Identifier: (GPL-2.0-only OR BSD-3-Clause)
+
+#include <linux/bpf_trace.h>
+#include <linux/dma-mapping.h>
+#include <linux/etherdevice.h>
+#include <linux/filter.h>
+#include <linux/irq.h>
+#include <linux/pci.h>
+#include <linux/skbuff.h>
+#include "funeth_txrx.h"
+#include "funeth.h"
+#include "fun_queue.h"
+
+#define CREATE_TRACE_POINTS
+#include "funeth_trace.h"
+
+/* Given the device's max supported MTU and pages of at least 4KB a packet can
+ * be scattered into at most 4 buffers.
+ */
+#define RX_MAX_FRAGS 4
+
+/* Per packet headroom in non-XDP mode. Present only for 1-frag packets. */
+#define FUN_RX_HEADROOM (NET_SKB_PAD + NET_IP_ALIGN)
+
+/* We try to reuse pages for our buffers. To avoid frequent page ref writes we
+ * take EXTRA_PAGE_REFS references at once and then hand them out one per packet
+ * occupying the buffer.
+ */
+#define EXTRA_PAGE_REFS 1000000
+#define MIN_PAGE_REFS 1000
+
+enum {
+	FUN_XDP_FLUSH_REDIR = 1,
+	FUN_XDP_FLUSH_TX = 2,
+};
+
+/* See if a page is running low on refs we are holding and if so take more. */
+static void refresh_refs(struct funeth_rxbuf *buf)
+{
+	if (unlikely(buf->pg_refs < MIN_PAGE_REFS)) {
+		buf->pg_refs += EXTRA_PAGE_REFS;
+		page_ref_add(buf->page, EXTRA_PAGE_REFS);
+	}
+}
+
+/* Offer a buffer to the Rx buffer cache. The cache will hold the buffer if its
+ * page is worth retaining and there's room for it. Otherwise the page is
+ * unmapped and our references released.
+ */
+static void cache_offer(struct funeth_rxq *q, const struct funeth_rxbuf *buf)
+{
+	struct funeth_rx_cache *c = &q->cache;
+
+	if (c->prod_cnt - c->cons_cnt <= c->mask && buf->node == numa_mem_id()) {
+		c->bufs[c->prod_cnt & c->mask] = *buf;
+		c->prod_cnt++;
+	} else {
+		dma_unmap_page_attrs(q->dma_dev, buf->dma_addr, PAGE_SIZE,
+				     DMA_FROM_DEVICE, DMA_ATTR_SKIP_CPU_SYNC);
+		__page_frag_cache_drain(buf->page, buf->pg_refs);
+	}
+}
+
+/* Get a page from the Rx buffer cache. We only consider the next available
+ * page and return it if we own all its references.
+ */
+static bool cache_get(struct funeth_rxq *q, struct funeth_rxbuf *rb)
+{
+	struct funeth_rx_cache *c = &q->cache;
+	struct funeth_rxbuf *buf;
+
+	if (c->prod_cnt == c->cons_cnt)
+		return false;             /* empty cache */
+
+	buf = &c->bufs[c->cons_cnt & c->mask];
+	if (page_ref_count(buf->page) == buf->pg_refs) {
+		dma_sync_single_for_device(q->dma_dev, buf->dma_addr,
+					   PAGE_SIZE, DMA_FROM_DEVICE);
+		*rb = *buf;
+		buf->page = NULL;
+		refresh_refs(rb);
+		c->cons_cnt++;
+		return true;
+	}
+
+	/* Page can't be reused. If the cache is full drop this page. */
+	if (c->prod_cnt - c->cons_cnt > c->mask) {
+		dma_unmap_page_attrs(q->dma_dev, buf->dma_addr, PAGE_SIZE,
+				     DMA_FROM_DEVICE, DMA_ATTR_SKIP_CPU_SYNC);
+		__page_frag_cache_drain(buf->page, buf->pg_refs);
+		buf->page = NULL;
+		c->cons_cnt++;
+	}
+	return false;
+}
+
+/* Allocate and DMA-map a page for receive. */
+static int funeth_alloc_page(struct funeth_rxq *q, struct funeth_rxbuf *rb,
+			     int node, gfp_t gfp)
+{
+	struct page *p;
+
+	if (cache_get(q, rb))
+		return 0;
+
+	p = __alloc_pages_node(node, gfp | __GFP_NOWARN, 0);
+	if (unlikely(!p))
+		return -ENOMEM;
+
+	rb->dma_addr = dma_map_page(q->dma_dev, p, 0, PAGE_SIZE,
+				    DMA_FROM_DEVICE);
+	if (unlikely(dma_mapping_error(q->dma_dev, rb->dma_addr))) {
+		FUN_QSTAT_INC(q, rx_map_err);
+		__free_page(p);
+		return -ENOMEM;
+	}
+
+	FUN_QSTAT_INC(q, rx_page_alloc);
+
+	rb->page = p;
+	rb->pg_refs = 1;
+	refresh_refs(rb);
+	rb->node = page_is_pfmemalloc(p) ? -1 : page_to_nid(p);
+	return 0;
+}
+
+static void funeth_free_page(struct funeth_rxq *q, struct funeth_rxbuf *rb)
+{
+	if (rb->page) {
+		dma_unmap_page(q->dma_dev, rb->dma_addr, PAGE_SIZE,
+			       DMA_FROM_DEVICE);
+		__page_frag_cache_drain(rb->page, rb->pg_refs);
+		rb->page = NULL;
+	}
+}
+
+/* Run the XDP program assigned to an Rx queue.
+ * Return %NULL if the buffer is consumed, or the virtual address of the packet
+ * to turn into an skb.
+ */
+static void *fun_run_xdp(struct funeth_rxq *q, skb_frag_t *frags, void *buf_va,
+			 int ref_ok, struct funeth_txq *xdp_q)
+{
+	struct bpf_prog *xdp_prog;
+	struct xdp_frame *xdpf;
+	struct xdp_buff xdp;
+	u32 act;
+
+	/* VA includes the headroom, frag size includes headroom + tailroom */
+	xdp_init_buff(&xdp, ALIGN(skb_frag_size(frags), FUN_EPRQ_PKT_ALIGN),
+		      &q->xdp_rxq);
+	xdp_prepare_buff(&xdp, buf_va, FUN_XDP_HEADROOM, skb_frag_size(frags) -
+			 (FUN_RX_TAILROOM + FUN_XDP_HEADROOM), false);
+
+	xdp_prog = READ_ONCE(q->xdp_prog);
+	act = bpf_prog_run_xdp(xdp_prog, &xdp);
+
+	switch (act) {
+	case XDP_PASS:
+		/* remove headroom, which may not be FUN_XDP_HEADROOM now */
+		skb_frag_size_set(frags, xdp.data_end - xdp.data);
+		skb_frag_off_add(frags, xdp.data - xdp.data_hard_start);
+		goto pass;
+	case XDP_TX:
+		if (unlikely(!ref_ok))
+			goto pass;
+
+		xdpf = xdp_convert_buff_to_frame(&xdp);
+		if (!xdpf || !fun_xdp_tx(xdp_q, xdpf))
+			goto xdp_error;
+		FUN_QSTAT_INC(q, xdp_tx);
+		q->xdp_flush |= FUN_XDP_FLUSH_TX;
+		break;
+	case XDP_REDIRECT:
+		if (unlikely(!ref_ok))
+			goto pass;
+		if (unlikely(xdp_do_redirect(q->netdev, &xdp, xdp_prog)))
+			goto xdp_error;
+		FUN_QSTAT_INC(q, xdp_redir);
+		q->xdp_flush |= FUN_XDP_FLUSH_REDIR;
+		break;
+	default:
+		bpf_warn_invalid_xdp_action(q->netdev, xdp_prog, act);
+		fallthrough;
+	case XDP_ABORTED:
+		trace_xdp_exception(q->netdev, xdp_prog, act);
+xdp_error:
+		q->cur_buf->pg_refs++; /* return frags' page reference */
+		FUN_QSTAT_INC(q, xdp_err);
+		break;
+	case XDP_DROP:
+		q->cur_buf->pg_refs++;
+		FUN_QSTAT_INC(q, xdp_drops);
+		break;
+	}
+	return NULL;
+
+pass:
+	return xdp.data;
+}
+
+/* A CQE contains a fixed completion structure along with optional metadata and
+ * even packet data. Given the start address of a CQE return the start of the
+ * contained fixed structure, which lies at the end.
+ */
+static const void *cqe_to_info(const void *cqe)
+{
+	return cqe + FUNETH_CQE_INFO_OFFSET;
+}
+
+/* The inverse of cqe_to_info(). */
+static const void *info_to_cqe(const void *cqe_info)
+{
+	return cqe_info - FUNETH_CQE_INFO_OFFSET;
+}
+
+/* Return the type of hash provided by the device based on the L3 and L4
+ * protocols it parsed for the packet.
+ */
+static enum pkt_hash_types cqe_to_pkt_hash_type(u16 pkt_parse)
+{
+	static const enum pkt_hash_types htype_map[] = {
+		PKT_HASH_TYPE_NONE, PKT_HASH_TYPE_L3,
+		PKT_HASH_TYPE_NONE, PKT_HASH_TYPE_L4,
+		PKT_HASH_TYPE_NONE, PKT_HASH_TYPE_L3,
+		PKT_HASH_TYPE_NONE, PKT_HASH_TYPE_L3
+	};
+	u16 key;
+
+	/* Build the key from the TCP/UDP and IP/IPv6 bits */
+	key = ((pkt_parse >> FUN_ETH_RX_CV_OL4_PROT_S) & 6) |
+	      ((pkt_parse >> (FUN_ETH_RX_CV_OL3_PROT_S + 1)) & 1);
+
+	return htype_map[key];
+}
+
+/* Each received packet can be scattered across several Rx buffers or can
+ * share a buffer with previously received packets depending on the buffer
+ * and packet sizes and the room available in the most recently used buffer.
+ *
+ * The rules are:
+ * - If the buffer at the head of an RQ has not been used it gets (part of) the
+ *   next incoming packet.
+ * - Otherwise, if the packet fully fits in the buffer's remaining space the
+ *   packet is written there.
+ * - Otherwise, the packet goes into the next Rx buffer.
+ *
+ * This function returns the Rx buffer for a packet or fragment thereof of the
+ * given length. If it isn't @buf it either recycles or frees that buffer
+ * before advancing the queue to the next buffer.
+ *
+ * If called repeatedly with the remaining length of a packet it will walk
+ * through all the buffers containing the packet.
+ */
+static struct funeth_rxbuf *
+get_buf(struct funeth_rxq *q, struct funeth_rxbuf *buf, unsigned int len)
+{
+	if (q->buf_offset + len <= PAGE_SIZE || !q->buf_offset)
+		return buf;            /* @buf holds (part of) the packet */
+
+	/* The packet occupies part of the next buffer. Move there after
+	 * replenishing the current buffer slot either with the spare page or
+	 * by reusing the slot's existing page. Note that if a spare page isn't
+	 * available and the current packet occupies @buf it is a multi-frag
+	 * packet that will be dropped leaving @buf available for reuse.
+	 */
+	if ((page_ref_count(buf->page) == buf->pg_refs &&
+	     buf->node == numa_mem_id()) || !q->spare_buf.page) {
+		dma_sync_single_for_device(q->dma_dev, buf->dma_addr,
+					   PAGE_SIZE, DMA_FROM_DEVICE);
+		refresh_refs(buf);
+	} else {
+		cache_offer(q, buf);
+		*buf = q->spare_buf;
+		q->spare_buf.page = NULL;
+		q->rqes[q->rq_cons & q->rq_mask] =
+			FUN_EPRQ_RQBUF_INIT(buf->dma_addr);
+	}
+	q->buf_offset = 0;
+	q->rq_cons++;
+	return &q->bufs[q->rq_cons & q->rq_mask];
+}
+
+/* Gather the page fragments making up the first Rx packet on @q. Its total
+ * length @tot_len includes optional head- and tail-rooms.
+ *
+ * Return 0 if the device retains ownership of at least some of the pages.
+ * In this case the caller may only copy the packet.
+ *
+ * A non-zero return value gives the caller permission to use references to the
+ * pages, e.g., attach them to skbs. Additionally, if the value is <0 at least
+ * one of the pages is PF_MEMALLOC.
+ *
+ * Regardless of outcome the caller is granted a reference to each of the pages.
+ */
+static int fun_gather_pkt(struct funeth_rxq *q, unsigned int tot_len,
+			  skb_frag_t *frags)
+{
+	struct funeth_rxbuf *buf = q->cur_buf;
+	unsigned int frag_len;
+	int ref_ok = 1;
+
+	for (;;) {
+		buf = get_buf(q, buf, tot_len);
+
+		/* We always keep the RQ full of buffers so before we can give
+		 * one of our pages to the stack we require that we can obtain
+		 * a replacement page. If we can't the packet will either be
+		 * copied or dropped so we can retain ownership of the page and
+		 * reuse it.
+		 */
+		if (!q->spare_buf.page &&
+		    funeth_alloc_page(q, &q->spare_buf, numa_mem_id(),
+				      GFP_ATOMIC | __GFP_MEMALLOC))
+			ref_ok = 0;
+
+		frag_len = min_t(unsigned int, tot_len,
+				 PAGE_SIZE - q->buf_offset);
+		dma_sync_single_for_cpu(q->dma_dev,
+					buf->dma_addr + q->buf_offset,
+					frag_len, DMA_FROM_DEVICE);
+		buf->pg_refs--;
+		if (ref_ok)
+			ref_ok |= buf->node;
+
+		skb_frag_fill_page_desc(frags++, buf->page, q->buf_offset,
+					frag_len);
+
+		tot_len -= frag_len;
+		if (!tot_len)
+			break;
+
+		q->buf_offset = PAGE_SIZE;
+	}
+	q->buf_offset = ALIGN(q->buf_offset + frag_len, FUN_EPRQ_PKT_ALIGN);
+	q->cur_buf = buf;
+	return ref_ok;
+}
+
+static bool rx_hwtstamp_enabled(const struct net_device *dev)
+{
+	const struct funeth_priv *d = netdev_priv(dev);
+
+	return d->hwtstamp_cfg.rx_filter == HWTSTAMP_FILTER_ALL;
+}
+
+/* Advance the CQ pointers and phase tag to the next CQE. */
+static void advance_cq(struct funeth_rxq *q)
+{
+	if (unlikely(q->cq_head == q->cq_mask)) {
+		q->cq_head = 0;
+		q->phase ^= 1;
+		q->next_cqe_info = cqe_to_info(q->cqes);
+	} else {
+		q->cq_head++;
+		q->next_cqe_info += FUNETH_CQE_SIZE;
+	}
+	prefetch(q->next_cqe_info);
+}
+
+/* Process the packet represented by the head CQE of @q. Gather the packet's
+ * fragments, run it through the optional XDP program, and if needed construct
+ * an skb and pass it to the stack.
+ */
+static void fun_handle_cqe_pkt(struct funeth_rxq *q, struct funeth_txq *xdp_q)
+{
+	const struct fun_eth_cqe *rxreq = info_to_cqe(q->next_cqe_info);
+	unsigned int i, tot_len, pkt_len = be32_to_cpu(rxreq->pkt_len);
+	struct net_device *ndev = q->netdev;
+	skb_frag_t frags[RX_MAX_FRAGS];
+	struct skb_shared_info *si;
+	unsigned int headroom;
+	gro_result_t gro_res;
+	struct sk_buff *skb;
+	int ref_ok;
+	void *va;
+	u16 cv;
+
+	u64_stats_update_begin(&q->syncp);
+	q->stats.rx_pkts++;
+	q->stats.rx_bytes += pkt_len;
+	u64_stats_update_end(&q->syncp);
+
+	advance_cq(q);
+
+	/* account for head- and tail-room, present only for 1-buffer packets */
+	tot_len = pkt_len;
+	headroom = be16_to_cpu(rxreq->headroom);
+	if (likely(headroom))
+		tot_len += FUN_RX_TAILROOM + headroom;
+
+	ref_ok = fun_gather_pkt(q, tot_len, frags);
+	va = skb_frag_address(frags);
+	if (xdp_q && headroom == FUN_XDP_HEADROOM) {
+		va = fun_run_xdp(q, frags, va, ref_ok, xdp_q);
+		if (!va)
+			return;
+		headroom = 0;   /* XDP_PASS trims it */
+	}
+	if (unlikely(!ref_ok))
+		goto no_mem;
+
+	if (likely(headroom)) {
+		/* headroom is either FUN_RX_HEADROOM or FUN_XDP_HEADROOM */
+		prefetch(va + headroom);
+		skb = napi_build_skb(va, ALIGN(tot_len, FUN_EPRQ_PKT_ALIGN));
+		if (unlikely(!skb))
+			goto no_mem;
+
+		skb_reserve(skb, headroom);
+		__skb_put(skb, pkt_len);
+		skb->protocol = eth_type_trans(skb, ndev);
+	} else {
+		prefetch(va);
+		skb = napi_get_frags(q->napi);
+		if (unlikely(!skb))
+			goto no_mem;
+
+		if (ref_ok < 0)
+			skb->pfmemalloc = 1;
+
+		si = skb_shinfo(skb);
+		si->nr_frags = rxreq->nsgl;
+		for (i = 0; i < si->nr_frags; i++)
+			si->frags[i] = frags[i];
+
+		skb->len = pkt_len;
+		skb->data_len = pkt_len;
+		skb->truesize += round_up(pkt_len, FUN_EPRQ_PKT_ALIGN);
+	}
+
+	skb_record_rx_queue(skb, q->qidx);
+	cv = be16_to_cpu(rxreq->pkt_cv);
+	if (likely((q->netdev->features & NETIF_F_RXHASH) && rxreq->hash))
+		skb_set_hash(skb, be32_to_cpu(rxreq->hash),
+			     cqe_to_pkt_hash_type(cv));
+	if (likely((q->netdev->features & NETIF_F_RXCSUM) && rxreq->csum)) {
+		FUN_QSTAT_INC(q, rx_cso);
+		skb->ip_summed = CHECKSUM_UNNECESSARY;
+		skb->csum_level = be16_to_cpu(rxreq->csum) - 1;
+	}
+	if (unlikely(rx_hwtstamp_enabled(q->netdev)))
+		skb_hwtstamps(skb)->hwtstamp = be64_to_cpu(rxreq->timestamp);
+
+	trace_funeth_rx(q, rxreq->nsgl, pkt_len, skb->hash, cv);
+
+	gro_res = skb->data_len ? napi_gro_frags(q->napi) :
+				  napi_gro_receive(q->napi, skb);
+	if (gro_res == GRO_MERGED || gro_res == GRO_MERGED_FREE)
+		FUN_QSTAT_INC(q, gro_merged);
+	else if (gro_res == GRO_HELD)
+		FUN_QSTAT_INC(q, gro_pkts);
+	return;
+
+no_mem:
+	FUN_QSTAT_INC(q, rx_mem_drops);
+
+	/* Release the references we've been granted for the frag pages.
+	 * We return the ref of the last frag and free the rest.
+	 */
+	q->cur_buf->pg_refs++;
+	for (i = 0; i < rxreq->nsgl - 1; i++)
+		__free_page(skb_frag_page(frags + i));
+}
+
+/* Return 0 if the phase tag of the CQE at the CQ's head matches expectations
+ * indicating the CQE is new.
+ */
+static u16 cqe_phase_mismatch(const struct fun_cqe_info *ci, u16 phase)
+{
+	u16 sf_p = be16_to_cpu(ci->sf_p);
+
+	return (sf_p & 1) ^ phase;
+}
+
+/* Walk through a CQ identifying and processing fresh CQEs up to the given
+ * budget. Return the remaining budget.
+ */
+static int fun_process_cqes(struct funeth_rxq *q, int budget)
+{
+	struct funeth_priv *fp = netdev_priv(q->netdev);
+	struct funeth_txq **xdpqs, *xdp_q = NULL;
+
+	xdpqs = rcu_dereference_bh(fp->xdpqs);
+	if (xdpqs)
+		xdp_q = xdpqs[smp_processor_id()];
+
+	while (budget && !cqe_phase_mismatch(q->next_cqe_info, q->phase)) {
+		/* access other descriptor fields after the phase check */
+		dma_rmb();
+
+		fun_handle_cqe_pkt(q, xdp_q);
+		budget--;
+	}
+
+	if (unlikely(q->xdp_flush)) {
+		if (q->xdp_flush & FUN_XDP_FLUSH_TX)
+			fun_txq_wr_db(xdp_q);
+		if (q->xdp_flush & FUN_XDP_FLUSH_REDIR)
+			xdp_do_flush();
+		q->xdp_flush = 0;
+	}
+
+	return budget;
+}
+
+/* NAPI handler for Rx queues. Calls the CQE processing loop and writes RQ/CQ
+ * doorbells as needed.
+ */
+int fun_rxq_napi_poll(struct napi_struct *napi, int budget)
+{
+	struct fun_irq *irq = container_of(napi, struct fun_irq, napi);
+	struct funeth_rxq *q = irq->rxq;
+	int work_done = budget - fun_process_cqes(q, budget);
+	u32 cq_db_val = q->cq_head;
+
+	if (unlikely(work_done >= budget))
+		FUN_QSTAT_INC(q, rx_budget);
+	else if (napi_complete_done(napi, work_done))
+		cq_db_val |= q->irq_db_val;
+
+	/* check whether to post new Rx buffers */
+	if (q->rq_cons - q->rq_cons_db >= q->rq_db_thres) {
+		u64_stats_update_begin(&q->syncp);
+		q->stats.rx_bufs += q->rq_cons - q->rq_cons_db;
+		u64_stats_update_end(&q->syncp);
+		q->rq_cons_db = q->rq_cons;
+		writel((q->rq_cons - 1) & q->rq_mask, q->rq_db);
+	}
+
+	writel(cq_db_val, q->cq_db);
+	return work_done;
+}
+
+/* Free the Rx buffers of an Rx queue. */
+static void fun_rxq_free_bufs(struct funeth_rxq *q)
+{
+	struct funeth_rxbuf *b = q->bufs;
+	unsigned int i;
+
+	for (i = 0; i <= q->rq_mask; i++, b++)
+		funeth_free_page(q, b);
+
+	funeth_free_page(q, &q->spare_buf);
+	q->cur_buf = NULL;
+}
+
+/* Initially provision an Rx queue with Rx buffers. */
+static int fun_rxq_alloc_bufs(struct funeth_rxq *q, int node)
+{
+	struct funeth_rxbuf *b = q->bufs;
+	unsigned int i;
+
+	for (i = 0; i <= q->rq_mask; i++, b++) {
+		if (funeth_alloc_page(q, b, node, GFP_KERNEL)) {
+			fun_rxq_free_bufs(q);
+			return -ENOMEM;
+		}
+		q->rqes[i] = FUN_EPRQ_RQBUF_INIT(b->dma_addr);
+	}
+	q->cur_buf = q->bufs;
+	return 0;
+}
+
+/* Initialize a used-buffer cache of the given depth. */
+static int fun_rxq_init_cache(struct funeth_rx_cache *c, unsigned int depth,
+			      int node)
+{
+	c->mask = depth - 1;
+	c->bufs = kvzalloc_node(depth * sizeof(*c->bufs), GFP_KERNEL, node);
+	return c->bufs ? 0 : -ENOMEM;
+}
+
+/* Deallocate an Rx queue's used-buffer cache and its contents. */
+static void fun_rxq_free_cache(struct funeth_rxq *q)
+{
+	struct funeth_rxbuf *b = q->cache.bufs;
+	unsigned int i;
+
+	for (i = 0; i <= q->cache.mask; i++, b++)
+		funeth_free_page(q, b);
+
+	kvfree(q->cache.bufs);
+	q->cache.bufs = NULL;
+}
+
+int fun_rxq_set_bpf(struct funeth_rxq *q, struct bpf_prog *prog)
+{
+	struct funeth_priv *fp = netdev_priv(q->netdev);
+	struct fun_admin_epcq_req cmd;
+	u16 headroom;
+	int err;
+
+	headroom = prog ? FUN_XDP_HEADROOM : FUN_RX_HEADROOM;
+	if (headroom != q->headroom) {
+		cmd.common = FUN_ADMIN_REQ_COMMON_INIT2(FUN_ADMIN_OP_EPCQ,
+							sizeof(cmd));
+		cmd.u.modify =
+			FUN_ADMIN_EPCQ_MODIFY_REQ_INIT(FUN_ADMIN_SUBOP_MODIFY,
+						       0, q->hw_cqid, headroom);
+		err = fun_submit_admin_sync_cmd(fp->fdev, &cmd.common, NULL, 0,
+						0);
+		if (err)
+			return err;
+		q->headroom = headroom;
+	}
+
+	WRITE_ONCE(q->xdp_prog, prog);
+	return 0;
+}
+
+/* Create an Rx queue, allocating the host memory it needs. */
+static struct funeth_rxq *fun_rxq_create_sw(struct net_device *dev,
+					    unsigned int qidx,
+					    unsigned int ncqe,
+					    unsigned int nrqe,
+					    struct fun_irq *irq)
+{
+	struct funeth_priv *fp = netdev_priv(dev);
+	struct funeth_rxq *q;
+	int err = -ENOMEM;
+	int numa_node;
+
+	numa_node = fun_irq_node(irq);
+	q = kzalloc_node(sizeof(*q), GFP_KERNEL, numa_node);
+	if (!q)
+		goto err;
+
+	q->qidx = qidx;
+	q->netdev = dev;
+	q->cq_mask = ncqe - 1;
+	q->rq_mask = nrqe - 1;
+	q->numa_node = numa_node;
+	q->rq_db_thres = nrqe / 4;
+	u64_stats_init(&q->syncp);
+	q->dma_dev = &fp->pdev->dev;
+
+	q->rqes = fun_alloc_ring_mem(q->dma_dev, nrqe, sizeof(*q->rqes),
+				     sizeof(*q->bufs), false, numa_node,
+				     &q->rq_dma_addr, (void **)&q->bufs, NULL);
+	if (!q->rqes)
+		goto free_q;
+
+	q->cqes = fun_alloc_ring_mem(q->dma_dev, ncqe, FUNETH_CQE_SIZE, 0,
+				     false, numa_node, &q->cq_dma_addr, NULL,
+				     NULL);
+	if (!q->cqes)
+		goto free_rqes;
+
+	err = fun_rxq_init_cache(&q->cache, nrqe, numa_node);
+	if (err)
+		goto free_cqes;
+
+	err = fun_rxq_alloc_bufs(q, numa_node);
+	if (err)
+		goto free_cache;
+
+	q->stats.rx_bufs = q->rq_mask;
+	q->init_state = FUN_QSTATE_INIT_SW;
+	return q;
+
+free_cache:
+	fun_rxq_free_cache(q);
+free_cqes:
+	dma_free_coherent(q->dma_dev, ncqe * FUNETH_CQE_SIZE, q->cqes,
+			  q->cq_dma_addr);
+free_rqes:
+	fun_free_ring_mem(q->dma_dev, nrqe, sizeof(*q->rqes), false, q->rqes,
+			  q->rq_dma_addr, q->bufs);
+free_q:
+	kfree(q);
+err:
+	netdev_err(dev, "Unable to allocate memory for Rx queue %u\n", qidx);
+	return ERR_PTR(err);
+}
+
+static void fun_rxq_free_sw(struct funeth_rxq *q)
+{
+	struct funeth_priv *fp = netdev_priv(q->netdev);
+
+	fun_rxq_free_cache(q);
+	fun_rxq_free_bufs(q);
+	fun_free_ring_mem(q->dma_dev, q->rq_mask + 1, sizeof(*q->rqes), false,
+			  q->rqes, q->rq_dma_addr, q->bufs);
+	dma_free_coherent(q->dma_dev, (q->cq_mask + 1) * FUNETH_CQE_SIZE,
+			  q->cqes, q->cq_dma_addr);
+
+	/* Before freeing the queue transfer key counters to the device. */
+	fp->rx_packets += q->stats.rx_pkts;
+	fp->rx_bytes   += q->stats.rx_bytes;
+	fp->rx_dropped += q->stats.rx_map_err + q->stats.rx_mem_drops;
+
+	kfree(q);
+}
+
+/* Create an Rx queue's resources on the device. */
+int fun_rxq_create_dev(struct funeth_rxq *q, struct fun_irq *irq)
+{
+	struct funeth_priv *fp = netdev_priv(q->netdev);
+	unsigned int ncqe = q->cq_mask + 1;
+	unsigned int nrqe = q->rq_mask + 1;
+	int err;
+
+	err = xdp_rxq_info_reg(&q->xdp_rxq, q->netdev, q->qidx,
+			       irq->napi.napi_id);
+	if (err)
+		goto out;
+
+	err = xdp_rxq_info_reg_mem_model(&q->xdp_rxq, MEM_TYPE_PAGE_SHARED,
+					 NULL);
+	if (err)
+		goto xdp_unreg;
+
+	q->phase = 1;
+	q->irq_cnt = 0;
+	q->cq_head = 0;
+	q->rq_cons = 0;
+	q->rq_cons_db = 0;
+	q->buf_offset = 0;
+	q->napi = &irq->napi;
+	q->irq_db_val = fp->cq_irq_db;
+	q->next_cqe_info = cqe_to_info(q->cqes);
+
+	q->xdp_prog = fp->xdp_prog;
+	q->headroom = fp->xdp_prog ? FUN_XDP_HEADROOM : FUN_RX_HEADROOM;
+
+	err = fun_sq_create(fp->fdev, FUN_ADMIN_RES_CREATE_FLAG_ALLOCATOR |
+			    FUN_ADMIN_EPSQ_CREATE_FLAG_RQ, 0,
+			    FUN_HCI_ID_INVALID, 0, nrqe, q->rq_dma_addr, 0, 0,
+			    0, 0, fp->fdev->kern_end_qid, PAGE_SHIFT,
+			    &q->hw_sqid, &q->rq_db);
+	if (err)
+		goto xdp_unreg;
+
+	err = fun_cq_create(fp->fdev, FUN_ADMIN_RES_CREATE_FLAG_ALLOCATOR |
+			    FUN_ADMIN_EPCQ_CREATE_FLAG_RQ, 0,
+			    q->hw_sqid, ilog2(FUNETH_CQE_SIZE), ncqe,
+			    q->cq_dma_addr, q->headroom, FUN_RX_TAILROOM, 0, 0,
+			    irq->irq_idx, 0, fp->fdev->kern_end_qid,
+			    &q->hw_cqid, &q->cq_db);
+	if (err)
+		goto free_rq;
+
+	irq->rxq = q;
+	writel(q->rq_mask, q->rq_db);
+	q->init_state = FUN_QSTATE_INIT_FULL;
+
+	netif_info(fp, ifup, q->netdev,
+		   "Rx queue %u, depth %u/%u, HW qid %u/%u, IRQ idx %u, node %d, headroom %u\n",
+		   q->qidx, ncqe, nrqe, q->hw_cqid, q->hw_sqid, irq->irq_idx,
+		   q->numa_node, q->headroom);
+	return 0;
+
+free_rq:
+	fun_destroy_sq(fp->fdev, q->hw_sqid);
+xdp_unreg:
+	xdp_rxq_info_unreg(&q->xdp_rxq);
+out:
+	netdev_err(q->netdev,
+		   "Failed to create Rx queue %u on device, error %d\n",
+		   q->qidx, err);
+	return err;
+}
+
+static void fun_rxq_free_dev(struct funeth_rxq *q)
+{
+	struct funeth_priv *fp = netdev_priv(q->netdev);
+	struct fun_irq *irq;
+
+	if (q->init_state < FUN_QSTATE_INIT_FULL)
+		return;
+
+	irq = container_of(q->napi, struct fun_irq, napi);
+	netif_info(fp, ifdown, q->netdev,
+		   "Freeing Rx queue %u (id %u/%u), IRQ %u\n",
+		   q->qidx, q->hw_cqid, q->hw_sqid, irq->irq_idx);
+
+	irq->rxq = NULL;
+	xdp_rxq_info_unreg(&q->xdp_rxq);
+	fun_destroy_sq(fp->fdev, q->hw_sqid);
+	fun_destroy_cq(fp->fdev, q->hw_cqid);
+	q->init_state = FUN_QSTATE_INIT_SW;
+}
+
+/* Create or advance an Rx queue, allocating all the host and device resources
+ * needed to reach the target state.
+ */
+int funeth_rxq_create(struct net_device *dev, unsigned int qidx,
+		      unsigned int ncqe, unsigned int nrqe, struct fun_irq *irq,
+		      int state, struct funeth_rxq **qp)
+{
+	struct funeth_rxq *q = *qp;
+	int err;
+
+	if (!q) {
+		q = fun_rxq_create_sw(dev, qidx, ncqe, nrqe, irq);
+		if (IS_ERR(q))
+			return PTR_ERR(q);
+	}
+
+	if (q->init_state >= state)
+		goto out;
+
+	err = fun_rxq_create_dev(q, irq);
+	if (err) {
+		if (!*qp)
+			fun_rxq_free_sw(q);
+		return err;
+	}
+
+out:
+	*qp = q;
+	return 0;
+}
+
+/* Free Rx queue resources until it reaches the target state. */
+struct funeth_rxq *funeth_rxq_free(struct funeth_rxq *q, int state)
+{
+	if (state < FUN_QSTATE_INIT_FULL)
+		fun_rxq_free_dev(q);
+
+	if (state == FUN_QSTATE_DESTROYED) {
+		fun_rxq_free_sw(q);
+		q = NULL;
+	}
+
+	return q;
+}
diff --git a/drivers/net/ethernet/fungible/funeth/funeth_trace.h b/drivers/net/ethernet/fungible/funeth/funeth_trace.h
new file mode 100644
index 000000000000..b9985900f30b
--- /dev/null
+++ b/drivers/net/ethernet/fungible/funeth/funeth_trace.h
@@ -0,0 +1,117 @@
+/* SPDX-License-Identifier: (GPL-2.0-only OR BSD-3-Clause) */
+
+#undef TRACE_SYSTEM
+#define TRACE_SYSTEM funeth
+
+#if !defined(_TRACE_FUNETH_H) || defined(TRACE_HEADER_MULTI_READ)
+#define _TRACE_FUNETH_H
+
+#include <linux/tracepoint.h>
+
+#include "funeth_txrx.h"
+
+TRACE_EVENT(funeth_tx,
+
+	TP_PROTO(const struct funeth_txq *txq,
+		 u32 len,
+		 u32 sqe_idx,
+		 u32 ngle),
+
+	TP_ARGS(txq, len, sqe_idx, ngle),
+
+	TP_STRUCT__entry(
+		__field(u32, qidx)
+		__field(u32, len)
+		__field(u32, sqe_idx)
+		__field(u32, ngle)
+		__string(devname, txq->netdev->name)
+	),
+
+	TP_fast_assign(
+		__entry->qidx = txq->qidx;
+		__entry->len = len;
+		__entry->sqe_idx = sqe_idx;
+		__entry->ngle = ngle;
+		__assign_str(devname);
+	),
+
+	TP_printk("%s: Txq %u, SQE idx %u, len %u, num GLEs %u",
+		  __get_str(devname), __entry->qidx, __entry->sqe_idx,
+		  __entry->len, __entry->ngle)
+);
+
+TRACE_EVENT(funeth_tx_free,
+
+	TP_PROTO(const struct funeth_txq *txq,
+		 u32 sqe_idx,
+		 u32 num_sqes,
+		 u32 hw_head),
+
+	TP_ARGS(txq, sqe_idx, num_sqes, hw_head),
+
+	TP_STRUCT__entry(
+		__field(u32, qidx)
+		__field(u32, sqe_idx)
+		__field(u32, num_sqes)
+		__field(u32, hw_head)
+		__string(devname, txq->netdev->name)
+	),
+
+	TP_fast_assign(
+		__entry->qidx = txq->qidx;
+		__entry->sqe_idx = sqe_idx;
+		__entry->num_sqes = num_sqes;
+		__entry->hw_head = hw_head;
+		__assign_str(devname);
+	),
+
+	TP_printk("%s: Txq %u, SQE idx %u, SQEs %u, HW head %u",
+		  __get_str(devname), __entry->qidx, __entry->sqe_idx,
+		  __entry->num_sqes, __entry->hw_head)
+);
+
+TRACE_EVENT(funeth_rx,
+
+	TP_PROTO(const struct funeth_rxq *rxq,
+		 u32 num_rqes,
+		 u32 pkt_len,
+		 u32 hash,
+		 u32 cls_vec),
+
+	TP_ARGS(rxq, num_rqes, pkt_len, hash, cls_vec),
+
+	TP_STRUCT__entry(
+		__field(u32, qidx)
+		__field(u32, cq_head)
+		__field(u32, num_rqes)
+		__field(u32, len)
+		__field(u32, hash)
+		__field(u32, cls_vec)
+		__string(devname, rxq->netdev->name)
+	),
+
+	TP_fast_assign(
+		__entry->qidx = rxq->qidx;
+		__entry->cq_head = rxq->cq_head;
+		__entry->num_rqes = num_rqes;
+		__entry->len = pkt_len;
+		__entry->hash = hash;
+		__entry->cls_vec = cls_vec;
+		__assign_str(devname);
+	),
+
+	TP_printk("%s: Rxq %u, CQ head %u, RQEs %u, len %u, hash %u, CV %#x",
+		  __get_str(devname), __entry->qidx, __entry->cq_head,
+		  __entry->num_rqes, __entry->len, __entry->hash,
+		  __entry->cls_vec)
+);
+
+#endif /* _TRACE_FUNETH_H */
+
+/* Below must be outside protection. */
+#undef TRACE_INCLUDE_PATH
+#define TRACE_INCLUDE_PATH .
+#undef TRACE_INCLUDE_FILE
+#define TRACE_INCLUDE_FILE funeth_trace
+
+#include <trace/define_trace.h>
diff --git a/drivers/net/ethernet/fungible/funeth/funeth_tx.c b/drivers/net/ethernet/fungible/funeth/funeth_tx.c
new file mode 100644
index 000000000000..8ddefd3ec15b
--- /dev/null
+++ b/drivers/net/ethernet/fungible/funeth/funeth_tx.c
@@ -0,0 +1,800 @@
+// SPDX-License-Identifier: (GPL-2.0-only OR BSD-3-Clause)
+
+#include <linux/dma-mapping.h>
+#include <linux/ip.h>
+#include <linux/pci.h>
+#include <linux/skbuff.h>
+#include <linux/tcp.h>
+#include <uapi/linux/udp.h>
+#include "funeth.h"
+#include "funeth_ktls.h"
+#include "funeth_txrx.h"
+#include "funeth_trace.h"
+#include "fun_queue.h"
+
+#define FUN_XDP_CLEAN_THRES 32
+#define FUN_XDP_CLEAN_BATCH 16
+
+/* DMA-map a packet and return the (length, DMA_address) pairs for its
+ * segments. If a mapping error occurs -ENOMEM is returned. The packet
+ * consists of an skb_shared_info and one additional address/length pair.
+ */
+static int fun_map_pkt(struct device *dev, const struct skb_shared_info *si,
+		       void *data, unsigned int data_len,
+		       dma_addr_t *addr, unsigned int *len)
+{
+	const skb_frag_t *fp, *end;
+
+	*len = data_len;
+	*addr = dma_map_single(dev, data, *len, DMA_TO_DEVICE);
+	if (dma_mapping_error(dev, *addr))
+		return -ENOMEM;
+
+	if (!si)
+		return 0;
+
+	for (fp = si->frags, end = fp + si->nr_frags; fp < end; fp++) {
+		*++len = skb_frag_size(fp);
+		*++addr = skb_frag_dma_map(dev, fp, 0, *len, DMA_TO_DEVICE);
+		if (dma_mapping_error(dev, *addr))
+			goto unwind;
+	}
+	return 0;
+
+unwind:
+	while (fp-- > si->frags)
+		dma_unmap_page(dev, *--addr, skb_frag_size(fp), DMA_TO_DEVICE);
+
+	dma_unmap_single(dev, addr[-1], data_len, DMA_TO_DEVICE);
+	return -ENOMEM;
+}
+
+/* Return the address just past the end of a Tx queue's descriptor ring.
+ * It exploits the fact that the HW writeback area is just after the end
+ * of the descriptor ring.
+ */
+static void *txq_end(const struct funeth_txq *q)
+{
+	return (void *)q->hw_wb;
+}
+
+/* Return the amount of space within a Tx ring from the given address to the
+ * end.
+ */
+static unsigned int txq_to_end(const struct funeth_txq *q, void *p)
+{
+	return txq_end(q) - p;
+}
+
+/* Return the number of Tx descriptors occupied by a Tx request. */
+static unsigned int tx_req_ndesc(const struct fun_eth_tx_req *req)
+{
+	return DIV_ROUND_UP(req->len8, FUNETH_SQE_SIZE / 8);
+}
+
+/* Write a gather list to the Tx descriptor at @req from @ngle address/length
+ * pairs.
+ */
+static struct fun_dataop_gl *fun_write_gl(const struct funeth_txq *q,
+					  struct fun_eth_tx_req *req,
+					  const dma_addr_t *addrs,
+					  const unsigned int *lens,
+					  unsigned int ngle)
+{
+	struct fun_dataop_gl *gle;
+	unsigned int i;
+
+	req->len8 = (sizeof(*req) + ngle * sizeof(*gle)) / 8;
+
+	for (i = 0, gle = (struct fun_dataop_gl *)req->dataop.imm;
+	     i < ngle && txq_to_end(q, gle); i++, gle++)
+		fun_dataop_gl_init(gle, 0, 0, lens[i], addrs[i]);
+
+	if (txq_to_end(q, gle) == 0) {
+		gle = (struct fun_dataop_gl *)q->desc;
+		for ( ; i < ngle; i++, gle++)
+			fun_dataop_gl_init(gle, 0, 0, lens[i], addrs[i]);
+	}
+
+	return gle;
+}
+
+static __be16 tcp_hdr_doff_flags(const struct tcphdr *th)
+{
+	return *(__be16 *)&tcp_flag_word(th);
+}
+
+static struct sk_buff *fun_tls_tx(struct sk_buff *skb, struct funeth_txq *q,
+				  unsigned int *tls_len)
+{
+#if IS_ENABLED(CONFIG_TLS_DEVICE)
+	const struct fun_ktls_tx_ctx *tls_ctx;
+	u32 datalen, seq;
+
+	datalen = skb->len - skb_tcp_all_headers(skb);
+	if (!datalen)
+		return skb;
+
+	if (likely(!tls_offload_tx_resync_pending(skb->sk))) {
+		seq = ntohl(tcp_hdr(skb)->seq);
+		tls_ctx = tls_driver_ctx(skb->sk, TLS_OFFLOAD_CTX_DIR_TX);
+
+		if (likely(tls_ctx->next_seq == seq)) {
+			*tls_len = datalen;
+			return skb;
+		}
+		if (seq - tls_ctx->next_seq < U32_MAX / 4) {
+			tls_offload_tx_resync_request(skb->sk, seq,
+						      tls_ctx->next_seq);
+		}
+	}
+
+	FUN_QSTAT_INC(q, tx_tls_fallback);
+	skb = tls_encrypt_skb(skb);
+	if (!skb)
+		FUN_QSTAT_INC(q, tx_tls_drops);
+
+	return skb;
+#else
+	return NULL;
+#endif
+}
+
+/* Write as many descriptors as needed for the supplied skb starting at the
+ * current producer location. The caller has made certain enough descriptors
+ * are available.
+ *
+ * Returns the number of descriptors written, 0 on error.
+ */
+static unsigned int write_pkt_desc(struct sk_buff *skb, struct funeth_txq *q,
+				   unsigned int tls_len)
+{
+	unsigned int extra_bytes = 0, extra_pkts = 0;
+	unsigned int idx = q->prod_cnt & q->mask;
+	const struct skb_shared_info *shinfo;
+	unsigned int lens[MAX_SKB_FRAGS + 1];
+	dma_addr_t addrs[MAX_SKB_FRAGS + 1];
+	struct fun_eth_tx_req *req;
+	struct fun_dataop_gl *gle;
+	const struct tcphdr *th;
+	unsigned int l4_hlen;
+	unsigned int ngle;
+	u16 flags;
+
+	shinfo = skb_shinfo(skb);
+	if (unlikely(fun_map_pkt(q->dma_dev, shinfo, skb->data,
+				 skb_headlen(skb), addrs, lens))) {
+		FUN_QSTAT_INC(q, tx_map_err);
+		return 0;
+	}
+
+	req = fun_tx_desc_addr(q, idx);
+	req->op = FUN_ETH_OP_TX;
+	req->len8 = 0;
+	req->flags = 0;
+	req->suboff8 = offsetof(struct fun_eth_tx_req, dataop);
+	req->repr_idn = 0;
+	req->encap_proto = 0;
+
+	if (likely(shinfo->gso_size)) {
+		if (skb->encapsulation) {
+			u16 ol4_ofst;
+
+			flags = FUN_ETH_OUTER_EN | FUN_ETH_INNER_LSO |
+				FUN_ETH_UPDATE_INNER_L4_CKSUM |
+				FUN_ETH_UPDATE_OUTER_L3_LEN;
+			if (shinfo->gso_type & (SKB_GSO_UDP_TUNNEL |
+						SKB_GSO_UDP_TUNNEL_CSUM)) {
+				flags |= FUN_ETH_UPDATE_OUTER_L4_LEN |
+					 FUN_ETH_OUTER_UDP;
+				if (shinfo->gso_type & SKB_GSO_UDP_TUNNEL_CSUM)
+					flags |= FUN_ETH_UPDATE_OUTER_L4_CKSUM;
+				ol4_ofst = skb_transport_offset(skb);
+			} else {
+				ol4_ofst = skb_inner_network_offset(skb);
+			}
+
+			if (ip_hdr(skb)->version == 4)
+				flags |= FUN_ETH_UPDATE_OUTER_L3_CKSUM;
+			else
+				flags |= FUN_ETH_OUTER_IPV6;
+
+			if (skb->inner_network_header) {
+				if (inner_ip_hdr(skb)->version == 4)
+					flags |= FUN_ETH_UPDATE_INNER_L3_CKSUM |
+						 FUN_ETH_UPDATE_INNER_L3_LEN;
+				else
+					flags |= FUN_ETH_INNER_IPV6 |
+						 FUN_ETH_UPDATE_INNER_L3_LEN;
+			}
+			th = inner_tcp_hdr(skb);
+			l4_hlen = __tcp_hdrlen(th);
+			fun_eth_offload_init(&req->offload, flags,
+					     shinfo->gso_size,
+					     tcp_hdr_doff_flags(th), 0,
+					     skb_inner_network_offset(skb),
+					     skb_inner_transport_offset(skb),
+					     skb_network_offset(skb), ol4_ofst);
+			FUN_QSTAT_INC(q, tx_encap_tso);
+		} else if (shinfo->gso_type & SKB_GSO_UDP_L4) {
+			flags = FUN_ETH_INNER_LSO | FUN_ETH_INNER_UDP |
+				FUN_ETH_UPDATE_INNER_L4_CKSUM |
+				FUN_ETH_UPDATE_INNER_L4_LEN |
+				FUN_ETH_UPDATE_INNER_L3_LEN;
+
+			if (ip_hdr(skb)->version == 4)
+				flags |= FUN_ETH_UPDATE_INNER_L3_CKSUM;
+			else
+				flags |= FUN_ETH_INNER_IPV6;
+
+			l4_hlen = sizeof(struct udphdr);
+			fun_eth_offload_init(&req->offload, flags,
+					     shinfo->gso_size,
+					     cpu_to_be16(l4_hlen << 10), 0,
+					     skb_network_offset(skb),
+					     skb_transport_offset(skb), 0, 0);
+			FUN_QSTAT_INC(q, tx_uso);
+		} else {
+			/* HW considers one set of headers as inner */
+			flags = FUN_ETH_INNER_LSO |
+				FUN_ETH_UPDATE_INNER_L4_CKSUM |
+				FUN_ETH_UPDATE_INNER_L3_LEN;
+			if (shinfo->gso_type & SKB_GSO_TCPV6)
+				flags |= FUN_ETH_INNER_IPV6;
+			else
+				flags |= FUN_ETH_UPDATE_INNER_L3_CKSUM;
+			th = tcp_hdr(skb);
+			l4_hlen = __tcp_hdrlen(th);
+			fun_eth_offload_init(&req->offload, flags,
+					     shinfo->gso_size,
+					     tcp_hdr_doff_flags(th), 0,
+					     skb_network_offset(skb),
+					     skb_transport_offset(skb), 0, 0);
+			FUN_QSTAT_INC(q, tx_tso);
+		}
+
+		u64_stats_update_begin(&q->syncp);
+		q->stats.tx_cso += shinfo->gso_segs;
+		u64_stats_update_end(&q->syncp);
+
+		extra_pkts = shinfo->gso_segs - 1;
+		extra_bytes = (be16_to_cpu(req->offload.inner_l4_off) +
+			       l4_hlen) * extra_pkts;
+	} else if (likely(skb->ip_summed == CHECKSUM_PARTIAL)) {
+		flags = FUN_ETH_UPDATE_INNER_L4_CKSUM;
+		if (skb->csum_offset == offsetof(struct udphdr, check))
+			flags |= FUN_ETH_INNER_UDP;
+		fun_eth_offload_init(&req->offload, flags, 0, 0, 0, 0,
+				     skb_checksum_start_offset(skb), 0, 0);
+		FUN_QSTAT_INC(q, tx_cso);
+	} else {
+		fun_eth_offload_init(&req->offload, 0, 0, 0, 0, 0, 0, 0, 0);
+	}
+
+	ngle = shinfo->nr_frags + 1;
+	req->dataop = FUN_DATAOP_HDR_INIT(ngle, 0, ngle, 0, skb->len);
+
+	gle = fun_write_gl(q, req, addrs, lens, ngle);
+
+	if (IS_ENABLED(CONFIG_TLS_DEVICE) && unlikely(tls_len)) {
+		struct fun_eth_tls *tls = (struct fun_eth_tls *)gle;
+		struct fun_ktls_tx_ctx *tls_ctx;
+
+		req->len8 += FUNETH_TLS_SZ / 8;
+		req->flags = cpu_to_be16(FUN_ETH_TX_TLS);
+
+		tls_ctx = tls_driver_ctx(skb->sk, TLS_OFFLOAD_CTX_DIR_TX);
+		tls->tlsid = tls_ctx->tlsid;
+		tls_ctx->next_seq += tls_len;
+
+		u64_stats_update_begin(&q->syncp);
+		q->stats.tx_tls_bytes += tls_len;
+		q->stats.tx_tls_pkts += 1 + extra_pkts;
+		u64_stats_update_end(&q->syncp);
+	}
+
+	u64_stats_update_begin(&q->syncp);
+	q->stats.tx_bytes += skb->len + extra_bytes;
+	q->stats.tx_pkts += 1 + extra_pkts;
+	u64_stats_update_end(&q->syncp);
+
+	q->info[idx].skb = skb;
+
+	trace_funeth_tx(q, skb->len, idx, req->dataop.ngather);
+	return tx_req_ndesc(req);
+}
+
+/* Return the number of available descriptors of a Tx queue.
+ * HW assumes head==tail means the ring is empty so we need to keep one
+ * descriptor unused.
+ */
+static unsigned int fun_txq_avail(const struct funeth_txq *q)
+{
+	return q->mask - q->prod_cnt + q->cons_cnt;
+}
+
+/* Stop a queue if it can't handle another worst-case packet. */
+static void fun_tx_check_stop(struct funeth_txq *q)
+{
+	if (likely(fun_txq_avail(q) >= FUNETH_MAX_PKT_DESC))
+		return;
+
+	netif_tx_stop_queue(q->ndq);
+
+	/* NAPI reclaim is freeing packets in parallel with us and we may race.
+	 * We have stopped the queue but check again after synchronizing with
+	 * reclaim.
+	 */
+	smp_mb();
+	if (likely(fun_txq_avail(q) < FUNETH_MAX_PKT_DESC))
+		FUN_QSTAT_INC(q, tx_nstops);
+	else
+		netif_tx_start_queue(q->ndq);
+}
+
+/* Return true if a queue has enough space to restart. Current condition is
+ * that the queue must be >= 1/4 empty.
+ */
+static bool fun_txq_may_restart(struct funeth_txq *q)
+{
+	return fun_txq_avail(q) >= q->mask / 4;
+}
+
+netdev_tx_t fun_start_xmit(struct sk_buff *skb, struct net_device *netdev)
+{
+	struct funeth_priv *fp = netdev_priv(netdev);
+	unsigned int qid = skb_get_queue_mapping(skb);
+	struct funeth_txq *q = fp->txqs[qid];
+	unsigned int tls_len = 0;
+	unsigned int ndesc;
+
+	if (tls_is_skb_tx_device_offloaded(skb)) {
+		skb = fun_tls_tx(skb, q, &tls_len);
+		if (unlikely(!skb))
+			goto dropped;
+	}
+
+	ndesc = write_pkt_desc(skb, q, tls_len);
+	if (unlikely(!ndesc)) {
+		dev_kfree_skb_any(skb);
+		goto dropped;
+	}
+
+	q->prod_cnt += ndesc;
+	fun_tx_check_stop(q);
+
+	skb_tx_timestamp(skb);
+
+	if (__netdev_tx_sent_queue(q->ndq, skb->len, netdev_xmit_more()))
+		fun_txq_wr_db(q);
+	else
+		FUN_QSTAT_INC(q, tx_more);
+
+	return NETDEV_TX_OK;
+
+dropped:
+	/* A dropped packet may be the last one in a xmit_more train,
+	 * ring the doorbell just in case.
+	 */
+	if (!netdev_xmit_more())
+		fun_txq_wr_db(q);
+	return NETDEV_TX_OK;
+}
+
+/* Return a Tx queue's HW head index written back to host memory. */
+static u16 txq_hw_head(const struct funeth_txq *q)
+{
+	return (u16)be64_to_cpu(*q->hw_wb);
+}
+
+/* Unmap the Tx packet starting at the given descriptor index and
+ * return the number of Tx descriptors it occupied.
+ */
+static unsigned int fun_unmap_pkt(const struct funeth_txq *q, unsigned int idx)
+{
+	const struct fun_eth_tx_req *req = fun_tx_desc_addr(q, idx);
+	unsigned int ngle = req->dataop.ngather;
+	struct fun_dataop_gl *gle;
+
+	if (ngle) {
+		gle = (struct fun_dataop_gl *)req->dataop.imm;
+		dma_unmap_single(q->dma_dev, be64_to_cpu(gle->sgl_data),
+				 be32_to_cpu(gle->sgl_len), DMA_TO_DEVICE);
+
+		for (gle++; --ngle && txq_to_end(q, gle); gle++)
+			dma_unmap_page(q->dma_dev, be64_to_cpu(gle->sgl_data),
+				       be32_to_cpu(gle->sgl_len),
+				       DMA_TO_DEVICE);
+
+		for (gle = (struct fun_dataop_gl *)q->desc; ngle; ngle--, gle++)
+			dma_unmap_page(q->dma_dev, be64_to_cpu(gle->sgl_data),
+				       be32_to_cpu(gle->sgl_len),
+				       DMA_TO_DEVICE);
+	}
+
+	return tx_req_ndesc(req);
+}
+
+/* Reclaim completed Tx descriptors and free their packets. Restart a stopped
+ * queue if we freed enough descriptors.
+ *
+ * Return true if we exhausted the budget while there is more work to be done.
+ */
+static bool fun_txq_reclaim(struct funeth_txq *q, int budget)
+{
+	unsigned int npkts = 0, nbytes = 0, ndesc = 0;
+	unsigned int head, limit, reclaim_idx;
+
+	/* budget may be 0, e.g., netpoll */
+	limit = budget ? budget : UINT_MAX;
+
+	for (head = txq_hw_head(q), reclaim_idx = q->cons_cnt & q->mask;
+	     head != reclaim_idx && npkts < limit; head = txq_hw_head(q)) {
+		/* The HW head is continually updated, ensure we don't read
+		 * descriptor state before the head tells us to reclaim it.
+		 * On the enqueue side the doorbell is an implicit write
+		 * barrier.
+		 */
+		rmb();
+
+		do {
+			unsigned int pkt_desc = fun_unmap_pkt(q, reclaim_idx);
+			struct sk_buff *skb = q->info[reclaim_idx].skb;
+
+			trace_funeth_tx_free(q, reclaim_idx, pkt_desc, head);
+
+			nbytes += skb->len;
+			napi_consume_skb(skb, budget);
+			ndesc += pkt_desc;
+			reclaim_idx = (reclaim_idx + pkt_desc) & q->mask;
+			npkts++;
+		} while (reclaim_idx != head && npkts < limit);
+	}
+
+	q->cons_cnt += ndesc;
+	netdev_tx_completed_queue(q->ndq, npkts, nbytes);
+	smp_mb(); /* pairs with the one in fun_tx_check_stop() */
+
+	if (unlikely(netif_tx_queue_stopped(q->ndq) &&
+		     fun_txq_may_restart(q))) {
+		netif_tx_wake_queue(q->ndq);
+		FUN_QSTAT_INC(q, tx_nrestarts);
+	}
+
+	return reclaim_idx != head;
+}
+
+/* The NAPI handler for Tx queues. */
+int fun_txq_napi_poll(struct napi_struct *napi, int budget)
+{
+	struct fun_irq *irq = container_of(napi, struct fun_irq, napi);
+	struct funeth_txq *q = irq->txq;
+	unsigned int db_val;
+
+	if (fun_txq_reclaim(q, budget))
+		return budget;               /* exhausted budget */
+
+	napi_complete(napi);                 /* exhausted pending work */
+	db_val = READ_ONCE(q->irq_db_val) | (q->cons_cnt & q->mask);
+	writel(db_val, q->db);
+	return 0;
+}
+
+/* Reclaim up to @budget completed Tx packets from a TX XDP queue. */
+static unsigned int fun_xdpq_clean(struct funeth_txq *q, unsigned int budget)
+{
+	unsigned int npkts = 0, ndesc = 0, head, reclaim_idx;
+
+	for (head = txq_hw_head(q), reclaim_idx = q->cons_cnt & q->mask;
+	     head != reclaim_idx && npkts < budget; head = txq_hw_head(q)) {
+		/* The HW head is continually updated, ensure we don't read
+		 * descriptor state before the head tells us to reclaim it.
+		 * On the enqueue side the doorbell is an implicit write
+		 * barrier.
+		 */
+		rmb();
+
+		do {
+			unsigned int pkt_desc = fun_unmap_pkt(q, reclaim_idx);
+
+			xdp_return_frame(q->info[reclaim_idx].xdpf);
+
+			trace_funeth_tx_free(q, reclaim_idx, pkt_desc, head);
+
+			reclaim_idx = (reclaim_idx + pkt_desc) & q->mask;
+			ndesc += pkt_desc;
+			npkts++;
+		} while (reclaim_idx != head && npkts < budget);
+	}
+
+	q->cons_cnt += ndesc;
+	return npkts;
+}
+
+bool fun_xdp_tx(struct funeth_txq *q, struct xdp_frame *xdpf)
+{
+	unsigned int idx, nfrags = 1, ndesc = 1, tot_len = xdpf->len;
+	const struct skb_shared_info *si = NULL;
+	unsigned int lens[MAX_SKB_FRAGS + 1];
+	dma_addr_t dma[MAX_SKB_FRAGS + 1];
+	struct fun_eth_tx_req *req;
+
+	if (fun_txq_avail(q) < FUN_XDP_CLEAN_THRES)
+		fun_xdpq_clean(q, FUN_XDP_CLEAN_BATCH);
+
+	if (unlikely(xdp_frame_has_frags(xdpf))) {
+		si = xdp_get_shared_info_from_frame(xdpf);
+		tot_len = xdp_get_frame_len(xdpf);
+		nfrags += si->nr_frags;
+		ndesc = DIV_ROUND_UP((sizeof(*req) + nfrags *
+				      sizeof(struct fun_dataop_gl)),
+				     FUNETH_SQE_SIZE);
+	}
+
+	if (unlikely(fun_txq_avail(q) < ndesc)) {
+		FUN_QSTAT_INC(q, tx_xdp_full);
+		return false;
+	}
+
+	if (unlikely(fun_map_pkt(q->dma_dev, si, xdpf->data, xdpf->len, dma,
+				 lens))) {
+		FUN_QSTAT_INC(q, tx_map_err);
+		return false;
+	}
+
+	idx = q->prod_cnt & q->mask;
+	req = fun_tx_desc_addr(q, idx);
+	req->op = FUN_ETH_OP_TX;
+	req->len8 = 0;
+	req->flags = 0;
+	req->suboff8 = offsetof(struct fun_eth_tx_req, dataop);
+	req->repr_idn = 0;
+	req->encap_proto = 0;
+	fun_eth_offload_init(&req->offload, 0, 0, 0, 0, 0, 0, 0, 0);
+	req->dataop = FUN_DATAOP_HDR_INIT(nfrags, 0, nfrags, 0, tot_len);
+
+	fun_write_gl(q, req, dma, lens, nfrags);
+
+	q->info[idx].xdpf = xdpf;
+
+	u64_stats_update_begin(&q->syncp);
+	q->stats.tx_bytes += tot_len;
+	q->stats.tx_pkts++;
+	u64_stats_update_end(&q->syncp);
+
+	trace_funeth_tx(q, tot_len, idx, nfrags);
+	q->prod_cnt += ndesc;
+
+	return true;
+}
+
+int fun_xdp_xmit_frames(struct net_device *dev, int n,
+			struct xdp_frame **frames, u32 flags)
+{
+	struct funeth_priv *fp = netdev_priv(dev);
+	struct funeth_txq *q, **xdpqs;
+	int i, q_idx;
+
+	if (unlikely(flags & ~XDP_XMIT_FLAGS_MASK))
+		return -EINVAL;
+
+	xdpqs = rcu_dereference_bh(fp->xdpqs);
+	if (unlikely(!xdpqs))
+		return -ENETDOWN;
+
+	q_idx = smp_processor_id();
+	if (unlikely(q_idx >= fp->num_xdpqs))
+		return -ENXIO;
+
+	for (q = xdpqs[q_idx], i = 0; i < n; i++)
+		if (!fun_xdp_tx(q, frames[i]))
+			break;
+
+	if (unlikely(flags & XDP_XMIT_FLUSH))
+		fun_txq_wr_db(q);
+	return i;
+}
+
+/* Purge a Tx queue of any queued packets. Should be called once HW access
+ * to the packets has been revoked, e.g., after the queue has been disabled.
+ */
+static void fun_txq_purge(struct funeth_txq *q)
+{
+	while (q->cons_cnt != q->prod_cnt) {
+		unsigned int idx = q->cons_cnt & q->mask;
+
+		q->cons_cnt += fun_unmap_pkt(q, idx);
+		dev_kfree_skb_any(q->info[idx].skb);
+	}
+	netdev_tx_reset_queue(q->ndq);
+}
+
+static void fun_xdpq_purge(struct funeth_txq *q)
+{
+	while (q->cons_cnt != q->prod_cnt) {
+		unsigned int idx = q->cons_cnt & q->mask;
+
+		q->cons_cnt += fun_unmap_pkt(q, idx);
+		xdp_return_frame(q->info[idx].xdpf);
+	}
+}
+
+/* Create a Tx queue, allocating all the host resources needed. */
+static struct funeth_txq *fun_txq_create_sw(struct net_device *dev,
+					    unsigned int qidx,
+					    unsigned int ndesc,
+					    struct fun_irq *irq)
+{
+	struct funeth_priv *fp = netdev_priv(dev);
+	struct funeth_txq *q;
+	int numa_node;
+
+	if (irq)
+		numa_node = fun_irq_node(irq); /* skb Tx queue */
+	else
+		numa_node = cpu_to_node(qidx); /* XDP Tx queue */
+
+	q = kzalloc_node(sizeof(*q), GFP_KERNEL, numa_node);
+	if (!q)
+		goto err;
+
+	q->dma_dev = &fp->pdev->dev;
+	q->desc = fun_alloc_ring_mem(q->dma_dev, ndesc, FUNETH_SQE_SIZE,
+				     sizeof(*q->info), true, numa_node,
+				     &q->dma_addr, (void **)&q->info,
+				     &q->hw_wb);
+	if (!q->desc)
+		goto free_q;
+
+	q->netdev = dev;
+	q->mask = ndesc - 1;
+	q->qidx = qidx;
+	q->numa_node = numa_node;
+	u64_stats_init(&q->syncp);
+	q->init_state = FUN_QSTATE_INIT_SW;
+	return q;
+
+free_q:
+	kfree(q);
+err:
+	netdev_err(dev, "Can't allocate memory for %s queue %u\n",
+		   irq ? "Tx" : "XDP", qidx);
+	return NULL;
+}
+
+static void fun_txq_free_sw(struct funeth_txq *q)
+{
+	struct funeth_priv *fp = netdev_priv(q->netdev);
+
+	fun_free_ring_mem(q->dma_dev, q->mask + 1, FUNETH_SQE_SIZE, true,
+			  q->desc, q->dma_addr, q->info);
+
+	fp->tx_packets += q->stats.tx_pkts;
+	fp->tx_bytes   += q->stats.tx_bytes;
+	fp->tx_dropped += q->stats.tx_map_err;
+
+	kfree(q);
+}
+
+/* Allocate the device portion of a Tx queue. */
+int fun_txq_create_dev(struct funeth_txq *q, struct fun_irq *irq)
+{
+	struct funeth_priv *fp = netdev_priv(q->netdev);
+	unsigned int irq_idx, ndesc = q->mask + 1;
+	int err;
+
+	q->irq = irq;
+	*q->hw_wb = 0;
+	q->prod_cnt = 0;
+	q->cons_cnt = 0;
+	irq_idx = irq ? irq->irq_idx : 0;
+
+	err = fun_sq_create(fp->fdev,
+			    FUN_ADMIN_EPSQ_CREATE_FLAG_HEAD_WB_ADDRESS |
+			    FUN_ADMIN_RES_CREATE_FLAG_ALLOCATOR, 0,
+			    FUN_HCI_ID_INVALID, ilog2(FUNETH_SQE_SIZE), ndesc,
+			    q->dma_addr, fp->tx_coal_count, fp->tx_coal_usec,
+			    irq_idx, 0, fp->fdev->kern_end_qid, 0,
+			    &q->hw_qid, &q->db);
+	if (err)
+		goto out;
+
+	err = fun_create_and_bind_tx(fp, q->hw_qid);
+	if (err < 0)
+		goto free_devq;
+	q->ethid = err;
+
+	if (irq) {
+		irq->txq = q;
+		q->ndq = netdev_get_tx_queue(q->netdev, q->qidx);
+		q->irq_db_val = FUN_IRQ_SQ_DB(fp->tx_coal_usec,
+					      fp->tx_coal_count);
+		writel(q->irq_db_val, q->db);
+	}
+
+	q->init_state = FUN_QSTATE_INIT_FULL;
+	netif_info(fp, ifup, q->netdev,
+		   "%s queue %u, depth %u, HW qid %u, IRQ idx %u, eth id %u, node %d\n",
+		   irq ? "Tx" : "XDP", q->qidx, ndesc, q->hw_qid, irq_idx,
+		   q->ethid, q->numa_node);
+	return 0;
+
+free_devq:
+	fun_destroy_sq(fp->fdev, q->hw_qid);
+out:
+	netdev_err(q->netdev,
+		   "Failed to create %s queue %u on device, error %d\n",
+		   irq ? "Tx" : "XDP", q->qidx, err);
+	return err;
+}
+
+static void fun_txq_free_dev(struct funeth_txq *q)
+{
+	struct funeth_priv *fp = netdev_priv(q->netdev);
+
+	if (q->init_state < FUN_QSTATE_INIT_FULL)
+		return;
+
+	netif_info(fp, ifdown, q->netdev,
+		   "Freeing %s queue %u (id %u), IRQ %u, ethid %u\n",
+		   q->irq ? "Tx" : "XDP", q->qidx, q->hw_qid,
+		   q->irq ? q->irq->irq_idx : 0, q->ethid);
+
+	fun_destroy_sq(fp->fdev, q->hw_qid);
+	fun_res_destroy(fp->fdev, FUN_ADMIN_OP_ETH, 0, q->ethid);
+
+	if (q->irq) {
+		q->irq->txq = NULL;
+		fun_txq_purge(q);
+	} else {
+		fun_xdpq_purge(q);
+	}
+
+	q->init_state = FUN_QSTATE_INIT_SW;
+}
+
+/* Create or advance a Tx queue, allocating all the host and device resources
+ * needed to reach the target state.
+ */
+int funeth_txq_create(struct net_device *dev, unsigned int qidx,
+		      unsigned int ndesc, struct fun_irq *irq, int state,
+		      struct funeth_txq **qp)
+{
+	struct funeth_txq *q = *qp;
+	int err;
+
+	if (!q)
+		q = fun_txq_create_sw(dev, qidx, ndesc, irq);
+	if (!q)
+		return -ENOMEM;
+
+	if (q->init_state >= state)
+		goto out;
+
+	err = fun_txq_create_dev(q, irq);
+	if (err) {
+		if (!*qp)
+			fun_txq_free_sw(q);
+		return err;
+	}
+
+out:
+	*qp = q;
+	return 0;
+}
+
+/* Free Tx queue resources until it reaches the target state.
+ * The queue must be already disconnected from the stack.
+ */
+struct funeth_txq *funeth_txq_free(struct funeth_txq *q, int state)
+{
+	if (state < FUN_QSTATE_INIT_FULL)
+		fun_txq_free_dev(q);
+
+	if (state == FUN_QSTATE_DESTROYED) {
+		fun_txq_free_sw(q);
+		q = NULL;
+	}
+
+	return q;
+}
diff --git a/drivers/net/ethernet/fungible/funeth/funeth_txrx.h b/drivers/net/ethernet/fungible/funeth/funeth_txrx.h
new file mode 100644
index 000000000000..5eec552a1f24
--- /dev/null
+++ b/drivers/net/ethernet/fungible/funeth/funeth_txrx.h
@@ -0,0 +1,266 @@
+/* SPDX-License-Identifier: (GPL-2.0-only OR BSD-3-Clause) */
+
+#ifndef _FUNETH_TXRX_H
+#define _FUNETH_TXRX_H
+
+#include <linux/netdevice.h>
+#include <linux/u64_stats_sync.h>
+#include <net/xdp.h>
+
+/* Tx descriptor size */
+#define FUNETH_SQE_SIZE 64U
+
+/* Size of device headers per Tx packet */
+#define FUNETH_FUNOS_HDR_SZ (sizeof(struct fun_eth_tx_req))
+
+/* Number of gather list entries per Tx descriptor */
+#define FUNETH_GLE_PER_DESC (FUNETH_SQE_SIZE / sizeof(struct fun_dataop_gl))
+
+/* Max gather list size in bytes for an sk_buff. */
+#define FUNETH_MAX_GL_SZ ((MAX_SKB_FRAGS + 1) * sizeof(struct fun_dataop_gl))
+
+#if IS_ENABLED(CONFIG_TLS_DEVICE)
+# define FUNETH_TLS_SZ sizeof(struct fun_eth_tls)
+#else
+# define FUNETH_TLS_SZ 0
+#endif
+
+/* Max number of Tx descriptors for an sk_buff using a gather list. */
+#define FUNETH_MAX_GL_DESC \
+	DIV_ROUND_UP((FUNETH_FUNOS_HDR_SZ + FUNETH_MAX_GL_SZ + FUNETH_TLS_SZ), \
+		     FUNETH_SQE_SIZE)
+
+/* Max number of Tx descriptors for any packet. */
+#define FUNETH_MAX_PKT_DESC FUNETH_MAX_GL_DESC
+
+/* Rx CQ descriptor size. */
+#define FUNETH_CQE_SIZE 64U
+
+/* Offset of cqe_info within a CQE. */
+#define FUNETH_CQE_INFO_OFFSET (FUNETH_CQE_SIZE - sizeof(struct fun_cqe_info))
+
+/* Construct the IRQ portion of a CQ doorbell. The resulting value arms the
+ * interrupt with the supplied time delay and packet count moderation settings.
+ */
+#define FUN_IRQ_CQ_DB(usec, pkts) \
+	(FUN_DB_IRQ_ARM_F | ((usec) << FUN_DB_INTCOAL_USEC_S) | \
+	 ((pkts) << FUN_DB_INTCOAL_ENTRIES_S))
+
+/* As above for SQ doorbells. */
+#define FUN_IRQ_SQ_DB(usec, pkts) \
+	(FUN_DB_IRQ_ARM_F | \
+	 ((usec) << FUN_DB_INTCOAL_USEC_S) | \
+	 ((pkts) << FUN_DB_INTCOAL_ENTRIES_S))
+
+/* Per packet tailroom. Present only for 1-frag packets. */
+#define FUN_RX_TAILROOM SKB_DATA_ALIGN(sizeof(struct skb_shared_info))
+
+/* Per packet headroom for XDP. Preferred over XDP_PACKET_HEADROOM to
+ * accommodate two packets per buffer for 4K pages and 1500B MTUs.
+ */
+#define FUN_XDP_HEADROOM 192
+
+/* Initialization state of a queue. */
+enum {
+	FUN_QSTATE_DESTROYED, /* what queue? */
+	FUN_QSTATE_INIT_SW,   /* exists in SW, not on the device */
+	FUN_QSTATE_INIT_FULL, /* exists both in SW and on device */
+};
+
+/* Initialization state of an interrupt. */
+enum {
+	FUN_IRQ_INIT,      /* initialized and in the XArray but inactive */
+	FUN_IRQ_REQUESTED, /* request_irq() done */
+	FUN_IRQ_ENABLED,   /* processing enabled */
+	FUN_IRQ_DISABLED,  /* processing disabled */
+};
+
+struct bpf_prog;
+
+struct funeth_txq_stats {  /* per Tx queue SW counters */
+	u64 tx_pkts;       /* # of Tx packets */
+	u64 tx_bytes;      /* total bytes of Tx packets */
+	u64 tx_cso;        /* # of packets with checksum offload */
+	u64 tx_tso;        /* # of non-encapsulated TSO super-packets */
+	u64 tx_encap_tso;  /* # of encapsulated TSO super-packets */
+	u64 tx_uso;        /* # of non-encapsulated UDP LSO super-packets */
+	u64 tx_more;       /* # of DBs elided due to xmit_more */
+	u64 tx_nstops;     /* # of times the queue has stopped */
+	u64 tx_nrestarts;  /* # of times the queue has restarted */
+	u64 tx_map_err;    /* # of packets dropped due to DMA mapping errors */
+	u64 tx_xdp_full;   /* # of XDP packets that could not be enqueued */
+	u64 tx_tls_pkts;   /* # of Tx TLS packets offloaded to HW */
+	u64 tx_tls_bytes;  /* Tx bytes of HW-handled TLS payload */
+	u64 tx_tls_fallback; /* attempted Tx TLS offloads punted to SW */
+	u64 tx_tls_drops;  /* attempted Tx TLS offloads dropped */
+};
+
+struct funeth_tx_info {      /* per Tx descriptor state */
+	union {
+		struct sk_buff *skb;    /* associated packet (sk_buff path) */
+		struct xdp_frame *xdpf; /* associated XDP frame (XDP path) */
+	};
+};
+
+struct funeth_txq {
+	/* RO cacheline of frequently accessed data */
+	u32 mask;               /* queue depth - 1 */
+	u32 hw_qid;             /* device ID of the queue */
+	void *desc;             /* base address of descriptor ring */
+	struct funeth_tx_info *info;
+	struct device *dma_dev; /* device for DMA mappings */
+	volatile __be64 *hw_wb; /* HW write-back location */
+	u32 __iomem *db;        /* SQ doorbell register address */
+	struct netdev_queue *ndq;
+	dma_addr_t dma_addr;    /* DMA address of descriptor ring */
+	/* producer R/W cacheline */
+	u16 qidx;               /* queue index within net_device */
+	u16 ethid;
+	u32 prod_cnt;           /* producer counter */
+	struct funeth_txq_stats stats;
+	/* shared R/W cacheline, primarily accessed by consumer */
+	u32 irq_db_val;         /* value written to IRQ doorbell */
+	u32 cons_cnt;           /* consumer (cleanup) counter */
+	struct net_device *netdev;
+	struct fun_irq *irq;
+	int numa_node;
+	u8 init_state;          /* queue initialization state */
+	struct u64_stats_sync syncp;
+};
+
+struct funeth_rxq_stats {  /* per Rx queue SW counters */
+	u64 rx_pkts;       /* # of received packets, including SW drops */
+	u64 rx_bytes;      /* total size of received packets */
+	u64 rx_cso;        /* # of packets with checksum offload */
+	u64 rx_bufs;       /* total # of Rx buffers provided to device */
+	u64 gro_pkts;      /* # of GRO superpackets */
+	u64 gro_merged;    /* # of pkts merged into existing GRO superpackets */
+	u64 rx_page_alloc; /* # of page allocations for Rx buffers */
+	u64 rx_budget;     /* NAPI iterations that exhausted their budget */
+	u64 rx_mem_drops;  /* # of packets dropped due to memory shortage */
+	u64 rx_map_err;    /* # of page DMA mapping errors */
+	u64 xdp_drops;     /* XDP_DROPped packets */
+	u64 xdp_tx;        /* successful XDP transmits */
+	u64 xdp_redir;     /* successful XDP redirects */
+	u64 xdp_err;       /* packets dropped due to XDP errors */
+};
+
+struct funeth_rxbuf {          /* per Rx buffer state */
+	struct page *page;     /* associated page */
+	dma_addr_t dma_addr;   /* DMA address of page start */
+	int pg_refs;           /* page refs held by driver */
+	int node;              /* page node, or -1 if it is PF_MEMALLOC */
+};
+
+struct funeth_rx_cache {       /* cache of DMA-mapped previously used buffers */
+	struct funeth_rxbuf *bufs; /* base of Rx buffer state ring */
+	unsigned int prod_cnt;     /* producer counter */
+	unsigned int cons_cnt;     /* consumer counter */
+	unsigned int mask;         /* depth - 1 */
+};
+
+/* An Rx queue consists of a CQ and an SQ used to provide Rx buffers. */
+struct funeth_rxq {
+	struct net_device *netdev;
+	struct napi_struct *napi;
+	struct device *dma_dev;    /* device for DMA mappings */
+	void *cqes;                /* base of CQ descriptor ring */
+	const void *next_cqe_info; /* fun_cqe_info of next CQE */
+	u32 __iomem *cq_db;        /* CQ doorbell register address */
+	unsigned int cq_head;      /* CQ head index */
+	unsigned int cq_mask;      /* CQ depth - 1 */
+	u16 phase;                 /* CQ phase tag */
+	u16 qidx;                  /* queue index within net_device */
+	unsigned int irq_db_val;   /* IRQ info for CQ doorbell */
+	struct fun_eprq_rqbuf *rqes; /* base of RQ descriptor ring */
+	struct funeth_rxbuf *bufs; /* base of Rx buffer state ring */
+	struct funeth_rxbuf *cur_buf; /* currently active buffer */
+	u32 __iomem *rq_db;        /* RQ doorbell register address */
+	unsigned int rq_cons;      /* RQ consumer counter */
+	unsigned int rq_mask;      /* RQ depth - 1 */
+	unsigned int buf_offset;   /* offset of next pkt in head buffer */
+	u8 xdp_flush;              /* XDP flush types needed at NAPI end */
+	u8 init_state;             /* queue initialization state */
+	u16 headroom;              /* per packet headroom */
+	unsigned int rq_cons_db;   /* value of rq_cons at last RQ db */
+	unsigned int rq_db_thres;  /* # of new buffers needed to write RQ db */
+	struct funeth_rxbuf spare_buf; /* spare for next buffer replacement */
+	struct funeth_rx_cache cache; /* used buffer cache */
+	struct bpf_prog *xdp_prog; /* optional XDP BPF program */
+	struct funeth_rxq_stats stats;
+	dma_addr_t cq_dma_addr;    /* DMA address of CQE ring */
+	dma_addr_t rq_dma_addr;    /* DMA address of RQE ring */
+	u16 irq_cnt;
+	u32 hw_cqid;               /* device ID of the queue's CQ */
+	u32 hw_sqid;               /* device ID of the queue's SQ */
+	int numa_node;
+	struct u64_stats_sync syncp;
+	struct xdp_rxq_info xdp_rxq;
+};
+
+#define FUN_QSTAT_INC(q, counter) \
+	do { \
+		u64_stats_update_begin(&(q)->syncp); \
+		(q)->stats.counter++; \
+		u64_stats_update_end(&(q)->syncp); \
+	} while (0)
+
+#define FUN_QSTAT_READ(q, seq, stats_copy) \
+	do { \
+		seq = u64_stats_fetch_begin(&(q)->syncp); \
+		stats_copy = (q)->stats; \
+	} while (u64_stats_fetch_retry(&(q)->syncp, (seq)))
+
+#define FUN_INT_NAME_LEN (IFNAMSIZ + 16)
+
+struct fun_irq {
+	struct napi_struct napi;
+	struct funeth_txq *txq;
+	struct funeth_rxq *rxq;
+	u8 state;
+	u16 irq_idx;              /* index of MSI-X interrupt */
+	int irq;                  /* Linux IRQ vector */
+	cpumask_t affinity_mask;  /* IRQ affinity */
+	struct irq_affinity_notify aff_notify;
+	char name[FUN_INT_NAME_LEN];
+} ____cacheline_internodealigned_in_smp;
+
+/* Return the start address of the idx-th Tx descriptor. */
+static inline void *fun_tx_desc_addr(const struct funeth_txq *q,
+				     unsigned int idx)
+{
+	return q->desc + idx * FUNETH_SQE_SIZE;
+}
+
+static inline void fun_txq_wr_db(const struct funeth_txq *q)
+{
+	unsigned int tail = q->prod_cnt & q->mask;
+
+	writel(tail, q->db);
+}
+
+static inline int fun_irq_node(const struct fun_irq *p)
+{
+	return cpu_to_mem(cpumask_first(&p->affinity_mask));
+}
+
+int fun_rxq_napi_poll(struct napi_struct *napi, int budget);
+int fun_txq_napi_poll(struct napi_struct *napi, int budget);
+netdev_tx_t fun_start_xmit(struct sk_buff *skb, struct net_device *netdev);
+bool fun_xdp_tx(struct funeth_txq *q, struct xdp_frame *xdpf);
+int fun_xdp_xmit_frames(struct net_device *dev, int n,
+			struct xdp_frame **frames, u32 flags);
+
+int funeth_txq_create(struct net_device *dev, unsigned int qidx,
+		      unsigned int ndesc, struct fun_irq *irq, int state,
+		      struct funeth_txq **qp);
+int fun_txq_create_dev(struct funeth_txq *q, struct fun_irq *irq);
+struct funeth_txq *funeth_txq_free(struct funeth_txq *q, int state);
+int funeth_rxq_create(struct net_device *dev, unsigned int qidx,
+		      unsigned int ncqe, unsigned int nrqe, struct fun_irq *irq,
+		      int state, struct funeth_rxq **qp);
+int fun_rxq_create_dev(struct funeth_rxq *q, struct fun_irq *irq);
+struct funeth_rxq *funeth_rxq_free(struct funeth_rxq *q, int state);
+int fun_rxq_set_bpf(struct funeth_rxq *q, struct bpf_prog *prog);
+
+#endif /* _FUNETH_TXRX_H */
diff --git a/drivers/net/ethernet/google/Kconfig b/drivers/net/ethernet/google/Kconfig
new file mode 100644
index 000000000000..14c9431e15e5
--- /dev/null
+++ b/drivers/net/ethernet/google/Kconfig
@@ -0,0 +1,29 @@
+#
+# Google network device configuration
+#
+
+config NET_VENDOR_GOOGLE
+	bool "Google Devices"
+	default y
+	help
+	  If you have a network (Ethernet) device belonging to this class, say Y.
+
+	  Note that the answer to this question doesn't directly affect the
+	  kernel: saying N will just cause the configurator to skip all
+	  the questions about Google devices. If you say Y, you will be asked
+	  for your specific device in the following questions.
+
+if NET_VENDOR_GOOGLE
+
+config GVE
+	tristate "Google Virtual NIC (gVNIC) support"
+	depends on (PCI_MSI && (X86 || CPU_LITTLE_ENDIAN))
+	depends on PTP_1588_CLOCK_OPTIONAL
+	select PAGE_POOL
+	help
+	  This driver supports Google Virtual NIC (gVNIC)"
+
+	  To compile this driver as a module, choose M here.
+	  The module will be called gve.
+
+endif #NET_VENDOR_GOOGLE
diff --git a/drivers/net/ethernet/google/Makefile b/drivers/net/ethernet/google/Makefile
new file mode 100644
index 000000000000..402cc3ba1639
--- /dev/null
+++ b/drivers/net/ethernet/google/Makefile
@@ -0,0 +1,5 @@
+#
+# Makefile for the Google network device drivers.
+#
+
+obj-$(CONFIG_GVE) += gve/
diff --git a/drivers/net/ethernet/google/gve/Makefile b/drivers/net/ethernet/google/gve/Makefile
new file mode 100644
index 000000000000..e0ec227a50f7
--- /dev/null
+++ b/drivers/net/ethernet/google/gve/Makefile
@@ -0,0 +1,7 @@
+# Makefile for the Google virtual Ethernet (gve) driver
+
+obj-$(CONFIG_GVE) += gve.o
+gve-y := gve_main.o gve_tx.o gve_tx_dqo.o gve_rx.o gve_rx_dqo.o gve_ethtool.o gve_adminq.o gve_utils.o gve_flow_rule.o \
+	    gve_buffer_mgmt_dqo.o
+
+gve-$(CONFIG_PTP_1588_CLOCK) += gve_ptp.o
diff --git a/drivers/net/ethernet/google/gve/gve.h b/drivers/net/ethernet/google/gve/gve.h
new file mode 100644
index 000000000000..bceaf9b05cb4
--- /dev/null
+++ b/drivers/net/ethernet/google/gve/gve.h
@@ -0,0 +1,1330 @@
+/* SPDX-License-Identifier: (GPL-2.0 OR MIT)
+ * Google virtual Ethernet (gve) driver
+ *
+ * Copyright (C) 2015-2024 Google LLC
+ */
+
+#ifndef _GVE_H_
+#define _GVE_H_
+
+#include <linux/dma-mapping.h>
+#include <linux/dmapool.h>
+#include <linux/ethtool_netlink.h>
+#include <linux/netdevice.h>
+#include <linux/net_tstamp.h>
+#include <linux/pci.h>
+#include <linux/ptp_clock_kernel.h>
+#include <linux/u64_stats_sync.h>
+#include <net/page_pool/helpers.h>
+#include <net/xdp.h>
+
+#include "gve_desc.h"
+#include "gve_desc_dqo.h"
+
+#ifndef PCI_VENDOR_ID_GOOGLE
+#define PCI_VENDOR_ID_GOOGLE	0x1ae0
+#endif
+
+#define PCI_DEV_ID_GVNIC	0x0042
+
+#define GVE_REGISTER_BAR	0
+#define GVE_DOORBELL_BAR	2
+
+/* Driver can alloc up to 2 segments for the header and 2 for the payload. */
+#define GVE_TX_MAX_IOVEC	4
+/* 1 for management, 1 for rx, 1 for tx */
+#define GVE_MIN_MSIX 3
+
+/* Numbers of gve tx/rx stats in stats report. */
+#define GVE_TX_STATS_REPORT_NUM	6
+#define GVE_RX_STATS_REPORT_NUM	2
+
+/* Interval to schedule a stats report update, 20000ms. */
+#define GVE_STATS_REPORT_TIMER_PERIOD	20000
+
+/* Numbers of NIC tx/rx stats in stats report. */
+#define NIC_TX_STATS_REPORT_NUM	0
+#define NIC_RX_STATS_REPORT_NUM	4
+
+#define GVE_ADMINQ_BUFFER_SIZE 4096
+
+#define GVE_DATA_SLOT_ADDR_PAGE_MASK (~(PAGE_SIZE - 1))
+
+/* PTYPEs are always 10 bits. */
+#define GVE_NUM_PTYPES	1024
+
+/* Default minimum ring size */
+#define GVE_DEFAULT_MIN_TX_RING_SIZE 256
+#define GVE_DEFAULT_MIN_RX_RING_SIZE 512
+
+#define GVE_DEFAULT_RX_BUFFER_SIZE 2048
+
+#define GVE_MAX_RX_BUFFER_SIZE 4096
+
+#define GVE_XDP_RX_BUFFER_SIZE_DQO 4096
+
+#define GVE_DEFAULT_RX_BUFFER_OFFSET 2048
+
+#define GVE_PAGE_POOL_SIZE_MULTIPLIER 4
+
+#define GVE_FLOW_RULES_CACHE_SIZE \
+	(GVE_ADMINQ_BUFFER_SIZE / sizeof(struct gve_adminq_queried_flow_rule))
+#define GVE_FLOW_RULE_IDS_CACHE_SIZE \
+	(GVE_ADMINQ_BUFFER_SIZE / sizeof(((struct gve_adminq_queried_flow_rule *)0)->location))
+
+#define GVE_RSS_KEY_SIZE	40
+#define GVE_RSS_INDIR_SIZE	128
+
+#define GVE_XDP_ACTIONS 5
+
+#define GVE_GQ_TX_MIN_PKT_DESC_BYTES 182
+
+#define GVE_DEFAULT_HEADER_BUFFER_SIZE 128
+
+#define DQO_QPL_DEFAULT_TX_PAGES 512
+
+/* Maximum TSO size supported on DQO */
+#define GVE_DQO_TX_MAX	0x3FFFF
+
+#define GVE_TX_BUF_SHIFT_DQO 11
+
+/* 2K buffers for DQO-QPL */
+#define GVE_TX_BUF_SIZE_DQO BIT(GVE_TX_BUF_SHIFT_DQO)
+#define GVE_TX_BUFS_PER_PAGE_DQO (PAGE_SIZE >> GVE_TX_BUF_SHIFT_DQO)
+#define GVE_MAX_TX_BUFS_PER_PKT (DIV_ROUND_UP(GVE_DQO_TX_MAX, GVE_TX_BUF_SIZE_DQO))
+
+/* If number of free/recyclable buffers are less than this threshold; driver
+ * allocs and uses a non-qpl page on the receive path of DQO QPL to free
+ * up buffers.
+ * Value is set big enough to post at least 3 64K LRO packet via 2K buffer to NIC.
+ */
+#define GVE_DQO_QPL_ONDEMAND_ALLOC_THRESHOLD 96
+
+/* Each slot in the desc ring has a 1:1 mapping to a slot in the data ring */
+struct gve_rx_desc_queue {
+	struct gve_rx_desc *desc_ring; /* the descriptor ring */
+	dma_addr_t bus; /* the bus for the desc_ring */
+	u8 seqno; /* the next expected seqno for this desc*/
+};
+
+/* The page info for a single slot in the RX data queue */
+struct gve_rx_slot_page_info {
+	/* netmem is used for DQO RDA mode
+	 * page is used in all other modes
+	 */
+	union {
+		struct page *page;
+		netmem_ref netmem;
+	};
+	void *page_address;
+	u32 page_offset; /* offset to write to in page */
+	unsigned int buf_size;
+	int pagecnt_bias; /* expected pagecnt if only the driver has a ref */
+	u16 pad; /* adjustment for rx padding */
+	u8 can_flip; /* tracks if the networking stack is using the page */
+};
+
+/* A list of pages registered with the device during setup and used by a queue
+ * as buffers
+ */
+struct gve_queue_page_list {
+	u32 id; /* unique id */
+	u32 num_entries;
+	struct page **pages; /* list of num_entries pages */
+	dma_addr_t *page_buses; /* the dma addrs of the pages */
+};
+
+/* Each slot in the data ring has a 1:1 mapping to a slot in the desc ring */
+struct gve_rx_data_queue {
+	union gve_rx_data_slot *data_ring; /* read by NIC */
+	dma_addr_t data_bus; /* dma mapping of the slots */
+	struct gve_rx_slot_page_info *page_info; /* page info of the buffers */
+	struct gve_queue_page_list *qpl; /* qpl assigned to this queue */
+	u8 raw_addressing; /* use raw_addressing? */
+};
+
+struct gve_priv;
+
+/* RX buffer queue for posting buffers to HW.
+ * Each RX (completion) queue has a corresponding buffer queue.
+ */
+struct gve_rx_buf_queue_dqo {
+	struct gve_rx_desc_dqo *desc_ring;
+	dma_addr_t bus;
+	u32 head; /* Pointer to start cleaning buffers at. */
+	u32 tail; /* Last posted buffer index + 1 */
+	u32 mask; /* Mask for indices to the size of the ring */
+};
+
+/* RX completion queue to receive packets from HW. */
+struct gve_rx_compl_queue_dqo {
+	struct gve_rx_compl_desc_dqo *desc_ring;
+	dma_addr_t bus;
+
+	/* Number of slots which did not have a buffer posted yet. We should not
+	 * post more buffers than the queue size to avoid HW overrunning the
+	 * queue.
+	 */
+	int num_free_slots;
+
+	/* HW uses a "generation bit" to notify SW of new descriptors. When a
+	 * descriptor's generation bit is different from the current generation,
+	 * that descriptor is ready to be consumed by SW.
+	 */
+	u8 cur_gen_bit;
+
+	/* Pointer into desc_ring where the next completion descriptor will be
+	 * received.
+	 */
+	u32 head;
+	u32 mask; /* Mask for indices to the size of the ring */
+};
+
+struct gve_header_buf {
+	u8 *data;
+	dma_addr_t addr;
+};
+
+/* Stores state for tracking buffers posted to HW */
+struct gve_rx_buf_state_dqo {
+	/* The page posted to HW. */
+	struct gve_rx_slot_page_info page_info;
+
+	/* XSK buffer */
+	struct xdp_buff *xsk_buff;
+
+	/* The DMA address corresponding to `page_info`. */
+	dma_addr_t addr;
+
+	/* Last offset into the page when it only had a single reference, at
+	 * which point every other offset is free to be reused.
+	 */
+	u32 last_single_ref_offset;
+
+	/* Linked list index to next element in the list, or -1 if none */
+	s16 next;
+};
+
+/* `head` and `tail` are indices into an array, or -1 if empty. */
+struct gve_index_list {
+	s16 head;
+	s16 tail;
+};
+
+/* A single received packet split across multiple buffers may be
+ * reconstructed using the information in this structure.
+ */
+struct gve_rx_ctx {
+	/* head and tail of skb chain for the current packet or NULL if none */
+	struct sk_buff *skb_head;
+	struct sk_buff *skb_tail;
+	u32 total_size;
+	u8 frag_cnt;
+	bool drop_pkt;
+};
+
+struct gve_rx_cnts {
+	u32 ok_pkt_bytes;
+	u16 ok_pkt_cnt;
+	u16 total_pkt_cnt;
+	u16 cont_pkt_cnt;
+	u16 desc_err_pkt_cnt;
+};
+
+/* Contains datapath state used to represent an RX queue. */
+struct gve_rx_ring {
+	struct gve_priv *gve;
+
+	u16 packet_buffer_size;		/* Size of buffer posted to NIC */
+	u16 packet_buffer_truesize;	/* Total size of RX buffer */
+	u16 rx_headroom;
+
+	union {
+		/* GQI fields */
+		struct {
+			struct gve_rx_desc_queue desc;
+			struct gve_rx_data_queue data;
+
+			/* threshold for posting new buffs and descs */
+			u32 db_threshold;
+
+			u32 qpl_copy_pool_mask;
+			u32 qpl_copy_pool_head;
+			struct gve_rx_slot_page_info *qpl_copy_pool;
+		};
+
+		/* DQO fields. */
+		struct {
+			struct gve_rx_buf_queue_dqo bufq;
+			struct gve_rx_compl_queue_dqo complq;
+
+			struct gve_rx_buf_state_dqo *buf_states;
+			u16 num_buf_states;
+
+			/* Linked list of gve_rx_buf_state_dqo. Index into
+			 * buf_states, or -1 if empty.
+			 */
+			s16 free_buf_states;
+
+			/* Linked list of gve_rx_buf_state_dqo. Indexes into
+			 * buf_states, or -1 if empty.
+			 *
+			 * This list contains buf_states which are pointing to
+			 * valid buffers.
+			 *
+			 * We use a FIFO here in order to increase the
+			 * probability that buffers can be reused by increasing
+			 * the time between usages.
+			 */
+			struct gve_index_list recycled_buf_states;
+
+			/* Linked list of gve_rx_buf_state_dqo. Indexes into
+			 * buf_states, or -1 if empty.
+			 *
+			 * This list contains buf_states which have buffers
+			 * which cannot be reused yet.
+			 */
+			struct gve_index_list used_buf_states;
+
+			/* qpl assigned to this queue */
+			struct gve_queue_page_list *qpl;
+
+			/* index into queue page list */
+			u32 next_qpl_page_idx;
+
+			/* track number of used buffers */
+			u16 used_buf_states_cnt;
+
+			/* Address info of the buffers for header-split */
+			struct gve_header_buf hdr_bufs;
+
+			struct page_pool *page_pool;
+		} dqo;
+	};
+
+	u64 rbytes; /* free-running bytes received */
+	u64 rx_hsplit_bytes; /* free-running header bytes received */
+	u64 rpackets; /* free-running packets received */
+	u32 cnt; /* free-running total number of completed packets */
+	u32 fill_cnt; /* free-running total number of descs and buffs posted */
+	u32 mask; /* masks the cnt and fill_cnt to the size of the ring */
+	u64 rx_hsplit_pkt; /* free-running packets with headers split */
+	u64 rx_copybreak_pkt; /* free-running count of copybreak packets */
+	u64 rx_copied_pkt; /* free-running total number of copied packets */
+	u64 rx_skb_alloc_fail; /* free-running count of skb alloc fails */
+	u64 rx_buf_alloc_fail; /* free-running count of buffer alloc fails */
+	u64 rx_desc_err_dropped_pkt; /* free-running count of packets dropped by descriptor error */
+	/* free-running count of unsplit packets due to header buffer overflow or hdr_len is 0 */
+	u64 rx_hsplit_unsplit_pkt;
+	u64 rx_cont_packet_cnt; /* free-running multi-fragment packets received */
+	u64 rx_frag_flip_cnt; /* free-running count of rx segments where page_flip was used */
+	u64 rx_frag_copy_cnt; /* free-running count of rx segments copied */
+	u64 rx_frag_alloc_cnt; /* free-running count of rx page allocations */
+	u64 xdp_tx_errors;
+	u64 xdp_redirect_errors;
+	u64 xdp_alloc_fails;
+	u64 xdp_actions[GVE_XDP_ACTIONS];
+	u32 q_num; /* queue index */
+	u32 ntfy_id; /* notification block index */
+	struct gve_queue_resources *q_resources; /* head and tail pointer idx */
+	dma_addr_t q_resources_bus; /* dma address for the queue resources */
+	struct u64_stats_sync statss; /* sync stats for 32bit archs */
+
+	struct gve_rx_ctx ctx; /* Info for packet currently being processed in this ring. */
+
+	/* XDP stuff */
+	struct xdp_rxq_info xdp_rxq;
+	struct xsk_buff_pool *xsk_pool;
+	struct page_frag_cache page_cache; /* Page cache to allocate XDP frames */
+};
+
+/* A TX desc ring entry */
+union gve_tx_desc {
+	struct gve_tx_pkt_desc pkt; /* first desc for a packet */
+	struct gve_tx_mtd_desc mtd; /* optional metadata descriptor */
+	struct gve_tx_seg_desc seg; /* subsequent descs for a packet */
+};
+
+/* Tracks the memory in the fifo occupied by a segment of a packet */
+struct gve_tx_iovec {
+	u32 iov_offset; /* offset into this segment */
+	u32 iov_len; /* length */
+	u32 iov_padding; /* padding associated with this segment */
+};
+
+/* Tracks the memory in the fifo occupied by the skb. Mapped 1:1 to a desc
+ * ring entry but only used for a pkt_desc not a seg_desc
+ */
+struct gve_tx_buffer_state {
+	union {
+		struct sk_buff *skb; /* skb for this pkt */
+		struct xdp_frame *xdp_frame; /* xdp_frame */
+	};
+	struct {
+		u16 size; /* size of xmitted xdp pkt */
+		u8 is_xsk; /* xsk buff */
+	} xdp;
+	union {
+		struct gve_tx_iovec iov[GVE_TX_MAX_IOVEC]; /* segments of this pkt */
+		struct {
+			DEFINE_DMA_UNMAP_ADDR(dma);
+			DEFINE_DMA_UNMAP_LEN(len);
+		};
+	};
+};
+
+/* A TX buffer - each queue has one */
+struct gve_tx_fifo {
+	void *base; /* address of base of FIFO */
+	u32 size; /* total size */
+	atomic_t available; /* how much space is still available */
+	u32 head; /* offset to write at */
+	struct gve_queue_page_list *qpl; /* QPL mapped into this FIFO */
+};
+
+/* TX descriptor for DQO format */
+union gve_tx_desc_dqo {
+	struct gve_tx_pkt_desc_dqo pkt;
+	struct gve_tx_tso_context_desc_dqo tso_ctx;
+	struct gve_tx_general_context_desc_dqo general_ctx;
+};
+
+enum gve_packet_state {
+	/* Packet is in free list, available to be allocated.
+	 * This should always be zero since state is not explicitly initialized.
+	 */
+	GVE_PACKET_STATE_UNALLOCATED,
+	/* Packet is expecting a regular data completion or miss completion */
+	GVE_PACKET_STATE_PENDING_DATA_COMPL,
+	/* Packet has received a miss completion and is expecting a
+	 * re-injection completion.
+	 */
+	GVE_PACKET_STATE_PENDING_REINJECT_COMPL,
+	/* No valid completion received within the specified timeout. */
+	GVE_PACKET_STATE_TIMED_OUT_COMPL,
+	/* XSK pending packet has received a packet/reinjection completion, or
+	 * has timed out. At this point, the pending packet can be counted by
+	 * xsk_tx_complete and freed.
+	 */
+	GVE_PACKET_STATE_XSK_COMPLETE,
+};
+
+enum gve_tx_pending_packet_dqo_type {
+	GVE_TX_PENDING_PACKET_DQO_SKB,
+	GVE_TX_PENDING_PACKET_DQO_XDP_FRAME,
+	GVE_TX_PENDING_PACKET_DQO_XSK,
+};
+
+struct gve_tx_pending_packet_dqo {
+	union {
+		struct sk_buff *skb;
+		struct xdp_frame *xdpf;
+	};
+
+	/* 0th element corresponds to the linear portion of `skb`, should be
+	 * unmapped with `dma_unmap_single`.
+	 *
+	 * All others correspond to `skb`'s frags and should be unmapped with
+	 * `dma_unmap_page`.
+	 */
+	union {
+		struct {
+			DEFINE_DMA_UNMAP_ADDR(dma[MAX_SKB_FRAGS + 1]);
+			DEFINE_DMA_UNMAP_LEN(len[MAX_SKB_FRAGS + 1]);
+		};
+		s16 tx_qpl_buf_ids[GVE_MAX_TX_BUFS_PER_PKT];
+	};
+
+	u16 num_bufs;
+
+	/* Linked list index to next element in the list, or -1 if none */
+	s16 next;
+
+	/* Linked list index to prev element in the list, or -1 if none.
+	 * Used for tracking either outstanding miss completions or prematurely
+	 * freed packets.
+	 */
+	s16 prev;
+
+	/* Identifies the current state of the packet as defined in
+	 * `enum gve_packet_state`.
+	 */
+	u8 state : 3;
+
+	/* gve_tx_pending_packet_dqo_type */
+	u8 type : 2;
+
+	/* If packet is an outstanding miss completion, then the packet is
+	 * freed if the corresponding re-injection completion is not received
+	 * before kernel jiffies exceeds timeout_jiffies.
+	 */
+	unsigned long timeout_jiffies;
+};
+
+/* Contains datapath state used to represent a TX queue. */
+struct gve_tx_ring {
+	/* Cacheline 0 -- Accessed & dirtied during transmit */
+	union {
+		/* GQI fields */
+		struct {
+			struct gve_tx_fifo tx_fifo;
+			u32 req; /* driver tracked head pointer */
+			u32 done; /* driver tracked tail pointer */
+		};
+
+		/* DQO fields. */
+		struct {
+			/* Spinlock for XDP tx traffic */
+			spinlock_t xdp_lock;
+
+			/* Linked list of gve_tx_pending_packet_dqo. Index into
+			 * pending_packets, or -1 if empty.
+			 *
+			 * This is a consumer list owned by the TX path. When it
+			 * runs out, the producer list is stolen from the
+			 * completion handling path
+			 * (dqo_compl.free_pending_packets).
+			 */
+			s16 free_pending_packets;
+
+			/* Cached value of `dqo_compl.hw_tx_head` */
+			u32 head;
+			u32 tail; /* Last posted buffer index + 1 */
+
+			/* Index of the last descriptor with "report event" bit
+			 * set.
+			 */
+			u32 last_re_idx;
+
+			/* free running number of packet buf descriptors posted */
+			u16 posted_packet_desc_cnt;
+			/* free running number of packet buf descriptors completed */
+			u16 completed_packet_desc_cnt;
+
+			/* QPL fields */
+			struct {
+			       /* Linked list of gve_tx_buf_dqo. Index into
+				* tx_qpl_buf_next, or -1 if empty.
+				*
+				* This is a consumer list owned by the TX path. When it
+				* runs out, the producer list is stolen from the
+				* completion handling path
+				* (dqo_compl.free_tx_qpl_buf_head).
+				*/
+				s16 free_tx_qpl_buf_head;
+
+			       /* Free running count of the number of QPL tx buffers
+				* allocated
+				*/
+				u32 alloc_tx_qpl_buf_cnt;
+
+				/* Cached value of `dqo_compl.free_tx_qpl_buf_cnt` */
+				u32 free_tx_qpl_buf_cnt;
+			};
+
+			atomic_t xsk_reorder_queue_tail;
+		} dqo_tx;
+	};
+
+	/* Cacheline 1 -- Accessed & dirtied during gve_clean_tx_done */
+	union {
+		/* GQI fields */
+		struct {
+			/* Spinlock for when cleanup in progress */
+			spinlock_t clean_lock;
+			/* Spinlock for XDP tx traffic */
+			spinlock_t xdp_lock;
+		};
+
+		/* DQO fields. */
+		struct {
+			u32 head; /* Last read on compl_desc */
+
+			/* Tracks the current gen bit of compl_q */
+			u8 cur_gen_bit;
+
+			/* Linked list of gve_tx_pending_packet_dqo. Index into
+			 * pending_packets, or -1 if empty.
+			 *
+			 * This is the producer list, owned by the completion
+			 * handling path. When the consumer list
+			 * (dqo_tx.free_pending_packets) is runs out, this list
+			 * will be stolen.
+			 */
+			atomic_t free_pending_packets;
+
+			/* Last TX ring index fetched by HW */
+			atomic_t hw_tx_head;
+
+			u16 xsk_reorder_queue_head;
+			u16 xsk_reorder_queue_tail;
+
+			/* List to track pending packets which received a miss
+			 * completion but not a corresponding reinjection.
+			 */
+			struct gve_index_list miss_completions;
+
+			/* List to track pending packets that were completed
+			 * before receiving a valid completion because they
+			 * reached a specified timeout.
+			 */
+			struct gve_index_list timed_out_completions;
+
+			/* QPL fields */
+			struct {
+				/* Linked list of gve_tx_buf_dqo. Index into
+				 * tx_qpl_buf_next, or -1 if empty.
+				 *
+				 * This is the producer list, owned by the completion
+				 * handling path. When the consumer list
+				 * (dqo_tx.free_tx_qpl_buf_head) is runs out, this list
+				 * will be stolen.
+				 */
+				atomic_t free_tx_qpl_buf_head;
+
+				/* Free running count of the number of tx buffers
+				 * freed
+				 */
+				atomic_t free_tx_qpl_buf_cnt;
+			};
+		} dqo_compl;
+	} ____cacheline_aligned;
+	u64 pkt_done; /* free-running - total packets completed */
+	u64 bytes_done; /* free-running - total bytes completed */
+	u64 dropped_pkt; /* free-running - total packets dropped */
+	u64 dma_mapping_error; /* count of dma mapping errors */
+
+	/* Cacheline 2 -- Read-mostly fields */
+	union {
+		/* GQI fields */
+		struct {
+			union gve_tx_desc *desc;
+
+			/* Maps 1:1 to a desc */
+			struct gve_tx_buffer_state *info;
+		};
+
+		/* DQO fields. */
+		struct {
+			union gve_tx_desc_dqo *tx_ring;
+			struct gve_tx_compl_desc *compl_ring;
+
+			struct gve_tx_pending_packet_dqo *pending_packets;
+			s16 num_pending_packets;
+
+			u16 *xsk_reorder_queue;
+
+			u32 complq_mask; /* complq size is complq_mask + 1 */
+
+			/* QPL fields */
+			struct {
+				/* qpl assigned to this queue */
+				struct gve_queue_page_list *qpl;
+
+				/* Each QPL page is divided into TX bounce buffers
+				 * of size GVE_TX_BUF_SIZE_DQO. tx_qpl_buf_next is
+				 * an array to manage linked lists of TX buffers.
+				 * An entry j at index i implies that j'th buffer
+				 * is next on the list after i
+				 */
+				s16 *tx_qpl_buf_next;
+				u32 num_tx_qpl_bufs;
+			};
+		} dqo;
+	} ____cacheline_aligned;
+	struct netdev_queue *netdev_txq;
+	struct gve_queue_resources *q_resources; /* head and tail pointer idx */
+	struct device *dev;
+	u32 mask; /* masks req and done down to queue size */
+	u8 raw_addressing; /* use raw_addressing? */
+
+	/* Slow-path fields */
+	u32 q_num ____cacheline_aligned; /* queue idx */
+	u32 stop_queue; /* count of queue stops */
+	u32 wake_queue; /* count of queue wakes */
+	u32 queue_timeout; /* count of queue timeouts */
+	u32 ntfy_id; /* notification block index */
+	u32 last_kick_msec; /* Last time the queue was kicked */
+	dma_addr_t bus; /* dma address of the descr ring */
+	dma_addr_t q_resources_bus; /* dma address of the queue resources */
+	dma_addr_t complq_bus_dqo; /* dma address of the dqo.compl_ring */
+	struct u64_stats_sync statss; /* sync stats for 32bit archs */
+	struct xsk_buff_pool *xsk_pool;
+	u64 xdp_xsk_sent;
+	u64 xdp_xmit;
+	u64 xdp_xmit_errors;
+} ____cacheline_aligned;
+
+/* Wraps the info for one irq including the napi struct and the queues
+ * associated with that irq.
+ */
+struct gve_notify_block {
+	__be32 *irq_db_index; /* pointer to idx into Bar2 */
+	char name[IFNAMSIZ + 16]; /* name registered with the kernel */
+	struct napi_struct napi; /* kernel napi struct for this block */
+	struct gve_priv *priv;
+	struct gve_tx_ring *tx; /* tx rings on this block */
+	struct gve_rx_ring *rx; /* rx rings on this block */
+	u32 irq;
+};
+
+/* Tracks allowed and current rx queue settings */
+struct gve_rx_queue_config {
+	u16 max_queues;
+	u16 num_queues;
+	u16 packet_buffer_size;
+};
+
+/* Tracks allowed and current tx queue settings */
+struct gve_tx_queue_config {
+	u16 max_queues;
+	u16 num_queues; /* number of TX queues, excluding XDP queues */
+	u16 num_xdp_queues;
+};
+
+/* Tracks the available and used qpl IDs */
+struct gve_qpl_config {
+	u32 qpl_map_size; /* map memory size */
+	unsigned long *qpl_id_map; /* bitmap of used qpl ids */
+};
+
+struct gve_irq_db {
+	__be32 index;
+} ____cacheline_aligned;
+
+struct gve_ptype {
+	u8 l3_type;  /* `gve_l3_type` in gve_adminq.h */
+	u8 l4_type;  /* `gve_l4_type` in gve_adminq.h */
+};
+
+struct gve_ptype_lut {
+	struct gve_ptype ptypes[GVE_NUM_PTYPES];
+};
+
+/* Parameters for allocating resources for tx queues */
+struct gve_tx_alloc_rings_cfg {
+	struct gve_tx_queue_config *qcfg;
+
+	u16 num_xdp_rings;
+
+	u16 ring_size;
+	bool raw_addressing;
+
+	/* Allocated resources are returned here */
+	struct gve_tx_ring *tx;
+};
+
+/* Parameters for allocating resources for rx queues */
+struct gve_rx_alloc_rings_cfg {
+	/* tx config is also needed to determine QPL ids */
+	struct gve_rx_queue_config *qcfg_rx;
+	struct gve_tx_queue_config *qcfg_tx;
+
+	u16 ring_size;
+	u16 packet_buffer_size;
+	bool raw_addressing;
+	bool enable_header_split;
+	bool reset_rss;
+	bool xdp;
+
+	/* Allocated resources are returned here */
+	struct gve_rx_ring *rx;
+};
+
+/* GVE_QUEUE_FORMAT_UNSPECIFIED must be zero since 0 is the default value
+ * when the entire configure_device_resources command is zeroed out and the
+ * queue_format is not specified.
+ */
+enum gve_queue_format {
+	GVE_QUEUE_FORMAT_UNSPECIFIED	= 0x0,
+	GVE_GQI_RDA_FORMAT		= 0x1,
+	GVE_GQI_QPL_FORMAT		= 0x2,
+	GVE_DQO_RDA_FORMAT		= 0x3,
+	GVE_DQO_QPL_FORMAT		= 0x4,
+};
+
+struct gve_flow_spec {
+	__be32 src_ip[4];
+	__be32 dst_ip[4];
+	union {
+		struct {
+			__be16 src_port;
+			__be16 dst_port;
+		};
+		__be32 spi;
+	};
+	union {
+		u8 tos;
+		u8 tclass;
+	};
+};
+
+struct gve_flow_rule {
+	u32 location;
+	u16 flow_type;
+	u16 action;
+	struct gve_flow_spec key;
+	struct gve_flow_spec mask;
+};
+
+struct gve_flow_rules_cache {
+	bool rules_cache_synced; /* False if the driver's rules_cache is outdated */
+	struct gve_adminq_queried_flow_rule *rules_cache;
+	__be32 *rule_ids_cache;
+	/* The total number of queried rules that stored in the caches */
+	u32 rules_cache_num;
+	u32 rule_ids_cache_num;
+};
+
+struct gve_rss_config {
+	u8 *hash_key;
+	u32 *hash_lut;
+};
+
+struct gve_ptp {
+	struct ptp_clock_info info;
+	struct ptp_clock *clock;
+	struct gve_priv *priv;
+};
+
+struct gve_priv {
+	struct net_device *dev;
+	struct gve_tx_ring *tx; /* array of tx_cfg.num_queues */
+	struct gve_rx_ring *rx; /* array of rx_cfg.num_queues */
+	struct gve_notify_block *ntfy_blocks; /* array of num_ntfy_blks */
+	struct gve_irq_db *irq_db_indices; /* array of num_ntfy_blks */
+	dma_addr_t irq_db_indices_bus;
+	struct msix_entry *msix_vectors; /* array of num_ntfy_blks + 1 */
+	char mgmt_msix_name[IFNAMSIZ + 16];
+	u32 mgmt_msix_idx;
+	__be32 *counter_array; /* array of num_event_counters */
+	dma_addr_t counter_array_bus;
+
+	u16 num_event_counters;
+	u16 tx_desc_cnt; /* num desc per ring */
+	u16 rx_desc_cnt; /* num desc per ring */
+	u16 max_tx_desc_cnt;
+	u16 max_rx_desc_cnt;
+	u16 min_tx_desc_cnt;
+	u16 min_rx_desc_cnt;
+	bool modify_ring_size_enabled;
+	bool default_min_ring_size;
+	u16 tx_pages_per_qpl; /* Suggested number of pages per qpl for TX queues by NIC */
+	u64 max_registered_pages;
+	u64 num_registered_pages; /* num pages registered with NIC */
+	struct bpf_prog *xdp_prog; /* XDP BPF program */
+	u32 rx_copybreak; /* copy packets smaller than this */
+	u16 default_num_queues; /* default num queues to set up */
+
+	struct gve_tx_queue_config tx_cfg;
+	struct gve_rx_queue_config rx_cfg;
+	unsigned long *xsk_pools; /* bitmap of RX queues with XSK pools */
+	u32 num_ntfy_blks; /* split between TX and RX so must be even */
+	int numa_node;
+
+	struct gve_registers __iomem *reg_bar0; /* see gve_register.h */
+	__be32 __iomem *db_bar2; /* "array" of doorbells */
+	u32 msg_enable;	/* level for netif* netdev print macros	*/
+	struct pci_dev *pdev;
+
+	/* metrics */
+	u32 tx_timeo_cnt;
+
+	/* Admin queue - see gve_adminq.h*/
+	union gve_adminq_command *adminq;
+	dma_addr_t adminq_bus_addr;
+	struct dma_pool *adminq_pool;
+	struct mutex adminq_lock; /* Protects adminq command execution */
+	u32 adminq_mask; /* masks prod_cnt to adminq size */
+	u32 adminq_prod_cnt; /* free-running count of AQ cmds executed */
+	u32 adminq_cmd_fail; /* free-running count of AQ cmds failed */
+	u32 adminq_timeouts; /* free-running count of AQ cmds timeouts */
+	/* free-running count of per AQ cmd executed */
+	u32 adminq_describe_device_cnt;
+	u32 adminq_cfg_device_resources_cnt;
+	u32 adminq_register_page_list_cnt;
+	u32 adminq_unregister_page_list_cnt;
+	u32 adminq_create_tx_queue_cnt;
+	u32 adminq_create_rx_queue_cnt;
+	u32 adminq_destroy_tx_queue_cnt;
+	u32 adminq_destroy_rx_queue_cnt;
+	u32 adminq_dcfg_device_resources_cnt;
+	u32 adminq_set_driver_parameter_cnt;
+	u32 adminq_report_stats_cnt;
+	u32 adminq_report_link_speed_cnt;
+	u32 adminq_report_nic_timestamp_cnt;
+	u32 adminq_get_ptype_map_cnt;
+	u32 adminq_verify_driver_compatibility_cnt;
+	u32 adminq_query_flow_rules_cnt;
+	u32 adminq_cfg_flow_rule_cnt;
+	u32 adminq_cfg_rss_cnt;
+	u32 adminq_query_rss_cnt;
+
+	/* Global stats */
+	u32 interface_up_cnt; /* count of times interface turned up since last reset */
+	u32 interface_down_cnt; /* count of times interface turned down since last reset */
+	u32 reset_cnt; /* count of reset */
+	u32 page_alloc_fail; /* count of page alloc fails */
+	u32 dma_mapping_error; /* count of dma mapping errors */
+	u32 stats_report_trigger_cnt; /* count of device-requested stats-reports since last reset */
+	u32 suspend_cnt; /* count of times suspended */
+	u32 resume_cnt; /* count of times resumed */
+	struct workqueue_struct *gve_wq;
+	struct work_struct service_task;
+	struct work_struct stats_report_task;
+	unsigned long service_task_flags;
+	unsigned long state_flags;
+
+	struct gve_stats_report *stats_report;
+	u64 stats_report_len;
+	dma_addr_t stats_report_bus; /* dma address for the stats report */
+	unsigned long ethtool_flags;
+
+	unsigned long stats_report_timer_period;
+	struct timer_list stats_report_timer;
+
+	/* Gvnic device link speed from hypervisor. */
+	u64 link_speed;
+	bool up_before_suspend; /* True if dev was up before suspend */
+
+	struct gve_ptype_lut *ptype_lut_dqo;
+
+	/* Must be a power of two. */
+	u16 max_rx_buffer_size; /* device limit */
+
+	enum gve_queue_format queue_format;
+
+	/* Interrupt coalescing settings */
+	u32 tx_coalesce_usecs;
+	u32 rx_coalesce_usecs;
+
+	u16 header_buf_size; /* device configured, header-split supported if non-zero */
+	bool header_split_enabled; /* True if the header split is enabled by the user */
+
+	u32 max_flow_rules;
+	u32 num_flow_rules;
+
+	struct gve_flow_rules_cache flow_rules_cache;
+
+	u16 rss_key_size;
+	u16 rss_lut_size;
+	bool cache_rss_config;
+	struct gve_rss_config rss_config;
+
+	/* True if the device supports reading the nic clock */
+	bool nic_timestamp_supported;
+	struct gve_ptp *ptp;
+	struct kernel_hwtstamp_config ts_config;
+	struct gve_nic_ts_report *nic_ts_report;
+	dma_addr_t nic_ts_report_bus;
+	u64 last_sync_nic_counter; /* Clock counter from last NIC TS report */
+};
+
+enum gve_service_task_flags_bit {
+	GVE_PRIV_FLAGS_DO_RESET			= 1,
+	GVE_PRIV_FLAGS_RESET_IN_PROGRESS	= 2,
+	GVE_PRIV_FLAGS_PROBE_IN_PROGRESS	= 3,
+	GVE_PRIV_FLAGS_DO_REPORT_STATS = 4,
+};
+
+enum gve_state_flags_bit {
+	GVE_PRIV_FLAGS_ADMIN_QUEUE_OK		= 1,
+	GVE_PRIV_FLAGS_DEVICE_RESOURCES_OK	= 2,
+	GVE_PRIV_FLAGS_DEVICE_RINGS_OK		= 3,
+	GVE_PRIV_FLAGS_NAPI_ENABLED		= 4,
+};
+
+enum gve_ethtool_flags_bit {
+	GVE_PRIV_FLAGS_REPORT_STATS		= 0,
+};
+
+static inline bool gve_get_do_reset(struct gve_priv *priv)
+{
+	return test_bit(GVE_PRIV_FLAGS_DO_RESET, &priv->service_task_flags);
+}
+
+static inline void gve_set_do_reset(struct gve_priv *priv)
+{
+	set_bit(GVE_PRIV_FLAGS_DO_RESET, &priv->service_task_flags);
+}
+
+static inline void gve_clear_do_reset(struct gve_priv *priv)
+{
+	clear_bit(GVE_PRIV_FLAGS_DO_RESET, &priv->service_task_flags);
+}
+
+static inline bool gve_get_reset_in_progress(struct gve_priv *priv)
+{
+	return test_bit(GVE_PRIV_FLAGS_RESET_IN_PROGRESS,
+			&priv->service_task_flags);
+}
+
+static inline void gve_set_reset_in_progress(struct gve_priv *priv)
+{
+	set_bit(GVE_PRIV_FLAGS_RESET_IN_PROGRESS, &priv->service_task_flags);
+}
+
+static inline void gve_clear_reset_in_progress(struct gve_priv *priv)
+{
+	clear_bit(GVE_PRIV_FLAGS_RESET_IN_PROGRESS, &priv->service_task_flags);
+}
+
+static inline bool gve_get_probe_in_progress(struct gve_priv *priv)
+{
+	return test_bit(GVE_PRIV_FLAGS_PROBE_IN_PROGRESS,
+			&priv->service_task_flags);
+}
+
+static inline void gve_set_probe_in_progress(struct gve_priv *priv)
+{
+	set_bit(GVE_PRIV_FLAGS_PROBE_IN_PROGRESS, &priv->service_task_flags);
+}
+
+static inline void gve_clear_probe_in_progress(struct gve_priv *priv)
+{
+	clear_bit(GVE_PRIV_FLAGS_PROBE_IN_PROGRESS, &priv->service_task_flags);
+}
+
+static inline bool gve_get_do_report_stats(struct gve_priv *priv)
+{
+	return test_bit(GVE_PRIV_FLAGS_DO_REPORT_STATS,
+			&priv->service_task_flags);
+}
+
+static inline void gve_set_do_report_stats(struct gve_priv *priv)
+{
+	set_bit(GVE_PRIV_FLAGS_DO_REPORT_STATS, &priv->service_task_flags);
+}
+
+static inline void gve_clear_do_report_stats(struct gve_priv *priv)
+{
+	clear_bit(GVE_PRIV_FLAGS_DO_REPORT_STATS, &priv->service_task_flags);
+}
+
+static inline bool gve_get_admin_queue_ok(struct gve_priv *priv)
+{
+	return test_bit(GVE_PRIV_FLAGS_ADMIN_QUEUE_OK, &priv->state_flags);
+}
+
+static inline void gve_set_admin_queue_ok(struct gve_priv *priv)
+{
+	set_bit(GVE_PRIV_FLAGS_ADMIN_QUEUE_OK, &priv->state_flags);
+}
+
+static inline void gve_clear_admin_queue_ok(struct gve_priv *priv)
+{
+	clear_bit(GVE_PRIV_FLAGS_ADMIN_QUEUE_OK, &priv->state_flags);
+}
+
+static inline bool gve_get_device_resources_ok(struct gve_priv *priv)
+{
+	return test_bit(GVE_PRIV_FLAGS_DEVICE_RESOURCES_OK, &priv->state_flags);
+}
+
+static inline void gve_set_device_resources_ok(struct gve_priv *priv)
+{
+	set_bit(GVE_PRIV_FLAGS_DEVICE_RESOURCES_OK, &priv->state_flags);
+}
+
+static inline void gve_clear_device_resources_ok(struct gve_priv *priv)
+{
+	clear_bit(GVE_PRIV_FLAGS_DEVICE_RESOURCES_OK, &priv->state_flags);
+}
+
+static inline bool gve_get_device_rings_ok(struct gve_priv *priv)
+{
+	return test_bit(GVE_PRIV_FLAGS_DEVICE_RINGS_OK, &priv->state_flags);
+}
+
+static inline void gve_set_device_rings_ok(struct gve_priv *priv)
+{
+	set_bit(GVE_PRIV_FLAGS_DEVICE_RINGS_OK, &priv->state_flags);
+}
+
+static inline void gve_clear_device_rings_ok(struct gve_priv *priv)
+{
+	clear_bit(GVE_PRIV_FLAGS_DEVICE_RINGS_OK, &priv->state_flags);
+}
+
+static inline bool gve_get_napi_enabled(struct gve_priv *priv)
+{
+	return test_bit(GVE_PRIV_FLAGS_NAPI_ENABLED, &priv->state_flags);
+}
+
+static inline void gve_set_napi_enabled(struct gve_priv *priv)
+{
+	set_bit(GVE_PRIV_FLAGS_NAPI_ENABLED, &priv->state_flags);
+}
+
+static inline void gve_clear_napi_enabled(struct gve_priv *priv)
+{
+	clear_bit(GVE_PRIV_FLAGS_NAPI_ENABLED, &priv->state_flags);
+}
+
+static inline bool gve_get_report_stats(struct gve_priv *priv)
+{
+	return test_bit(GVE_PRIV_FLAGS_REPORT_STATS, &priv->ethtool_flags);
+}
+
+static inline void gve_clear_report_stats(struct gve_priv *priv)
+{
+	clear_bit(GVE_PRIV_FLAGS_REPORT_STATS, &priv->ethtool_flags);
+}
+
+/* Returns the address of the ntfy_blocks irq doorbell
+ */
+static inline __be32 __iomem *gve_irq_doorbell(struct gve_priv *priv,
+					       struct gve_notify_block *block)
+{
+	return &priv->db_bar2[be32_to_cpu(*block->irq_db_index)];
+}
+
+/* Returns the index into ntfy_blocks of the given tx ring's block
+ */
+static inline u32 gve_tx_idx_to_ntfy(struct gve_priv *priv, u32 queue_idx)
+{
+	return queue_idx;
+}
+
+/* Returns the index into ntfy_blocks of the given rx ring's block
+ */
+static inline u32 gve_rx_idx_to_ntfy(struct gve_priv *priv, u32 queue_idx)
+{
+	return (priv->num_ntfy_blks / 2) + queue_idx;
+}
+
+static inline bool gve_is_qpl(struct gve_priv *priv)
+{
+	return priv->queue_format == GVE_GQI_QPL_FORMAT ||
+		priv->queue_format == GVE_DQO_QPL_FORMAT;
+}
+
+/* Returns the number of tx queue page lists */
+static inline u32 gve_num_tx_qpls(const struct gve_tx_queue_config *tx_cfg,
+				  bool is_qpl)
+{
+	if (!is_qpl)
+		return 0;
+	return tx_cfg->num_queues + tx_cfg->num_xdp_queues;
+}
+
+/* Returns the number of rx queue page lists */
+static inline u32 gve_num_rx_qpls(const struct gve_rx_queue_config *rx_cfg,
+				  bool is_qpl)
+{
+	if (!is_qpl)
+		return 0;
+	return rx_cfg->num_queues;
+}
+
+static inline u32 gve_tx_qpl_id(struct gve_priv *priv, int tx_qid)
+{
+	return tx_qid;
+}
+
+static inline u32 gve_rx_qpl_id(struct gve_priv *priv, int rx_qid)
+{
+	return priv->tx_cfg.max_queues + rx_qid;
+}
+
+static inline u32 gve_get_rx_qpl_id(const struct gve_tx_queue_config *tx_cfg,
+				    int rx_qid)
+{
+	return tx_cfg->max_queues + rx_qid;
+}
+
+static inline u32 gve_tx_start_qpl_id(struct gve_priv *priv)
+{
+	return gve_tx_qpl_id(priv, 0);
+}
+
+static inline u32 gve_rx_start_qpl_id(const struct gve_tx_queue_config *tx_cfg)
+{
+	return gve_get_rx_qpl_id(tx_cfg, 0);
+}
+
+static inline u32 gve_get_rx_pages_per_qpl_dqo(u32 rx_desc_cnt)
+{
+	/* For DQO, page count should be more than ring size for
+	 * out-of-order completions. Set it to two times of ring size.
+	 */
+	return 2 * rx_desc_cnt;
+}
+
+/* Returns the correct dma direction for tx and rx qpls */
+static inline enum dma_data_direction gve_qpl_dma_dir(struct gve_priv *priv,
+						      int id)
+{
+	if (id < gve_rx_start_qpl_id(&priv->tx_cfg))
+		return DMA_TO_DEVICE;
+	else
+		return DMA_FROM_DEVICE;
+}
+
+static inline bool gve_is_gqi(struct gve_priv *priv)
+{
+	return priv->queue_format == GVE_GQI_RDA_FORMAT ||
+		priv->queue_format == GVE_GQI_QPL_FORMAT;
+}
+
+static inline u32 gve_num_tx_queues(struct gve_priv *priv)
+{
+	return priv->tx_cfg.num_queues + priv->tx_cfg.num_xdp_queues;
+}
+
+static inline u32 gve_xdp_tx_queue_id(struct gve_priv *priv, u32 queue_id)
+{
+	return priv->tx_cfg.num_queues + queue_id;
+}
+
+static inline u32 gve_xdp_tx_start_queue_id(struct gve_priv *priv)
+{
+	return gve_xdp_tx_queue_id(priv, 0);
+}
+
+static inline bool gve_supports_xdp_xmit(struct gve_priv *priv)
+{
+	switch (priv->queue_format) {
+	case GVE_GQI_QPL_FORMAT:
+	case GVE_DQO_RDA_FORMAT:
+		return true;
+	default:
+		return false;
+	}
+}
+
+/* gqi napi handler defined in gve_main.c */
+int gve_napi_poll(struct napi_struct *napi, int budget);
+
+/* buffers */
+int gve_alloc_page(struct gve_priv *priv, struct device *dev,
+		   struct page **page, dma_addr_t *dma,
+		   enum dma_data_direction, gfp_t gfp_flags);
+void gve_free_page(struct device *dev, struct page *page, dma_addr_t dma,
+		   enum dma_data_direction);
+/* qpls */
+struct gve_queue_page_list *gve_alloc_queue_page_list(struct gve_priv *priv,
+						      u32 id, int pages);
+void gve_free_queue_page_list(struct gve_priv *priv,
+			      struct gve_queue_page_list *qpl,
+			      u32 id);
+/* tx handling */
+netdev_tx_t gve_tx(struct sk_buff *skb, struct net_device *dev);
+int gve_xdp_xmit_gqi(struct net_device *dev, int n, struct xdp_frame **frames,
+		     u32 flags);
+int gve_xdp_xmit_dqo(struct net_device *dev, int n, struct xdp_frame **frames,
+		     u32 flags);
+int gve_xdp_xmit_one(struct gve_priv *priv, struct gve_tx_ring *tx,
+		     void *data, int len, void *frame_p);
+void gve_xdp_tx_flush(struct gve_priv *priv, u32 xdp_qid);
+int gve_xdp_xmit_one_dqo(struct gve_priv *priv, struct gve_tx_ring *tx,
+			 struct xdp_frame *xdpf);
+bool gve_tx_poll(struct gve_notify_block *block, int budget);
+bool gve_xdp_poll(struct gve_notify_block *block, int budget);
+int gve_xsk_tx_poll(struct gve_notify_block *block, int budget);
+int gve_tx_alloc_rings_gqi(struct gve_priv *priv,
+			   struct gve_tx_alloc_rings_cfg *cfg);
+void gve_tx_free_rings_gqi(struct gve_priv *priv,
+			   struct gve_tx_alloc_rings_cfg *cfg);
+void gve_tx_start_ring_gqi(struct gve_priv *priv, int idx);
+void gve_tx_stop_ring_gqi(struct gve_priv *priv, int idx);
+u32 gve_tx_load_event_counter(struct gve_priv *priv,
+			      struct gve_tx_ring *tx);
+bool gve_tx_clean_pending(struct gve_priv *priv, struct gve_tx_ring *tx);
+/* rx handling */
+void gve_rx_write_doorbell(struct gve_priv *priv, struct gve_rx_ring *rx);
+int gve_rx_poll(struct gve_notify_block *block, int budget);
+bool gve_rx_work_pending(struct gve_rx_ring *rx);
+int gve_rx_alloc_ring_gqi(struct gve_priv *priv,
+			  struct gve_rx_alloc_rings_cfg *cfg,
+			  struct gve_rx_ring *rx,
+			  int idx);
+void gve_rx_free_ring_gqi(struct gve_priv *priv, struct gve_rx_ring *rx,
+			  struct gve_rx_alloc_rings_cfg *cfg);
+int gve_rx_alloc_rings_gqi(struct gve_priv *priv,
+			   struct gve_rx_alloc_rings_cfg *cfg);
+void gve_rx_free_rings_gqi(struct gve_priv *priv,
+			   struct gve_rx_alloc_rings_cfg *cfg);
+void gve_rx_start_ring_gqi(struct gve_priv *priv, int idx);
+void gve_rx_stop_ring_gqi(struct gve_priv *priv, int idx);
+u16 gve_get_pkt_buf_size(const struct gve_priv *priv, bool enable_hplit);
+bool gve_header_split_supported(const struct gve_priv *priv);
+int gve_set_hsplit_config(struct gve_priv *priv, u8 tcp_data_split);
+/* rx buffer handling */
+int gve_buf_ref_cnt(struct gve_rx_buf_state_dqo *bs);
+void gve_free_page_dqo(struct gve_priv *priv, struct gve_rx_buf_state_dqo *bs,
+		       bool free_page);
+struct gve_rx_buf_state_dqo *gve_alloc_buf_state(struct gve_rx_ring *rx);
+bool gve_buf_state_is_allocated(struct gve_rx_ring *rx,
+				struct gve_rx_buf_state_dqo *buf_state);
+void gve_free_buf_state(struct gve_rx_ring *rx,
+			struct gve_rx_buf_state_dqo *buf_state);
+struct gve_rx_buf_state_dqo *gve_dequeue_buf_state(struct gve_rx_ring *rx,
+						   struct gve_index_list *list);
+void gve_enqueue_buf_state(struct gve_rx_ring *rx, struct gve_index_list *list,
+			   struct gve_rx_buf_state_dqo *buf_state);
+struct gve_rx_buf_state_dqo *gve_get_recycled_buf_state(struct gve_rx_ring *rx);
+void gve_try_recycle_buf(struct gve_priv *priv, struct gve_rx_ring *rx,
+			 struct gve_rx_buf_state_dqo *buf_state);
+void gve_free_to_page_pool(struct gve_rx_ring *rx,
+			   struct gve_rx_buf_state_dqo *buf_state,
+			   bool allow_direct);
+int gve_alloc_qpl_page_dqo(struct gve_rx_ring *rx,
+			   struct gve_rx_buf_state_dqo *buf_state);
+void gve_free_qpl_page_dqo(struct gve_rx_buf_state_dqo *buf_state);
+void gve_reuse_buffer(struct gve_rx_ring *rx,
+		      struct gve_rx_buf_state_dqo *buf_state);
+void gve_free_buffer(struct gve_rx_ring *rx,
+		     struct gve_rx_buf_state_dqo *buf_state);
+int gve_alloc_buffer(struct gve_rx_ring *rx, struct gve_rx_desc_dqo *desc);
+struct page_pool *gve_rx_create_page_pool(struct gve_priv *priv,
+					  struct gve_rx_ring *rx,
+					  bool xdp);
+
+/* Reset */
+void gve_schedule_reset(struct gve_priv *priv);
+int gve_reset(struct gve_priv *priv, bool attempt_teardown);
+void gve_get_curr_alloc_cfgs(struct gve_priv *priv,
+			     struct gve_tx_alloc_rings_cfg *tx_alloc_cfg,
+			     struct gve_rx_alloc_rings_cfg *rx_alloc_cfg);
+int gve_adjust_config(struct gve_priv *priv,
+		      struct gve_tx_alloc_rings_cfg *tx_alloc_cfg,
+		      struct gve_rx_alloc_rings_cfg *rx_alloc_cfg);
+int gve_adjust_queues(struct gve_priv *priv,
+		      struct gve_rx_queue_config new_rx_config,
+		      struct gve_tx_queue_config new_tx_config,
+		      bool reset_rss);
+/* flow steering rule */
+int gve_get_flow_rule_entry(struct gve_priv *priv, struct ethtool_rxnfc *cmd);
+int gve_get_flow_rule_ids(struct gve_priv *priv, struct ethtool_rxnfc *cmd, u32 *rule_locs);
+int gve_add_flow_rule(struct gve_priv *priv, struct ethtool_rxnfc *cmd);
+int gve_del_flow_rule(struct gve_priv *priv, struct ethtool_rxnfc *cmd);
+int gve_flow_rules_reset(struct gve_priv *priv);
+/* RSS config */
+int gve_init_rss_config(struct gve_priv *priv, u16 num_queues);
+/* PTP and timestamping */
+#if IS_ENABLED(CONFIG_PTP_1588_CLOCK)
+int gve_clock_nic_ts_read(struct gve_priv *priv);
+int gve_init_clock(struct gve_priv *priv);
+void gve_teardown_clock(struct gve_priv *priv);
+#else /* CONFIG_PTP_1588_CLOCK */
+static inline int gve_clock_nic_ts_read(struct gve_priv *priv)
+{
+	return -EOPNOTSUPP;
+}
+
+static inline int gve_init_clock(struct gve_priv *priv)
+{
+	return 0;
+}
+
+static inline void gve_teardown_clock(struct gve_priv *priv) { }
+#endif /* CONFIG_PTP_1588_CLOCK */
+/* report stats handling */
+void gve_handle_report_stats(struct gve_priv *priv);
+/* exported by ethtool.c */
+extern const struct ethtool_ops gve_ethtool_ops;
+/* needed by ethtool */
+extern char gve_driver_name[];
+extern const char gve_version_str[];
+#endif /* _GVE_H_ */
diff --git a/drivers/net/ethernet/google/gve/gve_adminq.c b/drivers/net/ethernet/google/gve/gve_adminq.c
new file mode 100644
index 000000000000..4f33d094a2ef
--- /dev/null
+++ b/drivers/net/ethernet/google/gve/gve_adminq.c
@@ -0,0 +1,1604 @@
+// SPDX-License-Identifier: (GPL-2.0 OR MIT)
+/* Google virtual Ethernet (gve) driver
+ *
+ * Copyright (C) 2015-2021 Google, Inc.
+ */
+
+#include <linux/etherdevice.h>
+#include <linux/pci.h>
+#include "gve.h"
+#include "gve_adminq.h"
+#include "gve_register.h"
+
+#define GVE_MAX_ADMINQ_RELEASE_CHECK	500
+#define GVE_ADMINQ_SLEEP_LEN		20
+#define GVE_MAX_ADMINQ_EVENT_COUNTER_CHECK	100
+
+#define GVE_DEVICE_OPTION_ERROR_FMT "%s option error:\n" \
+"Expected: length=%d, feature_mask=%x.\n" \
+"Actual: length=%d, feature_mask=%x.\n"
+
+#define GVE_DEVICE_OPTION_TOO_BIG_FMT "Length of %s option larger than expected. Possible older version of guest driver.\n"
+
+static
+struct gve_device_option *gve_get_next_option(struct gve_device_descriptor *descriptor,
+					      struct gve_device_option *option)
+{
+	void *option_end, *descriptor_end;
+
+	option_end = (void *)(option + 1) + be16_to_cpu(option->option_length);
+	descriptor_end = (void *)descriptor + be16_to_cpu(descriptor->total_length);
+
+	return option_end > descriptor_end ? NULL : (struct gve_device_option *)option_end;
+}
+
+#define GVE_DEVICE_OPTION_NO_MIN_RING_SIZE	8
+
+static
+void gve_parse_device_option(struct gve_priv *priv,
+			     struct gve_device_descriptor *device_descriptor,
+			     struct gve_device_option *option,
+			     struct gve_device_option_gqi_rda **dev_op_gqi_rda,
+			     struct gve_device_option_gqi_qpl **dev_op_gqi_qpl,
+			     struct gve_device_option_dqo_rda **dev_op_dqo_rda,
+			     struct gve_device_option_jumbo_frames **dev_op_jumbo_frames,
+			     struct gve_device_option_dqo_qpl **dev_op_dqo_qpl,
+			     struct gve_device_option_buffer_sizes **dev_op_buffer_sizes,
+			     struct gve_device_option_flow_steering **dev_op_flow_steering,
+			     struct gve_device_option_rss_config **dev_op_rss_config,
+			     struct gve_device_option_nic_timestamp **dev_op_nic_timestamp,
+			     struct gve_device_option_modify_ring **dev_op_modify_ring)
+{
+	u32 req_feat_mask = be32_to_cpu(option->required_features_mask);
+	u16 option_length = be16_to_cpu(option->option_length);
+	u16 option_id = be16_to_cpu(option->option_id);
+
+	/* If the length or feature mask doesn't match, continue without
+	 * enabling the feature.
+	 */
+	switch (option_id) {
+	case GVE_DEV_OPT_ID_GQI_RAW_ADDRESSING:
+		if (option_length != GVE_DEV_OPT_LEN_GQI_RAW_ADDRESSING ||
+		    req_feat_mask != GVE_DEV_OPT_REQ_FEAT_MASK_GQI_RAW_ADDRESSING) {
+			dev_warn(&priv->pdev->dev, GVE_DEVICE_OPTION_ERROR_FMT,
+				 "Raw Addressing",
+				 GVE_DEV_OPT_LEN_GQI_RAW_ADDRESSING,
+				 GVE_DEV_OPT_REQ_FEAT_MASK_GQI_RAW_ADDRESSING,
+				 option_length, req_feat_mask);
+			break;
+		}
+
+		dev_info(&priv->pdev->dev,
+			 "Gqi raw addressing device option enabled.\n");
+		priv->queue_format = GVE_GQI_RDA_FORMAT;
+		break;
+	case GVE_DEV_OPT_ID_GQI_RDA:
+		if (option_length < sizeof(**dev_op_gqi_rda) ||
+		    req_feat_mask != GVE_DEV_OPT_REQ_FEAT_MASK_GQI_RDA) {
+			dev_warn(&priv->pdev->dev, GVE_DEVICE_OPTION_ERROR_FMT,
+				 "GQI RDA", (int)sizeof(**dev_op_gqi_rda),
+				 GVE_DEV_OPT_REQ_FEAT_MASK_GQI_RDA,
+				 option_length, req_feat_mask);
+			break;
+		}
+
+		if (option_length > sizeof(**dev_op_gqi_rda)) {
+			dev_warn(&priv->pdev->dev,
+				 GVE_DEVICE_OPTION_TOO_BIG_FMT, "GQI RDA");
+		}
+		*dev_op_gqi_rda = (void *)(option + 1);
+		break;
+	case GVE_DEV_OPT_ID_GQI_QPL:
+		if (option_length < sizeof(**dev_op_gqi_qpl) ||
+		    req_feat_mask != GVE_DEV_OPT_REQ_FEAT_MASK_GQI_QPL) {
+			dev_warn(&priv->pdev->dev, GVE_DEVICE_OPTION_ERROR_FMT,
+				 "GQI QPL", (int)sizeof(**dev_op_gqi_qpl),
+				 GVE_DEV_OPT_REQ_FEAT_MASK_GQI_QPL,
+				 option_length, req_feat_mask);
+			break;
+		}
+
+		if (option_length > sizeof(**dev_op_gqi_qpl)) {
+			dev_warn(&priv->pdev->dev,
+				 GVE_DEVICE_OPTION_TOO_BIG_FMT, "GQI QPL");
+		}
+		*dev_op_gqi_qpl = (void *)(option + 1);
+		break;
+	case GVE_DEV_OPT_ID_DQO_RDA:
+		if (option_length < sizeof(**dev_op_dqo_rda) ||
+		    req_feat_mask != GVE_DEV_OPT_REQ_FEAT_MASK_DQO_RDA) {
+			dev_warn(&priv->pdev->dev, GVE_DEVICE_OPTION_ERROR_FMT,
+				 "DQO RDA", (int)sizeof(**dev_op_dqo_rda),
+				 GVE_DEV_OPT_REQ_FEAT_MASK_DQO_RDA,
+				 option_length, req_feat_mask);
+			break;
+		}
+
+		if (option_length > sizeof(**dev_op_dqo_rda)) {
+			dev_warn(&priv->pdev->dev,
+				 GVE_DEVICE_OPTION_TOO_BIG_FMT, "DQO RDA");
+		}
+		*dev_op_dqo_rda = (void *)(option + 1);
+		break;
+	case GVE_DEV_OPT_ID_DQO_QPL:
+		if (option_length < sizeof(**dev_op_dqo_qpl) ||
+		    req_feat_mask != GVE_DEV_OPT_REQ_FEAT_MASK_DQO_QPL) {
+			dev_warn(&priv->pdev->dev, GVE_DEVICE_OPTION_ERROR_FMT,
+				 "DQO QPL", (int)sizeof(**dev_op_dqo_qpl),
+				 GVE_DEV_OPT_REQ_FEAT_MASK_DQO_QPL,
+				 option_length, req_feat_mask);
+			break;
+		}
+
+		if (option_length > sizeof(**dev_op_dqo_qpl)) {
+			dev_warn(&priv->pdev->dev,
+				 GVE_DEVICE_OPTION_TOO_BIG_FMT, "DQO QPL");
+		}
+		*dev_op_dqo_qpl = (void *)(option + 1);
+		break;
+	case GVE_DEV_OPT_ID_JUMBO_FRAMES:
+		if (option_length < sizeof(**dev_op_jumbo_frames) ||
+		    req_feat_mask != GVE_DEV_OPT_REQ_FEAT_MASK_JUMBO_FRAMES) {
+			dev_warn(&priv->pdev->dev, GVE_DEVICE_OPTION_ERROR_FMT,
+				 "Jumbo Frames",
+				 (int)sizeof(**dev_op_jumbo_frames),
+				 GVE_DEV_OPT_REQ_FEAT_MASK_JUMBO_FRAMES,
+				 option_length, req_feat_mask);
+			break;
+		}
+
+		if (option_length > sizeof(**dev_op_jumbo_frames)) {
+			dev_warn(&priv->pdev->dev,
+				 GVE_DEVICE_OPTION_TOO_BIG_FMT,
+				 "Jumbo Frames");
+		}
+		*dev_op_jumbo_frames = (void *)(option + 1);
+		break;
+	case GVE_DEV_OPT_ID_BUFFER_SIZES:
+		if (option_length < sizeof(**dev_op_buffer_sizes) ||
+		    req_feat_mask != GVE_DEV_OPT_REQ_FEAT_MASK_BUFFER_SIZES) {
+			dev_warn(&priv->pdev->dev, GVE_DEVICE_OPTION_ERROR_FMT,
+				 "Buffer Sizes",
+				 (int)sizeof(**dev_op_buffer_sizes),
+				 GVE_DEV_OPT_REQ_FEAT_MASK_BUFFER_SIZES,
+				 option_length, req_feat_mask);
+			break;
+		}
+
+		if (option_length > sizeof(**dev_op_buffer_sizes))
+			dev_warn(&priv->pdev->dev,
+				 GVE_DEVICE_OPTION_TOO_BIG_FMT,
+				 "Buffer Sizes");
+		*dev_op_buffer_sizes = (void *)(option + 1);
+		break;
+	case GVE_DEV_OPT_ID_MODIFY_RING:
+		if (option_length < GVE_DEVICE_OPTION_NO_MIN_RING_SIZE ||
+		    req_feat_mask != GVE_DEV_OPT_REQ_FEAT_MASK_MODIFY_RING) {
+			dev_warn(&priv->pdev->dev, GVE_DEVICE_OPTION_ERROR_FMT,
+				 "Modify Ring", (int)sizeof(**dev_op_modify_ring),
+				 GVE_DEV_OPT_REQ_FEAT_MASK_MODIFY_RING,
+				 option_length, req_feat_mask);
+			break;
+		}
+
+		if (option_length > sizeof(**dev_op_modify_ring)) {
+			dev_warn(&priv->pdev->dev,
+				 GVE_DEVICE_OPTION_TOO_BIG_FMT, "Modify Ring");
+		}
+
+		*dev_op_modify_ring = (void *)(option + 1);
+
+		/* device has not provided min ring size */
+		if (option_length == GVE_DEVICE_OPTION_NO_MIN_RING_SIZE)
+			priv->default_min_ring_size = true;
+		break;
+	case GVE_DEV_OPT_ID_FLOW_STEERING:
+		if (option_length < sizeof(**dev_op_flow_steering) ||
+		    req_feat_mask != GVE_DEV_OPT_REQ_FEAT_MASK_FLOW_STEERING) {
+			dev_warn(&priv->pdev->dev, GVE_DEVICE_OPTION_ERROR_FMT,
+				 "Flow Steering",
+				 (int)sizeof(**dev_op_flow_steering),
+				 GVE_DEV_OPT_REQ_FEAT_MASK_FLOW_STEERING,
+				 option_length, req_feat_mask);
+			break;
+		}
+
+		if (option_length > sizeof(**dev_op_flow_steering))
+			dev_warn(&priv->pdev->dev,
+				 GVE_DEVICE_OPTION_TOO_BIG_FMT,
+				 "Flow Steering");
+		*dev_op_flow_steering = (void *)(option + 1);
+		break;
+	case GVE_DEV_OPT_ID_RSS_CONFIG:
+		if (option_length < sizeof(**dev_op_rss_config) ||
+		    req_feat_mask != GVE_DEV_OPT_REQ_FEAT_MASK_RSS_CONFIG) {
+			dev_warn(&priv->pdev->dev, GVE_DEVICE_OPTION_ERROR_FMT,
+				 "RSS config",
+				 (int)sizeof(**dev_op_rss_config),
+				 GVE_DEV_OPT_REQ_FEAT_MASK_RSS_CONFIG,
+				 option_length, req_feat_mask);
+			break;
+		}
+
+		if (option_length > sizeof(**dev_op_rss_config))
+			dev_warn(&priv->pdev->dev,
+				 GVE_DEVICE_OPTION_TOO_BIG_FMT,
+				 "RSS config");
+		*dev_op_rss_config = (void *)(option + 1);
+		break;
+	case GVE_DEV_OPT_ID_NIC_TIMESTAMP:
+		if (option_length < sizeof(**dev_op_nic_timestamp) ||
+		    req_feat_mask != GVE_DEV_OPT_REQ_FEAT_MASK_NIC_TIMESTAMP) {
+			dev_warn(&priv->pdev->dev, GVE_DEVICE_OPTION_ERROR_FMT,
+				 "Nic Timestamp",
+				 (int)sizeof(**dev_op_nic_timestamp),
+				 GVE_DEV_OPT_REQ_FEAT_MASK_NIC_TIMESTAMP,
+				 option_length, req_feat_mask);
+			break;
+		}
+
+		if (option_length > sizeof(**dev_op_nic_timestamp))
+			dev_warn(&priv->pdev->dev,
+				 GVE_DEVICE_OPTION_TOO_BIG_FMT,
+				 "Nic Timestamp");
+		*dev_op_nic_timestamp = (void *)(option + 1);
+		break;
+	default:
+		/* If we don't recognize the option just continue
+		 * without doing anything.
+		 */
+		dev_dbg(&priv->pdev->dev, "Unrecognized device option 0x%hx not enabled.\n",
+			option_id);
+	}
+}
+
+/* Process all device options for a given describe device call. */
+static int
+gve_process_device_options(struct gve_priv *priv,
+			   struct gve_device_descriptor *descriptor,
+			   struct gve_device_option_gqi_rda **dev_op_gqi_rda,
+			   struct gve_device_option_gqi_qpl **dev_op_gqi_qpl,
+			   struct gve_device_option_dqo_rda **dev_op_dqo_rda,
+			   struct gve_device_option_jumbo_frames **dev_op_jumbo_frames,
+			   struct gve_device_option_dqo_qpl **dev_op_dqo_qpl,
+			   struct gve_device_option_buffer_sizes **dev_op_buffer_sizes,
+			   struct gve_device_option_flow_steering **dev_op_flow_steering,
+			   struct gve_device_option_rss_config **dev_op_rss_config,
+			   struct gve_device_option_nic_timestamp **dev_op_nic_timestamp,
+			   struct gve_device_option_modify_ring **dev_op_modify_ring)
+{
+	const int num_options = be16_to_cpu(descriptor->num_device_options);
+	struct gve_device_option *dev_opt;
+	int i;
+
+	/* The options struct directly follows the device descriptor. */
+	dev_opt = (void *)(descriptor + 1);
+	for (i = 0; i < num_options; i++) {
+		struct gve_device_option *next_opt;
+
+		next_opt = gve_get_next_option(descriptor, dev_opt);
+		if (!next_opt) {
+			dev_err(&priv->dev->dev,
+				"options exceed device_descriptor's total length.\n");
+			return -EINVAL;
+		}
+
+		gve_parse_device_option(priv, descriptor, dev_opt,
+					dev_op_gqi_rda, dev_op_gqi_qpl,
+					dev_op_dqo_rda, dev_op_jumbo_frames,
+					dev_op_dqo_qpl, dev_op_buffer_sizes,
+					dev_op_flow_steering, dev_op_rss_config,
+					dev_op_nic_timestamp,
+					dev_op_modify_ring);
+		dev_opt = next_opt;
+	}
+
+	return 0;
+}
+
+int gve_adminq_alloc(struct device *dev, struct gve_priv *priv)
+{
+	priv->adminq_pool = dma_pool_create("adminq_pool", dev,
+					    GVE_ADMINQ_BUFFER_SIZE, 0, 0);
+	if (unlikely(!priv->adminq_pool))
+		return -ENOMEM;
+	priv->adminq = dma_pool_alloc(priv->adminq_pool, GFP_KERNEL,
+				      &priv->adminq_bus_addr);
+	if (unlikely(!priv->adminq)) {
+		dma_pool_destroy(priv->adminq_pool);
+		return -ENOMEM;
+	}
+
+	priv->adminq_mask =
+		(GVE_ADMINQ_BUFFER_SIZE / sizeof(union gve_adminq_command)) - 1;
+	priv->adminq_prod_cnt = 0;
+	priv->adminq_cmd_fail = 0;
+	priv->adminq_timeouts = 0;
+	priv->adminq_describe_device_cnt = 0;
+	priv->adminq_cfg_device_resources_cnt = 0;
+	priv->adminq_register_page_list_cnt = 0;
+	priv->adminq_unregister_page_list_cnt = 0;
+	priv->adminq_create_tx_queue_cnt = 0;
+	priv->adminq_create_rx_queue_cnt = 0;
+	priv->adminq_destroy_tx_queue_cnt = 0;
+	priv->adminq_destroy_rx_queue_cnt = 0;
+	priv->adminq_dcfg_device_resources_cnt = 0;
+	priv->adminq_set_driver_parameter_cnt = 0;
+	priv->adminq_report_stats_cnt = 0;
+	priv->adminq_report_link_speed_cnt = 0;
+	priv->adminq_report_nic_timestamp_cnt = 0;
+	priv->adminq_get_ptype_map_cnt = 0;
+	priv->adminq_query_flow_rules_cnt = 0;
+	priv->adminq_cfg_flow_rule_cnt = 0;
+	priv->adminq_cfg_rss_cnt = 0;
+	priv->adminq_query_rss_cnt = 0;
+
+	/* Setup Admin queue with the device */
+	if (priv->pdev->revision < 0x1) {
+		iowrite32be(priv->adminq_bus_addr / PAGE_SIZE,
+			    &priv->reg_bar0->adminq_pfn);
+	} else {
+		iowrite16be(GVE_ADMINQ_BUFFER_SIZE,
+			    &priv->reg_bar0->adminq_length);
+#ifdef CONFIG_ARCH_DMA_ADDR_T_64BIT
+		iowrite32be(priv->adminq_bus_addr >> 32,
+			    &priv->reg_bar0->adminq_base_address_hi);
+#endif
+		iowrite32be(priv->adminq_bus_addr,
+			    &priv->reg_bar0->adminq_base_address_lo);
+		iowrite32be(GVE_DRIVER_STATUS_RUN_MASK, &priv->reg_bar0->driver_status);
+	}
+	mutex_init(&priv->adminq_lock);
+	gve_set_admin_queue_ok(priv);
+	return 0;
+}
+
+void gve_adminq_release(struct gve_priv *priv)
+{
+	int i = 0;
+
+	/* Tell the device the adminq is leaving */
+	if (priv->pdev->revision < 0x1) {
+		iowrite32be(0x0, &priv->reg_bar0->adminq_pfn);
+		while (ioread32be(&priv->reg_bar0->adminq_pfn)) {
+			/* If this is reached the device is unrecoverable and still
+			 * holding memory. Continue looping to avoid memory corruption,
+			 * but WARN so it is visible what is going on.
+			 */
+			if (i == GVE_MAX_ADMINQ_RELEASE_CHECK)
+				WARN(1, "Unrecoverable platform error!");
+			i++;
+			msleep(GVE_ADMINQ_SLEEP_LEN);
+		}
+	} else {
+		iowrite32be(GVE_DRIVER_STATUS_RESET_MASK, &priv->reg_bar0->driver_status);
+		while (!(ioread32be(&priv->reg_bar0->device_status)
+				& GVE_DEVICE_STATUS_DEVICE_IS_RESET)) {
+			if (i == GVE_MAX_ADMINQ_RELEASE_CHECK)
+				WARN(1, "Unrecoverable platform error!");
+			i++;
+			msleep(GVE_ADMINQ_SLEEP_LEN);
+		}
+	}
+	gve_clear_device_rings_ok(priv);
+	gve_clear_device_resources_ok(priv);
+	gve_clear_admin_queue_ok(priv);
+}
+
+void gve_adminq_free(struct device *dev, struct gve_priv *priv)
+{
+	if (!gve_get_admin_queue_ok(priv))
+		return;
+	gve_adminq_release(priv);
+	dma_pool_free(priv->adminq_pool, priv->adminq, priv->adminq_bus_addr);
+	dma_pool_destroy(priv->adminq_pool);
+	gve_clear_admin_queue_ok(priv);
+}
+
+static void gve_adminq_kick_cmd(struct gve_priv *priv, u32 prod_cnt)
+{
+	iowrite32be(prod_cnt, &priv->reg_bar0->adminq_doorbell);
+}
+
+static bool gve_adminq_wait_for_cmd(struct gve_priv *priv, u32 prod_cnt)
+{
+	int i;
+
+	for (i = 0; i < GVE_MAX_ADMINQ_EVENT_COUNTER_CHECK; i++) {
+		if (ioread32be(&priv->reg_bar0->adminq_event_counter)
+		    == prod_cnt)
+			return true;
+		msleep(GVE_ADMINQ_SLEEP_LEN);
+	}
+
+	return false;
+}
+
+static int gve_adminq_parse_err(struct gve_priv *priv, u32 status)
+{
+	if (status != GVE_ADMINQ_COMMAND_PASSED &&
+	    status != GVE_ADMINQ_COMMAND_UNSET) {
+		dev_err(&priv->pdev->dev, "AQ command failed with status %d\n", status);
+		priv->adminq_cmd_fail++;
+	}
+	switch (status) {
+	case GVE_ADMINQ_COMMAND_PASSED:
+		return 0;
+	case GVE_ADMINQ_COMMAND_UNSET:
+		dev_err(&priv->pdev->dev, "parse_aq_err: err and status both unset, this should not be possible.\n");
+		return -EINVAL;
+	case GVE_ADMINQ_COMMAND_ERROR_ABORTED:
+	case GVE_ADMINQ_COMMAND_ERROR_CANCELLED:
+	case GVE_ADMINQ_COMMAND_ERROR_DATALOSS:
+	case GVE_ADMINQ_COMMAND_ERROR_FAILED_PRECONDITION:
+	case GVE_ADMINQ_COMMAND_ERROR_UNAVAILABLE:
+		return -EAGAIN;
+	case GVE_ADMINQ_COMMAND_ERROR_ALREADY_EXISTS:
+	case GVE_ADMINQ_COMMAND_ERROR_INTERNAL_ERROR:
+	case GVE_ADMINQ_COMMAND_ERROR_INVALID_ARGUMENT:
+	case GVE_ADMINQ_COMMAND_ERROR_NOT_FOUND:
+	case GVE_ADMINQ_COMMAND_ERROR_OUT_OF_RANGE:
+	case GVE_ADMINQ_COMMAND_ERROR_UNKNOWN_ERROR:
+		return -EINVAL;
+	case GVE_ADMINQ_COMMAND_ERROR_DEADLINE_EXCEEDED:
+		return -ETIME;
+	case GVE_ADMINQ_COMMAND_ERROR_PERMISSION_DENIED:
+	case GVE_ADMINQ_COMMAND_ERROR_UNAUTHENTICATED:
+		return -EACCES;
+	case GVE_ADMINQ_COMMAND_ERROR_RESOURCE_EXHAUSTED:
+		return -ENOMEM;
+	case GVE_ADMINQ_COMMAND_ERROR_UNIMPLEMENTED:
+		return -EOPNOTSUPP;
+	default:
+		dev_err(&priv->pdev->dev, "parse_aq_err: unknown status code %d\n", status);
+		return -EINVAL;
+	}
+}
+
+/* Flushes all AQ commands currently queued and waits for them to complete.
+ * If there are failures, it will return the first error.
+ */
+static int gve_adminq_kick_and_wait(struct gve_priv *priv)
+{
+	int tail, head;
+	int i;
+
+	lockdep_assert_held(&priv->adminq_lock);
+
+	tail = ioread32be(&priv->reg_bar0->adminq_event_counter);
+	head = priv->adminq_prod_cnt;
+
+	gve_adminq_kick_cmd(priv, head);
+	if (!gve_adminq_wait_for_cmd(priv, head)) {
+		dev_err(&priv->pdev->dev, "AQ commands timed out, need to reset AQ\n");
+		priv->adminq_timeouts++;
+		return -ENOTRECOVERABLE;
+	}
+
+	for (i = tail; i < head; i++) {
+		union gve_adminq_command *cmd;
+		u32 status, err;
+
+		cmd = &priv->adminq[i & priv->adminq_mask];
+		status = be32_to_cpu(READ_ONCE(cmd->status));
+		err = gve_adminq_parse_err(priv, status);
+		if (err)
+			// Return the first error if we failed.
+			return err;
+	}
+
+	return 0;
+}
+
+static int gve_adminq_issue_cmd(struct gve_priv *priv,
+				union gve_adminq_command *cmd_orig)
+{
+	union gve_adminq_command *cmd;
+	u32 opcode;
+	u32 tail;
+
+	lockdep_assert_held(&priv->adminq_lock);
+
+	tail = ioread32be(&priv->reg_bar0->adminq_event_counter);
+
+	// Check if next command will overflow the buffer.
+	if (((priv->adminq_prod_cnt + 1) & priv->adminq_mask) ==
+	    (tail & priv->adminq_mask)) {
+		int err;
+
+		// Flush existing commands to make room.
+		err = gve_adminq_kick_and_wait(priv);
+		if (err)
+			return err;
+
+		// Retry.
+		tail = ioread32be(&priv->reg_bar0->adminq_event_counter);
+		if (((priv->adminq_prod_cnt + 1) & priv->adminq_mask) ==
+		    (tail & priv->adminq_mask)) {
+			// This should never happen. We just flushed the
+			// command queue so there should be enough space.
+			return -ENOMEM;
+		}
+	}
+
+	cmd = &priv->adminq[priv->adminq_prod_cnt & priv->adminq_mask];
+	priv->adminq_prod_cnt++;
+
+	memcpy(cmd, cmd_orig, sizeof(*cmd_orig));
+	opcode = be32_to_cpu(READ_ONCE(cmd->opcode));
+	if (opcode == GVE_ADMINQ_EXTENDED_COMMAND)
+		opcode = be32_to_cpu(cmd->extended_command.inner_opcode);
+
+	switch (opcode) {
+	case GVE_ADMINQ_DESCRIBE_DEVICE:
+		priv->adminq_describe_device_cnt++;
+		break;
+	case GVE_ADMINQ_CONFIGURE_DEVICE_RESOURCES:
+		priv->adminq_cfg_device_resources_cnt++;
+		break;
+	case GVE_ADMINQ_REGISTER_PAGE_LIST:
+		priv->adminq_register_page_list_cnt++;
+		break;
+	case GVE_ADMINQ_UNREGISTER_PAGE_LIST:
+		priv->adminq_unregister_page_list_cnt++;
+		break;
+	case GVE_ADMINQ_CREATE_TX_QUEUE:
+		priv->adminq_create_tx_queue_cnt++;
+		break;
+	case GVE_ADMINQ_CREATE_RX_QUEUE:
+		priv->adminq_create_rx_queue_cnt++;
+		break;
+	case GVE_ADMINQ_DESTROY_TX_QUEUE:
+		priv->adminq_destroy_tx_queue_cnt++;
+		break;
+	case GVE_ADMINQ_DESTROY_RX_QUEUE:
+		priv->adminq_destroy_rx_queue_cnt++;
+		break;
+	case GVE_ADMINQ_DECONFIGURE_DEVICE_RESOURCES:
+		priv->adminq_dcfg_device_resources_cnt++;
+		break;
+	case GVE_ADMINQ_SET_DRIVER_PARAMETER:
+		priv->adminq_set_driver_parameter_cnt++;
+		break;
+	case GVE_ADMINQ_REPORT_STATS:
+		priv->adminq_report_stats_cnt++;
+		break;
+	case GVE_ADMINQ_REPORT_LINK_SPEED:
+		priv->adminq_report_link_speed_cnt++;
+		break;
+	case GVE_ADMINQ_REPORT_NIC_TIMESTAMP:
+		priv->adminq_report_nic_timestamp_cnt++;
+		break;
+	case GVE_ADMINQ_GET_PTYPE_MAP:
+		priv->adminq_get_ptype_map_cnt++;
+		break;
+	case GVE_ADMINQ_VERIFY_DRIVER_COMPATIBILITY:
+		priv->adminq_verify_driver_compatibility_cnt++;
+		break;
+	case GVE_ADMINQ_QUERY_FLOW_RULES:
+		priv->adminq_query_flow_rules_cnt++;
+		break;
+	case GVE_ADMINQ_CONFIGURE_FLOW_RULE:
+		priv->adminq_cfg_flow_rule_cnt++;
+		break;
+	case GVE_ADMINQ_CONFIGURE_RSS:
+		priv->adminq_cfg_rss_cnt++;
+		break;
+	case GVE_ADMINQ_QUERY_RSS:
+		priv->adminq_query_rss_cnt++;
+		break;
+	default:
+		dev_err(&priv->pdev->dev, "unknown AQ command opcode %d\n", opcode);
+		return -EINVAL;
+	}
+
+	return 0;
+}
+
+static int gve_adminq_execute_cmd(struct gve_priv *priv,
+				  union gve_adminq_command *cmd_orig)
+{
+	u32 tail, head;
+	int err;
+
+	mutex_lock(&priv->adminq_lock);
+	tail = ioread32be(&priv->reg_bar0->adminq_event_counter);
+	head = priv->adminq_prod_cnt;
+	if (tail != head) {
+		err = -EINVAL;
+		goto out;
+	}
+
+	err = gve_adminq_issue_cmd(priv, cmd_orig);
+	if (err)
+		goto out;
+
+	err = gve_adminq_kick_and_wait(priv);
+
+out:
+	mutex_unlock(&priv->adminq_lock);
+	return err;
+}
+
+static int gve_adminq_execute_extended_cmd(struct gve_priv *priv, u32 opcode,
+					   size_t cmd_size, void *cmd_orig)
+{
+	union gve_adminq_command cmd;
+	dma_addr_t inner_cmd_bus;
+	void *inner_cmd;
+	int err;
+
+	inner_cmd = dma_alloc_coherent(&priv->pdev->dev, cmd_size,
+				       &inner_cmd_bus, GFP_KERNEL);
+	if (!inner_cmd)
+		return -ENOMEM;
+
+	memcpy(inner_cmd, cmd_orig, cmd_size);
+
+	memset(&cmd, 0, sizeof(cmd));
+	cmd.opcode = cpu_to_be32(GVE_ADMINQ_EXTENDED_COMMAND);
+	cmd.extended_command = (struct gve_adminq_extended_command) {
+		.inner_opcode = cpu_to_be32(opcode),
+		.inner_length = cpu_to_be32(cmd_size),
+		.inner_command_addr = cpu_to_be64(inner_cmd_bus),
+	};
+
+	err = gve_adminq_execute_cmd(priv, &cmd);
+
+	dma_free_coherent(&priv->pdev->dev, cmd_size, inner_cmd, inner_cmd_bus);
+	return err;
+}
+
+/* The device specifies that the management vector can either be the first irq
+ * or the last irq. ntfy_blk_msix_base_idx indicates the first irq assigned to
+ * the ntfy blks. If it is 0 then the management vector is last, if it is 1 then
+ * the management vector is first.
+ *
+ * gve arranges the msix vectors so that the management vector is last.
+ */
+#define GVE_NTFY_BLK_BASE_MSIX_IDX	0
+int gve_adminq_configure_device_resources(struct gve_priv *priv,
+					  dma_addr_t counter_array_bus_addr,
+					  u32 num_counters,
+					  dma_addr_t db_array_bus_addr,
+					  u32 num_ntfy_blks)
+{
+	union gve_adminq_command cmd;
+
+	memset(&cmd, 0, sizeof(cmd));
+	cmd.opcode = cpu_to_be32(GVE_ADMINQ_CONFIGURE_DEVICE_RESOURCES);
+	cmd.configure_device_resources =
+		(struct gve_adminq_configure_device_resources) {
+		.counter_array = cpu_to_be64(counter_array_bus_addr),
+		.num_counters = cpu_to_be32(num_counters),
+		.irq_db_addr = cpu_to_be64(db_array_bus_addr),
+		.num_irq_dbs = cpu_to_be32(num_ntfy_blks),
+		.irq_db_stride = cpu_to_be32(sizeof(*priv->irq_db_indices)),
+		.ntfy_blk_msix_base_idx =
+					cpu_to_be32(GVE_NTFY_BLK_BASE_MSIX_IDX),
+		.queue_format = priv->queue_format,
+	};
+
+	return gve_adminq_execute_cmd(priv, &cmd);
+}
+
+int gve_adminq_deconfigure_device_resources(struct gve_priv *priv)
+{
+	union gve_adminq_command cmd;
+
+	memset(&cmd, 0, sizeof(cmd));
+	cmd.opcode = cpu_to_be32(GVE_ADMINQ_DECONFIGURE_DEVICE_RESOURCES);
+
+	return gve_adminq_execute_cmd(priv, &cmd);
+}
+
+static int gve_adminq_create_tx_queue(struct gve_priv *priv, u32 queue_index)
+{
+	struct gve_tx_ring *tx = &priv->tx[queue_index];
+	union gve_adminq_command cmd;
+
+	memset(&cmd, 0, sizeof(cmd));
+	cmd.opcode = cpu_to_be32(GVE_ADMINQ_CREATE_TX_QUEUE);
+	cmd.create_tx_queue = (struct gve_adminq_create_tx_queue) {
+		.queue_id = cpu_to_be32(queue_index),
+		.queue_resources_addr =
+			cpu_to_be64(tx->q_resources_bus),
+		.tx_ring_addr = cpu_to_be64(tx->bus),
+		.ntfy_id = cpu_to_be32(tx->ntfy_id),
+		.tx_ring_size = cpu_to_be16(priv->tx_desc_cnt),
+	};
+
+	if (gve_is_gqi(priv)) {
+		u32 qpl_id = priv->queue_format == GVE_GQI_RDA_FORMAT ?
+			GVE_RAW_ADDRESSING_QPL_ID : tx->tx_fifo.qpl->id;
+
+		cmd.create_tx_queue.queue_page_list_id = cpu_to_be32(qpl_id);
+	} else {
+		u32 qpl_id = 0;
+
+		if (priv->queue_format == GVE_DQO_RDA_FORMAT)
+			qpl_id = GVE_RAW_ADDRESSING_QPL_ID;
+		else
+			qpl_id = tx->dqo.qpl->id;
+		cmd.create_tx_queue.queue_page_list_id = cpu_to_be32(qpl_id);
+		cmd.create_tx_queue.tx_comp_ring_addr =
+			cpu_to_be64(tx->complq_bus_dqo);
+		cmd.create_tx_queue.tx_comp_ring_size =
+			cpu_to_be16(priv->tx_desc_cnt);
+	}
+
+	return gve_adminq_issue_cmd(priv, &cmd);
+}
+
+int gve_adminq_create_tx_queues(struct gve_priv *priv, u32 start_id, u32 num_queues)
+{
+	int err;
+	int i;
+
+	mutex_lock(&priv->adminq_lock);
+
+	for (i = start_id; i < start_id + num_queues; i++) {
+		err = gve_adminq_create_tx_queue(priv, i);
+		if (err)
+			goto out;
+	}
+
+	err = gve_adminq_kick_and_wait(priv);
+
+out:
+	mutex_unlock(&priv->adminq_lock);
+	return err;
+}
+
+static void gve_adminq_get_create_rx_queue_cmd(struct gve_priv *priv,
+					       union gve_adminq_command *cmd,
+					       u32 queue_index)
+{
+	struct gve_rx_ring *rx = &priv->rx[queue_index];
+
+	memset(cmd, 0, sizeof(*cmd));
+	cmd->opcode = cpu_to_be32(GVE_ADMINQ_CREATE_RX_QUEUE);
+	cmd->create_rx_queue = (struct gve_adminq_create_rx_queue) {
+		.queue_id = cpu_to_be32(queue_index),
+		.ntfy_id = cpu_to_be32(rx->ntfy_id),
+		.queue_resources_addr = cpu_to_be64(rx->q_resources_bus),
+		.rx_ring_size = cpu_to_be16(priv->rx_desc_cnt),
+		.packet_buffer_size = cpu_to_be16(rx->packet_buffer_size),
+	};
+
+	if (gve_is_gqi(priv)) {
+		u32 qpl_id = priv->queue_format == GVE_GQI_RDA_FORMAT ?
+			GVE_RAW_ADDRESSING_QPL_ID : rx->data.qpl->id;
+
+		cmd->create_rx_queue.rx_desc_ring_addr =
+			cpu_to_be64(rx->desc.bus);
+		cmd->create_rx_queue.rx_data_ring_addr =
+			cpu_to_be64(rx->data.data_bus);
+		cmd->create_rx_queue.index = cpu_to_be32(queue_index);
+		cmd->create_rx_queue.queue_page_list_id = cpu_to_be32(qpl_id);
+	} else {
+		u32 qpl_id = 0;
+
+		if (priv->queue_format == GVE_DQO_RDA_FORMAT)
+			qpl_id = GVE_RAW_ADDRESSING_QPL_ID;
+		else
+			qpl_id = rx->dqo.qpl->id;
+		cmd->create_rx_queue.queue_page_list_id = cpu_to_be32(qpl_id);
+		cmd->create_rx_queue.rx_desc_ring_addr =
+			cpu_to_be64(rx->dqo.complq.bus);
+		cmd->create_rx_queue.rx_data_ring_addr =
+			cpu_to_be64(rx->dqo.bufq.bus);
+		cmd->create_rx_queue.rx_buff_ring_size =
+			cpu_to_be16(priv->rx_desc_cnt);
+		cmd->create_rx_queue.enable_rsc =
+			!!(priv->dev->features & NETIF_F_LRO);
+		if (priv->header_split_enabled)
+			cmd->create_rx_queue.header_buffer_size =
+				cpu_to_be16(priv->header_buf_size);
+	}
+}
+
+static int gve_adminq_create_rx_queue(struct gve_priv *priv, u32 queue_index)
+{
+	union gve_adminq_command cmd;
+
+	gve_adminq_get_create_rx_queue_cmd(priv, &cmd, queue_index);
+	return gve_adminq_issue_cmd(priv, &cmd);
+}
+
+/* Unlike gve_adminq_create_rx_queue, this actually rings the doorbell */
+int gve_adminq_create_single_rx_queue(struct gve_priv *priv, u32 queue_index)
+{
+	union gve_adminq_command cmd;
+
+	gve_adminq_get_create_rx_queue_cmd(priv, &cmd, queue_index);
+	return gve_adminq_execute_cmd(priv, &cmd);
+}
+
+int gve_adminq_create_rx_queues(struct gve_priv *priv, u32 num_queues)
+{
+	int err;
+	int i;
+
+	mutex_lock(&priv->adminq_lock);
+
+	for (i = 0; i < num_queues; i++) {
+		err = gve_adminq_create_rx_queue(priv, i);
+		if (err)
+			goto out;
+	}
+
+	err = gve_adminq_kick_and_wait(priv);
+
+out:
+	mutex_unlock(&priv->adminq_lock);
+	return err;
+}
+
+static int gve_adminq_destroy_tx_queue(struct gve_priv *priv, u32 queue_index)
+{
+	union gve_adminq_command cmd;
+	int err;
+
+	memset(&cmd, 0, sizeof(cmd));
+	cmd.opcode = cpu_to_be32(GVE_ADMINQ_DESTROY_TX_QUEUE);
+	cmd.destroy_tx_queue = (struct gve_adminq_destroy_tx_queue) {
+		.queue_id = cpu_to_be32(queue_index),
+	};
+
+	err = gve_adminq_issue_cmd(priv, &cmd);
+	if (err)
+		return err;
+
+	return 0;
+}
+
+int gve_adminq_destroy_tx_queues(struct gve_priv *priv, u32 start_id, u32 num_queues)
+{
+	int err;
+	int i;
+
+	mutex_lock(&priv->adminq_lock);
+
+	for (i = start_id; i < start_id + num_queues; i++) {
+		err = gve_adminq_destroy_tx_queue(priv, i);
+		if (err)
+			goto out;
+	}
+
+	err = gve_adminq_kick_and_wait(priv);
+
+out:
+	mutex_unlock(&priv->adminq_lock);
+	return err;
+}
+
+static void gve_adminq_make_destroy_rx_queue_cmd(union gve_adminq_command *cmd,
+						 u32 queue_index)
+{
+	memset(cmd, 0, sizeof(*cmd));
+	cmd->opcode = cpu_to_be32(GVE_ADMINQ_DESTROY_RX_QUEUE);
+	cmd->destroy_rx_queue = (struct gve_adminq_destroy_rx_queue) {
+		.queue_id = cpu_to_be32(queue_index),
+	};
+}
+
+static int gve_adminq_destroy_rx_queue(struct gve_priv *priv, u32 queue_index)
+{
+	union gve_adminq_command cmd;
+
+	gve_adminq_make_destroy_rx_queue_cmd(&cmd, queue_index);
+	return gve_adminq_issue_cmd(priv, &cmd);
+}
+
+/* Unlike gve_adminq_destroy_rx_queue, this actually rings the doorbell */
+int gve_adminq_destroy_single_rx_queue(struct gve_priv *priv, u32 queue_index)
+{
+	union gve_adminq_command cmd;
+
+	gve_adminq_make_destroy_rx_queue_cmd(&cmd, queue_index);
+	return gve_adminq_execute_cmd(priv, &cmd);
+}
+
+int gve_adminq_destroy_rx_queues(struct gve_priv *priv, u32 num_queues)
+{
+	int err;
+	int i;
+
+	mutex_lock(&priv->adminq_lock);
+
+	for (i = 0; i < num_queues; i++) {
+		err = gve_adminq_destroy_rx_queue(priv, i);
+		if (err)
+			goto out;
+	}
+
+	err = gve_adminq_kick_and_wait(priv);
+
+out:
+	mutex_unlock(&priv->adminq_lock);
+	return err;
+}
+
+static void gve_set_default_desc_cnt(struct gve_priv *priv,
+			const struct gve_device_descriptor *descriptor)
+{
+	priv->tx_desc_cnt = be16_to_cpu(descriptor->tx_queue_entries);
+	priv->rx_desc_cnt = be16_to_cpu(descriptor->rx_queue_entries);
+
+	/* set default ranges */
+	priv->max_tx_desc_cnt = priv->tx_desc_cnt;
+	priv->max_rx_desc_cnt = priv->rx_desc_cnt;
+	priv->min_tx_desc_cnt = priv->tx_desc_cnt;
+	priv->min_rx_desc_cnt = priv->rx_desc_cnt;
+}
+
+static void gve_set_default_rss_sizes(struct gve_priv *priv)
+{
+	if (!gve_is_gqi(priv)) {
+		priv->rss_key_size = GVE_RSS_KEY_SIZE;
+		priv->rss_lut_size = GVE_RSS_INDIR_SIZE;
+		priv->cache_rss_config = true;
+	}
+}
+
+static void gve_enable_supported_features(struct gve_priv *priv,
+					  u32 supported_features_mask,
+					  const struct gve_device_option_jumbo_frames
+					  *dev_op_jumbo_frames,
+					  const struct gve_device_option_dqo_qpl
+					  *dev_op_dqo_qpl,
+					  const struct gve_device_option_buffer_sizes
+					  *dev_op_buffer_sizes,
+					  const struct gve_device_option_flow_steering
+					  *dev_op_flow_steering,
+					  const struct gve_device_option_rss_config
+					  *dev_op_rss_config,
+					  const struct gve_device_option_nic_timestamp
+					  *dev_op_nic_timestamp,
+					  const struct gve_device_option_modify_ring
+					  *dev_op_modify_ring)
+{
+	/* Before control reaches this point, the page-size-capped max MTU from
+	 * the gve_device_descriptor field has already been stored in
+	 * priv->dev->max_mtu. We overwrite it with the true max MTU below.
+	 */
+	if (dev_op_jumbo_frames &&
+	    (supported_features_mask & GVE_SUP_JUMBO_FRAMES_MASK)) {
+		dev_info(&priv->pdev->dev,
+			 "JUMBO FRAMES device option enabled.\n");
+		priv->dev->max_mtu = be16_to_cpu(dev_op_jumbo_frames->max_mtu);
+	}
+
+	/* Override pages for qpl for DQO-QPL */
+	if (dev_op_dqo_qpl) {
+		priv->tx_pages_per_qpl =
+			be16_to_cpu(dev_op_dqo_qpl->tx_pages_per_qpl);
+		if (priv->tx_pages_per_qpl == 0)
+			priv->tx_pages_per_qpl = DQO_QPL_DEFAULT_TX_PAGES;
+	}
+
+	if (dev_op_buffer_sizes &&
+	    (supported_features_mask & GVE_SUP_BUFFER_SIZES_MASK)) {
+		priv->max_rx_buffer_size =
+			be16_to_cpu(dev_op_buffer_sizes->packet_buffer_size);
+		priv->header_buf_size =
+			be16_to_cpu(dev_op_buffer_sizes->header_buffer_size);
+		dev_info(&priv->pdev->dev,
+			 "BUFFER SIZES device option enabled with max_rx_buffer_size of %u, header_buf_size of %u.\n",
+			 priv->max_rx_buffer_size, priv->header_buf_size);
+	}
+
+	/* Read and store ring size ranges given by device */
+	if (dev_op_modify_ring &&
+	    (supported_features_mask & GVE_SUP_MODIFY_RING_MASK)) {
+		priv->modify_ring_size_enabled = true;
+
+		/* max ring size for DQO QPL should not be overwritten because of device limit */
+		if (priv->queue_format != GVE_DQO_QPL_FORMAT) {
+			priv->max_rx_desc_cnt = be16_to_cpu(dev_op_modify_ring->max_rx_ring_size);
+			priv->max_tx_desc_cnt = be16_to_cpu(dev_op_modify_ring->max_tx_ring_size);
+		}
+		if (priv->default_min_ring_size) {
+			/* If device hasn't provided minimums, use default minimums */
+			priv->min_tx_desc_cnt = GVE_DEFAULT_MIN_TX_RING_SIZE;
+			priv->min_rx_desc_cnt = GVE_DEFAULT_MIN_RX_RING_SIZE;
+		} else {
+			priv->min_rx_desc_cnt = be16_to_cpu(dev_op_modify_ring->min_rx_ring_size);
+			priv->min_tx_desc_cnt = be16_to_cpu(dev_op_modify_ring->min_tx_ring_size);
+		}
+	}
+
+	if (dev_op_flow_steering &&
+	    (supported_features_mask & GVE_SUP_FLOW_STEERING_MASK)) {
+		if (dev_op_flow_steering->max_flow_rules) {
+			priv->max_flow_rules =
+				be32_to_cpu(dev_op_flow_steering->max_flow_rules);
+			priv->dev->hw_features |= NETIF_F_NTUPLE;
+			dev_info(&priv->pdev->dev,
+				 "FLOW STEERING device option enabled with max rule limit of %u.\n",
+				 priv->max_flow_rules);
+		}
+	}
+
+	if (dev_op_rss_config &&
+	    (supported_features_mask & GVE_SUP_RSS_CONFIG_MASK)) {
+		priv->rss_key_size =
+			be16_to_cpu(dev_op_rss_config->hash_key_size);
+		priv->rss_lut_size =
+			be16_to_cpu(dev_op_rss_config->hash_lut_size);
+		priv->cache_rss_config = false;
+		dev_dbg(&priv->pdev->dev,
+			"RSS device option enabled with key size of %u, lut size of %u.\n",
+			priv->rss_key_size, priv->rss_lut_size);
+	}
+
+	if (dev_op_nic_timestamp &&
+	    (supported_features_mask & GVE_SUP_NIC_TIMESTAMP_MASK))
+		priv->nic_timestamp_supported = true;
+}
+
+int gve_adminq_describe_device(struct gve_priv *priv)
+{
+	struct gve_device_option_nic_timestamp *dev_op_nic_timestamp = NULL;
+	struct gve_device_option_flow_steering *dev_op_flow_steering = NULL;
+	struct gve_device_option_buffer_sizes *dev_op_buffer_sizes = NULL;
+	struct gve_device_option_jumbo_frames *dev_op_jumbo_frames = NULL;
+	struct gve_device_option_modify_ring *dev_op_modify_ring = NULL;
+	struct gve_device_option_rss_config *dev_op_rss_config = NULL;
+	struct gve_device_option_gqi_rda *dev_op_gqi_rda = NULL;
+	struct gve_device_option_gqi_qpl *dev_op_gqi_qpl = NULL;
+	struct gve_device_option_dqo_rda *dev_op_dqo_rda = NULL;
+	struct gve_device_option_dqo_qpl *dev_op_dqo_qpl = NULL;
+	struct gve_device_descriptor *descriptor;
+	u32 supported_features_mask = 0;
+	union gve_adminq_command cmd;
+	dma_addr_t descriptor_bus;
+	int err = 0;
+	u8 *mac;
+	u16 mtu;
+
+	memset(&cmd, 0, sizeof(cmd));
+	descriptor = dma_pool_alloc(priv->adminq_pool, GFP_KERNEL,
+				    &descriptor_bus);
+	if (!descriptor)
+		return -ENOMEM;
+	cmd.opcode = cpu_to_be32(GVE_ADMINQ_DESCRIBE_DEVICE);
+	cmd.describe_device.device_descriptor_addr =
+						cpu_to_be64(descriptor_bus);
+	cmd.describe_device.device_descriptor_version =
+			cpu_to_be32(GVE_ADMINQ_DEVICE_DESCRIPTOR_VERSION);
+	cmd.describe_device.available_length =
+		cpu_to_be32(GVE_ADMINQ_BUFFER_SIZE);
+
+	err = gve_adminq_execute_cmd(priv, &cmd);
+	if (err)
+		goto free_device_descriptor;
+
+	err = gve_process_device_options(priv, descriptor, &dev_op_gqi_rda,
+					 &dev_op_gqi_qpl, &dev_op_dqo_rda,
+					 &dev_op_jumbo_frames, &dev_op_dqo_qpl,
+					 &dev_op_buffer_sizes,
+					 &dev_op_flow_steering,
+					 &dev_op_rss_config,
+					 &dev_op_nic_timestamp,
+					 &dev_op_modify_ring);
+	if (err)
+		goto free_device_descriptor;
+
+	/* If the GQI_RAW_ADDRESSING option is not enabled and the queue format
+	 * is not set to GqiRda, choose the queue format in a priority order:
+	 * DqoRda, DqoQpl, GqiRda, GqiQpl. Use GqiQpl as default.
+	 */
+	if (dev_op_dqo_rda) {
+		priv->queue_format = GVE_DQO_RDA_FORMAT;
+		dev_info(&priv->pdev->dev,
+			 "Driver is running with DQO RDA queue format.\n");
+		supported_features_mask =
+			be32_to_cpu(dev_op_dqo_rda->supported_features_mask);
+	} else if (dev_op_dqo_qpl) {
+		priv->queue_format = GVE_DQO_QPL_FORMAT;
+		supported_features_mask =
+			be32_to_cpu(dev_op_dqo_qpl->supported_features_mask);
+	}  else if (dev_op_gqi_rda) {
+		priv->queue_format = GVE_GQI_RDA_FORMAT;
+		dev_info(&priv->pdev->dev,
+			 "Driver is running with GQI RDA queue format.\n");
+		supported_features_mask =
+			be32_to_cpu(dev_op_gqi_rda->supported_features_mask);
+	} else if (priv->queue_format == GVE_GQI_RDA_FORMAT) {
+		dev_info(&priv->pdev->dev,
+			 "Driver is running with GQI RDA queue format.\n");
+	} else {
+		priv->queue_format = GVE_GQI_QPL_FORMAT;
+		if (dev_op_gqi_qpl)
+			supported_features_mask =
+				be32_to_cpu(dev_op_gqi_qpl->supported_features_mask);
+		dev_info(&priv->pdev->dev,
+			 "Driver is running with GQI QPL queue format.\n");
+	}
+
+	/* set default descriptor counts */
+	gve_set_default_desc_cnt(priv, descriptor);
+
+	gve_set_default_rss_sizes(priv);
+
+	/* DQO supports LRO. */
+	if (!gve_is_gqi(priv))
+		priv->dev->hw_features |= NETIF_F_LRO;
+
+	priv->max_registered_pages =
+				be64_to_cpu(descriptor->max_registered_pages);
+	mtu = be16_to_cpu(descriptor->mtu);
+	if (mtu < ETH_MIN_MTU) {
+		dev_err(&priv->pdev->dev, "MTU %d below minimum MTU\n", mtu);
+		err = -EINVAL;
+		goto free_device_descriptor;
+	}
+	priv->dev->max_mtu = mtu;
+	priv->num_event_counters = be16_to_cpu(descriptor->counters);
+	eth_hw_addr_set(priv->dev, descriptor->mac);
+	mac = descriptor->mac;
+	dev_info(&priv->pdev->dev, "MAC addr: %pM\n", mac);
+	priv->tx_pages_per_qpl = be16_to_cpu(descriptor->tx_pages_per_qpl);
+	priv->default_num_queues = be16_to_cpu(descriptor->default_num_queues);
+
+	gve_enable_supported_features(priv, supported_features_mask,
+				      dev_op_jumbo_frames, dev_op_dqo_qpl,
+				      dev_op_buffer_sizes, dev_op_flow_steering,
+				      dev_op_rss_config, dev_op_nic_timestamp,
+				      dev_op_modify_ring);
+
+free_device_descriptor:
+	dma_pool_free(priv->adminq_pool, descriptor, descriptor_bus);
+	return err;
+}
+
+int gve_adminq_register_page_list(struct gve_priv *priv,
+				  struct gve_queue_page_list *qpl)
+{
+	struct device *hdev = &priv->pdev->dev;
+	u32 num_entries = qpl->num_entries;
+	u32 size = num_entries * sizeof(qpl->page_buses[0]);
+	union gve_adminq_command cmd;
+	dma_addr_t page_list_bus;
+	__be64 *page_list;
+	int err;
+	int i;
+
+	memset(&cmd, 0, sizeof(cmd));
+	page_list = dma_alloc_coherent(hdev, size, &page_list_bus, GFP_KERNEL);
+	if (!page_list)
+		return -ENOMEM;
+
+	for (i = 0; i < num_entries; i++)
+		page_list[i] = cpu_to_be64(qpl->page_buses[i]);
+
+	cmd.opcode = cpu_to_be32(GVE_ADMINQ_REGISTER_PAGE_LIST);
+	cmd.reg_page_list = (struct gve_adminq_register_page_list) {
+		.page_list_id = cpu_to_be32(qpl->id),
+		.num_pages = cpu_to_be32(num_entries),
+		.page_address_list_addr = cpu_to_be64(page_list_bus),
+		.page_size = cpu_to_be64(PAGE_SIZE),
+	};
+
+	err = gve_adminq_execute_cmd(priv, &cmd);
+	dma_free_coherent(hdev, size, page_list, page_list_bus);
+	return err;
+}
+
+int gve_adminq_unregister_page_list(struct gve_priv *priv, u32 page_list_id)
+{
+	union gve_adminq_command cmd;
+
+	memset(&cmd, 0, sizeof(cmd));
+	cmd.opcode = cpu_to_be32(GVE_ADMINQ_UNREGISTER_PAGE_LIST);
+	cmd.unreg_page_list = (struct gve_adminq_unregister_page_list) {
+		.page_list_id = cpu_to_be32(page_list_id),
+	};
+
+	return gve_adminq_execute_cmd(priv, &cmd);
+}
+
+int gve_adminq_report_stats(struct gve_priv *priv, u64 stats_report_len,
+			    dma_addr_t stats_report_addr, u64 interval)
+{
+	union gve_adminq_command cmd;
+
+	memset(&cmd, 0, sizeof(cmd));
+	cmd.opcode = cpu_to_be32(GVE_ADMINQ_REPORT_STATS);
+	cmd.report_stats = (struct gve_adminq_report_stats) {
+		.stats_report_len = cpu_to_be64(stats_report_len),
+		.stats_report_addr = cpu_to_be64(stats_report_addr),
+		.interval = cpu_to_be64(interval),
+	};
+
+	return gve_adminq_execute_cmd(priv, &cmd);
+}
+
+int gve_adminq_verify_driver_compatibility(struct gve_priv *priv,
+					   u64 driver_info_len,
+					   dma_addr_t driver_info_addr)
+{
+	union gve_adminq_command cmd;
+
+	memset(&cmd, 0, sizeof(cmd));
+	cmd.opcode = cpu_to_be32(GVE_ADMINQ_VERIFY_DRIVER_COMPATIBILITY);
+	cmd.verify_driver_compatibility = (struct gve_adminq_verify_driver_compatibility) {
+		.driver_info_len = cpu_to_be64(driver_info_len),
+		.driver_info_addr = cpu_to_be64(driver_info_addr),
+	};
+
+	return gve_adminq_execute_cmd(priv, &cmd);
+}
+
+int gve_adminq_report_link_speed(struct gve_priv *priv)
+{
+	union gve_adminq_command gvnic_cmd;
+	dma_addr_t link_speed_region_bus;
+	__be64 *link_speed_region;
+	int err;
+
+	link_speed_region =
+		dma_alloc_coherent(&priv->pdev->dev, sizeof(*link_speed_region),
+				   &link_speed_region_bus, GFP_KERNEL);
+
+	if (!link_speed_region)
+		return -ENOMEM;
+
+	memset(&gvnic_cmd, 0, sizeof(gvnic_cmd));
+	gvnic_cmd.opcode = cpu_to_be32(GVE_ADMINQ_REPORT_LINK_SPEED);
+	gvnic_cmd.report_link_speed.link_speed_address =
+		cpu_to_be64(link_speed_region_bus);
+
+	err = gve_adminq_execute_cmd(priv, &gvnic_cmd);
+
+	priv->link_speed = be64_to_cpu(*link_speed_region);
+	dma_free_coherent(&priv->pdev->dev, sizeof(*link_speed_region), link_speed_region,
+			  link_speed_region_bus);
+	return err;
+}
+
+int gve_adminq_report_nic_ts(struct gve_priv *priv,
+			     dma_addr_t nic_ts_report_addr)
+{
+	union gve_adminq_command cmd;
+
+	memset(&cmd, 0, sizeof(cmd));
+	cmd.opcode = cpu_to_be32(GVE_ADMINQ_REPORT_NIC_TIMESTAMP);
+	cmd.report_nic_ts = (struct gve_adminq_report_nic_ts) {
+		.nic_ts_report_len =
+			cpu_to_be64(sizeof(struct gve_nic_ts_report)),
+		.nic_ts_report_addr = cpu_to_be64(nic_ts_report_addr),
+	};
+
+	return gve_adminq_execute_cmd(priv, &cmd);
+}
+
+int gve_adminq_get_ptype_map_dqo(struct gve_priv *priv,
+				 struct gve_ptype_lut *ptype_lut)
+{
+	struct gve_ptype_map *ptype_map;
+	union gve_adminq_command cmd;
+	dma_addr_t ptype_map_bus;
+	int err = 0;
+	int i;
+
+	memset(&cmd, 0, sizeof(cmd));
+	ptype_map = dma_alloc_coherent(&priv->pdev->dev, sizeof(*ptype_map),
+				       &ptype_map_bus, GFP_KERNEL);
+	if (!ptype_map)
+		return -ENOMEM;
+
+	cmd.opcode = cpu_to_be32(GVE_ADMINQ_GET_PTYPE_MAP);
+	cmd.get_ptype_map = (struct gve_adminq_get_ptype_map) {
+		.ptype_map_len = cpu_to_be64(sizeof(*ptype_map)),
+		.ptype_map_addr = cpu_to_be64(ptype_map_bus),
+	};
+
+	err = gve_adminq_execute_cmd(priv, &cmd);
+	if (err)
+		goto err;
+
+	/* Populate ptype_lut. */
+	for (i = 0; i < GVE_NUM_PTYPES; i++) {
+		ptype_lut->ptypes[i].l3_type =
+			ptype_map->ptypes[i].l3_type;
+		ptype_lut->ptypes[i].l4_type =
+			ptype_map->ptypes[i].l4_type;
+	}
+err:
+	dma_free_coherent(&priv->pdev->dev, sizeof(*ptype_map), ptype_map,
+			  ptype_map_bus);
+	return err;
+}
+
+static int
+gve_adminq_configure_flow_rule(struct gve_priv *priv,
+			       struct gve_adminq_configure_flow_rule *flow_rule_cmd)
+{
+	int err = gve_adminq_execute_extended_cmd(priv,
+			GVE_ADMINQ_CONFIGURE_FLOW_RULE,
+			sizeof(struct gve_adminq_configure_flow_rule),
+			flow_rule_cmd);
+
+	if (err == -ETIME) {
+		dev_err(&priv->pdev->dev, "Timeout to configure the flow rule, trigger reset");
+		gve_reset(priv, true);
+	} else if (!err) {
+		priv->flow_rules_cache.rules_cache_synced = false;
+	}
+
+	return err;
+}
+
+int gve_adminq_add_flow_rule(struct gve_priv *priv, struct gve_adminq_flow_rule *rule, u32 loc)
+{
+	struct gve_adminq_configure_flow_rule flow_rule_cmd = {
+		.opcode = cpu_to_be16(GVE_FLOW_RULE_CFG_ADD),
+		.location = cpu_to_be32(loc),
+		.rule = *rule,
+	};
+
+	return gve_adminq_configure_flow_rule(priv, &flow_rule_cmd);
+}
+
+int gve_adminq_del_flow_rule(struct gve_priv *priv, u32 loc)
+{
+	struct gve_adminq_configure_flow_rule flow_rule_cmd = {
+		.opcode = cpu_to_be16(GVE_FLOW_RULE_CFG_DEL),
+		.location = cpu_to_be32(loc),
+	};
+
+	return gve_adminq_configure_flow_rule(priv, &flow_rule_cmd);
+}
+
+int gve_adminq_reset_flow_rules(struct gve_priv *priv)
+{
+	struct gve_adminq_configure_flow_rule flow_rule_cmd = {
+		.opcode = cpu_to_be16(GVE_FLOW_RULE_CFG_RESET),
+	};
+
+	return gve_adminq_configure_flow_rule(priv, &flow_rule_cmd);
+}
+
+int gve_adminq_configure_rss(struct gve_priv *priv, struct ethtool_rxfh_param *rxfh)
+{
+	const u32 *hash_lut_to_config = NULL;
+	const u8 *hash_key_to_config = NULL;
+	dma_addr_t lut_bus = 0, key_bus = 0;
+	union gve_adminq_command cmd;
+	__be32 *lut = NULL;
+	u8 hash_alg = 0;
+	u8 *key = NULL;
+	int err = 0;
+	u16 i;
+
+	switch (rxfh->hfunc) {
+	case ETH_RSS_HASH_NO_CHANGE:
+		fallthrough;
+	case ETH_RSS_HASH_TOP:
+		hash_alg = ETH_RSS_HASH_TOP;
+		break;
+	default:
+		return -EOPNOTSUPP;
+	}
+
+	if (rxfh->indir) {
+		if (rxfh->indir_size != priv->rss_lut_size)
+			return -EINVAL;
+
+		hash_lut_to_config = rxfh->indir;
+	} else if (priv->cache_rss_config) {
+		hash_lut_to_config = priv->rss_config.hash_lut;
+	}
+
+	if (hash_lut_to_config) {
+		lut = dma_alloc_coherent(&priv->pdev->dev,
+					 priv->rss_lut_size * sizeof(*lut),
+					 &lut_bus, GFP_KERNEL);
+		if (!lut)
+			return -ENOMEM;
+
+		for (i = 0; i < priv->rss_lut_size; i++)
+			lut[i] = cpu_to_be32(hash_lut_to_config[i]);
+	}
+
+	if (rxfh->key) {
+		if (rxfh->key_size != priv->rss_key_size) {
+			err = -EINVAL;
+			goto out;
+		}
+
+		hash_key_to_config = rxfh->key;
+	} else if (priv->cache_rss_config) {
+		hash_key_to_config = priv->rss_config.hash_key;
+	}
+
+	if (hash_key_to_config) {
+		key = dma_alloc_coherent(&priv->pdev->dev,
+					 priv->rss_key_size,
+					 &key_bus, GFP_KERNEL);
+		if (!key) {
+			err = -ENOMEM;
+			goto out;
+		}
+
+		memcpy(key, hash_key_to_config, priv->rss_key_size);
+	}
+
+	/* Zero-valued fields in the cmd.configure_rss instruct the device to
+	 * not update those fields.
+	 */
+	memset(&cmd, 0, sizeof(cmd));
+	cmd.opcode = cpu_to_be32(GVE_ADMINQ_CONFIGURE_RSS);
+	cmd.configure_rss = (struct gve_adminq_configure_rss) {
+		.hash_types = cpu_to_be16(BIT(GVE_RSS_HASH_TCPV4) |
+					  BIT(GVE_RSS_HASH_UDPV4) |
+					  BIT(GVE_RSS_HASH_TCPV6) |
+					  BIT(GVE_RSS_HASH_UDPV6)),
+		.hash_alg = hash_alg,
+		.hash_key_size =
+			cpu_to_be16((key_bus) ? priv->rss_key_size : 0),
+		.hash_lut_size =
+			cpu_to_be16((lut_bus) ? priv->rss_lut_size : 0),
+		.hash_key_addr = cpu_to_be64(key_bus),
+		.hash_lut_addr = cpu_to_be64(lut_bus),
+	};
+
+	err = gve_adminq_execute_cmd(priv, &cmd);
+
+out:
+	if (lut)
+		dma_free_coherent(&priv->pdev->dev,
+				  priv->rss_lut_size * sizeof(*lut),
+				  lut, lut_bus);
+	if (key)
+		dma_free_coherent(&priv->pdev->dev,
+				  priv->rss_key_size, key, key_bus);
+	return err;
+}
+
+/* In the dma memory that the driver allocated for the device to query the flow rules, the device
+ * will first write it with a struct of gve_query_flow_rules_descriptor. Next to it, the device
+ * will write an array of rules or rule ids with the count that specified in the descriptor.
+ * For GVE_FLOW_RULE_QUERY_STATS, the device will only write the descriptor.
+ */
+static int gve_adminq_process_flow_rules_query(struct gve_priv *priv, u16 query_opcode,
+					       struct gve_query_flow_rules_descriptor *descriptor)
+{
+	struct gve_flow_rules_cache *flow_rules_cache = &priv->flow_rules_cache;
+	u32 num_queried_rules, total_memory_len, rule_info_len;
+	void *rule_info;
+
+	total_memory_len = be32_to_cpu(descriptor->total_length);
+	num_queried_rules = be32_to_cpu(descriptor->num_queried_rules);
+	rule_info = (void *)(descriptor + 1);
+
+	switch (query_opcode) {
+	case GVE_FLOW_RULE_QUERY_RULES:
+		rule_info_len = num_queried_rules * sizeof(*flow_rules_cache->rules_cache);
+		if (sizeof(*descriptor) + rule_info_len != total_memory_len) {
+			dev_err(&priv->dev->dev, "flow rules query is out of memory.\n");
+			return -ENOMEM;
+		}
+
+		memcpy(flow_rules_cache->rules_cache, rule_info, rule_info_len);
+		flow_rules_cache->rules_cache_num = num_queried_rules;
+		break;
+	case GVE_FLOW_RULE_QUERY_IDS:
+		rule_info_len = num_queried_rules * sizeof(*flow_rules_cache->rule_ids_cache);
+		if (sizeof(*descriptor) + rule_info_len != total_memory_len) {
+			dev_err(&priv->dev->dev, "flow rule ids query is out of memory.\n");
+			return -ENOMEM;
+		}
+
+		memcpy(flow_rules_cache->rule_ids_cache, rule_info, rule_info_len);
+		flow_rules_cache->rule_ids_cache_num = num_queried_rules;
+		break;
+	case GVE_FLOW_RULE_QUERY_STATS:
+		priv->num_flow_rules = be32_to_cpu(descriptor->num_flow_rules);
+		priv->max_flow_rules = be32_to_cpu(descriptor->max_flow_rules);
+		return 0;
+	default:
+		return -EINVAL;
+	}
+
+	return  0;
+}
+
+int gve_adminq_query_flow_rules(struct gve_priv *priv, u16 query_opcode, u32 starting_loc)
+{
+	struct gve_query_flow_rules_descriptor *descriptor;
+	union gve_adminq_command cmd;
+	dma_addr_t descriptor_bus;
+	int err = 0;
+
+	memset(&cmd, 0, sizeof(cmd));
+	descriptor = dma_pool_alloc(priv->adminq_pool, GFP_KERNEL, &descriptor_bus);
+	if (!descriptor)
+		return -ENOMEM;
+
+	cmd.opcode = cpu_to_be32(GVE_ADMINQ_QUERY_FLOW_RULES);
+	cmd.query_flow_rules = (struct gve_adminq_query_flow_rules) {
+		.opcode = cpu_to_be16(query_opcode),
+		.starting_rule_id = cpu_to_be32(starting_loc),
+		.available_length = cpu_to_be64(GVE_ADMINQ_BUFFER_SIZE),
+		.rule_descriptor_addr = cpu_to_be64(descriptor_bus),
+	};
+	err = gve_adminq_execute_cmd(priv, &cmd);
+	if (err)
+		goto out;
+
+	err = gve_adminq_process_flow_rules_query(priv, query_opcode, descriptor);
+
+out:
+	dma_pool_free(priv->adminq_pool, descriptor, descriptor_bus);
+	return err;
+}
+
+static int gve_adminq_process_rss_query(struct gve_priv *priv,
+					struct gve_query_rss_descriptor *descriptor,
+					struct ethtool_rxfh_param *rxfh)
+{
+	u32 total_memory_length;
+	u16 hash_lut_length;
+	void *rss_info_addr;
+	__be32 *lut;
+	u16 i;
+
+	total_memory_length = be32_to_cpu(descriptor->total_length);
+	hash_lut_length = priv->rss_lut_size * sizeof(*rxfh->indir);
+
+	if (sizeof(*descriptor) + priv->rss_key_size + hash_lut_length != total_memory_length) {
+		dev_err(&priv->dev->dev,
+			"rss query desc from device has invalid length parameter.\n");
+		return -EINVAL;
+	}
+
+	rxfh->hfunc = descriptor->hash_alg;
+
+	rss_info_addr = (void *)(descriptor + 1);
+	if (rxfh->key) {
+		rxfh->key_size = priv->rss_key_size;
+		memcpy(rxfh->key, rss_info_addr, priv->rss_key_size);
+	}
+
+	rss_info_addr += priv->rss_key_size;
+	lut = (__be32 *)rss_info_addr;
+	if (rxfh->indir) {
+		rxfh->indir_size = priv->rss_lut_size;
+		for (i = 0; i < priv->rss_lut_size; i++)
+			rxfh->indir[i] = be32_to_cpu(lut[i]);
+	}
+
+	return 0;
+}
+
+int gve_adminq_query_rss_config(struct gve_priv *priv, struct ethtool_rxfh_param *rxfh)
+{
+	struct gve_query_rss_descriptor *descriptor;
+	union gve_adminq_command cmd;
+	dma_addr_t descriptor_bus;
+	int err = 0;
+
+	descriptor = dma_pool_alloc(priv->adminq_pool, GFP_KERNEL, &descriptor_bus);
+	if (!descriptor)
+		return -ENOMEM;
+
+	memset(&cmd, 0, sizeof(cmd));
+	cmd.opcode = cpu_to_be32(GVE_ADMINQ_QUERY_RSS);
+	cmd.query_rss = (struct gve_adminq_query_rss) {
+		.available_length = cpu_to_be64(GVE_ADMINQ_BUFFER_SIZE),
+		.rss_descriptor_addr = cpu_to_be64(descriptor_bus),
+	};
+	err = gve_adminq_execute_cmd(priv, &cmd);
+	if (err)
+		goto out;
+
+	err = gve_adminq_process_rss_query(priv, descriptor, rxfh);
+
+out:
+	dma_pool_free(priv->adminq_pool, descriptor, descriptor_bus);
+	return err;
+}
diff --git a/drivers/net/ethernet/google/gve/gve_adminq.h b/drivers/net/ethernet/google/gve/gve_adminq.h
new file mode 100644
index 000000000000..22a74b6aa17e
--- /dev/null
+++ b/drivers/net/ethernet/google/gve/gve_adminq.h
@@ -0,0 +1,660 @@
+/* SPDX-License-Identifier: (GPL-2.0 OR MIT)
+ * Google virtual Ethernet (gve) driver
+ *
+ * Copyright (C) 2015-2021 Google, Inc.
+ */
+
+#ifndef _GVE_ADMINQ_H
+#define _GVE_ADMINQ_H
+
+#include <linux/build_bug.h>
+
+/* Admin queue opcodes */
+enum gve_adminq_opcodes {
+	GVE_ADMINQ_DESCRIBE_DEVICE		= 0x1,
+	GVE_ADMINQ_CONFIGURE_DEVICE_RESOURCES	= 0x2,
+	GVE_ADMINQ_REGISTER_PAGE_LIST		= 0x3,
+	GVE_ADMINQ_UNREGISTER_PAGE_LIST		= 0x4,
+	GVE_ADMINQ_CREATE_TX_QUEUE		= 0x5,
+	GVE_ADMINQ_CREATE_RX_QUEUE		= 0x6,
+	GVE_ADMINQ_DESTROY_TX_QUEUE		= 0x7,
+	GVE_ADMINQ_DESTROY_RX_QUEUE		= 0x8,
+	GVE_ADMINQ_DECONFIGURE_DEVICE_RESOURCES	= 0x9,
+	GVE_ADMINQ_CONFIGURE_RSS		= 0xA,
+	GVE_ADMINQ_SET_DRIVER_PARAMETER		= 0xB,
+	GVE_ADMINQ_REPORT_STATS			= 0xC,
+	GVE_ADMINQ_REPORT_LINK_SPEED		= 0xD,
+	GVE_ADMINQ_GET_PTYPE_MAP		= 0xE,
+	GVE_ADMINQ_VERIFY_DRIVER_COMPATIBILITY	= 0xF,
+	GVE_ADMINQ_QUERY_FLOW_RULES		= 0x10,
+	GVE_ADMINQ_REPORT_NIC_TIMESTAMP		= 0x11,
+	GVE_ADMINQ_QUERY_RSS			= 0x12,
+
+	/* For commands that are larger than 56 bytes */
+	GVE_ADMINQ_EXTENDED_COMMAND		= 0xFF,
+};
+
+/* The normal adminq command is restricted to be 56 bytes at maximum. For the
+ * longer adminq command, it is wrapped by GVE_ADMINQ_EXTENDED_COMMAND with
+ * inner opcode of gve_adminq_extended_cmd_opcodes specified. The inner command
+ * is written in the dma memory allocated by GVE_ADMINQ_EXTENDED_COMMAND.
+ */
+enum gve_adminq_extended_cmd_opcodes {
+	GVE_ADMINQ_CONFIGURE_FLOW_RULE	= 0x101,
+};
+
+/* Admin queue status codes */
+enum gve_adminq_statuses {
+	GVE_ADMINQ_COMMAND_UNSET			= 0x0,
+	GVE_ADMINQ_COMMAND_PASSED			= 0x1,
+	GVE_ADMINQ_COMMAND_ERROR_ABORTED		= 0xFFFFFFF0,
+	GVE_ADMINQ_COMMAND_ERROR_ALREADY_EXISTS		= 0xFFFFFFF1,
+	GVE_ADMINQ_COMMAND_ERROR_CANCELLED		= 0xFFFFFFF2,
+	GVE_ADMINQ_COMMAND_ERROR_DATALOSS		= 0xFFFFFFF3,
+	GVE_ADMINQ_COMMAND_ERROR_DEADLINE_EXCEEDED	= 0xFFFFFFF4,
+	GVE_ADMINQ_COMMAND_ERROR_FAILED_PRECONDITION	= 0xFFFFFFF5,
+	GVE_ADMINQ_COMMAND_ERROR_INTERNAL_ERROR		= 0xFFFFFFF6,
+	GVE_ADMINQ_COMMAND_ERROR_INVALID_ARGUMENT	= 0xFFFFFFF7,
+	GVE_ADMINQ_COMMAND_ERROR_NOT_FOUND		= 0xFFFFFFF8,
+	GVE_ADMINQ_COMMAND_ERROR_OUT_OF_RANGE		= 0xFFFFFFF9,
+	GVE_ADMINQ_COMMAND_ERROR_PERMISSION_DENIED	= 0xFFFFFFFA,
+	GVE_ADMINQ_COMMAND_ERROR_UNAUTHENTICATED	= 0xFFFFFFFB,
+	GVE_ADMINQ_COMMAND_ERROR_RESOURCE_EXHAUSTED	= 0xFFFFFFFC,
+	GVE_ADMINQ_COMMAND_ERROR_UNAVAILABLE		= 0xFFFFFFFD,
+	GVE_ADMINQ_COMMAND_ERROR_UNIMPLEMENTED		= 0xFFFFFFFE,
+	GVE_ADMINQ_COMMAND_ERROR_UNKNOWN_ERROR		= 0xFFFFFFFF,
+};
+
+#define GVE_ADMINQ_DEVICE_DESCRIPTOR_VERSION 1
+
+/* All AdminQ command structs should be naturally packed. The static_assert
+ * calls make sure this is the case at compile time.
+ */
+
+struct gve_adminq_describe_device {
+	__be64 device_descriptor_addr;
+	__be32 device_descriptor_version;
+	__be32 available_length;
+};
+
+static_assert(sizeof(struct gve_adminq_describe_device) == 16);
+
+struct gve_device_descriptor {
+	__be64 max_registered_pages;
+	__be16 reserved1;
+	__be16 tx_queue_entries;
+	__be16 rx_queue_entries;
+	__be16 default_num_queues;
+	__be16 mtu;
+	__be16 counters;
+	__be16 tx_pages_per_qpl;
+	__be16 rx_pages_per_qpl;
+	u8  mac[ETH_ALEN];
+	__be16 num_device_options;
+	__be16 total_length;
+	u8  reserved2[6];
+};
+
+static_assert(sizeof(struct gve_device_descriptor) == 40);
+
+struct gve_device_option {
+	__be16 option_id;
+	__be16 option_length;
+	__be32 required_features_mask;
+};
+
+static_assert(sizeof(struct gve_device_option) == 8);
+
+struct gve_device_option_gqi_rda {
+	__be32 supported_features_mask;
+};
+
+static_assert(sizeof(struct gve_device_option_gqi_rda) == 4);
+
+struct gve_device_option_gqi_qpl {
+	__be32 supported_features_mask;
+};
+
+static_assert(sizeof(struct gve_device_option_gqi_qpl) == 4);
+
+struct gve_device_option_dqo_rda {
+	__be32 supported_features_mask;
+	__be32 reserved;
+};
+
+static_assert(sizeof(struct gve_device_option_dqo_rda) == 8);
+
+struct gve_device_option_dqo_qpl {
+	__be32 supported_features_mask;
+	__be16 tx_pages_per_qpl;
+	__be16 rx_pages_per_qpl;
+};
+
+static_assert(sizeof(struct gve_device_option_dqo_qpl) == 8);
+
+struct gve_device_option_jumbo_frames {
+	__be32 supported_features_mask;
+	__be16 max_mtu;
+	u8 padding[2];
+};
+
+static_assert(sizeof(struct gve_device_option_jumbo_frames) == 8);
+
+struct gve_device_option_buffer_sizes {
+	/* GVE_SUP_BUFFER_SIZES_MASK bit should be set */
+	__be32 supported_features_mask;
+	__be16 packet_buffer_size;
+	__be16 header_buffer_size;
+};
+
+static_assert(sizeof(struct gve_device_option_buffer_sizes) == 8);
+
+struct gve_device_option_modify_ring {
+	__be32 supported_featured_mask;
+	__be16 max_rx_ring_size;
+	__be16 max_tx_ring_size;
+	__be16 min_rx_ring_size;
+	__be16 min_tx_ring_size;
+};
+
+static_assert(sizeof(struct gve_device_option_modify_ring) == 12);
+
+struct gve_device_option_flow_steering {
+	__be32 supported_features_mask;
+	__be32 reserved;
+	__be32 max_flow_rules;
+};
+
+static_assert(sizeof(struct gve_device_option_flow_steering) == 12);
+
+struct gve_device_option_rss_config {
+	__be32 supported_features_mask;
+	__be16 hash_key_size;
+	__be16 hash_lut_size;
+};
+
+static_assert(sizeof(struct gve_device_option_rss_config) == 8);
+
+struct gve_device_option_nic_timestamp {
+	__be32 supported_features_mask;
+};
+
+static_assert(sizeof(struct gve_device_option_nic_timestamp) == 4);
+
+/* Terminology:
+ *
+ * RDA - Raw DMA Addressing - Buffers associated with SKBs are directly DMA
+ *       mapped and read/updated by the device.
+ *
+ * QPL - Queue Page Lists - Driver uses bounce buffers which are DMA mapped with
+ *       the device for read/write and data is copied from/to SKBs.
+ */
+enum gve_dev_opt_id {
+	GVE_DEV_OPT_ID_GQI_RAW_ADDRESSING	= 0x1,
+	GVE_DEV_OPT_ID_GQI_RDA			= 0x2,
+	GVE_DEV_OPT_ID_GQI_QPL			= 0x3,
+	GVE_DEV_OPT_ID_DQO_RDA			= 0x4,
+	GVE_DEV_OPT_ID_MODIFY_RING		= 0x6,
+	GVE_DEV_OPT_ID_DQO_QPL			= 0x7,
+	GVE_DEV_OPT_ID_JUMBO_FRAMES		= 0x8,
+	GVE_DEV_OPT_ID_BUFFER_SIZES		= 0xa,
+	GVE_DEV_OPT_ID_FLOW_STEERING		= 0xb,
+	GVE_DEV_OPT_ID_NIC_TIMESTAMP		= 0xd,
+	GVE_DEV_OPT_ID_RSS_CONFIG		= 0xe,
+};
+
+enum gve_dev_opt_req_feat_mask {
+	GVE_DEV_OPT_REQ_FEAT_MASK_GQI_RAW_ADDRESSING	= 0x0,
+	GVE_DEV_OPT_REQ_FEAT_MASK_GQI_RDA		= 0x0,
+	GVE_DEV_OPT_REQ_FEAT_MASK_GQI_QPL		= 0x0,
+	GVE_DEV_OPT_REQ_FEAT_MASK_DQO_RDA		= 0x0,
+	GVE_DEV_OPT_REQ_FEAT_MASK_JUMBO_FRAMES		= 0x0,
+	GVE_DEV_OPT_REQ_FEAT_MASK_DQO_QPL		= 0x0,
+	GVE_DEV_OPT_REQ_FEAT_MASK_BUFFER_SIZES		= 0x0,
+	GVE_DEV_OPT_REQ_FEAT_MASK_MODIFY_RING		= 0x0,
+	GVE_DEV_OPT_REQ_FEAT_MASK_FLOW_STEERING		= 0x0,
+	GVE_DEV_OPT_REQ_FEAT_MASK_RSS_CONFIG		= 0x0,
+	GVE_DEV_OPT_REQ_FEAT_MASK_NIC_TIMESTAMP		= 0x0,
+};
+
+enum gve_sup_feature_mask {
+	GVE_SUP_MODIFY_RING_MASK	= 1 << 0,
+	GVE_SUP_JUMBO_FRAMES_MASK	= 1 << 2,
+	GVE_SUP_BUFFER_SIZES_MASK	= 1 << 4,
+	GVE_SUP_FLOW_STEERING_MASK	= 1 << 5,
+	GVE_SUP_RSS_CONFIG_MASK		= 1 << 7,
+	GVE_SUP_NIC_TIMESTAMP_MASK	= 1 << 8,
+};
+
+#define GVE_DEV_OPT_LEN_GQI_RAW_ADDRESSING 0x0
+
+#define GVE_VERSION_STR_LEN 128
+
+enum gve_driver_capbility {
+	gve_driver_capability_gqi_qpl = 0,
+	gve_driver_capability_gqi_rda = 1,
+	gve_driver_capability_dqo_qpl = 2, /* reserved for future use */
+	gve_driver_capability_dqo_rda = 3,
+	gve_driver_capability_alt_miss_compl = 4,
+	gve_driver_capability_flexible_buffer_size = 5,
+	gve_driver_capability_flexible_rss_size = 6,
+};
+
+#define GVE_CAP1(a) BIT((int)a)
+#define GVE_CAP2(a) BIT(((int)a) - 64)
+#define GVE_CAP3(a) BIT(((int)a) - 128)
+#define GVE_CAP4(a) BIT(((int)a) - 192)
+
+#define GVE_DRIVER_CAPABILITY_FLAGS1 \
+	(GVE_CAP1(gve_driver_capability_gqi_qpl) | \
+	 GVE_CAP1(gve_driver_capability_gqi_rda) | \
+	 GVE_CAP1(gve_driver_capability_dqo_rda) | \
+	 GVE_CAP1(gve_driver_capability_alt_miss_compl) | \
+	 GVE_CAP1(gve_driver_capability_flexible_buffer_size) | \
+	 GVE_CAP1(gve_driver_capability_flexible_rss_size))
+
+#define GVE_DRIVER_CAPABILITY_FLAGS2 0x0
+#define GVE_DRIVER_CAPABILITY_FLAGS3 0x0
+#define GVE_DRIVER_CAPABILITY_FLAGS4 0x0
+
+struct gve_adminq_extended_command {
+	__be32 inner_opcode;
+	__be32 inner_length;
+	__be64 inner_command_addr;
+};
+
+static_assert(sizeof(struct gve_adminq_extended_command) == 16);
+
+struct gve_driver_info {
+	u8 os_type;	/* 0x01 = Linux */
+	u8 driver_major;
+	u8 driver_minor;
+	u8 driver_sub;
+	__be32 os_version_major;
+	__be32 os_version_minor;
+	__be32 os_version_sub;
+	__be64 driver_capability_flags[4];
+	u8 os_version_str1[GVE_VERSION_STR_LEN];
+	u8 os_version_str2[GVE_VERSION_STR_LEN];
+};
+
+struct gve_adminq_verify_driver_compatibility {
+	__be64 driver_info_len;
+	__be64 driver_info_addr;
+};
+
+static_assert(sizeof(struct gve_adminq_verify_driver_compatibility) == 16);
+
+struct gve_adminq_configure_device_resources {
+	__be64 counter_array;
+	__be64 irq_db_addr;
+	__be32 num_counters;
+	__be32 num_irq_dbs;
+	__be32 irq_db_stride;
+	__be32 ntfy_blk_msix_base_idx;
+	u8 queue_format;
+	u8 padding[7];
+};
+
+static_assert(sizeof(struct gve_adminq_configure_device_resources) == 40);
+
+struct gve_adminq_register_page_list {
+	__be32 page_list_id;
+	__be32 num_pages;
+	__be64 page_address_list_addr;
+	__be64 page_size;
+};
+
+static_assert(sizeof(struct gve_adminq_register_page_list) == 24);
+
+struct gve_adminq_unregister_page_list {
+	__be32 page_list_id;
+};
+
+static_assert(sizeof(struct gve_adminq_unregister_page_list) == 4);
+
+#define GVE_RAW_ADDRESSING_QPL_ID 0xFFFFFFFF
+
+struct gve_adminq_create_tx_queue {
+	__be32 queue_id;
+	__be32 reserved;
+	__be64 queue_resources_addr;
+	__be64 tx_ring_addr;
+	__be32 queue_page_list_id;
+	__be32 ntfy_id;
+	__be64 tx_comp_ring_addr;
+	__be16 tx_ring_size;
+	__be16 tx_comp_ring_size;
+	u8 padding[4];
+};
+
+static_assert(sizeof(struct gve_adminq_create_tx_queue) == 48);
+
+struct gve_adminq_create_rx_queue {
+	__be32 queue_id;
+	__be32 index;
+	__be32 reserved;
+	__be32 ntfy_id;
+	__be64 queue_resources_addr;
+	__be64 rx_desc_ring_addr;
+	__be64 rx_data_ring_addr;
+	__be32 queue_page_list_id;
+	__be16 rx_ring_size;
+	__be16 packet_buffer_size;
+	__be16 rx_buff_ring_size;
+	u8 enable_rsc;
+	u8 padding1;
+	__be16 header_buffer_size;
+	u8 padding2[2];
+};
+
+static_assert(sizeof(struct gve_adminq_create_rx_queue) == 56);
+
+/* Queue resources that are shared with the device */
+struct gve_queue_resources {
+	union {
+		struct {
+			__be32 db_index;	/* Device -> Guest */
+			__be32 counter_index;	/* Device -> Guest */
+		};
+		u8 reserved[64];
+	};
+};
+
+static_assert(sizeof(struct gve_queue_resources) == 64);
+
+struct gve_adminq_destroy_tx_queue {
+	__be32 queue_id;
+};
+
+static_assert(sizeof(struct gve_adminq_destroy_tx_queue) == 4);
+
+struct gve_adminq_destroy_rx_queue {
+	__be32 queue_id;
+};
+
+static_assert(sizeof(struct gve_adminq_destroy_rx_queue) == 4);
+
+/* GVE Set Driver Parameter Types */
+enum gve_set_driver_param_types {
+	GVE_SET_PARAM_MTU	= 0x1,
+};
+
+struct gve_adminq_set_driver_parameter {
+	__be32 parameter_type;
+	u8 reserved[4];
+	__be64 parameter_value;
+};
+
+static_assert(sizeof(struct gve_adminq_set_driver_parameter) == 16);
+
+struct gve_adminq_report_stats {
+	__be64 stats_report_len;
+	__be64 stats_report_addr;
+	__be64 interval;
+};
+
+static_assert(sizeof(struct gve_adminq_report_stats) == 24);
+
+struct gve_adminq_report_link_speed {
+	__be64 link_speed_address;
+};
+
+static_assert(sizeof(struct gve_adminq_report_link_speed) == 8);
+
+struct gve_adminq_report_nic_ts {
+	__be64 nic_ts_report_len;
+	__be64 nic_ts_report_addr;
+};
+
+static_assert(sizeof(struct gve_adminq_report_nic_ts) == 16);
+
+struct gve_nic_ts_report {
+	__be64 nic_timestamp; /* NIC clock in nanoseconds */
+	__be64 reserved1;
+	__be64 reserved2;
+	__be64 reserved3;
+	__be64 reserved4;
+};
+
+struct stats {
+	__be32 stat_name;
+	__be32 queue_id;
+	__be64 value;
+};
+
+static_assert(sizeof(struct stats) == 16);
+
+struct gve_stats_report {
+	__be64 written_count;
+	struct stats stats[];
+};
+
+static_assert(sizeof(struct gve_stats_report) == 8);
+
+enum gve_stat_names {
+	// stats from gve
+	TX_WAKE_CNT			= 1,
+	TX_STOP_CNT			= 2,
+	TX_FRAMES_SENT			= 3,
+	TX_BYTES_SENT			= 4,
+	TX_LAST_COMPLETION_PROCESSED	= 5,
+	RX_NEXT_EXPECTED_SEQUENCE	= 6,
+	RX_BUFFERS_POSTED		= 7,
+	TX_TIMEOUT_CNT			= 8,
+	// stats from NIC
+	RX_QUEUE_DROP_CNT		= 65,
+	RX_NO_BUFFERS_POSTED		= 66,
+	RX_DROPS_PACKET_OVER_MRU	= 67,
+	RX_DROPS_INVALID_CHECKSUM	= 68,
+};
+
+enum gve_l3_type {
+	/* Must be zero so zero initialized LUT is unknown. */
+	GVE_L3_TYPE_UNKNOWN = 0,
+	GVE_L3_TYPE_OTHER,
+	GVE_L3_TYPE_IPV4,
+	GVE_L3_TYPE_IPV6,
+};
+
+enum gve_l4_type {
+	/* Must be zero so zero initialized LUT is unknown. */
+	GVE_L4_TYPE_UNKNOWN = 0,
+	GVE_L4_TYPE_OTHER,
+	GVE_L4_TYPE_TCP,
+	GVE_L4_TYPE_UDP,
+	GVE_L4_TYPE_ICMP,
+	GVE_L4_TYPE_SCTP,
+};
+
+/* These are control path types for PTYPE which are the same as the data path
+ * types.
+ */
+struct gve_ptype_entry {
+	u8 l3_type;
+	u8 l4_type;
+};
+
+struct gve_ptype_map {
+	struct gve_ptype_entry ptypes[GVE_NUM_PTYPES]; /* PTYPES are always 10 bits. */
+};
+
+struct gve_adminq_get_ptype_map {
+	__be64 ptype_map_len;
+	__be64 ptype_map_addr;
+};
+
+/* Flow-steering related definitions */
+enum gve_adminq_flow_rule_cfg_opcode {
+	GVE_FLOW_RULE_CFG_ADD	= 0,
+	GVE_FLOW_RULE_CFG_DEL	= 1,
+	GVE_FLOW_RULE_CFG_RESET	= 2,
+};
+
+enum gve_adminq_flow_rule_query_opcode {
+	GVE_FLOW_RULE_QUERY_RULES	= 0,
+	GVE_FLOW_RULE_QUERY_IDS		= 1,
+	GVE_FLOW_RULE_QUERY_STATS	= 2,
+};
+
+enum gve_adminq_flow_type {
+	GVE_FLOW_TYPE_TCPV4,
+	GVE_FLOW_TYPE_UDPV4,
+	GVE_FLOW_TYPE_SCTPV4,
+	GVE_FLOW_TYPE_AHV4,
+	GVE_FLOW_TYPE_ESPV4,
+	GVE_FLOW_TYPE_TCPV6,
+	GVE_FLOW_TYPE_UDPV6,
+	GVE_FLOW_TYPE_SCTPV6,
+	GVE_FLOW_TYPE_AHV6,
+	GVE_FLOW_TYPE_ESPV6,
+};
+
+/* Flow-steering command */
+struct gve_adminq_flow_rule {
+	__be16 flow_type;
+	__be16 action; /* RX queue id */
+	struct gve_flow_spec key;
+	struct gve_flow_spec mask;
+};
+
+struct gve_adminq_configure_flow_rule {
+	__be16 opcode;
+	u8 padding[2];
+	struct gve_adminq_flow_rule rule;
+	__be32 location;
+};
+
+static_assert(sizeof(struct gve_adminq_configure_flow_rule) == 92);
+
+struct gve_query_flow_rules_descriptor {
+	__be32 num_flow_rules;
+	__be32 max_flow_rules;
+	__be32 num_queried_rules;
+	__be32 total_length;
+};
+
+struct gve_adminq_queried_flow_rule {
+	__be32 location;
+	struct gve_adminq_flow_rule flow_rule;
+};
+
+struct gve_adminq_query_flow_rules {
+	__be16 opcode;
+	u8 padding[2];
+	__be32 starting_rule_id;
+	__be64 available_length; /* The dma memory length that the driver allocated */
+	__be64 rule_descriptor_addr; /* The dma memory address */
+};
+
+static_assert(sizeof(struct gve_adminq_query_flow_rules) == 24);
+
+enum gve_rss_hash_type {
+	GVE_RSS_HASH_IPV4,
+	GVE_RSS_HASH_TCPV4,
+	GVE_RSS_HASH_IPV6,
+	GVE_RSS_HASH_IPV6_EX,
+	GVE_RSS_HASH_TCPV6,
+	GVE_RSS_HASH_TCPV6_EX,
+	GVE_RSS_HASH_UDPV4,
+	GVE_RSS_HASH_UDPV6,
+	GVE_RSS_HASH_UDPV6_EX,
+};
+
+struct gve_adminq_configure_rss {
+	__be16 hash_types;
+	u8 hash_alg;
+	u8 reserved;
+	__be16 hash_key_size;
+	__be16 hash_lut_size;
+	__be64 hash_key_addr;
+	__be64 hash_lut_addr;
+};
+
+static_assert(sizeof(struct gve_adminq_configure_rss) == 24);
+
+struct gve_query_rss_descriptor {
+	__be32 total_length;
+	__be16 hash_types;
+	u8 hash_alg;
+	u8 reserved;
+};
+
+struct gve_adminq_query_rss {
+	__be64 available_length;
+	__be64 rss_descriptor_addr;
+};
+
+static_assert(sizeof(struct gve_adminq_query_rss) == 16);
+
+union gve_adminq_command {
+	struct {
+		__be32 opcode;
+		__be32 status;
+		union {
+			struct gve_adminq_configure_device_resources
+						configure_device_resources;
+			struct gve_adminq_create_tx_queue create_tx_queue;
+			struct gve_adminq_create_rx_queue create_rx_queue;
+			struct gve_adminq_destroy_tx_queue destroy_tx_queue;
+			struct gve_adminq_destroy_rx_queue destroy_rx_queue;
+			struct gve_adminq_describe_device describe_device;
+			struct gve_adminq_register_page_list reg_page_list;
+			struct gve_adminq_unregister_page_list unreg_page_list;
+			struct gve_adminq_set_driver_parameter set_driver_param;
+			struct gve_adminq_report_stats report_stats;
+			struct gve_adminq_report_link_speed report_link_speed;
+			struct gve_adminq_get_ptype_map get_ptype_map;
+			struct gve_adminq_verify_driver_compatibility
+						verify_driver_compatibility;
+			struct gve_adminq_query_flow_rules query_flow_rules;
+			struct gve_adminq_configure_rss configure_rss;
+			struct gve_adminq_query_rss query_rss;
+			struct gve_adminq_report_nic_ts report_nic_ts;
+			struct gve_adminq_extended_command extended_command;
+		};
+	};
+	u8 reserved[64];
+};
+
+static_assert(sizeof(union gve_adminq_command) == 64);
+
+int gve_adminq_alloc(struct device *dev, struct gve_priv *priv);
+void gve_adminq_free(struct device *dev, struct gve_priv *priv);
+void gve_adminq_release(struct gve_priv *priv);
+int gve_adminq_describe_device(struct gve_priv *priv);
+int gve_adminq_configure_device_resources(struct gve_priv *priv,
+					  dma_addr_t counter_array_bus_addr,
+					  u32 num_counters,
+					  dma_addr_t db_array_bus_addr,
+					  u32 num_ntfy_blks);
+int gve_adminq_deconfigure_device_resources(struct gve_priv *priv);
+int gve_adminq_create_tx_queues(struct gve_priv *priv, u32 start_id, u32 num_queues);
+int gve_adminq_destroy_tx_queues(struct gve_priv *priv, u32 start_id, u32 num_queues);
+int gve_adminq_create_single_rx_queue(struct gve_priv *priv, u32 queue_index);
+int gve_adminq_create_rx_queues(struct gve_priv *priv, u32 num_queues);
+int gve_adminq_destroy_single_rx_queue(struct gve_priv *priv, u32 queue_index);
+int gve_adminq_destroy_rx_queues(struct gve_priv *priv, u32 queue_id);
+int gve_adminq_register_page_list(struct gve_priv *priv,
+				  struct gve_queue_page_list *qpl);
+int gve_adminq_unregister_page_list(struct gve_priv *priv, u32 page_list_id);
+int gve_adminq_report_stats(struct gve_priv *priv, u64 stats_report_len,
+			    dma_addr_t stats_report_addr, u64 interval);
+int gve_adminq_verify_driver_compatibility(struct gve_priv *priv,
+					   u64 driver_info_len,
+					   dma_addr_t driver_info_addr);
+int gve_adminq_report_link_speed(struct gve_priv *priv);
+int gve_adminq_add_flow_rule(struct gve_priv *priv, struct gve_adminq_flow_rule *rule, u32 loc);
+int gve_adminq_del_flow_rule(struct gve_priv *priv, u32 loc);
+int gve_adminq_reset_flow_rules(struct gve_priv *priv);
+int gve_adminq_query_flow_rules(struct gve_priv *priv, u16 query_opcode, u32 starting_loc);
+int gve_adminq_configure_rss(struct gve_priv *priv, struct ethtool_rxfh_param *rxfh);
+int gve_adminq_query_rss_config(struct gve_priv *priv, struct ethtool_rxfh_param *rxfh);
+int gve_adminq_report_nic_ts(struct gve_priv *priv,
+			     dma_addr_t nic_ts_report_addr);
+
+struct gve_ptype_lut;
+int gve_adminq_get_ptype_map_dqo(struct gve_priv *priv,
+				 struct gve_ptype_lut *ptype_lut);
+
+#endif /* _GVE_ADMINQ_H */
diff --git a/drivers/net/ethernet/google/gve/gve_buffer_mgmt_dqo.c b/drivers/net/ethernet/google/gve/gve_buffer_mgmt_dqo.c
new file mode 100644
index 000000000000..0e2b703c673a
--- /dev/null
+++ b/drivers/net/ethernet/google/gve/gve_buffer_mgmt_dqo.c
@@ -0,0 +1,344 @@
+// SPDX-License-Identifier: (GPL-2.0 OR MIT)
+/* Google virtual Ethernet (gve) driver
+ *
+ * Copyright (C) 2015-2024 Google, Inc.
+ */
+
+#include <net/xdp_sock_drv.h>
+#include "gve.h"
+#include "gve_utils.h"
+
+int gve_buf_ref_cnt(struct gve_rx_buf_state_dqo *bs)
+{
+	return page_count(bs->page_info.page) - bs->page_info.pagecnt_bias;
+}
+
+struct gve_rx_buf_state_dqo *gve_alloc_buf_state(struct gve_rx_ring *rx)
+{
+	struct gve_rx_buf_state_dqo *buf_state;
+	s16 buffer_id;
+
+	buffer_id = rx->dqo.free_buf_states;
+	if (unlikely(buffer_id == -1))
+		return NULL;
+
+	buf_state = &rx->dqo.buf_states[buffer_id];
+
+	/* Remove buf_state from free list */
+	rx->dqo.free_buf_states = buf_state->next;
+
+	/* Point buf_state to itself to mark it as allocated */
+	buf_state->next = buffer_id;
+
+	/* Clear the buffer pointers */
+	buf_state->page_info.page = NULL;
+	buf_state->xsk_buff = NULL;
+
+	return buf_state;
+}
+
+bool gve_buf_state_is_allocated(struct gve_rx_ring *rx,
+				struct gve_rx_buf_state_dqo *buf_state)
+{
+	s16 buffer_id = buf_state - rx->dqo.buf_states;
+
+	return buf_state->next == buffer_id;
+}
+
+void gve_free_buf_state(struct gve_rx_ring *rx,
+			struct gve_rx_buf_state_dqo *buf_state)
+{
+	s16 buffer_id = buf_state - rx->dqo.buf_states;
+
+	buf_state->next = rx->dqo.free_buf_states;
+	rx->dqo.free_buf_states = buffer_id;
+}
+
+struct gve_rx_buf_state_dqo *gve_dequeue_buf_state(struct gve_rx_ring *rx,
+						   struct gve_index_list *list)
+{
+	struct gve_rx_buf_state_dqo *buf_state;
+	s16 buffer_id;
+
+	buffer_id = list->head;
+	if (unlikely(buffer_id == -1))
+		return NULL;
+
+	buf_state = &rx->dqo.buf_states[buffer_id];
+
+	/* Remove buf_state from list */
+	list->head = buf_state->next;
+	if (buf_state->next == -1)
+		list->tail = -1;
+
+	/* Point buf_state to itself to mark it as allocated */
+	buf_state->next = buffer_id;
+
+	return buf_state;
+}
+
+void gve_enqueue_buf_state(struct gve_rx_ring *rx, struct gve_index_list *list,
+			   struct gve_rx_buf_state_dqo *buf_state)
+{
+	s16 buffer_id = buf_state - rx->dqo.buf_states;
+
+	buf_state->next = -1;
+
+	if (list->head == -1) {
+		list->head = buffer_id;
+		list->tail = buffer_id;
+	} else {
+		int tail = list->tail;
+
+		rx->dqo.buf_states[tail].next = buffer_id;
+		list->tail = buffer_id;
+	}
+}
+
+struct gve_rx_buf_state_dqo *gve_get_recycled_buf_state(struct gve_rx_ring *rx)
+{
+	struct gve_rx_buf_state_dqo *buf_state;
+	int i;
+
+	/* Recycled buf states are immediately usable. */
+	buf_state = gve_dequeue_buf_state(rx, &rx->dqo.recycled_buf_states);
+	if (likely(buf_state))
+		return buf_state;
+
+	if (unlikely(rx->dqo.used_buf_states.head == -1))
+		return NULL;
+
+	/* Used buf states are only usable when ref count reaches 0, which means
+	 * no SKBs refer to them.
+	 *
+	 * Search a limited number before giving up.
+	 */
+	for (i = 0; i < 5; i++) {
+		buf_state = gve_dequeue_buf_state(rx, &rx->dqo.used_buf_states);
+		if (gve_buf_ref_cnt(buf_state) == 0) {
+			rx->dqo.used_buf_states_cnt--;
+			return buf_state;
+		}
+
+		gve_enqueue_buf_state(rx, &rx->dqo.used_buf_states, buf_state);
+	}
+
+	return NULL;
+}
+
+int gve_alloc_qpl_page_dqo(struct gve_rx_ring *rx,
+			   struct gve_rx_buf_state_dqo *buf_state)
+{
+	struct gve_priv *priv = rx->gve;
+	u32 idx;
+
+	idx = rx->dqo.next_qpl_page_idx;
+	if (idx >= gve_get_rx_pages_per_qpl_dqo(priv->rx_desc_cnt)) {
+		net_err_ratelimited("%s: Out of QPL pages\n",
+				    priv->dev->name);
+		return -ENOMEM;
+	}
+	buf_state->page_info.page = rx->dqo.qpl->pages[idx];
+	buf_state->addr = rx->dqo.qpl->page_buses[idx];
+	rx->dqo.next_qpl_page_idx++;
+	buf_state->page_info.page_offset = 0;
+	buf_state->page_info.page_address =
+		page_address(buf_state->page_info.page);
+	buf_state->page_info.buf_size = rx->packet_buffer_truesize;
+	buf_state->page_info.pad = rx->rx_headroom;
+	buf_state->last_single_ref_offset = 0;
+
+	/* The page already has 1 ref. */
+	page_ref_add(buf_state->page_info.page, INT_MAX - 1);
+	buf_state->page_info.pagecnt_bias = INT_MAX;
+
+	return 0;
+}
+
+void gve_free_qpl_page_dqo(struct gve_rx_buf_state_dqo *buf_state)
+{
+	if (!buf_state->page_info.page)
+		return;
+
+	page_ref_sub(buf_state->page_info.page,
+		     buf_state->page_info.pagecnt_bias - 1);
+	buf_state->page_info.page = NULL;
+}
+
+void gve_try_recycle_buf(struct gve_priv *priv, struct gve_rx_ring *rx,
+			 struct gve_rx_buf_state_dqo *buf_state)
+{
+	const u16 data_buffer_size = rx->packet_buffer_truesize;
+	int pagecount;
+
+	/* Can't reuse if we only fit one buffer per page */
+	if (data_buffer_size * 2 > PAGE_SIZE)
+		goto mark_used;
+
+	pagecount = gve_buf_ref_cnt(buf_state);
+
+	/* Record the offset when we have a single remaining reference.
+	 *
+	 * When this happens, we know all of the other offsets of the page are
+	 * usable.
+	 */
+	if (pagecount == 1) {
+		buf_state->last_single_ref_offset =
+			buf_state->page_info.page_offset;
+	}
+
+	/* Use the next buffer sized chunk in the page. */
+	buf_state->page_info.page_offset += data_buffer_size;
+	buf_state->page_info.page_offset &= (PAGE_SIZE - 1);
+
+	/* If we wrap around to the same offset without ever dropping to 1
+	 * reference, then we don't know if this offset was ever freed.
+	 */
+	if (buf_state->page_info.page_offset ==
+	    buf_state->last_single_ref_offset) {
+		goto mark_used;
+	}
+
+	gve_enqueue_buf_state(rx, &rx->dqo.recycled_buf_states, buf_state);
+	return;
+
+mark_used:
+	gve_enqueue_buf_state(rx, &rx->dqo.used_buf_states, buf_state);
+	rx->dqo.used_buf_states_cnt++;
+}
+
+void gve_free_to_page_pool(struct gve_rx_ring *rx,
+			   struct gve_rx_buf_state_dqo *buf_state,
+			   bool allow_direct)
+{
+	netmem_ref netmem = buf_state->page_info.netmem;
+
+	if (!netmem)
+		return;
+
+	page_pool_put_full_netmem(netmem_get_pp(netmem), netmem, allow_direct);
+	buf_state->page_info.netmem = 0;
+}
+
+static int gve_alloc_from_page_pool(struct gve_rx_ring *rx,
+				    struct gve_rx_buf_state_dqo *buf_state)
+{
+	netmem_ref netmem;
+
+	buf_state->page_info.buf_size = rx->packet_buffer_truesize;
+	netmem = page_pool_alloc_netmem(rx->dqo.page_pool,
+					&buf_state->page_info.page_offset,
+					&buf_state->page_info.buf_size,
+					GFP_ATOMIC);
+
+	if (!netmem)
+		return -ENOMEM;
+
+	buf_state->page_info.netmem = netmem;
+	buf_state->page_info.page_address = netmem_address(netmem);
+	buf_state->addr = page_pool_get_dma_addr_netmem(netmem);
+	buf_state->page_info.pad = rx->dqo.page_pool->p.offset;
+
+	return 0;
+}
+
+struct page_pool *gve_rx_create_page_pool(struct gve_priv *priv,
+					  struct gve_rx_ring *rx,
+					  bool xdp)
+{
+	u32 ntfy_id = gve_rx_idx_to_ntfy(priv, rx->q_num);
+	struct page_pool_params pp = {
+		.flags = PP_FLAG_DMA_MAP | PP_FLAG_DMA_SYNC_DEV,
+		.order = 0,
+		.pool_size = GVE_PAGE_POOL_SIZE_MULTIPLIER * priv->rx_desc_cnt,
+		.nid = priv->numa_node,
+		.dev = &priv->pdev->dev,
+		.netdev = priv->dev,
+		.napi = &priv->ntfy_blocks[ntfy_id].napi,
+		.max_len = PAGE_SIZE,
+		.dma_dir = xdp ? DMA_BIDIRECTIONAL : DMA_FROM_DEVICE,
+		.offset = xdp ? XDP_PACKET_HEADROOM : 0,
+	};
+
+	if (priv->header_split_enabled) {
+		pp.flags |= PP_FLAG_ALLOW_UNREADABLE_NETMEM;
+		pp.queue_idx = rx->q_num;
+	}
+
+	return page_pool_create(&pp);
+}
+
+void gve_free_buffer(struct gve_rx_ring *rx,
+		     struct gve_rx_buf_state_dqo *buf_state)
+{
+	if (rx->dqo.page_pool) {
+		gve_free_to_page_pool(rx, buf_state, true);
+		gve_free_buf_state(rx, buf_state);
+	} else {
+		gve_enqueue_buf_state(rx, &rx->dqo.recycled_buf_states,
+				      buf_state);
+	}
+}
+
+void gve_reuse_buffer(struct gve_rx_ring *rx,
+		      struct gve_rx_buf_state_dqo *buf_state)
+{
+	if (rx->dqo.page_pool) {
+		buf_state->page_info.netmem = 0;
+		gve_free_buf_state(rx, buf_state);
+	} else {
+		gve_dec_pagecnt_bias(&buf_state->page_info);
+		gve_try_recycle_buf(rx->gve, rx, buf_state);
+	}
+}
+
+int gve_alloc_buffer(struct gve_rx_ring *rx, struct gve_rx_desc_dqo *desc)
+{
+	struct gve_rx_buf_state_dqo *buf_state;
+
+	if (rx->xsk_pool) {
+		buf_state = gve_alloc_buf_state(rx);
+		if (unlikely(!buf_state))
+			return -ENOMEM;
+
+		buf_state->xsk_buff = xsk_buff_alloc(rx->xsk_pool);
+		if (unlikely(!buf_state->xsk_buff)) {
+			xsk_set_rx_need_wakeup(rx->xsk_pool);
+			gve_free_buf_state(rx, buf_state);
+			return -ENOMEM;
+		}
+		/* Allocated xsk buffer. Clear wakeup in case it was set. */
+		xsk_clear_rx_need_wakeup(rx->xsk_pool);
+		desc->buf_id = cpu_to_le16(buf_state - rx->dqo.buf_states);
+		desc->buf_addr =
+			cpu_to_le64(xsk_buff_xdp_get_dma(buf_state->xsk_buff));
+		return 0;
+	} else if (rx->dqo.page_pool) {
+		buf_state = gve_alloc_buf_state(rx);
+		if (WARN_ON_ONCE(!buf_state))
+			return -ENOMEM;
+
+		if (gve_alloc_from_page_pool(rx, buf_state))
+			goto free_buf_state;
+	} else {
+		buf_state = gve_get_recycled_buf_state(rx);
+		if (unlikely(!buf_state)) {
+			buf_state = gve_alloc_buf_state(rx);
+			if (unlikely(!buf_state))
+				return -ENOMEM;
+
+			if (unlikely(gve_alloc_qpl_page_dqo(rx, buf_state)))
+				goto free_buf_state;
+		}
+	}
+	desc->buf_id = cpu_to_le16(buf_state - rx->dqo.buf_states);
+	desc->buf_addr = cpu_to_le64(buf_state->addr +
+				     buf_state->page_info.page_offset +
+				     buf_state->page_info.pad);
+
+	return 0;
+
+free_buf_state:
+	gve_free_buf_state(rx, buf_state);
+	return -ENOMEM;
+}
diff --git a/drivers/net/ethernet/google/gve/gve_desc.h b/drivers/net/ethernet/google/gve/gve_desc.h
new file mode 100644
index 000000000000..c2874cdcf40c
--- /dev/null
+++ b/drivers/net/ethernet/google/gve/gve_desc.h
@@ -0,0 +1,139 @@
+/* SPDX-License-Identifier: (GPL-2.0 OR MIT)
+ * Google virtual Ethernet (gve) driver
+ *
+ * Copyright (C) 2015-2019 Google, Inc.
+ */
+
+/* GVE Transmit Descriptor formats */
+
+#ifndef _GVE_DESC_H_
+#define _GVE_DESC_H_
+
+#include <linux/build_bug.h>
+
+/* A note on seg_addrs
+ *
+ * Base addresses encoded in seg_addr are not assumed to be physical
+ * addresses. The ring format assumes these come from some linear address
+ * space. This could be physical memory, kernel virtual memory, user virtual
+ * memory.
+ * If raw dma addressing is not supported then gVNIC uses lists of registered
+ * pages. Each queue is assumed to be associated with a single such linear
+ * address space to ensure a consistent meaning for seg_addrs posted to its
+ * rings.
+ */
+
+struct gve_tx_pkt_desc {
+	u8	type_flags;  /* desc type is lower 4 bits, flags upper */
+	u8	l4_csum_offset;  /* relative offset of L4 csum word */
+	u8	l4_hdr_offset;  /* Offset of start of L4 headers in packet */
+	u8	desc_cnt;  /* Total descriptors for this packet */
+	__be16	len;  /* Total length of this packet (in bytes) */
+	__be16	seg_len;  /* Length of this descriptor's segment */
+	__be64	seg_addr;  /* Base address (see note) of this segment */
+} __packed;
+
+struct gve_tx_mtd_desc {
+	u8      type_flags;     /* type is lower 4 bits, subtype upper  */
+	u8      path_state;     /* state is lower 4 bits, hash type upper */
+	__be16  reserved0;
+	__be32  path_hash;
+	__be64  reserved1;
+} __packed;
+
+struct gve_tx_seg_desc {
+	u8	type_flags;	/* type is lower 4 bits, flags upper	*/
+	u8	l3_offset;	/* TSO: 2 byte units to start of IPH	*/
+	__be16	reserved;
+	__be16	mss;		/* TSO MSS				*/
+	__be16	seg_len;
+	__be64	seg_addr;
+} __packed;
+
+/* GVE Transmit Descriptor Types */
+#define	GVE_TXD_STD		(0x0 << 4) /* Std with Host Address	*/
+#define	GVE_TXD_TSO		(0x1 << 4) /* TSO with Host Address	*/
+#define	GVE_TXD_SEG		(0x2 << 4) /* Seg with Host Address	*/
+#define	GVE_TXD_MTD		(0x3 << 4) /* Metadata			*/
+
+/* GVE Transmit Descriptor Flags for Std Pkts */
+#define	GVE_TXF_L4CSUM	BIT(0)	/* Need csum offload */
+#define	GVE_TXF_TSTAMP	BIT(2)	/* Timestamp required */
+
+/* GVE Transmit Descriptor Flags for TSO Segs */
+#define	GVE_TXSF_IPV6	BIT(1)	/* IPv6 TSO */
+
+/* GVE Transmit Descriptor Options for MTD Segs */
+#define GVE_MTD_SUBTYPE_PATH		0
+
+#define GVE_MTD_PATH_STATE_DEFAULT	0
+#define GVE_MTD_PATH_STATE_TIMEOUT	1
+#define GVE_MTD_PATH_STATE_CONGESTION	2
+#define GVE_MTD_PATH_STATE_RETRANSMIT	3
+
+#define GVE_MTD_PATH_HASH_NONE         (0x0 << 4)
+#define GVE_MTD_PATH_HASH_L4           (0x1 << 4)
+
+/* GVE Receive Packet Descriptor */
+/* The start of an ethernet packet comes 2 bytes into the rx buffer.
+ * gVNIC adds this padding so that both the DMA and the L3/4 protocol header
+ * access is aligned.
+ */
+#define GVE_RX_PAD 2
+
+struct gve_rx_desc {
+	u8	padding[48];
+	__be32	rss_hash;  /* Receive-side scaling hash (Toeplitz for gVNIC) */
+	__be16	mss;
+	__be16	reserved;  /* Reserved to zero */
+	u8	hdr_len;  /* Header length (L2-L4) including padding */
+	u8	hdr_off;  /* 64-byte-scaled offset into RX_DATA entry */
+	__sum16	csum;  /* 1's-complement partial checksum of L3+ bytes */
+	__be16	len;  /* Length of the received packet */
+	__be16	flags_seq;  /* Flags [15:3] and sequence number [2:0] (1-7) */
+} __packed;
+static_assert(sizeof(struct gve_rx_desc) == 64);
+
+/* If the device supports raw dma addressing then the addr in data slot is
+ * the dma address of the buffer.
+ * If the device only supports registered segments then the addr is a byte
+ * offset into the registered segment (an ordered list of pages) where the
+ * buffer is.
+ */
+union gve_rx_data_slot {
+	__be64 qpl_offset;
+	__be64 addr;
+};
+
+/* GVE Receive Packet Descriptor Seq No */
+#define GVE_SEQNO(x) (be16_to_cpu(x) & 0x7)
+
+/* GVE Receive Packet Descriptor Flags */
+#define GVE_RXFLG(x)	cpu_to_be16(1 << (3 + (x)))
+#define	GVE_RXF_FRAG		GVE_RXFLG(3)	/* IP Fragment			*/
+#define	GVE_RXF_IPV4		GVE_RXFLG(4)	/* IPv4				*/
+#define	GVE_RXF_IPV6		GVE_RXFLG(5)	/* IPv6				*/
+#define	GVE_RXF_TCP		GVE_RXFLG(6)	/* TCP Packet			*/
+#define	GVE_RXF_UDP		GVE_RXFLG(7)	/* UDP Packet			*/
+#define	GVE_RXF_ERR		GVE_RXFLG(8)	/* Packet Error Detected	*/
+#define	GVE_RXF_PKT_CONT	GVE_RXFLG(10)	/* Multi Fragment RX packet	*/
+
+/* GVE IRQ */
+#define GVE_IRQ_ACK	BIT(31)
+#define GVE_IRQ_MASK	BIT(30)
+#define GVE_IRQ_EVENT	BIT(29)
+
+static inline bool gve_needs_rss(__be16 flag)
+{
+	if (flag & GVE_RXF_FRAG)
+		return false;
+	if (flag & (GVE_RXF_IPV4 | GVE_RXF_IPV6))
+		return true;
+	return false;
+}
+
+static inline u8 gve_next_seqno(u8 seq)
+{
+	return (seq + 1) == 8 ? 1 : seq + 1;
+}
+#endif /* _GVE_DESC_H_ */
diff --git a/drivers/net/ethernet/google/gve/gve_desc_dqo.h b/drivers/net/ethernet/google/gve/gve_desc_dqo.h
new file mode 100644
index 000000000000..d17da841b5a0
--- /dev/null
+++ b/drivers/net/ethernet/google/gve/gve_desc_dqo.h
@@ -0,0 +1,262 @@
+/* SPDX-License-Identifier: (GPL-2.0 OR MIT)
+ * Google virtual Ethernet (gve) driver
+ *
+ * Copyright (C) 2015-2021 Google, Inc.
+ */
+
+/* GVE DQO Descriptor formats */
+
+#ifndef _GVE_DESC_DQO_H_
+#define _GVE_DESC_DQO_H_
+
+#include <linux/build_bug.h>
+
+#define GVE_TX_MAX_HDR_SIZE_DQO 255
+#define GVE_TX_MIN_TSO_MSS_DQO 88
+
+#ifndef __LITTLE_ENDIAN_BITFIELD
+#error "Only little endian supported"
+#endif
+
+/* Basic TX descriptor (DTYPE 0x0C) */
+struct gve_tx_pkt_desc_dqo {
+	__le64 buf_addr;
+
+	/* Must be GVE_TX_PKT_DESC_DTYPE_DQO (0xc) */
+	u8 dtype: 5;
+
+	/* Denotes the last descriptor of a packet. */
+	u8 end_of_packet: 1;
+	u8 checksum_offload_enable: 1;
+
+	/* If set, will generate a descriptor completion for this descriptor. */
+	u8 report_event: 1;
+	u8 reserved0;
+	__le16 reserved1;
+
+	/* The TX completion associated with this packet will contain this tag.
+	 */
+	__le16 compl_tag;
+	u16 buf_size: 14;
+	u16 reserved2: 2;
+} __packed;
+static_assert(sizeof(struct gve_tx_pkt_desc_dqo) == 16);
+
+#define GVE_TX_PKT_DESC_DTYPE_DQO 0xc
+#define GVE_TX_MAX_BUF_SIZE_DQO ((16 * 1024) - 1)
+
+/* Maximum number of data descriptors allowed per packet, or per-TSO segment. */
+#define GVE_TX_MAX_DATA_DESCS 10
+
+/* Min gap between tail and head to avoid cacheline overlap */
+#define GVE_TX_MIN_DESC_PREVENT_CACHE_OVERLAP 4
+
+/* "report_event" on TX packet descriptors may only be reported on the last
+ * descriptor of a TX packet, and they must be spaced apart with at least this
+ * value.
+ */
+#define GVE_TX_MIN_RE_INTERVAL 32
+
+struct gve_tx_context_cmd_dtype {
+	u8 dtype: 5;
+	u8 tso: 1;
+	u8 reserved1: 2;
+
+	u8 reserved2;
+};
+
+static_assert(sizeof(struct gve_tx_context_cmd_dtype) == 2);
+
+/* TX Native TSO Context DTYPE (0x05)
+ *
+ * "flex" fields allow the driver to send additional packet context to HW.
+ */
+struct gve_tx_tso_context_desc_dqo {
+	/* The L4 payload bytes that should be segmented. */
+	u32 tso_total_len: 24;
+	u32 flex10: 8;
+
+	/* Max segment size in TSO excluding headers. */
+	u16 mss: 14;
+	u16 reserved: 2;
+
+	u8 header_len; /* Header length to use for TSO offload */
+	u8 flex11;
+	struct gve_tx_context_cmd_dtype cmd_dtype;
+	u8 flex0;
+	u8 flex5;
+	u8 flex6;
+	u8 flex7;
+	u8 flex8;
+	u8 flex9;
+} __packed;
+static_assert(sizeof(struct gve_tx_tso_context_desc_dqo) == 16);
+
+#define GVE_TX_TSO_CTX_DESC_DTYPE_DQO 0x5
+
+/* General context descriptor for sending metadata. */
+struct gve_tx_general_context_desc_dqo {
+	u8 flex4;
+	u8 flex5;
+	u8 flex6;
+	u8 flex7;
+	u8 flex8;
+	u8 flex9;
+	u8 flex10;
+	u8 flex11;
+	struct gve_tx_context_cmd_dtype cmd_dtype;
+	u16 reserved;
+	u8 flex0;
+	u8 flex1;
+	u8 flex2;
+	u8 flex3;
+} __packed;
+static_assert(sizeof(struct gve_tx_general_context_desc_dqo) == 16);
+
+#define GVE_TX_GENERAL_CTX_DESC_DTYPE_DQO 0x4
+
+/* Logical structure of metadata which is packed into context descriptor flex
+ * fields.
+ */
+struct gve_tx_metadata_dqo {
+	union {
+		struct {
+			u8 version;
+
+			/* If `skb->l4_hash` is set, this value should be
+			 * derived from `skb->hash`.
+			 *
+			 * A zero value means no l4_hash was associated with the
+			 * skb.
+			 */
+			u16 path_hash: 15;
+
+			/* Should be set to 1 if the flow associated with the
+			 * skb had a rehash from the TCP stack.
+			 */
+			u16 rehash_event: 1;
+		}  __packed;
+		u8 bytes[12];
+	};
+}  __packed;
+static_assert(sizeof(struct gve_tx_metadata_dqo) == 12);
+
+#define GVE_TX_METADATA_VERSION_DQO 0
+
+/* TX completion descriptor */
+struct gve_tx_compl_desc {
+	/* For types 0-4 this is the TX queue ID associated with this
+	 * completion.
+	 */
+	u16 id: 11;
+
+	/* See: GVE_COMPL_TYPE_DQO* */
+	u16 type: 3;
+	u16 reserved0: 1;
+
+	/* Flipped by HW to notify the descriptor is populated. */
+	u16 generation: 1;
+	union {
+		/* For descriptor completions, this is the last index fetched
+		 * by HW + 1.
+		 */
+		__le16 tx_head;
+
+		/* For packet completions, this is the completion tag set on the
+		 * TX packet descriptors.
+		 */
+		__le16 completion_tag;
+	};
+	__le32 reserved1;
+} __packed;
+static_assert(sizeof(struct gve_tx_compl_desc) == 8);
+
+#define GVE_COMPL_TYPE_DQO_PKT 0x2 /* Packet completion */
+#define GVE_COMPL_TYPE_DQO_DESC 0x4 /* Descriptor completion */
+#define GVE_COMPL_TYPE_DQO_MISS 0x1 /* Miss path completion */
+#define GVE_COMPL_TYPE_DQO_REINJECTION 0x3 /* Re-injection completion */
+
+/* The most significant bit in the completion tag can change the completion
+ * type from packet completion to miss path completion.
+ */
+#define GVE_ALT_MISS_COMPL_BIT BIT(15)
+
+/* Descriptor to post buffers to HW on buffer queue. */
+struct gve_rx_desc_dqo {
+	__le16 buf_id; /* ID returned in Rx completion descriptor */
+	__le16 reserved0;
+	__le32 reserved1;
+	__le64 buf_addr; /* DMA address of the buffer */
+	__le64 header_buf_addr;
+	__le64 reserved2;
+} __packed;
+static_assert(sizeof(struct gve_rx_desc_dqo) == 32);
+
+/* Descriptor for HW to notify SW of new packets received on RX queue. */
+struct gve_rx_compl_desc_dqo {
+	/* Must be 1 */
+	u8 rxdid: 4;
+	u8 reserved0: 4;
+
+	/* Packet originated from this system rather than the network. */
+	u8 loopback: 1;
+	/* Set when IPv6 packet contains a destination options header or routing
+	 * header.
+	 */
+	u8 ipv6_ex_add: 1;
+	/* Invalid packet was received. */
+	u8 rx_error: 1;
+	u8 reserved1: 5;
+
+	u16 packet_type: 10;
+	u16 ip_hdr_err: 1;
+	u16 udp_len_err: 1;
+	u16 raw_cs_invalid: 1;
+	u16 reserved2: 3;
+
+	u16 packet_len: 14;
+	/* Flipped by HW to notify the descriptor is populated. */
+	u16 generation: 1;
+	/* Should be zero. */
+	u16 buffer_queue_id: 1;
+
+	u16 header_len: 10;
+	u16 rsc: 1;
+	u16 split_header: 1;
+	u16 reserved3: 4;
+
+	u8 descriptor_done: 1;
+	u8 end_of_packet: 1;
+	u8 header_buffer_overflow: 1;
+	u8 l3_l4_processed: 1;
+	u8 csum_ip_err: 1;
+	u8 csum_l4_err: 1;
+	u8 csum_external_ip_err: 1;
+	u8 csum_external_udp_err: 1;
+
+	u8 status_error1;
+
+	__le16 reserved5;
+	__le16 buf_id; /* Buffer ID which was sent on the buffer queue. */
+
+	union {
+		/* Packet checksum. */
+		__le16 raw_cs;
+		/* Segment length for RSC packets. */
+		__le16 rsc_seg_len;
+	};
+	__le32 hash;
+	__le32 reserved6;
+	__le32 reserved7;
+	__le32 ts; /* timestamp in nanosecs */
+} __packed;
+
+static_assert(sizeof(struct gve_rx_compl_desc_dqo) == 32);
+
+/* Ringing the doorbell too often can hurt performance.
+ *
+ * HW requires this value to be at least 8.
+ */
+#define GVE_RX_BUF_THRESH_DQO 32
+
+#endif /* _GVE_DESC_DQO_H_ */
diff --git a/drivers/net/ethernet/google/gve/gve_dqo.h b/drivers/net/ethernet/google/gve/gve_dqo.h
new file mode 100644
index 000000000000..6eb442096e02
--- /dev/null
+++ b/drivers/net/ethernet/google/gve/gve_dqo.h
@@ -0,0 +1,115 @@
+/* SPDX-License-Identifier: (GPL-2.0 OR MIT)
+ * Google virtual Ethernet (gve) driver
+ *
+ * Copyright (C) 2015-2021 Google, Inc.
+ */
+
+#ifndef _GVE_DQO_H_
+#define _GVE_DQO_H_
+
+#include "gve_adminq.h"
+
+#define GVE_ITR_ENABLE_BIT_DQO BIT(0)
+#define GVE_ITR_CLEAR_PBA_BIT_DQO BIT(1)
+#define GVE_ITR_NO_UPDATE_DQO (3 << 3)
+
+#define GVE_ITR_INTERVAL_DQO_SHIFT 5
+#define GVE_ITR_INTERVAL_DQO_MASK ((1 << 12) - 1)
+
+#define GVE_TX_IRQ_RATELIMIT_US_DQO 50
+#define GVE_RX_IRQ_RATELIMIT_US_DQO 20
+#define GVE_MAX_ITR_INTERVAL_DQO (GVE_ITR_INTERVAL_DQO_MASK * 2)
+
+/* Timeout in seconds to wait for a reinjection completion after receiving
+ * its corresponding miss completion.
+ */
+#define GVE_REINJECT_COMPL_TIMEOUT 1
+
+/* Timeout in seconds to deallocate the completion tag for a packet that was
+ * prematurely freed for not receiving a valid completion. This should be large
+ * enough to rule out the possibility of receiving the corresponding valid
+ * completion after this interval.
+ */
+#define GVE_DEALLOCATE_COMPL_TIMEOUT 60
+
+netdev_tx_t gve_tx_dqo(struct sk_buff *skb, struct net_device *dev);
+netdev_features_t gve_features_check_dqo(struct sk_buff *skb,
+					 struct net_device *dev,
+					 netdev_features_t features);
+bool gve_tx_poll_dqo(struct gve_notify_block *block, bool do_clean);
+bool gve_xdp_poll_dqo(struct gve_notify_block *block);
+bool gve_xsk_tx_poll_dqo(struct gve_notify_block *block, int budget);
+int gve_rx_poll_dqo(struct gve_notify_block *block, int budget);
+int gve_tx_alloc_rings_dqo(struct gve_priv *priv,
+			   struct gve_tx_alloc_rings_cfg *cfg);
+void gve_tx_free_rings_dqo(struct gve_priv *priv,
+			   struct gve_tx_alloc_rings_cfg *cfg);
+void gve_tx_start_ring_dqo(struct gve_priv *priv, int idx);
+void gve_tx_stop_ring_dqo(struct gve_priv *priv, int idx);
+int gve_rx_alloc_ring_dqo(struct gve_priv *priv,
+			  struct gve_rx_alloc_rings_cfg *cfg,
+			  struct gve_rx_ring *rx,
+			  int idx);
+void gve_rx_free_ring_dqo(struct gve_priv *priv, struct gve_rx_ring *rx,
+			  struct gve_rx_alloc_rings_cfg *cfg);
+int gve_rx_alloc_rings_dqo(struct gve_priv *priv,
+			   struct gve_rx_alloc_rings_cfg *cfg);
+void gve_rx_free_rings_dqo(struct gve_priv *priv,
+			   struct gve_rx_alloc_rings_cfg *cfg);
+void gve_rx_start_ring_dqo(struct gve_priv *priv, int idx);
+void gve_rx_stop_ring_dqo(struct gve_priv *priv, int idx);
+int gve_clean_tx_done_dqo(struct gve_priv *priv, struct gve_tx_ring *tx,
+			  struct napi_struct *napi);
+void gve_rx_post_buffers_dqo(struct gve_rx_ring *rx);
+void gve_rx_write_doorbell_dqo(const struct gve_priv *priv, int queue_idx);
+void gve_xdp_tx_flush_dqo(struct gve_priv *priv, u32 xdp_qid);
+
+static inline void
+gve_tx_put_doorbell_dqo(const struct gve_priv *priv,
+			const struct gve_queue_resources *q_resources, u32 val)
+{
+	u64 index;
+
+	index = be32_to_cpu(q_resources->db_index);
+	iowrite32(val, &priv->db_bar2[index]);
+}
+
+/* Builds register value to write to DQO IRQ doorbell to enable with specified
+ * ITR interval.
+ */
+static inline u32 gve_setup_itr_interval_dqo(u32 interval_us)
+{
+	u32 result = GVE_ITR_ENABLE_BIT_DQO;
+
+	/* Interval has 2us granularity. */
+	interval_us >>= 1;
+
+	interval_us &= GVE_ITR_INTERVAL_DQO_MASK;
+	result |= (interval_us << GVE_ITR_INTERVAL_DQO_SHIFT);
+
+	return result;
+}
+
+static inline void
+gve_write_irq_doorbell_dqo(const struct gve_priv *priv,
+			   const struct gve_notify_block *block, u32 val)
+{
+	u32 index = be32_to_cpu(*block->irq_db_index);
+
+	iowrite32(val, &priv->db_bar2[index]);
+}
+
+/* Sets interrupt throttling interval and enables interrupt
+ * by writing to IRQ doorbell.
+ */
+static inline void
+gve_set_itr_coalesce_usecs_dqo(struct gve_priv *priv,
+			       struct gve_notify_block *block,
+			       u32 usecs)
+{
+	gve_write_irq_doorbell_dqo(priv, block,
+				   gve_setup_itr_interval_dqo(usecs));
+}
+
+int gve_napi_poll_dqo(struct napi_struct *napi, int budget);
+#endif /* _GVE_DQO_H_ */
diff --git a/drivers/net/ethernet/google/gve/gve_ethtool.c b/drivers/net/ethernet/google/gve/gve_ethtool.c
new file mode 100644
index 000000000000..d0a223250845
--- /dev/null
+++ b/drivers/net/ethernet/google/gve/gve_ethtool.c
@@ -0,0 +1,976 @@
+// SPDX-License-Identifier: (GPL-2.0 OR MIT)
+/* Google virtual Ethernet (gve) driver
+ *
+ * Copyright (C) 2015-2024 Google LLC
+ */
+
+#include <linux/rtnetlink.h>
+#include "gve.h"
+#include "gve_adminq.h"
+#include "gve_dqo.h"
+#include "gve_utils.h"
+
+static void gve_get_drvinfo(struct net_device *netdev,
+			    struct ethtool_drvinfo *info)
+{
+	struct gve_priv *priv = netdev_priv(netdev);
+
+	strscpy(info->driver, gve_driver_name, sizeof(info->driver));
+	strscpy(info->version, gve_version_str, sizeof(info->version));
+	strscpy(info->bus_info, pci_name(priv->pdev), sizeof(info->bus_info));
+}
+
+static void gve_set_msglevel(struct net_device *netdev, u32 value)
+{
+	struct gve_priv *priv = netdev_priv(netdev);
+
+	priv->msg_enable = value;
+}
+
+static u32 gve_get_msglevel(struct net_device *netdev)
+{
+	struct gve_priv *priv = netdev_priv(netdev);
+
+	return priv->msg_enable;
+}
+
+/* For the following stats column string names, make sure the order
+ * matches how it is filled in the code. For xdp_aborted, xdp_drop,
+ * xdp_pass, xdp_tx, xdp_redirect, make sure it also matches the order
+ * as declared in enum xdp_action inside file uapi/linux/bpf.h .
+ */
+static const char gve_gstrings_main_stats[][ETH_GSTRING_LEN] = {
+	"rx_packets", "rx_hsplit_pkt", "tx_packets", "rx_bytes",
+	"tx_bytes", "rx_dropped", "tx_dropped", "tx_timeouts",
+	"rx_skb_alloc_fail", "rx_buf_alloc_fail", "rx_desc_err_dropped_pkt",
+	"rx_hsplit_unsplit_pkt",
+	"interface_up_cnt", "interface_down_cnt", "reset_cnt",
+	"page_alloc_fail", "dma_mapping_error", "stats_report_trigger_cnt",
+};
+
+static const char gve_gstrings_rx_stats[][ETH_GSTRING_LEN] = {
+	"rx_posted_desc[%u]", "rx_completed_desc[%u]", "rx_consumed_desc[%u]",
+	"rx_bytes[%u]", "rx_hsplit_bytes[%u]", "rx_cont_packet_cnt[%u]",
+	"rx_frag_flip_cnt[%u]", "rx_frag_copy_cnt[%u]", "rx_frag_alloc_cnt[%u]",
+	"rx_dropped_pkt[%u]", "rx_copybreak_pkt[%u]", "rx_copied_pkt[%u]",
+	"rx_queue_drop_cnt[%u]", "rx_no_buffers_posted[%u]",
+	"rx_drops_packet_over_mru[%u]", "rx_drops_invalid_checksum[%u]",
+	"rx_xdp_aborted[%u]", "rx_xdp_drop[%u]", "rx_xdp_pass[%u]",
+	"rx_xdp_tx[%u]", "rx_xdp_redirect[%u]",
+	"rx_xdp_tx_errors[%u]", "rx_xdp_redirect_errors[%u]", "rx_xdp_alloc_fails[%u]",
+};
+
+static const char gve_gstrings_tx_stats[][ETH_GSTRING_LEN] = {
+	"tx_posted_desc[%u]", "tx_completed_desc[%u]", "tx_consumed_desc[%u]", "tx_bytes[%u]",
+	"tx_wake[%u]", "tx_stop[%u]", "tx_event_counter[%u]",
+	"tx_dma_mapping_error[%u]",
+	"tx_xsk_sent[%u]", "tx_xdp_xmit[%u]", "tx_xdp_xmit_errors[%u]"
+};
+
+static const char gve_gstrings_adminq_stats[][ETH_GSTRING_LEN] __nonstring_array = {
+	"adminq_prod_cnt", "adminq_cmd_fail", "adminq_timeouts",
+	"adminq_describe_device_cnt", "adminq_cfg_device_resources_cnt",
+	"adminq_register_page_list_cnt", "adminq_unregister_page_list_cnt",
+	"adminq_create_tx_queue_cnt", "adminq_create_rx_queue_cnt",
+	"adminq_destroy_tx_queue_cnt", "adminq_destroy_rx_queue_cnt",
+	"adminq_dcfg_device_resources_cnt", "adminq_set_driver_parameter_cnt",
+	"adminq_report_stats_cnt", "adminq_report_link_speed_cnt", "adminq_get_ptype_map_cnt",
+	"adminq_query_flow_rules", "adminq_cfg_flow_rule", "adminq_cfg_rss_cnt",
+	"adminq_query_rss_cnt", "adminq_report_nic_timestamp_cnt",
+};
+
+static const char gve_gstrings_priv_flags[][ETH_GSTRING_LEN] = {
+	"report-stats",
+};
+
+#define GVE_MAIN_STATS_LEN  ARRAY_SIZE(gve_gstrings_main_stats)
+#define GVE_ADMINQ_STATS_LEN  ARRAY_SIZE(gve_gstrings_adminq_stats)
+#define NUM_GVE_TX_CNTS	ARRAY_SIZE(gve_gstrings_tx_stats)
+#define NUM_GVE_RX_CNTS	ARRAY_SIZE(gve_gstrings_rx_stats)
+#define GVE_PRIV_FLAGS_STR_LEN ARRAY_SIZE(gve_gstrings_priv_flags)
+
+static void gve_get_strings(struct net_device *netdev, u32 stringset, u8 *data)
+{
+	struct gve_priv *priv = netdev_priv(netdev);
+	u8 *s = (char *)data;
+	int num_tx_queues;
+	int i, j;
+
+	num_tx_queues = gve_num_tx_queues(priv);
+	switch (stringset) {
+	case ETH_SS_STATS:
+		for (i = 0; i < ARRAY_SIZE(gve_gstrings_main_stats); i++)
+			ethtool_puts(&s, gve_gstrings_main_stats[i]);
+
+		for (i = 0; i < priv->rx_cfg.num_queues; i++)
+			for (j = 0; j < NUM_GVE_RX_CNTS; j++)
+				ethtool_sprintf(&s, gve_gstrings_rx_stats[j],
+						i);
+
+		for (i = 0; i < num_tx_queues; i++)
+			for (j = 0; j < NUM_GVE_TX_CNTS; j++)
+				ethtool_sprintf(&s, gve_gstrings_tx_stats[j],
+						i);
+
+		for (i = 0; i < ARRAY_SIZE(gve_gstrings_adminq_stats); i++)
+			ethtool_cpy(&s, gve_gstrings_adminq_stats[i]);
+
+		break;
+
+	case ETH_SS_PRIV_FLAGS:
+		for (i = 0; i < ARRAY_SIZE(gve_gstrings_priv_flags); i++)
+			ethtool_puts(&s, gve_gstrings_priv_flags[i]);
+		break;
+
+	default:
+		break;
+	}
+}
+
+static int gve_get_sset_count(struct net_device *netdev, int sset)
+{
+	struct gve_priv *priv = netdev_priv(netdev);
+	int num_tx_queues;
+
+	num_tx_queues = gve_num_tx_queues(priv);
+	switch (sset) {
+	case ETH_SS_STATS:
+		return GVE_MAIN_STATS_LEN + GVE_ADMINQ_STATS_LEN +
+		       (priv->rx_cfg.num_queues * NUM_GVE_RX_CNTS) +
+		       (num_tx_queues * NUM_GVE_TX_CNTS);
+	case ETH_SS_PRIV_FLAGS:
+		return GVE_PRIV_FLAGS_STR_LEN;
+	default:
+		return -EOPNOTSUPP;
+	}
+}
+
+static void
+gve_get_ethtool_stats(struct net_device *netdev,
+		      struct ethtool_stats *stats, u64 *data)
+{
+	u64 tmp_rx_pkts, tmp_rx_hsplit_pkt, tmp_rx_bytes, tmp_rx_hsplit_bytes,
+		tmp_rx_skb_alloc_fail, tmp_rx_buf_alloc_fail,
+		tmp_rx_desc_err_dropped_pkt, tmp_rx_hsplit_unsplit_pkt,
+		tmp_tx_pkts, tmp_tx_bytes;
+	u64 rx_buf_alloc_fail, rx_desc_err_dropped_pkt, rx_hsplit_unsplit_pkt,
+		rx_pkts, rx_hsplit_pkt, rx_skb_alloc_fail, rx_bytes, tx_pkts, tx_bytes,
+		tx_dropped;
+	int stats_idx, base_stats_idx, max_stats_idx;
+	struct stats *report_stats;
+	int *rx_qid_to_stats_idx;
+	int *tx_qid_to_stats_idx;
+	int num_stopped_rxqs = 0;
+	int num_stopped_txqs = 0;
+	struct gve_priv *priv;
+	bool skip_nic_stats;
+	unsigned int start;
+	int num_tx_queues;
+	int ring;
+	int i, j;
+
+	ASSERT_RTNL();
+
+	priv = netdev_priv(netdev);
+	num_tx_queues = gve_num_tx_queues(priv);
+	report_stats = priv->stats_report->stats;
+	rx_qid_to_stats_idx = kmalloc_array(priv->rx_cfg.num_queues,
+					    sizeof(int), GFP_KERNEL);
+	if (!rx_qid_to_stats_idx)
+		return;
+	for (ring = 0; ring < priv->rx_cfg.num_queues; ring++) {
+		rx_qid_to_stats_idx[ring] = -1;
+		if (!gve_rx_was_added_to_block(priv, ring))
+			num_stopped_rxqs++;
+	}
+	tx_qid_to_stats_idx = kmalloc_array(num_tx_queues,
+					    sizeof(int), GFP_KERNEL);
+	if (!tx_qid_to_stats_idx) {
+		kfree(rx_qid_to_stats_idx);
+		return;
+	}
+	for (ring = 0; ring < num_tx_queues; ring++) {
+		tx_qid_to_stats_idx[ring] = -1;
+		if (!gve_tx_was_added_to_block(priv, ring))
+			num_stopped_txqs++;
+	}
+
+	for (rx_pkts = 0, rx_bytes = 0, rx_hsplit_pkt = 0,
+	     rx_skb_alloc_fail = 0, rx_buf_alloc_fail = 0,
+	     rx_desc_err_dropped_pkt = 0, rx_hsplit_unsplit_pkt = 0,
+	     ring = 0;
+	     ring < priv->rx_cfg.num_queues; ring++) {
+		if (priv->rx) {
+			do {
+				struct gve_rx_ring *rx = &priv->rx[ring];
+
+				start =
+				  u64_stats_fetch_begin(&priv->rx[ring].statss);
+				tmp_rx_pkts = rx->rpackets;
+				tmp_rx_hsplit_pkt = rx->rx_hsplit_pkt;
+				tmp_rx_bytes = rx->rbytes;
+				tmp_rx_skb_alloc_fail = rx->rx_skb_alloc_fail;
+				tmp_rx_buf_alloc_fail = rx->rx_buf_alloc_fail;
+				tmp_rx_desc_err_dropped_pkt =
+					rx->rx_desc_err_dropped_pkt;
+				tmp_rx_hsplit_unsplit_pkt =
+					rx->rx_hsplit_unsplit_pkt;
+			} while (u64_stats_fetch_retry(&priv->rx[ring].statss,
+						       start));
+			rx_pkts += tmp_rx_pkts;
+			rx_hsplit_pkt += tmp_rx_hsplit_pkt;
+			rx_bytes += tmp_rx_bytes;
+			rx_skb_alloc_fail += tmp_rx_skb_alloc_fail;
+			rx_buf_alloc_fail += tmp_rx_buf_alloc_fail;
+			rx_desc_err_dropped_pkt += tmp_rx_desc_err_dropped_pkt;
+			rx_hsplit_unsplit_pkt += tmp_rx_hsplit_unsplit_pkt;
+		}
+	}
+	for (tx_pkts = 0, tx_bytes = 0, tx_dropped = 0, ring = 0;
+	     ring < num_tx_queues; ring++) {
+		if (priv->tx) {
+			do {
+				start =
+				  u64_stats_fetch_begin(&priv->tx[ring].statss);
+				tmp_tx_pkts = priv->tx[ring].pkt_done;
+				tmp_tx_bytes = priv->tx[ring].bytes_done;
+			} while (u64_stats_fetch_retry(&priv->tx[ring].statss,
+						       start));
+			tx_pkts += tmp_tx_pkts;
+			tx_bytes += tmp_tx_bytes;
+			tx_dropped += priv->tx[ring].dropped_pkt;
+		}
+	}
+
+	i = 0;
+	data[i++] = rx_pkts;
+	data[i++] = rx_hsplit_pkt;
+	data[i++] = tx_pkts;
+	data[i++] = rx_bytes;
+	data[i++] = tx_bytes;
+	/* total rx dropped packets */
+	data[i++] = rx_skb_alloc_fail + rx_buf_alloc_fail +
+		    rx_desc_err_dropped_pkt;
+	data[i++] = tx_dropped;
+	data[i++] = priv->tx_timeo_cnt;
+	data[i++] = rx_skb_alloc_fail;
+	data[i++] = rx_buf_alloc_fail;
+	data[i++] = rx_desc_err_dropped_pkt;
+	data[i++] = rx_hsplit_unsplit_pkt;
+	data[i++] = priv->interface_up_cnt;
+	data[i++] = priv->interface_down_cnt;
+	data[i++] = priv->reset_cnt;
+	data[i++] = priv->page_alloc_fail;
+	data[i++] = priv->dma_mapping_error;
+	data[i++] = priv->stats_report_trigger_cnt;
+	i = GVE_MAIN_STATS_LEN;
+
+	/* For rx cross-reporting stats, start from nic rx stats in report */
+	base_stats_idx = GVE_TX_STATS_REPORT_NUM * num_tx_queues +
+		GVE_RX_STATS_REPORT_NUM * priv->rx_cfg.num_queues;
+	/* The boundary between driver stats and NIC stats shifts if there are
+	 * stopped queues.
+	 */
+	base_stats_idx += NIC_RX_STATS_REPORT_NUM * num_stopped_rxqs +
+		NIC_TX_STATS_REPORT_NUM * num_stopped_txqs;
+	max_stats_idx = NIC_RX_STATS_REPORT_NUM *
+		(priv->rx_cfg.num_queues - num_stopped_rxqs) +
+		base_stats_idx;
+	/* Preprocess the stats report for rx, map queue id to start index */
+	skip_nic_stats = false;
+	for (stats_idx = base_stats_idx; stats_idx < max_stats_idx;
+		stats_idx += NIC_RX_STATS_REPORT_NUM) {
+		u32 stat_name = be32_to_cpu(report_stats[stats_idx].stat_name);
+		u32 queue_id = be32_to_cpu(report_stats[stats_idx].queue_id);
+
+		if (stat_name == 0) {
+			/* no stats written by NIC yet */
+			skip_nic_stats = true;
+			break;
+		}
+		if (queue_id < 0 || queue_id >= priv->rx_cfg.num_queues) {
+			net_err_ratelimited("Invalid rxq id in NIC stats\n");
+			continue;
+		}
+		rx_qid_to_stats_idx[queue_id] = stats_idx;
+	}
+	/* walk RX rings */
+	if (priv->rx) {
+		for (ring = 0; ring < priv->rx_cfg.num_queues; ring++) {
+			struct gve_rx_ring *rx = &priv->rx[ring];
+
+			data[i++] = rx->fill_cnt;
+			data[i++] = rx->cnt;
+			data[i++] = rx->fill_cnt - rx->cnt;
+			do {
+				start =
+				  u64_stats_fetch_begin(&priv->rx[ring].statss);
+				tmp_rx_bytes = rx->rbytes;
+				tmp_rx_hsplit_bytes = rx->rx_hsplit_bytes;
+				tmp_rx_skb_alloc_fail = rx->rx_skb_alloc_fail;
+				tmp_rx_buf_alloc_fail = rx->rx_buf_alloc_fail;
+				tmp_rx_desc_err_dropped_pkt =
+					rx->rx_desc_err_dropped_pkt;
+			} while (u64_stats_fetch_retry(&priv->rx[ring].statss,
+						       start));
+			data[i++] = tmp_rx_bytes;
+			data[i++] = tmp_rx_hsplit_bytes;
+			data[i++] = rx->rx_cont_packet_cnt;
+			data[i++] = rx->rx_frag_flip_cnt;
+			data[i++] = rx->rx_frag_copy_cnt;
+			data[i++] = rx->rx_frag_alloc_cnt;
+			/* rx dropped packets */
+			data[i++] = tmp_rx_skb_alloc_fail +
+				tmp_rx_buf_alloc_fail +
+				tmp_rx_desc_err_dropped_pkt;
+			data[i++] = rx->rx_copybreak_pkt;
+			data[i++] = rx->rx_copied_pkt;
+			/* stats from NIC */
+			stats_idx = rx_qid_to_stats_idx[ring];
+			if (skip_nic_stats || stats_idx < 0) {
+				/* skip NIC rx stats */
+				i += NIC_RX_STATS_REPORT_NUM;
+			} else {
+				for (j = 0; j < NIC_RX_STATS_REPORT_NUM; j++) {
+					u64 value =
+						be64_to_cpu(report_stats[stats_idx + j].value);
+
+					data[i++] = value;
+				}
+			}
+			/* XDP rx counters */
+			do {
+				start =	u64_stats_fetch_begin(&priv->rx[ring].statss);
+				for (j = 0; j < GVE_XDP_ACTIONS; j++)
+					data[i + j] = rx->xdp_actions[j];
+				data[i + j++] = rx->xdp_tx_errors;
+				data[i + j++] = rx->xdp_redirect_errors;
+				data[i + j++] = rx->xdp_alloc_fails;
+			} while (u64_stats_fetch_retry(&priv->rx[ring].statss,
+						       start));
+			i += GVE_XDP_ACTIONS + 3; /* XDP rx counters */
+		}
+	} else {
+		i += priv->rx_cfg.num_queues * NUM_GVE_RX_CNTS;
+	}
+
+	/* For tx cross-reporting stats, start from nic tx stats in report */
+	base_stats_idx = max_stats_idx;
+	max_stats_idx = NIC_TX_STATS_REPORT_NUM *
+		(num_tx_queues - num_stopped_txqs) +
+		max_stats_idx;
+	/* Preprocess the stats report for tx, map queue id to start index */
+	skip_nic_stats = false;
+	for (stats_idx = base_stats_idx; stats_idx < max_stats_idx;
+		stats_idx += NIC_TX_STATS_REPORT_NUM) {
+		u32 stat_name = be32_to_cpu(report_stats[stats_idx].stat_name);
+		u32 queue_id = be32_to_cpu(report_stats[stats_idx].queue_id);
+
+		if (stat_name == 0) {
+			/* no stats written by NIC yet */
+			skip_nic_stats = true;
+			break;
+		}
+		if (queue_id < 0 || queue_id >= num_tx_queues) {
+			net_err_ratelimited("Invalid txq id in NIC stats\n");
+			continue;
+		}
+		tx_qid_to_stats_idx[queue_id] = stats_idx;
+	}
+	/* walk TX rings */
+	if (priv->tx) {
+		for (ring = 0; ring < num_tx_queues; ring++) {
+			struct gve_tx_ring *tx = &priv->tx[ring];
+
+			if (gve_is_gqi(priv)) {
+				data[i++] = tx->req;
+				data[i++] = tx->done;
+				data[i++] = tx->req - tx->done;
+			} else {
+				/* DQO doesn't currently support
+				 * posted/completed descriptor counts;
+				 */
+				data[i++] = 0;
+				data[i++] = 0;
+				data[i++] =
+					(tx->dqo_tx.tail - tx->dqo_tx.head) &
+					tx->mask;
+			}
+			do {
+				start =
+				  u64_stats_fetch_begin(&priv->tx[ring].statss);
+				tmp_tx_bytes = tx->bytes_done;
+			} while (u64_stats_fetch_retry(&priv->tx[ring].statss,
+						       start));
+			data[i++] = tmp_tx_bytes;
+			data[i++] = tx->wake_queue;
+			data[i++] = tx->stop_queue;
+			data[i++] = gve_tx_load_event_counter(priv, tx);
+			data[i++] = tx->dma_mapping_error;
+			/* stats from NIC */
+			stats_idx = tx_qid_to_stats_idx[ring];
+			if (skip_nic_stats || stats_idx < 0) {
+				/* skip NIC tx stats */
+				i += NIC_TX_STATS_REPORT_NUM;
+			} else {
+				for (j = 0; j < NIC_TX_STATS_REPORT_NUM; j++) {
+					u64 value =
+						be64_to_cpu(report_stats[stats_idx + j].value);
+					data[i++] = value;
+				}
+			}
+			/* XDP counters */
+			do {
+				start = u64_stats_fetch_begin(&priv->tx[ring].statss);
+				data[i] = tx->xdp_xsk_sent;
+				data[i + 1] = tx->xdp_xmit;
+				data[i + 2] = tx->xdp_xmit_errors;
+			} while (u64_stats_fetch_retry(&priv->tx[ring].statss,
+						       start));
+			i += 3; /* XDP tx counters */
+		}
+	} else {
+		i += num_tx_queues * NUM_GVE_TX_CNTS;
+	}
+
+	kfree(rx_qid_to_stats_idx);
+	kfree(tx_qid_to_stats_idx);
+	/* AQ Stats */
+	data[i++] = priv->adminq_prod_cnt;
+	data[i++] = priv->adminq_cmd_fail;
+	data[i++] = priv->adminq_timeouts;
+	data[i++] = priv->adminq_describe_device_cnt;
+	data[i++] = priv->adminq_cfg_device_resources_cnt;
+	data[i++] = priv->adminq_register_page_list_cnt;
+	data[i++] = priv->adminq_unregister_page_list_cnt;
+	data[i++] = priv->adminq_create_tx_queue_cnt;
+	data[i++] = priv->adminq_create_rx_queue_cnt;
+	data[i++] = priv->adminq_destroy_tx_queue_cnt;
+	data[i++] = priv->adminq_destroy_rx_queue_cnt;
+	data[i++] = priv->adminq_dcfg_device_resources_cnt;
+	data[i++] = priv->adminq_set_driver_parameter_cnt;
+	data[i++] = priv->adminq_report_stats_cnt;
+	data[i++] = priv->adminq_report_link_speed_cnt;
+	data[i++] = priv->adminq_get_ptype_map_cnt;
+	data[i++] = priv->adminq_query_flow_rules_cnt;
+	data[i++] = priv->adminq_cfg_flow_rule_cnt;
+	data[i++] = priv->adminq_cfg_rss_cnt;
+	data[i++] = priv->adminq_query_rss_cnt;
+	data[i++] = priv->adminq_report_nic_timestamp_cnt;
+}
+
+static void gve_get_channels(struct net_device *netdev,
+			     struct ethtool_channels *cmd)
+{
+	struct gve_priv *priv = netdev_priv(netdev);
+
+	cmd->max_rx = priv->rx_cfg.max_queues;
+	cmd->max_tx = priv->tx_cfg.max_queues;
+	cmd->max_other = 0;
+	cmd->max_combined = 0;
+	cmd->rx_count = priv->rx_cfg.num_queues;
+	cmd->tx_count = priv->tx_cfg.num_queues;
+	cmd->other_count = 0;
+	cmd->combined_count = 0;
+}
+
+static int gve_set_channels(struct net_device *netdev,
+			    struct ethtool_channels *cmd)
+{
+	struct gve_priv *priv = netdev_priv(netdev);
+	struct gve_tx_queue_config new_tx_cfg = priv->tx_cfg;
+	struct gve_rx_queue_config new_rx_cfg = priv->rx_cfg;
+	struct ethtool_channels old_settings;
+	int new_tx = cmd->tx_count;
+	int new_rx = cmd->rx_count;
+	bool reset_rss = false;
+
+	gve_get_channels(netdev, &old_settings);
+
+	/* Changing combined is not allowed */
+	if (cmd->combined_count != old_settings.combined_count)
+		return -EINVAL;
+
+	if (!new_rx || !new_tx)
+		return -EINVAL;
+
+	if (priv->xdp_prog) {
+		if (new_tx != new_rx ||
+		    (2 * new_tx > priv->tx_cfg.max_queues)) {
+			dev_err(&priv->pdev->dev, "The number of configured RX queues should be equal to the number of configured TX queues and the number of configured RX/TX queues should be less than or equal to half the maximum number of RX/TX queues when XDP program is installed");
+			return -EINVAL;
+		}
+
+		/* One XDP TX queue per RX queue. */
+		new_tx_cfg.num_xdp_queues = new_rx;
+	} else {
+		new_tx_cfg.num_xdp_queues = 0;
+	}
+
+	if (new_rx != priv->rx_cfg.num_queues &&
+	    priv->cache_rss_config && !netif_is_rxfh_configured(netdev))
+		reset_rss = true;
+
+	new_tx_cfg.num_queues = new_tx;
+	new_rx_cfg.num_queues = new_rx;
+
+	return gve_adjust_queues(priv, new_rx_cfg, new_tx_cfg, reset_rss);
+}
+
+static void gve_get_ringparam(struct net_device *netdev,
+			      struct ethtool_ringparam *cmd,
+			      struct kernel_ethtool_ringparam *kernel_cmd,
+			      struct netlink_ext_ack *extack)
+{
+	struct gve_priv *priv = netdev_priv(netdev);
+
+	cmd->rx_max_pending = priv->max_rx_desc_cnt;
+	cmd->tx_max_pending = priv->max_tx_desc_cnt;
+	cmd->rx_pending = priv->rx_desc_cnt;
+	cmd->tx_pending = priv->tx_desc_cnt;
+
+	if (!gve_header_split_supported(priv))
+		kernel_cmd->tcp_data_split = ETHTOOL_TCP_DATA_SPLIT_UNKNOWN;
+	else if (priv->header_split_enabled)
+		kernel_cmd->tcp_data_split = ETHTOOL_TCP_DATA_SPLIT_ENABLED;
+	else
+		kernel_cmd->tcp_data_split = ETHTOOL_TCP_DATA_SPLIT_DISABLED;
+}
+
+static int gve_adjust_ring_sizes(struct gve_priv *priv,
+				 u16 new_tx_desc_cnt,
+				 u16 new_rx_desc_cnt)
+{
+	struct gve_tx_alloc_rings_cfg tx_alloc_cfg = {0};
+	struct gve_rx_alloc_rings_cfg rx_alloc_cfg = {0};
+	int err;
+
+	/* get current queue configuration */
+	gve_get_curr_alloc_cfgs(priv, &tx_alloc_cfg, &rx_alloc_cfg);
+
+	/* copy over the new ring_size from ethtool */
+	tx_alloc_cfg.ring_size = new_tx_desc_cnt;
+	rx_alloc_cfg.ring_size = new_rx_desc_cnt;
+
+	if (netif_running(priv->dev)) {
+		err = gve_adjust_config(priv, &tx_alloc_cfg, &rx_alloc_cfg);
+		if (err)
+			return err;
+	}
+
+	/* Set new ring_size for the next up */
+	priv->tx_desc_cnt = new_tx_desc_cnt;
+	priv->rx_desc_cnt = new_rx_desc_cnt;
+
+	return 0;
+}
+
+static int gve_validate_req_ring_size(struct gve_priv *priv, u16 new_tx_desc_cnt,
+				      u16 new_rx_desc_cnt)
+{
+	/* check for valid range */
+	if (new_tx_desc_cnt < priv->min_tx_desc_cnt ||
+	    new_tx_desc_cnt > priv->max_tx_desc_cnt ||
+	    new_rx_desc_cnt < priv->min_rx_desc_cnt ||
+	    new_rx_desc_cnt > priv->max_rx_desc_cnt) {
+		dev_err(&priv->pdev->dev, "Requested descriptor count out of range\n");
+		return -EINVAL;
+	}
+
+	if (!is_power_of_2(new_tx_desc_cnt) || !is_power_of_2(new_rx_desc_cnt)) {
+		dev_err(&priv->pdev->dev, "Requested descriptor count has to be a power of 2\n");
+		return -EINVAL;
+	}
+	return 0;
+}
+
+static int gve_set_ringparam(struct net_device *netdev,
+			     struct ethtool_ringparam *cmd,
+			     struct kernel_ethtool_ringparam *kernel_cmd,
+			     struct netlink_ext_ack *extack)
+{
+	struct gve_priv *priv = netdev_priv(netdev);
+	u16 new_tx_cnt, new_rx_cnt;
+	int err;
+
+	err = gve_set_hsplit_config(priv, kernel_cmd->tcp_data_split);
+	if (err)
+		return err;
+
+	if (cmd->tx_pending == priv->tx_desc_cnt && cmd->rx_pending == priv->rx_desc_cnt)
+		return 0;
+
+	if (!priv->modify_ring_size_enabled) {
+		dev_err(&priv->pdev->dev, "Modify ring size is not supported.\n");
+		return -EOPNOTSUPP;
+	}
+
+	new_tx_cnt = cmd->tx_pending;
+	new_rx_cnt = cmd->rx_pending;
+
+	if (gve_validate_req_ring_size(priv, new_tx_cnt, new_rx_cnt))
+		return -EINVAL;
+
+	return gve_adjust_ring_sizes(priv, new_tx_cnt, new_rx_cnt);
+}
+
+static int gve_user_reset(struct net_device *netdev, u32 *flags)
+{
+	struct gve_priv *priv = netdev_priv(netdev);
+
+	if (*flags == ETH_RESET_ALL) {
+		*flags = 0;
+		return gve_reset(priv, true);
+	}
+
+	return -EOPNOTSUPP;
+}
+
+static int gve_get_tunable(struct net_device *netdev,
+			   const struct ethtool_tunable *etuna, void *value)
+{
+	struct gve_priv *priv = netdev_priv(netdev);
+
+	switch (etuna->id) {
+	case ETHTOOL_RX_COPYBREAK:
+		*(u32 *)value = priv->rx_copybreak;
+		return 0;
+	default:
+		return -EOPNOTSUPP;
+	}
+}
+
+static int gve_set_tunable(struct net_device *netdev,
+			   const struct ethtool_tunable *etuna,
+			   const void *value)
+{
+	struct gve_priv *priv = netdev_priv(netdev);
+	u32 len;
+
+	switch (etuna->id) {
+	case ETHTOOL_RX_COPYBREAK:
+	{
+		u32 max_copybreak = priv->rx_cfg.packet_buffer_size;
+
+		len = *(u32 *)value;
+		if (len > max_copybreak)
+			return -EINVAL;
+		priv->rx_copybreak = len;
+		return 0;
+	}
+	default:
+		return -EOPNOTSUPP;
+	}
+}
+
+static u32 gve_get_priv_flags(struct net_device *netdev)
+{
+	struct gve_priv *priv = netdev_priv(netdev);
+	u32 ret_flags = 0;
+
+	/* Only 1 flag exists currently: report-stats (BIT(0)), so set that flag. */
+	if (priv->ethtool_flags & BIT(0))
+		ret_flags |= BIT(0);
+	return ret_flags;
+}
+
+static int gve_set_priv_flags(struct net_device *netdev, u32 flags)
+{
+	struct gve_priv *priv = netdev_priv(netdev);
+	u64 ori_flags, new_flags;
+	int num_tx_queues;
+
+	num_tx_queues = gve_num_tx_queues(priv);
+	ori_flags = READ_ONCE(priv->ethtool_flags);
+	new_flags = ori_flags;
+
+	/* Only one priv flag exists: report-stats (BIT(0))*/
+	if (flags & BIT(0))
+		new_flags |= BIT(0);
+	else
+		new_flags &= ~(BIT(0));
+	priv->ethtool_flags = new_flags;
+	/* start report-stats timer when user turns report stats on. */
+	if (flags & BIT(0)) {
+		mod_timer(&priv->stats_report_timer,
+			  round_jiffies(jiffies +
+					msecs_to_jiffies(priv->stats_report_timer_period)));
+	}
+	/* Zero off gve stats when report-stats turned off and */
+	/* delete report stats timer. */
+	if (!(flags & BIT(0)) && (ori_flags & BIT(0))) {
+		int tx_stats_num = GVE_TX_STATS_REPORT_NUM *
+			num_tx_queues;
+		int rx_stats_num = GVE_RX_STATS_REPORT_NUM *
+			priv->rx_cfg.num_queues;
+
+		memset(priv->stats_report->stats, 0, (tx_stats_num + rx_stats_num) *
+				   sizeof(struct stats));
+		timer_delete_sync(&priv->stats_report_timer);
+	}
+	return 0;
+}
+
+static int gve_get_link_ksettings(struct net_device *netdev,
+				  struct ethtool_link_ksettings *cmd)
+{
+	struct gve_priv *priv = netdev_priv(netdev);
+	int err = 0;
+
+	if (priv->link_speed == 0)
+		err = gve_adminq_report_link_speed(priv);
+
+	cmd->base.speed = priv->link_speed;
+
+	cmd->base.duplex = DUPLEX_FULL;
+
+	return err;
+}
+
+static int gve_get_coalesce(struct net_device *netdev,
+			    struct ethtool_coalesce *ec,
+			    struct kernel_ethtool_coalesce *kernel_ec,
+			    struct netlink_ext_ack *extack)
+{
+	struct gve_priv *priv = netdev_priv(netdev);
+
+	if (gve_is_gqi(priv))
+		return -EOPNOTSUPP;
+	ec->tx_coalesce_usecs = priv->tx_coalesce_usecs;
+	ec->rx_coalesce_usecs = priv->rx_coalesce_usecs;
+
+	return 0;
+}
+
+static int gve_set_coalesce(struct net_device *netdev,
+			    struct ethtool_coalesce *ec,
+			    struct kernel_ethtool_coalesce *kernel_ec,
+			    struct netlink_ext_ack *extack)
+{
+	struct gve_priv *priv = netdev_priv(netdev);
+	u32 tx_usecs_orig = priv->tx_coalesce_usecs;
+	u32 rx_usecs_orig = priv->rx_coalesce_usecs;
+	int idx;
+
+	if (gve_is_gqi(priv))
+		return -EOPNOTSUPP;
+
+	if (ec->tx_coalesce_usecs > GVE_MAX_ITR_INTERVAL_DQO ||
+	    ec->rx_coalesce_usecs > GVE_MAX_ITR_INTERVAL_DQO)
+		return -EINVAL;
+	priv->tx_coalesce_usecs = ec->tx_coalesce_usecs;
+	priv->rx_coalesce_usecs = ec->rx_coalesce_usecs;
+
+	if (tx_usecs_orig != priv->tx_coalesce_usecs) {
+		for (idx = 0; idx < priv->tx_cfg.num_queues; idx++) {
+			int ntfy_idx = gve_tx_idx_to_ntfy(priv, idx);
+			struct gve_notify_block *block = &priv->ntfy_blocks[ntfy_idx];
+
+			gve_set_itr_coalesce_usecs_dqo(priv, block,
+						       priv->tx_coalesce_usecs);
+		}
+	}
+
+	if (rx_usecs_orig != priv->rx_coalesce_usecs) {
+		for (idx = 0; idx < priv->rx_cfg.num_queues; idx++) {
+			int ntfy_idx = gve_rx_idx_to_ntfy(priv, idx);
+			struct gve_notify_block *block = &priv->ntfy_blocks[ntfy_idx];
+
+			gve_set_itr_coalesce_usecs_dqo(priv, block,
+						       priv->rx_coalesce_usecs);
+		}
+	}
+
+	return 0;
+}
+
+static int gve_set_rxnfc(struct net_device *netdev, struct ethtool_rxnfc *cmd)
+{
+	struct gve_priv *priv = netdev_priv(netdev);
+	int err = 0;
+
+	if (!(netdev->features & NETIF_F_NTUPLE))
+		return -EOPNOTSUPP;
+
+	switch (cmd->cmd) {
+	case ETHTOOL_SRXCLSRLINS:
+		err = gve_add_flow_rule(priv, cmd);
+		break;
+	case ETHTOOL_SRXCLSRLDEL:
+		err = gve_del_flow_rule(priv, cmd);
+		break;
+	default:
+		err = -EOPNOTSUPP;
+		break;
+	}
+
+	return err;
+}
+
+static int gve_get_rxnfc(struct net_device *netdev, struct ethtool_rxnfc *cmd, u32 *rule_locs)
+{
+	struct gve_priv *priv = netdev_priv(netdev);
+	int err = 0;
+
+	switch (cmd->cmd) {
+	case ETHTOOL_GRXRINGS:
+		cmd->data = priv->rx_cfg.num_queues;
+		break;
+	case ETHTOOL_GRXCLSRLCNT:
+		if (!priv->max_flow_rules)
+			return -EOPNOTSUPP;
+
+		err = gve_adminq_query_flow_rules(priv, GVE_FLOW_RULE_QUERY_STATS, 0);
+		if (err)
+			return err;
+
+		cmd->rule_cnt = priv->num_flow_rules;
+		cmd->data = priv->max_flow_rules;
+		break;
+	case ETHTOOL_GRXCLSRULE:
+		err = gve_get_flow_rule_entry(priv, cmd);
+		break;
+	case ETHTOOL_GRXCLSRLALL:
+		err = gve_get_flow_rule_ids(priv, cmd, (u32 *)rule_locs);
+		break;
+	default:
+		err = -EOPNOTSUPP;
+		break;
+	}
+
+	return err;
+}
+
+static u32 gve_get_rxfh_key_size(struct net_device *netdev)
+{
+	struct gve_priv *priv = netdev_priv(netdev);
+
+	return priv->rss_key_size;
+}
+
+static u32 gve_get_rxfh_indir_size(struct net_device *netdev)
+{
+	struct gve_priv *priv = netdev_priv(netdev);
+
+	return priv->rss_lut_size;
+}
+
+static void gve_get_rss_config_cache(struct gve_priv *priv,
+				     struct ethtool_rxfh_param *rxfh)
+{
+	struct gve_rss_config *rss_config = &priv->rss_config;
+
+	rxfh->hfunc = ETH_RSS_HASH_TOP;
+
+	if (rxfh->key) {
+		rxfh->key_size = priv->rss_key_size;
+		memcpy(rxfh->key, rss_config->hash_key, priv->rss_key_size);
+	}
+
+	if (rxfh->indir) {
+		rxfh->indir_size = priv->rss_lut_size;
+		memcpy(rxfh->indir, rss_config->hash_lut,
+		       priv->rss_lut_size * sizeof(*rxfh->indir));
+	}
+}
+
+static int gve_get_rxfh(struct net_device *netdev, struct ethtool_rxfh_param *rxfh)
+{
+	struct gve_priv *priv = netdev_priv(netdev);
+
+	if (!priv->rss_key_size || !priv->rss_lut_size)
+		return -EOPNOTSUPP;
+
+	if (priv->cache_rss_config) {
+		gve_get_rss_config_cache(priv, rxfh);
+		return 0;
+	}
+
+	return gve_adminq_query_rss_config(priv, rxfh);
+}
+
+static void gve_set_rss_config_cache(struct gve_priv *priv,
+				     struct ethtool_rxfh_param *rxfh)
+{
+	struct gve_rss_config *rss_config = &priv->rss_config;
+
+	if (rxfh->key)
+		memcpy(rss_config->hash_key, rxfh->key, priv->rss_key_size);
+
+	if (rxfh->indir)
+		memcpy(rss_config->hash_lut, rxfh->indir,
+		       priv->rss_lut_size * sizeof(*rxfh->indir));
+}
+
+static int gve_set_rxfh(struct net_device *netdev, struct ethtool_rxfh_param *rxfh,
+			struct netlink_ext_ack *extack)
+{
+	struct gve_priv *priv = netdev_priv(netdev);
+	int err;
+
+	if (!priv->rss_key_size || !priv->rss_lut_size)
+		return -EOPNOTSUPP;
+
+	err = gve_adminq_configure_rss(priv, rxfh);
+	if (err) {
+		NL_SET_ERR_MSG_MOD(extack, "Fail to configure RSS config");
+		return err;
+	}
+
+	if (priv->cache_rss_config)
+		gve_set_rss_config_cache(priv, rxfh);
+
+	return 0;
+}
+
+static int gve_get_ts_info(struct net_device *netdev,
+			   struct kernel_ethtool_ts_info *info)
+{
+	struct gve_priv *priv = netdev_priv(netdev);
+
+	ethtool_op_get_ts_info(netdev, info);
+
+	if (priv->nic_timestamp_supported) {
+		info->so_timestamping |= SOF_TIMESTAMPING_RX_HARDWARE |
+					 SOF_TIMESTAMPING_RAW_HARDWARE;
+
+		info->rx_filters |= BIT(HWTSTAMP_FILTER_NONE) |
+				    BIT(HWTSTAMP_FILTER_ALL);
+
+		if (priv->ptp)
+			info->phc_index = ptp_clock_index(priv->ptp->clock);
+	}
+
+	return 0;
+}
+
+const struct ethtool_ops gve_ethtool_ops = {
+	.supported_coalesce_params = ETHTOOL_COALESCE_USECS,
+	.supported_ring_params = ETHTOOL_RING_USE_TCP_DATA_SPLIT,
+	.get_drvinfo = gve_get_drvinfo,
+	.get_strings = gve_get_strings,
+	.get_sset_count = gve_get_sset_count,
+	.get_ethtool_stats = gve_get_ethtool_stats,
+	.set_msglevel = gve_set_msglevel,
+	.get_msglevel = gve_get_msglevel,
+	.set_channels = gve_set_channels,
+	.get_channels = gve_get_channels,
+	.set_rxnfc = gve_set_rxnfc,
+	.get_rxnfc = gve_get_rxnfc,
+	.get_rxfh_indir_size = gve_get_rxfh_indir_size,
+	.get_rxfh_key_size = gve_get_rxfh_key_size,
+	.get_rxfh = gve_get_rxfh,
+	.set_rxfh = gve_set_rxfh,
+	.get_link = ethtool_op_get_link,
+	.get_coalesce = gve_get_coalesce,
+	.set_coalesce = gve_set_coalesce,
+	.get_ringparam = gve_get_ringparam,
+	.set_ringparam = gve_set_ringparam,
+	.reset = gve_user_reset,
+	.get_tunable = gve_get_tunable,
+	.set_tunable = gve_set_tunable,
+	.get_priv_flags = gve_get_priv_flags,
+	.set_priv_flags = gve_set_priv_flags,
+	.get_link_ksettings = gve_get_link_ksettings,
+	.get_ts_info = gve_get_ts_info,
+};
diff --git a/drivers/net/ethernet/google/gve/gve_flow_rule.c b/drivers/net/ethernet/google/gve/gve_flow_rule.c
new file mode 100644
index 000000000000..0bb8cd1876a3
--- /dev/null
+++ b/drivers/net/ethernet/google/gve/gve_flow_rule.c
@@ -0,0 +1,298 @@
+// SPDX-License-Identifier: (GPL-2.0 OR MIT)
+/* Google virtual Ethernet (gve) driver
+ *
+ * Copyright (C) 2015-2024 Google LLC
+ */
+
+#include "gve.h"
+#include "gve_adminq.h"
+
+static
+int gve_fill_ethtool_flow_spec(struct ethtool_rx_flow_spec *fsp,
+			       struct gve_adminq_queried_flow_rule *rule)
+{
+	struct gve_adminq_flow_rule *flow_rule = &rule->flow_rule;
+	static const u16 flow_type_lut[] = {
+		[GVE_FLOW_TYPE_TCPV4]	= TCP_V4_FLOW,
+		[GVE_FLOW_TYPE_UDPV4]	= UDP_V4_FLOW,
+		[GVE_FLOW_TYPE_SCTPV4]	= SCTP_V4_FLOW,
+		[GVE_FLOW_TYPE_AHV4]	= AH_V4_FLOW,
+		[GVE_FLOW_TYPE_ESPV4]	= ESP_V4_FLOW,
+		[GVE_FLOW_TYPE_TCPV6]	= TCP_V6_FLOW,
+		[GVE_FLOW_TYPE_UDPV6]	= UDP_V6_FLOW,
+		[GVE_FLOW_TYPE_SCTPV6]	= SCTP_V6_FLOW,
+		[GVE_FLOW_TYPE_AHV6]	= AH_V6_FLOW,
+		[GVE_FLOW_TYPE_ESPV6]	= ESP_V6_FLOW,
+	};
+
+	if (be16_to_cpu(flow_rule->flow_type) >= ARRAY_SIZE(flow_type_lut))
+		return -EINVAL;
+
+	fsp->flow_type = flow_type_lut[be16_to_cpu(flow_rule->flow_type)];
+
+	memset(&fsp->h_u, 0, sizeof(fsp->h_u));
+	memset(&fsp->h_ext, 0, sizeof(fsp->h_ext));
+	memset(&fsp->m_u, 0, sizeof(fsp->m_u));
+	memset(&fsp->m_ext, 0, sizeof(fsp->m_ext));
+
+	switch (fsp->flow_type) {
+	case TCP_V4_FLOW:
+	case UDP_V4_FLOW:
+	case SCTP_V4_FLOW:
+		fsp->h_u.tcp_ip4_spec.ip4src = flow_rule->key.src_ip[0];
+		fsp->h_u.tcp_ip4_spec.ip4dst = flow_rule->key.dst_ip[0];
+		fsp->h_u.tcp_ip4_spec.psrc = flow_rule->key.src_port;
+		fsp->h_u.tcp_ip4_spec.pdst = flow_rule->key.dst_port;
+		fsp->h_u.tcp_ip4_spec.tos = flow_rule->key.tos;
+		fsp->m_u.tcp_ip4_spec.ip4src = flow_rule->mask.src_ip[0];
+		fsp->m_u.tcp_ip4_spec.ip4dst = flow_rule->mask.dst_ip[0];
+		fsp->m_u.tcp_ip4_spec.psrc = flow_rule->mask.src_port;
+		fsp->m_u.tcp_ip4_spec.pdst = flow_rule->mask.dst_port;
+		fsp->m_u.tcp_ip4_spec.tos = flow_rule->mask.tos;
+		break;
+	case AH_V4_FLOW:
+	case ESP_V4_FLOW:
+		fsp->h_u.ah_ip4_spec.ip4src = flow_rule->key.src_ip[0];
+		fsp->h_u.ah_ip4_spec.ip4dst = flow_rule->key.dst_ip[0];
+		fsp->h_u.ah_ip4_spec.spi = flow_rule->key.spi;
+		fsp->h_u.ah_ip4_spec.tos = flow_rule->key.tos;
+		fsp->m_u.ah_ip4_spec.ip4src = flow_rule->mask.src_ip[0];
+		fsp->m_u.ah_ip4_spec.ip4dst = flow_rule->mask.dst_ip[0];
+		fsp->m_u.ah_ip4_spec.spi = flow_rule->mask.spi;
+		fsp->m_u.ah_ip4_spec.tos = flow_rule->mask.tos;
+		break;
+	case TCP_V6_FLOW:
+	case UDP_V6_FLOW:
+	case SCTP_V6_FLOW:
+		memcpy(fsp->h_u.tcp_ip6_spec.ip6src, &flow_rule->key.src_ip,
+		       sizeof(struct in6_addr));
+		memcpy(fsp->h_u.tcp_ip6_spec.ip6dst, &flow_rule->key.dst_ip,
+		       sizeof(struct in6_addr));
+		fsp->h_u.tcp_ip6_spec.psrc = flow_rule->key.src_port;
+		fsp->h_u.tcp_ip6_spec.pdst = flow_rule->key.dst_port;
+		fsp->h_u.tcp_ip6_spec.tclass = flow_rule->key.tclass;
+		memcpy(fsp->m_u.tcp_ip6_spec.ip6src, &flow_rule->mask.src_ip,
+		       sizeof(struct in6_addr));
+		memcpy(fsp->m_u.tcp_ip6_spec.ip6dst, &flow_rule->mask.dst_ip,
+		       sizeof(struct in6_addr));
+		fsp->m_u.tcp_ip6_spec.psrc = flow_rule->mask.src_port;
+		fsp->m_u.tcp_ip6_spec.pdst = flow_rule->mask.dst_port;
+		fsp->m_u.tcp_ip6_spec.tclass = flow_rule->mask.tclass;
+		break;
+	case AH_V6_FLOW:
+	case ESP_V6_FLOW:
+		memcpy(fsp->h_u.ah_ip6_spec.ip6src, &flow_rule->key.src_ip,
+		       sizeof(struct in6_addr));
+		memcpy(fsp->h_u.ah_ip6_spec.ip6dst, &flow_rule->key.dst_ip,
+		       sizeof(struct in6_addr));
+		fsp->h_u.ah_ip6_spec.spi = flow_rule->key.spi;
+		fsp->h_u.ah_ip6_spec.tclass = flow_rule->key.tclass;
+		memcpy(fsp->m_u.ah_ip6_spec.ip6src, &flow_rule->mask.src_ip,
+		       sizeof(struct in6_addr));
+		memcpy(fsp->m_u.ah_ip6_spec.ip6dst, &flow_rule->mask.dst_ip,
+		       sizeof(struct in6_addr));
+		fsp->m_u.ah_ip6_spec.spi = flow_rule->mask.spi;
+		fsp->m_u.ah_ip6_spec.tclass = flow_rule->mask.tclass;
+		break;
+	default:
+		return -EINVAL;
+	}
+
+	fsp->ring_cookie = be16_to_cpu(flow_rule->action);
+
+	return 0;
+}
+
+static int gve_generate_flow_rule(struct gve_priv *priv, struct ethtool_rx_flow_spec *fsp,
+				  struct gve_adminq_flow_rule *rule)
+{
+	static const u16 flow_type_lut[] = {
+		[TCP_V4_FLOW]	= GVE_FLOW_TYPE_TCPV4,
+		[UDP_V4_FLOW]	= GVE_FLOW_TYPE_UDPV4,
+		[SCTP_V4_FLOW]	= GVE_FLOW_TYPE_SCTPV4,
+		[AH_V4_FLOW]	= GVE_FLOW_TYPE_AHV4,
+		[ESP_V4_FLOW]	= GVE_FLOW_TYPE_ESPV4,
+		[TCP_V6_FLOW]	= GVE_FLOW_TYPE_TCPV6,
+		[UDP_V6_FLOW]	= GVE_FLOW_TYPE_UDPV6,
+		[SCTP_V6_FLOW]	= GVE_FLOW_TYPE_SCTPV6,
+		[AH_V6_FLOW]	= GVE_FLOW_TYPE_AHV6,
+		[ESP_V6_FLOW]	= GVE_FLOW_TYPE_ESPV6,
+	};
+	u32 flow_type;
+
+	if (fsp->ring_cookie == RX_CLS_FLOW_DISC)
+		return -EOPNOTSUPP;
+
+	if (fsp->ring_cookie >= priv->rx_cfg.num_queues)
+		return -EINVAL;
+
+	rule->action = cpu_to_be16(fsp->ring_cookie);
+
+	flow_type = fsp->flow_type & ~(FLOW_EXT | FLOW_MAC_EXT | FLOW_RSS);
+	if (!flow_type || flow_type >= ARRAY_SIZE(flow_type_lut))
+		return -EINVAL;
+
+	rule->flow_type = cpu_to_be16(flow_type_lut[flow_type]);
+
+	switch (flow_type) {
+	case TCP_V4_FLOW:
+	case UDP_V4_FLOW:
+	case SCTP_V4_FLOW:
+		rule->key.src_ip[0] = fsp->h_u.tcp_ip4_spec.ip4src;
+		rule->key.dst_ip[0] = fsp->h_u.tcp_ip4_spec.ip4dst;
+		rule->key.src_port = fsp->h_u.tcp_ip4_spec.psrc;
+		rule->key.dst_port = fsp->h_u.tcp_ip4_spec.pdst;
+		rule->mask.src_ip[0] = fsp->m_u.tcp_ip4_spec.ip4src;
+		rule->mask.dst_ip[0] = fsp->m_u.tcp_ip4_spec.ip4dst;
+		rule->mask.src_port = fsp->m_u.tcp_ip4_spec.psrc;
+		rule->mask.dst_port = fsp->m_u.tcp_ip4_spec.pdst;
+		break;
+	case AH_V4_FLOW:
+	case ESP_V4_FLOW:
+		rule->key.src_ip[0] = fsp->h_u.tcp_ip4_spec.ip4src;
+		rule->key.dst_ip[0] = fsp->h_u.tcp_ip4_spec.ip4dst;
+		rule->key.spi = fsp->h_u.ah_ip4_spec.spi;
+		rule->mask.src_ip[0] = fsp->m_u.tcp_ip4_spec.ip4src;
+		rule->mask.dst_ip[0] = fsp->m_u.tcp_ip4_spec.ip4dst;
+		rule->mask.spi = fsp->m_u.ah_ip4_spec.spi;
+		break;
+	case TCP_V6_FLOW:
+	case UDP_V6_FLOW:
+	case SCTP_V6_FLOW:
+		memcpy(&rule->key.src_ip, fsp->h_u.tcp_ip6_spec.ip6src,
+		       sizeof(struct in6_addr));
+		memcpy(&rule->key.dst_ip, fsp->h_u.tcp_ip6_spec.ip6dst,
+		       sizeof(struct in6_addr));
+		rule->key.src_port = fsp->h_u.tcp_ip6_spec.psrc;
+		rule->key.dst_port = fsp->h_u.tcp_ip6_spec.pdst;
+		memcpy(&rule->mask.src_ip, fsp->m_u.tcp_ip6_spec.ip6src,
+		       sizeof(struct in6_addr));
+		memcpy(&rule->mask.dst_ip, fsp->m_u.tcp_ip6_spec.ip6dst,
+		       sizeof(struct in6_addr));
+		rule->mask.src_port = fsp->m_u.tcp_ip6_spec.psrc;
+		rule->mask.dst_port = fsp->m_u.tcp_ip6_spec.pdst;
+		break;
+	case AH_V6_FLOW:
+	case ESP_V6_FLOW:
+		memcpy(&rule->key.src_ip, fsp->h_u.usr_ip6_spec.ip6src,
+		       sizeof(struct in6_addr));
+		memcpy(&rule->key.dst_ip, fsp->h_u.usr_ip6_spec.ip6dst,
+		       sizeof(struct in6_addr));
+		rule->key.spi = fsp->h_u.ah_ip6_spec.spi;
+		memcpy(&rule->mask.src_ip, fsp->m_u.usr_ip6_spec.ip6src,
+		       sizeof(struct in6_addr));
+		memcpy(&rule->mask.dst_ip, fsp->m_u.usr_ip6_spec.ip6dst,
+		       sizeof(struct in6_addr));
+		rule->key.spi = fsp->h_u.ah_ip6_spec.spi;
+		break;
+	default:
+		/* not doing un-parsed flow types */
+		return -EINVAL;
+	}
+
+	return 0;
+}
+
+int gve_get_flow_rule_entry(struct gve_priv *priv, struct ethtool_rxnfc *cmd)
+{
+	struct gve_adminq_queried_flow_rule *rules_cache = priv->flow_rules_cache.rules_cache;
+	struct ethtool_rx_flow_spec *fsp = (struct ethtool_rx_flow_spec *)&cmd->fs;
+	u32 *cache_num = &priv->flow_rules_cache.rules_cache_num;
+	struct gve_adminq_queried_flow_rule *rule = NULL;
+	int err = 0;
+	u32 i;
+
+	if (!priv->max_flow_rules)
+		return -EOPNOTSUPP;
+
+	if (!priv->flow_rules_cache.rules_cache_synced ||
+	    fsp->location < be32_to_cpu(rules_cache[0].location) ||
+	    fsp->location > be32_to_cpu(rules_cache[*cache_num - 1].location)) {
+		err = gve_adminq_query_flow_rules(priv, GVE_FLOW_RULE_QUERY_RULES, fsp->location);
+		if (err)
+			return err;
+
+		priv->flow_rules_cache.rules_cache_synced = true;
+	}
+
+	for (i = 0; i < *cache_num; i++) {
+		if (fsp->location == be32_to_cpu(rules_cache[i].location)) {
+			rule = &rules_cache[i];
+			break;
+		}
+	}
+
+	if (!rule)
+		return -EINVAL;
+
+	err = gve_fill_ethtool_flow_spec(fsp, rule);
+
+	return err;
+}
+
+int gve_get_flow_rule_ids(struct gve_priv *priv, struct ethtool_rxnfc *cmd, u32 *rule_locs)
+{
+	__be32 *rule_ids_cache = priv->flow_rules_cache.rule_ids_cache;
+	u32 *cache_num = &priv->flow_rules_cache.rule_ids_cache_num;
+	u32 starting_rule_id = 0;
+	u32 i = 0, j = 0;
+	int err = 0;
+
+	if (!priv->max_flow_rules)
+		return -EOPNOTSUPP;
+
+	do {
+		err = gve_adminq_query_flow_rules(priv, GVE_FLOW_RULE_QUERY_IDS,
+						  starting_rule_id);
+		if (err)
+			return err;
+
+		for (i = 0; i < *cache_num; i++) {
+			if (j >= cmd->rule_cnt)
+				return -EMSGSIZE;
+
+			rule_locs[j++] = be32_to_cpu(rule_ids_cache[i]);
+			starting_rule_id = be32_to_cpu(rule_ids_cache[i]) + 1;
+		}
+	} while (*cache_num != 0);
+	cmd->data = priv->max_flow_rules;
+
+	return err;
+}
+
+int gve_add_flow_rule(struct gve_priv *priv, struct ethtool_rxnfc *cmd)
+{
+	struct ethtool_rx_flow_spec *fsp = &cmd->fs;
+	struct gve_adminq_flow_rule *rule = NULL;
+	int err;
+
+	if (!priv->max_flow_rules)
+		return -EOPNOTSUPP;
+
+	rule = kvzalloc(sizeof(*rule), GFP_KERNEL);
+	if (!rule)
+		return -ENOMEM;
+
+	err = gve_generate_flow_rule(priv, fsp, rule);
+	if (err)
+		goto out;
+
+	err = gve_adminq_add_flow_rule(priv, rule, fsp->location);
+
+out:
+	kvfree(rule);
+	if (err)
+		dev_err(&priv->pdev->dev, "Failed to add the flow rule: %u", fsp->location);
+
+	return err;
+}
+
+int gve_del_flow_rule(struct gve_priv *priv, struct ethtool_rxnfc *cmd)
+{
+	struct ethtool_rx_flow_spec *fsp = (struct ethtool_rx_flow_spec *)&cmd->fs;
+
+	if (!priv->max_flow_rules)
+		return -EOPNOTSUPP;
+
+	return gve_adminq_del_flow_rule(priv, fsp->location);
+}
diff --git a/drivers/net/ethernet/google/gve/gve_main.c b/drivers/net/ethernet/google/gve/gve_main.c
new file mode 100644
index 000000000000..1be1b1ef31ee
--- /dev/null
+++ b/drivers/net/ethernet/google/gve/gve_main.c
@@ -0,0 +1,2950 @@
+// SPDX-License-Identifier: (GPL-2.0 OR MIT)
+/* Google virtual Ethernet (gve) driver
+ *
+ * Copyright (C) 2015-2024 Google LLC
+ */
+
+#include <linux/bitmap.h>
+#include <linux/bpf.h>
+#include <linux/cpumask.h>
+#include <linux/etherdevice.h>
+#include <linux/filter.h>
+#include <linux/interrupt.h>
+#include <linux/irq.h>
+#include <linux/module.h>
+#include <linux/pci.h>
+#include <linux/sched.h>
+#include <linux/timer.h>
+#include <linux/workqueue.h>
+#include <linux/utsname.h>
+#include <linux/version.h>
+#include <net/netdev_queues.h>
+#include <net/sch_generic.h>
+#include <net/xdp_sock_drv.h>
+#include "gve.h"
+#include "gve_dqo.h"
+#include "gve_adminq.h"
+#include "gve_register.h"
+#include "gve_utils.h"
+
+#define GVE_DEFAULT_RX_COPYBREAK	(256)
+
+#define DEFAULT_MSG_LEVEL	(NETIF_MSG_DRV | NETIF_MSG_LINK)
+#define GVE_VERSION		"1.0.0"
+#define GVE_VERSION_PREFIX	"GVE-"
+
+// Minimum amount of time between queue kicks in msec (10 seconds)
+#define MIN_TX_TIMEOUT_GAP (1000 * 10)
+
+char gve_driver_name[] = "gve";
+const char gve_version_str[] = GVE_VERSION;
+static const char gve_version_prefix[] = GVE_VERSION_PREFIX;
+
+static int gve_verify_driver_compatibility(struct gve_priv *priv)
+{
+	int err;
+	struct gve_driver_info *driver_info;
+	dma_addr_t driver_info_bus;
+
+	driver_info = dma_alloc_coherent(&priv->pdev->dev,
+					 sizeof(struct gve_driver_info),
+					 &driver_info_bus, GFP_KERNEL);
+	if (!driver_info)
+		return -ENOMEM;
+
+	*driver_info = (struct gve_driver_info) {
+		.os_type = 1, /* Linux */
+		.os_version_major = cpu_to_be32(LINUX_VERSION_MAJOR),
+		.os_version_minor = cpu_to_be32(LINUX_VERSION_SUBLEVEL),
+		.os_version_sub = cpu_to_be32(LINUX_VERSION_PATCHLEVEL),
+		.driver_capability_flags = {
+			cpu_to_be64(GVE_DRIVER_CAPABILITY_FLAGS1),
+			cpu_to_be64(GVE_DRIVER_CAPABILITY_FLAGS2),
+			cpu_to_be64(GVE_DRIVER_CAPABILITY_FLAGS3),
+			cpu_to_be64(GVE_DRIVER_CAPABILITY_FLAGS4),
+		},
+	};
+	strscpy(driver_info->os_version_str1, utsname()->release,
+		sizeof(driver_info->os_version_str1));
+	strscpy(driver_info->os_version_str2, utsname()->version,
+		sizeof(driver_info->os_version_str2));
+
+	err = gve_adminq_verify_driver_compatibility(priv,
+						     sizeof(struct gve_driver_info),
+						     driver_info_bus);
+
+	/* It's ok if the device doesn't support this */
+	if (err == -EOPNOTSUPP)
+		err = 0;
+
+	dma_free_coherent(&priv->pdev->dev,
+			  sizeof(struct gve_driver_info),
+			  driver_info, driver_info_bus);
+	return err;
+}
+
+static netdev_features_t gve_features_check(struct sk_buff *skb,
+					    struct net_device *dev,
+					    netdev_features_t features)
+{
+	struct gve_priv *priv = netdev_priv(dev);
+
+	if (!gve_is_gqi(priv))
+		return gve_features_check_dqo(skb, dev, features);
+
+	return features;
+}
+
+static netdev_tx_t gve_start_xmit(struct sk_buff *skb, struct net_device *dev)
+{
+	struct gve_priv *priv = netdev_priv(dev);
+
+	if (gve_is_gqi(priv))
+		return gve_tx(skb, dev);
+	else
+		return gve_tx_dqo(skb, dev);
+}
+
+static void gve_get_stats(struct net_device *dev, struct rtnl_link_stats64 *s)
+{
+	struct gve_priv *priv = netdev_priv(dev);
+	unsigned int start;
+	u64 packets, bytes;
+	int num_tx_queues;
+	int ring;
+
+	num_tx_queues = gve_num_tx_queues(priv);
+	if (priv->rx) {
+		for (ring = 0; ring < priv->rx_cfg.num_queues; ring++) {
+			do {
+				start =
+				  u64_stats_fetch_begin(&priv->rx[ring].statss);
+				packets = priv->rx[ring].rpackets;
+				bytes = priv->rx[ring].rbytes;
+			} while (u64_stats_fetch_retry(&priv->rx[ring].statss,
+						       start));
+			s->rx_packets += packets;
+			s->rx_bytes += bytes;
+		}
+	}
+	if (priv->tx) {
+		for (ring = 0; ring < num_tx_queues; ring++) {
+			do {
+				start =
+				  u64_stats_fetch_begin(&priv->tx[ring].statss);
+				packets = priv->tx[ring].pkt_done;
+				bytes = priv->tx[ring].bytes_done;
+			} while (u64_stats_fetch_retry(&priv->tx[ring].statss,
+						       start));
+			s->tx_packets += packets;
+			s->tx_bytes += bytes;
+		}
+	}
+}
+
+static int gve_alloc_flow_rule_caches(struct gve_priv *priv)
+{
+	struct gve_flow_rules_cache *flow_rules_cache = &priv->flow_rules_cache;
+	int err = 0;
+
+	if (!priv->max_flow_rules)
+		return 0;
+
+	flow_rules_cache->rules_cache =
+		kvcalloc(GVE_FLOW_RULES_CACHE_SIZE, sizeof(*flow_rules_cache->rules_cache),
+			 GFP_KERNEL);
+	if (!flow_rules_cache->rules_cache) {
+		dev_err(&priv->pdev->dev, "Cannot alloc flow rules cache\n");
+		return -ENOMEM;
+	}
+
+	flow_rules_cache->rule_ids_cache =
+		kvcalloc(GVE_FLOW_RULE_IDS_CACHE_SIZE, sizeof(*flow_rules_cache->rule_ids_cache),
+			 GFP_KERNEL);
+	if (!flow_rules_cache->rule_ids_cache) {
+		dev_err(&priv->pdev->dev, "Cannot alloc flow rule ids cache\n");
+		err = -ENOMEM;
+		goto free_rules_cache;
+	}
+
+	return 0;
+
+free_rules_cache:
+	kvfree(flow_rules_cache->rules_cache);
+	flow_rules_cache->rules_cache = NULL;
+	return err;
+}
+
+static void gve_free_flow_rule_caches(struct gve_priv *priv)
+{
+	struct gve_flow_rules_cache *flow_rules_cache = &priv->flow_rules_cache;
+
+	kvfree(flow_rules_cache->rule_ids_cache);
+	flow_rules_cache->rule_ids_cache = NULL;
+	kvfree(flow_rules_cache->rules_cache);
+	flow_rules_cache->rules_cache = NULL;
+}
+
+static int gve_alloc_rss_config_cache(struct gve_priv *priv)
+{
+	struct gve_rss_config *rss_config = &priv->rss_config;
+
+	if (!priv->cache_rss_config)
+		return 0;
+
+	rss_config->hash_key = kcalloc(priv->rss_key_size,
+				       sizeof(rss_config->hash_key[0]),
+				       GFP_KERNEL);
+	if (!rss_config->hash_key)
+		return -ENOMEM;
+
+	rss_config->hash_lut = kcalloc(priv->rss_lut_size,
+				       sizeof(rss_config->hash_lut[0]),
+				       GFP_KERNEL);
+	if (!rss_config->hash_lut)
+		goto free_rss_key_cache;
+
+	return 0;
+
+free_rss_key_cache:
+	kfree(rss_config->hash_key);
+	rss_config->hash_key = NULL;
+	return -ENOMEM;
+}
+
+static void gve_free_rss_config_cache(struct gve_priv *priv)
+{
+	struct gve_rss_config *rss_config = &priv->rss_config;
+
+	kfree(rss_config->hash_key);
+	kfree(rss_config->hash_lut);
+
+	memset(rss_config, 0, sizeof(*rss_config));
+}
+
+static int gve_alloc_counter_array(struct gve_priv *priv)
+{
+	priv->counter_array =
+		dma_alloc_coherent(&priv->pdev->dev,
+				   priv->num_event_counters *
+				   sizeof(*priv->counter_array),
+				   &priv->counter_array_bus, GFP_KERNEL);
+	if (!priv->counter_array)
+		return -ENOMEM;
+
+	return 0;
+}
+
+static void gve_free_counter_array(struct gve_priv *priv)
+{
+	if (!priv->counter_array)
+		return;
+
+	dma_free_coherent(&priv->pdev->dev,
+			  priv->num_event_counters *
+			  sizeof(*priv->counter_array),
+			  priv->counter_array, priv->counter_array_bus);
+	priv->counter_array = NULL;
+}
+
+/* NIC requests to report stats */
+static void gve_stats_report_task(struct work_struct *work)
+{
+	struct gve_priv *priv = container_of(work, struct gve_priv,
+					     stats_report_task);
+	if (gve_get_do_report_stats(priv)) {
+		gve_handle_report_stats(priv);
+		gve_clear_do_report_stats(priv);
+	}
+}
+
+static void gve_stats_report_schedule(struct gve_priv *priv)
+{
+	if (!gve_get_probe_in_progress(priv) &&
+	    !gve_get_reset_in_progress(priv)) {
+		gve_set_do_report_stats(priv);
+		queue_work(priv->gve_wq, &priv->stats_report_task);
+	}
+}
+
+static void gve_stats_report_timer(struct timer_list *t)
+{
+	struct gve_priv *priv = timer_container_of(priv, t,
+						   stats_report_timer);
+
+	mod_timer(&priv->stats_report_timer,
+		  round_jiffies(jiffies +
+		  msecs_to_jiffies(priv->stats_report_timer_period)));
+	gve_stats_report_schedule(priv);
+}
+
+static int gve_alloc_stats_report(struct gve_priv *priv)
+{
+	int tx_stats_num, rx_stats_num;
+
+	tx_stats_num = (GVE_TX_STATS_REPORT_NUM + NIC_TX_STATS_REPORT_NUM) *
+		       gve_num_tx_queues(priv);
+	rx_stats_num = (GVE_RX_STATS_REPORT_NUM + NIC_RX_STATS_REPORT_NUM) *
+		       priv->rx_cfg.num_queues;
+	priv->stats_report_len = struct_size(priv->stats_report, stats,
+					     size_add(tx_stats_num, rx_stats_num));
+	priv->stats_report =
+		dma_alloc_coherent(&priv->pdev->dev, priv->stats_report_len,
+				   &priv->stats_report_bus, GFP_KERNEL);
+	if (!priv->stats_report)
+		return -ENOMEM;
+	/* Set up timer for the report-stats task */
+	timer_setup(&priv->stats_report_timer, gve_stats_report_timer, 0);
+	priv->stats_report_timer_period = GVE_STATS_REPORT_TIMER_PERIOD;
+	return 0;
+}
+
+static void gve_free_stats_report(struct gve_priv *priv)
+{
+	if (!priv->stats_report)
+		return;
+
+	timer_delete_sync(&priv->stats_report_timer);
+	dma_free_coherent(&priv->pdev->dev, priv->stats_report_len,
+			  priv->stats_report, priv->stats_report_bus);
+	priv->stats_report = NULL;
+}
+
+static irqreturn_t gve_mgmnt_intr(int irq, void *arg)
+{
+	struct gve_priv *priv = arg;
+
+	queue_work(priv->gve_wq, &priv->service_task);
+	return IRQ_HANDLED;
+}
+
+static irqreturn_t gve_intr(int irq, void *arg)
+{
+	struct gve_notify_block *block = arg;
+	struct gve_priv *priv = block->priv;
+
+	iowrite32be(GVE_IRQ_MASK, gve_irq_doorbell(priv, block));
+	napi_schedule_irqoff(&block->napi);
+	return IRQ_HANDLED;
+}
+
+static irqreturn_t gve_intr_dqo(int irq, void *arg)
+{
+	struct gve_notify_block *block = arg;
+
+	/* Interrupts are automatically masked */
+	napi_schedule_irqoff(&block->napi);
+	return IRQ_HANDLED;
+}
+
+static int gve_is_napi_on_home_cpu(struct gve_priv *priv, u32 irq)
+{
+	int cpu_curr = smp_processor_id();
+	const struct cpumask *aff_mask;
+
+	aff_mask = irq_get_effective_affinity_mask(irq);
+	if (unlikely(!aff_mask))
+		return 1;
+
+	return cpumask_test_cpu(cpu_curr, aff_mask);
+}
+
+int gve_napi_poll(struct napi_struct *napi, int budget)
+{
+	struct gve_notify_block *block;
+	__be32 __iomem *irq_doorbell;
+	bool reschedule = false;
+	struct gve_priv *priv;
+	int work_done = 0;
+
+	block = container_of(napi, struct gve_notify_block, napi);
+	priv = block->priv;
+
+	if (block->tx) {
+		if (block->tx->q_num < priv->tx_cfg.num_queues)
+			reschedule |= gve_tx_poll(block, budget);
+		else if (budget)
+			reschedule |= gve_xdp_poll(block, budget);
+	}
+
+	if (!budget)
+		return 0;
+
+	if (block->rx) {
+		work_done = gve_rx_poll(block, budget);
+
+		/* Poll XSK TX as part of RX NAPI. Setup re-poll based on max of
+		 * TX and RX work done.
+		 */
+		if (priv->xdp_prog)
+			work_done = max_t(int, work_done,
+					  gve_xsk_tx_poll(block, budget));
+
+		reschedule |= work_done == budget;
+	}
+
+	if (reschedule)
+		return budget;
+
+       /* Complete processing - don't unmask irq if busy polling is enabled */
+	if (likely(napi_complete_done(napi, work_done))) {
+		irq_doorbell = gve_irq_doorbell(priv, block);
+		iowrite32be(GVE_IRQ_ACK | GVE_IRQ_EVENT, irq_doorbell);
+
+		/* Ensure IRQ ACK is visible before we check pending work.
+		 * If queue had issued updates, it would be truly visible.
+		 */
+		mb();
+
+		if (block->tx)
+			reschedule |= gve_tx_clean_pending(priv, block->tx);
+		if (block->rx)
+			reschedule |= gve_rx_work_pending(block->rx);
+
+		if (reschedule && napi_schedule(napi))
+			iowrite32be(GVE_IRQ_MASK, irq_doorbell);
+	}
+	return work_done;
+}
+
+int gve_napi_poll_dqo(struct napi_struct *napi, int budget)
+{
+	struct gve_notify_block *block =
+		container_of(napi, struct gve_notify_block, napi);
+	struct gve_priv *priv = block->priv;
+	bool reschedule = false;
+	int work_done = 0;
+
+	if (block->tx) {
+		if (block->tx->q_num < priv->tx_cfg.num_queues)
+			reschedule |= gve_tx_poll_dqo(block, /*do_clean=*/true);
+		else
+			reschedule |= gve_xdp_poll_dqo(block);
+	}
+
+	if (!budget)
+		return 0;
+
+	if (block->rx) {
+		work_done = gve_rx_poll_dqo(block, budget);
+
+		/* Poll XSK TX as part of RX NAPI. Setup re-poll based on if
+		 * either datapath has more work to do.
+		 */
+		if (priv->xdp_prog)
+			reschedule |= gve_xsk_tx_poll_dqo(block, budget);
+		reschedule |= work_done == budget;
+	}
+
+	if (reschedule) {
+		/* Reschedule by returning budget only if already on the correct
+		 * cpu.
+		 */
+		if (likely(gve_is_napi_on_home_cpu(priv, block->irq)))
+			return budget;
+
+		/* If not on the cpu with which this queue's irq has affinity
+		 * with, we avoid rescheduling napi and arm the irq instead so
+		 * that napi gets rescheduled back eventually onto the right
+		 * cpu.
+		 */
+		if (work_done == budget)
+			work_done--;
+	}
+
+	if (likely(napi_complete_done(napi, work_done))) {
+		/* Enable interrupts again.
+		 *
+		 * We don't need to repoll afterwards because HW supports the
+		 * PCI MSI-X PBA feature.
+		 *
+		 * Another interrupt would be triggered if a new event came in
+		 * since the last one.
+		 */
+		gve_write_irq_doorbell_dqo(priv, block,
+					   GVE_ITR_NO_UPDATE_DQO | GVE_ITR_ENABLE_BIT_DQO);
+	}
+
+	return work_done;
+}
+
+static const struct cpumask *gve_get_node_mask(struct gve_priv *priv)
+{
+	if (priv->numa_node == NUMA_NO_NODE)
+		return cpu_all_mask;
+	else
+		return cpumask_of_node(priv->numa_node);
+}
+
+static int gve_alloc_notify_blocks(struct gve_priv *priv)
+{
+	int num_vecs_requested = priv->num_ntfy_blks + 1;
+	const struct cpumask *node_mask;
+	unsigned int cur_cpu;
+	int vecs_enabled;
+	int i, j;
+	int err;
+
+	priv->msix_vectors = kvcalloc(num_vecs_requested,
+				      sizeof(*priv->msix_vectors), GFP_KERNEL);
+	if (!priv->msix_vectors)
+		return -ENOMEM;
+	for (i = 0; i < num_vecs_requested; i++)
+		priv->msix_vectors[i].entry = i;
+	vecs_enabled = pci_enable_msix_range(priv->pdev, priv->msix_vectors,
+					     GVE_MIN_MSIX, num_vecs_requested);
+	if (vecs_enabled < 0) {
+		dev_err(&priv->pdev->dev, "Could not enable min msix %d/%d\n",
+			GVE_MIN_MSIX, vecs_enabled);
+		err = vecs_enabled;
+		goto abort_with_msix_vectors;
+	}
+	if (vecs_enabled != num_vecs_requested) {
+		int new_num_ntfy_blks = (vecs_enabled - 1) & ~0x1;
+		int vecs_per_type = new_num_ntfy_blks / 2;
+		int vecs_left = new_num_ntfy_blks % 2;
+
+		priv->num_ntfy_blks = new_num_ntfy_blks;
+		priv->mgmt_msix_idx = priv->num_ntfy_blks;
+		priv->tx_cfg.max_queues = min_t(int, priv->tx_cfg.max_queues,
+						vecs_per_type);
+		priv->rx_cfg.max_queues = min_t(int, priv->rx_cfg.max_queues,
+						vecs_per_type + vecs_left);
+		dev_err(&priv->pdev->dev,
+			"Could not enable desired msix, only enabled %d, adjusting tx max queues to %d, and rx max queues to %d\n",
+			vecs_enabled, priv->tx_cfg.max_queues,
+			priv->rx_cfg.max_queues);
+		if (priv->tx_cfg.num_queues > priv->tx_cfg.max_queues)
+			priv->tx_cfg.num_queues = priv->tx_cfg.max_queues;
+		if (priv->rx_cfg.num_queues > priv->rx_cfg.max_queues)
+			priv->rx_cfg.num_queues = priv->rx_cfg.max_queues;
+	}
+
+	/* Setup Management Vector  - the last vector */
+	snprintf(priv->mgmt_msix_name, sizeof(priv->mgmt_msix_name), "gve-mgmnt@pci:%s",
+		 pci_name(priv->pdev));
+	err = request_irq(priv->msix_vectors[priv->mgmt_msix_idx].vector,
+			  gve_mgmnt_intr, 0, priv->mgmt_msix_name, priv);
+	if (err) {
+		dev_err(&priv->pdev->dev, "Did not receive management vector.\n");
+		goto abort_with_msix_enabled;
+	}
+	priv->irq_db_indices =
+		dma_alloc_coherent(&priv->pdev->dev,
+				   priv->num_ntfy_blks *
+				   sizeof(*priv->irq_db_indices),
+				   &priv->irq_db_indices_bus, GFP_KERNEL);
+	if (!priv->irq_db_indices) {
+		err = -ENOMEM;
+		goto abort_with_mgmt_vector;
+	}
+
+	priv->ntfy_blocks = kvzalloc(priv->num_ntfy_blks *
+				     sizeof(*priv->ntfy_blocks), GFP_KERNEL);
+	if (!priv->ntfy_blocks) {
+		err = -ENOMEM;
+		goto abort_with_irq_db_indices;
+	}
+
+	/* Setup the other blocks - the first n-1 vectors */
+	node_mask = gve_get_node_mask(priv);
+	cur_cpu = cpumask_first(node_mask);
+	for (i = 0; i < priv->num_ntfy_blks; i++) {
+		struct gve_notify_block *block = &priv->ntfy_blocks[i];
+		int msix_idx = i;
+
+		snprintf(block->name, sizeof(block->name), "gve-ntfy-blk%d@pci:%s",
+			 i, pci_name(priv->pdev));
+		block->priv = priv;
+		err = request_irq(priv->msix_vectors[msix_idx].vector,
+				  gve_is_gqi(priv) ? gve_intr : gve_intr_dqo,
+				  0, block->name, block);
+		if (err) {
+			dev_err(&priv->pdev->dev,
+				"Failed to receive msix vector %d\n", i);
+			goto abort_with_some_ntfy_blocks;
+		}
+		block->irq = priv->msix_vectors[msix_idx].vector;
+		irq_set_affinity_and_hint(block->irq,
+					  cpumask_of(cur_cpu));
+		block->irq_db_index = &priv->irq_db_indices[i].index;
+
+		cur_cpu = cpumask_next(cur_cpu, node_mask);
+		/* Wrap once CPUs in the node have been exhausted, or when
+		 * starting RX queue affinities. TX and RX queues of the same
+		 * index share affinity.
+		 */
+		if (cur_cpu >= nr_cpu_ids || (i + 1) == priv->tx_cfg.max_queues)
+			cur_cpu = cpumask_first(node_mask);
+	}
+	return 0;
+abort_with_some_ntfy_blocks:
+	for (j = 0; j < i; j++) {
+		struct gve_notify_block *block = &priv->ntfy_blocks[j];
+		int msix_idx = j;
+
+		irq_set_affinity_hint(priv->msix_vectors[msix_idx].vector,
+				      NULL);
+		free_irq(priv->msix_vectors[msix_idx].vector, block);
+		block->irq = 0;
+	}
+	kvfree(priv->ntfy_blocks);
+	priv->ntfy_blocks = NULL;
+abort_with_irq_db_indices:
+	dma_free_coherent(&priv->pdev->dev, priv->num_ntfy_blks *
+			  sizeof(*priv->irq_db_indices),
+			  priv->irq_db_indices, priv->irq_db_indices_bus);
+	priv->irq_db_indices = NULL;
+abort_with_mgmt_vector:
+	free_irq(priv->msix_vectors[priv->mgmt_msix_idx].vector, priv);
+abort_with_msix_enabled:
+	pci_disable_msix(priv->pdev);
+abort_with_msix_vectors:
+	kvfree(priv->msix_vectors);
+	priv->msix_vectors = NULL;
+	return err;
+}
+
+static void gve_free_notify_blocks(struct gve_priv *priv)
+{
+	int i;
+
+	if (!priv->msix_vectors)
+		return;
+
+	/* Free the irqs */
+	for (i = 0; i < priv->num_ntfy_blks; i++) {
+		struct gve_notify_block *block = &priv->ntfy_blocks[i];
+		int msix_idx = i;
+
+		irq_set_affinity_hint(priv->msix_vectors[msix_idx].vector,
+				      NULL);
+		free_irq(priv->msix_vectors[msix_idx].vector, block);
+		block->irq = 0;
+	}
+	free_irq(priv->msix_vectors[priv->mgmt_msix_idx].vector, priv);
+	kvfree(priv->ntfy_blocks);
+	priv->ntfy_blocks = NULL;
+	dma_free_coherent(&priv->pdev->dev, priv->num_ntfy_blks *
+			  sizeof(*priv->irq_db_indices),
+			  priv->irq_db_indices, priv->irq_db_indices_bus);
+	priv->irq_db_indices = NULL;
+	pci_disable_msix(priv->pdev);
+	kvfree(priv->msix_vectors);
+	priv->msix_vectors = NULL;
+}
+
+static int gve_setup_device_resources(struct gve_priv *priv)
+{
+	int err;
+
+	err = gve_alloc_flow_rule_caches(priv);
+	if (err)
+		return err;
+	err = gve_alloc_rss_config_cache(priv);
+	if (err)
+		goto abort_with_flow_rule_caches;
+	err = gve_alloc_counter_array(priv);
+	if (err)
+		goto abort_with_rss_config_cache;
+	err = gve_init_clock(priv);
+	if (err)
+		goto abort_with_counter;
+	err = gve_alloc_notify_blocks(priv);
+	if (err)
+		goto abort_with_clock;
+	err = gve_alloc_stats_report(priv);
+	if (err)
+		goto abort_with_ntfy_blocks;
+	err = gve_adminq_configure_device_resources(priv,
+						    priv->counter_array_bus,
+						    priv->num_event_counters,
+						    priv->irq_db_indices_bus,
+						    priv->num_ntfy_blks);
+	if (unlikely(err)) {
+		dev_err(&priv->pdev->dev,
+			"could not setup device_resources: err=%d\n", err);
+		err = -ENXIO;
+		goto abort_with_stats_report;
+	}
+
+	if (!gve_is_gqi(priv)) {
+		priv->ptype_lut_dqo = kvzalloc(sizeof(*priv->ptype_lut_dqo),
+					       GFP_KERNEL);
+		if (!priv->ptype_lut_dqo) {
+			err = -ENOMEM;
+			goto abort_with_stats_report;
+		}
+		err = gve_adminq_get_ptype_map_dqo(priv, priv->ptype_lut_dqo);
+		if (err) {
+			dev_err(&priv->pdev->dev,
+				"Failed to get ptype map: err=%d\n", err);
+			goto abort_with_ptype_lut;
+		}
+	}
+
+	err = gve_init_rss_config(priv, priv->rx_cfg.num_queues);
+	if (err) {
+		dev_err(&priv->pdev->dev, "Failed to init RSS config");
+		goto abort_with_ptype_lut;
+	}
+
+	err = gve_adminq_report_stats(priv, priv->stats_report_len,
+				      priv->stats_report_bus,
+				      GVE_STATS_REPORT_TIMER_PERIOD);
+	if (err)
+		dev_err(&priv->pdev->dev,
+			"Failed to report stats: err=%d\n", err);
+	gve_set_device_resources_ok(priv);
+	return 0;
+
+abort_with_ptype_lut:
+	kvfree(priv->ptype_lut_dqo);
+	priv->ptype_lut_dqo = NULL;
+abort_with_stats_report:
+	gve_free_stats_report(priv);
+abort_with_ntfy_blocks:
+	gve_free_notify_blocks(priv);
+abort_with_clock:
+	gve_teardown_clock(priv);
+abort_with_counter:
+	gve_free_counter_array(priv);
+abort_with_rss_config_cache:
+	gve_free_rss_config_cache(priv);
+abort_with_flow_rule_caches:
+	gve_free_flow_rule_caches(priv);
+
+	return err;
+}
+
+static void gve_trigger_reset(struct gve_priv *priv);
+
+static void gve_teardown_device_resources(struct gve_priv *priv)
+{
+	int err;
+
+	/* Tell device its resources are being freed */
+	if (gve_get_device_resources_ok(priv)) {
+		err = gve_flow_rules_reset(priv);
+		if (err) {
+			dev_err(&priv->pdev->dev,
+				"Failed to reset flow rules: err=%d\n", err);
+			gve_trigger_reset(priv);
+		}
+		/* detach the stats report */
+		err = gve_adminq_report_stats(priv, 0, 0x0, GVE_STATS_REPORT_TIMER_PERIOD);
+		if (err) {
+			dev_err(&priv->pdev->dev,
+				"Failed to detach stats report: err=%d\n", err);
+			gve_trigger_reset(priv);
+		}
+		err = gve_adminq_deconfigure_device_resources(priv);
+		if (err) {
+			dev_err(&priv->pdev->dev,
+				"Could not deconfigure device resources: err=%d\n",
+				err);
+			gve_trigger_reset(priv);
+		}
+	}
+
+	kvfree(priv->ptype_lut_dqo);
+	priv->ptype_lut_dqo = NULL;
+
+	gve_free_flow_rule_caches(priv);
+	gve_free_rss_config_cache(priv);
+	gve_free_counter_array(priv);
+	gve_free_notify_blocks(priv);
+	gve_free_stats_report(priv);
+	gve_teardown_clock(priv);
+	gve_clear_device_resources_ok(priv);
+}
+
+static int gve_unregister_qpl(struct gve_priv *priv,
+			      struct gve_queue_page_list *qpl)
+{
+	int err;
+
+	if (!qpl)
+		return 0;
+
+	err = gve_adminq_unregister_page_list(priv, qpl->id);
+	if (err) {
+		netif_err(priv, drv, priv->dev,
+			  "Failed to unregister queue page list %d\n",
+			  qpl->id);
+		return err;
+	}
+
+	priv->num_registered_pages -= qpl->num_entries;
+	return 0;
+}
+
+static int gve_register_qpl(struct gve_priv *priv,
+			    struct gve_queue_page_list *qpl)
+{
+	int pages;
+	int err;
+
+	if (!qpl)
+		return 0;
+
+	pages = qpl->num_entries;
+
+	if (pages + priv->num_registered_pages > priv->max_registered_pages) {
+		netif_err(priv, drv, priv->dev,
+			  "Reached max number of registered pages %llu > %llu\n",
+			  pages + priv->num_registered_pages,
+			  priv->max_registered_pages);
+		return -EINVAL;
+	}
+
+	err = gve_adminq_register_page_list(priv, qpl);
+	if (err) {
+		netif_err(priv, drv, priv->dev,
+			  "failed to register queue page list %d\n",
+			  qpl->id);
+		return err;
+	}
+
+	priv->num_registered_pages += pages;
+	return 0;
+}
+
+static struct gve_queue_page_list *gve_tx_get_qpl(struct gve_priv *priv, int idx)
+{
+	struct gve_tx_ring *tx = &priv->tx[idx];
+
+	if (gve_is_gqi(priv))
+		return tx->tx_fifo.qpl;
+	else
+		return tx->dqo.qpl;
+}
+
+static struct gve_queue_page_list *gve_rx_get_qpl(struct gve_priv *priv, int idx)
+{
+	struct gve_rx_ring *rx = &priv->rx[idx];
+
+	if (gve_is_gqi(priv))
+		return rx->data.qpl;
+	else
+		return rx->dqo.qpl;
+}
+
+static int gve_register_qpls(struct gve_priv *priv)
+{
+	int num_tx_qpls, num_rx_qpls;
+	int err;
+	int i;
+
+	num_tx_qpls = gve_num_tx_qpls(&priv->tx_cfg, gve_is_qpl(priv));
+	num_rx_qpls = gve_num_rx_qpls(&priv->rx_cfg, gve_is_qpl(priv));
+
+	for (i = 0; i < num_tx_qpls; i++) {
+		err = gve_register_qpl(priv, gve_tx_get_qpl(priv, i));
+		if (err)
+			return err;
+	}
+
+	for (i = 0; i < num_rx_qpls; i++) {
+		err = gve_register_qpl(priv, gve_rx_get_qpl(priv, i));
+		if (err)
+			return err;
+	}
+
+	return 0;
+}
+
+static int gve_unregister_qpls(struct gve_priv *priv)
+{
+	int num_tx_qpls, num_rx_qpls;
+	int err;
+	int i;
+
+	num_tx_qpls = gve_num_tx_qpls(&priv->tx_cfg, gve_is_qpl(priv));
+	num_rx_qpls = gve_num_rx_qpls(&priv->rx_cfg, gve_is_qpl(priv));
+
+	for (i = 0; i < num_tx_qpls; i++) {
+		err = gve_unregister_qpl(priv, gve_tx_get_qpl(priv, i));
+		/* This failure will trigger a reset - no need to clean */
+		if (err)
+			return err;
+	}
+
+	for (i = 0; i < num_rx_qpls; i++) {
+		err = gve_unregister_qpl(priv, gve_rx_get_qpl(priv, i));
+		/* This failure will trigger a reset - no need to clean */
+		if (err)
+			return err;
+	}
+	return 0;
+}
+
+static int gve_create_rings(struct gve_priv *priv)
+{
+	int num_tx_queues = gve_num_tx_queues(priv);
+	int err;
+	int i;
+
+	err = gve_adminq_create_tx_queues(priv, 0, num_tx_queues);
+	if (err) {
+		netif_err(priv, drv, priv->dev, "failed to create %d tx queues\n",
+			  num_tx_queues);
+		/* This failure will trigger a reset - no need to clean
+		 * up
+		 */
+		return err;
+	}
+	netif_dbg(priv, drv, priv->dev, "created %d tx queues\n",
+		  num_tx_queues);
+
+	err = gve_adminq_create_rx_queues(priv, priv->rx_cfg.num_queues);
+	if (err) {
+		netif_err(priv, drv, priv->dev, "failed to create %d rx queues\n",
+			  priv->rx_cfg.num_queues);
+		/* This failure will trigger a reset - no need to clean
+		 * up
+		 */
+		return err;
+	}
+	netif_dbg(priv, drv, priv->dev, "created %d rx queues\n",
+		  priv->rx_cfg.num_queues);
+
+	if (gve_is_gqi(priv)) {
+		/* Rx data ring has been prefilled with packet buffers at queue
+		 * allocation time.
+		 *
+		 * Write the doorbell to provide descriptor slots and packet
+		 * buffers to the NIC.
+		 */
+		for (i = 0; i < priv->rx_cfg.num_queues; i++)
+			gve_rx_write_doorbell(priv, &priv->rx[i]);
+	} else {
+		for (i = 0; i < priv->rx_cfg.num_queues; i++) {
+			/* Post buffers and ring doorbell. */
+			gve_rx_post_buffers_dqo(&priv->rx[i]);
+		}
+	}
+
+	return 0;
+}
+
+static void init_xdp_sync_stats(struct gve_priv *priv)
+{
+	int start_id = gve_xdp_tx_start_queue_id(priv);
+	int i;
+
+	/* Init stats */
+	for (i = start_id; i < start_id + priv->tx_cfg.num_xdp_queues; i++) {
+		int ntfy_idx = gve_tx_idx_to_ntfy(priv, i);
+
+		u64_stats_init(&priv->tx[i].statss);
+		priv->tx[i].ntfy_id = ntfy_idx;
+	}
+}
+
+static void gve_init_sync_stats(struct gve_priv *priv)
+{
+	int i;
+
+	for (i = 0; i < priv->tx_cfg.num_queues; i++)
+		u64_stats_init(&priv->tx[i].statss);
+
+	/* Init stats for XDP TX queues */
+	init_xdp_sync_stats(priv);
+
+	for (i = 0; i < priv->rx_cfg.num_queues; i++)
+		u64_stats_init(&priv->rx[i].statss);
+}
+
+static void gve_tx_get_curr_alloc_cfg(struct gve_priv *priv,
+				      struct gve_tx_alloc_rings_cfg *cfg)
+{
+	cfg->qcfg = &priv->tx_cfg;
+	cfg->raw_addressing = !gve_is_qpl(priv);
+	cfg->ring_size = priv->tx_desc_cnt;
+	cfg->num_xdp_rings = cfg->qcfg->num_xdp_queues;
+	cfg->tx = priv->tx;
+}
+
+static void gve_tx_stop_rings(struct gve_priv *priv, int num_rings)
+{
+	int i;
+
+	if (!priv->tx)
+		return;
+
+	for (i = 0; i < num_rings; i++) {
+		if (gve_is_gqi(priv))
+			gve_tx_stop_ring_gqi(priv, i);
+		else
+			gve_tx_stop_ring_dqo(priv, i);
+	}
+}
+
+static void gve_tx_start_rings(struct gve_priv *priv, int num_rings)
+{
+	int i;
+
+	for (i = 0; i < num_rings; i++) {
+		if (gve_is_gqi(priv))
+			gve_tx_start_ring_gqi(priv, i);
+		else
+			gve_tx_start_ring_dqo(priv, i);
+	}
+}
+
+static int gve_queues_mem_alloc(struct gve_priv *priv,
+				struct gve_tx_alloc_rings_cfg *tx_alloc_cfg,
+				struct gve_rx_alloc_rings_cfg *rx_alloc_cfg)
+{
+	int err;
+
+	if (gve_is_gqi(priv))
+		err = gve_tx_alloc_rings_gqi(priv, tx_alloc_cfg);
+	else
+		err = gve_tx_alloc_rings_dqo(priv, tx_alloc_cfg);
+	if (err)
+		return err;
+
+	if (gve_is_gqi(priv))
+		err = gve_rx_alloc_rings_gqi(priv, rx_alloc_cfg);
+	else
+		err = gve_rx_alloc_rings_dqo(priv, rx_alloc_cfg);
+	if (err)
+		goto free_tx;
+
+	return 0;
+
+free_tx:
+	if (gve_is_gqi(priv))
+		gve_tx_free_rings_gqi(priv, tx_alloc_cfg);
+	else
+		gve_tx_free_rings_dqo(priv, tx_alloc_cfg);
+	return err;
+}
+
+static int gve_destroy_rings(struct gve_priv *priv)
+{
+	int num_tx_queues = gve_num_tx_queues(priv);
+	int err;
+
+	err = gve_adminq_destroy_tx_queues(priv, 0, num_tx_queues);
+	if (err) {
+		netif_err(priv, drv, priv->dev,
+			  "failed to destroy tx queues\n");
+		/* This failure will trigger a reset - no need to clean up */
+		return err;
+	}
+	netif_dbg(priv, drv, priv->dev, "destroyed tx queues\n");
+	err = gve_adminq_destroy_rx_queues(priv, priv->rx_cfg.num_queues);
+	if (err) {
+		netif_err(priv, drv, priv->dev,
+			  "failed to destroy rx queues\n");
+		/* This failure will trigger a reset - no need to clean up */
+		return err;
+	}
+	netif_dbg(priv, drv, priv->dev, "destroyed rx queues\n");
+	return 0;
+}
+
+static void gve_queues_mem_free(struct gve_priv *priv,
+				struct gve_tx_alloc_rings_cfg *tx_cfg,
+				struct gve_rx_alloc_rings_cfg *rx_cfg)
+{
+	if (gve_is_gqi(priv)) {
+		gve_tx_free_rings_gqi(priv, tx_cfg);
+		gve_rx_free_rings_gqi(priv, rx_cfg);
+	} else {
+		gve_tx_free_rings_dqo(priv, tx_cfg);
+		gve_rx_free_rings_dqo(priv, rx_cfg);
+	}
+}
+
+int gve_alloc_page(struct gve_priv *priv, struct device *dev,
+		   struct page **page, dma_addr_t *dma,
+		   enum dma_data_direction dir, gfp_t gfp_flags)
+{
+	*page = alloc_pages_node(priv->numa_node, gfp_flags, 0);
+	if (!*page) {
+		priv->page_alloc_fail++;
+		return -ENOMEM;
+	}
+	*dma = dma_map_page(dev, *page, 0, PAGE_SIZE, dir);
+	if (dma_mapping_error(dev, *dma)) {
+		priv->dma_mapping_error++;
+		put_page(*page);
+		return -ENOMEM;
+	}
+	return 0;
+}
+
+struct gve_queue_page_list *gve_alloc_queue_page_list(struct gve_priv *priv,
+						      u32 id, int pages)
+{
+	struct gve_queue_page_list *qpl;
+	int err;
+	int i;
+
+	qpl = kvzalloc(sizeof(*qpl), GFP_KERNEL);
+	if (!qpl)
+		return NULL;
+
+	qpl->id = id;
+	qpl->num_entries = 0;
+	qpl->pages = kvcalloc(pages, sizeof(*qpl->pages), GFP_KERNEL);
+	if (!qpl->pages)
+		goto abort;
+
+	qpl->page_buses = kvcalloc(pages, sizeof(*qpl->page_buses), GFP_KERNEL);
+	if (!qpl->page_buses)
+		goto abort;
+
+	for (i = 0; i < pages; i++) {
+		err = gve_alloc_page(priv, &priv->pdev->dev, &qpl->pages[i],
+				     &qpl->page_buses[i],
+				     gve_qpl_dma_dir(priv, id), GFP_KERNEL);
+		if (err)
+			goto abort;
+		qpl->num_entries++;
+	}
+
+	return qpl;
+
+abort:
+	gve_free_queue_page_list(priv, qpl, id);
+	return NULL;
+}
+
+void gve_free_page(struct device *dev, struct page *page, dma_addr_t dma,
+		   enum dma_data_direction dir)
+{
+	if (!dma_mapping_error(dev, dma))
+		dma_unmap_page(dev, dma, PAGE_SIZE, dir);
+	if (page)
+		put_page(page);
+}
+
+void gve_free_queue_page_list(struct gve_priv *priv,
+			      struct gve_queue_page_list *qpl,
+			      u32 id)
+{
+	int i;
+
+	if (!qpl)
+		return;
+	if (!qpl->pages)
+		goto free_qpl;
+	if (!qpl->page_buses)
+		goto free_pages;
+
+	for (i = 0; i < qpl->num_entries; i++)
+		gve_free_page(&priv->pdev->dev, qpl->pages[i],
+			      qpl->page_buses[i], gve_qpl_dma_dir(priv, id));
+
+	kvfree(qpl->page_buses);
+	qpl->page_buses = NULL;
+free_pages:
+	kvfree(qpl->pages);
+	qpl->pages = NULL;
+free_qpl:
+	kvfree(qpl);
+}
+
+/* Use this to schedule a reset when the device is capable of continuing
+ * to handle other requests in its current state. If it is not, do a reset
+ * in thread instead.
+ */
+void gve_schedule_reset(struct gve_priv *priv)
+{
+	gve_set_do_reset(priv);
+	queue_work(priv->gve_wq, &priv->service_task);
+}
+
+static void gve_reset_and_teardown(struct gve_priv *priv, bool was_up);
+static int gve_reset_recovery(struct gve_priv *priv, bool was_up);
+static void gve_turndown(struct gve_priv *priv);
+static void gve_turnup(struct gve_priv *priv);
+
+static void gve_unreg_xsk_pool(struct gve_priv *priv, u16 qid)
+{
+	struct gve_rx_ring *rx;
+
+	if (!priv->rx)
+		return;
+
+	rx = &priv->rx[qid];
+	rx->xsk_pool = NULL;
+	if (xdp_rxq_info_is_reg(&rx->xdp_rxq))
+		xdp_rxq_info_unreg_mem_model(&rx->xdp_rxq);
+
+	if (!priv->tx)
+		return;
+	priv->tx[gve_xdp_tx_queue_id(priv, qid)].xsk_pool = NULL;
+}
+
+static int gve_reg_xsk_pool(struct gve_priv *priv, struct net_device *dev,
+			    struct xsk_buff_pool *pool, u16 qid)
+{
+	struct gve_rx_ring *rx;
+	u16 tx_qid;
+	int err;
+
+	rx = &priv->rx[qid];
+	err = xdp_rxq_info_reg_mem_model(&rx->xdp_rxq,
+					 MEM_TYPE_XSK_BUFF_POOL, pool);
+	if (err) {
+		gve_unreg_xsk_pool(priv, qid);
+		return err;
+	}
+
+	rx->xsk_pool = pool;
+
+	tx_qid = gve_xdp_tx_queue_id(priv, qid);
+	priv->tx[tx_qid].xsk_pool = pool;
+
+	return 0;
+}
+
+static void gve_unreg_xdp_info(struct gve_priv *priv)
+{
+	int i;
+
+	if (!priv->tx_cfg.num_xdp_queues || !priv->rx)
+		return;
+
+	for (i = 0; i < priv->rx_cfg.num_queues; i++) {
+		struct gve_rx_ring *rx = &priv->rx[i];
+
+		if (xdp_rxq_info_is_reg(&rx->xdp_rxq))
+			xdp_rxq_info_unreg(&rx->xdp_rxq);
+
+		gve_unreg_xsk_pool(priv, i);
+	}
+}
+
+static struct xsk_buff_pool *gve_get_xsk_pool(struct gve_priv *priv, int qid)
+{
+	if (!test_bit(qid, priv->xsk_pools))
+		return NULL;
+
+	return xsk_get_pool_from_qid(priv->dev, qid);
+}
+
+static int gve_reg_xdp_info(struct gve_priv *priv, struct net_device *dev)
+{
+	struct napi_struct *napi;
+	struct gve_rx_ring *rx;
+	int err = 0;
+	int i;
+
+	if (!priv->tx_cfg.num_xdp_queues)
+		return 0;
+
+	for (i = 0; i < priv->rx_cfg.num_queues; i++) {
+		struct xsk_buff_pool *xsk_pool;
+
+		rx = &priv->rx[i];
+		napi = &priv->ntfy_blocks[rx->ntfy_id].napi;
+
+		err = xdp_rxq_info_reg(&rx->xdp_rxq, dev, i,
+				       napi->napi_id);
+		if (err)
+			goto err;
+
+		xsk_pool = gve_get_xsk_pool(priv, i);
+		if (xsk_pool)
+			err = gve_reg_xsk_pool(priv, dev, xsk_pool, i);
+		else if (gve_is_qpl(priv))
+			err = xdp_rxq_info_reg_mem_model(&rx->xdp_rxq,
+							 MEM_TYPE_PAGE_SHARED,
+							 NULL);
+		else
+			err = xdp_rxq_info_reg_mem_model(&rx->xdp_rxq,
+							 MEM_TYPE_PAGE_POOL,
+							 rx->dqo.page_pool);
+		if (err)
+			goto err;
+	}
+	return 0;
+
+err:
+	gve_unreg_xdp_info(priv);
+	return err;
+}
+
+
+static void gve_drain_page_cache(struct gve_priv *priv)
+{
+	int i;
+
+	for (i = 0; i < priv->rx_cfg.num_queues; i++)
+		page_frag_cache_drain(&priv->rx[i].page_cache);
+}
+
+static void gve_rx_get_curr_alloc_cfg(struct gve_priv *priv,
+				      struct gve_rx_alloc_rings_cfg *cfg)
+{
+	cfg->qcfg_rx = &priv->rx_cfg;
+	cfg->qcfg_tx = &priv->tx_cfg;
+	cfg->raw_addressing = !gve_is_qpl(priv);
+	cfg->enable_header_split = priv->header_split_enabled;
+	cfg->ring_size = priv->rx_desc_cnt;
+	cfg->packet_buffer_size = priv->rx_cfg.packet_buffer_size;
+	cfg->rx = priv->rx;
+	cfg->xdp = !!cfg->qcfg_tx->num_xdp_queues;
+}
+
+void gve_get_curr_alloc_cfgs(struct gve_priv *priv,
+			     struct gve_tx_alloc_rings_cfg *tx_alloc_cfg,
+			     struct gve_rx_alloc_rings_cfg *rx_alloc_cfg)
+{
+	gve_tx_get_curr_alloc_cfg(priv, tx_alloc_cfg);
+	gve_rx_get_curr_alloc_cfg(priv, rx_alloc_cfg);
+}
+
+static void gve_rx_start_ring(struct gve_priv *priv, int i)
+{
+	if (gve_is_gqi(priv))
+		gve_rx_start_ring_gqi(priv, i);
+	else
+		gve_rx_start_ring_dqo(priv, i);
+}
+
+static void gve_rx_start_rings(struct gve_priv *priv, int num_rings)
+{
+	int i;
+
+	for (i = 0; i < num_rings; i++)
+		gve_rx_start_ring(priv, i);
+}
+
+static void gve_rx_stop_ring(struct gve_priv *priv, int i)
+{
+	if (gve_is_gqi(priv))
+		gve_rx_stop_ring_gqi(priv, i);
+	else
+		gve_rx_stop_ring_dqo(priv, i);
+}
+
+static void gve_rx_stop_rings(struct gve_priv *priv, int num_rings)
+{
+	int i;
+
+	if (!priv->rx)
+		return;
+
+	for (i = 0; i < num_rings; i++)
+		gve_rx_stop_ring(priv, i);
+}
+
+static void gve_queues_mem_remove(struct gve_priv *priv)
+{
+	struct gve_tx_alloc_rings_cfg tx_alloc_cfg = {0};
+	struct gve_rx_alloc_rings_cfg rx_alloc_cfg = {0};
+
+	gve_get_curr_alloc_cfgs(priv, &tx_alloc_cfg, &rx_alloc_cfg);
+	gve_queues_mem_free(priv, &tx_alloc_cfg, &rx_alloc_cfg);
+	priv->tx = NULL;
+	priv->rx = NULL;
+}
+
+/* The passed-in queue memory is stored into priv and the queues are made live.
+ * No memory is allocated. Passed-in memory is freed on errors.
+ */
+static int gve_queues_start(struct gve_priv *priv,
+			    struct gve_tx_alloc_rings_cfg *tx_alloc_cfg,
+			    struct gve_rx_alloc_rings_cfg *rx_alloc_cfg)
+{
+	struct net_device *dev = priv->dev;
+	int err;
+
+	/* Record new resources into priv */
+	priv->tx = tx_alloc_cfg->tx;
+	priv->rx = rx_alloc_cfg->rx;
+
+	/* Record new configs into priv */
+	priv->tx_cfg = *tx_alloc_cfg->qcfg;
+	priv->tx_cfg.num_xdp_queues = tx_alloc_cfg->num_xdp_rings;
+	priv->rx_cfg = *rx_alloc_cfg->qcfg_rx;
+	priv->tx_desc_cnt = tx_alloc_cfg->ring_size;
+	priv->rx_desc_cnt = rx_alloc_cfg->ring_size;
+
+	gve_tx_start_rings(priv, gve_num_tx_queues(priv));
+	gve_rx_start_rings(priv, rx_alloc_cfg->qcfg_rx->num_queues);
+	gve_init_sync_stats(priv);
+
+	err = netif_set_real_num_tx_queues(dev, priv->tx_cfg.num_queues);
+	if (err)
+		goto stop_and_free_rings;
+	err = netif_set_real_num_rx_queues(dev, priv->rx_cfg.num_queues);
+	if (err)
+		goto stop_and_free_rings;
+
+	err = gve_reg_xdp_info(priv, dev);
+	if (err)
+		goto stop_and_free_rings;
+
+	if (rx_alloc_cfg->reset_rss) {
+		err = gve_init_rss_config(priv, priv->rx_cfg.num_queues);
+		if (err)
+			goto reset;
+	}
+
+	err = gve_register_qpls(priv);
+	if (err)
+		goto reset;
+
+	priv->header_split_enabled = rx_alloc_cfg->enable_header_split;
+	priv->rx_cfg.packet_buffer_size = rx_alloc_cfg->packet_buffer_size;
+
+	err = gve_create_rings(priv);
+	if (err)
+		goto reset;
+
+	gve_set_device_rings_ok(priv);
+
+	if (gve_get_report_stats(priv))
+		mod_timer(&priv->stats_report_timer,
+			  round_jiffies(jiffies +
+				msecs_to_jiffies(priv->stats_report_timer_period)));
+
+	gve_turnup(priv);
+	queue_work(priv->gve_wq, &priv->service_task);
+	priv->interface_up_cnt++;
+	return 0;
+
+reset:
+	if (gve_get_reset_in_progress(priv))
+		goto stop_and_free_rings;
+	gve_reset_and_teardown(priv, true);
+	/* if this fails there is nothing we can do so just ignore the return */
+	gve_reset_recovery(priv, false);
+	/* return the original error */
+	return err;
+stop_and_free_rings:
+	gve_tx_stop_rings(priv, gve_num_tx_queues(priv));
+	gve_rx_stop_rings(priv, priv->rx_cfg.num_queues);
+	gve_queues_mem_remove(priv);
+	return err;
+}
+
+static int gve_open(struct net_device *dev)
+{
+	struct gve_tx_alloc_rings_cfg tx_alloc_cfg = {0};
+	struct gve_rx_alloc_rings_cfg rx_alloc_cfg = {0};
+	struct gve_priv *priv = netdev_priv(dev);
+	int err;
+
+	gve_get_curr_alloc_cfgs(priv, &tx_alloc_cfg, &rx_alloc_cfg);
+
+	err = gve_queues_mem_alloc(priv, &tx_alloc_cfg, &rx_alloc_cfg);
+	if (err)
+		return err;
+
+	/* No need to free on error: ownership of resources is lost after
+	 * calling gve_queues_start.
+	 */
+	err = gve_queues_start(priv, &tx_alloc_cfg, &rx_alloc_cfg);
+	if (err)
+		return err;
+
+	return 0;
+}
+
+static int gve_queues_stop(struct gve_priv *priv)
+{
+	int err;
+
+	netif_carrier_off(priv->dev);
+	if (gve_get_device_rings_ok(priv)) {
+		gve_turndown(priv);
+		gve_drain_page_cache(priv);
+		err = gve_destroy_rings(priv);
+		if (err)
+			goto err;
+		err = gve_unregister_qpls(priv);
+		if (err)
+			goto err;
+		gve_clear_device_rings_ok(priv);
+	}
+	timer_delete_sync(&priv->stats_report_timer);
+
+	gve_unreg_xdp_info(priv);
+
+	gve_tx_stop_rings(priv, gve_num_tx_queues(priv));
+	gve_rx_stop_rings(priv, priv->rx_cfg.num_queues);
+
+	priv->interface_down_cnt++;
+	return 0;
+
+err:
+	/* This must have been called from a reset due to the rtnl lock
+	 * so just return at this point.
+	 */
+	if (gve_get_reset_in_progress(priv))
+		return err;
+	/* Otherwise reset before returning */
+	gve_reset_and_teardown(priv, true);
+	return gve_reset_recovery(priv, false);
+}
+
+static int gve_close(struct net_device *dev)
+{
+	struct gve_priv *priv = netdev_priv(dev);
+	int err;
+
+	err = gve_queues_stop(priv);
+	if (err)
+		return err;
+
+	gve_queues_mem_remove(priv);
+	return 0;
+}
+
+static void gve_handle_link_status(struct gve_priv *priv, bool link_status)
+{
+	if (!gve_get_napi_enabled(priv))
+		return;
+
+	if (link_status == netif_carrier_ok(priv->dev))
+		return;
+
+	if (link_status) {
+		netdev_info(priv->dev, "Device link is up.\n");
+		netif_carrier_on(priv->dev);
+	} else {
+		netdev_info(priv->dev, "Device link is down.\n");
+		netif_carrier_off(priv->dev);
+	}
+}
+
+static int gve_configure_rings_xdp(struct gve_priv *priv,
+				   u16 num_xdp_rings)
+{
+	struct gve_tx_alloc_rings_cfg tx_alloc_cfg = {0};
+	struct gve_rx_alloc_rings_cfg rx_alloc_cfg = {0};
+
+	gve_get_curr_alloc_cfgs(priv, &tx_alloc_cfg, &rx_alloc_cfg);
+	tx_alloc_cfg.num_xdp_rings = num_xdp_rings;
+
+	rx_alloc_cfg.xdp = !!num_xdp_rings;
+	return gve_adjust_config(priv, &tx_alloc_cfg, &rx_alloc_cfg);
+}
+
+static int gve_set_xdp(struct gve_priv *priv, struct bpf_prog *prog,
+		       struct netlink_ext_ack *extack)
+{
+	struct bpf_prog *old_prog;
+	int err = 0;
+	u32 status;
+
+	old_prog = READ_ONCE(priv->xdp_prog);
+	if (!netif_running(priv->dev)) {
+		WRITE_ONCE(priv->xdp_prog, prog);
+		if (old_prog)
+			bpf_prog_put(old_prog);
+
+		/* Update priv XDP queue configuration */
+		priv->tx_cfg.num_xdp_queues = priv->xdp_prog ?
+			priv->rx_cfg.num_queues : 0;
+		return 0;
+	}
+
+	if (!old_prog && prog)
+		err = gve_configure_rings_xdp(priv, priv->rx_cfg.num_queues);
+	else if (old_prog && !prog)
+		err = gve_configure_rings_xdp(priv, 0);
+
+	if (err)
+		goto out;
+
+	WRITE_ONCE(priv->xdp_prog, prog);
+	if (old_prog)
+		bpf_prog_put(old_prog);
+
+out:
+	status = ioread32be(&priv->reg_bar0->device_status);
+	gve_handle_link_status(priv, GVE_DEVICE_STATUS_LINK_STATUS_MASK & status);
+	return err;
+}
+
+static int gve_xdp_xmit(struct net_device *dev, int n,
+			struct xdp_frame **frames, u32 flags)
+{
+	struct gve_priv *priv = netdev_priv(dev);
+
+	if (priv->queue_format == GVE_GQI_QPL_FORMAT)
+		return gve_xdp_xmit_gqi(dev, n, frames, flags);
+	else if (priv->queue_format == GVE_DQO_RDA_FORMAT)
+		return gve_xdp_xmit_dqo(dev, n, frames, flags);
+
+	return -EOPNOTSUPP;
+}
+
+static int gve_xsk_pool_enable(struct net_device *dev,
+			       struct xsk_buff_pool *pool,
+			       u16 qid)
+{
+	struct gve_priv *priv = netdev_priv(dev);
+	int err;
+
+	if (qid >= priv->rx_cfg.num_queues) {
+		dev_err(&priv->pdev->dev, "xsk pool invalid qid %d", qid);
+		return -EINVAL;
+	}
+	if (xsk_pool_get_rx_frame_size(pool) <
+	     priv->dev->max_mtu + sizeof(struct ethhdr)) {
+		dev_err(&priv->pdev->dev, "xsk pool frame_len too small");
+		return -EINVAL;
+	}
+
+	err = xsk_pool_dma_map(pool, &priv->pdev->dev,
+			       DMA_ATTR_SKIP_CPU_SYNC | DMA_ATTR_WEAK_ORDERING);
+	if (err)
+		return err;
+
+	set_bit(qid, priv->xsk_pools);
+
+	/* If XDP prog is not installed or interface is down, return. */
+	if (!priv->xdp_prog || !netif_running(dev))
+		return 0;
+
+	err = gve_reg_xsk_pool(priv, dev, pool, qid);
+	if (err)
+		goto err_xsk_pool_dma_mapped;
+
+	/* Stop and start RDA queues to repost buffers. */
+	if (!gve_is_qpl(priv)) {
+		err = gve_configure_rings_xdp(priv, priv->rx_cfg.num_queues);
+		if (err)
+			goto err_xsk_pool_registered;
+	}
+	return 0;
+
+err_xsk_pool_registered:
+	gve_unreg_xsk_pool(priv, qid);
+err_xsk_pool_dma_mapped:
+	clear_bit(qid, priv->xsk_pools);
+	xsk_pool_dma_unmap(pool,
+			   DMA_ATTR_SKIP_CPU_SYNC |
+			   DMA_ATTR_WEAK_ORDERING);
+	return err;
+}
+
+static int gve_xsk_pool_disable(struct net_device *dev,
+				u16 qid)
+{
+	struct gve_priv *priv = netdev_priv(dev);
+	struct napi_struct *napi_rx;
+	struct napi_struct *napi_tx;
+	struct xsk_buff_pool *pool;
+	int tx_qid;
+	int err;
+
+	if (qid >= priv->rx_cfg.num_queues)
+		return -EINVAL;
+
+	clear_bit(qid, priv->xsk_pools);
+
+	pool = xsk_get_pool_from_qid(dev, qid);
+	if (pool)
+		xsk_pool_dma_unmap(pool,
+				   DMA_ATTR_SKIP_CPU_SYNC |
+				   DMA_ATTR_WEAK_ORDERING);
+
+	if (!netif_running(dev) || !priv->tx_cfg.num_xdp_queues)
+		return 0;
+
+	/* Stop and start RDA queues to repost buffers. */
+	if (!gve_is_qpl(priv) && priv->xdp_prog) {
+		err = gve_configure_rings_xdp(priv, priv->rx_cfg.num_queues);
+		if (err)
+			return err;
+	}
+
+	napi_rx = &priv->ntfy_blocks[priv->rx[qid].ntfy_id].napi;
+	napi_disable(napi_rx); /* make sure current rx poll is done */
+
+	tx_qid = gve_xdp_tx_queue_id(priv, qid);
+	napi_tx = &priv->ntfy_blocks[priv->tx[tx_qid].ntfy_id].napi;
+	napi_disable(napi_tx); /* make sure current tx poll is done */
+
+	gve_unreg_xsk_pool(priv, qid);
+	smp_mb(); /* Make sure it is visible to the workers on datapath */
+
+	napi_enable(napi_rx);
+	napi_enable(napi_tx);
+	if (gve_is_gqi(priv)) {
+		if (gve_rx_work_pending(&priv->rx[qid]))
+			napi_schedule(napi_rx);
+
+		if (gve_tx_clean_pending(priv, &priv->tx[tx_qid]))
+			napi_schedule(napi_tx);
+	}
+
+	return 0;
+}
+
+static int gve_xsk_wakeup(struct net_device *dev, u32 queue_id, u32 flags)
+{
+	struct gve_priv *priv = netdev_priv(dev);
+	struct napi_struct *napi;
+
+	if (!gve_get_napi_enabled(priv))
+		return -ENETDOWN;
+
+	if (queue_id >= priv->rx_cfg.num_queues || !priv->xdp_prog)
+		return -EINVAL;
+
+	napi = &priv->ntfy_blocks[gve_rx_idx_to_ntfy(priv, queue_id)].napi;
+	if (!napi_if_scheduled_mark_missed(napi)) {
+		/* Call local_bh_enable to trigger SoftIRQ processing */
+		local_bh_disable();
+		napi_schedule(napi);
+		local_bh_enable();
+	}
+
+	return 0;
+}
+
+static int verify_xdp_configuration(struct net_device *dev)
+{
+	struct gve_priv *priv = netdev_priv(dev);
+	u16 max_xdp_mtu;
+
+	if (dev->features & NETIF_F_LRO) {
+		netdev_warn(dev, "XDP is not supported when LRO is on.\n");
+		return -EOPNOTSUPP;
+	}
+
+	if (priv->header_split_enabled) {
+		netdev_warn(dev, "XDP is not supported when header-data split is enabled.\n");
+		return -EOPNOTSUPP;
+	}
+
+	max_xdp_mtu = priv->rx_cfg.packet_buffer_size - sizeof(struct ethhdr);
+	if (priv->queue_format == GVE_GQI_QPL_FORMAT)
+		max_xdp_mtu -= GVE_RX_PAD;
+
+	if (dev->mtu > max_xdp_mtu) {
+		netdev_warn(dev, "XDP is not supported for mtu %d.\n",
+			    dev->mtu);
+		return -EOPNOTSUPP;
+	}
+
+	if (priv->rx_cfg.num_queues != priv->tx_cfg.num_queues ||
+	    (2 * priv->tx_cfg.num_queues > priv->tx_cfg.max_queues)) {
+		netdev_warn(dev, "XDP load failed: The number of configured RX queues %d should be equal to the number of configured TX queues %d and the number of configured RX/TX queues should be less than or equal to half the maximum number of RX/TX queues %d",
+			    priv->rx_cfg.num_queues,
+			    priv->tx_cfg.num_queues,
+			    priv->tx_cfg.max_queues);
+		return -EINVAL;
+	}
+	return 0;
+}
+
+static int gve_xdp(struct net_device *dev, struct netdev_bpf *xdp)
+{
+	struct gve_priv *priv = netdev_priv(dev);
+	int err;
+
+	err = verify_xdp_configuration(dev);
+	if (err)
+		return err;
+	switch (xdp->command) {
+	case XDP_SETUP_PROG:
+		return gve_set_xdp(priv, xdp->prog, xdp->extack);
+	case XDP_SETUP_XSK_POOL:
+		if (xdp->xsk.pool)
+			return gve_xsk_pool_enable(dev, xdp->xsk.pool, xdp->xsk.queue_id);
+		else
+			return gve_xsk_pool_disable(dev, xdp->xsk.queue_id);
+	default:
+		return -EINVAL;
+	}
+}
+
+int gve_init_rss_config(struct gve_priv *priv, u16 num_queues)
+{
+	struct gve_rss_config *rss_config = &priv->rss_config;
+	struct ethtool_rxfh_param rxfh = {0};
+	u16 i;
+
+	if (!priv->cache_rss_config)
+		return 0;
+
+	for (i = 0; i < priv->rss_lut_size; i++)
+		rss_config->hash_lut[i] =
+			ethtool_rxfh_indir_default(i, num_queues);
+
+	netdev_rss_key_fill(rss_config->hash_key, priv->rss_key_size);
+
+	rxfh.hfunc = ETH_RSS_HASH_TOP;
+
+	return gve_adminq_configure_rss(priv, &rxfh);
+}
+
+int gve_flow_rules_reset(struct gve_priv *priv)
+{
+	if (!priv->max_flow_rules)
+		return 0;
+
+	return gve_adminq_reset_flow_rules(priv);
+}
+
+int gve_adjust_config(struct gve_priv *priv,
+		      struct gve_tx_alloc_rings_cfg *tx_alloc_cfg,
+		      struct gve_rx_alloc_rings_cfg *rx_alloc_cfg)
+{
+	int err;
+
+	/* Allocate resources for the new configuration */
+	err = gve_queues_mem_alloc(priv, tx_alloc_cfg, rx_alloc_cfg);
+	if (err) {
+		netif_err(priv, drv, priv->dev,
+			  "Adjust config failed to alloc new queues");
+		return err;
+	}
+
+	/* Teardown the device and free existing resources */
+	err = gve_close(priv->dev);
+	if (err) {
+		netif_err(priv, drv, priv->dev,
+			  "Adjust config failed to close old queues");
+		gve_queues_mem_free(priv, tx_alloc_cfg, rx_alloc_cfg);
+		return err;
+	}
+
+	/* Bring the device back up again with the new resources. */
+	err = gve_queues_start(priv, tx_alloc_cfg, rx_alloc_cfg);
+	if (err) {
+		netif_err(priv, drv, priv->dev,
+			  "Adjust config failed to start new queues, !!! DISABLING ALL QUEUES !!!\n");
+		/* No need to free on error: ownership of resources is lost after
+		 * calling gve_queues_start.
+		 */
+		gve_turndown(priv);
+		return err;
+	}
+
+	return 0;
+}
+
+int gve_adjust_queues(struct gve_priv *priv,
+		      struct gve_rx_queue_config new_rx_config,
+		      struct gve_tx_queue_config new_tx_config,
+		      bool reset_rss)
+{
+	struct gve_tx_alloc_rings_cfg tx_alloc_cfg = {0};
+	struct gve_rx_alloc_rings_cfg rx_alloc_cfg = {0};
+	int err;
+
+	gve_get_curr_alloc_cfgs(priv, &tx_alloc_cfg, &rx_alloc_cfg);
+
+	/* Relay the new config from ethtool */
+	tx_alloc_cfg.qcfg = &new_tx_config;
+	rx_alloc_cfg.qcfg_tx = &new_tx_config;
+	rx_alloc_cfg.qcfg_rx = &new_rx_config;
+	rx_alloc_cfg.reset_rss = reset_rss;
+
+	if (netif_running(priv->dev)) {
+		err = gve_adjust_config(priv, &tx_alloc_cfg, &rx_alloc_cfg);
+		return err;
+	}
+	/* Set the config for the next up. */
+	if (reset_rss) {
+		err = gve_init_rss_config(priv, new_rx_config.num_queues);
+		if (err)
+			return err;
+	}
+	priv->tx_cfg = new_tx_config;
+	priv->rx_cfg = new_rx_config;
+
+	return 0;
+}
+
+static void gve_turndown(struct gve_priv *priv)
+{
+	int idx;
+
+	if (netif_carrier_ok(priv->dev))
+		netif_carrier_off(priv->dev);
+
+	if (!gve_get_napi_enabled(priv))
+		return;
+
+	/* Disable napi to prevent more work from coming in */
+	for (idx = 0; idx < gve_num_tx_queues(priv); idx++) {
+		int ntfy_idx = gve_tx_idx_to_ntfy(priv, idx);
+		struct gve_notify_block *block = &priv->ntfy_blocks[ntfy_idx];
+
+		if (!gve_tx_was_added_to_block(priv, idx))
+			continue;
+
+		if (idx < priv->tx_cfg.num_queues)
+			netif_queue_set_napi(priv->dev, idx,
+					     NETDEV_QUEUE_TYPE_TX, NULL);
+
+		napi_disable_locked(&block->napi);
+	}
+	for (idx = 0; idx < priv->rx_cfg.num_queues; idx++) {
+		int ntfy_idx = gve_rx_idx_to_ntfy(priv, idx);
+		struct gve_notify_block *block = &priv->ntfy_blocks[ntfy_idx];
+
+		if (!gve_rx_was_added_to_block(priv, idx))
+			continue;
+
+		netif_queue_set_napi(priv->dev, idx, NETDEV_QUEUE_TYPE_RX,
+				     NULL);
+		napi_disable_locked(&block->napi);
+	}
+
+	/* Stop tx queues */
+	netif_tx_disable(priv->dev);
+
+	xdp_features_clear_redirect_target_locked(priv->dev);
+
+	gve_clear_napi_enabled(priv);
+	gve_clear_report_stats(priv);
+
+	/* Make sure that all traffic is finished processing. */
+	synchronize_net();
+}
+
+static void gve_turnup(struct gve_priv *priv)
+{
+	int idx;
+
+	/* Start the tx queues */
+	netif_tx_start_all_queues(priv->dev);
+
+	/* Enable napi and unmask interrupts for all queues */
+	for (idx = 0; idx < gve_num_tx_queues(priv); idx++) {
+		int ntfy_idx = gve_tx_idx_to_ntfy(priv, idx);
+		struct gve_notify_block *block = &priv->ntfy_blocks[ntfy_idx];
+
+		if (!gve_tx_was_added_to_block(priv, idx))
+			continue;
+
+		napi_enable_locked(&block->napi);
+
+		if (idx < priv->tx_cfg.num_queues)
+			netif_queue_set_napi(priv->dev, idx,
+					     NETDEV_QUEUE_TYPE_TX,
+					     &block->napi);
+
+		if (gve_is_gqi(priv)) {
+			iowrite32be(0, gve_irq_doorbell(priv, block));
+		} else {
+			gve_set_itr_coalesce_usecs_dqo(priv, block,
+						       priv->tx_coalesce_usecs);
+		}
+
+		/* Any descs written by the NIC before this barrier will be
+		 * handled by the one-off napi schedule below. Whereas any
+		 * descs after the barrier will generate interrupts.
+		 */
+		mb();
+		napi_schedule(&block->napi);
+	}
+	for (idx = 0; idx < priv->rx_cfg.num_queues; idx++) {
+		int ntfy_idx = gve_rx_idx_to_ntfy(priv, idx);
+		struct gve_notify_block *block = &priv->ntfy_blocks[ntfy_idx];
+
+		if (!gve_rx_was_added_to_block(priv, idx))
+			continue;
+
+		napi_enable_locked(&block->napi);
+		netif_queue_set_napi(priv->dev, idx, NETDEV_QUEUE_TYPE_RX,
+				     &block->napi);
+
+		if (gve_is_gqi(priv)) {
+			iowrite32be(0, gve_irq_doorbell(priv, block));
+		} else {
+			gve_set_itr_coalesce_usecs_dqo(priv, block,
+						       priv->rx_coalesce_usecs);
+		}
+
+		/* Any descs written by the NIC before this barrier will be
+		 * handled by the one-off napi schedule below. Whereas any
+		 * descs after the barrier will generate interrupts.
+		 */
+		mb();
+		napi_schedule(&block->napi);
+	}
+
+	if (priv->tx_cfg.num_xdp_queues && gve_supports_xdp_xmit(priv))
+		xdp_features_set_redirect_target_locked(priv->dev, false);
+
+	gve_set_napi_enabled(priv);
+}
+
+static void gve_turnup_and_check_status(struct gve_priv *priv)
+{
+	u32 status;
+
+	gve_turnup(priv);
+	status = ioread32be(&priv->reg_bar0->device_status);
+	gve_handle_link_status(priv, GVE_DEVICE_STATUS_LINK_STATUS_MASK & status);
+}
+
+static struct gve_notify_block *gve_get_tx_notify_block(struct gve_priv *priv,
+							unsigned int txqueue)
+{
+	u32 ntfy_idx;
+
+	if (txqueue > priv->tx_cfg.num_queues)
+		return NULL;
+
+	ntfy_idx = gve_tx_idx_to_ntfy(priv, txqueue);
+	if (ntfy_idx >= priv->num_ntfy_blks)
+		return NULL;
+
+	return &priv->ntfy_blocks[ntfy_idx];
+}
+
+static bool gve_tx_timeout_try_q_kick(struct gve_priv *priv,
+				      unsigned int txqueue)
+{
+	struct gve_notify_block *block;
+	u32 current_time;
+
+	block = gve_get_tx_notify_block(priv, txqueue);
+
+	if (!block)
+		return false;
+
+	current_time = jiffies_to_msecs(jiffies);
+	if (block->tx->last_kick_msec + MIN_TX_TIMEOUT_GAP > current_time)
+		return false;
+
+	netdev_info(priv->dev, "Kicking queue %d", txqueue);
+	napi_schedule(&block->napi);
+	block->tx->last_kick_msec = current_time;
+	return true;
+}
+
+static void gve_tx_timeout(struct net_device *dev, unsigned int txqueue)
+{
+	struct gve_notify_block *block;
+	struct gve_priv *priv;
+
+	netdev_info(dev, "Timeout on tx queue, %d", txqueue);
+	priv = netdev_priv(dev);
+
+	if (!gve_tx_timeout_try_q_kick(priv, txqueue))
+		gve_schedule_reset(priv);
+
+	block = gve_get_tx_notify_block(priv, txqueue);
+	if (block)
+		block->tx->queue_timeout++;
+	priv->tx_timeo_cnt++;
+}
+
+u16 gve_get_pkt_buf_size(const struct gve_priv *priv, bool enable_hsplit)
+{
+	if (enable_hsplit && priv->max_rx_buffer_size >= GVE_MAX_RX_BUFFER_SIZE)
+		return GVE_MAX_RX_BUFFER_SIZE;
+	else
+		return GVE_DEFAULT_RX_BUFFER_SIZE;
+}
+
+/* Header split is only supported on DQ RDA queue format. If XDP is enabled,
+ * header split is not allowed.
+ */
+bool gve_header_split_supported(const struct gve_priv *priv)
+{
+	return priv->header_buf_size &&
+		priv->queue_format == GVE_DQO_RDA_FORMAT && !priv->xdp_prog;
+}
+
+int gve_set_hsplit_config(struct gve_priv *priv, u8 tcp_data_split)
+{
+	struct gve_tx_alloc_rings_cfg tx_alloc_cfg = {0};
+	struct gve_rx_alloc_rings_cfg rx_alloc_cfg = {0};
+	bool enable_hdr_split;
+	int err = 0;
+
+	if (tcp_data_split == ETHTOOL_TCP_DATA_SPLIT_UNKNOWN)
+		return 0;
+
+	if (!gve_header_split_supported(priv)) {
+		dev_err(&priv->pdev->dev, "Header-split not supported\n");
+		return -EOPNOTSUPP;
+	}
+
+	if (tcp_data_split == ETHTOOL_TCP_DATA_SPLIT_ENABLED)
+		enable_hdr_split = true;
+	else
+		enable_hdr_split = false;
+
+	if (enable_hdr_split == priv->header_split_enabled)
+		return 0;
+
+	gve_get_curr_alloc_cfgs(priv, &tx_alloc_cfg, &rx_alloc_cfg);
+
+	rx_alloc_cfg.enable_header_split = enable_hdr_split;
+	rx_alloc_cfg.packet_buffer_size = gve_get_pkt_buf_size(priv, enable_hdr_split);
+
+	if (netif_running(priv->dev))
+		err = gve_adjust_config(priv, &tx_alloc_cfg, &rx_alloc_cfg);
+	return err;
+}
+
+static int gve_set_features(struct net_device *netdev,
+			    netdev_features_t features)
+{
+	const netdev_features_t orig_features = netdev->features;
+	struct gve_tx_alloc_rings_cfg tx_alloc_cfg = {0};
+	struct gve_rx_alloc_rings_cfg rx_alloc_cfg = {0};
+	struct gve_priv *priv = netdev_priv(netdev);
+	int err;
+
+	gve_get_curr_alloc_cfgs(priv, &tx_alloc_cfg, &rx_alloc_cfg);
+
+	if ((netdev->features & NETIF_F_LRO) != (features & NETIF_F_LRO)) {
+		netdev->features ^= NETIF_F_LRO;
+		if (priv->xdp_prog && (netdev->features & NETIF_F_LRO)) {
+			netdev_warn(netdev,
+				    "XDP is not supported when LRO is on.\n");
+			err =  -EOPNOTSUPP;
+			goto revert_features;
+		}
+		if (netif_running(netdev)) {
+			err = gve_adjust_config(priv, &tx_alloc_cfg, &rx_alloc_cfg);
+			if (err)
+				goto revert_features;
+		}
+	}
+	if ((netdev->features & NETIF_F_NTUPLE) && !(features & NETIF_F_NTUPLE)) {
+		err = gve_flow_rules_reset(priv);
+		if (err)
+			goto revert_features;
+	}
+
+	return 0;
+
+revert_features:
+	netdev->features = orig_features;
+	return err;
+}
+
+static int gve_get_ts_config(struct net_device *dev,
+			     struct kernel_hwtstamp_config *kernel_config)
+{
+	struct gve_priv *priv = netdev_priv(dev);
+
+	*kernel_config = priv->ts_config;
+	return 0;
+}
+
+static int gve_set_ts_config(struct net_device *dev,
+			     struct kernel_hwtstamp_config *kernel_config,
+			     struct netlink_ext_ack *extack)
+{
+	struct gve_priv *priv = netdev_priv(dev);
+
+	if (kernel_config->tx_type != HWTSTAMP_TX_OFF) {
+		NL_SET_ERR_MSG_MOD(extack, "TX timestamping is not supported");
+		return -ERANGE;
+	}
+
+	if (kernel_config->rx_filter != HWTSTAMP_FILTER_NONE) {
+		if (!priv->nic_ts_report) {
+			NL_SET_ERR_MSG_MOD(extack,
+					   "RX timestamping is not supported");
+			kernel_config->rx_filter = HWTSTAMP_FILTER_NONE;
+			return -EOPNOTSUPP;
+		}
+
+		kernel_config->rx_filter = HWTSTAMP_FILTER_ALL;
+		gve_clock_nic_ts_read(priv);
+		ptp_schedule_worker(priv->ptp->clock, 0);
+	} else {
+		ptp_cancel_worker_sync(priv->ptp->clock);
+	}
+
+	priv->ts_config.rx_filter = kernel_config->rx_filter;
+
+	return 0;
+}
+
+static const struct net_device_ops gve_netdev_ops = {
+	.ndo_start_xmit		=	gve_start_xmit,
+	.ndo_features_check	=	gve_features_check,
+	.ndo_open		=	gve_open,
+	.ndo_stop		=	gve_close,
+	.ndo_get_stats64	=	gve_get_stats,
+	.ndo_tx_timeout         =       gve_tx_timeout,
+	.ndo_set_features	=	gve_set_features,
+	.ndo_bpf		=	gve_xdp,
+	.ndo_xdp_xmit		=	gve_xdp_xmit,
+	.ndo_xsk_wakeup		=	gve_xsk_wakeup,
+	.ndo_hwtstamp_get	=	gve_get_ts_config,
+	.ndo_hwtstamp_set	=	gve_set_ts_config,
+};
+
+static void gve_handle_status(struct gve_priv *priv, u32 status)
+{
+	if (GVE_DEVICE_STATUS_RESET_MASK & status) {
+		dev_info(&priv->pdev->dev, "Device requested reset.\n");
+		gve_set_do_reset(priv);
+	}
+	if (GVE_DEVICE_STATUS_REPORT_STATS_MASK & status) {
+		priv->stats_report_trigger_cnt++;
+		gve_set_do_report_stats(priv);
+	}
+}
+
+static void gve_handle_reset(struct gve_priv *priv)
+{
+	/* A service task will be scheduled at the end of probe to catch any
+	 * resets that need to happen, and we don't want to reset until
+	 * probe is done.
+	 */
+	if (gve_get_probe_in_progress(priv))
+		return;
+
+	if (gve_get_do_reset(priv)) {
+		rtnl_lock();
+		netdev_lock(priv->dev);
+		gve_reset(priv, false);
+		netdev_unlock(priv->dev);
+		rtnl_unlock();
+	}
+}
+
+void gve_handle_report_stats(struct gve_priv *priv)
+{
+	struct stats *stats = priv->stats_report->stats;
+	int idx, stats_idx = 0;
+	unsigned int start = 0;
+	u64 tx_bytes;
+
+	if (!gve_get_report_stats(priv))
+		return;
+
+	be64_add_cpu(&priv->stats_report->written_count, 1);
+	/* tx stats */
+	if (priv->tx) {
+		for (idx = 0; idx < gve_num_tx_queues(priv); idx++) {
+			u32 last_completion = 0;
+			u32 tx_frames = 0;
+
+			/* DQO doesn't currently support these metrics. */
+			if (gve_is_gqi(priv)) {
+				last_completion = priv->tx[idx].done;
+				tx_frames = priv->tx[idx].req;
+			}
+
+			do {
+				start = u64_stats_fetch_begin(&priv->tx[idx].statss);
+				tx_bytes = priv->tx[idx].bytes_done;
+			} while (u64_stats_fetch_retry(&priv->tx[idx].statss, start));
+			stats[stats_idx++] = (struct stats) {
+				.stat_name = cpu_to_be32(TX_WAKE_CNT),
+				.value = cpu_to_be64(priv->tx[idx].wake_queue),
+				.queue_id = cpu_to_be32(idx),
+			};
+			stats[stats_idx++] = (struct stats) {
+				.stat_name = cpu_to_be32(TX_STOP_CNT),
+				.value = cpu_to_be64(priv->tx[idx].stop_queue),
+				.queue_id = cpu_to_be32(idx),
+			};
+			stats[stats_idx++] = (struct stats) {
+				.stat_name = cpu_to_be32(TX_FRAMES_SENT),
+				.value = cpu_to_be64(tx_frames),
+				.queue_id = cpu_to_be32(idx),
+			};
+			stats[stats_idx++] = (struct stats) {
+				.stat_name = cpu_to_be32(TX_BYTES_SENT),
+				.value = cpu_to_be64(tx_bytes),
+				.queue_id = cpu_to_be32(idx),
+			};
+			stats[stats_idx++] = (struct stats) {
+				.stat_name = cpu_to_be32(TX_LAST_COMPLETION_PROCESSED),
+				.value = cpu_to_be64(last_completion),
+				.queue_id = cpu_to_be32(idx),
+			};
+			stats[stats_idx++] = (struct stats) {
+				.stat_name = cpu_to_be32(TX_TIMEOUT_CNT),
+				.value = cpu_to_be64(priv->tx[idx].queue_timeout),
+				.queue_id = cpu_to_be32(idx),
+			};
+		}
+	}
+	/* rx stats */
+	if (priv->rx) {
+		for (idx = 0; idx < priv->rx_cfg.num_queues; idx++) {
+			stats[stats_idx++] = (struct stats) {
+				.stat_name = cpu_to_be32(RX_NEXT_EXPECTED_SEQUENCE),
+				.value = cpu_to_be64(priv->rx[idx].desc.seqno),
+				.queue_id = cpu_to_be32(idx),
+			};
+			stats[stats_idx++] = (struct stats) {
+				.stat_name = cpu_to_be32(RX_BUFFERS_POSTED),
+				.value = cpu_to_be64(priv->rx[idx].fill_cnt),
+				.queue_id = cpu_to_be32(idx),
+			};
+		}
+	}
+}
+
+/* Handle NIC status register changes, reset requests and report stats */
+static void gve_service_task(struct work_struct *work)
+{
+	struct gve_priv *priv = container_of(work, struct gve_priv,
+					     service_task);
+	u32 status = ioread32be(&priv->reg_bar0->device_status);
+
+	gve_handle_status(priv, status);
+
+	gve_handle_reset(priv);
+	gve_handle_link_status(priv, GVE_DEVICE_STATUS_LINK_STATUS_MASK & status);
+}
+
+static void gve_set_netdev_xdp_features(struct gve_priv *priv)
+{
+	xdp_features_t xdp_features;
+
+	if (priv->queue_format == GVE_GQI_QPL_FORMAT) {
+		xdp_features = NETDEV_XDP_ACT_BASIC;
+		xdp_features |= NETDEV_XDP_ACT_REDIRECT;
+		xdp_features |= NETDEV_XDP_ACT_XSK_ZEROCOPY;
+	} else if (priv->queue_format == GVE_DQO_RDA_FORMAT) {
+		xdp_features = NETDEV_XDP_ACT_BASIC;
+		xdp_features |= NETDEV_XDP_ACT_REDIRECT;
+		xdp_features |= NETDEV_XDP_ACT_XSK_ZEROCOPY;
+	} else {
+		xdp_features = 0;
+	}
+
+	xdp_set_features_flag_locked(priv->dev, xdp_features);
+}
+
+static int gve_init_priv(struct gve_priv *priv, bool skip_describe_device)
+{
+	int num_ntfy;
+	int err;
+
+	/* Set up the adminq */
+	err = gve_adminq_alloc(&priv->pdev->dev, priv);
+	if (err) {
+		dev_err(&priv->pdev->dev,
+			"Failed to alloc admin queue: err=%d\n", err);
+		return err;
+	}
+
+	err = gve_verify_driver_compatibility(priv);
+	if (err) {
+		dev_err(&priv->pdev->dev,
+			"Could not verify driver compatibility: err=%d\n", err);
+		goto err;
+	}
+
+	priv->num_registered_pages = 0;
+
+	if (skip_describe_device)
+		goto setup_device;
+
+	priv->queue_format = GVE_QUEUE_FORMAT_UNSPECIFIED;
+	/* Get the initial information we need from the device */
+	err = gve_adminq_describe_device(priv);
+	if (err) {
+		dev_err(&priv->pdev->dev,
+			"Could not get device information: err=%d\n", err);
+		goto err;
+	}
+	priv->dev->mtu = priv->dev->max_mtu;
+	num_ntfy = pci_msix_vec_count(priv->pdev);
+	if (num_ntfy <= 0) {
+		dev_err(&priv->pdev->dev,
+			"could not count MSI-x vectors: err=%d\n", num_ntfy);
+		err = num_ntfy;
+		goto err;
+	} else if (num_ntfy < GVE_MIN_MSIX) {
+		dev_err(&priv->pdev->dev, "gve needs at least %d MSI-x vectors, but only has %d\n",
+			GVE_MIN_MSIX, num_ntfy);
+		err = -EINVAL;
+		goto err;
+	}
+
+	/* Big TCP is only supported on DQO */
+	if (!gve_is_gqi(priv))
+		netif_set_tso_max_size(priv->dev, GVE_DQO_TX_MAX);
+
+	priv->rx_copybreak = GVE_DEFAULT_RX_COPYBREAK;
+	/* gvnic has one Notification Block per MSI-x vector, except for the
+	 * management vector
+	 */
+	priv->num_ntfy_blks = (num_ntfy - 1) & ~0x1;
+	priv->mgmt_msix_idx = priv->num_ntfy_blks;
+	priv->numa_node = dev_to_node(&priv->pdev->dev);
+
+	priv->tx_cfg.max_queues =
+		min_t(int, priv->tx_cfg.max_queues, priv->num_ntfy_blks / 2);
+	priv->rx_cfg.max_queues =
+		min_t(int, priv->rx_cfg.max_queues, priv->num_ntfy_blks / 2);
+
+	priv->tx_cfg.num_queues = priv->tx_cfg.max_queues;
+	priv->rx_cfg.num_queues = priv->rx_cfg.max_queues;
+	if (priv->default_num_queues > 0) {
+		priv->tx_cfg.num_queues = min_t(int, priv->default_num_queues,
+						priv->tx_cfg.num_queues);
+		priv->rx_cfg.num_queues = min_t(int, priv->default_num_queues,
+						priv->rx_cfg.num_queues);
+	}
+	priv->tx_cfg.num_xdp_queues = 0;
+
+	dev_info(&priv->pdev->dev, "TX queues %d, RX queues %d\n",
+		 priv->tx_cfg.num_queues, priv->rx_cfg.num_queues);
+	dev_info(&priv->pdev->dev, "Max TX queues %d, Max RX queues %d\n",
+		 priv->tx_cfg.max_queues, priv->rx_cfg.max_queues);
+
+	if (!gve_is_gqi(priv)) {
+		priv->tx_coalesce_usecs = GVE_TX_IRQ_RATELIMIT_US_DQO;
+		priv->rx_coalesce_usecs = GVE_RX_IRQ_RATELIMIT_US_DQO;
+	}
+
+	priv->ts_config.tx_type = HWTSTAMP_TX_OFF;
+	priv->ts_config.rx_filter = HWTSTAMP_FILTER_NONE;
+
+setup_device:
+	priv->xsk_pools = bitmap_zalloc(priv->rx_cfg.max_queues, GFP_KERNEL);
+	if (!priv->xsk_pools) {
+		err = -ENOMEM;
+		goto err;
+	}
+
+	gve_set_netdev_xdp_features(priv);
+	err = gve_setup_device_resources(priv);
+	if (err)
+		goto err_free_xsk_bitmap;
+
+	return 0;
+
+err_free_xsk_bitmap:
+	bitmap_free(priv->xsk_pools);
+	priv->xsk_pools = NULL;
+err:
+	gve_adminq_free(&priv->pdev->dev, priv);
+	return err;
+}
+
+static void gve_teardown_priv_resources(struct gve_priv *priv)
+{
+	gve_teardown_device_resources(priv);
+	gve_adminq_free(&priv->pdev->dev, priv);
+	bitmap_free(priv->xsk_pools);
+	priv->xsk_pools = NULL;
+}
+
+static void gve_trigger_reset(struct gve_priv *priv)
+{
+	/* Reset the device by releasing the AQ */
+	gve_adminq_release(priv);
+}
+
+static void gve_reset_and_teardown(struct gve_priv *priv, bool was_up)
+{
+	gve_trigger_reset(priv);
+	/* With the reset having already happened, close cannot fail */
+	if (was_up)
+		gve_close(priv->dev);
+	gve_teardown_priv_resources(priv);
+}
+
+static int gve_reset_recovery(struct gve_priv *priv, bool was_up)
+{
+	int err;
+
+	err = gve_init_priv(priv, true);
+	if (err)
+		goto err;
+	if (was_up) {
+		err = gve_open(priv->dev);
+		if (err)
+			goto err;
+	}
+	return 0;
+err:
+	dev_err(&priv->pdev->dev, "Reset failed! !!! DISABLING ALL QUEUES !!!\n");
+	gve_turndown(priv);
+	return err;
+}
+
+int gve_reset(struct gve_priv *priv, bool attempt_teardown)
+{
+	bool was_up = netif_running(priv->dev);
+	int err;
+
+	dev_info(&priv->pdev->dev, "Performing reset\n");
+	gve_clear_do_reset(priv);
+	gve_set_reset_in_progress(priv);
+	/* If we aren't attempting to teardown normally, just go turndown and
+	 * reset right away.
+	 */
+	if (!attempt_teardown) {
+		gve_turndown(priv);
+		gve_reset_and_teardown(priv, was_up);
+	} else {
+		/* Otherwise attempt to close normally */
+		if (was_up) {
+			err = gve_close(priv->dev);
+			/* If that fails reset as we did above */
+			if (err)
+				gve_reset_and_teardown(priv, was_up);
+		}
+		/* Clean up any remaining resources */
+		gve_teardown_priv_resources(priv);
+	}
+
+	/* Set it all back up */
+	err = gve_reset_recovery(priv, was_up);
+	gve_clear_reset_in_progress(priv);
+	priv->reset_cnt++;
+	priv->interface_up_cnt = 0;
+	priv->interface_down_cnt = 0;
+	priv->stats_report_trigger_cnt = 0;
+	return err;
+}
+
+static void gve_write_version(u8 __iomem *driver_version_register)
+{
+	const char *c = gve_version_prefix;
+
+	while (*c) {
+		writeb(*c, driver_version_register);
+		c++;
+	}
+
+	c = gve_version_str;
+	while (*c) {
+		writeb(*c, driver_version_register);
+		c++;
+	}
+	writeb('\n', driver_version_register);
+}
+
+static int gve_rx_queue_stop(struct net_device *dev, void *per_q_mem, int idx)
+{
+	struct gve_priv *priv = netdev_priv(dev);
+	struct gve_rx_ring *gve_per_q_mem;
+	int err;
+
+	if (!priv->rx)
+		return -EAGAIN;
+
+	/* Destroying queue 0 while other queues exist is not supported in DQO */
+	if (!gve_is_gqi(priv) && idx == 0)
+		return -ERANGE;
+
+	/* Single-queue destruction requires quiescence on all queues */
+	gve_turndown(priv);
+
+	/* This failure will trigger a reset - no need to clean up */
+	err = gve_adminq_destroy_single_rx_queue(priv, idx);
+	if (err)
+		return err;
+
+	if (gve_is_qpl(priv)) {
+		/* This failure will trigger a reset - no need to clean up */
+		err = gve_unregister_qpl(priv, gve_rx_get_qpl(priv, idx));
+		if (err)
+			return err;
+	}
+
+	gve_rx_stop_ring(priv, idx);
+
+	/* Turn the unstopped queues back up */
+	gve_turnup_and_check_status(priv);
+
+	gve_per_q_mem = (struct gve_rx_ring *)per_q_mem;
+	*gve_per_q_mem = priv->rx[idx];
+	memset(&priv->rx[idx], 0, sizeof(priv->rx[idx]));
+	return 0;
+}
+
+static void gve_rx_queue_mem_free(struct net_device *dev, void *per_q_mem)
+{
+	struct gve_priv *priv = netdev_priv(dev);
+	struct gve_rx_alloc_rings_cfg cfg = {0};
+	struct gve_rx_ring *gve_per_q_mem;
+
+	gve_per_q_mem = (struct gve_rx_ring *)per_q_mem;
+	gve_rx_get_curr_alloc_cfg(priv, &cfg);
+
+	if (gve_is_gqi(priv))
+		gve_rx_free_ring_gqi(priv, gve_per_q_mem, &cfg);
+	else
+		gve_rx_free_ring_dqo(priv, gve_per_q_mem, &cfg);
+}
+
+static int gve_rx_queue_mem_alloc(struct net_device *dev, void *per_q_mem,
+				  int idx)
+{
+	struct gve_priv *priv = netdev_priv(dev);
+	struct gve_rx_alloc_rings_cfg cfg = {0};
+	struct gve_rx_ring *gve_per_q_mem;
+	int err;
+
+	if (!priv->rx)
+		return -EAGAIN;
+
+	gve_per_q_mem = (struct gve_rx_ring *)per_q_mem;
+	gve_rx_get_curr_alloc_cfg(priv, &cfg);
+
+	if (gve_is_gqi(priv))
+		err = gve_rx_alloc_ring_gqi(priv, &cfg, gve_per_q_mem, idx);
+	else
+		err = gve_rx_alloc_ring_dqo(priv, &cfg, gve_per_q_mem, idx);
+
+	return err;
+}
+
+static int gve_rx_queue_start(struct net_device *dev, void *per_q_mem, int idx)
+{
+	struct gve_priv *priv = netdev_priv(dev);
+	struct gve_rx_ring *gve_per_q_mem;
+	int err;
+
+	if (!priv->rx)
+		return -EAGAIN;
+
+	gve_per_q_mem = (struct gve_rx_ring *)per_q_mem;
+	priv->rx[idx] = *gve_per_q_mem;
+
+	/* Single-queue creation requires quiescence on all queues */
+	gve_turndown(priv);
+
+	gve_rx_start_ring(priv, idx);
+
+	if (gve_is_qpl(priv)) {
+		/* This failure will trigger a reset - no need to clean up */
+		err = gve_register_qpl(priv, gve_rx_get_qpl(priv, idx));
+		if (err)
+			goto abort;
+	}
+
+	/* This failure will trigger a reset - no need to clean up */
+	err = gve_adminq_create_single_rx_queue(priv, idx);
+	if (err)
+		goto abort;
+
+	if (gve_is_gqi(priv))
+		gve_rx_write_doorbell(priv, &priv->rx[idx]);
+	else
+		gve_rx_post_buffers_dqo(&priv->rx[idx]);
+
+	/* Turn the unstopped queues back up */
+	gve_turnup_and_check_status(priv);
+	return 0;
+
+abort:
+	gve_rx_stop_ring(priv, idx);
+
+	/* All failures in this func result in a reset, by clearing the struct
+	 * at idx, we prevent a double free when that reset runs. The reset,
+	 * which needs the rtnl lock, will not run till this func returns and
+	 * its caller gives up the lock.
+	 */
+	memset(&priv->rx[idx], 0, sizeof(priv->rx[idx]));
+	return err;
+}
+
+static const struct netdev_queue_mgmt_ops gve_queue_mgmt_ops = {
+	.ndo_queue_mem_size	=	sizeof(struct gve_rx_ring),
+	.ndo_queue_mem_alloc	=	gve_rx_queue_mem_alloc,
+	.ndo_queue_mem_free	=	gve_rx_queue_mem_free,
+	.ndo_queue_start	=	gve_rx_queue_start,
+	.ndo_queue_stop		=	gve_rx_queue_stop,
+};
+
+static void gve_get_rx_queue_stats(struct net_device *dev, int idx,
+				   struct netdev_queue_stats_rx *rx_stats)
+{
+	struct gve_priv *priv = netdev_priv(dev);
+	struct gve_rx_ring *rx = &priv->rx[idx];
+	unsigned int start;
+
+	do {
+		start = u64_stats_fetch_begin(&rx->statss);
+		rx_stats->packets = rx->rpackets;
+		rx_stats->bytes = rx->rbytes;
+		rx_stats->alloc_fail = rx->rx_skb_alloc_fail +
+				       rx->rx_buf_alloc_fail;
+	} while (u64_stats_fetch_retry(&rx->statss, start));
+}
+
+static void gve_get_tx_queue_stats(struct net_device *dev, int idx,
+				   struct netdev_queue_stats_tx *tx_stats)
+{
+	struct gve_priv *priv = netdev_priv(dev);
+	struct gve_tx_ring *tx = &priv->tx[idx];
+	unsigned int start;
+
+	do {
+		start = u64_stats_fetch_begin(&tx->statss);
+		tx_stats->packets = tx->pkt_done;
+		tx_stats->bytes = tx->bytes_done;
+	} while (u64_stats_fetch_retry(&tx->statss, start));
+}
+
+static void gve_get_base_stats(struct net_device *dev,
+			       struct netdev_queue_stats_rx *rx,
+			       struct netdev_queue_stats_tx *tx)
+{
+	rx->packets = 0;
+	rx->bytes = 0;
+	rx->alloc_fail = 0;
+
+	tx->packets = 0;
+	tx->bytes = 0;
+}
+
+static const struct netdev_stat_ops gve_stat_ops = {
+	.get_queue_stats_rx	= gve_get_rx_queue_stats,
+	.get_queue_stats_tx	= gve_get_tx_queue_stats,
+	.get_base_stats		= gve_get_base_stats,
+};
+
+static int gve_probe(struct pci_dev *pdev, const struct pci_device_id *ent)
+{
+	int max_tx_queues, max_rx_queues;
+	struct net_device *dev;
+	__be32 __iomem *db_bar;
+	struct gve_registers __iomem *reg_bar;
+	struct gve_priv *priv;
+	int err;
+
+	err = pci_enable_device(pdev);
+	if (err)
+		return err;
+
+	err = pci_request_regions(pdev, gve_driver_name);
+	if (err)
+		goto abort_with_enabled;
+
+	pci_set_master(pdev);
+
+	err = dma_set_mask_and_coherent(&pdev->dev, DMA_BIT_MASK(64));
+	if (err) {
+		dev_err(&pdev->dev, "Failed to set dma mask: err=%d\n", err);
+		goto abort_with_pci_region;
+	}
+
+	reg_bar = pci_iomap(pdev, GVE_REGISTER_BAR, 0);
+	if (!reg_bar) {
+		dev_err(&pdev->dev, "Failed to map pci bar!\n");
+		err = -ENOMEM;
+		goto abort_with_pci_region;
+	}
+
+	db_bar = pci_iomap(pdev, GVE_DOORBELL_BAR, 0);
+	if (!db_bar) {
+		dev_err(&pdev->dev, "Failed to map doorbell bar!\n");
+		err = -ENOMEM;
+		goto abort_with_reg_bar;
+	}
+
+	gve_write_version(&reg_bar->driver_version);
+	/* Get max queues to alloc etherdev */
+	max_tx_queues = ioread32be(&reg_bar->max_tx_queues);
+	max_rx_queues = ioread32be(&reg_bar->max_rx_queues);
+	/* Alloc and setup the netdev and priv */
+	dev = alloc_etherdev_mqs(sizeof(*priv), max_tx_queues, max_rx_queues);
+	if (!dev) {
+		dev_err(&pdev->dev, "could not allocate netdev\n");
+		err = -ENOMEM;
+		goto abort_with_db_bar;
+	}
+	SET_NETDEV_DEV(dev, &pdev->dev);
+	pci_set_drvdata(pdev, dev);
+	dev->ethtool_ops = &gve_ethtool_ops;
+	dev->netdev_ops = &gve_netdev_ops;
+	dev->queue_mgmt_ops = &gve_queue_mgmt_ops;
+	dev->stat_ops = &gve_stat_ops;
+
+	/* Set default and supported features.
+	 *
+	 * Features might be set in other locations as well (such as
+	 * `gve_adminq_describe_device`).
+	 */
+	dev->hw_features = NETIF_F_HIGHDMA;
+	dev->hw_features |= NETIF_F_SG;
+	dev->hw_features |= NETIF_F_HW_CSUM;
+	dev->hw_features |= NETIF_F_TSO;
+	dev->hw_features |= NETIF_F_TSO6;
+	dev->hw_features |= NETIF_F_TSO_ECN;
+	dev->hw_features |= NETIF_F_RXCSUM;
+	dev->hw_features |= NETIF_F_RXHASH;
+	dev->features = dev->hw_features;
+	dev->watchdog_timeo = 5 * HZ;
+	dev->min_mtu = ETH_MIN_MTU;
+	netif_carrier_off(dev);
+
+	priv = netdev_priv(dev);
+	priv->dev = dev;
+	priv->pdev = pdev;
+	priv->msg_enable = DEFAULT_MSG_LEVEL;
+	priv->reg_bar0 = reg_bar;
+	priv->db_bar2 = db_bar;
+	priv->service_task_flags = 0x0;
+	priv->state_flags = 0x0;
+	priv->ethtool_flags = 0x0;
+	priv->rx_cfg.packet_buffer_size = GVE_DEFAULT_RX_BUFFER_SIZE;
+	priv->max_rx_buffer_size = GVE_DEFAULT_RX_BUFFER_SIZE;
+
+	gve_set_probe_in_progress(priv);
+	priv->gve_wq = alloc_ordered_workqueue("gve", 0);
+	if (!priv->gve_wq) {
+		dev_err(&pdev->dev, "Could not allocate workqueue");
+		err = -ENOMEM;
+		goto abort_with_netdev;
+	}
+	INIT_WORK(&priv->service_task, gve_service_task);
+	INIT_WORK(&priv->stats_report_task, gve_stats_report_task);
+	priv->tx_cfg.max_queues = max_tx_queues;
+	priv->rx_cfg.max_queues = max_rx_queues;
+
+	err = gve_init_priv(priv, false);
+	if (err)
+		goto abort_with_wq;
+
+	if (!gve_is_gqi(priv) && !gve_is_qpl(priv))
+		dev->netmem_tx = true;
+
+	err = register_netdev(dev);
+	if (err)
+		goto abort_with_gve_init;
+
+	dev_info(&pdev->dev, "GVE version %s\n", gve_version_str);
+	dev_info(&pdev->dev, "GVE queue format %d\n", (int)priv->queue_format);
+	gve_clear_probe_in_progress(priv);
+	queue_work(priv->gve_wq, &priv->service_task);
+	return 0;
+
+abort_with_gve_init:
+	gve_teardown_priv_resources(priv);
+
+abort_with_wq:
+	destroy_workqueue(priv->gve_wq);
+
+abort_with_netdev:
+	free_netdev(dev);
+
+abort_with_db_bar:
+	pci_iounmap(pdev, db_bar);
+
+abort_with_reg_bar:
+	pci_iounmap(pdev, reg_bar);
+
+abort_with_pci_region:
+	pci_release_regions(pdev);
+
+abort_with_enabled:
+	pci_disable_device(pdev);
+	return err;
+}
+
+static void gve_remove(struct pci_dev *pdev)
+{
+	struct net_device *netdev = pci_get_drvdata(pdev);
+	struct gve_priv *priv = netdev_priv(netdev);
+	__be32 __iomem *db_bar = priv->db_bar2;
+	void __iomem *reg_bar = priv->reg_bar0;
+
+	unregister_netdev(netdev);
+	gve_teardown_priv_resources(priv);
+	destroy_workqueue(priv->gve_wq);
+	free_netdev(netdev);
+	pci_iounmap(pdev, db_bar);
+	pci_iounmap(pdev, reg_bar);
+	pci_release_regions(pdev);
+	pci_disable_device(pdev);
+}
+
+static void gve_shutdown(struct pci_dev *pdev)
+{
+	struct net_device *netdev = pci_get_drvdata(pdev);
+	struct gve_priv *priv = netdev_priv(netdev);
+	bool was_up = netif_running(priv->dev);
+
+	netif_device_detach(netdev);
+
+	rtnl_lock();
+	netdev_lock(netdev);
+	if (was_up && gve_close(priv->dev)) {
+		/* If the dev was up, attempt to close, if close fails, reset */
+		gve_reset_and_teardown(priv, was_up);
+	} else {
+		/* If the dev wasn't up or close worked, finish tearing down */
+		gve_teardown_priv_resources(priv);
+	}
+	netdev_unlock(netdev);
+	rtnl_unlock();
+}
+
+#ifdef CONFIG_PM
+static int gve_suspend(struct pci_dev *pdev, pm_message_t state)
+{
+	struct net_device *netdev = pci_get_drvdata(pdev);
+	struct gve_priv *priv = netdev_priv(netdev);
+	bool was_up = netif_running(priv->dev);
+
+	priv->suspend_cnt++;
+	rtnl_lock();
+	netdev_lock(netdev);
+	if (was_up && gve_close(priv->dev)) {
+		/* If the dev was up, attempt to close, if close fails, reset */
+		gve_reset_and_teardown(priv, was_up);
+	} else {
+		/* If the dev wasn't up or close worked, finish tearing down */
+		gve_teardown_priv_resources(priv);
+	}
+	priv->up_before_suspend = was_up;
+	netdev_unlock(netdev);
+	rtnl_unlock();
+	return 0;
+}
+
+static int gve_resume(struct pci_dev *pdev)
+{
+	struct net_device *netdev = pci_get_drvdata(pdev);
+	struct gve_priv *priv = netdev_priv(netdev);
+	int err;
+
+	priv->resume_cnt++;
+	rtnl_lock();
+	netdev_lock(netdev);
+	err = gve_reset_recovery(priv, priv->up_before_suspend);
+	netdev_unlock(netdev);
+	rtnl_unlock();
+	return err;
+}
+#endif /* CONFIG_PM */
+
+static const struct pci_device_id gve_id_table[] = {
+	{ PCI_DEVICE(PCI_VENDOR_ID_GOOGLE, PCI_DEV_ID_GVNIC) },
+	{ }
+};
+
+static struct pci_driver gve_driver = {
+	.name		= gve_driver_name,
+	.id_table	= gve_id_table,
+	.probe		= gve_probe,
+	.remove		= gve_remove,
+	.shutdown	= gve_shutdown,
+#ifdef CONFIG_PM
+	.suspend        = gve_suspend,
+	.resume         = gve_resume,
+#endif
+};
+
+module_pci_driver(gve_driver);
+
+MODULE_DEVICE_TABLE(pci, gve_id_table);
+MODULE_AUTHOR("Google, Inc.");
+MODULE_DESCRIPTION("Google Virtual NIC Driver");
+MODULE_LICENSE("Dual MIT/GPL");
+MODULE_VERSION(GVE_VERSION);
diff --git a/drivers/net/ethernet/google/gve/gve_ptp.c b/drivers/net/ethernet/google/gve/gve_ptp.c
new file mode 100644
index 000000000000..e96247c9d68d
--- /dev/null
+++ b/drivers/net/ethernet/google/gve/gve_ptp.c
@@ -0,0 +1,139 @@
+// SPDX-License-Identifier: (GPL-2.0 OR MIT)
+/* Google virtual Ethernet (gve) driver
+ *
+ * Copyright (C) 2025 Google LLC
+ */
+
+#include "gve.h"
+#include "gve_adminq.h"
+
+/* Interval to schedule a nic timestamp calibration, 250ms. */
+#define GVE_NIC_TS_SYNC_INTERVAL_MS 250
+
+/* Read the nic timestamp from hardware via the admin queue. */
+int gve_clock_nic_ts_read(struct gve_priv *priv)
+{
+	u64 nic_raw;
+	int err;
+
+	err = gve_adminq_report_nic_ts(priv, priv->nic_ts_report_bus);
+	if (err)
+		return err;
+
+	nic_raw = be64_to_cpu(priv->nic_ts_report->nic_timestamp);
+	WRITE_ONCE(priv->last_sync_nic_counter, nic_raw);
+
+	return 0;
+}
+
+static long gve_ptp_do_aux_work(struct ptp_clock_info *info)
+{
+	const struct gve_ptp *ptp = container_of(info, struct gve_ptp, info);
+	struct gve_priv *priv = ptp->priv;
+	int err;
+
+	if (gve_get_reset_in_progress(priv) || !gve_get_admin_queue_ok(priv))
+		goto out;
+
+	err = gve_clock_nic_ts_read(priv);
+	if (err && net_ratelimit())
+		dev_err(&priv->pdev->dev,
+			"%s read err %d\n", __func__, err);
+
+out:
+	return msecs_to_jiffies(GVE_NIC_TS_SYNC_INTERVAL_MS);
+}
+
+static const struct ptp_clock_info gve_ptp_caps = {
+	.owner          = THIS_MODULE,
+	.name		= "gve clock",
+	.do_aux_work	= gve_ptp_do_aux_work,
+};
+
+static int gve_ptp_init(struct gve_priv *priv)
+{
+	struct gve_ptp *ptp;
+	int err;
+
+	if (!priv->nic_timestamp_supported) {
+		dev_dbg(&priv->pdev->dev, "Device does not support PTP\n");
+		return -EOPNOTSUPP;
+	}
+
+	priv->ptp = kzalloc(sizeof(*priv->ptp), GFP_KERNEL);
+	if (!priv->ptp)
+		return -ENOMEM;
+
+	ptp = priv->ptp;
+	ptp->info = gve_ptp_caps;
+	ptp->clock = ptp_clock_register(&ptp->info, &priv->pdev->dev);
+
+	if (IS_ERR(ptp->clock)) {
+		dev_err(&priv->pdev->dev, "PTP clock registration failed\n");
+		err  = PTR_ERR(ptp->clock);
+		goto free_ptp;
+	}
+
+	ptp->priv = priv;
+	return 0;
+
+free_ptp:
+	kfree(ptp);
+	priv->ptp = NULL;
+	return err;
+}
+
+static void gve_ptp_release(struct gve_priv *priv)
+{
+	struct gve_ptp *ptp = priv->ptp;
+
+	if (!ptp)
+		return;
+
+	if (ptp->clock)
+		ptp_clock_unregister(ptp->clock);
+
+	kfree(ptp);
+	priv->ptp = NULL;
+}
+
+int gve_init_clock(struct gve_priv *priv)
+{
+	int err;
+
+	if (!priv->nic_timestamp_supported)
+		return 0;
+
+	err = gve_ptp_init(priv);
+	if (err)
+		return err;
+
+	priv->nic_ts_report =
+		dma_alloc_coherent(&priv->pdev->dev,
+				   sizeof(struct gve_nic_ts_report),
+				   &priv->nic_ts_report_bus,
+				   GFP_KERNEL);
+	if (!priv->nic_ts_report) {
+		dev_err(&priv->pdev->dev, "%s dma alloc error\n", __func__);
+		err = -ENOMEM;
+		goto release_ptp;
+	}
+
+	return 0;
+
+release_ptp:
+	gve_ptp_release(priv);
+	return err;
+}
+
+void gve_teardown_clock(struct gve_priv *priv)
+{
+	gve_ptp_release(priv);
+
+	if (priv->nic_ts_report) {
+		dma_free_coherent(&priv->pdev->dev,
+				  sizeof(struct gve_nic_ts_report),
+				  priv->nic_ts_report, priv->nic_ts_report_bus);
+		priv->nic_ts_report = NULL;
+	}
+}
diff --git a/drivers/net/ethernet/google/gve/gve_register.h b/drivers/net/ethernet/google/gve/gve_register.h
new file mode 100644
index 000000000000..8e72b97008d6
--- /dev/null
+++ b/drivers/net/ethernet/google/gve/gve_register.h
@@ -0,0 +1,37 @@
+/* SPDX-License-Identifier: (GPL-2.0 OR MIT)
+ * Google virtual Ethernet (gve) driver
+ *
+ * Copyright (C) 2015-2019 Google, Inc.
+ */
+
+#ifndef _GVE_REGISTER_H_
+#define _GVE_REGISTER_H_
+
+/* Fixed Configuration Registers */
+struct gve_registers {
+	__be32	device_status;
+	__be32	driver_status;
+	__be32	max_tx_queues;
+	__be32	max_rx_queues;
+	__be32	adminq_pfn;
+	__be32	adminq_doorbell;
+	__be32	adminq_event_counter;
+	u8	reserved[3];
+	u8	driver_version;
+	__be32	adminq_base_address_hi;
+	__be32	adminq_base_address_lo;
+	__be16	adminq_length;
+};
+
+enum gve_device_status_flags {
+	GVE_DEVICE_STATUS_RESET_MASK		= BIT(1),
+	GVE_DEVICE_STATUS_LINK_STATUS_MASK	= BIT(2),
+	GVE_DEVICE_STATUS_REPORT_STATS_MASK	= BIT(3),
+	GVE_DEVICE_STATUS_DEVICE_IS_RESET	= BIT(4),
+};
+
+enum gve_driver_status_flags {
+	GVE_DRIVER_STATUS_RUN_MASK		= BIT(0),
+	GVE_DRIVER_STATUS_RESET_MASK		= BIT(1),
+};
+#endif /* _GVE_REGISTER_H_ */
diff --git a/drivers/net/ethernet/google/gve/gve_rx.c b/drivers/net/ethernet/google/gve/gve_rx.c
new file mode 100644
index 000000000000..ec424d2f4f57
--- /dev/null
+++ b/drivers/net/ethernet/google/gve/gve_rx.c
@@ -0,0 +1,1098 @@
+// SPDX-License-Identifier: (GPL-2.0 OR MIT)
+/* Google virtual Ethernet (gve) driver
+ *
+ * Copyright (C) 2015-2021 Google, Inc.
+ */
+
+#include "gve.h"
+#include "gve_adminq.h"
+#include "gve_utils.h"
+#include <linux/etherdevice.h>
+#include <linux/filter.h>
+#include <net/xdp.h>
+#include <net/xdp_sock_drv.h>
+
+static void gve_rx_free_buffer(struct device *dev,
+			       struct gve_rx_slot_page_info *page_info,
+			       union gve_rx_data_slot *data_slot)
+{
+	dma_addr_t dma = (dma_addr_t)(be64_to_cpu(data_slot->addr) &
+				      GVE_DATA_SLOT_ADDR_PAGE_MASK);
+
+	page_ref_sub(page_info->page, page_info->pagecnt_bias - 1);
+	gve_free_page(dev, page_info->page, dma, DMA_FROM_DEVICE);
+}
+
+static void gve_rx_unfill_pages(struct gve_priv *priv,
+				struct gve_rx_ring *rx,
+				struct gve_rx_alloc_rings_cfg *cfg)
+{
+	u32 slots = rx->mask + 1;
+	int i;
+
+	if (!rx->data.page_info)
+		return;
+
+	if (rx->data.raw_addressing) {
+		for (i = 0; i < slots; i++)
+			gve_rx_free_buffer(&priv->pdev->dev, &rx->data.page_info[i],
+					   &rx->data.data_ring[i]);
+	} else {
+		for (i = 0; i < slots; i++)
+			page_ref_sub(rx->data.page_info[i].page,
+				     rx->data.page_info[i].pagecnt_bias - 1);
+
+		for (i = 0; i < rx->qpl_copy_pool_mask + 1; i++) {
+			page_ref_sub(rx->qpl_copy_pool[i].page,
+				     rx->qpl_copy_pool[i].pagecnt_bias - 1);
+			put_page(rx->qpl_copy_pool[i].page);
+		}
+	}
+	kvfree(rx->data.page_info);
+	rx->data.page_info = NULL;
+}
+
+static void gve_rx_ctx_clear(struct gve_rx_ctx *ctx)
+{
+	ctx->skb_head = NULL;
+	ctx->skb_tail = NULL;
+	ctx->total_size = 0;
+	ctx->frag_cnt = 0;
+	ctx->drop_pkt = false;
+}
+
+static void gve_rx_init_ring_state_gqi(struct gve_rx_ring *rx)
+{
+	rx->desc.seqno = 1;
+	rx->cnt = 0;
+	gve_rx_ctx_clear(&rx->ctx);
+}
+
+static void gve_rx_reset_ring_gqi(struct gve_priv *priv, int idx)
+{
+	struct gve_rx_ring *rx = &priv->rx[idx];
+	const u32 slots = priv->rx_desc_cnt;
+	size_t size;
+
+	/* Reset desc ring */
+	if (rx->desc.desc_ring) {
+		size = slots * sizeof(rx->desc.desc_ring[0]);
+		memset(rx->desc.desc_ring, 0, size);
+	}
+
+	/* Reset q_resources */
+	if (rx->q_resources)
+		memset(rx->q_resources, 0, sizeof(*rx->q_resources));
+
+	gve_rx_init_ring_state_gqi(rx);
+}
+
+void gve_rx_stop_ring_gqi(struct gve_priv *priv, int idx)
+{
+	int ntfy_idx = gve_rx_idx_to_ntfy(priv, idx);
+
+	if (!gve_rx_was_added_to_block(priv, idx))
+		return;
+
+	gve_remove_napi(priv, ntfy_idx);
+	gve_rx_remove_from_block(priv, idx);
+	gve_rx_reset_ring_gqi(priv, idx);
+}
+
+void gve_rx_free_ring_gqi(struct gve_priv *priv, struct gve_rx_ring *rx,
+			  struct gve_rx_alloc_rings_cfg *cfg)
+{
+	struct device *dev = &priv->pdev->dev;
+	u32 slots = rx->mask + 1;
+	int idx = rx->q_num;
+	size_t bytes;
+	u32 qpl_id;
+
+	if (rx->desc.desc_ring) {
+		bytes = sizeof(struct gve_rx_desc) * cfg->ring_size;
+		dma_free_coherent(dev, bytes, rx->desc.desc_ring, rx->desc.bus);
+		rx->desc.desc_ring = NULL;
+	}
+
+	if (rx->q_resources) {
+		dma_free_coherent(dev, sizeof(*rx->q_resources),
+				  rx->q_resources, rx->q_resources_bus);
+		rx->q_resources = NULL;
+	}
+
+	gve_rx_unfill_pages(priv, rx, cfg);
+
+	if (rx->data.data_ring) {
+		bytes = sizeof(*rx->data.data_ring) * slots;
+		dma_free_coherent(dev, bytes, rx->data.data_ring,
+				  rx->data.data_bus);
+		rx->data.data_ring = NULL;
+	}
+
+	kvfree(rx->qpl_copy_pool);
+	rx->qpl_copy_pool = NULL;
+
+	if (rx->data.qpl) {
+		qpl_id = gve_get_rx_qpl_id(cfg->qcfg_tx, idx);
+		gve_free_queue_page_list(priv, rx->data.qpl, qpl_id);
+		rx->data.qpl = NULL;
+	}
+
+	netif_dbg(priv, drv, priv->dev, "freed rx ring %d\n", idx);
+}
+
+static void gve_setup_rx_buffer(struct gve_rx_ring *rx,
+				struct gve_rx_slot_page_info *page_info,
+				dma_addr_t addr, struct page *page,
+				__be64 *slot_addr)
+{
+	page_info->page = page;
+	page_info->page_offset = 0;
+	page_info->page_address = page_address(page);
+	page_info->buf_size = rx->packet_buffer_size;
+	*slot_addr = cpu_to_be64(addr);
+	/* The page already has 1 ref */
+	page_ref_add(page, INT_MAX - 1);
+	page_info->pagecnt_bias = INT_MAX;
+}
+
+static int gve_rx_alloc_buffer(struct gve_priv *priv, struct device *dev,
+			       struct gve_rx_slot_page_info *page_info,
+			       union gve_rx_data_slot *data_slot,
+			       struct gve_rx_ring *rx)
+{
+	struct page *page;
+	dma_addr_t dma;
+	int err;
+
+	err = gve_alloc_page(priv, dev, &page, &dma, DMA_FROM_DEVICE,
+			     GFP_ATOMIC);
+	if (err) {
+		u64_stats_update_begin(&rx->statss);
+		rx->rx_buf_alloc_fail++;
+		u64_stats_update_end(&rx->statss);
+		return err;
+	}
+
+	gve_setup_rx_buffer(rx, page_info, dma, page, &data_slot->addr);
+	return 0;
+}
+
+static int gve_rx_prefill_pages(struct gve_rx_ring *rx,
+				struct gve_rx_alloc_rings_cfg *cfg)
+{
+	struct gve_priv *priv = rx->gve;
+	u32 slots;
+	int err;
+	int i;
+	int j;
+
+	/* Allocate one page per Rx queue slot. Each page is split into two
+	 * packet buffers, when possible we "page flip" between the two.
+	 */
+	slots = rx->mask + 1;
+
+	rx->data.page_info = kvcalloc_node(slots, sizeof(*rx->data.page_info),
+					   GFP_KERNEL, priv->numa_node);
+	if (!rx->data.page_info)
+		return -ENOMEM;
+
+	for (i = 0; i < slots; i++) {
+		if (!rx->data.raw_addressing) {
+			struct page *page = rx->data.qpl->pages[i];
+			dma_addr_t addr = i * PAGE_SIZE;
+
+			gve_setup_rx_buffer(rx, &rx->data.page_info[i], addr,
+					    page,
+					    &rx->data.data_ring[i].qpl_offset);
+			continue;
+		}
+		err = gve_rx_alloc_buffer(priv, &priv->pdev->dev,
+					  &rx->data.page_info[i],
+					  &rx->data.data_ring[i], rx);
+		if (err)
+			goto alloc_err_rda;
+	}
+
+	if (!rx->data.raw_addressing) {
+		for (j = 0; j < rx->qpl_copy_pool_mask + 1; j++) {
+			struct page *page = alloc_pages_node(priv->numa_node,
+							     GFP_KERNEL, 0);
+
+			if (!page) {
+				err = -ENOMEM;
+				goto alloc_err_qpl;
+			}
+
+			rx->qpl_copy_pool[j].page = page;
+			rx->qpl_copy_pool[j].page_offset = 0;
+			rx->qpl_copy_pool[j].page_address = page_address(page);
+			rx->qpl_copy_pool[j].buf_size = rx->packet_buffer_size;
+
+			/* The page already has 1 ref. */
+			page_ref_add(page, INT_MAX - 1);
+			rx->qpl_copy_pool[j].pagecnt_bias = INT_MAX;
+		}
+	}
+
+	return slots;
+
+alloc_err_qpl:
+	/* Fully free the copy pool pages. */
+	while (j--) {
+		page_ref_sub(rx->qpl_copy_pool[j].page,
+			     rx->qpl_copy_pool[j].pagecnt_bias - 1);
+		put_page(rx->qpl_copy_pool[j].page);
+	}
+
+	/* Do not fully free QPL pages - only remove the bias added in this
+	 * function with gve_setup_rx_buffer.
+	 */
+	while (i--)
+		page_ref_sub(rx->data.page_info[i].page,
+			     rx->data.page_info[i].pagecnt_bias - 1);
+
+	return err;
+
+alloc_err_rda:
+	while (i--)
+		gve_rx_free_buffer(&priv->pdev->dev,
+				   &rx->data.page_info[i],
+				   &rx->data.data_ring[i]);
+	return err;
+}
+
+void gve_rx_start_ring_gqi(struct gve_priv *priv, int idx)
+{
+	int ntfy_idx = gve_rx_idx_to_ntfy(priv, idx);
+
+	gve_rx_add_to_block(priv, idx);
+	gve_add_napi(priv, ntfy_idx, gve_napi_poll);
+}
+
+int gve_rx_alloc_ring_gqi(struct gve_priv *priv,
+			  struct gve_rx_alloc_rings_cfg *cfg,
+			  struct gve_rx_ring *rx,
+			  int idx)
+{
+	struct device *hdev = &priv->pdev->dev;
+	u32 slots = cfg->ring_size;
+	int filled_pages;
+	int qpl_page_cnt;
+	u32 qpl_id = 0;
+	size_t bytes;
+	int err;
+
+	netif_dbg(priv, drv, priv->dev, "allocating rx ring\n");
+	/* Make sure everything is zeroed to start with */
+	memset(rx, 0, sizeof(*rx));
+
+	rx->gve = priv;
+	rx->q_num = idx;
+	rx->packet_buffer_size = cfg->packet_buffer_size;
+
+	rx->mask = slots - 1;
+	rx->data.raw_addressing = cfg->raw_addressing;
+
+	/* alloc rx data ring */
+	bytes = sizeof(*rx->data.data_ring) * slots;
+	rx->data.data_ring = dma_alloc_coherent(hdev, bytes,
+						&rx->data.data_bus,
+						GFP_KERNEL);
+	if (!rx->data.data_ring)
+		return -ENOMEM;
+
+	rx->qpl_copy_pool_mask = min_t(u32, U32_MAX, slots * 2) - 1;
+	rx->qpl_copy_pool_head = 0;
+	rx->qpl_copy_pool = kvcalloc_node(rx->qpl_copy_pool_mask + 1,
+					  sizeof(rx->qpl_copy_pool[0]),
+					  GFP_KERNEL, priv->numa_node);
+	if (!rx->qpl_copy_pool) {
+		err = -ENOMEM;
+		goto abort_with_slots;
+	}
+
+	if (!rx->data.raw_addressing) {
+		qpl_id = gve_get_rx_qpl_id(cfg->qcfg_tx, rx->q_num);
+		qpl_page_cnt = cfg->ring_size;
+
+		rx->data.qpl = gve_alloc_queue_page_list(priv, qpl_id,
+							 qpl_page_cnt);
+		if (!rx->data.qpl) {
+			err = -ENOMEM;
+			goto abort_with_copy_pool;
+		}
+	}
+
+	filled_pages = gve_rx_prefill_pages(rx, cfg);
+	if (filled_pages < 0) {
+		err = -ENOMEM;
+		goto abort_with_qpl;
+	}
+	rx->fill_cnt = filled_pages;
+	/* Ensure data ring slots (packet buffers) are visible. */
+	dma_wmb();
+
+	/* Alloc gve_queue_resources */
+	rx->q_resources =
+		dma_alloc_coherent(hdev,
+				   sizeof(*rx->q_resources),
+				   &rx->q_resources_bus,
+				   GFP_KERNEL);
+	if (!rx->q_resources) {
+		err = -ENOMEM;
+		goto abort_filled;
+	}
+	netif_dbg(priv, drv, priv->dev, "rx[%d]->data.data_bus=%lx\n", idx,
+		  (unsigned long)rx->data.data_bus);
+
+	/* alloc rx desc ring */
+	bytes = sizeof(struct gve_rx_desc) * cfg->ring_size;
+	rx->desc.desc_ring = dma_alloc_coherent(hdev, bytes, &rx->desc.bus,
+						GFP_KERNEL);
+	if (!rx->desc.desc_ring) {
+		err = -ENOMEM;
+		goto abort_with_q_resources;
+	}
+	rx->db_threshold = slots / 2;
+	gve_rx_init_ring_state_gqi(rx);
+
+	gve_rx_ctx_clear(&rx->ctx);
+
+	return 0;
+
+abort_with_q_resources:
+	dma_free_coherent(hdev, sizeof(*rx->q_resources),
+			  rx->q_resources, rx->q_resources_bus);
+	rx->q_resources = NULL;
+abort_filled:
+	gve_rx_unfill_pages(priv, rx, cfg);
+abort_with_qpl:
+	if (!rx->data.raw_addressing) {
+		gve_free_queue_page_list(priv, rx->data.qpl, qpl_id);
+		rx->data.qpl = NULL;
+	}
+abort_with_copy_pool:
+	kvfree(rx->qpl_copy_pool);
+	rx->qpl_copy_pool = NULL;
+abort_with_slots:
+	bytes = sizeof(*rx->data.data_ring) * slots;
+	dma_free_coherent(hdev, bytes, rx->data.data_ring, rx->data.data_bus);
+	rx->data.data_ring = NULL;
+
+	return err;
+}
+
+int gve_rx_alloc_rings_gqi(struct gve_priv *priv,
+			   struct gve_rx_alloc_rings_cfg *cfg)
+{
+	struct gve_rx_ring *rx;
+	int err = 0;
+	int i, j;
+
+	rx = kvcalloc(cfg->qcfg_rx->max_queues, sizeof(struct gve_rx_ring),
+		      GFP_KERNEL);
+	if (!rx)
+		return -ENOMEM;
+
+	for (i = 0; i < cfg->qcfg_rx->num_queues; i++) {
+		err = gve_rx_alloc_ring_gqi(priv, cfg, &rx[i], i);
+		if (err) {
+			netif_err(priv, drv, priv->dev,
+				  "Failed to alloc rx ring=%d: err=%d\n",
+				  i, err);
+			goto cleanup;
+		}
+	}
+
+	cfg->rx = rx;
+	return 0;
+
+cleanup:
+	for (j = 0; j < i; j++)
+		gve_rx_free_ring_gqi(priv, &rx[j], cfg);
+	kvfree(rx);
+	return err;
+}
+
+void gve_rx_free_rings_gqi(struct gve_priv *priv,
+			   struct gve_rx_alloc_rings_cfg *cfg)
+{
+	struct gve_rx_ring *rx = cfg->rx;
+	int i;
+
+	if (!rx)
+		return;
+
+	for (i = 0; i < cfg->qcfg_rx->num_queues;  i++)
+		gve_rx_free_ring_gqi(priv, &rx[i], cfg);
+
+	kvfree(rx);
+	cfg->rx = NULL;
+}
+
+void gve_rx_write_doorbell(struct gve_priv *priv, struct gve_rx_ring *rx)
+{
+	u32 db_idx = be32_to_cpu(rx->q_resources->db_index);
+
+	iowrite32be(rx->fill_cnt, &priv->db_bar2[db_idx]);
+}
+
+static enum pkt_hash_types gve_rss_type(__be16 pkt_flags)
+{
+	if (likely(pkt_flags & (GVE_RXF_TCP | GVE_RXF_UDP)))
+		return PKT_HASH_TYPE_L4;
+	if (pkt_flags & (GVE_RXF_IPV4 | GVE_RXF_IPV6))
+		return PKT_HASH_TYPE_L3;
+	return PKT_HASH_TYPE_L2;
+}
+
+static struct sk_buff *gve_rx_add_frags(struct napi_struct *napi,
+					struct gve_rx_slot_page_info *page_info,
+					unsigned int truesize, u16 len,
+					struct gve_rx_ctx *ctx)
+{
+	u32 offset = page_info->page_offset + page_info->pad;
+	struct sk_buff *skb = ctx->skb_tail;
+	int num_frags = 0;
+
+	if (!skb) {
+		skb = napi_get_frags(napi);
+		if (unlikely(!skb))
+			return NULL;
+
+		ctx->skb_head = skb;
+		ctx->skb_tail = skb;
+	} else {
+		num_frags = skb_shinfo(ctx->skb_tail)->nr_frags;
+		if (num_frags == MAX_SKB_FRAGS) {
+			skb = napi_alloc_skb(napi, 0);
+			if (!skb)
+				return NULL;
+
+			// We will never chain more than two SKBs: 2 * 16 * 2k > 64k
+			// which is why we do not need to chain by using skb->next
+			skb_shinfo(ctx->skb_tail)->frag_list = skb;
+
+			ctx->skb_tail = skb;
+			num_frags = 0;
+		}
+	}
+
+	if (skb != ctx->skb_head) {
+		ctx->skb_head->len += len;
+		ctx->skb_head->data_len += len;
+		ctx->skb_head->truesize += truesize;
+	}
+	skb_add_rx_frag(skb, num_frags, page_info->page,
+			offset, len, truesize);
+
+	return ctx->skb_head;
+}
+
+static void gve_rx_flip_buff(struct gve_rx_slot_page_info *page_info, __be64 *slot_addr)
+{
+	const __be64 offset = cpu_to_be64(GVE_DEFAULT_RX_BUFFER_OFFSET);
+
+	/* "flip" to other packet buffer on this page */
+	page_info->page_offset ^= GVE_DEFAULT_RX_BUFFER_OFFSET;
+	*(slot_addr) ^= offset;
+}
+
+static int gve_rx_can_recycle_buffer(struct gve_rx_slot_page_info *page_info)
+{
+	int pagecount = page_count(page_info->page);
+
+	/* This page is not being used by any SKBs - reuse */
+	if (pagecount == page_info->pagecnt_bias)
+		return 1;
+	/* This page is still being used by an SKB - we can't reuse */
+	else if (pagecount > page_info->pagecnt_bias)
+		return 0;
+	WARN(pagecount < page_info->pagecnt_bias,
+	     "Pagecount should never be less than the bias.");
+	return -1;
+}
+
+static struct sk_buff *
+gve_rx_raw_addressing(struct device *dev, struct net_device *netdev,
+		      struct gve_rx_slot_page_info *page_info, u16 len,
+		      struct napi_struct *napi,
+		      union gve_rx_data_slot *data_slot,
+		      u16 packet_buffer_size, struct gve_rx_ctx *ctx)
+{
+	struct sk_buff *skb = gve_rx_add_frags(napi, page_info, packet_buffer_size, len, ctx);
+
+	if (!skb)
+		return NULL;
+
+	/* Optimistically stop the kernel from freeing the page.
+	 * We will check again in refill to determine if we need to alloc a
+	 * new page.
+	 */
+	gve_dec_pagecnt_bias(page_info);
+
+	return skb;
+}
+
+static struct sk_buff *gve_rx_copy_to_pool(struct gve_rx_ring *rx,
+					   struct gve_rx_slot_page_info *page_info,
+					   u16 len, struct napi_struct *napi)
+{
+	u32 pool_idx = rx->qpl_copy_pool_head & rx->qpl_copy_pool_mask;
+	void *src = page_info->page_address + page_info->page_offset;
+	struct gve_rx_slot_page_info *copy_page_info;
+	struct gve_rx_ctx *ctx = &rx->ctx;
+	bool alloc_page = false;
+	struct sk_buff *skb;
+	void *dst;
+
+	copy_page_info = &rx->qpl_copy_pool[pool_idx];
+	if (!copy_page_info->can_flip) {
+		int recycle = gve_rx_can_recycle_buffer(copy_page_info);
+
+		if (unlikely(recycle < 0)) {
+			gve_schedule_reset(rx->gve);
+			return NULL;
+		}
+		alloc_page = !recycle;
+	}
+
+	if (alloc_page) {
+		struct gve_rx_slot_page_info alloc_page_info;
+		struct page *page;
+
+		/* The least recently used page turned out to be
+		 * still in use by the kernel. Ignoring it and moving
+		 * on alleviates head-of-line blocking.
+		 */
+		rx->qpl_copy_pool_head++;
+
+		page = alloc_page(GFP_ATOMIC);
+		if (!page)
+			return NULL;
+
+		alloc_page_info.page = page;
+		alloc_page_info.page_offset = 0;
+		alloc_page_info.page_address = page_address(page);
+		alloc_page_info.pad = page_info->pad;
+
+		memcpy(alloc_page_info.page_address, src, page_info->pad + len);
+		skb = gve_rx_add_frags(napi, &alloc_page_info,
+				       PAGE_SIZE,
+				       len, ctx);
+
+		u64_stats_update_begin(&rx->statss);
+		rx->rx_frag_copy_cnt++;
+		rx->rx_frag_alloc_cnt++;
+		u64_stats_update_end(&rx->statss);
+
+		return skb;
+	}
+
+	dst = copy_page_info->page_address + copy_page_info->page_offset;
+	memcpy(dst, src, page_info->pad + len);
+	copy_page_info->pad = page_info->pad;
+
+	skb = gve_rx_add_frags(napi, copy_page_info,
+			       copy_page_info->buf_size, len, ctx);
+	if (unlikely(!skb))
+		return NULL;
+
+	gve_dec_pagecnt_bias(copy_page_info);
+	copy_page_info->page_offset ^= GVE_DEFAULT_RX_BUFFER_OFFSET;
+
+	if (copy_page_info->can_flip) {
+		/* We have used both halves of this copy page, it
+		 * is time for it to go to the back of the queue.
+		 */
+		copy_page_info->can_flip = false;
+		rx->qpl_copy_pool_head++;
+		prefetch(rx->qpl_copy_pool[rx->qpl_copy_pool_head & rx->qpl_copy_pool_mask].page);
+	} else {
+		copy_page_info->can_flip = true;
+	}
+
+	u64_stats_update_begin(&rx->statss);
+	rx->rx_frag_copy_cnt++;
+	u64_stats_update_end(&rx->statss);
+
+	return skb;
+}
+
+static struct sk_buff *
+gve_rx_qpl(struct device *dev, struct net_device *netdev,
+	   struct gve_rx_ring *rx, struct gve_rx_slot_page_info *page_info,
+	   u16 len, struct napi_struct *napi,
+	   union gve_rx_data_slot *data_slot)
+{
+	struct gve_rx_ctx *ctx = &rx->ctx;
+	struct sk_buff *skb;
+
+	/* if raw_addressing mode is not enabled gvnic can only receive into
+	 * registered segments. If the buffer can't be recycled, our only
+	 * choice is to copy the data out of it so that we can return it to the
+	 * device.
+	 */
+	if (page_info->can_flip) {
+		skb = gve_rx_add_frags(napi, page_info, page_info->buf_size,
+				       len, ctx);
+		/* No point in recycling if we didn't get the skb */
+		if (skb) {
+			/* Make sure that the page isn't freed. */
+			gve_dec_pagecnt_bias(page_info);
+			gve_rx_flip_buff(page_info, &data_slot->qpl_offset);
+		}
+	} else {
+		skb = gve_rx_copy_to_pool(rx, page_info, len, napi);
+	}
+	return skb;
+}
+
+static struct sk_buff *gve_rx_skb(struct gve_priv *priv, struct gve_rx_ring *rx,
+				  struct gve_rx_slot_page_info *page_info, struct napi_struct *napi,
+				  u16 len, union gve_rx_data_slot *data_slot,
+				  bool is_only_frag)
+{
+	struct net_device *netdev = priv->dev;
+	struct gve_rx_ctx *ctx = &rx->ctx;
+	struct sk_buff *skb = NULL;
+
+	if (len <= priv->rx_copybreak && is_only_frag)  {
+		/* Just copy small packets */
+		skb = gve_rx_copy(netdev, napi, page_info, len);
+		if (skb) {
+			u64_stats_update_begin(&rx->statss);
+			rx->rx_copied_pkt++;
+			rx->rx_frag_copy_cnt++;
+			rx->rx_copybreak_pkt++;
+			u64_stats_update_end(&rx->statss);
+		}
+	} else {
+		int recycle = gve_rx_can_recycle_buffer(page_info);
+
+		if (unlikely(recycle < 0)) {
+			gve_schedule_reset(priv);
+			return NULL;
+		}
+		page_info->can_flip = recycle;
+		if (page_info->can_flip) {
+			u64_stats_update_begin(&rx->statss);
+			rx->rx_frag_flip_cnt++;
+			u64_stats_update_end(&rx->statss);
+		}
+
+		if (rx->data.raw_addressing) {
+			skb = gve_rx_raw_addressing(&priv->pdev->dev, netdev,
+						    page_info, len, napi,
+						    data_slot,
+						    page_info->buf_size, ctx);
+		} else {
+			skb = gve_rx_qpl(&priv->pdev->dev, netdev, rx,
+					 page_info, len, napi, data_slot);
+		}
+	}
+	return skb;
+}
+
+static int gve_xsk_pool_redirect(struct net_device *dev,
+				 struct gve_rx_ring *rx,
+				 void *data, int len,
+				 struct bpf_prog *xdp_prog)
+{
+	struct xdp_buff *xdp;
+	int err;
+
+	if (rx->xsk_pool->frame_len < len)
+		return -E2BIG;
+	xdp = xsk_buff_alloc(rx->xsk_pool);
+	if (!xdp) {
+		u64_stats_update_begin(&rx->statss);
+		rx->xdp_alloc_fails++;
+		u64_stats_update_end(&rx->statss);
+		return -ENOMEM;
+	}
+	xdp->data_end = xdp->data + len;
+	memcpy(xdp->data, data, len);
+	err = xdp_do_redirect(dev, xdp, xdp_prog);
+	if (err)
+		xsk_buff_free(xdp);
+	return err;
+}
+
+static int gve_xdp_redirect(struct net_device *dev, struct gve_rx_ring *rx,
+			    struct xdp_buff *orig, struct bpf_prog *xdp_prog)
+{
+	int total_len, len = orig->data_end - orig->data;
+	int headroom = XDP_PACKET_HEADROOM;
+	struct xdp_buff new;
+	void *frame;
+	int err;
+
+	if (rx->xsk_pool)
+		return gve_xsk_pool_redirect(dev, rx, orig->data,
+					     len, xdp_prog);
+
+	total_len = headroom + SKB_DATA_ALIGN(len) +
+		SKB_DATA_ALIGN(sizeof(struct skb_shared_info));
+	frame = page_frag_alloc(&rx->page_cache, total_len, GFP_ATOMIC);
+	if (!frame) {
+		u64_stats_update_begin(&rx->statss);
+		rx->xdp_alloc_fails++;
+		u64_stats_update_end(&rx->statss);
+		return -ENOMEM;
+	}
+	xdp_init_buff(&new, total_len, &rx->xdp_rxq);
+	xdp_prepare_buff(&new, frame, headroom, len, false);
+	memcpy(new.data, orig->data, len);
+
+	err = xdp_do_redirect(dev, &new, xdp_prog);
+	if (err)
+		page_frag_free(frame);
+
+	return err;
+}
+
+static void gve_xdp_done(struct gve_priv *priv, struct gve_rx_ring *rx,
+			 struct xdp_buff *xdp, struct bpf_prog *xprog,
+			 int xdp_act)
+{
+	struct gve_tx_ring *tx;
+	int tx_qid;
+	int err;
+
+	switch (xdp_act) {
+	case XDP_ABORTED:
+	case XDP_DROP:
+	default:
+		break;
+	case XDP_TX:
+		tx_qid = gve_xdp_tx_queue_id(priv, rx->q_num);
+		tx = &priv->tx[tx_qid];
+		spin_lock(&tx->xdp_lock);
+		err = gve_xdp_xmit_one(priv, tx, xdp->data,
+				       xdp->data_end - xdp->data, NULL);
+		spin_unlock(&tx->xdp_lock);
+
+		if (unlikely(err)) {
+			u64_stats_update_begin(&rx->statss);
+			rx->xdp_tx_errors++;
+			u64_stats_update_end(&rx->statss);
+		}
+		break;
+	case XDP_REDIRECT:
+		err = gve_xdp_redirect(priv->dev, rx, xdp, xprog);
+
+		if (unlikely(err)) {
+			u64_stats_update_begin(&rx->statss);
+			rx->xdp_redirect_errors++;
+			u64_stats_update_end(&rx->statss);
+		}
+		break;
+	}
+	u64_stats_update_begin(&rx->statss);
+	if ((u32)xdp_act < GVE_XDP_ACTIONS)
+		rx->xdp_actions[xdp_act]++;
+	u64_stats_update_end(&rx->statss);
+}
+
+#define GVE_PKTCONT_BIT_IS_SET(x) (GVE_RXF_PKT_CONT & (x))
+static void gve_rx(struct gve_rx_ring *rx, netdev_features_t feat,
+		   struct gve_rx_desc *desc, u32 idx,
+		   struct gve_rx_cnts *cnts)
+{
+	bool is_last_frag = !GVE_PKTCONT_BIT_IS_SET(desc->flags_seq);
+	struct gve_rx_slot_page_info *page_info;
+	u16 frag_size = be16_to_cpu(desc->len);
+	struct gve_rx_ctx *ctx = &rx->ctx;
+	union gve_rx_data_slot *data_slot;
+	struct gve_priv *priv = rx->gve;
+	struct sk_buff *skb = NULL;
+	struct bpf_prog *xprog;
+	struct xdp_buff xdp;
+	dma_addr_t page_bus;
+	void *va;
+
+	u16 len = frag_size;
+	struct napi_struct *napi = &priv->ntfy_blocks[rx->ntfy_id].napi;
+	bool is_first_frag = ctx->frag_cnt == 0;
+
+	bool is_only_frag = is_first_frag && is_last_frag;
+
+	if (unlikely(ctx->drop_pkt))
+		goto finish_frag;
+
+	if (desc->flags_seq & GVE_RXF_ERR) {
+		ctx->drop_pkt = true;
+		cnts->desc_err_pkt_cnt++;
+		napi_free_frags(napi);
+		goto finish_frag;
+	}
+
+	if (unlikely(frag_size > rx->packet_buffer_size)) {
+		netdev_warn(priv->dev, "Unexpected frag size %d, can't exceed %d, scheduling reset",
+			    frag_size, rx->packet_buffer_size);
+		ctx->drop_pkt = true;
+		napi_free_frags(napi);
+		gve_schedule_reset(rx->gve);
+		goto finish_frag;
+	}
+
+	/* Prefetch two packet buffers ahead, we will need it soon. */
+	page_info = &rx->data.page_info[(idx + 2) & rx->mask];
+	va = page_info->page_address + page_info->page_offset;
+	prefetch(page_info->page); /* Kernel page struct. */
+	prefetch(va);              /* Packet header. */
+	prefetch(va + 64);         /* Next cacheline too. */
+
+	page_info = &rx->data.page_info[idx];
+	data_slot = &rx->data.data_ring[idx];
+	page_bus = (rx->data.raw_addressing) ?
+		be64_to_cpu(data_slot->addr) - page_info->page_offset :
+		rx->data.qpl->page_buses[idx];
+	dma_sync_single_for_cpu(&priv->pdev->dev, page_bus,
+				PAGE_SIZE, DMA_FROM_DEVICE);
+	page_info->pad = is_first_frag ? GVE_RX_PAD : 0;
+	len -= page_info->pad;
+	frag_size -= page_info->pad;
+
+	xprog = READ_ONCE(priv->xdp_prog);
+	if (xprog && is_only_frag) {
+		void *old_data;
+		int xdp_act;
+
+		xdp_init_buff(&xdp, page_info->buf_size, &rx->xdp_rxq);
+		xdp_prepare_buff(&xdp, page_info->page_address +
+				 page_info->page_offset, GVE_RX_PAD,
+				 len, false);
+		old_data = xdp.data;
+		xdp_act = bpf_prog_run_xdp(xprog, &xdp);
+		if (xdp_act != XDP_PASS) {
+			gve_xdp_done(priv, rx, &xdp, xprog, xdp_act);
+			ctx->total_size += frag_size;
+			goto finish_ok_pkt;
+		}
+
+		page_info->pad += xdp.data - old_data;
+		len = xdp.data_end - xdp.data;
+
+		u64_stats_update_begin(&rx->statss);
+		rx->xdp_actions[XDP_PASS]++;
+		u64_stats_update_end(&rx->statss);
+	}
+
+	skb = gve_rx_skb(priv, rx, page_info, napi, len,
+			 data_slot, is_only_frag);
+	if (!skb) {
+		u64_stats_update_begin(&rx->statss);
+		rx->rx_skb_alloc_fail++;
+		u64_stats_update_end(&rx->statss);
+
+		napi_free_frags(napi);
+		ctx->drop_pkt = true;
+		goto finish_frag;
+	}
+	ctx->total_size += frag_size;
+
+	if (is_first_frag) {
+		if (likely(feat & NETIF_F_RXCSUM)) {
+			/* NIC passes up the partial sum */
+			if (desc->csum)
+				skb->ip_summed = CHECKSUM_COMPLETE;
+			else
+				skb->ip_summed = CHECKSUM_NONE;
+			skb->csum = csum_unfold(desc->csum);
+		}
+
+		/* parse flags & pass relevant info up */
+		if (likely(feat & NETIF_F_RXHASH) &&
+		    gve_needs_rss(desc->flags_seq))
+			skb_set_hash(skb, be32_to_cpu(desc->rss_hash),
+				     gve_rss_type(desc->flags_seq));
+	}
+
+	if (is_last_frag) {
+		skb_record_rx_queue(skb, rx->q_num);
+		if (skb_is_nonlinear(skb))
+			napi_gro_frags(napi);
+		else
+			napi_gro_receive(napi, skb);
+		goto finish_ok_pkt;
+	}
+
+	goto finish_frag;
+
+finish_ok_pkt:
+	cnts->ok_pkt_bytes += ctx->total_size;
+	cnts->ok_pkt_cnt++;
+finish_frag:
+	ctx->frag_cnt++;
+	if (is_last_frag) {
+		cnts->total_pkt_cnt++;
+		cnts->cont_pkt_cnt += (ctx->frag_cnt > 1);
+		gve_rx_ctx_clear(ctx);
+	}
+}
+
+bool gve_rx_work_pending(struct gve_rx_ring *rx)
+{
+	struct gve_rx_desc *desc;
+	__be16 flags_seq;
+	u32 next_idx;
+
+	next_idx = rx->cnt & rx->mask;
+	desc = rx->desc.desc_ring + next_idx;
+
+	flags_seq = desc->flags_seq;
+
+	return (GVE_SEQNO(flags_seq) == rx->desc.seqno);
+}
+
+static bool gve_rx_refill_buffers(struct gve_priv *priv, struct gve_rx_ring *rx)
+{
+	int refill_target = rx->mask + 1;
+	u32 fill_cnt = rx->fill_cnt;
+
+	while (fill_cnt - rx->cnt < refill_target) {
+		struct gve_rx_slot_page_info *page_info;
+		u32 idx = fill_cnt & rx->mask;
+
+		page_info = &rx->data.page_info[idx];
+		if (page_info->can_flip) {
+			/* The other half of the page is free because it was
+			 * free when we processed the descriptor. Flip to it.
+			 */
+			union gve_rx_data_slot *data_slot =
+						&rx->data.data_ring[idx];
+
+			gve_rx_flip_buff(page_info, &data_slot->addr);
+			page_info->can_flip = 0;
+		} else {
+			/* It is possible that the networking stack has already
+			 * finished processing all outstanding packets in the buffer
+			 * and it can be reused.
+			 * Flipping is unnecessary here - if the networking stack still
+			 * owns half the page it is impossible to tell which half. Either
+			 * the whole page is free or it needs to be replaced.
+			 */
+			int recycle = gve_rx_can_recycle_buffer(page_info);
+
+			if (recycle < 0) {
+				if (!rx->data.raw_addressing)
+					gve_schedule_reset(priv);
+				return false;
+			}
+			if (!recycle) {
+				/* We can't reuse the buffer - alloc a new one*/
+				union gve_rx_data_slot *data_slot =
+						&rx->data.data_ring[idx];
+				struct device *dev = &priv->pdev->dev;
+				gve_rx_free_buffer(dev, page_info, data_slot);
+				page_info->page = NULL;
+				if (gve_rx_alloc_buffer(priv, dev, page_info,
+							data_slot, rx)) {
+					break;
+				}
+			}
+		}
+		fill_cnt++;
+	}
+	rx->fill_cnt = fill_cnt;
+	return true;
+}
+
+static int gve_clean_rx_done(struct gve_rx_ring *rx, int budget,
+			     netdev_features_t feat)
+{
+	u64 xdp_redirects = rx->xdp_actions[XDP_REDIRECT];
+	u64 xdp_txs = rx->xdp_actions[XDP_TX];
+	struct gve_rx_ctx *ctx = &rx->ctx;
+	struct gve_priv *priv = rx->gve;
+	struct gve_rx_cnts cnts = {0};
+	struct gve_rx_desc *next_desc;
+	u32 idx = rx->cnt & rx->mask;
+	u32 work_done = 0;
+
+	struct gve_rx_desc *desc = &rx->desc.desc_ring[idx];
+
+	// Exceed budget only if (and till) the inflight packet is consumed.
+	while ((GVE_SEQNO(desc->flags_seq) == rx->desc.seqno) &&
+	       (work_done < budget || ctx->frag_cnt)) {
+		next_desc = &rx->desc.desc_ring[(idx + 1) & rx->mask];
+		prefetch(next_desc);
+
+		gve_rx(rx, feat, desc, idx, &cnts);
+
+		rx->cnt++;
+		idx = rx->cnt & rx->mask;
+		desc = &rx->desc.desc_ring[idx];
+		rx->desc.seqno = gve_next_seqno(rx->desc.seqno);
+		work_done++;
+	}
+
+	// The device will only send whole packets.
+	if (unlikely(ctx->frag_cnt)) {
+		struct napi_struct *napi = &priv->ntfy_blocks[rx->ntfy_id].napi;
+
+		napi_free_frags(napi);
+		gve_rx_ctx_clear(&rx->ctx);
+		netdev_warn(priv->dev, "Unexpected seq number %d with incomplete packet, expected %d, scheduling reset",
+			    GVE_SEQNO(desc->flags_seq), rx->desc.seqno);
+		gve_schedule_reset(rx->gve);
+	}
+
+	if (!work_done && rx->fill_cnt - rx->cnt > rx->db_threshold)
+		return 0;
+
+	if (work_done) {
+		u64_stats_update_begin(&rx->statss);
+		rx->rpackets += cnts.ok_pkt_cnt;
+		rx->rbytes += cnts.ok_pkt_bytes;
+		rx->rx_cont_packet_cnt += cnts.cont_pkt_cnt;
+		rx->rx_desc_err_dropped_pkt += cnts.desc_err_pkt_cnt;
+		u64_stats_update_end(&rx->statss);
+	}
+
+	if (xdp_txs != rx->xdp_actions[XDP_TX])
+		gve_xdp_tx_flush(priv, rx->q_num);
+
+	if (xdp_redirects != rx->xdp_actions[XDP_REDIRECT])
+		xdp_do_flush();
+
+	/* restock ring slots */
+	if (!rx->data.raw_addressing) {
+		/* In QPL mode buffs are refilled as the desc are processed */
+		rx->fill_cnt += work_done;
+	} else if (rx->fill_cnt - rx->cnt <= rx->db_threshold) {
+		/* In raw addressing mode buffs are only refilled if the avail
+		 * falls below a threshold.
+		 */
+		if (!gve_rx_refill_buffers(priv, rx))
+			return 0;
+
+		/* If we were not able to completely refill buffers, we'll want
+		 * to schedule this queue for work again to refill buffers.
+		 */
+		if (rx->fill_cnt - rx->cnt <= rx->db_threshold) {
+			gve_rx_write_doorbell(priv, rx);
+			return budget;
+		}
+	}
+
+	gve_rx_write_doorbell(priv, rx);
+	return cnts.total_pkt_cnt;
+}
+
+int gve_rx_poll(struct gve_notify_block *block, int budget)
+{
+	struct gve_rx_ring *rx = block->rx;
+	netdev_features_t feat;
+	int work_done = 0;
+
+	feat = block->napi.dev->features;
+
+	if (budget > 0)
+		work_done = gve_clean_rx_done(rx, budget, feat);
+
+	return work_done;
+}
diff --git a/drivers/net/ethernet/google/gve/gve_rx_dqo.c b/drivers/net/ethernet/google/gve/gve_rx_dqo.c
new file mode 100644
index 000000000000..55393b784317
--- /dev/null
+++ b/drivers/net/ethernet/google/gve/gve_rx_dqo.c
@@ -0,0 +1,1076 @@
+// SPDX-License-Identifier: (GPL-2.0 OR MIT)
+/* Google virtual Ethernet (gve) driver
+ *
+ * Copyright (C) 2015-2021 Google, Inc.
+ */
+
+#include "gve.h"
+#include "gve_dqo.h"
+#include "gve_adminq.h"
+#include "gve_utils.h"
+#include <linux/bpf.h>
+#include <linux/ip.h>
+#include <linux/ipv6.h>
+#include <linux/skbuff.h>
+#include <linux/slab.h>
+#include <net/ip6_checksum.h>
+#include <net/ipv6.h>
+#include <net/tcp.h>
+#include <net/xdp_sock_drv.h>
+
+static void gve_rx_free_hdr_bufs(struct gve_priv *priv, struct gve_rx_ring *rx)
+{
+	struct device *hdev = &priv->pdev->dev;
+	int buf_count = rx->dqo.bufq.mask + 1;
+
+	if (rx->dqo.hdr_bufs.data) {
+		dma_free_coherent(hdev, priv->header_buf_size * buf_count,
+				  rx->dqo.hdr_bufs.data, rx->dqo.hdr_bufs.addr);
+		rx->dqo.hdr_bufs.data = NULL;
+	}
+}
+
+static void gve_rx_init_ring_state_dqo(struct gve_rx_ring *rx,
+				       const u32 buffer_queue_slots,
+				       const u32 completion_queue_slots)
+{
+	int i;
+
+	/* Set buffer queue state */
+	rx->dqo.bufq.mask = buffer_queue_slots - 1;
+	rx->dqo.bufq.head = 0;
+	rx->dqo.bufq.tail = 0;
+
+	/* Set completion queue state */
+	rx->dqo.complq.num_free_slots = completion_queue_slots;
+	rx->dqo.complq.mask = completion_queue_slots - 1;
+	rx->dqo.complq.cur_gen_bit = 0;
+	rx->dqo.complq.head = 0;
+
+	/* Set RX SKB context */
+	rx->ctx.skb_head = NULL;
+	rx->ctx.skb_tail = NULL;
+
+	/* Set up linked list of buffer IDs */
+	if (rx->dqo.buf_states) {
+		for (i = 0; i < rx->dqo.num_buf_states - 1; i++)
+			rx->dqo.buf_states[i].next = i + 1;
+		rx->dqo.buf_states[rx->dqo.num_buf_states - 1].next = -1;
+	}
+
+	rx->dqo.free_buf_states = 0;
+	rx->dqo.recycled_buf_states.head = -1;
+	rx->dqo.recycled_buf_states.tail = -1;
+	rx->dqo.used_buf_states.head = -1;
+	rx->dqo.used_buf_states.tail = -1;
+}
+
+static void gve_rx_reset_ring_dqo(struct gve_priv *priv, int idx)
+{
+	struct gve_rx_ring *rx = &priv->rx[idx];
+	size_t size;
+	int i;
+
+	const u32 buffer_queue_slots = priv->rx_desc_cnt;
+	const u32 completion_queue_slots = priv->rx_desc_cnt;
+
+	/* Reset buffer queue */
+	if (rx->dqo.bufq.desc_ring) {
+		size = sizeof(rx->dqo.bufq.desc_ring[0]) *
+			buffer_queue_slots;
+		memset(rx->dqo.bufq.desc_ring, 0, size);
+	}
+
+	/* Reset completion queue */
+	if (rx->dqo.complq.desc_ring) {
+		size = sizeof(rx->dqo.complq.desc_ring[0]) *
+			completion_queue_slots;
+		memset(rx->dqo.complq.desc_ring, 0, size);
+	}
+
+	/* Reset q_resources */
+	if (rx->q_resources)
+		memset(rx->q_resources, 0, sizeof(*rx->q_resources));
+
+	/* Reset buf states */
+	if (rx->dqo.buf_states) {
+		for (i = 0; i < rx->dqo.num_buf_states; i++) {
+			struct gve_rx_buf_state_dqo *bs = &rx->dqo.buf_states[i];
+
+			if (rx->dqo.page_pool)
+				gve_free_to_page_pool(rx, bs, false);
+			else
+				gve_free_qpl_page_dqo(bs);
+		}
+	}
+
+	gve_rx_init_ring_state_dqo(rx, buffer_queue_slots,
+				   completion_queue_slots);
+}
+
+void gve_rx_stop_ring_dqo(struct gve_priv *priv, int idx)
+{
+	int ntfy_idx = gve_rx_idx_to_ntfy(priv, idx);
+	struct gve_rx_ring *rx = &priv->rx[idx];
+
+	if (!gve_rx_was_added_to_block(priv, idx))
+		return;
+
+	if (rx->dqo.page_pool)
+		page_pool_disable_direct_recycling(rx->dqo.page_pool);
+	gve_remove_napi(priv, ntfy_idx);
+	gve_rx_remove_from_block(priv, idx);
+	gve_rx_reset_ring_dqo(priv, idx);
+}
+
+void gve_rx_free_ring_dqo(struct gve_priv *priv, struct gve_rx_ring *rx,
+			  struct gve_rx_alloc_rings_cfg *cfg)
+{
+	struct device *hdev = &priv->pdev->dev;
+	size_t completion_queue_slots;
+	size_t buffer_queue_slots;
+	int idx = rx->q_num;
+	size_t size;
+	u32 qpl_id;
+	int i;
+
+	completion_queue_slots = rx->dqo.complq.mask + 1;
+	buffer_queue_slots = rx->dqo.bufq.mask + 1;
+
+	if (rx->q_resources) {
+		dma_free_coherent(hdev, sizeof(*rx->q_resources),
+				  rx->q_resources, rx->q_resources_bus);
+		rx->q_resources = NULL;
+	}
+
+	for (i = 0; i < rx->dqo.num_buf_states; i++) {
+		struct gve_rx_buf_state_dqo *bs = &rx->dqo.buf_states[i];
+
+		if (rx->dqo.page_pool)
+			gve_free_to_page_pool(rx, bs, false);
+		else
+			gve_free_qpl_page_dqo(bs);
+		if (gve_buf_state_is_allocated(rx, bs) && bs->xsk_buff) {
+			xsk_buff_free(bs->xsk_buff);
+			bs->xsk_buff = NULL;
+		}
+	}
+
+	if (rx->dqo.qpl) {
+		qpl_id = gve_get_rx_qpl_id(cfg->qcfg_tx, rx->q_num);
+		gve_free_queue_page_list(priv, rx->dqo.qpl, qpl_id);
+		rx->dqo.qpl = NULL;
+	}
+
+	if (rx->dqo.bufq.desc_ring) {
+		size = sizeof(rx->dqo.bufq.desc_ring[0]) * buffer_queue_slots;
+		dma_free_coherent(hdev, size, rx->dqo.bufq.desc_ring,
+				  rx->dqo.bufq.bus);
+		rx->dqo.bufq.desc_ring = NULL;
+	}
+
+	if (rx->dqo.complq.desc_ring) {
+		size = sizeof(rx->dqo.complq.desc_ring[0]) *
+			completion_queue_slots;
+		dma_free_coherent(hdev, size, rx->dqo.complq.desc_ring,
+				  rx->dqo.complq.bus);
+		rx->dqo.complq.desc_ring = NULL;
+	}
+
+	kvfree(rx->dqo.buf_states);
+	rx->dqo.buf_states = NULL;
+
+	if (rx->dqo.page_pool) {
+		page_pool_destroy(rx->dqo.page_pool);
+		rx->dqo.page_pool = NULL;
+	}
+
+	gve_rx_free_hdr_bufs(priv, rx);
+
+	netif_dbg(priv, drv, priv->dev, "freed rx ring %d\n", idx);
+}
+
+static int gve_rx_alloc_hdr_bufs(struct gve_priv *priv, struct gve_rx_ring *rx,
+				 const u32 buf_count)
+{
+	struct device *hdev = &priv->pdev->dev;
+
+	rx->dqo.hdr_bufs.data = dma_alloc_coherent(hdev, priv->header_buf_size * buf_count,
+						   &rx->dqo.hdr_bufs.addr, GFP_KERNEL);
+	if (!rx->dqo.hdr_bufs.data)
+		return -ENOMEM;
+
+	return 0;
+}
+
+void gve_rx_start_ring_dqo(struct gve_priv *priv, int idx)
+{
+	int ntfy_idx = gve_rx_idx_to_ntfy(priv, idx);
+
+	gve_rx_add_to_block(priv, idx);
+	gve_add_napi(priv, ntfy_idx, gve_napi_poll_dqo);
+}
+
+int gve_rx_alloc_ring_dqo(struct gve_priv *priv,
+			  struct gve_rx_alloc_rings_cfg *cfg,
+			  struct gve_rx_ring *rx,
+			  int idx)
+{
+	struct device *hdev = &priv->pdev->dev;
+	struct page_pool *pool;
+	int qpl_page_cnt;
+	size_t size;
+	u32 qpl_id;
+
+	const u32 buffer_queue_slots = cfg->ring_size;
+	const u32 completion_queue_slots = cfg->ring_size;
+
+	netif_dbg(priv, drv, priv->dev, "allocating rx ring DQO\n");
+
+	memset(rx, 0, sizeof(*rx));
+	rx->gve = priv;
+	rx->q_num = idx;
+	rx->packet_buffer_size = cfg->packet_buffer_size;
+
+	if (cfg->xdp) {
+		rx->packet_buffer_truesize = GVE_XDP_RX_BUFFER_SIZE_DQO;
+		rx->rx_headroom = XDP_PACKET_HEADROOM;
+	} else {
+		rx->packet_buffer_truesize = rx->packet_buffer_size;
+		rx->rx_headroom = 0;
+	}
+
+	rx->dqo.num_buf_states = cfg->raw_addressing ? buffer_queue_slots :
+		gve_get_rx_pages_per_qpl_dqo(cfg->ring_size);
+	rx->dqo.buf_states = kvcalloc_node(rx->dqo.num_buf_states,
+					   sizeof(rx->dqo.buf_states[0]),
+					   GFP_KERNEL, priv->numa_node);
+	if (!rx->dqo.buf_states)
+		return -ENOMEM;
+
+	/* Allocate header buffers for header-split */
+	if (cfg->enable_header_split)
+		if (gve_rx_alloc_hdr_bufs(priv, rx, buffer_queue_slots))
+			goto err;
+
+	/* Allocate RX completion queue */
+	size = sizeof(rx->dqo.complq.desc_ring[0]) *
+		completion_queue_slots;
+	rx->dqo.complq.desc_ring =
+		dma_alloc_coherent(hdev, size, &rx->dqo.complq.bus, GFP_KERNEL);
+	if (!rx->dqo.complq.desc_ring)
+		goto err;
+
+	/* Allocate RX buffer queue */
+	size = sizeof(rx->dqo.bufq.desc_ring[0]) * buffer_queue_slots;
+	rx->dqo.bufq.desc_ring =
+		dma_alloc_coherent(hdev, size, &rx->dqo.bufq.bus, GFP_KERNEL);
+	if (!rx->dqo.bufq.desc_ring)
+		goto err;
+
+	if (cfg->raw_addressing) {
+		pool = gve_rx_create_page_pool(priv, rx, cfg->xdp);
+		if (IS_ERR(pool))
+			goto err;
+
+		rx->dqo.page_pool = pool;
+	} else {
+		qpl_id = gve_get_rx_qpl_id(cfg->qcfg_tx, rx->q_num);
+		qpl_page_cnt = gve_get_rx_pages_per_qpl_dqo(cfg->ring_size);
+
+		rx->dqo.qpl = gve_alloc_queue_page_list(priv, qpl_id,
+							qpl_page_cnt);
+		if (!rx->dqo.qpl)
+			goto err;
+		rx->dqo.next_qpl_page_idx = 0;
+	}
+
+	rx->q_resources = dma_alloc_coherent(hdev, sizeof(*rx->q_resources),
+					     &rx->q_resources_bus, GFP_KERNEL);
+	if (!rx->q_resources)
+		goto err;
+
+	gve_rx_init_ring_state_dqo(rx, buffer_queue_slots,
+				   completion_queue_slots);
+
+	return 0;
+
+err:
+	gve_rx_free_ring_dqo(priv, rx, cfg);
+	return -ENOMEM;
+}
+
+void gve_rx_write_doorbell_dqo(const struct gve_priv *priv, int queue_idx)
+{
+	const struct gve_rx_ring *rx = &priv->rx[queue_idx];
+	u64 index = be32_to_cpu(rx->q_resources->db_index);
+
+	iowrite32(rx->dqo.bufq.tail, &priv->db_bar2[index]);
+}
+
+int gve_rx_alloc_rings_dqo(struct gve_priv *priv,
+			   struct gve_rx_alloc_rings_cfg *cfg)
+{
+	struct gve_rx_ring *rx;
+	int err;
+	int i;
+
+	rx = kvcalloc(cfg->qcfg_rx->max_queues, sizeof(struct gve_rx_ring),
+		      GFP_KERNEL);
+	if (!rx)
+		return -ENOMEM;
+
+	for (i = 0; i < cfg->qcfg_rx->num_queues; i++) {
+		err = gve_rx_alloc_ring_dqo(priv, cfg, &rx[i], i);
+		if (err) {
+			netif_err(priv, drv, priv->dev,
+				  "Failed to alloc rx ring=%d: err=%d\n",
+				  i, err);
+			goto err;
+		}
+	}
+
+	cfg->rx = rx;
+	return 0;
+
+err:
+	for (i--; i >= 0; i--)
+		gve_rx_free_ring_dqo(priv, &rx[i], cfg);
+	kvfree(rx);
+	return err;
+}
+
+void gve_rx_free_rings_dqo(struct gve_priv *priv,
+			   struct gve_rx_alloc_rings_cfg *cfg)
+{
+	struct gve_rx_ring *rx = cfg->rx;
+	int i;
+
+	if (!rx)
+		return;
+
+	for (i = 0; i < cfg->qcfg_rx->num_queues;  i++)
+		gve_rx_free_ring_dqo(priv, &rx[i], cfg);
+
+	kvfree(rx);
+	cfg->rx = NULL;
+}
+
+void gve_rx_post_buffers_dqo(struct gve_rx_ring *rx)
+{
+	struct gve_rx_compl_queue_dqo *complq = &rx->dqo.complq;
+	struct gve_rx_buf_queue_dqo *bufq = &rx->dqo.bufq;
+	struct gve_priv *priv = rx->gve;
+	u32 num_avail_slots;
+	u32 num_full_slots;
+	u32 num_posted = 0;
+
+	num_full_slots = (bufq->tail - bufq->head) & bufq->mask;
+	num_avail_slots = bufq->mask - num_full_slots;
+
+	num_avail_slots = min_t(u32, num_avail_slots, complq->num_free_slots);
+	while (num_posted < num_avail_slots) {
+		struct gve_rx_desc_dqo *desc = &bufq->desc_ring[bufq->tail];
+
+		if (unlikely(gve_alloc_buffer(rx, desc))) {
+			u64_stats_update_begin(&rx->statss);
+			rx->rx_buf_alloc_fail++;
+			u64_stats_update_end(&rx->statss);
+			break;
+		}
+
+		if (rx->dqo.hdr_bufs.data)
+			desc->header_buf_addr =
+				cpu_to_le64(rx->dqo.hdr_bufs.addr +
+					    priv->header_buf_size * bufq->tail);
+
+		bufq->tail = (bufq->tail + 1) & bufq->mask;
+		complq->num_free_slots--;
+		num_posted++;
+
+		if ((bufq->tail & (GVE_RX_BUF_THRESH_DQO - 1)) == 0)
+			gve_rx_write_doorbell_dqo(priv, rx->q_num);
+	}
+
+	rx->fill_cnt += num_posted;
+}
+
+static void gve_rx_skb_csum(struct sk_buff *skb,
+			    const struct gve_rx_compl_desc_dqo *desc,
+			    struct gve_ptype ptype)
+{
+	skb->ip_summed = CHECKSUM_NONE;
+
+	/* HW did not identify and process L3 and L4 headers. */
+	if (unlikely(!desc->l3_l4_processed))
+		return;
+
+	if (ptype.l3_type == GVE_L3_TYPE_IPV4) {
+		if (unlikely(desc->csum_ip_err || desc->csum_external_ip_err))
+			return;
+	} else if (ptype.l3_type == GVE_L3_TYPE_IPV6) {
+		/* Checksum should be skipped if this flag is set. */
+		if (unlikely(desc->ipv6_ex_add))
+			return;
+	}
+
+	if (unlikely(desc->csum_l4_err))
+		return;
+
+	switch (ptype.l4_type) {
+	case GVE_L4_TYPE_TCP:
+	case GVE_L4_TYPE_UDP:
+	case GVE_L4_TYPE_ICMP:
+	case GVE_L4_TYPE_SCTP:
+		skb->ip_summed = CHECKSUM_UNNECESSARY;
+		break;
+	default:
+		break;
+	}
+}
+
+static void gve_rx_skb_hash(struct sk_buff *skb,
+			    const struct gve_rx_compl_desc_dqo *compl_desc,
+			    struct gve_ptype ptype)
+{
+	enum pkt_hash_types hash_type = PKT_HASH_TYPE_L2;
+
+	if (ptype.l4_type != GVE_L4_TYPE_UNKNOWN)
+		hash_type = PKT_HASH_TYPE_L4;
+	else if (ptype.l3_type != GVE_L3_TYPE_UNKNOWN)
+		hash_type = PKT_HASH_TYPE_L3;
+
+	skb_set_hash(skb, le32_to_cpu(compl_desc->hash), hash_type);
+}
+
+/* Expand the hardware timestamp to the full 64 bits of width, and add it to the
+ * skb.
+ *
+ * This algorithm works by using the passed hardware timestamp to generate a
+ * diff relative to the last read of the nic clock. This diff can be positive or
+ * negative, as it is possible that we have read the clock more recently than
+ * the hardware has received this packet. To detect this, we use the high bit of
+ * the diff, and assume that the read is more recent if the high bit is set. In
+ * this case we invert the process.
+ *
+ * Note that this means if the time delta between packet reception and the last
+ * clock read is greater than ~2 seconds, this will provide invalid results.
+ */
+static void gve_rx_skb_hwtstamp(struct gve_rx_ring *rx, u32 hwts)
+{
+	u64 last_read = READ_ONCE(rx->gve->last_sync_nic_counter);
+	struct sk_buff *skb = rx->ctx.skb_head;
+	u32 low = (u32)last_read;
+	s32 diff = hwts - low;
+
+	skb_hwtstamps(skb)->hwtstamp = ns_to_ktime(last_read + diff);
+}
+
+static void gve_rx_free_skb(struct napi_struct *napi, struct gve_rx_ring *rx)
+{
+	if (!rx->ctx.skb_head)
+		return;
+
+	if (rx->ctx.skb_head == napi->skb)
+		napi->skb = NULL;
+	dev_kfree_skb_any(rx->ctx.skb_head);
+	rx->ctx.skb_head = NULL;
+	rx->ctx.skb_tail = NULL;
+}
+
+static bool gve_rx_should_trigger_copy_ondemand(struct gve_rx_ring *rx)
+{
+	if (!rx->dqo.qpl)
+		return false;
+	if (rx->dqo.used_buf_states_cnt <
+		     (rx->dqo.num_buf_states -
+		     GVE_DQO_QPL_ONDEMAND_ALLOC_THRESHOLD))
+		return false;
+	return true;
+}
+
+static int gve_rx_copy_ondemand(struct gve_rx_ring *rx,
+				struct gve_rx_buf_state_dqo *buf_state,
+				u16 buf_len)
+{
+	struct page *page = alloc_pages_node(rx->gve->numa_node, GFP_ATOMIC, 0);
+	int num_frags;
+
+	if (!page)
+		return -ENOMEM;
+
+	memcpy(page_address(page),
+	       buf_state->page_info.page_address +
+	       buf_state->page_info.page_offset,
+	       buf_len);
+	num_frags = skb_shinfo(rx->ctx.skb_tail)->nr_frags;
+	skb_add_rx_frag(rx->ctx.skb_tail, num_frags, page,
+			0, buf_len, PAGE_SIZE);
+
+	u64_stats_update_begin(&rx->statss);
+	rx->rx_frag_alloc_cnt++;
+	u64_stats_update_end(&rx->statss);
+	/* Return unused buffer. */
+	gve_enqueue_buf_state(rx, &rx->dqo.recycled_buf_states, buf_state);
+	return 0;
+}
+
+static void gve_skb_add_rx_frag(struct gve_rx_ring *rx,
+				struct gve_rx_buf_state_dqo *buf_state,
+				int num_frags, u16 buf_len)
+{
+	if (rx->dqo.page_pool) {
+		skb_add_rx_frag_netmem(rx->ctx.skb_tail, num_frags,
+				       buf_state->page_info.netmem,
+				       buf_state->page_info.page_offset +
+				       buf_state->page_info.pad, buf_len,
+				       buf_state->page_info.buf_size);
+	} else {
+		skb_add_rx_frag(rx->ctx.skb_tail, num_frags,
+				buf_state->page_info.page,
+				buf_state->page_info.page_offset +
+				buf_state->page_info.pad, buf_len,
+				buf_state->page_info.buf_size);
+	}
+}
+
+/* Chains multi skbs for single rx packet.
+ * Returns 0 if buffer is appended, -1 otherwise.
+ */
+static int gve_rx_append_frags(struct napi_struct *napi,
+			       struct gve_rx_buf_state_dqo *buf_state,
+			       u16 buf_len, struct gve_rx_ring *rx,
+			       struct gve_priv *priv)
+{
+	int num_frags = skb_shinfo(rx->ctx.skb_tail)->nr_frags;
+
+	if (unlikely(num_frags == MAX_SKB_FRAGS)) {
+		struct sk_buff *skb;
+
+		skb = napi_alloc_skb(napi, 0);
+		if (!skb)
+			return -1;
+
+		if (rx->dqo.page_pool)
+			skb_mark_for_recycle(skb);
+
+		if (rx->ctx.skb_tail == rx->ctx.skb_head)
+			skb_shinfo(rx->ctx.skb_head)->frag_list = skb;
+		else
+			rx->ctx.skb_tail->next = skb;
+		rx->ctx.skb_tail = skb;
+		num_frags = 0;
+	}
+	if (rx->ctx.skb_tail != rx->ctx.skb_head) {
+		rx->ctx.skb_head->len += buf_len;
+		rx->ctx.skb_head->data_len += buf_len;
+		rx->ctx.skb_head->truesize += buf_state->page_info.buf_size;
+	}
+
+	/* Trigger ondemand page allocation if we are running low on buffers */
+	if (gve_rx_should_trigger_copy_ondemand(rx))
+		return gve_rx_copy_ondemand(rx, buf_state, buf_len);
+
+	gve_skb_add_rx_frag(rx, buf_state, num_frags, buf_len);
+	gve_reuse_buffer(rx, buf_state);
+	return 0;
+}
+
+static int gve_xdp_tx_dqo(struct gve_priv *priv, struct gve_rx_ring *rx,
+			  struct xdp_buff *xdp)
+{
+	struct gve_tx_ring *tx;
+	struct xdp_frame *xdpf;
+	u32 tx_qid;
+	int err;
+
+	xdpf = xdp_convert_buff_to_frame(xdp);
+	if (unlikely(!xdpf)) {
+		if (rx->xsk_pool)
+			xsk_buff_free(xdp);
+		return -ENOSPC;
+	}
+
+	tx_qid = gve_xdp_tx_queue_id(priv, rx->q_num);
+	tx = &priv->tx[tx_qid];
+	spin_lock(&tx->dqo_tx.xdp_lock);
+	err = gve_xdp_xmit_one_dqo(priv, tx, xdpf);
+	spin_unlock(&tx->dqo_tx.xdp_lock);
+
+	return err;
+}
+
+static void gve_xsk_done_dqo(struct gve_priv *priv, struct gve_rx_ring *rx,
+			     struct xdp_buff *xdp, struct bpf_prog *xprog,
+			     int xdp_act)
+{
+	switch (xdp_act) {
+	case XDP_ABORTED:
+	case XDP_DROP:
+	default:
+		xsk_buff_free(xdp);
+		break;
+	case XDP_TX:
+		if (unlikely(gve_xdp_tx_dqo(priv, rx, xdp)))
+			goto err;
+		break;
+	case XDP_REDIRECT:
+		if (unlikely(xdp_do_redirect(priv->dev, xdp, xprog)))
+			goto err;
+		break;
+	}
+
+	u64_stats_update_begin(&rx->statss);
+	if ((u32)xdp_act < GVE_XDP_ACTIONS)
+		rx->xdp_actions[xdp_act]++;
+	u64_stats_update_end(&rx->statss);
+	return;
+
+err:
+	u64_stats_update_begin(&rx->statss);
+	if (xdp_act == XDP_TX)
+		rx->xdp_tx_errors++;
+	if (xdp_act == XDP_REDIRECT)
+		rx->xdp_redirect_errors++;
+	u64_stats_update_end(&rx->statss);
+}
+
+static void gve_xdp_done_dqo(struct gve_priv *priv, struct gve_rx_ring *rx,
+			     struct xdp_buff *xdp, struct bpf_prog *xprog,
+			     int xdp_act,
+			     struct gve_rx_buf_state_dqo *buf_state)
+{
+	int err;
+	switch (xdp_act) {
+	case XDP_ABORTED:
+	case XDP_DROP:
+	default:
+		gve_free_buffer(rx, buf_state);
+		break;
+	case XDP_TX:
+		err = gve_xdp_tx_dqo(priv, rx, xdp);
+		if (unlikely(err))
+			goto err;
+		gve_reuse_buffer(rx, buf_state);
+		break;
+	case XDP_REDIRECT:
+		err = xdp_do_redirect(priv->dev, xdp, xprog);
+		if (unlikely(err))
+			goto err;
+		gve_reuse_buffer(rx, buf_state);
+		break;
+	}
+	u64_stats_update_begin(&rx->statss);
+	if ((u32)xdp_act < GVE_XDP_ACTIONS)
+		rx->xdp_actions[xdp_act]++;
+	u64_stats_update_end(&rx->statss);
+	return;
+err:
+	u64_stats_update_begin(&rx->statss);
+	if (xdp_act == XDP_TX)
+		rx->xdp_tx_errors++;
+	else if (xdp_act == XDP_REDIRECT)
+		rx->xdp_redirect_errors++;
+	u64_stats_update_end(&rx->statss);
+	gve_free_buffer(rx, buf_state);
+	return;
+}
+
+static int gve_rx_xsk_dqo(struct napi_struct *napi, struct gve_rx_ring *rx,
+			  struct gve_rx_buf_state_dqo *buf_state, int buf_len,
+			  struct bpf_prog *xprog)
+{
+	struct xdp_buff *xdp = buf_state->xsk_buff;
+	struct gve_priv *priv = rx->gve;
+	int xdp_act;
+
+	xdp->data_end = xdp->data + buf_len;
+	xsk_buff_dma_sync_for_cpu(xdp);
+
+	if (xprog) {
+		xdp_act = bpf_prog_run_xdp(xprog, xdp);
+		buf_len = xdp->data_end - xdp->data;
+		if (xdp_act != XDP_PASS) {
+			gve_xsk_done_dqo(priv, rx, xdp, xprog, xdp_act);
+			gve_free_buf_state(rx, buf_state);
+			return 0;
+		}
+	}
+
+	/* Copy the data to skb */
+	rx->ctx.skb_head = gve_rx_copy_data(priv->dev, napi,
+					    xdp->data, buf_len);
+	if (unlikely(!rx->ctx.skb_head)) {
+		xsk_buff_free(xdp);
+		gve_free_buf_state(rx, buf_state);
+		return -ENOMEM;
+	}
+	rx->ctx.skb_tail = rx->ctx.skb_head;
+
+	/* Free XSK buffer and Buffer state */
+	xsk_buff_free(xdp);
+	gve_free_buf_state(rx, buf_state);
+
+	/* Update Stats */
+	u64_stats_update_begin(&rx->statss);
+	rx->xdp_actions[XDP_PASS]++;
+	u64_stats_update_end(&rx->statss);
+	return 0;
+}
+
+static void gve_dma_sync(struct gve_priv *priv, struct gve_rx_ring *rx,
+			 struct gve_rx_buf_state_dqo *buf_state, u16 buf_len)
+{
+	struct gve_rx_slot_page_info *page_info = &buf_state->page_info;
+
+	if (rx->dqo.page_pool) {
+		page_pool_dma_sync_netmem_for_cpu(rx->dqo.page_pool,
+						  page_info->netmem,
+						  page_info->page_offset,
+						  buf_len);
+	} else {
+		dma_sync_single_range_for_cpu(&priv->pdev->dev, buf_state->addr,
+					      page_info->page_offset +
+					      page_info->pad,
+					      buf_len, DMA_FROM_DEVICE);
+	}
+}
+
+/* Returns 0 if descriptor is completed successfully.
+ * Returns -EINVAL if descriptor is invalid.
+ * Returns -ENOMEM if data cannot be copied to skb.
+ */
+static int gve_rx_dqo(struct napi_struct *napi, struct gve_rx_ring *rx,
+		      const struct gve_rx_compl_desc_dqo *compl_desc,
+		      u32 desc_idx, int queue_idx)
+{
+	const u16 buffer_id = le16_to_cpu(compl_desc->buf_id);
+	const bool hbo = compl_desc->header_buffer_overflow;
+	const bool eop = compl_desc->end_of_packet != 0;
+	const bool hsplit = compl_desc->split_header;
+	struct gve_rx_buf_state_dqo *buf_state;
+	struct gve_priv *priv = rx->gve;
+	struct bpf_prog *xprog;
+	u16 buf_len;
+	u16 hdr_len;
+
+	if (unlikely(buffer_id >= rx->dqo.num_buf_states)) {
+		net_err_ratelimited("%s: Invalid RX buffer_id=%u\n",
+				    priv->dev->name, buffer_id);
+		return -EINVAL;
+	}
+	buf_state = &rx->dqo.buf_states[buffer_id];
+	if (unlikely(!gve_buf_state_is_allocated(rx, buf_state))) {
+		net_err_ratelimited("%s: RX buffer_id is not allocated: %u\n",
+				    priv->dev->name, buffer_id);
+		return -EINVAL;
+	}
+
+	if (unlikely(compl_desc->rx_error)) {
+		gve_free_buffer(rx, buf_state);
+		return -EINVAL;
+	}
+
+	buf_len = compl_desc->packet_len;
+	hdr_len = compl_desc->header_len;
+
+	xprog = READ_ONCE(priv->xdp_prog);
+	if (buf_state->xsk_buff)
+		return gve_rx_xsk_dqo(napi, rx, buf_state, buf_len, xprog);
+
+	/* Page might have not been used for awhile and was likely last written
+	 * by a different thread.
+	 */
+	if (rx->dqo.page_pool) {
+		if (!netmem_is_net_iov(buf_state->page_info.netmem))
+			prefetch(netmem_to_page(buf_state->page_info.netmem));
+	} else {
+		prefetch(buf_state->page_info.page);
+	}
+
+	/* Copy the header into the skb in the case of header split */
+	if (hsplit) {
+		int unsplit = 0;
+
+		if (hdr_len && !hbo) {
+			rx->ctx.skb_head = gve_rx_copy_data(priv->dev, napi,
+							    rx->dqo.hdr_bufs.data +
+							    desc_idx * priv->header_buf_size,
+							    hdr_len);
+			if (unlikely(!rx->ctx.skb_head))
+				goto error;
+			rx->ctx.skb_tail = rx->ctx.skb_head;
+
+			if (rx->dqo.page_pool)
+				skb_mark_for_recycle(rx->ctx.skb_head);
+		} else {
+			unsplit = 1;
+		}
+		u64_stats_update_begin(&rx->statss);
+		rx->rx_hsplit_pkt++;
+		rx->rx_hsplit_unsplit_pkt += unsplit;
+		rx->rx_hsplit_bytes += hdr_len;
+		u64_stats_update_end(&rx->statss);
+	} else if (!rx->ctx.skb_head && rx->dqo.page_pool &&
+		   netmem_is_net_iov(buf_state->page_info.netmem)) {
+		/* when header split is disabled, the header went to the packet
+		 * buffer. If the packet buffer is a net_iov, those can't be
+		 * easily mapped into the kernel space to access the header
+		 * required to process the packet.
+		 */
+		goto error;
+	}
+
+	/* Sync the portion of dma buffer for CPU to read. */
+	gve_dma_sync(priv, rx, buf_state, buf_len);
+
+	/* Append to current skb if one exists. */
+	if (rx->ctx.skb_head) {
+		if (unlikely(gve_rx_append_frags(napi, buf_state, buf_len, rx,
+						 priv)) != 0) {
+			goto error;
+		}
+		return 0;
+	}
+
+	if (xprog) {
+		struct xdp_buff xdp;
+		void *old_data;
+		int xdp_act;
+
+		xdp_init_buff(&xdp, buf_state->page_info.buf_size,
+			      &rx->xdp_rxq);
+		xdp_prepare_buff(&xdp,
+				 buf_state->page_info.page_address +
+				 buf_state->page_info.page_offset,
+				 buf_state->page_info.pad,
+				 buf_len, false);
+		old_data = xdp.data;
+		xdp_act = bpf_prog_run_xdp(xprog, &xdp);
+		buf_state->page_info.pad += xdp.data - old_data;
+		buf_len = xdp.data_end - xdp.data;
+		if (xdp_act != XDP_PASS) {
+			gve_xdp_done_dqo(priv, rx, &xdp, xprog, xdp_act,
+					 buf_state);
+			return 0;
+		}
+
+		u64_stats_update_begin(&rx->statss);
+		rx->xdp_actions[XDP_PASS]++;
+		u64_stats_update_end(&rx->statss);
+	}
+
+	if (eop && buf_len <= priv->rx_copybreak &&
+	    !(rx->dqo.page_pool &&
+	      netmem_is_net_iov(buf_state->page_info.netmem))) {
+		rx->ctx.skb_head = gve_rx_copy(priv->dev, napi,
+					       &buf_state->page_info, buf_len);
+		if (unlikely(!rx->ctx.skb_head))
+			goto error;
+		rx->ctx.skb_tail = rx->ctx.skb_head;
+
+		u64_stats_update_begin(&rx->statss);
+		rx->rx_copied_pkt++;
+		rx->rx_copybreak_pkt++;
+		u64_stats_update_end(&rx->statss);
+
+		gve_free_buffer(rx, buf_state);
+		return 0;
+	}
+
+	rx->ctx.skb_head = napi_get_frags(napi);
+	if (unlikely(!rx->ctx.skb_head))
+		goto error;
+	rx->ctx.skb_tail = rx->ctx.skb_head;
+
+	if (gve_rx_should_trigger_copy_ondemand(rx)) {
+		if (gve_rx_copy_ondemand(rx, buf_state, buf_len) < 0)
+			goto error;
+		return 0;
+	}
+
+	if (rx->dqo.page_pool)
+		skb_mark_for_recycle(rx->ctx.skb_head);
+
+	gve_skb_add_rx_frag(rx, buf_state, 0, buf_len);
+	gve_reuse_buffer(rx, buf_state);
+	return 0;
+
+error:
+	gve_free_buffer(rx, buf_state);
+	return -ENOMEM;
+}
+
+static int gve_rx_complete_rsc(struct sk_buff *skb,
+			       const struct gve_rx_compl_desc_dqo *desc,
+			       struct gve_ptype ptype)
+{
+	struct skb_shared_info *shinfo = skb_shinfo(skb);
+
+	/* Only TCP is supported right now. */
+	if (ptype.l4_type != GVE_L4_TYPE_TCP)
+		return -EINVAL;
+
+	switch (ptype.l3_type) {
+	case GVE_L3_TYPE_IPV4:
+		shinfo->gso_type = SKB_GSO_TCPV4;
+		break;
+	case GVE_L3_TYPE_IPV6:
+		shinfo->gso_type = SKB_GSO_TCPV6;
+		break;
+	default:
+		return -EINVAL;
+	}
+
+	shinfo->gso_size = le16_to_cpu(desc->rsc_seg_len);
+	return 0;
+}
+
+/* Returns 0 if skb is completed successfully, -1 otherwise. */
+static int gve_rx_complete_skb(struct gve_rx_ring *rx, struct napi_struct *napi,
+			       const struct gve_rx_compl_desc_dqo *desc,
+			       netdev_features_t feat)
+{
+	struct gve_ptype ptype =
+		rx->gve->ptype_lut_dqo->ptypes[desc->packet_type];
+	int err;
+
+	skb_record_rx_queue(rx->ctx.skb_head, rx->q_num);
+
+	if (feat & NETIF_F_RXHASH)
+		gve_rx_skb_hash(rx->ctx.skb_head, desc, ptype);
+
+	if (feat & NETIF_F_RXCSUM)
+		gve_rx_skb_csum(rx->ctx.skb_head, desc, ptype);
+
+	if (rx->gve->ts_config.rx_filter == HWTSTAMP_FILTER_ALL)
+		gve_rx_skb_hwtstamp(rx, le32_to_cpu(desc->ts));
+
+	/* RSC packets must set gso_size otherwise the TCP stack will complain
+	 * that packets are larger than MTU.
+	 */
+	if (desc->rsc) {
+		err = gve_rx_complete_rsc(rx->ctx.skb_head, desc, ptype);
+		if (err < 0)
+			return err;
+	}
+
+	if (skb_headlen(rx->ctx.skb_head) == 0)
+		napi_gro_frags(napi);
+	else
+		napi_gro_receive(napi, rx->ctx.skb_head);
+
+	return 0;
+}
+
+int gve_rx_poll_dqo(struct gve_notify_block *block, int budget)
+{
+	struct gve_rx_compl_queue_dqo *complq;
+	struct napi_struct *napi;
+	netdev_features_t feat;
+	struct gve_rx_ring *rx;
+	struct gve_priv *priv;
+	u64 xdp_redirects;
+	u32 work_done = 0;
+	u64 bytes = 0;
+	u64 xdp_txs;
+	int err;
+
+	napi = &block->napi;
+	feat = napi->dev->features;
+
+	rx = block->rx;
+	priv = rx->gve;
+	complq = &rx->dqo.complq;
+
+	xdp_redirects = rx->xdp_actions[XDP_REDIRECT];
+	xdp_txs = rx->xdp_actions[XDP_TX];
+
+	while (work_done < budget) {
+		struct gve_rx_compl_desc_dqo *compl_desc =
+			&complq->desc_ring[complq->head];
+		u32 pkt_bytes;
+
+		/* No more new packets */
+		if (compl_desc->generation == complq->cur_gen_bit)
+			break;
+
+		/* Prefetch the next two descriptors. */
+		prefetch(&complq->desc_ring[(complq->head + 1) & complq->mask]);
+		prefetch(&complq->desc_ring[(complq->head + 2) & complq->mask]);
+
+		/* Do not read data until we own the descriptor */
+		dma_rmb();
+
+		err = gve_rx_dqo(napi, rx, compl_desc, complq->head, rx->q_num);
+		if (err < 0) {
+			gve_rx_free_skb(napi, rx);
+			u64_stats_update_begin(&rx->statss);
+			if (err == -ENOMEM)
+				rx->rx_skb_alloc_fail++;
+			else if (err == -EINVAL)
+				rx->rx_desc_err_dropped_pkt++;
+			u64_stats_update_end(&rx->statss);
+		}
+
+		complq->head = (complq->head + 1) & complq->mask;
+		complq->num_free_slots++;
+
+		/* When the ring wraps, the generation bit is flipped. */
+		complq->cur_gen_bit ^= (complq->head == 0);
+
+		/* Receiving a completion means we have space to post another
+		 * buffer on the buffer queue.
+		 */
+		{
+			struct gve_rx_buf_queue_dqo *bufq = &rx->dqo.bufq;
+
+			bufq->head = (bufq->head + 1) & bufq->mask;
+		}
+
+		/* Free running counter of completed descriptors */
+		rx->cnt++;
+
+		if (!rx->ctx.skb_head)
+			continue;
+
+		if (!compl_desc->end_of_packet)
+			continue;
+
+		work_done++;
+		pkt_bytes = rx->ctx.skb_head->len;
+		/* The ethernet header (first ETH_HLEN bytes) is snipped off
+		 * by eth_type_trans.
+		 */
+		if (skb_headlen(rx->ctx.skb_head))
+			pkt_bytes += ETH_HLEN;
+
+		/* gve_rx_complete_skb() will consume skb if successful */
+		if (gve_rx_complete_skb(rx, napi, compl_desc, feat) != 0) {
+			gve_rx_free_skb(napi, rx);
+			u64_stats_update_begin(&rx->statss);
+			rx->rx_desc_err_dropped_pkt++;
+			u64_stats_update_end(&rx->statss);
+			continue;
+		}
+
+		bytes += pkt_bytes;
+		rx->ctx.skb_head = NULL;
+		rx->ctx.skb_tail = NULL;
+	}
+
+	if (xdp_txs != rx->xdp_actions[XDP_TX])
+		gve_xdp_tx_flush_dqo(priv, rx->q_num);
+
+	if (xdp_redirects != rx->xdp_actions[XDP_REDIRECT])
+		xdp_do_flush();
+
+	gve_rx_post_buffers_dqo(rx);
+
+	u64_stats_update_begin(&rx->statss);
+	rx->rpackets += work_done;
+	rx->rbytes += bytes;
+	u64_stats_update_end(&rx->statss);
+
+	return work_done;
+}
diff --git a/drivers/net/ethernet/google/gve/gve_tx.c b/drivers/net/ethernet/google/gve/gve_tx.c
new file mode 100644
index 000000000000..c6ff0968929d
--- /dev/null
+++ b/drivers/net/ethernet/google/gve/gve_tx.c
@@ -0,0 +1,1040 @@
+// SPDX-License-Identifier: (GPL-2.0 OR MIT)
+/* Google virtual Ethernet (gve) driver
+ *
+ * Copyright (C) 2015-2021 Google, Inc.
+ */
+
+#include "gve.h"
+#include "gve_adminq.h"
+#include "gve_utils.h"
+#include <linux/ip.h>
+#include <linux/tcp.h>
+#include <linux/vmalloc.h>
+#include <linux/skbuff.h>
+#include <net/xdp_sock_drv.h>
+
+static inline void gve_tx_put_doorbell(struct gve_priv *priv,
+				       struct gve_queue_resources *q_resources,
+				       u32 val)
+{
+	iowrite32be(val, &priv->db_bar2[be32_to_cpu(q_resources->db_index)]);
+}
+
+void gve_xdp_tx_flush(struct gve_priv *priv, u32 xdp_qid)
+{
+	u32 tx_qid = gve_xdp_tx_queue_id(priv, xdp_qid);
+	struct gve_tx_ring *tx = &priv->tx[tx_qid];
+
+	gve_tx_put_doorbell(priv, tx->q_resources, tx->req);
+}
+
+/* gvnic can only transmit from a Registered Segment.
+ * We copy skb payloads into the registered segment before writing Tx
+ * descriptors and ringing the Tx doorbell.
+ *
+ * gve_tx_fifo_* manages the Registered Segment as a FIFO - clients must
+ * free allocations in the order they were allocated.
+ */
+
+static int gve_tx_fifo_init(struct gve_priv *priv, struct gve_tx_fifo *fifo)
+{
+	fifo->base = vmap(fifo->qpl->pages, fifo->qpl->num_entries, VM_MAP,
+			  PAGE_KERNEL);
+	if (unlikely(!fifo->base)) {
+		netif_err(priv, drv, priv->dev, "Failed to vmap fifo, qpl_id = %d\n",
+			  fifo->qpl->id);
+		return -ENOMEM;
+	}
+
+	fifo->size = fifo->qpl->num_entries * PAGE_SIZE;
+	atomic_set(&fifo->available, fifo->size);
+	fifo->head = 0;
+	return 0;
+}
+
+static void gve_tx_fifo_release(struct gve_priv *priv, struct gve_tx_fifo *fifo)
+{
+	WARN(atomic_read(&fifo->available) != fifo->size,
+	     "Releasing non-empty fifo");
+
+	vunmap(fifo->base);
+}
+
+static int gve_tx_fifo_pad_alloc_one_frag(struct gve_tx_fifo *fifo,
+					  size_t bytes)
+{
+	return (fifo->head + bytes < fifo->size) ? 0 : fifo->size - fifo->head;
+}
+
+static bool gve_tx_fifo_can_alloc(struct gve_tx_fifo *fifo, size_t bytes)
+{
+	return (atomic_read(&fifo->available) <= bytes) ? false : true;
+}
+
+/* gve_tx_alloc_fifo - Allocate fragment(s) from Tx FIFO
+ * @fifo: FIFO to allocate from
+ * @bytes: Allocation size
+ * @iov: Scatter-gather elements to fill with allocation fragment base/len
+ *
+ * Returns number of valid elements in iov[] or negative on error.
+ *
+ * Allocations from a given FIFO must be externally synchronized but concurrent
+ * allocation and frees are allowed.
+ */
+static int gve_tx_alloc_fifo(struct gve_tx_fifo *fifo, size_t bytes,
+			     struct gve_tx_iovec iov[2])
+{
+	size_t overflow, padding;
+	u32 aligned_head;
+	int nfrags = 0;
+
+	if (!bytes)
+		return 0;
+
+	/* This check happens before we know how much padding is needed to
+	 * align to a cacheline boundary for the payload, but that is fine,
+	 * because the FIFO head always start aligned, and the FIFO's boundaries
+	 * are aligned, so if there is space for the data, there is space for
+	 * the padding to the next alignment.
+	 */
+	WARN(!gve_tx_fifo_can_alloc(fifo, bytes),
+	     "Reached %s when there's not enough space in the fifo", __func__);
+
+	nfrags++;
+
+	iov[0].iov_offset = fifo->head;
+	iov[0].iov_len = bytes;
+	fifo->head += bytes;
+
+	if (fifo->head > fifo->size) {
+		/* If the allocation did not fit in the tail fragment of the
+		 * FIFO, also use the head fragment.
+		 */
+		nfrags++;
+		overflow = fifo->head - fifo->size;
+		iov[0].iov_len -= overflow;
+		iov[1].iov_offset = 0;	/* Start of fifo*/
+		iov[1].iov_len = overflow;
+
+		fifo->head = overflow;
+	}
+
+	/* Re-align to a cacheline boundary */
+	aligned_head = L1_CACHE_ALIGN(fifo->head);
+	padding = aligned_head - fifo->head;
+	iov[nfrags - 1].iov_padding = padding;
+	atomic_sub(bytes + padding, &fifo->available);
+	fifo->head = aligned_head;
+
+	if (fifo->head == fifo->size)
+		fifo->head = 0;
+
+	return nfrags;
+}
+
+/* gve_tx_free_fifo - Return space to Tx FIFO
+ * @fifo: FIFO to return fragments to
+ * @bytes: Bytes to free
+ */
+static void gve_tx_free_fifo(struct gve_tx_fifo *fifo, size_t bytes)
+{
+	atomic_add(bytes, &fifo->available);
+}
+
+static size_t gve_tx_clear_buffer_state(struct gve_tx_buffer_state *info)
+{
+	size_t space_freed = 0;
+	int i;
+
+	for (i = 0; i < ARRAY_SIZE(info->iov); i++) {
+		space_freed += info->iov[i].iov_len + info->iov[i].iov_padding;
+		info->iov[i].iov_len = 0;
+		info->iov[i].iov_padding = 0;
+	}
+	return space_freed;
+}
+
+static int gve_clean_xdp_done(struct gve_priv *priv, struct gve_tx_ring *tx,
+			      u32 to_do)
+{
+	struct gve_tx_buffer_state *info;
+	u64 pkts = 0, bytes = 0;
+	size_t space_freed = 0;
+	u32 xsk_complete = 0;
+	u32 idx;
+	int i;
+
+	for (i = 0; i < to_do; i++) {
+		idx = tx->done & tx->mask;
+		info = &tx->info[idx];
+		tx->done++;
+
+		if (unlikely(!info->xdp.size))
+			continue;
+
+		bytes += info->xdp.size;
+		pkts++;
+		xsk_complete += info->xdp.is_xsk;
+
+		info->xdp.size = 0;
+		if (info->xdp_frame) {
+			xdp_return_frame(info->xdp_frame);
+			info->xdp_frame = NULL;
+		}
+		space_freed += gve_tx_clear_buffer_state(info);
+	}
+
+	gve_tx_free_fifo(&tx->tx_fifo, space_freed);
+	if (xsk_complete > 0 && tx->xsk_pool)
+		xsk_tx_completed(tx->xsk_pool, xsk_complete);
+	u64_stats_update_begin(&tx->statss);
+	tx->bytes_done += bytes;
+	tx->pkt_done += pkts;
+	u64_stats_update_end(&tx->statss);
+	return pkts;
+}
+
+static int gve_clean_tx_done(struct gve_priv *priv, struct gve_tx_ring *tx,
+			     u32 to_do, bool try_to_wake);
+
+void gve_tx_stop_ring_gqi(struct gve_priv *priv, int idx)
+{
+	int ntfy_idx = gve_tx_idx_to_ntfy(priv, idx);
+	struct gve_tx_ring *tx = &priv->tx[idx];
+
+	if (!gve_tx_was_added_to_block(priv, idx))
+		return;
+
+	gve_remove_napi(priv, ntfy_idx);
+	if (tx->q_num < priv->tx_cfg.num_queues)
+		gve_clean_tx_done(priv, tx, priv->tx_desc_cnt, false);
+	else
+		gve_clean_xdp_done(priv, tx, priv->tx_desc_cnt);
+	netdev_tx_reset_queue(tx->netdev_txq);
+	gve_tx_remove_from_block(priv, idx);
+}
+
+static void gve_tx_free_ring_gqi(struct gve_priv *priv, struct gve_tx_ring *tx,
+				 struct gve_tx_alloc_rings_cfg *cfg)
+{
+	struct device *hdev = &priv->pdev->dev;
+	int idx = tx->q_num;
+	size_t bytes;
+	u32 qpl_id;
+	u32 slots;
+
+	slots = tx->mask + 1;
+	dma_free_coherent(hdev, sizeof(*tx->q_resources),
+			  tx->q_resources, tx->q_resources_bus);
+	tx->q_resources = NULL;
+
+	if (tx->tx_fifo.qpl) {
+		if (tx->tx_fifo.base)
+			gve_tx_fifo_release(priv, &tx->tx_fifo);
+
+		qpl_id = gve_tx_qpl_id(priv, tx->q_num);
+		gve_free_queue_page_list(priv, tx->tx_fifo.qpl, qpl_id);
+		tx->tx_fifo.qpl = NULL;
+	}
+
+	bytes = sizeof(*tx->desc) * slots;
+	dma_free_coherent(hdev, bytes, tx->desc, tx->bus);
+	tx->desc = NULL;
+
+	vfree(tx->info);
+	tx->info = NULL;
+
+	netif_dbg(priv, drv, priv->dev, "freed tx queue %d\n", idx);
+}
+
+void gve_tx_start_ring_gqi(struct gve_priv *priv, int idx)
+{
+	int ntfy_idx = gve_tx_idx_to_ntfy(priv, idx);
+	struct gve_tx_ring *tx = &priv->tx[idx];
+
+	gve_tx_add_to_block(priv, idx);
+
+	tx->netdev_txq = netdev_get_tx_queue(priv->dev, idx);
+	gve_add_napi(priv, ntfy_idx, gve_napi_poll);
+}
+
+static int gve_tx_alloc_ring_gqi(struct gve_priv *priv,
+				 struct gve_tx_alloc_rings_cfg *cfg,
+				 struct gve_tx_ring *tx,
+				 int idx)
+{
+	struct device *hdev = &priv->pdev->dev;
+	int qpl_page_cnt;
+	u32 qpl_id = 0;
+	size_t bytes;
+
+	/* Make sure everything is zeroed to start */
+	memset(tx, 0, sizeof(*tx));
+	spin_lock_init(&tx->clean_lock);
+	spin_lock_init(&tx->xdp_lock);
+	tx->q_num = idx;
+
+	tx->mask = cfg->ring_size - 1;
+
+	/* alloc metadata */
+	tx->info = vcalloc(cfg->ring_size, sizeof(*tx->info));
+	if (!tx->info)
+		return -ENOMEM;
+
+	/* alloc tx queue */
+	bytes = sizeof(*tx->desc) * cfg->ring_size;
+	tx->desc = dma_alloc_coherent(hdev, bytes, &tx->bus, GFP_KERNEL);
+	if (!tx->desc)
+		goto abort_with_info;
+
+	tx->raw_addressing = cfg->raw_addressing;
+	tx->dev = hdev;
+	if (!tx->raw_addressing) {
+		qpl_id = gve_tx_qpl_id(priv, tx->q_num);
+		qpl_page_cnt = priv->tx_pages_per_qpl;
+
+		tx->tx_fifo.qpl = gve_alloc_queue_page_list(priv, qpl_id,
+							    qpl_page_cnt);
+		if (!tx->tx_fifo.qpl)
+			goto abort_with_desc;
+
+		/* map Tx FIFO */
+		if (gve_tx_fifo_init(priv, &tx->tx_fifo))
+			goto abort_with_qpl;
+	}
+
+	tx->q_resources =
+		dma_alloc_coherent(hdev,
+				   sizeof(*tx->q_resources),
+				   &tx->q_resources_bus,
+				   GFP_KERNEL);
+	if (!tx->q_resources)
+		goto abort_with_fifo;
+
+	return 0;
+
+abort_with_fifo:
+	if (!tx->raw_addressing)
+		gve_tx_fifo_release(priv, &tx->tx_fifo);
+abort_with_qpl:
+	if (!tx->raw_addressing) {
+		gve_free_queue_page_list(priv, tx->tx_fifo.qpl, qpl_id);
+		tx->tx_fifo.qpl = NULL;
+	}
+abort_with_desc:
+	dma_free_coherent(hdev, bytes, tx->desc, tx->bus);
+	tx->desc = NULL;
+abort_with_info:
+	vfree(tx->info);
+	tx->info = NULL;
+	return -ENOMEM;
+}
+
+int gve_tx_alloc_rings_gqi(struct gve_priv *priv,
+			   struct gve_tx_alloc_rings_cfg *cfg)
+{
+	struct gve_tx_ring *tx = cfg->tx;
+	int total_queues;
+	int err = 0;
+	int i, j;
+
+	total_queues = cfg->qcfg->num_queues + cfg->num_xdp_rings;
+	if (total_queues > cfg->qcfg->max_queues) {
+		netif_err(priv, drv, priv->dev,
+			  "Cannot alloc more than the max num of Tx rings\n");
+		return -EINVAL;
+	}
+
+	tx = kvcalloc(cfg->qcfg->max_queues, sizeof(struct gve_tx_ring),
+		      GFP_KERNEL);
+	if (!tx)
+		return -ENOMEM;
+
+	for (i = 0; i < total_queues; i++) {
+		err = gve_tx_alloc_ring_gqi(priv, cfg, &tx[i], i);
+		if (err) {
+			netif_err(priv, drv, priv->dev,
+				  "Failed to alloc tx ring=%d: err=%d\n",
+				  i, err);
+			goto cleanup;
+		}
+	}
+
+	cfg->tx = tx;
+	return 0;
+
+cleanup:
+	for (j = 0; j < i; j++)
+		gve_tx_free_ring_gqi(priv, &tx[j], cfg);
+	kvfree(tx);
+	return err;
+}
+
+void gve_tx_free_rings_gqi(struct gve_priv *priv,
+			   struct gve_tx_alloc_rings_cfg *cfg)
+{
+	struct gve_tx_ring *tx = cfg->tx;
+	int i;
+
+	if (!tx)
+		return;
+
+	for (i = 0; i < cfg->qcfg->num_queues + cfg->qcfg->num_xdp_queues; i++)
+		gve_tx_free_ring_gqi(priv, &tx[i], cfg);
+
+	kvfree(tx);
+	cfg->tx = NULL;
+}
+
+/* gve_tx_avail - Calculates the number of slots available in the ring
+ * @tx: tx ring to check
+ *
+ * Returns the number of slots available
+ *
+ * The capacity of the queue is mask + 1. We don't need to reserve an entry.
+ **/
+static inline u32 gve_tx_avail(struct gve_tx_ring *tx)
+{
+	return tx->mask + 1 - (tx->req - tx->done);
+}
+
+static inline int gve_skb_fifo_bytes_required(struct gve_tx_ring *tx,
+					      struct sk_buff *skb)
+{
+	int pad_bytes, align_hdr_pad;
+	int bytes;
+	int hlen;
+
+	hlen = skb_is_gso(skb) ? skb_checksum_start_offset(skb) + tcp_hdrlen(skb) :
+				 min_t(int, GVE_GQ_TX_MIN_PKT_DESC_BYTES, skb->len);
+
+	pad_bytes = gve_tx_fifo_pad_alloc_one_frag(&tx->tx_fifo,
+						   hlen);
+	/* We need to take into account the header alignment padding. */
+	align_hdr_pad = L1_CACHE_ALIGN(hlen) - hlen;
+	bytes = align_hdr_pad + pad_bytes + skb->len;
+
+	return bytes;
+}
+
+/* The most descriptors we could need is MAX_SKB_FRAGS + 4 :
+ * 1 for each skb frag
+ * 1 for the skb linear portion
+ * 1 for when tcp hdr needs to be in separate descriptor
+ * 1 if the payload wraps to the beginning of the FIFO
+ * 1 for metadata descriptor
+ */
+#define MAX_TX_DESC_NEEDED	(MAX_SKB_FRAGS + 4)
+static void gve_tx_unmap_buf(struct device *dev, struct gve_tx_buffer_state *info)
+{
+	if (info->skb) {
+		dma_unmap_single(dev, dma_unmap_addr(info, dma),
+				 dma_unmap_len(info, len),
+				 DMA_TO_DEVICE);
+		dma_unmap_len_set(info, len, 0);
+	} else {
+		dma_unmap_page(dev, dma_unmap_addr(info, dma),
+			       dma_unmap_len(info, len),
+			       DMA_TO_DEVICE);
+		dma_unmap_len_set(info, len, 0);
+	}
+}
+
+/* Check if sufficient resources (descriptor ring space, FIFO space) are
+ * available to transmit the given number of bytes.
+ */
+static inline bool gve_can_tx(struct gve_tx_ring *tx, int bytes_required)
+{
+	bool can_alloc = true;
+
+	if (!tx->raw_addressing)
+		can_alloc = gve_tx_fifo_can_alloc(&tx->tx_fifo, bytes_required);
+
+	return (gve_tx_avail(tx) >= MAX_TX_DESC_NEEDED && can_alloc);
+}
+
+static_assert(NAPI_POLL_WEIGHT >= MAX_TX_DESC_NEEDED);
+
+/* Stops the queue if the skb cannot be transmitted. */
+static int gve_maybe_stop_tx(struct gve_priv *priv, struct gve_tx_ring *tx,
+			     struct sk_buff *skb)
+{
+	int bytes_required = 0;
+	u32 nic_done;
+	u32 to_do;
+	int ret;
+
+	if (!tx->raw_addressing)
+		bytes_required = gve_skb_fifo_bytes_required(tx, skb);
+
+	if (likely(gve_can_tx(tx, bytes_required)))
+		return 0;
+
+	ret = -EBUSY;
+	spin_lock(&tx->clean_lock);
+	nic_done = gve_tx_load_event_counter(priv, tx);
+	to_do = nic_done - tx->done;
+
+	/* Only try to clean if there is hope for TX */
+	if (to_do + gve_tx_avail(tx) >= MAX_TX_DESC_NEEDED) {
+		if (to_do > 0) {
+			to_do = min_t(u32, to_do, NAPI_POLL_WEIGHT);
+			gve_clean_tx_done(priv, tx, to_do, false);
+		}
+		if (likely(gve_can_tx(tx, bytes_required)))
+			ret = 0;
+	}
+	if (ret) {
+		/* No space, so stop the queue */
+		tx->stop_queue++;
+		netif_tx_stop_queue(tx->netdev_txq);
+	}
+	spin_unlock(&tx->clean_lock);
+
+	return ret;
+}
+
+static void gve_tx_fill_pkt_desc(union gve_tx_desc *pkt_desc,
+				 u16 csum_offset, u8 ip_summed, bool is_gso,
+				 int l4_hdr_offset, u32 desc_cnt,
+				 u16 hlen, u64 addr, u16 pkt_len)
+{
+	/* l4_hdr_offset and csum_offset are in units of 16-bit words */
+	if (is_gso) {
+		pkt_desc->pkt.type_flags = GVE_TXD_TSO | GVE_TXF_L4CSUM;
+		pkt_desc->pkt.l4_csum_offset = csum_offset >> 1;
+		pkt_desc->pkt.l4_hdr_offset = l4_hdr_offset >> 1;
+	} else if (likely(ip_summed == CHECKSUM_PARTIAL)) {
+		pkt_desc->pkt.type_flags = GVE_TXD_STD | GVE_TXF_L4CSUM;
+		pkt_desc->pkt.l4_csum_offset = csum_offset >> 1;
+		pkt_desc->pkt.l4_hdr_offset = l4_hdr_offset >> 1;
+	} else {
+		pkt_desc->pkt.type_flags = GVE_TXD_STD;
+		pkt_desc->pkt.l4_csum_offset = 0;
+		pkt_desc->pkt.l4_hdr_offset = 0;
+	}
+	pkt_desc->pkt.desc_cnt = desc_cnt;
+	pkt_desc->pkt.len = cpu_to_be16(pkt_len);
+	pkt_desc->pkt.seg_len = cpu_to_be16(hlen);
+	pkt_desc->pkt.seg_addr = cpu_to_be64(addr);
+}
+
+static void gve_tx_fill_mtd_desc(union gve_tx_desc *mtd_desc,
+				 struct sk_buff *skb)
+{
+	BUILD_BUG_ON(sizeof(mtd_desc->mtd) != sizeof(mtd_desc->pkt));
+
+	mtd_desc->mtd.type_flags = GVE_TXD_MTD | GVE_MTD_SUBTYPE_PATH;
+	mtd_desc->mtd.path_state = GVE_MTD_PATH_STATE_DEFAULT |
+				   GVE_MTD_PATH_HASH_L4;
+	mtd_desc->mtd.path_hash = cpu_to_be32(skb->hash);
+	mtd_desc->mtd.reserved0 = 0;
+	mtd_desc->mtd.reserved1 = 0;
+}
+
+static void gve_tx_fill_seg_desc(union gve_tx_desc *seg_desc,
+				 u16 l3_offset, u16 gso_size,
+				 bool is_gso_v6, bool is_gso,
+				 u16 len, u64 addr)
+{
+	seg_desc->seg.type_flags = GVE_TXD_SEG;
+	if (is_gso) {
+		if (is_gso_v6)
+			seg_desc->seg.type_flags |= GVE_TXSF_IPV6;
+		seg_desc->seg.l3_offset = l3_offset >> 1;
+		seg_desc->seg.mss = cpu_to_be16(gso_size);
+	}
+	seg_desc->seg.seg_len = cpu_to_be16(len);
+	seg_desc->seg.seg_addr = cpu_to_be64(addr);
+}
+
+static void gve_dma_sync_for_device(struct device *dev, dma_addr_t *page_buses,
+				    u64 iov_offset, u64 iov_len)
+{
+	u64 last_page = (iov_offset + iov_len - 1) / PAGE_SIZE;
+	u64 first_page = iov_offset / PAGE_SIZE;
+	u64 page;
+
+	for (page = first_page; page <= last_page; page++)
+		dma_sync_single_for_device(dev, page_buses[page], PAGE_SIZE, DMA_TO_DEVICE);
+}
+
+static int gve_tx_add_skb_copy(struct gve_priv *priv, struct gve_tx_ring *tx, struct sk_buff *skb)
+{
+	int pad_bytes, hlen, hdr_nfrags, payload_nfrags, l4_hdr_offset;
+	union gve_tx_desc *pkt_desc, *seg_desc;
+	struct gve_tx_buffer_state *info;
+	int mtd_desc_nr = !!skb->l4_hash;
+	bool is_gso = skb_is_gso(skb);
+	u32 idx = tx->req & tx->mask;
+	int payload_iov = 2;
+	int copy_offset;
+	u32 next_idx;
+	int i;
+
+	info = &tx->info[idx];
+	pkt_desc = &tx->desc[idx];
+
+	l4_hdr_offset = skb_checksum_start_offset(skb);
+	/* If the skb is gso, then we want the tcp header alone in the first segment
+	 * otherwise we want the minimum required by the gVNIC spec.
+	 */
+	hlen = is_gso ? l4_hdr_offset + tcp_hdrlen(skb) :
+			min_t(int, GVE_GQ_TX_MIN_PKT_DESC_BYTES, skb->len);
+
+	info->skb =  skb;
+	/* We don't want to split the header, so if necessary, pad to the end
+	 * of the fifo and then put the header at the beginning of the fifo.
+	 */
+	pad_bytes = gve_tx_fifo_pad_alloc_one_frag(&tx->tx_fifo, hlen);
+	hdr_nfrags = gve_tx_alloc_fifo(&tx->tx_fifo, hlen + pad_bytes,
+				       &info->iov[0]);
+	WARN(!hdr_nfrags, "hdr_nfrags should never be 0!");
+	payload_nfrags = gve_tx_alloc_fifo(&tx->tx_fifo, skb->len - hlen,
+					   &info->iov[payload_iov]);
+
+	gve_tx_fill_pkt_desc(pkt_desc, skb->csum_offset, skb->ip_summed,
+			     is_gso, l4_hdr_offset,
+			     1 + mtd_desc_nr + payload_nfrags, hlen,
+			     info->iov[hdr_nfrags - 1].iov_offset, skb->len);
+
+	skb_copy_bits(skb, 0,
+		      tx->tx_fifo.base + info->iov[hdr_nfrags - 1].iov_offset,
+		      hlen);
+	gve_dma_sync_for_device(&priv->pdev->dev, tx->tx_fifo.qpl->page_buses,
+				info->iov[hdr_nfrags - 1].iov_offset,
+				info->iov[hdr_nfrags - 1].iov_len);
+	copy_offset = hlen;
+
+	if (mtd_desc_nr) {
+		next_idx = (tx->req + 1) & tx->mask;
+		gve_tx_fill_mtd_desc(&tx->desc[next_idx], skb);
+	}
+
+	for (i = payload_iov; i < payload_nfrags + payload_iov; i++) {
+		next_idx = (tx->req + 1 + mtd_desc_nr + i - payload_iov) & tx->mask;
+		seg_desc = &tx->desc[next_idx];
+
+		gve_tx_fill_seg_desc(seg_desc, skb_network_offset(skb),
+				     skb_shinfo(skb)->gso_size,
+				     skb_is_gso_v6(skb), is_gso,
+				     info->iov[i].iov_len,
+				     info->iov[i].iov_offset);
+
+		skb_copy_bits(skb, copy_offset,
+			      tx->tx_fifo.base + info->iov[i].iov_offset,
+			      info->iov[i].iov_len);
+		gve_dma_sync_for_device(&priv->pdev->dev, tx->tx_fifo.qpl->page_buses,
+					info->iov[i].iov_offset,
+					info->iov[i].iov_len);
+		copy_offset += info->iov[i].iov_len;
+	}
+
+	return 1 + mtd_desc_nr + payload_nfrags;
+}
+
+static int gve_tx_add_skb_no_copy(struct gve_priv *priv, struct gve_tx_ring *tx,
+				  struct sk_buff *skb)
+{
+	const struct skb_shared_info *shinfo = skb_shinfo(skb);
+	int hlen, num_descriptors, l4_hdr_offset;
+	union gve_tx_desc *pkt_desc, *mtd_desc, *seg_desc;
+	struct gve_tx_buffer_state *info;
+	int mtd_desc_nr = !!skb->l4_hash;
+	bool is_gso = skb_is_gso(skb);
+	u32 idx = tx->req & tx->mask;
+	u64 addr;
+	u32 len;
+	int i;
+
+	info = &tx->info[idx];
+	pkt_desc = &tx->desc[idx];
+
+	l4_hdr_offset = skb_checksum_start_offset(skb);
+	/* If the skb is gso, then we want only up to the tcp header in the first segment
+	 * to efficiently replicate on each segment otherwise we want the linear portion
+	 * of the skb (which will contain the checksum because skb->csum_start and
+	 * skb->csum_offset are given relative to skb->head) in the first segment.
+	 */
+	hlen = is_gso ? l4_hdr_offset + tcp_hdrlen(skb) : skb_headlen(skb);
+	len = skb_headlen(skb);
+
+	info->skb =  skb;
+
+	addr = dma_map_single(tx->dev, skb->data, len, DMA_TO_DEVICE);
+	if (unlikely(dma_mapping_error(tx->dev, addr))) {
+		tx->dma_mapping_error++;
+		goto drop;
+	}
+	dma_unmap_len_set(info, len, len);
+	dma_unmap_addr_set(info, dma, addr);
+
+	num_descriptors = 1 + shinfo->nr_frags;
+	if (hlen < len)
+		num_descriptors++;
+	if (mtd_desc_nr)
+		num_descriptors++;
+
+	gve_tx_fill_pkt_desc(pkt_desc, skb->csum_offset, skb->ip_summed,
+			     is_gso, l4_hdr_offset,
+			     num_descriptors, hlen, addr, skb->len);
+
+	if (mtd_desc_nr) {
+		idx = (idx + 1) & tx->mask;
+		mtd_desc = &tx->desc[idx];
+		gve_tx_fill_mtd_desc(mtd_desc, skb);
+	}
+
+	if (hlen < len) {
+		/* For gso the rest of the linear portion of the skb needs to
+		 * be in its own descriptor.
+		 */
+		len -= hlen;
+		addr += hlen;
+		idx = (idx + 1) & tx->mask;
+		seg_desc = &tx->desc[idx];
+		gve_tx_fill_seg_desc(seg_desc, skb_network_offset(skb),
+				     skb_shinfo(skb)->gso_size,
+				     skb_is_gso_v6(skb), is_gso, len, addr);
+	}
+
+	for (i = 0; i < shinfo->nr_frags; i++) {
+		const skb_frag_t *frag = &shinfo->frags[i];
+
+		idx = (idx + 1) & tx->mask;
+		seg_desc = &tx->desc[idx];
+		len = skb_frag_size(frag);
+		addr = skb_frag_dma_map(tx->dev, frag, 0, len, DMA_TO_DEVICE);
+		if (unlikely(dma_mapping_error(tx->dev, addr))) {
+			tx->dma_mapping_error++;
+			goto unmap_drop;
+		}
+		tx->info[idx].skb = NULL;
+		dma_unmap_len_set(&tx->info[idx], len, len);
+		dma_unmap_addr_set(&tx->info[idx], dma, addr);
+
+		gve_tx_fill_seg_desc(seg_desc, skb_network_offset(skb),
+				     skb_shinfo(skb)->gso_size,
+				     skb_is_gso_v6(skb), is_gso, len, addr);
+	}
+
+	return num_descriptors;
+
+unmap_drop:
+	i += num_descriptors - shinfo->nr_frags;
+	while (i--) {
+		/* Skip metadata descriptor, if set */
+		if (i == 1 && mtd_desc_nr == 1)
+			continue;
+		idx--;
+		gve_tx_unmap_buf(tx->dev, &tx->info[idx & tx->mask]);
+	}
+drop:
+	tx->dropped_pkt++;
+	return 0;
+}
+
+netdev_tx_t gve_tx(struct sk_buff *skb, struct net_device *dev)
+{
+	struct gve_priv *priv = netdev_priv(dev);
+	struct gve_tx_ring *tx;
+	int nsegs;
+
+	WARN(skb_get_queue_mapping(skb) >= priv->tx_cfg.num_queues,
+	     "skb queue index out of range");
+	tx = &priv->tx[skb_get_queue_mapping(skb)];
+	if (unlikely(gve_maybe_stop_tx(priv, tx, skb))) {
+		/* We need to ring the txq doorbell -- we have stopped the Tx
+		 * queue for want of resources, but prior calls to gve_tx()
+		 * may have added descriptors without ringing the doorbell.
+		 */
+
+		gve_tx_put_doorbell(priv, tx->q_resources, tx->req);
+		return NETDEV_TX_BUSY;
+	}
+	if (tx->raw_addressing)
+		nsegs = gve_tx_add_skb_no_copy(priv, tx, skb);
+	else
+		nsegs = gve_tx_add_skb_copy(priv, tx, skb);
+
+	/* If the packet is getting sent, we need to update the skb */
+	if (nsegs) {
+		netdev_tx_sent_queue(tx->netdev_txq, skb->len);
+		skb_tx_timestamp(skb);
+		tx->req += nsegs;
+	} else {
+		dev_kfree_skb_any(skb);
+	}
+
+	if (!netif_xmit_stopped(tx->netdev_txq) && netdev_xmit_more())
+		return NETDEV_TX_OK;
+
+	/* Give packets to NIC. Even if this packet failed to send the doorbell
+	 * might need to be rung because of xmit_more.
+	 */
+	gve_tx_put_doorbell(priv, tx->q_resources, tx->req);
+	return NETDEV_TX_OK;
+}
+
+static int gve_tx_fill_xdp(struct gve_priv *priv, struct gve_tx_ring *tx,
+			   void *data, int len, void *frame_p, bool is_xsk)
+{
+	int pad, nfrags, ndescs, iovi, offset;
+	struct gve_tx_buffer_state *info;
+	u32 reqi = tx->req;
+
+	pad = gve_tx_fifo_pad_alloc_one_frag(&tx->tx_fifo, len);
+	if (pad >= GVE_GQ_TX_MIN_PKT_DESC_BYTES)
+		pad = 0;
+	info = &tx->info[reqi & tx->mask];
+	info->xdp_frame = frame_p;
+	info->xdp.size = len;
+	info->xdp.is_xsk = is_xsk;
+
+	nfrags = gve_tx_alloc_fifo(&tx->tx_fifo, pad + len,
+				   &info->iov[0]);
+	iovi = pad > 0;
+	ndescs = nfrags - iovi;
+	offset = 0;
+
+	while (iovi < nfrags) {
+		if (!offset)
+			gve_tx_fill_pkt_desc(&tx->desc[reqi & tx->mask], 0,
+					     CHECKSUM_NONE, false, 0, ndescs,
+					     info->iov[iovi].iov_len,
+					     info->iov[iovi].iov_offset, len);
+		else
+			gve_tx_fill_seg_desc(&tx->desc[reqi & tx->mask],
+					     0, 0, false, false,
+					     info->iov[iovi].iov_len,
+					     info->iov[iovi].iov_offset);
+
+		memcpy(tx->tx_fifo.base + info->iov[iovi].iov_offset,
+		       data + offset, info->iov[iovi].iov_len);
+		gve_dma_sync_for_device(&priv->pdev->dev,
+					tx->tx_fifo.qpl->page_buses,
+					info->iov[iovi].iov_offset,
+					info->iov[iovi].iov_len);
+		offset += info->iov[iovi].iov_len;
+		iovi++;
+		reqi++;
+	}
+
+	return ndescs;
+}
+
+int gve_xdp_xmit_gqi(struct net_device *dev, int n, struct xdp_frame **frames,
+		     u32 flags)
+{
+	struct gve_priv *priv = netdev_priv(dev);
+	struct gve_tx_ring *tx;
+	int i, err = 0, qid;
+
+	if (unlikely(flags & ~XDP_XMIT_FLAGS_MASK) || !priv->xdp_prog)
+		return -EINVAL;
+
+	if (!gve_get_napi_enabled(priv))
+		return -ENETDOWN;
+
+	qid = gve_xdp_tx_queue_id(priv,
+				  smp_processor_id() % priv->tx_cfg.num_xdp_queues);
+
+	tx = &priv->tx[qid];
+
+	spin_lock(&tx->xdp_lock);
+	for (i = 0; i < n; i++) {
+		err = gve_xdp_xmit_one(priv, tx, frames[i]->data,
+				       frames[i]->len, frames[i]);
+		if (err)
+			break;
+	}
+
+	if (flags & XDP_XMIT_FLUSH)
+		gve_tx_put_doorbell(priv, tx->q_resources, tx->req);
+
+	spin_unlock(&tx->xdp_lock);
+
+	u64_stats_update_begin(&tx->statss);
+	tx->xdp_xmit += n;
+	tx->xdp_xmit_errors += n - i;
+	u64_stats_update_end(&tx->statss);
+
+	return i ? i : err;
+}
+
+int gve_xdp_xmit_one(struct gve_priv *priv, struct gve_tx_ring *tx,
+		     void *data, int len, void *frame_p)
+{
+	int nsegs;
+
+	if (!gve_can_tx(tx, len + GVE_GQ_TX_MIN_PKT_DESC_BYTES - 1))
+		return -EBUSY;
+
+	nsegs = gve_tx_fill_xdp(priv, tx, data, len, frame_p, false);
+	tx->req += nsegs;
+
+	return 0;
+}
+
+#define GVE_TX_START_THRESH	4096
+
+static int gve_clean_tx_done(struct gve_priv *priv, struct gve_tx_ring *tx,
+			     u32 to_do, bool try_to_wake)
+{
+	struct gve_tx_buffer_state *info;
+	u64 pkts = 0, bytes = 0;
+	size_t space_freed = 0;
+	struct sk_buff *skb;
+	u32 idx;
+	int j;
+
+	for (j = 0; j < to_do; j++) {
+		idx = tx->done & tx->mask;
+		netif_info(priv, tx_done, priv->dev,
+			   "[%d] %s: idx=%d (req=%u done=%u)\n",
+			   tx->q_num, __func__, idx, tx->req, tx->done);
+		info = &tx->info[idx];
+		skb = info->skb;
+
+		/* Unmap the buffer */
+		if (tx->raw_addressing)
+			gve_tx_unmap_buf(tx->dev, info);
+		tx->done++;
+		/* Mark as free */
+		if (skb) {
+			info->skb = NULL;
+			bytes += skb->len;
+			pkts++;
+			dev_consume_skb_any(skb);
+			if (tx->raw_addressing)
+				continue;
+			space_freed += gve_tx_clear_buffer_state(info);
+		}
+	}
+
+	if (!tx->raw_addressing)
+		gve_tx_free_fifo(&tx->tx_fifo, space_freed);
+	u64_stats_update_begin(&tx->statss);
+	tx->bytes_done += bytes;
+	tx->pkt_done += pkts;
+	u64_stats_update_end(&tx->statss);
+	netdev_tx_completed_queue(tx->netdev_txq, pkts, bytes);
+
+	/* start the queue if we've stopped it */
+#ifndef CONFIG_BQL
+	/* Make sure that the doorbells are synced */
+	smp_mb();
+#endif
+	if (try_to_wake && netif_tx_queue_stopped(tx->netdev_txq) &&
+	    likely(gve_can_tx(tx, GVE_TX_START_THRESH))) {
+		tx->wake_queue++;
+		netif_tx_wake_queue(tx->netdev_txq);
+	}
+
+	return pkts;
+}
+
+u32 gve_tx_load_event_counter(struct gve_priv *priv,
+			      struct gve_tx_ring *tx)
+{
+	u32 counter_index = be32_to_cpu(tx->q_resources->counter_index);
+	__be32 counter = READ_ONCE(priv->counter_array[counter_index]);
+
+	return be32_to_cpu(counter);
+}
+
+static int gve_xsk_tx(struct gve_priv *priv, struct gve_tx_ring *tx,
+		      int budget)
+{
+	struct xdp_desc desc;
+	int sent = 0, nsegs;
+	void *data;
+
+	spin_lock(&tx->xdp_lock);
+	while (sent < budget) {
+		if (!gve_can_tx(tx, GVE_TX_START_THRESH) ||
+		    !xsk_tx_peek_desc(tx->xsk_pool, &desc))
+			goto out;
+
+		data = xsk_buff_raw_get_data(tx->xsk_pool, desc.addr);
+		nsegs = gve_tx_fill_xdp(priv, tx, data, desc.len, NULL, true);
+		tx->req += nsegs;
+		sent++;
+	}
+out:
+	if (sent > 0) {
+		gve_tx_put_doorbell(priv, tx->q_resources, tx->req);
+		xsk_tx_release(tx->xsk_pool);
+	}
+	spin_unlock(&tx->xdp_lock);
+	return sent;
+}
+
+int gve_xsk_tx_poll(struct gve_notify_block *rx_block, int budget)
+{
+	struct gve_rx_ring *rx = rx_block->rx;
+	struct gve_priv *priv = rx->gve;
+	struct gve_tx_ring *tx;
+	int sent = 0;
+
+	tx = &priv->tx[gve_xdp_tx_queue_id(priv, rx->q_num)];
+	if (tx->xsk_pool) {
+		sent = gve_xsk_tx(priv, tx, budget);
+
+		u64_stats_update_begin(&tx->statss);
+		tx->xdp_xsk_sent += sent;
+		u64_stats_update_end(&tx->statss);
+		if (xsk_uses_need_wakeup(tx->xsk_pool))
+			xsk_set_tx_need_wakeup(tx->xsk_pool);
+	}
+
+	return sent;
+}
+
+bool gve_xdp_poll(struct gve_notify_block *block, int budget)
+{
+	struct gve_priv *priv = block->priv;
+	struct gve_tx_ring *tx = block->tx;
+	u32 nic_done;
+	u32 to_do;
+
+	/* Find out how much work there is to be done */
+	nic_done = gve_tx_load_event_counter(priv, tx);
+	to_do = min_t(u32, (nic_done - tx->done), budget);
+	gve_clean_xdp_done(priv, tx, to_do);
+
+	/* If we still have work we want to repoll */
+	return nic_done != tx->done;
+}
+
+bool gve_tx_poll(struct gve_notify_block *block, int budget)
+{
+	struct gve_priv *priv = block->priv;
+	struct gve_tx_ring *tx = block->tx;
+	u32 nic_done;
+	u32 to_do;
+
+	/* If budget is 0, do all the work */
+	if (budget == 0)
+		budget = INT_MAX;
+
+	/* In TX path, it may try to clean completed pkts in order to xmit,
+	 * to avoid cleaning conflict, use spin_lock(), it yields better
+	 * concurrency between xmit/clean than netif's lock.
+	 */
+	spin_lock(&tx->clean_lock);
+	/* Find out how much work there is to be done */
+	nic_done = gve_tx_load_event_counter(priv, tx);
+	to_do = min_t(u32, (nic_done - tx->done), budget);
+	gve_clean_tx_done(priv, tx, to_do, true);
+	spin_unlock(&tx->clean_lock);
+	/* If we still have work we want to repoll */
+	return nic_done != tx->done;
+}
+
+bool gve_tx_clean_pending(struct gve_priv *priv, struct gve_tx_ring *tx)
+{
+	u32 nic_done = gve_tx_load_event_counter(priv, tx);
+
+	return nic_done != tx->done;
+}
diff --git a/drivers/net/ethernet/google/gve/gve_tx_dqo.c b/drivers/net/ethernet/google/gve/gve_tx_dqo.c
new file mode 100644
index 000000000000..6f1d515673d2
--- /dev/null
+++ b/drivers/net/ethernet/google/gve/gve_tx_dqo.c
@@ -0,0 +1,1609 @@
+// SPDX-License-Identifier: (GPL-2.0 OR MIT)
+/* Google virtual Ethernet (gve) driver
+ *
+ * Copyright (C) 2015-2021 Google, Inc.
+ */
+
+#include "gve.h"
+#include "gve_adminq.h"
+#include "gve_utils.h"
+#include "gve_dqo.h"
+#include <net/ip.h>
+#include <linux/bpf.h>
+#include <linux/tcp.h>
+#include <linux/slab.h>
+#include <linux/skbuff.h>
+#include <net/xdp_sock_drv.h>
+
+/* Returns true if tx_bufs are available. */
+static bool gve_has_free_tx_qpl_bufs(struct gve_tx_ring *tx, int count)
+{
+	int num_avail;
+
+	if (!tx->dqo.qpl)
+		return true;
+
+	num_avail = tx->dqo.num_tx_qpl_bufs -
+		(tx->dqo_tx.alloc_tx_qpl_buf_cnt -
+		 tx->dqo_tx.free_tx_qpl_buf_cnt);
+
+	if (count <= num_avail)
+		return true;
+
+	/* Update cached value from dqo_compl. */
+	tx->dqo_tx.free_tx_qpl_buf_cnt =
+		atomic_read_acquire(&tx->dqo_compl.free_tx_qpl_buf_cnt);
+
+	num_avail = tx->dqo.num_tx_qpl_bufs -
+		(tx->dqo_tx.alloc_tx_qpl_buf_cnt -
+		 tx->dqo_tx.free_tx_qpl_buf_cnt);
+
+	return count <= num_avail;
+}
+
+static s16
+gve_alloc_tx_qpl_buf(struct gve_tx_ring *tx)
+{
+	s16 index;
+
+	index = tx->dqo_tx.free_tx_qpl_buf_head;
+
+	/* No TX buffers available, try to steal the list from the
+	 * completion handler.
+	 */
+	if (unlikely(index == -1)) {
+		tx->dqo_tx.free_tx_qpl_buf_head =
+			atomic_xchg(&tx->dqo_compl.free_tx_qpl_buf_head, -1);
+		index = tx->dqo_tx.free_tx_qpl_buf_head;
+
+		if (unlikely(index == -1))
+			return index;
+	}
+
+	/* Remove TX buf from free list */
+	tx->dqo_tx.free_tx_qpl_buf_head = tx->dqo.tx_qpl_buf_next[index];
+
+	return index;
+}
+
+static void
+gve_free_tx_qpl_bufs(struct gve_tx_ring *tx,
+		     struct gve_tx_pending_packet_dqo *pkt)
+{
+	s16 index;
+	int i;
+
+	if (!pkt->num_bufs)
+		return;
+
+	index = pkt->tx_qpl_buf_ids[0];
+	/* Create a linked list of buffers to be added to the free list */
+	for (i = 1; i < pkt->num_bufs; i++) {
+		tx->dqo.tx_qpl_buf_next[index] = pkt->tx_qpl_buf_ids[i];
+		index = pkt->tx_qpl_buf_ids[i];
+	}
+
+	while (true) {
+		s16 old_head = atomic_read_acquire(&tx->dqo_compl.free_tx_qpl_buf_head);
+
+		tx->dqo.tx_qpl_buf_next[index] = old_head;
+		if (atomic_cmpxchg(&tx->dqo_compl.free_tx_qpl_buf_head,
+				   old_head,
+				   pkt->tx_qpl_buf_ids[0]) == old_head) {
+			break;
+		}
+	}
+
+	atomic_add(pkt->num_bufs, &tx->dqo_compl.free_tx_qpl_buf_cnt);
+	pkt->num_bufs = 0;
+}
+
+/* Returns true if a gve_tx_pending_packet_dqo object is available. */
+static bool gve_has_pending_packet(struct gve_tx_ring *tx)
+{
+	/* Check TX path's list. */
+	if (tx->dqo_tx.free_pending_packets != -1)
+		return true;
+
+	/* Check completion handler's list. */
+	if (atomic_read_acquire(&tx->dqo_compl.free_pending_packets) != -1)
+		return true;
+
+	return false;
+}
+
+void gve_xdp_tx_flush_dqo(struct gve_priv *priv, u32 xdp_qid)
+{
+	u32 tx_qid = gve_xdp_tx_queue_id(priv, xdp_qid);
+	struct gve_tx_ring *tx = &priv->tx[tx_qid];
+
+	gve_tx_put_doorbell_dqo(priv, tx->q_resources, tx->dqo_tx.tail);
+}
+
+static struct gve_tx_pending_packet_dqo *
+gve_alloc_pending_packet(struct gve_tx_ring *tx)
+{
+	struct gve_tx_pending_packet_dqo *pending_packet;
+	s16 index;
+
+	index = tx->dqo_tx.free_pending_packets;
+
+	/* No pending_packets available, try to steal the list from the
+	 * completion handler.
+	 */
+	if (unlikely(index == -1)) {
+		tx->dqo_tx.free_pending_packets =
+			atomic_xchg(&tx->dqo_compl.free_pending_packets, -1);
+		index = tx->dqo_tx.free_pending_packets;
+
+		if (unlikely(index == -1))
+			return NULL;
+	}
+
+	pending_packet = &tx->dqo.pending_packets[index];
+
+	/* Remove pending_packet from free list */
+	tx->dqo_tx.free_pending_packets = pending_packet->next;
+	pending_packet->state = GVE_PACKET_STATE_PENDING_DATA_COMPL;
+
+	return pending_packet;
+}
+
+static void
+gve_free_pending_packet(struct gve_tx_ring *tx,
+			struct gve_tx_pending_packet_dqo *pending_packet)
+{
+	s16 index = pending_packet - tx->dqo.pending_packets;
+
+	pending_packet->state = GVE_PACKET_STATE_UNALLOCATED;
+	while (true) {
+		s16 old_head = atomic_read_acquire(&tx->dqo_compl.free_pending_packets);
+
+		pending_packet->next = old_head;
+		if (atomic_cmpxchg(&tx->dqo_compl.free_pending_packets,
+				   old_head, index) == old_head) {
+			break;
+		}
+	}
+}
+
+/* gve_tx_free_desc - Cleans up all pending tx requests and buffers.
+ */
+static void gve_tx_clean_pending_packets(struct gve_tx_ring *tx)
+{
+	int i;
+
+	for (i = 0; i < tx->dqo.num_pending_packets; i++) {
+		struct gve_tx_pending_packet_dqo *cur_state =
+			&tx->dqo.pending_packets[i];
+		int j;
+
+		for (j = 0; j < cur_state->num_bufs; j++) {
+			if (j == 0) {
+				dma_unmap_single(tx->dev,
+					dma_unmap_addr(cur_state, dma[j]),
+					dma_unmap_len(cur_state, len[j]),
+					DMA_TO_DEVICE);
+			} else {
+				dma_unmap_page(tx->dev,
+					dma_unmap_addr(cur_state, dma[j]),
+					dma_unmap_len(cur_state, len[j]),
+					DMA_TO_DEVICE);
+			}
+		}
+		if (cur_state->skb) {
+			dev_consume_skb_any(cur_state->skb);
+			cur_state->skb = NULL;
+		}
+	}
+}
+
+void gve_tx_stop_ring_dqo(struct gve_priv *priv, int idx)
+{
+	int ntfy_idx = gve_tx_idx_to_ntfy(priv, idx);
+	struct gve_tx_ring *tx = &priv->tx[idx];
+
+	if (!gve_tx_was_added_to_block(priv, idx))
+		return;
+
+	gve_remove_napi(priv, ntfy_idx);
+	gve_clean_tx_done_dqo(priv, tx, /*napi=*/NULL);
+	if (tx->netdev_txq)
+		netdev_tx_reset_queue(tx->netdev_txq);
+	gve_tx_clean_pending_packets(tx);
+	gve_tx_remove_from_block(priv, idx);
+}
+
+static void gve_tx_free_ring_dqo(struct gve_priv *priv, struct gve_tx_ring *tx,
+				 struct gve_tx_alloc_rings_cfg *cfg)
+{
+	struct device *hdev = &priv->pdev->dev;
+	int idx = tx->q_num;
+	size_t bytes;
+	u32 qpl_id;
+
+	if (tx->q_resources) {
+		dma_free_coherent(hdev, sizeof(*tx->q_resources),
+				  tx->q_resources, tx->q_resources_bus);
+		tx->q_resources = NULL;
+	}
+
+	if (tx->dqo.compl_ring) {
+		bytes = sizeof(tx->dqo.compl_ring[0]) *
+			(tx->dqo.complq_mask + 1);
+		dma_free_coherent(hdev, bytes, tx->dqo.compl_ring,
+				  tx->complq_bus_dqo);
+		tx->dqo.compl_ring = NULL;
+	}
+
+	if (tx->dqo.tx_ring) {
+		bytes = sizeof(tx->dqo.tx_ring[0]) * (tx->mask + 1);
+		dma_free_coherent(hdev, bytes, tx->dqo.tx_ring, tx->bus);
+		tx->dqo.tx_ring = NULL;
+	}
+
+	kvfree(tx->dqo.xsk_reorder_queue);
+	tx->dqo.xsk_reorder_queue = NULL;
+
+	kvfree(tx->dqo.pending_packets);
+	tx->dqo.pending_packets = NULL;
+
+	kvfree(tx->dqo.tx_qpl_buf_next);
+	tx->dqo.tx_qpl_buf_next = NULL;
+
+	if (tx->dqo.qpl) {
+		qpl_id = gve_tx_qpl_id(priv, tx->q_num);
+		gve_free_queue_page_list(priv, tx->dqo.qpl, qpl_id);
+		tx->dqo.qpl = NULL;
+	}
+
+	netif_dbg(priv, drv, priv->dev, "freed tx queue %d\n", idx);
+}
+
+static int gve_tx_qpl_buf_init(struct gve_tx_ring *tx)
+{
+	int num_tx_qpl_bufs = GVE_TX_BUFS_PER_PAGE_DQO *
+		tx->dqo.qpl->num_entries;
+	int i;
+
+	tx->dqo.tx_qpl_buf_next = kvcalloc(num_tx_qpl_bufs,
+					   sizeof(tx->dqo.tx_qpl_buf_next[0]),
+					   GFP_KERNEL);
+	if (!tx->dqo.tx_qpl_buf_next)
+		return -ENOMEM;
+
+	tx->dqo.num_tx_qpl_bufs = num_tx_qpl_bufs;
+
+	/* Generate free TX buf list */
+	for (i = 0; i < num_tx_qpl_bufs - 1; i++)
+		tx->dqo.tx_qpl_buf_next[i] = i + 1;
+	tx->dqo.tx_qpl_buf_next[num_tx_qpl_bufs - 1] = -1;
+
+	atomic_set_release(&tx->dqo_compl.free_tx_qpl_buf_head, -1);
+	return 0;
+}
+
+void gve_tx_start_ring_dqo(struct gve_priv *priv, int idx)
+{
+	int ntfy_idx = gve_tx_idx_to_ntfy(priv, idx);
+	struct gve_tx_ring *tx = &priv->tx[idx];
+
+	gve_tx_add_to_block(priv, idx);
+
+	if (idx < priv->tx_cfg.num_queues)
+		tx->netdev_txq = netdev_get_tx_queue(priv->dev, idx);
+	gve_add_napi(priv, ntfy_idx, gve_napi_poll_dqo);
+}
+
+static int gve_tx_alloc_ring_dqo(struct gve_priv *priv,
+				 struct gve_tx_alloc_rings_cfg *cfg,
+				 struct gve_tx_ring *tx,
+				 int idx)
+{
+	struct device *hdev = &priv->pdev->dev;
+	int num_pending_packets;
+	int qpl_page_cnt;
+	size_t bytes;
+	u32 qpl_id;
+	int i;
+
+	memset(tx, 0, sizeof(*tx));
+	tx->q_num = idx;
+	tx->dev = hdev;
+	spin_lock_init(&tx->dqo_tx.xdp_lock);
+	atomic_set_release(&tx->dqo_compl.hw_tx_head, 0);
+
+	/* Queue sizes must be a power of 2 */
+	tx->mask = cfg->ring_size - 1;
+	tx->dqo.complq_mask = tx->mask;
+
+	/* The max number of pending packets determines the maximum number of
+	 * descriptors which maybe written to the completion queue.
+	 *
+	 * We must set the number small enough to make sure we never overrun the
+	 * completion queue.
+	 */
+	num_pending_packets = tx->dqo.complq_mask + 1;
+
+	/* Reserve space for descriptor completions, which will be reported at
+	 * most every GVE_TX_MIN_RE_INTERVAL packets.
+	 */
+	num_pending_packets -=
+		(tx->dqo.complq_mask + 1) / GVE_TX_MIN_RE_INTERVAL;
+
+	/* Each packet may have at most 2 buffer completions if it receives both
+	 * a miss and reinjection completion.
+	 */
+	num_pending_packets /= 2;
+
+	tx->dqo.num_pending_packets = min_t(int, num_pending_packets, S16_MAX);
+	tx->dqo.pending_packets = kvcalloc(tx->dqo.num_pending_packets,
+					   sizeof(tx->dqo.pending_packets[0]),
+					   GFP_KERNEL);
+	if (!tx->dqo.pending_packets)
+		goto err;
+
+	/* Set up linked list of pending packets */
+	for (i = 0; i < tx->dqo.num_pending_packets - 1; i++)
+		tx->dqo.pending_packets[i].next = i + 1;
+
+	tx->dqo.pending_packets[tx->dqo.num_pending_packets - 1].next = -1;
+	atomic_set_release(&tx->dqo_compl.free_pending_packets, -1);
+
+	/* Only alloc xsk pool for XDP queues */
+	if (idx >= cfg->qcfg->num_queues && cfg->num_xdp_rings) {
+		tx->dqo.xsk_reorder_queue =
+			kvcalloc(tx->dqo.complq_mask + 1,
+				 sizeof(tx->dqo.xsk_reorder_queue[0]),
+				 GFP_KERNEL);
+		if (!tx->dqo.xsk_reorder_queue)
+			goto err;
+	}
+
+	tx->dqo_compl.miss_completions.head = -1;
+	tx->dqo_compl.miss_completions.tail = -1;
+	tx->dqo_compl.timed_out_completions.head = -1;
+	tx->dqo_compl.timed_out_completions.tail = -1;
+
+	bytes = sizeof(tx->dqo.tx_ring[0]) * (tx->mask + 1);
+	tx->dqo.tx_ring = dma_alloc_coherent(hdev, bytes, &tx->bus, GFP_KERNEL);
+	if (!tx->dqo.tx_ring)
+		goto err;
+
+	bytes = sizeof(tx->dqo.compl_ring[0]) * (tx->dqo.complq_mask + 1);
+	tx->dqo.compl_ring = dma_alloc_coherent(hdev, bytes,
+						&tx->complq_bus_dqo,
+						GFP_KERNEL);
+	if (!tx->dqo.compl_ring)
+		goto err;
+
+	tx->q_resources = dma_alloc_coherent(hdev, sizeof(*tx->q_resources),
+					     &tx->q_resources_bus, GFP_KERNEL);
+	if (!tx->q_resources)
+		goto err;
+
+	if (!cfg->raw_addressing) {
+		qpl_id = gve_tx_qpl_id(priv, tx->q_num);
+		qpl_page_cnt = priv->tx_pages_per_qpl;
+
+		tx->dqo.qpl = gve_alloc_queue_page_list(priv, qpl_id,
+							qpl_page_cnt);
+		if (!tx->dqo.qpl)
+			goto err;
+
+		if (gve_tx_qpl_buf_init(tx))
+			goto err;
+	}
+
+	return 0;
+
+err:
+	gve_tx_free_ring_dqo(priv, tx, cfg);
+	return -ENOMEM;
+}
+
+int gve_tx_alloc_rings_dqo(struct gve_priv *priv,
+			   struct gve_tx_alloc_rings_cfg *cfg)
+{
+	struct gve_tx_ring *tx = cfg->tx;
+	int total_queues;
+	int err = 0;
+	int i, j;
+
+	total_queues = cfg->qcfg->num_queues + cfg->num_xdp_rings;
+	if (total_queues > cfg->qcfg->max_queues) {
+		netif_err(priv, drv, priv->dev,
+			  "Cannot alloc more than the max num of Tx rings\n");
+		return -EINVAL;
+	}
+
+	tx = kvcalloc(cfg->qcfg->max_queues, sizeof(struct gve_tx_ring),
+		      GFP_KERNEL);
+	if (!tx)
+		return -ENOMEM;
+
+	for (i = 0; i < total_queues; i++) {
+		err = gve_tx_alloc_ring_dqo(priv, cfg, &tx[i], i);
+		if (err) {
+			netif_err(priv, drv, priv->dev,
+				  "Failed to alloc tx ring=%d: err=%d\n",
+				  i, err);
+			goto err;
+		}
+	}
+
+	cfg->tx = tx;
+	return 0;
+
+err:
+	for (j = 0; j < i; j++)
+		gve_tx_free_ring_dqo(priv, &tx[j], cfg);
+	kvfree(tx);
+	return err;
+}
+
+void gve_tx_free_rings_dqo(struct gve_priv *priv,
+			   struct gve_tx_alloc_rings_cfg *cfg)
+{
+	struct gve_tx_ring *tx = cfg->tx;
+	int i;
+
+	if (!tx)
+		return;
+
+	for (i = 0; i < cfg->qcfg->num_queues + cfg->qcfg->num_xdp_queues; i++)
+		gve_tx_free_ring_dqo(priv, &tx[i], cfg);
+
+	kvfree(tx);
+	cfg->tx = NULL;
+}
+
+/* Returns the number of slots available in the ring */
+static u32 num_avail_tx_slots(const struct gve_tx_ring *tx)
+{
+	u32 num_used = (tx->dqo_tx.tail - tx->dqo_tx.head) & tx->mask;
+
+	return tx->mask - num_used;
+}
+
+/* Checks if the requested number of slots are available in the ring */
+static bool gve_has_tx_slots_available(struct gve_tx_ring *tx, u32 slots_req)
+{
+	u32 num_avail = num_avail_tx_slots(tx);
+
+	slots_req += GVE_TX_MIN_DESC_PREVENT_CACHE_OVERLAP;
+
+	if (num_avail >= slots_req)
+		return true;
+
+	/* Update cached TX head pointer */
+	tx->dqo_tx.head = atomic_read_acquire(&tx->dqo_compl.hw_tx_head);
+
+	return num_avail_tx_slots(tx) >= slots_req;
+}
+
+static bool gve_has_avail_slots_tx_dqo(struct gve_tx_ring *tx,
+				       int desc_count, int buf_count)
+{
+	return gve_has_pending_packet(tx) &&
+		gve_has_tx_slots_available(tx, desc_count) &&
+		gve_has_free_tx_qpl_bufs(tx, buf_count);
+}
+
+/* Stops the queue if available descriptors is less than 'count'.
+ * Return: 0 if stop is not required.
+ */
+static int gve_maybe_stop_tx_dqo(struct gve_tx_ring *tx,
+				 int desc_count, int buf_count)
+{
+	if (likely(gve_has_avail_slots_tx_dqo(tx, desc_count, buf_count)))
+		return 0;
+
+	/* No space, so stop the queue */
+	tx->stop_queue++;
+	netif_tx_stop_queue(tx->netdev_txq);
+
+	/* Sync with restarting queue in `gve_tx_poll_dqo()` */
+	mb();
+
+	/* After stopping queue, check if we can transmit again in order to
+	 * avoid TOCTOU bug.
+	 */
+	if (likely(!gve_has_avail_slots_tx_dqo(tx, desc_count, buf_count)))
+		return -EBUSY;
+
+	netif_tx_start_queue(tx->netdev_txq);
+	tx->wake_queue++;
+	return 0;
+}
+
+static void gve_extract_tx_metadata_dqo(const struct sk_buff *skb,
+					struct gve_tx_metadata_dqo *metadata)
+{
+	memset(metadata, 0, sizeof(*metadata));
+	metadata->version = GVE_TX_METADATA_VERSION_DQO;
+
+	if (skb->l4_hash) {
+		u16 path_hash = skb->hash ^ (skb->hash >> 16);
+
+		path_hash &= (1 << 15) - 1;
+		if (unlikely(path_hash == 0))
+			path_hash = ~path_hash;
+
+		metadata->path_hash = path_hash;
+	}
+}
+
+static void gve_tx_fill_pkt_desc_dqo(struct gve_tx_ring *tx, u32 *desc_idx,
+				     bool enable_csum, u32 len, u64 addr,
+				     s16 compl_tag, bool eop, bool is_gso)
+{
+	while (len > 0) {
+		struct gve_tx_pkt_desc_dqo *desc =
+			&tx->dqo.tx_ring[*desc_idx].pkt;
+		u32 cur_len = min_t(u32, len, GVE_TX_MAX_BUF_SIZE_DQO);
+		bool cur_eop = eop && cur_len == len;
+
+		*desc = (struct gve_tx_pkt_desc_dqo){
+			.buf_addr = cpu_to_le64(addr),
+			.dtype = GVE_TX_PKT_DESC_DTYPE_DQO,
+			.end_of_packet = cur_eop,
+			.checksum_offload_enable = enable_csum,
+			.compl_tag = cpu_to_le16(compl_tag),
+			.buf_size = cur_len,
+		};
+
+		addr += cur_len;
+		len -= cur_len;
+		*desc_idx = (*desc_idx + 1) & tx->mask;
+	}
+}
+
+/* Validates and prepares `skb` for TSO.
+ *
+ * Returns header length, or < 0 if invalid.
+ */
+static int gve_prep_tso(struct sk_buff *skb)
+{
+	struct tcphdr *tcp;
+	int header_len;
+	u32 paylen;
+	int err;
+
+	/* Note: HW requires MSS (gso_size) to be <= 9728 and the total length
+	 * of the TSO to be <= 262143.
+	 *
+	 * However, we don't validate these because:
+	 * - Hypervisor enforces a limit of 9K MTU
+	 * - Kernel will not produce a TSO larger than 64k
+	 */
+
+	if (unlikely(skb_shinfo(skb)->gso_size < GVE_TX_MIN_TSO_MSS_DQO))
+		return -1;
+
+	if (!(skb_shinfo(skb)->gso_type & (SKB_GSO_TCPV4 | SKB_GSO_TCPV6)))
+		return -EINVAL;
+
+	/* Needed because we will modify header. */
+	err = skb_cow_head(skb, 0);
+	if (err < 0)
+		return err;
+
+	tcp = tcp_hdr(skb);
+	paylen = skb->len - skb_transport_offset(skb);
+	csum_replace_by_diff(&tcp->check, (__force __wsum)htonl(paylen));
+	header_len = skb_tcp_all_headers(skb);
+
+	if (unlikely(header_len > GVE_TX_MAX_HDR_SIZE_DQO))
+		return -EINVAL;
+
+	return header_len;
+}
+
+static void gve_tx_fill_tso_ctx_desc(struct gve_tx_tso_context_desc_dqo *desc,
+				     const struct sk_buff *skb,
+				     const struct gve_tx_metadata_dqo *metadata,
+				     int header_len)
+{
+	*desc = (struct gve_tx_tso_context_desc_dqo){
+		.header_len = header_len,
+		.cmd_dtype = {
+			.dtype = GVE_TX_TSO_CTX_DESC_DTYPE_DQO,
+			.tso = 1,
+		},
+		.flex0 = metadata->bytes[0],
+		.flex5 = metadata->bytes[5],
+		.flex6 = metadata->bytes[6],
+		.flex7 = metadata->bytes[7],
+		.flex8 = metadata->bytes[8],
+		.flex9 = metadata->bytes[9],
+		.flex10 = metadata->bytes[10],
+		.flex11 = metadata->bytes[11],
+	};
+	desc->tso_total_len = skb->len - header_len;
+	desc->mss = skb_shinfo(skb)->gso_size;
+}
+
+static void
+gve_tx_fill_general_ctx_desc(struct gve_tx_general_context_desc_dqo *desc,
+			     const struct gve_tx_metadata_dqo *metadata)
+{
+	*desc = (struct gve_tx_general_context_desc_dqo){
+		.flex0 = metadata->bytes[0],
+		.flex1 = metadata->bytes[1],
+		.flex2 = metadata->bytes[2],
+		.flex3 = metadata->bytes[3],
+		.flex4 = metadata->bytes[4],
+		.flex5 = metadata->bytes[5],
+		.flex6 = metadata->bytes[6],
+		.flex7 = metadata->bytes[7],
+		.flex8 = metadata->bytes[8],
+		.flex9 = metadata->bytes[9],
+		.flex10 = metadata->bytes[10],
+		.flex11 = metadata->bytes[11],
+		.cmd_dtype = {.dtype = GVE_TX_GENERAL_CTX_DESC_DTYPE_DQO},
+	};
+}
+
+static void gve_tx_update_tail(struct gve_tx_ring *tx, u32 desc_idx)
+{
+	u32 last_desc_idx = (desc_idx - 1) & tx->mask;
+	u32 last_report_event_interval =
+			(last_desc_idx - tx->dqo_tx.last_re_idx) & tx->mask;
+
+	/* Commit the changes to our state */
+	tx->dqo_tx.tail = desc_idx;
+
+	/* Request a descriptor completion on the last descriptor of the
+	 * packet if we are allowed to by the HW enforced interval.
+	 */
+
+	if (unlikely(last_report_event_interval >= GVE_TX_MIN_RE_INTERVAL)) {
+		tx->dqo.tx_ring[last_desc_idx].pkt.report_event = true;
+		tx->dqo_tx.last_re_idx = last_desc_idx;
+	}
+}
+
+static int gve_tx_add_skb_no_copy_dqo(struct gve_tx_ring *tx,
+				      struct sk_buff *skb,
+				      struct gve_tx_pending_packet_dqo *pkt,
+				      s16 completion_tag,
+				      u32 *desc_idx,
+				      bool is_gso)
+{
+	bool enable_csum = skb->ip_summed == CHECKSUM_PARTIAL;
+	const struct skb_shared_info *shinfo = skb_shinfo(skb);
+	int i;
+
+	/* Note: HW requires that the size of a non-TSO packet be within the
+	 * range of [17, 9728].
+	 *
+	 * We don't double check because
+	 * - We limited `netdev->min_mtu` to ETH_MIN_MTU.
+	 * - Hypervisor won't allow MTU larger than 9216.
+	 */
+
+	pkt->num_bufs = 0;
+	/* Map the linear portion of skb */
+	{
+		u32 len = skb_headlen(skb);
+		dma_addr_t addr;
+
+		addr = dma_map_single(tx->dev, skb->data, len, DMA_TO_DEVICE);
+		if (unlikely(dma_mapping_error(tx->dev, addr)))
+			goto err;
+
+		dma_unmap_len_set(pkt, len[pkt->num_bufs], len);
+		dma_unmap_addr_set(pkt, dma[pkt->num_bufs], addr);
+		++pkt->num_bufs;
+
+		gve_tx_fill_pkt_desc_dqo(tx, desc_idx, enable_csum, len, addr,
+					 completion_tag,
+					 /*eop=*/shinfo->nr_frags == 0, is_gso);
+	}
+
+	for (i = 0; i < shinfo->nr_frags; i++) {
+		const skb_frag_t *frag = &shinfo->frags[i];
+		bool is_eop = i == (shinfo->nr_frags - 1);
+		u32 len = skb_frag_size(frag);
+		dma_addr_t addr;
+
+		addr = skb_frag_dma_map(tx->dev, frag, 0, len, DMA_TO_DEVICE);
+		if (unlikely(dma_mapping_error(tx->dev, addr)))
+			goto err;
+
+		dma_unmap_len_set(pkt, len[pkt->num_bufs], len);
+		netmem_dma_unmap_addr_set(skb_frag_netmem(frag), pkt,
+					  dma[pkt->num_bufs], addr);
+		++pkt->num_bufs;
+
+		gve_tx_fill_pkt_desc_dqo(tx, desc_idx, enable_csum, len, addr,
+					 completion_tag, is_eop, is_gso);
+	}
+
+	return 0;
+err:
+	for (i = 0; i < pkt->num_bufs; i++) {
+		if (i == 0) {
+			dma_unmap_single(tx->dev,
+					 dma_unmap_addr(pkt, dma[i]),
+					 dma_unmap_len(pkt, len[i]),
+					 DMA_TO_DEVICE);
+		} else {
+			dma_unmap_page(tx->dev,
+				       dma_unmap_addr(pkt, dma[i]),
+				       dma_unmap_len(pkt, len[i]),
+				       DMA_TO_DEVICE);
+		}
+	}
+	pkt->num_bufs = 0;
+	return -1;
+}
+
+/* Tx buffer i corresponds to
+ * qpl_page_id = i / GVE_TX_BUFS_PER_PAGE_DQO
+ * qpl_page_offset = (i % GVE_TX_BUFS_PER_PAGE_DQO) * GVE_TX_BUF_SIZE_DQO
+ */
+static void gve_tx_buf_get_addr(struct gve_tx_ring *tx,
+				s16 index,
+				void **va, dma_addr_t *dma_addr)
+{
+	int page_id = index >> (PAGE_SHIFT - GVE_TX_BUF_SHIFT_DQO);
+	int offset = (index & (GVE_TX_BUFS_PER_PAGE_DQO - 1)) << GVE_TX_BUF_SHIFT_DQO;
+
+	*va = page_address(tx->dqo.qpl->pages[page_id]) + offset;
+	*dma_addr = tx->dqo.qpl->page_buses[page_id] + offset;
+}
+
+static int gve_tx_add_skb_copy_dqo(struct gve_tx_ring *tx,
+				   struct sk_buff *skb,
+				   struct gve_tx_pending_packet_dqo *pkt,
+				   s16 completion_tag,
+				   u32 *desc_idx,
+				   bool is_gso)
+{
+	bool enable_csum = skb->ip_summed == CHECKSUM_PARTIAL;
+	u32 copy_offset = 0;
+	dma_addr_t dma_addr;
+	u32 copy_len;
+	s16 index;
+	void *va;
+
+	/* Break the packet into buffer size chunks */
+	pkt->num_bufs = 0;
+	while (copy_offset < skb->len) {
+		index = gve_alloc_tx_qpl_buf(tx);
+		if (unlikely(index == -1))
+			goto err;
+
+		gve_tx_buf_get_addr(tx, index, &va, &dma_addr);
+		copy_len = min_t(u32, GVE_TX_BUF_SIZE_DQO,
+				 skb->len - copy_offset);
+		skb_copy_bits(skb, copy_offset, va, copy_len);
+
+		copy_offset += copy_len;
+		dma_sync_single_for_device(tx->dev, dma_addr,
+					   copy_len, DMA_TO_DEVICE);
+		gve_tx_fill_pkt_desc_dqo(tx, desc_idx, enable_csum,
+					 copy_len,
+					 dma_addr,
+					 completion_tag,
+					 copy_offset == skb->len,
+					 is_gso);
+
+		pkt->tx_qpl_buf_ids[pkt->num_bufs] = index;
+		++tx->dqo_tx.alloc_tx_qpl_buf_cnt;
+		++pkt->num_bufs;
+	}
+
+	return 0;
+err:
+	/* Should not be here if gve_has_free_tx_qpl_bufs() check is correct */
+	gve_free_tx_qpl_bufs(tx, pkt);
+	return -ENOMEM;
+}
+
+/* Returns 0 on success, or < 0 on error.
+ *
+ * Before this function is called, the caller must ensure
+ * gve_has_pending_packet(tx) returns true.
+ */
+static int gve_tx_add_skb_dqo(struct gve_tx_ring *tx,
+			      struct sk_buff *skb)
+{
+	const bool is_gso = skb_is_gso(skb);
+	u32 desc_idx = tx->dqo_tx.tail;
+	struct gve_tx_pending_packet_dqo *pkt;
+	struct gve_tx_metadata_dqo metadata;
+	s16 completion_tag;
+
+	pkt = gve_alloc_pending_packet(tx);
+	if (!pkt)
+		return -ENOMEM;
+
+	pkt->skb = skb;
+	pkt->type = GVE_TX_PENDING_PACKET_DQO_SKB;
+	completion_tag = pkt - tx->dqo.pending_packets;
+
+	gve_extract_tx_metadata_dqo(skb, &metadata);
+	if (is_gso) {
+		int header_len = gve_prep_tso(skb);
+
+		if (unlikely(header_len < 0))
+			goto err;
+
+		gve_tx_fill_tso_ctx_desc(&tx->dqo.tx_ring[desc_idx].tso_ctx,
+					 skb, &metadata, header_len);
+		desc_idx = (desc_idx + 1) & tx->mask;
+	}
+
+	gve_tx_fill_general_ctx_desc(&tx->dqo.tx_ring[desc_idx].general_ctx,
+				     &metadata);
+	desc_idx = (desc_idx + 1) & tx->mask;
+
+	if (tx->dqo.qpl) {
+		if (gve_tx_add_skb_copy_dqo(tx, skb, pkt,
+					    completion_tag,
+					    &desc_idx, is_gso))
+			goto err;
+	}  else {
+		if (gve_tx_add_skb_no_copy_dqo(tx, skb, pkt,
+					       completion_tag,
+					       &desc_idx, is_gso))
+			goto err;
+	}
+
+	tx->dqo_tx.posted_packet_desc_cnt += pkt->num_bufs;
+
+	gve_tx_update_tail(tx, desc_idx);
+	return 0;
+
+err:
+	pkt->skb = NULL;
+	gve_free_pending_packet(tx, pkt);
+
+	return -1;
+}
+
+static int gve_num_descs_per_buf(size_t size)
+{
+	return DIV_ROUND_UP(size, GVE_TX_MAX_BUF_SIZE_DQO);
+}
+
+static int gve_num_buffer_descs_needed(const struct sk_buff *skb)
+{
+	const struct skb_shared_info *shinfo = skb_shinfo(skb);
+	int num_descs;
+	int i;
+
+	num_descs = gve_num_descs_per_buf(skb_headlen(skb));
+
+	for (i = 0; i < shinfo->nr_frags; i++) {
+		unsigned int frag_size = skb_frag_size(&shinfo->frags[i]);
+
+		num_descs += gve_num_descs_per_buf(frag_size);
+	}
+
+	return num_descs;
+}
+
+/* Returns true if HW is capable of sending TSO represented by `skb`.
+ *
+ * Each segment must not span more than GVE_TX_MAX_DATA_DESCS buffers.
+ * - The header is counted as one buffer for every single segment.
+ * - A buffer which is split between two segments is counted for both.
+ * - If a buffer contains both header and payload, it is counted as two buffers.
+ */
+static bool gve_can_send_tso(const struct sk_buff *skb)
+{
+	const int max_bufs_per_seg = GVE_TX_MAX_DATA_DESCS - 1;
+	const struct skb_shared_info *shinfo = skb_shinfo(skb);
+	const int header_len = skb_tcp_all_headers(skb);
+	const int gso_size = shinfo->gso_size;
+	int cur_seg_num_bufs;
+	int prev_frag_size;
+	int cur_seg_size;
+	int i;
+
+	cur_seg_size = skb_headlen(skb) - header_len;
+	prev_frag_size = skb_headlen(skb);
+	cur_seg_num_bufs = cur_seg_size > 0;
+
+	for (i = 0; i < shinfo->nr_frags; i++) {
+		if (cur_seg_size >= gso_size) {
+			cur_seg_size %= gso_size;
+			cur_seg_num_bufs = cur_seg_size > 0;
+
+			if (prev_frag_size > GVE_TX_MAX_BUF_SIZE_DQO) {
+				int prev_frag_remain = prev_frag_size %
+					GVE_TX_MAX_BUF_SIZE_DQO;
+
+				/* If the last descriptor of the previous frag
+				 * is less than cur_seg_size, the segment will
+				 * span two descriptors in the previous frag.
+				 * Since max gso size (9728) is less than
+				 * GVE_TX_MAX_BUF_SIZE_DQO, it is impossible
+				 * for the segment to span more than two
+				 * descriptors.
+				 */
+				if (prev_frag_remain &&
+				    cur_seg_size > prev_frag_remain)
+					cur_seg_num_bufs++;
+			}
+		}
+
+		if (unlikely(++cur_seg_num_bufs > max_bufs_per_seg))
+			return false;
+
+		prev_frag_size = skb_frag_size(&shinfo->frags[i]);
+		cur_seg_size += prev_frag_size;
+	}
+
+	return true;
+}
+
+netdev_features_t gve_features_check_dqo(struct sk_buff *skb,
+					 struct net_device *dev,
+					 netdev_features_t features)
+{
+	if (skb_is_gso(skb) && !gve_can_send_tso(skb))
+		return features & ~NETIF_F_GSO_MASK;
+
+	return features;
+}
+
+/* Attempt to transmit specified SKB.
+ *
+ * Returns 0 if the SKB was transmitted or dropped.
+ * Returns -1 if there is not currently enough space to transmit the SKB.
+ */
+static int gve_try_tx_skb(struct gve_priv *priv, struct gve_tx_ring *tx,
+			  struct sk_buff *skb)
+{
+	int num_buffer_descs;
+	int total_num_descs;
+
+	if (skb_is_gso(skb) && unlikely(ipv6_hopopt_jumbo_remove(skb)))
+		goto drop;
+
+	if (tx->dqo.qpl) {
+		/* We do not need to verify the number of buffers used per
+		 * packet or per segment in case of TSO as with 2K size buffers
+		 * none of the TX packet rules would be violated.
+		 *
+		 * gve_can_send_tso() checks that each TCP segment of gso_size is
+		 * not distributed over more than 9 SKB frags..
+		 */
+		num_buffer_descs = DIV_ROUND_UP(skb->len, GVE_TX_BUF_SIZE_DQO);
+	} else {
+		num_buffer_descs = gve_num_buffer_descs_needed(skb);
+		if (!skb_is_gso(skb)) {
+			if (unlikely(num_buffer_descs > GVE_TX_MAX_DATA_DESCS)) {
+				if (unlikely(skb_linearize(skb) < 0))
+					goto drop;
+
+				num_buffer_descs = 1;
+			}
+		}
+	}
+
+	/* Metadata + (optional TSO) + data descriptors. */
+	total_num_descs = 1 + skb_is_gso(skb) + num_buffer_descs;
+	if (unlikely(gve_maybe_stop_tx_dqo(tx, total_num_descs,
+					   num_buffer_descs))) {
+		return -1;
+	}
+
+	if (unlikely(gve_tx_add_skb_dqo(tx, skb) < 0))
+		goto drop;
+
+	netdev_tx_sent_queue(tx->netdev_txq, skb->len);
+	skb_tx_timestamp(skb);
+	return 0;
+
+drop:
+	tx->dropped_pkt++;
+	dev_kfree_skb_any(skb);
+	return 0;
+}
+
+static void gve_xsk_reorder_queue_push_dqo(struct gve_tx_ring *tx,
+					   u16 completion_tag)
+{
+	u32 tail = atomic_read(&tx->dqo_tx.xsk_reorder_queue_tail);
+
+	tx->dqo.xsk_reorder_queue[tail] = completion_tag;
+	tail = (tail + 1) & tx->dqo.complq_mask;
+	atomic_set_release(&tx->dqo_tx.xsk_reorder_queue_tail, tail);
+}
+
+static struct gve_tx_pending_packet_dqo *
+gve_xsk_reorder_queue_head(struct gve_tx_ring *tx)
+{
+	u32 head = tx->dqo_compl.xsk_reorder_queue_head;
+
+	if (head == tx->dqo_compl.xsk_reorder_queue_tail) {
+		tx->dqo_compl.xsk_reorder_queue_tail =
+			atomic_read_acquire(&tx->dqo_tx.xsk_reorder_queue_tail);
+
+		if (head == tx->dqo_compl.xsk_reorder_queue_tail)
+			return NULL;
+	}
+
+	return &tx->dqo.pending_packets[tx->dqo.xsk_reorder_queue[head]];
+}
+
+static void gve_xsk_reorder_queue_pop_dqo(struct gve_tx_ring *tx)
+{
+	tx->dqo_compl.xsk_reorder_queue_head++;
+	tx->dqo_compl.xsk_reorder_queue_head &= tx->dqo.complq_mask;
+}
+
+/* Transmit a given skb and ring the doorbell. */
+netdev_tx_t gve_tx_dqo(struct sk_buff *skb, struct net_device *dev)
+{
+	struct gve_priv *priv = netdev_priv(dev);
+	struct gve_tx_ring *tx;
+
+	tx = &priv->tx[skb_get_queue_mapping(skb)];
+	if (unlikely(gve_try_tx_skb(priv, tx, skb) < 0)) {
+		/* We need to ring the txq doorbell -- we have stopped the Tx
+		 * queue for want of resources, but prior calls to gve_tx()
+		 * may have added descriptors without ringing the doorbell.
+		 */
+		gve_tx_put_doorbell_dqo(priv, tx->q_resources, tx->dqo_tx.tail);
+		return NETDEV_TX_BUSY;
+	}
+
+	if (!netif_xmit_stopped(tx->netdev_txq) && netdev_xmit_more())
+		return NETDEV_TX_OK;
+
+	gve_tx_put_doorbell_dqo(priv, tx->q_resources, tx->dqo_tx.tail);
+	return NETDEV_TX_OK;
+}
+
+static bool gve_xsk_tx_dqo(struct gve_priv *priv, struct gve_tx_ring *tx,
+			   int budget)
+{
+	struct xsk_buff_pool *pool = tx->xsk_pool;
+	struct xdp_desc desc;
+	bool repoll = false;
+	int sent = 0;
+
+	spin_lock(&tx->dqo_tx.xdp_lock);
+	for (; sent < budget; sent++) {
+		struct gve_tx_pending_packet_dqo *pkt;
+		s16 completion_tag;
+		dma_addr_t addr;
+		u32 desc_idx;
+
+		if (unlikely(!gve_has_avail_slots_tx_dqo(tx, 1, 1))) {
+			repoll = true;
+			break;
+		}
+
+		if (!xsk_tx_peek_desc(pool, &desc))
+			break;
+
+		pkt = gve_alloc_pending_packet(tx);
+		pkt->type = GVE_TX_PENDING_PACKET_DQO_XSK;
+		pkt->num_bufs = 0;
+		completion_tag = pkt - tx->dqo.pending_packets;
+
+		addr = xsk_buff_raw_get_dma(pool, desc.addr);
+		xsk_buff_raw_dma_sync_for_device(pool, addr, desc.len);
+
+		desc_idx = tx->dqo_tx.tail;
+		gve_tx_fill_pkt_desc_dqo(tx, &desc_idx,
+					 true, desc.len,
+					 addr, completion_tag, true,
+					 false);
+		++pkt->num_bufs;
+		gve_tx_update_tail(tx, desc_idx);
+		tx->dqo_tx.posted_packet_desc_cnt += pkt->num_bufs;
+		gve_xsk_reorder_queue_push_dqo(tx, completion_tag);
+	}
+
+	if (sent) {
+		gve_tx_put_doorbell_dqo(priv, tx->q_resources, tx->dqo_tx.tail);
+		xsk_tx_release(pool);
+	}
+
+	spin_unlock(&tx->dqo_tx.xdp_lock);
+
+	u64_stats_update_begin(&tx->statss);
+	tx->xdp_xsk_sent += sent;
+	u64_stats_update_end(&tx->statss);
+
+	return (sent == budget) || repoll;
+}
+
+static void add_to_list(struct gve_tx_ring *tx, struct gve_index_list *list,
+			struct gve_tx_pending_packet_dqo *pending_packet)
+{
+	s16 old_tail, index;
+
+	index = pending_packet - tx->dqo.pending_packets;
+	old_tail = list->tail;
+	list->tail = index;
+	if (old_tail == -1)
+		list->head = index;
+	else
+		tx->dqo.pending_packets[old_tail].next = index;
+
+	pending_packet->next = -1;
+	pending_packet->prev = old_tail;
+}
+
+static void remove_from_list(struct gve_tx_ring *tx,
+			     struct gve_index_list *list,
+			     struct gve_tx_pending_packet_dqo *pkt)
+{
+	s16 prev_index, next_index;
+
+	prev_index = pkt->prev;
+	next_index = pkt->next;
+
+	if (prev_index == -1) {
+		/* Node is head */
+		list->head = next_index;
+	} else {
+		tx->dqo.pending_packets[prev_index].next = next_index;
+	}
+	if (next_index == -1) {
+		/* Node is tail */
+		list->tail = prev_index;
+	} else {
+		tx->dqo.pending_packets[next_index].prev = prev_index;
+	}
+}
+
+static void gve_unmap_packet(struct device *dev,
+			     struct gve_tx_pending_packet_dqo *pkt)
+{
+	int i;
+
+	/* SKB linear portion is guaranteed to be mapped */
+	dma_unmap_single(dev, dma_unmap_addr(pkt, dma[0]),
+			 dma_unmap_len(pkt, len[0]), DMA_TO_DEVICE);
+	for (i = 1; i < pkt->num_bufs; i++) {
+		netmem_dma_unmap_page_attrs(dev, dma_unmap_addr(pkt, dma[i]),
+					    dma_unmap_len(pkt, len[i]),
+					    DMA_TO_DEVICE, 0);
+	}
+	pkt->num_bufs = 0;
+}
+
+/* Completion types and expected behavior:
+ * No Miss compl + Packet compl = Packet completed normally.
+ * Miss compl + Re-inject compl = Packet completed normally.
+ * No Miss compl + Re-inject compl = Skipped i.e. packet not completed.
+ * Miss compl + Packet compl = Skipped i.e. packet not completed.
+ */
+static void gve_handle_packet_completion(struct gve_priv *priv,
+					 struct gve_tx_ring *tx, bool is_napi,
+					 u16 compl_tag, u64 *bytes, u64 *pkts,
+					 bool is_reinjection)
+{
+	struct gve_tx_pending_packet_dqo *pending_packet;
+
+	if (unlikely(compl_tag >= tx->dqo.num_pending_packets)) {
+		net_err_ratelimited("%s: Invalid TX completion tag: %d\n",
+				    priv->dev->name, (int)compl_tag);
+		return;
+	}
+
+	pending_packet = &tx->dqo.pending_packets[compl_tag];
+
+	if (unlikely(is_reinjection)) {
+		if (unlikely(pending_packet->state ==
+			     GVE_PACKET_STATE_TIMED_OUT_COMPL)) {
+			net_err_ratelimited("%s: Re-injection completion: %d received after timeout.\n",
+					    priv->dev->name, (int)compl_tag);
+			/* Packet was already completed as a result of timeout,
+			 * so just remove from list and free pending packet.
+			 */
+			remove_from_list(tx,
+					 &tx->dqo_compl.timed_out_completions,
+					 pending_packet);
+			gve_free_pending_packet(tx, pending_packet);
+			return;
+		}
+		if (unlikely(pending_packet->state !=
+			     GVE_PACKET_STATE_PENDING_REINJECT_COMPL)) {
+			/* No outstanding miss completion but packet allocated
+			 * implies packet receives a re-injection completion
+			 * without a prior miss completion. Return without
+			 * completing the packet.
+			 */
+			net_err_ratelimited("%s: Re-injection completion received without corresponding miss completion: %d\n",
+					    priv->dev->name, (int)compl_tag);
+			return;
+		}
+		remove_from_list(tx, &tx->dqo_compl.miss_completions,
+				 pending_packet);
+	} else {
+		/* Packet is allocated but not a pending data completion. */
+		if (unlikely(pending_packet->state !=
+			     GVE_PACKET_STATE_PENDING_DATA_COMPL)) {
+			net_err_ratelimited("%s: No pending data completion: %d\n",
+					    priv->dev->name, (int)compl_tag);
+			return;
+		}
+	}
+	tx->dqo_tx.completed_packet_desc_cnt += pending_packet->num_bufs;
+
+	switch (pending_packet->type) {
+	case GVE_TX_PENDING_PACKET_DQO_SKB:
+		if (tx->dqo.qpl)
+			gve_free_tx_qpl_bufs(tx, pending_packet);
+		else
+			gve_unmap_packet(tx->dev, pending_packet);
+		(*pkts)++;
+		*bytes += pending_packet->skb->len;
+
+		napi_consume_skb(pending_packet->skb, is_napi);
+		pending_packet->skb = NULL;
+		gve_free_pending_packet(tx, pending_packet);
+		break;
+	case GVE_TX_PENDING_PACKET_DQO_XDP_FRAME:
+		gve_unmap_packet(tx->dev, pending_packet);
+		(*pkts)++;
+		*bytes += pending_packet->xdpf->len;
+
+		xdp_return_frame(pending_packet->xdpf);
+		pending_packet->xdpf = NULL;
+		gve_free_pending_packet(tx, pending_packet);
+		break;
+	case GVE_TX_PENDING_PACKET_DQO_XSK:
+		pending_packet->state = GVE_PACKET_STATE_XSK_COMPLETE;
+		break;
+	default:
+		WARN_ON_ONCE(1);
+	}
+}
+
+static void gve_handle_miss_completion(struct gve_priv *priv,
+				       struct gve_tx_ring *tx, u16 compl_tag,
+				       u64 *bytes, u64 *pkts)
+{
+	struct gve_tx_pending_packet_dqo *pending_packet;
+
+	if (unlikely(compl_tag >= tx->dqo.num_pending_packets)) {
+		net_err_ratelimited("%s: Invalid TX completion tag: %d\n",
+				    priv->dev->name, (int)compl_tag);
+		return;
+	}
+
+	pending_packet = &tx->dqo.pending_packets[compl_tag];
+	if (unlikely(pending_packet->state !=
+				GVE_PACKET_STATE_PENDING_DATA_COMPL)) {
+		net_err_ratelimited("%s: Unexpected packet state: %d for completion tag : %d\n",
+				    priv->dev->name, (int)pending_packet->state,
+				    (int)compl_tag);
+		return;
+	}
+
+	pending_packet->state = GVE_PACKET_STATE_PENDING_REINJECT_COMPL;
+	/* jiffies can wraparound but time comparisons can handle overflows. */
+	pending_packet->timeout_jiffies =
+			jiffies +
+			secs_to_jiffies(GVE_REINJECT_COMPL_TIMEOUT);
+	add_to_list(tx, &tx->dqo_compl.miss_completions, pending_packet);
+
+	*bytes += pending_packet->skb->len;
+	(*pkts)++;
+}
+
+static void remove_miss_completions(struct gve_priv *priv,
+				    struct gve_tx_ring *tx)
+{
+	struct gve_tx_pending_packet_dqo *pending_packet;
+	s16 next_index;
+
+	next_index = tx->dqo_compl.miss_completions.head;
+	while (next_index != -1) {
+		pending_packet = &tx->dqo.pending_packets[next_index];
+		next_index = pending_packet->next;
+		/* Break early because packets should timeout in order. */
+		if (time_is_after_jiffies(pending_packet->timeout_jiffies))
+			break;
+
+		remove_from_list(tx, &tx->dqo_compl.miss_completions,
+				 pending_packet);
+		/* Unmap/free TX buffers and free skb but do not unallocate packet i.e.
+		 * the completion tag is not freed to ensure that the driver
+		 * can take appropriate action if a corresponding valid
+		 * completion is received later.
+		 */
+		if (tx->dqo.qpl)
+			gve_free_tx_qpl_bufs(tx, pending_packet);
+		else
+			gve_unmap_packet(tx->dev, pending_packet);
+
+		/* This indicates the packet was dropped. */
+		dev_kfree_skb_any(pending_packet->skb);
+		pending_packet->skb = NULL;
+		tx->dropped_pkt++;
+		net_err_ratelimited("%s: No reinjection completion was received for: %d.\n",
+				    priv->dev->name,
+				    (int)(pending_packet - tx->dqo.pending_packets));
+
+		pending_packet->state = GVE_PACKET_STATE_TIMED_OUT_COMPL;
+		pending_packet->timeout_jiffies =
+				jiffies +
+				secs_to_jiffies(GVE_DEALLOCATE_COMPL_TIMEOUT);
+		/* Maintain pending packet in another list so the packet can be
+		 * unallocated at a later time.
+		 */
+		add_to_list(tx, &tx->dqo_compl.timed_out_completions,
+			    pending_packet);
+	}
+}
+
+static void remove_timed_out_completions(struct gve_priv *priv,
+					 struct gve_tx_ring *tx)
+{
+	struct gve_tx_pending_packet_dqo *pending_packet;
+	s16 next_index;
+
+	next_index = tx->dqo_compl.timed_out_completions.head;
+	while (next_index != -1) {
+		pending_packet = &tx->dqo.pending_packets[next_index];
+		next_index = pending_packet->next;
+		/* Break early because packets should timeout in order. */
+		if (time_is_after_jiffies(pending_packet->timeout_jiffies))
+			break;
+
+		remove_from_list(tx, &tx->dqo_compl.timed_out_completions,
+				 pending_packet);
+
+		/* Need to count XSK packets in xsk_tx_completed. */
+		if (pending_packet->type == GVE_TX_PENDING_PACKET_DQO_XSK)
+			pending_packet->state = GVE_PACKET_STATE_XSK_COMPLETE;
+		else
+			gve_free_pending_packet(tx, pending_packet);
+	}
+}
+
+static void gve_tx_process_xsk_completions(struct gve_tx_ring *tx)
+{
+	u32 num_xsks = 0;
+
+	while (true) {
+		struct gve_tx_pending_packet_dqo *pending_packet =
+			gve_xsk_reorder_queue_head(tx);
+
+		if (!pending_packet ||
+		    pending_packet->state != GVE_PACKET_STATE_XSK_COMPLETE)
+			break;
+
+		num_xsks++;
+		gve_xsk_reorder_queue_pop_dqo(tx);
+		gve_free_pending_packet(tx, pending_packet);
+	}
+
+	if (num_xsks)
+		xsk_tx_completed(tx->xsk_pool, num_xsks);
+}
+
+int gve_clean_tx_done_dqo(struct gve_priv *priv, struct gve_tx_ring *tx,
+			  struct napi_struct *napi)
+{
+	u64 reinject_compl_bytes = 0;
+	u64 reinject_compl_pkts = 0;
+	int num_descs_cleaned = 0;
+	u64 miss_compl_bytes = 0;
+	u64 miss_compl_pkts = 0;
+	u64 pkt_compl_bytes = 0;
+	u64 pkt_compl_pkts = 0;
+
+	/* Limit in order to avoid blocking for too long */
+	while (!napi || pkt_compl_pkts < napi->weight) {
+		struct gve_tx_compl_desc *compl_desc =
+			&tx->dqo.compl_ring[tx->dqo_compl.head];
+		u16 type;
+
+		if (compl_desc->generation == tx->dqo_compl.cur_gen_bit)
+			break;
+
+		/* Prefetch the next descriptor. */
+		prefetch(&tx->dqo.compl_ring[(tx->dqo_compl.head + 1) &
+				tx->dqo.complq_mask]);
+
+		/* Do not read data until we own the descriptor */
+		dma_rmb();
+		type = compl_desc->type;
+
+		if (type == GVE_COMPL_TYPE_DQO_DESC) {
+			/* This is the last descriptor fetched by HW plus one */
+			u16 tx_head = le16_to_cpu(compl_desc->tx_head);
+
+			atomic_set_release(&tx->dqo_compl.hw_tx_head, tx_head);
+		} else if (type == GVE_COMPL_TYPE_DQO_PKT) {
+			u16 compl_tag = le16_to_cpu(compl_desc->completion_tag);
+			if (compl_tag & GVE_ALT_MISS_COMPL_BIT) {
+				compl_tag &= ~GVE_ALT_MISS_COMPL_BIT;
+				gve_handle_miss_completion(priv, tx, compl_tag,
+							   &miss_compl_bytes,
+							   &miss_compl_pkts);
+			} else {
+				gve_handle_packet_completion(priv, tx, !!napi,
+							     compl_tag,
+							     &pkt_compl_bytes,
+							     &pkt_compl_pkts,
+							     false);
+			}
+		} else if (type == GVE_COMPL_TYPE_DQO_MISS) {
+			u16 compl_tag = le16_to_cpu(compl_desc->completion_tag);
+
+			gve_handle_miss_completion(priv, tx, compl_tag,
+						   &miss_compl_bytes,
+						   &miss_compl_pkts);
+		} else if (type == GVE_COMPL_TYPE_DQO_REINJECTION) {
+			u16 compl_tag = le16_to_cpu(compl_desc->completion_tag);
+
+			gve_handle_packet_completion(priv, tx, !!napi,
+						     compl_tag,
+						     &reinject_compl_bytes,
+						     &reinject_compl_pkts,
+						     true);
+		}
+
+		tx->dqo_compl.head =
+			(tx->dqo_compl.head + 1) & tx->dqo.complq_mask;
+		/* Flip the generation bit when we wrap around */
+		tx->dqo_compl.cur_gen_bit ^= tx->dqo_compl.head == 0;
+		num_descs_cleaned++;
+	}
+
+	if (tx->netdev_txq)
+		netdev_tx_completed_queue(tx->netdev_txq,
+					  pkt_compl_pkts + miss_compl_pkts,
+					  pkt_compl_bytes + miss_compl_bytes);
+
+	remove_miss_completions(priv, tx);
+	remove_timed_out_completions(priv, tx);
+
+	if (tx->xsk_pool)
+		gve_tx_process_xsk_completions(tx);
+
+	u64_stats_update_begin(&tx->statss);
+	tx->bytes_done += pkt_compl_bytes + reinject_compl_bytes;
+	tx->pkt_done += pkt_compl_pkts + reinject_compl_pkts;
+	u64_stats_update_end(&tx->statss);
+	return num_descs_cleaned;
+}
+
+bool gve_tx_poll_dqo(struct gve_notify_block *block, bool do_clean)
+{
+	struct gve_tx_compl_desc *compl_desc;
+	struct gve_tx_ring *tx = block->tx;
+	struct gve_priv *priv = block->priv;
+
+	if (do_clean) {
+		int num_descs_cleaned = gve_clean_tx_done_dqo(priv, tx,
+							      &block->napi);
+
+		/* Sync with queue being stopped in `gve_maybe_stop_tx_dqo()` */
+		mb();
+
+		if (netif_tx_queue_stopped(tx->netdev_txq) &&
+		    num_descs_cleaned > 0) {
+			tx->wake_queue++;
+			netif_tx_wake_queue(tx->netdev_txq);
+		}
+	}
+
+	/* Return true if we still have work. */
+	compl_desc = &tx->dqo.compl_ring[tx->dqo_compl.head];
+	return compl_desc->generation != tx->dqo_compl.cur_gen_bit;
+}
+
+bool gve_xsk_tx_poll_dqo(struct gve_notify_block *rx_block, int budget)
+{
+	struct gve_rx_ring *rx = rx_block->rx;
+	struct gve_priv *priv = rx->gve;
+	struct gve_tx_ring *tx;
+
+	tx = &priv->tx[gve_xdp_tx_queue_id(priv, rx->q_num)];
+	if (tx->xsk_pool)
+		return gve_xsk_tx_dqo(priv, tx, budget);
+
+	return 0;
+}
+
+bool gve_xdp_poll_dqo(struct gve_notify_block *block)
+{
+	struct gve_tx_compl_desc *compl_desc;
+	struct gve_tx_ring *tx = block->tx;
+	struct gve_priv *priv = block->priv;
+
+	gve_clean_tx_done_dqo(priv, tx, &block->napi);
+
+	/* Return true if we still have work. */
+	compl_desc = &tx->dqo.compl_ring[tx->dqo_compl.head];
+	return compl_desc->generation != tx->dqo_compl.cur_gen_bit;
+}
+
+int gve_xdp_xmit_one_dqo(struct gve_priv *priv, struct gve_tx_ring *tx,
+			 struct xdp_frame *xdpf)
+{
+	struct gve_tx_pending_packet_dqo *pkt;
+	u32 desc_idx = tx->dqo_tx.tail;
+	s16 completion_tag;
+	int num_descs = 1;
+	dma_addr_t addr;
+	int err;
+
+	if (unlikely(!gve_has_tx_slots_available(tx, num_descs)))
+		return -EBUSY;
+
+	pkt = gve_alloc_pending_packet(tx);
+	if (unlikely(!pkt))
+		return -EBUSY;
+
+	pkt->type = GVE_TX_PENDING_PACKET_DQO_XDP_FRAME;
+	pkt->num_bufs = 0;
+	pkt->xdpf = xdpf;
+	completion_tag = pkt - tx->dqo.pending_packets;
+
+	/* Generate Packet Descriptor */
+	addr = dma_map_single(tx->dev, xdpf->data, xdpf->len, DMA_TO_DEVICE);
+	err = dma_mapping_error(tx->dev, addr);
+	if (unlikely(err))
+		goto err;
+
+	dma_unmap_len_set(pkt, len[pkt->num_bufs], xdpf->len);
+	dma_unmap_addr_set(pkt, dma[pkt->num_bufs], addr);
+	pkt->num_bufs++;
+
+	gve_tx_fill_pkt_desc_dqo(tx, &desc_idx,
+				 false, xdpf->len,
+				 addr, completion_tag, true,
+				 false);
+
+	gve_tx_update_tail(tx, desc_idx);
+	return 0;
+
+err:
+	pkt->xdpf = NULL;
+	pkt->num_bufs = 0;
+	gve_free_pending_packet(tx, pkt);
+	return err;
+}
+
+int gve_xdp_xmit_dqo(struct net_device *dev, int n, struct xdp_frame **frames,
+		     u32 flags)
+{
+	struct gve_priv *priv = netdev_priv(dev);
+	struct gve_tx_ring *tx;
+	int i, err = 0, qid;
+
+	if (unlikely(flags & ~XDP_XMIT_FLAGS_MASK))
+		return -EINVAL;
+
+	qid = gve_xdp_tx_queue_id(priv,
+				  smp_processor_id() % priv->tx_cfg.num_xdp_queues);
+
+	tx = &priv->tx[qid];
+
+	spin_lock(&tx->dqo_tx.xdp_lock);
+	for (i = 0; i < n; i++) {
+		err = gve_xdp_xmit_one_dqo(priv, tx, frames[i]);
+		if (err)
+			break;
+	}
+
+	if (flags & XDP_XMIT_FLUSH)
+		gve_tx_put_doorbell_dqo(priv, tx->q_resources, tx->dqo_tx.tail);
+
+	spin_unlock(&tx->dqo_tx.xdp_lock);
+
+	u64_stats_update_begin(&tx->statss);
+	tx->xdp_xmit += n;
+	tx->xdp_xmit_errors += n - i;
+	u64_stats_update_end(&tx->statss);
+
+	return i ? i : err;
+}
diff --git a/drivers/net/ethernet/google/gve/gve_utils.c b/drivers/net/ethernet/google/gve/gve_utils.c
new file mode 100644
index 000000000000..ace9b8698021
--- /dev/null
+++ b/drivers/net/ethernet/google/gve/gve_utils.c
@@ -0,0 +1,122 @@
+// SPDX-License-Identifier: (GPL-2.0 OR MIT)
+/* Google virtual Ethernet (gve) driver
+ *
+ * Copyright (C) 2015-2021 Google, Inc.
+ */
+
+#include "gve.h"
+#include "gve_adminq.h"
+#include "gve_utils.h"
+
+bool gve_tx_was_added_to_block(struct gve_priv *priv, int queue_idx)
+{
+	struct gve_notify_block *block =
+			&priv->ntfy_blocks[gve_tx_idx_to_ntfy(priv, queue_idx)];
+
+	return block->tx != NULL;
+}
+
+void gve_tx_remove_from_block(struct gve_priv *priv, int queue_idx)
+{
+	struct gve_notify_block *block =
+			&priv->ntfy_blocks[gve_tx_idx_to_ntfy(priv, queue_idx)];
+
+	block->tx = NULL;
+}
+
+void gve_tx_add_to_block(struct gve_priv *priv, int queue_idx)
+{
+	unsigned int active_cpus = min_t(int, priv->num_ntfy_blks / 2,
+					 num_online_cpus());
+	int ntfy_idx = gve_tx_idx_to_ntfy(priv, queue_idx);
+	struct gve_notify_block *block = &priv->ntfy_blocks[ntfy_idx];
+	struct gve_tx_ring *tx = &priv->tx[queue_idx];
+
+	block->tx = tx;
+	tx->ntfy_id = ntfy_idx;
+	netif_set_xps_queue(priv->dev, get_cpu_mask(ntfy_idx % active_cpus),
+			    queue_idx);
+}
+
+bool gve_rx_was_added_to_block(struct gve_priv *priv, int queue_idx)
+{
+	struct gve_notify_block *block =
+			&priv->ntfy_blocks[gve_rx_idx_to_ntfy(priv, queue_idx)];
+
+	return block->rx != NULL;
+}
+
+void gve_rx_remove_from_block(struct gve_priv *priv, int queue_idx)
+{
+	struct gve_notify_block *block =
+			&priv->ntfy_blocks[gve_rx_idx_to_ntfy(priv, queue_idx)];
+
+	block->rx = NULL;
+}
+
+void gve_rx_add_to_block(struct gve_priv *priv, int queue_idx)
+{
+	u32 ntfy_idx = gve_rx_idx_to_ntfy(priv, queue_idx);
+	struct gve_notify_block *block = &priv->ntfy_blocks[ntfy_idx];
+	struct gve_rx_ring *rx = &priv->rx[queue_idx];
+
+	block->rx = rx;
+	rx->ntfy_id = ntfy_idx;
+}
+
+struct sk_buff *gve_rx_copy_data(struct net_device *dev, struct napi_struct *napi,
+				 u8 *data, u16 len)
+{
+	struct sk_buff *skb;
+
+	skb = napi_alloc_skb(napi, len);
+	if (unlikely(!skb))
+		return NULL;
+
+	__skb_put(skb, len);
+	skb_copy_to_linear_data_offset(skb, 0, data, len);
+	skb->protocol = eth_type_trans(skb, dev);
+
+	return skb;
+}
+
+struct sk_buff *gve_rx_copy(struct net_device *dev, struct napi_struct *napi,
+			    struct gve_rx_slot_page_info *page_info, u16 len)
+{
+	void *va = page_info->page_address + page_info->page_offset +
+		page_info->pad;
+
+	return gve_rx_copy_data(dev, napi, va, len);
+}
+
+void gve_dec_pagecnt_bias(struct gve_rx_slot_page_info *page_info)
+{
+	page_info->pagecnt_bias--;
+	if (page_info->pagecnt_bias == 0) {
+		int pagecount = page_count(page_info->page);
+
+		/* If we have run out of bias - set it back up to INT_MAX
+		 * minus the existing refs.
+		 */
+		page_info->pagecnt_bias = INT_MAX - pagecount;
+
+		/* Set pagecount back up to max. */
+		page_ref_add(page_info->page, INT_MAX - pagecount);
+	}
+}
+
+void gve_add_napi(struct gve_priv *priv, int ntfy_idx,
+		  int (*gve_poll)(struct napi_struct *, int))
+{
+	struct gve_notify_block *block = &priv->ntfy_blocks[ntfy_idx];
+
+	netif_napi_add_locked(priv->dev, &block->napi, gve_poll);
+	netif_napi_set_irq_locked(&block->napi, block->irq);
+}
+
+void gve_remove_napi(struct gve_priv *priv, int ntfy_idx)
+{
+	struct gve_notify_block *block = &priv->ntfy_blocks[ntfy_idx];
+
+	netif_napi_del_locked(&block->napi);
+}
diff --git a/drivers/net/ethernet/google/gve/gve_utils.h b/drivers/net/ethernet/google/gve/gve_utils.h
new file mode 100644
index 000000000000..bf2e9a0adb36
--- /dev/null
+++ b/drivers/net/ethernet/google/gve/gve_utils.h
@@ -0,0 +1,35 @@
+/* SPDX-License-Identifier: (GPL-2.0 OR MIT)
+ * Google virtual Ethernet (gve) driver
+ *
+ * Copyright (C) 2015-2021 Google, Inc.
+ */
+
+#ifndef _GVE_UTILS_H
+#define _GVE_UTILS_H
+
+#include <linux/etherdevice.h>
+
+#include "gve.h"
+
+bool gve_tx_was_added_to_block(struct gve_priv *priv, int queue_idx);
+void gve_tx_remove_from_block(struct gve_priv *priv, int queue_idx);
+void gve_tx_add_to_block(struct gve_priv *priv, int queue_idx);
+
+bool gve_rx_was_added_to_block(struct gve_priv *priv, int queue_idx);
+void gve_rx_remove_from_block(struct gve_priv *priv, int queue_idx);
+void gve_rx_add_to_block(struct gve_priv *priv, int queue_idx);
+
+struct sk_buff *gve_rx_copy_data(struct net_device *dev, struct napi_struct *napi,
+				 u8 *data, u16 len);
+
+struct sk_buff *gve_rx_copy(struct net_device *dev, struct napi_struct *napi,
+			    struct gve_rx_slot_page_info *page_info, u16 len);
+
+/* Decrement pagecnt_bias. Set it back to INT_MAX if it reached zero. */
+void gve_dec_pagecnt_bias(struct gve_rx_slot_page_info *page_info);
+
+void gve_add_napi(struct gve_priv *priv, int ntfy_idx,
+		  int (*gve_poll)(struct napi_struct *, int));
+void gve_remove_napi(struct gve_priv *priv, int ntfy_idx);
+#endif /* _GVE_UTILS_H */
+
diff --git a/drivers/net/ethernet/hisilicon/Kconfig b/drivers/net/ethernet/hisilicon/Kconfig
index d11287e11371..65302c41bfb1 100644
--- a/drivers/net/ethernet/hisilicon/Kconfig
+++ b/drivers/net/ethernet/hisilicon/Kconfig
@@ -1,3 +1,4 @@
+# SPDX-License-Identifier: GPL-2.0-only
 #
 # HISILICON device configuration
 #
@@ -5,9 +6,8 @@
 config NET_VENDOR_HISILICON
 	bool "Hisilicon devices"
 	default y
-	depends on (OF || ACPI) && HAS_DMA
-	depends on ARM || ARM64 || COMPILE_TEST
-	---help---
+	depends on OF || ACPI
+	help
 	  If you have a network (Ethernet) card belonging to this class, say Y.
 
 	  Note that the answer to this question doesn't directly affect the
@@ -17,6 +17,8 @@ config NET_VENDOR_HISILICON
 
 if NET_VENDOR_HISILICON
 
+if ARM || ARM64 || COMPILE_TEST
+
 config HIX5HD2_GMAC
 	tristate "Hisilicon HIX5HD2 Family Network Device Support"
 	select PHYLIB
@@ -41,20 +43,30 @@ config HIP04_ETH
 	select MARVELL_PHY
 	select MFD_SYSCON
 	select HNS_MDIO
-	---help---
+	help
 	  If you wish to compile a kernel for a hardware with hisilicon p04 SoC and
 	  want to use the internal ethernet then you should answer Y to this.
 
+config HI13X1_GMAC
+	bool "Hisilicon HI13X1 Network Device Support"
+	depends on HIP04_ETH
+	help
+	  If you wish to compile a kernel for a hardware with hisilicon hi13x1_gamc
+	  then you should answer Y to this. This makes this driver suitable for use
+	  on certain boards such as the HI13X1.
+
+	  If you are unsure, say N.
+
 config HNS_MDIO
 	tristate
 	select PHYLIB
-	---help---
+	help
 	  This selects the HNS MDIO support. It is needed by HNS_DSAF to access
 	  the PHY
 
 config HNS
-	tristate "Hisilicon Network Subsystem Support (Framework)"
-	---help---
+	tristate
+	help
 	  This selects the framework support for Hisilicon Network Subsystem. It
 	  is needed by any driver which provides HNS acceleration engine or make
 	  use of the engine
@@ -63,7 +75,7 @@ config HNS_DSAF
 	tristate "Hisilicon HNS DSAF device Support"
 	select HNS
 	select HNS_MDIO
-	---help---
+	help
 	  This selects the DSAF (Distributed System Area Frabric) network
 	  acceleration engine support. The engine is used in Hisilicon hip05,
 	  Hi1610 and further ICT SoC
@@ -72,8 +84,77 @@ config HNS_ENET
 	tristate "Hisilicon HNS Ethernet Device Support"
 	select PHYLIB
 	select HNS
-	---help---
+	help
 	  This selects the general ethernet driver for HNS.  This module make
 	  use of any HNS AE driver, such as HNS_DSAF
 
+config HNS3
+	tristate "Hisilicon Network Subsystem Support HNS3 (Framework)"
+	depends on PCI
+	select NET_DEVLINK
+	select PAGE_POOL
+	help
+	  This selects the framework support for Hisilicon Network Subsystem 3.
+	  This layer facilitates clients like ENET, RoCE and user-space ethernet
+	  drivers(like ODP)to register with HNAE devices and their associated
+	  operations.
+
+if HNS3
+
+config HNS3_HCLGE
+	tristate "Hisilicon HNS3 HCLGE Acceleration Engine & Compatibility Layer Support"
+	default m
+	depends on PCI_MSI
+	depends on PTP_1588_CLOCK_OPTIONAL
+	help
+	  This selects the HNS3_HCLGE network acceleration engine & its hardware
+	  compatibility layer. The engine would be used in Hisilicon hip08 family of
+	  SoCs and further upcoming SoCs.
+
+config HNS3_DCB
+	bool "Hisilicon HNS3 Data Center Bridge Support"
+	default n
+	depends on HNS3_HCLGE && DCB
+	help
+	  Say Y here if you want to use Data Center Bridging (DCB) in the HNS3 driver.
+
+	  If unsure, say N.
+
+config HNS3_HCLGEVF
+	tristate "Hisilicon HNS3VF Acceleration Engine & Compatibility Layer Support"
+	depends on PCI_MSI
+	depends on HNS3_HCLGE
+	help
+	  This selects the HNS3 VF drivers network acceleration engine & its hardware
+	  compatibility layer. The engine would be used in Hisilicon hip08 family of
+	  SoCs and further upcoming SoCs.
+
+config HNS3_ENET
+	tristate "Hisilicon HNS3 Ethernet Device Support"
+	default m
+	depends on 64BIT && PCI
+	depends on INET
+	select DIMLIB
+	help
+	  This selects the Ethernet Driver for Hisilicon Network Subsystem 3 for hip08
+	  family of SoCs. This module depends upon HNAE3 driver to access the HNAE3
+	  devices and their associated operations.
+
+endif #HNS3
+
+endif # ARM || ARM64 || COMPILE_TEST
+
+config HIBMCGE
+	tristate "Hisilicon BMC Gigabit Ethernet Device Support"
+	depends on PCI && PCI_MSI
+	select PHYLIB
+	select MOTORCOMM_PHY
+	select REALTEK_PHY
+	help
+	  If you wish to compile a kernel for a BMC with HIBMC-xx_gmac
+	  then you should answer Y to this. This makes this driver suitable for use
+	  on certain boards such as the HIBMC-210.
+
+	  If you are unsure, say N.
+
 endif # NET_VENDOR_HISILICON
diff --git a/drivers/net/ethernet/hisilicon/Makefile b/drivers/net/ethernet/hisilicon/Makefile
index 8661695024dc..0e2cadfea8ff 100644
--- a/drivers/net/ethernet/hisilicon/Makefile
+++ b/drivers/net/ethernet/hisilicon/Makefile
@@ -1,3 +1,4 @@
+# SPDX-License-Identifier: GPL-2.0
 #
 # Makefile for the HISILICON network device drivers.
 #
@@ -6,4 +7,6 @@ obj-$(CONFIG_HIX5HD2_GMAC) += hix5hd2_gmac.o
 obj-$(CONFIG_HIP04_ETH) += hip04_eth.o
 obj-$(CONFIG_HNS_MDIO) += hns_mdio.o
 obj-$(CONFIG_HNS) += hns/
+obj-$(CONFIG_HNS3) += hns3/
 obj-$(CONFIG_HISI_FEMAC) += hisi_femac.o
+obj-$(CONFIG_HIBMCGE) += hibmcge/
diff --git a/drivers/net/ethernet/hisilicon/hibmcge/Makefile b/drivers/net/ethernet/hisilicon/hibmcge/Makefile
new file mode 100644
index 000000000000..1a9da564b306
--- /dev/null
+++ b/drivers/net/ethernet/hisilicon/hibmcge/Makefile
@@ -0,0 +1,9 @@
+# SPDX-License-Identifier: GPL-2.0+
+#
+# Makefile for the HISILICON BMC GE network device drivers.
+#
+
+obj-$(CONFIG_HIBMCGE) += hibmcge.o
+
+hibmcge-objs = hbg_main.o hbg_hw.o hbg_mdio.o hbg_irq.o hbg_txrx.o hbg_ethtool.o \
+		hbg_debugfs.o hbg_err.o hbg_diagnose.o
diff --git a/drivers/net/ethernet/hisilicon/hibmcge/hbg_common.h b/drivers/net/ethernet/hisilicon/hibmcge/hbg_common.h
new file mode 100644
index 000000000000..ea09a09c451b
--- /dev/null
+++ b/drivers/net/ethernet/hisilicon/hibmcge/hbg_common.h
@@ -0,0 +1,285 @@
+/* SPDX-License-Identifier: GPL-2.0+ */
+/* Copyright (c) 2024 Hisilicon Limited. */
+
+#ifndef __HBG_COMMON_H
+#define __HBG_COMMON_H
+
+#include <linux/ethtool.h>
+#include <linux/netdevice.h>
+#include <linux/pci.h>
+#include "hbg_reg.h"
+
+#define HBG_STATUS_DISABLE		0x0
+#define HBG_STATUS_ENABLE		0x1
+#define HBG_RX_SKIP1			0x00
+#define HBG_RX_SKIP2			0x01
+#define HBG_VECTOR_NUM			4
+#define HBG_PCU_CACHE_LINE_SIZE		32
+#define HBG_TX_TIMEOUT_BUF_LEN		1024
+#define HBG_RX_DESCR			0x01
+
+#define HBG_PACKET_HEAD_SIZE	((HBG_RX_SKIP1 + HBG_RX_SKIP2 + \
+				  HBG_RX_DESCR) * HBG_PCU_CACHE_LINE_SIZE)
+
+enum hbg_dir {
+	HBG_DIR_TX = 1 << 0,
+	HBG_DIR_RX = 1 << 1,
+	HBG_DIR_TX_RX = HBG_DIR_TX | HBG_DIR_RX,
+};
+
+enum hbg_tx_state {
+	HBG_TX_STATE_COMPLETE = 0, /* clear state, must fix to 0 */
+	HBG_TX_STATE_START,
+};
+
+enum hbg_nic_state {
+	HBG_NIC_STATE_EVENT_HANDLING = 0,
+	HBG_NIC_STATE_RESETTING,
+	HBG_NIC_STATE_RESET_FAIL,
+	HBG_NIC_STATE_NEED_RESET, /* trigger a reset in scheduled task */
+	HBG_NIC_STATE_NP_LINK_FAIL,
+};
+
+enum hbg_reset_type {
+	HBG_RESET_TYPE_NONE = 0,
+	HBG_RESET_TYPE_FLR,
+	HBG_RESET_TYPE_FUNCTION,
+};
+
+struct hbg_buffer {
+	u32 state;
+	dma_addr_t state_dma;
+
+	struct sk_buff *skb;
+	dma_addr_t skb_dma;
+	u32 skb_len;
+
+	enum hbg_dir dir;
+	struct hbg_ring *ring;
+	struct hbg_priv *priv;
+};
+
+struct hbg_ring {
+	struct hbg_buffer *queue;
+	dma_addr_t queue_dma;
+
+	union {
+		u32 head;
+		u32 ntc;
+	};
+	union {
+		u32 tail;
+		u32 ntu;
+	};
+	u32 len;
+
+	enum hbg_dir dir;
+	struct hbg_priv *priv;
+	struct napi_struct napi;
+	char *tout_log_buf; /* tx timeout log buffer */
+};
+
+enum hbg_hw_event_type {
+	HBG_HW_EVENT_NONE = 0,
+	HBG_HW_EVENT_INIT, /* driver is loading */
+	HBG_HW_EVENT_RESET,
+	HBG_HW_EVENT_CORE_RESET,
+};
+
+struct hbg_dev_specs {
+	u32 mac_id;
+	struct sockaddr mac_addr;
+	u32 phy_addr;
+	u32 mdio_frequency;
+	u32 rx_fifo_num;
+	u32 tx_fifo_num;
+	u32 vlan_layers;
+	u32 max_mtu;
+	u32 min_mtu;
+	u32 uc_mac_num;
+
+	u32 max_frame_len;
+	u32 rx_buf_size;
+};
+
+struct hbg_irq_info {
+	const char *name;
+	u32 mask;
+	bool re_enable;
+	bool need_print;
+	bool need_reset;
+
+	void (*irq_handle)(struct hbg_priv *priv,
+			   const struct hbg_irq_info *info);
+};
+
+struct hbg_vector {
+	char name[HBG_VECTOR_NUM][32];
+
+	u64 *stats_array;
+	const struct hbg_irq_info *info_array;
+	u32 info_array_len;
+};
+
+struct hbg_mac {
+	struct mii_bus *mdio_bus;
+	struct phy_device *phydev;
+	u8 phy_addr;
+
+	u32 speed;
+	u32 duplex;
+	u32 autoneg;
+	u32 link_status;
+	u32 pause_autoneg;
+};
+
+struct hbg_mac_table_entry {
+	u8 addr[ETH_ALEN];
+};
+
+struct hbg_mac_filter {
+	struct hbg_mac_table_entry *mac_table;
+	u32 table_max_len;
+	bool enabled;
+};
+
+/* saved for restore after rest */
+struct hbg_user_def {
+	struct ethtool_pauseparam pause_param;
+};
+
+struct hbg_stats {
+	u64 rx_desc_drop;
+	u64 rx_desc_l2_err_cnt;
+	u64 rx_desc_pkt_len_err_cnt;
+	u64 rx_desc_l3l4_err_cnt;
+	u64 rx_desc_l3_wrong_head_cnt;
+	u64 rx_desc_l3_csum_err_cnt;
+	u64 rx_desc_l3_len_err_cnt;
+	u64 rx_desc_l3_zero_ttl_cnt;
+	u64 rx_desc_l3_other_cnt;
+	u64 rx_desc_l4_err_cnt;
+	u64 rx_desc_l4_wrong_head_cnt;
+	u64 rx_desc_l4_len_err_cnt;
+	u64 rx_desc_l4_csum_err_cnt;
+	u64 rx_desc_l4_zero_port_num_cnt;
+	u64 rx_desc_l4_other_cnt;
+	u64 rx_desc_frag_cnt;
+	u64 rx_desc_ip_ver_err_cnt;
+	u64 rx_desc_ipv4_pkt_cnt;
+	u64 rx_desc_ipv6_pkt_cnt;
+	u64 rx_desc_no_ip_pkt_cnt;
+	u64 rx_desc_ip_pkt_cnt;
+	u64 rx_desc_tcp_pkt_cnt;
+	u64 rx_desc_udp_pkt_cnt;
+	u64 rx_desc_vlan_pkt_cnt;
+	u64 rx_desc_icmp_pkt_cnt;
+	u64 rx_desc_arp_pkt_cnt;
+	u64 rx_desc_rarp_pkt_cnt;
+	u64 rx_desc_multicast_pkt_cnt;
+	u64 rx_desc_broadcast_pkt_cnt;
+	u64 rx_desc_ipsec_pkt_cnt;
+	u64 rx_desc_ip_opt_pkt_cnt;
+	u64 rx_desc_key_not_match_cnt;
+
+	u64 rx_octets_total_ok_cnt;
+	u64 rx_uc_pkt_cnt;
+	u64 rx_mc_pkt_cnt;
+	u64 rx_bc_pkt_cnt;
+	u64 rx_vlan_pkt_cnt;
+	u64 rx_octets_bad_cnt;
+	u64 rx_octets_total_filt_cnt;
+	u64 rx_filt_pkt_cnt;
+	u64 rx_trans_pkt_cnt;
+	u64 rx_framesize_64;
+	u64 rx_framesize_65_127;
+	u64 rx_framesize_128_255;
+	u64 rx_framesize_256_511;
+	u64 rx_framesize_512_1023;
+	u64 rx_framesize_1024_1518;
+	u64 rx_framesize_bt_1518;
+	u64 rx_fcs_error_cnt;
+	u64 rx_data_error_cnt;
+	u64 rx_align_error_cnt;
+	u64 rx_pause_macctl_frame_cnt;
+	u64 rx_unknown_macctl_frame_cnt;
+	/* crc ok, > max_frm_size, < 2max_frm_size */
+	u64 rx_frame_long_err_cnt;
+	/* crc fail, > max_frm_size, < 2max_frm_size */
+	u64 rx_jabber_err_cnt;
+	/* > 2max_frm_size */
+	u64 rx_frame_very_long_err_cnt;
+	/* < 64byte, >= short_runts_thr */
+	u64 rx_frame_runt_err_cnt;
+	/* < short_runts_thr */
+	u64 rx_frame_short_err_cnt;
+	/* PCU: dropped when the RX FIFO is full.*/
+	u64 rx_overflow_cnt;
+	/* GMAC: the count of overflows of the RX FIFO */
+	u64 rx_overrun_cnt;
+	/* PCU: the count of buffer alloc errors in RX */
+	u64 rx_bufrq_err_cnt;
+	/* PCU: the count of write descriptor errors in RX */
+	u64 rx_we_err_cnt;
+	/* GMAC: the count of pkts that contain PAD but length is not 64 */
+	u64 rx_lengthfield_err_cnt;
+	u64 rx_fail_comma_cnt;
+
+	u64 rx_dma_err_cnt;
+	u64 rx_fifo_less_empty_thrsld_cnt;
+
+	u64 tx_octets_total_ok_cnt;
+	u64 tx_uc_pkt_cnt;
+	u64 tx_mc_pkt_cnt;
+	u64 tx_bc_pkt_cnt;
+	u64 tx_vlan_pkt_cnt;
+	u64 tx_octets_bad_cnt;
+	u64 tx_trans_pkt_cnt;
+	u64 tx_pause_frame_cnt;
+	u64 tx_framesize_64;
+	u64 tx_framesize_65_127;
+	u64 tx_framesize_128_255;
+	u64 tx_framesize_256_511;
+	u64 tx_framesize_512_1023;
+	u64 tx_framesize_1024_1518;
+	u64 tx_framesize_bt_1518;
+	/* GMAC: the count of times that frames fail to be transmitted
+	 *       due to internal errors.
+	 */
+	u64 tx_underrun_err_cnt;
+	u64 tx_add_cs_fail_cnt;
+	/* PCU: the count of buffer free errors in TX */
+	u64 tx_bufrl_err_cnt;
+	u64 tx_crc_err_cnt;
+	u64 tx_drop_cnt;
+	u64 tx_excessive_length_drop_cnt;
+
+	u64 tx_timeout_cnt;
+	u64 tx_dma_err_cnt;
+
+	u64 np_link_fail_cnt;
+	u64 reset_fail_cnt;
+};
+
+struct hbg_priv {
+	struct net_device *netdev;
+	struct pci_dev *pdev;
+	u8 __iomem *io_base;
+	struct hbg_dev_specs dev_specs;
+	unsigned long state;
+	struct hbg_mac mac;
+	struct hbg_vector vectors;
+	struct hbg_ring tx_ring;
+	struct hbg_ring rx_ring;
+	struct hbg_mac_filter filter;
+	enum hbg_reset_type reset_type;
+	struct hbg_user_def user_def;
+	struct hbg_stats stats;
+	unsigned long last_update_stats_time;
+	struct delayed_work service_task;
+};
+
+void hbg_err_reset_task_schedule(struct hbg_priv *priv);
+void hbg_np_link_fail_task_schedule(struct hbg_priv *priv);
+
+#endif
diff --git a/drivers/net/ethernet/hisilicon/hibmcge/hbg_debugfs.c b/drivers/net/ethernet/hisilicon/hibmcge/hbg_debugfs.c
new file mode 100644
index 000000000000..01ad82d2f5cc
--- /dev/null
+++ b/drivers/net/ethernet/hisilicon/hibmcge/hbg_debugfs.c
@@ -0,0 +1,168 @@
+// SPDX-License-Identifier: GPL-2.0+
+// Copyright (c) 2024 Hisilicon Limited.
+
+#include <linux/debugfs.h>
+#include <linux/device.h>
+#include <linux/etherdevice.h>
+#include <linux/seq_file.h>
+#include <linux/string_choices.h>
+#include "hbg_common.h"
+#include "hbg_debugfs.h"
+#include "hbg_hw.h"
+#include "hbg_irq.h"
+#include "hbg_txrx.h"
+
+static struct dentry *hbg_dbgfs_root;
+
+struct hbg_dbg_info {
+	const char *name;
+	int (*read)(struct seq_file *seq, void *data);
+};
+
+#define state_str_true_false(p, s) str_true_false(test_bit(s, &(p)->state))
+
+static void hbg_dbg_ring(struct hbg_priv *priv, struct hbg_ring *ring,
+			 struct seq_file *s)
+{
+	u32 irq_mask = ring->dir == HBG_DIR_TX ? HBG_INT_MSK_TX_B :
+						 HBG_INT_MSK_RX_B;
+
+	seq_printf(s, "ring used num: %u\n",
+		   hbg_get_queue_used_num(ring));
+	seq_printf(s, "ring max num: %u\n", ring->len);
+	seq_printf(s, "ring head: %u, tail: %u\n", ring->head, ring->tail);
+	seq_printf(s, "fifo used num: %u\n",
+		   hbg_hw_get_fifo_used_num(priv, ring->dir));
+	seq_printf(s, "fifo max num: %u\n",
+		   hbg_get_spec_fifo_max_num(priv, ring->dir));
+	seq_printf(s, "irq enabled: %s\n",
+		   str_true_false(hbg_hw_irq_is_enabled(priv, irq_mask)));
+}
+
+static int hbg_dbg_tx_ring(struct seq_file *s, void *unused)
+{
+	struct net_device *netdev = dev_get_drvdata(s->private);
+	struct hbg_priv *priv = netdev_priv(netdev);
+
+	hbg_dbg_ring(priv, &priv->tx_ring, s);
+	return 0;
+}
+
+static int hbg_dbg_rx_ring(struct seq_file *s, void *unused)
+{
+	struct net_device *netdev = dev_get_drvdata(s->private);
+	struct hbg_priv *priv = netdev_priv(netdev);
+
+	hbg_dbg_ring(priv, &priv->rx_ring, s);
+	return 0;
+}
+
+static int hbg_dbg_irq_info(struct seq_file *s, void *unused)
+{
+	struct net_device *netdev = dev_get_drvdata(s->private);
+	struct hbg_priv *priv = netdev_priv(netdev);
+	const struct hbg_irq_info *info;
+	u32 i;
+
+	for (i = 0; i < priv->vectors.info_array_len; i++) {
+		info = &priv->vectors.info_array[i];
+		seq_printf(s,
+			   "%-20s: enabled: %-5s, reset: %-5s, logged: %-5s, count: %llu\n",
+			   info->name,
+			   str_true_false(hbg_hw_irq_is_enabled(priv,
+								info->mask)),
+			   str_true_false(info->need_reset),
+			   str_true_false(info->need_print),
+			   priv->vectors.stats_array[i]);
+	}
+
+	return 0;
+}
+
+static int hbg_dbg_mac_table(struct seq_file *s, void *unused)
+{
+	struct net_device *netdev = dev_get_drvdata(s->private);
+	struct hbg_priv *priv = netdev_priv(netdev);
+	struct hbg_mac_filter *filter;
+	u32 i;
+
+	filter = &priv->filter;
+	seq_printf(s, "mac addr max count: %u\n", filter->table_max_len);
+	seq_printf(s, "filter enabled: %s\n", str_true_false(filter->enabled));
+
+	for (i = 0; i < filter->table_max_len; i++) {
+		if (is_zero_ether_addr(filter->mac_table[i].addr))
+			continue;
+
+		seq_printf(s, "[%u] %pM\n", i, filter->mac_table[i].addr);
+	}
+
+	return 0;
+}
+
+static const char * const reset_type_str[] = {"None", "FLR", "Function"};
+
+static int hbg_dbg_nic_state(struct seq_file *s, void *unused)
+{
+	struct net_device *netdev = dev_get_drvdata(s->private);
+	struct hbg_priv *priv = netdev_priv(netdev);
+	bool np_link_fail;
+
+	seq_printf(s, "event handling state: %s\n",
+		   state_str_true_false(priv, HBG_NIC_STATE_EVENT_HANDLING));
+	seq_printf(s, "resetting state: %s\n",
+		   state_str_true_false(priv, HBG_NIC_STATE_RESETTING));
+	seq_printf(s, "reset fail state: %s\n",
+		   state_str_true_false(priv, HBG_NIC_STATE_RESET_FAIL));
+	seq_printf(s, "last reset type: %s\n",
+		   reset_type_str[priv->reset_type]);
+	seq_printf(s, "need reset state: %s\n",
+		   state_str_true_false(priv, HBG_NIC_STATE_NEED_RESET));
+
+	np_link_fail = !hbg_reg_read_field(priv, HBG_REG_AN_NEG_STATE_ADDR,
+					   HBG_REG_AN_NEG_STATE_NP_LINK_OK_B);
+	seq_printf(s, "np_link fail state: %s\n", str_true_false(np_link_fail));
+
+	return 0;
+}
+
+static const struct hbg_dbg_info hbg_dbg_infos[] = {
+	{ "tx_ring", hbg_dbg_tx_ring },
+	{ "rx_ring", hbg_dbg_rx_ring },
+	{ "irq_info", hbg_dbg_irq_info },
+	{ "mac_table", hbg_dbg_mac_table },
+	{ "nic_state", hbg_dbg_nic_state },
+};
+
+static void hbg_debugfs_uninit(void *data)
+{
+	debugfs_remove_recursive((struct dentry *)data);
+}
+
+void hbg_debugfs_init(struct hbg_priv *priv)
+{
+	const char *name = pci_name(priv->pdev);
+	struct device *dev = &priv->pdev->dev;
+	struct dentry *root;
+	u32 i;
+
+	root = debugfs_create_dir(name, hbg_dbgfs_root);
+
+	for (i = 0; i < ARRAY_SIZE(hbg_dbg_infos); i++)
+		debugfs_create_devm_seqfile(dev, hbg_dbg_infos[i].name,
+					    root, hbg_dbg_infos[i].read);
+
+	/* Ignore the failure because debugfs is not a key feature. */
+	devm_add_action_or_reset(dev, hbg_debugfs_uninit, root);
+}
+
+void hbg_debugfs_register(void)
+{
+	hbg_dbgfs_root = debugfs_create_dir("hibmcge", NULL);
+}
+
+void hbg_debugfs_unregister(void)
+{
+	debugfs_remove_recursive(hbg_dbgfs_root);
+	hbg_dbgfs_root = NULL;
+}
diff --git a/drivers/net/ethernet/hisilicon/hibmcge/hbg_debugfs.h b/drivers/net/ethernet/hisilicon/hibmcge/hbg_debugfs.h
new file mode 100644
index 000000000000..80670d66bbeb
--- /dev/null
+++ b/drivers/net/ethernet/hisilicon/hibmcge/hbg_debugfs.h
@@ -0,0 +1,12 @@
+/* SPDX-License-Identifier: GPL-2.0+ */
+/* Copyright (c) 2024 Hisilicon Limited. */
+
+#ifndef __HBG_DEBUGFS_H
+#define __HBG_DEBUGFS_H
+
+void hbg_debugfs_register(void);
+void hbg_debugfs_unregister(void);
+
+void hbg_debugfs_init(struct hbg_priv *priv);
+
+#endif
diff --git a/drivers/net/ethernet/hisilicon/hibmcge/hbg_diagnose.c b/drivers/net/ethernet/hisilicon/hibmcge/hbg_diagnose.c
new file mode 100644
index 000000000000..c0ce74cf7382
--- /dev/null
+++ b/drivers/net/ethernet/hisilicon/hibmcge/hbg_diagnose.c
@@ -0,0 +1,349 @@
+// SPDX-License-Identifier: GPL-2.0+
+// Copyright (c) 2025 Hisilicon Limited.
+
+#include <linux/iopoll.h>
+#include <linux/phy.h>
+#include "hbg_common.h"
+#include "hbg_ethtool.h"
+#include "hbg_hw.h"
+#include "hbg_diagnose.h"
+
+#define HBG_MSG_DATA_MAX_NUM	64
+
+struct hbg_diagnose_message {
+	u32 opcode;
+	u32 status;
+	u32 data_num;
+	struct hbg_priv *priv;
+
+	u32 data[HBG_MSG_DATA_MAX_NUM];
+};
+
+#define HBG_HW_PUSH_WAIT_TIMEOUT_US	(2 * 1000 * 1000)
+#define HBG_HW_PUSH_WAIT_INTERVAL_US	(1 * 1000)
+
+enum hbg_push_cmd {
+	HBG_PUSH_CMD_IRQ = 0,
+	HBG_PUSH_CMD_STATS,
+	HBG_PUSH_CMD_LINK,
+};
+
+struct hbg_push_stats_info {
+	/* id is used to match the name of the current stats item.
+	 * and is used for pretty print on BMC
+	 */
+	u32 id;
+	u64 offset;
+};
+
+struct hbg_push_irq_info {
+	/* id is used to match the name of the current irq.
+	 * and is used for pretty print on BMC
+	 */
+	u32 id;
+	u32 mask;
+};
+
+#define HBG_PUSH_IRQ_I(name, id) {id, HBG_INT_MSK_##name##_B}
+static const struct hbg_push_irq_info hbg_push_irq_list[] = {
+	HBG_PUSH_IRQ_I(RX, 0),
+	HBG_PUSH_IRQ_I(TX, 1),
+	HBG_PUSH_IRQ_I(TX_PKT_CPL, 2),
+	HBG_PUSH_IRQ_I(MAC_MII_FIFO_ERR, 3),
+	HBG_PUSH_IRQ_I(MAC_PCS_RX_FIFO_ERR, 4),
+	HBG_PUSH_IRQ_I(MAC_PCS_TX_FIFO_ERR, 5),
+	HBG_PUSH_IRQ_I(MAC_APP_RX_FIFO_ERR, 6),
+	HBG_PUSH_IRQ_I(MAC_APP_TX_FIFO_ERR, 7),
+	HBG_PUSH_IRQ_I(SRAM_PARITY_ERR, 8),
+	HBG_PUSH_IRQ_I(TX_AHB_ERR, 9),
+	HBG_PUSH_IRQ_I(RX_BUF_AVL, 10),
+	HBG_PUSH_IRQ_I(REL_BUF_ERR, 11),
+	HBG_PUSH_IRQ_I(TXCFG_AVL, 12),
+	HBG_PUSH_IRQ_I(TX_DROP, 13),
+	HBG_PUSH_IRQ_I(RX_DROP, 14),
+	HBG_PUSH_IRQ_I(RX_AHB_ERR, 15),
+	HBG_PUSH_IRQ_I(MAC_FIFO_ERR, 16),
+	HBG_PUSH_IRQ_I(RBREQ_ERR, 17),
+	HBG_PUSH_IRQ_I(WE_ERR, 18),
+};
+
+#define HBG_PUSH_STATS_I(name, id) {id, HBG_STATS_FIELD_OFF(name)}
+static const struct hbg_push_stats_info hbg_push_stats_list[] = {
+	HBG_PUSH_STATS_I(rx_desc_drop, 0),
+	HBG_PUSH_STATS_I(rx_desc_l2_err_cnt, 1),
+	HBG_PUSH_STATS_I(rx_desc_pkt_len_err_cnt, 2),
+	HBG_PUSH_STATS_I(rx_desc_l3_wrong_head_cnt, 3),
+	HBG_PUSH_STATS_I(rx_desc_l3_csum_err_cnt, 4),
+	HBG_PUSH_STATS_I(rx_desc_l3_len_err_cnt, 5),
+	HBG_PUSH_STATS_I(rx_desc_l3_zero_ttl_cnt, 6),
+	HBG_PUSH_STATS_I(rx_desc_l3_other_cnt, 7),
+	HBG_PUSH_STATS_I(rx_desc_l4_err_cnt, 8),
+	HBG_PUSH_STATS_I(rx_desc_l4_wrong_head_cnt, 9),
+	HBG_PUSH_STATS_I(rx_desc_l4_len_err_cnt, 10),
+	HBG_PUSH_STATS_I(rx_desc_l4_csum_err_cnt, 11),
+	HBG_PUSH_STATS_I(rx_desc_l4_zero_port_num_cnt, 12),
+	HBG_PUSH_STATS_I(rx_desc_l4_other_cnt, 13),
+	HBG_PUSH_STATS_I(rx_desc_frag_cnt, 14),
+	HBG_PUSH_STATS_I(rx_desc_ip_ver_err_cnt, 15),
+	HBG_PUSH_STATS_I(rx_desc_ipv4_pkt_cnt, 16),
+	HBG_PUSH_STATS_I(rx_desc_ipv6_pkt_cnt, 17),
+	HBG_PUSH_STATS_I(rx_desc_no_ip_pkt_cnt, 18),
+	HBG_PUSH_STATS_I(rx_desc_ip_pkt_cnt, 19),
+	HBG_PUSH_STATS_I(rx_desc_tcp_pkt_cnt, 20),
+	HBG_PUSH_STATS_I(rx_desc_udp_pkt_cnt, 21),
+	HBG_PUSH_STATS_I(rx_desc_vlan_pkt_cnt, 22),
+	HBG_PUSH_STATS_I(rx_desc_icmp_pkt_cnt, 23),
+	HBG_PUSH_STATS_I(rx_desc_arp_pkt_cnt, 24),
+	HBG_PUSH_STATS_I(rx_desc_rarp_pkt_cnt, 25),
+	HBG_PUSH_STATS_I(rx_desc_multicast_pkt_cnt, 26),
+	HBG_PUSH_STATS_I(rx_desc_broadcast_pkt_cnt, 27),
+	HBG_PUSH_STATS_I(rx_desc_ipsec_pkt_cnt, 28),
+	HBG_PUSH_STATS_I(rx_desc_ip_opt_pkt_cnt, 29),
+	HBG_PUSH_STATS_I(rx_desc_key_not_match_cnt, 30),
+	HBG_PUSH_STATS_I(rx_octets_total_ok_cnt, 31),
+	HBG_PUSH_STATS_I(rx_uc_pkt_cnt, 32),
+	HBG_PUSH_STATS_I(rx_mc_pkt_cnt, 33),
+	HBG_PUSH_STATS_I(rx_bc_pkt_cnt, 34),
+	HBG_PUSH_STATS_I(rx_vlan_pkt_cnt, 35),
+	HBG_PUSH_STATS_I(rx_octets_bad_cnt, 36),
+	HBG_PUSH_STATS_I(rx_octets_total_filt_cnt, 37),
+	HBG_PUSH_STATS_I(rx_filt_pkt_cnt, 38),
+	HBG_PUSH_STATS_I(rx_trans_pkt_cnt, 39),
+	HBG_PUSH_STATS_I(rx_framesize_64, 40),
+	HBG_PUSH_STATS_I(rx_framesize_65_127, 41),
+	HBG_PUSH_STATS_I(rx_framesize_128_255, 42),
+	HBG_PUSH_STATS_I(rx_framesize_256_511, 43),
+	HBG_PUSH_STATS_I(rx_framesize_512_1023, 44),
+	HBG_PUSH_STATS_I(rx_framesize_1024_1518, 45),
+	HBG_PUSH_STATS_I(rx_framesize_bt_1518, 46),
+	HBG_PUSH_STATS_I(rx_fcs_error_cnt, 47),
+	HBG_PUSH_STATS_I(rx_data_error_cnt, 48),
+	HBG_PUSH_STATS_I(rx_align_error_cnt, 49),
+	HBG_PUSH_STATS_I(rx_frame_long_err_cnt, 50),
+	HBG_PUSH_STATS_I(rx_jabber_err_cnt, 51),
+	HBG_PUSH_STATS_I(rx_pause_macctl_frame_cnt, 52),
+	HBG_PUSH_STATS_I(rx_unknown_macctl_frame_cnt, 53),
+	HBG_PUSH_STATS_I(rx_frame_very_long_err_cnt, 54),
+	HBG_PUSH_STATS_I(rx_frame_runt_err_cnt, 55),
+	HBG_PUSH_STATS_I(rx_frame_short_err_cnt, 56),
+	HBG_PUSH_STATS_I(rx_overflow_cnt, 57),
+	HBG_PUSH_STATS_I(rx_bufrq_err_cnt, 58),
+	HBG_PUSH_STATS_I(rx_we_err_cnt, 59),
+	HBG_PUSH_STATS_I(rx_overrun_cnt, 60),
+	HBG_PUSH_STATS_I(rx_lengthfield_err_cnt, 61),
+	HBG_PUSH_STATS_I(rx_fail_comma_cnt, 62),
+	HBG_PUSH_STATS_I(rx_dma_err_cnt, 63),
+	HBG_PUSH_STATS_I(rx_fifo_less_empty_thrsld_cnt, 64),
+	HBG_PUSH_STATS_I(tx_octets_total_ok_cnt, 65),
+	HBG_PUSH_STATS_I(tx_uc_pkt_cnt, 66),
+	HBG_PUSH_STATS_I(tx_mc_pkt_cnt, 67),
+	HBG_PUSH_STATS_I(tx_bc_pkt_cnt, 68),
+	HBG_PUSH_STATS_I(tx_vlan_pkt_cnt, 69),
+	HBG_PUSH_STATS_I(tx_octets_bad_cnt, 70),
+	HBG_PUSH_STATS_I(tx_trans_pkt_cnt, 71),
+	HBG_PUSH_STATS_I(tx_pause_frame_cnt, 72),
+	HBG_PUSH_STATS_I(tx_framesize_64, 73),
+	HBG_PUSH_STATS_I(tx_framesize_65_127, 74),
+	HBG_PUSH_STATS_I(tx_framesize_128_255, 75),
+	HBG_PUSH_STATS_I(tx_framesize_256_511, 76),
+	HBG_PUSH_STATS_I(tx_framesize_512_1023, 77),
+	HBG_PUSH_STATS_I(tx_framesize_1024_1518, 78),
+	HBG_PUSH_STATS_I(tx_framesize_bt_1518, 79),
+	HBG_PUSH_STATS_I(tx_underrun_err_cnt, 80),
+	HBG_PUSH_STATS_I(tx_add_cs_fail_cnt, 81),
+	HBG_PUSH_STATS_I(tx_bufrl_err_cnt, 82),
+	HBG_PUSH_STATS_I(tx_crc_err_cnt, 83),
+	HBG_PUSH_STATS_I(tx_drop_cnt, 84),
+	HBG_PUSH_STATS_I(tx_excessive_length_drop_cnt, 85),
+	HBG_PUSH_STATS_I(tx_dma_err_cnt, 86),
+	HBG_PUSH_STATS_I(reset_fail_cnt, 87),
+};
+
+static int hbg_push_msg_send(struct hbg_priv *priv,
+			     struct hbg_diagnose_message *msg)
+{
+	u32 header = 0;
+	u32 i;
+
+	if (msg->data_num == 0)
+		return 0;
+
+	for (i = 0; i < msg->data_num && i < HBG_MSG_DATA_MAX_NUM; i++)
+		hbg_reg_write(priv,
+			      HBG_REG_MSG_DATA_BASE_ADDR + i * sizeof(u32),
+			      msg->data[i]);
+
+	hbg_field_modify(header, HBG_REG_MSG_HEADER_OPCODE_M, msg->opcode);
+	hbg_field_modify(header, HBG_REG_MSG_HEADER_DATA_NUM_M,  msg->data_num);
+	hbg_field_modify(header, HBG_REG_MSG_HEADER_RESP_CODE_M, ETIMEDOUT);
+
+	/* start status */
+	hbg_field_modify(header, HBG_REG_MSG_HEADER_STATUS_M, 1);
+
+	/* write header msg to start push */
+	hbg_reg_write(priv, HBG_REG_MSG_HEADER_ADDR, header);
+
+	/* wait done */
+	readl_poll_timeout(priv->io_base + HBG_REG_MSG_HEADER_ADDR, header,
+			   !FIELD_GET(HBG_REG_MSG_HEADER_STATUS_M, header),
+			   HBG_HW_PUSH_WAIT_INTERVAL_US,
+			   HBG_HW_PUSH_WAIT_TIMEOUT_US);
+
+	msg->status = FIELD_GET(HBG_REG_MSG_HEADER_STATUS_M, header);
+	return -(int)FIELD_GET(HBG_REG_MSG_HEADER_RESP_CODE_M, header);
+}
+
+static int hbg_push_data(struct hbg_priv *priv,
+			 u32 opcode, u32 *data, u32 data_num)
+{
+	struct hbg_diagnose_message msg = {0};
+	u32 data_left_num;
+	u32 i, j;
+	int ret;
+
+	msg.priv = priv;
+	msg.opcode = opcode;
+	for (i = 0; i < data_num / HBG_MSG_DATA_MAX_NUM + 1; i++) {
+		if (i * HBG_MSG_DATA_MAX_NUM >= data_num)
+			break;
+
+		data_left_num = data_num - i * HBG_MSG_DATA_MAX_NUM;
+		for (j = 0; j < data_left_num && j < HBG_MSG_DATA_MAX_NUM; j++)
+			msg.data[j] = data[i * HBG_MSG_DATA_MAX_NUM + j];
+
+		msg.data_num = j;
+		ret = hbg_push_msg_send(priv, &msg);
+		if (ret)
+			return ret;
+	}
+
+	return 0;
+}
+
+static int hbg_push_data_u64(struct hbg_priv *priv, u32 opcode,
+			     u64 *data, u32 data_num)
+{
+	/* The length of u64 is twice that of u32,
+	 * the data_num must be multiplied by 2.
+	 */
+	return hbg_push_data(priv, opcode, (u32 *)data, data_num * 2);
+}
+
+static u64 hbg_get_irq_stats(struct hbg_vector *vectors, u32 mask)
+{
+	u32 i = 0;
+
+	for (i = 0; i < vectors->info_array_len; i++)
+		if (vectors->info_array[i].mask == mask)
+			return vectors->stats_array[i];
+
+	return 0;
+}
+
+static int hbg_push_irq_cnt(struct hbg_priv *priv)
+{
+	/* An id needs to be added for each data.
+	 * Therefore, the data_num must be multiplied by 2.
+	 */
+	u32 data_num = ARRAY_SIZE(hbg_push_irq_list) * 2;
+	struct hbg_vector *vectors = &priv->vectors;
+	const struct hbg_push_irq_info *info;
+	u32 i, j = 0;
+	u64 *data;
+	int ret;
+
+	data = kcalloc(data_num, sizeof(u64), GFP_KERNEL);
+	if (!data)
+		return -ENOMEM;
+
+	/* An id needs to be added for each data.
+	 * So i + 2 for each loop.
+	 */
+	for (i = 0; i < data_num; i += 2) {
+		info = &hbg_push_irq_list[j++];
+		data[i] = info->id;
+		data[i + 1] = hbg_get_irq_stats(vectors, info->mask);
+	}
+
+	ret = hbg_push_data_u64(priv, HBG_PUSH_CMD_IRQ, data, data_num);
+	kfree(data);
+	return ret;
+}
+
+static int hbg_push_link_status(struct hbg_priv *priv)
+{
+	u32 link_status[2];
+
+	/* phy link status */
+	link_status[0] = priv->mac.phydev->link;
+	/* mac link status */
+	link_status[1] = hbg_reg_read_field(priv, HBG_REG_AN_NEG_STATE_ADDR,
+					    HBG_REG_AN_NEG_STATE_NP_LINK_OK_B);
+
+	return hbg_push_data(priv, HBG_PUSH_CMD_LINK,
+			     link_status, ARRAY_SIZE(link_status));
+}
+
+static int hbg_push_stats(struct hbg_priv *priv)
+{
+	/* An id needs to be added for each data.
+	 * Therefore, the data_num must be multiplied by 2.
+	 */
+	u64 data_num = ARRAY_SIZE(hbg_push_stats_list) * 2;
+	struct hbg_stats *stats = &priv->stats;
+	const struct hbg_push_stats_info *info;
+	u32 i, j = 0;
+	u64 *data;
+	int ret;
+
+	data = kcalloc(data_num, sizeof(u64), GFP_KERNEL);
+	if (!data)
+		return -ENOMEM;
+
+	/* An id needs to be added for each data.
+	 * So i + 2 for each loop.
+	 */
+	for (i = 0; i < data_num; i += 2) {
+		info = &hbg_push_stats_list[j++];
+		data[i] = info->id;
+		data[i + 1] = HBG_STATS_R(stats, info->offset);
+	}
+
+	ret = hbg_push_data_u64(priv, HBG_PUSH_CMD_STATS, data, data_num);
+	kfree(data);
+	return ret;
+}
+
+void hbg_diagnose_message_push(struct hbg_priv *priv)
+{
+	int ret;
+
+	if (test_bit(HBG_NIC_STATE_RESETTING, &priv->state))
+		return;
+
+	/* only 1 is the right value */
+	if (hbg_reg_read(priv, HBG_REG_PUSH_REQ_ADDR) != 1)
+		return;
+
+	ret = hbg_push_irq_cnt(priv);
+	if (ret) {
+		dev_err(&priv->pdev->dev,
+			"failed to push irq cnt, ret = %d\n", ret);
+		goto push_done;
+	}
+
+	ret = hbg_push_link_status(priv);
+	if (ret) {
+		dev_err(&priv->pdev->dev,
+			"failed to push link status, ret = %d\n", ret);
+		goto push_done;
+	}
+
+	ret = hbg_push_stats(priv);
+	if (ret)
+		dev_err(&priv->pdev->dev,
+			"failed to push stats, ret = %d\n", ret);
+
+push_done:
+	hbg_reg_write(priv, HBG_REG_PUSH_REQ_ADDR, 0);
+}
diff --git a/drivers/net/ethernet/hisilicon/hibmcge/hbg_diagnose.h b/drivers/net/ethernet/hisilicon/hibmcge/hbg_diagnose.h
new file mode 100644
index 000000000000..ba04c6d8c03d
--- /dev/null
+++ b/drivers/net/ethernet/hisilicon/hibmcge/hbg_diagnose.h
@@ -0,0 +1,11 @@
+/* SPDX-License-Identifier: GPL-2.0+ */
+/* Copyright (c) 2025 Hisilicon Limited. */
+
+#ifndef __HBG_DIAGNOSE_H
+#define __HBG_DIAGNOSE_H
+
+#include "hbg_common.h"
+
+void hbg_diagnose_message_push(struct hbg_priv *priv);
+
+#endif
diff --git a/drivers/net/ethernet/hisilicon/hibmcge/hbg_err.c b/drivers/net/ethernet/hisilicon/hibmcge/hbg_err.c
new file mode 100644
index 000000000000..83cf75bf7a17
--- /dev/null
+++ b/drivers/net/ethernet/hisilicon/hibmcge/hbg_err.c
@@ -0,0 +1,193 @@
+// SPDX-License-Identifier: GPL-2.0+
+// Copyright (c) 2024 Hisilicon Limited.
+
+#include <linux/etherdevice.h>
+#include <linux/netdevice.h>
+#include <linux/phy.h>
+#include <linux/rtnetlink.h>
+#include "hbg_common.h"
+#include "hbg_err.h"
+#include "hbg_hw.h"
+
+static void hbg_restore_mac_table(struct hbg_priv *priv)
+{
+	struct hbg_mac_filter *filter = &priv->filter;
+	u64 addr;
+	u32 i;
+
+	for (i = 0; i < filter->table_max_len; i++)
+		if (!is_zero_ether_addr(filter->mac_table[i].addr)) {
+			addr = ether_addr_to_u64(filter->mac_table[i].addr);
+			hbg_hw_set_uc_addr(priv, addr, i);
+		}
+
+	hbg_hw_set_mac_filter_enable(priv, priv->filter.enabled);
+}
+
+static void hbg_restore_user_def_settings(struct hbg_priv *priv)
+{
+	/* The index of host mac is always 0. */
+	u64 rx_pause_addr = ether_addr_to_u64(priv->filter.mac_table[0].addr);
+	struct ethtool_pauseparam *pause_param = &priv->user_def.pause_param;
+
+	hbg_restore_mac_table(priv);
+	hbg_hw_set_mtu(priv, priv->netdev->mtu);
+	hbg_hw_set_pause_enable(priv, pause_param->tx_pause,
+				pause_param->rx_pause);
+	hbg_hw_set_rx_pause_mac_addr(priv, rx_pause_addr);
+}
+
+int hbg_rebuild(struct hbg_priv *priv)
+{
+	int ret;
+
+	ret = hbg_hw_init(priv);
+	if (ret)
+		return ret;
+
+	hbg_restore_user_def_settings(priv);
+	return 0;
+}
+
+static int hbg_reset_prepare(struct hbg_priv *priv, enum hbg_reset_type type)
+{
+	int ret;
+
+	if (test_and_set_bit(HBG_NIC_STATE_RESETTING, &priv->state))
+		return -EBUSY;
+
+	if (netif_running(priv->netdev)) {
+		clear_bit(HBG_NIC_STATE_RESETTING, &priv->state);
+		dev_warn(&priv->pdev->dev,
+			 "failed to reset because port is up\n");
+		return -EBUSY;
+	}
+
+	netif_device_detach(priv->netdev);
+
+	priv->reset_type = type;
+	clear_bit(HBG_NIC_STATE_RESET_FAIL, &priv->state);
+	ret = hbg_hw_event_notify(priv, HBG_HW_EVENT_RESET);
+	if (ret) {
+		priv->stats.reset_fail_cnt++;
+		set_bit(HBG_NIC_STATE_RESET_FAIL, &priv->state);
+		clear_bit(HBG_NIC_STATE_RESETTING, &priv->state);
+	}
+
+	return ret;
+}
+
+static int hbg_reset_done(struct hbg_priv *priv, enum hbg_reset_type type)
+{
+	int ret;
+
+	if (!test_bit(HBG_NIC_STATE_RESETTING, &priv->state) ||
+	    type != priv->reset_type)
+		return 0;
+
+	ret = hbg_rebuild(priv);
+	if (ret) {
+		priv->stats.reset_fail_cnt++;
+		set_bit(HBG_NIC_STATE_RESET_FAIL, &priv->state);
+		clear_bit(HBG_NIC_STATE_RESETTING, &priv->state);
+		dev_err(&priv->pdev->dev, "failed to rebuild after reset\n");
+		return ret;
+	}
+
+	netif_device_attach(priv->netdev);
+	clear_bit(HBG_NIC_STATE_RESETTING, &priv->state);
+
+	dev_info(&priv->pdev->dev, "reset done\n");
+	return ret;
+}
+
+int hbg_reset(struct hbg_priv *priv)
+{
+	int ret;
+
+	ret = hbg_reset_prepare(priv, HBG_RESET_TYPE_FUNCTION);
+	if (ret)
+		return ret;
+
+	return hbg_reset_done(priv, HBG_RESET_TYPE_FUNCTION);
+}
+
+void hbg_err_reset(struct hbg_priv *priv)
+{
+	bool running;
+
+	rtnl_lock();
+	running = netif_running(priv->netdev);
+	if (running)
+		dev_close(priv->netdev);
+
+	if (hbg_reset(priv))
+		goto err_unlock;
+
+	if (running)
+		dev_open(priv->netdev, NULL);
+
+err_unlock:
+	rtnl_unlock();
+}
+
+static pci_ers_result_t hbg_pci_err_detected(struct pci_dev *pdev,
+					     pci_channel_state_t state)
+{
+	struct net_device *netdev = pci_get_drvdata(pdev);
+
+	netif_device_detach(netdev);
+
+	if (state == pci_channel_io_perm_failure)
+		return PCI_ERS_RESULT_DISCONNECT;
+
+	pci_disable_device(pdev);
+	return PCI_ERS_RESULT_NEED_RESET;
+}
+
+static pci_ers_result_t hbg_pci_err_slot_reset(struct pci_dev *pdev)
+{
+	struct net_device *netdev = pci_get_drvdata(pdev);
+	struct hbg_priv *priv = netdev_priv(netdev);
+
+	if (pci_enable_device(pdev)) {
+		dev_err(&pdev->dev,
+			"failed to re-enable PCI device after reset\n");
+		return PCI_ERS_RESULT_DISCONNECT;
+	}
+
+	pci_set_master(pdev);
+	pci_restore_state(pdev);
+	pci_save_state(pdev);
+
+	hbg_err_reset(priv);
+	return PCI_ERS_RESULT_RECOVERED;
+}
+
+static void hbg_pci_err_reset_prepare(struct pci_dev *pdev)
+{
+	struct net_device *netdev = pci_get_drvdata(pdev);
+	struct hbg_priv *priv = netdev_priv(netdev);
+
+	hbg_reset_prepare(priv, HBG_RESET_TYPE_FLR);
+}
+
+static void hbg_pci_err_reset_done(struct pci_dev *pdev)
+{
+	struct net_device *netdev = pci_get_drvdata(pdev);
+	struct hbg_priv *priv = netdev_priv(netdev);
+
+	hbg_reset_done(priv, HBG_RESET_TYPE_FLR);
+}
+
+static const struct pci_error_handlers hbg_pci_err_handler = {
+	.error_detected = hbg_pci_err_detected,
+	.slot_reset = hbg_pci_err_slot_reset,
+	.reset_prepare = hbg_pci_err_reset_prepare,
+	.reset_done = hbg_pci_err_reset_done,
+};
+
+void hbg_set_pci_err_handler(struct pci_driver *pdrv)
+{
+	pdrv->err_handler = &hbg_pci_err_handler;
+}
diff --git a/drivers/net/ethernet/hisilicon/hibmcge/hbg_err.h b/drivers/net/ethernet/hisilicon/hibmcge/hbg_err.h
new file mode 100644
index 000000000000..fb9fbe7004e8
--- /dev/null
+++ b/drivers/net/ethernet/hisilicon/hibmcge/hbg_err.h
@@ -0,0 +1,14 @@
+/* SPDX-License-Identifier: GPL-2.0+ */
+/* Copyright (c) 2024 Hisilicon Limited. */
+
+#ifndef __HBG_ERR_H
+#define __HBG_ERR_H
+
+#include <linux/pci.h>
+
+void hbg_set_pci_err_handler(struct pci_driver *pdrv);
+int hbg_reset(struct hbg_priv *priv);
+int hbg_rebuild(struct hbg_priv *priv);
+void hbg_err_reset(struct hbg_priv *priv);
+
+#endif
diff --git a/drivers/net/ethernet/hisilicon/hibmcge/hbg_ethtool.c b/drivers/net/ethernet/hisilicon/hibmcge/hbg_ethtool.c
new file mode 100644
index 000000000000..1d62ff913737
--- /dev/null
+++ b/drivers/net/ethernet/hisilicon/hibmcge/hbg_ethtool.c
@@ -0,0 +1,500 @@
+// SPDX-License-Identifier: GPL-2.0+
+// Copyright (c) 2024 Hisilicon Limited.
+
+#include <linux/ethtool.h>
+#include <linux/phy.h>
+#include <linux/rtnetlink.h>
+#include "hbg_common.h"
+#include "hbg_err.h"
+#include "hbg_ethtool.h"
+#include "hbg_hw.h"
+
+struct hbg_ethtool_stats {
+	char name[ETH_GSTRING_LEN];
+	unsigned long offset;
+	u32 reg; /* set to 0 if stats is not updated via dump reg */
+};
+
+#define HBG_STATS_I(stats) { #stats, HBG_STATS_FIELD_OFF(stats), 0}
+#define HBG_STATS_REG_I(stats, reg) { #stats, HBG_STATS_FIELD_OFF(stats), reg}
+
+static const struct hbg_ethtool_stats hbg_ethtool_stats_info[] = {
+	HBG_STATS_I(rx_desc_l2_err_cnt),
+	HBG_STATS_I(rx_desc_pkt_len_err_cnt),
+	HBG_STATS_I(rx_desc_l3_wrong_head_cnt),
+	HBG_STATS_I(rx_desc_l3_csum_err_cnt),
+	HBG_STATS_I(rx_desc_l3_len_err_cnt),
+	HBG_STATS_I(rx_desc_l3_zero_ttl_cnt),
+	HBG_STATS_I(rx_desc_l3_other_cnt),
+	HBG_STATS_I(rx_desc_l4_wrong_head_cnt),
+	HBG_STATS_I(rx_desc_l4_len_err_cnt),
+	HBG_STATS_I(rx_desc_l4_csum_err_cnt),
+	HBG_STATS_I(rx_desc_l4_zero_port_num_cnt),
+	HBG_STATS_I(rx_desc_l4_other_cnt),
+	HBG_STATS_I(rx_desc_ip_ver_err_cnt),
+	HBG_STATS_I(rx_desc_ipv4_pkt_cnt),
+	HBG_STATS_I(rx_desc_ipv6_pkt_cnt),
+	HBG_STATS_I(rx_desc_no_ip_pkt_cnt),
+	HBG_STATS_I(rx_desc_ip_pkt_cnt),
+	HBG_STATS_I(rx_desc_tcp_pkt_cnt),
+	HBG_STATS_I(rx_desc_udp_pkt_cnt),
+	HBG_STATS_I(rx_desc_vlan_pkt_cnt),
+	HBG_STATS_I(rx_desc_icmp_pkt_cnt),
+	HBG_STATS_I(rx_desc_arp_pkt_cnt),
+	HBG_STATS_I(rx_desc_rarp_pkt_cnt),
+	HBG_STATS_I(rx_desc_multicast_pkt_cnt),
+	HBG_STATS_I(rx_desc_broadcast_pkt_cnt),
+	HBG_STATS_I(rx_desc_ipsec_pkt_cnt),
+	HBG_STATS_I(rx_desc_ip_opt_pkt_cnt),
+	HBG_STATS_I(rx_desc_key_not_match_cnt),
+
+	HBG_STATS_REG_I(rx_octets_bad_cnt, HBG_REG_RX_OCTETS_BAD_ADDR),
+	HBG_STATS_REG_I(rx_octets_total_filt_cnt,
+			HBG_REG_RX_OCTETS_TOTAL_FILT_ADDR),
+	HBG_STATS_REG_I(rx_uc_pkt_cnt, HBG_REG_RX_UC_PKTS_ADDR),
+	HBG_STATS_REG_I(rx_vlan_pkt_cnt, HBG_REG_RX_TAGGED_ADDR),
+	HBG_STATS_REG_I(rx_filt_pkt_cnt, HBG_REG_RX_FILT_PKT_CNT_ADDR),
+	HBG_STATS_REG_I(rx_data_error_cnt, HBG_REG_RX_DATA_ERR_ADDR),
+	HBG_STATS_REG_I(rx_frame_long_err_cnt, HBG_REG_RX_LONG_ERRORS_ADDR),
+	HBG_STATS_REG_I(rx_jabber_err_cnt, HBG_REG_RX_JABBER_ERRORS_ADDR),
+	HBG_STATS_REG_I(rx_frame_very_long_err_cnt,
+			HBG_REG_RX_VERY_LONG_ERR_CNT_ADDR),
+	HBG_STATS_REG_I(rx_frame_runt_err_cnt, HBG_REG_RX_RUNT_ERR_CNT_ADDR),
+	HBG_STATS_REG_I(rx_frame_short_err_cnt, HBG_REG_RX_SHORT_ERR_CNT_ADDR),
+	HBG_STATS_REG_I(rx_overflow_cnt, HBG_REG_RX_OVER_FLOW_CNT_ADDR),
+	HBG_STATS_REG_I(rx_bufrq_err_cnt, HBG_REG_RX_BUFRQ_ERR_CNT_ADDR),
+	HBG_STATS_REG_I(rx_we_err_cnt, HBG_REG_RX_WE_ERR_CNT_ADDR),
+	HBG_STATS_REG_I(rx_overrun_cnt, HBG_REG_RX_OVERRUN_CNT_ADDR),
+	HBG_STATS_REG_I(rx_lengthfield_err_cnt,
+			HBG_REG_RX_LENGTHFIELD_ERR_CNT_ADDR),
+	HBG_STATS_REG_I(rx_fail_comma_cnt, HBG_REG_RX_FAIL_COMMA_CNT_ADDR),
+	HBG_STATS_I(rx_dma_err_cnt),
+	HBG_STATS_I(rx_fifo_less_empty_thrsld_cnt),
+
+	HBG_STATS_REG_I(tx_uc_pkt_cnt, HBG_REG_TX_UC_PKTS_ADDR),
+	HBG_STATS_REG_I(tx_vlan_pkt_cnt, HBG_REG_TX_TAGGED_ADDR),
+	HBG_STATS_REG_I(tx_octets_bad_cnt, HBG_REG_OCTETS_TRANSMITTED_BAD_ADDR),
+
+	HBG_STATS_REG_I(tx_underrun_err_cnt, HBG_REG_TX_UNDERRUN_ADDR),
+	HBG_STATS_REG_I(tx_add_cs_fail_cnt, HBG_REG_TX_CS_FAIL_CNT_ADDR),
+	HBG_STATS_REG_I(tx_bufrl_err_cnt, HBG_REG_TX_BUFRL_ERR_CNT_ADDR),
+	HBG_STATS_REG_I(tx_crc_err_cnt, HBG_REG_TX_CRC_ERROR_ADDR),
+	HBG_STATS_REG_I(tx_drop_cnt, HBG_REG_TX_DROP_CNT_ADDR),
+	HBG_STATS_REG_I(tx_excessive_length_drop_cnt,
+			HBG_REG_TX_EXCESSIVE_LENGTH_DROP_ADDR),
+	HBG_STATS_I(tx_dma_err_cnt),
+	HBG_STATS_I(tx_timeout_cnt),
+	HBG_STATS_I(reset_fail_cnt),
+};
+
+static const struct hbg_ethtool_stats hbg_ethtool_rmon_stats_info[] = {
+	HBG_STATS_I(rx_desc_frag_cnt),
+	HBG_STATS_REG_I(rx_framesize_64, HBG_REG_RX_PKTS_64OCTETS_ADDR),
+	HBG_STATS_REG_I(rx_framesize_65_127,
+			HBG_REG_RX_PKTS_65TO127OCTETS_ADDR),
+	HBG_STATS_REG_I(rx_framesize_128_255,
+			HBG_REG_RX_PKTS_128TO255OCTETS_ADDR),
+	HBG_STATS_REG_I(rx_framesize_256_511,
+			HBG_REG_RX_PKTS_256TO511OCTETS_ADDR),
+	HBG_STATS_REG_I(rx_framesize_512_1023,
+			HBG_REG_RX_PKTS_512TO1023OCTETS_ADDR),
+	HBG_STATS_REG_I(rx_framesize_1024_1518,
+			HBG_REG_RX_PKTS_1024TO1518OCTETS_ADDR),
+	HBG_STATS_REG_I(rx_framesize_bt_1518,
+			HBG_REG_RX_PKTS_1519TOMAXOCTETS_ADDR),
+	HBG_STATS_REG_I(tx_framesize_64, HBG_REG_TX_PKTS_64OCTETS_ADDR),
+	HBG_STATS_REG_I(tx_framesize_65_127,
+			HBG_REG_TX_PKTS_65TO127OCTETS_ADDR),
+	HBG_STATS_REG_I(tx_framesize_128_255,
+			HBG_REG_TX_PKTS_128TO255OCTETS_ADDR),
+	HBG_STATS_REG_I(tx_framesize_256_511,
+			HBG_REG_TX_PKTS_256TO511OCTETS_ADDR),
+	HBG_STATS_REG_I(tx_framesize_512_1023,
+			HBG_REG_TX_PKTS_512TO1023OCTETS_ADDR),
+	HBG_STATS_REG_I(tx_framesize_1024_1518,
+			HBG_REG_TX_PKTS_1024TO1518OCTETS_ADDR),
+	HBG_STATS_REG_I(tx_framesize_bt_1518,
+			HBG_REG_TX_PKTS_1519TOMAXOCTETS_ADDR),
+};
+
+static const struct hbg_ethtool_stats hbg_ethtool_mac_stats_info[] = {
+	HBG_STATS_REG_I(rx_mc_pkt_cnt, HBG_REG_RX_MC_PKTS_ADDR),
+	HBG_STATS_REG_I(rx_bc_pkt_cnt, HBG_REG_RX_BC_PKTS_ADDR),
+	HBG_STATS_REG_I(rx_align_error_cnt, HBG_REG_RX_ALIGN_ERRORS_ADDR),
+	HBG_STATS_REG_I(rx_octets_total_ok_cnt,
+			HBG_REG_RX_OCTETS_TOTAL_OK_ADDR),
+	HBG_STATS_REG_I(rx_trans_pkt_cnt, HBG_REG_RX_TRANS_PKG_CNT_ADDR),
+	HBG_STATS_REG_I(rx_fcs_error_cnt, HBG_REG_RX_FCS_ERRORS_ADDR),
+	HBG_STATS_REG_I(tx_mc_pkt_cnt, HBG_REG_TX_MC_PKTS_ADDR),
+	HBG_STATS_REG_I(tx_bc_pkt_cnt, HBG_REG_TX_BC_PKTS_ADDR),
+	HBG_STATS_REG_I(tx_octets_total_ok_cnt,
+			HBG_REG_OCTETS_TRANSMITTED_OK_ADDR),
+	HBG_STATS_REG_I(tx_trans_pkt_cnt, HBG_REG_TX_TRANS_PKG_CNT_ADDR),
+};
+
+static const struct hbg_ethtool_stats hbg_ethtool_ctrl_stats_info[] = {
+	HBG_STATS_REG_I(rx_pause_macctl_frame_cnt,
+			HBG_REG_RX_PAUSE_MACCTL_FRAMCOUNTER_ADDR),
+	HBG_STATS_REG_I(tx_pause_frame_cnt, HBG_REG_TX_PAUSE_FRAMES_ADDR),
+	HBG_STATS_REG_I(rx_unknown_macctl_frame_cnt,
+			HBG_REG_RX_UNKNOWN_MACCTL_FRAMCOUNTER_ADDR),
+};
+
+enum hbg_reg_dump_type {
+	HBG_DUMP_REG_TYPE_SPEC = 0,
+	HBG_DUMP_REG_TYPE_MDIO,
+	HBG_DUMP_REG_TYPE_GMAC,
+	HBG_DUMP_REG_TYPE_PCU,
+};
+
+struct hbg_reg_info {
+	u32 type;
+	u32 offset;
+	u32 val;
+};
+
+#define HBG_DUMP_SPEC_I(offset) {HBG_DUMP_REG_TYPE_SPEC, offset, 0}
+#define HBG_DUMP_MDIO_I(offset) {HBG_DUMP_REG_TYPE_MDIO, offset, 0}
+#define HBG_DUMP_GMAC_I(offset) {HBG_DUMP_REG_TYPE_GMAC, offset, 0}
+#define HBG_DUMP_PCU_I(offset) {HBG_DUMP_REG_TYPE_PCU, offset, 0}
+
+static const struct hbg_reg_info hbg_dump_reg_infos[] = {
+	/* dev specs */
+	HBG_DUMP_SPEC_I(HBG_REG_SPEC_VALID_ADDR),
+	HBG_DUMP_SPEC_I(HBG_REG_EVENT_REQ_ADDR),
+	HBG_DUMP_SPEC_I(HBG_REG_MAC_ID_ADDR),
+	HBG_DUMP_SPEC_I(HBG_REG_PHY_ID_ADDR),
+	HBG_DUMP_SPEC_I(HBG_REG_MAC_ADDR_ADDR),
+	HBG_DUMP_SPEC_I(HBG_REG_MAC_ADDR_HIGH_ADDR),
+	HBG_DUMP_SPEC_I(HBG_REG_UC_MAC_NUM_ADDR),
+	HBG_DUMP_SPEC_I(HBG_REG_MDIO_FREQ_ADDR),
+	HBG_DUMP_SPEC_I(HBG_REG_MAX_MTU_ADDR),
+	HBG_DUMP_SPEC_I(HBG_REG_MIN_MTU_ADDR),
+	HBG_DUMP_SPEC_I(HBG_REG_TX_FIFO_NUM_ADDR),
+	HBG_DUMP_SPEC_I(HBG_REG_RX_FIFO_NUM_ADDR),
+	HBG_DUMP_SPEC_I(HBG_REG_VLAN_LAYERS_ADDR),
+
+	/* mdio */
+	HBG_DUMP_MDIO_I(HBG_REG_MDIO_COMMAND_ADDR),
+	HBG_DUMP_MDIO_I(HBG_REG_MDIO_ADDR_ADDR),
+	HBG_DUMP_MDIO_I(HBG_REG_MDIO_WDATA_ADDR),
+	HBG_DUMP_MDIO_I(HBG_REG_MDIO_RDATA_ADDR),
+	HBG_DUMP_MDIO_I(HBG_REG_MDIO_STA_ADDR),
+
+	/* gmac */
+	HBG_DUMP_GMAC_I(HBG_REG_DUPLEX_TYPE_ADDR),
+	HBG_DUMP_GMAC_I(HBG_REG_FD_FC_TYPE_ADDR),
+	HBG_DUMP_GMAC_I(HBG_REG_FC_TX_TIMER_ADDR),
+	HBG_DUMP_GMAC_I(HBG_REG_FD_FC_ADDR_LOW_ADDR),
+	HBG_DUMP_GMAC_I(HBG_REG_FD_FC_ADDR_HIGH_ADDR),
+	HBG_DUMP_GMAC_I(HBG_REG_MAX_FRAME_SIZE_ADDR),
+	HBG_DUMP_GMAC_I(HBG_REG_PORT_MODE_ADDR),
+	HBG_DUMP_GMAC_I(HBG_REG_PORT_ENABLE_ADDR),
+	HBG_DUMP_GMAC_I(HBG_REG_PAUSE_ENABLE_ADDR),
+	HBG_DUMP_GMAC_I(HBG_REG_AN_NEG_STATE_ADDR),
+	HBG_DUMP_GMAC_I(HBG_REG_TRANSMIT_CTRL_ADDR),
+	HBG_DUMP_GMAC_I(HBG_REG_REC_FILT_CTRL_ADDR),
+	HBG_DUMP_GMAC_I(HBG_REG_LINE_LOOP_BACK_ADDR),
+	HBG_DUMP_GMAC_I(HBG_REG_CF_CRC_STRIP_ADDR),
+	HBG_DUMP_GMAC_I(HBG_REG_MODE_CHANGE_EN_ADDR),
+	HBG_DUMP_GMAC_I(HBG_REG_LOOP_REG_ADDR),
+	HBG_DUMP_GMAC_I(HBG_REG_RECV_CTRL_ADDR),
+	HBG_DUMP_GMAC_I(HBG_REG_VLAN_CODE_ADDR),
+	HBG_DUMP_GMAC_I(HBG_REG_STATION_ADDR_LOW_0_ADDR),
+	HBG_DUMP_GMAC_I(HBG_REG_STATION_ADDR_HIGH_0_ADDR),
+	HBG_DUMP_GMAC_I(HBG_REG_STATION_ADDR_LOW_1_ADDR),
+	HBG_DUMP_GMAC_I(HBG_REG_STATION_ADDR_HIGH_1_ADDR),
+	HBG_DUMP_GMAC_I(HBG_REG_STATION_ADDR_LOW_2_ADDR),
+	HBG_DUMP_GMAC_I(HBG_REG_STATION_ADDR_HIGH_2_ADDR),
+	HBG_DUMP_GMAC_I(HBG_REG_STATION_ADDR_LOW_3_ADDR),
+	HBG_DUMP_GMAC_I(HBG_REG_STATION_ADDR_HIGH_3_ADDR),
+	HBG_DUMP_GMAC_I(HBG_REG_STATION_ADDR_LOW_4_ADDR),
+	HBG_DUMP_GMAC_I(HBG_REG_STATION_ADDR_HIGH_4_ADDR),
+	HBG_DUMP_GMAC_I(HBG_REG_STATION_ADDR_LOW_5_ADDR),
+	HBG_DUMP_GMAC_I(HBG_REG_STATION_ADDR_HIGH_5_ADDR),
+
+	/* pcu */
+	HBG_DUMP_PCU_I(HBG_REG_TX_FIFO_THRSLD_ADDR),
+	HBG_DUMP_PCU_I(HBG_REG_RX_FIFO_THRSLD_ADDR),
+	HBG_DUMP_PCU_I(HBG_REG_CFG_FIFO_THRSLD_ADDR),
+	HBG_DUMP_PCU_I(HBG_REG_CF_INTRPT_MSK_ADDR),
+	HBG_DUMP_PCU_I(HBG_REG_CF_INTRPT_STAT_ADDR),
+	HBG_DUMP_PCU_I(HBG_REG_CF_INTRPT_CLR_ADDR),
+	HBG_DUMP_PCU_I(HBG_REG_TX_BUS_ERR_ADDR_ADDR),
+	HBG_DUMP_PCU_I(HBG_REG_RX_BUS_ERR_ADDR_ADDR),
+	HBG_DUMP_PCU_I(HBG_REG_MAX_FRAME_LEN_ADDR),
+	HBG_DUMP_PCU_I(HBG_REG_DEBUG_ST_MCH_ADDR),
+	HBG_DUMP_PCU_I(HBG_REG_FIFO_CURR_STATUS_ADDR),
+	HBG_DUMP_PCU_I(HBG_REG_FIFO_HIST_STATUS_ADDR),
+	HBG_DUMP_PCU_I(HBG_REG_CF_CFF_DATA_NUM_ADDR),
+	HBG_DUMP_PCU_I(HBG_REG_CF_TX_PAUSE_ADDR),
+	HBG_DUMP_PCU_I(HBG_REG_RX_CFF_ADDR_ADDR),
+	HBG_DUMP_PCU_I(HBG_REG_RX_BUF_SIZE_ADDR),
+	HBG_DUMP_PCU_I(HBG_REG_BUS_CTRL_ADDR),
+	HBG_DUMP_PCU_I(HBG_REG_RX_CTRL_ADDR),
+	HBG_DUMP_PCU_I(HBG_REG_RX_PKT_MODE_ADDR),
+	HBG_DUMP_PCU_I(HBG_REG_DBG_ST0_ADDR),
+	HBG_DUMP_PCU_I(HBG_REG_DBG_ST1_ADDR),
+	HBG_DUMP_PCU_I(HBG_REG_DBG_ST2_ADDR),
+	HBG_DUMP_PCU_I(HBG_REG_BUS_RST_EN_ADDR),
+	HBG_DUMP_PCU_I(HBG_REG_CF_IND_TXINT_MSK_ADDR),
+	HBG_DUMP_PCU_I(HBG_REG_CF_IND_TXINT_STAT_ADDR),
+	HBG_DUMP_PCU_I(HBG_REG_CF_IND_TXINT_CLR_ADDR),
+	HBG_DUMP_PCU_I(HBG_REG_CF_IND_RXINT_MSK_ADDR),
+	HBG_DUMP_PCU_I(HBG_REG_CF_IND_RXINT_STAT_ADDR),
+	HBG_DUMP_PCU_I(HBG_REG_CF_IND_RXINT_CLR_ADDR),
+};
+
+static const u32 hbg_dump_type_base_array[] = {
+	[HBG_DUMP_REG_TYPE_SPEC] = 0,
+	[HBG_DUMP_REG_TYPE_MDIO] = HBG_REG_MDIO_BASE,
+	[HBG_DUMP_REG_TYPE_GMAC] = HBG_REG_SGMII_BASE,
+	[HBG_DUMP_REG_TYPE_PCU] = HBG_REG_SGMII_BASE,
+};
+
+static int hbg_ethtool_get_regs_len(struct net_device *netdev)
+{
+	return ARRAY_SIZE(hbg_dump_reg_infos) * sizeof(struct hbg_reg_info);
+}
+
+static void hbg_ethtool_get_regs(struct net_device *netdev,
+				 struct ethtool_regs *regs, void *data)
+{
+	struct hbg_priv *priv = netdev_priv(netdev);
+	struct hbg_reg_info *info;
+	u32 i, offset = 0;
+
+	regs->version = 0;
+	for (i = 0; i < ARRAY_SIZE(hbg_dump_reg_infos); i++) {
+		info = data + offset;
+
+		*info = hbg_dump_reg_infos[i];
+		info->val = hbg_reg_read(priv, info->offset);
+		info->offset -= hbg_dump_type_base_array[info->type];
+
+		offset += sizeof(*info);
+	}
+}
+
+static void hbg_ethtool_get_pauseparam(struct net_device *net_dev,
+				       struct ethtool_pauseparam *param)
+{
+	struct hbg_priv *priv = netdev_priv(net_dev);
+
+	param->autoneg = priv->mac.pause_autoneg;
+	hbg_hw_get_pause_enable(priv, &param->tx_pause, &param->rx_pause);
+}
+
+static int hbg_ethtool_set_pauseparam(struct net_device *net_dev,
+				      struct ethtool_pauseparam *param)
+{
+	struct hbg_priv *priv = netdev_priv(net_dev);
+
+	priv->mac.pause_autoneg = param->autoneg;
+	phy_set_asym_pause(priv->mac.phydev, param->rx_pause, param->tx_pause);
+
+	if (!param->autoneg)
+		hbg_hw_set_pause_enable(priv, param->tx_pause, param->rx_pause);
+
+	priv->user_def.pause_param = *param;
+	return 0;
+}
+
+static int hbg_ethtool_reset(struct net_device *netdev, u32 *flags)
+{
+	struct hbg_priv *priv = netdev_priv(netdev);
+
+	if (*flags != ETH_RESET_DEDICATED)
+		return -EOPNOTSUPP;
+
+	*flags = 0;
+	return hbg_reset(priv);
+}
+
+static void hbg_update_stats_by_info(struct hbg_priv *priv,
+				     const struct hbg_ethtool_stats *info,
+				     u32 info_len)
+{
+	const struct hbg_ethtool_stats *stats;
+	u32 i;
+
+	if (test_bit(HBG_NIC_STATE_RESETTING, &priv->state))
+		return;
+
+	for (i = 0; i < info_len; i++) {
+		stats = &info[i];
+		if (!stats->reg)
+			continue;
+
+		HBG_STATS_U(&priv->stats, stats->offset,
+			    hbg_reg_read(priv, stats->reg));
+	}
+}
+
+void hbg_update_stats(struct hbg_priv *priv)
+{
+	hbg_update_stats_by_info(priv, hbg_ethtool_stats_info,
+				 ARRAY_SIZE(hbg_ethtool_stats_info));
+	hbg_update_stats_by_info(priv, hbg_ethtool_rmon_stats_info,
+				 ARRAY_SIZE(hbg_ethtool_rmon_stats_info));
+	hbg_update_stats_by_info(priv, hbg_ethtool_mac_stats_info,
+				 ARRAY_SIZE(hbg_ethtool_mac_stats_info));
+	hbg_update_stats_by_info(priv, hbg_ethtool_ctrl_stats_info,
+				 ARRAY_SIZE(hbg_ethtool_ctrl_stats_info));
+}
+
+static int hbg_ethtool_get_sset_count(struct net_device *netdev, int stringset)
+{
+	if (stringset != ETH_SS_STATS)
+		return -EOPNOTSUPP;
+
+	return ARRAY_SIZE(hbg_ethtool_stats_info);
+}
+
+static void hbg_ethtool_get_strings(struct net_device *netdev,
+				    u32 stringset, u8 *data)
+{
+	u32 i;
+
+	if (stringset != ETH_SS_STATS)
+		return;
+
+	for (i = 0; i < ARRAY_SIZE(hbg_ethtool_stats_info); i++)
+		ethtool_puts(&data, hbg_ethtool_stats_info[i].name);
+}
+
+static void hbg_ethtool_get_stats(struct net_device *netdev,
+				  struct ethtool_stats *stats, u64 *data)
+{
+	struct hbg_priv *priv = netdev_priv(netdev);
+	u32 i;
+
+	hbg_update_stats(priv);
+	for (i = 0; i < ARRAY_SIZE(hbg_ethtool_stats_info); i++)
+		*data++ = HBG_STATS_R(&priv->stats,
+				      hbg_ethtool_stats_info[i].offset);
+}
+
+static void hbg_ethtool_get_pause_stats(struct net_device *netdev,
+					struct ethtool_pause_stats *epstats)
+{
+	struct hbg_priv *priv = netdev_priv(netdev);
+	struct hbg_stats *stats = &priv->stats;
+
+	hbg_update_stats(priv);
+	epstats->rx_pause_frames = stats->rx_pause_macctl_frame_cnt;
+	epstats->tx_pause_frames = stats->tx_pause_frame_cnt;
+}
+
+static void hbg_ethtool_get_eth_mac_stats(struct net_device *netdev,
+					  struct ethtool_eth_mac_stats *emstats)
+{
+	struct hbg_priv *priv = netdev_priv(netdev);
+	struct hbg_stats *stats = &priv->stats;
+
+	hbg_update_stats(priv);
+	emstats->FramesTransmittedOK = stats->tx_trans_pkt_cnt;
+	emstats->FramesReceivedOK = stats->rx_trans_pkt_cnt;
+	emstats->FrameCheckSequenceErrors = stats->rx_fcs_error_cnt;
+	emstats->AlignmentErrors = stats->rx_align_error_cnt;
+	emstats->OctetsTransmittedOK = stats->tx_octets_total_ok_cnt;
+	emstats->OctetsReceivedOK = stats->rx_octets_total_ok_cnt;
+
+	emstats->MulticastFramesXmittedOK = stats->tx_mc_pkt_cnt;
+	emstats->BroadcastFramesXmittedOK = stats->tx_bc_pkt_cnt;
+	emstats->MulticastFramesReceivedOK = stats->rx_mc_pkt_cnt;
+	emstats->BroadcastFramesReceivedOK = stats->rx_bc_pkt_cnt;
+	emstats->InRangeLengthErrors = stats->rx_fcs_error_cnt +
+				       stats->rx_jabber_err_cnt +
+				       stats->rx_unknown_macctl_frame_cnt +
+				       stats->rx_bufrq_err_cnt +
+				       stats->rx_we_err_cnt;
+	emstats->OutOfRangeLengthField = stats->rx_frame_short_err_cnt +
+					 stats->rx_frame_runt_err_cnt +
+					 stats->rx_lengthfield_err_cnt +
+					 stats->rx_frame_long_err_cnt +
+					 stats->rx_frame_very_long_err_cnt;
+	emstats->FrameTooLongErrors = stats->rx_frame_long_err_cnt +
+				      stats->rx_frame_very_long_err_cnt;
+}
+
+static void
+hbg_ethtool_get_eth_ctrl_stats(struct net_device *netdev,
+			       struct ethtool_eth_ctrl_stats *ecstats)
+{
+	struct hbg_priv *priv = netdev_priv(netdev);
+	struct hbg_stats *s = &priv->stats;
+
+	hbg_update_stats(priv);
+	ecstats->MACControlFramesTransmitted = s->tx_pause_frame_cnt;
+	ecstats->MACControlFramesReceived = s->rx_pause_macctl_frame_cnt;
+	ecstats->UnsupportedOpcodesReceived = s->rx_unknown_macctl_frame_cnt;
+}
+
+static const struct ethtool_rmon_hist_range hbg_rmon_ranges[] = {
+	{    0,    64 },
+	{   65,   127 },
+	{  128,   255 },
+	{  256,   511 },
+	{  512,  1023 },
+	{ 1024,  1518 },
+	{ 1519,  4095 },
+};
+
+static void
+hbg_ethtool_get_rmon_stats(struct net_device *netdev,
+			   struct ethtool_rmon_stats *rmon_stats,
+			   const struct ethtool_rmon_hist_range **ranges)
+{
+	struct hbg_priv *priv = netdev_priv(netdev);
+	struct hbg_stats *stats = &priv->stats;
+
+	hbg_update_stats(priv);
+	rmon_stats->undersize_pkts = stats->rx_frame_short_err_cnt +
+				     stats->rx_frame_runt_err_cnt +
+				     stats->rx_lengthfield_err_cnt;
+	rmon_stats->oversize_pkts = stats->rx_frame_long_err_cnt +
+				    stats->rx_frame_very_long_err_cnt;
+	rmon_stats->fragments = stats->rx_desc_frag_cnt;
+	rmon_stats->hist[0] = stats->rx_framesize_64;
+	rmon_stats->hist[1] = stats->rx_framesize_65_127;
+	rmon_stats->hist[2] = stats->rx_framesize_128_255;
+	rmon_stats->hist[3] = stats->rx_framesize_256_511;
+	rmon_stats->hist[4] = stats->rx_framesize_512_1023;
+	rmon_stats->hist[5] = stats->rx_framesize_1024_1518;
+	rmon_stats->hist[6] = stats->rx_framesize_bt_1518;
+
+	rmon_stats->hist_tx[0] = stats->tx_framesize_64;
+	rmon_stats->hist_tx[1] = stats->tx_framesize_65_127;
+	rmon_stats->hist_tx[2] = stats->tx_framesize_128_255;
+	rmon_stats->hist_tx[3] = stats->tx_framesize_256_511;
+	rmon_stats->hist_tx[4] = stats->tx_framesize_512_1023;
+	rmon_stats->hist_tx[5] = stats->tx_framesize_1024_1518;
+	rmon_stats->hist_tx[6] = stats->tx_framesize_bt_1518;
+
+	*ranges = hbg_rmon_ranges;
+}
+
+static const struct ethtool_ops hbg_ethtool_ops = {
+	.get_link		= ethtool_op_get_link,
+	.get_link_ksettings	= phy_ethtool_get_link_ksettings,
+	.set_link_ksettings	= phy_ethtool_set_link_ksettings,
+	.get_regs_len		= hbg_ethtool_get_regs_len,
+	.get_regs		= hbg_ethtool_get_regs,
+	.get_pauseparam         = hbg_ethtool_get_pauseparam,
+	.set_pauseparam         = hbg_ethtool_set_pauseparam,
+	.reset			= hbg_ethtool_reset,
+	.nway_reset		= phy_ethtool_nway_reset,
+	.get_sset_count		= hbg_ethtool_get_sset_count,
+	.get_strings		= hbg_ethtool_get_strings,
+	.get_ethtool_stats	= hbg_ethtool_get_stats,
+	.get_pause_stats	= hbg_ethtool_get_pause_stats,
+	.get_eth_mac_stats	= hbg_ethtool_get_eth_mac_stats,
+	.get_eth_ctrl_stats	= hbg_ethtool_get_eth_ctrl_stats,
+	.get_rmon_stats		= hbg_ethtool_get_rmon_stats,
+};
+
+void hbg_ethtool_set_ops(struct net_device *netdev)
+{
+	netdev->ethtool_ops = &hbg_ethtool_ops;
+}
diff --git a/drivers/net/ethernet/hisilicon/hibmcge/hbg_ethtool.h b/drivers/net/ethernet/hisilicon/hibmcge/hbg_ethtool.h
new file mode 100644
index 000000000000..e173155b146a
--- /dev/null
+++ b/drivers/net/ethernet/hisilicon/hibmcge/hbg_ethtool.h
@@ -0,0 +1,16 @@
+/* SPDX-License-Identifier: GPL-2.0+ */
+/* Copyright (c) 2024 Hisilicon Limited. */
+
+#ifndef __HBG_ETHTOOL_H
+#define __HBG_ETHTOOL_H
+
+#include <linux/netdevice.h>
+
+#define HBG_STATS_FIELD_OFF(f) (offsetof(struct hbg_stats, f))
+#define HBG_STATS_R(p, offset) (*(u64 *)((u8 *)(p) + (offset)))
+#define HBG_STATS_U(p, offset, val) (HBG_STATS_R(p, offset) += (val))
+
+void hbg_ethtool_set_ops(struct net_device *netdev);
+void hbg_update_stats(struct hbg_priv *priv);
+
+#endif
diff --git a/drivers/net/ethernet/hisilicon/hibmcge/hbg_hw.c b/drivers/net/ethernet/hisilicon/hibmcge/hbg_hw.c
new file mode 100644
index 000000000000..d0aa0661ecd4
--- /dev/null
+++ b/drivers/net/ethernet/hisilicon/hibmcge/hbg_hw.c
@@ -0,0 +1,396 @@
+// SPDX-License-Identifier: GPL-2.0+
+// Copyright (c) 2024 Hisilicon Limited.
+
+#include <linux/etherdevice.h>
+#include <linux/ethtool.h>
+#include <linux/if_vlan.h>
+#include <linux/iopoll.h>
+#include <linux/minmax.h>
+#include "hbg_common.h"
+#include "hbg_hw.h"
+#include "hbg_reg.h"
+
+#define HBG_HW_EVENT_WAIT_TIMEOUT_US	(2 * 1000 * 1000)
+#define HBG_HW_EVENT_WAIT_INTERVAL_US	(10 * 1000)
+#define HBG_MAC_LINK_WAIT_TIMEOUT_US	(500 * 1000)
+#define HBG_MAC_LINK_WAIT_INTERVAL_US	(5 * 1000)
+/* little endian or big endian.
+ * ctrl means packet description, data means skb packet data
+ */
+#define HBG_ENDIAN_CTRL_LE_DATA_BE	0x0
+#define HBG_PCU_FRAME_LEN_PLUS 4
+
+#define HBG_FIFO_TX_FULL_THRSLD		0x3F0
+#define HBG_FIFO_TX_EMPTY_THRSLD	0x1F0
+#define HBG_FIFO_RX_FULL_THRSLD		0x240
+#define HBG_FIFO_RX_EMPTY_THRSLD	0x190
+#define HBG_CFG_FIFO_FULL_THRSLD	0x10
+#define HBG_CFG_FIFO_EMPTY_THRSLD	0x01
+
+static bool hbg_hw_spec_is_valid(struct hbg_priv *priv)
+{
+	return hbg_reg_read(priv, HBG_REG_SPEC_VALID_ADDR) &&
+	       !hbg_reg_read(priv, HBG_REG_EVENT_REQ_ADDR);
+}
+
+int hbg_hw_event_notify(struct hbg_priv *priv,
+			enum hbg_hw_event_type event_type)
+{
+	bool is_valid;
+	int ret;
+
+	if (test_and_set_bit(HBG_NIC_STATE_EVENT_HANDLING, &priv->state))
+		return -EBUSY;
+
+	/* notify */
+	hbg_reg_write(priv, HBG_REG_EVENT_REQ_ADDR, event_type);
+
+	ret = read_poll_timeout(hbg_hw_spec_is_valid, is_valid, is_valid,
+				HBG_HW_EVENT_WAIT_INTERVAL_US,
+				HBG_HW_EVENT_WAIT_TIMEOUT_US,
+				HBG_HW_EVENT_WAIT_INTERVAL_US, priv);
+
+	clear_bit(HBG_NIC_STATE_EVENT_HANDLING, &priv->state);
+
+	if (ret)
+		dev_err(&priv->pdev->dev,
+			"event %d wait timeout\n", event_type);
+
+	return ret;
+}
+
+static int hbg_hw_dev_specs_init(struct hbg_priv *priv)
+{
+	struct hbg_dev_specs *specs = &priv->dev_specs;
+	u64 mac_addr;
+
+	if (!hbg_hw_spec_is_valid(priv)) {
+		dev_err(&priv->pdev->dev, "dev_specs not init\n");
+		return -EINVAL;
+	}
+
+	specs->mac_id = hbg_reg_read(priv, HBG_REG_MAC_ID_ADDR);
+	specs->phy_addr = hbg_reg_read(priv, HBG_REG_PHY_ID_ADDR);
+	specs->mdio_frequency = hbg_reg_read(priv, HBG_REG_MDIO_FREQ_ADDR);
+	specs->max_mtu = hbg_reg_read(priv, HBG_REG_MAX_MTU_ADDR);
+	specs->min_mtu = hbg_reg_read(priv, HBG_REG_MIN_MTU_ADDR);
+	specs->vlan_layers = hbg_reg_read(priv, HBG_REG_VLAN_LAYERS_ADDR);
+	specs->rx_fifo_num = hbg_reg_read(priv, HBG_REG_RX_FIFO_NUM_ADDR);
+	specs->tx_fifo_num = hbg_reg_read(priv, HBG_REG_TX_FIFO_NUM_ADDR);
+	specs->uc_mac_num = hbg_reg_read(priv, HBG_REG_UC_MAC_NUM_ADDR);
+
+	mac_addr = hbg_reg_read64(priv, HBG_REG_MAC_ADDR_ADDR);
+	u64_to_ether_addr(mac_addr, (u8 *)specs->mac_addr.sa_data);
+
+	if (!is_valid_ether_addr((u8 *)specs->mac_addr.sa_data))
+		return -EADDRNOTAVAIL;
+
+	specs->max_frame_len = HBG_PCU_CACHE_LINE_SIZE + specs->max_mtu;
+	specs->rx_buf_size = HBG_PACKET_HEAD_SIZE + specs->max_frame_len;
+	return 0;
+}
+
+u32 hbg_hw_get_irq_status(struct hbg_priv *priv)
+{
+	u32 status;
+
+	status = hbg_reg_read(priv, HBG_REG_CF_INTRPT_STAT_ADDR);
+
+	hbg_field_modify(status, HBG_INT_MSK_TX_B,
+			 hbg_reg_read(priv, HBG_REG_CF_IND_TXINT_STAT_ADDR));
+	hbg_field_modify(status, HBG_INT_MSK_RX_B,
+			 hbg_reg_read(priv, HBG_REG_CF_IND_RXINT_STAT_ADDR));
+
+	return status;
+}
+
+void hbg_hw_irq_clear(struct hbg_priv *priv, u32 mask)
+{
+	if (FIELD_GET(HBG_INT_MSK_TX_B, mask))
+		return hbg_reg_write(priv, HBG_REG_CF_IND_TXINT_CLR_ADDR, 0x1);
+
+	if (FIELD_GET(HBG_INT_MSK_RX_B, mask))
+		return hbg_reg_write(priv, HBG_REG_CF_IND_RXINT_CLR_ADDR, 0x1);
+
+	return hbg_reg_write(priv, HBG_REG_CF_INTRPT_CLR_ADDR, mask);
+}
+
+bool hbg_hw_irq_is_enabled(struct hbg_priv *priv, u32 mask)
+{
+	if (FIELD_GET(HBG_INT_MSK_TX_B, mask))
+		return hbg_reg_read(priv, HBG_REG_CF_IND_TXINT_MSK_ADDR);
+
+	if (FIELD_GET(HBG_INT_MSK_RX_B, mask))
+		return hbg_reg_read(priv, HBG_REG_CF_IND_RXINT_MSK_ADDR);
+
+	return hbg_reg_read(priv, HBG_REG_CF_INTRPT_MSK_ADDR) & mask;
+}
+
+void hbg_hw_irq_enable(struct hbg_priv *priv, u32 mask, bool enable)
+{
+	u32 value;
+
+	if (FIELD_GET(HBG_INT_MSK_TX_B, mask))
+		return hbg_reg_write(priv,
+				     HBG_REG_CF_IND_TXINT_MSK_ADDR, enable);
+
+	if (FIELD_GET(HBG_INT_MSK_RX_B, mask))
+		return hbg_reg_write(priv,
+				     HBG_REG_CF_IND_RXINT_MSK_ADDR, enable);
+
+	value = hbg_reg_read(priv, HBG_REG_CF_INTRPT_MSK_ADDR);
+	if (enable)
+		value |= mask;
+	else
+		value &= ~mask;
+
+	hbg_reg_write(priv, HBG_REG_CF_INTRPT_MSK_ADDR, value);
+}
+
+void hbg_hw_set_uc_addr(struct hbg_priv *priv, u64 mac_addr, u32 index)
+{
+	u32 addr;
+
+	/* mac addr is u64, so the addr offset is 0x8 */
+	addr = HBG_REG_STATION_ADDR_LOW_2_ADDR + (index * 0x8);
+	hbg_reg_write64(priv, addr, mac_addr);
+}
+
+static void hbg_hw_set_pcu_max_frame_len(struct hbg_priv *priv,
+					 u16 max_frame_len)
+{
+	max_frame_len = max_t(u32, max_frame_len, ETH_DATA_LEN);
+
+	/* lower two bits of value must be set to 0 */
+	max_frame_len = round_up(max_frame_len, HBG_PCU_FRAME_LEN_PLUS);
+
+	hbg_reg_write_field(priv, HBG_REG_MAX_FRAME_LEN_ADDR,
+			    HBG_REG_MAX_FRAME_LEN_M, max_frame_len);
+}
+
+static void hbg_hw_set_mac_max_frame_len(struct hbg_priv *priv,
+					 u16 max_frame_size)
+{
+	hbg_reg_write_field(priv, HBG_REG_MAX_FRAME_SIZE_ADDR,
+			    HBG_REG_MAX_FRAME_LEN_M, max_frame_size);
+}
+
+void hbg_hw_set_mtu(struct hbg_priv *priv, u16 mtu)
+{
+	/* burst_len BIT(29) set to 1 can improve the TX performance.
+	 * But packet drop occurs when mtu > 2000.
+	 * So, BIT(29) reset to 0 when mtu > 2000.
+	 */
+	u32 burst_len_bit = (mtu > 2000) ? 0 : 1;
+	u32 frame_len;
+
+	frame_len = mtu + VLAN_HLEN * priv->dev_specs.vlan_layers +
+		    ETH_HLEN + ETH_FCS_LEN;
+
+	hbg_hw_set_pcu_max_frame_len(priv, frame_len);
+	hbg_hw_set_mac_max_frame_len(priv, frame_len);
+
+	hbg_reg_write_field(priv, HBG_REG_BRUST_LENGTH_ADDR,
+			    HBG_REG_BRUST_LENGTH_B, burst_len_bit);
+}
+
+void hbg_hw_mac_enable(struct hbg_priv *priv, u32 enable)
+{
+	hbg_reg_write_field(priv, HBG_REG_PORT_ENABLE_ADDR,
+			    HBG_REG_PORT_ENABLE_TX_B, enable);
+	hbg_reg_write_field(priv, HBG_REG_PORT_ENABLE_ADDR,
+			    HBG_REG_PORT_ENABLE_RX_B, enable);
+}
+
+u32 hbg_hw_get_fifo_used_num(struct hbg_priv *priv, enum hbg_dir dir)
+{
+	if (dir & HBG_DIR_TX)
+		return hbg_reg_read_field(priv, HBG_REG_CF_CFF_DATA_NUM_ADDR,
+					  HBG_REG_CF_CFF_DATA_NUM_ADDR_TX_M);
+
+	if (dir & HBG_DIR_RX)
+		return hbg_reg_read_field(priv, HBG_REG_CF_CFF_DATA_NUM_ADDR,
+					  HBG_REG_CF_CFF_DATA_NUM_ADDR_RX_M);
+
+	return 0;
+}
+
+void hbg_hw_set_tx_desc(struct hbg_priv *priv, struct hbg_tx_desc *tx_desc)
+{
+	hbg_reg_write(priv, HBG_REG_TX_CFF_ADDR_0_ADDR, tx_desc->word0);
+	hbg_reg_write(priv, HBG_REG_TX_CFF_ADDR_1_ADDR, tx_desc->word1);
+	hbg_reg_write(priv, HBG_REG_TX_CFF_ADDR_2_ADDR, tx_desc->word2);
+	hbg_reg_write(priv, HBG_REG_TX_CFF_ADDR_3_ADDR, tx_desc->word3);
+}
+
+void hbg_hw_fill_buffer(struct hbg_priv *priv, u32 buffer_dma_addr)
+{
+	hbg_reg_write(priv, HBG_REG_RX_CFF_ADDR_ADDR, buffer_dma_addr);
+}
+
+void hbg_hw_adjust_link(struct hbg_priv *priv, u32 speed, u32 duplex)
+{
+	u32 link_status;
+	int ret;
+
+	hbg_hw_mac_enable(priv, HBG_STATUS_DISABLE);
+
+	hbg_reg_write_field(priv, HBG_REG_PORT_MODE_ADDR,
+			    HBG_REG_PORT_MODE_M, speed);
+	hbg_reg_write_field(priv, HBG_REG_DUPLEX_TYPE_ADDR,
+			    HBG_REG_DUPLEX_B, duplex);
+
+	hbg_hw_event_notify(priv, HBG_HW_EVENT_CORE_RESET);
+
+	hbg_hw_mac_enable(priv, HBG_STATUS_ENABLE);
+
+	/* wait MAC link up */
+	ret = readl_poll_timeout(priv->io_base + HBG_REG_AN_NEG_STATE_ADDR,
+				 link_status,
+				 FIELD_GET(HBG_REG_AN_NEG_STATE_NP_LINK_OK_B,
+					   link_status),
+				 HBG_MAC_LINK_WAIT_INTERVAL_US,
+				 HBG_MAC_LINK_WAIT_TIMEOUT_US);
+	if (ret)
+		hbg_np_link_fail_task_schedule(priv);
+}
+
+/* only support uc filter */
+void hbg_hw_set_mac_filter_enable(struct hbg_priv *priv, u32 enable)
+{
+	hbg_reg_write_field(priv, HBG_REG_REC_FILT_CTRL_ADDR,
+			    HBG_REG_REC_FILT_CTRL_UC_MATCH_EN_B, enable);
+
+	/* only uc filter is supported, so set all bits of mc mask reg to 1 */
+	hbg_reg_write64(priv, HBG_REG_STATION_ADDR_LOW_MSK_0, U64_MAX);
+	hbg_reg_write64(priv, HBG_REG_STATION_ADDR_LOW_MSK_1, U64_MAX);
+}
+
+void hbg_hw_set_pause_enable(struct hbg_priv *priv, u32 tx_en, u32 rx_en)
+{
+	hbg_reg_write_field(priv, HBG_REG_PAUSE_ENABLE_ADDR,
+			    HBG_REG_PAUSE_ENABLE_TX_B, tx_en);
+	hbg_reg_write_field(priv, HBG_REG_PAUSE_ENABLE_ADDR,
+			    HBG_REG_PAUSE_ENABLE_RX_B, rx_en);
+
+	hbg_reg_write_field(priv, HBG_REG_REC_FILT_CTRL_ADDR,
+			    HBG_REG_REC_FILT_CTRL_PAUSE_FRM_PASS_B, rx_en);
+}
+
+void hbg_hw_get_pause_enable(struct hbg_priv *priv, u32 *tx_en, u32 *rx_en)
+{
+	*tx_en = hbg_reg_read_field(priv, HBG_REG_PAUSE_ENABLE_ADDR,
+				    HBG_REG_PAUSE_ENABLE_TX_B);
+	*rx_en = hbg_reg_read_field(priv, HBG_REG_PAUSE_ENABLE_ADDR,
+				    HBG_REG_PAUSE_ENABLE_RX_B);
+}
+
+void hbg_hw_set_rx_pause_mac_addr(struct hbg_priv *priv, u64 mac_addr)
+{
+	hbg_reg_write64(priv, HBG_REG_FD_FC_ADDR_LOW_ADDR, mac_addr);
+}
+
+static void hbg_hw_set_fifo_thrsld(struct hbg_priv *priv,
+				   u32 full, u32 empty, enum hbg_dir dir)
+{
+	u32 value = 0;
+
+	value |= FIELD_PREP(HBG_REG_FIFO_THRSLD_FULL_M, full);
+	value |= FIELD_PREP(HBG_REG_FIFO_THRSLD_EMPTY_M, empty);
+
+	if (dir & HBG_DIR_TX)
+		hbg_reg_write(priv, HBG_REG_TX_FIFO_THRSLD_ADDR, value);
+
+	if (dir & HBG_DIR_RX)
+		hbg_reg_write(priv, HBG_REG_RX_FIFO_THRSLD_ADDR, value);
+}
+
+static void hbg_hw_set_cfg_fifo_thrsld(struct hbg_priv *priv,
+				       u32 full, u32 empty, enum hbg_dir dir)
+{
+	u32 value;
+
+	value = hbg_reg_read(priv, HBG_REG_CFG_FIFO_THRSLD_ADDR);
+
+	if (dir & HBG_DIR_TX) {
+		value |= FIELD_PREP(HBG_REG_CFG_FIFO_THRSLD_TX_FULL_M, full);
+		value |= FIELD_PREP(HBG_REG_CFG_FIFO_THRSLD_TX_EMPTY_M, empty);
+	}
+
+	if (dir & HBG_DIR_RX) {
+		value |= FIELD_PREP(HBG_REG_CFG_FIFO_THRSLD_RX_FULL_M, full);
+		value |= FIELD_PREP(HBG_REG_CFG_FIFO_THRSLD_RX_EMPTY_M, empty);
+	}
+
+	hbg_reg_write(priv, HBG_REG_CFG_FIFO_THRSLD_ADDR, value);
+}
+
+static void hbg_hw_init_transmit_ctrl(struct hbg_priv *priv)
+{
+	u32 ctrl = 0;
+
+	ctrl |= FIELD_PREP(HBG_REG_TRANSMIT_CTRL_AN_EN_B, HBG_STATUS_ENABLE);
+	ctrl |= FIELD_PREP(HBG_REG_TRANSMIT_CTRL_CRC_ADD_B, HBG_STATUS_ENABLE);
+	ctrl |= FIELD_PREP(HBG_REG_TRANSMIT_CTRL_PAD_EN_B, HBG_STATUS_ENABLE);
+
+	hbg_reg_write(priv, HBG_REG_TRANSMIT_CTRL_ADDR, ctrl);
+}
+
+static void hbg_hw_init_rx_ctrl(struct hbg_priv *priv)
+{
+	u32 ctrl = 0;
+
+	ctrl |= FIELD_PREP(HBG_REG_RX_CTRL_RX_GET_ADDR_MODE_B,
+			   HBG_STATUS_ENABLE);
+	ctrl |= FIELD_PREP(HBG_REG_RX_CTRL_TIME_INF_EN_B, HBG_STATUS_DISABLE);
+	ctrl |= FIELD_PREP(HBG_REG_RX_CTRL_RXBUF_1ST_SKIP_SIZE_M, HBG_RX_SKIP1);
+	ctrl |= FIELD_PREP(HBG_REG_RX_CTRL_RXBUF_1ST_SKIP_SIZE2_M,
+			   HBG_RX_SKIP2);
+	ctrl |= FIELD_PREP(HBG_REG_RX_CTRL_RX_ALIGN_NUM_M, NET_IP_ALIGN);
+	ctrl |= FIELD_PREP(HBG_REG_RX_CTRL_PORT_NUM, priv->dev_specs.mac_id);
+
+	hbg_reg_write(priv, HBG_REG_RX_CTRL_ADDR, ctrl);
+}
+
+static void hbg_hw_init_rx_control(struct hbg_priv *priv)
+{
+	hbg_hw_init_rx_ctrl(priv);
+
+	/* parse from L2 layer */
+	hbg_reg_write_field(priv, HBG_REG_RX_PKT_MODE_ADDR,
+			    HBG_REG_RX_PKT_MODE_PARSE_MODE_M, 0x1);
+
+	hbg_reg_write_field(priv, HBG_REG_RECV_CTRL_ADDR,
+			    HBG_REG_RECV_CTRL_STRIP_PAD_EN_B,
+			    HBG_STATUS_ENABLE);
+	hbg_reg_write_field(priv, HBG_REG_RX_BUF_SIZE_ADDR,
+			    HBG_REG_RX_BUF_SIZE_M, priv->dev_specs.rx_buf_size);
+	hbg_reg_write_field(priv, HBG_REG_CF_CRC_STRIP_ADDR,
+			    HBG_REG_CF_CRC_STRIP_B, HBG_STATUS_DISABLE);
+}
+
+int hbg_hw_init(struct hbg_priv *priv)
+{
+	int ret;
+
+	ret = hbg_hw_dev_specs_init(priv);
+	if (ret)
+		return ret;
+
+	hbg_reg_write_field(priv, HBG_REG_BUS_CTRL_ADDR,
+			    HBG_REG_BUS_CTRL_ENDIAN_M,
+			    HBG_ENDIAN_CTRL_LE_DATA_BE);
+	hbg_reg_write_field(priv, HBG_REG_MODE_CHANGE_EN_ADDR,
+			    HBG_REG_MODE_CHANGE_EN_B, HBG_STATUS_ENABLE);
+
+	hbg_hw_init_rx_control(priv);
+	hbg_hw_init_transmit_ctrl(priv);
+
+	hbg_hw_set_fifo_thrsld(priv, HBG_FIFO_TX_FULL_THRSLD,
+			       HBG_FIFO_TX_EMPTY_THRSLD, HBG_DIR_TX);
+	hbg_hw_set_fifo_thrsld(priv, HBG_FIFO_RX_FULL_THRSLD,
+			       HBG_FIFO_RX_EMPTY_THRSLD, HBG_DIR_RX);
+	hbg_hw_set_cfg_fifo_thrsld(priv, HBG_CFG_FIFO_FULL_THRSLD,
+				   HBG_CFG_FIFO_EMPTY_THRSLD, HBG_DIR_TX_RX);
+	return 0;
+}
diff --git a/drivers/net/ethernet/hisilicon/hibmcge/hbg_hw.h b/drivers/net/ethernet/hisilicon/hibmcge/hbg_hw.h
new file mode 100644
index 000000000000..a4a049b5121d
--- /dev/null
+++ b/drivers/net/ethernet/hisilicon/hibmcge/hbg_hw.h
@@ -0,0 +1,63 @@
+/* SPDX-License-Identifier: GPL-2.0+ */
+/* Copyright (c) 2024 Hisilicon Limited. */
+
+#ifndef __HBG_HW_H
+#define __HBG_HW_H
+
+#include <linux/bitfield.h>
+#include <linux/io-64-nonatomic-lo-hi.h>
+
+static inline u32 hbg_reg_read(struct hbg_priv *priv, u32 addr)
+{
+	return readl(priv->io_base + addr);
+}
+
+static inline void hbg_reg_write(struct hbg_priv *priv, u32 addr, u32 value)
+{
+	writel(value, priv->io_base + addr);
+}
+
+static inline u64 hbg_reg_read64(struct hbg_priv *priv, u32 addr)
+{
+	return lo_hi_readq(priv->io_base + addr);
+}
+
+static inline void hbg_reg_write64(struct hbg_priv *priv, u32 addr, u64 value)
+{
+	lo_hi_writeq(value, priv->io_base + addr);
+}
+
+#define hbg_reg_read_field(priv, addr, mask) \
+		FIELD_GET(mask, hbg_reg_read(priv, addr))
+
+#define hbg_field_modify(reg_value, mask, value) ({	\
+		(reg_value) &= ~(mask);			\
+		(reg_value) |= FIELD_PREP(mask, value); })
+
+#define hbg_reg_write_field(priv, addr, mask, val) ({		\
+		typeof(priv) _priv = (priv);			\
+		typeof(addr) _addr = (addr);			\
+		u32 _value = hbg_reg_read(_priv, _addr);	\
+		hbg_field_modify(_value, mask, val);		\
+		hbg_reg_write(_priv, _addr, _value); })
+
+int hbg_hw_event_notify(struct hbg_priv *priv,
+			enum hbg_hw_event_type event_type);
+int hbg_hw_init(struct hbg_priv *priv);
+void hbg_hw_adjust_link(struct hbg_priv *priv, u32 speed, u32 duplex);
+u32 hbg_hw_get_irq_status(struct hbg_priv *priv);
+void hbg_hw_irq_clear(struct hbg_priv *priv, u32 mask);
+bool hbg_hw_irq_is_enabled(struct hbg_priv *priv, u32 mask);
+void hbg_hw_irq_enable(struct hbg_priv *priv, u32 mask, bool enable);
+void hbg_hw_set_mtu(struct hbg_priv *priv, u16 mtu);
+void hbg_hw_mac_enable(struct hbg_priv *priv, u32 enable);
+void hbg_hw_set_uc_addr(struct hbg_priv *priv, u64 mac_addr, u32 index);
+u32 hbg_hw_get_fifo_used_num(struct hbg_priv *priv, enum hbg_dir dir);
+void hbg_hw_set_tx_desc(struct hbg_priv *priv, struct hbg_tx_desc *tx_desc);
+void hbg_hw_fill_buffer(struct hbg_priv *priv, u32 buffer_dma_addr);
+void hbg_hw_set_mac_filter_enable(struct hbg_priv *priv, u32 enable);
+void hbg_hw_set_pause_enable(struct hbg_priv *priv, u32 tx_en, u32 rx_en);
+void hbg_hw_get_pause_enable(struct hbg_priv *priv, u32 *tx_en, u32 *rx_en);
+void hbg_hw_set_rx_pause_mac_addr(struct hbg_priv *priv, u64 mac_addr);
+
+#endif
diff --git a/drivers/net/ethernet/hisilicon/hibmcge/hbg_irq.c b/drivers/net/ethernet/hisilicon/hibmcge/hbg_irq.c
new file mode 100644
index 000000000000..8af0bc4cca21
--- /dev/null
+++ b/drivers/net/ethernet/hisilicon/hibmcge/hbg_irq.c
@@ -0,0 +1,144 @@
+// SPDX-License-Identifier: GPL-2.0+
+// Copyright (c) 2024 Hisilicon Limited.
+
+#include <linux/interrupt.h>
+#include "hbg_irq.h"
+#include "hbg_hw.h"
+
+static void hbg_irq_handle_err(struct hbg_priv *priv,
+			       const struct hbg_irq_info *irq_info)
+{
+	if (irq_info->need_print)
+		dev_err(&priv->pdev->dev,
+			"receive error interrupt: %s\n", irq_info->name);
+
+	if (irq_info->need_reset)
+		hbg_err_reset_task_schedule(priv);
+}
+
+static void hbg_irq_handle_tx(struct hbg_priv *priv,
+			      const struct hbg_irq_info *irq_info)
+{
+	napi_schedule(&priv->tx_ring.napi);
+}
+
+static void hbg_irq_handle_rx(struct hbg_priv *priv,
+			      const struct hbg_irq_info *irq_info)
+{
+	napi_schedule(&priv->rx_ring.napi);
+}
+
+static void hbg_irq_handle_rx_buf_val(struct hbg_priv *priv,
+				      const struct hbg_irq_info *irq_info)
+{
+	priv->stats.rx_fifo_less_empty_thrsld_cnt++;
+}
+
+#define HBG_IRQ_I(name, handle) \
+	{#name, HBG_INT_MSK_##name##_B, false, false, false, handle}
+#define HBG_ERR_IRQ_I(name, need_print, ndde_reset) \
+	{#name, HBG_INT_MSK_##name##_B, true, need_print, \
+	ndde_reset, hbg_irq_handle_err}
+
+static const struct hbg_irq_info hbg_irqs[] = {
+	HBG_IRQ_I(RX, hbg_irq_handle_rx),
+	HBG_IRQ_I(TX, hbg_irq_handle_tx),
+	HBG_ERR_IRQ_I(TX_PKT_CPL, true, true),
+	HBG_ERR_IRQ_I(MAC_MII_FIFO_ERR, true, true),
+	HBG_ERR_IRQ_I(MAC_PCS_RX_FIFO_ERR, true, true),
+	HBG_ERR_IRQ_I(MAC_PCS_TX_FIFO_ERR, true, true),
+	HBG_ERR_IRQ_I(MAC_APP_RX_FIFO_ERR, true, true),
+	HBG_ERR_IRQ_I(MAC_APP_TX_FIFO_ERR, true, true),
+	HBG_ERR_IRQ_I(SRAM_PARITY_ERR, true, false),
+	HBG_ERR_IRQ_I(TX_AHB_ERR, true, true),
+	HBG_IRQ_I(RX_BUF_AVL, hbg_irq_handle_rx_buf_val),
+	HBG_ERR_IRQ_I(REL_BUF_ERR, true, false),
+	HBG_ERR_IRQ_I(TXCFG_AVL, false, false),
+	HBG_ERR_IRQ_I(TX_DROP, false, false),
+	HBG_ERR_IRQ_I(RX_DROP, false, false),
+	HBG_ERR_IRQ_I(RX_AHB_ERR, true, false),
+	HBG_ERR_IRQ_I(MAC_FIFO_ERR, true, true),
+	HBG_ERR_IRQ_I(RBREQ_ERR, true, true),
+	HBG_ERR_IRQ_I(WE_ERR, true, true),
+};
+
+static irqreturn_t hbg_irq_handle(int irq_num, void *p)
+{
+	const struct hbg_irq_info *info;
+	struct hbg_priv *priv = p;
+	u32 status;
+	u32 i;
+
+	status = hbg_hw_get_irq_status(priv);
+	for (i = 0; i < priv->vectors.info_array_len; i++) {
+		info = &priv->vectors.info_array[i];
+		if (status & info->mask) {
+			if (!hbg_hw_irq_is_enabled(priv, info->mask))
+				continue;
+
+			hbg_hw_irq_enable(priv, info->mask, false);
+			hbg_hw_irq_clear(priv, info->mask);
+
+			priv->vectors.stats_array[i]++;
+			if (info->irq_handle)
+				info->irq_handle(priv, info);
+
+			if (info->re_enable)
+				hbg_hw_irq_enable(priv, info->mask, true);
+		}
+	}
+
+	return IRQ_HANDLED;
+}
+
+static const char *irq_names_map[HBG_VECTOR_NUM] = { "tx", "rx",
+						     "err", "mdio" };
+
+int hbg_irq_init(struct hbg_priv *priv)
+{
+	struct hbg_vector *vectors = &priv->vectors;
+	struct device *dev = &priv->pdev->dev;
+	int ret, id;
+	u32 i;
+
+	/* used pcim_enable_device(),  so the vectors become device managed */
+	ret = pci_alloc_irq_vectors(priv->pdev, HBG_VECTOR_NUM, HBG_VECTOR_NUM,
+				    PCI_IRQ_MSI | PCI_IRQ_MSIX);
+	if (ret < 0)
+		return dev_err_probe(dev, ret, "failed to allocate vectors\n");
+
+	if (ret != HBG_VECTOR_NUM)
+		return dev_err_probe(dev, -EINVAL,
+				     "requested %u MSI, but allocated %d MSI\n",
+				     HBG_VECTOR_NUM, ret);
+
+	/* mdio irq not requested, so the number of requested interrupts
+	 * is HBG_VECTOR_NUM - 1.
+	 */
+	for (i = 0; i < HBG_VECTOR_NUM - 1; i++) {
+		id = pci_irq_vector(priv->pdev, i);
+		if (id < 0)
+			return dev_err_probe(dev, id, "failed to get irq id\n");
+
+		snprintf(vectors->name[i], sizeof(vectors->name[i]), "%s-%s-%s",
+			 dev_driver_string(dev), pci_name(priv->pdev),
+			 irq_names_map[i]);
+
+		ret = devm_request_irq(dev, id, hbg_irq_handle, 0,
+				       vectors->name[i], priv);
+		if (ret)
+			return dev_err_probe(dev, ret,
+					     "failed to request irq: %s\n",
+					     irq_names_map[i]);
+	}
+
+	vectors->stats_array = devm_kcalloc(&priv->pdev->dev,
+					    ARRAY_SIZE(hbg_irqs),
+					    sizeof(u64), GFP_KERNEL);
+	if (!vectors->stats_array)
+		return -ENOMEM;
+
+	vectors->info_array = hbg_irqs;
+	vectors->info_array_len = ARRAY_SIZE(hbg_irqs);
+	return 0;
+}
diff --git a/drivers/net/ethernet/hisilicon/hibmcge/hbg_irq.h b/drivers/net/ethernet/hisilicon/hibmcge/hbg_irq.h
new file mode 100644
index 000000000000..5c5323cfc751
--- /dev/null
+++ b/drivers/net/ethernet/hisilicon/hibmcge/hbg_irq.h
@@ -0,0 +1,11 @@
+/* SPDX-License-Identifier: GPL-2.0+ */
+/* Copyright (c) 2024 Hisilicon Limited. */
+
+#ifndef __HBG_IRQ_H
+#define __HBG_IRQ_H
+
+#include "hbg_common.h"
+
+int hbg_irq_init(struct hbg_priv *priv);
+
+#endif
diff --git a/drivers/net/ethernet/hisilicon/hibmcge/hbg_main.c b/drivers/net/ethernet/hisilicon/hibmcge/hbg_main.c
new file mode 100644
index 000000000000..0b92a2e5e986
--- /dev/null
+++ b/drivers/net/ethernet/hisilicon/hibmcge/hbg_main.c
@@ -0,0 +1,511 @@
+// SPDX-License-Identifier: GPL-2.0+
+// Copyright (c) 2024 Hisilicon Limited.
+
+#include <linux/etherdevice.h>
+#include <linux/if_vlan.h>
+#include <linux/netdevice.h>
+#include <linux/pci.h>
+#include <linux/phy.h>
+#include "hbg_common.h"
+#include "hbg_diagnose.h"
+#include "hbg_err.h"
+#include "hbg_ethtool.h"
+#include "hbg_hw.h"
+#include "hbg_irq.h"
+#include "hbg_mdio.h"
+#include "hbg_txrx.h"
+#include "hbg_debugfs.h"
+
+#define HBG_SUPPORT_FEATURES (NETIF_F_IP_CSUM | NETIF_F_IPV6_CSUM | \
+			     NETIF_F_RXCSUM)
+
+static void hbg_all_irq_enable(struct hbg_priv *priv, bool enabled)
+{
+	const struct hbg_irq_info *info;
+	u32 i;
+
+	for (i = 0; i < priv->vectors.info_array_len; i++) {
+		info = &priv->vectors.info_array[i];
+		hbg_hw_irq_enable(priv, info->mask, enabled);
+	}
+}
+
+static int hbg_net_open(struct net_device *netdev)
+{
+	struct hbg_priv *priv = netdev_priv(netdev);
+	int ret;
+
+	ret = hbg_txrx_init(priv);
+	if (ret)
+		return ret;
+
+	hbg_all_irq_enable(priv, true);
+	hbg_hw_mac_enable(priv, HBG_STATUS_ENABLE);
+	netif_start_queue(netdev);
+	hbg_phy_start(priv);
+
+	return 0;
+}
+
+/* This function only can be called after hbg_txrx_uninit() */
+static int hbg_hw_txrx_clear(struct hbg_priv *priv)
+{
+	int ret;
+
+	/* After ring buffers have been released,
+	 * do a reset to release hw fifo rx ring buffer
+	 */
+	ret = hbg_hw_event_notify(priv, HBG_HW_EVENT_RESET);
+	if (ret)
+		return ret;
+
+	/* After reset, regs need to be reconfigured */
+	return hbg_rebuild(priv);
+}
+
+static int hbg_net_stop(struct net_device *netdev)
+{
+	struct hbg_priv *priv = netdev_priv(netdev);
+
+	hbg_phy_stop(priv);
+	netif_stop_queue(netdev);
+	hbg_hw_mac_enable(priv, HBG_STATUS_DISABLE);
+	hbg_all_irq_enable(priv, false);
+	hbg_txrx_uninit(priv);
+	return hbg_hw_txrx_clear(priv);
+}
+
+static void hbg_update_promisc_mode(struct net_device *netdev, bool overflow)
+{
+	struct hbg_priv *priv = netdev_priv(netdev);
+
+	/* Only when not table_overflow, and netdev->flags not set IFF_PROMISC,
+	 * The MAC filter will be enabled.
+	 * Otherwise the filter will be disabled.
+	 */
+	priv->filter.enabled = !(overflow || (netdev->flags & IFF_PROMISC));
+	hbg_hw_set_mac_filter_enable(priv, priv->filter.enabled);
+}
+
+static void hbg_set_mac_to_mac_table(struct hbg_priv *priv,
+				     u32 index, const u8 *addr)
+{
+	if (addr) {
+		ether_addr_copy(priv->filter.mac_table[index].addr, addr);
+		hbg_hw_set_uc_addr(priv, ether_addr_to_u64(addr), index);
+	} else {
+		eth_zero_addr(priv->filter.mac_table[index].addr);
+		hbg_hw_set_uc_addr(priv, 0, index);
+	}
+}
+
+static int hbg_get_index_from_mac_table(struct hbg_priv *priv,
+					const u8 *addr, u32 *index)
+{
+	u32 i;
+
+	for (i = 0; i < priv->filter.table_max_len; i++)
+		if (ether_addr_equal(priv->filter.mac_table[i].addr, addr)) {
+			*index = i;
+			return 0;
+		}
+
+	return -EINVAL;
+}
+
+static int hbg_add_mac_to_filter(struct hbg_priv *priv, const u8 *addr)
+{
+	u32 index;
+
+	/* already exists */
+	if (!hbg_get_index_from_mac_table(priv, addr, &index))
+		return 0;
+
+	for (index = 0; index < priv->filter.table_max_len; index++)
+		if (is_zero_ether_addr(priv->filter.mac_table[index].addr)) {
+			hbg_set_mac_to_mac_table(priv, index, addr);
+			return 0;
+		}
+
+	return -ENOSPC;
+}
+
+static void hbg_del_mac_from_filter(struct hbg_priv *priv, const u8 *addr)
+{
+	u32 index;
+
+	/* not exists */
+	if (hbg_get_index_from_mac_table(priv, addr, &index))
+		return;
+
+	hbg_set_mac_to_mac_table(priv, index, NULL);
+}
+
+static int hbg_uc_sync(struct net_device *netdev, const unsigned char *addr)
+{
+	struct hbg_priv *priv = netdev_priv(netdev);
+
+	return hbg_add_mac_to_filter(priv, addr);
+}
+
+static int hbg_uc_unsync(struct net_device *netdev, const unsigned char *addr)
+{
+	struct hbg_priv *priv = netdev_priv(netdev);
+
+	if (ether_addr_equal(netdev->dev_addr, (u8 *)addr))
+		return 0;
+
+	hbg_del_mac_from_filter(priv, addr);
+	return 0;
+}
+
+static void hbg_net_set_rx_mode(struct net_device *netdev)
+{
+	int ret;
+
+	ret = __dev_uc_sync(netdev, hbg_uc_sync, hbg_uc_unsync);
+
+	/* If ret != 0, overflow has occurred */
+	hbg_update_promisc_mode(netdev, !!ret);
+}
+
+static int hbg_net_set_mac_address(struct net_device *netdev, void *addr)
+{
+	struct hbg_priv *priv = netdev_priv(netdev);
+	u8 *mac_addr;
+	bool exists;
+	u32 index;
+
+	mac_addr = ((struct sockaddr *)addr)->sa_data;
+
+	if (!is_valid_ether_addr(mac_addr))
+		return -EADDRNOTAVAIL;
+
+	/* The index of host mac is always 0.
+	 * If new mac address already exists,
+	 * delete the existing mac address and
+	 * add it to the position with index 0.
+	 */
+	exists = !hbg_get_index_from_mac_table(priv, mac_addr, &index);
+	hbg_set_mac_to_mac_table(priv, 0, mac_addr);
+	if (exists)
+		hbg_set_mac_to_mac_table(priv, index, NULL);
+
+	hbg_hw_set_rx_pause_mac_addr(priv, ether_addr_to_u64(mac_addr));
+	dev_addr_set(netdev, mac_addr);
+	return 0;
+}
+
+static int hbg_net_change_mtu(struct net_device *netdev, int new_mtu)
+{
+	struct hbg_priv *priv = netdev_priv(netdev);
+
+	if (netif_running(netdev))
+		return -EBUSY;
+
+	dev_dbg(&priv->pdev->dev,
+		"change mtu from %u to %u\n", netdev->mtu, new_mtu);
+
+	hbg_hw_set_mtu(priv, new_mtu);
+	WRITE_ONCE(netdev->mtu, new_mtu);
+
+	return 0;
+}
+
+static void hbg_net_tx_timeout(struct net_device *netdev, unsigned int txqueue)
+{
+	struct hbg_priv *priv = netdev_priv(netdev);
+	struct hbg_ring *ring = &priv->tx_ring;
+	char *buf = ring->tout_log_buf;
+	u32 pos = 0;
+
+	priv->stats.tx_timeout_cnt++;
+
+	pos += scnprintf(buf + pos, HBG_TX_TIMEOUT_BUF_LEN - pos,
+			 "tx_timeout cnt: %llu\n", priv->stats.tx_timeout_cnt);
+	pos += scnprintf(buf + pos, HBG_TX_TIMEOUT_BUF_LEN - pos,
+			 "ring used num: %u, fifo used num: %u\n",
+			 hbg_get_queue_used_num(ring),
+			 hbg_hw_get_fifo_used_num(priv, HBG_DIR_TX));
+	pos += scnprintf(buf + pos, HBG_TX_TIMEOUT_BUF_LEN - pos,
+			 "ntc: %u, ntu: %u, irq enabled: %u\n",
+			 ring->ntc, ring->ntu,
+			 hbg_hw_irq_is_enabled(priv, HBG_INT_MSK_TX_B));
+
+	netdev_info(netdev, "%s", buf);
+}
+
+static void hbg_net_get_stats(struct net_device *netdev,
+			      struct rtnl_link_stats64 *stats)
+{
+	struct hbg_priv *priv = netdev_priv(netdev);
+	struct hbg_stats *h_stats = &priv->stats;
+
+	hbg_update_stats(priv);
+	dev_get_tstats64(netdev, stats);
+
+	/* fifo empty */
+	stats->tx_fifo_errors += h_stats->tx_drop_cnt;
+
+	stats->tx_dropped += h_stats->tx_excessive_length_drop_cnt +
+			     h_stats->tx_drop_cnt;
+	stats->tx_errors += h_stats->tx_add_cs_fail_cnt +
+			    h_stats->tx_bufrl_err_cnt +
+			    h_stats->tx_underrun_err_cnt +
+			    h_stats->tx_crc_err_cnt;
+	stats->rx_errors += h_stats->rx_data_error_cnt;
+	stats->multicast += h_stats->rx_mc_pkt_cnt;
+	stats->rx_dropped += h_stats->rx_desc_drop;
+	stats->rx_length_errors += h_stats->rx_frame_very_long_err_cnt +
+				   h_stats->rx_frame_long_err_cnt +
+				   h_stats->rx_frame_runt_err_cnt +
+				   h_stats->rx_frame_short_err_cnt +
+				   h_stats->rx_lengthfield_err_cnt;
+	stats->rx_frame_errors += h_stats->rx_desc_l2_err_cnt +
+				  h_stats->rx_desc_l3l4_err_cnt;
+	stats->rx_fifo_errors += h_stats->rx_overflow_cnt +
+				 h_stats->rx_overrun_cnt;
+	stats->rx_crc_errors += h_stats->rx_fcs_error_cnt;
+}
+
+static const struct net_device_ops hbg_netdev_ops = {
+	.ndo_open		= hbg_net_open,
+	.ndo_stop		= hbg_net_stop,
+	.ndo_start_xmit		= hbg_net_start_xmit,
+	.ndo_validate_addr	= eth_validate_addr,
+	.ndo_set_mac_address	= hbg_net_set_mac_address,
+	.ndo_change_mtu		= hbg_net_change_mtu,
+	.ndo_tx_timeout		= hbg_net_tx_timeout,
+	.ndo_set_rx_mode	= hbg_net_set_rx_mode,
+	.ndo_get_stats64	= hbg_net_get_stats,
+	.ndo_eth_ioctl		= phy_do_ioctl_running,
+};
+
+static void hbg_service_task(struct work_struct *work)
+{
+	struct hbg_priv *priv = container_of(work, struct hbg_priv,
+					     service_task.work);
+
+	if (test_and_clear_bit(HBG_NIC_STATE_NEED_RESET, &priv->state))
+		hbg_err_reset(priv);
+
+	if (test_and_clear_bit(HBG_NIC_STATE_NP_LINK_FAIL, &priv->state))
+		hbg_fix_np_link_fail(priv);
+
+	hbg_diagnose_message_push(priv);
+
+	/* The type of statistics register is u32,
+	 * To prevent the statistics register from overflowing,
+	 * the driver dumps the statistics every 30 seconds.
+	 */
+	if (time_after(jiffies, priv->last_update_stats_time + 30 * HZ)) {
+		hbg_update_stats(priv);
+		priv->last_update_stats_time = jiffies;
+	}
+
+	schedule_delayed_work(&priv->service_task,
+			      msecs_to_jiffies(MSEC_PER_SEC));
+}
+
+void hbg_err_reset_task_schedule(struct hbg_priv *priv)
+{
+	set_bit(HBG_NIC_STATE_NEED_RESET, &priv->state);
+	schedule_delayed_work(&priv->service_task, 0);
+}
+
+void hbg_np_link_fail_task_schedule(struct hbg_priv *priv)
+{
+	set_bit(HBG_NIC_STATE_NP_LINK_FAIL, &priv->state);
+	schedule_delayed_work(&priv->service_task, 0);
+}
+
+static void hbg_cancel_delayed_work_sync(void *data)
+{
+	cancel_delayed_work_sync(data);
+}
+
+static int hbg_delaywork_init(struct hbg_priv *priv)
+{
+	INIT_DELAYED_WORK(&priv->service_task, hbg_service_task);
+	schedule_delayed_work(&priv->service_task, 0);
+	return devm_add_action_or_reset(&priv->pdev->dev,
+					hbg_cancel_delayed_work_sync,
+					&priv->service_task);
+}
+
+static int hbg_mac_filter_init(struct hbg_priv *priv)
+{
+	struct hbg_dev_specs *dev_specs = &priv->dev_specs;
+	struct hbg_mac_filter *filter = &priv->filter;
+	struct hbg_mac_table_entry *tmp_table;
+
+	tmp_table = devm_kcalloc(&priv->pdev->dev, dev_specs->uc_mac_num,
+				 sizeof(*tmp_table), GFP_KERNEL);
+	if (!tmp_table)
+		return -ENOMEM;
+
+	filter->mac_table = tmp_table;
+	filter->table_max_len = dev_specs->uc_mac_num;
+	filter->enabled = true;
+
+	hbg_hw_set_mac_filter_enable(priv, filter->enabled);
+	return 0;
+}
+
+static void hbg_init_user_def(struct hbg_priv *priv)
+{
+	struct ethtool_pauseparam *pause_param = &priv->user_def.pause_param;
+
+	priv->mac.pause_autoneg = HBG_STATUS_ENABLE;
+
+	pause_param->autoneg = priv->mac.pause_autoneg;
+	hbg_hw_get_pause_enable(priv, &pause_param->tx_pause,
+				&pause_param->rx_pause);
+}
+
+static int hbg_init(struct hbg_priv *priv)
+{
+	int ret;
+
+	ret = hbg_hw_event_notify(priv, HBG_HW_EVENT_INIT);
+	if (ret)
+		return ret;
+
+	ret = hbg_hw_init(priv);
+	if (ret)
+		return ret;
+
+	ret = hbg_irq_init(priv);
+	if (ret)
+		return ret;
+
+	ret = hbg_mdio_init(priv);
+	if (ret)
+		return ret;
+
+	ret = hbg_mac_filter_init(priv);
+	if (ret)
+		return ret;
+
+	ret = hbg_delaywork_init(priv);
+	if (ret)
+		return ret;
+
+	hbg_debugfs_init(priv);
+	hbg_init_user_def(priv);
+	return 0;
+}
+
+static int hbg_pci_init(struct pci_dev *pdev)
+{
+	struct net_device *netdev = pci_get_drvdata(pdev);
+	struct hbg_priv *priv = netdev_priv(netdev);
+	struct device *dev = &pdev->dev;
+	int ret;
+
+	ret = pcim_enable_device(pdev);
+	if (ret)
+		return dev_err_probe(dev, ret, "failed to enable PCI device\n");
+
+	ret = dma_set_mask_and_coherent(dev, DMA_BIT_MASK(32));
+	if (ret)
+		return dev_err_probe(dev, ret, "failed to set PCI DMA mask\n");
+
+	ret = pcim_iomap_regions(pdev, BIT(0), dev_driver_string(dev));
+	if (ret)
+		return dev_err_probe(dev, ret, "failed to map PCI bar space\n");
+
+	priv->io_base = pcim_iomap_table(pdev)[0];
+	if (!priv->io_base)
+		return -ENOMEM;
+
+	pci_set_master(pdev);
+	return 0;
+}
+
+static int hbg_probe(struct pci_dev *pdev, const struct pci_device_id *ent)
+{
+	struct device *dev = &pdev->dev;
+	struct net_device *netdev;
+	struct hbg_priv *priv;
+	int ret;
+
+	netdev = devm_alloc_etherdev(dev, sizeof(struct hbg_priv));
+	if (!netdev)
+		return -ENOMEM;
+
+	pci_set_drvdata(pdev, netdev);
+	SET_NETDEV_DEV(netdev, dev);
+
+	priv = netdev_priv(netdev);
+	priv->netdev = netdev;
+	priv->pdev = pdev;
+
+	ret = hbg_pci_init(pdev);
+	if (ret)
+		return ret;
+
+	ret = hbg_init(priv);
+	if (ret)
+		return ret;
+
+	/* set default features */
+	netdev->features |= HBG_SUPPORT_FEATURES;
+	netdev->hw_features |= HBG_SUPPORT_FEATURES;
+	netdev->priv_flags |= IFF_UNICAST_FLT;
+
+	netdev->pcpu_stat_type = NETDEV_PCPU_STAT_TSTATS;
+	netdev->max_mtu = priv->dev_specs.max_mtu;
+	netdev->min_mtu = priv->dev_specs.min_mtu;
+	netdev->netdev_ops = &hbg_netdev_ops;
+	netdev->watchdog_timeo = 5 * HZ;
+
+	hbg_hw_set_mtu(priv, ETH_DATA_LEN);
+	hbg_net_set_mac_address(priv->netdev, &priv->dev_specs.mac_addr);
+	hbg_ethtool_set_ops(netdev);
+
+	ret = devm_register_netdev(dev, netdev);
+	if (ret)
+		return dev_err_probe(dev, ret, "failed to register netdev\n");
+
+	netif_carrier_off(netdev);
+	return 0;
+}
+
+static const struct pci_device_id hbg_pci_tbl[] = {
+	{PCI_VDEVICE(HUAWEI, 0x3730), 0},
+	{ }
+};
+MODULE_DEVICE_TABLE(pci, hbg_pci_tbl);
+
+static struct pci_driver hbg_driver = {
+	.name		= "hibmcge",
+	.id_table	= hbg_pci_tbl,
+	.probe		= hbg_probe,
+};
+
+static int __init hbg_module_init(void)
+{
+	int ret;
+
+	hbg_debugfs_register();
+	hbg_set_pci_err_handler(&hbg_driver);
+	ret = pci_register_driver(&hbg_driver);
+	if (ret)
+		hbg_debugfs_unregister();
+
+	return ret;
+}
+module_init(hbg_module_init);
+
+static void __exit hbg_module_exit(void)
+{
+	pci_unregister_driver(&hbg_driver);
+	hbg_debugfs_unregister();
+}
+module_exit(hbg_module_exit);
+
+MODULE_LICENSE("GPL");
+MODULE_AUTHOR("Huawei Tech. Co., Ltd.");
+MODULE_DESCRIPTION("hibmcge driver");
+MODULE_VERSION("1.0");
diff --git a/drivers/net/ethernet/hisilicon/hibmcge/hbg_mdio.c b/drivers/net/ethernet/hisilicon/hibmcge/hbg_mdio.c
new file mode 100644
index 000000000000..37791de47f6f
--- /dev/null
+++ b/drivers/net/ethernet/hisilicon/hibmcge/hbg_mdio.c
@@ -0,0 +1,305 @@
+// SPDX-License-Identifier: GPL-2.0+
+// Copyright (c) 2024 Hisilicon Limited.
+
+#include <linux/phy.h>
+#include <linux/phy_fixed.h>
+#include <linux/rtnetlink.h>
+#include "hbg_common.h"
+#include "hbg_hw.h"
+#include "hbg_mdio.h"
+#include "hbg_reg.h"
+
+#define HBG_MAC_GET_PRIV(mac) ((struct hbg_priv *)(mac)->mdio_bus->priv)
+#define HBG_MII_BUS_GET_MAC(bus) (&((struct hbg_priv *)(bus)->priv)->mac)
+
+#define HBG_MDIO_C22_MODE		0x1
+#define HBG_MDIO_C22_REG_WRITE		0x1
+#define HBG_MDIO_C22_REG_READ		0x2
+
+#define HBG_MDIO_OP_TIMEOUT_US		(1 * 1000 * 1000)
+#define HBG_MDIO_OP_INTERVAL_US		(5 * 1000)
+
+#define HBG_NP_LINK_FAIL_RETRY_TIMES	5
+#define HBG_NO_PHY			0xFF
+
+static void hbg_mdio_set_command(struct hbg_mac *mac, u32 cmd)
+{
+	hbg_reg_write(HBG_MAC_GET_PRIV(mac), HBG_REG_MDIO_COMMAND_ADDR, cmd);
+}
+
+static void hbg_mdio_get_command(struct hbg_mac *mac, u32 *cmd)
+{
+	*cmd = hbg_reg_read(HBG_MAC_GET_PRIV(mac), HBG_REG_MDIO_COMMAND_ADDR);
+}
+
+static void hbg_mdio_set_wdata_reg(struct hbg_mac *mac, u16 wdata_value)
+{
+	hbg_reg_write_field(HBG_MAC_GET_PRIV(mac), HBG_REG_MDIO_WDATA_ADDR,
+			    HBG_REG_MDIO_WDATA_M, wdata_value);
+}
+
+static u32 hbg_mdio_get_rdata_reg(struct hbg_mac *mac)
+{
+	return hbg_reg_read_field(HBG_MAC_GET_PRIV(mac),
+				  HBG_REG_MDIO_RDATA_ADDR,
+				  HBG_REG_MDIO_WDATA_M);
+}
+
+static int hbg_mdio_wait_ready(struct hbg_mac *mac)
+{
+	struct hbg_priv *priv = HBG_MAC_GET_PRIV(mac);
+	u32 cmd = 0;
+	int ret;
+
+	ret = readl_poll_timeout(priv->io_base + HBG_REG_MDIO_COMMAND_ADDR, cmd,
+				 !FIELD_GET(HBG_REG_MDIO_COMMAND_START_B, cmd),
+				 HBG_MDIO_OP_INTERVAL_US,
+				 HBG_MDIO_OP_TIMEOUT_US);
+
+	return ret ? -ETIMEDOUT : 0;
+}
+
+static int hbg_mdio_cmd_send(struct hbg_mac *mac, u32 prt_addr, u32 dev_addr,
+			     u32 type, u32 op_code)
+{
+	u32 cmd = 0;
+
+	hbg_mdio_get_command(mac, &cmd);
+	hbg_field_modify(cmd, HBG_REG_MDIO_COMMAND_ST_M, type);
+	hbg_field_modify(cmd, HBG_REG_MDIO_COMMAND_OP_M, op_code);
+	hbg_field_modify(cmd, HBG_REG_MDIO_COMMAND_PRTAD_M, prt_addr);
+	hbg_field_modify(cmd, HBG_REG_MDIO_COMMAND_DEVAD_M, dev_addr);
+
+	/* if auto scan enabled, this value need fix to 0 */
+	hbg_field_modify(cmd, HBG_REG_MDIO_COMMAND_START_B, 0x1);
+
+	hbg_mdio_set_command(mac, cmd);
+
+	/* wait operation complete and check the result */
+	return hbg_mdio_wait_ready(mac);
+}
+
+static int hbg_mdio_read22(struct mii_bus *bus, int phy_addr, int regnum)
+{
+	struct hbg_mac *mac = HBG_MII_BUS_GET_MAC(bus);
+	int ret;
+
+	ret = hbg_mdio_cmd_send(mac, phy_addr, regnum, HBG_MDIO_C22_MODE,
+				HBG_MDIO_C22_REG_READ);
+	if (ret)
+		return ret;
+
+	return hbg_mdio_get_rdata_reg(mac);
+}
+
+static int hbg_mdio_write22(struct mii_bus *bus, int phy_addr, int regnum,
+			    u16 val)
+{
+	struct hbg_mac *mac = HBG_MII_BUS_GET_MAC(bus);
+
+	hbg_mdio_set_wdata_reg(mac, val);
+	return hbg_mdio_cmd_send(mac, phy_addr, regnum, HBG_MDIO_C22_MODE,
+				 HBG_MDIO_C22_REG_WRITE);
+}
+
+static void hbg_mdio_init_hw(struct hbg_priv *priv)
+{
+	u32 freq = priv->dev_specs.mdio_frequency;
+	struct hbg_mac *mac = &priv->mac;
+	u32 cmd = 0;
+
+	cmd |= FIELD_PREP(HBG_REG_MDIO_COMMAND_ST_M, HBG_MDIO_C22_MODE);
+	cmd |= FIELD_PREP(HBG_REG_MDIO_COMMAND_AUTO_SCAN_B, HBG_STATUS_DISABLE);
+
+	/* freq use two bits, which are stored in clk_sel and clk_sel_exp */
+	cmd |= FIELD_PREP(HBG_REG_MDIO_COMMAND_CLK_SEL_B, freq & 0x1);
+	cmd |= FIELD_PREP(HBG_REG_MDIO_COMMAND_CLK_SEL_EXP_B,
+			  (freq >> 1) & 0x1);
+
+	hbg_mdio_set_command(mac, cmd);
+}
+
+static void hbg_flowctrl_cfg(struct hbg_priv *priv)
+{
+	struct phy_device *phydev = priv->mac.phydev;
+	bool rx_pause;
+	bool tx_pause;
+
+	if (!priv->mac.pause_autoneg)
+		return;
+
+	phy_get_pause(phydev, &tx_pause, &rx_pause);
+	hbg_hw_set_pause_enable(priv, tx_pause, rx_pause);
+}
+
+void hbg_fix_np_link_fail(struct hbg_priv *priv)
+{
+	struct device *dev = &priv->pdev->dev;
+
+	rtnl_lock();
+
+	if (priv->stats.np_link_fail_cnt >= HBG_NP_LINK_FAIL_RETRY_TIMES) {
+		dev_err(dev, "failed to fix the MAC link status\n");
+		priv->stats.np_link_fail_cnt = 0;
+		goto unlock;
+	}
+
+	if (!priv->mac.phydev->link)
+		goto unlock;
+
+	priv->stats.np_link_fail_cnt++;
+	dev_err(dev, "failed to link between MAC and PHY, try to fix...\n");
+
+	/* Replace phy_reset() with phy_stop() and phy_start(),
+	 * as suggested by Andrew.
+	 */
+	hbg_phy_stop(priv);
+	hbg_phy_start(priv);
+
+unlock:
+	rtnl_unlock();
+}
+
+static void hbg_phy_adjust_link(struct net_device *netdev)
+{
+	struct hbg_priv *priv = netdev_priv(netdev);
+	struct phy_device *phydev = netdev->phydev;
+	u32 speed;
+
+	if (phydev->link != priv->mac.link_status) {
+		if (phydev->link) {
+			switch (phydev->speed) {
+			case SPEED_10:
+				speed = HBG_PORT_MODE_SGMII_10M;
+				break;
+			case SPEED_100:
+				speed = HBG_PORT_MODE_SGMII_100M;
+				break;
+			case SPEED_1000:
+				speed = HBG_PORT_MODE_SGMII_1000M;
+				break;
+			default:
+				return;
+			}
+
+			priv->mac.speed = speed;
+			priv->mac.duplex = phydev->duplex;
+			priv->mac.autoneg = phydev->autoneg;
+			hbg_hw_adjust_link(priv, speed, phydev->duplex);
+			hbg_flowctrl_cfg(priv);
+		}
+
+		priv->mac.link_status = phydev->link;
+		phy_print_status(phydev);
+	}
+}
+
+static void hbg_phy_disconnect(void *data)
+{
+	phy_disconnect((struct phy_device *)data);
+}
+
+static int hbg_phy_connect(struct hbg_priv *priv)
+{
+	struct phy_device *phydev = priv->mac.phydev;
+	struct device *dev = &priv->pdev->dev;
+	int ret;
+
+	ret = phy_connect_direct(priv->netdev, phydev, hbg_phy_adjust_link,
+				 PHY_INTERFACE_MODE_SGMII);
+	if (ret)
+		return dev_err_probe(dev, ret, "failed to connect phy\n");
+
+	ret = devm_add_action_or_reset(dev, hbg_phy_disconnect, phydev);
+	if (ret)
+		return ret;
+
+	phy_remove_link_mode(phydev, ETHTOOL_LINK_MODE_1000baseT_Half_BIT);
+	phy_support_asym_pause(phydev);
+	phy_attached_info(phydev);
+
+	return 0;
+}
+
+void hbg_phy_start(struct hbg_priv *priv)
+{
+	phy_start(priv->mac.phydev);
+}
+
+void hbg_phy_stop(struct hbg_priv *priv)
+{
+	phy_stop(priv->mac.phydev);
+}
+
+static void hbg_fixed_phy_uninit(void *data)
+{
+	fixed_phy_unregister((struct phy_device *)data);
+}
+
+static int hbg_fixed_phy_init(struct hbg_priv *priv)
+{
+	struct fixed_phy_status hbg_fixed_phy_status = {
+		.link = 1,
+		.speed = SPEED_1000,
+		.duplex = DUPLEX_FULL,
+		.pause = 1,
+		.asym_pause = 1,
+	};
+	struct device *dev = &priv->pdev->dev;
+	struct phy_device *phydev;
+	int ret;
+
+	phydev = fixed_phy_register(&hbg_fixed_phy_status, NULL);
+	if (IS_ERR(phydev)) {
+		dev_err_probe(dev, PTR_ERR(phydev),
+			      "failed to register fixed PHY device\n");
+		return PTR_ERR(phydev);
+	}
+
+	ret = devm_add_action_or_reset(dev, hbg_fixed_phy_uninit, phydev);
+	if (ret)
+		return ret;
+
+	priv->mac.phydev = phydev;
+	return hbg_phy_connect(priv);
+}
+
+int hbg_mdio_init(struct hbg_priv *priv)
+{
+	struct device *dev = &priv->pdev->dev;
+	struct hbg_mac *mac = &priv->mac;
+	struct phy_device *phydev;
+	struct mii_bus *mdio_bus;
+	int ret;
+
+	mac->phy_addr = priv->dev_specs.phy_addr;
+	if (mac->phy_addr == HBG_NO_PHY)
+		return hbg_fixed_phy_init(priv);
+
+	mdio_bus = devm_mdiobus_alloc(dev);
+	if (!mdio_bus)
+		return -ENOMEM;
+
+	mdio_bus->parent = dev;
+	mdio_bus->priv = priv;
+	mdio_bus->phy_mask = ~(1 << mac->phy_addr);
+	mdio_bus->name = "hibmcge mii bus";
+	mac->mdio_bus = mdio_bus;
+
+	mdio_bus->read = hbg_mdio_read22;
+	mdio_bus->write = hbg_mdio_write22;
+	snprintf(mdio_bus->id, MII_BUS_ID_SIZE, "%s-%s", "mii", dev_name(dev));
+
+	ret = devm_mdiobus_register(dev, mdio_bus);
+	if (ret)
+		return dev_err_probe(dev, ret, "failed to register MDIO bus\n");
+
+	phydev = mdiobus_get_phy(mdio_bus, mac->phy_addr);
+	if (!phydev)
+		return dev_err_probe(dev, -ENODEV,
+				     "failed to get phy device\n");
+
+	mac->phydev = phydev;
+	hbg_mdio_init_hw(priv);
+	return hbg_phy_connect(priv);
+}
diff --git a/drivers/net/ethernet/hisilicon/hibmcge/hbg_mdio.h b/drivers/net/ethernet/hisilicon/hibmcge/hbg_mdio.h
new file mode 100644
index 000000000000..f3771c1bbd34
--- /dev/null
+++ b/drivers/net/ethernet/hisilicon/hibmcge/hbg_mdio.h
@@ -0,0 +1,14 @@
+/* SPDX-License-Identifier: GPL-2.0+ */
+/* Copyright (c) 2024 Hisilicon Limited. */
+
+#ifndef __HBG_MDIO_H
+#define __HBG_MDIO_H
+
+#include "hbg_common.h"
+
+int hbg_mdio_init(struct hbg_priv *priv);
+void hbg_phy_start(struct hbg_priv *priv);
+void hbg_phy_stop(struct hbg_priv *priv);
+void hbg_fix_np_link_fail(struct hbg_priv *priv);
+
+#endif
diff --git a/drivers/net/ethernet/hisilicon/hibmcge/hbg_reg.h b/drivers/net/ethernet/hisilicon/hibmcge/hbg_reg.h
new file mode 100644
index 000000000000..a39d1e796e4a
--- /dev/null
+++ b/drivers/net/ethernet/hisilicon/hibmcge/hbg_reg.h
@@ -0,0 +1,298 @@
+/* SPDX-License-Identifier: GPL-2.0+ */
+/* Copyright (c) 2024 Hisilicon Limited. */
+
+#ifndef __HBG_REG_H
+#define __HBG_REG_H
+
+/* DEV SPEC */
+#define HBG_REG_SPEC_VALID_ADDR			0x0000
+#define HBG_REG_EVENT_REQ_ADDR			0x0004
+#define HBG_REG_MAC_ID_ADDR			0x0008
+#define HBG_REG_PHY_ID_ADDR			0x000C
+#define HBG_REG_MAC_ADDR_ADDR			0x0010
+#define HBG_REG_MAC_ADDR_HIGH_ADDR		0x0014
+#define HBG_REG_UC_MAC_NUM_ADDR			0x0018
+#define HBG_REG_MDIO_FREQ_ADDR			0x0024
+#define HBG_REG_MAX_MTU_ADDR			0x0028
+#define HBG_REG_MIN_MTU_ADDR			0x002C
+#define HBG_REG_TX_FIFO_NUM_ADDR		0x0030
+#define HBG_REG_RX_FIFO_NUM_ADDR		0x0034
+#define HBG_REG_VLAN_LAYERS_ADDR		0x0038
+#define HBG_REG_PUSH_REQ_ADDR			0x00F0
+#define HBG_REG_MSG_HEADER_ADDR			0x00F4
+#define HBG_REG_MSG_HEADER_OPCODE_M		GENMASK(7, 0)
+#define HBG_REG_MSG_HEADER_STATUS_M		GENMASK(11, 8)
+#define HBG_REG_MSG_HEADER_DATA_NUM_M		GENMASK(19, 12)
+#define HBG_REG_MSG_HEADER_RESP_CODE_M		GENMASK(27, 20)
+#define HBG_REG_MSG_DATA_BASE_ADDR		0x0100
+
+/* MDIO */
+#define HBG_REG_MDIO_BASE			0x8000
+#define HBG_REG_MDIO_COMMAND_ADDR		(HBG_REG_MDIO_BASE + 0x0000)
+#define HBG_REG_MDIO_COMMAND_CLK_SEL_EXP_B	BIT(17)
+#define HBG_REG_MDIO_COMMAND_AUTO_SCAN_B	BIT(16)
+#define HBG_REG_MDIO_COMMAND_CLK_SEL_B		BIT(15)
+#define HBG_REG_MDIO_COMMAND_START_B		BIT(14)
+#define HBG_REG_MDIO_COMMAND_ST_M		GENMASK(13, 12)
+#define HBG_REG_MDIO_COMMAND_OP_M		GENMASK(11, 10)
+#define HBG_REG_MDIO_COMMAND_PRTAD_M		GENMASK(9, 5)
+#define HBG_REG_MDIO_COMMAND_DEVAD_M		GENMASK(4, 0)
+#define HBG_REG_MDIO_ADDR_ADDR			(HBG_REG_MDIO_BASE + 0x0004)
+#define HBG_REG_MDIO_WDATA_ADDR			(HBG_REG_MDIO_BASE + 0x0008)
+#define HBG_REG_MDIO_WDATA_M			GENMASK(15, 0)
+#define HBG_REG_MDIO_RDATA_ADDR			(HBG_REG_MDIO_BASE + 0x000C)
+#define HBG_REG_MDIO_STA_ADDR			(HBG_REG_MDIO_BASE + 0x0010)
+
+/* GMAC */
+#define HBG_REG_SGMII_BASE			0x10000
+#define HBG_REG_DUPLEX_TYPE_ADDR		(HBG_REG_SGMII_BASE + 0x0008)
+#define HBG_REG_FD_FC_TYPE_ADDR			(HBG_REG_SGMII_BASE + 0x000C)
+#define HBG_REG_FC_TX_TIMER_ADDR		(HBG_REG_SGMII_BASE + 0x001C)
+#define HBG_REG_FD_FC_ADDR_LOW_ADDR		(HBG_REG_SGMII_BASE + 0x0020)
+#define HBG_REG_FD_FC_ADDR_HIGH_ADDR		(HBG_REG_SGMII_BASE + 0x0024)
+#define HBG_REG_DUPLEX_B			BIT(0)
+#define HBG_REG_MAX_FRAME_SIZE_ADDR		(HBG_REG_SGMII_BASE + 0x003C)
+#define HBG_REG_PORT_MODE_ADDR			(HBG_REG_SGMII_BASE + 0x0040)
+#define HBG_REG_PORT_MODE_M			GENMASK(3, 0)
+#define HBG_REG_PORT_ENABLE_ADDR		(HBG_REG_SGMII_BASE + 0x0044)
+#define HBG_REG_PORT_ENABLE_RX_B		BIT(1)
+#define HBG_REG_PORT_ENABLE_TX_B		BIT(2)
+#define HBG_REG_PAUSE_ENABLE_ADDR		(HBG_REG_SGMII_BASE + 0x0048)
+#define HBG_REG_PAUSE_ENABLE_RX_B		BIT(0)
+#define HBG_REG_PAUSE_ENABLE_TX_B		BIT(1)
+#define HBG_REG_AN_NEG_STATE_ADDR		(HBG_REG_SGMII_BASE + 0x0058)
+#define HBG_REG_AN_NEG_STATE_NP_LINK_OK_B	BIT(15)
+#define HBG_REG_TRANSMIT_CTRL_ADDR		(HBG_REG_SGMII_BASE + 0x0060)
+#define HBG_REG_TRANSMIT_CTRL_PAD_EN_B		BIT(7)
+#define HBG_REG_TRANSMIT_CTRL_CRC_ADD_B		BIT(6)
+#define HBG_REG_TRANSMIT_CTRL_AN_EN_B		BIT(5)
+#define HBG_REG_REC_FILT_CTRL_ADDR		(HBG_REG_SGMII_BASE + 0x0064)
+#define HBG_REG_REC_FILT_CTRL_UC_MATCH_EN_B	BIT(0)
+#define HBG_REG_REC_FILT_CTRL_PAUSE_FRM_PASS_B	BIT(4)
+#define HBG_REG_RX_OCTETS_TOTAL_OK_ADDR		(HBG_REG_SGMII_BASE + 0x0080)
+#define HBG_REG_RX_OCTETS_BAD_ADDR		(HBG_REG_SGMII_BASE + 0x0084)
+#define HBG_REG_RX_UC_PKTS_ADDR			(HBG_REG_SGMII_BASE + 0x0088)
+#define HBG_REG_RX_MC_PKTS_ADDR			(HBG_REG_SGMII_BASE + 0x008C)
+#define HBG_REG_RX_BC_PKTS_ADDR			(HBG_REG_SGMII_BASE + 0x0090)
+#define HBG_REG_RX_PKTS_64OCTETS_ADDR		(HBG_REG_SGMII_BASE + 0x0094)
+#define HBG_REG_RX_PKTS_65TO127OCTETS_ADDR	(HBG_REG_SGMII_BASE + 0x0098)
+#define HBG_REG_RX_PKTS_128TO255OCTETS_ADDR	(HBG_REG_SGMII_BASE + 0x009C)
+#define HBG_REG_RX_PKTS_256TO511OCTETS_ADDR	(HBG_REG_SGMII_BASE + 0x00A0)
+#define HBG_REG_RX_PKTS_512TO1023OCTETS_ADDR	(HBG_REG_SGMII_BASE + 0x00A4)
+#define HBG_REG_RX_PKTS_1024TO1518OCTETS_ADDR	(HBG_REG_SGMII_BASE + 0x00A8)
+#define HBG_REG_RX_PKTS_1519TOMAXOCTETS_ADDR	(HBG_REG_SGMII_BASE + 0x00AC)
+#define HBG_REG_RX_FCS_ERRORS_ADDR		(HBG_REG_SGMII_BASE + 0x00B0)
+#define HBG_REG_RX_TAGGED_ADDR			(HBG_REG_SGMII_BASE + 0x00B4)
+#define HBG_REG_RX_DATA_ERR_ADDR		(HBG_REG_SGMII_BASE + 0x00B8)
+#define HBG_REG_RX_ALIGN_ERRORS_ADDR		(HBG_REG_SGMII_BASE + 0x00BC)
+#define HBG_REG_RX_LONG_ERRORS_ADDR		(HBG_REG_SGMII_BASE + 0x00C0)
+#define HBG_REG_RX_JABBER_ERRORS_ADDR		(HBG_REG_SGMII_BASE + 0x00C4)
+#define HBG_REG_RX_PAUSE_MACCTL_FRAMCOUNTER_ADDR   (HBG_REG_SGMII_BASE + 0x00C8)
+#define HBG_REG_RX_UNKNOWN_MACCTL_FRAMCOUNTER_ADDR (HBG_REG_SGMII_BASE + 0x00CC)
+#define HBG_REG_RX_VERY_LONG_ERR_CNT_ADDR	(HBG_REG_SGMII_BASE + 0x00D0)
+#define HBG_REG_RX_RUNT_ERR_CNT_ADDR		(HBG_REG_SGMII_BASE + 0x00D4)
+#define HBG_REG_RX_SHORT_ERR_CNT_ADDR		(HBG_REG_SGMII_BASE + 0x00D8)
+#define HBG_REG_RX_FILT_PKT_CNT_ADDR		(HBG_REG_SGMII_BASE + 0x00E8)
+#define HBG_REG_RX_OCTETS_TOTAL_FILT_ADDR	(HBG_REG_SGMII_BASE + 0x00EC)
+#define HBG_REG_OCTETS_TRANSMITTED_OK_ADDR	(HBG_REG_SGMII_BASE + 0x0100)
+#define HBG_REG_OCTETS_TRANSMITTED_BAD_ADDR	(HBG_REG_SGMII_BASE + 0x0104)
+#define HBG_REG_TX_UC_PKTS_ADDR			(HBG_REG_SGMII_BASE + 0x0108)
+#define HBG_REG_TX_MC_PKTS_ADDR			(HBG_REG_SGMII_BASE + 0x010C)
+#define HBG_REG_TX_BC_PKTS_ADDR			(HBG_REG_SGMII_BASE + 0x0110)
+#define HBG_REG_TX_PKTS_64OCTETS_ADDR		(HBG_REG_SGMII_BASE + 0x0114)
+#define HBG_REG_TX_PKTS_65TO127OCTETS_ADDR	(HBG_REG_SGMII_BASE + 0x0118)
+#define HBG_REG_TX_PKTS_128TO255OCTETS_ADDR	(HBG_REG_SGMII_BASE + 0x011C)
+#define HBG_REG_TX_PKTS_256TO511OCTETS_ADDR	(HBG_REG_SGMII_BASE + 0x0120)
+#define HBG_REG_TX_PKTS_512TO1023OCTETS_ADDR	(HBG_REG_SGMII_BASE + 0x0124)
+#define HBG_REG_TX_PKTS_1024TO1518OCTETS_ADDR	(HBG_REG_SGMII_BASE + 0x0128)
+#define HBG_REG_TX_PKTS_1519TOMAXOCTETS_ADDR	(HBG_REG_SGMII_BASE + 0x012C)
+#define HBG_REG_TX_EXCESSIVE_LENGTH_DROP_ADDR	(HBG_REG_SGMII_BASE + 0x014C)
+#define HBG_REG_TX_UNDERRUN_ADDR		(HBG_REG_SGMII_BASE + 0x0150)
+#define HBG_REG_TX_TAGGED_ADDR			(HBG_REG_SGMII_BASE + 0x0154)
+#define HBG_REG_TX_CRC_ERROR_ADDR		(HBG_REG_SGMII_BASE + 0x0158)
+#define HBG_REG_TX_PAUSE_FRAMES_ADDR		(HBG_REG_SGMII_BASE + 0x015C)
+#define HBG_REG_LINE_LOOP_BACK_ADDR		(HBG_REG_SGMII_BASE + 0x01A8)
+#define HBG_REG_CF_CRC_STRIP_ADDR		(HBG_REG_SGMII_BASE + 0x01B0)
+#define HBG_REG_CF_CRC_STRIP_B			BIT(0)
+#define HBG_REG_MODE_CHANGE_EN_ADDR		(HBG_REG_SGMII_BASE + 0x01B4)
+#define HBG_REG_MODE_CHANGE_EN_B		BIT(0)
+#define HBG_REG_LOOP_REG_ADDR			(HBG_REG_SGMII_BASE + 0x01DC)
+#define HBG_REG_RECV_CTRL_ADDR			(HBG_REG_SGMII_BASE + 0x01E0)
+#define HBG_REG_RECV_CTRL_STRIP_PAD_EN_B	BIT(3)
+#define HBG_REG_VLAN_CODE_ADDR			(HBG_REG_SGMII_BASE + 0x01E8)
+#define HBG_REG_RX_OVERRUN_CNT_ADDR		(HBG_REG_SGMII_BASE + 0x01EC)
+#define HBG_REG_RX_LENGTHFIELD_ERR_CNT_ADDR	(HBG_REG_SGMII_BASE + 0x01F4)
+#define HBG_REG_RX_FAIL_COMMA_CNT_ADDR		(HBG_REG_SGMII_BASE + 0x01F8)
+#define HBG_REG_STATION_ADDR_LOW_0_ADDR		(HBG_REG_SGMII_BASE + 0x0200)
+#define HBG_REG_STATION_ADDR_HIGH_0_ADDR	(HBG_REG_SGMII_BASE + 0x0204)
+#define HBG_REG_STATION_ADDR_LOW_1_ADDR		(HBG_REG_SGMII_BASE + 0x0208)
+#define HBG_REG_STATION_ADDR_HIGH_1_ADDR	(HBG_REG_SGMII_BASE + 0x020C)
+#define HBG_REG_STATION_ADDR_LOW_2_ADDR		(HBG_REG_SGMII_BASE + 0x0210)
+#define HBG_REG_STATION_ADDR_HIGH_2_ADDR	(HBG_REG_SGMII_BASE + 0x0214)
+#define HBG_REG_STATION_ADDR_LOW_3_ADDR		(HBG_REG_SGMII_BASE + 0x0218)
+#define HBG_REG_STATION_ADDR_HIGH_3_ADDR	(HBG_REG_SGMII_BASE + 0x021C)
+#define HBG_REG_STATION_ADDR_LOW_4_ADDR		(HBG_REG_SGMII_BASE + 0x0220)
+#define HBG_REG_STATION_ADDR_HIGH_4_ADDR	(HBG_REG_SGMII_BASE + 0x0224)
+#define HBG_REG_STATION_ADDR_LOW_5_ADDR		(HBG_REG_SGMII_BASE + 0x0228)
+#define HBG_REG_STATION_ADDR_HIGH_5_ADDR	(HBG_REG_SGMII_BASE + 0x022C)
+#define HBG_REG_STATION_ADDR_LOW_MSK_0		(HBG_REG_SGMII_BASE + 0x0230)
+#define HBG_REG_STATION_ADDR_LOW_MSK_1		(HBG_REG_SGMII_BASE + 0x0238)
+
+/* PCU */
+#define HBG_REG_TX_FIFO_THRSLD_ADDR		(HBG_REG_SGMII_BASE + 0x0420)
+#define HBG_REG_RX_FIFO_THRSLD_ADDR		(HBG_REG_SGMII_BASE + 0x0424)
+#define HBG_REG_FIFO_THRSLD_FULL_M		GENMASK(25, 16)
+#define HBG_REG_FIFO_THRSLD_EMPTY_M		GENMASK(9, 0)
+#define HBG_REG_CFG_FIFO_THRSLD_ADDR		(HBG_REG_SGMII_BASE + 0x0428)
+#define HBG_REG_CFG_FIFO_THRSLD_TX_FULL_M	GENMASK(31, 24)
+#define HBG_REG_CFG_FIFO_THRSLD_TX_EMPTY_M	GENMASK(23, 16)
+#define HBG_REG_CFG_FIFO_THRSLD_RX_FULL_M	GENMASK(15, 8)
+#define HBG_REG_CFG_FIFO_THRSLD_RX_EMPTY_M	GENMASK(7, 0)
+#define HBG_REG_CF_INTRPT_MSK_ADDR		(HBG_REG_SGMII_BASE + 0x042C)
+#define HBG_INT_MSK_WE_ERR_B			BIT(31)
+#define HBG_INT_MSK_RBREQ_ERR_B			BIT(30)
+#define HBG_INT_MSK_MAC_FIFO_ERR_B		BIT(29)
+#define HBG_INT_MSK_RX_AHB_ERR_B		BIT(28)
+#define HBG_INT_MSK_RX_DROP_B			BIT(26)
+#define HBG_INT_MSK_TX_DROP_B			BIT(25)
+#define HBG_INT_MSK_TXCFG_AVL_B			BIT(24)
+#define HBG_INT_MSK_REL_BUF_ERR_B		BIT(23)
+#define HBG_INT_MSK_RX_BUF_AVL_B		BIT(22)
+#define HBG_INT_MSK_TX_AHB_ERR_B		BIT(21)
+#define HBG_INT_MSK_SRAM_PARITY_ERR_B		BIT(20)
+#define HBG_INT_MSK_MAC_APP_TX_FIFO_ERR_B	BIT(19)
+#define HBG_INT_MSK_MAC_APP_RX_FIFO_ERR_B	BIT(18)
+#define HBG_INT_MSK_MAC_PCS_TX_FIFO_ERR_B	BIT(17)
+#define HBG_INT_MSK_MAC_PCS_RX_FIFO_ERR_B	BIT(16)
+#define HBG_INT_MSK_MAC_MII_FIFO_ERR_B		BIT(15)
+#define HBG_INT_MSK_TX_PKT_CPL_B		BIT(14)
+#define HBG_INT_MSK_TX_B			BIT(1) /* just used in driver */
+#define HBG_INT_MSK_RX_B			BIT(0) /* just used in driver */
+#define HBG_REG_CF_INTRPT_STAT_ADDR		(HBG_REG_SGMII_BASE + 0x0434)
+#define HBG_REG_CF_INTRPT_CLR_ADDR		(HBG_REG_SGMII_BASE + 0x0438)
+#define HBG_REG_TX_BUS_ERR_ADDR_ADDR		(HBG_REG_SGMII_BASE + 0x043C)
+#define HBG_REG_RX_BUS_ERR_ADDR_ADDR		(HBG_REG_SGMII_BASE + 0x0440)
+#define HBG_REG_MAX_FRAME_LEN_ADDR		(HBG_REG_SGMII_BASE + 0x0444)
+#define HBG_REG_MAX_FRAME_LEN_M			GENMASK(15, 0)
+#define HBG_REG_TX_DROP_CNT_ADDR		(HBG_REG_SGMII_BASE + 0x0448)
+#define HBG_REG_RX_OVER_FLOW_CNT_ADDR		(HBG_REG_SGMII_BASE + 0x044C)
+#define HBG_REG_DEBUG_ST_MCH_ADDR		(HBG_REG_SGMII_BASE + 0x0450)
+#define HBG_REG_FIFO_CURR_STATUS_ADDR		(HBG_REG_SGMII_BASE + 0x0454)
+#define HBG_REG_FIFO_HIST_STATUS_ADDR		(HBG_REG_SGMII_BASE + 0x0458)
+#define HBG_REG_CF_CFF_DATA_NUM_ADDR		(HBG_REG_SGMII_BASE + 0x045C)
+#define HBG_REG_CF_CFF_DATA_NUM_ADDR_TX_M	GENMASK(8, 0)
+#define HBG_REG_CF_CFF_DATA_NUM_ADDR_RX_M	GENMASK(24, 16)
+#define HBG_REG_TX_CS_FAIL_CNT_ADDR		(HBG_REG_SGMII_BASE + 0x0460)
+#define HBG_REG_RX_TRANS_PKG_CNT_ADDR		(HBG_REG_SGMII_BASE + 0x0464)
+#define HBG_REG_TX_TRANS_PKG_CNT_ADDR		(HBG_REG_SGMII_BASE + 0x0468)
+#define HBG_REG_CF_TX_PAUSE_ADDR		(HBG_REG_SGMII_BASE + 0x0470)
+#define HBG_REG_TX_CFF_ADDR_0_ADDR		(HBG_REG_SGMII_BASE + 0x0488)
+#define HBG_REG_TX_CFF_ADDR_1_ADDR		(HBG_REG_SGMII_BASE + 0x048C)
+#define HBG_REG_TX_CFF_ADDR_2_ADDR		(HBG_REG_SGMII_BASE + 0x0490)
+#define HBG_REG_TX_CFF_ADDR_3_ADDR		(HBG_REG_SGMII_BASE + 0x0494)
+#define HBG_REG_RX_CFF_ADDR_ADDR		(HBG_REG_SGMII_BASE + 0x04A0)
+#define HBG_REG_BRUST_LENGTH_ADDR		(HBG_REG_SGMII_BASE + 0x04C4)
+#define HBG_REG_BRUST_LENGTH_B			BIT(29)
+#define HBG_REG_RX_BUF_SIZE_ADDR		(HBG_REG_SGMII_BASE + 0x04E4)
+#define HBG_REG_RX_BUF_SIZE_M			GENMASK(15, 0)
+#define HBG_REG_BUS_CTRL_ADDR			(HBG_REG_SGMII_BASE + 0x04E8)
+#define HBG_REG_BUS_CTRL_ENDIAN_M		GENMASK(2, 1)
+#define HBG_REG_RX_CTRL_ADDR			(HBG_REG_SGMII_BASE + 0x04F0)
+#define HBG_REG_RX_CTRL_RXBUF_1ST_SKIP_SIZE_M	GENMASK(31, 28)
+#define HBG_REG_RX_CTRL_TIME_INF_EN_B		BIT(23)
+#define HBG_REG_RX_CTRL_RX_ALIGN_NUM_M		GENMASK(18, 17)
+#define HBG_REG_RX_CTRL_PORT_NUM		GENMASK(16, 13)
+#define HBG_REG_RX_CTRL_RX_GET_ADDR_MODE_B	BIT(12)
+#define HBG_REG_RX_CTRL_RXBUF_1ST_SKIP_SIZE2_M	GENMASK(3, 0)
+#define HBG_REG_RX_PKT_MODE_ADDR		(HBG_REG_SGMII_BASE + 0x04F4)
+#define HBG_REG_RX_PKT_MODE_PARSE_MODE_M	GENMASK(22, 21)
+#define HBG_REG_RX_BUFRQ_ERR_CNT_ADDR		(HBG_REG_SGMII_BASE + 0x058C)
+#define HBG_REG_TX_BUFRL_ERR_CNT_ADDR		(HBG_REG_SGMII_BASE + 0x0590)
+#define HBG_REG_RX_WE_ERR_CNT_ADDR		(HBG_REG_SGMII_BASE + 0x0594)
+#define HBG_REG_DBG_ST0_ADDR			(HBG_REG_SGMII_BASE + 0x05E4)
+#define HBG_REG_DBG_ST1_ADDR			(HBG_REG_SGMII_BASE + 0x05E8)
+#define HBG_REG_DBG_ST2_ADDR			(HBG_REG_SGMII_BASE + 0x05EC)
+#define HBG_REG_BUS_RST_EN_ADDR			(HBG_REG_SGMII_BASE + 0x0688)
+#define HBG_REG_CF_IND_TXINT_MSK_ADDR		(HBG_REG_SGMII_BASE + 0x0694)
+#define HBG_REG_IND_INTR_MASK_B			BIT(0)
+#define HBG_REG_CF_IND_TXINT_STAT_ADDR		(HBG_REG_SGMII_BASE + 0x0698)
+#define HBG_REG_CF_IND_TXINT_CLR_ADDR		(HBG_REG_SGMII_BASE + 0x069C)
+#define HBG_REG_CF_IND_RXINT_MSK_ADDR		(HBG_REG_SGMII_BASE + 0x06a0)
+#define HBG_REG_CF_IND_RXINT_STAT_ADDR		(HBG_REG_SGMII_BASE + 0x06a4)
+#define HBG_REG_CF_IND_RXINT_CLR_ADDR		(HBG_REG_SGMII_BASE + 0x06a8)
+
+enum hbg_port_mode {
+	/* 0x0 ~ 0x5 are reserved */
+	HBG_PORT_MODE_SGMII_10M = 0x6,
+	HBG_PORT_MODE_SGMII_100M = 0x7,
+	HBG_PORT_MODE_SGMII_1000M = 0x8,
+};
+
+struct hbg_tx_desc {
+	u32 word0;
+	u32 word1;
+	u32 word2; /* pkt_addr */
+	u32 word3; /* clear_addr */
+};
+
+#define HBG_TX_DESC_W0_IP_OFF_M		GENMASK(30, 26)
+#define HBG_TX_DESC_W0_l3_CS_B		BIT(2)
+#define HBG_TX_DESC_W0_WB_B		BIT(1)
+#define HBG_TX_DESC_W0_l4_CS_B		BIT(0)
+#define HBG_TX_DESC_W1_SEND_LEN_M	GENMASK(19, 4)
+
+struct hbg_rx_desc {
+	u32 word0;
+	u32 word1; /* tag */
+	u32 word2;
+	u32 word3;
+	u32 word4;
+	u32 word5;
+};
+
+#define HBG_RX_DESC_W2_PKT_LEN_M	GENMASK(31, 16)
+#define HBG_RX_DESC_W2_PORT_NUM_M	GENMASK(15, 12)
+#define HBG_RX_DESC_W4_IP_TCP_UDP_M	GENMASK(31, 30)
+#define HBG_RX_DESC_W4_IPSEC_B		BIT(29)
+#define HBG_RX_DESC_W4_IP_VERSION_B	BIT(28)
+#define HBG_RX_DESC_W4_L4_ERR_CODE_M	GENMASK(26, 23)
+#define HBG_RX_DESC_W4_FRAG_B		BIT(22)
+#define HBG_RX_DESC_W4_OPT_B		BIT(21)
+#define HBG_RX_DESC_W4_IP_VERSION_ERR_B	BIT(20)
+#define HBG_RX_DESC_W4_BRD_CST_B	BIT(19)
+#define HBG_RX_DESC_W4_MUL_CST_B	BIT(18)
+#define HBG_RX_DESC_W4_ARP_B		BIT(17)
+#define HBG_RX_DESC_W4_RARP_B		BIT(16)
+#define HBG_RX_DESC_W4_ICMP_B		BIT(15)
+#define HBG_RX_DESC_W4_VLAN_FLAG_B	BIT(14)
+#define HBG_RX_DESC_W4_DROP_B		BIT(13)
+#define HBG_RX_DESC_W4_L3_ERR_CODE_M	GENMASK(12, 9)
+#define HBG_RX_DESC_W4_L2_ERR_B		BIT(8)
+#define HBG_RX_DESC_W4_IDX_MATCH_B	BIT(7)
+
+enum hbg_l3_err_code {
+	HBG_L3_OK = 0,
+	HBG_L3_WRONG_HEAD,
+	HBG_L3_CSUM_ERR,
+	HBG_L3_LEN_ERR,
+	HBG_L3_ZERO_TTL,
+	HBG_L3_RSVD,
+};
+
+enum hbg_l4_err_code {
+	HBG_L4_OK = 0,
+	HBG_L4_WRONG_HEAD,
+	HBG_L4_LEN_ERR,
+	HBG_L4_CSUM_ERR,
+	HBG_L4_ZERO_PORT_NUM,
+	HBG_L4_RSVD,
+};
+
+enum hbg_pkt_type_code {
+	HBG_NO_IP_PKT = 0,
+	HBG_IP_PKT,
+	HBG_TCP_PKT,
+	HBG_UDP_PKT,
+};
+
+#endif
diff --git a/drivers/net/ethernet/hisilicon/hibmcge/hbg_txrx.c b/drivers/net/ethernet/hisilicon/hibmcge/hbg_txrx.c
new file mode 100644
index 000000000000..8d814c8f19ea
--- /dev/null
+++ b/drivers/net/ethernet/hisilicon/hibmcge/hbg_txrx.c
@@ -0,0 +1,586 @@
+// SPDX-License-Identifier: GPL-2.0+
+// Copyright (c) 2024 Hisilicon Limited.
+
+#include <net/netdev_queues.h>
+#include "hbg_common.h"
+#include "hbg_irq.h"
+#include "hbg_reg.h"
+#include "hbg_txrx.h"
+
+#define netdev_get_tx_ring(netdev) \
+			(&(((struct hbg_priv *)netdev_priv(netdev))->tx_ring))
+
+#define buffer_to_dma_dir(buffer) (((buffer)->dir == HBG_DIR_RX) ? \
+				   DMA_FROM_DEVICE : DMA_TO_DEVICE)
+
+#define hbg_queue_used_num(head, tail, ring) ({ \
+	typeof(ring) _ring = (ring); \
+	((tail) + _ring->len - (head)) % _ring->len; })
+#define hbg_queue_left_num(head, tail, ring) ({ \
+	typeof(ring) _r = (ring); \
+	_r->len - hbg_queue_used_num((head), (tail), _r) - 1; })
+#define hbg_queue_is_empty(head, tail, ring) \
+	(hbg_queue_used_num((head), (tail), (ring)) == 0)
+#define hbg_queue_is_full(head, tail, ring) \
+	(hbg_queue_left_num((head), (tail), (ring)) == 0)
+#define hbg_queue_next_prt(p, ring) (((p) + 1) % (ring)->len)
+#define hbg_queue_move_next(p, ring) ({ \
+	typeof(ring) _ring = (ring); \
+	_ring->p = hbg_queue_next_prt(_ring->p, _ring); })
+
+#define HBG_TX_STOP_THRS	2
+#define HBG_TX_START_THRS	(2 * HBG_TX_STOP_THRS)
+
+static int hbg_dma_map(struct hbg_buffer *buffer)
+{
+	struct hbg_priv *priv = buffer->priv;
+
+	buffer->skb_dma = dma_map_single(&priv->pdev->dev,
+					 buffer->skb->data, buffer->skb_len,
+					 buffer_to_dma_dir(buffer));
+	if (unlikely(dma_mapping_error(&priv->pdev->dev, buffer->skb_dma))) {
+		if (buffer->dir == HBG_DIR_RX)
+			priv->stats.rx_dma_err_cnt++;
+		else
+			priv->stats.tx_dma_err_cnt++;
+
+		return -ENOMEM;
+	}
+
+	return 0;
+}
+
+static void hbg_dma_unmap(struct hbg_buffer *buffer)
+{
+	struct hbg_priv *priv = buffer->priv;
+
+	if (unlikely(!buffer->skb_dma))
+		return;
+
+	dma_unmap_single(&priv->pdev->dev, buffer->skb_dma, buffer->skb_len,
+			 buffer_to_dma_dir(buffer));
+	buffer->skb_dma = 0;
+}
+
+static void hbg_init_tx_desc(struct hbg_buffer *buffer,
+			     struct hbg_tx_desc *tx_desc)
+{
+	u32 ip_offset = buffer->skb->network_header - buffer->skb->mac_header;
+	u32 word0 = 0;
+
+	word0 |= FIELD_PREP(HBG_TX_DESC_W0_WB_B, HBG_STATUS_ENABLE);
+	word0 |= FIELD_PREP(HBG_TX_DESC_W0_IP_OFF_M, ip_offset);
+	if (likely(buffer->skb->ip_summed == CHECKSUM_PARTIAL)) {
+		word0 |= FIELD_PREP(HBG_TX_DESC_W0_l3_CS_B, HBG_STATUS_ENABLE);
+		word0 |= FIELD_PREP(HBG_TX_DESC_W0_l4_CS_B, HBG_STATUS_ENABLE);
+	}
+
+	tx_desc->word0 = word0;
+	tx_desc->word1 = FIELD_PREP(HBG_TX_DESC_W1_SEND_LEN_M,
+				    buffer->skb->len);
+	tx_desc->word2 = buffer->skb_dma;
+	tx_desc->word3 = buffer->state_dma;
+}
+
+netdev_tx_t hbg_net_start_xmit(struct sk_buff *skb, struct net_device *netdev)
+{
+	struct hbg_ring *ring = netdev_get_tx_ring(netdev);
+	struct hbg_priv *priv = netdev_priv(netdev);
+	/* This smp_load_acquire() pairs with smp_store_release() in
+	 * hbg_napi_tx_recycle() called in tx interrupt handle process.
+	 */
+	u32 ntc = smp_load_acquire(&ring->ntc);
+	struct hbg_buffer *buffer;
+	struct hbg_tx_desc tx_desc;
+	u32 ntu = ring->ntu;
+
+	if (unlikely(!skb->len ||
+		     skb->len > hbg_spec_max_frame_len(priv, HBG_DIR_TX))) {
+		dev_kfree_skb_any(skb);
+		netdev->stats.tx_errors++;
+		return NETDEV_TX_OK;
+	}
+
+	if (!netif_subqueue_maybe_stop(netdev, 0,
+				       hbg_queue_left_num(ntc, ntu, ring),
+				       HBG_TX_STOP_THRS, HBG_TX_START_THRS))
+		return NETDEV_TX_BUSY;
+
+	buffer = &ring->queue[ntu];
+	buffer->skb = skb;
+	buffer->skb_len = skb->len;
+	if (unlikely(hbg_dma_map(buffer))) {
+		dev_kfree_skb_any(skb);
+		return NETDEV_TX_OK;
+	}
+
+	buffer->state = HBG_TX_STATE_START;
+	hbg_init_tx_desc(buffer, &tx_desc);
+	hbg_hw_set_tx_desc(priv, &tx_desc);
+
+	/* This smp_store_release() pairs with smp_load_acquire() in
+	 * hbg_napi_tx_recycle() called in tx interrupt handle process.
+	 */
+	smp_store_release(&ring->ntu, hbg_queue_next_prt(ntu, ring));
+	dev_sw_netstats_tx_add(netdev, 1, skb->len);
+	return NETDEV_TX_OK;
+}
+
+static void hbg_buffer_free_skb(struct hbg_buffer *buffer)
+{
+	if (unlikely(!buffer->skb))
+		return;
+
+	dev_kfree_skb_any(buffer->skb);
+	buffer->skb = NULL;
+}
+
+static int hbg_buffer_alloc_skb(struct hbg_buffer *buffer)
+{
+	u32 len = hbg_spec_max_frame_len(buffer->priv, buffer->dir);
+	struct hbg_priv *priv = buffer->priv;
+
+	buffer->skb = netdev_alloc_skb(priv->netdev, len);
+	if (unlikely(!buffer->skb))
+		return -ENOMEM;
+
+	buffer->skb_len = len;
+	memset(buffer->skb->data, 0, HBG_PACKET_HEAD_SIZE);
+	return 0;
+}
+
+static void hbg_buffer_free(struct hbg_buffer *buffer)
+{
+	hbg_dma_unmap(buffer);
+	hbg_buffer_free_skb(buffer);
+}
+
+static int hbg_napi_tx_recycle(struct napi_struct *napi, int budget)
+{
+	struct hbg_ring *ring = container_of(napi, struct hbg_ring, napi);
+	/* This smp_load_acquire() pairs with smp_store_release() in
+	 * hbg_net_start_xmit() called in xmit process.
+	 */
+	u32 ntu = smp_load_acquire(&ring->ntu);
+	struct hbg_priv *priv = ring->priv;
+	struct hbg_buffer *buffer;
+	u32 ntc = ring->ntc;
+	int packet_done = 0;
+
+	/* We need do cleanup even if budget is 0.
+	 * Per NAPI documentation budget is for Rx.
+	 * So We hardcode the amount of work Tx NAPI does to 128.
+	 */
+	budget = 128;
+	while (packet_done < budget) {
+		if (unlikely(hbg_queue_is_empty(ntc, ntu, ring)))
+			break;
+
+		/* make sure HW write desc complete */
+		dma_rmb();
+
+		buffer = &ring->queue[ntc];
+		if (buffer->state != HBG_TX_STATE_COMPLETE)
+			break;
+
+		hbg_buffer_free(buffer);
+		ntc = hbg_queue_next_prt(ntc, ring);
+		packet_done++;
+	}
+
+	/* This smp_store_release() pairs with smp_load_acquire() in
+	 * hbg_net_start_xmit() called in xmit process.
+	 */
+	smp_store_release(&ring->ntc, ntc);
+	netif_wake_queue(priv->netdev);
+
+	if (likely(packet_done < budget &&
+		   napi_complete_done(napi, packet_done)))
+		hbg_hw_irq_enable(priv, HBG_INT_MSK_TX_B, true);
+
+	return packet_done;
+}
+
+static bool hbg_rx_check_l3l4_error(struct hbg_priv *priv,
+				    struct hbg_rx_desc *desc,
+				    struct sk_buff *skb)
+{
+	bool rx_checksum_offload = !!(priv->netdev->features & NETIF_F_RXCSUM);
+
+	skb->ip_summed = rx_checksum_offload ?
+			 CHECKSUM_UNNECESSARY : CHECKSUM_NONE;
+
+	if (likely(!FIELD_GET(HBG_RX_DESC_W4_L3_ERR_CODE_M, desc->word4) &&
+		   !FIELD_GET(HBG_RX_DESC_W4_L4_ERR_CODE_M, desc->word4)))
+		return true;
+
+	switch (FIELD_GET(HBG_RX_DESC_W4_L3_ERR_CODE_M, desc->word4)) {
+	case HBG_L3_OK:
+		break;
+	case HBG_L3_WRONG_HEAD:
+		priv->stats.rx_desc_l3_wrong_head_cnt++;
+		return false;
+	case HBG_L3_CSUM_ERR:
+		skb->ip_summed = CHECKSUM_NONE;
+		priv->stats.rx_desc_l3_csum_err_cnt++;
+
+		/* Don't drop packets on csum validation failure,
+		 * suggest by Jakub
+		 */
+		break;
+	case HBG_L3_LEN_ERR:
+		priv->stats.rx_desc_l3_len_err_cnt++;
+		return false;
+	case HBG_L3_ZERO_TTL:
+		priv->stats.rx_desc_l3_zero_ttl_cnt++;
+		return false;
+	default:
+		priv->stats.rx_desc_l3_other_cnt++;
+		return false;
+	}
+
+	switch (FIELD_GET(HBG_RX_DESC_W4_L4_ERR_CODE_M, desc->word4)) {
+	case HBG_L4_OK:
+		break;
+	case HBG_L4_WRONG_HEAD:
+		priv->stats.rx_desc_l4_wrong_head_cnt++;
+		return false;
+	case HBG_L4_LEN_ERR:
+		priv->stats.rx_desc_l4_len_err_cnt++;
+		return false;
+	case HBG_L4_CSUM_ERR:
+		skb->ip_summed = CHECKSUM_NONE;
+		priv->stats.rx_desc_l4_csum_err_cnt++;
+
+		/* Don't drop packets on csum validation failure,
+		 * suggest by Jakub
+		 */
+		break;
+	case HBG_L4_ZERO_PORT_NUM:
+		priv->stats.rx_desc_l4_zero_port_num_cnt++;
+		return false;
+	default:
+		priv->stats.rx_desc_l4_other_cnt++;
+		return false;
+	}
+
+	return true;
+}
+
+static void hbg_update_rx_ip_protocol_stats(struct hbg_priv *priv,
+					    struct hbg_rx_desc *desc)
+{
+	if (unlikely(!FIELD_GET(HBG_RX_DESC_W4_IP_TCP_UDP_M, desc->word4))) {
+		priv->stats.rx_desc_no_ip_pkt_cnt++;
+		return;
+	}
+
+	if (unlikely(FIELD_GET(HBG_RX_DESC_W4_IP_VERSION_ERR_B, desc->word4))) {
+		priv->stats.rx_desc_ip_ver_err_cnt++;
+		return;
+	}
+
+	/* 0:ipv4, 1:ipv6 */
+	if (FIELD_GET(HBG_RX_DESC_W4_IP_VERSION_B, desc->word4))
+		priv->stats.rx_desc_ipv6_pkt_cnt++;
+	else
+		priv->stats.rx_desc_ipv4_pkt_cnt++;
+
+	switch (FIELD_GET(HBG_RX_DESC_W4_IP_TCP_UDP_M, desc->word4)) {
+	case HBG_IP_PKT:
+		priv->stats.rx_desc_ip_pkt_cnt++;
+		if (FIELD_GET(HBG_RX_DESC_W4_OPT_B, desc->word4))
+			priv->stats.rx_desc_ip_opt_pkt_cnt++;
+		if (FIELD_GET(HBG_RX_DESC_W4_FRAG_B, desc->word4))
+			priv->stats.rx_desc_frag_cnt++;
+
+		if (FIELD_GET(HBG_RX_DESC_W4_ICMP_B, desc->word4))
+			priv->stats.rx_desc_icmp_pkt_cnt++;
+		else if (FIELD_GET(HBG_RX_DESC_W4_IPSEC_B, desc->word4))
+			priv->stats.rx_desc_ipsec_pkt_cnt++;
+		break;
+	case HBG_TCP_PKT:
+		priv->stats.rx_desc_tcp_pkt_cnt++;
+		break;
+	case HBG_UDP_PKT:
+		priv->stats.rx_desc_udp_pkt_cnt++;
+		break;
+	default:
+		priv->stats.rx_desc_no_ip_pkt_cnt++;
+		break;
+	}
+}
+
+static void hbg_update_rx_protocol_stats(struct hbg_priv *priv,
+					 struct hbg_rx_desc *desc)
+{
+	if (unlikely(!FIELD_GET(HBG_RX_DESC_W4_IDX_MATCH_B, desc->word4))) {
+		priv->stats.rx_desc_key_not_match_cnt++;
+		return;
+	}
+
+	if (FIELD_GET(HBG_RX_DESC_W4_BRD_CST_B, desc->word4))
+		priv->stats.rx_desc_broadcast_pkt_cnt++;
+	else if (FIELD_GET(HBG_RX_DESC_W4_MUL_CST_B, desc->word4))
+		priv->stats.rx_desc_multicast_pkt_cnt++;
+
+	if (FIELD_GET(HBG_RX_DESC_W4_VLAN_FLAG_B, desc->word4))
+		priv->stats.rx_desc_vlan_pkt_cnt++;
+
+	if (FIELD_GET(HBG_RX_DESC_W4_ARP_B, desc->word4)) {
+		priv->stats.rx_desc_arp_pkt_cnt++;
+		return;
+	} else if (FIELD_GET(HBG_RX_DESC_W4_RARP_B, desc->word4)) {
+		priv->stats.rx_desc_rarp_pkt_cnt++;
+		return;
+	}
+
+	hbg_update_rx_ip_protocol_stats(priv, desc);
+}
+
+static bool hbg_rx_pkt_check(struct hbg_priv *priv, struct hbg_rx_desc *desc,
+			     struct sk_buff *skb)
+{
+	if (unlikely(FIELD_GET(HBG_RX_DESC_W2_PKT_LEN_M, desc->word2) >
+		     priv->dev_specs.max_frame_len)) {
+		priv->stats.rx_desc_pkt_len_err_cnt++;
+		return false;
+	}
+
+	if (unlikely(FIELD_GET(HBG_RX_DESC_W2_PORT_NUM_M, desc->word2) !=
+		     priv->dev_specs.mac_id ||
+		     FIELD_GET(HBG_RX_DESC_W4_DROP_B, desc->word4))) {
+		priv->stats.rx_desc_drop++;
+		return false;
+	}
+
+	if (unlikely(FIELD_GET(HBG_RX_DESC_W4_L2_ERR_B, desc->word4))) {
+		priv->stats.rx_desc_l2_err_cnt++;
+		return false;
+	}
+
+	if (unlikely(!hbg_rx_check_l3l4_error(priv, desc, skb))) {
+		priv->stats.rx_desc_l3l4_err_cnt++;
+		return false;
+	}
+
+	hbg_update_rx_protocol_stats(priv, desc);
+	return true;
+}
+
+static int hbg_rx_fill_one_buffer(struct hbg_priv *priv)
+{
+	struct hbg_ring *ring = &priv->rx_ring;
+	struct hbg_buffer *buffer;
+	int ret;
+
+	if (hbg_queue_is_full(ring->ntc, ring->ntu, ring))
+		return 0;
+
+	buffer = &ring->queue[ring->ntu];
+	ret = hbg_buffer_alloc_skb(buffer);
+	if (unlikely(ret))
+		return ret;
+
+	ret = hbg_dma_map(buffer);
+	if (unlikely(ret)) {
+		hbg_buffer_free_skb(buffer);
+		return ret;
+	}
+
+	hbg_hw_fill_buffer(priv, buffer->skb_dma);
+	hbg_queue_move_next(ntu, ring);
+	return 0;
+}
+
+static bool hbg_sync_data_from_hw(struct hbg_priv *priv,
+				  struct hbg_buffer *buffer)
+{
+	struct hbg_rx_desc *rx_desc;
+
+	/* make sure HW write desc complete */
+	dma_rmb();
+
+	dma_sync_single_for_cpu(&priv->pdev->dev, buffer->skb_dma,
+				buffer->skb_len, DMA_FROM_DEVICE);
+
+	rx_desc = (struct hbg_rx_desc *)buffer->skb->data;
+	return FIELD_GET(HBG_RX_DESC_W2_PKT_LEN_M, rx_desc->word2) != 0;
+}
+
+static int hbg_napi_rx_poll(struct napi_struct *napi, int budget)
+{
+	struct hbg_ring *ring = container_of(napi, struct hbg_ring, napi);
+	struct hbg_priv *priv = ring->priv;
+	struct hbg_rx_desc *rx_desc;
+	struct hbg_buffer *buffer;
+	u32 packet_done = 0;
+	u32 pkt_len;
+
+	while (packet_done < budget) {
+		if (unlikely(hbg_queue_is_empty(ring->ntc, ring->ntu, ring)))
+			break;
+
+		buffer = &ring->queue[ring->ntc];
+		if (unlikely(!buffer->skb))
+			goto next_buffer;
+
+		if (unlikely(!hbg_sync_data_from_hw(priv, buffer)))
+			break;
+		rx_desc = (struct hbg_rx_desc *)buffer->skb->data;
+		pkt_len = FIELD_GET(HBG_RX_DESC_W2_PKT_LEN_M, rx_desc->word2);
+
+		if (unlikely(!hbg_rx_pkt_check(priv, rx_desc, buffer->skb))) {
+			hbg_buffer_free(buffer);
+			goto next_buffer;
+		}
+
+		hbg_dma_unmap(buffer);
+		skb_reserve(buffer->skb, HBG_PACKET_HEAD_SIZE + NET_IP_ALIGN);
+		skb_put(buffer->skb, pkt_len);
+		buffer->skb->protocol = eth_type_trans(buffer->skb,
+						       priv->netdev);
+
+		dev_sw_netstats_rx_add(priv->netdev, pkt_len);
+		napi_gro_receive(napi, buffer->skb);
+		buffer->skb = NULL;
+
+next_buffer:
+		hbg_rx_fill_one_buffer(priv);
+		hbg_queue_move_next(ntc, ring);
+		packet_done++;
+	}
+
+	if (likely(packet_done < budget &&
+		   napi_complete_done(napi, packet_done)))
+		hbg_hw_irq_enable(priv, HBG_INT_MSK_RX_B, true);
+
+	return packet_done;
+}
+
+static void hbg_ring_uninit(struct hbg_ring *ring)
+{
+	struct hbg_buffer *buffer;
+	u32 i;
+
+	if (!ring->queue)
+		return;
+
+	napi_disable(&ring->napi);
+	netif_napi_del(&ring->napi);
+
+	for (i = 0; i < ring->len; i++) {
+		buffer = &ring->queue[i];
+		hbg_buffer_free(buffer);
+		buffer->ring = NULL;
+		buffer->priv = NULL;
+	}
+
+	dma_free_coherent(&ring->priv->pdev->dev,
+			  ring->len * sizeof(*ring->queue),
+			  ring->queue, ring->queue_dma);
+	ring->queue = NULL;
+	ring->queue_dma = 0;
+	ring->len = 0;
+	ring->priv = NULL;
+}
+
+static int hbg_ring_init(struct hbg_priv *priv, struct hbg_ring *ring,
+			 int (*napi_poll)(struct napi_struct *, int),
+			 enum hbg_dir dir)
+{
+	struct hbg_buffer *buffer;
+	u32 i, len;
+
+	len = hbg_get_spec_fifo_max_num(priv, dir) + 1;
+	ring->queue = dma_alloc_coherent(&priv->pdev->dev,
+					 len * sizeof(*ring->queue),
+					 &ring->queue_dma, GFP_KERNEL);
+	if (!ring->queue)
+		return -ENOMEM;
+
+	for (i = 0; i < len; i++) {
+		buffer = &ring->queue[i];
+		buffer->skb_len = 0;
+		buffer->dir = dir;
+		buffer->ring = ring;
+		buffer->priv = priv;
+		buffer->state_dma = ring->queue_dma + (i * sizeof(*buffer));
+	}
+
+	ring->dir = dir;
+	ring->priv = priv;
+	ring->ntc = 0;
+	ring->ntu = 0;
+	ring->len = len;
+
+	if (dir == HBG_DIR_TX)
+		netif_napi_add_tx(priv->netdev, &ring->napi, napi_poll);
+	else
+		netif_napi_add(priv->netdev, &ring->napi, napi_poll);
+
+	napi_enable(&ring->napi);
+	return 0;
+}
+
+static int hbg_tx_ring_init(struct hbg_priv *priv)
+{
+	struct hbg_ring *tx_ring = &priv->tx_ring;
+
+	if (!tx_ring->tout_log_buf)
+		tx_ring->tout_log_buf = devm_kmalloc(&priv->pdev->dev,
+						     HBG_TX_TIMEOUT_BUF_LEN,
+						     GFP_KERNEL);
+
+	if (!tx_ring->tout_log_buf)
+		return -ENOMEM;
+
+	return hbg_ring_init(priv, tx_ring, hbg_napi_tx_recycle, HBG_DIR_TX);
+}
+
+static int hbg_rx_ring_init(struct hbg_priv *priv)
+{
+	int ret;
+	u32 i;
+
+	ret = hbg_ring_init(priv, &priv->rx_ring, hbg_napi_rx_poll, HBG_DIR_RX);
+	if (ret)
+		return ret;
+
+	for (i = 0; i < priv->rx_ring.len - 1; i++) {
+		ret = hbg_rx_fill_one_buffer(priv);
+		if (ret) {
+			hbg_ring_uninit(&priv->rx_ring);
+			return ret;
+		}
+	}
+
+	return 0;
+}
+
+int hbg_txrx_init(struct hbg_priv *priv)
+{
+	int ret;
+
+	ret = hbg_tx_ring_init(priv);
+	if (ret) {
+		dev_err(&priv->pdev->dev,
+			"failed to init tx ring, ret = %d\n", ret);
+		return ret;
+	}
+
+	ret = hbg_rx_ring_init(priv);
+	if (ret) {
+		dev_err(&priv->pdev->dev,
+			"failed to init rx ring, ret = %d\n", ret);
+		hbg_ring_uninit(&priv->tx_ring);
+	}
+
+	return ret;
+}
+
+void hbg_txrx_uninit(struct hbg_priv *priv)
+{
+	hbg_ring_uninit(&priv->tx_ring);
+	hbg_ring_uninit(&priv->rx_ring);
+}
diff --git a/drivers/net/ethernet/hisilicon/hibmcge/hbg_txrx.h b/drivers/net/ethernet/hisilicon/hibmcge/hbg_txrx.h
new file mode 100644
index 000000000000..8b6110599e10
--- /dev/null
+++ b/drivers/net/ethernet/hisilicon/hibmcge/hbg_txrx.h
@@ -0,0 +1,44 @@
+/* SPDX-License-Identifier: GPL-2.0+ */
+/* Copyright (c) 2024 Hisilicon Limited. */
+
+#ifndef __HBG_TXRX_H
+#define __HBG_TXRX_H
+
+#include <linux/etherdevice.h>
+#include "hbg_hw.h"
+
+static inline u32 hbg_spec_max_frame_len(struct hbg_priv *priv,
+					 enum hbg_dir dir)
+{
+	return (dir == HBG_DIR_TX) ? priv->dev_specs.max_frame_len :
+		priv->dev_specs.rx_buf_size;
+}
+
+static inline u32 hbg_get_spec_fifo_max_num(struct hbg_priv *priv,
+					    enum hbg_dir dir)
+{
+	return (dir == HBG_DIR_TX) ? priv->dev_specs.tx_fifo_num :
+		priv->dev_specs.rx_fifo_num;
+}
+
+static inline bool hbg_fifo_is_full(struct hbg_priv *priv, enum hbg_dir dir)
+{
+	return hbg_hw_get_fifo_used_num(priv, dir) >=
+	       hbg_get_spec_fifo_max_num(priv, dir);
+}
+
+static inline u32 hbg_get_queue_used_num(struct hbg_ring *ring)
+{
+	u32 len = READ_ONCE(ring->len);
+
+	if (!len)
+		return 0;
+
+	return (READ_ONCE(ring->ntu) + len - READ_ONCE(ring->ntc)) % len;
+}
+
+netdev_tx_t hbg_net_start_xmit(struct sk_buff *skb, struct net_device *netdev);
+int hbg_txrx_init(struct hbg_priv *priv);
+void hbg_txrx_uninit(struct hbg_priv *priv);
+
+#endif
diff --git a/drivers/net/ethernet/hisilicon/hip04_eth.c b/drivers/net/ethernet/hisilicon/hip04_eth.c
index 0cec06bec63e..18376bcc718a 100644
--- a/drivers/net/ethernet/hisilicon/hip04_eth.c
+++ b/drivers/net/ethernet/hisilicon/hip04_eth.c
@@ -1,11 +1,7 @@
+// SPDX-License-Identifier: GPL-2.0-or-later
 
 /* Copyright (c) 2014 Linaro Ltd.
  * Copyright (c) 2014 Hisilicon Limited.
- *
- * This program is free software; you can redistribute it and/or modify
- * it under the terms of the GNU General Public License as published by
- * the Free Software Foundation; either version 2 of the License, or
- * (at your option) any later version.
  */
 
 #include <linux/module.h>
@@ -20,6 +16,8 @@
 #include <linux/mfd/syscon.h>
 #include <linux/regmap.h>
 
+#define SC_PPE_RESET_DREQ		0x026C
+
 #define PPE_CFG_RX_ADDR			0x100
 #define PPE_CFG_POOL_GRP		0x300
 #define PPE_CFG_RX_BUF_SIZE		0x400
@@ -37,10 +35,23 @@
 #define GE_MODE_CHANGE_REG		0x1b4
 #define GE_RECV_CONTROL_REG		0x1e0
 #define GE_STATION_MAC_ADDRESS		0x210
-#define PPE_CFG_CPU_ADD_ADDR		0x580
-#define PPE_CFG_MAX_FRAME_LEN_REG	0x408
+
 #define PPE_CFG_BUS_CTRL_REG		0x424
 #define PPE_CFG_RX_CTRL_REG		0x428
+
+#if defined(CONFIG_HI13X1_GMAC)
+#define PPE_CFG_CPU_ADD_ADDR		0x6D0
+#define PPE_CFG_MAX_FRAME_LEN_REG	0x500
+#define PPE_CFG_RX_PKT_MODE_REG		0x504
+#define PPE_CFG_QOS_VMID_GEN		0x520
+#define PPE_CFG_RX_PKT_INT		0x740
+#define PPE_INTEN			0x700
+#define PPE_INTSTS			0x708
+#define PPE_RINT			0x704
+#define PPE_CFG_STS_MODE		0x880
+#else
+#define PPE_CFG_CPU_ADD_ADDR		0x580
+#define PPE_CFG_MAX_FRAME_LEN_REG	0x408
 #define PPE_CFG_RX_PKT_MODE_REG		0x438
 #define PPE_CFG_QOS_VMID_GEN		0x500
 #define PPE_CFG_RX_PKT_INT		0x538
@@ -48,8 +59,12 @@
 #define PPE_INTSTS			0x608
 #define PPE_RINT			0x604
 #define PPE_CFG_STS_MODE		0x700
+#endif /* CONFIG_HI13X1_GMAC */
+
 #define PPE_HIS_RX_PKT_CNT		0x804
 
+#define RESET_DREQ_ALL			0xffffffff
+
 /* REG_INTERRUPT */
 #define RCV_INT				BIT(10)
 #define RCV_NOBUF			BIT(8)
@@ -61,8 +76,15 @@
 /* TX descriptor config */
 #define TX_FREE_MEM			BIT(0)
 #define TX_READ_ALLOC_L3		BIT(1)
-#define TX_FINISH_CACHE_INV		BIT(2)
+#if defined(CONFIG_HI13X1_GMAC)
+#define TX_CLEAR_WB			BIT(7)
+#define TX_RELEASE_TO_PPE		BIT(4)
+#define TX_FINISH_CACHE_INV		BIT(6)
+#define TX_POOL_SHIFT			16
+#else
 #define TX_CLEAR_WB			BIT(4)
+#define TX_FINISH_CACHE_INV		BIT(2)
+#endif
 #define TX_L3_CHECKSUM			BIT(5)
 #define TX_LOOP_BACK			BIT(11)
 
@@ -97,18 +119,35 @@
 #define GE_RX_PORT_EN			BIT(1)
 #define GE_TX_PORT_EN			BIT(2)
 
-#define PPE_CFG_STS_RX_PKT_CNT_RC	BIT(12)
-
 #define PPE_CFG_RX_PKT_ALIGN		BIT(18)
-#define PPE_CFG_QOS_VMID_MODE		BIT(14)
+
+#if defined(CONFIG_HI13X1_GMAC)
+#define PPE_CFG_QOS_VMID_GRP_SHIFT	4
+#define PPE_CFG_RX_CTRL_ALIGN_SHIFT	7
+#define PPE_CFG_STS_RX_PKT_CNT_RC	BIT(0)
+#define PPE_CFG_QOS_VMID_MODE		BIT(15)
+#define PPE_CFG_BUS_LOCAL_REL		(BIT(9) | BIT(15) | BIT(19) | BIT(23))
+
+/* buf unit size is cache_line_size, which is 64, so the shift is 6 */
+#define PPE_BUF_SIZE_SHIFT		6
+#define PPE_TX_BUF_HOLD			BIT(31)
+#define SOC_CACHE_LINE_MASK		0x3F
+#else
 #define PPE_CFG_QOS_VMID_GRP_SHIFT	8
+#define PPE_CFG_RX_CTRL_ALIGN_SHIFT	11
+#define PPE_CFG_STS_RX_PKT_CNT_RC	BIT(12)
+#define PPE_CFG_QOS_VMID_MODE		BIT(14)
+#define PPE_CFG_BUS_LOCAL_REL		BIT(14)
+
+/* buf unit size is 1, so the shift is 6 */
+#define PPE_BUF_SIZE_SHIFT		0
+#define PPE_TX_BUF_HOLD			0
+#endif /* CONFIG_HI13X1_GMAC */
 
 #define PPE_CFG_RX_FIFO_FSFU		BIT(11)
 #define PPE_CFG_RX_DEPTH_SHIFT		16
 #define PPE_CFG_RX_START_SHIFT		0
-#define PPE_CFG_RX_CTRL_ALIGN_SHIFT	11
 
-#define PPE_CFG_BUS_LOCAL_REL		BIT(14)
 #define PPE_CFG_BUS_BIG_ENDIEN		BIT(0)
 
 #define RX_DESC_NUM			128
@@ -132,31 +171,56 @@
 #define HIP04_MIN_TX_COALESCE_FRAMES	100
 
 struct tx_desc {
+#if defined(CONFIG_HI13X1_GMAC)
+	u32 reserved1[2];
+	u32 send_addr;
+	u16 send_size;
+	u16 data_offset;
+	u32 reserved2[7];
+	u32 cfg;
+	u32 wb_addr;
+	u32 reserved3[3];
+#else
 	u32 send_addr;
 	u32 send_size;
 	u32 next_addr;
 	u32 cfg;
 	u32 wb_addr;
+#endif
 } __aligned(64);
 
 struct rx_desc {
+#if defined(CONFIG_HI13X1_GMAC)
+	u32 reserved1[3];
+	u16 pkt_len;
+	u16 reserved_16;
+	u32 reserved2[6];
+	u32 pkt_err;
+	u32 reserved3[5];
+#else
 	u16 reserved_16;
 	u16 pkt_len;
 	u32 reserve1[3];
 	u32 pkt_err;
 	u32 reserve2[4];
+#endif
 };
 
 struct hip04_priv {
 	void __iomem *base;
-	int phy_mode;
+#if defined(CONFIG_HI13X1_GMAC)
+	void __iomem *sysctrl_base;
+#endif
+	phy_interface_t phy_mode;
 	int chan;
 	unsigned int port;
+	unsigned int group;
 	unsigned int speed;
 	unsigned int duplex;
 	unsigned int reg_inten;
 
 	struct napi_struct napi;
+	struct device *dev;
 	struct net_device *ndev;
 
 	struct tx_desc *tx_desc;
@@ -173,6 +237,7 @@ struct hip04_priv {
 	dma_addr_t rx_phys[RX_DESC_NUM];
 	unsigned int rx_head;
 	unsigned int rx_buf_size;
+	unsigned int rx_cnt_remaining;
 
 	struct device_node *phy_node;
 	struct phy_device *phy;
@@ -185,7 +250,7 @@ struct hip04_priv {
 
 static inline unsigned int tx_count(unsigned int head, unsigned int tail)
 {
-	return (head - tail) % (TX_DESC_NUM - 1);
+	return (head - tail) % TX_DESC_NUM;
 }
 
 static void hip04_config_port(struct net_device *ndev, u32 speed, u32 duplex)
@@ -225,6 +290,13 @@ static void hip04_config_port(struct net_device *ndev, u32 speed, u32 duplex)
 	writel_relaxed(val, priv->base + GE_MODE_CHANGE_REG);
 }
 
+static void hip04_reset_dreq(struct hip04_priv *priv)
+{
+#if defined(CONFIG_HI13X1_GMAC)
+	writel_relaxed(RESET_DREQ_ALL, priv->sysctrl_base + SC_PPE_RESET_DREQ);
+#endif
+}
+
 static void hip04_reset_ppe(struct hip04_priv *priv)
 {
 	u32 val, tmp, timeout = 0;
@@ -245,14 +317,14 @@ static void hip04_config_fifo(struct hip04_priv *priv)
 	val |= PPE_CFG_STS_RX_PKT_CNT_RC;
 	writel_relaxed(val, priv->base + PPE_CFG_STS_MODE);
 
-	val = BIT(priv->port);
+	val = BIT(priv->group);
 	regmap_write(priv->map, priv->port * 4 + PPE_CFG_POOL_GRP, val);
 
-	val = priv->port << PPE_CFG_QOS_VMID_GRP_SHIFT;
+	val = priv->group << PPE_CFG_QOS_VMID_GRP_SHIFT;
 	val |= PPE_CFG_QOS_VMID_MODE;
 	writel_relaxed(val, priv->base + PPE_CFG_QOS_VMID_GEN);
 
-	val = RX_BUF_SIZE;
+	val = RX_BUF_SIZE >> PPE_BUF_SIZE_SHIFT;
 	regmap_write(priv->map, priv->port * 4 + PPE_CFG_RX_BUF_SIZE, val);
 
 	val = RX_DESC_NUM << PPE_CFG_RX_DEPTH_SHIFT;
@@ -289,8 +361,10 @@ static void hip04_config_fifo(struct hip04_priv *priv)
 	val |= GE_RX_STRIP_PAD | GE_RX_PAD_EN;
 	writel_relaxed(val, priv->base + GE_RECV_CONTROL_REG);
 
+#ifndef CONFIG_HI13X1_GMAC
 	val = GE_AUTO_NEG_CTL;
 	writel_relaxed(val, priv->base + GE_TX_LOCAL_PAGE_REG);
+#endif
 }
 
 static void hip04_mac_enable(struct net_device *ndev)
@@ -333,12 +407,18 @@ static void hip04_mac_disable(struct net_device *ndev)
 
 static void hip04_set_xmit_desc(struct hip04_priv *priv, dma_addr_t phys)
 {
-	writel(phys, priv->base + PPE_CFG_CPU_ADD_ADDR);
+	u32 val;
+
+	val = phys >> PPE_BUF_SIZE_SHIFT | PPE_TX_BUF_HOLD;
+	writel(val, priv->base + PPE_CFG_CPU_ADD_ADDR);
 }
 
 static void hip04_set_recv_desc(struct hip04_priv *priv, dma_addr_t phys)
 {
-	regmap_write(priv->map, priv->port * 4 + PPE_CFG_RX_ADDR, phys);
+	u32 val;
+
+	val = phys >> PPE_BUF_SIZE_SHIFT;
+	regmap_write(priv->map, priv->port * 4 + PPE_CFG_RX_ADDR, val);
 }
 
 static u32 hip04_recv_cnt(struct hip04_priv *priv)
@@ -373,7 +453,7 @@ static int hip04_tx_reclaim(struct net_device *ndev, bool force)
 	unsigned int count;
 
 	smp_rmb();
-	count = tx_count(ACCESS_ONCE(priv->tx_head), tx_tail);
+	count = tx_count(READ_ONCE(priv->tx_head), tx_tail);
 	if (count == 0)
 		goto out;
 
@@ -387,7 +467,7 @@ static int hip04_tx_reclaim(struct net_device *ndev, bool force)
 		}
 
 		if (priv->tx_phys[tx_tail]) {
-			dma_unmap_single(&ndev->dev, priv->tx_phys[tx_tail],
+			dma_unmap_single(priv->dev, priv->tx_phys[tx_tail],
 					 priv->tx_skb[tx_tail]->len,
 					 DMA_TO_DEVICE);
 			priv->tx_phys[tx_tail] = 0;
@@ -422,7 +502,8 @@ static void hip04_start_tx_timer(struct hip04_priv *priv)
 			       ns, HRTIMER_MODE_REL);
 }
 
-static int hip04_mac_start_xmit(struct sk_buff *skb, struct net_device *ndev)
+static netdev_tx_t
+hip04_mac_start_xmit(struct sk_buff *skb, struct net_device *ndev)
 {
 	struct hip04_priv *priv = netdev_priv(ndev);
 	struct net_device_stats *stats = &ndev->stats;
@@ -431,31 +512,40 @@ static int hip04_mac_start_xmit(struct sk_buff *skb, struct net_device *ndev)
 	dma_addr_t phys;
 
 	smp_rmb();
-	count = tx_count(tx_head, ACCESS_ONCE(priv->tx_tail));
+	count = tx_count(tx_head, READ_ONCE(priv->tx_tail));
 	if (count == (TX_DESC_NUM - 1)) {
 		netif_stop_queue(ndev);
 		return NETDEV_TX_BUSY;
 	}
 
-	phys = dma_map_single(&ndev->dev, skb->data, skb->len, DMA_TO_DEVICE);
-	if (dma_mapping_error(&ndev->dev, phys)) {
+	phys = dma_map_single(priv->dev, skb->data, skb->len, DMA_TO_DEVICE);
+	if (dma_mapping_error(priv->dev, phys)) {
 		dev_kfree_skb(skb);
 		return NETDEV_TX_OK;
 	}
 
 	priv->tx_skb[tx_head] = skb;
 	priv->tx_phys[tx_head] = phys;
-	desc->send_addr = cpu_to_be32(phys);
-	desc->send_size = cpu_to_be32(skb->len);
-	desc->cfg = cpu_to_be32(TX_CLEAR_WB | TX_FINISH_CACHE_INV);
+
+	desc->send_size = (__force u32)cpu_to_be32(skb->len);
+#if defined(CONFIG_HI13X1_GMAC)
+	desc->cfg = (__force u32)cpu_to_be32(TX_CLEAR_WB | TX_FINISH_CACHE_INV
+		| TX_RELEASE_TO_PPE | priv->port << TX_POOL_SHIFT);
+	desc->data_offset = (__force u32)cpu_to_be32(phys & SOC_CACHE_LINE_MASK);
+	desc->send_addr =  (__force u32)cpu_to_be32(phys & ~SOC_CACHE_LINE_MASK);
+#else
+	desc->cfg = (__force u32)cpu_to_be32(TX_CLEAR_WB | TX_FINISH_CACHE_INV);
+	desc->send_addr = (__force u32)cpu_to_be32(phys);
+#endif
 	phys = priv->tx_desc_dma + tx_head * sizeof(struct tx_desc);
-	desc->wb_addr = cpu_to_be32(phys);
+	desc->wb_addr = (__force u32)cpu_to_be32(phys +
+		offsetof(struct tx_desc, send_addr));
 	skb_tx_timestamp(skb);
 
 	hip04_set_xmit_desc(priv, phys);
-	priv->tx_head = TX_NEXT(tx_head);
 	count++;
 	netdev_sent_queue(ndev, skb->len);
+	priv->tx_head = TX_NEXT(tx_head);
 
 	stats->tx_bytes += skb->len;
 	stats->tx_packets++;
@@ -486,7 +576,6 @@ static int hip04_rx_poll(struct napi_struct *napi, int budget)
 	struct hip04_priv *priv = container_of(napi, struct hip04_priv, napi);
 	struct net_device *ndev = priv->ndev;
 	struct net_device_stats *stats = &ndev->stats;
-	unsigned int cnt = hip04_recv_cnt(priv);
 	struct rx_desc *desc;
 	struct sk_buff *skb;
 	unsigned char *buf;
@@ -497,7 +586,10 @@ static int hip04_rx_poll(struct napi_struct *napi, int budget)
 	u16 len;
 	u32 err;
 
-	while (cnt && !last) {
+	/* clean up tx descriptors */
+	tx_remaining = hip04_tx_reclaim(ndev, false);
+	priv->rx_cnt_remaining += hip04_recv_cnt(priv);
+	while (priv->rx_cnt_remaining && !last) {
 		buf = priv->rx_buf[priv->rx_head];
 		skb = build_skb(buf, priv->rx_buf_size);
 		if (unlikely(!skb)) {
@@ -505,13 +597,13 @@ static int hip04_rx_poll(struct napi_struct *napi, int budget)
 			goto refill;
 		}
 
-		dma_unmap_single(&ndev->dev, priv->rx_phys[priv->rx_head],
+		dma_unmap_single(priv->dev, priv->rx_phys[priv->rx_head],
 				 RX_BUF_SIZE, DMA_FROM_DEVICE);
 		priv->rx_phys[priv->rx_head] = 0;
 
 		desc = (struct rx_desc *)skb->data;
-		len = be16_to_cpu(desc->pkt_len);
-		err = be32_to_cpu(desc->pkt_err);
+		len = be16_to_cpu((__force __be16)desc->pkt_len);
+		err = be32_to_cpu((__force __be32)desc->pkt_err);
 
 		if (0 == len) {
 			dev_kfree_skb_any(skb);
@@ -534,20 +626,22 @@ refill:
 		buf = netdev_alloc_frag(priv->rx_buf_size);
 		if (!buf)
 			goto done;
-		phys = dma_map_single(&ndev->dev, buf,
+		phys = dma_map_single(priv->dev, buf,
 				      RX_BUF_SIZE, DMA_FROM_DEVICE);
-		if (dma_mapping_error(&ndev->dev, phys))
+		if (dma_mapping_error(priv->dev, phys))
 			goto done;
 		priv->rx_buf[priv->rx_head] = buf;
 		priv->rx_phys[priv->rx_head] = phys;
 		hip04_set_recv_desc(priv, phys);
 
 		priv->rx_head = RX_NEXT(priv->rx_head);
-		if (rx >= budget)
+		if (rx >= budget) {
+			--priv->rx_cnt_remaining;
 			goto done;
+		}
 
-		if (--cnt == 0)
-			cnt = hip04_recv_cnt(priv);
+		if (--priv->rx_cnt_remaining == 0)
+			priv->rx_cnt_remaining += hip04_recv_cnt(priv);
 	}
 
 	if (!(priv->reg_inten & RCV_INT)) {
@@ -557,8 +651,7 @@ refill:
 	}
 	napi_complete_done(napi, rx);
 done:
-	/* clean up tx descriptors and start a new timer if necessary */
-	tx_remaining = hip04_tx_reclaim(ndev, false);
+	/* start a new timer if necessary */
 	if (rx < budget && tx_remaining)
 		hip04_start_tx_timer(priv);
 
@@ -633,6 +726,7 @@ static int hip04_mac_open(struct net_device *ndev)
 	int i;
 
 	priv->rx_head = 0;
+	priv->rx_cnt_remaining = 0;
 	priv->tx_head = 0;
 	priv->tx_tail = 0;
 	hip04_reset_ppe(priv);
@@ -640,9 +734,9 @@ static int hip04_mac_open(struct net_device *ndev)
 	for (i = 0; i < RX_DESC_NUM; i++) {
 		dma_addr_t phys;
 
-		phys = dma_map_single(&ndev->dev, priv->rx_buf[i],
+		phys = dma_map_single(priv->dev, priv->rx_buf[i],
 				      RX_BUF_SIZE, DMA_FROM_DEVICE);
-		if (dma_mapping_error(&ndev->dev, phys))
+		if (dma_mapping_error(priv->dev, phys))
 			return -EIO;
 
 		priv->rx_phys[i] = phys;
@@ -676,7 +770,7 @@ static int hip04_mac_stop(struct net_device *ndev)
 
 	for (i = 0; i < RX_DESC_NUM; i++) {
 		if (priv->rx_phys[i]) {
-			dma_unmap_single(&ndev->dev, priv->rx_phys[i],
+			dma_unmap_single(priv->dev, priv->rx_phys[i],
 					 RX_BUF_SIZE, DMA_FROM_DEVICE);
 			priv->rx_phys[i] = 0;
 		}
@@ -685,7 +779,7 @@ static int hip04_mac_stop(struct net_device *ndev)
 	return 0;
 }
 
-static void hip04_timeout(struct net_device *ndev)
+static void hip04_timeout(struct net_device *ndev, unsigned int txqueue)
 {
 	struct hip04_priv *priv = netdev_priv(ndev);
 
@@ -702,7 +796,9 @@ static void hip04_tx_timeout_task(struct work_struct *work)
 }
 
 static int hip04_get_coalesce(struct net_device *netdev,
-			      struct ethtool_coalesce *ec)
+			      struct ethtool_coalesce *ec,
+			      struct kernel_ethtool_coalesce *kernel_coal,
+			      struct netlink_ext_ack *extack)
 {
 	struct hip04_priv *priv = netdev_priv(netdev);
 
@@ -713,24 +809,12 @@ static int hip04_get_coalesce(struct net_device *netdev,
 }
 
 static int hip04_set_coalesce(struct net_device *netdev,
-			      struct ethtool_coalesce *ec)
+			      struct ethtool_coalesce *ec,
+			      struct kernel_ethtool_coalesce *kernel_coal,
+			      struct netlink_ext_ack *extack)
 {
 	struct hip04_priv *priv = netdev_priv(netdev);
 
-	/* Check not supported parameters  */
-	if ((ec->rx_max_coalesced_frames) || (ec->rx_coalesce_usecs_irq) ||
-	    (ec->rx_max_coalesced_frames_irq) || (ec->tx_coalesce_usecs_irq) ||
-	    (ec->use_adaptive_rx_coalesce) || (ec->use_adaptive_tx_coalesce) ||
-	    (ec->pkt_rate_low) || (ec->rx_coalesce_usecs_low) ||
-	    (ec->rx_max_coalesced_frames_low) || (ec->tx_coalesce_usecs_high) ||
-	    (ec->tx_max_coalesced_frames_low) || (ec->pkt_rate_high) ||
-	    (ec->tx_coalesce_usecs_low) || (ec->rx_coalesce_usecs_high) ||
-	    (ec->rx_max_coalesced_frames_high) || (ec->rx_coalesce_usecs) ||
-	    (ec->tx_max_coalesced_frames_irq) ||
-	    (ec->stats_block_coalesce_usecs) ||
-	    (ec->tx_max_coalesced_frames_high) || (ec->rate_sample_interval))
-		return -EOPNOTSUPP;
-
 	if ((ec->tx_coalesce_usecs > HIP04_MAX_TX_COALESCE_USECS ||
 	     ec->tx_coalesce_usecs < HIP04_MIN_TX_COALESCE_USECS) ||
 	    (ec->tx_max_coalesced_frames > HIP04_MAX_TX_COALESCE_FRAMES ||
@@ -746,11 +830,13 @@ static int hip04_set_coalesce(struct net_device *netdev,
 static void hip04_get_drvinfo(struct net_device *netdev,
 			      struct ethtool_drvinfo *drvinfo)
 {
-	strlcpy(drvinfo->driver, DRV_NAME, sizeof(drvinfo->driver));
-	strlcpy(drvinfo->version, DRV_VERSION, sizeof(drvinfo->version));
+	strscpy(drvinfo->driver, DRV_NAME, sizeof(drvinfo->driver));
+	strscpy(drvinfo->version, DRV_VERSION, sizeof(drvinfo->version));
 }
 
 static const struct ethtool_ops hip04_ethtool_ops = {
+	.supported_coalesce_params = ETHTOOL_COALESCE_TX_USECS |
+				     ETHTOOL_COALESCE_TX_MAX_FRAMES,
 	.get_coalesce		= hip04_get_coalesce,
 	.set_coalesce		= hip04_set_coalesce,
 	.get_drvinfo		= hip04_get_drvinfo,
@@ -811,7 +897,6 @@ static int hip04_mac_probe(struct platform_device *pdev)
 	struct of_phandle_args arg;
 	struct net_device *ndev;
 	struct hip04_priv *priv;
-	struct resource *res;
 	int irq;
 	int ret;
 
@@ -820,18 +905,26 @@ static int hip04_mac_probe(struct platform_device *pdev)
 		return -ENOMEM;
 
 	priv = netdev_priv(ndev);
+	priv->dev = d;
 	priv->ndev = ndev;
 	platform_set_drvdata(pdev, ndev);
 	SET_NETDEV_DEV(ndev, &pdev->dev);
 
-	res = platform_get_resource(pdev, IORESOURCE_MEM, 0);
-	priv->base = devm_ioremap_resource(d, res);
+	priv->base = devm_platform_ioremap_resource(pdev, 0);
 	if (IS_ERR(priv->base)) {
 		ret = PTR_ERR(priv->base);
 		goto init_fail;
 	}
 
-	ret = of_parse_phandle_with_fixed_args(node, "port-handle", 2, 0, &arg);
+#if defined(CONFIG_HI13X1_GMAC)
+	priv->sysctrl_base = devm_platform_ioremap_resource(pdev, 1);
+	if (IS_ERR(priv->sysctrl_base)) {
+		ret = PTR_ERR(priv->sysctrl_base);
+		goto init_fail;
+	}
+#endif
+
+	ret = of_parse_phandle_with_fixed_args(node, "port-handle", 3, 0, &arg);
 	if (ret < 0) {
 		dev_warn(d, "no port-handle\n");
 		goto init_fail;
@@ -839,8 +932,7 @@ static int hip04_mac_probe(struct platform_device *pdev)
 
 	priv->port = arg.args[0];
 	priv->chan = arg.args[1] * RX_DESC_NUM;
-
-	hrtimer_init(&priv->tx_coalesce_timer, CLOCK_MONOTONIC, HRTIMER_MODE_REL);
+	priv->group = arg.args[2];
 
 	/* BQL will try to keep the TX queue as short as possible, but it can't
 	 * be faster than tx_coalesce_usecs, so we need a fast timeout here,
@@ -850,25 +942,25 @@ static int hip04_mac_probe(struct platform_device *pdev)
 	 */
 	priv->tx_coalesce_frames = TX_DESC_NUM * 3 / 4;
 	priv->tx_coalesce_usecs = 200;
-	priv->tx_coalesce_timer.function = tx_done;
+	hrtimer_setup(&priv->tx_coalesce_timer, tx_done, CLOCK_MONOTONIC, HRTIMER_MODE_REL);
 
 	priv->map = syscon_node_to_regmap(arg.np);
+	of_node_put(arg.np);
 	if (IS_ERR(priv->map)) {
 		dev_warn(d, "no syscon hisilicon,hip04-ppe\n");
 		ret = PTR_ERR(priv->map);
 		goto init_fail;
 	}
 
-	priv->phy_mode = of_get_phy_mode(node);
-	if (priv->phy_mode < 0) {
+	ret = of_get_phy_mode(node, &priv->phy_mode);
+	if (ret) {
 		dev_warn(d, "not find phy-mode\n");
-		ret = -EINVAL;
 		goto init_fail;
 	}
 
 	irq = platform_get_irq(pdev, 0);
-	if (irq <= 0) {
-		ret = -EINVAL;
+	if (irq < 0) {
+		ret = irq;
 		goto init_fail;
 	}
 
@@ -897,14 +989,15 @@ static int hip04_mac_probe(struct platform_device *pdev)
 	ndev->watchdog_timeo = TX_TIMEOUT;
 	ndev->priv_flags |= IFF_UNICAST_FLT;
 	ndev->irq = irq;
-	netif_napi_add(ndev, &priv->napi, hip04_rx_poll, NAPI_POLL_WEIGHT);
+	netif_napi_add(ndev, &priv->napi, hip04_rx_poll);
 
+	hip04_reset_dreq(priv);
 	hip04_reset_ppe(priv);
 	if (priv->phy_mode == PHY_INTERFACE_MODE_MII)
 		hip04_config_port(ndev, SPEED_100, DUPLEX_FULL);
 
 	hip04_config_fifo(priv);
-	random_ether_addr(ndev->dev_addr);
+	eth_hw_addr_random(ndev);
 	hip04_update_mac_address(ndev);
 
 	ret = hip04_alloc_ring(ndev, d);
@@ -914,10 +1007,8 @@ static int hip04_mac_probe(struct platform_device *pdev)
 	}
 
 	ret = register_netdev(ndev);
-	if (ret) {
-		free_netdev(ndev);
+	if (ret)
 		goto alloc_fail;
-	}
 
 	return 0;
 
@@ -929,7 +1020,7 @@ init_fail:
 	return ret;
 }
 
-static int hip04_remove(struct platform_device *pdev)
+static void hip04_remove(struct platform_device *pdev)
 {
 	struct net_device *ndev = platform_get_drvdata(pdev);
 	struct hip04_priv *priv = netdev_priv(ndev);
@@ -940,12 +1031,9 @@ static int hip04_remove(struct platform_device *pdev)
 
 	hip04_free_ring(ndev, d);
 	unregister_netdev(ndev);
-	free_irq(ndev->irq, ndev);
 	of_node_put(priv->phy_node);
 	cancel_work_sync(&priv->tx_timeout_task);
 	free_netdev(ndev);
-
-	return 0;
 }
 
 static const struct of_device_id hip04_mac_match[] = {
@@ -957,7 +1045,7 @@ MODULE_DEVICE_TABLE(of, hip04_mac_match);
 
 static struct platform_driver hip04_mac_driver = {
 	.probe	= hip04_mac_probe,
-	.remove	= hip04_remove,
+	.remove = hip04_remove,
 	.driver	= {
 		.name		= DRV_NAME,
 		.of_match_table	= hip04_mac_match,
diff --git a/drivers/net/ethernet/hisilicon/hisi_femac.c b/drivers/net/ethernet/hisilicon/hisi_femac.c
index 2c2808830e95..d244a40df430 100644
--- a/drivers/net/ethernet/hisilicon/hisi_femac.c
+++ b/drivers/net/ethernet/hisilicon/hisi_femac.c
@@ -1,20 +1,8 @@
+// SPDX-License-Identifier: GPL-2.0-or-later
 /*
  * Hisilicon Fast Ethernet MAC Driver
  *
  * Copyright (c) 2016 HiSilicon Technologies Co., Ltd.
- *
- * This program is free software; you can redistribute it and/or modify
- * it under the terms of the GNU General Public License as published by
- * the Free Software Foundation; either version 2 of the License, or
- * (at your option) any later version.
- *
- * This program is distributed in the hope that it will be useful,
- * but WITHOUT ANY WARRANTY; without even the implied warranty of
- * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
- * GNU General Public License for more details.
- *
- * You should have received a copy of the GNU General Public License
- * along with this program. If not, see <http://www.gnu.org/licenses/>.
  */
 
 #include <linux/circ_buf.h>
@@ -295,7 +283,7 @@ static int hisi_femac_rx(struct net_device *dev, int limit)
 		skb->protocol = eth_type_trans(skb, dev);
 		napi_gro_receive(&priv->napi, skb);
 		dev->stats.rx_packets++;
-		dev->stats.rx_bytes += skb->len;
+		dev->stats.rx_bytes += len;
 next:
 		pos = (pos + 1) % rxq->num;
 		if (rx_pkts_num >= limit)
@@ -439,7 +427,7 @@ static void hisi_femac_free_skb_rings(struct hisi_femac_priv *priv)
 }
 
 static int hisi_femac_set_hw_mac_addr(struct hisi_femac_priv *priv,
-				      unsigned char *mac)
+				      const unsigned char *mac)
 {
 	u32 reg;
 
@@ -567,7 +555,7 @@ static int hisi_femac_set_mac_address(struct net_device *dev, void *p)
 	if (!is_valid_ether_addr(skaddr->sa_data))
 		return -EADDRNOTAVAIL;
 
-	memcpy(dev->dev_addr, skaddr->sa_data, dev->addr_len);
+	eth_hw_addr_set(dev, skaddr->sa_data);
 	dev->addr_assign_type &= ~NET_ADDR_RANDOM;
 
 	hisi_femac_set_hw_mac_addr(priv, dev->dev_addr);
@@ -687,18 +675,6 @@ static void hisi_femac_net_set_rx_mode(struct net_device *dev)
 	}
 }
 
-static int hisi_femac_net_ioctl(struct net_device *dev,
-				struct ifreq *ifreq, int cmd)
-{
-	if (!netif_running(dev))
-		return -EINVAL;
-
-	if (!dev->phydev)
-		return -EINVAL;
-
-	return phy_mii_ioctl(dev->phydev, ifreq, cmd);
-}
-
 static const struct ethtool_ops hisi_femac_ethtools_ops = {
 	.get_link		= ethtool_op_get_link,
 	.get_link_ksettings	= phy_ethtool_get_link_ksettings,
@@ -709,7 +685,7 @@ static const struct net_device_ops hisi_femac_netdev_ops = {
 	.ndo_open		= hisi_femac_net_open,
 	.ndo_stop		= hisi_femac_net_close,
 	.ndo_start_xmit		= hisi_femac_net_xmit,
-	.ndo_do_ioctl		= hisi_femac_net_ioctl,
+	.ndo_eth_ioctl		= phy_do_ioctl_running,
 	.ndo_set_mac_address	= hisi_femac_set_mac_address,
 	.ndo_set_rx_mode	= hisi_femac_net_set_rx_mode,
 };
@@ -793,11 +769,9 @@ static int hisi_femac_drv_probe(struct platform_device *pdev)
 {
 	struct device *dev = &pdev->dev;
 	struct device_node *node = dev->of_node;
-	struct resource *res;
 	struct net_device *ndev;
 	struct hisi_femac_priv *priv;
 	struct phy_device *phy;
-	const char *mac_addr;
 	int ret;
 
 	ndev = alloc_etherdev(sizeof(*priv));
@@ -811,15 +785,13 @@ static int hisi_femac_drv_probe(struct platform_device *pdev)
 	priv->dev = dev;
 	priv->ndev = ndev;
 
-	res = platform_get_resource(pdev, IORESOURCE_MEM, 0);
-	priv->port_base = devm_ioremap_resource(dev, res);
+	priv->port_base = devm_platform_ioremap_resource(pdev, 0);
 	if (IS_ERR(priv->port_base)) {
 		ret = PTR_ERR(priv->port_base);
 		goto out_free_netdev;
 	}
 
-	res = platform_get_resource(pdev, IORESOURCE_MEM, 1);
-	priv->glb_base = devm_ioremap_resource(dev, res);
+	priv->glb_base = devm_platform_ioremap_resource(pdev, 1);
 	if (IS_ERR(priv->glb_base)) {
 		ret = PTR_ERR(priv->glb_base);
 		goto out_free_netdev;
@@ -869,10 +841,8 @@ static int hisi_femac_drv_probe(struct platform_device *pdev)
 			   (unsigned long)phy->phy_id,
 			   phy_modes(phy->interface));
 
-	mac_addr = of_get_mac_address(node);
-	if (mac_addr)
-		ether_addr_copy(ndev->dev_addr, mac_addr);
-	if (!is_valid_ether_addr(ndev->dev_addr)) {
+	ret = of_get_ethdev_address(node, ndev);
+	if (ret) {
 		eth_hw_addr_random(ndev);
 		dev_warn(dev, "using random MAC address %pM\n",
 			 ndev->dev_addr);
@@ -882,7 +852,8 @@ static int hisi_femac_drv_probe(struct platform_device *pdev)
 	ndev->priv_flags |= IFF_UNICAST_FLT;
 	ndev->netdev_ops = &hisi_femac_netdev_ops;
 	ndev->ethtool_ops = &hisi_femac_ethtools_ops;
-	netif_napi_add(ndev, &priv->napi, hisi_femac_poll, FEMAC_POLL_WEIGHT);
+	netif_napi_add_weight(ndev, &priv->napi, hisi_femac_poll,
+			      FEMAC_POLL_WEIGHT);
 
 	hisi_femac_port_init(priv);
 
@@ -891,9 +862,8 @@ static int hisi_femac_drv_probe(struct platform_device *pdev)
 		goto out_disconnect_phy;
 
 	ndev->irq = platform_get_irq(pdev, 0);
-	if (ndev->irq <= 0) {
-		dev_err(dev, "No irq resource\n");
-		ret = -ENODEV;
+	if (ndev->irq < 0) {
+		ret = ndev->irq;
 		goto out_disconnect_phy;
 	}
 
@@ -923,7 +893,7 @@ out_free_netdev:
 	return ret;
 }
 
-static int hisi_femac_drv_remove(struct platform_device *pdev)
+static void hisi_femac_drv_remove(struct platform_device *pdev)
 {
 	struct net_device *ndev = platform_get_drvdata(pdev);
 	struct hisi_femac_priv *priv = netdev_priv(ndev);
@@ -934,8 +904,6 @@ static int hisi_femac_drv_remove(struct platform_device *pdev)
 	phy_disconnect(ndev->phydev);
 	clk_disable_unprepare(priv->clk);
 	free_netdev(ndev);
-
-	return 0;
 }
 
 #ifdef CONFIG_PM
diff --git a/drivers/net/ethernet/hisilicon/hix5hd2_gmac.c b/drivers/net/ethernet/hisilicon/hix5hd2_gmac.c
index 25a6c8722eca..e3e7f2270560 100644
--- a/drivers/net/ethernet/hisilicon/hix5hd2_gmac.c
+++ b/drivers/net/ethernet/hisilicon/hix5hd2_gmac.c
@@ -1,17 +1,14 @@
+// SPDX-License-Identifier: GPL-2.0-or-later
 /* Copyright (c) 2014 Linaro Ltd.
  * Copyright (c) 2014 Hisilicon Limited.
- *
- * This program is free software; you can redistribute it and/or modify
- * it under the terms of the GNU General Public License as published by
- * the Free Software Foundation; either version 2 of the License, or
- * (at your option) any later version.
  */
 
 #include <linux/module.h>
 #include <linux/interrupt.h>
 #include <linux/etherdevice.h>
 #include <linux/platform_device.h>
-#include <linux/of_device.h>
+#include <linux/property.h>
+#include <linux/of.h>
 #include <linux/of_net.h>
 #include <linux/of_mdio.h>
 #include <linux/reset.h>
@@ -433,7 +430,7 @@ static void hix5hd2_port_disable(struct hix5hd2_priv *priv)
 static void hix5hd2_hw_set_mac_addr(struct net_device *dev)
 {
 	struct hix5hd2_priv *priv = netdev_priv(dev);
-	unsigned char *mac = dev->dev_addr;
+	const unsigned char *mac = dev->dev_addr;
 	u32 val;
 
 	val = mac[1] | (mac[0] << 8);
@@ -554,7 +551,7 @@ static int hix5hd2_rx(struct net_device *dev, int limit)
 		skb->protocol = eth_type_trans(skb, dev);
 		napi_gro_receive(&priv->napi, skb);
 		dev->stats.rx_packets++;
-		dev->stats.rx_bytes += skb->len;
+		dev->stats.rx_bytes += len;
 next:
 		pos = dma_ring_incr(pos, RX_DESC_NUM);
 	}
@@ -723,7 +720,7 @@ static int hix5hd2_fill_sg_desc(struct hix5hd2_priv *priv,
 
 	for (i = 0; i < skb_shinfo(skb)->nr_frags; i++) {
 		skb_frag_t *frag = &skb_shinfo(skb)->frags[i];
-		int len = frag->size;
+		int len = skb_frag_size(frag);
 
 		addr = skb_frag_dma_map(priv->dev, frag, 0, len, DMA_TO_DEVICE);
 		ret = dma_mapping_error(priv->dev, addr);
@@ -736,7 +733,7 @@ static int hix5hd2_fill_sg_desc(struct hix5hd2_priv *priv,
 	return 0;
 }
 
-static int hix5hd2_net_xmit(struct sk_buff *skb, struct net_device *dev)
+static netdev_tx_t hix5hd2_net_xmit(struct sk_buff *skb, struct net_device *dev)
 {
 	struct hix5hd2_priv *priv = netdev_priv(dev);
 	struct hix5hd2_desc *desc;
@@ -897,7 +894,7 @@ static void hix5hd2_tx_timeout_task(struct work_struct *work)
 	hix5hd2_net_open(priv->netdev);
 }
 
-static void hix5hd2_net_timeout(struct net_device *dev)
+static void hix5hd2_net_timeout(struct net_device *dev, unsigned int txqueue)
 {
 	struct hix5hd2_priv *priv = netdev_priv(dev);
 
@@ -1011,7 +1008,6 @@ static int hix5hd2_init_hw_desc_queue(struct hix5hd2_priv *priv)
 		if (virt_addr == NULL)
 			goto error_free_pool;
 
-		memset(virt_addr, 0, size);
 		priv->pool[i].size = size;
 		priv->pool[i].desc = virt_addr;
 		priv->pool[i].phys_addr = phys_addr;
@@ -1029,9 +1025,9 @@ static int hix5hd2_init_sg_desc_queue(struct hix5hd2_priv *priv)
 	struct sg_desc *desc;
 	dma_addr_t phys_addr;
 
-	desc = (struct sg_desc *)dma_alloc_coherent(priv->dev,
-				TX_DESC_NUM * sizeof(struct sg_desc),
-				&phys_addr, GFP_KERNEL);
+	desc = dma_alloc_coherent(priv->dev,
+				  TX_DESC_NUM * sizeof(struct sg_desc),
+				  &phys_addr, GFP_KERNEL);
 	if (!desc)
 		return -ENOMEM;
 
@@ -1099,12 +1095,9 @@ static int hix5hd2_dev_probe(struct platform_device *pdev)
 {
 	struct device *dev = &pdev->dev;
 	struct device_node *node = dev->of_node;
-	const struct of_device_id *of_id = NULL;
 	struct net_device *ndev;
 	struct hix5hd2_priv *priv;
-	struct resource *res;
 	struct mii_bus *bus;
-	const char *mac_addr;
 	int ret;
 
 	ndev = alloc_etherdev(sizeof(struct hix5hd2_priv));
@@ -1117,22 +1110,15 @@ static int hix5hd2_dev_probe(struct platform_device *pdev)
 	priv->dev = dev;
 	priv->netdev = ndev;
 
-	of_id = of_match_device(hix5hd2_of_match, dev);
-	if (!of_id) {
-		ret = -EINVAL;
-		goto out_free_netdev;
-	}
-	priv->hw_cap = (unsigned long)of_id->data;
+	priv->hw_cap = (unsigned long)device_get_match_data(dev);
 
-	res = platform_get_resource(pdev, IORESOURCE_MEM, 0);
-	priv->base = devm_ioremap_resource(dev, res);
+	priv->base = devm_platform_ioremap_resource(pdev, 0);
 	if (IS_ERR(priv->base)) {
 		ret = PTR_ERR(priv->base);
 		goto out_free_netdev;
 	}
 
-	res = platform_get_resource(pdev, IORESOURCE_MEM, 1);
-	priv->ctrl_base = devm_ioremap_resource(dev, res);
+	priv->ctrl_base = devm_platform_ioremap_resource(pdev, 1);
 	if (IS_ERR(priv->ctrl_base)) {
 		ret = PTR_ERR(priv->ctrl_base);
 		goto out_free_netdev;
@@ -1201,10 +1187,9 @@ static int hix5hd2_dev_probe(struct platform_device *pdev)
 	if (ret)
 		goto err_free_mdio;
 
-	priv->phy_mode = of_get_phy_mode(node);
-	if (priv->phy_mode < 0) {
+	ret = of_get_phy_mode(node, &priv->phy_mode);
+	if (ret) {
 		netdev_err(ndev, "not find phy-mode\n");
-		ret = -EINVAL;
 		goto err_mdiobus;
 	}
 
@@ -1216,9 +1201,8 @@ static int hix5hd2_dev_probe(struct platform_device *pdev)
 	}
 
 	ndev->irq = platform_get_irq(pdev, 0);
-	if (ndev->irq <= 0) {
-		netdev_err(ndev, "No irq resource\n");
-		ret = -EINVAL;
+	if (ndev->irq < 0) {
+		ret = ndev->irq;
 		goto out_phy_node;
 	}
 
@@ -1229,10 +1213,8 @@ static int hix5hd2_dev_probe(struct platform_device *pdev)
 		goto out_phy_node;
 	}
 
-	mac_addr = of_get_mac_address(node);
-	if (mac_addr)
-		ether_addr_copy(ndev->dev_addr, mac_addr);
-	if (!is_valid_ether_addr(ndev->dev_addr)) {
+	ret = of_get_ethdev_address(node, ndev);
+	if (ret) {
 		eth_hw_addr_random(ndev);
 		netdev_warn(ndev, "using random MAC address %pM\n",
 			    ndev->dev_addr);
@@ -1255,7 +1237,7 @@ static int hix5hd2_dev_probe(struct platform_device *pdev)
 	if (ret)
 		goto out_phy_node;
 
-	netif_napi_add(ndev, &priv->napi, hix5hd2_poll, NAPI_POLL_WEIGHT);
+	netif_napi_add(ndev, &priv->napi, hix5hd2_poll);
 
 	if (HAS_CAP_TSO(priv->hw_cap)) {
 		ret = hix5hd2_init_sg_desc_queue(priv);
@@ -1295,7 +1277,7 @@ out_free_netdev:
 	return ret;
 }
 
-static int hix5hd2_dev_remove(struct platform_device *pdev)
+static void hix5hd2_dev_remove(struct platform_device *pdev)
 {
 	struct net_device *ndev = platform_get_drvdata(pdev);
 	struct hix5hd2_priv *priv = netdev_priv(ndev);
@@ -1311,8 +1293,6 @@ static int hix5hd2_dev_remove(struct platform_device *pdev)
 	of_node_put(priv->phy_node);
 	cancel_work_sync(&priv->tx_timeout_task);
 	free_netdev(ndev);
-
-	return 0;
 }
 
 static const struct of_device_id hix5hd2_of_match[] = {
diff --git a/drivers/net/ethernet/hisilicon/hns/Makefile b/drivers/net/ethernet/hisilicon/hns/Makefile
index 6010c83e38d8..7aa623b9c82a 100644
--- a/drivers/net/ethernet/hisilicon/hns/Makefile
+++ b/drivers/net/ethernet/hisilicon/hns/Makefile
@@ -1,3 +1,4 @@
+# SPDX-License-Identifier: GPL-2.0
 #
 # Makefile for the HISILICON network device drivers.
 #
diff --git a/drivers/net/ethernet/hisilicon/hns/hnae.c b/drivers/net/ethernet/hisilicon/hns/hnae.c
index 9d9b6e6dd988..d4293f76d69d 100644
--- a/drivers/net/ethernet/hisilicon/hns/hnae.c
+++ b/drivers/net/ethernet/hisilicon/hns/hnae.c
@@ -1,10 +1,6 @@
+// SPDX-License-Identifier: GPL-2.0-or-later
 /*
  * Copyright (c) 2014-2015 Hisilicon Limited.
- *
- * This program is free software; you can redistribute it and/or modify
- * it under the terms of the GNU General Public License as published by
- * the Free Software Foundation; either version 2 of the License, or
- * (at your option) any later version.
  */
 
 #include <linux/dma-mapping.h>
@@ -16,7 +12,9 @@
 
 #define cls_to_ae_dev(dev) container_of(dev, struct hnae_ae_dev, cls_dev)
 
-static struct class *hnae_class;
+static const struct class hnae_class = {
+	.name = "hnae",
+};
 
 static void
 hnae_list_add(spinlock_t *lock, struct list_head *node, struct list_head *head)
@@ -84,7 +82,7 @@ static void hnae_unmap_buffer(struct hnae_ring *ring, struct hnae_desc_cb *cb)
 	if (cb->type == DESC_TYPE_SKB)
 		dma_unmap_single(ring_to_dev(ring), cb->dma, cb->length,
 				 ring_to_dma_dir(ring));
-	else
+	else if (cb->length)
 		dma_unmap_page(ring_to_dev(ring), cb->dma, cb->length,
 			       ring_to_dma_dir(ring));
 }
@@ -115,7 +113,7 @@ static struct hnae_ae_dev *find_ae(const struct fwnode_handle *fwnode)
 
 	WARN_ON(!fwnode);
 
-	dev = class_find_device(hnae_class, NULL, fwnode, __ae_match);
+	dev = class_find_device(&hnae_class, NULL, fwnode, __ae_match);
 
 	return dev ? cls_to_ae_dev(dev) : NULL;
 }
@@ -150,7 +148,6 @@ out_buffer_fail:
 /* free desc along with its attached buffer */
 static void hnae_free_desc(struct hnae_ring *ring)
 {
-	hnae_free_buffers(ring);
 	dma_unmap_single(ring_to_dev(ring), ring->desc_dma_addr,
 			 ring->desc_num * sizeof(ring->desc[0]),
 			 ring_to_dma_dir(ring));
@@ -183,6 +180,9 @@ static int hnae_alloc_desc(struct hnae_ring *ring)
 /* fini ring, also free the buffer for the ring */
 static void hnae_fini_ring(struct hnae_ring *ring)
 {
+	if (is_rx_ring(ring))
+		hnae_free_buffers(ring);
+
 	hnae_free_desc(ring);
 	kfree(ring->desc_cb);
 	ring->desc_cb = NULL;
@@ -201,7 +201,7 @@ hnae_init_ring(struct hnae_queue *q, struct hnae_ring *ring, int flags)
 
 	ring->q = q;
 	ring->flags = flags;
-	spin_lock_init(&ring->lock);
+	ring->coal_param = q->handle->coal_param;
 	assert(!ring->desc && !ring->desc_cb && !ring->desc_dma_addr);
 
 	/* not matter for tx or rx ring, the ntc and ntc start from 0 */
@@ -272,7 +272,7 @@ static void hnae_fini_queue(struct hnae_queue *q)
 	hnae_fini_ring(&q->rx_ring);
 }
 
-/**
+/*
  * ae_chain - define ae chain head
  */
 static RAW_NOTIFIER_HEAD(ae_chain);
@@ -417,14 +417,14 @@ int hnae_ae_register(struct hnae_ae_dev *hdev, struct module *owner)
 	hdev->owner = owner;
 	hdev->id = (int)atomic_inc_return(&id);
 	hdev->cls_dev.parent = hdev->dev;
-	hdev->cls_dev.class = hnae_class;
+	hdev->cls_dev.class = &hnae_class;
 	hdev->cls_dev.release = hnae_release;
 	(void)dev_set_name(&hdev->cls_dev, "hnae%d", hdev->id);
 	ret = device_register(&hdev->cls_dev);
-	if (ret)
+	if (ret) {
+		put_device(&hdev->cls_dev);
 		return ret;
-
-	__module_get(THIS_MODULE);
+	}
 
 	INIT_LIST_HEAD(&hdev->handle_list);
 	spin_lock_init(&hdev->lock);
@@ -440,24 +440,22 @@ EXPORT_SYMBOL(hnae_ae_register);
 
 /**
  * hnae_ae_unregister - unregisters a HNAE AE engine
- * @cdev: the device to unregister
+ * @hdev: the device to unregister
  */
 void hnae_ae_unregister(struct hnae_ae_dev *hdev)
 {
 	device_unregister(&hdev->cls_dev);
-	module_put(THIS_MODULE);
 }
 EXPORT_SYMBOL(hnae_ae_unregister);
 
 static int __init hnae_init(void)
 {
-	hnae_class = class_create(THIS_MODULE, "hnae");
-	return PTR_ERR_OR_ZERO(hnae_class);
+	return class_register(&hnae_class);
 }
 
 static void __exit hnae_exit(void)
 {
-	class_destroy(hnae_class);
+	class_unregister(&hnae_class);
 }
 
 subsys_initcall(hnae_init);
diff --git a/drivers/net/ethernet/hisilicon/hns/hnae.h b/drivers/net/ethernet/hisilicon/hns/hnae.h
index 04211ac73b36..2ae34d01fd36 100644
--- a/drivers/net/ethernet/hisilicon/hns/hnae.h
+++ b/drivers/net/ethernet/hisilicon/hns/hnae.h
@@ -1,10 +1,6 @@
+/* SPDX-License-Identifier: GPL-2.0-or-later */
 /*
  * Copyright (c) 2014-2015 Hisilicon Limited.
- *
- * This program is free software; you can redistribute it and/or modify
- * it under the terms of the GNU General Public License as published by
- * the Free Software Foundation; either version 2 of the License, or
- * (at your option) any later version.
  */
 
 #ifndef __HNAE_H
@@ -87,7 +83,11 @@ do { \
 
 #define HNAE_AE_REGISTER 0x1
 
-#define RCB_RING_NAME_LEN 16
+#define RCB_RING_NAME_LEN (IFNAMSIZ + 4)
+
+#define HNAE_LOWEST_LATENCY_COAL_PARAM	30
+#define HNAE_LOW_LATENCY_COAL_PARAM	80
+#define HNAE_BULK_LATENCY_COAL_PARAM	150
 
 enum hnae_led_state {
 	HNAE_LED_INACTIVE,
@@ -216,10 +216,10 @@ struct hnae_desc_cb {
 
 	/* priv data for the desc, e.g. skb when use with ip stack*/
 	void *priv;
-	u16 page_offset;
-	u16 reuse_flag;
+	u32 page_offset;
+	u32 length;     /* length of the buffer */
 
-	u16 length;     /* length of the buffer */
+	u16 reuse_flag;
 
        /* desc type, used by the ring user to mark the type of the priv data */
 	u16 type;
@@ -274,9 +274,6 @@ struct hnae_ring {
 	/* statistic */
 	struct ring_stats stats;
 
-	/* ring lock for poll one */
-	spinlock_t lock;
-
 	dma_addr_t desc_dma_addr;
 	u32 buf_size;       /* size for hnae_desc->addr, preset by AE */
 	u16 desc_num;       /* total number of desc */
@@ -292,6 +289,12 @@ struct hnae_ring {
 
 	int flags;          /* ring attribute */
 	int irq_init_flag;
+
+	/* total rx bytes after last rx rate calucated */
+	u64 coal_last_rx_bytes;
+	unsigned long coal_last_jiffies;
+	u32 coal_param;
+	u32 coal_rx_rate;	/* rx rate in MB */
 };
 
 #define ring_ptr_move_fw(ring, p) \
@@ -347,7 +350,7 @@ struct hnae_buf_ops {
 };
 
 struct hnae_queue {
-	void __iomem *io_base;
+	u8 __iomem *io_base;
 	phys_addr_t phy_base;
 	struct hnae_ae_dev *dev;	/* the device who use this queue */
 	struct hnae_ring rx_ring ____cacheline_internodealigned_in_smp;
@@ -360,6 +363,7 @@ enum hnae_loop {
 	MAC_INTERNALLOOP_MAC = 0,
 	MAC_INTERNALLOOP_SERDES,
 	MAC_INTERNALLOOP_PHY,
+	MAC_LOOP_PHY_NONE,
 	MAC_LOOP_NONE,
 };
 
@@ -410,10 +414,6 @@ enum hnae_media_type {
  *   get ring bd number limit
  * get_pauseparam()
  *   get tx and rx of pause frame use
- * set_autoneg()
- *   set auto autonegotiation of pause frame use
- * get_autoneg()
- *   get auto autonegotiation of pause frame use
  * set_pauseparam()
  *   set tx and rx of pause frame use
  * get_coalesce_usecs()
@@ -475,14 +475,14 @@ struct hnae_ae_ops {
 			u8 *auto_neg, u16 *speed, u8 *duplex);
 	void (*toggle_ring_irq)(struct hnae_ring *ring, u32 val);
 	void (*adjust_link)(struct hnae_handle *handle, int speed, int duplex);
+	bool (*need_adjust_link)(struct hnae_handle *handle,
+				 int speed, int duplex);
 	int (*set_loopback)(struct hnae_handle *handle,
 			    enum hnae_loop loop_mode, int en);
 	void (*get_ring_bdnum_limit)(struct hnae_queue *queue,
 				     u32 *uplimit);
 	void (*get_pauseparam)(struct hnae_handle *handle,
 			       u32 *auto_neg, u32 *rx_en, u32 *tx_en);
-	int (*set_autoneg)(struct hnae_handle *handle, u8 enable);
-	int (*get_autoneg)(struct hnae_handle *handle);
 	int (*set_pauseparam)(struct hnae_handle *handle,
 			      u32 auto_neg, u32 rx_en, u32 tx_en);
 	void (*get_coalesce_usecs)(struct hnae_handle *handle,
@@ -499,7 +499,7 @@ struct hnae_ae_ops {
 				   u32 *tx_usecs_high, u32 *rx_usecs_high);
 	void (*set_promisc_mode)(struct hnae_handle *handle, u32 en);
 	int (*get_mac_addr)(struct hnae_handle *handle, void **p);
-	int (*set_mac_addr)(struct hnae_handle *handle, void *p);
+	int (*set_mac_addr)(struct hnae_handle *handle, const void *p);
 	int (*add_uc_addr)(struct hnae_handle *handle,
 			   const unsigned char *addr);
 	int (*rm_uc_addr)(struct hnae_handle *handle,
@@ -512,7 +512,7 @@ struct hnae_ae_ops {
 			     struct net_device_stats *net_stats);
 	void (*get_stats)(struct hnae_handle *handle, u64 *data);
 	void (*get_strings)(struct hnae_handle *handle,
-			    u32 stringset, u8 *data);
+			    u32 stringset, u8 **data);
 	int (*get_sset_count)(struct hnae_handle *handle, int stringset);
 	void (*update_led_status)(struct hnae_handle *handle);
 	int (*set_led_id)(struct hnae_handle *handle,
@@ -547,13 +547,18 @@ struct hnae_handle {
 	u32 if_support;
 	int q_num;
 	int vf_id;
+	unsigned long coal_last_jiffies;
+	u32 coal_param;		/* self adapt coalesce param */
+	/* the ring index of last ring that set coal param */
+	u32 coal_ring_idx;
 	u32 eport_id;
 	u32 dport_id;	/* v2 tx bd should fill the dport_id */
+	bool coal_adapt_en;
 	enum hnae_port_type port_type;
 	enum hnae_media_type media_type;
 	struct list_head node;    /* list to hnae_ae_dev->handle_list */
 	struct hnae_buf_ops *bops; /* operation for the buffer */
-	struct hnae_queue **qs;  /* array base of all queues */
+	struct hnae_queue *qs[];  /* flexible array of all queues */
 };
 
 #define ring_to_dev(ring) ((ring)->q->dev->dev)
diff --git a/drivers/net/ethernet/hisilicon/hns/hns_ae_adapt.c b/drivers/net/ethernet/hisilicon/hns/hns_ae_adapt.c
index ff864a187d5a..8ce910f8d0cc 100644
--- a/drivers/net/ethernet/hisilicon/hns/hns_ae_adapt.c
+++ b/drivers/net/ethernet/hisilicon/hns/hns_ae_adapt.c
@@ -1,10 +1,6 @@
+// SPDX-License-Identifier: GPL-2.0-or-later
 /*
  * Copyright (c) 2014-2015 Hisilicon Limited.
- *
- * This program is free software; you can redistribute it and/or modify
- * it under the terms of the GNU General Public License as published by
- * the Free Software Foundation; either version 2 of the License, or
- * (at your option) any later version.
  */
 
 #include <linux/etherdevice.h>
@@ -17,8 +13,6 @@
 #include "hns_dsaf_ppe.h"
 #include "hns_dsaf_rcb.h"
 
-#define AE_NAME_PORT_ID_IDX 6
-
 static struct hns_mac_cb *hns_get_mac_cb(struct hnae_handle *handle)
 {
 	struct  hnae_vf_cb *vf_cb = hns_ae_get_vf_cb(handle);
@@ -70,8 +64,8 @@ static struct ring_pair_cb *hns_ae_get_ring_pair(struct hnae_queue *q)
 	return container_of(q, struct ring_pair_cb, q);
 }
 
-struct hnae_handle *hns_ae_get_handle(struct hnae_ae_dev *dev,
-				      u32 port_id)
+static struct hnae_handle *hns_ae_get_handle(struct hnae_ae_dev *dev,
+					     u32 port_id)
 {
 	int vfnum_per_port;
 	int qnum_per_vf;
@@ -87,8 +81,8 @@ struct hnae_handle *hns_ae_get_handle(struct hnae_ae_dev *dev,
 	vfnum_per_port = hns_ae_get_vf_num_per_port(dsaf_dev, port_id);
 	qnum_per_vf = hns_ae_get_q_num_per_vf(dsaf_dev, port_id);
 
-	vf_cb = kzalloc(sizeof(*vf_cb) +
-			qnum_per_vf * sizeof(struct hnae_queue *), GFP_KERNEL);
+	vf_cb = kzalloc(struct_size(vf_cb, ae_handle.qs, qnum_per_vf),
+			GFP_KERNEL);
 	if (unlikely(!vf_cb)) {
 		dev_err(dsaf_dev->dev, "malloc vf_cb fail!\n");
 		ae_handle = ERR_PTR(-ENOMEM);
@@ -99,6 +93,7 @@ struct hnae_handle *hns_ae_get_handle(struct hnae_ae_dev *dev,
 	ae_handle->owner_dev = dsaf_dev->dev;
 	ae_handle->dev = dev;
 	ae_handle->q_num = qnum_per_vf;
+	ae_handle->coal_param = HNAE_LOWEST_LATENCY_COAL_PARAM;
 
 	/* find ring pair, and set vf id*/
 	for (ae_handle->vf_id = 0;
@@ -113,7 +108,6 @@ struct hnae_handle *hns_ae_get_handle(struct hnae_ae_dev *dev,
 		goto vf_id_err;
 	}
 
-	ae_handle->qs = (struct hnae_queue **)(&ae_handle->qs + 1);
 	for (i = 0; i < qnum_per_vf; i++) {
 		ae_handle->qs[i] = &ring_pair_cb->q;
 		ae_handle->qs[i]->rx_ring.q = ae_handle->qs[i];
@@ -146,12 +140,45 @@ static void hns_ae_put_handle(struct hnae_handle *handle)
 	struct hnae_vf_cb *vf_cb = hns_ae_get_vf_cb(handle);
 	int i;
 
-	vf_cb->mac_cb	 = NULL;
+	for (i = 0; i < handle->q_num; i++)
+		hns_ae_get_ring_pair(handle->qs[i])->used_by_vf = 0;
 
 	kfree(vf_cb);
+}
 
-	for (i = 0; i < handle->q_num; i++)
-		hns_ae_get_ring_pair(handle->qs[i])->used_by_vf = 0;
+static int hns_ae_wait_flow_down(struct hnae_handle *handle)
+{
+	struct dsaf_device *dsaf_dev;
+	struct hns_ppe_cb *ppe_cb;
+	struct hnae_vf_cb *vf_cb;
+	int ret;
+	int i;
+
+	for (i = 0; i < handle->q_num; i++) {
+		ret = hns_rcb_wait_tx_ring_clean(handle->qs[i]);
+		if (ret)
+			return ret;
+	}
+
+	ppe_cb = hns_get_ppe_cb(handle);
+	ret = hns_ppe_wait_tx_fifo_clean(ppe_cb);
+	if (ret)
+		return ret;
+
+	dsaf_dev = hns_ae_get_dsaf_dev(handle->dev);
+	if (!dsaf_dev)
+		return -EINVAL;
+	ret = hns_dsaf_wait_pkt_clean(dsaf_dev, handle->dport_id);
+	if (ret)
+		return ret;
+
+	vf_cb = hns_ae_get_vf_cb(handle);
+	ret = hns_mac_wait_fifo_clean(vf_cb->mac_cb);
+	if (ret)
+		return ret;
+
+	mdelay(10);
+	return 0;
 }
 
 static void hns_ae_ring_enable_all(struct hnae_handle *handle, int val)
@@ -179,7 +206,7 @@ static void hns_ae_fini_queue(struct hnae_queue *q)
 		hns_rcb_reset_ring_hw(q);
 }
 
-static int hns_ae_set_mac_address(struct hnae_handle *handle, void *p)
+static int hns_ae_set_mac_address(struct hnae_handle *handle, const void *p)
 {
 	int ret;
 	struct hns_mac_cb *mac_cb = hns_get_mac_cb(handle);
@@ -328,7 +355,7 @@ static int hns_ae_start(struct hnae_handle *handle)
 	return 0;
 }
 
-void hns_ae_stop(struct hnae_handle *handle)
+static void hns_ae_stop(struct hnae_handle *handle)
 {
 	struct hns_mac_cb *mac_cb = hns_get_mac_cb(handle);
 
@@ -343,6 +370,9 @@ void hns_ae_stop(struct hnae_handle *handle)
 
 	hns_ae_ring_enable_all(handle, 0);
 
+	/* clean rx fbd. */
+	hns_rcb_wait_fbd_clean(handle->qs, handle->q_num, RCB_INT_FLAG_RX);
+
 	(void)hns_mac_vm_config_bc_en(mac_cb, 0, false);
 }
 
@@ -356,7 +386,7 @@ static void hns_ae_reset(struct hnae_handle *handle)
 	}
 }
 
-void hns_ae_toggle_ring_irq(struct hnae_ring *ring, u32 mask)
+static void hns_ae_toggle_ring_irq(struct hnae_ring *ring, u32 mask)
 {
 	u32 flag;
 
@@ -398,12 +428,39 @@ static int hns_ae_get_mac_info(struct hnae_handle *handle,
 	return hns_mac_get_port_info(mac_cb, auto_neg, speed, duplex);
 }
 
+static bool hns_ae_need_adjust_link(struct hnae_handle *handle, int speed,
+				    int duplex)
+{
+	struct hns_mac_cb *mac_cb = hns_get_mac_cb(handle);
+
+	return hns_mac_need_adjust_link(mac_cb, speed, duplex);
+}
+
 static void hns_ae_adjust_link(struct hnae_handle *handle, int speed,
 			       int duplex)
 {
 	struct hns_mac_cb *mac_cb = hns_get_mac_cb(handle);
 
-	hns_mac_adjust_link(mac_cb, speed, duplex);
+	switch (mac_cb->dsaf_dev->dsaf_ver) {
+	case AE_VERSION_1:
+		hns_mac_adjust_link(mac_cb, speed, duplex);
+		break;
+
+	case AE_VERSION_2:
+		/* chip need to clear all pkt inside */
+		hns_mac_disable(mac_cb, MAC_COMM_MODE_RX);
+		if (hns_ae_wait_flow_down(handle)) {
+			hns_mac_enable(mac_cb, MAC_COMM_MODE_RX);
+			break;
+		}
+
+		hns_mac_adjust_link(mac_cb, speed, duplex);
+		hns_mac_enable(mac_cb, MAC_COMM_MODE_RX);
+		break;
+
+	default:
+		break;
+	}
 }
 
 static void hns_ae_get_ring_bdnum_limit(struct hnae_queue *queue,
@@ -427,13 +484,6 @@ static void hns_ae_get_pauseparam(struct hnae_handle *handle,
 		hns_dsaf_get_rx_mac_pause_en(dsaf_dev, mac_cb->mac_id, rx_en);
 }
 
-static int hns_ae_set_autoneg(struct hnae_handle *handle, u8 enable)
-{
-	assert(handle);
-
-	return hns_mac_set_autoneg(hns_get_mac_cb(handle), enable);
-}
-
 static void hns_ae_set_promisc_mode(struct hnae_handle *handle, u32 en)
 {
 	struct hns_mac_cb *mac_cb = hns_get_mac_cb(handle);
@@ -442,17 +492,6 @@ static void hns_ae_set_promisc_mode(struct hnae_handle *handle, u32 en)
 	hns_mac_set_promisc(mac_cb, (u8)!!en);
 }
 
-static int hns_ae_get_autoneg(struct hnae_handle *handle)
-{
-	u32     auto_neg;
-
-	assert(handle);
-
-	hns_mac_get_autoneg(hns_get_mac_cb(handle), &auto_neg);
-
-	return auto_neg;
-}
-
 static int hns_ae_set_pauseparam(struct hnae_handle *handle,
 				 u32 autoneg, u32 rx_en, u32 tx_en)
 {
@@ -576,8 +615,8 @@ static void hns_ae_get_coalesce_range(struct hnae_handle *handle,
 	*rx_usecs_high  = HNS_RCB_RX_USECS_HIGH;
 }
 
-void hns_ae_update_stats(struct hnae_handle *handle,
-			 struct net_device_stats *net_stats)
+static void hns_ae_update_stats(struct hnae_handle *handle,
+				struct net_device_stats *net_stats)
 {
 	int port;
 	int idx;
@@ -588,7 +627,7 @@ void hns_ae_update_stats(struct hnae_handle *handle,
 	struct hnae_vf_cb *vf_cb = hns_ae_get_vf_cb(handle);
 	u64 tx_bytes = 0, rx_bytes = 0, tx_packets = 0, rx_packets = 0;
 	u64 rx_errors = 0, tx_errors = 0, tx_dropped = 0;
-	u64 rx_missed_errors = 0;
+	u64 rx_missed_errors;
 
 	dsaf_dev = hns_ae_get_dsaf_dev(handle->dev);
 	if (!dsaf_dev)
@@ -659,7 +698,7 @@ void hns_ae_update_stats(struct hnae_handle *handle,
 	net_stats->multicast = mac_cb->hw_stats.rx_mc_pkts;
 }
 
-void hns_ae_get_stats(struct hnae_handle *handle, u64 *data)
+static void hns_ae_get_stats(struct hnae_handle *handle, u64 *data)
 {
 	int idx;
 	struct hns_mac_cb *mac_cb;
@@ -691,15 +730,14 @@ void hns_ae_get_stats(struct hnae_handle *handle, u64 *data)
 		hns_dsaf_get_stats(vf_cb->dsaf_dev, p, vf_cb->port_index);
 }
 
-void hns_ae_get_strings(struct hnae_handle *handle,
-			u32 stringset, u8 *data)
+static void hns_ae_get_strings(struct hnae_handle *handle, u32 stringset,
+			       u8 **data)
 {
 	int port;
 	int idx;
 	struct hns_mac_cb *mac_cb;
 	struct hns_ppe_cb *ppe_cb;
 	struct dsaf_device *dsaf_dev = hns_ae_get_dsaf_dev(handle->dev);
-	u8 *p = data;
 	struct	hnae_vf_cb *vf_cb;
 
 	assert(handle);
@@ -709,22 +747,17 @@ void hns_ae_get_strings(struct hnae_handle *handle,
 	mac_cb = hns_get_mac_cb(handle);
 	ppe_cb = hns_get_ppe_cb(handle);
 
-	for (idx = 0; idx < handle->q_num; idx++) {
-		hns_rcb_get_strings(stringset, p, idx);
-		p += ETH_GSTRING_LEN * hns_rcb_get_ring_sset_count(stringset);
-	}
+	for (idx = 0; idx < handle->q_num; idx++)
+		hns_rcb_get_strings(stringset, data, idx);
 
-	hns_ppe_get_strings(ppe_cb, stringset, p);
-	p += ETH_GSTRING_LEN * hns_ppe_get_sset_count(stringset);
-
-	hns_mac_get_strings(mac_cb, stringset, p);
-	p += ETH_GSTRING_LEN * hns_mac_get_sset_count(mac_cb, stringset);
+	hns_ppe_get_strings(ppe_cb, stringset, data);
+	hns_mac_get_strings(mac_cb, stringset, data);
 
 	if (mac_cb->mac_type == HNAE_PORT_SERVICE)
-		hns_dsaf_get_strings(stringset, p, port, dsaf_dev);
+		hns_dsaf_get_strings(stringset, data, port, dsaf_dev);
 }
 
-int hns_ae_get_sset_count(struct hnae_handle *handle, int stringset)
+static int hns_ae_get_sset_count(struct hnae_handle *handle, int stringset)
 {
 	u32 sset_count = 0;
 	struct hns_mac_cb *mac_cb;
@@ -770,19 +803,20 @@ static int hns_ae_config_loopback(struct hnae_handle *handle,
 	return ret;
 }
 
-void hns_ae_update_led_status(struct hnae_handle *handle)
+static void hns_ae_update_led_status(struct hnae_handle *handle)
 {
 	struct hns_mac_cb *mac_cb;
 
 	assert(handle);
 	mac_cb = hns_get_mac_cb(handle);
-	if (!mac_cb->cpld_ctrl)
+	if (mac_cb->media_type != HNAE_MEDIA_TYPE_FIBER)
 		return;
+
 	hns_set_led_opt(mac_cb);
 }
 
-int hns_ae_cpld_set_led_id(struct hnae_handle *handle,
-			   enum hnae_led_state status)
+static int hns_ae_cpld_set_led_id(struct hnae_handle *handle,
+				  enum hnae_led_state status)
 {
 	struct hns_mac_cb *mac_cb;
 
@@ -793,7 +827,7 @@ int hns_ae_cpld_set_led_id(struct hnae_handle *handle,
 	return hns_cpld_led_set_id(mac_cb, status);
 }
 
-void hns_ae_get_regs(struct hnae_handle *handle, void *data)
+static void hns_ae_get_regs(struct hnae_handle *handle, void *data)
 {
 	u32 *p = data;
 	int i;
@@ -818,7 +852,7 @@ void hns_ae_get_regs(struct hnae_handle *handle, void *data)
 		hns_dsaf_get_regs(vf_cb->dsaf_dev, vf_cb->port_index, p);
 }
 
-int hns_ae_get_regs_len(struct hnae_handle *handle)
+static int hns_ae_get_regs_len(struct hnae_handle *handle)
 {
 	u32 total_num;
 	struct hnae_vf_cb *vf_cb = hns_ae_get_vf_cb(handle);
@@ -900,11 +934,10 @@ static struct hnae_ae_ops hns_dsaf_ops = {
 	.get_status = hns_ae_get_link_status,
 	.get_info = hns_ae_get_mac_info,
 	.adjust_link = hns_ae_adjust_link,
+	.need_adjust_link = hns_ae_need_adjust_link,
 	.set_loopback = hns_ae_config_loopback,
 	.get_ring_bdnum_limit = hns_ae_get_ring_bdnum_limit,
 	.get_pauseparam = hns_ae_get_pauseparam,
-	.set_autoneg = hns_ae_set_autoneg,
-	.get_autoneg = hns_ae_get_autoneg,
 	.set_pauseparam = hns_ae_set_pauseparam,
 	.get_coalesce_usecs = hns_ae_get_coalesce_usecs,
 	.get_max_coalesced_frames = hns_ae_get_max_coalesced_frames,
diff --git a/drivers/net/ethernet/hisilicon/hns/hns_dsaf_gmac.c b/drivers/net/ethernet/hisilicon/hns/hns_dsaf_gmac.c
index 86944bc3b273..400933ca1a29 100644
--- a/drivers/net/ethernet/hisilicon/hns/hns_dsaf_gmac.c
+++ b/drivers/net/ethernet/hisilicon/hns/hns_dsaf_gmac.c
@@ -1,10 +1,6 @@
+// SPDX-License-Identifier: GPL-2.0-or-later
 /*
  * Copyright (c) 2014-2015 Hisilicon Limited.
- *
- * This program is free software; you can redistribute it and/or modify
- * it under the terms of the GNU General Public License as published by
- * the Free Software Foundation; either version 2 of the License, or
- * (at your option) any later version.
  */
 
 #include <linux/delay.h>
@@ -67,11 +63,14 @@ static void hns_gmac_enable(void *mac_drv, enum mac_commom_mode mode)
 	struct mac_driver *drv = (struct mac_driver *)mac_drv;
 
 	/*enable GE rX/tX */
-	if ((mode == MAC_COMM_MODE_TX) || (mode == MAC_COMM_MODE_RX_AND_TX))
+	if (mode == MAC_COMM_MODE_TX || mode == MAC_COMM_MODE_RX_AND_TX)
 		dsaf_set_dev_bit(drv, GMAC_PORT_EN_REG, GMAC_PORT_TX_EN_B, 1);
 
-	if ((mode == MAC_COMM_MODE_RX) || (mode == MAC_COMM_MODE_RX_AND_TX))
+	if (mode == MAC_COMM_MODE_RX || mode == MAC_COMM_MODE_RX_AND_TX) {
+		/* enable rx pcs */
+		dsaf_set_dev_bit(drv, GMAC_PCS_RX_EN_REG, 0, 0);
 		dsaf_set_dev_bit(drv, GMAC_PORT_EN_REG, GMAC_PORT_RX_EN_B, 1);
+	}
 }
 
 static void hns_gmac_disable(void *mac_drv, enum mac_commom_mode mode)
@@ -79,11 +78,14 @@ static void hns_gmac_disable(void *mac_drv, enum mac_commom_mode mode)
 	struct mac_driver *drv = (struct mac_driver *)mac_drv;
 
 	/*disable GE rX/tX */
-	if ((mode == MAC_COMM_MODE_TX) || (mode == MAC_COMM_MODE_RX_AND_TX))
+	if (mode == MAC_COMM_MODE_TX || mode == MAC_COMM_MODE_RX_AND_TX)
 		dsaf_set_dev_bit(drv, GMAC_PORT_EN_REG, GMAC_PORT_TX_EN_B, 0);
 
-	if ((mode == MAC_COMM_MODE_RX) || (mode == MAC_COMM_MODE_RX_AND_TX))
+	if (mode == MAC_COMM_MODE_RX || mode == MAC_COMM_MODE_RX_AND_TX) {
+		/* disable rx pcs */
+		dsaf_set_dev_bit(drv, GMAC_PCS_RX_EN_REG, 0, 1);
 		dsaf_set_dev_bit(drv, GMAC_PORT_EN_REG, GMAC_PORT_RX_EN_B, 0);
+	}
 }
 
 /* hns_gmac_get_en - get port enable
@@ -128,14 +130,6 @@ static void hns_gmac_get_tx_auto_pause_frames(void *mac_drv, u16 *newval)
 				     GMAC_FC_TX_TIMER_M, GMAC_FC_TX_TIMER_S);
 }
 
-static void hns_gmac_set_rx_auto_pause_frames(void *mac_drv, u32 newval)
-{
-	struct mac_driver *drv = (struct mac_driver *)mac_drv;
-
-	dsaf_set_dev_bit(drv, GMAC_PAUSE_EN_REG,
-			 GMAC_PAUSE_EN_RX_FDFC_B, !!newval);
-}
-
 static void hns_gmac_config_max_frame_length(void *mac_drv, u16 newval)
 {
 	struct mac_driver *drv = (struct mac_driver *)mac_drv;
@@ -177,14 +171,6 @@ static void hns_gmac_tx_loop_pkt_dis(void *mac_drv)
 	dsaf_write_dev(drv, GMAC_TX_LOOP_PKT_PRI_REG, tx_loop_pkt_pri);
 }
 
-static void hns_gmac_set_duplex_type(void *mac_drv, u8 newval)
-{
-	struct mac_driver *drv = (struct mac_driver *)mac_drv;
-
-	dsaf_set_dev_bit(drv, GMAC_DUPLEX_TYPE_REG,
-			 GMAC_DUPLEX_TYPE_B, !!newval);
-}
-
 static void hns_gmac_get_duplex_type(void *mac_drv,
 				     enum hns_gmac_duplex_mdoe *duplex_mode)
 {
@@ -257,6 +243,16 @@ static void hns_gmac_get_pausefrm_cfg(void *mac_drv, u32 *rx_pause_en,
 	*tx_pause_en = dsaf_get_bit(pause_en, GMAC_PAUSE_EN_TX_FDFC_B);
 }
 
+static bool hns_gmac_need_adjust_link(void *mac_drv, enum mac_speed speed,
+				      int duplex)
+{
+	struct mac_driver *drv = (struct mac_driver *)mac_drv;
+	struct hns_mac_cb *mac_cb = drv->mac_cb;
+
+	return (mac_cb->speed != speed) ||
+		(mac_cb->half_duplex == duplex);
+}
+
 static int hns_gmac_adjust_link(void *mac_drv, enum mac_speed speed,
 				u32 full_duplex)
 {
@@ -309,6 +305,30 @@ static void hns_gmac_set_promisc(void *mac_drv, u8 en)
 		hns_gmac_set_uc_match(mac_drv, en);
 }
 
+static int hns_gmac_wait_fifo_clean(void *mac_drv)
+{
+	struct mac_driver *drv = (struct mac_driver *)mac_drv;
+	int wait_cnt;
+	u32 val;
+
+	wait_cnt = 0;
+	while (wait_cnt++ < HNS_MAX_WAIT_CNT) {
+		val = dsaf_read_dev(drv, GMAC_FIFO_STATE_REG);
+		/* bit5~bit0 is not send complete pkts */
+		if ((val & 0x3f) == 0)
+			break;
+		usleep_range(100, 200);
+	}
+
+	if (wait_cnt >= HNS_MAX_WAIT_CNT) {
+		dev_err(drv->dev,
+			"hns ge %d fifo was not idle.\n", drv->mac_id);
+		return -EBUSY;
+	}
+
+	return 0;
+}
+
 static void hns_gmac_init(void *mac_drv)
 {
 	u32 port;
@@ -339,7 +359,7 @@ static void hns_gmac_init(void *mac_drv)
 			   GMAC_TX_WATER_LINE_SHIFT, 8);
 }
 
-void hns_gmac_update_stats(void *mac_drv)
+static void hns_gmac_update_stats(void *mac_drv)
 {
 	struct mac_hw_stats *hw_stats = NULL;
 	struct mac_driver *drv = (struct mac_driver *)mac_drv;
@@ -430,7 +450,7 @@ void hns_gmac_update_stats(void *mac_drv)
 		+= dsaf_read_dev(drv, GMAC_TX_PAUSE_FRAMES_REG);
 }
 
-static void hns_gmac_set_mac_addr(void *mac_drv, char *mac_addr)
+static void hns_gmac_set_mac_addr(void *mac_drv, const char *mac_addr)
 {
 	struct mac_driver *drv = (struct mac_driver *)mac_drv;
 
@@ -649,24 +669,20 @@ static void hns_gmac_get_stats(void *mac_drv, u64 *data)
 	}
 }
 
-static void hns_gmac_get_strings(u32 stringset, u8 *data)
+static void hns_gmac_get_strings(u32 stringset, u8 **data)
 {
-	char *buff = (char *)data;
 	u32 i;
 
 	if (stringset != ETH_SS_STATS)
 		return;
 
-	for (i = 0; i < ARRAY_SIZE(g_gmac_stats_string); i++) {
-		snprintf(buff, ETH_GSTRING_LEN, "%s",
-			 g_gmac_stats_string[i].desc);
-		buff = buff + ETH_GSTRING_LEN;
-	}
+	for (i = 0; i < ARRAY_SIZE(g_gmac_stats_string); i++)
+		ethtool_puts(data, g_gmac_stats_string[i].desc);
 }
 
 static int hns_gmac_get_sset_count(int stringset)
 {
-	if (stringset == ETH_SS_STATS || stringset == ETH_SS_PRIV_FLAGS)
+	if (stringset == ETH_SS_STATS)
 		return ARRAY_SIZE(g_gmac_stats_string);
 
 	return 0;
@@ -690,6 +706,7 @@ void *hns_gmac_config(struct hns_mac_cb *mac_cb, struct mac_params *mac_param)
 	mac_drv->mac_disable = hns_gmac_disable;
 	mac_drv->mac_free = hns_gmac_free;
 	mac_drv->adjust_link = hns_gmac_adjust_link;
+	mac_drv->need_adjust_link = hns_gmac_need_adjust_link;
 	mac_drv->set_tx_auto_pause_frames = hns_gmac_set_tx_auto_pause_frames;
 	mac_drv->config_max_frame_length = hns_gmac_config_max_frame_length;
 	mac_drv->mac_pausefrm_cfg = hns_gmac_pause_frm_cfg;
@@ -704,8 +721,6 @@ void *hns_gmac_config(struct hns_mac_cb *mac_cb, struct mac_params *mac_param)
 	mac_drv->set_an_mode = hns_gmac_config_an_mode;
 	mac_drv->config_loopback = hns_gmac_config_loopback;
 	mac_drv->config_pad_and_crc = hns_gmac_config_pad_and_crc;
-	mac_drv->config_half_duplex = hns_gmac_set_duplex_type;
-	mac_drv->set_rx_ignore_pause_frames = hns_gmac_set_rx_auto_pause_frames;
 	mac_drv->get_info = hns_gmac_get_info;
 	mac_drv->autoneg_stat = hns_gmac_autoneg_stat;
 	mac_drv->get_pause_enable = hns_gmac_get_pausefrm_cfg;
@@ -717,6 +732,7 @@ void *hns_gmac_config(struct hns_mac_cb *mac_cb, struct mac_params *mac_param)
 	mac_drv->get_strings = hns_gmac_get_strings;
 	mac_drv->update_stats = hns_gmac_update_stats;
 	mac_drv->set_promiscuous = hns_gmac_set_promisc;
+	mac_drv->wait_fifo_clean = hns_gmac_wait_fifo_clean;
 
 	return (void *)mac_drv;
 }
diff --git a/drivers/net/ethernet/hisilicon/hns/hns_dsaf_gmac.h b/drivers/net/ethernet/hisilicon/hns/hns_dsaf_gmac.h
index 44fe3010dc6d..ec266e7fff83 100644
--- a/drivers/net/ethernet/hisilicon/hns/hns_dsaf_gmac.h
+++ b/drivers/net/ethernet/hisilicon/hns/hns_dsaf_gmac.h
@@ -1,10 +1,6 @@
+/* SPDX-License-Identifier: GPL-2.0-or-later */
 /*
  * Copyright (c) 2014-2015 Hisilicon Limited.
- *
- * This program is free software; you can redistribute it and/or modify
- * it under the terms of the GNU General Public License as published by
- * the Free Software Foundation; either version 2 of the License, or
- * (at your option) any later version.
  */
 
 #ifndef _HNS_GMAC_H
diff --git a/drivers/net/ethernet/hisilicon/hns/hns_dsaf_mac.c b/drivers/net/ethernet/hisilicon/hns/hns_dsaf_mac.c
index 8b5cdf490850..bc6b269be299 100644
--- a/drivers/net/ethernet/hisilicon/hns/hns_dsaf_mac.c
+++ b/drivers/net/ethernet/hisilicon/hns/hns_dsaf_mac.c
@@ -1,10 +1,6 @@
+// SPDX-License-Identifier: GPL-2.0-or-later
 /*
  * Copyright (c) 2014-2015 Hisilicon Limited.
- *
- * This program is free software; you can redistribute it and/or modify
- * it under the terms of the GNU General Public License as published by
- * the Free Software Foundation; either version 2 of the License, or
- * (at your option) any later version.
  */
 
 #include <linux/acpi.h>
@@ -70,6 +66,27 @@ static enum mac_mode hns_get_enet_interface(const struct hns_mac_cb *mac_cb)
 	}
 }
 
+static u32 hns_mac_link_anti_shake(struct mac_driver *mac_ctrl_drv)
+{
+#define HNS_MAC_LINK_WAIT_TIME 5
+#define HNS_MAC_LINK_WAIT_CNT 40
+
+	u32 link_status = 0;
+	int i;
+
+	if (!mac_ctrl_drv->get_link_status)
+		return link_status;
+
+	for (i = 0; i < HNS_MAC_LINK_WAIT_CNT; i++) {
+		msleep(HNS_MAC_LINK_WAIT_TIME);
+		mac_ctrl_drv->get_link_status(mac_ctrl_drv, &link_status);
+		if (!link_status)
+			break;
+	}
+
+	return link_status;
+}
+
 void hns_mac_get_link_status(struct hns_mac_cb *mac_cb, u32 *link_status)
 {
 	struct mac_driver *mac_ctrl_drv;
@@ -87,6 +104,14 @@ void hns_mac_get_link_status(struct hns_mac_cb *mac_cb, u32 *link_status)
 							       &sfp_prsnt);
 		if (!ret)
 			*link_status = *link_status && sfp_prsnt;
+
+		/* for FIBER port, it may have a fake link up.
+		 * when the link status changes from down to up, we need to do
+		 * anti-shake. the anti-shake time is base on tests.
+		 * only FIBER port need to do this.
+		 */
+		if (*link_status && !mac_cb->link)
+			*link_status = hns_mac_link_anti_shake(mac_ctrl_drv);
 	}
 
 	mac_cb->link = *link_status;
@@ -114,6 +139,26 @@ int hns_mac_get_port_info(struct hns_mac_cb *mac_cb,
 	return 0;
 }
 
+/**
+ *hns_mac_need_adjust_link - check is need change mac speed and duplex register
+ *@mac_cb: mac device
+ *@speed: phy device speed
+ *@duplex:phy device duplex
+ *
+ */
+bool hns_mac_need_adjust_link(struct hns_mac_cb *mac_cb, int speed, int duplex)
+{
+	struct mac_driver *mac_ctrl_drv;
+
+	mac_ctrl_drv = (struct mac_driver *)(mac_cb->priv.mac);
+
+	if (mac_ctrl_drv->need_adjust_link)
+		return mac_ctrl_drv->need_adjust_link(mac_ctrl_drv,
+			(enum mac_speed)speed, duplex);
+	else
+		return true;
+}
+
 void hns_mac_adjust_link(struct hns_mac_cb *mac_cb, int speed, int duplex)
 {
 	int ret;
@@ -224,7 +269,7 @@ int hns_mac_get_inner_port_num(struct hns_mac_cb *mac_cb, u8 vmid, u8 *port_num)
  *@addr:mac address
  */
 int hns_mac_change_vf_addr(struct hns_mac_cb *mac_cb,
-			   u32 vmid, char *addr)
+			   u32 vmid, const char *addr)
 {
 	int ret;
 	struct mac_driver *mac_ctrl_drv = hns_mac_get_drv(mac_cb);
@@ -350,7 +395,7 @@ int hns_mac_clr_multicast(struct hns_mac_cb *mac_cb, int vfn)
 static void hns_mac_param_get(struct mac_params *param,
 			      struct hns_mac_cb *mac_cb)
 {
-	param->vaddr = (void *)mac_cb->vaddr;
+	param->vaddr = mac_cb->vaddr;
 	param->mac_mode = hns_get_enet_interface(mac_cb);
 	ether_addr_copy(param->addr, mac_cb->addr_entry_idx[0].addr);
 	param->mac_id = mac_cb->mac_id;
@@ -358,18 +403,18 @@ static void hns_mac_param_get(struct mac_params *param,
 }
 
 /**
- *hns_mac_queue_config_bc_en - set broadcast rx&tx enable
- *@mac_cb: mac device
- *@queue: queue number
- *@en:enable
- *retuen 0 - success , negative --fail
+ * hns_mac_port_config_bc_en - set broadcast rx&tx enable
+ * @mac_cb: mac device
+ * @port_num: queue number
+ * @vlan_id: vlan id`
+ * @enable: enable
+ * return 0 - success , negative --fail
  */
 static int hns_mac_port_config_bc_en(struct hns_mac_cb *mac_cb,
 				     u32 port_num, u16 vlan_id, bool enable)
 {
 	int ret;
 	struct dsaf_device *dsaf_dev = mac_cb->dsaf_dev;
-	u8 addr[ETH_ALEN] = {0xff, 0xff, 0xff, 0xff, 0xff, 0xff};
 	struct dsaf_drv_mac_single_dest_entry mac_entry;
 
 	/* directy return ok in debug network mode */
@@ -377,7 +422,7 @@ static int hns_mac_port_config_bc_en(struct hns_mac_cb *mac_cb,
 		return 0;
 
 	if (!HNS_DSAF_IS_DEBUG(dsaf_dev)) {
-		memcpy(mac_entry.addr, addr, sizeof(mac_entry.addr));
+		eth_broadcast_addr(mac_entry.addr);
 		mac_entry.in_vlan_id = vlan_id;
 		mac_entry.in_port_num = mac_cb->mac_id;
 		mac_entry.port_num = port_num;
@@ -393,18 +438,17 @@ static int hns_mac_port_config_bc_en(struct hns_mac_cb *mac_cb,
 }
 
 /**
- *hns_mac_vm_config_bc_en - set broadcast rx&tx enable
- *@mac_cb: mac device
- *@vmid: vm id
- *@en:enable
- *retuen 0 - success , negative --fail
+ * hns_mac_vm_config_bc_en - set broadcast rx&tx enable
+ * @mac_cb: mac device
+ * @vmid: vm id
+ * @enable: enable
+ * return 0 - success , negative --fail
  */
 int hns_mac_vm_config_bc_en(struct hns_mac_cb *mac_cb, u32 vmid, bool enable)
 {
 	int ret;
 	struct dsaf_device *dsaf_dev = mac_cb->dsaf_dev;
 	u8 port_num;
-	u8 addr[ETH_ALEN] = {0xff, 0xff, 0xff, 0xff, 0xff, 0xff};
 	struct mac_entry_idx *uc_mac_entry;
 	struct dsaf_drv_mac_single_dest_entry mac_entry;
 
@@ -414,7 +458,7 @@ int hns_mac_vm_config_bc_en(struct hns_mac_cb *mac_cb, u32 vmid, bool enable)
 	uc_mac_entry = &mac_cb->addr_entry_idx[vmid];
 
 	if (!HNS_DSAF_IS_DEBUG(dsaf_dev))  {
-		memcpy(mac_entry.addr, addr, sizeof(mac_entry.addr));
+		eth_broadcast_addr(mac_entry.addr);
 		mac_entry.in_vlan_id = uc_mac_entry->vlan_id;
 		mac_entry.in_port_num = mac_cb->mac_id;
 		ret = hns_mac_get_inner_port_num(mac_cb, vmid, &port_num);
@@ -432,6 +476,16 @@ int hns_mac_vm_config_bc_en(struct hns_mac_cb *mac_cb, u32 vmid, bool enable)
 	return 0;
 }
 
+int hns_mac_wait_fifo_clean(struct hns_mac_cb *mac_cb)
+{
+	struct mac_driver *drv = hns_mac_get_drv(mac_cb);
+
+	if (drv->wait_fifo_clean)
+		return drv->wait_fifo_clean(drv);
+
+	return 0;
+}
+
 void hns_mac_reset(struct hns_mac_cb *mac_cb)
 {
 	struct mac_driver *drv = hns_mac_get_drv(mac_cb);
@@ -460,11 +514,6 @@ int hns_mac_set_mtu(struct hns_mac_cb *mac_cb, u32 new_mtu, u32 buf_size)
 {
 	struct mac_driver *drv = hns_mac_get_drv(mac_cb);
 	u32 new_frm = new_mtu + ETH_HLEN + ETH_FCS_LEN + VLAN_HLEN;
-	u32 max_frm = AE_IS_VER1(mac_cb->dsaf_dev->dsaf_ver) ?
-			MAC_MAX_MTU : MAC_MAX_MTU_V2;
-
-	if (mac_cb->mac_type == HNAE_PORT_DEBUG)
-		max_frm = MAC_MAX_MTU_DBG;
 
 	if (new_frm > HNS_RCB_RING_MAX_BD_PER_PKT * buf_size)
 		return -EINVAL;
@@ -523,8 +572,8 @@ void hns_mac_stop(struct hns_mac_cb *mac_cb)
 /**
  * hns_mac_get_autoneg - get auto autonegotiation
  * @mac_cb: mac control block
- * @enable: enable or not
- * retuen 0 - success , negative --fail
+ * @auto_neg: output pointer to autoneg result
+ * return 0 - success , negative --fail
  */
 void hns_mac_get_autoneg(struct hns_mac_cb *mac_cb, u32 *auto_neg)
 {
@@ -541,7 +590,7 @@ void hns_mac_get_autoneg(struct hns_mac_cb *mac_cb, u32 *auto_neg)
  * @mac_cb: mac control block
  * @rx_en: rx enable status
  * @tx_en: tx enable status
- * retuen 0 - success , negative --fail
+ * return 0 - success , negative --fail
  */
 void hns_mac_get_pauseparam(struct hns_mac_cb *mac_cb, u32 *rx_en, u32 *tx_en)
 {
@@ -559,7 +608,7 @@ void hns_mac_get_pauseparam(struct hns_mac_cb *mac_cb, u32 *rx_en, u32 *tx_en)
  * hns_mac_set_autoneg - set auto autonegotiation
  * @mac_cb: mac control block
  * @enable: enable or not
- * retuen 0 - success , negative --fail
+ * return 0 - success , negative --fail
  */
 int hns_mac_set_autoneg(struct hns_mac_cb *mac_cb, u8 enable)
 {
@@ -577,7 +626,7 @@ int hns_mac_set_autoneg(struct hns_mac_cb *mac_cb, u8 enable)
 }
 
 /**
- * hns_mac_set_autoneg - set rx & tx pause parameter
+ * hns_mac_set_pauseparam - set rx & tx pause parameter
  * @mac_cb: mac control block
  * @rx_en: rx enable or not
  * @tx_en: tx enable or not
@@ -604,7 +653,7 @@ int hns_mac_set_pauseparam(struct hns_mac_cb *mac_cb, u32 rx_en, u32 tx_en)
 /**
  * hns_mac_init_ex - mac init
  * @mac_cb: mac control block
- * retuen 0 - success , negative --fail
+ * return 0 - success , negative --fail
  */
 static int hns_mac_init_ex(struct hns_mac_cb *mac_cb)
 {
@@ -678,14 +727,14 @@ hns_mac_register_phydev(struct mii_bus *mdio, struct hns_mac_cb *mac_cb,
 		return rc;
 
 	if (!strcmp(phy_type, phy_modes(PHY_INTERFACE_MODE_XGMII)))
-		is_c45 = 1;
+		is_c45 = true;
 	else if (!strcmp(phy_type, phy_modes(PHY_INTERFACE_MODE_SGMII)))
-		is_c45 = 0;
+		is_c45 = false;
 	else
 		return -ENODATA;
 
 	phy = get_phy_device(mdio, addr, is_c45);
-	if (!phy || IS_ERR(phy))
+	if (IS_ERR_OR_NULL(phy))
 		return -EIO;
 
 	phy->irq = mdio->irq[addr];
@@ -710,7 +759,7 @@ hns_mac_register_phydev(struct mii_bus *mdio, struct hns_mac_cb *mac_cb,
 
 static int hns_mac_register_phy(struct hns_mac_cb *mac_cb)
 {
-	struct acpi_reference_args args;
+	struct fwnode_reference_args args;
 	struct platform_device *pdev;
 	struct mii_bus *mii_bus;
 	int rc;
@@ -724,13 +773,15 @@ static int hns_mac_register_phy(struct hns_mac_cb *mac_cb)
 			mac_cb->fw_port, "mdio-node", 0, &args);
 	if (rc)
 		return rc;
+	if (!is_acpi_device_node(args.fwnode))
+		return -EINVAL;
 
 	addr = hns_mac_phy_parse_addr(mac_cb->dev, mac_cb->fw_port);
 	if (addr < 0)
 		return addr;
 
 	/* dev address in adev */
-	pdev = hns_dsaf_find_platform_device(acpi_fwnode_handle(args.adev));
+	pdev = hns_dsaf_find_platform_device(args.fwnode);
 	if (!pdev) {
 		dev_err(mac_cb->dev, "mac%d mdio pdev is NULL\n",
 			mac_cb->mac_id);
@@ -753,6 +804,17 @@ static int hns_mac_register_phy(struct hns_mac_cb *mac_cb)
 	return rc;
 }
 
+static void hns_mac_remove_phydev(struct hns_mac_cb *mac_cb)
+{
+	if (!to_acpi_device_node(mac_cb->fw_port) || !mac_cb->phy_dev)
+		return;
+
+	phy_device_remove(mac_cb->phy_dev);
+	phy_device_free(mac_cb->phy_dev);
+
+	mac_cb->phy_dev = NULL;
+}
+
 #define MAC_MEDIA_TYPE_MAX_LEN		16
 
 static const struct {
@@ -768,7 +830,6 @@ static const struct {
 /**
  *hns_mac_get_info  - get mac information from device node
  *@mac_cb: mac device
- *@np:device node
  * return: 0 --success, negative --fail
  */
 static int hns_mac_get_info(struct hns_mac_cb *mac_cb)
@@ -812,8 +873,8 @@ static int hns_mac_get_info(struct hns_mac_cb *mac_cb)
 			 */
 			put_device(&mac_cb->phy_dev->mdio.dev);
 
-			dev_dbg(mac_cb->dev, "mac%d phy_node: %s\n",
-				mac_cb->mac_id, np->name);
+			dev_dbg(mac_cb->dev, "mac%d phy_node: %pOFn\n",
+				mac_cb->mac_id, np);
 		}
 		of_node_put(np);
 
@@ -830,8 +891,8 @@ static int hns_mac_get_info(struct hns_mac_cb *mac_cb)
 			 * if the phy_dev is found
 			 */
 			put_device(&mac_cb->phy_dev->mdio.dev);
-			dev_dbg(mac_cb->dev, "mac%d phy_node: %s\n",
-				mac_cb->mac_id, np->name);
+			dev_dbg(mac_cb->dev, "mac%d phy_node: %pOFn\n",
+				mac_cb->mac_id, np);
 		}
 		of_node_put(np);
 
@@ -872,6 +933,7 @@ static int hns_mac_get_info(struct hns_mac_cb *mac_cb)
 			mac_cb->cpld_ctrl = NULL;
 		} else {
 			syscon = syscon_node_to_regmap(cpld_args.np);
+			of_node_put(cpld_args.np);
 			if (IS_ERR_OR_NULL(syscon)) {
 				dev_dbg(mac_cb->dev, "no cpld-syscon found!\n");
 				mac_cb->cpld_ctrl = NULL;
@@ -882,8 +944,7 @@ static int hns_mac_get_info(struct hns_mac_cb *mac_cb)
 		}
 	} else if (is_acpi_node(mac_cb->fw_port)) {
 		ret = hns_mac_register_phy(mac_cb);
-		/*
-		 * Mac can work well if there is phy or not.If the port don't
+		/* Mac can work well if there is phy or not.If the port don't
 		 * connect with phy, the return value will be ignored. Only
 		 * when there is phy but can't find mdio bus, the return value
 		 * will be handled.
@@ -919,7 +980,7 @@ static int hns_mac_get_info(struct hns_mac_cb *mac_cb)
 /**
  * hns_mac_get_mode - get mac mode
  * @phy_if: phy interface
- * retuen 0 - gmac, 1 - xgmac , negative --fail
+ * return 0 - gmac, 1 - xgmac , negative --fail
  */
 static int hns_mac_get_mode(phy_interface_t phy_if)
 {
@@ -933,8 +994,9 @@ static int hns_mac_get_mode(phy_interface_t phy_if)
 	}
 }
 
-u8 __iomem *hns_mac_get_vaddr(struct dsaf_device *dsaf_dev,
-			      struct hns_mac_cb *mac_cb, u32 mac_mode_idx)
+static u8 __iomem *
+hns_mac_get_vaddr(struct dsaf_device *dsaf_dev,
+		  struct hns_mac_cb *mac_cb, u32 mac_mode_idx)
 {
 	u8 __iomem *base = dsaf_dev->io_base;
 	int mac_id = mac_cb->mac_id;
@@ -952,7 +1014,8 @@ u8 __iomem *hns_mac_get_vaddr(struct dsaf_device *dsaf_dev,
  * @mac_cb: mac control block
  * return 0 - success , negative --fail
  */
-int hns_mac_get_cfg(struct dsaf_device *dsaf_dev, struct hns_mac_cb *mac_cb)
+static int
+hns_mac_get_cfg(struct dsaf_device *dsaf_dev, struct hns_mac_cb *mac_cb)
 {
 	int ret;
 	u32 mac_mode_idx;
@@ -1001,6 +1064,20 @@ static int hns_mac_get_max_port_num(struct dsaf_device *dsaf_dev)
 		return  DSAF_MAX_PORT_NUM;
 }
 
+void hns_mac_enable(struct hns_mac_cb *mac_cb, enum mac_commom_mode mode)
+{
+	struct mac_driver *mac_ctrl_drv = hns_mac_get_drv(mac_cb);
+
+	mac_ctrl_drv->mac_enable(mac_cb->priv.mac, mode);
+}
+
+void hns_mac_disable(struct hns_mac_cb *mac_cb, enum mac_commom_mode mode)
+{
+	struct mac_driver *mac_ctrl_drv = hns_mac_get_drv(mac_cb);
+
+	mac_ctrl_drv->mac_disable(mac_cb->priv.mac, mode);
+}
+
 /**
  * hns_mac_init - init mac
  * @dsaf_dev: dsa fabric device struct pointer
@@ -1013,9 +1090,8 @@ int hns_mac_init(struct dsaf_device *dsaf_dev)
 	u32 port_id;
 	int max_port_num = hns_mac_get_max_port_num(dsaf_dev);
 	struct hns_mac_cb *mac_cb;
-	struct fwnode_handle *child;
 
-	device_for_each_child_node(dsaf_dev->dev, child) {
+	device_for_each_child_node_scoped(dsaf_dev->dev, child) {
 		ret = fwnode_property_read_u32(child, "reg", &port_id);
 		if (ret) {
 			dev_err(dsaf_dev->dev,
@@ -1031,6 +1107,7 @@ int hns_mac_init(struct dsaf_device *dsaf_dev)
 				      GFP_KERNEL);
 		if (!mac_cb)
 			return -ENOMEM;
+
 		mac_cb->fw_port = child;
 		mac_cb->mac_id = (u8)port_id;
 		dsaf_dev->mac_cb[port_id] = mac_cb;
@@ -1076,7 +1153,11 @@ void hns_mac_uninit(struct dsaf_device *dsaf_dev)
 	int max_port_num = hns_mac_get_max_port_num(dsaf_dev);
 
 	for (i = 0; i < max_port_num; i++) {
+		if (!dsaf_dev->mac_cb[i])
+			continue;
+
 		dsaf_dev->misc_op->cpld_reset_led(dsaf_dev->mac_cb[i]);
+		hns_mac_remove_phydev(dsaf_dev->mac_cb[i]);
 		dsaf_dev->mac_cb[i] = NULL;
 	}
 }
@@ -1109,8 +1190,7 @@ void hns_mac_get_stats(struct hns_mac_cb *mac_cb, u64 *data)
 	mac_ctrl_drv->get_ethtool_stats(mac_ctrl_drv, data);
 }
 
-void hns_mac_get_strings(struct hns_mac_cb *mac_cb,
-			 int stringset, u8 *data)
+void hns_mac_get_strings(struct hns_mac_cb *mac_cb, int stringset, u8 **data)
 {
 	struct mac_driver *mac_ctrl_drv = hns_mac_get_drv(mac_cb);
 
@@ -1150,7 +1230,7 @@ void hns_mac_get_regs(struct hns_mac_cb *mac_cb, void *data)
 
 void hns_set_led_opt(struct hns_mac_cb *mac_cb)
 {
-	int nic_data = 0;
+	int nic_data;
 	int txpkts, rxpkts;
 
 	txpkts = mac_cb->txpkt_for_led - mac_cb->hw_stats.tx_good_pkts;
@@ -1168,7 +1248,7 @@ void hns_set_led_opt(struct hns_mac_cb *mac_cb)
 int hns_cpld_led_set_id(struct hns_mac_cb *mac_cb,
 			enum hnae_led_state status)
 {
-	if (!mac_cb || !mac_cb->cpld_ctrl)
+	if (!mac_cb)
 		return 0;
 
 	return mac_cb->dsaf_dev->misc_op->cpld_set_led_id(mac_cb, status);
diff --git a/drivers/net/ethernet/hisilicon/hns/hns_dsaf_mac.h b/drivers/net/ethernet/hisilicon/hns/hns_dsaf_mac.h
index bbc0a98e7ca3..630f01cf7a71 100644
--- a/drivers/net/ethernet/hisilicon/hns/hns_dsaf_mac.h
+++ b/drivers/net/ethernet/hisilicon/hns/hns_dsaf_mac.h
@@ -1,10 +1,6 @@
+/* SPDX-License-Identifier: GPL-2.0-or-later */
 /*
  * Copyright (c) 2014-2015 Hisilicon Limited.
- *
- * This program is free software; you can redistribute it and/or modify
- * it under the terms of the GNU General Public License as published by
- * the Free Software Foundation; either version 2 of the License, or
- * (at your option) any later version.
  */
 
 #ifndef _HNS_DSAF_MAC_H
@@ -187,7 +183,7 @@ struct mac_statistics {
 /*mac para struct ,mac get param from nic or dsaf when initialize*/
 struct mac_params {
 	char addr[ETH_ALEN];
-	void *vaddr; /*virtual address*/
+	u8 __iomem *vaddr; /*virtual address*/
 	struct device *dev;
 	u8 mac_id;
 	/**< Ethernet operation mode (MAC-PHY interface and speed) */
@@ -352,10 +348,13 @@ struct mac_driver {
 	/*disable mac when disable nic or dsaf*/
 	void (*mac_disable)(void *mac_drv, enum mac_commom_mode mode);
 	/* config mac address*/
-	void (*set_mac_addr)(void *mac_drv,	char *mac_addr);
+	void (*set_mac_addr)(void *mac_drv,	const char *mac_addr);
 	/*adjust mac mode of port,include speed and duplex*/
 	int (*adjust_link)(void *mac_drv, enum mac_speed speed,
 			   u32 full_duplex);
+	/* need adjust link */
+	bool (*need_adjust_link)(void *mac_drv, enum mac_speed speed,
+				 int duplex);
 	/* config autoegotaite mode of port*/
 	void (*set_an_mode)(void *mac_drv, u8 enable);
 	/* config loopbank mode */
@@ -365,12 +364,8 @@ struct mac_driver {
 	void (*config_max_frame_length)(void *mac_drv, u16 newval);
 	/*config PAD and CRC enable */
 	void (*config_pad_and_crc)(void *mac_drv, u8 newval);
-	/* config duplex mode*/
-	void (*config_half_duplex)(void *mac_drv, u8 newval);
 	/*config tx pause time,if pause_time is zero,disable tx pause enable*/
 	void (*set_tx_auto_pause_frames)(void *mac_drv, u16 pause_time);
-	/*config rx pause enable*/
-	void (*set_rx_ignore_pause_frames)(void *mac_drv, u32 enable);
 	/* config rx mode for promiscuous*/
 	void (*set_promiscuous)(void *mac_drv, u8 enable);
 	void (*mac_pausefrm_cfg)(void *mac_drv, u32 rx_en, u32 tx_en);
@@ -383,7 +378,7 @@ struct mac_driver {
 	void (*get_regs)(void *mac_drv, void *data);
 	int (*get_regs_count)(void);
 	/* get strings name for ethtool statistic */
-	void (*get_strings)(u32 stringset, u8 *data);
+	void (*get_strings)(u32 stringset, u8 **data);
 	/* get the number of strings*/
 	int (*get_sset_count)(int stringset);
 
@@ -394,11 +389,12 @@ struct mac_driver {
 	void (*get_info)(void *mac_drv, struct mac_info *mac_info);
 
 	void (*update_stats)(void *mac_drv);
+	int (*wait_fifo_clean)(void *mac_drv);
 
 	enum mac_mode mac_mode;
 	u8 mac_id;
 	struct hns_mac_cb *mac_cb;
-	void __iomem *io_base;
+	u8 __iomem *io_base;
 	unsigned int mac_en_flg;/*you'd better don't enable mac twice*/
 	unsigned int virt_dev_num;
 	struct device *dev;
@@ -426,9 +422,10 @@ void *hns_xgmac_config(struct hns_mac_cb *mac_cb,
 		       struct mac_params *mac_param);
 
 int hns_mac_init(struct dsaf_device *dsaf_dev);
-void mac_adjust_link(struct net_device *net_dev);
+bool hns_mac_need_adjust_link(struct hns_mac_cb *mac_cb, int speed, int duplex);
 void hns_mac_get_link_status(struct hns_mac_cb *mac_cb,	u32 *link_status);
-int hns_mac_change_vf_addr(struct hns_mac_cb *mac_cb, u32 vmid, char *addr);
+int hns_mac_change_vf_addr(struct hns_mac_cb *mac_cb, u32 vmid,
+			   const char *addr);
 int hns_mac_set_multi(struct hns_mac_cb *mac_cb,
 		      u32 port_num, char *addr, bool enable);
 int hns_mac_vm_config_bc_en(struct hns_mac_cb *mac_cb, u32 vm, bool enable);
@@ -448,7 +445,7 @@ int hns_mac_config_mac_loopback(struct hns_mac_cb *mac_cb,
 				enum hnae_loop loop, int en);
 void hns_mac_update_stats(struct hns_mac_cb *mac_cb);
 void hns_mac_get_stats(struct hns_mac_cb *mac_cb, u64 *data);
-void hns_mac_get_strings(struct hns_mac_cb *mac_cb, int stringset, u8 *data);
+void hns_mac_get_strings(struct hns_mac_cb *mac_cb, int stringset, u8 **data);
 int hns_mac_get_sset_count(struct hns_mac_cb *mac_cb, int stringset);
 void hns_mac_get_regs(struct hns_mac_cb *mac_cb, void *data);
 int hns_mac_get_regs_count(struct hns_mac_cb *mac_cb);
@@ -463,5 +460,8 @@ int hns_mac_add_uc_addr(struct hns_mac_cb *mac_cb, u8 vf_id,
 int hns_mac_rm_uc_addr(struct hns_mac_cb *mac_cb, u8 vf_id,
 		       const unsigned char *addr);
 int hns_mac_clr_multicast(struct hns_mac_cb *mac_cb, int vfn);
+void hns_mac_enable(struct hns_mac_cb *mac_cb, enum mac_commom_mode mode);
+void hns_mac_disable(struct hns_mac_cb *mac_cb, enum mac_commom_mode mode);
+int hns_mac_wait_fifo_clean(struct hns_mac_cb *mac_cb);
 
 #endif /* _HNS_DSAF_MAC_H */
diff --git a/drivers/net/ethernet/hisilicon/hns/hns_dsaf_main.c b/drivers/net/ethernet/hisilicon/hns/hns_dsaf_main.c
index e0bc79ea3d88..6b6ced37e490 100644
--- a/drivers/net/ethernet/hisilicon/hns/hns_dsaf_main.c
+++ b/drivers/net/ethernet/hisilicon/hns/hns_dsaf_main.c
@@ -1,10 +1,6 @@
+// SPDX-License-Identifier: GPL-2.0-or-later
 /*
  * Copyright (c) 2014-2015 Hisilicon Limited.
- *
- * This program is free software; you can redistribute it and/or modify
- * it under the terms of the GNU General Public License as published by
- * the Free Software Foundation; either version 2 of the License, or
- * (at your option) any later version.
  */
 
 #include <linux/acpi.h>
@@ -28,7 +24,7 @@
 #include "hns_dsaf_rcb.h"
 #include "hns_dsaf_misc.h"
 
-const char *g_dsaf_mode_match[DSAF_MODE_MAX] = {
+static const char *g_dsaf_mode_match[DSAF_MODE_MAX] = {
 	[DSAF_MODE_DISABLE_2PORT_64VM] = "2port-64vf",
 	[DSAF_MODE_DISABLE_6PORT_0VM] = "6port-16rss",
 	[DSAF_MODE_DISABLE_6PORT_16VM] = "6port-16vf",
@@ -42,7 +38,7 @@ static const struct acpi_device_id hns_dsaf_acpi_match[] = {
 };
 MODULE_DEVICE_TABLE(acpi, hns_dsaf_acpi_match);
 
-int hns_dsaf_get_cfg(struct dsaf_device *dsaf_dev)
+static int hns_dsaf_get_cfg(struct dsaf_device *dsaf_dev)
 {
 	int ret, i;
 	u32 desc_num;
@@ -211,7 +207,7 @@ int hns_dsaf_get_cfg(struct dsaf_device *dsaf_dev)
 
 /**
  * hns_dsaf_sbm_link_sram_init_en - config dsaf_sbm_init_en
- * @dsaf_id: dsa fabric id
+ * @dsaf_dev: dsa fabric id
  */
 static void hns_dsaf_sbm_link_sram_init_en(struct dsaf_device *dsaf_dev)
 {
@@ -220,8 +216,8 @@ static void hns_dsaf_sbm_link_sram_init_en(struct dsaf_device *dsaf_dev)
 
 /**
  * hns_dsaf_reg_cnt_clr_ce - config hns_dsaf_reg_cnt_clr_ce
- * @dsaf_id: dsa fabric id
- * @hns_dsaf_reg_cnt_clr_ce: config value
+ * @dsaf_dev: dsa fabric id
+ * @reg_cnt_clr_ce: config value
  */
 static void
 hns_dsaf_reg_cnt_clr_ce(struct dsaf_device *dsaf_dev, u32 reg_cnt_clr_ce)
@@ -231,9 +227,9 @@ hns_dsaf_reg_cnt_clr_ce(struct dsaf_device *dsaf_dev, u32 reg_cnt_clr_ce)
 }
 
 /**
- * hns_ppe_qid_cfg - config ppe qid
- * @dsaf_id: dsa fabric id
- * @pppe_qid_cfg: value array
+ * hns_dsaf_ppe_qid_cfg - config ppe qid
+ * @dsaf_dev: dsa fabric id
+ * @qid_cfg: value array
  */
 static void
 hns_dsaf_ppe_qid_cfg(struct dsaf_device *dsaf_dev, u32 qid_cfg)
@@ -289,8 +285,8 @@ static void hns_dsaf_inner_qid_cfg(struct dsaf_device *dsaf_dev)
 
 /**
  * hns_dsaf_sw_port_type_cfg - cfg sw type
- * @dsaf_id: dsa fabric id
- * @psw_port_type: array
+ * @dsaf_dev: dsa fabric id
+ * @port_type: array
  */
 static void hns_dsaf_sw_port_type_cfg(struct dsaf_device *dsaf_dev,
 				      enum dsaf_sw_port_type port_type)
@@ -307,8 +303,8 @@ static void hns_dsaf_sw_port_type_cfg(struct dsaf_device *dsaf_dev,
 
 /**
  * hns_dsaf_stp_port_type_cfg - cfg stp type
- * @dsaf_id: dsa fabric id
- * @pstp_port_type: array
+ * @dsaf_dev: dsa fabric id
+ * @port_type: array
  */
 static void hns_dsaf_stp_port_type_cfg(struct dsaf_device *dsaf_dev,
 				       enum dsaf_stp_port_type port_type)
@@ -327,7 +323,7 @@ static void hns_dsaf_stp_port_type_cfg(struct dsaf_device *dsaf_dev,
 	(AE_IS_VER1((dev)->dsaf_ver) ? DSAF_SBM_NUM : DSAFV2_SBM_NUM)
 /**
  * hns_dsaf_sbm_cfg - config sbm
- * @dsaf_id: dsa fabric id
+ * @dsaf_dev: dsa fabric id
  */
 static void hns_dsaf_sbm_cfg(struct dsaf_device *dsaf_dev)
 {
@@ -346,7 +342,7 @@ static void hns_dsaf_sbm_cfg(struct dsaf_device *dsaf_dev)
 
 /**
  * hns_dsaf_sbm_cfg_mib_en - config sbm
- * @dsaf_id: dsa fabric id
+ * @dsaf_dev: dsa fabric id
  */
 static int hns_dsaf_sbm_cfg_mib_en(struct dsaf_device *dsaf_dev)
 {
@@ -391,7 +387,7 @@ static int hns_dsaf_sbm_cfg_mib_en(struct dsaf_device *dsaf_dev)
 
 /**
  * hns_dsaf_sbm_bp_wl_cfg - config sbm
- * @dsaf_id: dsa fabric id
+ * @dsaf_dev: dsa fabric id
  */
 static void hns_dsaf_sbm_bp_wl_cfg(struct dsaf_device *dsaf_dev)
 {
@@ -560,7 +556,7 @@ static void hns_dsafv2_sbm_bp_wl_cfg(struct dsaf_device *dsaf_dev)
 
 /**
  * hns_dsaf_voq_bp_all_thrd_cfg -  voq
- * @dsaf_id: dsa fabric id
+ * @dsaf_dev: dsa fabric id
  */
 static void hns_dsaf_voq_bp_all_thrd_cfg(struct dsaf_device *dsaf_dev)
 {
@@ -603,7 +599,7 @@ static void hns_dsaf_tbl_tcam_match_cfg(
 
 /**
  * hns_dsaf_tbl_tcam_data_cfg - tbl
- * @dsaf_id: dsa fabric id
+ * @dsaf_dev: dsa fabric id
  * @ptbl_tcam_data: addr
  */
 static void hns_dsaf_tbl_tcam_data_cfg(
@@ -617,9 +613,9 @@ static void hns_dsaf_tbl_tcam_data_cfg(
 }
 
 /**
- * dsaf_tbl_tcam_mcast_cfg - tbl
- * @dsaf_id: dsa fabric id
- * @ptbl_tcam_mcast: addr
+ * hns_dsaf_tbl_tcam_mcast_cfg - tbl
+ * @dsaf_dev: dsa fabric id
+ * @mcast: addr
  */
 static void hns_dsaf_tbl_tcam_mcast_cfg(
 	struct dsaf_device *dsaf_dev,
@@ -652,8 +648,8 @@ static void hns_dsaf_tbl_tcam_mcast_cfg(
 
 /**
  * hns_dsaf_tbl_tcam_ucast_cfg - tbl
- * @dsaf_id: dsa fabric id
- * @ptbl_tcam_ucast: addr
+ * @dsaf_dev: dsa fabric id
+ * @tbl_tcam_ucast: addr
  */
 static void hns_dsaf_tbl_tcam_ucast_cfg(
 	struct dsaf_device *dsaf_dev,
@@ -678,8 +674,8 @@ static void hns_dsaf_tbl_tcam_ucast_cfg(
 
 /**
  * hns_dsaf_tbl_line_cfg - tbl
- * @dsaf_id: dsa fabric id
- * @ptbl_lin: addr
+ * @dsaf_dev: dsa fabric id
+ * @tbl_lin: addr
  */
 static void hns_dsaf_tbl_line_cfg(struct dsaf_device *dsaf_dev,
 				  struct dsaf_tbl_line_cfg *tbl_lin)
@@ -699,7 +695,7 @@ static void hns_dsaf_tbl_line_cfg(struct dsaf_device *dsaf_dev,
 
 /**
  * hns_dsaf_tbl_tcam_mcast_pul - tbl
- * @dsaf_id: dsa fabric id
+ * @dsaf_dev: dsa fabric id
  */
 static void hns_dsaf_tbl_tcam_mcast_pul(struct dsaf_device *dsaf_dev)
 {
@@ -714,7 +710,7 @@ static void hns_dsaf_tbl_tcam_mcast_pul(struct dsaf_device *dsaf_dev)
 
 /**
  * hns_dsaf_tbl_line_pul - tbl
- * @dsaf_id: dsa fabric id
+ * @dsaf_dev: dsa fabric id
  */
 static void hns_dsaf_tbl_line_pul(struct dsaf_device *dsaf_dev)
 {
@@ -729,7 +725,7 @@ static void hns_dsaf_tbl_line_pul(struct dsaf_device *dsaf_dev)
 
 /**
  * hns_dsaf_tbl_tcam_data_mcast_pul - tbl
- * @dsaf_id: dsa fabric id
+ * @dsaf_dev: dsa fabric id
  */
 static void hns_dsaf_tbl_tcam_data_mcast_pul(
 	struct dsaf_device *dsaf_dev)
@@ -747,7 +743,7 @@ static void hns_dsaf_tbl_tcam_data_mcast_pul(
 
 /**
  * hns_dsaf_tbl_tcam_data_ucast_pul - tbl
- * @dsaf_id: dsa fabric id
+ * @dsaf_dev: dsa fabric id
  */
 static void hns_dsaf_tbl_tcam_data_ucast_pul(
 	struct dsaf_device *dsaf_dev)
@@ -772,8 +768,7 @@ void hns_dsaf_set_promisc_mode(struct dsaf_device *dsaf_dev, u32 en)
 
 /**
  * hns_dsaf_tbl_stat_en - tbl
- * @dsaf_id: dsa fabric id
- * @ptbl_stat_en: addr
+ * @dsaf_dev: dsa fabric id
  */
 static void hns_dsaf_tbl_stat_en(struct dsaf_device *dsaf_dev)
 {
@@ -789,7 +784,7 @@ static void hns_dsaf_tbl_stat_en(struct dsaf_device *dsaf_dev)
 
 /**
  * hns_dsaf_rocee_bp_en - rocee back press enable
- * @dsaf_id: dsa fabric id
+ * @dsaf_dev: dsa fabric id
  */
 static void hns_dsaf_rocee_bp_en(struct dsaf_device *dsaf_dev)
 {
@@ -856,9 +851,9 @@ static void hns_dsaf_int_tbl_src_clr(struct dsaf_device *dsaf_dev,
 
 /**
  * hns_dsaf_single_line_tbl_cfg - INT
- * @dsaf_id: dsa fabric id
- * @address:
- * @ptbl_line:
+ * @dsaf_dev: dsa fabric id
+ * @address: the address
+ * @ptbl_line: the line
  */
 static void hns_dsaf_single_line_tbl_cfg(
 	struct dsaf_device *dsaf_dev,
@@ -880,9 +875,10 @@ static void hns_dsaf_single_line_tbl_cfg(
 
 /**
  * hns_dsaf_tcam_uc_cfg - INT
- * @dsaf_id: dsa fabric id
- * @address,
- * @ptbl_tcam_data,
+ * @dsaf_dev: dsa fabric id
+ * @address: the address
+ * @ptbl_tcam_data: the data
+ * @ptbl_tcam_ucast: unicast
  */
 static void hns_dsaf_tcam_uc_cfg(
 	struct dsaf_device *dsaf_dev, u32 address,
@@ -908,7 +904,8 @@ static void hns_dsaf_tcam_uc_cfg(
  * @dsaf_dev: dsa fabric device struct pointer
  * @address: tcam index
  * @ptbl_tcam_data: tcam data struct pointer
- * @ptbl_tcam_mcast: tcam mask struct pointer, it must be null for HNSv1
+ * @ptbl_tcam_mask: tcam mask struct pointer, it must be null for HNSv1
+ * @ptbl_tcam_mcast: tcam data struct pointer
  */
 static void hns_dsaf_tcam_mc_cfg(
 	struct dsaf_device *dsaf_dev, u32 address,
@@ -935,9 +932,67 @@ static void hns_dsaf_tcam_mc_cfg(
 }
 
 /**
+ * hns_dsaf_tcam_uc_cfg_vague - INT
+ * @dsaf_dev: dsa fabric device struct pointer
+ * @address: the address
+ * @tcam_data: the data
+ * @tcam_mask: the mask
+ * @tcam_uc: the unicast data
+ */
+static void hns_dsaf_tcam_uc_cfg_vague(struct dsaf_device *dsaf_dev,
+				       u32 address,
+				       struct dsaf_tbl_tcam_data *tcam_data,
+				       struct dsaf_tbl_tcam_data *tcam_mask,
+				       struct dsaf_tbl_tcam_ucast_cfg *tcam_uc)
+{
+	spin_lock_bh(&dsaf_dev->tcam_lock);
+	hns_dsaf_tbl_tcam_addr_cfg(dsaf_dev, address);
+	hns_dsaf_tbl_tcam_data_cfg(dsaf_dev, tcam_data);
+	hns_dsaf_tbl_tcam_ucast_cfg(dsaf_dev, tcam_uc);
+	hns_dsaf_tbl_tcam_match_cfg(dsaf_dev, tcam_mask);
+	hns_dsaf_tbl_tcam_data_ucast_pul(dsaf_dev);
+
+	/*Restore Match Data*/
+	tcam_mask->tbl_tcam_data_high = 0xffffffff;
+	tcam_mask->tbl_tcam_data_low = 0xffffffff;
+	hns_dsaf_tbl_tcam_match_cfg(dsaf_dev, tcam_mask);
+
+	spin_unlock_bh(&dsaf_dev->tcam_lock);
+}
+
+/**
+ * hns_dsaf_tcam_mc_cfg_vague - INT
+ * @dsaf_dev: dsa fabric device struct pointer
+ * @address: the address
+ * @tcam_data: the data
+ * @tcam_mask: the mask
+ * @tcam_mc: the multicast data
+ */
+static void hns_dsaf_tcam_mc_cfg_vague(struct dsaf_device *dsaf_dev,
+				       u32 address,
+				       struct dsaf_tbl_tcam_data *tcam_data,
+				       struct dsaf_tbl_tcam_data *tcam_mask,
+				       struct dsaf_tbl_tcam_mcast_cfg *tcam_mc)
+{
+	spin_lock_bh(&dsaf_dev->tcam_lock);
+	hns_dsaf_tbl_tcam_addr_cfg(dsaf_dev, address);
+	hns_dsaf_tbl_tcam_data_cfg(dsaf_dev, tcam_data);
+	hns_dsaf_tbl_tcam_mcast_cfg(dsaf_dev, tcam_mc);
+	hns_dsaf_tbl_tcam_match_cfg(dsaf_dev, tcam_mask);
+	hns_dsaf_tbl_tcam_data_mcast_pul(dsaf_dev);
+
+	/*Restore Match Data*/
+	tcam_mask->tbl_tcam_data_high = 0xffffffff;
+	tcam_mask->tbl_tcam_data_low = 0xffffffff;
+	hns_dsaf_tbl_tcam_match_cfg(dsaf_dev, tcam_mask);
+
+	spin_unlock_bh(&dsaf_dev->tcam_lock);
+}
+
+/**
  * hns_dsaf_tcam_mc_invld - INT
- * @dsaf_id: dsa fabric id
- * @address
+ * @dsaf_dev: dsa fabric id
+ * @address: the address
  */
 static void hns_dsaf_tcam_mc_invld(struct dsaf_device *dsaf_dev, u32 address)
 {
@@ -959,7 +1014,8 @@ static void hns_dsaf_tcam_mc_invld(struct dsaf_device *dsaf_dev, u32 address)
 	spin_unlock_bh(&dsaf_dev->tcam_lock);
 }
 
-void hns_dsaf_tcam_addr_get(struct dsaf_drv_tbl_tcam_key *mac_key, u8 *addr)
+static void
+hns_dsaf_tcam_addr_get(struct dsaf_drv_tbl_tcam_key *mac_key, u8 *addr)
 {
 	addr[0] = mac_key->high.bits.mac_0;
 	addr[1] = mac_key->high.bits.mac_1;
@@ -971,10 +1027,10 @@ void hns_dsaf_tcam_addr_get(struct dsaf_drv_tbl_tcam_key *mac_key, u8 *addr)
 
 /**
  * hns_dsaf_tcam_uc_get - INT
- * @dsaf_id: dsa fabric id
- * @address
- * @ptbl_tcam_data
- * @ptbl_tcam_ucast
+ * @dsaf_dev: dsa fabric id
+ * @address: the address
+ * @ptbl_tcam_data: the data
+ * @ptbl_tcam_ucast: unicast
  */
 static void hns_dsaf_tcam_uc_get(
 	struct dsaf_device *dsaf_dev, u32 address,
@@ -1024,10 +1080,10 @@ static void hns_dsaf_tcam_uc_get(
 
 /**
  * hns_dsaf_tcam_mc_get - INT
- * @dsaf_id: dsa fabric id
- * @address
- * @ptbl_tcam_data
- * @ptbl_tcam_ucast
+ * @dsaf_dev: dsa fabric id
+ * @address: the address
+ * @ptbl_tcam_data: the data
+ * @ptbl_tcam_mcast: tcam multicast data
  */
 static void hns_dsaf_tcam_mc_get(
 	struct dsaf_device *dsaf_dev, u32 address,
@@ -1074,7 +1130,7 @@ static void hns_dsaf_tcam_mc_get(
 
 /**
  * hns_dsaf_tbl_line_init - INT
- * @dsaf_id: dsa fabric id
+ * @dsaf_dev: dsa fabric id
  */
 static void hns_dsaf_tbl_line_init(struct dsaf_device *dsaf_dev)
 {
@@ -1088,7 +1144,7 @@ static void hns_dsaf_tbl_line_init(struct dsaf_device *dsaf_dev)
 
 /**
  * hns_dsaf_tbl_tcam_init - INT
- * @dsaf_id: dsa fabric id
+ * @dsaf_dev: dsa fabric id
  */
 static void hns_dsaf_tbl_tcam_init(struct dsaf_device *dsaf_dev)
 {
@@ -1103,7 +1159,9 @@ static void hns_dsaf_tbl_tcam_init(struct dsaf_device *dsaf_dev)
 
 /**
  * hns_dsaf_pfc_en_cfg - dsaf pfc pause cfg
- * @mac_cb: mac contrl block
+ * @dsaf_dev: dsa fabric id
+ * @mac_id: mac contrl block
+ * @tc_en: traffic class
  */
 static void hns_dsaf_pfc_en_cfg(struct dsaf_device *dsaf_dev,
 				int mac_id, int tc_en)
@@ -1155,9 +1213,8 @@ void hns_dsaf_get_rx_mac_pause_en(struct dsaf_device *dsaf_dev, int mac_id,
 }
 
 /**
- * hns_dsaf_tbl_tcam_init - INT
- * @dsaf_id: dsa fabric id
- * @dsaf_mode
+ * hns_dsaf_comm_init - INT
+ * @dsaf_dev: dsa fabric id
  */
 static void hns_dsaf_comm_init(struct dsaf_device *dsaf_dev)
 {
@@ -1210,7 +1267,7 @@ static void hns_dsaf_comm_init(struct dsaf_device *dsaf_dev)
 
 /**
  * hns_dsaf_inode_init - INT
- * @dsaf_id: dsa fabric id
+ * @dsaf_dev: dsa fabric id
  */
 static void hns_dsaf_inode_init(struct dsaf_device *dsaf_dev)
 {
@@ -1262,7 +1319,7 @@ static void hns_dsaf_inode_init(struct dsaf_device *dsaf_dev)
 
 /**
  * hns_dsaf_sbm_init - INT
- * @dsaf_id: dsa fabric id
+ * @dsaf_dev: dsa fabric id
  */
 static int hns_dsaf_sbm_init(struct dsaf_device *dsaf_dev)
 {
@@ -1316,7 +1373,7 @@ static int hns_dsaf_sbm_init(struct dsaf_device *dsaf_dev)
 
 /**
  * hns_dsaf_tbl_init - INT
- * @dsaf_id: dsa fabric id
+ * @dsaf_dev: dsa fabric id
  */
 static void hns_dsaf_tbl_init(struct dsaf_device *dsaf_dev)
 {
@@ -1328,7 +1385,7 @@ static void hns_dsaf_tbl_init(struct dsaf_device *dsaf_dev)
 
 /**
  * hns_dsaf_voq_init - INT
- * @dsaf_id: dsa fabric id
+ * @dsaf_dev: dsa fabric id
  */
 static void hns_dsaf_voq_init(struct dsaf_device *dsaf_dev)
 {
@@ -1382,7 +1439,7 @@ static void hns_dsaf_remove_hw(struct dsaf_device *dsaf_dev)
 /**
  * hns_dsaf_init - init dsa fabric
  * @dsaf_dev: dsa fabric device struct pointer
- * retuen 0 - success , negative --fail
+ * return 0 - success , negative --fail
  */
 static int hns_dsaf_init(struct dsaf_device *dsaf_dev)
 {
@@ -1406,8 +1463,8 @@ static int hns_dsaf_init(struct dsaf_device *dsaf_dev)
 		return ret;
 
 	/* malloc mem for tcam mac key(vlan+mac) */
-	priv->soft_mac_tbl = vzalloc(sizeof(*priv->soft_mac_tbl)
-		  * DSAF_TCAM_SUM);
+	priv->soft_mac_tbl = vzalloc(array_size(DSAF_TCAM_SUM,
+						sizeof(*priv->soft_mac_tbl)));
 	if (!priv->soft_mac_tbl) {
 		ret = -ENOMEM;
 		goto remove_hw;
@@ -1492,6 +1549,27 @@ static u16 hns_dsaf_find_empty_mac_entry(struct dsaf_device *dsaf_dev)
 }
 
 /**
+ * hns_dsaf_find_empty_mac_entry_reverse
+ * search dsa fabric soft empty-entry from the end
+ * @dsaf_dev: dsa fabric device struct pointer
+ */
+static u16 hns_dsaf_find_empty_mac_entry_reverse(struct dsaf_device *dsaf_dev)
+{
+	struct dsaf_drv_priv *priv = hns_dsaf_dev_priv(dsaf_dev);
+	struct dsaf_drv_soft_mac_tbl *soft_mac_entry;
+	int i;
+
+	soft_mac_entry = priv->soft_mac_tbl + (DSAF_TCAM_SUM - 1);
+	for (i = (DSAF_TCAM_SUM - 1); i > 0; i--) {
+		/* search all entry from end to start.*/
+		if (soft_mac_entry->index == DSAF_INVALID_ENTRY_IDX)
+			return i;
+		soft_mac_entry--;
+	}
+	return DSAF_INVALID_ENTRY_IDX;
+}
+
+/**
  * hns_dsaf_set_mac_key - set mac key
  * @dsaf_dev: dsa fabric device struct pointer
  * @mac_key: tcam key pointer
@@ -1524,8 +1602,6 @@ static void hns_dsaf_set_mac_key(
 		       DSAF_TBL_TCAM_KEY_VLAN_S, vlan_id);
 	dsaf_set_field(mac_key->low.bits.port_vlan, DSAF_TBL_TCAM_KEY_PORT_M,
 		       DSAF_TBL_TCAM_KEY_PORT_S, port);
-
-	mac_key->low.bits.port_vlan = le16_to_cpu(mac_key->low.bits.port_vlan);
 }
 
 /**
@@ -1537,7 +1613,7 @@ int hns_dsaf_set_mac_uc_entry(
 	struct dsaf_device *dsaf_dev,
 	struct dsaf_drv_mac_single_dest_entry *mac_entry)
 {
-	u16 entry_index = DSAF_INVALID_ENTRY_IDX;
+	u16 entry_index;
 	struct dsaf_drv_tbl_tcam_key mac_key;
 	struct dsaf_tbl_tcam_ucast_cfg mac_data;
 	struct dsaf_drv_priv *priv =
@@ -1585,8 +1661,8 @@ int hns_dsaf_set_mac_uc_entry(
 	/* default config dvc to 0 */
 	mac_data.tbl_ucast_dvc = 0;
 	mac_data.tbl_ucast_out_port = mac_entry->port_num;
-	tcam_data.tbl_tcam_data_high = cpu_to_le32(mac_key.high.val);
-	tcam_data.tbl_tcam_data_low = cpu_to_le32(mac_key.low.val);
+	tcam_data.tbl_tcam_data_high = mac_key.high.val;
+	tcam_data.tbl_tcam_data_low = mac_key.low.val;
 
 	hns_dsaf_tcam_uc_cfg(dsaf_dev, entry_index, &tcam_data, &mac_data);
 
@@ -1603,7 +1679,7 @@ int hns_dsaf_rm_mac_addr(
 	struct dsaf_device *dsaf_dev,
 	struct dsaf_drv_mac_single_dest_entry *mac_entry)
 {
-	u16 entry_index = DSAF_INVALID_ENTRY_IDX;
+	u16 entry_index;
 	struct dsaf_tbl_tcam_ucast_cfg mac_data;
 	struct dsaf_drv_tbl_tcam_key mac_key;
 
@@ -1648,6 +1724,15 @@ int hns_dsaf_rm_mac_addr(
 				      mac_entry->addr);
 }
 
+static void hns_dsaf_setup_mc_mask(struct dsaf_device *dsaf_dev,
+				   u8 port_num, u8 *mask, u8 *addr)
+{
+	if (MAC_IS_BROADCAST(addr))
+		eth_broadcast_addr(mask);
+	else
+		memcpy(mask, dsaf_dev->mac_cb[port_num]->mc_mask, ETH_ALEN);
+}
+
 static void hns_dsaf_mc_mask_bit_clear(char *dst, const char *src)
 {
 	u16 *a = (u16 *)dst;
@@ -1666,17 +1751,15 @@ static void hns_dsaf_mc_mask_bit_clear(char *dst, const char *src)
 int hns_dsaf_add_mac_mc_port(struct dsaf_device *dsaf_dev,
 			     struct dsaf_drv_mac_single_dest_entry *mac_entry)
 {
-	u16 entry_index = DSAF_INVALID_ENTRY_IDX;
+	u16 entry_index;
 	struct dsaf_drv_tbl_tcam_key mac_key;
 	struct dsaf_drv_tbl_tcam_key mask_key;
 	struct dsaf_tbl_tcam_data *pmask_key = NULL;
 	struct dsaf_tbl_tcam_mcast_cfg mac_data;
 	struct dsaf_drv_priv *priv = hns_dsaf_dev_priv(dsaf_dev);
 	struct dsaf_drv_soft_mac_tbl *soft_mac_entry = priv->soft_mac_tbl;
-	struct dsaf_drv_tbl_tcam_key tmp_mac_key;
 	struct dsaf_tbl_tcam_data tcam_data;
 	u8 mc_addr[ETH_ALEN];
-	u8 *mc_mask;
 	int mskid;
 
 	/*chechk mac addr */
@@ -1687,9 +1770,12 @@ int hns_dsaf_add_mac_mc_port(struct dsaf_device *dsaf_dev,
 	}
 
 	ether_addr_copy(mc_addr, mac_entry->addr);
-	mc_mask = dsaf_dev->mac_cb[mac_entry->in_port_num]->mc_mask;
 	if (!AE_IS_VER1(dsaf_dev->dsaf_ver)) {
+		u8 mc_mask[ETH_ALEN];
+
 		/* prepare for key data setting */
+		hns_dsaf_setup_mc_mask(dsaf_dev, mac_entry->in_port_num,
+				       mc_mask, mac_entry->addr);
 		hns_dsaf_mc_mask_bit_clear(mc_addr, mc_mask);
 
 		/* config key mask */
@@ -1698,9 +1784,6 @@ int hns_dsaf_add_mac_mc_port(struct dsaf_device *dsaf_dev,
 				     0xff,
 				     mc_mask);
 
-		mask_key.high.val = le32_to_cpu(mask_key.high.val);
-		mask_key.low.val = le32_to_cpu(mask_key.low.val);
-
 		pmask_key = (struct dsaf_tbl_tcam_data *)(&mask_key);
 	}
 
@@ -1728,10 +1811,6 @@ int hns_dsaf_add_mac_mc_port(struct dsaf_device *dsaf_dev,
 		/* if exist, add in */
 		hns_dsaf_tcam_mc_get(dsaf_dev, entry_index, &tcam_data,
 				     &mac_data);
-
-		tmp_mac_key.high.val =
-			le32_to_cpu(tcam_data.tbl_tcam_data_high);
-		tmp_mac_key.low.val = le32_to_cpu(tcam_data.tbl_tcam_data_low);
 	}
 
 	/* config hardware entry */
@@ -1756,8 +1835,8 @@ int hns_dsaf_add_mac_mc_port(struct dsaf_device *dsaf_dev,
 		dsaf_dev->ae_dev.name, mac_key.high.val,
 		mac_key.low.val, entry_index);
 
-	tcam_data.tbl_tcam_data_high = cpu_to_le32(mac_key.high.val);
-	tcam_data.tbl_tcam_data_low = cpu_to_le32(mac_key.low.val);
+	tcam_data.tbl_tcam_data_high = mac_key.high.val;
+	tcam_data.tbl_tcam_data_low = mac_key.low.val;
 
 	/* config mc entry with mask */
 	hns_dsaf_tcam_mc_cfg(dsaf_dev, entry_index, &tcam_data,
@@ -1782,7 +1861,7 @@ int hns_dsaf_add_mac_mc_port(struct dsaf_device *dsaf_dev,
 int hns_dsaf_del_mac_entry(struct dsaf_device *dsaf_dev, u16 vlan_id,
 			   u8 in_port_num, u8 *addr)
 {
-	u16 entry_index = DSAF_INVALID_ENTRY_IDX;
+	u16 entry_index;
 	struct dsaf_drv_tbl_tcam_key mac_key;
 	struct dsaf_drv_priv *priv =
 	    (struct dsaf_drv_priv *)hns_dsaf_dev_priv(dsaf_dev);
@@ -1831,7 +1910,7 @@ int hns_dsaf_del_mac_entry(struct dsaf_device *dsaf_dev, u16 vlan_id,
 int hns_dsaf_del_mac_mc_port(struct dsaf_device *dsaf_dev,
 			     struct dsaf_drv_mac_single_dest_entry *mac_entry)
 {
-	u16 entry_index = DSAF_INVALID_ENTRY_IDX;
+	u16 entry_index;
 	struct dsaf_drv_tbl_tcam_key mac_key;
 	struct dsaf_drv_priv *priv = hns_dsaf_dev_priv(dsaf_dev);
 	struct dsaf_drv_soft_mac_tbl *soft_mac_entry = priv->soft_mac_tbl;
@@ -1841,10 +1920,9 @@ int hns_dsaf_del_mac_mc_port(struct dsaf_device *dsaf_dev,
 	struct dsaf_tbl_tcam_data tcam_data;
 	int mskid;
 	const u8 empty_msk[sizeof(mac_data.tbl_mcast_port_msk)] = {0};
-	struct dsaf_drv_tbl_tcam_key mask_key, tmp_mac_key;
+	struct dsaf_drv_tbl_tcam_key mask_key;
 	struct dsaf_tbl_tcam_data *pmask_key = NULL;
 	u8 mc_addr[ETH_ALEN];
-	u8 *mc_mask;
 
 	if (!(void *)mac_entry) {
 		dev_err(dsaf_dev->dev,
@@ -1861,17 +1939,17 @@ int hns_dsaf_del_mac_mc_port(struct dsaf_device *dsaf_dev,
 
 	/* always mask vlan_id field */
 	ether_addr_copy(mc_addr, mac_entry->addr);
-	mc_mask = dsaf_dev->mac_cb[mac_entry->in_port_num]->mc_mask;
 
 	if (!AE_IS_VER1(dsaf_dev->dsaf_ver)) {
+		u8 mc_mask[ETH_ALEN];
+
 		/* prepare for key data setting */
+		hns_dsaf_setup_mc_mask(dsaf_dev, mac_entry->in_port_num,
+				       mc_mask, mac_entry->addr);
 		hns_dsaf_mc_mask_bit_clear(mc_addr, mc_mask);
 
 		/* config key mask */
-		hns_dsaf_set_mac_key(dsaf_dev, &mask_key, 0x00, 0xff, mc_addr);
-
-		mask_key.high.val = le32_to_cpu(mask_key.high.val);
-		mask_key.low.val = le32_to_cpu(mask_key.low.val);
+		hns_dsaf_set_mac_key(dsaf_dev, &mask_key, 0x00, 0xff, mc_mask);
 
 		pmask_key = (struct dsaf_tbl_tcam_data *)(&mask_key);
 	}
@@ -1902,9 +1980,6 @@ int hns_dsaf_del_mac_mc_port(struct dsaf_device *dsaf_dev,
 	/* read entry */
 	hns_dsaf_tcam_mc_get(dsaf_dev, entry_index, &tcam_data, &mac_data);
 
-	tmp_mac_key.high.val = le32_to_cpu(tcam_data.tbl_tcam_data_high);
-	tmp_mac_key.low.val = le32_to_cpu(tcam_data.tbl_tcam_data_low);
-
 	/*del the port*/
 	if (mac_entry->port_num < DSAF_SERVICE_NW_NUM) {
 		mskid = mac_entry->port_num;
@@ -1929,8 +2004,8 @@ int hns_dsaf_del_mac_mc_port(struct dsaf_device *dsaf_dev,
 		soft_mac_entry += entry_index;
 		soft_mac_entry->index = DSAF_INVALID_ENTRY_IDX;
 	} else { /* not zero, just del port, update */
-		tcam_data.tbl_tcam_data_high = cpu_to_le32(mac_key.high.val);
-		tcam_data.tbl_tcam_data_low = cpu_to_le32(mac_key.low.val);
+		tcam_data.tbl_tcam_data_high = mac_key.high.val;
+		tcam_data.tbl_tcam_data_low = mac_key.low.val;
 
 		hns_dsaf_tcam_mc_cfg(dsaf_dev, entry_index,
 				     &tcam_data,
@@ -2028,7 +2103,7 @@ static struct dsaf_device *hns_dsaf_alloc_dev(struct device *dev,
 
 /**
  * hns_dsaf_free_dev - free dev mem
- * @dev: struct device pointer
+ * @dsaf_dev: struct device pointer
  */
 static void hns_dsaf_free_dev(struct dsaf_device *dsaf_dev)
 {
@@ -2036,10 +2111,10 @@ static void hns_dsaf_free_dev(struct dsaf_device *dsaf_dev)
 }
 
 /**
- * dsaf_pfc_unit_cnt - set pfc unit count
- * @dsaf_id: dsa fabric id
- * @pport_rate:  value array
- * @pdsaf_pfc_unit_cnt:  value array
+ * hns_dsaf_pfc_unit_cnt - set pfc unit count
+ * @dsaf_dev: dsa fabric id
+ * @mac_id: id in use
+ * @rate:  value array
  */
 static void hns_dsaf_pfc_unit_cnt(struct dsaf_device *dsaf_dev, int  mac_id,
 				  enum dsaf_port_rate_mode rate)
@@ -2067,12 +2142,14 @@ static void hns_dsaf_pfc_unit_cnt(struct dsaf_device *dsaf_dev, int  mac_id,
 }
 
 /**
- * dsaf_port_work_rate_cfg - fifo
- * @dsaf_id: dsa fabric id
- * @xge_ge_work_mode
+ * hns_dsaf_port_work_rate_cfg - fifo
+ * @dsaf_dev: dsa fabric id
+ * @mac_id: mac contrl block
+ * @rate_mode: value array
  */
-void hns_dsaf_port_work_rate_cfg(struct dsaf_device *dsaf_dev, int mac_id,
-				 enum dsaf_port_rate_mode rate_mode)
+static void
+hns_dsaf_port_work_rate_cfg(struct dsaf_device *dsaf_dev, int mac_id,
+			    enum dsaf_port_rate_mode rate_mode)
 {
 	u32 port_work_mode;
 
@@ -2159,9 +2236,9 @@ void hns_dsaf_update_stats(struct dsaf_device *dsaf_dev, u32 node_num)
 		DSAF_INODE_LOCAL_ADDR_FALSE_NUM_0_REG + 0x80 * (u64)node_num);
 
 	hw_stats->vlan_drop += dsaf_read_dev(dsaf_dev,
-		DSAF_INODE_SW_VLAN_TAG_DISC_0_REG + 0x80 * (u64)node_num);
+		DSAF_INODE_SW_VLAN_TAG_DISC_0_REG + 4 * (u64)node_num);
 	hw_stats->stp_drop += dsaf_read_dev(dsaf_dev,
-		DSAF_INODE_IN_DATA_STP_DISC_0_REG + 0x80 * (u64)node_num);
+		DSAF_INODE_IN_DATA_STP_DISC_0_REG + 4 * (u64)node_num);
 
 	/* pfc pause frame statistics stored in dsaf inode*/
 	if ((node_num < DSAF_SERVICE_NW_NUM) && !is_ver1) {
@@ -2181,12 +2258,13 @@ void hns_dsaf_update_stats(struct dsaf_device *dsaf_dev, u32 node_num)
 
 /**
  *hns_dsaf_get_regs - dump dsaf regs
- *@dsaf_dev: dsaf device
+ *@ddev: dsaf device
+ *@port: port
  *@data:data for value of regs
  */
 void hns_dsaf_get_regs(struct dsaf_device *ddev, u32 port, void *data)
 {
-	u32 i = 0;
+	u32 i;
 	u32 j;
 	u32 *p = data;
 	u32 reg_tmp;
@@ -2278,289 +2356,268 @@ void hns_dsaf_get_regs(struct dsaf_device *ddev, u32 port, void *data)
 				DSAF_INODE_BD_ORDER_STATUS_0_REG + j * 4);
 		p[223 + i] = dsaf_read_dev(ddev,
 				DSAF_INODE_SW_VLAN_TAG_DISC_0_REG + j * 4);
-		p[224 + i] = dsaf_read_dev(ddev,
+		p[226 + i] = dsaf_read_dev(ddev,
 				DSAF_INODE_IN_DATA_STP_DISC_0_REG + j * 4);
 	}
 
-	p[227] = dsaf_read_dev(ddev, DSAF_INODE_GE_FC_EN_0_REG + port * 4);
+	p[229] = dsaf_read_dev(ddev, DSAF_INODE_GE_FC_EN_0_REG + port * 4);
 
 	for (i = 0; i < DSAF_INODE_NUM / DSAF_COMM_CHN; i++) {
 		j = i * DSAF_COMM_CHN + port;
-		p[228 + i] = dsaf_read_dev(ddev,
+		p[230 + i] = dsaf_read_dev(ddev,
 				DSAF_INODE_VC0_IN_PKT_NUM_0_REG + j * 4);
 	}
 
-	p[231] = dsaf_read_dev(ddev,
-		DSAF_INODE_VC1_IN_PKT_NUM_0_REG + port * 4);
+	p[233] = dsaf_read_dev(ddev,
+		DSAF_INODE_VC1_IN_PKT_NUM_0_REG + port * 0x80);
 
 	/* dsaf inode registers */
 	for (i = 0; i < HNS_DSAF_SBM_NUM(ddev) / DSAF_COMM_CHN; i++) {
 		j = i * DSAF_COMM_CHN + port;
-		p[232 + i] = dsaf_read_dev(ddev,
+		p[234 + i] = dsaf_read_dev(ddev,
 				DSAF_SBM_CFG_REG_0_REG + j * 0x80);
-		p[235 + i] = dsaf_read_dev(ddev,
+		p[237 + i] = dsaf_read_dev(ddev,
 				DSAF_SBM_BP_CFG_0_XGE_REG_0_REG + j * 0x80);
-		p[238 + i] = dsaf_read_dev(ddev,
+		p[240 + i] = dsaf_read_dev(ddev,
 				DSAF_SBM_BP_CFG_1_REG_0_REG + j * 0x80);
-		p[241 + i] = dsaf_read_dev(ddev,
+		p[243 + i] = dsaf_read_dev(ddev,
 				DSAF_SBM_BP_CFG_2_XGE_REG_0_REG + j * 0x80);
-		p[244 + i] = dsaf_read_dev(ddev,
+		p[246 + i] = dsaf_read_dev(ddev,
 				DSAF_SBM_FREE_CNT_0_0_REG + j * 0x80);
-		p[245 + i] = dsaf_read_dev(ddev,
+		p[249 + i] = dsaf_read_dev(ddev,
 				DSAF_SBM_FREE_CNT_1_0_REG + j * 0x80);
-		p[248 + i] = dsaf_read_dev(ddev,
+		p[252 + i] = dsaf_read_dev(ddev,
 				DSAF_SBM_BP_CNT_0_0_REG + j * 0x80);
-		p[251 + i] = dsaf_read_dev(ddev,
+		p[255 + i] = dsaf_read_dev(ddev,
 				DSAF_SBM_BP_CNT_1_0_REG + j * 0x80);
-		p[254 + i] = dsaf_read_dev(ddev,
+		p[258 + i] = dsaf_read_dev(ddev,
 				DSAF_SBM_BP_CNT_2_0_REG + j * 0x80);
-		p[257 + i] = dsaf_read_dev(ddev,
+		p[261 + i] = dsaf_read_dev(ddev,
 				DSAF_SBM_BP_CNT_3_0_REG + j * 0x80);
-		p[260 + i] = dsaf_read_dev(ddev,
+		p[264 + i] = dsaf_read_dev(ddev,
 				DSAF_SBM_INER_ST_0_REG + j * 0x80);
-		p[263 + i] = dsaf_read_dev(ddev,
+		p[267 + i] = dsaf_read_dev(ddev,
 				DSAF_SBM_MIB_REQ_FAILED_TC_0_REG + j * 0x80);
-		p[266 + i] = dsaf_read_dev(ddev,
+		p[270 + i] = dsaf_read_dev(ddev,
 				DSAF_SBM_LNK_INPORT_CNT_0_REG + j * 0x80);
-		p[269 + i] = dsaf_read_dev(ddev,
+		p[273 + i] = dsaf_read_dev(ddev,
 				DSAF_SBM_LNK_DROP_CNT_0_REG + j * 0x80);
-		p[272 + i] = dsaf_read_dev(ddev,
+		p[276 + i] = dsaf_read_dev(ddev,
 				DSAF_SBM_INF_OUTPORT_CNT_0_REG + j * 0x80);
-		p[275 + i] = dsaf_read_dev(ddev,
+		p[279 + i] = dsaf_read_dev(ddev,
 				DSAF_SBM_LNK_INPORT_TC0_CNT_0_REG + j * 0x80);
-		p[278 + i] = dsaf_read_dev(ddev,
+		p[282 + i] = dsaf_read_dev(ddev,
 				DSAF_SBM_LNK_INPORT_TC1_CNT_0_REG + j * 0x80);
-		p[281 + i] = dsaf_read_dev(ddev,
+		p[285 + i] = dsaf_read_dev(ddev,
 				DSAF_SBM_LNK_INPORT_TC2_CNT_0_REG + j * 0x80);
-		p[284 + i] = dsaf_read_dev(ddev,
+		p[288 + i] = dsaf_read_dev(ddev,
 				DSAF_SBM_LNK_INPORT_TC3_CNT_0_REG + j * 0x80);
-		p[287 + i] = dsaf_read_dev(ddev,
+		p[291 + i] = dsaf_read_dev(ddev,
 				DSAF_SBM_LNK_INPORT_TC4_CNT_0_REG + j * 0x80);
-		p[290 + i] = dsaf_read_dev(ddev,
+		p[294 + i] = dsaf_read_dev(ddev,
 				DSAF_SBM_LNK_INPORT_TC5_CNT_0_REG + j * 0x80);
-		p[293 + i] = dsaf_read_dev(ddev,
+		p[297 + i] = dsaf_read_dev(ddev,
 				DSAF_SBM_LNK_INPORT_TC6_CNT_0_REG + j * 0x80);
-		p[296 + i] = dsaf_read_dev(ddev,
+		p[300 + i] = dsaf_read_dev(ddev,
 				DSAF_SBM_LNK_INPORT_TC7_CNT_0_REG + j * 0x80);
-		p[299 + i] = dsaf_read_dev(ddev,
+		p[303 + i] = dsaf_read_dev(ddev,
 				DSAF_SBM_LNK_REQ_CNT_0_REG + j * 0x80);
-		p[302 + i] = dsaf_read_dev(ddev,
+		p[306 + i] = dsaf_read_dev(ddev,
 				DSAF_SBM_LNK_RELS_CNT_0_REG + j * 0x80);
-		p[305 + i] = dsaf_read_dev(ddev,
+		p[309 + i] = dsaf_read_dev(ddev,
 				DSAF_SBM_BP_CFG_3_REG_0_REG + j * 0x80);
-		p[308 + i] = dsaf_read_dev(ddev,
+		p[312 + i] = dsaf_read_dev(ddev,
 				DSAF_SBM_BP_CFG_4_REG_0_REG + j * 0x80);
 	}
 
 	/* dsaf onode registers */
 	for (i = 0; i < DSAF_XOD_NUM; i++) {
-		p[311 + i] = dsaf_read_dev(ddev,
+		p[315 + i] = dsaf_read_dev(ddev,
 				DSAF_XOD_ETS_TSA_TC0_TC3_CFG_0_REG + i * 0x90);
-		p[319 + i] = dsaf_read_dev(ddev,
+		p[323 + i] = dsaf_read_dev(ddev,
 				DSAF_XOD_ETS_TSA_TC4_TC7_CFG_0_REG + i * 0x90);
-		p[327 + i] = dsaf_read_dev(ddev,
+		p[331 + i] = dsaf_read_dev(ddev,
 				DSAF_XOD_ETS_BW_TC0_TC3_CFG_0_REG + i * 0x90);
-		p[335 + i] = dsaf_read_dev(ddev,
+		p[339 + i] = dsaf_read_dev(ddev,
 				DSAF_XOD_ETS_BW_TC4_TC7_CFG_0_REG + i * 0x90);
-		p[343 + i] = dsaf_read_dev(ddev,
+		p[347 + i] = dsaf_read_dev(ddev,
 				DSAF_XOD_ETS_BW_OFFSET_CFG_0_REG + i * 0x90);
-		p[351 + i] = dsaf_read_dev(ddev,
+		p[355 + i] = dsaf_read_dev(ddev,
 				DSAF_XOD_ETS_TOKEN_CFG_0_REG + i * 0x90);
 	}
 
-	p[359] = dsaf_read_dev(ddev, DSAF_XOD_PFS_CFG_0_0_REG + port * 0x90);
-	p[360] = dsaf_read_dev(ddev, DSAF_XOD_PFS_CFG_1_0_REG + port * 0x90);
-	p[361] = dsaf_read_dev(ddev, DSAF_XOD_PFS_CFG_2_0_REG + port * 0x90);
+	p[363] = dsaf_read_dev(ddev, DSAF_XOD_PFS_CFG_0_0_REG + port * 0x90);
+	p[364] = dsaf_read_dev(ddev, DSAF_XOD_PFS_CFG_1_0_REG + port * 0x90);
+	p[365] = dsaf_read_dev(ddev, DSAF_XOD_PFS_CFG_2_0_REG + port * 0x90);
 
 	for (i = 0; i < DSAF_XOD_BIG_NUM / DSAF_COMM_CHN; i++) {
 		j = i * DSAF_COMM_CHN + port;
-		p[362 + i] = dsaf_read_dev(ddev,
+		p[366 + i] = dsaf_read_dev(ddev,
 				DSAF_XOD_GNT_L_0_REG + j * 0x90);
-		p[365 + i] = dsaf_read_dev(ddev,
+		p[369 + i] = dsaf_read_dev(ddev,
 				DSAF_XOD_GNT_H_0_REG + j * 0x90);
-		p[368 + i] = dsaf_read_dev(ddev,
+		p[372 + i] = dsaf_read_dev(ddev,
 				DSAF_XOD_CONNECT_STATE_0_REG + j * 0x90);
-		p[371 + i] = dsaf_read_dev(ddev,
+		p[375 + i] = dsaf_read_dev(ddev,
 				DSAF_XOD_RCVPKT_CNT_0_REG + j * 0x90);
-		p[374 + i] = dsaf_read_dev(ddev,
+		p[378 + i] = dsaf_read_dev(ddev,
 				DSAF_XOD_RCVTC0_CNT_0_REG + j * 0x90);
-		p[377 + i] = dsaf_read_dev(ddev,
+		p[381 + i] = dsaf_read_dev(ddev,
 				DSAF_XOD_RCVTC1_CNT_0_REG + j * 0x90);
-		p[380 + i] = dsaf_read_dev(ddev,
+		p[384 + i] = dsaf_read_dev(ddev,
 				DSAF_XOD_RCVTC2_CNT_0_REG + j * 0x90);
-		p[383 + i] = dsaf_read_dev(ddev,
+		p[387 + i] = dsaf_read_dev(ddev,
 				DSAF_XOD_RCVTC3_CNT_0_REG + j * 0x90);
-		p[386 + i] = dsaf_read_dev(ddev,
+		p[390 + i] = dsaf_read_dev(ddev,
 				DSAF_XOD_RCVVC0_CNT_0_REG + j * 0x90);
-		p[389 + i] = dsaf_read_dev(ddev,
+		p[393 + i] = dsaf_read_dev(ddev,
 				DSAF_XOD_RCVVC1_CNT_0_REG + j * 0x90);
 	}
 
-	p[392] = dsaf_read_dev(ddev,
+	p[396] = dsaf_read_dev(ddev,
 		DSAF_XOD_XGE_RCVIN0_CNT_0_REG + port * 0x90);
-	p[393] = dsaf_read_dev(ddev,
+	p[397] = dsaf_read_dev(ddev,
 		DSAF_XOD_XGE_RCVIN1_CNT_0_REG + port * 0x90);
-	p[394] = dsaf_read_dev(ddev,
+	p[398] = dsaf_read_dev(ddev,
 		DSAF_XOD_XGE_RCVIN2_CNT_0_REG + port * 0x90);
-	p[395] = dsaf_read_dev(ddev,
+	p[399] = dsaf_read_dev(ddev,
 		DSAF_XOD_XGE_RCVIN3_CNT_0_REG + port * 0x90);
-	p[396] = dsaf_read_dev(ddev,
+	p[400] = dsaf_read_dev(ddev,
 		DSAF_XOD_XGE_RCVIN4_CNT_0_REG + port * 0x90);
-	p[397] = dsaf_read_dev(ddev,
+	p[401] = dsaf_read_dev(ddev,
 		DSAF_XOD_XGE_RCVIN5_CNT_0_REG + port * 0x90);
-	p[398] = dsaf_read_dev(ddev,
+	p[402] = dsaf_read_dev(ddev,
 		DSAF_XOD_XGE_RCVIN6_CNT_0_REG + port * 0x90);
-	p[399] = dsaf_read_dev(ddev,
+	p[403] = dsaf_read_dev(ddev,
 		DSAF_XOD_XGE_RCVIN7_CNT_0_REG + port * 0x90);
-	p[400] = dsaf_read_dev(ddev,
+	p[404] = dsaf_read_dev(ddev,
 		DSAF_XOD_PPE_RCVIN0_CNT_0_REG + port * 0x90);
-	p[401] = dsaf_read_dev(ddev,
+	p[405] = dsaf_read_dev(ddev,
 		DSAF_XOD_PPE_RCVIN1_CNT_0_REG + port * 0x90);
-	p[402] = dsaf_read_dev(ddev,
+	p[406] = dsaf_read_dev(ddev,
 		DSAF_XOD_ROCEE_RCVIN0_CNT_0_REG + port * 0x90);
-	p[403] = dsaf_read_dev(ddev,
+	p[407] = dsaf_read_dev(ddev,
 		DSAF_XOD_ROCEE_RCVIN1_CNT_0_REG + port * 0x90);
-	p[404] = dsaf_read_dev(ddev,
+	p[408] = dsaf_read_dev(ddev,
 		DSAF_XOD_FIFO_STATUS_0_REG + port * 0x90);
 
 	/* dsaf voq registers */
 	for (i = 0; i < DSAF_VOQ_NUM / DSAF_COMM_CHN; i++) {
 		j = (i * DSAF_COMM_CHN + port) * 0x90;
-		p[405 + i] = dsaf_read_dev(ddev,
+		p[409 + i] = dsaf_read_dev(ddev,
 			DSAF_VOQ_ECC_INVERT_EN_0_REG + j);
-		p[408 + i] = dsaf_read_dev(ddev,
+		p[412 + i] = dsaf_read_dev(ddev,
 			DSAF_VOQ_SRAM_PKT_NUM_0_REG + j);
-		p[411 + i] = dsaf_read_dev(ddev, DSAF_VOQ_IN_PKT_NUM_0_REG + j);
-		p[414 + i] = dsaf_read_dev(ddev,
+		p[415 + i] = dsaf_read_dev(ddev, DSAF_VOQ_IN_PKT_NUM_0_REG + j);
+		p[418 + i] = dsaf_read_dev(ddev,
 			DSAF_VOQ_OUT_PKT_NUM_0_REG + j);
-		p[417 + i] = dsaf_read_dev(ddev,
+		p[421 + i] = dsaf_read_dev(ddev,
 			DSAF_VOQ_ECC_ERR_ADDR_0_REG + j);
-		p[420 + i] = dsaf_read_dev(ddev, DSAF_VOQ_BP_STATUS_0_REG + j);
-		p[423 + i] = dsaf_read_dev(ddev, DSAF_VOQ_SPUP_IDLE_0_REG + j);
-		p[426 + i] = dsaf_read_dev(ddev,
+		p[424 + i] = dsaf_read_dev(ddev, DSAF_VOQ_BP_STATUS_0_REG + j);
+		p[427 + i] = dsaf_read_dev(ddev, DSAF_VOQ_SPUP_IDLE_0_REG + j);
+		p[430 + i] = dsaf_read_dev(ddev,
 			DSAF_VOQ_XGE_XOD_REQ_0_0_REG + j);
-		p[429 + i] = dsaf_read_dev(ddev,
+		p[433 + i] = dsaf_read_dev(ddev,
 			DSAF_VOQ_XGE_XOD_REQ_1_0_REG + j);
-		p[432 + i] = dsaf_read_dev(ddev,
+		p[436 + i] = dsaf_read_dev(ddev,
 			DSAF_VOQ_PPE_XOD_REQ_0_REG + j);
-		p[435 + i] = dsaf_read_dev(ddev,
+		p[439 + i] = dsaf_read_dev(ddev,
 			DSAF_VOQ_ROCEE_XOD_REQ_0_REG + j);
-		p[438 + i] = dsaf_read_dev(ddev,
+		p[442 + i] = dsaf_read_dev(ddev,
 			DSAF_VOQ_BP_ALL_THRD_0_REG + j);
 	}
 
 	/* dsaf tbl registers */
-	p[441] = dsaf_read_dev(ddev, DSAF_TBL_CTRL_0_REG);
-	p[442] = dsaf_read_dev(ddev, DSAF_TBL_INT_MSK_0_REG);
-	p[443] = dsaf_read_dev(ddev, DSAF_TBL_INT_SRC_0_REG);
-	p[444] = dsaf_read_dev(ddev, DSAF_TBL_INT_STS_0_REG);
-	p[445] = dsaf_read_dev(ddev, DSAF_TBL_TCAM_ADDR_0_REG);
-	p[446] = dsaf_read_dev(ddev, DSAF_TBL_LINE_ADDR_0_REG);
-	p[447] = dsaf_read_dev(ddev, DSAF_TBL_TCAM_HIGH_0_REG);
-	p[448] = dsaf_read_dev(ddev, DSAF_TBL_TCAM_LOW_0_REG);
-	p[449] = dsaf_read_dev(ddev, DSAF_TBL_TCAM_MCAST_CFG_4_0_REG);
-	p[450] = dsaf_read_dev(ddev, DSAF_TBL_TCAM_MCAST_CFG_3_0_REG);
-	p[451] = dsaf_read_dev(ddev, DSAF_TBL_TCAM_MCAST_CFG_2_0_REG);
-	p[452] = dsaf_read_dev(ddev, DSAF_TBL_TCAM_MCAST_CFG_1_0_REG);
-	p[453] = dsaf_read_dev(ddev, DSAF_TBL_TCAM_MCAST_CFG_0_0_REG);
-	p[454] = dsaf_read_dev(ddev, DSAF_TBL_TCAM_UCAST_CFG_0_REG);
-	p[455] = dsaf_read_dev(ddev, DSAF_TBL_LIN_CFG_0_REG);
-	p[456] = dsaf_read_dev(ddev, DSAF_TBL_TCAM_RDATA_HIGH_0_REG);
-	p[457] = dsaf_read_dev(ddev, DSAF_TBL_TCAM_RDATA_LOW_0_REG);
-	p[458] = dsaf_read_dev(ddev, DSAF_TBL_TCAM_RAM_RDATA4_0_REG);
-	p[459] = dsaf_read_dev(ddev, DSAF_TBL_TCAM_RAM_RDATA3_0_REG);
-	p[460] = dsaf_read_dev(ddev, DSAF_TBL_TCAM_RAM_RDATA2_0_REG);
-	p[461] = dsaf_read_dev(ddev, DSAF_TBL_TCAM_RAM_RDATA1_0_REG);
-	p[462] = dsaf_read_dev(ddev, DSAF_TBL_TCAM_RAM_RDATA0_0_REG);
-	p[463] = dsaf_read_dev(ddev, DSAF_TBL_LIN_RDATA_0_REG);
+	p[445] = dsaf_read_dev(ddev, DSAF_TBL_CTRL_0_REG);
+	p[446] = dsaf_read_dev(ddev, DSAF_TBL_INT_MSK_0_REG);
+	p[447] = dsaf_read_dev(ddev, DSAF_TBL_INT_SRC_0_REG);
+	p[448] = dsaf_read_dev(ddev, DSAF_TBL_INT_STS_0_REG);
+	p[449] = dsaf_read_dev(ddev, DSAF_TBL_TCAM_ADDR_0_REG);
+	p[450] = dsaf_read_dev(ddev, DSAF_TBL_LINE_ADDR_0_REG);
+	p[451] = dsaf_read_dev(ddev, DSAF_TBL_TCAM_HIGH_0_REG);
+	p[452] = dsaf_read_dev(ddev, DSAF_TBL_TCAM_LOW_0_REG);
+	p[453] = dsaf_read_dev(ddev, DSAF_TBL_TCAM_MCAST_CFG_4_0_REG);
+	p[454] = dsaf_read_dev(ddev, DSAF_TBL_TCAM_MCAST_CFG_3_0_REG);
+	p[455] = dsaf_read_dev(ddev, DSAF_TBL_TCAM_MCAST_CFG_2_0_REG);
+	p[456] = dsaf_read_dev(ddev, DSAF_TBL_TCAM_MCAST_CFG_1_0_REG);
+	p[457] = dsaf_read_dev(ddev, DSAF_TBL_TCAM_MCAST_CFG_0_0_REG);
+	p[458] = dsaf_read_dev(ddev, DSAF_TBL_TCAM_UCAST_CFG_0_REG);
+	p[459] = dsaf_read_dev(ddev, DSAF_TBL_LIN_CFG_0_REG);
+	p[460] = dsaf_read_dev(ddev, DSAF_TBL_TCAM_RDATA_HIGH_0_REG);
+	p[461] = dsaf_read_dev(ddev, DSAF_TBL_TCAM_RDATA_LOW_0_REG);
+	p[462] = dsaf_read_dev(ddev, DSAF_TBL_TCAM_RAM_RDATA4_0_REG);
+	p[463] = dsaf_read_dev(ddev, DSAF_TBL_TCAM_RAM_RDATA3_0_REG);
+	p[464] = dsaf_read_dev(ddev, DSAF_TBL_TCAM_RAM_RDATA2_0_REG);
+	p[465] = dsaf_read_dev(ddev, DSAF_TBL_TCAM_RAM_RDATA1_0_REG);
+	p[466] = dsaf_read_dev(ddev, DSAF_TBL_TCAM_RAM_RDATA0_0_REG);
+	p[467] = dsaf_read_dev(ddev, DSAF_TBL_LIN_RDATA_0_REG);
 
 	for (i = 0; i < DSAF_SW_PORT_NUM; i++) {
 		j = i * 0x8;
-		p[464 + 2 * i] = dsaf_read_dev(ddev,
+		p[468 + 2 * i] = dsaf_read_dev(ddev,
 			DSAF_TBL_DA0_MIS_INFO1_0_REG + j);
-		p[465 + 2 * i] = dsaf_read_dev(ddev,
+		p[469 + 2 * i] = dsaf_read_dev(ddev,
 			DSAF_TBL_DA0_MIS_INFO0_0_REG + j);
 	}
 
-	p[480] = dsaf_read_dev(ddev, DSAF_TBL_SA_MIS_INFO2_0_REG);
-	p[481] = dsaf_read_dev(ddev, DSAF_TBL_SA_MIS_INFO1_0_REG);
-	p[482] = dsaf_read_dev(ddev, DSAF_TBL_SA_MIS_INFO0_0_REG);
-	p[483] = dsaf_read_dev(ddev, DSAF_TBL_PUL_0_REG);
-	p[484] = dsaf_read_dev(ddev, DSAF_TBL_OLD_RSLT_0_REG);
-	p[485] = dsaf_read_dev(ddev, DSAF_TBL_OLD_SCAN_VAL_0_REG);
-	p[486] = dsaf_read_dev(ddev, DSAF_TBL_DFX_CTRL_0_REG);
-	p[487] = dsaf_read_dev(ddev, DSAF_TBL_DFX_STAT_0_REG);
-	p[488] = dsaf_read_dev(ddev, DSAF_TBL_DFX_STAT_2_0_REG);
-	p[489] = dsaf_read_dev(ddev, DSAF_TBL_LKUP_NUM_I_0_REG);
-	p[490] = dsaf_read_dev(ddev, DSAF_TBL_LKUP_NUM_O_0_REG);
-	p[491] = dsaf_read_dev(ddev, DSAF_TBL_UCAST_BCAST_MIS_INFO_0_0_REG);
+	p[484] = dsaf_read_dev(ddev, DSAF_TBL_SA_MIS_INFO2_0_REG);
+	p[485] = dsaf_read_dev(ddev, DSAF_TBL_SA_MIS_INFO1_0_REG);
+	p[486] = dsaf_read_dev(ddev, DSAF_TBL_SA_MIS_INFO0_0_REG);
+	p[487] = dsaf_read_dev(ddev, DSAF_TBL_PUL_0_REG);
+	p[488] = dsaf_read_dev(ddev, DSAF_TBL_OLD_RSLT_0_REG);
+	p[489] = dsaf_read_dev(ddev, DSAF_TBL_OLD_SCAN_VAL_0_REG);
+	p[490] = dsaf_read_dev(ddev, DSAF_TBL_DFX_CTRL_0_REG);
+	p[491] = dsaf_read_dev(ddev, DSAF_TBL_DFX_STAT_0_REG);
+	p[492] = dsaf_read_dev(ddev, DSAF_TBL_DFX_STAT_2_0_REG);
+	p[493] = dsaf_read_dev(ddev, DSAF_TBL_LKUP_NUM_I_0_REG);
+	p[494] = dsaf_read_dev(ddev, DSAF_TBL_LKUP_NUM_O_0_REG);
+	p[495] = dsaf_read_dev(ddev, DSAF_TBL_UCAST_BCAST_MIS_INFO_0_0_REG);
 
 	/* dsaf other registers */
-	p[492] = dsaf_read_dev(ddev, DSAF_INODE_FIFO_WL_0_REG + port * 0x4);
-	p[493] = dsaf_read_dev(ddev, DSAF_ONODE_FIFO_WL_0_REG + port * 0x4);
-	p[494] = dsaf_read_dev(ddev, DSAF_XGE_GE_WORK_MODE_0_REG + port * 0x4);
-	p[495] = dsaf_read_dev(ddev,
+	p[496] = dsaf_read_dev(ddev, DSAF_INODE_FIFO_WL_0_REG + port * 0x4);
+	p[497] = dsaf_read_dev(ddev, DSAF_ONODE_FIFO_WL_0_REG + port * 0x4);
+	p[498] = dsaf_read_dev(ddev, DSAF_XGE_GE_WORK_MODE_0_REG + port * 0x4);
+	p[499] = dsaf_read_dev(ddev,
 		DSAF_XGE_APP_RX_LINK_UP_0_REG + port * 0x4);
-	p[496] = dsaf_read_dev(ddev, DSAF_NETPORT_CTRL_SIG_0_REG + port * 0x4);
-	p[497] = dsaf_read_dev(ddev, DSAF_XGE_CTRL_SIG_CFG_0_REG + port * 0x4);
+	p[500] = dsaf_read_dev(ddev, DSAF_NETPORT_CTRL_SIG_0_REG + port * 0x4);
+	p[501] = dsaf_read_dev(ddev, DSAF_XGE_CTRL_SIG_CFG_0_REG + port * 0x4);
 
 	if (!is_ver1)
-		p[498] = dsaf_read_dev(ddev, DSAF_PAUSE_CFG_REG + port * 0x4);
+		p[502] = dsaf_read_dev(ddev, DSAF_PAUSE_CFG_REG + port * 0x4);
 
 	/* mark end of dsaf regs */
-	for (i = 499; i < 504; i++)
+	for (i = 503; i < 504; i++)
 		p[i] = 0xdddddddd;
 }
 
-static char *hns_dsaf_get_node_stats_strings(char *data, int node,
-					     struct dsaf_device *dsaf_dev)
+static void hns_dsaf_get_node_stats_strings(u8 **data, int node,
+					    struct dsaf_device *dsaf_dev)
 {
-	char *buff = data;
-	int i;
 	bool is_ver1 = AE_IS_VER1(dsaf_dev->dsaf_ver);
+	int i;
 
-	snprintf(buff, ETH_GSTRING_LEN, "innod%d_pad_drop_pkts", node);
-	buff += ETH_GSTRING_LEN;
-	snprintf(buff, ETH_GSTRING_LEN, "innod%d_manage_pkts", node);
-	buff += ETH_GSTRING_LEN;
-	snprintf(buff, ETH_GSTRING_LEN, "innod%d_rx_pkts", node);
-	buff += ETH_GSTRING_LEN;
-	snprintf(buff, ETH_GSTRING_LEN, "innod%d_rx_pkt_id", node);
-	buff += ETH_GSTRING_LEN;
-	snprintf(buff, ETH_GSTRING_LEN, "innod%d_rx_pause_frame", node);
-	buff += ETH_GSTRING_LEN;
-	snprintf(buff, ETH_GSTRING_LEN, "innod%d_release_buf_num", node);
-	buff += ETH_GSTRING_LEN;
-	snprintf(buff, ETH_GSTRING_LEN, "innod%d_sbm_drop_pkts", node);
-	buff += ETH_GSTRING_LEN;
-	snprintf(buff, ETH_GSTRING_LEN, "innod%d_crc_false_pkts", node);
-	buff += ETH_GSTRING_LEN;
-	snprintf(buff, ETH_GSTRING_LEN, "innod%d_bp_drop_pkts", node);
-	buff += ETH_GSTRING_LEN;
-	snprintf(buff, ETH_GSTRING_LEN, "innod%d_lookup_rslt_drop_pkts", node);
-	buff += ETH_GSTRING_LEN;
-	snprintf(buff, ETH_GSTRING_LEN, "innod%d_local_rslt_fail_pkts", node);
-	buff += ETH_GSTRING_LEN;
-	snprintf(buff, ETH_GSTRING_LEN, "innod%d_vlan_drop_pkts", node);
-	buff += ETH_GSTRING_LEN;
-	snprintf(buff, ETH_GSTRING_LEN, "innod%d_stp_drop_pkts", node);
-	buff += ETH_GSTRING_LEN;
+	ethtool_sprintf(data, "innod%d_pad_drop_pkts", node);
+	ethtool_sprintf(data, "innod%d_manage_pkts", node);
+	ethtool_sprintf(data, "innod%d_rx_pkts", node);
+	ethtool_sprintf(data, "innod%d_rx_pkt_id", node);
+	ethtool_sprintf(data, "innod%d_rx_pause_frame", node);
+	ethtool_sprintf(data, "innod%d_release_buf_num", node);
+	ethtool_sprintf(data, "innod%d_sbm_drop_pkts", node);
+	ethtool_sprintf(data, "innod%d_crc_false_pkts", node);
+	ethtool_sprintf(data, "innod%d_bp_drop_pkts", node);
+	ethtool_sprintf(data, "innod%d_lookup_rslt_drop_pkts", node);
+	ethtool_sprintf(data, "innod%d_local_rslt_fail_pkts", node);
+	ethtool_sprintf(data, "innod%d_vlan_drop_pkts", node);
+	ethtool_sprintf(data, "innod%d_stp_drop_pkts", node);
 	if (node < DSAF_SERVICE_NW_NUM && !is_ver1) {
 		for (i = 0; i < DSAF_PRIO_NR; i++) {
-			snprintf(buff + 0 * ETH_GSTRING_LEN * DSAF_PRIO_NR,
-				 ETH_GSTRING_LEN, "inod%d_pfc_prio%d_pkts",
-				 node, i);
-			snprintf(buff + 1 * ETH_GSTRING_LEN * DSAF_PRIO_NR,
-				 ETH_GSTRING_LEN, "onod%d_pfc_prio%d_pkts",
-				 node, i);
-			buff += ETH_GSTRING_LEN;
+			ethtool_sprintf(data, "inod%d_pfc_prio%d_pkts", node,
+					i);
+			ethtool_sprintf(data, "onod%d_pfc_prio%d_pkts", node,
+					i);
 		}
-		buff += 1 * DSAF_PRIO_NR * ETH_GSTRING_LEN;
 	}
-	snprintf(buff, ETH_GSTRING_LEN, "onnod%d_tx_pkts", node);
-	buff += ETH_GSTRING_LEN;
-
-	return buff;
+	ethtool_sprintf(data, "onnod%d_tx_pkts", node);
 }
 
 static u64 *hns_dsaf_get_node_stats(struct dsaf_device *ddev, u64 *data,
@@ -2618,6 +2675,7 @@ void hns_dsaf_get_stats(struct dsaf_device *ddev, u64 *data, int port)
 
 /**
  *hns_dsaf_get_sset_count - get dsaf string set count
+ *@dsaf_dev: dsaf device
  *@stringset: type of values in data
  *return dsaf string name count
  */
@@ -2639,26 +2697,26 @@ int hns_dsaf_get_sset_count(struct dsaf_device *dsaf_dev, int stringset)
  *@stringset:srting set index
  *@data:strings name value
  *@port:port index
+ *@dsaf_dev: dsaf device
  */
-void hns_dsaf_get_strings(int stringset, u8 *data, int port,
+void hns_dsaf_get_strings(int stringset, u8 **data, int port,
 			  struct dsaf_device *dsaf_dev)
 {
-	char *buff = (char *)data;
 	int node = port;
 
 	if (stringset != ETH_SS_STATS)
 		return;
 
 	/* for ge/xge node info */
-	buff = hns_dsaf_get_node_stats_strings(buff, node, dsaf_dev);
+	hns_dsaf_get_node_stats_strings(data, node, dsaf_dev);
 
 	/* for ppe node info */
 	node = port + DSAF_PPE_INODE_BASE;
-	(void)hns_dsaf_get_node_stats_strings(buff, node, dsaf_dev);
+	hns_dsaf_get_node_stats_strings(data, node, dsaf_dev);
 }
 
 /**
- *hns_dsaf_get_sset_count - get dsaf regs count
+ *hns_dsaf_get_regs_count - get dsaf regs count
  *return dsaf regs count
  */
 int hns_dsaf_get_regs_count(void)
@@ -2666,64 +2724,212 @@ int hns_dsaf_get_regs_count(void)
 	return DSAF_DUMP_REGS_NUM;
 }
 
-/* Reserve the last TCAM entry for promisc support */
-#define dsaf_promisc_tcam_entry(port) \
-	(DSAF_TCAM_SUM - DSAFV2_MAC_FUZZY_TCAM_NUM + (port))
-void hns_dsaf_set_promisc_tcam(struct dsaf_device *dsaf_dev,
-			       u32 port, bool enable)
+static int hns_dsaf_get_port_id(u8 port)
+{
+	if (port < DSAF_SERVICE_NW_NUM)
+		return port;
+
+	if (port >= DSAF_BASE_INNER_PORT_NUM)
+		return port - DSAF_BASE_INNER_PORT_NUM + DSAF_SERVICE_NW_NUM;
+
+	return -EINVAL;
+}
+
+static void set_promisc_tcam_enable(struct dsaf_device *dsaf_dev, u32 port)
 {
+	struct dsaf_tbl_tcam_ucast_cfg tbl_tcam_ucast = {0, 1, 0, 0, 0x80};
+	struct dsaf_tbl_tcam_data tbl_tcam_data_mc = {0x01000000, port};
+	struct dsaf_tbl_tcam_data tbl_tcam_mask_uc = {0x01000000, 0xf};
+	struct dsaf_tbl_tcam_mcast_cfg tbl_tcam_mcast = {0, 0, {0} };
 	struct dsaf_drv_priv *priv = hns_dsaf_dev_priv(dsaf_dev);
-	struct dsaf_drv_soft_mac_tbl *soft_mac_entry = priv->soft_mac_tbl;
+	struct dsaf_tbl_tcam_data tbl_tcam_data_uc = {0, port};
+	struct dsaf_drv_mac_single_dest_entry mask_entry;
+	struct dsaf_drv_tbl_tcam_key temp_key, mask_key;
+	struct dsaf_drv_soft_mac_tbl *soft_mac_entry;
 	u16 entry_index;
-	struct dsaf_drv_tbl_tcam_key tbl_tcam_data, tbl_tcam_mask;
-	struct dsaf_tbl_tcam_mcast_cfg mac_data = {0};
+	struct dsaf_drv_tbl_tcam_key mac_key;
+	struct hns_mac_cb *mac_cb;
+	u8 addr[ETH_ALEN] = {0};
+	u8 port_num;
+	int mskid;
 
-	if ((AE_IS_VER1(dsaf_dev->dsaf_ver)) || HNS_DSAF_IS_DEBUG(dsaf_dev))
+	/* promisc use vague table match with vlanid = 0 & macaddr = 0 */
+	hns_dsaf_set_mac_key(dsaf_dev, &mac_key, 0x00, port, addr);
+	entry_index = hns_dsaf_find_soft_mac_entry(dsaf_dev, &mac_key);
+	if (entry_index != DSAF_INVALID_ENTRY_IDX)
 		return;
 
-	/* find the tcam entry index for promisc */
-	entry_index = dsaf_promisc_tcam_entry(port);
-
-	memset(&tbl_tcam_data, 0, sizeof(tbl_tcam_data));
-	memset(&tbl_tcam_mask, 0, sizeof(tbl_tcam_mask));
-
-	/* config key mask */
-	if (enable) {
-		dsaf_set_field(tbl_tcam_data.low.bits.port_vlan,
-			       DSAF_TBL_TCAM_KEY_PORT_M,
-			       DSAF_TBL_TCAM_KEY_PORT_S, port);
-		dsaf_set_field(tbl_tcam_mask.low.bits.port_vlan,
-			       DSAF_TBL_TCAM_KEY_PORT_M,
-			       DSAF_TBL_TCAM_KEY_PORT_S, 0xf);
-
-		/* SUB_QID */
-		dsaf_set_bit(mac_data.tbl_mcast_port_msk[0],
-			     DSAF_SERVICE_NW_NUM, true);
-		mac_data.tbl_mcast_item_vld = true;	/* item_vld bit */
-	} else {
-		mac_data.tbl_mcast_item_vld = false;	/* item_vld bit */
+	/* put promisc tcam entry in the end. */
+	/* 1. set promisc unicast vague tcam entry. */
+	entry_index = hns_dsaf_find_empty_mac_entry_reverse(dsaf_dev);
+	if (entry_index == DSAF_INVALID_ENTRY_IDX) {
+		dev_err(dsaf_dev->dev,
+			"enable uc promisc failed (port:%#x)\n",
+			port);
+		return;
 	}
 
-	dev_dbg(dsaf_dev->dev,
-		"set_promisc_entry, %s Mac key(%#x:%#x) entry_index%d\n",
-		dsaf_dev->ae_dev.name, tbl_tcam_data.high.val,
-		tbl_tcam_data.low.val, entry_index);
+	mac_cb = dsaf_dev->mac_cb[port];
+	(void)hns_mac_get_inner_port_num(mac_cb, 0, &port_num);
+	tbl_tcam_ucast.tbl_ucast_out_port = port_num;
 
-	/* config promisc entry with mask */
-	hns_dsaf_tcam_mc_cfg(dsaf_dev, entry_index,
-			     (struct dsaf_tbl_tcam_data *)&tbl_tcam_data,
-			     (struct dsaf_tbl_tcam_data *)&tbl_tcam_mask,
-			     &mac_data);
+	/* config uc vague table */
+	hns_dsaf_tcam_uc_cfg_vague(dsaf_dev, entry_index, &tbl_tcam_data_uc,
+				   &tbl_tcam_mask_uc, &tbl_tcam_ucast);
 
-	/* config software entry */
+	/* update software entry */
+	soft_mac_entry = priv->soft_mac_tbl;
 	soft_mac_entry += entry_index;
-	soft_mac_entry->index = enable ? entry_index : DSAF_INVALID_ENTRY_IDX;
+	soft_mac_entry->index = entry_index;
+	soft_mac_entry->tcam_key.high.val = mac_key.high.val;
+	soft_mac_entry->tcam_key.low.val = mac_key.low.val;
+	/* step back to the START for mc. */
+	soft_mac_entry = priv->soft_mac_tbl;
+
+	/* 2. set promisc multicast vague tcam entry. */
+	entry_index = hns_dsaf_find_empty_mac_entry_reverse(dsaf_dev);
+	if (entry_index == DSAF_INVALID_ENTRY_IDX) {
+		dev_err(dsaf_dev->dev,
+			"enable mc promisc failed (port:%#x)\n",
+			port);
+		return;
+	}
+
+	memset(&mask_entry, 0x0, sizeof(mask_entry));
+	memset(&mask_key, 0x0, sizeof(mask_key));
+	memset(&temp_key, 0x0, sizeof(temp_key));
+	mask_entry.addr[0] = 0x01;
+	hns_dsaf_set_mac_key(dsaf_dev, &mask_key, mask_entry.in_vlan_id,
+			     0xf, mask_entry.addr);
+	tbl_tcam_mcast.tbl_mcast_item_vld = 1;
+	tbl_tcam_mcast.tbl_mcast_old_en = 0;
+
+	/* set MAC port to handle multicast */
+	mskid = hns_dsaf_get_port_id(port);
+	if (mskid == -EINVAL) {
+		dev_err(dsaf_dev->dev, "%s,pnum(%d)error,key(%#x:%#x)\n",
+			dsaf_dev->ae_dev.name, port,
+			mask_key.high.val, mask_key.low.val);
+		return;
+	}
+	dsaf_set_bit(tbl_tcam_mcast.tbl_mcast_port_msk[mskid / 32],
+		     mskid % 32, 1);
+
+	/* set pool bit map to handle multicast */
+	mskid = hns_dsaf_get_port_id(port_num);
+	if (mskid == -EINVAL) {
+		dev_err(dsaf_dev->dev,
+			"%s, pool bit map pnum(%d)error,key(%#x:%#x)\n",
+			dsaf_dev->ae_dev.name, port_num,
+			mask_key.high.val, mask_key.low.val);
+		return;
+	}
+	dsaf_set_bit(tbl_tcam_mcast.tbl_mcast_port_msk[mskid / 32],
+		     mskid % 32, 1);
+
+	memcpy(&temp_key, &mask_key, sizeof(mask_key));
+	hns_dsaf_tcam_mc_cfg_vague(dsaf_dev, entry_index, &tbl_tcam_data_mc,
+				   (struct dsaf_tbl_tcam_data *)(&mask_key),
+				   &tbl_tcam_mcast);
+
+	/* update software entry */
+	soft_mac_entry += entry_index;
+	soft_mac_entry->index = entry_index;
+	soft_mac_entry->tcam_key.high.val = temp_key.high.val;
+	soft_mac_entry->tcam_key.low.val = temp_key.low.val;
+}
+
+static void set_promisc_tcam_disable(struct dsaf_device *dsaf_dev, u32 port)
+{
+	struct dsaf_tbl_tcam_data tbl_tcam_data_mc = {0x01000000, port};
+	struct dsaf_tbl_tcam_ucast_cfg tbl_tcam_ucast = {0, 0, 0, 0, 0};
+	struct dsaf_tbl_tcam_mcast_cfg tbl_tcam_mcast = {0, 0, {0} };
+	struct dsaf_drv_priv *priv = hns_dsaf_dev_priv(dsaf_dev);
+	struct dsaf_tbl_tcam_data tbl_tcam_data_uc = {0, 0};
+	struct dsaf_tbl_tcam_data tbl_tcam_mask = {0, 0};
+	struct dsaf_drv_soft_mac_tbl *soft_mac_entry;
+	u16 entry_index;
+	struct dsaf_drv_tbl_tcam_key mac_key;
+	u8 addr[ETH_ALEN] = {0};
+
+	/* 1. delete uc vague tcam entry. */
+	/* promisc use vague table match with vlanid = 0 & macaddr = 0 */
+	hns_dsaf_set_mac_key(dsaf_dev, &mac_key, 0x00, port, addr);
+	entry_index = hns_dsaf_find_soft_mac_entry(dsaf_dev, &mac_key);
+
+	if (entry_index == DSAF_INVALID_ENTRY_IDX)
+		return;
+
+	/* config uc vague table */
+	hns_dsaf_tcam_uc_cfg_vague(dsaf_dev, entry_index, &tbl_tcam_data_uc,
+				   &tbl_tcam_mask, &tbl_tcam_ucast);
+	/* update soft management table. */
+	soft_mac_entry = priv->soft_mac_tbl;
+	soft_mac_entry += entry_index;
+	soft_mac_entry->index = DSAF_INVALID_ENTRY_IDX;
+	/* step back to the START for mc. */
+	soft_mac_entry = priv->soft_mac_tbl;
+
+	/* 2. delete mc vague tcam entry. */
+	addr[0] = 0x01;
+	memset(&mac_key, 0x0, sizeof(mac_key));
+	hns_dsaf_set_mac_key(dsaf_dev, &mac_key, 0x00, port, addr);
+	entry_index = hns_dsaf_find_soft_mac_entry(dsaf_dev, &mac_key);
+
+	if (entry_index == DSAF_INVALID_ENTRY_IDX)
+		return;
+
+	/* config mc vague table */
+	hns_dsaf_tcam_mc_cfg_vague(dsaf_dev, entry_index, &tbl_tcam_data_mc,
+				   &tbl_tcam_mask, &tbl_tcam_mcast);
+	/* update soft management table. */
+	soft_mac_entry += entry_index;
+	soft_mac_entry->index = DSAF_INVALID_ENTRY_IDX;
+}
+
+/* Reserve the last TCAM entry for promisc support */
+void hns_dsaf_set_promisc_tcam(struct dsaf_device *dsaf_dev,
+			       u32 port, bool enable)
+{
+	if (enable)
+		set_promisc_tcam_enable(dsaf_dev, port);
+	else
+		set_promisc_tcam_disable(dsaf_dev, port);
+}
+
+int hns_dsaf_wait_pkt_clean(struct dsaf_device *dsaf_dev, int port)
+{
+	u32 val, val_tmp;
+	int wait_cnt;
+
+	if (port >= DSAF_SERVICE_NW_NUM)
+		return 0;
+
+	wait_cnt = 0;
+	while (wait_cnt++ < HNS_MAX_WAIT_CNT) {
+		val = dsaf_read_dev(dsaf_dev, DSAF_VOQ_IN_PKT_NUM_0_REG +
+			(port + DSAF_XGE_NUM) * 0x40);
+		val_tmp = dsaf_read_dev(dsaf_dev, DSAF_VOQ_OUT_PKT_NUM_0_REG +
+			(port + DSAF_XGE_NUM) * 0x40);
+		if (val == val_tmp)
+			break;
+
+		usleep_range(100, 200);
+	}
+
+	if (wait_cnt >= HNS_MAX_WAIT_CNT) {
+		dev_err(dsaf_dev->dev, "hns dsaf clean wait timeout(%u - %u).\n",
+			val, val_tmp);
+		return -EBUSY;
+	}
+
+	return 0;
 }
 
 /**
- * dsaf_probe - probo dsaf dev
+ * hns_dsaf_probe - probo dsaf dev
  * @pdev: dasf platform device
- * retuen 0 - success , negative --fail
+ * return 0 - success , negative --fail
  */
 static int hns_dsaf_probe(struct platform_device *pdev)
 {
@@ -2776,10 +2982,10 @@ free_dev:
 }
 
 /**
- * dsaf_remove - remove dsaf dev
+ * hns_dsaf_remove - remove dsaf dev
  * @pdev: dasf platform device
  */
-static int hns_dsaf_remove(struct platform_device *pdev)
+static void hns_dsaf_remove(struct platform_device *pdev)
 {
 	struct dsaf_device *dsaf_dev = dev_get_drvdata(&pdev->dev);
 
@@ -2792,8 +2998,6 @@ static int hns_dsaf_remove(struct platform_device *pdev)
 	hns_dsaf_free(dsaf_dev);
 
 	hns_dsaf_free_dev(dsaf_dev);
-
-	return 0;
 }
 
 static const struct of_device_id g_dsaf_match[] = {
@@ -2815,110 +3019,6 @@ static struct platform_driver g_dsaf_driver = {
 
 module_platform_driver(g_dsaf_driver);
 
-/**
- * hns_dsaf_roce_reset - reset dsaf and roce
- * @dsaf_fwnode: Pointer to framework node for the dasf
- * @enable: false - request reset , true - drop reset
- * retuen 0 - success , negative -fail
- */
-int hns_dsaf_roce_reset(struct fwnode_handle *dsaf_fwnode, bool dereset)
-{
-	struct dsaf_device *dsaf_dev;
-	struct platform_device *pdev;
-	u32 mp;
-	u32 sl;
-	u32 credit;
-	int i;
-	const u32 port_map[DSAF_ROCE_CREDIT_CHN][DSAF_ROCE_CHAN_MODE_NUM] = {
-		{DSAF_ROCE_PORT_0, DSAF_ROCE_PORT_0, DSAF_ROCE_PORT_0},
-		{DSAF_ROCE_PORT_1, DSAF_ROCE_PORT_0, DSAF_ROCE_PORT_0},
-		{DSAF_ROCE_PORT_2, DSAF_ROCE_PORT_1, DSAF_ROCE_PORT_0},
-		{DSAF_ROCE_PORT_3, DSAF_ROCE_PORT_1, DSAF_ROCE_PORT_0},
-		{DSAF_ROCE_PORT_4, DSAF_ROCE_PORT_2, DSAF_ROCE_PORT_1},
-		{DSAF_ROCE_PORT_4, DSAF_ROCE_PORT_2, DSAF_ROCE_PORT_1},
-		{DSAF_ROCE_PORT_5, DSAF_ROCE_PORT_3, DSAF_ROCE_PORT_1},
-		{DSAF_ROCE_PORT_5, DSAF_ROCE_PORT_3, DSAF_ROCE_PORT_1},
-	};
-	const u32 sl_map[DSAF_ROCE_CREDIT_CHN][DSAF_ROCE_CHAN_MODE_NUM] = {
-		{DSAF_ROCE_SL_0, DSAF_ROCE_SL_0, DSAF_ROCE_SL_0},
-		{DSAF_ROCE_SL_0, DSAF_ROCE_SL_1, DSAF_ROCE_SL_1},
-		{DSAF_ROCE_SL_0, DSAF_ROCE_SL_0, DSAF_ROCE_SL_2},
-		{DSAF_ROCE_SL_0, DSAF_ROCE_SL_1, DSAF_ROCE_SL_3},
-		{DSAF_ROCE_SL_0, DSAF_ROCE_SL_0, DSAF_ROCE_SL_0},
-		{DSAF_ROCE_SL_1, DSAF_ROCE_SL_1, DSAF_ROCE_SL_1},
-		{DSAF_ROCE_SL_0, DSAF_ROCE_SL_0, DSAF_ROCE_SL_2},
-		{DSAF_ROCE_SL_1, DSAF_ROCE_SL_1, DSAF_ROCE_SL_3},
-	};
-
-	/* find the platform device corresponding to fwnode */
-	if (is_of_node(dsaf_fwnode)) {
-		pdev = of_find_device_by_node(to_of_node(dsaf_fwnode));
-	} else if (is_acpi_device_node(dsaf_fwnode)) {
-		pdev = hns_dsaf_find_platform_device(dsaf_fwnode);
-	} else {
-		pr_err("fwnode is neither OF or ACPI type\n");
-		return -EINVAL;
-	}
-
-	/* check if we were a success in fetching pdev */
-	if (!pdev) {
-		pr_err("couldn't find platform device for node\n");
-		return -ENODEV;
-	}
-
-	/* retrieve the dsaf_device from the driver data */
-	dsaf_dev = dev_get_drvdata(&pdev->dev);
-	if (!dsaf_dev) {
-		dev_err(&pdev->dev, "dsaf_dev is NULL\n");
-		return -ENODEV;
-	}
-
-	/* now, make sure we are running on compatible SoC */
-	if (AE_IS_VER1(dsaf_dev->dsaf_ver)) {
-		dev_err(dsaf_dev->dev, "%s v1 chip doesn't support RoCE!\n",
-			dsaf_dev->ae_dev.name);
-		return -ENODEV;
-	}
-
-	/* do reset or de-reset according to the flag */
-	if (!dereset) {
-		/* reset rocee-channels in dsaf and rocee */
-		dsaf_dev->misc_op->hns_dsaf_srst_chns(dsaf_dev, DSAF_CHNS_MASK,
-						      false);
-		dsaf_dev->misc_op->hns_dsaf_roce_srst(dsaf_dev, false);
-	} else {
-		/* configure dsaf tx roce correspond to port map and sl map */
-		mp = dsaf_read_dev(dsaf_dev, DSAF_ROCE_PORT_MAP_REG);
-		for (i = 0; i < DSAF_ROCE_CREDIT_CHN; i++)
-			dsaf_set_field(mp, 7 << i * 3, i * 3,
-				       port_map[i][DSAF_ROCE_6PORT_MODE]);
-		dsaf_set_field(mp, 3 << i * 3, i * 3, 0);
-		dsaf_write_dev(dsaf_dev, DSAF_ROCE_PORT_MAP_REG, mp);
-
-		sl = dsaf_read_dev(dsaf_dev, DSAF_ROCE_SL_MAP_REG);
-		for (i = 0; i < DSAF_ROCE_CREDIT_CHN; i++)
-			dsaf_set_field(sl, 3 << i * 2, i * 2,
-				       sl_map[i][DSAF_ROCE_6PORT_MODE]);
-		dsaf_write_dev(dsaf_dev, DSAF_ROCE_SL_MAP_REG, sl);
-
-		/* de-reset rocee-channels in dsaf and rocee */
-		dsaf_dev->misc_op->hns_dsaf_srst_chns(dsaf_dev, DSAF_CHNS_MASK,
-						      true);
-		msleep(SRST_TIME_INTERVAL);
-		dsaf_dev->misc_op->hns_dsaf_roce_srst(dsaf_dev, true);
-
-		/* enable dsaf channel rocee credit */
-		credit = dsaf_read_dev(dsaf_dev, DSAF_SBM_ROCEE_CFG_REG_REG);
-		dsaf_set_bit(credit, DSAF_SBM_ROCEE_CFG_CRD_EN_B, 0);
-		dsaf_write_dev(dsaf_dev, DSAF_SBM_ROCEE_CFG_REG_REG, credit);
-
-		dsaf_set_bit(credit, DSAF_SBM_ROCEE_CFG_CRD_EN_B, 1);
-		dsaf_write_dev(dsaf_dev, DSAF_SBM_ROCEE_CFG_REG_REG, credit);
-	}
-	return 0;
-}
-EXPORT_SYMBOL(hns_dsaf_roce_reset);
-
 MODULE_LICENSE("GPL");
 MODULE_AUTHOR("Huawei Tech. Co., Ltd.");
 MODULE_DESCRIPTION("HNS DSAF driver");
diff --git a/drivers/net/ethernet/hisilicon/hns/hns_dsaf_main.h b/drivers/net/ethernet/hisilicon/hns/hns_dsaf_main.h
index 4507e8222683..653dfbb25d1b 100644
--- a/drivers/net/ethernet/hisilicon/hns/hns_dsaf_main.h
+++ b/drivers/net/ethernet/hisilicon/hns/hns_dsaf_main.h
@@ -1,10 +1,6 @@
+/* SPDX-License-Identifier: GPL-2.0-or-later */
 /*
  * Copyright (c) 2014-2015 Hisilicon Limited.
- *
- * This program is free software; you can redistribute it and/or modify
- * it under the terms of the GNU General Public License as published by
- * the Free Software Foundation; either version 2 of the License, or
- * (at your option) any later version.
  */
 
 #ifndef __HNS_DSAF_MAIN_H
@@ -44,28 +40,7 @@ struct hns_mac_cb;
 #define DSAF_ROCE_CREDIT_CHN	8
 #define DSAF_ROCE_CHAN_MODE	3
 
-enum dsaf_roce_port_mode {
-	DSAF_ROCE_6PORT_MODE,
-	DSAF_ROCE_4PORT_MODE,
-	DSAF_ROCE_2PORT_MODE,
-	DSAF_ROCE_CHAN_MODE_NUM,
-};
-
-enum dsaf_roce_port_num {
-	DSAF_ROCE_PORT_0,
-	DSAF_ROCE_PORT_1,
-	DSAF_ROCE_PORT_2,
-	DSAF_ROCE_PORT_3,
-	DSAF_ROCE_PORT_4,
-	DSAF_ROCE_PORT_5,
-};
-
-enum dsaf_roce_qos_sl {
-	DSAF_ROCE_SL_0,
-	DSAF_ROCE_SL_1,
-	DSAF_ROCE_SL_2,
-	DSAF_ROCE_SL_3,
-};
+#define HNS_MAX_WAIT_CNT 10000
 
 #define DSAF_STATS_READ(p, offset) (*((u64 *)((u8 *)(p) + (offset))))
 #define HNS_DSAF_IS_DEBUG(dev) ((dev)->dsaf_mode == DSAF_MODE_DISABLE_SP)
@@ -212,7 +187,7 @@ struct hnae_vf_cb {
 	u8 port_index;
 	struct hns_mac_cb *mac_cb;
 	struct dsaf_device *dsaf_dev;
-	struct hnae_handle  ae_handle; /* must be the last number */
+	struct hnae_handle  ae_handle; /* must be the last member */
 };
 
 struct dsaf_int_xge_src {
@@ -309,9 +284,6 @@ struct dsaf_misc_op {
 	void (*ge_srst)(struct dsaf_device *dsaf_dev, u32 port, bool dereset);
 	void (*ppe_srst)(struct dsaf_device *dsaf_dev, u32 port, bool dereset);
 	void (*ppe_comm_srst)(struct dsaf_device *dsaf_dev, bool dereset);
-	void (*hns_dsaf_srst_chns)(struct dsaf_device *dsaf_dev, u32 msk,
-				   bool dereset);
-	void (*hns_dsaf_roce_srst)(struct dsaf_device *dsaf_dev, bool dereset);
 
 	phy_interface_t (*get_phy_if)(struct hns_mac_cb *mac_cb);
 	int (*get_sfp_prsnt)(struct hns_mac_cb *mac_cb, int *sfp_prsnt);
@@ -444,7 +416,7 @@ void hns_dsaf_update_stats(struct dsaf_device *dsaf_dev, u32 inode_num);
 
 int hns_dsaf_get_sset_count(struct dsaf_device *dsaf_dev, int stringset);
 void hns_dsaf_get_stats(struct dsaf_device *ddev, u64 *data, int port);
-void hns_dsaf_get_strings(int stringset, u8 *data, int port,
+void hns_dsaf_get_strings(int stringset, u8 **data, int port,
 			  struct dsaf_device *dsaf_dev);
 
 void hns_dsaf_get_regs(struct dsaf_device *ddev, u32 port, void *data);
@@ -463,5 +435,6 @@ int hns_dsaf_rm_mac_addr(
 
 int hns_dsaf_clr_mac_mc_port(struct dsaf_device *dsaf_dev,
 			     u8 mac_id, u8 port_num);
+int hns_dsaf_wait_pkt_clean(struct dsaf_device *dsaf_dev, int port);
 
 #endif /* __HNS_DSAF_MAIN_H__ */
diff --git a/drivers/net/ethernet/hisilicon/hns/hns_dsaf_misc.c b/drivers/net/ethernet/hisilicon/hns/hns_dsaf_misc.c
index e13aa064a8e9..91391a49fcea 100644
--- a/drivers/net/ethernet/hisilicon/hns/hns_dsaf_misc.c
+++ b/drivers/net/ethernet/hisilicon/hns/hns_dsaf_misc.c
@@ -1,10 +1,6 @@
+// SPDX-License-Identifier: GPL-2.0-or-later
 /*
  * Copyright (c) 2014-2015 Hisilicon Limited.
- *
- * This program is free software; you can redistribute it and/or modify
- * it under the terms of the GNU General Public License as published by
- * the Free Software Foundation; either version 2 of the License, or
- * (at your option) any later version.
  */
 
 #include "hns_dsaf_mac.h"
@@ -18,6 +14,7 @@ enum _dsm_op_index {
 	HNS_OP_LED_SET_FUNC             = 0x3,
 	HNS_OP_GET_PORT_TYPE_FUNC       = 0x4,
 	HNS_OP_GET_SFP_STAT_FUNC        = 0x5,
+	HNS_OP_LOCATE_LED_SET_FUNC      = 0x6,
 };
 
 enum _dsm_rst_type {
@@ -29,10 +26,9 @@ enum _dsm_rst_type {
 	HNS_ROCE_RESET_FUNC     = 0x7,
 };
 
-const u8 hns_dsaf_acpi_dsm_uuid[] = {
-	0x1A, 0xAA, 0x85, 0x1A, 0x93, 0xE2, 0x5E, 0x41,
-	0x8E, 0x28, 0x8D, 0x69, 0x0A, 0x0F, 0x82, 0x0A
-};
+static const guid_t hns_dsaf_acpi_dsm_guid =
+	GUID_INIT(0x1A85AA1A, 0xE293, 0x415E,
+		  0x8E, 0x28, 0x8D, 0x69, 0x0A, 0x0F, 0x82, 0x0A);
 
 static void dsaf_write_sub(struct dsaf_device *dsaf_dev, u32 reg, u32 val)
 {
@@ -44,16 +40,76 @@ static void dsaf_write_sub(struct dsaf_device *dsaf_dev, u32 reg, u32 val)
 
 static u32 dsaf_read_sub(struct dsaf_device *dsaf_dev, u32 reg)
 {
-	u32 ret;
-
-	if (dsaf_dev->sub_ctrl)
-		ret = dsaf_read_syscon(dsaf_dev->sub_ctrl, reg);
-	else
+	u32 ret = 0;
+	int err;
+
+	if (dsaf_dev->sub_ctrl) {
+		err = dsaf_read_syscon(dsaf_dev->sub_ctrl, reg, &ret);
+		if (err)
+			dev_err(dsaf_dev->dev, "dsaf_read_syscon error %d!\n",
+				err);
+	} else {
 		ret = dsaf_read_reg(dsaf_dev->sc_base, reg);
+	}
 
 	return ret;
 }
 
+static void hns_dsaf_acpi_ledctrl_by_port(struct hns_mac_cb *mac_cb, u8 op_type,
+					  u32 link, u32 port, u32 act)
+{
+	union acpi_object *obj;
+	union acpi_object obj_args[3], argv4;
+
+	obj_args[0].integer.type = ACPI_TYPE_INTEGER;
+	obj_args[0].integer.value = link;
+	obj_args[1].integer.type = ACPI_TYPE_INTEGER;
+	obj_args[1].integer.value = port;
+	obj_args[2].integer.type = ACPI_TYPE_INTEGER;
+	obj_args[2].integer.value = act;
+
+	argv4.type = ACPI_TYPE_PACKAGE;
+	argv4.package.count = 3;
+	argv4.package.elements = obj_args;
+
+	obj = acpi_evaluate_dsm(ACPI_HANDLE(mac_cb->dev),
+				&hns_dsaf_acpi_dsm_guid, 0, op_type, &argv4);
+	if (!obj) {
+		dev_warn(mac_cb->dev, "ledctrl fail, link:%d port:%d act:%d!\n",
+			 link, port, act);
+		return;
+	}
+
+	ACPI_FREE(obj);
+}
+
+static void hns_dsaf_acpi_locate_ledctrl_by_port(struct hns_mac_cb *mac_cb,
+						 u8 op_type, u32 locate,
+						 u32 port)
+{
+	union acpi_object obj_args[2], argv4;
+	union acpi_object *obj;
+
+	obj_args[0].integer.type = ACPI_TYPE_INTEGER;
+	obj_args[0].integer.value = locate;
+	obj_args[1].integer.type = ACPI_TYPE_INTEGER;
+	obj_args[1].integer.value = port;
+
+	argv4.type = ACPI_TYPE_PACKAGE;
+	argv4.package.count = 2;
+	argv4.package.elements = obj_args;
+
+	obj = acpi_evaluate_dsm(ACPI_HANDLE(mac_cb->dev),
+				&hns_dsaf_acpi_dsm_guid, 0, op_type, &argv4);
+	if (!obj) {
+		dev_err(mac_cb->dev, "ledctrl fail, locate:%d port:%d!\n",
+			locate, port);
+		return;
+	}
+
+	ACPI_FREE(obj);
+}
+
 static void hns_cpld_set_led(struct hns_mac_cb *mac_cb, int link_status,
 			     u16 speed, int data)
 {
@@ -94,6 +150,18 @@ static void hns_cpld_set_led(struct hns_mac_cb *mac_cb, int link_status,
 	}
 }
 
+static void hns_cpld_set_led_acpi(struct hns_mac_cb *mac_cb, int link_status,
+				  u16 speed, int data)
+{
+	if (!mac_cb) {
+		pr_err("cpld_led_set mac_cb is null!\n");
+		return;
+	}
+
+	hns_dsaf_acpi_ledctrl_by_port(mac_cb, HNS_OP_LED_SET_FUNC,
+				      link_status, mac_cb->mac_id, data);
+}
+
 static void cpld_led_reset(struct hns_mac_cb *mac_cb)
 {
 	if (!mac_cb || !mac_cb->cpld_ctrl)
@@ -104,18 +172,40 @@ static void cpld_led_reset(struct hns_mac_cb *mac_cb)
 	mac_cb->cpld_led_value = CPLD_LED_DEFAULT_VALUE;
 }
 
+static void cpld_led_reset_acpi(struct hns_mac_cb *mac_cb)
+{
+	if (!mac_cb) {
+		pr_err("cpld_led_reset mac_cb is null!\n");
+		return;
+	}
+
+	if (mac_cb->media_type != HNAE_MEDIA_TYPE_FIBER)
+		return;
+
+	hns_dsaf_acpi_ledctrl_by_port(mac_cb, HNS_OP_LED_SET_FUNC,
+				      0, mac_cb->mac_id, 0);
+}
+
 static int cpld_set_led_id(struct hns_mac_cb *mac_cb,
 			   enum hnae_led_state status)
 {
+	u32 val = 0;
+	int ret;
+
+	if (!mac_cb->cpld_ctrl)
+		return 0;
+
 	switch (status) {
 	case HNAE_LED_ACTIVE:
-		mac_cb->cpld_led_value =
-			dsaf_read_syscon(mac_cb->cpld_ctrl,
-					 mac_cb->cpld_ctrl_reg);
-		dsaf_set_bit(mac_cb->cpld_led_value, DSAF_LED_ANCHOR_B,
-			     CPLD_LED_ON_VALUE);
+		ret = dsaf_read_syscon(mac_cb->cpld_ctrl, mac_cb->cpld_ctrl_reg,
+				       &val);
+		if (ret)
+			return ret;
+
+		dsaf_set_bit(val, DSAF_LED_ANCHOR_B, CPLD_LED_ON_VALUE);
 		dsaf_write_syscon(mac_cb->cpld_ctrl, mac_cb->cpld_ctrl_reg,
-				  mac_cb->cpld_led_value);
+				  val);
+		mac_cb->cpld_led_value = val;
 		break;
 	case HNAE_LED_INACTIVE:
 		dsaf_set_bit(mac_cb->cpld_led_value, DSAF_LED_ANCHOR_B,
@@ -131,6 +221,30 @@ static int cpld_set_led_id(struct hns_mac_cb *mac_cb,
 	return 0;
 }
 
+static int cpld_set_led_id_acpi(struct hns_mac_cb *mac_cb,
+				enum hnae_led_state status)
+{
+	switch (status) {
+	case HNAE_LED_ACTIVE:
+		hns_dsaf_acpi_locate_ledctrl_by_port(mac_cb,
+						     HNS_OP_LOCATE_LED_SET_FUNC,
+						     CPLD_LED_ON_VALUE,
+						     mac_cb->mac_id);
+		break;
+	case HNAE_LED_INACTIVE:
+		hns_dsaf_acpi_locate_ledctrl_by_port(mac_cb,
+						     HNS_OP_LOCATE_LED_SET_FUNC,
+						     CPLD_LED_DEFAULT_VALUE,
+						     mac_cb->mac_id);
+		break;
+	default:
+		dev_err(mac_cb->dev, "invalid led state: %d!", status);
+		return -EINVAL;
+	}
+
+	return 0;
+}
+
 #define RESET_REQ_OR_DREQ 1
 
 static void hns_dsaf_acpi_srst_by_port(struct dsaf_device *dsaf_dev, u8 op_type,
@@ -151,7 +265,7 @@ static void hns_dsaf_acpi_srst_by_port(struct dsaf_device *dsaf_dev, u8 op_type,
 	argv4.package.elements = obj_args;
 
 	obj = acpi_evaluate_dsm(ACPI_HANDLE(dsaf_dev->dev),
-				hns_dsaf_acpi_dsm_uuid, 0, op_type, &argv4);
+				&hns_dsaf_acpi_dsm_guid, 0, op_type, &argv4);
 	if (!obj) {
 		dev_warn(dsaf_dev->dev, "reset port_type%d port%d fail!",
 			 port_type, port);
@@ -212,66 +326,6 @@ static void hns_dsaf_xge_srst_by_port_acpi(struct dsaf_device *dsaf_dev,
 				   HNS_XGE_RESET_FUNC, port, dereset);
 }
 
-/**
- * hns_dsaf_srst_chns - reset dsaf channels
- * @dsaf_dev: dsaf device struct pointer
- * @msk: xbar channels mask value:
- * bit0-5 for xge0-5
- * bit6-11 for ppe0-5
- * bit12-17 for roce0-5
- * bit18-19 for com/dfx
- * @enable: false - request reset , true - drop reset
- */
-void hns_dsaf_srst_chns(struct dsaf_device *dsaf_dev, u32 msk, bool dereset)
-{
-	u32 reg_addr;
-
-	if (!dereset)
-		reg_addr = DSAF_SUB_SC_DSAF_RESET_REQ_REG;
-	else
-		reg_addr = DSAF_SUB_SC_DSAF_RESET_DREQ_REG;
-
-	dsaf_write_sub(dsaf_dev, reg_addr, msk);
-}
-
-/**
- * hns_dsaf_srst_chns - reset dsaf channels
- * @dsaf_dev: dsaf device struct pointer
- * @msk: xbar channels mask value:
- * bit0-5 for xge0-5
- * bit6-11 for ppe0-5
- * bit12-17 for roce0-5
- * bit18-19 for com/dfx
- * @enable: false - request reset , true - drop reset
- */
-void
-hns_dsaf_srst_chns_acpi(struct dsaf_device *dsaf_dev, u32 msk, bool dereset)
-{
-	hns_dsaf_acpi_srst_by_port(dsaf_dev, HNS_OP_RESET_FUNC,
-				   HNS_DSAF_CHN_RESET_FUNC,
-				   msk, dereset);
-}
-
-void hns_dsaf_roce_srst(struct dsaf_device *dsaf_dev, bool dereset)
-{
-	if (!dereset) {
-		dsaf_write_sub(dsaf_dev, DSAF_SUB_SC_ROCEE_RESET_REQ_REG, 1);
-	} else {
-		dsaf_write_sub(dsaf_dev,
-			       DSAF_SUB_SC_ROCEE_CLK_DIS_REG, 1);
-		dsaf_write_sub(dsaf_dev,
-			       DSAF_SUB_SC_ROCEE_RESET_DREQ_REG, 1);
-		msleep(20);
-		dsaf_write_sub(dsaf_dev, DSAF_SUB_SC_ROCEE_CLK_EN_REG, 1);
-	}
-}
-
-void hns_dsaf_roce_srst_acpi(struct dsaf_device *dsaf_dev, bool dereset)
-{
-	hns_dsaf_acpi_srst_by_port(dsaf_dev, HNS_OP_RESET_FUNC,
-				   HNS_ROCE_RESET_FUNC, 0, dereset);
-}
-
 static void hns_dsaf_ge_srst_by_port(struct dsaf_device *dsaf_dev, u32 port,
 				     bool dereset)
 {
@@ -283,6 +337,10 @@ static void hns_dsaf_ge_srst_by_port(struct dsaf_device *dsaf_dev, u32 port,
 		return;
 
 	if (!HNS_DSAF_IS_DEBUG(dsaf_dev)) {
+		/* DSAF_MAX_PORT_NUM is 6, but DSAF_GE_NUM is 8.
+		   We need check to prevent array overflow */
+		if (port >= DSAF_MAX_PORT_NUM)
+			return;
 		reg_val_1  = 0x1 << port;
 		port_rst_off = dsaf_dev->mac_cb[port]->port_rst_off;
 		/* there is difference between V1 and V2 in register.*/
@@ -384,7 +442,7 @@ static void hns_ppe_com_srst(struct dsaf_device *dsaf_dev, bool dereset)
 }
 
 /**
- * hns_mac_get_sds_mode - get phy ifterface form serdes mode
+ * hns_mac_get_phy_if - get phy ifterface form serdes mode
  * @mac_cb: mac control block
  * retuen phy interface
  */
@@ -404,7 +462,7 @@ static phy_interface_t hns_mac_get_phy_if(struct hns_mac_cb *mac_cb)
 			reg = HNS_MAC_HILINK4_REG;
 		else
 			reg = HNS_MAC_HILINK3_REG;
-	} else{
+	} else {
 		if (!HNS_DSAF_IS_DEBUG(mac_cb->dsaf_dev) && mac_id <= 3)
 			reg = HNS_MAC_HILINK4V2_REG;
 		else
@@ -429,15 +487,15 @@ static phy_interface_t hns_mac_get_phy_if_acpi(struct hns_mac_cb *mac_cb)
 	obj_args.integer.type = ACPI_TYPE_INTEGER;
 	obj_args.integer.value = mac_cb->mac_id;
 
-	argv4.type = ACPI_TYPE_PACKAGE,
-	argv4.package.count = 1,
-	argv4.package.elements = &obj_args,
-
-	obj = acpi_evaluate_dsm(ACPI_HANDLE(mac_cb->dev),
-				hns_dsaf_acpi_dsm_uuid, 0,
-				HNS_OP_GET_PORT_TYPE_FUNC, &argv4);
-
-	if (!obj || obj->type != ACPI_TYPE_INTEGER)
+	argv4.type = ACPI_TYPE_PACKAGE;
+	argv4.package.count = 1;
+	argv4.package.elements = &obj_args;
+
+	obj = acpi_evaluate_dsm_typed(ACPI_HANDLE(mac_cb->dev),
+				      &hns_dsaf_acpi_dsm_guid, 0,
+				      HNS_OP_GET_PORT_TYPE_FUNC, &argv4,
+				      ACPI_TYPE_INTEGER);
+	if (!obj)
 		return phy_if;
 
 	phy_if = obj->integer.value ?
@@ -450,18 +508,25 @@ static phy_interface_t hns_mac_get_phy_if_acpi(struct hns_mac_cb *mac_cb)
 	return phy_if;
 }
 
-int hns_mac_get_sfp_prsnt(struct hns_mac_cb *mac_cb, int *sfp_prsnt)
+static int hns_mac_get_sfp_prsnt(struct hns_mac_cb *mac_cb, int *sfp_prsnt)
 {
+	u32 val = 0;
+	int ret;
+
 	if (!mac_cb->cpld_ctrl)
 		return -ENODEV;
 
-	*sfp_prsnt = !dsaf_read_syscon(mac_cb->cpld_ctrl, mac_cb->cpld_ctrl_reg
-					+ MAC_SFP_PORT_OFFSET);
+	ret = dsaf_read_syscon(mac_cb->cpld_ctrl,
+			       mac_cb->cpld_ctrl_reg + MAC_SFP_PORT_OFFSET,
+			       &val);
+	if (ret)
+		return ret;
 
+	*sfp_prsnt = !val;
 	return 0;
 }
 
-int hns_mac_get_sfp_prsnt_acpi(struct hns_mac_cb *mac_cb, int *sfp_prsnt)
+static int hns_mac_get_sfp_prsnt_acpi(struct hns_mac_cb *mac_cb, int *sfp_prsnt)
 {
 	union acpi_object *obj;
 	union acpi_object obj_args, argv4;
@@ -469,15 +534,15 @@ int hns_mac_get_sfp_prsnt_acpi(struct hns_mac_cb *mac_cb, int *sfp_prsnt)
 	obj_args.integer.type = ACPI_TYPE_INTEGER;
 	obj_args.integer.value = mac_cb->mac_id;
 
-	argv4.type = ACPI_TYPE_PACKAGE,
-	argv4.package.count = 1,
-	argv4.package.elements = &obj_args,
-
-	obj = acpi_evaluate_dsm(ACPI_HANDLE(mac_cb->dev),
-				hns_dsaf_acpi_dsm_uuid, 0,
-				HNS_OP_GET_SFP_STAT_FUNC, &argv4);
-
-	if (!obj || obj->type != ACPI_TYPE_INTEGER)
+	argv4.type = ACPI_TYPE_PACKAGE;
+	argv4.package.count = 1;
+	argv4.package.elements = &obj_args;
+
+	obj = acpi_evaluate_dsm_typed(ACPI_HANDLE(mac_cb->dev),
+				      &hns_dsaf_acpi_dsm_guid, 0,
+				      HNS_OP_GET_SFP_STAT_FUNC, &argv4,
+				      ACPI_TYPE_INTEGER);
+	if (!obj)
 		return -ENODEV;
 
 	*sfp_prsnt = obj->integer.value;
@@ -490,7 +555,8 @@ int hns_mac_get_sfp_prsnt_acpi(struct hns_mac_cb *mac_cb, int *sfp_prsnt)
 /**
  * hns_mac_config_sds_loopback - set loop back for serdes
  * @mac_cb: mac control block
- * retuen 0 == success
+ * @en: enable or disable
+ * return 0 == success
  */
 static int hns_mac_config_sds_loopback(struct hns_mac_cb *mac_cb, bool en)
 {
@@ -507,7 +573,7 @@ static int hns_mac_config_sds_loopback(struct hns_mac_cb *mac_cb, bool en)
 #define RX_CSR(lane, reg) ((0x4080 + (reg) * 0x0002 + (lane) * 0x0200) * 2)
 	u64 reg_offset = RX_CSR(lane_id[mac_cb->mac_id], 0);
 
-	int sfp_prsnt;
+	int sfp_prsnt = 0;
 	int ret = hns_mac_get_sfp_prsnt(mac_cb, &sfp_prsnt);
 
 	if (!mac_cb->phy_dev) {
@@ -519,7 +585,7 @@ static int hns_mac_config_sds_loopback(struct hns_mac_cb *mac_cb, bool en)
 	}
 
 	if (mac_cb->serdes_ctrl) {
-		u32 origin;
+		u32 origin = 0;
 
 		if (!AE_IS_VER1(mac_cb->dsaf_dev->dsaf_ver)) {
 #define HILINK_ACCESS_SEL_CFG		0x40008
@@ -536,12 +602,15 @@ static int hns_mac_config_sds_loopback(struct hns_mac_cb *mac_cb, bool en)
 						  HILINK_ACCESS_SEL_CFG, 3);
 		}
 
-		origin = dsaf_read_syscon(mac_cb->serdes_ctrl, reg_offset);
+		ret = dsaf_read_syscon(mac_cb->serdes_ctrl, reg_offset,
+				       &origin);
+		if (ret)
+			return ret;
 
 		dsaf_set_field(origin, 1ull << 10, 10, en);
 		dsaf_write_syscon(mac_cb->serdes_ctrl, reg_offset, origin);
 	} else {
-		u8 *base_addr = (u8 *)mac_cb->serdes_vaddr +
+		u8 __iomem *base_addr = mac_cb->serdes_vaddr +
 				(mac_cb->mac_id <= 3 ? 0x00280000 : 0x00200000);
 		dsaf_set_reg_field(base_addr, reg_offset, 1ull << 10, 10, en);
 	}
@@ -558,14 +627,14 @@ hns_mac_config_sds_loopback_acpi(struct hns_mac_cb *mac_cb, bool en)
 	obj_args[0].integer.type = ACPI_TYPE_INTEGER;
 	obj_args[0].integer.value = mac_cb->mac_id;
 	obj_args[1].integer.type = ACPI_TYPE_INTEGER;
-	obj_args[1].integer.value = !!en;
+	obj_args[1].integer.value = en;
 
 	argv4.type = ACPI_TYPE_PACKAGE;
 	argv4.package.count = 2;
 	argv4.package.elements = obj_args;
 
 	obj = acpi_evaluate_dsm(ACPI_HANDLE(mac_cb->dsaf_dev->dev),
-				hns_dsaf_acpi_dsm_uuid, 0,
+				&hns_dsaf_acpi_dsm_guid, 0,
 				HNS_OP_SERDES_LP_FUNC, &argv4);
 	if (!obj) {
 		dev_warn(mac_cb->dsaf_dev->dev, "set port%d serdes lp fail!",
@@ -597,25 +666,21 @@ struct dsaf_misc_op *hns_misc_op_get(struct dsaf_device *dsaf_dev)
 		misc_op->ge_srst = hns_dsaf_ge_srst_by_port;
 		misc_op->ppe_srst = hns_ppe_srst_by_port;
 		misc_op->ppe_comm_srst = hns_ppe_com_srst;
-		misc_op->hns_dsaf_srst_chns = hns_dsaf_srst_chns;
-		misc_op->hns_dsaf_roce_srst = hns_dsaf_roce_srst;
 
 		misc_op->get_phy_if = hns_mac_get_phy_if;
 		misc_op->get_sfp_prsnt = hns_mac_get_sfp_prsnt;
 
 		misc_op->cfg_serdes_loopback = hns_mac_config_sds_loopback;
 	} else if (is_acpi_node(dsaf_dev->dev->fwnode)) {
-		misc_op->cpld_set_led = hns_cpld_set_led;
-		misc_op->cpld_reset_led = cpld_led_reset;
-		misc_op->cpld_set_led_id = cpld_set_led_id;
+		misc_op->cpld_set_led = hns_cpld_set_led_acpi;
+		misc_op->cpld_reset_led = cpld_led_reset_acpi;
+		misc_op->cpld_set_led_id = cpld_set_led_id_acpi;
 
 		misc_op->dsaf_reset = hns_dsaf_rst_acpi;
 		misc_op->xge_srst = hns_dsaf_xge_srst_by_port_acpi;
 		misc_op->ge_srst = hns_dsaf_ge_srst_by_port_acpi;
 		misc_op->ppe_srst = hns_ppe_srst_by_port_acpi;
 		misc_op->ppe_comm_srst = hns_ppe_com_srst;
-		misc_op->hns_dsaf_srst_chns = hns_dsaf_srst_chns_acpi;
-		misc_op->hns_dsaf_roce_srst = hns_dsaf_roce_srst_acpi;
 
 		misc_op->get_phy_if = hns_mac_get_phy_if_acpi;
 		misc_op->get_sfp_prsnt = hns_mac_get_sfp_prsnt_acpi;
@@ -629,17 +694,11 @@ struct dsaf_misc_op *hns_misc_op_get(struct dsaf_device *dsaf_dev)
 	return (void *)misc_op;
 }
 
-static int hns_dsaf_dev_match(struct device *dev, void *fwnode)
-{
-	return dev->fwnode == fwnode;
-}
-
 struct
 platform_device *hns_dsaf_find_platform_device(struct fwnode_handle *fwnode)
 {
 	struct device *dev;
 
-	dev = bus_find_device(&platform_bus_type, NULL,
-			      fwnode, hns_dsaf_dev_match);
+	dev = bus_find_device_by_fwnode(&platform_bus_type, fwnode);
 	return dev ? to_platform_device(dev) : NULL;
 }
diff --git a/drivers/net/ethernet/hisilicon/hns/hns_dsaf_misc.h b/drivers/net/ethernet/hisilicon/hns/hns_dsaf_misc.h
index 310e80261366..f64c6667dd05 100644
--- a/drivers/net/ethernet/hisilicon/hns/hns_dsaf_misc.h
+++ b/drivers/net/ethernet/hisilicon/hns/hns_dsaf_misc.h
@@ -1,10 +1,6 @@
+/* SPDX-License-Identifier: GPL-2.0-or-later */
 /*
  * Copyright (c) 2014-2015 Hisilicon Limited.
- *
- * This program is free software; you can redistribute it and/or modify
- * it under the terms of the GNU General Public License as published by
- * the Free Software Foundation; either version 2 of the License, or
- * (at your option) any later version.
  */
 
 #ifndef _HNS_DSAF_MISC_H
diff --git a/drivers/net/ethernet/hisilicon/hns/hns_dsaf_ppe.c b/drivers/net/ethernet/hisilicon/hns/hns_dsaf_ppe.c
index b62816c1574e..5013beb4d282 100644
--- a/drivers/net/ethernet/hisilicon/hns/hns_dsaf_ppe.c
+++ b/drivers/net/ethernet/hisilicon/hns/hns_dsaf_ppe.c
@@ -1,10 +1,6 @@
+// SPDX-License-Identifier: GPL-2.0-or-later
 /*
  * Copyright (c) 2014-2015 Hisilicon Limited.
- *
- * This program is free software; you can redistribute it and/or modify
- * it under the terms of the GNU General Public License as published by
- * the Free Software Foundation; either version 2 of the License, or
- * (at your option) any later version.
  */
 
 #include <linux/module.h>
@@ -13,9 +9,6 @@
 #include <linux/netdevice.h>
 #include <linux/etherdevice.h>
 #include <linux/platform_device.h>
-#include <linux/of.h>
-#include <linux/of_address.h>
-#include <linux/of_platform.h>
 
 #include "hns_dsaf_ppe.h"
 
@@ -61,7 +54,7 @@ void hns_ppe_set_indir_table(struct hns_ppe_cb *ppe_cb,
 	}
 }
 
-static void __iomem *
+static u8 __iomem *
 hns_ppe_common_get_ioaddr(struct ppe_common_cb *ppe_common)
 {
 	return ppe_common->dsaf_dev->ppe_base + PPE_COMMON_REG_OFFSET;
@@ -70,10 +63,10 @@ hns_ppe_common_get_ioaddr(struct ppe_common_cb *ppe_common)
 /**
  * hns_ppe_common_get_cfg - get ppe common config
  * @dsaf_dev: dasf device
- * comm_index: common index
- * retuen 0 - success , negative --fail
+ * @comm_index: common index
+ * return 0 - success , negative --fail
  */
-int hns_ppe_common_get_cfg(struct dsaf_device *dsaf_dev, int comm_index)
+static int hns_ppe_common_get_cfg(struct dsaf_device *dsaf_dev, int comm_index)
 {
 	struct ppe_common_cb *ppe_common;
 	int ppe_num;
@@ -83,8 +76,9 @@ int hns_ppe_common_get_cfg(struct dsaf_device *dsaf_dev, int comm_index)
 	else
 		ppe_num = HNS_PPE_DEBUG_NW_ENGINE_NUM;
 
-	ppe_common = devm_kzalloc(dsaf_dev->dev, sizeof(*ppe_common) +
-		ppe_num * sizeof(struct hns_ppe_cb), GFP_KERNEL);
+	ppe_common = devm_kzalloc(dsaf_dev->dev,
+				  struct_size(ppe_common, ppe_cb, ppe_num),
+				  GFP_KERNEL);
 	if (!ppe_common)
 		return -ENOMEM;
 
@@ -104,13 +98,14 @@ int hns_ppe_common_get_cfg(struct dsaf_device *dsaf_dev, int comm_index)
 	return 0;
 }
 
-void hns_ppe_common_free_cfg(struct dsaf_device *dsaf_dev, u32 comm_index)
+static void
+hns_ppe_common_free_cfg(struct dsaf_device *dsaf_dev, u32 comm_index)
 {
 	dsaf_dev->ppe_common[comm_index] = NULL;
 }
 
-static void __iomem *hns_ppe_get_iobase(struct ppe_common_cb *ppe_common,
-					int ppe_idx)
+static u8 __iomem *hns_ppe_get_iobase(struct ppe_common_cb *ppe_common,
+				      int ppe_idx)
 {
 	return ppe_common->dsaf_dev->ppe_base + ppe_idx * PPE_REG_OFFSET;
 }
@@ -145,7 +140,7 @@ static void hns_ppe_set_vlan_strip(struct hns_ppe_cb *ppe_cb, int en)
 
 /**
  * hns_ppe_checksum_hw - set ppe checksum caculate
- * @ppe_device: ppe device
+ * @ppe_cb: ppe device
  * @value: value
  */
 static void hns_ppe_checksum_hw(struct hns_ppe_cb *ppe_cb, u32 value)
@@ -181,7 +176,7 @@ static void hns_ppe_set_qid(struct ppe_common_cb *ppe_common, u32 qid)
 
 /**
  * hns_ppe_set_port_mode - set port mode
- * @ppe_device: ppe device
+ * @ppe_cb: ppe device
  * @mode: port mode
  */
 static void hns_ppe_set_port_mode(struct hns_ppe_cb *ppe_cb,
@@ -203,9 +198,9 @@ static int hns_ppe_common_init_hw(struct ppe_common_cb *ppe_common)
 	enum dsaf_mode dsaf_mode = dsaf_dev->dsaf_mode;
 
 	dsaf_dev->misc_op->ppe_comm_srst(dsaf_dev, 0);
-	mdelay(100);
+	msleep(100);
 	dsaf_dev->misc_op->ppe_comm_srst(dsaf_dev, 1);
-	mdelay(100);
+	msleep(100);
 
 	if (ppe_common->ppe_mode == PPE_COMMON_MODE_SERVICE) {
 		switch (dsaf_mode) {
@@ -274,8 +269,31 @@ static void hns_ppe_exc_irq_en(struct hns_ppe_cb *ppe_cb, int en)
 	dsaf_write_dev(ppe_cb, PPE_INTEN_REG, msk_vlue & vld_msk);
 }
 
+int hns_ppe_wait_tx_fifo_clean(struct hns_ppe_cb *ppe_cb)
+{
+	int wait_cnt;
+	u32 val;
+
+	wait_cnt = 0;
+	while (wait_cnt++ < HNS_MAX_WAIT_CNT) {
+		val = dsaf_read_dev(ppe_cb, PPE_CURR_TX_FIFO0_REG) & 0x3ffU;
+		if (!val)
+			break;
+
+		usleep_range(100, 200);
+	}
+
+	if (wait_cnt >= HNS_MAX_WAIT_CNT) {
+		dev_err(ppe_cb->dev, "hns ppe tx fifo clean wait timeout, still has %u pkt.\n",
+			val);
+		return -EBUSY;
+	}
+
+	return 0;
+}
+
 /**
- * ppe_init_hw - init ppe
+ * hns_ppe_init_hw - init ppe
  * @ppe_cb: ppe device
  */
 static void hns_ppe_init_hw(struct hns_ppe_cb *ppe_cb)
@@ -322,8 +340,8 @@ static void hns_ppe_init_hw(struct hns_ppe_cb *ppe_cb)
 }
 
 /**
- * ppe_uninit_hw - uninit ppe
- * @ppe_device: ppe device
+ * hns_ppe_uninit_hw - uninit ppe
+ * @ppe_cb: ppe device
  */
 static void hns_ppe_uninit_hw(struct hns_ppe_cb *ppe_cb)
 {
@@ -337,7 +355,7 @@ static void hns_ppe_uninit_hw(struct hns_ppe_cb *ppe_cb)
 	}
 }
 
-void hns_ppe_uninit_ex(struct ppe_common_cb *ppe_common)
+static void hns_ppe_uninit_ex(struct ppe_common_cb *ppe_common)
 {
 	u32 i;
 
@@ -361,9 +379,10 @@ void hns_ppe_uninit(struct dsaf_device *dsaf_dev)
 }
 
 /**
- * hns_ppe_reset - reinit ppe/rcb hw
+ * hns_ppe_reset_common - reinit ppe/rcb hw
  * @dsaf_dev: dasf device
- * retuen void
+ * @ppe_common_index: the index
+ * return void
  */
 void hns_ppe_reset_common(struct dsaf_device *dsaf_dev, u8 ppe_common_index)
 {
@@ -422,7 +441,7 @@ void hns_ppe_update_stats(struct hns_ppe_cb *ppe_cb)
 
 int hns_ppe_get_sset_count(int stringset)
 {
-	if (stringset == ETH_SS_STATS || stringset == ETH_SS_PRIV_FLAGS)
+	if (stringset == ETH_SS_STATS)
 		return ETH_PPE_STATIC_NUM;
 	return 0;
 }
@@ -433,40 +452,28 @@ int hns_ppe_get_regs_count(void)
 }
 
 /**
- * ppe_get_strings - get ppe srting
- * @ppe_device: ppe device
+ * hns_ppe_get_strings - get ppe srting
+ * @ppe_cb: ppe device
  * @stringset: string set type
  * @data: output string
  */
-void hns_ppe_get_strings(struct hns_ppe_cb *ppe_cb, int stringset, u8 *data)
+void hns_ppe_get_strings(struct hns_ppe_cb *ppe_cb, int stringset, u8 **data)
 {
-	char *buff = (char *)data;
 	int index = ppe_cb->index;
 
-	snprintf(buff, ETH_GSTRING_LEN, "ppe%d_rx_sw_pkt", index);
-	buff = buff + ETH_GSTRING_LEN;
-	snprintf(buff, ETH_GSTRING_LEN, "ppe%d_rx_pkt_ok", index);
-	buff = buff + ETH_GSTRING_LEN;
-	snprintf(buff, ETH_GSTRING_LEN, "ppe%d_rx_drop_pkt_no_bd", index);
-	buff = buff + ETH_GSTRING_LEN;
-	snprintf(buff, ETH_GSTRING_LEN, "ppe%d_rx_alloc_buf_fail", index);
-	buff = buff + ETH_GSTRING_LEN;
-	snprintf(buff, ETH_GSTRING_LEN, "ppe%d_rx_alloc_buf_wait", index);
-	buff = buff + ETH_GSTRING_LEN;
-	snprintf(buff, ETH_GSTRING_LEN, "ppe%d_rx_pkt_drop_no_buf", index);
-	buff = buff + ETH_GSTRING_LEN;
-	snprintf(buff, ETH_GSTRING_LEN, "ppe%d_rx_pkt_err_fifo_full", index);
-	buff = buff + ETH_GSTRING_LEN;
-
-	snprintf(buff, ETH_GSTRING_LEN, "ppe%d_tx_bd", index);
-	buff = buff + ETH_GSTRING_LEN;
-	snprintf(buff, ETH_GSTRING_LEN, "ppe%d_tx_pkt", index);
-	buff = buff + ETH_GSTRING_LEN;
-	snprintf(buff, ETH_GSTRING_LEN, "ppe%d_tx_pkt_ok", index);
-	buff = buff + ETH_GSTRING_LEN;
-	snprintf(buff, ETH_GSTRING_LEN, "ppe%d_tx_pkt_err_fifo_empty", index);
-	buff = buff + ETH_GSTRING_LEN;
-	snprintf(buff, ETH_GSTRING_LEN, "ppe%d_tx_pkt_err_csum_fail", index);
+	ethtool_sprintf(data, "ppe%d_rx_sw_pkt", index);
+	ethtool_sprintf(data, "ppe%d_rx_pkt_ok", index);
+	ethtool_sprintf(data, "ppe%d_rx_drop_pkt_no_bd", index);
+	ethtool_sprintf(data, "ppe%d_rx_alloc_buf_fail", index);
+	ethtool_sprintf(data, "ppe%d_rx_alloc_buf_wait", index);
+	ethtool_sprintf(data, "ppe%d_rx_pkt_drop_no_buf", index);
+	ethtool_sprintf(data, "ppe%d_rx_pkt_err_fifo_full", index);
+
+	ethtool_sprintf(data, "ppe%d_tx_bd", index);
+	ethtool_sprintf(data, "ppe%d_tx_pkt", index);
+	ethtool_sprintf(data, "ppe%d_tx_pkt_ok", index);
+	ethtool_sprintf(data, "ppe%d_tx_pkt_err_fifo_empty", index);
+	ethtool_sprintf(data, "ppe%d_tx_pkt_err_csum_fail", index);
 }
 
 void hns_ppe_get_stats(struct hns_ppe_cb *ppe_cb, u64 *data)
@@ -492,7 +499,7 @@ void hns_ppe_get_stats(struct hns_ppe_cb *ppe_cb, u64 *data)
 /**
  * hns_ppe_init - init ppe device
  * @dsaf_dev: dasf device
- * retuen 0 - success , negative --fail
+ * return 0 - success , negative --fail
  */
 int hns_ppe_init(struct dsaf_device *dsaf_dev)
 {
diff --git a/drivers/net/ethernet/hisilicon/hns/hns_dsaf_ppe.h b/drivers/net/ethernet/hisilicon/hns/hns_dsaf_ppe.h
index 9d8e643e8aa6..602c8e971fe4 100644
--- a/drivers/net/ethernet/hisilicon/hns/hns_dsaf_ppe.h
+++ b/drivers/net/ethernet/hisilicon/hns/hns_dsaf_ppe.h
@@ -1,10 +1,6 @@
+/* SPDX-License-Identifier: GPL-2.0-or-later */
 /*
  * Copyright (c) 2014-2015 Hisilicon Limited.
- *
- * This program is free software; you can redistribute it and/or modify
- * it under the terms of the GNU General Public License as published by
- * the Free Software Foundation; either version 2 of the License, or
- * (at your option) any later version.
  */
 
 #ifndef _HNS_DSAF_PPE_H
@@ -80,7 +76,7 @@ struct hns_ppe_cb {
 	struct hns_ppe_hw_stats hw_stats;
 
 	u8 index;	/* index in a ppe common device */
-	void __iomem *io_base;
+	u8 __iomem *io_base;
 	int virq;
 	u32 rss_indir_table[HNS_PPEV2_RSS_IND_TBL_SIZE]; /*shadow indir tab */
 	u32 rss_key[HNS_PPEV2_RSS_KEY_NUM]; /* rss hash key */
@@ -89,17 +85,18 @@ struct hns_ppe_cb {
 struct ppe_common_cb {
 	struct device *dev;
 	struct dsaf_device *dsaf_dev;
-	void __iomem *io_base;
+	u8 __iomem *io_base;
 
 	enum ppe_common_mode ppe_mode;
 
 	u8 comm_index;   /*ppe_common index*/
 
 	u32 ppe_num;
-	struct hns_ppe_cb ppe_cb[0];
+	struct hns_ppe_cb ppe_cb[] __counted_by(ppe_num);
 
 };
 
+int hns_ppe_wait_tx_fifo_clean(struct hns_ppe_cb *ppe_cb);
 int hns_ppe_init(struct dsaf_device *dsaf_dev);
 
 void hns_ppe_uninit(struct dsaf_device *dsaf_dev);
@@ -112,7 +109,7 @@ int hns_ppe_get_sset_count(int stringset);
 int hns_ppe_get_regs_count(void);
 void hns_ppe_get_regs(struct hns_ppe_cb *ppe_cb, void *data);
 
-void hns_ppe_get_strings(struct hns_ppe_cb *ppe_cb, int stringset, u8 *data);
+void hns_ppe_get_strings(struct hns_ppe_cb *ppe_cb, int stringset, u8 **data);
 void hns_ppe_get_stats(struct hns_ppe_cb *ppe_cb, u64 *data);
 void hns_ppe_set_tso_enable(struct hns_ppe_cb *ppe_cb, u32 value);
 void hns_ppe_set_rss_key(struct hns_ppe_cb *ppe_cb,
diff --git a/drivers/net/ethernet/hisilicon/hns/hns_dsaf_rcb.c b/drivers/net/ethernet/hisilicon/hns/hns_dsaf_rcb.c
index 6f3570cfb501..635b3a95dd82 100644
--- a/drivers/net/ethernet/hisilicon/hns/hns_dsaf_rcb.c
+++ b/drivers/net/ethernet/hisilicon/hns/hns_dsaf_rcb.c
@@ -1,10 +1,6 @@
+// SPDX-License-Identifier: GPL-2.0-or-later
 /*
  * Copyright (c) 2014-2015 Hisilicon Limited.
- *
- * This program is free software; you can redistribute it and/or modify
- * it under the terms of the GNU General Public License as published by
- * the Free Software Foundation; either version 2 of the License, or
- * (at your option) any later version.
  */
 
 #include <linux/cdev.h>
@@ -15,10 +11,6 @@
 #include <linux/etherdevice.h>
 #include <asm/cacheflush.h>
 #include <linux/platform_device.h>
-#include <linux/of.h>
-#include <linux/of_address.h>
-#include <linux/of_platform.h>
-#include <linux/of_irq.h>
 #include <linux/spinlock.h>
 
 #include "hns_dsaf_main.h"
@@ -38,7 +30,7 @@
 /**
  *hns_rcb_wait_fbd_clean - clean fbd
  *@qs: ring struct pointer array
- *@qnum: num of array
+ *@q_num: num of array
  *@flag: tx or rx flag
  */
 void hns_rcb_wait_fbd_clean(struct hnae_queue **qs, int q_num, u32 flag)
@@ -66,6 +58,29 @@ void hns_rcb_wait_fbd_clean(struct hnae_queue **qs, int q_num, u32 flag)
 			"queue(%d) wait fbd(%d) clean fail!!\n", i, fbd_num);
 }
 
+int hns_rcb_wait_tx_ring_clean(struct hnae_queue *qs)
+{
+	u32 head, tail;
+	int wait_cnt;
+
+	tail = dsaf_read_dev(&qs->tx_ring, RCB_REG_TAIL);
+	wait_cnt = 0;
+	while (wait_cnt++ < HNS_MAX_WAIT_CNT) {
+		head = dsaf_read_dev(&qs->tx_ring, RCB_REG_HEAD);
+		if (tail == head)
+			break;
+
+		usleep_range(100, 200);
+	}
+
+	if (wait_cnt >= HNS_MAX_WAIT_CNT) {
+		dev_err(qs->dev->dev, "rcb wait timeout, head not equal to tail.\n");
+		return -EBUSY;
+	}
+
+	return 0;
+}
+
 /**
  *hns_rcb_reset_ring_hw - ring reset
  *@q: ring struct pointer
@@ -172,18 +187,14 @@ void hns_rcbv2_int_clr_hw(struct hnae_queue *q, u32 flag)
 
 /**
  *hns_rcb_ring_enable_hw - enable ring
- *@ring: rcb ring
+ *@q: rcb ring
+ *@val: value to write
  */
 void hns_rcb_ring_enable_hw(struct hnae_queue *q, u32 val)
 {
 	dsaf_write_dev(q, RCB_RING_PREFETCH_EN_REG, !!val);
 }
 
-void hns_rcb_start(struct hnae_queue *q, u32 val)
-{
-	hns_rcb_ring_enable_hw(q, val);
-}
-
 /**
  *hns_rcb_common_init_commit_hw - make rcb common init completed
  *@rcb_common: rcb common device
@@ -435,7 +446,7 @@ static void hns_rcb_ring_get_cfg(struct hnae_queue *q, int ring_type)
 		mdnum_ppkt = HNS_RCB_RING_MAX_BD_PER_PKT;
 	} else {
 		ring = &q->tx_ring;
-		ring->io_base = (u8 __iomem *)ring_pair_cb->q.io_base +
+		ring->io_base = ring_pair_cb->q.io_base +
 			HNS_RCB_TX_REG_OFFSET;
 		irq_idx = HNS_RCB_IRQ_IDX_TX;
 		mdnum_ppkt = is_ver1 ? HNS_RCB_RING_MAX_TXBD_PER_PKT :
@@ -705,7 +716,7 @@ void hns_rcb_get_queue_mode(enum dsaf_mode dsaf_mode, u16 *max_vfn,
 	}
 }
 
-int hns_rcb_get_ring_num(struct dsaf_device *dsaf_dev)
+static int hns_rcb_get_ring_num(struct dsaf_device *dsaf_dev)
 {
 	switch (dsaf_dev->dsaf_mode) {
 	case DSAF_MODE_ENABLE_FIX:
@@ -741,7 +752,7 @@ int hns_rcb_get_ring_num(struct dsaf_device *dsaf_dev)
 	}
 }
 
-void __iomem *hns_rcb_common_get_vaddr(struct rcb_common_cb *rcb_common)
+static u8 __iomem *hns_rcb_common_get_vaddr(struct rcb_common_cb *rcb_common)
 {
 	struct dsaf_device *dsaf_dev = rcb_common->dsaf_dev;
 
@@ -765,8 +776,9 @@ int hns_rcb_common_get_cfg(struct dsaf_device *dsaf_dev,
 	int ring_num = hns_rcb_get_ring_num(dsaf_dev);
 
 	rcb_common =
-		devm_kzalloc(dsaf_dev->dev, sizeof(*rcb_common) +
-			ring_num * sizeof(struct ring_pair_cb), GFP_KERNEL);
+		devm_kzalloc(dsaf_dev->dev,
+			     struct_size(rcb_common, ring_pair_cb, ring_num),
+			     GFP_KERNEL);
 	if (!rcb_common) {
 		dev_err(dsaf_dev->dev, "rcb common devm_kzalloc fail!\n");
 		return -ENOMEM;
@@ -824,7 +836,7 @@ void hns_rcb_update_stats(struct hnae_queue *queue)
 
 /**
  *hns_rcb_get_stats - get rcb statistic
- *@ring: rcb ring
+ *@queue: rcb ring
  *@data:statistic value
  */
 void hns_rcb_get_stats(struct hnae_queue *queue, u64 *data)
@@ -876,7 +888,7 @@ void hns_rcb_get_stats(struct hnae_queue *queue, u64 *data)
  */
 int hns_rcb_get_ring_sset_count(int stringset)
 {
-	if (stringset == ETH_SS_STATS || stringset == ETH_SS_PRIV_FLAGS)
+	if (stringset == ETH_SS_STATS)
 		return HNS_RING_STATIC_REG_NUM;
 
 	return 0;
@@ -892,7 +904,7 @@ int hns_rcb_get_common_regs_count(void)
 }
 
 /**
- *rcb_get_sset_count - rcb ring regs count
+ *hns_rcb_get_ring_regs_count - rcb ring regs count
  *return regs count
  */
 int hns_rcb_get_ring_regs_count(void)
@@ -906,71 +918,42 @@ int hns_rcb_get_ring_regs_count(void)
  *@data:strings name value
  *@index:queue index
  */
-void hns_rcb_get_strings(int stringset, u8 *data, int index)
+void hns_rcb_get_strings(int stringset, u8 **data, int index)
 {
-	char *buff = (char *)data;
-
 	if (stringset != ETH_SS_STATS)
 		return;
 
-	snprintf(buff, ETH_GSTRING_LEN, "tx_ring%d_rcb_pkt_num", index);
-	buff = buff + ETH_GSTRING_LEN;
-	snprintf(buff, ETH_GSTRING_LEN, "tx_ring%d_ppe_tx_pkt_num", index);
-	buff = buff + ETH_GSTRING_LEN;
-	snprintf(buff, ETH_GSTRING_LEN, "tx_ring%d_ppe_drop_pkt_num", index);
-	buff = buff + ETH_GSTRING_LEN;
-	snprintf(buff, ETH_GSTRING_LEN, "tx_ring%d_fbd_num", index);
-	buff = buff + ETH_GSTRING_LEN;
-
-	snprintf(buff, ETH_GSTRING_LEN, "tx_ring%d_pkt_num", index);
-	buff = buff + ETH_GSTRING_LEN;
-	snprintf(buff, ETH_GSTRING_LEN, "tx_ring%d_bytes", index);
-	buff = buff + ETH_GSTRING_LEN;
-	snprintf(buff, ETH_GSTRING_LEN, "tx_ring%d_err_cnt", index);
-	buff = buff + ETH_GSTRING_LEN;
-	snprintf(buff, ETH_GSTRING_LEN, "tx_ring%d_io_err", index);
-	buff = buff + ETH_GSTRING_LEN;
-	snprintf(buff, ETH_GSTRING_LEN, "tx_ring%d_sw_err", index);
-	buff = buff + ETH_GSTRING_LEN;
-	snprintf(buff, ETH_GSTRING_LEN, "tx_ring%d_seg_pkt", index);
-	buff = buff + ETH_GSTRING_LEN;
-	snprintf(buff, ETH_GSTRING_LEN, "tx_ring%d_restart_queue", index);
-	buff = buff + ETH_GSTRING_LEN;
-	snprintf(buff, ETH_GSTRING_LEN, "tx_ring%d_tx_busy", index);
-	buff = buff + ETH_GSTRING_LEN;
-
-	snprintf(buff, ETH_GSTRING_LEN, "rx_ring%d_rcb_pkt_num", index);
-	buff = buff + ETH_GSTRING_LEN;
-	snprintf(buff, ETH_GSTRING_LEN, "rx_ring%d_ppe_pkt_num", index);
-	buff = buff + ETH_GSTRING_LEN;
-	snprintf(buff, ETH_GSTRING_LEN, "rx_ring%d_ppe_drop_pkt_num", index);
-	buff = buff + ETH_GSTRING_LEN;
-	snprintf(buff, ETH_GSTRING_LEN, "rx_ring%d_fbd_num", index);
-	buff = buff + ETH_GSTRING_LEN;
-
-	snprintf(buff, ETH_GSTRING_LEN, "rx_ring%d_pkt_num", index);
-	buff = buff + ETH_GSTRING_LEN;
-	snprintf(buff, ETH_GSTRING_LEN, "rx_ring%d_bytes", index);
-	buff = buff + ETH_GSTRING_LEN;
-	snprintf(buff, ETH_GSTRING_LEN, "rx_ring%d_err_cnt", index);
-	buff = buff + ETH_GSTRING_LEN;
-	snprintf(buff, ETH_GSTRING_LEN, "rx_ring%d_io_err", index);
-	buff = buff + ETH_GSTRING_LEN;
-	snprintf(buff, ETH_GSTRING_LEN, "rx_ring%d_sw_err", index);
-	buff = buff + ETH_GSTRING_LEN;
-	snprintf(buff, ETH_GSTRING_LEN, "rx_ring%d_seg_pkt", index);
-	buff = buff + ETH_GSTRING_LEN;
-	snprintf(buff, ETH_GSTRING_LEN, "rx_ring%d_reuse_pg", index);
-	buff = buff + ETH_GSTRING_LEN;
-	snprintf(buff, ETH_GSTRING_LEN, "rx_ring%d_len_err", index);
-	buff = buff + ETH_GSTRING_LEN;
-	snprintf(buff, ETH_GSTRING_LEN, "rx_ring%d_non_vld_desc_err", index);
-	buff = buff + ETH_GSTRING_LEN;
-	snprintf(buff, ETH_GSTRING_LEN, "rx_ring%d_bd_num_err", index);
-	buff = buff + ETH_GSTRING_LEN;
-	snprintf(buff, ETH_GSTRING_LEN, "rx_ring%d_l2_err", index);
-	buff = buff + ETH_GSTRING_LEN;
-	snprintf(buff, ETH_GSTRING_LEN, "rx_ring%d_l3l4csum_err", index);
+	ethtool_sprintf(data, "tx_ring%d_rcb_pkt_num", index);
+	ethtool_sprintf(data, "tx_ring%d_ppe_tx_pkt_num", index);
+	ethtool_sprintf(data, "tx_ring%d_ppe_drop_pkt_num", index);
+	ethtool_sprintf(data, "tx_ring%d_fbd_num", index);
+
+	ethtool_sprintf(data, "tx_ring%d_pkt_num", index);
+	ethtool_sprintf(data, "tx_ring%d_bytes", index);
+	ethtool_sprintf(data, "tx_ring%d_err_cnt", index);
+	ethtool_sprintf(data, "tx_ring%d_io_err", index);
+	ethtool_sprintf(data, "tx_ring%d_sw_err", index);
+	ethtool_sprintf(data, "tx_ring%d_seg_pkt", index);
+	ethtool_sprintf(data, "tx_ring%d_restart_queue", index);
+	ethtool_sprintf(data, "tx_ring%d_tx_busy", index);
+
+	ethtool_sprintf(data, "rx_ring%d_rcb_pkt_num", index);
+	ethtool_sprintf(data, "rx_ring%d_ppe_pkt_num", index);
+	ethtool_sprintf(data, "rx_ring%d_ppe_drop_pkt_num", index);
+	ethtool_sprintf(data, "rx_ring%d_fbd_num", index);
+
+	ethtool_sprintf(data, "rx_ring%d_pkt_num", index);
+	ethtool_sprintf(data, "rx_ring%d_bytes", index);
+	ethtool_sprintf(data, "rx_ring%d_err_cnt", index);
+	ethtool_sprintf(data, "rx_ring%d_io_err", index);
+	ethtool_sprintf(data, "rx_ring%d_sw_err", index);
+	ethtool_sprintf(data, "rx_ring%d_seg_pkt", index);
+	ethtool_sprintf(data, "rx_ring%d_reuse_pg", index);
+	ethtool_sprintf(data, "rx_ring%d_len_err", index);
+	ethtool_sprintf(data, "rx_ring%d_non_vld_desc_err", index);
+	ethtool_sprintf(data, "rx_ring%d_bd_num_err", index);
+	ethtool_sprintf(data, "rx_ring%d_l2_err", index);
+	ethtool_sprintf(data, "rx_ring%d_l3l4csum_err", index);
 }
 
 void hns_rcb_get_common_regs(struct rcb_common_cb *rcb_com, void *data)
@@ -980,7 +963,7 @@ void hns_rcb_get_common_regs(struct rcb_common_cb *rcb_com, void *data)
 	bool is_dbg = HNS_DSAF_IS_DEBUG(rcb_com->dsaf_dev);
 	u32 reg_tmp;
 	u32 reg_num_tmp;
-	u32 i = 0;
+	u32 i;
 
 	/*rcb common registers */
 	regs[0] = dsaf_read_dev(rcb_com, RCB_COM_CFG_ENDIAN_REG);
@@ -1051,7 +1034,7 @@ void hns_rcb_get_ring_regs(struct hnae_queue *queue, void *data)
 	u32 *regs = data;
 	struct ring_pair_cb *ring_pair
 		= container_of(queue, struct ring_pair_cb, q);
-	u32 i = 0;
+	u32 i;
 
 	/*rcb ring registers */
 	regs[0] = dsaf_read_dev(queue, RCB_RING_RX_RING_BASEADDR_L_REG);
diff --git a/drivers/net/ethernet/hisilicon/hns/hns_dsaf_rcb.h b/drivers/net/ethernet/hisilicon/hns/hns_dsaf_rcb.h
index 602816498c8d..68f81547dfb4 100644
--- a/drivers/net/ethernet/hisilicon/hns/hns_dsaf_rcb.h
+++ b/drivers/net/ethernet/hisilicon/hns/hns_dsaf_rcb.h
@@ -1,10 +1,6 @@
+/* SPDX-License-Identifier: GPL-2.0-or-later */
 /*
  * Copyright (c) 2014-2015 Hisilicon Limited.
- *
- * This program is free software; you can redistribute it and/or modify
- * it under the terms of the GNU General Public License as published by
- * the Free Software Foundation; either version 2 of the License, or
- * (at your option) any later version.
  */
 
 #ifndef _HNS_DSAF_RCB_H
@@ -112,7 +108,7 @@ struct rcb_common_cb {
 	u32 ring_num;
 	u32 desc_num; /*  desc num per queue*/
 
-	struct ring_pair_cb ring_pair_cb[0];
+	struct ring_pair_cb ring_pair_cb[] __counted_by(ring_num);
 };
 
 int hns_rcb_buf_size2type(u32 buf_size);
@@ -120,7 +116,6 @@ int hns_rcb_buf_size2type(u32 buf_size);
 int hns_rcb_common_get_cfg(struct dsaf_device *dsaf_dev, int comm_index);
 void hns_rcb_common_free_cfg(struct dsaf_device *dsaf_dev, u32 comm_index);
 int hns_rcb_common_init_hw(struct rcb_common_cb *rcb_common);
-void hns_rcb_start(struct hnae_queue *q, u32 val);
 int hns_rcb_get_cfg(struct rcb_common_cb *rcb_common);
 void hns_rcb_get_queue_mode(enum dsaf_mode dsaf_mode,
 			    u16 *max_vfn, u16 *max_q_per_vf);
@@ -136,6 +131,7 @@ void hns_rcbv2_int_clr_hw(struct hnae_queue *q, u32 flag);
 void hns_rcb_init_hw(struct ring_pair_cb *ring);
 void hns_rcb_reset_ring_hw(struct hnae_queue *q);
 void hns_rcb_wait_fbd_clean(struct hnae_queue **qs, int q_num, u32 flag);
+int hns_rcb_wait_tx_ring_clean(struct hnae_queue *qs);
 u32 hns_rcb_get_rx_coalesced_frames(
 	struct rcb_common_cb *rcb_common, u32 port_idx);
 u32 hns_rcb_get_tx_coalesced_frames(
@@ -160,7 +156,7 @@ int hns_rcb_get_ring_regs_count(void);
 
 void hns_rcb_get_ring_regs(struct hnae_queue *queue, void *data);
 
-void hns_rcb_get_strings(int stringset, u8 *data, int index);
+void hns_rcb_get_strings(int stringset, u8 **data, int index);
 void hns_rcb_set_rx_ring_bs(struct hnae_queue *q, u32 buf_size);
 void hns_rcb_set_tx_ring_bs(struct hnae_queue *q, u32 buf_size);
 
diff --git a/drivers/net/ethernet/hisilicon/hns/hns_dsaf_reg.h b/drivers/net/ethernet/hisilicon/hns/hns_dsaf_reg.h
index 46a52d9bb196..47ccb6e0fcaa 100644
--- a/drivers/net/ethernet/hisilicon/hns/hns_dsaf_reg.h
+++ b/drivers/net/ethernet/hisilicon/hns/hns_dsaf_reg.h
@@ -1,10 +1,6 @@
+/* SPDX-License-Identifier: GPL-2.0-or-later */
 /*
  * Copyright (c) 2014-2015 Hisilicon Limited.
- *
- * This program is free software; you can redistribute it and/or modify
- * it under the terms of the GNU General Public License as published by
- * the Free Software Foundation; either version 2 of the License, or
- * (at your option) any later version.
  */
 
 #ifndef _DSAF_REG_H_
@@ -176,7 +172,7 @@
 #define DSAF_INODE_IN_DATA_STP_DISC_0_REG	0x1A50
 #define DSAF_INODE_GE_FC_EN_0_REG		0x1B00
 #define DSAF_INODE_VC0_IN_PKT_NUM_0_REG		0x1B50
-#define DSAF_INODE_VC1_IN_PKT_NUM_0_REG		0x1C00
+#define DSAF_INODE_VC1_IN_PKT_NUM_0_REG		0x103C
 #define DSAF_INODE_IN_PRIO_PAUSE_BASE_REG	0x1C00
 #define DSAF_INODE_IN_PRIO_PAUSE_BASE_OFFSET	0x100
 #define DSAF_INODE_IN_PRIO_PAUSE_OFFSET		0x50
@@ -404,11 +400,11 @@
 #define RCB_ECC_ERR_ADDR4_REG			0x460
 #define RCB_ECC_ERR_ADDR5_REG			0x464
 
-#define RCB_COM_SF_CFG_INTMASK_RING		0x480
-#define RCB_COM_SF_CFG_RING_STS			0x484
-#define RCB_COM_SF_CFG_RING			0x488
-#define RCB_COM_SF_CFG_INTMASK_BD		0x48C
-#define RCB_COM_SF_CFG_BD_RINT_STS		0x470
+#define RCB_COM_SF_CFG_INTMASK_RING		0x470
+#define RCB_COM_SF_CFG_RING_STS			0x474
+#define RCB_COM_SF_CFG_RING			0x478
+#define RCB_COM_SF_CFG_INTMASK_BD		0x47C
+#define RCB_COM_SF_CFG_BD_RINT_STS		0x480
 #define RCB_COM_RCB_RD_BD_BUSY			0x490
 #define RCB_COM_RCB_FBD_CRT_EN			0x494
 #define RCB_COM_AXI_WR_ERR_INTMASK		0x498
@@ -464,6 +460,7 @@
 #define RCB_RING_INTMSK_TX_OVERTIME_REG		0x000C4
 #define RCB_RING_INTSTS_TX_OVERTIME_REG		0x000C8
 
+#define GMAC_FIFO_STATE_REG			0x0000UL
 #define GMAC_DUPLEX_TYPE_REG			0x0008UL
 #define GMAC_FD_FC_TYPE_REG			0x000CUL
 #define GMAC_TX_WATER_LINE_REG			0x0010UL
@@ -533,6 +530,7 @@
 #define GMAC_LD_LINK_COUNTER_REG		0x01D0UL
 #define GMAC_LOOP_REG				0x01DCUL
 #define GMAC_RECV_CONTROL_REG			0x01E0UL
+#define GMAC_PCS_RX_EN_REG			0x01E4UL
 #define GMAC_VLAN_CODE_REG			0x01E8UL
 #define GMAC_RX_OVERRUN_CNT_REG			0x01ECUL
 #define GMAC_RX_LENGTHFIELD_ERR_CNT_REG		0x01F4UL
@@ -1016,7 +1014,7 @@
 #define XGMAC_PAUSE_CTL_RSP_MODE_B	2
 #define XGMAC_PAUSE_CTL_TX_XOFF_B	3
 
-static inline void dsaf_write_reg(void __iomem *base, u32 reg, u32 value)
+static inline void dsaf_write_reg(u8 __iomem *base, u32 reg, u32 value)
 {
 	writel(value, base + reg);
 }
@@ -1034,12 +1032,9 @@ static inline void dsaf_write_syscon(struct regmap *base, u32 reg, u32 value)
 	regmap_write(base, reg, value);
 }
 
-static inline u32 dsaf_read_syscon(struct regmap *base, u32 reg)
+static inline int dsaf_read_syscon(struct regmap *base, u32 reg, u32 *val)
 {
-	unsigned int val;
-
-	regmap_read(base, reg, &val);
-	return val;
+	return regmap_read(base, reg, val);
 }
 
 #define dsaf_read_dev(a, reg) \
@@ -1054,7 +1049,7 @@ static inline u32 dsaf_read_syscon(struct regmap *base, u32 reg)
 #define dsaf_set_bit(origin, shift, val) \
 	dsaf_set_field((origin), (1ull << (shift)), (shift), (val))
 
-static inline void dsaf_set_reg_field(void __iomem *base, u32 reg, u32 mask,
+static inline void dsaf_set_reg_field(u8 __iomem *base, u32 reg, u32 mask,
 				      u32 shift, u32 val)
 {
 	u32 origin = dsaf_read_reg(base, reg);
@@ -1074,7 +1069,7 @@ static inline void dsaf_set_reg_field(void __iomem *base, u32 reg, u32 mask,
 #define dsaf_get_bit(origin, shift) \
 	dsaf_get_field((origin), (1ull << (shift)), (shift))
 
-static inline u32 dsaf_get_reg_field(void __iomem *base, u32 reg, u32 mask,
+static inline u32 dsaf_get_reg_field(u8 __iomem *base, u32 reg, u32 mask,
 				     u32 shift)
 {
 	u32 origin;
@@ -1090,11 +1085,11 @@ static inline u32 dsaf_get_reg_field(void __iomem *base, u32 reg, u32 mask,
 	dsaf_get_reg_field((dev)->io_base, (reg), (1ull << (bit)), (bit))
 
 #define dsaf_write_b(addr, data)\
-	writeb((data), (__iomem unsigned char *)(addr))
+	writeb((data), (__iomem u8 *)(addr))
 #define dsaf_read_b(addr)\
-	readb((__iomem unsigned char *)(addr))
+	readb((__iomem u8 *)(addr))
 
 #define hns_mac_reg_read64(drv, offset) \
-	readq((__iomem void *)(((u8 *)(drv)->io_base + 0xc00 + (offset))))
+	readq((__iomem void *)(((drv)->io_base + 0xc00 + (offset))))
 
 #endif	/* _DSAF_REG_H */
diff --git a/drivers/net/ethernet/hisilicon/hns/hns_dsaf_xgmac.c b/drivers/net/ethernet/hisilicon/hns/hns_dsaf_xgmac.c
index 51e7e9f5af49..dbc44c2c26c2 100644
--- a/drivers/net/ethernet/hisilicon/hns/hns_dsaf_xgmac.c
+++ b/drivers/net/ethernet/hisilicon/hns/hns_dsaf_xgmac.c
@@ -1,10 +1,6 @@
+// SPDX-License-Identifier: GPL-2.0-or-later
 /*
  * Copyright (c) 2014-2015 Hisilicon Limited.
- *
- * This program is free software; you can redistribute it and/or modify
- * it under the terms of the GNU General Public License as published by
- * the Free Software Foundation; either version 2 of the License, or
- * (at your option) any later version.
  */
 
 #include <linux/io-64-nonatomic-hi-lo.h>
@@ -108,7 +104,7 @@ static void hns_xgmac_rx_enable(struct mac_driver *drv, u32 value)
 }
 
 /**
- * hns_xgmac_tx_lf_rf_insert - insert lf rf control about xgmac
+ * hns_xgmac_lf_rf_insert - insert lf rf control about xgmac
  * @mac_drv: mac driver
  * @mode: inserf rf or lf
  */
@@ -119,7 +115,7 @@ static void hns_xgmac_lf_rf_insert(struct mac_driver *mac_drv, u32 mode)
 }
 
 /**
- * hns_xgmac__lf_rf_control_init - initial the lf rf control register
+ * hns_xgmac_lf_rf_control_init - initial the lf rf control register
  * @mac_drv: mac driver
  */
 static void hns_xgmac_lf_rf_control_init(struct mac_driver *mac_drv)
@@ -129,12 +125,12 @@ static void hns_xgmac_lf_rf_control_init(struct mac_driver *mac_drv)
 	dsaf_set_bit(val, XGMAC_UNIDIR_EN_B, 0);
 	dsaf_set_bit(val, XGMAC_RF_TX_EN_B, 1);
 	dsaf_set_field(val, XGMAC_LF_RF_INSERT_M, XGMAC_LF_RF_INSERT_S, 0);
-	dsaf_write_reg(mac_drv, XGMAC_MAC_TX_LF_RF_CONTROL_REG, val);
+	dsaf_write_dev(mac_drv, XGMAC_MAC_TX_LF_RF_CONTROL_REG, val);
 }
 
 /**
  *hns_xgmac_enable - enable xgmac port
- *@drv: mac driver
+ *@mac_drv: mac driver
  *@mode: mode of mac port
  */
 static void hns_xgmac_enable(void *mac_drv, enum mac_commom_mode mode)
@@ -215,10 +211,10 @@ static void hns_xgmac_init(void *mac_drv)
 	u32 port = drv->mac_id;
 
 	dsaf_dev->misc_op->xge_srst(dsaf_dev, port, 0);
-	mdelay(100);
+	msleep(100);
 	dsaf_dev->misc_op->xge_srst(dsaf_dev, port, 1);
 
-	mdelay(100);
+	msleep(100);
 	hns_xgmac_lf_rf_control_init(drv);
 	hns_xgmac_exc_irq_en(drv, 0);
 
@@ -246,7 +242,8 @@ static void hns_xgmac_config_pad_and_crc(void *mac_drv, u8 newval)
 /**
  *hns_xgmac_pausefrm_cfg - set pause param about xgmac
  *@mac_drv: mac driver
- *@newval:enable of pad and crc
+ *@rx_en: enable receive
+ *@tx_en: enable transmit
  */
 static void hns_xgmac_pausefrm_cfg(void *mac_drv, u32 rx_en, u32 tx_en)
 {
@@ -258,7 +255,7 @@ static void hns_xgmac_pausefrm_cfg(void *mac_drv, u32 rx_en, u32 tx_en)
 	dsaf_write_dev(drv, XGMAC_MAC_PAUSE_CTRL_REG, origin);
 }
 
-static void hns_xgmac_set_pausefrm_mac_addr(void *mac_drv, char *mac_addr)
+static void hns_xgmac_set_pausefrm_mac_addr(void *mac_drv, const char *mac_addr)
 {
 	struct mac_driver *drv = (struct mac_driver *)mac_drv;
 
@@ -270,19 +267,6 @@ static void hns_xgmac_set_pausefrm_mac_addr(void *mac_drv, char *mac_addr)
 }
 
 /**
- *hns_xgmac_set_rx_ignore_pause_frames - set rx pause param about xgmac
- *@mac_drv: mac driver
- *@enable:enable rx pause param
- */
-static void hns_xgmac_set_rx_ignore_pause_frames(void *mac_drv, u32 enable)
-{
-	struct mac_driver *drv = (struct mac_driver *)mac_drv;
-
-	dsaf_set_dev_bit(drv, XGMAC_MAC_PAUSE_CTRL_REG,
-			 XGMAC_PAUSE_CTL_RX_B, !!enable);
-}
-
-/**
  *hns_xgmac_set_tx_auto_pause_frames - set tx pause param about xgmac
  *@mac_drv: mac driver
  *@enable:enable tx pause param
@@ -311,7 +295,7 @@ static void hns_xgmac_config_max_frame_length(void *mac_drv, u16 newval)
 	dsaf_write_dev(drv, XGMAC_MAC_MAX_PKT_SIZE_REG, newval);
 }
 
-void hns_xgmac_update_stats(void *mac_drv)
+static void hns_xgmac_update_stats(void *mac_drv)
 {
 	struct mac_driver *drv = (struct mac_driver *)mac_drv;
 	struct mac_hw_stats *hw_stats = &drv->mac_cb->hw_stats;
@@ -494,12 +478,11 @@ static void hns_xgmac_get_link_status(void *mac_drv, u32 *link_stat)
 /**
  *hns_xgmac_get_regs - dump xgmac regs
  *@mac_drv: mac driver
- *@cmd:ethtool cmd
  *@data:data for value of regs
  */
 static void hns_xgmac_get_regs(void *mac_drv, void *data)
 {
-	u32 i = 0;
+	u32 i;
 	struct mac_driver *drv = (struct mac_driver *)mac_drv;
 	u32 *regs = data;
 	u64 qtmp;
@@ -760,18 +743,15 @@ static void hns_xgmac_get_stats(void *mac_drv, u64 *data)
  *@stringset: type of values in data
  *@data:data for value of string name
  */
-static void hns_xgmac_get_strings(u32 stringset, u8 *data)
+static void hns_xgmac_get_strings(u32 stringset, u8 **data)
 {
-	char *buff = (char *)data;
 	u32 i;
 
 	if (stringset != ETH_SS_STATS)
 		return;
 
-	for (i = 0; i < ARRAY_SIZE(g_xgmac_stats_string); i++) {
-		snprintf(buff, ETH_GSTRING_LEN, g_xgmac_stats_string[i].desc);
-		buff = buff + ETH_GSTRING_LEN;
-	}
+	for (i = 0; i < ARRAY_SIZE(g_xgmac_stats_string); i++)
+		ethtool_puts(data, g_xgmac_stats_string[i].desc);
 }
 
 /**
@@ -818,9 +798,6 @@ void *hns_xgmac_config(struct hns_mac_cb *mac_cb, struct mac_params *mac_param)
 	mac_drv->set_an_mode = NULL;
 	mac_drv->config_loopback = NULL;
 	mac_drv->config_pad_and_crc = hns_xgmac_config_pad_and_crc;
-	mac_drv->config_half_duplex = NULL;
-	mac_drv->set_rx_ignore_pause_frames =
-		hns_xgmac_set_rx_ignore_pause_frames;
 	mac_drv->mac_free = hns_xgmac_free;
 	mac_drv->adjust_link = NULL;
 	mac_drv->set_tx_auto_pause_frames = hns_xgmac_set_tx_auto_pause_frames;
diff --git a/drivers/net/ethernet/hisilicon/hns/hns_dsaf_xgmac.h b/drivers/net/ethernet/hisilicon/hns/hns_dsaf_xgmac.h
index da6c5343d3e1..e1b3db980712 100644
--- a/drivers/net/ethernet/hisilicon/hns/hns_dsaf_xgmac.h
+++ b/drivers/net/ethernet/hisilicon/hns/hns_dsaf_xgmac.h
@@ -1,10 +1,6 @@
+/* SPDX-License-Identifier: GPL-2.0-or-later */
 /*
  * Copyright (c) 2014-2015 Hisilicon Limited.
- *
- * This program is free software; you can redistribute it and/or modify
- * it under the terms of the GNU General Public License as published by
- * the Free Software Foundation; either version 2 of the License, or
- * (at your option) any later version.
  */
 
 #ifndef _HNS_XGMAC_H
diff --git a/drivers/net/ethernet/hisilicon/hns/hns_enet.c b/drivers/net/ethernet/hisilicon/hns/hns_enet.c
index c6700b91a2df..e905f10b894e 100644
--- a/drivers/net/ethernet/hisilicon/hns/hns_enet.c
+++ b/drivers/net/ethernet/hisilicon/hns/hns_enet.c
@@ -1,10 +1,6 @@
+// SPDX-License-Identifier: GPL-2.0-or-later
 /*
  * Copyright (c) 2014-2015 Hisilicon Limited.
- *
- * This program is free software; you can redistribute it and/or modify
- * it under the terms of the GNU General Public License as published by
- * the Free Software Foundation; either version 2 of the License, or
- * (at your option) any later version.
  */
 
 #include <linux/clk.h>
@@ -15,6 +11,7 @@
 #include <linux/io.h>
 #include <linux/ip.h>
 #include <linux/ipv6.h>
+#include <linux/irq.h>
 #include <linux/module.h>
 #include <linux/phy.h>
 #include <linux/platform_device.h>
@@ -29,20 +26,15 @@
 
 #define SERVICE_TIMER_HZ (1 * HZ)
 
-#define NIC_TX_CLEAN_MAX_NUM 256
-#define NIC_RX_CLEAN_MAX_NUM 64
-
 #define RCB_IRQ_NOT_INITED 0
 #define RCB_IRQ_INITED 1
 #define HNS_BUFFER_SIZE_2048 2048
 
 #define BD_MAX_SEND_SIZE 8191
-#define SKB_TMP_LEN(SKB) \
-	(((SKB)->transport_header - (SKB)->mac_header) + tcp_hdrlen(SKB))
 
-static void fill_v2_desc(struct hnae_ring *ring, void *priv,
-			 int size, dma_addr_t dma, int frag_end,
-			 int buf_num, enum hns_desc_type type, int mtu)
+static void fill_v2_desc_hw(struct hnae_ring *ring, void *priv, int size,
+			    int send_sz, dma_addr_t dma, int frag_end,
+			    int buf_num, enum hns_desc_type type, int mtu)
 {
 	struct hnae_desc *desc = &ring->desc[ring->next_to_use];
 	struct hnae_desc_cb *desc_cb = &ring->desc_cb[ring->next_to_use];
@@ -64,7 +56,7 @@ static void fill_v2_desc(struct hnae_ring *ring, void *priv,
 	desc_cb->type = type;
 
 	desc->addr = cpu_to_le64(dma);
-	desc->tx.send_size = cpu_to_le16((u16)size);
+	desc->tx.send_size = cpu_to_le16((u16)send_sz);
 
 	/* config bd buffer end */
 	hnae_set_bit(rrcfv, HNSV2_TXD_VLD_B, 1);
@@ -100,7 +92,7 @@ static void fill_v2_desc(struct hnae_ring *ring, void *priv,
 						     HNSV2_TXD_TSE_B, 1);
 					l4_len = tcp_hdrlen(skb);
 					mss = skb_shinfo(skb)->gso_size;
-					paylen = skb->len - SKB_TMP_LEN(skb);
+					paylen = skb->len - skb_tcp_all_headers(skb);
 				}
 			} else if (skb->protocol == htons(ETH_P_IPV6)) {
 				hnae_set_bit(tvsvsn, HNSV2_TXD_IPV6_B, 1);
@@ -114,7 +106,7 @@ static void fill_v2_desc(struct hnae_ring *ring, void *priv,
 						     HNSV2_TXD_TSE_B, 1);
 					l4_len = tcp_hdrlen(skb);
 					mss = skb_shinfo(skb)->gso_size;
-					paylen = skb->len - SKB_TMP_LEN(skb);
+					paylen = skb->len - skb_tcp_all_headers(skb);
 				}
 			}
 			desc->tx.ip_offset = ip_offset;
@@ -133,6 +125,14 @@ static void fill_v2_desc(struct hnae_ring *ring, void *priv,
 	ring_ptr_move_fw(ring, next_to_use);
 }
 
+static void fill_v2_desc(struct hnae_ring *ring, void *priv,
+			 int size, dma_addr_t dma, int frag_end,
+			 int buf_num, enum hns_desc_type type, int mtu)
+{
+	fill_v2_desc_hw(ring, priv, size, size, dma, frag_end,
+			buf_num, type, mtu);
+}
+
 static const struct acpi_device_id hns_enet_acpi_match[] = {
 	{ "HISI00C1", 0 },
 	{ "HISI00C2", 0 },
@@ -142,7 +142,8 @@ MODULE_DEVICE_TABLE(acpi, hns_enet_acpi_match);
 
 static void fill_desc(struct hnae_ring *ring, void *priv,
 		      int size, dma_addr_t dma, int frag_end,
-		      int buf_num, enum hns_desc_type type, int mtu)
+		      int buf_num, enum hns_desc_type type, int mtu,
+		      bool is_gso)
 {
 	struct hnae_desc *desc = &ring->desc[ring->next_to_use];
 	struct hnae_desc_cb *desc_cb = &ring->desc_cb[ring->next_to_use];
@@ -244,7 +245,7 @@ static int hns_nic_maybe_stop_tso(
 	int frag_num;
 	struct sk_buff *skb = *out_skb;
 	struct sk_buff *new_skb = NULL;
-	struct skb_frag_struct *frag;
+	skb_frag_t *frag;
 
 	size = skb_headlen(skb);
 	buf_num = (size + BD_MAX_SEND_SIZE - 1) / BD_MAX_SEND_SIZE;
@@ -275,6 +276,15 @@ static int hns_nic_maybe_stop_tso(
 	return 0;
 }
 
+static int hns_nic_maybe_stop_tx_v2(struct sk_buff **out_skb, int *bnum,
+				    struct hnae_ring *ring)
+{
+	if (skb_is_gso(*out_skb))
+		return hns_nic_maybe_stop_tso(out_skb, bnum, ring);
+	else
+		return hns_nic_maybe_stop_tx(out_skb, bnum, ring);
+}
+
 static void fill_tso_desc(struct hnae_ring *ring, void *priv,
 			  int size, dma_addr_t dma, int frag_end,
 			  int buf_num, enum hns_desc_type type, int mtu)
@@ -289,30 +299,44 @@ static void fill_tso_desc(struct hnae_ring *ring, void *priv,
 
 	/* when the frag size is bigger than hardware, split this frag */
 	for (k = 0; k < frag_buf_num; k++)
-		fill_v2_desc(ring, priv,
-			     (k == frag_buf_num - 1) ?
+		fill_v2_desc_hw(ring, priv, k == 0 ? size : 0,
+				(k == frag_buf_num - 1) ?
 					sizeoflast : BD_MAX_SEND_SIZE,
-			     dma + BD_MAX_SEND_SIZE * k,
-			     frag_end && (k == frag_buf_num - 1) ? 1 : 0,
-			     buf_num,
-			     (type == DESC_TYPE_SKB && !k) ?
+				dma + BD_MAX_SEND_SIZE * k,
+				frag_end && (k == frag_buf_num - 1) ? 1 : 0,
+				buf_num,
+				(type == DESC_TYPE_SKB && !k) ?
 					DESC_TYPE_SKB : DESC_TYPE_PAGE,
+				mtu);
+}
+
+static void fill_desc_v2(struct hnae_ring *ring, void *priv,
+			 int size, dma_addr_t dma, int frag_end,
+			 int buf_num, enum hns_desc_type type, int mtu,
+			 bool is_gso)
+{
+	if (is_gso)
+		fill_tso_desc(ring, priv, size, dma, frag_end, buf_num, type,
+			      mtu);
+	else
+		fill_v2_desc(ring, priv, size, dma, frag_end, buf_num, type,
 			     mtu);
 }
 
-int hns_nic_net_xmit_hw(struct net_device *ndev,
-			struct sk_buff *skb,
-			struct hns_nic_ring_data *ring_data)
+netdev_tx_t hns_nic_net_xmit_hw(struct net_device *ndev,
+				struct sk_buff *skb,
+				struct hns_nic_ring_data *ring_data)
 {
 	struct hns_nic_priv *priv = netdev_priv(ndev);
 	struct hnae_ring *ring = ring_data->ring;
 	struct device *dev = ring_to_dev(ring);
 	struct netdev_queue *dev_queue;
-	struct skb_frag_struct *frag;
+	skb_frag_t *frag;
 	int buf_num;
 	int seg_num;
 	dma_addr_t dma;
 	int size, next_to_use;
+	bool is_gso;
 	int i;
 
 	switch (priv->ops.maybe_stop_tx(&skb, &buf_num, ring)) {
@@ -339,8 +363,9 @@ int hns_nic_net_xmit_hw(struct net_device *ndev,
 		ring->stats.sw_err_cnt++;
 		goto out_err_tx_ok;
 	}
+	is_gso = skb_is_gso(skb);
 	priv->ops.fill_desc(ring, skb, size, dma, seg_num == 1 ? 1 : 0,
-			    buf_num, DESC_TYPE_SKB, ndev->mtu);
+			    buf_num, DESC_TYPE_SKB, ndev->mtu, is_gso);
 
 	/* fill the fragments */
 	for (i = 1; i < seg_num; i++) {
@@ -354,18 +379,20 @@ int hns_nic_net_xmit_hw(struct net_device *ndev,
 		}
 		priv->ops.fill_desc(ring, skb_frag_page(frag), size, dma,
 				    seg_num - 1 == i ? 1 : 0, buf_num,
-				    DESC_TYPE_PAGE, ndev->mtu);
+				    DESC_TYPE_PAGE, ndev->mtu, is_gso);
 	}
 
 	/*complete translate all packets*/
 	dev_queue = netdev_get_tx_queue(ndev, skb->queue_mapping);
 	netdev_tx_sent_queue(dev_queue, skb->len);
 
+	netif_trans_update(ndev);
+	ndev->stats.tx_bytes += skb->len;
+	ndev->stats.tx_packets++;
+
 	wmb(); /* commit all data before submit */
 	assert(skb->queue_mapping < priv->ae_handle->q_num);
 	hnae_queue_xmit(priv->ae_handle->qs[skb->queue_mapping], buf_num);
-	ring->stats.tx_pkts++;
-	ring->stats.tx_bytes += skb->len;
 
 	return NETDEV_TX_OK;
 
@@ -402,113 +429,13 @@ out_net_tx_busy:
 	return NETDEV_TX_BUSY;
 }
 
-/**
- * hns_nic_get_headlen - determine size of header for RSC/LRO/GRO/FCOE
- * @data: pointer to the start of the headers
- * @max: total length of section to find headers in
- *
- * This function is meant to determine the length of headers that will
- * be recognized by hardware for LRO, GRO, and RSC offloads.  The main
- * motivation of doing this is to only perform one pull for IPv4 TCP
- * packets so that we can do basic things like calculating the gso_size
- * based on the average data per packet.
- **/
-static unsigned int hns_nic_get_headlen(unsigned char *data, u32 flag,
-					unsigned int max_size)
-{
-	unsigned char *network;
-	u8 hlen;
-
-	/* this should never happen, but better safe than sorry */
-	if (max_size < ETH_HLEN)
-		return max_size;
-
-	/* initialize network frame pointer */
-	network = data;
-
-	/* set first protocol and move network header forward */
-	network += ETH_HLEN;
-
-	/* handle any vlan tag if present */
-	if (hnae_get_field(flag, HNS_RXD_VLAN_M, HNS_RXD_VLAN_S)
-		== HNS_RX_FLAG_VLAN_PRESENT) {
-		if ((typeof(max_size))(network - data) > (max_size - VLAN_HLEN))
-			return max_size;
-
-		network += VLAN_HLEN;
-	}
-
-	/* handle L3 protocols */
-	if (hnae_get_field(flag, HNS_RXD_L3ID_M, HNS_RXD_L3ID_S)
-		== HNS_RX_FLAG_L3ID_IPV4) {
-		if ((typeof(max_size))(network - data) >
-		    (max_size - sizeof(struct iphdr)))
-			return max_size;
-
-		/* access ihl as a u8 to avoid unaligned access on ia64 */
-		hlen = (network[0] & 0x0F) << 2;
-
-		/* verify hlen meets minimum size requirements */
-		if (hlen < sizeof(struct iphdr))
-			return network - data;
-
-		/* record next protocol if header is present */
-	} else if (hnae_get_field(flag, HNS_RXD_L3ID_M, HNS_RXD_L3ID_S)
-		== HNS_RX_FLAG_L3ID_IPV6) {
-		if ((typeof(max_size))(network - data) >
-		    (max_size - sizeof(struct ipv6hdr)))
-			return max_size;
-
-		/* record next protocol */
-		hlen = sizeof(struct ipv6hdr);
-	} else {
-		return network - data;
-	}
-
-	/* relocate pointer to start of L4 header */
-	network += hlen;
-
-	/* finally sort out TCP/UDP */
-	if (hnae_get_field(flag, HNS_RXD_L4ID_M, HNS_RXD_L4ID_S)
-		== HNS_RX_FLAG_L4ID_TCP) {
-		if ((typeof(max_size))(network - data) >
-		    (max_size - sizeof(struct tcphdr)))
-			return max_size;
-
-		/* access doff as a u8 to avoid unaligned access on ia64 */
-		hlen = (network[12] & 0xF0) >> 2;
-
-		/* verify hlen meets minimum size requirements */
-		if (hlen < sizeof(struct tcphdr))
-			return network - data;
-
-		network += hlen;
-	} else if (hnae_get_field(flag, HNS_RXD_L4ID_M, HNS_RXD_L4ID_S)
-		== HNS_RX_FLAG_L4ID_UDP) {
-		if ((typeof(max_size))(network - data) >
-		    (max_size - sizeof(struct udphdr)))
-			return max_size;
-
-		network += sizeof(struct udphdr);
-	}
-
-	/* If everything has gone correctly network should be the
-	 * data section of the packet and will be the end of the header.
-	 * If not then it probably represents the end of the last recognized
-	 * header.
-	 */
-	if ((typeof(max_size))(network - data) < max_size)
-		return network - data;
-	else
-		return max_size;
-}
-
 static void hns_nic_reuse_page(struct sk_buff *skb, int i,
 			       struct hnae_ring *ring, int pull_len,
 			       struct hnae_desc_cb *desc_cb)
 {
 	struct hnae_desc *desc;
-	int truesize, size;
+	u32 truesize;
+	int size;
 	int last_offset;
 	bool twobufs;
 
@@ -526,7 +453,7 @@ static void hns_nic_reuse_page(struct sk_buff *skb, int i,
 	}
 
 	skb_add_rx_frag(skb, i, desc_cb->priv, desc_cb->page_offset + pull_len,
-			size - pull_len, truesize - pull_len);
+			size - pull_len, truesize);
 
 	 /* avoid re-using remote pages,flag default unreuse */
 	if (unlikely(page_to_nid(desc_cb->priv) != numa_node_id()))
@@ -654,15 +581,11 @@ static int hns_nic_poll_rx_skb(struct hns_nic_ring_data *ring_data,
 	va = (unsigned char *)desc_cb->buf + desc_cb->page_offset;
 
 	/* prefetch first cache line of first page */
-	prefetch(va);
-#if L1_CACHE_BYTES < 128
-	prefetch(va + L1_CACHE_BYTES);
-#endif
+	net_prefetch(va);
 
 	skb = *out_skb = napi_alloc_skb(&ring_data->napi,
 					HNS_RX_HEAD_SIZE);
 	if (unlikely(!skb)) {
-		netdev_err(ndev, "alloc rx skb fail\n");
 		ring->stats.sw_err_cnt++;
 		return -ENOMEM;
 	}
@@ -691,7 +614,7 @@ static int hns_nic_poll_rx_skb(struct hns_nic_ring_data *ring_data,
 	} else {
 		ring->stats.seg_pkt_cnt++;
 
-		pull_len = hns_nic_get_headlen(va, bnum_flag, HNS_RX_HEAD_SIZE);
+		pull_len = eth_get_headlen(ndev, va, HNS_RX_HEAD_SIZE);
 		memcpy(__skb_put(skb, pull_len), va,
 		       ALIGN(pull_len, sizeof(long)));
 
@@ -797,7 +720,7 @@ static void hns_nic_rx_up_pro(struct hns_nic_ring_data *ring_data,
 	struct net_device *ndev = ring_data->napi.dev;
 
 	skb->protocol = eth_type_trans(skb, ndev);
-	(void)napi_gro_receive(&ring_data->napi, skb);
+	napi_gro_receive(&ring_data->napi, skb);
 }
 
 static int hns_desc_unused(struct hnae_ring *ring)
@@ -808,6 +731,115 @@ static int hns_desc_unused(struct hnae_ring *ring)
 	return ((ntc >= ntu) ? 0 : ring->desc_num) + ntc - ntu;
 }
 
+#define HNS_LOWEST_LATENCY_RATE		27	/* 27 MB/s */
+#define HNS_LOW_LATENCY_RATE			80	/* 80 MB/s */
+
+#define HNS_COAL_BDNUM			3
+
+static u32 hns_coal_rx_bdnum(struct hnae_ring *ring)
+{
+	bool coal_enable = ring->q->handle->coal_adapt_en;
+
+	if (coal_enable &&
+	    ring->coal_last_rx_bytes > HNS_LOWEST_LATENCY_RATE)
+		return HNS_COAL_BDNUM;
+	else
+		return 0;
+}
+
+static void hns_update_rx_rate(struct hnae_ring *ring)
+{
+	bool coal_enable = ring->q->handle->coal_adapt_en;
+	u32 time_passed_ms;
+	u64 total_bytes;
+
+	if (!coal_enable ||
+	    time_before(jiffies, ring->coal_last_jiffies + (HZ >> 4)))
+		return;
+
+	/* ring->stats.rx_bytes overflowed */
+	if (ring->coal_last_rx_bytes > ring->stats.rx_bytes) {
+		ring->coal_last_rx_bytes = ring->stats.rx_bytes;
+		ring->coal_last_jiffies = jiffies;
+		return;
+	}
+
+	total_bytes = ring->stats.rx_bytes - ring->coal_last_rx_bytes;
+	time_passed_ms = jiffies_to_msecs(jiffies - ring->coal_last_jiffies);
+	do_div(total_bytes, time_passed_ms);
+	ring->coal_rx_rate = total_bytes >> 10;
+
+	ring->coal_last_rx_bytes = ring->stats.rx_bytes;
+	ring->coal_last_jiffies = jiffies;
+}
+
+/**
+ * smooth_alg - smoothing algrithm for adjusting coalesce parameter
+ * @new_param: new value
+ * @old_param: old value
+ **/
+static u32 smooth_alg(u32 new_param, u32 old_param)
+{
+	u32 gap = (new_param > old_param) ? new_param - old_param
+					  : old_param - new_param;
+
+	if (gap > 8)
+		gap >>= 3;
+
+	if (new_param > old_param)
+		return old_param + gap;
+	else
+		return old_param - gap;
+}
+
+/**
+ * hns_nic_adpt_coalesce - self adapte coalesce according to rx rate
+ * @ring_data: pointer to hns_nic_ring_data
+ **/
+static void hns_nic_adpt_coalesce(struct hns_nic_ring_data *ring_data)
+{
+	struct hnae_ring *ring = ring_data->ring;
+	struct hnae_handle *handle = ring->q->handle;
+	u32 new_coal_param, old_coal_param = ring->coal_param;
+
+	if (ring->coal_rx_rate < HNS_LOWEST_LATENCY_RATE)
+		new_coal_param = HNAE_LOWEST_LATENCY_COAL_PARAM;
+	else if (ring->coal_rx_rate < HNS_LOW_LATENCY_RATE)
+		new_coal_param = HNAE_LOW_LATENCY_COAL_PARAM;
+	else
+		new_coal_param = HNAE_BULK_LATENCY_COAL_PARAM;
+
+	if (new_coal_param == old_coal_param &&
+	    new_coal_param == handle->coal_param)
+		return;
+
+	new_coal_param = smooth_alg(new_coal_param, old_coal_param);
+	ring->coal_param = new_coal_param;
+
+	/**
+	 * Because all ring in one port has one coalesce param, when one ring
+	 * calculate its own coalesce param, it cannot write to hardware at
+	 * once. There are three conditions as follows:
+	 *       1. current ring's coalesce param is larger than the hardware.
+	 *       2. or ring which adapt last time can change again.
+	 *       3. timeout.
+	 */
+	if (new_coal_param == handle->coal_param) {
+		handle->coal_last_jiffies = jiffies;
+		handle->coal_ring_idx = ring_data->queue_index;
+	} else if (new_coal_param > handle->coal_param ||
+		   handle->coal_ring_idx == ring_data->queue_index ||
+		   time_after(jiffies, handle->coal_last_jiffies + (HZ >> 4))) {
+		handle->dev->ops->set_coalesce_usecs(handle,
+					new_coal_param);
+		handle->dev->ops->set_coalesce_frames(handle,
+					1, new_coal_param);
+		handle->coal_param = new_coal_param;
+		handle->coal_ring_idx = ring_data->queue_index;
+		handle->coal_last_jiffies = jiffies;
+	}
+}
+
 static int hns_nic_rx_poll_one(struct hns_nic_ring_data *ring_data,
 			       int budget, void *v)
 {
@@ -863,21 +895,28 @@ out:
 static bool hns_nic_rx_fini_pro(struct hns_nic_ring_data *ring_data)
 {
 	struct hnae_ring *ring = ring_data->ring;
-	int num = 0;
+	int num;
+	bool rx_stopped;
 
-	ring_data->ring->q->handle->dev->ops->toggle_ring_irq(ring, 0);
+	hns_update_rx_rate(ring);
 
 	/* for hardware bug fixed */
+	ring_data->ring->q->handle->dev->ops->toggle_ring_irq(ring, 0);
 	num = readl_relaxed(ring->io_base + RCB_REG_FBDNUM);
 
-	if (num > 0) {
+	if (num <= hns_coal_rx_bdnum(ring)) {
+		if (ring->q->handle->coal_adapt_en)
+			hns_nic_adpt_coalesce(ring_data);
+
+		rx_stopped = true;
+	} else {
 		ring_data->ring->q->handle->dev->ops->toggle_ring_irq(
 			ring_data->ring, 1);
 
-		return false;
-	} else {
-		return true;
+		rx_stopped = false;
 	}
+
+	return rx_stopped;
 }
 
 static bool hns_nic_rx_fini_pro_v2(struct hns_nic_ring_data *ring_data)
@@ -885,12 +924,17 @@ static bool hns_nic_rx_fini_pro_v2(struct hns_nic_ring_data *ring_data)
 	struct hnae_ring *ring = ring_data->ring;
 	int num;
 
+	hns_update_rx_rate(ring);
 	num = readl_relaxed(ring->io_base + RCB_REG_FBDNUM);
 
-	if (!num)
+	if (num <= hns_coal_rx_bdnum(ring)) {
+		if (ring->q->handle->coal_adapt_en)
+			hns_nic_adpt_coalesce(ring_data);
+
 		return true;
-	else
-		return false;
+	}
+
+	return false;
 }
 
 static inline void hns_nic_reclaim_one_desc(struct hnae_ring *ring,
@@ -921,15 +965,6 @@ static int is_valid_clean_head(struct hnae_ring *ring, int h)
 	return u > c ? (h > c && h <= u) : (h > c || h <= u);
 }
 
-/* netif_tx_lock will turn down the performance, set only when necessary */
-#ifdef CONFIG_NET_POLL_CONTROLLER
-#define NETIF_TX_LOCK(ring) spin_lock(&(ring)->lock)
-#define NETIF_TX_UNLOCK(ring) spin_unlock(&(ring)->lock)
-#else
-#define NETIF_TX_LOCK(ring)
-#define NETIF_TX_UNLOCK(ring)
-#endif
-
 /* reclaim all desc in one budget
  * return error or number of desc left
  */
@@ -943,21 +978,16 @@ static int hns_nic_tx_poll_one(struct hns_nic_ring_data *ring_data,
 	int head;
 	int bytes, pkts;
 
-	NETIF_TX_LOCK(ring);
-
 	head = readl_relaxed(ring->io_base + RCB_REG_HEAD);
 	rmb(); /* make sure head is ready before touch any data */
 
-	if (is_ring_empty(ring) || head == ring->next_to_clean) {
-		NETIF_TX_UNLOCK(ring);
+	if (is_ring_empty(ring) || head == ring->next_to_clean)
 		return 0; /* no data to poll */
-	}
 
 	if (!is_valid_clean_head(ring, head)) {
 		netdev_err(ndev, "wrong head (%d, %d-%d)\n", head,
 			   ring->next_to_use, ring->next_to_clean);
 		ring->stats.io_err_cnt++;
-		NETIF_TX_UNLOCK(ring);
 		return -EIO;
 	}
 
@@ -968,8 +998,9 @@ static int hns_nic_tx_poll_one(struct hns_nic_ring_data *ring_data,
 		/* issue prefetch for next Tx descriptor */
 		prefetch(&ring->desc_cb[ring->next_to_clean]);
 	}
-
-	NETIF_TX_UNLOCK(ring);
+	/* update tx ring statistics. */
+	ring->stats.tx_pkts += pkts;
+	ring->stats.tx_bytes += bytes;
 
 	dev_queue = netdev_get_tx_queue(ndev, ring_data->queue_index);
 	netdev_tx_completed_queue(dev_queue, pkts, bytes);
@@ -1030,16 +1061,12 @@ static void hns_nic_tx_clr_all_bufs(struct hns_nic_ring_data *ring_data)
 	int head;
 	int bytes, pkts;
 
-	NETIF_TX_LOCK(ring);
-
 	head = ring->next_to_use; /* ntu :soft setted ring position*/
 	bytes = 0;
 	pkts = 0;
 	while (head != ring->next_to_clean)
 		hns_nic_reclaim_one_desc(ring, &bytes, &pkts);
 
-	NETIF_TX_UNLOCK(ring);
-
 	dev_queue = netdev_get_tx_queue(ndev, ring_data->queue_index);
 	netdev_tx_reset_queue(dev_queue);
 }
@@ -1051,7 +1078,6 @@ static int hns_nic_common_poll(struct napi_struct *napi, int budget)
 		container_of(napi, struct hns_nic_ring_data, napi);
 	struct hnae_ring *ring = ring_data->ring;
 
-try_again:
 	clean_complete += ring_data->poll_one(
 				ring_data, budget - clean_complete,
 				ring_data->ex_process);
@@ -1061,7 +1087,7 @@ try_again:
 			napi_complete(napi);
 			ring->q->handle->dev->ops->toggle_ring_irq(ring, 0);
 		} else {
-			goto try_again;
+			return budget;
 		}
 	}
 
@@ -1089,11 +1115,26 @@ static void hns_nic_adjust_link(struct net_device *ndev)
 	struct hnae_handle *h = priv->ae_handle;
 	int state = 1;
 
+	/* If there is no phy, do not need adjust link */
 	if (ndev->phydev) {
-		h->dev->ops->adjust_link(h, ndev->phydev->speed,
-					 ndev->phydev->duplex);
-		state = ndev->phydev->link;
+		/* When phy link down, do nothing */
+		if (ndev->phydev->link == 0)
+			return;
+
+		if (h->dev->ops->need_adjust_link(h, ndev->phydev->speed,
+						  ndev->phydev->duplex)) {
+			/* because Hi161X chip don't support to change gmac
+			 * speed and duplex with traffic. Delay 200ms to
+			 * make sure there is no more data in chip FIFO.
+			 */
+			netif_carrier_off(ndev);
+			msleep(200);
+			h->dev->ops->adjust_link(h, ndev->phydev->speed,
+						 ndev->phydev->duplex);
+			netif_carrier_on(ndev);
+		}
 	}
+
 	state = state && h->dev->ops->get_status(h);
 
 	if (state != priv->link) {
@@ -1117,12 +1158,20 @@ static void hns_nic_adjust_link(struct net_device *ndev)
  */
 int hns_nic_init_phy(struct net_device *ndev, struct hnae_handle *h)
 {
+	__ETHTOOL_DECLARE_LINK_MODE_MASK(supported) = { 0, };
 	struct phy_device *phy_dev = h->phy_dev;
 	int ret;
 
 	if (!h->phy_dev)
 		return 0;
 
+	ethtool_convert_legacy_u32_to_link_mode(supported, h->if_support);
+	linkmode_and(phy_dev->supported, phy_dev->supported, supported);
+	linkmode_copy(phy_dev->advertising, phy_dev->supported);
+
+	if (h->phy_if == PHY_INTERFACE_MODE_XGMII)
+		phy_dev->autoneg = false;
+
 	if (h->phy_if != PHY_INTERFACE_MODE_XGMII) {
 		phy_dev->dev_flags = 0;
 
@@ -1134,11 +1183,7 @@ int hns_nic_init_phy(struct net_device *ndev, struct hnae_handle *h)
 	if (unlikely(ret))
 		return -ENODEV;
 
-	phy_dev->supported &= h->if_support;
-	phy_dev->advertising = phy_dev->supported;
-
-	if (h->phy_if == PHY_INTERFACE_MODE_XGMII)
-		phy_dev->autoneg = false;
+	phy_attached_info(phy_dev);
 
 	return 0;
 }
@@ -1172,12 +1217,12 @@ static int hns_nic_net_set_mac_address(struct net_device *ndev, void *p)
 		return ret;
 	}
 
-	memcpy(ndev->dev_addr, mac_addr->sa_data, ndev->addr_len);
+	eth_hw_addr_set(ndev, mac_addr->sa_data);
 
 	return 0;
 }
 
-void hns_nic_update_stats(struct net_device *netdev)
+static void hns_nic_update_stats(struct net_device *netdev)
 {
 	struct hns_nic_priv *priv = netdev_priv(netdev);
 	struct hnae_handle *h = priv->ae_handle;
@@ -1190,7 +1235,7 @@ static void hns_init_mac_addr(struct net_device *ndev)
 {
 	struct hns_nic_priv *priv = netdev_priv(ndev);
 
-	if (!device_get_mac_address(priv->dev, ndev->dev_addr, ETH_ALEN)) {
+	if (device_get_ethdev_address(priv->dev, ndev)) {
 		eth_hw_addr_random(ndev);
 		dev_warn(priv->dev, "No valid mac, use random mac %pM",
 			 ndev->dev_addr);
@@ -1213,7 +1258,7 @@ static int hns_nic_init_affinity_mask(int q_num, int ring_idx,
 {
 	int cpu;
 
-	/* Diffrent irq banlance between 16core and 32core.
+	/* Different irq balance between 16core and 32core.
 	 * The cpu mask set by ring index according to the ring flag
 	 * which indicate the ring is tx or rx.
 	 */
@@ -1235,6 +1280,22 @@ static int hns_nic_init_affinity_mask(int q_num, int ring_idx,
 	return cpu;
 }
 
+static void hns_nic_free_irq(int q_num, struct hns_nic_priv *priv)
+{
+	int i;
+
+	for (i = 0; i < q_num * 2; i++) {
+		if (priv->ring_data[i].ring->irq_init_flag == RCB_IRQ_INITED) {
+			irq_set_affinity_hint(priv->ring_data[i].ring->irq,
+					      NULL);
+			free_irq(priv->ring_data[i].ring->irq,
+				 &priv->ring_data[i]);
+			priv->ring_data[i].ring->irq_init_flag =
+				RCB_IRQ_NOT_INITED;
+		}
+	}
+}
+
 static int hns_nic_init_irq(struct hns_nic_priv *priv)
 {
 	struct hnae_handle *h = priv->ae_handle;
@@ -1255,14 +1316,14 @@ static int hns_nic_init_irq(struct hns_nic_priv *priv)
 
 		rd->ring->ring_name[RCB_RING_NAME_LEN - 1] = '\0';
 
+		irq_set_status_flags(rd->ring->irq, IRQ_NOAUTOEN);
 		ret = request_irq(rd->ring->irq,
 				  hns_irq_handle, 0, rd->ring->ring_name, rd);
 		if (ret) {
 			netdev_err(priv->netdev, "request irq(%d) fail\n",
 				   rd->ring->irq);
-			return ret;
+			goto out_free_irq;
 		}
-		disable_irq(rd->ring->irq);
 
 		cpu = hns_nic_init_affinity_mask(h->q_num, i,
 						 rd->ring, &rd->mask);
@@ -1275,6 +1336,10 @@ static int hns_nic_init_irq(struct hns_nic_priv *priv)
 	}
 
 	return 0;
+
+out_free_irq:
+	hns_nic_free_irq(h->q_num, priv);
+	return ret;
 }
 
 static int hns_nic_net_up(struct net_device *ndev)
@@ -1284,6 +1349,9 @@ static int hns_nic_net_up(struct net_device *ndev)
 	int i, j;
 	int ret;
 
+	if (!test_bit(NIC_STATE_DOWN, &priv->state))
+		return 0;
+
 	ret = hns_nic_init_irq(priv);
 	if (ret != 0) {
 		netdev_err(ndev, "hns init irq failed! ret=%d\n", ret);
@@ -1319,6 +1387,7 @@ out_has_some_queues:
 	for (j = i - 1; j >= 0; j--)
 		hns_nic_ring_close(ndev, j);
 
+	hns_nic_free_irq(h->q_num, priv);
 	set_bit(NIC_STATE_DOWN, &priv->state);
 
 	return ret;
@@ -1333,7 +1402,7 @@ static void hns_nic_net_down(struct net_device *ndev)
 	if (test_and_set_bit(NIC_STATE_DOWN, &priv->state))
 		return;
 
-	(void)del_timer_sync(&priv->service_timer);
+	(void) timer_delete_sync(&priv->service_timer);
 	netif_tx_stop_all_queues(ndev);
 	netif_carrier_off(ndev);
 	netif_tx_disable(ndev);
@@ -1374,13 +1443,20 @@ void hns_nic_net_reset(struct net_device *ndev)
 void hns_nic_net_reinit(struct net_device *netdev)
 {
 	struct hns_nic_priv *priv = netdev_priv(netdev);
+	enum hnae_port_type type = priv->ae_handle->port_type;
 
 	netif_trans_update(priv->netdev);
 	while (test_and_set_bit(NIC_STATE_REINITING, &priv->state))
 		usleep_range(1000, 2000);
 
 	hns_nic_net_down(netdev);
-	hns_nic_net_reset(netdev);
+
+	/* Only do hns_nic_net_reset in debug mode
+	 * because of hardware limitation.
+	 */
+	if (type == HNAE_PORT_DEBUG)
+		hns_nic_net_reset(netdev);
+
 	(void)hns_nic_net_up(netdev);
 	clear_bit(NIC_STATE_REINITING, &priv->state);
 }
@@ -1429,57 +1505,30 @@ static int hns_nic_net_stop(struct net_device *ndev)
 }
 
 static void hns_tx_timeout_reset(struct hns_nic_priv *priv);
-static void hns_nic_net_timeout(struct net_device *ndev)
+#define HNS_TX_TIMEO_LIMIT (40 * HZ)
+static void hns_nic_net_timeout(struct net_device *ndev, unsigned int txqueue)
 {
 	struct hns_nic_priv *priv = netdev_priv(ndev);
 
-	hns_tx_timeout_reset(priv);
-}
-
-static int hns_nic_do_ioctl(struct net_device *netdev, struct ifreq *ifr,
-			    int cmd)
-{
-	struct phy_device *phy_dev = netdev->phydev;
-
-	if (!netif_running(netdev))
-		return -EINVAL;
-
-	if (!phy_dev)
-		return -ENOTSUPP;
-
-	return phy_mii_ioctl(phy_dev, ifr, cmd);
-}
-
-/* use only for netconsole to poll with the device without interrupt */
-#ifdef CONFIG_NET_POLL_CONTROLLER
-void hns_nic_poll_controller(struct net_device *ndev)
-{
-	struct hns_nic_priv *priv = netdev_priv(ndev);
-	unsigned long flags;
-	int i;
-
-	local_irq_save(flags);
-	for (i = 0; i < priv->ae_handle->q_num * 2; i++)
-		napi_schedule(&priv->ring_data[i].napi);
-	local_irq_restore(flags);
+	if (ndev->watchdog_timeo < HNS_TX_TIMEO_LIMIT) {
+		ndev->watchdog_timeo *= 2;
+		netdev_info(ndev, "watchdog_timo changed to %d.\n",
+			    ndev->watchdog_timeo);
+	} else {
+		ndev->watchdog_timeo = HNS_NIC_TX_TIMEOUT;
+		hns_tx_timeout_reset(priv);
+	}
 }
-#endif
 
 static netdev_tx_t hns_nic_net_xmit(struct sk_buff *skb,
 				    struct net_device *ndev)
 {
 	struct hns_nic_priv *priv = netdev_priv(ndev);
-	int ret;
 
-	assert(skb->queue_mapping < ndev->ae_handle->q_num);
-	ret = hns_nic_net_xmit_hw(ndev, skb,
-				  &tx_ring_data(priv, skb->queue_mapping));
-	if (ret == NETDEV_TX_OK) {
-		netif_trans_update(ndev);
-		ndev->stats.tx_bytes += skb->len;
-		ndev->stats.tx_packets++;
-	}
-	return (netdev_tx_t)ret;
+	assert(skb->queue_mapping < priv->ae_handle->q_num);
+
+	return hns_nic_net_xmit_hw(ndev, skb,
+				   &tx_ring_data(priv, skb->queue_mapping));
 }
 
 static void hns_nic_drop_rx_fetch(struct hns_nic_ring_data *ring_data,
@@ -1566,7 +1615,7 @@ static void hns_disable_serdes_lb(struct net_device *ndev)
  *       which buffer size is 4096.
  *    2. we set the chip serdes loopback and set rss indirection to the ring.
  *    3. construct 64-bytes ip broadcast packages, wait the associated rx ring
- *       recieving all packages and it will fetch new descriptions.
+ *       receiving all packages and it will fetch new descriptions.
  *    4. recover to the original state.
  *
  *@ndev: net device
@@ -1595,7 +1644,7 @@ static int hns_nic_clear_all_rx_fetch(struct net_device *ndev)
 	if (!org_indir)
 		return -ENOMEM;
 
-	/* store the orginal indirection */
+	/* store the original indirection */
 	ops->get_rss(h, org_indir, NULL, NULL);
 
 	cur_indir = kzalloc(indir_size, GFP_KERNEL);
@@ -1637,8 +1686,10 @@ static int hns_nic_clear_all_rx_fetch(struct net_device *ndev)
 			for (j = 0; j < fetch_num; j++) {
 				/* alloc one skb and init */
 				skb = hns_assemble_skb(ndev);
-				if (!skb)
+				if (!skb) {
+					ret = -ENOMEM;
 					goto out;
+				}
 				rd = &tx_ring_data(priv, skb->queue_mapping);
 				hns_nic_net_xmit_hw(ndev, skb, rd);
 
@@ -1726,7 +1777,7 @@ static int hns_nic_change_mtu(struct net_device *ndev, int new_mtu)
 	}
 
 	/* finally, set new mtu to netdevice */
-	ndev->mtu = new_mtu;
+	WRITE_ONCE(ndev->mtu, new_mtu);
 
 out:
 	if (if_running) {
@@ -1750,15 +1801,6 @@ static int hns_nic_set_features(struct net_device *netdev,
 			netdev_info(netdev, "enet v1 do not support tso!\n");
 		break;
 	default:
-		if (features & (NETIF_F_TSO | NETIF_F_TSO6)) {
-			priv->ops.fill_desc = fill_tso_desc;
-			priv->ops.maybe_stop_tx = hns_nic_maybe_stop_tso;
-			/* The chip only support 7*4096 */
-			netif_set_gso_max_size(netdev, 7 * 4096);
-		} else {
-			priv->ops.fill_desc = fill_v2_desc;
-			priv->ops.maybe_stop_tx = hns_nic_maybe_stop_tx;
-		}
 		break;
 	}
 	netdev->features = features;
@@ -1805,13 +1847,12 @@ static int hns_nic_uc_unsync(struct net_device *netdev,
 }
 
 /**
- * nic_set_multicast_list - set mutl mac address
- * @netdev: net device
- * @p: mac address
+ * hns_set_multicast_list - set mutl mac address
+ * @ndev: net device
  *
  * return void
  */
-void hns_set_multicast_list(struct net_device *ndev)
+static void hns_set_multicast_list(struct net_device *ndev)
 {
 	struct hns_nic_priv *priv = netdev_priv(ndev);
 	struct hnae_handle *h = priv->ae_handle;
@@ -1833,7 +1874,7 @@ void hns_set_multicast_list(struct net_device *ndev)
 	}
 }
 
-void hns_nic_set_rx_mode(struct net_device *ndev)
+static void hns_nic_set_rx_mode(struct net_device *ndev)
 {
 	struct hns_nic_priv *priv = netdev_priv(ndev);
 	struct hnae_handle *h = priv->ae_handle;
@@ -1854,7 +1895,7 @@ void hns_nic_set_rx_mode(struct net_device *ndev)
 static void hns_nic_get_stats64(struct net_device *ndev,
 				struct rtnl_link_stats64 *stats)
 {
-	int idx = 0;
+	int idx;
 	u64 tx_bytes = 0;
 	u64 rx_bytes = 0;
 	u64 tx_pkts = 0;
@@ -1898,7 +1939,7 @@ static void hns_nic_get_stats64(struct net_device *ndev,
 
 static u16
 hns_nic_select_queue(struct net_device *ndev, struct sk_buff *skb,
-		     void *accel_priv, select_queue_fallback_t fallback)
+		     struct net_device *sb_dev)
 {
 	struct ethhdr *eth_hdr = (struct ethhdr *)skb->data;
 	struct hns_nic_priv *priv = netdev_priv(ndev);
@@ -1908,7 +1949,7 @@ hns_nic_select_queue(struct net_device *ndev, struct sk_buff *skb,
 	    is_multicast_ether_addr(eth_hdr->h_dest))
 		return 0;
 	else
-		return fallback(ndev, skb);
+		return netdev_pick_tx(ndev, skb, NULL);
 }
 
 static const struct net_device_ops hns_nic_netdev_ops = {
@@ -1918,13 +1959,10 @@ static const struct net_device_ops hns_nic_netdev_ops = {
 	.ndo_tx_timeout = hns_nic_net_timeout,
 	.ndo_set_mac_address = hns_nic_net_set_mac_address,
 	.ndo_change_mtu = hns_nic_change_mtu,
-	.ndo_do_ioctl = hns_nic_do_ioctl,
+	.ndo_eth_ioctl = phy_do_ioctl_running,
 	.ndo_set_features = hns_nic_set_features,
 	.ndo_fix_features = hns_nic_fix_features,
 	.ndo_get_stats64 = hns_nic_get_stats64,
-#ifdef CONFIG_NET_POLL_CONTROLLER
-	.ndo_poll_controller = hns_nic_poll_controller,
-#endif
 	.ndo_set_rx_mode = hns_nic_set_rx_mode,
 	.ndo_select_queue = hns_nic_select_queue,
 };
@@ -1999,13 +2037,8 @@ static void hns_nic_reset_subtask(struct hns_nic_priv *priv)
 	rtnl_lock();
 	/* put off any impending NetWatchDogTimeout */
 	netif_trans_update(priv->netdev);
+	hns_nic_net_reinit(priv->netdev);
 
-	if (type == HNAE_PORT_DEBUG) {
-		hns_nic_net_reinit(priv->netdev);
-	} else {
-		netif_carrier_off(priv->netdev);
-		netif_tx_disable(priv->netdev);
-	}
 	rtnl_unlock();
 }
 
@@ -2024,11 +2057,11 @@ static void hns_nic_service_task(struct work_struct *work)
 		= container_of(work, struct hns_nic_priv, service_task);
 	struct hnae_handle *h = priv->ae_handle;
 
+	hns_nic_reset_subtask(priv);
 	hns_nic_update_link_status(priv->netdev);
 	h->dev->ops->update_led_status(h);
 	hns_nic_update_stats(priv->netdev);
 
-	hns_nic_reset_subtask(priv);
 	hns_nic_service_event_complete(priv);
 }
 
@@ -2040,9 +2073,9 @@ static void hns_nic_task_schedule(struct hns_nic_priv *priv)
 		(void)schedule_work(&priv->service_task);
 }
 
-static void hns_nic_service_timer(unsigned long data)
+static void hns_nic_service_timer(struct timer_list *t)
 {
-	struct hns_nic_priv *priv = (struct hns_nic_priv *)data;
+	struct hns_nic_priv *priv = timer_container_of(priv, t, service_timer);
 
 	(void)mod_timer(&priv->service_timer, jiffies + SERVICE_TIMER_HZ);
 
@@ -2078,7 +2111,8 @@ static int hns_nic_init_ring_data(struct hns_nic_priv *priv)
 		return -EINVAL;
 	}
 
-	priv->ring_data = kzalloc(h->q_num * sizeof(*priv->ring_data) * 2,
+	priv->ring_data = kzalloc(array3_size(h->q_num,
+					      sizeof(*priv->ring_data), 2),
 				  GFP_KERNEL);
 	if (!priv->ring_data)
 		return -ENOMEM;
@@ -2091,8 +2125,7 @@ static int hns_nic_init_ring_data(struct hns_nic_priv *priv)
 		rd->fini_process = is_ver1 ? hns_nic_tx_fini_pro :
 			hns_nic_tx_fini_pro_v2;
 
-		netif_napi_add(priv->netdev, &rd->napi,
-			       hns_nic_common_poll, NIC_TX_CLEAN_MAX_NUM);
+		netif_napi_add(priv->netdev, &rd->napi, hns_nic_common_poll);
 		rd->ring->irq_init_flag = RCB_IRQ_NOT_INITED;
 	}
 	for (i = h->q_num; i < h->q_num * 2; i++) {
@@ -2104,8 +2137,7 @@ static int hns_nic_init_ring_data(struct hns_nic_priv *priv)
 		rd->fini_process = is_ver1 ? hns_nic_rx_fini_pro :
 			hns_nic_rx_fini_pro_v2;
 
-		netif_napi_add(priv->netdev, &rd->napi,
-			       hns_nic_common_poll, NIC_RX_CLEAN_MAX_NUM);
+		netif_napi_add(priv->netdev, &rd->napi, hns_nic_common_poll);
 		rd->ring->irq_init_flag = RCB_IRQ_NOT_INITED;
 	}
 
@@ -2143,16 +2175,9 @@ static void hns_nic_set_priv_ops(struct net_device *netdev)
 		priv->ops.maybe_stop_tx = hns_nic_maybe_stop_tx;
 	} else {
 		priv->ops.get_rxd_bnum = get_v2rx_desc_bnum;
-		if ((netdev->features & NETIF_F_TSO) ||
-		    (netdev->features & NETIF_F_TSO6)) {
-			priv->ops.fill_desc = fill_tso_desc;
-			priv->ops.maybe_stop_tx = hns_nic_maybe_stop_tso;
-			/* This chip only support 7*4096 */
-			netif_set_gso_max_size(netdev, 7 * 4096);
-		} else {
-			priv->ops.fill_desc = fill_v2_desc;
-			priv->ops.maybe_stop_tx = hns_nic_maybe_stop_tx;
-		}
+		priv->ops.fill_desc = fill_desc_v2;
+		priv->ops.maybe_stop_tx = hns_nic_maybe_stop_tx_v2;
+		netif_set_tso_max_size(netdev, 7 * 4096);
 		/* enable tso when init
 		 * control tso on/off through TSE bit in bd
 		 */
@@ -2250,21 +2275,23 @@ static int hns_nic_dev_probe(struct platform_device *pdev)
 			priv->enet_ver = AE_VERSION_2;
 
 		ae_node = of_parse_phandle(dev->of_node, "ae-handle", 0);
-		if (IS_ERR_OR_NULL(ae_node)) {
-			ret = PTR_ERR(ae_node);
+		if (!ae_node) {
+			ret = -ENODEV;
 			dev_err(dev, "not find ae-handle\n");
 			goto out_read_prop_fail;
 		}
 		priv->fwnode = &ae_node->fwnode;
 	} else if (is_acpi_node(dev->fwnode)) {
-		struct acpi_reference_args args;
+		struct fwnode_reference_args args;
 
 		if (acpi_dev_found(hns_enet_acpi_match[0].id))
 			priv->enet_ver = AE_VERSION_1;
 		else if (acpi_dev_found(hns_enet_acpi_match[1].id))
 			priv->enet_ver = AE_VERSION_2;
-		else
-			return -ENXIO;
+		else {
+			ret = -ENXIO;
+			goto out_read_prop_fail;
+		}
 
 		/* try to find port-idx-in-ae first */
 		ret = acpi_node_get_property_reference(dev->fwnode,
@@ -2273,10 +2300,15 @@ static int hns_nic_dev_probe(struct platform_device *pdev)
 			dev_err(dev, "not find ae-handle\n");
 			goto out_read_prop_fail;
 		}
-		priv->fwnode = acpi_fwnode_handle(args.adev);
+		if (!is_acpi_device_node(args.fwnode)) {
+			ret = -EINVAL;
+			goto out_read_prop_fail;
+		}
+		priv->fwnode = args.fwnode;
 	} else {
 		dev_err(dev, "cannot read cfg data from OF or acpi\n");
-		return -ENXIO;
+		ret = -ENXIO;
+		goto out_read_prop_fail;
 	}
 
 	ret = device_property_read_u32(dev, "port-idx-in-ae", &port_id);
@@ -2309,10 +2341,11 @@ static int hns_nic_dev_probe(struct platform_device *pdev)
 	ndev->min_mtu = MAC_MIN_MTU;
 	switch (priv->enet_ver) {
 	case AE_VERSION_2:
-		ndev->features |= NETIF_F_TSO | NETIF_F_TSO6;
+		ndev->features |= NETIF_F_TSO | NETIF_F_TSO6 | NETIF_F_NTUPLE;
 		ndev->hw_features |= NETIF_F_IP_CSUM | NETIF_F_IPV6_CSUM |
 			NETIF_F_RXCSUM | NETIF_F_SG | NETIF_F_GSO |
 			NETIF_F_GRO | NETIF_F_TSO | NETIF_F_TSO6;
+		ndev->vlan_features |= NETIF_F_TSO | NETIF_F_TSO6;
 		ndev->max_mtu = MAC_MAX_MTU_V2 -
 				(ETH_HLEN + ETH_FCS_LEN + VLAN_HLEN);
 		break;
@@ -2332,8 +2365,7 @@ static int hns_nic_dev_probe(struct platform_device *pdev)
 	/* carrier off reporting is important to ethtool even BEFORE open */
 	netif_carrier_off(ndev);
 
-	setup_timer(&priv->service_timer, hns_nic_service_timer,
-		    (unsigned long)priv);
+	timer_setup(&priv->service_timer, hns_nic_service_timer, 0);
 	INIT_WORK(&priv->service_task, hns_nic_service_task);
 
 	set_bit(NIC_STATE_SERVICE_INITED, &priv->state);
@@ -2355,11 +2387,13 @@ static int hns_nic_dev_probe(struct platform_device *pdev)
 out_notify_fail:
 	(void)cancel_work_sync(&priv->service_task);
 out_read_prop_fail:
+	/* safe for ACPI FW */
+	of_node_put(to_of_node(priv->fwnode));
 	free_netdev(ndev);
 	return ret;
 }
 
-static int hns_nic_dev_remove(struct platform_device *pdev)
+static void hns_nic_dev_remove(struct platform_device *pdev)
 {
 	struct net_device *ndev = platform_get_drvdata(pdev);
 	struct hns_nic_priv *priv = netdev_priv(ndev);
@@ -2384,8 +2418,10 @@ static int hns_nic_dev_remove(struct platform_device *pdev)
 	set_bit(NIC_STATE_REMOVING, &priv->state);
 	(void)cancel_work_sync(&priv->service_task);
 
+	/* safe for ACPI FW */
+	of_node_put(to_of_node(priv->fwnode));
+
 	free_netdev(ndev);
-	return 0;
 }
 
 static const struct of_device_id hns_enet_of_match[] = {
diff --git a/drivers/net/ethernet/hisilicon/hns/hns_enet.h b/drivers/net/ethernet/hisilicon/hns/hns_enet.h
index 1b83232082b2..3f3ee032f631 100644
--- a/drivers/net/ethernet/hisilicon/hns/hns_enet.h
+++ b/drivers/net/ethernet/hisilicon/hns/hns_enet.h
@@ -1,10 +1,6 @@
+/* SPDX-License-Identifier: GPL-2.0-or-later */
 /*
  * Copyright (c) 2014-2015 Hisilicon Limited.
- *
- * This program is free software; you can redistribute it and/or modify
- * it under the terms of the GNU General Public License as published by
- * the Free Software Foundation; either version 2 of the License, or
- * (at your option) any later version.
  */
 
 #ifndef __HNS_ENET_H
@@ -38,7 +34,7 @@ struct hns_nic_ring_data {
 	struct hnae_ring *ring;
 	struct napi_struct napi;
 	cpumask_t mask; /* affinity mask */
-	int queue_index;
+	u32 queue_index;
 	int (*poll_one)(struct hns_nic_ring_data *, int, void *);
 	void (*ex_process)(struct hns_nic_ring_data *, struct sk_buff *);
 	bool (*fini_process)(struct hns_nic_ring_data *);
@@ -48,7 +44,8 @@ struct hns_nic_ring_data {
 struct hns_nic_ops {
 	void (*fill_desc)(struct hnae_ring *ring, void *priv,
 			  int size, dma_addr_t dma, int frag_end,
-			  int buf_num, enum hns_desc_type type, int mtu);
+			  int buf_num, enum hns_desc_type type, int mtu,
+			  bool is_gso);
 	int (*maybe_stop_tx)(struct sk_buff **out_skb,
 			     int *bnum, struct hnae_ring *ring);
 	void (*get_rxd_bnum)(u32 bnum_flag, int *out_bnum);
@@ -92,8 +89,8 @@ void hns_ethtool_set_ops(struct net_device *ndev);
 void hns_nic_net_reset(struct net_device *ndev);
 void hns_nic_net_reinit(struct net_device *netdev);
 int hns_nic_init_phy(struct net_device *ndev, struct hnae_handle *h);
-int hns_nic_net_xmit_hw(struct net_device *ndev,
-			struct sk_buff *skb,
-			struct hns_nic_ring_data *ring_data);
+netdev_tx_t hns_nic_net_xmit_hw(struct net_device *ndev,
+				struct sk_buff *skb,
+				struct hns_nic_ring_data *ring_data);
 
 #endif	/**__HNS_ENET_H */
diff --git a/drivers/net/ethernet/hisilicon/hns/hns_ethtool.c b/drivers/net/ethernet/hisilicon/hns/hns_ethtool.c
index b8fab149690f..60a586a951a0 100644
--- a/drivers/net/ethernet/hisilicon/hns/hns_ethtool.c
+++ b/drivers/net/ethernet/hisilicon/hns/hns_ethtool.c
@@ -1,10 +1,6 @@
+// SPDX-License-Identifier: GPL-2.0-or-later
 /*
  * Copyright (c) 2014-2015 Hisilicon Limited.
- *
- * This program is free software; you can redistribute it and/or modify
- * it under the terms of the GNU General Public License as published by
- * the Free Software Foundation; either version 2 of the License, or
- * (at your option) any later version.
  */
 
 #include <linux/etherdevice.h>
@@ -21,7 +17,6 @@
 #define HNS_PHY_CSC_REG		16	/* Copper Specific Control Register */
 #define HNS_PHY_CSS_REG		17	/* Copper Specific Status Register */
 #define HNS_LED_FC_REG		16	/* LED Function Control Reg. */
-#define HNS_LED_PC_REG		17	/* LED Polarity Control Reg. */
 
 #define HNS_LED_FORCE_ON	9
 #define HNS_LED_FORCE_OFF	8
@@ -150,7 +145,7 @@ static int hns_nic_get_link_ksettings(struct net_device *net_dev,
 	cmd->base.duplex = duplex;
 
 	if (net_dev->phydev)
-		(void)phy_ethtool_ksettings_get(net_dev->phydev, cmd);
+		phy_ethtool_ksettings_get(net_dev->phydev, cmd);
 
 	link_stat = hns_nic_get_link(net_dev);
 	if (!link_stat) {
@@ -197,7 +192,7 @@ static int hns_nic_get_link_ksettings(struct net_device *net_dev,
 }
 
 /**
- *hns_nic_set_link_settings - implement ethtool set link ksettings
+ *hns_nic_set_link_ksettings - implement ethtool set link ksettings
  *@net_dev: net_device
  *@cmd: ethtool_link_ksettings
  *retuen 0 - success , negative --fail
@@ -243,7 +238,9 @@ static int hns_nic_set_link_ksettings(struct net_device *net_dev,
 	}
 
 	if (h->dev->ops->adjust_link) {
+		netif_carrier_off(net_dev);
 		h->dev->ops->adjust_link(h, (int)speed, cmd->base.duplex);
+		netif_carrier_on(net_dev);
 		return 0;
 	}
 
@@ -259,55 +256,27 @@ static const char hns_nic_test_strs[][ETH_GSTRING_LEN] = {
 
 static int hns_nic_config_phy_loopback(struct phy_device *phy_dev, u8 en)
 {
-#define COPPER_CONTROL_REG 0
-#define PHY_POWER_DOWN BIT(11)
-#define PHY_LOOP_BACK BIT(14)
-	u16 val = 0;
-
-	if (phy_dev->is_c45) /* c45 branch adding for XGE PHY */
-		return -ENOTSUPP;
+	int err;
 
 	if (en) {
-		/* speed : 1000M */
-		phy_write(phy_dev, HNS_PHY_PAGE_REG, 2);
-		phy_write(phy_dev, 21, 0x1046);
-
-		phy_write(phy_dev, HNS_PHY_PAGE_REG, 0);
-		/* Force Master */
-		phy_write(phy_dev, 9, 0x1F00);
-
-		/* Soft-reset */
-		phy_write(phy_dev, 0, 0x9140);
-		/* If autoneg disabled,two soft-reset operations */
-		phy_write(phy_dev, 0, 0x9140);
-
-		phy_write(phy_dev, HNS_PHY_PAGE_REG, 0xFA);
-
-		/* Default is 0x0400 */
-		phy_write(phy_dev, 1, 0x418);
-
-		/* Force 1000M Link, Default is 0x0200 */
-		phy_write(phy_dev, 7, 0x20C);
-		phy_write(phy_dev, HNS_PHY_PAGE_REG, 0);
-
-		/* Enable PHY loop-back */
-		val = phy_read(phy_dev, COPPER_CONTROL_REG);
-		val |= PHY_LOOP_BACK;
-		val &= ~PHY_POWER_DOWN;
-		phy_write(phy_dev, COPPER_CONTROL_REG, val);
+		/* Doing phy loopback in offline state, phy resuming is
+		 * needed to power up the device.
+		 */
+		err = phy_resume(phy_dev);
+		if (err)
+			goto out;
+
+		err = phy_loopback(phy_dev, true, 0);
 	} else {
-		phy_write(phy_dev, HNS_PHY_PAGE_REG, 0xFA);
-		phy_write(phy_dev, 1, 0x400);
-		phy_write(phy_dev, 7, 0x200);
-		phy_write(phy_dev, HNS_PHY_PAGE_REG, 0);
-		phy_write(phy_dev, 9, 0xF00);
-
-		val = phy_read(phy_dev, COPPER_CONTROL_REG);
-		val &= ~PHY_LOOP_BACK;
-		val |= PHY_POWER_DOWN;
-		phy_write(phy_dev, COPPER_CONTROL_REG, val);
+		err = phy_loopback(phy_dev, false, 0);
+		if (err)
+			goto out;
+
+		err = phy_suspend(phy_dev);
 	}
-	return 0;
+
+out:
+	return err;
 }
 
 static int __lb_setup(struct net_device *ndev,
@@ -320,10 +289,9 @@ static int __lb_setup(struct net_device *ndev,
 
 	switch (loop) {
 	case MAC_INTERNALLOOP_PHY:
-		if ((phy_dev) && (!phy_dev->is_c45)) {
-			ret = hns_nic_config_phy_loopback(phy_dev, 0x1);
-			ret |= h->dev->ops->set_loopback(h, loop, 0x1);
-		}
+		ret = hns_nic_config_phy_loopback(phy_dev, 0x1);
+		if (!ret)
+			ret = h->dev->ops->set_loopback(h, loop, 0x1);
 		break;
 	case MAC_INTERNALLOOP_MAC:
 		if ((h->dev->ops->set_loopback) &&
@@ -334,17 +302,18 @@ static int __lb_setup(struct net_device *ndev,
 		if (h->dev->ops->set_loopback)
 			ret = h->dev->ops->set_loopback(h, loop, 0x1);
 		break;
+	case MAC_LOOP_PHY_NONE:
+		ret = hns_nic_config_phy_loopback(phy_dev, 0x0);
+		fallthrough;
 	case MAC_LOOP_NONE:
-		if ((phy_dev) && (!phy_dev->is_c45))
-			ret |= hns_nic_config_phy_loopback(phy_dev, 0x0);
-
-		if (h->dev->ops->set_loopback) {
+		if (!ret && h->dev->ops->set_loopback) {
 			if (priv->ae_handle->phy_if != PHY_INTERFACE_MODE_XGMII)
-				ret |= h->dev->ops->set_loopback(h,
+				ret = h->dev->ops->set_loopback(h,
 					MAC_INTERNALLOOP_MAC, 0x0);
 
-			ret |= h->dev->ops->set_loopback(h,
-				MAC_INTERNALLOOP_SERDES, 0x0);
+			if (!ret)
+				ret = h->dev->ops->set_loopback(h,
+					MAC_INTERNALLOOP_SERDES, 0x0);
 		}
 		break;
 	default:
@@ -365,6 +334,7 @@ static int __lb_setup(struct net_device *ndev,
 static int __lb_up(struct net_device *ndev,
 		   enum hnae_loop loop_mode)
 {
+#define NIC_LB_TEST_WAIT_PHY_LINK_TIME 300
 	struct hns_nic_priv *priv = netdev_priv(ndev);
 	struct hnae_handle *h = priv->ae_handle;
 	int speed, duplex;
@@ -391,6 +361,9 @@ static int __lb_up(struct net_device *ndev,
 
 	h->dev->ops->adjust_link(h, speed, duplex);
 
+	/* wait adjust link done and phy ready */
+	msleep(NIC_LB_TEST_WAIT_PHY_LINK_TIME);
+
 	return 0;
 }
 
@@ -441,6 +414,10 @@ static void __lb_other_process(struct hns_nic_ring_data *ring_data,
 	/* for mutl buffer*/
 	new_skb = skb_copy(skb, GFP_ATOMIC);
 	dev_kfree_skb_any(skb);
+	if (!new_skb) {
+		netdev_err(ndev, "skb alloc failed\n");
+		return;
+	}
 	skb = new_skb;
 
 	check_ok = 0;
@@ -488,9 +465,9 @@ static int __lb_clean_rings(struct hns_nic_priv *priv,
 }
 
 /**
- * nic_run_loopback_test -  run loopback test
- * @nic_dev: net device
- * @loopback_type: loopback type
+ * __lb_run_test -  run loopback test
+ * @ndev: net device
+ * @loop_mode: loopback mode
  */
 static int __lb_run_test(struct net_device *ndev,
 			 enum hnae_loop loop_mode)
@@ -502,7 +479,7 @@ static int __lb_run_test(struct net_device *ndev,
 #define NIC_LB_TEST_NO_MEM_ERR 1
 #define NIC_LB_TEST_TX_CNT_ERR 2
 #define NIC_LB_TEST_RX_CNT_ERR 3
-#define NIC_LB_TEST_RX_PKG_ERR 4
+
 	struct hns_nic_priv *priv = netdev_priv(ndev);
 	struct hnae_handle *h = priv->ae_handle;
 	int i, j, lc, good_cnt, ret_val = 0;
@@ -570,13 +547,16 @@ static int __lb_run_test(struct net_device *ndev,
 	return ret_val;
 }
 
-static int __lb_down(struct net_device *ndev)
+static int __lb_down(struct net_device *ndev, enum hnae_loop loop)
 {
 	struct hns_nic_priv *priv = netdev_priv(ndev);
 	struct hnae_handle *h = priv->ae_handle;
 	int ret;
 
-	ret = __lb_setup(ndev, MAC_LOOP_NONE);
+	if (loop == MAC_INTERNALLOOP_PHY)
+		ret = __lb_setup(ndev, MAC_LOOP_PHY_NONE);
+	else
+		ret = __lb_setup(ndev, MAC_LOOP_NONE);
 	if (ret)
 		netdev_err(ndev, "%s: __lb_setup return error(%d)!\n",
 			   __func__,
@@ -595,7 +575,7 @@ static int __lb_down(struct net_device *ndev)
 
 /**
  * hns_nic_self_test - self test
- * @dev: net device
+ * @ndev: net device
  * @eth_test: test cmd
  * @data: test result
  */
@@ -621,7 +601,7 @@ static void hns_nic_self_test(struct net_device *ndev,
 		set_bit(NIC_STATE_TESTING, &priv->state);
 
 		if (if_running)
-			(void)dev_close(ndev);
+			dev_close(ndev);
 
 		for (i = 0; i < SELF_TEST_TPYE_NUM; i++) {
 			if (!st_param[i][1])
@@ -632,7 +612,8 @@ static void hns_nic_self_test(struct net_device *ndev,
 			if (!data[test_index]) {
 				data[test_index] = __lb_run_test(
 					ndev, (enum hnae_loop)st_param[i][0]);
-				(void)__lb_down(ndev);
+				(void)__lb_down(ndev,
+						(enum hnae_loop)st_param[i][0]);
 			}
 
 			if (data[test_index])
@@ -646,7 +627,7 @@ static void hns_nic_self_test(struct net_device *ndev,
 		clear_bit(NIC_STATE_TESTING, &priv->state);
 
 		if (if_running)
-			(void)dev_open(ndev);
+			(void)dev_open(ndev, NULL);
 	}
 	/* Online tests aren't run; pass by default */
 
@@ -655,7 +636,7 @@ static void hns_nic_self_test(struct net_device *ndev,
 
 /**
  * hns_nic_get_drvinfo - get net driver info
- * @dev: net device
+ * @net_dev: net device
  * @drvinfo: driver info
  */
 static void hns_nic_get_drvinfo(struct net_device *net_dev,
@@ -663,28 +644,29 @@ static void hns_nic_get_drvinfo(struct net_device *net_dev,
 {
 	struct hns_nic_priv *priv = netdev_priv(net_dev);
 
-	strncpy(drvinfo->version, HNAE_DRIVER_VERSION,
+	strscpy(drvinfo->version, HNAE_DRIVER_VERSION,
 		sizeof(drvinfo->version));
-	drvinfo->version[sizeof(drvinfo->version) - 1] = '\0';
 
-	strncpy(drvinfo->driver, HNAE_DRIVER_NAME, sizeof(drvinfo->driver));
-	drvinfo->driver[sizeof(drvinfo->driver) - 1] = '\0';
+	strscpy(drvinfo->driver, HNAE_DRIVER_NAME, sizeof(drvinfo->driver));
 
-	strncpy(drvinfo->bus_info, priv->dev->bus->name,
+	strscpy(drvinfo->bus_info, priv->dev->bus->name,
 		sizeof(drvinfo->bus_info));
-	drvinfo->bus_info[ETHTOOL_BUSINFO_LEN - 1] = '\0';
 
-	strncpy(drvinfo->fw_version, "N/A", ETHTOOL_FWVERS_LEN);
+	strscpy(drvinfo->fw_version, "N/A", ETHTOOL_FWVERS_LEN);
 	drvinfo->eedump_len = 0;
 }
 
 /**
  * hns_get_ringparam - get ring parameter
- * @dev: net device
+ * @net_dev: net device
  * @param: ethtool parameter
+ * @kernel_param: ethtool external parameter
+ * @extack: netlink extended ACK report struct
  */
-void hns_get_ringparam(struct net_device *net_dev,
-		       struct ethtool_ringparam *param)
+static void hns_get_ringparam(struct net_device *net_dev,
+			      struct ethtool_ringparam *param,
+			      struct kernel_ethtool_ringparam *kernel_param,
+			      struct netlink_ext_ack *extack)
 {
 	struct hns_nic_priv *priv = netdev_priv(net_dev);
 	struct hnae_ae_ops *ops;
@@ -705,7 +687,7 @@ void hns_get_ringparam(struct net_device *net_dev,
 
 /**
  * hns_get_pauseparam - get pause parameter
- * @dev: net device
+ * @net_dev: net device
  * @param: pause parameter
  */
 static void hns_get_pauseparam(struct net_device *net_dev,
@@ -723,7 +705,7 @@ static void hns_get_pauseparam(struct net_device *net_dev,
 
 /**
  * hns_set_pauseparam - set pause parameter
- * @dev: net device
+ * @net_dev: net device
  * @param: pause parameter
  *
  * Return 0 on success, negative on failure
@@ -747,21 +729,25 @@ static int hns_set_pauseparam(struct net_device *net_dev,
 
 /**
  * hns_get_coalesce - get coalesce info.
- * @dev: net device
+ * @net_dev: net device
  * @ec: coalesce info.
+ * @kernel_coal: ethtool CQE mode setting structure
+ * @extack: extack for reporting error messages
  *
  * Return 0 on success, negative on failure.
  */
 static int hns_get_coalesce(struct net_device *net_dev,
-			    struct ethtool_coalesce *ec)
+			    struct ethtool_coalesce *ec,
+			    struct kernel_ethtool_coalesce *kernel_coal,
+			    struct netlink_ext_ack *extack)
 {
 	struct hns_nic_priv *priv = netdev_priv(net_dev);
 	struct hnae_ae_ops *ops;
 
 	ops = priv->ae_handle->dev->ops;
 
-	ec->use_adaptive_rx_coalesce = 1;
-	ec->use_adaptive_tx_coalesce = 1;
+	ec->use_adaptive_rx_coalesce = priv->ae_handle->coal_adapt_en;
+	ec->use_adaptive_tx_coalesce = priv->ae_handle->coal_adapt_en;
 
 	if ((!ops->get_coalesce_usecs) ||
 	    (!ops->get_max_coalesced_frames))
@@ -791,13 +777,17 @@ static int hns_get_coalesce(struct net_device *net_dev,
 
 /**
  * hns_set_coalesce - set coalesce info.
- * @dev: net device
+ * @net_dev: net device
  * @ec: coalesce info.
+ * @kernel_coal: ethtool CQE mode setting structure
+ * @extack: extack for reporting error messages
  *
  * Return 0 on success, negative on failure.
  */
 static int hns_set_coalesce(struct net_device *net_dev,
-			    struct ethtool_coalesce *ec)
+			    struct ethtool_coalesce *ec,
+			    struct kernel_ethtool_coalesce *kernel_coal,
+			    struct netlink_ext_ack *extack)
 {
 	struct hns_nic_priv *priv = netdev_priv(net_dev);
 	struct hnae_ae_ops *ops;
@@ -812,6 +802,9 @@ static int hns_set_coalesce(struct net_device *net_dev,
 	    (!ops->set_coalesce_frames))
 		return -ESRCH;
 
+	if (ec->use_adaptive_rx_coalesce != priv->ae_handle->coal_adapt_en)
+		priv->ae_handle->coal_adapt_en = ec->use_adaptive_rx_coalesce;
+
 	rc1 = ops->set_coalesce_usecs(priv->ae_handle,
 				      ec->rx_coalesce_usecs);
 
@@ -827,10 +820,11 @@ static int hns_set_coalesce(struct net_device *net_dev,
 
 /**
  * hns_get_channels - get channel info.
- * @dev: net device
+ * @net_dev: net device
  * @ch: channel info.
  */
-void hns_get_channels(struct net_device *net_dev, struct ethtool_channels *ch)
+static void
+hns_get_channels(struct net_device *net_dev, struct ethtool_channels *ch)
 {
 	struct hns_nic_priv *priv = netdev_priv(net_dev);
 
@@ -842,13 +836,13 @@ void hns_get_channels(struct net_device *net_dev, struct ethtool_channels *ch)
 }
 
 /**
- * get_ethtool_stats - get detail statistics.
- * @dev: net device
+ * hns_get_ethtool_stats - get detail statistics.
+ * @netdev: net device
  * @stats: statistics info.
  * @data: statistics data.
  */
-void hns_get_ethtool_stats(struct net_device *netdev,
-			   struct ethtool_stats *stats, u64 *data)
+static void hns_get_ethtool_stats(struct net_device *netdev,
+				  struct ethtool_stats *stats, u64 *data)
 {
 	u64 *p = data;
 	struct hns_nic_priv *priv = netdev_priv(netdev);
@@ -890,8 +884,8 @@ void hns_get_ethtool_stats(struct net_device *netdev,
 	p[21] = net_stats->rx_compressed;
 	p[22] = net_stats->tx_compressed;
 
-	p[23] = netdev->rx_dropped.counter;
-	p[24] = netdev->tx_dropped.counter;
+	p[23] = 0; /* was netdev->rx_dropped.counter */
+	p[24] = 0; /* was netdev->tx_dropped.counter */
 
 	p[25] = priv->tx_timeout_count;
 
@@ -900,16 +894,15 @@ void hns_get_ethtool_stats(struct net_device *netdev,
 }
 
 /**
- * get_strings: Return a set of strings that describe the requested objects
- * @dev: net device
- * @stats: string set ID.
+ * hns_get_strings: Return a set of strings that describe the requested objects
+ * @netdev: net device
+ * @stringset: string set ID.
  * @data: objects data.
  */
-void hns_get_strings(struct net_device *netdev, u32 stringset, u8 *data)
+static void hns_get_strings(struct net_device *netdev, u32 stringset, u8 *data)
 {
 	struct hns_nic_priv *priv = netdev_priv(netdev);
 	struct hnae_handle *h = priv->ae_handle;
-	char *buff = (char *)data;
 
 	if (!h->dev->ops->get_strings) {
 		netdev_err(netdev, "h->dev->ops->get_strings is null!\n");
@@ -917,85 +910,55 @@ void hns_get_strings(struct net_device *netdev, u32 stringset, u8 *data)
 	}
 
 	if (stringset == ETH_SS_TEST) {
-		if (priv->ae_handle->phy_if != PHY_INTERFACE_MODE_XGMII) {
-			memcpy(buff, hns_nic_test_strs[MAC_INTERNALLOOP_MAC],
-			       ETH_GSTRING_LEN);
-			buff += ETH_GSTRING_LEN;
-		}
-		memcpy(buff, hns_nic_test_strs[MAC_INTERNALLOOP_SERDES],
-		       ETH_GSTRING_LEN);
-		buff += ETH_GSTRING_LEN;
+		if (priv->ae_handle->phy_if != PHY_INTERFACE_MODE_XGMII)
+			ethtool_puts(&data,
+				     hns_nic_test_strs[MAC_INTERNALLOOP_MAC]);
+		ethtool_puts(&data, hns_nic_test_strs[MAC_INTERNALLOOP_SERDES]);
 		if ((netdev->phydev) && (!netdev->phydev->is_c45))
-			memcpy(buff, hns_nic_test_strs[MAC_INTERNALLOOP_PHY],
-			       ETH_GSTRING_LEN);
+			ethtool_puts(&data,
+				     hns_nic_test_strs[MAC_INTERNALLOOP_PHY]);
 
 	} else {
-		snprintf(buff, ETH_GSTRING_LEN, "rx_packets");
-		buff = buff + ETH_GSTRING_LEN;
-		snprintf(buff, ETH_GSTRING_LEN, "tx_packets");
-		buff = buff + ETH_GSTRING_LEN;
-		snprintf(buff, ETH_GSTRING_LEN, "rx_bytes");
-		buff = buff + ETH_GSTRING_LEN;
-		snprintf(buff, ETH_GSTRING_LEN, "tx_bytes");
-		buff = buff + ETH_GSTRING_LEN;
-		snprintf(buff, ETH_GSTRING_LEN, "rx_errors");
-		buff = buff + ETH_GSTRING_LEN;
-		snprintf(buff, ETH_GSTRING_LEN, "tx_errors");
-		buff = buff + ETH_GSTRING_LEN;
-		snprintf(buff, ETH_GSTRING_LEN, "rx_dropped");
-		buff = buff + ETH_GSTRING_LEN;
-		snprintf(buff, ETH_GSTRING_LEN, "tx_dropped");
-		buff = buff + ETH_GSTRING_LEN;
-		snprintf(buff, ETH_GSTRING_LEN, "multicast");
-		buff = buff + ETH_GSTRING_LEN;
-		snprintf(buff, ETH_GSTRING_LEN, "collisions");
-		buff = buff + ETH_GSTRING_LEN;
-		snprintf(buff, ETH_GSTRING_LEN, "rx_over_errors");
-		buff = buff + ETH_GSTRING_LEN;
-		snprintf(buff, ETH_GSTRING_LEN, "rx_crc_errors");
-		buff = buff + ETH_GSTRING_LEN;
-		snprintf(buff, ETH_GSTRING_LEN, "rx_frame_errors");
-		buff = buff + ETH_GSTRING_LEN;
-		snprintf(buff, ETH_GSTRING_LEN, "rx_fifo_errors");
-		buff = buff + ETH_GSTRING_LEN;
-		snprintf(buff, ETH_GSTRING_LEN, "rx_missed_errors");
-		buff = buff + ETH_GSTRING_LEN;
-		snprintf(buff, ETH_GSTRING_LEN, "tx_aborted_errors");
-		buff = buff + ETH_GSTRING_LEN;
-		snprintf(buff, ETH_GSTRING_LEN, "tx_carrier_errors");
-		buff = buff + ETH_GSTRING_LEN;
-		snprintf(buff, ETH_GSTRING_LEN, "tx_fifo_errors");
-		buff = buff + ETH_GSTRING_LEN;
-		snprintf(buff, ETH_GSTRING_LEN, "tx_heartbeat_errors");
-		buff = buff + ETH_GSTRING_LEN;
-		snprintf(buff, ETH_GSTRING_LEN, "rx_length_errors");
-		buff = buff + ETH_GSTRING_LEN;
-		snprintf(buff, ETH_GSTRING_LEN, "tx_window_errors");
-		buff = buff + ETH_GSTRING_LEN;
-		snprintf(buff, ETH_GSTRING_LEN, "rx_compressed");
-		buff = buff + ETH_GSTRING_LEN;
-		snprintf(buff, ETH_GSTRING_LEN, "tx_compressed");
-		buff = buff + ETH_GSTRING_LEN;
-		snprintf(buff, ETH_GSTRING_LEN, "netdev_rx_dropped");
-		buff = buff + ETH_GSTRING_LEN;
-		snprintf(buff, ETH_GSTRING_LEN, "netdev_tx_dropped");
-		buff = buff + ETH_GSTRING_LEN;
-
-		snprintf(buff, ETH_GSTRING_LEN, "netdev_tx_timeout");
-		buff = buff + ETH_GSTRING_LEN;
-
-		h->dev->ops->get_strings(h, stringset, (u8 *)buff);
+		ethtool_puts(&data, "rx_packets");
+		ethtool_puts(&data, "tx_packets");
+		ethtool_puts(&data, "rx_bytes");
+		ethtool_puts(&data, "tx_bytes");
+		ethtool_puts(&data, "rx_errors");
+		ethtool_puts(&data, "tx_errors");
+		ethtool_puts(&data, "rx_dropped");
+		ethtool_puts(&data, "tx_dropped");
+		ethtool_puts(&data, "multicast");
+		ethtool_puts(&data, "collisions");
+		ethtool_puts(&data, "rx_over_errors");
+		ethtool_puts(&data, "rx_crc_errors");
+		ethtool_puts(&data, "rx_frame_errors");
+		ethtool_puts(&data, "rx_fifo_errors");
+		ethtool_puts(&data, "rx_missed_errors");
+		ethtool_puts(&data, "tx_aborted_errors");
+		ethtool_puts(&data, "tx_carrier_errors");
+		ethtool_puts(&data, "tx_fifo_errors");
+		ethtool_puts(&data, "tx_heartbeat_errors");
+		ethtool_puts(&data, "rx_length_errors");
+		ethtool_puts(&data, "tx_window_errors");
+		ethtool_puts(&data, "rx_compressed");
+		ethtool_puts(&data, "tx_compressed");
+		ethtool_puts(&data, "netdev_rx_dropped");
+		ethtool_puts(&data, "netdev_tx_dropped");
+
+		ethtool_puts(&data, "netdev_tx_timeout");
+
+		h->dev->ops->get_strings(h, stringset, &data);
 	}
 }
 
 /**
- * nic_get_sset_count - get string set count witch returned by nic_get_strings.
- * @dev: net device
+ * hns_get_sset_count - get string set count returned by nic_get_strings
+ * @netdev: net device
  * @stringset: string set index, 0: self test string; 1: statistics string.
  *
  * Return string set count.
  */
-int hns_get_sset_count(struct net_device *netdev, int stringset)
+static int hns_get_sset_count(struct net_device *netdev, int stringset)
 {
 	struct hns_nic_priv *priv = netdev_priv(netdev);
 	struct hnae_handle *h = priv->ae_handle;
@@ -1006,7 +969,7 @@ int hns_get_sset_count(struct net_device *netdev, int stringset)
 		return -EOPNOTSUPP;
 	}
 	if (stringset == ETH_SS_TEST) {
-		u32 cnt = (sizeof(hns_nic_test_strs) / ETH_GSTRING_LEN);
+		u32 cnt = ARRAY_SIZE(hns_nic_test_strs);
 
 		if (priv->ae_handle->phy_if == PHY_INTERFACE_MODE_XGMII)
 			cnt--;
@@ -1015,19 +978,21 @@ int hns_get_sset_count(struct net_device *netdev, int stringset)
 			cnt--;
 
 		return cnt;
-	} else {
+	} else if (stringset == ETH_SS_STATS) {
 		return (HNS_NET_STATS_CNT + ops->get_sset_count(h, stringset));
+	} else {
+		return -EOPNOTSUPP;
 	}
 }
 
 /**
  * hns_phy_led_set - set phy LED status.
- * @dev: net device
+ * @netdev: net device
  * @value: LED state.
  *
  * Return 0 on success, negative on failure.
  */
-int hns_phy_led_set(struct net_device *netdev, int value)
+static int hns_phy_led_set(struct net_device *netdev, int value)
 {
 	int retval;
 	struct phy_device *phy_dev = netdev->phydev;
@@ -1043,13 +1008,14 @@ int hns_phy_led_set(struct net_device *netdev, int value)
 }
 
 /**
- * nic_set_phys_id - set phy identify LED.
- * @dev: net device
+ * hns_set_phys_id - set phy identify LED.
+ * @netdev: net device
  * @state: LED state.
  *
  * Return 0 on success, negative on failure.
  */
-int hns_set_phys_id(struct net_device *netdev, enum ethtool_phys_id_state state)
+static int
+hns_set_phys_id(struct net_device *netdev, enum ethtool_phys_id_state state)
 {
 	struct hns_nic_priv *priv = netdev_priv(netdev);
 	struct hnae_handle *h = priv->ae_handle;
@@ -1119,12 +1085,12 @@ int hns_set_phys_id(struct net_device *netdev, enum ethtool_phys_id_state state)
 
 /**
  * hns_get_regs - get net device register
- * @dev: net device
+ * @net_dev: net device
  * @cmd: ethtool cmd
- * @date: register data
+ * @data: register data
  */
-void hns_get_regs(struct net_device *net_dev, struct ethtool_regs *cmd,
-		  void *data)
+static void hns_get_regs(struct net_device *net_dev, struct ethtool_regs *cmd,
+			 void *data)
 {
 	struct hns_nic_priv *priv = netdev_priv(net_dev);
 	struct hnae_ae_ops *ops;
@@ -1140,8 +1106,8 @@ void hns_get_regs(struct net_device *net_dev, struct ethtool_regs *cmd,
 }
 
 /**
- * nic_get_regs_len - get total register len.
- * @dev: net device
+ * hns_get_regs_len - get total register len.
+ * @net_dev: net device
  *
  * Return total register len.
  */
@@ -1166,22 +1132,24 @@ static int hns_get_regs_len(struct net_device *net_dev)
 
 /**
  * hns_nic_nway_reset - nway reset
- * @dev: net device
+ * @netdev: net device
  *
  * Return 0 on success, negative on failure
  */
 static int hns_nic_nway_reset(struct net_device *netdev)
 {
-	int ret = 0;
 	struct phy_device *phy = netdev->phydev;
 
-	if (netif_running(netdev)) {
-		/* if autoneg is disabled, don't restart auto-negotiation */
-		if (phy && phy->autoneg == AUTONEG_ENABLE)
-			ret = genphy_restart_aneg(phy);
-	}
+	if (!netif_running(netdev))
+		return 0;
 
-	return ret;
+	if (!phy)
+		return -EOPNOTSUPP;
+
+	if (phy->autoneg != AUTONEG_ENABLE)
+		return -EINVAL;
+
+	return genphy_restart_aneg(phy);
 }
 
 static u32
@@ -1217,7 +1185,7 @@ hns_get_rss_indir_size(struct net_device *netdev)
 }
 
 static int
-hns_get_rss(struct net_device *netdev, u32 *indir, u8 *key, u8 *hfunc)
+hns_get_rss(struct net_device *netdev, struct ethtool_rxfh_param *rxfh)
 {
 	struct hns_nic_priv *priv = netdev_priv(netdev);
 	struct hnae_ae_ops *ops;
@@ -1230,15 +1198,16 @@ hns_get_rss(struct net_device *netdev, u32 *indir, u8 *key, u8 *hfunc)
 
 	ops = priv->ae_handle->dev->ops;
 
-	if (!indir)
+	if (!rxfh->indir)
 		return 0;
 
-	return ops->get_rss(priv->ae_handle, indir, key, hfunc);
+	return ops->get_rss(priv->ae_handle,
+			    rxfh->indir, rxfh->key, &rxfh->hfunc);
 }
 
 static int
-hns_set_rss(struct net_device *netdev, const u32 *indir, const u8 *key,
-	    const u8 hfunc)
+hns_set_rss(struct net_device *netdev, struct ethtool_rxfh_param *rxfh,
+	    struct netlink_ext_ack *extack)
 {
 	struct hns_nic_priv *priv = netdev_priv(netdev);
 	struct hnae_ae_ops *ops;
@@ -1251,12 +1220,14 @@ hns_set_rss(struct net_device *netdev, const u32 *indir, const u8 *key,
 
 	ops = priv->ae_handle->dev->ops;
 
-	if (hfunc != ETH_RSS_HASH_NO_CHANGE && hfunc != ETH_RSS_HASH_TOP) {
+	if (rxfh->hfunc != ETH_RSS_HASH_NO_CHANGE &&
+	    rxfh->hfunc != ETH_RSS_HASH_TOP) {
 		netdev_err(netdev, "Invalid hfunc!\n");
 		return -EOPNOTSUPP;
 	}
 
-	return ops->set_rss(priv->ae_handle, indir, key, hfunc);
+	return ops->set_rss(priv->ae_handle,
+			    rxfh->indir, rxfh->key, rxfh->hfunc);
 }
 
 static int hns_get_rxnfc(struct net_device *netdev,
@@ -1277,6 +1248,11 @@ static int hns_get_rxnfc(struct net_device *netdev,
 }
 
 static const struct ethtool_ops hns_ethtool_ops = {
+	.supported_coalesce_params = ETHTOOL_COALESCE_USECS |
+				     ETHTOOL_COALESCE_MAX_FRAMES |
+				     ETHTOOL_COALESCE_USE_ADAPTIVE |
+				     ETHTOOL_COALESCE_USECS_LOW_HIGH |
+				     ETHTOOL_COALESCE_MAX_FRAMES_LOW_HIGH,
 	.get_drvinfo = hns_nic_get_drvinfo,
 	.get_link  = hns_nic_get_link,
 	.get_ringparam = hns_get_ringparam,
diff --git a/drivers/net/ethernet/hisilicon/hns3/Makefile b/drivers/net/ethernet/hisilicon/hns3/Makefile
new file mode 100644
index 000000000000..e8af26da1fc1
--- /dev/null
+++ b/drivers/net/ethernet/hisilicon/hns3/Makefile
@@ -0,0 +1,28 @@
+# SPDX-License-Identifier: GPL-2.0+
+#
+# Makefile for the HISILICON network device drivers.
+#
+
+ccflags-y += -I$(src)
+ccflags-y += -I$(srctree)/drivers/net/ethernet/hisilicon/hns3/hns3pf
+ccflags-y += -I$(srctree)/drivers/net/ethernet/hisilicon/hns3/hns3vf
+ccflags-y += -I$(srctree)/drivers/net/ethernet/hisilicon/hns3/hns3_common
+
+obj-$(CONFIG_HNS3) += hnae3.o
+
+obj-$(CONFIG_HNS3_ENET) += hns3.o
+hns3-objs = hns3_enet.o hns3_ethtool.o hns3_debugfs.o
+
+hns3-$(CONFIG_HNS3_DCB) += hns3_dcbnl.o
+
+obj-$(CONFIG_HNS3_HCLGEVF) += hclgevf.o hclge-common.o
+
+hclge-common-objs += hns3_common/hclge_comm_cmd.o hns3_common/hclge_comm_rss.o hns3_common/hclge_comm_tqp_stats.o
+
+hclgevf-objs = hns3vf/hclgevf_main.o hns3vf/hclgevf_mbx.o  hns3vf/hclgevf_devlink.o hns3vf/hclgevf_regs.o
+
+obj-$(CONFIG_HNS3_HCLGE) += hclge.o hclge-common.o
+hclge-objs = hns3pf/hclge_main.o hns3pf/hclge_mdio.o hns3pf/hclge_tm.o hns3pf/hclge_regs.o \
+		hns3pf/hclge_mbx.o hns3pf/hclge_err.o  hns3pf/hclge_debugfs.o hns3pf/hclge_ptp.o hns3pf/hclge_devlink.o \
+
+hclge-$(CONFIG_HNS3_DCB) += hns3pf/hclge_dcb.o
diff --git a/drivers/net/ethernet/hisilicon/hns3/hclge_mbx.h b/drivers/net/ethernet/hisilicon/hns3/hclge_mbx.h
new file mode 100644
index 000000000000..abcd7877f7d2
--- /dev/null
+++ b/drivers/net/ethernet/hisilicon/hns3/hclge_mbx.h
@@ -0,0 +1,256 @@
+/* SPDX-License-Identifier: GPL-2.0+ */
+/* Copyright (c) 2016-2017 Hisilicon Limited. */
+
+#ifndef __HCLGE_MBX_H
+#define __HCLGE_MBX_H
+#include <linux/init.h>
+#include <linux/mutex.h>
+#include <linux/types.h>
+
+enum HCLGE_MBX_OPCODE {
+	HCLGE_MBX_RESET = 0x01,		/* (VF -> PF) assert reset */
+	HCLGE_MBX_ASSERTING_RESET,	/* (PF -> VF) PF is asserting reset */
+	HCLGE_MBX_SET_UNICAST,		/* (VF -> PF) set UC addr */
+	HCLGE_MBX_SET_MULTICAST,	/* (VF -> PF) set MC addr */
+	HCLGE_MBX_SET_VLAN,		/* (VF -> PF) set VLAN */
+	HCLGE_MBX_MAP_RING_TO_VECTOR,	/* (VF -> PF) map ring-to-vector */
+	HCLGE_MBX_UNMAP_RING_TO_VECTOR,	/* (VF -> PF) unamp ring-to-vector */
+	HCLGE_MBX_SET_PROMISC_MODE,	/* (VF -> PF) set promiscuous mode */
+	HCLGE_MBX_SET_MACVLAN,		/* (VF -> PF) set unicast filter */
+	HCLGE_MBX_API_NEGOTIATE,	/* (VF -> PF) negotiate API version */
+	HCLGE_MBX_GET_QINFO,		/* (VF -> PF) get queue config */
+	HCLGE_MBX_GET_QDEPTH,		/* (VF -> PF) get queue depth */
+	HCLGE_MBX_GET_BASIC_INFO,	/* (VF -> PF) get basic info */
+	HCLGE_MBX_GET_RETA,		/* (VF -> PF) get RETA */
+	HCLGE_MBX_GET_RSS_KEY,		/* (VF -> PF) get RSS key */
+	HCLGE_MBX_GET_MAC_ADDR,		/* (VF -> PF) get MAC addr */
+	HCLGE_MBX_PF_VF_RESP,		/* (PF -> VF) generate response to VF */
+	HCLGE_MBX_GET_BDNUM,		/* (VF -> PF) get BD num */
+	HCLGE_MBX_GET_BUFSIZE,		/* (VF -> PF) get buffer size */
+	HCLGE_MBX_GET_STREAMID,		/* (VF -> PF) get stream id */
+	HCLGE_MBX_SET_AESTART,		/* (VF -> PF) start ae */
+	HCLGE_MBX_SET_TSOSTATS,		/* (VF -> PF) get tso stats */
+	HCLGE_MBX_LINK_STAT_CHANGE,	/* (PF -> VF) link status has changed */
+	HCLGE_MBX_GET_BASE_CONFIG,	/* (VF -> PF) get config */
+	HCLGE_MBX_BIND_FUNC_QUEUE,	/* (VF -> PF) bind function and queue */
+	HCLGE_MBX_GET_LINK_STATUS,	/* (VF -> PF) get link status */
+	HCLGE_MBX_QUEUE_RESET,		/* (VF -> PF) reset queue */
+	HCLGE_MBX_KEEP_ALIVE,		/* (VF -> PF) send keep alive cmd */
+	HCLGE_MBX_SET_ALIVE,		/* (VF -> PF) set alive state */
+	HCLGE_MBX_SET_MTU,		/* (VF -> PF) set mtu */
+	HCLGE_MBX_GET_QID_IN_PF,	/* (VF -> PF) get queue id in pf */
+	HCLGE_MBX_LINK_STAT_MODE,	/* (PF -> VF) link mode has changed */
+	HCLGE_MBX_GET_LINK_MODE,	/* (VF -> PF) get the link mode of pf */
+	HCLGE_MBX_PUSH_VLAN_INFO,	/* (PF -> VF) push port base vlan */
+	HCLGE_MBX_GET_MEDIA_TYPE,       /* (VF -> PF) get media type */
+	HCLGE_MBX_PUSH_PROMISC_INFO,	/* (PF -> VF) push vf promisc info */
+	HCLGE_MBX_VF_UNINIT,            /* (VF -> PF) vf is unintializing */
+	HCLGE_MBX_HANDLE_VF_TBL,	/* (VF -> PF) store/clear hw table */
+	HCLGE_MBX_GET_RING_VECTOR_MAP,	/* (VF -> PF) get ring-to-vector map */
+
+	HCLGE_MBX_GET_VF_FLR_STATUS = 200, /* (M7 -> PF) get vf flr status */
+	HCLGE_MBX_PUSH_LINK_STATUS,	/* (M7 -> PF) get port link status */
+	HCLGE_MBX_NCSI_ERROR,		/* (M7 -> PF) receive a NCSI error */
+};
+
+/* below are per-VF mac-vlan subcodes */
+enum hclge_mbx_mac_vlan_subcode {
+	HCLGE_MBX_MAC_VLAN_UC_MODIFY = 0,	/* modify UC mac addr */
+	HCLGE_MBX_MAC_VLAN_UC_ADD,		/* add a new UC mac addr */
+	HCLGE_MBX_MAC_VLAN_UC_REMOVE,		/* remove a new UC mac addr */
+	HCLGE_MBX_MAC_VLAN_MC_MODIFY,		/* modify MC mac addr */
+	HCLGE_MBX_MAC_VLAN_MC_ADD,		/* add new MC mac addr */
+	HCLGE_MBX_MAC_VLAN_MC_REMOVE,		/* remove MC mac addr */
+};
+
+/* below are per-VF vlan cfg subcodes */
+enum hclge_mbx_vlan_cfg_subcode {
+	HCLGE_MBX_VLAN_FILTER = 0,	/* set vlan filter */
+	HCLGE_MBX_VLAN_TX_OFF_CFG,	/* set tx side vlan offload */
+	HCLGE_MBX_VLAN_RX_OFF_CFG,	/* set rx side vlan offload */
+	HCLGE_MBX_PORT_BASE_VLAN_CFG,	/* set port based vlan configuration */
+	HCLGE_MBX_GET_PORT_BASE_VLAN_STATE,	/* get port based vlan state */
+	HCLGE_MBX_ENABLE_VLAN_FILTER,
+};
+
+enum hclge_mbx_tbl_cfg_subcode {
+	HCLGE_MBX_VPORT_LIST_CLEAR,
+};
+
+#define HCLGE_MBX_MAX_MSG_SIZE	14
+#define HCLGE_MBX_MAX_RESP_DATA_SIZE	8U
+#define HCLGE_MBX_MAX_RING_CHAIN_PARAM_NUM	4
+
+#define HCLGE_RESET_SCHED_TIMEOUT	(3 * HZ)
+#define HCLGE_MBX_SCHED_TIMEOUT	(HZ / 2)
+
+struct hclge_ring_chain_param {
+	u8 ring_type;
+	u8 tqp_index;
+	u8 int_gl_index;
+};
+
+struct hclge_basic_info {
+	u8 hw_tc_map;
+	u8 rsv;
+	__le16 mbx_api_version;
+	__le32 pf_caps;
+};
+
+struct hclgevf_mbx_resp_status {
+	struct mutex mbx_mutex; /* protects against contending sync cmd resp */
+	u32 origin_mbx_msg;
+	bool received_resp;
+	int resp_status;
+	u16 match_id;
+	u8 additional_info[HCLGE_MBX_MAX_RESP_DATA_SIZE];
+};
+
+struct hclge_respond_to_vf_msg {
+	int status;
+	u8 data[HCLGE_MBX_MAX_RESP_DATA_SIZE];
+	u16 len;
+};
+
+struct hclge_vf_to_pf_msg {
+	u8 code;
+	union {
+		struct {
+			u8 subcode;
+			u8 data[HCLGE_MBX_MAX_MSG_SIZE];
+		};
+		struct {
+			u8 en_bc;
+			u8 en_uc;
+			u8 en_mc;
+			u8 en_limit_promisc;
+		};
+		struct {
+			u8 vector_id;
+			u8 ring_num;
+			struct hclge_ring_chain_param
+				param[HCLGE_MBX_MAX_RING_CHAIN_PARAM_NUM];
+		};
+	};
+};
+
+struct hclge_pf_to_vf_msg {
+	__le16 code;
+	union {
+		/* used for mbx response */
+		struct {
+			__le16 vf_mbx_msg_code;
+			__le16 vf_mbx_msg_subcode;
+			__le16 resp_status;
+			u8 resp_data[HCLGE_MBX_MAX_RESP_DATA_SIZE];
+		};
+		/* used for general mbx */
+		struct {
+			u8 msg_data[HCLGE_MBX_MAX_MSG_SIZE];
+		};
+	};
+};
+
+struct hclge_mbx_vf_to_pf_cmd {
+	u8 rsv;
+	u8 mbx_src_vfid; /* Auto filled by IMP */
+	u8 mbx_need_resp;
+	u8 rsv1[1];
+	u8 msg_len;
+	u8 rsv2;
+	__le16 match_id;
+	struct hclge_vf_to_pf_msg msg;
+};
+
+#define HCLGE_MBX_NEED_RESP_B		0
+
+struct hclge_mbx_pf_to_vf_cmd {
+	u8 dest_vfid;
+	u8 rsv[3];
+	u8 msg_len;
+	u8 rsv1;
+	__le16 match_id;
+	struct hclge_pf_to_vf_msg msg;
+};
+
+struct hclge_vf_rst_cmd {
+	u8 dest_vfid;
+	u8 vf_rst;
+	u8 rsv[22];
+};
+
+#pragma pack(1)
+struct hclge_mbx_link_status {
+	__le16 link_status;
+	__le32 speed;
+	__le16 duplex;
+	u8 flag;
+};
+
+struct hclge_mbx_link_mode {
+	__le16 idx;
+	__le64 link_mode;
+};
+
+struct hclge_mbx_port_base_vlan {
+	__le16 state;
+	__le16 vlan_proto;
+	__le16 qos;
+	__le16 vlan_tag;
+};
+
+struct hclge_mbx_vf_queue_info {
+	__le16 num_tqps;
+	__le16 rss_size;
+	__le16 rx_buf_len;
+};
+
+struct hclge_mbx_vf_queue_depth {
+	__le16 num_tx_desc;
+	__le16 num_rx_desc;
+};
+
+struct hclge_mbx_vlan_filter {
+	u8 is_kill;
+	__le16 vlan_id;
+	__le16 proto;
+};
+
+struct hclge_mbx_mtu_info {
+	__le32 mtu;
+};
+
+#pragma pack()
+
+/* used by VF to store the received Async responses from PF */
+struct hclgevf_mbx_arq_ring {
+#define HCLGE_MBX_MAX_ARQ_MSG_SIZE	8
+#define HCLGE_MBX_MAX_ARQ_MSG_NUM	1024
+	struct hclgevf_dev *hdev;
+	u32 head;
+	u32 tail;
+	atomic_t count;
+	__le16 msg_q[HCLGE_MBX_MAX_ARQ_MSG_NUM][HCLGE_MBX_MAX_ARQ_MSG_SIZE];
+};
+
+struct hclge_dev;
+
+#define HCLGE_MBX_OPCODE_MAX 256
+struct hclge_mbx_ops_param {
+	struct hclge_vport *vport;
+	struct hclge_mbx_vf_to_pf_cmd *req;
+	struct hclge_respond_to_vf_msg *resp_msg;
+};
+
+typedef int (*hclge_mbx_ops_fn)(struct hclge_mbx_ops_param *param);
+
+#define hclge_mbx_ring_ptr_move_crq(crq) \
+	(crq->next_to_use = (crq->next_to_use + 1) % crq->desc_num)
+#define hclge_mbx_tail_ptr_move_arq(arq) \
+		(arq.tail = (arq.tail + 1) % HCLGE_MBX_MAX_ARQ_MSG_NUM)
+#define hclge_mbx_head_ptr_move_arq(arq) \
+		(arq.head = (arq.head + 1) % HCLGE_MBX_MAX_ARQ_MSG_NUM)
+
+/* PF immediately push link status to VFs when link status changed */
+#define HCLGE_MBX_PUSH_LINK_STATUS_EN			BIT(0)
+#endif
diff --git a/drivers/net/ethernet/hisilicon/hns3/hnae3.c b/drivers/net/ethernet/hisilicon/hns3/hnae3.c
new file mode 100644
index 000000000000..b25fb400f476
--- /dev/null
+++ b/drivers/net/ethernet/hisilicon/hns3/hnae3.c
@@ -0,0 +1,410 @@
+// SPDX-License-Identifier: GPL-2.0+
+// Copyright (c) 2016-2017 Hisilicon Limited.
+
+#include <linux/list.h>
+#include <linux/spinlock.h>
+
+#include "hnae3.h"
+
+static LIST_HEAD(hnae3_ae_algo_list);
+static LIST_HEAD(hnae3_client_list);
+static LIST_HEAD(hnae3_ae_dev_list);
+
+void hnae3_unregister_ae_algo_prepare(struct hnae3_ae_algo *ae_algo)
+{
+	const struct pci_device_id *pci_id;
+	struct hnae3_ae_dev *ae_dev;
+
+	if (!ae_algo)
+		return;
+
+	list_for_each_entry(ae_dev, &hnae3_ae_dev_list, node) {
+		if (!hnae3_get_bit(ae_dev->flag, HNAE3_DEV_INITED_B))
+			continue;
+
+		pci_id = pci_match_id(ae_algo->pdev_id_table, ae_dev->pdev);
+		if (!pci_id)
+			continue;
+		if (IS_ENABLED(CONFIG_PCI_IOV)) {
+			device_lock(&ae_dev->pdev->dev);
+			pci_disable_sriov(ae_dev->pdev);
+			device_unlock(&ae_dev->pdev->dev);
+		}
+	}
+}
+EXPORT_SYMBOL(hnae3_unregister_ae_algo_prepare);
+
+/* we are keeping things simple and using single lock for all the
+ * list. This is a non-critical code so other updations, if happen
+ * in parallel, can wait.
+ */
+static DEFINE_MUTEX(hnae3_common_lock);
+
+/* ensure the drivers being unloaded one by one */
+static DEFINE_MUTEX(hnae3_unload_lock);
+
+void hnae3_acquire_unload_lock(void)
+{
+	mutex_lock(&hnae3_unload_lock);
+}
+EXPORT_SYMBOL(hnae3_acquire_unload_lock);
+
+void hnae3_release_unload_lock(void)
+{
+	mutex_unlock(&hnae3_unload_lock);
+}
+EXPORT_SYMBOL(hnae3_release_unload_lock);
+
+static bool hnae3_client_match(enum hnae3_client_type client_type)
+{
+	if (client_type == HNAE3_CLIENT_KNIC ||
+	    client_type == HNAE3_CLIENT_ROCE)
+		return true;
+
+	return false;
+}
+
+void hnae3_set_client_init_flag(struct hnae3_client *client,
+				struct hnae3_ae_dev *ae_dev,
+				unsigned int inited)
+{
+	if (!client || !ae_dev)
+		return;
+
+	switch (client->type) {
+	case HNAE3_CLIENT_KNIC:
+		hnae3_set_bit(ae_dev->flag, HNAE3_KNIC_CLIENT_INITED_B, inited);
+		break;
+	case HNAE3_CLIENT_ROCE:
+		hnae3_set_bit(ae_dev->flag, HNAE3_ROCE_CLIENT_INITED_B, inited);
+		break;
+	default:
+		break;
+	}
+}
+EXPORT_SYMBOL(hnae3_set_client_init_flag);
+
+static int hnae3_get_client_init_flag(struct hnae3_client *client,
+				      struct hnae3_ae_dev *ae_dev)
+{
+	int inited = 0;
+
+	switch (client->type) {
+	case HNAE3_CLIENT_KNIC:
+		inited = hnae3_get_bit(ae_dev->flag,
+				       HNAE3_KNIC_CLIENT_INITED_B);
+		break;
+	case HNAE3_CLIENT_ROCE:
+		inited = hnae3_get_bit(ae_dev->flag,
+				       HNAE3_ROCE_CLIENT_INITED_B);
+		break;
+	default:
+		break;
+	}
+
+	return inited;
+}
+
+static int hnae3_init_client_instance(struct hnae3_client *client,
+				      struct hnae3_ae_dev *ae_dev)
+{
+	int ret;
+
+	/* check if this client matches the type of ae_dev */
+	if (!(hnae3_client_match(client->type) &&
+	      hnae3_get_bit(ae_dev->flag, HNAE3_DEV_INITED_B))) {
+		return 0;
+	}
+
+	ret = ae_dev->ops->init_client_instance(client, ae_dev);
+	if (ret)
+		dev_err(&ae_dev->pdev->dev,
+			"fail to instantiate client, ret = %d\n", ret);
+
+	return ret;
+}
+
+static void hnae3_uninit_client_instance(struct hnae3_client *client,
+					 struct hnae3_ae_dev *ae_dev)
+{
+	/* check if this client matches the type of ae_dev */
+	if (!(hnae3_client_match(client->type) &&
+	      hnae3_get_bit(ae_dev->flag, HNAE3_DEV_INITED_B)))
+		return;
+
+	if (hnae3_get_client_init_flag(client, ae_dev)) {
+		ae_dev->ops->uninit_client_instance(client, ae_dev);
+
+		hnae3_set_client_init_flag(client, ae_dev, 0);
+	}
+}
+
+int hnae3_register_client(struct hnae3_client *client)
+{
+	struct hnae3_client *client_tmp;
+	struct hnae3_ae_dev *ae_dev;
+
+	if (!client)
+		return -ENODEV;
+
+	mutex_lock(&hnae3_common_lock);
+	/* one system should only have one client for every type */
+	list_for_each_entry(client_tmp, &hnae3_client_list, node) {
+		if (client_tmp->type == client->type)
+			goto exit;
+	}
+
+	list_add_tail(&client->node, &hnae3_client_list);
+
+	/* initialize the client on every matched port */
+	list_for_each_entry(ae_dev, &hnae3_ae_dev_list, node) {
+		/* if the client could not be initialized on current port, for
+		 * any error reasons, move on to next available port
+		 */
+		int ret = hnae3_init_client_instance(client, ae_dev);
+		if (ret)
+			dev_err(&ae_dev->pdev->dev,
+				"match and instantiation failed for port, ret = %d\n",
+				ret);
+	}
+
+exit:
+	mutex_unlock(&hnae3_common_lock);
+
+	return 0;
+}
+EXPORT_SYMBOL(hnae3_register_client);
+
+void hnae3_unregister_client(struct hnae3_client *client)
+{
+	struct hnae3_client *client_tmp;
+	struct hnae3_ae_dev *ae_dev;
+	bool existed = false;
+
+	if (!client)
+		return;
+
+	mutex_lock(&hnae3_common_lock);
+	/* one system should only have one client for every type */
+	list_for_each_entry(client_tmp, &hnae3_client_list, node) {
+		if (client_tmp->type == client->type) {
+			existed = true;
+			break;
+		}
+	}
+
+	if (!existed) {
+		mutex_unlock(&hnae3_common_lock);
+		pr_err("client %s does not exist!\n", client->name);
+		return;
+	}
+
+	/* un-initialize the client on every matched port */
+	list_for_each_entry(ae_dev, &hnae3_ae_dev_list, node) {
+		hnae3_uninit_client_instance(client, ae_dev);
+	}
+
+	list_del(&client->node);
+	mutex_unlock(&hnae3_common_lock);
+}
+EXPORT_SYMBOL(hnae3_unregister_client);
+
+/* hnae3_register_ae_algo - register a AE algorithm to hnae3 framework
+ * @ae_algo: AE algorithm
+ * NOTE: the duplicated name will not be checked
+ */
+void hnae3_register_ae_algo(struct hnae3_ae_algo *ae_algo)
+{
+	const struct pci_device_id *id;
+	struct hnae3_ae_dev *ae_dev;
+	struct hnae3_client *client;
+	int ret;
+
+	if (!ae_algo)
+		return;
+
+	mutex_lock(&hnae3_common_lock);
+
+	list_add_tail(&ae_algo->node, &hnae3_ae_algo_list);
+
+	/* Check if this algo/ops matches the list of ae_devs */
+	list_for_each_entry(ae_dev, &hnae3_ae_dev_list, node) {
+		id = pci_match_id(ae_algo->pdev_id_table, ae_dev->pdev);
+		if (!id)
+			continue;
+
+		if (!ae_algo->ops) {
+			dev_err(&ae_dev->pdev->dev, "ae_algo ops are null\n");
+			continue;
+		}
+		ae_dev->ops = ae_algo->ops;
+
+		ret = ae_algo->ops->init_ae_dev(ae_dev);
+		if (ret) {
+			dev_err(&ae_dev->pdev->dev,
+				"init ae_dev error, ret = %d\n", ret);
+			continue;
+		}
+
+		/* ae_dev init should set flag */
+		hnae3_set_bit(ae_dev->flag, HNAE3_DEV_INITED_B, 1);
+
+		/* check the client list for the match with this ae_dev type and
+		 * initialize the figure out client instance
+		 */
+		list_for_each_entry(client, &hnae3_client_list, node) {
+			ret = hnae3_init_client_instance(client, ae_dev);
+			if (ret)
+				dev_err(&ae_dev->pdev->dev,
+					"match and instantiation failed, ret = %d\n",
+					ret);
+		}
+	}
+
+	mutex_unlock(&hnae3_common_lock);
+}
+EXPORT_SYMBOL(hnae3_register_ae_algo);
+
+/* hnae3_unregister_ae_algo - unregisters a AE algorithm
+ * @ae_algo: the AE algorithm to unregister
+ */
+void hnae3_unregister_ae_algo(struct hnae3_ae_algo *ae_algo)
+{
+	const struct pci_device_id *id;
+	struct hnae3_ae_dev *ae_dev;
+	struct hnae3_client *client;
+
+	if (!ae_algo)
+		return;
+
+	mutex_lock(&hnae3_common_lock);
+	/* Check if there are matched ae_dev */
+	list_for_each_entry(ae_dev, &hnae3_ae_dev_list, node) {
+		if (!hnae3_get_bit(ae_dev->flag, HNAE3_DEV_INITED_B))
+			continue;
+
+		id = pci_match_id(ae_algo->pdev_id_table, ae_dev->pdev);
+		if (!id)
+			continue;
+
+		/* check the client list for the match with this ae_dev type and
+		 * un-initialize the figure out client instance
+		 */
+		list_for_each_entry(client, &hnae3_client_list, node)
+			hnae3_uninit_client_instance(client, ae_dev);
+
+		ae_algo->ops->uninit_ae_dev(ae_dev);
+		hnae3_set_bit(ae_dev->flag, HNAE3_DEV_INITED_B, 0);
+		ae_dev->ops = NULL;
+	}
+
+	list_del(&ae_algo->node);
+	mutex_unlock(&hnae3_common_lock);
+}
+EXPORT_SYMBOL(hnae3_unregister_ae_algo);
+
+/* hnae3_register_ae_dev - registers a AE device to hnae3 framework
+ * @ae_dev: the AE device
+ * NOTE: the duplicated name will not be checked
+ */
+int hnae3_register_ae_dev(struct hnae3_ae_dev *ae_dev)
+{
+	const struct pci_device_id *id;
+	struct hnae3_ae_algo *ae_algo;
+	struct hnae3_client *client;
+	int ret;
+
+	if (!ae_dev)
+		return -ENODEV;
+
+	mutex_lock(&hnae3_common_lock);
+
+	list_add_tail(&ae_dev->node, &hnae3_ae_dev_list);
+
+	/* Check if there are matched ae_algo */
+	list_for_each_entry(ae_algo, &hnae3_ae_algo_list, node) {
+		id = pci_match_id(ae_algo->pdev_id_table, ae_dev->pdev);
+		if (!id)
+			continue;
+
+		if (!ae_algo->ops) {
+			dev_err(&ae_dev->pdev->dev, "ae_algo ops are null\n");
+			ret = -EOPNOTSUPP;
+			goto out_err;
+		}
+		ae_dev->ops = ae_algo->ops;
+
+		ret = ae_dev->ops->init_ae_dev(ae_dev);
+		if (ret) {
+			dev_err(&ae_dev->pdev->dev,
+				"init ae_dev error, ret = %d\n", ret);
+			goto out_err;
+		}
+
+		/* ae_dev init should set flag */
+		hnae3_set_bit(ae_dev->flag, HNAE3_DEV_INITED_B, 1);
+		break;
+	}
+
+	/* check the client list for the match with this ae_dev type and
+	 * initialize the figure out client instance
+	 */
+	list_for_each_entry(client, &hnae3_client_list, node) {
+		ret = hnae3_init_client_instance(client, ae_dev);
+		if (ret)
+			dev_err(&ae_dev->pdev->dev,
+				"match and instantiation failed, ret = %d\n",
+				ret);
+	}
+
+	mutex_unlock(&hnae3_common_lock);
+
+	return 0;
+
+out_err:
+	list_del(&ae_dev->node);
+	mutex_unlock(&hnae3_common_lock);
+
+	return ret;
+}
+EXPORT_SYMBOL(hnae3_register_ae_dev);
+
+/* hnae3_unregister_ae_dev - unregisters a AE device
+ * @ae_dev: the AE device to unregister
+ */
+void hnae3_unregister_ae_dev(struct hnae3_ae_dev *ae_dev)
+{
+	const struct pci_device_id *id;
+	struct hnae3_ae_algo *ae_algo;
+	struct hnae3_client *client;
+
+	if (!ae_dev)
+		return;
+
+	mutex_lock(&hnae3_common_lock);
+	/* Check if there are matched ae_algo */
+	list_for_each_entry(ae_algo, &hnae3_ae_algo_list, node) {
+		if (!hnae3_get_bit(ae_dev->flag, HNAE3_DEV_INITED_B))
+			continue;
+
+		id = pci_match_id(ae_algo->pdev_id_table, ae_dev->pdev);
+		if (!id)
+			continue;
+
+		list_for_each_entry(client, &hnae3_client_list, node)
+			hnae3_uninit_client_instance(client, ae_dev);
+
+		ae_algo->ops->uninit_ae_dev(ae_dev);
+		hnae3_set_bit(ae_dev->flag, HNAE3_DEV_INITED_B, 0);
+		ae_dev->ops = NULL;
+	}
+
+	list_del(&ae_dev->node);
+	mutex_unlock(&hnae3_common_lock);
+}
+EXPORT_SYMBOL(hnae3_unregister_ae_dev);
+
+MODULE_AUTHOR("Huawei Tech. Co., Ltd.");
+MODULE_LICENSE("GPL");
+MODULE_DESCRIPTION("HNAE3(Hisilicon Network Acceleration Engine) Framework");
+MODULE_VERSION(HNAE3_MOD_VERSION);
diff --git a/drivers/net/ethernet/hisilicon/hns3/hnae3.h b/drivers/net/ethernet/hisilicon/hns3/hnae3.h
new file mode 100644
index 000000000000..3b548f71fa8a
--- /dev/null
+++ b/drivers/net/ethernet/hisilicon/hns3/hnae3.h
@@ -0,0 +1,976 @@
+/* SPDX-License-Identifier: GPL-2.0+ */
+// Copyright (c) 2016-2017 Hisilicon Limited.
+
+#ifndef __HNAE3_H
+#define __HNAE3_H
+
+/* Names used in this framework:
+ *      ae handle (handle):
+ *        a set of queues provided by AE
+ *      ring buffer queue (rbq):
+ *        the channel between upper layer and the AE, can do tx and rx
+ *      ring:
+ *        a tx or rx channel within a rbq
+ *      ring description (desc):
+ *        an element in the ring with packet information
+ *      buffer:
+ *        a memory region referred by desc with the full packet payload
+ *
+ * "num" means a static number set as a parameter, "count" mean a dynamic
+ *   number set while running
+ * "cb" means control block
+ */
+
+#include <linux/acpi.h>
+#include <linux/dcbnl.h>
+#include <linux/delay.h>
+#include <linux/device.h>
+#include <linux/ethtool.h>
+#include <linux/module.h>
+#include <linux/netdevice.h>
+#include <linux/pci.h>
+#include <linux/pkt_sched.h>
+#include <linux/types.h>
+#include <linux/bitmap.h>
+#include <net/pkt_cls.h>
+#include <net/pkt_sched.h>
+
+#define HNAE3_MOD_VERSION "1.0"
+
+#define HNAE3_MIN_VECTOR_NUM	2 /* first one for misc, another for IO */
+
+/* Device version */
+#define HNAE3_DEVICE_VERSION_V1   0x00020
+#define HNAE3_DEVICE_VERSION_V2   0x00021
+#define HNAE3_DEVICE_VERSION_V3   0x00030
+
+#define HNAE3_PCI_REVISION_BIT_SIZE		8
+
+/* Device IDs */
+#define HNAE3_DEV_ID_GE				0xA220
+#define HNAE3_DEV_ID_25GE			0xA221
+#define HNAE3_DEV_ID_25GE_RDMA			0xA222
+#define HNAE3_DEV_ID_25GE_RDMA_MACSEC		0xA223
+#define HNAE3_DEV_ID_50GE_RDMA			0xA224
+#define HNAE3_DEV_ID_50GE_RDMA_MACSEC		0xA225
+#define HNAE3_DEV_ID_100G_RDMA_MACSEC		0xA226
+#define HNAE3_DEV_ID_200G_RDMA			0xA228
+#define HNAE3_DEV_ID_VF				0xA22E
+#define HNAE3_DEV_ID_RDMA_DCB_PFC_VF		0xA22F
+
+#define HNAE3_CLASS_NAME_SIZE 16
+
+#define HNAE3_DEV_INITED_B			0x0
+#define HNAE3_DEV_SUPPORT_ROCE_B		0x1
+#define HNAE3_DEV_SUPPORT_DCB_B			0x2
+#define HNAE3_KNIC_CLIENT_INITED_B		0x3
+#define HNAE3_UNIC_CLIENT_INITED_B		0x4
+#define HNAE3_ROCE_CLIENT_INITED_B		0x5
+
+#define HNAE3_DEV_SUPPORT_ROCE_DCB_BITS (BIT(HNAE3_DEV_SUPPORT_DCB_B) | \
+		BIT(HNAE3_DEV_SUPPORT_ROCE_B))
+
+#define hnae3_dev_roce_supported(hdev) \
+	hnae3_get_bit((hdev)->ae_dev->flag, HNAE3_DEV_SUPPORT_ROCE_B)
+
+#define hnae3_dev_dcb_supported(hdev) \
+	hnae3_get_bit((hdev)->ae_dev->flag, HNAE3_DEV_SUPPORT_DCB_B)
+
+enum HNAE3_DEV_CAP_BITS {
+	HNAE3_DEV_SUPPORT_FD_B,
+	HNAE3_DEV_SUPPORT_GRO_B,
+	HNAE3_DEV_SUPPORT_FEC_B,
+	HNAE3_DEV_SUPPORT_UDP_GSO_B,
+	HNAE3_DEV_SUPPORT_QB_B,
+	HNAE3_DEV_SUPPORT_FD_FORWARD_TC_B,
+	HNAE3_DEV_SUPPORT_PTP_B,
+	HNAE3_DEV_SUPPORT_INT_QL_B,
+	HNAE3_DEV_SUPPORT_HW_TX_CSUM_B,
+	HNAE3_DEV_SUPPORT_TX_PUSH_B,
+	HNAE3_DEV_SUPPORT_PHY_IMP_B,
+	HNAE3_DEV_SUPPORT_TQP_TXRX_INDEP_B,
+	HNAE3_DEV_SUPPORT_HW_PAD_B,
+	HNAE3_DEV_SUPPORT_STASH_B,
+	HNAE3_DEV_SUPPORT_UDP_TUNNEL_CSUM_B,
+	HNAE3_DEV_SUPPORT_PAUSE_B,
+	HNAE3_DEV_SUPPORT_RAS_IMP_B,
+	HNAE3_DEV_SUPPORT_RXD_ADV_LAYOUT_B,
+	HNAE3_DEV_SUPPORT_PORT_VLAN_BYPASS_B,
+	HNAE3_DEV_SUPPORT_VLAN_FLTR_MDF_B,
+	HNAE3_DEV_SUPPORT_MC_MAC_MNG_B,
+	HNAE3_DEV_SUPPORT_CQ_B,
+	HNAE3_DEV_SUPPORT_FEC_STATS_B,
+	HNAE3_DEV_SUPPORT_LANE_NUM_B,
+	HNAE3_DEV_SUPPORT_WOL_B,
+	HNAE3_DEV_SUPPORT_TM_FLUSH_B,
+	HNAE3_DEV_SUPPORT_VF_FAULT_B,
+	HNAE3_DEV_SUPPORT_ERR_MOD_GEN_REG_B,
+};
+
+#define hnae3_ae_dev_fd_supported(ae_dev) \
+		test_bit(HNAE3_DEV_SUPPORT_FD_B, (ae_dev)->caps)
+
+#define hnae3_ae_dev_gro_supported(ae_dev) \
+		test_bit(HNAE3_DEV_SUPPORT_GRO_B, (ae_dev)->caps)
+
+#define hnae3_dev_fec_supported(hdev) \
+	test_bit(HNAE3_DEV_SUPPORT_FEC_B, (hdev)->ae_dev->caps)
+
+#define hnae3_dev_udp_gso_supported(hdev) \
+	test_bit(HNAE3_DEV_SUPPORT_UDP_GSO_B, (hdev)->ae_dev->caps)
+
+#define hnae3_dev_qb_supported(hdev) \
+	test_bit(HNAE3_DEV_SUPPORT_QB_B, (hdev)->ae_dev->caps)
+
+#define hnae3_dev_fd_forward_tc_supported(hdev) \
+	test_bit(HNAE3_DEV_SUPPORT_FD_FORWARD_TC_B, (hdev)->ae_dev->caps)
+
+#define hnae3_dev_ptp_supported(hdev) \
+	test_bit(HNAE3_DEV_SUPPORT_PTP_B, (hdev)->ae_dev->caps)
+
+#define hnae3_dev_int_ql_supported(hdev) \
+	test_bit(HNAE3_DEV_SUPPORT_INT_QL_B, (hdev)->ae_dev->caps)
+
+#define hnae3_dev_hw_csum_supported(hdev) \
+	test_bit(HNAE3_DEV_SUPPORT_HW_TX_CSUM_B, (hdev)->ae_dev->caps)
+
+#define hnae3_dev_tx_push_supported(hdev) \
+	test_bit(HNAE3_DEV_SUPPORT_TX_PUSH_B, (hdev)->ae_dev->caps)
+
+#define hnae3_dev_phy_imp_supported(hdev) \
+	test_bit(HNAE3_DEV_SUPPORT_PHY_IMP_B, (hdev)->ae_dev->caps)
+
+#define hnae3_dev_ras_imp_supported(hdev) \
+	test_bit(HNAE3_DEV_SUPPORT_RAS_IMP_B, (hdev)->ae_dev->caps)
+
+#define hnae3_dev_tqp_txrx_indep_supported(hdev) \
+	test_bit(HNAE3_DEV_SUPPORT_TQP_TXRX_INDEP_B, (hdev)->ae_dev->caps)
+
+#define hnae3_dev_hw_pad_supported(hdev) \
+	test_bit(HNAE3_DEV_SUPPORT_HW_PAD_B, (hdev)->ae_dev->caps)
+
+#define hnae3_dev_stash_supported(hdev) \
+	test_bit(HNAE3_DEV_SUPPORT_STASH_B, (hdev)->ae_dev->caps)
+
+#define hnae3_dev_pause_supported(hdev) \
+	test_bit(HNAE3_DEV_SUPPORT_PAUSE_B, (hdev)->ae_dev->caps)
+
+#define hnae3_ae_dev_tqp_txrx_indep_supported(ae_dev) \
+	test_bit(HNAE3_DEV_SUPPORT_TQP_TXRX_INDEP_B, (ae_dev)->caps)
+
+#define hnae3_ae_dev_rxd_adv_layout_supported(ae_dev) \
+	test_bit(HNAE3_DEV_SUPPORT_RXD_ADV_LAYOUT_B, (ae_dev)->caps)
+
+#define hnae3_ae_dev_mc_mac_mng_supported(ae_dev) \
+	test_bit(HNAE3_DEV_SUPPORT_MC_MAC_MNG_B, (ae_dev)->caps)
+
+#define hnae3_ae_dev_cq_supported(ae_dev) \
+	test_bit(HNAE3_DEV_SUPPORT_CQ_B, (ae_dev)->caps)
+
+#define hnae3_ae_dev_fec_stats_supported(ae_dev) \
+	test_bit(HNAE3_DEV_SUPPORT_FEC_STATS_B, (ae_dev)->caps)
+
+#define hnae3_ae_dev_lane_num_supported(ae_dev) \
+	test_bit(HNAE3_DEV_SUPPORT_LANE_NUM_B, (ae_dev)->caps)
+
+#define hnae3_ae_dev_wol_supported(ae_dev) \
+	test_bit(HNAE3_DEV_SUPPORT_WOL_B, (ae_dev)->caps)
+
+#define hnae3_ae_dev_tm_flush_supported(hdev) \
+	test_bit(HNAE3_DEV_SUPPORT_TM_FLUSH_B, (hdev)->ae_dev->caps)
+
+#define hnae3_ae_dev_vf_fault_supported(ae_dev) \
+	test_bit(HNAE3_DEV_SUPPORT_VF_FAULT_B, (ae_dev)->caps)
+
+#define hnae3_ae_dev_gen_reg_dfx_supported(hdev) \
+	test_bit(HNAE3_DEV_SUPPORT_ERR_MOD_GEN_REG_B, (hdev)->ae_dev->caps)
+
+enum HNAE3_PF_CAP_BITS {
+	HNAE3_PF_SUPPORT_VLAN_FLTR_MDF_B = 0,
+};
+#define ring_ptr_move_fw(ring, p) \
+	((ring)->p = ((ring)->p + 1) % (ring)->desc_num)
+#define ring_ptr_move_bw(ring, p) \
+	((ring)->p = ((ring)->p - 1 + (ring)->desc_num) % (ring)->desc_num)
+
+struct hnae3_handle;
+
+struct hnae3_queue {
+	void __iomem *io_base;
+	void __iomem *mem_base;
+	struct hnae3_ae_algo *ae_algo;
+	struct hnae3_handle *handle;
+	int tqp_index;		/* index in a handle */
+	u32 buf_size;		/* size for hnae_desc->addr, preset by AE */
+	u16 tx_desc_num;	/* total number of tx desc */
+	u16 rx_desc_num;	/* total number of rx desc */
+};
+
+struct hns3_mac_stats {
+	u64 tx_pause_cnt;
+	u64 rx_pause_cnt;
+};
+
+/* hnae3 loop mode */
+enum hnae3_loop {
+	HNAE3_LOOP_EXTERNAL,
+	HNAE3_LOOP_APP,
+	HNAE3_LOOP_SERIAL_SERDES,
+	HNAE3_LOOP_PARALLEL_SERDES,
+	HNAE3_LOOP_PHY,
+	HNAE3_LOOP_NONE,
+};
+
+enum hnae3_client_type {
+	HNAE3_CLIENT_KNIC,
+	HNAE3_CLIENT_ROCE,
+};
+
+/* mac media type */
+enum hnae3_media_type {
+	HNAE3_MEDIA_TYPE_UNKNOWN,
+	HNAE3_MEDIA_TYPE_FIBER,
+	HNAE3_MEDIA_TYPE_COPPER,
+	HNAE3_MEDIA_TYPE_BACKPLANE,
+	HNAE3_MEDIA_TYPE_NONE,
+};
+
+/* must be consistent with definition in firmware */
+enum hnae3_module_type {
+	HNAE3_MODULE_TYPE_UNKNOWN	= 0x00,
+	HNAE3_MODULE_TYPE_FIBRE_LR	= 0x01,
+	HNAE3_MODULE_TYPE_FIBRE_SR	= 0x02,
+	HNAE3_MODULE_TYPE_AOC		= 0x03,
+	HNAE3_MODULE_TYPE_CR		= 0x04,
+	HNAE3_MODULE_TYPE_KR		= 0x05,
+	HNAE3_MODULE_TYPE_TP		= 0x06,
+};
+
+enum hnae3_fec_mode {
+	HNAE3_FEC_AUTO = 0,
+	HNAE3_FEC_BASER,
+	HNAE3_FEC_RS,
+	HNAE3_FEC_LLRS,
+	HNAE3_FEC_NONE,
+	HNAE3_FEC_USER_DEF,
+};
+
+enum hnae3_reset_notify_type {
+	HNAE3_UP_CLIENT,
+	HNAE3_DOWN_CLIENT,
+	HNAE3_INIT_CLIENT,
+	HNAE3_UNINIT_CLIENT,
+};
+
+enum hnae3_hw_error_type {
+	HNAE3_PPU_POISON_ERROR,
+	HNAE3_CMDQ_ECC_ERROR,
+	HNAE3_IMP_RD_POISON_ERROR,
+	HNAE3_ROCEE_AXI_RESP_ERROR,
+};
+
+enum hnae3_reset_type {
+	HNAE3_VF_RESET,
+	HNAE3_VF_FUNC_RESET,
+	HNAE3_VF_PF_FUNC_RESET,
+	HNAE3_VF_FULL_RESET,
+	HNAE3_FLR_RESET,
+	HNAE3_FUNC_RESET,
+	HNAE3_GLOBAL_RESET,
+	HNAE3_IMP_RESET,
+	HNAE3_NONE_RESET,
+	HNAE3_VF_EXP_RESET,
+	HNAE3_MAX_RESET,
+};
+
+enum hnae3_port_base_vlan_state {
+	HNAE3_PORT_BASE_VLAN_DISABLE,
+	HNAE3_PORT_BASE_VLAN_ENABLE,
+	HNAE3_PORT_BASE_VLAN_MODIFY,
+	HNAE3_PORT_BASE_VLAN_NOCHANGE,
+};
+
+enum hnae3_dbg_cmd {
+	HNAE3_DBG_CMD_TM_NODES,
+	HNAE3_DBG_CMD_TM_PRI,
+	HNAE3_DBG_CMD_TM_QSET,
+	HNAE3_DBG_CMD_TM_MAP,
+	HNAE3_DBG_CMD_TM_PG,
+	HNAE3_DBG_CMD_TM_PORT,
+	HNAE3_DBG_CMD_TC_SCH_INFO,
+	HNAE3_DBG_CMD_QOS_PAUSE_CFG,
+	HNAE3_DBG_CMD_QOS_PRI_MAP,
+	HNAE3_DBG_CMD_QOS_DSCP_MAP,
+	HNAE3_DBG_CMD_QOS_BUF_CFG,
+	HNAE3_DBG_CMD_DEV_INFO,
+	HNAE3_DBG_CMD_TX_BD,
+	HNAE3_DBG_CMD_RX_BD,
+	HNAE3_DBG_CMD_MAC_UC,
+	HNAE3_DBG_CMD_MAC_MC,
+	HNAE3_DBG_CMD_MNG_TBL,
+	HNAE3_DBG_CMD_LOOPBACK,
+	HNAE3_DBG_CMD_PTP_INFO,
+	HNAE3_DBG_CMD_INTERRUPT_INFO,
+	HNAE3_DBG_CMD_RESET_INFO,
+	HNAE3_DBG_CMD_IMP_INFO,
+	HNAE3_DBG_CMD_NCL_CONFIG,
+	HNAE3_DBG_CMD_REG_BIOS_COMMON,
+	HNAE3_DBG_CMD_REG_SSU,
+	HNAE3_DBG_CMD_REG_IGU_EGU,
+	HNAE3_DBG_CMD_REG_RPU,
+	HNAE3_DBG_CMD_REG_NCSI,
+	HNAE3_DBG_CMD_REG_RTC,
+	HNAE3_DBG_CMD_REG_PPP,
+	HNAE3_DBG_CMD_REG_RCB,
+	HNAE3_DBG_CMD_REG_TQP,
+	HNAE3_DBG_CMD_REG_MAC,
+	HNAE3_DBG_CMD_REG_DCB,
+	HNAE3_DBG_CMD_VLAN_CONFIG,
+	HNAE3_DBG_CMD_QUEUE_MAP,
+	HNAE3_DBG_CMD_RX_QUEUE_INFO,
+	HNAE3_DBG_CMD_TX_QUEUE_INFO,
+	HNAE3_DBG_CMD_FD_TCAM,
+	HNAE3_DBG_CMD_FD_COUNTER,
+	HNAE3_DBG_CMD_MAC_TNL_STATUS,
+	HNAE3_DBG_CMD_SERV_INFO,
+	HNAE3_DBG_CMD_UMV_INFO,
+	HNAE3_DBG_CMD_PAGE_POOL_INFO,
+	HNAE3_DBG_CMD_COAL_INFO,
+	HNAE3_DBG_CMD_UNKNOWN,
+};
+
+#define hnae3_seq_file_to_ae_dev(s)	(dev_get_drvdata((s)->private))
+#define hnae3_seq_file_to_handle(s)	\
+		(((struct hnae3_ae_dev *)hnae3_seq_file_to_ae_dev(s))->handle)
+
+enum hnae3_tc_map_mode {
+	HNAE3_TC_MAP_MODE_PRIO,
+	HNAE3_TC_MAP_MODE_DSCP,
+};
+
+struct hnae3_vector_info {
+	u8 __iomem *io_addr;
+	int vector;
+};
+
+#define HNAE3_RING_TYPE_B 0
+#define HNAE3_RING_TYPE_TX 0
+#define HNAE3_RING_TYPE_RX 1
+#define HNAE3_RING_GL_IDX_S 0
+#define HNAE3_RING_GL_IDX_M GENMASK(1, 0)
+#define HNAE3_RING_GL_RX 0
+#define HNAE3_RING_GL_TX 1
+
+#define HNAE3_FW_VERSION_BYTE3_SHIFT	24
+#define HNAE3_FW_VERSION_BYTE3_MASK	GENMASK(31, 24)
+#define HNAE3_FW_VERSION_BYTE2_SHIFT	16
+#define HNAE3_FW_VERSION_BYTE2_MASK	GENMASK(23, 16)
+#define HNAE3_FW_VERSION_BYTE1_SHIFT	8
+#define HNAE3_FW_VERSION_BYTE1_MASK	GENMASK(15, 8)
+#define HNAE3_FW_VERSION_BYTE0_SHIFT	0
+#define HNAE3_FW_VERSION_BYTE0_MASK	GENMASK(7, 0)
+
+#define HNAE3_SCC_VERSION_BYTE3_SHIFT	24
+#define HNAE3_SCC_VERSION_BYTE3_MASK	GENMASK(31, 24)
+#define HNAE3_SCC_VERSION_BYTE2_SHIFT	16
+#define HNAE3_SCC_VERSION_BYTE2_MASK	GENMASK(23, 16)
+#define HNAE3_SCC_VERSION_BYTE1_SHIFT	8
+#define HNAE3_SCC_VERSION_BYTE1_MASK	GENMASK(15, 8)
+#define HNAE3_SCC_VERSION_BYTE0_SHIFT	0
+#define HNAE3_SCC_VERSION_BYTE0_MASK	GENMASK(7, 0)
+
+struct hnae3_ring_chain_node {
+	struct hnae3_ring_chain_node *next;
+	u32 tqp_index;
+	u32 flag;
+	u32 int_gl_idx;
+};
+
+#define HNAE3_IS_TX_RING(node) \
+	(((node)->flag & 1 << HNAE3_RING_TYPE_B) == HNAE3_RING_TYPE_TX)
+
+/* device specification info from firmware */
+struct hnae3_dev_specs {
+	u32 mac_entry_num; /* number of mac-vlan table entry */
+	u32 mng_entry_num; /* number of manager table entry */
+	u32 max_tm_rate;
+	u16 rss_ind_tbl_size;
+	u16 rss_key_size;
+	u16 int_ql_max; /* max value of interrupt coalesce based on INT_QL */
+	u16 max_int_gl; /* max value of interrupt coalesce based on INT_GL */
+	u8 max_non_tso_bd_num; /* max BD number of one non-TSO packet */
+	u16 max_frm_size;
+	u16 max_qset_num;
+	u16 umv_size;
+	u16 mc_mac_size;
+	u32 mac_stats_num;
+	u8 tnl_num;
+	u8 hilink_version;
+};
+
+struct hnae3_client_ops {
+	int (*init_instance)(struct hnae3_handle *handle);
+	void (*uninit_instance)(struct hnae3_handle *handle, bool reset);
+	void (*link_status_change)(struct hnae3_handle *handle, bool state);
+	int (*reset_notify)(struct hnae3_handle *handle,
+			    enum hnae3_reset_notify_type type);
+	void (*process_hw_error)(struct hnae3_handle *handle,
+				 enum hnae3_hw_error_type);
+};
+
+#define HNAE3_CLIENT_NAME_LENGTH 16
+struct hnae3_client {
+	char name[HNAE3_CLIENT_NAME_LENGTH];
+	unsigned long state;
+	enum hnae3_client_type type;
+	const struct hnae3_client_ops *ops;
+	struct list_head node;
+};
+
+#define HNAE3_DEV_CAPS_MAX_NUM	96
+struct hnae3_ae_dev {
+	struct pci_dev *pdev;
+	const struct hnae3_ae_ops *ops;
+	struct list_head node;
+	u32 flag;
+	unsigned long hw_err_reset_req;
+	struct hnae3_dev_specs dev_specs;
+	u32 dev_version;
+	DECLARE_BITMAP(caps, HNAE3_DEV_CAPS_MAX_NUM);
+	void *priv;
+	struct hnae3_handle *handle;
+};
+
+typedef int (*read_func)(struct seq_file *s, void *data);
+
+/* This struct defines the operation on the handle.
+ *
+ * init_ae_dev(): (mandatory)
+ *   Get PF configure from pci_dev and initialize PF hardware
+ * uninit_ae_dev()
+ *   Disable PF device and release PF resource
+ * register_client
+ *   Register client to ae_dev
+ * unregister_client()
+ *   Unregister client from ae_dev
+ * start()
+ *   Enable the hardware
+ * stop()
+ *   Disable the hardware
+ * start_client()
+ *   Inform the hclge that client has been started
+ * stop_client()
+ *   Inform the hclge that client has been stopped
+ * get_status()
+ *   Get the carrier state of the back channel of the handle, 1 for ok, 0 for
+ *   non-ok
+ * get_ksettings_an_result()
+ *   Get negotiation status,speed and duplex
+ * get_media_type()
+ *   Get media type of MAC
+ * check_port_speed()
+ *   Check target speed whether is supported
+ * adjust_link()
+ *   Adjust link status
+ * set_loopback()
+ *   Set loopback
+ * set_promisc_mode
+ *   Set promisc mode
+ * request_update_promisc_mode
+ *   request to hclge(vf) to update promisc mode
+ * set_mtu()
+ *   set mtu
+ * get_pauseparam()
+ *   get tx and rx of pause frame use
+ * set_pauseparam()
+ *   set tx and rx of pause frame use
+ * set_autoneg()
+ *   set auto autonegotiation of pause frame use
+ * get_autoneg()
+ *   get auto autonegotiation of pause frame use
+ * restart_autoneg()
+ *   restart autonegotiation
+ * halt_autoneg()
+ *   halt/resume autonegotiation when autonegotiation on
+ * get_coalesce_usecs()
+ *   get usecs to delay a TX interrupt after a packet is sent
+ * get_rx_max_coalesced_frames()
+ *   get Maximum number of packets to be sent before a TX interrupt.
+ * set_coalesce_usecs()
+ *   set usecs to delay a TX interrupt after a packet is sent
+ * set_coalesce_frames()
+ *   set Maximum number of packets to be sent before a TX interrupt.
+ * get_mac_addr()
+ *   get mac address
+ * set_mac_addr()
+ *   set mac address
+ * add_uc_addr
+ *   Add unicast addr to mac table
+ * rm_uc_addr
+ *   Remove unicast addr from mac table
+ * set_mc_addr()
+ *   Set multicast address
+ * add_mc_addr
+ *   Add multicast address to mac table
+ * rm_mc_addr
+ *   Remove multicast address from mac table
+ * update_stats()
+ *   Update Old network device statistics
+ * get_mac_stats()
+ *   get mac pause statistics including tx_cnt and rx_cnt
+ * get_ethtool_stats()
+ *   Get ethtool network device statistics
+ * get_strings()
+ *   Get a set of strings that describe the requested objects
+ * get_sset_count()
+ *   Get number of strings that @get_strings will write
+ * update_led_status()
+ *   Update the led status
+ * set_led_id()
+ *   Set led id
+ * get_regs()
+ *   Get regs dump
+ * get_regs_len()
+ *   Get the len of the regs dump
+ * get_rss_key_size()
+ *   Get rss key size
+ * get_rss()
+ *   Get rss table
+ * set_rss()
+ *   Set rss table
+ * get_tc_size()
+ *   Get tc size of handle
+ * get_vector()
+ *   Get vector number and vector information
+ * put_vector()
+ *   Put the vector in hdev
+ * map_ring_to_vector()
+ *   Map rings to vector
+ * unmap_ring_from_vector()
+ *   Unmap rings from vector
+ * reset_queue()
+ *   Reset queue
+ * get_fw_version()
+ *   Get firmware version
+ * get_mdix_mode()
+ *   Get media typr of phy
+ * enable_vlan_filter()
+ *   Enable vlan filter
+ * set_vlan_filter()
+ *   Set vlan filter config of Ports
+ * set_vf_vlan_filter()
+ *   Set vlan filter config of vf
+ * enable_hw_strip_rxvtag()
+ *   Enable/disable hardware strip vlan tag of packets received
+ * set_gro_en
+ *   Enable/disable HW GRO
+ * add_arfs_entry
+ *   Check the 5-tuples of flow, and create flow director rule
+ * get_vf_config
+ *   Get the VF configuration setting by the host
+ * set_vf_link_state
+ *   Set VF link status
+ * set_vf_spoofchk
+ *   Enable/disable spoof check for specified vf
+ * set_vf_trust
+ *   Enable/disable trust for specified vf, if the vf being trusted, then
+ *   it can enable promisc mode
+ * set_vf_rate
+ *   Set the max tx rate of specified vf.
+ * set_vf_mac
+ *   Configure the default MAC for specified VF
+ * get_module_eeprom
+ *   Get the optical module eeprom info.
+ * add_cls_flower
+ *   Add clsflower rule
+ * del_cls_flower
+ *   Delete clsflower rule
+ * cls_flower_active
+ *   Check if any cls flower rule exist
+ * set_tx_hwts_info
+ *   Save information for 1588 tx packet
+ * get_rx_hwts
+ *   Get 1588 rx hwstamp
+ * get_ts_info
+ *   Get phc info
+ * clean_vf_config
+ *   Clean residual vf info after disable sriov
+ * get_wol
+ *   Get wake on lan info
+ * set_wol
+ *   Config wake on lan
+ * dbg_get_read_func
+ *   Return the read func for debugfs seq file
+ */
+struct hnae3_ae_ops {
+	int (*init_ae_dev)(struct hnae3_ae_dev *ae_dev);
+	void (*uninit_ae_dev)(struct hnae3_ae_dev *ae_dev);
+	void (*reset_prepare)(struct hnae3_ae_dev *ae_dev,
+			      enum hnae3_reset_type rst_type);
+	void (*reset_done)(struct hnae3_ae_dev *ae_dev);
+	int (*init_client_instance)(struct hnae3_client *client,
+				    struct hnae3_ae_dev *ae_dev);
+	void (*uninit_client_instance)(struct hnae3_client *client,
+				       struct hnae3_ae_dev *ae_dev);
+	int (*start)(struct hnae3_handle *handle);
+	void (*stop)(struct hnae3_handle *handle);
+	int (*client_start)(struct hnae3_handle *handle);
+	void (*client_stop)(struct hnae3_handle *handle);
+	int (*get_status)(struct hnae3_handle *handle);
+	void (*get_ksettings_an_result)(struct hnae3_handle *handle,
+					u8 *auto_neg, u32 *speed, u8 *duplex,
+					u32 *lane_num);
+
+	int (*cfg_mac_speed_dup_h)(struct hnae3_handle *handle, int speed,
+				   u8 duplex, u8 lane_num);
+
+	void (*get_media_type)(struct hnae3_handle *handle, u8 *media_type,
+			       u8 *module_type);
+	int (*check_port_speed)(struct hnae3_handle *handle, u32 speed);
+	void (*get_fec_stats)(struct hnae3_handle *handle,
+			      struct ethtool_fec_stats *fec_stats);
+	void (*get_fec)(struct hnae3_handle *handle, u8 *fec_ability,
+			u8 *fec_mode);
+	int (*set_fec)(struct hnae3_handle *handle, u32 fec_mode);
+	void (*adjust_link)(struct hnae3_handle *handle, int speed, int duplex);
+	int (*set_loopback)(struct hnae3_handle *handle,
+			    enum hnae3_loop loop_mode, bool en);
+
+	int (*set_promisc_mode)(struct hnae3_handle *handle, bool en_uc_pmc,
+				bool en_mc_pmc);
+	void (*request_update_promisc_mode)(struct hnae3_handle *handle);
+	int (*set_mtu)(struct hnae3_handle *handle, int new_mtu);
+
+	void (*get_pauseparam)(struct hnae3_handle *handle,
+			       u32 *auto_neg, u32 *rx_en, u32 *tx_en);
+	int (*set_pauseparam)(struct hnae3_handle *handle,
+			      u32 auto_neg, u32 rx_en, u32 tx_en);
+
+	int (*set_autoneg)(struct hnae3_handle *handle, bool enable);
+	int (*get_autoneg)(struct hnae3_handle *handle);
+	int (*restart_autoneg)(struct hnae3_handle *handle);
+	int (*halt_autoneg)(struct hnae3_handle *handle, bool halt);
+
+	void (*get_coalesce_usecs)(struct hnae3_handle *handle,
+				   u32 *tx_usecs, u32 *rx_usecs);
+	void (*get_rx_max_coalesced_frames)(struct hnae3_handle *handle,
+					    u32 *tx_frames, u32 *rx_frames);
+	int (*set_coalesce_usecs)(struct hnae3_handle *handle, u32 timeout);
+	int (*set_coalesce_frames)(struct hnae3_handle *handle,
+				   u32 coalesce_frames);
+	void (*get_coalesce_range)(struct hnae3_handle *handle,
+				   u32 *tx_frames_low, u32 *rx_frames_low,
+				   u32 *tx_frames_high, u32 *rx_frames_high,
+				   u32 *tx_usecs_low, u32 *rx_usecs_low,
+				   u32 *tx_usecs_high, u32 *rx_usecs_high);
+
+	void (*get_mac_addr)(struct hnae3_handle *handle, u8 *p);
+	int (*set_mac_addr)(struct hnae3_handle *handle, const void *p,
+			    bool is_first);
+	int (*do_ioctl)(struct hnae3_handle *handle,
+			struct ifreq *ifr, int cmd);
+	int (*add_uc_addr)(struct hnae3_handle *handle,
+			   const unsigned char *addr);
+	int (*rm_uc_addr)(struct hnae3_handle *handle,
+			  const unsigned char *addr);
+	int (*set_mc_addr)(struct hnae3_handle *handle, void *addr);
+	int (*add_mc_addr)(struct hnae3_handle *handle,
+			   const unsigned char *addr);
+	int (*rm_mc_addr)(struct hnae3_handle *handle,
+			  const unsigned char *addr);
+	void (*set_tso_stats)(struct hnae3_handle *handle, int enable);
+	void (*update_stats)(struct hnae3_handle *handle);
+	void (*get_stats)(struct hnae3_handle *handle, u64 *data);
+	void (*get_mac_stats)(struct hnae3_handle *handle,
+			      struct hns3_mac_stats *mac_stats);
+	void (*get_strings)(struct hnae3_handle *handle,
+			    u32 stringset, u8 **data);
+	int (*get_sset_count)(struct hnae3_handle *handle, int stringset);
+
+	void (*get_regs)(struct hnae3_handle *handle, u32 *version,
+			 void *data);
+	int (*get_regs_len)(struct hnae3_handle *handle);
+
+	u32 (*get_rss_key_size)(struct hnae3_handle *handle);
+	int (*get_rss)(struct hnae3_handle *handle, u32 *indir, u8 *key,
+		       u8 *hfunc);
+	int (*set_rss)(struct hnae3_handle *handle, const u32 *indir,
+		       const u8 *key, const u8 hfunc);
+	int (*set_rss_tuple)(struct hnae3_handle *handle,
+			     const struct ethtool_rxfh_fields *cmd);
+	int (*get_rss_tuple)(struct hnae3_handle *handle,
+			     struct ethtool_rxfh_fields *cmd);
+
+	int (*get_tc_size)(struct hnae3_handle *handle);
+
+	int (*get_vector)(struct hnae3_handle *handle, u16 vector_num,
+			  struct hnae3_vector_info *vector_info);
+	int (*put_vector)(struct hnae3_handle *handle, int vector_num);
+	int (*map_ring_to_vector)(struct hnae3_handle *handle,
+				  int vector_num,
+				  struct hnae3_ring_chain_node *vr_chain);
+	int (*unmap_ring_from_vector)(struct hnae3_handle *handle,
+				      int vector_num,
+				      struct hnae3_ring_chain_node *vr_chain);
+
+	int (*reset_queue)(struct hnae3_handle *handle);
+	u32 (*get_fw_version)(struct hnae3_handle *handle);
+	void (*get_mdix_mode)(struct hnae3_handle *handle,
+			      u8 *tp_mdix_ctrl, u8 *tp_mdix);
+
+	int (*enable_vlan_filter)(struct hnae3_handle *handle, bool enable);
+	int (*set_vlan_filter)(struct hnae3_handle *handle, __be16 proto,
+			       u16 vlan_id, bool is_kill);
+	int (*set_vf_vlan_filter)(struct hnae3_handle *handle, int vfid,
+				  u16 vlan, u8 qos, __be16 proto);
+	int (*enable_hw_strip_rxvtag)(struct hnae3_handle *handle, bool enable);
+	void (*reset_event)(struct pci_dev *pdev, struct hnae3_handle *handle);
+	enum hnae3_reset_type (*get_reset_level)(struct hnae3_ae_dev *ae_dev,
+						 unsigned long *addr);
+	void (*set_default_reset_request)(struct hnae3_ae_dev *ae_dev,
+					  enum hnae3_reset_type rst_type);
+	void (*get_channels)(struct hnae3_handle *handle,
+			     struct ethtool_channels *ch);
+	void (*get_tqps_and_rss_info)(struct hnae3_handle *h,
+				      u16 *alloc_tqps, u16 *max_rss_size);
+	int (*set_channels)(struct hnae3_handle *handle, u32 new_tqps_num,
+			    bool rxfh_configured);
+	void (*get_flowctrl_adv)(struct hnae3_handle *handle,
+				 u32 *flowctrl_adv);
+	int (*set_led_id)(struct hnae3_handle *handle,
+			  enum ethtool_phys_id_state status);
+	void (*get_link_mode)(struct hnae3_handle *handle,
+			      unsigned long *supported,
+			      unsigned long *advertising);
+	int (*add_fd_entry)(struct hnae3_handle *handle,
+			    struct ethtool_rxnfc *cmd);
+	int (*del_fd_entry)(struct hnae3_handle *handle,
+			    struct ethtool_rxnfc *cmd);
+	int (*get_fd_rule_cnt)(struct hnae3_handle *handle,
+			       struct ethtool_rxnfc *cmd);
+	int (*get_fd_rule_info)(struct hnae3_handle *handle,
+				struct ethtool_rxnfc *cmd);
+	int (*get_fd_all_rules)(struct hnae3_handle *handle,
+				struct ethtool_rxnfc *cmd, u32 *rule_locs);
+	void (*enable_fd)(struct hnae3_handle *handle, bool enable);
+	int (*add_arfs_entry)(struct hnae3_handle *handle, u16 queue_id,
+			      u16 flow_id, struct flow_keys *fkeys);
+	pci_ers_result_t (*handle_hw_ras_error)(struct hnae3_ae_dev *ae_dev);
+	bool (*get_hw_reset_stat)(struct hnae3_handle *handle);
+	bool (*ae_dev_resetting)(struct hnae3_handle *handle);
+	unsigned long (*ae_dev_reset_cnt)(struct hnae3_handle *handle);
+	int (*set_gro_en)(struct hnae3_handle *handle, bool enable);
+	u16 (*get_global_queue_id)(struct hnae3_handle *handle, u16 queue_id);
+	void (*set_timer_task)(struct hnae3_handle *handle, bool enable);
+	int (*mac_connect_phy)(struct hnae3_handle *handle);
+	void (*mac_disconnect_phy)(struct hnae3_handle *handle);
+	int (*get_vf_config)(struct hnae3_handle *handle, int vf,
+			     struct ifla_vf_info *ivf);
+	int (*set_vf_link_state)(struct hnae3_handle *handle, int vf,
+				 int link_state);
+	int (*set_vf_spoofchk)(struct hnae3_handle *handle, int vf,
+			       bool enable);
+	int (*set_vf_trust)(struct hnae3_handle *handle, int vf, bool enable);
+	int (*set_vf_rate)(struct hnae3_handle *handle, int vf,
+			   int min_tx_rate, int max_tx_rate, bool force);
+	int (*set_vf_mac)(struct hnae3_handle *handle, int vf, u8 *p);
+	int (*get_module_eeprom)(struct hnae3_handle *handle, u32 offset,
+				 u32 len, u8 *data);
+	bool (*get_cmdq_stat)(struct hnae3_handle *handle);
+	int (*add_cls_flower)(struct hnae3_handle *handle,
+			      struct flow_cls_offload *cls_flower, int tc);
+	int (*del_cls_flower)(struct hnae3_handle *handle,
+			      struct flow_cls_offload *cls_flower);
+	bool (*cls_flower_active)(struct hnae3_handle *handle);
+	int (*get_phy_link_ksettings)(struct hnae3_handle *handle,
+				      struct ethtool_link_ksettings *cmd);
+	int (*set_phy_link_ksettings)(struct hnae3_handle *handle,
+				      const struct ethtool_link_ksettings *cmd);
+	bool (*set_tx_hwts_info)(struct hnae3_handle *handle,
+				 struct sk_buff *skb);
+	void (*get_rx_hwts)(struct hnae3_handle *handle, struct sk_buff *skb,
+			    u32 nsec, u32 sec);
+	int (*get_ts_info)(struct hnae3_handle *handle,
+			   struct kernel_ethtool_ts_info *info);
+	int (*get_link_diagnosis_info)(struct hnae3_handle *handle,
+				       u32 *status_code);
+	void (*clean_vf_config)(struct hnae3_ae_dev *ae_dev, int num_vfs);
+	int (*get_dscp_prio)(struct hnae3_handle *handle, u8 dscp,
+			     u8 *tc_map_mode, u8 *priority);
+	void (*get_wol)(struct hnae3_handle *handle,
+			struct ethtool_wolinfo *wol);
+	int (*set_wol)(struct hnae3_handle *handle,
+		       struct ethtool_wolinfo *wol);
+	int (*dbg_get_read_func)(struct hnae3_handle *handle,
+				 enum hnae3_dbg_cmd cmd,
+				 read_func *func);
+};
+
+struct hnae3_dcb_ops {
+	/* IEEE 802.1Qaz std */
+	int (*ieee_getets)(struct hnae3_handle *, struct ieee_ets *);
+	int (*ieee_setets)(struct hnae3_handle *, struct ieee_ets *);
+	int (*ieee_getpfc)(struct hnae3_handle *, struct ieee_pfc *);
+	int (*ieee_setpfc)(struct hnae3_handle *, struct ieee_pfc *);
+	int (*ieee_setapp)(struct hnae3_handle *h, struct dcb_app *app);
+	int (*ieee_delapp)(struct hnae3_handle *h, struct dcb_app *app);
+
+	/* DCBX configuration */
+	u8   (*getdcbx)(struct hnae3_handle *);
+	u8   (*setdcbx)(struct hnae3_handle *, u8);
+
+	int (*setup_tc)(struct hnae3_handle *handle,
+			struct tc_mqprio_qopt_offload *mqprio_qopt);
+};
+
+struct hnae3_ae_algo {
+	const struct hnae3_ae_ops *ops;
+	struct list_head node;
+	const struct pci_device_id *pdev_id_table;
+};
+
+#define HNAE3_INT_NAME_LEN        32
+#define HNAE3_ITR_COUNTDOWN_START 100
+
+#define HNAE3_MAX_TC		8
+#define HNAE3_MAX_USER_PRIO	8
+struct hnae3_tc_info {
+	u8 prio_tc[HNAE3_MAX_USER_PRIO]; /* TC indexed by prio */
+	u16 tqp_count[HNAE3_MAX_TC];
+	u16 tqp_offset[HNAE3_MAX_TC];
+	u8 max_tc; /* Total number of TCs */
+	u8 num_tc; /* Total number of enabled TCs */
+	bool mqprio_active;
+	bool mqprio_destroy;
+	bool dcb_ets_active;
+};
+
+#define HNAE3_MAX_DSCP			64
+#define HNAE3_PRIO_ID_INVALID		0xff
+struct hnae3_knic_private_info {
+	struct net_device *netdev; /* Set by KNIC client when init instance */
+	u16 rss_size;		   /* Allocated RSS queues */
+	u16 req_rss_size;
+	u16 rx_buf_len;
+	u16 num_tx_desc;
+	u16 num_rx_desc;
+	u32 tx_spare_buf_size;
+
+	struct hnae3_tc_info tc_info;
+	u8 tc_map_mode;
+	u8 dscp_app_cnt;
+	u8 dscp_prio[HNAE3_MAX_DSCP];
+
+	u16 num_tqps;		  /* total number of TQPs in this handle */
+	struct hnae3_queue **tqp;  /* array base of all TQPs in this instance */
+	const struct hnae3_dcb_ops *dcb_ops;
+
+	u16 int_rl_setting;
+	void __iomem *io_base;
+};
+
+struct hnae3_roce_private_info {
+	struct net_device *netdev;
+	void __iomem *roce_io_base;
+	void __iomem *roce_mem_base;
+	int base_vector;
+	int num_vectors;
+
+	/* The below attributes defined for RoCE client, hnae3 gives
+	 * initial values to them, and RoCE client can modify and use
+	 * them.
+	 */
+	unsigned long reset_state;
+	unsigned long instance_state;
+	unsigned long state;
+};
+
+#define HNAE3_SUPPORT_APP_LOOPBACK    BIT(0)
+#define HNAE3_SUPPORT_PHY_LOOPBACK    BIT(1)
+#define HNAE3_SUPPORT_SERDES_SERIAL_LOOPBACK	BIT(2)
+#define HNAE3_SUPPORT_VF	      BIT(3)
+#define HNAE3_SUPPORT_SERDES_PARALLEL_LOOPBACK	BIT(4)
+#define HNAE3_SUPPORT_EXTERNAL_LOOPBACK	BIT(5)
+
+#define HNAE3_USER_UPE		BIT(0)	/* unicast promisc enabled by user */
+#define HNAE3_USER_MPE		BIT(1)	/* mulitcast promisc enabled by user */
+#define HNAE3_BPE		BIT(2)	/* broadcast promisc enable */
+#define HNAE3_OVERFLOW_UPE	BIT(3)	/* unicast mac vlan overflow */
+#define HNAE3_OVERFLOW_MPE	BIT(4)	/* multicast mac vlan overflow */
+#define HNAE3_UPE		(HNAE3_USER_UPE | HNAE3_OVERFLOW_UPE)
+#define HNAE3_MPE		(HNAE3_USER_MPE | HNAE3_OVERFLOW_MPE)
+
+enum hnae3_pflag {
+	HNAE3_PFLAG_LIMIT_PROMISC,
+	HNAE3_PFLAG_MAX
+};
+
+struct hnae3_handle {
+	struct hnae3_client *client;
+	struct pci_dev *pdev;
+	void *priv;
+	struct hnae3_ae_algo *ae_algo;  /* the class who provides this handle */
+	u64 flags; /* Indicate the capabilities for this handle */
+
+	union {
+		struct net_device *netdev; /* first member */
+		struct hnae3_knic_private_info kinfo;
+		struct hnae3_roce_private_info rinfo;
+	};
+
+	nodemask_t numa_node_mask; /* for multi-chip support */
+
+	enum hnae3_port_base_vlan_state port_base_vlan_state;
+
+	u8 netdev_flags;
+	struct dentry *hnae3_dbgfs;
+
+	/* Network interface message level enabled bits */
+	u32 msg_enable;
+
+	unsigned long supported_pflags;
+	unsigned long priv_flags;
+};
+
+#define hnae3_set_field(origin, mask, shift, val) \
+	do { \
+		(origin) &= (~(mask)); \
+		(origin) |= ((val) << (shift)) & (mask); \
+	} while (0)
+#define hnae3_get_field(origin, mask, shift) (((origin) & (mask)) >> (shift))
+
+#define hnae3_set_bit(origin, shift, val) \
+	hnae3_set_field(origin, 0x1 << (shift), shift, val)
+#define hnae3_get_bit(origin, shift) \
+	hnae3_get_field(origin, 0x1 << (shift), shift)
+
+#define HNAE3_FORMAT_MAC_ADDR_LEN	18
+#define HNAE3_FORMAT_MAC_ADDR_OFFSET_0	0
+#define HNAE3_FORMAT_MAC_ADDR_OFFSET_4	4
+#define HNAE3_FORMAT_MAC_ADDR_OFFSET_5	5
+
+static inline void hnae3_format_mac_addr(char *format_mac_addr,
+					 const u8 *mac_addr)
+{
+	snprintf(format_mac_addr, HNAE3_FORMAT_MAC_ADDR_LEN, "%02x:**:**:**:%02x:%02x",
+		 mac_addr[HNAE3_FORMAT_MAC_ADDR_OFFSET_0],
+		 mac_addr[HNAE3_FORMAT_MAC_ADDR_OFFSET_4],
+		 mac_addr[HNAE3_FORMAT_MAC_ADDR_OFFSET_5]);
+}
+
+int hnae3_register_ae_dev(struct hnae3_ae_dev *ae_dev);
+void hnae3_unregister_ae_dev(struct hnae3_ae_dev *ae_dev);
+
+void hnae3_unregister_ae_algo_prepare(struct hnae3_ae_algo *ae_algo);
+void hnae3_unregister_ae_algo(struct hnae3_ae_algo *ae_algo);
+void hnae3_register_ae_algo(struct hnae3_ae_algo *ae_algo);
+
+void hnae3_unregister_client(struct hnae3_client *client);
+int hnae3_register_client(struct hnae3_client *client);
+
+void hnae3_set_client_init_flag(struct hnae3_client *client,
+				struct hnae3_ae_dev *ae_dev,
+				unsigned int inited);
+void hnae3_acquire_unload_lock(void);
+void hnae3_release_unload_lock(void);
+#endif
diff --git a/drivers/net/ethernet/hisilicon/hns3/hns3_common/hclge_comm_cmd.c b/drivers/net/ethernet/hisilicon/hns3/hns3_common/hclge_comm_cmd.c
new file mode 100644
index 000000000000..37396ca4ecfc
--- /dev/null
+++ b/drivers/net/ethernet/hisilicon/hns3/hns3_common/hclge_comm_cmd.c
@@ -0,0 +1,685 @@
+// SPDX-License-Identifier: GPL-2.0+
+// Copyright (c) 2021-2021 Hisilicon Limited.
+
+#include "hnae3.h"
+#include "hclge_comm_cmd.h"
+
+static void hclge_comm_cmd_config_regs(struct hclge_comm_hw *hw,
+				       struct hclge_comm_cmq_ring *ring)
+{
+	dma_addr_t dma = ring->desc_dma_addr;
+	u32 reg_val;
+
+	if (ring->ring_type == HCLGE_COMM_TYPE_CSQ) {
+		hclge_comm_write_dev(hw, HCLGE_COMM_NIC_CSQ_BASEADDR_L_REG,
+				     lower_32_bits(dma));
+		hclge_comm_write_dev(hw, HCLGE_COMM_NIC_CSQ_BASEADDR_H_REG,
+				     upper_32_bits(dma));
+		reg_val = hclge_comm_read_dev(hw, HCLGE_COMM_NIC_CSQ_DEPTH_REG);
+		reg_val &= HCLGE_COMM_NIC_SW_RST_RDY;
+		reg_val |= ring->desc_num >> HCLGE_COMM_NIC_CMQ_DESC_NUM_S;
+		hclge_comm_write_dev(hw, HCLGE_COMM_NIC_CSQ_DEPTH_REG, reg_val);
+		hclge_comm_write_dev(hw, HCLGE_COMM_NIC_CSQ_HEAD_REG, 0);
+		hclge_comm_write_dev(hw, HCLGE_COMM_NIC_CSQ_TAIL_REG, 0);
+	} else {
+		hclge_comm_write_dev(hw, HCLGE_COMM_NIC_CRQ_BASEADDR_L_REG,
+				     lower_32_bits(dma));
+		hclge_comm_write_dev(hw, HCLGE_COMM_NIC_CRQ_BASEADDR_H_REG,
+				     upper_32_bits(dma));
+		reg_val = ring->desc_num >> HCLGE_COMM_NIC_CMQ_DESC_NUM_S;
+		hclge_comm_write_dev(hw, HCLGE_COMM_NIC_CRQ_DEPTH_REG, reg_val);
+		hclge_comm_write_dev(hw, HCLGE_COMM_NIC_CRQ_HEAD_REG, 0);
+		hclge_comm_write_dev(hw, HCLGE_COMM_NIC_CRQ_TAIL_REG, 0);
+	}
+}
+
+void hclge_comm_cmd_init_regs(struct hclge_comm_hw *hw)
+{
+	hclge_comm_cmd_config_regs(hw, &hw->cmq.csq);
+	hclge_comm_cmd_config_regs(hw, &hw->cmq.crq);
+}
+
+void hclge_comm_cmd_reuse_desc(struct hclge_desc *desc, bool is_read)
+{
+	desc->flag = cpu_to_le16(HCLGE_COMM_CMD_FLAG_NO_INTR |
+				 HCLGE_COMM_CMD_FLAG_IN);
+	if (is_read)
+		desc->flag |= cpu_to_le16(HCLGE_COMM_CMD_FLAG_WR);
+	else
+		desc->flag &= cpu_to_le16(~HCLGE_COMM_CMD_FLAG_WR);
+}
+EXPORT_SYMBOL_GPL(hclge_comm_cmd_reuse_desc);
+
+static void hclge_comm_set_default_capability(struct hnae3_ae_dev *ae_dev,
+					      bool is_pf)
+{
+	set_bit(HNAE3_DEV_SUPPORT_GRO_B, ae_dev->caps);
+	if (is_pf) {
+		set_bit(HNAE3_DEV_SUPPORT_FD_B, ae_dev->caps);
+		set_bit(HNAE3_DEV_SUPPORT_FEC_B, ae_dev->caps);
+		set_bit(HNAE3_DEV_SUPPORT_PAUSE_B, ae_dev->caps);
+	}
+}
+
+void hclge_comm_cmd_setup_basic_desc(struct hclge_desc *desc,
+				     enum hclge_opcode_type opcode,
+				     bool is_read)
+{
+	memset((void *)desc, 0, sizeof(struct hclge_desc));
+	desc->opcode = cpu_to_le16(opcode);
+	desc->flag = cpu_to_le16(HCLGE_COMM_CMD_FLAG_NO_INTR |
+				 HCLGE_COMM_CMD_FLAG_IN);
+
+	if (is_read)
+		desc->flag |= cpu_to_le16(HCLGE_COMM_CMD_FLAG_WR);
+}
+EXPORT_SYMBOL_GPL(hclge_comm_cmd_setup_basic_desc);
+
+int hclge_comm_firmware_compat_config(struct hnae3_ae_dev *ae_dev,
+				      struct hclge_comm_hw *hw, bool en)
+{
+	struct hclge_comm_firmware_compat_cmd *req;
+	struct hclge_desc desc;
+	u32 compat = 0;
+
+	hclge_comm_cmd_setup_basic_desc(&desc, HCLGE_OPC_IMP_COMPAT_CFG, false);
+
+	if (en) {
+		req = (struct hclge_comm_firmware_compat_cmd *)desc.data;
+
+		hnae3_set_bit(compat, HCLGE_COMM_LINK_EVENT_REPORT_EN_B, 1);
+		hnae3_set_bit(compat, HCLGE_COMM_NCSI_ERROR_REPORT_EN_B, 1);
+		if (hclge_comm_dev_phy_imp_supported(ae_dev))
+			hnae3_set_bit(compat, HCLGE_COMM_PHY_IMP_EN_B, 1);
+		hnae3_set_bit(compat, HCLGE_COMM_MAC_STATS_EXT_EN_B, 1);
+		hnae3_set_bit(compat, HCLGE_COMM_SYNC_RX_RING_HEAD_EN_B, 1);
+		hnae3_set_bit(compat, HCLGE_COMM_LLRS_FEC_EN_B, 1);
+
+		req->compat = cpu_to_le32(compat);
+	}
+
+	return hclge_comm_cmd_send(hw, &desc, 1);
+}
+
+void hclge_comm_free_cmd_desc(struct hclge_comm_cmq_ring *ring)
+{
+	int size  = ring->desc_num * sizeof(struct hclge_desc);
+
+	if (!ring->desc)
+		return;
+
+	dma_free_coherent(&ring->pdev->dev, size,
+			  ring->desc, ring->desc_dma_addr);
+	ring->desc = NULL;
+}
+
+static int hclge_comm_alloc_cmd_desc(struct hclge_comm_cmq_ring *ring)
+{
+	int size  = ring->desc_num * sizeof(struct hclge_desc);
+
+	ring->desc = dma_alloc_coherent(&ring->pdev->dev,
+					size, &ring->desc_dma_addr, GFP_KERNEL);
+	if (!ring->desc)
+		return -ENOMEM;
+
+	return 0;
+}
+
+static __le32 hclge_comm_build_api_caps(void)
+{
+	u32 api_caps = 0;
+
+	hnae3_set_bit(api_caps, HCLGE_COMM_API_CAP_FLEX_RSS_TBL_B, 1);
+
+	return cpu_to_le32(api_caps);
+}
+
+static const struct hclge_comm_caps_bit_map hclge_pf_cmd_caps[] = {
+	{HCLGE_COMM_CAP_UDP_GSO_B, HNAE3_DEV_SUPPORT_UDP_GSO_B},
+	{HCLGE_COMM_CAP_PTP_B, HNAE3_DEV_SUPPORT_PTP_B},
+	{HCLGE_COMM_CAP_INT_QL_B, HNAE3_DEV_SUPPORT_INT_QL_B},
+	{HCLGE_COMM_CAP_TQP_TXRX_INDEP_B, HNAE3_DEV_SUPPORT_TQP_TXRX_INDEP_B},
+	{HCLGE_COMM_CAP_HW_TX_CSUM_B, HNAE3_DEV_SUPPORT_HW_TX_CSUM_B},
+	{HCLGE_COMM_CAP_UDP_TUNNEL_CSUM_B, HNAE3_DEV_SUPPORT_UDP_TUNNEL_CSUM_B},
+	{HCLGE_COMM_CAP_FD_FORWARD_TC_B, HNAE3_DEV_SUPPORT_FD_FORWARD_TC_B},
+	{HCLGE_COMM_CAP_FEC_B, HNAE3_DEV_SUPPORT_FEC_B},
+	{HCLGE_COMM_CAP_PAUSE_B, HNAE3_DEV_SUPPORT_PAUSE_B},
+	{HCLGE_COMM_CAP_PHY_IMP_B, HNAE3_DEV_SUPPORT_PHY_IMP_B},
+	{HCLGE_COMM_CAP_QB_B, HNAE3_DEV_SUPPORT_QB_B},
+	{HCLGE_COMM_CAP_TX_PUSH_B, HNAE3_DEV_SUPPORT_TX_PUSH_B},
+	{HCLGE_COMM_CAP_RAS_IMP_B, HNAE3_DEV_SUPPORT_RAS_IMP_B},
+	{HCLGE_COMM_CAP_RXD_ADV_LAYOUT_B, HNAE3_DEV_SUPPORT_RXD_ADV_LAYOUT_B},
+	{HCLGE_COMM_CAP_PORT_VLAN_BYPASS_B,
+	 HNAE3_DEV_SUPPORT_PORT_VLAN_BYPASS_B},
+	{HCLGE_COMM_CAP_PORT_VLAN_BYPASS_B, HNAE3_DEV_SUPPORT_VLAN_FLTR_MDF_B},
+	{HCLGE_COMM_CAP_CQ_B, HNAE3_DEV_SUPPORT_CQ_B},
+	{HCLGE_COMM_CAP_GRO_B, HNAE3_DEV_SUPPORT_GRO_B},
+	{HCLGE_COMM_CAP_FD_B, HNAE3_DEV_SUPPORT_FD_B},
+	{HCLGE_COMM_CAP_FEC_STATS_B, HNAE3_DEV_SUPPORT_FEC_STATS_B},
+	{HCLGE_COMM_CAP_LANE_NUM_B, HNAE3_DEV_SUPPORT_LANE_NUM_B},
+	{HCLGE_COMM_CAP_WOL_B, HNAE3_DEV_SUPPORT_WOL_B},
+	{HCLGE_COMM_CAP_TM_FLUSH_B, HNAE3_DEV_SUPPORT_TM_FLUSH_B},
+	{HCLGE_COMM_CAP_VF_FAULT_B, HNAE3_DEV_SUPPORT_VF_FAULT_B},
+	{HCLGE_COMM_CAP_ERR_MOD_GEN_REG_B, HNAE3_DEV_SUPPORT_ERR_MOD_GEN_REG_B},
+};
+
+static const struct hclge_comm_caps_bit_map hclge_vf_cmd_caps[] = {
+	{HCLGE_COMM_CAP_UDP_GSO_B, HNAE3_DEV_SUPPORT_UDP_GSO_B},
+	{HCLGE_COMM_CAP_INT_QL_B, HNAE3_DEV_SUPPORT_INT_QL_B},
+	{HCLGE_COMM_CAP_TQP_TXRX_INDEP_B, HNAE3_DEV_SUPPORT_TQP_TXRX_INDEP_B},
+	{HCLGE_COMM_CAP_HW_TX_CSUM_B, HNAE3_DEV_SUPPORT_HW_TX_CSUM_B},
+	{HCLGE_COMM_CAP_UDP_TUNNEL_CSUM_B, HNAE3_DEV_SUPPORT_UDP_TUNNEL_CSUM_B},
+	{HCLGE_COMM_CAP_QB_B, HNAE3_DEV_SUPPORT_QB_B},
+	{HCLGE_COMM_CAP_TX_PUSH_B, HNAE3_DEV_SUPPORT_TX_PUSH_B},
+	{HCLGE_COMM_CAP_RXD_ADV_LAYOUT_B, HNAE3_DEV_SUPPORT_RXD_ADV_LAYOUT_B},
+	{HCLGE_COMM_CAP_CQ_B, HNAE3_DEV_SUPPORT_CQ_B},
+	{HCLGE_COMM_CAP_GRO_B, HNAE3_DEV_SUPPORT_GRO_B},
+};
+
+static void
+hclge_comm_capability_to_bitmap(unsigned long *bitmap, __le32 *caps)
+{
+	const unsigned int words = HCLGE_COMM_QUERY_CAP_LENGTH;
+	u32 val[HCLGE_COMM_QUERY_CAP_LENGTH];
+	unsigned int i;
+
+	for (i = 0; i < words; i++)
+		val[i] = __le32_to_cpu(caps[i]);
+
+	bitmap_from_arr32(bitmap, val,
+			  HCLGE_COMM_QUERY_CAP_LENGTH * BITS_PER_TYPE(u32));
+}
+
+static void
+hclge_comm_parse_capability(struct hnae3_ae_dev *ae_dev, bool is_pf,
+			    struct hclge_comm_query_version_cmd *cmd)
+{
+	const struct hclge_comm_caps_bit_map *caps_map =
+				is_pf ? hclge_pf_cmd_caps : hclge_vf_cmd_caps;
+	u32 size = is_pf ? ARRAY_SIZE(hclge_pf_cmd_caps) :
+				ARRAY_SIZE(hclge_vf_cmd_caps);
+	DECLARE_BITMAP(caps, HCLGE_COMM_QUERY_CAP_LENGTH * BITS_PER_TYPE(u32));
+	u32 i;
+
+	hclge_comm_capability_to_bitmap(caps, cmd->caps);
+	for (i = 0; i < size; i++)
+		if (test_bit(caps_map[i].imp_bit, caps))
+			set_bit(caps_map[i].local_bit, ae_dev->caps);
+}
+
+int hclge_comm_alloc_cmd_queue(struct hclge_comm_hw *hw, int ring_type)
+{
+	struct hclge_comm_cmq_ring *ring =
+		(ring_type == HCLGE_COMM_TYPE_CSQ) ? &hw->cmq.csq :
+						     &hw->cmq.crq;
+	int ret;
+
+	ring->ring_type = ring_type;
+
+	ret = hclge_comm_alloc_cmd_desc(ring);
+	if (ret)
+		dev_err(&ring->pdev->dev, "descriptor %s alloc error %d\n",
+			(ring_type == HCLGE_COMM_TYPE_CSQ) ? "CSQ" : "CRQ",
+			ret);
+
+	return ret;
+}
+
+int hclge_comm_cmd_query_version_and_capability(struct hnae3_ae_dev *ae_dev,
+						struct hclge_comm_hw *hw,
+						u32 *fw_version, bool is_pf)
+{
+	struct hclge_comm_query_version_cmd *resp;
+	struct hclge_desc desc;
+	int ret;
+
+	hclge_comm_cmd_setup_basic_desc(&desc, HCLGE_OPC_QUERY_FW_VER, 1);
+	resp = (struct hclge_comm_query_version_cmd *)desc.data;
+	resp->api_caps = hclge_comm_build_api_caps();
+
+	ret = hclge_comm_cmd_send(hw, &desc, 1);
+	if (ret)
+		return ret;
+
+	*fw_version = le32_to_cpu(resp->firmware);
+
+	ae_dev->dev_version = le32_to_cpu(resp->hardware) <<
+					 HNAE3_PCI_REVISION_BIT_SIZE;
+	ae_dev->dev_version |= ae_dev->pdev->revision;
+
+	if (ae_dev->dev_version == HNAE3_DEVICE_VERSION_V2) {
+		hclge_comm_set_default_capability(ae_dev, is_pf);
+		return 0;
+	}
+
+	hclge_comm_parse_capability(ae_dev, is_pf, resp);
+
+	return ret;
+}
+
+static const u16 spec_opcode[] = { HCLGE_OPC_STATS_64_BIT,
+				   HCLGE_OPC_STATS_32_BIT,
+				   HCLGE_OPC_STATS_MAC,
+				   HCLGE_OPC_STATS_MAC_ALL,
+				   HCLGE_OPC_QUERY_32_BIT_REG,
+				   HCLGE_OPC_QUERY_64_BIT_REG,
+				   HCLGE_QUERY_CLEAR_MPF_RAS_INT,
+				   HCLGE_QUERY_CLEAR_PF_RAS_INT,
+				   HCLGE_QUERY_CLEAR_ALL_MPF_MSIX_INT,
+				   HCLGE_QUERY_CLEAR_ALL_PF_MSIX_INT,
+				   HCLGE_QUERY_ALL_ERR_INFO };
+
+static bool hclge_comm_is_special_opcode(u16 opcode)
+{
+	/* these commands have several descriptors,
+	 * and use the first one to save opcode and return value
+	 */
+	u32 i;
+
+	for (i = 0; i < ARRAY_SIZE(spec_opcode); i++)
+		if (spec_opcode[i] == opcode)
+			return true;
+
+	return false;
+}
+
+static int hclge_comm_ring_space(struct hclge_comm_cmq_ring *ring)
+{
+	int ntc = ring->next_to_clean;
+	int ntu = ring->next_to_use;
+	int used = (ntu - ntc + ring->desc_num) % ring->desc_num;
+
+	return ring->desc_num - used - 1;
+}
+
+static void hclge_comm_cmd_copy_desc(struct hclge_comm_hw *hw,
+				     struct hclge_desc *desc, int num)
+{
+	struct hclge_desc *desc_to_use;
+	int handle = 0;
+
+	while (handle < num) {
+		desc_to_use = &hw->cmq.csq.desc[hw->cmq.csq.next_to_use];
+		*desc_to_use = desc[handle];
+		(hw->cmq.csq.next_to_use)++;
+		if (hw->cmq.csq.next_to_use >= hw->cmq.csq.desc_num)
+			hw->cmq.csq.next_to_use = 0;
+		handle++;
+	}
+}
+
+static int hclge_comm_is_valid_csq_clean_head(struct hclge_comm_cmq_ring *ring,
+					      int head)
+{
+	int ntc = ring->next_to_clean;
+	int ntu = ring->next_to_use;
+
+	if (ntu > ntc)
+		return head >= ntc && head <= ntu;
+
+	return head >= ntc || head <= ntu;
+}
+
+static int hclge_comm_cmd_csq_clean(struct hclge_comm_hw *hw)
+{
+	struct hclge_comm_cmq_ring *csq = &hw->cmq.csq;
+	int clean;
+	u32 head;
+
+	head = hclge_comm_read_dev(hw, HCLGE_COMM_NIC_CSQ_HEAD_REG);
+	rmb(); /* Make sure head is ready before touch any data */
+
+	if (!hclge_comm_is_valid_csq_clean_head(csq, head)) {
+		dev_warn(&hw->cmq.csq.pdev->dev, "wrong cmd head (%u, %d-%d)\n",
+			 head, csq->next_to_use, csq->next_to_clean);
+		dev_warn(&hw->cmq.csq.pdev->dev,
+			 "Disabling any further commands to IMP firmware\n");
+		set_bit(HCLGE_COMM_STATE_CMD_DISABLE, &hw->comm_state);
+		dev_warn(&hw->cmq.csq.pdev->dev,
+			 "IMP firmware watchdog reset soon expected!\n");
+		return -EIO;
+	}
+
+	clean = (head - csq->next_to_clean + csq->desc_num) % csq->desc_num;
+	csq->next_to_clean = head;
+	return clean;
+}
+
+static int hclge_comm_cmd_csq_done(struct hclge_comm_hw *hw)
+{
+	u32 head = hclge_comm_read_dev(hw, HCLGE_COMM_NIC_CSQ_HEAD_REG);
+	return head == (u32)hw->cmq.csq.next_to_use;
+}
+
+static u32 hclge_get_cmdq_tx_timeout(u16 opcode, u32 tx_timeout)
+{
+	static const struct hclge_cmdq_tx_timeout_map cmdq_tx_timeout_map[] = {
+		{HCLGE_OPC_CFG_RST_TRIGGER, HCLGE_COMM_CMDQ_CFG_RST_TIMEOUT},
+	};
+	u32 i;
+
+	for (i = 0; i < ARRAY_SIZE(cmdq_tx_timeout_map); i++)
+		if (cmdq_tx_timeout_map[i].opcode == opcode)
+			return cmdq_tx_timeout_map[i].tx_timeout;
+
+	return tx_timeout;
+}
+
+static void hclge_comm_wait_for_resp(struct hclge_comm_hw *hw, u16 opcode,
+				     bool *is_completed)
+{
+	u32 cmdq_tx_timeout = hclge_get_cmdq_tx_timeout(opcode,
+							hw->cmq.tx_timeout);
+	u32 timeout = 0;
+
+	do {
+		if (hclge_comm_cmd_csq_done(hw)) {
+			*is_completed = true;
+			break;
+		}
+		udelay(1);
+		timeout++;
+	} while (timeout < cmdq_tx_timeout);
+}
+
+static int hclge_comm_cmd_convert_err_code(u16 desc_ret)
+{
+	struct hclge_comm_errcode hclge_comm_cmd_errcode[] = {
+		{ HCLGE_COMM_CMD_EXEC_SUCCESS, 0 },
+		{ HCLGE_COMM_CMD_NO_AUTH, -EPERM },
+		{ HCLGE_COMM_CMD_NOT_SUPPORTED, -EOPNOTSUPP },
+		{ HCLGE_COMM_CMD_QUEUE_FULL, -EXFULL },
+		{ HCLGE_COMM_CMD_NEXT_ERR, -ENOSR },
+		{ HCLGE_COMM_CMD_UNEXE_ERR, -ENOTBLK },
+		{ HCLGE_COMM_CMD_PARA_ERR, -EINVAL },
+		{ HCLGE_COMM_CMD_RESULT_ERR, -ERANGE },
+		{ HCLGE_COMM_CMD_TIMEOUT, -ETIME },
+		{ HCLGE_COMM_CMD_HILINK_ERR, -ENOLINK },
+		{ HCLGE_COMM_CMD_QUEUE_ILLEGAL, -ENXIO },
+		{ HCLGE_COMM_CMD_INVALID, -EBADR },
+	};
+	u32 errcode_count = ARRAY_SIZE(hclge_comm_cmd_errcode);
+	u32 i;
+
+	for (i = 0; i < errcode_count; i++)
+		if (hclge_comm_cmd_errcode[i].imp_errcode == desc_ret)
+			return hclge_comm_cmd_errcode[i].common_errno;
+
+	return -EIO;
+}
+
+static int hclge_comm_cmd_check_retval(struct hclge_comm_hw *hw,
+				       struct hclge_desc *desc, int num,
+				       int ntc)
+{
+	u16 opcode, desc_ret;
+	int handle;
+
+	opcode = le16_to_cpu(desc[0].opcode);
+	for (handle = 0; handle < num; handle++) {
+		desc[handle] = hw->cmq.csq.desc[ntc];
+		ntc++;
+		if (ntc >= hw->cmq.csq.desc_num)
+			ntc = 0;
+	}
+	if (likely(!hclge_comm_is_special_opcode(opcode)))
+		desc_ret = le16_to_cpu(desc[num - 1].retval);
+	else
+		desc_ret = le16_to_cpu(desc[0].retval);
+
+	hw->cmq.last_status = desc_ret;
+
+	return hclge_comm_cmd_convert_err_code(desc_ret);
+}
+
+static int hclge_comm_cmd_check_result(struct hclge_comm_hw *hw,
+				       struct hclge_desc *desc,
+				       int num, int ntc)
+{
+	bool is_completed = false;
+	int handle, ret;
+
+	/* If the command is sync, wait for the firmware to write back,
+	 * if multi descriptors to be sent, use the first one to check
+	 */
+	if (HCLGE_COMM_SEND_SYNC(le16_to_cpu(desc->flag)))
+		hclge_comm_wait_for_resp(hw, le16_to_cpu(desc->opcode),
+					 &is_completed);
+
+	if (!is_completed)
+		ret = -EBADE;
+	else
+		ret = hclge_comm_cmd_check_retval(hw, desc, num, ntc);
+
+	/* Clean the command send queue */
+	handle = hclge_comm_cmd_csq_clean(hw);
+	if (handle < 0)
+		ret = handle;
+	else if (handle != num)
+		dev_warn(&hw->cmq.csq.pdev->dev,
+			 "cleaned %d, need to clean %d\n", handle, num);
+	return ret;
+}
+
+/**
+ * hclge_comm_cmd_send - send command to command queue
+ * @hw: pointer to the hw struct
+ * @desc: prefilled descriptor for describing the command
+ * @num : the number of descriptors to be sent
+ *
+ * This is the main send command for command queue, it
+ * sends the queue, cleans the queue, etc
+ **/
+int hclge_comm_cmd_send(struct hclge_comm_hw *hw, struct hclge_desc *desc,
+			int num)
+{
+	bool is_special = hclge_comm_is_special_opcode(le16_to_cpu(desc->opcode));
+	struct hclge_comm_cmq_ring *csq = &hw->cmq.csq;
+	int ret;
+	int ntc;
+
+	if (hw->cmq.ops.trace_cmd_send)
+		hw->cmq.ops.trace_cmd_send(hw, desc, num, is_special);
+
+	spin_lock_bh(&hw->cmq.csq.lock);
+
+	if (test_bit(HCLGE_COMM_STATE_CMD_DISABLE, &hw->comm_state)) {
+		spin_unlock_bh(&hw->cmq.csq.lock);
+		return -EBUSY;
+	}
+
+	if (num > hclge_comm_ring_space(&hw->cmq.csq)) {
+		/* If CMDQ ring is full, SW HEAD and HW HEAD may be different,
+		 * need update the SW HEAD pointer csq->next_to_clean
+		 */
+		csq->next_to_clean =
+			hclge_comm_read_dev(hw, HCLGE_COMM_NIC_CSQ_HEAD_REG);
+		spin_unlock_bh(&hw->cmq.csq.lock);
+		return -EBUSY;
+	}
+
+	/**
+	 * Record the location of desc in the ring for this time
+	 * which will be use for hardware to write back
+	 */
+	ntc = hw->cmq.csq.next_to_use;
+
+	hclge_comm_cmd_copy_desc(hw, desc, num);
+
+	/* Write to hardware */
+	hclge_comm_write_dev(hw, HCLGE_COMM_NIC_CSQ_TAIL_REG,
+			     hw->cmq.csq.next_to_use);
+
+	ret = hclge_comm_cmd_check_result(hw, desc, num, ntc);
+
+	spin_unlock_bh(&hw->cmq.csq.lock);
+
+	if (hw->cmq.ops.trace_cmd_get)
+		hw->cmq.ops.trace_cmd_get(hw, desc, num, is_special);
+
+	return ret;
+}
+EXPORT_SYMBOL_GPL(hclge_comm_cmd_send);
+
+static void hclge_comm_cmd_uninit_regs(struct hclge_comm_hw *hw)
+{
+	hclge_comm_write_dev(hw, HCLGE_COMM_NIC_CSQ_BASEADDR_L_REG, 0);
+	hclge_comm_write_dev(hw, HCLGE_COMM_NIC_CSQ_BASEADDR_H_REG, 0);
+	hclge_comm_write_dev(hw, HCLGE_COMM_NIC_CSQ_DEPTH_REG, 0);
+	hclge_comm_write_dev(hw, HCLGE_COMM_NIC_CSQ_HEAD_REG, 0);
+	hclge_comm_write_dev(hw, HCLGE_COMM_NIC_CSQ_TAIL_REG, 0);
+	hclge_comm_write_dev(hw, HCLGE_COMM_NIC_CRQ_BASEADDR_L_REG, 0);
+	hclge_comm_write_dev(hw, HCLGE_COMM_NIC_CRQ_BASEADDR_H_REG, 0);
+	hclge_comm_write_dev(hw, HCLGE_COMM_NIC_CRQ_DEPTH_REG, 0);
+	hclge_comm_write_dev(hw, HCLGE_COMM_NIC_CRQ_HEAD_REG, 0);
+	hclge_comm_write_dev(hw, HCLGE_COMM_NIC_CRQ_TAIL_REG, 0);
+}
+
+void hclge_comm_cmd_uninit(struct hnae3_ae_dev *ae_dev,
+			   struct hclge_comm_hw *hw)
+{
+	struct hclge_comm_cmq *cmdq = &hw->cmq;
+
+	hclge_comm_firmware_compat_config(ae_dev, hw, false);
+	set_bit(HCLGE_COMM_STATE_CMD_DISABLE, &hw->comm_state);
+
+	/* wait to ensure that the firmware completes the possible left
+	 * over commands.
+	 */
+	msleep(HCLGE_COMM_CMDQ_CLEAR_WAIT_TIME);
+	spin_lock_bh(&cmdq->csq.lock);
+	spin_lock(&cmdq->crq.lock);
+	hclge_comm_cmd_uninit_regs(hw);
+	spin_unlock(&cmdq->crq.lock);
+	spin_unlock_bh(&cmdq->csq.lock);
+
+	hclge_comm_free_cmd_desc(&cmdq->csq);
+	hclge_comm_free_cmd_desc(&cmdq->crq);
+}
+EXPORT_SYMBOL_GPL(hclge_comm_cmd_uninit);
+
+int hclge_comm_cmd_queue_init(struct pci_dev *pdev, struct hclge_comm_hw *hw)
+{
+	struct hclge_comm_cmq *cmdq = &hw->cmq;
+	int ret;
+
+	/* Setup the lock for command queue */
+	spin_lock_init(&cmdq->csq.lock);
+	spin_lock_init(&cmdq->crq.lock);
+
+	cmdq->csq.pdev = pdev;
+	cmdq->crq.pdev = pdev;
+
+	/* Setup the queue entries for use cmd queue */
+	cmdq->csq.desc_num = HCLGE_COMM_NIC_CMQ_DESC_NUM;
+	cmdq->crq.desc_num = HCLGE_COMM_NIC_CMQ_DESC_NUM;
+
+	/* Setup Tx write back timeout */
+	cmdq->tx_timeout = HCLGE_COMM_CMDQ_TX_TIMEOUT_DEFAULT;
+
+	/* Setup queue rings */
+	ret = hclge_comm_alloc_cmd_queue(hw, HCLGE_COMM_TYPE_CSQ);
+	if (ret) {
+		dev_err(&pdev->dev, "CSQ ring setup error %d\n", ret);
+		return ret;
+	}
+
+	ret = hclge_comm_alloc_cmd_queue(hw, HCLGE_COMM_TYPE_CRQ);
+	if (ret) {
+		dev_err(&pdev->dev, "CRQ ring setup error %d\n", ret);
+		goto err_csq;
+	}
+
+	return 0;
+err_csq:
+	hclge_comm_free_cmd_desc(&hw->cmq.csq);
+	return ret;
+}
+EXPORT_SYMBOL_GPL(hclge_comm_cmd_queue_init);
+
+void hclge_comm_cmd_init_ops(struct hclge_comm_hw *hw,
+			     const struct hclge_comm_cmq_ops *ops)
+{
+	struct hclge_comm_cmq *cmdq = &hw->cmq;
+
+	if (ops) {
+		cmdq->ops.trace_cmd_send = ops->trace_cmd_send;
+		cmdq->ops.trace_cmd_get = ops->trace_cmd_get;
+	}
+}
+EXPORT_SYMBOL_GPL(hclge_comm_cmd_init_ops);
+
+int hclge_comm_cmd_init(struct hnae3_ae_dev *ae_dev, struct hclge_comm_hw *hw,
+			u32 *fw_version, bool is_pf,
+			unsigned long reset_pending)
+{
+	struct hclge_comm_cmq *cmdq = &hw->cmq;
+	int ret;
+
+	spin_lock_bh(&cmdq->csq.lock);
+	spin_lock(&cmdq->crq.lock);
+
+	cmdq->csq.next_to_clean = 0;
+	cmdq->csq.next_to_use = 0;
+	cmdq->crq.next_to_clean = 0;
+	cmdq->crq.next_to_use = 0;
+
+	hclge_comm_cmd_init_regs(hw);
+
+	spin_unlock(&cmdq->crq.lock);
+	spin_unlock_bh(&cmdq->csq.lock);
+
+	clear_bit(HCLGE_COMM_STATE_CMD_DISABLE, &hw->comm_state);
+
+	/* Check if there is new reset pending, because the higher level
+	 * reset may happen when lower level reset is being processed.
+	 */
+	if (reset_pending) {
+		ret = -EBUSY;
+		goto err_cmd_init;
+	}
+
+	/* get version and device capabilities */
+	ret = hclge_comm_cmd_query_version_and_capability(ae_dev, hw,
+							  fw_version, is_pf);
+	if (ret) {
+		dev_err(&ae_dev->pdev->dev,
+			"failed to query version and capabilities, ret = %d\n",
+			ret);
+		goto err_cmd_init;
+	}
+
+	dev_info(&ae_dev->pdev->dev,
+		 "The firmware version is %lu.%lu.%lu.%lu\n",
+		 hnae3_get_field(*fw_version, HNAE3_FW_VERSION_BYTE3_MASK,
+				 HNAE3_FW_VERSION_BYTE3_SHIFT),
+		 hnae3_get_field(*fw_version, HNAE3_FW_VERSION_BYTE2_MASK,
+				 HNAE3_FW_VERSION_BYTE2_SHIFT),
+		 hnae3_get_field(*fw_version, HNAE3_FW_VERSION_BYTE1_MASK,
+				 HNAE3_FW_VERSION_BYTE1_SHIFT),
+		 hnae3_get_field(*fw_version, HNAE3_FW_VERSION_BYTE0_MASK,
+				 HNAE3_FW_VERSION_BYTE0_SHIFT));
+
+	if (!is_pf && ae_dev->dev_version < HNAE3_DEVICE_VERSION_V3)
+		return 0;
+
+	/* ask the firmware to enable some features, driver can work without
+	 * it.
+	 */
+	ret = hclge_comm_firmware_compat_config(ae_dev, hw, true);
+	if (ret)
+		dev_warn(&ae_dev->pdev->dev,
+			 "Firmware compatible features not enabled(%d).\n",
+			 ret);
+	return 0;
+
+err_cmd_init:
+	set_bit(HCLGE_COMM_STATE_CMD_DISABLE, &hw->comm_state);
+
+	return ret;
+}
+EXPORT_SYMBOL_GPL(hclge_comm_cmd_init);
+
+MODULE_LICENSE("GPL");
+MODULE_DESCRIPTION("HNS3: Hisilicon Ethernet PF/VF Common Library");
+MODULE_AUTHOR("Huawei Tech. Co., Ltd.");
diff --git a/drivers/net/ethernet/hisilicon/hns3/hns3_common/hclge_comm_cmd.h b/drivers/net/ethernet/hisilicon/hns3/hns3_common/hclge_comm_cmd.h
new file mode 100644
index 000000000000..2c2a2f1e0d7a
--- /dev/null
+++ b/drivers/net/ethernet/hisilicon/hns3/hns3_common/hclge_comm_cmd.h
@@ -0,0 +1,500 @@
+/* SPDX-License-Identifier: GPL-2.0+ */
+// Copyright (c) 2021-2021 Hisilicon Limited.
+
+#ifndef __HCLGE_COMM_CMD_H
+#define __HCLGE_COMM_CMD_H
+#include <linux/types.h>
+
+#include "hnae3.h"
+
+#define HCLGE_COMM_CMD_FLAG_IN			BIT(0)
+#define HCLGE_COMM_CMD_FLAG_NEXT		BIT(2)
+#define HCLGE_COMM_CMD_FLAG_WR			BIT(3)
+#define HCLGE_COMM_CMD_FLAG_NO_INTR		BIT(4)
+
+#define HCLGE_COMM_SEND_SYNC(flag) \
+	((flag) & HCLGE_COMM_CMD_FLAG_NO_INTR)
+
+#define HCLGE_COMM_LINK_EVENT_REPORT_EN_B	0
+#define HCLGE_COMM_NCSI_ERROR_REPORT_EN_B	1
+#define HCLGE_COMM_PHY_IMP_EN_B			2
+#define HCLGE_COMM_MAC_STATS_EXT_EN_B		3
+#define HCLGE_COMM_SYNC_RX_RING_HEAD_EN_B	4
+#define HCLGE_COMM_LLRS_FEC_EN_B		5
+
+#define hclge_comm_dev_phy_imp_supported(ae_dev) \
+	test_bit(HNAE3_DEV_SUPPORT_PHY_IMP_B, (ae_dev)->caps)
+
+#define HCLGE_COMM_TYPE_CRQ			0
+#define HCLGE_COMM_TYPE_CSQ			1
+
+#define HCLGE_COMM_CMDQ_CLEAR_WAIT_TIME		200
+
+/* bar registers for cmdq */
+#define HCLGE_COMM_NIC_CSQ_BASEADDR_L_REG	0x27000
+#define HCLGE_COMM_NIC_CSQ_BASEADDR_H_REG	0x27004
+#define HCLGE_COMM_NIC_CSQ_DEPTH_REG		0x27008
+#define HCLGE_COMM_NIC_CSQ_TAIL_REG		0x27010
+#define HCLGE_COMM_NIC_CSQ_HEAD_REG		0x27014
+#define HCLGE_COMM_NIC_CRQ_BASEADDR_L_REG	0x27018
+#define HCLGE_COMM_NIC_CRQ_BASEADDR_H_REG	0x2701C
+#define HCLGE_COMM_NIC_CRQ_DEPTH_REG		0x27020
+#define HCLGE_COMM_NIC_CRQ_TAIL_REG		0x27024
+#define HCLGE_COMM_NIC_CRQ_HEAD_REG		0x27028
+/* Vector0 interrupt CMDQ event source register(RW) */
+#define HCLGE_COMM_VECTOR0_CMDQ_SRC_REG		0x27100
+/* Vector0 interrupt CMDQ event status register(RO) */
+#define HCLGE_COMM_VECTOR0_CMDQ_STATE_REG	0x27104
+#define HCLGE_COMM_CMDQ_INTR_EN_REG		0x27108
+#define HCLGE_COMM_CMDQ_INTR_GEN_REG		0x2710C
+#define HCLGE_COMM_CMDQ_INTR_STS_REG		0x27104
+
+/* this bit indicates that the driver is ready for hardware reset */
+#define HCLGE_COMM_NIC_SW_RST_RDY_B		16
+#define HCLGE_COMM_NIC_SW_RST_RDY		BIT(HCLGE_COMM_NIC_SW_RST_RDY_B)
+#define HCLGE_COMM_NIC_CMQ_DESC_NUM_S		3
+#define HCLGE_COMM_NIC_CMQ_DESC_NUM		1024
+#define HCLGE_COMM_CMDQ_TX_TIMEOUT_DEFAULT	30000
+#define HCLGE_COMM_CMDQ_CFG_RST_TIMEOUT		1000000
+
+enum hclge_opcode_type {
+	/* Generic commands */
+	HCLGE_OPC_QUERY_FW_VER		= 0x0001,
+	HCLGE_OPC_CFG_RST_TRIGGER	= 0x0020,
+	HCLGE_OPC_GBL_RST_STATUS	= 0x0021,
+	HCLGE_OPC_QUERY_FUNC_STATUS	= 0x0022,
+	HCLGE_OPC_QUERY_PF_RSRC		= 0x0023,
+	HCLGE_OPC_QUERY_VF_RSRC		= 0x0024,
+	HCLGE_OPC_GET_CFG_PARAM		= 0x0025,
+	HCLGE_OPC_PF_RST_DONE		= 0x0026,
+	HCLGE_OPC_QUERY_VF_RST_RDY	= 0x0027,
+
+	HCLGE_OPC_STATS_64_BIT		= 0x0030,
+	HCLGE_OPC_STATS_32_BIT		= 0x0031,
+	HCLGE_OPC_STATS_MAC		= 0x0032,
+	HCLGE_OPC_QUERY_MAC_REG_NUM	= 0x0033,
+	HCLGE_OPC_STATS_MAC_ALL		= 0x0034,
+
+	HCLGE_OPC_QUERY_REG_NUM		= 0x0040,
+	HCLGE_OPC_QUERY_32_BIT_REG	= 0x0041,
+	HCLGE_OPC_QUERY_64_BIT_REG	= 0x0042,
+	HCLGE_OPC_DFX_BD_NUM		= 0x0043,
+	HCLGE_OPC_DFX_BIOS_COMMON_REG	= 0x0044,
+	HCLGE_OPC_DFX_SSU_REG_0		= 0x0045,
+	HCLGE_OPC_DFX_SSU_REG_1		= 0x0046,
+	HCLGE_OPC_DFX_IGU_EGU_REG	= 0x0047,
+	HCLGE_OPC_DFX_RPU_REG_0		= 0x0048,
+	HCLGE_OPC_DFX_RPU_REG_1		= 0x0049,
+	HCLGE_OPC_DFX_NCSI_REG		= 0x004A,
+	HCLGE_OPC_DFX_RTC_REG		= 0x004B,
+	HCLGE_OPC_DFX_PPP_REG		= 0x004C,
+	HCLGE_OPC_DFX_RCB_REG		= 0x004D,
+	HCLGE_OPC_DFX_TQP_REG		= 0x004E,
+	HCLGE_OPC_DFX_SSU_REG_2		= 0x004F,
+	HCLGE_OPC_DFX_GEN_REG		= 0x7038,
+
+	HCLGE_OPC_QUERY_DEV_SPECS	= 0x0050,
+	HCLGE_OPC_GET_QUEUE_ERR_VF      = 0x0067,
+
+	/* MAC command */
+	HCLGE_OPC_CONFIG_MAC_MODE	= 0x0301,
+	HCLGE_OPC_CONFIG_AN_MODE	= 0x0304,
+	HCLGE_OPC_QUERY_LINK_STATUS	= 0x0307,
+	HCLGE_OPC_CONFIG_MAX_FRM_SIZE	= 0x0308,
+	HCLGE_OPC_CONFIG_SPEED_DUP	= 0x0309,
+	HCLGE_OPC_QUERY_MAC_TNL_INT	= 0x0310,
+	HCLGE_OPC_MAC_TNL_INT_EN	= 0x0311,
+	HCLGE_OPC_CLEAR_MAC_TNL_INT	= 0x0312,
+	HCLGE_OPC_COMMON_LOOPBACK       = 0x0315,
+	HCLGE_OPC_QUERY_FEC_STATS	= 0x0316,
+	HCLGE_OPC_CONFIG_FEC_MODE	= 0x031A,
+	HCLGE_OPC_QUERY_ROH_TYPE_INFO	= 0x0389,
+
+	/* PTP commands */
+	HCLGE_OPC_PTP_INT_EN		= 0x0501,
+	HCLGE_OPC_PTP_MODE_CFG		= 0x0507,
+
+	/* PFC/Pause commands */
+	HCLGE_OPC_CFG_MAC_PAUSE_EN      = 0x0701,
+	HCLGE_OPC_CFG_PFC_PAUSE_EN      = 0x0702,
+	HCLGE_OPC_CFG_MAC_PARA          = 0x0703,
+	HCLGE_OPC_CFG_PFC_PARA          = 0x0704,
+	HCLGE_OPC_QUERY_MAC_TX_PKT_CNT  = 0x0705,
+	HCLGE_OPC_QUERY_MAC_RX_PKT_CNT  = 0x0706,
+	HCLGE_OPC_QUERY_PFC_TX_PKT_CNT  = 0x0707,
+	HCLGE_OPC_QUERY_PFC_RX_PKT_CNT  = 0x0708,
+	HCLGE_OPC_PRI_TO_TC_MAPPING     = 0x0709,
+	HCLGE_OPC_QOS_MAP               = 0x070A,
+
+	/* ETS/scheduler commands */
+	HCLGE_OPC_TM_PG_TO_PRI_LINK	= 0x0804,
+	HCLGE_OPC_TM_QS_TO_PRI_LINK     = 0x0805,
+	HCLGE_OPC_TM_NQ_TO_QS_LINK      = 0x0806,
+	HCLGE_OPC_TM_RQ_TO_QS_LINK      = 0x0807,
+	HCLGE_OPC_TM_PORT_WEIGHT        = 0x0808,
+	HCLGE_OPC_TM_PG_WEIGHT          = 0x0809,
+	HCLGE_OPC_TM_QS_WEIGHT          = 0x080A,
+	HCLGE_OPC_TM_PRI_WEIGHT         = 0x080B,
+	HCLGE_OPC_TM_PRI_C_SHAPPING     = 0x080C,
+	HCLGE_OPC_TM_PRI_P_SHAPPING     = 0x080D,
+	HCLGE_OPC_TM_PG_C_SHAPPING      = 0x080E,
+	HCLGE_OPC_TM_PG_P_SHAPPING      = 0x080F,
+	HCLGE_OPC_TM_PORT_SHAPPING      = 0x0810,
+	HCLGE_OPC_TM_PG_SCH_MODE_CFG    = 0x0812,
+	HCLGE_OPC_TM_PRI_SCH_MODE_CFG   = 0x0813,
+	HCLGE_OPC_TM_QS_SCH_MODE_CFG    = 0x0814,
+	HCLGE_OPC_TM_BP_TO_QSET_MAPPING = 0x0815,
+	HCLGE_OPC_TM_NODES		= 0x0816,
+	HCLGE_OPC_ETS_TC_WEIGHT		= 0x0843,
+	HCLGE_OPC_QSET_DFX_STS		= 0x0844,
+	HCLGE_OPC_PRI_DFX_STS		= 0x0845,
+	HCLGE_OPC_PG_DFX_STS		= 0x0846,
+	HCLGE_OPC_PORT_DFX_STS		= 0x0847,
+	HCLGE_OPC_SCH_NQ_CNT		= 0x0848,
+	HCLGE_OPC_SCH_RQ_CNT		= 0x0849,
+	HCLGE_OPC_TM_INTERNAL_STS	= 0x0850,
+	HCLGE_OPC_TM_INTERNAL_CNT	= 0x0851,
+	HCLGE_OPC_TM_INTERNAL_STS_1	= 0x0852,
+	HCLGE_OPC_TM_FLUSH		= 0x0872,
+
+	/* Packet buffer allocate commands */
+	HCLGE_OPC_TX_BUFF_ALLOC		= 0x0901,
+	HCLGE_OPC_RX_PRIV_BUFF_ALLOC	= 0x0902,
+	HCLGE_OPC_RX_PRIV_WL_ALLOC	= 0x0903,
+	HCLGE_OPC_RX_COM_THRD_ALLOC	= 0x0904,
+	HCLGE_OPC_RX_COM_WL_ALLOC	= 0x0905,
+	HCLGE_OPC_RX_GBL_PKT_CNT	= 0x0906,
+
+	/* TQP management command */
+	HCLGE_OPC_SET_TQP_MAP		= 0x0A01,
+
+	/* TQP commands */
+	HCLGE_OPC_CFG_TX_QUEUE		= 0x0B01,
+	HCLGE_OPC_QUERY_TX_POINTER	= 0x0B02,
+	HCLGE_OPC_QUERY_TX_STATS	= 0x0B03,
+	HCLGE_OPC_TQP_TX_QUEUE_TC	= 0x0B04,
+	HCLGE_OPC_CFG_RX_QUEUE		= 0x0B11,
+	HCLGE_OPC_QUERY_RX_POINTER	= 0x0B12,
+	HCLGE_OPC_QUERY_RX_STATS	= 0x0B13,
+	HCLGE_OPC_STASH_RX_QUEUE_LRO	= 0x0B16,
+	HCLGE_OPC_CFG_RX_QUEUE_LRO	= 0x0B17,
+	HCLGE_OPC_CFG_COM_TQP_QUEUE	= 0x0B20,
+	HCLGE_OPC_RESET_TQP_QUEUE	= 0x0B22,
+
+	/* PPU commands */
+	HCLGE_OPC_PPU_PF_OTHER_INT_DFX	= 0x0B4A,
+
+	/* TSO command */
+	HCLGE_OPC_TSO_GENERIC_CONFIG	= 0x0C01,
+	HCLGE_OPC_GRO_GENERIC_CONFIG    = 0x0C10,
+
+	/* RSS commands */
+	HCLGE_OPC_RSS_GENERIC_CONFIG	= 0x0D01,
+	HCLGE_OPC_RSS_INDIR_TABLE	= 0x0D07,
+	HCLGE_OPC_RSS_TC_MODE		= 0x0D08,
+	HCLGE_OPC_RSS_INPUT_TUPLE	= 0x0D02,
+
+	/* Promisuous mode command */
+	HCLGE_OPC_CFG_PROMISC_MODE	= 0x0E01,
+
+	/* Vlan offload commands */
+	HCLGE_OPC_VLAN_PORT_TX_CFG	= 0x0F01,
+	HCLGE_OPC_VLAN_PORT_RX_CFG	= 0x0F02,
+
+	/* Interrupts commands */
+	HCLGE_OPC_ADD_RING_TO_VECTOR	= 0x1503,
+	HCLGE_OPC_DEL_RING_TO_VECTOR	= 0x1504,
+
+	/* MAC commands */
+	HCLGE_OPC_MAC_VLAN_ADD		    = 0x1000,
+	HCLGE_OPC_MAC_VLAN_REMOVE	    = 0x1001,
+	HCLGE_OPC_MAC_VLAN_TYPE_ID	    = 0x1002,
+	HCLGE_OPC_MAC_VLAN_INSERT	    = 0x1003,
+	HCLGE_OPC_MAC_VLAN_ALLOCATE	    = 0x1004,
+	HCLGE_OPC_MAC_ETHTYPE_ADD	    = 0x1010,
+	HCLGE_OPC_MAC_ETHTYPE_REMOVE	= 0x1011,
+
+	/* MAC VLAN commands */
+	HCLGE_OPC_MAC_VLAN_SWITCH_PARAM	= 0x1033,
+
+	/* VLAN commands */
+	HCLGE_OPC_VLAN_FILTER_CTRL	    = 0x1100,
+	HCLGE_OPC_VLAN_FILTER_PF_CFG	= 0x1101,
+	HCLGE_OPC_VLAN_FILTER_VF_CFG	= 0x1102,
+	HCLGE_OPC_PORT_VLAN_BYPASS	= 0x1103,
+
+	/* Flow Director commands */
+	HCLGE_OPC_FD_MODE_CTRL		= 0x1200,
+	HCLGE_OPC_FD_GET_ALLOCATION	= 0x1201,
+	HCLGE_OPC_FD_KEY_CONFIG		= 0x1202,
+	HCLGE_OPC_FD_TCAM_OP		= 0x1203,
+	HCLGE_OPC_FD_AD_OP		= 0x1204,
+	HCLGE_OPC_FD_CNT_OP		= 0x1205,
+	HCLGE_OPC_FD_USER_DEF_OP	= 0x1207,
+	HCLGE_OPC_FD_QB_CTRL		= 0x1210,
+	HCLGE_OPC_FD_QB_AD_OP		= 0x1211,
+
+	/* MDIO command */
+	HCLGE_OPC_MDIO_CONFIG		= 0x1900,
+
+	/* QCN commands */
+	HCLGE_OPC_QCN_MOD_CFG		= 0x1A01,
+	HCLGE_OPC_QCN_GRP_TMPLT_CFG	= 0x1A02,
+	HCLGE_OPC_QCN_SHAPPING_CFG	= 0x1A03,
+	HCLGE_OPC_QCN_SHAPPING_BS_CFG	= 0x1A04,
+	HCLGE_OPC_QCN_QSET_LINK_CFG	= 0x1A05,
+	HCLGE_OPC_QCN_RP_STATUS_GET	= 0x1A06,
+	HCLGE_OPC_QCN_AJUST_INIT	= 0x1A07,
+	HCLGE_OPC_QCN_DFX_CNT_STATUS    = 0x1A08,
+
+	/* SCC commands */
+	HCLGE_OPC_QUERY_SCC_VER		= 0x1A84,
+
+	/* Mailbox command */
+	HCLGEVF_OPC_MBX_PF_TO_VF	= 0x2000,
+	HCLGEVF_OPC_MBX_VF_TO_PF	= 0x2001,
+
+	/* Led command */
+	HCLGE_OPC_LED_STATUS_CFG	= 0xB000,
+
+	/* clear hardware resource command */
+	HCLGE_OPC_CLEAR_HW_RESOURCE	= 0x700B,
+
+	/* NCL config command */
+	HCLGE_OPC_QUERY_NCL_CONFIG	= 0x7011,
+
+	/* IMP stats command */
+	HCLGE_OPC_IMP_STATS_BD		= 0x7012,
+	HCLGE_OPC_IMP_STATS_INFO		= 0x7013,
+	HCLGE_OPC_IMP_COMPAT_CFG		= 0x701A,
+
+	/* SFP command */
+	HCLGE_OPC_GET_SFP_EEPROM	= 0x7100,
+	HCLGE_OPC_GET_SFP_EXIST		= 0x7101,
+	HCLGE_OPC_GET_SFP_INFO		= 0x7104,
+
+	/* Error INT commands */
+	HCLGE_MAC_COMMON_INT_EN		= 0x030E,
+	HCLGE_TM_SCH_ECC_INT_EN		= 0x0829,
+	HCLGE_SSU_ECC_INT_CMD		= 0x0989,
+	HCLGE_SSU_COMMON_INT_CMD	= 0x098C,
+	HCLGE_PPU_MPF_ECC_INT_CMD	= 0x0B40,
+	HCLGE_PPU_MPF_OTHER_INT_CMD	= 0x0B41,
+	HCLGE_PPU_PF_OTHER_INT_CMD	= 0x0B42,
+	HCLGE_COMMON_ECC_INT_CFG	= 0x1505,
+	HCLGE_QUERY_RAS_INT_STS_BD_NUM	= 0x1510,
+	HCLGE_QUERY_CLEAR_MPF_RAS_INT	= 0x1511,
+	HCLGE_QUERY_CLEAR_PF_RAS_INT	= 0x1512,
+	HCLGE_QUERY_MSIX_INT_STS_BD_NUM	= 0x1513,
+	HCLGE_QUERY_CLEAR_ALL_MPF_MSIX_INT	= 0x1514,
+	HCLGE_QUERY_CLEAR_ALL_PF_MSIX_INT	= 0x1515,
+	HCLGE_QUERY_ALL_ERR_BD_NUM		= 0x1516,
+	HCLGE_QUERY_ALL_ERR_INFO		= 0x1517,
+	HCLGE_CONFIG_ROCEE_RAS_INT_EN	= 0x1580,
+	HCLGE_QUERY_CLEAR_ROCEE_RAS_INT = 0x1581,
+	HCLGE_ROCEE_PF_RAS_INT_CMD	= 0x1584,
+	HCLGE_QUERY_ROCEE_ECC_RAS_INFO_CMD	= 0x1585,
+	HCLGE_QUERY_ROCEE_AXI_RAS_INFO_CMD	= 0x1586,
+	HCLGE_IGU_EGU_TNL_INT_EN	= 0x1803,
+	HCLGE_IGU_COMMON_INT_EN		= 0x1806,
+	HCLGE_TM_QCN_MEM_INT_CFG	= 0x1A14,
+	HCLGE_PPP_CMD0_INT_CMD		= 0x2100,
+	HCLGE_PPP_CMD1_INT_CMD		= 0x2101,
+	HCLGE_MAC_ETHERTYPE_IDX_RD      = 0x2105,
+	HCLGE_OPC_WOL_GET_SUPPORTED_MODE	= 0x2201,
+	HCLGE_OPC_WOL_CFG		= 0x2202,
+	HCLGE_NCSI_INT_EN		= 0x2401,
+
+	/* ROH MAC commands */
+	HCLGE_OPC_MAC_ADDR_CHECK	= 0x9004,
+
+	/* PHY command */
+	HCLGE_OPC_PHY_LINK_KSETTING	= 0x7025,
+	HCLGE_OPC_PHY_REG		= 0x7026,
+
+	/* Query link diagnosis info command */
+	HCLGE_OPC_QUERY_LINK_DIAGNOSIS	= 0x702A,
+};
+
+enum hclge_comm_cmd_return_status {
+	HCLGE_COMM_CMD_EXEC_SUCCESS	= 0,
+	HCLGE_COMM_CMD_NO_AUTH		= 1,
+	HCLGE_COMM_CMD_NOT_SUPPORTED	= 2,
+	HCLGE_COMM_CMD_QUEUE_FULL	= 3,
+	HCLGE_COMM_CMD_NEXT_ERR		= 4,
+	HCLGE_COMM_CMD_UNEXE_ERR	= 5,
+	HCLGE_COMM_CMD_PARA_ERR		= 6,
+	HCLGE_COMM_CMD_RESULT_ERR	= 7,
+	HCLGE_COMM_CMD_TIMEOUT		= 8,
+	HCLGE_COMM_CMD_HILINK_ERR	= 9,
+	HCLGE_COMM_CMD_QUEUE_ILLEGAL	= 10,
+	HCLGE_COMM_CMD_INVALID		= 11,
+};
+
+enum HCLGE_COMM_CAP_BITS {
+	HCLGE_COMM_CAP_UDP_GSO_B,
+	HCLGE_COMM_CAP_QB_B,
+	HCLGE_COMM_CAP_FD_FORWARD_TC_B,
+	HCLGE_COMM_CAP_PTP_B,
+	HCLGE_COMM_CAP_INT_QL_B,
+	HCLGE_COMM_CAP_HW_TX_CSUM_B,
+	HCLGE_COMM_CAP_TX_PUSH_B,
+	HCLGE_COMM_CAP_PHY_IMP_B,
+	HCLGE_COMM_CAP_TQP_TXRX_INDEP_B,
+	HCLGE_COMM_CAP_HW_PAD_B,
+	HCLGE_COMM_CAP_STASH_B,
+	HCLGE_COMM_CAP_UDP_TUNNEL_CSUM_B,
+	HCLGE_COMM_CAP_RAS_IMP_B = 12,
+	HCLGE_COMM_CAP_FEC_B = 13,
+	HCLGE_COMM_CAP_PAUSE_B = 14,
+	HCLGE_COMM_CAP_RXD_ADV_LAYOUT_B = 15,
+	HCLGE_COMM_CAP_PORT_VLAN_BYPASS_B = 17,
+	HCLGE_COMM_CAP_CQ_B = 18,
+	HCLGE_COMM_CAP_GRO_B = 20,
+	HCLGE_COMM_CAP_FD_B = 21,
+	HCLGE_COMM_CAP_FEC_STATS_B = 25,
+	HCLGE_COMM_CAP_VF_FAULT_B = 26,
+	HCLGE_COMM_CAP_LANE_NUM_B = 27,
+	HCLGE_COMM_CAP_WOL_B = 28,
+	HCLGE_COMM_CAP_TM_FLUSH_B = 31,
+	HCLGE_COMM_CAP_ERR_MOD_GEN_REG_B = 32,
+};
+
+enum HCLGE_COMM_API_CAP_BITS {
+	HCLGE_COMM_API_CAP_FLEX_RSS_TBL_B,
+};
+
+/* capabilities bits map between imp firmware and local driver */
+struct hclge_comm_caps_bit_map {
+	u16 imp_bit;
+	u16 local_bit;
+};
+
+struct hclge_cmdq_tx_timeout_map {
+	u32 opcode;
+	u32 tx_timeout;
+};
+
+struct hclge_comm_firmware_compat_cmd {
+	__le32 compat;
+	u8 rsv[20];
+};
+
+enum hclge_comm_cmd_state {
+	HCLGE_COMM_STATE_CMD_DISABLE,
+};
+
+struct hclge_comm_errcode {
+	u32 imp_errcode;
+	int common_errno;
+};
+
+#define HCLGE_COMM_QUERY_CAP_LENGTH		3
+struct hclge_comm_query_version_cmd {
+	__le32 firmware;
+	__le32 hardware;
+	__le32 api_caps;
+	__le32 caps[HCLGE_COMM_QUERY_CAP_LENGTH]; /* capabilities of device */
+};
+
+struct hclge_comm_query_scc_cmd {
+	__le32 scc_version;
+	u8 rsv[20];
+};
+
+#define HCLGE_DESC_DATA_LEN		6
+struct hclge_desc {
+	__le16 opcode;
+	__le16 flag;
+	__le16 retval;
+	__le16 rsv;
+	__le32 data[HCLGE_DESC_DATA_LEN];
+};
+
+struct hclge_comm_cmq_ring {
+	dma_addr_t desc_dma_addr;
+	struct hclge_desc *desc;
+	struct pci_dev *pdev;
+	u32 head;
+	u32 tail;
+
+	u16 buf_size;
+	u16 desc_num;
+	int next_to_use;
+	int next_to_clean;
+	u8 ring_type; /* cmq ring type */
+	spinlock_t lock; /* Command queue lock */
+};
+
+enum hclge_comm_cmd_status {
+	HCLGE_COMM_STATUS_SUCCESS	= 0,
+	HCLGE_COMM_ERR_CSQ_FULL		= -1,
+	HCLGE_COMM_ERR_CSQ_TIMEOUT	= -2,
+	HCLGE_COMM_ERR_CSQ_ERROR	= -3,
+};
+
+struct hclge_comm_hw;
+struct hclge_comm_cmq_ops {
+	void (*trace_cmd_send)(struct hclge_comm_hw *hw,
+			       struct hclge_desc *desc,
+			       int num, bool is_special);
+	void (*trace_cmd_get)(struct hclge_comm_hw *hw,
+			      struct hclge_desc *desc,
+			      int num, bool is_special);
+};
+
+struct hclge_comm_cmq {
+	struct hclge_comm_cmq_ring csq;
+	struct hclge_comm_cmq_ring crq;
+	u16 tx_timeout;
+	enum hclge_comm_cmd_status last_status;
+	struct hclge_comm_cmq_ops ops;
+};
+
+struct hclge_comm_hw {
+	void __iomem *io_base;
+	void __iomem *mem_base;
+	struct hclge_comm_cmq cmq;
+	unsigned long comm_state;
+};
+
+static inline void hclge_comm_write_reg(void __iomem *base, u32 reg, u32 value)
+{
+	writel(value, base + reg);
+}
+
+static inline u32 hclge_comm_read_reg(u8 __iomem *base, u32 reg)
+{
+	u8 __iomem *reg_addr = READ_ONCE(base);
+
+	return readl(reg_addr + reg);
+}
+
+#define hclge_comm_write_dev(a, reg, value) \
+	hclge_comm_write_reg((a)->io_base, reg, value)
+#define hclge_comm_read_dev(a, reg) \
+	hclge_comm_read_reg((a)->io_base, reg)
+
+void hclge_comm_cmd_init_regs(struct hclge_comm_hw *hw);
+int hclge_comm_cmd_query_version_and_capability(struct hnae3_ae_dev *ae_dev,
+						struct hclge_comm_hw *hw,
+						u32 *fw_version, bool is_pf);
+int hclge_comm_alloc_cmd_queue(struct hclge_comm_hw *hw, int ring_type);
+int hclge_comm_cmd_send(struct hclge_comm_hw *hw, struct hclge_desc *desc,
+			int num);
+void hclge_comm_cmd_reuse_desc(struct hclge_desc *desc, bool is_read);
+int hclge_comm_firmware_compat_config(struct hnae3_ae_dev *ae_dev,
+				      struct hclge_comm_hw *hw, bool en);
+void hclge_comm_free_cmd_desc(struct hclge_comm_cmq_ring *ring);
+void hclge_comm_cmd_setup_basic_desc(struct hclge_desc *desc,
+				     enum hclge_opcode_type opcode,
+				     bool is_read);
+void hclge_comm_cmd_uninit(struct hnae3_ae_dev *ae_dev,
+			   struct hclge_comm_hw *hw);
+int hclge_comm_cmd_queue_init(struct pci_dev *pdev, struct hclge_comm_hw *hw);
+int hclge_comm_cmd_init(struct hnae3_ae_dev *ae_dev, struct hclge_comm_hw *hw,
+			u32 *fw_version, bool is_pf,
+			unsigned long reset_pending);
+void hclge_comm_cmd_init_ops(struct hclge_comm_hw *hw,
+			     const struct hclge_comm_cmq_ops *ops);
+#endif
diff --git a/drivers/net/ethernet/hisilicon/hns3/hns3_common/hclge_comm_rss.c b/drivers/net/ethernet/hisilicon/hns3/hns3_common/hclge_comm_rss.c
new file mode 100644
index 000000000000..1eca53aaf598
--- /dev/null
+++ b/drivers/net/ethernet/hisilicon/hns3/hns3_common/hclge_comm_rss.c
@@ -0,0 +1,518 @@
+// SPDX-License-Identifier: GPL-2.0+
+// Copyright (c) 2021-2021 Hisilicon Limited.
+#include <linux/skbuff.h>
+
+#include "hnae3.h"
+#include "hclge_comm_cmd.h"
+#include "hclge_comm_rss.h"
+
+static const u8 hclge_comm_hash_key[] = {
+	0x6D, 0x5A, 0x56, 0xDA, 0x25, 0x5B, 0x0E, 0xC2,
+	0x41, 0x67, 0x25, 0x3D, 0x43, 0xA3, 0x8F, 0xB0,
+	0xD0, 0xCA, 0x2B, 0xCB, 0xAE, 0x7B, 0x30, 0xB4,
+	0x77, 0xCB, 0x2D, 0xA3, 0x80, 0x30, 0xF2, 0x0C,
+	0x6A, 0x42, 0xB7, 0x3B, 0xBE, 0xAC, 0x01, 0xFA
+};
+
+static void
+hclge_comm_init_rss_tuple(struct hnae3_ae_dev *ae_dev,
+			  struct hclge_comm_rss_tuple_cfg *rss_tuple_cfg)
+{
+	rss_tuple_cfg->ipv4_tcp_en = HCLGE_COMM_RSS_INPUT_TUPLE_OTHER;
+	rss_tuple_cfg->ipv4_udp_en = HCLGE_COMM_RSS_INPUT_TUPLE_OTHER;
+	rss_tuple_cfg->ipv4_sctp_en = HCLGE_COMM_RSS_INPUT_TUPLE_SCTP;
+	rss_tuple_cfg->ipv4_fragment_en = HCLGE_COMM_RSS_INPUT_TUPLE_OTHER;
+	rss_tuple_cfg->ipv6_tcp_en = HCLGE_COMM_RSS_INPUT_TUPLE_OTHER;
+	rss_tuple_cfg->ipv6_udp_en = HCLGE_COMM_RSS_INPUT_TUPLE_OTHER;
+	rss_tuple_cfg->ipv6_sctp_en =
+		ae_dev->dev_version <= HNAE3_DEVICE_VERSION_V2 ?
+		HCLGE_COMM_RSS_INPUT_TUPLE_SCTP_NO_PORT :
+		HCLGE_COMM_RSS_INPUT_TUPLE_SCTP;
+	rss_tuple_cfg->ipv6_fragment_en = HCLGE_COMM_RSS_INPUT_TUPLE_OTHER;
+}
+
+int hclge_comm_rss_init_cfg(struct hnae3_handle *nic,
+			    struct hnae3_ae_dev *ae_dev,
+			    struct hclge_comm_rss_cfg *rss_cfg)
+{
+	u16 rss_ind_tbl_size = ae_dev->dev_specs.rss_ind_tbl_size;
+	int rss_algo = HCLGE_COMM_RSS_HASH_ALGO_TOEPLITZ;
+	u16 *rss_ind_tbl;
+
+	if (nic->flags & HNAE3_SUPPORT_VF)
+		rss_cfg->rss_size = nic->kinfo.rss_size;
+
+	if (ae_dev->dev_version >= HNAE3_DEVICE_VERSION_V2)
+		rss_algo = HCLGE_COMM_RSS_HASH_ALGO_SIMPLE;
+
+	hclge_comm_init_rss_tuple(ae_dev, &rss_cfg->rss_tuple_sets);
+
+	rss_cfg->rss_algo = rss_algo;
+
+	rss_ind_tbl = devm_kcalloc(&ae_dev->pdev->dev, rss_ind_tbl_size,
+				   sizeof(*rss_ind_tbl), GFP_KERNEL);
+	if (!rss_ind_tbl)
+		return -ENOMEM;
+
+	rss_cfg->rss_indirection_tbl = rss_ind_tbl;
+	memcpy(rss_cfg->rss_hash_key, hclge_comm_hash_key,
+	       HCLGE_COMM_RSS_KEY_SIZE);
+
+	hclge_comm_rss_indir_init_cfg(ae_dev, rss_cfg);
+
+	return 0;
+}
+EXPORT_SYMBOL_GPL(hclge_comm_rss_init_cfg);
+
+void hclge_comm_get_rss_tc_info(u16 rss_size, u8 hw_tc_map, u16 *tc_offset,
+				u16 *tc_valid, u16 *tc_size)
+{
+	u16 roundup_size;
+	u32 i;
+
+	roundup_size = roundup_pow_of_two(rss_size);
+	roundup_size = ilog2(roundup_size);
+
+	for (i = 0; i < HCLGE_COMM_MAX_TC_NUM; i++) {
+		tc_valid[i] = 1;
+		tc_size[i] = roundup_size;
+		tc_offset[i] = (hw_tc_map & BIT(i)) ? rss_size * i : 0;
+	}
+}
+EXPORT_SYMBOL_GPL(hclge_comm_get_rss_tc_info);
+
+int hclge_comm_set_rss_tc_mode(struct hclge_comm_hw *hw, u16 *tc_offset,
+			       u16 *tc_valid, u16 *tc_size)
+{
+	struct hclge_comm_rss_tc_mode_cmd *req;
+	struct hclge_desc desc;
+	unsigned int i;
+	int ret;
+
+	req = (struct hclge_comm_rss_tc_mode_cmd *)desc.data;
+
+	hclge_comm_cmd_setup_basic_desc(&desc, HCLGE_OPC_RSS_TC_MODE, false);
+	for (i = 0; i < HCLGE_COMM_MAX_TC_NUM; i++) {
+		u16 mode = 0;
+
+		hnae3_set_bit(mode, HCLGE_COMM_RSS_TC_VALID_B,
+			      (tc_valid[i] & 0x1));
+		hnae3_set_field(mode, HCLGE_COMM_RSS_TC_SIZE_M,
+				HCLGE_COMM_RSS_TC_SIZE_S, tc_size[i]);
+		hnae3_set_bit(mode, HCLGE_COMM_RSS_TC_SIZE_MSB_B,
+			      tc_size[i] >> HCLGE_COMM_RSS_TC_SIZE_MSB_OFFSET &
+			      0x1);
+		hnae3_set_field(mode, HCLGE_COMM_RSS_TC_OFFSET_M,
+				HCLGE_COMM_RSS_TC_OFFSET_S, tc_offset[i]);
+
+		req->rss_tc_mode[i] = cpu_to_le16(mode);
+	}
+
+	ret = hclge_comm_cmd_send(hw, &desc, 1);
+	if (ret)
+		dev_err(&hw->cmq.csq.pdev->dev,
+			"failed to set rss tc mode, ret = %d.\n", ret);
+
+	return ret;
+}
+EXPORT_SYMBOL_GPL(hclge_comm_set_rss_tc_mode);
+
+int hclge_comm_set_rss_hash_key(struct hclge_comm_rss_cfg *rss_cfg,
+				struct hclge_comm_hw *hw, const u8 *key,
+				const u8 hfunc)
+{
+	u8 hash_algo;
+	int ret;
+
+	ret = hclge_comm_parse_rss_hfunc(rss_cfg, hfunc, &hash_algo);
+	if (ret)
+		return ret;
+
+	/* Set the RSS Hash Key if specififed by the user */
+	if (key) {
+		ret = hclge_comm_set_rss_algo_key(hw, hash_algo, key);
+		if (ret)
+			return ret;
+
+		/* Update the shadow RSS key with user specified qids */
+		memcpy(rss_cfg->rss_hash_key, key, HCLGE_COMM_RSS_KEY_SIZE);
+	} else {
+		ret = hclge_comm_set_rss_algo_key(hw, hash_algo,
+						  rss_cfg->rss_hash_key);
+		if (ret)
+			return ret;
+	}
+	rss_cfg->rss_algo = hash_algo;
+
+	return 0;
+}
+EXPORT_SYMBOL_GPL(hclge_comm_set_rss_hash_key);
+
+int hclge_comm_set_rss_tuple(struct hnae3_ae_dev *ae_dev,
+			     struct hclge_comm_hw *hw,
+			     struct hclge_comm_rss_cfg *rss_cfg,
+			     const struct ethtool_rxfh_fields *nfc)
+{
+	struct hclge_comm_rss_input_tuple_cmd *req;
+	struct hclge_desc desc;
+	int ret;
+
+	if (nfc->data &
+	    ~(RXH_IP_SRC | RXH_IP_DST | RXH_L4_B_0_1 | RXH_L4_B_2_3))
+		return -EINVAL;
+
+	req = (struct hclge_comm_rss_input_tuple_cmd *)desc.data;
+	hclge_comm_cmd_setup_basic_desc(&desc, HCLGE_OPC_RSS_INPUT_TUPLE,
+					false);
+
+	ret = hclge_comm_init_rss_tuple_cmd(rss_cfg, nfc, ae_dev, req);
+	if (ret) {
+		dev_err(&hw->cmq.csq.pdev->dev,
+			"failed to init rss tuple cmd, ret = %d.\n", ret);
+		return ret;
+	}
+
+	ret = hclge_comm_cmd_send(hw, &desc, 1);
+	if (ret) {
+		dev_err(&hw->cmq.csq.pdev->dev,
+			"failed to set rss tuple, ret = %d.\n", ret);
+		return ret;
+	}
+
+	rss_cfg->rss_tuple_sets.ipv4_tcp_en = req->ipv4_tcp_en;
+	rss_cfg->rss_tuple_sets.ipv4_udp_en = req->ipv4_udp_en;
+	rss_cfg->rss_tuple_sets.ipv4_sctp_en = req->ipv4_sctp_en;
+	rss_cfg->rss_tuple_sets.ipv4_fragment_en = req->ipv4_fragment_en;
+	rss_cfg->rss_tuple_sets.ipv6_tcp_en = req->ipv6_tcp_en;
+	rss_cfg->rss_tuple_sets.ipv6_udp_en = req->ipv6_udp_en;
+	rss_cfg->rss_tuple_sets.ipv6_sctp_en = req->ipv6_sctp_en;
+	rss_cfg->rss_tuple_sets.ipv6_fragment_en = req->ipv6_fragment_en;
+	return 0;
+}
+EXPORT_SYMBOL_GPL(hclge_comm_set_rss_tuple);
+
+u32 hclge_comm_get_rss_key_size(struct hnae3_handle *handle)
+{
+	return HCLGE_COMM_RSS_KEY_SIZE;
+}
+EXPORT_SYMBOL_GPL(hclge_comm_get_rss_key_size);
+
+int hclge_comm_parse_rss_hfunc(struct hclge_comm_rss_cfg *rss_cfg,
+			       const u8 hfunc, u8 *hash_algo)
+{
+	switch (hfunc) {
+	case ETH_RSS_HASH_TOP:
+		*hash_algo = HCLGE_COMM_RSS_HASH_ALGO_TOEPLITZ;
+		return 0;
+	case ETH_RSS_HASH_XOR:
+		*hash_algo = HCLGE_COMM_RSS_HASH_ALGO_SIMPLE;
+		return 0;
+	case ETH_RSS_HASH_NO_CHANGE:
+		*hash_algo = rss_cfg->rss_algo;
+		return 0;
+	default:
+		return -EINVAL;
+	}
+}
+
+void hclge_comm_rss_indir_init_cfg(struct hnae3_ae_dev *ae_dev,
+				   struct hclge_comm_rss_cfg *rss_cfg)
+{
+	u16 i;
+	/* Initialize RSS indirect table */
+	for (i = 0; i < ae_dev->dev_specs.rss_ind_tbl_size; i++)
+		rss_cfg->rss_indirection_tbl[i] = i % rss_cfg->rss_size;
+}
+EXPORT_SYMBOL_GPL(hclge_comm_rss_indir_init_cfg);
+
+int hclge_comm_get_rss_tuple(struct hclge_comm_rss_cfg *rss_cfg, int flow_type,
+			     u8 *tuple_sets)
+{
+	switch (flow_type) {
+	case TCP_V4_FLOW:
+		*tuple_sets = rss_cfg->rss_tuple_sets.ipv4_tcp_en;
+		break;
+	case UDP_V4_FLOW:
+		*tuple_sets = rss_cfg->rss_tuple_sets.ipv4_udp_en;
+		break;
+	case TCP_V6_FLOW:
+		*tuple_sets = rss_cfg->rss_tuple_sets.ipv6_tcp_en;
+		break;
+	case UDP_V6_FLOW:
+		*tuple_sets = rss_cfg->rss_tuple_sets.ipv6_udp_en;
+		break;
+	case SCTP_V4_FLOW:
+		*tuple_sets = rss_cfg->rss_tuple_sets.ipv4_sctp_en;
+		break;
+	case SCTP_V6_FLOW:
+		*tuple_sets = rss_cfg->rss_tuple_sets.ipv6_sctp_en;
+		break;
+	case IPV4_FLOW:
+	case IPV6_FLOW:
+		*tuple_sets = HCLGE_COMM_S_IP_BIT | HCLGE_COMM_D_IP_BIT;
+		break;
+	default:
+		return -EINVAL;
+	}
+
+	return 0;
+}
+EXPORT_SYMBOL_GPL(hclge_comm_get_rss_tuple);
+
+static void
+hclge_comm_append_rss_msb_info(struct hclge_comm_rss_ind_tbl_cmd *req,
+			       u16 qid, u32 j)
+{
+	u8 rss_msb_oft;
+	u8 rss_msb_val;
+
+	rss_msb_oft =
+		j * HCLGE_COMM_RSS_CFG_TBL_BW_H / BITS_PER_BYTE;
+	rss_msb_val = (qid >> HCLGE_COMM_RSS_CFG_TBL_BW_L & 0x1) <<
+		(j * HCLGE_COMM_RSS_CFG_TBL_BW_H % BITS_PER_BYTE);
+	req->rss_qid_h[rss_msb_oft] |= rss_msb_val;
+}
+
+int hclge_comm_set_rss_indir_table(struct hnae3_ae_dev *ae_dev,
+				   struct hclge_comm_hw *hw, const u16 *indir)
+{
+	struct hclge_comm_rss_ind_tbl_cmd *req;
+	struct hclge_desc desc;
+	u16 rss_cfg_tbl_num;
+	int ret;
+	u16 qid;
+	u16 i;
+	u32 j;
+
+	req = (struct hclge_comm_rss_ind_tbl_cmd *)desc.data;
+	rss_cfg_tbl_num = ae_dev->dev_specs.rss_ind_tbl_size /
+			  HCLGE_COMM_RSS_CFG_TBL_SIZE;
+
+	for (i = 0; i < rss_cfg_tbl_num; i++) {
+		hclge_comm_cmd_setup_basic_desc(&desc,
+						HCLGE_OPC_RSS_INDIR_TABLE,
+						false);
+
+		req->start_table_index =
+			cpu_to_le16(i * HCLGE_COMM_RSS_CFG_TBL_SIZE);
+		req->rss_set_bitmap =
+			cpu_to_le16(HCLGE_COMM_RSS_SET_BITMAP_MSK);
+		for (j = 0; j < HCLGE_COMM_RSS_CFG_TBL_SIZE; j++) {
+			qid = indir[i * HCLGE_COMM_RSS_CFG_TBL_SIZE + j];
+			req->rss_qid_l[j] = qid & 0xff;
+			hclge_comm_append_rss_msb_info(req, qid, j);
+		}
+		ret = hclge_comm_cmd_send(hw, &desc, 1);
+		if (ret) {
+			dev_err(&hw->cmq.csq.pdev->dev,
+				"failed to configure rss table, ret = %d.\n",
+				ret);
+			return ret;
+		}
+	}
+	return 0;
+}
+EXPORT_SYMBOL_GPL(hclge_comm_set_rss_indir_table);
+
+int hclge_comm_set_rss_input_tuple(struct hclge_comm_hw *hw,
+				   struct hclge_comm_rss_cfg *rss_cfg)
+{
+	struct hclge_comm_rss_input_tuple_cmd *req;
+	struct hclge_desc desc;
+	int ret;
+
+	hclge_comm_cmd_setup_basic_desc(&desc, HCLGE_OPC_RSS_INPUT_TUPLE,
+					false);
+
+	req = (struct hclge_comm_rss_input_tuple_cmd *)desc.data;
+
+	req->ipv4_tcp_en = rss_cfg->rss_tuple_sets.ipv4_tcp_en;
+	req->ipv4_udp_en = rss_cfg->rss_tuple_sets.ipv4_udp_en;
+	req->ipv4_sctp_en = rss_cfg->rss_tuple_sets.ipv4_sctp_en;
+	req->ipv4_fragment_en = rss_cfg->rss_tuple_sets.ipv4_fragment_en;
+	req->ipv6_tcp_en = rss_cfg->rss_tuple_sets.ipv6_tcp_en;
+	req->ipv6_udp_en = rss_cfg->rss_tuple_sets.ipv6_udp_en;
+	req->ipv6_sctp_en = rss_cfg->rss_tuple_sets.ipv6_sctp_en;
+	req->ipv6_fragment_en = rss_cfg->rss_tuple_sets.ipv6_fragment_en;
+
+	ret = hclge_comm_cmd_send(hw, &desc, 1);
+	if (ret)
+		dev_err(&hw->cmq.csq.pdev->dev,
+			"failed to configure rss input, ret = %d.\n", ret);
+	return ret;
+}
+EXPORT_SYMBOL_GPL(hclge_comm_set_rss_input_tuple);
+
+void hclge_comm_get_rss_hash_info(struct hclge_comm_rss_cfg *rss_cfg, u8 *key,
+				  u8 *hfunc)
+{
+	/* Get hash algorithm */
+	if (hfunc) {
+		switch (rss_cfg->rss_algo) {
+		case HCLGE_COMM_RSS_HASH_ALGO_TOEPLITZ:
+			*hfunc = ETH_RSS_HASH_TOP;
+			break;
+		case HCLGE_COMM_RSS_HASH_ALGO_SIMPLE:
+			*hfunc = ETH_RSS_HASH_XOR;
+			break;
+		default:
+			*hfunc = ETH_RSS_HASH_UNKNOWN;
+			break;
+		}
+	}
+
+	/* Get the RSS Key required by the user */
+	if (key)
+		memcpy(key, rss_cfg->rss_hash_key, HCLGE_COMM_RSS_KEY_SIZE);
+}
+EXPORT_SYMBOL_GPL(hclge_comm_get_rss_hash_info);
+
+void hclge_comm_get_rss_indir_tbl(struct hclge_comm_rss_cfg *rss_cfg,
+				  u32 *indir, u16 rss_ind_tbl_size)
+{
+	u16 i;
+
+	if (!indir)
+		return;
+
+	for (i = 0; i < rss_ind_tbl_size; i++)
+		indir[i] = rss_cfg->rss_indirection_tbl[i];
+}
+EXPORT_SYMBOL_GPL(hclge_comm_get_rss_indir_tbl);
+
+int hclge_comm_set_rss_algo_key(struct hclge_comm_hw *hw, const u8 hfunc,
+				const u8 *key)
+{
+	struct hclge_comm_rss_config_cmd *req;
+	unsigned int key_offset = 0;
+	struct hclge_desc desc;
+	int key_counts;
+	int key_size;
+	int ret;
+
+	key_counts = HCLGE_COMM_RSS_KEY_SIZE;
+	req = (struct hclge_comm_rss_config_cmd *)desc.data;
+
+	while (key_counts) {
+		hclge_comm_cmd_setup_basic_desc(&desc,
+						HCLGE_OPC_RSS_GENERIC_CONFIG,
+						false);
+
+		req->hash_config |= (hfunc & HCLGE_COMM_RSS_HASH_ALGO_MASK);
+		req->hash_config |=
+			(key_offset << HCLGE_COMM_RSS_HASH_KEY_OFFSET_B);
+
+		key_size = min(HCLGE_COMM_RSS_HASH_KEY_NUM, key_counts);
+		memcpy(req->hash_key,
+		       key + key_offset * HCLGE_COMM_RSS_HASH_KEY_NUM,
+		       key_size);
+
+		key_counts -= key_size;
+		key_offset++;
+		ret = hclge_comm_cmd_send(hw, &desc, 1);
+		if (ret) {
+			dev_err(&hw->cmq.csq.pdev->dev,
+				"failed to configure RSS key, ret = %d.\n",
+				ret);
+			return ret;
+		}
+	}
+
+	return 0;
+}
+EXPORT_SYMBOL_GPL(hclge_comm_set_rss_algo_key);
+
+static u8 hclge_comm_get_rss_hash_bits(const struct ethtool_rxfh_fields *nfc)
+{
+	u8 hash_sets = nfc->data & RXH_L4_B_0_1 ? HCLGE_COMM_S_PORT_BIT : 0;
+
+	if (nfc->data & RXH_L4_B_2_3)
+		hash_sets |= HCLGE_COMM_D_PORT_BIT;
+	else
+		hash_sets &= ~HCLGE_COMM_D_PORT_BIT;
+
+	if (nfc->data & RXH_IP_SRC)
+		hash_sets |= HCLGE_COMM_S_IP_BIT;
+	else
+		hash_sets &= ~HCLGE_COMM_S_IP_BIT;
+
+	if (nfc->data & RXH_IP_DST)
+		hash_sets |= HCLGE_COMM_D_IP_BIT;
+	else
+		hash_sets &= ~HCLGE_COMM_D_IP_BIT;
+
+	if (nfc->flow_type == SCTP_V4_FLOW || nfc->flow_type == SCTP_V6_FLOW)
+		hash_sets |= HCLGE_COMM_V_TAG_BIT;
+
+	return hash_sets;
+}
+
+int hclge_comm_init_rss_tuple_cmd(struct hclge_comm_rss_cfg *rss_cfg,
+				  const struct ethtool_rxfh_fields *nfc,
+				  struct hnae3_ae_dev *ae_dev,
+				  struct hclge_comm_rss_input_tuple_cmd *req)
+{
+	u8 tuple_sets;
+
+	req->ipv4_tcp_en = rss_cfg->rss_tuple_sets.ipv4_tcp_en;
+	req->ipv4_udp_en = rss_cfg->rss_tuple_sets.ipv4_udp_en;
+	req->ipv4_sctp_en = rss_cfg->rss_tuple_sets.ipv4_sctp_en;
+	req->ipv4_fragment_en = rss_cfg->rss_tuple_sets.ipv4_fragment_en;
+	req->ipv6_tcp_en = rss_cfg->rss_tuple_sets.ipv6_tcp_en;
+	req->ipv6_udp_en = rss_cfg->rss_tuple_sets.ipv6_udp_en;
+	req->ipv6_sctp_en = rss_cfg->rss_tuple_sets.ipv6_sctp_en;
+	req->ipv6_fragment_en = rss_cfg->rss_tuple_sets.ipv6_fragment_en;
+
+	tuple_sets = hclge_comm_get_rss_hash_bits(nfc);
+	switch (nfc->flow_type) {
+	case TCP_V4_FLOW:
+		req->ipv4_tcp_en = tuple_sets;
+		break;
+	case TCP_V6_FLOW:
+		req->ipv6_tcp_en = tuple_sets;
+		break;
+	case UDP_V4_FLOW:
+		req->ipv4_udp_en = tuple_sets;
+		break;
+	case UDP_V6_FLOW:
+		req->ipv6_udp_en = tuple_sets;
+		break;
+	case SCTP_V4_FLOW:
+		req->ipv4_sctp_en = tuple_sets;
+		break;
+	case SCTP_V6_FLOW:
+		if (ae_dev->dev_version <= HNAE3_DEVICE_VERSION_V2 &&
+		    (nfc->data & (RXH_L4_B_0_1 | RXH_L4_B_2_3)))
+			return -EINVAL;
+
+		req->ipv6_sctp_en = tuple_sets;
+		break;
+	case IPV4_FLOW:
+		req->ipv4_fragment_en = HCLGE_COMM_RSS_INPUT_TUPLE_OTHER;
+		break;
+	case IPV6_FLOW:
+		req->ipv6_fragment_en = HCLGE_COMM_RSS_INPUT_TUPLE_OTHER;
+		break;
+	default:
+		return -EINVAL;
+	}
+
+	return 0;
+}
+
+u64 hclge_comm_convert_rss_tuple(u8 tuple_sets)
+{
+	u64 tuple_data = 0;
+
+	if (tuple_sets & HCLGE_COMM_D_PORT_BIT)
+		tuple_data |= RXH_L4_B_2_3;
+	if (tuple_sets & HCLGE_COMM_S_PORT_BIT)
+		tuple_data |= RXH_L4_B_0_1;
+	if (tuple_sets & HCLGE_COMM_D_IP_BIT)
+		tuple_data |= RXH_IP_DST;
+	if (tuple_sets & HCLGE_COMM_S_IP_BIT)
+		tuple_data |= RXH_IP_SRC;
+
+	return tuple_data;
+}
+EXPORT_SYMBOL_GPL(hclge_comm_convert_rss_tuple);
diff --git a/drivers/net/ethernet/hisilicon/hns3/hns3_common/hclge_comm_rss.h b/drivers/net/ethernet/hisilicon/hns3/hns3_common/hclge_comm_rss.h
new file mode 100644
index 000000000000..cbc02b50c6e7
--- /dev/null
+++ b/drivers/net/ethernet/hisilicon/hns3/hns3_common/hclge_comm_rss.h
@@ -0,0 +1,133 @@
+/* SPDX-License-Identifier: GPL-2.0+ */
+// Copyright (c) 2021-2021 Hisilicon Limited.
+
+#ifndef __HCLGE_COMM_RSS_H
+#define __HCLGE_COMM_RSS_H
+#include <linux/types.h>
+
+#include "hnae3.h"
+#include "hclge_comm_cmd.h"
+
+#define HCLGE_COMM_RSS_HASH_ALGO_TOEPLITZ	0
+#define HCLGE_COMM_RSS_HASH_ALGO_SIMPLE		1
+#define HCLGE_COMM_RSS_HASH_ALGO_SYMMETRIC	2
+
+#define HCLGE_COMM_RSS_INPUT_TUPLE_OTHER	GENMASK(3, 0)
+#define HCLGE_COMM_RSS_INPUT_TUPLE_SCTP		GENMASK(4, 0)
+
+#define HCLGE_COMM_D_PORT_BIT		BIT(0)
+#define HCLGE_COMM_S_PORT_BIT		BIT(1)
+#define HCLGE_COMM_D_IP_BIT		BIT(2)
+#define HCLGE_COMM_S_IP_BIT		BIT(3)
+#define HCLGE_COMM_V_TAG_BIT		BIT(4)
+#define HCLGE_COMM_RSS_INPUT_TUPLE_SCTP_NO_PORT	\
+	(HCLGE_COMM_D_IP_BIT | HCLGE_COMM_S_IP_BIT | HCLGE_COMM_V_TAG_BIT)
+#define HCLGE_COMM_MAX_TC_NUM		8
+
+#define HCLGE_COMM_RSS_TC_OFFSET_S		0
+#define HCLGE_COMM_RSS_TC_OFFSET_M		GENMASK(10, 0)
+#define HCLGE_COMM_RSS_TC_SIZE_MSB_B	11
+#define HCLGE_COMM_RSS_TC_SIZE_S		12
+#define HCLGE_COMM_RSS_TC_SIZE_M		GENMASK(14, 12)
+#define HCLGE_COMM_RSS_TC_VALID_B		15
+#define HCLGE_COMM_RSS_TC_SIZE_MSB_OFFSET	3
+
+struct hclge_comm_rss_tuple_cfg {
+	u8 ipv4_tcp_en;
+	u8 ipv4_udp_en;
+	u8 ipv4_sctp_en;
+	u8 ipv4_fragment_en;
+	u8 ipv6_tcp_en;
+	u8 ipv6_udp_en;
+	u8 ipv6_sctp_en;
+	u8 ipv6_fragment_en;
+};
+
+#define HCLGE_COMM_RSS_KEY_SIZE		40
+#define HCLGE_COMM_RSS_CFG_TBL_SIZE	16
+#define HCLGE_COMM_RSS_CFG_TBL_BW_H	2U
+#define HCLGE_COMM_RSS_CFG_TBL_BW_L	8U
+#define HCLGE_COMM_RSS_CFG_TBL_SIZE_H	4
+#define HCLGE_COMM_RSS_SET_BITMAP_MSK	GENMASK(15, 0)
+#define HCLGE_COMM_RSS_HASH_ALGO_MASK	GENMASK(3, 0)
+#define HCLGE_COMM_RSS_HASH_KEY_OFFSET_B	4
+
+#define HCLGE_COMM_RSS_HASH_KEY_NUM	16
+struct hclge_comm_rss_config_cmd {
+	u8 hash_config;
+	u8 rsv[7];
+	u8 hash_key[HCLGE_COMM_RSS_HASH_KEY_NUM];
+};
+
+struct hclge_comm_rss_cfg {
+	u8 rss_hash_key[HCLGE_COMM_RSS_KEY_SIZE]; /* user configured hash keys */
+
+	/* shadow table */
+	u16 *rss_indirection_tbl;
+	u32 rss_algo;
+
+	struct hclge_comm_rss_tuple_cfg rss_tuple_sets;
+	u32 rss_size;
+};
+
+struct hclge_comm_rss_input_tuple_cmd {
+	u8 ipv4_tcp_en;
+	u8 ipv4_udp_en;
+	u8 ipv4_sctp_en;
+	u8 ipv4_fragment_en;
+	u8 ipv6_tcp_en;
+	u8 ipv6_udp_en;
+	u8 ipv6_sctp_en;
+	u8 ipv6_fragment_en;
+	u8 rsv[16];
+};
+
+struct hclge_comm_rss_ind_tbl_cmd {
+	__le16 start_table_index;
+	__le16 rss_set_bitmap;
+	u8 rss_qid_h[HCLGE_COMM_RSS_CFG_TBL_SIZE_H];
+	u8 rss_qid_l[HCLGE_COMM_RSS_CFG_TBL_SIZE];
+};
+
+struct hclge_comm_rss_tc_mode_cmd {
+	__le16 rss_tc_mode[HCLGE_COMM_MAX_TC_NUM];
+	u8 rsv[8];
+};
+
+u32 hclge_comm_get_rss_key_size(struct hnae3_handle *handle);
+void hclge_comm_rss_indir_init_cfg(struct hnae3_ae_dev *ae_dev,
+				   struct hclge_comm_rss_cfg *rss_cfg);
+int hclge_comm_get_rss_tuple(struct hclge_comm_rss_cfg *rss_cfg, int flow_type,
+			     u8 *tuple_sets);
+int hclge_comm_parse_rss_hfunc(struct hclge_comm_rss_cfg *rss_cfg,
+			       const u8 hfunc, u8 *hash_algo);
+void hclge_comm_get_rss_hash_info(struct hclge_comm_rss_cfg *rss_cfg, u8 *key,
+				  u8 *hfunc);
+void hclge_comm_get_rss_indir_tbl(struct hclge_comm_rss_cfg *rss_cfg,
+				  u32 *indir, u16 rss_ind_tbl_size);
+int hclge_comm_set_rss_algo_key(struct hclge_comm_hw *hw, const u8 hfunc,
+				const u8 *key);
+int hclge_comm_init_rss_tuple_cmd(struct hclge_comm_rss_cfg *rss_cfg,
+				  const struct ethtool_rxfh_fields *nfc,
+				  struct hnae3_ae_dev *ae_dev,
+				  struct hclge_comm_rss_input_tuple_cmd *req);
+u64 hclge_comm_convert_rss_tuple(u8 tuple_sets);
+int hclge_comm_set_rss_input_tuple(struct hclge_comm_hw *hw,
+				   struct hclge_comm_rss_cfg *rss_cfg);
+int hclge_comm_set_rss_indir_table(struct hnae3_ae_dev *ae_dev,
+				   struct hclge_comm_hw *hw, const u16 *indir);
+int hclge_comm_rss_init_cfg(struct hnae3_handle *nic,
+			    struct hnae3_ae_dev *ae_dev,
+			    struct hclge_comm_rss_cfg *rss_cfg);
+void hclge_comm_get_rss_tc_info(u16 rss_size, u8 hw_tc_map, u16 *tc_offset,
+				u16 *tc_valid, u16 *tc_size);
+int hclge_comm_set_rss_tc_mode(struct hclge_comm_hw *hw, u16 *tc_offset,
+			       u16 *tc_valid, u16 *tc_size);
+int hclge_comm_set_rss_hash_key(struct hclge_comm_rss_cfg *rss_cfg,
+				struct hclge_comm_hw *hw, const u8 *key,
+				const u8 hfunc);
+int hclge_comm_set_rss_tuple(struct hnae3_ae_dev *ae_dev,
+			     struct hclge_comm_hw *hw,
+			     struct hclge_comm_rss_cfg *rss_cfg,
+			     const struct ethtool_rxfh_fields *nfc);
+#endif
diff --git a/drivers/net/ethernet/hisilicon/hns3/hns3_common/hclge_comm_tqp_stats.c b/drivers/net/ethernet/hisilicon/hns3/hns3_common/hclge_comm_tqp_stats.c
new file mode 100644
index 000000000000..f9a3d6fc4416
--- /dev/null
+++ b/drivers/net/ethernet/hisilicon/hns3/hns3_common/hclge_comm_tqp_stats.c
@@ -0,0 +1,115 @@
+// SPDX-License-Identifier: GPL-2.0+
+// Copyright (c) 2021-2021 Hisilicon Limited.
+
+#include <linux/err.h>
+
+#include "hnae3.h"
+#include "hclge_comm_cmd.h"
+#include "hclge_comm_tqp_stats.h"
+
+u64 *hclge_comm_tqps_get_stats(struct hnae3_handle *handle, u64 *data)
+{
+	struct hnae3_knic_private_info *kinfo = &handle->kinfo;
+	struct hclge_comm_tqp *tqp;
+	u64 *buff = data;
+	u16 i;
+
+	for (i = 0; i < kinfo->num_tqps; i++) {
+		tqp = container_of(kinfo->tqp[i], struct hclge_comm_tqp, q);
+		*buff++ = tqp->tqp_stats.rcb_tx_ring_pktnum_rcd;
+	}
+
+	for (i = 0; i < kinfo->num_tqps; i++) {
+		tqp = container_of(kinfo->tqp[i], struct hclge_comm_tqp, q);
+		*buff++ = tqp->tqp_stats.rcb_rx_ring_pktnum_rcd;
+	}
+
+	return buff;
+}
+EXPORT_SYMBOL_GPL(hclge_comm_tqps_get_stats);
+
+int hclge_comm_tqps_get_sset_count(struct hnae3_handle *handle)
+{
+	struct hnae3_knic_private_info *kinfo = &handle->kinfo;
+
+	return kinfo->num_tqps * HCLGE_COMM_QUEUE_PAIR_SIZE;
+}
+EXPORT_SYMBOL_GPL(hclge_comm_tqps_get_sset_count);
+
+void hclge_comm_tqps_get_strings(struct hnae3_handle *handle, u8 **data)
+{
+	struct hnae3_knic_private_info *kinfo = &handle->kinfo;
+	u16 i;
+
+	for (i = 0; i < kinfo->num_tqps; i++) {
+		struct hclge_comm_tqp *tqp =
+			container_of(kinfo->tqp[i], struct hclge_comm_tqp, q);
+		ethtool_sprintf(data, "txq%u_pktnum_rcd", tqp->index);
+	}
+
+	for (i = 0; i < kinfo->num_tqps; i++) {
+		struct hclge_comm_tqp *tqp =
+			container_of(kinfo->tqp[i], struct hclge_comm_tqp, q);
+		ethtool_sprintf(data, "rxq%u_pktnum_rcd", tqp->index);
+	}
+}
+EXPORT_SYMBOL_GPL(hclge_comm_tqps_get_strings);
+
+int hclge_comm_tqps_update_stats(struct hnae3_handle *handle,
+				 struct hclge_comm_hw *hw)
+{
+	struct hnae3_knic_private_info *kinfo = &handle->kinfo;
+	struct hclge_comm_tqp *tqp;
+	struct hclge_desc desc;
+	int ret;
+	u16 i;
+
+	for (i = 0; i < kinfo->num_tqps; i++) {
+		tqp = container_of(kinfo->tqp[i], struct hclge_comm_tqp, q);
+		hclge_comm_cmd_setup_basic_desc(&desc, HCLGE_OPC_QUERY_RX_STATS,
+						true);
+
+		desc.data[0] = cpu_to_le32(tqp->index);
+		ret = hclge_comm_cmd_send(hw, &desc, 1);
+		if (ret) {
+			dev_err(&hw->cmq.csq.pdev->dev,
+				"failed to get tqp stat, ret = %d, rx = %u.\n",
+				ret, i);
+			return ret;
+		}
+		tqp->tqp_stats.rcb_rx_ring_pktnum_rcd +=
+			le32_to_cpu(desc.data[1]);
+
+		hclge_comm_cmd_setup_basic_desc(&desc, HCLGE_OPC_QUERY_TX_STATS,
+						true);
+
+		desc.data[0] = cpu_to_le32(tqp->index);
+		ret = hclge_comm_cmd_send(hw, &desc, 1);
+		if (ret) {
+			dev_err(&hw->cmq.csq.pdev->dev,
+				"failed to get tqp stat, ret = %d, tx = %u.\n",
+				ret, i);
+			return ret;
+		}
+		tqp->tqp_stats.rcb_tx_ring_pktnum_rcd +=
+			le32_to_cpu(desc.data[1]);
+	}
+
+	return 0;
+}
+EXPORT_SYMBOL_GPL(hclge_comm_tqps_update_stats);
+
+void hclge_comm_reset_tqp_stats(struct hnae3_handle *handle)
+{
+	struct hnae3_knic_private_info *kinfo = &handle->kinfo;
+	struct hclge_comm_tqp *tqp;
+	struct hnae3_queue *queue;
+	u16 i;
+
+	for (i = 0; i < kinfo->num_tqps; i++) {
+		queue = kinfo->tqp[i];
+		tqp = container_of(queue, struct hclge_comm_tqp, q);
+		memset(&tqp->tqp_stats, 0, sizeof(tqp->tqp_stats));
+	}
+}
+EXPORT_SYMBOL_GPL(hclge_comm_reset_tqp_stats);
diff --git a/drivers/net/ethernet/hisilicon/hns3/hns3_common/hclge_comm_tqp_stats.h b/drivers/net/ethernet/hisilicon/hns3/hns3_common/hclge_comm_tqp_stats.h
new file mode 100644
index 000000000000..b9ff424c0bc2
--- /dev/null
+++ b/drivers/net/ethernet/hisilicon/hns3/hns3_common/hclge_comm_tqp_stats.h
@@ -0,0 +1,39 @@
+/* SPDX-License-Identifier: GPL-2.0+ */
+// Copyright (c) 2021-2021 Hisilicon Limited.
+
+#ifndef __HCLGE_COMM_TQP_STATS_H
+#define __HCLGE_COMM_TQP_STATS_H
+#include <linux/types.h>
+#include <linux/etherdevice.h>
+#include "hnae3.h"
+
+/* each tqp has TX & RX two queues */
+#define HCLGE_COMM_QUEUE_PAIR_SIZE 2
+
+/* TQP stats */
+struct hclge_comm_tqp_stats {
+	/* query_tqp_tx_queue_statistics ,opcode id:  0x0B03 */
+	u64 rcb_tx_ring_pktnum_rcd; /* 32bit */
+	/* query_tqp_rx_queue_statistics ,opcode id:  0x0B13 */
+	u64 rcb_rx_ring_pktnum_rcd; /* 32bit */
+};
+
+struct hclge_comm_tqp {
+	/* copy of device pointer from pci_dev,
+	 * used when perform DMA mapping
+	 */
+	struct device *dev;
+	struct hnae3_queue q;
+	struct hclge_comm_tqp_stats tqp_stats;
+	u16 index;	/* Global index in a NIC controller */
+
+	bool alloced;
+};
+
+u64 *hclge_comm_tqps_get_stats(struct hnae3_handle *handle, u64 *data);
+int hclge_comm_tqps_get_sset_count(struct hnae3_handle *handle);
+void hclge_comm_tqps_get_strings(struct hnae3_handle *handle, u8 **data);
+void hclge_comm_reset_tqp_stats(struct hnae3_handle *handle);
+int hclge_comm_tqps_update_stats(struct hnae3_handle *handle,
+				 struct hclge_comm_hw *hw);
+#endif
diff --git a/drivers/net/ethernet/hisilicon/hns3/hns3_dcbnl.c b/drivers/net/ethernet/hisilicon/hns3/hns3_dcbnl.c
new file mode 100644
index 000000000000..f72dc0cee30e
--- /dev/null
+++ b/drivers/net/ethernet/hisilicon/hns3/hns3_dcbnl.c
@@ -0,0 +1,130 @@
+// SPDX-License-Identifier: GPL-2.0+
+// Copyright (c) 2016-2017 Hisilicon Limited.
+
+#include "hnae3.h"
+#include "hns3_enet.h"
+
+static int hns3_dcbnl_ieee_getets(struct net_device *ndev, struct ieee_ets *ets)
+{
+	struct hnae3_handle *h = hns3_get_handle(ndev);
+
+	if (hns3_nic_resetting(ndev))
+		return -EBUSY;
+
+	if (h->kinfo.dcb_ops->ieee_getets)
+		return h->kinfo.dcb_ops->ieee_getets(h, ets);
+
+	return -EOPNOTSUPP;
+}
+
+static int hns3_dcbnl_ieee_setets(struct net_device *ndev, struct ieee_ets *ets)
+{
+	struct hnae3_handle *h = hns3_get_handle(ndev);
+
+	if (hns3_nic_resetting(ndev))
+		return -EBUSY;
+
+	if (h->kinfo.dcb_ops->ieee_setets)
+		return h->kinfo.dcb_ops->ieee_setets(h, ets);
+
+	return -EOPNOTSUPP;
+}
+
+static int hns3_dcbnl_ieee_getpfc(struct net_device *ndev, struct ieee_pfc *pfc)
+{
+	struct hnae3_handle *h = hns3_get_handle(ndev);
+
+	if (hns3_nic_resetting(ndev))
+		return -EBUSY;
+
+	if (h->kinfo.dcb_ops->ieee_getpfc)
+		return h->kinfo.dcb_ops->ieee_getpfc(h, pfc);
+
+	return -EOPNOTSUPP;
+}
+
+static int hns3_dcbnl_ieee_setpfc(struct net_device *ndev, struct ieee_pfc *pfc)
+{
+	struct hnae3_handle *h = hns3_get_handle(ndev);
+
+	if (hns3_nic_resetting(ndev))
+		return -EBUSY;
+
+	if (h->kinfo.dcb_ops->ieee_setpfc)
+		return h->kinfo.dcb_ops->ieee_setpfc(h, pfc);
+
+	return -EOPNOTSUPP;
+}
+
+static int hns3_dcbnl_ieee_setapp(struct net_device *ndev, struct dcb_app *app)
+{
+	struct hnae3_handle *h = hns3_get_handle(ndev);
+
+	if (hns3_nic_resetting(ndev))
+		return -EBUSY;
+
+	if (h->kinfo.dcb_ops->ieee_setapp)
+		return h->kinfo.dcb_ops->ieee_setapp(h, app);
+
+	return -EOPNOTSUPP;
+}
+
+static int hns3_dcbnl_ieee_delapp(struct net_device *ndev, struct dcb_app *app)
+{
+	struct hnae3_handle *h = hns3_get_handle(ndev);
+
+	if (hns3_nic_resetting(ndev))
+		return -EBUSY;
+
+	if (h->kinfo.dcb_ops->ieee_delapp)
+		return h->kinfo.dcb_ops->ieee_delapp(h, app);
+
+	return -EOPNOTSUPP;
+}
+
+/* DCBX configuration */
+static u8 hns3_dcbnl_getdcbx(struct net_device *ndev)
+{
+	struct hnae3_handle *h = hns3_get_handle(ndev);
+
+	if (h->kinfo.dcb_ops->getdcbx)
+		return h->kinfo.dcb_ops->getdcbx(h);
+
+	return 0;
+}
+
+/* return 0 if successful, otherwise fail */
+static u8 hns3_dcbnl_setdcbx(struct net_device *ndev, u8 mode)
+{
+	struct hnae3_handle *h = hns3_get_handle(ndev);
+
+	if (h->kinfo.dcb_ops->setdcbx)
+		return h->kinfo.dcb_ops->setdcbx(h, mode);
+
+	return 1;
+}
+
+static const struct dcbnl_rtnl_ops hns3_dcbnl_ops = {
+	.ieee_getets	= hns3_dcbnl_ieee_getets,
+	.ieee_setets	= hns3_dcbnl_ieee_setets,
+	.ieee_getpfc	= hns3_dcbnl_ieee_getpfc,
+	.ieee_setpfc	= hns3_dcbnl_ieee_setpfc,
+	.ieee_setapp    = hns3_dcbnl_ieee_setapp,
+	.ieee_delapp    = hns3_dcbnl_ieee_delapp,
+	.getdcbx	= hns3_dcbnl_getdcbx,
+	.setdcbx	= hns3_dcbnl_setdcbx,
+};
+
+/* hclge_dcbnl_setup - DCBNL setup
+ * @handle: the corresponding vport handle
+ * Set up DCBNL
+ */
+void hns3_dcbnl_setup(struct hnae3_handle *handle)
+{
+	struct net_device *dev = handle->kinfo.netdev;
+
+	if ((!handle->kinfo.dcb_ops) || (handle->flags & HNAE3_SUPPORT_VF))
+		return;
+
+	dev->dcbnl_ops = &hns3_dcbnl_ops;
+}
diff --git a/drivers/net/ethernet/hisilicon/hns3/hns3_debugfs.c b/drivers/net/ethernet/hisilicon/hns3/hns3_debugfs.c
new file mode 100644
index 000000000000..4cce4f4ba6b0
--- /dev/null
+++ b/drivers/net/ethernet/hisilicon/hns3/hns3_debugfs.c
@@ -0,0 +1,985 @@
+// SPDX-License-Identifier: GPL-2.0+
+/* Copyright (c) 2018-2019 Hisilicon Limited. */
+
+#include <linux/debugfs.h>
+#include <linux/device.h>
+#include <linux/seq_file.h>
+#include <linux/string_choices.h>
+
+#include "hnae3.h"
+#include "hns3_debugfs.h"
+#include "hns3_enet.h"
+
+static struct dentry *hns3_dbgfs_root;
+
+static struct hns3_dbg_dentry_info hns3_dbg_dentry[] = {
+	{
+		.name = "tm"
+	},
+	{
+		.name = "tx_bd_info"
+	},
+	{
+		.name = "rx_bd_info"
+	},
+	{
+		.name = "mac_list"
+	},
+	{
+		.name = "reg"
+	},
+	{
+		.name = "queue"
+	},
+	{
+		.name = "fd"
+	},
+	/* keep common at the bottom and add new directory above */
+	{
+		.name = "common"
+	},
+};
+
+static int hns3_dbg_bd_file_init(struct hnae3_handle *handle, u32 cmd);
+static int hns3_dbg_common_init_t1(struct hnae3_handle *handle, u32 cmd);
+static int hns3_dbg_common_init_t2(struct hnae3_handle *handle, u32 cmd);
+
+static struct hns3_dbg_cmd_info hns3_dbg_cmd[] = {
+	{
+		.name = "tm_nodes",
+		.cmd = HNAE3_DBG_CMD_TM_NODES,
+		.dentry = HNS3_DBG_DENTRY_TM,
+		.init = hns3_dbg_common_init_t2,
+	},
+	{
+		.name = "tm_priority",
+		.cmd = HNAE3_DBG_CMD_TM_PRI,
+		.dentry = HNS3_DBG_DENTRY_TM,
+		.init = hns3_dbg_common_init_t2,
+	},
+	{
+		.name = "tm_qset",
+		.cmd = HNAE3_DBG_CMD_TM_QSET,
+		.dentry = HNS3_DBG_DENTRY_TM,
+		.init = hns3_dbg_common_init_t2,
+	},
+	{
+		.name = "tm_map",
+		.cmd = HNAE3_DBG_CMD_TM_MAP,
+		.dentry = HNS3_DBG_DENTRY_TM,
+		.init = hns3_dbg_common_init_t2,
+	},
+	{
+		.name = "tm_pg",
+		.cmd = HNAE3_DBG_CMD_TM_PG,
+		.dentry = HNS3_DBG_DENTRY_TM,
+		.init = hns3_dbg_common_init_t2,
+	},
+	{
+		.name = "tm_port",
+		.cmd = HNAE3_DBG_CMD_TM_PORT,
+		.dentry = HNS3_DBG_DENTRY_TM,
+		.init = hns3_dbg_common_init_t2,
+	},
+	{
+		.name = "tc_sch_info",
+		.cmd = HNAE3_DBG_CMD_TC_SCH_INFO,
+		.dentry = HNS3_DBG_DENTRY_TM,
+		.init = hns3_dbg_common_init_t2,
+	},
+	{
+		.name = "qos_pause_cfg",
+		.cmd = HNAE3_DBG_CMD_QOS_PAUSE_CFG,
+		.dentry = HNS3_DBG_DENTRY_TM,
+		.init = hns3_dbg_common_init_t2,
+	},
+	{
+		.name = "qos_pri_map",
+		.cmd = HNAE3_DBG_CMD_QOS_PRI_MAP,
+		.dentry = HNS3_DBG_DENTRY_TM,
+		.init = hns3_dbg_common_init_t2,
+	},
+	{
+		.name = "qos_dscp_map",
+		.cmd = HNAE3_DBG_CMD_QOS_DSCP_MAP,
+		.dentry = HNS3_DBG_DENTRY_TM,
+		.init = hns3_dbg_common_init_t2,
+	},
+	{
+		.name = "qos_buf_cfg",
+		.cmd = HNAE3_DBG_CMD_QOS_BUF_CFG,
+		.dentry = HNS3_DBG_DENTRY_TM,
+		.init = hns3_dbg_common_init_t2,
+	},
+	{
+		.name = "dev_info",
+		.cmd = HNAE3_DBG_CMD_DEV_INFO,
+		.dentry = HNS3_DBG_DENTRY_COMMON,
+		.init = hns3_dbg_common_init_t1,
+	},
+	{
+		.name = "tx_bd_queue",
+		.cmd = HNAE3_DBG_CMD_TX_BD,
+		.dentry = HNS3_DBG_DENTRY_TX_BD,
+		.init = hns3_dbg_bd_file_init,
+	},
+	{
+		.name = "rx_bd_queue",
+		.cmd = HNAE3_DBG_CMD_RX_BD,
+		.dentry = HNS3_DBG_DENTRY_RX_BD,
+		.init = hns3_dbg_bd_file_init,
+	},
+	{
+		.name = "uc",
+		.cmd = HNAE3_DBG_CMD_MAC_UC,
+		.dentry = HNS3_DBG_DENTRY_MAC,
+		.init = hns3_dbg_common_init_t2,
+	},
+	{
+		.name = "mc",
+		.cmd = HNAE3_DBG_CMD_MAC_MC,
+		.dentry = HNS3_DBG_DENTRY_MAC,
+		.init = hns3_dbg_common_init_t2,
+	},
+	{
+		.name = "mng_tbl",
+		.cmd = HNAE3_DBG_CMD_MNG_TBL,
+		.dentry = HNS3_DBG_DENTRY_COMMON,
+		.init = hns3_dbg_common_init_t2,
+	},
+	{
+		.name = "loopback",
+		.cmd = HNAE3_DBG_CMD_LOOPBACK,
+		.dentry = HNS3_DBG_DENTRY_COMMON,
+		.init = hns3_dbg_common_init_t2,
+	},
+	{
+		.name = "interrupt_info",
+		.cmd = HNAE3_DBG_CMD_INTERRUPT_INFO,
+		.dentry = HNS3_DBG_DENTRY_COMMON,
+		.init = hns3_dbg_common_init_t2,
+	},
+	{
+		.name = "reset_info",
+		.cmd = HNAE3_DBG_CMD_RESET_INFO,
+		.dentry = HNS3_DBG_DENTRY_COMMON,
+		.init = hns3_dbg_common_init_t2,
+	},
+	{
+		.name = "imp_info",
+		.cmd = HNAE3_DBG_CMD_IMP_INFO,
+		.dentry = HNS3_DBG_DENTRY_COMMON,
+		.init = hns3_dbg_common_init_t2,
+	},
+	{
+		.name = "ncl_config",
+		.cmd = HNAE3_DBG_CMD_NCL_CONFIG,
+		.dentry = HNS3_DBG_DENTRY_COMMON,
+		.init = hns3_dbg_common_init_t2,
+	},
+	{
+		.name = "mac_tnl_status",
+		.cmd = HNAE3_DBG_CMD_MAC_TNL_STATUS,
+		.dentry = HNS3_DBG_DENTRY_COMMON,
+		.init = hns3_dbg_common_init_t2,
+	},
+	{
+		.name = "bios_common",
+		.cmd = HNAE3_DBG_CMD_REG_BIOS_COMMON,
+		.dentry = HNS3_DBG_DENTRY_REG,
+		.init = hns3_dbg_common_init_t2,
+	},
+	{
+		.name = "ssu",
+		.cmd = HNAE3_DBG_CMD_REG_SSU,
+		.dentry = HNS3_DBG_DENTRY_REG,
+		.init = hns3_dbg_common_init_t2,
+	},
+	{
+		.name = "igu_egu",
+		.cmd = HNAE3_DBG_CMD_REG_IGU_EGU,
+		.dentry = HNS3_DBG_DENTRY_REG,
+		.init = hns3_dbg_common_init_t2,
+	},
+	{
+		.name = "rpu",
+		.cmd = HNAE3_DBG_CMD_REG_RPU,
+		.dentry = HNS3_DBG_DENTRY_REG,
+		.init = hns3_dbg_common_init_t2,
+	},
+	{
+		.name = "ncsi",
+		.cmd = HNAE3_DBG_CMD_REG_NCSI,
+		.dentry = HNS3_DBG_DENTRY_REG,
+		.init = hns3_dbg_common_init_t2,
+	},
+	{
+		.name = "rtc",
+		.cmd = HNAE3_DBG_CMD_REG_RTC,
+		.dentry = HNS3_DBG_DENTRY_REG,
+		.init = hns3_dbg_common_init_t2,
+	},
+	{
+		.name = "ppp",
+		.cmd = HNAE3_DBG_CMD_REG_PPP,
+		.dentry = HNS3_DBG_DENTRY_REG,
+		.init = hns3_dbg_common_init_t2,
+	},
+	{
+		.name = "rcb",
+		.cmd = HNAE3_DBG_CMD_REG_RCB,
+		.dentry = HNS3_DBG_DENTRY_REG,
+		.init = hns3_dbg_common_init_t2,
+	},
+	{
+		.name = "tqp",
+		.cmd = HNAE3_DBG_CMD_REG_TQP,
+		.dentry = HNS3_DBG_DENTRY_REG,
+		.init = hns3_dbg_common_init_t2,
+	},
+	{
+		.name = "mac",
+		.cmd = HNAE3_DBG_CMD_REG_MAC,
+		.dentry = HNS3_DBG_DENTRY_REG,
+		.init = hns3_dbg_common_init_t2,
+	},
+	{
+		.name = "dcb",
+		.cmd = HNAE3_DBG_CMD_REG_DCB,
+		.dentry = HNS3_DBG_DENTRY_REG,
+		.init = hns3_dbg_common_init_t2,
+	},
+	{
+		.name = "queue_map",
+		.cmd = HNAE3_DBG_CMD_QUEUE_MAP,
+		.dentry = HNS3_DBG_DENTRY_QUEUE,
+		.init = hns3_dbg_common_init_t1,
+	},
+	{
+		.name = "rx_queue_info",
+		.cmd = HNAE3_DBG_CMD_RX_QUEUE_INFO,
+		.dentry = HNS3_DBG_DENTRY_QUEUE,
+		.init = hns3_dbg_common_init_t1,
+	},
+	{
+		.name = "tx_queue_info",
+		.cmd = HNAE3_DBG_CMD_TX_QUEUE_INFO,
+		.dentry = HNS3_DBG_DENTRY_QUEUE,
+		.init = hns3_dbg_common_init_t1,
+	},
+	{
+		.name = "fd_tcam",
+		.cmd = HNAE3_DBG_CMD_FD_TCAM,
+		.dentry = HNS3_DBG_DENTRY_FD,
+		.init = hns3_dbg_common_init_t2,
+	},
+	{
+		.name = "service_task_info",
+		.cmd = HNAE3_DBG_CMD_SERV_INFO,
+		.dentry = HNS3_DBG_DENTRY_COMMON,
+		.init = hns3_dbg_common_init_t2,
+	},
+	{
+		.name = "vlan_config",
+		.cmd = HNAE3_DBG_CMD_VLAN_CONFIG,
+		.dentry = HNS3_DBG_DENTRY_COMMON,
+		.init = hns3_dbg_common_init_t2,
+	},
+	{
+		.name = "ptp_info",
+		.cmd = HNAE3_DBG_CMD_PTP_INFO,
+		.dentry = HNS3_DBG_DENTRY_COMMON,
+		.init = hns3_dbg_common_init_t2,
+	},
+	{
+		.name = "fd_counter",
+		.cmd = HNAE3_DBG_CMD_FD_COUNTER,
+		.dentry = HNS3_DBG_DENTRY_FD,
+		.init = hns3_dbg_common_init_t2,
+	},
+	{
+		.name = "umv_info",
+		.cmd = HNAE3_DBG_CMD_UMV_INFO,
+		.dentry = HNS3_DBG_DENTRY_COMMON,
+		.init = hns3_dbg_common_init_t2,
+	},
+	{
+		.name = "page_pool_info",
+		.cmd = HNAE3_DBG_CMD_PAGE_POOL_INFO,
+		.dentry = HNS3_DBG_DENTRY_COMMON,
+		.init = hns3_dbg_common_init_t1,
+	},
+	{
+		.name = "coalesce_info",
+		.cmd = HNAE3_DBG_CMD_COAL_INFO,
+		.dentry = HNS3_DBG_DENTRY_COMMON,
+		.init = hns3_dbg_common_init_t1,
+	},
+};
+
+static struct hns3_dbg_cap_info hns3_dbg_cap[] = {
+	{
+		.name = "support FD",
+		.cap_bit = HNAE3_DEV_SUPPORT_FD_B,
+	}, {
+		.name = "support GRO",
+		.cap_bit = HNAE3_DEV_SUPPORT_GRO_B,
+	}, {
+		.name = "support FEC",
+		.cap_bit = HNAE3_DEV_SUPPORT_FEC_B,
+	}, {
+		.name = "support UDP GSO",
+		.cap_bit = HNAE3_DEV_SUPPORT_UDP_GSO_B,
+	}, {
+		.name = "support PTP",
+		.cap_bit = HNAE3_DEV_SUPPORT_PTP_B,
+	}, {
+		.name = "support INT QL",
+		.cap_bit = HNAE3_DEV_SUPPORT_INT_QL_B,
+	}, {
+		.name = "support HW TX csum",
+		.cap_bit = HNAE3_DEV_SUPPORT_HW_TX_CSUM_B,
+	}, {
+		.name = "support UDP tunnel csum",
+		.cap_bit = HNAE3_DEV_SUPPORT_UDP_TUNNEL_CSUM_B,
+	}, {
+		.name = "support TX push",
+		.cap_bit = HNAE3_DEV_SUPPORT_TX_PUSH_B,
+	}, {
+		.name = "support imp-controlled PHY",
+		.cap_bit = HNAE3_DEV_SUPPORT_PHY_IMP_B,
+	}, {
+		.name = "support imp-controlled RAS",
+		.cap_bit = HNAE3_DEV_SUPPORT_RAS_IMP_B,
+	}, {
+		.name = "support rxd advanced layout",
+		.cap_bit = HNAE3_DEV_SUPPORT_RXD_ADV_LAYOUT_B,
+	}, {
+		.name = "support port vlan bypass",
+		.cap_bit = HNAE3_DEV_SUPPORT_PORT_VLAN_BYPASS_B,
+	}, {
+		.name = "support modify vlan filter state",
+		.cap_bit = HNAE3_DEV_SUPPORT_VLAN_FLTR_MDF_B,
+	}, {
+		.name = "support FEC statistics",
+		.cap_bit = HNAE3_DEV_SUPPORT_FEC_STATS_B,
+	}, {
+		.name = "support lane num",
+		.cap_bit = HNAE3_DEV_SUPPORT_LANE_NUM_B,
+	}, {
+		.name = "support wake on lan",
+		.cap_bit = HNAE3_DEV_SUPPORT_WOL_B,
+	}, {
+		.name = "support tm flush",
+		.cap_bit = HNAE3_DEV_SUPPORT_TM_FLUSH_B,
+	}, {
+		.name = "support vf fault detect",
+		.cap_bit = HNAE3_DEV_SUPPORT_VF_FAULT_B,
+	}
+};
+
+static const char * const dim_cqe_mode_str[] = { "EQE", "CQE" };
+static const char * const dim_state_str[] = { "START", "IN_PROG", "APPLY" };
+static const char * const
+dim_tune_stat_str[] = { "ON_TOP", "TIRED", "RIGHT", "LEFT" };
+
+static void hns3_get_coal_info(struct hns3_enet_tqp_vector *tqp_vector,
+			       struct seq_file *s, int i, bool is_tx)
+{
+	unsigned int gl_offset, ql_offset;
+	struct hns3_enet_coalesce *coal;
+	unsigned int reg_val;
+	struct dim *dim;
+	bool ql_enable;
+
+	if (is_tx) {
+		coal = &tqp_vector->tx_group.coal;
+		dim = &tqp_vector->tx_group.dim;
+		gl_offset = HNS3_VECTOR_GL1_OFFSET;
+		ql_offset = HNS3_VECTOR_TX_QL_OFFSET;
+		ql_enable = tqp_vector->tx_group.coal.ql_enable;
+	} else {
+		coal = &tqp_vector->rx_group.coal;
+		dim = &tqp_vector->rx_group.dim;
+		gl_offset = HNS3_VECTOR_GL0_OFFSET;
+		ql_offset = HNS3_VECTOR_RX_QL_OFFSET;
+		ql_enable = tqp_vector->rx_group.coal.ql_enable;
+	}
+
+	seq_printf(s, "%-8d", i);
+	seq_printf(s, "%-12s", dim->state < ARRAY_SIZE(dim_state_str) ?
+		   dim_state_str[dim->state] : "unknown");
+	seq_printf(s, "%-12u", dim->profile_ix);
+	seq_printf(s, "%-10s", dim->mode < ARRAY_SIZE(dim_cqe_mode_str) ?
+		   dim_cqe_mode_str[dim->mode] : "unknown");
+	seq_printf(s, "%-12s", dim->tune_state < ARRAY_SIZE(dim_tune_stat_str) ?
+		   dim_tune_stat_str[dim->tune_state] : "unknown");
+	seq_printf(s, "%-12u%-13u%-7u%-7u%-7u", dim->steps_left,
+		   dim->steps_right, dim->tired, coal->int_gl, coal->int_ql);
+	reg_val = readl(tqp_vector->mask_addr + gl_offset) &
+		  HNS3_VECTOR_GL_MASK;
+	seq_printf(s, "%-7u", reg_val);
+	if (ql_enable) {
+		reg_val = readl(tqp_vector->mask_addr + ql_offset) &
+			  HNS3_VECTOR_QL_MASK;
+		seq_printf(s, "%u\n", reg_val);
+	} else {
+		seq_puts(s, "NA\n");
+	}
+}
+
+static void hns3_dump_coal_info(struct seq_file *s, bool is_tx)
+{
+	struct hnae3_handle *h = hnae3_seq_file_to_handle(s);
+	struct hns3_enet_tqp_vector *tqp_vector;
+	struct hns3_nic_priv *priv = h->priv;
+	unsigned int i;
+
+	seq_printf(s, "%s interrupt coalesce info:\n", is_tx ? "tx" : "rx");
+
+	seq_puts(s, "VEC_ID  ALGO_STATE  PROFILE_ID  CQE_MODE  TUNE_STATE  ");
+	seq_puts(s, "STEPS_LEFT  STEPS_RIGHT  TIRED  SW_GL  SW_QL  ");
+	seq_puts(s, "HW_GL  HW_QL\n");
+
+	for (i = 0; i < priv->vector_num; i++) {
+		tqp_vector = &priv->tqp_vector[i];
+		hns3_get_coal_info(tqp_vector, s, i, is_tx);
+	}
+}
+
+static int hns3_dbg_coal_info(struct seq_file *s, void *data)
+{
+	hns3_dump_coal_info(s, true);
+	seq_puts(s, "\n");
+	hns3_dump_coal_info(s, false);
+
+	return 0;
+}
+
+static void hns3_dump_rx_queue_info(struct hns3_enet_ring *ring,
+				    struct seq_file *s, u32 index)
+{
+	struct hnae3_ae_dev *ae_dev = hnae3_seq_file_to_ae_dev(s);
+	void __iomem *base = ring->tqp->io_base;
+	u32 base_add_l, base_add_h;
+
+	seq_printf(s, "%-10u", index);
+	seq_printf(s, "%-8u",
+		   readl_relaxed(base + HNS3_RING_RX_RING_BD_NUM_REG));
+	seq_printf(s, "%-8u",
+		   readl_relaxed(base + HNS3_RING_RX_RING_BD_LEN_REG));
+	seq_printf(s, "%-6u",
+		   readl_relaxed(base + HNS3_RING_RX_RING_TAIL_REG));
+	seq_printf(s, "%-6u",
+		   readl_relaxed(base + HNS3_RING_RX_RING_HEAD_REG));
+	seq_printf(s, "%-8u",
+		   readl_relaxed(base + HNS3_RING_RX_RING_FBDNUM_REG));
+	seq_printf(s, "%-11u", readl_relaxed(base +
+		   HNS3_RING_RX_RING_PKTNUM_RECORD_REG));
+	seq_printf(s, "%-11u", ring->rx_copybreak);
+	seq_printf(s, "%-9s",
+		   str_on_off(readl_relaxed(base + HNS3_RING_EN_REG)));
+
+	if (hnae3_ae_dev_tqp_txrx_indep_supported(ae_dev))
+		seq_printf(s, "%-12s", str_on_off(readl_relaxed(base +
+						  HNS3_RING_RX_EN_REG)));
+	else
+		seq_printf(s, "%-12s", "NA");
+
+	base_add_h = readl_relaxed(base + HNS3_RING_RX_RING_BASEADDR_H_REG);
+	base_add_l = readl_relaxed(base + HNS3_RING_RX_RING_BASEADDR_L_REG);
+	seq_printf(s, "0x%08x%08x\n", base_add_h, base_add_l);
+}
+
+static int hns3_dbg_rx_queue_info(struct seq_file *s, void *data)
+{
+	struct hnae3_handle *h = hnae3_seq_file_to_handle(s);
+	struct hns3_nic_priv *priv = h->priv;
+	struct hns3_enet_ring *ring;
+	u32 i;
+
+	if (!priv->ring) {
+		dev_err(&h->pdev->dev, "priv->ring is NULL\n");
+		return -EFAULT;
+	}
+
+	seq_puts(s, "QUEUE_ID  BD_NUM  BD_LEN  TAIL  HEAD  FBDNUM  ");
+	seq_puts(s, "PKTNUM     COPYBREAK  RING_EN  RX_RING_EN  BASE_ADDR\n");
+
+	for (i = 0; i < h->kinfo.num_tqps; i++) {
+		/* Each cycle needs to determine whether the instance is reset,
+		 * to prevent reference to invalid memory. And need to ensure
+		 * that the following code is executed within 100ms.
+		 */
+		if (!test_bit(HNS3_NIC_STATE_INITED, &priv->state) ||
+		    test_bit(HNS3_NIC_STATE_RESETTING, &priv->state))
+			return -EPERM;
+
+		ring = &priv->ring[(u32)(i + h->kinfo.num_tqps)];
+		hns3_dump_rx_queue_info(ring, s, i);
+	}
+
+	return 0;
+}
+
+static void hns3_dump_tx_queue_info(struct hns3_enet_ring *ring,
+				    struct seq_file *s, u32 index)
+{
+	struct hnae3_ae_dev *ae_dev = hnae3_seq_file_to_ae_dev(s);
+	void __iomem *base = ring->tqp->io_base;
+	u32 base_add_l, base_add_h;
+
+	seq_printf(s, "%-10u", index);
+	seq_printf(s, "%-8u",
+		   readl_relaxed(base + HNS3_RING_TX_RING_BD_NUM_REG));
+	seq_printf(s, "%-4u", readl_relaxed(base + HNS3_RING_TX_RING_TC_REG));
+	seq_printf(s, "%-6u", readl_relaxed(base + HNS3_RING_TX_RING_TAIL_REG));
+	seq_printf(s, "%-6u", readl_relaxed(base + HNS3_RING_TX_RING_HEAD_REG));
+	seq_printf(s, "%-8u",
+		   readl_relaxed(base + HNS3_RING_TX_RING_FBDNUM_REG));
+	seq_printf(s, "%-8u",
+		   readl_relaxed(base + HNS3_RING_TX_RING_OFFSET_REG));
+	seq_printf(s, "%-11u",
+		   readl_relaxed(base + HNS3_RING_TX_RING_PKTNUM_RECORD_REG));
+	seq_printf(s, "%-9s",
+		   str_on_off(readl_relaxed(base + HNS3_RING_EN_REG)));
+
+	if (hnae3_ae_dev_tqp_txrx_indep_supported(ae_dev))
+		seq_printf(s, "%-12s",
+			   str_on_off(readl_relaxed(base +
+						    HNS3_RING_TX_EN_REG)));
+	else
+		seq_printf(s, "%-12s", "NA");
+
+	base_add_h = readl_relaxed(base + HNS3_RING_TX_RING_BASEADDR_H_REG);
+	base_add_l = readl_relaxed(base + HNS3_RING_TX_RING_BASEADDR_L_REG);
+	seq_printf(s, "0x%08x%08x\n", base_add_h, base_add_l);
+}
+
+static int hns3_dbg_tx_queue_info(struct seq_file *s, void *data)
+{
+	struct hnae3_handle *h = hnae3_seq_file_to_handle(s);
+	struct hns3_nic_priv *priv = h->priv;
+	struct hns3_enet_ring *ring;
+	u32 i;
+
+	if (!priv->ring) {
+		dev_err(&h->pdev->dev, "priv->ring is NULL\n");
+		return -EFAULT;
+	}
+
+	seq_puts(s, "QUEUE_ID  BD_NUM  TC  TAIL  HEAD  FBDNUM  OFFSET  ");
+	seq_puts(s, "PKTNUM     RING_EN  TX_RING_EN  BASE_ADDR\n");
+
+	for (i = 0; i < h->kinfo.num_tqps; i++) {
+		/* Each cycle needs to determine whether the instance is reset,
+		 * to prevent reference to invalid memory. And need to ensure
+		 * that the following code is executed within 100ms.
+		 */
+		if (!test_bit(HNS3_NIC_STATE_INITED, &priv->state) ||
+		    test_bit(HNS3_NIC_STATE_RESETTING, &priv->state))
+			return -EPERM;
+
+		ring = &priv->ring[i];
+		hns3_dump_tx_queue_info(ring, s, i);
+	}
+
+	return 0;
+}
+
+static int hns3_dbg_queue_map(struct seq_file *s, void *data)
+{
+	struct hnae3_handle *h = hnae3_seq_file_to_handle(s);
+	struct hns3_nic_priv *priv = h->priv;
+	u32 i;
+
+	if (!h->ae_algo->ops->get_global_queue_id)
+		return -EOPNOTSUPP;
+
+	seq_puts(s, "local_queue_id  global_queue_id  vector_id\n");
+
+	for (i = 0; i < h->kinfo.num_tqps; i++) {
+		if (!priv->ring || !priv->ring[i].tqp_vector)
+			continue;
+		seq_printf(s, "%-16u%-17u%d\n", i,
+			   h->ae_algo->ops->get_global_queue_id(h, i),
+			   priv->ring[i].tqp_vector->vector_irq);
+	}
+
+	return 0;
+}
+
+static void hns3_dump_rx_bd_info(struct hns3_nic_priv *priv,
+				 struct hns3_desc *desc, struct seq_file *s,
+				 int idx)
+{
+	seq_printf(s, "%-9d%#-11x%-10u%-8u%#-12x%-7u%-10u%-17u%-13u%#-14x",
+		   idx, le32_to_cpu(desc->rx.l234_info),
+		   le16_to_cpu(desc->rx.pkt_len), le16_to_cpu(desc->rx.size),
+		   le32_to_cpu(desc->rx.rss_hash), le16_to_cpu(desc->rx.fd_id),
+		   le16_to_cpu(desc->rx.vlan_tag),
+		   le16_to_cpu(desc->rx.o_dm_vlan_id_fb),
+		   le16_to_cpu(desc->rx.ot_vlan_tag),
+		   le32_to_cpu(desc->rx.bd_base_info));
+
+	if (test_bit(HNS3_NIC_STATE_RXD_ADV_LAYOUT_ENABLE, &priv->state)) {
+		u32 ol_info = le32_to_cpu(desc->rx.ol_info);
+
+		seq_printf(s, "%-7lu%-9u\n",
+			   hnae3_get_field(ol_info, HNS3_RXD_PTYPE_M,
+					   HNS3_RXD_PTYPE_S),
+			   le16_to_cpu(desc->csum));
+	} else {
+		seq_puts(s, "NA     NA\n");
+	}
+}
+
+static int hns3_dbg_rx_bd_info(struct seq_file *s, void *private)
+{
+	struct hns3_dbg_data *data = s->private;
+	struct hnae3_handle *h = data->handle;
+	struct hns3_nic_priv *priv = h->priv;
+	struct hns3_enet_ring *ring;
+	struct hns3_desc *desc;
+	unsigned int i;
+
+	if (data->qid >= h->kinfo.num_tqps) {
+		dev_err(&h->pdev->dev, "queue%u is not in use\n", data->qid);
+		return -EINVAL;
+	}
+
+	seq_printf(s, "Queue %u rx bd info:\n", data->qid);
+	seq_puts(s, "BD_IDX   L234_INFO  PKT_LEN   SIZE    ");
+	seq_puts(s, "RSS_HASH    FD_ID  VLAN_TAG  O_DM_VLAN_ID_FB  ");
+	seq_puts(s, "OT_VLAN_TAG  BD_BASE_INFO  PTYPE  HW_CSUM\n");
+
+	ring = &priv->ring[data->qid + data->handle->kinfo.num_tqps];
+	for (i = 0; i < ring->desc_num; i++) {
+		desc = &ring->desc[i];
+
+		hns3_dump_rx_bd_info(priv, desc, s, i);
+	}
+
+	return 0;
+}
+
+static void hns3_dump_tx_bd_info(struct hns3_desc *desc, struct seq_file *s,
+				 int idx)
+{
+	seq_printf(s, "%-8d%#-20llx%-10u%-6u%#-15x%-14u%-7u%-16u%#-14x%#-14x%-11u\n",
+		   idx, le64_to_cpu(desc->addr),
+		   le16_to_cpu(desc->tx.vlan_tag),
+		   le16_to_cpu(desc->tx.send_size),
+		   le32_to_cpu(desc->tx.type_cs_vlan_tso_len),
+		   le16_to_cpu(desc->tx.outer_vlan_tag),
+		   le16_to_cpu(desc->tx.tv),
+		   le32_to_cpu(desc->tx.ol_type_vlan_len_msec),
+		   le32_to_cpu(desc->tx.paylen_ol4cs),
+		   le16_to_cpu(desc->tx.bdtp_fe_sc_vld_ra_ri),
+		   le16_to_cpu(desc->tx.mss_hw_csum));
+}
+
+static int hns3_dbg_tx_bd_info(struct seq_file *s, void *private)
+{
+	struct hns3_dbg_data *data = s->private;
+	struct hnae3_handle *h = data->handle;
+	struct hns3_nic_priv *priv = h->priv;
+	struct hns3_enet_ring *ring;
+	struct hns3_desc *desc;
+	unsigned int i;
+
+	if (data->qid >= h->kinfo.num_tqps) {
+		dev_err(&h->pdev->dev, "queue%u is not in use\n", data->qid);
+		return -EINVAL;
+	}
+
+	seq_printf(s, "Queue %u tx bd info:\n", data->qid);
+	seq_puts(s, "BD_IDX  ADDRESS             VLAN_TAG  SIZE  ");
+	seq_puts(s, "T_CS_VLAN_TSO  OT_VLAN_TAG   TV     OLT_VLAN_LEN  ");
+	seq_puts(s, "PAYLEN_OL4CS  BD_FE_SC_VLD   MSS_HW_CSUM\n");
+
+	ring = &priv->ring[data->qid];
+	for (i = 0; i < ring->desc_num; i++) {
+		desc = &ring->desc[i];
+
+		hns3_dump_tx_bd_info(desc, s, i);
+	}
+
+	return 0;
+}
+
+static void hns3_dbg_dev_caps(struct hnae3_handle *h, struct seq_file *s)
+{
+	struct hnae3_ae_dev *ae_dev = hns3_get_ae_dev(h);
+	unsigned long *caps = ae_dev->caps;
+	u32 i, state;
+
+	seq_puts(s, "dev capability:\n");
+
+	for (i = 0; i < ARRAY_SIZE(hns3_dbg_cap); i++) {
+		state = test_bit(hns3_dbg_cap[i].cap_bit, caps);
+		seq_printf(s, "%s: %s\n", hns3_dbg_cap[i].name,
+			   str_yes_no(state));
+	}
+
+	seq_puts(s, "\n");
+}
+
+static void hns3_dbg_dev_specs(struct hnae3_handle *h, struct seq_file *s)
+{
+	struct hnae3_ae_dev *ae_dev = pci_get_drvdata(h->pdev);
+	struct hnae3_dev_specs *dev_specs = &ae_dev->dev_specs;
+	struct hnae3_knic_private_info *kinfo = &h->kinfo;
+	struct net_device *dev = kinfo->netdev;
+
+	seq_puts(s, "dev_spec:\n");
+	seq_printf(s, "MAC entry num: %u\n", dev_specs->mac_entry_num);
+	seq_printf(s, "MNG entry num: %u\n", dev_specs->mng_entry_num);
+	seq_printf(s, "MAX non tso bd num: %u\n",
+		   dev_specs->max_non_tso_bd_num);
+	seq_printf(s, "RSS ind tbl size: %u\n", dev_specs->rss_ind_tbl_size);
+	seq_printf(s, "RSS key size: %u\n", dev_specs->rss_key_size);
+	seq_printf(s, "RSS size: %u\n", kinfo->rss_size);
+	seq_printf(s, "Allocated RSS size: %u\n", kinfo->req_rss_size);
+	seq_printf(s, "Task queue pairs numbers: %u\n", kinfo->num_tqps);
+	seq_printf(s, "RX buffer length: %u\n", kinfo->rx_buf_len);
+	seq_printf(s, "Desc num per TX queue: %u\n", kinfo->num_tx_desc);
+	seq_printf(s, "Desc num per RX queue: %u\n", kinfo->num_rx_desc);
+	seq_printf(s, "Total number of enabled TCs: %u\n",
+		   kinfo->tc_info.num_tc);
+	seq_printf(s, "MAX INT QL: %u\n", dev_specs->int_ql_max);
+	seq_printf(s, "MAX INT GL: %u\n", dev_specs->max_int_gl);
+	seq_printf(s, "MAX TM RATE: %u\n", dev_specs->max_tm_rate);
+	seq_printf(s, "MAX QSET number: %u\n", dev_specs->max_qset_num);
+	seq_printf(s, "umv size: %u\n", dev_specs->umv_size);
+	seq_printf(s, "mc mac size: %u\n", dev_specs->mc_mac_size);
+	seq_printf(s, "MAC statistics number: %u\n", dev_specs->mac_stats_num);
+	seq_printf(s, "TX timeout threshold: %d seconds\n",
+		   dev->watchdog_timeo / HZ);
+	seq_printf(s, "mac tunnel number: %u\n", dev_specs->tnl_num);
+	seq_printf(s, "Hilink Version: %u\n", dev_specs->hilink_version);
+}
+
+static int hns3_dbg_dev_info(struct seq_file *s, void *data)
+{
+	struct hnae3_handle *h = hnae3_seq_file_to_handle(s);
+
+	hns3_dbg_dev_caps(h, s);
+	hns3_dbg_dev_specs(h, s);
+
+	return 0;
+}
+
+static void hns3_dump_page_pool_info(struct hns3_enet_ring *ring,
+				     struct seq_file *s, u32 index)
+{
+	seq_printf(s, "%-10u%-14u%-14d%-21u%-7u%-9d%uK\n",
+		   index,
+		   READ_ONCE(ring->page_pool->pages_state_hold_cnt),
+		   atomic_read(&ring->page_pool->pages_state_release_cnt),
+		   ring->page_pool->p.pool_size,
+		   ring->page_pool->p.order,
+		   ring->page_pool->p.nid,
+		   ring->page_pool->p.max_len / 1024);
+}
+
+static int hns3_dbg_page_pool_info(struct seq_file *s, void *data)
+{
+	struct hnae3_handle *h = hnae3_seq_file_to_handle(s);
+	struct hns3_nic_priv *priv = h->priv;
+	struct hns3_enet_ring *ring;
+	u32 i;
+
+	if (!priv->ring) {
+		dev_err(&h->pdev->dev, "priv->ring is NULL\n");
+		return -EFAULT;
+	}
+
+	if (!priv->ring[h->kinfo.num_tqps].page_pool) {
+		dev_err(&h->pdev->dev, "page pool is not initialized\n");
+		return -EFAULT;
+	}
+
+	seq_puts(s, "QUEUE_ID  ALLOCATE_CNT  FREE_CNT      ");
+	seq_puts(s, "POOL_SIZE(PAGE_NUM)  ORDER  NUMA_ID  MAX_LEN\n");
+
+	for (i = 0; i < h->kinfo.num_tqps; i++) {
+		if (!test_bit(HNS3_NIC_STATE_INITED, &priv->state) ||
+		    test_bit(HNS3_NIC_STATE_RESETTING, &priv->state))
+			return -EPERM;
+
+		ring = &priv->ring[(u32)(i + h->kinfo.num_tqps)];
+		hns3_dump_page_pool_info(ring, s, i);
+	}
+
+	return 0;
+}
+
+static int hns3_dbg_bd_info_show(struct seq_file *s, void *private)
+{
+	struct hns3_dbg_data *data = s->private;
+	struct hnae3_handle *h = data->handle;
+	struct hns3_nic_priv *priv = h->priv;
+
+	if (!test_bit(HNS3_NIC_STATE_INITED, &priv->state) ||
+	    test_bit(HNS3_NIC_STATE_RESETTING, &priv->state))
+		return -EBUSY;
+
+	if (data->cmd == HNAE3_DBG_CMD_TX_BD)
+		return hns3_dbg_tx_bd_info(s, private);
+	else if (data->cmd == HNAE3_DBG_CMD_RX_BD)
+		return hns3_dbg_rx_bd_info(s, private);
+
+	return -EOPNOTSUPP;
+}
+DEFINE_SHOW_ATTRIBUTE(hns3_dbg_bd_info);
+
+static int hns3_dbg_bd_file_init(struct hnae3_handle *handle, u32 cmd)
+{
+	struct hns3_dbg_data *data;
+	struct dentry *entry_dir;
+	u16 max_queue_num;
+	unsigned int i;
+
+	entry_dir = hns3_dbg_dentry[hns3_dbg_cmd[cmd].dentry].dentry;
+	max_queue_num = hns3_get_max_available_channels(handle);
+	data = devm_kcalloc(&handle->pdev->dev, max_queue_num, sizeof(*data),
+			    GFP_KERNEL);
+	if (!data)
+		return -ENOMEM;
+
+	for (i = 0; i < max_queue_num; i++) {
+		char name[HNS3_DBG_FILE_NAME_LEN];
+
+		data[i].handle = handle;
+		data[i].cmd = hns3_dbg_cmd[cmd].cmd;
+		data[i].qid = i;
+		sprintf(name, "%s%u", hns3_dbg_cmd[cmd].name, i);
+		debugfs_create_file(name, 0400, entry_dir, &data[i],
+				    &hns3_dbg_bd_info_fops);
+	}
+
+	return 0;
+}
+
+static int hns3_dbg_common_init_t1(struct hnae3_handle *handle, u32 cmd)
+{
+	struct device *dev = &handle->pdev->dev;
+	struct dentry *entry_dir;
+	read_func func = NULL;
+
+	switch (hns3_dbg_cmd[cmd].cmd) {
+	case HNAE3_DBG_CMD_TX_QUEUE_INFO:
+		func = hns3_dbg_tx_queue_info;
+		break;
+	case HNAE3_DBG_CMD_RX_QUEUE_INFO:
+		func = hns3_dbg_rx_queue_info;
+		break;
+	case HNAE3_DBG_CMD_QUEUE_MAP:
+		func = hns3_dbg_queue_map;
+		break;
+	case HNAE3_DBG_CMD_PAGE_POOL_INFO:
+		func = hns3_dbg_page_pool_info;
+		break;
+	case HNAE3_DBG_CMD_COAL_INFO:
+		func = hns3_dbg_coal_info;
+		break;
+	case HNAE3_DBG_CMD_DEV_INFO:
+		func = hns3_dbg_dev_info;
+		break;
+	default:
+		return -EINVAL;
+	}
+
+	entry_dir = hns3_dbg_dentry[hns3_dbg_cmd[cmd].dentry].dentry;
+	debugfs_create_devm_seqfile(dev, hns3_dbg_cmd[cmd].name, entry_dir,
+				    func);
+
+	return 0;
+}
+
+static int hns3_dbg_common_init_t2(struct hnae3_handle *handle, u32 cmd)
+{
+	const struct hnae3_ae_ops *ops = hns3_get_ops(handle);
+	struct device *dev = &handle->pdev->dev;
+	struct dentry *entry_dir;
+	read_func func;
+	int ret;
+
+	if (!ops->dbg_get_read_func)
+		return 0;
+
+	ret = ops->dbg_get_read_func(handle, hns3_dbg_cmd[cmd].cmd, &func);
+	if (ret)
+		return ret;
+
+	entry_dir = hns3_dbg_dentry[hns3_dbg_cmd[cmd].dentry].dentry;
+	debugfs_create_devm_seqfile(dev, hns3_dbg_cmd[cmd].name, entry_dir,
+				    func);
+
+	return 0;
+}
+
+int hns3_dbg_init(struct hnae3_handle *handle)
+{
+	struct hnae3_ae_dev *ae_dev = hns3_get_ae_dev(handle);
+	const char *name = pci_name(handle->pdev);
+	int ret;
+	u32 i;
+
+	hns3_dbg_dentry[HNS3_DBG_DENTRY_COMMON].dentry =
+				debugfs_create_dir(name, hns3_dbgfs_root);
+	handle->hnae3_dbgfs = hns3_dbg_dentry[HNS3_DBG_DENTRY_COMMON].dentry;
+
+	for (i = 0; i < HNS3_DBG_DENTRY_COMMON; i++)
+		hns3_dbg_dentry[i].dentry =
+			debugfs_create_dir(hns3_dbg_dentry[i].name,
+					   handle->hnae3_dbgfs);
+
+	for (i = 0; i < ARRAY_SIZE(hns3_dbg_cmd); i++) {
+		if ((hns3_dbg_cmd[i].cmd == HNAE3_DBG_CMD_TM_NODES &&
+		     ae_dev->dev_version <= HNAE3_DEVICE_VERSION_V2) ||
+		    (hns3_dbg_cmd[i].cmd == HNAE3_DBG_CMD_PTP_INFO &&
+		     !test_bit(HNAE3_DEV_SUPPORT_PTP_B, ae_dev->caps)))
+			continue;
+
+		if (!hns3_dbg_cmd[i].init) {
+			dev_err(&handle->pdev->dev,
+				"cmd %s lack of init func\n",
+				hns3_dbg_cmd[i].name);
+			ret = -EINVAL;
+			goto out;
+		}
+
+		ret = hns3_dbg_cmd[i].init(handle, i);
+		if (ret) {
+			dev_err(&handle->pdev->dev, "failed to init cmd %s\n",
+				hns3_dbg_cmd[i].name);
+			goto out;
+		}
+	}
+
+	return 0;
+
+out:
+	debugfs_remove_recursive(handle->hnae3_dbgfs);
+	handle->hnae3_dbgfs = NULL;
+	return ret;
+}
+
+void hns3_dbg_uninit(struct hnae3_handle *handle)
+{
+	debugfs_remove_recursive(handle->hnae3_dbgfs);
+	handle->hnae3_dbgfs = NULL;
+}
+
+void hns3_dbg_register_debugfs(const char *debugfs_dir_name)
+{
+	hns3_dbgfs_root = debugfs_create_dir(debugfs_dir_name, NULL);
+}
+
+void hns3_dbg_unregister_debugfs(void)
+{
+	debugfs_remove_recursive(hns3_dbgfs_root);
+	hns3_dbgfs_root = NULL;
+}
diff --git a/drivers/net/ethernet/hisilicon/hns3/hns3_debugfs.h b/drivers/net/ethernet/hisilicon/hns3/hns3_debugfs.h
new file mode 100644
index 000000000000..57c9d3fc1b27
--- /dev/null
+++ b/drivers/net/ethernet/hisilicon/hns3/hns3_debugfs.h
@@ -0,0 +1,51 @@
+/* SPDX-License-Identifier: GPL-2.0+ */
+/* Copyright (c) 2021 Hisilicon Limited. */
+
+#ifndef __HNS3_DEBUGFS_H
+#define __HNS3_DEBUGFS_H
+
+#include "hnae3.h"
+
+#define HNS3_DBG_ITEM_NAME_LEN	32
+#define HNS3_DBG_FILE_NAME_LEN	16
+
+struct hns3_dbg_item {
+	char name[HNS3_DBG_ITEM_NAME_LEN];
+	u16 interval; /* blank numbers after the item */
+};
+
+struct hns3_dbg_data {
+	struct hnae3_handle *handle;
+	enum hnae3_dbg_cmd cmd;
+	u16 qid;
+};
+
+enum hns3_dbg_dentry_type {
+	HNS3_DBG_DENTRY_TM,
+	HNS3_DBG_DENTRY_TX_BD,
+	HNS3_DBG_DENTRY_RX_BD,
+	HNS3_DBG_DENTRY_MAC,
+	HNS3_DBG_DENTRY_REG,
+	HNS3_DBG_DENTRY_QUEUE,
+	HNS3_DBG_DENTRY_FD,
+	HNS3_DBG_DENTRY_COMMON,
+};
+
+struct hns3_dbg_dentry_info {
+	const char *name;
+	struct dentry *dentry;
+};
+
+struct hns3_dbg_cmd_info {
+	const char *name;
+	enum hnae3_dbg_cmd cmd;
+	enum hns3_dbg_dentry_type dentry;
+	int (*init)(struct hnae3_handle *handle, unsigned int cmd);
+};
+
+struct hns3_dbg_cap_info {
+	const char *name;
+	enum HNAE3_DEV_CAP_BITS cap_bit;
+};
+
+#endif
diff --git a/drivers/net/ethernet/hisilicon/hns3/hns3_enet.c b/drivers/net/ethernet/hisilicon/hns3/hns3_enet.c
new file mode 100644
index 000000000000..bfa5568baa92
--- /dev/null
+++ b/drivers/net/ethernet/hisilicon/hns3/hns3_enet.c
@@ -0,0 +1,6045 @@
+// SPDX-License-Identifier: GPL-2.0+
+// Copyright (c) 2016-2017 Hisilicon Limited.
+
+#include <linux/dma-mapping.h>
+#include <linux/etherdevice.h>
+#include <linux/interrupt.h>
+#ifdef CONFIG_RFS_ACCEL
+#include <linux/cpu_rmap.h>
+#endif
+#include <linux/if_vlan.h>
+#include <linux/irq.h>
+#include <linux/ip.h>
+#include <linux/ipv6.h>
+#include <linux/iommu.h>
+#include <linux/module.h>
+#include <linux/pci.h>
+#include <linux/skbuff.h>
+#include <linux/sctp.h>
+#include <net/gre.h>
+#include <net/gro.h>
+#include <net/ip6_checksum.h>
+#include <net/page_pool/helpers.h>
+#include <net/pkt_cls.h>
+#include <net/pkt_sched.h>
+#include <net/tcp.h>
+#include <net/vxlan.h>
+#include <net/geneve.h>
+
+#include "hnae3.h"
+#include "hns3_enet.h"
+/* All hns3 tracepoints are defined by the include below, which
+ * must be included exactly once across the whole kernel with
+ * CREATE_TRACE_POINTS defined
+ */
+#define CREATE_TRACE_POINTS
+#include "hns3_trace.h"
+
+#define hns3_set_field(origin, shift, val)	((origin) |= (val) << (shift))
+#define hns3_tx_bd_count(S)	DIV_ROUND_UP(S, HNS3_MAX_BD_SIZE)
+
+#define hns3_rl_err(fmt, ...)						\
+	do {								\
+		if (net_ratelimit())					\
+			netdev_err(fmt, ##__VA_ARGS__);			\
+	} while (0)
+
+static void hns3_clear_all_ring(struct hnae3_handle *h, bool force);
+
+static const char hns3_driver_name[] = "hns3";
+static const char hns3_driver_string[] =
+			"Hisilicon Ethernet Network Driver for Hip08 Family";
+static const char hns3_copyright[] = "Copyright (c) 2017 Huawei Corporation.";
+static struct hnae3_client client;
+
+static int debug = -1;
+module_param(debug, int, 0);
+MODULE_PARM_DESC(debug, " Network interface message level setting");
+
+static unsigned int tx_sgl = 1;
+module_param(tx_sgl, uint, 0600);
+MODULE_PARM_DESC(tx_sgl, "Minimum number of frags when using dma_map_sg() to optimize the IOMMU mapping");
+
+static bool page_pool_enabled = true;
+module_param(page_pool_enabled, bool, 0400);
+
+#define HNS3_SGL_SIZE(nfrag)	(sizeof(struct scatterlist) * (nfrag) +	\
+				 sizeof(struct sg_table))
+#define HNS3_MAX_SGL_SIZE	ALIGN(HNS3_SGL_SIZE(HNS3_MAX_TSO_BD_NUM), \
+				      dma_get_cache_alignment())
+
+#define DEFAULT_MSG_LEVEL (NETIF_MSG_PROBE | NETIF_MSG_LINK | \
+			   NETIF_MSG_IFDOWN | NETIF_MSG_IFUP)
+
+#define HNS3_INNER_VLAN_TAG	1
+#define HNS3_OUTER_VLAN_TAG	2
+
+#define HNS3_MIN_TX_LEN		33U
+#define HNS3_MIN_TUN_PKT_LEN	65U
+
+/* hns3_pci_tbl - PCI Device ID Table
+ *
+ * Last entry must be all 0s
+ *
+ * { Vendor ID, Device ID, SubVendor ID, SubDevice ID,
+ *   Class, Class Mask, private data (not used) }
+ */
+static const struct pci_device_id hns3_pci_tbl[] = {
+	{PCI_VDEVICE(HUAWEI, HNAE3_DEV_ID_GE), 0},
+	{PCI_VDEVICE(HUAWEI, HNAE3_DEV_ID_25GE), 0},
+	{PCI_VDEVICE(HUAWEI, HNAE3_DEV_ID_25GE_RDMA),
+	 HNAE3_DEV_SUPPORT_ROCE_DCB_BITS},
+	{PCI_VDEVICE(HUAWEI, HNAE3_DEV_ID_25GE_RDMA_MACSEC),
+	 HNAE3_DEV_SUPPORT_ROCE_DCB_BITS},
+	{PCI_VDEVICE(HUAWEI, HNAE3_DEV_ID_50GE_RDMA),
+	 HNAE3_DEV_SUPPORT_ROCE_DCB_BITS},
+	{PCI_VDEVICE(HUAWEI, HNAE3_DEV_ID_50GE_RDMA_MACSEC),
+	 HNAE3_DEV_SUPPORT_ROCE_DCB_BITS},
+	{PCI_VDEVICE(HUAWEI, HNAE3_DEV_ID_100G_RDMA_MACSEC),
+	 HNAE3_DEV_SUPPORT_ROCE_DCB_BITS},
+	{PCI_VDEVICE(HUAWEI, HNAE3_DEV_ID_200G_RDMA),
+	 HNAE3_DEV_SUPPORT_ROCE_DCB_BITS},
+	{PCI_VDEVICE(HUAWEI, HNAE3_DEV_ID_VF), 0},
+	{PCI_VDEVICE(HUAWEI, HNAE3_DEV_ID_RDMA_DCB_PFC_VF),
+	 HNAE3_DEV_SUPPORT_ROCE_DCB_BITS},
+	/* required last entry */
+	{0,}
+};
+MODULE_DEVICE_TABLE(pci, hns3_pci_tbl);
+
+#define HNS3_RX_PTYPE_ENTRY(ptype, l, s, t, h) \
+	{	ptype, \
+		l, \
+		CHECKSUM_##s, \
+		HNS3_L3_TYPE_##t, \
+		1, \
+		h}
+
+#define HNS3_RX_PTYPE_UNUSED_ENTRY(ptype) \
+		{ ptype, 0, CHECKSUM_NONE, HNS3_L3_TYPE_PARSE_FAIL, 0, \
+		  PKT_HASH_TYPE_NONE }
+
+static const struct hns3_rx_ptype hns3_rx_ptype_tbl[] = {
+	HNS3_RX_PTYPE_UNUSED_ENTRY(0),
+	HNS3_RX_PTYPE_ENTRY(1, 0, COMPLETE, ARP, PKT_HASH_TYPE_NONE),
+	HNS3_RX_PTYPE_ENTRY(2, 0, COMPLETE, RARP, PKT_HASH_TYPE_NONE),
+	HNS3_RX_PTYPE_ENTRY(3, 0, COMPLETE, LLDP, PKT_HASH_TYPE_NONE),
+	HNS3_RX_PTYPE_ENTRY(4, 0, COMPLETE, PARSE_FAIL, PKT_HASH_TYPE_NONE),
+	HNS3_RX_PTYPE_ENTRY(5, 0, COMPLETE, PARSE_FAIL, PKT_HASH_TYPE_NONE),
+	HNS3_RX_PTYPE_ENTRY(6, 0, COMPLETE, PARSE_FAIL, PKT_HASH_TYPE_NONE),
+	HNS3_RX_PTYPE_ENTRY(7, 0, COMPLETE, CNM, PKT_HASH_TYPE_NONE),
+	HNS3_RX_PTYPE_ENTRY(8, 0, NONE, PARSE_FAIL, PKT_HASH_TYPE_NONE),
+	HNS3_RX_PTYPE_UNUSED_ENTRY(9),
+	HNS3_RX_PTYPE_UNUSED_ENTRY(10),
+	HNS3_RX_PTYPE_UNUSED_ENTRY(11),
+	HNS3_RX_PTYPE_UNUSED_ENTRY(12),
+	HNS3_RX_PTYPE_UNUSED_ENTRY(13),
+	HNS3_RX_PTYPE_UNUSED_ENTRY(14),
+	HNS3_RX_PTYPE_UNUSED_ENTRY(15),
+	HNS3_RX_PTYPE_ENTRY(16, 0, COMPLETE, PARSE_FAIL, PKT_HASH_TYPE_NONE),
+	HNS3_RX_PTYPE_ENTRY(17, 0, COMPLETE, IPV4, PKT_HASH_TYPE_NONE),
+	HNS3_RX_PTYPE_ENTRY(18, 0, COMPLETE, IPV4, PKT_HASH_TYPE_NONE),
+	HNS3_RX_PTYPE_ENTRY(19, 0, UNNECESSARY, IPV4, PKT_HASH_TYPE_L4),
+	HNS3_RX_PTYPE_ENTRY(20, 0, UNNECESSARY, IPV4, PKT_HASH_TYPE_L4),
+	HNS3_RX_PTYPE_ENTRY(21, 0, NONE, IPV4, PKT_HASH_TYPE_NONE),
+	HNS3_RX_PTYPE_ENTRY(22, 0, UNNECESSARY, IPV4, PKT_HASH_TYPE_L4),
+	HNS3_RX_PTYPE_ENTRY(23, 0, NONE, IPV4, PKT_HASH_TYPE_L3),
+	HNS3_RX_PTYPE_ENTRY(24, 0, NONE, IPV4, PKT_HASH_TYPE_L3),
+	HNS3_RX_PTYPE_ENTRY(25, 0, UNNECESSARY, IPV4, PKT_HASH_TYPE_L4),
+	HNS3_RX_PTYPE_UNUSED_ENTRY(26),
+	HNS3_RX_PTYPE_UNUSED_ENTRY(27),
+	HNS3_RX_PTYPE_UNUSED_ENTRY(28),
+	HNS3_RX_PTYPE_ENTRY(29, 0, COMPLETE, PARSE_FAIL, PKT_HASH_TYPE_NONE),
+	HNS3_RX_PTYPE_ENTRY(30, 0, COMPLETE, PARSE_FAIL, PKT_HASH_TYPE_NONE),
+	HNS3_RX_PTYPE_ENTRY(31, 0, COMPLETE, IPV4, PKT_HASH_TYPE_L3),
+	HNS3_RX_PTYPE_ENTRY(32, 0, COMPLETE, IPV4, PKT_HASH_TYPE_L3),
+	HNS3_RX_PTYPE_ENTRY(33, 1, UNNECESSARY, IPV4, PKT_HASH_TYPE_L4),
+	HNS3_RX_PTYPE_ENTRY(34, 1, UNNECESSARY, IPV4, PKT_HASH_TYPE_L4),
+	HNS3_RX_PTYPE_ENTRY(35, 1, UNNECESSARY, IPV4, PKT_HASH_TYPE_L4),
+	HNS3_RX_PTYPE_ENTRY(36, 0, COMPLETE, IPV4, PKT_HASH_TYPE_L3),
+	HNS3_RX_PTYPE_ENTRY(37, 0, COMPLETE, IPV4, PKT_HASH_TYPE_L3),
+	HNS3_RX_PTYPE_UNUSED_ENTRY(38),
+	HNS3_RX_PTYPE_ENTRY(39, 0, COMPLETE, IPV6, PKT_HASH_TYPE_L3),
+	HNS3_RX_PTYPE_ENTRY(40, 0, COMPLETE, IPV6, PKT_HASH_TYPE_L3),
+	HNS3_RX_PTYPE_ENTRY(41, 1, UNNECESSARY, IPV6, PKT_HASH_TYPE_L4),
+	HNS3_RX_PTYPE_ENTRY(42, 1, UNNECESSARY, IPV6, PKT_HASH_TYPE_L4),
+	HNS3_RX_PTYPE_ENTRY(43, 1, UNNECESSARY, IPV6, PKT_HASH_TYPE_L4),
+	HNS3_RX_PTYPE_ENTRY(44, 0, COMPLETE, IPV6, PKT_HASH_TYPE_L3),
+	HNS3_RX_PTYPE_ENTRY(45, 0, COMPLETE, IPV6, PKT_HASH_TYPE_L3),
+	HNS3_RX_PTYPE_UNUSED_ENTRY(46),
+	HNS3_RX_PTYPE_UNUSED_ENTRY(47),
+	HNS3_RX_PTYPE_UNUSED_ENTRY(48),
+	HNS3_RX_PTYPE_UNUSED_ENTRY(49),
+	HNS3_RX_PTYPE_UNUSED_ENTRY(50),
+	HNS3_RX_PTYPE_UNUSED_ENTRY(51),
+	HNS3_RX_PTYPE_UNUSED_ENTRY(52),
+	HNS3_RX_PTYPE_UNUSED_ENTRY(53),
+	HNS3_RX_PTYPE_UNUSED_ENTRY(54),
+	HNS3_RX_PTYPE_UNUSED_ENTRY(55),
+	HNS3_RX_PTYPE_UNUSED_ENTRY(56),
+	HNS3_RX_PTYPE_UNUSED_ENTRY(57),
+	HNS3_RX_PTYPE_UNUSED_ENTRY(58),
+	HNS3_RX_PTYPE_UNUSED_ENTRY(59),
+	HNS3_RX_PTYPE_UNUSED_ENTRY(60),
+	HNS3_RX_PTYPE_UNUSED_ENTRY(61),
+	HNS3_RX_PTYPE_UNUSED_ENTRY(62),
+	HNS3_RX_PTYPE_UNUSED_ENTRY(63),
+	HNS3_RX_PTYPE_UNUSED_ENTRY(64),
+	HNS3_RX_PTYPE_UNUSED_ENTRY(65),
+	HNS3_RX_PTYPE_UNUSED_ENTRY(66),
+	HNS3_RX_PTYPE_UNUSED_ENTRY(67),
+	HNS3_RX_PTYPE_UNUSED_ENTRY(68),
+	HNS3_RX_PTYPE_UNUSED_ENTRY(69),
+	HNS3_RX_PTYPE_UNUSED_ENTRY(70),
+	HNS3_RX_PTYPE_UNUSED_ENTRY(71),
+	HNS3_RX_PTYPE_UNUSED_ENTRY(72),
+	HNS3_RX_PTYPE_UNUSED_ENTRY(73),
+	HNS3_RX_PTYPE_UNUSED_ENTRY(74),
+	HNS3_RX_PTYPE_UNUSED_ENTRY(75),
+	HNS3_RX_PTYPE_UNUSED_ENTRY(76),
+	HNS3_RX_PTYPE_UNUSED_ENTRY(77),
+	HNS3_RX_PTYPE_UNUSED_ENTRY(78),
+	HNS3_RX_PTYPE_UNUSED_ENTRY(79),
+	HNS3_RX_PTYPE_UNUSED_ENTRY(80),
+	HNS3_RX_PTYPE_UNUSED_ENTRY(81),
+	HNS3_RX_PTYPE_UNUSED_ENTRY(82),
+	HNS3_RX_PTYPE_UNUSED_ENTRY(83),
+	HNS3_RX_PTYPE_UNUSED_ENTRY(84),
+	HNS3_RX_PTYPE_UNUSED_ENTRY(85),
+	HNS3_RX_PTYPE_UNUSED_ENTRY(86),
+	HNS3_RX_PTYPE_UNUSED_ENTRY(87),
+	HNS3_RX_PTYPE_UNUSED_ENTRY(88),
+	HNS3_RX_PTYPE_UNUSED_ENTRY(89),
+	HNS3_RX_PTYPE_UNUSED_ENTRY(90),
+	HNS3_RX_PTYPE_UNUSED_ENTRY(91),
+	HNS3_RX_PTYPE_UNUSED_ENTRY(92),
+	HNS3_RX_PTYPE_UNUSED_ENTRY(93),
+	HNS3_RX_PTYPE_UNUSED_ENTRY(94),
+	HNS3_RX_PTYPE_UNUSED_ENTRY(95),
+	HNS3_RX_PTYPE_UNUSED_ENTRY(96),
+	HNS3_RX_PTYPE_UNUSED_ENTRY(97),
+	HNS3_RX_PTYPE_UNUSED_ENTRY(98),
+	HNS3_RX_PTYPE_UNUSED_ENTRY(99),
+	HNS3_RX_PTYPE_UNUSED_ENTRY(100),
+	HNS3_RX_PTYPE_UNUSED_ENTRY(101),
+	HNS3_RX_PTYPE_UNUSED_ENTRY(102),
+	HNS3_RX_PTYPE_UNUSED_ENTRY(103),
+	HNS3_RX_PTYPE_UNUSED_ENTRY(104),
+	HNS3_RX_PTYPE_UNUSED_ENTRY(105),
+	HNS3_RX_PTYPE_UNUSED_ENTRY(106),
+	HNS3_RX_PTYPE_UNUSED_ENTRY(107),
+	HNS3_RX_PTYPE_UNUSED_ENTRY(108),
+	HNS3_RX_PTYPE_UNUSED_ENTRY(109),
+	HNS3_RX_PTYPE_UNUSED_ENTRY(110),
+	HNS3_RX_PTYPE_ENTRY(111, 0, COMPLETE, IPV6, PKT_HASH_TYPE_L3),
+	HNS3_RX_PTYPE_ENTRY(112, 0, COMPLETE, IPV6, PKT_HASH_TYPE_L3),
+	HNS3_RX_PTYPE_ENTRY(113, 0, UNNECESSARY, IPV6, PKT_HASH_TYPE_L4),
+	HNS3_RX_PTYPE_ENTRY(114, 0, UNNECESSARY, IPV6, PKT_HASH_TYPE_L4),
+	HNS3_RX_PTYPE_ENTRY(115, 0, NONE, IPV6, PKT_HASH_TYPE_L3),
+	HNS3_RX_PTYPE_ENTRY(116, 0, UNNECESSARY, IPV6, PKT_HASH_TYPE_L4),
+	HNS3_RX_PTYPE_ENTRY(117, 0, NONE, IPV6, PKT_HASH_TYPE_L3),
+	HNS3_RX_PTYPE_ENTRY(118, 0, NONE, IPV6, PKT_HASH_TYPE_L3),
+	HNS3_RX_PTYPE_ENTRY(119, 0, UNNECESSARY, IPV6, PKT_HASH_TYPE_L4),
+	HNS3_RX_PTYPE_UNUSED_ENTRY(120),
+	HNS3_RX_PTYPE_UNUSED_ENTRY(121),
+	HNS3_RX_PTYPE_UNUSED_ENTRY(122),
+	HNS3_RX_PTYPE_ENTRY(123, 0, COMPLETE, PARSE_FAIL, PKT_HASH_TYPE_NONE),
+	HNS3_RX_PTYPE_ENTRY(124, 0, COMPLETE, PARSE_FAIL, PKT_HASH_TYPE_NONE),
+	HNS3_RX_PTYPE_ENTRY(125, 0, COMPLETE, IPV4, PKT_HASH_TYPE_L3),
+	HNS3_RX_PTYPE_ENTRY(126, 0, COMPLETE, IPV4, PKT_HASH_TYPE_L3),
+	HNS3_RX_PTYPE_ENTRY(127, 1, UNNECESSARY, IPV4, PKT_HASH_TYPE_L4),
+	HNS3_RX_PTYPE_ENTRY(128, 1, UNNECESSARY, IPV4, PKT_HASH_TYPE_L4),
+	HNS3_RX_PTYPE_ENTRY(129, 1, UNNECESSARY, IPV4, PKT_HASH_TYPE_L4),
+	HNS3_RX_PTYPE_ENTRY(130, 0, COMPLETE, IPV4, PKT_HASH_TYPE_L3),
+	HNS3_RX_PTYPE_ENTRY(131, 0, COMPLETE, IPV4, PKT_HASH_TYPE_L3),
+	HNS3_RX_PTYPE_UNUSED_ENTRY(132),
+	HNS3_RX_PTYPE_ENTRY(133, 0, COMPLETE, IPV6, PKT_HASH_TYPE_L3),
+	HNS3_RX_PTYPE_ENTRY(134, 0, COMPLETE, IPV6, PKT_HASH_TYPE_L3),
+	HNS3_RX_PTYPE_ENTRY(135, 1, UNNECESSARY, IPV6, PKT_HASH_TYPE_L4),
+	HNS3_RX_PTYPE_ENTRY(136, 1, UNNECESSARY, IPV6, PKT_HASH_TYPE_L4),
+	HNS3_RX_PTYPE_ENTRY(137, 1, UNNECESSARY, IPV6, PKT_HASH_TYPE_L4),
+	HNS3_RX_PTYPE_ENTRY(138, 0, COMPLETE, IPV6, PKT_HASH_TYPE_L3),
+	HNS3_RX_PTYPE_ENTRY(139, 0, COMPLETE, IPV6, PKT_HASH_TYPE_L3),
+	HNS3_RX_PTYPE_UNUSED_ENTRY(140),
+	HNS3_RX_PTYPE_UNUSED_ENTRY(141),
+	HNS3_RX_PTYPE_UNUSED_ENTRY(142),
+	HNS3_RX_PTYPE_UNUSED_ENTRY(143),
+	HNS3_RX_PTYPE_UNUSED_ENTRY(144),
+	HNS3_RX_PTYPE_UNUSED_ENTRY(145),
+	HNS3_RX_PTYPE_UNUSED_ENTRY(146),
+	HNS3_RX_PTYPE_UNUSED_ENTRY(147),
+	HNS3_RX_PTYPE_UNUSED_ENTRY(148),
+	HNS3_RX_PTYPE_UNUSED_ENTRY(149),
+	HNS3_RX_PTYPE_UNUSED_ENTRY(150),
+	HNS3_RX_PTYPE_UNUSED_ENTRY(151),
+	HNS3_RX_PTYPE_UNUSED_ENTRY(152),
+	HNS3_RX_PTYPE_UNUSED_ENTRY(153),
+	HNS3_RX_PTYPE_UNUSED_ENTRY(154),
+	HNS3_RX_PTYPE_UNUSED_ENTRY(155),
+	HNS3_RX_PTYPE_UNUSED_ENTRY(156),
+	HNS3_RX_PTYPE_UNUSED_ENTRY(157),
+	HNS3_RX_PTYPE_UNUSED_ENTRY(158),
+	HNS3_RX_PTYPE_UNUSED_ENTRY(159),
+	HNS3_RX_PTYPE_UNUSED_ENTRY(160),
+	HNS3_RX_PTYPE_UNUSED_ENTRY(161),
+	HNS3_RX_PTYPE_UNUSED_ENTRY(162),
+	HNS3_RX_PTYPE_UNUSED_ENTRY(163),
+	HNS3_RX_PTYPE_UNUSED_ENTRY(164),
+	HNS3_RX_PTYPE_UNUSED_ENTRY(165),
+	HNS3_RX_PTYPE_UNUSED_ENTRY(166),
+	HNS3_RX_PTYPE_UNUSED_ENTRY(167),
+	HNS3_RX_PTYPE_UNUSED_ENTRY(168),
+	HNS3_RX_PTYPE_UNUSED_ENTRY(169),
+	HNS3_RX_PTYPE_UNUSED_ENTRY(170),
+	HNS3_RX_PTYPE_UNUSED_ENTRY(171),
+	HNS3_RX_PTYPE_UNUSED_ENTRY(172),
+	HNS3_RX_PTYPE_UNUSED_ENTRY(173),
+	HNS3_RX_PTYPE_UNUSED_ENTRY(174),
+	HNS3_RX_PTYPE_UNUSED_ENTRY(175),
+	HNS3_RX_PTYPE_UNUSED_ENTRY(176),
+	HNS3_RX_PTYPE_UNUSED_ENTRY(177),
+	HNS3_RX_PTYPE_UNUSED_ENTRY(178),
+	HNS3_RX_PTYPE_UNUSED_ENTRY(179),
+	HNS3_RX_PTYPE_UNUSED_ENTRY(180),
+	HNS3_RX_PTYPE_UNUSED_ENTRY(181),
+	HNS3_RX_PTYPE_UNUSED_ENTRY(182),
+	HNS3_RX_PTYPE_UNUSED_ENTRY(183),
+	HNS3_RX_PTYPE_UNUSED_ENTRY(184),
+	HNS3_RX_PTYPE_UNUSED_ENTRY(185),
+	HNS3_RX_PTYPE_UNUSED_ENTRY(186),
+	HNS3_RX_PTYPE_UNUSED_ENTRY(187),
+	HNS3_RX_PTYPE_UNUSED_ENTRY(188),
+	HNS3_RX_PTYPE_UNUSED_ENTRY(189),
+	HNS3_RX_PTYPE_UNUSED_ENTRY(190),
+	HNS3_RX_PTYPE_UNUSED_ENTRY(191),
+	HNS3_RX_PTYPE_UNUSED_ENTRY(192),
+	HNS3_RX_PTYPE_UNUSED_ENTRY(193),
+	HNS3_RX_PTYPE_UNUSED_ENTRY(194),
+	HNS3_RX_PTYPE_UNUSED_ENTRY(195),
+	HNS3_RX_PTYPE_UNUSED_ENTRY(196),
+	HNS3_RX_PTYPE_UNUSED_ENTRY(197),
+	HNS3_RX_PTYPE_UNUSED_ENTRY(198),
+	HNS3_RX_PTYPE_UNUSED_ENTRY(199),
+	HNS3_RX_PTYPE_UNUSED_ENTRY(200),
+	HNS3_RX_PTYPE_UNUSED_ENTRY(201),
+	HNS3_RX_PTYPE_UNUSED_ENTRY(202),
+	HNS3_RX_PTYPE_UNUSED_ENTRY(203),
+	HNS3_RX_PTYPE_UNUSED_ENTRY(204),
+	HNS3_RX_PTYPE_UNUSED_ENTRY(205),
+	HNS3_RX_PTYPE_UNUSED_ENTRY(206),
+	HNS3_RX_PTYPE_UNUSED_ENTRY(207),
+	HNS3_RX_PTYPE_UNUSED_ENTRY(208),
+	HNS3_RX_PTYPE_UNUSED_ENTRY(209),
+	HNS3_RX_PTYPE_UNUSED_ENTRY(210),
+	HNS3_RX_PTYPE_UNUSED_ENTRY(211),
+	HNS3_RX_PTYPE_UNUSED_ENTRY(212),
+	HNS3_RX_PTYPE_UNUSED_ENTRY(213),
+	HNS3_RX_PTYPE_UNUSED_ENTRY(214),
+	HNS3_RX_PTYPE_UNUSED_ENTRY(215),
+	HNS3_RX_PTYPE_UNUSED_ENTRY(216),
+	HNS3_RX_PTYPE_UNUSED_ENTRY(217),
+	HNS3_RX_PTYPE_UNUSED_ENTRY(218),
+	HNS3_RX_PTYPE_UNUSED_ENTRY(219),
+	HNS3_RX_PTYPE_UNUSED_ENTRY(220),
+	HNS3_RX_PTYPE_UNUSED_ENTRY(221),
+	HNS3_RX_PTYPE_UNUSED_ENTRY(222),
+	HNS3_RX_PTYPE_UNUSED_ENTRY(223),
+	HNS3_RX_PTYPE_UNUSED_ENTRY(224),
+	HNS3_RX_PTYPE_UNUSED_ENTRY(225),
+	HNS3_RX_PTYPE_UNUSED_ENTRY(226),
+	HNS3_RX_PTYPE_UNUSED_ENTRY(227),
+	HNS3_RX_PTYPE_UNUSED_ENTRY(228),
+	HNS3_RX_PTYPE_UNUSED_ENTRY(229),
+	HNS3_RX_PTYPE_UNUSED_ENTRY(230),
+	HNS3_RX_PTYPE_UNUSED_ENTRY(231),
+	HNS3_RX_PTYPE_UNUSED_ENTRY(232),
+	HNS3_RX_PTYPE_UNUSED_ENTRY(233),
+	HNS3_RX_PTYPE_UNUSED_ENTRY(234),
+	HNS3_RX_PTYPE_UNUSED_ENTRY(235),
+	HNS3_RX_PTYPE_UNUSED_ENTRY(236),
+	HNS3_RX_PTYPE_UNUSED_ENTRY(237),
+	HNS3_RX_PTYPE_UNUSED_ENTRY(238),
+	HNS3_RX_PTYPE_UNUSED_ENTRY(239),
+	HNS3_RX_PTYPE_UNUSED_ENTRY(240),
+	HNS3_RX_PTYPE_UNUSED_ENTRY(241),
+	HNS3_RX_PTYPE_UNUSED_ENTRY(242),
+	HNS3_RX_PTYPE_UNUSED_ENTRY(243),
+	HNS3_RX_PTYPE_UNUSED_ENTRY(244),
+	HNS3_RX_PTYPE_UNUSED_ENTRY(245),
+	HNS3_RX_PTYPE_UNUSED_ENTRY(246),
+	HNS3_RX_PTYPE_UNUSED_ENTRY(247),
+	HNS3_RX_PTYPE_UNUSED_ENTRY(248),
+	HNS3_RX_PTYPE_UNUSED_ENTRY(249),
+	HNS3_RX_PTYPE_UNUSED_ENTRY(250),
+	HNS3_RX_PTYPE_UNUSED_ENTRY(251),
+	HNS3_RX_PTYPE_UNUSED_ENTRY(252),
+	HNS3_RX_PTYPE_UNUSED_ENTRY(253),
+	HNS3_RX_PTYPE_UNUSED_ENTRY(254),
+	HNS3_RX_PTYPE_UNUSED_ENTRY(255),
+};
+
+#define HNS3_INVALID_PTYPE \
+		ARRAY_SIZE(hns3_rx_ptype_tbl)
+
+static irqreturn_t hns3_irq_handle(int irq, void *vector)
+{
+	struct hns3_enet_tqp_vector *tqp_vector = vector;
+
+	napi_schedule_irqoff(&tqp_vector->napi);
+	tqp_vector->event_cnt++;
+
+	return IRQ_HANDLED;
+}
+
+static void hns3_nic_uninit_irq(struct hns3_nic_priv *priv)
+{
+	struct hns3_enet_tqp_vector *tqp_vectors;
+	unsigned int i;
+
+	for (i = 0; i < priv->vector_num; i++) {
+		tqp_vectors = &priv->tqp_vector[i];
+
+		if (tqp_vectors->irq_init_flag != HNS3_VECTOR_INITED)
+			continue;
+
+		/* clear the affinity mask */
+		irq_set_affinity_hint(tqp_vectors->vector_irq, NULL);
+
+		/* release the irq resource */
+		free_irq(tqp_vectors->vector_irq, tqp_vectors);
+		tqp_vectors->irq_init_flag = HNS3_VECTOR_NOT_INITED;
+	}
+}
+
+static int hns3_nic_init_irq(struct hns3_nic_priv *priv)
+{
+	struct hns3_enet_tqp_vector *tqp_vectors;
+	int txrx_int_idx = 0;
+	int rx_int_idx = 0;
+	int tx_int_idx = 0;
+	unsigned int i;
+	int ret;
+
+	for (i = 0; i < priv->vector_num; i++) {
+		tqp_vectors = &priv->tqp_vector[i];
+
+		if (tqp_vectors->irq_init_flag == HNS3_VECTOR_INITED)
+			continue;
+
+		if (tqp_vectors->tx_group.ring && tqp_vectors->rx_group.ring) {
+			snprintf(tqp_vectors->name, HNAE3_INT_NAME_LEN,
+				 "%s-%s-%s-%d", hns3_driver_name,
+				 pci_name(priv->ae_handle->pdev),
+				 "TxRx", txrx_int_idx++);
+			txrx_int_idx++;
+		} else if (tqp_vectors->rx_group.ring) {
+			snprintf(tqp_vectors->name, HNAE3_INT_NAME_LEN,
+				 "%s-%s-%s-%d", hns3_driver_name,
+				 pci_name(priv->ae_handle->pdev),
+				 "Rx", rx_int_idx++);
+		} else if (tqp_vectors->tx_group.ring) {
+			snprintf(tqp_vectors->name, HNAE3_INT_NAME_LEN,
+				 "%s-%s-%s-%d", hns3_driver_name,
+				 pci_name(priv->ae_handle->pdev),
+				 "Tx", tx_int_idx++);
+		} else {
+			/* Skip this unused q_vector */
+			continue;
+		}
+
+		tqp_vectors->name[HNAE3_INT_NAME_LEN - 1] = '\0';
+
+		irq_set_status_flags(tqp_vectors->vector_irq, IRQ_NOAUTOEN);
+		ret = request_irq(tqp_vectors->vector_irq, hns3_irq_handle, 0,
+				  tqp_vectors->name, tqp_vectors);
+		if (ret) {
+			netdev_err(priv->netdev, "request irq(%d) fail\n",
+				   tqp_vectors->vector_irq);
+			hns3_nic_uninit_irq(priv);
+			return ret;
+		}
+
+		irq_set_affinity_hint(tqp_vectors->vector_irq,
+				      &tqp_vectors->affinity_mask);
+
+		tqp_vectors->irq_init_flag = HNS3_VECTOR_INITED;
+	}
+
+	return 0;
+}
+
+static void hns3_mask_vector_irq(struct hns3_enet_tqp_vector *tqp_vector,
+				 u32 mask_en)
+{
+	writel(mask_en, tqp_vector->mask_addr);
+}
+
+static void hns3_irq_enable(struct hns3_enet_tqp_vector *tqp_vector)
+{
+	napi_enable(&tqp_vector->napi);
+	enable_irq(tqp_vector->vector_irq);
+}
+
+static void hns3_irq_disable(struct hns3_enet_tqp_vector *tqp_vector)
+{
+	disable_irq(tqp_vector->vector_irq);
+	napi_disable(&tqp_vector->napi);
+	cancel_work_sync(&tqp_vector->rx_group.dim.work);
+	cancel_work_sync(&tqp_vector->tx_group.dim.work);
+}
+
+void hns3_set_vector_coalesce_rl(struct hns3_enet_tqp_vector *tqp_vector,
+				 u32 rl_value)
+{
+	u32 rl_reg = hns3_rl_usec_to_reg(rl_value);
+
+	/* this defines the configuration for RL (Interrupt Rate Limiter).
+	 * Rl defines rate of interrupts i.e. number of interrupts-per-second
+	 * GL and RL(Rate Limiter) are 2 ways to acheive interrupt coalescing
+	 */
+	if (rl_reg > 0 && !tqp_vector->tx_group.coal.adapt_enable &&
+	    !tqp_vector->rx_group.coal.adapt_enable)
+		/* According to the hardware, the range of rl_reg is
+		 * 0-59 and the unit is 4.
+		 */
+		rl_reg |=  HNS3_INT_RL_ENABLE_MASK;
+
+	writel(rl_reg, tqp_vector->mask_addr + HNS3_VECTOR_RL_OFFSET);
+}
+
+void hns3_set_vector_coalesce_rx_gl(struct hns3_enet_tqp_vector *tqp_vector,
+				    u32 gl_value)
+{
+	u32 new_val;
+
+	if (tqp_vector->rx_group.coal.unit_1us)
+		new_val = gl_value | HNS3_INT_GL_1US;
+	else
+		new_val = hns3_gl_usec_to_reg(gl_value);
+
+	writel(new_val, tqp_vector->mask_addr + HNS3_VECTOR_GL0_OFFSET);
+}
+
+void hns3_set_vector_coalesce_tx_gl(struct hns3_enet_tqp_vector *tqp_vector,
+				    u32 gl_value)
+{
+	u32 new_val;
+
+	if (tqp_vector->tx_group.coal.unit_1us)
+		new_val = gl_value | HNS3_INT_GL_1US;
+	else
+		new_val = hns3_gl_usec_to_reg(gl_value);
+
+	writel(new_val, tqp_vector->mask_addr + HNS3_VECTOR_GL1_OFFSET);
+}
+
+void hns3_set_vector_coalesce_tx_ql(struct hns3_enet_tqp_vector *tqp_vector,
+				    u32 ql_value)
+{
+	writel(ql_value, tqp_vector->mask_addr + HNS3_VECTOR_TX_QL_OFFSET);
+}
+
+void hns3_set_vector_coalesce_rx_ql(struct hns3_enet_tqp_vector *tqp_vector,
+				    u32 ql_value)
+{
+	writel(ql_value, tqp_vector->mask_addr + HNS3_VECTOR_RX_QL_OFFSET);
+}
+
+static void hns3_vector_coalesce_init(struct hns3_enet_tqp_vector *tqp_vector,
+				      struct hns3_nic_priv *priv)
+{
+	struct hns3_enet_coalesce *tx_coal = &tqp_vector->tx_group.coal;
+	struct hns3_enet_coalesce *rx_coal = &tqp_vector->rx_group.coal;
+	struct hnae3_ae_dev *ae_dev = hns3_get_ae_dev(priv->ae_handle);
+	struct hns3_enet_coalesce *ptx_coal = &priv->tx_coal;
+	struct hns3_enet_coalesce *prx_coal = &priv->rx_coal;
+
+	tx_coal->adapt_enable = ptx_coal->adapt_enable;
+	rx_coal->adapt_enable = prx_coal->adapt_enable;
+
+	tx_coal->int_gl = ptx_coal->int_gl;
+	rx_coal->int_gl = prx_coal->int_gl;
+
+	rx_coal->flow_level = prx_coal->flow_level;
+	tx_coal->flow_level = ptx_coal->flow_level;
+
+	/* device version above V3(include V3), GL can configure 1us
+	 * unit, so uses 1us unit.
+	 */
+	if (ae_dev->dev_version >= HNAE3_DEVICE_VERSION_V3) {
+		tx_coal->unit_1us = 1;
+		rx_coal->unit_1us = 1;
+	}
+
+	if (ae_dev->dev_specs.int_ql_max) {
+		tx_coal->ql_enable = 1;
+		rx_coal->ql_enable = 1;
+		tx_coal->int_ql_max = ae_dev->dev_specs.int_ql_max;
+		rx_coal->int_ql_max = ae_dev->dev_specs.int_ql_max;
+		tx_coal->int_ql = ptx_coal->int_ql;
+		rx_coal->int_ql = prx_coal->int_ql;
+	}
+}
+
+static void
+hns3_vector_coalesce_init_hw(struct hns3_enet_tqp_vector *tqp_vector,
+			     struct hns3_nic_priv *priv)
+{
+	struct hns3_enet_coalesce *tx_coal = &tqp_vector->tx_group.coal;
+	struct hns3_enet_coalesce *rx_coal = &tqp_vector->rx_group.coal;
+	struct hnae3_handle *h = priv->ae_handle;
+
+	hns3_set_vector_coalesce_tx_gl(tqp_vector, tx_coal->int_gl);
+	hns3_set_vector_coalesce_rx_gl(tqp_vector, rx_coal->int_gl);
+	hns3_set_vector_coalesce_rl(tqp_vector, h->kinfo.int_rl_setting);
+
+	if (tx_coal->ql_enable)
+		hns3_set_vector_coalesce_tx_ql(tqp_vector, tx_coal->int_ql);
+
+	if (rx_coal->ql_enable)
+		hns3_set_vector_coalesce_rx_ql(tqp_vector, rx_coal->int_ql);
+}
+
+static int hns3_nic_set_real_num_queue(struct net_device *netdev)
+{
+	struct hnae3_handle *h = hns3_get_handle(netdev);
+	struct hnae3_knic_private_info *kinfo = &h->kinfo;
+	struct hnae3_tc_info *tc_info = &kinfo->tc_info;
+	unsigned int queue_size = kinfo->num_tqps;
+	int i, ret;
+
+	if (tc_info->num_tc <= 1 && !tc_info->mqprio_active) {
+		netdev_reset_tc(netdev);
+	} else {
+		ret = netdev_set_num_tc(netdev, tc_info->num_tc);
+		if (ret) {
+			netdev_err(netdev,
+				   "netdev_set_num_tc fail, ret=%d!\n", ret);
+			return ret;
+		}
+
+		for (i = 0; i < tc_info->num_tc; i++)
+			netdev_set_tc_queue(netdev, i, tc_info->tqp_count[i],
+					    tc_info->tqp_offset[i]);
+	}
+
+	ret = netif_set_real_num_tx_queues(netdev, queue_size);
+	if (ret) {
+		netdev_err(netdev,
+			   "netif_set_real_num_tx_queues fail, ret=%d!\n", ret);
+		return ret;
+	}
+
+	ret = netif_set_real_num_rx_queues(netdev, queue_size);
+	if (ret) {
+		netdev_err(netdev,
+			   "netif_set_real_num_rx_queues fail, ret=%d!\n", ret);
+		return ret;
+	}
+
+	return 0;
+}
+
+u16 hns3_get_max_available_channels(struct hnae3_handle *h)
+{
+	u16 alloc_tqps, max_rss_size, rss_size;
+
+	h->ae_algo->ops->get_tqps_and_rss_info(h, &alloc_tqps, &max_rss_size);
+	rss_size = alloc_tqps / h->kinfo.tc_info.num_tc;
+
+	return min_t(u16, rss_size, max_rss_size);
+}
+
+static void hns3_tqp_enable(struct hnae3_queue *tqp)
+{
+	u32 rcb_reg;
+
+	rcb_reg = hns3_read_dev(tqp, HNS3_RING_EN_REG);
+	rcb_reg |= BIT(HNS3_RING_EN_B);
+	hns3_write_dev(tqp, HNS3_RING_EN_REG, rcb_reg);
+}
+
+static void hns3_tqp_disable(struct hnae3_queue *tqp)
+{
+	u32 rcb_reg;
+
+	rcb_reg = hns3_read_dev(tqp, HNS3_RING_EN_REG);
+	rcb_reg &= ~BIT(HNS3_RING_EN_B);
+	hns3_write_dev(tqp, HNS3_RING_EN_REG, rcb_reg);
+}
+
+static void hns3_free_rx_cpu_rmap(struct net_device *netdev)
+{
+#ifdef CONFIG_RFS_ACCEL
+	free_irq_cpu_rmap(netdev->rx_cpu_rmap);
+	netdev->rx_cpu_rmap = NULL;
+#endif
+}
+
+static int hns3_set_rx_cpu_rmap(struct net_device *netdev)
+{
+#ifdef CONFIG_RFS_ACCEL
+	struct hns3_nic_priv *priv = netdev_priv(netdev);
+	struct hns3_enet_tqp_vector *tqp_vector;
+	int i, ret;
+
+	if (!netdev->rx_cpu_rmap) {
+		netdev->rx_cpu_rmap = alloc_irq_cpu_rmap(priv->vector_num);
+		if (!netdev->rx_cpu_rmap)
+			return -ENOMEM;
+	}
+
+	for (i = 0; i < priv->vector_num; i++) {
+		tqp_vector = &priv->tqp_vector[i];
+		ret = irq_cpu_rmap_add(netdev->rx_cpu_rmap,
+				       tqp_vector->vector_irq);
+		if (ret) {
+			hns3_free_rx_cpu_rmap(netdev);
+			return ret;
+		}
+	}
+#endif
+	return 0;
+}
+
+static void hns3_enable_irqs_and_tqps(struct net_device *netdev)
+{
+	struct hns3_nic_priv *priv = netdev_priv(netdev);
+	struct hnae3_handle *h = priv->ae_handle;
+	u16 i;
+
+	for (i = 0; i < priv->vector_num; i++)
+		hns3_irq_enable(&priv->tqp_vector[i]);
+
+	for (i = 0; i < priv->vector_num; i++)
+		hns3_mask_vector_irq(&priv->tqp_vector[i], 1);
+
+	for (i = 0; i < h->kinfo.num_tqps; i++)
+		hns3_tqp_enable(h->kinfo.tqp[i]);
+}
+
+static void hns3_disable_irqs_and_tqps(struct net_device *netdev)
+{
+	struct hns3_nic_priv *priv = netdev_priv(netdev);
+	struct hnae3_handle *h = priv->ae_handle;
+	u16 i;
+
+	for (i = 0; i < h->kinfo.num_tqps; i++)
+		hns3_tqp_disable(h->kinfo.tqp[i]);
+
+	for (i = 0; i < priv->vector_num; i++)
+		hns3_mask_vector_irq(&priv->tqp_vector[i], 0);
+
+	for (i = 0; i < priv->vector_num; i++)
+		hns3_irq_disable(&priv->tqp_vector[i]);
+}
+
+static int hns3_nic_net_up(struct net_device *netdev)
+{
+	struct hns3_nic_priv *priv = netdev_priv(netdev);
+	struct hnae3_handle *h = priv->ae_handle;
+	int ret;
+
+	ret = hns3_nic_reset_all_ring(h);
+	if (ret)
+		return ret;
+
+	clear_bit(HNS3_NIC_STATE_DOWN, &priv->state);
+
+	hns3_enable_irqs_and_tqps(netdev);
+
+	/* start the ae_dev */
+	ret = h->ae_algo->ops->start ? h->ae_algo->ops->start(h) : 0;
+	if (ret) {
+		set_bit(HNS3_NIC_STATE_DOWN, &priv->state);
+		hns3_disable_irqs_and_tqps(netdev);
+	}
+
+	return ret;
+}
+
+static void hns3_config_xps(struct hns3_nic_priv *priv)
+{
+	int i;
+
+	for (i = 0; i < priv->vector_num; i++) {
+		struct hns3_enet_tqp_vector *tqp_vector = &priv->tqp_vector[i];
+		struct hns3_enet_ring *ring = tqp_vector->tx_group.ring;
+
+		while (ring) {
+			int ret;
+
+			ret = netif_set_xps_queue(priv->netdev,
+						  &tqp_vector->affinity_mask,
+						  ring->tqp->tqp_index);
+			if (ret)
+				netdev_warn(priv->netdev,
+					    "set xps queue failed: %d", ret);
+
+			ring = ring->next;
+		}
+	}
+}
+
+static int hns3_nic_net_open(struct net_device *netdev)
+{
+	struct hns3_nic_priv *priv = netdev_priv(netdev);
+	struct hnae3_handle *h = hns3_get_handle(netdev);
+	struct hnae3_knic_private_info *kinfo;
+	int i, ret;
+
+	if (hns3_nic_resetting(netdev))
+		return -EBUSY;
+
+	if (!test_bit(HNS3_NIC_STATE_DOWN, &priv->state)) {
+		netdev_warn(netdev, "net open repeatedly!\n");
+		return 0;
+	}
+
+	netif_carrier_off(netdev);
+
+	ret = hns3_nic_set_real_num_queue(netdev);
+	if (ret)
+		return ret;
+
+	ret = hns3_nic_net_up(netdev);
+	if (ret) {
+		netdev_err(netdev, "net up fail, ret=%d!\n", ret);
+		return ret;
+	}
+
+	kinfo = &h->kinfo;
+	for (i = 0; i < HNAE3_MAX_USER_PRIO; i++)
+		netdev_set_prio_tc_map(netdev, i, kinfo->tc_info.prio_tc[i]);
+
+	if (h->ae_algo->ops->set_timer_task)
+		h->ae_algo->ops->set_timer_task(priv->ae_handle, true);
+
+	hns3_config_xps(priv);
+
+	netif_dbg(h, drv, netdev, "net open\n");
+
+	return 0;
+}
+
+static void hns3_reset_tx_queue(struct hnae3_handle *h)
+{
+	struct net_device *ndev = h->kinfo.netdev;
+	struct hns3_nic_priv *priv = netdev_priv(ndev);
+	struct netdev_queue *dev_queue;
+	u32 i;
+
+	for (i = 0; i < h->kinfo.num_tqps; i++) {
+		dev_queue = netdev_get_tx_queue(ndev,
+						priv->ring[i].queue_index);
+		netdev_tx_reset_queue(dev_queue);
+	}
+}
+
+static void hns3_nic_net_down(struct net_device *netdev)
+{
+	struct hns3_nic_priv *priv = netdev_priv(netdev);
+	const struct hnae3_ae_ops *ops;
+
+	hns3_disable_irqs_and_tqps(netdev);
+
+	/* stop ae_dev */
+	ops = priv->ae_handle->ae_algo->ops;
+	if (ops->stop)
+		ops->stop(priv->ae_handle);
+
+	/* delay ring buffer clearing to hns3_reset_notify_uninit_enet
+	 * during reset process, because driver may not be able
+	 * to disable the ring through firmware when downing the netdev.
+	 */
+	if (!hns3_nic_resetting(netdev))
+		hns3_clear_all_ring(priv->ae_handle, false);
+
+	hns3_reset_tx_queue(priv->ae_handle);
+}
+
+static int hns3_nic_net_stop(struct net_device *netdev)
+{
+	struct hns3_nic_priv *priv = netdev_priv(netdev);
+	struct hnae3_handle *h = hns3_get_handle(netdev);
+
+	if (test_and_set_bit(HNS3_NIC_STATE_DOWN, &priv->state))
+		return 0;
+
+	netif_dbg(h, drv, netdev, "net stop\n");
+
+	if (h->ae_algo->ops->set_timer_task)
+		h->ae_algo->ops->set_timer_task(priv->ae_handle, false);
+
+	netif_carrier_off(netdev);
+	netif_tx_disable(netdev);
+
+	hns3_nic_net_down(netdev);
+
+	return 0;
+}
+
+static int hns3_nic_uc_sync(struct net_device *netdev,
+			    const unsigned char *addr)
+{
+	struct hnae3_handle *h = hns3_get_handle(netdev);
+
+	if (h->ae_algo->ops->add_uc_addr)
+		return h->ae_algo->ops->add_uc_addr(h, addr);
+
+	return 0;
+}
+
+static int hns3_nic_uc_unsync(struct net_device *netdev,
+			      const unsigned char *addr)
+{
+	struct hnae3_handle *h = hns3_get_handle(netdev);
+
+	/* need ignore the request of removing device address, because
+	 * we store the device address and other addresses of uc list
+	 * in the function's mac filter list.
+	 */
+	if (ether_addr_equal(addr, netdev->dev_addr))
+		return 0;
+
+	if (h->ae_algo->ops->rm_uc_addr)
+		return h->ae_algo->ops->rm_uc_addr(h, addr);
+
+	return 0;
+}
+
+static int hns3_nic_mc_sync(struct net_device *netdev,
+			    const unsigned char *addr)
+{
+	struct hnae3_handle *h = hns3_get_handle(netdev);
+
+	if (h->ae_algo->ops->add_mc_addr)
+		return h->ae_algo->ops->add_mc_addr(h, addr);
+
+	return 0;
+}
+
+static int hns3_nic_mc_unsync(struct net_device *netdev,
+			      const unsigned char *addr)
+{
+	struct hnae3_handle *h = hns3_get_handle(netdev);
+
+	if (h->ae_algo->ops->rm_mc_addr)
+		return h->ae_algo->ops->rm_mc_addr(h, addr);
+
+	return 0;
+}
+
+static u8 hns3_get_netdev_flags(struct net_device *netdev)
+{
+	u8 flags = 0;
+
+	if (netdev->flags & IFF_PROMISC)
+		flags = HNAE3_USER_UPE | HNAE3_USER_MPE | HNAE3_BPE;
+	else if (netdev->flags & IFF_ALLMULTI)
+		flags = HNAE3_USER_MPE;
+
+	return flags;
+}
+
+static void hns3_nic_set_rx_mode(struct net_device *netdev)
+{
+	struct hnae3_handle *h = hns3_get_handle(netdev);
+	u8 new_flags;
+
+	new_flags = hns3_get_netdev_flags(netdev);
+
+	__dev_uc_sync(netdev, hns3_nic_uc_sync, hns3_nic_uc_unsync);
+	__dev_mc_sync(netdev, hns3_nic_mc_sync, hns3_nic_mc_unsync);
+
+	/* User mode Promisc mode enable and vlan filtering is disabled to
+	 * let all packets in.
+	 */
+	h->netdev_flags = new_flags;
+	hns3_request_update_promisc_mode(h);
+}
+
+void hns3_request_update_promisc_mode(struct hnae3_handle *handle)
+{
+	const struct hnae3_ae_ops *ops = hns3_get_ops(handle);
+
+	if (ops->request_update_promisc_mode)
+		ops->request_update_promisc_mode(handle);
+}
+
+static u32 hns3_tx_spare_space(struct hns3_enet_ring *ring)
+{
+	struct hns3_tx_spare *tx_spare = ring->tx_spare;
+	u32 ntc, ntu;
+
+	/* This smp_load_acquire() pairs with smp_store_release() in
+	 * hns3_tx_spare_update() called in tx desc cleaning process.
+	 */
+	ntc = smp_load_acquire(&tx_spare->last_to_clean);
+	ntu = tx_spare->next_to_use;
+
+	if (ntc > ntu)
+		return ntc - ntu - 1;
+
+	/* The free tx buffer is divided into two part, so pick the
+	 * larger one.
+	 */
+	return max(ntc, tx_spare->len - ntu) - 1;
+}
+
+static void hns3_tx_spare_update(struct hns3_enet_ring *ring)
+{
+	struct hns3_tx_spare *tx_spare = ring->tx_spare;
+
+	if (!tx_spare ||
+	    tx_spare->last_to_clean == tx_spare->next_to_clean)
+		return;
+
+	/* This smp_store_release() pairs with smp_load_acquire() in
+	 * hns3_tx_spare_space() called in xmit process.
+	 */
+	smp_store_release(&tx_spare->last_to_clean,
+			  tx_spare->next_to_clean);
+}
+
+static bool hns3_can_use_tx_bounce(struct hns3_enet_ring *ring,
+				   struct sk_buff *skb,
+				   u32 space)
+{
+	u32 len = skb->len <= ring->tx_copybreak ? skb->len :
+				skb_headlen(skb);
+
+	if (len > ring->tx_copybreak)
+		return false;
+
+	if (ALIGN(len, dma_get_cache_alignment()) > space) {
+		hns3_ring_stats_update(ring, tx_spare_full);
+		return false;
+	}
+
+	return true;
+}
+
+static bool hns3_can_use_tx_sgl(struct hns3_enet_ring *ring,
+				struct sk_buff *skb,
+				u32 space)
+{
+	if (skb->len <= ring->tx_copybreak || !tx_sgl ||
+	    (!skb_has_frag_list(skb) &&
+	     skb_shinfo(skb)->nr_frags < tx_sgl))
+		return false;
+
+	if (space < HNS3_MAX_SGL_SIZE) {
+		hns3_ring_stats_update(ring, tx_spare_full);
+		return false;
+	}
+
+	return true;
+}
+
+static void hns3_init_tx_spare_buffer(struct hns3_enet_ring *ring)
+{
+	u32 alloc_size = ring->tqp->handle->kinfo.tx_spare_buf_size;
+	struct net_device *netdev = ring_to_netdev(ring);
+	struct hns3_nic_priv *priv = netdev_priv(netdev);
+	struct hns3_tx_spare *tx_spare;
+	struct page *page;
+	dma_addr_t dma;
+	int order;
+
+	if (!alloc_size)
+		return;
+
+	order = get_order(alloc_size);
+	if (order > MAX_PAGE_ORDER) {
+		if (net_ratelimit())
+			dev_warn(ring_to_dev(ring), "failed to allocate tx spare buffer, exceed to max order\n");
+		return;
+	}
+
+	tx_spare = devm_kzalloc(ring_to_dev(ring), sizeof(*tx_spare),
+				GFP_KERNEL);
+	if (!tx_spare) {
+		/* The driver still work without the tx spare buffer */
+		dev_warn(ring_to_dev(ring), "failed to allocate hns3_tx_spare\n");
+		goto devm_kzalloc_error;
+	}
+
+	page = alloc_pages_node(dev_to_node(ring_to_dev(ring)),
+				GFP_KERNEL, order);
+	if (!page) {
+		dev_warn(ring_to_dev(ring), "failed to allocate tx spare pages\n");
+		goto alloc_pages_error;
+	}
+
+	dma = dma_map_page(ring_to_dev(ring), page, 0,
+			   PAGE_SIZE << order, DMA_TO_DEVICE);
+	if (dma_mapping_error(ring_to_dev(ring), dma)) {
+		dev_warn(ring_to_dev(ring), "failed to map pages for tx spare\n");
+		goto dma_mapping_error;
+	}
+
+	tx_spare->dma = dma;
+	tx_spare->buf = page_address(page);
+	tx_spare->len = PAGE_SIZE << order;
+	ring->tx_spare = tx_spare;
+	ring->tx_copybreak = priv->tx_copybreak;
+	return;
+
+dma_mapping_error:
+	put_page(page);
+alloc_pages_error:
+	devm_kfree(ring_to_dev(ring), tx_spare);
+devm_kzalloc_error:
+	ring->tqp->handle->kinfo.tx_spare_buf_size = 0;
+}
+
+/* Use hns3_tx_spare_space() to make sure there is enough buffer
+ * before calling below function to allocate tx buffer.
+ */
+static void *hns3_tx_spare_alloc(struct hns3_enet_ring *ring,
+				 unsigned int size, dma_addr_t *dma,
+				 u32 *cb_len)
+{
+	struct hns3_tx_spare *tx_spare = ring->tx_spare;
+	u32 ntu = tx_spare->next_to_use;
+
+	size = ALIGN(size, dma_get_cache_alignment());
+	*cb_len = size;
+
+	/* Tx spare buffer wraps back here because the end of
+	 * freed tx buffer is not enough.
+	 */
+	if (ntu + size > tx_spare->len) {
+		*cb_len += (tx_spare->len - ntu);
+		ntu = 0;
+	}
+
+	tx_spare->next_to_use = ntu + size;
+	if (tx_spare->next_to_use == tx_spare->len)
+		tx_spare->next_to_use = 0;
+
+	*dma = tx_spare->dma + ntu;
+
+	return tx_spare->buf + ntu;
+}
+
+static void hns3_tx_spare_rollback(struct hns3_enet_ring *ring, u32 len)
+{
+	struct hns3_tx_spare *tx_spare = ring->tx_spare;
+
+	if (len > tx_spare->next_to_use) {
+		len -= tx_spare->next_to_use;
+		tx_spare->next_to_use = tx_spare->len - len;
+	} else {
+		tx_spare->next_to_use -= len;
+	}
+}
+
+static void hns3_tx_spare_reclaim_cb(struct hns3_enet_ring *ring,
+				     struct hns3_desc_cb *cb)
+{
+	struct hns3_tx_spare *tx_spare = ring->tx_spare;
+	u32 ntc = tx_spare->next_to_clean;
+	u32 len = cb->length;
+
+	tx_spare->next_to_clean += len;
+
+	if (tx_spare->next_to_clean >= tx_spare->len) {
+		tx_spare->next_to_clean -= tx_spare->len;
+
+		if (tx_spare->next_to_clean) {
+			ntc = 0;
+			len = tx_spare->next_to_clean;
+		}
+	}
+
+	/* This tx spare buffer is only really reclaimed after calling
+	 * hns3_tx_spare_update(), so it is still safe to use the info in
+	 * the tx buffer to do the dma sync or sg unmapping after
+	 * tx_spare->next_to_clean is moved forword.
+	 */
+	if (cb->type & (DESC_TYPE_BOUNCE_HEAD | DESC_TYPE_BOUNCE_ALL)) {
+		dma_addr_t dma = tx_spare->dma + ntc;
+
+		dma_sync_single_for_cpu(ring_to_dev(ring), dma, len,
+					DMA_TO_DEVICE);
+	} else {
+		struct sg_table *sgt = tx_spare->buf + ntc;
+
+		dma_unmap_sg(ring_to_dev(ring), sgt->sgl, sgt->orig_nents,
+			     DMA_TO_DEVICE);
+	}
+}
+
+static int hns3_set_tso(struct sk_buff *skb, u32 *paylen_fdop_ol4cs,
+			u16 *mss, u32 *type_cs_vlan_tso, u32 *send_bytes)
+{
+	u32 l4_offset, hdr_len;
+	union l3_hdr_info l3;
+	union l4_hdr_info l4;
+	u32 l4_paylen;
+	int ret;
+
+	if (!skb_is_gso(skb))
+		return 0;
+
+	ret = skb_cow_head(skb, 0);
+	if (unlikely(ret < 0))
+		return ret;
+
+	l3.hdr = skb_network_header(skb);
+	l4.hdr = skb_transport_header(skb);
+
+	/* Software should clear the IPv4's checksum field when tso is
+	 * needed.
+	 */
+	if (l3.v4->version == 4)
+		l3.v4->check = 0;
+
+	/* tunnel packet */
+	if (skb_shinfo(skb)->gso_type & (SKB_GSO_GRE |
+					 SKB_GSO_GRE_CSUM |
+					 SKB_GSO_UDP_TUNNEL |
+					 SKB_GSO_UDP_TUNNEL_CSUM)) {
+		/* reset l3&l4 pointers from outer to inner headers */
+		l3.hdr = skb_inner_network_header(skb);
+		l4.hdr = skb_inner_transport_header(skb);
+
+		/* Software should clear the IPv4's checksum field when
+		 * tso is needed.
+		 */
+		if (l3.v4->version == 4)
+			l3.v4->check = 0;
+	}
+
+	/* normal or tunnel packet */
+	l4_offset = l4.hdr - skb->data;
+
+	/* remove payload length from inner pseudo checksum when tso */
+	l4_paylen = skb->len - l4_offset;
+
+	if (skb_shinfo(skb)->gso_type & SKB_GSO_UDP_L4) {
+		hdr_len = sizeof(*l4.udp) + l4_offset;
+		csum_replace_by_diff(&l4.udp->check,
+				     (__force __wsum)htonl(l4_paylen));
+	} else {
+		hdr_len = (l4.tcp->doff << 2) + l4_offset;
+		csum_replace_by_diff(&l4.tcp->check,
+				     (__force __wsum)htonl(l4_paylen));
+	}
+
+	*send_bytes = (skb_shinfo(skb)->gso_segs - 1) * hdr_len + skb->len;
+
+	/* find the txbd field values */
+	*paylen_fdop_ol4cs = skb->len - hdr_len;
+	hns3_set_field(*type_cs_vlan_tso, HNS3_TXD_TSO_B, 1);
+
+	/* offload outer UDP header checksum */
+	if (skb_shinfo(skb)->gso_type & SKB_GSO_UDP_TUNNEL_CSUM)
+		hns3_set_field(*paylen_fdop_ol4cs, HNS3_TXD_OL4CS_B, 1);
+
+	/* get MSS for TSO */
+	*mss = skb_shinfo(skb)->gso_size;
+
+	trace_hns3_tso(skb);
+
+	return 0;
+}
+
+static int hns3_get_l4_protocol(struct sk_buff *skb, u8 *ol4_proto,
+				u8 *il4_proto)
+{
+	union l3_hdr_info l3;
+	unsigned char *l4_hdr;
+	unsigned char *exthdr;
+	u8 l4_proto_tmp;
+	__be16 frag_off;
+
+	/* find outer header point */
+	l3.hdr = skb_network_header(skb);
+	l4_hdr = skb_transport_header(skb);
+
+	if (skb->protocol == htons(ETH_P_IPV6)) {
+		exthdr = l3.hdr + sizeof(*l3.v6);
+		l4_proto_tmp = l3.v6->nexthdr;
+		if (l4_hdr != exthdr)
+			ipv6_skip_exthdr(skb, exthdr - skb->data,
+					 &l4_proto_tmp, &frag_off);
+	} else if (skb->protocol == htons(ETH_P_IP)) {
+		l4_proto_tmp = l3.v4->protocol;
+	} else {
+		return -EINVAL;
+	}
+
+	*ol4_proto = l4_proto_tmp;
+
+	/* tunnel packet */
+	if (!skb->encapsulation) {
+		*il4_proto = 0;
+		return 0;
+	}
+
+	/* find inner header point */
+	l3.hdr = skb_inner_network_header(skb);
+	l4_hdr = skb_inner_transport_header(skb);
+
+	if (l3.v6->version == 6) {
+		exthdr = l3.hdr + sizeof(*l3.v6);
+		l4_proto_tmp = l3.v6->nexthdr;
+		if (l4_hdr != exthdr)
+			ipv6_skip_exthdr(skb, exthdr - skb->data,
+					 &l4_proto_tmp, &frag_off);
+	} else if (l3.v4->version == 4) {
+		l4_proto_tmp = l3.v4->protocol;
+	}
+
+	*il4_proto = l4_proto_tmp;
+
+	return 0;
+}
+
+/* when skb->encapsulation is 0, skb->ip_summed is CHECKSUM_PARTIAL
+ * and it is udp packet, which has a dest port as the IANA assigned.
+ * the hardware is expected to do the checksum offload, but the
+ * hardware will not do the checksum offload when udp dest port is
+ * 4789, 4790 or 6081.
+ */
+static bool hns3_tunnel_csum_bug(struct sk_buff *skb)
+{
+	struct hns3_nic_priv *priv = netdev_priv(skb->dev);
+	struct hnae3_ae_dev *ae_dev = hns3_get_ae_dev(priv->ae_handle);
+	union l4_hdr_info l4;
+
+	/* device version above V3(include V3), the hardware can
+	 * do this checksum offload.
+	 */
+	if (ae_dev->dev_version >= HNAE3_DEVICE_VERSION_V3)
+		return false;
+
+	l4.hdr = skb_transport_header(skb);
+
+	if (!(!skb->encapsulation &&
+	      (l4.udp->dest == htons(IANA_VXLAN_UDP_PORT) ||
+	      l4.udp->dest == htons(GENEVE_UDP_PORT) ||
+	      l4.udp->dest == htons(IANA_VXLAN_GPE_UDP_PORT))))
+		return false;
+
+	return true;
+}
+
+static void hns3_set_outer_l2l3l4(struct sk_buff *skb, u8 ol4_proto,
+				  u32 *ol_type_vlan_len_msec)
+{
+	u32 l2_len, l3_len, l4_len;
+	unsigned char *il2_hdr;
+	union l3_hdr_info l3;
+	union l4_hdr_info l4;
+
+	l3.hdr = skb_network_header(skb);
+	l4.hdr = skb_transport_header(skb);
+
+	/* compute OL2 header size, defined in 2 Bytes */
+	l2_len = l3.hdr - skb->data;
+	hns3_set_field(*ol_type_vlan_len_msec, HNS3_TXD_L2LEN_S, l2_len >> 1);
+
+	/* compute OL3 header size, defined in 4 Bytes */
+	l3_len = l4.hdr - l3.hdr;
+	hns3_set_field(*ol_type_vlan_len_msec, HNS3_TXD_L3LEN_S, l3_len >> 2);
+
+	il2_hdr = skb_inner_mac_header(skb);
+	/* compute OL4 header size, defined in 4 Bytes */
+	l4_len = il2_hdr - l4.hdr;
+	hns3_set_field(*ol_type_vlan_len_msec, HNS3_TXD_L4LEN_S, l4_len >> 2);
+
+	/* define outer network header type */
+	if (skb->protocol == htons(ETH_P_IP)) {
+		if (skb_is_gso(skb))
+			hns3_set_field(*ol_type_vlan_len_msec,
+				       HNS3_TXD_OL3T_S,
+				       HNS3_OL3T_IPV4_CSUM);
+		else
+			hns3_set_field(*ol_type_vlan_len_msec,
+				       HNS3_TXD_OL3T_S,
+				       HNS3_OL3T_IPV4_NO_CSUM);
+	} else if (skb->protocol == htons(ETH_P_IPV6)) {
+		hns3_set_field(*ol_type_vlan_len_msec, HNS3_TXD_OL3T_S,
+			       HNS3_OL3T_IPV6);
+	}
+
+	if (ol4_proto == IPPROTO_UDP)
+		hns3_set_field(*ol_type_vlan_len_msec, HNS3_TXD_TUNTYPE_S,
+			       HNS3_TUN_MAC_IN_UDP);
+	else if (ol4_proto == IPPROTO_GRE)
+		hns3_set_field(*ol_type_vlan_len_msec, HNS3_TXD_TUNTYPE_S,
+			       HNS3_TUN_NVGRE);
+}
+
+static void hns3_set_l3_type(struct sk_buff *skb, union l3_hdr_info l3,
+			     u32 *type_cs_vlan_tso)
+{
+	if (l3.v4->version == 4) {
+		hns3_set_field(*type_cs_vlan_tso, HNS3_TXD_L3T_S,
+			       HNS3_L3T_IPV4);
+
+		/* the stack computes the IP header already, the only time we
+		 * need the hardware to recompute it is in the case of TSO.
+		 */
+		if (skb_is_gso(skb))
+			hns3_set_field(*type_cs_vlan_tso, HNS3_TXD_L3CS_B, 1);
+	} else if (l3.v6->version == 6) {
+		hns3_set_field(*type_cs_vlan_tso, HNS3_TXD_L3T_S,
+			       HNS3_L3T_IPV6);
+	}
+}
+
+static int hns3_set_l4_csum_length(struct sk_buff *skb, union l4_hdr_info l4,
+				   u32 l4_proto, u32 *type_cs_vlan_tso)
+{
+	/* compute inner(/normal) L4 header size, defined in 4 Bytes */
+	switch (l4_proto) {
+	case IPPROTO_TCP:
+		hns3_set_field(*type_cs_vlan_tso, HNS3_TXD_L4CS_B, 1);
+		hns3_set_field(*type_cs_vlan_tso, HNS3_TXD_L4T_S,
+			       HNS3_L4T_TCP);
+		hns3_set_field(*type_cs_vlan_tso, HNS3_TXD_L4LEN_S,
+			       l4.tcp->doff);
+		break;
+	case IPPROTO_UDP:
+		if (hns3_tunnel_csum_bug(skb)) {
+			int ret = skb_put_padto(skb, HNS3_MIN_TUN_PKT_LEN);
+
+			return ret ? ret : skb_checksum_help(skb);
+		}
+
+		hns3_set_field(*type_cs_vlan_tso, HNS3_TXD_L4CS_B, 1);
+		hns3_set_field(*type_cs_vlan_tso, HNS3_TXD_L4T_S,
+			       HNS3_L4T_UDP);
+		hns3_set_field(*type_cs_vlan_tso, HNS3_TXD_L4LEN_S,
+			       (sizeof(struct udphdr) >> 2));
+		break;
+	case IPPROTO_SCTP:
+		hns3_set_field(*type_cs_vlan_tso, HNS3_TXD_L4CS_B, 1);
+		hns3_set_field(*type_cs_vlan_tso, HNS3_TXD_L4T_S,
+			       HNS3_L4T_SCTP);
+		hns3_set_field(*type_cs_vlan_tso, HNS3_TXD_L4LEN_S,
+			       (sizeof(struct sctphdr) >> 2));
+		break;
+	default:
+		/* drop the skb tunnel packet if hardware don't support,
+		 * because hardware can't calculate csum when TSO.
+		 */
+		if (skb_is_gso(skb))
+			return -EDOM;
+
+		/* the stack computes the IP header already,
+		 * driver calculate l4 checksum when not TSO.
+		 */
+		return skb_checksum_help(skb);
+	}
+
+	return 0;
+}
+
+static int hns3_set_l2l3l4(struct sk_buff *skb, u8 ol4_proto,
+			   u8 il4_proto, u32 *type_cs_vlan_tso,
+			   u32 *ol_type_vlan_len_msec)
+{
+	unsigned char *l2_hdr = skb->data;
+	u32 l4_proto = ol4_proto;
+	union l4_hdr_info l4;
+	union l3_hdr_info l3;
+	u32 l2_len, l3_len;
+
+	l4.hdr = skb_transport_header(skb);
+	l3.hdr = skb_network_header(skb);
+
+	/* handle encapsulation skb */
+	if (skb->encapsulation) {
+		/* If this is a not UDP/GRE encapsulation skb */
+		if (!(ol4_proto == IPPROTO_UDP || ol4_proto == IPPROTO_GRE)) {
+			/* drop the skb tunnel packet if hardware don't support,
+			 * because hardware can't calculate csum when TSO.
+			 */
+			if (skb_is_gso(skb))
+				return -EDOM;
+
+			/* the stack computes the IP header already,
+			 * driver calculate l4 checksum when not TSO.
+			 */
+			return skb_checksum_help(skb);
+		}
+
+		hns3_set_outer_l2l3l4(skb, ol4_proto, ol_type_vlan_len_msec);
+
+		/* switch to inner header */
+		l2_hdr = skb_inner_mac_header(skb);
+		l3.hdr = skb_inner_network_header(skb);
+		l4.hdr = skb_inner_transport_header(skb);
+		l4_proto = il4_proto;
+	}
+
+	hns3_set_l3_type(skb, l3, type_cs_vlan_tso);
+
+	/* compute inner(/normal) L2 header size, defined in 2 Bytes */
+	l2_len = l3.hdr - l2_hdr;
+	hns3_set_field(*type_cs_vlan_tso, HNS3_TXD_L2LEN_S, l2_len >> 1);
+
+	/* compute inner(/normal) L3 header size, defined in 4 Bytes */
+	l3_len = l4.hdr - l3.hdr;
+	hns3_set_field(*type_cs_vlan_tso, HNS3_TXD_L3LEN_S, l3_len >> 2);
+
+	return hns3_set_l4_csum_length(skb, l4, l4_proto, type_cs_vlan_tso);
+}
+
+static int hns3_handle_vtags(struct hns3_enet_ring *tx_ring,
+			     struct sk_buff *skb)
+{
+	struct hnae3_handle *handle = tx_ring->tqp->handle;
+	struct hnae3_ae_dev *ae_dev;
+	struct vlan_ethhdr *vhdr;
+	int rc;
+
+	if (!(skb->protocol == htons(ETH_P_8021Q) ||
+	      skb_vlan_tag_present(skb)))
+		return 0;
+
+	/* For HW limitation on HNAE3_DEVICE_VERSION_V2, if port based insert
+	 * VLAN enabled, only one VLAN header is allowed in skb, otherwise it
+	 * will cause RAS error.
+	 */
+	ae_dev = hns3_get_ae_dev(handle);
+	if (unlikely(skb_vlan_tagged_multi(skb) &&
+		     ae_dev->dev_version <= HNAE3_DEVICE_VERSION_V2 &&
+		     handle->port_base_vlan_state ==
+		     HNAE3_PORT_BASE_VLAN_ENABLE))
+		return -EINVAL;
+
+	if (skb->protocol == htons(ETH_P_8021Q) &&
+	    !(handle->kinfo.netdev->features & NETIF_F_HW_VLAN_CTAG_TX)) {
+		/* When HW VLAN acceleration is turned off, and the stack
+		 * sets the protocol to 802.1q, the driver just need to
+		 * set the protocol to the encapsulated ethertype.
+		 */
+		skb->protocol = vlan_get_protocol(skb);
+		return 0;
+	}
+
+	if (skb_vlan_tag_present(skb)) {
+		/* Based on hw strategy, use out_vtag in two layer tag case,
+		 * and use inner_vtag in one tag case.
+		 */
+		if (skb->protocol == htons(ETH_P_8021Q) &&
+		    handle->port_base_vlan_state ==
+		    HNAE3_PORT_BASE_VLAN_DISABLE)
+			rc = HNS3_OUTER_VLAN_TAG;
+		else
+			rc = HNS3_INNER_VLAN_TAG;
+
+		skb->protocol = vlan_get_protocol(skb);
+		return rc;
+	}
+
+	rc = skb_cow_head(skb, 0);
+	if (unlikely(rc < 0))
+		return rc;
+
+	vhdr = skb_vlan_eth_hdr(skb);
+	vhdr->h_vlan_TCI |= cpu_to_be16((skb->priority << VLAN_PRIO_SHIFT)
+					 & VLAN_PRIO_MASK);
+
+	skb->protocol = vlan_get_protocol(skb);
+	return 0;
+}
+
+/* check if the hardware is capable of checksum offloading */
+static bool hns3_check_hw_tx_csum(struct sk_buff *skb)
+{
+	struct hns3_nic_priv *priv = netdev_priv(skb->dev);
+
+	/* Kindly note, due to backward compatibility of the TX descriptor,
+	 * HW checksum of the non-IP packets and GSO packets is handled at
+	 * different place in the following code
+	 */
+	if (skb_csum_is_sctp(skb) || skb_is_gso(skb) ||
+	    !test_bit(HNS3_NIC_STATE_HW_TX_CSUM_ENABLE, &priv->state))
+		return false;
+
+	return true;
+}
+
+struct hns3_desc_param {
+	u32 paylen_ol4cs;
+	u32 ol_type_vlan_len_msec;
+	u32 type_cs_vlan_tso;
+	u16 mss_hw_csum;
+	u16 inner_vtag;
+	u16 out_vtag;
+};
+
+static void hns3_init_desc_data(struct sk_buff *skb, struct hns3_desc_param *pa)
+{
+	pa->paylen_ol4cs = skb->len;
+	pa->ol_type_vlan_len_msec = 0;
+	pa->type_cs_vlan_tso = 0;
+	pa->mss_hw_csum = 0;
+	pa->inner_vtag = 0;
+	pa->out_vtag = 0;
+}
+
+static int hns3_handle_vlan_info(struct hns3_enet_ring *ring,
+				 struct sk_buff *skb,
+				 struct hns3_desc_param *param)
+{
+	int ret;
+
+	ret = hns3_handle_vtags(ring, skb);
+	if (unlikely(ret < 0)) {
+		hns3_ring_stats_update(ring, tx_vlan_err);
+		return ret;
+	} else if (ret == HNS3_INNER_VLAN_TAG) {
+		param->inner_vtag = skb_vlan_tag_get(skb);
+		param->inner_vtag |= (skb->priority << VLAN_PRIO_SHIFT) &
+				VLAN_PRIO_MASK;
+		hns3_set_field(param->type_cs_vlan_tso, HNS3_TXD_VLAN_B, 1);
+	} else if (ret == HNS3_OUTER_VLAN_TAG) {
+		param->out_vtag = skb_vlan_tag_get(skb);
+		param->out_vtag |= (skb->priority << VLAN_PRIO_SHIFT) &
+				VLAN_PRIO_MASK;
+		hns3_set_field(param->ol_type_vlan_len_msec, HNS3_TXD_OVLAN_B,
+			       1);
+	}
+	return 0;
+}
+
+static int hns3_handle_csum_partial(struct hns3_enet_ring *ring,
+				    struct sk_buff *skb,
+				    struct hns3_desc_cb *desc_cb,
+				    struct hns3_desc_param *param)
+{
+	u8 ol4_proto, il4_proto;
+	int ret;
+
+	if (hns3_check_hw_tx_csum(skb)) {
+		/* set checksum start and offset, defined in 2 Bytes */
+		hns3_set_field(param->type_cs_vlan_tso, HNS3_TXD_CSUM_START_S,
+			       skb_checksum_start_offset(skb) >> 1);
+		hns3_set_field(param->ol_type_vlan_len_msec,
+			       HNS3_TXD_CSUM_OFFSET_S,
+			       skb->csum_offset >> 1);
+		param->mss_hw_csum |= BIT(HNS3_TXD_HW_CS_B);
+		return 0;
+	}
+
+	skb_reset_mac_len(skb);
+
+	ret = hns3_get_l4_protocol(skb, &ol4_proto, &il4_proto);
+	if (unlikely(ret < 0)) {
+		hns3_ring_stats_update(ring, tx_l4_proto_err);
+		return ret;
+	}
+
+	ret = hns3_set_l2l3l4(skb, ol4_proto, il4_proto,
+			      &param->type_cs_vlan_tso,
+			      &param->ol_type_vlan_len_msec);
+	if (unlikely(ret < 0)) {
+		hns3_ring_stats_update(ring, tx_l2l3l4_err);
+		return ret;
+	}
+
+	ret = hns3_set_tso(skb, &param->paylen_ol4cs, &param->mss_hw_csum,
+			   &param->type_cs_vlan_tso, &desc_cb->send_bytes);
+	if (unlikely(ret < 0)) {
+		hns3_ring_stats_update(ring, tx_tso_err);
+		return ret;
+	}
+	return 0;
+}
+
+static int hns3_fill_skb_desc(struct hns3_enet_ring *ring,
+			      struct sk_buff *skb, struct hns3_desc *desc,
+			      struct hns3_desc_cb *desc_cb)
+{
+	struct hns3_desc_param param;
+	int ret;
+
+	hns3_init_desc_data(skb, &param);
+	ret = hns3_handle_vlan_info(ring, skb, &param);
+	if (unlikely(ret < 0))
+		return ret;
+
+	desc_cb->send_bytes = skb->len;
+
+	if (skb->ip_summed == CHECKSUM_PARTIAL) {
+		ret = hns3_handle_csum_partial(ring, skb, desc_cb, &param);
+		if (ret)
+			return ret;
+	}
+
+	/* Set txbd */
+	desc->tx.ol_type_vlan_len_msec =
+		cpu_to_le32(param.ol_type_vlan_len_msec);
+	desc->tx.type_cs_vlan_tso_len = cpu_to_le32(param.type_cs_vlan_tso);
+	desc->tx.paylen_ol4cs = cpu_to_le32(param.paylen_ol4cs);
+	desc->tx.mss_hw_csum = cpu_to_le16(param.mss_hw_csum);
+	desc->tx.vlan_tag = cpu_to_le16(param.inner_vtag);
+	desc->tx.outer_vlan_tag = cpu_to_le16(param.out_vtag);
+
+	return 0;
+}
+
+static int hns3_fill_desc(struct hns3_enet_ring *ring, dma_addr_t dma,
+			  unsigned int size)
+{
+#define HNS3_LIKELY_BD_NUM	1
+
+	struct hns3_desc *desc = &ring->desc[ring->next_to_use];
+	unsigned int frag_buf_num, k;
+	int sizeoflast;
+
+	if (likely(size <= HNS3_MAX_BD_SIZE)) {
+		desc->addr = cpu_to_le64(dma);
+		desc->tx.send_size = cpu_to_le16(size);
+		desc->tx.bdtp_fe_sc_vld_ra_ri =
+			cpu_to_le16(BIT(HNS3_TXD_VLD_B));
+
+		trace_hns3_tx_desc(ring, ring->next_to_use);
+		ring_ptr_move_fw(ring, next_to_use);
+		return HNS3_LIKELY_BD_NUM;
+	}
+
+	frag_buf_num = hns3_tx_bd_count(size);
+	sizeoflast = size % HNS3_MAX_BD_SIZE;
+	sizeoflast = sizeoflast ? sizeoflast : HNS3_MAX_BD_SIZE;
+
+	/* When frag size is bigger than hardware limit, split this frag */
+	for (k = 0; k < frag_buf_num; k++) {
+		/* now, fill the descriptor */
+		desc->addr = cpu_to_le64(dma + HNS3_MAX_BD_SIZE * k);
+		desc->tx.send_size = cpu_to_le16((k == frag_buf_num - 1) ?
+				     (u16)sizeoflast : (u16)HNS3_MAX_BD_SIZE);
+		desc->tx.bdtp_fe_sc_vld_ra_ri =
+				cpu_to_le16(BIT(HNS3_TXD_VLD_B));
+
+		trace_hns3_tx_desc(ring, ring->next_to_use);
+		/* move ring pointer to next */
+		ring_ptr_move_fw(ring, next_to_use);
+
+		desc = &ring->desc[ring->next_to_use];
+	}
+
+	return frag_buf_num;
+}
+
+static int hns3_map_and_fill_desc(struct hns3_enet_ring *ring, void *priv,
+				  unsigned int type)
+{
+	struct hns3_desc_cb *desc_cb = &ring->desc_cb[ring->next_to_use];
+	struct device *dev = ring_to_dev(ring);
+	unsigned int size;
+	dma_addr_t dma;
+
+	if (type & (DESC_TYPE_FRAGLIST_SKB | DESC_TYPE_SKB)) {
+		struct sk_buff *skb = (struct sk_buff *)priv;
+
+		size = skb_headlen(skb);
+		if (!size)
+			return 0;
+
+		dma = dma_map_single(dev, skb->data, size, DMA_TO_DEVICE);
+	} else if (type & DESC_TYPE_BOUNCE_HEAD) {
+		/* Head data has been filled in hns3_handle_tx_bounce(),
+		 * just return 0 here.
+		 */
+		return 0;
+	} else {
+		skb_frag_t *frag = (skb_frag_t *)priv;
+
+		size = skb_frag_size(frag);
+		if (!size)
+			return 0;
+
+		dma = skb_frag_dma_map(dev, frag, 0, size, DMA_TO_DEVICE);
+	}
+
+	if (unlikely(dma_mapping_error(dev, dma))) {
+		hns3_ring_stats_update(ring, sw_err_cnt);
+		return -ENOMEM;
+	}
+
+	desc_cb->priv = priv;
+	desc_cb->length = size;
+	desc_cb->dma = dma;
+	desc_cb->type = type;
+
+	return hns3_fill_desc(ring, dma, size);
+}
+
+static unsigned int hns3_skb_bd_num(struct sk_buff *skb, unsigned int *bd_size,
+				    unsigned int bd_num)
+{
+	unsigned int size;
+	int i;
+
+	size = skb_headlen(skb);
+	while (size > HNS3_MAX_BD_SIZE) {
+		bd_size[bd_num++] = HNS3_MAX_BD_SIZE;
+		size -= HNS3_MAX_BD_SIZE;
+
+		if (bd_num > HNS3_MAX_TSO_BD_NUM)
+			return bd_num;
+	}
+
+	if (size) {
+		bd_size[bd_num++] = size;
+		if (bd_num > HNS3_MAX_TSO_BD_NUM)
+			return bd_num;
+	}
+
+	for (i = 0; i < skb_shinfo(skb)->nr_frags; i++) {
+		skb_frag_t *frag = &skb_shinfo(skb)->frags[i];
+		size = skb_frag_size(frag);
+		if (!size)
+			continue;
+
+		while (size > HNS3_MAX_BD_SIZE) {
+			bd_size[bd_num++] = HNS3_MAX_BD_SIZE;
+			size -= HNS3_MAX_BD_SIZE;
+
+			if (bd_num > HNS3_MAX_TSO_BD_NUM)
+				return bd_num;
+		}
+
+		bd_size[bd_num++] = size;
+		if (bd_num > HNS3_MAX_TSO_BD_NUM)
+			return bd_num;
+	}
+
+	return bd_num;
+}
+
+static unsigned int hns3_tx_bd_num(struct sk_buff *skb, unsigned int *bd_size,
+				   u8 max_non_tso_bd_num, unsigned int bd_num,
+				   unsigned int recursion_level)
+{
+#define HNS3_MAX_RECURSION_LEVEL	24
+
+	struct sk_buff *frag_skb;
+
+	/* If the total len is within the max bd limit */
+	if (likely(skb->len <= HNS3_MAX_BD_SIZE && !recursion_level &&
+		   !skb_has_frag_list(skb) &&
+		   skb_shinfo(skb)->nr_frags < max_non_tso_bd_num))
+		return skb_shinfo(skb)->nr_frags + 1U;
+
+	if (unlikely(recursion_level >= HNS3_MAX_RECURSION_LEVEL))
+		return UINT_MAX;
+
+	bd_num = hns3_skb_bd_num(skb, bd_size, bd_num);
+	if (!skb_has_frag_list(skb) || bd_num > HNS3_MAX_TSO_BD_NUM)
+		return bd_num;
+
+	skb_walk_frags(skb, frag_skb) {
+		bd_num = hns3_tx_bd_num(frag_skb, bd_size, max_non_tso_bd_num,
+					bd_num, recursion_level + 1);
+		if (bd_num > HNS3_MAX_TSO_BD_NUM)
+			return bd_num;
+	}
+
+	return bd_num;
+}
+
+static unsigned int hns3_gso_hdr_len(struct sk_buff *skb)
+{
+	if (!skb->encapsulation)
+		return skb_tcp_all_headers(skb);
+
+	return skb_inner_tcp_all_headers(skb);
+}
+
+/* HW need every continuous max_non_tso_bd_num buffer data to be larger
+ * than MSS, we simplify it by ensuring skb_headlen + the first continuous
+ * max_non_tso_bd_num - 1 frags to be larger than gso header len + mss,
+ * and the remaining continuous max_non_tso_bd_num - 1 frags to be larger
+ * than MSS except the last max_non_tso_bd_num - 1 frags.
+ */
+static bool hns3_skb_need_linearized(struct sk_buff *skb, unsigned int *bd_size,
+				     unsigned int bd_num, u8 max_non_tso_bd_num)
+{
+	unsigned int tot_len = 0;
+	unsigned int i;
+
+	for (i = 0; i < max_non_tso_bd_num - 1U; i++)
+		tot_len += bd_size[i];
+
+	/* ensure the first max_non_tso_bd_num frags is greater than
+	 * mss + header
+	 */
+	if (tot_len + bd_size[max_non_tso_bd_num - 1U] <
+	    skb_shinfo(skb)->gso_size + hns3_gso_hdr_len(skb))
+		return true;
+
+	/* ensure every continuous max_non_tso_bd_num - 1 buffer is greater
+	 * than mss except the last one.
+	 */
+	for (i = 0; i < bd_num - max_non_tso_bd_num; i++) {
+		tot_len -= bd_size[i];
+		tot_len += bd_size[i + max_non_tso_bd_num - 1U];
+
+		if (tot_len < skb_shinfo(skb)->gso_size)
+			return true;
+	}
+
+	return false;
+}
+
+void hns3_shinfo_pack(struct skb_shared_info *shinfo, __u32 *size)
+{
+	u32 i;
+
+	for (i = 0; i < MAX_SKB_FRAGS; i++)
+		size[i] = skb_frag_size(&shinfo->frags[i]);
+}
+
+static int hns3_skb_linearize(struct hns3_enet_ring *ring,
+			      struct sk_buff *skb,
+			      unsigned int bd_num)
+{
+	/* 'bd_num == UINT_MAX' means the skb' fraglist has a
+	 * recursion level of over HNS3_MAX_RECURSION_LEVEL.
+	 */
+	if (bd_num == UINT_MAX) {
+		hns3_ring_stats_update(ring, over_max_recursion);
+		return -ENOMEM;
+	}
+
+	/* The skb->len has exceeded the hw limitation, linearization
+	 * will not help.
+	 */
+	if (skb->len > HNS3_MAX_TSO_SIZE ||
+	    (!skb_is_gso(skb) && skb->len > HNS3_MAX_NON_TSO_SIZE)) {
+		hns3_ring_stats_update(ring, hw_limitation);
+		return -ENOMEM;
+	}
+
+	if (__skb_linearize(skb)) {
+		hns3_ring_stats_update(ring, sw_err_cnt);
+		return -ENOMEM;
+	}
+
+	return 0;
+}
+
+static int hns3_nic_maybe_stop_tx(struct hns3_enet_ring *ring,
+				  struct net_device *netdev,
+				  struct sk_buff *skb)
+{
+	struct hns3_nic_priv *priv = netdev_priv(netdev);
+	u8 max_non_tso_bd_num = priv->max_non_tso_bd_num;
+	unsigned int bd_size[HNS3_MAX_TSO_BD_NUM + 1U];
+	unsigned int bd_num;
+
+	bd_num = hns3_tx_bd_num(skb, bd_size, max_non_tso_bd_num, 0, 0);
+	if (unlikely(bd_num > max_non_tso_bd_num)) {
+		if (bd_num <= HNS3_MAX_TSO_BD_NUM && skb_is_gso(skb) &&
+		    !hns3_skb_need_linearized(skb, bd_size, bd_num,
+					      max_non_tso_bd_num)) {
+			trace_hns3_over_max_bd(skb);
+			goto out;
+		}
+
+		if (hns3_skb_linearize(ring, skb, bd_num))
+			return -ENOMEM;
+
+		bd_num = hns3_tx_bd_count(skb->len);
+
+		hns3_ring_stats_update(ring, tx_copy);
+	}
+
+out:
+	if (likely(ring_space(ring) >= bd_num))
+		return bd_num;
+
+	netif_stop_subqueue(netdev, ring->queue_index);
+	smp_mb(); /* Memory barrier before checking ring_space */
+
+	/* Start queue in case hns3_clean_tx_ring has just made room
+	 * available and has not seen the queue stopped state performed
+	 * by netif_stop_subqueue above.
+	 */
+	if (ring_space(ring) >= bd_num && netif_carrier_ok(netdev) &&
+	    !test_bit(HNS3_NIC_STATE_DOWN, &priv->state)) {
+		netif_start_subqueue(netdev, ring->queue_index);
+		return bd_num;
+	}
+
+	hns3_ring_stats_update(ring, tx_busy);
+
+	return -EBUSY;
+}
+
+static void hns3_clear_desc(struct hns3_enet_ring *ring, int next_to_use_orig)
+{
+	struct device *dev = ring_to_dev(ring);
+	unsigned int i;
+
+	for (i = 0; i < ring->desc_num; i++) {
+		struct hns3_desc *desc = &ring->desc[ring->next_to_use];
+		struct hns3_desc_cb *desc_cb;
+
+		memset(desc, 0, sizeof(*desc));
+
+		/* check if this is where we started */
+		if (ring->next_to_use == next_to_use_orig)
+			break;
+
+		/* rollback one */
+		ring_ptr_move_bw(ring, next_to_use);
+
+		desc_cb = &ring->desc_cb[ring->next_to_use];
+
+		if (!desc_cb->dma)
+			continue;
+
+		/* unmap the descriptor dma address */
+		if (desc_cb->type & (DESC_TYPE_SKB | DESC_TYPE_FRAGLIST_SKB))
+			dma_unmap_single(dev, desc_cb->dma, desc_cb->length,
+					 DMA_TO_DEVICE);
+		else if (desc_cb->type &
+			 (DESC_TYPE_BOUNCE_HEAD | DESC_TYPE_BOUNCE_ALL))
+			hns3_tx_spare_rollback(ring, desc_cb->length);
+		else if (desc_cb->length)
+			dma_unmap_page(dev, desc_cb->dma, desc_cb->length,
+				       DMA_TO_DEVICE);
+
+		desc_cb->length = 0;
+		desc_cb->dma = 0;
+		desc_cb->type = DESC_TYPE_UNKNOWN;
+	}
+}
+
+static int hns3_fill_skb_to_desc(struct hns3_enet_ring *ring,
+				 struct sk_buff *skb, unsigned int type)
+{
+	struct sk_buff *frag_skb;
+	int i, ret, bd_num = 0;
+
+	ret = hns3_map_and_fill_desc(ring, skb, type);
+	if (unlikely(ret < 0))
+		return ret;
+
+	bd_num += ret;
+
+	for (i = 0; i < skb_shinfo(skb)->nr_frags; i++) {
+		skb_frag_t *frag = &skb_shinfo(skb)->frags[i];
+
+		ret = hns3_map_and_fill_desc(ring, frag, DESC_TYPE_PAGE);
+		if (unlikely(ret < 0))
+			return ret;
+
+		bd_num += ret;
+	}
+
+	skb_walk_frags(skb, frag_skb) {
+		ret = hns3_fill_skb_to_desc(ring, frag_skb,
+					    DESC_TYPE_FRAGLIST_SKB);
+		if (unlikely(ret < 0))
+			return ret;
+
+		bd_num += ret;
+	}
+
+	return bd_num;
+}
+
+static void hns3_tx_push_bd(struct hns3_enet_ring *ring, int num)
+{
+#define HNS3_BYTES_PER_64BIT		8
+
+	struct hns3_desc desc[HNS3_MAX_PUSH_BD_NUM] = {};
+	int offset = 0;
+
+	/* make sure everything is visible to device before
+	 * excuting tx push or updating doorbell
+	 */
+	dma_wmb();
+
+	do {
+		int idx = (ring->next_to_use - num + ring->desc_num) %
+			  ring->desc_num;
+
+		u64_stats_update_begin(&ring->syncp);
+		ring->stats.tx_push++;
+		u64_stats_update_end(&ring->syncp);
+		memcpy(&desc[offset], &ring->desc[idx],
+		       sizeof(struct hns3_desc));
+		offset++;
+	} while (--num);
+
+	__iowrite64_copy(ring->tqp->mem_base, desc,
+			 (sizeof(struct hns3_desc) * HNS3_MAX_PUSH_BD_NUM) /
+			 HNS3_BYTES_PER_64BIT);
+}
+
+static void hns3_tx_mem_doorbell(struct hns3_enet_ring *ring)
+{
+#define HNS3_MEM_DOORBELL_OFFSET	64
+
+	__le64 bd_num = cpu_to_le64((u64)ring->pending_buf);
+
+	/* make sure everything is visible to device before
+	 * excuting tx push or updating doorbell
+	 */
+	dma_wmb();
+
+	__iowrite64_copy(ring->tqp->mem_base + HNS3_MEM_DOORBELL_OFFSET,
+			 &bd_num, 1);
+	u64_stats_update_begin(&ring->syncp);
+	ring->stats.tx_mem_doorbell += ring->pending_buf;
+	u64_stats_update_end(&ring->syncp);
+}
+
+static void hns3_tx_doorbell(struct hns3_enet_ring *ring, int num,
+			     bool doorbell)
+{
+	struct net_device *netdev = ring_to_netdev(ring);
+	struct hns3_nic_priv *priv = netdev_priv(netdev);
+
+	/* when tx push is enabled, the packet whose number of BD below
+	 * HNS3_MAX_PUSH_BD_NUM can be pushed directly.
+	 */
+	if (test_bit(HNS3_NIC_STATE_TX_PUSH_ENABLE, &priv->state) && num &&
+	    !ring->pending_buf && num <= HNS3_MAX_PUSH_BD_NUM && doorbell) {
+		/* This smp_store_release() pairs with smp_load_acquire() in
+		 * hns3_nic_reclaim_desc(). Ensure that the BD valid bit
+		 * is updated.
+		 */
+		smp_store_release(&ring->last_to_use, ring->next_to_use);
+		hns3_tx_push_bd(ring, num);
+		return;
+	}
+
+	ring->pending_buf += num;
+
+	if (!doorbell) {
+		hns3_ring_stats_update(ring, tx_more);
+		return;
+	}
+
+	/* This smp_store_release() pairs with smp_load_acquire() in
+	 * hns3_nic_reclaim_desc(). Ensure that the BD valid bit is updated.
+	 */
+	smp_store_release(&ring->last_to_use, ring->next_to_use);
+
+	if (ring->tqp->mem_base)
+		hns3_tx_mem_doorbell(ring);
+	else
+		writel(ring->pending_buf,
+		       ring->tqp->io_base + HNS3_RING_TX_RING_TAIL_REG);
+
+	ring->pending_buf = 0;
+}
+
+static void hns3_tsyn(struct net_device *netdev, struct sk_buff *skb,
+		      struct hns3_desc *desc)
+{
+	struct hnae3_handle *h = hns3_get_handle(netdev);
+
+	if (!(h->ae_algo->ops->set_tx_hwts_info &&
+	      h->ae_algo->ops->set_tx_hwts_info(h, skb)))
+		return;
+
+	desc->tx.bdtp_fe_sc_vld_ra_ri |= cpu_to_le16(BIT(HNS3_TXD_TSYN_B));
+}
+
+static int hns3_handle_tx_bounce(struct hns3_enet_ring *ring,
+				 struct sk_buff *skb)
+{
+	struct hns3_desc_cb *desc_cb = &ring->desc_cb[ring->next_to_use];
+	unsigned int type = DESC_TYPE_BOUNCE_HEAD;
+	unsigned int size = skb_headlen(skb);
+	dma_addr_t dma;
+	int bd_num = 0;
+	u32 cb_len;
+	void *buf;
+	int ret;
+
+	if (skb->len <= ring->tx_copybreak) {
+		size = skb->len;
+		type = DESC_TYPE_BOUNCE_ALL;
+	}
+
+	/* hns3_can_use_tx_bounce() is called to ensure the below
+	 * function can always return the tx buffer.
+	 */
+	buf = hns3_tx_spare_alloc(ring, size, &dma, &cb_len);
+
+	ret = skb_copy_bits(skb, 0, buf, size);
+	if (unlikely(ret < 0)) {
+		hns3_tx_spare_rollback(ring, cb_len);
+		hns3_ring_stats_update(ring, copy_bits_err);
+		return ret;
+	}
+
+	desc_cb->priv = skb;
+	desc_cb->length = cb_len;
+	desc_cb->dma = dma;
+	desc_cb->type = type;
+
+	bd_num += hns3_fill_desc(ring, dma, size);
+
+	if (type == DESC_TYPE_BOUNCE_HEAD) {
+		ret = hns3_fill_skb_to_desc(ring, skb,
+					    DESC_TYPE_BOUNCE_HEAD);
+		if (unlikely(ret < 0))
+			return ret;
+
+		bd_num += ret;
+	}
+
+	dma_sync_single_for_device(ring_to_dev(ring), dma, size,
+				   DMA_TO_DEVICE);
+
+	hns3_ring_stats_update(ring, tx_bounce);
+
+	return bd_num;
+}
+
+static int hns3_handle_tx_sgl(struct hns3_enet_ring *ring,
+			      struct sk_buff *skb)
+{
+	struct hns3_desc_cb *desc_cb = &ring->desc_cb[ring->next_to_use];
+	u32 nfrag = skb_shinfo(skb)->nr_frags + 1;
+	struct sg_table *sgt;
+	int bd_num = 0;
+	dma_addr_t dma;
+	u32 cb_len, i;
+	int nents;
+
+	if (skb_has_frag_list(skb))
+		nfrag = HNS3_MAX_TSO_BD_NUM;
+
+	/* hns3_can_use_tx_sgl() is called to ensure the below
+	 * function can always return the tx buffer.
+	 */
+	sgt = hns3_tx_spare_alloc(ring, HNS3_SGL_SIZE(nfrag),
+				  &dma, &cb_len);
+
+	/* scatterlist follows by the sg table */
+	sgt->sgl = (struct scatterlist *)(sgt + 1);
+	sg_init_table(sgt->sgl, nfrag);
+	nents = skb_to_sgvec(skb, sgt->sgl, 0, skb->len);
+	if (unlikely(nents < 0)) {
+		hns3_tx_spare_rollback(ring, cb_len);
+		hns3_ring_stats_update(ring, skb2sgl_err);
+		return -ENOMEM;
+	}
+
+	sgt->orig_nents = nents;
+	sgt->nents = dma_map_sg(ring_to_dev(ring), sgt->sgl, sgt->orig_nents,
+				DMA_TO_DEVICE);
+	if (unlikely(!sgt->nents)) {
+		hns3_tx_spare_rollback(ring, cb_len);
+		hns3_ring_stats_update(ring, map_sg_err);
+		return -ENOMEM;
+	}
+
+	desc_cb->priv = skb;
+	desc_cb->length = cb_len;
+	desc_cb->dma = dma;
+	desc_cb->type = DESC_TYPE_SGL_SKB;
+
+	for (i = 0; i < sgt->nents; i++)
+		bd_num += hns3_fill_desc(ring, sg_dma_address(sgt->sgl + i),
+					 sg_dma_len(sgt->sgl + i));
+	hns3_ring_stats_update(ring, tx_sgl);
+
+	return bd_num;
+}
+
+static int hns3_handle_desc_filling(struct hns3_enet_ring *ring,
+				    struct sk_buff *skb)
+{
+	u32 space;
+
+	if (!ring->tx_spare)
+		goto out;
+
+	space = hns3_tx_spare_space(ring);
+
+	if (hns3_can_use_tx_sgl(ring, skb, space))
+		return hns3_handle_tx_sgl(ring, skb);
+
+	if (hns3_can_use_tx_bounce(ring, skb, space))
+		return hns3_handle_tx_bounce(ring, skb);
+
+out:
+	return hns3_fill_skb_to_desc(ring, skb, DESC_TYPE_SKB);
+}
+
+static int hns3_handle_skb_desc(struct hns3_enet_ring *ring,
+				struct sk_buff *skb,
+				struct hns3_desc_cb *desc_cb,
+				int next_to_use_head)
+{
+	int ret;
+
+	ret = hns3_fill_skb_desc(ring, skb, &ring->desc[ring->next_to_use],
+				 desc_cb);
+	if (unlikely(ret < 0))
+		goto fill_err;
+
+	/* 'ret < 0' means filling error, 'ret == 0' means skb->len is
+	 * zero, which is unlikely, and 'ret > 0' means how many tx desc
+	 * need to be notified to the hw.
+	 */
+	ret = hns3_handle_desc_filling(ring, skb);
+	if (likely(ret > 0))
+		return ret;
+
+fill_err:
+	hns3_clear_desc(ring, next_to_use_head);
+	return ret;
+}
+
+netdev_tx_t hns3_nic_net_xmit(struct sk_buff *skb, struct net_device *netdev)
+{
+	struct hns3_nic_priv *priv = netdev_priv(netdev);
+	struct hns3_enet_ring *ring = &priv->ring[skb->queue_mapping];
+	struct hns3_desc_cb *desc_cb = &ring->desc_cb[ring->next_to_use];
+	struct netdev_queue *dev_queue;
+	int pre_ntu, ret;
+	bool doorbell;
+
+	/* Hardware can only handle short frames above 32 bytes */
+	if (skb_put_padto(skb, HNS3_MIN_TX_LEN)) {
+		hns3_tx_doorbell(ring, 0, !netdev_xmit_more());
+
+		hns3_ring_stats_update(ring, sw_err_cnt);
+
+		return NETDEV_TX_OK;
+	}
+
+	/* Prefetch the data used later */
+	prefetch(skb->data);
+
+	ret = hns3_nic_maybe_stop_tx(ring, netdev, skb);
+	if (unlikely(ret <= 0)) {
+		if (ret == -EBUSY) {
+			hns3_tx_doorbell(ring, 0, true);
+			return NETDEV_TX_BUSY;
+		}
+
+		hns3_rl_err(netdev, "xmit error: %d!\n", ret);
+		goto out_err_tx_ok;
+	}
+
+	ret = hns3_handle_skb_desc(ring, skb, desc_cb, ring->next_to_use);
+	if (unlikely(ret <= 0))
+		goto out_err_tx_ok;
+
+	pre_ntu = ring->next_to_use ? (ring->next_to_use - 1) :
+					(ring->desc_num - 1);
+
+	if (unlikely(skb_shinfo(skb)->tx_flags & SKBTX_HW_TSTAMP))
+		hns3_tsyn(netdev, skb, &ring->desc[pre_ntu]);
+
+	ring->desc[pre_ntu].tx.bdtp_fe_sc_vld_ra_ri |=
+				cpu_to_le16(BIT(HNS3_TXD_FE_B));
+	trace_hns3_tx_desc(ring, pre_ntu);
+
+	skb_tx_timestamp(skb);
+
+	/* Complete translate all packets */
+	dev_queue = netdev_get_tx_queue(netdev, ring->queue_index);
+	doorbell = __netdev_tx_sent_queue(dev_queue, desc_cb->send_bytes,
+					  netdev_xmit_more());
+	hns3_tx_doorbell(ring, ret, doorbell);
+
+	return NETDEV_TX_OK;
+
+out_err_tx_ok:
+	dev_kfree_skb_any(skb);
+	hns3_tx_doorbell(ring, 0, !netdev_xmit_more());
+	return NETDEV_TX_OK;
+}
+
+static int hns3_nic_net_set_mac_address(struct net_device *netdev, void *p)
+{
+	char format_mac_addr_perm[HNAE3_FORMAT_MAC_ADDR_LEN];
+	char format_mac_addr_sa[HNAE3_FORMAT_MAC_ADDR_LEN];
+	struct hnae3_handle *h = hns3_get_handle(netdev);
+	struct sockaddr *mac_addr = p;
+	int ret;
+
+	if (!mac_addr || !is_valid_ether_addr((const u8 *)mac_addr->sa_data))
+		return -EADDRNOTAVAIL;
+
+	if (ether_addr_equal(netdev->dev_addr, mac_addr->sa_data)) {
+		hnae3_format_mac_addr(format_mac_addr_sa, mac_addr->sa_data);
+		netdev_info(netdev, "already using mac address %s\n",
+			    format_mac_addr_sa);
+		return 0;
+	}
+
+	/* For VF device, if there is a perm_addr, then the user will not
+	 * be allowed to change the address.
+	 */
+	if (!hns3_is_phys_func(h->pdev) &&
+	    !is_zero_ether_addr(netdev->perm_addr)) {
+		hnae3_format_mac_addr(format_mac_addr_perm, netdev->perm_addr);
+		hnae3_format_mac_addr(format_mac_addr_sa, mac_addr->sa_data);
+		netdev_err(netdev, "has permanent MAC %s, user MAC %s not allow\n",
+			   format_mac_addr_perm, format_mac_addr_sa);
+		return -EPERM;
+	}
+
+	ret = h->ae_algo->ops->set_mac_addr(h, mac_addr->sa_data, false);
+	if (ret) {
+		netdev_err(netdev, "set_mac_address fail, ret=%d!\n", ret);
+		return ret;
+	}
+
+	eth_hw_addr_set(netdev, mac_addr->sa_data);
+
+	return 0;
+}
+
+static int hns3_nic_do_ioctl(struct net_device *netdev,
+			     struct ifreq *ifr, int cmd)
+{
+	struct hnae3_handle *h = hns3_get_handle(netdev);
+
+	if (!netif_running(netdev))
+		return -EINVAL;
+
+	if (!h->ae_algo->ops->do_ioctl)
+		return -EOPNOTSUPP;
+
+	return h->ae_algo->ops->do_ioctl(h, ifr, cmd);
+}
+
+static int hns3_nic_set_features(struct net_device *netdev,
+				 netdev_features_t features)
+{
+	netdev_features_t changed = netdev->features ^ features;
+	struct hns3_nic_priv *priv = netdev_priv(netdev);
+	struct hnae3_handle *h = priv->ae_handle;
+	bool enable;
+	int ret;
+
+	if (changed & (NETIF_F_GRO_HW) && h->ae_algo->ops->set_gro_en) {
+		enable = !!(features & NETIF_F_GRO_HW);
+		ret = h->ae_algo->ops->set_gro_en(h, enable);
+		if (ret)
+			return ret;
+	}
+
+	if ((changed & NETIF_F_HW_VLAN_CTAG_RX) &&
+	    h->ae_algo->ops->enable_hw_strip_rxvtag) {
+		enable = !!(features & NETIF_F_HW_VLAN_CTAG_RX);
+		ret = h->ae_algo->ops->enable_hw_strip_rxvtag(h, enable);
+		if (ret)
+			return ret;
+	}
+
+	if ((changed & NETIF_F_NTUPLE) && h->ae_algo->ops->enable_fd) {
+		enable = !!(features & NETIF_F_NTUPLE);
+		h->ae_algo->ops->enable_fd(h, enable);
+	}
+
+	if ((netdev->features & NETIF_F_HW_TC) > (features & NETIF_F_HW_TC) &&
+	    h->ae_algo->ops->cls_flower_active(h)) {
+		netdev_err(netdev,
+			   "there are offloaded TC filters active, cannot disable HW TC offload\n");
+		return -EINVAL;
+	}
+
+	if ((changed & NETIF_F_HW_VLAN_CTAG_FILTER) &&
+	    h->ae_algo->ops->enable_vlan_filter) {
+		enable = !!(features & NETIF_F_HW_VLAN_CTAG_FILTER);
+		ret = h->ae_algo->ops->enable_vlan_filter(h, enable);
+		if (ret)
+			return ret;
+	}
+
+	return 0;
+}
+
+static netdev_features_t hns3_features_check(struct sk_buff *skb,
+					     struct net_device *dev,
+					     netdev_features_t features)
+{
+#define HNS3_MAX_HDR_LEN	480U
+#define HNS3_MAX_L4_HDR_LEN	60U
+
+	size_t len;
+
+	if (skb->ip_summed != CHECKSUM_PARTIAL)
+		return features;
+
+	if (skb->encapsulation)
+		len = skb_inner_transport_offset(skb);
+	else
+		len = skb_transport_offset(skb);
+
+	/* Assume L4 is 60 byte as TCP is the only protocol with a
+	 * a flexible value, and it's max len is 60 bytes.
+	 */
+	len += HNS3_MAX_L4_HDR_LEN;
+
+	/* Hardware only supports checksum on the skb with a max header
+	 * len of 480 bytes.
+	 */
+	if (len > HNS3_MAX_HDR_LEN)
+		features &= ~(NETIF_F_CSUM_MASK | NETIF_F_GSO_MASK);
+
+	return features;
+}
+
+static void hns3_fetch_stats(struct rtnl_link_stats64 *stats,
+			     struct hns3_enet_ring *ring, bool is_tx)
+{
+	unsigned int start;
+
+	do {
+		start = u64_stats_fetch_begin(&ring->syncp);
+		if (is_tx) {
+			stats->tx_bytes += ring->stats.tx_bytes;
+			stats->tx_packets += ring->stats.tx_pkts;
+			stats->tx_dropped += ring->stats.sw_err_cnt;
+			stats->tx_dropped += ring->stats.tx_vlan_err;
+			stats->tx_dropped += ring->stats.tx_l4_proto_err;
+			stats->tx_dropped += ring->stats.tx_l2l3l4_err;
+			stats->tx_dropped += ring->stats.tx_tso_err;
+			stats->tx_dropped += ring->stats.over_max_recursion;
+			stats->tx_dropped += ring->stats.hw_limitation;
+			stats->tx_dropped += ring->stats.copy_bits_err;
+			stats->tx_dropped += ring->stats.skb2sgl_err;
+			stats->tx_dropped += ring->stats.map_sg_err;
+			stats->tx_errors += ring->stats.sw_err_cnt;
+			stats->tx_errors += ring->stats.tx_vlan_err;
+			stats->tx_errors += ring->stats.tx_l4_proto_err;
+			stats->tx_errors += ring->stats.tx_l2l3l4_err;
+			stats->tx_errors += ring->stats.tx_tso_err;
+			stats->tx_errors += ring->stats.over_max_recursion;
+			stats->tx_errors += ring->stats.hw_limitation;
+			stats->tx_errors += ring->stats.copy_bits_err;
+			stats->tx_errors += ring->stats.skb2sgl_err;
+			stats->tx_errors += ring->stats.map_sg_err;
+		} else {
+			stats->rx_bytes += ring->stats.rx_bytes;
+			stats->rx_packets += ring->stats.rx_pkts;
+			stats->rx_dropped += ring->stats.l2_err;
+			stats->rx_errors += ring->stats.l2_err;
+			stats->rx_errors += ring->stats.l3l4_csum_err;
+			stats->rx_crc_errors += ring->stats.l2_err;
+			stats->multicast += ring->stats.rx_multicast;
+			stats->rx_length_errors += ring->stats.err_pkt_len;
+		}
+	} while (u64_stats_fetch_retry(&ring->syncp, start));
+}
+
+static void hns3_nic_get_stats64(struct net_device *netdev,
+				 struct rtnl_link_stats64 *stats)
+{
+	struct hns3_nic_priv *priv = netdev_priv(netdev);
+	int queue_num = priv->ae_handle->kinfo.num_tqps;
+	struct hnae3_handle *handle = priv->ae_handle;
+	struct rtnl_link_stats64 ring_total_stats;
+	struct hns3_enet_ring *ring;
+	int idx;
+
+	if (test_bit(HNS3_NIC_STATE_DOWN, &priv->state))
+		return;
+
+	handle->ae_algo->ops->update_stats(handle);
+
+	memset(&ring_total_stats, 0, sizeof(ring_total_stats));
+	for (idx = 0; idx < queue_num; idx++) {
+		/* fetch the tx stats */
+		ring = &priv->ring[idx];
+		hns3_fetch_stats(&ring_total_stats, ring, true);
+
+		/* fetch the rx stats */
+		ring = &priv->ring[idx + queue_num];
+		hns3_fetch_stats(&ring_total_stats, ring, false);
+	}
+
+	stats->tx_bytes = ring_total_stats.tx_bytes;
+	stats->tx_packets = ring_total_stats.tx_packets;
+	stats->rx_bytes = ring_total_stats.rx_bytes;
+	stats->rx_packets = ring_total_stats.rx_packets;
+
+	stats->rx_errors = ring_total_stats.rx_errors;
+	stats->multicast = ring_total_stats.multicast;
+	stats->rx_length_errors = ring_total_stats.rx_length_errors;
+	stats->rx_crc_errors = ring_total_stats.rx_crc_errors;
+	stats->rx_missed_errors = netdev->stats.rx_missed_errors;
+
+	stats->tx_errors = ring_total_stats.tx_errors;
+	stats->rx_dropped = ring_total_stats.rx_dropped;
+	stats->tx_dropped = ring_total_stats.tx_dropped;
+	stats->collisions = netdev->stats.collisions;
+	stats->rx_over_errors = netdev->stats.rx_over_errors;
+	stats->rx_frame_errors = netdev->stats.rx_frame_errors;
+	stats->rx_fifo_errors = netdev->stats.rx_fifo_errors;
+	stats->tx_aborted_errors = netdev->stats.tx_aborted_errors;
+	stats->tx_carrier_errors = netdev->stats.tx_carrier_errors;
+	stats->tx_fifo_errors = netdev->stats.tx_fifo_errors;
+	stats->tx_heartbeat_errors = netdev->stats.tx_heartbeat_errors;
+	stats->tx_window_errors = netdev->stats.tx_window_errors;
+	stats->rx_compressed = netdev->stats.rx_compressed;
+	stats->tx_compressed = netdev->stats.tx_compressed;
+}
+
+static int hns3_setup_tc(struct net_device *netdev, void *type_data)
+{
+	struct tc_mqprio_qopt_offload *mqprio_qopt = type_data;
+	struct hnae3_knic_private_info *kinfo;
+	u8 tc = mqprio_qopt->qopt.num_tc;
+	u16 mode = mqprio_qopt->mode;
+	u8 hw = mqprio_qopt->qopt.hw;
+	struct hnae3_handle *h;
+
+	if (!((hw == TC_MQPRIO_HW_OFFLOAD_TCS &&
+	       mode == TC_MQPRIO_MODE_CHANNEL) || (!hw && tc == 0)))
+		return -EOPNOTSUPP;
+
+	if (tc > HNAE3_MAX_TC)
+		return -EINVAL;
+
+	if (!netdev)
+		return -EINVAL;
+
+	h = hns3_get_handle(netdev);
+	kinfo = &h->kinfo;
+
+	netif_dbg(h, drv, netdev, "setup tc: num_tc=%u\n", tc);
+
+	return (kinfo->dcb_ops && kinfo->dcb_ops->setup_tc) ?
+		kinfo->dcb_ops->setup_tc(h, mqprio_qopt) : -EOPNOTSUPP;
+}
+
+static int hns3_setup_tc_cls_flower(struct hns3_nic_priv *priv,
+				    struct flow_cls_offload *flow)
+{
+	int tc = tc_classid_to_hwtc(priv->netdev, flow->classid);
+	struct hnae3_handle *h = hns3_get_handle(priv->netdev);
+
+	switch (flow->command) {
+	case FLOW_CLS_REPLACE:
+		if (h->ae_algo->ops->add_cls_flower)
+			return h->ae_algo->ops->add_cls_flower(h, flow, tc);
+		break;
+	case FLOW_CLS_DESTROY:
+		if (h->ae_algo->ops->del_cls_flower)
+			return h->ae_algo->ops->del_cls_flower(h, flow);
+		break;
+	default:
+		break;
+	}
+
+	return -EOPNOTSUPP;
+}
+
+static int hns3_setup_tc_block_cb(enum tc_setup_type type, void *type_data,
+				  void *cb_priv)
+{
+	struct hns3_nic_priv *priv = cb_priv;
+
+	if (!tc_cls_can_offload_and_chain0(priv->netdev, type_data))
+		return -EOPNOTSUPP;
+
+	switch (type) {
+	case TC_SETUP_CLSFLOWER:
+		return hns3_setup_tc_cls_flower(priv, type_data);
+	default:
+		return -EOPNOTSUPP;
+	}
+}
+
+static LIST_HEAD(hns3_block_cb_list);
+
+static int hns3_nic_setup_tc(struct net_device *dev, enum tc_setup_type type,
+			     void *type_data)
+{
+	struct hns3_nic_priv *priv = netdev_priv(dev);
+	int ret;
+
+	switch (type) {
+	case TC_SETUP_QDISC_MQPRIO:
+		ret = hns3_setup_tc(dev, type_data);
+		break;
+	case TC_SETUP_BLOCK:
+		ret = flow_block_cb_setup_simple(type_data,
+						 &hns3_block_cb_list,
+						 hns3_setup_tc_block_cb,
+						 priv, priv, true);
+		break;
+	default:
+		return -EOPNOTSUPP;
+	}
+
+	return ret;
+}
+
+static int hns3_vlan_rx_add_vid(struct net_device *netdev,
+				__be16 proto, u16 vid)
+{
+	struct hnae3_handle *h = hns3_get_handle(netdev);
+	int ret = -EIO;
+
+	if (h->ae_algo->ops->set_vlan_filter)
+		ret = h->ae_algo->ops->set_vlan_filter(h, proto, vid, false);
+
+	return ret;
+}
+
+static int hns3_vlan_rx_kill_vid(struct net_device *netdev,
+				 __be16 proto, u16 vid)
+{
+	struct hnae3_handle *h = hns3_get_handle(netdev);
+	int ret = -EIO;
+
+	if (h->ae_algo->ops->set_vlan_filter)
+		ret = h->ae_algo->ops->set_vlan_filter(h, proto, vid, true);
+
+	return ret;
+}
+
+static int hns3_ndo_set_vf_vlan(struct net_device *netdev, int vf, u16 vlan,
+				u8 qos, __be16 vlan_proto)
+{
+	struct hnae3_handle *h = hns3_get_handle(netdev);
+	int ret = -EIO;
+
+	netif_dbg(h, drv, netdev,
+		  "set vf vlan: vf=%d, vlan=%u, qos=%u, vlan_proto=0x%x\n",
+		  vf, vlan, qos, ntohs(vlan_proto));
+
+	if (h->ae_algo->ops->set_vf_vlan_filter)
+		ret = h->ae_algo->ops->set_vf_vlan_filter(h, vf, vlan,
+							  qos, vlan_proto);
+
+	return ret;
+}
+
+static int hns3_set_vf_spoofchk(struct net_device *netdev, int vf, bool enable)
+{
+	struct hnae3_handle *handle = hns3_get_handle(netdev);
+
+	if (hns3_nic_resetting(netdev))
+		return -EBUSY;
+
+	if (!handle->ae_algo->ops->set_vf_spoofchk)
+		return -EOPNOTSUPP;
+
+	return handle->ae_algo->ops->set_vf_spoofchk(handle, vf, enable);
+}
+
+static int hns3_set_vf_trust(struct net_device *netdev, int vf, bool enable)
+{
+	struct hnae3_handle *handle = hns3_get_handle(netdev);
+
+	if (!handle->ae_algo->ops->set_vf_trust)
+		return -EOPNOTSUPP;
+
+	return handle->ae_algo->ops->set_vf_trust(handle, vf, enable);
+}
+
+static int hns3_nic_change_mtu(struct net_device *netdev, int new_mtu)
+{
+	struct hnae3_handle *h = hns3_get_handle(netdev);
+	int ret;
+
+	if (hns3_nic_resetting(netdev))
+		return -EBUSY;
+
+	if (!h->ae_algo->ops->set_mtu)
+		return -EOPNOTSUPP;
+
+	netif_dbg(h, drv, netdev,
+		  "change mtu from %u to %d\n", netdev->mtu, new_mtu);
+
+	ret = h->ae_algo->ops->set_mtu(h, new_mtu);
+	if (ret)
+		netdev_err(netdev, "failed to change MTU in hardware %d\n",
+			   ret);
+	else
+		WRITE_ONCE(netdev->mtu, new_mtu);
+
+	return ret;
+}
+
+static int hns3_get_timeout_queue(struct net_device *ndev)
+{
+	unsigned int i;
+
+	/* Find the stopped queue the same way the stack does */
+	for (i = 0; i < ndev->num_tx_queues; i++) {
+		struct netdev_queue *q;
+		unsigned long trans_start;
+
+		q = netdev_get_tx_queue(ndev, i);
+		trans_start = READ_ONCE(q->trans_start);
+		if (netif_xmit_stopped(q) &&
+		    time_after(jiffies,
+			       (trans_start + ndev->watchdog_timeo))) {
+#ifdef CONFIG_BQL
+			struct dql *dql = &q->dql;
+
+			netdev_info(ndev, "DQL info last_cnt: %u, queued: %u, adj_limit: %u, completed: %u\n",
+				    dql->last_obj_cnt, dql->num_queued,
+				    dql->adj_limit, dql->num_completed);
+#endif
+			netdev_info(ndev, "queue state: 0x%lx, delta msecs: %u\n",
+				    q->state,
+				    jiffies_to_msecs(jiffies - trans_start));
+			break;
+		}
+	}
+
+	return i;
+}
+
+static void hns3_dump_queue_stats(struct net_device *ndev,
+				  struct hns3_enet_ring *tx_ring,
+				  int timeout_queue)
+{
+	struct napi_struct *napi = &tx_ring->tqp_vector->napi;
+	struct hns3_nic_priv *priv = netdev_priv(ndev);
+
+	netdev_info(ndev,
+		    "tx_timeout count: %llu, queue id: %d, SW_NTU: 0x%x, SW_NTC: 0x%x, napi state: %lu\n",
+		    priv->tx_timeout_count, timeout_queue, tx_ring->next_to_use,
+		    tx_ring->next_to_clean, napi->state);
+
+	netdev_info(ndev,
+		    "tx_pkts: %llu, tx_bytes: %llu, sw_err_cnt: %llu, tx_pending: %d\n",
+		    tx_ring->stats.tx_pkts, tx_ring->stats.tx_bytes,
+		    tx_ring->stats.sw_err_cnt, tx_ring->pending_buf);
+
+	netdev_info(ndev,
+		    "seg_pkt_cnt: %llu, tx_more: %llu, restart_queue: %llu, tx_busy: %llu\n",
+		    tx_ring->stats.seg_pkt_cnt, tx_ring->stats.tx_more,
+		    tx_ring->stats.restart_queue, tx_ring->stats.tx_busy);
+
+	netdev_info(ndev, "tx_push: %llu, tx_mem_doorbell: %llu\n",
+		    tx_ring->stats.tx_push, tx_ring->stats.tx_mem_doorbell);
+}
+
+static void hns3_dump_queue_reg(struct net_device *ndev,
+				struct hns3_enet_ring *tx_ring)
+{
+	netdev_info(ndev,
+		    "BD_NUM: 0x%x HW_HEAD: 0x%x, HW_TAIL: 0x%x, BD_ERR: 0x%x, INT: 0x%x\n",
+		    hns3_tqp_read_reg(tx_ring, HNS3_RING_TX_RING_BD_NUM_REG),
+		    hns3_tqp_read_reg(tx_ring, HNS3_RING_TX_RING_HEAD_REG),
+		    hns3_tqp_read_reg(tx_ring, HNS3_RING_TX_RING_TAIL_REG),
+		    hns3_tqp_read_reg(tx_ring, HNS3_RING_TX_RING_BD_ERR_REG),
+		    readl(tx_ring->tqp_vector->mask_addr));
+	netdev_info(ndev,
+		    "RING_EN: 0x%x, TC: 0x%x, FBD_NUM: 0x%x FBD_OFT: 0x%x, EBD_NUM: 0x%x, EBD_OFT: 0x%x\n",
+		    hns3_tqp_read_reg(tx_ring, HNS3_RING_EN_REG),
+		    hns3_tqp_read_reg(tx_ring, HNS3_RING_TX_RING_TC_REG),
+		    hns3_tqp_read_reg(tx_ring, HNS3_RING_TX_RING_FBDNUM_REG),
+		    hns3_tqp_read_reg(tx_ring, HNS3_RING_TX_RING_OFFSET_REG),
+		    hns3_tqp_read_reg(tx_ring, HNS3_RING_TX_RING_EBDNUM_REG),
+		    hns3_tqp_read_reg(tx_ring,
+				      HNS3_RING_TX_RING_EBD_OFFSET_REG));
+}
+
+static bool hns3_get_tx_timeo_queue_info(struct net_device *ndev)
+{
+	struct hns3_nic_priv *priv = netdev_priv(ndev);
+	struct hnae3_handle *h = hns3_get_handle(ndev);
+	struct hns3_enet_ring *tx_ring;
+	u32 timeout_queue;
+
+	timeout_queue = hns3_get_timeout_queue(ndev);
+	if (timeout_queue >= ndev->num_tx_queues) {
+		netdev_info(ndev,
+			    "no netdev TX timeout queue found, timeout count: %llu\n",
+			    priv->tx_timeout_count);
+		return false;
+	}
+
+	priv->tx_timeout_count++;
+
+	tx_ring = &priv->ring[timeout_queue];
+	hns3_dump_queue_stats(ndev, tx_ring, timeout_queue);
+
+	/* When mac received many pause frames continuous, it's unable to send
+	 * packets, which may cause tx timeout
+	 */
+	if (h->ae_algo->ops->get_mac_stats) {
+		struct hns3_mac_stats mac_stats;
+
+		h->ae_algo->ops->get_mac_stats(h, &mac_stats);
+		netdev_info(ndev, "tx_pause_cnt: %llu, rx_pause_cnt: %llu\n",
+			    mac_stats.tx_pause_cnt, mac_stats.rx_pause_cnt);
+	}
+
+	hns3_dump_queue_reg(ndev, tx_ring);
+
+	return true;
+}
+
+static void hns3_nic_net_timeout(struct net_device *ndev, unsigned int txqueue)
+{
+	struct hns3_nic_priv *priv = netdev_priv(ndev);
+	struct hnae3_handle *h = priv->ae_handle;
+
+	if (!hns3_get_tx_timeo_queue_info(ndev))
+		return;
+
+	/* request the reset, and let the hclge to determine
+	 * which reset level should be done
+	 */
+	if (h->ae_algo->ops->reset_event)
+		h->ae_algo->ops->reset_event(h->pdev, h);
+}
+
+#ifdef CONFIG_RFS_ACCEL
+static int hns3_rx_flow_steer(struct net_device *dev, const struct sk_buff *skb,
+			      u16 rxq_index, u32 flow_id)
+{
+	struct hnae3_handle *h = hns3_get_handle(dev);
+	struct flow_keys fkeys;
+
+	if (!h->ae_algo->ops->add_arfs_entry)
+		return -EOPNOTSUPP;
+
+	if (skb->encapsulation)
+		return -EPROTONOSUPPORT;
+
+	if (!skb_flow_dissect_flow_keys(skb, &fkeys, 0))
+		return -EPROTONOSUPPORT;
+
+	if ((fkeys.basic.n_proto != htons(ETH_P_IP) &&
+	     fkeys.basic.n_proto != htons(ETH_P_IPV6)) ||
+	    (fkeys.basic.ip_proto != IPPROTO_TCP &&
+	     fkeys.basic.ip_proto != IPPROTO_UDP))
+		return -EPROTONOSUPPORT;
+
+	return h->ae_algo->ops->add_arfs_entry(h, rxq_index, flow_id, &fkeys);
+}
+#endif
+
+static int hns3_nic_get_vf_config(struct net_device *ndev, int vf,
+				  struct ifla_vf_info *ivf)
+{
+	struct hnae3_handle *h = hns3_get_handle(ndev);
+
+	if (!h->ae_algo->ops->get_vf_config)
+		return -EOPNOTSUPP;
+
+	return h->ae_algo->ops->get_vf_config(h, vf, ivf);
+}
+
+static int hns3_nic_set_vf_link_state(struct net_device *ndev, int vf,
+				      int link_state)
+{
+	struct hnae3_handle *h = hns3_get_handle(ndev);
+
+	if (!h->ae_algo->ops->set_vf_link_state)
+		return -EOPNOTSUPP;
+
+	return h->ae_algo->ops->set_vf_link_state(h, vf, link_state);
+}
+
+static int hns3_nic_set_vf_rate(struct net_device *ndev, int vf,
+				int min_tx_rate, int max_tx_rate)
+{
+	struct hnae3_handle *h = hns3_get_handle(ndev);
+
+	if (!h->ae_algo->ops->set_vf_rate)
+		return -EOPNOTSUPP;
+
+	return h->ae_algo->ops->set_vf_rate(h, vf, min_tx_rate, max_tx_rate,
+					    false);
+}
+
+static int hns3_nic_set_vf_mac(struct net_device *netdev, int vf_id, u8 *mac)
+{
+	struct hnae3_handle *h = hns3_get_handle(netdev);
+	char format_mac_addr[HNAE3_FORMAT_MAC_ADDR_LEN];
+
+	if (!h->ae_algo->ops->set_vf_mac)
+		return -EOPNOTSUPP;
+
+	if (is_multicast_ether_addr(mac)) {
+		hnae3_format_mac_addr(format_mac_addr, mac);
+		netdev_err(netdev,
+			   "Invalid MAC:%s specified. Could not set MAC\n",
+			   format_mac_addr);
+		return -EINVAL;
+	}
+
+	return h->ae_algo->ops->set_vf_mac(h, vf_id, mac);
+}
+
+#define HNS3_INVALID_DSCP		0xff
+#define HNS3_DSCP_SHIFT			2
+
+static u8 hns3_get_skb_dscp(struct sk_buff *skb)
+{
+	__be16 protocol = skb->protocol;
+	u8 dscp = HNS3_INVALID_DSCP;
+
+	if (protocol == htons(ETH_P_8021Q))
+		protocol = vlan_get_protocol(skb);
+
+	if (protocol == htons(ETH_P_IP))
+		dscp = ipv4_get_dsfield(ip_hdr(skb)) >> HNS3_DSCP_SHIFT;
+	else if (protocol == htons(ETH_P_IPV6))
+		dscp = ipv6_get_dsfield(ipv6_hdr(skb)) >> HNS3_DSCP_SHIFT;
+
+	return dscp;
+}
+
+static u16 hns3_nic_select_queue(struct net_device *netdev,
+				 struct sk_buff *skb,
+				 struct net_device *sb_dev)
+{
+	struct hnae3_handle *h = hns3_get_handle(netdev);
+	u8 dscp;
+
+	if (h->kinfo.tc_map_mode != HNAE3_TC_MAP_MODE_DSCP ||
+	    !h->ae_algo->ops->get_dscp_prio)
+		goto out;
+
+	dscp = hns3_get_skb_dscp(skb);
+	if (unlikely(dscp >= HNAE3_MAX_DSCP))
+		goto out;
+
+	skb->priority = h->kinfo.dscp_prio[dscp];
+	if (skb->priority == HNAE3_PRIO_ID_INVALID)
+		skb->priority = 0;
+
+out:
+	return netdev_pick_tx(netdev, skb, sb_dev);
+}
+
+static const struct net_device_ops hns3_nic_netdev_ops = {
+	.ndo_open		= hns3_nic_net_open,
+	.ndo_stop		= hns3_nic_net_stop,
+	.ndo_start_xmit		= hns3_nic_net_xmit,
+	.ndo_tx_timeout		= hns3_nic_net_timeout,
+	.ndo_set_mac_address	= hns3_nic_net_set_mac_address,
+	.ndo_eth_ioctl		= hns3_nic_do_ioctl,
+	.ndo_change_mtu		= hns3_nic_change_mtu,
+	.ndo_set_features	= hns3_nic_set_features,
+	.ndo_features_check	= hns3_features_check,
+	.ndo_get_stats64	= hns3_nic_get_stats64,
+	.ndo_setup_tc		= hns3_nic_setup_tc,
+	.ndo_set_rx_mode	= hns3_nic_set_rx_mode,
+	.ndo_vlan_rx_add_vid	= hns3_vlan_rx_add_vid,
+	.ndo_vlan_rx_kill_vid	= hns3_vlan_rx_kill_vid,
+	.ndo_set_vf_vlan	= hns3_ndo_set_vf_vlan,
+	.ndo_set_vf_spoofchk	= hns3_set_vf_spoofchk,
+	.ndo_set_vf_trust	= hns3_set_vf_trust,
+#ifdef CONFIG_RFS_ACCEL
+	.ndo_rx_flow_steer	= hns3_rx_flow_steer,
+#endif
+	.ndo_get_vf_config	= hns3_nic_get_vf_config,
+	.ndo_set_vf_link_state	= hns3_nic_set_vf_link_state,
+	.ndo_set_vf_rate	= hns3_nic_set_vf_rate,
+	.ndo_set_vf_mac		= hns3_nic_set_vf_mac,
+	.ndo_select_queue	= hns3_nic_select_queue,
+};
+
+bool hns3_is_phys_func(struct pci_dev *pdev)
+{
+	u32 dev_id = pdev->device;
+
+	switch (dev_id) {
+	case HNAE3_DEV_ID_GE:
+	case HNAE3_DEV_ID_25GE:
+	case HNAE3_DEV_ID_25GE_RDMA:
+	case HNAE3_DEV_ID_25GE_RDMA_MACSEC:
+	case HNAE3_DEV_ID_50GE_RDMA:
+	case HNAE3_DEV_ID_50GE_RDMA_MACSEC:
+	case HNAE3_DEV_ID_100G_RDMA_MACSEC:
+	case HNAE3_DEV_ID_200G_RDMA:
+		return true;
+	case HNAE3_DEV_ID_VF:
+	case HNAE3_DEV_ID_RDMA_DCB_PFC_VF:
+		return false;
+	default:
+		dev_warn(&pdev->dev, "un-recognized pci device-id %u",
+			 dev_id);
+	}
+
+	return false;
+}
+
+static void hns3_disable_sriov(struct pci_dev *pdev)
+{
+	/* If our VFs are assigned we cannot shut down SR-IOV
+	 * without causing issues, so just leave the hardware
+	 * available but disabled
+	 */
+	if (pci_vfs_assigned(pdev)) {
+		dev_warn(&pdev->dev,
+			 "disabling driver while VFs are assigned\n");
+		return;
+	}
+
+	pci_disable_sriov(pdev);
+}
+
+/* hns3_probe - Device initialization routine
+ * @pdev: PCI device information struct
+ * @ent: entry in hns3_pci_tbl
+ *
+ * hns3_probe initializes a PF identified by a pci_dev structure.
+ * The OS initialization, configuring of the PF private structure,
+ * and a hardware reset occur.
+ *
+ * Returns 0 on success, negative on failure
+ */
+static int hns3_probe(struct pci_dev *pdev, const struct pci_device_id *ent)
+{
+	struct hnae3_ae_dev *ae_dev;
+	int ret;
+
+	ae_dev = devm_kzalloc(&pdev->dev, sizeof(*ae_dev), GFP_KERNEL);
+	if (!ae_dev)
+		return -ENOMEM;
+
+	ae_dev->pdev = pdev;
+	ae_dev->flag = ent->driver_data;
+	pci_set_drvdata(pdev, ae_dev);
+
+	ret = hnae3_register_ae_dev(ae_dev);
+	if (ret)
+		pci_set_drvdata(pdev, NULL);
+
+	return ret;
+}
+
+/**
+ * hns3_clean_vf_config
+ * @pdev: pointer to a pci_dev structure
+ * @num_vfs: number of VFs allocated
+ *
+ * Clean residual vf config after disable sriov
+ **/
+static void hns3_clean_vf_config(struct pci_dev *pdev, int num_vfs)
+{
+	struct hnae3_ae_dev *ae_dev = pci_get_drvdata(pdev);
+
+	if (ae_dev->ops->clean_vf_config)
+		ae_dev->ops->clean_vf_config(ae_dev, num_vfs);
+}
+
+/* hns3_remove - Device removal routine
+ * @pdev: PCI device information struct
+ */
+static void hns3_remove(struct pci_dev *pdev)
+{
+	struct hnae3_ae_dev *ae_dev = pci_get_drvdata(pdev);
+
+	if (hns3_is_phys_func(pdev) && IS_ENABLED(CONFIG_PCI_IOV))
+		hns3_disable_sriov(pdev);
+
+	hnae3_unregister_ae_dev(ae_dev);
+	pci_set_drvdata(pdev, NULL);
+}
+
+/**
+ * hns3_pci_sriov_configure
+ * @pdev: pointer to a pci_dev structure
+ * @num_vfs: number of VFs to allocate
+ *
+ * Enable or change the number of VFs. Called when the user updates the number
+ * of VFs in sysfs.
+ **/
+static int hns3_pci_sriov_configure(struct pci_dev *pdev, int num_vfs)
+{
+	int ret;
+
+	if (!(hns3_is_phys_func(pdev) && IS_ENABLED(CONFIG_PCI_IOV))) {
+		dev_warn(&pdev->dev, "Can not config SRIOV\n");
+		return -EINVAL;
+	}
+
+	if (num_vfs) {
+		ret = pci_enable_sriov(pdev, num_vfs);
+		if (ret)
+			dev_err(&pdev->dev, "SRIOV enable failed %d\n", ret);
+		else
+			return num_vfs;
+	} else if (!pci_vfs_assigned(pdev)) {
+		int num_vfs_pre = pci_num_vf(pdev);
+
+		pci_disable_sriov(pdev);
+		hns3_clean_vf_config(pdev, num_vfs_pre);
+	} else {
+		dev_warn(&pdev->dev,
+			 "Unable to free VFs because some are assigned to VMs.\n");
+	}
+
+	return 0;
+}
+
+static void hns3_shutdown(struct pci_dev *pdev)
+{
+	struct hnae3_ae_dev *ae_dev = pci_get_drvdata(pdev);
+
+	hnae3_unregister_ae_dev(ae_dev);
+	pci_set_drvdata(pdev, NULL);
+
+	if (system_state == SYSTEM_POWER_OFF)
+		pci_set_power_state(pdev, PCI_D3hot);
+}
+
+static int __maybe_unused hns3_suspend(struct device *dev)
+{
+	struct hnae3_ae_dev *ae_dev = dev_get_drvdata(dev);
+
+	if (ae_dev && hns3_is_phys_func(ae_dev->pdev)) {
+		dev_info(dev, "Begin to suspend.\n");
+		if (ae_dev->ops && ae_dev->ops->reset_prepare)
+			ae_dev->ops->reset_prepare(ae_dev, HNAE3_FUNC_RESET);
+	}
+
+	return 0;
+}
+
+static int __maybe_unused hns3_resume(struct device *dev)
+{
+	struct hnae3_ae_dev *ae_dev = dev_get_drvdata(dev);
+
+	if (ae_dev && hns3_is_phys_func(ae_dev->pdev)) {
+		dev_info(dev, "Begin to resume.\n");
+		if (ae_dev->ops && ae_dev->ops->reset_done)
+			ae_dev->ops->reset_done(ae_dev);
+	}
+
+	return 0;
+}
+
+static pci_ers_result_t hns3_error_detected(struct pci_dev *pdev,
+					    pci_channel_state_t state)
+{
+	struct hnae3_ae_dev *ae_dev = pci_get_drvdata(pdev);
+	pci_ers_result_t ret;
+
+	dev_info(&pdev->dev, "PCI error detected, state(=%u)!!\n", state);
+
+	if (state == pci_channel_io_perm_failure)
+		return PCI_ERS_RESULT_DISCONNECT;
+
+	if (!ae_dev || !ae_dev->ops) {
+		dev_err(&pdev->dev,
+			"Can't recover - error happened before device initialized\n");
+		return PCI_ERS_RESULT_NONE;
+	}
+
+	if (ae_dev->ops->handle_hw_ras_error)
+		ret = ae_dev->ops->handle_hw_ras_error(ae_dev);
+	else
+		return PCI_ERS_RESULT_NONE;
+
+	return ret;
+}
+
+static pci_ers_result_t hns3_slot_reset(struct pci_dev *pdev)
+{
+	struct hnae3_ae_dev *ae_dev = pci_get_drvdata(pdev);
+	const struct hnae3_ae_ops *ops;
+	enum hnae3_reset_type reset_type;
+	struct device *dev = &pdev->dev;
+
+	if (!ae_dev || !ae_dev->ops)
+		return PCI_ERS_RESULT_NONE;
+
+	ops = ae_dev->ops;
+	/* request the reset */
+	if (ops->reset_event && ops->get_reset_level &&
+	    ops->set_default_reset_request) {
+		if (ae_dev->hw_err_reset_req) {
+			reset_type = ops->get_reset_level(ae_dev,
+						&ae_dev->hw_err_reset_req);
+			ops->set_default_reset_request(ae_dev, reset_type);
+			dev_info(dev, "requesting reset due to PCI error\n");
+			ops->reset_event(pdev, NULL);
+		}
+
+		return PCI_ERS_RESULT_RECOVERED;
+	}
+
+	return PCI_ERS_RESULT_DISCONNECT;
+}
+
+static void hns3_reset_prepare(struct pci_dev *pdev)
+{
+	struct hnae3_ae_dev *ae_dev = pci_get_drvdata(pdev);
+
+	dev_info(&pdev->dev, "FLR prepare\n");
+	if (ae_dev && ae_dev->ops && ae_dev->ops->reset_prepare)
+		ae_dev->ops->reset_prepare(ae_dev, HNAE3_FLR_RESET);
+}
+
+static void hns3_reset_done(struct pci_dev *pdev)
+{
+	struct hnae3_ae_dev *ae_dev = pci_get_drvdata(pdev);
+
+	dev_info(&pdev->dev, "FLR done\n");
+	if (ae_dev && ae_dev->ops && ae_dev->ops->reset_done)
+		ae_dev->ops->reset_done(ae_dev);
+}
+
+static const struct pci_error_handlers hns3_err_handler = {
+	.error_detected = hns3_error_detected,
+	.slot_reset     = hns3_slot_reset,
+	.reset_prepare	= hns3_reset_prepare,
+	.reset_done	= hns3_reset_done,
+};
+
+static SIMPLE_DEV_PM_OPS(hns3_pm_ops, hns3_suspend, hns3_resume);
+
+static struct pci_driver hns3_driver = {
+	.name     = hns3_driver_name,
+	.id_table = hns3_pci_tbl,
+	.probe    = hns3_probe,
+	.remove   = hns3_remove,
+	.shutdown = hns3_shutdown,
+	.driver.pm  = &hns3_pm_ops,
+	.sriov_configure = hns3_pci_sriov_configure,
+	.err_handler    = &hns3_err_handler,
+};
+
+/* set default feature to hns3 */
+static void hns3_set_default_feature(struct net_device *netdev)
+{
+	struct hnae3_handle *h = hns3_get_handle(netdev);
+	struct pci_dev *pdev = h->pdev;
+	struct hnae3_ae_dev *ae_dev = pci_get_drvdata(pdev);
+
+	netdev->priv_flags |= IFF_UNICAST_FLT;
+
+	netdev->features |= NETIF_F_HW_VLAN_CTAG_FILTER |
+		NETIF_F_HW_VLAN_CTAG_TX | NETIF_F_HW_VLAN_CTAG_RX |
+		NETIF_F_RXCSUM | NETIF_F_SG | NETIF_F_GSO |
+		NETIF_F_GRO | NETIF_F_TSO | NETIF_F_TSO6 | NETIF_F_GSO_GRE |
+		NETIF_F_GSO_GRE_CSUM | NETIF_F_GSO_UDP_TUNNEL |
+		NETIF_F_SCTP_CRC | NETIF_F_FRAGLIST;
+
+	if (hnae3_ae_dev_gro_supported(ae_dev))
+		netdev->features |= NETIF_F_GRO_HW;
+
+	if (hnae3_ae_dev_fd_supported(ae_dev))
+		netdev->features |= NETIF_F_NTUPLE;
+
+	if (test_bit(HNAE3_DEV_SUPPORT_UDP_GSO_B, ae_dev->caps))
+		netdev->features |= NETIF_F_GSO_UDP_L4;
+
+	if (test_bit(HNAE3_DEV_SUPPORT_HW_TX_CSUM_B, ae_dev->caps))
+		netdev->features |= NETIF_F_HW_CSUM;
+	else
+		netdev->features |= NETIF_F_IP_CSUM | NETIF_F_IPV6_CSUM;
+
+	if (test_bit(HNAE3_DEV_SUPPORT_UDP_TUNNEL_CSUM_B, ae_dev->caps))
+		netdev->features |= NETIF_F_GSO_UDP_TUNNEL_CSUM;
+
+	if (test_bit(HNAE3_DEV_SUPPORT_FD_FORWARD_TC_B, ae_dev->caps))
+		netdev->features |= NETIF_F_HW_TC;
+
+	netdev->hw_features |= netdev->features;
+	if (!test_bit(HNAE3_DEV_SUPPORT_VLAN_FLTR_MDF_B, ae_dev->caps))
+		netdev->hw_features &= ~NETIF_F_HW_VLAN_CTAG_FILTER;
+
+	netdev->vlan_features |= netdev->features &
+		~(NETIF_F_HW_VLAN_CTAG_FILTER | NETIF_F_HW_VLAN_CTAG_TX |
+		  NETIF_F_HW_VLAN_CTAG_RX | NETIF_F_GRO_HW | NETIF_F_NTUPLE |
+		  NETIF_F_HW_TC);
+
+	netdev->hw_enc_features |= netdev->vlan_features | NETIF_F_TSO_MANGLEID;
+
+	/* The device_version V3 hardware can't offload the checksum for IP in
+	 * GRE packets, but can do it for NvGRE. So default to disable the
+	 * checksum and GSO offload for GRE.
+	 */
+	if (ae_dev->dev_version > HNAE3_DEVICE_VERSION_V2) {
+		netdev->features &= ~NETIF_F_GSO_GRE;
+		netdev->features &= ~NETIF_F_GSO_GRE_CSUM;
+	}
+}
+
+static int hns3_alloc_buffer(struct hns3_enet_ring *ring,
+			     struct hns3_desc_cb *cb)
+{
+	unsigned int order = hns3_page_order(ring);
+	struct page *p;
+
+	if (ring->page_pool) {
+		p = page_pool_dev_alloc_frag(ring->page_pool,
+					     &cb->page_offset,
+					     hns3_buf_size(ring));
+		if (unlikely(!p))
+			return -ENOMEM;
+
+		cb->priv = p;
+		cb->buf = page_address(p);
+		cb->dma = page_pool_get_dma_addr(p);
+		cb->type = DESC_TYPE_PP_FRAG;
+		cb->reuse_flag = 0;
+		return 0;
+	}
+
+	p = dev_alloc_pages(order);
+	if (!p)
+		return -ENOMEM;
+
+	cb->priv = p;
+	cb->page_offset = 0;
+	cb->reuse_flag = 0;
+	cb->buf  = page_address(p);
+	cb->length = hns3_page_size(ring);
+	cb->type = DESC_TYPE_PAGE;
+	page_ref_add(p, USHRT_MAX - 1);
+	cb->pagecnt_bias = USHRT_MAX;
+
+	return 0;
+}
+
+static void hns3_free_buffer(struct hns3_enet_ring *ring,
+			     struct hns3_desc_cb *cb, int budget)
+{
+	if (cb->type & (DESC_TYPE_SKB | DESC_TYPE_BOUNCE_HEAD |
+			DESC_TYPE_BOUNCE_ALL | DESC_TYPE_SGL_SKB))
+		napi_consume_skb(cb->priv, budget);
+	else if (!HNAE3_IS_TX_RING(ring)) {
+		if (cb->type & DESC_TYPE_PAGE && cb->pagecnt_bias)
+			__page_frag_cache_drain(cb->priv, cb->pagecnt_bias);
+		else if (cb->type & DESC_TYPE_PP_FRAG)
+			page_pool_put_full_page(ring->page_pool, cb->priv,
+						false);
+	}
+	memset(cb, 0, sizeof(*cb));
+}
+
+static int hns3_map_buffer(struct hns3_enet_ring *ring, struct hns3_desc_cb *cb)
+{
+	cb->dma = dma_map_page(ring_to_dev(ring), cb->priv, 0,
+			       cb->length, ring_to_dma_dir(ring));
+
+	if (unlikely(dma_mapping_error(ring_to_dev(ring), cb->dma)))
+		return -EIO;
+
+	return 0;
+}
+
+static void hns3_unmap_buffer(struct hns3_enet_ring *ring,
+			      struct hns3_desc_cb *cb)
+{
+	if (cb->type & (DESC_TYPE_SKB | DESC_TYPE_FRAGLIST_SKB))
+		dma_unmap_single(ring_to_dev(ring), cb->dma, cb->length,
+				 ring_to_dma_dir(ring));
+	else if ((cb->type & DESC_TYPE_PAGE) && cb->length)
+		dma_unmap_page(ring_to_dev(ring), cb->dma, cb->length,
+			       ring_to_dma_dir(ring));
+	else if (cb->type & (DESC_TYPE_BOUNCE_ALL | DESC_TYPE_BOUNCE_HEAD |
+			     DESC_TYPE_SGL_SKB))
+		hns3_tx_spare_reclaim_cb(ring, cb);
+}
+
+static void hns3_buffer_detach(struct hns3_enet_ring *ring, int i)
+{
+	hns3_unmap_buffer(ring, &ring->desc_cb[i]);
+	ring->desc[i].addr = 0;
+	ring->desc_cb[i].refill = 0;
+}
+
+static void hns3_free_buffer_detach(struct hns3_enet_ring *ring, int i,
+				    int budget)
+{
+	struct hns3_desc_cb *cb = &ring->desc_cb[i];
+
+	if (!ring->desc_cb[i].dma)
+		return;
+
+	hns3_buffer_detach(ring, i);
+	hns3_free_buffer(ring, cb, budget);
+}
+
+static void hns3_free_buffers(struct hns3_enet_ring *ring)
+{
+	int i;
+
+	for (i = 0; i < ring->desc_num; i++)
+		hns3_free_buffer_detach(ring, i, 0);
+}
+
+/* free desc along with its attached buffer */
+static void hns3_free_desc(struct hns3_enet_ring *ring)
+{
+	int size = ring->desc_num * sizeof(ring->desc[0]);
+
+	hns3_free_buffers(ring);
+
+	if (ring->desc) {
+		dma_free_coherent(ring_to_dev(ring), size,
+				  ring->desc, ring->desc_dma_addr);
+		ring->desc = NULL;
+	}
+}
+
+static int hns3_alloc_desc(struct hns3_enet_ring *ring)
+{
+	int size = ring->desc_num * sizeof(ring->desc[0]);
+
+	ring->desc = dma_alloc_coherent(ring_to_dev(ring), size,
+					&ring->desc_dma_addr, GFP_KERNEL);
+	if (!ring->desc)
+		return -ENOMEM;
+
+	return 0;
+}
+
+static int hns3_alloc_and_map_buffer(struct hns3_enet_ring *ring,
+				   struct hns3_desc_cb *cb)
+{
+	int ret;
+
+	ret = hns3_alloc_buffer(ring, cb);
+	if (ret || ring->page_pool)
+		goto out;
+
+	ret = hns3_map_buffer(ring, cb);
+	if (ret)
+		goto out_with_buf;
+
+	return 0;
+
+out_with_buf:
+	hns3_free_buffer(ring, cb, 0);
+out:
+	return ret;
+}
+
+static int hns3_alloc_and_attach_buffer(struct hns3_enet_ring *ring, int i)
+{
+	int ret = hns3_alloc_and_map_buffer(ring, &ring->desc_cb[i]);
+
+	if (ret)
+		return ret;
+
+	ring->desc[i].addr = cpu_to_le64(ring->desc_cb[i].dma +
+					 ring->desc_cb[i].page_offset);
+	ring->desc_cb[i].refill = 1;
+
+	return 0;
+}
+
+/* Allocate memory for raw pkg, and map with dma */
+static int hns3_alloc_ring_buffers(struct hns3_enet_ring *ring)
+{
+	int i, j, ret;
+
+	for (i = 0; i < ring->desc_num; i++) {
+		ret = hns3_alloc_and_attach_buffer(ring, i);
+		if (ret)
+			goto out_buffer_fail;
+
+		if (!(i % HNS3_RESCHED_BD_NUM))
+			cond_resched();
+	}
+
+	return 0;
+
+out_buffer_fail:
+	for (j = i - 1; j >= 0; j--)
+		hns3_free_buffer_detach(ring, j, 0);
+	return ret;
+}
+
+/* detach a in-used buffer and replace with a reserved one */
+static void hns3_replace_buffer(struct hns3_enet_ring *ring, int i,
+				struct hns3_desc_cb *res_cb)
+{
+	hns3_unmap_buffer(ring, &ring->desc_cb[i]);
+	ring->desc_cb[i] = *res_cb;
+	ring->desc_cb[i].refill = 1;
+	ring->desc[i].addr = cpu_to_le64(ring->desc_cb[i].dma +
+					 ring->desc_cb[i].page_offset);
+	ring->desc[i].rx.bd_base_info = 0;
+}
+
+static void hns3_reuse_buffer(struct hns3_enet_ring *ring, int i)
+{
+	ring->desc_cb[i].reuse_flag = 0;
+	ring->desc_cb[i].refill = 1;
+	ring->desc[i].addr = cpu_to_le64(ring->desc_cb[i].dma +
+					 ring->desc_cb[i].page_offset);
+	ring->desc[i].rx.bd_base_info = 0;
+
+	dma_sync_single_for_device(ring_to_dev(ring),
+			ring->desc_cb[i].dma + ring->desc_cb[i].page_offset,
+			hns3_buf_size(ring),
+			DMA_FROM_DEVICE);
+}
+
+static bool hns3_nic_reclaim_desc(struct hns3_enet_ring *ring,
+				  int *bytes, int *pkts, int budget)
+{
+	/* This smp_load_acquire() pairs with smp_store_release() in
+	 * hns3_tx_doorbell().
+	 */
+	int ltu = smp_load_acquire(&ring->last_to_use);
+	int ntc = ring->next_to_clean;
+	struct hns3_desc_cb *desc_cb;
+	bool reclaimed = false;
+	struct hns3_desc *desc;
+
+	while (ltu != ntc) {
+		desc = &ring->desc[ntc];
+
+		if (le16_to_cpu(desc->tx.bdtp_fe_sc_vld_ra_ri) &
+				BIT(HNS3_TXD_VLD_B))
+			break;
+
+		desc_cb = &ring->desc_cb[ntc];
+
+		if (desc_cb->type & (DESC_TYPE_SKB | DESC_TYPE_BOUNCE_ALL |
+				     DESC_TYPE_BOUNCE_HEAD |
+				     DESC_TYPE_SGL_SKB)) {
+			(*pkts)++;
+			(*bytes) += desc_cb->send_bytes;
+		}
+
+		/* desc_cb will be cleaned, after hnae3_free_buffer_detach */
+		hns3_free_buffer_detach(ring, ntc, budget);
+
+		if (++ntc == ring->desc_num)
+			ntc = 0;
+
+		/* Issue prefetch for next Tx descriptor */
+		prefetch(&ring->desc_cb[ntc]);
+		reclaimed = true;
+	}
+
+	if (unlikely(!reclaimed))
+		return false;
+
+	/* This smp_store_release() pairs with smp_load_acquire() in
+	 * ring_space called by hns3_nic_net_xmit.
+	 */
+	smp_store_release(&ring->next_to_clean, ntc);
+
+	hns3_tx_spare_update(ring);
+
+	return true;
+}
+
+void hns3_clean_tx_ring(struct hns3_enet_ring *ring, int budget)
+{
+	struct net_device *netdev = ring_to_netdev(ring);
+	struct hns3_nic_priv *priv = netdev_priv(netdev);
+	struct netdev_queue *dev_queue;
+	int bytes, pkts;
+
+	bytes = 0;
+	pkts = 0;
+
+	if (unlikely(!hns3_nic_reclaim_desc(ring, &bytes, &pkts, budget)))
+		return;
+
+	ring->tqp_vector->tx_group.total_bytes += bytes;
+	ring->tqp_vector->tx_group.total_packets += pkts;
+
+	u64_stats_update_begin(&ring->syncp);
+	ring->stats.tx_bytes += bytes;
+	ring->stats.tx_pkts += pkts;
+	u64_stats_update_end(&ring->syncp);
+
+	dev_queue = netdev_get_tx_queue(netdev, ring->tqp->tqp_index);
+	netdev_tx_completed_queue(dev_queue, pkts, bytes);
+
+	if (unlikely(netif_carrier_ok(netdev) &&
+		     ring_space(ring) > HNS3_MAX_TSO_BD_NUM)) {
+		/* Make sure that anybody stopping the queue after this
+		 * sees the new next_to_clean.
+		 */
+		smp_mb();
+		if (netif_tx_queue_stopped(dev_queue) &&
+		    !test_bit(HNS3_NIC_STATE_DOWN, &priv->state)) {
+			netif_tx_wake_queue(dev_queue);
+			ring->stats.restart_queue++;
+		}
+	}
+}
+
+static int hns3_desc_unused(struct hns3_enet_ring *ring)
+{
+	int ntc = ring->next_to_clean;
+	int ntu = ring->next_to_use;
+
+	if (unlikely(ntc == ntu && !ring->desc_cb[ntc].refill))
+		return ring->desc_num;
+
+	return ((ntc >= ntu) ? 0 : ring->desc_num) + ntc - ntu;
+}
+
+/* Return true if there is any allocation failure */
+static bool hns3_nic_alloc_rx_buffers(struct hns3_enet_ring *ring,
+				      int cleand_count)
+{
+	struct hns3_desc_cb *desc_cb;
+	struct hns3_desc_cb res_cbs;
+	int i, ret;
+
+	for (i = 0; i < cleand_count; i++) {
+		desc_cb = &ring->desc_cb[ring->next_to_use];
+		if (desc_cb->reuse_flag) {
+			hns3_ring_stats_update(ring, reuse_pg_cnt);
+
+			hns3_reuse_buffer(ring, ring->next_to_use);
+		} else {
+			ret = hns3_alloc_and_map_buffer(ring, &res_cbs);
+			if (ret) {
+				hns3_ring_stats_update(ring, sw_err_cnt);
+
+				hns3_rl_err(ring_to_netdev(ring),
+					    "alloc rx buffer failed: %d\n",
+					    ret);
+
+				writel(i, ring->tqp->io_base +
+				       HNS3_RING_RX_RING_HEAD_REG);
+				return true;
+			}
+			hns3_replace_buffer(ring, ring->next_to_use, &res_cbs);
+
+			hns3_ring_stats_update(ring, non_reuse_pg);
+		}
+
+		ring_ptr_move_fw(ring, next_to_use);
+	}
+
+	writel(i, ring->tqp->io_base + HNS3_RING_RX_RING_HEAD_REG);
+	return false;
+}
+
+static bool hns3_can_reuse_page(struct hns3_desc_cb *cb)
+{
+	return page_count(cb->priv) == cb->pagecnt_bias;
+}
+
+static int hns3_handle_rx_copybreak(struct sk_buff *skb, int i,
+				    struct hns3_enet_ring *ring,
+				    int pull_len,
+				    struct hns3_desc_cb *desc_cb)
+{
+	struct hns3_desc *desc = &ring->desc[ring->next_to_clean];
+	u32 frag_offset = desc_cb->page_offset + pull_len;
+	int size = le16_to_cpu(desc->rx.size);
+	u32 frag_size = size - pull_len;
+	void *frag = napi_alloc_frag(frag_size);
+
+	if (unlikely(!frag)) {
+		hns3_ring_stats_update(ring, frag_alloc_err);
+
+		hns3_rl_err(ring_to_netdev(ring),
+			    "failed to allocate rx frag\n");
+		return -ENOMEM;
+	}
+
+	desc_cb->reuse_flag = 1;
+	memcpy(frag, desc_cb->buf + frag_offset, frag_size);
+	skb_add_rx_frag(skb, i, virt_to_page(frag),
+			offset_in_page(frag), frag_size, frag_size);
+
+	hns3_ring_stats_update(ring, frag_alloc);
+	return 0;
+}
+
+static void hns3_nic_reuse_page(struct sk_buff *skb, int i,
+				struct hns3_enet_ring *ring, int pull_len,
+				struct hns3_desc_cb *desc_cb)
+{
+	struct hns3_desc *desc = &ring->desc[ring->next_to_clean];
+	u32 frag_offset = desc_cb->page_offset + pull_len;
+	int size = le16_to_cpu(desc->rx.size);
+	u32 truesize = hns3_buf_size(ring);
+	u32 frag_size = size - pull_len;
+	int ret = 0;
+	bool reused;
+
+	if (ring->page_pool) {
+		skb_add_rx_frag(skb, i, desc_cb->priv, frag_offset,
+				frag_size, truesize);
+		return;
+	}
+
+	/* Avoid re-using remote or pfmem page */
+	if (unlikely(!dev_page_is_reusable(desc_cb->priv)))
+		goto out;
+
+	reused = hns3_can_reuse_page(desc_cb);
+
+	/* Rx page can be reused when:
+	 * 1. Rx page is only owned by the driver when page_offset
+	 *    is zero, which means 0 @ truesize will be used by
+	 *    stack after skb_add_rx_frag() is called, and the rest
+	 *    of rx page can be reused by driver.
+	 * Or
+	 * 2. Rx page is only owned by the driver when page_offset
+	 *    is non-zero, which means page_offset @ truesize will
+	 *    be used by stack after skb_add_rx_frag() is called,
+	 *    and 0 @ truesize can be reused by driver.
+	 */
+	if ((!desc_cb->page_offset && reused) ||
+	    ((desc_cb->page_offset + truesize + truesize) <=
+	     hns3_page_size(ring) && desc_cb->page_offset)) {
+		desc_cb->page_offset += truesize;
+		desc_cb->reuse_flag = 1;
+	} else if (desc_cb->page_offset && reused) {
+		desc_cb->page_offset = 0;
+		desc_cb->reuse_flag = 1;
+	} else if (frag_size <= ring->rx_copybreak) {
+		ret = hns3_handle_rx_copybreak(skb, i, ring, pull_len, desc_cb);
+		if (!ret)
+			return;
+	}
+
+out:
+	desc_cb->pagecnt_bias--;
+
+	if (unlikely(!desc_cb->pagecnt_bias)) {
+		page_ref_add(desc_cb->priv, USHRT_MAX);
+		desc_cb->pagecnt_bias = USHRT_MAX;
+	}
+
+	skb_add_rx_frag(skb, i, desc_cb->priv, frag_offset,
+			frag_size, truesize);
+
+	if (unlikely(!desc_cb->reuse_flag))
+		__page_frag_cache_drain(desc_cb->priv, desc_cb->pagecnt_bias);
+}
+
+static int hns3_gro_complete(struct sk_buff *skb, u32 l234info)
+{
+	__be16 type = skb->protocol;
+	struct tcphdr *th;
+	u32 depth = 0;
+
+	while (eth_type_vlan(type)) {
+		struct vlan_hdr *vh;
+
+		if ((depth + VLAN_HLEN) > skb_headlen(skb))
+			return -EFAULT;
+
+		vh = (struct vlan_hdr *)(skb->data + depth);
+		type = vh->h_vlan_encapsulated_proto;
+		depth += VLAN_HLEN;
+	}
+
+	skb_set_network_header(skb, depth);
+
+	if (type == htons(ETH_P_IP)) {
+		const struct iphdr *iph = ip_hdr(skb);
+
+		depth += sizeof(struct iphdr);
+		skb_set_transport_header(skb, depth);
+		th = tcp_hdr(skb);
+		th->check = ~tcp_v4_check(skb->len - depth, iph->saddr,
+					  iph->daddr, 0);
+	} else if (type == htons(ETH_P_IPV6)) {
+		const struct ipv6hdr *iph = ipv6_hdr(skb);
+
+		depth += sizeof(struct ipv6hdr);
+		skb_set_transport_header(skb, depth);
+		th = tcp_hdr(skb);
+		th->check = ~tcp_v6_check(skb->len - depth, &iph->saddr,
+					  &iph->daddr, 0);
+	} else {
+		hns3_rl_err(skb->dev,
+			    "Error: FW GRO supports only IPv4/IPv6, not 0x%04x, depth: %d\n",
+			    be16_to_cpu(type), depth);
+		return -EFAULT;
+	}
+
+	skb_shinfo(skb)->gso_segs = NAPI_GRO_CB(skb)->count;
+	if (th->cwr)
+		skb_shinfo(skb)->gso_type |= SKB_GSO_TCP_ECN;
+
+	if (l234info & BIT(HNS3_RXD_GRO_FIXID_B))
+		skb_shinfo(skb)->gso_type |= SKB_GSO_TCP_FIXEDID;
+
+	skb->csum_start = (unsigned char *)th - skb->head;
+	skb->csum_offset = offsetof(struct tcphdr, check);
+	skb->ip_summed = CHECKSUM_PARTIAL;
+
+	trace_hns3_gro(skb);
+
+	return 0;
+}
+
+static void hns3_checksum_complete(struct hns3_enet_ring *ring,
+				   struct sk_buff *skb, u32 ptype, u16 csum)
+{
+	if (ptype == HNS3_INVALID_PTYPE ||
+	    hns3_rx_ptype_tbl[ptype].ip_summed != CHECKSUM_COMPLETE)
+		return;
+
+	hns3_ring_stats_update(ring, csum_complete);
+	skb->ip_summed = CHECKSUM_COMPLETE;
+	skb->csum = csum_unfold((__force __sum16)csum);
+}
+
+static void hns3_rx_handle_csum(struct sk_buff *skb, u32 l234info,
+				u32 ol_info, u32 ptype)
+{
+	int l3_type, l4_type;
+	int ol4_type;
+
+	if (ptype != HNS3_INVALID_PTYPE) {
+		skb->csum_level = hns3_rx_ptype_tbl[ptype].csum_level;
+		skb->ip_summed = hns3_rx_ptype_tbl[ptype].ip_summed;
+
+		return;
+	}
+
+	ol4_type = hnae3_get_field(ol_info, HNS3_RXD_OL4ID_M,
+				   HNS3_RXD_OL4ID_S);
+	switch (ol4_type) {
+	case HNS3_OL4_TYPE_MAC_IN_UDP:
+	case HNS3_OL4_TYPE_NVGRE:
+		skb->csum_level = 1;
+		fallthrough;
+	case HNS3_OL4_TYPE_NO_TUN:
+		l3_type = hnae3_get_field(l234info, HNS3_RXD_L3ID_M,
+					  HNS3_RXD_L3ID_S);
+		l4_type = hnae3_get_field(l234info, HNS3_RXD_L4ID_M,
+					  HNS3_RXD_L4ID_S);
+		/* Can checksum ipv4 or ipv6 + UDP/TCP/SCTP packets */
+		if ((l3_type == HNS3_L3_TYPE_IPV4 ||
+		     l3_type == HNS3_L3_TYPE_IPV6) &&
+		    (l4_type == HNS3_L4_TYPE_UDP ||
+		     l4_type == HNS3_L4_TYPE_TCP ||
+		     l4_type == HNS3_L4_TYPE_SCTP))
+			skb->ip_summed = CHECKSUM_UNNECESSARY;
+		break;
+	default:
+		break;
+	}
+}
+
+static void hns3_rx_checksum(struct hns3_enet_ring *ring, struct sk_buff *skb,
+			     u32 l234info, u32 bd_base_info, u32 ol_info,
+			     u16 csum)
+{
+	struct net_device *netdev = ring_to_netdev(ring);
+	struct hns3_nic_priv *priv = netdev_priv(netdev);
+	u32 ptype = HNS3_INVALID_PTYPE;
+
+	skb->ip_summed = CHECKSUM_NONE;
+
+	skb_checksum_none_assert(skb);
+
+	if (!(netdev->features & NETIF_F_RXCSUM))
+		return;
+
+	if (test_bit(HNS3_NIC_STATE_RXD_ADV_LAYOUT_ENABLE, &priv->state))
+		ptype = hnae3_get_field(ol_info, HNS3_RXD_PTYPE_M,
+					HNS3_RXD_PTYPE_S);
+
+	hns3_checksum_complete(ring, skb, ptype, csum);
+
+	/* check if hardware has done checksum */
+	if (!(bd_base_info & BIT(HNS3_RXD_L3L4P_B)))
+		return;
+
+	if (unlikely(l234info & (BIT(HNS3_RXD_L3E_B) | BIT(HNS3_RXD_L4E_B) |
+				 BIT(HNS3_RXD_OL3E_B) |
+				 BIT(HNS3_RXD_OL4E_B)))) {
+		skb->ip_summed = CHECKSUM_NONE;
+		hns3_ring_stats_update(ring, l3l4_csum_err);
+
+		return;
+	}
+
+	hns3_rx_handle_csum(skb, l234info, ol_info, ptype);
+}
+
+static void hns3_rx_skb(struct hns3_enet_ring *ring, struct sk_buff *skb)
+{
+	if (skb_has_frag_list(skb))
+		napi_gro_flush(&ring->tqp_vector->napi, false);
+
+	napi_gro_receive(&ring->tqp_vector->napi, skb);
+}
+
+static bool hns3_parse_vlan_tag(struct hns3_enet_ring *ring,
+				struct hns3_desc *desc, u32 l234info,
+				u16 *vlan_tag)
+{
+	struct hnae3_handle *handle = ring->tqp->handle;
+	struct pci_dev *pdev = ring->tqp->handle->pdev;
+	struct hnae3_ae_dev *ae_dev = pci_get_drvdata(pdev);
+
+	if (unlikely(ae_dev->dev_version < HNAE3_DEVICE_VERSION_V2)) {
+		*vlan_tag = le16_to_cpu(desc->rx.ot_vlan_tag);
+		if (!(*vlan_tag & VLAN_VID_MASK))
+			*vlan_tag = le16_to_cpu(desc->rx.vlan_tag);
+
+		return (*vlan_tag != 0);
+	}
+
+#define HNS3_STRP_OUTER_VLAN	0x1
+#define HNS3_STRP_INNER_VLAN	0x2
+#define HNS3_STRP_BOTH		0x3
+
+	/* Hardware always insert VLAN tag into RX descriptor when
+	 * remove the tag from packet, driver needs to determine
+	 * reporting which tag to stack.
+	 */
+	switch (hnae3_get_field(l234info, HNS3_RXD_STRP_TAGP_M,
+				HNS3_RXD_STRP_TAGP_S)) {
+	case HNS3_STRP_OUTER_VLAN:
+		if (handle->port_base_vlan_state !=
+				HNAE3_PORT_BASE_VLAN_DISABLE)
+			return false;
+
+		*vlan_tag = le16_to_cpu(desc->rx.ot_vlan_tag);
+		return true;
+	case HNS3_STRP_INNER_VLAN:
+		if (handle->port_base_vlan_state !=
+				HNAE3_PORT_BASE_VLAN_DISABLE)
+			return false;
+
+		*vlan_tag = le16_to_cpu(desc->rx.vlan_tag);
+		return true;
+	case HNS3_STRP_BOTH:
+		if (handle->port_base_vlan_state ==
+				HNAE3_PORT_BASE_VLAN_DISABLE)
+			*vlan_tag = le16_to_cpu(desc->rx.ot_vlan_tag);
+		else
+			*vlan_tag = le16_to_cpu(desc->rx.vlan_tag);
+
+		return true;
+	default:
+		return false;
+	}
+}
+
+static void hns3_rx_ring_move_fw(struct hns3_enet_ring *ring)
+{
+	ring->desc[ring->next_to_clean].rx.bd_base_info &=
+		cpu_to_le32(~BIT(HNS3_RXD_VLD_B));
+	ring->desc_cb[ring->next_to_clean].refill = 0;
+	ring->next_to_clean += 1;
+
+	if (unlikely(ring->next_to_clean == ring->desc_num))
+		ring->next_to_clean = 0;
+}
+
+static int hns3_alloc_skb(struct hns3_enet_ring *ring, unsigned int length,
+			  unsigned char *va)
+{
+	struct hns3_desc_cb *desc_cb = &ring->desc_cb[ring->next_to_clean];
+	struct net_device *netdev = ring_to_netdev(ring);
+	struct sk_buff *skb;
+
+	ring->skb = napi_alloc_skb(&ring->tqp_vector->napi, HNS3_RX_HEAD_SIZE);
+	skb = ring->skb;
+	if (unlikely(!skb)) {
+		hns3_rl_err(netdev, "alloc rx skb fail\n");
+		hns3_ring_stats_update(ring, sw_err_cnt);
+
+		return -ENOMEM;
+	}
+
+	trace_hns3_rx_desc(ring);
+	prefetchw(skb->data);
+
+	ring->pending_buf = 1;
+	ring->frag_num = 0;
+	ring->tail_skb = NULL;
+	if (length <= HNS3_RX_HEAD_SIZE) {
+		memcpy(__skb_put(skb, length), va, ALIGN(length, sizeof(long)));
+
+		/* We can reuse buffer as-is, just make sure it is reusable */
+		if (dev_page_is_reusable(desc_cb->priv))
+			desc_cb->reuse_flag = 1;
+		else if (desc_cb->type & DESC_TYPE_PP_FRAG)
+			page_pool_put_full_page(ring->page_pool, desc_cb->priv,
+						false);
+		else /* This page cannot be reused so discard it */
+			__page_frag_cache_drain(desc_cb->priv,
+						desc_cb->pagecnt_bias);
+
+		hns3_rx_ring_move_fw(ring);
+		return 0;
+	}
+
+	if (ring->page_pool)
+		skb_mark_for_recycle(skb);
+
+	hns3_ring_stats_update(ring, seg_pkt_cnt);
+
+	ring->pull_len = eth_get_headlen(netdev, va, HNS3_RX_HEAD_SIZE);
+	__skb_put(skb, ring->pull_len);
+	hns3_nic_reuse_page(skb, ring->frag_num++, ring, ring->pull_len,
+			    desc_cb);
+	hns3_rx_ring_move_fw(ring);
+
+	return 0;
+}
+
+static int hns3_add_frag(struct hns3_enet_ring *ring)
+{
+	struct sk_buff *skb = ring->skb;
+	struct sk_buff *head_skb = skb;
+	struct sk_buff *new_skb;
+	struct hns3_desc_cb *desc_cb;
+	struct hns3_desc *desc;
+	u32 bd_base_info;
+
+	do {
+		desc = &ring->desc[ring->next_to_clean];
+		desc_cb = &ring->desc_cb[ring->next_to_clean];
+		bd_base_info = le32_to_cpu(desc->rx.bd_base_info);
+		/* make sure HW write desc complete */
+		dma_rmb();
+		if (!(bd_base_info & BIT(HNS3_RXD_VLD_B)))
+			return -ENXIO;
+
+		if (unlikely(ring->frag_num >= MAX_SKB_FRAGS)) {
+			new_skb = napi_alloc_skb(&ring->tqp_vector->napi, 0);
+			if (unlikely(!new_skb)) {
+				hns3_rl_err(ring_to_netdev(ring),
+					    "alloc rx fraglist skb fail\n");
+				return -ENXIO;
+			}
+
+			if (ring->page_pool)
+				skb_mark_for_recycle(new_skb);
+
+			ring->frag_num = 0;
+
+			if (ring->tail_skb) {
+				ring->tail_skb->next = new_skb;
+				ring->tail_skb = new_skb;
+			} else {
+				skb_shinfo(skb)->frag_list = new_skb;
+				ring->tail_skb = new_skb;
+			}
+		}
+
+		if (ring->tail_skb) {
+			head_skb->truesize += hns3_buf_size(ring);
+			head_skb->data_len += le16_to_cpu(desc->rx.size);
+			head_skb->len += le16_to_cpu(desc->rx.size);
+			skb = ring->tail_skb;
+		}
+
+		dma_sync_single_for_cpu(ring_to_dev(ring),
+				desc_cb->dma + desc_cb->page_offset,
+				hns3_buf_size(ring),
+				DMA_FROM_DEVICE);
+
+		hns3_nic_reuse_page(skb, ring->frag_num++, ring, 0, desc_cb);
+		trace_hns3_rx_desc(ring);
+		hns3_rx_ring_move_fw(ring);
+		ring->pending_buf++;
+	} while (!(bd_base_info & BIT(HNS3_RXD_FE_B)));
+
+	return 0;
+}
+
+static int hns3_set_gro_and_checksum(struct hns3_enet_ring *ring,
+				     struct sk_buff *skb, u32 l234info,
+				     u32 bd_base_info, u32 ol_info, u16 csum)
+{
+	struct net_device *netdev = ring_to_netdev(ring);
+	struct hns3_nic_priv *priv = netdev_priv(netdev);
+	u32 l3_type;
+
+	skb_shinfo(skb)->gso_size = hnae3_get_field(bd_base_info,
+						    HNS3_RXD_GRO_SIZE_M,
+						    HNS3_RXD_GRO_SIZE_S);
+	/* if there is no HW GRO, do not set gro params */
+	if (!skb_shinfo(skb)->gso_size) {
+		hns3_rx_checksum(ring, skb, l234info, bd_base_info, ol_info,
+				 csum);
+		return 0;
+	}
+
+	NAPI_GRO_CB(skb)->count = hnae3_get_field(l234info,
+						  HNS3_RXD_GRO_COUNT_M,
+						  HNS3_RXD_GRO_COUNT_S);
+
+	if (test_bit(HNS3_NIC_STATE_RXD_ADV_LAYOUT_ENABLE, &priv->state)) {
+		u32 ptype = hnae3_get_field(ol_info, HNS3_RXD_PTYPE_M,
+					    HNS3_RXD_PTYPE_S);
+
+		l3_type = hns3_rx_ptype_tbl[ptype].l3_type;
+	} else {
+		l3_type = hnae3_get_field(l234info, HNS3_RXD_L3ID_M,
+					  HNS3_RXD_L3ID_S);
+	}
+
+	if (l3_type == HNS3_L3_TYPE_IPV4)
+		skb_shinfo(skb)->gso_type = SKB_GSO_TCPV4;
+	else if (l3_type == HNS3_L3_TYPE_IPV6)
+		skb_shinfo(skb)->gso_type = SKB_GSO_TCPV6;
+	else
+		return -EFAULT;
+
+	return  hns3_gro_complete(skb, l234info);
+}
+
+static void hns3_set_rx_skb_rss_type(struct hns3_enet_ring *ring,
+				     struct sk_buff *skb, u32 rss_hash,
+				     u32 l234info, u32 ol_info)
+{
+	enum pkt_hash_types rss_type = PKT_HASH_TYPE_NONE;
+	struct net_device *netdev = ring_to_netdev(ring);
+	struct hns3_nic_priv *priv = netdev_priv(netdev);
+
+	if (test_bit(HNS3_NIC_STATE_RXD_ADV_LAYOUT_ENABLE, &priv->state)) {
+		u32 ptype = hnae3_get_field(ol_info, HNS3_RXD_PTYPE_M,
+					    HNS3_RXD_PTYPE_S);
+
+		rss_type = hns3_rx_ptype_tbl[ptype].hash_type;
+	} else {
+		int l3_type = hnae3_get_field(l234info, HNS3_RXD_L3ID_M,
+					      HNS3_RXD_L3ID_S);
+		int l4_type = hnae3_get_field(l234info, HNS3_RXD_L4ID_M,
+					      HNS3_RXD_L4ID_S);
+
+		if (l3_type == HNS3_L3_TYPE_IPV4 ||
+		    l3_type == HNS3_L3_TYPE_IPV6) {
+			if (l4_type == HNS3_L4_TYPE_UDP ||
+			    l4_type == HNS3_L4_TYPE_TCP ||
+			    l4_type == HNS3_L4_TYPE_SCTP)
+				rss_type = PKT_HASH_TYPE_L4;
+			else if (l4_type == HNS3_L4_TYPE_IGMP ||
+				 l4_type == HNS3_L4_TYPE_ICMP)
+				rss_type = PKT_HASH_TYPE_L3;
+		}
+	}
+
+	skb_set_hash(skb, rss_hash, rss_type);
+}
+
+static void hns3_handle_rx_ts_info(struct net_device *netdev,
+				   struct hns3_desc *desc, struct sk_buff *skb,
+				   u32 bd_base_info)
+{
+	if (unlikely(bd_base_info & BIT(HNS3_RXD_TS_VLD_B))) {
+		struct hnae3_handle *h = hns3_get_handle(netdev);
+		u32 nsec = le32_to_cpu(desc->ts_nsec);
+		u32 sec = le32_to_cpu(desc->ts_sec);
+
+		if (h->ae_algo->ops->get_rx_hwts)
+			h->ae_algo->ops->get_rx_hwts(h, skb, nsec, sec);
+	}
+}
+
+static void hns3_handle_rx_vlan_tag(struct hns3_enet_ring *ring,
+				    struct hns3_desc *desc, struct sk_buff *skb,
+				    u32 l234info)
+{
+	struct net_device *netdev = ring_to_netdev(ring);
+
+	/* Based on hw strategy, the tag offloaded will be stored at
+	 * ot_vlan_tag in two layer tag case, and stored at vlan_tag
+	 * in one layer tag case.
+	 */
+	if (netdev->features & NETIF_F_HW_VLAN_CTAG_RX) {
+		u16 vlan_tag;
+
+		if (hns3_parse_vlan_tag(ring, desc, l234info, &vlan_tag))
+			__vlan_hwaccel_put_tag(skb, htons(ETH_P_8021Q),
+					       vlan_tag);
+	}
+}
+
+static int hns3_handle_bdinfo(struct hns3_enet_ring *ring, struct sk_buff *skb)
+{
+	struct net_device *netdev = ring_to_netdev(ring);
+	enum hns3_pkt_l2t_type l2_frame_type;
+	u32 bd_base_info, l234info, ol_info;
+	struct hns3_desc *desc;
+	unsigned int len;
+	int pre_ntc, ret;
+	u16 csum;
+
+	/* bdinfo handled below is only valid on the last BD of the
+	 * current packet, and ring->next_to_clean indicates the first
+	 * descriptor of next packet, so need - 1 below.
+	 */
+	pre_ntc = ring->next_to_clean ? (ring->next_to_clean - 1) :
+					(ring->desc_num - 1);
+	desc = &ring->desc[pre_ntc];
+	bd_base_info = le32_to_cpu(desc->rx.bd_base_info);
+	l234info = le32_to_cpu(desc->rx.l234_info);
+	ol_info = le32_to_cpu(desc->rx.ol_info);
+	csum = le16_to_cpu(desc->csum);
+
+	hns3_handle_rx_ts_info(netdev, desc, skb, bd_base_info);
+
+	hns3_handle_rx_vlan_tag(ring, desc, skb, l234info);
+
+	if (unlikely(!desc->rx.pkt_len || (l234info & (BIT(HNS3_RXD_TRUNCAT_B) |
+				  BIT(HNS3_RXD_L2E_B))))) {
+		u64_stats_update_begin(&ring->syncp);
+		if (l234info & BIT(HNS3_RXD_L2E_B))
+			ring->stats.l2_err++;
+		else
+			ring->stats.err_pkt_len++;
+		u64_stats_update_end(&ring->syncp);
+
+		return -EFAULT;
+	}
+
+	len = skb->len;
+
+	/* Do update ip stack process */
+	skb->protocol = eth_type_trans(skb, netdev);
+
+	/* This is needed in order to enable forwarding support */
+	ret = hns3_set_gro_and_checksum(ring, skb, l234info,
+					bd_base_info, ol_info, csum);
+	if (unlikely(ret)) {
+		hns3_ring_stats_update(ring, rx_err_cnt);
+		return ret;
+	}
+
+	l2_frame_type = hnae3_get_field(l234info, HNS3_RXD_DMAC_M,
+					HNS3_RXD_DMAC_S);
+
+	u64_stats_update_begin(&ring->syncp);
+	ring->stats.rx_pkts++;
+	ring->stats.rx_bytes += len;
+
+	if (l2_frame_type == HNS3_L2_TYPE_MULTICAST)
+		ring->stats.rx_multicast++;
+
+	u64_stats_update_end(&ring->syncp);
+
+	ring->tqp_vector->rx_group.total_bytes += len;
+
+	hns3_set_rx_skb_rss_type(ring, skb, le32_to_cpu(desc->rx.rss_hash),
+				 l234info, ol_info);
+	return 0;
+}
+
+static int hns3_handle_rx_bd(struct hns3_enet_ring *ring)
+{
+	struct sk_buff *skb = ring->skb;
+	struct hns3_desc_cb *desc_cb;
+	struct hns3_desc *desc;
+	unsigned int length;
+	u32 bd_base_info;
+	int ret;
+
+	desc = &ring->desc[ring->next_to_clean];
+	desc_cb = &ring->desc_cb[ring->next_to_clean];
+
+	prefetch(desc);
+
+	if (!skb) {
+		bd_base_info = le32_to_cpu(desc->rx.bd_base_info);
+		/* Check valid BD */
+		if (unlikely(!(bd_base_info & BIT(HNS3_RXD_VLD_B))))
+			return -ENXIO;
+
+		dma_rmb();
+		length = le16_to_cpu(desc->rx.size);
+
+		ring->va = desc_cb->buf + desc_cb->page_offset;
+
+		dma_sync_single_for_cpu(ring_to_dev(ring),
+				desc_cb->dma + desc_cb->page_offset,
+				hns3_buf_size(ring),
+				DMA_FROM_DEVICE);
+
+		/* Prefetch first cache line of first page.
+		 * Idea is to cache few bytes of the header of the packet.
+		 * Our L1 Cache line size is 64B so need to prefetch twice to make
+		 * it 128B. But in actual we can have greater size of caches with
+		 * 128B Level 1 cache lines. In such a case, single fetch would
+		 * suffice to cache in the relevant part of the header.
+		 */
+		net_prefetch(ring->va);
+
+		ret = hns3_alloc_skb(ring, length, ring->va);
+		skb = ring->skb;
+
+		if (ret < 0) /* alloc buffer fail */
+			return ret;
+		if (!(bd_base_info & BIT(HNS3_RXD_FE_B))) { /* need add frag */
+			ret = hns3_add_frag(ring);
+			if (ret)
+				return ret;
+		}
+	} else {
+		ret = hns3_add_frag(ring);
+		if (ret)
+			return ret;
+	}
+
+	/* As the head data may be changed when GRO enable, copy
+	 * the head data in after other data rx completed
+	 */
+	if (skb->len > HNS3_RX_HEAD_SIZE)
+		memcpy(skb->data, ring->va,
+		       ALIGN(ring->pull_len, sizeof(long)));
+
+	ret = hns3_handle_bdinfo(ring, skb);
+	if (unlikely(ret)) {
+		dev_kfree_skb_any(skb);
+		return ret;
+	}
+
+	skb_record_rx_queue(skb, ring->tqp->tqp_index);
+	return 0;
+}
+
+int hns3_clean_rx_ring(struct hns3_enet_ring *ring, int budget,
+		       void (*rx_fn)(struct hns3_enet_ring *, struct sk_buff *))
+{
+#define RCB_NOF_ALLOC_RX_BUFF_ONCE 16
+	int unused_count = hns3_desc_unused(ring);
+	bool failure = false;
+	int recv_pkts = 0;
+	int err;
+
+	unused_count -= ring->pending_buf;
+
+	while (recv_pkts < budget) {
+		/* Reuse or realloc buffers */
+		if (unused_count >= RCB_NOF_ALLOC_RX_BUFF_ONCE) {
+			failure = failure ||
+				hns3_nic_alloc_rx_buffers(ring, unused_count);
+			unused_count = 0;
+		}
+
+		/* Poll one pkt */
+		err = hns3_handle_rx_bd(ring);
+		/* Do not get FE for the packet or failed to alloc skb */
+		if (unlikely(!ring->skb || err == -ENXIO)) {
+			goto out;
+		} else if (likely(!err)) {
+			rx_fn(ring, ring->skb);
+			recv_pkts++;
+		}
+
+		unused_count += ring->pending_buf;
+		ring->skb = NULL;
+		ring->pending_buf = 0;
+	}
+
+out:
+	/* sync head pointer before exiting, since hardware will calculate
+	 * FBD number with head pointer
+	 */
+	if (unused_count > 0)
+		failure = failure ||
+			  hns3_nic_alloc_rx_buffers(ring, unused_count);
+
+	return failure ? budget : recv_pkts;
+}
+
+static void hns3_update_rx_int_coalesce(struct hns3_enet_tqp_vector *tqp_vector)
+{
+	struct hns3_enet_ring_group *rx_group = &tqp_vector->rx_group;
+	struct dim_sample sample = {};
+
+	if (!rx_group->coal.adapt_enable)
+		return;
+
+	dim_update_sample(tqp_vector->event_cnt, rx_group->total_packets,
+			  rx_group->total_bytes, &sample);
+	net_dim(&rx_group->dim, &sample);
+}
+
+static void hns3_update_tx_int_coalesce(struct hns3_enet_tqp_vector *tqp_vector)
+{
+	struct hns3_enet_ring_group *tx_group = &tqp_vector->tx_group;
+	struct dim_sample sample = {};
+
+	if (!tx_group->coal.adapt_enable)
+		return;
+
+	dim_update_sample(tqp_vector->event_cnt, tx_group->total_packets,
+			  tx_group->total_bytes, &sample);
+	net_dim(&tx_group->dim, &sample);
+}
+
+static int hns3_nic_common_poll(struct napi_struct *napi, int budget)
+{
+	struct hns3_nic_priv *priv = netdev_priv(napi->dev);
+	struct hns3_enet_ring *ring;
+	int rx_pkt_total = 0;
+
+	struct hns3_enet_tqp_vector *tqp_vector =
+		container_of(napi, struct hns3_enet_tqp_vector, napi);
+	bool clean_complete = true;
+	int rx_budget = budget;
+
+	if (unlikely(test_bit(HNS3_NIC_STATE_DOWN, &priv->state))) {
+		napi_complete(napi);
+		return 0;
+	}
+
+	/* Since the actual Tx work is minimal, we can give the Tx a larger
+	 * budget and be more aggressive about cleaning up the Tx descriptors.
+	 */
+	hns3_for_each_ring(ring, tqp_vector->tx_group)
+		hns3_clean_tx_ring(ring, budget);
+
+	/* make sure rx ring budget not smaller than 1 */
+	if (tqp_vector->num_tqps > 1)
+		rx_budget = max(budget / tqp_vector->num_tqps, 1);
+
+	hns3_for_each_ring(ring, tqp_vector->rx_group) {
+		int rx_cleaned = hns3_clean_rx_ring(ring, rx_budget,
+						    hns3_rx_skb);
+		if (rx_cleaned >= rx_budget)
+			clean_complete = false;
+
+		rx_pkt_total += rx_cleaned;
+	}
+
+	tqp_vector->rx_group.total_packets += rx_pkt_total;
+
+	if (!clean_complete)
+		return budget;
+
+	if (napi_complete(napi) &&
+	    likely(!test_bit(HNS3_NIC_STATE_DOWN, &priv->state))) {
+		hns3_update_rx_int_coalesce(tqp_vector);
+		hns3_update_tx_int_coalesce(tqp_vector);
+
+		hns3_mask_vector_irq(tqp_vector, 1);
+	}
+
+	return rx_pkt_total;
+}
+
+static int hns3_create_ring_chain(struct hns3_enet_tqp_vector *tqp_vector,
+				  struct hnae3_ring_chain_node **head,
+				  bool is_tx)
+{
+	u32 bit_value = is_tx ? HNAE3_RING_TYPE_TX : HNAE3_RING_TYPE_RX;
+	u32 field_value = is_tx ? HNAE3_RING_GL_TX : HNAE3_RING_GL_RX;
+	struct hnae3_ring_chain_node *cur_chain = *head;
+	struct pci_dev *pdev = tqp_vector->handle->pdev;
+	struct hnae3_ring_chain_node *chain;
+	struct hns3_enet_ring *ring;
+
+	ring = is_tx ? tqp_vector->tx_group.ring : tqp_vector->rx_group.ring;
+
+	if (cur_chain) {
+		while (cur_chain->next)
+			cur_chain = cur_chain->next;
+	}
+
+	while (ring) {
+		chain = devm_kzalloc(&pdev->dev, sizeof(*chain), GFP_KERNEL);
+		if (!chain)
+			return -ENOMEM;
+		if (cur_chain)
+			cur_chain->next = chain;
+		else
+			*head = chain;
+		chain->tqp_index = ring->tqp->tqp_index;
+		hnae3_set_bit(chain->flag, HNAE3_RING_TYPE_B,
+				bit_value);
+		hnae3_set_field(chain->int_gl_idx,
+				HNAE3_RING_GL_IDX_M,
+				HNAE3_RING_GL_IDX_S, field_value);
+
+		cur_chain = chain;
+
+		ring = ring->next;
+	}
+
+	return 0;
+}
+
+static struct hnae3_ring_chain_node *
+hns3_get_vector_ring_chain(struct hns3_enet_tqp_vector *tqp_vector)
+{
+	struct pci_dev *pdev = tqp_vector->handle->pdev;
+	struct hnae3_ring_chain_node *cur_chain = NULL;
+	struct hnae3_ring_chain_node *chain;
+
+	if (hns3_create_ring_chain(tqp_vector, &cur_chain, true))
+		goto err_free_chain;
+
+	if (hns3_create_ring_chain(tqp_vector, &cur_chain, false))
+		goto err_free_chain;
+
+	return cur_chain;
+
+err_free_chain:
+	while (cur_chain) {
+		chain = cur_chain->next;
+		devm_kfree(&pdev->dev, cur_chain);
+		cur_chain = chain;
+	}
+
+	return NULL;
+}
+
+static void hns3_free_vector_ring_chain(struct hns3_enet_tqp_vector *tqp_vector,
+					struct hnae3_ring_chain_node *head)
+{
+	struct pci_dev *pdev = tqp_vector->handle->pdev;
+	struct hnae3_ring_chain_node *chain_tmp, *chain;
+
+	chain = head;
+
+	while (chain) {
+		chain_tmp = chain->next;
+		devm_kfree(&pdev->dev, chain);
+		chain = chain_tmp;
+	}
+}
+
+static void hns3_add_ring_to_group(struct hns3_enet_ring_group *group,
+				   struct hns3_enet_ring *ring)
+{
+	ring->next = group->ring;
+	group->ring = ring;
+
+	group->count++;
+}
+
+static void hns3_nic_set_cpumask(struct hns3_nic_priv *priv)
+{
+	struct pci_dev *pdev = priv->ae_handle->pdev;
+	struct hns3_enet_tqp_vector *tqp_vector;
+	int num_vectors = priv->vector_num;
+	int numa_node;
+	int vector_i;
+
+	numa_node = dev_to_node(&pdev->dev);
+
+	for (vector_i = 0; vector_i < num_vectors; vector_i++) {
+		tqp_vector = &priv->tqp_vector[vector_i];
+		cpumask_set_cpu(cpumask_local_spread(vector_i, numa_node),
+				&tqp_vector->affinity_mask);
+	}
+}
+
+static void hns3_rx_dim_work(struct work_struct *work)
+{
+	struct dim *dim = container_of(work, struct dim, work);
+	struct hns3_enet_ring_group *group = container_of(dim,
+		struct hns3_enet_ring_group, dim);
+	struct hns3_enet_tqp_vector *tqp_vector = group->ring->tqp_vector;
+	struct dim_cq_moder cur_moder =
+		net_dim_get_rx_moderation(dim->mode, dim->profile_ix);
+
+	hns3_set_vector_coalesce_rx_gl(group->ring->tqp_vector, cur_moder.usec);
+	tqp_vector->rx_group.coal.int_gl = cur_moder.usec;
+
+	if (cur_moder.pkts < tqp_vector->rx_group.coal.int_ql_max) {
+		hns3_set_vector_coalesce_rx_ql(tqp_vector, cur_moder.pkts);
+		tqp_vector->rx_group.coal.int_ql = cur_moder.pkts;
+	}
+
+	dim->state = DIM_START_MEASURE;
+}
+
+static void hns3_tx_dim_work(struct work_struct *work)
+{
+	struct dim *dim = container_of(work, struct dim, work);
+	struct hns3_enet_ring_group *group = container_of(dim,
+		struct hns3_enet_ring_group, dim);
+	struct hns3_enet_tqp_vector *tqp_vector = group->ring->tqp_vector;
+	struct dim_cq_moder cur_moder =
+		net_dim_get_tx_moderation(dim->mode, dim->profile_ix);
+
+	hns3_set_vector_coalesce_tx_gl(tqp_vector, cur_moder.usec);
+	tqp_vector->tx_group.coal.int_gl = cur_moder.usec;
+
+	if (cur_moder.pkts < tqp_vector->tx_group.coal.int_ql_max) {
+		hns3_set_vector_coalesce_tx_ql(tqp_vector, cur_moder.pkts);
+		tqp_vector->tx_group.coal.int_ql = cur_moder.pkts;
+	}
+
+	dim->state = DIM_START_MEASURE;
+}
+
+static void hns3_nic_init_dim(struct hns3_enet_tqp_vector *tqp_vector)
+{
+	INIT_WORK(&tqp_vector->rx_group.dim.work, hns3_rx_dim_work);
+	INIT_WORK(&tqp_vector->tx_group.dim.work, hns3_tx_dim_work);
+}
+
+static int hns3_nic_init_vector_data(struct hns3_nic_priv *priv)
+{
+	struct hnae3_handle *h = priv->ae_handle;
+	struct hns3_enet_tqp_vector *tqp_vector;
+	int ret;
+	int i;
+
+	hns3_nic_set_cpumask(priv);
+
+	for (i = 0; i < priv->vector_num; i++) {
+		tqp_vector = &priv->tqp_vector[i];
+		hns3_vector_coalesce_init_hw(tqp_vector, priv);
+		tqp_vector->num_tqps = 0;
+		hns3_nic_init_dim(tqp_vector);
+	}
+
+	for (i = 0; i < h->kinfo.num_tqps; i++) {
+		u16 vector_i = i % priv->vector_num;
+		u16 tqp_num = h->kinfo.num_tqps;
+
+		tqp_vector = &priv->tqp_vector[vector_i];
+
+		hns3_add_ring_to_group(&tqp_vector->tx_group,
+				       &priv->ring[i]);
+
+		hns3_add_ring_to_group(&tqp_vector->rx_group,
+				       &priv->ring[i + tqp_num]);
+
+		priv->ring[i].tqp_vector = tqp_vector;
+		priv->ring[i + tqp_num].tqp_vector = tqp_vector;
+		tqp_vector->num_tqps++;
+	}
+
+	for (i = 0; i < priv->vector_num; i++) {
+		struct hnae3_ring_chain_node *vector_ring_chain;
+
+		tqp_vector = &priv->tqp_vector[i];
+
+		tqp_vector->rx_group.total_bytes = 0;
+		tqp_vector->rx_group.total_packets = 0;
+		tqp_vector->tx_group.total_bytes = 0;
+		tqp_vector->tx_group.total_packets = 0;
+		tqp_vector->handle = h;
+
+		vector_ring_chain = hns3_get_vector_ring_chain(tqp_vector);
+		if (!vector_ring_chain) {
+			ret = -ENOMEM;
+			goto map_ring_fail;
+		}
+
+		ret = h->ae_algo->ops->map_ring_to_vector(h,
+			tqp_vector->vector_irq, vector_ring_chain);
+
+		hns3_free_vector_ring_chain(tqp_vector, vector_ring_chain);
+
+		if (ret)
+			goto map_ring_fail;
+
+		netif_napi_add(priv->netdev, &tqp_vector->napi,
+			       hns3_nic_common_poll);
+	}
+
+	return 0;
+
+map_ring_fail:
+	while (i--)
+		netif_napi_del(&priv->tqp_vector[i].napi);
+
+	return ret;
+}
+
+static void hns3_nic_init_coal_cfg(struct hns3_nic_priv *priv)
+{
+	struct hnae3_ae_dev *ae_dev = hns3_get_ae_dev(priv->ae_handle);
+	struct hns3_enet_coalesce *tx_coal = &priv->tx_coal;
+	struct hns3_enet_coalesce *rx_coal = &priv->rx_coal;
+
+	/* initialize the configuration for interrupt coalescing.
+	 * 1. GL (Interrupt Gap Limiter)
+	 * 2. RL (Interrupt Rate Limiter)
+	 * 3. QL (Interrupt Quantity Limiter)
+	 *
+	 * Default: enable interrupt coalescing self-adaptive and GL
+	 */
+	tx_coal->adapt_enable = 1;
+	rx_coal->adapt_enable = 1;
+
+	tx_coal->int_gl = HNS3_INT_GL_50K;
+	rx_coal->int_gl = HNS3_INT_GL_50K;
+
+	rx_coal->flow_level = HNS3_FLOW_LOW;
+	tx_coal->flow_level = HNS3_FLOW_LOW;
+
+	if (ae_dev->dev_specs.int_ql_max) {
+		tx_coal->int_ql = HNS3_INT_QL_DEFAULT_CFG;
+		rx_coal->int_ql = HNS3_INT_QL_DEFAULT_CFG;
+	}
+}
+
+static int hns3_nic_alloc_vector_data(struct hns3_nic_priv *priv)
+{
+	struct hnae3_handle *h = priv->ae_handle;
+	struct hns3_enet_tqp_vector *tqp_vector;
+	struct hnae3_vector_info *vector;
+	struct pci_dev *pdev = h->pdev;
+	u16 tqp_num = h->kinfo.num_tqps;
+	u16 vector_num;
+	int ret = 0;
+	u16 i;
+
+	/* RSS size, cpu online and vector_num should be the same */
+	/* Should consider 2p/4p later */
+	vector_num = min_t(u16, num_online_cpus(), tqp_num);
+
+	vector = devm_kcalloc(&pdev->dev, vector_num, sizeof(*vector),
+			      GFP_KERNEL);
+	if (!vector)
+		return -ENOMEM;
+
+	/* save the actual available vector number */
+	vector_num = h->ae_algo->ops->get_vector(h, vector_num, vector);
+
+	priv->vector_num = vector_num;
+	priv->tqp_vector = (struct hns3_enet_tqp_vector *)
+		devm_kcalloc(&pdev->dev, vector_num, sizeof(*priv->tqp_vector),
+			     GFP_KERNEL);
+	if (!priv->tqp_vector) {
+		ret = -ENOMEM;
+		goto out;
+	}
+
+	for (i = 0; i < priv->vector_num; i++) {
+		tqp_vector = &priv->tqp_vector[i];
+		tqp_vector->idx = i;
+		tqp_vector->mask_addr = vector[i].io_addr;
+		tqp_vector->vector_irq = vector[i].vector;
+		hns3_vector_coalesce_init(tqp_vector, priv);
+	}
+
+out:
+	devm_kfree(&pdev->dev, vector);
+	return ret;
+}
+
+static void hns3_clear_ring_group(struct hns3_enet_ring_group *group)
+{
+	group->ring = NULL;
+	group->count = 0;
+}
+
+static void hns3_nic_uninit_vector_data(struct hns3_nic_priv *priv)
+{
+	struct hnae3_ring_chain_node *vector_ring_chain;
+	struct hnae3_handle *h = priv->ae_handle;
+	struct hns3_enet_tqp_vector *tqp_vector;
+	int i;
+
+	for (i = 0; i < priv->vector_num; i++) {
+		tqp_vector = &priv->tqp_vector[i];
+
+		if (!tqp_vector->rx_group.ring && !tqp_vector->tx_group.ring)
+			continue;
+
+		/* Since the mapping can be overwritten, when fail to get the
+		 * chain between vector and ring, we should go on to deal with
+		 * the remaining options.
+		 */
+		vector_ring_chain = hns3_get_vector_ring_chain(tqp_vector);
+		if (!vector_ring_chain)
+			dev_warn(priv->dev, "failed to get ring chain\n");
+
+		h->ae_algo->ops->unmap_ring_from_vector(h,
+			tqp_vector->vector_irq, vector_ring_chain);
+
+		hns3_free_vector_ring_chain(tqp_vector, vector_ring_chain);
+
+		hns3_clear_ring_group(&tqp_vector->rx_group);
+		hns3_clear_ring_group(&tqp_vector->tx_group);
+		netif_napi_del(&priv->tqp_vector[i].napi);
+	}
+}
+
+static void hns3_nic_dealloc_vector_data(struct hns3_nic_priv *priv)
+{
+	struct hnae3_handle *h = priv->ae_handle;
+	struct pci_dev *pdev = h->pdev;
+	int i, ret;
+
+	for (i = 0; i < priv->vector_num; i++) {
+		struct hns3_enet_tqp_vector *tqp_vector;
+
+		tqp_vector = &priv->tqp_vector[i];
+		ret = h->ae_algo->ops->put_vector(h, tqp_vector->vector_irq);
+		if (ret)
+			return;
+	}
+
+	devm_kfree(&pdev->dev, priv->tqp_vector);
+}
+
+static void hns3_update_tx_spare_buf_config(struct hns3_nic_priv *priv)
+{
+#define HNS3_MIN_SPARE_BUF_SIZE (2 * 1024 * 1024)
+#define HNS3_MAX_PACKET_SIZE (64 * 1024)
+
+	struct iommu_domain *domain = iommu_get_domain_for_dev(priv->dev);
+	struct hnae3_ae_dev *ae_dev = hns3_get_ae_dev(priv->ae_handle);
+	struct hnae3_handle *handle = priv->ae_handle;
+
+	if (ae_dev->dev_version < HNAE3_DEVICE_VERSION_V3)
+		return;
+
+	if (!(domain && iommu_is_dma_domain(domain)))
+		return;
+
+	priv->min_tx_copybreak = HNS3_MAX_PACKET_SIZE;
+	priv->min_tx_spare_buf_size = HNS3_MIN_SPARE_BUF_SIZE;
+
+	if (priv->tx_copybreak < priv->min_tx_copybreak)
+		priv->tx_copybreak = priv->min_tx_copybreak;
+	if (handle->kinfo.tx_spare_buf_size < priv->min_tx_spare_buf_size)
+		handle->kinfo.tx_spare_buf_size = priv->min_tx_spare_buf_size;
+}
+
+static void hns3_ring_get_cfg(struct hnae3_queue *q, struct hns3_nic_priv *priv,
+			      unsigned int ring_type)
+{
+	int queue_num = priv->ae_handle->kinfo.num_tqps;
+	struct hns3_enet_ring *ring;
+	int desc_num;
+
+	if (ring_type == HNAE3_RING_TYPE_TX) {
+		ring = &priv->ring[q->tqp_index];
+		desc_num = priv->ae_handle->kinfo.num_tx_desc;
+		ring->queue_index = q->tqp_index;
+		ring->tx_copybreak = priv->tx_copybreak;
+		ring->last_to_use = 0;
+	} else {
+		ring = &priv->ring[q->tqp_index + queue_num];
+		desc_num = priv->ae_handle->kinfo.num_rx_desc;
+		ring->queue_index = q->tqp_index;
+		ring->rx_copybreak = priv->rx_copybreak;
+	}
+
+	hnae3_set_bit(ring->flag, HNAE3_RING_TYPE_B, ring_type);
+
+	ring->tqp = q;
+	ring->desc = NULL;
+	ring->desc_cb = NULL;
+	ring->dev = priv->dev;
+	ring->desc_dma_addr = 0;
+	ring->buf_size = q->buf_size;
+	ring->desc_num = desc_num;
+	ring->next_to_use = 0;
+	ring->next_to_clean = 0;
+}
+
+static void hns3_queue_to_ring(struct hnae3_queue *tqp,
+			       struct hns3_nic_priv *priv)
+{
+	hns3_ring_get_cfg(tqp, priv, HNAE3_RING_TYPE_TX);
+	hns3_ring_get_cfg(tqp, priv, HNAE3_RING_TYPE_RX);
+}
+
+static int hns3_get_ring_config(struct hns3_nic_priv *priv)
+{
+	struct hnae3_handle *h = priv->ae_handle;
+	struct pci_dev *pdev = h->pdev;
+	int i;
+
+	priv->ring = devm_kzalloc(&pdev->dev,
+				  array3_size(h->kinfo.num_tqps,
+					      sizeof(*priv->ring), 2),
+				  GFP_KERNEL);
+	if (!priv->ring)
+		return -ENOMEM;
+
+	for (i = 0; i < h->kinfo.num_tqps; i++)
+		hns3_queue_to_ring(h->kinfo.tqp[i], priv);
+
+	return 0;
+}
+
+static void hns3_put_ring_config(struct hns3_nic_priv *priv)
+{
+	if (!priv->ring)
+		return;
+
+	devm_kfree(priv->dev, priv->ring);
+	priv->ring = NULL;
+}
+
+static void hns3_alloc_page_pool(struct hns3_enet_ring *ring)
+{
+	struct page_pool_params pp_params = {
+		.flags = PP_FLAG_DMA_MAP | PP_FLAG_DMA_SYNC_DEV,
+		.order = hns3_page_order(ring),
+		.pool_size = ring->desc_num * hns3_buf_size(ring) /
+				(PAGE_SIZE << hns3_page_order(ring)),
+		.nid = dev_to_node(ring_to_dev(ring)),
+		.dev = ring_to_dev(ring),
+		.dma_dir = DMA_FROM_DEVICE,
+		.offset = 0,
+		.max_len = PAGE_SIZE << hns3_page_order(ring),
+	};
+
+	ring->page_pool = page_pool_create(&pp_params);
+	if (IS_ERR(ring->page_pool)) {
+		dev_warn(ring_to_dev(ring), "page pool creation failed: %ld\n",
+			 PTR_ERR(ring->page_pool));
+		ring->page_pool = NULL;
+	}
+}
+
+static int hns3_alloc_ring_memory(struct hns3_enet_ring *ring)
+{
+	int ret;
+
+	if (ring->desc_num <= 0 || ring->buf_size <= 0)
+		return -EINVAL;
+
+	ring->desc_cb = devm_kcalloc(ring_to_dev(ring), ring->desc_num,
+				     sizeof(ring->desc_cb[0]), GFP_KERNEL);
+	if (!ring->desc_cb) {
+		ret = -ENOMEM;
+		goto out;
+	}
+
+	ret = hns3_alloc_desc(ring);
+	if (ret)
+		goto out_with_desc_cb;
+
+	if (!HNAE3_IS_TX_RING(ring)) {
+		if (page_pool_enabled)
+			hns3_alloc_page_pool(ring);
+
+		ret = hns3_alloc_ring_buffers(ring);
+		if (ret)
+			goto out_with_desc;
+	} else {
+		hns3_init_tx_spare_buffer(ring);
+	}
+
+	return 0;
+
+out_with_desc:
+	hns3_free_desc(ring);
+out_with_desc_cb:
+	devm_kfree(ring_to_dev(ring), ring->desc_cb);
+	ring->desc_cb = NULL;
+out:
+	return ret;
+}
+
+void hns3_fini_ring(struct hns3_enet_ring *ring)
+{
+	hns3_free_desc(ring);
+	devm_kfree(ring_to_dev(ring), ring->desc_cb);
+	ring->desc_cb = NULL;
+	ring->next_to_clean = 0;
+	ring->next_to_use = 0;
+	ring->last_to_use = 0;
+	ring->pending_buf = 0;
+	if (!HNAE3_IS_TX_RING(ring) && ring->skb) {
+		dev_kfree_skb_any(ring->skb);
+		ring->skb = NULL;
+	} else if (HNAE3_IS_TX_RING(ring) && ring->tx_spare) {
+		struct hns3_tx_spare *tx_spare = ring->tx_spare;
+
+		dma_unmap_page(ring_to_dev(ring), tx_spare->dma, tx_spare->len,
+			       DMA_TO_DEVICE);
+		free_pages((unsigned long)tx_spare->buf,
+			   get_order(tx_spare->len));
+		devm_kfree(ring_to_dev(ring), tx_spare);
+		ring->tx_spare = NULL;
+	}
+
+	if (!HNAE3_IS_TX_RING(ring) && ring->page_pool) {
+		page_pool_destroy(ring->page_pool);
+		ring->page_pool = NULL;
+	}
+}
+
+static int hns3_buf_size2type(u32 buf_size)
+{
+	int bd_size_type;
+
+	switch (buf_size) {
+	case 512:
+		bd_size_type = HNS3_BD_SIZE_512_TYPE;
+		break;
+	case 1024:
+		bd_size_type = HNS3_BD_SIZE_1024_TYPE;
+		break;
+	case 2048:
+		bd_size_type = HNS3_BD_SIZE_2048_TYPE;
+		break;
+	case 4096:
+		bd_size_type = HNS3_BD_SIZE_4096_TYPE;
+		break;
+	default:
+		bd_size_type = HNS3_BD_SIZE_2048_TYPE;
+	}
+
+	return bd_size_type;
+}
+
+static void hns3_init_ring_hw(struct hns3_enet_ring *ring)
+{
+	dma_addr_t dma = ring->desc_dma_addr;
+	struct hnae3_queue *q = ring->tqp;
+
+	if (!HNAE3_IS_TX_RING(ring)) {
+		hns3_write_dev(q, HNS3_RING_RX_RING_BASEADDR_L_REG, (u32)dma);
+		hns3_write_dev(q, HNS3_RING_RX_RING_BASEADDR_H_REG,
+			       (u32)((dma >> 31) >> 1));
+
+		hns3_write_dev(q, HNS3_RING_RX_RING_BD_LEN_REG,
+			       hns3_buf_size2type(ring->buf_size));
+		hns3_write_dev(q, HNS3_RING_RX_RING_BD_NUM_REG,
+			       ring->desc_num / 8 - 1);
+	} else {
+		hns3_write_dev(q, HNS3_RING_TX_RING_BASEADDR_L_REG,
+			       (u32)dma);
+		hns3_write_dev(q, HNS3_RING_TX_RING_BASEADDR_H_REG,
+			       (u32)((dma >> 31) >> 1));
+
+		hns3_write_dev(q, HNS3_RING_TX_RING_BD_NUM_REG,
+			       ring->desc_num / 8 - 1);
+	}
+}
+
+static void hns3_init_tx_ring_tc(struct hns3_nic_priv *priv)
+{
+	struct hnae3_knic_private_info *kinfo = &priv->ae_handle->kinfo;
+	struct hnae3_tc_info *tc_info = &kinfo->tc_info;
+	int i;
+
+	for (i = 0; i < tc_info->num_tc; i++) {
+		int j;
+
+		for (j = 0; j < tc_info->tqp_count[i]; j++) {
+			struct hnae3_queue *q;
+
+			q = priv->ring[tc_info->tqp_offset[i] + j].tqp;
+			hns3_write_dev(q, HNS3_RING_TX_RING_TC_REG, i);
+		}
+	}
+}
+
+int hns3_init_all_ring(struct hns3_nic_priv *priv)
+{
+	struct hnae3_handle *h = priv->ae_handle;
+	int ring_num = h->kinfo.num_tqps * 2;
+	int i, j;
+	int ret;
+
+	hns3_update_tx_spare_buf_config(priv);
+	for (i = 0; i < ring_num; i++) {
+		ret = hns3_alloc_ring_memory(&priv->ring[i]);
+		if (ret) {
+			dev_err(priv->dev,
+				"Alloc ring memory fail! ret=%d\n", ret);
+			goto out_when_alloc_ring_memory;
+		}
+
+		u64_stats_init(&priv->ring[i].syncp);
+		cond_resched();
+	}
+
+	return 0;
+
+out_when_alloc_ring_memory:
+	for (j = i - 1; j >= 0; j--)
+		hns3_fini_ring(&priv->ring[j]);
+
+	return -ENOMEM;
+}
+
+static void hns3_uninit_all_ring(struct hns3_nic_priv *priv)
+{
+	struct hnae3_handle *h = priv->ae_handle;
+	int i;
+
+	for (i = 0; i < h->kinfo.num_tqps; i++) {
+		hns3_fini_ring(&priv->ring[i]);
+		hns3_fini_ring(&priv->ring[i + h->kinfo.num_tqps]);
+	}
+}
+
+/* Set mac addr if it is configured. or leave it to the AE driver */
+static int hns3_init_mac_addr(struct net_device *netdev)
+{
+	struct hns3_nic_priv *priv = netdev_priv(netdev);
+	char format_mac_addr[HNAE3_FORMAT_MAC_ADDR_LEN];
+	struct hnae3_handle *h = priv->ae_handle;
+	u8 mac_addr_temp[ETH_ALEN] = {0};
+	int ret = 0;
+
+	if (h->ae_algo->ops->get_mac_addr)
+		h->ae_algo->ops->get_mac_addr(h, mac_addr_temp);
+
+	/* Check if the MAC address is valid, if not get a random one */
+	if (!is_valid_ether_addr(mac_addr_temp)) {
+		eth_hw_addr_random(netdev);
+		hnae3_format_mac_addr(format_mac_addr, netdev->dev_addr);
+		dev_warn(priv->dev, "using random MAC address %s\n",
+			 format_mac_addr);
+	} else if (!ether_addr_equal(netdev->dev_addr, mac_addr_temp)) {
+		eth_hw_addr_set(netdev, mac_addr_temp);
+		ether_addr_copy(netdev->perm_addr, mac_addr_temp);
+	} else {
+		return 0;
+	}
+
+	if (h->ae_algo->ops->set_mac_addr)
+		ret = h->ae_algo->ops->set_mac_addr(h, netdev->dev_addr, true);
+
+	return ret;
+}
+
+static int hns3_init_phy(struct net_device *netdev)
+{
+	struct hnae3_handle *h = hns3_get_handle(netdev);
+	int ret = 0;
+
+	if (h->ae_algo->ops->mac_connect_phy)
+		ret = h->ae_algo->ops->mac_connect_phy(h);
+
+	return ret;
+}
+
+static void hns3_uninit_phy(struct net_device *netdev)
+{
+	struct hnae3_handle *h = hns3_get_handle(netdev);
+
+	if (h->ae_algo->ops->mac_disconnect_phy)
+		h->ae_algo->ops->mac_disconnect_phy(h);
+}
+
+static int hns3_client_start(struct hnae3_handle *handle)
+{
+	if (!handle->ae_algo->ops->client_start)
+		return 0;
+
+	return handle->ae_algo->ops->client_start(handle);
+}
+
+static void hns3_client_stop(struct hnae3_handle *handle)
+{
+	if (!handle->ae_algo->ops->client_stop)
+		return;
+
+	handle->ae_algo->ops->client_stop(handle);
+}
+
+static void hns3_info_show(struct hns3_nic_priv *priv)
+{
+	struct hnae3_knic_private_info *kinfo = &priv->ae_handle->kinfo;
+	char format_mac_addr[HNAE3_FORMAT_MAC_ADDR_LEN];
+
+	hnae3_format_mac_addr(format_mac_addr, priv->netdev->dev_addr);
+	dev_info(priv->dev, "MAC address: %s\n", format_mac_addr);
+	dev_info(priv->dev, "Task queue pairs numbers: %u\n", kinfo->num_tqps);
+	dev_info(priv->dev, "RSS size: %u\n", kinfo->rss_size);
+	dev_info(priv->dev, "Allocated RSS size: %u\n", kinfo->req_rss_size);
+	dev_info(priv->dev, "RX buffer length: %u\n", kinfo->rx_buf_len);
+	dev_info(priv->dev, "Desc num per TX queue: %u\n", kinfo->num_tx_desc);
+	dev_info(priv->dev, "Desc num per RX queue: %u\n", kinfo->num_rx_desc);
+	dev_info(priv->dev, "Total number of enabled TCs: %u\n",
+		 kinfo->tc_info.num_tc);
+	dev_info(priv->dev, "Max mtu size: %u\n", priv->netdev->max_mtu);
+}
+
+static void hns3_set_cq_period_mode(struct hns3_nic_priv *priv,
+				    enum dim_cq_period_mode mode, bool is_tx)
+{
+	struct hnae3_ae_dev *ae_dev = hns3_get_ae_dev(priv->ae_handle);
+	struct hnae3_handle *handle = priv->ae_handle;
+	int i;
+
+	if (is_tx) {
+		priv->tx_cqe_mode = mode;
+
+		for (i = 0; i < priv->vector_num; i++)
+			priv->tqp_vector[i].tx_group.dim.mode = mode;
+	} else {
+		priv->rx_cqe_mode = mode;
+
+		for (i = 0; i < priv->vector_num; i++)
+			priv->tqp_vector[i].rx_group.dim.mode = mode;
+	}
+
+	if (hnae3_ae_dev_cq_supported(ae_dev)) {
+		u32 new_mode;
+		u64 reg;
+
+		new_mode = (mode == DIM_CQ_PERIOD_MODE_START_FROM_CQE) ?
+			HNS3_CQ_MODE_CQE : HNS3_CQ_MODE_EQE;
+		reg = is_tx ? HNS3_GL1_CQ_MODE_REG : HNS3_GL0_CQ_MODE_REG;
+
+		writel(new_mode, handle->kinfo.io_base + reg);
+	}
+}
+
+void hns3_cq_period_mode_init(struct hns3_nic_priv *priv,
+			      enum dim_cq_period_mode tx_mode,
+			      enum dim_cq_period_mode rx_mode)
+{
+	hns3_set_cq_period_mode(priv, tx_mode, true);
+	hns3_set_cq_period_mode(priv, rx_mode, false);
+}
+
+static void hns3_state_init(struct hnae3_handle *handle)
+{
+	struct hnae3_ae_dev *ae_dev = hns3_get_ae_dev(handle);
+	struct net_device *netdev = handle->kinfo.netdev;
+	struct hns3_nic_priv *priv = netdev_priv(netdev);
+
+	set_bit(HNS3_NIC_STATE_INITED, &priv->state);
+
+	if (test_bit(HNAE3_DEV_SUPPORT_TX_PUSH_B, ae_dev->caps))
+		set_bit(HNS3_NIC_STATE_TX_PUSH_ENABLE, &priv->state);
+
+	if (ae_dev->dev_version >= HNAE3_DEVICE_VERSION_V3)
+		set_bit(HNAE3_PFLAG_LIMIT_PROMISC, &handle->supported_pflags);
+
+	if (test_bit(HNAE3_DEV_SUPPORT_HW_TX_CSUM_B, ae_dev->caps))
+		set_bit(HNS3_NIC_STATE_HW_TX_CSUM_ENABLE, &priv->state);
+
+	if (hnae3_ae_dev_rxd_adv_layout_supported(ae_dev))
+		set_bit(HNS3_NIC_STATE_RXD_ADV_LAYOUT_ENABLE, &priv->state);
+}
+
+static void hns3_state_uninit(struct hnae3_handle *handle)
+{
+	struct hns3_nic_priv *priv  = handle->priv;
+
+	clear_bit(HNS3_NIC_STATE_INITED, &priv->state);
+}
+
+static int hns3_client_init(struct hnae3_handle *handle)
+{
+	struct pci_dev *pdev = handle->pdev;
+	struct hnae3_ae_dev *ae_dev = pci_get_drvdata(pdev);
+	u16 alloc_tqps, max_rss_size;
+	struct hns3_nic_priv *priv;
+	struct net_device *netdev;
+	int ret;
+
+	ae_dev->handle = handle;
+
+	handle->ae_algo->ops->get_tqps_and_rss_info(handle, &alloc_tqps,
+						    &max_rss_size);
+	netdev = alloc_etherdev_mq(sizeof(struct hns3_nic_priv), alloc_tqps);
+	if (!netdev)
+		return -ENOMEM;
+
+	priv = netdev_priv(netdev);
+	priv->dev = &pdev->dev;
+	priv->netdev = netdev;
+	priv->ae_handle = handle;
+	priv->tx_timeout_count = 0;
+	priv->max_non_tso_bd_num = ae_dev->dev_specs.max_non_tso_bd_num;
+	priv->min_tx_copybreak = 0;
+	priv->min_tx_spare_buf_size = 0;
+	set_bit(HNS3_NIC_STATE_DOWN, &priv->state);
+
+	handle->msg_enable = netif_msg_init(debug, DEFAULT_MSG_LEVEL);
+
+	handle->kinfo.netdev = netdev;
+	handle->priv = (void *)priv;
+
+	hns3_init_mac_addr(netdev);
+
+	hns3_set_default_feature(netdev);
+
+	netdev->watchdog_timeo = HNS3_TX_TIMEOUT;
+	netdev->priv_flags |= IFF_UNICAST_FLT;
+	netdev->netdev_ops = &hns3_nic_netdev_ops;
+	SET_NETDEV_DEV(netdev, &pdev->dev);
+	hns3_ethtool_set_ops(netdev);
+
+	/* Carrier off reporting is important to ethtool even BEFORE open */
+	netif_carrier_off(netdev);
+
+	ret = hns3_get_ring_config(priv);
+	if (ret) {
+		ret = -ENOMEM;
+		goto out_get_ring_cfg;
+	}
+
+	hns3_nic_init_coal_cfg(priv);
+
+	ret = hns3_nic_alloc_vector_data(priv);
+	if (ret) {
+		ret = -ENOMEM;
+		goto out_alloc_vector_data;
+	}
+
+	ret = hns3_nic_init_vector_data(priv);
+	if (ret) {
+		ret = -ENOMEM;
+		goto out_init_vector_data;
+	}
+
+	ret = hns3_init_all_ring(priv);
+	if (ret) {
+		ret = -ENOMEM;
+		goto out_init_ring;
+	}
+
+	hns3_cq_period_mode_init(priv, DIM_CQ_PERIOD_MODE_START_FROM_EQE,
+				 DIM_CQ_PERIOD_MODE_START_FROM_EQE);
+
+	ret = hns3_init_phy(netdev);
+	if (ret)
+		goto out_init_phy;
+
+	/* the device can work without cpu rmap, only aRFS needs it */
+	ret = hns3_set_rx_cpu_rmap(netdev);
+	if (ret)
+		dev_warn(priv->dev, "set rx cpu rmap fail, ret=%d\n", ret);
+
+	ret = hns3_nic_init_irq(priv);
+	if (ret) {
+		dev_err(priv->dev, "init irq failed! ret=%d\n", ret);
+		hns3_free_rx_cpu_rmap(netdev);
+		goto out_init_irq_fail;
+	}
+
+	ret = hns3_client_start(handle);
+	if (ret) {
+		dev_err(priv->dev, "hns3_client_start fail! ret=%d\n", ret);
+		goto out_client_start;
+	}
+
+	hns3_dcbnl_setup(handle);
+
+	ret = hns3_dbg_init(handle);
+	if (ret) {
+		dev_err(priv->dev, "failed to init debugfs, ret = %d\n",
+			ret);
+		goto out_client_start;
+	}
+
+	netdev->max_mtu = HNS3_MAX_MTU(ae_dev->dev_specs.max_frm_size);
+
+	hns3_state_init(handle);
+
+	ret = register_netdev(netdev);
+	if (ret) {
+		dev_err(priv->dev, "probe register netdev fail!\n");
+		goto out_reg_netdev_fail;
+	}
+
+	if (netif_msg_drv(handle))
+		hns3_info_show(priv);
+
+	return ret;
+
+out_reg_netdev_fail:
+	hns3_state_uninit(handle);
+	hns3_dbg_uninit(handle);
+	hns3_client_stop(handle);
+out_client_start:
+	hns3_free_rx_cpu_rmap(netdev);
+	hns3_nic_uninit_irq(priv);
+out_init_irq_fail:
+	hns3_uninit_phy(netdev);
+out_init_phy:
+	hns3_uninit_all_ring(priv);
+out_init_ring:
+	hns3_nic_uninit_vector_data(priv);
+out_init_vector_data:
+	hns3_nic_dealloc_vector_data(priv);
+out_alloc_vector_data:
+	priv->ring = NULL;
+out_get_ring_cfg:
+	priv->ae_handle = NULL;
+	free_netdev(netdev);
+	return ret;
+}
+
+static void hns3_client_uninit(struct hnae3_handle *handle, bool reset)
+{
+	struct net_device *netdev = handle->kinfo.netdev;
+	struct hns3_nic_priv *priv = netdev_priv(netdev);
+
+	if (netdev->reg_state != NETREG_UNINITIALIZED)
+		unregister_netdev(netdev);
+
+	hns3_client_stop(handle);
+
+	hns3_uninit_phy(netdev);
+
+	if (!test_and_clear_bit(HNS3_NIC_STATE_INITED, &priv->state)) {
+		netdev_warn(netdev, "already uninitialized\n");
+		goto out_netdev_free;
+	}
+
+	hns3_free_rx_cpu_rmap(netdev);
+
+	hns3_nic_uninit_irq(priv);
+
+	hns3_clear_all_ring(handle, true);
+
+	hns3_nic_uninit_vector_data(priv);
+
+	hns3_nic_dealloc_vector_data(priv);
+
+	hns3_uninit_all_ring(priv);
+
+	hns3_put_ring_config(priv);
+
+out_netdev_free:
+	hns3_dbg_uninit(handle);
+	free_netdev(netdev);
+}
+
+static void hns3_link_status_change(struct hnae3_handle *handle, bool linkup)
+{
+	struct net_device *netdev = handle->kinfo.netdev;
+
+	if (!netdev)
+		return;
+
+	if (linkup) {
+		netif_tx_wake_all_queues(netdev);
+		netif_carrier_on(netdev);
+		if (netif_msg_link(handle))
+			netdev_info(netdev, "link up\n");
+	} else {
+		netif_carrier_off(netdev);
+		netif_tx_stop_all_queues(netdev);
+		if (netif_msg_link(handle))
+			netdev_info(netdev, "link down\n");
+	}
+}
+
+static void hns3_clear_tx_ring(struct hns3_enet_ring *ring)
+{
+	while (ring->next_to_clean != ring->next_to_use) {
+		ring->desc[ring->next_to_clean].tx.bdtp_fe_sc_vld_ra_ri = 0;
+		hns3_free_buffer_detach(ring, ring->next_to_clean, 0);
+		ring_ptr_move_fw(ring, next_to_clean);
+	}
+
+	ring->pending_buf = 0;
+}
+
+static int hns3_clear_rx_ring(struct hns3_enet_ring *ring)
+{
+	struct hns3_desc_cb res_cbs;
+	int ret;
+
+	while (ring->next_to_use != ring->next_to_clean) {
+		/* When a buffer is not reused, it's memory has been
+		 * freed in hns3_handle_rx_bd or will be freed by
+		 * stack, so we need to replace the buffer here.
+		 */
+		if (!ring->desc_cb[ring->next_to_use].reuse_flag) {
+			ret = hns3_alloc_and_map_buffer(ring, &res_cbs);
+			if (ret) {
+				hns3_ring_stats_update(ring, sw_err_cnt);
+				/* if alloc new buffer fail, exit directly
+				 * and reclear in up flow.
+				 */
+				netdev_warn(ring_to_netdev(ring),
+					    "reserve buffer map failed, ret = %d\n",
+					    ret);
+				return ret;
+			}
+			hns3_replace_buffer(ring, ring->next_to_use, &res_cbs);
+		}
+		ring_ptr_move_fw(ring, next_to_use);
+	}
+
+	/* Free the pending skb in rx ring */
+	if (ring->skb) {
+		dev_kfree_skb_any(ring->skb);
+		ring->skb = NULL;
+		ring->pending_buf = 0;
+	}
+
+	return 0;
+}
+
+static void hns3_force_clear_rx_ring(struct hns3_enet_ring *ring)
+{
+	while (ring->next_to_use != ring->next_to_clean) {
+		/* When a buffer is not reused, it's memory has been
+		 * freed in hns3_handle_rx_bd or will be freed by
+		 * stack, so only need to unmap the buffer here.
+		 */
+		if (!ring->desc_cb[ring->next_to_use].reuse_flag) {
+			hns3_unmap_buffer(ring,
+					  &ring->desc_cb[ring->next_to_use]);
+			ring->desc_cb[ring->next_to_use].dma = 0;
+		}
+
+		ring_ptr_move_fw(ring, next_to_use);
+	}
+}
+
+static void hns3_clear_all_ring(struct hnae3_handle *h, bool force)
+{
+	struct net_device *ndev = h->kinfo.netdev;
+	struct hns3_nic_priv *priv = netdev_priv(ndev);
+	u32 i;
+
+	for (i = 0; i < h->kinfo.num_tqps; i++) {
+		struct hns3_enet_ring *ring;
+
+		ring = &priv->ring[i];
+		hns3_clear_tx_ring(ring);
+
+		ring = &priv->ring[i + h->kinfo.num_tqps];
+		/* Continue to clear other rings even if clearing some
+		 * rings failed.
+		 */
+		if (force)
+			hns3_force_clear_rx_ring(ring);
+		else
+			hns3_clear_rx_ring(ring);
+	}
+}
+
+int hns3_nic_reset_all_ring(struct hnae3_handle *h)
+{
+	struct net_device *ndev = h->kinfo.netdev;
+	struct hns3_nic_priv *priv = netdev_priv(ndev);
+	struct hns3_enet_ring *rx_ring;
+	int i, j;
+	int ret;
+
+	ret = h->ae_algo->ops->reset_queue(h);
+	if (ret)
+		return ret;
+
+	for (i = 0; i < h->kinfo.num_tqps; i++) {
+		hns3_init_ring_hw(&priv->ring[i]);
+
+		/* We need to clear tx ring here because self test will
+		 * use the ring and will not run down before up
+		 */
+		hns3_clear_tx_ring(&priv->ring[i]);
+		priv->ring[i].next_to_clean = 0;
+		priv->ring[i].next_to_use = 0;
+		priv->ring[i].last_to_use = 0;
+
+		rx_ring = &priv->ring[i + h->kinfo.num_tqps];
+		hns3_init_ring_hw(rx_ring);
+		ret = hns3_clear_rx_ring(rx_ring);
+		if (ret)
+			return ret;
+
+		/* We can not know the hardware head and tail when this
+		 * function is called in reset flow, so we reuse all desc.
+		 */
+		for (j = 0; j < rx_ring->desc_num; j++)
+			hns3_reuse_buffer(rx_ring, j);
+
+		rx_ring->next_to_clean = 0;
+		rx_ring->next_to_use = 0;
+	}
+
+	hns3_init_tx_ring_tc(priv);
+
+	return 0;
+}
+
+static int hns3_reset_notify_down_enet(struct hnae3_handle *handle)
+{
+	struct hnae3_knic_private_info *kinfo = &handle->kinfo;
+	struct net_device *ndev = kinfo->netdev;
+	struct hns3_nic_priv *priv = netdev_priv(ndev);
+
+	if (test_and_set_bit(HNS3_NIC_STATE_RESETTING, &priv->state))
+		return 0;
+
+	if (!netif_running(ndev))
+		return 0;
+
+	return hns3_nic_net_stop(ndev);
+}
+
+static int hns3_reset_notify_up_enet(struct hnae3_handle *handle)
+{
+	struct hnae3_knic_private_info *kinfo = &handle->kinfo;
+	struct hns3_nic_priv *priv = netdev_priv(kinfo->netdev);
+	int ret = 0;
+
+	if (!test_bit(HNS3_NIC_STATE_INITED, &priv->state)) {
+		netdev_err(kinfo->netdev, "device is not initialized yet\n");
+		return -EFAULT;
+	}
+
+	clear_bit(HNS3_NIC_STATE_RESETTING, &priv->state);
+
+	if (netif_running(kinfo->netdev)) {
+		ret = hns3_nic_net_open(kinfo->netdev);
+		if (ret) {
+			set_bit(HNS3_NIC_STATE_RESETTING, &priv->state);
+			netdev_err(kinfo->netdev,
+				   "net up fail, ret=%d!\n", ret);
+			return ret;
+		}
+	}
+
+	return ret;
+}
+
+static int hns3_reset_notify_init_enet(struct hnae3_handle *handle)
+{
+	struct net_device *netdev = handle->kinfo.netdev;
+	struct hns3_nic_priv *priv = netdev_priv(netdev);
+	int ret;
+
+	/* Carrier off reporting is important to ethtool even BEFORE open */
+	netif_carrier_off(netdev);
+
+	ret = hns3_get_ring_config(priv);
+	if (ret)
+		return ret;
+
+	ret = hns3_nic_alloc_vector_data(priv);
+	if (ret)
+		goto err_put_ring;
+
+	ret = hns3_nic_init_vector_data(priv);
+	if (ret)
+		goto err_dealloc_vector;
+
+	ret = hns3_init_all_ring(priv);
+	if (ret)
+		goto err_uninit_vector;
+
+	hns3_cq_period_mode_init(priv, priv->tx_cqe_mode, priv->rx_cqe_mode);
+
+	/* the device can work without cpu rmap, only aRFS needs it */
+	ret = hns3_set_rx_cpu_rmap(netdev);
+	if (ret)
+		dev_warn(priv->dev, "set rx cpu rmap fail, ret=%d\n", ret);
+
+	ret = hns3_nic_init_irq(priv);
+	if (ret) {
+		dev_err(priv->dev, "init irq failed! ret=%d\n", ret);
+		hns3_free_rx_cpu_rmap(netdev);
+		goto err_init_irq_fail;
+	}
+
+	if (!hns3_is_phys_func(handle->pdev))
+		hns3_init_mac_addr(netdev);
+
+	ret = hns3_client_start(handle);
+	if (ret) {
+		dev_err(priv->dev, "hns3_client_start fail! ret=%d\n", ret);
+		goto err_client_start_fail;
+	}
+
+	set_bit(HNS3_NIC_STATE_INITED, &priv->state);
+
+	return ret;
+
+err_client_start_fail:
+	hns3_free_rx_cpu_rmap(netdev);
+	hns3_nic_uninit_irq(priv);
+err_init_irq_fail:
+	hns3_uninit_all_ring(priv);
+err_uninit_vector:
+	hns3_nic_uninit_vector_data(priv);
+err_dealloc_vector:
+	hns3_nic_dealloc_vector_data(priv);
+err_put_ring:
+	hns3_put_ring_config(priv);
+
+	return ret;
+}
+
+static int hns3_reset_notify_uninit_enet(struct hnae3_handle *handle)
+{
+	struct net_device *netdev = handle->kinfo.netdev;
+	struct hns3_nic_priv *priv = netdev_priv(netdev);
+
+	if (!test_bit(HNS3_NIC_STATE_DOWN, &priv->state))
+		hns3_nic_net_stop(netdev);
+
+	if (!test_and_clear_bit(HNS3_NIC_STATE_INITED, &priv->state)) {
+		netdev_warn(netdev, "already uninitialized\n");
+		return 0;
+	}
+
+	hns3_free_rx_cpu_rmap(netdev);
+	hns3_nic_uninit_irq(priv);
+	hns3_clear_all_ring(handle, true);
+	hns3_reset_tx_queue(priv->ae_handle);
+
+	hns3_nic_uninit_vector_data(priv);
+
+	hns3_nic_dealloc_vector_data(priv);
+
+	hns3_uninit_all_ring(priv);
+
+	hns3_put_ring_config(priv);
+
+	return 0;
+}
+
+int hns3_reset_notify(struct hnae3_handle *handle,
+		      enum hnae3_reset_notify_type type)
+{
+	int ret = 0;
+
+	switch (type) {
+	case HNAE3_UP_CLIENT:
+		ret = hns3_reset_notify_up_enet(handle);
+		break;
+	case HNAE3_DOWN_CLIENT:
+		ret = hns3_reset_notify_down_enet(handle);
+		break;
+	case HNAE3_INIT_CLIENT:
+		ret = hns3_reset_notify_init_enet(handle);
+		break;
+	case HNAE3_UNINIT_CLIENT:
+		ret = hns3_reset_notify_uninit_enet(handle);
+		break;
+	default:
+		break;
+	}
+
+	return ret;
+}
+
+static int hns3_change_channels(struct hnae3_handle *handle, u32 new_tqp_num,
+				bool rxfh_configured)
+{
+	int ret;
+
+	ret = handle->ae_algo->ops->set_channels(handle, new_tqp_num,
+						 rxfh_configured);
+	if (ret) {
+		dev_err(&handle->pdev->dev,
+			"Change tqp num(%u) fail.\n", new_tqp_num);
+		return ret;
+	}
+
+	ret = hns3_reset_notify(handle, HNAE3_INIT_CLIENT);
+	if (ret)
+		return ret;
+
+	ret =  hns3_reset_notify(handle, HNAE3_UP_CLIENT);
+	if (ret)
+		hns3_reset_notify(handle, HNAE3_UNINIT_CLIENT);
+
+	return ret;
+}
+
+int hns3_set_channels(struct net_device *netdev,
+		      struct ethtool_channels *ch)
+{
+	struct hnae3_handle *h = hns3_get_handle(netdev);
+	struct hnae3_knic_private_info *kinfo = &h->kinfo;
+	bool rxfh_configured = netif_is_rxfh_configured(netdev);
+	u32 new_tqp_num = ch->combined_count;
+	u16 org_tqp_num;
+	int ret;
+
+	if (hns3_nic_resetting(netdev))
+		return -EBUSY;
+
+	if (ch->rx_count || ch->tx_count)
+		return -EINVAL;
+
+	if (kinfo->tc_info.mqprio_active) {
+		dev_err(&netdev->dev,
+			"it's not allowed to set channels via ethtool when MQPRIO mode is on\n");
+		return -EINVAL;
+	}
+
+	if (new_tqp_num > hns3_get_max_available_channels(h) ||
+	    new_tqp_num < 1) {
+		dev_err(&netdev->dev,
+			"Change tqps fail, the tqp range is from 1 to %u",
+			hns3_get_max_available_channels(h));
+		return -EINVAL;
+	}
+
+	if (kinfo->rss_size == new_tqp_num)
+		return 0;
+
+	netif_dbg(h, drv, netdev,
+		  "set channels: tqp_num=%u, rxfh=%d\n",
+		  new_tqp_num, rxfh_configured);
+
+	ret = hns3_reset_notify(h, HNAE3_DOWN_CLIENT);
+	if (ret)
+		return ret;
+
+	ret = hns3_reset_notify(h, HNAE3_UNINIT_CLIENT);
+	if (ret)
+		return ret;
+
+	org_tqp_num = h->kinfo.num_tqps;
+	ret = hns3_change_channels(h, new_tqp_num, rxfh_configured);
+	if (ret) {
+		int ret1;
+
+		netdev_warn(netdev,
+			    "Change channels fail, revert to old value\n");
+		ret1 = hns3_change_channels(h, org_tqp_num, rxfh_configured);
+		if (ret1) {
+			netdev_err(netdev,
+				   "revert to old channel fail\n");
+			return ret1;
+		}
+
+		return ret;
+	}
+
+	return 0;
+}
+
+void hns3_external_lb_prepare(struct net_device *ndev, bool if_running)
+{
+	struct hns3_nic_priv *priv = netdev_priv(ndev);
+
+	if (!if_running)
+		return;
+
+	if (test_and_set_bit(HNS3_NIC_STATE_DOWN, &priv->state))
+		return;
+
+	netif_carrier_off(ndev);
+	netif_tx_disable(ndev);
+
+	hns3_disable_irqs_and_tqps(ndev);
+
+	/* delay ring buffer clearing to hns3_reset_notify_uninit_enet
+	 * during reset process, because driver may not be able
+	 * to disable the ring through firmware when downing the netdev.
+	 */
+	if (!hns3_nic_resetting(ndev))
+		hns3_nic_reset_all_ring(priv->ae_handle);
+
+	hns3_reset_tx_queue(priv->ae_handle);
+}
+
+void hns3_external_lb_restore(struct net_device *ndev, bool if_running)
+{
+	struct hns3_nic_priv *priv = netdev_priv(ndev);
+	struct hnae3_handle *h = priv->ae_handle;
+
+	if (!if_running)
+		return;
+
+	if (hns3_nic_resetting(ndev))
+		return;
+
+	if (!test_bit(HNS3_NIC_STATE_DOWN, &priv->state))
+		return;
+
+	if (hns3_nic_reset_all_ring(priv->ae_handle))
+		return;
+
+	clear_bit(HNS3_NIC_STATE_DOWN, &priv->state);
+
+	hns3_enable_irqs_and_tqps(ndev);
+
+	netif_tx_wake_all_queues(ndev);
+
+	if (h->ae_algo->ops->get_status(h))
+		netif_carrier_on(ndev);
+}
+
+static const struct hns3_hw_error_info hns3_hw_err[] = {
+	{ .type = HNAE3_PPU_POISON_ERROR,
+	  .msg = "PPU poison" },
+	{ .type = HNAE3_CMDQ_ECC_ERROR,
+	  .msg = "IMP CMDQ error" },
+	{ .type = HNAE3_IMP_RD_POISON_ERROR,
+	  .msg = "IMP RD poison" },
+	{ .type = HNAE3_ROCEE_AXI_RESP_ERROR,
+	  .msg = "ROCEE AXI RESP error" },
+};
+
+static void hns3_process_hw_error(struct hnae3_handle *handle,
+				  enum hnae3_hw_error_type type)
+{
+	u32 i;
+
+	for (i = 0; i < ARRAY_SIZE(hns3_hw_err); i++) {
+		if (hns3_hw_err[i].type == type) {
+			dev_err(&handle->pdev->dev, "Detected %s!\n",
+				hns3_hw_err[i].msg);
+			break;
+		}
+	}
+}
+
+static const struct hnae3_client_ops client_ops = {
+	.init_instance = hns3_client_init,
+	.uninit_instance = hns3_client_uninit,
+	.link_status_change = hns3_link_status_change,
+	.reset_notify = hns3_reset_notify,
+	.process_hw_error = hns3_process_hw_error,
+};
+
+/* hns3_init_module - Driver registration routine
+ * hns3_init_module is the first routine called when the driver is
+ * loaded. All it does is register with the PCI subsystem.
+ */
+static int __init hns3_init_module(void)
+{
+	int ret;
+
+	pr_debug("%s: %s - version\n", hns3_driver_name, hns3_driver_string);
+	pr_debug("%s: %s\n", hns3_driver_name, hns3_copyright);
+
+	client.type = HNAE3_CLIENT_KNIC;
+	snprintf(client.name, HNAE3_CLIENT_NAME_LENGTH, "%s",
+		 hns3_driver_name);
+
+	client.ops = &client_ops;
+
+	INIT_LIST_HEAD(&client.node);
+
+	hns3_dbg_register_debugfs(hns3_driver_name);
+
+	ret = hnae3_register_client(&client);
+	if (ret)
+		goto err_reg_client;
+
+	ret = pci_register_driver(&hns3_driver);
+	if (ret)
+		goto err_reg_driver;
+
+	return ret;
+
+err_reg_driver:
+	hnae3_unregister_client(&client);
+err_reg_client:
+	hns3_dbg_unregister_debugfs();
+	return ret;
+}
+module_init(hns3_init_module);
+
+/* hns3_exit_module - Driver exit cleanup routine
+ * hns3_exit_module is called just before the driver is removed
+ * from memory.
+ */
+static void __exit hns3_exit_module(void)
+{
+	hnae3_acquire_unload_lock();
+	pci_unregister_driver(&hns3_driver);
+	hnae3_unregister_client(&client);
+	hns3_dbg_unregister_debugfs();
+	hnae3_release_unload_lock();
+}
+module_exit(hns3_exit_module);
+
+MODULE_DESCRIPTION("HNS3: Hisilicon Ethernet Driver");
+MODULE_AUTHOR("Huawei Tech. Co., Ltd.");
+MODULE_LICENSE("GPL");
+MODULE_ALIAS("pci:hns-nic");
diff --git a/drivers/net/ethernet/hisilicon/hns3/hns3_enet.h b/drivers/net/ethernet/hisilicon/hns3/hns3_enet.h
new file mode 100644
index 000000000000..933e3527ed82
--- /dev/null
+++ b/drivers/net/ethernet/hisilicon/hns3/hns3_enet.h
@@ -0,0 +1,761 @@
+/* SPDX-License-Identifier: GPL-2.0+ */
+// Copyright (c) 2016-2017 Hisilicon Limited.
+
+#ifndef __HNS3_ENET_H
+#define __HNS3_ENET_H
+
+#include <linux/dim.h>
+#include <linux/if_vlan.h>
+#include <net/page_pool/types.h>
+#include <asm/barrier.h>
+
+#include "hnae3.h"
+
+struct iphdr;
+struct ipv6hdr;
+
+enum hns3_nic_state {
+	HNS3_NIC_STATE_TESTING,
+	HNS3_NIC_STATE_RESETTING,
+	HNS3_NIC_STATE_INITED,
+	HNS3_NIC_STATE_DOWN,
+	HNS3_NIC_STATE_DISABLED,
+	HNS3_NIC_STATE_REMOVING,
+	HNS3_NIC_STATE_SERVICE_INITED,
+	HNS3_NIC_STATE_SERVICE_SCHED,
+	HNS3_NIC_STATE2_RESET_REQUESTED,
+	HNS3_NIC_STATE_HW_TX_CSUM_ENABLE,
+	HNS3_NIC_STATE_RXD_ADV_LAYOUT_ENABLE,
+	HNS3_NIC_STATE_TX_PUSH_ENABLE,
+	HNS3_NIC_STATE_MAX
+};
+
+#define HNS3_MAX_PUSH_BD_NUM		2
+
+#define HNS3_RING_RX_RING_BASEADDR_L_REG	0x00000
+#define HNS3_RING_RX_RING_BASEADDR_H_REG	0x00004
+#define HNS3_RING_RX_RING_BD_NUM_REG		0x00008
+#define HNS3_RING_RX_RING_BD_LEN_REG		0x0000C
+#define HNS3_RING_RX_RING_TAIL_REG		0x00018
+#define HNS3_RING_RX_RING_HEAD_REG		0x0001C
+#define HNS3_RING_RX_RING_FBDNUM_REG		0x00020
+#define HNS3_RING_RX_RING_PKTNUM_RECORD_REG	0x0002C
+
+#define HNS3_RING_TX_RING_BASEADDR_L_REG	0x00040
+#define HNS3_RING_TX_RING_BASEADDR_H_REG	0x00044
+#define HNS3_RING_TX_RING_BD_NUM_REG		0x00048
+#define HNS3_RING_TX_RING_TC_REG		0x00050
+#define HNS3_RING_TX_RING_TAIL_REG		0x00058
+#define HNS3_RING_TX_RING_HEAD_REG		0x0005C
+#define HNS3_RING_TX_RING_FBDNUM_REG		0x00060
+#define HNS3_RING_TX_RING_OFFSET_REG		0x00064
+#define HNS3_RING_TX_RING_EBDNUM_REG		0x00068
+#define HNS3_RING_TX_RING_PKTNUM_RECORD_REG	0x0006C
+#define HNS3_RING_TX_RING_EBD_OFFSET_REG	0x00070
+#define HNS3_RING_TX_RING_BD_ERR_REG		0x00074
+#define HNS3_RING_EN_REG			0x00090
+#define HNS3_RING_RX_EN_REG			0x00098
+#define HNS3_RING_TX_EN_REG			0x000D4
+
+#define HNS3_RX_HEAD_SIZE			256
+
+#define HNS3_TX_TIMEOUT (5 * HZ)
+#define HNS3_RING_NAME_LEN			16
+#define HNS3_BUFFER_SIZE_2048			2048
+#define HNS3_RING_MAX_PENDING			32760
+#define HNS3_RING_MIN_PENDING			72
+#define HNS3_RING_BD_MULTIPLE			8
+/* max frame size of mac */
+#define HNS3_MAX_MTU(max_frm_size) \
+	((max_frm_size) - (ETH_HLEN + ETH_FCS_LEN + 2 * VLAN_HLEN))
+
+#define HNS3_BD_SIZE_512_TYPE			0
+#define HNS3_BD_SIZE_1024_TYPE			1
+#define HNS3_BD_SIZE_2048_TYPE			2
+#define HNS3_BD_SIZE_4096_TYPE			3
+
+#define HNS3_RX_FLAG_VLAN_PRESENT		0x1
+#define HNS3_RX_FLAG_L3ID_IPV4			0x0
+#define HNS3_RX_FLAG_L3ID_IPV6			0x1
+#define HNS3_RX_FLAG_L4ID_UDP			0x0
+#define HNS3_RX_FLAG_L4ID_TCP			0x1
+
+#define HNS3_RXD_DMAC_S				0
+#define HNS3_RXD_DMAC_M				(0x3 << HNS3_RXD_DMAC_S)
+#define HNS3_RXD_VLAN_S				2
+#define HNS3_RXD_VLAN_M				(0x3 << HNS3_RXD_VLAN_S)
+#define HNS3_RXD_L3ID_S				4
+#define HNS3_RXD_L3ID_M				(0xf << HNS3_RXD_L3ID_S)
+#define HNS3_RXD_L4ID_S				8
+#define HNS3_RXD_L4ID_M				(0xf << HNS3_RXD_L4ID_S)
+#define HNS3_RXD_FRAG_B				12
+#define HNS3_RXD_STRP_TAGP_S			13
+#define HNS3_RXD_STRP_TAGP_M			(0x3 << HNS3_RXD_STRP_TAGP_S)
+
+#define HNS3_RXD_L2E_B				16
+#define HNS3_RXD_L3E_B				17
+#define HNS3_RXD_L4E_B				18
+#define HNS3_RXD_TRUNCAT_B			19
+#define HNS3_RXD_HOI_B				20
+#define HNS3_RXD_DOI_B				21
+#define HNS3_RXD_OL3E_B				22
+#define HNS3_RXD_OL4E_B				23
+#define HNS3_RXD_GRO_COUNT_S			24
+#define HNS3_RXD_GRO_COUNT_M			(0x3f << HNS3_RXD_GRO_COUNT_S)
+#define HNS3_RXD_GRO_FIXID_B			30
+#define HNS3_RXD_GRO_ECN_B			31
+
+#define HNS3_RXD_ODMAC_S			0
+#define HNS3_RXD_ODMAC_M			(0x3 << HNS3_RXD_ODMAC_S)
+#define HNS3_RXD_OVLAN_S			2
+#define HNS3_RXD_OVLAN_M			(0x3 << HNS3_RXD_OVLAN_S)
+#define HNS3_RXD_OL3ID_S			4
+#define HNS3_RXD_OL3ID_M			(0xf << HNS3_RXD_OL3ID_S)
+#define HNS3_RXD_OL4ID_S			8
+#define HNS3_RXD_OL4ID_M			(0xf << HNS3_RXD_OL4ID_S)
+#define HNS3_RXD_FBHI_S				12
+#define HNS3_RXD_FBHI_M				(0x3 << HNS3_RXD_FBHI_S)
+#define HNS3_RXD_FBLI_S				14
+#define HNS3_RXD_FBLI_M				(0x3 << HNS3_RXD_FBLI_S)
+
+#define HNS3_RXD_PTYPE_S			4
+#define HNS3_RXD_PTYPE_M			GENMASK(11, 4)
+
+#define HNS3_RXD_BDTYPE_S			0
+#define HNS3_RXD_BDTYPE_M			(0xf << HNS3_RXD_BDTYPE_S)
+#define HNS3_RXD_VLD_B				4
+#define HNS3_RXD_UDP0_B				5
+#define HNS3_RXD_EXTEND_B			7
+#define HNS3_RXD_FE_B				8
+#define HNS3_RXD_LUM_B				9
+#define HNS3_RXD_CRCP_B				10
+#define HNS3_RXD_L3L4P_B			11
+#define HNS3_RXD_TSIDX_S			12
+#define HNS3_RXD_TSIDX_M			(0x3 << HNS3_RXD_TSIDX_S)
+#define HNS3_RXD_TS_VLD_B			14
+#define HNS3_RXD_LKBK_B				15
+#define HNS3_RXD_GRO_SIZE_S			16
+#define HNS3_RXD_GRO_SIZE_M			(0x3fff << HNS3_RXD_GRO_SIZE_S)
+
+#define HNS3_TXD_L3T_S				0
+#define HNS3_TXD_L3T_M				(0x3 << HNS3_TXD_L3T_S)
+#define HNS3_TXD_L4T_S				2
+#define HNS3_TXD_L4T_M				(0x3 << HNS3_TXD_L4T_S)
+#define HNS3_TXD_L3CS_B				4
+#define HNS3_TXD_L4CS_B				5
+#define HNS3_TXD_VLAN_B				6
+#define HNS3_TXD_TSO_B				7
+
+#define HNS3_TXD_L2LEN_S			8
+#define HNS3_TXD_L2LEN_M			(0xff << HNS3_TXD_L2LEN_S)
+#define HNS3_TXD_L3LEN_S			16
+#define HNS3_TXD_L3LEN_M			(0xff << HNS3_TXD_L3LEN_S)
+#define HNS3_TXD_L4LEN_S			24
+#define HNS3_TXD_L4LEN_M			(0xff << HNS3_TXD_L4LEN_S)
+
+#define HNS3_TXD_CSUM_START_S		8
+#define HNS3_TXD_CSUM_START_M		(0xffff << HNS3_TXD_CSUM_START_S)
+
+#define HNS3_TXD_OL3T_S				0
+#define HNS3_TXD_OL3T_M				(0x3 << HNS3_TXD_OL3T_S)
+#define HNS3_TXD_OVLAN_B			2
+#define HNS3_TXD_MACSEC_B			3
+#define HNS3_TXD_TUNTYPE_S			4
+#define HNS3_TXD_TUNTYPE_M			(0xf << HNS3_TXD_TUNTYPE_S)
+
+#define HNS3_TXD_CSUM_OFFSET_S		8
+#define HNS3_TXD_CSUM_OFFSET_M		(0xffff << HNS3_TXD_CSUM_OFFSET_S)
+
+#define HNS3_TXD_BDTYPE_S			0
+#define HNS3_TXD_BDTYPE_M			(0xf << HNS3_TXD_BDTYPE_S)
+#define HNS3_TXD_FE_B				4
+#define HNS3_TXD_SC_S				5
+#define HNS3_TXD_SC_M				(0x3 << HNS3_TXD_SC_S)
+#define HNS3_TXD_EXTEND_B			7
+#define HNS3_TXD_VLD_B				8
+#define HNS3_TXD_RI_B				9
+#define HNS3_TXD_RA_B				10
+#define HNS3_TXD_TSYN_B				11
+#define HNS3_TXD_DECTTL_S			12
+#define HNS3_TXD_DECTTL_M			(0xf << HNS3_TXD_DECTTL_S)
+
+#define HNS3_TXD_OL4CS_B			22
+
+#define HNS3_TXD_MSS_S				0
+#define HNS3_TXD_MSS_M				(0x3fff << HNS3_TXD_MSS_S)
+#define HNS3_TXD_HW_CS_B			14
+
+#define HNS3_VECTOR_TX_IRQ			BIT_ULL(0)
+#define HNS3_VECTOR_RX_IRQ			BIT_ULL(1)
+
+#define HNS3_VECTOR_NOT_INITED			0
+#define HNS3_VECTOR_INITED			1
+
+#define HNS3_MAX_BD_SIZE			65535
+#define HNS3_MAX_TSO_BD_NUM			63U
+#define HNS3_MAX_TSO_SIZE			1048576U
+#define HNS3_MAX_NON_TSO_SIZE			9728U
+
+#define HNS3_VECTOR_GL_MASK			GENMASK(11, 0)
+#define HNS3_VECTOR_GL0_OFFSET			0x100
+#define HNS3_VECTOR_GL1_OFFSET			0x200
+#define HNS3_VECTOR_GL2_OFFSET			0x300
+#define HNS3_VECTOR_RL_OFFSET			0x900
+#define HNS3_VECTOR_RL_EN_B			6
+#define HNS3_VECTOR_QL_MASK			GENMASK(9, 0)
+#define HNS3_VECTOR_TX_QL_OFFSET		0xe00
+#define HNS3_VECTOR_RX_QL_OFFSET		0xf00
+
+#define HNS3_RING_EN_B				0
+
+#define HNS3_GL0_CQ_MODE_REG			0x20d00
+#define HNS3_GL1_CQ_MODE_REG			0x20d04
+#define HNS3_GL2_CQ_MODE_REG			0x20d08
+#define HNS3_CQ_MODE_EQE			1U
+#define HNS3_CQ_MODE_CQE			0U
+
+#define HNS3_RESCHED_BD_NUM			1024
+
+enum hns3_pkt_l2t_type {
+	HNS3_L2_TYPE_UNICAST,
+	HNS3_L2_TYPE_MULTICAST,
+	HNS3_L2_TYPE_BROADCAST,
+	HNS3_L2_TYPE_INVALID,
+};
+
+enum hns3_pkt_l3t_type {
+	HNS3_L3T_NONE,
+	HNS3_L3T_IPV6,
+	HNS3_L3T_IPV4,
+	HNS3_L3T_RESERVED
+};
+
+enum hns3_pkt_l4t_type {
+	HNS3_L4T_UNKNOWN,
+	HNS3_L4T_TCP,
+	HNS3_L4T_UDP,
+	HNS3_L4T_SCTP
+};
+
+enum hns3_pkt_ol3t_type {
+	HNS3_OL3T_NONE,
+	HNS3_OL3T_IPV6,
+	HNS3_OL3T_IPV4_NO_CSUM,
+	HNS3_OL3T_IPV4_CSUM
+};
+
+enum hns3_pkt_tun_type {
+	HNS3_TUN_NONE,
+	HNS3_TUN_MAC_IN_UDP,
+	HNS3_TUN_NVGRE,
+	HNS3_TUN_OTHER
+};
+
+/* hardware spec ring buffer format */
+struct __packed hns3_desc {
+	union {
+		__le64 addr;
+		__le16 csum;
+		struct {
+			__le32 ts_nsec;
+			__le32 ts_sec;
+		};
+	};
+	union {
+		struct {
+			__le16 vlan_tag;
+			__le16 send_size;
+			union {
+				__le32 type_cs_vlan_tso_len;
+				struct {
+					__u8 type_cs_vlan_tso;
+					__u8 l2_len;
+					__u8 l3_len;
+					__u8 l4_len;
+				};
+			};
+			__le16 outer_vlan_tag;
+			__le16 tv;
+
+		union {
+			__le32 ol_type_vlan_len_msec;
+			struct {
+				__u8 ol_type_vlan_msec;
+				__u8 ol2_len;
+				__u8 ol3_len;
+				__u8 ol4_len;
+			};
+		};
+
+			__le32 paylen_ol4cs;
+			__le16 bdtp_fe_sc_vld_ra_ri;
+			__le16 mss_hw_csum;
+		} tx;
+
+		struct {
+			__le32 l234_info;
+			__le16 pkt_len;
+			__le16 size;
+
+			__le32 rss_hash;
+			__le16 fd_id;
+			__le16 vlan_tag;
+
+			union {
+				__le32 ol_info;
+				struct {
+					__le16 o_dm_vlan_id_fb;
+					__le16 ot_vlan_tag;
+				};
+			};
+
+			__le32 bd_base_info;
+		} rx;
+	};
+};
+
+enum hns3_desc_type {
+	DESC_TYPE_UNKNOWN		= 0,
+	DESC_TYPE_SKB			= 1 << 0,
+	DESC_TYPE_FRAGLIST_SKB		= 1 << 1,
+	DESC_TYPE_PAGE			= 1 << 2,
+	DESC_TYPE_BOUNCE_ALL		= 1 << 3,
+	DESC_TYPE_BOUNCE_HEAD		= 1 << 4,
+	DESC_TYPE_SGL_SKB		= 1 << 5,
+	DESC_TYPE_PP_FRAG		= 1 << 6,
+};
+
+struct hns3_desc_cb {
+	dma_addr_t dma; /* dma address of this desc */
+	void *buf;      /* cpu addr for a desc */
+
+	/* priv data for the desc, e.g. skb when use with ip stack */
+	void *priv;
+
+	union {
+		u32 page_offset;	/* for rx */
+		u32 send_bytes;		/* for tx */
+	};
+
+	u32 length;     /* length of the buffer */
+
+	u16 reuse_flag;
+	u16 refill;
+
+	/* desc type, used by the ring user to mark the type of the priv data */
+	u16 type;
+	u16 pagecnt_bias;
+};
+
+enum hns3_pkt_l3type {
+	HNS3_L3_TYPE_IPV4,
+	HNS3_L3_TYPE_IPV6,
+	HNS3_L3_TYPE_ARP,
+	HNS3_L3_TYPE_RARP,
+	HNS3_L3_TYPE_IPV4_OPT,
+	HNS3_L3_TYPE_IPV6_EXT,
+	HNS3_L3_TYPE_LLDP,
+	HNS3_L3_TYPE_BPDU,
+	HNS3_L3_TYPE_MAC_PAUSE,
+	HNS3_L3_TYPE_PFC_PAUSE, /* 0x9 */
+
+	/* reserved for 0xA~0xB */
+
+	HNS3_L3_TYPE_CNM = 0xc,
+
+	/* reserved for 0xD~0xE */
+
+	HNS3_L3_TYPE_PARSE_FAIL	= 0xf /* must be last */
+};
+
+enum hns3_pkt_l4type {
+	HNS3_L4_TYPE_UDP,
+	HNS3_L4_TYPE_TCP,
+	HNS3_L4_TYPE_GRE,
+	HNS3_L4_TYPE_SCTP,
+	HNS3_L4_TYPE_IGMP,
+	HNS3_L4_TYPE_ICMP,
+
+	/* reserved for 0x6~0xE */
+
+	HNS3_L4_TYPE_PARSE_FAIL	= 0xf /* must be last */
+};
+
+enum hns3_pkt_ol3type {
+	HNS3_OL3_TYPE_IPV4 = 0,
+	HNS3_OL3_TYPE_IPV6,
+	/* reserved for 0x2~0x3 */
+	HNS3_OL3_TYPE_IPV4_OPT = 4,
+	HNS3_OL3_TYPE_IPV6_EXT,
+
+	/* reserved for 0x6~0xE */
+
+	HNS3_OL3_TYPE_PARSE_FAIL = 0xf	/* must be last */
+};
+
+enum hns3_pkt_ol4type {
+	HNS3_OL4_TYPE_NO_TUN,
+	HNS3_OL4_TYPE_MAC_IN_UDP,
+	HNS3_OL4_TYPE_NVGRE,
+	HNS3_OL4_TYPE_UNKNOWN
+};
+
+struct hns3_rx_ptype {
+	u32 ptype : 8;
+	u32 csum_level : 2;
+	u32 ip_summed : 2;
+	u32 l3_type : 4;
+	u32 valid : 1;
+	u32 hash_type: 3;
+};
+
+struct ring_stats {
+	u64 sw_err_cnt;
+	u64 seg_pkt_cnt;
+	union {
+		struct {
+			u64 tx_pkts;
+			u64 tx_bytes;
+			u64 tx_more;
+			u64 tx_push;
+			u64 tx_mem_doorbell;
+			u64 restart_queue;
+			u64 tx_busy;
+			u64 tx_copy;
+			u64 tx_vlan_err;
+			u64 tx_l4_proto_err;
+			u64 tx_l2l3l4_err;
+			u64 tx_tso_err;
+			u64 over_max_recursion;
+			u64 hw_limitation;
+			u64 tx_bounce;
+			u64 tx_spare_full;
+			u64 copy_bits_err;
+			u64 tx_sgl;
+			u64 skb2sgl_err;
+			u64 map_sg_err;
+		};
+		struct {
+			u64 rx_pkts;
+			u64 rx_bytes;
+			u64 rx_err_cnt;
+			u64 reuse_pg_cnt;
+			u64 err_pkt_len;
+			u64 err_bd_num;
+			u64 l2_err;
+			u64 l3l4_csum_err;
+			u64 csum_complete;
+			u64 rx_multicast;
+			u64 non_reuse_pg;
+			u64 frag_alloc_err;
+			u64 frag_alloc;
+		};
+		__le16 csum;
+	};
+};
+
+struct hns3_tx_spare {
+	dma_addr_t dma;
+	void *buf;
+	u32 next_to_use;
+	u32 next_to_clean;
+	u32 last_to_clean;
+	u32 len;
+};
+
+struct hns3_enet_ring {
+	struct hns3_desc *desc; /* dma map address space */
+	struct hns3_desc_cb *desc_cb;
+	struct hns3_enet_ring *next;
+	struct hns3_enet_tqp_vector *tqp_vector;
+	struct hnae3_queue *tqp;
+	int queue_index;
+	struct device *dev; /* will be used for DMA mapping of descriptors */
+	struct page_pool *page_pool;
+
+	/* statistic */
+	struct ring_stats stats;
+	struct u64_stats_sync syncp;
+
+	dma_addr_t desc_dma_addr;
+	u32 buf_size;       /* size for hnae_desc->addr, preset by AE */
+	u16 desc_num;       /* total number of desc */
+	int next_to_use;    /* idx of next spare desc */
+
+	/* idx of lastest sent desc, the ring is empty when equal to
+	 * next_to_use
+	 */
+	int next_to_clean;
+	u32 flag;          /* ring attribute */
+
+	int pending_buf;
+	union {
+		/* for Tx ring */
+		struct {
+			u32 fd_qb_tx_sample;
+			int last_to_use;        /* last idx used by xmit */
+			u32 tx_copybreak;
+			struct hns3_tx_spare *tx_spare;
+		};
+
+		/* for Rx ring */
+		struct {
+			u32 pull_len;   /* memcpy len for current rx packet */
+			u32 rx_copybreak;
+			u32 frag_num;
+			/* first buffer address for current packet */
+			unsigned char *va;
+			struct sk_buff *skb;
+			struct sk_buff *tail_skb;
+		};
+	};
+} ____cacheline_internodealigned_in_smp;
+
+enum hns3_flow_level_range {
+	HNS3_FLOW_LOW = 0,
+	HNS3_FLOW_MID = 1,
+	HNS3_FLOW_HIGH = 2,
+	HNS3_FLOW_ULTRA = 3,
+};
+
+#define HNS3_INT_GL_50K			0x0014
+#define HNS3_INT_GL_20K			0x0032
+#define HNS3_INT_GL_18K			0x0036
+#define HNS3_INT_GL_8K			0x007C
+
+#define HNS3_INT_GL_1US			BIT(31)
+
+#define HNS3_INT_RL_MAX			0x00EC
+#define HNS3_INT_RL_ENABLE_MASK		0x40
+
+#define HNS3_INT_QL_DEFAULT_CFG		0x20
+
+struct hns3_enet_coalesce {
+	u16 int_gl;
+	u16 int_ql;
+	u16 int_ql_max;
+	u8 adapt_enable : 1;
+	u8 ql_enable : 1;
+	u8 unit_1us : 1;
+	enum hns3_flow_level_range flow_level;
+};
+
+struct hns3_enet_ring_group {
+	/* array of pointers to rings */
+	struct hns3_enet_ring *ring;
+	u64 total_bytes;	/* total bytes processed this group */
+	u64 total_packets;	/* total packets processed this group */
+	u16 count;
+	struct hns3_enet_coalesce coal;
+	struct dim dim;
+};
+
+struct hns3_enet_tqp_vector {
+	struct hnae3_handle *handle;
+	u8 __iomem *mask_addr;
+	int vector_irq;
+	int irq_init_flag;
+
+	u16 idx;		/* index in the TQP vector array per handle. */
+
+	struct napi_struct napi;
+
+	struct hns3_enet_ring_group rx_group;
+	struct hns3_enet_ring_group tx_group;
+
+	cpumask_t affinity_mask;
+	u16 num_tqps;	/* total number of tqps in TQP vector */
+	struct irq_affinity_notify affinity_notify;
+
+	char name[HNAE3_INT_NAME_LEN];
+
+	u64 event_cnt;
+} ____cacheline_internodealigned_in_smp;
+
+struct hns3_nic_priv {
+	struct hnae3_handle *ae_handle;
+	struct net_device *netdev;
+	struct device *dev;
+
+	/**
+	 * the cb for nic to manage the ring buffer, the first half of the
+	 * array is for tx_ring and vice versa for the second half
+	 */
+	struct hns3_enet_ring *ring;
+	struct hns3_enet_tqp_vector *tqp_vector;
+	u16 vector_num;
+	u8 max_non_tso_bd_num;
+
+	u64 tx_timeout_count;
+
+	unsigned long state;
+
+	enum dim_cq_period_mode tx_cqe_mode;
+	enum dim_cq_period_mode rx_cqe_mode;
+	struct hns3_enet_coalesce tx_coal;
+	struct hns3_enet_coalesce rx_coal;
+	u32 tx_copybreak;
+	u32 rx_copybreak;
+	u32 min_tx_copybreak;
+	u32 min_tx_spare_buf_size;
+};
+
+union l3_hdr_info {
+	struct iphdr *v4;
+	struct ipv6hdr *v6;
+	unsigned char *hdr;
+};
+
+union l4_hdr_info {
+	struct tcphdr *tcp;
+	struct udphdr *udp;
+	struct gre_base_hdr *gre;
+	unsigned char *hdr;
+};
+
+struct hns3_hw_error_info {
+	enum hnae3_hw_error_type type;
+	const char *msg;
+};
+
+struct hns3_reset_type_map {
+	enum ethtool_reset_flags rst_flags;
+	enum hnae3_reset_type rst_type;
+};
+
+static inline u32 ring_space(struct hns3_enet_ring *ring)
+{
+	/* This smp_load_acquire() pairs with smp_store_release() in
+	 * hns3_nic_reclaim_one_desc called by hns3_clean_tx_ring.
+	 */
+	int begin = smp_load_acquire(&ring->next_to_clean);
+	int end = READ_ONCE(ring->next_to_use);
+
+	return ((end >= begin) ? (ring->desc_num - end + begin) :
+			(begin - end)) - 1;
+}
+
+static inline u32 hns3_tqp_read_reg(struct hns3_enet_ring *ring, u32 reg)
+{
+	return readl_relaxed(ring->tqp->io_base + reg);
+}
+
+static inline u32 hns3_read_reg(void __iomem *base, u32 reg)
+{
+	return readl(base + reg);
+}
+
+static inline void hns3_write_reg(void __iomem *base, u32 reg, u32 value)
+{
+	u8 __iomem *reg_addr = READ_ONCE(base);
+
+	writel(value, reg_addr + reg);
+}
+
+#define hns3_read_dev(a, reg) \
+	hns3_read_reg((a)->io_base, reg)
+
+static inline bool hns3_nic_resetting(struct net_device *netdev)
+{
+	struct hns3_nic_priv *priv = netdev_priv(netdev);
+
+	return test_bit(HNS3_NIC_STATE_RESETTING, &priv->state);
+}
+
+#define hns3_write_dev(a, reg, value) \
+	hns3_write_reg((a)->io_base, reg, value)
+
+#define ring_to_dev(ring) ((ring)->dev)
+
+#define ring_to_netdev(ring)	((ring)->tqp_vector->napi.dev)
+
+#define ring_to_dma_dir(ring) (HNAE3_IS_TX_RING(ring) ? \
+	DMA_TO_DEVICE : DMA_FROM_DEVICE)
+
+#define hns3_buf_size(_ring) ((_ring)->buf_size)
+
+#define hns3_ring_stats_update(ring, cnt) do { \
+	typeof(ring) (tmp) = (ring); \
+	u64_stats_update_begin(&(tmp)->syncp); \
+	((tmp)->stats.cnt)++; \
+	u64_stats_update_end(&(tmp)->syncp); \
+} while (0) \
+
+static inline unsigned int hns3_page_order(struct hns3_enet_ring *ring)
+{
+#if (PAGE_SIZE < 8192)
+	if (ring->buf_size > (PAGE_SIZE / 2))
+		return 1;
+#endif
+	return 0;
+}
+
+#define hns3_page_size(_ring) (PAGE_SIZE << hns3_page_order(_ring))
+
+/* iterator for handling rings in ring group */
+#define hns3_for_each_ring(pos, head) \
+	for ((pos) = (head).ring; (pos); (pos) = (pos)->next)
+
+#define hns3_get_handle(ndev) \
+	(((struct hns3_nic_priv *)netdev_priv(ndev))->ae_handle)
+
+#define hns3_get_ae_dev(handle) \
+	(pci_get_drvdata((handle)->pdev))
+
+#define hns3_get_ops(handle) \
+	((handle)->ae_algo->ops)
+
+#define hns3_gl_usec_to_reg(int_gl) ((int_gl) >> 1)
+#define hns3_gl_round_down(int_gl) round_down(int_gl, 2)
+
+#define hns3_rl_usec_to_reg(int_rl) ((int_rl) >> 2)
+#define hns3_rl_round_down(int_rl) round_down(int_rl, 4)
+
+void hns3_ethtool_set_ops(struct net_device *netdev);
+int hns3_set_channels(struct net_device *netdev,
+		      struct ethtool_channels *ch);
+
+void hns3_clean_tx_ring(struct hns3_enet_ring *ring, int budget);
+int hns3_init_all_ring(struct hns3_nic_priv *priv);
+int hns3_nic_reset_all_ring(struct hnae3_handle *h);
+void hns3_fini_ring(struct hns3_enet_ring *ring);
+netdev_tx_t hns3_nic_net_xmit(struct sk_buff *skb, struct net_device *netdev);
+bool hns3_is_phys_func(struct pci_dev *pdev);
+int hns3_clean_rx_ring(
+		struct hns3_enet_ring *ring, int budget,
+		void (*rx_fn)(struct hns3_enet_ring *, struct sk_buff *));
+
+void hns3_set_vector_coalesce_rx_gl(struct hns3_enet_tqp_vector *tqp_vector,
+				    u32 gl_value);
+void hns3_set_vector_coalesce_tx_gl(struct hns3_enet_tqp_vector *tqp_vector,
+				    u32 gl_value);
+void hns3_set_vector_coalesce_rl(struct hns3_enet_tqp_vector *tqp_vector,
+				 u32 rl_value);
+void hns3_set_vector_coalesce_rx_ql(struct hns3_enet_tqp_vector *tqp_vector,
+				    u32 ql_value);
+void hns3_set_vector_coalesce_tx_ql(struct hns3_enet_tqp_vector *tqp_vector,
+				    u32 ql_value);
+
+void hns3_request_update_promisc_mode(struct hnae3_handle *handle);
+int hns3_reset_notify(struct hnae3_handle *handle,
+		      enum hnae3_reset_notify_type type);
+
+#ifdef CONFIG_HNS3_DCB
+void hns3_dcbnl_setup(struct hnae3_handle *handle);
+#else
+static inline void hns3_dcbnl_setup(struct hnae3_handle *handle) {}
+#endif
+
+int hns3_dbg_init(struct hnae3_handle *handle);
+void hns3_dbg_uninit(struct hnae3_handle *handle);
+void hns3_dbg_register_debugfs(const char *debugfs_dir_name);
+void hns3_dbg_unregister_debugfs(void);
+void hns3_shinfo_pack(struct skb_shared_info *shinfo, __u32 *size);
+u16 hns3_get_max_available_channels(struct hnae3_handle *h);
+void hns3_cq_period_mode_init(struct hns3_nic_priv *priv,
+			      enum dim_cq_period_mode tx_mode,
+			      enum dim_cq_period_mode rx_mode);
+
+void hns3_external_lb_prepare(struct net_device *ndev, bool if_running);
+void hns3_external_lb_restore(struct net_device *ndev, bool if_running);
+#endif
diff --git a/drivers/net/ethernet/hisilicon/hns3/hns3_ethtool.c b/drivers/net/ethernet/hisilicon/hns3/hns3_ethtool.c
new file mode 100644
index 000000000000..a5eefa28454c
--- /dev/null
+++ b/drivers/net/ethernet/hisilicon/hns3/hns3_ethtool.c
@@ -0,0 +1,2232 @@
+// SPDX-License-Identifier: GPL-2.0+
+// Copyright (c) 2016-2017 Hisilicon Limited.
+
+#include <linux/etherdevice.h>
+#include <linux/string.h>
+#include <linux/string_choices.h>
+#include <linux/phy.h>
+#include <linux/sfp.h>
+
+#include "hns3_enet.h"
+#include "hns3_ethtool.h"
+
+/* tqp related stats */
+#define HNS3_TQP_STAT(_string, _member)	{			\
+	.stats_string = _string,				\
+	.stats_offset = offsetof(struct hns3_enet_ring, stats) +\
+			offsetof(struct ring_stats, _member),   \
+}
+
+static const struct hns3_stats hns3_txq_stats[] = {
+	/* Tx per-queue statistics */
+	HNS3_TQP_STAT("dropped", sw_err_cnt),
+	HNS3_TQP_STAT("seg_pkt_cnt", seg_pkt_cnt),
+	HNS3_TQP_STAT("packets", tx_pkts),
+	HNS3_TQP_STAT("bytes", tx_bytes),
+	HNS3_TQP_STAT("more", tx_more),
+	HNS3_TQP_STAT("push", tx_push),
+	HNS3_TQP_STAT("mem_doorbell", tx_mem_doorbell),
+	HNS3_TQP_STAT("wake", restart_queue),
+	HNS3_TQP_STAT("busy", tx_busy),
+	HNS3_TQP_STAT("copy", tx_copy),
+	HNS3_TQP_STAT("vlan_err", tx_vlan_err),
+	HNS3_TQP_STAT("l4_proto_err", tx_l4_proto_err),
+	HNS3_TQP_STAT("l2l3l4_err", tx_l2l3l4_err),
+	HNS3_TQP_STAT("tso_err", tx_tso_err),
+	HNS3_TQP_STAT("over_max_recursion", over_max_recursion),
+	HNS3_TQP_STAT("hw_limitation", hw_limitation),
+	HNS3_TQP_STAT("bounce", tx_bounce),
+	HNS3_TQP_STAT("spare_full", tx_spare_full),
+	HNS3_TQP_STAT("copy_bits_err", copy_bits_err),
+	HNS3_TQP_STAT("sgl", tx_sgl),
+	HNS3_TQP_STAT("skb2sgl_err", skb2sgl_err),
+	HNS3_TQP_STAT("map_sg_err", map_sg_err),
+};
+
+#define HNS3_TXQ_STATS_COUNT ARRAY_SIZE(hns3_txq_stats)
+
+static const struct hns3_stats hns3_rxq_stats[] = {
+	/* Rx per-queue statistics */
+	HNS3_TQP_STAT("dropped", sw_err_cnt),
+	HNS3_TQP_STAT("seg_pkt_cnt", seg_pkt_cnt),
+	HNS3_TQP_STAT("packets", rx_pkts),
+	HNS3_TQP_STAT("bytes", rx_bytes),
+	HNS3_TQP_STAT("errors", rx_err_cnt),
+	HNS3_TQP_STAT("reuse_pg_cnt", reuse_pg_cnt),
+	HNS3_TQP_STAT("err_pkt_len", err_pkt_len),
+	HNS3_TQP_STAT("err_bd_num", err_bd_num),
+	HNS3_TQP_STAT("l2_err", l2_err),
+	HNS3_TQP_STAT("l3l4_csum_err", l3l4_csum_err),
+	HNS3_TQP_STAT("csum_complete", csum_complete),
+	HNS3_TQP_STAT("multicast", rx_multicast),
+	HNS3_TQP_STAT("non_reuse_pg", non_reuse_pg),
+	HNS3_TQP_STAT("frag_alloc_err", frag_alloc_err),
+	HNS3_TQP_STAT("frag_alloc", frag_alloc),
+};
+
+#define HNS3_PRIV_FLAGS_LEN ARRAY_SIZE(hns3_priv_flags)
+
+#define HNS3_RXQ_STATS_COUNT ARRAY_SIZE(hns3_rxq_stats)
+
+#define HNS3_TQP_STATS_COUNT (HNS3_TXQ_STATS_COUNT + HNS3_RXQ_STATS_COUNT)
+
+#define HNS3_NIC_LB_TEST_PKT_NUM	1
+#define HNS3_NIC_LB_TEST_RING_ID	0
+#define HNS3_NIC_LB_TEST_PACKET_SIZE	128
+#define HNS3_NIC_LB_SETUP_USEC		10000
+
+/* Nic loopback test err  */
+#define HNS3_NIC_LB_TEST_NO_MEM_ERR	1
+#define HNS3_NIC_LB_TEST_TX_CNT_ERR	2
+#define HNS3_NIC_LB_TEST_RX_CNT_ERR	3
+#define HNS3_NIC_LB_TEST_UNEXECUTED	4
+
+static int hns3_get_sset_count(struct net_device *netdev, int stringset);
+
+static int hns3_lp_setup(struct net_device *ndev, enum hnae3_loop loop, bool en)
+{
+	struct hnae3_handle *h = hns3_get_handle(ndev);
+	struct hnae3_ae_dev *ae_dev = hns3_get_ae_dev(h);
+	int ret;
+
+	if (!h->ae_algo->ops->set_loopback ||
+	    !h->ae_algo->ops->set_promisc_mode)
+		return -EOPNOTSUPP;
+
+	switch (loop) {
+	case HNAE3_LOOP_SERIAL_SERDES:
+	case HNAE3_LOOP_PARALLEL_SERDES:
+	case HNAE3_LOOP_APP:
+	case HNAE3_LOOP_PHY:
+	case HNAE3_LOOP_EXTERNAL:
+		ret = h->ae_algo->ops->set_loopback(h, loop, en);
+		break;
+	default:
+		ret = -ENOTSUPP;
+		break;
+	}
+
+	if (ret || ae_dev->dev_version >= HNAE3_DEVICE_VERSION_V2)
+		return ret;
+
+	if (en)
+		h->ae_algo->ops->set_promisc_mode(h, true, true);
+	else
+		/* recover promisc mode before loopback test */
+		hns3_request_update_promisc_mode(h);
+
+	return ret;
+}
+
+static int hns3_lp_up(struct net_device *ndev, enum hnae3_loop loop_mode)
+{
+	struct hnae3_handle *h = hns3_get_handle(ndev);
+	int ret;
+
+	ret = hns3_nic_reset_all_ring(h);
+	if (ret)
+		return ret;
+
+	ret = hns3_lp_setup(ndev, loop_mode, true);
+	usleep_range(HNS3_NIC_LB_SETUP_USEC, HNS3_NIC_LB_SETUP_USEC * 2);
+
+	return ret;
+}
+
+static int hns3_lp_down(struct net_device *ndev, enum hnae3_loop loop_mode)
+{
+	int ret;
+
+	ret = hns3_lp_setup(ndev, loop_mode, false);
+	if (ret) {
+		netdev_err(ndev, "lb_setup return error: %d\n", ret);
+		return ret;
+	}
+
+	usleep_range(HNS3_NIC_LB_SETUP_USEC, HNS3_NIC_LB_SETUP_USEC * 2);
+
+	return 0;
+}
+
+static void hns3_lp_setup_skb(struct sk_buff *skb)
+{
+#define	HNS3_NIC_LB_DST_MAC_ADDR	0x1f
+
+	struct net_device *ndev = skb->dev;
+	struct hnae3_handle *handle;
+	struct hnae3_ae_dev *ae_dev;
+	unsigned char *packet;
+	struct ethhdr *ethh;
+	unsigned int i;
+
+	skb_reserve(skb, NET_IP_ALIGN);
+	ethh = skb_put(skb, sizeof(struct ethhdr));
+	packet = skb_put(skb, HNS3_NIC_LB_TEST_PACKET_SIZE);
+
+	memcpy(ethh->h_dest, ndev->dev_addr, ETH_ALEN);
+
+	/* The dst mac addr of loopback packet is the same as the host'
+	 * mac addr, the SSU component may loop back the packet to host
+	 * before the packet reaches mac or serdes, which will defect
+	 * the purpose of mac or serdes selftest.
+	 */
+	handle = hns3_get_handle(ndev);
+	ae_dev = hns3_get_ae_dev(handle);
+	if (ae_dev->dev_version < HNAE3_DEVICE_VERSION_V2)
+		ethh->h_dest[5] += HNS3_NIC_LB_DST_MAC_ADDR;
+	eth_zero_addr(ethh->h_source);
+	ethh->h_proto = htons(ETH_P_ARP);
+	skb_reset_mac_header(skb);
+
+	for (i = 0; i < HNS3_NIC_LB_TEST_PACKET_SIZE; i++)
+		packet[i] = (unsigned char)(i & 0xff);
+}
+
+static void hns3_lb_check_skb_data(struct hns3_enet_ring *ring,
+				   struct sk_buff *skb)
+{
+	struct hns3_enet_tqp_vector *tqp_vector = ring->tqp_vector;
+	unsigned char *packet = skb->data;
+	u32 len = skb_headlen(skb);
+	u32 i;
+
+	len = min_t(u32, len, HNS3_NIC_LB_TEST_PACKET_SIZE);
+
+	for (i = 0; i < len; i++)
+		if (packet[i] != (unsigned char)(i & 0xff))
+			break;
+
+	/* The packet is correctly received */
+	if (i == HNS3_NIC_LB_TEST_PACKET_SIZE)
+		tqp_vector->rx_group.total_packets++;
+	else
+		print_hex_dump(KERN_ERR, "selftest:", DUMP_PREFIX_OFFSET, 16, 1,
+			       skb->data, len, true);
+
+	dev_kfree_skb_any(skb);
+}
+
+static u32 hns3_lb_check_rx_ring(struct hns3_nic_priv *priv, u32 budget)
+{
+	struct hnae3_handle *h = priv->ae_handle;
+	struct hnae3_knic_private_info *kinfo;
+	u32 i, rcv_good_pkt_total = 0;
+
+	kinfo = &h->kinfo;
+	for (i = kinfo->num_tqps; i < kinfo->num_tqps * 2; i++) {
+		struct hns3_enet_ring *ring = &priv->ring[i];
+		struct hns3_enet_ring_group *rx_group;
+		u64 pre_rx_pkt;
+
+		rx_group = &ring->tqp_vector->rx_group;
+		pre_rx_pkt = rx_group->total_packets;
+
+		preempt_disable();
+		hns3_clean_rx_ring(ring, budget, hns3_lb_check_skb_data);
+		preempt_enable();
+
+		rcv_good_pkt_total += (rx_group->total_packets - pre_rx_pkt);
+		rx_group->total_packets = pre_rx_pkt;
+	}
+	return rcv_good_pkt_total;
+}
+
+static void hns3_lb_clear_tx_ring(struct hns3_nic_priv *priv, u32 start_ringid,
+				  u32 end_ringid)
+{
+	u32 i;
+
+	for (i = start_ringid; i <= end_ringid; i++) {
+		struct hns3_enet_ring *ring = &priv->ring[i];
+
+		hns3_clean_tx_ring(ring, 0);
+	}
+}
+
+/**
+ * hns3_lp_run_test - run loopback test
+ * @ndev: net device
+ * @mode: loopback type
+ *
+ * Return: %0 for success or a NIC loopback test error code on failure
+ */
+static int hns3_lp_run_test(struct net_device *ndev, enum hnae3_loop mode)
+{
+	struct hns3_nic_priv *priv = netdev_priv(ndev);
+	struct sk_buff *skb;
+	u32 i, good_cnt;
+	int ret_val = 0;
+
+	skb = alloc_skb(HNS3_NIC_LB_TEST_PACKET_SIZE + ETH_HLEN + NET_IP_ALIGN,
+			GFP_KERNEL);
+	if (!skb)
+		return HNS3_NIC_LB_TEST_NO_MEM_ERR;
+
+	skb->dev = ndev;
+	hns3_lp_setup_skb(skb);
+	skb->queue_mapping = HNS3_NIC_LB_TEST_RING_ID;
+
+	good_cnt = 0;
+	for (i = 0; i < HNS3_NIC_LB_TEST_PKT_NUM; i++) {
+		netdev_tx_t tx_ret;
+
+		skb_get(skb);
+		tx_ret = hns3_nic_net_xmit(skb, ndev);
+		if (tx_ret == NETDEV_TX_OK) {
+			good_cnt++;
+		} else {
+			kfree_skb(skb);
+			netdev_err(ndev, "hns3_lb_run_test xmit failed: %d\n",
+				   tx_ret);
+		}
+	}
+	if (good_cnt != HNS3_NIC_LB_TEST_PKT_NUM) {
+		ret_val = HNS3_NIC_LB_TEST_TX_CNT_ERR;
+		netdev_err(ndev, "mode %d sent fail, cnt=0x%x, budget=0x%x\n",
+			   mode, good_cnt, HNS3_NIC_LB_TEST_PKT_NUM);
+		goto out;
+	}
+
+	/* Allow 200 milliseconds for packets to go from Tx to Rx */
+	msleep(200);
+
+	good_cnt = hns3_lb_check_rx_ring(priv, HNS3_NIC_LB_TEST_PKT_NUM);
+	if (good_cnt != HNS3_NIC_LB_TEST_PKT_NUM) {
+		ret_val = HNS3_NIC_LB_TEST_RX_CNT_ERR;
+		netdev_err(ndev, "mode %d recv fail, cnt=0x%x, budget=0x%x\n",
+			   mode, good_cnt, HNS3_NIC_LB_TEST_PKT_NUM);
+	}
+
+out:
+	hns3_lb_clear_tx_ring(priv, HNS3_NIC_LB_TEST_RING_ID,
+			      HNS3_NIC_LB_TEST_RING_ID);
+
+	kfree_skb(skb);
+	return ret_val;
+}
+
+static void hns3_set_selftest_param(struct hnae3_handle *h, int (*st_param)[2])
+{
+	st_param[HNAE3_LOOP_EXTERNAL][0] = HNAE3_LOOP_EXTERNAL;
+	st_param[HNAE3_LOOP_EXTERNAL][1] =
+			h->flags & HNAE3_SUPPORT_EXTERNAL_LOOPBACK;
+
+	st_param[HNAE3_LOOP_APP][0] = HNAE3_LOOP_APP;
+	st_param[HNAE3_LOOP_APP][1] =
+			h->flags & HNAE3_SUPPORT_APP_LOOPBACK;
+
+	st_param[HNAE3_LOOP_SERIAL_SERDES][0] = HNAE3_LOOP_SERIAL_SERDES;
+	st_param[HNAE3_LOOP_SERIAL_SERDES][1] =
+			h->flags & HNAE3_SUPPORT_SERDES_SERIAL_LOOPBACK;
+
+	st_param[HNAE3_LOOP_PARALLEL_SERDES][0] =
+			HNAE3_LOOP_PARALLEL_SERDES;
+	st_param[HNAE3_LOOP_PARALLEL_SERDES][1] =
+			h->flags & HNAE3_SUPPORT_SERDES_PARALLEL_LOOPBACK;
+
+	st_param[HNAE3_LOOP_PHY][0] = HNAE3_LOOP_PHY;
+	st_param[HNAE3_LOOP_PHY][1] =
+			h->flags & HNAE3_SUPPORT_PHY_LOOPBACK;
+}
+
+static void hns3_selftest_prepare(struct net_device *ndev, bool if_running)
+{
+	struct hns3_nic_priv *priv = netdev_priv(ndev);
+	struct hnae3_handle *h = priv->ae_handle;
+
+	if (if_running)
+		ndev->netdev_ops->ndo_stop(ndev);
+
+#if IS_ENABLED(CONFIG_VLAN_8021Q)
+	/* Disable the vlan filter for selftest does not support it */
+	if (h->ae_algo->ops->enable_vlan_filter &&
+	    ndev->features & NETIF_F_HW_VLAN_CTAG_FILTER)
+		h->ae_algo->ops->enable_vlan_filter(h, false);
+#endif
+
+	/* Tell firmware to stop mac autoneg before loopback test start,
+	 * otherwise loopback test may be failed when the port is still
+	 * negotiating.
+	 */
+	if (h->ae_algo->ops->halt_autoneg)
+		h->ae_algo->ops->halt_autoneg(h, true);
+
+	set_bit(HNS3_NIC_STATE_TESTING, &priv->state);
+}
+
+static void hns3_selftest_restore(struct net_device *ndev, bool if_running)
+{
+	struct hns3_nic_priv *priv = netdev_priv(ndev);
+	struct hnae3_handle *h = priv->ae_handle;
+
+	clear_bit(HNS3_NIC_STATE_TESTING, &priv->state);
+
+	if (h->ae_algo->ops->halt_autoneg)
+		h->ae_algo->ops->halt_autoneg(h, false);
+
+#if IS_ENABLED(CONFIG_VLAN_8021Q)
+	if (h->ae_algo->ops->enable_vlan_filter &&
+	    ndev->features & NETIF_F_HW_VLAN_CTAG_FILTER)
+		h->ae_algo->ops->enable_vlan_filter(h, true);
+#endif
+
+	if (if_running)
+		ndev->netdev_ops->ndo_open(ndev);
+}
+
+static void hns3_do_selftest(struct net_device *ndev, int (*st_param)[2],
+			     struct ethtool_test *eth_test, u64 *data)
+{
+	int test_index = HNAE3_LOOP_APP;
+	u32 i;
+
+	for (i = HNAE3_LOOP_APP; i < HNAE3_LOOP_NONE; i++) {
+		enum hnae3_loop loop_type = (enum hnae3_loop)st_param[i][0];
+
+		if (!st_param[i][1])
+			continue;
+
+		data[test_index] = hns3_lp_up(ndev, loop_type);
+		if (!data[test_index])
+			data[test_index] = hns3_lp_run_test(ndev, loop_type);
+
+		hns3_lp_down(ndev, loop_type);
+
+		if (data[test_index])
+			eth_test->flags |= ETH_TEST_FL_FAILED;
+
+		test_index++;
+	}
+}
+
+static void hns3_do_external_lb(struct net_device *ndev,
+				struct ethtool_test *eth_test, u64 *data)
+{
+	data[HNAE3_LOOP_EXTERNAL] = hns3_lp_up(ndev, HNAE3_LOOP_EXTERNAL);
+	if (!data[HNAE3_LOOP_EXTERNAL])
+		data[HNAE3_LOOP_EXTERNAL] = hns3_lp_run_test(ndev, HNAE3_LOOP_EXTERNAL);
+	hns3_lp_down(ndev, HNAE3_LOOP_EXTERNAL);
+
+	if (data[HNAE3_LOOP_EXTERNAL])
+		eth_test->flags |= ETH_TEST_FL_FAILED;
+
+	eth_test->flags |= ETH_TEST_FL_EXTERNAL_LB_DONE;
+}
+
+/**
+ * hns3_self_test - self test
+ * @ndev: net device
+ * @eth_test: test cmd
+ * @data: test result
+ */
+static void hns3_self_test(struct net_device *ndev,
+			   struct ethtool_test *eth_test, u64 *data)
+{
+	int cnt = hns3_get_sset_count(ndev, ETH_SS_TEST);
+	struct hns3_nic_priv *priv = netdev_priv(ndev);
+	struct hnae3_handle *h = priv->ae_handle;
+	int st_param[HNAE3_LOOP_NONE][2];
+	bool if_running = netif_running(ndev);
+	int i;
+
+	/* initialize the loopback test result, avoid marking an unexcuted
+	 * loopback test as PASS.
+	 */
+	for (i = 0; i < cnt; i++)
+		data[i] = HNS3_NIC_LB_TEST_UNEXECUTED;
+
+	if (hns3_nic_resetting(ndev)) {
+		netdev_err(ndev, "dev resetting!\n");
+		goto failure;
+	}
+
+	if (!(eth_test->flags & ETH_TEST_FL_OFFLINE))
+		goto failure;
+
+	if (netif_msg_ifdown(h))
+		netdev_info(ndev, "self test start\n");
+
+	hns3_set_selftest_param(h, st_param);
+
+	/* external loopback test requires that the link is up and the duplex is
+	 * full, do external test first to reduce the whole test time
+	 */
+	if (eth_test->flags & ETH_TEST_FL_EXTERNAL_LB) {
+		hns3_external_lb_prepare(ndev, if_running);
+		hns3_do_external_lb(ndev, eth_test, data);
+		hns3_external_lb_restore(ndev, if_running);
+	}
+
+	hns3_selftest_prepare(ndev, if_running);
+	hns3_do_selftest(ndev, st_param, eth_test, data);
+	hns3_selftest_restore(ndev, if_running);
+
+	if (netif_msg_ifdown(h))
+		netdev_info(ndev, "self test end\n");
+	return;
+
+failure:
+	eth_test->flags |= ETH_TEST_FL_FAILED;
+}
+
+static void hns3_update_limit_promisc_mode(struct net_device *netdev,
+					   bool enable)
+{
+	struct hnae3_handle *handle = hns3_get_handle(netdev);
+
+	if (enable)
+		set_bit(HNAE3_PFLAG_LIMIT_PROMISC, &handle->priv_flags);
+	else
+		clear_bit(HNAE3_PFLAG_LIMIT_PROMISC, &handle->priv_flags);
+
+	hns3_request_update_promisc_mode(handle);
+}
+
+static const struct hns3_pflag_desc hns3_priv_flags[HNAE3_PFLAG_MAX] = {
+	{ "limit_promisc",	hns3_update_limit_promisc_mode }
+};
+
+static int hns3_get_sset_count(struct net_device *netdev, int stringset)
+{
+	struct hnae3_handle *h = hns3_get_handle(netdev);
+	const struct hnae3_ae_ops *ops = hns3_get_ops(h);
+
+	if (!ops->get_sset_count)
+		return -EOPNOTSUPP;
+
+	switch (stringset) {
+	case ETH_SS_STATS:
+		return ((HNS3_TQP_STATS_COUNT * h->kinfo.num_tqps) +
+			ops->get_sset_count(h, stringset));
+
+	case ETH_SS_TEST:
+		return ops->get_sset_count(h, stringset);
+
+	case ETH_SS_PRIV_FLAGS:
+		return HNAE3_PFLAG_MAX;
+
+	default:
+		return -EOPNOTSUPP;
+	}
+}
+
+static void hns3_update_strings(u8 **data, const struct hns3_stats *stats,
+				u32 stat_count, u32 num_tqps,
+				const char *prefix)
+{
+	u32 i, j;
+
+	for (i = 0; i < num_tqps; i++)
+		for (j = 0; j < stat_count; j++)
+			ethtool_sprintf(data, "%s%u_%s", prefix, i,
+					stats[j].stats_string);
+}
+
+static void hns3_get_strings_tqps(struct hnae3_handle *handle, u8 **data)
+{
+	struct hnae3_knic_private_info *kinfo = &handle->kinfo;
+	const char tx_prefix[] = "txq";
+	const char rx_prefix[] = "rxq";
+
+	/* get strings for Tx */
+	hns3_update_strings(data, hns3_txq_stats, HNS3_TXQ_STATS_COUNT,
+			    kinfo->num_tqps, tx_prefix);
+
+	/* get strings for Rx */
+	hns3_update_strings(data, hns3_rxq_stats, HNS3_RXQ_STATS_COUNT,
+			    kinfo->num_tqps, rx_prefix);
+}
+
+static void hns3_get_strings(struct net_device *netdev, u32 stringset, u8 *data)
+{
+	struct hnae3_handle *h = hns3_get_handle(netdev);
+	const struct hnae3_ae_ops *ops = hns3_get_ops(h);
+	u32 i;
+
+	if (!ops->get_strings)
+		return;
+
+	switch (stringset) {
+	case ETH_SS_STATS:
+		hns3_get_strings_tqps(h, &data);
+		ops->get_strings(h, stringset, &data);
+		break;
+	case ETH_SS_TEST:
+		ops->get_strings(h, stringset, &data);
+		break;
+	case ETH_SS_PRIV_FLAGS:
+		for (i = 0; i < HNS3_PRIV_FLAGS_LEN; i++)
+			ethtool_puts(&data, hns3_priv_flags[i].name);
+		break;
+	default:
+		break;
+	}
+}
+
+static u64 *hns3_get_stats_tqps(struct hnae3_handle *handle, u64 *data)
+{
+	struct hnae3_knic_private_info *kinfo = &handle->kinfo;
+	struct hns3_nic_priv *nic_priv = handle->priv;
+	struct hns3_enet_ring *ring;
+	u8 *stat;
+	u32 i, j;
+
+	/* get stats for Tx */
+	for (i = 0; i < kinfo->num_tqps; i++) {
+		ring = &nic_priv->ring[i];
+		for (j = 0; j < HNS3_TXQ_STATS_COUNT; j++) {
+			stat = (u8 *)ring + hns3_txq_stats[j].stats_offset;
+			*data++ = *(u64 *)stat;
+		}
+	}
+
+	/* get stats for Rx */
+	for (i = 0; i < kinfo->num_tqps; i++) {
+		ring = &nic_priv->ring[i + kinfo->num_tqps];
+		for (j = 0; j < HNS3_RXQ_STATS_COUNT; j++) {
+			stat = (u8 *)ring + hns3_rxq_stats[j].stats_offset;
+			*data++ = *(u64 *)stat;
+		}
+	}
+
+	return data;
+}
+
+/* hns3_get_stats - get detail statistics.
+ * @netdev: net device
+ * @stats: statistics info.
+ * @data: statistics data.
+ */
+static void hns3_get_stats(struct net_device *netdev,
+			   struct ethtool_stats *stats, u64 *data)
+{
+	struct hnae3_handle *h = hns3_get_handle(netdev);
+	u64 *p = data;
+
+	if (hns3_nic_resetting(netdev)) {
+		netdev_err(netdev, "dev resetting, could not get stats\n");
+		return;
+	}
+
+	if (!h->ae_algo->ops->get_stats || !h->ae_algo->ops->update_stats) {
+		netdev_err(netdev, "could not get any statistics\n");
+		return;
+	}
+
+	h->ae_algo->ops->update_stats(h);
+
+	/* get per-queue stats */
+	p = hns3_get_stats_tqps(h, p);
+
+	/* get MAC & other misc hardware stats */
+	h->ae_algo->ops->get_stats(h, p);
+}
+
+static void hns3_get_drvinfo(struct net_device *netdev,
+			     struct ethtool_drvinfo *drvinfo)
+{
+	struct hns3_nic_priv *priv = netdev_priv(netdev);
+	struct hnae3_handle *h = priv->ae_handle;
+	u32 fw_version;
+
+	if (!h->ae_algo->ops->get_fw_version) {
+		netdev_err(netdev, "could not get fw version!\n");
+		return;
+	}
+
+	strscpy(drvinfo->driver, dev_driver_string(&h->pdev->dev),
+		sizeof(drvinfo->driver));
+
+	strscpy(drvinfo->bus_info, pci_name(h->pdev),
+		sizeof(drvinfo->bus_info));
+
+	fw_version = priv->ae_handle->ae_algo->ops->get_fw_version(h);
+
+	snprintf(drvinfo->fw_version, sizeof(drvinfo->fw_version),
+		 "%lu.%lu.%lu.%lu",
+		 hnae3_get_field(fw_version, HNAE3_FW_VERSION_BYTE3_MASK,
+				 HNAE3_FW_VERSION_BYTE3_SHIFT),
+		 hnae3_get_field(fw_version, HNAE3_FW_VERSION_BYTE2_MASK,
+				 HNAE3_FW_VERSION_BYTE2_SHIFT),
+		 hnae3_get_field(fw_version, HNAE3_FW_VERSION_BYTE1_MASK,
+				 HNAE3_FW_VERSION_BYTE1_SHIFT),
+		 hnae3_get_field(fw_version, HNAE3_FW_VERSION_BYTE0_MASK,
+				 HNAE3_FW_VERSION_BYTE0_SHIFT));
+}
+
+static u32 hns3_get_link(struct net_device *netdev)
+{
+	struct hnae3_handle *h = hns3_get_handle(netdev);
+
+	if (h->ae_algo->ops->get_status)
+		return h->ae_algo->ops->get_status(h);
+	else
+		return 0;
+}
+
+static void hns3_get_ringparam(struct net_device *netdev,
+			       struct ethtool_ringparam *param,
+			       struct kernel_ethtool_ringparam *kernel_param,
+			       struct netlink_ext_ack *extack)
+{
+	struct hns3_nic_priv *priv = netdev_priv(netdev);
+	struct hnae3_handle *h = priv->ae_handle;
+	int rx_queue_index = h->kinfo.num_tqps;
+
+	if (hns3_nic_resetting(netdev) || !priv->ring) {
+		netdev_err(netdev, "failed to get ringparam value, due to dev resetting or uninited\n");
+		return;
+	}
+
+	param->tx_max_pending = HNS3_RING_MAX_PENDING;
+	param->rx_max_pending = HNS3_RING_MAX_PENDING;
+
+	param->tx_pending = priv->ring[0].desc_num;
+	param->rx_pending = priv->ring[rx_queue_index].desc_num;
+	kernel_param->rx_buf_len = priv->ring[rx_queue_index].buf_size;
+	kernel_param->tx_push = test_bit(HNS3_NIC_STATE_TX_PUSH_ENABLE,
+					 &priv->state);
+}
+
+static void hns3_get_pauseparam(struct net_device *netdev,
+				struct ethtool_pauseparam *param)
+{
+	struct hnae3_handle *h = hns3_get_handle(netdev);
+	struct hnae3_ae_dev *ae_dev = hns3_get_ae_dev(h);
+
+	if (!test_bit(HNAE3_DEV_SUPPORT_PAUSE_B, ae_dev->caps))
+		return;
+
+	if (h->ae_algo->ops->get_pauseparam)
+		h->ae_algo->ops->get_pauseparam(h, &param->autoneg,
+			&param->rx_pause, &param->tx_pause);
+}
+
+static int hns3_set_pauseparam(struct net_device *netdev,
+			       struct ethtool_pauseparam *param)
+{
+	struct hnae3_handle *h = hns3_get_handle(netdev);
+	struct hnae3_ae_dev *ae_dev = hns3_get_ae_dev(h);
+
+	if (!test_bit(HNAE3_DEV_SUPPORT_PAUSE_B, ae_dev->caps))
+		return -EOPNOTSUPP;
+
+	netif_dbg(h, drv, netdev,
+		  "set pauseparam: autoneg=%u, rx:%u, tx:%u\n",
+		  param->autoneg, param->rx_pause, param->tx_pause);
+
+	if (h->ae_algo->ops->set_pauseparam)
+		return h->ae_algo->ops->set_pauseparam(h, param->autoneg,
+						       param->rx_pause,
+						       param->tx_pause);
+	return -EOPNOTSUPP;
+}
+
+static void hns3_get_ksettings(struct hnae3_handle *h,
+			       struct ethtool_link_ksettings *cmd)
+{
+	const struct hnae3_ae_ops *ops = hns3_get_ops(h);
+
+	/* 1.auto_neg & speed & duplex from cmd */
+	if (ops->get_ksettings_an_result)
+		ops->get_ksettings_an_result(h,
+					     &cmd->base.autoneg,
+					     &cmd->base.speed,
+					     &cmd->base.duplex,
+					     &cmd->lanes);
+
+	/* 2.get link mode */
+	if (ops->get_link_mode)
+		ops->get_link_mode(h,
+				   cmd->link_modes.supported,
+				   cmd->link_modes.advertising);
+
+	/* 3.mdix_ctrl&mdix get from phy reg */
+	if (ops->get_mdix_mode)
+		ops->get_mdix_mode(h, &cmd->base.eth_tp_mdix_ctrl,
+				   &cmd->base.eth_tp_mdix);
+}
+
+static int hns3_get_link_ksettings(struct net_device *netdev,
+				   struct ethtool_link_ksettings *cmd)
+{
+	struct hnae3_handle *h = hns3_get_handle(netdev);
+	struct hnae3_ae_dev *ae_dev = hns3_get_ae_dev(h);
+	const struct hnae3_ae_ops *ops;
+	u8 module_type;
+	u8 media_type;
+	u8 link_stat;
+
+	ops = h->ae_algo->ops;
+	if (ops->get_media_type)
+		ops->get_media_type(h, &media_type, &module_type);
+	else
+		return -EOPNOTSUPP;
+
+	switch (media_type) {
+	case HNAE3_MEDIA_TYPE_NONE:
+		cmd->base.port = PORT_NONE;
+		hns3_get_ksettings(h, cmd);
+		break;
+	case HNAE3_MEDIA_TYPE_FIBER:
+		if (module_type == HNAE3_MODULE_TYPE_UNKNOWN)
+			cmd->base.port = PORT_OTHER;
+		else if (module_type == HNAE3_MODULE_TYPE_CR)
+			cmd->base.port = PORT_DA;
+		else
+			cmd->base.port = PORT_FIBRE;
+
+		hns3_get_ksettings(h, cmd);
+		break;
+	case HNAE3_MEDIA_TYPE_BACKPLANE:
+		cmd->base.port = PORT_NONE;
+		hns3_get_ksettings(h, cmd);
+		break;
+	case HNAE3_MEDIA_TYPE_COPPER:
+		cmd->base.port = PORT_TP;
+		if (test_bit(HNAE3_DEV_SUPPORT_PHY_IMP_B, ae_dev->caps) &&
+		    ops->get_phy_link_ksettings)
+			ops->get_phy_link_ksettings(h, cmd);
+		else if (!netdev->phydev)
+			hns3_get_ksettings(h, cmd);
+		else
+			phy_ethtool_ksettings_get(netdev->phydev, cmd);
+		break;
+	default:
+
+		netdev_warn(netdev, "Unknown media type\n");
+		return 0;
+	}
+
+	/* mdio_support */
+	cmd->base.mdio_support = ETH_MDIO_SUPPORTS_C22;
+
+	link_stat = hns3_get_link(netdev);
+	if (!link_stat) {
+		cmd->base.speed = SPEED_UNKNOWN;
+		cmd->base.duplex = DUPLEX_UNKNOWN;
+	}
+
+	return 0;
+}
+
+static int hns3_check_ksettings_param(const struct net_device *netdev,
+				      const struct ethtool_link_ksettings *cmd)
+{
+	struct hnae3_handle *handle = hns3_get_handle(netdev);
+	const struct hnae3_ae_ops *ops = hns3_get_ops(handle);
+	u8 module_type = HNAE3_MODULE_TYPE_UNKNOWN;
+	u8 media_type = HNAE3_MEDIA_TYPE_UNKNOWN;
+	u32 lane_num;
+	u8 autoneg;
+	u32 speed;
+	u8 duplex;
+	int ret;
+
+	/* hw doesn't support use specified speed and duplex to negotiate,
+	 * unnecessary to check them when autoneg on.
+	 */
+	if (cmd->base.autoneg)
+		return 0;
+
+	if (ops->get_ksettings_an_result) {
+		ops->get_ksettings_an_result(handle, &autoneg, &speed, &duplex, &lane_num);
+		if (cmd->base.autoneg == autoneg && cmd->base.speed == speed &&
+		    cmd->base.duplex == duplex && cmd->lanes == lane_num)
+			return 0;
+	}
+
+	if (ops->get_media_type)
+		ops->get_media_type(handle, &media_type, &module_type);
+
+	if (cmd->base.duplex == DUPLEX_HALF &&
+	    media_type != HNAE3_MEDIA_TYPE_COPPER) {
+		netdev_err(netdev,
+			   "only copper port supports half duplex!\n");
+		return -EINVAL;
+	}
+
+	if (ops->check_port_speed) {
+		ret = ops->check_port_speed(handle, cmd->base.speed);
+		if (ret) {
+			netdev_err(netdev, "unsupported speed\n");
+			return ret;
+		}
+	}
+
+	return 0;
+}
+
+static int hns3_set_link_ksettings(struct net_device *netdev,
+				   const struct ethtool_link_ksettings *cmd)
+{
+	struct hnae3_handle *handle = hns3_get_handle(netdev);
+	struct hnae3_ae_dev *ae_dev = hns3_get_ae_dev(handle);
+	const struct hnae3_ae_ops *ops = hns3_get_ops(handle);
+	int ret;
+
+	/* Chip don't support this mode. */
+	if (cmd->base.speed == SPEED_1000 && cmd->base.duplex == DUPLEX_HALF)
+		return -EINVAL;
+
+	if (cmd->lanes && !hnae3_ae_dev_lane_num_supported(ae_dev))
+		return -EOPNOTSUPP;
+
+	netif_dbg(handle, drv, netdev,
+		  "set link(%s): autoneg=%u, speed=%u, duplex=%u, lanes=%u\n",
+		  netdev->phydev ? "phy" : "mac",
+		  cmd->base.autoneg, cmd->base.speed, cmd->base.duplex,
+		  cmd->lanes);
+
+	/* Only support ksettings_set for netdev with phy attached for now */
+	if (netdev->phydev) {
+		if (cmd->base.speed == SPEED_1000 &&
+		    cmd->base.autoneg == AUTONEG_DISABLE)
+			return -EINVAL;
+
+		return phy_ethtool_ksettings_set(netdev->phydev, cmd);
+	} else if (test_bit(HNAE3_DEV_SUPPORT_PHY_IMP_B, ae_dev->caps) &&
+		   ops->set_phy_link_ksettings) {
+		return ops->set_phy_link_ksettings(handle, cmd);
+	}
+
+	if (ae_dev->dev_version < HNAE3_DEVICE_VERSION_V2)
+		return -EOPNOTSUPP;
+
+	ret = hns3_check_ksettings_param(netdev, cmd);
+	if (ret)
+		return ret;
+
+	if (ops->set_autoneg) {
+		ret = ops->set_autoneg(handle, cmd->base.autoneg);
+		if (ret)
+			return ret;
+	}
+
+	/* hw doesn't support use specified speed and duplex to negotiate,
+	 * ignore them when autoneg on.
+	 */
+	if (cmd->base.autoneg) {
+		netdev_info(netdev,
+			    "autoneg is on, ignore the speed and duplex\n");
+		return 0;
+	}
+
+	if (ops->cfg_mac_speed_dup_h)
+		ret = ops->cfg_mac_speed_dup_h(handle, cmd->base.speed,
+					       cmd->base.duplex, (u8)(cmd->lanes));
+
+	return ret;
+}
+
+static u32 hns3_get_rss_key_size(struct net_device *netdev)
+{
+	struct hnae3_handle *h = hns3_get_handle(netdev);
+
+	if (!h->ae_algo->ops->get_rss_key_size)
+		return 0;
+
+	return h->ae_algo->ops->get_rss_key_size(h);
+}
+
+static u32 hns3_get_rss_indir_size(struct net_device *netdev)
+{
+	struct hnae3_handle *h = hns3_get_handle(netdev);
+	struct hnae3_ae_dev *ae_dev = hns3_get_ae_dev(h);
+
+	return ae_dev->dev_specs.rss_ind_tbl_size;
+}
+
+static int hns3_get_rss(struct net_device *netdev,
+			struct ethtool_rxfh_param *rxfh)
+{
+	struct hnae3_handle *h = hns3_get_handle(netdev);
+
+	if (!h->ae_algo->ops->get_rss)
+		return -EOPNOTSUPP;
+
+	return h->ae_algo->ops->get_rss(h, rxfh->indir, rxfh->key,
+					&rxfh->hfunc);
+}
+
+static int hns3_set_rss(struct net_device *netdev,
+			struct ethtool_rxfh_param *rxfh,
+			struct netlink_ext_ack *extack)
+{
+	struct hnae3_handle *h = hns3_get_handle(netdev);
+	struct hnae3_ae_dev *ae_dev = hns3_get_ae_dev(h);
+
+	if (!h->ae_algo->ops->set_rss)
+		return -EOPNOTSUPP;
+
+	if ((ae_dev->dev_version < HNAE3_DEVICE_VERSION_V2 &&
+	     rxfh->hfunc != ETH_RSS_HASH_TOP) ||
+	    (rxfh->hfunc != ETH_RSS_HASH_NO_CHANGE &&
+	     rxfh->hfunc != ETH_RSS_HASH_TOP &&
+	     rxfh->hfunc != ETH_RSS_HASH_XOR)) {
+		netdev_err(netdev, "hash func not supported\n");
+		return -EOPNOTSUPP;
+	}
+
+	if (!rxfh->indir) {
+		netdev_err(netdev,
+			   "set rss failed for indir is empty\n");
+		return -EOPNOTSUPP;
+	}
+
+	return h->ae_algo->ops->set_rss(h, rxfh->indir, rxfh->key,
+					rxfh->hfunc);
+}
+
+static int hns3_get_rxfh_fields(struct net_device *netdev,
+				struct ethtool_rxfh_fields *cmd)
+{
+	struct hnae3_handle *h = hns3_get_handle(netdev);
+
+	if (h->ae_algo->ops->get_rss_tuple)
+		return h->ae_algo->ops->get_rss_tuple(h, cmd);
+	return -EOPNOTSUPP;
+}
+
+static int hns3_get_rxnfc(struct net_device *netdev,
+			  struct ethtool_rxnfc *cmd,
+			  u32 *rule_locs)
+{
+	struct hnae3_handle *h = hns3_get_handle(netdev);
+
+	switch (cmd->cmd) {
+	case ETHTOOL_GRXRINGS:
+		cmd->data = h->kinfo.num_tqps;
+		return 0;
+	case ETHTOOL_GRXCLSRLCNT:
+		if (h->ae_algo->ops->get_fd_rule_cnt)
+			return h->ae_algo->ops->get_fd_rule_cnt(h, cmd);
+		return -EOPNOTSUPP;
+	case ETHTOOL_GRXCLSRULE:
+		if (h->ae_algo->ops->get_fd_rule_info)
+			return h->ae_algo->ops->get_fd_rule_info(h, cmd);
+		return -EOPNOTSUPP;
+	case ETHTOOL_GRXCLSRLALL:
+		if (h->ae_algo->ops->get_fd_all_rules)
+			return h->ae_algo->ops->get_fd_all_rules(h, cmd,
+								 rule_locs);
+		return -EOPNOTSUPP;
+	default:
+		return -EOPNOTSUPP;
+	}
+}
+
+static const struct hns3_reset_type_map hns3_reset_type[] = {
+	{ETH_RESET_MGMT, HNAE3_IMP_RESET},
+	{ETH_RESET_ALL, HNAE3_GLOBAL_RESET},
+	{ETH_RESET_DEDICATED, HNAE3_FUNC_RESET},
+};
+
+static const struct hns3_reset_type_map hns3vf_reset_type[] = {
+	{ETH_RESET_DEDICATED, HNAE3_VF_FUNC_RESET},
+};
+
+static int hns3_set_reset(struct net_device *netdev, u32 *flags)
+{
+	enum hnae3_reset_type rst_type = HNAE3_NONE_RESET;
+	struct hnae3_handle *h = hns3_get_handle(netdev);
+	struct hnae3_ae_dev *ae_dev = hns3_get_ae_dev(h);
+	const struct hnae3_ae_ops *ops = hns3_get_ops(h);
+	const struct hns3_reset_type_map *rst_type_map;
+	enum ethtool_reset_flags rst_flags;
+	u32 i, size;
+
+	if (ops->ae_dev_resetting && ops->ae_dev_resetting(h))
+		return -EBUSY;
+
+	if (!ops->set_default_reset_request || !ops->reset_event)
+		return -EOPNOTSUPP;
+
+	if (h->flags & HNAE3_SUPPORT_VF) {
+		rst_type_map = hns3vf_reset_type;
+		size = ARRAY_SIZE(hns3vf_reset_type);
+	} else {
+		rst_type_map = hns3_reset_type;
+		size = ARRAY_SIZE(hns3_reset_type);
+	}
+
+	for (i = 0; i < size; i++) {
+		if (rst_type_map[i].rst_flags == *flags) {
+			rst_type = rst_type_map[i].rst_type;
+			rst_flags = rst_type_map[i].rst_flags;
+			break;
+		}
+	}
+
+	if (rst_type == HNAE3_NONE_RESET ||
+	    (rst_type == HNAE3_IMP_RESET &&
+	     ae_dev->dev_version <= HNAE3_DEVICE_VERSION_V2))
+		return -EOPNOTSUPP;
+
+	netdev_info(netdev, "Setting reset type %d\n", rst_type);
+
+	ops->set_default_reset_request(ae_dev, rst_type);
+
+	ops->reset_event(h->pdev, h);
+
+	*flags &= ~rst_flags;
+
+	return 0;
+}
+
+static void hns3_change_all_ring_bd_num(struct hns3_nic_priv *priv,
+					u32 tx_desc_num, u32 rx_desc_num)
+{
+	struct hnae3_handle *h = priv->ae_handle;
+	int i;
+
+	h->kinfo.num_tx_desc = tx_desc_num;
+	h->kinfo.num_rx_desc = rx_desc_num;
+
+	for (i = 0; i < h->kinfo.num_tqps; i++) {
+		priv->ring[i].desc_num = tx_desc_num;
+		priv->ring[i + h->kinfo.num_tqps].desc_num = rx_desc_num;
+	}
+}
+
+static struct hns3_enet_ring *hns3_backup_ringparam(struct hns3_nic_priv *priv)
+{
+	struct hnae3_handle *handle = priv->ae_handle;
+	struct hns3_enet_ring *tmp_rings;
+	int i;
+
+	tmp_rings = kcalloc(handle->kinfo.num_tqps * 2,
+			    sizeof(struct hns3_enet_ring), GFP_KERNEL);
+	if (!tmp_rings)
+		return NULL;
+
+	for (i = 0; i < handle->kinfo.num_tqps * 2; i++) {
+		memcpy(&tmp_rings[i], &priv->ring[i],
+		       sizeof(struct hns3_enet_ring));
+		tmp_rings[i].skb = NULL;
+	}
+
+	return tmp_rings;
+}
+
+static int hns3_check_ringparam(struct net_device *ndev,
+				struct ethtool_ringparam *param,
+				struct kernel_ethtool_ringparam *kernel_param)
+{
+#define RX_BUF_LEN_2K 2048
+#define RX_BUF_LEN_4K 4096
+
+	struct hns3_nic_priv *priv = netdev_priv(ndev);
+
+	if (hns3_nic_resetting(ndev) || !priv->ring) {
+		netdev_err(ndev, "failed to set ringparam value, due to dev resetting or uninited\n");
+		return -EBUSY;
+	}
+
+
+	if (param->rx_mini_pending || param->rx_jumbo_pending)
+		return -EINVAL;
+
+	if (kernel_param->rx_buf_len != RX_BUF_LEN_2K &&
+	    kernel_param->rx_buf_len != RX_BUF_LEN_4K) {
+		netdev_err(ndev, "Rx buf len only support 2048 and 4096\n");
+		return -EINVAL;
+	}
+
+	if (param->tx_pending > HNS3_RING_MAX_PENDING ||
+	    param->tx_pending < HNS3_RING_MIN_PENDING ||
+	    param->rx_pending > HNS3_RING_MAX_PENDING ||
+	    param->rx_pending < HNS3_RING_MIN_PENDING) {
+		netdev_err(ndev, "Queue depth out of range [%d-%d]\n",
+			   HNS3_RING_MIN_PENDING, HNS3_RING_MAX_PENDING);
+		return -EINVAL;
+	}
+
+	return 0;
+}
+
+static bool
+hns3_is_ringparam_changed(struct net_device *ndev,
+			  struct ethtool_ringparam *param,
+			  struct kernel_ethtool_ringparam *kernel_param,
+			  struct hns3_ring_param *old_ringparam,
+			  struct hns3_ring_param *new_ringparam)
+{
+	struct hns3_nic_priv *priv = netdev_priv(ndev);
+	struct hnae3_handle *h = priv->ae_handle;
+	u16 queue_num = h->kinfo.num_tqps;
+
+	new_ringparam->tx_desc_num = ALIGN(param->tx_pending,
+					   HNS3_RING_BD_MULTIPLE);
+	new_ringparam->rx_desc_num = ALIGN(param->rx_pending,
+					   HNS3_RING_BD_MULTIPLE);
+	old_ringparam->tx_desc_num = priv->ring[0].desc_num;
+	old_ringparam->rx_desc_num = priv->ring[queue_num].desc_num;
+	old_ringparam->rx_buf_len = priv->ring[queue_num].buf_size;
+	new_ringparam->rx_buf_len = kernel_param->rx_buf_len;
+
+	if (old_ringparam->tx_desc_num == new_ringparam->tx_desc_num &&
+	    old_ringparam->rx_desc_num == new_ringparam->rx_desc_num &&
+	    old_ringparam->rx_buf_len == new_ringparam->rx_buf_len) {
+		netdev_info(ndev, "descriptor number and rx buffer length not changed\n");
+		return false;
+	}
+
+	return true;
+}
+
+static int hns3_change_rx_buf_len(struct net_device *ndev, u32 rx_buf_len)
+{
+	struct hns3_nic_priv *priv = netdev_priv(ndev);
+	struct hnae3_handle *h = priv->ae_handle;
+	int i;
+
+	h->kinfo.rx_buf_len = rx_buf_len;
+
+	for (i = 0; i < h->kinfo.num_tqps; i++) {
+		h->kinfo.tqp[i]->buf_size = rx_buf_len;
+		priv->ring[i + h->kinfo.num_tqps].buf_size = rx_buf_len;
+	}
+
+	return 0;
+}
+
+static int hns3_set_tx_push(struct net_device *netdev, u32 tx_push)
+{
+	struct hns3_nic_priv *priv = netdev_priv(netdev);
+	struct hnae3_handle *h = hns3_get_handle(netdev);
+	struct hnae3_ae_dev *ae_dev = hns3_get_ae_dev(h);
+	u32 old_state = test_bit(HNS3_NIC_STATE_TX_PUSH_ENABLE, &priv->state);
+
+	if (!test_bit(HNAE3_DEV_SUPPORT_TX_PUSH_B, ae_dev->caps) && tx_push)
+		return -EOPNOTSUPP;
+
+	if (tx_push == old_state)
+		return 0;
+
+	netdev_dbg(netdev, "Changing tx push from %s to %s\n",
+		   str_on_off(old_state), str_on_off(tx_push));
+
+	if (tx_push)
+		set_bit(HNS3_NIC_STATE_TX_PUSH_ENABLE, &priv->state);
+	else
+		clear_bit(HNS3_NIC_STATE_TX_PUSH_ENABLE, &priv->state);
+
+	return 0;
+}
+
+static int hns3_set_ringparam(struct net_device *ndev,
+			      struct ethtool_ringparam *param,
+			      struct kernel_ethtool_ringparam *kernel_param,
+			      struct netlink_ext_ack *extack)
+{
+	struct hns3_ring_param old_ringparam, new_ringparam;
+	struct hns3_nic_priv *priv = netdev_priv(ndev);
+	struct hnae3_handle *h = priv->ae_handle;
+	struct hns3_enet_ring *tmp_rings;
+	bool if_running = netif_running(ndev);
+	int ret, i;
+
+	ret = hns3_check_ringparam(ndev, param, kernel_param);
+	if (ret)
+		return ret;
+
+	ret = hns3_set_tx_push(ndev, kernel_param->tx_push);
+	if (ret)
+		return ret;
+
+	if (!hns3_is_ringparam_changed(ndev, param, kernel_param,
+				       &old_ringparam, &new_ringparam))
+		return 0;
+
+	tmp_rings = hns3_backup_ringparam(priv);
+	if (!tmp_rings) {
+		netdev_err(ndev, "backup ring param failed by allocating memory fail\n");
+		return -ENOMEM;
+	}
+
+	netdev_info(ndev,
+		    "Changing Tx/Rx ring depth from %u/%u to %u/%u, Changing rx buffer len from %u to %u\n",
+		    old_ringparam.tx_desc_num, old_ringparam.rx_desc_num,
+		    new_ringparam.tx_desc_num, new_ringparam.rx_desc_num,
+		    old_ringparam.rx_buf_len, new_ringparam.rx_buf_len);
+
+	if (if_running)
+		ndev->netdev_ops->ndo_stop(ndev);
+
+	hns3_change_all_ring_bd_num(priv, new_ringparam.tx_desc_num,
+				    new_ringparam.rx_desc_num);
+	hns3_change_rx_buf_len(ndev, new_ringparam.rx_buf_len);
+	ret = hns3_init_all_ring(priv);
+	if (ret) {
+		netdev_err(ndev, "set ringparam fail, revert to old value(%d)\n",
+			   ret);
+
+		hns3_change_rx_buf_len(ndev, old_ringparam.rx_buf_len);
+		hns3_change_all_ring_bd_num(priv, old_ringparam.tx_desc_num,
+					    old_ringparam.rx_desc_num);
+		for (i = 0; i < h->kinfo.num_tqps * 2; i++)
+			memcpy(&priv->ring[i], &tmp_rings[i],
+			       sizeof(struct hns3_enet_ring));
+	} else {
+		for (i = 0; i < h->kinfo.num_tqps * 2; i++)
+			hns3_fini_ring(&tmp_rings[i]);
+	}
+
+	kfree(tmp_rings);
+
+	if (if_running)
+		ret = ndev->netdev_ops->ndo_open(ndev);
+
+	return ret;
+}
+
+static int hns3_set_rxfh_fields(struct net_device *netdev,
+				const struct ethtool_rxfh_fields *cmd,
+				struct netlink_ext_ack *extack)
+{
+	struct hnae3_handle *h = hns3_get_handle(netdev);
+
+	if (h->ae_algo->ops->set_rss_tuple)
+		return h->ae_algo->ops->set_rss_tuple(h, cmd);
+	return -EOPNOTSUPP;
+}
+
+static int hns3_set_rxnfc(struct net_device *netdev, struct ethtool_rxnfc *cmd)
+{
+	struct hnae3_handle *h = hns3_get_handle(netdev);
+
+	switch (cmd->cmd) {
+	case ETHTOOL_SRXCLSRLINS:
+		if (h->ae_algo->ops->add_fd_entry)
+			return h->ae_algo->ops->add_fd_entry(h, cmd);
+		return -EOPNOTSUPP;
+	case ETHTOOL_SRXCLSRLDEL:
+		if (h->ae_algo->ops->del_fd_entry)
+			return h->ae_algo->ops->del_fd_entry(h, cmd);
+		return -EOPNOTSUPP;
+	default:
+		return -EOPNOTSUPP;
+	}
+}
+
+static int hns3_nway_reset(struct net_device *netdev)
+{
+	struct hnae3_handle *handle = hns3_get_handle(netdev);
+	const struct hnae3_ae_ops *ops = hns3_get_ops(handle);
+	struct phy_device *phy = netdev->phydev;
+	int autoneg;
+
+	if (!netif_running(netdev))
+		return 0;
+
+	if (hns3_nic_resetting(netdev)) {
+		netdev_err(netdev, "dev resetting!\n");
+		return -EBUSY;
+	}
+
+	if (!ops->get_autoneg || !ops->restart_autoneg)
+		return -EOPNOTSUPP;
+
+	autoneg = ops->get_autoneg(handle);
+	if (autoneg != AUTONEG_ENABLE) {
+		netdev_err(netdev,
+			   "Autoneg is off, don't support to restart it\n");
+		return -EINVAL;
+	}
+
+	netif_dbg(handle, drv, netdev,
+		  "nway reset (using %s)\n", phy ? "phy" : "mac");
+
+	if (phy)
+		return genphy_restart_aneg(phy);
+
+	return ops->restart_autoneg(handle);
+}
+
+static void hns3_get_channels(struct net_device *netdev,
+			      struct ethtool_channels *ch)
+{
+	struct hnae3_handle *h = hns3_get_handle(netdev);
+
+	if (h->ae_algo->ops->get_channels)
+		h->ae_algo->ops->get_channels(h, ch);
+}
+
+static int hns3_get_coalesce(struct net_device *netdev,
+			     struct ethtool_coalesce *cmd,
+			     struct kernel_ethtool_coalesce *kernel_coal,
+			     struct netlink_ext_ack *extack)
+{
+	struct hns3_nic_priv *priv = netdev_priv(netdev);
+	struct hns3_enet_coalesce *tx_coal = &priv->tx_coal;
+	struct hns3_enet_coalesce *rx_coal = &priv->rx_coal;
+	struct hnae3_handle *h = priv->ae_handle;
+
+	if (hns3_nic_resetting(netdev))
+		return -EBUSY;
+
+	cmd->use_adaptive_tx_coalesce = tx_coal->adapt_enable;
+	cmd->use_adaptive_rx_coalesce = rx_coal->adapt_enable;
+
+	cmd->tx_coalesce_usecs = tx_coal->int_gl;
+	cmd->rx_coalesce_usecs = rx_coal->int_gl;
+
+	cmd->tx_coalesce_usecs_high = h->kinfo.int_rl_setting;
+	cmd->rx_coalesce_usecs_high = h->kinfo.int_rl_setting;
+
+	cmd->tx_max_coalesced_frames = tx_coal->int_ql;
+	cmd->rx_max_coalesced_frames = rx_coal->int_ql;
+
+	kernel_coal->use_cqe_mode_tx = (priv->tx_cqe_mode ==
+					DIM_CQ_PERIOD_MODE_START_FROM_CQE);
+	kernel_coal->use_cqe_mode_rx = (priv->rx_cqe_mode ==
+					DIM_CQ_PERIOD_MODE_START_FROM_CQE);
+
+	return 0;
+}
+
+static int hns3_check_gl_coalesce_para(struct net_device *netdev,
+				       struct ethtool_coalesce *cmd)
+{
+	struct hnae3_handle *handle = hns3_get_handle(netdev);
+	struct hnae3_ae_dev *ae_dev = hns3_get_ae_dev(handle);
+	u32 rx_gl, tx_gl;
+
+	if (cmd->rx_coalesce_usecs > ae_dev->dev_specs.max_int_gl) {
+		netdev_err(netdev,
+			   "invalid rx-usecs value, rx-usecs range is 0-%u\n",
+			   ae_dev->dev_specs.max_int_gl);
+		return -EINVAL;
+	}
+
+	if (cmd->tx_coalesce_usecs > ae_dev->dev_specs.max_int_gl) {
+		netdev_err(netdev,
+			   "invalid tx-usecs value, tx-usecs range is 0-%u\n",
+			   ae_dev->dev_specs.max_int_gl);
+		return -EINVAL;
+	}
+
+	/* device version above V3(include V3), GL uses 1us unit,
+	 * so the round down is not needed.
+	 */
+	if (ae_dev->dev_version >= HNAE3_DEVICE_VERSION_V3)
+		return 0;
+
+	rx_gl = hns3_gl_round_down(cmd->rx_coalesce_usecs);
+	if (rx_gl != cmd->rx_coalesce_usecs) {
+		netdev_info(netdev,
+			    "rx_usecs(%u) rounded down to %u, because it must be multiple of 2.\n",
+			    cmd->rx_coalesce_usecs, rx_gl);
+	}
+
+	tx_gl = hns3_gl_round_down(cmd->tx_coalesce_usecs);
+	if (tx_gl != cmd->tx_coalesce_usecs) {
+		netdev_info(netdev,
+			    "tx_usecs(%u) rounded down to %u, because it must be multiple of 2.\n",
+			    cmd->tx_coalesce_usecs, tx_gl);
+	}
+
+	return 0;
+}
+
+static int hns3_check_rl_coalesce_para(struct net_device *netdev,
+				       struct ethtool_coalesce *cmd)
+{
+	u32 rl;
+
+	if (cmd->tx_coalesce_usecs_high != cmd->rx_coalesce_usecs_high) {
+		netdev_err(netdev,
+			   "tx_usecs_high must be same as rx_usecs_high.\n");
+		return -EINVAL;
+	}
+
+	if (cmd->rx_coalesce_usecs_high > HNS3_INT_RL_MAX) {
+		netdev_err(netdev,
+			   "Invalid usecs_high value, usecs_high range is 0-%d\n",
+			   HNS3_INT_RL_MAX);
+		return -EINVAL;
+	}
+
+	rl = hns3_rl_round_down(cmd->rx_coalesce_usecs_high);
+	if (rl != cmd->rx_coalesce_usecs_high) {
+		netdev_info(netdev,
+			    "usecs_high(%u) rounded down to %u, because it must be multiple of 4.\n",
+			    cmd->rx_coalesce_usecs_high, rl);
+	}
+
+	return 0;
+}
+
+static int hns3_check_ql_coalesce_param(struct net_device *netdev,
+					struct ethtool_coalesce *cmd)
+{
+	struct hnae3_handle *handle = hns3_get_handle(netdev);
+	struct hnae3_ae_dev *ae_dev = hns3_get_ae_dev(handle);
+
+	if ((cmd->tx_max_coalesced_frames || cmd->rx_max_coalesced_frames) &&
+	    !ae_dev->dev_specs.int_ql_max) {
+		netdev_err(netdev, "coalesced frames is not supported\n");
+		return -EOPNOTSUPP;
+	}
+
+	if (cmd->tx_max_coalesced_frames > ae_dev->dev_specs.int_ql_max ||
+	    cmd->rx_max_coalesced_frames > ae_dev->dev_specs.int_ql_max) {
+		netdev_err(netdev,
+			   "invalid coalesced_frames value, range is 0-%u\n",
+			   ae_dev->dev_specs.int_ql_max);
+		return -ERANGE;
+	}
+
+	return 0;
+}
+
+static int
+hns3_check_cqe_coalesce_param(struct net_device *netdev,
+			      struct kernel_ethtool_coalesce *kernel_coal)
+{
+	struct hnae3_handle *handle = hns3_get_handle(netdev);
+	struct hnae3_ae_dev *ae_dev = hns3_get_ae_dev(handle);
+
+	if ((kernel_coal->use_cqe_mode_tx || kernel_coal->use_cqe_mode_rx) &&
+	    !hnae3_ae_dev_cq_supported(ae_dev)) {
+		netdev_err(netdev, "coalesced cqe mode is not supported\n");
+		return -EOPNOTSUPP;
+	}
+
+	return 0;
+}
+
+static int
+hns3_check_coalesce_para(struct net_device *netdev,
+			 struct ethtool_coalesce *cmd,
+			 struct kernel_ethtool_coalesce *kernel_coal)
+{
+	int ret;
+
+	ret = hns3_check_cqe_coalesce_param(netdev, kernel_coal);
+	if (ret)
+		return ret;
+
+	ret = hns3_check_gl_coalesce_para(netdev, cmd);
+	if (ret) {
+		netdev_err(netdev,
+			   "Check gl coalesce param fail. ret = %d\n", ret);
+		return ret;
+	}
+
+	ret = hns3_check_rl_coalesce_para(netdev, cmd);
+	if (ret) {
+		netdev_err(netdev,
+			   "Check rl coalesce param fail. ret = %d\n", ret);
+		return ret;
+	}
+
+	return hns3_check_ql_coalesce_param(netdev, cmd);
+}
+
+static void hns3_set_coalesce_per_queue(struct net_device *netdev,
+					struct ethtool_coalesce *cmd,
+					u32 queue)
+{
+	struct hns3_enet_tqp_vector *tx_vector, *rx_vector;
+	struct hns3_nic_priv *priv = netdev_priv(netdev);
+	struct hnae3_handle *h = priv->ae_handle;
+	int queue_num = h->kinfo.num_tqps;
+
+	tx_vector = priv->ring[queue].tqp_vector;
+	rx_vector = priv->ring[queue_num + queue].tqp_vector;
+
+	tx_vector->tx_group.coal.adapt_enable =
+				cmd->use_adaptive_tx_coalesce;
+	rx_vector->rx_group.coal.adapt_enable =
+				cmd->use_adaptive_rx_coalesce;
+
+	tx_vector->tx_group.coal.int_gl = cmd->tx_coalesce_usecs;
+	rx_vector->rx_group.coal.int_gl = cmd->rx_coalesce_usecs;
+
+	tx_vector->tx_group.coal.int_ql = cmd->tx_max_coalesced_frames;
+	rx_vector->rx_group.coal.int_ql = cmd->rx_max_coalesced_frames;
+
+	hns3_set_vector_coalesce_tx_gl(tx_vector,
+				       tx_vector->tx_group.coal.int_gl);
+	hns3_set_vector_coalesce_rx_gl(rx_vector,
+				       rx_vector->rx_group.coal.int_gl);
+
+	hns3_set_vector_coalesce_rl(tx_vector, h->kinfo.int_rl_setting);
+	hns3_set_vector_coalesce_rl(rx_vector, h->kinfo.int_rl_setting);
+
+	if (tx_vector->tx_group.coal.ql_enable)
+		hns3_set_vector_coalesce_tx_ql(tx_vector,
+					       tx_vector->tx_group.coal.int_ql);
+	if (rx_vector->rx_group.coal.ql_enable)
+		hns3_set_vector_coalesce_rx_ql(rx_vector,
+					       rx_vector->rx_group.coal.int_ql);
+}
+
+static int hns3_set_coalesce(struct net_device *netdev,
+			     struct ethtool_coalesce *cmd,
+			     struct kernel_ethtool_coalesce *kernel_coal,
+			     struct netlink_ext_ack *extack)
+{
+	struct hnae3_handle *h = hns3_get_handle(netdev);
+	struct hns3_nic_priv *priv = netdev_priv(netdev);
+	struct hns3_enet_coalesce *tx_coal = &priv->tx_coal;
+	struct hns3_enet_coalesce *rx_coal = &priv->rx_coal;
+	u16 queue_num = h->kinfo.num_tqps;
+	enum dim_cq_period_mode tx_mode;
+	enum dim_cq_period_mode rx_mode;
+	int ret;
+	int i;
+
+	if (hns3_nic_resetting(netdev))
+		return -EBUSY;
+
+	ret = hns3_check_coalesce_para(netdev, cmd, kernel_coal);
+	if (ret)
+		return ret;
+
+	h->kinfo.int_rl_setting =
+		hns3_rl_round_down(cmd->rx_coalesce_usecs_high);
+
+	tx_coal->adapt_enable = cmd->use_adaptive_tx_coalesce;
+	rx_coal->adapt_enable = cmd->use_adaptive_rx_coalesce;
+
+	tx_coal->int_gl = cmd->tx_coalesce_usecs;
+	rx_coal->int_gl = cmd->rx_coalesce_usecs;
+
+	tx_coal->int_ql = cmd->tx_max_coalesced_frames;
+	rx_coal->int_ql = cmd->rx_max_coalesced_frames;
+
+	for (i = 0; i < queue_num; i++)
+		hns3_set_coalesce_per_queue(netdev, cmd, i);
+
+	tx_mode = kernel_coal->use_cqe_mode_tx ?
+		  DIM_CQ_PERIOD_MODE_START_FROM_CQE :
+		  DIM_CQ_PERIOD_MODE_START_FROM_EQE;
+	rx_mode = kernel_coal->use_cqe_mode_rx ?
+		  DIM_CQ_PERIOD_MODE_START_FROM_CQE :
+		  DIM_CQ_PERIOD_MODE_START_FROM_EQE;
+	hns3_cq_period_mode_init(priv, tx_mode, rx_mode);
+
+	return 0;
+}
+
+static int hns3_get_regs_len(struct net_device *netdev)
+{
+	struct hnae3_handle *h = hns3_get_handle(netdev);
+
+	if (!h->ae_algo->ops->get_regs_len)
+		return -EOPNOTSUPP;
+
+	return h->ae_algo->ops->get_regs_len(h);
+}
+
+static void hns3_get_regs(struct net_device *netdev,
+			  struct ethtool_regs *cmd, void *data)
+{
+	struct hnae3_handle *h = hns3_get_handle(netdev);
+
+	if (!h->ae_algo->ops->get_regs)
+		return;
+
+	h->ae_algo->ops->get_regs(h, &cmd->version, data);
+}
+
+static int hns3_set_phys_id(struct net_device *netdev,
+			    enum ethtool_phys_id_state state)
+{
+	struct hnae3_handle *h = hns3_get_handle(netdev);
+
+	if (!h->ae_algo->ops->set_led_id)
+		return -EOPNOTSUPP;
+
+	return h->ae_algo->ops->set_led_id(h, state);
+}
+
+static u32 hns3_get_msglevel(struct net_device *netdev)
+{
+	struct hnae3_handle *h = hns3_get_handle(netdev);
+
+	return h->msg_enable;
+}
+
+static void hns3_set_msglevel(struct net_device *netdev, u32 msg_level)
+{
+	struct hnae3_handle *h = hns3_get_handle(netdev);
+
+	h->msg_enable = msg_level;
+}
+
+static void hns3_get_fec_stats(struct net_device *netdev,
+			       struct ethtool_fec_stats *fec_stats,
+			       struct ethtool_fec_hist *hist)
+{
+	struct hnae3_handle *handle = hns3_get_handle(netdev);
+	struct hnae3_ae_dev *ae_dev = hns3_get_ae_dev(handle);
+	const struct hnae3_ae_ops *ops = hns3_get_ops(handle);
+
+	if (!hnae3_ae_dev_fec_stats_supported(ae_dev) || !ops->get_fec_stats)
+		return;
+
+	ops->get_fec_stats(handle, fec_stats);
+}
+
+/* Translate local fec value into ethtool value. */
+static unsigned int loc_to_eth_fec(u8 loc_fec)
+{
+	u32 eth_fec = 0;
+
+	if (loc_fec & BIT(HNAE3_FEC_AUTO))
+		eth_fec |= ETHTOOL_FEC_AUTO;
+	if (loc_fec & BIT(HNAE3_FEC_RS))
+		eth_fec |= ETHTOOL_FEC_RS;
+	if (loc_fec & BIT(HNAE3_FEC_LLRS))
+		eth_fec |= ETHTOOL_FEC_LLRS;
+	if (loc_fec & BIT(HNAE3_FEC_BASER))
+		eth_fec |= ETHTOOL_FEC_BASER;
+	if (loc_fec & BIT(HNAE3_FEC_NONE))
+		eth_fec |= ETHTOOL_FEC_OFF;
+
+	return eth_fec;
+}
+
+/* Translate ethtool fec value into local value. */
+static unsigned int eth_to_loc_fec(unsigned int eth_fec)
+{
+	u32 loc_fec = 0;
+
+	if (eth_fec & ETHTOOL_FEC_OFF)
+		loc_fec |= BIT(HNAE3_FEC_NONE);
+	if (eth_fec & ETHTOOL_FEC_AUTO)
+		loc_fec |= BIT(HNAE3_FEC_AUTO);
+	if (eth_fec & ETHTOOL_FEC_RS)
+		loc_fec |= BIT(HNAE3_FEC_RS);
+	if (eth_fec & ETHTOOL_FEC_LLRS)
+		loc_fec |= BIT(HNAE3_FEC_LLRS);
+	if (eth_fec & ETHTOOL_FEC_BASER)
+		loc_fec |= BIT(HNAE3_FEC_BASER);
+
+	return loc_fec;
+}
+
+static int hns3_get_fecparam(struct net_device *netdev,
+			     struct ethtool_fecparam *fec)
+{
+	struct hnae3_handle *handle = hns3_get_handle(netdev);
+	struct hnae3_ae_dev *ae_dev = hns3_get_ae_dev(handle);
+	const struct hnae3_ae_ops *ops = hns3_get_ops(handle);
+	u8 fec_ability;
+	u8 fec_mode;
+
+	if (!test_bit(HNAE3_DEV_SUPPORT_FEC_B, ae_dev->caps))
+		return -EOPNOTSUPP;
+
+	if (!ops->get_fec)
+		return -EOPNOTSUPP;
+
+	ops->get_fec(handle, &fec_ability, &fec_mode);
+
+	fec->fec = loc_to_eth_fec(fec_ability);
+	fec->active_fec = loc_to_eth_fec(fec_mode);
+	if (!fec->active_fec)
+		fec->active_fec = ETHTOOL_FEC_OFF;
+
+	return 0;
+}
+
+static int hns3_set_fecparam(struct net_device *netdev,
+			     struct ethtool_fecparam *fec)
+{
+	struct hnae3_handle *handle = hns3_get_handle(netdev);
+	struct hnae3_ae_dev *ae_dev = hns3_get_ae_dev(handle);
+	const struct hnae3_ae_ops *ops = hns3_get_ops(handle);
+	u32 fec_mode;
+
+	if (!test_bit(HNAE3_DEV_SUPPORT_FEC_B, ae_dev->caps))
+		return -EOPNOTSUPP;
+
+	if (!ops->set_fec)
+		return -EOPNOTSUPP;
+	fec_mode = eth_to_loc_fec(fec->fec);
+
+	netif_dbg(handle, drv, netdev, "set fecparam: mode=%u\n", fec_mode);
+
+	return ops->set_fec(handle, fec_mode);
+}
+
+static int hns3_get_module_info(struct net_device *netdev,
+				struct ethtool_modinfo *modinfo)
+{
+#define HNS3_SFF_8636_V1_3 0x03
+
+	struct hnae3_handle *handle = hns3_get_handle(netdev);
+	struct hnae3_ae_dev *ae_dev = hns3_get_ae_dev(handle);
+	const struct hnae3_ae_ops *ops = hns3_get_ops(handle);
+	struct hns3_sfp_type sfp_type;
+	int ret;
+
+	if (ae_dev->dev_version < HNAE3_DEVICE_VERSION_V2 ||
+	    !ops->get_module_eeprom)
+		return -EOPNOTSUPP;
+
+	memset(&sfp_type, 0, sizeof(sfp_type));
+	ret = ops->get_module_eeprom(handle, 0, sizeof(sfp_type) / sizeof(u8),
+				     (u8 *)&sfp_type);
+	if (ret)
+		return ret;
+
+	switch (sfp_type.type) {
+	case SFF8024_ID_SFP:
+		modinfo->type = ETH_MODULE_SFF_8472;
+		modinfo->eeprom_len = ETH_MODULE_SFF_8472_LEN;
+		break;
+	case SFF8024_ID_QSFP_8438:
+		modinfo->type = ETH_MODULE_SFF_8436;
+		modinfo->eeprom_len = ETH_MODULE_SFF_8436_MAX_LEN;
+		break;
+	case SFF8024_ID_QSFP_8436_8636:
+		if (sfp_type.ext_type < HNS3_SFF_8636_V1_3) {
+			modinfo->type = ETH_MODULE_SFF_8436;
+			modinfo->eeprom_len = ETH_MODULE_SFF_8436_MAX_LEN;
+		} else {
+			modinfo->type = ETH_MODULE_SFF_8636;
+			modinfo->eeprom_len = ETH_MODULE_SFF_8636_MAX_LEN;
+		}
+		break;
+	case SFF8024_ID_QSFP28_8636:
+		modinfo->type = ETH_MODULE_SFF_8636;
+		modinfo->eeprom_len = ETH_MODULE_SFF_8636_MAX_LEN;
+		break;
+	default:
+		netdev_err(netdev, "Optical module unknown: %#x\n",
+			   sfp_type.type);
+		return -EINVAL;
+	}
+
+	return 0;
+}
+
+static int hns3_get_module_eeprom(struct net_device *netdev,
+				  struct ethtool_eeprom *ee, u8 *data)
+{
+	struct hnae3_handle *handle = hns3_get_handle(netdev);
+	struct hnae3_ae_dev *ae_dev = hns3_get_ae_dev(handle);
+	const struct hnae3_ae_ops *ops = hns3_get_ops(handle);
+
+	if (ae_dev->dev_version < HNAE3_DEVICE_VERSION_V2 ||
+	    !ops->get_module_eeprom)
+		return -EOPNOTSUPP;
+
+	if (!ee->len)
+		return -EINVAL;
+
+	memset(data, 0, ee->len);
+
+	return ops->get_module_eeprom(handle, ee->offset, ee->len, data);
+}
+
+static u32 hns3_get_priv_flags(struct net_device *netdev)
+{
+	struct hnae3_handle *handle = hns3_get_handle(netdev);
+
+	return handle->priv_flags;
+}
+
+static int hns3_check_priv_flags(struct hnae3_handle *h, u32 changed)
+{
+	u32 i;
+
+	for (i = 0; i < HNAE3_PFLAG_MAX; i++)
+		if ((changed & BIT(i)) && !test_bit(i, &h->supported_pflags)) {
+			netdev_err(h->netdev, "%s is unsupported\n",
+				   hns3_priv_flags[i].name);
+			return -EOPNOTSUPP;
+		}
+
+	return 0;
+}
+
+static int hns3_set_priv_flags(struct net_device *netdev, u32 pflags)
+{
+	struct hnae3_handle *handle = hns3_get_handle(netdev);
+	u32 changed = pflags ^ handle->priv_flags;
+	int ret;
+	u32 i;
+
+	ret = hns3_check_priv_flags(handle, changed);
+	if (ret)
+		return ret;
+
+	for (i = 0; i < HNAE3_PFLAG_MAX; i++) {
+		if (changed & BIT(i)) {
+			bool enable = !(handle->priv_flags & BIT(i));
+
+			if (enable)
+				handle->priv_flags |= BIT(i);
+			else
+				handle->priv_flags &= ~BIT(i);
+			hns3_priv_flags[i].handler(netdev, enable);
+		}
+	}
+
+	return 0;
+}
+
+static int hns3_get_tunable(struct net_device *netdev,
+			    const struct ethtool_tunable *tuna,
+			    void *data)
+{
+	struct hns3_nic_priv *priv = netdev_priv(netdev);
+	struct hnae3_handle *h = priv->ae_handle;
+	int ret = 0;
+
+	switch (tuna->id) {
+	case ETHTOOL_TX_COPYBREAK:
+		/* all the tx rings have the same tx_copybreak */
+		*(u32 *)data = priv->tx_copybreak;
+		break;
+	case ETHTOOL_RX_COPYBREAK:
+		*(u32 *)data = priv->rx_copybreak;
+		break;
+	case ETHTOOL_TX_COPYBREAK_BUF_SIZE:
+		*(u32 *)data = h->kinfo.tx_spare_buf_size;
+		break;
+	default:
+		ret = -EOPNOTSUPP;
+		break;
+	}
+
+	return ret;
+}
+
+static int hns3_set_tx_spare_buf_size(struct net_device *netdev,
+				      u32 data)
+{
+	struct hns3_nic_priv *priv = netdev_priv(netdev);
+	struct hnae3_handle *h = priv->ae_handle;
+	int ret;
+
+	h->kinfo.tx_spare_buf_size = data;
+
+	ret = hns3_reset_notify(h, HNAE3_DOWN_CLIENT);
+	if (ret)
+		return ret;
+
+	ret = hns3_reset_notify(h, HNAE3_UNINIT_CLIENT);
+	if (ret)
+		return ret;
+
+	ret = hns3_reset_notify(h, HNAE3_INIT_CLIENT);
+	if (ret)
+		return ret;
+
+	ret = hns3_reset_notify(h, HNAE3_UP_CLIENT);
+	if (ret)
+		hns3_reset_notify(h, HNAE3_UNINIT_CLIENT);
+
+	return ret;
+}
+
+static int hns3_check_tx_copybreak(struct net_device *netdev, u32 copybreak)
+{
+	struct hns3_nic_priv *priv = netdev_priv(netdev);
+
+	if (copybreak < priv->min_tx_copybreak) {
+		netdev_err(netdev, "tx copybreak %u should be no less than %u!\n",
+			   copybreak, priv->min_tx_copybreak);
+		return -EINVAL;
+	}
+	return 0;
+}
+
+static int hns3_check_tx_spare_buf_size(struct net_device *netdev, u32 buf_size)
+{
+	struct hns3_nic_priv *priv = netdev_priv(netdev);
+
+	if (buf_size < priv->min_tx_spare_buf_size) {
+		netdev_err(netdev,
+			   "tx spare buf size %u should be no less than %u!\n",
+			   buf_size, priv->min_tx_spare_buf_size);
+		return -EINVAL;
+	}
+	return 0;
+}
+
+static int hns3_set_tunable(struct net_device *netdev,
+			    const struct ethtool_tunable *tuna,
+			    const void *data)
+{
+	struct hns3_nic_priv *priv = netdev_priv(netdev);
+	u32 old_tx_spare_buf_size, new_tx_spare_buf_size;
+	struct hnae3_handle *h = priv->ae_handle;
+	int i, ret = 0;
+
+	if (hns3_nic_resetting(netdev) || !priv->ring) {
+		netdev_err(netdev, "failed to set tunable value, dev resetting!\n");
+		return -EBUSY;
+	}
+
+	switch (tuna->id) {
+	case ETHTOOL_TX_COPYBREAK:
+		ret = hns3_check_tx_copybreak(netdev, *(u32 *)data);
+		if (ret)
+			return ret;
+
+		priv->tx_copybreak = *(u32 *)data;
+
+		for (i = 0; i < h->kinfo.num_tqps; i++)
+			priv->ring[i].tx_copybreak = priv->tx_copybreak;
+
+		break;
+	case ETHTOOL_RX_COPYBREAK:
+		priv->rx_copybreak = *(u32 *)data;
+
+		for (i = h->kinfo.num_tqps; i < h->kinfo.num_tqps * 2; i++)
+			priv->ring[i].rx_copybreak = priv->rx_copybreak;
+
+		break;
+	case ETHTOOL_TX_COPYBREAK_BUF_SIZE:
+		ret = hns3_check_tx_spare_buf_size(netdev, *(u32 *)data);
+		if (ret)
+			return ret;
+
+		old_tx_spare_buf_size = h->kinfo.tx_spare_buf_size;
+		new_tx_spare_buf_size = *(u32 *)data;
+		netdev_info(netdev, "request to set tx spare buf size from %u to %u\n",
+			    old_tx_spare_buf_size, new_tx_spare_buf_size);
+		ret = hns3_set_tx_spare_buf_size(netdev, new_tx_spare_buf_size);
+		if (ret ||
+		    (!priv->ring->tx_spare && new_tx_spare_buf_size != 0)) {
+			int ret1;
+
+			netdev_warn(netdev, "change tx spare buf size fail, revert to old value\n");
+			ret1 = hns3_set_tx_spare_buf_size(netdev,
+							  old_tx_spare_buf_size);
+			if (ret1) {
+				netdev_err(netdev, "revert to old tx spare buf size fail\n");
+				return ret1;
+			}
+
+			return ret;
+		}
+
+		if (!priv->ring->tx_spare)
+			netdev_info(netdev, "the active tx spare buf size is 0, disable tx spare buffer\n");
+		else
+			netdev_info(netdev, "the active tx spare buf size is %u, due to page order\n",
+				    priv->ring->tx_spare->len);
+
+		break;
+	default:
+		ret = -EOPNOTSUPP;
+		break;
+	}
+
+	return ret;
+}
+
+#define HNS3_ETHTOOL_COALESCE	(ETHTOOL_COALESCE_USECS |		\
+				 ETHTOOL_COALESCE_USE_ADAPTIVE |	\
+				 ETHTOOL_COALESCE_RX_USECS_HIGH |	\
+				 ETHTOOL_COALESCE_TX_USECS_HIGH |	\
+				 ETHTOOL_COALESCE_MAX_FRAMES |		\
+				 ETHTOOL_COALESCE_USE_CQE)
+
+#define HNS3_ETHTOOL_RING	(ETHTOOL_RING_USE_RX_BUF_LEN |		\
+				 ETHTOOL_RING_USE_TX_PUSH)
+
+static int hns3_get_ts_info(struct net_device *netdev,
+			    struct kernel_ethtool_ts_info *info)
+{
+	struct hnae3_handle *handle = hns3_get_handle(netdev);
+
+	if (handle->ae_algo->ops->get_ts_info)
+		return handle->ae_algo->ops->get_ts_info(handle, info);
+
+	return ethtool_op_get_ts_info(netdev, info);
+}
+
+static const struct hns3_ethtool_link_ext_state_mapping
+hns3_link_ext_state_map[] = {
+	{1, ETHTOOL_LINK_EXT_STATE_AUTONEG,
+		ETHTOOL_LINK_EXT_SUBSTATE_AN_NO_HCD},
+	{2, ETHTOOL_LINK_EXT_STATE_AUTONEG,
+		ETHTOOL_LINK_EXT_SUBSTATE_AN_ACK_NOT_RECEIVED},
+
+	{256, ETHTOOL_LINK_EXT_STATE_LINK_TRAINING_FAILURE,
+		ETHTOOL_LINK_EXT_SUBSTATE_LT_KR_LINK_INHIBIT_TIMEOUT},
+	{257, ETHTOOL_LINK_EXT_STATE_LINK_TRAINING_FAILURE,
+		ETHTOOL_LINK_EXT_SUBSTATE_LT_KR_LINK_PARTNER_DID_NOT_SET_RECEIVER_READY},
+	{512, ETHTOOL_LINK_EXT_STATE_LINK_TRAINING_FAILURE,
+		ETHTOOL_LINK_EXT_SUBSTATE_LT_REMOTE_FAULT},
+
+	{513, ETHTOOL_LINK_EXT_STATE_LINK_LOGICAL_MISMATCH,
+		ETHTOOL_LINK_EXT_SUBSTATE_LLM_PCS_DID_NOT_ACQUIRE_BLOCK_LOCK},
+	{514, ETHTOOL_LINK_EXT_STATE_LINK_LOGICAL_MISMATCH,
+		ETHTOOL_LINK_EXT_SUBSTATE_LLM_FC_FEC_IS_NOT_LOCKED},
+	{515, ETHTOOL_LINK_EXT_STATE_LINK_LOGICAL_MISMATCH,
+		ETHTOOL_LINK_EXT_SUBSTATE_LLM_RS_FEC_IS_NOT_LOCKED},
+
+	{768, ETHTOOL_LINK_EXT_STATE_BAD_SIGNAL_INTEGRITY,
+		ETHTOOL_LINK_EXT_SUBSTATE_BSI_LARGE_NUMBER_OF_PHYSICAL_ERRORS},
+	{769, ETHTOOL_LINK_EXT_STATE_BAD_SIGNAL_INTEGRITY,
+		ETHTOOL_LINK_EXT_SUBSTATE_BSI_SERDES_REFERENCE_CLOCK_LOST},
+	{770, ETHTOOL_LINK_EXT_STATE_BAD_SIGNAL_INTEGRITY,
+		ETHTOOL_LINK_EXT_SUBSTATE_BSI_SERDES_ALOS},
+
+	{1024, ETHTOOL_LINK_EXT_STATE_NO_CABLE, 0},
+	{1025, ETHTOOL_LINK_EXT_STATE_CABLE_ISSUE,
+		ETHTOOL_LINK_EXT_SUBSTATE_CI_UNSUPPORTED_CABLE},
+
+	{1026, ETHTOOL_LINK_EXT_STATE_EEPROM_ISSUE, 0},
+};
+
+static int hns3_get_link_ext_state(struct net_device *netdev,
+				   struct ethtool_link_ext_state_info *info)
+{
+	const struct hns3_ethtool_link_ext_state_mapping *map;
+	struct hnae3_handle *h = hns3_get_handle(netdev);
+	u32 status_code, i;
+	int ret;
+
+	if (netif_carrier_ok(netdev))
+		return -ENODATA;
+
+	if (!h->ae_algo->ops->get_link_diagnosis_info)
+		return -EOPNOTSUPP;
+
+	ret = h->ae_algo->ops->get_link_diagnosis_info(h, &status_code);
+	if (ret)
+		return ret;
+
+	for (i = 0; i < ARRAY_SIZE(hns3_link_ext_state_map); i++) {
+		map = &hns3_link_ext_state_map[i];
+		if (map->status_code == status_code) {
+			info->link_ext_state = map->link_ext_state;
+			info->__link_ext_substate = map->link_ext_substate;
+			return 0;
+		}
+	}
+
+	return -ENODATA;
+}
+
+static void hns3_get_wol(struct net_device *netdev, struct ethtool_wolinfo *wol)
+{
+	struct hnae3_handle *handle = hns3_get_handle(netdev);
+	const struct hnae3_ae_ops *ops = hns3_get_ops(handle);
+	struct hnae3_ae_dev *ae_dev = hns3_get_ae_dev(handle);
+
+	if (!hnae3_ae_dev_wol_supported(ae_dev))
+		return;
+
+	ops->get_wol(handle, wol);
+}
+
+static int hns3_set_wol(struct net_device *netdev,
+			struct ethtool_wolinfo *wol)
+{
+	struct hnae3_handle *handle = hns3_get_handle(netdev);
+	const struct hnae3_ae_ops *ops = hns3_get_ops(handle);
+	struct hnae3_ae_dev *ae_dev = hns3_get_ae_dev(handle);
+
+	if (!hnae3_ae_dev_wol_supported(ae_dev))
+		return -EOPNOTSUPP;
+
+	return ops->set_wol(handle, wol);
+}
+
+static const struct ethtool_ops hns3vf_ethtool_ops = {
+	.supported_coalesce_params = HNS3_ETHTOOL_COALESCE,
+	.supported_ring_params = HNS3_ETHTOOL_RING,
+	.get_drvinfo = hns3_get_drvinfo,
+	.get_ringparam = hns3_get_ringparam,
+	.set_ringparam = hns3_set_ringparam,
+	.get_strings = hns3_get_strings,
+	.get_ethtool_stats = hns3_get_stats,
+	.get_sset_count = hns3_get_sset_count,
+	.get_rxnfc = hns3_get_rxnfc,
+	.set_rxnfc = hns3_set_rxnfc,
+	.get_rxfh_key_size = hns3_get_rss_key_size,
+	.get_rxfh_indir_size = hns3_get_rss_indir_size,
+	.get_rxfh = hns3_get_rss,
+	.set_rxfh = hns3_set_rss,
+	.get_rxfh_fields = hns3_get_rxfh_fields,
+	.set_rxfh_fields = hns3_set_rxfh_fields,
+	.get_link_ksettings = hns3_get_link_ksettings,
+	.get_channels = hns3_get_channels,
+	.set_channels = hns3_set_channels,
+	.get_coalesce = hns3_get_coalesce,
+	.set_coalesce = hns3_set_coalesce,
+	.get_regs_len = hns3_get_regs_len,
+	.get_regs = hns3_get_regs,
+	.get_link = hns3_get_link,
+	.get_msglevel = hns3_get_msglevel,
+	.set_msglevel = hns3_set_msglevel,
+	.get_priv_flags = hns3_get_priv_flags,
+	.set_priv_flags = hns3_set_priv_flags,
+	.get_tunable = hns3_get_tunable,
+	.set_tunable = hns3_set_tunable,
+	.reset = hns3_set_reset,
+};
+
+static const struct ethtool_ops hns3_ethtool_ops = {
+	.supported_coalesce_params = HNS3_ETHTOOL_COALESCE,
+	.supported_ring_params = HNS3_ETHTOOL_RING,
+	.cap_link_lanes_supported = true,
+	.self_test = hns3_self_test,
+	.get_drvinfo = hns3_get_drvinfo,
+	.get_link = hns3_get_link,
+	.get_ringparam = hns3_get_ringparam,
+	.set_ringparam = hns3_set_ringparam,
+	.get_pauseparam = hns3_get_pauseparam,
+	.set_pauseparam = hns3_set_pauseparam,
+	.get_strings = hns3_get_strings,
+	.get_ethtool_stats = hns3_get_stats,
+	.get_sset_count = hns3_get_sset_count,
+	.get_rxnfc = hns3_get_rxnfc,
+	.set_rxnfc = hns3_set_rxnfc,
+	.get_rxfh_key_size = hns3_get_rss_key_size,
+	.get_rxfh_indir_size = hns3_get_rss_indir_size,
+	.get_rxfh = hns3_get_rss,
+	.set_rxfh = hns3_set_rss,
+	.get_rxfh_fields = hns3_get_rxfh_fields,
+	.set_rxfh_fields = hns3_set_rxfh_fields,
+	.get_link_ksettings = hns3_get_link_ksettings,
+	.set_link_ksettings = hns3_set_link_ksettings,
+	.nway_reset = hns3_nway_reset,
+	.get_channels = hns3_get_channels,
+	.set_channels = hns3_set_channels,
+	.get_coalesce = hns3_get_coalesce,
+	.set_coalesce = hns3_set_coalesce,
+	.get_regs_len = hns3_get_regs_len,
+	.get_regs = hns3_get_regs,
+	.set_phys_id = hns3_set_phys_id,
+	.get_msglevel = hns3_get_msglevel,
+	.set_msglevel = hns3_set_msglevel,
+	.get_fecparam = hns3_get_fecparam,
+	.set_fecparam = hns3_set_fecparam,
+	.get_fec_stats = hns3_get_fec_stats,
+	.get_module_info = hns3_get_module_info,
+	.get_module_eeprom = hns3_get_module_eeprom,
+	.get_priv_flags = hns3_get_priv_flags,
+	.set_priv_flags = hns3_set_priv_flags,
+	.get_ts_info = hns3_get_ts_info,
+	.get_tunable = hns3_get_tunable,
+	.set_tunable = hns3_set_tunable,
+	.reset = hns3_set_reset,
+	.get_link_ext_state = hns3_get_link_ext_state,
+	.get_wol = hns3_get_wol,
+	.set_wol = hns3_set_wol,
+};
+
+void hns3_ethtool_set_ops(struct net_device *netdev)
+{
+	struct hnae3_handle *h = hns3_get_handle(netdev);
+
+	if (h->flags & HNAE3_SUPPORT_VF)
+		netdev->ethtool_ops = &hns3vf_ethtool_ops;
+	else
+		netdev->ethtool_ops = &hns3_ethtool_ops;
+}
diff --git a/drivers/net/ethernet/hisilicon/hns3/hns3_ethtool.h b/drivers/net/ethernet/hisilicon/hns3/hns3_ethtool.h
new file mode 100644
index 000000000000..da207d1d9aa9
--- /dev/null
+++ b/drivers/net/ethernet/hisilicon/hns3/hns3_ethtool.h
@@ -0,0 +1,37 @@
+/* SPDX-License-Identifier: GPL-2.0+ */
+// Copyright (c) 2021 Hisilicon Limited.
+
+#ifndef __HNS3_ETHTOOL_H
+#define __HNS3_ETHTOOL_H
+
+#include <linux/ethtool.h>
+#include <linux/netdevice.h>
+
+struct hns3_stats {
+	char stats_string[ETH_GSTRING_LEN];
+	int stats_offset;
+};
+
+struct hns3_sfp_type {
+	u8 type;
+	u8 ext_type;
+};
+
+struct hns3_pflag_desc {
+	char name[ETH_GSTRING_LEN];
+	void (*handler)(struct net_device *netdev, bool enable);
+};
+
+struct hns3_ethtool_link_ext_state_mapping {
+	u32 status_code;
+	enum ethtool_link_ext_state link_ext_state;
+	u8 link_ext_substate;
+};
+
+struct hns3_ring_param {
+	u32 tx_desc_num;
+	u32 rx_desc_num;
+	u32 rx_buf_len;
+};
+
+#endif
diff --git a/drivers/net/ethernet/hisilicon/hns3/hns3_trace.h b/drivers/net/ethernet/hisilicon/hns3/hns3_trace.h
new file mode 100644
index 000000000000..3362b8d14d4f
--- /dev/null
+++ b/drivers/net/ethernet/hisilicon/hns3/hns3_trace.h
@@ -0,0 +1,138 @@
+/* SPDX-License-Identifier: GPL-2.0+ */
+/* Copyright (c) 2018-2019 Hisilicon Limited. */
+
+/* This must be outside ifdef _HNS3_TRACE_H */
+#undef TRACE_SYSTEM
+#define TRACE_SYSTEM hns3
+
+#if !defined(_HNS3_TRACE_H_) || defined(TRACE_HEADER_MULTI_READ)
+#define _HNS3_TRACE_H_
+
+#include <linux/tracepoint.h>
+
+#define DESC_NR		(sizeof(struct hns3_desc) / sizeof(u32))
+
+DECLARE_EVENT_CLASS(hns3_skb_template,
+	TP_PROTO(struct sk_buff *skb),
+	TP_ARGS(skb),
+
+	TP_STRUCT__entry(
+		__field(unsigned int, headlen)
+		__field(unsigned int, len)
+		__field(__u8, nr_frags)
+		__field(__u8, ip_summed)
+		__field(unsigned int, hdr_len)
+		__field(unsigned short, gso_size)
+		__field(unsigned short, gso_segs)
+		__field(unsigned int, gso_type)
+		__field(bool, fraglist)
+		__array(__u32, size, MAX_SKB_FRAGS)
+	),
+
+	TP_fast_assign(
+		__entry->headlen = skb_headlen(skb);
+		__entry->len = skb->len;
+		__entry->nr_frags = skb_shinfo(skb)->nr_frags;
+		__entry->gso_size = skb_shinfo(skb)->gso_size;
+		__entry->gso_segs = skb_shinfo(skb)->gso_segs;
+		__entry->gso_type = skb_shinfo(skb)->gso_type;
+		__entry->hdr_len = skb->encapsulation ?
+		skb_inner_tcp_all_headers(skb) : skb_tcp_all_headers(skb);
+		__entry->ip_summed = skb->ip_summed;
+		__entry->fraglist = skb_has_frag_list(skb);
+		hns3_shinfo_pack(skb_shinfo(skb), __entry->size);
+	),
+
+	TP_printk(
+		"len: %u, %u, %u, cs: %u, gso: %u, %u, %x, frag(%d %u): %s",
+		__entry->headlen, __entry->len, __entry->hdr_len,
+		__entry->ip_summed, __entry->gso_size, __entry->gso_segs,
+		__entry->gso_type, __entry->fraglist, __entry->nr_frags,
+		__print_array(__entry->size, MAX_SKB_FRAGS, sizeof(__u32))
+	)
+);
+
+DEFINE_EVENT(hns3_skb_template, hns3_over_max_bd,
+	TP_PROTO(struct sk_buff *skb),
+	TP_ARGS(skb));
+
+DEFINE_EVENT(hns3_skb_template, hns3_gro,
+	TP_PROTO(struct sk_buff *skb),
+	TP_ARGS(skb));
+
+DEFINE_EVENT(hns3_skb_template, hns3_tso,
+	TP_PROTO(struct sk_buff *skb),
+	TP_ARGS(skb));
+
+TRACE_EVENT(hns3_tx_desc,
+	TP_PROTO(struct hns3_enet_ring *ring, int cur_ntu),
+	TP_ARGS(ring, cur_ntu),
+
+	TP_STRUCT__entry(
+		__field(int, index)
+		__field(int, ntu)
+		__field(int, ntc)
+		__field(dma_addr_t, desc_dma)
+		__array(u32, desc, DESC_NR)
+		__string(devname, ring->tqp->handle->kinfo.netdev->name)
+	),
+
+	TP_fast_assign(
+		__entry->index = ring->tqp->tqp_index;
+		__entry->ntu = ring->next_to_use;
+		__entry->ntc = ring->next_to_clean;
+		__entry->desc_dma = ring->desc_dma_addr,
+		memcpy(__entry->desc, &ring->desc[cur_ntu],
+		       sizeof(struct hns3_desc));
+		__assign_str(devname);
+	),
+
+	TP_printk(
+		"%s-%d-%d/%d desc(%pad): %s",
+		__get_str(devname), __entry->index, __entry->ntu,
+		__entry->ntc, &__entry->desc_dma,
+		__print_array(__entry->desc, DESC_NR, sizeof(u32))
+	)
+);
+
+TRACE_EVENT(hns3_rx_desc,
+	TP_PROTO(struct hns3_enet_ring *ring),
+	TP_ARGS(ring),
+
+	TP_STRUCT__entry(
+		__field(int, index)
+		__field(int, ntu)
+		__field(int, ntc)
+		__field(dma_addr_t, desc_dma)
+		__field(dma_addr_t, buf_dma)
+		__array(u32, desc, DESC_NR)
+		__string(devname, ring->tqp->handle->kinfo.netdev->name)
+	),
+
+	TP_fast_assign(
+		__entry->index = ring->tqp->tqp_index;
+		__entry->ntu = ring->next_to_use;
+		__entry->ntc = ring->next_to_clean;
+		__entry->desc_dma = ring->desc_dma_addr;
+		__entry->buf_dma = ring->desc_cb[ring->next_to_clean].dma;
+		memcpy(__entry->desc, &ring->desc[ring->next_to_clean],
+		       sizeof(struct hns3_desc));
+		__assign_str(devname);
+	),
+
+	TP_printk(
+		"%s-%d-%d/%d desc(%pad) buf(%pad): %s",
+		__get_str(devname), __entry->index, __entry->ntu,
+		__entry->ntc, &__entry->desc_dma, &__entry->buf_dma,
+		__print_array(__entry->desc, DESC_NR, sizeof(u32))
+	)
+);
+
+#endif /* _HNS3_TRACE_H_ */
+
+/* This must be outside ifdef _HNS3_TRACE_H */
+#undef TRACE_INCLUDE_PATH
+#define TRACE_INCLUDE_PATH .
+#undef TRACE_INCLUDE_FILE
+#define TRACE_INCLUDE_FILE hns3_trace
+#include <trace/define_trace.h>
diff --git a/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_cmd.h b/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_cmd.h
new file mode 100644
index 000000000000..9bb708fa42f2
--- /dev/null
+++ b/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_cmd.h
@@ -0,0 +1,892 @@
+/* SPDX-License-Identifier: GPL-2.0+ */
+// Copyright (c) 2016-2017 Hisilicon Limited.
+
+#ifndef __HCLGE_CMD_H
+#define __HCLGE_CMD_H
+#include <linux/types.h>
+#include <linux/io.h>
+#include <linux/etherdevice.h>
+#include "hnae3.h"
+#include "hclge_comm_cmd.h"
+
+struct hclge_dev;
+
+#define HCLGE_CMDQ_RX_INVLD_B		0
+#define HCLGE_CMDQ_RX_OUTVLD_B		1
+
+struct hclge_misc_vector {
+	u8 __iomem *addr;
+	int vector_irq;
+	char name[HNAE3_INT_NAME_LEN];
+};
+
+#define hclge_cmd_setup_basic_desc(desc, opcode, is_read) \
+	hclge_comm_cmd_setup_basic_desc(desc, opcode, is_read)
+
+#define HCLGE_TQP_REG_OFFSET		0x80000
+#define HCLGE_TQP_REG_SIZE		0x200
+
+#define HCLGE_TQP_MAX_SIZE_DEV_V2	1024
+#define HCLGE_TQP_EXT_REG_OFFSET	0x100
+
+#define HCLGE_RCB_INIT_QUERY_TIMEOUT	10
+#define HCLGE_RCB_INIT_FLAG_EN_B	0
+#define HCLGE_RCB_INIT_FLAG_FINI_B	8
+struct hclge_config_rcb_init_cmd {
+	__le16 rcb_init_flag;
+	u8 rsv[22];
+};
+
+struct hclge_tqp_map_cmd {
+	__le16 tqp_id;	/* Absolute tqp id for in this pf */
+	u8 tqp_vf;	/* VF id */
+#define HCLGE_TQP_MAP_TYPE_PF		0
+#define HCLGE_TQP_MAP_TYPE_VF		1
+#define HCLGE_TQP_MAP_TYPE_B		0
+#define HCLGE_TQP_MAP_EN_B		1
+	u8 tqp_flag;	/* Indicate it's pf or vf tqp */
+	__le16 tqp_vid; /* Virtual id in this pf/vf */
+	u8 rsv[18];
+};
+
+#define HCLGE_VECTOR_ELEMENTS_PER_CMD	10
+
+enum hclge_int_type {
+	HCLGE_INT_TX,
+	HCLGE_INT_RX,
+	HCLGE_INT_EVENT,
+};
+
+struct hclge_ctrl_vector_chain_cmd {
+#define HCLGE_VECTOR_ID_L_S	0
+#define HCLGE_VECTOR_ID_L_M	GENMASK(7, 0)
+	u8 int_vector_id_l;
+	u8 int_cause_num;
+#define HCLGE_INT_TYPE_S	0
+#define HCLGE_INT_TYPE_M	GENMASK(1, 0)
+#define HCLGE_TQP_ID_S		2
+#define HCLGE_TQP_ID_M		GENMASK(12, 2)
+#define HCLGE_INT_GL_IDX_S	13
+#define HCLGE_INT_GL_IDX_M	GENMASK(14, 13)
+	__le16 tqp_type_and_id[HCLGE_VECTOR_ELEMENTS_PER_CMD];
+	u8 vfid;
+#define HCLGE_VECTOR_ID_H_S	8
+#define HCLGE_VECTOR_ID_H_M	GENMASK(15, 8)
+	u8 int_vector_id_h;
+};
+
+#define HCLGE_MAX_TC_NUM		8
+#define HCLGE_TC0_PRI_BUF_EN_B	15 /* Bit 15 indicate enable or not */
+#define HCLGE_BUF_UNIT_S	7  /* Buf size is united by 128 bytes */
+struct hclge_tx_buff_alloc_cmd {
+	__le16 tx_pkt_buff[HCLGE_MAX_TC_NUM];
+	u8 tx_buff_rsv[8];
+};
+
+struct hclge_rx_priv_buff_cmd {
+	__le16 buf_num[HCLGE_MAX_TC_NUM];
+	__le16 shared_buf;
+	u8 rsv[6];
+};
+
+#define HCLGE_RX_PRIV_EN_B	15
+#define HCLGE_TC_NUM_ONE_DESC	4
+struct hclge_priv_wl {
+	__le16 high;
+	__le16 low;
+};
+
+struct hclge_rx_priv_wl_buf {
+	struct hclge_priv_wl tc_wl[HCLGE_TC_NUM_ONE_DESC];
+};
+
+struct hclge_rx_com_thrd {
+	struct hclge_priv_wl com_thrd[HCLGE_TC_NUM_ONE_DESC];
+};
+
+struct hclge_rx_com_wl {
+	struct hclge_priv_wl com_wl;
+};
+
+struct hclge_waterline {
+	u32 low;
+	u32 high;
+};
+
+struct hclge_tc_thrd {
+	u32 low;
+	u32 high;
+};
+
+struct hclge_priv_buf {
+	struct hclge_waterline wl;	/* Waterline for low and high */
+	u32 buf_size;	/* TC private buffer size */
+	u32 tx_buf_size;
+	u32 enable;	/* Enable TC private buffer or not */
+};
+
+struct hclge_shared_buf {
+	struct hclge_waterline self;
+	struct hclge_tc_thrd tc_thrd[HCLGE_MAX_TC_NUM];
+	u32 buf_size;
+};
+
+struct hclge_pkt_buf_alloc {
+	struct hclge_priv_buf priv_buf[HCLGE_MAX_TC_NUM];
+	struct hclge_shared_buf s_buf;
+};
+
+#define HCLGE_RX_COM_WL_EN_B	15
+struct hclge_rx_com_wl_buf_cmd {
+	__le16 high_wl;
+	__le16 low_wl;
+	u8 rsv[20];
+};
+
+#define HCLGE_RX_PKT_EN_B	15
+struct hclge_rx_pkt_buf_cmd {
+	__le16 high_pkt;
+	__le16 low_pkt;
+	u8 rsv[20];
+};
+
+#define HCLGE_PF_STATE_DONE_B	0
+#define HCLGE_PF_STATE_MAIN_B	1
+#define HCLGE_PF_STATE_BOND_B	2
+#define HCLGE_PF_STATE_MAC_N_B	6
+#define HCLGE_PF_MAC_NUM_MASK	0x3
+#define HCLGE_PF_STATE_MAIN	BIT(HCLGE_PF_STATE_MAIN_B)
+#define HCLGE_PF_STATE_DONE	BIT(HCLGE_PF_STATE_DONE_B)
+#define HCLGE_VF_RST_STATUS_CMD	4
+
+struct hclge_func_status_cmd {
+	__le32  vf_rst_state[HCLGE_VF_RST_STATUS_CMD];
+	u8 pf_state;
+	u8 mac_id;
+	u8 rsv1;
+	u8 pf_cnt_in_mac;
+	u8 pf_num;
+	u8 vf_num;
+	u8 rsv[2];
+};
+
+struct hclge_pf_res_cmd {
+	__le16 tqp_num;
+	__le16 buf_size;
+	__le16 msixcap_localid_ba_nic;
+	__le16 msixcap_localid_number_nic;
+	__le16 pf_intr_vector_number_roce;
+	__le16 pf_own_fun_number;
+	__le16 tx_buf_size;
+	__le16 dv_buf_size;
+	__le16 ext_tqp_num;
+	u8 rsv[6];
+};
+
+#define HCLGE_CFG_OFFSET_S	0
+#define HCLGE_CFG_OFFSET_M	GENMASK(19, 0)
+#define HCLGE_CFG_RD_LEN_S	24
+#define HCLGE_CFG_RD_LEN_M	GENMASK(27, 24)
+#define HCLGE_CFG_RD_LEN_BYTES	16
+#define HCLGE_CFG_RD_LEN_UNIT	4
+
+#define HCLGE_CFG_TC_NUM_S	8
+#define HCLGE_CFG_TC_NUM_M	GENMASK(15, 8)
+#define HCLGE_CFG_TQP_DESC_N_S	16
+#define HCLGE_CFG_TQP_DESC_N_M	GENMASK(31, 16)
+#define HCLGE_CFG_PHY_ADDR_S	0
+#define HCLGE_CFG_PHY_ADDR_M	GENMASK(7, 0)
+#define HCLGE_CFG_MEDIA_TP_S	8
+#define HCLGE_CFG_MEDIA_TP_M	GENMASK(15, 8)
+#define HCLGE_CFG_RX_BUF_LEN_S	16
+#define HCLGE_CFG_RX_BUF_LEN_M	GENMASK(31, 16)
+#define HCLGE_CFG_MAC_ADDR_H_S	0
+#define HCLGE_CFG_MAC_ADDR_H_M	GENMASK(15, 0)
+#define HCLGE_CFG_DEFAULT_SPEED_S	16
+#define HCLGE_CFG_DEFAULT_SPEED_M	GENMASK(23, 16)
+#define HCLGE_CFG_RSS_SIZE_S	24
+#define HCLGE_CFG_RSS_SIZE_M	GENMASK(31, 24)
+#define HCLGE_CFG_SPEED_ABILITY_S	0
+#define HCLGE_CFG_SPEED_ABILITY_M	GENMASK(7, 0)
+#define HCLGE_CFG_SPEED_ABILITY_EXT_S	10
+#define HCLGE_CFG_SPEED_ABILITY_EXT_M	GENMASK(15, 10)
+#define HCLGE_CFG_VLAN_FLTR_CAP_S	8
+#define HCLGE_CFG_VLAN_FLTR_CAP_M	GENMASK(9, 8)
+#define HCLGE_CFG_UMV_TBL_SPACE_S	16
+#define HCLGE_CFG_UMV_TBL_SPACE_M	GENMASK(31, 16)
+#define HCLGE_CFG_PF_RSS_SIZE_S		0
+#define HCLGE_CFG_PF_RSS_SIZE_M		GENMASK(3, 0)
+#define HCLGE_CFG_TX_SPARE_BUF_SIZE_S	4
+#define HCLGE_CFG_TX_SPARE_BUF_SIZE_M	GENMASK(15, 4)
+
+#define HCLGE_CFG_CMD_CNT		4
+
+struct hclge_cfg_param_cmd {
+	__le32 offset;
+	__le32 rsv;
+	__le32 param[HCLGE_CFG_CMD_CNT];
+};
+
+#define HCLGE_MAC_MODE		0x0
+#define HCLGE_DESC_NUM		0x40
+
+#define HCLGE_ALLOC_VALID_B	0
+struct hclge_vf_num_cmd {
+	u8 alloc_valid;
+	u8 rsv[23];
+};
+
+#define HCLGE_RSS_DEFAULT_OUTPORT_B	4
+
+#define HCLGE_RSS_CFG_TBL_SIZE_H	4
+#define HCLGE_RSS_CFG_TBL_BW_L		8U
+
+#define HCLGE_RSS_TC_OFFSET_S		0
+#define HCLGE_RSS_TC_OFFSET_M		GENMASK(10, 0)
+#define HCLGE_RSS_TC_SIZE_MSB_B		11
+#define HCLGE_RSS_TC_SIZE_S		12
+#define HCLGE_RSS_TC_SIZE_M		GENMASK(14, 12)
+#define HCLGE_RSS_TC_SIZE_MSB_OFFSET	3
+#define HCLGE_RSS_TC_VALID_B		15
+
+#define HCLGE_LINK_STATUS_UP_B	0
+#define HCLGE_LINK_STATUS_UP_M	BIT(HCLGE_LINK_STATUS_UP_B)
+struct hclge_link_status_cmd {
+	u8 status;
+	u8 rsv[23];
+};
+
+/* for DEVICE_VERSION_V1/2, reference to promisc cmd byte8 */
+#define HCLGE_PROMISC_EN_UC	1
+#define HCLGE_PROMISC_EN_MC	2
+#define HCLGE_PROMISC_EN_BC	3
+#define HCLGE_PROMISC_TX_EN	4
+#define HCLGE_PROMISC_RX_EN	5
+
+/* for DEVICE_VERSION_V3, reference to promisc cmd byte10 */
+#define HCLGE_PROMISC_UC_RX_EN	2
+#define HCLGE_PROMISC_MC_RX_EN	3
+#define HCLGE_PROMISC_BC_RX_EN	4
+#define HCLGE_PROMISC_UC_TX_EN	5
+#define HCLGE_PROMISC_MC_TX_EN	6
+#define HCLGE_PROMISC_BC_TX_EN	7
+
+struct hclge_promisc_cfg_cmd {
+	u8 promisc;
+	u8 vf_id;
+	u8 extend_promisc;
+	u8 rsv0[21];
+};
+
+enum hclge_promisc_type {
+	HCLGE_UNICAST	= 1,
+	HCLGE_MULTICAST	= 2,
+	HCLGE_BROADCAST	= 3,
+};
+
+#define HCLGE_MAC_TX_EN_B	6
+#define HCLGE_MAC_RX_EN_B	7
+#define HCLGE_MAC_PAD_TX_B	11
+#define HCLGE_MAC_PAD_RX_B	12
+#define HCLGE_MAC_1588_TX_B	13
+#define HCLGE_MAC_1588_RX_B	14
+#define HCLGE_MAC_APP_LP_B	15
+#define HCLGE_MAC_LINE_LP_B	16
+#define HCLGE_MAC_FCS_TX_B	17
+#define HCLGE_MAC_RX_OVERSIZE_TRUNCATE_B	18
+#define HCLGE_MAC_RX_FCS_STRIP_B	19
+#define HCLGE_MAC_RX_FCS_B	20
+#define HCLGE_MAC_TX_UNDER_MIN_ERR_B		21
+#define HCLGE_MAC_TX_OVERSIZE_TRUNCATE_B	22
+
+struct hclge_config_mac_mode_cmd {
+	__le32 txrx_pad_fcs_loop_en;
+	u8 rsv[20];
+};
+
+struct hclge_pf_rst_sync_cmd {
+#define HCLGE_PF_RST_ALL_VF_RDY_B	0
+	u8 all_vf_ready;
+	u8 rsv[23];
+};
+
+#define HCLGE_CFG_SPEED_S		0
+#define HCLGE_CFG_SPEED_M		GENMASK(5, 0)
+
+#define HCLGE_CFG_DUPLEX_B		7
+#define HCLGE_CFG_DUPLEX_M		BIT(HCLGE_CFG_DUPLEX_B)
+
+struct hclge_config_mac_speed_dup_cmd {
+	u8 speed_dup;
+
+#define HCLGE_CFG_MAC_SPEED_CHANGE_EN_B	0
+	u8 mac_change_fec_en;
+	u8 rsv[4];
+	u8 lane_num;
+	u8 rsv1[17];
+};
+
+#define HCLGE_TQP_ENABLE_B		0
+
+#define HCLGE_MAC_CFG_AN_EN_B		0
+#define HCLGE_MAC_CFG_AN_INT_EN_B	1
+#define HCLGE_MAC_CFG_AN_INT_MSK_B	2
+#define HCLGE_MAC_CFG_AN_INT_CLR_B	3
+#define HCLGE_MAC_CFG_AN_RST_B		4
+
+#define HCLGE_MAC_CFG_AN_EN	BIT(HCLGE_MAC_CFG_AN_EN_B)
+
+struct hclge_config_auto_neg_cmd {
+	__le32  cfg_an_cmd_flag;
+	u8      rsv[20];
+};
+
+struct hclge_sfp_info_cmd {
+	__le32 speed;
+	u8 query_type; /* 0: sfp speed, 1: active speed */
+	u8 active_fec;
+	u8 autoneg; /* autoneg state */
+	u8 autoneg_ability; /* whether support autoneg */
+	__le32 speed_ability; /* speed ability for current media */
+	__le32 module_type;
+	u8 fec_ability;
+	u8 lane_num;
+	u8 rsv[6];
+};
+
+#define HCLGE_MAC_CFG_FEC_AUTO_EN_B	0
+#define HCLGE_MAC_CFG_FEC_MODE_S	1
+#define HCLGE_MAC_CFG_FEC_MODE_M	GENMASK(3, 1)
+#define HCLGE_MAC_CFG_FEC_SET_DEF_B	0
+#define HCLGE_MAC_CFG_FEC_CLR_DEF_B	1
+
+#define HCLGE_MAC_FEC_OFF		0
+#define HCLGE_MAC_FEC_BASER		1
+#define HCLGE_MAC_FEC_RS		2
+#define HCLGE_MAC_FEC_LLRS		3
+struct hclge_config_fec_cmd {
+	u8 fec_mode;
+	u8 default_config;
+	u8 rsv[22];
+};
+
+#define HCLGE_FEC_STATS_CMD_NUM 4
+
+struct hclge_query_fec_stats_cmd {
+	/* fec rs mode total stats */
+	__le32 rs_fec_corr_blocks;
+	__le32 rs_fec_uncorr_blocks;
+	__le32 rs_fec_error_blocks;
+	/* fec base-r mode per lanes stats */
+	u8 base_r_lane_num;
+	u8 rsv[3];
+	__le32 base_r_fec_corr_blocks;
+	__le32 base_r_fec_uncorr_blocks;
+};
+
+#define HCLGE_MAC_UPLINK_PORT		0x100
+
+struct hclge_config_max_frm_size_cmd {
+	__le16  max_frm_size;
+	u8      min_frm_size;
+	u8      rsv[21];
+};
+
+enum hclge_mac_vlan_tbl_opcode {
+	HCLGE_MAC_VLAN_ADD,	/* Add new or modify mac_vlan */
+	HCLGE_MAC_VLAN_UPDATE,  /* Modify other fields of this table */
+	HCLGE_MAC_VLAN_REMOVE,  /* Remove a entry through mac_vlan key */
+	HCLGE_MAC_VLAN_LKUP,    /* Lookup a entry through mac_vlan key */
+};
+
+enum hclge_mac_vlan_add_resp_code {
+	HCLGE_ADD_UC_OVERFLOW = 2,	/* ADD failed for UC overflow */
+	HCLGE_ADD_MC_OVERFLOW,		/* ADD failed for MC overflow */
+};
+
+#define HCLGE_MAC_VLAN_BIT0_EN_B	0
+#define HCLGE_MAC_VLAN_BIT1_EN_B	1
+#define HCLGE_MAC_EPORT_SW_EN_B		12
+#define HCLGE_MAC_EPORT_TYPE_B		11
+#define HCLGE_MAC_EPORT_VFID_S		3
+#define HCLGE_MAC_EPORT_VFID_M		GENMASK(10, 3)
+#define HCLGE_MAC_EPORT_PFID_S		0
+#define HCLGE_MAC_EPORT_PFID_M		GENMASK(2, 0)
+struct hclge_mac_vlan_tbl_entry_cmd {
+	u8	flags;
+	u8      resp_code;
+	__le16  vlan_tag;
+	__le32  mac_addr_hi32;
+	__le16  mac_addr_lo16;
+	__le16  rsv1;
+	u8      entry_type;
+	u8      mc_mac_en;
+	__le16  egress_port;
+	__le16  egress_queue;
+	u8      rsv2[6];
+};
+
+#define HCLGE_UMV_SPC_ALC_B	0
+struct hclge_umv_spc_alc_cmd {
+	u8 allocate;
+	u8 rsv1[3];
+	__le32 space_size;
+	u8 rsv2[16];
+};
+
+#define HCLGE_MAC_MGR_MASK_VLAN_B		BIT(0)
+#define HCLGE_MAC_MGR_MASK_MAC_B		BIT(1)
+#define HCLGE_MAC_MGR_MASK_ETHERTYPE_B		BIT(2)
+
+struct hclge_mac_mgr_tbl_entry_cmd {
+	u8      flags;
+	u8      resp_code;
+	__le16  vlan_tag;
+	u8      mac_addr[ETH_ALEN];
+	__le16  rsv1;
+	__le16  ethter_type;
+	__le16  egress_port;
+	__le16  egress_queue;
+	u8      sw_port_id_aware;
+	u8      rsv2;
+	u8      i_port_bitmap;
+	u8      i_port_direction;
+	u8      rsv3[2];
+};
+
+struct hclge_vlan_filter_ctrl_cmd {
+	u8 vlan_type;
+	u8 vlan_fe;
+	u8 rsv1[2];
+	u8 vf_id;
+	u8 rsv2[19];
+};
+
+#define HCLGE_VLAN_ID_OFFSET_STEP	160
+#define HCLGE_VLAN_BYTE_SIZE		8
+#define	HCLGE_VLAN_OFFSET_BITMAP \
+	(HCLGE_VLAN_ID_OFFSET_STEP / HCLGE_VLAN_BYTE_SIZE)
+
+struct hclge_vlan_filter_pf_cfg_cmd {
+	u8 vlan_offset;
+	u8 vlan_cfg;
+	u8 rsv[2];
+	u8 vlan_offset_bitmap[HCLGE_VLAN_OFFSET_BITMAP];
+};
+
+#define HCLGE_MAX_VF_BYTES  16
+
+struct hclge_vlan_filter_vf_cfg_cmd {
+	__le16 vlan_id;
+	u8  resp_code;
+	u8  rsv;
+	u8  vlan_cfg;
+	u8  rsv1[3];
+	u8  vf_bitmap[HCLGE_MAX_VF_BYTES];
+};
+
+#define HCLGE_INGRESS_BYPASS_B		0
+struct hclge_port_vlan_filter_bypass_cmd {
+	u8 bypass_state;
+	u8 rsv1[3];
+	u8 vf_id;
+	u8 rsv2[19];
+};
+
+#define HCLGE_SWITCH_ANTI_SPOOF_B	0U
+#define HCLGE_SWITCH_ALW_LPBK_B		1U
+#define HCLGE_SWITCH_ALW_LCL_LPBK_B	2U
+#define HCLGE_SWITCH_ALW_DST_OVRD_B	3U
+#define HCLGE_SWITCH_NO_MASK		0x0
+#define HCLGE_SWITCH_ANTI_SPOOF_MASK	0xFE
+#define HCLGE_SWITCH_ALW_LPBK_MASK	0xFD
+#define HCLGE_SWITCH_ALW_LCL_LPBK_MASK	0xFB
+#define HCLGE_SWITCH_LW_DST_OVRD_MASK	0xF7
+
+struct hclge_mac_vlan_switch_cmd {
+	u8 roce_sel;
+	u8 rsv1[3];
+	__le32 func_id;
+	u8 switch_param;
+	u8 rsv2[3];
+	u8 param_mask;
+	u8 rsv3[11];
+};
+
+enum hclge_mac_vlan_cfg_sel {
+	HCLGE_MAC_VLAN_NIC_SEL = 0,
+	HCLGE_MAC_VLAN_ROCE_SEL,
+};
+
+#define HCLGE_ACCEPT_TAG1_B		0
+#define HCLGE_ACCEPT_UNTAG1_B		1
+#define HCLGE_PORT_INS_TAG1_EN_B	2
+#define HCLGE_PORT_INS_TAG2_EN_B	3
+#define HCLGE_CFG_NIC_ROCE_SEL_B	4
+#define HCLGE_ACCEPT_TAG2_B		5
+#define HCLGE_ACCEPT_UNTAG2_B		6
+#define HCLGE_TAG_SHIFT_MODE_EN_B	7
+#define HCLGE_VF_NUM_PER_BYTE		8
+
+struct hclge_vport_vtag_tx_cfg_cmd {
+	u8 vport_vlan_cfg;
+	u8 vf_offset;
+	u8 rsv1[2];
+	__le16 def_vlan_tag1;
+	__le16 def_vlan_tag2;
+	u8 vf_bitmap[HCLGE_VF_NUM_PER_BYTE];
+	u8 rsv2[8];
+};
+
+#define HCLGE_REM_TAG1_EN_B		0
+#define HCLGE_REM_TAG2_EN_B		1
+#define HCLGE_SHOW_TAG1_EN_B		2
+#define HCLGE_SHOW_TAG2_EN_B		3
+#define HCLGE_DISCARD_TAG1_EN_B		5
+#define HCLGE_DISCARD_TAG2_EN_B		6
+struct hclge_vport_vtag_rx_cfg_cmd {
+	u8 vport_vlan_cfg;
+	u8 vf_offset;
+	u8 rsv1[6];
+	u8 vf_bitmap[HCLGE_VF_NUM_PER_BYTE];
+	u8 rsv2[8];
+};
+
+struct hclge_tx_vlan_type_cfg_cmd {
+	__le16 ot_vlan_type;
+	__le16 in_vlan_type;
+	u8 rsv[20];
+};
+
+struct hclge_rx_vlan_type_cfg_cmd {
+	__le16 ot_fst_vlan_type;
+	__le16 ot_sec_vlan_type;
+	__le16 in_fst_vlan_type;
+	__le16 in_sec_vlan_type;
+	u8 rsv[16];
+};
+
+struct hclge_cfg_com_tqp_queue_cmd {
+	__le16 tqp_id;
+	__le16 stream_id;
+	u8 enable;
+	u8 rsv[19];
+};
+
+struct hclge_cfg_tx_queue_pointer_cmd {
+	__le16 tqp_id;
+	__le16 tx_tail;
+	__le16 tx_head;
+	__le16 fbd_num;
+	__le16 ring_offset;
+	u8 rsv[14];
+};
+
+#pragma pack(1)
+struct hclge_mac_ethertype_idx_rd_cmd {
+	u8	flags;
+	u8	resp_code;
+	__le16  vlan_tag;
+	u8      mac_addr[ETH_ALEN];
+	__le16  index;
+	__le16	ethter_type;
+	__le16  egress_port;
+	__le16  egress_queue;
+	__le16  rev0;
+	u8	i_port_bitmap;
+	u8	i_port_direction;
+	u8	rev1[2];
+};
+
+#pragma pack()
+
+#define HCLGE_TSO_MSS_MIN_S	0
+#define HCLGE_TSO_MSS_MIN_M	GENMASK(13, 0)
+
+#define HCLGE_TSO_MSS_MAX_S	16
+#define HCLGE_TSO_MSS_MAX_M	GENMASK(29, 16)
+
+struct hclge_cfg_tso_status_cmd {
+	__le16 tso_mss_min;
+	__le16 tso_mss_max;
+	u8 rsv[20];
+};
+
+#define HCLGE_GRO_EN_B		0
+struct hclge_cfg_gro_status_cmd {
+	u8 gro_en;
+	u8 rsv[23];
+};
+
+#define HCLGE_TSO_MSS_MIN	256
+#define HCLGE_TSO_MSS_MAX	9668
+
+#define HCLGE_TQP_RESET_B	0
+struct hclge_reset_tqp_queue_cmd {
+	__le16 tqp_id;
+	u8 reset_req;
+	u8 ready_to_reset;
+	u8 rsv[20];
+};
+
+#define HCLGE_CFG_RESET_MAC_B		3
+#define HCLGE_CFG_RESET_FUNC_B		7
+#define HCLGE_CFG_RESET_RCB_B		1
+struct hclge_reset_cmd {
+	u8 mac_func_reset;
+	u8 fun_reset_vfid;
+	u8 fun_reset_rcb;
+	u8 rsv;
+	__le16 fun_reset_rcb_vqid_start;
+	__le16 fun_reset_rcb_vqid_num;
+	u8 fun_reset_rcb_return_status;
+	u8 rsv1[15];
+};
+
+#define HCLGE_PF_RESET_DONE_BIT		BIT(0)
+
+struct hclge_pf_rst_done_cmd {
+	u8 pf_rst_done;
+	u8 rsv[23];
+};
+
+#define HCLGE_CMD_SERDES_SERIAL_INNER_LOOP_B	BIT(0)
+#define HCLGE_CMD_SERDES_PARALLEL_INNER_LOOP_B	BIT(2)
+#define HCLGE_CMD_GE_PHY_INNER_LOOP_B		BIT(3)
+#define HCLGE_CMD_COMMON_LB_DONE_B		BIT(0)
+#define HCLGE_CMD_COMMON_LB_SUCCESS_B		BIT(1)
+struct hclge_common_lb_cmd {
+	u8 mask;
+	u8 enable;
+	u8 result;
+	u8 rsv[21];
+};
+
+#define HCLGE_DEFAULT_TX_BUF		0x4000	 /* 16k  bytes */
+#define HCLGE_TOTAL_PKT_BUF		0x108000 /* 1.03125M bytes */
+#define HCLGE_DEFAULT_DV		0xA000	 /* 40k byte */
+#define HCLGE_DEFAULT_NON_DCB_DV	0x7800	/* 30K byte */
+#define HCLGE_NON_DCB_ADDITIONAL_BUF	0x1400	/* 5120 byte */
+
+#define HCLGE_LED_LOCATE_STATE_S	0
+#define HCLGE_LED_LOCATE_STATE_M	GENMASK(1, 0)
+
+struct hclge_set_led_state_cmd {
+	u8 rsv1[3];
+	u8 locate_led_config;
+	u8 rsv2[20];
+};
+
+struct hclge_get_fd_mode_cmd {
+	u8 mode;
+	u8 enable;
+	u8 rsv[22];
+};
+
+struct hclge_get_fd_allocation_cmd {
+	__le32 stage1_entry_num;
+	__le32 stage2_entry_num;
+	__le16 stage1_counter_num;
+	__le16 stage2_counter_num;
+	u8 rsv[12];
+};
+
+struct hclge_set_fd_key_config_cmd {
+	u8 stage;
+	u8 key_select;
+	u8 inner_sipv6_word_en;
+	u8 inner_dipv6_word_en;
+	u8 outer_sipv6_word_en;
+	u8 outer_dipv6_word_en;
+	u8 rsv1[2];
+	__le32 tuple_mask;
+	__le32 meta_data_mask;
+	u8 rsv2[8];
+};
+
+#define HCLGE_FD_EPORT_SW_EN_B		0
+struct hclge_fd_tcam_config_1_cmd {
+	u8 stage;
+	u8 xy_sel;
+	u8 port_info;
+	u8 rsv1[1];
+	__le32 index;
+	u8 entry_vld;
+	u8 rsv2[7];
+	u8 tcam_data[8];
+};
+
+struct hclge_fd_tcam_config_2_cmd {
+	u8 tcam_data[24];
+};
+
+struct hclge_fd_tcam_config_3_cmd {
+	u8 tcam_data[20];
+	u8 rsv[4];
+};
+
+#define HCLGE_FD_AD_DROP_B		0
+#define HCLGE_FD_AD_DIRECT_QID_B	1
+#define HCLGE_FD_AD_QID_S		2
+#define HCLGE_FD_AD_QID_M		GENMASK(11, 2)
+#define HCLGE_FD_AD_USE_COUNTER_B	12
+#define HCLGE_FD_AD_COUNTER_NUM_S	13
+#define HCLGE_FD_AD_COUNTER_NUM_M	GENMASK(20, 13)
+#define HCLGE_FD_AD_NXT_STEP_B		20
+#define HCLGE_FD_AD_NXT_KEY_S		21
+#define HCLGE_FD_AD_NXT_KEY_M		GENMASK(25, 21)
+#define HCLGE_FD_AD_WR_RULE_ID_B	0
+#define HCLGE_FD_AD_RULE_ID_S		1
+#define HCLGE_FD_AD_RULE_ID_M		GENMASK(12, 1)
+#define HCLGE_FD_AD_TC_OVRD_B		16
+#define HCLGE_FD_AD_TC_SIZE_S		17
+#define HCLGE_FD_AD_TC_SIZE_M		GENMASK(20, 17)
+
+struct hclge_fd_ad_config_cmd {
+	u8 stage;
+	u8 rsv1[3];
+	__le32 index;
+	__le64 ad_data;
+	u8 rsv2[8];
+};
+
+struct hclge_fd_ad_cnt_read_cmd {
+	u8 rsv0[4];
+	__le16 index;
+	u8 rsv1[2];
+	__le64 cnt;
+	u8 rsv2[8];
+};
+
+#define HCLGE_FD_USER_DEF_OFT_S		0
+#define HCLGE_FD_USER_DEF_OFT_M		GENMASK(14, 0)
+#define HCLGE_FD_USER_DEF_EN_B		15
+struct hclge_fd_user_def_cfg_cmd {
+	__le16 ol2_cfg;
+	__le16 l2_cfg;
+	__le16 ol3_cfg;
+	__le16 l3_cfg;
+	__le16 ol4_cfg;
+	__le16 l4_cfg;
+	u8 rsv[12];
+};
+
+struct hclge_get_imp_bd_cmd {
+	__le32 bd_num;
+	u8 rsv[20];
+};
+
+struct hclge_query_ppu_pf_other_int_dfx_cmd {
+	__le16 over_8bd_no_fe_qid;
+	__le16 over_8bd_no_fe_vf_id;
+	__le16 tso_mss_cmp_min_err_qid;
+	__le16 tso_mss_cmp_min_err_vf_id;
+	__le16 tso_mss_cmp_max_err_qid;
+	__le16 tso_mss_cmp_max_err_vf_id;
+	__le16 tx_rd_fbd_poison_qid;
+	__le16 tx_rd_fbd_poison_vf_id;
+	__le16 rx_rd_fbd_poison_qid;
+	__le16 rx_rd_fbd_poison_vf_id;
+	u8 rsv[4];
+};
+
+#define HCLGE_SFP_INFO_CMD_NUM	6
+#define HCLGE_SFP_INFO_BD0_LEN	20
+#define HCLGE_SFP_INFO_BDX_LEN	24
+#define HCLGE_SFP_INFO_MAX_LEN \
+	(HCLGE_SFP_INFO_BD0_LEN + \
+	(HCLGE_SFP_INFO_CMD_NUM - 1) * HCLGE_SFP_INFO_BDX_LEN)
+
+struct hclge_sfp_info_bd0_cmd {
+	__le16 offset;
+	__le16 read_len;
+	u8 data[HCLGE_SFP_INFO_BD0_LEN];
+};
+
+#define HCLGE_QUERY_DEV_SPECS_BD_NUM		4
+
+struct hclge_dev_specs_0_cmd {
+	__le32 rsv0;
+	__le32 mac_entry_num;
+	__le32 mng_entry_num;
+	__le16 rss_ind_tbl_size;
+	__le16 rss_key_size;
+	__le16 int_ql_max;
+	u8 max_non_tso_bd_num;
+	u8 rsv1;
+	__le32 max_tm_rate;
+};
+
+#define HCLGE_DEF_MAX_INT_GL		0x1FE0U
+
+struct hclge_dev_specs_1_cmd {
+	__le16 max_frm_size;
+	__le16 max_qset_num;
+	__le16 max_int_gl;
+	u8 rsv0[2];
+	__le16 umv_size;
+	__le16 mc_mac_size;
+	u8 rsv1[6];
+	u8 tnl_num;
+	u8 hilink_version;
+	u8 rsv2[4];
+};
+
+/* mac speed type defined in firmware command */
+enum HCLGE_FIRMWARE_MAC_SPEED {
+	HCLGE_FW_MAC_SPEED_1G,
+	HCLGE_FW_MAC_SPEED_10G,
+	HCLGE_FW_MAC_SPEED_25G,
+	HCLGE_FW_MAC_SPEED_40G,
+	HCLGE_FW_MAC_SPEED_50G,
+	HCLGE_FW_MAC_SPEED_100G,
+	HCLGE_FW_MAC_SPEED_10M,
+	HCLGE_FW_MAC_SPEED_100M,
+	HCLGE_FW_MAC_SPEED_200G,
+};
+
+#define HCLGE_PHY_LINK_SETTING_BD_NUM		2
+
+struct hclge_phy_link_ksetting_0_cmd {
+	__le32 speed;
+	u8 duplex;
+	u8 autoneg;
+	u8 eth_tp_mdix;
+	u8 eth_tp_mdix_ctrl;
+	u8 port;
+	u8 transceiver;
+	u8 phy_address;
+	u8 rsv;
+	__le32 supported;
+	__le32 advertising;
+	__le32 lp_advertising;
+};
+
+struct hclge_phy_link_ksetting_1_cmd {
+	u8 master_slave_cfg;
+	u8 master_slave_state;
+	u8 rsv[22];
+};
+
+struct hclge_phy_reg_cmd {
+	__le16 reg_addr;
+	u8 rsv0[2];
+	__le16 reg_val;
+	u8 rsv1[18];
+};
+
+struct hclge_wol_cfg_cmd {
+	__le32 wake_on_lan_mode;
+	u8 sopass[SOPASS_MAX];
+	u8 sopass_size;
+	u8 rsv[13];
+};
+
+struct hclge_query_wol_supported_cmd {
+	__le32 supported_wake_mode;
+	u8 rsv[20];
+};
+
+struct hclge_hw;
+int hclge_cmd_send(struct hclge_hw *hw, struct hclge_desc *desc, int num);
+#endif
diff --git a/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_dcb.c b/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_dcb.c
new file mode 100644
index 000000000000..eabbacb1c714
--- /dev/null
+++ b/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_dcb.c
@@ -0,0 +1,682 @@
+// SPDX-License-Identifier: GPL-2.0+
+// Copyright (c) 2016-2017 Hisilicon Limited.
+
+#include "hclge_main.h"
+#include "hclge_dcb.h"
+#include "hclge_tm.h"
+#include "hnae3.h"
+
+#define BW_PERCENT	100
+
+static int hclge_ieee_ets_to_tm_info(struct hclge_dev *hdev,
+				     struct ieee_ets *ets)
+{
+	u8 i;
+
+	for (i = 0; i < HNAE3_MAX_TC; i++) {
+		switch (ets->tc_tsa[i]) {
+		case IEEE_8021QAZ_TSA_STRICT:
+			hdev->tm_info.tc_info[i].tc_sch_mode =
+				HCLGE_SCH_MODE_SP;
+			hdev->tm_info.pg_info[0].tc_dwrr[i] = 0;
+			break;
+		case IEEE_8021QAZ_TSA_ETS:
+			hdev->tm_info.tc_info[i].tc_sch_mode =
+				HCLGE_SCH_MODE_DWRR;
+			hdev->tm_info.pg_info[0].tc_dwrr[i] =
+				ets->tc_tx_bw[i];
+			break;
+		default:
+			/* Hardware only supports SP (strict priority)
+			 * or ETS (enhanced transmission selection)
+			 * algorithms, if we receive some other value
+			 * from dcbnl, then throw an error.
+			 */
+			return -EINVAL;
+		}
+	}
+
+	hclge_tm_prio_tc_info_update(hdev, ets->prio_tc);
+
+	return 0;
+}
+
+static void hclge_tm_info_to_ieee_ets(struct hclge_dev *hdev,
+				      struct ieee_ets *ets)
+{
+	u32 i;
+
+	memset(ets, 0, sizeof(*ets));
+	ets->willing = 1;
+	ets->ets_cap = hdev->tc_max;
+
+	for (i = 0; i < HNAE3_MAX_TC; i++) {
+		ets->prio_tc[i] = hdev->tm_info.prio_tc[i];
+		if (i < hdev->tm_info.num_tc)
+			ets->tc_tx_bw[i] = hdev->tm_info.pg_info[0].tc_dwrr[i];
+		else
+			ets->tc_tx_bw[i] = 0;
+
+		if (hdev->tm_info.tc_info[i].tc_sch_mode ==
+		    HCLGE_SCH_MODE_SP)
+			ets->tc_tsa[i] = IEEE_8021QAZ_TSA_STRICT;
+		else
+			ets->tc_tsa[i] = IEEE_8021QAZ_TSA_ETS;
+	}
+}
+
+/* IEEE std */
+static int hclge_ieee_getets(struct hnae3_handle *h, struct ieee_ets *ets)
+{
+	struct hclge_vport *vport = hclge_get_vport(h);
+	struct hclge_dev *hdev = vport->back;
+
+	hclge_tm_info_to_ieee_ets(hdev, ets);
+
+	return 0;
+}
+
+static int hclge_dcb_common_validate(struct hclge_dev *hdev, u8 num_tc,
+				     u8 *prio_tc)
+{
+	int i;
+
+	if (num_tc > hdev->tc_max) {
+		dev_err(&hdev->pdev->dev,
+			"tc num checking failed, %u > tc_max(%u)\n",
+			num_tc, hdev->tc_max);
+		return -EINVAL;
+	}
+
+	for (i = 0; i < HNAE3_MAX_USER_PRIO; i++) {
+		if (prio_tc[i] >= num_tc) {
+			dev_err(&hdev->pdev->dev,
+				"prio_tc[%d] checking failed, %u >= num_tc(%u)\n",
+				i, prio_tc[i], num_tc);
+			return -EINVAL;
+		}
+	}
+
+	if (num_tc > hdev->vport[0].alloc_tqps) {
+		dev_err(&hdev->pdev->dev,
+			"allocated tqp checking failed, %u > tqp(%u)\n",
+			num_tc, hdev->vport[0].alloc_tqps);
+		return -EINVAL;
+	}
+
+	return 0;
+}
+
+static u8 hclge_ets_tc_changed(struct hclge_dev *hdev, struct ieee_ets *ets,
+			       bool *changed)
+{
+	u8 max_tc_id = 0;
+	u8 i;
+
+	for (i = 0; i < HNAE3_MAX_USER_PRIO; i++) {
+		if (ets->prio_tc[i] != hdev->tm_info.prio_tc[i])
+			*changed = true;
+
+		if (ets->prio_tc[i] > max_tc_id)
+			max_tc_id = ets->prio_tc[i];
+	}
+
+	/* return max tc number, max tc id need to plus 1 */
+	return max_tc_id + 1;
+}
+
+static int hclge_ets_sch_mode_validate(struct hclge_dev *hdev,
+				       struct ieee_ets *ets, bool *changed,
+				       u8 tc_num)
+{
+	bool has_ets_tc = false;
+	u32 total_ets_bw = 0;
+	u8 i;
+
+	for (i = 0; i < HNAE3_MAX_TC; i++) {
+		switch (ets->tc_tsa[i]) {
+		case IEEE_8021QAZ_TSA_STRICT:
+			if (hdev->tm_info.tc_info[i].tc_sch_mode !=
+				HCLGE_SCH_MODE_SP)
+				*changed = true;
+			break;
+		case IEEE_8021QAZ_TSA_ETS:
+			if (i >= tc_num) {
+				dev_err(&hdev->pdev->dev,
+					"tc%u is disabled, cannot set ets bw\n",
+					i);
+				return -EINVAL;
+			}
+
+			/* The hardware will switch to sp mode if bandwidth is
+			 * 0, so limit ets bandwidth must be greater than 0.
+			 */
+			if (!ets->tc_tx_bw[i]) {
+				dev_err(&hdev->pdev->dev,
+					"tc%u ets bw cannot be 0\n", i);
+				return -EINVAL;
+			}
+
+			if (hdev->tm_info.tc_info[i].tc_sch_mode !=
+				HCLGE_SCH_MODE_DWRR)
+				*changed = true;
+
+			total_ets_bw += ets->tc_tx_bw[i];
+			has_ets_tc = true;
+			break;
+		default:
+			return -EINVAL;
+		}
+	}
+
+	if (has_ets_tc && total_ets_bw != BW_PERCENT)
+		return -EINVAL;
+
+	return 0;
+}
+
+static int hclge_ets_validate(struct hclge_dev *hdev, struct ieee_ets *ets,
+			      u8 *tc, bool *changed)
+{
+	u8 tc_num;
+	int ret;
+
+	tc_num = hclge_ets_tc_changed(hdev, ets, changed);
+
+	ret = hclge_dcb_common_validate(hdev, tc_num, ets->prio_tc);
+	if (ret)
+		return ret;
+
+	ret = hclge_ets_sch_mode_validate(hdev, ets, changed, tc_num);
+	if (ret)
+		return ret;
+
+	*tc = tc_num;
+	if (*tc != hdev->tm_info.num_tc)
+		*changed = true;
+
+	return 0;
+}
+
+static int hclge_map_update(struct hclge_dev *hdev)
+{
+	int ret;
+
+	ret = hclge_tm_schd_setup_hw(hdev);
+	if (ret)
+		return ret;
+
+	ret = hclge_pause_setup_hw(hdev, false);
+	if (ret)
+		return ret;
+
+	ret = hclge_buffer_alloc(hdev);
+	if (ret)
+		return ret;
+
+	hclge_comm_rss_indir_init_cfg(hdev->ae_dev, &hdev->rss_cfg);
+
+	return hclge_rss_init_hw(hdev);
+}
+
+static int hclge_notify_down_uinit(struct hclge_dev *hdev)
+{
+	int ret;
+
+	ret = hclge_notify_client(hdev, HNAE3_DOWN_CLIENT);
+	if (ret)
+		return ret;
+
+	ret = hclge_tm_flush_cfg(hdev, true);
+	if (ret)
+		return ret;
+
+	return hclge_notify_client(hdev, HNAE3_UNINIT_CLIENT);
+}
+
+static int hclge_notify_init_up(struct hclge_dev *hdev)
+{
+	int ret;
+
+	ret = hclge_notify_client(hdev, HNAE3_INIT_CLIENT);
+	if (ret)
+		return ret;
+
+	ret = hclge_tm_flush_cfg(hdev, false);
+	if (ret)
+		return ret;
+
+	return hclge_notify_client(hdev, HNAE3_UP_CLIENT);
+}
+
+static int hclge_ieee_setets(struct hnae3_handle *h, struct ieee_ets *ets)
+{
+	struct hclge_vport *vport = hclge_get_vport(h);
+	struct net_device *netdev = h->kinfo.netdev;
+	struct hclge_dev *hdev = vport->back;
+	bool map_changed = false;
+	u8 num_tc = 0;
+	int ret;
+
+	if (!(hdev->dcbx_cap & DCB_CAP_DCBX_VER_IEEE) ||
+	    h->kinfo.tc_info.mqprio_active)
+		return -EINVAL;
+
+	ret = hclge_ets_validate(hdev, ets, &num_tc, &map_changed);
+	if (ret)
+		return ret;
+
+	if (map_changed) {
+		netif_dbg(h, drv, netdev, "set ets\n");
+
+		ret = hclge_notify_down_uinit(hdev);
+		if (ret)
+			return ret;
+	}
+
+	hclge_tm_schd_info_update(hdev, num_tc);
+	h->kinfo.tc_info.dcb_ets_active = num_tc > 1;
+
+	ret = hclge_ieee_ets_to_tm_info(hdev, ets);
+	if (ret)
+		goto err_out;
+
+	if (map_changed) {
+		ret = hclge_map_update(hdev);
+		if (ret)
+			goto err_out;
+
+		return hclge_notify_init_up(hdev);
+	}
+
+	return hclge_tm_dwrr_cfg(hdev);
+
+err_out:
+	if (!map_changed)
+		return ret;
+
+	hclge_notify_init_up(hdev);
+
+	return ret;
+}
+
+static int hclge_ieee_getpfc(struct hnae3_handle *h, struct ieee_pfc *pfc)
+{
+	struct hclge_vport *vport = hclge_get_vport(h);
+	struct hclge_dev *hdev = vport->back;
+	int ret;
+
+	memset(pfc, 0, sizeof(*pfc));
+	pfc->pfc_cap = hdev->pfc_max;
+	pfc->pfc_en = hdev->tm_info.pfc_en;
+
+	ret = hclge_mac_update_stats(hdev);
+	if (ret) {
+		dev_err(&hdev->pdev->dev,
+			"failed to update MAC stats, ret = %d.\n", ret);
+		return ret;
+	}
+
+	hclge_pfc_tx_stats_get(hdev, pfc->requests);
+	hclge_pfc_rx_stats_get(hdev, pfc->indications);
+
+	return 0;
+}
+
+static int hclge_ieee_setpfc(struct hnae3_handle *h, struct ieee_pfc *pfc)
+{
+	struct hclge_vport *vport = hclge_get_vport(h);
+	struct net_device *netdev = h->kinfo.netdev;
+	struct hclge_dev *hdev = vport->back;
+	u8 i, j, pfc_map, *prio_tc;
+	int last_bad_ret = 0;
+	int ret;
+
+	if (!(hdev->dcbx_cap & DCB_CAP_DCBX_VER_IEEE))
+		return -EINVAL;
+
+	if (pfc->pfc_en == hdev->tm_info.pfc_en)
+		return 0;
+
+	prio_tc = hdev->tm_info.prio_tc;
+	pfc_map = 0;
+
+	for (i = 0; i < hdev->tm_info.num_tc; i++) {
+		for (j = 0; j < HNAE3_MAX_USER_PRIO; j++) {
+			if ((prio_tc[j] == i) && (pfc->pfc_en & BIT(j))) {
+				pfc_map |= BIT(i);
+				break;
+			}
+		}
+	}
+
+	hdev->tm_info.hw_pfc_map = pfc_map;
+	hdev->tm_info.pfc_en = pfc->pfc_en;
+
+	netif_dbg(h, drv, netdev,
+		  "set pfc: pfc_en=%x, pfc_map=%x, num_tc=%u\n",
+		  pfc->pfc_en, pfc_map, hdev->tm_info.num_tc);
+
+	hclge_tm_pfc_info_update(hdev);
+
+	ret = hclge_pause_setup_hw(hdev, false);
+	if (ret)
+		return ret;
+
+	ret = hclge_notify_client(hdev, HNAE3_DOWN_CLIENT);
+	if (ret)
+		return ret;
+
+	ret = hclge_tm_flush_cfg(hdev, true);
+	if (ret)
+		return ret;
+
+	/* No matter whether the following operations are performed
+	 * successfully or not, disabling the tm flush and notify
+	 * the network status to up are necessary.
+	 * Do not return immediately.
+	 */
+	ret = hclge_buffer_alloc(hdev);
+	if (ret)
+		last_bad_ret = ret;
+
+	ret = hclge_tm_flush_cfg(hdev, false);
+	if (ret)
+		last_bad_ret = ret;
+
+	ret = hclge_notify_client(hdev, HNAE3_UP_CLIENT);
+	if (ret)
+		last_bad_ret = ret;
+
+	return last_bad_ret;
+}
+
+static int hclge_ieee_setapp(struct hnae3_handle *h, struct dcb_app *app)
+{
+	struct hclge_vport *vport = hclge_get_vport(h);
+	struct net_device *netdev = h->kinfo.netdev;
+	struct hclge_dev *hdev = vport->back;
+	struct dcb_app old_app;
+	int ret;
+
+	if (app->selector != IEEE_8021QAZ_APP_SEL_DSCP ||
+	    app->protocol >= HNAE3_MAX_DSCP ||
+	    app->priority >= HNAE3_MAX_USER_PRIO)
+		return -EINVAL;
+
+	dev_info(&hdev->pdev->dev, "setapp dscp=%u priority=%u\n",
+		 app->protocol, app->priority);
+
+	if (app->priority == h->kinfo.dscp_prio[app->protocol])
+		return 0;
+
+	ret = dcb_ieee_setapp(netdev, app);
+	if (ret)
+		return ret;
+
+	old_app.selector = IEEE_8021QAZ_APP_SEL_DSCP;
+	old_app.protocol = app->protocol;
+	old_app.priority = h->kinfo.dscp_prio[app->protocol];
+
+	h->kinfo.dscp_prio[app->protocol] = app->priority;
+	ret = hclge_dscp_to_tc_map(hdev);
+	if (ret) {
+		dev_err(&hdev->pdev->dev,
+			"failed to set dscp to tc map, ret = %d\n", ret);
+		h->kinfo.dscp_prio[app->protocol] = old_app.priority;
+		(void)dcb_ieee_delapp(netdev, app);
+		return ret;
+	}
+
+	vport->nic.kinfo.tc_map_mode = HNAE3_TC_MAP_MODE_DSCP;
+	if (old_app.priority == HNAE3_PRIO_ID_INVALID)
+		h->kinfo.dscp_app_cnt++;
+	else
+		ret = dcb_ieee_delapp(netdev, &old_app);
+
+	return ret;
+}
+
+static int hclge_ieee_delapp(struct hnae3_handle *h, struct dcb_app *app)
+{
+	struct hclge_vport *vport = hclge_get_vport(h);
+	struct net_device *netdev = h->kinfo.netdev;
+	struct hclge_dev *hdev = vport->back;
+	int ret;
+
+	if (app->selector != IEEE_8021QAZ_APP_SEL_DSCP ||
+	    app->protocol >= HNAE3_MAX_DSCP ||
+	    app->priority >= HNAE3_MAX_USER_PRIO ||
+	    app->priority != h->kinfo.dscp_prio[app->protocol])
+		return -EINVAL;
+
+	dev_info(&hdev->pdev->dev, "delapp dscp=%u priority=%u\n",
+		 app->protocol, app->priority);
+
+	ret = dcb_ieee_delapp(netdev, app);
+	if (ret)
+		return ret;
+
+	h->kinfo.dscp_prio[app->protocol] = HNAE3_PRIO_ID_INVALID;
+	ret = hclge_dscp_to_tc_map(hdev);
+	if (ret) {
+		dev_err(&hdev->pdev->dev,
+			"failed to del dscp to tc map, ret = %d\n", ret);
+		h->kinfo.dscp_prio[app->protocol] = app->priority;
+		(void)dcb_ieee_setapp(netdev, app);
+		return ret;
+	}
+
+	if (h->kinfo.dscp_app_cnt)
+		h->kinfo.dscp_app_cnt--;
+
+	if (!h->kinfo.dscp_app_cnt) {
+		vport->nic.kinfo.tc_map_mode = HNAE3_TC_MAP_MODE_PRIO;
+		ret = hclge_up_to_tc_map(hdev);
+	}
+
+	return ret;
+}
+
+/* DCBX configuration */
+static u8 hclge_getdcbx(struct hnae3_handle *h)
+{
+	struct hclge_vport *vport = hclge_get_vport(h);
+	struct hclge_dev *hdev = vport->back;
+
+	if (h->kinfo.tc_info.mqprio_active)
+		return 0;
+
+	return hdev->dcbx_cap;
+}
+
+static u8 hclge_setdcbx(struct hnae3_handle *h, u8 mode)
+{
+	struct hclge_vport *vport = hclge_get_vport(h);
+	struct net_device *netdev = h->kinfo.netdev;
+	struct hclge_dev *hdev = vport->back;
+
+	netif_dbg(h, drv, netdev, "set dcbx: mode=%u\n", mode);
+
+	/* No support for LLD_MANAGED modes or CEE */
+	if ((mode & DCB_CAP_DCBX_LLD_MANAGED) ||
+	    (mode & DCB_CAP_DCBX_VER_CEE) ||
+	    !(mode & DCB_CAP_DCBX_HOST))
+		return 1;
+
+	hdev->dcbx_cap = mode;
+
+	return 0;
+}
+
+static int hclge_mqprio_qopt_check(struct hclge_dev *hdev,
+				   struct tc_mqprio_qopt_offload *mqprio_qopt)
+{
+	u16 queue_sum = 0;
+	int ret;
+	int i;
+
+	if (!mqprio_qopt->qopt.num_tc) {
+		mqprio_qopt->qopt.num_tc = 1;
+		return 0;
+	}
+
+	ret = hclge_dcb_common_validate(hdev, mqprio_qopt->qopt.num_tc,
+					mqprio_qopt->qopt.prio_tc_map);
+	if (ret)
+		return ret;
+
+	for (i = 0; i < mqprio_qopt->qopt.num_tc; i++) {
+		if (!is_power_of_2(mqprio_qopt->qopt.count[i])) {
+			dev_err(&hdev->pdev->dev,
+				"qopt queue count must be power of 2\n");
+			return -EINVAL;
+		}
+
+		if (mqprio_qopt->qopt.count[i] > hdev->pf_rss_size_max) {
+			dev_err(&hdev->pdev->dev,
+				"qopt queue count should be no more than %u\n",
+				hdev->pf_rss_size_max);
+			return -EINVAL;
+		}
+
+		if (mqprio_qopt->qopt.offset[i] != queue_sum) {
+			dev_err(&hdev->pdev->dev,
+				"qopt queue offset must start from 0, and being continuous\n");
+			return -EINVAL;
+		}
+
+		if (mqprio_qopt->min_rate[i] || mqprio_qopt->max_rate[i]) {
+			dev_err(&hdev->pdev->dev,
+				"qopt tx_rate is not supported\n");
+			return -EOPNOTSUPP;
+		}
+
+		queue_sum = mqprio_qopt->qopt.offset[i];
+		queue_sum += mqprio_qopt->qopt.count[i];
+	}
+	if (hdev->vport[0].alloc_tqps < queue_sum) {
+		dev_err(&hdev->pdev->dev,
+			"qopt queue count sum should be less than %u\n",
+			hdev->vport[0].alloc_tqps);
+		return -EINVAL;
+	}
+
+	return 0;
+}
+
+static void hclge_sync_mqprio_qopt(struct hnae3_tc_info *tc_info,
+				   struct tc_mqprio_qopt_offload *mqprio_qopt)
+{
+	memset(tc_info, 0, sizeof(*tc_info));
+	tc_info->num_tc = mqprio_qopt->qopt.num_tc;
+	memcpy(tc_info->prio_tc, mqprio_qopt->qopt.prio_tc_map,
+	       sizeof_field(struct hnae3_tc_info, prio_tc));
+	memcpy(tc_info->tqp_count, mqprio_qopt->qopt.count,
+	       sizeof_field(struct hnae3_tc_info, tqp_count));
+	memcpy(tc_info->tqp_offset, mqprio_qopt->qopt.offset,
+	       sizeof_field(struct hnae3_tc_info, tqp_offset));
+}
+
+static int hclge_config_tc(struct hclge_dev *hdev,
+			   struct hnae3_tc_info *tc_info)
+{
+	int i;
+
+	hclge_tm_schd_info_update(hdev, tc_info->num_tc);
+	for (i = 0; i < HNAE3_MAX_USER_PRIO; i++)
+		hdev->tm_info.prio_tc[i] = tc_info->prio_tc[i];
+
+	return hclge_map_update(hdev);
+}
+
+/* Set up TC for hardware offloaded mqprio in channel mode */
+static int hclge_setup_tc(struct hnae3_handle *h,
+			  struct tc_mqprio_qopt_offload *mqprio_qopt)
+{
+	struct hclge_vport *vport = hclge_get_vport(h);
+	struct hnae3_knic_private_info *kinfo;
+	struct hclge_dev *hdev = vport->back;
+	struct hnae3_tc_info old_tc_info;
+	u8 tc = mqprio_qopt->qopt.num_tc;
+	int ret;
+
+	/* if client unregistered, it's not allowed to change
+	 * mqprio configuration, which may cause uninit ring
+	 * fail.
+	 */
+	if (!test_bit(HCLGE_STATE_NIC_REGISTERED, &hdev->state))
+		return -EBUSY;
+
+	kinfo = &vport->nic.kinfo;
+	if (kinfo->tc_info.dcb_ets_active)
+		return -EINVAL;
+
+	ret = hclge_mqprio_qopt_check(hdev, mqprio_qopt);
+	if (ret) {
+		dev_err(&hdev->pdev->dev,
+			"failed to check mqprio qopt params, ret = %d\n", ret);
+		return ret;
+	}
+
+	kinfo->tc_info.mqprio_destroy = !tc;
+
+	ret = hclge_notify_down_uinit(hdev);
+	if (ret)
+		return ret;
+
+	memcpy(&old_tc_info, &kinfo->tc_info, sizeof(old_tc_info));
+	hclge_sync_mqprio_qopt(&kinfo->tc_info, mqprio_qopt);
+	kinfo->tc_info.mqprio_active = tc > 0;
+
+	ret = hclge_config_tc(hdev, &kinfo->tc_info);
+	if (ret)
+		goto err_out;
+
+	return hclge_notify_init_up(hdev);
+
+err_out:
+	if (!tc) {
+		dev_warn(&hdev->pdev->dev,
+			 "failed to destroy mqprio, will active after reset, ret = %d\n",
+			 ret);
+	} else {
+		/* roll-back */
+		memcpy(&kinfo->tc_info, &old_tc_info, sizeof(old_tc_info));
+		if (hclge_config_tc(hdev, &kinfo->tc_info))
+			dev_err(&hdev->pdev->dev,
+				"failed to roll back tc configuration\n");
+	}
+	hclge_notify_init_up(hdev);
+
+	return ret;
+}
+
+static const struct hnae3_dcb_ops hns3_dcb_ops = {
+	.ieee_getets	= hclge_ieee_getets,
+	.ieee_setets	= hclge_ieee_setets,
+	.ieee_getpfc	= hclge_ieee_getpfc,
+	.ieee_setpfc	= hclge_ieee_setpfc,
+	.ieee_setapp    = hclge_ieee_setapp,
+	.ieee_delapp    = hclge_ieee_delapp,
+	.getdcbx	= hclge_getdcbx,
+	.setdcbx	= hclge_setdcbx,
+	.setup_tc	= hclge_setup_tc,
+};
+
+void hclge_dcb_ops_set(struct hclge_dev *hdev)
+{
+	struct hclge_vport *vport = hdev->vport;
+	struct hnae3_knic_private_info *kinfo;
+
+	/* Hdev does not support DCB or vport is
+	 * not a pf, then dcb_ops is not set.
+	 */
+	if (!hnae3_dev_dcb_supported(hdev) ||
+	    vport->vport_id != 0)
+		return;
+
+	kinfo = &vport->nic.kinfo;
+	kinfo->dcb_ops = &hns3_dcb_ops;
+	hdev->dcbx_cap = DCB_CAP_DCBX_VER_IEEE | DCB_CAP_DCBX_HOST;
+}
diff --git a/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_dcb.h b/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_dcb.h
new file mode 100644
index 000000000000..b04702e65689
--- /dev/null
+++ b/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_dcb.h
@@ -0,0 +1,15 @@
+/* SPDX-License-Identifier: GPL-2.0+ */
+// Copyright (c) 2016-2017 Hisilicon Limited.
+
+#ifndef __HCLGE_DCB_H__
+#define __HCLGE_DCB_H__
+
+#include "hclge_main.h"
+
+#ifdef CONFIG_HNS3_DCB
+void hclge_dcb_ops_set(struct hclge_dev *hdev);
+#else
+static inline void hclge_dcb_ops_set(struct hclge_dev *hdev) {}
+#endif
+
+#endif /* __HCLGE_DCB_H__ */
diff --git a/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_debugfs.c b/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_debugfs.c
new file mode 100644
index 000000000000..b76d25074e99
--- /dev/null
+++ b/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_debugfs.c
@@ -0,0 +1,2956 @@
+// SPDX-License-Identifier: GPL-2.0+
+/* Copyright (c) 2018-2019 Hisilicon Limited. */
+
+#include <linux/device.h>
+#include <linux/sched/clock.h>
+#include <linux/string_choices.h>
+
+#include "hclge_debugfs.h"
+#include "hclge_err.h"
+#include "hclge_main.h"
+#include "hclge_regs.h"
+#include "hclge_tm.h"
+#include "hnae3.h"
+
+#define hclge_seq_file_to_hdev(s)	\
+		(((struct hnae3_ae_dev *)hnae3_seq_file_to_ae_dev(s))->priv)
+
+static const char * const hclge_mac_state_str[] = {
+	"TO_ADD", "TO_DEL", "ACTIVE"
+};
+
+static const char * const tc_map_mode_str[] = { "PRIO", "DSCP" };
+
+static const struct hclge_dbg_dfx_message hclge_dbg_bios_common_reg[] = {
+	{false, "Reserved"},
+	{true,	"BP_CPU_STATE"},
+	{true,	"DFX_MSIX_INFO_NIC_0"},
+	{true,	"DFX_MSIX_INFO_NIC_1"},
+	{true,	"DFX_MSIX_INFO_NIC_2"},
+	{true,	"DFX_MSIX_INFO_NIC_3"},
+
+	{true,	"DFX_MSIX_INFO_ROC_0"},
+	{true,	"DFX_MSIX_INFO_ROC_1"},
+	{true,	"DFX_MSIX_INFO_ROC_2"},
+	{true,	"DFX_MSIX_INFO_ROC_3"},
+	{false, "Reserved"},
+	{false, "Reserved"},
+};
+
+static const struct hclge_dbg_dfx_message hclge_dbg_ssu_reg_0[] = {
+	{false, "Reserved"},
+	{true,	"SSU_ETS_PORT_STATUS"},
+	{true,	"SSU_ETS_TCG_STATUS"},
+	{false, "Reserved"},
+	{false, "Reserved"},
+	{true,	"SSU_BP_STATUS_0"},
+
+	{true,	"SSU_BP_STATUS_1"},
+	{true,	"SSU_BP_STATUS_2"},
+	{true,	"SSU_BP_STATUS_3"},
+	{true,	"SSU_BP_STATUS_4"},
+	{true,	"SSU_BP_STATUS_5"},
+	{true,	"SSU_MAC_TX_PFC_IND"},
+
+	{true,	"MAC_SSU_RX_PFC_IND"},
+	{true,	"BTMP_AGEING_ST_B0"},
+	{true,	"BTMP_AGEING_ST_B1"},
+	{true,	"BTMP_AGEING_ST_B2"},
+	{false, "Reserved"},
+	{false, "Reserved"},
+
+	{true,	"FULL_DROP_NUM"},
+	{true,	"PART_DROP_NUM"},
+	{true,	"PPP_KEY_DROP_NUM"},
+	{true,	"PPP_RLT_DROP_NUM"},
+	{true,	"LO_PRI_UNICAST_RLT_DROP_NUM"},
+	{true,	"HI_PRI_MULTICAST_RLT_DROP_NUM"},
+
+	{true,	"LO_PRI_MULTICAST_RLT_DROP_NUM"},
+	{true,	"NCSI_PACKET_CURR_BUFFER_CNT"},
+	{true,	"BTMP_AGEING_RLS_CNT_BANK0"},
+	{true,	"BTMP_AGEING_RLS_CNT_BANK1"},
+	{true,	"BTMP_AGEING_RLS_CNT_BANK2"},
+	{true,	"SSU_MB_RD_RLT_DROP_CNT"},
+
+	{true,	"SSU_PPP_MAC_KEY_NUM_L"},
+	{true,	"SSU_PPP_MAC_KEY_NUM_H"},
+	{true,	"SSU_PPP_HOST_KEY_NUM_L"},
+	{true,	"SSU_PPP_HOST_KEY_NUM_H"},
+	{true,	"PPP_SSU_MAC_RLT_NUM_L"},
+	{true,	"PPP_SSU_MAC_RLT_NUM_H"},
+
+	{true,	"PPP_SSU_HOST_RLT_NUM_L"},
+	{true,	"PPP_SSU_HOST_RLT_NUM_H"},
+	{true,	"NCSI_RX_PACKET_IN_CNT_L"},
+	{true,	"NCSI_RX_PACKET_IN_CNT_H"},
+	{true,	"NCSI_TX_PACKET_OUT_CNT_L"},
+	{true,	"NCSI_TX_PACKET_OUT_CNT_H"},
+
+	{true,	"SSU_KEY_DROP_NUM"},
+	{true,	"MB_UNCOPY_NUM"},
+	{true,	"RX_OQ_DROP_PKT_CNT"},
+	{true,	"TX_OQ_DROP_PKT_CNT"},
+	{true,	"BANK_UNBALANCE_DROP_CNT"},
+	{true,	"BANK_UNBALANCE_RX_DROP_CNT"},
+
+	{true,	"NIC_L2_ERR_DROP_PKT_CNT"},
+	{true,	"ROC_L2_ERR_DROP_PKT_CNT"},
+	{true,	"NIC_L2_ERR_DROP_PKT_CNT_RX"},
+	{true,	"ROC_L2_ERR_DROP_PKT_CNT_RX"},
+	{true,	"RX_OQ_GLB_DROP_PKT_CNT"},
+	{false, "Reserved"},
+
+	{true,	"LO_PRI_UNICAST_CUR_CNT"},
+	{true,	"HI_PRI_MULTICAST_CUR_CNT"},
+	{true,	"LO_PRI_MULTICAST_CUR_CNT"},
+	{false, "Reserved"},
+	{false, "Reserved"},
+	{false, "Reserved"},
+};
+
+static const struct hclge_dbg_dfx_message hclge_dbg_ssu_reg_1[] = {
+	{true,	"prt_id"},
+	{true,	"PACKET_TC_CURR_BUFFER_CNT_0"},
+	{true,	"PACKET_TC_CURR_BUFFER_CNT_1"},
+	{true,	"PACKET_TC_CURR_BUFFER_CNT_2"},
+	{true,	"PACKET_TC_CURR_BUFFER_CNT_3"},
+	{true,	"PACKET_TC_CURR_BUFFER_CNT_4"},
+
+	{true,	"PACKET_TC_CURR_BUFFER_CNT_5"},
+	{true,	"PACKET_TC_CURR_BUFFER_CNT_6"},
+	{true,	"PACKET_TC_CURR_BUFFER_CNT_7"},
+	{true,	"PACKET_CURR_BUFFER_CNT"},
+	{false, "Reserved"},
+	{false, "Reserved"},
+
+	{true,	"RX_PACKET_IN_CNT_L"},
+	{true,	"RX_PACKET_IN_CNT_H"},
+	{true,	"RX_PACKET_OUT_CNT_L"},
+	{true,	"RX_PACKET_OUT_CNT_H"},
+	{true,	"TX_PACKET_IN_CNT_L"},
+	{true,	"TX_PACKET_IN_CNT_H"},
+
+	{true,	"TX_PACKET_OUT_CNT_L"},
+	{true,	"TX_PACKET_OUT_CNT_H"},
+	{true,	"ROC_RX_PACKET_IN_CNT_L"},
+	{true,	"ROC_RX_PACKET_IN_CNT_H"},
+	{true,	"ROC_TX_PACKET_OUT_CNT_L"},
+	{true,	"ROC_TX_PACKET_OUT_CNT_H"},
+
+	{true,	"RX_PACKET_TC_IN_CNT_0_L"},
+	{true,	"RX_PACKET_TC_IN_CNT_0_H"},
+	{true,	"RX_PACKET_TC_IN_CNT_1_L"},
+	{true,	"RX_PACKET_TC_IN_CNT_1_H"},
+	{true,	"RX_PACKET_TC_IN_CNT_2_L"},
+	{true,	"RX_PACKET_TC_IN_CNT_2_H"},
+
+	{true,	"RX_PACKET_TC_IN_CNT_3_L"},
+	{true,	"RX_PACKET_TC_IN_CNT_3_H"},
+	{true,	"RX_PACKET_TC_IN_CNT_4_L"},
+	{true,	"RX_PACKET_TC_IN_CNT_4_H"},
+	{true,	"RX_PACKET_TC_IN_CNT_5_L"},
+	{true,	"RX_PACKET_TC_IN_CNT_5_H"},
+
+	{true,	"RX_PACKET_TC_IN_CNT_6_L"},
+	{true,	"RX_PACKET_TC_IN_CNT_6_H"},
+	{true,	"RX_PACKET_TC_IN_CNT_7_L"},
+	{true,	"RX_PACKET_TC_IN_CNT_7_H"},
+	{true,	"RX_PACKET_TC_OUT_CNT_0_L"},
+	{true,	"RX_PACKET_TC_OUT_CNT_0_H"},
+
+	{true,	"RX_PACKET_TC_OUT_CNT_1_L"},
+	{true,	"RX_PACKET_TC_OUT_CNT_1_H"},
+	{true,	"RX_PACKET_TC_OUT_CNT_2_L"},
+	{true,	"RX_PACKET_TC_OUT_CNT_2_H"},
+	{true,	"RX_PACKET_TC_OUT_CNT_3_L"},
+	{true,	"RX_PACKET_TC_OUT_CNT_3_H"},
+
+	{true,	"RX_PACKET_TC_OUT_CNT_4_L"},
+	{true,	"RX_PACKET_TC_OUT_CNT_4_H"},
+	{true,	"RX_PACKET_TC_OUT_CNT_5_L"},
+	{true,	"RX_PACKET_TC_OUT_CNT_5_H"},
+	{true,	"RX_PACKET_TC_OUT_CNT_6_L"},
+	{true,	"RX_PACKET_TC_OUT_CNT_6_H"},
+
+	{true,	"RX_PACKET_TC_OUT_CNT_7_L"},
+	{true,	"RX_PACKET_TC_OUT_CNT_7_H"},
+	{true,	"TX_PACKET_TC_IN_CNT_0_L"},
+	{true,	"TX_PACKET_TC_IN_CNT_0_H"},
+	{true,	"TX_PACKET_TC_IN_CNT_1_L"},
+	{true,	"TX_PACKET_TC_IN_CNT_1_H"},
+
+	{true,	"TX_PACKET_TC_IN_CNT_2_L"},
+	{true,	"TX_PACKET_TC_IN_CNT_2_H"},
+	{true,	"TX_PACKET_TC_IN_CNT_3_L"},
+	{true,	"TX_PACKET_TC_IN_CNT_3_H"},
+	{true,	"TX_PACKET_TC_IN_CNT_4_L"},
+	{true,	"TX_PACKET_TC_IN_CNT_4_H"},
+
+	{true,	"TX_PACKET_TC_IN_CNT_5_L"},
+	{true,	"TX_PACKET_TC_IN_CNT_5_H"},
+	{true,	"TX_PACKET_TC_IN_CNT_6_L"},
+	{true,	"TX_PACKET_TC_IN_CNT_6_H"},
+	{true,	"TX_PACKET_TC_IN_CNT_7_L"},
+	{true,	"TX_PACKET_TC_IN_CNT_7_H"},
+
+	{true,	"TX_PACKET_TC_OUT_CNT_0_L"},
+	{true,	"TX_PACKET_TC_OUT_CNT_0_H"},
+	{true,	"TX_PACKET_TC_OUT_CNT_1_L"},
+	{true,	"TX_PACKET_TC_OUT_CNT_1_H"},
+	{true,	"TX_PACKET_TC_OUT_CNT_2_L"},
+	{true,	"TX_PACKET_TC_OUT_CNT_2_H"},
+
+	{true,	"TX_PACKET_TC_OUT_CNT_3_L"},
+	{true,	"TX_PACKET_TC_OUT_CNT_3_H"},
+	{true,	"TX_PACKET_TC_OUT_CNT_4_L"},
+	{true,	"TX_PACKET_TC_OUT_CNT_4_H"},
+	{true,	"TX_PACKET_TC_OUT_CNT_5_L"},
+	{true,	"TX_PACKET_TC_OUT_CNT_5_H"},
+
+	{true,	"TX_PACKET_TC_OUT_CNT_6_L"},
+	{true,	"TX_PACKET_TC_OUT_CNT_6_H"},
+	{true,	"TX_PACKET_TC_OUT_CNT_7_L"},
+	{true,	"TX_PACKET_TC_OUT_CNT_7_H"},
+	{false, "Reserved"},
+	{false, "Reserved"},
+};
+
+static const struct hclge_dbg_dfx_message hclge_dbg_ssu_reg_2[] = {
+	{true,	"OQ_INDEX"},
+	{true,	"QUEUE_CNT"},
+	{false, "Reserved"},
+	{false, "Reserved"},
+	{false, "Reserved"},
+	{false, "Reserved"},
+};
+
+static const struct hclge_dbg_dfx_message hclge_dbg_igu_egu_reg[] = {
+	{true,	"prt_id"},
+	{true,	"IGU_RX_ERR_PKT"},
+	{true,	"IGU_RX_NO_SOF_PKT"},
+	{true,	"EGU_TX_1588_SHORT_PKT"},
+	{true,	"EGU_TX_1588_PKT"},
+	{true,	"EGU_TX_ERR_PKT"},
+
+	{true,	"IGU_RX_OUT_L2_PKT"},
+	{true,	"IGU_RX_OUT_L3_PKT"},
+	{true,	"IGU_RX_OUT_L4_PKT"},
+	{true,	"IGU_RX_IN_L2_PKT"},
+	{true,	"IGU_RX_IN_L3_PKT"},
+	{true,	"IGU_RX_IN_L4_PKT"},
+
+	{true,	"IGU_RX_EL3E_PKT"},
+	{true,	"IGU_RX_EL4E_PKT"},
+	{true,	"IGU_RX_L3E_PKT"},
+	{true,	"IGU_RX_L4E_PKT"},
+	{true,	"IGU_RX_ROCEE_PKT"},
+	{true,	"IGU_RX_OUT_UDP0_PKT"},
+
+	{true,	"IGU_RX_IN_UDP0_PKT"},
+	{true,	"IGU_MC_CAR_DROP_PKT_L"},
+	{true,	"IGU_MC_CAR_DROP_PKT_H"},
+	{true,	"IGU_BC_CAR_DROP_PKT_L"},
+	{true,	"IGU_BC_CAR_DROP_PKT_H"},
+	{false, "Reserved"},
+
+	{true,	"IGU_RX_OVERSIZE_PKT_L"},
+	{true,	"IGU_RX_OVERSIZE_PKT_H"},
+	{true,	"IGU_RX_UNDERSIZE_PKT_L"},
+	{true,	"IGU_RX_UNDERSIZE_PKT_H"},
+	{true,	"IGU_RX_OUT_ALL_PKT_L"},
+	{true,	"IGU_RX_OUT_ALL_PKT_H"},
+
+	{true,	"IGU_TX_OUT_ALL_PKT_L"},
+	{true,	"IGU_TX_OUT_ALL_PKT_H"},
+	{true,	"IGU_RX_UNI_PKT_L"},
+	{true,	"IGU_RX_UNI_PKT_H"},
+	{true,	"IGU_RX_MULTI_PKT_L"},
+	{true,	"IGU_RX_MULTI_PKT_H"},
+
+	{true,	"IGU_RX_BROAD_PKT_L"},
+	{true,	"IGU_RX_BROAD_PKT_H"},
+	{true,	"EGU_TX_OUT_ALL_PKT_L"},
+	{true,	"EGU_TX_OUT_ALL_PKT_H"},
+	{true,	"EGU_TX_UNI_PKT_L"},
+	{true,	"EGU_TX_UNI_PKT_H"},
+
+	{true,	"EGU_TX_MULTI_PKT_L"},
+	{true,	"EGU_TX_MULTI_PKT_H"},
+	{true,	"EGU_TX_BROAD_PKT_L"},
+	{true,	"EGU_TX_BROAD_PKT_H"},
+	{true,	"IGU_TX_KEY_NUM_L"},
+	{true,	"IGU_TX_KEY_NUM_H"},
+
+	{true,	"IGU_RX_NON_TUN_PKT_L"},
+	{true,	"IGU_RX_NON_TUN_PKT_H"},
+	{true,	"IGU_RX_TUN_PKT_L"},
+	{true,	"IGU_RX_TUN_PKT_H"},
+	{false,	"Reserved"},
+	{false,	"Reserved"},
+};
+
+static const struct hclge_dbg_dfx_message hclge_dbg_rpu_reg_0[] = {
+	{true, "tc_queue_num"},
+	{true, "FSM_DFX_ST0"},
+	{true, "FSM_DFX_ST1"},
+	{true, "RPU_RX_PKT_DROP_CNT"},
+	{true, "BUF_WAIT_TIMEOUT"},
+	{true, "BUF_WAIT_TIMEOUT_QID"},
+};
+
+static const struct hclge_dbg_dfx_message hclge_dbg_rpu_reg_1[] = {
+	{false, "Reserved"},
+	{true,	"FIFO_DFX_ST0"},
+	{true,	"FIFO_DFX_ST1"},
+	{true,	"FIFO_DFX_ST2"},
+	{true,	"FIFO_DFX_ST3"},
+	{true,	"FIFO_DFX_ST4"},
+
+	{true,	"FIFO_DFX_ST5"},
+	{false, "Reserved"},
+	{false, "Reserved"},
+	{false, "Reserved"},
+	{false, "Reserved"},
+	{false, "Reserved"},
+};
+
+static const struct hclge_dbg_dfx_message hclge_dbg_ncsi_reg[] = {
+	{false, "Reserved"},
+	{true,	"NCSI_EGU_TX_FIFO_STS"},
+	{true,	"NCSI_PAUSE_STATUS"},
+	{true,	"NCSI_RX_CTRL_DMAC_ERR_CNT"},
+	{true,	"NCSI_RX_CTRL_SMAC_ERR_CNT"},
+	{true,	"NCSI_RX_CTRL_CKS_ERR_CNT"},
+
+	{true,	"NCSI_RX_CTRL_PKT_CNT"},
+	{true,	"NCSI_RX_PT_DMAC_ERR_CNT"},
+	{true,	"NCSI_RX_PT_SMAC_ERR_CNT"},
+	{true,	"NCSI_RX_PT_PKT_CNT"},
+	{true,	"NCSI_RX_FCS_ERR_CNT"},
+	{true,	"NCSI_TX_CTRL_DMAC_ERR_CNT"},
+
+	{true,	"NCSI_TX_CTRL_SMAC_ERR_CNT"},
+	{true,	"NCSI_TX_CTRL_PKT_CNT"},
+	{true,	"NCSI_TX_PT_DMAC_ERR_CNT"},
+	{true,	"NCSI_TX_PT_SMAC_ERR_CNT"},
+	{true,	"NCSI_TX_PT_PKT_CNT"},
+	{true,	"NCSI_TX_PT_PKT_TRUNC_CNT"},
+
+	{true,	"NCSI_TX_PT_PKT_ERR_CNT"},
+	{true,	"NCSI_TX_CTRL_PKT_ERR_CNT"},
+	{true,	"NCSI_RX_CTRL_PKT_TRUNC_CNT"},
+	{true,	"NCSI_RX_CTRL_PKT_CFLIT_CNT"},
+	{false, "Reserved"},
+	{false, "Reserved"},
+
+	{true,	"NCSI_MAC_RX_OCTETS_OK"},
+	{true,	"NCSI_MAC_RX_OCTETS_BAD"},
+	{true,	"NCSI_MAC_RX_UC_PKTS"},
+	{true,	"NCSI_MAC_RX_MC_PKTS"},
+	{true,	"NCSI_MAC_RX_BC_PKTS"},
+	{true,	"NCSI_MAC_RX_PKTS_64OCTETS"},
+
+	{true,	"NCSI_MAC_RX_PKTS_65TO127OCTETS"},
+	{true,	"NCSI_MAC_RX_PKTS_128TO255OCTETS"},
+	{true,	"NCSI_MAC_RX_PKTS_255TO511OCTETS"},
+	{true,	"NCSI_MAC_RX_PKTS_512TO1023OCTETS"},
+	{true,	"NCSI_MAC_RX_PKTS_1024TO1518OCTETS"},
+	{true,	"NCSI_MAC_RX_PKTS_1519TOMAXOCTETS"},
+
+	{true,	"NCSI_MAC_RX_FCS_ERRORS"},
+	{true,	"NCSI_MAC_RX_LONG_ERRORS"},
+	{true,	"NCSI_MAC_RX_JABBER_ERRORS"},
+	{true,	"NCSI_MAC_RX_RUNT_ERR_CNT"},
+	{true,	"NCSI_MAC_RX_SHORT_ERR_CNT"},
+	{true,	"NCSI_MAC_RX_FILT_PKT_CNT"},
+
+	{true,	"NCSI_MAC_RX_OCTETS_TOTAL_FILT"},
+	{true,	"NCSI_MAC_TX_OCTETS_OK"},
+	{true,	"NCSI_MAC_TX_OCTETS_BAD"},
+	{true,	"NCSI_MAC_TX_UC_PKTS"},
+	{true,	"NCSI_MAC_TX_MC_PKTS"},
+	{true,	"NCSI_MAC_TX_BC_PKTS"},
+
+	{true,	"NCSI_MAC_TX_PKTS_64OCTETS"},
+	{true,	"NCSI_MAC_TX_PKTS_65TO127OCTETS"},
+	{true,	"NCSI_MAC_TX_PKTS_128TO255OCTETS"},
+	{true,	"NCSI_MAC_TX_PKTS_256TO511OCTETS"},
+	{true,	"NCSI_MAC_TX_PKTS_512TO1023OCTETS"},
+	{true,	"NCSI_MAC_TX_PKTS_1024TO1518OCTETS"},
+
+	{true,	"NCSI_MAC_TX_PKTS_1519TOMAXOCTETS"},
+	{true,	"NCSI_MAC_TX_UNDERRUN"},
+	{true,	"NCSI_MAC_TX_CRC_ERROR"},
+	{true,	"NCSI_MAC_TX_PAUSE_FRAMES"},
+	{true,	"NCSI_MAC_RX_PAD_PKTS"},
+	{true,	"NCSI_MAC_RX_PAUSE_FRAMES"},
+};
+
+static const struct hclge_dbg_dfx_message hclge_dbg_rtc_reg[] = {
+	{false, "Reserved"},
+	{true,	"LGE_IGU_AFIFO_DFX_0"},
+	{true,	"LGE_IGU_AFIFO_DFX_1"},
+	{true,	"LGE_IGU_AFIFO_DFX_2"},
+	{true,	"LGE_IGU_AFIFO_DFX_3"},
+	{true,	"LGE_IGU_AFIFO_DFX_4"},
+
+	{true,	"LGE_IGU_AFIFO_DFX_5"},
+	{true,	"LGE_IGU_AFIFO_DFX_6"},
+	{true,	"LGE_IGU_AFIFO_DFX_7"},
+	{true,	"LGE_EGU_AFIFO_DFX_0"},
+	{true,	"LGE_EGU_AFIFO_DFX_1"},
+	{true,	"LGE_EGU_AFIFO_DFX_2"},
+
+	{true,	"LGE_EGU_AFIFO_DFX_3"},
+	{true,	"LGE_EGU_AFIFO_DFX_4"},
+	{true,	"LGE_EGU_AFIFO_DFX_5"},
+	{true,	"LGE_EGU_AFIFO_DFX_6"},
+	{true,	"LGE_EGU_AFIFO_DFX_7"},
+	{true,	"CGE_IGU_AFIFO_DFX_0"},
+
+	{true,	"CGE_IGU_AFIFO_DFX_1"},
+	{true,	"CGE_EGU_AFIFO_DFX_0"},
+	{true,	"CGE_EGU_AFIFO_DFX_1"},
+	{false, "Reserved"},
+	{false, "Reserved"},
+	{false, "Reserved"},
+};
+
+static const struct hclge_dbg_dfx_message hclge_dbg_ppp_reg[] = {
+	{false, "Reserved"},
+	{true,	"DROP_FROM_PRT_PKT_CNT"},
+	{true,	"DROP_FROM_HOST_PKT_CNT"},
+	{true,	"DROP_TX_VLAN_PROC_CNT"},
+	{true,	"DROP_MNG_CNT"},
+	{true,	"DROP_FD_CNT"},
+
+	{true,	"DROP_NO_DST_CNT"},
+	{true,	"DROP_MC_MBID_FULL_CNT"},
+	{true,	"DROP_SC_FILTERED"},
+	{true,	"PPP_MC_DROP_PKT_CNT"},
+	{true,	"DROP_PT_CNT"},
+	{true,	"DROP_MAC_ANTI_SPOOF_CNT"},
+
+	{true,	"DROP_IG_VFV_CNT"},
+	{true,	"DROP_IG_PRTV_CNT"},
+	{true,	"DROP_CNM_PFC_PAUSE_CNT"},
+	{true,	"DROP_TORUS_TC_CNT"},
+	{true,	"DROP_TORUS_LPBK_CNT"},
+	{true,	"PPP_HFS_STS"},
+
+	{true,	"PPP_MC_RSLT_STS"},
+	{true,	"PPP_P3U_STS"},
+	{true,	"PPP_RSLT_DESCR_STS"},
+	{true,	"PPP_UMV_STS_0"},
+	{true,	"PPP_UMV_STS_1"},
+	{true,	"PPP_VFV_STS"},
+
+	{true,	"PPP_GRO_KEY_CNT"},
+	{true,	"PPP_GRO_INFO_CNT"},
+	{true,	"PPP_GRO_DROP_CNT"},
+	{true,	"PPP_GRO_OUT_CNT"},
+	{true,	"PPP_GRO_KEY_MATCH_DATA_CNT"},
+	{true,	"PPP_GRO_KEY_MATCH_TCAM_CNT"},
+
+	{true,	"PPP_GRO_INFO_MATCH_CNT"},
+	{true,	"PPP_GRO_FREE_ENTRY_CNT"},
+	{true,	"PPP_GRO_INNER_DFX_SIGNAL"},
+	{false, "Reserved"},
+	{false, "Reserved"},
+	{false, "Reserved"},
+
+	{true,	"GET_RX_PKT_CNT_L"},
+	{true,	"GET_RX_PKT_CNT_H"},
+	{true,	"GET_TX_PKT_CNT_L"},
+	{true,	"GET_TX_PKT_CNT_H"},
+	{true,	"SEND_UC_PRT2HOST_PKT_CNT_L"},
+	{true,	"SEND_UC_PRT2HOST_PKT_CNT_H"},
+
+	{true,	"SEND_UC_PRT2PRT_PKT_CNT_L"},
+	{true,	"SEND_UC_PRT2PRT_PKT_CNT_H"},
+	{true,	"SEND_UC_HOST2HOST_PKT_CNT_L"},
+	{true,	"SEND_UC_HOST2HOST_PKT_CNT_H"},
+	{true,	"SEND_UC_HOST2PRT_PKT_CNT_L"},
+	{true,	"SEND_UC_HOST2PRT_PKT_CNT_H"},
+
+	{true,	"SEND_MC_FROM_PRT_CNT_L"},
+	{true,	"SEND_MC_FROM_PRT_CNT_H"},
+	{true,	"SEND_MC_FROM_HOST_CNT_L"},
+	{true,	"SEND_MC_FROM_HOST_CNT_H"},
+	{true,	"SSU_MC_RD_CNT_L"},
+	{true,	"SSU_MC_RD_CNT_H"},
+
+	{true,	"SSU_MC_DROP_CNT_L"},
+	{true,	"SSU_MC_DROP_CNT_H"},
+	{true,	"SSU_MC_RD_PKT_CNT_L"},
+	{true,	"SSU_MC_RD_PKT_CNT_H"},
+	{true,	"PPP_MC_2HOST_PKT_CNT_L"},
+	{true,	"PPP_MC_2HOST_PKT_CNT_H"},
+
+	{true,	"PPP_MC_2PRT_PKT_CNT_L"},
+	{true,	"PPP_MC_2PRT_PKT_CNT_H"},
+	{true,	"NTSNOS_PKT_CNT_L"},
+	{true,	"NTSNOS_PKT_CNT_H"},
+	{true,	"NTUP_PKT_CNT_L"},
+	{true,	"NTUP_PKT_CNT_H"},
+
+	{true,	"NTLCL_PKT_CNT_L"},
+	{true,	"NTLCL_PKT_CNT_H"},
+	{true,	"NTTGT_PKT_CNT_L"},
+	{true,	"NTTGT_PKT_CNT_H"},
+	{true,	"RTNS_PKT_CNT_L"},
+	{true,	"RTNS_PKT_CNT_H"},
+
+	{true,	"RTLPBK_PKT_CNT_L"},
+	{true,	"RTLPBK_PKT_CNT_H"},
+	{true,	"NR_PKT_CNT_L"},
+	{true,	"NR_PKT_CNT_H"},
+	{true,	"RR_PKT_CNT_L"},
+	{true,	"RR_PKT_CNT_H"},
+
+	{true,	"MNG_TBL_HIT_CNT_L"},
+	{true,	"MNG_TBL_HIT_CNT_H"},
+	{true,	"FD_TBL_HIT_CNT_L"},
+	{true,	"FD_TBL_HIT_CNT_H"},
+	{true,	"FD_LKUP_CNT_L"},
+	{true,	"FD_LKUP_CNT_H"},
+
+	{true,	"BC_HIT_CNT_L"},
+	{true,	"BC_HIT_CNT_H"},
+	{true,	"UM_TBL_UC_HIT_CNT_L"},
+	{true,	"UM_TBL_UC_HIT_CNT_H"},
+	{true,	"UM_TBL_MC_HIT_CNT_L"},
+	{true,	"UM_TBL_MC_HIT_CNT_H"},
+
+	{true,	"UM_TBL_VMDQ1_HIT_CNT_L"},
+	{true,	"UM_TBL_VMDQ1_HIT_CNT_H"},
+	{true,	"MTA_TBL_HIT_CNT_L"},
+	{true,	"MTA_TBL_HIT_CNT_H"},
+	{true,	"FWD_BONDING_HIT_CNT_L"},
+	{true,	"FWD_BONDING_HIT_CNT_H"},
+
+	{true,	"PROMIS_TBL_HIT_CNT_L"},
+	{true,	"PROMIS_TBL_HIT_CNT_H"},
+	{true,	"GET_TUNL_PKT_CNT_L"},
+	{true,	"GET_TUNL_PKT_CNT_H"},
+	{true,	"GET_BMC_PKT_CNT_L"},
+	{true,	"GET_BMC_PKT_CNT_H"},
+
+	{true,	"SEND_UC_PRT2BMC_PKT_CNT_L"},
+	{true,	"SEND_UC_PRT2BMC_PKT_CNT_H"},
+	{true,	"SEND_UC_HOST2BMC_PKT_CNT_L"},
+	{true,	"SEND_UC_HOST2BMC_PKT_CNT_H"},
+	{true,	"SEND_UC_BMC2HOST_PKT_CNT_L"},
+	{true,	"SEND_UC_BMC2HOST_PKT_CNT_H"},
+
+	{true,	"SEND_UC_BMC2PRT_PKT_CNT_L"},
+	{true,	"SEND_UC_BMC2PRT_PKT_CNT_H"},
+	{true,	"PPP_MC_2BMC_PKT_CNT_L"},
+	{true,	"PPP_MC_2BMC_PKT_CNT_H"},
+	{true,	"VLAN_MIRR_CNT_L"},
+	{true,	"VLAN_MIRR_CNT_H"},
+
+	{true,	"IG_MIRR_CNT_L"},
+	{true,	"IG_MIRR_CNT_H"},
+	{true,	"EG_MIRR_CNT_L"},
+	{true,	"EG_MIRR_CNT_H"},
+	{true,	"RX_DEFAULT_HOST_HIT_CNT_L"},
+	{true,	"RX_DEFAULT_HOST_HIT_CNT_H"},
+
+	{true,	"LAN_PAIR_CNT_L"},
+	{true,	"LAN_PAIR_CNT_H"},
+	{true,	"UM_TBL_MC_HIT_PKT_CNT_L"},
+	{true,	"UM_TBL_MC_HIT_PKT_CNT_H"},
+	{true,	"MTA_TBL_HIT_PKT_CNT_L"},
+	{true,	"MTA_TBL_HIT_PKT_CNT_H"},
+
+	{true,	"PROMIS_TBL_HIT_PKT_CNT_L"},
+	{true,	"PROMIS_TBL_HIT_PKT_CNT_H"},
+	{false, "Reserved"},
+	{false, "Reserved"},
+	{false, "Reserved"},
+	{false, "Reserved"},
+};
+
+static const struct hclge_dbg_dfx_message hclge_dbg_rcb_reg[] = {
+	{false, "Reserved"},
+	{true,	"FSM_DFX_ST0"},
+	{true,	"FSM_DFX_ST1"},
+	{true,	"FSM_DFX_ST2"},
+	{true,	"FIFO_DFX_ST0"},
+	{true,	"FIFO_DFX_ST1"},
+
+	{true,	"FIFO_DFX_ST2"},
+	{true,	"FIFO_DFX_ST3"},
+	{true,	"FIFO_DFX_ST4"},
+	{true,	"FIFO_DFX_ST5"},
+	{true,	"FIFO_DFX_ST6"},
+	{true,	"FIFO_DFX_ST7"},
+
+	{true,	"FIFO_DFX_ST8"},
+	{true,	"FIFO_DFX_ST9"},
+	{true,	"FIFO_DFX_ST10"},
+	{true,	"FIFO_DFX_ST11"},
+	{true,	"Q_CREDIT_VLD_0"},
+	{true,	"Q_CREDIT_VLD_1"},
+
+	{true,	"Q_CREDIT_VLD_2"},
+	{true,	"Q_CREDIT_VLD_3"},
+	{true,	"Q_CREDIT_VLD_4"},
+	{true,	"Q_CREDIT_VLD_5"},
+	{true,	"Q_CREDIT_VLD_6"},
+	{true,	"Q_CREDIT_VLD_7"},
+
+	{true,	"Q_CREDIT_VLD_8"},
+	{true,	"Q_CREDIT_VLD_9"},
+	{true,	"Q_CREDIT_VLD_10"},
+	{true,	"Q_CREDIT_VLD_11"},
+	{true,	"Q_CREDIT_VLD_12"},
+	{true,	"Q_CREDIT_VLD_13"},
+
+	{true,	"Q_CREDIT_VLD_14"},
+	{true,	"Q_CREDIT_VLD_15"},
+	{true,	"Q_CREDIT_VLD_16"},
+	{true,	"Q_CREDIT_VLD_17"},
+	{true,	"Q_CREDIT_VLD_18"},
+	{true,	"Q_CREDIT_VLD_19"},
+
+	{true,	"Q_CREDIT_VLD_20"},
+	{true,	"Q_CREDIT_VLD_21"},
+	{true,	"Q_CREDIT_VLD_22"},
+	{true,	"Q_CREDIT_VLD_23"},
+	{true,	"Q_CREDIT_VLD_24"},
+	{true,	"Q_CREDIT_VLD_25"},
+
+	{true,	"Q_CREDIT_VLD_26"},
+	{true,	"Q_CREDIT_VLD_27"},
+	{true,	"Q_CREDIT_VLD_28"},
+	{true,	"Q_CREDIT_VLD_29"},
+	{true,	"Q_CREDIT_VLD_30"},
+	{true,	"Q_CREDIT_VLD_31"},
+
+	{true,	"GRO_BD_SERR_CNT"},
+	{true,	"GRO_CONTEXT_SERR_CNT"},
+	{true,	"RX_STASH_CFG_SERR_CNT"},
+	{true,	"AXI_RD_FBD_SERR_CNT"},
+	{true,	"GRO_BD_MERR_CNT"},
+	{true,	"GRO_CONTEXT_MERR_CNT"},
+
+	{true,	"RX_STASH_CFG_MERR_CNT"},
+	{true,	"AXI_RD_FBD_MERR_CNT"},
+	{false, "Reserved"},
+	{false, "Reserved"},
+	{false, "Reserved"},
+	{false, "Reserved"},
+};
+
+static const struct hclge_dbg_dfx_message hclge_dbg_tqp_reg[] = {
+	{true, "q_num"},
+	{true, "RCB_CFG_RX_RING_TAIL"},
+	{true, "RCB_CFG_RX_RING_HEAD"},
+	{true, "RCB_CFG_RX_RING_FBDNUM"},
+	{true, "RCB_CFG_RX_RING_OFFSET"},
+	{true, "RCB_CFG_RX_RING_FBDOFFSET"},
+
+	{true, "RCB_CFG_RX_RING_PKTNUM_RECORD"},
+	{true, "RCB_CFG_TX_RING_TAIL"},
+	{true, "RCB_CFG_TX_RING_HEAD"},
+	{true, "RCB_CFG_TX_RING_FBDNUM"},
+	{true, "RCB_CFG_TX_RING_OFFSET"},
+	{true, "RCB_CFG_TX_RING_EBDNUM"},
+};
+
+static const struct hclge_dbg_reg_type_info hclge_dbg_reg_info[] = {
+	{ .cmd = HNAE3_DBG_CMD_REG_BIOS_COMMON,
+	  .dfx_msg = &hclge_dbg_bios_common_reg[0],
+	  .reg_msg = { .msg_num = ARRAY_SIZE(hclge_dbg_bios_common_reg),
+		       .offset = HCLGE_DBG_DFX_BIOS_OFFSET,
+		       .cmd = HCLGE_OPC_DFX_BIOS_COMMON_REG } },
+	{ .cmd = HNAE3_DBG_CMD_REG_SSU,
+	  .dfx_msg = &hclge_dbg_ssu_reg_0[0],
+	  .reg_msg = { .msg_num = ARRAY_SIZE(hclge_dbg_ssu_reg_0),
+		       .offset = HCLGE_DBG_DFX_SSU_0_OFFSET,
+		       .cmd = HCLGE_OPC_DFX_SSU_REG_0 } },
+	{ .cmd = HNAE3_DBG_CMD_REG_SSU,
+	  .dfx_msg = &hclge_dbg_ssu_reg_1[0],
+	  .reg_msg = { .msg_num = ARRAY_SIZE(hclge_dbg_ssu_reg_1),
+		       .offset = HCLGE_DBG_DFX_SSU_1_OFFSET,
+		       .cmd = HCLGE_OPC_DFX_SSU_REG_1 } },
+	{ .cmd = HNAE3_DBG_CMD_REG_SSU,
+	  .dfx_msg = &hclge_dbg_ssu_reg_2[0],
+	  .reg_msg = { .msg_num = ARRAY_SIZE(hclge_dbg_ssu_reg_2),
+		       .offset = HCLGE_DBG_DFX_SSU_2_OFFSET,
+		       .cmd = HCLGE_OPC_DFX_SSU_REG_2 } },
+	{ .cmd = HNAE3_DBG_CMD_REG_IGU_EGU,
+	  .dfx_msg = &hclge_dbg_igu_egu_reg[0],
+	  .reg_msg = { .msg_num = ARRAY_SIZE(hclge_dbg_igu_egu_reg),
+		       .offset = HCLGE_DBG_DFX_IGU_OFFSET,
+		       .cmd = HCLGE_OPC_DFX_IGU_EGU_REG } },
+	{ .cmd = HNAE3_DBG_CMD_REG_RPU,
+	  .dfx_msg = &hclge_dbg_rpu_reg_0[0],
+	  .reg_msg = { .msg_num = ARRAY_SIZE(hclge_dbg_rpu_reg_0),
+		       .offset = HCLGE_DBG_DFX_RPU_0_OFFSET,
+		       .cmd = HCLGE_OPC_DFX_RPU_REG_0 } },
+	{ .cmd = HNAE3_DBG_CMD_REG_RPU,
+	  .dfx_msg = &hclge_dbg_rpu_reg_1[0],
+	  .reg_msg = { .msg_num = ARRAY_SIZE(hclge_dbg_rpu_reg_1),
+		       .offset = HCLGE_DBG_DFX_RPU_1_OFFSET,
+		       .cmd = HCLGE_OPC_DFX_RPU_REG_1 } },
+	{ .cmd = HNAE3_DBG_CMD_REG_NCSI,
+	  .dfx_msg = &hclge_dbg_ncsi_reg[0],
+	  .reg_msg = { .msg_num = ARRAY_SIZE(hclge_dbg_ncsi_reg),
+		       .offset = HCLGE_DBG_DFX_NCSI_OFFSET,
+		       .cmd = HCLGE_OPC_DFX_NCSI_REG } },
+	{ .cmd = HNAE3_DBG_CMD_REG_RTC,
+	  .dfx_msg = &hclge_dbg_rtc_reg[0],
+	  .reg_msg = { .msg_num = ARRAY_SIZE(hclge_dbg_rtc_reg),
+		       .offset = HCLGE_DBG_DFX_RTC_OFFSET,
+		       .cmd = HCLGE_OPC_DFX_RTC_REG } },
+	{ .cmd = HNAE3_DBG_CMD_REG_PPP,
+	  .dfx_msg = &hclge_dbg_ppp_reg[0],
+	  .reg_msg = { .msg_num = ARRAY_SIZE(hclge_dbg_ppp_reg),
+		       .offset = HCLGE_DBG_DFX_PPP_OFFSET,
+		       .cmd = HCLGE_OPC_DFX_PPP_REG } },
+	{ .cmd = HNAE3_DBG_CMD_REG_RCB,
+	  .dfx_msg = &hclge_dbg_rcb_reg[0],
+	  .reg_msg = { .msg_num = ARRAY_SIZE(hclge_dbg_rcb_reg),
+		       .offset = HCLGE_DBG_DFX_RCB_OFFSET,
+		       .cmd = HCLGE_OPC_DFX_RCB_REG } },
+	{ .cmd = HNAE3_DBG_CMD_REG_TQP,
+	  .dfx_msg = &hclge_dbg_tqp_reg[0],
+	  .reg_msg = { .msg_num = ARRAY_SIZE(hclge_dbg_tqp_reg),
+		       .offset = HCLGE_DBG_DFX_TQP_OFFSET,
+		       .cmd = HCLGE_OPC_DFX_TQP_REG } },
+};
+
+static char *hclge_dbg_get_func_id_str(char *buf, u8 id)
+{
+	if (id)
+		sprintf(buf, "vf%u", id - 1U);
+	else
+		sprintf(buf, "pf");
+
+	return buf;
+}
+
+static int hclge_dbg_get_dfx_bd_num(struct hclge_dev *hdev, int offset,
+				    u32 *bd_num)
+{
+	struct hclge_desc desc[HCLGE_GET_DFX_REG_TYPE_CNT];
+	int entries_per_desc;
+	int index;
+	int ret;
+
+	ret = hclge_query_bd_num_cmd_send(hdev, desc);
+	if (ret) {
+		dev_err(&hdev->pdev->dev,
+			"failed to get dfx bd_num, offset = %d, ret = %d\n",
+			offset, ret);
+		return ret;
+	}
+
+	entries_per_desc = ARRAY_SIZE(desc[0].data);
+	index = offset % entries_per_desc;
+
+	*bd_num = le32_to_cpu(desc[offset / entries_per_desc].data[index]);
+	if (!(*bd_num)) {
+		dev_err(&hdev->pdev->dev, "The value of dfx bd_num is 0!\n");
+		return -EINVAL;
+	}
+
+	return 0;
+}
+
+int hclge_dbg_cmd_send(struct hclge_dev *hdev, struct hclge_desc *desc_src,
+		       int index, int bd_num, enum hclge_opcode_type cmd)
+{
+	struct hclge_desc *desc = desc_src;
+	int ret, i;
+
+	hclge_cmd_setup_basic_desc(desc, cmd, true);
+	desc->data[0] = cpu_to_le32(index);
+
+	for (i = 1; i < bd_num; i++) {
+		desc->flag |= cpu_to_le16(HCLGE_COMM_CMD_FLAG_NEXT);
+		desc++;
+		hclge_cmd_setup_basic_desc(desc, cmd, true);
+	}
+
+	ret = hclge_cmd_send(&hdev->hw, desc_src, bd_num);
+	if (ret)
+		dev_err(&hdev->pdev->dev,
+			"cmd(0x%x) send fail, ret = %d\n", cmd, ret);
+	return ret;
+}
+
+static int
+hclge_dbg_dump_reg_tqp(struct hclge_dev *hdev,
+		       const struct hclge_dbg_reg_type_info *reg_info,
+		       struct seq_file *s)
+{
+	const struct hclge_dbg_dfx_message *dfx_message = reg_info->dfx_msg;
+	const struct hclge_dbg_reg_common_msg *reg_msg = &reg_info->reg_msg;
+	u32 index, entry, i, cnt, min_num;
+	struct hclge_desc *desc_src;
+	struct hclge_desc *desc;
+	int bd_num, ret;
+
+	ret = hclge_dbg_get_dfx_bd_num(hdev, reg_msg->offset, &bd_num);
+	if (ret)
+		return ret;
+
+	desc_src = kcalloc(bd_num, sizeof(struct hclge_desc), GFP_KERNEL);
+	if (!desc_src)
+		return -ENOMEM;
+
+	min_num = min_t(int, bd_num * HCLGE_DESC_DATA_LEN, reg_msg->msg_num);
+
+	for (i = 0, cnt = 0; i < min_num; i++, dfx_message++)
+		seq_printf(s, "item%u = %s\n", cnt++, dfx_message->message);
+
+	for (i = 0; i < cnt; i++)
+		seq_printf(s, "item%u\t", i);
+
+	seq_puts(s, "\n");
+
+	for (index = 0; index < hdev->vport[0].alloc_tqps; index++) {
+		dfx_message = reg_info->dfx_msg;
+		desc = desc_src;
+		ret = hclge_dbg_cmd_send(hdev, desc, index, bd_num,
+					 reg_msg->cmd);
+		if (ret)
+			break;
+
+		for (i = 0; i < min_num; i++, dfx_message++) {
+			entry = i % HCLGE_DESC_DATA_LEN;
+			if (i > 0 && !entry)
+				desc++;
+
+			seq_printf(s, "%#x\t", le32_to_cpu(desc->data[entry]));
+		}
+		seq_puts(s, "\n");
+	}
+
+	kfree(desc_src);
+	return ret;
+}
+
+static int
+hclge_dbg_dump_reg_common(struct hclge_dev *hdev,
+			  const struct hclge_dbg_reg_type_info *reg_info,
+			  struct seq_file *s)
+{
+	const struct hclge_dbg_reg_common_msg *reg_msg = &reg_info->reg_msg;
+	const struct hclge_dbg_dfx_message *dfx_message = reg_info->dfx_msg;
+	struct hclge_desc *desc_src;
+	int bd_num, min_num, ret, i;
+	struct hclge_desc *desc;
+	u32 entry;
+
+	ret = hclge_dbg_get_dfx_bd_num(hdev, reg_msg->offset, &bd_num);
+	if (ret)
+		return ret;
+
+	desc_src = kcalloc(bd_num, sizeof(struct hclge_desc), GFP_KERNEL);
+	if (!desc_src)
+		return -ENOMEM;
+
+	desc = desc_src;
+
+	ret = hclge_dbg_cmd_send(hdev, desc, 0, bd_num, reg_msg->cmd);
+	if (ret) {
+		kfree(desc);
+		return ret;
+	}
+
+	min_num = min_t(int, bd_num * HCLGE_DESC_DATA_LEN, reg_msg->msg_num);
+
+	for (i = 0; i < min_num; i++, dfx_message++) {
+		entry = i % HCLGE_DESC_DATA_LEN;
+		if (i > 0 && !entry)
+			desc++;
+		if (!dfx_message->flag)
+			continue;
+
+		seq_printf(s, "%s: %#x\n", dfx_message->message,
+			   le32_to_cpu(desc->data[entry]));
+	}
+
+	kfree(desc_src);
+	return 0;
+}
+
+static const struct hclge_dbg_status_dfx_info hclge_dbg_mac_en_status[] = {
+	{HCLGE_MAC_TX_EN_B,  "mac_trans_en"},
+	{HCLGE_MAC_RX_EN_B,  "mac_rcv_en"},
+	{HCLGE_MAC_PAD_TX_B, "pad_trans_en"},
+	{HCLGE_MAC_PAD_RX_B, "pad_rcv_en"},
+	{HCLGE_MAC_1588_TX_B, "1588_trans_en"},
+	{HCLGE_MAC_1588_RX_B, "1588_rcv_en"},
+	{HCLGE_MAC_APP_LP_B,  "mac_app_loop_en"},
+	{HCLGE_MAC_LINE_LP_B, "mac_line_loop_en"},
+	{HCLGE_MAC_FCS_TX_B,  "mac_fcs_tx_en"},
+	{HCLGE_MAC_RX_OVERSIZE_TRUNCATE_B, "mac_rx_oversize_truncate_en"},
+	{HCLGE_MAC_RX_FCS_STRIP_B, "mac_rx_fcs_strip_en"},
+	{HCLGE_MAC_RX_FCS_B, "mac_rx_fcs_en"},
+	{HCLGE_MAC_TX_UNDER_MIN_ERR_B, "mac_tx_under_min_err_en"},
+	{HCLGE_MAC_TX_OVERSIZE_TRUNCATE_B, "mac_tx_oversize_truncate_en"}
+};
+
+static int hclge_dbg_dump_mac_enable_status(struct hclge_dev *hdev,
+					    struct seq_file *s)
+{
+	struct hclge_config_mac_mode_cmd *req;
+	struct hclge_desc desc;
+	u32 loop_en, i, offset;
+	int ret;
+
+	hclge_cmd_setup_basic_desc(&desc, HCLGE_OPC_CONFIG_MAC_MODE, true);
+
+	ret = hclge_cmd_send(&hdev->hw, &desc, 1);
+	if (ret) {
+		dev_err(&hdev->pdev->dev,
+			"failed to dump mac enable status, ret = %d\n", ret);
+		return ret;
+	}
+
+	req = (struct hclge_config_mac_mode_cmd *)desc.data;
+	loop_en = le32_to_cpu(req->txrx_pad_fcs_loop_en);
+
+	for (i = 0; i < ARRAY_SIZE(hclge_dbg_mac_en_status); i++) {
+		offset = hclge_dbg_mac_en_status[i].offset;
+		seq_printf(s, "%s: %#x\n", hclge_dbg_mac_en_status[i].message,
+			   hnae3_get_bit(loop_en, offset));
+	}
+
+	return 0;
+}
+
+static int hclge_dbg_dump_mac_frame_size(struct hclge_dev *hdev,
+					 struct seq_file *s)
+{
+	struct hclge_config_max_frm_size_cmd *req;
+	struct hclge_desc desc;
+	int ret;
+
+	hclge_cmd_setup_basic_desc(&desc, HCLGE_OPC_CONFIG_MAX_FRM_SIZE, true);
+
+	ret = hclge_cmd_send(&hdev->hw, &desc, 1);
+	if (ret) {
+		dev_err(&hdev->pdev->dev,
+			"failed to dump mac frame size, ret = %d\n", ret);
+		return ret;
+	}
+
+	req = (struct hclge_config_max_frm_size_cmd *)desc.data;
+
+	seq_printf(s, "max_frame_size: %u\n", le16_to_cpu(req->max_frm_size));
+	seq_printf(s, "min_frame_size: %u\n", req->min_frm_size);
+
+	return 0;
+}
+
+static int hclge_dbg_dump_mac_speed_duplex(struct hclge_dev *hdev,
+					   struct seq_file *s)
+{
+#define HCLGE_MAC_SPEED_SHIFT	0
+#define HCLGE_MAC_SPEED_MASK	GENMASK(5, 0)
+#define HCLGE_MAC_DUPLEX_SHIFT	7
+
+	struct hclge_config_mac_speed_dup_cmd *req;
+	struct hclge_desc desc;
+	int ret;
+
+	hclge_cmd_setup_basic_desc(&desc, HCLGE_OPC_CONFIG_SPEED_DUP, true);
+
+	ret = hclge_cmd_send(&hdev->hw, &desc, 1);
+	if (ret) {
+		dev_err(&hdev->pdev->dev,
+			"failed to dump mac speed duplex, ret = %d\n", ret);
+		return ret;
+	}
+
+	req = (struct hclge_config_mac_speed_dup_cmd *)desc.data;
+
+	seq_printf(s, "speed: %#lx\n",
+		   hnae3_get_field(req->speed_dup, HCLGE_MAC_SPEED_MASK,
+				   HCLGE_MAC_SPEED_SHIFT));
+	seq_printf(s, "duplex: %#x\n",
+		   hnae3_get_bit(req->speed_dup, HCLGE_MAC_DUPLEX_SHIFT));
+	return 0;
+}
+
+static int hclge_dbg_dump_mac(struct seq_file *s, void *data)
+{
+	struct hclge_dev *hdev = hclge_seq_file_to_hdev(s);
+	int ret;
+
+	ret = hclge_dbg_dump_mac_enable_status(hdev, s);
+	if (ret)
+		return ret;
+
+	ret = hclge_dbg_dump_mac_frame_size(hdev, s);
+	if (ret)
+		return ret;
+
+	return hclge_dbg_dump_mac_speed_duplex(hdev, s);
+}
+
+static int hclge_dbg_dump_dcb_qset(struct hclge_dev *hdev, struct seq_file *s)
+{
+	struct hclge_dbg_bitmap_cmd req;
+	struct hclge_desc desc;
+	u16 qset_id, qset_num;
+	int ret;
+
+	ret = hclge_tm_get_qset_num(hdev, &qset_num);
+	if (ret)
+		return ret;
+
+	seq_puts(s, "qset_id  roce_qset_mask  nic_qset_mask  ");
+	seq_puts(s, "qset_shaping_pass  qset_bp_status\n");
+	for (qset_id = 0; qset_id < qset_num; qset_id++) {
+		ret = hclge_dbg_cmd_send(hdev, &desc, qset_id, 1,
+					 HCLGE_OPC_QSET_DFX_STS);
+		if (ret)
+			return ret;
+
+		req.bitmap = (u8)le32_to_cpu(desc.data[1]);
+
+		seq_printf(s, "%04u     %#-16x%#-15x%#-19x%#-x\n",
+			   qset_id, req.bit0, req.bit1, req.bit2, req.bit3);
+	}
+
+	return 0;
+}
+
+static int hclge_dbg_dump_dcb_pri(struct hclge_dev *hdev, struct seq_file *s)
+{
+	struct hclge_dbg_bitmap_cmd req;
+	struct hclge_desc desc;
+	u8 pri_id, pri_num;
+	int ret;
+
+	ret = hclge_tm_get_pri_num(hdev, &pri_num);
+	if (ret)
+		return ret;
+
+	seq_puts(s, "pri_id  pri_mask  pri_cshaping_pass  pri_pshaping_pass\n");
+	for (pri_id = 0; pri_id < pri_num; pri_id++) {
+		ret = hclge_dbg_cmd_send(hdev, &desc, pri_id, 1,
+					 HCLGE_OPC_PRI_DFX_STS);
+		if (ret)
+			return ret;
+
+		req.bitmap = (u8)le32_to_cpu(desc.data[1]);
+
+		seq_printf(s, "%03u     %#-10x%#-19x%#-x\n",
+			   pri_id, req.bit0, req.bit1, req.bit2);
+	}
+
+	return 0;
+}
+
+static int hclge_dbg_dump_dcb_pg(struct hclge_dev *hdev, struct seq_file *s)
+{
+	struct hclge_dbg_bitmap_cmd req;
+	struct hclge_desc desc;
+	u8 pg_id;
+	int ret;
+
+	seq_puts(s, "pg_id  pg_mask  pg_cshaping_pass  pg_pshaping_pass\n");
+	for (pg_id = 0; pg_id < hdev->tm_info.num_pg; pg_id++) {
+		ret = hclge_dbg_cmd_send(hdev, &desc, pg_id, 1,
+					 HCLGE_OPC_PG_DFX_STS);
+		if (ret)
+			return ret;
+
+		req.bitmap = (u8)le32_to_cpu(desc.data[1]);
+
+		seq_printf(s, "%03u    %#-9x%#-18x%#-x\n",
+			   pg_id, req.bit0, req.bit1, req.bit2);
+	}
+
+	return 0;
+}
+
+static int hclge_dbg_dump_dcb_queue(struct hclge_dev *hdev, struct seq_file *s)
+{
+	struct hclge_desc desc;
+	u16 nq_id;
+	int ret;
+
+	seq_puts(s, "nq_id  sch_nic_queue_cnt  sch_roce_queue_cnt\n");
+	for (nq_id = 0; nq_id < hdev->num_tqps; nq_id++) {
+		ret = hclge_dbg_cmd_send(hdev, &desc, nq_id, 1,
+					 HCLGE_OPC_SCH_NQ_CNT);
+		if (ret)
+			return ret;
+
+		seq_printf(s, "%04u   %#-19x",
+			   nq_id, le32_to_cpu(desc.data[1]));
+
+		ret = hclge_dbg_cmd_send(hdev, &desc, nq_id, 1,
+					 HCLGE_OPC_SCH_RQ_CNT);
+		if (ret)
+			return ret;
+
+		seq_printf(s, "%#-x\n", le32_to_cpu(desc.data[1]));
+	}
+
+	return 0;
+}
+
+static int hclge_dbg_dump_dcb_port(struct hclge_dev *hdev, struct seq_file *s)
+{
+	struct hclge_dbg_bitmap_cmd req;
+	struct hclge_desc desc;
+	u8 port_id = 0;
+	int ret;
+
+	ret = hclge_dbg_cmd_send(hdev, &desc, port_id, 1,
+				 HCLGE_OPC_PORT_DFX_STS);
+	if (ret)
+		return ret;
+
+	req.bitmap = (u8)le32_to_cpu(desc.data[1]);
+
+	seq_printf(s, "port_mask: %#x\n", req.bit0);
+	seq_printf(s, "port_shaping_pass: %#x\n", req.bit1);
+
+	return 0;
+}
+
+static int hclge_dbg_dump_dcb_tm(struct hclge_dev *hdev, struct seq_file *s)
+{
+	struct hclge_desc desc[2];
+	u8 port_id = 0;
+	int ret;
+
+	ret = hclge_dbg_cmd_send(hdev, desc, port_id, 1,
+				 HCLGE_OPC_TM_INTERNAL_CNT);
+	if (ret)
+		return ret;
+
+	seq_printf(s, "SCH_NIC_NUM: %#x\n", le32_to_cpu(desc[0].data[1]));
+	seq_printf(s, "SCH_ROCE_NUM: %#x\n", le32_to_cpu(desc[0].data[2]));
+
+	ret = hclge_dbg_cmd_send(hdev, desc, port_id, 2,
+				 HCLGE_OPC_TM_INTERNAL_STS);
+	if (ret)
+		return ret;
+
+	seq_printf(s, "pri_bp: %#x\n", le32_to_cpu(desc[0].data[1]));
+	seq_printf(s, "fifo_dfx_info: %#x\n", le32_to_cpu(desc[0].data[2]));
+	seq_printf(s, "sch_roce_fifo_afull_gap: %#x\n",
+		   le32_to_cpu(desc[0].data[3]));
+	seq_printf(s, "tx_private_waterline: %#x\n",
+		   le32_to_cpu(desc[0].data[4]));
+	seq_printf(s, "tm_bypass_en: %#x\n", le32_to_cpu(desc[0].data[5]));
+	seq_printf(s, "SSU_TM_BYPASS_EN: %#x\n", le32_to_cpu(desc[1].data[0]));
+	seq_printf(s, "SSU_RESERVE_CFG: %#x\n", le32_to_cpu(desc[1].data[1]));
+
+	if (hdev->hw.mac.media_type == HNAE3_MEDIA_TYPE_COPPER)
+		return 0;
+
+	ret = hclge_dbg_cmd_send(hdev, desc, port_id, 1,
+				 HCLGE_OPC_TM_INTERNAL_STS_1);
+	if (ret)
+		return ret;
+
+	seq_printf(s, "TC_MAP_SEL: %#x\n", le32_to_cpu(desc[0].data[1]));
+	seq_printf(s, "IGU_PFC_PRI_EN: %#x\n", le32_to_cpu(desc[0].data[2]));
+	seq_printf(s, "MAC_PFC_PRI_EN: %#x\n", le32_to_cpu(desc[0].data[3]));
+	seq_printf(s, "IGU_PRI_MAP_TC_CFG: %#x\n",
+		   le32_to_cpu(desc[0].data[4]));
+	seq_printf(s, "IGU_TX_PRI_MAP_TC_CFG: %#x\n",
+		   le32_to_cpu(desc[0].data[5]));
+
+	return 0;
+}
+
+static int hclge_dbg_dump_dcb(struct seq_file *s, void *data)
+{
+	struct hclge_dev *hdev = hclge_seq_file_to_hdev(s);
+	int ret;
+
+	ret = hclge_dbg_dump_dcb_qset(hdev, s);
+	if (ret)
+		return ret;
+
+	ret = hclge_dbg_dump_dcb_pri(hdev, s);
+	if (ret)
+		return ret;
+
+	ret = hclge_dbg_dump_dcb_pg(hdev, s);
+	if (ret)
+		return ret;
+
+	ret = hclge_dbg_dump_dcb_queue(hdev, s);
+	if (ret)
+		return ret;
+
+	ret = hclge_dbg_dump_dcb_port(hdev, s);
+	if (ret)
+		return ret;
+
+	return hclge_dbg_dump_dcb_tm(hdev, s);
+}
+
+static int hclge_dbg_dump_reg_cmd(enum hnae3_dbg_cmd cmd, struct seq_file *s)
+{
+	struct hclge_dev *hdev = hclge_seq_file_to_hdev(s);
+	const struct hclge_dbg_reg_type_info *reg_info;
+	int ret = 0;
+	u32 i;
+
+	for (i = 0; i < ARRAY_SIZE(hclge_dbg_reg_info); i++) {
+		reg_info = &hclge_dbg_reg_info[i];
+		if (cmd == reg_info->cmd) {
+			if (cmd == HNAE3_DBG_CMD_REG_TQP)
+				return hclge_dbg_dump_reg_tqp(hdev,
+							      reg_info, s);
+
+			ret = hclge_dbg_dump_reg_common(hdev, reg_info, s);
+			if (ret)
+				break;
+		}
+	}
+
+	return ret;
+}
+
+static int hclge_dbg_dump_bios_reg_cmd(struct seq_file *s, void *data)
+{
+	return hclge_dbg_dump_reg_cmd(HNAE3_DBG_CMD_REG_BIOS_COMMON, s);
+}
+
+static int hclge_dbg_dump_ssu_reg_cmd(struct seq_file *s, void *data)
+{
+	return hclge_dbg_dump_reg_cmd(HNAE3_DBG_CMD_REG_SSU, s);
+}
+
+static int hclge_dbg_dump_igu_egu_reg_cmd(struct seq_file *s, void *data)
+{
+	return hclge_dbg_dump_reg_cmd(HNAE3_DBG_CMD_REG_IGU_EGU, s);
+}
+
+static int hclge_dbg_dump_rpu_reg_cmd(struct seq_file *s, void *data)
+{
+	return hclge_dbg_dump_reg_cmd(HNAE3_DBG_CMD_REG_RPU, s);
+}
+
+static int hclge_dbg_dump_ncsi_reg_cmd(struct seq_file *s, void *data)
+{
+	return hclge_dbg_dump_reg_cmd(HNAE3_DBG_CMD_REG_NCSI, s);
+}
+
+static int hclge_dbg_dump_rtc_reg_cmd(struct seq_file *s, void *data)
+{
+	return hclge_dbg_dump_reg_cmd(HNAE3_DBG_CMD_REG_RTC, s);
+}
+
+static int hclge_dbg_dump_ppp_reg_cmd(struct seq_file *s, void *data)
+{
+	return hclge_dbg_dump_reg_cmd(HNAE3_DBG_CMD_REG_PPP, s);
+}
+
+static int hclge_dbg_dump_rcb_reg_cmd(struct seq_file *s, void *data)
+{
+	return hclge_dbg_dump_reg_cmd(HNAE3_DBG_CMD_REG_RCB, s);
+}
+
+static int hclge_dbg_dump_tqp_reg_cmd(struct seq_file *s, void *data)
+{
+	return hclge_dbg_dump_reg_cmd(HNAE3_DBG_CMD_REG_TQP, s);
+}
+
+static int hclge_dbg_dump_tc(struct seq_file *s, void *data)
+{
+	struct hclge_dev *hdev = hclge_seq_file_to_hdev(s);
+	struct hclge_ets_tc_weight_cmd *ets_weight;
+	const char *sch_mode_str;
+	struct hclge_desc desc;
+	int ret;
+	u8 i;
+
+	if (!hnae3_dev_dcb_supported(hdev)) {
+		dev_err(&hdev->pdev->dev,
+			"Only DCB-supported dev supports tc\n");
+		return -EOPNOTSUPP;
+	}
+
+	hclge_cmd_setup_basic_desc(&desc, HCLGE_OPC_ETS_TC_WEIGHT, true);
+	ret = hclge_cmd_send(&hdev->hw, &desc, 1);
+	if (ret) {
+		dev_err(&hdev->pdev->dev, "failed to get tc weight, ret = %d\n",
+			ret);
+		return ret;
+	}
+
+	ets_weight = (struct hclge_ets_tc_weight_cmd *)desc.data;
+
+	seq_printf(s, "enabled tc number: %u\n", hdev->tm_info.num_tc);
+	seq_printf(s, "weight_offset: %u\n", ets_weight->weight_offset);
+
+	seq_puts(s, "TC    MODE  WEIGHT\n");
+	for (i = 0; i < HNAE3_MAX_TC; i++) {
+		sch_mode_str = ets_weight->tc_weight[i] ? "dwrr" : "sp";
+		seq_printf(s, "%u     %4s    %3u\n", i, sch_mode_str,
+			   ets_weight->tc_weight[i]);
+	}
+
+	return 0;
+}
+
+static void hclge_dbg_fill_shaper_content(struct seq_file *s,
+					  struct hclge_tm_shaper_para *para)
+{
+	seq_printf(s, "%-8u%-8u%-8u%-8u%-8u%-8u%-14u", para->ir_b, para->ir_u,
+		   para->ir_s, para->bs_b, para->bs_s, para->flag, para->rate);
+}
+
+static int hclge_dbg_dump_tm_pg(struct seq_file *s, void *data)
+{
+	struct hclge_tm_shaper_para c_shaper_para, p_shaper_para;
+	struct hclge_dev *hdev = hclge_seq_file_to_hdev(s);
+	u8 pg_id, sch_mode, weight, pri_bit_map;
+	const char *sch_mode_str;
+	int ret;
+
+	seq_puts(s, "ID  PRI_MAP  MODE  DWRR  C_IR_B  C_IR_U  C_IR_S  C_BS_B  ");
+	seq_puts(s, "C_BS_S  C_FLAG  C_RATE(Mbps)  P_IR_B  P_IR_U  P_IR_S  ");
+	seq_puts(s, "P_BS_B  P_BS_S  P_FLAG  P_RATE(Mbps)\n");
+
+	for (pg_id = 0; pg_id < hdev->tm_info.num_pg; pg_id++) {
+		ret = hclge_tm_get_pg_to_pri_map(hdev, pg_id, &pri_bit_map);
+		if (ret)
+			return ret;
+
+		ret = hclge_tm_get_pg_sch_mode(hdev, pg_id, &sch_mode);
+		if (ret)
+			return ret;
+
+		ret = hclge_tm_get_pg_weight(hdev, pg_id, &weight);
+		if (ret)
+			return ret;
+
+		ret = hclge_tm_get_pg_shaper(hdev, pg_id,
+					     HCLGE_OPC_TM_PG_C_SHAPPING,
+					     &c_shaper_para);
+		if (ret)
+			return ret;
+
+		ret = hclge_tm_get_pg_shaper(hdev, pg_id,
+					     HCLGE_OPC_TM_PG_P_SHAPPING,
+					     &p_shaper_para);
+		if (ret)
+			return ret;
+
+		sch_mode_str = sch_mode & HCLGE_TM_TX_SCHD_DWRR_MSK ? "dwrr" :
+				       "sp";
+
+		seq_printf(s, "%02u  0x%-7x%-6s%-6u", pg_id, pri_bit_map,
+			   sch_mode_str, weight);
+		hclge_dbg_fill_shaper_content(s, &c_shaper_para);
+		hclge_dbg_fill_shaper_content(s, &p_shaper_para);
+		seq_puts(s, "\n");
+	}
+
+	return 0;
+}
+
+static int hclge_dbg_dump_tm_port(struct seq_file *s, void *data)
+{
+	struct hclge_dev *hdev = hclge_seq_file_to_hdev(s);
+	struct hclge_tm_shaper_para shaper_para;
+	int ret;
+
+	ret = hclge_tm_get_port_shaper(hdev, &shaper_para);
+	if (ret)
+		return ret;
+
+	seq_puts(s, "IR_B  IR_U  IR_S  BS_B  BS_S  FLAG  RATE(Mbps)\n");
+	seq_printf(s, "%3u   %3u   %3u   %3u   %3u     %1u   %6u\n",
+		   shaper_para.ir_b, shaper_para.ir_u, shaper_para.ir_s,
+		   shaper_para.bs_b, shaper_para.bs_s, shaper_para.flag,
+		   shaper_para.rate);
+
+	return 0;
+}
+
+static int hclge_dbg_dump_tm_bp_qset_map(struct hclge_dev *hdev, u8 tc_id,
+					 struct seq_file *s)
+{
+	u32 qset_mapping[HCLGE_BP_EXT_GRP_NUM];
+	struct hclge_bp_to_qs_map_cmd *map;
+	struct hclge_desc desc;
+	u8 group_id;
+	u8 grp_num;
+	u16 i = 0;
+	int ret;
+
+	grp_num = hdev->num_tqps <= HCLGE_TQP_MAX_SIZE_DEV_V2 ?
+		  HCLGE_BP_GRP_NUM : HCLGE_BP_EXT_GRP_NUM;
+	map = (struct hclge_bp_to_qs_map_cmd *)desc.data;
+	for (group_id = 0; group_id < grp_num; group_id++) {
+		hclge_cmd_setup_basic_desc(&desc,
+					   HCLGE_OPC_TM_BP_TO_QSET_MAPPING,
+					   true);
+		map->tc_id = tc_id;
+		map->qs_group_id = group_id;
+		ret = hclge_cmd_send(&hdev->hw, &desc, 1);
+		if (ret) {
+			dev_err(&hdev->pdev->dev,
+				"failed to get bp to qset map, ret = %d\n",
+				ret);
+			return ret;
+		}
+
+		qset_mapping[group_id] = le32_to_cpu(map->qs_bit_map);
+	}
+
+	seq_puts(s, "INDEX | TM BP QSET MAPPING:\n");
+	for (group_id = 0; group_id < grp_num / 8; group_id++) {
+		seq_printf(s,
+			   "%04d  | %08x:%08x:%08x:%08x:%08x:%08x:%08x:%08x\n",
+			   group_id * 256, qset_mapping[i + 7],
+			   qset_mapping[i + 6], qset_mapping[i + 5],
+			   qset_mapping[i + 4], qset_mapping[i + 3],
+			   qset_mapping[i + 2], qset_mapping[i + 1],
+			   qset_mapping[i]);
+		i += 8;
+	}
+
+	return 0;
+}
+
+static int hclge_dbg_dump_tm_map(struct seq_file *s, void *data)
+{
+	struct hclge_dev *hdev = hclge_seq_file_to_hdev(s);
+	u16 queue_id;
+	u16 qset_id;
+	u8 link_vld;
+	u8 pri_id;
+	u8 tc_id;
+	int ret;
+
+	for (queue_id = 0; queue_id < hdev->num_tqps; queue_id++) {
+		ret = hclge_tm_get_q_to_qs_map(hdev, queue_id, &qset_id);
+		if (ret)
+			return ret;
+
+		ret = hclge_tm_get_qset_map_pri(hdev, qset_id, &pri_id,
+						&link_vld);
+		if (ret)
+			return ret;
+
+		ret = hclge_tm_get_q_to_tc(hdev, queue_id, &tc_id);
+		if (ret)
+			return ret;
+
+		seq_puts(s, "QUEUE_ID   QSET_ID   PRI_ID   TC_ID\n");
+		seq_printf(s, "%04u        %4u       %3u      %2u\n",
+			   queue_id, qset_id, pri_id, tc_id);
+
+		if (!hnae3_dev_dcb_supported(hdev))
+			continue;
+
+		ret = hclge_dbg_dump_tm_bp_qset_map(hdev, tc_id, s);
+		if (ret < 0)
+			return ret;
+
+		seq_puts(s, "\n");
+	}
+
+	return 0;
+}
+
+static int hclge_dbg_dump_tm_nodes(struct seq_file *s, void *data)
+{
+	struct hclge_dev *hdev = hclge_seq_file_to_hdev(s);
+	struct hclge_tm_nodes_cmd *nodes;
+	struct hclge_desc desc;
+	int ret;
+
+	hclge_cmd_setup_basic_desc(&desc, HCLGE_OPC_TM_NODES, true);
+	ret = hclge_cmd_send(&hdev->hw, &desc, 1);
+	if (ret) {
+		dev_err(&hdev->pdev->dev,
+			"failed to dump tm nodes, ret = %d\n", ret);
+		return ret;
+	}
+
+	nodes = (struct hclge_tm_nodes_cmd *)desc.data;
+
+	seq_puts(s, "       BASE_ID  MAX_NUM\n");
+	seq_printf(s, "PG      %4u      %4u\n", nodes->pg_base_id,
+		   nodes->pg_num);
+	seq_printf(s, "PRI     %4u      %4u\n", nodes->pri_base_id,
+		   nodes->pri_num);
+	seq_printf(s, "QSET    %4u      %4u\n",
+		   le16_to_cpu(nodes->qset_base_id),
+		   le16_to_cpu(nodes->qset_num));
+	seq_printf(s, "QUEUE   %4u      %4u\n",
+		   le16_to_cpu(nodes->queue_base_id),
+		   le16_to_cpu(nodes->queue_num));
+
+	return 0;
+}
+
+static int hclge_dbg_dump_tm_pri(struct seq_file *s, void *data)
+{
+	struct hclge_tm_shaper_para c_shaper_para, p_shaper_para;
+	struct hclge_dev *hdev = hclge_seq_file_to_hdev(s);
+	u8 pri_num, sch_mode, weight, i;
+	const char *sch_mode_str;
+	int ret;
+
+	ret = hclge_tm_get_pri_num(hdev, &pri_num);
+	if (ret)
+		return ret;
+
+	seq_puts(s, "ID  MODE  DWRR  C_IR_B  C_IR_U  C_IR_S  C_BS_B  ");
+	seq_puts(s, "C_BS_S  C_FLAG  C_RATE(Mbps)  P_IR_B  P_IR_U  P_IR_S  ");
+	seq_puts(s, "P_BS_B  P_BS_S  P_FLAG  P_RATE(Mbps)\n");
+
+	for (i = 0; i < pri_num; i++) {
+		ret = hclge_tm_get_pri_sch_mode(hdev, i, &sch_mode);
+		if (ret)
+			goto out;
+
+		ret = hclge_tm_get_pri_weight(hdev, i, &weight);
+		if (ret)
+			goto out;
+
+		ret = hclge_tm_get_pri_shaper(hdev, i,
+					      HCLGE_OPC_TM_PRI_C_SHAPPING,
+					      &c_shaper_para);
+		if (ret)
+			goto out;
+
+		ret = hclge_tm_get_pri_shaper(hdev, i,
+					      HCLGE_OPC_TM_PRI_P_SHAPPING,
+					      &p_shaper_para);
+		if (ret)
+			goto out;
+
+		sch_mode_str = sch_mode & HCLGE_TM_TX_SCHD_DWRR_MSK ? "dwrr" :
+			       "sp";
+
+		seq_printf(s, "%04u  %-6s%-6u", i, sch_mode_str, weight);
+		hclge_dbg_fill_shaper_content(s, &c_shaper_para);
+		hclge_dbg_fill_shaper_content(s, &p_shaper_para);
+		seq_puts(s, "\n");
+	}
+
+out:
+	return ret;
+}
+
+static int hclge_dbg_dump_tm_qset(struct seq_file *s, void *data)
+{
+	struct hclge_dev *hdev = hclge_seq_file_to_hdev(s);
+	u8 priority, link_vld, sch_mode, weight;
+	struct hclge_tm_shaper_para shaper_para;
+	const char *sch_mode_str;
+	u16 qset_num, i;
+	int ret;
+
+	ret = hclge_tm_get_qset_num(hdev, &qset_num);
+	if (ret)
+		return ret;
+
+	seq_puts(s, "ID    MAP_PRI  LINK_VLD  MODE  DWRR  IR_B  IR_U  IR_S  ");
+	seq_puts(s, "BS_B  BS_S  FLAG  RATE(Mbps)\n");
+
+	for (i = 0; i < qset_num; i++) {
+		ret = hclge_tm_get_qset_map_pri(hdev, i, &priority, &link_vld);
+		if (ret)
+			return ret;
+
+		ret = hclge_tm_get_qset_sch_mode(hdev, i, &sch_mode);
+		if (ret)
+			return ret;
+
+		ret = hclge_tm_get_qset_weight(hdev, i, &weight);
+		if (ret)
+			return ret;
+
+		ret = hclge_tm_get_qset_shaper(hdev, i, &shaper_para);
+		if (ret)
+			return ret;
+
+		sch_mode_str = sch_mode & HCLGE_TM_TX_SCHD_DWRR_MSK ? "dwrr" :
+			       "sp";
+
+		seq_printf(s, "%04u  %-9u%-10u%-6s%-6u", i, priority, link_vld,
+			   sch_mode_str, weight);
+		seq_printf(s, "%-6u%-6u%-6u%-6u%-6u%-6u%-14u\n",
+			   shaper_para.ir_b, shaper_para.ir_u, shaper_para.ir_s,
+			   shaper_para.bs_b, shaper_para.bs_s, shaper_para.flag,
+			   shaper_para.rate);
+	}
+
+	return 0;
+}
+
+static int hclge_dbg_dump_qos_pause_cfg(struct seq_file *s, void *data)
+{
+	struct hclge_dev *hdev = hclge_seq_file_to_hdev(s);
+	struct hclge_cfg_pause_param_cmd *pause_param;
+	struct hclge_desc desc;
+	int ret;
+
+	hclge_cmd_setup_basic_desc(&desc, HCLGE_OPC_CFG_MAC_PARA, true);
+	ret = hclge_cmd_send(&hdev->hw, &desc, 1);
+	if (ret) {
+		dev_err(&hdev->pdev->dev,
+			"failed to dump qos pause, ret = %d\n", ret);
+		return ret;
+	}
+
+	pause_param = (struct hclge_cfg_pause_param_cmd *)desc.data;
+
+	seq_printf(s, "pause_trans_gap: 0x%x\n", pause_param->pause_trans_gap);
+	seq_printf(s, "pause_trans_time: 0x%x\n",
+		   le16_to_cpu(pause_param->pause_trans_time));
+	return 0;
+}
+
+#define HCLGE_DBG_TC_MASK		0x0F
+
+static int hclge_dbg_dump_qos_pri_map(struct seq_file *s, void *data)
+{
+#define HCLGE_DBG_TC_BIT_WIDTH		4
+
+	struct hclge_dev *hdev = hclge_seq_file_to_hdev(s);
+	struct hclge_qos_pri_map_cmd *pri_map;
+	struct hclge_desc desc;
+	u8 *pri_tc;
+	u8 tc, i;
+	int ret;
+
+	hclge_cmd_setup_basic_desc(&desc, HCLGE_OPC_PRI_TO_TC_MAPPING, true);
+	ret = hclge_cmd_send(&hdev->hw, &desc, 1);
+	if (ret) {
+		dev_err(&hdev->pdev->dev,
+			"failed to dump qos pri map, ret = %d\n", ret);
+		return ret;
+	}
+
+	pri_map = (struct hclge_qos_pri_map_cmd *)desc.data;
+
+	seq_printf(s, "vlan_to_pri: 0x%x\n", pri_map->vlan_pri);
+	seq_puts(s, "PRI  TC\n");
+
+	pri_tc = (u8 *)pri_map;
+	for (i = 0; i < HNAE3_MAX_TC; i++) {
+		tc = pri_tc[i >> 1] >> ((i & 1) * HCLGE_DBG_TC_BIT_WIDTH);
+		tc &= HCLGE_DBG_TC_MASK;
+		seq_printf(s, "%u     %u\n", i, tc);
+	}
+
+	return 0;
+}
+
+static int hclge_dbg_dump_qos_dscp_map(struct seq_file *s, void *data)
+{
+	struct hclge_dev *hdev = hclge_seq_file_to_hdev(s);
+	struct hclge_desc desc[HCLGE_DSCP_MAP_TC_BD_NUM];
+	struct hnae3_knic_private_info *kinfo;
+	u8 *req0 = (u8 *)desc[0].data;
+	u8 *req1 = (u8 *)desc[1].data;
+	u8 dscp_tc[HNAE3_MAX_DSCP];
+	int ret;
+	u8 i, j;
+
+	kinfo = &hdev->vport[0].nic.kinfo;
+
+	seq_printf(s, "tc map mode: %s\n", tc_map_mode_str[kinfo->tc_map_mode]);
+
+	if (kinfo->tc_map_mode != HNAE3_TC_MAP_MODE_DSCP)
+		return 0;
+
+	hclge_cmd_setup_basic_desc(&desc[0], HCLGE_OPC_QOS_MAP, true);
+	desc[0].flag |= cpu_to_le16(HCLGE_COMM_CMD_FLAG_NEXT);
+	hclge_cmd_setup_basic_desc(&desc[1], HCLGE_OPC_QOS_MAP, true);
+	ret = hclge_cmd_send(&hdev->hw, desc, HCLGE_DSCP_MAP_TC_BD_NUM);
+	if (ret) {
+		dev_err(&hdev->pdev->dev,
+			"failed to dump qos dscp map, ret = %d\n", ret);
+		return ret;
+	}
+
+	seq_puts(s, "\nDSCP  PRIO  TC\n");
+
+	/* The low 32 dscp setting use bd0, high 32 dscp setting use bd1 */
+	for (i = 0; i < HNAE3_MAX_DSCP / HCLGE_DSCP_MAP_TC_BD_NUM; i++) {
+		j = i + HNAE3_MAX_DSCP / HCLGE_DSCP_MAP_TC_BD_NUM;
+		/* Each dscp setting has 4 bits, so each byte saves two dscp
+		 * setting
+		 */
+		dscp_tc[i] = req0[i >> 1] >> HCLGE_DSCP_TC_SHIFT(i);
+		dscp_tc[j] = req1[i >> 1] >> HCLGE_DSCP_TC_SHIFT(i);
+		dscp_tc[i] &= HCLGE_DBG_TC_MASK;
+		dscp_tc[j] &= HCLGE_DBG_TC_MASK;
+	}
+
+	for (i = 0; i < HNAE3_MAX_DSCP; i++) {
+		if (kinfo->dscp_prio[i] == HNAE3_PRIO_ID_INVALID)
+			continue;
+
+		seq_printf(s, " %2u    %u    %u\n", i, kinfo->dscp_prio[i],
+			   dscp_tc[i]);
+	}
+
+	return 0;
+}
+
+static int hclge_dbg_dump_tx_buf_cfg(struct hclge_dev *hdev, struct seq_file *s)
+{
+	struct hclge_tx_buff_alloc_cmd *tx_buf_cmd;
+	struct hclge_desc desc;
+	int i, ret;
+
+	hclge_cmd_setup_basic_desc(&desc, HCLGE_OPC_TX_BUFF_ALLOC, true);
+	ret = hclge_cmd_send(&hdev->hw, &desc, 1);
+	if (ret) {
+		dev_err(&hdev->pdev->dev,
+			"failed to dump tx buf, ret = %d\n", ret);
+		return ret;
+	}
+
+	tx_buf_cmd = (struct hclge_tx_buff_alloc_cmd *)desc.data;
+	for (i = 0; i < HCLGE_MAX_TC_NUM; i++)
+		seq_printf(s, "tx_packet_buf_tc_%d: 0x%x\n", i,
+			   le16_to_cpu(tx_buf_cmd->tx_pkt_buff[i]));
+
+	return 0;
+}
+
+static int hclge_dbg_dump_rx_priv_buf_cfg(struct hclge_dev *hdev,
+					  struct seq_file *s)
+{
+	struct hclge_rx_priv_buff_cmd *rx_buf_cmd;
+	struct hclge_desc desc;
+	int i, ret;
+
+	hclge_cmd_setup_basic_desc(&desc, HCLGE_OPC_RX_PRIV_BUFF_ALLOC, true);
+	ret = hclge_cmd_send(&hdev->hw, &desc, 1);
+	if (ret) {
+		dev_err(&hdev->pdev->dev,
+			"failed to dump rx priv buf, ret = %d\n", ret);
+		return ret;
+	}
+
+	seq_puts(s, "\n");
+
+	rx_buf_cmd = (struct hclge_rx_priv_buff_cmd *)desc.data;
+	for (i = 0; i < HCLGE_MAX_TC_NUM; i++)
+		seq_printf(s, "rx_packet_buf_tc_%d: 0x%x\n", i,
+			   le16_to_cpu(rx_buf_cmd->buf_num[i]));
+
+	seq_printf(s, "rx_share_buf: 0x%x\n",
+		   le16_to_cpu(rx_buf_cmd->shared_buf));
+
+	return 0;
+}
+
+static int hclge_dbg_dump_rx_common_wl_cfg(struct hclge_dev *hdev,
+					   struct seq_file *s)
+{
+	struct hclge_rx_com_wl *rx_com_wl;
+	struct hclge_desc desc;
+	int ret;
+
+	hclge_cmd_setup_basic_desc(&desc, HCLGE_OPC_RX_COM_WL_ALLOC, true);
+	ret = hclge_cmd_send(&hdev->hw, &desc, 1);
+	if (ret) {
+		dev_err(&hdev->pdev->dev,
+			"failed to dump rx common wl, ret = %d\n", ret);
+		return ret;
+	}
+
+	rx_com_wl = (struct hclge_rx_com_wl *)desc.data;
+	seq_puts(s, "\n");
+	seq_printf(s, "rx_com_wl: high: 0x%x, low: 0x%x\n",
+		   le16_to_cpu(rx_com_wl->com_wl.high),
+		   le16_to_cpu(rx_com_wl->com_wl.low));
+
+	return 0;
+}
+
+static int hclge_dbg_dump_rx_global_pkt_cnt(struct hclge_dev *hdev,
+					    struct seq_file *s)
+{
+	struct hclge_rx_com_wl *rx_packet_cnt;
+	struct hclge_desc desc;
+	int ret;
+
+	hclge_cmd_setup_basic_desc(&desc, HCLGE_OPC_RX_GBL_PKT_CNT, true);
+	ret = hclge_cmd_send(&hdev->hw, &desc, 1);
+	if (ret) {
+		dev_err(&hdev->pdev->dev,
+			"failed to dump rx global pkt cnt, ret = %d\n", ret);
+		return ret;
+	}
+
+	rx_packet_cnt = (struct hclge_rx_com_wl *)desc.data;
+	seq_printf(s, "rx_global_packet_cnt: high: 0x%x, low: 0x%x\n",
+		   le16_to_cpu(rx_packet_cnt->com_wl.high),
+		   le16_to_cpu(rx_packet_cnt->com_wl.low));
+
+	return 0;
+}
+
+static int hclge_dbg_dump_rx_priv_wl_buf_cfg(struct hclge_dev *hdev,
+					     struct seq_file *s)
+{
+	struct hclge_rx_priv_wl_buf *rx_priv_wl;
+	struct hclge_desc desc[2];
+	int i, ret;
+
+	hclge_cmd_setup_basic_desc(&desc[0], HCLGE_OPC_RX_PRIV_WL_ALLOC, true);
+	desc[0].flag |= cpu_to_le16(HCLGE_COMM_CMD_FLAG_NEXT);
+	hclge_cmd_setup_basic_desc(&desc[1], HCLGE_OPC_RX_PRIV_WL_ALLOC, true);
+	ret = hclge_cmd_send(&hdev->hw, desc, 2);
+	if (ret) {
+		dev_err(&hdev->pdev->dev,
+			"failed to dump rx priv wl buf, ret = %d\n", ret);
+		return ret;
+	}
+
+	rx_priv_wl = (struct hclge_rx_priv_wl_buf *)desc[0].data;
+	for (i = 0; i < HCLGE_TC_NUM_ONE_DESC; i++)
+		seq_printf(s, "rx_priv_wl_tc_%d: high: 0x%x, low: 0x%x\n", i,
+			   le16_to_cpu(rx_priv_wl->tc_wl[i].high),
+			   le16_to_cpu(rx_priv_wl->tc_wl[i].low));
+
+	rx_priv_wl = (struct hclge_rx_priv_wl_buf *)desc[1].data;
+	for (i = 0; i < HCLGE_TC_NUM_ONE_DESC; i++)
+		seq_printf(s, "rx_priv_wl_tc_%d: high: 0x%x, low: 0x%x\n",
+			   i + HCLGE_TC_NUM_ONE_DESC,
+			   le16_to_cpu(rx_priv_wl->tc_wl[i].high),
+			   le16_to_cpu(rx_priv_wl->tc_wl[i].low));
+
+	return 0;
+}
+
+static int hclge_dbg_dump_rx_common_threshold_cfg(struct hclge_dev *hdev,
+						  struct seq_file *s)
+{
+	struct hclge_rx_com_thrd *rx_com_thrd;
+	struct hclge_desc desc[2];
+	int i, ret;
+
+	hclge_cmd_setup_basic_desc(&desc[0], HCLGE_OPC_RX_COM_THRD_ALLOC, true);
+	desc[0].flag |= cpu_to_le16(HCLGE_COMM_CMD_FLAG_NEXT);
+	hclge_cmd_setup_basic_desc(&desc[1], HCLGE_OPC_RX_COM_THRD_ALLOC, true);
+	ret = hclge_cmd_send(&hdev->hw, desc, 2);
+	if (ret) {
+		dev_err(&hdev->pdev->dev,
+			"failed to dump rx common threshold, ret = %d\n", ret);
+		return ret;
+	}
+
+	seq_puts(s, "\n");
+	rx_com_thrd = (struct hclge_rx_com_thrd *)desc[0].data;
+	for (i = 0; i < HCLGE_TC_NUM_ONE_DESC; i++)
+		seq_printf(s, "rx_com_thrd_tc_%d: high: 0x%x, low: 0x%x\n", i,
+			   le16_to_cpu(rx_com_thrd->com_thrd[i].high),
+			   le16_to_cpu(rx_com_thrd->com_thrd[i].low));
+
+	rx_com_thrd = (struct hclge_rx_com_thrd *)desc[1].data;
+	for (i = 0; i < HCLGE_TC_NUM_ONE_DESC; i++)
+		seq_printf(s, "rx_com_thrd_tc_%d: high: 0x%x, low: 0x%x\n",
+			   i + HCLGE_TC_NUM_ONE_DESC,
+			   le16_to_cpu(rx_com_thrd->com_thrd[i].high),
+			   le16_to_cpu(rx_com_thrd->com_thrd[i].low));
+
+	return 0;
+}
+
+static int hclge_dbg_dump_qos_buf_cfg(struct seq_file *s, void *data)
+{
+	struct hclge_dev *hdev = hclge_seq_file_to_hdev(s);
+	int ret;
+
+	ret = hclge_dbg_dump_tx_buf_cfg(hdev, s);
+	if (ret < 0)
+		return ret;
+
+	ret = hclge_dbg_dump_rx_priv_buf_cfg(hdev, s);
+	if (ret < 0)
+		return ret;
+
+	ret = hclge_dbg_dump_rx_common_wl_cfg(hdev, s);
+	if (ret < 0)
+		return ret;
+
+	ret = hclge_dbg_dump_rx_global_pkt_cnt(hdev, s);
+	if (ret < 0)
+		return ret;
+
+	seq_puts(s, "\n");
+	if (!hnae3_dev_dcb_supported(hdev))
+		return 0;
+
+	ret = hclge_dbg_dump_rx_priv_wl_buf_cfg(hdev, s);
+	if (ret < 0)
+		return ret;
+
+	ret = hclge_dbg_dump_rx_common_threshold_cfg(hdev, s);
+	if (ret < 0)
+		return ret;
+
+	return 0;
+}
+
+static int hclge_dbg_dump_mng_table(struct seq_file *s, void *data)
+{
+	struct hclge_dev *hdev = hclge_seq_file_to_hdev(s);
+	struct hclge_mac_ethertype_idx_rd_cmd *req0;
+	struct hclge_desc desc;
+	u32 msg_egress_port;
+	int ret, i;
+
+	seq_puts(s, "entry  mac_addr          mask  ether  ");
+	seq_puts(s, "mask  vlan  mask  i_map  i_dir  e_type  ");
+	seq_puts(s, "pf_id  vf_id  q_id  drop\n");
+
+	for (i = 0; i < HCLGE_DBG_MNG_TBL_MAX; i++) {
+		hclge_cmd_setup_basic_desc(&desc, HCLGE_MAC_ETHERTYPE_IDX_RD,
+					   true);
+		req0 = (struct hclge_mac_ethertype_idx_rd_cmd *)desc.data;
+		req0->index = cpu_to_le16(i);
+
+		ret = hclge_cmd_send(&hdev->hw, &desc, 1);
+		if (ret) {
+			dev_err(&hdev->pdev->dev,
+				"failed to dump manage table, ret = %d\n", ret);
+			return ret;
+		}
+
+		if (!req0->resp_code)
+			continue;
+
+		seq_printf(s, "%02u     %pM ",
+			   le16_to_cpu(req0->index), req0->mac_addr);
+
+		seq_printf(s, "%x     %04x   %x     %04x  ",
+			   !!(req0->flags & HCLGE_DBG_MNG_MAC_MASK_B),
+			   le16_to_cpu(req0->ethter_type),
+			   !!(req0->flags & HCLGE_DBG_MNG_ETHER_MASK_B),
+			   le16_to_cpu(req0->vlan_tag) &
+			   HCLGE_DBG_MNG_VLAN_TAG);
+
+		seq_printf(s, "%x     %02x     %02x     ",
+			   !!(req0->flags & HCLGE_DBG_MNG_VLAN_MASK_B),
+			   req0->i_port_bitmap, req0->i_port_direction);
+
+		msg_egress_port = le16_to_cpu(req0->egress_port);
+		seq_printf(s, "%x       %x      %02x     %04x  %x\n",
+			   !!(msg_egress_port & HCLGE_DBG_MNG_E_TYPE_B),
+			   msg_egress_port & HCLGE_DBG_MNG_PF_ID,
+			   (msg_egress_port >> 3) & HCLGE_DBG_MNG_VF_ID,
+			   le16_to_cpu(req0->egress_queue),
+			   !!(msg_egress_port & HCLGE_DBG_MNG_DROP_B));
+	}
+
+	return 0;
+}
+
+static int hclge_dbg_fd_tcam_read(struct hclge_dev *hdev, bool sel_x,
+				  struct seq_file *s,
+				  struct hclge_dbg_tcam_msg tcam_msg)
+{
+	struct hclge_fd_tcam_config_1_cmd *req1;
+	struct hclge_fd_tcam_config_2_cmd *req2;
+	struct hclge_fd_tcam_config_3_cmd *req3;
+	struct hclge_desc desc[3];
+	int ret, i;
+	__le32 *req;
+
+	hclge_cmd_setup_basic_desc(&desc[0], HCLGE_OPC_FD_TCAM_OP, true);
+	desc[0].flag |= cpu_to_le16(HCLGE_COMM_CMD_FLAG_NEXT);
+	hclge_cmd_setup_basic_desc(&desc[1], HCLGE_OPC_FD_TCAM_OP, true);
+	desc[1].flag |= cpu_to_le16(HCLGE_COMM_CMD_FLAG_NEXT);
+	hclge_cmd_setup_basic_desc(&desc[2], HCLGE_OPC_FD_TCAM_OP, true);
+
+	req1 = (struct hclge_fd_tcam_config_1_cmd *)desc[0].data;
+	req2 = (struct hclge_fd_tcam_config_2_cmd *)desc[1].data;
+	req3 = (struct hclge_fd_tcam_config_3_cmd *)desc[2].data;
+
+	req1->stage  = tcam_msg.stage;
+	req1->xy_sel = sel_x ? 1 : 0;
+	req1->index  = cpu_to_le32(tcam_msg.loc);
+
+	ret = hclge_cmd_send(&hdev->hw, desc, 3);
+	if (ret)
+		return ret;
+
+	seq_printf(s, "read result tcam key %s(%u):\n",
+		   sel_x ? "x" : "y", tcam_msg.loc);
+
+	/* tcam_data0 ~ tcam_data1 */
+	req = (__le32 *)req1->tcam_data;
+	for (i = 0; i < 2; i++)
+		seq_printf(s, "%08x\n", le32_to_cpu(*req++));
+
+	/* tcam_data2 ~ tcam_data7 */
+	req = (__le32 *)req2->tcam_data;
+	for (i = 0; i < 6; i++)
+		seq_printf(s, "%08x\n", le32_to_cpu(*req++));
+
+	/* tcam_data8 ~ tcam_data12 */
+	req = (__le32 *)req3->tcam_data;
+	for (i = 0; i < 5; i++)
+		seq_printf(s, "%08x\n", le32_to_cpu(*req++));
+
+	return ret;
+}
+
+static int hclge_dbg_get_rules_location(struct hclge_dev *hdev, u16 *rule_locs)
+{
+	struct hclge_fd_rule *rule;
+	struct hlist_node *node;
+	int cnt = 0;
+
+	spin_lock_bh(&hdev->fd_rule_lock);
+	hlist_for_each_entry_safe(rule, node, &hdev->fd_rule_list, rule_node) {
+		rule_locs[cnt] = rule->location;
+		cnt++;
+	}
+	spin_unlock_bh(&hdev->fd_rule_lock);
+
+	if (cnt != hdev->hclge_fd_rule_num || cnt == 0)
+		return -EINVAL;
+
+	return cnt;
+}
+
+static int hclge_dbg_dump_fd_tcam(struct seq_file *s, void *data)
+{
+	struct hclge_dev *hdev = hclge_seq_file_to_hdev(s);
+	struct hclge_dbg_tcam_msg tcam_msg;
+	int i, ret, rule_cnt;
+	u16 *rule_locs;
+	u32 rule_num;
+
+	if (!hnae3_ae_dev_fd_supported(hdev->ae_dev)) {
+		dev_err(&hdev->pdev->dev,
+			"Only FD-supported dev supports dump fd tcam\n");
+		return -EOPNOTSUPP;
+	}
+
+	rule_num = hdev->fd_cfg.rule_num[HCLGE_FD_STAGE_1];
+	if (!hdev->hclge_fd_rule_num || !rule_num)
+		return 0;
+
+	rule_locs = kcalloc(rule_num, sizeof(u16), GFP_KERNEL);
+	if (!rule_locs)
+		return -ENOMEM;
+
+	rule_cnt = hclge_dbg_get_rules_location(hdev, rule_locs);
+	if (rule_cnt < 0) {
+		ret = rule_cnt;
+		dev_err(&hdev->pdev->dev,
+			"failed to get rule number, ret = %d\n", ret);
+		goto out;
+	}
+
+	ret = 0;
+	for (i = 0; i < rule_cnt; i++) {
+		tcam_msg.stage = HCLGE_FD_STAGE_1;
+		tcam_msg.loc = rule_locs[i];
+
+		ret = hclge_dbg_fd_tcam_read(hdev, true, s, tcam_msg);
+		if (ret) {
+			dev_err(&hdev->pdev->dev,
+				"failed to get fd tcam key x, ret = %d\n", ret);
+			goto out;
+		}
+
+		ret = hclge_dbg_fd_tcam_read(hdev, false, s, tcam_msg);
+		if (ret) {
+			dev_err(&hdev->pdev->dev,
+				"failed to get fd tcam key y, ret = %d\n", ret);
+			goto out;
+		}
+
+	}
+
+out:
+	kfree(rule_locs);
+	return ret;
+}
+
+static int hclge_dbg_dump_fd_counter(struct seq_file *s, void *data)
+{
+	struct hclge_dev *hdev = hclge_seq_file_to_hdev(s);
+	u8 func_num = pci_num_vf(hdev->pdev) + 1; /* pf and enabled vf num */
+	struct hclge_fd_ad_cnt_read_cmd *req;
+	char str_id[HCLGE_DBG_ID_LEN];
+	struct hclge_desc desc;
+	int ret;
+	u64 cnt;
+	u8 i;
+
+	if (!hnae3_ae_dev_fd_supported(hdev->ae_dev))
+		return -EOPNOTSUPP;
+
+	seq_puts(s, "func_id\thit_times\n");
+
+	for (i = 0; i < func_num; i++) {
+		hclge_cmd_setup_basic_desc(&desc, HCLGE_OPC_FD_CNT_OP, true);
+		req = (struct hclge_fd_ad_cnt_read_cmd *)desc.data;
+		req->index = cpu_to_le16(i);
+		ret = hclge_cmd_send(&hdev->hw, &desc, 1);
+		if (ret) {
+			dev_err(&hdev->pdev->dev, "failed to get fd counter, ret = %d\n",
+				ret);
+			return ret;
+		}
+		cnt = le64_to_cpu(req->cnt);
+		hclge_dbg_get_func_id_str(str_id, i);
+		seq_printf(s, "%s\t%llu\n", str_id, cnt);
+	}
+
+	return 0;
+}
+
+static const struct hclge_dbg_status_dfx_info hclge_dbg_rst_info[] = {
+	{HCLGE_MISC_VECTOR_REG_BASE, "vector0 interrupt enable status"},
+	{HCLGE_MISC_RESET_STS_REG,   "reset interrupt source"},
+	{HCLGE_MISC_VECTOR_INT_STS,  "reset interrupt status"},
+	{HCLGE_RAS_PF_OTHER_INT_STS_REG, "RAS interrupt status"},
+	{HCLGE_GLOBAL_RESET_REG,  "hardware reset status"},
+	{HCLGE_NIC_CSQ_DEPTH_REG, "handshake status"},
+	{HCLGE_FUN_RST_ING, "function reset status"}
+};
+
+int hclge_dbg_dump_rst_info(struct hclge_dev *hdev, char *buf, int len)
+{
+	u32 i, offset;
+	int pos = 0;
+
+	pos += scnprintf(buf + pos, len - pos, "PF reset count: %u\n",
+			 hdev->rst_stats.pf_rst_cnt);
+	pos += scnprintf(buf + pos, len - pos, "FLR reset count: %u\n",
+			 hdev->rst_stats.flr_rst_cnt);
+	pos += scnprintf(buf + pos, len - pos, "GLOBAL reset count: %u\n",
+			 hdev->rst_stats.global_rst_cnt);
+	pos += scnprintf(buf + pos, len - pos, "IMP reset count: %u\n",
+			 hdev->rst_stats.imp_rst_cnt);
+	pos += scnprintf(buf + pos, len - pos, "reset done count: %u\n",
+			 hdev->rst_stats.reset_done_cnt);
+	pos += scnprintf(buf + pos, len - pos, "HW reset done count: %u\n",
+			 hdev->rst_stats.hw_reset_done_cnt);
+	pos += scnprintf(buf + pos, len - pos, "reset count: %u\n",
+			 hdev->rst_stats.reset_cnt);
+	pos += scnprintf(buf + pos, len - pos, "reset fail count: %u\n",
+			 hdev->rst_stats.reset_fail_cnt);
+
+	for (i = 0; i < ARRAY_SIZE(hclge_dbg_rst_info); i++) {
+		offset = hclge_dbg_rst_info[i].offset;
+		pos += scnprintf(buf + pos, len - pos, "%s: 0x%x\n",
+				 hclge_dbg_rst_info[i].message,
+				 hclge_read_dev(&hdev->hw, offset));
+	}
+
+	pos += scnprintf(buf + pos, len - pos, "hdev state: 0x%lx\n",
+			 hdev->state);
+
+	return 0;
+}
+
+static int hclge_dbg_seq_dump_rst_info(struct seq_file *s, void *data)
+{
+	struct hclge_dev *hdev = hclge_seq_file_to_hdev(s);
+	u32 i, offset;
+
+	seq_printf(s, "PF reset count: %u\n", hdev->rst_stats.pf_rst_cnt);
+	seq_printf(s, "FLR reset count: %u\n", hdev->rst_stats.flr_rst_cnt);
+	seq_printf(s, "GLOBAL reset count: %u\n",
+		   hdev->rst_stats.global_rst_cnt);
+	seq_printf(s, "IMP reset count: %u\n", hdev->rst_stats.imp_rst_cnt);
+	seq_printf(s, "reset done count: %u\n", hdev->rst_stats.reset_done_cnt);
+	seq_printf(s, "HW reset done count: %u\n",
+		   hdev->rst_stats.hw_reset_done_cnt);
+	seq_printf(s, "reset count: %u\n", hdev->rst_stats.reset_cnt);
+	seq_printf(s, "reset fail count: %u\n", hdev->rst_stats.reset_fail_cnt);
+
+	for (i = 0; i < ARRAY_SIZE(hclge_dbg_rst_info); i++) {
+		offset = hclge_dbg_rst_info[i].offset;
+		seq_printf(s, "%s: 0x%x\n",
+			   hclge_dbg_rst_info[i].message,
+			   hclge_read_dev(&hdev->hw, offset));
+	}
+
+	seq_printf(s, "hdev state: 0x%lx\n", hdev->state);
+
+	return 0;
+}
+
+static int hclge_dbg_dump_serv_info(struct seq_file *s, void *data)
+{
+	struct hclge_dev *hdev = hclge_seq_file_to_hdev(s);
+	unsigned long rem_nsec;
+	u64 lc;
+
+	lc = local_clock();
+	rem_nsec = do_div(lc, HCLGE_BILLION_NANO_SECONDS);
+
+	seq_printf(s, "local_clock: [%5lu.%06lu]\n",
+		   (unsigned long)lc, rem_nsec / 1000);
+	seq_printf(s, "delta: %u(ms)\n",
+		   jiffies_to_msecs(jiffies - hdev->last_serv_processed));
+	seq_printf(s, "last_service_task_processed: %lu(jiffies)\n",
+		   hdev->last_serv_processed);
+	seq_printf(s, "last_service_task_cnt: %lu\n", hdev->serv_processed_cnt);
+
+	return 0;
+}
+
+static int hclge_dbg_dump_interrupt(struct seq_file *s, void *data)
+{
+	struct hclge_dev *hdev = hclge_seq_file_to_hdev(s);
+
+	seq_printf(s, "num_nic_msi: %u\n", hdev->num_nic_msi);
+	seq_printf(s, "num_roce_msi: %u\n", hdev->num_roce_msi);
+	seq_printf(s, "num_msi_used: %u\n", hdev->num_msi_used);
+	seq_printf(s, "num_msi_left: %u\n", hdev->num_msi_left);
+
+	return 0;
+}
+
+static void hclge_dbg_imp_info_data_print(struct seq_file *s,
+					  struct hclge_desc *desc_src,
+					  u32 bd_num)
+{
+#define HCLGE_DBG_IMP_INFO_PRINT_OFFSET 0x2
+
+	struct hclge_desc *desc_index = desc_src;
+	u32 offset = 0;
+	u32 i, j;
+
+	seq_puts(s, "offset | data\n");
+
+	for (i = 0; i < bd_num; i++) {
+		j = 0;
+		while (j < HCLGE_DESC_DATA_LEN - 1) {
+			seq_printf(s, "0x%04x | ", offset);
+			seq_printf(s, "0x%08x  ",
+				   le32_to_cpu(desc_index->data[j++]));
+			seq_printf(s, "0x%08x\n",
+				   le32_to_cpu(desc_index->data[j++]));
+			offset += sizeof(u32) * HCLGE_DBG_IMP_INFO_PRINT_OFFSET;
+		}
+		desc_index++;
+	}
+}
+
+static int hclge_dbg_get_imp_stats_info(struct seq_file *s, void *data)
+{
+	struct hclge_dev *hdev = hclge_seq_file_to_hdev(s);
+	struct hclge_get_imp_bd_cmd *req;
+	struct hclge_desc *desc_src;
+	struct hclge_desc desc;
+	u32 bd_num;
+	int ret;
+
+	hclge_cmd_setup_basic_desc(&desc, HCLGE_OPC_IMP_STATS_BD, true);
+
+	req = (struct hclge_get_imp_bd_cmd *)desc.data;
+	ret = hclge_cmd_send(&hdev->hw, &desc, 1);
+	if (ret) {
+		dev_err(&hdev->pdev->dev,
+			"failed to get imp statistics bd number, ret = %d\n",
+			ret);
+		return ret;
+	}
+
+	bd_num = le32_to_cpu(req->bd_num);
+	if (!bd_num) {
+		dev_err(&hdev->pdev->dev, "imp statistics bd number is 0!\n");
+		return -EINVAL;
+	}
+
+	desc_src = kcalloc(bd_num, sizeof(struct hclge_desc), GFP_KERNEL);
+	if (!desc_src)
+		return -ENOMEM;
+
+	ret  = hclge_dbg_cmd_send(hdev, desc_src, 0, bd_num,
+				  HCLGE_OPC_IMP_STATS_INFO);
+	if (ret) {
+		kfree(desc_src);
+		dev_err(&hdev->pdev->dev,
+			"failed to get imp statistics, ret = %d\n", ret);
+		return ret;
+	}
+
+	hclge_dbg_imp_info_data_print(s, desc_src, bd_num);
+
+	kfree(desc_src);
+
+	return 0;
+}
+
+#define HCLGE_CMD_NCL_CONFIG_BD_NUM	5
+#define HCLGE_MAX_NCL_CONFIG_LENGTH	16384
+
+static void hclge_ncl_config_data_print(struct hclge_desc *desc, int *index,
+					struct seq_file *s)
+{
+#define HCLGE_CMD_DATA_NUM		6
+
+	int offset = HCLGE_MAX_NCL_CONFIG_LENGTH - *index;
+	int i, j;
+
+	for (i = 0; i < HCLGE_CMD_NCL_CONFIG_BD_NUM; i++) {
+		for (j = 0; j < HCLGE_CMD_DATA_NUM; j++) {
+			if (i == 0 && j == 0)
+				continue;
+
+			seq_printf(s, "0x%04x | 0x%08x\n", offset,
+				   le32_to_cpu(desc[i].data[j]));
+
+			offset += sizeof(u32);
+			*index -= sizeof(u32);
+
+			if (*index <= 0)
+				return;
+		}
+	}
+}
+
+static int hclge_dbg_dump_ncl_config(struct seq_file *s, void *data)
+{
+#define HCLGE_NCL_CONFIG_LENGTH_IN_EACH_CMD	(20 + 24 * 4)
+
+	struct hclge_desc desc[HCLGE_CMD_NCL_CONFIG_BD_NUM];
+	struct hclge_dev *hdev = hclge_seq_file_to_hdev(s);
+	int bd_num = HCLGE_CMD_NCL_CONFIG_BD_NUM;
+	int index = HCLGE_MAX_NCL_CONFIG_LENGTH;
+	u32 data0;
+	int ret;
+
+	seq_puts(s, "offset | data\n");
+
+	while (index > 0) {
+		data0 = HCLGE_MAX_NCL_CONFIG_LENGTH - index;
+		if (index >= HCLGE_NCL_CONFIG_LENGTH_IN_EACH_CMD)
+			data0 |= HCLGE_NCL_CONFIG_LENGTH_IN_EACH_CMD << 16;
+		else
+			data0 |= (u32)index << 16;
+		ret = hclge_dbg_cmd_send(hdev, desc, data0, bd_num,
+					 HCLGE_OPC_QUERY_NCL_CONFIG);
+		if (ret)
+			return ret;
+
+		hclge_ncl_config_data_print(desc, &index, s);
+	}
+
+	return 0;
+}
+
+static int hclge_dbg_dump_loopback(struct seq_file *s, void *data)
+{
+	struct hclge_dev *hdev = hclge_seq_file_to_hdev(s);
+	struct phy_device *phydev = hdev->hw.mac.phydev;
+	struct hclge_config_mac_mode_cmd *req_app;
+	struct hclge_common_lb_cmd *req_common;
+	struct hclge_desc desc;
+	u8 loopback_en;
+	int ret;
+
+	req_app = (struct hclge_config_mac_mode_cmd *)desc.data;
+	req_common = (struct hclge_common_lb_cmd *)desc.data;
+
+	seq_printf(s, "mac id: %u\n", hdev->hw.mac.mac_id);
+
+	hclge_cmd_setup_basic_desc(&desc, HCLGE_OPC_CONFIG_MAC_MODE, true);
+	ret = hclge_cmd_send(&hdev->hw, &desc, 1);
+	if (ret) {
+		dev_err(&hdev->pdev->dev,
+			"failed to dump app loopback status, ret = %d\n", ret);
+		return ret;
+	}
+
+	loopback_en = hnae3_get_bit(le32_to_cpu(req_app->txrx_pad_fcs_loop_en),
+				    HCLGE_MAC_APP_LP_B);
+	seq_printf(s, "app loopback: %s\n", str_on_off(loopback_en));
+
+	hclge_cmd_setup_basic_desc(&desc, HCLGE_OPC_COMMON_LOOPBACK, true);
+	ret = hclge_cmd_send(&hdev->hw, &desc, 1);
+	if (ret) {
+		dev_err(&hdev->pdev->dev,
+			"failed to dump common loopback status, ret = %d\n",
+			ret);
+		return ret;
+	}
+
+	loopback_en = req_common->enable &
+		      HCLGE_CMD_SERDES_SERIAL_INNER_LOOP_B;
+	seq_printf(s, "serdes serial loopback: %s\n", str_on_off(loopback_en));
+
+	loopback_en = req_common->enable &
+		      HCLGE_CMD_SERDES_PARALLEL_INNER_LOOP_B ? 1 : 0;
+	seq_printf(s, "serdes parallel loopback: %s\n",
+		   str_on_off(loopback_en));
+
+	if (phydev) {
+		loopback_en = phydev->loopback_enabled;
+		seq_printf(s, "phy loopback: %s\n", str_on_off(loopback_en));
+	} else if (hnae3_dev_phy_imp_supported(hdev)) {
+		loopback_en = req_common->enable &
+			      HCLGE_CMD_GE_PHY_INNER_LOOP_B;
+		seq_printf(s, "phy loopback: %s\n", str_on_off(loopback_en));
+	}
+
+	return 0;
+}
+
+/* hclge_dbg_dump_mac_tnl_status: print message about mac tnl interrupt
+ * @hdev: pointer to struct hclge_dev
+ */
+static int hclge_dbg_dump_mac_tnl_status(struct seq_file *s, void *data)
+{
+	struct hclge_dev *hdev = hclge_seq_file_to_hdev(s);
+	struct hclge_mac_tnl_stats stats;
+	unsigned long rem_nsec;
+
+	seq_puts(s, "Recently generated mac tnl interruption:\n");
+
+	while (kfifo_get(&hdev->mac_tnl_log, &stats)) {
+		rem_nsec = do_div(stats.time, HCLGE_BILLION_NANO_SECONDS);
+
+		seq_printf(s, "[%07lu.%03lu] status = 0x%x\n",
+			   (unsigned long)stats.time, rem_nsec / 1000,
+			   stats.status);
+	}
+
+	return 0;
+}
+
+static void hclge_dbg_dump_mac_list(struct seq_file *s, bool is_unicast)
+{
+	struct hclge_dev *hdev = hclge_seq_file_to_hdev(s);
+	struct hclge_mac_node *mac_node, *tmp;
+	struct hclge_vport *vport;
+	struct list_head *list;
+	u32 func_id;
+
+	seq_printf(s, "%s MAC_LIST:\n", is_unicast ? "UC" : "MC");
+	seq_puts(s, "FUNC_ID  MAC_ADDR            STATE\n");
+
+	for (func_id = 0; func_id < hdev->num_alloc_vport; func_id++) {
+		vport = &hdev->vport[func_id];
+		list = is_unicast ? &vport->uc_mac_list : &vport->mc_mac_list;
+		spin_lock_bh(&vport->mac_list_lock);
+		list_for_each_entry_safe(mac_node, tmp, list, node) {
+			if (func_id)
+				seq_printf(s, "vf%-7u", func_id - 1U);
+			else
+				seq_puts(s, "pf       ");
+			seq_printf(s, "%pM   ", mac_node->mac_addr);
+			seq_printf(s, "%5s\n",
+				   hclge_mac_state_str[mac_node->state]);
+		}
+		spin_unlock_bh(&vport->mac_list_lock);
+	}
+}
+
+static int hclge_dbg_dump_umv_info(struct seq_file *s, void *data)
+{
+	struct hclge_dev *hdev = hclge_seq_file_to_hdev(s);
+	u8 func_num = pci_num_vf(hdev->pdev) + 1;
+	struct hclge_vport *vport;
+	u8 i;
+
+	seq_printf(s, "num_alloc_vport   : %u\n", hdev->num_alloc_vport);
+	seq_printf(s, "max_umv_size     : %u\n", hdev->max_umv_size);
+	seq_printf(s, "wanted_umv_size  : %u\n", hdev->wanted_umv_size);
+	seq_printf(s, "priv_umv_size    : %u\n", hdev->priv_umv_size);
+
+	mutex_lock(&hdev->vport_lock);
+	seq_printf(s, "share_umv_size   : %u\n", hdev->share_umv_size);
+	for (i = 0; i < func_num; i++) {
+		vport = &hdev->vport[i];
+		seq_printf(s, "vport(%u) used_umv_num : %u\n",
+			   i, vport->used_umv_num);
+	}
+	mutex_unlock(&hdev->vport_lock);
+
+	seq_printf(s, "used_mc_mac_num  : %u\n", hdev->used_mc_mac_num);
+
+	return 0;
+}
+
+static int hclge_get_vlan_rx_offload_cfg(struct hclge_dev *hdev, u8 vf_id,
+					 struct hclge_dbg_vlan_cfg *vlan_cfg)
+{
+	struct hclge_vport_vtag_rx_cfg_cmd *req;
+	struct hclge_desc desc;
+	u16 bmap_index;
+	u8 rx_cfg;
+	int ret;
+
+	hclge_cmd_setup_basic_desc(&desc, HCLGE_OPC_VLAN_PORT_RX_CFG, true);
+
+	req = (struct hclge_vport_vtag_rx_cfg_cmd *)desc.data;
+	req->vf_offset = vf_id / HCLGE_VF_NUM_PER_CMD;
+	bmap_index = vf_id % HCLGE_VF_NUM_PER_CMD / HCLGE_VF_NUM_PER_BYTE;
+	req->vf_bitmap[bmap_index] = 1U << (vf_id % HCLGE_VF_NUM_PER_BYTE);
+
+	ret = hclge_cmd_send(&hdev->hw, &desc, 1);
+	if (ret) {
+		dev_err(&hdev->pdev->dev,
+			"failed to get vport%u rxvlan cfg, ret = %d\n",
+			vf_id, ret);
+		return ret;
+	}
+
+	rx_cfg = req->vport_vlan_cfg;
+	vlan_cfg->strip_tag1 = hnae3_get_bit(rx_cfg, HCLGE_REM_TAG1_EN_B);
+	vlan_cfg->strip_tag2 = hnae3_get_bit(rx_cfg, HCLGE_REM_TAG2_EN_B);
+	vlan_cfg->drop_tag1 = hnae3_get_bit(rx_cfg, HCLGE_DISCARD_TAG1_EN_B);
+	vlan_cfg->drop_tag2 = hnae3_get_bit(rx_cfg, HCLGE_DISCARD_TAG2_EN_B);
+	vlan_cfg->pri_only1 = hnae3_get_bit(rx_cfg, HCLGE_SHOW_TAG1_EN_B);
+	vlan_cfg->pri_only2 = hnae3_get_bit(rx_cfg, HCLGE_SHOW_TAG2_EN_B);
+
+	return 0;
+}
+
+static int hclge_get_vlan_tx_offload_cfg(struct hclge_dev *hdev, u8 vf_id,
+					 struct hclge_dbg_vlan_cfg *vlan_cfg)
+{
+	struct hclge_vport_vtag_tx_cfg_cmd *req;
+	struct hclge_desc desc;
+	u16 bmap_index;
+	u8 tx_cfg;
+	int ret;
+
+	hclge_cmd_setup_basic_desc(&desc, HCLGE_OPC_VLAN_PORT_TX_CFG, true);
+	req = (struct hclge_vport_vtag_tx_cfg_cmd *)desc.data;
+	req->vf_offset = vf_id / HCLGE_VF_NUM_PER_CMD;
+	bmap_index = vf_id % HCLGE_VF_NUM_PER_CMD / HCLGE_VF_NUM_PER_BYTE;
+	req->vf_bitmap[bmap_index] = 1U << (vf_id % HCLGE_VF_NUM_PER_BYTE);
+
+	ret = hclge_cmd_send(&hdev->hw, &desc, 1);
+	if (ret) {
+		dev_err(&hdev->pdev->dev,
+			"failed to get vport%u txvlan cfg, ret = %d\n",
+			vf_id, ret);
+		return ret;
+	}
+
+	tx_cfg = req->vport_vlan_cfg;
+	vlan_cfg->pvid = le16_to_cpu(req->def_vlan_tag1);
+
+	vlan_cfg->accept_tag1 = hnae3_get_bit(tx_cfg, HCLGE_ACCEPT_TAG1_B);
+	vlan_cfg->accept_tag2 = hnae3_get_bit(tx_cfg, HCLGE_ACCEPT_TAG2_B);
+	vlan_cfg->accept_untag1 = hnae3_get_bit(tx_cfg, HCLGE_ACCEPT_UNTAG1_B);
+	vlan_cfg->accept_untag2 = hnae3_get_bit(tx_cfg, HCLGE_ACCEPT_UNTAG2_B);
+	vlan_cfg->insert_tag1 = hnae3_get_bit(tx_cfg, HCLGE_PORT_INS_TAG1_EN_B);
+	vlan_cfg->insert_tag2 = hnae3_get_bit(tx_cfg, HCLGE_PORT_INS_TAG2_EN_B);
+	vlan_cfg->shift_tag = hnae3_get_bit(tx_cfg, HCLGE_TAG_SHIFT_MODE_EN_B);
+
+	return 0;
+}
+
+static int hclge_get_vlan_filter_config_cmd(struct hclge_dev *hdev,
+					    u8 vlan_type, u8 vf_id,
+					    struct hclge_desc *desc)
+{
+	struct hclge_vlan_filter_ctrl_cmd *req;
+	int ret;
+
+	hclge_cmd_setup_basic_desc(desc, HCLGE_OPC_VLAN_FILTER_CTRL, true);
+	req = (struct hclge_vlan_filter_ctrl_cmd *)desc->data;
+	req->vlan_type = vlan_type;
+	req->vf_id = vf_id;
+
+	ret = hclge_cmd_send(&hdev->hw, desc, 1);
+	if (ret)
+		dev_err(&hdev->pdev->dev,
+			"failed to get vport%u vlan filter config, ret = %d.\n",
+			vf_id, ret);
+
+	return ret;
+}
+
+static int hclge_get_vlan_filter_state(struct hclge_dev *hdev, u8 vlan_type,
+				       u8 vf_id, u8 *vlan_fe)
+{
+	struct hclge_vlan_filter_ctrl_cmd *req;
+	struct hclge_desc desc;
+	int ret;
+
+	ret = hclge_get_vlan_filter_config_cmd(hdev, vlan_type, vf_id, &desc);
+	if (ret)
+		return ret;
+
+	req = (struct hclge_vlan_filter_ctrl_cmd *)desc.data;
+	*vlan_fe = req->vlan_fe;
+
+	return 0;
+}
+
+static int hclge_get_port_vlan_filter_bypass_state(struct hclge_dev *hdev,
+						   u8 vf_id, u8 *bypass_en)
+{
+	struct hclge_port_vlan_filter_bypass_cmd *req;
+	struct hclge_desc desc;
+	int ret;
+
+	if (!test_bit(HNAE3_DEV_SUPPORT_PORT_VLAN_BYPASS_B, hdev->ae_dev->caps))
+		return 0;
+
+	hclge_cmd_setup_basic_desc(&desc, HCLGE_OPC_PORT_VLAN_BYPASS, true);
+	req = (struct hclge_port_vlan_filter_bypass_cmd *)desc.data;
+	req->vf_id = vf_id;
+
+	ret = hclge_cmd_send(&hdev->hw, &desc, 1);
+	if (ret) {
+		dev_err(&hdev->pdev->dev,
+			"failed to get vport%u port vlan filter bypass state, ret = %d.\n",
+			vf_id, ret);
+		return ret;
+	}
+
+	*bypass_en = hnae3_get_bit(req->bypass_state, HCLGE_INGRESS_BYPASS_B);
+
+	return 0;
+}
+
+static int hclge_dbg_dump_vlan_filter_config(struct hclge_dev *hdev,
+					     struct seq_file *s)
+{
+	u8 func_num = pci_num_vf(hdev->pdev) + 1; /* pf and enabled vf num */
+	u8 i, vlan_fe, bypass, ingress, egress;
+	char str_id[HCLGE_DBG_ID_LEN];
+	int ret;
+
+	ret = hclge_get_vlan_filter_state(hdev, HCLGE_FILTER_TYPE_PORT, 0,
+					  &vlan_fe);
+	if (ret)
+		return ret;
+	ingress = vlan_fe & HCLGE_FILTER_FE_NIC_INGRESS_B;
+	egress = vlan_fe & HCLGE_FILTER_FE_NIC_EGRESS_B ? 1 : 0;
+
+	seq_printf(s, "I_PORT_VLAN_FILTER: %s\n", str_on_off(ingress));
+	seq_printf(s, "E_PORT_VLAN_FILTER: %s\n", str_on_off(egress));
+
+	seq_puts(s, "FUNC_ID  I_VF_VLAN_FILTER  E_VF_VLAN_FILTER  ");
+	seq_puts(s, "PORT_VLAN_FILTER_BYPASS\n");
+
+	for (i = 0; i < func_num; i++) {
+		ret = hclge_get_vlan_filter_state(hdev, HCLGE_FILTER_TYPE_VF, i,
+						  &vlan_fe);
+		if (ret)
+			return ret;
+
+		ingress = vlan_fe & HCLGE_FILTER_FE_NIC_INGRESS_B;
+		egress = vlan_fe & HCLGE_FILTER_FE_NIC_EGRESS_B ? 1 : 0;
+		ret = hclge_get_port_vlan_filter_bypass_state(hdev, i, &bypass);
+		if (ret)
+			return ret;
+
+		seq_printf(s, "%-9s%-18s%-18s%s\n",
+			   hclge_dbg_get_func_id_str(str_id, i),
+			   str_on_off(ingress), str_on_off(egress),
+			   test_bit(HNAE3_DEV_SUPPORT_PORT_VLAN_BYPASS_B,
+				    hdev->ae_dev->caps) ?
+						str_on_off(bypass) : "NA");
+	}
+	seq_puts(s, "\n");
+
+	return 0;
+}
+
+static int hclge_dbg_dump_vlan_offload_config(struct hclge_dev *hdev,
+					      struct seq_file *s)
+{
+	u8 func_num = pci_num_vf(hdev->pdev) + 1; /* pf and enabled vf num */
+	struct hclge_dbg_vlan_cfg vlan_cfg;
+	char str_id[HCLGE_DBG_ID_LEN];
+	int ret;
+	u8 i;
+
+	seq_puts(s, "FUNC_ID  PVID  ACCEPT_TAG1  ACCEPT_TAG2 ACCEPT_UNTAG1  ");
+	seq_puts(s, "ACCEPT_UNTAG2  INSERT_TAG1  INSERT_TAG2  SHIFT_TAG  ");
+	seq_puts(s, "STRIP_TAG1  STRIP_TAG2  DROP_TAG1  DROP_TAG2  ");
+	seq_puts(s, "PRI_ONLY_TAG1  PRI_ONLY_TAG2\n");
+
+	for (i = 0; i < func_num; i++) {
+		ret = hclge_get_vlan_tx_offload_cfg(hdev, i, &vlan_cfg);
+		if (ret)
+			return ret;
+
+		ret = hclge_get_vlan_rx_offload_cfg(hdev, i, &vlan_cfg);
+		if (ret)
+			return ret;
+
+		seq_printf(s, "%-9s", hclge_dbg_get_func_id_str(str_id, i));
+		seq_printf(s, "%-6u", vlan_cfg.pvid);
+		seq_printf(s, "%-13s", str_on_off(vlan_cfg.accept_tag1));
+		seq_printf(s, "%-12s", str_on_off(vlan_cfg.accept_tag2));
+		seq_printf(s, "%-15s", str_on_off(vlan_cfg.accept_untag1));
+		seq_printf(s, "%-15s", str_on_off(vlan_cfg.accept_untag2));
+		seq_printf(s, "%-13s", str_on_off(vlan_cfg.insert_tag1));
+		seq_printf(s, "%-13s", str_on_off(vlan_cfg.insert_tag2));
+		seq_printf(s, "%-11s", str_on_off(vlan_cfg.shift_tag));
+		seq_printf(s, "%-12s", str_on_off(vlan_cfg.strip_tag1));
+		seq_printf(s, "%-12s", str_on_off(vlan_cfg.strip_tag2));
+		seq_printf(s, "%-11s", str_on_off(vlan_cfg.drop_tag1));
+		seq_printf(s, "%-11s", str_on_off(vlan_cfg.drop_tag2));
+		seq_printf(s, "%-15s", str_on_off(vlan_cfg.pri_only1));
+		seq_printf(s, "%s\n", str_on_off(vlan_cfg.pri_only2));
+	}
+
+	return 0;
+}
+
+static int hclge_dbg_dump_vlan_config(struct seq_file *s, void *data)
+{
+	struct hclge_dev *hdev = hclge_seq_file_to_hdev(s);
+	int ret;
+
+	ret = hclge_dbg_dump_vlan_filter_config(hdev, s);
+	if (ret)
+		return ret;
+
+	return hclge_dbg_dump_vlan_offload_config(hdev, s);
+}
+
+static int hclge_dbg_dump_ptp_info(struct seq_file *s, void *data)
+{
+	struct hclge_dev *hdev = hclge_seq_file_to_hdev(s);
+	struct hclge_ptp *ptp = hdev->ptp;
+	u32 sw_cfg = ptp->ptp_cfg;
+	unsigned int tx_start;
+	unsigned int last_rx;
+	u32 hw_cfg;
+	int ret;
+
+	seq_printf(s, "phc %s's debug info:\n", ptp->info.name);
+	seq_printf(s, "ptp enable: %s\n",
+		   str_yes_no(test_bit(HCLGE_PTP_FLAG_EN, &ptp->flags)));
+	seq_printf(s, "ptp tx enable: %s\n",
+		   str_yes_no(test_bit(HCLGE_PTP_FLAG_TX_EN, &ptp->flags)));
+	seq_printf(s, "ptp rx enable: %s\n",
+		   str_yes_no(test_bit(HCLGE_PTP_FLAG_RX_EN, &ptp->flags)));
+
+	last_rx = jiffies_to_msecs(ptp->last_rx);
+	seq_printf(s, "last rx time: %lu.%lu\n",
+		   last_rx / MSEC_PER_SEC, last_rx % MSEC_PER_SEC);
+	seq_printf(s, "rx count: %lu\n", ptp->rx_cnt);
+
+	tx_start = jiffies_to_msecs(ptp->tx_start);
+	seq_printf(s, "last tx start time: %lu.%lu\n",
+		   tx_start / MSEC_PER_SEC, tx_start % MSEC_PER_SEC);
+	seq_printf(s, "tx count: %lu\n", ptp->tx_cnt);
+	seq_printf(s, "tx skipped count: %lu\n", ptp->tx_skipped);
+	seq_printf(s, "tx timeout count: %lu\n", ptp->tx_timeout);
+	seq_printf(s, "last tx seqid: %u\n", ptp->last_tx_seqid);
+
+
+	ret = hclge_ptp_cfg_qry(hdev, &hw_cfg);
+	if (ret)
+		return ret;
+
+	seq_printf(s, "sw_cfg: %#x, hw_cfg: %#x\n", sw_cfg, hw_cfg);
+
+	seq_printf(s, "tx type: %d, rx filter: %d\n",
+		   ptp->ts_cfg.tx_type, ptp->ts_cfg.rx_filter);
+
+	return 0;
+}
+
+static int hclge_dbg_dump_mac_uc(struct seq_file *s, void *data)
+{
+	hclge_dbg_dump_mac_list(s, true);
+
+	return 0;
+}
+
+static int hclge_dbg_dump_mac_mc(struct seq_file *s, void *data)
+{
+	hclge_dbg_dump_mac_list(s, false);
+
+	return 0;
+}
+
+static const struct hclge_dbg_func hclge_dbg_cmd_func[] = {
+	{
+		.cmd = HNAE3_DBG_CMD_TM_NODES,
+		.dbg_read_func = hclge_dbg_dump_tm_nodes,
+	},
+	{
+		.cmd = HNAE3_DBG_CMD_TM_PRI,
+		.dbg_read_func = hclge_dbg_dump_tm_pri,
+	},
+	{
+		.cmd = HNAE3_DBG_CMD_TM_QSET,
+		.dbg_read_func = hclge_dbg_dump_tm_qset,
+	},
+	{
+		.cmd = HNAE3_DBG_CMD_TM_MAP,
+		.dbg_read_func = hclge_dbg_dump_tm_map,
+	},
+	{
+		.cmd = HNAE3_DBG_CMD_TM_PG,
+		.dbg_read_func = hclge_dbg_dump_tm_pg,
+	},
+	{
+		.cmd = HNAE3_DBG_CMD_TM_PORT,
+		.dbg_read_func = hclge_dbg_dump_tm_port,
+	},
+	{
+		.cmd = HNAE3_DBG_CMD_TC_SCH_INFO,
+		.dbg_read_func = hclge_dbg_dump_tc,
+	},
+	{
+		.cmd = HNAE3_DBG_CMD_QOS_PAUSE_CFG,
+		.dbg_read_func = hclge_dbg_dump_qos_pause_cfg,
+	},
+	{
+		.cmd = HNAE3_DBG_CMD_QOS_PRI_MAP,
+		.dbg_read_func = hclge_dbg_dump_qos_pri_map,
+	},
+	{
+		.cmd = HNAE3_DBG_CMD_QOS_DSCP_MAP,
+		.dbg_read_func = hclge_dbg_dump_qos_dscp_map,
+	},
+	{
+		.cmd = HNAE3_DBG_CMD_QOS_BUF_CFG,
+		.dbg_read_func = hclge_dbg_dump_qos_buf_cfg,
+	},
+	{
+		.cmd = HNAE3_DBG_CMD_MAC_UC,
+		.dbg_read_func = hclge_dbg_dump_mac_uc,
+	},
+	{
+		.cmd = HNAE3_DBG_CMD_MAC_MC,
+		.dbg_read_func = hclge_dbg_dump_mac_mc,
+	},
+	{
+		.cmd = HNAE3_DBG_CMD_MNG_TBL,
+		.dbg_read_func = hclge_dbg_dump_mng_table,
+	},
+	{
+		.cmd = HNAE3_DBG_CMD_LOOPBACK,
+		.dbg_read_func = hclge_dbg_dump_loopback,
+	},
+	{
+		.cmd = HNAE3_DBG_CMD_PTP_INFO,
+		.dbg_read_func = hclge_dbg_dump_ptp_info,
+	},
+	{
+		.cmd = HNAE3_DBG_CMD_INTERRUPT_INFO,
+		.dbg_read_func = hclge_dbg_dump_interrupt,
+	},
+	{
+		.cmd = HNAE3_DBG_CMD_RESET_INFO,
+		.dbg_read_func = hclge_dbg_seq_dump_rst_info,
+	},
+	{
+		.cmd = HNAE3_DBG_CMD_IMP_INFO,
+		.dbg_read_func = hclge_dbg_get_imp_stats_info,
+	},
+	{
+		.cmd = HNAE3_DBG_CMD_NCL_CONFIG,
+		.dbg_read_func = hclge_dbg_dump_ncl_config,
+	},
+	{
+		.cmd = HNAE3_DBG_CMD_REG_BIOS_COMMON,
+		.dbg_read_func = hclge_dbg_dump_bios_reg_cmd,
+	},
+	{
+		.cmd = HNAE3_DBG_CMD_REG_SSU,
+		.dbg_read_func = hclge_dbg_dump_ssu_reg_cmd,
+	},
+	{
+		.cmd = HNAE3_DBG_CMD_REG_IGU_EGU,
+		.dbg_read_func = hclge_dbg_dump_igu_egu_reg_cmd,
+	},
+	{
+		.cmd = HNAE3_DBG_CMD_REG_RPU,
+		.dbg_read_func = hclge_dbg_dump_rpu_reg_cmd,
+	},
+	{
+		.cmd = HNAE3_DBG_CMD_REG_NCSI,
+		.dbg_read_func = hclge_dbg_dump_ncsi_reg_cmd,
+	},
+	{
+		.cmd = HNAE3_DBG_CMD_REG_RTC,
+		.dbg_read_func = hclge_dbg_dump_rtc_reg_cmd,
+	},
+	{
+		.cmd = HNAE3_DBG_CMD_REG_PPP,
+		.dbg_read_func = hclge_dbg_dump_ppp_reg_cmd,
+	},
+	{
+		.cmd = HNAE3_DBG_CMD_REG_RCB,
+		.dbg_read_func = hclge_dbg_dump_rcb_reg_cmd,
+	},
+	{
+		.cmd = HNAE3_DBG_CMD_REG_TQP,
+		.dbg_read_func = hclge_dbg_dump_tqp_reg_cmd,
+	},
+	{
+		.cmd = HNAE3_DBG_CMD_REG_MAC,
+		.dbg_read_func = hclge_dbg_dump_mac,
+	},
+	{
+		.cmd = HNAE3_DBG_CMD_REG_DCB,
+		.dbg_read_func = hclge_dbg_dump_dcb,
+	},
+	{
+		.cmd = HNAE3_DBG_CMD_FD_TCAM,
+		.dbg_read_func = hclge_dbg_dump_fd_tcam,
+	},
+	{
+		.cmd = HNAE3_DBG_CMD_MAC_TNL_STATUS,
+		.dbg_read_func = hclge_dbg_dump_mac_tnl_status,
+	},
+	{
+		.cmd = HNAE3_DBG_CMD_SERV_INFO,
+		.dbg_read_func = hclge_dbg_dump_serv_info,
+	},
+	{
+		.cmd = HNAE3_DBG_CMD_VLAN_CONFIG,
+		.dbg_read_func = hclge_dbg_dump_vlan_config,
+	},
+	{
+		.cmd = HNAE3_DBG_CMD_FD_COUNTER,
+		.dbg_read_func = hclge_dbg_dump_fd_counter,
+	},
+	{
+		.cmd = HNAE3_DBG_CMD_UMV_INFO,
+		.dbg_read_func = hclge_dbg_dump_umv_info,
+	},
+};
+
+int hclge_dbg_get_read_func(struct hnae3_handle *handle, enum hnae3_dbg_cmd cmd,
+			    read_func *func)
+{
+	struct hclge_vport *vport = hclge_get_vport(handle);
+	const struct hclge_dbg_func *cmd_func;
+	struct hclge_dev *hdev = vport->back;
+	u32 i;
+
+	for (i = 0; i < ARRAY_SIZE(hclge_dbg_cmd_func); i++) {
+		if (cmd == hclge_dbg_cmd_func[i].cmd) {
+			cmd_func = &hclge_dbg_cmd_func[i];
+			*func = cmd_func->dbg_read_func;
+			return 0;
+		}
+	}
+
+	dev_err(&hdev->pdev->dev, "invalid command(%d)\n", cmd);
+	return -EINVAL;
+}
diff --git a/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_debugfs.h b/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_debugfs.h
new file mode 100644
index 000000000000..317f79efd54c
--- /dev/null
+++ b/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_debugfs.h
@@ -0,0 +1,138 @@
+/* SPDX-License-Identifier: GPL-2.0+ */
+/* Copyright (c) 2018-2019 Hisilicon Limited. */
+
+#ifndef __HCLGE_DEBUGFS_H
+#define __HCLGE_DEBUGFS_H
+
+#include <linux/etherdevice.h>
+#include "hclge_cmd.h"
+
+#define HCLGE_DBG_MNG_TBL_MAX	   64
+
+#define HCLGE_DBG_MNG_VLAN_MASK_B  BIT(0)
+#define HCLGE_DBG_MNG_MAC_MASK_B   BIT(1)
+#define HCLGE_DBG_MNG_ETHER_MASK_B BIT(2)
+#define HCLGE_DBG_MNG_E_TYPE_B	   BIT(11)
+#define HCLGE_DBG_MNG_DROP_B	   BIT(13)
+#define HCLGE_DBG_MNG_VLAN_TAG	   0x0FFF
+#define HCLGE_DBG_MNG_PF_ID	   0x0007
+#define HCLGE_DBG_MNG_VF_ID	   0x00FF
+
+/* Get DFX BD number offset */
+#define HCLGE_DBG_DFX_BIOS_OFFSET  1
+#define HCLGE_DBG_DFX_SSU_0_OFFSET 2
+#define HCLGE_DBG_DFX_SSU_1_OFFSET 3
+#define HCLGE_DBG_DFX_IGU_OFFSET   4
+#define HCLGE_DBG_DFX_RPU_0_OFFSET 5
+
+#define HCLGE_DBG_DFX_RPU_1_OFFSET 6
+#define HCLGE_DBG_DFX_NCSI_OFFSET  7
+#define HCLGE_DBG_DFX_RTC_OFFSET   8
+#define HCLGE_DBG_DFX_PPP_OFFSET   9
+#define HCLGE_DBG_DFX_RCB_OFFSET   10
+#define HCLGE_DBG_DFX_TQP_OFFSET   11
+
+#define HCLGE_DBG_DFX_SSU_2_OFFSET 12
+
+struct hclge_qos_pri_map_cmd {
+	u8 pri0_tc  : 4,
+	   pri1_tc  : 4;
+	u8 pri2_tc  : 4,
+	   pri3_tc  : 4;
+	u8 pri4_tc  : 4,
+	   pri5_tc  : 4;
+	u8 pri6_tc  : 4,
+	   pri7_tc  : 4;
+	u8 vlan_pri : 4,
+	   rev	    : 4;
+};
+
+struct hclge_dbg_bitmap_cmd {
+	union {
+		u8 bitmap;
+		struct {
+			u8 bit0 : 1,
+			   bit1 : 1,
+			   bit2 : 1,
+			   bit3 : 1,
+			   bit4 : 1,
+			   bit5 : 1,
+			   bit6 : 1,
+			   bit7 : 1;
+		};
+	};
+};
+
+struct hclge_dbg_reg_common_msg {
+	int msg_num;
+	int offset;
+	enum hclge_opcode_type cmd;
+};
+
+struct hclge_dbg_tcam_msg {
+	u8 stage;
+	u32 loc;
+};
+
+#define	HCLGE_DBG_MAX_DFX_MSG_LEN	60
+struct hclge_dbg_dfx_message {
+	int flag;
+	char message[HCLGE_DBG_MAX_DFX_MSG_LEN];
+};
+
+#define HCLGE_DBG_MAC_REG_TYPE_LEN	32
+struct hclge_dbg_reg_type_info {
+	enum hnae3_dbg_cmd cmd;
+	const struct hclge_dbg_dfx_message *dfx_msg;
+	struct hclge_dbg_reg_common_msg reg_msg;
+};
+
+struct hclge_dbg_func {
+	enum hnae3_dbg_cmd cmd;
+	int (*dbg_dump)(struct hclge_dev *hdev, char *buf, int len);
+	int (*dbg_dump_reg)(struct hclge_dev *hdev, enum hnae3_dbg_cmd cmd,
+			    char *buf, int len);
+	read_func dbg_read_func;
+};
+
+struct hclge_dbg_status_dfx_info {
+	u32  offset;
+	char message[HCLGE_DBG_MAX_DFX_MSG_LEN];
+};
+
+#define HCLGE_DBG_INFO_LEN			256
+#define HCLGE_DBG_VLAN_FLTR_INFO_LEN		256
+#define HCLGE_DBG_VLAN_OFFLOAD_INFO_LEN		512
+#define HCLGE_DBG_ID_LEN			16
+#define HCLGE_DBG_ITEM_NAME_LEN			32
+#define HCLGE_DBG_DATA_STR_LEN			32
+#define HCLGE_DBG_TM_INFO_LEN			256
+
+#define HCLGE_BILLION_NANO_SECONDS	1000000000
+
+struct hclge_dbg_item {
+	char name[HCLGE_DBG_ITEM_NAME_LEN];
+	u16 interval; /* blank numbers after the item */
+};
+
+struct hclge_dbg_vlan_cfg {
+	u16 pvid;
+	u8 accept_tag1;
+	u8 accept_tag2;
+	u8 accept_untag1;
+	u8 accept_untag2;
+	u8 insert_tag1;
+	u8 insert_tag2;
+	u8 shift_tag;
+	u8 strip_tag1;
+	u8 strip_tag2;
+	u8 drop_tag1;
+	u8 drop_tag2;
+	u8 pri_only1;
+	u8 pri_only2;
+};
+
+int hclge_dbg_cmd_send(struct hclge_dev *hdev, struct hclge_desc *desc_src,
+		       int index, int bd_num, enum hclge_opcode_type cmd);
+
+#endif
diff --git a/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_devlink.c b/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_devlink.c
new file mode 100644
index 000000000000..a1571c108678
--- /dev/null
+++ b/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_devlink.c
@@ -0,0 +1,166 @@
+// SPDX-License-Identifier: GPL-2.0+
+/* Copyright (c) 2021 Hisilicon Limited. */
+
+#include <net/devlink.h>
+
+#include "hclge_devlink.h"
+
+static int hclge_devlink_scc_info_get(struct devlink *devlink,
+				      struct devlink_info_req *req)
+{
+	struct hclge_devlink_priv *priv = devlink_priv(devlink);
+	char scc_version[HCLGE_DEVLINK_FW_SCC_LEN];
+	struct hclge_dev *hdev = priv->hdev;
+	u32 scc_version_tmp;
+	int ret;
+
+	ret = hclge_query_scc_version(hdev, &scc_version_tmp);
+	if (ret) {
+		dev_err(&hdev->pdev->dev,
+			"failed to get scc version, ret = %d\n", ret);
+		return ret;
+	}
+
+	snprintf(scc_version, sizeof(scc_version), "%lu.%lu.%lu.%lu",
+		 hnae3_get_field(scc_version_tmp, HNAE3_SCC_VERSION_BYTE3_MASK,
+				 HNAE3_FW_VERSION_BYTE3_SHIFT),
+		 hnae3_get_field(scc_version_tmp, HNAE3_SCC_VERSION_BYTE2_MASK,
+				 HNAE3_FW_VERSION_BYTE2_SHIFT),
+		 hnae3_get_field(scc_version_tmp, HNAE3_SCC_VERSION_BYTE1_MASK,
+				 HNAE3_FW_VERSION_BYTE1_SHIFT),
+		 hnae3_get_field(scc_version_tmp, HNAE3_SCC_VERSION_BYTE0_MASK,
+				 HNAE3_FW_VERSION_BYTE0_SHIFT));
+	return devlink_info_version_running_put(req, "fw.scc", scc_version);
+}
+
+static int hclge_devlink_info_get(struct devlink *devlink,
+				  struct devlink_info_req *req,
+				  struct netlink_ext_ack *extack)
+{
+#define	HCLGE_DEVLINK_FW_STRING_LEN	32
+	struct hclge_devlink_priv *priv = devlink_priv(devlink);
+	char version_str[HCLGE_DEVLINK_FW_STRING_LEN];
+	struct hclge_dev *hdev = priv->hdev;
+	int ret;
+
+	snprintf(version_str, sizeof(version_str), "%lu.%lu.%lu.%lu",
+		 hnae3_get_field(hdev->fw_version, HNAE3_FW_VERSION_BYTE3_MASK,
+				 HNAE3_FW_VERSION_BYTE3_SHIFT),
+		 hnae3_get_field(hdev->fw_version, HNAE3_FW_VERSION_BYTE2_MASK,
+				 HNAE3_FW_VERSION_BYTE2_SHIFT),
+		 hnae3_get_field(hdev->fw_version, HNAE3_FW_VERSION_BYTE1_MASK,
+				 HNAE3_FW_VERSION_BYTE1_SHIFT),
+		 hnae3_get_field(hdev->fw_version, HNAE3_FW_VERSION_BYTE0_MASK,
+				 HNAE3_FW_VERSION_BYTE0_SHIFT));
+
+	ret = devlink_info_version_running_put(req,
+					       DEVLINK_INFO_VERSION_GENERIC_FW,
+					       version_str);
+	if (ret) {
+		dev_err(&hdev->pdev->dev, "failed to set running version of fw\n");
+		return ret;
+	}
+
+	if (hdev->pdev->revision > HNAE3_DEVICE_VERSION_V2)
+		ret = hclge_devlink_scc_info_get(devlink, req);
+
+	return ret;
+}
+
+static int hclge_devlink_reload_down(struct devlink *devlink, bool netns_change,
+				     enum devlink_reload_action action,
+				     enum devlink_reload_limit limit,
+				     struct netlink_ext_ack *extack)
+{
+	struct hclge_devlink_priv *priv = devlink_priv(devlink);
+	struct hclge_dev *hdev = priv->hdev;
+	struct hnae3_handle *h = &hdev->vport->nic;
+	struct pci_dev *pdev = hdev->pdev;
+	int ret;
+
+	if (test_bit(HCLGE_STATE_RST_HANDLING, &hdev->state)) {
+		dev_err(&pdev->dev, "reset is handling\n");
+		return -EBUSY;
+	}
+
+	switch (action) {
+	case DEVLINK_RELOAD_ACTION_DRIVER_REINIT:
+		rtnl_lock();
+		ret = hdev->nic_client->ops->reset_notify(h, HNAE3_DOWN_CLIENT);
+		if (ret) {
+			rtnl_unlock();
+			return ret;
+		}
+
+		ret = hdev->nic_client->ops->reset_notify(h,
+							  HNAE3_UNINIT_CLIENT);
+		rtnl_unlock();
+		return ret;
+	default:
+		return -EOPNOTSUPP;
+	}
+}
+
+static int hclge_devlink_reload_up(struct devlink *devlink,
+				   enum devlink_reload_action action,
+				   enum devlink_reload_limit limit,
+				   u32 *actions_performed,
+				   struct netlink_ext_ack *extack)
+{
+	struct hclge_devlink_priv *priv = devlink_priv(devlink);
+	struct hclge_dev *hdev = priv->hdev;
+	struct hnae3_handle *h = &hdev->vport->nic;
+	int ret;
+
+	*actions_performed = BIT(action);
+	switch (action) {
+	case DEVLINK_RELOAD_ACTION_DRIVER_REINIT:
+		rtnl_lock();
+		ret = hdev->nic_client->ops->reset_notify(h, HNAE3_INIT_CLIENT);
+		if (ret) {
+			rtnl_unlock();
+			return ret;
+		}
+
+		ret = hdev->nic_client->ops->reset_notify(h, HNAE3_UP_CLIENT);
+		rtnl_unlock();
+		return ret;
+	default:
+		return -EOPNOTSUPP;
+	}
+}
+
+static const struct devlink_ops hclge_devlink_ops = {
+	.info_get = hclge_devlink_info_get,
+	.reload_actions = BIT(DEVLINK_RELOAD_ACTION_DRIVER_REINIT),
+	.reload_down = hclge_devlink_reload_down,
+	.reload_up = hclge_devlink_reload_up,
+};
+
+int hclge_devlink_init(struct hclge_dev *hdev)
+{
+	struct pci_dev *pdev = hdev->pdev;
+	struct hclge_devlink_priv *priv;
+	struct devlink *devlink;
+
+	devlink = devlink_alloc(&hclge_devlink_ops,
+				sizeof(struct hclge_devlink_priv), &pdev->dev);
+	if (!devlink)
+		return -ENOMEM;
+
+	priv = devlink_priv(devlink);
+	priv->hdev = hdev;
+	hdev->devlink = devlink;
+
+	devlink_register(devlink);
+	return 0;
+}
+
+void hclge_devlink_uninit(struct hclge_dev *hdev)
+{
+	struct devlink *devlink = hdev->devlink;
+
+	devlink_unregister(devlink);
+
+	devlink_free(devlink);
+}
diff --git a/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_devlink.h b/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_devlink.h
new file mode 100644
index 000000000000..148effa5ea89
--- /dev/null
+++ b/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_devlink.h
@@ -0,0 +1,17 @@
+/* SPDX-License-Identifier: GPL-2.0+ */
+/* Copyright (c) 2021 Hisilicon Limited. */
+
+#ifndef __HCLGE_DEVLINK_H
+#define __HCLGE_DEVLINK_H
+
+#include "hclge_main.h"
+
+#define	HCLGE_DEVLINK_FW_SCC_LEN	32
+
+struct hclge_devlink_priv {
+	struct hclge_dev *hdev;
+};
+
+int hclge_devlink_init(struct hclge_dev *hdev);
+void hclge_devlink_uninit(struct hclge_dev *hdev);
+#endif
diff --git a/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_err.c b/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_err.c
new file mode 100644
index 000000000000..cc7f46c0b35f
--- /dev/null
+++ b/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_err.c
@@ -0,0 +1,3473 @@
+// SPDX-License-Identifier: GPL-2.0+
+/* Copyright (c) 2016-2017 Hisilicon Limited. */
+
+#include <linux/sched/clock.h>
+
+#include "hclge_err.h"
+
+static const struct hclge_hw_error hclge_imp_tcm_ecc_int[] = {
+	{
+		.int_msk = BIT(1),
+		.msg = "imp_itcm0_ecc_mbit_err",
+		.reset_level = HNAE3_NONE_RESET
+	}, {
+		.int_msk = BIT(3),
+		.msg = "imp_itcm1_ecc_mbit_err",
+		.reset_level = HNAE3_NONE_RESET
+	}, {
+		.int_msk = BIT(5),
+		.msg = "imp_itcm2_ecc_mbit_err",
+		.reset_level = HNAE3_NONE_RESET
+	}, {
+		.int_msk = BIT(7),
+		.msg = "imp_itcm3_ecc_mbit_err",
+		.reset_level = HNAE3_NONE_RESET
+	}, {
+		.int_msk = BIT(9),
+		.msg = "imp_dtcm0_mem0_ecc_mbit_err",
+		.reset_level = HNAE3_NONE_RESET
+	}, {
+		.int_msk = BIT(11),
+		.msg = "imp_dtcm0_mem1_ecc_mbit_err",
+		.reset_level = HNAE3_NONE_RESET
+	}, {
+		.int_msk = BIT(13),
+		.msg = "imp_dtcm1_mem0_ecc_mbit_err",
+		.reset_level = HNAE3_NONE_RESET
+	}, {
+		.int_msk = BIT(15),
+		.msg = "imp_dtcm1_mem1_ecc_mbit_err",
+		.reset_level = HNAE3_NONE_RESET
+	}, {
+		.int_msk = BIT(17),
+		.msg = "imp_itcm4_ecc_mbit_err",
+		.reset_level = HNAE3_NONE_RESET
+	}, {
+		/* sentinel */
+	}
+};
+
+static const struct hclge_hw_error hclge_cmdq_nic_mem_ecc_int[] = {
+	{
+		.int_msk = BIT(1),
+		.msg = "cmdq_nic_rx_depth_ecc_mbit_err",
+		.reset_level = HNAE3_NONE_RESET
+	}, {
+		.int_msk = BIT(3),
+		.msg = "cmdq_nic_tx_depth_ecc_mbit_err",
+		.reset_level = HNAE3_NONE_RESET
+	}, {
+		.int_msk = BIT(5),
+		.msg = "cmdq_nic_rx_tail_ecc_mbit_err",
+		.reset_level = HNAE3_NONE_RESET
+	}, {
+		.int_msk = BIT(7),
+		.msg = "cmdq_nic_tx_tail_ecc_mbit_err",
+		.reset_level = HNAE3_NONE_RESET
+	}, {
+		.int_msk = BIT(9),
+		.msg = "cmdq_nic_rx_head_ecc_mbit_err",
+		.reset_level = HNAE3_NONE_RESET
+	}, {
+		.int_msk = BIT(11),
+		.msg = "cmdq_nic_tx_head_ecc_mbit_err",
+		.reset_level = HNAE3_NONE_RESET
+	}, {
+		.int_msk = BIT(13),
+		.msg = "cmdq_nic_rx_addr_ecc_mbit_err",
+		.reset_level = HNAE3_NONE_RESET
+	}, {
+		.int_msk = BIT(15),
+		.msg = "cmdq_nic_tx_addr_ecc_mbit_err",
+		.reset_level = HNAE3_NONE_RESET
+	}, {
+		.int_msk = BIT(17),
+		.msg = "cmdq_rocee_rx_depth_ecc_mbit_err",
+		.reset_level = HNAE3_NONE_RESET
+	}, {
+		.int_msk = BIT(19),
+		.msg = "cmdq_rocee_tx_depth_ecc_mbit_err",
+		.reset_level = HNAE3_NONE_RESET
+	}, {
+		.int_msk = BIT(21),
+		.msg = "cmdq_rocee_rx_tail_ecc_mbit_err",
+		.reset_level = HNAE3_NONE_RESET
+	}, {
+		.int_msk = BIT(23),
+		.msg = "cmdq_rocee_tx_tail_ecc_mbit_err",
+		.reset_level = HNAE3_NONE_RESET
+	}, {
+		.int_msk = BIT(25),
+		.msg = "cmdq_rocee_rx_head_ecc_mbit_err",
+		.reset_level = HNAE3_NONE_RESET
+	}, {
+		.int_msk = BIT(27),
+		.msg = "cmdq_rocee_tx_head_ecc_mbit_err",
+		.reset_level = HNAE3_NONE_RESET
+	}, {
+		.int_msk = BIT(29),
+		.msg = "cmdq_rocee_rx_addr_ecc_mbit_err",
+		.reset_level = HNAE3_NONE_RESET
+	}, {
+		.int_msk = BIT(31),
+		.msg = "cmdq_rocee_tx_addr_ecc_mbit_err",
+		.reset_level = HNAE3_NONE_RESET
+	}, {
+		/* sentinel */
+	}
+};
+
+static const struct hclge_hw_error hclge_tqp_int_ecc_int[] = {
+	{
+		.int_msk = BIT(6),
+		.msg = "tqp_int_cfg_even_ecc_mbit_err",
+		.reset_level = HNAE3_NONE_RESET
+	}, {
+		.int_msk = BIT(7),
+		.msg = "tqp_int_cfg_odd_ecc_mbit_err",
+		.reset_level = HNAE3_NONE_RESET
+	}, {
+		.int_msk = BIT(8),
+		.msg = "tqp_int_ctrl_even_ecc_mbit_err",
+		.reset_level = HNAE3_NONE_RESET
+	}, {
+		.int_msk = BIT(9),
+		.msg = "tqp_int_ctrl_odd_ecc_mbit_err",
+		.reset_level = HNAE3_NONE_RESET
+	}, {
+		.int_msk = BIT(10),
+		.msg = "tx_que_scan_int_ecc_mbit_err",
+		.reset_level = HNAE3_NONE_RESET
+	}, {
+		.int_msk = BIT(11),
+		.msg = "rx_que_scan_int_ecc_mbit_err",
+		.reset_level = HNAE3_NONE_RESET
+	}, {
+		/* sentinel */
+	}
+};
+
+static const struct hclge_hw_error hclge_msix_sram_ecc_int[] = {
+	{
+		.int_msk = BIT(1),
+		.msg = "msix_nic_ecc_mbit_err",
+		.reset_level = HNAE3_NONE_RESET
+	}, {
+		.int_msk = BIT(3),
+		.msg = "msix_rocee_ecc_mbit_err",
+		.reset_level = HNAE3_NONE_RESET
+	}, {
+		/* sentinel */
+	}
+};
+
+static const struct hclge_hw_error hclge_igu_int[] = {
+	{
+		.int_msk = BIT(0),
+		.msg = "igu_rx_buf0_ecc_mbit_err",
+		.reset_level = HNAE3_GLOBAL_RESET
+	}, {
+		.int_msk = BIT(2),
+		.msg = "igu_rx_buf1_ecc_mbit_err",
+		.reset_level = HNAE3_GLOBAL_RESET
+	}, {
+		/* sentinel */
+	}
+};
+
+static const struct hclge_hw_error hclge_igu_egu_tnl_int[] = {
+	{
+		.int_msk = BIT(0),
+		.msg = "rx_buf_overflow",
+		.reset_level = HNAE3_GLOBAL_RESET
+	}, {
+		.int_msk = BIT(1),
+		.msg = "rx_stp_fifo_overflow",
+		.reset_level = HNAE3_GLOBAL_RESET
+	}, {
+		.int_msk = BIT(2),
+		.msg = "rx_stp_fifo_underflow",
+		.reset_level = HNAE3_GLOBAL_RESET
+	}, {
+		.int_msk = BIT(3),
+		.msg = "tx_buf_overflow",
+		.reset_level = HNAE3_GLOBAL_RESET
+	}, {
+		.int_msk = BIT(4),
+		.msg = "tx_buf_underrun",
+		.reset_level = HNAE3_GLOBAL_RESET
+	}, {
+		.int_msk = BIT(5),
+		.msg = "rx_stp_buf_overflow",
+		.reset_level = HNAE3_GLOBAL_RESET
+	}, {
+		/* sentinel */
+	}
+};
+
+static const struct hclge_hw_error hclge_ncsi_err_int[] = {
+	{
+		.int_msk = BIT(1),
+		.msg = "ncsi_tx_ecc_mbit_err",
+		.reset_level = HNAE3_NONE_RESET
+	}, {
+		/* sentinel */
+	}
+};
+
+static const struct hclge_hw_error hclge_ppp_mpf_abnormal_int_st1[] = {
+	{
+		.int_msk = BIT(0),
+		.msg = "vf_vlan_ad_mem_ecc_mbit_err",
+		.reset_level = HNAE3_GLOBAL_RESET
+	}, {
+		.int_msk = BIT(1),
+		.msg = "umv_mcast_group_mem_ecc_mbit_err",
+		.reset_level = HNAE3_GLOBAL_RESET
+	}, {
+		.int_msk = BIT(2),
+		.msg = "umv_key_mem0_ecc_mbit_err",
+		.reset_level = HNAE3_GLOBAL_RESET
+	}, {
+		.int_msk = BIT(3),
+		.msg = "umv_key_mem1_ecc_mbit_err",
+		.reset_level = HNAE3_GLOBAL_RESET
+	}, {
+		.int_msk = BIT(4),
+		.msg = "umv_key_mem2_ecc_mbit_err",
+		.reset_level = HNAE3_GLOBAL_RESET
+	}, {
+		.int_msk = BIT(5),
+		.msg = "umv_key_mem3_ecc_mbit_err",
+		.reset_level = HNAE3_GLOBAL_RESET
+	}, {
+		.int_msk = BIT(6),
+		.msg = "umv_ad_mem_ecc_mbit_err",
+		.reset_level = HNAE3_GLOBAL_RESET
+	}, {
+		.int_msk = BIT(7),
+		.msg = "rss_tc_mode_mem_ecc_mbit_err",
+		.reset_level = HNAE3_GLOBAL_RESET
+	}, {
+		.int_msk = BIT(8),
+		.msg = "rss_idt_mem0_ecc_mbit_err",
+		.reset_level = HNAE3_GLOBAL_RESET
+	}, {
+		.int_msk = BIT(9),
+		.msg = "rss_idt_mem1_ecc_mbit_err",
+		.reset_level = HNAE3_GLOBAL_RESET
+	}, {
+		.int_msk = BIT(10),
+		.msg = "rss_idt_mem2_ecc_mbit_err",
+		.reset_level = HNAE3_GLOBAL_RESET
+	}, {
+		.int_msk = BIT(11),
+		.msg = "rss_idt_mem3_ecc_mbit_err",
+		.reset_level = HNAE3_GLOBAL_RESET
+	}, {
+		.int_msk = BIT(12),
+		.msg = "rss_idt_mem4_ecc_mbit_err",
+		.reset_level = HNAE3_GLOBAL_RESET
+	}, {
+		.int_msk = BIT(13),
+		.msg = "rss_idt_mem5_ecc_mbit_err",
+		.reset_level = HNAE3_GLOBAL_RESET
+	}, {
+		.int_msk = BIT(14),
+		.msg = "rss_idt_mem6_ecc_mbit_err",
+		.reset_level = HNAE3_GLOBAL_RESET
+	}, {
+		.int_msk = BIT(15),
+		.msg = "rss_idt_mem7_ecc_mbit_err",
+		.reset_level = HNAE3_GLOBAL_RESET
+	}, {
+		.int_msk = BIT(16),
+		.msg = "rss_idt_mem8_ecc_mbit_err",
+		.reset_level = HNAE3_GLOBAL_RESET
+	}, {
+		.int_msk = BIT(17),
+		.msg = "rss_idt_mem9_ecc_mbit_err",
+		.reset_level = HNAE3_GLOBAL_RESET
+	}, {
+		.int_msk = BIT(18),
+		.msg = "rss_idt_mem10_ecc_mbit_err",
+		.reset_level = HNAE3_GLOBAL_RESET
+	}, {
+		.int_msk = BIT(19),
+		.msg = "rss_idt_mem11_ecc_mbit_err",
+		.reset_level = HNAE3_GLOBAL_RESET
+	}, {
+		.int_msk = BIT(20),
+		.msg = "rss_idt_mem12_ecc_mbit_err",
+		.reset_level = HNAE3_GLOBAL_RESET
+	}, {
+		.int_msk = BIT(21),
+		.msg = "rss_idt_mem13_ecc_mbit_err",
+		.reset_level = HNAE3_GLOBAL_RESET
+	}, {
+		.int_msk = BIT(22),
+		.msg = "rss_idt_mem14_ecc_mbit_err",
+		.reset_level = HNAE3_GLOBAL_RESET
+	}, {
+		.int_msk = BIT(23),
+		.msg = "rss_idt_mem15_ecc_mbit_err",
+		.reset_level = HNAE3_GLOBAL_RESET
+	}, {
+		.int_msk = BIT(24),
+		.msg = "port_vlan_mem_ecc_mbit_err",
+		.reset_level = HNAE3_GLOBAL_RESET
+	}, {
+		.int_msk = BIT(25),
+		.msg = "mcast_linear_table_mem_ecc_mbit_err",
+		.reset_level = HNAE3_GLOBAL_RESET
+	}, {
+		.int_msk = BIT(26),
+		.msg = "mcast_result_mem_ecc_mbit_err",
+		.reset_level = HNAE3_GLOBAL_RESET
+	}, {
+		.int_msk = BIT(27),
+		.msg = "flow_director_ad_mem0_ecc_mbit_err",
+		.reset_level = HNAE3_GLOBAL_RESET
+	}, {
+		.int_msk = BIT(28),
+		.msg = "flow_director_ad_mem1_ecc_mbit_err",
+		.reset_level = HNAE3_GLOBAL_RESET
+	}, {
+		.int_msk = BIT(29),
+		.msg = "rx_vlan_tag_memory_ecc_mbit_err",
+		.reset_level = HNAE3_GLOBAL_RESET
+	}, {
+		.int_msk = BIT(30),
+		.msg = "Tx_UP_mapping_config_mem_ecc_mbit_err",
+		.reset_level = HNAE3_GLOBAL_RESET
+	}, {
+		/* sentinel */
+	}
+};
+
+static const struct hclge_hw_error hclge_ppp_pf_abnormal_int[] = {
+	{
+		.int_msk = BIT(0),
+		.msg = "tx_vlan_tag_err",
+		.reset_level = HNAE3_NONE_RESET
+	}, {
+		.int_msk = BIT(1),
+		.msg = "rss_list_tc_unassigned_queue_err",
+		.reset_level = HNAE3_NONE_RESET
+	}, {
+		/* sentinel */
+	}
+};
+
+static const struct hclge_hw_error hclge_ppp_mpf_abnormal_int_st3[] = {
+	{
+		.int_msk = BIT(0),
+		.msg = "hfs_fifo_mem_ecc_mbit_err",
+		.reset_level = HNAE3_GLOBAL_RESET
+	}, {
+		.int_msk = BIT(1),
+		.msg = "rslt_descr_fifo_mem_ecc_mbit_err",
+		.reset_level = HNAE3_GLOBAL_RESET
+	}, {
+		.int_msk = BIT(2),
+		.msg = "tx_vlan_tag_mem_ecc_mbit_err",
+		.reset_level = HNAE3_GLOBAL_RESET
+	}, {
+		.int_msk = BIT(3),
+		.msg = "FD_CN0_memory_ecc_mbit_err",
+		.reset_level = HNAE3_GLOBAL_RESET
+	}, {
+		.int_msk = BIT(4),
+		.msg = "FD_CN1_memory_ecc_mbit_err",
+		.reset_level = HNAE3_GLOBAL_RESET
+	}, {
+		.int_msk = BIT(5),
+		.msg = "GRO_AD_memory_ecc_mbit_err",
+		.reset_level = HNAE3_GLOBAL_RESET
+	}, {
+		/* sentinel */
+	}
+};
+
+static const struct hclge_hw_error hclge_tm_sch_rint[] = {
+	{
+		.int_msk = BIT(1),
+		.msg = "tm_sch_ecc_mbit_err",
+		.reset_level = HNAE3_GLOBAL_RESET
+	}, {
+		.int_msk = BIT(2),
+		.msg = "tm_sch_port_shap_sub_fifo_wr_err",
+		.reset_level = HNAE3_GLOBAL_RESET
+	}, {
+		.int_msk = BIT(3),
+		.msg = "tm_sch_port_shap_sub_fifo_rd_err",
+		.reset_level = HNAE3_GLOBAL_RESET
+	}, {
+		.int_msk = BIT(4),
+		.msg = "tm_sch_pg_pshap_sub_fifo_wr_err",
+		.reset_level = HNAE3_GLOBAL_RESET
+	}, {
+		.int_msk = BIT(5),
+		.msg = "tm_sch_pg_pshap_sub_fifo_rd_err",
+		.reset_level = HNAE3_GLOBAL_RESET
+	}, {
+		.int_msk = BIT(6),
+		.msg = "tm_sch_pg_cshap_sub_fifo_wr_err",
+		.reset_level = HNAE3_GLOBAL_RESET
+	}, {
+		.int_msk = BIT(7),
+		.msg = "tm_sch_pg_cshap_sub_fifo_rd_err",
+		.reset_level = HNAE3_GLOBAL_RESET
+	}, {
+		.int_msk = BIT(8),
+		.msg = "tm_sch_pri_pshap_sub_fifo_wr_err",
+		.reset_level = HNAE3_GLOBAL_RESET
+	}, {
+		.int_msk = BIT(9),
+		.msg = "tm_sch_pri_pshap_sub_fifo_rd_err",
+		.reset_level = HNAE3_GLOBAL_RESET
+	}, {
+		.int_msk = BIT(10),
+		.msg = "tm_sch_pri_cshap_sub_fifo_wr_err",
+		.reset_level = HNAE3_GLOBAL_RESET
+	}, {
+		.int_msk = BIT(11),
+		.msg = "tm_sch_pri_cshap_sub_fifo_rd_err",
+		.reset_level = HNAE3_GLOBAL_RESET
+	}, {
+		.int_msk = BIT(12),
+		.msg = "tm_sch_port_shap_offset_fifo_wr_err",
+		.reset_level = HNAE3_GLOBAL_RESET
+	}, {
+		.int_msk = BIT(13),
+		.msg = "tm_sch_port_shap_offset_fifo_rd_err",
+		.reset_level = HNAE3_GLOBAL_RESET
+	}, {
+		.int_msk = BIT(14),
+		.msg = "tm_sch_pg_pshap_offset_fifo_wr_err",
+		.reset_level = HNAE3_GLOBAL_RESET
+	}, {
+		.int_msk = BIT(15),
+		.msg = "tm_sch_pg_pshap_offset_fifo_rd_err",
+		.reset_level = HNAE3_GLOBAL_RESET
+	}, {
+		.int_msk = BIT(16),
+		.msg = "tm_sch_pg_cshap_offset_fifo_wr_err",
+		.reset_level = HNAE3_GLOBAL_RESET
+	}, {
+		.int_msk = BIT(17),
+		.msg = "tm_sch_pg_cshap_offset_fifo_rd_err",
+		.reset_level = HNAE3_GLOBAL_RESET
+	}, {
+		.int_msk = BIT(18),
+		.msg = "tm_sch_pri_pshap_offset_fifo_wr_err",
+		.reset_level = HNAE3_GLOBAL_RESET
+	}, {
+		.int_msk = BIT(19),
+		.msg = "tm_sch_pri_pshap_offset_fifo_rd_err",
+		.reset_level = HNAE3_GLOBAL_RESET
+	}, {
+		.int_msk = BIT(20),
+		.msg = "tm_sch_pri_cshap_offset_fifo_wr_err",
+		.reset_level = HNAE3_GLOBAL_RESET
+	}, {
+		.int_msk = BIT(21),
+		.msg = "tm_sch_pri_cshap_offset_fifo_rd_err",
+		.reset_level = HNAE3_GLOBAL_RESET
+	}, {
+		.int_msk = BIT(22),
+		.msg = "tm_sch_rq_fifo_wr_err",
+		.reset_level = HNAE3_GLOBAL_RESET
+	}, {
+		.int_msk = BIT(23),
+		.msg = "tm_sch_rq_fifo_rd_err",
+		.reset_level = HNAE3_GLOBAL_RESET
+	}, {
+		.int_msk = BIT(24),
+		.msg = "tm_sch_nq_fifo_wr_err",
+		.reset_level = HNAE3_GLOBAL_RESET
+	}, {
+		.int_msk = BIT(25),
+		.msg = "tm_sch_nq_fifo_rd_err",
+		.reset_level = HNAE3_GLOBAL_RESET
+	}, {
+		.int_msk = BIT(26),
+		.msg = "tm_sch_roce_up_fifo_wr_err",
+		.reset_level = HNAE3_GLOBAL_RESET
+	}, {
+		.int_msk = BIT(27),
+		.msg = "tm_sch_roce_up_fifo_rd_err",
+		.reset_level = HNAE3_GLOBAL_RESET
+	}, {
+		.int_msk = BIT(28),
+		.msg = "tm_sch_rcb_byte_fifo_wr_err",
+		.reset_level = HNAE3_GLOBAL_RESET
+	}, {
+		.int_msk = BIT(29),
+		.msg = "tm_sch_rcb_byte_fifo_rd_err",
+		.reset_level = HNAE3_GLOBAL_RESET
+	}, {
+		.int_msk = BIT(30),
+		.msg = "tm_sch_ssu_byte_fifo_wr_err",
+		.reset_level = HNAE3_GLOBAL_RESET
+	}, {
+		.int_msk = BIT(31),
+		.msg = "tm_sch_ssu_byte_fifo_rd_err",
+		.reset_level = HNAE3_GLOBAL_RESET
+	}, {
+		/* sentinel */
+	}
+};
+
+static const struct hclge_hw_error hclge_qcn_fifo_rint[] = {
+	{
+		.int_msk = BIT(0),
+		.msg = "qcn_shap_gp0_sch_fifo_rd_err",
+		.reset_level = HNAE3_GLOBAL_RESET
+	}, {
+		.int_msk = BIT(1),
+		.msg = "qcn_shap_gp0_sch_fifo_wr_err",
+		.reset_level = HNAE3_GLOBAL_RESET
+	}, {
+		.int_msk = BIT(2),
+		.msg = "qcn_shap_gp1_sch_fifo_rd_err",
+		.reset_level = HNAE3_GLOBAL_RESET
+	}, {
+		.int_msk = BIT(3),
+		.msg = "qcn_shap_gp1_sch_fifo_wr_err",
+		.reset_level = HNAE3_GLOBAL_RESET
+	}, {
+		.int_msk = BIT(4),
+		.msg = "qcn_shap_gp2_sch_fifo_rd_err",
+		.reset_level = HNAE3_GLOBAL_RESET
+	}, {
+		.int_msk = BIT(5),
+		.msg = "qcn_shap_gp2_sch_fifo_wr_err",
+		.reset_level = HNAE3_GLOBAL_RESET
+	}, {
+		.int_msk = BIT(6),
+		.msg = "qcn_shap_gp3_sch_fifo_rd_err",
+		.reset_level = HNAE3_GLOBAL_RESET
+	}, {
+		.int_msk = BIT(7),
+		.msg = "qcn_shap_gp3_sch_fifo_wr_err",
+		.reset_level = HNAE3_GLOBAL_RESET
+	}, {
+		.int_msk = BIT(8),
+		.msg = "qcn_shap_gp0_offset_fifo_rd_err",
+		.reset_level = HNAE3_GLOBAL_RESET
+	}, {
+		.int_msk = BIT(9),
+		.msg = "qcn_shap_gp0_offset_fifo_wr_err",
+		.reset_level = HNAE3_GLOBAL_RESET
+	}, {
+		.int_msk = BIT(10),
+		.msg = "qcn_shap_gp1_offset_fifo_rd_err",
+		.reset_level = HNAE3_GLOBAL_RESET
+	}, {
+		.int_msk = BIT(11),
+		.msg = "qcn_shap_gp1_offset_fifo_wr_err",
+		.reset_level = HNAE3_GLOBAL_RESET
+	}, {
+		.int_msk = BIT(12),
+		.msg = "qcn_shap_gp2_offset_fifo_rd_err",
+		.reset_level = HNAE3_GLOBAL_RESET
+	}, {
+		.int_msk = BIT(13),
+		.msg = "qcn_shap_gp2_offset_fifo_wr_err",
+		.reset_level = HNAE3_GLOBAL_RESET
+	}, {
+		.int_msk = BIT(14),
+		.msg = "qcn_shap_gp3_offset_fifo_rd_err",
+		.reset_level = HNAE3_GLOBAL_RESET
+	}, {
+		.int_msk = BIT(15),
+		.msg = "qcn_shap_gp3_offset_fifo_wr_err",
+		.reset_level = HNAE3_GLOBAL_RESET
+	}, {
+		.int_msk = BIT(16),
+		.msg = "qcn_byte_info_fifo_rd_err",
+		.reset_level = HNAE3_GLOBAL_RESET
+	}, {
+		.int_msk = BIT(17),
+		.msg = "qcn_byte_info_fifo_wr_err",
+		.reset_level = HNAE3_GLOBAL_RESET
+	}, {
+		/* sentinel */
+	}
+};
+
+static const struct hclge_hw_error hclge_qcn_ecc_rint[] = {
+	{
+		.int_msk = BIT(1),
+		.msg = "qcn_byte_mem_ecc_mbit_err",
+		.reset_level = HNAE3_GLOBAL_RESET
+	}, {
+		.int_msk = BIT(3),
+		.msg = "qcn_time_mem_ecc_mbit_err",
+		.reset_level = HNAE3_GLOBAL_RESET
+	}, {
+		.int_msk = BIT(5),
+		.msg = "qcn_fb_mem_ecc_mbit_err",
+		.reset_level = HNAE3_GLOBAL_RESET
+	}, {
+		.int_msk = BIT(7),
+		.msg = "qcn_link_mem_ecc_mbit_err",
+		.reset_level = HNAE3_GLOBAL_RESET
+	}, {
+		.int_msk = BIT(9),
+		.msg = "qcn_rate_mem_ecc_mbit_err",
+		.reset_level = HNAE3_GLOBAL_RESET
+	}, {
+		.int_msk = BIT(11),
+		.msg = "qcn_tmplt_mem_ecc_mbit_err",
+		.reset_level = HNAE3_GLOBAL_RESET
+	}, {
+		.int_msk = BIT(13),
+		.msg = "qcn_shap_cfg_mem_ecc_mbit_err",
+		.reset_level = HNAE3_GLOBAL_RESET
+	}, {
+		.int_msk = BIT(15),
+		.msg = "qcn_gp0_barrel_mem_ecc_mbit_err",
+		.reset_level = HNAE3_GLOBAL_RESET
+	}, {
+		.int_msk = BIT(17),
+		.msg = "qcn_gp1_barrel_mem_ecc_mbit_err",
+		.reset_level = HNAE3_GLOBAL_RESET
+	}, {
+		.int_msk = BIT(19),
+		.msg = "qcn_gp2_barrel_mem_ecc_mbit_err",
+		.reset_level = HNAE3_GLOBAL_RESET
+	}, {
+		.int_msk = BIT(21),
+		.msg = "qcn_gp3_barral_mem_ecc_mbit_err",
+		.reset_level = HNAE3_GLOBAL_RESET
+	}, {
+		/* sentinel */
+	}
+};
+
+static const struct hclge_hw_error hclge_mac_afifo_tnl_int[] = {
+	{
+		.int_msk = BIT(0),
+		.msg = "egu_cge_afifo_ecc_1bit_err",
+		.reset_level = HNAE3_NONE_RESET
+	}, {
+		.int_msk = BIT(1),
+		.msg = "egu_cge_afifo_ecc_mbit_err",
+		.reset_level = HNAE3_GLOBAL_RESET
+	}, {
+		.int_msk = BIT(2),
+		.msg = "egu_lge_afifo_ecc_1bit_err",
+		.reset_level = HNAE3_NONE_RESET
+	}, {
+		.int_msk = BIT(3),
+		.msg = "egu_lge_afifo_ecc_mbit_err",
+		.reset_level = HNAE3_GLOBAL_RESET
+	}, {
+		.int_msk = BIT(4),
+		.msg = "cge_igu_afifo_ecc_1bit_err",
+		.reset_level = HNAE3_NONE_RESET
+	}, {
+		.int_msk = BIT(5),
+		.msg = "cge_igu_afifo_ecc_mbit_err",
+		.reset_level = HNAE3_GLOBAL_RESET
+	}, {
+		.int_msk = BIT(6),
+		.msg = "lge_igu_afifo_ecc_1bit_err",
+		.reset_level = HNAE3_NONE_RESET
+	}, {
+		.int_msk = BIT(7),
+		.msg = "lge_igu_afifo_ecc_mbit_err",
+		.reset_level = HNAE3_GLOBAL_RESET
+	}, {
+		.int_msk = BIT(8),
+		.msg = "cge_igu_afifo_overflow_err",
+		.reset_level = HNAE3_GLOBAL_RESET
+	}, {
+		.int_msk = BIT(9),
+		.msg = "lge_igu_afifo_overflow_err",
+		.reset_level = HNAE3_GLOBAL_RESET
+	}, {
+		.int_msk = BIT(10),
+		.msg = "egu_cge_afifo_underrun_err",
+		.reset_level = HNAE3_GLOBAL_RESET
+	}, {
+		.int_msk = BIT(11),
+		.msg = "egu_lge_afifo_underrun_err",
+		.reset_level = HNAE3_GLOBAL_RESET
+	}, {
+		.int_msk = BIT(12),
+		.msg = "egu_ge_afifo_underrun_err",
+		.reset_level = HNAE3_GLOBAL_RESET
+	}, {
+		.int_msk = BIT(13),
+		.msg = "ge_igu_afifo_overflow_err",
+		.reset_level = HNAE3_GLOBAL_RESET
+	}, {
+		/* sentinel */
+	}
+};
+
+static const struct hclge_hw_error hclge_ppu_mpf_abnormal_int_st2[] = {
+	{
+		.int_msk = BIT(13),
+		.msg = "rpu_rx_pkt_bit32_ecc_mbit_err",
+		.reset_level = HNAE3_GLOBAL_RESET
+	}, {
+		.int_msk = BIT(14),
+		.msg = "rpu_rx_pkt_bit33_ecc_mbit_err",
+		.reset_level = HNAE3_GLOBAL_RESET
+	}, {
+		.int_msk = BIT(15),
+		.msg = "rpu_rx_pkt_bit34_ecc_mbit_err",
+		.reset_level = HNAE3_GLOBAL_RESET
+	}, {
+		.int_msk = BIT(16),
+		.msg = "rpu_rx_pkt_bit35_ecc_mbit_err",
+		.reset_level = HNAE3_GLOBAL_RESET
+	}, {
+		.int_msk = BIT(17),
+		.msg = "rcb_tx_ring_ecc_mbit_err",
+		.reset_level = HNAE3_GLOBAL_RESET
+	}, {
+		.int_msk = BIT(18),
+		.msg = "rcb_rx_ring_ecc_mbit_err",
+		.reset_level = HNAE3_GLOBAL_RESET
+	}, {
+		.int_msk = BIT(19),
+		.msg = "rcb_tx_fbd_ecc_mbit_err",
+		.reset_level = HNAE3_GLOBAL_RESET
+	}, {
+		.int_msk = BIT(20),
+		.msg = "rcb_rx_ebd_ecc_mbit_err",
+		.reset_level = HNAE3_GLOBAL_RESET
+	}, {
+		.int_msk = BIT(21),
+		.msg = "rcb_tso_info_ecc_mbit_err",
+		.reset_level = HNAE3_GLOBAL_RESET
+	}, {
+		.int_msk = BIT(22),
+		.msg = "rcb_tx_int_info_ecc_mbit_err",
+		.reset_level = HNAE3_GLOBAL_RESET
+	}, {
+		.int_msk = BIT(23),
+		.msg = "rcb_rx_int_info_ecc_mbit_err",
+		.reset_level = HNAE3_GLOBAL_RESET
+	}, {
+		.int_msk = BIT(24),
+		.msg = "tpu_tx_pkt_0_ecc_mbit_err",
+		.reset_level = HNAE3_GLOBAL_RESET
+	}, {
+		.int_msk = BIT(25),
+		.msg = "tpu_tx_pkt_1_ecc_mbit_err",
+		.reset_level = HNAE3_GLOBAL_RESET
+	}, {
+		.int_msk = BIT(26),
+		.msg = "rd_bus_err",
+		.reset_level = HNAE3_GLOBAL_RESET
+	}, {
+		.int_msk = BIT(27),
+		.msg = "wr_bus_err",
+		.reset_level = HNAE3_GLOBAL_RESET
+	}, {
+		.int_msk = BIT(28),
+		.msg = "reg_search_miss",
+		.reset_level = HNAE3_GLOBAL_RESET
+	}, {
+		.int_msk = BIT(29),
+		.msg = "rx_q_search_miss",
+		.reset_level = HNAE3_NONE_RESET
+	}, {
+		.int_msk = BIT(30),
+		.msg = "ooo_ecc_err_detect",
+		.reset_level = HNAE3_NONE_RESET
+	}, {
+		.int_msk = BIT(31),
+		.msg = "ooo_ecc_err_multpl",
+		.reset_level = HNAE3_GLOBAL_RESET
+	}, {
+		/* sentinel */
+	}
+};
+
+static const struct hclge_hw_error hclge_ppu_mpf_abnormal_int_st3[] = {
+	{
+		.int_msk = BIT(4),
+		.msg = "gro_bd_ecc_mbit_err",
+		.reset_level = HNAE3_GLOBAL_RESET
+	}, {
+		.int_msk = BIT(5),
+		.msg = "gro_context_ecc_mbit_err",
+		.reset_level = HNAE3_GLOBAL_RESET
+	}, {
+		.int_msk = BIT(6),
+		.msg = "rx_stash_cfg_ecc_mbit_err",
+		.reset_level = HNAE3_GLOBAL_RESET
+	}, {
+		.int_msk = BIT(7),
+		.msg = "axi_rd_fbd_ecc_mbit_err",
+		.reset_level = HNAE3_GLOBAL_RESET
+	}, {
+		/* sentinel */
+	}
+};
+
+static const struct hclge_hw_error hclge_ppu_pf_abnormal_int[] = {
+	{
+		.int_msk = BIT(0),
+		.msg = "over_8bd_no_fe",
+		.reset_level = HNAE3_FUNC_RESET
+	}, {
+		.int_msk = BIT(1),
+		.msg = "tso_mss_cmp_min_err",
+		.reset_level = HNAE3_NONE_RESET
+	}, {
+		.int_msk = BIT(2),
+		.msg = "tso_mss_cmp_max_err",
+		.reset_level = HNAE3_NONE_RESET
+	}, {
+		.int_msk = BIT(3),
+		.msg = "tx_rd_fbd_poison",
+		.reset_level = HNAE3_FUNC_RESET
+	}, {
+		.int_msk = BIT(4),
+		.msg = "rx_rd_ebd_poison",
+		.reset_level = HNAE3_FUNC_RESET
+	}, {
+		.int_msk = BIT(5),
+		.msg = "buf_wait_timeout",
+		.reset_level = HNAE3_NONE_RESET
+	}, {
+		/* sentinel */
+	}
+};
+
+static const struct hclge_hw_error hclge_ssu_com_err_int[] = {
+	{
+		.int_msk = BIT(0),
+		.msg = "buf_sum_err",
+		.reset_level = HNAE3_NONE_RESET
+	}, {
+		.int_msk = BIT(1),
+		.msg = "ppp_mb_num_err",
+		.reset_level = HNAE3_NONE_RESET
+	}, {
+		.int_msk = BIT(2),
+		.msg = "ppp_mbid_err",
+		.reset_level = HNAE3_GLOBAL_RESET
+	}, {
+		.int_msk = BIT(3),
+		.msg = "ppp_rlt_mac_err",
+		.reset_level = HNAE3_GLOBAL_RESET
+	}, {
+		.int_msk = BIT(4),
+		.msg = "ppp_rlt_host_err",
+		.reset_level = HNAE3_GLOBAL_RESET
+	}, {
+		.int_msk = BIT(5),
+		.msg = "cks_edit_position_err",
+		.reset_level = HNAE3_GLOBAL_RESET
+	}, {
+		.int_msk = BIT(6),
+		.msg = "cks_edit_condition_err",
+		.reset_level = HNAE3_GLOBAL_RESET
+	}, {
+		.int_msk = BIT(7),
+		.msg = "vlan_edit_condition_err",
+		.reset_level = HNAE3_GLOBAL_RESET
+	}, {
+		.int_msk = BIT(8),
+		.msg = "vlan_num_ot_err",
+		.reset_level = HNAE3_GLOBAL_RESET
+	}, {
+		.int_msk = BIT(9),
+		.msg = "vlan_num_in_err",
+		.reset_level = HNAE3_GLOBAL_RESET
+	}, {
+		/* sentinel */
+	}
+};
+
+#define HCLGE_SSU_MEM_ECC_ERR(x) \
+{ \
+	.int_msk = BIT(x), \
+	.msg = "ssu_mem" #x "_ecc_mbit_err", \
+	.reset_level = HNAE3_GLOBAL_RESET \
+}
+
+static const struct hclge_hw_error hclge_ssu_mem_ecc_err_int[] = {
+	HCLGE_SSU_MEM_ECC_ERR(0),
+	HCLGE_SSU_MEM_ECC_ERR(1),
+	HCLGE_SSU_MEM_ECC_ERR(2),
+	HCLGE_SSU_MEM_ECC_ERR(3),
+	HCLGE_SSU_MEM_ECC_ERR(4),
+	HCLGE_SSU_MEM_ECC_ERR(5),
+	HCLGE_SSU_MEM_ECC_ERR(6),
+	HCLGE_SSU_MEM_ECC_ERR(7),
+	HCLGE_SSU_MEM_ECC_ERR(8),
+	HCLGE_SSU_MEM_ECC_ERR(9),
+	HCLGE_SSU_MEM_ECC_ERR(10),
+	HCLGE_SSU_MEM_ECC_ERR(11),
+	HCLGE_SSU_MEM_ECC_ERR(12),
+	HCLGE_SSU_MEM_ECC_ERR(13),
+	HCLGE_SSU_MEM_ECC_ERR(14),
+	HCLGE_SSU_MEM_ECC_ERR(15),
+	HCLGE_SSU_MEM_ECC_ERR(16),
+	HCLGE_SSU_MEM_ECC_ERR(17),
+	HCLGE_SSU_MEM_ECC_ERR(18),
+	HCLGE_SSU_MEM_ECC_ERR(19),
+	HCLGE_SSU_MEM_ECC_ERR(20),
+	HCLGE_SSU_MEM_ECC_ERR(21),
+	HCLGE_SSU_MEM_ECC_ERR(22),
+	HCLGE_SSU_MEM_ECC_ERR(23),
+	HCLGE_SSU_MEM_ECC_ERR(24),
+	HCLGE_SSU_MEM_ECC_ERR(25),
+	HCLGE_SSU_MEM_ECC_ERR(26),
+	HCLGE_SSU_MEM_ECC_ERR(27),
+	HCLGE_SSU_MEM_ECC_ERR(28),
+	HCLGE_SSU_MEM_ECC_ERR(29),
+	HCLGE_SSU_MEM_ECC_ERR(30),
+	HCLGE_SSU_MEM_ECC_ERR(31),
+	{ /* sentinel */ }
+};
+
+static const struct hclge_hw_error hclge_ssu_port_based_err_int[] = {
+	{
+		.int_msk = BIT(0),
+		.msg = "roc_pkt_without_key_port",
+		.reset_level = HNAE3_FUNC_RESET
+	}, {
+		.int_msk = BIT(1),
+		.msg = "tpu_pkt_without_key_port",
+		.reset_level = HNAE3_GLOBAL_RESET
+	}, {
+		.int_msk = BIT(2),
+		.msg = "igu_pkt_without_key_port",
+		.reset_level = HNAE3_GLOBAL_RESET
+	}, {
+		.int_msk = BIT(3),
+		.msg = "roc_eof_mis_match_port",
+		.reset_level = HNAE3_GLOBAL_RESET
+	}, {
+		.int_msk = BIT(4),
+		.msg = "tpu_eof_mis_match_port",
+		.reset_level = HNAE3_GLOBAL_RESET
+	}, {
+		.int_msk = BIT(5),
+		.msg = "igu_eof_mis_match_port",
+		.reset_level = HNAE3_GLOBAL_RESET
+	}, {
+		.int_msk = BIT(6),
+		.msg = "roc_sof_mis_match_port",
+		.reset_level = HNAE3_GLOBAL_RESET
+	}, {
+		.int_msk = BIT(7),
+		.msg = "tpu_sof_mis_match_port",
+		.reset_level = HNAE3_GLOBAL_RESET
+	}, {
+		.int_msk = BIT(8),
+		.msg = "igu_sof_mis_match_port",
+		.reset_level = HNAE3_GLOBAL_RESET
+	}, {
+		.int_msk = BIT(11),
+		.msg = "ets_rd_int_rx_port",
+		.reset_level = HNAE3_GLOBAL_RESET
+	}, {
+		.int_msk = BIT(12),
+		.msg = "ets_wr_int_rx_port",
+		.reset_level = HNAE3_GLOBAL_RESET
+	}, {
+		.int_msk = BIT(13),
+		.msg = "ets_rd_int_tx_port",
+		.reset_level = HNAE3_GLOBAL_RESET
+	}, {
+		.int_msk = BIT(14),
+		.msg = "ets_wr_int_tx_port",
+		.reset_level = HNAE3_GLOBAL_RESET
+	}, {
+		/* sentinel */
+	}
+};
+
+static const struct hclge_hw_error hclge_ssu_fifo_overflow_int[] = {
+	{
+		.int_msk = BIT(0),
+		.msg = "ig_mac_inf_int",
+		.reset_level = HNAE3_GLOBAL_RESET
+	}, {
+		.int_msk = BIT(1),
+		.msg = "ig_host_inf_int",
+		.reset_level = HNAE3_GLOBAL_RESET
+	}, {
+		.int_msk = BIT(2),
+		.msg = "ig_roc_buf_int",
+		.reset_level = HNAE3_GLOBAL_RESET
+	}, {
+		.int_msk = BIT(3),
+		.msg = "ig_host_data_fifo_int",
+		.reset_level = HNAE3_GLOBAL_RESET
+	}, {
+		.int_msk = BIT(4),
+		.msg = "ig_host_key_fifo_int",
+		.reset_level = HNAE3_GLOBAL_RESET
+	}, {
+		.int_msk = BIT(5),
+		.msg = "tx_qcn_fifo_int",
+		.reset_level = HNAE3_GLOBAL_RESET
+	}, {
+		.int_msk = BIT(6),
+		.msg = "rx_qcn_fifo_int",
+		.reset_level = HNAE3_GLOBAL_RESET
+	}, {
+		.int_msk = BIT(7),
+		.msg = "tx_pf_rd_fifo_int",
+		.reset_level = HNAE3_GLOBAL_RESET
+	}, {
+		.int_msk = BIT(8),
+		.msg = "rx_pf_rd_fifo_int",
+		.reset_level = HNAE3_GLOBAL_RESET
+	}, {
+		.int_msk = BIT(9),
+		.msg = "qm_eof_fifo_int",
+		.reset_level = HNAE3_GLOBAL_RESET
+	}, {
+		.int_msk = BIT(10),
+		.msg = "mb_rlt_fifo_int",
+		.reset_level = HNAE3_GLOBAL_RESET
+	}, {
+		.int_msk = BIT(11),
+		.msg = "dup_uncopy_fifo_int",
+		.reset_level = HNAE3_GLOBAL_RESET
+	}, {
+		.int_msk = BIT(12),
+		.msg = "dup_cnt_rd_fifo_int",
+		.reset_level = HNAE3_GLOBAL_RESET
+	}, {
+		.int_msk = BIT(13),
+		.msg = "dup_cnt_drop_fifo_int",
+		.reset_level = HNAE3_GLOBAL_RESET
+	}, {
+		.int_msk = BIT(14),
+		.msg = "dup_cnt_wrb_fifo_int",
+		.reset_level = HNAE3_GLOBAL_RESET
+	}, {
+		.int_msk = BIT(15),
+		.msg = "host_cmd_fifo_int",
+		.reset_level = HNAE3_GLOBAL_RESET
+	}, {
+		.int_msk = BIT(16),
+		.msg = "mac_cmd_fifo_int",
+		.reset_level = HNAE3_GLOBAL_RESET
+	}, {
+		.int_msk = BIT(17),
+		.msg = "host_cmd_bitmap_empty_int",
+		.reset_level = HNAE3_GLOBAL_RESET
+	}, {
+		.int_msk = BIT(18),
+		.msg = "mac_cmd_bitmap_empty_int",
+		.reset_level = HNAE3_GLOBAL_RESET
+	}, {
+		.int_msk = BIT(19),
+		.msg = "dup_bitmap_empty_int",
+		.reset_level = HNAE3_GLOBAL_RESET
+	}, {
+		.int_msk = BIT(20),
+		.msg = "out_queue_bitmap_empty_int",
+		.reset_level = HNAE3_GLOBAL_RESET
+	}, {
+		.int_msk = BIT(21),
+		.msg = "bank2_bitmap_empty_int",
+		.reset_level = HNAE3_GLOBAL_RESET
+	}, {
+		.int_msk = BIT(22),
+		.msg = "bank1_bitmap_empty_int",
+		.reset_level = HNAE3_GLOBAL_RESET
+	}, {
+		.int_msk = BIT(23),
+		.msg = "bank0_bitmap_empty_int",
+		.reset_level = HNAE3_GLOBAL_RESET
+	}, {
+		/* sentinel */
+	}
+};
+
+static const struct hclge_hw_error hclge_ssu_ets_tcg_int[] = {
+	{
+		.int_msk = BIT(0),
+		.msg = "ets_rd_int_rx_tcg",
+		.reset_level = HNAE3_GLOBAL_RESET
+	}, {
+		.int_msk = BIT(1),
+		.msg = "ets_wr_int_rx_tcg",
+		.reset_level = HNAE3_GLOBAL_RESET
+	}, {
+		.int_msk = BIT(2),
+		.msg = "ets_rd_int_tx_tcg",
+		.reset_level = HNAE3_GLOBAL_RESET
+	}, {
+		.int_msk = BIT(3),
+		.msg = "ets_wr_int_tx_tcg",
+		.reset_level = HNAE3_GLOBAL_RESET
+	}, {
+		/* sentinel */
+	}
+};
+
+static const struct hclge_hw_error hclge_ssu_port_based_pf_int[] = {
+	{
+		.int_msk = BIT(0),
+		.msg = "roc_pkt_without_key_port",
+		.reset_level = HNAE3_FUNC_RESET
+	}, {
+		.int_msk = BIT(9),
+		.msg = "low_water_line_err_port",
+		.reset_level = HNAE3_NONE_RESET
+	}, {
+		.int_msk = BIT(10),
+		.msg = "hi_water_line_err_port",
+		.reset_level = HNAE3_GLOBAL_RESET
+	}, {
+		/* sentinel */
+	}
+};
+
+static const struct hclge_hw_error hclge_rocee_qmm_ovf_err_int[] = {
+	{
+		.int_msk = 0,
+		.msg = "rocee qmm ovf: sgid invalid err"
+	}, {
+		.int_msk = 0x4,
+		.msg = "rocee qmm ovf: sgid ovf err"
+	}, {
+		.int_msk = 0x8,
+		.msg = "rocee qmm ovf: smac invalid err"
+	}, {
+		.int_msk = 0xC,
+		.msg = "rocee qmm ovf: smac ovf err"
+	}, {
+		.int_msk = 0x10,
+		.msg = "rocee qmm ovf: cqc invalid err"
+	}, {
+		.int_msk = 0x11,
+		.msg = "rocee qmm ovf: cqc ovf err"
+	}, {
+		.int_msk = 0x12,
+		.msg = "rocee qmm ovf: cqc hopnum err"
+	}, {
+		.int_msk = 0x13,
+		.msg = "rocee qmm ovf: cqc ba0 err"
+	}, {
+		.int_msk = 0x14,
+		.msg = "rocee qmm ovf: srqc invalid err"
+	}, {
+		.int_msk = 0x15,
+		.msg = "rocee qmm ovf: srqc ovf err"
+	}, {
+		.int_msk = 0x16,
+		.msg = "rocee qmm ovf: srqc hopnum err"
+	}, {
+		.int_msk = 0x17,
+		.msg = "rocee qmm ovf: srqc ba0 err"
+	}, {
+		.int_msk = 0x18,
+		.msg = "rocee qmm ovf: mpt invalid err"
+	}, {
+		.int_msk = 0x19,
+		.msg = "rocee qmm ovf: mpt ovf err"
+	}, {
+		.int_msk = 0x1A,
+		.msg = "rocee qmm ovf: mpt hopnum err"
+	}, {
+		.int_msk = 0x1B,
+		.msg = "rocee qmm ovf: mpt ba0 err"
+	}, {
+		.int_msk = 0x1C,
+		.msg = "rocee qmm ovf: qpc invalid err"
+	}, {
+		.int_msk = 0x1D,
+		.msg = "rocee qmm ovf: qpc ovf err"
+	}, {
+		.int_msk = 0x1E,
+		.msg = "rocee qmm ovf: qpc hopnum err"
+	}, {
+		.int_msk = 0x1F,
+		.msg = "rocee qmm ovf: qpc ba0 err"
+	}, {
+		/* sentinel */
+	}
+};
+
+static const struct hclge_mod_reg_info hclge_ssu_reg_0_info[] = {
+	{
+		.reg_name = "SSU_BP_STATUS_0~5",
+		.reg_offset_group = { 5, 6, 7, 8, 9, 10},
+		.group_size = 6
+	}, {
+		.reg_name = "LO_PRI_UNICAST_CUR_CNT",
+		.reg_offset_group = {54},
+		.group_size = 1
+	}, {
+		.reg_name = "HI/LO_PRI_MULTICAST_CUR_CNT",
+		.reg_offset_group = {55, 56},
+		.group_size = 2
+	}, {
+		.reg_name = "SSU_MB_RD_RLT_DROP_CNT",
+		.reg_offset_group = {29},
+		.group_size = 1
+	}, {
+		.reg_name = "SSU_PPP_MAC_KEY_NUM",
+		.reg_offset_group = {31, 30},
+		.group_size = 2
+	}, {
+		.reg_name = "SSU_PPP_HOST_KEY_NUM",
+		.reg_offset_group = {33, 32},
+		.group_size = 2
+	}, {
+		.reg_name = "PPP_SSU_MAC/HOST_RLT_NUM",
+		.reg_offset_group = {35, 34, 37, 36},
+		.group_size = 4
+	}, {
+		.reg_name = "FULL/PART_DROP_NUM",
+		.reg_offset_group = {18, 19},
+		.group_size = 2
+	}, {
+		.reg_name = "PPP_KEY/RLT_DROP_NUM",
+		.reg_offset_group = {20, 21},
+		.group_size = 2
+	}, {
+		.reg_name = "NIC/ROC_L2_ERR_DROP_PKT_CNT",
+		.reg_offset_group = {48, 49},
+		.group_size = 2
+	}, {
+		.reg_name = "NIC/ROC_L2_ERR_DROP_PKT_CNT_RX",
+		.reg_offset_group = {50, 51},
+		.group_size = 2
+	},
+};
+
+static const struct hclge_mod_reg_info hclge_ssu_reg_1_info[] = {
+	{
+		.reg_name = "RX_PACKET_IN/OUT_CNT",
+		.reg_offset_group = {13, 12, 15, 14},
+		.group_size = 4
+	}, {
+		.reg_name = "TX_PACKET_IN/OUT_CNT",
+		.reg_offset_group = {17, 16, 19, 18},
+		.group_size = 4
+	}, {
+		.reg_name = "RX_PACKET_TC0_IN/OUT_CNT",
+		.reg_offset_group = {25, 24, 41, 40},
+		.group_size = 4
+	}, {
+		.reg_name = "RX_PACKET_TC1_IN/OUT_CNT",
+		.reg_offset_group = {27, 26, 43, 42},
+		.group_size = 4
+	}, {
+		.reg_name = "RX_PACKET_TC2_IN/OUT_CNT",
+		.reg_offset_group = {29, 28, 45, 44},
+		.group_size = 4
+	}, {
+		.reg_name = "RX_PACKET_TC3_IN/OUT_CNT",
+		.reg_offset_group = {31, 30, 47, 46},
+		.group_size = 4
+	}, {
+		.reg_name = "RX_PACKET_TC4_IN/OUT_CNT",
+		.reg_offset_group = {33, 32, 49, 48},
+		.group_size = 4
+	}, {
+		.reg_name = "RX_PACKET_TC5_IN/OUT_CNT",
+		.reg_offset_group = {35, 34, 51, 50},
+		.group_size = 4
+	}, {
+		.reg_name = "RX_PACKET_TC6_IN/OUT_CNT",
+		.reg_offset_group = {37, 36, 53, 52},
+		.group_size = 4
+	}, {
+		.reg_name = "RX_PACKET_TC7_IN/OUT_CNT",
+		.reg_offset_group = {39, 38, 55, 54},
+		.group_size = 4
+	}, {
+		.reg_name = "TX_PACKET_TC0_IN/OUT_CNT",
+		.reg_offset_group = {57, 56, 73, 72},
+		.group_size = 4
+	}, {
+		.reg_name = "TX_PACKET_TC1_IN/OUT_CNT",
+		.reg_offset_group = {59, 58, 75, 74},
+		.group_size = 4
+	}, {
+		.reg_name = "TX_PACKET_TC2_IN/OUT_CNT",
+		.reg_offset_group = {61, 60, 77, 76},
+		.group_size = 4
+	}, {
+		.reg_name = "TX_PACKET_TC3_IN/OUT_CNT",
+		.reg_offset_group = {63, 62, 79, 78},
+		.group_size = 4
+	}, {
+		.reg_name = "TX_PACKET_TC4_IN/OUT_CNT",
+		.reg_offset_group = {65, 64, 81, 80},
+		.group_size = 4
+	}, {
+		.reg_name = "TX_PACKET_TC5_IN/OUT_CNT",
+		.reg_offset_group = {67, 66, 83, 82},
+		.group_size = 4
+	}, {
+		.reg_name = "TX_PACKET_TC6_IN/OUT_CNT",
+		.reg_offset_group = {69, 68, 85, 84},
+		.group_size = 4
+	}, {
+		.reg_name = "TX_PACKET_TC7_IN/OUT_CNT",
+		.reg_offset_group = {71, 70, 87, 86},
+		.group_size = 4
+	}, {
+		.reg_name = "PACKET_TC0~3_CURR_BUFFER_CNT",
+		.reg_offset_group = {1, 2, 3, 4},
+		.group_size = 4
+	}, {
+		.reg_name = "PACKET_TC4~7_CURR_BUFFER_CNT",
+		.reg_offset_group = {5, 6, 7, 8},
+		.group_size = 4
+	}, {
+		.reg_name = "ROC_RX_PACKET_IN_CNT",
+		.reg_offset_group = {21, 20},
+		.group_size = 2
+	}, {
+		.reg_name = "ROC_TX_PACKET_OUT_CNT",
+		.reg_offset_group = {23, 22},
+		.group_size = 2
+	}
+};
+
+static const struct hclge_mod_reg_info hclge_rpu_reg_0_info[] = {
+	{
+		.reg_name = "RPU_FSM_DFX_ST0/ST1_TNL",
+		.has_suffix = true,
+		.reg_offset_group = {1, 2},
+		.group_size = 2
+	}, {
+		.reg_name = "RPU_RX_PKT_DROP_CNT_TNL",
+		.has_suffix = true,
+		.reg_offset_group = {3},
+		.group_size = 1
+	}
+};
+
+static const struct hclge_mod_reg_info hclge_rpu_reg_1_info[] = {
+	{
+		.reg_name = "FIFO_DFX_ST0_1_2_4",
+		.reg_offset_group = {1, 2, 3, 5},
+		.group_size = 4
+	}
+};
+
+static const struct hclge_mod_reg_info hclge_igu_egu_reg_info[] = {
+	{
+		.reg_name = "IGU_RX_ERR_PKT",
+		.reg_offset_group = {1},
+		.group_size = 1
+	}, {
+		.reg_name = "IGU_RX_OUT_ALL_PKT",
+		.reg_offset_group = {29, 28},
+		.group_size = 2
+	}, {
+		.reg_name = "EGU_TX_OUT_ALL_PKT",
+		.reg_offset_group = {39, 38},
+		.group_size = 2
+	}, {
+		.reg_name = "EGU_TX_ERR_PKT",
+		.reg_offset_group = {5},
+		.group_size = 1
+	}
+};
+
+static const struct hclge_mod_reg_info hclge_gen_reg_info_tnl[] = {
+	{
+		.reg_name = "SSU2RPU_TNL_WR_PKT_CNT_TNL",
+		.has_suffix = true,
+		.reg_offset_group = {1},
+		.group_size = 1
+	}, {
+		.reg_name = "RPU2HST_TNL_WR_PKT_CNT_TNL",
+		.has_suffix = true,
+		.reg_offset_group = {12},
+		.group_size = 1
+	}
+};
+
+static const struct hclge_mod_reg_info hclge_gen_reg_info[] = {
+	{
+		.reg_name = "SSU_OVERSIZE_DROP_CNT",
+		.reg_offset_group = {12},
+		.group_size = 1
+	}, {
+		.reg_name = "ROCE_RX_BYPASS_5NS_DROP_NUM",
+		.reg_offset_group = {13},
+		.group_size = 1
+	}, {
+		.reg_name = "RX_PKT_IN/OUT_ERR_CNT",
+		.reg_offset_group = {15, 14, 19, 18},
+		.group_size = 4
+	}, {
+		.reg_name = "TX_PKT_IN/OUT_ERR_CNT",
+		.reg_offset_group = {17, 16, 21, 20},
+		.group_size = 4
+	}, {
+		.reg_name = "ETS_TC_READY",
+		.reg_offset_group = {22},
+		.group_size = 1
+	}, {
+		.reg_name = "MIB_TX/RX_BAD_PKTS",
+		.reg_offset_group = {19, 18, 29, 28},
+		.group_size = 4
+	}, {
+		.reg_name = "MIB_TX/RX_GOOD_PKTS",
+		.reg_offset_group = {21, 20, 31, 30},
+		.group_size = 4
+	}, {
+		.reg_name = "MIB_TX/RX_TOTAL_PKTS",
+		.reg_offset_group = {23, 22, 33, 32},
+		.group_size = 4
+	}, {
+		.reg_name = "MIB_TX/RX_PAUSE_PKTS",
+		.reg_offset_group = {25, 24, 35, 34},
+		.group_size = 4
+	}, {
+		.reg_name = "MIB_TX_ERR_ALL_PKTS",
+		.reg_offset_group = {27, 26},
+		.group_size = 2
+	}, {
+		.reg_name = "MIB_RX_FCS_ERR_PKTS",
+		.reg_offset_group = {37, 36},
+		.group_size = 2
+	}, {
+		.reg_name = "IGU_EGU_AUTO_GATE_EN",
+		.reg_offset_group = {42},
+		.group_size = 1
+	}, {
+		.reg_name = "IGU_EGU_INT_SRC",
+		.reg_offset_group = {43},
+		.group_size = 1
+	}, {
+		.reg_name = "EGU_READY_NUM_CFG",
+		.reg_offset_group = {44},
+		.group_size = 1
+	}, {
+		.reg_name = "IGU_EGU_TNL_DFX",
+		.reg_offset_group = {45},
+		.group_size = 1
+	}, {
+		.reg_name = "TX_TNL_NOTE_PKT",
+		.reg_offset_group = {46},
+		.group_size = 1
+	}
+};
+
+static const struct hclge_mod_reg_common_msg hclge_ssu_reg_common_msg[] = {
+	{
+		.cmd = HCLGE_OPC_DFX_SSU_REG_0,
+		.result_regs = hclge_ssu_reg_0_info,
+		.bd_num = HCLGE_BD_NUM_SSU_REG_0,
+		.result_regs_size = ARRAY_SIZE(hclge_ssu_reg_0_info)
+	}, {
+		.cmd = HCLGE_OPC_DFX_SSU_REG_1,
+		.result_regs = hclge_ssu_reg_1_info,
+		.bd_num = HCLGE_BD_NUM_SSU_REG_1,
+		.result_regs_size = ARRAY_SIZE(hclge_ssu_reg_1_info)
+	}, {
+		.cmd = HCLGE_OPC_DFX_RPU_REG_0,
+		.result_regs = hclge_rpu_reg_0_info,
+		.bd_num = HCLGE_BD_NUM_RPU_REG_0,
+		.result_regs_size = ARRAY_SIZE(hclge_rpu_reg_0_info),
+		.need_para = true
+	}, {
+		.cmd = HCLGE_OPC_DFX_RPU_REG_1,
+		.result_regs = hclge_rpu_reg_1_info,
+		.bd_num = HCLGE_BD_NUM_RPU_REG_1,
+		.result_regs_size = ARRAY_SIZE(hclge_rpu_reg_1_info)
+	}, {
+		.cmd = HCLGE_OPC_DFX_IGU_EGU_REG,
+		.result_regs = hclge_igu_egu_reg_info,
+		.bd_num = HCLGE_BD_NUM_IGU_EGU_REG,
+		.result_regs_size = ARRAY_SIZE(hclge_igu_egu_reg_info)
+	}, {
+		.cmd = HCLGE_OPC_DFX_GEN_REG,
+		.result_regs = hclge_gen_reg_info_tnl,
+		.bd_num = HCLGE_BD_NUM_GEN_REG,
+		.result_regs_size = ARRAY_SIZE(hclge_gen_reg_info_tnl),
+		.need_para = true
+	}, {
+		.cmd = HCLGE_OPC_DFX_GEN_REG,
+		.result_regs = hclge_gen_reg_info,
+		.bd_num = HCLGE_BD_NUM_GEN_REG,
+		.result_regs_size = ARRAY_SIZE(hclge_gen_reg_info)
+	}
+};
+
+static int
+hclge_print_mod_reg_info(struct device *dev, struct hclge_desc *desc,
+			 const struct hclge_mod_reg_info *reg_info, int size)
+{
+	int i, j, pos, actual_len;
+	u8 offset, bd_idx, index;
+	char *buf;
+
+	buf = kzalloc(HCLGE_MOD_REG_INFO_LEN_MAX, GFP_KERNEL);
+	if (!buf)
+		return -ENOMEM;
+
+	for (i = 0; i < size; i++) {
+		actual_len = strlen(reg_info[i].reg_name) +
+			     HCLGE_MOD_REG_EXTRA_LEN +
+			     HCLGE_MOD_REG_VALUE_LEN * reg_info[i].group_size;
+		if (actual_len > HCLGE_MOD_REG_INFO_LEN_MAX) {
+			dev_info(dev, "length of reg(%s) is invalid, len=%d\n",
+				 reg_info[i].reg_name, actual_len);
+			continue;
+		}
+
+		pos = scnprintf(buf, HCLGE_MOD_REG_INFO_LEN_MAX, "%s",
+				reg_info[i].reg_name);
+		if (reg_info[i].has_suffix)
+			pos += scnprintf(buf + pos,
+					 HCLGE_MOD_REG_INFO_LEN_MAX - pos, "%u",
+					 le32_to_cpu(desc->data[0]));
+		pos += scnprintf(buf + pos,
+				 HCLGE_MOD_REG_INFO_LEN_MAX - pos,
+				 ":");
+		for (j = 0; j < reg_info[i].group_size; j++) {
+			offset = reg_info[i].reg_offset_group[j];
+			index = offset % HCLGE_DESC_DATA_LEN;
+			bd_idx = offset / HCLGE_DESC_DATA_LEN;
+			pos += scnprintf(buf + pos,
+					 HCLGE_MOD_REG_INFO_LEN_MAX - pos,
+					 " %08x",
+					 le32_to_cpu(desc[bd_idx].data[index]));
+		}
+		dev_info(dev, "%s\n", buf);
+	}
+
+	kfree(buf);
+	return 0;
+}
+
+static bool hclge_err_mod_check_support_cmd(enum hclge_opcode_type opcode,
+					    struct hclge_dev *hdev)
+{
+	if (opcode == HCLGE_OPC_DFX_GEN_REG &&
+	    !hnae3_ae_dev_gen_reg_dfx_supported(hdev))
+		return false;
+	return true;
+}
+
+/* For each common msg, send cmdq to IMP and print result reg info.
+ * If there is a parameter, loop it and request.
+ */
+static void
+hclge_query_reg_info(struct hclge_dev *hdev,
+		     struct hclge_mod_reg_common_msg *msg, u32 loop_time,
+		     u32 *loop_para)
+{
+	int desc_len, i, ret;
+
+	desc_len = msg->bd_num * sizeof(struct hclge_desc);
+	msg->desc = kzalloc(desc_len, GFP_KERNEL);
+	if (!msg->desc) {
+		dev_err(&hdev->pdev->dev, "failed to query reg info, ret=%d",
+			-ENOMEM);
+		return;
+	}
+
+	for (i = 0; i < loop_time; i++) {
+		ret = hclge_dbg_cmd_send(hdev, msg->desc, *loop_para,
+					 msg->bd_num, msg->cmd);
+		loop_para++;
+		if (ret)
+			continue;
+		ret = hclge_print_mod_reg_info(&hdev->pdev->dev, msg->desc,
+					       msg->result_regs,
+					       msg->result_regs_size);
+		if (ret)
+			dev_err(&hdev->pdev->dev, "failed to print mod reg info, ret=%d\n",
+				ret);
+	}
+
+	kfree(msg->desc);
+}
+
+static void hclge_query_reg_info_of_ssu(struct hclge_dev *hdev)
+{
+	u32 loop_para[HCLGE_MOD_MSG_PARA_ARRAY_MAX_SIZE] = {0};
+	struct hclge_mod_reg_common_msg msg;
+	u8 i, j, num, loop_time;
+
+	num = ARRAY_SIZE(hclge_ssu_reg_common_msg);
+	for (i = 0; i < num; i++) {
+		msg = hclge_ssu_reg_common_msg[i];
+		if (!hclge_err_mod_check_support_cmd(msg.cmd, hdev))
+			continue;
+		loop_time = 1;
+		loop_para[0] = 0;
+		if (msg.need_para) {
+			loop_time = min(hdev->ae_dev->dev_specs.tnl_num,
+					HCLGE_MOD_MSG_PARA_ARRAY_MAX_SIZE);
+			for (j = 0; j < loop_time; j++)
+				loop_para[j] = j + 1;
+		}
+		hclge_query_reg_info(hdev, &msg, loop_time, loop_para);
+	}
+}
+
+static const struct hclge_hw_module_id hclge_hw_module_id_st[] = {
+	{
+		.module_id = MODULE_NONE,
+		.msg = "MODULE_NONE"
+	}, {
+		.module_id = MODULE_BIOS_COMMON,
+		.msg = "MODULE_BIOS_COMMON"
+	}, {
+		.module_id = MODULE_GE,
+		.msg = "MODULE_GE"
+	}, {
+		.module_id = MODULE_IGU_EGU,
+		.msg = "MODULE_IGU_EGU",
+		.query_reg_info = hclge_query_reg_info_of_ssu
+	}, {
+		.module_id = MODULE_LGE,
+		.msg = "MODULE_LGE"
+	}, {
+		.module_id = MODULE_NCSI,
+		.msg = "MODULE_NCSI"
+	}, {
+		.module_id = MODULE_PPP,
+		.msg = "MODULE_PPP"
+	}, {
+		.module_id = MODULE_QCN,
+		.msg = "MODULE_QCN"
+	}, {
+		.module_id = MODULE_RCB_RX,
+		.msg = "MODULE_RCB_RX"
+	}, {
+		.module_id = MODULE_RTC,
+		.msg = "MODULE_RTC"
+	}, {
+		.module_id = MODULE_SSU,
+		.msg = "MODULE_SSU",
+		.query_reg_info = hclge_query_reg_info_of_ssu
+	}, {
+		.module_id = MODULE_TM,
+		.msg = "MODULE_TM"
+	}, {
+		.module_id = MODULE_RCB_TX,
+		.msg = "MODULE_RCB_TX"
+	}, {
+		.module_id = MODULE_TXDMA,
+		.msg = "MODULE_TXDMA"
+	}, {
+		.module_id = MODULE_MASTER,
+		.msg = "MODULE_MASTER"
+	}, {
+		.module_id = MODULE_HIMAC,
+		.msg = "MODULE_HIMAC"
+	}, {
+		.module_id = MODULE_ROCEE_TOP,
+		.msg = "MODULE_ROCEE_TOP"
+	}, {
+		.module_id = MODULE_ROCEE_TIMER,
+		.msg = "MODULE_ROCEE_TIMER"
+	}, {
+		.module_id = MODULE_ROCEE_MDB,
+		.msg = "MODULE_ROCEE_MDB"
+	}, {
+		.module_id = MODULE_ROCEE_TSP,
+		.msg = "MODULE_ROCEE_TSP"
+	}, {
+		.module_id = MODULE_ROCEE_TRP,
+		.msg = "MODULE_ROCEE_TRP"
+	}, {
+		.module_id = MODULE_ROCEE_SCC,
+		.msg = "MODULE_ROCEE_SCC"
+	}, {
+		.module_id = MODULE_ROCEE_CAEP,
+		.msg = "MODULE_ROCEE_CAEP"
+	}, {
+		.module_id = MODULE_ROCEE_GEN_AC,
+		.msg = "MODULE_ROCEE_GEN_AC"
+	}, {
+		.module_id = MODULE_ROCEE_QMM,
+		.msg = "MODULE_ROCEE_QMM"
+	}, {
+		.module_id = MODULE_ROCEE_LSAN,
+		.msg = "MODULE_ROCEE_LSAN"
+	}
+};
+
+static const struct hclge_hw_type_id hclge_hw_type_id_st[] = {
+	{
+		.type_id = NONE_ERROR,
+		.msg = "none_error"
+	}, {
+		.type_id = FIFO_ERROR,
+		.msg = "fifo_error"
+	}, {
+		.type_id = MEMORY_ERROR,
+		.msg = "memory_error"
+	}, {
+		.type_id = POISON_ERROR,
+		.msg = "poison_error"
+	}, {
+		.type_id = MSIX_ECC_ERROR,
+		.msg = "msix_ecc_error"
+	}, {
+		.type_id = TQP_INT_ECC_ERROR,
+		.msg = "tqp_int_ecc_error"
+	}, {
+		.type_id = PF_ABNORMAL_INT_ERROR,
+		.msg = "pf_abnormal_int_error",
+		.cause_by_vf = true
+	}, {
+		.type_id = MPF_ABNORMAL_INT_ERROR,
+		.msg = "mpf_abnormal_int_error",
+		.cause_by_vf = true
+	}, {
+		.type_id = COMMON_ERROR,
+		.msg = "common_error"
+	}, {
+		.type_id = PORT_ERROR,
+		.msg = "port_error"
+	}, {
+		.type_id = ETS_ERROR,
+		.msg = "ets_error"
+	}, {
+		.type_id = NCSI_ERROR,
+		.msg = "ncsi_error"
+	}, {
+		.type_id = GLB_ERROR,
+		.msg = "glb_error"
+	}, {
+		.type_id = LINK_ERROR,
+		.msg = "link_error"
+	}, {
+		.type_id = PTP_ERROR,
+		.msg = "ptp_error"
+	}, {
+		.type_id = ROCEE_NORMAL_ERR,
+		.msg = "rocee_normal_error"
+	}, {
+		.type_id = ROCEE_OVF_ERR,
+		.msg = "rocee_ovf_error"
+	}, {
+		.type_id = ROCEE_BUS_ERR,
+		.msg = "rocee_bus_error"
+	},
+};
+
+static void hclge_log_error(struct device *dev, char *reg,
+			    const struct hclge_hw_error *err,
+			    u32 err_sts, unsigned long *reset_requests)
+{
+	while (err->msg) {
+		if (err->int_msk & err_sts) {
+			dev_err(dev, "%s %s found [error status=0x%x]\n",
+				reg, err->msg, err_sts);
+			if (err->reset_level &&
+			    err->reset_level != HNAE3_NONE_RESET)
+				set_bit(err->reset_level, reset_requests);
+		}
+		err++;
+	}
+}
+
+/* hclge_cmd_query_error: read the error information
+ * @hdev: pointer to struct hclge_dev
+ * @desc: descriptor for describing the command
+ * @cmd:  command opcode
+ * @flag: flag for extended command structure
+ *
+ * This function query the error info from hw register/s using command
+ */
+static int hclge_cmd_query_error(struct hclge_dev *hdev,
+				 struct hclge_desc *desc, u32 cmd, u16 flag)
+{
+	struct device *dev = &hdev->pdev->dev;
+	int desc_num = 1;
+	int ret;
+
+	hclge_cmd_setup_basic_desc(&desc[0], cmd, true);
+	if (flag) {
+		desc[0].flag |= cpu_to_le16(flag);
+		hclge_cmd_setup_basic_desc(&desc[1], cmd, true);
+		desc_num = 2;
+	}
+
+	ret = hclge_cmd_send(&hdev->hw, &desc[0], desc_num);
+	if (ret)
+		dev_err(dev, "query error cmd failed (%d)\n", ret);
+
+	return ret;
+}
+
+static int hclge_clear_mac_tnl_int(struct hclge_dev *hdev)
+{
+	struct hclge_desc desc;
+
+	hclge_cmd_setup_basic_desc(&desc, HCLGE_OPC_CLEAR_MAC_TNL_INT, false);
+	desc.data[0] = cpu_to_le32(HCLGE_MAC_TNL_INT_CLR);
+
+	return hclge_cmd_send(&hdev->hw, &desc, 1);
+}
+
+static int hclge_config_common_hw_err_int(struct hclge_dev *hdev, bool en)
+{
+	struct device *dev = &hdev->pdev->dev;
+	struct hclge_desc desc[2];
+	int ret;
+
+	/* configure common error interrupts */
+	hclge_cmd_setup_basic_desc(&desc[0], HCLGE_COMMON_ECC_INT_CFG, false);
+	desc[0].flag |= cpu_to_le16(HCLGE_COMM_CMD_FLAG_NEXT);
+	hclge_cmd_setup_basic_desc(&desc[1], HCLGE_COMMON_ECC_INT_CFG, false);
+
+	if (en) {
+		desc[0].data[0] = cpu_to_le32(HCLGE_IMP_TCM_ECC_ERR_INT_EN);
+		desc[0].data[2] = cpu_to_le32(HCLGE_CMDQ_NIC_ECC_ERR_INT_EN |
+					HCLGE_CMDQ_ROCEE_ECC_ERR_INT_EN);
+		desc[0].data[3] = cpu_to_le32(HCLGE_IMP_RD_POISON_ERR_INT_EN);
+		desc[0].data[4] = cpu_to_le32(HCLGE_TQP_ECC_ERR_INT_EN |
+					      HCLGE_MSIX_SRAM_ECC_ERR_INT_EN);
+		desc[0].data[5] = cpu_to_le32(HCLGE_IMP_ITCM4_ECC_ERR_INT_EN);
+	}
+
+	desc[1].data[0] = cpu_to_le32(HCLGE_IMP_TCM_ECC_ERR_INT_EN_MASK);
+	desc[1].data[2] = cpu_to_le32(HCLGE_CMDQ_NIC_ECC_ERR_INT_EN_MASK |
+				HCLGE_CMDQ_ROCEE_ECC_ERR_INT_EN_MASK);
+	desc[1].data[3] = cpu_to_le32(HCLGE_IMP_RD_POISON_ERR_INT_EN_MASK);
+	desc[1].data[4] = cpu_to_le32(HCLGE_TQP_ECC_ERR_INT_EN_MASK |
+				      HCLGE_MSIX_SRAM_ECC_ERR_INT_EN_MASK);
+	desc[1].data[5] = cpu_to_le32(HCLGE_IMP_ITCM4_ECC_ERR_INT_EN_MASK);
+
+	ret = hclge_cmd_send(&hdev->hw, &desc[0], 2);
+	if (ret)
+		dev_err(dev,
+			"fail(%d) to configure common err interrupts\n", ret);
+
+	return ret;
+}
+
+static int hclge_config_ncsi_hw_err_int(struct hclge_dev *hdev, bool en)
+{
+	struct device *dev = &hdev->pdev->dev;
+	struct hclge_desc desc;
+	int ret;
+
+	if (hdev->ae_dev->dev_version < HNAE3_DEVICE_VERSION_V2)
+		return 0;
+
+	/* configure NCSI error interrupts */
+	hclge_cmd_setup_basic_desc(&desc, HCLGE_NCSI_INT_EN, false);
+	if (en)
+		desc.data[0] = cpu_to_le32(HCLGE_NCSI_ERR_INT_EN);
+
+	ret = hclge_cmd_send(&hdev->hw, &desc, 1);
+	if (ret)
+		dev_err(dev,
+			"fail(%d) to configure  NCSI error interrupts\n", ret);
+
+	return ret;
+}
+
+static int hclge_config_igu_egu_hw_err_int(struct hclge_dev *hdev, bool en)
+{
+	struct device *dev = &hdev->pdev->dev;
+	struct hclge_desc desc;
+	int ret;
+
+	/* configure IGU,EGU error interrupts */
+	hclge_cmd_setup_basic_desc(&desc, HCLGE_IGU_COMMON_INT_EN, false);
+	desc.data[0] = cpu_to_le32(HCLGE_IGU_ERR_INT_TYPE);
+	if (en)
+		desc.data[0] |= cpu_to_le32(HCLGE_IGU_ERR_INT_EN);
+
+	desc.data[1] = cpu_to_le32(HCLGE_IGU_ERR_INT_EN_MASK);
+
+	ret = hclge_cmd_send(&hdev->hw, &desc, 1);
+	if (ret) {
+		dev_err(dev,
+			"fail(%d) to configure IGU common interrupts\n", ret);
+		return ret;
+	}
+
+	hclge_cmd_setup_basic_desc(&desc, HCLGE_IGU_EGU_TNL_INT_EN, false);
+	if (en)
+		desc.data[0] = cpu_to_le32(HCLGE_IGU_TNL_ERR_INT_EN);
+
+	desc.data[1] = cpu_to_le32(HCLGE_IGU_TNL_ERR_INT_EN_MASK);
+
+	ret = hclge_cmd_send(&hdev->hw, &desc, 1);
+	if (ret) {
+		dev_err(dev,
+			"fail(%d) to configure IGU-EGU TNL interrupts\n", ret);
+		return ret;
+	}
+
+	ret = hclge_config_ncsi_hw_err_int(hdev, en);
+
+	return ret;
+}
+
+static int hclge_config_ppp_error_interrupt(struct hclge_dev *hdev, u32 cmd,
+					    bool en)
+{
+	struct device *dev = &hdev->pdev->dev;
+	struct hclge_desc desc[2];
+	int ret;
+
+	/* configure PPP error interrupts */
+	hclge_cmd_setup_basic_desc(&desc[0], cmd, false);
+	desc[0].flag |= cpu_to_le16(HCLGE_COMM_CMD_FLAG_NEXT);
+	hclge_cmd_setup_basic_desc(&desc[1], cmd, false);
+
+	if (cmd == HCLGE_PPP_CMD0_INT_CMD) {
+		if (en) {
+			desc[0].data[0] =
+				cpu_to_le32(HCLGE_PPP_MPF_ECC_ERR_INT0_EN);
+			desc[0].data[1] =
+				cpu_to_le32(HCLGE_PPP_MPF_ECC_ERR_INT1_EN);
+			desc[0].data[4] = cpu_to_le32(HCLGE_PPP_PF_ERR_INT_EN);
+		}
+
+		desc[1].data[0] =
+			cpu_to_le32(HCLGE_PPP_MPF_ECC_ERR_INT0_EN_MASK);
+		desc[1].data[1] =
+			cpu_to_le32(HCLGE_PPP_MPF_ECC_ERR_INT1_EN_MASK);
+		if (hdev->ae_dev->dev_version >= HNAE3_DEVICE_VERSION_V2)
+			desc[1].data[2] =
+				cpu_to_le32(HCLGE_PPP_PF_ERR_INT_EN_MASK);
+	} else if (cmd == HCLGE_PPP_CMD1_INT_CMD) {
+		if (en) {
+			desc[0].data[0] =
+				cpu_to_le32(HCLGE_PPP_MPF_ECC_ERR_INT2_EN);
+			desc[0].data[1] =
+				cpu_to_le32(HCLGE_PPP_MPF_ECC_ERR_INT3_EN);
+		}
+
+		desc[1].data[0] =
+				cpu_to_le32(HCLGE_PPP_MPF_ECC_ERR_INT2_EN_MASK);
+		desc[1].data[1] =
+				cpu_to_le32(HCLGE_PPP_MPF_ECC_ERR_INT3_EN_MASK);
+	}
+
+	ret = hclge_cmd_send(&hdev->hw, &desc[0], 2);
+	if (ret)
+		dev_err(dev, "fail(%d) to configure PPP error intr\n", ret);
+
+	return ret;
+}
+
+static int hclge_config_ppp_hw_err_int(struct hclge_dev *hdev, bool en)
+{
+	int ret;
+
+	ret = hclge_config_ppp_error_interrupt(hdev, HCLGE_PPP_CMD0_INT_CMD,
+					       en);
+	if (ret)
+		return ret;
+
+	ret = hclge_config_ppp_error_interrupt(hdev, HCLGE_PPP_CMD1_INT_CMD,
+					       en);
+
+	return ret;
+}
+
+static int hclge_config_tm_hw_err_int(struct hclge_dev *hdev, bool en)
+{
+	struct device *dev = &hdev->pdev->dev;
+	struct hclge_desc desc;
+	int ret;
+
+	/* configure TM SCH hw errors */
+	hclge_cmd_setup_basic_desc(&desc, HCLGE_TM_SCH_ECC_INT_EN, false);
+	if (en)
+		desc.data[0] = cpu_to_le32(HCLGE_TM_SCH_ECC_ERR_INT_EN);
+
+	ret = hclge_cmd_send(&hdev->hw, &desc, 1);
+	if (ret) {
+		dev_err(dev, "fail(%d) to configure TM SCH errors\n", ret);
+		return ret;
+	}
+
+	/* configure TM QCN hw errors */
+	hclge_cmd_setup_basic_desc(&desc, HCLGE_TM_QCN_MEM_INT_CFG, false);
+	desc.data[0] = cpu_to_le32(HCLGE_TM_QCN_ERR_INT_TYPE);
+	if (en) {
+		desc.data[0] |= cpu_to_le32(HCLGE_TM_QCN_FIFO_INT_EN);
+		desc.data[1] = cpu_to_le32(HCLGE_TM_QCN_MEM_ERR_INT_EN);
+	}
+
+	ret = hclge_cmd_send(&hdev->hw, &desc, 1);
+	if (ret)
+		dev_err(dev,
+			"fail(%d) to configure TM QCN mem errors\n", ret);
+
+	return ret;
+}
+
+static int hclge_config_mac_err_int(struct hclge_dev *hdev, bool en)
+{
+	struct device *dev = &hdev->pdev->dev;
+	struct hclge_desc desc;
+	int ret;
+
+	/* configure MAC common error interrupts */
+	hclge_cmd_setup_basic_desc(&desc, HCLGE_MAC_COMMON_INT_EN, false);
+	if (en)
+		desc.data[0] = cpu_to_le32(HCLGE_MAC_COMMON_ERR_INT_EN);
+
+	desc.data[1] = cpu_to_le32(HCLGE_MAC_COMMON_ERR_INT_EN_MASK);
+
+	ret = hclge_cmd_send(&hdev->hw, &desc, 1);
+	if (ret)
+		dev_err(dev,
+			"fail(%d) to configure MAC COMMON error intr\n", ret);
+
+	return ret;
+}
+
+int hclge_config_mac_tnl_int(struct hclge_dev *hdev, bool en)
+{
+	struct hclge_desc desc;
+
+	hclge_cmd_setup_basic_desc(&desc, HCLGE_OPC_MAC_TNL_INT_EN, false);
+	if (en)
+		desc.data[0] = cpu_to_le32(HCLGE_MAC_TNL_INT_EN);
+	else
+		desc.data[0] = 0;
+
+	desc.data[1] = cpu_to_le32(HCLGE_MAC_TNL_INT_EN_MASK);
+
+	return hclge_cmd_send(&hdev->hw, &desc, 1);
+}
+
+static int hclge_config_ppu_error_interrupts(struct hclge_dev *hdev, u32 cmd,
+					     bool en)
+{
+	struct device *dev = &hdev->pdev->dev;
+	struct hclge_desc desc[2];
+	int desc_num = 1;
+	int ret;
+
+	/* configure PPU error interrupts */
+	if (cmd == HCLGE_PPU_MPF_ECC_INT_CMD) {
+		hclge_cmd_setup_basic_desc(&desc[0], cmd, false);
+		desc[0].flag |= cpu_to_le16(HCLGE_COMM_CMD_FLAG_NEXT);
+		hclge_cmd_setup_basic_desc(&desc[1], cmd, false);
+		if (en) {
+			desc[0].data[0] =
+				cpu_to_le32(HCLGE_PPU_MPF_ABNORMAL_INT0_EN);
+			desc[0].data[1] =
+				cpu_to_le32(HCLGE_PPU_MPF_ABNORMAL_INT1_EN);
+			desc[1].data[3] =
+				cpu_to_le32(HCLGE_PPU_MPF_ABNORMAL_INT3_EN);
+			desc[1].data[4] =
+				cpu_to_le32(HCLGE_PPU_MPF_ABNORMAL_INT2_EN);
+		}
+
+		desc[1].data[0] =
+			cpu_to_le32(HCLGE_PPU_MPF_ABNORMAL_INT0_EN_MASK);
+		desc[1].data[1] =
+			cpu_to_le32(HCLGE_PPU_MPF_ABNORMAL_INT1_EN_MASK);
+		desc[1].data[2] =
+			cpu_to_le32(HCLGE_PPU_MPF_ABNORMAL_INT2_EN_MASK);
+		desc[1].data[3] |=
+			cpu_to_le32(HCLGE_PPU_MPF_ABNORMAL_INT3_EN_MASK);
+		desc_num = 2;
+	} else if (cmd == HCLGE_PPU_MPF_OTHER_INT_CMD) {
+		hclge_cmd_setup_basic_desc(&desc[0], cmd, false);
+		if (en)
+			desc[0].data[0] =
+				cpu_to_le32(HCLGE_PPU_MPF_ABNORMAL_INT2_EN2);
+
+		desc[0].data[2] =
+			cpu_to_le32(HCLGE_PPU_MPF_ABNORMAL_INT2_EN2_MASK);
+	} else if (cmd == HCLGE_PPU_PF_OTHER_INT_CMD) {
+		hclge_cmd_setup_basic_desc(&desc[0], cmd, false);
+		if (en)
+			desc[0].data[0] =
+				cpu_to_le32(HCLGE_PPU_PF_ABNORMAL_INT_EN);
+
+		desc[0].data[2] =
+			cpu_to_le32(HCLGE_PPU_PF_ABNORMAL_INT_EN_MASK);
+	} else {
+		dev_err(dev, "Invalid cmd to configure PPU error interrupts\n");
+		return -EINVAL;
+	}
+
+	ret = hclge_cmd_send(&hdev->hw, &desc[0], desc_num);
+
+	return ret;
+}
+
+static int hclge_config_ppu_hw_err_int(struct hclge_dev *hdev, bool en)
+{
+	struct device *dev = &hdev->pdev->dev;
+	int ret;
+
+	ret = hclge_config_ppu_error_interrupts(hdev, HCLGE_PPU_MPF_ECC_INT_CMD,
+						en);
+	if (ret) {
+		dev_err(dev, "fail(%d) to configure PPU MPF ECC error intr\n",
+			ret);
+		return ret;
+	}
+
+	ret = hclge_config_ppu_error_interrupts(hdev,
+						HCLGE_PPU_MPF_OTHER_INT_CMD,
+						en);
+	if (ret) {
+		dev_err(dev, "fail(%d) to configure PPU MPF other intr\n", ret);
+		return ret;
+	}
+
+	ret = hclge_config_ppu_error_interrupts(hdev,
+						HCLGE_PPU_PF_OTHER_INT_CMD, en);
+	if (ret)
+		dev_err(dev, "fail(%d) to configure PPU PF error interrupts\n",
+			ret);
+	return ret;
+}
+
+static int hclge_config_ssu_hw_err_int(struct hclge_dev *hdev, bool en)
+{
+	struct device *dev = &hdev->pdev->dev;
+	struct hclge_desc desc[2];
+	int ret;
+
+	/* configure SSU ecc error interrupts */
+	hclge_cmd_setup_basic_desc(&desc[0], HCLGE_SSU_ECC_INT_CMD, false);
+	desc[0].flag |= cpu_to_le16(HCLGE_COMM_CMD_FLAG_NEXT);
+	hclge_cmd_setup_basic_desc(&desc[1], HCLGE_SSU_ECC_INT_CMD, false);
+	if (en) {
+		desc[0].data[0] = cpu_to_le32(HCLGE_SSU_1BIT_ECC_ERR_INT_EN);
+		desc[0].data[1] =
+			cpu_to_le32(HCLGE_SSU_MULTI_BIT_ECC_ERR_INT_EN);
+		desc[0].data[4] = cpu_to_le32(HCLGE_SSU_BIT32_ECC_ERR_INT_EN);
+	}
+
+	desc[1].data[0] = cpu_to_le32(HCLGE_SSU_1BIT_ECC_ERR_INT_EN_MASK);
+	desc[1].data[1] = cpu_to_le32(HCLGE_SSU_MULTI_BIT_ECC_ERR_INT_EN_MASK);
+	desc[1].data[2] = cpu_to_le32(HCLGE_SSU_BIT32_ECC_ERR_INT_EN_MASK);
+
+	ret = hclge_cmd_send(&hdev->hw, &desc[0], 2);
+	if (ret) {
+		dev_err(dev,
+			"fail(%d) to configure SSU ECC error interrupt\n", ret);
+		return ret;
+	}
+
+	/* configure SSU common error interrupts */
+	hclge_cmd_setup_basic_desc(&desc[0], HCLGE_SSU_COMMON_INT_CMD, false);
+	desc[0].flag |= cpu_to_le16(HCLGE_COMM_CMD_FLAG_NEXT);
+	hclge_cmd_setup_basic_desc(&desc[1], HCLGE_SSU_COMMON_INT_CMD, false);
+
+	if (en) {
+		if (hdev->ae_dev->dev_version >= HNAE3_DEVICE_VERSION_V2)
+			desc[0].data[0] =
+				cpu_to_le32(HCLGE_SSU_COMMON_INT_EN);
+		else
+			desc[0].data[0] =
+				cpu_to_le32(HCLGE_SSU_COMMON_INT_EN & ~BIT(5));
+		desc[0].data[1] = cpu_to_le32(HCLGE_SSU_PORT_BASED_ERR_INT_EN);
+		desc[0].data[2] =
+			cpu_to_le32(HCLGE_SSU_FIFO_OVERFLOW_ERR_INT_EN);
+	}
+
+	desc[1].data[0] = cpu_to_le32(HCLGE_SSU_COMMON_INT_EN_MASK |
+				HCLGE_SSU_PORT_BASED_ERR_INT_EN_MASK);
+	desc[1].data[1] = cpu_to_le32(HCLGE_SSU_FIFO_OVERFLOW_ERR_INT_EN_MASK);
+
+	ret = hclge_cmd_send(&hdev->hw, &desc[0], 2);
+	if (ret)
+		dev_err(dev,
+			"fail(%d) to configure SSU COMMON error intr\n", ret);
+
+	return ret;
+}
+
+/* hclge_query_bd_num: query number of buffer descriptors
+ * @hdev: pointer to struct hclge_dev
+ * @is_ras: true for ras, false for msix
+ * @mpf_bd_num: number of main PF interrupt buffer descriptors
+ * @pf_bd_num: number of not main PF interrupt buffer descriptors
+ *
+ * This function querys number of mpf and pf buffer descriptors.
+ */
+static int hclge_query_bd_num(struct hclge_dev *hdev, bool is_ras,
+			      u32 *mpf_bd_num, u32 *pf_bd_num)
+{
+	struct device *dev = &hdev->pdev->dev;
+	u32 mpf_min_bd_num, pf_min_bd_num;
+	enum hclge_opcode_type opcode;
+	struct hclge_desc desc_bd;
+	int ret;
+
+	if (is_ras) {
+		opcode = HCLGE_QUERY_RAS_INT_STS_BD_NUM;
+		mpf_min_bd_num = HCLGE_MPF_RAS_INT_MIN_BD_NUM;
+		pf_min_bd_num = HCLGE_PF_RAS_INT_MIN_BD_NUM;
+	} else {
+		opcode = HCLGE_QUERY_MSIX_INT_STS_BD_NUM;
+		mpf_min_bd_num = HCLGE_MPF_MSIX_INT_MIN_BD_NUM;
+		pf_min_bd_num = HCLGE_PF_MSIX_INT_MIN_BD_NUM;
+	}
+
+	hclge_cmd_setup_basic_desc(&desc_bd, opcode, true);
+	ret = hclge_cmd_send(&hdev->hw, &desc_bd, 1);
+	if (ret) {
+		dev_err(dev, "fail(%d) to query msix int status bd num\n",
+			ret);
+		return ret;
+	}
+
+	*mpf_bd_num = le32_to_cpu(desc_bd.data[0]);
+	*pf_bd_num = le32_to_cpu(desc_bd.data[1]);
+	if (*mpf_bd_num < mpf_min_bd_num || *pf_bd_num < pf_min_bd_num) {
+		dev_err(dev, "Invalid bd num: mpf(%u), pf(%u)\n",
+			*mpf_bd_num, *pf_bd_num);
+		return -EINVAL;
+	}
+
+	return 0;
+}
+
+/* hclge_handle_mpf_ras_error: handle all main PF RAS errors
+ * @hdev: pointer to struct hclge_dev
+ * @desc: descriptor for describing the command
+ * @num:  number of extended command structures
+ *
+ * This function handles all the main PF RAS errors in the
+ * hw register/s using command.
+ */
+static int hclge_handle_mpf_ras_error(struct hclge_dev *hdev,
+				      struct hclge_desc *desc,
+				      int num)
+{
+	struct hnae3_ae_dev *ae_dev = hdev->ae_dev;
+	struct device *dev = &hdev->pdev->dev;
+	__le32 *desc_data;
+	u32 status;
+	int ret;
+
+	/* query all main PF RAS errors */
+	hclge_cmd_setup_basic_desc(&desc[0], HCLGE_QUERY_CLEAR_MPF_RAS_INT,
+				   true);
+	ret = hclge_cmd_send(&hdev->hw, &desc[0], num);
+	if (ret) {
+		dev_err(dev, "query all mpf ras int cmd failed (%d)\n", ret);
+		return ret;
+	}
+
+	/* log HNS common errors */
+	status = le32_to_cpu(desc[0].data[0]);
+	if (status)
+		hclge_log_error(dev, "IMP_TCM_ECC_INT_STS",
+				&hclge_imp_tcm_ecc_int[0], status,
+				&ae_dev->hw_err_reset_req);
+
+	status = le32_to_cpu(desc[0].data[1]);
+	if (status)
+		hclge_log_error(dev, "CMDQ_MEM_ECC_INT_STS",
+				&hclge_cmdq_nic_mem_ecc_int[0], status,
+				&ae_dev->hw_err_reset_req);
+
+	if ((le32_to_cpu(desc[0].data[2])) & BIT(0))
+		dev_warn(dev, "imp_rd_data_poison_err found\n");
+
+	status = le32_to_cpu(desc[0].data[3]);
+	if (status)
+		hclge_log_error(dev, "TQP_INT_ECC_INT_STS",
+				&hclge_tqp_int_ecc_int[0], status,
+				&ae_dev->hw_err_reset_req);
+
+	status = le32_to_cpu(desc[0].data[4]);
+	if (status)
+		hclge_log_error(dev, "MSIX_ECC_INT_STS",
+				&hclge_msix_sram_ecc_int[0], status,
+				&ae_dev->hw_err_reset_req);
+
+	/* log SSU(Storage Switch Unit) errors */
+	desc_data = (__le32 *)&desc[2];
+	status = le32_to_cpu(*(desc_data + 2));
+	if (status)
+		hclge_log_error(dev, "SSU_ECC_MULTI_BIT_INT_0",
+				&hclge_ssu_mem_ecc_err_int[0], status,
+				&ae_dev->hw_err_reset_req);
+
+	status = le32_to_cpu(*(desc_data + 3)) & BIT(0);
+	if (status) {
+		dev_err(dev, "SSU_ECC_MULTI_BIT_INT_1 ssu_mem32_ecc_mbit_err found [error status=0x%x]\n",
+			status);
+		set_bit(HNAE3_GLOBAL_RESET, &ae_dev->hw_err_reset_req);
+	}
+
+	status = le32_to_cpu(*(desc_data + 4)) & HCLGE_SSU_COMMON_ERR_INT_MASK;
+	if (status)
+		hclge_log_error(dev, "SSU_COMMON_ERR_INT",
+				&hclge_ssu_com_err_int[0], status,
+				&ae_dev->hw_err_reset_req);
+
+	/* log IGU(Ingress Unit) errors */
+	desc_data = (__le32 *)&desc[3];
+	status = le32_to_cpu(*desc_data) & HCLGE_IGU_INT_MASK;
+	if (status)
+		hclge_log_error(dev, "IGU_INT_STS",
+				&hclge_igu_int[0], status,
+				&ae_dev->hw_err_reset_req);
+
+	/* log PPP(Programmable Packet Process) errors */
+	desc_data = (__le32 *)&desc[4];
+	status = le32_to_cpu(*(desc_data + 1));
+	if (status)
+		hclge_log_error(dev, "PPP_MPF_ABNORMAL_INT_ST1",
+				&hclge_ppp_mpf_abnormal_int_st1[0], status,
+				&ae_dev->hw_err_reset_req);
+
+	status = le32_to_cpu(*(desc_data + 3)) & HCLGE_PPP_MPF_INT_ST3_MASK;
+	if (status)
+		hclge_log_error(dev, "PPP_MPF_ABNORMAL_INT_ST3",
+				&hclge_ppp_mpf_abnormal_int_st3[0], status,
+				&ae_dev->hw_err_reset_req);
+
+	/* log PPU(RCB) errors */
+	desc_data = (__le32 *)&desc[5];
+	status = le32_to_cpu(*(desc_data + 1));
+	if (status) {
+		dev_err(dev,
+			"PPU_MPF_ABNORMAL_INT_ST1 rpu_rx_pkt_ecc_mbit_err found\n");
+		set_bit(HNAE3_GLOBAL_RESET, &ae_dev->hw_err_reset_req);
+	}
+
+	status = le32_to_cpu(*(desc_data + 2));
+	if (status)
+		hclge_log_error(dev, "PPU_MPF_ABNORMAL_INT_ST2",
+				&hclge_ppu_mpf_abnormal_int_st2[0], status,
+				&ae_dev->hw_err_reset_req);
+
+	status = le32_to_cpu(*(desc_data + 3)) & HCLGE_PPU_MPF_INT_ST3_MASK;
+	if (status)
+		hclge_log_error(dev, "PPU_MPF_ABNORMAL_INT_ST3",
+				&hclge_ppu_mpf_abnormal_int_st3[0], status,
+				&ae_dev->hw_err_reset_req);
+
+	/* log TM(Traffic Manager) errors */
+	desc_data = (__le32 *)&desc[6];
+	status = le32_to_cpu(*desc_data);
+	if (status)
+		hclge_log_error(dev, "TM_SCH_RINT",
+				&hclge_tm_sch_rint[0], status,
+				&ae_dev->hw_err_reset_req);
+
+	/* log QCN(Quantized Congestion Control) errors */
+	desc_data = (__le32 *)&desc[7];
+	status = le32_to_cpu(*desc_data) & HCLGE_QCN_FIFO_INT_MASK;
+	if (status)
+		hclge_log_error(dev, "QCN_FIFO_RINT",
+				&hclge_qcn_fifo_rint[0], status,
+				&ae_dev->hw_err_reset_req);
+
+	status = le32_to_cpu(*(desc_data + 1)) & HCLGE_QCN_ECC_INT_MASK;
+	if (status)
+		hclge_log_error(dev, "QCN_ECC_RINT",
+				&hclge_qcn_ecc_rint[0], status,
+				&ae_dev->hw_err_reset_req);
+
+	/* log NCSI errors */
+	desc_data = (__le32 *)&desc[9];
+	status = le32_to_cpu(*desc_data) & HCLGE_NCSI_ECC_INT_MASK;
+	if (status)
+		hclge_log_error(dev, "NCSI_ECC_INT_RPT",
+				&hclge_ncsi_err_int[0], status,
+				&ae_dev->hw_err_reset_req);
+
+	/* clear all main PF RAS errors */
+	hclge_comm_cmd_reuse_desc(&desc[0], false);
+	ret = hclge_cmd_send(&hdev->hw, &desc[0], num);
+	if (ret)
+		dev_err(dev, "clear all mpf ras int cmd failed (%d)\n", ret);
+
+	return ret;
+}
+
+/* hclge_handle_pf_ras_error: handle all PF RAS errors
+ * @hdev: pointer to struct hclge_dev
+ * @desc: descriptor for describing the command
+ * @num:  number of extended command structures
+ *
+ * This function handles all the PF RAS errors in the
+ * hw registers using command.
+ */
+static int hclge_handle_pf_ras_error(struct hclge_dev *hdev,
+				     struct hclge_desc *desc,
+				     int num)
+{
+	struct hnae3_ae_dev *ae_dev = hdev->ae_dev;
+	struct device *dev = &hdev->pdev->dev;
+	__le32 *desc_data;
+	u32 status;
+	int ret;
+
+	/* query all PF RAS errors */
+	hclge_cmd_setup_basic_desc(&desc[0], HCLGE_QUERY_CLEAR_PF_RAS_INT,
+				   true);
+	ret = hclge_cmd_send(&hdev->hw, &desc[0], num);
+	if (ret) {
+		dev_err(dev, "query all pf ras int cmd failed (%d)\n", ret);
+		return ret;
+	}
+
+	/* log SSU(Storage Switch Unit) errors */
+	status = le32_to_cpu(desc[0].data[0]);
+	if (status)
+		hclge_log_error(dev, "SSU_PORT_BASED_ERR_INT",
+				&hclge_ssu_port_based_err_int[0], status,
+				&ae_dev->hw_err_reset_req);
+
+	status = le32_to_cpu(desc[0].data[1]);
+	if (status)
+		hclge_log_error(dev, "SSU_FIFO_OVERFLOW_INT",
+				&hclge_ssu_fifo_overflow_int[0], status,
+				&ae_dev->hw_err_reset_req);
+
+	status = le32_to_cpu(desc[0].data[2]);
+	if (status)
+		hclge_log_error(dev, "SSU_ETS_TCG_INT",
+				&hclge_ssu_ets_tcg_int[0], status,
+				&ae_dev->hw_err_reset_req);
+
+	/* log IGU(Ingress Unit) EGU(Egress Unit) TNL errors */
+	desc_data = (__le32 *)&desc[1];
+	status = le32_to_cpu(*desc_data) & HCLGE_IGU_EGU_TNL_INT_MASK;
+	if (status)
+		hclge_log_error(dev, "IGU_EGU_TNL_INT_STS",
+				&hclge_igu_egu_tnl_int[0], status,
+				&ae_dev->hw_err_reset_req);
+
+	/* log PPU(RCB) errors */
+	desc_data = (__le32 *)&desc[3];
+	status = le32_to_cpu(*desc_data) & HCLGE_PPU_PF_INT_RAS_MASK;
+	if (status) {
+		hclge_log_error(dev, "PPU_PF_ABNORMAL_INT_ST0",
+				&hclge_ppu_pf_abnormal_int[0], status,
+				&ae_dev->hw_err_reset_req);
+		hclge_report_hw_error(hdev, HNAE3_PPU_POISON_ERROR);
+	}
+
+	/* clear all PF RAS errors */
+	hclge_comm_cmd_reuse_desc(&desc[0], false);
+	ret = hclge_cmd_send(&hdev->hw, &desc[0], num);
+	if (ret)
+		dev_err(dev, "clear all pf ras int cmd failed (%d)\n", ret);
+
+	return ret;
+}
+
+static int hclge_handle_all_ras_errors(struct hclge_dev *hdev)
+{
+	u32 mpf_bd_num, pf_bd_num, bd_num;
+	struct hclge_desc *desc;
+	int ret;
+
+	/* query the number of registers in the RAS int status */
+	ret = hclge_query_bd_num(hdev, true, &mpf_bd_num, &pf_bd_num);
+	if (ret)
+		return ret;
+
+	bd_num = max_t(u32, mpf_bd_num, pf_bd_num);
+	desc = kcalloc(bd_num, sizeof(struct hclge_desc), GFP_KERNEL);
+	if (!desc)
+		return -ENOMEM;
+
+	/* handle all main PF RAS errors */
+	ret = hclge_handle_mpf_ras_error(hdev, desc, mpf_bd_num);
+	if (ret) {
+		kfree(desc);
+		return ret;
+	}
+	memset(desc, 0, bd_num * sizeof(struct hclge_desc));
+
+	/* handle all PF RAS errors */
+	ret = hclge_handle_pf_ras_error(hdev, desc, pf_bd_num);
+	kfree(desc);
+
+	return ret;
+}
+
+static int hclge_log_rocee_axi_error(struct hclge_dev *hdev)
+{
+	struct device *dev = &hdev->pdev->dev;
+	struct hclge_desc desc[3];
+	int ret;
+
+	hclge_cmd_setup_basic_desc(&desc[0], HCLGE_QUERY_ROCEE_AXI_RAS_INFO_CMD,
+				   true);
+	hclge_cmd_setup_basic_desc(&desc[1], HCLGE_QUERY_ROCEE_AXI_RAS_INFO_CMD,
+				   true);
+	hclge_cmd_setup_basic_desc(&desc[2], HCLGE_QUERY_ROCEE_AXI_RAS_INFO_CMD,
+				   true);
+	desc[0].flag |= cpu_to_le16(HCLGE_COMM_CMD_FLAG_NEXT);
+	desc[1].flag |= cpu_to_le16(HCLGE_COMM_CMD_FLAG_NEXT);
+
+	ret = hclge_cmd_send(&hdev->hw, &desc[0], 3);
+	if (ret) {
+		dev_err(dev, "failed(%d) to query ROCEE AXI error sts\n", ret);
+		return ret;
+	}
+
+	dev_err(dev, "AXI1: %08X %08X %08X %08X %08X %08X\n",
+		le32_to_cpu(desc[0].data[0]), le32_to_cpu(desc[0].data[1]),
+		le32_to_cpu(desc[0].data[2]), le32_to_cpu(desc[0].data[3]),
+		le32_to_cpu(desc[0].data[4]), le32_to_cpu(desc[0].data[5]));
+	dev_err(dev, "AXI2: %08X %08X %08X %08X %08X %08X\n",
+		le32_to_cpu(desc[1].data[0]), le32_to_cpu(desc[1].data[1]),
+		le32_to_cpu(desc[1].data[2]), le32_to_cpu(desc[1].data[3]),
+		le32_to_cpu(desc[1].data[4]), le32_to_cpu(desc[1].data[5]));
+	dev_err(dev, "AXI3: %08X %08X %08X %08X\n",
+		le32_to_cpu(desc[2].data[0]), le32_to_cpu(desc[2].data[1]),
+		le32_to_cpu(desc[2].data[2]), le32_to_cpu(desc[2].data[3]));
+
+	return 0;
+}
+
+static int hclge_log_rocee_ecc_error(struct hclge_dev *hdev)
+{
+	struct device *dev = &hdev->pdev->dev;
+	struct hclge_desc desc[2];
+	int ret;
+
+	ret = hclge_cmd_query_error(hdev, &desc[0],
+				    HCLGE_QUERY_ROCEE_ECC_RAS_INFO_CMD,
+				    HCLGE_COMM_CMD_FLAG_NEXT);
+	if (ret) {
+		dev_err(dev, "failed(%d) to query ROCEE ECC error sts\n", ret);
+		return ret;
+	}
+
+	dev_err(dev, "ECC1: %08X %08X %08X %08X %08X %08X\n",
+		le32_to_cpu(desc[0].data[0]), le32_to_cpu(desc[0].data[1]),
+		le32_to_cpu(desc[0].data[2]), le32_to_cpu(desc[0].data[3]),
+		le32_to_cpu(desc[0].data[4]), le32_to_cpu(desc[0].data[5]));
+	dev_err(dev, "ECC2: %08X %08X %08X\n", le32_to_cpu(desc[1].data[0]),
+		le32_to_cpu(desc[1].data[1]), le32_to_cpu(desc[1].data[2]));
+
+	return 0;
+}
+
+static int hclge_log_rocee_ovf_error(struct hclge_dev *hdev)
+{
+	struct device *dev = &hdev->pdev->dev;
+	struct hclge_desc desc[2];
+	int ret;
+
+	/* read overflow error status */
+	ret = hclge_cmd_query_error(hdev, &desc[0], HCLGE_ROCEE_PF_RAS_INT_CMD,
+				    0);
+	if (ret) {
+		dev_err(dev, "failed(%d) to query ROCEE OVF error sts\n", ret);
+		return ret;
+	}
+
+	/* log overflow error */
+	if (le32_to_cpu(desc[0].data[0]) & HCLGE_ROCEE_OVF_ERR_INT_MASK) {
+		const struct hclge_hw_error *err;
+		u32 err_sts;
+
+		err = &hclge_rocee_qmm_ovf_err_int[0];
+		err_sts = HCLGE_ROCEE_OVF_ERR_TYPE_MASK &
+			  le32_to_cpu(desc[0].data[0]);
+		while (err->msg) {
+			if (err->int_msk == err_sts) {
+				dev_err(dev, "%s [error status=0x%x] found\n",
+					err->msg,
+					le32_to_cpu(desc[0].data[0]));
+				break;
+			}
+			err++;
+		}
+	}
+
+	if (le32_to_cpu(desc[0].data[1]) & HCLGE_ROCEE_OVF_ERR_INT_MASK) {
+		dev_err(dev, "ROCEE TSP OVF [error status=0x%x] found\n",
+			le32_to_cpu(desc[0].data[1]));
+	}
+
+	if (le32_to_cpu(desc[0].data[2]) & HCLGE_ROCEE_OVF_ERR_INT_MASK) {
+		dev_err(dev, "ROCEE SCC OVF [error status=0x%x] found\n",
+			le32_to_cpu(desc[0].data[2]));
+	}
+
+	return 0;
+}
+
+static enum hnae3_reset_type
+hclge_log_and_clear_rocee_ras_error(struct hclge_dev *hdev)
+{
+	enum hnae3_reset_type reset_type = HNAE3_NONE_RESET;
+	struct device *dev = &hdev->pdev->dev;
+	struct hclge_desc desc[2];
+	unsigned int status;
+	int ret;
+
+	/* read RAS error interrupt status */
+	ret = hclge_cmd_query_error(hdev, &desc[0],
+				    HCLGE_QUERY_CLEAR_ROCEE_RAS_INT, 0);
+	if (ret) {
+		dev_err(dev, "failed(%d) to query ROCEE RAS INT SRC\n", ret);
+		/* reset everything for now */
+		return HNAE3_GLOBAL_RESET;
+	}
+
+	status = le32_to_cpu(desc[0].data[0]);
+	if (status & HCLGE_ROCEE_AXI_ERR_INT_MASK) {
+		if (status & HCLGE_ROCEE_RERR_INT_MASK)
+			dev_err(dev, "ROCEE RAS AXI rresp error\n");
+
+		if (status & HCLGE_ROCEE_BERR_INT_MASK)
+			dev_err(dev, "ROCEE RAS AXI bresp error\n");
+
+		reset_type = HNAE3_FUNC_RESET;
+
+		hclge_report_hw_error(hdev, HNAE3_ROCEE_AXI_RESP_ERROR);
+
+		ret = hclge_log_rocee_axi_error(hdev);
+		if (ret)
+			return HNAE3_GLOBAL_RESET;
+	}
+
+	if (status & HCLGE_ROCEE_ECC_INT_MASK) {
+		dev_err(dev, "ROCEE RAS 2bit ECC error\n");
+		reset_type = HNAE3_GLOBAL_RESET;
+
+		ret = hclge_log_rocee_ecc_error(hdev);
+		if (ret)
+			return HNAE3_GLOBAL_RESET;
+	}
+
+	if (status & HCLGE_ROCEE_OVF_INT_MASK) {
+		ret = hclge_log_rocee_ovf_error(hdev);
+		if (ret) {
+			dev_err(dev, "failed(%d) to process ovf error\n", ret);
+			/* reset everything for now */
+			return HNAE3_GLOBAL_RESET;
+		}
+	}
+
+	/* clear error status */
+	hclge_comm_cmd_reuse_desc(&desc[0], false);
+	ret = hclge_cmd_send(&hdev->hw, &desc[0], 1);
+	if (ret) {
+		dev_err(dev, "failed(%d) to clear ROCEE RAS error\n", ret);
+		/* reset everything for now */
+		return HNAE3_GLOBAL_RESET;
+	}
+
+	return reset_type;
+}
+
+int hclge_config_rocee_ras_interrupt(struct hclge_dev *hdev, bool en)
+{
+	struct device *dev = &hdev->pdev->dev;
+	struct hclge_desc desc;
+	int ret;
+
+	if (hdev->ae_dev->dev_version < HNAE3_DEVICE_VERSION_V2 ||
+	    !hnae3_dev_roce_supported(hdev))
+		return 0;
+
+	hclge_cmd_setup_basic_desc(&desc, HCLGE_CONFIG_ROCEE_RAS_INT_EN, false);
+	if (en) {
+		/* enable ROCEE hw error interrupts */
+		desc.data[0] = cpu_to_le32(HCLGE_ROCEE_RAS_NFE_INT_EN);
+		desc.data[1] = cpu_to_le32(HCLGE_ROCEE_RAS_CE_INT_EN);
+
+		hclge_log_and_clear_rocee_ras_error(hdev);
+	}
+	desc.data[2] = cpu_to_le32(HCLGE_ROCEE_RAS_NFE_INT_EN_MASK);
+	desc.data[3] = cpu_to_le32(HCLGE_ROCEE_RAS_CE_INT_EN_MASK);
+
+	ret = hclge_cmd_send(&hdev->hw, &desc, 1);
+	if (ret)
+		dev_err(dev, "failed(%d) to config ROCEE RAS interrupt\n", ret);
+
+	return ret;
+}
+
+static void hclge_handle_rocee_ras_error(struct hnae3_ae_dev *ae_dev)
+{
+	struct hclge_dev *hdev = ae_dev->priv;
+	enum hnae3_reset_type reset_type;
+
+	if (test_bit(HCLGE_STATE_RST_HANDLING, &hdev->state))
+		return;
+
+	reset_type = hclge_log_and_clear_rocee_ras_error(hdev);
+	if (reset_type != HNAE3_NONE_RESET)
+		set_bit(reset_type, &ae_dev->hw_err_reset_req);
+}
+
+static const struct hclge_hw_blk hw_blk[] = {
+	{
+		.msk = BIT(0),
+		.name = "IGU_EGU",
+		.config_err_int = hclge_config_igu_egu_hw_err_int,
+	}, {
+		.msk = BIT(1),
+		.name = "PPP",
+		.config_err_int = hclge_config_ppp_hw_err_int,
+	}, {
+		.msk = BIT(2),
+		.name = "SSU",
+		.config_err_int = hclge_config_ssu_hw_err_int,
+	}, {
+		.msk = BIT(3),
+		.name = "PPU",
+		.config_err_int = hclge_config_ppu_hw_err_int,
+	}, {
+		.msk = BIT(4),
+		.name = "TM",
+		.config_err_int = hclge_config_tm_hw_err_int,
+	}, {
+		.msk = BIT(5),
+		.name = "COMMON",
+		.config_err_int = hclge_config_common_hw_err_int,
+	}, {
+		.msk = BIT(8),
+		.name = "MAC",
+		.config_err_int = hclge_config_mac_err_int,
+	}, {
+		/* sentinel */
+	}
+};
+
+static void hclge_config_all_msix_error(struct hclge_dev *hdev, bool enable)
+{
+	u32 reg_val;
+
+	reg_val = hclge_read_dev(&hdev->hw, HCLGE_PF_OTHER_INT_REG);
+
+	if (enable)
+		reg_val |= BIT(HCLGE_VECTOR0_ALL_MSIX_ERR_B);
+	else
+		reg_val &= ~BIT(HCLGE_VECTOR0_ALL_MSIX_ERR_B);
+
+	hclge_write_dev(&hdev->hw, HCLGE_PF_OTHER_INT_REG, reg_val);
+}
+
+int hclge_config_nic_hw_error(struct hclge_dev *hdev, bool state)
+{
+	const struct hclge_hw_blk *module = hw_blk;
+	int ret = 0;
+
+	hclge_config_all_msix_error(hdev, state);
+
+	while (module->name) {
+		if (module->config_err_int) {
+			ret = module->config_err_int(hdev, state);
+			if (ret)
+				return ret;
+		}
+		module++;
+	}
+
+	return ret;
+}
+
+pci_ers_result_t hclge_handle_hw_ras_error(struct hnae3_ae_dev *ae_dev)
+{
+	struct hclge_dev *hdev = ae_dev->priv;
+	struct device *dev = &hdev->pdev->dev;
+	u32 status;
+
+	if (!test_bit(HCLGE_STATE_SERVICE_INITED, &hdev->state)) {
+		dev_err(dev,
+			"Can't recover - RAS error reported during dev init\n");
+		return PCI_ERS_RESULT_NONE;
+	}
+
+	status = hclge_read_dev(&hdev->hw, HCLGE_RAS_PF_OTHER_INT_STS_REG);
+	if (status & HCLGE_RAS_REG_NFE_MASK ||
+	    status & HCLGE_RAS_REG_ROCEE_ERR_MASK)
+		ae_dev->hw_err_reset_req = 0;
+	else
+		goto out;
+
+	/* Handling Non-fatal HNS RAS errors */
+	if (status & HCLGE_RAS_REG_NFE_MASK) {
+		dev_err(dev,
+			"HNS Non-Fatal RAS error(status=0x%x) identified\n",
+			status);
+		hclge_handle_all_ras_errors(hdev);
+	}
+
+	/* Handling Non-fatal Rocee RAS errors */
+	if (hdev->ae_dev->dev_version >= HNAE3_DEVICE_VERSION_V2 &&
+	    status & HCLGE_RAS_REG_ROCEE_ERR_MASK) {
+		dev_err(dev, "ROCEE Non-Fatal RAS error identified\n");
+		hclge_handle_rocee_ras_error(ae_dev);
+	}
+
+	if (ae_dev->hw_err_reset_req)
+		return PCI_ERS_RESULT_NEED_RESET;
+
+out:
+	return PCI_ERS_RESULT_RECOVERED;
+}
+
+static int hclge_clear_hw_msix_error(struct hclge_dev *hdev,
+				     struct hclge_desc *desc, bool is_mpf,
+				     u32 bd_num)
+{
+	if (is_mpf)
+		desc[0].opcode =
+			cpu_to_le16(HCLGE_QUERY_CLEAR_ALL_MPF_MSIX_INT);
+	else
+		desc[0].opcode = cpu_to_le16(HCLGE_QUERY_CLEAR_ALL_PF_MSIX_INT);
+
+	desc[0].flag = cpu_to_le16(HCLGE_COMM_CMD_FLAG_NO_INTR |
+				   HCLGE_COMM_CMD_FLAG_IN);
+
+	return hclge_cmd_send(&hdev->hw, &desc[0], bd_num);
+}
+
+/* hclge_query_8bd_info: query information about over_8bd_nfe_err
+ * @hdev: pointer to struct hclge_dev
+ * @vf_id: Index of the virtual function with error
+ * @q_id: Physical index of the queue with error
+ *
+ * This function get specific index of queue and function which causes
+ * over_8bd_nfe_err by using command. If vf_id is 0, it means error is
+ * caused by PF instead of VF.
+ */
+static int hclge_query_over_8bd_err_info(struct hclge_dev *hdev, u16 *vf_id,
+					 u16 *q_id)
+{
+	struct hclge_query_ppu_pf_other_int_dfx_cmd *req;
+	struct hclge_desc desc;
+	int ret;
+
+	hclge_cmd_setup_basic_desc(&desc, HCLGE_OPC_PPU_PF_OTHER_INT_DFX, true);
+	ret = hclge_cmd_send(&hdev->hw, &desc, 1);
+	if (ret)
+		return ret;
+
+	req = (struct hclge_query_ppu_pf_other_int_dfx_cmd *)desc.data;
+	*vf_id = le16_to_cpu(req->over_8bd_no_fe_vf_id);
+	*q_id = le16_to_cpu(req->over_8bd_no_fe_qid);
+
+	return 0;
+}
+
+/* hclge_handle_over_8bd_err: handle MSI-X error named over_8bd_nfe_err
+ * @hdev: pointer to struct hclge_dev
+ * @reset_requests: reset level that we need to trigger later
+ *
+ * over_8bd_nfe_err is a special MSI-X because it may caused by a VF, in
+ * that case, we need to trigger VF reset. Otherwise, a PF reset is needed.
+ */
+static void hclge_handle_over_8bd_err(struct hclge_dev *hdev,
+				      unsigned long *reset_requests)
+{
+	struct device *dev = &hdev->pdev->dev;
+	u16 vf_id;
+	u16 q_id;
+	int ret;
+
+	ret = hclge_query_over_8bd_err_info(hdev, &vf_id, &q_id);
+	if (ret) {
+		dev_err(dev, "fail(%d) to query over_8bd_no_fe info\n",
+			ret);
+		return;
+	}
+
+	dev_err(dev, "PPU_PF_ABNORMAL_INT_ST over_8bd_no_fe found, vport(%u), queue_id(%u)\n",
+		vf_id, q_id);
+
+	if (vf_id) {
+		if (vf_id >= hdev->num_alloc_vport) {
+			dev_err(dev, "invalid vport(%u)\n", vf_id);
+			return;
+		}
+
+		/* If we need to trigger other reset whose level is higher
+		 * than HNAE3_VF_FUNC_RESET, no need to trigger a VF reset
+		 * here.
+		 */
+		if (*reset_requests != 0)
+			return;
+
+		ret = hclge_inform_reset_assert_to_vf(&hdev->vport[vf_id]);
+		if (ret)
+			dev_err(dev, "inform reset to vport(%u) failed %d!\n",
+				vf_id, ret);
+	} else {
+		set_bit(HNAE3_FUNC_RESET, reset_requests);
+	}
+}
+
+/* hclge_handle_mpf_msix_error: handle all main PF MSI-X errors
+ * @hdev: pointer to struct hclge_dev
+ * @desc: descriptor for describing the command
+ * @mpf_bd_num: number of extended command structures
+ * @reset_requests: record of the reset level that we need
+ *
+ * This function handles all the main PF MSI-X errors in the hw register/s
+ * using command.
+ */
+static int hclge_handle_mpf_msix_error(struct hclge_dev *hdev,
+				       struct hclge_desc *desc,
+				       int mpf_bd_num,
+				       unsigned long *reset_requests)
+{
+	struct device *dev = &hdev->pdev->dev;
+	__le32 *desc_data;
+	u32 status;
+	int ret;
+	/* query all main PF MSIx errors */
+	hclge_cmd_setup_basic_desc(&desc[0], HCLGE_QUERY_CLEAR_ALL_MPF_MSIX_INT,
+				   true);
+	ret = hclge_cmd_send(&hdev->hw, &desc[0], mpf_bd_num);
+	if (ret) {
+		dev_err(dev, "query all mpf msix int cmd failed (%d)\n", ret);
+		return ret;
+	}
+
+	/* log MAC errors */
+	desc_data = (__le32 *)&desc[1];
+	status = le32_to_cpu(*desc_data);
+	if (status)
+		hclge_log_error(dev, "MAC_AFIFO_TNL_INT_R",
+				&hclge_mac_afifo_tnl_int[0], status,
+				reset_requests);
+
+	/* log PPU(RCB) MPF errors */
+	desc_data = (__le32 *)&desc[5];
+	status = le32_to_cpu(*(desc_data + 2)) &
+			HCLGE_PPU_MPF_INT_ST2_MSIX_MASK;
+	if (status)
+		dev_err(dev, "PPU_MPF_ABNORMAL_INT_ST2 rx_q_search_miss found [dfx status=0x%x\n]",
+			status);
+
+	/* clear all main PF MSIx errors */
+	ret = hclge_clear_hw_msix_error(hdev, desc, true, mpf_bd_num);
+	if (ret)
+		dev_err(dev, "clear all mpf msix int cmd failed (%d)\n", ret);
+
+	return ret;
+}
+
+/* hclge_handle_pf_msix_error: handle all PF MSI-X errors
+ * @hdev: pointer to struct hclge_dev
+ * @desc: descriptor for describing the command
+ * @mpf_bd_num: number of extended command structures
+ * @reset_requests: record of the reset level that we need
+ *
+ * This function handles all the PF MSI-X errors in the hw register/s using
+ * command.
+ */
+static int hclge_handle_pf_msix_error(struct hclge_dev *hdev,
+				      struct hclge_desc *desc,
+				      int pf_bd_num,
+				      unsigned long *reset_requests)
+{
+	struct device *dev = &hdev->pdev->dev;
+	__le32 *desc_data;
+	u32 status;
+	int ret;
+
+	/* query all PF MSIx errors */
+	hclge_cmd_setup_basic_desc(&desc[0], HCLGE_QUERY_CLEAR_ALL_PF_MSIX_INT,
+				   true);
+	ret = hclge_cmd_send(&hdev->hw, &desc[0], pf_bd_num);
+	if (ret) {
+		dev_err(dev, "query all pf msix int cmd failed (%d)\n", ret);
+		return ret;
+	}
+
+	/* log SSU PF errors */
+	status = le32_to_cpu(desc[0].data[0]) & HCLGE_SSU_PORT_INT_MSIX_MASK;
+	if (status)
+		hclge_log_error(dev, "SSU_PORT_BASED_ERR_INT",
+				&hclge_ssu_port_based_pf_int[0],
+				status, reset_requests);
+
+	/* read and log PPP PF errors */
+	desc_data = (__le32 *)&desc[2];
+	status = le32_to_cpu(*desc_data);
+	if (status)
+		hclge_log_error(dev, "PPP_PF_ABNORMAL_INT_ST0",
+				&hclge_ppp_pf_abnormal_int[0],
+				status, reset_requests);
+
+	/* log PPU(RCB) PF errors */
+	desc_data = (__le32 *)&desc[3];
+	status = le32_to_cpu(*desc_data) & HCLGE_PPU_PF_INT_MSIX_MASK;
+	if (status)
+		hclge_log_error(dev, "PPU_PF_ABNORMAL_INT_ST",
+				&hclge_ppu_pf_abnormal_int[0],
+				status, reset_requests);
+
+	status = le32_to_cpu(*desc_data) & HCLGE_PPU_PF_OVER_8BD_ERR_MASK;
+	if (status)
+		hclge_handle_over_8bd_err(hdev, reset_requests);
+
+	/* clear all PF MSIx errors */
+	ret = hclge_clear_hw_msix_error(hdev, desc, false, pf_bd_num);
+	if (ret)
+		dev_err(dev, "clear all pf msix int cmd failed (%d)\n", ret);
+
+	return ret;
+}
+
+static int hclge_handle_all_hw_msix_error(struct hclge_dev *hdev,
+					  unsigned long *reset_requests)
+{
+	u32 mpf_bd_num, pf_bd_num, bd_num;
+	struct hclge_desc *desc;
+	int ret;
+
+	/* query the number of bds for the MSIx int status */
+	ret = hclge_query_bd_num(hdev, false, &mpf_bd_num, &pf_bd_num);
+	if (ret)
+		goto out;
+
+	bd_num = max_t(u32, mpf_bd_num, pf_bd_num);
+	desc = kcalloc(bd_num, sizeof(struct hclge_desc), GFP_KERNEL);
+	if (!desc)
+		return -ENOMEM;
+
+	ret = hclge_handle_mpf_msix_error(hdev, desc, mpf_bd_num,
+					  reset_requests);
+	if (ret)
+		goto msi_error;
+
+	memset(desc, 0, bd_num * sizeof(struct hclge_desc));
+	ret = hclge_handle_pf_msix_error(hdev, desc, pf_bd_num, reset_requests);
+	if (ret)
+		goto msi_error;
+
+	ret = hclge_handle_mac_tnl(hdev);
+
+msi_error:
+	kfree(desc);
+out:
+	return ret;
+}
+
+int hclge_handle_hw_msix_error(struct hclge_dev *hdev,
+			       unsigned long *reset_requests)
+{
+	struct device *dev = &hdev->pdev->dev;
+
+	if (!test_bit(HCLGE_STATE_SERVICE_INITED, &hdev->state)) {
+		dev_err(dev,
+			"failed to handle msix error during dev init\n");
+		return -EAGAIN;
+	}
+
+	return hclge_handle_all_hw_msix_error(hdev, reset_requests);
+}
+
+int hclge_handle_mac_tnl(struct hclge_dev *hdev)
+{
+	struct hclge_mac_tnl_stats mac_tnl_stats;
+	struct device *dev = &hdev->pdev->dev;
+	struct hclge_desc desc;
+	u32 status;
+	int ret;
+
+	/* query and clear mac tnl interruptions */
+	hclge_cmd_setup_basic_desc(&desc, HCLGE_OPC_QUERY_MAC_TNL_INT, true);
+	ret = hclge_cmd_send(&hdev->hw, &desc, 1);
+	if (ret) {
+		dev_err(dev, "failed to query mac tnl int, ret = %d.\n", ret);
+		return ret;
+	}
+
+	status = le32_to_cpu(desc.data[0]);
+	if (status) {
+		/* When mac tnl interrupt occurs, we record current time and
+		 * register status here in a fifo, then clear the status. So
+		 * that if link status changes suddenly at some time, we can
+		 * query them by debugfs.
+		 */
+		mac_tnl_stats.time = local_clock();
+		mac_tnl_stats.status = status;
+		kfifo_put(&hdev->mac_tnl_log, mac_tnl_stats);
+		ret = hclge_clear_mac_tnl_int(hdev);
+		if (ret)
+			dev_err(dev, "failed to clear mac tnl int, ret = %d.\n",
+				ret);
+	}
+
+	return ret;
+}
+
+void hclge_handle_all_hns_hw_errors(struct hnae3_ae_dev *ae_dev)
+{
+	struct hclge_dev *hdev = ae_dev->priv;
+	struct device *dev = &hdev->pdev->dev;
+	u32 mpf_bd_num, pf_bd_num, bd_num;
+	struct hclge_desc *desc;
+	u32 status;
+	int ret;
+
+	ae_dev->hw_err_reset_req = 0;
+	status = hclge_read_dev(&hdev->hw, HCLGE_RAS_PF_OTHER_INT_STS_REG);
+
+	/* query the number of bds for the MSIx int status */
+	ret = hclge_query_bd_num(hdev, false, &mpf_bd_num, &pf_bd_num);
+	if (ret)
+		return;
+
+	bd_num = max_t(u32, mpf_bd_num, pf_bd_num);
+	desc = kcalloc(bd_num, sizeof(struct hclge_desc), GFP_KERNEL);
+	if (!desc)
+		return;
+
+	/* Clear HNS hw errors reported through msix  */
+	memset(&desc[0].data[0], 0xFF, mpf_bd_num * sizeof(struct hclge_desc) -
+	       HCLGE_DESC_NO_DATA_LEN);
+	ret = hclge_clear_hw_msix_error(hdev, desc, true, mpf_bd_num);
+	if (ret) {
+		dev_err(dev, "fail(%d) to clear mpf msix int during init\n",
+			ret);
+		goto msi_error;
+	}
+
+	memset(&desc[0].data[0], 0xFF, pf_bd_num * sizeof(struct hclge_desc) -
+	       HCLGE_DESC_NO_DATA_LEN);
+	ret = hclge_clear_hw_msix_error(hdev, desc, false, pf_bd_num);
+	if (ret) {
+		dev_err(dev, "fail(%d) to clear pf msix int during init\n",
+			ret);
+		goto msi_error;
+	}
+
+	/* Handle Non-fatal HNS RAS errors */
+	if (status & HCLGE_RAS_REG_NFE_MASK) {
+		dev_err(dev, "HNS hw error(RAS) identified during init\n");
+		hclge_handle_all_ras_errors(hdev);
+	}
+
+msi_error:
+	kfree(desc);
+}
+
+bool hclge_find_error_source(struct hclge_dev *hdev)
+{
+	u32 msix_src_flag, hw_err_src_flag;
+
+	msix_src_flag = hclge_read_dev(&hdev->hw, HCLGE_MISC_VECTOR_INT_STS) &
+			HCLGE_VECTOR0_REG_MSIX_MASK;
+
+	hw_err_src_flag = hclge_read_dev(&hdev->hw,
+					 HCLGE_RAS_PF_OTHER_INT_STS_REG) &
+			  HCLGE_RAS_REG_ERR_MASK;
+
+	return msix_src_flag || hw_err_src_flag;
+}
+
+void hclge_handle_occurred_error(struct hclge_dev *hdev)
+{
+	struct hnae3_ae_dev *ae_dev = pci_get_drvdata(hdev->pdev);
+
+	if (hclge_find_error_source(hdev))
+		hclge_handle_error_info_log(ae_dev);
+}
+
+static bool
+hclge_handle_error_type_reg_log(struct hclge_dev *hdev,
+				struct hclge_mod_err_info *mod_info,
+				struct hclge_type_reg_err_info *type_reg_info)
+{
+#define HCLGE_ERR_TYPE_MASK 0x7F
+#define HCLGE_ERR_TYPE_IS_RAS_OFFSET 7
+
+	u8 mod_id, total_module, type_id, total_type, i, is_ras;
+	struct device *dev = &hdev->pdev->dev;
+	u8 index_module = MODULE_NONE;
+	u8 index_type = NONE_ERROR;
+	bool cause_by_vf = false;
+
+	mod_id = mod_info->mod_id;
+	type_id = type_reg_info->type_id & HCLGE_ERR_TYPE_MASK;
+	is_ras = type_reg_info->type_id >> HCLGE_ERR_TYPE_IS_RAS_OFFSET;
+
+	total_module = ARRAY_SIZE(hclge_hw_module_id_st);
+	total_type = ARRAY_SIZE(hclge_hw_type_id_st);
+
+	for (i = 0; i < total_module; i++) {
+		if (mod_id == hclge_hw_module_id_st[i].module_id) {
+			index_module = i;
+			break;
+		}
+	}
+
+	for (i = 0; i < total_type; i++) {
+		if (type_id == hclge_hw_type_id_st[i].type_id) {
+			index_type = i;
+			cause_by_vf = hclge_hw_type_id_st[i].cause_by_vf;
+			break;
+		}
+	}
+
+	if (index_module != MODULE_NONE && index_type != NONE_ERROR)
+		dev_err(dev,
+			"found %s %s, is %s error.\n",
+			hclge_hw_module_id_st[index_module].msg,
+			hclge_hw_type_id_st[index_type].msg,
+			is_ras ? "ras" : "msix");
+	else
+		dev_err(dev,
+			"unknown module[%u] or type[%u].\n", mod_id, type_id);
+
+	dev_err(dev, "reg_value:\n");
+	for (i = 0; i < type_reg_info->reg_num; i++)
+		dev_err(dev, "0x%08x\n", type_reg_info->hclge_reg[i]);
+
+	if (hclge_hw_module_id_st[index_module].query_reg_info)
+		hclge_hw_module_id_st[index_module].query_reg_info(hdev);
+
+	return cause_by_vf;
+}
+
+static void hclge_handle_error_module_log(struct hnae3_ae_dev *ae_dev,
+					  const u32 *buf, u32 buf_size)
+{
+	struct hclge_type_reg_err_info *type_reg_info;
+	struct hclge_dev *hdev = ae_dev->priv;
+	struct device *dev = &hdev->pdev->dev;
+	struct hclge_mod_err_info *mod_info;
+	struct hclge_sum_err_info *sum_info;
+	bool cause_by_vf = false;
+	u8 mod_num, err_num, i;
+	u32 offset = 0;
+
+	sum_info = (struct hclge_sum_err_info *)&buf[offset++];
+	if (sum_info->reset_type &&
+	    sum_info->reset_type != HNAE3_NONE_RESET)
+		set_bit(sum_info->reset_type, &ae_dev->hw_err_reset_req);
+	mod_num = sum_info->mod_num;
+
+	while (mod_num--) {
+		if (offset >= buf_size) {
+			dev_err(dev, "The offset(%u) exceeds buf's size(%u).\n",
+				offset, buf_size);
+			return;
+		}
+		mod_info = (struct hclge_mod_err_info *)&buf[offset++];
+		err_num = mod_info->err_num;
+
+		for (i = 0; i < err_num; i++) {
+			if (offset >= buf_size) {
+				dev_err(dev,
+					"The offset(%u) exceeds buf size(%u).\n",
+					offset, buf_size);
+				return;
+			}
+
+			type_reg_info = (struct hclge_type_reg_err_info *)
+					    &buf[offset++];
+			if (hclge_handle_error_type_reg_log(hdev, mod_info,
+							    type_reg_info))
+				cause_by_vf = true;
+
+			offset += type_reg_info->reg_num;
+		}
+	}
+
+	if (hnae3_ae_dev_vf_fault_supported(hdev->ae_dev) && cause_by_vf)
+		set_bit(HNAE3_VF_EXP_RESET, &ae_dev->hw_err_reset_req);
+}
+
+static int hclge_query_all_err_bd_num(struct hclge_dev *hdev, u32 *bd_num)
+{
+	struct device *dev = &hdev->pdev->dev;
+	struct hclge_desc desc_bd;
+	int ret;
+
+	hclge_cmd_setup_basic_desc(&desc_bd, HCLGE_QUERY_ALL_ERR_BD_NUM, true);
+	ret = hclge_cmd_send(&hdev->hw, &desc_bd, 1);
+	if (ret) {
+		dev_err(dev, "failed to query error bd_num, ret = %d.\n", ret);
+		return ret;
+	}
+
+	*bd_num = le32_to_cpu(desc_bd.data[0]);
+	if (!(*bd_num)) {
+		dev_err(dev, "The value of bd_num is 0!\n");
+		return -EINVAL;
+	}
+
+	return 0;
+}
+
+static int hclge_query_all_err_info(struct hclge_dev *hdev,
+				    struct hclge_desc *desc, u32 bd_num)
+{
+	struct device *dev = &hdev->pdev->dev;
+	int ret;
+
+	hclge_cmd_setup_basic_desc(desc, HCLGE_QUERY_ALL_ERR_INFO, true);
+	ret = hclge_cmd_send(&hdev->hw, desc, bd_num);
+	if (ret)
+		dev_err(dev, "failed to query error info, ret = %d.\n", ret);
+
+	return ret;
+}
+
+int hclge_handle_error_info_log(struct hnae3_ae_dev *ae_dev)
+{
+	u32 bd_num, desc_len, buf_len, buf_size, i;
+	struct hclge_dev *hdev = ae_dev->priv;
+	struct hclge_desc *desc;
+	__le32 *desc_data;
+	u32 *buf;
+	int ret;
+
+	ret = hclge_query_all_err_bd_num(hdev, &bd_num);
+	if (ret)
+		goto out;
+
+	desc_len = bd_num * sizeof(struct hclge_desc);
+	desc = kzalloc(desc_len, GFP_KERNEL);
+	if (!desc) {
+		ret = -ENOMEM;
+		goto out;
+	}
+
+	ret = hclge_query_all_err_info(hdev, desc, bd_num);
+	if (ret)
+		goto err_desc;
+
+	buf_len = bd_num * sizeof(struct hclge_desc) - HCLGE_DESC_NO_DATA_LEN;
+	buf_size = buf_len / sizeof(u32);
+
+	desc_data = kzalloc(buf_len, GFP_KERNEL);
+	if (!desc_data) {
+		ret = -ENOMEM;
+		goto err_desc;
+	}
+
+	buf = kzalloc(buf_len, GFP_KERNEL);
+	if (!buf) {
+		ret = -ENOMEM;
+		goto err_buf_alloc;
+	}
+
+	memcpy(desc_data, &desc[0].data[0], buf_len);
+	for (i = 0; i < buf_size; i++)
+		buf[i] = le32_to_cpu(desc_data[i]);
+
+	hclge_handle_error_module_log(ae_dev, buf, buf_size);
+	kfree(buf);
+
+err_buf_alloc:
+	kfree(desc_data);
+err_desc:
+	kfree(desc);
+out:
+	return ret;
+}
+
+static bool hclge_reset_vf_in_bitmap(struct hclge_dev *hdev,
+				     unsigned long *bitmap)
+{
+	struct hclge_vport *vport;
+	bool exist_set = false;
+	int func_id;
+	int ret;
+
+	func_id = find_first_bit(bitmap, HCLGE_VPORT_NUM);
+	if (func_id == PF_VPORT_ID)
+		return false;
+
+	while (func_id != HCLGE_VPORT_NUM) {
+		vport = hclge_get_vf_vport(hdev,
+					   func_id - HCLGE_VF_VPORT_START_NUM);
+		if (!vport) {
+			dev_err(&hdev->pdev->dev, "invalid func id(%d)\n",
+				func_id);
+			return false;
+		}
+
+		dev_info(&hdev->pdev->dev, "do function %d recovery.", func_id);
+
+		ret = hclge_reset_tqp(&vport->nic);
+		if (ret) {
+			dev_err(&hdev->pdev->dev,
+				"failed to reset tqp, ret = %d.", ret);
+			return false;
+		}
+
+		ret = hclge_inform_vf_reset(vport, HNAE3_VF_FUNC_RESET);
+		if (ret) {
+			dev_err(&hdev->pdev->dev,
+				"failed to reset func %d, ret = %d.",
+				func_id, ret);
+			return false;
+		}
+
+		exist_set = true;
+		clear_bit(func_id, bitmap);
+		func_id = find_first_bit(bitmap, HCLGE_VPORT_NUM);
+	}
+
+	return exist_set;
+}
+
+static void hclge_get_vf_fault_bitmap(struct hclge_desc *desc,
+				      unsigned long *bitmap)
+{
+#define HCLGE_FIR_FAULT_BYTES	24
+#define HCLGE_SEC_FAULT_BYTES	8
+
+	u8 *buff;
+
+	BUILD_BUG_ON(HCLGE_FIR_FAULT_BYTES + HCLGE_SEC_FAULT_BYTES !=
+		     BITS_TO_BYTES(HCLGE_VPORT_NUM));
+
+	memcpy(bitmap, desc[0].data, HCLGE_FIR_FAULT_BYTES);
+	buff = (u8 *)bitmap + HCLGE_FIR_FAULT_BYTES;
+	memcpy(buff, desc[1].data, HCLGE_SEC_FAULT_BYTES);
+}
+
+int hclge_handle_vf_queue_err_ras(struct hclge_dev *hdev)
+{
+	unsigned long vf_fault_bitmap[BITS_TO_LONGS(HCLGE_VPORT_NUM)];
+	struct hclge_desc desc[2];
+	bool cause_by_vf = false;
+	int ret;
+
+	if (!test_and_clear_bit(HNAE3_VF_EXP_RESET,
+				&hdev->ae_dev->hw_err_reset_req) ||
+	    !hnae3_ae_dev_vf_fault_supported(hdev->ae_dev))
+		return 0;
+
+	hclge_comm_cmd_setup_basic_desc(&desc[0], HCLGE_OPC_GET_QUEUE_ERR_VF,
+					true);
+	desc[0].flag |= cpu_to_le16(HCLGE_COMM_CMD_FLAG_NEXT);
+	hclge_comm_cmd_setup_basic_desc(&desc[1], HCLGE_OPC_GET_QUEUE_ERR_VF,
+					true);
+
+	ret = hclge_comm_cmd_send(&hdev->hw.hw, desc, 2);
+	if (ret) {
+		dev_err(&hdev->pdev->dev,
+			"failed to get vf bitmap, ret = %d.\n", ret);
+		return ret;
+	}
+	hclge_get_vf_fault_bitmap(desc, vf_fault_bitmap);
+
+	cause_by_vf = hclge_reset_vf_in_bitmap(hdev, vf_fault_bitmap);
+	if (cause_by_vf)
+		hdev->ae_dev->hw_err_reset_req = 0;
+
+	return 0;
+}
diff --git a/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_err.h b/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_err.h
new file mode 100644
index 000000000000..45a783a50643
--- /dev/null
+++ b/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_err.h
@@ -0,0 +1,269 @@
+/* SPDX-License-Identifier: GPL-2.0+ */
+/*  Copyright (c) 2016-2017 Hisilicon Limited. */
+
+#ifndef __HCLGE_ERR_H
+#define __HCLGE_ERR_H
+
+#include "hclge_main.h"
+#include "hclge_debugfs.h"
+#include "hnae3.h"
+
+#define HCLGE_MPF_RAS_INT_MIN_BD_NUM	10
+#define HCLGE_PF_RAS_INT_MIN_BD_NUM	4
+#define HCLGE_MPF_MSIX_INT_MIN_BD_NUM	10
+#define HCLGE_PF_MSIX_INT_MIN_BD_NUM	4
+
+#define HCLGE_RAS_PF_OTHER_INT_STS_REG   0x20B00
+#define HCLGE_RAS_REG_NFE_MASK   0xFF00
+#define HCLGE_RAS_REG_ROCEE_ERR_MASK   0x3000000
+#define HCLGE_RAS_REG_ERR_MASK \
+	(HCLGE_RAS_REG_NFE_MASK | HCLGE_RAS_REG_ROCEE_ERR_MASK)
+
+#define HCLGE_VECTOR0_REG_MSIX_MASK   0x1FF00
+
+#define HCLGE_IMP_TCM_ECC_ERR_INT_EN	0xFFFF0000
+#define HCLGE_IMP_TCM_ECC_ERR_INT_EN_MASK	0xFFFF0000
+#define HCLGE_IMP_ITCM4_ECC_ERR_INT_EN	0x300
+#define HCLGE_IMP_ITCM4_ECC_ERR_INT_EN_MASK	0x300
+#define HCLGE_CMDQ_NIC_ECC_ERR_INT_EN	0xFFFF
+#define HCLGE_CMDQ_NIC_ECC_ERR_INT_EN_MASK	0xFFFF
+#define HCLGE_CMDQ_ROCEE_ECC_ERR_INT_EN	0xFFFF0000
+#define HCLGE_CMDQ_ROCEE_ECC_ERR_INT_EN_MASK	0xFFFF0000
+#define HCLGE_IMP_RD_POISON_ERR_INT_EN	0x0100
+#define HCLGE_IMP_RD_POISON_ERR_INT_EN_MASK	0x0100
+#define HCLGE_TQP_ECC_ERR_INT_EN	0x0FFF
+#define HCLGE_TQP_ECC_ERR_INT_EN_MASK	0x0FFF
+#define HCLGE_MSIX_SRAM_ECC_ERR_INT_EN_MASK	0x0F000000
+#define HCLGE_MSIX_SRAM_ECC_ERR_INT_EN	0x0F000000
+#define HCLGE_IGU_ERR_INT_EN	0x0000000F
+#define HCLGE_IGU_ERR_INT_TYPE	0x00000660
+#define HCLGE_IGU_ERR_INT_EN_MASK	0x000F
+#define HCLGE_IGU_TNL_ERR_INT_EN    0x0002AABF
+#define HCLGE_IGU_TNL_ERR_INT_EN_MASK  0x003F
+#define HCLGE_PPP_MPF_ECC_ERR_INT0_EN	0xFFFFFFFF
+#define HCLGE_PPP_MPF_ECC_ERR_INT0_EN_MASK	0xFFFFFFFF
+#define HCLGE_PPP_MPF_ECC_ERR_INT1_EN	0xFFFFFFFF
+#define HCLGE_PPP_MPF_ECC_ERR_INT1_EN_MASK	0xFFFFFFFF
+#define HCLGE_PPP_PF_ERR_INT_EN	0x0003
+#define HCLGE_PPP_PF_ERR_INT_EN_MASK	0x0003
+#define HCLGE_PPP_MPF_ECC_ERR_INT2_EN	0x003F
+#define HCLGE_PPP_MPF_ECC_ERR_INT2_EN_MASK	0x003F
+#define HCLGE_PPP_MPF_ECC_ERR_INT3_EN	0x003F
+#define HCLGE_PPP_MPF_ECC_ERR_INT3_EN_MASK	0x003F
+#define HCLGE_TM_SCH_ECC_ERR_INT_EN	0x3
+#define HCLGE_TM_QCN_ERR_INT_TYPE	0x29
+#define HCLGE_TM_QCN_FIFO_INT_EN	0xFFFF00
+#define HCLGE_TM_QCN_MEM_ERR_INT_EN	0xFFFFFF
+#define HCLGE_NCSI_ERR_INT_EN	0x3
+#define HCLGE_NCSI_ERR_INT_TYPE	0x9
+#define HCLGE_MAC_COMMON_ERR_INT_EN		0x107FF
+#define HCLGE_MAC_COMMON_ERR_INT_EN_MASK	0x107FF
+#define HCLGE_MAC_TNL_INT_EN			GENMASK(9, 0)
+#define HCLGE_MAC_TNL_INT_EN_MASK		GENMASK(9, 0)
+#define HCLGE_MAC_TNL_INT_CLR			GENMASK(9, 0)
+#define HCLGE_PPU_MPF_ABNORMAL_INT0_EN		GENMASK(31, 0)
+#define HCLGE_PPU_MPF_ABNORMAL_INT0_EN_MASK	GENMASK(31, 0)
+#define HCLGE_PPU_MPF_ABNORMAL_INT1_EN		GENMASK(31, 0)
+#define HCLGE_PPU_MPF_ABNORMAL_INT1_EN_MASK	GENMASK(31, 0)
+#define HCLGE_PPU_MPF_ABNORMAL_INT2_EN		0x3FFF3FFF
+#define HCLGE_PPU_MPF_ABNORMAL_INT2_EN_MASK	0x3FFF3FFF
+#define HCLGE_PPU_MPF_ABNORMAL_INT2_EN2		0xB
+#define HCLGE_PPU_MPF_ABNORMAL_INT2_EN2_MASK	0xB
+#define HCLGE_PPU_MPF_ABNORMAL_INT3_EN		GENMASK(7, 0)
+#define HCLGE_PPU_MPF_ABNORMAL_INT3_EN_MASK	GENMASK(23, 16)
+#define HCLGE_PPU_PF_ABNORMAL_INT_EN		GENMASK(5, 0)
+#define HCLGE_PPU_PF_ABNORMAL_INT_EN_MASK	GENMASK(5, 0)
+#define HCLGE_SSU_1BIT_ECC_ERR_INT_EN		GENMASK(31, 0)
+#define HCLGE_SSU_1BIT_ECC_ERR_INT_EN_MASK	GENMASK(31, 0)
+#define HCLGE_SSU_MULTI_BIT_ECC_ERR_INT_EN	GENMASK(31, 0)
+#define HCLGE_SSU_MULTI_BIT_ECC_ERR_INT_EN_MASK	GENMASK(31, 0)
+#define HCLGE_SSU_BIT32_ECC_ERR_INT_EN		0x0101
+#define HCLGE_SSU_BIT32_ECC_ERR_INT_EN_MASK	0x0101
+#define HCLGE_SSU_COMMON_INT_EN			GENMASK(9, 0)
+#define HCLGE_SSU_COMMON_INT_EN_MASK		GENMASK(9, 0)
+#define HCLGE_SSU_PORT_BASED_ERR_INT_EN		0x0BFF
+#define HCLGE_SSU_PORT_BASED_ERR_INT_EN_MASK	0x0BFF0000
+#define HCLGE_SSU_FIFO_OVERFLOW_ERR_INT_EN	GENMASK(23, 0)
+#define HCLGE_SSU_FIFO_OVERFLOW_ERR_INT_EN_MASK	GENMASK(23, 0)
+
+#define HCLGE_SSU_COMMON_ERR_INT_MASK	GENMASK(9, 0)
+#define HCLGE_SSU_PORT_INT_MSIX_MASK	0x7BFF
+#define HCLGE_IGU_INT_MASK		GENMASK(3, 0)
+#define HCLGE_IGU_EGU_TNL_INT_MASK	GENMASK(5, 0)
+#define HCLGE_PPP_MPF_INT_ST3_MASK	GENMASK(5, 0)
+#define HCLGE_PPU_MPF_INT_ST3_MASK	GENMASK(7, 0)
+#define HCLGE_PPU_MPF_INT_ST2_MSIX_MASK	BIT(29)
+#define HCLGE_PPU_PF_INT_RAS_MASK	0x18
+#define HCLGE_PPU_PF_INT_MSIX_MASK	0x26
+#define HCLGE_PPU_PF_OVER_8BD_ERR_MASK	0x01
+#define HCLGE_QCN_FIFO_INT_MASK		GENMASK(17, 0)
+#define HCLGE_QCN_ECC_INT_MASK		GENMASK(21, 0)
+#define HCLGE_NCSI_ECC_INT_MASK		GENMASK(1, 0)
+
+#define HCLGE_ROCEE_RAS_NFE_INT_EN		0xF
+#define HCLGE_ROCEE_RAS_CE_INT_EN		0x1
+#define HCLGE_ROCEE_RAS_NFE_INT_EN_MASK		0xF
+#define HCLGE_ROCEE_RAS_CE_INT_EN_MASK		0x1
+#define HCLGE_ROCEE_RERR_INT_MASK		BIT(0)
+#define HCLGE_ROCEE_BERR_INT_MASK		BIT(1)
+#define HCLGE_ROCEE_AXI_ERR_INT_MASK		GENMASK(1, 0)
+#define HCLGE_ROCEE_ECC_INT_MASK		BIT(2)
+#define HCLGE_ROCEE_OVF_INT_MASK		BIT(3)
+#define HCLGE_ROCEE_OVF_ERR_INT_MASK		0x10000
+#define HCLGE_ROCEE_OVF_ERR_TYPE_MASK		0x3F
+
+#define HCLGE_DESC_DATA_MAX			8
+#define HCLGE_REG_NUM_MAX			256
+#define HCLGE_DESC_NO_DATA_LEN			8
+
+#define HCLGE_BD_NUM_SSU_REG_0		10
+#define HCLGE_BD_NUM_SSU_REG_1		15
+#define HCLGE_BD_NUM_RPU_REG_0		1
+#define HCLGE_BD_NUM_RPU_REG_1		2
+#define HCLGE_BD_NUM_IGU_EGU_REG	9
+#define HCLGE_BD_NUM_GEN_REG		8
+#define HCLGE_MOD_REG_INFO_LEN_MAX	256
+#define HCLGE_MOD_REG_EXTRA_LEN		11
+#define HCLGE_MOD_REG_VALUE_LEN		9
+#define HCLGE_MOD_REG_GROUP_MAX_SIZE	6
+#define HCLGE_MOD_MSG_PARA_ARRAY_MAX_SIZE	8
+
+enum hclge_err_int_type {
+	HCLGE_ERR_INT_MSIX = 0,
+	HCLGE_ERR_INT_RAS_CE = 1,
+	HCLGE_ERR_INT_RAS_NFE = 2,
+	HCLGE_ERR_INT_RAS_FE = 3,
+};
+
+enum hclge_mod_name_list {
+	MODULE_NONE		= 0,
+	MODULE_BIOS_COMMON	= 1,
+	MODULE_GE		= 2,
+	MODULE_IGU_EGU		= 3,
+	MODULE_LGE		= 4,
+	MODULE_NCSI		= 5,
+	MODULE_PPP		= 6,
+	MODULE_QCN		= 7,
+	MODULE_RCB_RX		= 8,
+	MODULE_RTC		= 9,
+	MODULE_SSU		= 10,
+	MODULE_TM		= 11,
+	MODULE_RCB_TX		= 12,
+	MODULE_TXDMA		= 13,
+	MODULE_MASTER		= 14,
+	MODULE_HIMAC		= 15,
+	/* add new MODULE NAME for NIC here in order */
+	MODULE_ROCEE_TOP	= 40,
+	MODULE_ROCEE_TIMER	= 41,
+	MODULE_ROCEE_MDB	= 42,
+	MODULE_ROCEE_TSP	= 43,
+	MODULE_ROCEE_TRP	= 44,
+	MODULE_ROCEE_SCC	= 45,
+	MODULE_ROCEE_CAEP	= 46,
+	MODULE_ROCEE_GEN_AC	= 47,
+	MODULE_ROCEE_QMM	= 48,
+	MODULE_ROCEE_LSAN	= 49,
+	/* add new MODULE NAME for RoCEE here in order */
+};
+
+enum hclge_err_type_list {
+	NONE_ERROR		= 0,
+	FIFO_ERROR		= 1,
+	MEMORY_ERROR		= 2,
+	POISON_ERROR		= 3,
+	MSIX_ECC_ERROR		= 4,
+	TQP_INT_ECC_ERROR	= 5,
+	PF_ABNORMAL_INT_ERROR	= 6,
+	MPF_ABNORMAL_INT_ERROR	= 7,
+	COMMON_ERROR		= 8,
+	PORT_ERROR		= 9,
+	ETS_ERROR		= 10,
+	NCSI_ERROR		= 11,
+	GLB_ERROR		= 12,
+	LINK_ERROR		= 13,
+	PTP_ERROR		= 14,
+	/* add new ERROR TYPE for NIC here in order */
+	ROCEE_NORMAL_ERR	= 40,
+	ROCEE_OVF_ERR		= 41,
+	ROCEE_BUS_ERR		= 42,
+	/* add new ERROR TYPE for ROCEE here in order */
+};
+
+struct hclge_hw_blk {
+	u32 msk;
+	const char *name;
+	int (*config_err_int)(struct hclge_dev *hdev, bool en);
+};
+
+struct hclge_hw_error {
+	u32 int_msk;
+	const char *msg;
+	enum hnae3_reset_type reset_level;
+};
+
+struct hclge_hw_module_id {
+	enum hclge_mod_name_list module_id;
+	const char *msg;
+	void (*query_reg_info)(struct hclge_dev *hdev);
+};
+
+struct hclge_hw_type_id {
+	enum hclge_err_type_list type_id;
+	const char *msg;
+	bool cause_by_vf; /* indicate the error may from vf exception */
+};
+
+struct hclge_sum_err_info {
+	u8 reset_type;
+	u8 mod_num;
+	u8 rsv[2];
+};
+
+struct hclge_mod_err_info {
+	u8 mod_id;
+	u8 err_num;
+	u8 rsv[2];
+};
+
+struct hclge_type_reg_err_info {
+	u8 type_id;
+	u8 reg_num;
+	u8 rsv[2];
+	u32 hclge_reg[HCLGE_REG_NUM_MAX];
+};
+
+struct hclge_mod_reg_info {
+	const char *reg_name;
+	bool has_suffix; /* add suffix for register name */
+	/* the positions of reg values in hclge_desc.data */
+	u8 reg_offset_group[HCLGE_MOD_REG_GROUP_MAX_SIZE];
+	u8 group_size;
+};
+
+/* This structure defines cmdq used to query the hardware module debug
+ * regisgers.
+ */
+struct hclge_mod_reg_common_msg {
+	enum hclge_opcode_type cmd;
+	struct hclge_desc *desc;
+	u8 bd_num; /* the bd number of hclge_desc used */
+	bool need_para; /* whether this cmdq needs to add para */
+
+	/* the regs need to print */
+	const struct hclge_mod_reg_info *result_regs;
+	u16 result_regs_size;
+};
+
+int hclge_config_mac_tnl_int(struct hclge_dev *hdev, bool en);
+int hclge_config_nic_hw_error(struct hclge_dev *hdev, bool state);
+int hclge_config_rocee_ras_interrupt(struct hclge_dev *hdev, bool en);
+void hclge_handle_all_hns_hw_errors(struct hnae3_ae_dev *ae_dev);
+bool hclge_find_error_source(struct hclge_dev *hdev);
+void hclge_handle_occurred_error(struct hclge_dev *hdev);
+pci_ers_result_t hclge_handle_hw_ras_error(struct hnae3_ae_dev *ae_dev);
+int hclge_handle_hw_msix_error(struct hclge_dev *hdev,
+			       unsigned long *reset_requests);
+int hclge_handle_error_info_log(struct hnae3_ae_dev *ae_dev);
+int hclge_handle_mac_tnl(struct hclge_dev *hdev);
+int hclge_handle_vf_queue_err_ras(struct hclge_dev *hdev);
+#endif
diff --git a/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_main.c b/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_main.c
new file mode 100644
index 000000000000..9d34d28ff168
--- /dev/null
+++ b/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_main.c
@@ -0,0 +1,12941 @@
+// SPDX-License-Identifier: GPL-2.0+
+// Copyright (c) 2016-2017 Hisilicon Limited.
+
+#include <linux/acpi.h>
+#include <linux/device.h>
+#include <linux/etherdevice.h>
+#include <linux/init.h>
+#include <linux/interrupt.h>
+#include <linux/irq.h>
+#include <linux/kernel.h>
+#include <linux/module.h>
+#include <linux/netdevice.h>
+#include <linux/pci.h>
+#include <linux/platform_device.h>
+#include <linux/if_vlan.h>
+#include <linux/crash_dump.h>
+
+#include <net/rtnetlink.h>
+
+#include "hclge_cmd.h"
+#include "hclge_dcb.h"
+#include "hclge_main.h"
+#include "hclge_mbx.h"
+#include "hclge_mdio.h"
+#include "hclge_regs.h"
+#include "hclge_tm.h"
+#include "hclge_err.h"
+#include "hnae3.h"
+#include "hclge_devlink.h"
+#include "hclge_comm_cmd.h"
+
+#include "hclge_trace.h"
+
+#define HCLGE_NAME			"hclge"
+
+#define HCLGE_BUF_SIZE_UNIT	256U
+#define HCLGE_BUF_MUL_BY	2
+#define HCLGE_BUF_DIV_BY	2
+#define NEED_RESERVE_TC_NUM	2
+#define BUF_MAX_PERCENT		100
+#define BUF_RESERVE_PERCENT	90
+
+#define HCLGE_RESET_MAX_FAIL_CNT	5
+#define HCLGE_RESET_SYNC_TIME		100
+#define HCLGE_PF_RESET_SYNC_TIME	20
+#define HCLGE_PF_RESET_SYNC_CNT		1500
+
+#define HCLGE_LINK_STATUS_MS	10
+
+static int hclge_set_mac_mtu(struct hclge_dev *hdev, int new_mps);
+static int hclge_init_vlan_config(struct hclge_dev *hdev);
+static void hclge_sync_vlan_filter(struct hclge_dev *hdev);
+static int hclge_reset_ae_dev(struct hnae3_ae_dev *ae_dev);
+static bool hclge_get_hw_reset_stat(struct hnae3_handle *handle);
+static void hclge_rfs_filter_expire(struct hclge_dev *hdev);
+static int hclge_clear_arfs_rules(struct hclge_dev *hdev);
+static enum hnae3_reset_type hclge_get_reset_level(struct hnae3_ae_dev *ae_dev,
+						   unsigned long *addr);
+static int hclge_set_default_loopback(struct hclge_dev *hdev);
+
+static void hclge_sync_mac_table(struct hclge_dev *hdev);
+static void hclge_restore_hw_table(struct hclge_dev *hdev);
+static void hclge_sync_promisc_mode(struct hclge_dev *hdev);
+static void hclge_sync_fd_table(struct hclge_dev *hdev);
+static void hclge_update_fec_stats(struct hclge_dev *hdev);
+static int hclge_mac_link_status_wait(struct hclge_dev *hdev, int link_ret,
+				      int wait_cnt);
+static int hclge_update_port_info(struct hclge_dev *hdev);
+
+static struct hnae3_ae_algo ae_algo;
+
+static struct workqueue_struct *hclge_wq;
+
+static const struct pci_device_id ae_algo_pci_tbl[] = {
+	{PCI_VDEVICE(HUAWEI, HNAE3_DEV_ID_GE), 0},
+	{PCI_VDEVICE(HUAWEI, HNAE3_DEV_ID_25GE), 0},
+	{PCI_VDEVICE(HUAWEI, HNAE3_DEV_ID_25GE_RDMA), 0},
+	{PCI_VDEVICE(HUAWEI, HNAE3_DEV_ID_25GE_RDMA_MACSEC), 0},
+	{PCI_VDEVICE(HUAWEI, HNAE3_DEV_ID_50GE_RDMA), 0},
+	{PCI_VDEVICE(HUAWEI, HNAE3_DEV_ID_50GE_RDMA_MACSEC), 0},
+	{PCI_VDEVICE(HUAWEI, HNAE3_DEV_ID_100G_RDMA_MACSEC), 0},
+	{PCI_VDEVICE(HUAWEI, HNAE3_DEV_ID_200G_RDMA), 0},
+	/* required last entry */
+	{0, }
+};
+
+MODULE_DEVICE_TABLE(pci, ae_algo_pci_tbl);
+
+static const char hns3_nic_test_strs[][ETH_GSTRING_LEN] = {
+	"External Loopback test",
+	"App      Loopback test",
+	"Serdes   serial Loopback test",
+	"Serdes   parallel Loopback test",
+	"Phy      Loopback test"
+};
+
+static const struct hclge_comm_stats_str g_mac_stats_string[] = {
+	{"mac_tx_mac_pause_num", HCLGE_MAC_STATS_MAX_NUM_V1,
+		HCLGE_MAC_STATS_FIELD_OFF(mac_tx_mac_pause_num)},
+	{"mac_rx_mac_pause_num", HCLGE_MAC_STATS_MAX_NUM_V1,
+		HCLGE_MAC_STATS_FIELD_OFF(mac_rx_mac_pause_num)},
+	{"mac_tx_pause_xoff_time", HCLGE_MAC_STATS_MAX_NUM_V2,
+		HCLGE_MAC_STATS_FIELD_OFF(mac_tx_pause_xoff_time)},
+	{"mac_rx_pause_xoff_time", HCLGE_MAC_STATS_MAX_NUM_V2,
+		HCLGE_MAC_STATS_FIELD_OFF(mac_rx_pause_xoff_time)},
+	{"mac_tx_control_pkt_num", HCLGE_MAC_STATS_MAX_NUM_V1,
+		HCLGE_MAC_STATS_FIELD_OFF(mac_tx_ctrl_pkt_num)},
+	{"mac_rx_control_pkt_num", HCLGE_MAC_STATS_MAX_NUM_V1,
+		HCLGE_MAC_STATS_FIELD_OFF(mac_rx_ctrl_pkt_num)},
+	{"mac_tx_pfc_pkt_num", HCLGE_MAC_STATS_MAX_NUM_V1,
+		HCLGE_MAC_STATS_FIELD_OFF(mac_tx_pfc_pause_pkt_num)},
+	{"mac_tx_pfc_pri0_pkt_num", HCLGE_MAC_STATS_MAX_NUM_V1,
+		HCLGE_MAC_STATS_FIELD_OFF(mac_tx_pfc_pri0_pkt_num)},
+	{"mac_tx_pfc_pri1_pkt_num", HCLGE_MAC_STATS_MAX_NUM_V1,
+		HCLGE_MAC_STATS_FIELD_OFF(mac_tx_pfc_pri1_pkt_num)},
+	{"mac_tx_pfc_pri2_pkt_num", HCLGE_MAC_STATS_MAX_NUM_V1,
+		HCLGE_MAC_STATS_FIELD_OFF(mac_tx_pfc_pri2_pkt_num)},
+	{"mac_tx_pfc_pri3_pkt_num", HCLGE_MAC_STATS_MAX_NUM_V1,
+		HCLGE_MAC_STATS_FIELD_OFF(mac_tx_pfc_pri3_pkt_num)},
+	{"mac_tx_pfc_pri4_pkt_num", HCLGE_MAC_STATS_MAX_NUM_V1,
+		HCLGE_MAC_STATS_FIELD_OFF(mac_tx_pfc_pri4_pkt_num)},
+	{"mac_tx_pfc_pri5_pkt_num", HCLGE_MAC_STATS_MAX_NUM_V1,
+		HCLGE_MAC_STATS_FIELD_OFF(mac_tx_pfc_pri5_pkt_num)},
+	{"mac_tx_pfc_pri6_pkt_num", HCLGE_MAC_STATS_MAX_NUM_V1,
+		HCLGE_MAC_STATS_FIELD_OFF(mac_tx_pfc_pri6_pkt_num)},
+	{"mac_tx_pfc_pri7_pkt_num", HCLGE_MAC_STATS_MAX_NUM_V1,
+		HCLGE_MAC_STATS_FIELD_OFF(mac_tx_pfc_pri7_pkt_num)},
+	{"mac_tx_pfc_pri0_xoff_time", HCLGE_MAC_STATS_MAX_NUM_V2,
+		HCLGE_MAC_STATS_FIELD_OFF(mac_tx_pfc_pri0_xoff_time)},
+	{"mac_tx_pfc_pri1_xoff_time", HCLGE_MAC_STATS_MAX_NUM_V2,
+		HCLGE_MAC_STATS_FIELD_OFF(mac_tx_pfc_pri1_xoff_time)},
+	{"mac_tx_pfc_pri2_xoff_time", HCLGE_MAC_STATS_MAX_NUM_V2,
+		HCLGE_MAC_STATS_FIELD_OFF(mac_tx_pfc_pri2_xoff_time)},
+	{"mac_tx_pfc_pri3_xoff_time", HCLGE_MAC_STATS_MAX_NUM_V2,
+		HCLGE_MAC_STATS_FIELD_OFF(mac_tx_pfc_pri3_xoff_time)},
+	{"mac_tx_pfc_pri4_xoff_time", HCLGE_MAC_STATS_MAX_NUM_V2,
+		HCLGE_MAC_STATS_FIELD_OFF(mac_tx_pfc_pri4_xoff_time)},
+	{"mac_tx_pfc_pri5_xoff_time", HCLGE_MAC_STATS_MAX_NUM_V2,
+		HCLGE_MAC_STATS_FIELD_OFF(mac_tx_pfc_pri5_xoff_time)},
+	{"mac_tx_pfc_pri6_xoff_time", HCLGE_MAC_STATS_MAX_NUM_V2,
+		HCLGE_MAC_STATS_FIELD_OFF(mac_tx_pfc_pri6_xoff_time)},
+	{"mac_tx_pfc_pri7_xoff_time", HCLGE_MAC_STATS_MAX_NUM_V2,
+		HCLGE_MAC_STATS_FIELD_OFF(mac_tx_pfc_pri7_xoff_time)},
+	{"mac_rx_pfc_pkt_num", HCLGE_MAC_STATS_MAX_NUM_V1,
+		HCLGE_MAC_STATS_FIELD_OFF(mac_rx_pfc_pause_pkt_num)},
+	{"mac_rx_pfc_pri0_pkt_num", HCLGE_MAC_STATS_MAX_NUM_V1,
+		HCLGE_MAC_STATS_FIELD_OFF(mac_rx_pfc_pri0_pkt_num)},
+	{"mac_rx_pfc_pri1_pkt_num", HCLGE_MAC_STATS_MAX_NUM_V1,
+		HCLGE_MAC_STATS_FIELD_OFF(mac_rx_pfc_pri1_pkt_num)},
+	{"mac_rx_pfc_pri2_pkt_num", HCLGE_MAC_STATS_MAX_NUM_V1,
+		HCLGE_MAC_STATS_FIELD_OFF(mac_rx_pfc_pri2_pkt_num)},
+	{"mac_rx_pfc_pri3_pkt_num", HCLGE_MAC_STATS_MAX_NUM_V1,
+		HCLGE_MAC_STATS_FIELD_OFF(mac_rx_pfc_pri3_pkt_num)},
+	{"mac_rx_pfc_pri4_pkt_num", HCLGE_MAC_STATS_MAX_NUM_V1,
+		HCLGE_MAC_STATS_FIELD_OFF(mac_rx_pfc_pri4_pkt_num)},
+	{"mac_rx_pfc_pri5_pkt_num", HCLGE_MAC_STATS_MAX_NUM_V1,
+		HCLGE_MAC_STATS_FIELD_OFF(mac_rx_pfc_pri5_pkt_num)},
+	{"mac_rx_pfc_pri6_pkt_num", HCLGE_MAC_STATS_MAX_NUM_V1,
+		HCLGE_MAC_STATS_FIELD_OFF(mac_rx_pfc_pri6_pkt_num)},
+	{"mac_rx_pfc_pri7_pkt_num", HCLGE_MAC_STATS_MAX_NUM_V1,
+		HCLGE_MAC_STATS_FIELD_OFF(mac_rx_pfc_pri7_pkt_num)},
+	{"mac_rx_pfc_pri0_xoff_time", HCLGE_MAC_STATS_MAX_NUM_V2,
+		HCLGE_MAC_STATS_FIELD_OFF(mac_rx_pfc_pri0_xoff_time)},
+	{"mac_rx_pfc_pri1_xoff_time", HCLGE_MAC_STATS_MAX_NUM_V2,
+		HCLGE_MAC_STATS_FIELD_OFF(mac_rx_pfc_pri1_xoff_time)},
+	{"mac_rx_pfc_pri2_xoff_time", HCLGE_MAC_STATS_MAX_NUM_V2,
+		HCLGE_MAC_STATS_FIELD_OFF(mac_rx_pfc_pri2_xoff_time)},
+	{"mac_rx_pfc_pri3_xoff_time", HCLGE_MAC_STATS_MAX_NUM_V2,
+		HCLGE_MAC_STATS_FIELD_OFF(mac_rx_pfc_pri3_xoff_time)},
+	{"mac_rx_pfc_pri4_xoff_time", HCLGE_MAC_STATS_MAX_NUM_V2,
+		HCLGE_MAC_STATS_FIELD_OFF(mac_rx_pfc_pri4_xoff_time)},
+	{"mac_rx_pfc_pri5_xoff_time", HCLGE_MAC_STATS_MAX_NUM_V2,
+		HCLGE_MAC_STATS_FIELD_OFF(mac_rx_pfc_pri5_xoff_time)},
+	{"mac_rx_pfc_pri6_xoff_time", HCLGE_MAC_STATS_MAX_NUM_V2,
+		HCLGE_MAC_STATS_FIELD_OFF(mac_rx_pfc_pri6_xoff_time)},
+	{"mac_rx_pfc_pri7_xoff_time", HCLGE_MAC_STATS_MAX_NUM_V2,
+		HCLGE_MAC_STATS_FIELD_OFF(mac_rx_pfc_pri7_xoff_time)},
+	{"mac_tx_total_pkt_num", HCLGE_MAC_STATS_MAX_NUM_V1,
+		HCLGE_MAC_STATS_FIELD_OFF(mac_tx_total_pkt_num)},
+	{"mac_tx_total_oct_num", HCLGE_MAC_STATS_MAX_NUM_V1,
+		HCLGE_MAC_STATS_FIELD_OFF(mac_tx_total_oct_num)},
+	{"mac_tx_good_pkt_num", HCLGE_MAC_STATS_MAX_NUM_V1,
+		HCLGE_MAC_STATS_FIELD_OFF(mac_tx_good_pkt_num)},
+	{"mac_tx_bad_pkt_num", HCLGE_MAC_STATS_MAX_NUM_V1,
+		HCLGE_MAC_STATS_FIELD_OFF(mac_tx_bad_pkt_num)},
+	{"mac_tx_good_oct_num", HCLGE_MAC_STATS_MAX_NUM_V1,
+		HCLGE_MAC_STATS_FIELD_OFF(mac_tx_good_oct_num)},
+	{"mac_tx_bad_oct_num", HCLGE_MAC_STATS_MAX_NUM_V1,
+		HCLGE_MAC_STATS_FIELD_OFF(mac_tx_bad_oct_num)},
+	{"mac_tx_uni_pkt_num", HCLGE_MAC_STATS_MAX_NUM_V1,
+		HCLGE_MAC_STATS_FIELD_OFF(mac_tx_uni_pkt_num)},
+	{"mac_tx_multi_pkt_num", HCLGE_MAC_STATS_MAX_NUM_V1,
+		HCLGE_MAC_STATS_FIELD_OFF(mac_tx_multi_pkt_num)},
+	{"mac_tx_broad_pkt_num", HCLGE_MAC_STATS_MAX_NUM_V1,
+		HCLGE_MAC_STATS_FIELD_OFF(mac_tx_broad_pkt_num)},
+	{"mac_tx_undersize_pkt_num", HCLGE_MAC_STATS_MAX_NUM_V1,
+		HCLGE_MAC_STATS_FIELD_OFF(mac_tx_undersize_pkt_num)},
+	{"mac_tx_oversize_pkt_num", HCLGE_MAC_STATS_MAX_NUM_V1,
+		HCLGE_MAC_STATS_FIELD_OFF(mac_tx_oversize_pkt_num)},
+	{"mac_tx_64_oct_pkt_num", HCLGE_MAC_STATS_MAX_NUM_V1,
+		HCLGE_MAC_STATS_FIELD_OFF(mac_tx_64_oct_pkt_num)},
+	{"mac_tx_65_127_oct_pkt_num", HCLGE_MAC_STATS_MAX_NUM_V1,
+		HCLGE_MAC_STATS_FIELD_OFF(mac_tx_65_127_oct_pkt_num)},
+	{"mac_tx_128_255_oct_pkt_num", HCLGE_MAC_STATS_MAX_NUM_V1,
+		HCLGE_MAC_STATS_FIELD_OFF(mac_tx_128_255_oct_pkt_num)},
+	{"mac_tx_256_511_oct_pkt_num", HCLGE_MAC_STATS_MAX_NUM_V1,
+		HCLGE_MAC_STATS_FIELD_OFF(mac_tx_256_511_oct_pkt_num)},
+	{"mac_tx_512_1023_oct_pkt_num", HCLGE_MAC_STATS_MAX_NUM_V1,
+		HCLGE_MAC_STATS_FIELD_OFF(mac_tx_512_1023_oct_pkt_num)},
+	{"mac_tx_1024_1518_oct_pkt_num", HCLGE_MAC_STATS_MAX_NUM_V1,
+		HCLGE_MAC_STATS_FIELD_OFF(mac_tx_1024_1518_oct_pkt_num)},
+	{"mac_tx_1519_2047_oct_pkt_num", HCLGE_MAC_STATS_MAX_NUM_V1,
+		HCLGE_MAC_STATS_FIELD_OFF(mac_tx_1519_2047_oct_pkt_num)},
+	{"mac_tx_2048_4095_oct_pkt_num", HCLGE_MAC_STATS_MAX_NUM_V1,
+		HCLGE_MAC_STATS_FIELD_OFF(mac_tx_2048_4095_oct_pkt_num)},
+	{"mac_tx_4096_8191_oct_pkt_num", HCLGE_MAC_STATS_MAX_NUM_V1,
+		HCLGE_MAC_STATS_FIELD_OFF(mac_tx_4096_8191_oct_pkt_num)},
+	{"mac_tx_8192_9216_oct_pkt_num", HCLGE_MAC_STATS_MAX_NUM_V1,
+		HCLGE_MAC_STATS_FIELD_OFF(mac_tx_8192_9216_oct_pkt_num)},
+	{"mac_tx_9217_12287_oct_pkt_num", HCLGE_MAC_STATS_MAX_NUM_V1,
+		HCLGE_MAC_STATS_FIELD_OFF(mac_tx_9217_12287_oct_pkt_num)},
+	{"mac_tx_12288_16383_oct_pkt_num", HCLGE_MAC_STATS_MAX_NUM_V1,
+		HCLGE_MAC_STATS_FIELD_OFF(mac_tx_12288_16383_oct_pkt_num)},
+	{"mac_tx_1519_max_good_pkt_num", HCLGE_MAC_STATS_MAX_NUM_V1,
+		HCLGE_MAC_STATS_FIELD_OFF(mac_tx_1519_max_good_oct_pkt_num)},
+	{"mac_tx_1519_max_bad_pkt_num", HCLGE_MAC_STATS_MAX_NUM_V1,
+		HCLGE_MAC_STATS_FIELD_OFF(mac_tx_1519_max_bad_oct_pkt_num)},
+	{"mac_rx_total_pkt_num", HCLGE_MAC_STATS_MAX_NUM_V1,
+		HCLGE_MAC_STATS_FIELD_OFF(mac_rx_total_pkt_num)},
+	{"mac_rx_total_oct_num", HCLGE_MAC_STATS_MAX_NUM_V1,
+		HCLGE_MAC_STATS_FIELD_OFF(mac_rx_total_oct_num)},
+	{"mac_rx_good_pkt_num", HCLGE_MAC_STATS_MAX_NUM_V1,
+		HCLGE_MAC_STATS_FIELD_OFF(mac_rx_good_pkt_num)},
+	{"mac_rx_bad_pkt_num", HCLGE_MAC_STATS_MAX_NUM_V1,
+		HCLGE_MAC_STATS_FIELD_OFF(mac_rx_bad_pkt_num)},
+	{"mac_rx_good_oct_num", HCLGE_MAC_STATS_MAX_NUM_V1,
+		HCLGE_MAC_STATS_FIELD_OFF(mac_rx_good_oct_num)},
+	{"mac_rx_bad_oct_num", HCLGE_MAC_STATS_MAX_NUM_V1,
+		HCLGE_MAC_STATS_FIELD_OFF(mac_rx_bad_oct_num)},
+	{"mac_rx_uni_pkt_num", HCLGE_MAC_STATS_MAX_NUM_V1,
+		HCLGE_MAC_STATS_FIELD_OFF(mac_rx_uni_pkt_num)},
+	{"mac_rx_multi_pkt_num", HCLGE_MAC_STATS_MAX_NUM_V1,
+		HCLGE_MAC_STATS_FIELD_OFF(mac_rx_multi_pkt_num)},
+	{"mac_rx_broad_pkt_num", HCLGE_MAC_STATS_MAX_NUM_V1,
+		HCLGE_MAC_STATS_FIELD_OFF(mac_rx_broad_pkt_num)},
+	{"mac_rx_undersize_pkt_num", HCLGE_MAC_STATS_MAX_NUM_V1,
+		HCLGE_MAC_STATS_FIELD_OFF(mac_rx_undersize_pkt_num)},
+	{"mac_rx_oversize_pkt_num", HCLGE_MAC_STATS_MAX_NUM_V1,
+		HCLGE_MAC_STATS_FIELD_OFF(mac_rx_oversize_pkt_num)},
+	{"mac_rx_64_oct_pkt_num", HCLGE_MAC_STATS_MAX_NUM_V1,
+		HCLGE_MAC_STATS_FIELD_OFF(mac_rx_64_oct_pkt_num)},
+	{"mac_rx_65_127_oct_pkt_num", HCLGE_MAC_STATS_MAX_NUM_V1,
+		HCLGE_MAC_STATS_FIELD_OFF(mac_rx_65_127_oct_pkt_num)},
+	{"mac_rx_128_255_oct_pkt_num", HCLGE_MAC_STATS_MAX_NUM_V1,
+		HCLGE_MAC_STATS_FIELD_OFF(mac_rx_128_255_oct_pkt_num)},
+	{"mac_rx_256_511_oct_pkt_num", HCLGE_MAC_STATS_MAX_NUM_V1,
+		HCLGE_MAC_STATS_FIELD_OFF(mac_rx_256_511_oct_pkt_num)},
+	{"mac_rx_512_1023_oct_pkt_num", HCLGE_MAC_STATS_MAX_NUM_V1,
+		HCLGE_MAC_STATS_FIELD_OFF(mac_rx_512_1023_oct_pkt_num)},
+	{"mac_rx_1024_1518_oct_pkt_num", HCLGE_MAC_STATS_MAX_NUM_V1,
+		HCLGE_MAC_STATS_FIELD_OFF(mac_rx_1024_1518_oct_pkt_num)},
+	{"mac_rx_1519_2047_oct_pkt_num", HCLGE_MAC_STATS_MAX_NUM_V1,
+		HCLGE_MAC_STATS_FIELD_OFF(mac_rx_1519_2047_oct_pkt_num)},
+	{"mac_rx_2048_4095_oct_pkt_num", HCLGE_MAC_STATS_MAX_NUM_V1,
+		HCLGE_MAC_STATS_FIELD_OFF(mac_rx_2048_4095_oct_pkt_num)},
+	{"mac_rx_4096_8191_oct_pkt_num", HCLGE_MAC_STATS_MAX_NUM_V1,
+		HCLGE_MAC_STATS_FIELD_OFF(mac_rx_4096_8191_oct_pkt_num)},
+	{"mac_rx_8192_9216_oct_pkt_num", HCLGE_MAC_STATS_MAX_NUM_V1,
+		HCLGE_MAC_STATS_FIELD_OFF(mac_rx_8192_9216_oct_pkt_num)},
+	{"mac_rx_9217_12287_oct_pkt_num", HCLGE_MAC_STATS_MAX_NUM_V1,
+		HCLGE_MAC_STATS_FIELD_OFF(mac_rx_9217_12287_oct_pkt_num)},
+	{"mac_rx_12288_16383_oct_pkt_num", HCLGE_MAC_STATS_MAX_NUM_V1,
+		HCLGE_MAC_STATS_FIELD_OFF(mac_rx_12288_16383_oct_pkt_num)},
+	{"mac_rx_1519_max_good_pkt_num", HCLGE_MAC_STATS_MAX_NUM_V1,
+		HCLGE_MAC_STATS_FIELD_OFF(mac_rx_1519_max_good_oct_pkt_num)},
+	{"mac_rx_1519_max_bad_pkt_num", HCLGE_MAC_STATS_MAX_NUM_V1,
+		HCLGE_MAC_STATS_FIELD_OFF(mac_rx_1519_max_bad_oct_pkt_num)},
+
+	{"mac_tx_fragment_pkt_num", HCLGE_MAC_STATS_MAX_NUM_V1,
+		HCLGE_MAC_STATS_FIELD_OFF(mac_tx_fragment_pkt_num)},
+	{"mac_tx_undermin_pkt_num", HCLGE_MAC_STATS_MAX_NUM_V1,
+		HCLGE_MAC_STATS_FIELD_OFF(mac_tx_undermin_pkt_num)},
+	{"mac_tx_jabber_pkt_num", HCLGE_MAC_STATS_MAX_NUM_V1,
+		HCLGE_MAC_STATS_FIELD_OFF(mac_tx_jabber_pkt_num)},
+	{"mac_tx_err_all_pkt_num", HCLGE_MAC_STATS_MAX_NUM_V1,
+		HCLGE_MAC_STATS_FIELD_OFF(mac_tx_err_all_pkt_num)},
+	{"mac_tx_from_app_good_pkt_num", HCLGE_MAC_STATS_MAX_NUM_V1,
+		HCLGE_MAC_STATS_FIELD_OFF(mac_tx_from_app_good_pkt_num)},
+	{"mac_tx_from_app_bad_pkt_num", HCLGE_MAC_STATS_MAX_NUM_V1,
+		HCLGE_MAC_STATS_FIELD_OFF(mac_tx_from_app_bad_pkt_num)},
+	{"mac_rx_fragment_pkt_num", HCLGE_MAC_STATS_MAX_NUM_V1,
+		HCLGE_MAC_STATS_FIELD_OFF(mac_rx_fragment_pkt_num)},
+	{"mac_rx_undermin_pkt_num", HCLGE_MAC_STATS_MAX_NUM_V1,
+		HCLGE_MAC_STATS_FIELD_OFF(mac_rx_undermin_pkt_num)},
+	{"mac_rx_jabber_pkt_num", HCLGE_MAC_STATS_MAX_NUM_V1,
+		HCLGE_MAC_STATS_FIELD_OFF(mac_rx_jabber_pkt_num)},
+	{"mac_rx_fcs_err_pkt_num", HCLGE_MAC_STATS_MAX_NUM_V1,
+		HCLGE_MAC_STATS_FIELD_OFF(mac_rx_fcs_err_pkt_num)},
+	{"mac_rx_send_app_good_pkt_num", HCLGE_MAC_STATS_MAX_NUM_V1,
+		HCLGE_MAC_STATS_FIELD_OFF(mac_rx_send_app_good_pkt_num)},
+	{"mac_rx_send_app_bad_pkt_num", HCLGE_MAC_STATS_MAX_NUM_V1,
+		HCLGE_MAC_STATS_FIELD_OFF(mac_rx_send_app_bad_pkt_num)}
+};
+
+static const struct hclge_mac_mgr_tbl_entry_cmd hclge_mgr_table[] = {
+	{
+		.flags = HCLGE_MAC_MGR_MASK_VLAN_B,
+		.ethter_type = cpu_to_le16(ETH_P_LLDP),
+		.mac_addr = {0x01, 0x80, 0xc2, 0x00, 0x00, 0x0e},
+		.i_port_bitmap = 0x1,
+	},
+};
+
+static const struct key_info meta_data_key_info[] = {
+	{ PACKET_TYPE_ID, 6 },
+	{ IP_FRAGEMENT, 1 },
+	{ ROCE_TYPE, 1 },
+	{ NEXT_KEY, 5 },
+	{ VLAN_NUMBER, 2 },
+	{ SRC_VPORT, 12 },
+	{ DST_VPORT, 12 },
+	{ TUNNEL_PACKET, 1 },
+};
+
+static const struct key_info tuple_key_info[] = {
+	{ OUTER_DST_MAC, 48, KEY_OPT_MAC, -1, -1 },
+	{ OUTER_SRC_MAC, 48, KEY_OPT_MAC, -1, -1 },
+	{ OUTER_VLAN_TAG_FST, 16, KEY_OPT_LE16, -1, -1 },
+	{ OUTER_VLAN_TAG_SEC, 16, KEY_OPT_LE16, -1, -1 },
+	{ OUTER_ETH_TYPE, 16, KEY_OPT_LE16, -1, -1 },
+	{ OUTER_L2_RSV, 16, KEY_OPT_LE16, -1, -1 },
+	{ OUTER_IP_TOS, 8, KEY_OPT_U8, -1, -1 },
+	{ OUTER_IP_PROTO, 8, KEY_OPT_U8, -1, -1 },
+	{ OUTER_SRC_IP, 32, KEY_OPT_IP, -1, -1 },
+	{ OUTER_DST_IP, 32, KEY_OPT_IP, -1, -1 },
+	{ OUTER_L3_RSV, 16, KEY_OPT_LE16, -1, -1 },
+	{ OUTER_SRC_PORT, 16, KEY_OPT_LE16, -1, -1 },
+	{ OUTER_DST_PORT, 16, KEY_OPT_LE16, -1, -1 },
+	{ OUTER_L4_RSV, 32, KEY_OPT_LE32, -1, -1 },
+	{ OUTER_TUN_VNI, 24, KEY_OPT_VNI, -1, -1 },
+	{ OUTER_TUN_FLOW_ID, 8, KEY_OPT_U8, -1, -1 },
+	{ INNER_DST_MAC, 48, KEY_OPT_MAC,
+	  offsetof(struct hclge_fd_rule, tuples.dst_mac),
+	  offsetof(struct hclge_fd_rule, tuples_mask.dst_mac) },
+	{ INNER_SRC_MAC, 48, KEY_OPT_MAC,
+	  offsetof(struct hclge_fd_rule, tuples.src_mac),
+	  offsetof(struct hclge_fd_rule, tuples_mask.src_mac) },
+	{ INNER_VLAN_TAG_FST, 16, KEY_OPT_LE16,
+	  offsetof(struct hclge_fd_rule, tuples.vlan_tag1),
+	  offsetof(struct hclge_fd_rule, tuples_mask.vlan_tag1) },
+	{ INNER_VLAN_TAG_SEC, 16, KEY_OPT_LE16, -1, -1 },
+	{ INNER_ETH_TYPE, 16, KEY_OPT_LE16,
+	  offsetof(struct hclge_fd_rule, tuples.ether_proto),
+	  offsetof(struct hclge_fd_rule, tuples_mask.ether_proto) },
+	{ INNER_L2_RSV, 16, KEY_OPT_LE16,
+	  offsetof(struct hclge_fd_rule, tuples.l2_user_def),
+	  offsetof(struct hclge_fd_rule, tuples_mask.l2_user_def) },
+	{ INNER_IP_TOS, 8, KEY_OPT_U8,
+	  offsetof(struct hclge_fd_rule, tuples.ip_tos),
+	  offsetof(struct hclge_fd_rule, tuples_mask.ip_tos) },
+	{ INNER_IP_PROTO, 8, KEY_OPT_U8,
+	  offsetof(struct hclge_fd_rule, tuples.ip_proto),
+	  offsetof(struct hclge_fd_rule, tuples_mask.ip_proto) },
+	{ INNER_SRC_IP, 32, KEY_OPT_IP,
+	  offsetof(struct hclge_fd_rule, tuples.src_ip),
+	  offsetof(struct hclge_fd_rule, tuples_mask.src_ip) },
+	{ INNER_DST_IP, 32, KEY_OPT_IP,
+	  offsetof(struct hclge_fd_rule, tuples.dst_ip),
+	  offsetof(struct hclge_fd_rule, tuples_mask.dst_ip) },
+	{ INNER_L3_RSV, 16, KEY_OPT_LE16,
+	  offsetof(struct hclge_fd_rule, tuples.l3_user_def),
+	  offsetof(struct hclge_fd_rule, tuples_mask.l3_user_def) },
+	{ INNER_SRC_PORT, 16, KEY_OPT_LE16,
+	  offsetof(struct hclge_fd_rule, tuples.src_port),
+	  offsetof(struct hclge_fd_rule, tuples_mask.src_port) },
+	{ INNER_DST_PORT, 16, KEY_OPT_LE16,
+	  offsetof(struct hclge_fd_rule, tuples.dst_port),
+	  offsetof(struct hclge_fd_rule, tuples_mask.dst_port) },
+	{ INNER_L4_RSV, 32, KEY_OPT_LE32,
+	  offsetof(struct hclge_fd_rule, tuples.l4_user_def),
+	  offsetof(struct hclge_fd_rule, tuples_mask.l4_user_def) },
+};
+
+/**
+ * hclge_cmd_send - send command to command queue
+ * @hw: pointer to the hw struct
+ * @desc: prefilled descriptor for describing the command
+ * @num : the number of descriptors to be sent
+ *
+ * This is the main send command for command queue, it
+ * sends the queue, cleans the queue, etc
+ **/
+int hclge_cmd_send(struct hclge_hw *hw, struct hclge_desc *desc, int num)
+{
+	return hclge_comm_cmd_send(&hw->hw, desc, num);
+}
+
+static void hclge_trace_cmd_send(struct hclge_comm_hw *hw, struct hclge_desc *desc,
+				 int num, bool is_special)
+{
+	int i;
+
+	trace_hclge_pf_cmd_send(hw, desc, 0, num);
+
+	if (!is_special) {
+		for (i = 1; i < num; i++)
+			trace_hclge_pf_cmd_send(hw, &desc[i], i, num);
+	} else {
+		for (i = 1; i < num; i++)
+			trace_hclge_pf_special_cmd_send(hw, (__le32 *)&desc[i],
+							i, num);
+	}
+}
+
+static void hclge_trace_cmd_get(struct hclge_comm_hw *hw, struct hclge_desc *desc,
+				int num, bool is_special)
+{
+	int i;
+
+	if (!HCLGE_COMM_SEND_SYNC(le16_to_cpu(desc->flag)))
+		return;
+
+	trace_hclge_pf_cmd_get(hw, desc, 0, num);
+
+	if (!is_special) {
+		for (i = 1; i < num; i++)
+			trace_hclge_pf_cmd_get(hw, &desc[i], i, num);
+	} else {
+		for (i = 1; i < num; i++)
+			trace_hclge_pf_special_cmd_get(hw, (__le32 *)&desc[i],
+						       i, num);
+	}
+}
+
+static const struct hclge_comm_cmq_ops hclge_cmq_ops = {
+	.trace_cmd_send = hclge_trace_cmd_send,
+	.trace_cmd_get = hclge_trace_cmd_get,
+};
+
+static int hclge_mac_update_stats_defective(struct hclge_dev *hdev)
+{
+#define HCLGE_MAC_CMD_NUM 21
+
+	u64 *data = (u64 *)(&hdev->mac_stats);
+	struct hclge_desc desc[HCLGE_MAC_CMD_NUM];
+	__le64 *desc_data;
+	u32 data_size;
+	int ret;
+	u32 i;
+
+	hclge_cmd_setup_basic_desc(&desc[0], HCLGE_OPC_STATS_MAC, true);
+	ret = hclge_cmd_send(&hdev->hw, desc, HCLGE_MAC_CMD_NUM);
+	if (ret) {
+		dev_err(&hdev->pdev->dev,
+			"Get MAC pkt stats fail, status = %d.\n", ret);
+
+		return ret;
+	}
+
+	/* The first desc has a 64-bit header, so data size need to minus 1 */
+	data_size = sizeof(desc) / (sizeof(u64)) - 1;
+
+	desc_data = (__le64 *)(&desc[0].data[0]);
+	for (i = 0; i < data_size; i++) {
+		/* data memory is continuous becase only the first desc has a
+		 * header in this command
+		 */
+		*data += le64_to_cpu(*desc_data);
+		data++;
+		desc_data++;
+	}
+
+	return 0;
+}
+
+static int hclge_mac_update_stats_complete(struct hclge_dev *hdev)
+{
+#define HCLGE_REG_NUM_PER_DESC		4
+
+	u32 reg_num = hdev->ae_dev->dev_specs.mac_stats_num;
+	u64 *data = (u64 *)(&hdev->mac_stats);
+	struct hclge_desc *desc;
+	__le64 *desc_data;
+	u32 data_size;
+	u32 desc_num;
+	int ret;
+	u32 i;
+
+	/* The first desc has a 64-bit header, so need to consider it */
+	desc_num = reg_num / HCLGE_REG_NUM_PER_DESC + 1;
+
+	/* This may be called inside atomic sections,
+	 * so GFP_ATOMIC is more suitable here
+	 */
+	desc = kcalloc(desc_num, sizeof(struct hclge_desc), GFP_ATOMIC);
+	if (!desc)
+		return -ENOMEM;
+
+	hclge_cmd_setup_basic_desc(&desc[0], HCLGE_OPC_STATS_MAC_ALL, true);
+	ret = hclge_cmd_send(&hdev->hw, desc, desc_num);
+	if (ret) {
+		kfree(desc);
+		return ret;
+	}
+
+	data_size = min_t(u32, sizeof(hdev->mac_stats) / sizeof(u64), reg_num);
+
+	desc_data = (__le64 *)(&desc[0].data[0]);
+	for (i = 0; i < data_size; i++) {
+		/* data memory is continuous becase only the first desc has a
+		 * header in this command
+		 */
+		*data += le64_to_cpu(*desc_data);
+		data++;
+		desc_data++;
+	}
+
+	kfree(desc);
+
+	return 0;
+}
+
+static int hclge_mac_query_reg_num(struct hclge_dev *hdev, u32 *reg_num)
+{
+	struct hclge_desc desc;
+	int ret;
+
+	/* Driver needs total register number of both valid registers and
+	 * reserved registers, but the old firmware only returns number
+	 * of valid registers in device V2. To be compatible with these
+	 * devices, driver uses a fixed value.
+	 */
+	if (hdev->ae_dev->dev_version == HNAE3_DEVICE_VERSION_V2) {
+		*reg_num = HCLGE_MAC_STATS_MAX_NUM_V1;
+		return 0;
+	}
+
+	hclge_cmd_setup_basic_desc(&desc, HCLGE_OPC_QUERY_MAC_REG_NUM, true);
+	ret = hclge_cmd_send(&hdev->hw, &desc, 1);
+	if (ret) {
+		dev_err(&hdev->pdev->dev,
+			"failed to query mac statistic reg number, ret = %d\n",
+			ret);
+		return ret;
+	}
+
+	*reg_num = le32_to_cpu(desc.data[0]);
+	if (*reg_num == 0) {
+		dev_err(&hdev->pdev->dev,
+			"mac statistic reg number is invalid!\n");
+		return -ENODATA;
+	}
+
+	return 0;
+}
+
+int hclge_mac_update_stats(struct hclge_dev *hdev)
+{
+	/* The firmware supports the new statistics acquisition method */
+	if (hdev->ae_dev->dev_specs.mac_stats_num)
+		return hclge_mac_update_stats_complete(hdev);
+	else
+		return hclge_mac_update_stats_defective(hdev);
+}
+
+static int hclge_comm_get_count(struct hclge_dev *hdev,
+				const struct hclge_comm_stats_str strs[],
+				u32 size)
+{
+	int count = 0;
+	u32 i;
+
+	for (i = 0; i < size; i++)
+		if (strs[i].stats_num <= hdev->ae_dev->dev_specs.mac_stats_num)
+			count++;
+
+	return count;
+}
+
+static u64 *hclge_comm_get_stats(struct hclge_dev *hdev,
+				 const struct hclge_comm_stats_str strs[],
+				 int size, u64 *data)
+{
+	u64 *buf = data;
+	int i;
+
+	for (i = 0; i < size; i++) {
+		if (strs[i].stats_num > hdev->ae_dev->dev_specs.mac_stats_num)
+			continue;
+
+		*buf = HCLGE_STATS_READ(&hdev->mac_stats, strs[i].offset);
+		buf++;
+	}
+
+	return buf;
+}
+
+static void hclge_comm_get_strings(struct hclge_dev *hdev, u32 stringset,
+				   const struct hclge_comm_stats_str strs[],
+				   int size, u8 **data)
+{
+	int i;
+
+	if (stringset != ETH_SS_STATS)
+		return;
+
+	for (i = 0; i < size; i++) {
+		if (strs[i].stats_num > hdev->ae_dev->dev_specs.mac_stats_num)
+			continue;
+
+		ethtool_puts(data, strs[i].desc);
+	}
+}
+
+static void hclge_update_stats_for_all(struct hclge_dev *hdev)
+{
+	struct hnae3_handle *handle;
+	int status;
+
+	handle = &hdev->vport[0].nic;
+	if (handle->client) {
+		status = hclge_comm_tqps_update_stats(handle, &hdev->hw.hw);
+		if (status) {
+			dev_err(&hdev->pdev->dev,
+				"Update TQPS stats fail, status = %d.\n",
+				status);
+		}
+	}
+
+	hclge_update_fec_stats(hdev);
+
+	status = hclge_mac_update_stats(hdev);
+	if (status)
+		dev_err(&hdev->pdev->dev,
+			"Update MAC stats fail, status = %d.\n", status);
+}
+
+static void hclge_update_stats(struct hnae3_handle *handle)
+{
+	struct hclge_vport *vport = hclge_get_vport(handle);
+	struct hclge_dev *hdev = vport->back;
+	int status;
+
+	if (test_and_set_bit(HCLGE_STATE_STATISTICS_UPDATING, &hdev->state))
+		return;
+
+	status = hclge_mac_update_stats(hdev);
+	if (status)
+		dev_err(&hdev->pdev->dev,
+			"Update MAC stats fail, status = %d.\n",
+			status);
+
+	status = hclge_comm_tqps_update_stats(handle, &hdev->hw.hw);
+	if (status)
+		dev_err(&hdev->pdev->dev,
+			"Update TQPS stats fail, status = %d.\n",
+			status);
+
+	clear_bit(HCLGE_STATE_STATISTICS_UPDATING, &hdev->state);
+}
+
+static int hclge_get_sset_count(struct hnae3_handle *handle, int stringset)
+{
+#define HCLGE_LOOPBACK_TEST_FLAGS (HNAE3_SUPPORT_APP_LOOPBACK | \
+		HNAE3_SUPPORT_PHY_LOOPBACK | \
+		HNAE3_SUPPORT_SERDES_SERIAL_LOOPBACK | \
+		HNAE3_SUPPORT_SERDES_PARALLEL_LOOPBACK | \
+		HNAE3_SUPPORT_EXTERNAL_LOOPBACK)
+
+	struct hclge_vport *vport = hclge_get_vport(handle);
+	struct hclge_dev *hdev = vport->back;
+	int count = 0;
+
+	/* Loopback test support rules:
+	 * mac: only GE mode support
+	 * serdes: all mac mode will support include GE/XGE/LGE/CGE
+	 * phy: only support when phy device exist on board
+	 */
+	if (stringset == ETH_SS_TEST) {
+		/* clear loopback bit flags at first */
+		handle->flags = (handle->flags & (~HCLGE_LOOPBACK_TEST_FLAGS));
+		if (hdev->ae_dev->dev_version >= HNAE3_DEVICE_VERSION_V2 ||
+		    hdev->hw.mac.speed == HCLGE_MAC_SPEED_10M ||
+		    hdev->hw.mac.speed == HCLGE_MAC_SPEED_100M ||
+		    hdev->hw.mac.speed == HCLGE_MAC_SPEED_1G) {
+			count += 1;
+			handle->flags |= HNAE3_SUPPORT_APP_LOOPBACK;
+		}
+
+		if (hdev->ae_dev->dev_specs.hilink_version !=
+		    HCLGE_HILINK_H60) {
+			count += 1;
+			handle->flags |= HNAE3_SUPPORT_SERDES_SERIAL_LOOPBACK;
+		}
+
+		count += 1;
+		handle->flags |= HNAE3_SUPPORT_SERDES_PARALLEL_LOOPBACK;
+		count += 1;
+		handle->flags |= HNAE3_SUPPORT_EXTERNAL_LOOPBACK;
+
+		if ((hdev->hw.mac.phydev && hdev->hw.mac.phydev->drv &&
+		     hdev->hw.mac.phydev->drv->set_loopback) ||
+		    hnae3_dev_phy_imp_supported(hdev)) {
+			count += 1;
+			handle->flags |= HNAE3_SUPPORT_PHY_LOOPBACK;
+		}
+	} else if (stringset == ETH_SS_STATS) {
+		count = hclge_comm_get_count(hdev, g_mac_stats_string,
+					     ARRAY_SIZE(g_mac_stats_string)) +
+			hclge_comm_tqps_get_sset_count(handle);
+	}
+
+	return count;
+}
+
+static void hclge_get_strings(struct hnae3_handle *handle, u32 stringset,
+			      u8 **data)
+{
+	struct hclge_vport *vport = hclge_get_vport(handle);
+	struct hclge_dev *hdev = vport->back;
+	const char *str;
+	int size;
+
+	if (stringset == ETH_SS_STATS) {
+		size = ARRAY_SIZE(g_mac_stats_string);
+		hclge_comm_get_strings(hdev, stringset, g_mac_stats_string,
+				       size, data);
+		hclge_comm_tqps_get_strings(handle, data);
+	} else if (stringset == ETH_SS_TEST) {
+		if (handle->flags & HNAE3_SUPPORT_EXTERNAL_LOOPBACK) {
+			str = hns3_nic_test_strs[HNAE3_LOOP_EXTERNAL];
+			ethtool_puts(data, str);
+		}
+		if (handle->flags & HNAE3_SUPPORT_APP_LOOPBACK) {
+			str = hns3_nic_test_strs[HNAE3_LOOP_APP];
+			ethtool_puts(data, str);
+		}
+		if (handle->flags & HNAE3_SUPPORT_SERDES_SERIAL_LOOPBACK) {
+			str = hns3_nic_test_strs[HNAE3_LOOP_SERIAL_SERDES];
+			ethtool_puts(data, str);
+		}
+		if (handle->flags & HNAE3_SUPPORT_SERDES_PARALLEL_LOOPBACK) {
+			str = hns3_nic_test_strs[HNAE3_LOOP_PARALLEL_SERDES];
+			ethtool_puts(data, str);
+		}
+		if (handle->flags & HNAE3_SUPPORT_PHY_LOOPBACK) {
+			str = hns3_nic_test_strs[HNAE3_LOOP_PHY];
+			ethtool_puts(data, str);
+		}
+	}
+}
+
+static void hclge_get_stats(struct hnae3_handle *handle, u64 *data)
+{
+	struct hclge_vport *vport = hclge_get_vport(handle);
+	struct hclge_dev *hdev = vport->back;
+	u64 *p;
+
+	p = hclge_comm_get_stats(hdev, g_mac_stats_string,
+				 ARRAY_SIZE(g_mac_stats_string), data);
+	p = hclge_comm_tqps_get_stats(handle, p);
+}
+
+static void hclge_get_mac_stat(struct hnae3_handle *handle,
+			       struct hns3_mac_stats *mac_stats)
+{
+	struct hclge_vport *vport = hclge_get_vport(handle);
+	struct hclge_dev *hdev = vport->back;
+
+	hclge_update_stats(handle);
+
+	mac_stats->tx_pause_cnt = hdev->mac_stats.mac_tx_mac_pause_num;
+	mac_stats->rx_pause_cnt = hdev->mac_stats.mac_rx_mac_pause_num;
+}
+
+static int hclge_parse_func_status(struct hclge_dev *hdev,
+				   struct hclge_func_status_cmd *status)
+{
+#define HCLGE_MAC_ID_MASK	0xF
+
+	if (!(status->pf_state & HCLGE_PF_STATE_DONE))
+		return -EINVAL;
+
+	/* Set the pf to main pf */
+	if (status->pf_state & HCLGE_PF_STATE_MAIN)
+		hdev->flag |= HCLGE_FLAG_MAIN;
+	else
+		hdev->flag &= ~HCLGE_FLAG_MAIN;
+
+	hdev->hw.mac.mac_id = status->mac_id & HCLGE_MAC_ID_MASK;
+	return 0;
+}
+
+static int hclge_query_function_status(struct hclge_dev *hdev)
+{
+#define HCLGE_QUERY_MAX_CNT	5
+
+	struct hclge_func_status_cmd *req;
+	struct hclge_desc desc;
+	int timeout = 0;
+	int ret;
+
+	hclge_cmd_setup_basic_desc(&desc, HCLGE_OPC_QUERY_FUNC_STATUS, true);
+	req = (struct hclge_func_status_cmd *)desc.data;
+
+	do {
+		ret = hclge_cmd_send(&hdev->hw, &desc, 1);
+		if (ret) {
+			dev_err(&hdev->pdev->dev,
+				"query function status failed %d.\n", ret);
+			return ret;
+		}
+
+		/* Check pf reset is done */
+		if (req->pf_state)
+			break;
+		usleep_range(1000, 2000);
+	} while (timeout++ < HCLGE_QUERY_MAX_CNT);
+
+	return hclge_parse_func_status(hdev, req);
+}
+
+static int hclge_query_pf_resource(struct hclge_dev *hdev)
+{
+	struct hclge_pf_res_cmd *req;
+	struct hclge_desc desc;
+	int ret;
+
+	hclge_cmd_setup_basic_desc(&desc, HCLGE_OPC_QUERY_PF_RSRC, true);
+	ret = hclge_cmd_send(&hdev->hw, &desc, 1);
+	if (ret) {
+		dev_err(&hdev->pdev->dev,
+			"query pf resource failed %d.\n", ret);
+		return ret;
+	}
+
+	req = (struct hclge_pf_res_cmd *)desc.data;
+	hdev->num_tqps = le16_to_cpu(req->tqp_num) +
+			 le16_to_cpu(req->ext_tqp_num);
+	hdev->pkt_buf_size = le16_to_cpu(req->buf_size) << HCLGE_BUF_UNIT_S;
+
+	if (req->tx_buf_size)
+		hdev->tx_buf_size =
+			le16_to_cpu(req->tx_buf_size) << HCLGE_BUF_UNIT_S;
+	else
+		hdev->tx_buf_size = HCLGE_DEFAULT_TX_BUF;
+
+	hdev->tx_buf_size = roundup(hdev->tx_buf_size, HCLGE_BUF_SIZE_UNIT);
+
+	if (req->dv_buf_size)
+		hdev->dv_buf_size =
+			le16_to_cpu(req->dv_buf_size) << HCLGE_BUF_UNIT_S;
+	else
+		hdev->dv_buf_size = HCLGE_DEFAULT_DV;
+
+	hdev->dv_buf_size = roundup(hdev->dv_buf_size, HCLGE_BUF_SIZE_UNIT);
+
+	hdev->num_nic_msi = le16_to_cpu(req->msixcap_localid_number_nic);
+	if (hdev->num_nic_msi < HNAE3_MIN_VECTOR_NUM) {
+		dev_err(&hdev->pdev->dev,
+			"only %u msi resources available, not enough for pf(min:2).\n",
+			hdev->num_nic_msi);
+		return -EINVAL;
+	}
+
+	if (hnae3_dev_roce_supported(hdev)) {
+		hdev->num_roce_msi =
+			le16_to_cpu(req->pf_intr_vector_number_roce);
+
+		/* PF should have NIC vectors and Roce vectors,
+		 * NIC vectors are queued before Roce vectors.
+		 */
+		hdev->num_msi = hdev->num_nic_msi + hdev->num_roce_msi;
+	} else {
+		hdev->num_msi = hdev->num_nic_msi;
+	}
+
+	return 0;
+}
+
+static int hclge_parse_speed(u8 speed_cmd, u32 *speed)
+{
+	switch (speed_cmd) {
+	case HCLGE_FW_MAC_SPEED_10M:
+		*speed = HCLGE_MAC_SPEED_10M;
+		break;
+	case HCLGE_FW_MAC_SPEED_100M:
+		*speed = HCLGE_MAC_SPEED_100M;
+		break;
+	case HCLGE_FW_MAC_SPEED_1G:
+		*speed = HCLGE_MAC_SPEED_1G;
+		break;
+	case HCLGE_FW_MAC_SPEED_10G:
+		*speed = HCLGE_MAC_SPEED_10G;
+		break;
+	case HCLGE_FW_MAC_SPEED_25G:
+		*speed = HCLGE_MAC_SPEED_25G;
+		break;
+	case HCLGE_FW_MAC_SPEED_40G:
+		*speed = HCLGE_MAC_SPEED_40G;
+		break;
+	case HCLGE_FW_MAC_SPEED_50G:
+		*speed = HCLGE_MAC_SPEED_50G;
+		break;
+	case HCLGE_FW_MAC_SPEED_100G:
+		*speed = HCLGE_MAC_SPEED_100G;
+		break;
+	case HCLGE_FW_MAC_SPEED_200G:
+		*speed = HCLGE_MAC_SPEED_200G;
+		break;
+	default:
+		return -EINVAL;
+	}
+
+	return 0;
+}
+
+static const struct hclge_speed_bit_map speed_bit_map[] = {
+	{HCLGE_MAC_SPEED_10M, HCLGE_SUPPORT_10M_BIT},
+	{HCLGE_MAC_SPEED_100M, HCLGE_SUPPORT_100M_BIT},
+	{HCLGE_MAC_SPEED_1G, HCLGE_SUPPORT_1G_BIT},
+	{HCLGE_MAC_SPEED_10G, HCLGE_SUPPORT_10G_BIT},
+	{HCLGE_MAC_SPEED_25G, HCLGE_SUPPORT_25G_BIT},
+	{HCLGE_MAC_SPEED_40G, HCLGE_SUPPORT_40G_BIT},
+	{HCLGE_MAC_SPEED_50G, HCLGE_SUPPORT_50G_BITS},
+	{HCLGE_MAC_SPEED_100G, HCLGE_SUPPORT_100G_BITS},
+	{HCLGE_MAC_SPEED_200G, HCLGE_SUPPORT_200G_BITS},
+};
+
+static int hclge_get_speed_bit(u32 speed, u32 *speed_bit)
+{
+	u16 i;
+
+	for (i = 0; i < ARRAY_SIZE(speed_bit_map); i++) {
+		if (speed == speed_bit_map[i].speed) {
+			*speed_bit = speed_bit_map[i].speed_bit;
+			return 0;
+		}
+	}
+
+	return -EINVAL;
+}
+
+static int hclge_check_port_speed(struct hnae3_handle *handle, u32 speed)
+{
+	struct hclge_vport *vport = hclge_get_vport(handle);
+	struct hclge_dev *hdev = vport->back;
+	u32 speed_ability = hdev->hw.mac.speed_ability;
+	u32 speed_bit = 0;
+	int ret;
+
+	ret = hclge_get_speed_bit(speed, &speed_bit);
+	if (ret)
+		return ret;
+
+	if (speed_bit & speed_ability)
+		return 0;
+
+	return -EINVAL;
+}
+
+static void hclge_update_fec_support(struct hclge_mac *mac)
+{
+	linkmode_clear_bit(ETHTOOL_LINK_MODE_FEC_BASER_BIT, mac->supported);
+	linkmode_clear_bit(ETHTOOL_LINK_MODE_FEC_RS_BIT, mac->supported);
+	linkmode_clear_bit(ETHTOOL_LINK_MODE_FEC_LLRS_BIT, mac->supported);
+	linkmode_clear_bit(ETHTOOL_LINK_MODE_FEC_NONE_BIT, mac->supported);
+
+	if (mac->fec_ability & BIT(HNAE3_FEC_BASER))
+		linkmode_set_bit(ETHTOOL_LINK_MODE_FEC_BASER_BIT,
+				 mac->supported);
+	if (mac->fec_ability & BIT(HNAE3_FEC_RS))
+		linkmode_set_bit(ETHTOOL_LINK_MODE_FEC_RS_BIT,
+				 mac->supported);
+	if (mac->fec_ability & BIT(HNAE3_FEC_LLRS))
+		linkmode_set_bit(ETHTOOL_LINK_MODE_FEC_LLRS_BIT,
+				 mac->supported);
+	if (mac->fec_ability & BIT(HNAE3_FEC_NONE))
+		linkmode_set_bit(ETHTOOL_LINK_MODE_FEC_NONE_BIT,
+				 mac->supported);
+}
+
+static const struct hclge_link_mode_bmap hclge_sr_link_mode_bmap[] = {
+	{HCLGE_SUPPORT_10G_BIT, ETHTOOL_LINK_MODE_10000baseSR_Full_BIT},
+	{HCLGE_SUPPORT_25G_BIT, ETHTOOL_LINK_MODE_25000baseSR_Full_BIT},
+	{HCLGE_SUPPORT_40G_BIT, ETHTOOL_LINK_MODE_40000baseSR4_Full_BIT},
+	{HCLGE_SUPPORT_50G_R2_BIT, ETHTOOL_LINK_MODE_50000baseSR2_Full_BIT},
+	{HCLGE_SUPPORT_50G_R1_BIT, ETHTOOL_LINK_MODE_50000baseSR_Full_BIT},
+	{HCLGE_SUPPORT_100G_R4_BIT, ETHTOOL_LINK_MODE_100000baseSR4_Full_BIT},
+	{HCLGE_SUPPORT_100G_R2_BIT, ETHTOOL_LINK_MODE_100000baseSR2_Full_BIT},
+	{HCLGE_SUPPORT_200G_R4_EXT_BIT,
+	 ETHTOOL_LINK_MODE_200000baseSR4_Full_BIT},
+	{HCLGE_SUPPORT_200G_R4_BIT, ETHTOOL_LINK_MODE_200000baseSR4_Full_BIT},
+};
+
+static const struct hclge_link_mode_bmap hclge_lr_link_mode_bmap[] = {
+	{HCLGE_SUPPORT_10G_BIT, ETHTOOL_LINK_MODE_10000baseLR_Full_BIT},
+	{HCLGE_SUPPORT_40G_BIT, ETHTOOL_LINK_MODE_40000baseLR4_Full_BIT},
+	{HCLGE_SUPPORT_50G_R1_BIT, ETHTOOL_LINK_MODE_50000baseLR_ER_FR_Full_BIT},
+	{HCLGE_SUPPORT_100G_R4_BIT,
+	 ETHTOOL_LINK_MODE_100000baseLR4_ER4_Full_BIT},
+	{HCLGE_SUPPORT_100G_R2_BIT,
+	 ETHTOOL_LINK_MODE_100000baseLR2_ER2_FR2_Full_BIT},
+	{HCLGE_SUPPORT_200G_R4_EXT_BIT,
+	 ETHTOOL_LINK_MODE_200000baseLR4_ER4_FR4_Full_BIT},
+	{HCLGE_SUPPORT_200G_R4_BIT,
+	 ETHTOOL_LINK_MODE_200000baseLR4_ER4_FR4_Full_BIT},
+};
+
+static const struct hclge_link_mode_bmap hclge_cr_link_mode_bmap[] = {
+	{HCLGE_SUPPORT_10G_BIT, ETHTOOL_LINK_MODE_10000baseCR_Full_BIT},
+	{HCLGE_SUPPORT_25G_BIT, ETHTOOL_LINK_MODE_25000baseCR_Full_BIT},
+	{HCLGE_SUPPORT_40G_BIT, ETHTOOL_LINK_MODE_40000baseCR4_Full_BIT},
+	{HCLGE_SUPPORT_50G_R2_BIT, ETHTOOL_LINK_MODE_50000baseCR2_Full_BIT},
+	{HCLGE_SUPPORT_50G_R1_BIT, ETHTOOL_LINK_MODE_50000baseCR_Full_BIT},
+	{HCLGE_SUPPORT_100G_R4_BIT, ETHTOOL_LINK_MODE_100000baseCR4_Full_BIT},
+	{HCLGE_SUPPORT_100G_R2_BIT, ETHTOOL_LINK_MODE_100000baseCR2_Full_BIT},
+	{HCLGE_SUPPORT_200G_R4_EXT_BIT,
+	 ETHTOOL_LINK_MODE_200000baseCR4_Full_BIT},
+	{HCLGE_SUPPORT_200G_R4_BIT, ETHTOOL_LINK_MODE_200000baseCR4_Full_BIT},
+};
+
+static const struct hclge_link_mode_bmap hclge_kr_link_mode_bmap[] = {
+	{HCLGE_SUPPORT_1G_BIT, ETHTOOL_LINK_MODE_1000baseKX_Full_BIT},
+	{HCLGE_SUPPORT_10G_BIT, ETHTOOL_LINK_MODE_10000baseKR_Full_BIT},
+	{HCLGE_SUPPORT_25G_BIT, ETHTOOL_LINK_MODE_25000baseKR_Full_BIT},
+	{HCLGE_SUPPORT_40G_BIT, ETHTOOL_LINK_MODE_40000baseKR4_Full_BIT},
+	{HCLGE_SUPPORT_50G_R2_BIT, ETHTOOL_LINK_MODE_50000baseKR2_Full_BIT},
+	{HCLGE_SUPPORT_50G_R1_BIT, ETHTOOL_LINK_MODE_50000baseKR_Full_BIT},
+	{HCLGE_SUPPORT_100G_R4_BIT, ETHTOOL_LINK_MODE_100000baseKR4_Full_BIT},
+	{HCLGE_SUPPORT_100G_R2_BIT, ETHTOOL_LINK_MODE_100000baseKR2_Full_BIT},
+	{HCLGE_SUPPORT_200G_R4_EXT_BIT,
+	 ETHTOOL_LINK_MODE_200000baseKR4_Full_BIT},
+	{HCLGE_SUPPORT_200G_R4_BIT, ETHTOOL_LINK_MODE_200000baseKR4_Full_BIT},
+};
+
+static void hclge_convert_setting_sr(u16 speed_ability,
+				     unsigned long *link_mode)
+{
+	int i;
+
+	for (i = 0; i < ARRAY_SIZE(hclge_sr_link_mode_bmap); i++) {
+		if (speed_ability & hclge_sr_link_mode_bmap[i].support_bit)
+			linkmode_set_bit(hclge_sr_link_mode_bmap[i].link_mode,
+					 link_mode);
+	}
+}
+
+static void hclge_convert_setting_lr(u16 speed_ability,
+				     unsigned long *link_mode)
+{
+	int i;
+
+	for (i = 0; i < ARRAY_SIZE(hclge_lr_link_mode_bmap); i++) {
+		if (speed_ability & hclge_lr_link_mode_bmap[i].support_bit)
+			linkmode_set_bit(hclge_lr_link_mode_bmap[i].link_mode,
+					 link_mode);
+	}
+}
+
+static void hclge_convert_setting_cr(u16 speed_ability,
+				     unsigned long *link_mode)
+{
+	int i;
+
+	for (i = 0; i < ARRAY_SIZE(hclge_cr_link_mode_bmap); i++) {
+		if (speed_ability & hclge_cr_link_mode_bmap[i].support_bit)
+			linkmode_set_bit(hclge_cr_link_mode_bmap[i].link_mode,
+					 link_mode);
+	}
+}
+
+static void hclge_convert_setting_kr(u16 speed_ability,
+				     unsigned long *link_mode)
+{
+	int i;
+
+	for (i = 0; i < ARRAY_SIZE(hclge_kr_link_mode_bmap); i++) {
+		if (speed_ability & hclge_kr_link_mode_bmap[i].support_bit)
+			linkmode_set_bit(hclge_kr_link_mode_bmap[i].link_mode,
+					 link_mode);
+	}
+}
+
+static void hclge_convert_setting_fec(struct hclge_mac *mac)
+{
+	/* If firmware has reported fec_ability, don't need to convert by speed */
+	if (mac->fec_ability)
+		goto out;
+
+	switch (mac->speed) {
+	case HCLGE_MAC_SPEED_10G:
+	case HCLGE_MAC_SPEED_40G:
+		mac->fec_ability = BIT(HNAE3_FEC_BASER) | BIT(HNAE3_FEC_AUTO) |
+				   BIT(HNAE3_FEC_NONE);
+		break;
+	case HCLGE_MAC_SPEED_25G:
+	case HCLGE_MAC_SPEED_50G:
+		mac->fec_ability = BIT(HNAE3_FEC_BASER) | BIT(HNAE3_FEC_RS) |
+				   BIT(HNAE3_FEC_AUTO) | BIT(HNAE3_FEC_NONE);
+		break;
+	case HCLGE_MAC_SPEED_100G:
+		mac->fec_ability = BIT(HNAE3_FEC_RS) | BIT(HNAE3_FEC_AUTO) |
+				   BIT(HNAE3_FEC_NONE);
+		break;
+	case HCLGE_MAC_SPEED_200G:
+		mac->fec_ability = BIT(HNAE3_FEC_RS) | BIT(HNAE3_FEC_AUTO) |
+				   BIT(HNAE3_FEC_LLRS);
+		break;
+	default:
+		mac->fec_ability = 0;
+		break;
+	}
+
+out:
+	hclge_update_fec_support(mac);
+}
+
+static void hclge_parse_fiber_link_mode(struct hclge_dev *hdev,
+					u16 speed_ability)
+{
+	struct hclge_mac *mac = &hdev->hw.mac;
+
+	if (speed_ability & HCLGE_SUPPORT_1G_BIT)
+		linkmode_set_bit(ETHTOOL_LINK_MODE_1000baseX_Full_BIT,
+				 mac->supported);
+
+	hclge_convert_setting_sr(speed_ability, mac->supported);
+	hclge_convert_setting_lr(speed_ability, mac->supported);
+	hclge_convert_setting_cr(speed_ability, mac->supported);
+	if (hnae3_dev_fec_supported(hdev))
+		hclge_convert_setting_fec(mac);
+
+	if (hnae3_dev_pause_supported(hdev))
+		linkmode_set_bit(ETHTOOL_LINK_MODE_Pause_BIT, mac->supported);
+
+	linkmode_set_bit(ETHTOOL_LINK_MODE_FIBRE_BIT, mac->supported);
+	linkmode_set_bit(ETHTOOL_LINK_MODE_FEC_NONE_BIT, mac->supported);
+}
+
+static void hclge_parse_backplane_link_mode(struct hclge_dev *hdev,
+					    u16 speed_ability)
+{
+	struct hclge_mac *mac = &hdev->hw.mac;
+
+	hclge_convert_setting_kr(speed_ability, mac->supported);
+	if (hnae3_dev_fec_supported(hdev))
+		hclge_convert_setting_fec(mac);
+
+	if (hnae3_dev_pause_supported(hdev))
+		linkmode_set_bit(ETHTOOL_LINK_MODE_Pause_BIT, mac->supported);
+
+	linkmode_set_bit(ETHTOOL_LINK_MODE_Backplane_BIT, mac->supported);
+	linkmode_set_bit(ETHTOOL_LINK_MODE_FEC_NONE_BIT, mac->supported);
+}
+
+static void hclge_parse_copper_link_mode(struct hclge_dev *hdev,
+					 u16 speed_ability)
+{
+	unsigned long *supported = hdev->hw.mac.supported;
+
+	/* default to support all speed for GE port */
+	if (!speed_ability)
+		speed_ability = HCLGE_SUPPORT_GE;
+
+	if (speed_ability & HCLGE_SUPPORT_1G_BIT)
+		linkmode_set_bit(ETHTOOL_LINK_MODE_1000baseT_Full_BIT,
+				 supported);
+
+	if (speed_ability & HCLGE_SUPPORT_100M_BIT) {
+		linkmode_set_bit(ETHTOOL_LINK_MODE_100baseT_Full_BIT,
+				 supported);
+		linkmode_set_bit(ETHTOOL_LINK_MODE_100baseT_Half_BIT,
+				 supported);
+	}
+
+	if (speed_ability & HCLGE_SUPPORT_10M_BIT) {
+		linkmode_set_bit(ETHTOOL_LINK_MODE_10baseT_Full_BIT, supported);
+		linkmode_set_bit(ETHTOOL_LINK_MODE_10baseT_Half_BIT, supported);
+	}
+
+	if (hnae3_dev_pause_supported(hdev)) {
+		linkmode_set_bit(ETHTOOL_LINK_MODE_Pause_BIT, supported);
+		linkmode_set_bit(ETHTOOL_LINK_MODE_Asym_Pause_BIT, supported);
+	}
+
+	linkmode_set_bit(ETHTOOL_LINK_MODE_Autoneg_BIT, supported);
+	linkmode_set_bit(ETHTOOL_LINK_MODE_TP_BIT, supported);
+}
+
+static void hclge_parse_link_mode(struct hclge_dev *hdev, u16 speed_ability)
+{
+	u8 media_type = hdev->hw.mac.media_type;
+
+	if (media_type == HNAE3_MEDIA_TYPE_FIBER)
+		hclge_parse_fiber_link_mode(hdev, speed_ability);
+	else if (media_type == HNAE3_MEDIA_TYPE_COPPER)
+		hclge_parse_copper_link_mode(hdev, speed_ability);
+	else if (media_type == HNAE3_MEDIA_TYPE_BACKPLANE)
+		hclge_parse_backplane_link_mode(hdev, speed_ability);
+}
+
+static u32 hclge_get_max_speed(u16 speed_ability)
+{
+	if (speed_ability & HCLGE_SUPPORT_200G_BITS)
+		return HCLGE_MAC_SPEED_200G;
+
+	if (speed_ability & HCLGE_SUPPORT_100G_BITS)
+		return HCLGE_MAC_SPEED_100G;
+
+	if (speed_ability & HCLGE_SUPPORT_50G_BITS)
+		return HCLGE_MAC_SPEED_50G;
+
+	if (speed_ability & HCLGE_SUPPORT_40G_BIT)
+		return HCLGE_MAC_SPEED_40G;
+
+	if (speed_ability & HCLGE_SUPPORT_25G_BIT)
+		return HCLGE_MAC_SPEED_25G;
+
+	if (speed_ability & HCLGE_SUPPORT_10G_BIT)
+		return HCLGE_MAC_SPEED_10G;
+
+	if (speed_ability & HCLGE_SUPPORT_1G_BIT)
+		return HCLGE_MAC_SPEED_1G;
+
+	if (speed_ability & HCLGE_SUPPORT_100M_BIT)
+		return HCLGE_MAC_SPEED_100M;
+
+	if (speed_ability & HCLGE_SUPPORT_10M_BIT)
+		return HCLGE_MAC_SPEED_10M;
+
+	return HCLGE_MAC_SPEED_1G;
+}
+
+static void hclge_parse_cfg(struct hclge_cfg *cfg, struct hclge_desc *desc)
+{
+#define HCLGE_TX_SPARE_SIZE_UNIT		4096
+#define SPEED_ABILITY_EXT_SHIFT			8
+
+	struct hclge_cfg_param_cmd *req;
+	u64 mac_addr_tmp_high;
+	u16 speed_ability_ext;
+	u64 mac_addr_tmp;
+	unsigned int i;
+
+	req = (struct hclge_cfg_param_cmd *)desc[0].data;
+
+	/* get the configuration */
+	cfg->tc_num = hnae3_get_field(__le32_to_cpu(req->param[0]),
+				      HCLGE_CFG_TC_NUM_M, HCLGE_CFG_TC_NUM_S);
+	cfg->tqp_desc_num = hnae3_get_field(__le32_to_cpu(req->param[0]),
+					    HCLGE_CFG_TQP_DESC_N_M,
+					    HCLGE_CFG_TQP_DESC_N_S);
+
+	cfg->phy_addr = hnae3_get_field(__le32_to_cpu(req->param[1]),
+					HCLGE_CFG_PHY_ADDR_M,
+					HCLGE_CFG_PHY_ADDR_S);
+	cfg->media_type = hnae3_get_field(__le32_to_cpu(req->param[1]),
+					  HCLGE_CFG_MEDIA_TP_M,
+					  HCLGE_CFG_MEDIA_TP_S);
+	cfg->rx_buf_len = hnae3_get_field(__le32_to_cpu(req->param[1]),
+					  HCLGE_CFG_RX_BUF_LEN_M,
+					  HCLGE_CFG_RX_BUF_LEN_S);
+	/* get mac_address */
+	mac_addr_tmp = __le32_to_cpu(req->param[2]);
+	mac_addr_tmp_high = hnae3_get_field(__le32_to_cpu(req->param[3]),
+					    HCLGE_CFG_MAC_ADDR_H_M,
+					    HCLGE_CFG_MAC_ADDR_H_S);
+
+	mac_addr_tmp |= (mac_addr_tmp_high << 31) << 1;
+
+	cfg->default_speed = hnae3_get_field(__le32_to_cpu(req->param[3]),
+					     HCLGE_CFG_DEFAULT_SPEED_M,
+					     HCLGE_CFG_DEFAULT_SPEED_S);
+	cfg->vf_rss_size_max = hnae3_get_field(__le32_to_cpu(req->param[3]),
+					       HCLGE_CFG_RSS_SIZE_M,
+					       HCLGE_CFG_RSS_SIZE_S);
+
+	for (i = 0; i < ETH_ALEN; i++)
+		cfg->mac_addr[i] = (mac_addr_tmp >> (8 * i)) & 0xff;
+
+	req = (struct hclge_cfg_param_cmd *)desc[1].data;
+	cfg->numa_node_map = __le32_to_cpu(req->param[0]);
+
+	cfg->speed_ability = hnae3_get_field(__le32_to_cpu(req->param[1]),
+					     HCLGE_CFG_SPEED_ABILITY_M,
+					     HCLGE_CFG_SPEED_ABILITY_S);
+	speed_ability_ext = hnae3_get_field(__le32_to_cpu(req->param[1]),
+					    HCLGE_CFG_SPEED_ABILITY_EXT_M,
+					    HCLGE_CFG_SPEED_ABILITY_EXT_S);
+	cfg->speed_ability |= speed_ability_ext << SPEED_ABILITY_EXT_SHIFT;
+
+	cfg->vlan_fliter_cap = hnae3_get_field(__le32_to_cpu(req->param[1]),
+					       HCLGE_CFG_VLAN_FLTR_CAP_M,
+					       HCLGE_CFG_VLAN_FLTR_CAP_S);
+
+	cfg->umv_space = hnae3_get_field(__le32_to_cpu(req->param[1]),
+					 HCLGE_CFG_UMV_TBL_SPACE_M,
+					 HCLGE_CFG_UMV_TBL_SPACE_S);
+
+	cfg->pf_rss_size_max = hnae3_get_field(__le32_to_cpu(req->param[2]),
+					       HCLGE_CFG_PF_RSS_SIZE_M,
+					       HCLGE_CFG_PF_RSS_SIZE_S);
+
+	/* HCLGE_CFG_PF_RSS_SIZE_M is the PF max rss size, which is a
+	 * power of 2, instead of reading out directly. This would
+	 * be more flexible for future changes and expansions.
+	 * When VF max  rss size field is HCLGE_CFG_RSS_SIZE_S,
+	 * it does not make sense if PF's field is 0. In this case, PF and VF
+	 * has the same max rss size filed: HCLGE_CFG_RSS_SIZE_S.
+	 */
+	cfg->pf_rss_size_max = cfg->pf_rss_size_max ?
+			       1U << cfg->pf_rss_size_max :
+			       cfg->vf_rss_size_max;
+
+	/* The unit of the tx spare buffer size queried from configuration
+	 * file is HCLGE_TX_SPARE_SIZE_UNIT(4096) bytes, so a conversion is
+	 * needed here.
+	 */
+	cfg->tx_spare_buf_size = hnae3_get_field(__le32_to_cpu(req->param[2]),
+						 HCLGE_CFG_TX_SPARE_BUF_SIZE_M,
+						 HCLGE_CFG_TX_SPARE_BUF_SIZE_S);
+	cfg->tx_spare_buf_size *= HCLGE_TX_SPARE_SIZE_UNIT;
+}
+
+/* hclge_get_cfg: query the static parameter from flash
+ * @hdev: pointer to struct hclge_dev
+ * @hcfg: the config structure to be getted
+ */
+static int hclge_get_cfg(struct hclge_dev *hdev, struct hclge_cfg *hcfg)
+{
+	struct hclge_desc desc[HCLGE_PF_CFG_DESC_NUM];
+	struct hclge_cfg_param_cmd *req;
+	unsigned int i;
+	int ret;
+
+	for (i = 0; i < HCLGE_PF_CFG_DESC_NUM; i++) {
+		u32 offset = 0;
+
+		req = (struct hclge_cfg_param_cmd *)desc[i].data;
+		hclge_cmd_setup_basic_desc(&desc[i], HCLGE_OPC_GET_CFG_PARAM,
+					   true);
+		hnae3_set_field(offset, HCLGE_CFG_OFFSET_M,
+				HCLGE_CFG_OFFSET_S, i * HCLGE_CFG_RD_LEN_BYTES);
+		/* Len should be united by 4 bytes when send to hardware */
+		hnae3_set_field(offset, HCLGE_CFG_RD_LEN_M, HCLGE_CFG_RD_LEN_S,
+				HCLGE_CFG_RD_LEN_BYTES / HCLGE_CFG_RD_LEN_UNIT);
+		req->offset = cpu_to_le32(offset);
+	}
+
+	ret = hclge_cmd_send(&hdev->hw, desc, HCLGE_PF_CFG_DESC_NUM);
+	if (ret) {
+		dev_err(&hdev->pdev->dev, "get config failed %d.\n", ret);
+		return ret;
+	}
+
+	hclge_parse_cfg(hcfg, desc);
+
+	return 0;
+}
+
+static void hclge_set_default_dev_specs(struct hclge_dev *hdev)
+{
+#define HCLGE_MAX_NON_TSO_BD_NUM			8U
+
+	struct hnae3_ae_dev *ae_dev = pci_get_drvdata(hdev->pdev);
+
+	ae_dev->dev_specs.max_non_tso_bd_num = HCLGE_MAX_NON_TSO_BD_NUM;
+	ae_dev->dev_specs.rss_ind_tbl_size = HCLGE_RSS_IND_TBL_SIZE;
+	ae_dev->dev_specs.rss_key_size = HCLGE_COMM_RSS_KEY_SIZE;
+	ae_dev->dev_specs.max_tm_rate = HCLGE_ETHER_MAX_RATE;
+	ae_dev->dev_specs.max_int_gl = HCLGE_DEF_MAX_INT_GL;
+	ae_dev->dev_specs.max_frm_size = HCLGE_MAC_MAX_FRAME;
+	ae_dev->dev_specs.max_qset_num = HCLGE_MAX_QSET_NUM;
+	ae_dev->dev_specs.umv_size = HCLGE_DEFAULT_UMV_SPACE_PER_PF;
+	ae_dev->dev_specs.tnl_num = 0;
+}
+
+static void hclge_parse_dev_specs(struct hclge_dev *hdev,
+				  struct hclge_desc *desc)
+{
+	struct hnae3_ae_dev *ae_dev = pci_get_drvdata(hdev->pdev);
+	struct hclge_dev_specs_0_cmd *req0;
+	struct hclge_dev_specs_1_cmd *req1;
+
+	req0 = (struct hclge_dev_specs_0_cmd *)desc[0].data;
+	req1 = (struct hclge_dev_specs_1_cmd *)desc[1].data;
+
+	ae_dev->dev_specs.max_non_tso_bd_num = req0->max_non_tso_bd_num;
+	ae_dev->dev_specs.rss_ind_tbl_size =
+		le16_to_cpu(req0->rss_ind_tbl_size);
+	ae_dev->dev_specs.int_ql_max = le16_to_cpu(req0->int_ql_max);
+	ae_dev->dev_specs.rss_key_size = le16_to_cpu(req0->rss_key_size);
+	ae_dev->dev_specs.max_tm_rate = le32_to_cpu(req0->max_tm_rate);
+	ae_dev->dev_specs.max_qset_num = le16_to_cpu(req1->max_qset_num);
+	ae_dev->dev_specs.max_int_gl = le16_to_cpu(req1->max_int_gl);
+	ae_dev->dev_specs.max_frm_size = le16_to_cpu(req1->max_frm_size);
+	ae_dev->dev_specs.umv_size = le16_to_cpu(req1->umv_size);
+	ae_dev->dev_specs.mc_mac_size = le16_to_cpu(req1->mc_mac_size);
+	ae_dev->dev_specs.tnl_num = req1->tnl_num;
+	ae_dev->dev_specs.hilink_version = req1->hilink_version;
+}
+
+static void hclge_check_dev_specs(struct hclge_dev *hdev)
+{
+	struct hnae3_dev_specs *dev_specs = &hdev->ae_dev->dev_specs;
+
+	if (!dev_specs->max_non_tso_bd_num)
+		dev_specs->max_non_tso_bd_num = HCLGE_MAX_NON_TSO_BD_NUM;
+	if (!dev_specs->rss_ind_tbl_size)
+		dev_specs->rss_ind_tbl_size = HCLGE_RSS_IND_TBL_SIZE;
+	if (!dev_specs->rss_key_size)
+		dev_specs->rss_key_size = HCLGE_COMM_RSS_KEY_SIZE;
+	if (!dev_specs->max_tm_rate)
+		dev_specs->max_tm_rate = HCLGE_ETHER_MAX_RATE;
+	if (!dev_specs->max_qset_num)
+		dev_specs->max_qset_num = HCLGE_MAX_QSET_NUM;
+	if (!dev_specs->max_int_gl)
+		dev_specs->max_int_gl = HCLGE_DEF_MAX_INT_GL;
+	if (!dev_specs->max_frm_size)
+		dev_specs->max_frm_size = HCLGE_MAC_MAX_FRAME;
+	if (!dev_specs->umv_size)
+		dev_specs->umv_size = HCLGE_DEFAULT_UMV_SPACE_PER_PF;
+}
+
+static int hclge_query_mac_stats_num(struct hclge_dev *hdev)
+{
+	u32 reg_num = 0;
+	int ret;
+
+	ret = hclge_mac_query_reg_num(hdev, &reg_num);
+	if (ret && ret != -EOPNOTSUPP)
+		return ret;
+
+	hdev->ae_dev->dev_specs.mac_stats_num = reg_num;
+	return 0;
+}
+
+static int hclge_query_dev_specs(struct hclge_dev *hdev)
+{
+	struct hclge_desc desc[HCLGE_QUERY_DEV_SPECS_BD_NUM];
+	int ret;
+	int i;
+
+	ret = hclge_query_mac_stats_num(hdev);
+	if (ret)
+		return ret;
+
+	/* set default specifications as devices lower than version V3 do not
+	 * support querying specifications from firmware.
+	 */
+	if (hdev->ae_dev->dev_version < HNAE3_DEVICE_VERSION_V3) {
+		hclge_set_default_dev_specs(hdev);
+		return 0;
+	}
+
+	for (i = 0; i < HCLGE_QUERY_DEV_SPECS_BD_NUM - 1; i++) {
+		hclge_cmd_setup_basic_desc(&desc[i], HCLGE_OPC_QUERY_DEV_SPECS,
+					   true);
+		desc[i].flag |= cpu_to_le16(HCLGE_COMM_CMD_FLAG_NEXT);
+	}
+	hclge_cmd_setup_basic_desc(&desc[i], HCLGE_OPC_QUERY_DEV_SPECS, true);
+
+	ret = hclge_cmd_send(&hdev->hw, desc, HCLGE_QUERY_DEV_SPECS_BD_NUM);
+	if (ret)
+		return ret;
+
+	hclge_parse_dev_specs(hdev, desc);
+	hclge_check_dev_specs(hdev);
+
+	return 0;
+}
+
+static int hclge_get_cap(struct hclge_dev *hdev)
+{
+	int ret;
+
+	ret = hclge_query_function_status(hdev);
+	if (ret) {
+		dev_err(&hdev->pdev->dev,
+			"query function status error %d.\n", ret);
+		return ret;
+	}
+
+	/* get pf resource */
+	return hclge_query_pf_resource(hdev);
+}
+
+static void hclge_init_kdump_kernel_config(struct hclge_dev *hdev)
+{
+#define HCLGE_MIN_TX_DESC	64
+#define HCLGE_MIN_RX_DESC	64
+
+	if (!is_kdump_kernel())
+		return;
+
+	dev_info(&hdev->pdev->dev,
+		 "Running kdump kernel. Using minimal resources\n");
+
+	/* minimal queue pairs equals to the number of vports */
+	hdev->num_tqps = hdev->num_req_vfs + 1;
+	hdev->num_tx_desc = HCLGE_MIN_TX_DESC;
+	hdev->num_rx_desc = HCLGE_MIN_RX_DESC;
+}
+
+static void hclge_init_tc_config(struct hclge_dev *hdev)
+{
+	unsigned int i;
+
+	if (hdev->tc_max > HNAE3_MAX_TC ||
+	    hdev->tc_max < 1) {
+		dev_warn(&hdev->pdev->dev, "TC num = %u.\n",
+			 hdev->tc_max);
+		hdev->tc_max = 1;
+	}
+
+	/* Dev does not support DCB */
+	if (!hnae3_dev_dcb_supported(hdev)) {
+		hdev->tc_max = 1;
+		hdev->pfc_max = 0;
+	} else {
+		hdev->pfc_max = hdev->tc_max;
+	}
+
+	hdev->tm_info.num_tc = 1;
+
+	/* Currently not support uncontiuous tc */
+	for (i = 0; i < hdev->tm_info.num_tc; i++)
+		hnae3_set_bit(hdev->hw_tc_map, i, 1);
+
+	hdev->tx_sch_mode = HCLGE_FLAG_TC_BASE_SCH_MODE;
+}
+
+static int hclge_configure(struct hclge_dev *hdev)
+{
+	struct hnae3_ae_dev *ae_dev = pci_get_drvdata(hdev->pdev);
+	struct hclge_cfg cfg;
+	int ret;
+
+	ret = hclge_get_cfg(hdev, &cfg);
+	if (ret)
+		return ret;
+
+	hdev->base_tqp_pid = 0;
+	hdev->vf_rss_size_max = cfg.vf_rss_size_max;
+	hdev->pf_rss_size_max = cfg.pf_rss_size_max;
+	hdev->rx_buf_len = cfg.rx_buf_len;
+	ether_addr_copy(hdev->hw.mac.mac_addr, cfg.mac_addr);
+	hdev->hw.mac.media_type = cfg.media_type;
+	hdev->hw.mac.phy_addr = cfg.phy_addr;
+	hdev->num_tx_desc = cfg.tqp_desc_num;
+	hdev->num_rx_desc = cfg.tqp_desc_num;
+	hdev->tm_info.num_pg = 1;
+	hdev->tc_max = cfg.tc_num;
+	hdev->tm_info.hw_pfc_map = 0;
+	if (cfg.umv_space)
+		hdev->wanted_umv_size = cfg.umv_space;
+	else
+		hdev->wanted_umv_size = hdev->ae_dev->dev_specs.umv_size;
+	hdev->tx_spare_buf_size = cfg.tx_spare_buf_size;
+	hdev->gro_en = true;
+	if (cfg.vlan_fliter_cap == HCLGE_VLAN_FLTR_CAN_MDF)
+		set_bit(HNAE3_DEV_SUPPORT_VLAN_FLTR_MDF_B, ae_dev->caps);
+
+	if (hnae3_ae_dev_fd_supported(hdev->ae_dev)) {
+		hdev->fd_en = true;
+		hdev->fd_active_type = HCLGE_FD_RULE_NONE;
+	}
+
+	ret = hclge_parse_speed(cfg.default_speed, &hdev->hw.mac.speed);
+	if (ret) {
+		dev_err(&hdev->pdev->dev, "failed to parse speed %u, ret = %d\n",
+			cfg.default_speed, ret);
+		return ret;
+	}
+	hdev->hw.mac.req_speed = hdev->hw.mac.speed;
+	hdev->hw.mac.req_autoneg = AUTONEG_ENABLE;
+	hdev->hw.mac.req_duplex = DUPLEX_FULL;
+
+	hclge_parse_link_mode(hdev, cfg.speed_ability);
+
+	hdev->hw.mac.max_speed = hclge_get_max_speed(cfg.speed_ability);
+
+	hclge_init_tc_config(hdev);
+	hclge_init_kdump_kernel_config(hdev);
+
+	return ret;
+}
+
+static int hclge_config_tso(struct hclge_dev *hdev, u16 tso_mss_min,
+			    u16 tso_mss_max)
+{
+	struct hclge_cfg_tso_status_cmd *req;
+	struct hclge_desc desc;
+
+	hclge_cmd_setup_basic_desc(&desc, HCLGE_OPC_TSO_GENERIC_CONFIG, false);
+
+	req = (struct hclge_cfg_tso_status_cmd *)desc.data;
+	req->tso_mss_min = cpu_to_le16(tso_mss_min);
+	req->tso_mss_max = cpu_to_le16(tso_mss_max);
+
+	return hclge_cmd_send(&hdev->hw, &desc, 1);
+}
+
+static int hclge_config_gro(struct hclge_dev *hdev)
+{
+	struct hclge_cfg_gro_status_cmd *req;
+	struct hclge_desc desc;
+	int ret;
+
+	if (!hnae3_ae_dev_gro_supported(hdev->ae_dev))
+		return 0;
+
+	hclge_cmd_setup_basic_desc(&desc, HCLGE_OPC_GRO_GENERIC_CONFIG, false);
+	req = (struct hclge_cfg_gro_status_cmd *)desc.data;
+
+	req->gro_en = hdev->gro_en ? 1 : 0;
+
+	ret = hclge_cmd_send(&hdev->hw, &desc, 1);
+	if (ret)
+		dev_err(&hdev->pdev->dev,
+			"GRO hardware config cmd failed, ret = %d\n", ret);
+
+	return ret;
+}
+
+static int hclge_alloc_tqps(struct hclge_dev *hdev)
+{
+	struct hnae3_ae_dev *ae_dev = pci_get_drvdata(hdev->pdev);
+	struct hclge_comm_tqp *tqp;
+	int i;
+
+	hdev->htqp = devm_kcalloc(&hdev->pdev->dev, hdev->num_tqps,
+				  sizeof(struct hclge_comm_tqp), GFP_KERNEL);
+	if (!hdev->htqp)
+		return -ENOMEM;
+
+	tqp = hdev->htqp;
+
+	for (i = 0; i < hdev->num_tqps; i++) {
+		tqp->dev = &hdev->pdev->dev;
+		tqp->index = i;
+
+		tqp->q.ae_algo = &ae_algo;
+		tqp->q.buf_size = hdev->rx_buf_len;
+		tqp->q.tx_desc_num = hdev->num_tx_desc;
+		tqp->q.rx_desc_num = hdev->num_rx_desc;
+
+		/* need an extended offset to configure queues >=
+		 * HCLGE_TQP_MAX_SIZE_DEV_V2
+		 */
+		if (i < HCLGE_TQP_MAX_SIZE_DEV_V2)
+			tqp->q.io_base = hdev->hw.hw.io_base +
+					 HCLGE_TQP_REG_OFFSET +
+					 i * HCLGE_TQP_REG_SIZE;
+		else
+			tqp->q.io_base = hdev->hw.hw.io_base +
+					 HCLGE_TQP_REG_OFFSET +
+					 HCLGE_TQP_EXT_REG_OFFSET +
+					 (i - HCLGE_TQP_MAX_SIZE_DEV_V2) *
+					 HCLGE_TQP_REG_SIZE;
+
+		/* when device supports tx push and has device memory,
+		 * the queue can execute push mode or doorbell mode on
+		 * device memory.
+		 */
+		if (test_bit(HNAE3_DEV_SUPPORT_TX_PUSH_B, ae_dev->caps))
+			tqp->q.mem_base = hdev->hw.hw.mem_base +
+					  HCLGE_TQP_MEM_OFFSET(hdev, i);
+
+		tqp++;
+	}
+
+	return 0;
+}
+
+static int hclge_map_tqps_to_func(struct hclge_dev *hdev, u16 func_id,
+				  u16 tqp_pid, u16 tqp_vid, bool is_pf)
+{
+	struct hclge_tqp_map_cmd *req;
+	struct hclge_desc desc;
+	int ret;
+
+	hclge_cmd_setup_basic_desc(&desc, HCLGE_OPC_SET_TQP_MAP, false);
+
+	req = (struct hclge_tqp_map_cmd *)desc.data;
+	req->tqp_id = cpu_to_le16(tqp_pid);
+	req->tqp_vf = func_id;
+	req->tqp_flag = 1U << HCLGE_TQP_MAP_EN_B;
+	if (!is_pf)
+		req->tqp_flag |= 1U << HCLGE_TQP_MAP_TYPE_B;
+	req->tqp_vid = cpu_to_le16(tqp_vid);
+
+	ret = hclge_cmd_send(&hdev->hw, &desc, 1);
+	if (ret)
+		dev_err(&hdev->pdev->dev, "TQP map failed %d.\n", ret);
+
+	return ret;
+}
+
+static int  hclge_assign_tqp(struct hclge_vport *vport, u16 num_tqps)
+{
+	struct hnae3_knic_private_info *kinfo = &vport->nic.kinfo;
+	struct hclge_dev *hdev = vport->back;
+	int i, alloced;
+
+	for (i = 0, alloced = 0; i < hdev->num_tqps &&
+	     alloced < num_tqps; i++) {
+		if (!hdev->htqp[i].alloced) {
+			hdev->htqp[i].q.handle = &vport->nic;
+			hdev->htqp[i].q.tqp_index = alloced;
+			hdev->htqp[i].q.tx_desc_num = kinfo->num_tx_desc;
+			hdev->htqp[i].q.rx_desc_num = kinfo->num_rx_desc;
+			kinfo->tqp[alloced] = &hdev->htqp[i].q;
+			hdev->htqp[i].alloced = true;
+			alloced++;
+		}
+	}
+	vport->alloc_tqps = alloced;
+	kinfo->rss_size = min_t(u16, hdev->pf_rss_size_max,
+				vport->alloc_tqps / hdev->tm_info.num_tc);
+
+	/* ensure one to one mapping between irq and queue at default */
+	kinfo->rss_size = min_t(u16, kinfo->rss_size,
+				(hdev->num_nic_msi - 1) / hdev->tm_info.num_tc);
+
+	return 0;
+}
+
+static int hclge_knic_setup(struct hclge_vport *vport, u16 num_tqps,
+			    u16 num_tx_desc, u16 num_rx_desc)
+
+{
+	struct hnae3_handle *nic = &vport->nic;
+	struct hnae3_knic_private_info *kinfo = &nic->kinfo;
+	struct hclge_dev *hdev = vport->back;
+	int ret;
+
+	kinfo->num_tx_desc = num_tx_desc;
+	kinfo->num_rx_desc = num_rx_desc;
+
+	kinfo->rx_buf_len = hdev->rx_buf_len;
+	kinfo->tx_spare_buf_size = hdev->tx_spare_buf_size;
+
+	kinfo->tqp = devm_kcalloc(&hdev->pdev->dev, num_tqps,
+				  sizeof(struct hnae3_queue *), GFP_KERNEL);
+	if (!kinfo->tqp)
+		return -ENOMEM;
+
+	ret = hclge_assign_tqp(vport, num_tqps);
+	if (ret)
+		dev_err(&hdev->pdev->dev, "fail to assign TQPs %d.\n", ret);
+
+	return ret;
+}
+
+static int hclge_map_tqp_to_vport(struct hclge_dev *hdev,
+				  struct hclge_vport *vport)
+{
+	struct hnae3_handle *nic = &vport->nic;
+	struct hnae3_knic_private_info *kinfo;
+	u16 i;
+
+	kinfo = &nic->kinfo;
+	for (i = 0; i < vport->alloc_tqps; i++) {
+		struct hclge_comm_tqp *q =
+			container_of(kinfo->tqp[i], struct hclge_comm_tqp, q);
+		bool is_pf;
+		int ret;
+
+		is_pf = !(vport->vport_id);
+		ret = hclge_map_tqps_to_func(hdev, vport->vport_id, q->index,
+					     i, is_pf);
+		if (ret)
+			return ret;
+	}
+
+	return 0;
+}
+
+static int hclge_map_tqp(struct hclge_dev *hdev)
+{
+	struct hclge_vport *vport = hdev->vport;
+	u16 i, num_vport;
+
+	num_vport = hdev->num_req_vfs + 1;
+	for (i = 0; i < num_vport; i++) {
+		int ret;
+
+		ret = hclge_map_tqp_to_vport(hdev, vport);
+		if (ret)
+			return ret;
+
+		vport++;
+	}
+
+	return 0;
+}
+
+static int hclge_vport_setup(struct hclge_vport *vport, u16 num_tqps)
+{
+	struct hnae3_handle *nic = &vport->nic;
+	struct hclge_dev *hdev = vport->back;
+	int ret;
+
+	nic->pdev = hdev->pdev;
+	nic->ae_algo = &ae_algo;
+	bitmap_copy(nic->numa_node_mask.bits, hdev->numa_node_mask.bits,
+		    MAX_NUMNODES);
+	nic->kinfo.io_base = hdev->hw.hw.io_base;
+
+	ret = hclge_knic_setup(vport, num_tqps,
+			       hdev->num_tx_desc, hdev->num_rx_desc);
+	if (ret)
+		dev_err(&hdev->pdev->dev, "knic setup failed %d\n", ret);
+
+	return ret;
+}
+
+static int hclge_alloc_vport(struct hclge_dev *hdev)
+{
+	struct pci_dev *pdev = hdev->pdev;
+	struct hclge_vport *vport;
+	u32 tqp_main_vport;
+	u32 tqp_per_vport;
+	int num_vport, i;
+	int ret;
+
+	/* We need to alloc a vport for main NIC of PF */
+	num_vport = hdev->num_req_vfs + 1;
+
+	if (hdev->num_tqps < num_vport) {
+		dev_err(&hdev->pdev->dev, "tqps(%u) is less than vports(%d)",
+			hdev->num_tqps, num_vport);
+		return -EINVAL;
+	}
+
+	/* Alloc the same number of TQPs for every vport */
+	tqp_per_vport = hdev->num_tqps / num_vport;
+	tqp_main_vport = tqp_per_vport + hdev->num_tqps % num_vport;
+
+	vport = devm_kcalloc(&pdev->dev, num_vport, sizeof(struct hclge_vport),
+			     GFP_KERNEL);
+	if (!vport)
+		return -ENOMEM;
+
+	hdev->vport = vport;
+	hdev->num_alloc_vport = num_vport;
+
+	if (IS_ENABLED(CONFIG_PCI_IOV))
+		hdev->num_alloc_vfs = hdev->num_req_vfs;
+
+	for (i = 0; i < num_vport; i++) {
+		vport->back = hdev;
+		vport->vport_id = i;
+		vport->vf_info.link_state = IFLA_VF_LINK_STATE_AUTO;
+		vport->mps = HCLGE_MAC_DEFAULT_FRAME;
+		vport->port_base_vlan_cfg.state = HNAE3_PORT_BASE_VLAN_DISABLE;
+		vport->port_base_vlan_cfg.tbl_sta = true;
+		vport->rxvlan_cfg.rx_vlan_offload_en = true;
+		vport->req_vlan_fltr_en = true;
+		INIT_LIST_HEAD(&vport->vlan_list);
+		INIT_LIST_HEAD(&vport->uc_mac_list);
+		INIT_LIST_HEAD(&vport->mc_mac_list);
+		spin_lock_init(&vport->mac_list_lock);
+
+		if (i == 0)
+			ret = hclge_vport_setup(vport, tqp_main_vport);
+		else
+			ret = hclge_vport_setup(vport, tqp_per_vport);
+		if (ret) {
+			dev_err(&pdev->dev,
+				"vport setup failed for vport %d, %d\n",
+				i, ret);
+			return ret;
+		}
+
+		vport++;
+	}
+
+	return 0;
+}
+
+static int  hclge_cmd_alloc_tx_buff(struct hclge_dev *hdev,
+				    struct hclge_pkt_buf_alloc *buf_alloc)
+{
+/* TX buffer size is unit by 128 byte */
+#define HCLGE_BUF_SIZE_UNIT_SHIFT	7
+#define HCLGE_BUF_SIZE_UPDATE_EN_MSK	BIT(15)
+	struct hclge_tx_buff_alloc_cmd *req;
+	struct hclge_desc desc;
+	int ret;
+	u8 i;
+
+	req = (struct hclge_tx_buff_alloc_cmd *)desc.data;
+
+	hclge_cmd_setup_basic_desc(&desc, HCLGE_OPC_TX_BUFF_ALLOC, 0);
+	for (i = 0; i < HCLGE_MAX_TC_NUM; i++) {
+		u32 buf_size = buf_alloc->priv_buf[i].tx_buf_size;
+
+		req->tx_pkt_buff[i] =
+			cpu_to_le16((buf_size >> HCLGE_BUF_SIZE_UNIT_SHIFT) |
+				     HCLGE_BUF_SIZE_UPDATE_EN_MSK);
+	}
+
+	ret = hclge_cmd_send(&hdev->hw, &desc, 1);
+	if (ret)
+		dev_err(&hdev->pdev->dev, "tx buffer alloc cmd failed %d.\n",
+			ret);
+
+	return ret;
+}
+
+static int hclge_tx_buffer_alloc(struct hclge_dev *hdev,
+				 struct hclge_pkt_buf_alloc *buf_alloc)
+{
+	int ret = hclge_cmd_alloc_tx_buff(hdev, buf_alloc);
+
+	if (ret)
+		dev_err(&hdev->pdev->dev, "tx buffer alloc failed %d\n", ret);
+
+	return ret;
+}
+
+static u32 hclge_get_tc_num(struct hclge_dev *hdev)
+{
+	unsigned int i;
+	u32 cnt = 0;
+
+	for (i = 0; i < HCLGE_MAX_TC_NUM; i++)
+		if (hdev->hw_tc_map & BIT(i))
+			cnt++;
+	return cnt;
+}
+
+/* Get the number of pfc enabled TCs, which have private buffer */
+static int hclge_get_pfc_priv_num(struct hclge_dev *hdev,
+				  struct hclge_pkt_buf_alloc *buf_alloc)
+{
+	struct hclge_priv_buf *priv;
+	unsigned int i;
+	int cnt = 0;
+
+	for (i = 0; i < HCLGE_MAX_TC_NUM; i++) {
+		priv = &buf_alloc->priv_buf[i];
+		if ((hdev->tm_info.hw_pfc_map & BIT(i)) &&
+		    priv->enable)
+			cnt++;
+	}
+
+	return cnt;
+}
+
+/* Get the number of pfc disabled TCs, which have private buffer */
+static int hclge_get_no_pfc_priv_num(struct hclge_dev *hdev,
+				     struct hclge_pkt_buf_alloc *buf_alloc)
+{
+	struct hclge_priv_buf *priv;
+	unsigned int i;
+	int cnt = 0;
+
+	for (i = 0; i < HCLGE_MAX_TC_NUM; i++) {
+		priv = &buf_alloc->priv_buf[i];
+		if (hdev->hw_tc_map & BIT(i) &&
+		    !(hdev->tm_info.hw_pfc_map & BIT(i)) &&
+		    priv->enable)
+			cnt++;
+	}
+
+	return cnt;
+}
+
+static u32 hclge_get_rx_priv_buff_alloced(struct hclge_pkt_buf_alloc *buf_alloc)
+{
+	struct hclge_priv_buf *priv;
+	u32 rx_priv = 0;
+	int i;
+
+	for (i = 0; i < HCLGE_MAX_TC_NUM; i++) {
+		priv = &buf_alloc->priv_buf[i];
+		if (priv->enable)
+			rx_priv += priv->buf_size;
+	}
+	return rx_priv;
+}
+
+static u32 hclge_get_tx_buff_alloced(struct hclge_pkt_buf_alloc *buf_alloc)
+{
+	u32 i, total_tx_size = 0;
+
+	for (i = 0; i < HCLGE_MAX_TC_NUM; i++)
+		total_tx_size += buf_alloc->priv_buf[i].tx_buf_size;
+
+	return total_tx_size;
+}
+
+static bool  hclge_is_rx_buf_ok(struct hclge_dev *hdev,
+				struct hclge_pkt_buf_alloc *buf_alloc,
+				u32 rx_all)
+{
+	u32 shared_buf_min, shared_buf_tc, shared_std, hi_thrd, lo_thrd;
+	u32 tc_num = hclge_get_tc_num(hdev);
+	u32 shared_buf, aligned_mps;
+	u32 rx_priv;
+	int i;
+
+	aligned_mps = roundup(hdev->mps, HCLGE_BUF_SIZE_UNIT);
+
+	if (hnae3_dev_dcb_supported(hdev))
+		shared_buf_min = HCLGE_BUF_MUL_BY * aligned_mps +
+					hdev->dv_buf_size;
+	else
+		shared_buf_min = aligned_mps + HCLGE_NON_DCB_ADDITIONAL_BUF
+					+ hdev->dv_buf_size;
+
+	shared_buf_tc = tc_num * aligned_mps + aligned_mps;
+	shared_std = roundup(max_t(u32, shared_buf_min, shared_buf_tc),
+			     HCLGE_BUF_SIZE_UNIT);
+
+	rx_priv = hclge_get_rx_priv_buff_alloced(buf_alloc);
+	if (rx_all < rx_priv + shared_std)
+		return false;
+
+	shared_buf = rounddown(rx_all - rx_priv, HCLGE_BUF_SIZE_UNIT);
+	buf_alloc->s_buf.buf_size = shared_buf;
+	if (hnae3_dev_dcb_supported(hdev)) {
+		buf_alloc->s_buf.self.high = shared_buf - hdev->dv_buf_size;
+		buf_alloc->s_buf.self.low = buf_alloc->s_buf.self.high
+			- roundup(aligned_mps / HCLGE_BUF_DIV_BY,
+				  HCLGE_BUF_SIZE_UNIT);
+	} else {
+		buf_alloc->s_buf.self.high = aligned_mps +
+						HCLGE_NON_DCB_ADDITIONAL_BUF;
+		buf_alloc->s_buf.self.low = aligned_mps;
+	}
+
+	if (hnae3_dev_dcb_supported(hdev)) {
+		hi_thrd = shared_buf - hdev->dv_buf_size;
+
+		if (tc_num <= NEED_RESERVE_TC_NUM)
+			hi_thrd = hi_thrd * BUF_RESERVE_PERCENT
+					/ BUF_MAX_PERCENT;
+
+		if (tc_num)
+			hi_thrd = hi_thrd / tc_num;
+
+		hi_thrd = max_t(u32, hi_thrd, HCLGE_BUF_MUL_BY * aligned_mps);
+		hi_thrd = rounddown(hi_thrd, HCLGE_BUF_SIZE_UNIT);
+		lo_thrd = hi_thrd - aligned_mps / HCLGE_BUF_DIV_BY;
+	} else {
+		hi_thrd = aligned_mps + HCLGE_NON_DCB_ADDITIONAL_BUF;
+		lo_thrd = aligned_mps;
+	}
+
+	for (i = 0; i < HCLGE_MAX_TC_NUM; i++) {
+		buf_alloc->s_buf.tc_thrd[i].low = lo_thrd;
+		buf_alloc->s_buf.tc_thrd[i].high = hi_thrd;
+	}
+
+	return true;
+}
+
+static int hclge_tx_buffer_calc(struct hclge_dev *hdev,
+				struct hclge_pkt_buf_alloc *buf_alloc)
+{
+	u32 i, total_size;
+
+	total_size = hdev->pkt_buf_size;
+
+	/* alloc tx buffer for all enabled tc */
+	for (i = 0; i < HCLGE_MAX_TC_NUM; i++) {
+		struct hclge_priv_buf *priv = &buf_alloc->priv_buf[i];
+
+		if (hdev->hw_tc_map & BIT(i)) {
+			if (total_size < hdev->tx_buf_size)
+				return -ENOMEM;
+
+			priv->tx_buf_size = hdev->tx_buf_size;
+		} else {
+			priv->tx_buf_size = 0;
+		}
+
+		total_size -= priv->tx_buf_size;
+	}
+
+	return 0;
+}
+
+static bool hclge_rx_buf_calc_all(struct hclge_dev *hdev, bool max,
+				  struct hclge_pkt_buf_alloc *buf_alloc)
+{
+	u32 rx_all = hdev->pkt_buf_size - hclge_get_tx_buff_alloced(buf_alloc);
+	u32 aligned_mps = round_up(hdev->mps, HCLGE_BUF_SIZE_UNIT);
+	unsigned int i;
+
+	for (i = 0; i < HCLGE_MAX_TC_NUM; i++) {
+		struct hclge_priv_buf *priv = &buf_alloc->priv_buf[i];
+
+		priv->enable = 0;
+		priv->wl.low = 0;
+		priv->wl.high = 0;
+		priv->buf_size = 0;
+
+		if (!(hdev->hw_tc_map & BIT(i)))
+			continue;
+
+		priv->enable = 1;
+
+		if (hdev->tm_info.hw_pfc_map & BIT(i)) {
+			priv->wl.low = max ? aligned_mps : HCLGE_BUF_SIZE_UNIT;
+			priv->wl.high = roundup(priv->wl.low + aligned_mps,
+						HCLGE_BUF_SIZE_UNIT);
+		} else {
+			priv->wl.low = 0;
+			priv->wl.high = max ? (aligned_mps * HCLGE_BUF_MUL_BY) :
+					aligned_mps;
+		}
+
+		priv->buf_size = priv->wl.high + hdev->dv_buf_size;
+	}
+
+	return hclge_is_rx_buf_ok(hdev, buf_alloc, rx_all);
+}
+
+static bool hclge_drop_nopfc_buf_till_fit(struct hclge_dev *hdev,
+					  struct hclge_pkt_buf_alloc *buf_alloc)
+{
+	u32 rx_all = hdev->pkt_buf_size - hclge_get_tx_buff_alloced(buf_alloc);
+	int no_pfc_priv_num = hclge_get_no_pfc_priv_num(hdev, buf_alloc);
+	int i;
+
+	/* let the last to be cleared first */
+	for (i = HCLGE_MAX_TC_NUM - 1; i >= 0; i--) {
+		struct hclge_priv_buf *priv = &buf_alloc->priv_buf[i];
+		unsigned int mask = BIT((unsigned int)i);
+
+		if (hdev->hw_tc_map & mask &&
+		    !(hdev->tm_info.hw_pfc_map & mask)) {
+			/* Clear the no pfc TC private buffer */
+			priv->wl.low = 0;
+			priv->wl.high = 0;
+			priv->buf_size = 0;
+			priv->enable = 0;
+			no_pfc_priv_num--;
+		}
+
+		if (hclge_is_rx_buf_ok(hdev, buf_alloc, rx_all) ||
+		    no_pfc_priv_num == 0)
+			break;
+	}
+
+	return hclge_is_rx_buf_ok(hdev, buf_alloc, rx_all);
+}
+
+static bool hclge_drop_pfc_buf_till_fit(struct hclge_dev *hdev,
+					struct hclge_pkt_buf_alloc *buf_alloc)
+{
+	u32 rx_all = hdev->pkt_buf_size - hclge_get_tx_buff_alloced(buf_alloc);
+	int pfc_priv_num = hclge_get_pfc_priv_num(hdev, buf_alloc);
+	int i;
+
+	/* let the last to be cleared first */
+	for (i = HCLGE_MAX_TC_NUM - 1; i >= 0; i--) {
+		struct hclge_priv_buf *priv = &buf_alloc->priv_buf[i];
+		unsigned int mask = BIT((unsigned int)i);
+
+		if (hdev->hw_tc_map & mask &&
+		    hdev->tm_info.hw_pfc_map & mask) {
+			/* Reduce the number of pfc TC with private buffer */
+			priv->wl.low = 0;
+			priv->enable = 0;
+			priv->wl.high = 0;
+			priv->buf_size = 0;
+			pfc_priv_num--;
+		}
+
+		if (hclge_is_rx_buf_ok(hdev, buf_alloc, rx_all) ||
+		    pfc_priv_num == 0)
+			break;
+	}
+
+	return hclge_is_rx_buf_ok(hdev, buf_alloc, rx_all);
+}
+
+static bool hclge_only_alloc_priv_buff(struct hclge_dev *hdev,
+				       struct hclge_pkt_buf_alloc *buf_alloc)
+{
+#define COMPENSATE_BUFFER	0x3C00
+#define COMPENSATE_HALF_MPS_NUM	5
+#define PRIV_WL_GAP		0x1800
+
+	u32 rx_priv = hdev->pkt_buf_size - hclge_get_tx_buff_alloced(buf_alloc);
+	u32 tc_num = hclge_get_tc_num(hdev);
+	u32 half_mps = hdev->mps >> 1;
+	u32 min_rx_priv;
+	unsigned int i;
+
+	if (tc_num)
+		rx_priv = rx_priv / tc_num;
+
+	if (tc_num <= NEED_RESERVE_TC_NUM)
+		rx_priv = rx_priv * BUF_RESERVE_PERCENT / BUF_MAX_PERCENT;
+
+	min_rx_priv = hdev->dv_buf_size + COMPENSATE_BUFFER +
+			COMPENSATE_HALF_MPS_NUM * half_mps;
+	min_rx_priv = round_up(min_rx_priv, HCLGE_BUF_SIZE_UNIT);
+	rx_priv = round_down(rx_priv, HCLGE_BUF_SIZE_UNIT);
+	if (rx_priv < min_rx_priv)
+		return false;
+
+	for (i = 0; i < HCLGE_MAX_TC_NUM; i++) {
+		struct hclge_priv_buf *priv = &buf_alloc->priv_buf[i];
+
+		priv->enable = 0;
+		priv->wl.low = 0;
+		priv->wl.high = 0;
+		priv->buf_size = 0;
+
+		if (!(hdev->hw_tc_map & BIT(i)))
+			continue;
+
+		priv->enable = 1;
+		priv->buf_size = rx_priv;
+		priv->wl.high = rx_priv - hdev->dv_buf_size;
+		priv->wl.low = priv->wl.high - PRIV_WL_GAP;
+	}
+
+	buf_alloc->s_buf.buf_size = 0;
+
+	return true;
+}
+
+/* hclge_rx_buffer_calc: calculate the rx private buffer size for all TCs
+ * @hdev: pointer to struct hclge_dev
+ * @buf_alloc: pointer to buffer calculation data
+ * @return: 0: calculate successful, negative: fail
+ */
+static int hclge_rx_buffer_calc(struct hclge_dev *hdev,
+				struct hclge_pkt_buf_alloc *buf_alloc)
+{
+	/* When DCB is not supported, rx private buffer is not allocated. */
+	if (!hnae3_dev_dcb_supported(hdev)) {
+		u32 rx_all = hdev->pkt_buf_size;
+
+		rx_all -= hclge_get_tx_buff_alloced(buf_alloc);
+		if (!hclge_is_rx_buf_ok(hdev, buf_alloc, rx_all))
+			return -ENOMEM;
+
+		return 0;
+	}
+
+	if (hclge_only_alloc_priv_buff(hdev, buf_alloc))
+		return 0;
+
+	if (hclge_rx_buf_calc_all(hdev, true, buf_alloc))
+		return 0;
+
+	/* try to decrease the buffer size */
+	if (hclge_rx_buf_calc_all(hdev, false, buf_alloc))
+		return 0;
+
+	if (hclge_drop_nopfc_buf_till_fit(hdev, buf_alloc))
+		return 0;
+
+	if (hclge_drop_pfc_buf_till_fit(hdev, buf_alloc))
+		return 0;
+
+	return -ENOMEM;
+}
+
+static int hclge_rx_priv_buf_alloc(struct hclge_dev *hdev,
+				   struct hclge_pkt_buf_alloc *buf_alloc)
+{
+	struct hclge_rx_priv_buff_cmd *req;
+	struct hclge_desc desc;
+	int ret;
+	int i;
+
+	hclge_cmd_setup_basic_desc(&desc, HCLGE_OPC_RX_PRIV_BUFF_ALLOC, false);
+	req = (struct hclge_rx_priv_buff_cmd *)desc.data;
+
+	/* Alloc private buffer TCs */
+	for (i = 0; i < HCLGE_MAX_TC_NUM; i++) {
+		struct hclge_priv_buf *priv = &buf_alloc->priv_buf[i];
+
+		req->buf_num[i] =
+			cpu_to_le16(priv->buf_size >> HCLGE_BUF_UNIT_S);
+		req->buf_num[i] |=
+			cpu_to_le16(1 << HCLGE_TC0_PRI_BUF_EN_B);
+	}
+
+	req->shared_buf =
+		cpu_to_le16((buf_alloc->s_buf.buf_size >> HCLGE_BUF_UNIT_S) |
+			    (1 << HCLGE_TC0_PRI_BUF_EN_B));
+
+	ret = hclge_cmd_send(&hdev->hw, &desc, 1);
+	if (ret)
+		dev_err(&hdev->pdev->dev,
+			"rx private buffer alloc cmd failed %d\n", ret);
+
+	return ret;
+}
+
+static int hclge_rx_priv_wl_config(struct hclge_dev *hdev,
+				   struct hclge_pkt_buf_alloc *buf_alloc)
+{
+	struct hclge_rx_priv_wl_buf *req;
+	struct hclge_priv_buf *priv;
+	struct hclge_desc desc[2];
+	int i, j;
+	int ret;
+
+	for (i = 0; i < 2; i++) {
+		hclge_cmd_setup_basic_desc(&desc[i], HCLGE_OPC_RX_PRIV_WL_ALLOC,
+					   false);
+		req = (struct hclge_rx_priv_wl_buf *)desc[i].data;
+
+		/* The first descriptor set the NEXT bit to 1 */
+		if (i == 0)
+			desc[i].flag |= cpu_to_le16(HCLGE_COMM_CMD_FLAG_NEXT);
+		else
+			desc[i].flag &= ~cpu_to_le16(HCLGE_COMM_CMD_FLAG_NEXT);
+
+		for (j = 0; j < HCLGE_TC_NUM_ONE_DESC; j++) {
+			u32 idx = i * HCLGE_TC_NUM_ONE_DESC + j;
+
+			priv = &buf_alloc->priv_buf[idx];
+			req->tc_wl[j].high =
+				cpu_to_le16(priv->wl.high >> HCLGE_BUF_UNIT_S);
+			req->tc_wl[j].high |=
+				cpu_to_le16(BIT(HCLGE_RX_PRIV_EN_B));
+			req->tc_wl[j].low =
+				cpu_to_le16(priv->wl.low >> HCLGE_BUF_UNIT_S);
+			req->tc_wl[j].low |=
+				 cpu_to_le16(BIT(HCLGE_RX_PRIV_EN_B));
+		}
+	}
+
+	/* Send 2 descriptor at one time */
+	ret = hclge_cmd_send(&hdev->hw, desc, 2);
+	if (ret)
+		dev_err(&hdev->pdev->dev,
+			"rx private waterline config cmd failed %d\n",
+			ret);
+	return ret;
+}
+
+static int hclge_common_thrd_config(struct hclge_dev *hdev,
+				    struct hclge_pkt_buf_alloc *buf_alloc)
+{
+	struct hclge_shared_buf *s_buf = &buf_alloc->s_buf;
+	struct hclge_rx_com_thrd *req;
+	struct hclge_desc desc[2];
+	struct hclge_tc_thrd *tc;
+	int i, j;
+	int ret;
+
+	for (i = 0; i < 2; i++) {
+		hclge_cmd_setup_basic_desc(&desc[i],
+					   HCLGE_OPC_RX_COM_THRD_ALLOC, false);
+		req = (struct hclge_rx_com_thrd *)desc[i].data;
+
+		/* The first descriptor set the NEXT bit to 1 */
+		if (i == 0)
+			desc[i].flag |= cpu_to_le16(HCLGE_COMM_CMD_FLAG_NEXT);
+		else
+			desc[i].flag &= ~cpu_to_le16(HCLGE_COMM_CMD_FLAG_NEXT);
+
+		for (j = 0; j < HCLGE_TC_NUM_ONE_DESC; j++) {
+			tc = &s_buf->tc_thrd[i * HCLGE_TC_NUM_ONE_DESC + j];
+
+			req->com_thrd[j].high =
+				cpu_to_le16(tc->high >> HCLGE_BUF_UNIT_S);
+			req->com_thrd[j].high |=
+				 cpu_to_le16(BIT(HCLGE_RX_PRIV_EN_B));
+			req->com_thrd[j].low =
+				cpu_to_le16(tc->low >> HCLGE_BUF_UNIT_S);
+			req->com_thrd[j].low |=
+				 cpu_to_le16(BIT(HCLGE_RX_PRIV_EN_B));
+		}
+	}
+
+	/* Send 2 descriptors at one time */
+	ret = hclge_cmd_send(&hdev->hw, desc, 2);
+	if (ret)
+		dev_err(&hdev->pdev->dev,
+			"common threshold config cmd failed %d\n", ret);
+	return ret;
+}
+
+static int hclge_common_wl_config(struct hclge_dev *hdev,
+				  struct hclge_pkt_buf_alloc *buf_alloc)
+{
+	struct hclge_shared_buf *buf = &buf_alloc->s_buf;
+	struct hclge_rx_com_wl *req;
+	struct hclge_desc desc;
+	int ret;
+
+	hclge_cmd_setup_basic_desc(&desc, HCLGE_OPC_RX_COM_WL_ALLOC, false);
+
+	req = (struct hclge_rx_com_wl *)desc.data;
+	req->com_wl.high = cpu_to_le16(buf->self.high >> HCLGE_BUF_UNIT_S);
+	req->com_wl.high |=  cpu_to_le16(BIT(HCLGE_RX_PRIV_EN_B));
+
+	req->com_wl.low = cpu_to_le16(buf->self.low >> HCLGE_BUF_UNIT_S);
+	req->com_wl.low |=  cpu_to_le16(BIT(HCLGE_RX_PRIV_EN_B));
+
+	ret = hclge_cmd_send(&hdev->hw, &desc, 1);
+	if (ret)
+		dev_err(&hdev->pdev->dev,
+			"common waterline config cmd failed %d\n", ret);
+
+	return ret;
+}
+
+int hclge_buffer_alloc(struct hclge_dev *hdev)
+{
+	struct hclge_pkt_buf_alloc *pkt_buf;
+	int ret;
+
+	pkt_buf = kzalloc(sizeof(*pkt_buf), GFP_KERNEL);
+	if (!pkt_buf)
+		return -ENOMEM;
+
+	ret = hclge_tx_buffer_calc(hdev, pkt_buf);
+	if (ret) {
+		dev_err(&hdev->pdev->dev,
+			"could not calc tx buffer size for all TCs %d\n", ret);
+		goto out;
+	}
+
+	ret = hclge_tx_buffer_alloc(hdev, pkt_buf);
+	if (ret) {
+		dev_err(&hdev->pdev->dev,
+			"could not alloc tx buffers %d\n", ret);
+		goto out;
+	}
+
+	ret = hclge_rx_buffer_calc(hdev, pkt_buf);
+	if (ret) {
+		dev_err(&hdev->pdev->dev,
+			"could not calc rx priv buffer size for all TCs %d\n",
+			ret);
+		goto out;
+	}
+
+	ret = hclge_rx_priv_buf_alloc(hdev, pkt_buf);
+	if (ret) {
+		dev_err(&hdev->pdev->dev, "could not alloc rx priv buffer %d\n",
+			ret);
+		goto out;
+	}
+
+	if (hnae3_dev_dcb_supported(hdev)) {
+		ret = hclge_rx_priv_wl_config(hdev, pkt_buf);
+		if (ret) {
+			dev_err(&hdev->pdev->dev,
+				"could not configure rx private waterline %d\n",
+				ret);
+			goto out;
+		}
+
+		ret = hclge_common_thrd_config(hdev, pkt_buf);
+		if (ret) {
+			dev_err(&hdev->pdev->dev,
+				"could not configure common threshold %d\n",
+				ret);
+			goto out;
+		}
+	}
+
+	ret = hclge_common_wl_config(hdev, pkt_buf);
+	if (ret)
+		dev_err(&hdev->pdev->dev,
+			"could not configure common waterline %d\n", ret);
+
+out:
+	kfree(pkt_buf);
+	return ret;
+}
+
+static int hclge_init_roce_base_info(struct hclge_vport *vport)
+{
+	struct hnae3_handle *roce = &vport->roce;
+	struct hnae3_handle *nic = &vport->nic;
+	struct hclge_dev *hdev = vport->back;
+
+	roce->rinfo.num_vectors = vport->back->num_roce_msi;
+
+	if (hdev->num_msi < hdev->num_nic_msi + hdev->num_roce_msi)
+		return -EINVAL;
+
+	roce->rinfo.base_vector = hdev->num_nic_msi;
+
+	roce->rinfo.netdev = nic->kinfo.netdev;
+	roce->rinfo.roce_io_base = hdev->hw.hw.io_base;
+	roce->rinfo.roce_mem_base = hdev->hw.hw.mem_base;
+
+	roce->pdev = nic->pdev;
+	roce->ae_algo = nic->ae_algo;
+	bitmap_copy(roce->numa_node_mask.bits, nic->numa_node_mask.bits,
+		    MAX_NUMNODES);
+
+	return 0;
+}
+
+static int hclge_init_msi(struct hclge_dev *hdev)
+{
+	struct pci_dev *pdev = hdev->pdev;
+	int vectors;
+	int i;
+
+	vectors = pci_alloc_irq_vectors(pdev, HNAE3_MIN_VECTOR_NUM,
+					hdev->num_msi,
+					PCI_IRQ_MSI | PCI_IRQ_MSIX);
+	if (vectors < 0) {
+		dev_err(&pdev->dev,
+			"failed(%d) to allocate MSI/MSI-X vectors\n",
+			vectors);
+		return vectors;
+	}
+	if (vectors < hdev->num_msi)
+		dev_warn(&hdev->pdev->dev,
+			 "requested %u MSI/MSI-X, but allocated %d MSI/MSI-X\n",
+			 hdev->num_msi, vectors);
+
+	hdev->num_msi = vectors;
+	hdev->num_msi_left = vectors;
+
+	hdev->vector_status = devm_kcalloc(&pdev->dev, hdev->num_msi,
+					   sizeof(u16), GFP_KERNEL);
+	if (!hdev->vector_status) {
+		pci_free_irq_vectors(pdev);
+		return -ENOMEM;
+	}
+
+	for (i = 0; i < hdev->num_msi; i++)
+		hdev->vector_status[i] = HCLGE_INVALID_VPORT;
+
+	hdev->vector_irq = devm_kcalloc(&pdev->dev, hdev->num_msi,
+					sizeof(int), GFP_KERNEL);
+	if (!hdev->vector_irq) {
+		pci_free_irq_vectors(pdev);
+		return -ENOMEM;
+	}
+
+	return 0;
+}
+
+static u8 hclge_check_speed_dup(u8 duplex, int speed)
+{
+	if (!(speed == HCLGE_MAC_SPEED_10M || speed == HCLGE_MAC_SPEED_100M))
+		duplex = HCLGE_MAC_FULL;
+
+	return duplex;
+}
+
+static struct hclge_mac_speed_map hclge_mac_speed_map_to_fw[] = {
+	{HCLGE_MAC_SPEED_10M, HCLGE_FW_MAC_SPEED_10M},
+	{HCLGE_MAC_SPEED_100M, HCLGE_FW_MAC_SPEED_100M},
+	{HCLGE_MAC_SPEED_1G, HCLGE_FW_MAC_SPEED_1G},
+	{HCLGE_MAC_SPEED_10G, HCLGE_FW_MAC_SPEED_10G},
+	{HCLGE_MAC_SPEED_25G, HCLGE_FW_MAC_SPEED_25G},
+	{HCLGE_MAC_SPEED_40G, HCLGE_FW_MAC_SPEED_40G},
+	{HCLGE_MAC_SPEED_50G, HCLGE_FW_MAC_SPEED_50G},
+	{HCLGE_MAC_SPEED_100G, HCLGE_FW_MAC_SPEED_100G},
+	{HCLGE_MAC_SPEED_200G, HCLGE_FW_MAC_SPEED_200G},
+};
+
+static int hclge_convert_to_fw_speed(u32 speed_drv, u32 *speed_fw)
+{
+	u16 i;
+
+	for (i = 0; i < ARRAY_SIZE(hclge_mac_speed_map_to_fw); i++) {
+		if (hclge_mac_speed_map_to_fw[i].speed_drv == speed_drv) {
+			*speed_fw = hclge_mac_speed_map_to_fw[i].speed_fw;
+			return 0;
+		}
+	}
+
+	return -EINVAL;
+}
+
+static int hclge_cfg_mac_speed_dup_hw(struct hclge_dev *hdev, int speed,
+				      u8 duplex, u8 lane_num)
+{
+	struct hclge_config_mac_speed_dup_cmd *req;
+	struct hclge_desc desc;
+	u32 speed_fw;
+	int ret;
+
+	req = (struct hclge_config_mac_speed_dup_cmd *)desc.data;
+
+	hclge_cmd_setup_basic_desc(&desc, HCLGE_OPC_CONFIG_SPEED_DUP, false);
+
+	if (duplex)
+		hnae3_set_bit(req->speed_dup, HCLGE_CFG_DUPLEX_B, 1);
+
+	ret = hclge_convert_to_fw_speed(speed, &speed_fw);
+	if (ret) {
+		dev_err(&hdev->pdev->dev, "invalid speed (%d)\n", speed);
+		return ret;
+	}
+
+	hnae3_set_field(req->speed_dup, HCLGE_CFG_SPEED_M, HCLGE_CFG_SPEED_S,
+			speed_fw);
+	hnae3_set_bit(req->mac_change_fec_en, HCLGE_CFG_MAC_SPEED_CHANGE_EN_B,
+		      1);
+	req->lane_num = lane_num;
+
+	ret = hclge_cmd_send(&hdev->hw, &desc, 1);
+	if (ret) {
+		dev_err(&hdev->pdev->dev,
+			"mac speed/duplex config cmd failed %d.\n", ret);
+		return ret;
+	}
+
+	return 0;
+}
+
+int hclge_cfg_mac_speed_dup(struct hclge_dev *hdev, int speed, u8 duplex, u8 lane_num)
+{
+	struct hclge_mac *mac = &hdev->hw.mac;
+	int ret;
+
+	duplex = hclge_check_speed_dup(duplex, speed);
+	if (!mac->support_autoneg && mac->speed == (u32)speed &&
+	    mac->duplex == duplex && (mac->lane_num == lane_num || lane_num == 0))
+		return 0;
+
+	ret = hclge_cfg_mac_speed_dup_hw(hdev, speed, duplex, lane_num);
+	if (ret)
+		return ret;
+
+	hdev->hw.mac.speed = speed;
+	hdev->hw.mac.duplex = duplex;
+	if (!lane_num)
+		hdev->hw.mac.lane_num = lane_num;
+
+	return 0;
+}
+
+static int hclge_cfg_mac_speed_dup_h(struct hnae3_handle *handle, int speed,
+				     u8 duplex, u8 lane_num)
+{
+	struct hclge_vport *vport = hclge_get_vport(handle);
+	struct hclge_dev *hdev = vport->back;
+	int ret;
+
+	ret = hclge_cfg_mac_speed_dup(hdev, speed, duplex, lane_num);
+
+	if (ret)
+		return ret;
+
+	hdev->hw.mac.req_speed = (u32)speed;
+	hdev->hw.mac.req_duplex = duplex;
+
+	return 0;
+}
+
+static int hclge_set_autoneg_en(struct hclge_dev *hdev, bool enable)
+{
+	struct hclge_config_auto_neg_cmd *req;
+	struct hclge_desc desc;
+	u32 flag = 0;
+	int ret;
+
+	hclge_cmd_setup_basic_desc(&desc, HCLGE_OPC_CONFIG_AN_MODE, false);
+
+	req = (struct hclge_config_auto_neg_cmd *)desc.data;
+	if (enable)
+		hnae3_set_bit(flag, HCLGE_MAC_CFG_AN_EN_B, 1U);
+	req->cfg_an_cmd_flag = cpu_to_le32(flag);
+
+	ret = hclge_cmd_send(&hdev->hw, &desc, 1);
+	if (ret)
+		dev_err(&hdev->pdev->dev, "auto neg set cmd failed %d.\n",
+			ret);
+
+	return ret;
+}
+
+static int hclge_set_autoneg(struct hnae3_handle *handle, bool enable)
+{
+	struct hclge_vport *vport = hclge_get_vport(handle);
+	struct hclge_dev *hdev = vport->back;
+
+	if (!hdev->hw.mac.support_autoneg) {
+		if (enable) {
+			dev_err(&hdev->pdev->dev,
+				"autoneg is not supported by current port\n");
+			return -EOPNOTSUPP;
+		} else {
+			return 0;
+		}
+	}
+
+	return hclge_set_autoneg_en(hdev, enable);
+}
+
+static int hclge_get_autoneg(struct hnae3_handle *handle)
+{
+	struct hclge_vport *vport = hclge_get_vport(handle);
+	struct hclge_dev *hdev = vport->back;
+	struct phy_device *phydev = hdev->hw.mac.phydev;
+
+	if (phydev)
+		return phydev->autoneg;
+
+	return hdev->hw.mac.autoneg;
+}
+
+static int hclge_restart_autoneg(struct hnae3_handle *handle)
+{
+	struct hclge_vport *vport = hclge_get_vport(handle);
+	struct hclge_dev *hdev = vport->back;
+	int ret;
+
+	dev_dbg(&hdev->pdev->dev, "restart autoneg\n");
+
+	ret = hclge_notify_client(hdev, HNAE3_DOWN_CLIENT);
+	if (ret)
+		return ret;
+	return hclge_notify_client(hdev, HNAE3_UP_CLIENT);
+}
+
+static int hclge_halt_autoneg(struct hnae3_handle *handle, bool halt)
+{
+	struct hclge_vport *vport = hclge_get_vport(handle);
+	struct hclge_dev *hdev = vport->back;
+
+	if (hdev->hw.mac.support_autoneg && hdev->hw.mac.autoneg)
+		return hclge_set_autoneg_en(hdev, !halt);
+
+	return 0;
+}
+
+static void hclge_parse_fec_stats_lanes(struct hclge_dev *hdev,
+					struct hclge_desc *desc, u32 desc_len)
+{
+	u32 lane_size = HCLGE_FEC_STATS_MAX_LANES * 2;
+	u32 desc_index = 0;
+	u32 data_index = 0;
+	u32 i;
+
+	for (i = 0; i < lane_size; i++) {
+		if (data_index >= HCLGE_DESC_DATA_LEN) {
+			desc_index++;
+			data_index = 0;
+		}
+
+		if (desc_index >= desc_len)
+			return;
+
+		hdev->fec_stats.per_lanes[i] +=
+			le32_to_cpu(desc[desc_index].data[data_index]);
+		data_index++;
+	}
+}
+
+static void hclge_parse_fec_stats(struct hclge_dev *hdev,
+				  struct hclge_desc *desc, u32 desc_len)
+{
+	struct hclge_query_fec_stats_cmd *req;
+
+	req = (struct hclge_query_fec_stats_cmd *)desc[0].data;
+
+	hdev->fec_stats.base_r_lane_num = req->base_r_lane_num;
+	hdev->fec_stats.rs_corr_blocks +=
+		le32_to_cpu(req->rs_fec_corr_blocks);
+	hdev->fec_stats.rs_uncorr_blocks +=
+		le32_to_cpu(req->rs_fec_uncorr_blocks);
+	hdev->fec_stats.rs_error_blocks +=
+		le32_to_cpu(req->rs_fec_error_blocks);
+	hdev->fec_stats.base_r_corr_blocks +=
+		le32_to_cpu(req->base_r_fec_corr_blocks);
+	hdev->fec_stats.base_r_uncorr_blocks +=
+		le32_to_cpu(req->base_r_fec_uncorr_blocks);
+
+	hclge_parse_fec_stats_lanes(hdev, &desc[1], desc_len - 1);
+}
+
+static int hclge_update_fec_stats_hw(struct hclge_dev *hdev)
+{
+	struct hclge_desc desc[HCLGE_FEC_STATS_CMD_NUM];
+	int ret;
+	u32 i;
+
+	for (i = 0; i < HCLGE_FEC_STATS_CMD_NUM; i++) {
+		hclge_cmd_setup_basic_desc(&desc[i], HCLGE_OPC_QUERY_FEC_STATS,
+					   true);
+		if (i != (HCLGE_FEC_STATS_CMD_NUM - 1))
+			desc[i].flag |= cpu_to_le16(HCLGE_COMM_CMD_FLAG_NEXT);
+	}
+
+	ret = hclge_cmd_send(&hdev->hw, desc, HCLGE_FEC_STATS_CMD_NUM);
+	if (ret)
+		return ret;
+
+	hclge_parse_fec_stats(hdev, desc, HCLGE_FEC_STATS_CMD_NUM);
+
+	return 0;
+}
+
+static void hclge_update_fec_stats(struct hclge_dev *hdev)
+{
+	struct hnae3_ae_dev *ae_dev = pci_get_drvdata(hdev->pdev);
+	int ret;
+
+	if (!hnae3_ae_dev_fec_stats_supported(ae_dev) ||
+	    test_and_set_bit(HCLGE_STATE_FEC_STATS_UPDATING, &hdev->state))
+		return;
+
+	ret = hclge_update_fec_stats_hw(hdev);
+	if (ret)
+		dev_err(&hdev->pdev->dev,
+			"failed to update fec stats, ret = %d\n", ret);
+
+	clear_bit(HCLGE_STATE_FEC_STATS_UPDATING, &hdev->state);
+}
+
+static void hclge_get_fec_stats_total(struct hclge_dev *hdev,
+				      struct ethtool_fec_stats *fec_stats)
+{
+	fec_stats->corrected_blocks.total = hdev->fec_stats.rs_corr_blocks;
+	fec_stats->uncorrectable_blocks.total =
+		hdev->fec_stats.rs_uncorr_blocks;
+}
+
+static void hclge_get_fec_stats_lanes(struct hclge_dev *hdev,
+				      struct ethtool_fec_stats *fec_stats)
+{
+	u32 i;
+
+	if (hdev->fec_stats.base_r_lane_num == 0 ||
+	    hdev->fec_stats.base_r_lane_num > HCLGE_FEC_STATS_MAX_LANES) {
+		dev_err(&hdev->pdev->dev,
+			"fec stats lane number(%llu) is invalid\n",
+			hdev->fec_stats.base_r_lane_num);
+		return;
+	}
+
+	for (i = 0; i < hdev->fec_stats.base_r_lane_num; i++) {
+		fec_stats->corrected_blocks.lanes[i] =
+			hdev->fec_stats.base_r_corr_per_lanes[i];
+		fec_stats->uncorrectable_blocks.lanes[i] =
+			hdev->fec_stats.base_r_uncorr_per_lanes[i];
+	}
+}
+
+static void hclge_comm_get_fec_stats(struct hclge_dev *hdev,
+				     struct ethtool_fec_stats *fec_stats)
+{
+	u32 fec_mode = hdev->hw.mac.fec_mode;
+
+	switch (fec_mode) {
+	case BIT(HNAE3_FEC_RS):
+	case BIT(HNAE3_FEC_LLRS):
+		hclge_get_fec_stats_total(hdev, fec_stats);
+		break;
+	case BIT(HNAE3_FEC_BASER):
+		hclge_get_fec_stats_lanes(hdev, fec_stats);
+		break;
+	default:
+		dev_err(&hdev->pdev->dev,
+			"fec stats is not supported by current fec mode(0x%x)\n",
+			fec_mode);
+		break;
+	}
+}
+
+static void hclge_get_fec_stats(struct hnae3_handle *handle,
+				struct ethtool_fec_stats *fec_stats)
+{
+	struct hclge_vport *vport = hclge_get_vport(handle);
+	struct hclge_dev *hdev = vport->back;
+	u32 fec_mode = hdev->hw.mac.fec_mode;
+
+	if (fec_mode == BIT(HNAE3_FEC_NONE) ||
+	    fec_mode == BIT(HNAE3_FEC_AUTO) ||
+	    fec_mode == BIT(HNAE3_FEC_USER_DEF))
+		return;
+
+	hclge_update_fec_stats(hdev);
+
+	hclge_comm_get_fec_stats(hdev, fec_stats);
+}
+
+static int hclge_set_fec_hw(struct hclge_dev *hdev, u32 fec_mode)
+{
+	struct hclge_config_fec_cmd *req;
+	struct hclge_desc desc;
+	int ret;
+
+	hclge_cmd_setup_basic_desc(&desc, HCLGE_OPC_CONFIG_FEC_MODE, false);
+
+	req = (struct hclge_config_fec_cmd *)desc.data;
+	if (fec_mode & BIT(HNAE3_FEC_AUTO))
+		hnae3_set_bit(req->fec_mode, HCLGE_MAC_CFG_FEC_AUTO_EN_B, 1);
+	if (fec_mode & BIT(HNAE3_FEC_RS))
+		hnae3_set_field(req->fec_mode, HCLGE_MAC_CFG_FEC_MODE_M,
+				HCLGE_MAC_CFG_FEC_MODE_S, HCLGE_MAC_FEC_RS);
+	if (fec_mode & BIT(HNAE3_FEC_LLRS))
+		hnae3_set_field(req->fec_mode, HCLGE_MAC_CFG_FEC_MODE_M,
+				HCLGE_MAC_CFG_FEC_MODE_S, HCLGE_MAC_FEC_LLRS);
+	if (fec_mode & BIT(HNAE3_FEC_BASER))
+		hnae3_set_field(req->fec_mode, HCLGE_MAC_CFG_FEC_MODE_M,
+				HCLGE_MAC_CFG_FEC_MODE_S, HCLGE_MAC_FEC_BASER);
+
+	ret = hclge_cmd_send(&hdev->hw, &desc, 1);
+	if (ret)
+		dev_err(&hdev->pdev->dev, "set fec mode failed %d.\n", ret);
+
+	return ret;
+}
+
+static int hclge_set_fec(struct hnae3_handle *handle, u32 fec_mode)
+{
+	struct hclge_vport *vport = hclge_get_vport(handle);
+	struct hclge_dev *hdev = vport->back;
+	struct hclge_mac *mac = &hdev->hw.mac;
+	int ret;
+
+	if (fec_mode && !(mac->fec_ability & fec_mode)) {
+		dev_err(&hdev->pdev->dev, "unsupported fec mode\n");
+		return -EINVAL;
+	}
+
+	ret = hclge_set_fec_hw(hdev, fec_mode);
+	if (ret)
+		return ret;
+
+	mac->user_fec_mode = fec_mode | BIT(HNAE3_FEC_USER_DEF);
+	return 0;
+}
+
+static void hclge_get_fec(struct hnae3_handle *handle, u8 *fec_ability,
+			  u8 *fec_mode)
+{
+	struct hclge_vport *vport = hclge_get_vport(handle);
+	struct hclge_dev *hdev = vport->back;
+	struct hclge_mac *mac = &hdev->hw.mac;
+
+	if (fec_ability)
+		*fec_ability = mac->fec_ability;
+	if (fec_mode)
+		*fec_mode = mac->fec_mode;
+}
+
+static int hclge_mac_init(struct hclge_dev *hdev)
+{
+	struct hclge_mac *mac = &hdev->hw.mac;
+	int ret;
+
+	hdev->support_sfp_query = true;
+
+	if (!test_bit(HCLGE_STATE_RST_HANDLING, &hdev->state))
+		hdev->hw.mac.duplex = HCLGE_MAC_FULL;
+
+	if (hdev->hw.mac.support_autoneg) {
+		ret = hclge_set_autoneg_en(hdev, hdev->hw.mac.autoneg);
+		if (ret)
+			return ret;
+	}
+
+	if (!hdev->hw.mac.autoneg) {
+		ret = hclge_cfg_mac_speed_dup_hw(hdev, hdev->hw.mac.req_speed,
+						 hdev->hw.mac.req_duplex,
+						 hdev->hw.mac.lane_num);
+		if (ret)
+			return ret;
+	}
+
+	mac->link = 0;
+
+	if (mac->user_fec_mode & BIT(HNAE3_FEC_USER_DEF)) {
+		ret = hclge_set_fec_hw(hdev, mac->user_fec_mode);
+		if (ret)
+			return ret;
+	}
+
+	ret = hclge_set_mac_mtu(hdev, hdev->mps);
+	if (ret) {
+		dev_err(&hdev->pdev->dev, "set mtu failed ret=%d\n", ret);
+		return ret;
+	}
+
+	ret = hclge_set_default_loopback(hdev);
+	if (ret)
+		return ret;
+
+	ret = hclge_buffer_alloc(hdev);
+	if (ret)
+		dev_err(&hdev->pdev->dev,
+			"allocate buffer fail, ret=%d\n", ret);
+
+	return ret;
+}
+
+static void hclge_mbx_task_schedule(struct hclge_dev *hdev)
+{
+	if (!test_bit(HCLGE_STATE_REMOVING, &hdev->state) &&
+	    !test_and_set_bit(HCLGE_STATE_MBX_SERVICE_SCHED, &hdev->state)) {
+		hdev->last_mbx_scheduled = jiffies;
+		mod_delayed_work(hclge_wq, &hdev->service_task, 0);
+	}
+}
+
+static void hclge_reset_task_schedule(struct hclge_dev *hdev)
+{
+	if (!test_bit(HCLGE_STATE_REMOVING, &hdev->state) &&
+	    test_bit(HCLGE_STATE_SERVICE_INITED, &hdev->state) &&
+	    !test_and_set_bit(HCLGE_STATE_RST_SERVICE_SCHED, &hdev->state)) {
+		hdev->last_rst_scheduled = jiffies;
+		mod_delayed_work(hclge_wq, &hdev->service_task, 0);
+	}
+}
+
+static void hclge_errhand_task_schedule(struct hclge_dev *hdev)
+{
+	if (!test_bit(HCLGE_STATE_REMOVING, &hdev->state) &&
+	    !test_and_set_bit(HCLGE_STATE_ERR_SERVICE_SCHED, &hdev->state))
+		mod_delayed_work(hclge_wq, &hdev->service_task, 0);
+}
+
+void hclge_task_schedule(struct hclge_dev *hdev, unsigned long delay_time)
+{
+	if (!test_bit(HCLGE_STATE_REMOVING, &hdev->state) &&
+	    !test_bit(HCLGE_STATE_RST_FAIL, &hdev->state))
+		mod_delayed_work(hclge_wq, &hdev->service_task, delay_time);
+}
+
+static int hclge_get_mac_link_status(struct hclge_dev *hdev, int *link_status)
+{
+	struct hclge_link_status_cmd *req;
+	struct hclge_desc desc;
+	int ret;
+
+	hclge_cmd_setup_basic_desc(&desc, HCLGE_OPC_QUERY_LINK_STATUS, true);
+	ret = hclge_cmd_send(&hdev->hw, &desc, 1);
+	if (ret) {
+		dev_err(&hdev->pdev->dev, "get link status cmd failed %d\n",
+			ret);
+		return ret;
+	}
+
+	req = (struct hclge_link_status_cmd *)desc.data;
+	*link_status = (req->status & HCLGE_LINK_STATUS_UP_M) > 0 ?
+		HCLGE_LINK_STATUS_UP : HCLGE_LINK_STATUS_DOWN;
+
+	return 0;
+}
+
+static int hclge_get_mac_phy_link(struct hclge_dev *hdev, int *link_status)
+{
+	struct phy_device *phydev = hdev->hw.mac.phydev;
+
+	*link_status = HCLGE_LINK_STATUS_DOWN;
+
+	if (test_bit(HCLGE_STATE_DOWN, &hdev->state))
+		return 0;
+
+	if (phydev && (phydev->state != PHY_RUNNING || !phydev->link))
+		return 0;
+
+	return hclge_get_mac_link_status(hdev, link_status);
+}
+
+static void hclge_push_link_status(struct hclge_dev *hdev)
+{
+	struct hclge_vport *vport;
+	int ret;
+	u16 i;
+
+	for (i = 0; i < pci_num_vf(hdev->pdev); i++) {
+		vport = &hdev->vport[i + HCLGE_VF_VPORT_START_NUM];
+
+		if (!test_bit(HCLGE_VPORT_STATE_ALIVE, &vport->state) ||
+		    vport->vf_info.link_state != IFLA_VF_LINK_STATE_AUTO)
+			continue;
+
+		ret = hclge_push_vf_link_status(vport);
+		if (ret) {
+			dev_err(&hdev->pdev->dev,
+				"failed to push link status to vf%u, ret = %d\n",
+				i, ret);
+		}
+	}
+}
+
+static void hclge_update_link_status(struct hclge_dev *hdev)
+{
+	struct hnae3_handle *handle = &hdev->vport[0].nic;
+	struct hnae3_client *client = hdev->nic_client;
+	int state;
+	int ret;
+
+	if (!client)
+		return;
+
+	if (test_and_set_bit(HCLGE_STATE_LINK_UPDATING, &hdev->state))
+		return;
+
+	ret = hclge_get_mac_phy_link(hdev, &state);
+	if (ret) {
+		clear_bit(HCLGE_STATE_LINK_UPDATING, &hdev->state);
+		return;
+	}
+
+	if (state != hdev->hw.mac.link) {
+		hdev->hw.mac.link = state;
+		if (state == HCLGE_LINK_STATUS_UP)
+			hclge_update_port_info(hdev);
+
+		client->ops->link_status_change(handle, state);
+		hclge_config_mac_tnl_int(hdev, state);
+
+		if (test_bit(HCLGE_STATE_ROCE_REGISTERED, &hdev->state)) {
+			struct hnae3_handle *rhandle = &hdev->vport[0].roce;
+			struct hnae3_client *rclient = hdev->roce_client;
+
+			if (rclient && rclient->ops->link_status_change)
+				rclient->ops->link_status_change(rhandle,
+								 state);
+		}
+
+		hclge_push_link_status(hdev);
+	}
+
+	clear_bit(HCLGE_STATE_LINK_UPDATING, &hdev->state);
+}
+
+static void hclge_update_speed_advertising(struct hclge_mac *mac)
+{
+	u32 speed_ability;
+
+	if (hclge_get_speed_bit(mac->speed, &speed_ability))
+		return;
+
+	switch (mac->module_type) {
+	case HNAE3_MODULE_TYPE_FIBRE_LR:
+		hclge_convert_setting_lr(speed_ability, mac->advertising);
+		break;
+	case HNAE3_MODULE_TYPE_FIBRE_SR:
+	case HNAE3_MODULE_TYPE_AOC:
+		hclge_convert_setting_sr(speed_ability, mac->advertising);
+		break;
+	case HNAE3_MODULE_TYPE_CR:
+		hclge_convert_setting_cr(speed_ability, mac->advertising);
+		break;
+	case HNAE3_MODULE_TYPE_KR:
+		hclge_convert_setting_kr(speed_ability, mac->advertising);
+		break;
+	default:
+		break;
+	}
+}
+
+static void hclge_update_fec_advertising(struct hclge_mac *mac)
+{
+	if (mac->fec_mode & BIT(HNAE3_FEC_RS))
+		linkmode_set_bit(ETHTOOL_LINK_MODE_FEC_RS_BIT,
+				 mac->advertising);
+	else if (mac->fec_mode & BIT(HNAE3_FEC_LLRS))
+		linkmode_set_bit(ETHTOOL_LINK_MODE_FEC_LLRS_BIT,
+				 mac->advertising);
+	else if (mac->fec_mode & BIT(HNAE3_FEC_BASER))
+		linkmode_set_bit(ETHTOOL_LINK_MODE_FEC_BASER_BIT,
+				 mac->advertising);
+	else
+		linkmode_set_bit(ETHTOOL_LINK_MODE_FEC_NONE_BIT,
+				 mac->advertising);
+}
+
+static void hclge_update_pause_advertising(struct hclge_dev *hdev)
+{
+	struct hclge_mac *mac = &hdev->hw.mac;
+	bool rx_en, tx_en;
+
+	switch (hdev->fc_mode_last_time) {
+	case HCLGE_FC_RX_PAUSE:
+		rx_en = true;
+		tx_en = false;
+		break;
+	case HCLGE_FC_TX_PAUSE:
+		rx_en = false;
+		tx_en = true;
+		break;
+	case HCLGE_FC_FULL:
+		rx_en = true;
+		tx_en = true;
+		break;
+	default:
+		rx_en = false;
+		tx_en = false;
+		break;
+	}
+
+	linkmode_set_pause(mac->advertising, tx_en, rx_en);
+}
+
+static void hclge_update_advertising(struct hclge_dev *hdev)
+{
+	struct hclge_mac *mac = &hdev->hw.mac;
+
+	linkmode_zero(mac->advertising);
+	hclge_update_speed_advertising(mac);
+	hclge_update_fec_advertising(mac);
+	hclge_update_pause_advertising(hdev);
+}
+
+static void hclge_update_port_capability(struct hclge_dev *hdev,
+					 struct hclge_mac *mac)
+{
+	if (hnae3_dev_fec_supported(hdev))
+		hclge_convert_setting_fec(mac);
+
+	/* firmware can not identify back plane type, the media type
+	 * read from configuration can help deal it
+	 */
+	if (mac->media_type == HNAE3_MEDIA_TYPE_BACKPLANE &&
+	    mac->module_type == HNAE3_MODULE_TYPE_UNKNOWN)
+		mac->module_type = HNAE3_MODULE_TYPE_KR;
+	else if (mac->media_type == HNAE3_MEDIA_TYPE_COPPER)
+		mac->module_type = HNAE3_MODULE_TYPE_TP;
+
+	if (mac->support_autoneg) {
+		linkmode_set_bit(ETHTOOL_LINK_MODE_Autoneg_BIT, mac->supported);
+		linkmode_copy(mac->advertising, mac->supported);
+	} else {
+		linkmode_clear_bit(ETHTOOL_LINK_MODE_Autoneg_BIT,
+				   mac->supported);
+		hclge_update_advertising(hdev);
+	}
+}
+
+static int hclge_get_sfp_speed(struct hclge_dev *hdev, u32 *speed)
+{
+	struct hclge_sfp_info_cmd *resp;
+	struct hclge_desc desc;
+	int ret;
+
+	hclge_cmd_setup_basic_desc(&desc, HCLGE_OPC_GET_SFP_INFO, true);
+	resp = (struct hclge_sfp_info_cmd *)desc.data;
+	ret = hclge_cmd_send(&hdev->hw, &desc, 1);
+	if (ret == -EOPNOTSUPP) {
+		dev_warn(&hdev->pdev->dev,
+			 "IMP do not support get SFP speed %d\n", ret);
+		return ret;
+	} else if (ret) {
+		dev_err(&hdev->pdev->dev, "get sfp speed failed %d\n", ret);
+		return ret;
+	}
+
+	*speed = le32_to_cpu(resp->speed);
+
+	return 0;
+}
+
+static int hclge_get_sfp_info(struct hclge_dev *hdev, struct hclge_mac *mac)
+{
+	struct hclge_sfp_info_cmd *resp;
+	struct hclge_desc desc;
+	int ret;
+
+	hclge_cmd_setup_basic_desc(&desc, HCLGE_OPC_GET_SFP_INFO, true);
+	resp = (struct hclge_sfp_info_cmd *)desc.data;
+
+	resp->query_type = QUERY_ACTIVE_SPEED;
+
+	ret = hclge_cmd_send(&hdev->hw, &desc, 1);
+	if (ret == -EOPNOTSUPP) {
+		dev_warn(&hdev->pdev->dev,
+			 "IMP does not support get SFP info %d\n", ret);
+		return ret;
+	} else if (ret) {
+		dev_err(&hdev->pdev->dev, "get sfp info failed %d\n", ret);
+		return ret;
+	}
+
+	/* In some case, mac speed get from IMP may be 0, it shouldn't be
+	 * set to mac->speed.
+	 */
+	if (!le32_to_cpu(resp->speed))
+		return 0;
+
+	mac->speed = le32_to_cpu(resp->speed);
+	/* if resp->speed_ability is 0, it means it's an old version
+	 * firmware, do not update these params
+	 */
+	if (resp->speed_ability) {
+		mac->module_type = le32_to_cpu(resp->module_type);
+		mac->speed_ability = le32_to_cpu(resp->speed_ability);
+		mac->autoneg = resp->autoneg;
+		mac->support_autoneg = resp->autoneg_ability;
+		mac->speed_type = QUERY_ACTIVE_SPEED;
+		mac->lane_num = resp->lane_num;
+		if (!resp->active_fec)
+			mac->fec_mode = 0;
+		else
+			mac->fec_mode = BIT(resp->active_fec);
+		mac->fec_ability = resp->fec_ability;
+	} else {
+		mac->speed_type = QUERY_SFP_SPEED;
+	}
+
+	return 0;
+}
+
+static int hclge_get_phy_link_ksettings(struct hnae3_handle *handle,
+					struct ethtool_link_ksettings *cmd)
+{
+	struct hclge_desc desc[HCLGE_PHY_LINK_SETTING_BD_NUM];
+	struct hclge_vport *vport = hclge_get_vport(handle);
+	struct hclge_phy_link_ksetting_0_cmd *req0;
+	struct hclge_phy_link_ksetting_1_cmd *req1;
+	u32 supported, advertising, lp_advertising;
+	struct hclge_dev *hdev = vport->back;
+	int ret;
+
+	hclge_cmd_setup_basic_desc(&desc[0], HCLGE_OPC_PHY_LINK_KSETTING,
+				   true);
+	desc[0].flag |= cpu_to_le16(HCLGE_COMM_CMD_FLAG_NEXT);
+	hclge_cmd_setup_basic_desc(&desc[1], HCLGE_OPC_PHY_LINK_KSETTING,
+				   true);
+
+	ret = hclge_cmd_send(&hdev->hw, desc, HCLGE_PHY_LINK_SETTING_BD_NUM);
+	if (ret) {
+		dev_err(&hdev->pdev->dev,
+			"failed to get phy link ksetting, ret = %d.\n", ret);
+		return ret;
+	}
+
+	req0 = (struct hclge_phy_link_ksetting_0_cmd *)desc[0].data;
+	cmd->base.autoneg = req0->autoneg;
+	cmd->base.speed = le32_to_cpu(req0->speed);
+	cmd->base.duplex = req0->duplex;
+	cmd->base.port = req0->port;
+	cmd->base.transceiver = req0->transceiver;
+	cmd->base.phy_address = req0->phy_address;
+	cmd->base.eth_tp_mdix = req0->eth_tp_mdix;
+	cmd->base.eth_tp_mdix_ctrl = req0->eth_tp_mdix_ctrl;
+	supported = le32_to_cpu(req0->supported);
+	advertising = le32_to_cpu(req0->advertising);
+	lp_advertising = le32_to_cpu(req0->lp_advertising);
+	ethtool_convert_legacy_u32_to_link_mode(cmd->link_modes.supported,
+						supported);
+	ethtool_convert_legacy_u32_to_link_mode(cmd->link_modes.advertising,
+						advertising);
+	ethtool_convert_legacy_u32_to_link_mode(cmd->link_modes.lp_advertising,
+						lp_advertising);
+
+	req1 = (struct hclge_phy_link_ksetting_1_cmd *)desc[1].data;
+	cmd->base.master_slave_cfg = req1->master_slave_cfg;
+	cmd->base.master_slave_state = req1->master_slave_state;
+
+	return 0;
+}
+
+static int
+hclge_set_phy_link_ksettings(struct hnae3_handle *handle,
+			     const struct ethtool_link_ksettings *cmd)
+{
+	struct hclge_desc desc[HCLGE_PHY_LINK_SETTING_BD_NUM];
+	struct hclge_vport *vport = hclge_get_vport(handle);
+	struct hclge_phy_link_ksetting_0_cmd *req0;
+	struct hclge_phy_link_ksetting_1_cmd *req1;
+	struct hclge_dev *hdev = vport->back;
+	u32 advertising;
+	int ret;
+
+	if (cmd->base.autoneg == AUTONEG_DISABLE &&
+	    ((cmd->base.speed != SPEED_100 && cmd->base.speed != SPEED_10) ||
+	     (cmd->base.duplex != DUPLEX_HALF &&
+	      cmd->base.duplex != DUPLEX_FULL)))
+		return -EINVAL;
+
+	hclge_cmd_setup_basic_desc(&desc[0], HCLGE_OPC_PHY_LINK_KSETTING,
+				   false);
+	desc[0].flag |= cpu_to_le16(HCLGE_COMM_CMD_FLAG_NEXT);
+	hclge_cmd_setup_basic_desc(&desc[1], HCLGE_OPC_PHY_LINK_KSETTING,
+				   false);
+
+	req0 = (struct hclge_phy_link_ksetting_0_cmd *)desc[0].data;
+	req0->autoneg = cmd->base.autoneg;
+	req0->speed = cpu_to_le32(cmd->base.speed);
+	req0->duplex = cmd->base.duplex;
+	ethtool_convert_link_mode_to_legacy_u32(&advertising,
+						cmd->link_modes.advertising);
+	req0->advertising = cpu_to_le32(advertising);
+	req0->eth_tp_mdix_ctrl = cmd->base.eth_tp_mdix_ctrl;
+
+	req1 = (struct hclge_phy_link_ksetting_1_cmd *)desc[1].data;
+	req1->master_slave_cfg = cmd->base.master_slave_cfg;
+
+	ret = hclge_cmd_send(&hdev->hw, desc, HCLGE_PHY_LINK_SETTING_BD_NUM);
+	if (ret) {
+		dev_err(&hdev->pdev->dev,
+			"failed to set phy link ksettings, ret = %d.\n", ret);
+		return ret;
+	}
+
+	hdev->hw.mac.req_autoneg = cmd->base.autoneg;
+	hdev->hw.mac.req_speed = cmd->base.speed;
+	hdev->hw.mac.req_duplex = cmd->base.duplex;
+	linkmode_copy(hdev->hw.mac.advertising, cmd->link_modes.advertising);
+
+	return 0;
+}
+
+static int hclge_update_tp_port_info(struct hclge_dev *hdev)
+{
+	struct ethtool_link_ksettings cmd;
+	int ret;
+
+	if (!hnae3_dev_phy_imp_supported(hdev))
+		return 0;
+
+	ret = hclge_get_phy_link_ksettings(&hdev->vport->nic, &cmd);
+	if (ret)
+		return ret;
+
+	hdev->hw.mac.autoneg = cmd.base.autoneg;
+	hdev->hw.mac.speed = cmd.base.speed;
+	hdev->hw.mac.duplex = cmd.base.duplex;
+	linkmode_copy(hdev->hw.mac.advertising, cmd.link_modes.advertising);
+
+	return 0;
+}
+
+static int hclge_tp_port_init(struct hclge_dev *hdev)
+{
+	struct ethtool_link_ksettings cmd;
+
+	if (!hnae3_dev_phy_imp_supported(hdev))
+		return 0;
+
+	cmd.base.autoneg = hdev->hw.mac.req_autoneg;
+	cmd.base.speed = hdev->hw.mac.req_speed;
+	cmd.base.duplex = hdev->hw.mac.req_duplex;
+	linkmode_copy(cmd.link_modes.advertising, hdev->hw.mac.advertising);
+
+	return hclge_set_phy_link_ksettings(&hdev->vport->nic, &cmd);
+}
+
+static int hclge_update_port_info(struct hclge_dev *hdev)
+{
+	struct hclge_mac *mac = &hdev->hw.mac;
+	u32 speed;
+	int ret;
+
+	/* get the port info from SFP cmd if not copper port */
+	if (mac->media_type == HNAE3_MEDIA_TYPE_COPPER)
+		return hclge_update_tp_port_info(hdev);
+
+	/* if IMP does not support get SFP/qSFP info, return directly */
+	if (!hdev->support_sfp_query)
+		return 0;
+
+	if (hdev->ae_dev->dev_version >= HNAE3_DEVICE_VERSION_V2) {
+		speed = mac->speed;
+		ret = hclge_get_sfp_info(hdev, mac);
+	} else {
+		speed = HCLGE_MAC_SPEED_UNKNOWN;
+		ret = hclge_get_sfp_speed(hdev, &speed);
+	}
+
+	if (ret == -EOPNOTSUPP) {
+		hdev->support_sfp_query = false;
+		return ret;
+	} else if (ret) {
+		return ret;
+	}
+
+	if (hdev->ae_dev->dev_version >= HNAE3_DEVICE_VERSION_V2) {
+		if (mac->speed_type == QUERY_ACTIVE_SPEED) {
+			hclge_update_port_capability(hdev, mac);
+			if (mac->speed != speed)
+				(void)hclge_tm_port_shaper_cfg(hdev);
+			return 0;
+		}
+		return hclge_cfg_mac_speed_dup(hdev, mac->speed,
+					       HCLGE_MAC_FULL, mac->lane_num);
+	} else {
+		if (speed == HCLGE_MAC_SPEED_UNKNOWN)
+			return 0; /* do nothing if no SFP */
+
+		/* must config full duplex for SFP */
+		return hclge_cfg_mac_speed_dup(hdev, speed, HCLGE_MAC_FULL, 0);
+	}
+}
+
+static int hclge_get_status(struct hnae3_handle *handle)
+{
+	struct hclge_vport *vport = hclge_get_vport(handle);
+	struct hclge_dev *hdev = vport->back;
+
+	hclge_update_link_status(hdev);
+
+	return hdev->hw.mac.link;
+}
+
+struct hclge_vport *hclge_get_vf_vport(struct hclge_dev *hdev, int vf)
+{
+	if (!pci_num_vf(hdev->pdev)) {
+		dev_err(&hdev->pdev->dev,
+			"SRIOV is disabled, can not get vport(%d) info.\n", vf);
+		return NULL;
+	}
+
+	if (vf < 0 || vf >= pci_num_vf(hdev->pdev)) {
+		dev_err(&hdev->pdev->dev,
+			"vf id(%d) is out of range(0 <= vfid < %d)\n",
+			vf, pci_num_vf(hdev->pdev));
+		return NULL;
+	}
+
+	/* VF start from 1 in vport */
+	vf += HCLGE_VF_VPORT_START_NUM;
+	return &hdev->vport[vf];
+}
+
+static int hclge_get_vf_config(struct hnae3_handle *handle, int vf,
+			       struct ifla_vf_info *ivf)
+{
+	struct hclge_vport *vport = hclge_get_vport(handle);
+	struct hclge_dev *hdev = vport->back;
+
+	vport = hclge_get_vf_vport(hdev, vf);
+	if (!vport)
+		return -EINVAL;
+
+	ivf->vf = vf;
+	ivf->linkstate = vport->vf_info.link_state;
+	ivf->spoofchk = vport->vf_info.spoofchk;
+	ivf->trusted = vport->vf_info.trusted;
+	ivf->min_tx_rate = 0;
+	ivf->max_tx_rate = vport->vf_info.max_tx_rate;
+	ivf->vlan = vport->port_base_vlan_cfg.vlan_info.vlan_tag;
+	ivf->vlan_proto = htons(vport->port_base_vlan_cfg.vlan_info.vlan_proto);
+	ivf->qos = vport->port_base_vlan_cfg.vlan_info.qos;
+	ether_addr_copy(ivf->mac, vport->vf_info.mac);
+
+	return 0;
+}
+
+static int hclge_set_vf_link_state(struct hnae3_handle *handle, int vf,
+				   int link_state)
+{
+	struct hclge_vport *vport = hclge_get_vport(handle);
+	struct hclge_dev *hdev = vport->back;
+	int link_state_old;
+	int ret;
+
+	vport = hclge_get_vf_vport(hdev, vf);
+	if (!vport)
+		return -EINVAL;
+
+	link_state_old = vport->vf_info.link_state;
+	vport->vf_info.link_state = link_state;
+
+	/* return success directly if the VF is unalive, VF will
+	 * query link state itself when it starts work.
+	 */
+	if (!test_bit(HCLGE_VPORT_STATE_ALIVE, &vport->state))
+		return 0;
+
+	ret = hclge_push_vf_link_status(vport);
+	if (ret) {
+		vport->vf_info.link_state = link_state_old;
+		dev_err(&hdev->pdev->dev,
+			"failed to push vf%d link status, ret = %d\n", vf, ret);
+	}
+
+	return ret;
+}
+
+static void hclge_set_reset_pending(struct hclge_dev *hdev,
+				    enum hnae3_reset_type reset_type)
+{
+	/* When an incorrect reset type is executed, the get_reset_level
+	 * function generates the HNAE3_NONE_RESET flag. As a result, this
+	 * type do not need to pending.
+	 */
+	if (reset_type != HNAE3_NONE_RESET)
+		set_bit(reset_type, &hdev->reset_pending);
+}
+
+static u32 hclge_check_event_cause(struct hclge_dev *hdev, u32 *clearval)
+{
+	u32 cmdq_src_reg, msix_src_reg, hw_err_src_reg;
+
+	/* fetch the events from their corresponding regs */
+	cmdq_src_reg = hclge_read_dev(&hdev->hw, HCLGE_VECTOR0_CMDQ_SRC_REG);
+	msix_src_reg = hclge_read_dev(&hdev->hw, HCLGE_MISC_VECTOR_INT_STS);
+	hw_err_src_reg = hclge_read_dev(&hdev->hw,
+					HCLGE_RAS_PF_OTHER_INT_STS_REG);
+
+	/* Assumption: If by any chance reset and mailbox events are reported
+	 * together then we will only process reset event in this go and will
+	 * defer the processing of the mailbox events. Since, we would have not
+	 * cleared RX CMDQ event this time we would receive again another
+	 * interrupt from H/W just for the mailbox.
+	 *
+	 * check for vector0 reset event sources
+	 */
+	if (BIT(HCLGE_VECTOR0_IMPRESET_INT_B) & msix_src_reg) {
+		dev_info(&hdev->pdev->dev, "IMP reset interrupt\n");
+		hclge_set_reset_pending(hdev, HNAE3_IMP_RESET);
+		set_bit(HCLGE_COMM_STATE_CMD_DISABLE, &hdev->hw.hw.comm_state);
+		*clearval = BIT(HCLGE_VECTOR0_IMPRESET_INT_B);
+		hdev->rst_stats.imp_rst_cnt++;
+		return HCLGE_VECTOR0_EVENT_RST;
+	}
+
+	if (BIT(HCLGE_VECTOR0_GLOBALRESET_INT_B) & msix_src_reg) {
+		dev_info(&hdev->pdev->dev, "global reset interrupt\n");
+		set_bit(HCLGE_COMM_STATE_CMD_DISABLE, &hdev->hw.hw.comm_state);
+		hclge_set_reset_pending(hdev, HNAE3_GLOBAL_RESET);
+		*clearval = BIT(HCLGE_VECTOR0_GLOBALRESET_INT_B);
+		hdev->rst_stats.global_rst_cnt++;
+		return HCLGE_VECTOR0_EVENT_RST;
+	}
+
+	/* check for vector0 msix event and hardware error event source */
+	if (msix_src_reg & HCLGE_VECTOR0_REG_MSIX_MASK ||
+	    hw_err_src_reg & HCLGE_RAS_REG_ERR_MASK)
+		return HCLGE_VECTOR0_EVENT_ERR;
+
+	/* check for vector0 ptp event source */
+	if (BIT(HCLGE_VECTOR0_REG_PTP_INT_B) & msix_src_reg) {
+		*clearval = msix_src_reg;
+		return HCLGE_VECTOR0_EVENT_PTP;
+	}
+
+	/* check for vector0 mailbox(=CMDQ RX) event source */
+	if (BIT(HCLGE_VECTOR0_RX_CMDQ_INT_B) & cmdq_src_reg) {
+		cmdq_src_reg &= ~BIT(HCLGE_VECTOR0_RX_CMDQ_INT_B);
+		*clearval = cmdq_src_reg;
+		return HCLGE_VECTOR0_EVENT_MBX;
+	}
+
+	/* print other vector0 event source */
+	dev_info(&hdev->pdev->dev,
+		 "INT status: CMDQ(%#x) HW errors(%#x) other(%#x)\n",
+		 cmdq_src_reg, hw_err_src_reg, msix_src_reg);
+
+	return HCLGE_VECTOR0_EVENT_OTHER;
+}
+
+static void hclge_clear_event_cause(struct hclge_dev *hdev, u32 event_type,
+				    u32 regclr)
+{
+#define HCLGE_IMP_RESET_DELAY		5
+
+	switch (event_type) {
+	case HCLGE_VECTOR0_EVENT_PTP:
+	case HCLGE_VECTOR0_EVENT_RST:
+		if (regclr == BIT(HCLGE_VECTOR0_IMPRESET_INT_B))
+			mdelay(HCLGE_IMP_RESET_DELAY);
+
+		hclge_write_dev(&hdev->hw, HCLGE_MISC_RESET_STS_REG, regclr);
+		break;
+	case HCLGE_VECTOR0_EVENT_MBX:
+		hclge_write_dev(&hdev->hw, HCLGE_VECTOR0_CMDQ_SRC_REG, regclr);
+		break;
+	default:
+		break;
+	}
+}
+
+static void hclge_clear_all_event_cause(struct hclge_dev *hdev)
+{
+	hclge_clear_event_cause(hdev, HCLGE_VECTOR0_EVENT_RST,
+				BIT(HCLGE_VECTOR0_GLOBALRESET_INT_B) |
+				BIT(HCLGE_VECTOR0_CORERESET_INT_B) |
+				BIT(HCLGE_VECTOR0_IMPRESET_INT_B));
+	hclge_clear_event_cause(hdev, HCLGE_VECTOR0_EVENT_MBX, 0);
+}
+
+static void hclge_enable_vector(struct hclge_misc_vector *vector, bool enable)
+{
+	writel(enable ? 1 : 0, vector->addr);
+}
+
+static irqreturn_t hclge_misc_irq_handle(int irq, void *data)
+{
+	struct hclge_dev *hdev = data;
+	unsigned long flags;
+	u32 clearval = 0;
+	u32 event_cause;
+
+	hclge_enable_vector(&hdev->misc_vector, false);
+	event_cause = hclge_check_event_cause(hdev, &clearval);
+
+	/* vector 0 interrupt is shared with reset and mailbox source events. */
+	switch (event_cause) {
+	case HCLGE_VECTOR0_EVENT_ERR:
+		hclge_errhand_task_schedule(hdev);
+		break;
+	case HCLGE_VECTOR0_EVENT_RST:
+		hclge_reset_task_schedule(hdev);
+		break;
+	case HCLGE_VECTOR0_EVENT_PTP:
+		spin_lock_irqsave(&hdev->ptp->lock, flags);
+		hclge_ptp_clean_tx_hwts(hdev);
+		spin_unlock_irqrestore(&hdev->ptp->lock, flags);
+		break;
+	case HCLGE_VECTOR0_EVENT_MBX:
+		/* If we are here then,
+		 * 1. Either we are not handling any mbx task and we are not
+		 *    scheduled as well
+		 *                        OR
+		 * 2. We could be handling a mbx task but nothing more is
+		 *    scheduled.
+		 * In both cases, we should schedule mbx task as there are more
+		 * mbx messages reported by this interrupt.
+		 */
+		hclge_mbx_task_schedule(hdev);
+		break;
+	default:
+		dev_warn(&hdev->pdev->dev,
+			 "received unknown or unhandled event of vector0\n");
+		break;
+	}
+
+	hclge_clear_event_cause(hdev, event_cause, clearval);
+
+	/* Enable interrupt if it is not caused by reset event or error event */
+	if (event_cause == HCLGE_VECTOR0_EVENT_PTP ||
+	    event_cause == HCLGE_VECTOR0_EVENT_MBX ||
+	    event_cause == HCLGE_VECTOR0_EVENT_OTHER)
+		hclge_enable_vector(&hdev->misc_vector, true);
+
+	return IRQ_HANDLED;
+}
+
+static void hclge_free_vector(struct hclge_dev *hdev, int vector_id)
+{
+	if (hdev->vector_status[vector_id] == HCLGE_INVALID_VPORT) {
+		dev_warn(&hdev->pdev->dev,
+			 "vector(vector_id %d) has been freed.\n", vector_id);
+		return;
+	}
+
+	hdev->vector_status[vector_id] = HCLGE_INVALID_VPORT;
+	hdev->num_msi_left += 1;
+	hdev->num_msi_used -= 1;
+}
+
+static void hclge_get_misc_vector(struct hclge_dev *hdev)
+{
+	struct hclge_misc_vector *vector = &hdev->misc_vector;
+
+	vector->vector_irq = pci_irq_vector(hdev->pdev, 0);
+
+	vector->addr = hdev->hw.hw.io_base + HCLGE_MISC_VECTOR_REG_BASE;
+	hdev->vector_status[0] = 0;
+
+	hdev->num_msi_left -= 1;
+	hdev->num_msi_used += 1;
+}
+
+static int hclge_misc_irq_init(struct hclge_dev *hdev)
+{
+	int ret;
+
+	hclge_get_misc_vector(hdev);
+
+	/* this would be explicitly freed in the end */
+	snprintf(hdev->misc_vector.name, HNAE3_INT_NAME_LEN, "%s-misc-%s",
+		 HCLGE_NAME, pci_name(hdev->pdev));
+	ret = request_irq(hdev->misc_vector.vector_irq, hclge_misc_irq_handle,
+			  IRQF_NO_AUTOEN, hdev->misc_vector.name, hdev);
+	if (ret) {
+		hclge_free_vector(hdev, 0);
+		dev_err(&hdev->pdev->dev, "request misc irq(%d) fail\n",
+			hdev->misc_vector.vector_irq);
+	}
+
+	return ret;
+}
+
+static void hclge_misc_irq_uninit(struct hclge_dev *hdev)
+{
+	free_irq(hdev->misc_vector.vector_irq, hdev);
+	hclge_free_vector(hdev, 0);
+}
+
+int hclge_notify_client(struct hclge_dev *hdev,
+			enum hnae3_reset_notify_type type)
+{
+	struct hnae3_handle *handle = &hdev->vport[0].nic;
+	struct hnae3_client *client = hdev->nic_client;
+	int ret;
+
+	if (!test_bit(HCLGE_STATE_NIC_REGISTERED, &hdev->state) || !client)
+		return 0;
+
+	if (!client->ops->reset_notify)
+		return -EOPNOTSUPP;
+
+	ret = client->ops->reset_notify(handle, type);
+	if (ret)
+		dev_err(&hdev->pdev->dev, "notify nic client failed %d(%d)\n",
+			type, ret);
+
+	return ret;
+}
+
+static int hclge_notify_roce_client(struct hclge_dev *hdev,
+				    enum hnae3_reset_notify_type type)
+{
+	struct hnae3_handle *handle = &hdev->vport[0].roce;
+	struct hnae3_client *client = hdev->roce_client;
+	int ret;
+
+	if (!test_bit(HCLGE_STATE_ROCE_REGISTERED, &hdev->state) || !client)
+		return 0;
+
+	if (!client->ops->reset_notify)
+		return -EOPNOTSUPP;
+
+	ret = client->ops->reset_notify(handle, type);
+	if (ret)
+		dev_err(&hdev->pdev->dev, "notify roce client failed %d(%d)",
+			type, ret);
+
+	return ret;
+}
+
+static int hclge_reset_wait(struct hclge_dev *hdev)
+{
+#define HCLGE_RESET_WATI_MS	100
+#define HCLGE_RESET_WAIT_CNT	350
+
+	u32 val, reg, reg_bit;
+	u32 cnt = 0;
+
+	switch (hdev->reset_type) {
+	case HNAE3_IMP_RESET:
+		reg = HCLGE_GLOBAL_RESET_REG;
+		reg_bit = HCLGE_IMP_RESET_BIT;
+		break;
+	case HNAE3_GLOBAL_RESET:
+		reg = HCLGE_GLOBAL_RESET_REG;
+		reg_bit = HCLGE_GLOBAL_RESET_BIT;
+		break;
+	case HNAE3_FUNC_RESET:
+		reg = HCLGE_FUN_RST_ING;
+		reg_bit = HCLGE_FUN_RST_ING_B;
+		break;
+	default:
+		dev_err(&hdev->pdev->dev,
+			"Wait for unsupported reset type: %d\n",
+			hdev->reset_type);
+		return -EINVAL;
+	}
+
+	val = hclge_read_dev(&hdev->hw, reg);
+	while (hnae3_get_bit(val, reg_bit) && cnt < HCLGE_RESET_WAIT_CNT) {
+		msleep(HCLGE_RESET_WATI_MS);
+		val = hclge_read_dev(&hdev->hw, reg);
+		cnt++;
+	}
+
+	if (cnt >= HCLGE_RESET_WAIT_CNT) {
+		dev_warn(&hdev->pdev->dev,
+			 "Wait for reset timeout: %d\n", hdev->reset_type);
+		return -EBUSY;
+	}
+
+	return 0;
+}
+
+static int hclge_set_vf_rst(struct hclge_dev *hdev, int func_id, bool reset)
+{
+	struct hclge_vf_rst_cmd *req;
+	struct hclge_desc desc;
+
+	req = (struct hclge_vf_rst_cmd *)desc.data;
+	hclge_cmd_setup_basic_desc(&desc, HCLGE_OPC_GBL_RST_STATUS, false);
+	req->dest_vfid = func_id;
+
+	if (reset)
+		req->vf_rst = 0x1;
+
+	return hclge_cmd_send(&hdev->hw, &desc, 1);
+}
+
+static int hclge_set_all_vf_rst(struct hclge_dev *hdev, bool reset)
+{
+	int i;
+
+	for (i = HCLGE_VF_VPORT_START_NUM; i < hdev->num_alloc_vport; i++) {
+		struct hclge_vport *vport = &hdev->vport[i];
+		int ret;
+
+		/* Send cmd to set/clear VF's FUNC_RST_ING */
+		ret = hclge_set_vf_rst(hdev, vport->vport_id, reset);
+		if (ret) {
+			dev_err(&hdev->pdev->dev,
+				"set vf(%u) rst failed %d!\n",
+				vport->vport_id - HCLGE_VF_VPORT_START_NUM,
+				ret);
+			return ret;
+		}
+
+		if (!reset ||
+		    !test_bit(HCLGE_VPORT_STATE_INITED, &vport->state))
+			continue;
+
+		if (!test_bit(HCLGE_VPORT_STATE_ALIVE, &vport->state) &&
+		    hdev->reset_type == HNAE3_FUNC_RESET) {
+			set_bit(HCLGE_VPORT_NEED_NOTIFY_RESET,
+				&vport->need_notify);
+			continue;
+		}
+
+		/* Inform VF to process the reset.
+		 * hclge_inform_reset_assert_to_vf may fail if VF
+		 * driver is not loaded.
+		 */
+		ret = hclge_inform_reset_assert_to_vf(vport);
+		if (ret)
+			dev_warn(&hdev->pdev->dev,
+				 "inform reset to vf(%u) failed %d!\n",
+				 vport->vport_id - HCLGE_VF_VPORT_START_NUM,
+				 ret);
+	}
+
+	return 0;
+}
+
+static void hclge_mailbox_service_task(struct hclge_dev *hdev)
+{
+	if (!test_and_clear_bit(HCLGE_STATE_MBX_SERVICE_SCHED, &hdev->state) ||
+	    test_bit(HCLGE_COMM_STATE_CMD_DISABLE, &hdev->hw.hw.comm_state) ||
+	    test_and_set_bit(HCLGE_STATE_MBX_HANDLING, &hdev->state))
+		return;
+
+	if (time_is_before_jiffies(hdev->last_mbx_scheduled +
+				   HCLGE_MBX_SCHED_TIMEOUT))
+		dev_warn(&hdev->pdev->dev,
+			 "mbx service task is scheduled after %ums on cpu%u!\n",
+			 jiffies_to_msecs(jiffies - hdev->last_mbx_scheduled),
+			 smp_processor_id());
+
+	hclge_mbx_handler(hdev);
+
+	clear_bit(HCLGE_STATE_MBX_HANDLING, &hdev->state);
+}
+
+static void hclge_func_reset_sync_vf(struct hclge_dev *hdev)
+{
+	struct hclge_pf_rst_sync_cmd *req;
+	struct hclge_desc desc;
+	int cnt = 0;
+	int ret;
+
+	req = (struct hclge_pf_rst_sync_cmd *)desc.data;
+	hclge_cmd_setup_basic_desc(&desc, HCLGE_OPC_QUERY_VF_RST_RDY, true);
+
+	do {
+		/* vf need to down netdev by mbx during PF or FLR reset */
+		hclge_mailbox_service_task(hdev);
+
+		ret = hclge_cmd_send(&hdev->hw, &desc, 1);
+		/* for compatible with old firmware, wait
+		 * 100 ms for VF to stop IO
+		 */
+		if (ret == -EOPNOTSUPP) {
+			msleep(HCLGE_RESET_SYNC_TIME);
+			return;
+		} else if (ret) {
+			dev_warn(&hdev->pdev->dev, "sync with VF fail %d!\n",
+				 ret);
+			return;
+		} else if (req->all_vf_ready) {
+			return;
+		}
+		msleep(HCLGE_PF_RESET_SYNC_TIME);
+		hclge_comm_cmd_reuse_desc(&desc, true);
+	} while (cnt++ < HCLGE_PF_RESET_SYNC_CNT);
+
+	dev_warn(&hdev->pdev->dev, "sync with VF timeout!\n");
+}
+
+void hclge_report_hw_error(struct hclge_dev *hdev,
+			   enum hnae3_hw_error_type type)
+{
+	struct hnae3_client *client = hdev->nic_client;
+
+	if (!client || !client->ops->process_hw_error ||
+	    !test_bit(HCLGE_STATE_NIC_REGISTERED, &hdev->state))
+		return;
+
+	client->ops->process_hw_error(&hdev->vport[0].nic, type);
+}
+
+static void hclge_handle_imp_error(struct hclge_dev *hdev)
+{
+	u32 reg_val;
+
+	reg_val = hclge_read_dev(&hdev->hw, HCLGE_PF_OTHER_INT_REG);
+	if (reg_val & BIT(HCLGE_VECTOR0_IMP_RD_POISON_B)) {
+		hclge_report_hw_error(hdev, HNAE3_IMP_RD_POISON_ERROR);
+		reg_val &= ~BIT(HCLGE_VECTOR0_IMP_RD_POISON_B);
+		hclge_write_dev(&hdev->hw, HCLGE_PF_OTHER_INT_REG, reg_val);
+	}
+
+	if (reg_val & BIT(HCLGE_VECTOR0_IMP_CMDQ_ERR_B)) {
+		hclge_report_hw_error(hdev, HNAE3_CMDQ_ECC_ERROR);
+		reg_val &= ~BIT(HCLGE_VECTOR0_IMP_CMDQ_ERR_B);
+		hclge_write_dev(&hdev->hw, HCLGE_PF_OTHER_INT_REG, reg_val);
+	}
+}
+
+int hclge_func_reset_cmd(struct hclge_dev *hdev, int func_id)
+{
+	struct hclge_desc desc;
+	struct hclge_reset_cmd *req = (struct hclge_reset_cmd *)desc.data;
+	int ret;
+
+	hclge_cmd_setup_basic_desc(&desc, HCLGE_OPC_CFG_RST_TRIGGER, false);
+	hnae3_set_bit(req->mac_func_reset, HCLGE_CFG_RESET_FUNC_B, 1);
+	req->fun_reset_vfid = func_id;
+
+	ret = hclge_cmd_send(&hdev->hw, &desc, 1);
+	if (ret)
+		dev_err(&hdev->pdev->dev,
+			"send function reset cmd fail, status =%d\n", ret);
+
+	return ret;
+}
+
+static void hclge_do_reset(struct hclge_dev *hdev)
+{
+	struct hnae3_handle *handle = &hdev->vport[0].nic;
+	struct pci_dev *pdev = hdev->pdev;
+	u32 val;
+
+	if (hclge_get_hw_reset_stat(handle)) {
+		dev_info(&pdev->dev, "hardware reset not finish\n");
+		dev_info(&pdev->dev, "func_rst_reg:0x%x, global_rst_reg:0x%x\n",
+			 hclge_read_dev(&hdev->hw, HCLGE_FUN_RST_ING),
+			 hclge_read_dev(&hdev->hw, HCLGE_GLOBAL_RESET_REG));
+		return;
+	}
+
+	switch (hdev->reset_type) {
+	case HNAE3_IMP_RESET:
+		dev_info(&pdev->dev, "IMP reset requested\n");
+		val = hclge_read_dev(&hdev->hw, HCLGE_PF_OTHER_INT_REG);
+		hnae3_set_bit(val, HCLGE_TRIGGER_IMP_RESET_B, 1);
+		hclge_write_dev(&hdev->hw, HCLGE_PF_OTHER_INT_REG, val);
+		break;
+	case HNAE3_GLOBAL_RESET:
+		dev_info(&pdev->dev, "global reset requested\n");
+		val = hclge_read_dev(&hdev->hw, HCLGE_GLOBAL_RESET_REG);
+		hnae3_set_bit(val, HCLGE_GLOBAL_RESET_BIT, 1);
+		hclge_write_dev(&hdev->hw, HCLGE_GLOBAL_RESET_REG, val);
+		break;
+	case HNAE3_FUNC_RESET:
+		dev_info(&pdev->dev, "PF reset requested\n");
+		/* schedule again to check later */
+		hclge_set_reset_pending(hdev, HNAE3_FUNC_RESET);
+		hclge_reset_task_schedule(hdev);
+		break;
+	default:
+		dev_warn(&pdev->dev,
+			 "unsupported reset type: %d\n", hdev->reset_type);
+		break;
+	}
+}
+
+static enum hnae3_reset_type hclge_get_reset_level(struct hnae3_ae_dev *ae_dev,
+						   unsigned long *addr)
+{
+	enum hnae3_reset_type rst_level = HNAE3_NONE_RESET;
+	struct hclge_dev *hdev = ae_dev->priv;
+
+	/* return the highest priority reset level amongst all */
+	if (test_bit(HNAE3_IMP_RESET, addr)) {
+		rst_level = HNAE3_IMP_RESET;
+		clear_bit(HNAE3_IMP_RESET, addr);
+		clear_bit(HNAE3_GLOBAL_RESET, addr);
+		clear_bit(HNAE3_FUNC_RESET, addr);
+	} else if (test_bit(HNAE3_GLOBAL_RESET, addr)) {
+		rst_level = HNAE3_GLOBAL_RESET;
+		clear_bit(HNAE3_GLOBAL_RESET, addr);
+		clear_bit(HNAE3_FUNC_RESET, addr);
+	} else if (test_bit(HNAE3_FUNC_RESET, addr)) {
+		rst_level = HNAE3_FUNC_RESET;
+		clear_bit(HNAE3_FUNC_RESET, addr);
+	} else if (test_bit(HNAE3_FLR_RESET, addr)) {
+		rst_level = HNAE3_FLR_RESET;
+		clear_bit(HNAE3_FLR_RESET, addr);
+	}
+
+	clear_bit(HNAE3_NONE_RESET, addr);
+
+	if (hdev->reset_type != HNAE3_NONE_RESET &&
+	    rst_level < hdev->reset_type)
+		return HNAE3_NONE_RESET;
+
+	return rst_level;
+}
+
+static void hclge_clear_reset_cause(struct hclge_dev *hdev)
+{
+	u32 clearval = 0;
+
+	switch (hdev->reset_type) {
+	case HNAE3_IMP_RESET:
+		clearval = BIT(HCLGE_VECTOR0_IMPRESET_INT_B);
+		break;
+	case HNAE3_GLOBAL_RESET:
+		clearval = BIT(HCLGE_VECTOR0_GLOBALRESET_INT_B);
+		break;
+	default:
+		break;
+	}
+
+	if (!clearval)
+		return;
+
+	/* For revision 0x20, the reset interrupt source
+	 * can only be cleared after hardware reset done
+	 */
+	if (hdev->ae_dev->dev_version < HNAE3_DEVICE_VERSION_V2)
+		hclge_write_dev(&hdev->hw, HCLGE_MISC_RESET_STS_REG,
+				clearval);
+
+	hclge_enable_vector(&hdev->misc_vector, true);
+}
+
+static void hclge_reset_handshake(struct hclge_dev *hdev, bool enable)
+{
+	u32 reg_val;
+
+	reg_val = hclge_read_dev(&hdev->hw, HCLGE_COMM_NIC_CSQ_DEPTH_REG);
+	if (enable)
+		reg_val |= HCLGE_COMM_NIC_SW_RST_RDY;
+	else
+		reg_val &= ~HCLGE_COMM_NIC_SW_RST_RDY;
+
+	hclge_write_dev(&hdev->hw, HCLGE_COMM_NIC_CSQ_DEPTH_REG, reg_val);
+}
+
+static int hclge_func_reset_notify_vf(struct hclge_dev *hdev)
+{
+	int ret;
+
+	ret = hclge_set_all_vf_rst(hdev, true);
+	if (ret)
+		return ret;
+
+	hclge_func_reset_sync_vf(hdev);
+
+	return 0;
+}
+
+static int hclge_reset_prepare_wait(struct hclge_dev *hdev)
+{
+	u32 reg_val;
+	int ret = 0;
+
+	switch (hdev->reset_type) {
+	case HNAE3_FUNC_RESET:
+		ret = hclge_func_reset_notify_vf(hdev);
+		if (ret)
+			return ret;
+
+		ret = hclge_func_reset_cmd(hdev, 0);
+		if (ret) {
+			dev_err(&hdev->pdev->dev,
+				"asserting function reset fail %d!\n", ret);
+			return ret;
+		}
+
+		/* After performaning pf reset, it is not necessary to do the
+		 * mailbox handling or send any command to firmware, because
+		 * any mailbox handling or command to firmware is only valid
+		 * after hclge_comm_cmd_init is called.
+		 */
+		set_bit(HCLGE_COMM_STATE_CMD_DISABLE, &hdev->hw.hw.comm_state);
+		hdev->rst_stats.pf_rst_cnt++;
+		break;
+	case HNAE3_FLR_RESET:
+		ret = hclge_func_reset_notify_vf(hdev);
+		if (ret)
+			return ret;
+		break;
+	case HNAE3_IMP_RESET:
+		hclge_handle_imp_error(hdev);
+		reg_val = hclge_read_dev(&hdev->hw, HCLGE_PF_OTHER_INT_REG);
+		hclge_write_dev(&hdev->hw, HCLGE_PF_OTHER_INT_REG,
+				BIT(HCLGE_VECTOR0_IMP_RESET_INT_B) | reg_val);
+		break;
+	default:
+		break;
+	}
+
+	/* inform hardware that preparatory work is done */
+	msleep(HCLGE_RESET_SYNC_TIME);
+	hclge_reset_handshake(hdev, true);
+	dev_info(&hdev->pdev->dev, "prepare wait ok\n");
+
+	return ret;
+}
+
+static void hclge_show_rst_info(struct hclge_dev *hdev)
+{
+	char *buf;
+
+	buf = kzalloc(HCLGE_DBG_RESET_INFO_LEN, GFP_KERNEL);
+	if (!buf)
+		return;
+
+	hclge_dbg_dump_rst_info(hdev, buf, HCLGE_DBG_RESET_INFO_LEN);
+
+	dev_info(&hdev->pdev->dev, "dump reset info:\n%s", buf);
+
+	kfree(buf);
+}
+
+static bool hclge_reset_err_handle(struct hclge_dev *hdev)
+{
+#define MAX_RESET_FAIL_CNT 5
+
+	if (hdev->reset_pending) {
+		dev_info(&hdev->pdev->dev, "Reset pending %lu\n",
+			 hdev->reset_pending);
+		return true;
+	} else if (hclge_read_dev(&hdev->hw, HCLGE_MISC_VECTOR_INT_STS) &
+		   HCLGE_RESET_INT_M) {
+		dev_info(&hdev->pdev->dev,
+			 "reset failed because new reset interrupt\n");
+		hclge_clear_reset_cause(hdev);
+		return false;
+	} else if (hdev->rst_stats.reset_fail_cnt < MAX_RESET_FAIL_CNT) {
+		hdev->rst_stats.reset_fail_cnt++;
+		hclge_set_reset_pending(hdev, hdev->reset_type);
+		dev_info(&hdev->pdev->dev,
+			 "re-schedule reset task(%u)\n",
+			 hdev->rst_stats.reset_fail_cnt);
+		return true;
+	}
+
+	hclge_clear_reset_cause(hdev);
+
+	/* recover the handshake status when reset fail */
+	hclge_reset_handshake(hdev, true);
+
+	dev_err(&hdev->pdev->dev, "Reset fail!\n");
+
+	hclge_show_rst_info(hdev);
+
+	set_bit(HCLGE_STATE_RST_FAIL, &hdev->state);
+
+	return false;
+}
+
+static void hclge_update_reset_level(struct hclge_dev *hdev)
+{
+	struct hnae3_ae_dev *ae_dev = pci_get_drvdata(hdev->pdev);
+	enum hnae3_reset_type reset_level;
+
+	/* reset request will not be set during reset, so clear
+	 * pending reset request to avoid unnecessary reset
+	 * caused by the same reason.
+	 */
+	hclge_get_reset_level(ae_dev, &hdev->reset_request);
+
+	/* if default_reset_request has a higher level reset request,
+	 * it should be handled as soon as possible. since some errors
+	 * need this kind of reset to fix.
+	 */
+	reset_level = hclge_get_reset_level(ae_dev,
+					    &hdev->default_reset_request);
+	if (reset_level != HNAE3_NONE_RESET)
+		set_bit(reset_level, &hdev->reset_request);
+}
+
+static int hclge_set_rst_done(struct hclge_dev *hdev)
+{
+	struct hclge_pf_rst_done_cmd *req;
+	struct hclge_desc desc;
+	int ret;
+
+	req = (struct hclge_pf_rst_done_cmd *)desc.data;
+	hclge_cmd_setup_basic_desc(&desc, HCLGE_OPC_PF_RST_DONE, false);
+	req->pf_rst_done |= HCLGE_PF_RESET_DONE_BIT;
+
+	ret = hclge_cmd_send(&hdev->hw, &desc, 1);
+	/* To be compatible with the old firmware, which does not support
+	 * command HCLGE_OPC_PF_RST_DONE, just print a warning and
+	 * return success
+	 */
+	if (ret == -EOPNOTSUPP) {
+		dev_warn(&hdev->pdev->dev,
+			 "current firmware does not support command(0x%x)!\n",
+			 HCLGE_OPC_PF_RST_DONE);
+		return 0;
+	} else if (ret) {
+		dev_err(&hdev->pdev->dev, "assert PF reset done fail %d!\n",
+			ret);
+	}
+
+	return ret;
+}
+
+static int hclge_reset_prepare_up(struct hclge_dev *hdev)
+{
+	int ret = 0;
+
+	switch (hdev->reset_type) {
+	case HNAE3_FUNC_RESET:
+	case HNAE3_FLR_RESET:
+		ret = hclge_set_all_vf_rst(hdev, false);
+		break;
+	case HNAE3_GLOBAL_RESET:
+	case HNAE3_IMP_RESET:
+		ret = hclge_set_rst_done(hdev);
+		break;
+	default:
+		break;
+	}
+
+	/* clear up the handshake status after re-initialize done */
+	hclge_reset_handshake(hdev, false);
+
+	return ret;
+}
+
+static int hclge_reset_stack(struct hclge_dev *hdev)
+{
+	int ret;
+
+	ret = hclge_notify_client(hdev, HNAE3_UNINIT_CLIENT);
+	if (ret)
+		return ret;
+
+	ret = hclge_reset_ae_dev(hdev->ae_dev);
+	if (ret)
+		return ret;
+
+	return hclge_notify_client(hdev, HNAE3_INIT_CLIENT);
+}
+
+static int hclge_reset_prepare(struct hclge_dev *hdev)
+{
+	int ret;
+
+	hdev->rst_stats.reset_cnt++;
+	/* perform reset of the stack & ae device for a client */
+	ret = hclge_notify_roce_client(hdev, HNAE3_DOWN_CLIENT);
+	if (ret)
+		return ret;
+
+	rtnl_lock();
+	ret = hclge_notify_client(hdev, HNAE3_DOWN_CLIENT);
+	rtnl_unlock();
+	if (ret)
+		return ret;
+
+	return hclge_reset_prepare_wait(hdev);
+}
+
+static int hclge_reset_rebuild(struct hclge_dev *hdev)
+{
+	int ret;
+
+	hdev->rst_stats.hw_reset_done_cnt++;
+
+	ret = hclge_notify_roce_client(hdev, HNAE3_UNINIT_CLIENT);
+	if (ret)
+		return ret;
+
+	rtnl_lock();
+	ret = hclge_reset_stack(hdev);
+	rtnl_unlock();
+	if (ret)
+		return ret;
+
+	hclge_clear_reset_cause(hdev);
+
+	ret = hclge_notify_roce_client(hdev, HNAE3_INIT_CLIENT);
+	/* ignore RoCE notify error if it fails HCLGE_RESET_MAX_FAIL_CNT - 1
+	 * times
+	 */
+	if (ret &&
+	    hdev->rst_stats.reset_fail_cnt < HCLGE_RESET_MAX_FAIL_CNT - 1)
+		return ret;
+
+	ret = hclge_reset_prepare_up(hdev);
+	if (ret)
+		return ret;
+
+	rtnl_lock();
+	ret = hclge_notify_client(hdev, HNAE3_UP_CLIENT);
+	rtnl_unlock();
+	if (ret)
+		return ret;
+
+	ret = hclge_notify_roce_client(hdev, HNAE3_UP_CLIENT);
+	if (ret)
+		return ret;
+
+	hdev->last_reset_time = jiffies;
+	hdev->rst_stats.reset_fail_cnt = 0;
+	hdev->rst_stats.reset_done_cnt++;
+	clear_bit(HCLGE_STATE_RST_FAIL, &hdev->state);
+
+	hclge_update_reset_level(hdev);
+
+	return 0;
+}
+
+static void hclge_reset(struct hclge_dev *hdev)
+{
+	if (hclge_reset_prepare(hdev))
+		goto err_reset;
+
+	if (hclge_reset_wait(hdev))
+		goto err_reset;
+
+	if (hclge_reset_rebuild(hdev))
+		goto err_reset;
+
+	return;
+
+err_reset:
+	if (hclge_reset_err_handle(hdev))
+		hclge_reset_task_schedule(hdev);
+}
+
+static void hclge_reset_event(struct pci_dev *pdev, struct hnae3_handle *handle)
+{
+	struct hnae3_ae_dev *ae_dev = pci_get_drvdata(pdev);
+	struct hclge_dev *hdev = ae_dev->priv;
+
+	/* We might end up getting called broadly because of 2 below cases:
+	 * 1. Recoverable error was conveyed through APEI and only way to bring
+	 *    normalcy is to reset.
+	 * 2. A new reset request from the stack due to timeout
+	 *
+	 * check if this is a new reset request and we are not here just because
+	 * last reset attempt did not succeed and watchdog hit us again. We will
+	 * know this if last reset request did not occur very recently (watchdog
+	 * timer = 5*HZ, let us check after sufficiently large time, say 4*5*Hz)
+	 * In case of new request we reset the "reset level" to PF reset.
+	 * And if it is a repeat reset request of the most recent one then we
+	 * want to make sure we throttle the reset request. Therefore, we will
+	 * not allow it again before 3*HZ times.
+	 */
+
+	if (time_before(jiffies, (hdev->last_reset_time +
+				  HCLGE_RESET_INTERVAL))) {
+		mod_timer(&hdev->reset_timer, jiffies + HCLGE_RESET_INTERVAL);
+		return;
+	}
+
+	if (hdev->default_reset_request) {
+		hdev->reset_level =
+			hclge_get_reset_level(ae_dev,
+					      &hdev->default_reset_request);
+	} else if (time_after(jiffies, (hdev->last_reset_time + 4 * 5 * HZ))) {
+		hdev->reset_level = HNAE3_FUNC_RESET;
+	}
+
+	dev_info(&hdev->pdev->dev, "received reset event, reset type is %d\n",
+		 hdev->reset_level);
+
+	/* request reset & schedule reset task */
+	set_bit(hdev->reset_level, &hdev->reset_request);
+	hclge_reset_task_schedule(hdev);
+
+	if (hdev->reset_level < HNAE3_GLOBAL_RESET)
+		hdev->reset_level++;
+}
+
+static void hclge_set_def_reset_request(struct hnae3_ae_dev *ae_dev,
+					enum hnae3_reset_type rst_type)
+{
+#define HCLGE_SUPPORT_RESET_TYPE \
+	(BIT(HNAE3_FLR_RESET) | BIT(HNAE3_FUNC_RESET) | \
+	BIT(HNAE3_GLOBAL_RESET) | BIT(HNAE3_IMP_RESET))
+
+	struct hclge_dev *hdev = ae_dev->priv;
+
+	if (!(BIT(rst_type) & HCLGE_SUPPORT_RESET_TYPE)) {
+		/* To prevent reset triggered by hclge_reset_event */
+		set_bit(HNAE3_NONE_RESET, &hdev->default_reset_request);
+		dev_warn(&hdev->pdev->dev, "unsupported reset type %d\n",
+			 rst_type);
+		return;
+	}
+
+	set_bit(rst_type, &hdev->default_reset_request);
+}
+
+static void hclge_reset_timer(struct timer_list *t)
+{
+	struct hclge_dev *hdev = timer_container_of(hdev, t, reset_timer);
+
+	/* if default_reset_request has no value, it means that this reset
+	 * request has already be handled, so just return here
+	 */
+	if (!hdev->default_reset_request)
+		return;
+
+	dev_info(&hdev->pdev->dev,
+		 "triggering reset in reset timer\n");
+	hclge_reset_event(hdev->pdev, NULL);
+}
+
+static void hclge_reset_subtask(struct hclge_dev *hdev)
+{
+	struct hnae3_ae_dev *ae_dev = pci_get_drvdata(hdev->pdev);
+
+	/* check if there is any ongoing reset in the hardware. This status can
+	 * be checked from reset_pending. If there is then, we need to wait for
+	 * hardware to complete reset.
+	 *    a. If we are able to figure out in reasonable time that hardware
+	 *       has fully resetted then, we can proceed with driver, client
+	 *       reset.
+	 *    b. else, we can come back later to check this status so re-sched
+	 *       now.
+	 */
+	hdev->last_reset_time = jiffies;
+	hdev->reset_type = hclge_get_reset_level(ae_dev, &hdev->reset_pending);
+	if (hdev->reset_type != HNAE3_NONE_RESET)
+		hclge_reset(hdev);
+
+	/* check if we got any *new* reset requests to be honored */
+	hdev->reset_type = hclge_get_reset_level(ae_dev, &hdev->reset_request);
+	if (hdev->reset_type != HNAE3_NONE_RESET)
+		hclge_do_reset(hdev);
+
+	hdev->reset_type = HNAE3_NONE_RESET;
+}
+
+static void hclge_handle_err_reset_request(struct hclge_dev *hdev)
+{
+	struct hnae3_ae_dev *ae_dev = pci_get_drvdata(hdev->pdev);
+	enum hnae3_reset_type reset_type;
+
+	if (ae_dev->hw_err_reset_req) {
+		reset_type = hclge_get_reset_level(ae_dev,
+						   &ae_dev->hw_err_reset_req);
+		hclge_set_def_reset_request(ae_dev, reset_type);
+	}
+
+	if (hdev->default_reset_request && ae_dev->ops->reset_event)
+		ae_dev->ops->reset_event(hdev->pdev, NULL);
+
+	/* enable interrupt after error handling complete */
+	hclge_enable_vector(&hdev->misc_vector, true);
+}
+
+static void hclge_handle_err_recovery(struct hclge_dev *hdev)
+{
+	struct hnae3_ae_dev *ae_dev = pci_get_drvdata(hdev->pdev);
+
+	ae_dev->hw_err_reset_req = 0;
+
+	if (hclge_find_error_source(hdev)) {
+		hclge_handle_error_info_log(ae_dev);
+		hclge_handle_mac_tnl(hdev);
+		hclge_handle_vf_queue_err_ras(hdev);
+	}
+
+	hclge_handle_err_reset_request(hdev);
+}
+
+static void hclge_misc_err_recovery(struct hclge_dev *hdev)
+{
+	struct hnae3_ae_dev *ae_dev = pci_get_drvdata(hdev->pdev);
+	struct device *dev = &hdev->pdev->dev;
+	u32 msix_sts_reg;
+
+	msix_sts_reg = hclge_read_dev(&hdev->hw, HCLGE_MISC_VECTOR_INT_STS);
+	if (msix_sts_reg & HCLGE_VECTOR0_REG_MSIX_MASK) {
+		if (hclge_handle_hw_msix_error
+				(hdev, &hdev->default_reset_request))
+			dev_info(dev, "received msix interrupt 0x%x\n",
+				 msix_sts_reg);
+	}
+
+	hclge_handle_hw_ras_error(ae_dev);
+
+	hclge_handle_err_reset_request(hdev);
+}
+
+static void hclge_errhand_service_task(struct hclge_dev *hdev)
+{
+	if (!test_and_clear_bit(HCLGE_STATE_ERR_SERVICE_SCHED, &hdev->state))
+		return;
+
+	if (hnae3_dev_ras_imp_supported(hdev))
+		hclge_handle_err_recovery(hdev);
+	else
+		hclge_misc_err_recovery(hdev);
+}
+
+static void hclge_reset_service_task(struct hclge_dev *hdev)
+{
+	if (!test_and_clear_bit(HCLGE_STATE_RST_SERVICE_SCHED, &hdev->state))
+		return;
+
+	if (time_is_before_jiffies(hdev->last_rst_scheduled +
+				   HCLGE_RESET_SCHED_TIMEOUT))
+		dev_warn(&hdev->pdev->dev,
+			 "reset service task is scheduled after %ums on cpu%u!\n",
+			 jiffies_to_msecs(jiffies - hdev->last_rst_scheduled),
+			 smp_processor_id());
+
+	down(&hdev->reset_sem);
+	set_bit(HCLGE_STATE_RST_HANDLING, &hdev->state);
+
+	hclge_reset_subtask(hdev);
+
+	clear_bit(HCLGE_STATE_RST_HANDLING, &hdev->state);
+	up(&hdev->reset_sem);
+}
+
+static void hclge_update_vport_alive(struct hclge_dev *hdev)
+{
+#define HCLGE_ALIVE_SECONDS_NORMAL		8
+
+	unsigned long alive_time = HCLGE_ALIVE_SECONDS_NORMAL * HZ;
+	int i;
+
+	/* start from vport 1 for PF is always alive */
+	for (i = 1; i < hdev->num_alloc_vport; i++) {
+		struct hclge_vport *vport = &hdev->vport[i];
+
+		if (!test_bit(HCLGE_VPORT_STATE_INITED, &vport->state) ||
+		    !test_bit(HCLGE_VPORT_STATE_ALIVE, &vport->state))
+			continue;
+		if (time_after(jiffies, vport->last_active_jiffies +
+			       alive_time)) {
+			clear_bit(HCLGE_VPORT_STATE_ALIVE, &vport->state);
+			dev_warn(&hdev->pdev->dev,
+				 "VF %u heartbeat timeout\n",
+				 i - HCLGE_VF_VPORT_START_NUM);
+		}
+	}
+}
+
+static void hclge_periodic_service_task(struct hclge_dev *hdev)
+{
+	unsigned long delta = round_jiffies_relative(HZ);
+
+	if (test_bit(HCLGE_STATE_RST_FAIL, &hdev->state))
+		return;
+
+	/* Always handle the link updating to make sure link state is
+	 * updated when it is triggered by mbx.
+	 */
+	hclge_update_link_status(hdev);
+	hclge_sync_mac_table(hdev);
+	hclge_sync_promisc_mode(hdev);
+	hclge_sync_fd_table(hdev);
+
+	if (time_is_after_jiffies(hdev->last_serv_processed + HZ)) {
+		delta = jiffies - hdev->last_serv_processed;
+
+		if (delta < round_jiffies_relative(HZ)) {
+			delta = round_jiffies_relative(HZ) - delta;
+			goto out;
+		}
+	}
+
+	hdev->serv_processed_cnt++;
+	hclge_update_vport_alive(hdev);
+
+	if (test_bit(HCLGE_STATE_DOWN, &hdev->state)) {
+		hdev->last_serv_processed = jiffies;
+		goto out;
+	}
+
+	if (!(hdev->serv_processed_cnt % HCLGE_STATS_TIMER_INTERVAL))
+		hclge_update_stats_for_all(hdev);
+
+	hclge_update_port_info(hdev);
+	hclge_sync_vlan_filter(hdev);
+
+	if (!(hdev->serv_processed_cnt % HCLGE_ARFS_EXPIRE_INTERVAL))
+		hclge_rfs_filter_expire(hdev);
+
+	hdev->last_serv_processed = jiffies;
+
+out:
+	hclge_task_schedule(hdev, delta);
+}
+
+static void hclge_ptp_service_task(struct hclge_dev *hdev)
+{
+	unsigned long flags;
+
+	if (!test_bit(HCLGE_STATE_PTP_EN, &hdev->state) ||
+	    !test_bit(HCLGE_STATE_PTP_TX_HANDLING, &hdev->state) ||
+	    !time_is_before_jiffies(hdev->ptp->tx_start + HZ))
+		return;
+
+	/* to prevent concurrence with the irq handler */
+	spin_lock_irqsave(&hdev->ptp->lock, flags);
+
+	/* check HCLGE_STATE_PTP_TX_HANDLING here again, since the irq
+	 * handler may handle it just before spin_lock_irqsave().
+	 */
+	if (test_bit(HCLGE_STATE_PTP_TX_HANDLING, &hdev->state))
+		hclge_ptp_clean_tx_hwts(hdev);
+
+	spin_unlock_irqrestore(&hdev->ptp->lock, flags);
+}
+
+static void hclge_service_task(struct work_struct *work)
+{
+	struct hclge_dev *hdev =
+		container_of(work, struct hclge_dev, service_task.work);
+
+	hclge_errhand_service_task(hdev);
+	hclge_reset_service_task(hdev);
+	hclge_ptp_service_task(hdev);
+	hclge_mailbox_service_task(hdev);
+	hclge_periodic_service_task(hdev);
+
+	/* Handle error recovery, reset and mbx again in case periodical task
+	 * delays the handling by calling hclge_task_schedule() in
+	 * hclge_periodic_service_task().
+	 */
+	hclge_errhand_service_task(hdev);
+	hclge_reset_service_task(hdev);
+	hclge_mailbox_service_task(hdev);
+}
+
+struct hclge_vport *hclge_get_vport(struct hnae3_handle *handle)
+{
+	/* VF handle has no client */
+	if (!handle->client)
+		return container_of(handle, struct hclge_vport, nic);
+	else if (handle->client->type == HNAE3_CLIENT_ROCE)
+		return container_of(handle, struct hclge_vport, roce);
+	else
+		return container_of(handle, struct hclge_vport, nic);
+}
+
+static void hclge_get_vector_info(struct hclge_dev *hdev, u16 idx,
+				  struct hnae3_vector_info *vector_info)
+{
+#define HCLGE_PF_MAX_VECTOR_NUM_DEV_V2	64
+
+	vector_info->vector = pci_irq_vector(hdev->pdev, idx);
+
+	/* need an extend offset to config vector >= 64 */
+	if (idx - 1 < HCLGE_PF_MAX_VECTOR_NUM_DEV_V2)
+		vector_info->io_addr = hdev->hw.hw.io_base +
+				HCLGE_VECTOR_REG_BASE +
+				(idx - 1) * HCLGE_VECTOR_REG_OFFSET;
+	else
+		vector_info->io_addr = hdev->hw.hw.io_base +
+				HCLGE_VECTOR_EXT_REG_BASE +
+				(idx - 1) / HCLGE_PF_MAX_VECTOR_NUM_DEV_V2 *
+				HCLGE_VECTOR_REG_OFFSET_H +
+				(idx - 1) % HCLGE_PF_MAX_VECTOR_NUM_DEV_V2 *
+				HCLGE_VECTOR_REG_OFFSET;
+
+	hdev->vector_status[idx] = hdev->vport[0].vport_id;
+	hdev->vector_irq[idx] = vector_info->vector;
+}
+
+static int hclge_get_vector(struct hnae3_handle *handle, u16 vector_num,
+			    struct hnae3_vector_info *vector_info)
+{
+	struct hclge_vport *vport = hclge_get_vport(handle);
+	struct hnae3_vector_info *vector = vector_info;
+	struct hclge_dev *hdev = vport->back;
+	int alloc = 0;
+	u16 i = 0;
+	u16 j;
+
+	vector_num = min_t(u16, hdev->num_nic_msi - 1, vector_num);
+	vector_num = min(hdev->num_msi_left, vector_num);
+
+	for (j = 0; j < vector_num; j++) {
+		while (++i < hdev->num_nic_msi) {
+			if (hdev->vector_status[i] == HCLGE_INVALID_VPORT) {
+				hclge_get_vector_info(hdev, i, vector);
+				vector++;
+				alloc++;
+
+				break;
+			}
+		}
+	}
+	hdev->num_msi_left -= alloc;
+	hdev->num_msi_used += alloc;
+
+	return alloc;
+}
+
+static int hclge_get_vector_index(struct hclge_dev *hdev, int vector)
+{
+	int i;
+
+	for (i = 0; i < hdev->num_msi; i++)
+		if (vector == hdev->vector_irq[i])
+			return i;
+
+	return -EINVAL;
+}
+
+static int hclge_put_vector(struct hnae3_handle *handle, int vector)
+{
+	struct hclge_vport *vport = hclge_get_vport(handle);
+	struct hclge_dev *hdev = vport->back;
+	int vector_id;
+
+	vector_id = hclge_get_vector_index(hdev, vector);
+	if (vector_id < 0) {
+		dev_err(&hdev->pdev->dev,
+			"Get vector index fail. vector = %d\n", vector);
+		return vector_id;
+	}
+
+	hclge_free_vector(hdev, vector_id);
+
+	return 0;
+}
+
+static int hclge_get_rss(struct hnae3_handle *handle, u32 *indir,
+			 u8 *key, u8 *hfunc)
+{
+	struct hnae3_ae_dev *ae_dev = pci_get_drvdata(handle->pdev);
+	struct hclge_vport *vport = hclge_get_vport(handle);
+	struct hclge_comm_rss_cfg *rss_cfg = &vport->back->rss_cfg;
+
+	hclge_comm_get_rss_hash_info(rss_cfg, key, hfunc);
+
+	hclge_comm_get_rss_indir_tbl(rss_cfg, indir,
+				     ae_dev->dev_specs.rss_ind_tbl_size);
+
+	return 0;
+}
+
+static int hclge_set_rss(struct hnae3_handle *handle, const u32 *indir,
+			 const  u8 *key, const  u8 hfunc)
+{
+	struct hnae3_ae_dev *ae_dev = pci_get_drvdata(handle->pdev);
+	struct hclge_vport *vport = hclge_get_vport(handle);
+	struct hclge_dev *hdev = vport->back;
+	struct hclge_comm_rss_cfg *rss_cfg = &hdev->rss_cfg;
+	int ret, i;
+
+	ret = hclge_comm_set_rss_hash_key(rss_cfg, &hdev->hw.hw, key, hfunc);
+	if (ret) {
+		dev_err(&hdev->pdev->dev, "invalid hfunc type %u\n", hfunc);
+		return ret;
+	}
+
+	/* Update the shadow RSS table with user specified qids */
+	for (i = 0; i < ae_dev->dev_specs.rss_ind_tbl_size; i++)
+		rss_cfg->rss_indirection_tbl[i] = indir[i];
+
+	/* Update the hardware */
+	return hclge_comm_set_rss_indir_table(ae_dev, &hdev->hw.hw,
+					      rss_cfg->rss_indirection_tbl);
+}
+
+static int hclge_set_rss_tuple(struct hnae3_handle *handle,
+			       const struct ethtool_rxfh_fields *nfc)
+{
+	struct hclge_vport *vport = hclge_get_vport(handle);
+	struct hclge_dev *hdev = vport->back;
+	int ret;
+
+	ret = hclge_comm_set_rss_tuple(hdev->ae_dev, &hdev->hw.hw,
+				       &hdev->rss_cfg, nfc);
+	if (ret) {
+		dev_err(&hdev->pdev->dev,
+			"failed to set rss tuple, ret = %d.\n", ret);
+		return ret;
+	}
+
+	return 0;
+}
+
+static int hclge_get_rss_tuple(struct hnae3_handle *handle,
+			       struct ethtool_rxfh_fields *nfc)
+{
+	struct hclge_vport *vport = hclge_get_vport(handle);
+	u8 tuple_sets;
+	int ret;
+
+	nfc->data = 0;
+
+	ret = hclge_comm_get_rss_tuple(&vport->back->rss_cfg, nfc->flow_type,
+				       &tuple_sets);
+	if (ret || !tuple_sets)
+		return ret;
+
+	nfc->data = hclge_comm_convert_rss_tuple(tuple_sets);
+
+	return 0;
+}
+
+static int hclge_get_tc_size(struct hnae3_handle *handle)
+{
+	struct hclge_vport *vport = hclge_get_vport(handle);
+	struct hclge_dev *hdev = vport->back;
+
+	return hdev->pf_rss_size_max;
+}
+
+static int hclge_init_rss_tc_mode(struct hclge_dev *hdev)
+{
+	struct hnae3_ae_dev *ae_dev = hdev->ae_dev;
+	struct hclge_vport *vport = hdev->vport;
+	u16 tc_offset[HCLGE_MAX_TC_NUM] = {0};
+	u16 tc_valid[HCLGE_MAX_TC_NUM] = {0};
+	u16 tc_size[HCLGE_MAX_TC_NUM] = {0};
+	struct hnae3_tc_info *tc_info;
+	u16 roundup_size;
+	u16 rss_size;
+	int i;
+
+	tc_info = &vport->nic.kinfo.tc_info;
+	for (i = 0; i < HCLGE_MAX_TC_NUM; i++) {
+		rss_size = tc_info->tqp_count[i];
+		tc_valid[i] = 0;
+
+		if (!(hdev->hw_tc_map & BIT(i)))
+			continue;
+
+		/* tc_size set to hardware is the log2 of roundup power of two
+		 * of rss_size, the acutal queue size is limited by indirection
+		 * table.
+		 */
+		if (rss_size > ae_dev->dev_specs.rss_ind_tbl_size ||
+		    rss_size == 0) {
+			dev_err(&hdev->pdev->dev,
+				"Configure rss tc size failed, invalid TC_SIZE = %u\n",
+				rss_size);
+			return -EINVAL;
+		}
+
+		roundup_size = roundup_pow_of_two(rss_size);
+		roundup_size = ilog2(roundup_size);
+
+		tc_valid[i] = 1;
+		tc_size[i] = roundup_size;
+		tc_offset[i] = tc_info->tqp_offset[i];
+	}
+
+	return hclge_comm_set_rss_tc_mode(&hdev->hw.hw, tc_offset, tc_valid,
+					  tc_size);
+}
+
+int hclge_rss_init_hw(struct hclge_dev *hdev)
+{
+	u16 *rss_indir = hdev->rss_cfg.rss_indirection_tbl;
+	u8 *key = hdev->rss_cfg.rss_hash_key;
+	u8 hfunc = hdev->rss_cfg.rss_algo;
+	int ret;
+
+	ret = hclge_comm_set_rss_indir_table(hdev->ae_dev, &hdev->hw.hw,
+					     rss_indir);
+	if (ret)
+		return ret;
+
+	ret = hclge_comm_set_rss_algo_key(&hdev->hw.hw, hfunc, key);
+	if (ret)
+		return ret;
+
+	ret = hclge_comm_set_rss_input_tuple(&hdev->hw.hw, &hdev->rss_cfg);
+	if (ret)
+		return ret;
+
+	return hclge_init_rss_tc_mode(hdev);
+}
+
+int hclge_bind_ring_with_vector(struct hclge_vport *vport,
+				int vector_id, bool en,
+				struct hnae3_ring_chain_node *ring_chain)
+{
+	struct hclge_dev *hdev = vport->back;
+	struct hnae3_ring_chain_node *node;
+	struct hclge_desc desc;
+	struct hclge_ctrl_vector_chain_cmd *req =
+		(struct hclge_ctrl_vector_chain_cmd *)desc.data;
+	enum hclge_comm_cmd_status status;
+	enum hclge_opcode_type op;
+	u16 tqp_type_and_id;
+	int i;
+
+	op = en ? HCLGE_OPC_ADD_RING_TO_VECTOR : HCLGE_OPC_DEL_RING_TO_VECTOR;
+	hclge_cmd_setup_basic_desc(&desc, op, false);
+	req->int_vector_id_l = hnae3_get_field(vector_id,
+					       HCLGE_VECTOR_ID_L_M,
+					       HCLGE_VECTOR_ID_L_S);
+	req->int_vector_id_h = hnae3_get_field(vector_id,
+					       HCLGE_VECTOR_ID_H_M,
+					       HCLGE_VECTOR_ID_H_S);
+
+	i = 0;
+	for (node = ring_chain; node; node = node->next) {
+		tqp_type_and_id = le16_to_cpu(req->tqp_type_and_id[i]);
+		hnae3_set_field(tqp_type_and_id,  HCLGE_INT_TYPE_M,
+				HCLGE_INT_TYPE_S,
+				hnae3_get_bit(node->flag, HNAE3_RING_TYPE_B));
+		hnae3_set_field(tqp_type_and_id, HCLGE_TQP_ID_M,
+				HCLGE_TQP_ID_S, node->tqp_index);
+		hnae3_set_field(tqp_type_and_id, HCLGE_INT_GL_IDX_M,
+				HCLGE_INT_GL_IDX_S,
+				hnae3_get_field(node->int_gl_idx,
+						HNAE3_RING_GL_IDX_M,
+						HNAE3_RING_GL_IDX_S));
+		req->tqp_type_and_id[i] = cpu_to_le16(tqp_type_and_id);
+		if (++i >= HCLGE_VECTOR_ELEMENTS_PER_CMD) {
+			req->int_cause_num = HCLGE_VECTOR_ELEMENTS_PER_CMD;
+			req->vfid = vport->vport_id;
+
+			status = hclge_cmd_send(&hdev->hw, &desc, 1);
+			if (status) {
+				dev_err(&hdev->pdev->dev,
+					"Map TQP fail, status is %d.\n",
+					status);
+				return -EIO;
+			}
+			i = 0;
+
+			hclge_cmd_setup_basic_desc(&desc,
+						   op,
+						   false);
+			req->int_vector_id_l =
+				hnae3_get_field(vector_id,
+						HCLGE_VECTOR_ID_L_M,
+						HCLGE_VECTOR_ID_L_S);
+			req->int_vector_id_h =
+				hnae3_get_field(vector_id,
+						HCLGE_VECTOR_ID_H_M,
+						HCLGE_VECTOR_ID_H_S);
+		}
+	}
+
+	if (i > 0) {
+		req->int_cause_num = i;
+		req->vfid = vport->vport_id;
+		status = hclge_cmd_send(&hdev->hw, &desc, 1);
+		if (status) {
+			dev_err(&hdev->pdev->dev,
+				"Map TQP fail, status is %d.\n", status);
+			return -EIO;
+		}
+	}
+
+	return 0;
+}
+
+static int hclge_map_ring_to_vector(struct hnae3_handle *handle, int vector,
+				    struct hnae3_ring_chain_node *ring_chain)
+{
+	struct hclge_vport *vport = hclge_get_vport(handle);
+	struct hclge_dev *hdev = vport->back;
+	int vector_id;
+
+	vector_id = hclge_get_vector_index(hdev, vector);
+	if (vector_id < 0) {
+		dev_err(&hdev->pdev->dev,
+			"failed to get vector index. vector=%d\n", vector);
+		return vector_id;
+	}
+
+	return hclge_bind_ring_with_vector(vport, vector_id, true, ring_chain);
+}
+
+static int hclge_unmap_ring_frm_vector(struct hnae3_handle *handle, int vector,
+				       struct hnae3_ring_chain_node *ring_chain)
+{
+	struct hclge_vport *vport = hclge_get_vport(handle);
+	struct hclge_dev *hdev = vport->back;
+	int vector_id, ret;
+
+	if (test_bit(HCLGE_STATE_RST_HANDLING, &hdev->state))
+		return 0;
+
+	vector_id = hclge_get_vector_index(hdev, vector);
+	if (vector_id < 0) {
+		dev_err(&handle->pdev->dev,
+			"Get vector index fail. ret =%d\n", vector_id);
+		return vector_id;
+	}
+
+	ret = hclge_bind_ring_with_vector(vport, vector_id, false, ring_chain);
+	if (ret)
+		dev_err(&handle->pdev->dev,
+			"Unmap ring from vector fail. vectorid=%d, ret =%d\n",
+			vector_id, ret);
+
+	return ret;
+}
+
+static int hclge_cmd_set_promisc_mode(struct hclge_dev *hdev, u8 vf_id,
+				      bool en_uc, bool en_mc, bool en_bc)
+{
+	struct hclge_vport *vport = &hdev->vport[vf_id];
+	struct hnae3_handle *handle = &vport->nic;
+	struct hclge_promisc_cfg_cmd *req;
+	struct hclge_desc desc;
+	bool uc_tx_en = en_uc;
+	u8 promisc_cfg = 0;
+	int ret;
+
+	hclge_cmd_setup_basic_desc(&desc, HCLGE_OPC_CFG_PROMISC_MODE, false);
+
+	req = (struct hclge_promisc_cfg_cmd *)desc.data;
+	req->vf_id = vf_id;
+
+	if (test_bit(HNAE3_PFLAG_LIMIT_PROMISC, &handle->priv_flags))
+		uc_tx_en = false;
+
+	hnae3_set_bit(promisc_cfg, HCLGE_PROMISC_UC_RX_EN, en_uc ? 1 : 0);
+	hnae3_set_bit(promisc_cfg, HCLGE_PROMISC_MC_RX_EN, en_mc ? 1 : 0);
+	hnae3_set_bit(promisc_cfg, HCLGE_PROMISC_BC_RX_EN, en_bc ? 1 : 0);
+	hnae3_set_bit(promisc_cfg, HCLGE_PROMISC_UC_TX_EN, uc_tx_en ? 1 : 0);
+	hnae3_set_bit(promisc_cfg, HCLGE_PROMISC_MC_TX_EN, en_mc ? 1 : 0);
+	hnae3_set_bit(promisc_cfg, HCLGE_PROMISC_BC_TX_EN, en_bc ? 1 : 0);
+	req->extend_promisc = promisc_cfg;
+
+	/* to be compatible with DEVICE_VERSION_V1/2 */
+	promisc_cfg = 0;
+	hnae3_set_bit(promisc_cfg, HCLGE_PROMISC_EN_UC, en_uc ? 1 : 0);
+	hnae3_set_bit(promisc_cfg, HCLGE_PROMISC_EN_MC, en_mc ? 1 : 0);
+	hnae3_set_bit(promisc_cfg, HCLGE_PROMISC_EN_BC, en_bc ? 1 : 0);
+	hnae3_set_bit(promisc_cfg, HCLGE_PROMISC_TX_EN, 1);
+	hnae3_set_bit(promisc_cfg, HCLGE_PROMISC_RX_EN, 1);
+	req->promisc = promisc_cfg;
+
+	ret = hclge_cmd_send(&hdev->hw, &desc, 1);
+	if (ret)
+		dev_err(&hdev->pdev->dev,
+			"failed to set vport %u promisc mode, ret = %d.\n",
+			vf_id, ret);
+
+	return ret;
+}
+
+int hclge_set_vport_promisc_mode(struct hclge_vport *vport, bool en_uc_pmc,
+				 bool en_mc_pmc, bool en_bc_pmc)
+{
+	return hclge_cmd_set_promisc_mode(vport->back, vport->vport_id,
+					  en_uc_pmc, en_mc_pmc, en_bc_pmc);
+}
+
+static int hclge_set_promisc_mode(struct hnae3_handle *handle, bool en_uc_pmc,
+				  bool en_mc_pmc)
+{
+	struct hclge_vport *vport = hclge_get_vport(handle);
+	struct hclge_dev *hdev = vport->back;
+	bool en_bc_pmc = true;
+
+	/* For device whose version below V2, if broadcast promisc enabled,
+	 * vlan filter is always bypassed. So broadcast promisc should be
+	 * disabled until user enable promisc mode
+	 */
+	if (hdev->ae_dev->dev_version < HNAE3_DEVICE_VERSION_V2)
+		en_bc_pmc = handle->netdev_flags & HNAE3_BPE ? true : false;
+
+	return hclge_set_vport_promisc_mode(vport, en_uc_pmc, en_mc_pmc,
+					    en_bc_pmc);
+}
+
+static void hclge_request_update_promisc_mode(struct hnae3_handle *handle)
+{
+	struct hclge_vport *vport = hclge_get_vport(handle);
+
+	set_bit(HCLGE_VPORT_STATE_PROMISC_CHANGE, &vport->state);
+}
+
+static void hclge_sync_fd_state(struct hclge_dev *hdev)
+{
+	if (hlist_empty(&hdev->fd_rule_list))
+		hdev->fd_active_type = HCLGE_FD_RULE_NONE;
+}
+
+static void hclge_fd_inc_rule_cnt(struct hclge_dev *hdev, u16 location)
+{
+	if (!test_bit(location, hdev->fd_bmap)) {
+		set_bit(location, hdev->fd_bmap);
+		hdev->hclge_fd_rule_num++;
+	}
+}
+
+static void hclge_fd_dec_rule_cnt(struct hclge_dev *hdev, u16 location)
+{
+	if (test_bit(location, hdev->fd_bmap)) {
+		clear_bit(location, hdev->fd_bmap);
+		hdev->hclge_fd_rule_num--;
+	}
+}
+
+static void hclge_fd_free_node(struct hclge_dev *hdev,
+			       struct hclge_fd_rule *rule)
+{
+	hlist_del(&rule->rule_node);
+	kfree(rule);
+	hclge_sync_fd_state(hdev);
+}
+
+static void hclge_update_fd_rule_node(struct hclge_dev *hdev,
+				      struct hclge_fd_rule *old_rule,
+				      struct hclge_fd_rule *new_rule,
+				      enum HCLGE_FD_NODE_STATE state)
+{
+	switch (state) {
+	case HCLGE_FD_TO_ADD:
+	case HCLGE_FD_ACTIVE:
+		/* 1) if the new state is TO_ADD, just replace the old rule
+		 * with the same location, no matter its state, because the
+		 * new rule will be configured to the hardware.
+		 * 2) if the new state is ACTIVE, it means the new rule
+		 * has been configured to the hardware, so just replace
+		 * the old rule node with the same location.
+		 * 3) for it doesn't add a new node to the list, so it's
+		 * unnecessary to update the rule number and fd_bmap.
+		 */
+		new_rule->rule_node.next = old_rule->rule_node.next;
+		new_rule->rule_node.pprev = old_rule->rule_node.pprev;
+		memcpy(old_rule, new_rule, sizeof(*old_rule));
+		kfree(new_rule);
+		break;
+	case HCLGE_FD_DELETED:
+		hclge_fd_dec_rule_cnt(hdev, old_rule->location);
+		hclge_fd_free_node(hdev, old_rule);
+		break;
+	case HCLGE_FD_TO_DEL:
+		/* if new request is TO_DEL, and old rule is existent
+		 * 1) the state of old rule is TO_DEL, we need do nothing,
+		 * because we delete rule by location, other rule content
+		 * is unncessary.
+		 * 2) the state of old rule is ACTIVE, we need to change its
+		 * state to TO_DEL, so the rule will be deleted when periodic
+		 * task being scheduled.
+		 * 3) the state of old rule is TO_ADD, it means the rule hasn't
+		 * been added to hardware, so we just delete the rule node from
+		 * fd_rule_list directly.
+		 */
+		if (old_rule->state == HCLGE_FD_TO_ADD) {
+			hclge_fd_dec_rule_cnt(hdev, old_rule->location);
+			hclge_fd_free_node(hdev, old_rule);
+			return;
+		}
+		old_rule->state = HCLGE_FD_TO_DEL;
+		break;
+	}
+}
+
+static struct hclge_fd_rule *hclge_find_fd_rule(struct hlist_head *hlist,
+						u16 location,
+						struct hclge_fd_rule **parent)
+{
+	struct hclge_fd_rule *rule;
+	struct hlist_node *node;
+
+	hlist_for_each_entry_safe(rule, node, hlist, rule_node) {
+		if (rule->location == location)
+			return rule;
+		else if (rule->location > location)
+			return NULL;
+		/* record the parent node, use to keep the nodes in fd_rule_list
+		 * in ascend order.
+		 */
+		*parent = rule;
+	}
+
+	return NULL;
+}
+
+/* insert fd rule node in ascend order according to rule->location */
+static void hclge_fd_insert_rule_node(struct hlist_head *hlist,
+				      struct hclge_fd_rule *rule,
+				      struct hclge_fd_rule *parent)
+{
+	INIT_HLIST_NODE(&rule->rule_node);
+
+	if (parent)
+		hlist_add_behind(&rule->rule_node, &parent->rule_node);
+	else
+		hlist_add_head(&rule->rule_node, hlist);
+}
+
+static int hclge_fd_set_user_def_cmd(struct hclge_dev *hdev,
+				     struct hclge_fd_user_def_cfg *cfg)
+{
+	struct hclge_fd_user_def_cfg_cmd *req;
+	struct hclge_desc desc;
+	u16 data = 0;
+	int ret;
+
+	hclge_cmd_setup_basic_desc(&desc, HCLGE_OPC_FD_USER_DEF_OP, false);
+
+	req = (struct hclge_fd_user_def_cfg_cmd *)desc.data;
+
+	hnae3_set_bit(data, HCLGE_FD_USER_DEF_EN_B, cfg[0].ref_cnt > 0);
+	hnae3_set_field(data, HCLGE_FD_USER_DEF_OFT_M,
+			HCLGE_FD_USER_DEF_OFT_S, cfg[0].offset);
+	req->ol2_cfg = cpu_to_le16(data);
+
+	data = 0;
+	hnae3_set_bit(data, HCLGE_FD_USER_DEF_EN_B, cfg[1].ref_cnt > 0);
+	hnae3_set_field(data, HCLGE_FD_USER_DEF_OFT_M,
+			HCLGE_FD_USER_DEF_OFT_S, cfg[1].offset);
+	req->ol3_cfg = cpu_to_le16(data);
+
+	data = 0;
+	hnae3_set_bit(data, HCLGE_FD_USER_DEF_EN_B, cfg[2].ref_cnt > 0);
+	hnae3_set_field(data, HCLGE_FD_USER_DEF_OFT_M,
+			HCLGE_FD_USER_DEF_OFT_S, cfg[2].offset);
+	req->ol4_cfg = cpu_to_le16(data);
+
+	ret = hclge_cmd_send(&hdev->hw, &desc, 1);
+	if (ret)
+		dev_err(&hdev->pdev->dev,
+			"failed to set fd user def data, ret= %d\n", ret);
+	return ret;
+}
+
+static void hclge_sync_fd_user_def_cfg(struct hclge_dev *hdev, bool locked)
+{
+	int ret;
+
+	if (!test_and_clear_bit(HCLGE_STATE_FD_USER_DEF_CHANGED, &hdev->state))
+		return;
+
+	if (!locked)
+		spin_lock_bh(&hdev->fd_rule_lock);
+
+	ret = hclge_fd_set_user_def_cmd(hdev, hdev->fd_cfg.user_def_cfg);
+	if (ret)
+		set_bit(HCLGE_STATE_FD_USER_DEF_CHANGED, &hdev->state);
+
+	if (!locked)
+		spin_unlock_bh(&hdev->fd_rule_lock);
+}
+
+static int hclge_fd_check_user_def_refcnt(struct hclge_dev *hdev,
+					  struct hclge_fd_rule *rule)
+{
+	struct hlist_head *hlist = &hdev->fd_rule_list;
+	struct hclge_fd_rule *fd_rule, *parent = NULL;
+	struct hclge_fd_user_def_info *info, *old_info;
+	struct hclge_fd_user_def_cfg *cfg;
+
+	if (!rule || rule->rule_type != HCLGE_FD_EP_ACTIVE ||
+	    rule->ep.user_def.layer == HCLGE_FD_USER_DEF_NONE)
+		return 0;
+
+	/* for valid layer is start from 1, so need minus 1 to get the cfg */
+	cfg = &hdev->fd_cfg.user_def_cfg[rule->ep.user_def.layer - 1];
+	info = &rule->ep.user_def;
+
+	if (!cfg->ref_cnt || cfg->offset == info->offset)
+		return 0;
+
+	if (cfg->ref_cnt > 1)
+		goto error;
+
+	fd_rule = hclge_find_fd_rule(hlist, rule->location, &parent);
+	if (fd_rule) {
+		old_info = &fd_rule->ep.user_def;
+		if (info->layer == old_info->layer)
+			return 0;
+	}
+
+error:
+	dev_err(&hdev->pdev->dev,
+		"No available offset for layer%d fd rule, each layer only support one user def offset.\n",
+		info->layer + 1);
+	return -ENOSPC;
+}
+
+static void hclge_fd_inc_user_def_refcnt(struct hclge_dev *hdev,
+					 struct hclge_fd_rule *rule)
+{
+	struct hclge_fd_user_def_cfg *cfg;
+
+	if (!rule || rule->rule_type != HCLGE_FD_EP_ACTIVE ||
+	    rule->ep.user_def.layer == HCLGE_FD_USER_DEF_NONE)
+		return;
+
+	cfg = &hdev->fd_cfg.user_def_cfg[rule->ep.user_def.layer - 1];
+	if (!cfg->ref_cnt) {
+		cfg->offset = rule->ep.user_def.offset;
+		set_bit(HCLGE_STATE_FD_USER_DEF_CHANGED, &hdev->state);
+	}
+	cfg->ref_cnt++;
+}
+
+static void hclge_fd_dec_user_def_refcnt(struct hclge_dev *hdev,
+					 struct hclge_fd_rule *rule)
+{
+	struct hclge_fd_user_def_cfg *cfg;
+
+	if (!rule || rule->rule_type != HCLGE_FD_EP_ACTIVE ||
+	    rule->ep.user_def.layer == HCLGE_FD_USER_DEF_NONE)
+		return;
+
+	cfg = &hdev->fd_cfg.user_def_cfg[rule->ep.user_def.layer - 1];
+	if (!cfg->ref_cnt)
+		return;
+
+	cfg->ref_cnt--;
+	if (!cfg->ref_cnt) {
+		cfg->offset = 0;
+		set_bit(HCLGE_STATE_FD_USER_DEF_CHANGED, &hdev->state);
+	}
+}
+
+static void hclge_update_fd_list(struct hclge_dev *hdev,
+				 enum HCLGE_FD_NODE_STATE state, u16 location,
+				 struct hclge_fd_rule *new_rule)
+{
+	struct hlist_head *hlist = &hdev->fd_rule_list;
+	struct hclge_fd_rule *fd_rule, *parent = NULL;
+
+	fd_rule = hclge_find_fd_rule(hlist, location, &parent);
+	if (fd_rule) {
+		hclge_fd_dec_user_def_refcnt(hdev, fd_rule);
+		if (state == HCLGE_FD_ACTIVE)
+			hclge_fd_inc_user_def_refcnt(hdev, new_rule);
+		hclge_sync_fd_user_def_cfg(hdev, true);
+
+		hclge_update_fd_rule_node(hdev, fd_rule, new_rule, state);
+		return;
+	}
+
+	/* it's unlikely to fail here, because we have checked the rule
+	 * exist before.
+	 */
+	if (unlikely(state == HCLGE_FD_TO_DEL || state == HCLGE_FD_DELETED)) {
+		dev_warn(&hdev->pdev->dev,
+			 "failed to delete fd rule %u, it's inexistent\n",
+			 location);
+		return;
+	}
+
+	hclge_fd_inc_user_def_refcnt(hdev, new_rule);
+	hclge_sync_fd_user_def_cfg(hdev, true);
+
+	hclge_fd_insert_rule_node(hlist, new_rule, parent);
+	hclge_fd_inc_rule_cnt(hdev, new_rule->location);
+
+	if (state == HCLGE_FD_TO_ADD) {
+		set_bit(HCLGE_STATE_FD_TBL_CHANGED, &hdev->state);
+		hclge_task_schedule(hdev, 0);
+	}
+}
+
+static int hclge_get_fd_mode(struct hclge_dev *hdev, u8 *fd_mode)
+{
+	struct hclge_get_fd_mode_cmd *req;
+	struct hclge_desc desc;
+	int ret;
+
+	hclge_cmd_setup_basic_desc(&desc, HCLGE_OPC_FD_MODE_CTRL, true);
+
+	req = (struct hclge_get_fd_mode_cmd *)desc.data;
+
+	ret = hclge_cmd_send(&hdev->hw, &desc, 1);
+	if (ret) {
+		dev_err(&hdev->pdev->dev, "get fd mode fail, ret=%d\n", ret);
+		return ret;
+	}
+
+	*fd_mode = req->mode;
+
+	return ret;
+}
+
+static int hclge_get_fd_allocation(struct hclge_dev *hdev,
+				   u32 *stage1_entry_num,
+				   u32 *stage2_entry_num,
+				   u16 *stage1_counter_num,
+				   u16 *stage2_counter_num)
+{
+	struct hclge_get_fd_allocation_cmd *req;
+	struct hclge_desc desc;
+	int ret;
+
+	hclge_cmd_setup_basic_desc(&desc, HCLGE_OPC_FD_GET_ALLOCATION, true);
+
+	req = (struct hclge_get_fd_allocation_cmd *)desc.data;
+
+	ret = hclge_cmd_send(&hdev->hw, &desc, 1);
+	if (ret) {
+		dev_err(&hdev->pdev->dev, "query fd allocation fail, ret=%d\n",
+			ret);
+		return ret;
+	}
+
+	*stage1_entry_num = le32_to_cpu(req->stage1_entry_num);
+	*stage2_entry_num = le32_to_cpu(req->stage2_entry_num);
+	*stage1_counter_num = le16_to_cpu(req->stage1_counter_num);
+	*stage2_counter_num = le16_to_cpu(req->stage2_counter_num);
+
+	return ret;
+}
+
+static int hclge_set_fd_key_config(struct hclge_dev *hdev,
+				   enum HCLGE_FD_STAGE stage_num)
+{
+	struct hclge_set_fd_key_config_cmd *req;
+	struct hclge_fd_key_cfg *stage;
+	struct hclge_desc desc;
+	int ret;
+
+	hclge_cmd_setup_basic_desc(&desc, HCLGE_OPC_FD_KEY_CONFIG, false);
+
+	req = (struct hclge_set_fd_key_config_cmd *)desc.data;
+	stage = &hdev->fd_cfg.key_cfg[stage_num];
+	req->stage = stage_num;
+	req->key_select = stage->key_sel;
+	req->inner_sipv6_word_en = stage->inner_sipv6_word_en;
+	req->inner_dipv6_word_en = stage->inner_dipv6_word_en;
+	req->outer_sipv6_word_en = stage->outer_sipv6_word_en;
+	req->outer_dipv6_word_en = stage->outer_dipv6_word_en;
+	req->tuple_mask = cpu_to_le32(~stage->tuple_active);
+	req->meta_data_mask = cpu_to_le32(~stage->meta_data_active);
+
+	ret = hclge_cmd_send(&hdev->hw, &desc, 1);
+	if (ret)
+		dev_err(&hdev->pdev->dev, "set fd key fail, ret=%d\n", ret);
+
+	return ret;
+}
+
+static void hclge_fd_disable_user_def(struct hclge_dev *hdev)
+{
+	struct hclge_fd_user_def_cfg *cfg = hdev->fd_cfg.user_def_cfg;
+
+	spin_lock_bh(&hdev->fd_rule_lock);
+	memset(cfg, 0, sizeof(hdev->fd_cfg.user_def_cfg));
+	spin_unlock_bh(&hdev->fd_rule_lock);
+
+	hclge_fd_set_user_def_cmd(hdev, cfg);
+}
+
+static int hclge_init_fd_config(struct hclge_dev *hdev)
+{
+#define LOW_2_WORDS		0x03
+	struct hclge_fd_key_cfg *key_cfg;
+	int ret;
+
+	if (!hnae3_ae_dev_fd_supported(hdev->ae_dev))
+		return 0;
+
+	ret = hclge_get_fd_mode(hdev, &hdev->fd_cfg.fd_mode);
+	if (ret)
+		return ret;
+
+	switch (hdev->fd_cfg.fd_mode) {
+	case HCLGE_FD_MODE_DEPTH_2K_WIDTH_400B_STAGE_1:
+		hdev->fd_cfg.max_key_length = MAX_KEY_LENGTH;
+		break;
+	case HCLGE_FD_MODE_DEPTH_4K_WIDTH_200B_STAGE_1:
+		hdev->fd_cfg.max_key_length = MAX_KEY_LENGTH / 2;
+		break;
+	default:
+		dev_err(&hdev->pdev->dev,
+			"Unsupported flow director mode %u\n",
+			hdev->fd_cfg.fd_mode);
+		return -EOPNOTSUPP;
+	}
+
+	key_cfg = &hdev->fd_cfg.key_cfg[HCLGE_FD_STAGE_1];
+	key_cfg->key_sel = HCLGE_FD_KEY_BASE_ON_TUPLE;
+	key_cfg->inner_sipv6_word_en = LOW_2_WORDS;
+	key_cfg->inner_dipv6_word_en = LOW_2_WORDS;
+	key_cfg->outer_sipv6_word_en = 0;
+	key_cfg->outer_dipv6_word_en = 0;
+
+	key_cfg->tuple_active = BIT(INNER_VLAN_TAG_FST) | BIT(INNER_ETH_TYPE) |
+				BIT(INNER_IP_PROTO) | BIT(INNER_IP_TOS) |
+				BIT(INNER_SRC_IP) | BIT(INNER_DST_IP) |
+				BIT(INNER_SRC_PORT) | BIT(INNER_DST_PORT);
+
+	/* If use max 400bit key, we can support tuples for ether type */
+	if (hdev->fd_cfg.fd_mode == HCLGE_FD_MODE_DEPTH_2K_WIDTH_400B_STAGE_1) {
+		key_cfg->tuple_active |=
+				BIT(INNER_DST_MAC) | BIT(INNER_SRC_MAC);
+		if (hdev->ae_dev->dev_version >= HNAE3_DEVICE_VERSION_V3)
+			key_cfg->tuple_active |= HCLGE_FD_TUPLE_USER_DEF_TUPLES;
+	}
+
+	/* roce_type is used to filter roce frames
+	 * dst_vport is used to specify the rule
+	 */
+	key_cfg->meta_data_active = BIT(ROCE_TYPE) | BIT(DST_VPORT);
+
+	ret = hclge_get_fd_allocation(hdev,
+				      &hdev->fd_cfg.rule_num[HCLGE_FD_STAGE_1],
+				      &hdev->fd_cfg.rule_num[HCLGE_FD_STAGE_2],
+				      &hdev->fd_cfg.cnt_num[HCLGE_FD_STAGE_1],
+				      &hdev->fd_cfg.cnt_num[HCLGE_FD_STAGE_2]);
+	if (ret)
+		return ret;
+
+	return hclge_set_fd_key_config(hdev, HCLGE_FD_STAGE_1);
+}
+
+static int hclge_fd_tcam_config(struct hclge_dev *hdev, u8 stage, bool sel_x,
+				int loc, u8 *key, bool is_add)
+{
+	struct hclge_fd_tcam_config_1_cmd *req1;
+	struct hclge_fd_tcam_config_2_cmd *req2;
+	struct hclge_fd_tcam_config_3_cmd *req3;
+	struct hclge_desc desc[3];
+	int ret;
+
+	hclge_cmd_setup_basic_desc(&desc[0], HCLGE_OPC_FD_TCAM_OP, false);
+	desc[0].flag |= cpu_to_le16(HCLGE_COMM_CMD_FLAG_NEXT);
+	hclge_cmd_setup_basic_desc(&desc[1], HCLGE_OPC_FD_TCAM_OP, false);
+	desc[1].flag |= cpu_to_le16(HCLGE_COMM_CMD_FLAG_NEXT);
+	hclge_cmd_setup_basic_desc(&desc[2], HCLGE_OPC_FD_TCAM_OP, false);
+
+	req1 = (struct hclge_fd_tcam_config_1_cmd *)desc[0].data;
+	req2 = (struct hclge_fd_tcam_config_2_cmd *)desc[1].data;
+	req3 = (struct hclge_fd_tcam_config_3_cmd *)desc[2].data;
+
+	req1->stage = stage;
+	req1->xy_sel = sel_x ? 1 : 0;
+	hnae3_set_bit(req1->port_info, HCLGE_FD_EPORT_SW_EN_B, 0);
+	req1->index = cpu_to_le32(loc);
+	req1->entry_vld = sel_x ? is_add : 0;
+
+	if (key) {
+		memcpy(req1->tcam_data, &key[0], sizeof(req1->tcam_data));
+		memcpy(req2->tcam_data, &key[sizeof(req1->tcam_data)],
+		       sizeof(req2->tcam_data));
+		memcpy(req3->tcam_data, &key[sizeof(req1->tcam_data) +
+		       sizeof(req2->tcam_data)], sizeof(req3->tcam_data));
+	}
+
+	ret = hclge_cmd_send(&hdev->hw, desc, 3);
+	if (ret)
+		dev_err(&hdev->pdev->dev,
+			"config tcam key fail, ret=%d\n",
+			ret);
+
+	return ret;
+}
+
+static int hclge_fd_ad_config(struct hclge_dev *hdev, u8 stage, int loc,
+			      struct hclge_fd_ad_data *action)
+{
+	struct hnae3_ae_dev *ae_dev = pci_get_drvdata(hdev->pdev);
+	struct hclge_fd_ad_config_cmd *req;
+	struct hclge_desc desc;
+	u64 ad_data = 0;
+	int ret;
+
+	hclge_cmd_setup_basic_desc(&desc, HCLGE_OPC_FD_AD_OP, false);
+
+	req = (struct hclge_fd_ad_config_cmd *)desc.data;
+	req->index = cpu_to_le32(loc);
+	req->stage = stage;
+
+	hnae3_set_bit(ad_data, HCLGE_FD_AD_WR_RULE_ID_B,
+		      action->write_rule_id_to_bd);
+	hnae3_set_field(ad_data, HCLGE_FD_AD_RULE_ID_M, HCLGE_FD_AD_RULE_ID_S,
+			action->rule_id);
+	if (test_bit(HNAE3_DEV_SUPPORT_FD_FORWARD_TC_B, ae_dev->caps)) {
+		hnae3_set_bit(ad_data, HCLGE_FD_AD_TC_OVRD_B,
+			      action->override_tc);
+		hnae3_set_field(ad_data, HCLGE_FD_AD_TC_SIZE_M,
+				HCLGE_FD_AD_TC_SIZE_S, (u32)action->tc_size);
+	}
+	ad_data <<= 32;
+	hnae3_set_bit(ad_data, HCLGE_FD_AD_DROP_B, action->drop_packet);
+	hnae3_set_bit(ad_data, HCLGE_FD_AD_DIRECT_QID_B,
+		      action->forward_to_direct_queue);
+	hnae3_set_field(ad_data, HCLGE_FD_AD_QID_M, HCLGE_FD_AD_QID_S,
+			action->queue_id);
+	hnae3_set_bit(ad_data, HCLGE_FD_AD_USE_COUNTER_B, action->use_counter);
+	hnae3_set_field(ad_data, HCLGE_FD_AD_COUNTER_NUM_M,
+			HCLGE_FD_AD_COUNTER_NUM_S, action->counter_id);
+	hnae3_set_bit(ad_data, HCLGE_FD_AD_NXT_STEP_B, action->use_next_stage);
+	hnae3_set_field(ad_data, HCLGE_FD_AD_NXT_KEY_M, HCLGE_FD_AD_NXT_KEY_S,
+			action->counter_id);
+
+	req->ad_data = cpu_to_le64(ad_data);
+	ret = hclge_cmd_send(&hdev->hw, &desc, 1);
+	if (ret)
+		dev_err(&hdev->pdev->dev, "fd ad config fail, ret=%d\n", ret);
+
+	return ret;
+}
+
+static bool hclge_fd_convert_tuple(u32 tuple_bit, u8 *key_x, u8 *key_y,
+				   struct hclge_fd_rule *rule)
+{
+	int offset, moffset, ip_offset;
+	enum HCLGE_FD_KEY_OPT key_opt;
+	u16 tmp_x_s, tmp_y_s;
+	u32 tmp_x_l, tmp_y_l;
+	u8 *p = (u8 *)rule;
+	int i;
+
+	if (rule->unused_tuple & BIT(tuple_bit))
+		return true;
+
+	key_opt = tuple_key_info[tuple_bit].key_opt;
+	offset = tuple_key_info[tuple_bit].offset;
+	moffset = tuple_key_info[tuple_bit].moffset;
+
+	switch (key_opt) {
+	case KEY_OPT_U8:
+		calc_x(*key_x, p[offset], p[moffset]);
+		calc_y(*key_y, p[offset], p[moffset]);
+
+		return true;
+	case KEY_OPT_LE16:
+		calc_x(tmp_x_s, *(u16 *)(&p[offset]), *(u16 *)(&p[moffset]));
+		calc_y(tmp_y_s, *(u16 *)(&p[offset]), *(u16 *)(&p[moffset]));
+		*(__le16 *)key_x = cpu_to_le16(tmp_x_s);
+		*(__le16 *)key_y = cpu_to_le16(tmp_y_s);
+
+		return true;
+	case KEY_OPT_LE32:
+		calc_x(tmp_x_l, *(u32 *)(&p[offset]), *(u32 *)(&p[moffset]));
+		calc_y(tmp_y_l, *(u32 *)(&p[offset]), *(u32 *)(&p[moffset]));
+		*(__le32 *)key_x = cpu_to_le32(tmp_x_l);
+		*(__le32 *)key_y = cpu_to_le32(tmp_y_l);
+
+		return true;
+	case KEY_OPT_MAC:
+		for (i = 0; i < ETH_ALEN; i++) {
+			calc_x(key_x[ETH_ALEN - 1 - i], p[offset + i],
+			       p[moffset + i]);
+			calc_y(key_y[ETH_ALEN - 1 - i], p[offset + i],
+			       p[moffset + i]);
+		}
+
+		return true;
+	case KEY_OPT_IP:
+		ip_offset = IPV4_INDEX * sizeof(u32);
+		calc_x(tmp_x_l, *(u32 *)(&p[offset + ip_offset]),
+		       *(u32 *)(&p[moffset + ip_offset]));
+		calc_y(tmp_y_l, *(u32 *)(&p[offset + ip_offset]),
+		       *(u32 *)(&p[moffset + ip_offset]));
+		*(__le32 *)key_x = cpu_to_le32(tmp_x_l);
+		*(__le32 *)key_y = cpu_to_le32(tmp_y_l);
+
+		return true;
+	default:
+		return false;
+	}
+}
+
+static u32 hclge_get_port_number(enum HLCGE_PORT_TYPE port_type, u8 pf_id,
+				 u8 vf_id, u8 network_port_id)
+{
+	u32 port_number = 0;
+
+	if (port_type == HOST_PORT) {
+		hnae3_set_field(port_number, HCLGE_PF_ID_M, HCLGE_PF_ID_S,
+				pf_id);
+		hnae3_set_field(port_number, HCLGE_VF_ID_M, HCLGE_VF_ID_S,
+				vf_id);
+		hnae3_set_bit(port_number, HCLGE_PORT_TYPE_B, HOST_PORT);
+	} else {
+		hnae3_set_field(port_number, HCLGE_NETWORK_PORT_ID_M,
+				HCLGE_NETWORK_PORT_ID_S, network_port_id);
+		hnae3_set_bit(port_number, HCLGE_PORT_TYPE_B, NETWORK_PORT);
+	}
+
+	return port_number;
+}
+
+static void hclge_fd_convert_meta_data(struct hclge_fd_key_cfg *key_cfg,
+				       __le32 *key_x, __le32 *key_y,
+				       struct hclge_fd_rule *rule)
+{
+	u32 tuple_bit, meta_data = 0, tmp_x, tmp_y, port_number;
+	u8 cur_pos = 0, tuple_size, shift_bits;
+	unsigned int i;
+
+	for (i = 0; i < MAX_META_DATA; i++) {
+		tuple_size = meta_data_key_info[i].key_length;
+		tuple_bit = key_cfg->meta_data_active & BIT(i);
+
+		switch (tuple_bit) {
+		case BIT(ROCE_TYPE):
+			hnae3_set_bit(meta_data, cur_pos, NIC_PACKET);
+			cur_pos += tuple_size;
+			break;
+		case BIT(DST_VPORT):
+			port_number = hclge_get_port_number(HOST_PORT, 0,
+							    rule->vf_id, 0);
+			hnae3_set_field(meta_data,
+					GENMASK(cur_pos + tuple_size, cur_pos),
+					cur_pos, port_number);
+			cur_pos += tuple_size;
+			break;
+		default:
+			break;
+		}
+	}
+
+	calc_x(tmp_x, meta_data, 0xFFFFFFFF);
+	calc_y(tmp_y, meta_data, 0xFFFFFFFF);
+	shift_bits = sizeof(meta_data) * 8 - cur_pos;
+
+	*key_x = cpu_to_le32(tmp_x << shift_bits);
+	*key_y = cpu_to_le32(tmp_y << shift_bits);
+}
+
+/* A complete key is combined with meta data key and tuple key.
+ * Meta data key is stored at the MSB region, and tuple key is stored at
+ * the LSB region, unused bits will be filled 0.
+ */
+static int hclge_config_key(struct hclge_dev *hdev, u8 stage,
+			    struct hclge_fd_rule *rule)
+{
+	struct hclge_fd_key_cfg *key_cfg = &hdev->fd_cfg.key_cfg[stage];
+	u8 key_x[MAX_KEY_BYTES], key_y[MAX_KEY_BYTES];
+	u8 *cur_key_x, *cur_key_y;
+	u8 meta_data_region;
+	u8 tuple_size;
+	int ret;
+	u32 i;
+
+	memset(key_x, 0, sizeof(key_x));
+	memset(key_y, 0, sizeof(key_y));
+	cur_key_x = key_x;
+	cur_key_y = key_y;
+
+	for (i = 0; i < MAX_TUPLE; i++) {
+		bool tuple_valid;
+
+		tuple_size = tuple_key_info[i].key_length / 8;
+		if (!(key_cfg->tuple_active & BIT(i)))
+			continue;
+
+		tuple_valid = hclge_fd_convert_tuple(i, cur_key_x,
+						     cur_key_y, rule);
+		if (tuple_valid) {
+			cur_key_x += tuple_size;
+			cur_key_y += tuple_size;
+		}
+	}
+
+	meta_data_region = hdev->fd_cfg.max_key_length / 8 -
+			MAX_META_DATA_LENGTH / 8;
+
+	hclge_fd_convert_meta_data(key_cfg,
+				   (__le32 *)(key_x + meta_data_region),
+				   (__le32 *)(key_y + meta_data_region),
+				   rule);
+
+	ret = hclge_fd_tcam_config(hdev, stage, false, rule->location, key_y,
+				   true);
+	if (ret) {
+		dev_err(&hdev->pdev->dev,
+			"fd key_y config fail, loc=%u, ret=%d\n",
+			rule->queue_id, ret);
+		return ret;
+	}
+
+	ret = hclge_fd_tcam_config(hdev, stage, true, rule->location, key_x,
+				   true);
+	if (ret)
+		dev_err(&hdev->pdev->dev,
+			"fd key_x config fail, loc=%u, ret=%d\n",
+			rule->queue_id, ret);
+	return ret;
+}
+
+static int hclge_config_action(struct hclge_dev *hdev, u8 stage,
+			       struct hclge_fd_rule *rule)
+{
+	struct hclge_vport *vport = hdev->vport;
+	struct hnae3_knic_private_info *kinfo = &vport->nic.kinfo;
+	struct hclge_fd_ad_data ad_data;
+
+	memset(&ad_data, 0, sizeof(struct hclge_fd_ad_data));
+	ad_data.ad_id = rule->location;
+
+	if (rule->action == HCLGE_FD_ACTION_DROP_PACKET) {
+		ad_data.drop_packet = true;
+	} else if (rule->action == HCLGE_FD_ACTION_SELECT_TC) {
+		ad_data.override_tc = true;
+		ad_data.queue_id =
+			kinfo->tc_info.tqp_offset[rule->cls_flower.tc];
+		ad_data.tc_size =
+			ilog2(kinfo->tc_info.tqp_count[rule->cls_flower.tc]);
+	} else {
+		ad_data.forward_to_direct_queue = true;
+		ad_data.queue_id = rule->queue_id;
+	}
+
+	if (hdev->fd_cfg.cnt_num[HCLGE_FD_STAGE_1]) {
+		ad_data.use_counter = true;
+		ad_data.counter_id = rule->vf_id %
+				     hdev->fd_cfg.cnt_num[HCLGE_FD_STAGE_1];
+	} else {
+		ad_data.use_counter = false;
+		ad_data.counter_id = 0;
+	}
+
+	ad_data.use_next_stage = false;
+	ad_data.next_input_key = 0;
+
+	ad_data.write_rule_id_to_bd = true;
+	ad_data.rule_id = rule->location;
+
+	return hclge_fd_ad_config(hdev, stage, ad_data.ad_id, &ad_data);
+}
+
+static int hclge_fd_check_tcpip4_tuple(struct ethtool_tcpip4_spec *spec,
+				       u32 *unused_tuple)
+{
+	if (!spec || !unused_tuple)
+		return -EINVAL;
+
+	*unused_tuple |= BIT(INNER_SRC_MAC) | BIT(INNER_DST_MAC);
+
+	if (!spec->ip4src)
+		*unused_tuple |= BIT(INNER_SRC_IP);
+
+	if (!spec->ip4dst)
+		*unused_tuple |= BIT(INNER_DST_IP);
+
+	if (!spec->psrc)
+		*unused_tuple |= BIT(INNER_SRC_PORT);
+
+	if (!spec->pdst)
+		*unused_tuple |= BIT(INNER_DST_PORT);
+
+	if (!spec->tos)
+		*unused_tuple |= BIT(INNER_IP_TOS);
+
+	return 0;
+}
+
+static int hclge_fd_check_ip4_tuple(struct ethtool_usrip4_spec *spec,
+				    u32 *unused_tuple)
+{
+	if (!spec || !unused_tuple)
+		return -EINVAL;
+
+	*unused_tuple |= BIT(INNER_SRC_MAC) | BIT(INNER_DST_MAC) |
+		BIT(INNER_SRC_PORT) | BIT(INNER_DST_PORT);
+
+	if (!spec->ip4src)
+		*unused_tuple |= BIT(INNER_SRC_IP);
+
+	if (!spec->ip4dst)
+		*unused_tuple |= BIT(INNER_DST_IP);
+
+	if (!spec->tos)
+		*unused_tuple |= BIT(INNER_IP_TOS);
+
+	if (!spec->proto)
+		*unused_tuple |= BIT(INNER_IP_PROTO);
+
+	if (spec->l4_4_bytes)
+		return -EOPNOTSUPP;
+
+	if (spec->ip_ver != ETH_RX_NFC_IP4)
+		return -EOPNOTSUPP;
+
+	return 0;
+}
+
+static int hclge_fd_check_tcpip6_tuple(struct ethtool_tcpip6_spec *spec,
+				       u32 *unused_tuple)
+{
+	if (!spec || !unused_tuple)
+		return -EINVAL;
+
+	*unused_tuple |= BIT(INNER_SRC_MAC) | BIT(INNER_DST_MAC);
+
+	/* check whether src/dst ip address used */
+	if (ipv6_addr_any((struct in6_addr *)spec->ip6src))
+		*unused_tuple |= BIT(INNER_SRC_IP);
+
+	if (ipv6_addr_any((struct in6_addr *)spec->ip6dst))
+		*unused_tuple |= BIT(INNER_DST_IP);
+
+	if (!spec->psrc)
+		*unused_tuple |= BIT(INNER_SRC_PORT);
+
+	if (!spec->pdst)
+		*unused_tuple |= BIT(INNER_DST_PORT);
+
+	if (!spec->tclass)
+		*unused_tuple |= BIT(INNER_IP_TOS);
+
+	return 0;
+}
+
+static int hclge_fd_check_ip6_tuple(struct ethtool_usrip6_spec *spec,
+				    u32 *unused_tuple)
+{
+	if (!spec || !unused_tuple)
+		return -EINVAL;
+
+	*unused_tuple |= BIT(INNER_SRC_MAC) | BIT(INNER_DST_MAC) |
+			BIT(INNER_SRC_PORT) | BIT(INNER_DST_PORT);
+
+	/* check whether src/dst ip address used */
+	if (ipv6_addr_any((struct in6_addr *)spec->ip6src))
+		*unused_tuple |= BIT(INNER_SRC_IP);
+
+	if (ipv6_addr_any((struct in6_addr *)spec->ip6dst))
+		*unused_tuple |= BIT(INNER_DST_IP);
+
+	if (!spec->l4_proto)
+		*unused_tuple |= BIT(INNER_IP_PROTO);
+
+	if (!spec->tclass)
+		*unused_tuple |= BIT(INNER_IP_TOS);
+
+	if (spec->l4_4_bytes)
+		return -EOPNOTSUPP;
+
+	return 0;
+}
+
+static int hclge_fd_check_ether_tuple(struct ethhdr *spec, u32 *unused_tuple)
+{
+	if (!spec || !unused_tuple)
+		return -EINVAL;
+
+	*unused_tuple |= BIT(INNER_SRC_IP) | BIT(INNER_DST_IP) |
+		BIT(INNER_SRC_PORT) | BIT(INNER_DST_PORT) |
+		BIT(INNER_IP_TOS) | BIT(INNER_IP_PROTO);
+
+	if (is_zero_ether_addr(spec->h_source))
+		*unused_tuple |= BIT(INNER_SRC_MAC);
+
+	if (is_zero_ether_addr(spec->h_dest))
+		*unused_tuple |= BIT(INNER_DST_MAC);
+
+	if (!spec->h_proto)
+		*unused_tuple |= BIT(INNER_ETH_TYPE);
+
+	return 0;
+}
+
+static int hclge_fd_check_ext_tuple(struct hclge_dev *hdev,
+				    struct ethtool_rx_flow_spec *fs,
+				    u32 *unused_tuple)
+{
+	if (fs->flow_type & FLOW_EXT) {
+		if (fs->h_ext.vlan_etype) {
+			dev_err(&hdev->pdev->dev, "vlan-etype is not supported!\n");
+			return -EOPNOTSUPP;
+		}
+
+		if (!fs->h_ext.vlan_tci)
+			*unused_tuple |= BIT(INNER_VLAN_TAG_FST);
+
+		if (fs->m_ext.vlan_tci &&
+		    be16_to_cpu(fs->h_ext.vlan_tci) >= VLAN_N_VID) {
+			dev_err(&hdev->pdev->dev,
+				"failed to config vlan_tci, invalid vlan_tci: %u, max is %d.\n",
+				ntohs(fs->h_ext.vlan_tci), VLAN_N_VID - 1);
+			return -EINVAL;
+		}
+	} else {
+		*unused_tuple |= BIT(INNER_VLAN_TAG_FST);
+	}
+
+	if (fs->flow_type & FLOW_MAC_EXT) {
+		if (hdev->fd_cfg.fd_mode !=
+		    HCLGE_FD_MODE_DEPTH_2K_WIDTH_400B_STAGE_1) {
+			dev_err(&hdev->pdev->dev,
+				"FLOW_MAC_EXT is not supported in current fd mode!\n");
+			return -EOPNOTSUPP;
+		}
+
+		if (is_zero_ether_addr(fs->h_ext.h_dest))
+			*unused_tuple |= BIT(INNER_DST_MAC);
+		else
+			*unused_tuple &= ~BIT(INNER_DST_MAC);
+	}
+
+	return 0;
+}
+
+static int hclge_fd_get_user_def_layer(u32 flow_type, u32 *unused_tuple,
+				       struct hclge_fd_user_def_info *info)
+{
+	switch (flow_type) {
+	case ETHER_FLOW:
+		info->layer = HCLGE_FD_USER_DEF_L2;
+		*unused_tuple &= ~BIT(INNER_L2_RSV);
+		break;
+	case IP_USER_FLOW:
+	case IPV6_USER_FLOW:
+		info->layer = HCLGE_FD_USER_DEF_L3;
+		*unused_tuple &= ~BIT(INNER_L3_RSV);
+		break;
+	case TCP_V4_FLOW:
+	case UDP_V4_FLOW:
+	case TCP_V6_FLOW:
+	case UDP_V6_FLOW:
+		info->layer = HCLGE_FD_USER_DEF_L4;
+		*unused_tuple &= ~BIT(INNER_L4_RSV);
+		break;
+	default:
+		return -EOPNOTSUPP;
+	}
+
+	return 0;
+}
+
+static bool hclge_fd_is_user_def_all_masked(struct ethtool_rx_flow_spec *fs)
+{
+	return be32_to_cpu(fs->m_ext.data[1] | fs->m_ext.data[0]) == 0;
+}
+
+static int hclge_fd_parse_user_def_field(struct hclge_dev *hdev,
+					 struct ethtool_rx_flow_spec *fs,
+					 u32 *unused_tuple,
+					 struct hclge_fd_user_def_info *info)
+{
+	u32 tuple_active = hdev->fd_cfg.key_cfg[HCLGE_FD_STAGE_1].tuple_active;
+	u32 flow_type = fs->flow_type & ~(FLOW_EXT | FLOW_MAC_EXT);
+	u16 data, offset, data_mask, offset_mask;
+	int ret;
+
+	info->layer = HCLGE_FD_USER_DEF_NONE;
+	*unused_tuple |= HCLGE_FD_TUPLE_USER_DEF_TUPLES;
+
+	if (!(fs->flow_type & FLOW_EXT) || hclge_fd_is_user_def_all_masked(fs))
+		return 0;
+
+	/* user-def data from ethtool is 64 bit value, the bit0~15 is used
+	 * for data, and bit32~47 is used for offset.
+	 */
+	data = be32_to_cpu(fs->h_ext.data[1]) & HCLGE_FD_USER_DEF_DATA;
+	data_mask = be32_to_cpu(fs->m_ext.data[1]) & HCLGE_FD_USER_DEF_DATA;
+	offset = be32_to_cpu(fs->h_ext.data[0]) & HCLGE_FD_USER_DEF_OFFSET;
+	offset_mask = be32_to_cpu(fs->m_ext.data[0]) & HCLGE_FD_USER_DEF_OFFSET;
+
+	if (!(tuple_active & HCLGE_FD_TUPLE_USER_DEF_TUPLES)) {
+		dev_err(&hdev->pdev->dev, "user-def bytes are not supported\n");
+		return -EOPNOTSUPP;
+	}
+
+	if (offset > HCLGE_FD_MAX_USER_DEF_OFFSET) {
+		dev_err(&hdev->pdev->dev,
+			"user-def offset[%u] should be no more than %u\n",
+			offset, HCLGE_FD_MAX_USER_DEF_OFFSET);
+		return -EINVAL;
+	}
+
+	if (offset_mask != HCLGE_FD_USER_DEF_OFFSET_UNMASK) {
+		dev_err(&hdev->pdev->dev, "user-def offset can't be masked\n");
+		return -EINVAL;
+	}
+
+	ret = hclge_fd_get_user_def_layer(flow_type, unused_tuple, info);
+	if (ret) {
+		dev_err(&hdev->pdev->dev,
+			"unsupported flow type for user-def bytes, ret = %d\n",
+			ret);
+		return ret;
+	}
+
+	info->data = data;
+	info->data_mask = data_mask;
+	info->offset = offset;
+
+	return 0;
+}
+
+static int hclge_fd_check_spec(struct hclge_dev *hdev,
+			       struct ethtool_rx_flow_spec *fs,
+			       u32 *unused_tuple,
+			       struct hclge_fd_user_def_info *info)
+{
+	u32 flow_type;
+	int ret;
+
+	if (fs->location >= hdev->fd_cfg.rule_num[HCLGE_FD_STAGE_1]) {
+		dev_err(&hdev->pdev->dev,
+			"failed to config fd rules, invalid rule location: %u, max is %u\n.",
+			fs->location,
+			hdev->fd_cfg.rule_num[HCLGE_FD_STAGE_1] - 1);
+		return -EINVAL;
+	}
+
+	ret = hclge_fd_parse_user_def_field(hdev, fs, unused_tuple, info);
+	if (ret)
+		return ret;
+
+	flow_type = fs->flow_type & ~(FLOW_EXT | FLOW_MAC_EXT);
+	switch (flow_type) {
+	case SCTP_V4_FLOW:
+	case TCP_V4_FLOW:
+	case UDP_V4_FLOW:
+		ret = hclge_fd_check_tcpip4_tuple(&fs->h_u.tcp_ip4_spec,
+						  unused_tuple);
+		break;
+	case IP_USER_FLOW:
+		ret = hclge_fd_check_ip4_tuple(&fs->h_u.usr_ip4_spec,
+					       unused_tuple);
+		break;
+	case SCTP_V6_FLOW:
+	case TCP_V6_FLOW:
+	case UDP_V6_FLOW:
+		ret = hclge_fd_check_tcpip6_tuple(&fs->h_u.tcp_ip6_spec,
+						  unused_tuple);
+		break;
+	case IPV6_USER_FLOW:
+		ret = hclge_fd_check_ip6_tuple(&fs->h_u.usr_ip6_spec,
+					       unused_tuple);
+		break;
+	case ETHER_FLOW:
+		if (hdev->fd_cfg.fd_mode !=
+			HCLGE_FD_MODE_DEPTH_2K_WIDTH_400B_STAGE_1) {
+			dev_err(&hdev->pdev->dev,
+				"ETHER_FLOW is not supported in current fd mode!\n");
+			return -EOPNOTSUPP;
+		}
+
+		ret = hclge_fd_check_ether_tuple(&fs->h_u.ether_spec,
+						 unused_tuple);
+		break;
+	default:
+		dev_err(&hdev->pdev->dev,
+			"unsupported protocol type, protocol type = %#x\n",
+			flow_type);
+		return -EOPNOTSUPP;
+	}
+
+	if (ret) {
+		dev_err(&hdev->pdev->dev,
+			"failed to check flow union tuple, ret = %d\n",
+			ret);
+		return ret;
+	}
+
+	return hclge_fd_check_ext_tuple(hdev, fs, unused_tuple);
+}
+
+static void hclge_fd_get_tcpip4_tuple(struct ethtool_rx_flow_spec *fs,
+				      struct hclge_fd_rule *rule, u8 ip_proto)
+{
+	rule->tuples.src_ip[IPV4_INDEX] =
+			be32_to_cpu(fs->h_u.tcp_ip4_spec.ip4src);
+	rule->tuples_mask.src_ip[IPV4_INDEX] =
+			be32_to_cpu(fs->m_u.tcp_ip4_spec.ip4src);
+
+	rule->tuples.dst_ip[IPV4_INDEX] =
+			be32_to_cpu(fs->h_u.tcp_ip4_spec.ip4dst);
+	rule->tuples_mask.dst_ip[IPV4_INDEX] =
+			be32_to_cpu(fs->m_u.tcp_ip4_spec.ip4dst);
+
+	rule->tuples.src_port = be16_to_cpu(fs->h_u.tcp_ip4_spec.psrc);
+	rule->tuples_mask.src_port = be16_to_cpu(fs->m_u.tcp_ip4_spec.psrc);
+
+	rule->tuples.dst_port = be16_to_cpu(fs->h_u.tcp_ip4_spec.pdst);
+	rule->tuples_mask.dst_port = be16_to_cpu(fs->m_u.tcp_ip4_spec.pdst);
+
+	rule->tuples.ip_tos = fs->h_u.tcp_ip4_spec.tos;
+	rule->tuples_mask.ip_tos = fs->m_u.tcp_ip4_spec.tos;
+
+	rule->tuples.ether_proto = ETH_P_IP;
+	rule->tuples_mask.ether_proto = 0xFFFF;
+
+	rule->tuples.ip_proto = ip_proto;
+	rule->tuples_mask.ip_proto = 0xFF;
+}
+
+static void hclge_fd_get_ip4_tuple(struct ethtool_rx_flow_spec *fs,
+				   struct hclge_fd_rule *rule)
+{
+	rule->tuples.src_ip[IPV4_INDEX] =
+			be32_to_cpu(fs->h_u.usr_ip4_spec.ip4src);
+	rule->tuples_mask.src_ip[IPV4_INDEX] =
+			be32_to_cpu(fs->m_u.usr_ip4_spec.ip4src);
+
+	rule->tuples.dst_ip[IPV4_INDEX] =
+			be32_to_cpu(fs->h_u.usr_ip4_spec.ip4dst);
+	rule->tuples_mask.dst_ip[IPV4_INDEX] =
+			be32_to_cpu(fs->m_u.usr_ip4_spec.ip4dst);
+
+	rule->tuples.ip_tos = fs->h_u.usr_ip4_spec.tos;
+	rule->tuples_mask.ip_tos = fs->m_u.usr_ip4_spec.tos;
+
+	rule->tuples.ip_proto = fs->h_u.usr_ip4_spec.proto;
+	rule->tuples_mask.ip_proto = fs->m_u.usr_ip4_spec.proto;
+
+	rule->tuples.ether_proto = ETH_P_IP;
+	rule->tuples_mask.ether_proto = 0xFFFF;
+}
+
+static void hclge_fd_get_tcpip6_tuple(struct ethtool_rx_flow_spec *fs,
+				      struct hclge_fd_rule *rule, u8 ip_proto)
+{
+	ipv6_addr_be32_to_cpu(rule->tuples.src_ip,
+			      fs->h_u.tcp_ip6_spec.ip6src);
+	ipv6_addr_be32_to_cpu(rule->tuples_mask.src_ip,
+			      fs->m_u.tcp_ip6_spec.ip6src);
+
+	ipv6_addr_be32_to_cpu(rule->tuples.dst_ip,
+			      fs->h_u.tcp_ip6_spec.ip6dst);
+	ipv6_addr_be32_to_cpu(rule->tuples_mask.dst_ip,
+			      fs->m_u.tcp_ip6_spec.ip6dst);
+
+	rule->tuples.src_port = be16_to_cpu(fs->h_u.tcp_ip6_spec.psrc);
+	rule->tuples_mask.src_port = be16_to_cpu(fs->m_u.tcp_ip6_spec.psrc);
+
+	rule->tuples.dst_port = be16_to_cpu(fs->h_u.tcp_ip6_spec.pdst);
+	rule->tuples_mask.dst_port = be16_to_cpu(fs->m_u.tcp_ip6_spec.pdst);
+
+	rule->tuples.ether_proto = ETH_P_IPV6;
+	rule->tuples_mask.ether_proto = 0xFFFF;
+
+	rule->tuples.ip_tos = fs->h_u.tcp_ip6_spec.tclass;
+	rule->tuples_mask.ip_tos = fs->m_u.tcp_ip6_spec.tclass;
+
+	rule->tuples.ip_proto = ip_proto;
+	rule->tuples_mask.ip_proto = 0xFF;
+}
+
+static void hclge_fd_get_ip6_tuple(struct ethtool_rx_flow_spec *fs,
+				   struct hclge_fd_rule *rule)
+{
+	ipv6_addr_be32_to_cpu(rule->tuples.src_ip,
+			      fs->h_u.usr_ip6_spec.ip6src);
+	ipv6_addr_be32_to_cpu(rule->tuples_mask.src_ip,
+			      fs->m_u.usr_ip6_spec.ip6src);
+
+	ipv6_addr_be32_to_cpu(rule->tuples.dst_ip,
+			      fs->h_u.usr_ip6_spec.ip6dst);
+	ipv6_addr_be32_to_cpu(rule->tuples_mask.dst_ip,
+			      fs->m_u.usr_ip6_spec.ip6dst);
+
+	rule->tuples.ip_proto = fs->h_u.usr_ip6_spec.l4_proto;
+	rule->tuples_mask.ip_proto = fs->m_u.usr_ip6_spec.l4_proto;
+
+	rule->tuples.ip_tos = fs->h_u.tcp_ip6_spec.tclass;
+	rule->tuples_mask.ip_tos = fs->m_u.tcp_ip6_spec.tclass;
+
+	rule->tuples.ether_proto = ETH_P_IPV6;
+	rule->tuples_mask.ether_proto = 0xFFFF;
+}
+
+static void hclge_fd_get_ether_tuple(struct ethtool_rx_flow_spec *fs,
+				     struct hclge_fd_rule *rule)
+{
+	ether_addr_copy(rule->tuples.src_mac, fs->h_u.ether_spec.h_source);
+	ether_addr_copy(rule->tuples_mask.src_mac, fs->m_u.ether_spec.h_source);
+
+	ether_addr_copy(rule->tuples.dst_mac, fs->h_u.ether_spec.h_dest);
+	ether_addr_copy(rule->tuples_mask.dst_mac, fs->m_u.ether_spec.h_dest);
+
+	rule->tuples.ether_proto = be16_to_cpu(fs->h_u.ether_spec.h_proto);
+	rule->tuples_mask.ether_proto = be16_to_cpu(fs->m_u.ether_spec.h_proto);
+}
+
+static void hclge_fd_get_user_def_tuple(struct hclge_fd_user_def_info *info,
+					struct hclge_fd_rule *rule)
+{
+	switch (info->layer) {
+	case HCLGE_FD_USER_DEF_L2:
+		rule->tuples.l2_user_def = info->data;
+		rule->tuples_mask.l2_user_def = info->data_mask;
+		break;
+	case HCLGE_FD_USER_DEF_L3:
+		rule->tuples.l3_user_def = info->data;
+		rule->tuples_mask.l3_user_def = info->data_mask;
+		break;
+	case HCLGE_FD_USER_DEF_L4:
+		rule->tuples.l4_user_def = (u32)info->data << 16;
+		rule->tuples_mask.l4_user_def = (u32)info->data_mask << 16;
+		break;
+	default:
+		break;
+	}
+
+	rule->ep.user_def = *info;
+}
+
+static int hclge_fd_get_tuple(struct ethtool_rx_flow_spec *fs,
+			      struct hclge_fd_rule *rule,
+			      struct hclge_fd_user_def_info *info)
+{
+	u32 flow_type = fs->flow_type & ~(FLOW_EXT | FLOW_MAC_EXT);
+
+	switch (flow_type) {
+	case SCTP_V4_FLOW:
+		hclge_fd_get_tcpip4_tuple(fs, rule, IPPROTO_SCTP);
+		break;
+	case TCP_V4_FLOW:
+		hclge_fd_get_tcpip4_tuple(fs, rule, IPPROTO_TCP);
+		break;
+	case UDP_V4_FLOW:
+		hclge_fd_get_tcpip4_tuple(fs, rule, IPPROTO_UDP);
+		break;
+	case IP_USER_FLOW:
+		hclge_fd_get_ip4_tuple(fs, rule);
+		break;
+	case SCTP_V6_FLOW:
+		hclge_fd_get_tcpip6_tuple(fs, rule, IPPROTO_SCTP);
+		break;
+	case TCP_V6_FLOW:
+		hclge_fd_get_tcpip6_tuple(fs, rule, IPPROTO_TCP);
+		break;
+	case UDP_V6_FLOW:
+		hclge_fd_get_tcpip6_tuple(fs, rule, IPPROTO_UDP);
+		break;
+	case IPV6_USER_FLOW:
+		hclge_fd_get_ip6_tuple(fs, rule);
+		break;
+	case ETHER_FLOW:
+		hclge_fd_get_ether_tuple(fs, rule);
+		break;
+	default:
+		return -EOPNOTSUPP;
+	}
+
+	if (fs->flow_type & FLOW_EXT) {
+		rule->tuples.vlan_tag1 = be16_to_cpu(fs->h_ext.vlan_tci);
+		rule->tuples_mask.vlan_tag1 = be16_to_cpu(fs->m_ext.vlan_tci);
+		hclge_fd_get_user_def_tuple(info, rule);
+	}
+
+	if (fs->flow_type & FLOW_MAC_EXT) {
+		ether_addr_copy(rule->tuples.dst_mac, fs->h_ext.h_dest);
+		ether_addr_copy(rule->tuples_mask.dst_mac, fs->m_ext.h_dest);
+	}
+
+	return 0;
+}
+
+static int hclge_fd_config_rule(struct hclge_dev *hdev,
+				struct hclge_fd_rule *rule)
+{
+	int ret;
+
+	ret = hclge_config_action(hdev, HCLGE_FD_STAGE_1, rule);
+	if (ret)
+		return ret;
+
+	return hclge_config_key(hdev, HCLGE_FD_STAGE_1, rule);
+}
+
+static int hclge_add_fd_entry_common(struct hclge_dev *hdev,
+				     struct hclge_fd_rule *rule)
+{
+	int ret;
+
+	spin_lock_bh(&hdev->fd_rule_lock);
+
+	if (hdev->fd_active_type != rule->rule_type &&
+	    (hdev->fd_active_type == HCLGE_FD_TC_FLOWER_ACTIVE ||
+	     hdev->fd_active_type == HCLGE_FD_EP_ACTIVE)) {
+		dev_err(&hdev->pdev->dev,
+			"mode conflict(new type %d, active type %d), please delete existent rules first\n",
+			rule->rule_type, hdev->fd_active_type);
+		spin_unlock_bh(&hdev->fd_rule_lock);
+		return -EINVAL;
+	}
+
+	ret = hclge_fd_check_user_def_refcnt(hdev, rule);
+	if (ret)
+		goto out;
+
+	ret = hclge_clear_arfs_rules(hdev);
+	if (ret)
+		goto out;
+
+	ret = hclge_fd_config_rule(hdev, rule);
+	if (ret)
+		goto out;
+
+	rule->state = HCLGE_FD_ACTIVE;
+	hdev->fd_active_type = rule->rule_type;
+	hclge_update_fd_list(hdev, rule->state, rule->location, rule);
+
+out:
+	spin_unlock_bh(&hdev->fd_rule_lock);
+	return ret;
+}
+
+static bool hclge_is_cls_flower_active(struct hnae3_handle *handle)
+{
+	struct hclge_vport *vport = hclge_get_vport(handle);
+	struct hclge_dev *hdev = vport->back;
+
+	return hdev->fd_active_type == HCLGE_FD_TC_FLOWER_ACTIVE;
+}
+
+static int hclge_fd_parse_ring_cookie(struct hclge_dev *hdev, u64 ring_cookie,
+				      u16 *vport_id, u8 *action, u16 *queue_id)
+{
+	struct hclge_vport *vport = hdev->vport;
+
+	if (ring_cookie == RX_CLS_FLOW_DISC) {
+		*action = HCLGE_FD_ACTION_DROP_PACKET;
+	} else {
+		u32 ring = ethtool_get_flow_spec_ring(ring_cookie);
+		u8 vf = ethtool_get_flow_spec_ring_vf(ring_cookie);
+		u16 tqps;
+
+		/* To keep consistent with user's configuration, minus 1 when
+		 * printing 'vf', because vf id from ethtool is added 1 for vf.
+		 */
+		if (vf > hdev->num_req_vfs) {
+			dev_err(&hdev->pdev->dev,
+				"Error: vf id (%u) should be less than %u\n",
+				vf - 1U, hdev->num_req_vfs);
+			return -EINVAL;
+		}
+
+		*vport_id = vf ? hdev->vport[vf].vport_id : vport->vport_id;
+		tqps = hdev->vport[vf].nic.kinfo.num_tqps;
+
+		if (ring >= tqps) {
+			dev_err(&hdev->pdev->dev,
+				"Error: queue id (%u) > max tqp num (%u)\n",
+				ring, tqps - 1U);
+			return -EINVAL;
+		}
+
+		*action = HCLGE_FD_ACTION_SELECT_QUEUE;
+		*queue_id = ring;
+	}
+
+	return 0;
+}
+
+static int hclge_add_fd_entry(struct hnae3_handle *handle,
+			      struct ethtool_rxnfc *cmd)
+{
+	struct hclge_vport *vport = hclge_get_vport(handle);
+	struct hclge_dev *hdev = vport->back;
+	struct hclge_fd_user_def_info info;
+	u16 dst_vport_id = 0, q_index = 0;
+	struct ethtool_rx_flow_spec *fs;
+	struct hclge_fd_rule *rule;
+	u32 unused = 0;
+	u8 action;
+	int ret;
+
+	if (!hnae3_ae_dev_fd_supported(hdev->ae_dev)) {
+		dev_err(&hdev->pdev->dev,
+			"flow table director is not supported\n");
+		return -EOPNOTSUPP;
+	}
+
+	if (!hdev->fd_en) {
+		dev_err(&hdev->pdev->dev,
+			"please enable flow director first\n");
+		return -EOPNOTSUPP;
+	}
+
+	fs = (struct ethtool_rx_flow_spec *)&cmd->fs;
+
+	ret = hclge_fd_check_spec(hdev, fs, &unused, &info);
+	if (ret)
+		return ret;
+
+	ret = hclge_fd_parse_ring_cookie(hdev, fs->ring_cookie, &dst_vport_id,
+					 &action, &q_index);
+	if (ret)
+		return ret;
+
+	rule = kzalloc(sizeof(*rule), GFP_KERNEL);
+	if (!rule)
+		return -ENOMEM;
+
+	ret = hclge_fd_get_tuple(fs, rule, &info);
+	if (ret) {
+		kfree(rule);
+		return ret;
+	}
+
+	rule->flow_type = fs->flow_type;
+	rule->location = fs->location;
+	rule->unused_tuple = unused;
+	rule->vf_id = dst_vport_id;
+	rule->queue_id = q_index;
+	rule->action = action;
+	rule->rule_type = HCLGE_FD_EP_ACTIVE;
+
+	ret = hclge_add_fd_entry_common(hdev, rule);
+	if (ret)
+		kfree(rule);
+
+	return ret;
+}
+
+static int hclge_del_fd_entry(struct hnae3_handle *handle,
+			      struct ethtool_rxnfc *cmd)
+{
+	struct hclge_vport *vport = hclge_get_vport(handle);
+	struct hclge_dev *hdev = vport->back;
+	struct ethtool_rx_flow_spec *fs;
+	int ret;
+
+	if (!hnae3_ae_dev_fd_supported(hdev->ae_dev))
+		return -EOPNOTSUPP;
+
+	fs = (struct ethtool_rx_flow_spec *)&cmd->fs;
+
+	if (fs->location >= hdev->fd_cfg.rule_num[HCLGE_FD_STAGE_1])
+		return -EINVAL;
+
+	spin_lock_bh(&hdev->fd_rule_lock);
+	if (hdev->fd_active_type == HCLGE_FD_TC_FLOWER_ACTIVE ||
+	    !test_bit(fs->location, hdev->fd_bmap)) {
+		dev_err(&hdev->pdev->dev,
+			"Delete fail, rule %u is inexistent\n", fs->location);
+		spin_unlock_bh(&hdev->fd_rule_lock);
+		return -ENOENT;
+	}
+
+	ret = hclge_fd_tcam_config(hdev, HCLGE_FD_STAGE_1, true, fs->location,
+				   NULL, false);
+	if (ret)
+		goto out;
+
+	hclge_update_fd_list(hdev, HCLGE_FD_DELETED, fs->location, NULL);
+
+out:
+	spin_unlock_bh(&hdev->fd_rule_lock);
+	return ret;
+}
+
+static void hclge_clear_fd_rules_in_list(struct hclge_dev *hdev,
+					 bool clear_list)
+{
+	struct hclge_fd_rule *rule;
+	struct hlist_node *node;
+	u16 location;
+
+	spin_lock_bh(&hdev->fd_rule_lock);
+
+	for_each_set_bit(location, hdev->fd_bmap,
+			 hdev->fd_cfg.rule_num[HCLGE_FD_STAGE_1])
+		hclge_fd_tcam_config(hdev, HCLGE_FD_STAGE_1, true, location,
+				     NULL, false);
+
+	if (clear_list) {
+		hlist_for_each_entry_safe(rule, node, &hdev->fd_rule_list,
+					  rule_node) {
+			hlist_del(&rule->rule_node);
+			kfree(rule);
+		}
+		hdev->fd_active_type = HCLGE_FD_RULE_NONE;
+		hdev->hclge_fd_rule_num = 0;
+		bitmap_zero(hdev->fd_bmap,
+			    hdev->fd_cfg.rule_num[HCLGE_FD_STAGE_1]);
+	}
+
+	spin_unlock_bh(&hdev->fd_rule_lock);
+}
+
+static void hclge_del_all_fd_entries(struct hclge_dev *hdev)
+{
+	if (!hnae3_ae_dev_fd_supported(hdev->ae_dev))
+		return;
+
+	hclge_clear_fd_rules_in_list(hdev, true);
+	hclge_fd_disable_user_def(hdev);
+}
+
+static int hclge_restore_fd_entries(struct hnae3_handle *handle)
+{
+	struct hclge_vport *vport = hclge_get_vport(handle);
+	struct hclge_dev *hdev = vport->back;
+	struct hclge_fd_rule *rule;
+	struct hlist_node *node;
+
+	/* Return ok here, because reset error handling will check this
+	 * return value. If error is returned here, the reset process will
+	 * fail.
+	 */
+	if (!hnae3_ae_dev_fd_supported(hdev->ae_dev))
+		return 0;
+
+	/* if fd is disabled, should not restore it when reset */
+	if (!hdev->fd_en)
+		return 0;
+
+	spin_lock_bh(&hdev->fd_rule_lock);
+	hlist_for_each_entry_safe(rule, node, &hdev->fd_rule_list, rule_node) {
+		if (rule->state == HCLGE_FD_ACTIVE)
+			rule->state = HCLGE_FD_TO_ADD;
+	}
+	spin_unlock_bh(&hdev->fd_rule_lock);
+	set_bit(HCLGE_STATE_FD_TBL_CHANGED, &hdev->state);
+
+	return 0;
+}
+
+static int hclge_get_fd_rule_cnt(struct hnae3_handle *handle,
+				 struct ethtool_rxnfc *cmd)
+{
+	struct hclge_vport *vport = hclge_get_vport(handle);
+	struct hclge_dev *hdev = vport->back;
+
+	if (!hnae3_ae_dev_fd_supported(hdev->ae_dev) || hclge_is_cls_flower_active(handle))
+		return -EOPNOTSUPP;
+
+	cmd->rule_cnt = hdev->hclge_fd_rule_num;
+	cmd->data = hdev->fd_cfg.rule_num[HCLGE_FD_STAGE_1];
+
+	return 0;
+}
+
+static void hclge_fd_get_tcpip4_info(struct hclge_fd_rule *rule,
+				     struct ethtool_tcpip4_spec *spec,
+				     struct ethtool_tcpip4_spec *spec_mask)
+{
+	spec->ip4src = cpu_to_be32(rule->tuples.src_ip[IPV4_INDEX]);
+	spec_mask->ip4src = rule->unused_tuple & BIT(INNER_SRC_IP) ?
+			0 : cpu_to_be32(rule->tuples_mask.src_ip[IPV4_INDEX]);
+
+	spec->ip4dst = cpu_to_be32(rule->tuples.dst_ip[IPV4_INDEX]);
+	spec_mask->ip4dst = rule->unused_tuple & BIT(INNER_DST_IP) ?
+			0 : cpu_to_be32(rule->tuples_mask.dst_ip[IPV4_INDEX]);
+
+	spec->psrc = cpu_to_be16(rule->tuples.src_port);
+	spec_mask->psrc = rule->unused_tuple & BIT(INNER_SRC_PORT) ?
+			0 : cpu_to_be16(rule->tuples_mask.src_port);
+
+	spec->pdst = cpu_to_be16(rule->tuples.dst_port);
+	spec_mask->pdst = rule->unused_tuple & BIT(INNER_DST_PORT) ?
+			0 : cpu_to_be16(rule->tuples_mask.dst_port);
+
+	spec->tos = rule->tuples.ip_tos;
+	spec_mask->tos = rule->unused_tuple & BIT(INNER_IP_TOS) ?
+			0 : rule->tuples_mask.ip_tos;
+}
+
+static void hclge_fd_get_ip4_info(struct hclge_fd_rule *rule,
+				  struct ethtool_usrip4_spec *spec,
+				  struct ethtool_usrip4_spec *spec_mask)
+{
+	spec->ip4src = cpu_to_be32(rule->tuples.src_ip[IPV4_INDEX]);
+	spec_mask->ip4src = rule->unused_tuple & BIT(INNER_SRC_IP) ?
+			0 : cpu_to_be32(rule->tuples_mask.src_ip[IPV4_INDEX]);
+
+	spec->ip4dst = cpu_to_be32(rule->tuples.dst_ip[IPV4_INDEX]);
+	spec_mask->ip4dst = rule->unused_tuple & BIT(INNER_DST_IP) ?
+			0 : cpu_to_be32(rule->tuples_mask.dst_ip[IPV4_INDEX]);
+
+	spec->tos = rule->tuples.ip_tos;
+	spec_mask->tos = rule->unused_tuple & BIT(INNER_IP_TOS) ?
+			0 : rule->tuples_mask.ip_tos;
+
+	spec->proto = rule->tuples.ip_proto;
+	spec_mask->proto = rule->unused_tuple & BIT(INNER_IP_PROTO) ?
+			0 : rule->tuples_mask.ip_proto;
+
+	spec->ip_ver = ETH_RX_NFC_IP4;
+}
+
+static void hclge_fd_get_tcpip6_info(struct hclge_fd_rule *rule,
+				     struct ethtool_tcpip6_spec *spec,
+				     struct ethtool_tcpip6_spec *spec_mask)
+{
+	ipv6_addr_cpu_to_be32(spec->ip6src, rule->tuples.src_ip);
+	ipv6_addr_cpu_to_be32(spec->ip6dst, rule->tuples.dst_ip);
+	if (rule->unused_tuple & BIT(INNER_SRC_IP))
+		memset(spec_mask->ip6src, 0, sizeof(spec_mask->ip6src));
+	else
+		ipv6_addr_cpu_to_be32(spec_mask->ip6src,
+				      rule->tuples_mask.src_ip);
+
+	if (rule->unused_tuple & BIT(INNER_DST_IP))
+		memset(spec_mask->ip6dst, 0, sizeof(spec_mask->ip6dst));
+	else
+		ipv6_addr_cpu_to_be32(spec_mask->ip6dst,
+				      rule->tuples_mask.dst_ip);
+
+	spec->tclass = rule->tuples.ip_tos;
+	spec_mask->tclass = rule->unused_tuple & BIT(INNER_IP_TOS) ?
+			0 : rule->tuples_mask.ip_tos;
+
+	spec->psrc = cpu_to_be16(rule->tuples.src_port);
+	spec_mask->psrc = rule->unused_tuple & BIT(INNER_SRC_PORT) ?
+			0 : cpu_to_be16(rule->tuples_mask.src_port);
+
+	spec->pdst = cpu_to_be16(rule->tuples.dst_port);
+	spec_mask->pdst = rule->unused_tuple & BIT(INNER_DST_PORT) ?
+			0 : cpu_to_be16(rule->tuples_mask.dst_port);
+}
+
+static void hclge_fd_get_ip6_info(struct hclge_fd_rule *rule,
+				  struct ethtool_usrip6_spec *spec,
+				  struct ethtool_usrip6_spec *spec_mask)
+{
+	ipv6_addr_cpu_to_be32(spec->ip6src, rule->tuples.src_ip);
+	ipv6_addr_cpu_to_be32(spec->ip6dst, rule->tuples.dst_ip);
+	if (rule->unused_tuple & BIT(INNER_SRC_IP))
+		memset(spec_mask->ip6src, 0, sizeof(spec_mask->ip6src));
+	else
+		ipv6_addr_cpu_to_be32(spec_mask->ip6src,
+				      rule->tuples_mask.src_ip);
+
+	if (rule->unused_tuple & BIT(INNER_DST_IP))
+		memset(spec_mask->ip6dst, 0, sizeof(spec_mask->ip6dst));
+	else
+		ipv6_addr_cpu_to_be32(spec_mask->ip6dst,
+				      rule->tuples_mask.dst_ip);
+
+	spec->tclass = rule->tuples.ip_tos;
+	spec_mask->tclass = rule->unused_tuple & BIT(INNER_IP_TOS) ?
+			0 : rule->tuples_mask.ip_tos;
+
+	spec->l4_proto = rule->tuples.ip_proto;
+	spec_mask->l4_proto = rule->unused_tuple & BIT(INNER_IP_PROTO) ?
+			0 : rule->tuples_mask.ip_proto;
+}
+
+static void hclge_fd_get_ether_info(struct hclge_fd_rule *rule,
+				    struct ethhdr *spec,
+				    struct ethhdr *spec_mask)
+{
+	ether_addr_copy(spec->h_source, rule->tuples.src_mac);
+	ether_addr_copy(spec->h_dest, rule->tuples.dst_mac);
+
+	if (rule->unused_tuple & BIT(INNER_SRC_MAC))
+		eth_zero_addr(spec_mask->h_source);
+	else
+		ether_addr_copy(spec_mask->h_source, rule->tuples_mask.src_mac);
+
+	if (rule->unused_tuple & BIT(INNER_DST_MAC))
+		eth_zero_addr(spec_mask->h_dest);
+	else
+		ether_addr_copy(spec_mask->h_dest, rule->tuples_mask.dst_mac);
+
+	spec->h_proto = cpu_to_be16(rule->tuples.ether_proto);
+	spec_mask->h_proto = rule->unused_tuple & BIT(INNER_ETH_TYPE) ?
+			0 : cpu_to_be16(rule->tuples_mask.ether_proto);
+}
+
+static void hclge_fd_get_user_def_info(struct ethtool_rx_flow_spec *fs,
+				       struct hclge_fd_rule *rule)
+{
+	if ((rule->unused_tuple & HCLGE_FD_TUPLE_USER_DEF_TUPLES) ==
+	    HCLGE_FD_TUPLE_USER_DEF_TUPLES) {
+		fs->h_ext.data[0] = 0;
+		fs->h_ext.data[1] = 0;
+		fs->m_ext.data[0] = 0;
+		fs->m_ext.data[1] = 0;
+	} else {
+		fs->h_ext.data[0] = cpu_to_be32(rule->ep.user_def.offset);
+		fs->h_ext.data[1] = cpu_to_be32(rule->ep.user_def.data);
+		fs->m_ext.data[0] =
+				cpu_to_be32(HCLGE_FD_USER_DEF_OFFSET_UNMASK);
+		fs->m_ext.data[1] = cpu_to_be32(rule->ep.user_def.data_mask);
+	}
+}
+
+static void hclge_fd_get_ext_info(struct ethtool_rx_flow_spec *fs,
+				  struct hclge_fd_rule *rule)
+{
+	if (fs->flow_type & FLOW_EXT) {
+		fs->h_ext.vlan_tci = cpu_to_be16(rule->tuples.vlan_tag1);
+		fs->m_ext.vlan_tci =
+				rule->unused_tuple & BIT(INNER_VLAN_TAG_FST) ?
+				0 : cpu_to_be16(rule->tuples_mask.vlan_tag1);
+
+		hclge_fd_get_user_def_info(fs, rule);
+	}
+
+	if (fs->flow_type & FLOW_MAC_EXT) {
+		ether_addr_copy(fs->h_ext.h_dest, rule->tuples.dst_mac);
+		if (rule->unused_tuple & BIT(INNER_DST_MAC))
+			eth_zero_addr(fs->m_u.ether_spec.h_dest);
+		else
+			ether_addr_copy(fs->m_u.ether_spec.h_dest,
+					rule->tuples_mask.dst_mac);
+	}
+}
+
+static struct hclge_fd_rule *hclge_get_fd_rule(struct hclge_dev *hdev,
+					       u16 location)
+{
+	struct hclge_fd_rule *rule = NULL;
+	struct hlist_node *node2;
+
+	hlist_for_each_entry_safe(rule, node2, &hdev->fd_rule_list, rule_node) {
+		if (rule->location == location)
+			return rule;
+		else if (rule->location > location)
+			return NULL;
+	}
+
+	return NULL;
+}
+
+static void hclge_fd_get_ring_cookie(struct ethtool_rx_flow_spec *fs,
+				     struct hclge_fd_rule *rule)
+{
+	if (rule->action == HCLGE_FD_ACTION_DROP_PACKET) {
+		fs->ring_cookie = RX_CLS_FLOW_DISC;
+	} else {
+		u64 vf_id;
+
+		fs->ring_cookie = rule->queue_id;
+		vf_id = rule->vf_id;
+		vf_id <<= ETHTOOL_RX_FLOW_SPEC_RING_VF_OFF;
+		fs->ring_cookie |= vf_id;
+	}
+}
+
+static int hclge_get_fd_rule_info(struct hnae3_handle *handle,
+				  struct ethtool_rxnfc *cmd)
+{
+	struct hclge_vport *vport = hclge_get_vport(handle);
+	struct hclge_fd_rule *rule = NULL;
+	struct hclge_dev *hdev = vport->back;
+	struct ethtool_rx_flow_spec *fs;
+
+	if (!hnae3_ae_dev_fd_supported(hdev->ae_dev))
+		return -EOPNOTSUPP;
+
+	fs = (struct ethtool_rx_flow_spec *)&cmd->fs;
+
+	spin_lock_bh(&hdev->fd_rule_lock);
+
+	rule = hclge_get_fd_rule(hdev, fs->location);
+	if (!rule) {
+		spin_unlock_bh(&hdev->fd_rule_lock);
+		return -ENOENT;
+	}
+
+	fs->flow_type = rule->flow_type;
+	switch (fs->flow_type & ~(FLOW_EXT | FLOW_MAC_EXT)) {
+	case SCTP_V4_FLOW:
+	case TCP_V4_FLOW:
+	case UDP_V4_FLOW:
+		hclge_fd_get_tcpip4_info(rule, &fs->h_u.tcp_ip4_spec,
+					 &fs->m_u.tcp_ip4_spec);
+		break;
+	case IP_USER_FLOW:
+		hclge_fd_get_ip4_info(rule, &fs->h_u.usr_ip4_spec,
+				      &fs->m_u.usr_ip4_spec);
+		break;
+	case SCTP_V6_FLOW:
+	case TCP_V6_FLOW:
+	case UDP_V6_FLOW:
+		hclge_fd_get_tcpip6_info(rule, &fs->h_u.tcp_ip6_spec,
+					 &fs->m_u.tcp_ip6_spec);
+		break;
+	case IPV6_USER_FLOW:
+		hclge_fd_get_ip6_info(rule, &fs->h_u.usr_ip6_spec,
+				      &fs->m_u.usr_ip6_spec);
+		break;
+	/* The flow type of fd rule has been checked before adding in to rule
+	 * list. As other flow types have been handled, it must be ETHER_FLOW
+	 * for the default case
+	 */
+	default:
+		hclge_fd_get_ether_info(rule, &fs->h_u.ether_spec,
+					&fs->m_u.ether_spec);
+		break;
+	}
+
+	hclge_fd_get_ext_info(fs, rule);
+
+	hclge_fd_get_ring_cookie(fs, rule);
+
+	spin_unlock_bh(&hdev->fd_rule_lock);
+
+	return 0;
+}
+
+static int hclge_get_all_rules(struct hnae3_handle *handle,
+			       struct ethtool_rxnfc *cmd, u32 *rule_locs)
+{
+	struct hclge_vport *vport = hclge_get_vport(handle);
+	struct hclge_dev *hdev = vport->back;
+	struct hclge_fd_rule *rule;
+	struct hlist_node *node2;
+	u32 cnt = 0;
+
+	if (!hnae3_ae_dev_fd_supported(hdev->ae_dev))
+		return -EOPNOTSUPP;
+
+	cmd->data = hdev->fd_cfg.rule_num[HCLGE_FD_STAGE_1];
+
+	spin_lock_bh(&hdev->fd_rule_lock);
+	hlist_for_each_entry_safe(rule, node2,
+				  &hdev->fd_rule_list, rule_node) {
+		if (cnt == cmd->rule_cnt) {
+			spin_unlock_bh(&hdev->fd_rule_lock);
+			return -EMSGSIZE;
+		}
+
+		if (rule->state == HCLGE_FD_TO_DEL)
+			continue;
+
+		rule_locs[cnt] = rule->location;
+		cnt++;
+	}
+
+	spin_unlock_bh(&hdev->fd_rule_lock);
+
+	cmd->rule_cnt = cnt;
+
+	return 0;
+}
+
+static void hclge_fd_get_flow_tuples(const struct flow_keys *fkeys,
+				     struct hclge_fd_rule_tuples *tuples)
+{
+#define flow_ip6_src fkeys->addrs.v6addrs.src.in6_u.u6_addr32
+#define flow_ip6_dst fkeys->addrs.v6addrs.dst.in6_u.u6_addr32
+
+	tuples->ether_proto = be16_to_cpu(fkeys->basic.n_proto);
+	tuples->ip_proto = fkeys->basic.ip_proto;
+	tuples->dst_port = be16_to_cpu(fkeys->ports.dst);
+
+	if (fkeys->basic.n_proto == htons(ETH_P_IP)) {
+		tuples->src_ip[3] = be32_to_cpu(fkeys->addrs.v4addrs.src);
+		tuples->dst_ip[3] = be32_to_cpu(fkeys->addrs.v4addrs.dst);
+	} else {
+		int i;
+
+		for (i = 0; i < IPV6_ADDR_WORDS; i++) {
+			tuples->src_ip[i] = be32_to_cpu(flow_ip6_src[i]);
+			tuples->dst_ip[i] = be32_to_cpu(flow_ip6_dst[i]);
+		}
+	}
+}
+
+/* traverse all rules, check whether an existed rule has the same tuples */
+static struct hclge_fd_rule *
+hclge_fd_search_flow_keys(struct hclge_dev *hdev,
+			  const struct hclge_fd_rule_tuples *tuples)
+{
+	struct hclge_fd_rule *rule = NULL;
+	struct hlist_node *node;
+
+	hlist_for_each_entry_safe(rule, node, &hdev->fd_rule_list, rule_node) {
+		if (!memcmp(tuples, &rule->tuples, sizeof(*tuples)))
+			return rule;
+	}
+
+	return NULL;
+}
+
+static void hclge_fd_build_arfs_rule(const struct hclge_fd_rule_tuples *tuples,
+				     struct hclge_fd_rule *rule)
+{
+	rule->unused_tuple = BIT(INNER_SRC_MAC) | BIT(INNER_DST_MAC) |
+			     BIT(INNER_VLAN_TAG_FST) | BIT(INNER_IP_TOS) |
+			     BIT(INNER_SRC_PORT);
+	rule->action = 0;
+	rule->vf_id = 0;
+	rule->rule_type = HCLGE_FD_ARFS_ACTIVE;
+	rule->state = HCLGE_FD_TO_ADD;
+	if (tuples->ether_proto == ETH_P_IP) {
+		if (tuples->ip_proto == IPPROTO_TCP)
+			rule->flow_type = TCP_V4_FLOW;
+		else
+			rule->flow_type = UDP_V4_FLOW;
+	} else {
+		if (tuples->ip_proto == IPPROTO_TCP)
+			rule->flow_type = TCP_V6_FLOW;
+		else
+			rule->flow_type = UDP_V6_FLOW;
+	}
+	memcpy(&rule->tuples, tuples, sizeof(rule->tuples));
+	memset(&rule->tuples_mask, 0xFF, sizeof(rule->tuples_mask));
+}
+
+static int hclge_add_fd_entry_by_arfs(struct hnae3_handle *handle, u16 queue_id,
+				      u16 flow_id, struct flow_keys *fkeys)
+{
+	struct hclge_vport *vport = hclge_get_vport(handle);
+	struct hclge_fd_rule_tuples new_tuples = {};
+	struct hclge_dev *hdev = vport->back;
+	struct hclge_fd_rule *rule;
+	u16 bit_id;
+
+	if (!hnae3_ae_dev_fd_supported(hdev->ae_dev))
+		return -EOPNOTSUPP;
+
+	/* when there is already fd rule existed add by user,
+	 * arfs should not work
+	 */
+	spin_lock_bh(&hdev->fd_rule_lock);
+	if (hdev->fd_active_type != HCLGE_FD_ARFS_ACTIVE &&
+	    hdev->fd_active_type != HCLGE_FD_RULE_NONE) {
+		spin_unlock_bh(&hdev->fd_rule_lock);
+		return -EOPNOTSUPP;
+	}
+
+	hclge_fd_get_flow_tuples(fkeys, &new_tuples);
+
+	/* check is there flow director filter existed for this flow,
+	 * if not, create a new filter for it;
+	 * if filter exist with different queue id, modify the filter;
+	 * if filter exist with same queue id, do nothing
+	 */
+	rule = hclge_fd_search_flow_keys(hdev, &new_tuples);
+	if (!rule) {
+		bit_id = find_first_zero_bit(hdev->fd_bmap, MAX_FD_FILTER_NUM);
+		if (bit_id >= hdev->fd_cfg.rule_num[HCLGE_FD_STAGE_1]) {
+			spin_unlock_bh(&hdev->fd_rule_lock);
+			return -ENOSPC;
+		}
+
+		rule = kzalloc(sizeof(*rule), GFP_ATOMIC);
+		if (!rule) {
+			spin_unlock_bh(&hdev->fd_rule_lock);
+			return -ENOMEM;
+		}
+
+		rule->location = bit_id;
+		rule->arfs.flow_id = flow_id;
+		rule->queue_id = queue_id;
+		hclge_fd_build_arfs_rule(&new_tuples, rule);
+		hclge_update_fd_list(hdev, rule->state, rule->location, rule);
+		hdev->fd_active_type = HCLGE_FD_ARFS_ACTIVE;
+	} else if (rule->queue_id != queue_id) {
+		rule->queue_id = queue_id;
+		rule->state = HCLGE_FD_TO_ADD;
+		set_bit(HCLGE_STATE_FD_TBL_CHANGED, &hdev->state);
+		hclge_task_schedule(hdev, 0);
+	}
+	spin_unlock_bh(&hdev->fd_rule_lock);
+	return rule->location;
+}
+
+static void hclge_rfs_filter_expire(struct hclge_dev *hdev)
+{
+#ifdef CONFIG_RFS_ACCEL
+	struct hnae3_handle *handle = &hdev->vport[0].nic;
+	struct hclge_fd_rule *rule;
+	struct hlist_node *node;
+
+	spin_lock_bh(&hdev->fd_rule_lock);
+	if (hdev->fd_active_type != HCLGE_FD_ARFS_ACTIVE) {
+		spin_unlock_bh(&hdev->fd_rule_lock);
+		return;
+	}
+	hlist_for_each_entry_safe(rule, node, &hdev->fd_rule_list, rule_node) {
+		if (rule->state != HCLGE_FD_ACTIVE)
+			continue;
+		if (rps_may_expire_flow(handle->netdev, rule->queue_id,
+					rule->arfs.flow_id, rule->location)) {
+			rule->state = HCLGE_FD_TO_DEL;
+			set_bit(HCLGE_STATE_FD_TBL_CHANGED, &hdev->state);
+		}
+	}
+	spin_unlock_bh(&hdev->fd_rule_lock);
+#endif
+}
+
+/* make sure being called after lock up with fd_rule_lock */
+static int hclge_clear_arfs_rules(struct hclge_dev *hdev)
+{
+#ifdef CONFIG_RFS_ACCEL
+	struct hclge_fd_rule *rule;
+	struct hlist_node *node;
+	int ret;
+
+	if (hdev->fd_active_type != HCLGE_FD_ARFS_ACTIVE)
+		return 0;
+
+	hlist_for_each_entry_safe(rule, node, &hdev->fd_rule_list, rule_node) {
+		switch (rule->state) {
+		case HCLGE_FD_TO_DEL:
+		case HCLGE_FD_ACTIVE:
+			ret = hclge_fd_tcam_config(hdev, HCLGE_FD_STAGE_1, true,
+						   rule->location, NULL, false);
+			if (ret)
+				return ret;
+			fallthrough;
+		case HCLGE_FD_TO_ADD:
+			hclge_fd_dec_rule_cnt(hdev, rule->location);
+			hlist_del(&rule->rule_node);
+			kfree(rule);
+			break;
+		default:
+			break;
+		}
+	}
+	hclge_sync_fd_state(hdev);
+
+#endif
+	return 0;
+}
+
+static void hclge_get_cls_key_basic(const struct flow_rule *flow,
+				    struct hclge_fd_rule *rule)
+{
+	if (flow_rule_match_key(flow, FLOW_DISSECTOR_KEY_BASIC)) {
+		struct flow_match_basic match;
+		u16 ethtype_key, ethtype_mask;
+
+		flow_rule_match_basic(flow, &match);
+		ethtype_key = ntohs(match.key->n_proto);
+		ethtype_mask = ntohs(match.mask->n_proto);
+
+		if (ethtype_key == ETH_P_ALL) {
+			ethtype_key = 0;
+			ethtype_mask = 0;
+		}
+		rule->tuples.ether_proto = ethtype_key;
+		rule->tuples_mask.ether_proto = ethtype_mask;
+		rule->tuples.ip_proto = match.key->ip_proto;
+		rule->tuples_mask.ip_proto = match.mask->ip_proto;
+	} else {
+		rule->unused_tuple |= BIT(INNER_IP_PROTO);
+		rule->unused_tuple |= BIT(INNER_ETH_TYPE);
+	}
+}
+
+static void hclge_get_cls_key_mac(const struct flow_rule *flow,
+				  struct hclge_fd_rule *rule)
+{
+	if (flow_rule_match_key(flow, FLOW_DISSECTOR_KEY_ETH_ADDRS)) {
+		struct flow_match_eth_addrs match;
+
+		flow_rule_match_eth_addrs(flow, &match);
+		ether_addr_copy(rule->tuples.dst_mac, match.key->dst);
+		ether_addr_copy(rule->tuples_mask.dst_mac, match.mask->dst);
+		ether_addr_copy(rule->tuples.src_mac, match.key->src);
+		ether_addr_copy(rule->tuples_mask.src_mac, match.mask->src);
+	} else {
+		rule->unused_tuple |= BIT(INNER_DST_MAC);
+		rule->unused_tuple |= BIT(INNER_SRC_MAC);
+	}
+}
+
+static void hclge_get_cls_key_vlan(const struct flow_rule *flow,
+				   struct hclge_fd_rule *rule)
+{
+	if (flow_rule_match_key(flow, FLOW_DISSECTOR_KEY_VLAN)) {
+		struct flow_match_vlan match;
+
+		flow_rule_match_vlan(flow, &match);
+		rule->tuples.vlan_tag1 = match.key->vlan_id |
+				(match.key->vlan_priority << VLAN_PRIO_SHIFT);
+		rule->tuples_mask.vlan_tag1 = match.mask->vlan_id |
+				(match.mask->vlan_priority << VLAN_PRIO_SHIFT);
+	} else {
+		rule->unused_tuple |= BIT(INNER_VLAN_TAG_FST);
+	}
+}
+
+static int hclge_get_cls_key_ip(const struct flow_rule *flow,
+				struct hclge_fd_rule *rule,
+				struct netlink_ext_ack *extack)
+{
+	u16 addr_type = 0;
+
+	if (flow_rule_match_key(flow, FLOW_DISSECTOR_KEY_CONTROL)) {
+		struct flow_match_control match;
+
+		flow_rule_match_control(flow, &match);
+		addr_type = match.key->addr_type;
+
+		if (flow_rule_has_control_flags(match.mask->flags, extack))
+			return -EOPNOTSUPP;
+	}
+
+	if (addr_type == FLOW_DISSECTOR_KEY_IPV4_ADDRS) {
+		struct flow_match_ipv4_addrs match;
+
+		flow_rule_match_ipv4_addrs(flow, &match);
+		rule->tuples.src_ip[IPV4_INDEX] = be32_to_cpu(match.key->src);
+		rule->tuples_mask.src_ip[IPV4_INDEX] =
+						be32_to_cpu(match.mask->src);
+		rule->tuples.dst_ip[IPV4_INDEX] = be32_to_cpu(match.key->dst);
+		rule->tuples_mask.dst_ip[IPV4_INDEX] =
+						be32_to_cpu(match.mask->dst);
+	} else if (addr_type == FLOW_DISSECTOR_KEY_IPV6_ADDRS) {
+		struct flow_match_ipv6_addrs match;
+
+		flow_rule_match_ipv6_addrs(flow, &match);
+		ipv6_addr_be32_to_cpu(rule->tuples.src_ip,
+				      match.key->src.s6_addr32);
+		ipv6_addr_be32_to_cpu(rule->tuples_mask.src_ip,
+				      match.mask->src.s6_addr32);
+		ipv6_addr_be32_to_cpu(rule->tuples.dst_ip,
+				      match.key->dst.s6_addr32);
+		ipv6_addr_be32_to_cpu(rule->tuples_mask.dst_ip,
+				      match.mask->dst.s6_addr32);
+	} else {
+		rule->unused_tuple |= BIT(INNER_SRC_IP);
+		rule->unused_tuple |= BIT(INNER_DST_IP);
+	}
+
+	return 0;
+}
+
+static void hclge_get_cls_key_port(const struct flow_rule *flow,
+				   struct hclge_fd_rule *rule)
+{
+	if (flow_rule_match_key(flow, FLOW_DISSECTOR_KEY_PORTS)) {
+		struct flow_match_ports match;
+
+		flow_rule_match_ports(flow, &match);
+
+		rule->tuples.src_port = be16_to_cpu(match.key->src);
+		rule->tuples_mask.src_port = be16_to_cpu(match.mask->src);
+		rule->tuples.dst_port = be16_to_cpu(match.key->dst);
+		rule->tuples_mask.dst_port = be16_to_cpu(match.mask->dst);
+	} else {
+		rule->unused_tuple |= BIT(INNER_SRC_PORT);
+		rule->unused_tuple |= BIT(INNER_DST_PORT);
+	}
+}
+
+static int hclge_parse_cls_flower(struct hclge_dev *hdev,
+				  struct flow_cls_offload *cls_flower,
+				  struct hclge_fd_rule *rule)
+{
+	struct flow_rule *flow = flow_cls_offload_flow_rule(cls_flower);
+	struct netlink_ext_ack *extack = cls_flower->common.extack;
+	struct flow_dissector *dissector = flow->match.dissector;
+	int ret;
+
+	if (dissector->used_keys &
+	    ~(BIT_ULL(FLOW_DISSECTOR_KEY_CONTROL) |
+	      BIT_ULL(FLOW_DISSECTOR_KEY_BASIC) |
+	      BIT_ULL(FLOW_DISSECTOR_KEY_ETH_ADDRS) |
+	      BIT_ULL(FLOW_DISSECTOR_KEY_VLAN) |
+	      BIT_ULL(FLOW_DISSECTOR_KEY_IPV4_ADDRS) |
+	      BIT_ULL(FLOW_DISSECTOR_KEY_IPV6_ADDRS) |
+	      BIT_ULL(FLOW_DISSECTOR_KEY_PORTS))) {
+		dev_err(&hdev->pdev->dev, "unsupported key set: %#llx\n",
+			dissector->used_keys);
+		return -EOPNOTSUPP;
+	}
+
+	hclge_get_cls_key_basic(flow, rule);
+	hclge_get_cls_key_mac(flow, rule);
+	hclge_get_cls_key_vlan(flow, rule);
+
+	ret = hclge_get_cls_key_ip(flow, rule, extack);
+	if (ret)
+		return ret;
+
+	hclge_get_cls_key_port(flow, rule);
+
+	return 0;
+}
+
+static int hclge_check_cls_flower(struct hclge_dev *hdev,
+				  struct flow_cls_offload *cls_flower, int tc)
+{
+	u32 prio = cls_flower->common.prio;
+
+	if (tc < 0 || tc > hdev->tc_max) {
+		dev_err(&hdev->pdev->dev, "invalid traffic class\n");
+		return -EINVAL;
+	}
+
+	if (prio == 0 ||
+	    prio > hdev->fd_cfg.rule_num[HCLGE_FD_STAGE_1]) {
+		dev_err(&hdev->pdev->dev,
+			"prio %u should be in range[1, %u]\n",
+			prio, hdev->fd_cfg.rule_num[HCLGE_FD_STAGE_1]);
+		return -EINVAL;
+	}
+
+	if (test_bit(prio - 1, hdev->fd_bmap)) {
+		dev_err(&hdev->pdev->dev, "prio %u is already used\n", prio);
+		return -EINVAL;
+	}
+	return 0;
+}
+
+static int hclge_add_cls_flower(struct hnae3_handle *handle,
+				struct flow_cls_offload *cls_flower,
+				int tc)
+{
+	struct hclge_vport *vport = hclge_get_vport(handle);
+	struct hclge_dev *hdev = vport->back;
+	struct hclge_fd_rule *rule;
+	int ret;
+
+	if (!hnae3_ae_dev_fd_supported(hdev->ae_dev)) {
+		dev_err(&hdev->pdev->dev,
+			"cls flower is not supported\n");
+		return -EOPNOTSUPP;
+	}
+
+	ret = hclge_check_cls_flower(hdev, cls_flower, tc);
+	if (ret) {
+		dev_err(&hdev->pdev->dev,
+			"failed to check cls flower params, ret = %d\n", ret);
+		return ret;
+	}
+
+	rule = kzalloc(sizeof(*rule), GFP_KERNEL);
+	if (!rule)
+		return -ENOMEM;
+
+	ret = hclge_parse_cls_flower(hdev, cls_flower, rule);
+	if (ret) {
+		kfree(rule);
+		return ret;
+	}
+
+	rule->action = HCLGE_FD_ACTION_SELECT_TC;
+	rule->cls_flower.tc = tc;
+	rule->location = cls_flower->common.prio - 1;
+	rule->vf_id = 0;
+	rule->cls_flower.cookie = cls_flower->cookie;
+	rule->rule_type = HCLGE_FD_TC_FLOWER_ACTIVE;
+
+	ret = hclge_add_fd_entry_common(hdev, rule);
+	if (ret)
+		kfree(rule);
+
+	return ret;
+}
+
+static struct hclge_fd_rule *hclge_find_cls_flower(struct hclge_dev *hdev,
+						   unsigned long cookie)
+{
+	struct hclge_fd_rule *rule;
+	struct hlist_node *node;
+
+	hlist_for_each_entry_safe(rule, node, &hdev->fd_rule_list, rule_node) {
+		if (rule->cls_flower.cookie == cookie)
+			return rule;
+	}
+
+	return NULL;
+}
+
+static int hclge_del_cls_flower(struct hnae3_handle *handle,
+				struct flow_cls_offload *cls_flower)
+{
+	struct hclge_vport *vport = hclge_get_vport(handle);
+	struct hclge_dev *hdev = vport->back;
+	struct hclge_fd_rule *rule;
+	int ret;
+
+	if (!hnae3_ae_dev_fd_supported(hdev->ae_dev))
+		return -EOPNOTSUPP;
+
+	spin_lock_bh(&hdev->fd_rule_lock);
+
+	rule = hclge_find_cls_flower(hdev, cls_flower->cookie);
+	if (!rule) {
+		spin_unlock_bh(&hdev->fd_rule_lock);
+		return -EINVAL;
+	}
+
+	ret = hclge_fd_tcam_config(hdev, HCLGE_FD_STAGE_1, true, rule->location,
+				   NULL, false);
+	if (ret) {
+		/* if tcam config fail, set rule state to TO_DEL,
+		 * so the rule will be deleted when periodic
+		 * task being scheduled.
+		 */
+		hclge_update_fd_list(hdev, HCLGE_FD_TO_DEL, rule->location, NULL);
+		set_bit(HCLGE_STATE_FD_TBL_CHANGED, &hdev->state);
+		spin_unlock_bh(&hdev->fd_rule_lock);
+		return ret;
+	}
+
+	hclge_update_fd_list(hdev, HCLGE_FD_DELETED, rule->location, NULL);
+	spin_unlock_bh(&hdev->fd_rule_lock);
+
+	return 0;
+}
+
+static void hclge_sync_fd_list(struct hclge_dev *hdev, struct hlist_head *hlist)
+{
+	struct hclge_fd_rule *rule;
+	struct hlist_node *node;
+	int ret = 0;
+
+	if (!test_and_clear_bit(HCLGE_STATE_FD_TBL_CHANGED, &hdev->state))
+		return;
+
+	spin_lock_bh(&hdev->fd_rule_lock);
+
+	hlist_for_each_entry_safe(rule, node, hlist, rule_node) {
+		switch (rule->state) {
+		case HCLGE_FD_TO_ADD:
+			ret = hclge_fd_config_rule(hdev, rule);
+			if (ret)
+				goto out;
+			rule->state = HCLGE_FD_ACTIVE;
+			break;
+		case HCLGE_FD_TO_DEL:
+			ret = hclge_fd_tcam_config(hdev, HCLGE_FD_STAGE_1, true,
+						   rule->location, NULL, false);
+			if (ret)
+				goto out;
+			hclge_fd_dec_rule_cnt(hdev, rule->location);
+			hclge_fd_free_node(hdev, rule);
+			break;
+		default:
+			break;
+		}
+	}
+
+out:
+	if (ret)
+		set_bit(HCLGE_STATE_FD_TBL_CHANGED, &hdev->state);
+
+	spin_unlock_bh(&hdev->fd_rule_lock);
+}
+
+static void hclge_sync_fd_table(struct hclge_dev *hdev)
+{
+	if (!hnae3_ae_dev_fd_supported(hdev->ae_dev))
+		return;
+
+	if (test_and_clear_bit(HCLGE_STATE_FD_CLEAR_ALL, &hdev->state)) {
+		bool clear_list = hdev->fd_active_type == HCLGE_FD_ARFS_ACTIVE;
+
+		hclge_clear_fd_rules_in_list(hdev, clear_list);
+	}
+
+	hclge_sync_fd_user_def_cfg(hdev, false);
+
+	hclge_sync_fd_list(hdev, &hdev->fd_rule_list);
+}
+
+static bool hclge_get_hw_reset_stat(struct hnae3_handle *handle)
+{
+	struct hclge_vport *vport = hclge_get_vport(handle);
+	struct hclge_dev *hdev = vport->back;
+
+	return hclge_read_dev(&hdev->hw, HCLGE_GLOBAL_RESET_REG) ||
+	       hclge_read_dev(&hdev->hw, HCLGE_FUN_RST_ING);
+}
+
+static bool hclge_get_cmdq_stat(struct hnae3_handle *handle)
+{
+	struct hclge_vport *vport = hclge_get_vport(handle);
+	struct hclge_dev *hdev = vport->back;
+
+	return test_bit(HCLGE_COMM_STATE_CMD_DISABLE, &hdev->hw.hw.comm_state);
+}
+
+static bool hclge_ae_dev_resetting(struct hnae3_handle *handle)
+{
+	struct hclge_vport *vport = hclge_get_vport(handle);
+	struct hclge_dev *hdev = vport->back;
+
+	return test_bit(HCLGE_STATE_RST_HANDLING, &hdev->state);
+}
+
+static unsigned long hclge_ae_dev_reset_cnt(struct hnae3_handle *handle)
+{
+	struct hclge_vport *vport = hclge_get_vport(handle);
+	struct hclge_dev *hdev = vport->back;
+
+	return hdev->rst_stats.hw_reset_done_cnt;
+}
+
+static void hclge_enable_fd(struct hnae3_handle *handle, bool enable)
+{
+	struct hclge_vport *vport = hclge_get_vport(handle);
+	struct hclge_dev *hdev = vport->back;
+
+	hdev->fd_en = enable;
+
+	if (!enable)
+		set_bit(HCLGE_STATE_FD_CLEAR_ALL, &hdev->state);
+	else
+		hclge_restore_fd_entries(handle);
+
+	hclge_task_schedule(hdev, 0);
+}
+
+static void hclge_cfg_mac_mode(struct hclge_dev *hdev, bool enable)
+{
+#define HCLGE_LINK_STATUS_WAIT_CNT  3
+
+	struct hclge_desc desc;
+	struct hclge_config_mac_mode_cmd *req =
+		(struct hclge_config_mac_mode_cmd *)desc.data;
+	u32 loop_en = 0;
+	int ret;
+
+	hclge_cmd_setup_basic_desc(&desc, HCLGE_OPC_CONFIG_MAC_MODE, false);
+
+	if (enable) {
+		hnae3_set_bit(loop_en, HCLGE_MAC_TX_EN_B, 1U);
+		hnae3_set_bit(loop_en, HCLGE_MAC_RX_EN_B, 1U);
+		hnae3_set_bit(loop_en, HCLGE_MAC_PAD_TX_B, 1U);
+		hnae3_set_bit(loop_en, HCLGE_MAC_PAD_RX_B, 1U);
+		hnae3_set_bit(loop_en, HCLGE_MAC_FCS_TX_B, 1U);
+		hnae3_set_bit(loop_en, HCLGE_MAC_RX_FCS_B, 1U);
+		hnae3_set_bit(loop_en, HCLGE_MAC_RX_FCS_STRIP_B, 1U);
+		hnae3_set_bit(loop_en, HCLGE_MAC_TX_OVERSIZE_TRUNCATE_B, 1U);
+		hnae3_set_bit(loop_en, HCLGE_MAC_RX_OVERSIZE_TRUNCATE_B, 1U);
+		hnae3_set_bit(loop_en, HCLGE_MAC_TX_UNDER_MIN_ERR_B, 1U);
+	}
+
+	req->txrx_pad_fcs_loop_en = cpu_to_le32(loop_en);
+
+	ret = hclge_cmd_send(&hdev->hw, &desc, 1);
+	if (ret) {
+		dev_err(&hdev->pdev->dev,
+			"mac enable fail, ret =%d.\n", ret);
+		return;
+	}
+
+	if (!enable)
+		hclge_mac_link_status_wait(hdev, HCLGE_LINK_STATUS_DOWN,
+					   HCLGE_LINK_STATUS_WAIT_CNT);
+}
+
+static int hclge_config_switch_param(struct hclge_dev *hdev, int vfid,
+				     u8 switch_param, u8 param_mask)
+{
+	struct hclge_mac_vlan_switch_cmd *req;
+	struct hclge_desc desc;
+	u32 func_id;
+	int ret;
+
+	func_id = hclge_get_port_number(HOST_PORT, 0, vfid, 0);
+	req = (struct hclge_mac_vlan_switch_cmd *)desc.data;
+
+	/* read current config parameter */
+	hclge_cmd_setup_basic_desc(&desc, HCLGE_OPC_MAC_VLAN_SWITCH_PARAM,
+				   true);
+	req->roce_sel = HCLGE_MAC_VLAN_NIC_SEL;
+	req->func_id = cpu_to_le32(func_id);
+
+	ret = hclge_cmd_send(&hdev->hw, &desc, 1);
+	if (ret) {
+		dev_err(&hdev->pdev->dev,
+			"read mac vlan switch parameter fail, ret = %d\n", ret);
+		return ret;
+	}
+
+	/* modify and write new config parameter */
+	hclge_comm_cmd_reuse_desc(&desc, false);
+	req->switch_param = (req->switch_param & param_mask) | switch_param;
+	req->param_mask = param_mask;
+
+	ret = hclge_cmd_send(&hdev->hw, &desc, 1);
+	if (ret)
+		dev_err(&hdev->pdev->dev,
+			"set mac vlan switch parameter fail, ret = %d\n", ret);
+	return ret;
+}
+
+static void hclge_phy_link_status_wait(struct hclge_dev *hdev,
+				       int link_ret)
+{
+#define HCLGE_PHY_LINK_STATUS_NUM  200
+
+	struct phy_device *phydev = hdev->hw.mac.phydev;
+	int i = 0;
+	int ret;
+
+	do {
+		ret = phy_read_status(phydev);
+		if (ret) {
+			dev_err(&hdev->pdev->dev,
+				"phy update link status fail, ret = %d\n", ret);
+			return;
+		}
+
+		if (phydev->link == link_ret)
+			break;
+
+		msleep(HCLGE_LINK_STATUS_MS);
+	} while (++i < HCLGE_PHY_LINK_STATUS_NUM);
+}
+
+static int hclge_mac_link_status_wait(struct hclge_dev *hdev, int link_ret,
+				      int wait_cnt)
+{
+	int link_status;
+	int i = 0;
+	int ret;
+
+	do {
+		ret = hclge_get_mac_link_status(hdev, &link_status);
+		if (ret)
+			return ret;
+		if (link_status == link_ret)
+			return 0;
+
+		msleep(HCLGE_LINK_STATUS_MS);
+	} while (++i < wait_cnt);
+	return -EBUSY;
+}
+
+static int hclge_mac_phy_link_status_wait(struct hclge_dev *hdev, bool en,
+					  bool is_phy)
+{
+#define HCLGE_MAC_LINK_STATUS_NUM  100
+
+	int link_ret;
+
+	link_ret = en ? HCLGE_LINK_STATUS_UP : HCLGE_LINK_STATUS_DOWN;
+
+	if (is_phy)
+		hclge_phy_link_status_wait(hdev, link_ret);
+
+	return hclge_mac_link_status_wait(hdev, link_ret,
+					  HCLGE_MAC_LINK_STATUS_NUM);
+}
+
+static int hclge_set_app_loopback(struct hclge_dev *hdev, bool en)
+{
+	struct hclge_config_mac_mode_cmd *req;
+	struct hclge_desc desc;
+	u32 loop_en;
+	int ret;
+
+	req = (struct hclge_config_mac_mode_cmd *)&desc.data[0];
+	/* 1 Read out the MAC mode config at first */
+	hclge_cmd_setup_basic_desc(&desc, HCLGE_OPC_CONFIG_MAC_MODE, true);
+	ret = hclge_cmd_send(&hdev->hw, &desc, 1);
+	if (ret) {
+		dev_err(&hdev->pdev->dev,
+			"mac loopback get fail, ret =%d.\n", ret);
+		return ret;
+	}
+
+	/* 2 Then setup the loopback flag */
+	loop_en = le32_to_cpu(req->txrx_pad_fcs_loop_en);
+	hnae3_set_bit(loop_en, HCLGE_MAC_APP_LP_B, en ? 1 : 0);
+
+	req->txrx_pad_fcs_loop_en = cpu_to_le32(loop_en);
+
+	/* 3 Config mac work mode with loopback flag
+	 * and its original configure parameters
+	 */
+	hclge_comm_cmd_reuse_desc(&desc, false);
+	ret = hclge_cmd_send(&hdev->hw, &desc, 1);
+	if (ret)
+		dev_err(&hdev->pdev->dev,
+			"mac loopback set fail, ret =%d.\n", ret);
+	return ret;
+}
+
+static int hclge_cfg_common_loopback_cmd_send(struct hclge_dev *hdev, bool en,
+					      enum hnae3_loop loop_mode)
+{
+	struct hclge_common_lb_cmd *req;
+	struct hclge_desc desc;
+	u8 loop_mode_b;
+	int ret;
+
+	req = (struct hclge_common_lb_cmd *)desc.data;
+	hclge_cmd_setup_basic_desc(&desc, HCLGE_OPC_COMMON_LOOPBACK, false);
+
+	switch (loop_mode) {
+	case HNAE3_LOOP_SERIAL_SERDES:
+		loop_mode_b = HCLGE_CMD_SERDES_SERIAL_INNER_LOOP_B;
+		break;
+	case HNAE3_LOOP_PARALLEL_SERDES:
+		loop_mode_b = HCLGE_CMD_SERDES_PARALLEL_INNER_LOOP_B;
+		break;
+	case HNAE3_LOOP_PHY:
+		loop_mode_b = HCLGE_CMD_GE_PHY_INNER_LOOP_B;
+		break;
+	default:
+		dev_err(&hdev->pdev->dev,
+			"unsupported loopback mode %d\n", loop_mode);
+		return -ENOTSUPP;
+	}
+
+	req->mask = loop_mode_b;
+	if (en)
+		req->enable = loop_mode_b;
+
+	ret = hclge_cmd_send(&hdev->hw, &desc, 1);
+	if (ret)
+		dev_err(&hdev->pdev->dev,
+			"failed to send loopback cmd, loop_mode = %d, ret = %d\n",
+			loop_mode, ret);
+
+	return ret;
+}
+
+static int hclge_cfg_common_loopback_wait(struct hclge_dev *hdev)
+{
+#define HCLGE_COMMON_LB_RETRY_MS	10
+#define HCLGE_COMMON_LB_RETRY_NUM	100
+
+	struct hclge_common_lb_cmd *req;
+	struct hclge_desc desc;
+	u32 i = 0;
+	int ret;
+
+	req = (struct hclge_common_lb_cmd *)desc.data;
+
+	do {
+		msleep(HCLGE_COMMON_LB_RETRY_MS);
+		hclge_cmd_setup_basic_desc(&desc, HCLGE_OPC_COMMON_LOOPBACK,
+					   true);
+		ret = hclge_cmd_send(&hdev->hw, &desc, 1);
+		if (ret) {
+			dev_err(&hdev->pdev->dev,
+				"failed to get loopback done status, ret = %d\n",
+				ret);
+			return ret;
+		}
+	} while (++i < HCLGE_COMMON_LB_RETRY_NUM &&
+		 !(req->result & HCLGE_CMD_COMMON_LB_DONE_B));
+
+	if (!(req->result & HCLGE_CMD_COMMON_LB_DONE_B)) {
+		dev_err(&hdev->pdev->dev, "wait loopback timeout\n");
+		return -EBUSY;
+	} else if (!(req->result & HCLGE_CMD_COMMON_LB_SUCCESS_B)) {
+		dev_err(&hdev->pdev->dev, "failed to do loopback test\n");
+		return -EIO;
+	}
+
+	return 0;
+}
+
+static int hclge_cfg_common_loopback(struct hclge_dev *hdev, bool en,
+				     enum hnae3_loop loop_mode)
+{
+	int ret;
+
+	ret = hclge_cfg_common_loopback_cmd_send(hdev, en, loop_mode);
+	if (ret)
+		return ret;
+
+	return hclge_cfg_common_loopback_wait(hdev);
+}
+
+static int hclge_set_common_loopback(struct hclge_dev *hdev, bool en,
+				     enum hnae3_loop loop_mode)
+{
+	int ret;
+
+	ret = hclge_cfg_common_loopback(hdev, en, loop_mode);
+	if (ret)
+		return ret;
+
+	hclge_cfg_mac_mode(hdev, en);
+
+	ret = hclge_mac_phy_link_status_wait(hdev, en, false);
+	if (ret)
+		dev_err(&hdev->pdev->dev,
+			"serdes loopback config mac mode timeout\n");
+
+	return ret;
+}
+
+static int hclge_enable_phy_loopback(struct hclge_dev *hdev,
+				     struct phy_device *phydev)
+{
+	int ret;
+
+	if (!phydev->suspended) {
+		ret = phy_suspend(phydev);
+		if (ret)
+			return ret;
+	}
+
+	ret = phy_resume(phydev);
+	if (ret)
+		return ret;
+
+	return phy_loopback(phydev, true, 0);
+}
+
+static int hclge_disable_phy_loopback(struct hclge_dev *hdev,
+				      struct phy_device *phydev)
+{
+	int ret;
+
+	ret = phy_loopback(phydev, false, 0);
+	if (ret)
+		return ret;
+
+	return phy_suspend(phydev);
+}
+
+static int hclge_set_phy_loopback(struct hclge_dev *hdev, bool en)
+{
+	struct phy_device *phydev = hdev->hw.mac.phydev;
+	int ret;
+
+	if (!phydev) {
+		if (hnae3_dev_phy_imp_supported(hdev))
+			return hclge_set_common_loopback(hdev, en,
+							 HNAE3_LOOP_PHY);
+		return -ENOTSUPP;
+	}
+
+	if (en)
+		ret = hclge_enable_phy_loopback(hdev, phydev);
+	else
+		ret = hclge_disable_phy_loopback(hdev, phydev);
+	if (ret) {
+		dev_err(&hdev->pdev->dev,
+			"set phy loopback fail, ret = %d\n", ret);
+		return ret;
+	}
+
+	hclge_cfg_mac_mode(hdev, en);
+
+	ret = hclge_mac_phy_link_status_wait(hdev, en, true);
+	if (ret)
+		dev_err(&hdev->pdev->dev,
+			"phy loopback config mac mode timeout\n");
+
+	return ret;
+}
+
+static int hclge_tqp_enable_cmd_send(struct hclge_dev *hdev, u16 tqp_id,
+				     u16 stream_id, bool enable)
+{
+	struct hclge_desc desc;
+	struct hclge_cfg_com_tqp_queue_cmd *req =
+		(struct hclge_cfg_com_tqp_queue_cmd *)desc.data;
+
+	hclge_cmd_setup_basic_desc(&desc, HCLGE_OPC_CFG_COM_TQP_QUEUE, false);
+	req->tqp_id = cpu_to_le16(tqp_id);
+	req->stream_id = cpu_to_le16(stream_id);
+	if (enable)
+		req->enable |= 1U << HCLGE_TQP_ENABLE_B;
+
+	return hclge_cmd_send(&hdev->hw, &desc, 1);
+}
+
+static int hclge_tqp_enable(struct hnae3_handle *handle, bool enable)
+{
+	struct hclge_vport *vport = hclge_get_vport(handle);
+	struct hclge_dev *hdev = vport->back;
+	int ret;
+	u16 i;
+
+	for (i = 0; i < handle->kinfo.num_tqps; i++) {
+		ret = hclge_tqp_enable_cmd_send(hdev, i, 0, enable);
+		if (ret)
+			return ret;
+	}
+	return 0;
+}
+
+static int hclge_set_loopback(struct hnae3_handle *handle,
+			      enum hnae3_loop loop_mode, bool en)
+{
+	struct hclge_vport *vport = hclge_get_vport(handle);
+	struct hclge_dev *hdev = vport->back;
+	int ret = 0;
+
+	/* Loopback can be enabled in three places: SSU, MAC, and serdes. By
+	 * default, SSU loopback is enabled, so if the SMAC and the DMAC are
+	 * the same, the packets are looped back in the SSU. If SSU loopback
+	 * is disabled, packets can reach MAC even if SMAC is the same as DMAC.
+	 */
+	if (hdev->ae_dev->dev_version >= HNAE3_DEVICE_VERSION_V2) {
+		u8 switch_param = en ? 0 : BIT(HCLGE_SWITCH_ALW_LPBK_B);
+
+		ret = hclge_config_switch_param(hdev, PF_VPORT_ID, switch_param,
+						HCLGE_SWITCH_ALW_LPBK_MASK);
+		if (ret)
+			return ret;
+	}
+
+	switch (loop_mode) {
+	case HNAE3_LOOP_APP:
+		ret = hclge_set_app_loopback(hdev, en);
+		break;
+	case HNAE3_LOOP_SERIAL_SERDES:
+	case HNAE3_LOOP_PARALLEL_SERDES:
+		ret = hclge_set_common_loopback(hdev, en, loop_mode);
+		break;
+	case HNAE3_LOOP_PHY:
+		ret = hclge_set_phy_loopback(hdev, en);
+		break;
+	case HNAE3_LOOP_EXTERNAL:
+		break;
+	default:
+		ret = -ENOTSUPP;
+		dev_err(&hdev->pdev->dev,
+			"loop_mode %d is not supported\n", loop_mode);
+		break;
+	}
+
+	if (ret)
+		return ret;
+
+	ret = hclge_tqp_enable(handle, en);
+	if (ret)
+		dev_err(&hdev->pdev->dev, "failed to %s tqp in loopback, ret = %d\n",
+			str_enable_disable(en), ret);
+
+	return ret;
+}
+
+static int hclge_set_default_loopback(struct hclge_dev *hdev)
+{
+	int ret;
+
+	ret = hclge_set_app_loopback(hdev, false);
+	if (ret)
+		return ret;
+
+	ret = hclge_cfg_common_loopback(hdev, false, HNAE3_LOOP_SERIAL_SERDES);
+	if (ret)
+		return ret;
+
+	return hclge_cfg_common_loopback(hdev, false,
+					 HNAE3_LOOP_PARALLEL_SERDES);
+}
+
+static void hclge_flush_link_update(struct hclge_dev *hdev)
+{
+#define HCLGE_FLUSH_LINK_TIMEOUT	100000
+
+	unsigned long last = hdev->serv_processed_cnt;
+	int i = 0;
+
+	while (test_bit(HCLGE_STATE_LINK_UPDATING, &hdev->state) &&
+	       i++ < HCLGE_FLUSH_LINK_TIMEOUT &&
+	       last == hdev->serv_processed_cnt)
+		usleep_range(1, 1);
+}
+
+static void hclge_set_timer_task(struct hnae3_handle *handle, bool enable)
+{
+	struct hclge_vport *vport = hclge_get_vport(handle);
+	struct hclge_dev *hdev = vport->back;
+
+	if (enable) {
+		hclge_task_schedule(hdev, 0);
+	} else {
+		/* Set the DOWN flag here to disable link updating */
+		set_bit(HCLGE_STATE_DOWN, &hdev->state);
+
+		smp_mb__after_atomic(); /* flush memory to make sure DOWN is seen by service task */
+		hclge_flush_link_update(hdev);
+	}
+}
+
+static int hclge_ae_start(struct hnae3_handle *handle)
+{
+	struct hclge_vport *vport = hclge_get_vport(handle);
+	struct hclge_dev *hdev = vport->back;
+
+	/* mac enable */
+	hclge_cfg_mac_mode(hdev, true);
+	clear_bit(HCLGE_STATE_DOWN, &hdev->state);
+	hdev->hw.mac.link = 0;
+
+	/* reset tqp stats */
+	hclge_comm_reset_tqp_stats(handle);
+
+	hclge_mac_start_phy(hdev);
+
+	return 0;
+}
+
+static void hclge_ae_stop(struct hnae3_handle *handle)
+{
+	struct hclge_vport *vport = hclge_get_vport(handle);
+	struct hclge_dev *hdev = vport->back;
+
+	set_bit(HCLGE_STATE_DOWN, &hdev->state);
+	spin_lock_bh(&hdev->fd_rule_lock);
+	hclge_clear_arfs_rules(hdev);
+	spin_unlock_bh(&hdev->fd_rule_lock);
+
+	/* If it is not PF reset or FLR, the firmware will disable the MAC,
+	 * so it only need to stop phy here.
+	 */
+	if (test_bit(HCLGE_STATE_RST_HANDLING, &hdev->state)) {
+		hclge_pfc_pause_en_cfg(hdev, HCLGE_PFC_TX_RX_DISABLE,
+				       HCLGE_PFC_DISABLE);
+		if (hdev->reset_type != HNAE3_FUNC_RESET &&
+		    hdev->reset_type != HNAE3_FLR_RESET) {
+			hclge_mac_stop_phy(hdev);
+			hclge_update_link_status(hdev);
+			return;
+		}
+	}
+
+	hclge_reset_tqp(handle);
+
+	hclge_config_mac_tnl_int(hdev, false);
+
+	/* Mac disable */
+	hclge_cfg_mac_mode(hdev, false);
+
+	hclge_mac_stop_phy(hdev);
+
+	/* reset tqp stats */
+	hclge_comm_reset_tqp_stats(handle);
+	hclge_update_link_status(hdev);
+}
+
+int hclge_vport_start(struct hclge_vport *vport)
+{
+	struct hclge_dev *hdev = vport->back;
+
+	set_bit(HCLGE_VPORT_STATE_INITED, &vport->state);
+	set_bit(HCLGE_VPORT_STATE_ALIVE, &vport->state);
+	set_bit(HCLGE_VPORT_STATE_PROMISC_CHANGE, &vport->state);
+	vport->last_active_jiffies = jiffies;
+	vport->need_notify = 0;
+
+	if (test_bit(vport->vport_id, hdev->vport_config_block)) {
+		if (vport->vport_id) {
+			hclge_restore_mac_table_common(vport);
+			hclge_restore_vport_vlan_table(vport);
+		} else {
+			hclge_restore_hw_table(hdev);
+		}
+	}
+
+	clear_bit(vport->vport_id, hdev->vport_config_block);
+
+	return 0;
+}
+
+void hclge_vport_stop(struct hclge_vport *vport)
+{
+	clear_bit(HCLGE_VPORT_STATE_INITED, &vport->state);
+	clear_bit(HCLGE_VPORT_STATE_ALIVE, &vport->state);
+	vport->need_notify = 0;
+}
+
+static int hclge_client_start(struct hnae3_handle *handle)
+{
+	struct hclge_vport *vport = hclge_get_vport(handle);
+
+	return hclge_vport_start(vport);
+}
+
+static void hclge_client_stop(struct hnae3_handle *handle)
+{
+	struct hclge_vport *vport = hclge_get_vport(handle);
+
+	hclge_vport_stop(vport);
+}
+
+static int hclge_get_mac_vlan_cmd_status(struct hclge_vport *vport,
+					 u16 cmdq_resp, u8  resp_code,
+					 enum hclge_mac_vlan_tbl_opcode op)
+{
+	struct hclge_dev *hdev = vport->back;
+
+	if (cmdq_resp) {
+		dev_err(&hdev->pdev->dev,
+			"cmdq execute failed for get_mac_vlan_cmd_status,status=%u.\n",
+			cmdq_resp);
+		return -EIO;
+	}
+
+	if (op == HCLGE_MAC_VLAN_ADD) {
+		if (!resp_code || resp_code == 1)
+			return 0;
+		else if (resp_code == HCLGE_ADD_UC_OVERFLOW ||
+			 resp_code == HCLGE_ADD_MC_OVERFLOW)
+			return -ENOSPC;
+
+		dev_err(&hdev->pdev->dev,
+			"add mac addr failed for undefined, code=%u.\n",
+			resp_code);
+		return -EIO;
+	} else if (op == HCLGE_MAC_VLAN_REMOVE) {
+		if (!resp_code) {
+			return 0;
+		} else if (resp_code == 1) {
+			dev_dbg(&hdev->pdev->dev,
+				"remove mac addr failed for miss.\n");
+			return -ENOENT;
+		}
+
+		dev_err(&hdev->pdev->dev,
+			"remove mac addr failed for undefined, code=%u.\n",
+			resp_code);
+		return -EIO;
+	} else if (op == HCLGE_MAC_VLAN_LKUP) {
+		if (!resp_code) {
+			return 0;
+		} else if (resp_code == 1) {
+			dev_dbg(&hdev->pdev->dev,
+				"lookup mac addr failed for miss.\n");
+			return -ENOENT;
+		}
+
+		dev_err(&hdev->pdev->dev,
+			"lookup mac addr failed for undefined, code=%u.\n",
+			resp_code);
+		return -EIO;
+	}
+
+	dev_err(&hdev->pdev->dev,
+		"unknown opcode for get_mac_vlan_cmd_status, opcode=%d.\n", op);
+
+	return -EINVAL;
+}
+
+static int hclge_update_desc_vfid(struct hclge_desc *desc, int vfid, bool clr)
+{
+#define HCLGE_VF_NUM_IN_FIRST_DESC 192
+
+	unsigned int word_num;
+	unsigned int bit_num;
+
+	if (vfid > 255 || vfid < 0)
+		return -EIO;
+
+	if (vfid >= 0 && vfid < HCLGE_VF_NUM_IN_FIRST_DESC) {
+		word_num = vfid / 32;
+		bit_num  = vfid % 32;
+		if (clr)
+			desc[1].data[word_num] &= cpu_to_le32(~(1U << bit_num));
+		else
+			desc[1].data[word_num] |= cpu_to_le32(1 << bit_num);
+	} else {
+		word_num = (vfid - HCLGE_VF_NUM_IN_FIRST_DESC) / 32;
+		bit_num  = vfid % 32;
+		if (clr)
+			desc[2].data[word_num] &= cpu_to_le32(~(1U << bit_num));
+		else
+			desc[2].data[word_num] |= cpu_to_le32(1 << bit_num);
+	}
+
+	return 0;
+}
+
+static bool hclge_is_all_function_id_zero(struct hclge_desc *desc)
+{
+#define HCLGE_DESC_NUMBER 3
+#define HCLGE_FUNC_NUMBER_PER_DESC 6
+	int i, j;
+
+	for (i = 1; i < HCLGE_DESC_NUMBER; i++)
+		for (j = 0; j < HCLGE_FUNC_NUMBER_PER_DESC; j++)
+			if (desc[i].data[j])
+				return false;
+
+	return true;
+}
+
+static void hclge_prepare_mac_addr(struct hclge_mac_vlan_tbl_entry_cmd *new_req,
+				   const u8 *addr, bool is_mc)
+{
+	const unsigned char *mac_addr = addr;
+	u32 high_val = mac_addr[2] << 16 | (mac_addr[3] << 24) |
+		       (mac_addr[0]) | (mac_addr[1] << 8);
+	u32 low_val  = mac_addr[4] | (mac_addr[5] << 8);
+
+	hnae3_set_bit(new_req->flags, HCLGE_MAC_VLAN_BIT0_EN_B, 1);
+	if (is_mc) {
+		hnae3_set_bit(new_req->entry_type, HCLGE_MAC_VLAN_BIT1_EN_B, 1);
+		hnae3_set_bit(new_req->mc_mac_en, HCLGE_MAC_VLAN_BIT0_EN_B, 1);
+	}
+
+	new_req->mac_addr_hi32 = cpu_to_le32(high_val);
+	new_req->mac_addr_lo16 = cpu_to_le16(low_val & 0xffff);
+}
+
+static int hclge_remove_mac_vlan_tbl(struct hclge_vport *vport,
+				     struct hclge_mac_vlan_tbl_entry_cmd *req)
+{
+	struct hclge_dev *hdev = vport->back;
+	struct hclge_desc desc;
+	u8 resp_code;
+	u16 retval;
+	int ret;
+
+	hclge_cmd_setup_basic_desc(&desc, HCLGE_OPC_MAC_VLAN_REMOVE, false);
+
+	memcpy(desc.data, req, sizeof(struct hclge_mac_vlan_tbl_entry_cmd));
+
+	ret = hclge_cmd_send(&hdev->hw, &desc, 1);
+	if (ret) {
+		dev_err(&hdev->pdev->dev,
+			"del mac addr failed for cmd_send, ret =%d.\n",
+			ret);
+		return ret;
+	}
+	resp_code = (le32_to_cpu(desc.data[0]) >> 8) & 0xff;
+	retval = le16_to_cpu(desc.retval);
+
+	return hclge_get_mac_vlan_cmd_status(vport, retval, resp_code,
+					     HCLGE_MAC_VLAN_REMOVE);
+}
+
+static int hclge_lookup_mac_vlan_tbl(struct hclge_vport *vport,
+				     struct hclge_mac_vlan_tbl_entry_cmd *req,
+				     struct hclge_desc *desc,
+				     bool is_mc)
+{
+	struct hclge_dev *hdev = vport->back;
+	u8 resp_code;
+	u16 retval;
+	int ret;
+
+	hclge_cmd_setup_basic_desc(&desc[0], HCLGE_OPC_MAC_VLAN_ADD, true);
+	if (is_mc) {
+		desc[0].flag |= cpu_to_le16(HCLGE_COMM_CMD_FLAG_NEXT);
+		memcpy(desc[0].data,
+		       req,
+		       sizeof(struct hclge_mac_vlan_tbl_entry_cmd));
+		hclge_cmd_setup_basic_desc(&desc[1],
+					   HCLGE_OPC_MAC_VLAN_ADD,
+					   true);
+		desc[1].flag |= cpu_to_le16(HCLGE_COMM_CMD_FLAG_NEXT);
+		hclge_cmd_setup_basic_desc(&desc[2],
+					   HCLGE_OPC_MAC_VLAN_ADD,
+					   true);
+		ret = hclge_cmd_send(&hdev->hw, desc, 3);
+	} else {
+		memcpy(desc[0].data,
+		       req,
+		       sizeof(struct hclge_mac_vlan_tbl_entry_cmd));
+		ret = hclge_cmd_send(&hdev->hw, desc, 1);
+	}
+	if (ret) {
+		dev_err(&hdev->pdev->dev,
+			"lookup mac addr failed for cmd_send, ret =%d.\n",
+			ret);
+		return ret;
+	}
+	resp_code = (le32_to_cpu(desc[0].data[0]) >> 8) & 0xff;
+	retval = le16_to_cpu(desc[0].retval);
+
+	return hclge_get_mac_vlan_cmd_status(vport, retval, resp_code,
+					     HCLGE_MAC_VLAN_LKUP);
+}
+
+static int hclge_add_mac_vlan_tbl(struct hclge_vport *vport,
+				  struct hclge_mac_vlan_tbl_entry_cmd *req,
+				  struct hclge_desc *mc_desc)
+{
+	struct hclge_dev *hdev = vport->back;
+	int cfg_status;
+	u8 resp_code;
+	u16 retval;
+	int ret;
+
+	if (!mc_desc) {
+		struct hclge_desc desc;
+
+		hclge_cmd_setup_basic_desc(&desc,
+					   HCLGE_OPC_MAC_VLAN_ADD,
+					   false);
+		memcpy(desc.data, req,
+		       sizeof(struct hclge_mac_vlan_tbl_entry_cmd));
+		ret = hclge_cmd_send(&hdev->hw, &desc, 1);
+		resp_code = (le32_to_cpu(desc.data[0]) >> 8) & 0xff;
+		retval = le16_to_cpu(desc.retval);
+
+		cfg_status = hclge_get_mac_vlan_cmd_status(vport, retval,
+							   resp_code,
+							   HCLGE_MAC_VLAN_ADD);
+	} else {
+		hclge_comm_cmd_reuse_desc(&mc_desc[0], false);
+		mc_desc[0].flag |= cpu_to_le16(HCLGE_COMM_CMD_FLAG_NEXT);
+		hclge_comm_cmd_reuse_desc(&mc_desc[1], false);
+		mc_desc[1].flag |= cpu_to_le16(HCLGE_COMM_CMD_FLAG_NEXT);
+		hclge_comm_cmd_reuse_desc(&mc_desc[2], false);
+		mc_desc[2].flag &= cpu_to_le16(~HCLGE_COMM_CMD_FLAG_NEXT);
+		memcpy(mc_desc[0].data, req,
+		       sizeof(struct hclge_mac_vlan_tbl_entry_cmd));
+		ret = hclge_cmd_send(&hdev->hw, mc_desc, 3);
+		resp_code = (le32_to_cpu(mc_desc[0].data[0]) >> 8) & 0xff;
+		retval = le16_to_cpu(mc_desc[0].retval);
+
+		cfg_status = hclge_get_mac_vlan_cmd_status(vport, retval,
+							   resp_code,
+							   HCLGE_MAC_VLAN_ADD);
+	}
+
+	if (ret) {
+		dev_err(&hdev->pdev->dev,
+			"add mac addr failed for cmd_send, ret =%d.\n",
+			ret);
+		return ret;
+	}
+
+	return cfg_status;
+}
+
+static int hclge_set_umv_space(struct hclge_dev *hdev, u16 space_size,
+			       u16 *allocated_size)
+{
+	struct hclge_umv_spc_alc_cmd *req;
+	struct hclge_desc desc;
+	int ret;
+
+	req = (struct hclge_umv_spc_alc_cmd *)desc.data;
+	hclge_cmd_setup_basic_desc(&desc, HCLGE_OPC_MAC_VLAN_ALLOCATE, false);
+
+	req->space_size = cpu_to_le32(space_size);
+
+	ret = hclge_cmd_send(&hdev->hw, &desc, 1);
+	if (ret) {
+		dev_err(&hdev->pdev->dev, "failed to set umv space, ret = %d\n",
+			ret);
+		return ret;
+	}
+
+	*allocated_size = le32_to_cpu(desc.data[1]);
+
+	return 0;
+}
+
+static int hclge_init_umv_space(struct hclge_dev *hdev)
+{
+	u16 allocated_size = 0;
+	int ret;
+
+	ret = hclge_set_umv_space(hdev, hdev->wanted_umv_size, &allocated_size);
+	if (ret)
+		return ret;
+
+	if (allocated_size < hdev->wanted_umv_size)
+		dev_warn(&hdev->pdev->dev,
+			 "failed to alloc umv space, want %u, get %u\n",
+			 hdev->wanted_umv_size, allocated_size);
+
+	hdev->max_umv_size = allocated_size;
+	hdev->priv_umv_size = hdev->max_umv_size / (hdev->num_alloc_vport + 1);
+	hdev->share_umv_size = hdev->priv_umv_size +
+			hdev->max_umv_size % (hdev->num_alloc_vport + 1);
+
+	if (hdev->ae_dev->dev_specs.mc_mac_size)
+		set_bit(HNAE3_DEV_SUPPORT_MC_MAC_MNG_B, hdev->ae_dev->caps);
+
+	return 0;
+}
+
+static void hclge_reset_umv_space(struct hclge_dev *hdev)
+{
+	struct hclge_vport *vport;
+	int i;
+
+	for (i = 0; i < hdev->num_alloc_vport; i++) {
+		vport = &hdev->vport[i];
+		vport->used_umv_num = 0;
+	}
+
+	mutex_lock(&hdev->vport_lock);
+	hdev->share_umv_size = hdev->priv_umv_size +
+			hdev->max_umv_size % (hdev->num_alloc_vport + 1);
+	mutex_unlock(&hdev->vport_lock);
+
+	hdev->used_mc_mac_num = 0;
+}
+
+static bool hclge_is_umv_space_full(struct hclge_vport *vport, bool need_lock)
+{
+	struct hclge_dev *hdev = vport->back;
+	bool is_full;
+
+	if (need_lock)
+		mutex_lock(&hdev->vport_lock);
+
+	is_full = (vport->used_umv_num >= hdev->priv_umv_size &&
+		   hdev->share_umv_size == 0);
+
+	if (need_lock)
+		mutex_unlock(&hdev->vport_lock);
+
+	return is_full;
+}
+
+static void hclge_update_umv_space(struct hclge_vport *vport, bool is_free)
+{
+	struct hclge_dev *hdev = vport->back;
+
+	if (is_free) {
+		if (vport->used_umv_num > hdev->priv_umv_size)
+			hdev->share_umv_size++;
+
+		if (vport->used_umv_num > 0)
+			vport->used_umv_num--;
+	} else {
+		if (vport->used_umv_num >= hdev->priv_umv_size &&
+		    hdev->share_umv_size > 0)
+			hdev->share_umv_size--;
+		vport->used_umv_num++;
+	}
+}
+
+static struct hclge_mac_node *hclge_find_mac_node(struct list_head *list,
+						  const u8 *mac_addr)
+{
+	struct hclge_mac_node *mac_node, *tmp;
+
+	list_for_each_entry_safe(mac_node, tmp, list, node)
+		if (ether_addr_equal(mac_addr, mac_node->mac_addr))
+			return mac_node;
+
+	return NULL;
+}
+
+static void hclge_update_mac_node(struct hclge_mac_node *mac_node,
+				  enum HCLGE_MAC_NODE_STATE state)
+{
+	switch (state) {
+	/* from set_rx_mode or tmp_add_list */
+	case HCLGE_MAC_TO_ADD:
+		if (mac_node->state == HCLGE_MAC_TO_DEL)
+			mac_node->state = HCLGE_MAC_ACTIVE;
+		break;
+	/* only from set_rx_mode */
+	case HCLGE_MAC_TO_DEL:
+		if (mac_node->state == HCLGE_MAC_TO_ADD) {
+			list_del(&mac_node->node);
+			kfree(mac_node);
+		} else {
+			mac_node->state = HCLGE_MAC_TO_DEL;
+		}
+		break;
+	/* only from tmp_add_list, the mac_node->state won't be
+	 * ACTIVE.
+	 */
+	case HCLGE_MAC_ACTIVE:
+		if (mac_node->state == HCLGE_MAC_TO_ADD)
+			mac_node->state = HCLGE_MAC_ACTIVE;
+
+		break;
+	}
+}
+
+int hclge_update_mac_list(struct hclge_vport *vport,
+			  enum HCLGE_MAC_NODE_STATE state,
+			  enum HCLGE_MAC_ADDR_TYPE mac_type,
+			  const unsigned char *addr)
+{
+	char format_mac_addr[HNAE3_FORMAT_MAC_ADDR_LEN];
+	struct hclge_dev *hdev = vport->back;
+	struct hclge_mac_node *mac_node;
+	struct list_head *list;
+
+	list = (mac_type == HCLGE_MAC_ADDR_UC) ?
+		&vport->uc_mac_list : &vport->mc_mac_list;
+
+	spin_lock_bh(&vport->mac_list_lock);
+
+	/* if the mac addr is already in the mac list, no need to add a new
+	 * one into it, just check the mac addr state, convert it to a new
+	 * state, or just remove it, or do nothing.
+	 */
+	mac_node = hclge_find_mac_node(list, addr);
+	if (mac_node) {
+		hclge_update_mac_node(mac_node, state);
+		spin_unlock_bh(&vport->mac_list_lock);
+		set_bit(HCLGE_VPORT_STATE_MAC_TBL_CHANGE, &vport->state);
+		return 0;
+	}
+
+	/* if this address is never added, unnecessary to delete */
+	if (state == HCLGE_MAC_TO_DEL) {
+		spin_unlock_bh(&vport->mac_list_lock);
+		hnae3_format_mac_addr(format_mac_addr, addr);
+		dev_err(&hdev->pdev->dev,
+			"failed to delete address %s from mac list\n",
+			format_mac_addr);
+		return -ENOENT;
+	}
+
+	mac_node = kzalloc(sizeof(*mac_node), GFP_ATOMIC);
+	if (!mac_node) {
+		spin_unlock_bh(&vport->mac_list_lock);
+		return -ENOMEM;
+	}
+
+	set_bit(HCLGE_VPORT_STATE_MAC_TBL_CHANGE, &vport->state);
+
+	mac_node->state = state;
+	ether_addr_copy(mac_node->mac_addr, addr);
+	list_add_tail(&mac_node->node, list);
+
+	spin_unlock_bh(&vport->mac_list_lock);
+
+	return 0;
+}
+
+static int hclge_add_uc_addr(struct hnae3_handle *handle,
+			     const unsigned char *addr)
+{
+	struct hclge_vport *vport = hclge_get_vport(handle);
+
+	return hclge_update_mac_list(vport, HCLGE_MAC_TO_ADD, HCLGE_MAC_ADDR_UC,
+				     addr);
+}
+
+int hclge_add_uc_addr_common(struct hclge_vport *vport,
+			     const unsigned char *addr)
+{
+	char format_mac_addr[HNAE3_FORMAT_MAC_ADDR_LEN];
+	struct hclge_dev *hdev = vport->back;
+	struct hclge_mac_vlan_tbl_entry_cmd req;
+	struct hclge_desc desc;
+	u16 egress_port = 0;
+	int ret;
+
+	/* mac addr check */
+	if (is_zero_ether_addr(addr) ||
+	    is_broadcast_ether_addr(addr) ||
+	    is_multicast_ether_addr(addr)) {
+		hnae3_format_mac_addr(format_mac_addr, addr);
+		dev_err(&hdev->pdev->dev,
+			"Set_uc mac err! invalid mac:%s. is_zero:%d,is_br=%d,is_mul=%d\n",
+			 format_mac_addr, is_zero_ether_addr(addr),
+			 is_broadcast_ether_addr(addr),
+			 is_multicast_ether_addr(addr));
+		return -EINVAL;
+	}
+
+	memset(&req, 0, sizeof(req));
+
+	hnae3_set_field(egress_port, HCLGE_MAC_EPORT_VFID_M,
+			HCLGE_MAC_EPORT_VFID_S, vport->vport_id);
+
+	req.egress_port = cpu_to_le16(egress_port);
+
+	hclge_prepare_mac_addr(&req, addr, false);
+
+	/* Lookup the mac address in the mac_vlan table, and add
+	 * it if the entry is inexistent. Repeated unicast entry
+	 * is not allowed in the mac vlan table.
+	 */
+	ret = hclge_lookup_mac_vlan_tbl(vport, &req, &desc, false);
+	if (ret == -ENOENT) {
+		mutex_lock(&hdev->vport_lock);
+		if (!hclge_is_umv_space_full(vport, false)) {
+			ret = hclge_add_mac_vlan_tbl(vport, &req, NULL);
+			if (!ret)
+				hclge_update_umv_space(vport, false);
+			mutex_unlock(&hdev->vport_lock);
+			return ret;
+		}
+		mutex_unlock(&hdev->vport_lock);
+
+		if (!(vport->overflow_promisc_flags & HNAE3_OVERFLOW_UPE))
+			dev_err(&hdev->pdev->dev, "UC MAC table full(%u)\n",
+				hdev->priv_umv_size);
+
+		return -ENOSPC;
+	}
+
+	/* check if we just hit the duplicate */
+	if (!ret)
+		return -EEXIST;
+
+	return ret;
+}
+
+static int hclge_rm_uc_addr(struct hnae3_handle *handle,
+			    const unsigned char *addr)
+{
+	struct hclge_vport *vport = hclge_get_vport(handle);
+
+	return hclge_update_mac_list(vport, HCLGE_MAC_TO_DEL, HCLGE_MAC_ADDR_UC,
+				     addr);
+}
+
+int hclge_rm_uc_addr_common(struct hclge_vport *vport,
+			    const unsigned char *addr)
+{
+	char format_mac_addr[HNAE3_FORMAT_MAC_ADDR_LEN];
+	struct hclge_dev *hdev = vport->back;
+	struct hclge_mac_vlan_tbl_entry_cmd req;
+	int ret;
+
+	/* mac addr check */
+	if (is_zero_ether_addr(addr) ||
+	    is_broadcast_ether_addr(addr) ||
+	    is_multicast_ether_addr(addr)) {
+		hnae3_format_mac_addr(format_mac_addr, addr);
+		dev_dbg(&hdev->pdev->dev, "Remove mac err! invalid mac:%s.\n",
+			format_mac_addr);
+		return -EINVAL;
+	}
+
+	memset(&req, 0, sizeof(req));
+	hnae3_set_bit(req.entry_type, HCLGE_MAC_VLAN_BIT0_EN_B, 0);
+	hclge_prepare_mac_addr(&req, addr, false);
+	ret = hclge_remove_mac_vlan_tbl(vport, &req);
+	if (!ret || ret == -ENOENT) {
+		mutex_lock(&hdev->vport_lock);
+		hclge_update_umv_space(vport, true);
+		mutex_unlock(&hdev->vport_lock);
+		return 0;
+	}
+
+	return ret;
+}
+
+static int hclge_add_mc_addr(struct hnae3_handle *handle,
+			     const unsigned char *addr)
+{
+	struct hclge_vport *vport = hclge_get_vport(handle);
+
+	return hclge_update_mac_list(vport, HCLGE_MAC_TO_ADD, HCLGE_MAC_ADDR_MC,
+				     addr);
+}
+
+int hclge_add_mc_addr_common(struct hclge_vport *vport,
+			     const unsigned char *addr)
+{
+	char format_mac_addr[HNAE3_FORMAT_MAC_ADDR_LEN];
+	struct hclge_dev *hdev = vport->back;
+	struct hclge_mac_vlan_tbl_entry_cmd req;
+	struct hclge_desc desc[3];
+	bool is_new_addr = false;
+	int status;
+
+	/* mac addr check */
+	if (!is_multicast_ether_addr(addr)) {
+		hnae3_format_mac_addr(format_mac_addr, addr);
+		dev_err(&hdev->pdev->dev,
+			"Add mc mac err! invalid mac:%s.\n",
+			 format_mac_addr);
+		return -EINVAL;
+	}
+	memset(&req, 0, sizeof(req));
+	hclge_prepare_mac_addr(&req, addr, true);
+	status = hclge_lookup_mac_vlan_tbl(vport, &req, desc, true);
+	if (status) {
+		if (hnae3_ae_dev_mc_mac_mng_supported(hdev->ae_dev) &&
+		    hdev->used_mc_mac_num >=
+		    hdev->ae_dev->dev_specs.mc_mac_size)
+			goto err_no_space;
+
+		is_new_addr = true;
+
+		/* This mac addr do not exist, add new entry for it */
+		memset(desc[0].data, 0, sizeof(desc[0].data));
+		memset(desc[1].data, 0, sizeof(desc[0].data));
+		memset(desc[2].data, 0, sizeof(desc[0].data));
+	}
+	status = hclge_update_desc_vfid(desc, vport->vport_id, false);
+	if (status)
+		return status;
+	status = hclge_add_mac_vlan_tbl(vport, &req, desc);
+	if (status == -ENOSPC)
+		goto err_no_space;
+	else if (!status && is_new_addr)
+		hdev->used_mc_mac_num++;
+
+	return status;
+
+err_no_space:
+	/* if already overflow, not to print each time */
+	if (!(vport->overflow_promisc_flags & HNAE3_OVERFLOW_MPE)) {
+		vport->overflow_promisc_flags |= HNAE3_OVERFLOW_MPE;
+		dev_err(&hdev->pdev->dev, "mc mac vlan table is full\n");
+	}
+
+	return -ENOSPC;
+}
+
+static int hclge_rm_mc_addr(struct hnae3_handle *handle,
+			    const unsigned char *addr)
+{
+	struct hclge_vport *vport = hclge_get_vport(handle);
+
+	return hclge_update_mac_list(vport, HCLGE_MAC_TO_DEL, HCLGE_MAC_ADDR_MC,
+				     addr);
+}
+
+int hclge_rm_mc_addr_common(struct hclge_vport *vport,
+			    const unsigned char *addr)
+{
+	char format_mac_addr[HNAE3_FORMAT_MAC_ADDR_LEN];
+	struct hclge_dev *hdev = vport->back;
+	struct hclge_mac_vlan_tbl_entry_cmd req;
+	enum hclge_comm_cmd_status status;
+	struct hclge_desc desc[3];
+
+	/* mac addr check */
+	if (!is_multicast_ether_addr(addr)) {
+		hnae3_format_mac_addr(format_mac_addr, addr);
+		dev_dbg(&hdev->pdev->dev,
+			"Remove mc mac err! invalid mac:%s.\n",
+			 format_mac_addr);
+		return -EINVAL;
+	}
+
+	memset(&req, 0, sizeof(req));
+	hclge_prepare_mac_addr(&req, addr, true);
+	status = hclge_lookup_mac_vlan_tbl(vport, &req, desc, true);
+	if (!status) {
+		/* This mac addr exist, remove this handle's VFID for it */
+		status = hclge_update_desc_vfid(desc, vport->vport_id, true);
+		if (status)
+			return status;
+
+		if (hclge_is_all_function_id_zero(desc)) {
+			/* All the vfid is zero, so need to delete this entry */
+			status = hclge_remove_mac_vlan_tbl(vport, &req);
+			if (!status)
+				hdev->used_mc_mac_num--;
+		} else {
+			/* Not all the vfid is zero, update the vfid */
+			status = hclge_add_mac_vlan_tbl(vport, &req, desc);
+		}
+	} else if (status == -ENOENT) {
+		status = 0;
+	}
+
+	return status;
+}
+
+static void hclge_sync_vport_mac_list(struct hclge_vport *vport,
+				      struct list_head *list,
+				      enum HCLGE_MAC_ADDR_TYPE mac_type)
+{
+	int (*sync)(struct hclge_vport *vport, const unsigned char *addr);
+	struct hclge_mac_node *mac_node, *tmp;
+	int ret;
+
+	if (mac_type == HCLGE_MAC_ADDR_UC)
+		sync = hclge_add_uc_addr_common;
+	else
+		sync = hclge_add_mc_addr_common;
+
+	list_for_each_entry_safe(mac_node, tmp, list, node) {
+		ret = sync(vport, mac_node->mac_addr);
+		if (!ret) {
+			mac_node->state = HCLGE_MAC_ACTIVE;
+		} else {
+			set_bit(HCLGE_VPORT_STATE_MAC_TBL_CHANGE,
+				&vport->state);
+
+			/* If one unicast mac address is existing in hardware,
+			 * we need to try whether other unicast mac addresses
+			 * are new addresses that can be added.
+			 * Multicast mac address can be reusable, even though
+			 * there is no space to add new multicast mac address,
+			 * we should check whether other mac addresses are
+			 * existing in hardware for reuse.
+			 */
+			if ((mac_type == HCLGE_MAC_ADDR_UC && ret != -EEXIST) ||
+			    (mac_type == HCLGE_MAC_ADDR_MC && ret != -ENOSPC))
+				break;
+		}
+	}
+}
+
+static void hclge_unsync_vport_mac_list(struct hclge_vport *vport,
+					struct list_head *list,
+					enum HCLGE_MAC_ADDR_TYPE mac_type)
+{
+	int (*unsync)(struct hclge_vport *vport, const unsigned char *addr);
+	struct hclge_mac_node *mac_node, *tmp;
+	int ret;
+
+	if (mac_type == HCLGE_MAC_ADDR_UC)
+		unsync = hclge_rm_uc_addr_common;
+	else
+		unsync = hclge_rm_mc_addr_common;
+
+	list_for_each_entry_safe(mac_node, tmp, list, node) {
+		ret = unsync(vport, mac_node->mac_addr);
+		if (!ret || ret == -ENOENT) {
+			list_del(&mac_node->node);
+			kfree(mac_node);
+		} else {
+			set_bit(HCLGE_VPORT_STATE_MAC_TBL_CHANGE,
+				&vport->state);
+			break;
+		}
+	}
+}
+
+static bool hclge_sync_from_add_list(struct list_head *add_list,
+				     struct list_head *mac_list)
+{
+	struct hclge_mac_node *mac_node, *tmp, *new_node;
+	bool all_added = true;
+
+	list_for_each_entry_safe(mac_node, tmp, add_list, node) {
+		if (mac_node->state == HCLGE_MAC_TO_ADD)
+			all_added = false;
+
+		/* if the mac address from tmp_add_list is not in the
+		 * uc/mc_mac_list, it means have received a TO_DEL request
+		 * during the time window of adding the mac address into mac
+		 * table. if mac_node state is ACTIVE, then change it to TO_DEL,
+		 * then it will be removed at next time. else it must be TO_ADD,
+		 * this address hasn't been added into mac table,
+		 * so just remove the mac node.
+		 */
+		new_node = hclge_find_mac_node(mac_list, mac_node->mac_addr);
+		if (new_node) {
+			hclge_update_mac_node(new_node, mac_node->state);
+			list_del(&mac_node->node);
+			kfree(mac_node);
+		} else if (mac_node->state == HCLGE_MAC_ACTIVE) {
+			mac_node->state = HCLGE_MAC_TO_DEL;
+			list_move_tail(&mac_node->node, mac_list);
+		} else {
+			list_del(&mac_node->node);
+			kfree(mac_node);
+		}
+	}
+
+	return all_added;
+}
+
+static void hclge_sync_from_del_list(struct list_head *del_list,
+				     struct list_head *mac_list)
+{
+	struct hclge_mac_node *mac_node, *tmp, *new_node;
+
+	list_for_each_entry_safe(mac_node, tmp, del_list, node) {
+		new_node = hclge_find_mac_node(mac_list, mac_node->mac_addr);
+		if (new_node) {
+			/* If the mac addr exists in the mac list, it means
+			 * received a new TO_ADD request during the time window
+			 * of configuring the mac address. For the mac node
+			 * state is TO_ADD, and the address is already in the
+			 * in the hardware(due to delete fail), so we just need
+			 * to change the mac node state to ACTIVE.
+			 */
+			new_node->state = HCLGE_MAC_ACTIVE;
+			list_del(&mac_node->node);
+			kfree(mac_node);
+		} else {
+			list_move_tail(&mac_node->node, mac_list);
+		}
+	}
+}
+
+static void hclge_update_overflow_flags(struct hclge_vport *vport,
+					enum HCLGE_MAC_ADDR_TYPE mac_type,
+					bool is_all_added)
+{
+	if (mac_type == HCLGE_MAC_ADDR_UC) {
+		if (is_all_added)
+			vport->overflow_promisc_flags &= ~HNAE3_OVERFLOW_UPE;
+		else if (hclge_is_umv_space_full(vport, true))
+			vport->overflow_promisc_flags |= HNAE3_OVERFLOW_UPE;
+	} else {
+		if (is_all_added)
+			vport->overflow_promisc_flags &= ~HNAE3_OVERFLOW_MPE;
+		else
+			vport->overflow_promisc_flags |= HNAE3_OVERFLOW_MPE;
+	}
+}
+
+static void hclge_sync_vport_mac_table(struct hclge_vport *vport,
+				       enum HCLGE_MAC_ADDR_TYPE mac_type)
+{
+	struct hclge_mac_node *mac_node, *tmp, *new_node;
+	struct list_head tmp_add_list, tmp_del_list;
+	struct list_head *list;
+	bool all_added;
+
+	INIT_LIST_HEAD(&tmp_add_list);
+	INIT_LIST_HEAD(&tmp_del_list);
+
+	/* move the mac addr to the tmp_add_list and tmp_del_list, then
+	 * we can add/delete these mac addr outside the spin lock
+	 */
+	list = (mac_type == HCLGE_MAC_ADDR_UC) ?
+		&vport->uc_mac_list : &vport->mc_mac_list;
+
+	spin_lock_bh(&vport->mac_list_lock);
+
+	list_for_each_entry_safe(mac_node, tmp, list, node) {
+		switch (mac_node->state) {
+		case HCLGE_MAC_TO_DEL:
+			list_move_tail(&mac_node->node, &tmp_del_list);
+			break;
+		case HCLGE_MAC_TO_ADD:
+			new_node = kzalloc(sizeof(*new_node), GFP_ATOMIC);
+			if (!new_node)
+				goto stop_traverse;
+			ether_addr_copy(new_node->mac_addr, mac_node->mac_addr);
+			new_node->state = mac_node->state;
+			list_add_tail(&new_node->node, &tmp_add_list);
+			break;
+		default:
+			break;
+		}
+	}
+
+stop_traverse:
+	spin_unlock_bh(&vport->mac_list_lock);
+
+	/* delete first, in order to get max mac table space for adding */
+	hclge_unsync_vport_mac_list(vport, &tmp_del_list, mac_type);
+	hclge_sync_vport_mac_list(vport, &tmp_add_list, mac_type);
+
+	/* if some mac addresses were added/deleted fail, move back to the
+	 * mac_list, and retry at next time.
+	 */
+	spin_lock_bh(&vport->mac_list_lock);
+
+	hclge_sync_from_del_list(&tmp_del_list, list);
+	all_added = hclge_sync_from_add_list(&tmp_add_list, list);
+
+	spin_unlock_bh(&vport->mac_list_lock);
+
+	hclge_update_overflow_flags(vport, mac_type, all_added);
+}
+
+static bool hclge_need_sync_mac_table(struct hclge_vport *vport)
+{
+	struct hclge_dev *hdev = vport->back;
+
+	if (test_bit(vport->vport_id, hdev->vport_config_block))
+		return false;
+
+	if (test_and_clear_bit(HCLGE_VPORT_STATE_MAC_TBL_CHANGE, &vport->state))
+		return true;
+
+	return false;
+}
+
+static void hclge_sync_mac_table(struct hclge_dev *hdev)
+{
+	int i;
+
+	for (i = 0; i < hdev->num_alloc_vport; i++) {
+		struct hclge_vport *vport = &hdev->vport[i];
+
+		if (!hclge_need_sync_mac_table(vport))
+			continue;
+
+		hclge_sync_vport_mac_table(vport, HCLGE_MAC_ADDR_UC);
+		hclge_sync_vport_mac_table(vport, HCLGE_MAC_ADDR_MC);
+	}
+}
+
+static void hclge_build_del_list(struct list_head *list,
+				 bool is_del_list,
+				 struct list_head *tmp_del_list)
+{
+	struct hclge_mac_node *mac_cfg, *tmp;
+
+	list_for_each_entry_safe(mac_cfg, tmp, list, node) {
+		switch (mac_cfg->state) {
+		case HCLGE_MAC_TO_DEL:
+		case HCLGE_MAC_ACTIVE:
+			list_move_tail(&mac_cfg->node, tmp_del_list);
+			break;
+		case HCLGE_MAC_TO_ADD:
+			if (is_del_list) {
+				list_del(&mac_cfg->node);
+				kfree(mac_cfg);
+			}
+			break;
+		}
+	}
+}
+
+static void hclge_unsync_del_list(struct hclge_vport *vport,
+				  int (*unsync)(struct hclge_vport *vport,
+						const unsigned char *addr),
+				  bool is_del_list,
+				  struct list_head *tmp_del_list)
+{
+	struct hclge_mac_node *mac_cfg, *tmp;
+	int ret;
+
+	list_for_each_entry_safe(mac_cfg, tmp, tmp_del_list, node) {
+		ret = unsync(vport, mac_cfg->mac_addr);
+		if (!ret || ret == -ENOENT) {
+			/* clear all mac addr from hardware, but remain these
+			 * mac addr in the mac list, and restore them after
+			 * vf reset finished.
+			 */
+			if (!is_del_list &&
+			    mac_cfg->state == HCLGE_MAC_ACTIVE) {
+				mac_cfg->state = HCLGE_MAC_TO_ADD;
+			} else {
+				list_del(&mac_cfg->node);
+				kfree(mac_cfg);
+			}
+		} else if (is_del_list) {
+			mac_cfg->state = HCLGE_MAC_TO_DEL;
+		}
+	}
+}
+
+void hclge_rm_vport_all_mac_table(struct hclge_vport *vport, bool is_del_list,
+				  enum HCLGE_MAC_ADDR_TYPE mac_type)
+{
+	int (*unsync)(struct hclge_vport *vport, const unsigned char *addr);
+	struct hclge_dev *hdev = vport->back;
+	struct list_head tmp_del_list, *list;
+
+	if (mac_type == HCLGE_MAC_ADDR_UC) {
+		list = &vport->uc_mac_list;
+		unsync = hclge_rm_uc_addr_common;
+	} else {
+		list = &vport->mc_mac_list;
+		unsync = hclge_rm_mc_addr_common;
+	}
+
+	INIT_LIST_HEAD(&tmp_del_list);
+
+	if (!is_del_list)
+		set_bit(vport->vport_id, hdev->vport_config_block);
+
+	spin_lock_bh(&vport->mac_list_lock);
+
+	hclge_build_del_list(list, is_del_list, &tmp_del_list);
+
+	spin_unlock_bh(&vport->mac_list_lock);
+
+	hclge_unsync_del_list(vport, unsync, is_del_list, &tmp_del_list);
+
+	spin_lock_bh(&vport->mac_list_lock);
+
+	hclge_sync_from_del_list(&tmp_del_list, list);
+
+	spin_unlock_bh(&vport->mac_list_lock);
+}
+
+/* remove all mac address when uninitailize */
+static void hclge_uninit_vport_mac_list(struct hclge_vport *vport,
+					enum HCLGE_MAC_ADDR_TYPE mac_type)
+{
+	struct hclge_mac_node *mac_node, *tmp;
+	struct hclge_dev *hdev = vport->back;
+	struct list_head tmp_del_list, *list;
+
+	INIT_LIST_HEAD(&tmp_del_list);
+
+	list = (mac_type == HCLGE_MAC_ADDR_UC) ?
+		&vport->uc_mac_list : &vport->mc_mac_list;
+
+	spin_lock_bh(&vport->mac_list_lock);
+
+	list_for_each_entry_safe(mac_node, tmp, list, node) {
+		switch (mac_node->state) {
+		case HCLGE_MAC_TO_DEL:
+		case HCLGE_MAC_ACTIVE:
+			list_move_tail(&mac_node->node, &tmp_del_list);
+			break;
+		case HCLGE_MAC_TO_ADD:
+			list_del(&mac_node->node);
+			kfree(mac_node);
+			break;
+		}
+	}
+
+	spin_unlock_bh(&vport->mac_list_lock);
+
+	hclge_unsync_vport_mac_list(vport, &tmp_del_list, mac_type);
+
+	if (!list_empty(&tmp_del_list))
+		dev_warn(&hdev->pdev->dev,
+			 "uninit %s mac list for vport %u not completely.\n",
+			 mac_type == HCLGE_MAC_ADDR_UC ? "uc" : "mc",
+			 vport->vport_id);
+
+	list_for_each_entry_safe(mac_node, tmp, &tmp_del_list, node) {
+		list_del(&mac_node->node);
+		kfree(mac_node);
+	}
+}
+
+static void hclge_uninit_mac_table(struct hclge_dev *hdev)
+{
+	struct hclge_vport *vport;
+	int i;
+
+	for (i = 0; i < hdev->num_alloc_vport; i++) {
+		vport = &hdev->vport[i];
+		hclge_uninit_vport_mac_list(vport, HCLGE_MAC_ADDR_UC);
+		hclge_uninit_vport_mac_list(vport, HCLGE_MAC_ADDR_MC);
+	}
+}
+
+static int hclge_get_mac_ethertype_cmd_status(struct hclge_dev *hdev,
+					      u16 cmdq_resp, u8 resp_code)
+{
+#define HCLGE_ETHERTYPE_SUCCESS_ADD		0
+#define HCLGE_ETHERTYPE_ALREADY_ADD		1
+#define HCLGE_ETHERTYPE_MGR_TBL_OVERFLOW	2
+#define HCLGE_ETHERTYPE_KEY_CONFLICT		3
+
+	int return_status;
+
+	if (cmdq_resp) {
+		dev_err(&hdev->pdev->dev,
+			"cmdq execute failed for get_mac_ethertype_cmd_status, status=%u.\n",
+			cmdq_resp);
+		return -EIO;
+	}
+
+	switch (resp_code) {
+	case HCLGE_ETHERTYPE_SUCCESS_ADD:
+	case HCLGE_ETHERTYPE_ALREADY_ADD:
+		return_status = 0;
+		break;
+	case HCLGE_ETHERTYPE_MGR_TBL_OVERFLOW:
+		dev_err(&hdev->pdev->dev,
+			"add mac ethertype failed for manager table overflow.\n");
+		return_status = -EIO;
+		break;
+	case HCLGE_ETHERTYPE_KEY_CONFLICT:
+		dev_err(&hdev->pdev->dev,
+			"add mac ethertype failed for key conflict.\n");
+		return_status = -EIO;
+		break;
+	default:
+		dev_err(&hdev->pdev->dev,
+			"add mac ethertype failed for undefined, code=%u.\n",
+			resp_code);
+		return_status = -EIO;
+	}
+
+	return return_status;
+}
+
+static int hclge_set_vf_mac(struct hnae3_handle *handle, int vf,
+			    u8 *mac_addr)
+{
+	struct hclge_vport *vport = hclge_get_vport(handle);
+	char format_mac_addr[HNAE3_FORMAT_MAC_ADDR_LEN];
+	struct hclge_dev *hdev = vport->back;
+
+	vport = hclge_get_vf_vport(hdev, vf);
+	if (!vport)
+		return -EINVAL;
+
+	hnae3_format_mac_addr(format_mac_addr, mac_addr);
+	if (ether_addr_equal(mac_addr, vport->vf_info.mac)) {
+		dev_info(&hdev->pdev->dev,
+			 "Specified MAC(=%s) is same as before, no change committed!\n",
+			 format_mac_addr);
+		return 0;
+	}
+
+	ether_addr_copy(vport->vf_info.mac, mac_addr);
+
+	/* there is a timewindow for PF to know VF unalive, it may
+	 * cause send mailbox fail, but it doesn't matter, VF will
+	 * query it when reinit.
+	 */
+	if (test_bit(HCLGE_VPORT_STATE_ALIVE, &vport->state)) {
+		dev_info(&hdev->pdev->dev,
+			 "MAC of VF %d has been set to %s, and it will be reinitialized!\n",
+			 vf, format_mac_addr);
+		(void)hclge_inform_reset_assert_to_vf(vport);
+		return 0;
+	}
+
+	dev_info(&hdev->pdev->dev,
+		 "MAC of VF %d has been set to %s, will be active after VF reset\n",
+		 vf, format_mac_addr);
+	return 0;
+}
+
+static int hclge_add_mgr_tbl(struct hclge_dev *hdev,
+			     const struct hclge_mac_mgr_tbl_entry_cmd *req)
+{
+	struct hclge_desc desc;
+	u8 resp_code;
+	u16 retval;
+	int ret;
+
+	hclge_cmd_setup_basic_desc(&desc, HCLGE_OPC_MAC_ETHTYPE_ADD, false);
+	memcpy(desc.data, req, sizeof(struct hclge_mac_mgr_tbl_entry_cmd));
+
+	ret = hclge_cmd_send(&hdev->hw, &desc, 1);
+	if (ret) {
+		dev_err(&hdev->pdev->dev,
+			"add mac ethertype failed for cmd_send, ret =%d.\n",
+			ret);
+		return ret;
+	}
+
+	resp_code = (le32_to_cpu(desc.data[0]) >> 8) & 0xff;
+	retval = le16_to_cpu(desc.retval);
+
+	return hclge_get_mac_ethertype_cmd_status(hdev, retval, resp_code);
+}
+
+static int init_mgr_tbl(struct hclge_dev *hdev)
+{
+	int ret;
+	u32 i;
+
+	for (i = 0; i < ARRAY_SIZE(hclge_mgr_table); i++) {
+		ret = hclge_add_mgr_tbl(hdev, &hclge_mgr_table[i]);
+		if (ret) {
+			dev_err(&hdev->pdev->dev,
+				"add mac ethertype failed, ret =%d.\n",
+				ret);
+			return ret;
+		}
+	}
+
+	return 0;
+}
+
+static void hclge_get_mac_addr(struct hnae3_handle *handle, u8 *p)
+{
+	struct hclge_vport *vport = hclge_get_vport(handle);
+	struct hclge_dev *hdev = vport->back;
+
+	ether_addr_copy(p, hdev->hw.mac.mac_addr);
+}
+
+int hclge_update_mac_node_for_dev_addr(struct hclge_vport *vport,
+				       const u8 *old_addr, const u8 *new_addr)
+{
+	struct list_head *list = &vport->uc_mac_list;
+	struct hclge_mac_node *old_node, *new_node;
+
+	new_node = hclge_find_mac_node(list, new_addr);
+	if (!new_node) {
+		new_node = kzalloc(sizeof(*new_node), GFP_ATOMIC);
+		if (!new_node)
+			return -ENOMEM;
+
+		new_node->state = HCLGE_MAC_TO_ADD;
+		ether_addr_copy(new_node->mac_addr, new_addr);
+		list_add(&new_node->node, list);
+	} else {
+		if (new_node->state == HCLGE_MAC_TO_DEL)
+			new_node->state = HCLGE_MAC_ACTIVE;
+
+		/* make sure the new addr is in the list head, avoid dev
+		 * addr may be not re-added into mac table for the umv space
+		 * limitation after global/imp reset which will clear mac
+		 * table by hardware.
+		 */
+		list_move(&new_node->node, list);
+	}
+
+	if (old_addr && !ether_addr_equal(old_addr, new_addr)) {
+		old_node = hclge_find_mac_node(list, old_addr);
+		if (old_node) {
+			if (old_node->state == HCLGE_MAC_TO_ADD) {
+				list_del(&old_node->node);
+				kfree(old_node);
+			} else {
+				old_node->state = HCLGE_MAC_TO_DEL;
+			}
+		}
+	}
+
+	set_bit(HCLGE_VPORT_STATE_MAC_TBL_CHANGE, &vport->state);
+
+	return 0;
+}
+
+static int hclge_set_mac_addr(struct hnae3_handle *handle, const void *p,
+			      bool is_first)
+{
+	const unsigned char *new_addr = (const unsigned char *)p;
+	struct hclge_vport *vport = hclge_get_vport(handle);
+	char format_mac_addr[HNAE3_FORMAT_MAC_ADDR_LEN];
+	struct hclge_dev *hdev = vport->back;
+	unsigned char *old_addr = NULL;
+	int ret;
+
+	/* mac addr check */
+	if (is_zero_ether_addr(new_addr) ||
+	    is_broadcast_ether_addr(new_addr) ||
+	    is_multicast_ether_addr(new_addr)) {
+		hnae3_format_mac_addr(format_mac_addr, new_addr);
+		dev_err(&hdev->pdev->dev,
+			"change uc mac err! invalid mac: %s.\n",
+			 format_mac_addr);
+		return -EINVAL;
+	}
+
+	ret = hclge_pause_addr_cfg(hdev, new_addr);
+	if (ret) {
+		dev_err(&hdev->pdev->dev,
+			"failed to configure mac pause address, ret = %d\n",
+			ret);
+		return ret;
+	}
+
+	if (!is_first)
+		old_addr = hdev->hw.mac.mac_addr;
+
+	spin_lock_bh(&vport->mac_list_lock);
+	ret = hclge_update_mac_node_for_dev_addr(vport, old_addr, new_addr);
+	if (ret) {
+		hnae3_format_mac_addr(format_mac_addr, new_addr);
+		dev_err(&hdev->pdev->dev,
+			"failed to change the mac addr:%s, ret = %d\n",
+			format_mac_addr, ret);
+		spin_unlock_bh(&vport->mac_list_lock);
+
+		if (!is_first)
+			hclge_pause_addr_cfg(hdev, old_addr);
+
+		return ret;
+	}
+	/* we must update dev addr with spin lock protect, preventing dev addr
+	 * being removed by set_rx_mode path.
+	 */
+	ether_addr_copy(hdev->hw.mac.mac_addr, new_addr);
+	spin_unlock_bh(&vport->mac_list_lock);
+
+	hclge_task_schedule(hdev, 0);
+
+	return 0;
+}
+
+static int hclge_mii_ioctl(struct hclge_dev *hdev, struct ifreq *ifr, int cmd)
+{
+	struct mii_ioctl_data *data = if_mii(ifr);
+
+	if (!hnae3_dev_phy_imp_supported(hdev))
+		return -EOPNOTSUPP;
+
+	switch (cmd) {
+	case SIOCGMIIPHY:
+		data->phy_id = hdev->hw.mac.phy_addr;
+		/* this command reads phy id and register at the same time */
+		fallthrough;
+	case SIOCGMIIREG:
+		data->val_out = hclge_read_phy_reg(hdev, data->reg_num);
+		return 0;
+
+	case SIOCSMIIREG:
+		return hclge_write_phy_reg(hdev, data->reg_num, data->val_in);
+	default:
+		return -EOPNOTSUPP;
+	}
+}
+
+static int hclge_do_ioctl(struct hnae3_handle *handle, struct ifreq *ifr,
+			  int cmd)
+{
+	struct hclge_vport *vport = hclge_get_vport(handle);
+	struct hclge_dev *hdev = vport->back;
+
+	switch (cmd) {
+	case SIOCGHWTSTAMP:
+		return hclge_ptp_get_cfg(hdev, ifr);
+	case SIOCSHWTSTAMP:
+		return hclge_ptp_set_cfg(hdev, ifr);
+	default:
+		if (!hdev->hw.mac.phydev)
+			return hclge_mii_ioctl(hdev, ifr, cmd);
+	}
+
+	return phy_mii_ioctl(hdev->hw.mac.phydev, ifr, cmd);
+}
+
+static int hclge_set_port_vlan_filter_bypass(struct hclge_dev *hdev, u8 vf_id,
+					     bool bypass_en)
+{
+	struct hclge_port_vlan_filter_bypass_cmd *req;
+	struct hclge_desc desc;
+	int ret;
+
+	hclge_cmd_setup_basic_desc(&desc, HCLGE_OPC_PORT_VLAN_BYPASS, false);
+	req = (struct hclge_port_vlan_filter_bypass_cmd *)desc.data;
+	req->vf_id = vf_id;
+	hnae3_set_bit(req->bypass_state, HCLGE_INGRESS_BYPASS_B,
+		      bypass_en ? 1 : 0);
+
+	ret = hclge_cmd_send(&hdev->hw, &desc, 1);
+	if (ret)
+		dev_err(&hdev->pdev->dev,
+			"failed to set vport%u port vlan filter bypass state, ret = %d.\n",
+			vf_id, ret);
+
+	return ret;
+}
+
+static int hclge_set_vlan_filter_ctrl(struct hclge_dev *hdev, u8 vlan_type,
+				      u8 fe_type, bool filter_en, u8 vf_id)
+{
+	struct hclge_vlan_filter_ctrl_cmd *req;
+	struct hclge_desc desc;
+	int ret;
+
+	/* read current vlan filter parameter */
+	hclge_cmd_setup_basic_desc(&desc, HCLGE_OPC_VLAN_FILTER_CTRL, true);
+	req = (struct hclge_vlan_filter_ctrl_cmd *)desc.data;
+	req->vlan_type = vlan_type;
+	req->vf_id = vf_id;
+
+	ret = hclge_cmd_send(&hdev->hw, &desc, 1);
+	if (ret) {
+		dev_err(&hdev->pdev->dev, "failed to get vport%u vlan filter config, ret = %d.\n",
+			vf_id, ret);
+		return ret;
+	}
+
+	/* modify and write new config parameter */
+	hclge_comm_cmd_reuse_desc(&desc, false);
+	req->vlan_fe = filter_en ?
+			(req->vlan_fe | fe_type) : (req->vlan_fe & ~fe_type);
+
+	ret = hclge_cmd_send(&hdev->hw, &desc, 1);
+	if (ret)
+		dev_err(&hdev->pdev->dev, "failed to set vport%u vlan filter, ret = %d.\n",
+			vf_id, ret);
+
+	return ret;
+}
+
+static int hclge_set_vport_vlan_filter(struct hclge_vport *vport, bool enable)
+{
+	struct hclge_dev *hdev = vport->back;
+	struct hnae3_ae_dev *ae_dev = hdev->ae_dev;
+	int ret;
+
+	if (hdev->ae_dev->dev_version < HNAE3_DEVICE_VERSION_V2)
+		return hclge_set_vlan_filter_ctrl(hdev, HCLGE_FILTER_TYPE_VF,
+						  HCLGE_FILTER_FE_EGRESS_V1_B,
+						  enable, vport->vport_id);
+
+	ret = hclge_set_vlan_filter_ctrl(hdev, HCLGE_FILTER_TYPE_VF,
+					 HCLGE_FILTER_FE_EGRESS, enable,
+					 vport->vport_id);
+	if (ret)
+		return ret;
+
+	if (test_bit(HNAE3_DEV_SUPPORT_PORT_VLAN_BYPASS_B, ae_dev->caps)) {
+		ret = hclge_set_port_vlan_filter_bypass(hdev, vport->vport_id,
+							!enable);
+	} else if (!vport->vport_id) {
+		if (test_bit(HNAE3_DEV_SUPPORT_VLAN_FLTR_MDF_B, ae_dev->caps))
+			enable = false;
+
+		ret = hclge_set_vlan_filter_ctrl(hdev, HCLGE_FILTER_TYPE_PORT,
+						 HCLGE_FILTER_FE_INGRESS,
+						 enable, 0);
+	}
+
+	return ret;
+}
+
+static bool hclge_need_enable_vport_vlan_filter(struct hclge_vport *vport)
+{
+	struct hnae3_handle *handle = &vport->nic;
+	struct hclge_vport_vlan_cfg *vlan, *tmp;
+	struct hclge_dev *hdev = vport->back;
+
+	if (vport->vport_id) {
+		if (vport->port_base_vlan_cfg.state !=
+			HNAE3_PORT_BASE_VLAN_DISABLE)
+			return true;
+
+		if (vport->vf_info.trusted && vport->vf_info.request_uc_en)
+			return false;
+	} else if (handle->netdev_flags & HNAE3_USER_UPE) {
+		return false;
+	}
+
+	if (!vport->req_vlan_fltr_en)
+		return false;
+
+	/* compatible with former device, always enable vlan filter */
+	if (!test_bit(HNAE3_DEV_SUPPORT_VLAN_FLTR_MDF_B, hdev->ae_dev->caps))
+		return true;
+
+	list_for_each_entry_safe(vlan, tmp, &vport->vlan_list, node)
+		if (vlan->vlan_id != 0)
+			return true;
+
+	return false;
+}
+
+static int __hclge_enable_vport_vlan_filter(struct hclge_vport *vport,
+					    bool request_en)
+{
+	bool need_en;
+	int ret;
+
+	need_en = hclge_need_enable_vport_vlan_filter(vport);
+	if (need_en == vport->cur_vlan_fltr_en)
+		return 0;
+
+	ret = hclge_set_vport_vlan_filter(vport, need_en);
+	if (ret)
+		return ret;
+
+	vport->cur_vlan_fltr_en = need_en;
+
+	return 0;
+}
+
+int hclge_enable_vport_vlan_filter(struct hclge_vport *vport, bool request_en)
+{
+	struct hclge_dev *hdev = vport->back;
+	int ret;
+
+	mutex_lock(&hdev->vport_lock);
+	vport->req_vlan_fltr_en = request_en;
+	ret = __hclge_enable_vport_vlan_filter(vport, request_en);
+	mutex_unlock(&hdev->vport_lock);
+
+	return ret;
+}
+
+static int hclge_enable_vlan_filter(struct hnae3_handle *handle, bool enable)
+{
+	struct hclge_vport *vport = hclge_get_vport(handle);
+
+	return hclge_enable_vport_vlan_filter(vport, enable);
+}
+
+static int hclge_set_vf_vlan_filter_cmd(struct hclge_dev *hdev, u16 vfid,
+					bool is_kill, u16 vlan,
+					struct hclge_desc *desc)
+{
+	struct hclge_vlan_filter_vf_cfg_cmd *req0;
+	struct hclge_vlan_filter_vf_cfg_cmd *req1;
+	u8 vf_byte_val;
+	u8 vf_byte_off;
+	int ret;
+
+	hclge_cmd_setup_basic_desc(&desc[0],
+				   HCLGE_OPC_VLAN_FILTER_VF_CFG, false);
+	hclge_cmd_setup_basic_desc(&desc[1],
+				   HCLGE_OPC_VLAN_FILTER_VF_CFG, false);
+
+	desc[0].flag |= cpu_to_le16(HCLGE_COMM_CMD_FLAG_NEXT);
+
+	vf_byte_off = vfid / 8;
+	vf_byte_val = 1 << (vfid % 8);
+
+	req0 = (struct hclge_vlan_filter_vf_cfg_cmd *)desc[0].data;
+	req1 = (struct hclge_vlan_filter_vf_cfg_cmd *)desc[1].data;
+
+	req0->vlan_id  = cpu_to_le16(vlan);
+	req0->vlan_cfg = is_kill;
+
+	if (vf_byte_off < HCLGE_MAX_VF_BYTES)
+		req0->vf_bitmap[vf_byte_off] = vf_byte_val;
+	else
+		req1->vf_bitmap[vf_byte_off - HCLGE_MAX_VF_BYTES] = vf_byte_val;
+
+	ret = hclge_cmd_send(&hdev->hw, desc, 2);
+	if (ret) {
+		dev_err(&hdev->pdev->dev,
+			"Send vf vlan command fail, ret =%d.\n",
+			ret);
+		return ret;
+	}
+
+	return 0;
+}
+
+static int hclge_check_vf_vlan_cmd_status(struct hclge_dev *hdev, u16 vfid,
+					  bool is_kill, struct hclge_desc *desc)
+{
+	struct hclge_vlan_filter_vf_cfg_cmd *req;
+
+	req = (struct hclge_vlan_filter_vf_cfg_cmd *)desc[0].data;
+
+	if (!is_kill) {
+#define HCLGE_VF_VLAN_NO_ENTRY	2
+		if (!req->resp_code || req->resp_code == 1)
+			return 0;
+
+		if (req->resp_code == HCLGE_VF_VLAN_NO_ENTRY) {
+			set_bit(vfid, hdev->vf_vlan_full);
+			dev_warn(&hdev->pdev->dev,
+				 "vf vlan table is full, vf vlan filter is disabled\n");
+			return 0;
+		}
+
+		dev_err(&hdev->pdev->dev,
+			"Add vf vlan filter fail, ret =%u.\n",
+			req->resp_code);
+	} else {
+#define HCLGE_VF_VLAN_DEL_NO_FOUND	1
+		if (!req->resp_code)
+			return 0;
+
+		/* vf vlan filter is disabled when vf vlan table is full,
+		 * then new vlan id will not be added into vf vlan table.
+		 * Just return 0 without warning, avoid massive verbose
+		 * print logs when unload.
+		 */
+		if (req->resp_code == HCLGE_VF_VLAN_DEL_NO_FOUND)
+			return 0;
+
+		dev_err(&hdev->pdev->dev,
+			"Kill vf vlan filter fail, ret =%u.\n",
+			req->resp_code);
+	}
+
+	return -EIO;
+}
+
+static int hclge_set_vf_vlan_common(struct hclge_dev *hdev, u16 vfid,
+				    bool is_kill, u16 vlan)
+{
+	struct hclge_vport *vport = &hdev->vport[vfid];
+	struct hclge_desc desc[2];
+	int ret;
+
+	/* if vf vlan table is full, firmware will close vf vlan filter, it
+	 * is unable and unnecessary to add new vlan id to vf vlan filter.
+	 * If spoof check is enable, and vf vlan is full, it shouldn't add
+	 * new vlan, because tx packets with these vlan id will be dropped.
+	 */
+	if (test_bit(vfid, hdev->vf_vlan_full) && !is_kill) {
+		if (vport->vf_info.spoofchk && vlan) {
+			dev_err(&hdev->pdev->dev,
+				"Can't add vlan due to spoof check is on and vf vlan table is full\n");
+			return -EPERM;
+		}
+		return 0;
+	}
+
+	ret = hclge_set_vf_vlan_filter_cmd(hdev, vfid, is_kill, vlan, desc);
+	if (ret)
+		return ret;
+
+	return hclge_check_vf_vlan_cmd_status(hdev, vfid, is_kill, desc);
+}
+
+static int hclge_set_port_vlan_filter(struct hclge_dev *hdev, __be16 proto,
+				      u16 vlan_id, bool is_kill)
+{
+	struct hclge_vlan_filter_pf_cfg_cmd *req;
+	struct hclge_desc desc;
+	u8 vlan_offset_byte_val;
+	u8 vlan_offset_byte;
+	u8 vlan_offset_160;
+	int ret;
+
+	hclge_cmd_setup_basic_desc(&desc, HCLGE_OPC_VLAN_FILTER_PF_CFG, false);
+
+	vlan_offset_160 = vlan_id / HCLGE_VLAN_ID_OFFSET_STEP;
+	vlan_offset_byte = (vlan_id % HCLGE_VLAN_ID_OFFSET_STEP) /
+			   HCLGE_VLAN_BYTE_SIZE;
+	vlan_offset_byte_val = 1 << (vlan_id % HCLGE_VLAN_BYTE_SIZE);
+
+	req = (struct hclge_vlan_filter_pf_cfg_cmd *)desc.data;
+	req->vlan_offset = vlan_offset_160;
+	req->vlan_cfg = is_kill;
+	req->vlan_offset_bitmap[vlan_offset_byte] = vlan_offset_byte_val;
+
+	ret = hclge_cmd_send(&hdev->hw, &desc, 1);
+	if (ret)
+		dev_err(&hdev->pdev->dev,
+			"port vlan command, send fail, ret =%d.\n", ret);
+	return ret;
+}
+
+static bool hclge_need_update_port_vlan(struct hclge_dev *hdev, u16 vport_id,
+					u16 vlan_id, bool is_kill)
+{
+	/* vlan 0 may be added twice when 8021q module is enabled */
+	if (!is_kill && !vlan_id &&
+	    test_bit(vport_id, hdev->vlan_table[vlan_id]))
+		return false;
+
+	if (!is_kill && test_and_set_bit(vport_id, hdev->vlan_table[vlan_id])) {
+		dev_warn(&hdev->pdev->dev,
+			 "Add port vlan failed, vport %u is already in vlan %u\n",
+			 vport_id, vlan_id);
+		return false;
+	}
+
+	if (is_kill &&
+	    !test_and_clear_bit(vport_id, hdev->vlan_table[vlan_id])) {
+		dev_warn(&hdev->pdev->dev,
+			 "Delete port vlan failed, vport %u is not in vlan %u\n",
+			 vport_id, vlan_id);
+		return false;
+	}
+
+	return true;
+}
+
+static int hclge_set_vlan_filter_hw(struct hclge_dev *hdev, __be16 proto,
+				    u16 vport_id, u16 vlan_id,
+				    bool is_kill)
+{
+	u16 vport_idx, vport_num = 0;
+	int ret;
+
+	if (is_kill && !vlan_id)
+		return 0;
+
+	if (vlan_id >= VLAN_N_VID)
+		return -EINVAL;
+
+	ret = hclge_set_vf_vlan_common(hdev, vport_id, is_kill, vlan_id);
+	if (ret) {
+		dev_err(&hdev->pdev->dev,
+			"Set %u vport vlan filter config fail, ret =%d.\n",
+			vport_id, ret);
+		return ret;
+	}
+
+	if (!hclge_need_update_port_vlan(hdev, vport_id, vlan_id, is_kill))
+		return 0;
+
+	for_each_set_bit(vport_idx, hdev->vlan_table[vlan_id], HCLGE_VPORT_NUM)
+		vport_num++;
+
+	if ((is_kill && vport_num == 0) || (!is_kill && vport_num == 1))
+		ret = hclge_set_port_vlan_filter(hdev, proto, vlan_id,
+						 is_kill);
+
+	return ret;
+}
+
+static int hclge_set_vlan_tx_offload_cfg(struct hclge_vport *vport)
+{
+	struct hclge_tx_vtag_cfg *vcfg = &vport->txvlan_cfg;
+	struct hclge_vport_vtag_tx_cfg_cmd *req;
+	struct hclge_dev *hdev = vport->back;
+	struct hclge_desc desc;
+	u16 bmap_index;
+	int status;
+
+	hclge_cmd_setup_basic_desc(&desc, HCLGE_OPC_VLAN_PORT_TX_CFG, false);
+
+	req = (struct hclge_vport_vtag_tx_cfg_cmd *)desc.data;
+	req->def_vlan_tag1 = cpu_to_le16(vcfg->default_tag1);
+	req->def_vlan_tag2 = cpu_to_le16(vcfg->default_tag2);
+	hnae3_set_bit(req->vport_vlan_cfg, HCLGE_ACCEPT_TAG1_B,
+		      vcfg->accept_tag1 ? 1 : 0);
+	hnae3_set_bit(req->vport_vlan_cfg, HCLGE_ACCEPT_UNTAG1_B,
+		      vcfg->accept_untag1 ? 1 : 0);
+	hnae3_set_bit(req->vport_vlan_cfg, HCLGE_ACCEPT_TAG2_B,
+		      vcfg->accept_tag2 ? 1 : 0);
+	hnae3_set_bit(req->vport_vlan_cfg, HCLGE_ACCEPT_UNTAG2_B,
+		      vcfg->accept_untag2 ? 1 : 0);
+	hnae3_set_bit(req->vport_vlan_cfg, HCLGE_PORT_INS_TAG1_EN_B,
+		      vcfg->insert_tag1_en ? 1 : 0);
+	hnae3_set_bit(req->vport_vlan_cfg, HCLGE_PORT_INS_TAG2_EN_B,
+		      vcfg->insert_tag2_en ? 1 : 0);
+	hnae3_set_bit(req->vport_vlan_cfg, HCLGE_TAG_SHIFT_MODE_EN_B,
+		      vcfg->tag_shift_mode_en ? 1 : 0);
+	hnae3_set_bit(req->vport_vlan_cfg, HCLGE_CFG_NIC_ROCE_SEL_B, 0);
+
+	req->vf_offset = vport->vport_id / HCLGE_VF_NUM_PER_CMD;
+	bmap_index = vport->vport_id % HCLGE_VF_NUM_PER_CMD /
+			HCLGE_VF_NUM_PER_BYTE;
+	req->vf_bitmap[bmap_index] =
+		1U << (vport->vport_id % HCLGE_VF_NUM_PER_BYTE);
+
+	status = hclge_cmd_send(&hdev->hw, &desc, 1);
+	if (status)
+		dev_err(&hdev->pdev->dev,
+			"Send port txvlan cfg command fail, ret =%d\n",
+			status);
+
+	return status;
+}
+
+static int hclge_set_vlan_rx_offload_cfg(struct hclge_vport *vport)
+{
+	struct hclge_rx_vtag_cfg *vcfg = &vport->rxvlan_cfg;
+	struct hclge_vport_vtag_rx_cfg_cmd *req;
+	struct hclge_dev *hdev = vport->back;
+	struct hclge_desc desc;
+	u16 bmap_index;
+	int status;
+
+	hclge_cmd_setup_basic_desc(&desc, HCLGE_OPC_VLAN_PORT_RX_CFG, false);
+
+	req = (struct hclge_vport_vtag_rx_cfg_cmd *)desc.data;
+	hnae3_set_bit(req->vport_vlan_cfg, HCLGE_REM_TAG1_EN_B,
+		      vcfg->strip_tag1_en ? 1 : 0);
+	hnae3_set_bit(req->vport_vlan_cfg, HCLGE_REM_TAG2_EN_B,
+		      vcfg->strip_tag2_en ? 1 : 0);
+	hnae3_set_bit(req->vport_vlan_cfg, HCLGE_SHOW_TAG1_EN_B,
+		      vcfg->vlan1_vlan_prionly ? 1 : 0);
+	hnae3_set_bit(req->vport_vlan_cfg, HCLGE_SHOW_TAG2_EN_B,
+		      vcfg->vlan2_vlan_prionly ? 1 : 0);
+	hnae3_set_bit(req->vport_vlan_cfg, HCLGE_DISCARD_TAG1_EN_B,
+		      vcfg->strip_tag1_discard_en ? 1 : 0);
+	hnae3_set_bit(req->vport_vlan_cfg, HCLGE_DISCARD_TAG2_EN_B,
+		      vcfg->strip_tag2_discard_en ? 1 : 0);
+
+	req->vf_offset = vport->vport_id / HCLGE_VF_NUM_PER_CMD;
+	bmap_index = vport->vport_id % HCLGE_VF_NUM_PER_CMD /
+			HCLGE_VF_NUM_PER_BYTE;
+	req->vf_bitmap[bmap_index] =
+		1U << (vport->vport_id % HCLGE_VF_NUM_PER_BYTE);
+
+	status = hclge_cmd_send(&hdev->hw, &desc, 1);
+	if (status)
+		dev_err(&hdev->pdev->dev,
+			"Send port rxvlan cfg command fail, ret =%d\n",
+			status);
+
+	return status;
+}
+
+static int hclge_vlan_offload_cfg(struct hclge_vport *vport,
+				  u16 port_base_vlan_state,
+				  u16 vlan_tag, u8 qos)
+{
+	int ret;
+
+	if (port_base_vlan_state == HNAE3_PORT_BASE_VLAN_DISABLE) {
+		vport->txvlan_cfg.accept_tag1 = true;
+		vport->txvlan_cfg.insert_tag1_en = false;
+		vport->txvlan_cfg.default_tag1 = 0;
+	} else {
+		struct hnae3_ae_dev *ae_dev = pci_get_drvdata(vport->nic.pdev);
+
+		vport->txvlan_cfg.accept_tag1 =
+			ae_dev->dev_version >= HNAE3_DEVICE_VERSION_V3;
+		vport->txvlan_cfg.insert_tag1_en = true;
+		vport->txvlan_cfg.default_tag1 = (qos << VLAN_PRIO_SHIFT) |
+						 vlan_tag;
+	}
+
+	vport->txvlan_cfg.accept_untag1 = true;
+
+	/* accept_tag2 and accept_untag2 are not supported on
+	 * pdev revision(0x20), new revision support them,
+	 * this two fields can not be configured by user.
+	 */
+	vport->txvlan_cfg.accept_tag2 = true;
+	vport->txvlan_cfg.accept_untag2 = true;
+	vport->txvlan_cfg.insert_tag2_en = false;
+	vport->txvlan_cfg.default_tag2 = 0;
+	vport->txvlan_cfg.tag_shift_mode_en = true;
+
+	if (port_base_vlan_state == HNAE3_PORT_BASE_VLAN_DISABLE) {
+		vport->rxvlan_cfg.strip_tag1_en = false;
+		vport->rxvlan_cfg.strip_tag2_en =
+				vport->rxvlan_cfg.rx_vlan_offload_en;
+		vport->rxvlan_cfg.strip_tag2_discard_en = false;
+	} else {
+		vport->rxvlan_cfg.strip_tag1_en =
+				vport->rxvlan_cfg.rx_vlan_offload_en;
+		vport->rxvlan_cfg.strip_tag2_en = true;
+		vport->rxvlan_cfg.strip_tag2_discard_en = true;
+	}
+
+	vport->rxvlan_cfg.strip_tag1_discard_en = false;
+	vport->rxvlan_cfg.vlan1_vlan_prionly = false;
+	vport->rxvlan_cfg.vlan2_vlan_prionly = false;
+
+	ret = hclge_set_vlan_tx_offload_cfg(vport);
+	if (ret)
+		return ret;
+
+	return hclge_set_vlan_rx_offload_cfg(vport);
+}
+
+static int hclge_set_vlan_protocol_type(struct hclge_dev *hdev)
+{
+	struct hclge_rx_vlan_type_cfg_cmd *rx_req;
+	struct hclge_tx_vlan_type_cfg_cmd *tx_req;
+	struct hclge_desc desc;
+	int status;
+
+	hclge_cmd_setup_basic_desc(&desc, HCLGE_OPC_MAC_VLAN_TYPE_ID, false);
+	rx_req = (struct hclge_rx_vlan_type_cfg_cmd *)desc.data;
+	rx_req->ot_fst_vlan_type =
+		cpu_to_le16(hdev->vlan_type_cfg.rx_ot_fst_vlan_type);
+	rx_req->ot_sec_vlan_type =
+		cpu_to_le16(hdev->vlan_type_cfg.rx_ot_sec_vlan_type);
+	rx_req->in_fst_vlan_type =
+		cpu_to_le16(hdev->vlan_type_cfg.rx_in_fst_vlan_type);
+	rx_req->in_sec_vlan_type =
+		cpu_to_le16(hdev->vlan_type_cfg.rx_in_sec_vlan_type);
+
+	status = hclge_cmd_send(&hdev->hw, &desc, 1);
+	if (status) {
+		dev_err(&hdev->pdev->dev,
+			"Send rxvlan protocol type command fail, ret =%d\n",
+			status);
+		return status;
+	}
+
+	hclge_cmd_setup_basic_desc(&desc, HCLGE_OPC_MAC_VLAN_INSERT, false);
+
+	tx_req = (struct hclge_tx_vlan_type_cfg_cmd *)desc.data;
+	tx_req->ot_vlan_type = cpu_to_le16(hdev->vlan_type_cfg.tx_ot_vlan_type);
+	tx_req->in_vlan_type = cpu_to_le16(hdev->vlan_type_cfg.tx_in_vlan_type);
+
+	status = hclge_cmd_send(&hdev->hw, &desc, 1);
+	if (status)
+		dev_err(&hdev->pdev->dev,
+			"Send txvlan protocol type command fail, ret =%d\n",
+			status);
+
+	return status;
+}
+
+static int hclge_init_vlan_filter(struct hclge_dev *hdev)
+{
+	struct hclge_vport *vport;
+	bool enable = true;
+	int ret;
+	int i;
+
+	if (hdev->ae_dev->dev_version < HNAE3_DEVICE_VERSION_V2)
+		return hclge_set_vlan_filter_ctrl(hdev, HCLGE_FILTER_TYPE_VF,
+						  HCLGE_FILTER_FE_EGRESS_V1_B,
+						  true, 0);
+
+	/* for revision 0x21, vf vlan filter is per function */
+	for (i = 0; i < hdev->num_alloc_vport; i++) {
+		vport = &hdev->vport[i];
+		ret = hclge_set_vlan_filter_ctrl(hdev, HCLGE_FILTER_TYPE_VF,
+						 HCLGE_FILTER_FE_EGRESS, true,
+						 vport->vport_id);
+		if (ret)
+			return ret;
+		vport->cur_vlan_fltr_en = true;
+	}
+
+	if (test_bit(HNAE3_DEV_SUPPORT_VLAN_FLTR_MDF_B, hdev->ae_dev->caps) &&
+	    !test_bit(HNAE3_DEV_SUPPORT_PORT_VLAN_BYPASS_B, hdev->ae_dev->caps))
+		enable = false;
+
+	return hclge_set_vlan_filter_ctrl(hdev, HCLGE_FILTER_TYPE_PORT,
+					  HCLGE_FILTER_FE_INGRESS, enable, 0);
+}
+
+static int hclge_init_vlan_type(struct hclge_dev *hdev)
+{
+	hdev->vlan_type_cfg.rx_in_fst_vlan_type = ETH_P_8021Q;
+	hdev->vlan_type_cfg.rx_in_sec_vlan_type = ETH_P_8021Q;
+	hdev->vlan_type_cfg.rx_ot_fst_vlan_type = ETH_P_8021Q;
+	hdev->vlan_type_cfg.rx_ot_sec_vlan_type = ETH_P_8021Q;
+	hdev->vlan_type_cfg.tx_ot_vlan_type = ETH_P_8021Q;
+	hdev->vlan_type_cfg.tx_in_vlan_type = ETH_P_8021Q;
+
+	return hclge_set_vlan_protocol_type(hdev);
+}
+
+static int hclge_init_vport_vlan_offload(struct hclge_dev *hdev)
+{
+	struct hclge_port_base_vlan_config *cfg;
+	struct hclge_vport *vport;
+	int ret;
+	int i;
+
+	for (i = 0; i < hdev->num_alloc_vport; i++) {
+		vport = &hdev->vport[i];
+		cfg = &vport->port_base_vlan_cfg;
+
+		ret = hclge_vlan_offload_cfg(vport, cfg->state,
+					     cfg->vlan_info.vlan_tag,
+					     cfg->vlan_info.qos);
+		if (ret)
+			return ret;
+	}
+	return 0;
+}
+
+static int hclge_init_vlan_config(struct hclge_dev *hdev)
+{
+	struct hnae3_handle *handle = &hdev->vport[0].nic;
+	int ret;
+
+	ret = hclge_init_vlan_filter(hdev);
+	if (ret)
+		return ret;
+
+	ret = hclge_init_vlan_type(hdev);
+	if (ret)
+		return ret;
+
+	ret = hclge_init_vport_vlan_offload(hdev);
+	if (ret)
+		return ret;
+
+	return hclge_set_vlan_filter(handle, htons(ETH_P_8021Q), 0, false);
+}
+
+static void hclge_add_vport_vlan_table(struct hclge_vport *vport, u16 vlan_id,
+				       bool writen_to_tbl)
+{
+	struct hclge_vport_vlan_cfg *vlan, *tmp;
+	struct hclge_dev *hdev = vport->back;
+
+	mutex_lock(&hdev->vport_lock);
+
+	list_for_each_entry_safe(vlan, tmp, &vport->vlan_list, node) {
+		if (vlan->vlan_id == vlan_id) {
+			mutex_unlock(&hdev->vport_lock);
+			return;
+		}
+	}
+
+	vlan = kzalloc(sizeof(*vlan), GFP_KERNEL);
+	if (!vlan) {
+		mutex_unlock(&hdev->vport_lock);
+		return;
+	}
+
+	vlan->hd_tbl_status = writen_to_tbl;
+	vlan->vlan_id = vlan_id;
+
+	list_add_tail(&vlan->node, &vport->vlan_list);
+	mutex_unlock(&hdev->vport_lock);
+}
+
+static int hclge_add_vport_all_vlan_table(struct hclge_vport *vport)
+{
+	struct hclge_vport_vlan_cfg *vlan, *tmp;
+	struct hclge_dev *hdev = vport->back;
+	int ret;
+
+	mutex_lock(&hdev->vport_lock);
+
+	list_for_each_entry_safe(vlan, tmp, &vport->vlan_list, node) {
+		if (!vlan->hd_tbl_status) {
+			ret = hclge_set_vlan_filter_hw(hdev, htons(ETH_P_8021Q),
+						       vport->vport_id,
+						       vlan->vlan_id, false);
+			if (ret) {
+				dev_err(&hdev->pdev->dev,
+					"restore vport vlan list failed, ret=%d\n",
+					ret);
+
+				mutex_unlock(&hdev->vport_lock);
+				return ret;
+			}
+		}
+		vlan->hd_tbl_status = true;
+	}
+
+	mutex_unlock(&hdev->vport_lock);
+
+	return 0;
+}
+
+static void hclge_rm_vport_vlan_table(struct hclge_vport *vport, u16 vlan_id,
+				      bool is_write_tbl)
+{
+	struct hclge_vport_vlan_cfg *vlan, *tmp;
+	struct hclge_dev *hdev = vport->back;
+
+	list_for_each_entry_safe(vlan, tmp, &vport->vlan_list, node) {
+		if (vlan->vlan_id == vlan_id) {
+			if (is_write_tbl && vlan->hd_tbl_status)
+				hclge_set_vlan_filter_hw(hdev,
+							 htons(ETH_P_8021Q),
+							 vport->vport_id,
+							 vlan_id,
+							 true);
+
+			list_del(&vlan->node);
+			kfree(vlan);
+			break;
+		}
+	}
+}
+
+void hclge_rm_vport_all_vlan_table(struct hclge_vport *vport, bool is_del_list)
+{
+	struct hclge_vport_vlan_cfg *vlan, *tmp;
+	struct hclge_dev *hdev = vport->back;
+
+	mutex_lock(&hdev->vport_lock);
+
+	list_for_each_entry_safe(vlan, tmp, &vport->vlan_list, node) {
+		if (vlan->hd_tbl_status)
+			hclge_set_vlan_filter_hw(hdev,
+						 htons(ETH_P_8021Q),
+						 vport->vport_id,
+						 vlan->vlan_id,
+						 true);
+
+		vlan->hd_tbl_status = false;
+		if (is_del_list) {
+			list_del(&vlan->node);
+			kfree(vlan);
+		}
+	}
+	clear_bit(vport->vport_id, hdev->vf_vlan_full);
+	mutex_unlock(&hdev->vport_lock);
+}
+
+void hclge_uninit_vport_vlan_table(struct hclge_dev *hdev)
+{
+	struct hclge_vport_vlan_cfg *vlan, *tmp;
+	struct hclge_vport *vport;
+	int i;
+
+	mutex_lock(&hdev->vport_lock);
+
+	for (i = 0; i < hdev->num_alloc_vport; i++) {
+		vport = &hdev->vport[i];
+		list_for_each_entry_safe(vlan, tmp, &vport->vlan_list, node) {
+			list_del(&vlan->node);
+			kfree(vlan);
+		}
+	}
+
+	mutex_unlock(&hdev->vport_lock);
+}
+
+void hclge_restore_vport_port_base_vlan_config(struct hclge_dev *hdev)
+{
+	struct hclge_vlan_info *vlan_info;
+	struct hclge_vport *vport;
+	u16 vlan_proto;
+	u16 vlan_id;
+	u16 state;
+	int vf_id;
+	int ret;
+
+	/* PF should restore all vfs port base vlan */
+	for (vf_id = 0; vf_id < hdev->num_alloc_vfs; vf_id++) {
+		vport = &hdev->vport[vf_id + HCLGE_VF_VPORT_START_NUM];
+		vlan_info = vport->port_base_vlan_cfg.tbl_sta ?
+			    &vport->port_base_vlan_cfg.vlan_info :
+			    &vport->port_base_vlan_cfg.old_vlan_info;
+
+		vlan_id = vlan_info->vlan_tag;
+		vlan_proto = vlan_info->vlan_proto;
+		state = vport->port_base_vlan_cfg.state;
+
+		if (state != HNAE3_PORT_BASE_VLAN_DISABLE) {
+			clear_bit(vport->vport_id, hdev->vlan_table[vlan_id]);
+			ret = hclge_set_vlan_filter_hw(hdev, htons(vlan_proto),
+						       vport->vport_id,
+						       vlan_id, false);
+			vport->port_base_vlan_cfg.tbl_sta = ret == 0;
+		}
+	}
+}
+
+void hclge_restore_vport_vlan_table(struct hclge_vport *vport)
+{
+	struct hclge_vport_vlan_cfg *vlan, *tmp;
+	struct hclge_dev *hdev = vport->back;
+	int ret;
+
+	mutex_lock(&hdev->vport_lock);
+
+	if (vport->port_base_vlan_cfg.state == HNAE3_PORT_BASE_VLAN_DISABLE) {
+		list_for_each_entry_safe(vlan, tmp, &vport->vlan_list, node) {
+			ret = hclge_set_vlan_filter_hw(hdev, htons(ETH_P_8021Q),
+						       vport->vport_id,
+						       vlan->vlan_id, false);
+			if (ret)
+				break;
+			vlan->hd_tbl_status = true;
+		}
+	}
+
+	mutex_unlock(&hdev->vport_lock);
+}
+
+/* For global reset and imp reset, hardware will clear the mac table,
+ * so we change the mac address state from ACTIVE to TO_ADD, then they
+ * can be restored in the service task after reset complete. Furtherly,
+ * the mac addresses with state TO_DEL or DEL_FAIL are unnecessary to
+ * be restored after reset, so just remove these mac nodes from mac_list.
+ */
+static void hclge_mac_node_convert_for_reset(struct list_head *list)
+{
+	struct hclge_mac_node *mac_node, *tmp;
+
+	list_for_each_entry_safe(mac_node, tmp, list, node) {
+		if (mac_node->state == HCLGE_MAC_ACTIVE) {
+			mac_node->state = HCLGE_MAC_TO_ADD;
+		} else if (mac_node->state == HCLGE_MAC_TO_DEL) {
+			list_del(&mac_node->node);
+			kfree(mac_node);
+		}
+	}
+}
+
+void hclge_restore_mac_table_common(struct hclge_vport *vport)
+{
+	spin_lock_bh(&vport->mac_list_lock);
+
+	hclge_mac_node_convert_for_reset(&vport->uc_mac_list);
+	hclge_mac_node_convert_for_reset(&vport->mc_mac_list);
+	set_bit(HCLGE_VPORT_STATE_MAC_TBL_CHANGE, &vport->state);
+
+	spin_unlock_bh(&vport->mac_list_lock);
+}
+
+static void hclge_restore_hw_table(struct hclge_dev *hdev)
+{
+	struct hclge_vport *vport = &hdev->vport[0];
+	struct hnae3_handle *handle = &vport->nic;
+
+	hclge_restore_mac_table_common(vport);
+	hclge_restore_vport_port_base_vlan_config(hdev);
+	hclge_restore_vport_vlan_table(vport);
+	set_bit(HCLGE_STATE_FD_USER_DEF_CHANGED, &hdev->state);
+	hclge_restore_fd_entries(handle);
+}
+
+int hclge_en_hw_strip_rxvtag(struct hnae3_handle *handle, bool enable)
+{
+	struct hclge_vport *vport = hclge_get_vport(handle);
+
+	if (vport->port_base_vlan_cfg.state == HNAE3_PORT_BASE_VLAN_DISABLE) {
+		vport->rxvlan_cfg.strip_tag1_en = false;
+		vport->rxvlan_cfg.strip_tag2_en = enable;
+		vport->rxvlan_cfg.strip_tag2_discard_en = false;
+	} else {
+		vport->rxvlan_cfg.strip_tag1_en = enable;
+		vport->rxvlan_cfg.strip_tag2_en = true;
+		vport->rxvlan_cfg.strip_tag2_discard_en = true;
+	}
+
+	vport->rxvlan_cfg.strip_tag1_discard_en = false;
+	vport->rxvlan_cfg.vlan1_vlan_prionly = false;
+	vport->rxvlan_cfg.vlan2_vlan_prionly = false;
+	vport->rxvlan_cfg.rx_vlan_offload_en = enable;
+
+	return hclge_set_vlan_rx_offload_cfg(vport);
+}
+
+static void hclge_set_vport_vlan_fltr_change(struct hclge_vport *vport)
+{
+	struct hclge_dev *hdev = vport->back;
+
+	if (test_bit(HNAE3_DEV_SUPPORT_VLAN_FLTR_MDF_B, hdev->ae_dev->caps))
+		set_bit(HCLGE_VPORT_STATE_VLAN_FLTR_CHANGE, &vport->state);
+}
+
+static int hclge_update_vlan_filter_entries(struct hclge_vport *vport,
+					    u16 port_base_vlan_state,
+					    struct hclge_vlan_info *new_info,
+					    struct hclge_vlan_info *old_info)
+{
+	struct hclge_dev *hdev = vport->back;
+	int ret;
+
+	if (port_base_vlan_state == HNAE3_PORT_BASE_VLAN_ENABLE) {
+		hclge_rm_vport_all_vlan_table(vport, false);
+		/* force clear VLAN 0 */
+		ret = hclge_set_vf_vlan_common(hdev, vport->vport_id, true, 0);
+		if (ret)
+			return ret;
+		return hclge_set_vlan_filter_hw(hdev,
+						 htons(new_info->vlan_proto),
+						 vport->vport_id,
+						 new_info->vlan_tag,
+						 false);
+	}
+
+	vport->port_base_vlan_cfg.tbl_sta = false;
+
+	/* force add VLAN 0 */
+	ret = hclge_set_vf_vlan_common(hdev, vport->vport_id, false, 0);
+	if (ret)
+		return ret;
+
+	ret = hclge_set_vlan_filter_hw(hdev, htons(old_info->vlan_proto),
+				       vport->vport_id, old_info->vlan_tag,
+				       true);
+	if (ret)
+		return ret;
+
+	return hclge_add_vport_all_vlan_table(vport);
+}
+
+static bool hclge_need_update_vlan_filter(const struct hclge_vlan_info *new_cfg,
+					  const struct hclge_vlan_info *old_cfg)
+{
+	if (new_cfg->vlan_tag != old_cfg->vlan_tag)
+		return true;
+
+	if (new_cfg->vlan_tag == 0 && (new_cfg->qos == 0 || old_cfg->qos == 0))
+		return true;
+
+	return false;
+}
+
+static int hclge_modify_port_base_vlan_tag(struct hclge_vport *vport,
+					   struct hclge_vlan_info *new_info,
+					   struct hclge_vlan_info *old_info)
+{
+	struct hclge_dev *hdev = vport->back;
+	int ret;
+
+	/* add new VLAN tag */
+	ret = hclge_set_vlan_filter_hw(hdev, htons(new_info->vlan_proto),
+				       vport->vport_id, new_info->vlan_tag,
+				       false);
+	if (ret)
+		return ret;
+
+	vport->port_base_vlan_cfg.tbl_sta = false;
+	/* remove old VLAN tag */
+	if (old_info->vlan_tag == 0)
+		ret = hclge_set_vf_vlan_common(hdev, vport->vport_id,
+					       true, 0);
+	else
+		ret = hclge_set_vlan_filter_hw(hdev, htons(ETH_P_8021Q),
+					       vport->vport_id,
+					       old_info->vlan_tag, true);
+	if (ret)
+		dev_err(&hdev->pdev->dev,
+			"failed to clear vport%u port base vlan %u, ret = %d.\n",
+			vport->vport_id, old_info->vlan_tag, ret);
+
+	return ret;
+}
+
+int hclge_update_port_base_vlan_cfg(struct hclge_vport *vport, u16 state,
+				    struct hclge_vlan_info *vlan_info)
+{
+	struct hnae3_handle *nic = &vport->nic;
+	struct hclge_vlan_info *old_vlan_info;
+	int ret;
+
+	old_vlan_info = &vport->port_base_vlan_cfg.vlan_info;
+
+	ret = hclge_vlan_offload_cfg(vport, state, vlan_info->vlan_tag,
+				     vlan_info->qos);
+	if (ret)
+		return ret;
+
+	if (!hclge_need_update_vlan_filter(vlan_info, old_vlan_info))
+		goto out;
+
+	if (state == HNAE3_PORT_BASE_VLAN_MODIFY)
+		ret = hclge_modify_port_base_vlan_tag(vport, vlan_info,
+						      old_vlan_info);
+	else
+		ret = hclge_update_vlan_filter_entries(vport, state, vlan_info,
+						       old_vlan_info);
+	if (ret)
+		return ret;
+
+out:
+	vport->port_base_vlan_cfg.state = state;
+	if (state == HNAE3_PORT_BASE_VLAN_DISABLE)
+		nic->port_base_vlan_state = HNAE3_PORT_BASE_VLAN_DISABLE;
+	else
+		nic->port_base_vlan_state = HNAE3_PORT_BASE_VLAN_ENABLE;
+
+	vport->port_base_vlan_cfg.old_vlan_info = *old_vlan_info;
+	vport->port_base_vlan_cfg.vlan_info = *vlan_info;
+	vport->port_base_vlan_cfg.tbl_sta = true;
+	hclge_set_vport_vlan_fltr_change(vport);
+
+	return 0;
+}
+
+static u16 hclge_get_port_base_vlan_state(struct hclge_vport *vport,
+					  enum hnae3_port_base_vlan_state state,
+					  u16 vlan, u8 qos)
+{
+	if (state == HNAE3_PORT_BASE_VLAN_DISABLE) {
+		if (!vlan && !qos)
+			return HNAE3_PORT_BASE_VLAN_NOCHANGE;
+
+		return HNAE3_PORT_BASE_VLAN_ENABLE;
+	}
+
+	if (!vlan && !qos)
+		return HNAE3_PORT_BASE_VLAN_DISABLE;
+
+	if (vport->port_base_vlan_cfg.vlan_info.vlan_tag == vlan &&
+	    vport->port_base_vlan_cfg.vlan_info.qos == qos)
+		return HNAE3_PORT_BASE_VLAN_NOCHANGE;
+
+	return HNAE3_PORT_BASE_VLAN_MODIFY;
+}
+
+static int hclge_set_vf_vlan_filter(struct hnae3_handle *handle, int vfid,
+				    u16 vlan, u8 qos, __be16 proto)
+{
+	struct hnae3_ae_dev *ae_dev = pci_get_drvdata(handle->pdev);
+	struct hclge_vport *vport = hclge_get_vport(handle);
+	struct hclge_dev *hdev = vport->back;
+	struct hclge_vlan_info vlan_info;
+	u16 state;
+	int ret;
+
+	if (hdev->ae_dev->dev_version < HNAE3_DEVICE_VERSION_V2)
+		return -EOPNOTSUPP;
+
+	vport = hclge_get_vf_vport(hdev, vfid);
+	if (!vport)
+		return -EINVAL;
+
+	/* qos is a 3 bits value, so can not be bigger than 7 */
+	if (vlan > VLAN_N_VID - 1 || qos > 7)
+		return -EINVAL;
+	if (proto != htons(ETH_P_8021Q))
+		return -EPROTONOSUPPORT;
+
+	state = hclge_get_port_base_vlan_state(vport,
+					       vport->port_base_vlan_cfg.state,
+					       vlan, qos);
+	if (state == HNAE3_PORT_BASE_VLAN_NOCHANGE)
+		return 0;
+
+	vlan_info.vlan_tag = vlan;
+	vlan_info.qos = qos;
+	vlan_info.vlan_proto = ntohs(proto);
+
+	ret = hclge_update_port_base_vlan_cfg(vport, state, &vlan_info);
+	if (ret) {
+		dev_err(&hdev->pdev->dev,
+			"failed to update port base vlan for vf %d, ret = %d\n",
+			vfid, ret);
+		return ret;
+	}
+
+	/* there is a timewindow for PF to know VF unalive, it may
+	 * cause send mailbox fail, but it doesn't matter, VF will
+	 * query it when reinit.
+	 * for DEVICE_VERSION_V3, vf doesn't need to know about the port based
+	 * VLAN state.
+	 */
+	if (ae_dev->dev_version < HNAE3_DEVICE_VERSION_V3) {
+		if (test_bit(HCLGE_VPORT_STATE_ALIVE, &vport->state))
+			(void)hclge_push_vf_port_base_vlan_info(&hdev->vport[0],
+								vport->vport_id,
+								state,
+								&vlan_info);
+		else
+			set_bit(HCLGE_VPORT_NEED_NOTIFY_VF_VLAN,
+				&vport->need_notify);
+	}
+	return 0;
+}
+
+static void hclge_clear_vf_vlan(struct hclge_dev *hdev)
+{
+	struct hclge_vlan_info *vlan_info;
+	struct hclge_vport *vport;
+	int ret;
+	int vf;
+
+	/* clear port base vlan for all vf */
+	for (vf = HCLGE_VF_VPORT_START_NUM; vf < hdev->num_alloc_vport; vf++) {
+		vport = &hdev->vport[vf];
+		vlan_info = &vport->port_base_vlan_cfg.vlan_info;
+
+		ret = hclge_set_vlan_filter_hw(hdev, htons(ETH_P_8021Q),
+					       vport->vport_id,
+					       vlan_info->vlan_tag, true);
+		if (ret)
+			dev_err(&hdev->pdev->dev,
+				"failed to clear vf vlan for vf%d, ret = %d\n",
+				vf - HCLGE_VF_VPORT_START_NUM, ret);
+	}
+}
+
+int hclge_set_vlan_filter(struct hnae3_handle *handle, __be16 proto,
+			  u16 vlan_id, bool is_kill)
+{
+	struct hclge_vport *vport = hclge_get_vport(handle);
+	struct hclge_dev *hdev = vport->back;
+	bool writen_to_tbl = false;
+	int ret = 0;
+
+	/* When device is resetting or reset failed, firmware is unable to
+	 * handle mailbox. Just record the vlan id, and remove it after
+	 * reset finished.
+	 */
+	mutex_lock(&hdev->vport_lock);
+	if ((test_bit(HCLGE_STATE_RST_HANDLING, &hdev->state) ||
+	     test_bit(HCLGE_STATE_RST_FAIL, &hdev->state)) && is_kill) {
+		set_bit(vlan_id, vport->vlan_del_fail_bmap);
+		mutex_unlock(&hdev->vport_lock);
+		return -EBUSY;
+	} else if (!is_kill && test_bit(vlan_id, vport->vlan_del_fail_bmap)) {
+		clear_bit(vlan_id, vport->vlan_del_fail_bmap);
+	}
+	mutex_unlock(&hdev->vport_lock);
+
+	/* when port base vlan enabled, we use port base vlan as the vlan
+	 * filter entry. In this case, we don't update vlan filter table
+	 * when user add new vlan or remove exist vlan, just update the vport
+	 * vlan list. The vlan id in vlan list will be writen in vlan filter
+	 * table until port base vlan disabled
+	 */
+	if (handle->port_base_vlan_state == HNAE3_PORT_BASE_VLAN_DISABLE) {
+		ret = hclge_set_vlan_filter_hw(hdev, proto, vport->vport_id,
+					       vlan_id, is_kill);
+		writen_to_tbl = true;
+	}
+
+	if (!ret) {
+		if (!is_kill) {
+			hclge_add_vport_vlan_table(vport, vlan_id,
+						   writen_to_tbl);
+		} else if (is_kill && vlan_id != 0) {
+			mutex_lock(&hdev->vport_lock);
+			hclge_rm_vport_vlan_table(vport, vlan_id, false);
+			mutex_unlock(&hdev->vport_lock);
+		}
+	} else if (is_kill) {
+		/* when remove hw vlan filter failed, record the vlan id,
+		 * and try to remove it from hw later, to be consistence
+		 * with stack
+		 */
+		mutex_lock(&hdev->vport_lock);
+		set_bit(vlan_id, vport->vlan_del_fail_bmap);
+		mutex_unlock(&hdev->vport_lock);
+	}
+
+	hclge_set_vport_vlan_fltr_change(vport);
+
+	return ret;
+}
+
+static void hclge_sync_vlan_fltr_state(struct hclge_dev *hdev)
+{
+	struct hclge_vport *vport;
+	int ret;
+	u16 i;
+
+	for (i = 0; i < hdev->num_alloc_vport; i++) {
+		vport = &hdev->vport[i];
+		if (!test_and_clear_bit(HCLGE_VPORT_STATE_VLAN_FLTR_CHANGE,
+					&vport->state))
+			continue;
+
+		mutex_lock(&hdev->vport_lock);
+		ret = __hclge_enable_vport_vlan_filter(vport,
+						       vport->req_vlan_fltr_en);
+		if (ret) {
+			dev_err(&hdev->pdev->dev,
+				"failed to sync vlan filter state for vport%u, ret = %d\n",
+				vport->vport_id, ret);
+			set_bit(HCLGE_VPORT_STATE_VLAN_FLTR_CHANGE,
+				&vport->state);
+			mutex_unlock(&hdev->vport_lock);
+			return;
+		}
+		mutex_unlock(&hdev->vport_lock);
+	}
+}
+
+static void hclge_sync_vlan_filter(struct hclge_dev *hdev)
+{
+#define HCLGE_MAX_SYNC_COUNT	60
+
+	int i, ret, sync_cnt = 0;
+	u16 vlan_id;
+
+	mutex_lock(&hdev->vport_lock);
+	/* start from vport 1 for PF is always alive */
+	for (i = 0; i < hdev->num_alloc_vport; i++) {
+		struct hclge_vport *vport = &hdev->vport[i];
+
+		vlan_id = find_first_bit(vport->vlan_del_fail_bmap,
+					 VLAN_N_VID);
+		while (vlan_id != VLAN_N_VID) {
+			ret = hclge_set_vlan_filter_hw(hdev, htons(ETH_P_8021Q),
+						       vport->vport_id, vlan_id,
+						       true);
+			if (ret && ret != -EINVAL) {
+				mutex_unlock(&hdev->vport_lock);
+				return;
+			}
+
+			clear_bit(vlan_id, vport->vlan_del_fail_bmap);
+			hclge_rm_vport_vlan_table(vport, vlan_id, false);
+			hclge_set_vport_vlan_fltr_change(vport);
+
+			sync_cnt++;
+			if (sync_cnt >= HCLGE_MAX_SYNC_COUNT) {
+				mutex_unlock(&hdev->vport_lock);
+				return;
+			}
+
+			vlan_id = find_first_bit(vport->vlan_del_fail_bmap,
+						 VLAN_N_VID);
+		}
+	}
+	mutex_unlock(&hdev->vport_lock);
+
+	hclge_sync_vlan_fltr_state(hdev);
+}
+
+static int hclge_set_mac_mtu(struct hclge_dev *hdev, int new_mps)
+{
+	struct hclge_config_max_frm_size_cmd *req;
+	struct hclge_desc desc;
+
+	hclge_cmd_setup_basic_desc(&desc, HCLGE_OPC_CONFIG_MAX_FRM_SIZE, false);
+
+	req = (struct hclge_config_max_frm_size_cmd *)desc.data;
+	req->max_frm_size = cpu_to_le16(new_mps);
+	req->min_frm_size = HCLGE_MAC_MIN_FRAME;
+
+	return hclge_cmd_send(&hdev->hw, &desc, 1);
+}
+
+static int hclge_set_mtu(struct hnae3_handle *handle, int new_mtu)
+{
+	struct hclge_vport *vport = hclge_get_vport(handle);
+
+	return hclge_set_vport_mtu(vport, new_mtu);
+}
+
+int hclge_set_vport_mtu(struct hclge_vport *vport, int new_mtu)
+{
+	struct hclge_dev *hdev = vport->back;
+	int i, max_frm_size, ret;
+
+	/* HW supprt 2 layer vlan */
+	max_frm_size = new_mtu + ETH_HLEN + ETH_FCS_LEN + 2 * VLAN_HLEN;
+	if (max_frm_size < HCLGE_MAC_MIN_FRAME ||
+	    max_frm_size > hdev->ae_dev->dev_specs.max_frm_size)
+		return -EINVAL;
+
+	max_frm_size = max(max_frm_size, HCLGE_MAC_DEFAULT_FRAME);
+	mutex_lock(&hdev->vport_lock);
+	/* VF's mps must fit within hdev->mps */
+	if (vport->vport_id && (u32)max_frm_size > hdev->mps) {
+		mutex_unlock(&hdev->vport_lock);
+		return -EINVAL;
+	} else if (vport->vport_id) {
+		vport->mps = max_frm_size;
+		mutex_unlock(&hdev->vport_lock);
+		return 0;
+	}
+
+	/* PF's mps must be greater then VF's mps */
+	for (i = 1; i < hdev->num_alloc_vport; i++)
+		if ((u32)max_frm_size < hdev->vport[i].mps) {
+			dev_err(&hdev->pdev->dev,
+				"failed to set pf mtu for less than vport %d, mps = %u.\n",
+				i, hdev->vport[i].mps);
+			mutex_unlock(&hdev->vport_lock);
+			return -EINVAL;
+		}
+
+	hclge_notify_client(hdev, HNAE3_DOWN_CLIENT);
+
+	ret = hclge_set_mac_mtu(hdev, max_frm_size);
+	if (ret) {
+		dev_err(&hdev->pdev->dev,
+			"Change mtu fail, ret =%d\n", ret);
+		goto out;
+	}
+
+	hdev->mps = max_frm_size;
+	vport->mps = max_frm_size;
+
+	ret = hclge_buffer_alloc(hdev);
+	if (ret)
+		dev_err(&hdev->pdev->dev,
+			"Allocate buffer fail, ret =%d\n", ret);
+
+out:
+	hclge_notify_client(hdev, HNAE3_UP_CLIENT);
+	mutex_unlock(&hdev->vport_lock);
+	return ret;
+}
+
+static int hclge_reset_tqp_cmd_send(struct hclge_dev *hdev, u16 queue_id,
+				    bool enable)
+{
+	struct hclge_reset_tqp_queue_cmd *req;
+	struct hclge_desc desc;
+	int ret;
+
+	hclge_cmd_setup_basic_desc(&desc, HCLGE_OPC_RESET_TQP_QUEUE, false);
+
+	req = (struct hclge_reset_tqp_queue_cmd *)desc.data;
+	req->tqp_id = cpu_to_le16(queue_id);
+	if (enable)
+		hnae3_set_bit(req->reset_req, HCLGE_TQP_RESET_B, 1U);
+
+	ret = hclge_cmd_send(&hdev->hw, &desc, 1);
+	if (ret) {
+		dev_err(&hdev->pdev->dev,
+			"Send tqp reset cmd error, status =%d\n", ret);
+		return ret;
+	}
+
+	return 0;
+}
+
+static int hclge_get_reset_status(struct hclge_dev *hdev, u16 queue_id,
+				  u8 *reset_status)
+{
+	struct hclge_reset_tqp_queue_cmd *req;
+	struct hclge_desc desc;
+	int ret;
+
+	hclge_cmd_setup_basic_desc(&desc, HCLGE_OPC_RESET_TQP_QUEUE, true);
+
+	req = (struct hclge_reset_tqp_queue_cmd *)desc.data;
+	req->tqp_id = cpu_to_le16(queue_id);
+
+	ret = hclge_cmd_send(&hdev->hw, &desc, 1);
+	if (ret) {
+		dev_err(&hdev->pdev->dev,
+			"Get reset status error, status =%d\n", ret);
+		return ret;
+	}
+
+	*reset_status = hnae3_get_bit(req->ready_to_reset, HCLGE_TQP_RESET_B);
+
+	return 0;
+}
+
+u16 hclge_covert_handle_qid_global(struct hnae3_handle *handle, u16 queue_id)
+{
+	struct hclge_comm_tqp *tqp;
+	struct hnae3_queue *queue;
+
+	queue = handle->kinfo.tqp[queue_id];
+	tqp = container_of(queue, struct hclge_comm_tqp, q);
+
+	return tqp->index;
+}
+
+static int hclge_reset_tqp_cmd(struct hnae3_handle *handle)
+{
+	struct hclge_vport *vport = hclge_get_vport(handle);
+	struct hclge_dev *hdev = vport->back;
+	u16 reset_try_times = 0;
+	u8 reset_status;
+	u16 queue_gid;
+	int ret;
+	u16 i;
+
+	for (i = 0; i < handle->kinfo.num_tqps; i++) {
+		queue_gid = hclge_covert_handle_qid_global(handle, i);
+		ret = hclge_reset_tqp_cmd_send(hdev, queue_gid, true);
+		if (ret) {
+			dev_err(&hdev->pdev->dev,
+				"failed to send reset tqp cmd, ret = %d\n",
+				ret);
+			return ret;
+		}
+
+		while (reset_try_times++ < HCLGE_TQP_RESET_TRY_TIMES) {
+			ret = hclge_get_reset_status(hdev, queue_gid,
+						     &reset_status);
+			if (ret)
+				return ret;
+
+			if (reset_status)
+				break;
+
+			/* Wait for tqp hw reset */
+			usleep_range(1000, 1200);
+		}
+
+		if (reset_try_times >= HCLGE_TQP_RESET_TRY_TIMES) {
+			dev_err(&hdev->pdev->dev,
+				"wait for tqp hw reset timeout\n");
+			return -ETIME;
+		}
+
+		ret = hclge_reset_tqp_cmd_send(hdev, queue_gid, false);
+		if (ret) {
+			dev_err(&hdev->pdev->dev,
+				"failed to deassert soft reset, ret = %d\n",
+				ret);
+			return ret;
+		}
+		reset_try_times = 0;
+	}
+	return 0;
+}
+
+static int hclge_reset_rcb(struct hnae3_handle *handle)
+{
+#define HCLGE_RESET_RCB_NOT_SUPPORT	0U
+#define HCLGE_RESET_RCB_SUCCESS		1U
+
+	struct hclge_vport *vport = hclge_get_vport(handle);
+	struct hclge_dev *hdev = vport->back;
+	struct hclge_reset_cmd *req;
+	struct hclge_desc desc;
+	u8 return_status;
+	u16 queue_gid;
+	int ret;
+
+	queue_gid = hclge_covert_handle_qid_global(handle, 0);
+
+	req = (struct hclge_reset_cmd *)desc.data;
+	hclge_cmd_setup_basic_desc(&desc, HCLGE_OPC_CFG_RST_TRIGGER, false);
+	hnae3_set_bit(req->fun_reset_rcb, HCLGE_CFG_RESET_RCB_B, 1);
+	req->fun_reset_rcb_vqid_start = cpu_to_le16(queue_gid);
+	req->fun_reset_rcb_vqid_num = cpu_to_le16(handle->kinfo.num_tqps);
+
+	ret = hclge_cmd_send(&hdev->hw, &desc, 1);
+	if (ret) {
+		dev_err(&hdev->pdev->dev,
+			"failed to send rcb reset cmd, ret = %d\n", ret);
+		return ret;
+	}
+
+	return_status = req->fun_reset_rcb_return_status;
+	if (return_status == HCLGE_RESET_RCB_SUCCESS)
+		return 0;
+
+	if (return_status != HCLGE_RESET_RCB_NOT_SUPPORT) {
+		dev_err(&hdev->pdev->dev, "failed to reset rcb, ret = %u\n",
+			return_status);
+		return -EIO;
+	}
+
+	/* if reset rcb cmd is unsupported, we need to send reset tqp cmd
+	 * again to reset all tqps
+	 */
+	return hclge_reset_tqp_cmd(handle);
+}
+
+int hclge_reset_tqp(struct hnae3_handle *handle)
+{
+	struct hclge_vport *vport = hclge_get_vport(handle);
+	struct hclge_dev *hdev = vport->back;
+	int ret;
+
+	/* only need to disable PF's tqp */
+	if (!vport->vport_id) {
+		ret = hclge_tqp_enable(handle, false);
+		if (ret) {
+			dev_err(&hdev->pdev->dev,
+				"failed to disable tqp, ret = %d\n", ret);
+			return ret;
+		}
+	}
+
+	return hclge_reset_rcb(handle);
+}
+
+static u32 hclge_get_fw_version(struct hnae3_handle *handle)
+{
+	struct hclge_vport *vport = hclge_get_vport(handle);
+	struct hclge_dev *hdev = vport->back;
+
+	return hdev->fw_version;
+}
+
+int hclge_query_scc_version(struct hclge_dev *hdev, u32 *scc_version)
+{
+	struct hclge_comm_query_scc_cmd *resp;
+	struct hclge_desc desc;
+	int ret;
+
+	hclge_cmd_setup_basic_desc(&desc, HCLGE_OPC_QUERY_SCC_VER, 1);
+	resp = (struct hclge_comm_query_scc_cmd *)desc.data;
+
+	ret = hclge_cmd_send(&hdev->hw, &desc, 1);
+	if (ret)
+		return ret;
+
+	*scc_version = le32_to_cpu(resp->scc_version);
+
+	return 0;
+}
+
+static void hclge_set_flowctrl_adv(struct hclge_dev *hdev, u32 rx_en, u32 tx_en)
+{
+	struct phy_device *phydev = hdev->hw.mac.phydev;
+
+	if (!phydev)
+		return;
+
+	phy_set_asym_pause(phydev, rx_en, tx_en);
+}
+
+static int hclge_cfg_pauseparam(struct hclge_dev *hdev, u32 rx_en, u32 tx_en)
+{
+	int ret;
+
+	if (hdev->tm_info.fc_mode == HCLGE_FC_PFC)
+		return 0;
+
+	ret = hclge_mac_pause_en_cfg(hdev, tx_en, rx_en);
+	if (ret)
+		dev_err(&hdev->pdev->dev,
+			"configure pauseparam error, ret = %d.\n", ret);
+
+	return ret;
+}
+
+int hclge_cfg_flowctrl(struct hclge_dev *hdev)
+{
+	struct phy_device *phydev = hdev->hw.mac.phydev;
+	u16 remote_advertising = 0;
+	u16 local_advertising;
+	u32 rx_pause, tx_pause;
+	u8 flowctl;
+
+	if (!phydev->link)
+		return 0;
+
+	if (!phydev->autoneg)
+		return hclge_mac_pause_setup_hw(hdev);
+
+	local_advertising = linkmode_adv_to_lcl_adv_t(phydev->advertising);
+
+	if (phydev->pause)
+		remote_advertising = LPA_PAUSE_CAP;
+
+	if (phydev->asym_pause)
+		remote_advertising |= LPA_PAUSE_ASYM;
+
+	flowctl = mii_resolve_flowctrl_fdx(local_advertising,
+					   remote_advertising);
+	tx_pause = flowctl & FLOW_CTRL_TX;
+	rx_pause = flowctl & FLOW_CTRL_RX;
+
+	if (phydev->duplex == HCLGE_MAC_HALF) {
+		tx_pause = 0;
+		rx_pause = 0;
+	}
+
+	return hclge_cfg_pauseparam(hdev, rx_pause, tx_pause);
+}
+
+static void hclge_get_pauseparam(struct hnae3_handle *handle, u32 *auto_neg,
+				 u32 *rx_en, u32 *tx_en)
+{
+	struct hclge_vport *vport = hclge_get_vport(handle);
+	struct hclge_dev *hdev = vport->back;
+	u8 media_type = hdev->hw.mac.media_type;
+
+	*auto_neg = (media_type == HNAE3_MEDIA_TYPE_COPPER) ?
+		    hclge_get_autoneg(handle) : 0;
+
+	if (hdev->tm_info.fc_mode == HCLGE_FC_PFC) {
+		*rx_en = 0;
+		*tx_en = 0;
+		return;
+	}
+
+	if (hdev->tm_info.fc_mode == HCLGE_FC_RX_PAUSE) {
+		*rx_en = 1;
+		*tx_en = 0;
+	} else if (hdev->tm_info.fc_mode == HCLGE_FC_TX_PAUSE) {
+		*tx_en = 1;
+		*rx_en = 0;
+	} else if (hdev->tm_info.fc_mode == HCLGE_FC_FULL) {
+		*rx_en = 1;
+		*tx_en = 1;
+	} else {
+		*rx_en = 0;
+		*tx_en = 0;
+	}
+}
+
+static void hclge_record_user_pauseparam(struct hclge_dev *hdev,
+					 u32 rx_en, u32 tx_en)
+{
+	if (rx_en && tx_en)
+		hdev->fc_mode_last_time = HCLGE_FC_FULL;
+	else if (rx_en && !tx_en)
+		hdev->fc_mode_last_time = HCLGE_FC_RX_PAUSE;
+	else if (!rx_en && tx_en)
+		hdev->fc_mode_last_time = HCLGE_FC_TX_PAUSE;
+	else
+		hdev->fc_mode_last_time = HCLGE_FC_NONE;
+
+	hdev->tm_info.fc_mode = hdev->fc_mode_last_time;
+}
+
+static int hclge_set_pauseparam(struct hnae3_handle *handle, u32 auto_neg,
+				u32 rx_en, u32 tx_en)
+{
+	struct hclge_vport *vport = hclge_get_vport(handle);
+	struct hclge_dev *hdev = vport->back;
+	struct phy_device *phydev = hdev->hw.mac.phydev;
+	u32 fc_autoneg;
+
+	if (phydev || hnae3_dev_phy_imp_supported(hdev)) {
+		fc_autoneg = hclge_get_autoneg(handle);
+		if (auto_neg != fc_autoneg) {
+			dev_info(&hdev->pdev->dev,
+				 "To change autoneg please use: ethtool -s <dev> autoneg <on|off>\n");
+			return -EOPNOTSUPP;
+		}
+	}
+
+	if (hdev->tm_info.fc_mode == HCLGE_FC_PFC) {
+		dev_info(&hdev->pdev->dev,
+			 "Priority flow control enabled. Cannot set link flow control.\n");
+		return -EOPNOTSUPP;
+	}
+
+	hclge_set_flowctrl_adv(hdev, rx_en, tx_en);
+
+	hclge_record_user_pauseparam(hdev, rx_en, tx_en);
+
+	if (!auto_neg || hnae3_dev_phy_imp_supported(hdev))
+		return hclge_cfg_pauseparam(hdev, rx_en, tx_en);
+
+	if (phydev)
+		return phy_start_aneg(phydev);
+
+	return -EOPNOTSUPP;
+}
+
+static void hclge_get_ksettings_an_result(struct hnae3_handle *handle,
+					  u8 *auto_neg, u32 *speed, u8 *duplex, u32 *lane_num)
+{
+	struct hclge_vport *vport = hclge_get_vport(handle);
+	struct hclge_dev *hdev = vport->back;
+
+	if (speed)
+		*speed = hdev->hw.mac.speed;
+	if (duplex)
+		*duplex = hdev->hw.mac.duplex;
+	if (auto_neg)
+		*auto_neg = hdev->hw.mac.autoneg;
+	if (lane_num)
+		*lane_num = hdev->hw.mac.lane_num;
+}
+
+static void hclge_get_media_type(struct hnae3_handle *handle, u8 *media_type,
+				 u8 *module_type)
+{
+	struct hclge_vport *vport = hclge_get_vport(handle);
+	struct hclge_dev *hdev = vport->back;
+
+	/* When nic is down, the service task is not running, doesn't update
+	 * the port information per second. Query the port information before
+	 * return the media type, ensure getting the correct media information.
+	 */
+	hclge_update_port_info(hdev);
+
+	if (media_type)
+		*media_type = hdev->hw.mac.media_type;
+
+	if (module_type)
+		*module_type = hdev->hw.mac.module_type;
+}
+
+static void hclge_get_mdix_mode(struct hnae3_handle *handle,
+				u8 *tp_mdix_ctrl, u8 *tp_mdix)
+{
+	struct hclge_vport *vport = hclge_get_vport(handle);
+	struct hclge_dev *hdev = vport->back;
+	struct phy_device *phydev = hdev->hw.mac.phydev;
+	int mdix_ctrl, mdix, is_resolved;
+	unsigned int retval;
+
+	if (!phydev) {
+		*tp_mdix_ctrl = ETH_TP_MDI_INVALID;
+		*tp_mdix = ETH_TP_MDI_INVALID;
+		return;
+	}
+
+	phy_write(phydev, HCLGE_PHY_PAGE_REG, HCLGE_PHY_PAGE_MDIX);
+
+	retval = phy_read(phydev, HCLGE_PHY_CSC_REG);
+	mdix_ctrl = hnae3_get_field(retval, HCLGE_PHY_MDIX_CTRL_M,
+				    HCLGE_PHY_MDIX_CTRL_S);
+
+	retval = phy_read(phydev, HCLGE_PHY_CSS_REG);
+	mdix = hnae3_get_bit(retval, HCLGE_PHY_MDIX_STATUS_B);
+	is_resolved = hnae3_get_bit(retval, HCLGE_PHY_SPEED_DUP_RESOLVE_B);
+
+	phy_write(phydev, HCLGE_PHY_PAGE_REG, HCLGE_PHY_PAGE_COPPER);
+
+	switch (mdix_ctrl) {
+	case 0x0:
+		*tp_mdix_ctrl = ETH_TP_MDI;
+		break;
+	case 0x1:
+		*tp_mdix_ctrl = ETH_TP_MDI_X;
+		break;
+	case 0x3:
+		*tp_mdix_ctrl = ETH_TP_MDI_AUTO;
+		break;
+	default:
+		*tp_mdix_ctrl = ETH_TP_MDI_INVALID;
+		break;
+	}
+
+	if (!is_resolved)
+		*tp_mdix = ETH_TP_MDI_INVALID;
+	else if (mdix)
+		*tp_mdix = ETH_TP_MDI_X;
+	else
+		*tp_mdix = ETH_TP_MDI;
+}
+
+static void hclge_info_show(struct hclge_dev *hdev)
+{
+	struct hnae3_handle *handle = &hdev->vport->nic;
+	struct device *dev = &hdev->pdev->dev;
+
+	dev_info(dev, "PF info begin:\n");
+
+	dev_info(dev, "Task queue pairs numbers: %u\n", hdev->num_tqps);
+	dev_info(dev, "Desc num per TX queue: %u\n", hdev->num_tx_desc);
+	dev_info(dev, "Desc num per RX queue: %u\n", hdev->num_rx_desc);
+	dev_info(dev, "Numbers of vports: %u\n", hdev->num_alloc_vport);
+	dev_info(dev, "Numbers of VF for this PF: %u\n", hdev->num_req_vfs);
+	dev_info(dev, "HW tc map: 0x%x\n", hdev->hw_tc_map);
+	dev_info(dev, "Total buffer size for TX/RX: %u\n", hdev->pkt_buf_size);
+	dev_info(dev, "TX buffer size for each TC: %u\n", hdev->tx_buf_size);
+	dev_info(dev, "DV buffer size for each TC: %u\n", hdev->dv_buf_size);
+	dev_info(dev, "This is %s PF\n",
+		 hdev->flag & HCLGE_FLAG_MAIN ? "main" : "not main");
+	dev_info(dev, "DCB %s\n",
+		 str_enable_disable(handle->kinfo.tc_info.dcb_ets_active));
+	dev_info(dev, "MQPRIO %s\n",
+		 str_enable_disable(handle->kinfo.tc_info.mqprio_active));
+	dev_info(dev, "Default tx spare buffer size: %u\n",
+		 hdev->tx_spare_buf_size);
+
+	dev_info(dev, "PF info end.\n");
+}
+
+static int hclge_init_nic_client_instance(struct hnae3_ae_dev *ae_dev,
+					  struct hclge_vport *vport)
+{
+	struct hnae3_client *client = vport->nic.client;
+	struct hclge_dev *hdev = ae_dev->priv;
+	u32 rst_cnt = hdev->rst_stats.reset_cnt;
+	int ret;
+
+	ret = client->ops->init_instance(&vport->nic);
+	if (ret)
+		return ret;
+
+	set_bit(HCLGE_STATE_NIC_REGISTERED, &hdev->state);
+	if (test_bit(HCLGE_STATE_RST_HANDLING, &hdev->state) ||
+	    rst_cnt != hdev->rst_stats.reset_cnt) {
+		ret = -EBUSY;
+		goto init_nic_err;
+	}
+
+	/* Enable nic hw error interrupts */
+	ret = hclge_config_nic_hw_error(hdev, true);
+	if (ret) {
+		dev_err(&ae_dev->pdev->dev,
+			"fail(%d) to enable hw error interrupts\n", ret);
+		goto init_nic_err;
+	}
+
+	hnae3_set_client_init_flag(client, ae_dev, 1);
+
+	if (netif_msg_drv(&hdev->vport->nic))
+		hclge_info_show(hdev);
+
+	return ret;
+
+init_nic_err:
+	clear_bit(HCLGE_STATE_NIC_REGISTERED, &hdev->state);
+	while (test_bit(HCLGE_STATE_RST_HANDLING, &hdev->state))
+		msleep(HCLGE_WAIT_RESET_DONE);
+
+	client->ops->uninit_instance(&vport->nic, 0);
+
+	return ret;
+}
+
+static int hclge_init_roce_client_instance(struct hnae3_ae_dev *ae_dev,
+					   struct hclge_vport *vport)
+{
+	struct hclge_dev *hdev = ae_dev->priv;
+	struct hnae3_client *client;
+	u32 rst_cnt;
+	int ret;
+
+	if (!hnae3_dev_roce_supported(hdev) || !hdev->roce_client ||
+	    !hdev->nic_client)
+		return 0;
+
+	client = hdev->roce_client;
+	ret = hclge_init_roce_base_info(vport);
+	if (ret)
+		return ret;
+
+	rst_cnt = hdev->rst_stats.reset_cnt;
+	ret = client->ops->init_instance(&vport->roce);
+	if (ret)
+		return ret;
+
+	set_bit(HCLGE_STATE_ROCE_REGISTERED, &hdev->state);
+	if (test_bit(HCLGE_STATE_RST_HANDLING, &hdev->state) ||
+	    rst_cnt != hdev->rst_stats.reset_cnt) {
+		ret = -EBUSY;
+		goto init_roce_err;
+	}
+
+	/* Enable roce ras interrupts */
+	ret = hclge_config_rocee_ras_interrupt(hdev, true);
+	if (ret) {
+		dev_err(&ae_dev->pdev->dev,
+			"fail(%d) to enable roce ras interrupts\n", ret);
+		goto init_roce_err;
+	}
+
+	hnae3_set_client_init_flag(client, ae_dev, 1);
+
+	return 0;
+
+init_roce_err:
+	clear_bit(HCLGE_STATE_ROCE_REGISTERED, &hdev->state);
+	while (test_bit(HCLGE_STATE_RST_HANDLING, &hdev->state))
+		msleep(HCLGE_WAIT_RESET_DONE);
+
+	hdev->roce_client->ops->uninit_instance(&vport->roce, 0);
+
+	return ret;
+}
+
+static int hclge_init_client_instance(struct hnae3_client *client,
+				      struct hnae3_ae_dev *ae_dev)
+{
+	struct hclge_dev *hdev = ae_dev->priv;
+	struct hclge_vport *vport = &hdev->vport[0];
+	int ret;
+
+	switch (client->type) {
+	case HNAE3_CLIENT_KNIC:
+		hdev->nic_client = client;
+		vport->nic.client = client;
+		ret = hclge_init_nic_client_instance(ae_dev, vport);
+		if (ret)
+			goto clear_nic;
+
+		ret = hclge_init_roce_client_instance(ae_dev, vport);
+		if (ret)
+			goto clear_roce;
+
+		break;
+	case HNAE3_CLIENT_ROCE:
+		if (hnae3_dev_roce_supported(hdev)) {
+			hdev->roce_client = client;
+			vport->roce.client = client;
+		}
+
+		ret = hclge_init_roce_client_instance(ae_dev, vport);
+		if (ret)
+			goto clear_roce;
+
+		break;
+	default:
+		return -EINVAL;
+	}
+
+	return 0;
+
+clear_nic:
+	hdev->nic_client = NULL;
+	vport->nic.client = NULL;
+	return ret;
+clear_roce:
+	hdev->roce_client = NULL;
+	vport->roce.client = NULL;
+	return ret;
+}
+
+static bool hclge_uninit_need_wait(struct hclge_dev *hdev)
+{
+	return test_bit(HCLGE_STATE_RST_HANDLING, &hdev->state) ||
+	       test_bit(HCLGE_STATE_LINK_UPDATING, &hdev->state);
+}
+
+static void hclge_uninit_client_instance(struct hnae3_client *client,
+					 struct hnae3_ae_dev *ae_dev)
+{
+	struct hclge_dev *hdev = ae_dev->priv;
+	struct hclge_vport *vport = &hdev->vport[0];
+
+	if (hdev->roce_client) {
+		clear_bit(HCLGE_STATE_ROCE_REGISTERED, &hdev->state);
+		while (hclge_uninit_need_wait(hdev))
+			msleep(HCLGE_WAIT_RESET_DONE);
+
+		hdev->roce_client->ops->uninit_instance(&vport->roce, 0);
+		hdev->roce_client = NULL;
+		vport->roce.client = NULL;
+	}
+	if (client->type == HNAE3_CLIENT_ROCE)
+		return;
+	if (hdev->nic_client && client->ops->uninit_instance) {
+		clear_bit(HCLGE_STATE_NIC_REGISTERED, &hdev->state);
+		while (test_bit(HCLGE_STATE_RST_HANDLING, &hdev->state))
+			msleep(HCLGE_WAIT_RESET_DONE);
+
+		client->ops->uninit_instance(&vport->nic, 0);
+		hdev->nic_client = NULL;
+		vport->nic.client = NULL;
+	}
+}
+
+static int hclge_dev_mem_map(struct hclge_dev *hdev)
+{
+	struct pci_dev *pdev = hdev->pdev;
+	struct hclge_hw *hw = &hdev->hw;
+
+	/* for device does not have device memory, return directly */
+	if (!(pci_select_bars(pdev, IORESOURCE_MEM) & BIT(HCLGE_MEM_BAR)))
+		return 0;
+
+	hw->hw.mem_base =
+		devm_ioremap_wc(&pdev->dev,
+				pci_resource_start(pdev, HCLGE_MEM_BAR),
+				pci_resource_len(pdev, HCLGE_MEM_BAR));
+	if (!hw->hw.mem_base) {
+		dev_err(&pdev->dev, "failed to map device memory\n");
+		return -EFAULT;
+	}
+
+	return 0;
+}
+
+static int hclge_pci_init(struct hclge_dev *hdev)
+{
+	struct pci_dev *pdev = hdev->pdev;
+	struct hclge_hw *hw;
+	int ret;
+
+	ret = pci_enable_device(pdev);
+	if (ret) {
+		dev_err(&pdev->dev, "failed to enable PCI device\n");
+		return ret;
+	}
+
+	ret = dma_set_mask_and_coherent(&pdev->dev, DMA_BIT_MASK(64));
+	if (ret) {
+		ret = dma_set_mask_and_coherent(&pdev->dev, DMA_BIT_MASK(32));
+		if (ret) {
+			dev_err(&pdev->dev,
+				"can't set consistent PCI DMA\n");
+			goto err_disable_device;
+		}
+		dev_warn(&pdev->dev, "set DMA mask to 32 bits\n");
+	}
+
+	ret = pci_request_regions(pdev, HCLGE_DRIVER_NAME);
+	if (ret) {
+		dev_err(&pdev->dev, "PCI request regions failed %d\n", ret);
+		goto err_disable_device;
+	}
+
+	pci_set_master(pdev);
+	hw = &hdev->hw;
+	hw->hw.io_base = pcim_iomap(pdev, 2, 0);
+	if (!hw->hw.io_base) {
+		dev_err(&pdev->dev, "Can't map configuration register space\n");
+		ret = -ENOMEM;
+		goto err_release_regions;
+	}
+
+	ret = hclge_dev_mem_map(hdev);
+	if (ret)
+		goto err_unmap_io_base;
+
+	hdev->num_req_vfs = pci_sriov_get_totalvfs(pdev);
+
+	return 0;
+
+err_unmap_io_base:
+	pcim_iounmap(pdev, hdev->hw.hw.io_base);
+err_release_regions:
+	pci_release_regions(pdev);
+err_disable_device:
+	pci_disable_device(pdev);
+
+	return ret;
+}
+
+static void hclge_pci_uninit(struct hclge_dev *hdev)
+{
+	struct pci_dev *pdev = hdev->pdev;
+
+	if (hdev->hw.hw.mem_base)
+		devm_iounmap(&pdev->dev, hdev->hw.hw.mem_base);
+
+	pcim_iounmap(pdev, hdev->hw.hw.io_base);
+	pci_free_irq_vectors(pdev);
+	pci_release_regions(pdev);
+	pci_disable_device(pdev);
+}
+
+static void hclge_state_init(struct hclge_dev *hdev)
+{
+	set_bit(HCLGE_STATE_SERVICE_INITED, &hdev->state);
+	set_bit(HCLGE_STATE_DOWN, &hdev->state);
+	clear_bit(HCLGE_STATE_RST_SERVICE_SCHED, &hdev->state);
+	clear_bit(HCLGE_STATE_RST_HANDLING, &hdev->state);
+	clear_bit(HCLGE_STATE_RST_FAIL, &hdev->state);
+	clear_bit(HCLGE_STATE_MBX_SERVICE_SCHED, &hdev->state);
+	clear_bit(HCLGE_STATE_MBX_HANDLING, &hdev->state);
+}
+
+static void hclge_state_uninit(struct hclge_dev *hdev)
+{
+	set_bit(HCLGE_STATE_DOWN, &hdev->state);
+	set_bit(HCLGE_STATE_REMOVING, &hdev->state);
+
+	if (hdev->reset_timer.function)
+		timer_delete_sync(&hdev->reset_timer);
+	if (hdev->service_task.work.func)
+		cancel_delayed_work_sync(&hdev->service_task);
+}
+
+static void hclge_reset_prepare_general(struct hnae3_ae_dev *ae_dev,
+					enum hnae3_reset_type rst_type)
+{
+#define HCLGE_RESET_RETRY_WAIT_MS	500
+#define HCLGE_RESET_RETRY_CNT	5
+
+	struct hclge_dev *hdev = ae_dev->priv;
+	int retry_cnt = 0;
+	int ret;
+
+	while (retry_cnt++ < HCLGE_RESET_RETRY_CNT) {
+		down(&hdev->reset_sem);
+		set_bit(HCLGE_STATE_RST_HANDLING, &hdev->state);
+		hdev->reset_type = rst_type;
+		ret = hclge_reset_prepare(hdev);
+		if (!ret && !hdev->reset_pending)
+			break;
+
+		dev_err(&hdev->pdev->dev,
+			"failed to prepare to reset, ret=%d, reset_pending:0x%lx, retry_cnt:%d\n",
+			ret, hdev->reset_pending, retry_cnt);
+		clear_bit(HCLGE_STATE_RST_HANDLING, &hdev->state);
+		up(&hdev->reset_sem);
+		msleep(HCLGE_RESET_RETRY_WAIT_MS);
+	}
+
+	/* disable misc vector before reset done */
+	hclge_enable_vector(&hdev->misc_vector, false);
+	set_bit(HCLGE_COMM_STATE_CMD_DISABLE, &hdev->hw.hw.comm_state);
+
+	if (hdev->reset_type == HNAE3_FLR_RESET)
+		hdev->rst_stats.flr_rst_cnt++;
+}
+
+static void hclge_reset_done(struct hnae3_ae_dev *ae_dev)
+{
+	struct hclge_dev *hdev = ae_dev->priv;
+	int ret;
+
+	hclge_enable_vector(&hdev->misc_vector, true);
+
+	ret = hclge_reset_rebuild(hdev);
+	if (ret)
+		dev_err(&hdev->pdev->dev, "fail to rebuild, ret=%d\n", ret);
+
+	hdev->reset_type = HNAE3_NONE_RESET;
+	if (test_and_clear_bit(HCLGE_STATE_RST_HANDLING, &hdev->state))
+		up(&hdev->reset_sem);
+}
+
+static void hclge_clear_resetting_state(struct hclge_dev *hdev)
+{
+	u16 i;
+
+	for (i = 0; i < hdev->num_alloc_vport; i++) {
+		struct hclge_vport *vport = &hdev->vport[i];
+		int ret;
+
+		 /* Send cmd to clear vport's FUNC_RST_ING */
+		ret = hclge_set_vf_rst(hdev, vport->vport_id, false);
+		if (ret)
+			dev_warn(&hdev->pdev->dev,
+				 "clear vport(%u) rst failed %d!\n",
+				 vport->vport_id, ret);
+	}
+}
+
+static int hclge_clear_hw_resource(struct hclge_dev *hdev)
+{
+	struct hclge_desc desc;
+	int ret;
+
+	hclge_cmd_setup_basic_desc(&desc, HCLGE_OPC_CLEAR_HW_RESOURCE, false);
+
+	ret = hclge_cmd_send(&hdev->hw, &desc, 1);
+	/* This new command is only supported by new firmware, it will
+	 * fail with older firmware. Error value -EOPNOSUPP can only be
+	 * returned by older firmware running this command, to keep code
+	 * backward compatible we will override this value and return
+	 * success.
+	 */
+	if (ret && ret != -EOPNOTSUPP) {
+		dev_err(&hdev->pdev->dev,
+			"failed to clear hw resource, ret = %d\n", ret);
+		return ret;
+	}
+	return 0;
+}
+
+static void hclge_init_rxd_adv_layout(struct hclge_dev *hdev)
+{
+	if (hnae3_ae_dev_rxd_adv_layout_supported(hdev->ae_dev))
+		hclge_write_dev(&hdev->hw, HCLGE_RXD_ADV_LAYOUT_EN_REG, 1);
+}
+
+static void hclge_uninit_rxd_adv_layout(struct hclge_dev *hdev)
+{
+	if (hnae3_ae_dev_rxd_adv_layout_supported(hdev->ae_dev))
+		hclge_write_dev(&hdev->hw, HCLGE_RXD_ADV_LAYOUT_EN_REG, 0);
+}
+
+static struct hclge_wol_info *hclge_get_wol_info(struct hnae3_handle *handle)
+{
+	struct hclge_vport *vport = hclge_get_vport(handle);
+
+	return &vport->back->hw.mac.wol;
+}
+
+static int hclge_get_wol_supported_mode(struct hclge_dev *hdev,
+					u32 *wol_supported)
+{
+	struct hclge_query_wol_supported_cmd *wol_supported_cmd;
+	struct hclge_desc desc;
+	int ret;
+
+	hclge_cmd_setup_basic_desc(&desc, HCLGE_OPC_WOL_GET_SUPPORTED_MODE,
+				   true);
+	wol_supported_cmd = (struct hclge_query_wol_supported_cmd *)desc.data;
+
+	ret = hclge_cmd_send(&hdev->hw, &desc, 1);
+	if (ret) {
+		dev_err(&hdev->pdev->dev,
+			"failed to query wol supported, ret = %d\n", ret);
+		return ret;
+	}
+
+	*wol_supported = le32_to_cpu(wol_supported_cmd->supported_wake_mode);
+
+	return 0;
+}
+
+static int hclge_set_wol_cfg(struct hclge_dev *hdev,
+			     struct hclge_wol_info *wol_info)
+{
+	struct hclge_wol_cfg_cmd *wol_cfg_cmd;
+	struct hclge_desc desc;
+	int ret;
+
+	hclge_cmd_setup_basic_desc(&desc, HCLGE_OPC_WOL_CFG, false);
+	wol_cfg_cmd = (struct hclge_wol_cfg_cmd *)desc.data;
+	wol_cfg_cmd->wake_on_lan_mode = cpu_to_le32(wol_info->wol_current_mode);
+	wol_cfg_cmd->sopass_size = wol_info->wol_sopass_size;
+	memcpy(wol_cfg_cmd->sopass, wol_info->wol_sopass, SOPASS_MAX);
+
+	ret = hclge_cmd_send(&hdev->hw, &desc, 1);
+	if (ret)
+		dev_err(&hdev->pdev->dev,
+			"failed to set wol config, ret = %d\n", ret);
+
+	return ret;
+}
+
+static int hclge_update_wol(struct hclge_dev *hdev)
+{
+	struct hclge_wol_info *wol_info = &hdev->hw.mac.wol;
+
+	if (!hnae3_ae_dev_wol_supported(hdev->ae_dev))
+		return 0;
+
+	return hclge_set_wol_cfg(hdev, wol_info);
+}
+
+static int hclge_init_wol(struct hclge_dev *hdev)
+{
+	struct hclge_wol_info *wol_info = &hdev->hw.mac.wol;
+	int ret;
+
+	if (!hnae3_ae_dev_wol_supported(hdev->ae_dev))
+		return 0;
+
+	memset(wol_info, 0, sizeof(struct hclge_wol_info));
+	ret = hclge_get_wol_supported_mode(hdev,
+					   &wol_info->wol_support_mode);
+	if (ret) {
+		wol_info->wol_support_mode = 0;
+		return ret;
+	}
+
+	return hclge_update_wol(hdev);
+}
+
+static void hclge_get_wol(struct hnae3_handle *handle,
+			  struct ethtool_wolinfo *wol)
+{
+	struct hclge_wol_info *wol_info = hclge_get_wol_info(handle);
+
+	wol->supported = wol_info->wol_support_mode;
+	wol->wolopts = wol_info->wol_current_mode;
+	if (wol_info->wol_current_mode & WAKE_MAGICSECURE)
+		memcpy(wol->sopass, wol_info->wol_sopass, SOPASS_MAX);
+}
+
+static int hclge_set_wol(struct hnae3_handle *handle,
+			 struct ethtool_wolinfo *wol)
+{
+	struct hclge_wol_info *wol_info = hclge_get_wol_info(handle);
+	struct hclge_vport *vport = hclge_get_vport(handle);
+	u32 wol_mode;
+	int ret;
+
+	wol_mode = wol->wolopts;
+	if (wol_mode & ~wol_info->wol_support_mode)
+		return -EINVAL;
+
+	wol_info->wol_current_mode = wol_mode;
+	if (wol_mode & WAKE_MAGICSECURE) {
+		memcpy(wol_info->wol_sopass, wol->sopass, SOPASS_MAX);
+		wol_info->wol_sopass_size = SOPASS_MAX;
+	} else {
+		wol_info->wol_sopass_size = 0;
+	}
+
+	ret = hclge_set_wol_cfg(vport->back, wol_info);
+	if (ret)
+		wol_info->wol_current_mode = 0;
+
+	return ret;
+}
+
+static int hclge_init_ae_dev(struct hnae3_ae_dev *ae_dev)
+{
+	struct pci_dev *pdev = ae_dev->pdev;
+	struct hclge_dev *hdev;
+	int ret;
+
+	hdev = devm_kzalloc(&pdev->dev, sizeof(*hdev), GFP_KERNEL);
+	if (!hdev)
+		return -ENOMEM;
+
+	hdev->pdev = pdev;
+	hdev->ae_dev = ae_dev;
+	hdev->reset_type = HNAE3_NONE_RESET;
+	hdev->reset_level = HNAE3_FUNC_RESET;
+	ae_dev->priv = hdev;
+
+	/* HW supprt 2 layer vlan */
+	hdev->mps = ETH_FRAME_LEN + ETH_FCS_LEN + 2 * VLAN_HLEN;
+
+	mutex_init(&hdev->vport_lock);
+	spin_lock_init(&hdev->fd_rule_lock);
+	sema_init(&hdev->reset_sem, 1);
+
+	ret = hclge_pci_init(hdev);
+	if (ret)
+		goto out;
+
+	/* Firmware command queue initialize */
+	ret = hclge_comm_cmd_queue_init(hdev->pdev, &hdev->hw.hw);
+	if (ret)
+		goto err_pci_uninit;
+
+	/* Firmware command initialize */
+	hclge_comm_cmd_init_ops(&hdev->hw.hw, &hclge_cmq_ops);
+	ret = hclge_comm_cmd_init(hdev->ae_dev, &hdev->hw.hw, &hdev->fw_version,
+				  true, hdev->reset_pending);
+	if (ret)
+		goto err_cmd_uninit;
+
+	ret  = hclge_clear_hw_resource(hdev);
+	if (ret)
+		goto err_cmd_uninit;
+
+	ret = hclge_get_cap(hdev);
+	if (ret)
+		goto err_cmd_uninit;
+
+	ret = hclge_query_dev_specs(hdev);
+	if (ret) {
+		dev_err(&pdev->dev, "failed to query dev specifications, ret = %d.\n",
+			ret);
+		goto err_cmd_uninit;
+	}
+
+	ret = hclge_configure(hdev);
+	if (ret) {
+		dev_err(&pdev->dev, "Configure dev error, ret = %d.\n", ret);
+		goto err_cmd_uninit;
+	}
+
+	ret = hclge_init_msi(hdev);
+	if (ret) {
+		dev_err(&pdev->dev, "Init MSI/MSI-X error, ret = %d.\n", ret);
+		goto err_cmd_uninit;
+	}
+
+	ret = hclge_misc_irq_init(hdev);
+	if (ret)
+		goto err_msi_uninit;
+
+	ret = hclge_alloc_tqps(hdev);
+	if (ret) {
+		dev_err(&pdev->dev, "Allocate TQPs error, ret = %d.\n", ret);
+		goto err_msi_irq_uninit;
+	}
+
+	ret = hclge_alloc_vport(hdev);
+	if (ret)
+		goto err_msi_irq_uninit;
+
+	ret = hclge_map_tqp(hdev);
+	if (ret)
+		goto err_msi_irq_uninit;
+
+	if (hdev->hw.mac.media_type == HNAE3_MEDIA_TYPE_COPPER) {
+		clear_bit(HNAE3_DEV_SUPPORT_FEC_B, ae_dev->caps);
+		if (hnae3_dev_phy_imp_supported(hdev))
+			ret = hclge_update_tp_port_info(hdev);
+		else
+			ret = hclge_mac_mdio_config(hdev);
+
+		if (ret)
+			goto err_msi_irq_uninit;
+	}
+
+	ret = hclge_init_umv_space(hdev);
+	if (ret)
+		goto err_mdiobus_unreg;
+
+	ret = hclge_mac_init(hdev);
+	if (ret) {
+		dev_err(&pdev->dev, "Mac init error, ret = %d\n", ret);
+		goto err_mdiobus_unreg;
+	}
+
+	ret = hclge_config_tso(hdev, HCLGE_TSO_MSS_MIN, HCLGE_TSO_MSS_MAX);
+	if (ret) {
+		dev_err(&pdev->dev, "Enable tso fail, ret =%d\n", ret);
+		goto err_mdiobus_unreg;
+	}
+
+	ret = hclge_config_gro(hdev);
+	if (ret)
+		goto err_mdiobus_unreg;
+
+	ret = hclge_init_vlan_config(hdev);
+	if (ret) {
+		dev_err(&pdev->dev, "VLAN init fail, ret =%d\n", ret);
+		goto err_mdiobus_unreg;
+	}
+
+	ret = hclge_tm_schd_init(hdev);
+	if (ret) {
+		dev_err(&pdev->dev, "tm schd init fail, ret =%d\n", ret);
+		goto err_mdiobus_unreg;
+	}
+
+	ret = hclge_comm_rss_init_cfg(&hdev->vport->nic, hdev->ae_dev,
+				      &hdev->rss_cfg);
+	if (ret) {
+		dev_err(&pdev->dev, "failed to init rss cfg, ret = %d\n", ret);
+		goto err_mdiobus_unreg;
+	}
+
+	ret = hclge_rss_init_hw(hdev);
+	if (ret) {
+		dev_err(&pdev->dev, "Rss init fail, ret =%d\n", ret);
+		goto err_mdiobus_unreg;
+	}
+
+	ret = init_mgr_tbl(hdev);
+	if (ret) {
+		dev_err(&pdev->dev, "manager table init fail, ret =%d\n", ret);
+		goto err_mdiobus_unreg;
+	}
+
+	ret = hclge_init_fd_config(hdev);
+	if (ret) {
+		dev_err(&pdev->dev,
+			"fd table init fail, ret=%d\n", ret);
+		goto err_mdiobus_unreg;
+	}
+
+	ret = hclge_ptp_init(hdev);
+	if (ret)
+		goto err_mdiobus_unreg;
+
+	ret = hclge_update_port_info(hdev);
+	if (ret)
+		goto err_ptp_uninit;
+
+	INIT_KFIFO(hdev->mac_tnl_log);
+
+	hclge_dcb_ops_set(hdev);
+
+	timer_setup(&hdev->reset_timer, hclge_reset_timer, 0);
+	INIT_DELAYED_WORK(&hdev->service_task, hclge_service_task);
+
+	hclge_clear_all_event_cause(hdev);
+	hclge_clear_resetting_state(hdev);
+
+	/* Log and clear the hw errors those already occurred */
+	if (hnae3_dev_ras_imp_supported(hdev))
+		hclge_handle_occurred_error(hdev);
+	else
+		hclge_handle_all_hns_hw_errors(ae_dev);
+
+	/* request delayed reset for the error recovery because an immediate
+	 * global reset on a PF affecting pending initialization of other PFs
+	 */
+	if (ae_dev->hw_err_reset_req) {
+		enum hnae3_reset_type reset_level;
+
+		reset_level = hclge_get_reset_level(ae_dev,
+						    &ae_dev->hw_err_reset_req);
+		hclge_set_def_reset_request(ae_dev, reset_level);
+		mod_timer(&hdev->reset_timer, jiffies + HCLGE_RESET_INTERVAL);
+	}
+
+	hclge_init_rxd_adv_layout(hdev);
+
+	ret = hclge_init_wol(hdev);
+	if (ret)
+		dev_warn(&pdev->dev,
+			 "failed to wake on lan init, ret = %d\n", ret);
+
+	ret = hclge_devlink_init(hdev);
+	if (ret)
+		goto err_ptp_uninit;
+
+	hclge_state_init(hdev);
+	hdev->last_reset_time = jiffies;
+
+	/* Enable MISC vector(vector0) */
+	enable_irq(hdev->misc_vector.vector_irq);
+	hclge_enable_vector(&hdev->misc_vector, true);
+
+	dev_info(&hdev->pdev->dev, "%s driver initialization finished.\n",
+		 HCLGE_DRIVER_NAME);
+
+	hclge_task_schedule(hdev, round_jiffies_relative(HZ));
+	return 0;
+
+err_ptp_uninit:
+	hclge_ptp_uninit(hdev);
+err_mdiobus_unreg:
+	if (hdev->hw.mac.phydev)
+		mdiobus_unregister(hdev->hw.mac.mdio_bus);
+err_msi_irq_uninit:
+	hclge_misc_irq_uninit(hdev);
+err_msi_uninit:
+	pci_free_irq_vectors(pdev);
+err_cmd_uninit:
+	hclge_comm_cmd_uninit(hdev->ae_dev, &hdev->hw.hw);
+err_pci_uninit:
+	pcim_iounmap(pdev, hdev->hw.hw.io_base);
+	pci_release_regions(pdev);
+	pci_disable_device(pdev);
+out:
+	mutex_destroy(&hdev->vport_lock);
+	return ret;
+}
+
+static void hclge_stats_clear(struct hclge_dev *hdev)
+{
+	memset(&hdev->mac_stats, 0, sizeof(hdev->mac_stats));
+	memset(&hdev->fec_stats, 0, sizeof(hdev->fec_stats));
+}
+
+static int hclge_set_mac_spoofchk(struct hclge_dev *hdev, int vf, bool enable)
+{
+	return hclge_config_switch_param(hdev, vf, enable,
+					 HCLGE_SWITCH_ANTI_SPOOF_MASK);
+}
+
+static int hclge_set_vlan_spoofchk(struct hclge_dev *hdev, int vf, bool enable)
+{
+	return hclge_set_vlan_filter_ctrl(hdev, HCLGE_FILTER_TYPE_VF,
+					  HCLGE_FILTER_FE_NIC_INGRESS_B,
+					  enable, vf);
+}
+
+static int hclge_set_vf_spoofchk_hw(struct hclge_dev *hdev, int vf, bool enable)
+{
+	int ret;
+
+	ret = hclge_set_mac_spoofchk(hdev, vf, enable);
+	if (ret) {
+		dev_err(&hdev->pdev->dev,
+			"Set vf %d mac spoof check %s failed, ret=%d\n",
+			vf, str_on_off(enable), ret);
+		return ret;
+	}
+
+	ret = hclge_set_vlan_spoofchk(hdev, vf, enable);
+	if (ret)
+		dev_err(&hdev->pdev->dev,
+			"Set vf %d vlan spoof check %s failed, ret=%d\n",
+			vf, str_on_off(enable), ret);
+
+	return ret;
+}
+
+static int hclge_set_vf_spoofchk(struct hnae3_handle *handle, int vf,
+				 bool enable)
+{
+	struct hclge_vport *vport = hclge_get_vport(handle);
+	struct hclge_dev *hdev = vport->back;
+	u32 new_spoofchk = enable ? 1 : 0;
+	int ret;
+
+	if (hdev->ae_dev->dev_version < HNAE3_DEVICE_VERSION_V2)
+		return -EOPNOTSUPP;
+
+	vport = hclge_get_vf_vport(hdev, vf);
+	if (!vport)
+		return -EINVAL;
+
+	if (vport->vf_info.spoofchk == new_spoofchk)
+		return 0;
+
+	if (enable && test_bit(vport->vport_id, hdev->vf_vlan_full))
+		dev_warn(&hdev->pdev->dev,
+			 "vf %d vlan table is full, enable spoof check may cause its packet send fail\n",
+			 vf);
+	else if (enable && hclge_is_umv_space_full(vport, true))
+		dev_warn(&hdev->pdev->dev,
+			 "vf %d mac table is full, enable spoof check may cause its packet send fail\n",
+			 vf);
+
+	ret = hclge_set_vf_spoofchk_hw(hdev, vport->vport_id, enable);
+	if (ret)
+		return ret;
+
+	vport->vf_info.spoofchk = new_spoofchk;
+	return 0;
+}
+
+static int hclge_reset_vport_spoofchk(struct hclge_dev *hdev)
+{
+	struct hclge_vport *vport = hdev->vport;
+	int ret;
+	int i;
+
+	if (hdev->ae_dev->dev_version < HNAE3_DEVICE_VERSION_V2)
+		return 0;
+
+	/* resume the vf spoof check state after reset */
+	for (i = 0; i < hdev->num_alloc_vport; i++) {
+		ret = hclge_set_vf_spoofchk_hw(hdev, vport->vport_id,
+					       vport->vf_info.spoofchk);
+		if (ret)
+			return ret;
+
+		vport++;
+	}
+
+	return 0;
+}
+
+static int hclge_set_vf_trust(struct hnae3_handle *handle, int vf, bool enable)
+{
+	struct hclge_vport *vport = hclge_get_vport(handle);
+	struct hclge_dev *hdev = vport->back;
+	u32 new_trusted = enable ? 1 : 0;
+
+	vport = hclge_get_vf_vport(hdev, vf);
+	if (!vport)
+		return -EINVAL;
+
+	if (vport->vf_info.trusted == new_trusted)
+		return 0;
+
+	vport->vf_info.trusted = new_trusted;
+	set_bit(HCLGE_VPORT_STATE_PROMISC_CHANGE, &vport->state);
+	hclge_task_schedule(hdev, 0);
+
+	return 0;
+}
+
+static void hclge_reset_vf_rate(struct hclge_dev *hdev)
+{
+	int ret;
+	int vf;
+
+	/* reset vf rate to default value */
+	for (vf = HCLGE_VF_VPORT_START_NUM; vf < hdev->num_alloc_vport; vf++) {
+		struct hclge_vport *vport = &hdev->vport[vf];
+
+		vport->vf_info.max_tx_rate = 0;
+		ret = hclge_tm_qs_shaper_cfg(vport, vport->vf_info.max_tx_rate);
+		if (ret)
+			dev_err(&hdev->pdev->dev,
+				"vf%d failed to reset to default, ret=%d\n",
+				vf - HCLGE_VF_VPORT_START_NUM, ret);
+	}
+}
+
+static int hclge_vf_rate_param_check(struct hclge_dev *hdev,
+				     int min_tx_rate, int max_tx_rate)
+{
+	if (min_tx_rate != 0 ||
+	    max_tx_rate < 0 || (u32)max_tx_rate > hdev->hw.mac.max_speed) {
+		dev_err(&hdev->pdev->dev,
+			"min_tx_rate:%d [0], max_tx_rate:%d [0, %u]\n",
+			min_tx_rate, max_tx_rate, hdev->hw.mac.max_speed);
+		return -EINVAL;
+	}
+
+	return 0;
+}
+
+static int hclge_set_vf_rate(struct hnae3_handle *handle, int vf,
+			     int min_tx_rate, int max_tx_rate, bool force)
+{
+	struct hclge_vport *vport = hclge_get_vport(handle);
+	struct hclge_dev *hdev = vport->back;
+	int ret;
+
+	ret = hclge_vf_rate_param_check(hdev, min_tx_rate, max_tx_rate);
+	if (ret)
+		return ret;
+
+	vport = hclge_get_vf_vport(hdev, vf);
+	if (!vport)
+		return -EINVAL;
+
+	if (!force && (u32)max_tx_rate == vport->vf_info.max_tx_rate)
+		return 0;
+
+	ret = hclge_tm_qs_shaper_cfg(vport, max_tx_rate);
+	if (ret)
+		return ret;
+
+	vport->vf_info.max_tx_rate = max_tx_rate;
+
+	return 0;
+}
+
+static int hclge_resume_vf_rate(struct hclge_dev *hdev)
+{
+	struct hnae3_handle *handle = &hdev->vport->nic;
+	struct hclge_vport *vport;
+	int ret;
+	int vf;
+
+	/* resume the vf max_tx_rate after reset */
+	for (vf = 0; vf < pci_num_vf(hdev->pdev); vf++) {
+		vport = hclge_get_vf_vport(hdev, vf);
+		if (!vport)
+			return -EINVAL;
+
+		/* zero means max rate, after reset, firmware already set it to
+		 * max rate, so just continue.
+		 */
+		if (!vport->vf_info.max_tx_rate)
+			continue;
+
+		ret = hclge_set_vf_rate(handle, vf, 0,
+					vport->vf_info.max_tx_rate, true);
+		if (ret) {
+			dev_err(&hdev->pdev->dev,
+				"vf%d failed to resume tx_rate:%u, ret=%d\n",
+				vf, vport->vf_info.max_tx_rate, ret);
+			return ret;
+		}
+	}
+
+	return 0;
+}
+
+static void hclge_reset_vport_state(struct hclge_dev *hdev)
+{
+	struct hclge_vport *vport = hdev->vport;
+	int i;
+
+	for (i = 0; i < hdev->num_alloc_vport; i++) {
+		clear_bit(HCLGE_VPORT_STATE_ALIVE, &vport->state);
+		vport++;
+	}
+}
+
+static int hclge_reset_ae_dev(struct hnae3_ae_dev *ae_dev)
+{
+	struct hclge_dev *hdev = ae_dev->priv;
+	struct pci_dev *pdev = ae_dev->pdev;
+	int ret;
+
+	set_bit(HCLGE_STATE_DOWN, &hdev->state);
+
+	hclge_stats_clear(hdev);
+	/* NOTE: pf reset needn't to clear or restore pf and vf table entry.
+	 * so here should not clean table in memory.
+	 */
+	if (hdev->reset_type == HNAE3_IMP_RESET ||
+	    hdev->reset_type == HNAE3_GLOBAL_RESET) {
+		memset(hdev->vlan_table, 0, sizeof(hdev->vlan_table));
+		memset(hdev->vf_vlan_full, 0, sizeof(hdev->vf_vlan_full));
+		bitmap_set(hdev->vport_config_block, 0, hdev->num_alloc_vport);
+		hclge_reset_umv_space(hdev);
+	}
+
+	ret = hclge_comm_cmd_init(hdev->ae_dev, &hdev->hw.hw, &hdev->fw_version,
+				  true, hdev->reset_pending);
+	if (ret) {
+		dev_err(&pdev->dev, "Cmd queue init failed\n");
+		return ret;
+	}
+
+	ret = hclge_map_tqp(hdev);
+	if (ret) {
+		dev_err(&pdev->dev, "Map tqp error, ret = %d.\n", ret);
+		return ret;
+	}
+
+	ret = hclge_mac_init(hdev);
+	if (ret) {
+		dev_err(&pdev->dev, "Mac init error, ret = %d\n", ret);
+		return ret;
+	}
+
+	ret = hclge_tp_port_init(hdev);
+	if (ret) {
+		dev_err(&pdev->dev, "failed to init tp port, ret = %d\n",
+			ret);
+		return ret;
+	}
+
+	ret = hclge_config_tso(hdev, HCLGE_TSO_MSS_MIN, HCLGE_TSO_MSS_MAX);
+	if (ret) {
+		dev_err(&pdev->dev, "Enable tso fail, ret =%d\n", ret);
+		return ret;
+	}
+
+	ret = hclge_config_gro(hdev);
+	if (ret)
+		return ret;
+
+	ret = hclge_init_vlan_config(hdev);
+	if (ret) {
+		dev_err(&pdev->dev, "VLAN init fail, ret =%d\n", ret);
+		return ret;
+	}
+
+	hclge_reset_tc_config(hdev);
+
+	ret = hclge_tm_init_hw(hdev, true);
+	if (ret) {
+		dev_err(&pdev->dev, "tm init hw fail, ret =%d\n", ret);
+		return ret;
+	}
+
+	ret = hclge_rss_init_hw(hdev);
+	if (ret) {
+		dev_err(&pdev->dev, "Rss init fail, ret =%d\n", ret);
+		return ret;
+	}
+
+	ret = init_mgr_tbl(hdev);
+	if (ret) {
+		dev_err(&pdev->dev,
+			"failed to reinit manager table, ret = %d\n", ret);
+		return ret;
+	}
+
+	ret = hclge_init_fd_config(hdev);
+	if (ret) {
+		dev_err(&pdev->dev, "fd table init fail, ret=%d\n", ret);
+		return ret;
+	}
+
+	ret = hclge_ptp_init(hdev);
+	if (ret)
+		return ret;
+
+	/* Log and clear the hw errors those already occurred */
+	if (hnae3_dev_ras_imp_supported(hdev))
+		hclge_handle_occurred_error(hdev);
+	else
+		hclge_handle_all_hns_hw_errors(ae_dev);
+
+	/* Re-enable the hw error interrupts because
+	 * the interrupts get disabled on global reset.
+	 */
+	ret = hclge_config_nic_hw_error(hdev, true);
+	if (ret) {
+		dev_err(&pdev->dev,
+			"fail(%d) to re-enable NIC hw error interrupts\n",
+			ret);
+		return ret;
+	}
+
+	if (hdev->roce_client) {
+		ret = hclge_config_rocee_ras_interrupt(hdev, true);
+		if (ret) {
+			dev_err(&pdev->dev,
+				"fail(%d) to re-enable roce ras interrupts\n",
+				ret);
+			return ret;
+		}
+	}
+
+	hclge_reset_vport_state(hdev);
+	ret = hclge_reset_vport_spoofchk(hdev);
+	if (ret)
+		return ret;
+
+	ret = hclge_resume_vf_rate(hdev);
+	if (ret)
+		return ret;
+
+	hclge_init_rxd_adv_layout(hdev);
+
+	ret = hclge_update_wol(hdev);
+	if (ret)
+		dev_warn(&pdev->dev,
+			 "failed to update wol config, ret = %d\n", ret);
+
+	dev_info(&pdev->dev, "Reset done, %s driver initialization finished.\n",
+		 HCLGE_DRIVER_NAME);
+
+	return 0;
+}
+
+static void hclge_uninit_ae_dev(struct hnae3_ae_dev *ae_dev)
+{
+	struct hclge_dev *hdev = ae_dev->priv;
+	struct hclge_mac *mac = &hdev->hw.mac;
+
+	hclge_reset_vf_rate(hdev);
+	hclge_clear_vf_vlan(hdev);
+	hclge_state_uninit(hdev);
+	hclge_ptp_uninit(hdev);
+	hclge_uninit_rxd_adv_layout(hdev);
+	hclge_uninit_mac_table(hdev);
+	hclge_del_all_fd_entries(hdev);
+
+	if (mac->phydev)
+		mdiobus_unregister(mac->mdio_bus);
+
+	/* Disable MISC vector(vector0) */
+	hclge_enable_vector(&hdev->misc_vector, false);
+	disable_irq(hdev->misc_vector.vector_irq);
+
+	/* Disable all hw interrupts */
+	hclge_config_mac_tnl_int(hdev, false);
+	hclge_config_nic_hw_error(hdev, false);
+	hclge_config_rocee_ras_interrupt(hdev, false);
+
+	hclge_comm_cmd_uninit(hdev->ae_dev, &hdev->hw.hw);
+	hclge_misc_irq_uninit(hdev);
+	hclge_devlink_uninit(hdev);
+	hclge_pci_uninit(hdev);
+	hclge_uninit_vport_vlan_table(hdev);
+	mutex_destroy(&hdev->vport_lock);
+	ae_dev->priv = NULL;
+}
+
+static u32 hclge_get_max_channels(struct hnae3_handle *handle)
+{
+	struct hclge_vport *vport = hclge_get_vport(handle);
+	struct hclge_dev *hdev = vport->back;
+
+	return min_t(u32, hdev->pf_rss_size_max, vport->alloc_tqps);
+}
+
+static void hclge_get_channels(struct hnae3_handle *handle,
+			       struct ethtool_channels *ch)
+{
+	ch->max_combined = hclge_get_max_channels(handle);
+	ch->other_count = 1;
+	ch->max_other = 1;
+	ch->combined_count = handle->kinfo.rss_size;
+}
+
+static void hclge_get_tqps_and_rss_info(struct hnae3_handle *handle,
+					u16 *alloc_tqps, u16 *max_rss_size)
+{
+	struct hclge_vport *vport = hclge_get_vport(handle);
+	struct hclge_dev *hdev = vport->back;
+
+	*alloc_tqps = vport->alloc_tqps;
+	*max_rss_size = hdev->pf_rss_size_max;
+}
+
+static int hclge_set_rss_tc_mode_cfg(struct hnae3_handle *handle)
+{
+	struct hclge_vport *vport = hclge_get_vport(handle);
+	u16 tc_offset[HCLGE_MAX_TC_NUM] = {0};
+	struct hclge_dev *hdev = vport->back;
+	u16 tc_size[HCLGE_MAX_TC_NUM] = {0};
+	u16 tc_valid[HCLGE_MAX_TC_NUM];
+	u16 roundup_size;
+	unsigned int i;
+
+	roundup_size = roundup_pow_of_two(vport->nic.kinfo.rss_size);
+	roundup_size = ilog2(roundup_size);
+	/* Set the RSS TC mode according to the new RSS size */
+	for (i = 0; i < HCLGE_MAX_TC_NUM; i++) {
+		tc_valid[i] = 0;
+
+		if (!(hdev->hw_tc_map & BIT(i)))
+			continue;
+
+		tc_valid[i] = 1;
+		tc_size[i] = roundup_size;
+		tc_offset[i] = vport->nic.kinfo.rss_size * i;
+	}
+
+	return hclge_comm_set_rss_tc_mode(&hdev->hw.hw, tc_offset, tc_valid,
+					  tc_size);
+}
+
+static int hclge_set_channels(struct hnae3_handle *handle, u32 new_tqps_num,
+			      bool rxfh_configured)
+{
+	struct hnae3_ae_dev *ae_dev = pci_get_drvdata(handle->pdev);
+	struct hclge_vport *vport = hclge_get_vport(handle);
+	struct hnae3_knic_private_info *kinfo = &vport->nic.kinfo;
+	struct hclge_dev *hdev = vport->back;
+	u16 cur_rss_size = kinfo->rss_size;
+	u16 cur_tqps = kinfo->num_tqps;
+	u32 *rss_indir;
+	unsigned int i;
+	int ret;
+
+	kinfo->req_rss_size = new_tqps_num;
+
+	ret = hclge_tm_vport_map_update(hdev);
+	if (ret) {
+		dev_err(&hdev->pdev->dev, "tm vport map fail, ret =%d\n", ret);
+		return ret;
+	}
+
+	ret = hclge_set_rss_tc_mode_cfg(handle);
+	if (ret)
+		return ret;
+
+	/* RSS indirection table has been configured by user */
+	if (rxfh_configured)
+		goto out;
+
+	/* Reinitializes the rss indirect table according to the new RSS size */
+	rss_indir = kcalloc(ae_dev->dev_specs.rss_ind_tbl_size, sizeof(u32),
+			    GFP_KERNEL);
+	if (!rss_indir)
+		return -ENOMEM;
+
+	for (i = 0; i < ae_dev->dev_specs.rss_ind_tbl_size; i++)
+		rss_indir[i] = i % kinfo->rss_size;
+
+	ret = hclge_set_rss(handle, rss_indir, NULL, 0);
+	if (ret)
+		dev_err(&hdev->pdev->dev, "set rss indir table fail, ret=%d\n",
+			ret);
+
+	kfree(rss_indir);
+
+out:
+	if (!ret)
+		dev_info(&hdev->pdev->dev,
+			 "Channels changed, rss_size from %u to %u, tqps from %u to %u",
+			 cur_rss_size, kinfo->rss_size,
+			 cur_tqps, kinfo->rss_size * kinfo->tc_info.num_tc);
+
+	return ret;
+}
+
+static int hclge_set_led_status(struct hclge_dev *hdev, u8 locate_led_status)
+{
+	struct hclge_set_led_state_cmd *req;
+	struct hclge_desc desc;
+	int ret;
+
+	hclge_cmd_setup_basic_desc(&desc, HCLGE_OPC_LED_STATUS_CFG, false);
+
+	req = (struct hclge_set_led_state_cmd *)desc.data;
+	hnae3_set_field(req->locate_led_config, HCLGE_LED_LOCATE_STATE_M,
+			HCLGE_LED_LOCATE_STATE_S, locate_led_status);
+
+	ret = hclge_cmd_send(&hdev->hw, &desc, 1);
+	if (ret)
+		dev_err(&hdev->pdev->dev,
+			"Send set led state cmd error, ret =%d\n", ret);
+
+	return ret;
+}
+
+enum hclge_led_status {
+	HCLGE_LED_OFF,
+	HCLGE_LED_ON,
+	HCLGE_LED_NO_CHANGE = 0xFF,
+};
+
+static int hclge_set_led_id(struct hnae3_handle *handle,
+			    enum ethtool_phys_id_state status)
+{
+	struct hclge_vport *vport = hclge_get_vport(handle);
+	struct hclge_dev *hdev = vport->back;
+
+	switch (status) {
+	case ETHTOOL_ID_ACTIVE:
+		return hclge_set_led_status(hdev, HCLGE_LED_ON);
+	case ETHTOOL_ID_INACTIVE:
+		return hclge_set_led_status(hdev, HCLGE_LED_OFF);
+	default:
+		return -EINVAL;
+	}
+}
+
+static void hclge_get_link_mode(struct hnae3_handle *handle,
+				unsigned long *supported,
+				unsigned long *advertising)
+{
+	unsigned int size = BITS_TO_LONGS(__ETHTOOL_LINK_MODE_MASK_NBITS);
+	struct hclge_vport *vport = hclge_get_vport(handle);
+	struct hclge_dev *hdev = vport->back;
+	unsigned int idx = 0;
+
+	for (; idx < size; idx++) {
+		supported[idx] = hdev->hw.mac.supported[idx];
+		advertising[idx] = hdev->hw.mac.advertising[idx];
+	}
+}
+
+static int hclge_gro_en(struct hnae3_handle *handle, bool enable)
+{
+	struct hclge_vport *vport = hclge_get_vport(handle);
+	struct hclge_dev *hdev = vport->back;
+	bool gro_en_old = hdev->gro_en;
+	int ret;
+
+	hdev->gro_en = enable;
+	ret = hclge_config_gro(hdev);
+	if (ret)
+		hdev->gro_en = gro_en_old;
+
+	return ret;
+}
+
+static int hclge_sync_vport_promisc_mode(struct hclge_vport *vport)
+{
+	struct hnae3_handle *handle = &vport->nic;
+	struct hclge_dev *hdev = vport->back;
+	bool uc_en = false;
+	bool mc_en = false;
+	u8 tmp_flags;
+	bool bc_en;
+	int ret;
+
+	if (vport->last_promisc_flags != vport->overflow_promisc_flags) {
+		set_bit(HCLGE_VPORT_STATE_PROMISC_CHANGE, &vport->state);
+		vport->last_promisc_flags = vport->overflow_promisc_flags;
+	}
+
+	if (!test_and_clear_bit(HCLGE_VPORT_STATE_PROMISC_CHANGE,
+				&vport->state))
+		return 0;
+
+	/* for PF */
+	if (!vport->vport_id) {
+		tmp_flags = handle->netdev_flags | vport->last_promisc_flags;
+		ret = hclge_set_promisc_mode(handle, tmp_flags & HNAE3_UPE,
+					     tmp_flags & HNAE3_MPE);
+		if (!ret)
+			set_bit(HCLGE_VPORT_STATE_VLAN_FLTR_CHANGE,
+				&vport->state);
+		else
+			set_bit(HCLGE_VPORT_STATE_PROMISC_CHANGE,
+				&vport->state);
+		return ret;
+	}
+
+	/* for VF */
+	if (vport->vf_info.trusted) {
+		uc_en = vport->vf_info.request_uc_en > 0 ||
+			vport->overflow_promisc_flags & HNAE3_OVERFLOW_UPE;
+		mc_en = vport->vf_info.request_mc_en > 0 ||
+			vport->overflow_promisc_flags & HNAE3_OVERFLOW_MPE;
+	}
+	bc_en = vport->vf_info.request_bc_en > 0;
+
+	ret = hclge_cmd_set_promisc_mode(hdev, vport->vport_id, uc_en,
+					 mc_en, bc_en);
+	if (ret) {
+		set_bit(HCLGE_VPORT_STATE_PROMISC_CHANGE, &vport->state);
+		return ret;
+	}
+	hclge_set_vport_vlan_fltr_change(vport);
+
+	return 0;
+}
+
+static void hclge_sync_promisc_mode(struct hclge_dev *hdev)
+{
+	struct hclge_vport *vport;
+	int ret;
+	u16 i;
+
+	for (i = 0; i < hdev->num_alloc_vport; i++) {
+		vport = &hdev->vport[i];
+
+		ret = hclge_sync_vport_promisc_mode(vport);
+		if (ret)
+			return;
+	}
+}
+
+static bool hclge_module_existed(struct hclge_dev *hdev)
+{
+	struct hclge_desc desc;
+	u32 existed;
+	int ret;
+
+	hclge_cmd_setup_basic_desc(&desc, HCLGE_OPC_GET_SFP_EXIST, true);
+	ret = hclge_cmd_send(&hdev->hw, &desc, 1);
+	if (ret) {
+		dev_err(&hdev->pdev->dev,
+			"failed to get SFP exist state, ret = %d\n", ret);
+		return false;
+	}
+
+	existed = le32_to_cpu(desc.data[0]);
+
+	return existed != 0;
+}
+
+/* need 6 bds(total 140 bytes) in one reading
+ * return the number of bytes actually read, 0 means read failed.
+ */
+static u16 hclge_get_sfp_eeprom_info(struct hclge_dev *hdev, u32 offset,
+				     u32 len, u8 *data)
+{
+	struct hclge_desc desc[HCLGE_SFP_INFO_CMD_NUM];
+	struct hclge_sfp_info_bd0_cmd *sfp_info_bd0;
+	u16 read_len;
+	u16 copy_len;
+	int ret;
+	int i;
+
+	/* setup all 6 bds to read module eeprom info. */
+	for (i = 0; i < HCLGE_SFP_INFO_CMD_NUM; i++) {
+		hclge_cmd_setup_basic_desc(&desc[i], HCLGE_OPC_GET_SFP_EEPROM,
+					   true);
+
+		/* bd0~bd4 need next flag */
+		if (i < HCLGE_SFP_INFO_CMD_NUM - 1)
+			desc[i].flag |= cpu_to_le16(HCLGE_COMM_CMD_FLAG_NEXT);
+	}
+
+	/* setup bd0, this bd contains offset and read length. */
+	sfp_info_bd0 = (struct hclge_sfp_info_bd0_cmd *)desc[0].data;
+	sfp_info_bd0->offset = cpu_to_le16((u16)offset);
+	read_len = min_t(u16, len, HCLGE_SFP_INFO_MAX_LEN);
+	sfp_info_bd0->read_len = cpu_to_le16(read_len);
+
+	ret = hclge_cmd_send(&hdev->hw, desc, i);
+	if (ret) {
+		dev_err(&hdev->pdev->dev,
+			"failed to get SFP eeprom info, ret = %d\n", ret);
+		return 0;
+	}
+
+	/* copy sfp info from bd0 to out buffer. */
+	copy_len = min_t(u16, len, HCLGE_SFP_INFO_BD0_LEN);
+	memcpy(data, sfp_info_bd0->data, copy_len);
+	read_len = copy_len;
+
+	/* copy sfp info from bd1~bd5 to out buffer if needed. */
+	for (i = 1; i < HCLGE_SFP_INFO_CMD_NUM; i++) {
+		if (read_len >= len)
+			return read_len;
+
+		copy_len = min_t(u16, len - read_len, HCLGE_SFP_INFO_BDX_LEN);
+		memcpy(data + read_len, desc[i].data, copy_len);
+		read_len += copy_len;
+	}
+
+	return read_len;
+}
+
+static int hclge_get_module_eeprom(struct hnae3_handle *handle, u32 offset,
+				   u32 len, u8 *data)
+{
+	struct hclge_vport *vport = hclge_get_vport(handle);
+	struct hclge_dev *hdev = vport->back;
+	u32 read_len = 0;
+	u16 data_len;
+
+	if (hdev->hw.mac.media_type != HNAE3_MEDIA_TYPE_FIBER)
+		return -EOPNOTSUPP;
+
+	if (!hclge_module_existed(hdev))
+		return -ENXIO;
+
+	while (read_len < len) {
+		data_len = hclge_get_sfp_eeprom_info(hdev,
+						     offset + read_len,
+						     len - read_len,
+						     data + read_len);
+		if (!data_len)
+			return -EIO;
+
+		read_len += data_len;
+	}
+
+	return 0;
+}
+
+static int hclge_get_link_diagnosis_info(struct hnae3_handle *handle,
+					 u32 *status_code)
+{
+	struct hclge_vport *vport = hclge_get_vport(handle);
+	struct hclge_dev *hdev = vport->back;
+	struct hclge_desc desc;
+	int ret;
+
+	if (hdev->ae_dev->dev_version <= HNAE3_DEVICE_VERSION_V2)
+		return -EOPNOTSUPP;
+
+	hclge_cmd_setup_basic_desc(&desc, HCLGE_OPC_QUERY_LINK_DIAGNOSIS, true);
+	ret = hclge_cmd_send(&hdev->hw, &desc, 1);
+	if (ret) {
+		dev_err(&hdev->pdev->dev,
+			"failed to query link diagnosis info, ret = %d\n", ret);
+		return ret;
+	}
+
+	*status_code = le32_to_cpu(desc.data[0]);
+	return 0;
+}
+
+/* After disable sriov, VF still has some config and info need clean,
+ * which configed by PF.
+ */
+static void hclge_clear_vport_vf_info(struct hclge_vport *vport, int vfid)
+{
+	struct hclge_dev *hdev = vport->back;
+	struct hclge_vlan_info vlan_info;
+	int ret;
+
+	clear_bit(HCLGE_VPORT_STATE_INITED, &vport->state);
+	clear_bit(HCLGE_VPORT_STATE_ALIVE, &vport->state);
+	vport->need_notify = 0;
+	vport->mps = 0;
+
+	/* after disable sriov, clean VF rate configured by PF */
+	ret = hclge_tm_qs_shaper_cfg(vport, 0);
+	if (ret)
+		dev_err(&hdev->pdev->dev,
+			"failed to clean vf%d rate config, ret = %d\n",
+			vfid, ret);
+
+	vlan_info.vlan_tag = 0;
+	vlan_info.qos = 0;
+	vlan_info.vlan_proto = ETH_P_8021Q;
+	ret = hclge_update_port_base_vlan_cfg(vport,
+					      HNAE3_PORT_BASE_VLAN_DISABLE,
+					      &vlan_info);
+	if (ret)
+		dev_err(&hdev->pdev->dev,
+			"failed to clean vf%d port base vlan, ret = %d\n",
+			vfid, ret);
+
+	ret = hclge_set_vf_spoofchk_hw(hdev, vport->vport_id, false);
+	if (ret)
+		dev_err(&hdev->pdev->dev,
+			"failed to clean vf%d spoof config, ret = %d\n",
+			vfid, ret);
+
+	memset(&vport->vf_info, 0, sizeof(vport->vf_info));
+}
+
+static void hclge_clean_vport_config(struct hnae3_ae_dev *ae_dev, int num_vfs)
+{
+	struct hclge_dev *hdev = ae_dev->priv;
+	struct hclge_vport *vport;
+	int i;
+
+	for (i = 0; i < num_vfs; i++) {
+		vport = &hdev->vport[i + HCLGE_VF_VPORT_START_NUM];
+
+		hclge_clear_vport_vf_info(vport, i);
+	}
+}
+
+static int hclge_get_dscp_prio(struct hnae3_handle *h, u8 dscp, u8 *tc_mode,
+			       u8 *priority)
+{
+	struct hclge_vport *vport = hclge_get_vport(h);
+
+	if (dscp >= HNAE3_MAX_DSCP)
+		return -EINVAL;
+
+	if (tc_mode)
+		*tc_mode = vport->nic.kinfo.tc_map_mode;
+	if (priority)
+		*priority = vport->nic.kinfo.dscp_prio[dscp] == HNAE3_PRIO_ID_INVALID ? 0 :
+			    vport->nic.kinfo.dscp_prio[dscp];
+
+	return 0;
+}
+
+static const struct hnae3_ae_ops hclge_ops = {
+	.init_ae_dev = hclge_init_ae_dev,
+	.uninit_ae_dev = hclge_uninit_ae_dev,
+	.reset_prepare = hclge_reset_prepare_general,
+	.reset_done = hclge_reset_done,
+	.init_client_instance = hclge_init_client_instance,
+	.uninit_client_instance = hclge_uninit_client_instance,
+	.map_ring_to_vector = hclge_map_ring_to_vector,
+	.unmap_ring_from_vector = hclge_unmap_ring_frm_vector,
+	.get_vector = hclge_get_vector,
+	.put_vector = hclge_put_vector,
+	.set_promisc_mode = hclge_set_promisc_mode,
+	.request_update_promisc_mode = hclge_request_update_promisc_mode,
+	.set_loopback = hclge_set_loopback,
+	.start = hclge_ae_start,
+	.stop = hclge_ae_stop,
+	.client_start = hclge_client_start,
+	.client_stop = hclge_client_stop,
+	.get_status = hclge_get_status,
+	.get_ksettings_an_result = hclge_get_ksettings_an_result,
+	.cfg_mac_speed_dup_h = hclge_cfg_mac_speed_dup_h,
+	.get_media_type = hclge_get_media_type,
+	.check_port_speed = hclge_check_port_speed,
+	.get_fec_stats = hclge_get_fec_stats,
+	.get_fec = hclge_get_fec,
+	.set_fec = hclge_set_fec,
+	.get_rss_key_size = hclge_comm_get_rss_key_size,
+	.get_rss = hclge_get_rss,
+	.set_rss = hclge_set_rss,
+	.set_rss_tuple = hclge_set_rss_tuple,
+	.get_rss_tuple = hclge_get_rss_tuple,
+	.get_tc_size = hclge_get_tc_size,
+	.get_mac_addr = hclge_get_mac_addr,
+	.set_mac_addr = hclge_set_mac_addr,
+	.do_ioctl = hclge_do_ioctl,
+	.add_uc_addr = hclge_add_uc_addr,
+	.rm_uc_addr = hclge_rm_uc_addr,
+	.add_mc_addr = hclge_add_mc_addr,
+	.rm_mc_addr = hclge_rm_mc_addr,
+	.set_autoneg = hclge_set_autoneg,
+	.get_autoneg = hclge_get_autoneg,
+	.restart_autoneg = hclge_restart_autoneg,
+	.halt_autoneg = hclge_halt_autoneg,
+	.get_pauseparam = hclge_get_pauseparam,
+	.set_pauseparam = hclge_set_pauseparam,
+	.set_mtu = hclge_set_mtu,
+	.reset_queue = hclge_reset_tqp,
+	.get_stats = hclge_get_stats,
+	.get_mac_stats = hclge_get_mac_stat,
+	.update_stats = hclge_update_stats,
+	.get_strings = hclge_get_strings,
+	.get_sset_count = hclge_get_sset_count,
+	.get_fw_version = hclge_get_fw_version,
+	.get_mdix_mode = hclge_get_mdix_mode,
+	.enable_vlan_filter = hclge_enable_vlan_filter,
+	.set_vlan_filter = hclge_set_vlan_filter,
+	.set_vf_vlan_filter = hclge_set_vf_vlan_filter,
+	.enable_hw_strip_rxvtag = hclge_en_hw_strip_rxvtag,
+	.reset_event = hclge_reset_event,
+	.get_reset_level = hclge_get_reset_level,
+	.set_default_reset_request = hclge_set_def_reset_request,
+	.get_tqps_and_rss_info = hclge_get_tqps_and_rss_info,
+	.set_channels = hclge_set_channels,
+	.get_channels = hclge_get_channels,
+	.get_regs_len = hclge_get_regs_len,
+	.get_regs = hclge_get_regs,
+	.set_led_id = hclge_set_led_id,
+	.get_link_mode = hclge_get_link_mode,
+	.add_fd_entry = hclge_add_fd_entry,
+	.del_fd_entry = hclge_del_fd_entry,
+	.get_fd_rule_cnt = hclge_get_fd_rule_cnt,
+	.get_fd_rule_info = hclge_get_fd_rule_info,
+	.get_fd_all_rules = hclge_get_all_rules,
+	.enable_fd = hclge_enable_fd,
+	.add_arfs_entry = hclge_add_fd_entry_by_arfs,
+	.dbg_get_read_func = hclge_dbg_get_read_func,
+	.handle_hw_ras_error = hclge_handle_hw_ras_error,
+	.get_hw_reset_stat = hclge_get_hw_reset_stat,
+	.ae_dev_resetting = hclge_ae_dev_resetting,
+	.ae_dev_reset_cnt = hclge_ae_dev_reset_cnt,
+	.set_gro_en = hclge_gro_en,
+	.get_global_queue_id = hclge_covert_handle_qid_global,
+	.set_timer_task = hclge_set_timer_task,
+	.mac_connect_phy = hclge_mac_connect_phy,
+	.mac_disconnect_phy = hclge_mac_disconnect_phy,
+	.get_vf_config = hclge_get_vf_config,
+	.set_vf_link_state = hclge_set_vf_link_state,
+	.set_vf_spoofchk = hclge_set_vf_spoofchk,
+	.set_vf_trust = hclge_set_vf_trust,
+	.set_vf_rate = hclge_set_vf_rate,
+	.set_vf_mac = hclge_set_vf_mac,
+	.get_module_eeprom = hclge_get_module_eeprom,
+	.get_cmdq_stat = hclge_get_cmdq_stat,
+	.add_cls_flower = hclge_add_cls_flower,
+	.del_cls_flower = hclge_del_cls_flower,
+	.cls_flower_active = hclge_is_cls_flower_active,
+	.get_phy_link_ksettings = hclge_get_phy_link_ksettings,
+	.set_phy_link_ksettings = hclge_set_phy_link_ksettings,
+	.set_tx_hwts_info = hclge_ptp_set_tx_info,
+	.get_rx_hwts = hclge_ptp_get_rx_hwts,
+	.get_ts_info = hclge_ptp_get_ts_info,
+	.get_link_diagnosis_info = hclge_get_link_diagnosis_info,
+	.clean_vf_config = hclge_clean_vport_config,
+	.get_dscp_prio = hclge_get_dscp_prio,
+	.get_wol = hclge_get_wol,
+	.set_wol = hclge_set_wol,
+};
+
+static struct hnae3_ae_algo ae_algo = {
+	.ops = &hclge_ops,
+	.pdev_id_table = ae_algo_pci_tbl,
+};
+
+static int __init hclge_init(void)
+{
+	pr_debug("%s is initializing\n", HCLGE_NAME);
+
+	hclge_wq = alloc_workqueue("%s", WQ_UNBOUND, 0,
+				   HCLGE_NAME);
+	if (!hclge_wq) {
+		pr_err("%s: failed to create workqueue\n", HCLGE_NAME);
+		return -ENOMEM;
+	}
+
+	hnae3_register_ae_algo(&ae_algo);
+
+	return 0;
+}
+
+static void __exit hclge_exit(void)
+{
+	hnae3_acquire_unload_lock();
+	hnae3_unregister_ae_algo_prepare(&ae_algo);
+	hnae3_unregister_ae_algo(&ae_algo);
+	destroy_workqueue(hclge_wq);
+	hnae3_release_unload_lock();
+}
+module_init(hclge_init);
+module_exit(hclge_exit);
+
+MODULE_LICENSE("GPL");
+MODULE_AUTHOR("Huawei Tech. Co., Ltd.");
+MODULE_DESCRIPTION("HCLGE Driver");
+MODULE_VERSION(HCLGE_MOD_VERSION);
diff --git a/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_main.h b/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_main.h
new file mode 100644
index 000000000000..032b472d2368
--- /dev/null
+++ b/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_main.h
@@ -0,0 +1,1178 @@
+/* SPDX-License-Identifier: GPL-2.0+ */
+// Copyright (c) 2016-2017 Hisilicon Limited.
+
+#ifndef __HCLGE_MAIN_H
+#define __HCLGE_MAIN_H
+#include <linux/fs.h>
+#include <linux/types.h>
+#include <linux/phy.h>
+#include <linux/if_vlan.h>
+#include <linux/kfifo.h>
+
+#include <net/devlink.h>
+#include <net/ipv6.h>
+
+#include "hclge_cmd.h"
+#include "hclge_ptp.h"
+#include "hnae3.h"
+#include "hclge_comm_rss.h"
+#include "hclge_comm_tqp_stats.h"
+
+#define HCLGE_MOD_VERSION "1.0"
+#define HCLGE_DRIVER_NAME "hclge"
+
+#define HCLGE_MAX_PF_NUM		8
+
+#define HCLGE_VF_VPORT_START_NUM	1
+
+#define HCLGE_RD_FIRST_STATS_NUM        2
+#define HCLGE_RD_OTHER_STATS_NUM        4
+
+#define HCLGE_INVALID_VPORT 0xffff
+
+#define HCLGE_PF_CFG_BLOCK_SIZE		32
+#define HCLGE_PF_CFG_DESC_NUM \
+	(HCLGE_PF_CFG_BLOCK_SIZE / HCLGE_CFG_RD_LEN_BYTES)
+
+#define HCLGE_VECTOR_REG_BASE		0x20000
+#define HCLGE_VECTOR_EXT_REG_BASE	0x30000
+#define HCLGE_MISC_VECTOR_REG_BASE	0x20400
+
+#define HCLGE_VECTOR_REG_OFFSET		0x4
+#define HCLGE_VECTOR_REG_OFFSET_H	0x1000
+#define HCLGE_VECTOR_VF_OFFSET		0x100000
+
+#define HCLGE_NIC_CSQ_DEPTH_REG		0x27008
+
+/* bar registers for common func */
+#define HCLGE_GRO_EN_REG		0x28000
+#define HCLGE_RXD_ADV_LAYOUT_EN_REG	0x28008
+
+/* bar registers for rcb */
+#define HCLGE_RING_RX_ADDR_L_REG	0x80000
+#define HCLGE_RING_RX_ADDR_H_REG	0x80004
+#define HCLGE_RING_RX_BD_NUM_REG	0x80008
+#define HCLGE_RING_RX_BD_LENGTH_REG	0x8000C
+#define HCLGE_RING_RX_MERGE_EN_REG	0x80014
+#define HCLGE_RING_RX_TAIL_REG		0x80018
+#define HCLGE_RING_RX_HEAD_REG		0x8001C
+#define HCLGE_RING_RX_FBD_NUM_REG	0x80020
+#define HCLGE_RING_RX_OFFSET_REG	0x80024
+#define HCLGE_RING_RX_FBD_OFFSET_REG	0x80028
+#define HCLGE_RING_RX_STASH_REG		0x80030
+#define HCLGE_RING_RX_BD_ERR_REG	0x80034
+#define HCLGE_RING_TX_ADDR_L_REG	0x80040
+#define HCLGE_RING_TX_ADDR_H_REG	0x80044
+#define HCLGE_RING_TX_BD_NUM_REG	0x80048
+#define HCLGE_RING_TX_PRIORITY_REG	0x8004C
+#define HCLGE_RING_TX_TC_REG		0x80050
+#define HCLGE_RING_TX_MERGE_EN_REG	0x80054
+#define HCLGE_RING_TX_TAIL_REG		0x80058
+#define HCLGE_RING_TX_HEAD_REG		0x8005C
+#define HCLGE_RING_TX_FBD_NUM_REG	0x80060
+#define HCLGE_RING_TX_OFFSET_REG	0x80064
+#define HCLGE_RING_TX_EBD_NUM_REG	0x80068
+#define HCLGE_RING_TX_EBD_OFFSET_REG	0x80070
+#define HCLGE_RING_TX_BD_ERR_REG	0x80074
+#define HCLGE_RING_EN_REG		0x80090
+
+/* bar registers for tqp interrupt */
+#define HCLGE_TQP_INTR_CTRL_REG		0x20000
+#define HCLGE_TQP_INTR_GL0_REG		0x20100
+#define HCLGE_TQP_INTR_GL1_REG		0x20200
+#define HCLGE_TQP_INTR_GL2_REG		0x20300
+#define HCLGE_TQP_INTR_RL_REG		0x20900
+
+#define HCLGE_RSS_IND_TBL_SIZE		512
+
+#define HCLGE_RSS_TC_SIZE_0		1
+#define HCLGE_RSS_TC_SIZE_1		2
+#define HCLGE_RSS_TC_SIZE_2		4
+#define HCLGE_RSS_TC_SIZE_3		8
+#define HCLGE_RSS_TC_SIZE_4		16
+#define HCLGE_RSS_TC_SIZE_5		32
+#define HCLGE_RSS_TC_SIZE_6		64
+#define HCLGE_RSS_TC_SIZE_7		128
+
+#define HCLGE_UMV_TBL_SIZE		3072
+#define HCLGE_DEFAULT_UMV_SPACE_PER_PF \
+	(HCLGE_UMV_TBL_SIZE / HCLGE_MAX_PF_NUM)
+
+#define HCLGE_TQP_RESET_TRY_TIMES	200
+
+#define HCLGE_PHY_PAGE_MDIX		0
+#define HCLGE_PHY_PAGE_COPPER		0
+
+/* Page Selection Reg. */
+#define HCLGE_PHY_PAGE_REG		22
+
+/* Copper Specific Control Register */
+#define HCLGE_PHY_CSC_REG		16
+
+/* Copper Specific Status Register */
+#define HCLGE_PHY_CSS_REG		17
+
+#define HCLGE_PHY_MDIX_CTRL_S		5
+#define HCLGE_PHY_MDIX_CTRL_M		GENMASK(6, 5)
+
+#define HCLGE_PHY_MDIX_STATUS_B		6
+#define HCLGE_PHY_SPEED_DUP_RESOLVE_B	11
+
+#define HCLGE_GET_DFX_REG_TYPE_CNT	4
+
+/* Factor used to calculate offset and bitmap of VF num */
+#define HCLGE_VF_NUM_PER_CMD           64
+
+#define HCLGE_MAX_QSET_NUM		1024
+
+#define HCLGE_DBG_RESET_INFO_LEN	1024
+
+enum HLCGE_PORT_TYPE {
+	HOST_PORT,
+	NETWORK_PORT
+};
+
+#define PF_VPORT_ID			0
+
+#define HCLGE_PF_ID_S			0
+#define HCLGE_PF_ID_M			GENMASK(2, 0)
+#define HCLGE_VF_ID_S			3
+#define HCLGE_VF_ID_M			GENMASK(10, 3)
+#define HCLGE_PORT_TYPE_B		11
+#define HCLGE_NETWORK_PORT_ID_S		0
+#define HCLGE_NETWORK_PORT_ID_M		GENMASK(3, 0)
+
+/* Reset related Registers */
+#define HCLGE_PF_OTHER_INT_REG		0x20600
+#define HCLGE_MISC_RESET_STS_REG	0x20700
+#define HCLGE_MISC_VECTOR_INT_STS	0x20800
+#define HCLGE_GLOBAL_RESET_REG		0x20A00
+#define HCLGE_GLOBAL_RESET_BIT		0
+#define HCLGE_CORE_RESET_BIT		1
+#define HCLGE_IMP_RESET_BIT		2
+#define HCLGE_RESET_INT_M		GENMASK(7, 5)
+#define HCLGE_FUN_RST_ING		0x20C00
+#define HCLGE_FUN_RST_ING_B		0
+
+/* Vector0 register bits define */
+#define HCLGE_VECTOR0_REG_PTP_INT_B	0
+#define HCLGE_VECTOR0_GLOBALRESET_INT_B	5
+#define HCLGE_VECTOR0_CORERESET_INT_B	6
+#define HCLGE_VECTOR0_IMPRESET_INT_B	7
+
+/* Vector0 interrupt CMDQ event source register(RW) */
+#define HCLGE_VECTOR0_CMDQ_SRC_REG	0x27100
+/* CMDQ register bits for RX event(=MBX event) */
+#define HCLGE_VECTOR0_RX_CMDQ_INT_B	1
+
+#define HCLGE_VECTOR0_IMP_RESET_INT_B	1
+#define HCLGE_VECTOR0_IMP_CMDQ_ERR_B	4U
+#define HCLGE_VECTOR0_IMP_RD_POISON_B	5U
+#define HCLGE_VECTOR0_ALL_MSIX_ERR_B	6U
+#define HCLGE_TRIGGER_IMP_RESET_B	7U
+
+#define HCLGE_TQP_MEM_SIZE		0x10000
+#define HCLGE_MEM_BAR			4
+/* in the bar4, the first half is for roce, and the second half is for nic */
+#define HCLGE_NIC_MEM_OFFSET(hdev)	\
+	(pci_resource_len((hdev)->pdev, HCLGE_MEM_BAR) >> 1)
+#define HCLGE_TQP_MEM_OFFSET(hdev, i)	\
+	(HCLGE_NIC_MEM_OFFSET(hdev) + HCLGE_TQP_MEM_SIZE * (i))
+
+#define HCLGE_MAC_DEFAULT_FRAME \
+	(ETH_HLEN + ETH_FCS_LEN + 2 * VLAN_HLEN + ETH_DATA_LEN)
+#define HCLGE_MAC_MIN_FRAME		64
+#define HCLGE_MAC_MAX_FRAME		9728
+
+#define HCLGE_SUPPORT_1G_BIT		BIT(0)
+#define HCLGE_SUPPORT_10G_BIT		BIT(1)
+#define HCLGE_SUPPORT_25G_BIT		BIT(2)
+#define HCLGE_SUPPORT_50G_R2_BIT	BIT(3)
+#define HCLGE_SUPPORT_100G_R4_BIT	BIT(4)
+/* to be compatible with exsit board */
+#define HCLGE_SUPPORT_40G_BIT		BIT(5)
+#define HCLGE_SUPPORT_100M_BIT		BIT(6)
+#define HCLGE_SUPPORT_10M_BIT		BIT(7)
+#define HCLGE_SUPPORT_200G_R4_EXT_BIT	BIT(8)
+#define HCLGE_SUPPORT_50G_R1_BIT	BIT(9)
+#define HCLGE_SUPPORT_100G_R2_BIT	BIT(10)
+#define HCLGE_SUPPORT_200G_R4_BIT	BIT(11)
+
+#define HCLGE_SUPPORT_GE \
+	(HCLGE_SUPPORT_1G_BIT | HCLGE_SUPPORT_100M_BIT | HCLGE_SUPPORT_10M_BIT)
+#define HCLGE_SUPPORT_50G_BITS \
+	(HCLGE_SUPPORT_50G_R2_BIT | HCLGE_SUPPORT_50G_R1_BIT)
+#define HCLGE_SUPPORT_100G_BITS \
+	(HCLGE_SUPPORT_100G_R4_BIT | HCLGE_SUPPORT_100G_R2_BIT)
+#define HCLGE_SUPPORT_200G_BITS \
+	(HCLGE_SUPPORT_200G_R4_EXT_BIT | HCLGE_SUPPORT_200G_R4_BIT)
+
+enum HCLGE_DEV_STATE {
+	HCLGE_STATE_REINITING,
+	HCLGE_STATE_DOWN,
+	HCLGE_STATE_DISABLED,
+	HCLGE_STATE_REMOVING,
+	HCLGE_STATE_NIC_REGISTERED,
+	HCLGE_STATE_ROCE_REGISTERED,
+	HCLGE_STATE_SERVICE_INITED,
+	HCLGE_STATE_RST_SERVICE_SCHED,
+	HCLGE_STATE_RST_HANDLING,
+	HCLGE_STATE_MBX_SERVICE_SCHED,
+	HCLGE_STATE_MBX_HANDLING,
+	HCLGE_STATE_ERR_SERVICE_SCHED,
+	HCLGE_STATE_STATISTICS_UPDATING,
+	HCLGE_STATE_LINK_UPDATING,
+	HCLGE_STATE_RST_FAIL,
+	HCLGE_STATE_FD_TBL_CHANGED,
+	HCLGE_STATE_FD_CLEAR_ALL,
+	HCLGE_STATE_FD_USER_DEF_CHANGED,
+	HCLGE_STATE_PTP_EN,
+	HCLGE_STATE_PTP_TX_HANDLING,
+	HCLGE_STATE_FEC_STATS_UPDATING,
+	HCLGE_STATE_MAX
+};
+
+enum hclge_evt_cause {
+	HCLGE_VECTOR0_EVENT_RST,
+	HCLGE_VECTOR0_EVENT_MBX,
+	HCLGE_VECTOR0_EVENT_ERR,
+	HCLGE_VECTOR0_EVENT_PTP,
+	HCLGE_VECTOR0_EVENT_OTHER,
+};
+
+enum HCLGE_MAC_SPEED {
+	HCLGE_MAC_SPEED_UNKNOWN = 0,		/* unknown */
+	HCLGE_MAC_SPEED_10M	= 10,		/* 10 Mbps */
+	HCLGE_MAC_SPEED_100M	= 100,		/* 100 Mbps */
+	HCLGE_MAC_SPEED_1G	= 1000,		/* 1000 Mbps   = 1 Gbps */
+	HCLGE_MAC_SPEED_10G	= 10000,	/* 10000 Mbps  = 10 Gbps */
+	HCLGE_MAC_SPEED_25G	= 25000,	/* 25000 Mbps  = 25 Gbps */
+	HCLGE_MAC_SPEED_40G	= 40000,	/* 40000 Mbps  = 40 Gbps */
+	HCLGE_MAC_SPEED_50G	= 50000,	/* 50000 Mbps  = 50 Gbps */
+	HCLGE_MAC_SPEED_100G	= 100000,	/* 100000 Mbps = 100 Gbps */
+	HCLGE_MAC_SPEED_200G	= 200000	/* 200000 Mbps = 200 Gbps */
+};
+
+enum HCLGE_MAC_DUPLEX {
+	HCLGE_MAC_HALF,
+	HCLGE_MAC_FULL
+};
+
+/* hilink version */
+enum hclge_hilink_version {
+	HCLGE_HILINK_H32 = 0,
+	HCLGE_HILINK_H60 = 1,
+};
+
+#define QUERY_SFP_SPEED		0
+#define QUERY_ACTIVE_SPEED	1
+
+struct hclge_wol_info {
+	u32 wol_support_mode; /* store the wake on lan info */
+	u32 wol_current_mode;
+	u8 wol_sopass[SOPASS_MAX];
+	u8 wol_sopass_size;
+};
+
+struct hclge_mac {
+	u8 mac_id;
+	u8 phy_addr;
+	u8 flag;
+	u8 media_type;	/* port media type, e.g. fibre/copper/backplane */
+	u8 mac_addr[ETH_ALEN];
+	u8 autoneg;
+	u8 req_autoneg;
+	u8 duplex;
+	u8 req_duplex;
+	u8 support_autoneg;
+	u8 speed_type;	/* 0: sfp speed, 1: active speed */
+	u8 lane_num;
+	u32 speed;
+	u32 req_speed;
+	u32 max_speed;
+	u32 speed_ability; /* speed ability supported by current media */
+	u32 module_type; /* sub media type, e.g. kr/cr/sr/lr */
+	u32 fec_mode; /* active fec mode */
+	u32 user_fec_mode;
+	u32 fec_ability;
+	int link;	/* store the link status of mac & phy (if phy exists) */
+	struct hclge_wol_info wol;
+	struct phy_device *phydev;
+	struct mii_bus *mdio_bus;
+	phy_interface_t phy_if;
+	__ETHTOOL_DECLARE_LINK_MODE_MASK(supported);
+	__ETHTOOL_DECLARE_LINK_MODE_MASK(advertising);
+};
+
+struct hclge_hw {
+	struct hclge_comm_hw hw;
+	struct hclge_mac mac;
+	int num_vec;
+};
+
+enum hclge_fc_mode {
+	HCLGE_FC_NONE,
+	HCLGE_FC_RX_PAUSE,
+	HCLGE_FC_TX_PAUSE,
+	HCLGE_FC_FULL,
+	HCLGE_FC_PFC,
+	HCLGE_FC_DEFAULT
+};
+
+#define HCLGE_FILTER_TYPE_VF		0
+#define HCLGE_FILTER_TYPE_PORT		1
+#define HCLGE_FILTER_FE_EGRESS_V1_B	BIT(0)
+#define HCLGE_FILTER_FE_NIC_INGRESS_B	BIT(0)
+#define HCLGE_FILTER_FE_NIC_EGRESS_B	BIT(1)
+#define HCLGE_FILTER_FE_ROCE_INGRESS_B	BIT(2)
+#define HCLGE_FILTER_FE_ROCE_EGRESS_B	BIT(3)
+#define HCLGE_FILTER_FE_EGRESS		(HCLGE_FILTER_FE_NIC_EGRESS_B \
+					| HCLGE_FILTER_FE_ROCE_EGRESS_B)
+#define HCLGE_FILTER_FE_INGRESS		(HCLGE_FILTER_FE_NIC_INGRESS_B \
+					| HCLGE_FILTER_FE_ROCE_INGRESS_B)
+
+enum hclge_vlan_fltr_cap {
+	HCLGE_VLAN_FLTR_DEF,
+	HCLGE_VLAN_FLTR_CAN_MDF,
+};
+enum hclge_link_fail_code {
+	HCLGE_LF_NORMAL,
+	HCLGE_LF_REF_CLOCK_LOST,
+	HCLGE_LF_XSFP_TX_DISABLE,
+	HCLGE_LF_XSFP_ABSENT,
+};
+
+#define HCLGE_LINK_STATUS_DOWN 0
+#define HCLGE_LINK_STATUS_UP   1
+
+#define HCLGE_PG_NUM		4
+#define HCLGE_SCH_MODE_SP	0
+#define HCLGE_SCH_MODE_DWRR	1
+struct hclge_pg_info {
+	u8 pg_id;
+	u8 pg_sch_mode;		/* 0: sp; 1: dwrr */
+	u8 tc_bit_map;
+	u32 bw_limit;
+	u8 tc_dwrr[HNAE3_MAX_TC];
+};
+
+struct hclge_tc_info {
+	u8 tc_id;
+	u8 tc_sch_mode;		/* 0: sp; 1: dwrr */
+	u8 pgid;
+	u32 bw_limit;
+};
+
+struct hclge_cfg {
+	u8 tc_num;
+	u8 vlan_fliter_cap;
+	u16 tqp_desc_num;
+	u16 rx_buf_len;
+	u16 vf_rss_size_max;
+	u16 pf_rss_size_max;
+	u8 phy_addr;
+	u8 media_type;
+	u8 mac_addr[ETH_ALEN];
+	u8 default_speed;
+	u32 numa_node_map;
+	u32 tx_spare_buf_size;
+	u16 speed_ability;
+	u16 umv_space;
+};
+
+struct hclge_tm_info {
+	u8 num_tc;
+	u8 num_pg;      /* It must be 1 if vNET-Base schd */
+	u8 pg_dwrr[HCLGE_PG_NUM];
+	u8 prio_tc[HNAE3_MAX_USER_PRIO];
+	struct hclge_pg_info pg_info[HCLGE_PG_NUM];
+	struct hclge_tc_info tc_info[HNAE3_MAX_TC];
+	enum hclge_fc_mode fc_mode;
+	u8 hw_pfc_map; /* Allow for packet drop or not on this TC */
+	u8 pfc_en;	/* PFC enabled or not for user priority */
+};
+
+/* max number of mac statistics on each version */
+#define HCLGE_MAC_STATS_MAX_NUM_V1		87
+#define HCLGE_MAC_STATS_MAX_NUM_V2		105
+
+struct hclge_comm_stats_str {
+	char desc[ETH_GSTRING_LEN];
+	u32 stats_num;
+	unsigned long offset;
+};
+
+/* mac stats ,opcode id: 0x0032 */
+struct hclge_mac_stats {
+	u64 mac_tx_mac_pause_num;
+	u64 mac_rx_mac_pause_num;
+	u64 rsv0;
+	u64 mac_tx_pfc_pri0_pkt_num;
+	u64 mac_tx_pfc_pri1_pkt_num;
+	u64 mac_tx_pfc_pri2_pkt_num;
+	u64 mac_tx_pfc_pri3_pkt_num;
+	u64 mac_tx_pfc_pri4_pkt_num;
+	u64 mac_tx_pfc_pri5_pkt_num;
+	u64 mac_tx_pfc_pri6_pkt_num;
+	u64 mac_tx_pfc_pri7_pkt_num;
+	u64 mac_rx_pfc_pri0_pkt_num;
+	u64 mac_rx_pfc_pri1_pkt_num;
+	u64 mac_rx_pfc_pri2_pkt_num;
+	u64 mac_rx_pfc_pri3_pkt_num;
+	u64 mac_rx_pfc_pri4_pkt_num;
+	u64 mac_rx_pfc_pri5_pkt_num;
+	u64 mac_rx_pfc_pri6_pkt_num;
+	u64 mac_rx_pfc_pri7_pkt_num;
+	u64 mac_tx_total_pkt_num;
+	u64 mac_tx_total_oct_num;
+	u64 mac_tx_good_pkt_num;
+	u64 mac_tx_bad_pkt_num;
+	u64 mac_tx_good_oct_num;
+	u64 mac_tx_bad_oct_num;
+	u64 mac_tx_uni_pkt_num;
+	u64 mac_tx_multi_pkt_num;
+	u64 mac_tx_broad_pkt_num;
+	u64 mac_tx_undersize_pkt_num;
+	u64 mac_tx_oversize_pkt_num;
+	u64 mac_tx_64_oct_pkt_num;
+	u64 mac_tx_65_127_oct_pkt_num;
+	u64 mac_tx_128_255_oct_pkt_num;
+	u64 mac_tx_256_511_oct_pkt_num;
+	u64 mac_tx_512_1023_oct_pkt_num;
+	u64 mac_tx_1024_1518_oct_pkt_num;
+	u64 mac_tx_1519_2047_oct_pkt_num;
+	u64 mac_tx_2048_4095_oct_pkt_num;
+	u64 mac_tx_4096_8191_oct_pkt_num;
+	u64 rsv1;
+	u64 mac_tx_8192_9216_oct_pkt_num;
+	u64 mac_tx_9217_12287_oct_pkt_num;
+	u64 mac_tx_12288_16383_oct_pkt_num;
+	u64 mac_tx_1519_max_good_oct_pkt_num;
+	u64 mac_tx_1519_max_bad_oct_pkt_num;
+
+	u64 mac_rx_total_pkt_num;
+	u64 mac_rx_total_oct_num;
+	u64 mac_rx_good_pkt_num;
+	u64 mac_rx_bad_pkt_num;
+	u64 mac_rx_good_oct_num;
+	u64 mac_rx_bad_oct_num;
+	u64 mac_rx_uni_pkt_num;
+	u64 mac_rx_multi_pkt_num;
+	u64 mac_rx_broad_pkt_num;
+	u64 mac_rx_undersize_pkt_num;
+	u64 mac_rx_oversize_pkt_num;
+	u64 mac_rx_64_oct_pkt_num;
+	u64 mac_rx_65_127_oct_pkt_num;
+	u64 mac_rx_128_255_oct_pkt_num;
+	u64 mac_rx_256_511_oct_pkt_num;
+	u64 mac_rx_512_1023_oct_pkt_num;
+	u64 mac_rx_1024_1518_oct_pkt_num;
+	u64 mac_rx_1519_2047_oct_pkt_num;
+	u64 mac_rx_2048_4095_oct_pkt_num;
+	u64 mac_rx_4096_8191_oct_pkt_num;
+	u64 rsv2;
+	u64 mac_rx_8192_9216_oct_pkt_num;
+	u64 mac_rx_9217_12287_oct_pkt_num;
+	u64 mac_rx_12288_16383_oct_pkt_num;
+	u64 mac_rx_1519_max_good_oct_pkt_num;
+	u64 mac_rx_1519_max_bad_oct_pkt_num;
+
+	u64 mac_tx_fragment_pkt_num;
+	u64 mac_tx_undermin_pkt_num;
+	u64 mac_tx_jabber_pkt_num;
+	u64 mac_tx_err_all_pkt_num;
+	u64 mac_tx_from_app_good_pkt_num;
+	u64 mac_tx_from_app_bad_pkt_num;
+	u64 mac_rx_fragment_pkt_num;
+	u64 mac_rx_undermin_pkt_num;
+	u64 mac_rx_jabber_pkt_num;
+	u64 mac_rx_fcs_err_pkt_num;
+	u64 mac_rx_send_app_good_pkt_num;
+	u64 mac_rx_send_app_bad_pkt_num;
+	u64 mac_tx_pfc_pause_pkt_num;
+	u64 mac_rx_pfc_pause_pkt_num;
+	u64 mac_tx_ctrl_pkt_num;
+	u64 mac_rx_ctrl_pkt_num;
+
+	/* duration of pfc */
+	u64 mac_tx_pfc_pri0_xoff_time;
+	u64 mac_tx_pfc_pri1_xoff_time;
+	u64 mac_tx_pfc_pri2_xoff_time;
+	u64 mac_tx_pfc_pri3_xoff_time;
+	u64 mac_tx_pfc_pri4_xoff_time;
+	u64 mac_tx_pfc_pri5_xoff_time;
+	u64 mac_tx_pfc_pri6_xoff_time;
+	u64 mac_tx_pfc_pri7_xoff_time;
+	u64 mac_rx_pfc_pri0_xoff_time;
+	u64 mac_rx_pfc_pri1_xoff_time;
+	u64 mac_rx_pfc_pri2_xoff_time;
+	u64 mac_rx_pfc_pri3_xoff_time;
+	u64 mac_rx_pfc_pri4_xoff_time;
+	u64 mac_rx_pfc_pri5_xoff_time;
+	u64 mac_rx_pfc_pri6_xoff_time;
+	u64 mac_rx_pfc_pri7_xoff_time;
+
+	/* duration of pause */
+	u64 mac_tx_pause_xoff_time;
+	u64 mac_rx_pause_xoff_time;
+};
+
+#define HCLGE_STATS_TIMER_INTERVAL	300UL
+
+/* fec stats ,opcode id: 0x0316 */
+#define HCLGE_FEC_STATS_MAX_LANES	8
+struct hclge_fec_stats {
+	/* fec rs mode total stats */
+	u64 rs_corr_blocks;
+	u64 rs_uncorr_blocks;
+	u64 rs_error_blocks;
+	/* fec base-r mode per lanes stats */
+	u64 base_r_lane_num;
+	u64 base_r_corr_blocks;
+	u64 base_r_uncorr_blocks;
+	union {
+		struct {
+			u64 base_r_corr_per_lanes[HCLGE_FEC_STATS_MAX_LANES];
+			u64 base_r_uncorr_per_lanes[HCLGE_FEC_STATS_MAX_LANES];
+		};
+		u64 per_lanes[HCLGE_FEC_STATS_MAX_LANES * 2];
+	};
+};
+
+struct hclge_vlan_type_cfg {
+	u16 rx_ot_fst_vlan_type;
+	u16 rx_ot_sec_vlan_type;
+	u16 rx_in_fst_vlan_type;
+	u16 rx_in_sec_vlan_type;
+	u16 tx_ot_vlan_type;
+	u16 tx_in_vlan_type;
+};
+
+enum HCLGE_FD_MODE {
+	HCLGE_FD_MODE_DEPTH_2K_WIDTH_400B_STAGE_1,
+	HCLGE_FD_MODE_DEPTH_1K_WIDTH_400B_STAGE_2,
+	HCLGE_FD_MODE_DEPTH_4K_WIDTH_200B_STAGE_1,
+	HCLGE_FD_MODE_DEPTH_2K_WIDTH_200B_STAGE_2,
+};
+
+enum HCLGE_FD_KEY_TYPE {
+	HCLGE_FD_KEY_BASE_ON_PTYPE,
+	HCLGE_FD_KEY_BASE_ON_TUPLE,
+};
+
+enum HCLGE_FD_STAGE {
+	HCLGE_FD_STAGE_1,
+	HCLGE_FD_STAGE_2,
+	MAX_STAGE_NUM,
+};
+
+/* OUTER_XXX indicates tuples in tunnel header of tunnel packet
+ * INNER_XXX indicate tuples in tunneled header of tunnel packet or
+ *           tuples of non-tunnel packet
+ */
+enum HCLGE_FD_TUPLE {
+	OUTER_DST_MAC,
+	OUTER_SRC_MAC,
+	OUTER_VLAN_TAG_FST,
+	OUTER_VLAN_TAG_SEC,
+	OUTER_ETH_TYPE,
+	OUTER_L2_RSV,
+	OUTER_IP_TOS,
+	OUTER_IP_PROTO,
+	OUTER_SRC_IP,
+	OUTER_DST_IP,
+	OUTER_L3_RSV,
+	OUTER_SRC_PORT,
+	OUTER_DST_PORT,
+	OUTER_L4_RSV,
+	OUTER_TUN_VNI,
+	OUTER_TUN_FLOW_ID,
+	INNER_DST_MAC,
+	INNER_SRC_MAC,
+	INNER_VLAN_TAG_FST,
+	INNER_VLAN_TAG_SEC,
+	INNER_ETH_TYPE,
+	INNER_L2_RSV,
+	INNER_IP_TOS,
+	INNER_IP_PROTO,
+	INNER_SRC_IP,
+	INNER_DST_IP,
+	INNER_L3_RSV,
+	INNER_SRC_PORT,
+	INNER_DST_PORT,
+	INNER_L4_RSV,
+	MAX_TUPLE,
+};
+
+#define HCLGE_FD_TUPLE_USER_DEF_TUPLES \
+	(BIT(INNER_L2_RSV) | BIT(INNER_L3_RSV) | BIT(INNER_L4_RSV))
+
+enum HCLGE_FD_META_DATA {
+	PACKET_TYPE_ID,
+	IP_FRAGEMENT,
+	ROCE_TYPE,
+	NEXT_KEY,
+	VLAN_NUMBER,
+	SRC_VPORT,
+	DST_VPORT,
+	TUNNEL_PACKET,
+	MAX_META_DATA,
+};
+
+enum HCLGE_FD_KEY_OPT {
+	KEY_OPT_U8,
+	KEY_OPT_LE16,
+	KEY_OPT_LE32,
+	KEY_OPT_MAC,
+	KEY_OPT_IP,
+	KEY_OPT_VNI,
+};
+
+struct key_info {
+	u8 key_type;
+	u8 key_length; /* use bit as unit */
+	enum HCLGE_FD_KEY_OPT key_opt;
+	int offset;
+	int moffset;
+};
+
+#define MAX_KEY_LENGTH	400
+#define MAX_KEY_DWORDS	DIV_ROUND_UP(MAX_KEY_LENGTH / 8, 4)
+#define MAX_KEY_BYTES	(MAX_KEY_DWORDS * 4)
+#define MAX_META_DATA_LENGTH	32
+
+#define HCLGE_FD_MAX_USER_DEF_OFFSET	9000
+#define HCLGE_FD_USER_DEF_DATA		GENMASK(15, 0)
+#define HCLGE_FD_USER_DEF_OFFSET	GENMASK(15, 0)
+#define HCLGE_FD_USER_DEF_OFFSET_UNMASK	GENMASK(15, 0)
+
+/* assigned by firmware, the real filter number for each pf may be less */
+#define MAX_FD_FILTER_NUM	4096
+#define HCLGE_ARFS_EXPIRE_INTERVAL	5UL
+
+#define hclge_read_dev(a, reg) \
+	hclge_comm_read_reg((a)->hw.io_base, reg)
+#define hclge_write_dev(a, reg, value) \
+	hclge_comm_write_reg((a)->hw.io_base, reg, value)
+
+enum HCLGE_FD_ACTIVE_RULE_TYPE {
+	HCLGE_FD_RULE_NONE,
+	HCLGE_FD_ARFS_ACTIVE,
+	HCLGE_FD_EP_ACTIVE,
+	HCLGE_FD_TC_FLOWER_ACTIVE,
+};
+
+enum HCLGE_FD_PACKET_TYPE {
+	NIC_PACKET,
+	ROCE_PACKET,
+};
+
+enum HCLGE_FD_ACTION {
+	HCLGE_FD_ACTION_SELECT_QUEUE,
+	HCLGE_FD_ACTION_DROP_PACKET,
+	HCLGE_FD_ACTION_SELECT_TC,
+};
+
+enum HCLGE_FD_NODE_STATE {
+	HCLGE_FD_TO_ADD,
+	HCLGE_FD_TO_DEL,
+	HCLGE_FD_ACTIVE,
+	HCLGE_FD_DELETED,
+};
+
+enum HCLGE_FD_USER_DEF_LAYER {
+	HCLGE_FD_USER_DEF_NONE,
+	HCLGE_FD_USER_DEF_L2,
+	HCLGE_FD_USER_DEF_L3,
+	HCLGE_FD_USER_DEF_L4,
+};
+
+#define HCLGE_FD_USER_DEF_LAYER_NUM 3
+struct hclge_fd_user_def_cfg {
+	u16 ref_cnt;
+	u16 offset;
+};
+
+struct hclge_fd_user_def_info {
+	enum HCLGE_FD_USER_DEF_LAYER layer;
+	u16 data;
+	u16 data_mask;
+	u16 offset;
+};
+
+struct hclge_fd_key_cfg {
+	u8 key_sel;
+	u8 inner_sipv6_word_en;
+	u8 inner_dipv6_word_en;
+	u8 outer_sipv6_word_en;
+	u8 outer_dipv6_word_en;
+	u32 tuple_active;
+	u32 meta_data_active;
+};
+
+struct hclge_fd_cfg {
+	u8 fd_mode;
+	u16 max_key_length; /* use bit as unit */
+	u32 rule_num[MAX_STAGE_NUM]; /* rule entry number */
+	u16 cnt_num[MAX_STAGE_NUM]; /* rule hit counter number */
+	struct hclge_fd_key_cfg key_cfg[MAX_STAGE_NUM];
+	struct hclge_fd_user_def_cfg user_def_cfg[HCLGE_FD_USER_DEF_LAYER_NUM];
+};
+
+#define IPV4_INDEX	3
+
+struct hclge_fd_rule_tuples {
+	u8 src_mac[ETH_ALEN];
+	u8 dst_mac[ETH_ALEN];
+	/* Be compatible for ip address of both ipv4 and ipv6.
+	 * For ipv4 address, we store it in src/dst_ip[3].
+	 */
+	u32 src_ip[IPV6_ADDR_WORDS];
+	u32 dst_ip[IPV6_ADDR_WORDS];
+	u16 src_port;
+	u16 dst_port;
+	u16 vlan_tag1;
+	u16 ether_proto;
+	u16 l2_user_def;
+	u16 l3_user_def;
+	u32 l4_user_def;
+	u8 ip_tos;
+	u8 ip_proto;
+};
+
+struct hclge_fd_rule {
+	struct hlist_node rule_node;
+	struct hclge_fd_rule_tuples tuples;
+	struct hclge_fd_rule_tuples tuples_mask;
+	u32 unused_tuple;
+	u32 flow_type;
+	union {
+		struct {
+			unsigned long cookie;
+			u8 tc;
+		} cls_flower;
+		struct {
+			u16 flow_id; /* only used for arfs */
+		} arfs;
+		struct {
+			struct hclge_fd_user_def_info user_def;
+		} ep;
+	};
+	u16 queue_id;
+	u16 vf_id;
+	u16 location;
+	enum HCLGE_FD_ACTIVE_RULE_TYPE rule_type;
+	enum HCLGE_FD_NODE_STATE state;
+	u8 action;
+};
+
+struct hclge_fd_ad_data {
+	u16 ad_id;
+	u8 drop_packet;
+	u8 forward_to_direct_queue;
+	u16 queue_id;
+	u8 use_counter;
+	u8 counter_id;
+	u8 use_next_stage;
+	u8 write_rule_id_to_bd;
+	u8 next_input_key;
+	u16 rule_id;
+	u16 tc_size;
+	u8 override_tc;
+};
+
+enum HCLGE_MAC_NODE_STATE {
+	HCLGE_MAC_TO_ADD,
+	HCLGE_MAC_TO_DEL,
+	HCLGE_MAC_ACTIVE
+};
+
+struct hclge_mac_node {
+	struct list_head node;
+	enum HCLGE_MAC_NODE_STATE state;
+	u8 mac_addr[ETH_ALEN];
+};
+
+enum HCLGE_MAC_ADDR_TYPE {
+	HCLGE_MAC_ADDR_UC,
+	HCLGE_MAC_ADDR_MC
+};
+
+struct hclge_vport_vlan_cfg {
+	struct list_head node;
+	int hd_tbl_status;
+	u16 vlan_id;
+};
+
+struct hclge_rst_stats {
+	u32 reset_done_cnt;	/* the number of reset has completed */
+	u32 hw_reset_done_cnt;	/* the number of HW reset has completed */
+	u32 pf_rst_cnt;		/* the number of PF reset */
+	u32 flr_rst_cnt;	/* the number of FLR */
+	u32 global_rst_cnt;	/* the number of GLOBAL */
+	u32 imp_rst_cnt;	/* the number of IMP reset */
+	u32 reset_cnt;		/* the number of reset */
+	u32 reset_fail_cnt;	/* the number of reset fail */
+};
+
+/* time and register status when mac tunnel interruption occur */
+struct hclge_mac_tnl_stats {
+	u64 time;
+	u32 status;
+};
+
+#define HCLGE_RESET_INTERVAL	(10 * HZ)
+#define HCLGE_WAIT_RESET_DONE	100
+
+#pragma pack(1)
+struct hclge_vf_vlan_cfg {
+	u8 mbx_cmd;
+	u8 subcode;
+	union {
+		struct {
+			u8 is_kill;
+			__le16 vlan;
+			__le16 proto;
+		};
+		u8 enable;
+	};
+};
+
+#pragma pack()
+
+/* For each bit of TCAM entry, it uses a pair of 'x' and
+ * 'y' to indicate which value to match, like below:
+ * ----------------------------------
+ * | bit x | bit y |  search value  |
+ * ----------------------------------
+ * |   0   |   0   |   always hit   |
+ * ----------------------------------
+ * |   1   |   0   |   match '0'    |
+ * ----------------------------------
+ * |   0   |   1   |   match '1'    |
+ * ----------------------------------
+ * |   1   |   1   |   invalid      |
+ * ----------------------------------
+ * Then for input key(k) and mask(v), we can calculate the value by
+ * the formulae:
+ *	x = (~k) & v
+ *	y = k & v
+ */
+#define calc_x(x, k, v) ((x) = ~(k) & (v))
+#define calc_y(y, k, v) ((y) = (k) & (v))
+
+#define HCLGE_MAC_STATS_FIELD_OFF(f) (offsetof(struct hclge_mac_stats, f))
+#define HCLGE_STATS_READ(p, offset) (*(u64 *)((u8 *)(p) + (offset)))
+
+#define HCLGE_MAC_TNL_LOG_SIZE	8
+#define HCLGE_VPORT_NUM 256
+struct hclge_dev {
+	struct pci_dev *pdev;
+	struct hnae3_ae_dev *ae_dev;
+	struct hclge_hw hw;
+	struct hclge_misc_vector misc_vector;
+	struct hclge_mac_stats mac_stats;
+	struct hclge_fec_stats fec_stats;
+	unsigned long state;
+	unsigned long flr_state;
+	unsigned long last_reset_time;
+
+	enum hnae3_reset_type reset_type;
+	enum hnae3_reset_type reset_level;
+	unsigned long default_reset_request;
+	unsigned long reset_request;	/* reset has been requested */
+	unsigned long reset_pending;	/* client rst is pending to be served */
+	struct hclge_rst_stats rst_stats;
+	struct semaphore reset_sem;	/* protect reset process */
+	u32 fw_version;
+	u16 num_tqps;			/* Num task queue pairs of this PF */
+	u16 num_req_vfs;		/* Num VFs requested for this PF */
+
+	u16 base_tqp_pid;	/* Base task tqp physical id of this PF */
+	u16 alloc_rss_size;		/* Allocated RSS task queue */
+	u16 vf_rss_size_max;		/* HW defined VF max RSS task queue */
+	u16 pf_rss_size_max;		/* HW defined PF max RSS task queue */
+	u32 tx_spare_buf_size;		/* HW defined TX spare buffer size */
+
+	u16 fdir_pf_filter_count; /* Num of guaranteed filters for this PF */
+	u16 num_alloc_vport;		/* Num vports this driver supports */
+	nodemask_t numa_node_mask;
+	u16 rx_buf_len;
+	u16 num_tx_desc;		/* desc num of per tx queue */
+	u16 num_rx_desc;		/* desc num of per rx queue */
+	u8 hw_tc_map;
+	enum hclge_fc_mode fc_mode_last_time;
+	u8 support_sfp_query;
+
+#define HCLGE_FLAG_TC_BASE_SCH_MODE		1
+#define HCLGE_FLAG_VNET_BASE_SCH_MODE		2
+	u8 tx_sch_mode;
+	u8 tc_max;
+	u8 pfc_max;
+
+	u8 default_up;
+	u8 dcbx_cap;
+	struct hclge_tm_info tm_info;
+
+	u16 num_msi;
+	u16 num_msi_left;
+	u16 num_msi_used;
+	u16 *vector_status;
+	int *vector_irq;
+	u16 num_nic_msi;	/* Num of nic vectors for this PF */
+	u16 num_roce_msi;	/* Num of roce vectors for this PF */
+
+	unsigned long service_timer_period;
+	unsigned long service_timer_previous;
+	struct timer_list reset_timer;
+	struct delayed_work service_task;
+
+	bool cur_promisc;
+	int num_alloc_vfs;	/* Actual number of VFs allocated */
+
+	struct hclge_comm_tqp *htqp;
+	struct hclge_vport *vport;
+
+	struct dentry *hclge_dbgfs;
+
+	struct hnae3_client *nic_client;
+	struct hnae3_client *roce_client;
+
+#define HCLGE_FLAG_MAIN			BIT(0)
+#define HCLGE_FLAG_DCB_CAPABLE		BIT(1)
+	u32 flag;
+
+	u32 pkt_buf_size; /* Total pf buf size for tx/rx */
+	u32 tx_buf_size; /* Tx buffer size for each TC */
+	u32 dv_buf_size; /* Dv buffer size for each TC */
+
+	u32 mps; /* Max packet size */
+	/* vport_lock protect resource shared by vports */
+	struct mutex vport_lock;
+
+	struct hclge_vlan_type_cfg vlan_type_cfg;
+
+	unsigned long vlan_table[VLAN_N_VID][BITS_TO_LONGS(HCLGE_VPORT_NUM)];
+	unsigned long vf_vlan_full[BITS_TO_LONGS(HCLGE_VPORT_NUM)];
+
+	unsigned long vport_config_block[BITS_TO_LONGS(HCLGE_VPORT_NUM)];
+
+	struct hclge_fd_cfg fd_cfg;
+	struct hlist_head fd_rule_list;
+	spinlock_t fd_rule_lock; /* protect fd_rule_list and fd_bmap */
+	u16 hclge_fd_rule_num;
+	unsigned long serv_processed_cnt;
+	unsigned long last_serv_processed;
+	unsigned long last_rst_scheduled;
+	unsigned long last_mbx_scheduled;
+	unsigned long fd_bmap[BITS_TO_LONGS(MAX_FD_FILTER_NUM)];
+	enum HCLGE_FD_ACTIVE_RULE_TYPE fd_active_type;
+	u8 fd_en;
+	bool gro_en;
+
+	u16 wanted_umv_size;
+	/* max available unicast mac vlan space */
+	u16 max_umv_size;
+	/* private unicast mac vlan space, it's same for PF and its VFs */
+	u16 priv_umv_size;
+	/* unicast mac vlan space shared by PF and its VFs */
+	u16 share_umv_size;
+	/* multicast mac address number used by PF and its VFs */
+	u16 used_mc_mac_num;
+
+	DECLARE_KFIFO(mac_tnl_log, struct hclge_mac_tnl_stats,
+		      HCLGE_MAC_TNL_LOG_SIZE);
+
+	struct hclge_ptp *ptp;
+	struct devlink *devlink;
+	struct hclge_comm_rss_cfg rss_cfg;
+};
+
+/* VPort level vlan tag configuration for TX direction */
+struct hclge_tx_vtag_cfg {
+	bool accept_tag1;	/* Whether accept tag1 packet from host */
+	bool accept_untag1;	/* Whether accept untag1 packet from host */
+	bool accept_tag2;
+	bool accept_untag2;
+	bool insert_tag1_en;	/* Whether insert inner vlan tag */
+	bool insert_tag2_en;	/* Whether insert outer vlan tag */
+	u16  default_tag1;	/* The default inner vlan tag to insert */
+	u16  default_tag2;	/* The default outer vlan tag to insert */
+	bool tag_shift_mode_en;
+};
+
+/* VPort level vlan tag configuration for RX direction */
+struct hclge_rx_vtag_cfg {
+	bool rx_vlan_offload_en; /* Whether enable rx vlan offload */
+	bool strip_tag1_en;	 /* Whether strip inner vlan tag */
+	bool strip_tag2_en;	 /* Whether strip outer vlan tag */
+	bool vlan1_vlan_prionly; /* Inner vlan tag up to descriptor enable */
+	bool vlan2_vlan_prionly; /* Outer vlan tag up to descriptor enable */
+	bool strip_tag1_discard_en; /* Inner vlan tag discard for BD enable */
+	bool strip_tag2_discard_en; /* Outer vlan tag discard for BD enable */
+};
+
+enum HCLGE_VPORT_STATE {
+	HCLGE_VPORT_STATE_ALIVE,
+	HCLGE_VPORT_STATE_MAC_TBL_CHANGE,
+	HCLGE_VPORT_STATE_PROMISC_CHANGE,
+	HCLGE_VPORT_STATE_VLAN_FLTR_CHANGE,
+	HCLGE_VPORT_STATE_INITED,
+	HCLGE_VPORT_STATE_MAX
+};
+
+enum HCLGE_VPORT_NEED_NOTIFY {
+	HCLGE_VPORT_NEED_NOTIFY_RESET,
+	HCLGE_VPORT_NEED_NOTIFY_VF_VLAN,
+};
+
+struct hclge_vlan_info {
+	u16 vlan_proto; /* so far support 802.1Q only */
+	u16 qos;
+	u16 vlan_tag;
+};
+
+struct hclge_port_base_vlan_config {
+	u16 state;
+	bool tbl_sta;
+	struct hclge_vlan_info vlan_info;
+	struct hclge_vlan_info old_vlan_info;
+};
+
+struct hclge_vf_info {
+	int link_state;
+	u8 mac[ETH_ALEN];
+	u32 spoofchk;
+	u32 max_tx_rate;
+	u32 trusted;
+	u8 request_uc_en;
+	u8 request_mc_en;
+	u8 request_bc_en;
+};
+
+struct hclge_vport {
+	u16 alloc_tqps;	/* Allocated Tx/Rx queues */
+
+	u16 qs_offset;
+	u32 bw_limit;		/* VSI BW Limit (0 = disabled) */
+	u8  dwrr;
+
+	bool req_vlan_fltr_en;
+	bool cur_vlan_fltr_en;
+	unsigned long vlan_del_fail_bmap[BITS_TO_LONGS(VLAN_N_VID)];
+	struct hclge_port_base_vlan_config port_base_vlan_cfg;
+	struct hclge_tx_vtag_cfg  txvlan_cfg;
+	struct hclge_rx_vtag_cfg  rxvlan_cfg;
+
+	u16 used_umv_num;
+
+	u16 vport_id;
+	struct hclge_dev *back;  /* Back reference to associated dev */
+	struct hnae3_handle nic;
+	struct hnae3_handle roce;
+
+	unsigned long state;
+	unsigned long need_notify;
+	unsigned long last_active_jiffies;
+	u32 mps; /* Max packet size */
+	struct hclge_vf_info vf_info;
+
+	u8 overflow_promisc_flags;
+	u8 last_promisc_flags;
+
+	spinlock_t mac_list_lock; /* protect mac address need to add/detele */
+	struct list_head uc_mac_list;   /* Store VF unicast table */
+	struct list_head mc_mac_list;   /* Store VF multicast table */
+
+	struct list_head vlan_list;     /* Store VF vlan table */
+};
+
+struct hclge_speed_bit_map {
+	u32 speed;
+	u32 speed_bit;
+};
+
+struct hclge_mac_speed_map {
+	u32 speed_drv; /* speed defined in driver */
+	u32 speed_fw; /* speed defined in firmware */
+};
+
+struct hclge_link_mode_bmap {
+	u16 support_bit;
+	enum ethtool_link_mode_bit_indices link_mode;
+};
+
+int hclge_set_vport_promisc_mode(struct hclge_vport *vport, bool en_uc_pmc,
+				 bool en_mc_pmc, bool en_bc_pmc);
+int hclge_add_uc_addr_common(struct hclge_vport *vport,
+			     const unsigned char *addr);
+int hclge_rm_uc_addr_common(struct hclge_vport *vport,
+			    const unsigned char *addr);
+int hclge_add_mc_addr_common(struct hclge_vport *vport,
+			     const unsigned char *addr);
+int hclge_rm_mc_addr_common(struct hclge_vport *vport,
+			    const unsigned char *addr);
+
+struct hclge_vport *hclge_get_vport(struct hnae3_handle *handle);
+int hclge_bind_ring_with_vector(struct hclge_vport *vport,
+				int vector_id, bool en,
+				struct hnae3_ring_chain_node *ring_chain);
+
+static inline int hclge_get_queue_id(struct hnae3_queue *queue)
+{
+	struct hclge_comm_tqp *tqp =
+			container_of(queue, struct hclge_comm_tqp, q);
+
+	return tqp->index;
+}
+
+int hclge_inform_reset_assert_to_vf(struct hclge_vport *vport);
+int hclge_cfg_mac_speed_dup(struct hclge_dev *hdev, int speed, u8 duplex, u8 lane_num);
+int hclge_set_vlan_filter(struct hnae3_handle *handle, __be16 proto,
+			  u16 vlan_id, bool is_kill);
+int hclge_en_hw_strip_rxvtag(struct hnae3_handle *handle, bool enable);
+
+int hclge_buffer_alloc(struct hclge_dev *hdev);
+int hclge_rss_init_hw(struct hclge_dev *hdev);
+
+void hclge_mbx_handler(struct hclge_dev *hdev);
+int hclge_reset_tqp(struct hnae3_handle *handle);
+int hclge_cfg_flowctrl(struct hclge_dev *hdev);
+int hclge_func_reset_cmd(struct hclge_dev *hdev, int func_id);
+int hclge_vport_start(struct hclge_vport *vport);
+void hclge_vport_stop(struct hclge_vport *vport);
+int hclge_set_vport_mtu(struct hclge_vport *vport, int new_mtu);
+int hclge_dbg_get_read_func(struct hnae3_handle *handle, enum hnae3_dbg_cmd cmd,
+			    read_func *func);
+u16 hclge_covert_handle_qid_global(struct hnae3_handle *handle, u16 queue_id);
+int hclge_notify_client(struct hclge_dev *hdev,
+			enum hnae3_reset_notify_type type);
+int hclge_update_mac_list(struct hclge_vport *vport,
+			  enum HCLGE_MAC_NODE_STATE state,
+			  enum HCLGE_MAC_ADDR_TYPE mac_type,
+			  const unsigned char *addr);
+int hclge_update_mac_node_for_dev_addr(struct hclge_vport *vport,
+				       const u8 *old_addr, const u8 *new_addr);
+void hclge_rm_vport_all_mac_table(struct hclge_vport *vport, bool is_del_list,
+				  enum HCLGE_MAC_ADDR_TYPE mac_type);
+void hclge_rm_vport_all_vlan_table(struct hclge_vport *vport, bool is_del_list);
+void hclge_uninit_vport_vlan_table(struct hclge_dev *hdev);
+void hclge_restore_mac_table_common(struct hclge_vport *vport);
+void hclge_restore_vport_port_base_vlan_config(struct hclge_dev *hdev);
+void hclge_restore_vport_vlan_table(struct hclge_vport *vport);
+int hclge_update_port_base_vlan_cfg(struct hclge_vport *vport, u16 state,
+				    struct hclge_vlan_info *vlan_info);
+int hclge_push_vf_port_base_vlan_info(struct hclge_vport *vport, u8 vfid,
+				      u16 state,
+				      struct hclge_vlan_info *vlan_info);
+void hclge_task_schedule(struct hclge_dev *hdev, unsigned long delay_time);
+void hclge_report_hw_error(struct hclge_dev *hdev,
+			   enum hnae3_hw_error_type type);
+int hclge_dbg_dump_rst_info(struct hclge_dev *hdev, char *buf, int len);
+int hclge_push_vf_link_status(struct hclge_vport *vport);
+int hclge_enable_vport_vlan_filter(struct hclge_vport *vport, bool request_en);
+int hclge_mac_update_stats(struct hclge_dev *hdev);
+struct hclge_vport *hclge_get_vf_vport(struct hclge_dev *hdev, int vf);
+int hclge_inform_vf_reset(struct hclge_vport *vport, u16 reset_type);
+int hclge_query_scc_version(struct hclge_dev *hdev, u32 *scc_version);
+#endif
diff --git a/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_mbx.c b/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_mbx.c
new file mode 100644
index 000000000000..c7ff12a6c076
--- /dev/null
+++ b/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_mbx.c
@@ -0,0 +1,1155 @@
+// SPDX-License-Identifier: GPL-2.0+
+// Copyright (c) 2016-2017 Hisilicon Limited.
+
+#include "hclge_main.h"
+#include "hclge_mbx.h"
+#include "hnae3.h"
+#include "hclge_comm_rss.h"
+
+#define CREATE_TRACE_POINTS
+#include "hclge_trace.h"
+
+static u16 hclge_errno_to_resp(int errno)
+{
+	int resp = abs(errno);
+
+	/* The status for pf to vf msg cmd is u16, constrainted by HW.
+	 * We need to keep the same type with it.
+	 * The intput errno is the stander error code, it's safely to
+	 * use a u16 to store the abs(errno).
+	 */
+	return (u16)resp;
+}
+
+/* hclge_gen_resp_to_vf: used to generate a synchronous response to VF when PF
+ * receives a mailbox message from VF.
+ * @vport: pointer to struct hclge_vport
+ * @vf_to_pf_req: pointer to hclge_mbx_vf_to_pf_cmd of the original mailbox
+ *		  message
+ * @resp_status: indicate to VF whether its request success(0) or failed.
+ */
+static int hclge_gen_resp_to_vf(struct hclge_vport *vport,
+				struct hclge_mbx_vf_to_pf_cmd *vf_to_pf_req,
+				struct hclge_respond_to_vf_msg *resp_msg)
+{
+	struct hclge_mbx_pf_to_vf_cmd *resp_pf_to_vf;
+	struct hclge_dev *hdev = vport->back;
+	enum hclge_comm_cmd_status status;
+	struct hclge_desc desc;
+	u16 resp;
+
+	resp_pf_to_vf = (struct hclge_mbx_pf_to_vf_cmd *)desc.data;
+
+	if (resp_msg->len > HCLGE_MBX_MAX_RESP_DATA_SIZE) {
+		dev_err(&hdev->pdev->dev,
+			"PF fail to gen resp to VF len %u exceeds max len %u\n",
+			resp_msg->len,
+			HCLGE_MBX_MAX_RESP_DATA_SIZE);
+		/* If resp_msg->len is too long, set the value to max length
+		 * and return the msg to VF
+		 */
+		resp_msg->len = HCLGE_MBX_MAX_RESP_DATA_SIZE;
+	}
+
+	hclge_cmd_setup_basic_desc(&desc, HCLGEVF_OPC_MBX_PF_TO_VF, false);
+
+	resp_pf_to_vf->dest_vfid = vf_to_pf_req->mbx_src_vfid;
+	resp_pf_to_vf->msg_len = vf_to_pf_req->msg_len;
+	resp_pf_to_vf->match_id = vf_to_pf_req->match_id;
+
+	resp_pf_to_vf->msg.code = cpu_to_le16(HCLGE_MBX_PF_VF_RESP);
+	resp_pf_to_vf->msg.vf_mbx_msg_code =
+				cpu_to_le16(vf_to_pf_req->msg.code);
+	resp_pf_to_vf->msg.vf_mbx_msg_subcode =
+				cpu_to_le16(vf_to_pf_req->msg.subcode);
+	resp = hclge_errno_to_resp(resp_msg->status);
+	if (resp < SHRT_MAX) {
+		resp_pf_to_vf->msg.resp_status = cpu_to_le16(resp);
+	} else {
+		dev_warn(&hdev->pdev->dev,
+			 "failed to send response to VF, response status %u is out-of-bound\n",
+			 resp);
+		resp_pf_to_vf->msg.resp_status = cpu_to_le16(EIO);
+	}
+
+	if (resp_msg->len > 0)
+		memcpy(resp_pf_to_vf->msg.resp_data, resp_msg->data,
+		       resp_msg->len);
+
+	trace_hclge_pf_mbx_send(hdev, resp_pf_to_vf);
+
+	status = hclge_cmd_send(&hdev->hw, &desc, 1);
+	if (status)
+		dev_err(&hdev->pdev->dev,
+			"failed to send response to VF, status: %d, vfid: %u, code: %u, subcode: %u.\n",
+			status, vf_to_pf_req->mbx_src_vfid,
+			vf_to_pf_req->msg.code, vf_to_pf_req->msg.subcode);
+
+	return status;
+}
+
+static int hclge_send_mbx_msg(struct hclge_vport *vport, u8 *msg, u16 msg_len,
+			      u16 mbx_opcode, u8 dest_vfid)
+{
+	struct hclge_mbx_pf_to_vf_cmd *resp_pf_to_vf;
+	struct hclge_dev *hdev = vport->back;
+	enum hclge_comm_cmd_status status;
+	struct hclge_desc desc;
+
+	if (msg_len > HCLGE_MBX_MAX_MSG_SIZE) {
+		dev_err(&hdev->pdev->dev,
+			"msg data length(=%u) exceeds maximum(=%u)\n",
+			msg_len, HCLGE_MBX_MAX_MSG_SIZE);
+		return -EMSGSIZE;
+	}
+
+	resp_pf_to_vf = (struct hclge_mbx_pf_to_vf_cmd *)desc.data;
+
+	hclge_cmd_setup_basic_desc(&desc, HCLGEVF_OPC_MBX_PF_TO_VF, false);
+
+	resp_pf_to_vf->dest_vfid = dest_vfid;
+	resp_pf_to_vf->msg_len = msg_len;
+	resp_pf_to_vf->msg.code = cpu_to_le16(mbx_opcode);
+
+	memcpy(resp_pf_to_vf->msg.msg_data, msg, msg_len);
+
+	trace_hclge_pf_mbx_send(hdev, resp_pf_to_vf);
+
+	status = hclge_cmd_send(&hdev->hw, &desc, 1);
+	if (status)
+		dev_err(&hdev->pdev->dev,
+			"failed to send mailbox to VF, status: %d, vfid: %u, opcode: %u\n",
+			status, dest_vfid, mbx_opcode);
+
+	return status;
+}
+
+int hclge_inform_vf_reset(struct hclge_vport *vport, u16 reset_type)
+{
+	__le16 msg_data;
+	u8 dest_vfid;
+
+	dest_vfid = (u8)vport->vport_id;
+	msg_data = cpu_to_le16(reset_type);
+
+	/* send this requested info to VF */
+	return hclge_send_mbx_msg(vport, (u8 *)&msg_data, sizeof(msg_data),
+				  HCLGE_MBX_ASSERTING_RESET, dest_vfid);
+}
+
+int hclge_inform_reset_assert_to_vf(struct hclge_vport *vport)
+{
+	struct hclge_dev *hdev = vport->back;
+	u16 reset_type;
+
+	BUILD_BUG_ON(HNAE3_MAX_RESET > U16_MAX);
+
+	if (hdev->reset_type == HNAE3_FUNC_RESET)
+		reset_type = HNAE3_VF_PF_FUNC_RESET;
+	else if (hdev->reset_type == HNAE3_FLR_RESET)
+		reset_type = HNAE3_VF_FULL_RESET;
+	else
+		reset_type = HNAE3_VF_FUNC_RESET;
+
+	return hclge_inform_vf_reset(vport, reset_type);
+}
+
+static void hclge_free_vector_ring_chain(struct hnae3_ring_chain_node *head)
+{
+	struct hnae3_ring_chain_node *chain_tmp, *chain;
+
+	chain = head->next;
+
+	while (chain) {
+		chain_tmp = chain->next;
+		kfree_sensitive(chain);
+		chain = chain_tmp;
+	}
+}
+
+/* hclge_get_ring_chain_from_mbx: get ring type & tqp id & int_gl idx
+ * from mailbox message
+ * msg[0]: opcode
+ * msg[1]: <not relevant to this function>
+ * msg[2]: ring_num
+ * msg[3]: first ring type (TX|RX)
+ * msg[4]: first tqp id
+ * msg[5]: first int_gl idx
+ * msg[6] ~ msg[14]: other ring type, tqp id and int_gl idx
+ */
+static int hclge_get_ring_chain_from_mbx(
+			struct hclge_mbx_vf_to_pf_cmd *req,
+			struct hnae3_ring_chain_node *ring_chain,
+			struct hclge_vport *vport)
+{
+	struct hnae3_ring_chain_node *cur_chain, *new_chain;
+	struct hclge_dev *hdev = vport->back;
+	int ring_num;
+	int i;
+
+	ring_num = req->msg.ring_num;
+
+	if (ring_num > HCLGE_MBX_MAX_RING_CHAIN_PARAM_NUM)
+		return -EINVAL;
+
+	for (i = 0; i < ring_num; i++) {
+		if (req->msg.param[i].tqp_index >= vport->nic.kinfo.rss_size) {
+			dev_err(&hdev->pdev->dev, "tqp index(%u) is out of range(0-%u)\n",
+				req->msg.param[i].tqp_index,
+				vport->nic.kinfo.rss_size - 1U);
+			return -EINVAL;
+		}
+	}
+
+	hnae3_set_bit(ring_chain->flag, HNAE3_RING_TYPE_B,
+		      req->msg.param[0].ring_type);
+	ring_chain->tqp_index =
+		hclge_get_queue_id(vport->nic.kinfo.tqp
+				   [req->msg.param[0].tqp_index]);
+	hnae3_set_field(ring_chain->int_gl_idx, HNAE3_RING_GL_IDX_M,
+			HNAE3_RING_GL_IDX_S, req->msg.param[0].int_gl_index);
+
+	cur_chain = ring_chain;
+
+	for (i = 1; i < ring_num; i++) {
+		new_chain = kzalloc(sizeof(*new_chain), GFP_KERNEL);
+		if (!new_chain)
+			goto err;
+
+		hnae3_set_bit(new_chain->flag, HNAE3_RING_TYPE_B,
+			      req->msg.param[i].ring_type);
+
+		new_chain->tqp_index =
+		hclge_get_queue_id(vport->nic.kinfo.tqp
+			[req->msg.param[i].tqp_index]);
+
+		hnae3_set_field(new_chain->int_gl_idx, HNAE3_RING_GL_IDX_M,
+				HNAE3_RING_GL_IDX_S,
+				req->msg.param[i].int_gl_index);
+
+		cur_chain->next = new_chain;
+		cur_chain = new_chain;
+	}
+
+	return 0;
+err:
+	hclge_free_vector_ring_chain(ring_chain);
+	return -ENOMEM;
+}
+
+static int hclge_map_unmap_ring_to_vf_vector(struct hclge_vport *vport, bool en,
+					     struct hclge_mbx_vf_to_pf_cmd *req)
+{
+	struct hnae3_ring_chain_node ring_chain;
+	int vector_id = req->msg.vector_id;
+	int ret;
+
+	memset(&ring_chain, 0, sizeof(ring_chain));
+	ret = hclge_get_ring_chain_from_mbx(req, &ring_chain, vport);
+	if (ret)
+		return ret;
+
+	ret = hclge_bind_ring_with_vector(vport, vector_id, en, &ring_chain);
+
+	hclge_free_vector_ring_chain(&ring_chain);
+
+	return ret;
+}
+
+static int hclge_query_ring_vector_map(struct hclge_vport *vport,
+				       struct hnae3_ring_chain_node *ring_chain,
+				       struct hclge_desc *desc)
+{
+	struct hclge_ctrl_vector_chain_cmd *req =
+		(struct hclge_ctrl_vector_chain_cmd *)desc->data;
+	struct hclge_dev *hdev = vport->back;
+	u16 tqp_type_and_id;
+	int status;
+
+	hclge_cmd_setup_basic_desc(desc, HCLGE_OPC_ADD_RING_TO_VECTOR, true);
+
+	tqp_type_and_id = le16_to_cpu(req->tqp_type_and_id[0]);
+	hnae3_set_field(tqp_type_and_id, HCLGE_INT_TYPE_M, HCLGE_INT_TYPE_S,
+			hnae3_get_bit(ring_chain->flag, HNAE3_RING_TYPE_B));
+	hnae3_set_field(tqp_type_and_id, HCLGE_TQP_ID_M, HCLGE_TQP_ID_S,
+			ring_chain->tqp_index);
+	req->tqp_type_and_id[0] = cpu_to_le16(tqp_type_and_id);
+	req->vfid = vport->vport_id;
+
+	status = hclge_cmd_send(&hdev->hw, desc, 1);
+	if (status)
+		dev_err(&hdev->pdev->dev,
+			"Get VF ring vector map info fail, status is %d.\n",
+			status);
+
+	return status;
+}
+
+static int hclge_get_vf_ring_vector_map(struct hclge_vport *vport,
+					struct hclge_mbx_vf_to_pf_cmd *req,
+					struct hclge_respond_to_vf_msg *resp)
+{
+#define HCLGE_LIMIT_RING_NUM			1
+#define HCLGE_RING_TYPE_OFFSET			0
+#define HCLGE_TQP_INDEX_OFFSET			1
+#define HCLGE_INT_GL_INDEX_OFFSET		2
+#define HCLGE_VECTOR_ID_OFFSET			3
+#define HCLGE_RING_VECTOR_MAP_INFO_LEN		4
+	struct hnae3_ring_chain_node ring_chain;
+	struct hclge_desc desc;
+	struct hclge_ctrl_vector_chain_cmd *data =
+		(struct hclge_ctrl_vector_chain_cmd *)desc.data;
+	u16 tqp_type_and_id;
+	u8 int_gl_index;
+	int ret;
+
+	req->msg.ring_num = HCLGE_LIMIT_RING_NUM;
+
+	memset(&ring_chain, 0, sizeof(ring_chain));
+	ret = hclge_get_ring_chain_from_mbx(req, &ring_chain, vport);
+	if (ret)
+		return ret;
+
+	ret = hclge_query_ring_vector_map(vport, &ring_chain, &desc);
+	if (ret) {
+		hclge_free_vector_ring_chain(&ring_chain);
+		return ret;
+	}
+
+	tqp_type_and_id = le16_to_cpu(data->tqp_type_and_id[0]);
+	int_gl_index = hnae3_get_field(tqp_type_and_id,
+				       HCLGE_INT_GL_IDX_M, HCLGE_INT_GL_IDX_S);
+
+	resp->data[HCLGE_RING_TYPE_OFFSET] = req->msg.param[0].ring_type;
+	resp->data[HCLGE_TQP_INDEX_OFFSET] = req->msg.param[0].tqp_index;
+	resp->data[HCLGE_INT_GL_INDEX_OFFSET] = int_gl_index;
+	resp->data[HCLGE_VECTOR_ID_OFFSET] = data->int_vector_id_l;
+	resp->len = HCLGE_RING_VECTOR_MAP_INFO_LEN;
+
+	hclge_free_vector_ring_chain(&ring_chain);
+
+	return ret;
+}
+
+static void hclge_set_vf_promisc_mode(struct hclge_vport *vport,
+				      struct hclge_mbx_vf_to_pf_cmd *req)
+{
+	struct hnae3_handle *handle = &vport->nic;
+	struct hclge_dev *hdev = vport->back;
+
+	vport->vf_info.request_uc_en = req->msg.en_uc;
+	vport->vf_info.request_mc_en = req->msg.en_mc;
+	vport->vf_info.request_bc_en = req->msg.en_bc;
+
+	if (req->msg.en_limit_promisc)
+		set_bit(HNAE3_PFLAG_LIMIT_PROMISC, &handle->priv_flags);
+	else
+		clear_bit(HNAE3_PFLAG_LIMIT_PROMISC,
+			  &handle->priv_flags);
+
+	set_bit(HCLGE_VPORT_STATE_PROMISC_CHANGE, &vport->state);
+	hclge_task_schedule(hdev, 0);
+}
+
+static int hclge_set_vf_uc_mac_addr(struct hclge_vport *vport,
+				    struct hclge_mbx_vf_to_pf_cmd *mbx_req)
+{
+#define HCLGE_MBX_VF_OLD_MAC_ADDR_OFFSET	6
+
+	const u8 *mac_addr = (const u8 *)(mbx_req->msg.data);
+	struct hclge_dev *hdev = vport->back;
+	int status;
+
+	if (mbx_req->msg.subcode == HCLGE_MBX_MAC_VLAN_UC_MODIFY) {
+		const u8 *old_addr = (const u8 *)
+		(&mbx_req->msg.data[HCLGE_MBX_VF_OLD_MAC_ADDR_OFFSET]);
+
+		/* If VF MAC has been configured by the host then it
+		 * cannot be overridden by the MAC specified by the VM.
+		 */
+		if (!is_zero_ether_addr(vport->vf_info.mac) &&
+		    !ether_addr_equal(mac_addr, vport->vf_info.mac))
+			return -EPERM;
+
+		if (!is_valid_ether_addr(mac_addr))
+			return -EINVAL;
+
+		spin_lock_bh(&vport->mac_list_lock);
+		status = hclge_update_mac_node_for_dev_addr(vport, old_addr,
+							    mac_addr);
+		spin_unlock_bh(&vport->mac_list_lock);
+		hclge_task_schedule(hdev, 0);
+	} else if (mbx_req->msg.subcode == HCLGE_MBX_MAC_VLAN_UC_ADD) {
+		status = hclge_update_mac_list(vport, HCLGE_MAC_TO_ADD,
+					       HCLGE_MAC_ADDR_UC, mac_addr);
+	} else if (mbx_req->msg.subcode == HCLGE_MBX_MAC_VLAN_UC_REMOVE) {
+		status = hclge_update_mac_list(vport, HCLGE_MAC_TO_DEL,
+					       HCLGE_MAC_ADDR_UC, mac_addr);
+	} else {
+		dev_err(&hdev->pdev->dev,
+			"failed to set unicast mac addr, unknown subcode %u\n",
+			mbx_req->msg.subcode);
+		return -EIO;
+	}
+
+	return status;
+}
+
+static int hclge_set_vf_mc_mac_addr(struct hclge_vport *vport,
+				    struct hclge_mbx_vf_to_pf_cmd *mbx_req)
+{
+	const u8 *mac_addr = (const u8 *)(mbx_req->msg.data);
+	struct hclge_dev *hdev = vport->back;
+
+	if (mbx_req->msg.subcode == HCLGE_MBX_MAC_VLAN_MC_ADD) {
+		hclge_update_mac_list(vport, HCLGE_MAC_TO_ADD,
+				      HCLGE_MAC_ADDR_MC, mac_addr);
+	} else if (mbx_req->msg.subcode == HCLGE_MBX_MAC_VLAN_MC_REMOVE) {
+		hclge_update_mac_list(vport, HCLGE_MAC_TO_DEL,
+				      HCLGE_MAC_ADDR_MC, mac_addr);
+	} else {
+		dev_err(&hdev->pdev->dev,
+			"failed to set mcast mac addr, unknown subcode %u\n",
+			mbx_req->msg.subcode);
+		return -EIO;
+	}
+
+	return 0;
+}
+
+int hclge_push_vf_port_base_vlan_info(struct hclge_vport *vport, u8 vfid,
+				      u16 state,
+				      struct hclge_vlan_info *vlan_info)
+{
+	struct hclge_mbx_port_base_vlan base_vlan;
+
+	base_vlan.state = cpu_to_le16(state);
+	base_vlan.vlan_proto = cpu_to_le16(vlan_info->vlan_proto);
+	base_vlan.qos = cpu_to_le16(vlan_info->qos);
+	base_vlan.vlan_tag = cpu_to_le16(vlan_info->vlan_tag);
+
+	return hclge_send_mbx_msg(vport, (u8 *)&base_vlan, sizeof(base_vlan),
+				  HCLGE_MBX_PUSH_VLAN_INFO, vfid);
+}
+
+static int hclge_set_vf_vlan_cfg(struct hclge_vport *vport,
+				 struct hclge_mbx_vf_to_pf_cmd *mbx_req,
+				 struct hclge_respond_to_vf_msg *resp_msg)
+{
+#define HCLGE_MBX_VLAN_STATE_OFFSET	0
+#define HCLGE_MBX_VLAN_INFO_OFFSET	2
+
+	struct hnae3_handle *handle = &vport->nic;
+	struct hclge_dev *hdev = vport->back;
+	struct hclge_vf_vlan_cfg *msg_cmd;
+	__be16 proto;
+	u16 vlan_id;
+
+	msg_cmd = (struct hclge_vf_vlan_cfg *)&mbx_req->msg;
+	switch (msg_cmd->subcode) {
+	case HCLGE_MBX_VLAN_FILTER:
+		proto = cpu_to_be16(le16_to_cpu(msg_cmd->proto));
+		vlan_id = le16_to_cpu(msg_cmd->vlan);
+		return hclge_set_vlan_filter(handle, proto, vlan_id,
+					     msg_cmd->is_kill);
+	case HCLGE_MBX_VLAN_RX_OFF_CFG:
+		return hclge_en_hw_strip_rxvtag(handle, msg_cmd->enable);
+	case HCLGE_MBX_GET_PORT_BASE_VLAN_STATE:
+		/* vf does not need to know about the port based VLAN state
+		 * on device HNAE3_DEVICE_VERSION_V3. So always return disable
+		 * on device HNAE3_DEVICE_VERSION_V3 if vf queries the port
+		 * based VLAN state.
+		 */
+		resp_msg->data[0] =
+			hdev->ae_dev->dev_version >= HNAE3_DEVICE_VERSION_V3 ?
+			HNAE3_PORT_BASE_VLAN_DISABLE :
+			vport->port_base_vlan_cfg.state;
+		resp_msg->len = sizeof(u8);
+		return 0;
+	case HCLGE_MBX_ENABLE_VLAN_FILTER:
+		return hclge_enable_vport_vlan_filter(vport, msg_cmd->enable);
+	default:
+		return 0;
+	}
+}
+
+static int hclge_set_vf_alive(struct hclge_vport *vport,
+			      struct hclge_mbx_vf_to_pf_cmd *mbx_req)
+{
+	bool alive = !!mbx_req->msg.data[0];
+	int ret = 0;
+
+	if (alive)
+		ret = hclge_vport_start(vport);
+	else
+		hclge_vport_stop(vport);
+
+	return ret;
+}
+
+static void hclge_get_basic_info(struct hclge_vport *vport,
+				 struct hclge_respond_to_vf_msg *resp_msg)
+{
+	struct hnae3_knic_private_info *kinfo = &vport->nic.kinfo;
+	struct hnae3_ae_dev *ae_dev = vport->back->ae_dev;
+	struct hclge_basic_info *basic_info;
+	unsigned int i;
+	u32 pf_caps;
+
+	basic_info = (struct hclge_basic_info *)resp_msg->data;
+	for (i = 0; i < kinfo->tc_info.num_tc; i++)
+		basic_info->hw_tc_map |= BIT(i);
+
+	pf_caps = le32_to_cpu(basic_info->pf_caps);
+	if (test_bit(HNAE3_DEV_SUPPORT_VLAN_FLTR_MDF_B, ae_dev->caps))
+		hnae3_set_bit(pf_caps, HNAE3_PF_SUPPORT_VLAN_FLTR_MDF_B, 1);
+
+	basic_info->pf_caps = cpu_to_le32(pf_caps);
+	resp_msg->len = HCLGE_MBX_MAX_RESP_DATA_SIZE;
+}
+
+static void hclge_get_vf_queue_info(struct hclge_vport *vport,
+				    struct hclge_respond_to_vf_msg *resp_msg)
+{
+#define HCLGE_TQPS_RSS_INFO_LEN		6
+
+	struct hclge_mbx_vf_queue_info *queue_info;
+	struct hclge_dev *hdev = vport->back;
+
+	/* get the queue related info */
+	queue_info = (struct hclge_mbx_vf_queue_info *)resp_msg->data;
+	queue_info->num_tqps = cpu_to_le16(vport->alloc_tqps);
+	queue_info->rss_size = cpu_to_le16(vport->nic.kinfo.rss_size);
+	queue_info->rx_buf_len = cpu_to_le16(hdev->rx_buf_len);
+	resp_msg->len = HCLGE_TQPS_RSS_INFO_LEN;
+}
+
+static void hclge_get_vf_mac_addr(struct hclge_vport *vport,
+				  struct hclge_respond_to_vf_msg *resp_msg)
+{
+	ether_addr_copy(resp_msg->data, vport->vf_info.mac);
+	resp_msg->len = ETH_ALEN;
+}
+
+static void hclge_get_vf_queue_depth(struct hclge_vport *vport,
+				     struct hclge_respond_to_vf_msg *resp_msg)
+{
+#define HCLGE_TQPS_DEPTH_INFO_LEN	4
+
+	struct hclge_mbx_vf_queue_depth *queue_depth;
+	struct hclge_dev *hdev = vport->back;
+
+	/* get the queue depth info */
+	queue_depth = (struct hclge_mbx_vf_queue_depth *)resp_msg->data;
+	queue_depth->num_tx_desc = cpu_to_le16(hdev->num_tx_desc);
+	queue_depth->num_rx_desc = cpu_to_le16(hdev->num_rx_desc);
+
+	resp_msg->len = HCLGE_TQPS_DEPTH_INFO_LEN;
+}
+
+static void hclge_get_vf_media_type(struct hclge_vport *vport,
+				    struct hclge_respond_to_vf_msg *resp_msg)
+{
+#define HCLGE_VF_MEDIA_TYPE_OFFSET	0
+#define HCLGE_VF_MODULE_TYPE_OFFSET	1
+#define HCLGE_VF_MEDIA_TYPE_LENGTH	2
+
+	struct hclge_dev *hdev = vport->back;
+
+	resp_msg->data[HCLGE_VF_MEDIA_TYPE_OFFSET] =
+		hdev->hw.mac.media_type;
+	resp_msg->data[HCLGE_VF_MODULE_TYPE_OFFSET] =
+		hdev->hw.mac.module_type;
+	resp_msg->len = HCLGE_VF_MEDIA_TYPE_LENGTH;
+}
+
+int hclge_push_vf_link_status(struct hclge_vport *vport)
+{
+#define HCLGE_VF_LINK_STATE_UP		1U
+#define HCLGE_VF_LINK_STATE_DOWN	0U
+
+	struct hclge_mbx_link_status link_info;
+	struct hclge_dev *hdev = vport->back;
+	u16 link_status;
+
+	/* mac.link can only be 0 or 1 */
+	switch (vport->vf_info.link_state) {
+	case IFLA_VF_LINK_STATE_ENABLE:
+		link_status = HCLGE_VF_LINK_STATE_UP;
+		break;
+	case IFLA_VF_LINK_STATE_DISABLE:
+		link_status = HCLGE_VF_LINK_STATE_DOWN;
+		break;
+	case IFLA_VF_LINK_STATE_AUTO:
+	default:
+		link_status = (u16)hdev->hw.mac.link;
+		break;
+	}
+
+	link_info.link_status = cpu_to_le16(link_status);
+	link_info.speed = cpu_to_le32(hdev->hw.mac.speed);
+	link_info.duplex = cpu_to_le16(hdev->hw.mac.duplex);
+	link_info.flag = HCLGE_MBX_PUSH_LINK_STATUS_EN;
+
+	/* send this requested info to VF */
+	return hclge_send_mbx_msg(vport, (u8 *)&link_info, sizeof(link_info),
+				  HCLGE_MBX_LINK_STAT_CHANGE, vport->vport_id);
+}
+
+static void hclge_get_link_mode(struct hclge_vport *vport,
+				struct hclge_mbx_vf_to_pf_cmd *mbx_req)
+{
+#define HCLGE_SUPPORTED   1
+	struct hclge_mbx_link_mode link_mode;
+	struct hclge_dev *hdev = vport->back;
+	unsigned long advertising;
+	unsigned long supported;
+	unsigned long send_data;
+	u8 dest_vfid;
+
+	advertising = hdev->hw.mac.advertising[0];
+	supported = hdev->hw.mac.supported[0];
+	dest_vfid = mbx_req->mbx_src_vfid;
+	send_data = mbx_req->msg.data[0] == HCLGE_SUPPORTED ? supported :
+							      advertising;
+	link_mode.idx = cpu_to_le16((u16)mbx_req->msg.data[0]);
+	link_mode.link_mode = cpu_to_le64(send_data);
+
+	hclge_send_mbx_msg(vport, (u8 *)&link_mode, sizeof(link_mode),
+			   HCLGE_MBX_LINK_STAT_MODE, dest_vfid);
+}
+
+static int hclge_mbx_reset_vf_queue(struct hclge_vport *vport,
+				    struct hclge_mbx_vf_to_pf_cmd *mbx_req,
+				    struct hclge_respond_to_vf_msg *resp_msg)
+{
+#define HCLGE_RESET_ALL_QUEUE_DONE	1U
+	struct hnae3_handle *handle = &vport->nic;
+	struct hclge_dev *hdev = vport->back;
+	u16 queue_id;
+	int ret;
+
+	queue_id = le16_to_cpu(*(__le16 *)mbx_req->msg.data);
+	resp_msg->data[0] = HCLGE_RESET_ALL_QUEUE_DONE;
+	resp_msg->len = sizeof(u8);
+
+	/* pf will reset vf's all queues at a time. So it is unnecessary
+	 * to reset queues if queue_id > 0, just return success.
+	 */
+	if (queue_id > 0)
+		return 0;
+
+	ret = hclge_reset_tqp(handle);
+	if (ret)
+		dev_err(&hdev->pdev->dev, "failed to reset vf %u queue, ret = %d\n",
+			vport->vport_id - HCLGE_VF_VPORT_START_NUM, ret);
+
+	return ret;
+}
+
+static int hclge_reset_vf(struct hclge_vport *vport)
+{
+	struct hclge_dev *hdev = vport->back;
+
+	dev_warn(&hdev->pdev->dev, "PF received VF reset request from VF %u!",
+		 vport->vport_id - HCLGE_VF_VPORT_START_NUM);
+
+	return hclge_func_reset_cmd(hdev, vport->vport_id);
+}
+
+static void hclge_notify_vf_config(struct hclge_vport *vport)
+{
+	struct hclge_dev *hdev = vport->back;
+	struct hnae3_ae_dev *ae_dev = pci_get_drvdata(hdev->pdev);
+	struct hclge_port_base_vlan_config *vlan_cfg;
+	int ret;
+
+	hclge_push_vf_link_status(vport);
+	if (test_bit(HCLGE_VPORT_NEED_NOTIFY_RESET, &vport->need_notify)) {
+		ret = hclge_inform_vf_reset(vport, HNAE3_VF_PF_FUNC_RESET);
+		if (ret) {
+			dev_err(&hdev->pdev->dev,
+				"failed to inform VF %u reset!",
+				vport->vport_id - HCLGE_VF_VPORT_START_NUM);
+			return;
+		}
+		vport->need_notify = 0;
+		return;
+	}
+
+	if (ae_dev->dev_version < HNAE3_DEVICE_VERSION_V3 &&
+	    test_bit(HCLGE_VPORT_NEED_NOTIFY_VF_VLAN, &vport->need_notify)) {
+		vlan_cfg = &vport->port_base_vlan_cfg;
+		ret = hclge_push_vf_port_base_vlan_info(&hdev->vport[0],
+							vport->vport_id,
+							vlan_cfg->state,
+							&vlan_cfg->vlan_info);
+		if (ret) {
+			dev_err(&hdev->pdev->dev,
+				"failed to inform VF %u port base vlan!",
+				vport->vport_id - HCLGE_VF_VPORT_START_NUM);
+			return;
+		}
+		clear_bit(HCLGE_VPORT_NEED_NOTIFY_VF_VLAN, &vport->need_notify);
+	}
+}
+
+static void hclge_vf_keep_alive(struct hclge_vport *vport)
+{
+	struct hclge_dev *hdev = vport->back;
+
+	vport->last_active_jiffies = jiffies;
+
+	if (test_bit(HCLGE_VPORT_STATE_INITED, &vport->state) &&
+	    !test_bit(HCLGE_VPORT_STATE_ALIVE, &vport->state)) {
+		set_bit(HCLGE_VPORT_STATE_ALIVE, &vport->state);
+		dev_info(&hdev->pdev->dev, "VF %u is alive!",
+			 vport->vport_id - HCLGE_VF_VPORT_START_NUM);
+		hclge_notify_vf_config(vport);
+	}
+}
+
+static int hclge_set_vf_mtu(struct hclge_vport *vport,
+			    struct hclge_mbx_vf_to_pf_cmd *mbx_req)
+{
+	struct hclge_mbx_mtu_info *mtu_info;
+	u32 mtu;
+
+	mtu_info = (struct hclge_mbx_mtu_info *)mbx_req->msg.data;
+	mtu = le32_to_cpu(mtu_info->mtu);
+
+	return hclge_set_vport_mtu(vport, mtu);
+}
+
+static int hclge_get_queue_id_in_pf(struct hclge_vport *vport,
+				    struct hclge_mbx_vf_to_pf_cmd *mbx_req,
+				    struct hclge_respond_to_vf_msg *resp_msg)
+{
+	struct hnae3_handle *handle = &vport->nic;
+	struct hclge_dev *hdev = vport->back;
+	u16 queue_id, qid_in_pf;
+
+	queue_id = le16_to_cpu(*(__le16 *)mbx_req->msg.data);
+	if (queue_id >= handle->kinfo.num_tqps) {
+		dev_err(&hdev->pdev->dev, "Invalid queue id(%u) from VF %u\n",
+			queue_id, mbx_req->mbx_src_vfid);
+		return -EINVAL;
+	}
+
+	qid_in_pf = hclge_covert_handle_qid_global(&vport->nic, queue_id);
+	*(__le16 *)resp_msg->data = cpu_to_le16(qid_in_pf);
+	resp_msg->len = sizeof(qid_in_pf);
+	return 0;
+}
+
+static int hclge_get_rss_key(struct hclge_vport *vport,
+			     struct hclge_mbx_vf_to_pf_cmd *mbx_req,
+			     struct hclge_respond_to_vf_msg *resp_msg)
+{
+#define HCLGE_RSS_MBX_RESP_LEN	8
+	struct hclge_dev *hdev = vport->back;
+	struct hclge_comm_rss_cfg *rss_cfg;
+	int rss_hash_key_size;
+	u8 index;
+
+	index = mbx_req->msg.data[0];
+	rss_cfg = &hdev->rss_cfg;
+	rss_hash_key_size = sizeof(rss_cfg->rss_hash_key);
+
+	/* Check the query index of rss_hash_key from VF, make sure no
+	 * more than the size of rss_hash_key.
+	 */
+	if (((index + 1) * HCLGE_RSS_MBX_RESP_LEN) > rss_hash_key_size) {
+		dev_warn(&hdev->pdev->dev,
+			 "failed to get the rss hash key, the index(%u) invalid !\n",
+			 index);
+		return -EINVAL;
+	}
+
+	memcpy(resp_msg->data,
+	       &rss_cfg->rss_hash_key[index * HCLGE_RSS_MBX_RESP_LEN],
+	       HCLGE_RSS_MBX_RESP_LEN);
+	resp_msg->len = HCLGE_RSS_MBX_RESP_LEN;
+	return 0;
+}
+
+static void hclge_link_fail_parse(struct hclge_dev *hdev, u8 link_fail_code)
+{
+	switch (link_fail_code) {
+	case HCLGE_LF_REF_CLOCK_LOST:
+		dev_warn(&hdev->pdev->dev, "Reference clock lost!\n");
+		break;
+	case HCLGE_LF_XSFP_TX_DISABLE:
+		dev_warn(&hdev->pdev->dev, "SFP tx is disabled!\n");
+		break;
+	case HCLGE_LF_XSFP_ABSENT:
+		dev_warn(&hdev->pdev->dev, "SFP is absent!\n");
+		break;
+	default:
+		break;
+	}
+}
+
+static void hclge_handle_link_change_event(struct hclge_dev *hdev,
+					   struct hclge_mbx_vf_to_pf_cmd *req)
+{
+	hclge_task_schedule(hdev, 0);
+
+	if (!req->msg.subcode)
+		hclge_link_fail_parse(hdev, req->msg.data[0]);
+}
+
+static bool hclge_cmd_crq_empty(struct hclge_hw *hw)
+{
+	int tail = hclge_read_dev(hw, HCLGE_COMM_NIC_CRQ_TAIL_REG);
+
+	return tail == hw->hw.cmq.crq.next_to_use;
+}
+
+static void hclge_handle_ncsi_error(struct hclge_dev *hdev)
+{
+	struct hnae3_ae_dev *ae_dev = hdev->ae_dev;
+
+	ae_dev->ops->set_default_reset_request(ae_dev, HNAE3_GLOBAL_RESET);
+	dev_warn(&hdev->pdev->dev, "requesting reset due to NCSI error\n");
+	ae_dev->ops->reset_event(hdev->pdev, NULL);
+}
+
+static void hclge_handle_vf_tbl(struct hclge_vport *vport,
+				struct hclge_mbx_vf_to_pf_cmd *mbx_req)
+{
+	struct hclge_dev *hdev = vport->back;
+	struct hclge_vf_vlan_cfg *msg_cmd;
+
+	msg_cmd = (struct hclge_vf_vlan_cfg *)&mbx_req->msg;
+	if (msg_cmd->subcode == HCLGE_MBX_VPORT_LIST_CLEAR) {
+		hclge_rm_vport_all_mac_table(vport, true, HCLGE_MAC_ADDR_UC);
+		hclge_rm_vport_all_mac_table(vport, true, HCLGE_MAC_ADDR_MC);
+		hclge_rm_vport_all_vlan_table(vport, true);
+	} else {
+		dev_warn(&hdev->pdev->dev, "Invalid cmd(%u)\n",
+			 msg_cmd->subcode);
+	}
+}
+
+static int
+hclge_mbx_map_ring_to_vector_handler(struct hclge_mbx_ops_param *param)
+{
+	return hclge_map_unmap_ring_to_vf_vector(param->vport, true,
+						 param->req);
+}
+
+static int
+hclge_mbx_unmap_ring_to_vector_handler(struct hclge_mbx_ops_param *param)
+{
+	return hclge_map_unmap_ring_to_vf_vector(param->vport, false,
+						 param->req);
+}
+
+static int
+hclge_mbx_get_ring_vector_map_handler(struct hclge_mbx_ops_param *param)
+{
+	int ret;
+
+	ret = hclge_get_vf_ring_vector_map(param->vport, param->req,
+					   param->resp_msg);
+	if (ret)
+		dev_err(&param->vport->back->pdev->dev,
+			"PF fail(%d) to get VF ring vector map\n",
+			ret);
+	return ret;
+}
+
+static int hclge_mbx_set_promisc_mode_handler(struct hclge_mbx_ops_param *param)
+{
+	hclge_set_vf_promisc_mode(param->vport, param->req);
+	return 0;
+}
+
+static int hclge_mbx_set_unicast_handler(struct hclge_mbx_ops_param *param)
+{
+	int ret;
+
+	ret = hclge_set_vf_uc_mac_addr(param->vport, param->req);
+	if (ret)
+		dev_err(&param->vport->back->pdev->dev,
+			"PF fail(%d) to set VF UC MAC Addr\n",
+			ret);
+	return ret;
+}
+
+static int hclge_mbx_set_multicast_handler(struct hclge_mbx_ops_param *param)
+{
+	int ret;
+
+	ret = hclge_set_vf_mc_mac_addr(param->vport, param->req);
+	if (ret)
+		dev_err(&param->vport->back->pdev->dev,
+			"PF fail(%d) to set VF MC MAC Addr\n",
+			ret);
+	return ret;
+}
+
+static int hclge_mbx_set_vlan_handler(struct hclge_mbx_ops_param *param)
+{
+	int ret;
+
+	ret = hclge_set_vf_vlan_cfg(param->vport, param->req, param->resp_msg);
+	if (ret)
+		dev_err(&param->vport->back->pdev->dev,
+			"PF failed(%d) to config VF's VLAN\n",
+			ret);
+	return ret;
+}
+
+static int hclge_mbx_set_alive_handler(struct hclge_mbx_ops_param *param)
+{
+	int ret;
+
+	ret = hclge_set_vf_alive(param->vport, param->req);
+	if (ret)
+		dev_err(&param->vport->back->pdev->dev,
+			"PF failed(%d) to set VF's ALIVE\n",
+			ret);
+	return ret;
+}
+
+static int hclge_mbx_get_qinfo_handler(struct hclge_mbx_ops_param *param)
+{
+	hclge_get_vf_queue_info(param->vport, param->resp_msg);
+	return 0;
+}
+
+static int hclge_mbx_get_qdepth_handler(struct hclge_mbx_ops_param *param)
+{
+	hclge_get_vf_queue_depth(param->vport, param->resp_msg);
+	return 0;
+}
+
+static int hclge_mbx_get_basic_info_handler(struct hclge_mbx_ops_param *param)
+{
+	hclge_get_basic_info(param->vport, param->resp_msg);
+	return 0;
+}
+
+static int hclge_mbx_get_link_status_handler(struct hclge_mbx_ops_param *param)
+{
+	int ret;
+
+	ret = hclge_push_vf_link_status(param->vport);
+	if (ret)
+		dev_err(&param->vport->back->pdev->dev,
+			"failed to inform link stat to VF, ret = %d\n",
+			ret);
+	return ret;
+}
+
+static int hclge_mbx_queue_reset_handler(struct hclge_mbx_ops_param *param)
+{
+	return hclge_mbx_reset_vf_queue(param->vport, param->req,
+					param->resp_msg);
+}
+
+static int hclge_mbx_reset_handler(struct hclge_mbx_ops_param *param)
+{
+	return hclge_reset_vf(param->vport);
+}
+
+static int hclge_mbx_keep_alive_handler(struct hclge_mbx_ops_param *param)
+{
+	hclge_vf_keep_alive(param->vport);
+	return 0;
+}
+
+static int hclge_mbx_set_mtu_handler(struct hclge_mbx_ops_param *param)
+{
+	int ret;
+
+	ret = hclge_set_vf_mtu(param->vport, param->req);
+	if (ret)
+		dev_err(&param->vport->back->pdev->dev,
+			"VF fail(%d) to set mtu\n", ret);
+	return ret;
+}
+
+static int hclge_mbx_get_qid_in_pf_handler(struct hclge_mbx_ops_param *param)
+{
+	return hclge_get_queue_id_in_pf(param->vport, param->req,
+					param->resp_msg);
+}
+
+static int hclge_mbx_get_rss_key_handler(struct hclge_mbx_ops_param *param)
+{
+	return hclge_get_rss_key(param->vport, param->req, param->resp_msg);
+}
+
+static int hclge_mbx_get_link_mode_handler(struct hclge_mbx_ops_param *param)
+{
+	hclge_get_link_mode(param->vport, param->req);
+	return 0;
+}
+
+static int
+hclge_mbx_get_vf_flr_status_handler(struct hclge_mbx_ops_param *param)
+{
+	hclge_rm_vport_all_mac_table(param->vport, false,
+				     HCLGE_MAC_ADDR_UC);
+	hclge_rm_vport_all_mac_table(param->vport, false,
+				     HCLGE_MAC_ADDR_MC);
+	hclge_rm_vport_all_vlan_table(param->vport, false);
+	return 0;
+}
+
+static int hclge_mbx_vf_uninit_handler(struct hclge_mbx_ops_param *param)
+{
+	hclge_rm_vport_all_mac_table(param->vport, true,
+				     HCLGE_MAC_ADDR_UC);
+	hclge_rm_vport_all_mac_table(param->vport, true,
+				     HCLGE_MAC_ADDR_MC);
+	hclge_rm_vport_all_vlan_table(param->vport, true);
+	param->vport->mps = 0;
+	return 0;
+}
+
+static int hclge_mbx_get_media_type_handler(struct hclge_mbx_ops_param *param)
+{
+	hclge_get_vf_media_type(param->vport, param->resp_msg);
+	return 0;
+}
+
+static int hclge_mbx_push_link_status_handler(struct hclge_mbx_ops_param *param)
+{
+	hclge_handle_link_change_event(param->vport->back, param->req);
+	return 0;
+}
+
+static int hclge_mbx_get_mac_addr_handler(struct hclge_mbx_ops_param *param)
+{
+	hclge_get_vf_mac_addr(param->vport, param->resp_msg);
+	return 0;
+}
+
+static int hclge_mbx_ncsi_error_handler(struct hclge_mbx_ops_param *param)
+{
+	hclge_handle_ncsi_error(param->vport->back);
+	return 0;
+}
+
+static int hclge_mbx_handle_vf_tbl_handler(struct hclge_mbx_ops_param *param)
+{
+	hclge_handle_vf_tbl(param->vport, param->req);
+	return 0;
+}
+
+static const hclge_mbx_ops_fn hclge_mbx_ops_list[HCLGE_MBX_OPCODE_MAX] = {
+	[HCLGE_MBX_RESET]   = hclge_mbx_reset_handler,
+	[HCLGE_MBX_SET_UNICAST] = hclge_mbx_set_unicast_handler,
+	[HCLGE_MBX_SET_MULTICAST] = hclge_mbx_set_multicast_handler,
+	[HCLGE_MBX_SET_VLAN] = hclge_mbx_set_vlan_handler,
+	[HCLGE_MBX_MAP_RING_TO_VECTOR] = hclge_mbx_map_ring_to_vector_handler,
+	[HCLGE_MBX_UNMAP_RING_TO_VECTOR] = hclge_mbx_unmap_ring_to_vector_handler,
+	[HCLGE_MBX_SET_PROMISC_MODE] = hclge_mbx_set_promisc_mode_handler,
+	[HCLGE_MBX_GET_QINFO] = hclge_mbx_get_qinfo_handler,
+	[HCLGE_MBX_GET_QDEPTH] = hclge_mbx_get_qdepth_handler,
+	[HCLGE_MBX_GET_BASIC_INFO] = hclge_mbx_get_basic_info_handler,
+	[HCLGE_MBX_GET_RSS_KEY] = hclge_mbx_get_rss_key_handler,
+	[HCLGE_MBX_GET_MAC_ADDR] = hclge_mbx_get_mac_addr_handler,
+	[HCLGE_MBX_GET_LINK_STATUS] = hclge_mbx_get_link_status_handler,
+	[HCLGE_MBX_QUEUE_RESET] = hclge_mbx_queue_reset_handler,
+	[HCLGE_MBX_KEEP_ALIVE] = hclge_mbx_keep_alive_handler,
+	[HCLGE_MBX_SET_ALIVE] = hclge_mbx_set_alive_handler,
+	[HCLGE_MBX_SET_MTU] = hclge_mbx_set_mtu_handler,
+	[HCLGE_MBX_GET_QID_IN_PF] = hclge_mbx_get_qid_in_pf_handler,
+	[HCLGE_MBX_GET_LINK_MODE] = hclge_mbx_get_link_mode_handler,
+	[HCLGE_MBX_GET_MEDIA_TYPE] = hclge_mbx_get_media_type_handler,
+	[HCLGE_MBX_VF_UNINIT] = hclge_mbx_vf_uninit_handler,
+	[HCLGE_MBX_HANDLE_VF_TBL] = hclge_mbx_handle_vf_tbl_handler,
+	[HCLGE_MBX_GET_RING_VECTOR_MAP] = hclge_mbx_get_ring_vector_map_handler,
+	[HCLGE_MBX_GET_VF_FLR_STATUS] = hclge_mbx_get_vf_flr_status_handler,
+	[HCLGE_MBX_PUSH_LINK_STATUS] = hclge_mbx_push_link_status_handler,
+	[HCLGE_MBX_NCSI_ERROR] = hclge_mbx_ncsi_error_handler,
+};
+
+static void hclge_mbx_request_handling(struct hclge_mbx_ops_param *param)
+{
+	hclge_mbx_ops_fn cmd_func = NULL;
+	struct hclge_dev *hdev;
+	int ret = 0;
+
+	hdev = param->vport->back;
+	cmd_func = hclge_mbx_ops_list[param->req->msg.code];
+	if (!cmd_func) {
+		dev_err(&hdev->pdev->dev,
+			"un-supported mailbox message, code = %u\n",
+			param->req->msg.code);
+		return;
+	}
+	ret = cmd_func(param);
+
+	/* PF driver should not reply IMP */
+	if (hnae3_get_bit(param->req->mbx_need_resp, HCLGE_MBX_NEED_RESP_B) &&
+	    param->req->msg.code < HCLGE_MBX_GET_VF_FLR_STATUS) {
+		param->resp_msg->status = ret;
+		if (time_is_before_jiffies(hdev->last_mbx_scheduled +
+					   HCLGE_MBX_SCHED_TIMEOUT))
+			dev_warn(&hdev->pdev->dev,
+				 "resp vport%u mbx(%u,%u) late\n",
+				 param->req->mbx_src_vfid,
+				 param->req->msg.code,
+				 param->req->msg.subcode);
+
+		hclge_gen_resp_to_vf(param->vport, param->req, param->resp_msg);
+	}
+}
+
+void hclge_mbx_handler(struct hclge_dev *hdev)
+{
+	struct hclge_comm_cmq_ring *crq = &hdev->hw.hw.cmq.crq;
+	struct hclge_respond_to_vf_msg resp_msg;
+	struct hclge_mbx_vf_to_pf_cmd *req;
+	struct hclge_mbx_ops_param param;
+	struct hclge_desc *desc;
+	unsigned int flag;
+
+	param.resp_msg = &resp_msg;
+	/* handle all the mailbox requests in the queue */
+	while (!hclge_cmd_crq_empty(&hdev->hw)) {
+		if (test_bit(HCLGE_COMM_STATE_CMD_DISABLE,
+			     &hdev->hw.hw.comm_state)) {
+			dev_warn(&hdev->pdev->dev,
+				 "command queue needs re-initializing\n");
+			return;
+		}
+
+		desc = &crq->desc[crq->next_to_use];
+		req = (struct hclge_mbx_vf_to_pf_cmd *)desc->data;
+
+		flag = le16_to_cpu(crq->desc[crq->next_to_use].flag);
+		if (unlikely(!hnae3_get_bit(flag, HCLGE_CMDQ_RX_OUTVLD_B) ||
+			     req->mbx_src_vfid > hdev->num_req_vfs)) {
+			dev_warn(&hdev->pdev->dev,
+				 "dropped invalid mailbox message, code = %u, vfid = %u\n",
+				 req->msg.code, req->mbx_src_vfid);
+
+			/* dropping/not processing this invalid message */
+			crq->desc[crq->next_to_use].flag = 0;
+			hclge_mbx_ring_ptr_move_crq(crq);
+			continue;
+		}
+
+		trace_hclge_pf_mbx_get(hdev, req);
+
+		/* clear the resp_msg before processing every mailbox message */
+		memset(&resp_msg, 0, sizeof(resp_msg));
+		param.vport = &hdev->vport[req->mbx_src_vfid];
+		param.req = req;
+		hclge_mbx_request_handling(&param);
+
+		crq->desc[crq->next_to_use].flag = 0;
+		hclge_mbx_ring_ptr_move_crq(crq);
+	}
+
+	/* Write back CMDQ_RQ header pointer, M7 need this pointer */
+	hclge_write_dev(&hdev->hw, HCLGE_COMM_NIC_CRQ_HEAD_REG,
+			crq->next_to_use);
+}
diff --git a/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_mdio.c b/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_mdio.c
new file mode 100644
index 000000000000..96553109f44c
--- /dev/null
+++ b/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_mdio.c
@@ -0,0 +1,314 @@
+// SPDX-License-Identifier: GPL-2.0+
+// Copyright (c) 2016-2017 Hisilicon Limited.
+
+#include <linux/etherdevice.h>
+#include <linux/kernel.h>
+#include <linux/marvell_phy.h>
+
+#include "hclge_cmd.h"
+#include "hclge_main.h"
+#include "hclge_mdio.h"
+
+enum hclge_mdio_c22_op_seq {
+	HCLGE_MDIO_C22_WRITE = 1,
+	HCLGE_MDIO_C22_READ = 2
+};
+
+#define HCLGE_MDIO_CTRL_START_B		0
+#define HCLGE_MDIO_CTRL_ST_S		1
+#define HCLGE_MDIO_CTRL_ST_M		(0x3 << HCLGE_MDIO_CTRL_ST_S)
+#define HCLGE_MDIO_CTRL_OP_S		3
+#define HCLGE_MDIO_CTRL_OP_M		(0x3 << HCLGE_MDIO_CTRL_OP_S)
+
+#define HCLGE_MDIO_PHYID_S		0
+#define HCLGE_MDIO_PHYID_M		(0x1f << HCLGE_MDIO_PHYID_S)
+
+#define HCLGE_MDIO_PHYREG_S		0
+#define HCLGE_MDIO_PHYREG_M		(0x1f << HCLGE_MDIO_PHYREG_S)
+
+#define HCLGE_MDIO_STA_B		0
+
+struct hclge_mdio_cfg_cmd {
+	u8 ctrl_bit;
+	u8 phyid;
+	u8 phyad;
+	u8 rsvd;
+	__le16 reserve;
+	__le16 data_wr;
+	__le16 data_rd;
+	__le16 sta;
+};
+
+static int hclge_mdio_write(struct mii_bus *bus, int phyid, int regnum,
+			    u16 data)
+{
+	struct hclge_mdio_cfg_cmd *mdio_cmd;
+	struct hclge_dev *hdev = bus->priv;
+	struct hclge_desc desc;
+	int ret;
+
+	if (test_bit(HCLGE_COMM_STATE_CMD_DISABLE, &hdev->hw.hw.comm_state))
+		return -EBUSY;
+
+	hclge_cmd_setup_basic_desc(&desc, HCLGE_OPC_MDIO_CONFIG, false);
+
+	mdio_cmd = (struct hclge_mdio_cfg_cmd *)desc.data;
+
+	hnae3_set_field(mdio_cmd->phyid, HCLGE_MDIO_PHYID_M,
+			HCLGE_MDIO_PHYID_S, (u32)phyid);
+	hnae3_set_field(mdio_cmd->phyad, HCLGE_MDIO_PHYREG_M,
+			HCLGE_MDIO_PHYREG_S, (u32)regnum);
+
+	hnae3_set_bit(mdio_cmd->ctrl_bit, HCLGE_MDIO_CTRL_START_B, 1);
+	hnae3_set_field(mdio_cmd->ctrl_bit, HCLGE_MDIO_CTRL_ST_M,
+			HCLGE_MDIO_CTRL_ST_S, 1);
+	hnae3_set_field(mdio_cmd->ctrl_bit, HCLGE_MDIO_CTRL_OP_M,
+			HCLGE_MDIO_CTRL_OP_S, HCLGE_MDIO_C22_WRITE);
+
+	mdio_cmd->data_wr = cpu_to_le16(data);
+
+	ret = hclge_cmd_send(&hdev->hw, &desc, 1);
+	if (ret) {
+		dev_err(&hdev->pdev->dev,
+			"mdio write fail when sending cmd, status is %d.\n",
+			ret);
+		return ret;
+	}
+
+	return 0;
+}
+
+static int hclge_mdio_read(struct mii_bus *bus, int phyid, int regnum)
+{
+	struct hclge_mdio_cfg_cmd *mdio_cmd;
+	struct hclge_dev *hdev = bus->priv;
+	struct hclge_desc desc;
+	int ret;
+
+	if (test_bit(HCLGE_COMM_STATE_CMD_DISABLE, &hdev->hw.hw.comm_state))
+		return -EBUSY;
+
+	hclge_cmd_setup_basic_desc(&desc, HCLGE_OPC_MDIO_CONFIG, true);
+
+	mdio_cmd = (struct hclge_mdio_cfg_cmd *)desc.data;
+
+	hnae3_set_field(mdio_cmd->phyid, HCLGE_MDIO_PHYID_M,
+			HCLGE_MDIO_PHYID_S, (u32)phyid);
+	hnae3_set_field(mdio_cmd->phyad, HCLGE_MDIO_PHYREG_M,
+			HCLGE_MDIO_PHYREG_S, (u32)regnum);
+
+	hnae3_set_bit(mdio_cmd->ctrl_bit, HCLGE_MDIO_CTRL_START_B, 1);
+	hnae3_set_field(mdio_cmd->ctrl_bit, HCLGE_MDIO_CTRL_ST_M,
+			HCLGE_MDIO_CTRL_ST_S, 1);
+	hnae3_set_field(mdio_cmd->ctrl_bit, HCLGE_MDIO_CTRL_OP_M,
+			HCLGE_MDIO_CTRL_OP_S, HCLGE_MDIO_C22_READ);
+
+	/* Read out phy data */
+	ret = hclge_cmd_send(&hdev->hw, &desc, 1);
+	if (ret) {
+		dev_err(&hdev->pdev->dev,
+			"mdio read fail when get data, status is %d.\n",
+			ret);
+		return ret;
+	}
+
+	if (hnae3_get_bit(le16_to_cpu(mdio_cmd->sta), HCLGE_MDIO_STA_B)) {
+		dev_err(&hdev->pdev->dev, "mdio read data error\n");
+		return -EIO;
+	}
+
+	return le16_to_cpu(mdio_cmd->data_rd);
+}
+
+int hclge_mac_mdio_config(struct hclge_dev *hdev)
+{
+#define PHY_INEXISTENT	255
+
+	struct hclge_mac *mac = &hdev->hw.mac;
+	struct phy_device *phydev;
+	struct mii_bus *mdio_bus;
+	int ret;
+
+	if (hdev->hw.mac.phy_addr == PHY_INEXISTENT) {
+		dev_info(&hdev->pdev->dev,
+			 "no phy device is connected to mdio bus\n");
+		return 0;
+	} else if (hdev->hw.mac.phy_addr >= PHY_MAX_ADDR) {
+		dev_err(&hdev->pdev->dev, "phy_addr(%u) is too large.\n",
+			hdev->hw.mac.phy_addr);
+		return -EINVAL;
+	}
+
+	mdio_bus = devm_mdiobus_alloc(&hdev->pdev->dev);
+	if (!mdio_bus)
+		return -ENOMEM;
+
+	mdio_bus->name = "hisilicon MII bus";
+	mdio_bus->read = hclge_mdio_read;
+	mdio_bus->write = hclge_mdio_write;
+	snprintf(mdio_bus->id, MII_BUS_ID_SIZE, "%s-%s", "mii",
+		 dev_name(&hdev->pdev->dev));
+
+	mdio_bus->parent = &hdev->pdev->dev;
+	mdio_bus->priv = hdev;
+	mdio_bus->phy_mask = ~(1U << mac->phy_addr);
+	ret = mdiobus_register(mdio_bus);
+	if (ret) {
+		dev_err(mdio_bus->parent,
+			"failed to register MDIO bus, ret = %d\n", ret);
+		return ret;
+	}
+
+	phydev = mdiobus_get_phy(mdio_bus, mac->phy_addr);
+	if (!phydev) {
+		dev_err(mdio_bus->parent, "Failed to get phy device\n");
+		mdiobus_unregister(mdio_bus);
+		return -EIO;
+	}
+
+	mac->phydev = phydev;
+	mac->mdio_bus = mdio_bus;
+
+	return 0;
+}
+
+static void hclge_mac_adjust_link(struct net_device *netdev)
+{
+	struct hnae3_handle *h = *((void **)netdev_priv(netdev));
+	struct hclge_vport *vport = hclge_get_vport(h);
+	struct hclge_dev *hdev = vport->back;
+	int duplex, speed;
+	int ret;
+
+	/* When phy link down, do nothing */
+	if (netdev->phydev->link == 0)
+		return;
+
+	speed = netdev->phydev->speed;
+	duplex = netdev->phydev->duplex;
+
+	ret = hclge_cfg_mac_speed_dup(hdev, speed, duplex, 0);
+	if (ret)
+		netdev_err(netdev, "failed to adjust link.\n");
+
+	hdev->hw.mac.req_speed = (u32)speed;
+	hdev->hw.mac.req_duplex = (u8)duplex;
+
+	ret = hclge_cfg_flowctrl(hdev);
+	if (ret)
+		netdev_err(netdev, "failed to configure flow control.\n");
+}
+
+int hclge_mac_connect_phy(struct hnae3_handle *handle)
+{
+	struct hclge_vport *vport = hclge_get_vport(handle);
+	struct hclge_dev *hdev = vport->back;
+	struct net_device *netdev = hdev->vport[0].nic.netdev;
+	struct phy_device *phydev = hdev->hw.mac.phydev;
+	__ETHTOOL_DECLARE_LINK_MODE_MASK(mask) = { 0, };
+	int ret;
+
+	if (!phydev)
+		return 0;
+
+	linkmode_clear_bit(ETHTOOL_LINK_MODE_FIBRE_BIT, phydev->supported);
+
+	phydev->dev_flags |= MARVELL_PHY_LED0_LINK_LED1_ACTIVE;
+
+	ret = phy_connect_direct(netdev, phydev,
+				 hclge_mac_adjust_link,
+				 PHY_INTERFACE_MODE_SGMII);
+	if (ret) {
+		netdev_err(netdev, "phy_connect_direct err.\n");
+		return ret;
+	}
+
+	linkmode_copy(mask, hdev->hw.mac.supported);
+	linkmode_and(phydev->supported, phydev->supported, mask);
+	linkmode_copy(phydev->advertising, phydev->supported);
+
+	/* supported flag is Pause and Asym Pause, but default advertising
+	 * should be rx on, tx on, so need clear Asym Pause in advertising
+	 * flag
+	 */
+	linkmode_clear_bit(ETHTOOL_LINK_MODE_Asym_Pause_BIT,
+			   phydev->advertising);
+
+	phy_attached_info(phydev);
+
+	return 0;
+}
+
+void hclge_mac_disconnect_phy(struct hnae3_handle *handle)
+{
+	struct hclge_vport *vport = hclge_get_vport(handle);
+	struct hclge_dev *hdev = vport->back;
+	struct phy_device *phydev = hdev->hw.mac.phydev;
+
+	if (!phydev)
+		return;
+
+	phy_disconnect(phydev);
+}
+
+void hclge_mac_start_phy(struct hclge_dev *hdev)
+{
+	struct phy_device *phydev = hdev->hw.mac.phydev;
+
+	if (!phydev)
+		return;
+
+	phy_loopback(phydev, false, 0);
+
+	phy_start(phydev);
+}
+
+void hclge_mac_stop_phy(struct hclge_dev *hdev)
+{
+	struct net_device *netdev = hdev->vport[0].nic.netdev;
+	struct phy_device *phydev = netdev->phydev;
+
+	if (!phydev)
+		return;
+
+	phy_stop(phydev);
+}
+
+u16 hclge_read_phy_reg(struct hclge_dev *hdev, u16 reg_addr)
+{
+	struct hclge_phy_reg_cmd *req;
+	struct hclge_desc desc;
+	int ret;
+
+	hclge_cmd_setup_basic_desc(&desc, HCLGE_OPC_PHY_REG, true);
+
+	req = (struct hclge_phy_reg_cmd *)desc.data;
+	req->reg_addr = cpu_to_le16(reg_addr);
+
+	ret = hclge_cmd_send(&hdev->hw, &desc, 1);
+	if (ret)
+		dev_err(&hdev->pdev->dev,
+			"failed to read phy reg, ret = %d.\n", ret);
+
+	return le16_to_cpu(req->reg_val);
+}
+
+int hclge_write_phy_reg(struct hclge_dev *hdev, u16 reg_addr, u16 val)
+{
+	struct hclge_phy_reg_cmd *req;
+	struct hclge_desc desc;
+	int ret;
+
+	hclge_cmd_setup_basic_desc(&desc, HCLGE_OPC_PHY_REG, false);
+
+	req = (struct hclge_phy_reg_cmd *)desc.data;
+	req->reg_addr = cpu_to_le16(reg_addr);
+	req->reg_val = cpu_to_le16(val);
+
+	ret = hclge_cmd_send(&hdev->hw, &desc, 1);
+	if (ret)
+		dev_err(&hdev->pdev->dev,
+			"failed to write phy reg, ret = %d.\n", ret);
+
+	return ret;
+}
diff --git a/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_mdio.h b/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_mdio.h
new file mode 100644
index 000000000000..4200d0b6d931
--- /dev/null
+++ b/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_mdio.h
@@ -0,0 +1,19 @@
+/* SPDX-License-Identifier: GPL-2.0+ */
+// Copyright (c) 2016-2017 Hisilicon Limited.
+
+#ifndef __HCLGE_MDIO_H
+#define __HCLGE_MDIO_H
+
+#include "hnae3.h"
+
+struct hclge_dev;
+
+int hclge_mac_mdio_config(struct hclge_dev *hdev);
+int hclge_mac_connect_phy(struct hnae3_handle *handle);
+void hclge_mac_disconnect_phy(struct hnae3_handle *handle);
+void hclge_mac_start_phy(struct hclge_dev *hdev);
+void hclge_mac_stop_phy(struct hclge_dev *hdev);
+u16 hclge_read_phy_reg(struct hclge_dev *hdev, u16 reg_addr);
+int hclge_write_phy_reg(struct hclge_dev *hdev, u16 reg_addr, u16 val);
+
+#endif
diff --git a/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_ptp.c b/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_ptp.c
new file mode 100644
index 000000000000..4bd52eab3914
--- /dev/null
+++ b/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_ptp.c
@@ -0,0 +1,556 @@
+// SPDX-License-Identifier: GPL-2.0+
+// Copyright (c) 2021 Hisilicon Limited.
+
+#include <linux/skbuff.h>
+#include <linux/string_choices.h>
+#include "hclge_main.h"
+#include "hnae3.h"
+
+static int hclge_ptp_get_cycle(struct hclge_dev *hdev)
+{
+	struct hclge_ptp *ptp = hdev->ptp;
+
+	ptp->cycle.quo = readl(hdev->ptp->io_base + HCLGE_PTP_CYCLE_QUO_REG) &
+			 HCLGE_PTP_CYCLE_QUO_MASK;
+	ptp->cycle.numer = readl(hdev->ptp->io_base + HCLGE_PTP_CYCLE_NUM_REG);
+	ptp->cycle.den = readl(hdev->ptp->io_base + HCLGE_PTP_CYCLE_DEN_REG);
+
+	if (ptp->cycle.den == 0) {
+		dev_err(&hdev->pdev->dev, "invalid ptp cycle denominator!\n");
+		return -EINVAL;
+	}
+
+	return 0;
+}
+
+static int hclge_ptp_adjfine(struct ptp_clock_info *ptp, long scaled_ppm)
+{
+	struct hclge_dev *hdev = hclge_ptp_get_hdev(ptp);
+	struct hclge_ptp_cycle *cycle = &hdev->ptp->cycle;
+	u64 adj_val, adj_base;
+	unsigned long flags;
+	u32 quo, numerator;
+
+	adj_base = (u64)cycle->quo * (u64)cycle->den + (u64)cycle->numer;
+	adj_val = adjust_by_scaled_ppm(adj_base, scaled_ppm);
+
+	/* This clock cycle is defined by three part: quotient, numerator
+	 * and denominator. For example, 2.5ns, the quotient is 2,
+	 * denominator is fixed to ptp->cycle.den, and numerator
+	 * is 0.5 * ptp->cycle.den.
+	 */
+	quo = div_u64_rem(adj_val, cycle->den, &numerator);
+
+	spin_lock_irqsave(&hdev->ptp->lock, flags);
+	writel(quo & HCLGE_PTP_CYCLE_QUO_MASK,
+	       hdev->ptp->io_base + HCLGE_PTP_CYCLE_QUO_REG);
+	writel(numerator, hdev->ptp->io_base + HCLGE_PTP_CYCLE_NUM_REG);
+	writel(cycle->den, hdev->ptp->io_base + HCLGE_PTP_CYCLE_DEN_REG);
+	writel(HCLGE_PTP_CYCLE_ADJ_EN,
+	       hdev->ptp->io_base + HCLGE_PTP_CYCLE_CFG_REG);
+	spin_unlock_irqrestore(&hdev->ptp->lock, flags);
+
+	return 0;
+}
+
+bool hclge_ptp_set_tx_info(struct hnae3_handle *handle, struct sk_buff *skb)
+{
+	struct hclge_vport *vport = hclge_get_vport(handle);
+	struct hclge_dev *hdev = vport->back;
+	struct hclge_ptp *ptp = hdev->ptp;
+
+	if (!ptp)
+		return false;
+
+	if (!test_bit(HCLGE_PTP_FLAG_TX_EN, &ptp->flags) ||
+	    test_and_set_bit(HCLGE_STATE_PTP_TX_HANDLING, &hdev->state)) {
+		ptp->tx_skipped++;
+		return false;
+	}
+
+	ptp->tx_start = jiffies;
+	ptp->tx_skb = skb_get(skb);
+	ptp->tx_cnt++;
+
+	return true;
+}
+
+void hclge_ptp_clean_tx_hwts(struct hclge_dev *hdev)
+{
+	struct sk_buff *skb = hdev->ptp->tx_skb;
+	struct skb_shared_hwtstamps hwts;
+	u32 hi, lo;
+	u64 ns;
+
+	ns = readl(hdev->ptp->io_base + HCLGE_PTP_TX_TS_NSEC_REG) &
+	     HCLGE_PTP_TX_TS_NSEC_MASK;
+	lo = readl(hdev->ptp->io_base + HCLGE_PTP_TX_TS_SEC_L_REG);
+	hi = readl(hdev->ptp->io_base + HCLGE_PTP_TX_TS_SEC_H_REG) &
+	     HCLGE_PTP_TX_TS_SEC_H_MASK;
+	hdev->ptp->last_tx_seqid = readl(hdev->ptp->io_base +
+		HCLGE_PTP_TX_TS_SEQID_REG);
+
+	if (skb) {
+		hdev->ptp->tx_skb = NULL;
+		hdev->ptp->tx_cleaned++;
+
+		ns += (((u64)hi) << 32 | lo) * NSEC_PER_SEC;
+		hwts.hwtstamp = ns_to_ktime(ns);
+		skb_tstamp_tx(skb, &hwts);
+		dev_kfree_skb_any(skb);
+	}
+
+	clear_bit(HCLGE_STATE_PTP_TX_HANDLING, &hdev->state);
+}
+
+void hclge_ptp_get_rx_hwts(struct hnae3_handle *handle, struct sk_buff *skb,
+			   u32 nsec, u32 sec)
+{
+	struct hclge_vport *vport = hclge_get_vport(handle);
+	struct hclge_dev *hdev = vport->back;
+	unsigned long flags;
+	u64 ns = nsec;
+	u32 sec_h;
+
+	if (!hdev->ptp || !test_bit(HCLGE_PTP_FLAG_RX_EN, &hdev->ptp->flags))
+		return;
+
+	/* Since the BD does not have enough space for the higher 16 bits of
+	 * second, and this part will not change frequently, so read it
+	 * from register.
+	 */
+	spin_lock_irqsave(&hdev->ptp->lock, flags);
+	sec_h = readl(hdev->ptp->io_base + HCLGE_PTP_CUR_TIME_SEC_H_REG);
+	spin_unlock_irqrestore(&hdev->ptp->lock, flags);
+
+	ns += (((u64)sec_h) << HCLGE_PTP_SEC_H_OFFSET | sec) * NSEC_PER_SEC;
+	skb_hwtstamps(skb)->hwtstamp = ns_to_ktime(ns);
+	hdev->ptp->last_rx = jiffies;
+	hdev->ptp->rx_cnt++;
+}
+
+static int hclge_ptp_gettimex(struct ptp_clock_info *ptp, struct timespec64 *ts,
+			      struct ptp_system_timestamp *sts)
+{
+	struct hclge_dev *hdev = hclge_ptp_get_hdev(ptp);
+	unsigned long flags;
+	u32 hi, lo;
+	u64 ns;
+
+	spin_lock_irqsave(&hdev->ptp->lock, flags);
+	ns = readl(hdev->ptp->io_base + HCLGE_PTP_CUR_TIME_NSEC_REG);
+	hi = readl(hdev->ptp->io_base + HCLGE_PTP_CUR_TIME_SEC_H_REG);
+	lo = readl(hdev->ptp->io_base + HCLGE_PTP_CUR_TIME_SEC_L_REG);
+	spin_unlock_irqrestore(&hdev->ptp->lock, flags);
+
+	ns += (((u64)hi) << HCLGE_PTP_SEC_H_OFFSET | lo) * NSEC_PER_SEC;
+	*ts = ns_to_timespec64(ns);
+
+	return 0;
+}
+
+static int hclge_ptp_settime(struct ptp_clock_info *ptp,
+			     const struct timespec64 *ts)
+{
+	struct hclge_dev *hdev = hclge_ptp_get_hdev(ptp);
+	unsigned long flags;
+
+	spin_lock_irqsave(&hdev->ptp->lock, flags);
+	writel(ts->tv_nsec, hdev->ptp->io_base + HCLGE_PTP_TIME_NSEC_REG);
+	writel(ts->tv_sec >> HCLGE_PTP_SEC_H_OFFSET,
+	       hdev->ptp->io_base + HCLGE_PTP_TIME_SEC_H_REG);
+	writel(ts->tv_sec & HCLGE_PTP_SEC_L_MASK,
+	       hdev->ptp->io_base + HCLGE_PTP_TIME_SEC_L_REG);
+	/* synchronize the time of phc */
+	writel(HCLGE_PTP_TIME_SYNC_EN,
+	       hdev->ptp->io_base + HCLGE_PTP_TIME_SYNC_REG);
+	spin_unlock_irqrestore(&hdev->ptp->lock, flags);
+
+	return 0;
+}
+
+static int hclge_ptp_adjtime(struct ptp_clock_info *ptp, s64 delta)
+{
+	struct hclge_dev *hdev = hclge_ptp_get_hdev(ptp);
+	unsigned long flags;
+	bool is_neg = false;
+	u32 adj_val = 0;
+
+	if (delta < 0) {
+		adj_val |= HCLGE_PTP_TIME_NSEC_NEG;
+		delta = -delta;
+		is_neg = true;
+	}
+
+	if (delta > HCLGE_PTP_TIME_NSEC_MASK) {
+		struct timespec64 ts;
+		s64 ns;
+
+		hclge_ptp_gettimex(ptp, &ts, NULL);
+		ns = timespec64_to_ns(&ts);
+		ns = is_neg ? ns - delta : ns + delta;
+		ts = ns_to_timespec64(ns);
+		return hclge_ptp_settime(ptp, &ts);
+	}
+
+	adj_val |= delta & HCLGE_PTP_TIME_NSEC_MASK;
+
+	spin_lock_irqsave(&hdev->ptp->lock, flags);
+	writel(adj_val, hdev->ptp->io_base + HCLGE_PTP_TIME_NSEC_REG);
+	writel(HCLGE_PTP_TIME_ADJ_EN,
+	       hdev->ptp->io_base + HCLGE_PTP_TIME_ADJ_REG);
+	spin_unlock_irqrestore(&hdev->ptp->lock, flags);
+
+	return 0;
+}
+
+int hclge_ptp_get_cfg(struct hclge_dev *hdev, struct ifreq *ifr)
+{
+	if (!test_bit(HCLGE_STATE_PTP_EN, &hdev->state))
+		return -EOPNOTSUPP;
+
+	return copy_to_user(ifr->ifr_data, &hdev->ptp->ts_cfg,
+		sizeof(struct hwtstamp_config)) ? -EFAULT : 0;
+}
+
+static int hclge_ptp_int_en(struct hclge_dev *hdev, bool en)
+{
+	struct hclge_ptp_int_cmd *req;
+	struct hclge_desc desc;
+	int ret;
+
+	req = (struct hclge_ptp_int_cmd *)desc.data;
+	hclge_cmd_setup_basic_desc(&desc, HCLGE_OPC_PTP_INT_EN, false);
+	req->int_en = en ? 1 : 0;
+
+	ret = hclge_cmd_send(&hdev->hw, &desc, 1);
+	if (ret)
+		dev_err(&hdev->pdev->dev,
+			"failed to %s ptp interrupt, ret = %d\n",
+			str_enable_disable(en), ret);
+
+	return ret;
+}
+
+int hclge_ptp_cfg_qry(struct hclge_dev *hdev, u32 *cfg)
+{
+	struct hclge_ptp_cfg_cmd *req;
+	struct hclge_desc desc;
+	int ret;
+
+	req = (struct hclge_ptp_cfg_cmd *)desc.data;
+	hclge_cmd_setup_basic_desc(&desc, HCLGE_OPC_PTP_MODE_CFG, true);
+	ret = hclge_cmd_send(&hdev->hw, &desc, 1);
+	if (ret) {
+		dev_err(&hdev->pdev->dev,
+			"failed to query ptp config, ret = %d\n", ret);
+		return ret;
+	}
+
+	*cfg = le32_to_cpu(req->cfg);
+
+	return 0;
+}
+
+static int hclge_ptp_cfg(struct hclge_dev *hdev, u32 cfg)
+{
+	struct hclge_ptp_cfg_cmd *req;
+	struct hclge_desc desc;
+	int ret;
+
+	req = (struct hclge_ptp_cfg_cmd *)desc.data;
+	hclge_cmd_setup_basic_desc(&desc, HCLGE_OPC_PTP_MODE_CFG, false);
+	req->cfg = cpu_to_le32(cfg);
+	ret = hclge_cmd_send(&hdev->hw, &desc, 1);
+	if (ret)
+		dev_err(&hdev->pdev->dev,
+			"failed to config ptp, ret = %d\n", ret);
+
+	return ret;
+}
+
+static int hclge_ptp_set_tx_mode(struct hwtstamp_config *cfg,
+				 unsigned long *flags, u32 *ptp_cfg)
+{
+	switch (cfg->tx_type) {
+	case HWTSTAMP_TX_OFF:
+		clear_bit(HCLGE_PTP_FLAG_TX_EN, flags);
+		break;
+	case HWTSTAMP_TX_ON:
+		set_bit(HCLGE_PTP_FLAG_TX_EN, flags);
+		*ptp_cfg |= HCLGE_PTP_TX_EN_B;
+		break;
+	default:
+		return -ERANGE;
+	}
+
+	return 0;
+}
+
+static int hclge_ptp_set_rx_mode(struct hwtstamp_config *cfg,
+				 unsigned long *flags, u32 *ptp_cfg)
+{
+	int rx_filter = cfg->rx_filter;
+
+	switch (cfg->rx_filter) {
+	case HWTSTAMP_FILTER_NONE:
+		clear_bit(HCLGE_PTP_FLAG_RX_EN, flags);
+		break;
+	case HWTSTAMP_FILTER_PTP_V1_L4_SYNC:
+	case HWTSTAMP_FILTER_PTP_V1_L4_DELAY_REQ:
+	case HWTSTAMP_FILTER_PTP_V1_L4_EVENT:
+		set_bit(HCLGE_PTP_FLAG_RX_EN, flags);
+		*ptp_cfg |= HCLGE_PTP_RX_EN_B;
+		*ptp_cfg |= HCLGE_PTP_UDP_FULL_TYPE << HCLGE_PTP_UDP_EN_SHIFT;
+		rx_filter = HWTSTAMP_FILTER_PTP_V1_L4_EVENT;
+		break;
+	case HWTSTAMP_FILTER_PTP_V2_EVENT:
+	case HWTSTAMP_FILTER_PTP_V2_L4_EVENT:
+	case HWTSTAMP_FILTER_PTP_V2_SYNC:
+	case HWTSTAMP_FILTER_PTP_V2_L4_SYNC:
+	case HWTSTAMP_FILTER_PTP_V2_DELAY_REQ:
+	case HWTSTAMP_FILTER_PTP_V2_L4_DELAY_REQ:
+	case HWTSTAMP_FILTER_PTP_V2_L2_EVENT:
+	case HWTSTAMP_FILTER_PTP_V2_L2_SYNC:
+	case HWTSTAMP_FILTER_PTP_V2_L2_DELAY_REQ:
+		set_bit(HCLGE_PTP_FLAG_RX_EN, flags);
+		*ptp_cfg |= HCLGE_PTP_RX_EN_B;
+		*ptp_cfg |= HCLGE_PTP_UDP_FULL_TYPE << HCLGE_PTP_UDP_EN_SHIFT;
+		*ptp_cfg |= HCLGE_PTP_MSG1_V2_DEFAULT << HCLGE_PTP_MSG1_SHIFT;
+		*ptp_cfg |= HCLGE_PTP_MSG0_V2_EVENT << HCLGE_PTP_MSG0_SHIFT;
+		*ptp_cfg |= HCLGE_PTP_MSG_TYPE_V2 << HCLGE_PTP_MSG_TYPE_SHIFT;
+		rx_filter = HWTSTAMP_FILTER_PTP_V2_EVENT;
+		break;
+	case HWTSTAMP_FILTER_ALL:
+	default:
+		return -ERANGE;
+	}
+
+	cfg->rx_filter = rx_filter;
+
+	return 0;
+}
+
+static int hclge_ptp_set_ts_mode(struct hclge_dev *hdev,
+				 struct hwtstamp_config *cfg)
+{
+	unsigned long flags = hdev->ptp->flags;
+	u32 ptp_cfg = 0;
+	int ret;
+
+	if (test_bit(HCLGE_PTP_FLAG_EN, &hdev->ptp->flags))
+		ptp_cfg |= HCLGE_PTP_EN_B;
+
+	ret = hclge_ptp_set_tx_mode(cfg, &flags, &ptp_cfg);
+	if (ret)
+		return ret;
+
+	ret = hclge_ptp_set_rx_mode(cfg, &flags, &ptp_cfg);
+	if (ret)
+		return ret;
+
+	ret = hclge_ptp_cfg(hdev, ptp_cfg);
+	if (ret)
+		return ret;
+
+	hdev->ptp->flags = flags;
+	hdev->ptp->ptp_cfg = ptp_cfg;
+
+	return 0;
+}
+
+int hclge_ptp_set_cfg(struct hclge_dev *hdev, struct ifreq *ifr)
+{
+	struct hwtstamp_config cfg;
+	int ret;
+
+	if (!test_bit(HCLGE_STATE_PTP_EN, &hdev->state)) {
+		dev_err(&hdev->pdev->dev, "phc is unsupported\n");
+		return -EOPNOTSUPP;
+	}
+
+	if (copy_from_user(&cfg, ifr->ifr_data, sizeof(cfg)))
+		return -EFAULT;
+
+	ret = hclge_ptp_set_ts_mode(hdev, &cfg);
+	if (ret)
+		return ret;
+
+	hdev->ptp->ts_cfg = cfg;
+
+	return copy_to_user(ifr->ifr_data, &cfg, sizeof(cfg)) ? -EFAULT : 0;
+}
+
+int hclge_ptp_get_ts_info(struct hnae3_handle *handle,
+			  struct kernel_ethtool_ts_info *info)
+{
+	struct hclge_vport *vport = hclge_get_vport(handle);
+	struct hclge_dev *hdev = vport->back;
+
+	if (!test_bit(HCLGE_STATE_PTP_EN, &hdev->state)) {
+		dev_err(&hdev->pdev->dev, "phc is unsupported\n");
+		return -EOPNOTSUPP;
+	}
+
+	info->so_timestamping = SOF_TIMESTAMPING_TX_SOFTWARE |
+				SOF_TIMESTAMPING_TX_HARDWARE |
+				SOF_TIMESTAMPING_RX_HARDWARE |
+				SOF_TIMESTAMPING_RAW_HARDWARE;
+
+	if (hdev->ptp->clock)
+		info->phc_index = ptp_clock_index(hdev->ptp->clock);
+
+	info->tx_types = BIT(HWTSTAMP_TX_OFF) | BIT(HWTSTAMP_TX_ON);
+
+	info->rx_filters = BIT(HWTSTAMP_FILTER_NONE) |
+			   BIT(HWTSTAMP_FILTER_PTP_V2_L2_EVENT) |
+			   BIT(HWTSTAMP_FILTER_PTP_V2_L2_SYNC) |
+			   BIT(HWTSTAMP_FILTER_PTP_V2_L2_DELAY_REQ);
+
+	info->rx_filters |= BIT(HWTSTAMP_FILTER_PTP_V1_L4_SYNC) |
+			    BIT(HWTSTAMP_FILTER_PTP_V1_L4_DELAY_REQ) |
+			    BIT(HWTSTAMP_FILTER_PTP_V2_EVENT) |
+			    BIT(HWTSTAMP_FILTER_PTP_V2_L4_EVENT) |
+			    BIT(HWTSTAMP_FILTER_PTP_V2_SYNC) |
+			    BIT(HWTSTAMP_FILTER_PTP_V2_L4_SYNC) |
+			    BIT(HWTSTAMP_FILTER_PTP_V2_DELAY_REQ) |
+			    BIT(HWTSTAMP_FILTER_PTP_V2_L4_DELAY_REQ);
+
+	return 0;
+}
+
+static int hclge_ptp_create_clock(struct hclge_dev *hdev)
+{
+	struct hclge_ptp *ptp;
+
+	ptp = devm_kzalloc(&hdev->pdev->dev, sizeof(*ptp), GFP_KERNEL);
+	if (!ptp)
+		return -ENOMEM;
+
+	ptp->hdev = hdev;
+	snprintf(ptp->info.name, sizeof(ptp->info.name), "%s",
+		 HCLGE_DRIVER_NAME);
+	ptp->info.owner = THIS_MODULE;
+	ptp->info.max_adj = HCLGE_PTP_CYCLE_ADJ_MAX;
+	ptp->info.n_ext_ts = 0;
+	ptp->info.pps = 0;
+	ptp->info.adjfine = hclge_ptp_adjfine;
+	ptp->info.adjtime = hclge_ptp_adjtime;
+	ptp->info.gettimex64 = hclge_ptp_gettimex;
+	ptp->info.settime64 = hclge_ptp_settime;
+
+	ptp->info.n_alarm = 0;
+
+	spin_lock_init(&ptp->lock);
+	ptp->io_base = hdev->hw.hw.io_base + HCLGE_PTP_REG_OFFSET;
+	ptp->ts_cfg.rx_filter = HWTSTAMP_FILTER_NONE;
+	ptp->ts_cfg.tx_type = HWTSTAMP_TX_OFF;
+	hdev->ptp = ptp;
+
+	ptp->clock = ptp_clock_register(&ptp->info, &hdev->pdev->dev);
+	if (IS_ERR(ptp->clock)) {
+		dev_err(&hdev->pdev->dev,
+			"%d failed to register ptp clock, ret = %ld\n",
+			ptp->info.n_alarm, PTR_ERR(ptp->clock));
+		return -ENODEV;
+	} else if (!ptp->clock) {
+		dev_err(&hdev->pdev->dev, "failed to register ptp clock\n");
+		return -ENODEV;
+	}
+
+	return 0;
+}
+
+static void hclge_ptp_destroy_clock(struct hclge_dev *hdev)
+{
+	ptp_clock_unregister(hdev->ptp->clock);
+	hdev->ptp->clock = NULL;
+	devm_kfree(&hdev->pdev->dev, hdev->ptp);
+	hdev->ptp = NULL;
+}
+
+int hclge_ptp_init(struct hclge_dev *hdev)
+{
+	struct hnae3_ae_dev *ae_dev = pci_get_drvdata(hdev->pdev);
+	struct timespec64 ts;
+	int ret;
+
+	if (!test_bit(HNAE3_DEV_SUPPORT_PTP_B, ae_dev->caps))
+		return 0;
+
+	if (!hdev->ptp) {
+		ret = hclge_ptp_create_clock(hdev);
+		if (ret)
+			return ret;
+
+		ret = hclge_ptp_get_cycle(hdev);
+		if (ret)
+			goto out;
+	}
+
+	ret = hclge_ptp_int_en(hdev, true);
+	if (ret)
+		goto out;
+
+	set_bit(HCLGE_PTP_FLAG_EN, &hdev->ptp->flags);
+	ret = hclge_ptp_adjfine(&hdev->ptp->info, 0);
+	if (ret) {
+		dev_err(&hdev->pdev->dev,
+			"failed to init freq, ret = %d\n", ret);
+		goto out_clear_int;
+	}
+
+	ret = hclge_ptp_set_ts_mode(hdev, &hdev->ptp->ts_cfg);
+	if (ret) {
+		dev_err(&hdev->pdev->dev,
+			"failed to init ts mode, ret = %d\n", ret);
+		goto out_clear_int;
+	}
+
+	ktime_get_real_ts64(&ts);
+	ret = hclge_ptp_settime(&hdev->ptp->info, &ts);
+	if (ret) {
+		dev_err(&hdev->pdev->dev,
+			"failed to init ts time, ret = %d\n", ret);
+		goto out_clear_int;
+	}
+
+	set_bit(HCLGE_STATE_PTP_EN, &hdev->state);
+	dev_info(&hdev->pdev->dev, "phc initializes ok!\n");
+
+	return 0;
+
+out_clear_int:
+	clear_bit(HCLGE_PTP_FLAG_EN, &hdev->ptp->flags);
+	hclge_ptp_int_en(hdev, false);
+out:
+	hclge_ptp_destroy_clock(hdev);
+
+	return ret;
+}
+
+void hclge_ptp_uninit(struct hclge_dev *hdev)
+{
+	struct hclge_ptp *ptp = hdev->ptp;
+
+	if (!ptp)
+		return;
+
+	hclge_ptp_int_en(hdev, false);
+	clear_bit(HCLGE_STATE_PTP_EN, &hdev->state);
+	clear_bit(HCLGE_PTP_FLAG_EN, &ptp->flags);
+	ptp->ts_cfg.rx_filter = HWTSTAMP_FILTER_NONE;
+	ptp->ts_cfg.tx_type = HWTSTAMP_TX_OFF;
+
+	if (hclge_ptp_set_ts_mode(hdev, &ptp->ts_cfg))
+		dev_err(&hdev->pdev->dev, "failed to disable phc\n");
+
+	if (ptp->tx_skb) {
+		struct sk_buff *skb = ptp->tx_skb;
+
+		ptp->tx_skb = NULL;
+		dev_kfree_skb_any(skb);
+	}
+
+	hclge_ptp_destroy_clock(hdev);
+}
diff --git a/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_ptp.h b/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_ptp.h
new file mode 100644
index 000000000000..61faddcc3dd0
--- /dev/null
+++ b/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_ptp.h
@@ -0,0 +1,143 @@
+/* SPDX-License-Identifier: GPL-2.0+ */
+// Copyright (c) 2021 Hisilicon Limited.
+
+#ifndef __HCLGE_PTP_H
+#define __HCLGE_PTP_H
+
+#include <linux/ptp_clock_kernel.h>
+#include <linux/net_tstamp.h>
+#include <linux/types.h>
+
+struct hclge_dev;
+struct ifreq;
+
+#define HCLGE_PTP_REG_OFFSET	0x29000
+
+#define HCLGE_PTP_TX_TS_SEQID_REG	0x0
+#define HCLGE_PTP_TX_TS_NSEC_REG	0x4
+#define HCLGE_PTP_TX_TS_NSEC_MASK	GENMASK(29, 0)
+#define HCLGE_PTP_TX_TS_SEC_L_REG	0x8
+#define HCLGE_PTP_TX_TS_SEC_H_REG	0xC
+#define HCLGE_PTP_TX_TS_SEC_H_MASK	GENMASK(15, 0)
+#define HCLGE_PTP_TX_TS_CNT_REG		0x30
+
+#define HCLGE_PTP_TIME_SEC_H_REG	0x50
+#define HCLGE_PTP_TIME_SEC_H_MASK	GENMASK(15, 0)
+#define HCLGE_PTP_TIME_SEC_L_REG	0x54
+#define HCLGE_PTP_TIME_NSEC_REG		0x58
+#define HCLGE_PTP_TIME_NSEC_MASK	0x3fffffffLL
+#define HCLGE_PTP_TIME_NSEC_NEG		BIT(31)
+#define HCLGE_PTP_TIME_SYNC_REG		0x5C
+#define HCLGE_PTP_TIME_SYNC_EN		BIT(0)
+#define HCLGE_PTP_TIME_ADJ_REG		0x60
+#define HCLGE_PTP_TIME_ADJ_EN		BIT(0)
+#define HCLGE_PTP_CYCLE_QUO_REG		0x64
+#define HCLGE_PTP_CYCLE_QUO_MASK	GENMASK(7, 0)
+#define HCLGE_PTP_CYCLE_DEN_REG		0x68
+#define HCLGE_PTP_CYCLE_NUM_REG		0x6C
+#define HCLGE_PTP_CYCLE_CFG_REG		0x70
+#define HCLGE_PTP_CYCLE_ADJ_EN		BIT(0)
+#define HCLGE_PTP_CUR_TIME_SEC_H_REG	0x74
+#define HCLGE_PTP_CUR_TIME_SEC_L_REG	0x78
+#define HCLGE_PTP_CUR_TIME_NSEC_REG	0x7C
+
+#define HCLGE_PTP_CYCLE_ADJ_MAX		500000000
+#define HCLGE_PTP_SEC_H_OFFSET		32u
+#define HCLGE_PTP_SEC_L_MASK		GENMASK(31, 0)
+
+#define HCLGE_PTP_FLAG_EN		0
+#define HCLGE_PTP_FLAG_TX_EN		1
+#define HCLGE_PTP_FLAG_RX_EN		2
+
+struct hclge_ptp_cycle {
+	u32 quo;
+	u32 numer;
+	u32 den;
+};
+
+struct hclge_ptp {
+	struct hclge_dev *hdev;
+	struct ptp_clock *clock;
+	struct sk_buff *tx_skb;
+	unsigned long flags;
+	void __iomem *io_base;
+	struct ptp_clock_info info;
+	struct hwtstamp_config ts_cfg;
+	spinlock_t lock;	/* protects ptp registers */
+	u32 ptp_cfg;
+	u32 last_tx_seqid;
+	struct hclge_ptp_cycle cycle;
+	unsigned long tx_start;
+	unsigned long tx_cnt;
+	unsigned long tx_skipped;
+	unsigned long tx_cleaned;
+	unsigned long last_rx;
+	unsigned long rx_cnt;
+	unsigned long tx_timeout;
+};
+
+struct hclge_ptp_int_cmd {
+#define HCLGE_PTP_INT_EN_B	BIT(0)
+
+	u8 int_en;
+	u8 rsvd[23];
+};
+
+enum hclge_ptp_udp_type {
+	HCLGE_PTP_UDP_NOT_TYPE,
+	HCLGE_PTP_UDP_P13F_TYPE,
+	HCLGE_PTP_UDP_P140_TYPE,
+	HCLGE_PTP_UDP_FULL_TYPE,
+};
+
+enum hclge_ptp_msg_type {
+	HCLGE_PTP_MSG_TYPE_V2_L2,
+	HCLGE_PTP_MSG_TYPE_V2,
+	HCLGE_PTP_MSG_TYPE_V2_EVENT,
+};
+
+enum hclge_ptp_msg0_type {
+	HCLGE_PTP_MSG0_V2_DELAY_REQ = 1,
+	HCLGE_PTP_MSG0_V2_PDELAY_REQ,
+	HCLGE_PTP_MSG0_V2_DELAY_RESP,
+	HCLGE_PTP_MSG0_V2_EVENT = 0xF,
+};
+
+#define HCLGE_PTP_MSG1_V2_DEFAULT	1
+
+struct hclge_ptp_cfg_cmd {
+#define HCLGE_PTP_EN_B			BIT(0)
+#define HCLGE_PTP_TX_EN_B		BIT(1)
+#define HCLGE_PTP_RX_EN_B		BIT(2)
+#define HCLGE_PTP_UDP_EN_SHIFT		3
+#define HCLGE_PTP_UDP_EN_MASK		GENMASK(4, 3)
+#define HCLGE_PTP_MSG_TYPE_SHIFT	8
+#define HCLGE_PTP_MSG_TYPE_MASK		GENMASK(9, 8)
+#define HCLGE_PTP_MSG1_SHIFT		16
+#define HCLGE_PTP_MSG1_MASK		GENMASK(19, 16)
+#define HCLGE_PTP_MSG0_SHIFT		24
+#define HCLGE_PTP_MSG0_MASK		GENMASK(27, 24)
+
+	__le32 cfg;
+	u8 rsvd[20];
+};
+
+static inline struct hclge_dev *hclge_ptp_get_hdev(struct ptp_clock_info *info)
+{
+	struct hclge_ptp *ptp = container_of(info, struct hclge_ptp, info);
+
+	return ptp->hdev;
+}
+
+bool hclge_ptp_set_tx_info(struct hnae3_handle *handle, struct sk_buff *skb);
+void hclge_ptp_clean_tx_hwts(struct hclge_dev *hdev);
+void hclge_ptp_get_rx_hwts(struct hnae3_handle *handle, struct sk_buff *skb,
+			   u32 nsec, u32 sec);
+int hclge_ptp_get_cfg(struct hclge_dev *hdev, struct ifreq *ifr);
+int hclge_ptp_set_cfg(struct hclge_dev *hdev, struct ifreq *ifr);
+int hclge_ptp_init(struct hclge_dev *hdev);
+void hclge_ptp_uninit(struct hclge_dev *hdev);
+int hclge_ptp_get_ts_info(struct hnae3_handle *handle,
+			  struct kernel_ethtool_ts_info *info);
+int hclge_ptp_cfg_qry(struct hclge_dev *hdev, u32 *cfg);
+#endif
diff --git a/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_regs.c b/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_regs.c
new file mode 100644
index 000000000000..8c057192aae6
--- /dev/null
+++ b/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_regs.c
@@ -0,0 +1,669 @@
+// SPDX-License-Identifier: GPL-2.0+
+// Copyright (c) 2023 Hisilicon Limited.
+
+#include "hclge_cmd.h"
+#include "hclge_main.h"
+#include "hclge_regs.h"
+#include "hnae3.h"
+
+static const u32 cmdq_reg_addr_list[] = {HCLGE_COMM_NIC_CSQ_BASEADDR_L_REG,
+					 HCLGE_COMM_NIC_CSQ_BASEADDR_H_REG,
+					 HCLGE_COMM_NIC_CSQ_DEPTH_REG,
+					 HCLGE_COMM_NIC_CSQ_TAIL_REG,
+					 HCLGE_COMM_NIC_CSQ_HEAD_REG,
+					 HCLGE_COMM_NIC_CRQ_BASEADDR_L_REG,
+					 HCLGE_COMM_NIC_CRQ_BASEADDR_H_REG,
+					 HCLGE_COMM_NIC_CRQ_DEPTH_REG,
+					 HCLGE_COMM_NIC_CRQ_TAIL_REG,
+					 HCLGE_COMM_NIC_CRQ_HEAD_REG,
+					 HCLGE_COMM_VECTOR0_CMDQ_SRC_REG,
+					 HCLGE_COMM_CMDQ_INTR_STS_REG,
+					 HCLGE_COMM_CMDQ_INTR_EN_REG,
+					 HCLGE_COMM_CMDQ_INTR_GEN_REG};
+
+static const u32 common_reg_addr_list[] = {HCLGE_MISC_VECTOR_REG_BASE,
+					   HCLGE_PF_OTHER_INT_REG,
+					   HCLGE_MISC_RESET_STS_REG,
+					   HCLGE_MISC_VECTOR_INT_STS,
+					   HCLGE_GLOBAL_RESET_REG,
+					   HCLGE_FUN_RST_ING,
+					   HCLGE_GRO_EN_REG};
+
+static const u32 ring_reg_addr_list[] = {HCLGE_RING_RX_ADDR_L_REG,
+					 HCLGE_RING_RX_ADDR_H_REG,
+					 HCLGE_RING_RX_BD_NUM_REG,
+					 HCLGE_RING_RX_BD_LENGTH_REG,
+					 HCLGE_RING_RX_MERGE_EN_REG,
+					 HCLGE_RING_RX_TAIL_REG,
+					 HCLGE_RING_RX_HEAD_REG,
+					 HCLGE_RING_RX_FBD_NUM_REG,
+					 HCLGE_RING_RX_OFFSET_REG,
+					 HCLGE_RING_RX_FBD_OFFSET_REG,
+					 HCLGE_RING_RX_STASH_REG,
+					 HCLGE_RING_RX_BD_ERR_REG,
+					 HCLGE_RING_TX_ADDR_L_REG,
+					 HCLGE_RING_TX_ADDR_H_REG,
+					 HCLGE_RING_TX_BD_NUM_REG,
+					 HCLGE_RING_TX_PRIORITY_REG,
+					 HCLGE_RING_TX_TC_REG,
+					 HCLGE_RING_TX_MERGE_EN_REG,
+					 HCLGE_RING_TX_TAIL_REG,
+					 HCLGE_RING_TX_HEAD_REG,
+					 HCLGE_RING_TX_FBD_NUM_REG,
+					 HCLGE_RING_TX_OFFSET_REG,
+					 HCLGE_RING_TX_EBD_NUM_REG,
+					 HCLGE_RING_TX_EBD_OFFSET_REG,
+					 HCLGE_RING_TX_BD_ERR_REG,
+					 HCLGE_RING_EN_REG};
+
+static const u32 tqp_intr_reg_addr_list[] = {HCLGE_TQP_INTR_CTRL_REG,
+					     HCLGE_TQP_INTR_GL0_REG,
+					     HCLGE_TQP_INTR_GL1_REG,
+					     HCLGE_TQP_INTR_GL2_REG,
+					     HCLGE_TQP_INTR_RL_REG};
+
+/* Get DFX BD number offset */
+#define HCLGE_DFX_BIOS_BD_OFFSET        1
+#define HCLGE_DFX_SSU_0_BD_OFFSET       2
+#define HCLGE_DFX_SSU_1_BD_OFFSET       3
+#define HCLGE_DFX_IGU_BD_OFFSET         4
+#define HCLGE_DFX_RPU_0_BD_OFFSET       5
+#define HCLGE_DFX_RPU_1_BD_OFFSET       6
+#define HCLGE_DFX_NCSI_BD_OFFSET        7
+#define HCLGE_DFX_RTC_BD_OFFSET         8
+#define HCLGE_DFX_PPP_BD_OFFSET         9
+#define HCLGE_DFX_RCB_BD_OFFSET         10
+#define HCLGE_DFX_TQP_BD_OFFSET         11
+#define HCLGE_DFX_SSU_2_BD_OFFSET       12
+
+static const u32 hclge_dfx_bd_offset_list[] = {
+	HCLGE_DFX_BIOS_BD_OFFSET,
+	HCLGE_DFX_SSU_0_BD_OFFSET,
+	HCLGE_DFX_SSU_1_BD_OFFSET,
+	HCLGE_DFX_IGU_BD_OFFSET,
+	HCLGE_DFX_RPU_0_BD_OFFSET,
+	HCLGE_DFX_RPU_1_BD_OFFSET,
+	HCLGE_DFX_NCSI_BD_OFFSET,
+	HCLGE_DFX_RTC_BD_OFFSET,
+	HCLGE_DFX_PPP_BD_OFFSET,
+	HCLGE_DFX_RCB_BD_OFFSET,
+	HCLGE_DFX_TQP_BD_OFFSET,
+	HCLGE_DFX_SSU_2_BD_OFFSET
+};
+
+static const enum hclge_opcode_type hclge_dfx_reg_opcode_list[] = {
+	HCLGE_OPC_DFX_BIOS_COMMON_REG,
+	HCLGE_OPC_DFX_SSU_REG_0,
+	HCLGE_OPC_DFX_SSU_REG_1,
+	HCLGE_OPC_DFX_IGU_EGU_REG,
+	HCLGE_OPC_DFX_RPU_REG_0,
+	HCLGE_OPC_DFX_RPU_REG_1,
+	HCLGE_OPC_DFX_NCSI_REG,
+	HCLGE_OPC_DFX_RTC_REG,
+	HCLGE_OPC_DFX_PPP_REG,
+	HCLGE_OPC_DFX_RCB_REG,
+	HCLGE_OPC_DFX_TQP_REG,
+	HCLGE_OPC_DFX_SSU_REG_2
+};
+
+enum hclge_reg_tag {
+	HCLGE_REG_TAG_CMDQ = 0,
+	HCLGE_REG_TAG_COMMON,
+	HCLGE_REG_TAG_RING,
+	HCLGE_REG_TAG_TQP_INTR,
+	HCLGE_REG_TAG_QUERY_32_BIT,
+	HCLGE_REG_TAG_QUERY_64_BIT,
+	HCLGE_REG_TAG_DFX_BIOS_COMMON,
+	HCLGE_REG_TAG_DFX_SSU_0,
+	HCLGE_REG_TAG_DFX_SSU_1,
+	HCLGE_REG_TAG_DFX_IGU_EGU,
+	HCLGE_REG_TAG_DFX_RPU_0,
+	HCLGE_REG_TAG_DFX_RPU_1,
+	HCLGE_REG_TAG_DFX_NCSI,
+	HCLGE_REG_TAG_DFX_RTC,
+	HCLGE_REG_TAG_DFX_PPP,
+	HCLGE_REG_TAG_DFX_RCB,
+	HCLGE_REG_TAG_DFX_TQP,
+	HCLGE_REG_TAG_DFX_SSU_2,
+	HCLGE_REG_TAG_RPU_TNL,
+};
+
+#pragma pack(4)
+struct hclge_reg_tlv {
+	u16 tag;
+	u16 len;
+};
+
+struct hclge_reg_header {
+	u64 magic_number;
+	u8 is_vf;
+	u8 rsv[7];
+};
+
+#pragma pack()
+
+#define HCLGE_REG_TLV_SIZE	sizeof(struct hclge_reg_tlv)
+#define HCLGE_REG_HEADER_SIZE	sizeof(struct hclge_reg_header)
+#define HCLGE_REG_TLV_SPACE	(sizeof(struct hclge_reg_tlv) / sizeof(u32))
+#define HCLGE_REG_HEADER_SPACE	(sizeof(struct hclge_reg_header) / sizeof(u32))
+#define HCLGE_REG_MAGIC_NUMBER	0x686e733372656773 /* meaning is hns3regs */
+
+#define HCLGE_REG_RPU_TNL_ID_0	1
+
+static u32 hclge_reg_get_header(void *data)
+{
+	struct hclge_reg_header *header = data;
+
+	header->magic_number = HCLGE_REG_MAGIC_NUMBER;
+	header->is_vf = 0x0;
+
+	return HCLGE_REG_HEADER_SPACE;
+}
+
+static u32 hclge_reg_get_tlv(u32 tag, u32 regs_num, void *data)
+{
+	struct hclge_reg_tlv *tlv = data;
+
+	tlv->tag = tag;
+	tlv->len = regs_num * sizeof(u32) + HCLGE_REG_TLV_SIZE;
+
+	return HCLGE_REG_TLV_SPACE;
+}
+
+static int hclge_get_32_bit_regs(struct hclge_dev *hdev, u32 regs_num,
+				 void *data)
+{
+#define HCLGE_32_BIT_REG_RTN_DATANUM 8
+#define HCLGE_32_BIT_DESC_NODATA_LEN 2
+
+	struct hclge_desc *desc;
+	u32 *reg_val = data;
+	__le32 *desc_data;
+	int nodata_num;
+	int cmd_num;
+	int i, k, n;
+	int ret;
+
+	if (regs_num == 0)
+		return 0;
+
+	nodata_num = HCLGE_32_BIT_DESC_NODATA_LEN;
+	cmd_num = DIV_ROUND_UP(regs_num + nodata_num,
+			       HCLGE_32_BIT_REG_RTN_DATANUM);
+	desc = kcalloc(cmd_num, sizeof(struct hclge_desc), GFP_KERNEL);
+	if (!desc)
+		return -ENOMEM;
+
+	hclge_cmd_setup_basic_desc(&desc[0], HCLGE_OPC_QUERY_32_BIT_REG, true);
+	ret = hclge_cmd_send(&hdev->hw, desc, cmd_num);
+	if (ret) {
+		dev_err(&hdev->pdev->dev,
+			"Query 32 bit register cmd failed, ret = %d.\n", ret);
+		kfree(desc);
+		return ret;
+	}
+
+	for (i = 0; i < cmd_num; i++) {
+		if (i == 0) {
+			desc_data = (__le32 *)(&desc[i].data[0]);
+			n = HCLGE_32_BIT_REG_RTN_DATANUM - nodata_num;
+		} else {
+			desc_data = (__le32 *)(&desc[i]);
+			n = HCLGE_32_BIT_REG_RTN_DATANUM;
+		}
+		for (k = 0; k < n; k++) {
+			*reg_val++ = le32_to_cpu(*desc_data++);
+
+			regs_num--;
+			if (!regs_num)
+				break;
+		}
+	}
+
+	kfree(desc);
+	return 0;
+}
+
+static int hclge_get_64_bit_regs(struct hclge_dev *hdev, u32 regs_num,
+				 void *data)
+{
+#define HCLGE_64_BIT_REG_RTN_DATANUM 4
+#define HCLGE_64_BIT_DESC_NODATA_LEN 1
+
+	struct hclge_desc *desc;
+	u64 *reg_val = data;
+	__le64 *desc_data;
+	int nodata_len;
+	int cmd_num;
+	int i, k, n;
+	int ret;
+
+	if (regs_num == 0)
+		return 0;
+
+	nodata_len = HCLGE_64_BIT_DESC_NODATA_LEN;
+	cmd_num = DIV_ROUND_UP(regs_num + nodata_len,
+			       HCLGE_64_BIT_REG_RTN_DATANUM);
+	desc = kcalloc(cmd_num, sizeof(struct hclge_desc), GFP_KERNEL);
+	if (!desc)
+		return -ENOMEM;
+
+	hclge_cmd_setup_basic_desc(&desc[0], HCLGE_OPC_QUERY_64_BIT_REG, true);
+	ret = hclge_cmd_send(&hdev->hw, desc, cmd_num);
+	if (ret) {
+		dev_err(&hdev->pdev->dev,
+			"Query 64 bit register cmd failed, ret = %d.\n", ret);
+		kfree(desc);
+		return ret;
+	}
+
+	for (i = 0; i < cmd_num; i++) {
+		if (i == 0) {
+			desc_data = (__le64 *)(&desc[i].data[0]);
+			n = HCLGE_64_BIT_REG_RTN_DATANUM - nodata_len;
+		} else {
+			desc_data = (__le64 *)(&desc[i]);
+			n = HCLGE_64_BIT_REG_RTN_DATANUM;
+		}
+		for (k = 0; k < n; k++) {
+			*reg_val++ = le64_to_cpu(*desc_data++);
+
+			regs_num--;
+			if (!regs_num)
+				break;
+		}
+	}
+
+	kfree(desc);
+	return 0;
+}
+
+int hclge_query_bd_num_cmd_send(struct hclge_dev *hdev, struct hclge_desc *desc)
+{
+	int i;
+
+	/* initialize command BD except the last one */
+	for (i = 0; i < HCLGE_GET_DFX_REG_TYPE_CNT - 1; i++) {
+		hclge_cmd_setup_basic_desc(&desc[i], HCLGE_OPC_DFX_BD_NUM,
+					   true);
+		desc[i].flag |= cpu_to_le16(HCLGE_COMM_CMD_FLAG_NEXT);
+	}
+
+	/* initialize the last command BD */
+	hclge_cmd_setup_basic_desc(&desc[i], HCLGE_OPC_DFX_BD_NUM, true);
+
+	return hclge_cmd_send(&hdev->hw, desc, HCLGE_GET_DFX_REG_TYPE_CNT);
+}
+
+static int hclge_get_dfx_reg_bd_num(struct hclge_dev *hdev,
+				    int *bd_num_list,
+				    u32 type_num)
+{
+	u32 entries_per_desc, desc_index, index, offset, i;
+	struct hclge_desc desc[HCLGE_GET_DFX_REG_TYPE_CNT];
+	int ret;
+
+	ret = hclge_query_bd_num_cmd_send(hdev, desc);
+	if (ret) {
+		dev_err(&hdev->pdev->dev,
+			"Get dfx bd num fail, status is %d.\n", ret);
+		return ret;
+	}
+
+	entries_per_desc = ARRAY_SIZE(desc[0].data);
+	for (i = 0; i < type_num; i++) {
+		offset = hclge_dfx_bd_offset_list[i];
+		index = offset % entries_per_desc;
+		desc_index = offset / entries_per_desc;
+		bd_num_list[i] = le32_to_cpu(desc[desc_index].data[index]);
+	}
+
+	return ret;
+}
+
+static int hclge_dfx_reg_cmd_send(struct hclge_dev *hdev,
+				  struct hclge_desc *desc_src, int bd_num,
+				  enum hclge_opcode_type cmd)
+{
+	struct hclge_desc *desc = desc_src;
+	int i, ret;
+
+	hclge_cmd_setup_basic_desc(desc, cmd, true);
+	for (i = 0; i < bd_num - 1; i++) {
+		desc->flag |= cpu_to_le16(HCLGE_COMM_CMD_FLAG_NEXT);
+		desc++;
+		hclge_cmd_setup_basic_desc(desc, cmd, true);
+	}
+
+	desc = desc_src;
+	ret = hclge_cmd_send(&hdev->hw, desc, bd_num);
+	if (ret)
+		dev_err(&hdev->pdev->dev,
+			"Query dfx reg cmd(0x%x) send fail, status is %d.\n",
+			cmd, ret);
+
+	return ret;
+}
+
+/* tnl_id = 0 means get sum of all tnl reg's value */
+static int hclge_dfx_reg_rpu_tnl_cmd_send(struct hclge_dev *hdev, u32 tnl_id,
+					  struct hclge_desc *desc, int bd_num)
+{
+	int i, ret;
+
+	for (i = 0; i < bd_num; i++) {
+		hclge_cmd_setup_basic_desc(&desc[i], HCLGE_OPC_DFX_RPU_REG_0,
+					   true);
+		if (i != bd_num - 1)
+			desc[i].flag |= cpu_to_le16(HCLGE_COMM_CMD_FLAG_NEXT);
+	}
+
+	desc[0].data[0] = cpu_to_le32(tnl_id);
+	ret = hclge_cmd_send(&hdev->hw, desc, bd_num);
+	if (ret)
+		dev_err(&hdev->pdev->dev,
+			"failed to query dfx rpu tnl reg, ret = %d\n",
+			ret);
+	return ret;
+}
+
+static int hclge_dfx_reg_fetch_data(struct hclge_desc *desc_src, int bd_num,
+				    void *data)
+{
+	int entries_per_desc, reg_num, desc_index, index, i;
+	struct hclge_desc *desc = desc_src;
+	u32 *reg = data;
+
+	entries_per_desc = ARRAY_SIZE(desc->data);
+	reg_num = entries_per_desc * bd_num;
+	for (i = 0; i < reg_num; i++) {
+		index = i % entries_per_desc;
+		desc_index = i / entries_per_desc;
+		*reg++ = le32_to_cpu(desc[desc_index].data[index]);
+	}
+
+	return reg_num;
+}
+
+static int hclge_get_dfx_reg_len(struct hclge_dev *hdev, int *len)
+{
+	u32 dfx_reg_type_num = ARRAY_SIZE(hclge_dfx_bd_offset_list);
+	struct hnae3_ae_dev *ae_dev = pci_get_drvdata(hdev->pdev);
+	int data_len_per_desc;
+	int *bd_num_list;
+	int ret;
+	u32 i;
+
+	bd_num_list = kcalloc(dfx_reg_type_num, sizeof(int), GFP_KERNEL);
+	if (!bd_num_list)
+		return -ENOMEM;
+
+	ret = hclge_get_dfx_reg_bd_num(hdev, bd_num_list, dfx_reg_type_num);
+	if (ret) {
+		dev_err(&hdev->pdev->dev,
+			"Get dfx reg bd num fail, status is %d.\n", ret);
+		goto out;
+	}
+
+	data_len_per_desc = sizeof_field(struct hclge_desc, data);
+	*len = 0;
+	for (i = 0; i < dfx_reg_type_num; i++)
+		*len += bd_num_list[i] * data_len_per_desc + HCLGE_REG_TLV_SIZE;
+
+	/**
+	 * the num of dfx_rpu_0 is reused by each dfx_rpu_tnl
+	 * HCLGE_DFX_BD_OFFSET is starting at 1, but the array subscript is
+	 * starting at 0, so offset need '- 1'.
+	 */
+	*len += (bd_num_list[HCLGE_DFX_RPU_0_BD_OFFSET - 1] * data_len_per_desc +
+		 HCLGE_REG_TLV_SIZE) * ae_dev->dev_specs.tnl_num;
+
+out:
+	kfree(bd_num_list);
+	return ret;
+}
+
+static int hclge_get_dfx_rpu_tnl_reg(struct hclge_dev *hdev, u32 *reg,
+				     struct hclge_desc *desc_src,
+				     int bd_num)
+{
+	struct hnae3_ae_dev *ae_dev = pci_get_drvdata(hdev->pdev);
+	int ret = 0;
+	u8 i;
+
+	for (i = HCLGE_REG_RPU_TNL_ID_0; i <= ae_dev->dev_specs.tnl_num; i++) {
+		ret = hclge_dfx_reg_rpu_tnl_cmd_send(hdev, i, desc_src, bd_num);
+		if (ret)
+			break;
+
+		reg += hclge_reg_get_tlv(HCLGE_REG_TAG_RPU_TNL,
+					 ARRAY_SIZE(desc_src->data) * bd_num,
+					 reg);
+		reg += hclge_dfx_reg_fetch_data(desc_src, bd_num, reg);
+	}
+
+	return ret;
+}
+
+static int hclge_get_dfx_reg(struct hclge_dev *hdev, void *data)
+{
+	u32 dfx_reg_type_num = ARRAY_SIZE(hclge_dfx_bd_offset_list);
+	int bd_num, bd_num_max, buf_len;
+	struct hclge_desc *desc_src;
+	int *bd_num_list;
+	u32 *reg = data;
+	int ret;
+	u32 i;
+
+	bd_num_list = kcalloc(dfx_reg_type_num, sizeof(int), GFP_KERNEL);
+	if (!bd_num_list)
+		return -ENOMEM;
+
+	ret = hclge_get_dfx_reg_bd_num(hdev, bd_num_list, dfx_reg_type_num);
+	if (ret) {
+		dev_err(&hdev->pdev->dev,
+			"Get dfx reg bd num fail, status is %d.\n", ret);
+		goto out;
+	}
+
+	bd_num_max = bd_num_list[0];
+	for (i = 1; i < dfx_reg_type_num; i++)
+		bd_num_max = max_t(int, bd_num_max, bd_num_list[i]);
+
+	buf_len = sizeof(*desc_src) * bd_num_max;
+	desc_src = kzalloc(buf_len, GFP_KERNEL);
+	if (!desc_src) {
+		ret = -ENOMEM;
+		goto out;
+	}
+
+	for (i = 0; i < dfx_reg_type_num; i++) {
+		bd_num = bd_num_list[i];
+		ret = hclge_dfx_reg_cmd_send(hdev, desc_src, bd_num,
+					     hclge_dfx_reg_opcode_list[i]);
+		if (ret) {
+			dev_err(&hdev->pdev->dev,
+				"Get dfx reg fail, status is %d.\n", ret);
+			goto free;
+		}
+
+		reg += hclge_reg_get_tlv(HCLGE_REG_TAG_DFX_BIOS_COMMON + i,
+					 ARRAY_SIZE(desc_src->data) * bd_num,
+					 reg);
+		reg += hclge_dfx_reg_fetch_data(desc_src, bd_num, reg);
+	}
+
+	/**
+	 * HCLGE_DFX_BD_OFFSET is starting at 1, but the array subscript is
+	 * starting at 0, so offset need '- 1'.
+	 */
+	bd_num = bd_num_list[HCLGE_DFX_RPU_0_BD_OFFSET - 1];
+	ret = hclge_get_dfx_rpu_tnl_reg(hdev, reg, desc_src, bd_num);
+
+free:
+	kfree(desc_src);
+out:
+	kfree(bd_num_list);
+	return ret;
+}
+
+static int hclge_fetch_pf_reg(struct hclge_dev *hdev, void *data,
+			      struct hnae3_knic_private_info *kinfo)
+{
+#define HCLGE_RING_INT_REG_OFFSET	0x4
+
+	struct hnae3_queue *tqp;
+	int i, j, reg_num;
+	int data_num_sum;
+	u32 *reg = data;
+
+	/* fetching per-PF registers valus from PF PCIe register space */
+	reg_num = ARRAY_SIZE(cmdq_reg_addr_list);
+	reg += hclge_reg_get_tlv(HCLGE_REG_TAG_CMDQ, reg_num, reg);
+	for (i = 0; i < reg_num; i++)
+		*reg++ = hclge_read_dev(&hdev->hw, cmdq_reg_addr_list[i]);
+	data_num_sum = reg_num + HCLGE_REG_TLV_SPACE;
+
+	reg_num = ARRAY_SIZE(common_reg_addr_list);
+	reg += hclge_reg_get_tlv(HCLGE_REG_TAG_COMMON, reg_num, reg);
+	for (i = 0; i < reg_num; i++)
+		*reg++ = hclge_read_dev(&hdev->hw, common_reg_addr_list[i]);
+	data_num_sum += reg_num + HCLGE_REG_TLV_SPACE;
+
+	reg_num = ARRAY_SIZE(ring_reg_addr_list);
+	for (j = 0; j < kinfo->num_tqps; j++) {
+		reg += hclge_reg_get_tlv(HCLGE_REG_TAG_RING, reg_num, reg);
+		tqp = kinfo->tqp[j];
+		for (i = 0; i < reg_num; i++)
+			*reg++ = readl_relaxed(tqp->io_base -
+					       HCLGE_TQP_REG_OFFSET +
+					       ring_reg_addr_list[i]);
+	}
+	data_num_sum += (reg_num + HCLGE_REG_TLV_SPACE) * kinfo->num_tqps;
+
+	reg_num = ARRAY_SIZE(tqp_intr_reg_addr_list);
+	for (j = 0; j < hdev->num_msi_used - 1; j++) {
+		reg += hclge_reg_get_tlv(HCLGE_REG_TAG_TQP_INTR, reg_num, reg);
+		for (i = 0; i < reg_num; i++)
+			*reg++ = hclge_read_dev(&hdev->hw,
+						tqp_intr_reg_addr_list[i] +
+						HCLGE_RING_INT_REG_OFFSET * j);
+	}
+	data_num_sum += (reg_num + HCLGE_REG_TLV_SPACE) *
+			(hdev->num_msi_used - 1);
+
+	return data_num_sum;
+}
+
+static int hclge_get_regs_num(struct hclge_dev *hdev, u32 *regs_num_32_bit,
+			      u32 *regs_num_64_bit)
+{
+	struct hclge_desc desc;
+	u32 total_num;
+	int ret;
+
+	hclge_cmd_setup_basic_desc(&desc, HCLGE_OPC_QUERY_REG_NUM, true);
+	ret = hclge_cmd_send(&hdev->hw, &desc, 1);
+	if (ret) {
+		dev_err(&hdev->pdev->dev,
+			"Query register number cmd failed, ret = %d.\n", ret);
+		return ret;
+	}
+
+	*regs_num_32_bit = le32_to_cpu(desc.data[0]);
+	*regs_num_64_bit = le32_to_cpu(desc.data[1]);
+
+	total_num = *regs_num_32_bit + *regs_num_64_bit;
+	if (!total_num)
+		return -EINVAL;
+
+	return 0;
+}
+
+int hclge_get_regs_len(struct hnae3_handle *handle)
+{
+	struct hnae3_knic_private_info *kinfo = &handle->kinfo;
+	struct hclge_vport *vport = hclge_get_vport(handle);
+	int regs_num_32_bit, regs_num_64_bit, dfx_regs_len;
+	int cmdq_len, common_len, ring_len, tqp_intr_len;
+	int regs_len_32_bit, regs_len_64_bit;
+	struct hclge_dev *hdev = vport->back;
+	int ret;
+
+	ret = hclge_get_regs_num(hdev, &regs_num_32_bit, &regs_num_64_bit);
+	if (ret) {
+		dev_err(&hdev->pdev->dev,
+			"Get register number failed, ret = %d.\n", ret);
+		return ret;
+	}
+
+	ret = hclge_get_dfx_reg_len(hdev, &dfx_regs_len);
+	if (ret) {
+		dev_err(&hdev->pdev->dev,
+			"Get dfx reg len failed, ret = %d.\n", ret);
+		return ret;
+	}
+
+	cmdq_len = HCLGE_REG_TLV_SIZE + sizeof(cmdq_reg_addr_list);
+	common_len = HCLGE_REG_TLV_SIZE + sizeof(common_reg_addr_list);
+	ring_len = HCLGE_REG_TLV_SIZE + sizeof(ring_reg_addr_list);
+	tqp_intr_len = HCLGE_REG_TLV_SIZE + sizeof(tqp_intr_reg_addr_list);
+	regs_len_32_bit = HCLGE_REG_TLV_SIZE + regs_num_32_bit * sizeof(u32);
+	regs_len_64_bit = HCLGE_REG_TLV_SIZE + regs_num_64_bit * sizeof(u64);
+
+	/* return the total length of all register values */
+	return HCLGE_REG_HEADER_SIZE + cmdq_len + common_len + ring_len *
+		kinfo->num_tqps + tqp_intr_len * (hdev->num_msi_used - 1) +
+		regs_len_32_bit + regs_len_64_bit + dfx_regs_len;
+}
+
+void hclge_get_regs(struct hnae3_handle *handle, u32 *version,
+		    void *data)
+{
+#define HCLGE_REG_64_BIT_SPACE_MULTIPLE		2
+
+	struct hnae3_knic_private_info *kinfo = &handle->kinfo;
+	struct hclge_vport *vport = hclge_get_vport(handle);
+	struct hclge_dev *hdev = vport->back;
+	u32 regs_num_32_bit, regs_num_64_bit;
+	u32 *reg = data;
+	int ret;
+
+	*version = hdev->fw_version;
+
+	ret = hclge_get_regs_num(hdev, &regs_num_32_bit, &regs_num_64_bit);
+	if (ret) {
+		dev_err(&hdev->pdev->dev,
+			"Get register number failed, ret = %d.\n", ret);
+		return;
+	}
+
+	reg += hclge_reg_get_header(reg);
+	reg += hclge_fetch_pf_reg(hdev, reg, kinfo);
+
+	reg += hclge_reg_get_tlv(HCLGE_REG_TAG_QUERY_32_BIT,
+				 regs_num_32_bit, reg);
+	ret = hclge_get_32_bit_regs(hdev, regs_num_32_bit, reg);
+	if (ret) {
+		dev_err(&hdev->pdev->dev,
+			"Get 32 bit register failed, ret = %d.\n", ret);
+		return;
+	}
+	reg += regs_num_32_bit;
+
+	reg += hclge_reg_get_tlv(HCLGE_REG_TAG_QUERY_64_BIT,
+				 regs_num_64_bit *
+				 HCLGE_REG_64_BIT_SPACE_MULTIPLE, reg);
+	ret = hclge_get_64_bit_regs(hdev, regs_num_64_bit, reg);
+	if (ret) {
+		dev_err(&hdev->pdev->dev,
+			"Get 64 bit register failed, ret = %d.\n", ret);
+		return;
+	}
+	reg += regs_num_64_bit * HCLGE_REG_64_BIT_SPACE_MULTIPLE;
+
+	ret = hclge_get_dfx_reg(hdev, reg);
+	if (ret)
+		dev_err(&hdev->pdev->dev,
+			"Get dfx register failed, ret = %d.\n", ret);
+}
diff --git a/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_regs.h b/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_regs.h
new file mode 100644
index 000000000000..b6bc1ecb8054
--- /dev/null
+++ b/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_regs.h
@@ -0,0 +1,17 @@
+/* SPDX-License-Identifier: GPL-2.0+ */
+// Copyright (c) 2023 Hisilicon Limited.
+
+#ifndef __HCLGE_REGS_H
+#define __HCLGE_REGS_H
+#include <linux/types.h>
+#include "hclge_comm_cmd.h"
+
+struct hnae3_handle;
+struct hclge_dev;
+
+int hclge_query_bd_num_cmd_send(struct hclge_dev *hdev,
+				struct hclge_desc *desc);
+int hclge_get_regs_len(struct hnae3_handle *handle);
+void hclge_get_regs(struct hnae3_handle *handle, u32 *version,
+		    void *data);
+#endif
diff --git a/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_tm.c b/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_tm.c
new file mode 100644
index 000000000000..00c3f2548bf6
--- /dev/null
+++ b/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_tm.c
@@ -0,0 +1,2161 @@
+// SPDX-License-Identifier: GPL-2.0+
+// Copyright (c) 2016-2017 Hisilicon Limited.
+
+#include <linux/etherdevice.h>
+
+#include "hclge_cmd.h"
+#include "hclge_main.h"
+#include "hclge_tm.h"
+
+enum hclge_shaper_level {
+	HCLGE_SHAPER_LVL_PRI	= 0,
+	HCLGE_SHAPER_LVL_PG	= 1,
+	HCLGE_SHAPER_LVL_PORT	= 2,
+	HCLGE_SHAPER_LVL_QSET	= 3,
+	HCLGE_SHAPER_LVL_CNT	= 4,
+	HCLGE_SHAPER_LVL_VF	= 0,
+	HCLGE_SHAPER_LVL_PF	= 1,
+};
+
+#define HCLGE_TM_PFC_PKT_GET_CMD_NUM	3
+#define HCLGE_TM_PFC_NUM_GET_PER_CMD	3
+
+#define HCLGE_SHAPER_BS_U_DEF	5
+#define HCLGE_SHAPER_BS_S_DEF	20
+
+/* hclge_shaper_para_calc: calculate ir parameter for the shaper
+ * @ir: Rate to be config, its unit is Mbps
+ * @shaper_level: the shaper level. eg: port, pg, priority, queueset
+ * @ir_para: parameters of IR shaper
+ * @max_tm_rate: max tm rate is available to config
+ *
+ * the formula:
+ *
+ *		IR_b * (2 ^ IR_u) * 8
+ * IR(Mbps) = -------------------------  *  CLOCK(1000Mbps)
+ *		Tick * (2 ^ IR_s)
+ *
+ * @return: 0: calculate sucessful, negative: fail
+ */
+static int hclge_shaper_para_calc(u32 ir, u8 shaper_level,
+				  struct hclge_shaper_ir_para *ir_para,
+				  u32 max_tm_rate)
+{
+#define DEFAULT_SHAPER_IR_B	126
+#define DIVISOR_CLK		(1000 * 8)
+#define DEFAULT_DIVISOR_IR_B	(DEFAULT_SHAPER_IR_B * DIVISOR_CLK)
+
+	static const u16 tick_array[HCLGE_SHAPER_LVL_CNT] = {
+		6 * 256,        /* Prioriy level */
+		6 * 32,         /* Prioriy group level */
+		6 * 8,          /* Port level */
+		6 * 256         /* Qset level */
+	};
+	u8 ir_u_calc = 0;
+	u8 ir_s_calc = 0;
+	u32 ir_calc;
+	u32 tick;
+
+	/* Calc tick */
+	if (shaper_level >= HCLGE_SHAPER_LVL_CNT ||
+	    ir > max_tm_rate)
+		return -EINVAL;
+
+	tick = tick_array[shaper_level];
+
+	/**
+	 * Calc the speed if ir_b = 126, ir_u = 0 and ir_s = 0
+	 * the formula is changed to:
+	 *		126 * 1 * 8
+	 * ir_calc = ---------------- * 1000
+	 *		tick * 1
+	 */
+	ir_calc = (DEFAULT_DIVISOR_IR_B + (tick >> 1) - 1) / tick;
+
+	if (ir_calc == ir) {
+		ir_para->ir_b = DEFAULT_SHAPER_IR_B;
+		ir_para->ir_u = 0;
+		ir_para->ir_s = 0;
+
+		return 0;
+	} else if (ir_calc > ir) {
+		/* Increasing the denominator to select ir_s value */
+		while (ir_calc >= ir && ir) {
+			ir_s_calc++;
+			ir_calc = DEFAULT_DIVISOR_IR_B /
+				  (tick * (1 << ir_s_calc));
+		}
+
+		ir_para->ir_b = (ir * tick * (1 << ir_s_calc) +
+				(DIVISOR_CLK >> 1)) / DIVISOR_CLK;
+	} else {
+		/* Increasing the numerator to select ir_u value */
+		u32 numerator;
+
+		while (ir_calc < ir) {
+			ir_u_calc++;
+			numerator = DEFAULT_DIVISOR_IR_B * (1 << ir_u_calc);
+			ir_calc = (numerator + (tick >> 1)) / tick;
+		}
+
+		if (ir_calc == ir) {
+			ir_para->ir_b = DEFAULT_SHAPER_IR_B;
+		} else {
+			u32 denominator = DIVISOR_CLK * (1 << --ir_u_calc);
+			ir_para->ir_b = (ir * tick + (denominator >> 1)) /
+					denominator;
+		}
+	}
+
+	ir_para->ir_u = ir_u_calc;
+	ir_para->ir_s = ir_s_calc;
+
+	return 0;
+}
+
+static const u16 hclge_pfc_tx_stats_offset[] = {
+	HCLGE_MAC_STATS_FIELD_OFF(mac_tx_pfc_pri0_pkt_num),
+	HCLGE_MAC_STATS_FIELD_OFF(mac_tx_pfc_pri1_pkt_num),
+	HCLGE_MAC_STATS_FIELD_OFF(mac_tx_pfc_pri2_pkt_num),
+	HCLGE_MAC_STATS_FIELD_OFF(mac_tx_pfc_pri3_pkt_num),
+	HCLGE_MAC_STATS_FIELD_OFF(mac_tx_pfc_pri4_pkt_num),
+	HCLGE_MAC_STATS_FIELD_OFF(mac_tx_pfc_pri5_pkt_num),
+	HCLGE_MAC_STATS_FIELD_OFF(mac_tx_pfc_pri6_pkt_num),
+	HCLGE_MAC_STATS_FIELD_OFF(mac_tx_pfc_pri7_pkt_num)
+};
+
+static const u16 hclge_pfc_rx_stats_offset[] = {
+	HCLGE_MAC_STATS_FIELD_OFF(mac_rx_pfc_pri0_pkt_num),
+	HCLGE_MAC_STATS_FIELD_OFF(mac_rx_pfc_pri1_pkt_num),
+	HCLGE_MAC_STATS_FIELD_OFF(mac_rx_pfc_pri2_pkt_num),
+	HCLGE_MAC_STATS_FIELD_OFF(mac_rx_pfc_pri3_pkt_num),
+	HCLGE_MAC_STATS_FIELD_OFF(mac_rx_pfc_pri4_pkt_num),
+	HCLGE_MAC_STATS_FIELD_OFF(mac_rx_pfc_pri5_pkt_num),
+	HCLGE_MAC_STATS_FIELD_OFF(mac_rx_pfc_pri6_pkt_num),
+	HCLGE_MAC_STATS_FIELD_OFF(mac_rx_pfc_pri7_pkt_num)
+};
+
+static void hclge_pfc_stats_get(struct hclge_dev *hdev, bool tx, u64 *stats)
+{
+	const u16 *offset;
+	int i;
+
+	if (tx)
+		offset = hclge_pfc_tx_stats_offset;
+	else
+		offset = hclge_pfc_rx_stats_offset;
+
+	for (i = 0; i < HCLGE_MAX_TC_NUM; i++)
+		stats[i] = HCLGE_STATS_READ(&hdev->mac_stats, offset[i]);
+}
+
+void hclge_pfc_rx_stats_get(struct hclge_dev *hdev, u64 *stats)
+{
+	hclge_pfc_stats_get(hdev, false, stats);
+}
+
+void hclge_pfc_tx_stats_get(struct hclge_dev *hdev, u64 *stats)
+{
+	hclge_pfc_stats_get(hdev, true, stats);
+}
+
+int hclge_mac_pause_en_cfg(struct hclge_dev *hdev, bool tx, bool rx)
+{
+	struct hclge_desc desc;
+
+	hclge_cmd_setup_basic_desc(&desc, HCLGE_OPC_CFG_MAC_PAUSE_EN, false);
+
+	desc.data[0] = cpu_to_le32((tx ? HCLGE_TX_MAC_PAUSE_EN_MSK : 0) |
+		(rx ? HCLGE_RX_MAC_PAUSE_EN_MSK : 0));
+
+	return hclge_cmd_send(&hdev->hw, &desc, 1);
+}
+
+int hclge_pfc_pause_en_cfg(struct hclge_dev *hdev, u8 tx_rx_bitmap,
+			   u8 pfc_bitmap)
+{
+	struct hclge_desc desc;
+	struct hclge_pfc_en_cmd *pfc = (struct hclge_pfc_en_cmd *)desc.data;
+
+	hclge_cmd_setup_basic_desc(&desc, HCLGE_OPC_CFG_PFC_PAUSE_EN, false);
+
+	pfc->tx_rx_en_bitmap = tx_rx_bitmap;
+	pfc->pri_en_bitmap = pfc_bitmap;
+
+	return hclge_cmd_send(&hdev->hw, &desc, 1);
+}
+
+static int hclge_pause_param_cfg(struct hclge_dev *hdev, const u8 *addr,
+				 u8 pause_trans_gap, u16 pause_trans_time)
+{
+	struct hclge_cfg_pause_param_cmd *pause_param;
+	struct hclge_desc desc;
+
+	pause_param = (struct hclge_cfg_pause_param_cmd *)desc.data;
+
+	hclge_cmd_setup_basic_desc(&desc, HCLGE_OPC_CFG_MAC_PARA, false);
+
+	ether_addr_copy(pause_param->mac_addr, addr);
+	ether_addr_copy(pause_param->mac_addr_extra, addr);
+	pause_param->pause_trans_gap = pause_trans_gap;
+	pause_param->pause_trans_time = cpu_to_le16(pause_trans_time);
+
+	return hclge_cmd_send(&hdev->hw, &desc, 1);
+}
+
+int hclge_pause_addr_cfg(struct hclge_dev *hdev, const u8 *mac_addr)
+{
+	struct hclge_cfg_pause_param_cmd *pause_param;
+	struct hclge_desc desc;
+	u16 trans_time;
+	u8 trans_gap;
+	int ret;
+
+	pause_param = (struct hclge_cfg_pause_param_cmd *)desc.data;
+
+	hclge_cmd_setup_basic_desc(&desc, HCLGE_OPC_CFG_MAC_PARA, true);
+
+	ret = hclge_cmd_send(&hdev->hw, &desc, 1);
+	if (ret)
+		return ret;
+
+	trans_gap = pause_param->pause_trans_gap;
+	trans_time = le16_to_cpu(pause_param->pause_trans_time);
+
+	return hclge_pause_param_cfg(hdev, mac_addr, trans_gap, trans_time);
+}
+
+static int hclge_fill_pri_array(struct hclge_dev *hdev, u8 *pri, u8 pri_id)
+{
+	u8 tc;
+
+	tc = hdev->tm_info.prio_tc[pri_id];
+
+	if (tc >= hdev->tm_info.num_tc)
+		return -EINVAL;
+
+	/**
+	 * the register for priority has four bytes, the first bytes includes
+	 *  priority0 and priority1, the higher 4bit stands for priority1
+	 *  while the lower 4bit stands for priority0, as below:
+	 * first byte:	| pri_1 | pri_0 |
+	 * second byte:	| pri_3 | pri_2 |
+	 * third byte:	| pri_5 | pri_4 |
+	 * fourth byte:	| pri_7 | pri_6 |
+	 */
+	pri[pri_id >> 1] |= tc << ((pri_id & 1) * 4);
+
+	return 0;
+}
+
+int hclge_up_to_tc_map(struct hclge_dev *hdev)
+{
+	struct hclge_desc desc;
+	u8 *pri = (u8 *)desc.data;
+	u8 pri_id;
+	int ret;
+
+	hclge_cmd_setup_basic_desc(&desc, HCLGE_OPC_PRI_TO_TC_MAPPING, false);
+
+	for (pri_id = 0; pri_id < HNAE3_MAX_USER_PRIO; pri_id++) {
+		ret = hclge_fill_pri_array(hdev, pri, pri_id);
+		if (ret)
+			return ret;
+	}
+
+	return hclge_cmd_send(&hdev->hw, &desc, 1);
+}
+
+static void hclge_dscp_to_prio_map_init(struct hclge_dev *hdev)
+{
+	u8 i;
+
+	hdev->vport[0].nic.kinfo.tc_map_mode = HNAE3_TC_MAP_MODE_PRIO;
+	hdev->vport[0].nic.kinfo.dscp_app_cnt = 0;
+	for (i = 0; i < HNAE3_MAX_DSCP; i++)
+		hdev->vport[0].nic.kinfo.dscp_prio[i] = HNAE3_PRIO_ID_INVALID;
+}
+
+int hclge_dscp_to_tc_map(struct hclge_dev *hdev)
+{
+	struct hclge_desc desc[HCLGE_DSCP_MAP_TC_BD_NUM];
+	u8 *req0 = (u8 *)desc[0].data;
+	u8 *req1 = (u8 *)desc[1].data;
+	u8 pri_id, tc_id, i, j;
+
+	hclge_cmd_setup_basic_desc(&desc[0], HCLGE_OPC_QOS_MAP, false);
+	desc[0].flag |= cpu_to_le16(HCLGE_COMM_CMD_FLAG_NEXT);
+	hclge_cmd_setup_basic_desc(&desc[1], HCLGE_OPC_QOS_MAP, false);
+
+	/* The low 32 dscp setting use bd0, high 32 dscp setting use bd1 */
+	for (i = 0; i < HNAE3_MAX_DSCP / HCLGE_DSCP_MAP_TC_BD_NUM; i++) {
+		pri_id = hdev->vport[0].nic.kinfo.dscp_prio[i];
+		pri_id = pri_id == HNAE3_PRIO_ID_INVALID ? 0 : pri_id;
+		tc_id = hdev->tm_info.prio_tc[pri_id];
+		/* Each dscp setting has 4 bits, so each byte saves two dscp
+		 * setting
+		 */
+		req0[i >> 1] |= tc_id << HCLGE_DSCP_TC_SHIFT(i);
+
+		j = i + HNAE3_MAX_DSCP / HCLGE_DSCP_MAP_TC_BD_NUM;
+		pri_id = hdev->vport[0].nic.kinfo.dscp_prio[j];
+		pri_id = pri_id == HNAE3_PRIO_ID_INVALID ? 0 : pri_id;
+		tc_id = hdev->tm_info.prio_tc[pri_id];
+		req1[i >> 1] |= tc_id << HCLGE_DSCP_TC_SHIFT(i);
+	}
+
+	return hclge_cmd_send(&hdev->hw, desc, HCLGE_DSCP_MAP_TC_BD_NUM);
+}
+
+static int hclge_tm_pg_to_pri_map_cfg(struct hclge_dev *hdev,
+				      u8 pg_id, u8 pri_bit_map)
+{
+	struct hclge_pg_to_pri_link_cmd *map;
+	struct hclge_desc desc;
+
+	hclge_cmd_setup_basic_desc(&desc, HCLGE_OPC_TM_PG_TO_PRI_LINK, false);
+
+	map = (struct hclge_pg_to_pri_link_cmd *)desc.data;
+
+	map->pg_id = pg_id;
+	map->pri_bit_map = pri_bit_map;
+
+	return hclge_cmd_send(&hdev->hw, &desc, 1);
+}
+
+static int hclge_tm_qs_to_pri_map_cfg(struct hclge_dev *hdev, u16 qs_id, u8 pri,
+				      bool link_vld)
+{
+	struct hclge_qs_to_pri_link_cmd *map;
+	struct hclge_desc desc;
+
+	hclge_cmd_setup_basic_desc(&desc, HCLGE_OPC_TM_QS_TO_PRI_LINK, false);
+
+	map = (struct hclge_qs_to_pri_link_cmd *)desc.data;
+
+	map->qs_id = cpu_to_le16(qs_id);
+	map->priority = pri;
+	map->link_vld = link_vld ? HCLGE_TM_QS_PRI_LINK_VLD_MSK : 0;
+
+	return hclge_cmd_send(&hdev->hw, &desc, 1);
+}
+
+static int hclge_tm_q_to_qs_map_cfg(struct hclge_dev *hdev,
+				    u16 q_id, u16 qs_id)
+{
+	struct hclge_nq_to_qs_link_cmd *map;
+	struct hclge_desc desc;
+	u16 qs_id_l;
+	u16 qs_id_h;
+
+	hclge_cmd_setup_basic_desc(&desc, HCLGE_OPC_TM_NQ_TO_QS_LINK, false);
+
+	map = (struct hclge_nq_to_qs_link_cmd *)desc.data;
+
+	map->nq_id = cpu_to_le16(q_id);
+
+	/* convert qs_id to the following format to support qset_id >= 1024
+	 * qs_id: | 15 | 14 ~ 10 |  9 ~ 0   |
+	 *            /         / \         \
+	 *           /         /   \         \
+	 * qset_id: | 15 ~ 11 |  10 |  9 ~ 0  |
+	 *          | qs_id_h | vld | qs_id_l |
+	 */
+	qs_id_l = hnae3_get_field(qs_id, HCLGE_TM_QS_ID_L_MSK,
+				  HCLGE_TM_QS_ID_L_S);
+	qs_id_h = hnae3_get_field(qs_id, HCLGE_TM_QS_ID_H_MSK,
+				  HCLGE_TM_QS_ID_H_S);
+	hnae3_set_field(qs_id, HCLGE_TM_QS_ID_L_MSK, HCLGE_TM_QS_ID_L_S,
+			qs_id_l);
+	hnae3_set_field(qs_id, HCLGE_TM_QS_ID_H_EXT_MSK, HCLGE_TM_QS_ID_H_EXT_S,
+			qs_id_h);
+	map->qset_id = cpu_to_le16(qs_id | HCLGE_TM_Q_QS_LINK_VLD_MSK);
+
+	return hclge_cmd_send(&hdev->hw, &desc, 1);
+}
+
+static int hclge_tm_pg_weight_cfg(struct hclge_dev *hdev, u8 pg_id,
+				  u8 dwrr)
+{
+	struct hclge_pg_weight_cmd *weight;
+	struct hclge_desc desc;
+
+	hclge_cmd_setup_basic_desc(&desc, HCLGE_OPC_TM_PG_WEIGHT, false);
+
+	weight = (struct hclge_pg_weight_cmd *)desc.data;
+
+	weight->pg_id = pg_id;
+	weight->dwrr = dwrr;
+
+	return hclge_cmd_send(&hdev->hw, &desc, 1);
+}
+
+static int hclge_tm_pri_weight_cfg(struct hclge_dev *hdev, u8 pri_id,
+				   u8 dwrr)
+{
+	struct hclge_priority_weight_cmd *weight;
+	struct hclge_desc desc;
+
+	hclge_cmd_setup_basic_desc(&desc, HCLGE_OPC_TM_PRI_WEIGHT, false);
+
+	weight = (struct hclge_priority_weight_cmd *)desc.data;
+
+	weight->pri_id = pri_id;
+	weight->dwrr = dwrr;
+
+	return hclge_cmd_send(&hdev->hw, &desc, 1);
+}
+
+static int hclge_tm_qs_weight_cfg(struct hclge_dev *hdev, u16 qs_id,
+				  u8 dwrr)
+{
+	struct hclge_qs_weight_cmd *weight;
+	struct hclge_desc desc;
+
+	hclge_cmd_setup_basic_desc(&desc, HCLGE_OPC_TM_QS_WEIGHT, false);
+
+	weight = (struct hclge_qs_weight_cmd *)desc.data;
+
+	weight->qs_id = cpu_to_le16(qs_id);
+	weight->dwrr = dwrr;
+
+	return hclge_cmd_send(&hdev->hw, &desc, 1);
+}
+
+static u32 hclge_tm_get_shapping_para(u8 ir_b, u8 ir_u, u8 ir_s,
+				      u8 bs_b, u8 bs_s)
+{
+	u32 shapping_para = 0;
+
+	hclge_tm_set_field(shapping_para, IR_B, ir_b);
+	hclge_tm_set_field(shapping_para, IR_U, ir_u);
+	hclge_tm_set_field(shapping_para, IR_S, ir_s);
+	hclge_tm_set_field(shapping_para, BS_B, bs_b);
+	hclge_tm_set_field(shapping_para, BS_S, bs_s);
+
+	return shapping_para;
+}
+
+static int hclge_tm_pg_shapping_cfg(struct hclge_dev *hdev,
+				    enum hclge_shap_bucket bucket, u8 pg_id,
+				    u32 shapping_para, u32 rate)
+{
+	struct hclge_pg_shapping_cmd *shap_cfg_cmd;
+	enum hclge_opcode_type opcode;
+	struct hclge_desc desc;
+
+	opcode = bucket ? HCLGE_OPC_TM_PG_P_SHAPPING :
+		 HCLGE_OPC_TM_PG_C_SHAPPING;
+	hclge_cmd_setup_basic_desc(&desc, opcode, false);
+
+	shap_cfg_cmd = (struct hclge_pg_shapping_cmd *)desc.data;
+
+	shap_cfg_cmd->pg_id = pg_id;
+
+	shap_cfg_cmd->pg_shapping_para = cpu_to_le32(shapping_para);
+
+	hnae3_set_bit(shap_cfg_cmd->flag, HCLGE_TM_RATE_VLD, 1);
+
+	shap_cfg_cmd->pg_rate = cpu_to_le32(rate);
+
+	return hclge_cmd_send(&hdev->hw, &desc, 1);
+}
+
+int hclge_tm_port_shaper_cfg(struct hclge_dev *hdev)
+{
+	struct hclge_port_shapping_cmd *shap_cfg_cmd;
+	struct hclge_shaper_ir_para ir_para;
+	struct hclge_desc desc;
+	u32 shapping_para;
+	int ret;
+
+	ret = hclge_shaper_para_calc(hdev->hw.mac.speed, HCLGE_SHAPER_LVL_PORT,
+				     &ir_para,
+				     hdev->ae_dev->dev_specs.max_tm_rate);
+	if (ret)
+		return ret;
+
+	hclge_cmd_setup_basic_desc(&desc, HCLGE_OPC_TM_PORT_SHAPPING, false);
+	shap_cfg_cmd = (struct hclge_port_shapping_cmd *)desc.data;
+
+	shapping_para = hclge_tm_get_shapping_para(ir_para.ir_b, ir_para.ir_u,
+						   ir_para.ir_s,
+						   HCLGE_SHAPER_BS_U_DEF,
+						   HCLGE_SHAPER_BS_S_DEF);
+
+	shap_cfg_cmd->port_shapping_para = cpu_to_le32(shapping_para);
+
+	hnae3_set_bit(shap_cfg_cmd->flag, HCLGE_TM_RATE_VLD, 1);
+
+	shap_cfg_cmd->port_rate = cpu_to_le32(hdev->hw.mac.speed);
+
+	return hclge_cmd_send(&hdev->hw, &desc, 1);
+}
+
+static int hclge_tm_pri_shapping_cfg(struct hclge_dev *hdev,
+				     enum hclge_shap_bucket bucket, u8 pri_id,
+				     u32 shapping_para, u32 rate)
+{
+	struct hclge_pri_shapping_cmd *shap_cfg_cmd;
+	enum hclge_opcode_type opcode;
+	struct hclge_desc desc;
+
+	opcode = bucket ? HCLGE_OPC_TM_PRI_P_SHAPPING :
+		 HCLGE_OPC_TM_PRI_C_SHAPPING;
+
+	hclge_cmd_setup_basic_desc(&desc, opcode, false);
+
+	shap_cfg_cmd = (struct hclge_pri_shapping_cmd *)desc.data;
+
+	shap_cfg_cmd->pri_id = pri_id;
+
+	shap_cfg_cmd->pri_shapping_para = cpu_to_le32(shapping_para);
+
+	hnae3_set_bit(shap_cfg_cmd->flag, HCLGE_TM_RATE_VLD, 1);
+
+	shap_cfg_cmd->pri_rate = cpu_to_le32(rate);
+
+	return hclge_cmd_send(&hdev->hw, &desc, 1);
+}
+
+static int hclge_tm_pg_schd_mode_cfg(struct hclge_dev *hdev, u8 pg_id)
+{
+	struct hclge_desc desc;
+
+	hclge_cmd_setup_basic_desc(&desc, HCLGE_OPC_TM_PG_SCH_MODE_CFG, false);
+
+	if (hdev->tm_info.pg_info[pg_id].pg_sch_mode == HCLGE_SCH_MODE_DWRR)
+		desc.data[1] = cpu_to_le32(HCLGE_TM_TX_SCHD_DWRR_MSK);
+	else
+		desc.data[1] = 0;
+
+	desc.data[0] = cpu_to_le32(pg_id);
+
+	return hclge_cmd_send(&hdev->hw, &desc, 1);
+}
+
+static int hclge_tm_pri_schd_mode_cfg(struct hclge_dev *hdev, u8 pri_id)
+{
+	struct hclge_desc desc;
+
+	hclge_cmd_setup_basic_desc(&desc, HCLGE_OPC_TM_PRI_SCH_MODE_CFG, false);
+
+	if (hdev->tm_info.tc_info[pri_id].tc_sch_mode == HCLGE_SCH_MODE_DWRR)
+		desc.data[1] = cpu_to_le32(HCLGE_TM_TX_SCHD_DWRR_MSK);
+	else
+		desc.data[1] = 0;
+
+	desc.data[0] = cpu_to_le32(pri_id);
+
+	return hclge_cmd_send(&hdev->hw, &desc, 1);
+}
+
+static int hclge_tm_qs_schd_mode_cfg(struct hclge_dev *hdev, u16 qs_id, u8 mode)
+{
+	struct hclge_desc desc;
+
+	hclge_cmd_setup_basic_desc(&desc, HCLGE_OPC_TM_QS_SCH_MODE_CFG, false);
+
+	if (mode == HCLGE_SCH_MODE_DWRR)
+		desc.data[1] = cpu_to_le32(HCLGE_TM_TX_SCHD_DWRR_MSK);
+	else
+		desc.data[1] = 0;
+
+	desc.data[0] = cpu_to_le32(qs_id);
+
+	return hclge_cmd_send(&hdev->hw, &desc, 1);
+}
+
+static int hclge_tm_qs_bp_cfg(struct hclge_dev *hdev, u8 tc, u8 grp_id,
+			      u32 bit_map)
+{
+	struct hclge_bp_to_qs_map_cmd *bp_to_qs_map_cmd;
+	struct hclge_desc desc;
+
+	hclge_cmd_setup_basic_desc(&desc, HCLGE_OPC_TM_BP_TO_QSET_MAPPING,
+				   false);
+
+	bp_to_qs_map_cmd = (struct hclge_bp_to_qs_map_cmd *)desc.data;
+
+	bp_to_qs_map_cmd->tc_id = tc;
+	bp_to_qs_map_cmd->qs_group_id = grp_id;
+	bp_to_qs_map_cmd->qs_bit_map = cpu_to_le32(bit_map);
+
+	return hclge_cmd_send(&hdev->hw, &desc, 1);
+}
+
+int hclge_tm_qs_shaper_cfg(struct hclge_vport *vport, int max_tx_rate)
+{
+	struct hnae3_knic_private_info *kinfo = &vport->nic.kinfo;
+	struct hclge_qs_shapping_cmd *shap_cfg_cmd;
+	struct hclge_shaper_ir_para ir_para;
+	struct hclge_dev *hdev = vport->back;
+	struct hclge_desc desc;
+	u32 shaper_para;
+	int ret, i;
+
+	if (!max_tx_rate)
+		max_tx_rate = hdev->ae_dev->dev_specs.max_tm_rate;
+
+	ret = hclge_shaper_para_calc(max_tx_rate, HCLGE_SHAPER_LVL_QSET,
+				     &ir_para,
+				     hdev->ae_dev->dev_specs.max_tm_rate);
+	if (ret)
+		return ret;
+
+	shaper_para = hclge_tm_get_shapping_para(ir_para.ir_b, ir_para.ir_u,
+						 ir_para.ir_s,
+						 HCLGE_SHAPER_BS_U_DEF,
+						 HCLGE_SHAPER_BS_S_DEF);
+
+	for (i = 0; i < kinfo->tc_info.num_tc; i++) {
+		hclge_cmd_setup_basic_desc(&desc, HCLGE_OPC_QCN_SHAPPING_CFG,
+					   false);
+
+		shap_cfg_cmd = (struct hclge_qs_shapping_cmd *)desc.data;
+		shap_cfg_cmd->qs_id = cpu_to_le16(vport->qs_offset + i);
+		shap_cfg_cmd->qs_shapping_para = cpu_to_le32(shaper_para);
+
+		hnae3_set_bit(shap_cfg_cmd->flag, HCLGE_TM_RATE_VLD, 1);
+		shap_cfg_cmd->qs_rate = cpu_to_le32(max_tx_rate);
+
+		ret = hclge_cmd_send(&hdev->hw, &desc, 1);
+		if (ret) {
+			dev_err(&hdev->pdev->dev,
+				"vport%u, qs%u failed to set tx_rate:%d, ret=%d\n",
+				vport->vport_id, shap_cfg_cmd->qs_id,
+				max_tx_rate, ret);
+			return ret;
+		}
+	}
+
+	return 0;
+}
+
+static u16 hclge_vport_get_max_rss_size(struct hclge_vport *vport)
+{
+	struct hnae3_knic_private_info *kinfo = &vport->nic.kinfo;
+	struct hnae3_tc_info *tc_info = &kinfo->tc_info;
+	struct hclge_dev *hdev = vport->back;
+	u16 max_rss_size = 0;
+	int i;
+
+	if (!tc_info->mqprio_active)
+		return vport->alloc_tqps / tc_info->num_tc;
+
+	for (i = 0; i < HNAE3_MAX_TC; i++) {
+		if (!(hdev->hw_tc_map & BIT(i)) || i >= tc_info->num_tc)
+			continue;
+		if (max_rss_size < tc_info->tqp_count[i])
+			max_rss_size = tc_info->tqp_count[i];
+	}
+
+	return max_rss_size;
+}
+
+static u16 hclge_vport_get_tqp_num(struct hclge_vport *vport)
+{
+	struct hnae3_knic_private_info *kinfo = &vport->nic.kinfo;
+	struct hnae3_tc_info *tc_info = &kinfo->tc_info;
+	struct hclge_dev *hdev = vport->back;
+	int sum = 0;
+	int i;
+
+	if (!tc_info->mqprio_active)
+		return kinfo->rss_size * tc_info->num_tc;
+
+	for (i = 0; i < HNAE3_MAX_TC; i++) {
+		if (hdev->hw_tc_map & BIT(i) && i < tc_info->num_tc)
+			sum += tc_info->tqp_count[i];
+	}
+
+	return sum;
+}
+
+static void hclge_tm_update_kinfo_rss_size(struct hclge_vport *vport)
+{
+	struct hnae3_knic_private_info *kinfo = &vport->nic.kinfo;
+	struct hclge_dev *hdev = vport->back;
+	u16 vport_max_rss_size;
+	u16 max_rss_size;
+
+	/* TC configuration is shared by PF/VF in one port, only allow
+	 * one tc for VF for simplicity. VF's vport_id is non zero.
+	 */
+	if (vport->vport_id) {
+		kinfo->tc_info.max_tc = 1;
+		kinfo->tc_info.num_tc = 1;
+		vport->qs_offset = HNAE3_MAX_TC +
+				   vport->vport_id - HCLGE_VF_VPORT_START_NUM;
+		vport_max_rss_size = hdev->vf_rss_size_max;
+	} else {
+		kinfo->tc_info.max_tc = hdev->tc_max;
+		kinfo->tc_info.num_tc =
+			min_t(u16, vport->alloc_tqps, hdev->tm_info.num_tc);
+		vport->qs_offset = 0;
+		vport_max_rss_size = hdev->pf_rss_size_max;
+	}
+
+	max_rss_size = min_t(u16, vport_max_rss_size,
+			     hclge_vport_get_max_rss_size(vport));
+
+	/* Set to user value, no larger than max_rss_size. */
+	if (kinfo->req_rss_size != kinfo->rss_size && kinfo->req_rss_size &&
+	    kinfo->req_rss_size <= max_rss_size) {
+		dev_info(&hdev->pdev->dev, "rss changes from %u to %u\n",
+			 kinfo->rss_size, kinfo->req_rss_size);
+		kinfo->rss_size = kinfo->req_rss_size;
+	} else if (kinfo->rss_size > max_rss_size ||
+		   (!kinfo->req_rss_size && kinfo->rss_size < max_rss_size)) {
+		/* Set to the maximum specification value (max_rss_size). */
+		kinfo->rss_size = max_rss_size;
+	}
+}
+
+static void hclge_tm_vport_tc_info_update(struct hclge_vport *vport)
+{
+	struct hnae3_knic_private_info *kinfo = &vport->nic.kinfo;
+	struct hclge_dev *hdev = vport->back;
+	u8 i;
+
+	hclge_tm_update_kinfo_rss_size(vport);
+	kinfo->num_tqps = hclge_vport_get_tqp_num(vport);
+	vport->dwrr = 100;  /* 100 percent as init */
+	vport->bw_limit = hdev->tm_info.pg_info[0].bw_limit;
+
+	if (vport->vport_id == PF_VPORT_ID)
+		hdev->rss_cfg.rss_size = kinfo->rss_size;
+
+	/* when enable mqprio, the tc_info has been updated. */
+	if (kinfo->tc_info.mqprio_active)
+		return;
+
+	for (i = 0; i < HNAE3_MAX_TC; i++) {
+		if (hdev->hw_tc_map & BIT(i) && i < kinfo->tc_info.num_tc) {
+			kinfo->tc_info.tqp_offset[i] = i * kinfo->rss_size;
+			kinfo->tc_info.tqp_count[i] = kinfo->rss_size;
+		} else {
+			/* Set to default queue if TC is disable */
+			kinfo->tc_info.tqp_offset[i] = 0;
+			kinfo->tc_info.tqp_count[i] = 1;
+		}
+	}
+
+	memcpy(kinfo->tc_info.prio_tc, hdev->tm_info.prio_tc,
+	       sizeof_field(struct hnae3_tc_info, prio_tc));
+}
+
+static void hclge_tm_vport_info_update(struct hclge_dev *hdev)
+{
+	struct hclge_vport *vport = hdev->vport;
+	u32 i;
+
+	for (i = 0; i < hdev->num_alloc_vport; i++) {
+		hclge_tm_vport_tc_info_update(vport);
+
+		vport++;
+	}
+}
+
+static void hclge_tm_tc_info_init(struct hclge_dev *hdev)
+{
+	u8 i, tc_sch_mode;
+	u32 bw_limit;
+
+	for (i = 0; i < hdev->tc_max; i++) {
+		if (i < hdev->tm_info.num_tc) {
+			tc_sch_mode = HCLGE_SCH_MODE_DWRR;
+			bw_limit = hdev->tm_info.pg_info[0].bw_limit;
+		} else {
+			tc_sch_mode = HCLGE_SCH_MODE_SP;
+			bw_limit = 0;
+		}
+
+		hdev->tm_info.tc_info[i].tc_id = i;
+		hdev->tm_info.tc_info[i].tc_sch_mode = tc_sch_mode;
+		hdev->tm_info.tc_info[i].pgid = 0;
+		hdev->tm_info.tc_info[i].bw_limit = bw_limit;
+	}
+
+	for (i = 0; i < HNAE3_MAX_USER_PRIO; i++)
+		hdev->tm_info.prio_tc[i] =
+			(i >= hdev->tm_info.num_tc) ? 0 : i;
+}
+
+static void hclge_tm_pg_info_init(struct hclge_dev *hdev)
+{
+#define BW_PERCENT	100
+#define DEFAULT_BW_WEIGHT	1
+
+	u8 i;
+
+	for (i = 0; i < hdev->tm_info.num_pg; i++) {
+		int k;
+
+		hdev->tm_info.pg_dwrr[i] = i ? 0 : BW_PERCENT;
+
+		hdev->tm_info.pg_info[i].pg_id = i;
+		hdev->tm_info.pg_info[i].pg_sch_mode = HCLGE_SCH_MODE_DWRR;
+
+		hdev->tm_info.pg_info[i].bw_limit =
+					hdev->ae_dev->dev_specs.max_tm_rate;
+
+		if (i != 0)
+			continue;
+
+		hdev->tm_info.pg_info[i].tc_bit_map = hdev->hw_tc_map;
+		for (k = 0; k < hdev->tm_info.num_tc; k++)
+			hdev->tm_info.pg_info[i].tc_dwrr[k] = BW_PERCENT;
+		for (; k < HNAE3_MAX_TC; k++)
+			hdev->tm_info.pg_info[i].tc_dwrr[k] = DEFAULT_BW_WEIGHT;
+	}
+}
+
+static void hclge_update_fc_mode_by_dcb_flag(struct hclge_dev *hdev)
+{
+	if (hdev->tm_info.num_tc == 1 && !hdev->tm_info.pfc_en) {
+		if (hdev->fc_mode_last_time == HCLGE_FC_PFC)
+			dev_warn(&hdev->pdev->dev,
+				 "Only 1 tc used, but last mode is FC_PFC\n");
+
+		hdev->tm_info.fc_mode = hdev->fc_mode_last_time;
+	} else if (hdev->tm_info.fc_mode != HCLGE_FC_PFC) {
+		/* fc_mode_last_time record the last fc_mode when
+		 * DCB is enabled, so that fc_mode can be set to
+		 * the correct value when DCB is disabled.
+		 */
+		hdev->fc_mode_last_time = hdev->tm_info.fc_mode;
+		hdev->tm_info.fc_mode = HCLGE_FC_PFC;
+	}
+}
+
+static void hclge_update_fc_mode(struct hclge_dev *hdev)
+{
+	if (!hdev->tm_info.pfc_en) {
+		hdev->tm_info.fc_mode = hdev->fc_mode_last_time;
+		return;
+	}
+
+	if (hdev->tm_info.fc_mode != HCLGE_FC_PFC) {
+		hdev->fc_mode_last_time = hdev->tm_info.fc_mode;
+		hdev->tm_info.fc_mode = HCLGE_FC_PFC;
+	}
+}
+
+void hclge_tm_pfc_info_update(struct hclge_dev *hdev)
+{
+	if (hdev->ae_dev->dev_version >= HNAE3_DEVICE_VERSION_V3)
+		hclge_update_fc_mode(hdev);
+	else
+		hclge_update_fc_mode_by_dcb_flag(hdev);
+}
+
+static void hclge_tm_schd_info_init(struct hclge_dev *hdev)
+{
+	hclge_tm_pg_info_init(hdev);
+
+	hclge_tm_tc_info_init(hdev);
+
+	hclge_tm_vport_info_update(hdev);
+
+	hclge_tm_pfc_info_update(hdev);
+}
+
+static int hclge_tm_pg_to_pri_map(struct hclge_dev *hdev)
+{
+	int ret;
+	u32 i;
+
+	if (hdev->tx_sch_mode != HCLGE_FLAG_TC_BASE_SCH_MODE)
+		return 0;
+
+	for (i = 0; i < hdev->tm_info.num_pg; i++) {
+		/* Cfg mapping */
+		ret = hclge_tm_pg_to_pri_map_cfg(
+			hdev, i, hdev->tm_info.pg_info[i].tc_bit_map);
+		if (ret)
+			return ret;
+	}
+
+	return 0;
+}
+
+static int hclge_tm_pg_shaper_cfg(struct hclge_dev *hdev)
+{
+	u32 max_tm_rate = hdev->ae_dev->dev_specs.max_tm_rate;
+	struct hclge_shaper_ir_para ir_para;
+	u32 shaper_para;
+	int ret;
+	u32 i;
+
+	/* Cfg pg schd */
+	if (hdev->tx_sch_mode != HCLGE_FLAG_TC_BASE_SCH_MODE)
+		return 0;
+
+	/* Pg to pri */
+	for (i = 0; i < hdev->tm_info.num_pg; i++) {
+		u32 rate = hdev->tm_info.pg_info[i].bw_limit;
+
+		/* Calc shaper para */
+		ret = hclge_shaper_para_calc(rate, HCLGE_SHAPER_LVL_PG,
+					     &ir_para, max_tm_rate);
+		if (ret)
+			return ret;
+
+		shaper_para = hclge_tm_get_shapping_para(0, 0, 0,
+							 HCLGE_SHAPER_BS_U_DEF,
+							 HCLGE_SHAPER_BS_S_DEF);
+		ret = hclge_tm_pg_shapping_cfg(hdev,
+					       HCLGE_TM_SHAP_C_BUCKET, i,
+					       shaper_para, rate);
+		if (ret)
+			return ret;
+
+		shaper_para = hclge_tm_get_shapping_para(ir_para.ir_b,
+							 ir_para.ir_u,
+							 ir_para.ir_s,
+							 HCLGE_SHAPER_BS_U_DEF,
+							 HCLGE_SHAPER_BS_S_DEF);
+		ret = hclge_tm_pg_shapping_cfg(hdev,
+					       HCLGE_TM_SHAP_P_BUCKET, i,
+					       shaper_para, rate);
+		if (ret)
+			return ret;
+	}
+
+	return 0;
+}
+
+static int hclge_tm_pg_dwrr_cfg(struct hclge_dev *hdev)
+{
+	int ret;
+	u32 i;
+
+	/* cfg pg schd */
+	if (hdev->tx_sch_mode != HCLGE_FLAG_TC_BASE_SCH_MODE)
+		return 0;
+
+	/* pg to prio */
+	for (i = 0; i < hdev->tm_info.num_pg; i++) {
+		/* Cfg dwrr */
+		ret = hclge_tm_pg_weight_cfg(hdev, i, hdev->tm_info.pg_dwrr[i]);
+		if (ret)
+			return ret;
+	}
+
+	return 0;
+}
+
+static int hclge_vport_q_to_qs_map(struct hclge_dev *hdev,
+				   struct hclge_vport *vport)
+{
+	struct hnae3_knic_private_info *kinfo = &vport->nic.kinfo;
+	struct hnae3_tc_info *tc_info = &kinfo->tc_info;
+	struct hnae3_queue **tqp = kinfo->tqp;
+	u32 i, j;
+	int ret;
+
+	for (i = 0; i < tc_info->num_tc; i++) {
+		for (j = 0; j < tc_info->tqp_count[i]; j++) {
+			struct hnae3_queue *q = tqp[tc_info->tqp_offset[i] + j];
+
+			ret = hclge_tm_q_to_qs_map_cfg(hdev,
+						       hclge_get_queue_id(q),
+						       vport->qs_offset + i);
+			if (ret)
+				return ret;
+		}
+	}
+
+	return 0;
+}
+
+static int hclge_tm_pri_q_qs_cfg_tc_base(struct hclge_dev *hdev)
+{
+	struct hclge_vport *vport = hdev->vport;
+	u16 i, k;
+	int ret;
+
+	/* Cfg qs -> pri mapping, one by one mapping */
+	for (k = 0; k < hdev->num_alloc_vport; k++) {
+		struct hnae3_knic_private_info *kinfo = &vport[k].nic.kinfo;
+
+		for (i = 0; i < kinfo->tc_info.max_tc; i++) {
+			u8 pri = i < kinfo->tc_info.num_tc ? i : 0;
+			bool link_vld = i < kinfo->tc_info.num_tc;
+
+			ret = hclge_tm_qs_to_pri_map_cfg(hdev,
+							 vport[k].qs_offset + i,
+							 pri, link_vld);
+			if (ret)
+				return ret;
+		}
+	}
+
+	return 0;
+}
+
+static int hclge_tm_pri_q_qs_cfg_vnet_base(struct hclge_dev *hdev)
+{
+	struct hclge_vport *vport = hdev->vport;
+	u16 i, k;
+	int ret;
+
+	/* Cfg qs -> pri mapping,  qs = tc, pri = vf, 8 qs -> 1 pri */
+	for (k = 0; k < hdev->num_alloc_vport; k++)
+		for (i = 0; i < HNAE3_MAX_TC; i++) {
+			ret = hclge_tm_qs_to_pri_map_cfg(hdev,
+							 vport[k].qs_offset + i,
+							 k, true);
+			if (ret)
+				return ret;
+		}
+
+	return 0;
+}
+
+static int hclge_tm_pri_q_qs_cfg(struct hclge_dev *hdev)
+{
+	struct hclge_vport *vport = hdev->vport;
+	int ret;
+	u32 i;
+
+	if (hdev->tx_sch_mode == HCLGE_FLAG_TC_BASE_SCH_MODE)
+		ret = hclge_tm_pri_q_qs_cfg_tc_base(hdev);
+	else if (hdev->tx_sch_mode == HCLGE_FLAG_VNET_BASE_SCH_MODE)
+		ret = hclge_tm_pri_q_qs_cfg_vnet_base(hdev);
+	else
+		return -EINVAL;
+
+	if (ret)
+		return ret;
+
+	/* Cfg q -> qs mapping */
+	for (i = 0; i < hdev->num_alloc_vport; i++) {
+		ret = hclge_vport_q_to_qs_map(hdev, vport);
+		if (ret)
+			return ret;
+
+		vport++;
+	}
+
+	return 0;
+}
+
+static int hclge_tm_pri_tc_base_shaper_cfg(struct hclge_dev *hdev)
+{
+	u32 max_tm_rate = hdev->ae_dev->dev_specs.max_tm_rate;
+	struct hclge_shaper_ir_para ir_para;
+	u32 shaper_para_c, shaper_para_p;
+	int ret;
+	u32 i;
+
+	for (i = 0; i < hdev->tc_max; i++) {
+		u32 rate = hdev->tm_info.tc_info[i].bw_limit;
+
+		if (rate) {
+			ret = hclge_shaper_para_calc(rate, HCLGE_SHAPER_LVL_PRI,
+						     &ir_para, max_tm_rate);
+			if (ret)
+				return ret;
+
+			shaper_para_c = hclge_tm_get_shapping_para(0, 0, 0,
+								   HCLGE_SHAPER_BS_U_DEF,
+								   HCLGE_SHAPER_BS_S_DEF);
+			shaper_para_p = hclge_tm_get_shapping_para(ir_para.ir_b,
+								   ir_para.ir_u,
+								   ir_para.ir_s,
+								   HCLGE_SHAPER_BS_U_DEF,
+								   HCLGE_SHAPER_BS_S_DEF);
+		} else {
+			shaper_para_c = 0;
+			shaper_para_p = 0;
+		}
+
+		ret = hclge_tm_pri_shapping_cfg(hdev, HCLGE_TM_SHAP_C_BUCKET, i,
+						shaper_para_c, rate);
+		if (ret)
+			return ret;
+
+		ret = hclge_tm_pri_shapping_cfg(hdev, HCLGE_TM_SHAP_P_BUCKET, i,
+						shaper_para_p, rate);
+		if (ret)
+			return ret;
+	}
+
+	return 0;
+}
+
+static int hclge_tm_pri_vnet_base_shaper_pri_cfg(struct hclge_vport *vport)
+{
+	struct hclge_dev *hdev = vport->back;
+	struct hclge_shaper_ir_para ir_para;
+	u32 shaper_para;
+	int ret;
+
+	ret = hclge_shaper_para_calc(vport->bw_limit, HCLGE_SHAPER_LVL_VF,
+				     &ir_para,
+				     hdev->ae_dev->dev_specs.max_tm_rate);
+	if (ret)
+		return ret;
+
+	shaper_para = hclge_tm_get_shapping_para(0, 0, 0,
+						 HCLGE_SHAPER_BS_U_DEF,
+						 HCLGE_SHAPER_BS_S_DEF);
+	ret = hclge_tm_pri_shapping_cfg(hdev, HCLGE_TM_SHAP_C_BUCKET,
+					vport->vport_id, shaper_para,
+					vport->bw_limit);
+	if (ret)
+		return ret;
+
+	shaper_para = hclge_tm_get_shapping_para(ir_para.ir_b, ir_para.ir_u,
+						 ir_para.ir_s,
+						 HCLGE_SHAPER_BS_U_DEF,
+						 HCLGE_SHAPER_BS_S_DEF);
+	ret = hclge_tm_pri_shapping_cfg(hdev, HCLGE_TM_SHAP_P_BUCKET,
+					vport->vport_id, shaper_para,
+					vport->bw_limit);
+	if (ret)
+		return ret;
+
+	return 0;
+}
+
+static int hclge_tm_pri_vnet_base_shaper_qs_cfg(struct hclge_vport *vport)
+{
+	struct hnae3_knic_private_info *kinfo = &vport->nic.kinfo;
+	struct hclge_dev *hdev = vport->back;
+	u32 max_tm_rate = hdev->ae_dev->dev_specs.max_tm_rate;
+	struct hclge_shaper_ir_para ir_para;
+	u32 i;
+	int ret;
+
+	for (i = 0; i < kinfo->tc_info.num_tc; i++) {
+		ret = hclge_shaper_para_calc(hdev->tm_info.tc_info[i].bw_limit,
+					     HCLGE_SHAPER_LVL_QSET,
+					     &ir_para, max_tm_rate);
+		if (ret)
+			return ret;
+	}
+
+	return 0;
+}
+
+static int hclge_tm_pri_vnet_base_shaper_cfg(struct hclge_dev *hdev)
+{
+	struct hclge_vport *vport = hdev->vport;
+	int ret;
+	u32 i;
+
+	/* Need config vport shaper */
+	for (i = 0; i < hdev->num_alloc_vport; i++) {
+		ret = hclge_tm_pri_vnet_base_shaper_pri_cfg(vport);
+		if (ret)
+			return ret;
+
+		ret = hclge_tm_pri_vnet_base_shaper_qs_cfg(vport);
+		if (ret)
+			return ret;
+
+		vport++;
+	}
+
+	return 0;
+}
+
+static int hclge_tm_pri_shaper_cfg(struct hclge_dev *hdev)
+{
+	int ret;
+
+	if (hdev->tx_sch_mode == HCLGE_FLAG_TC_BASE_SCH_MODE) {
+		ret = hclge_tm_pri_tc_base_shaper_cfg(hdev);
+		if (ret)
+			return ret;
+	} else {
+		ret = hclge_tm_pri_vnet_base_shaper_cfg(hdev);
+		if (ret)
+			return ret;
+	}
+
+	return 0;
+}
+
+static int hclge_tm_pri_tc_base_dwrr_cfg(struct hclge_dev *hdev)
+{
+	struct hclge_vport *vport = hdev->vport;
+	struct hclge_pg_info *pg_info;
+	u8 dwrr;
+	int ret;
+	u32 i, k;
+
+	for (i = 0; i < hdev->tc_max; i++) {
+		pg_info =
+			&hdev->tm_info.pg_info[hdev->tm_info.tc_info[i].pgid];
+		dwrr = pg_info->tc_dwrr[i];
+
+		ret = hclge_tm_pri_weight_cfg(hdev, i, dwrr);
+		if (ret)
+			return ret;
+
+		for (k = 0; k < hdev->num_alloc_vport; k++) {
+			struct hnae3_knic_private_info *kinfo = &vport[k].nic.kinfo;
+
+			if (i >= kinfo->tc_info.max_tc)
+				continue;
+
+			dwrr = i < kinfo->tc_info.num_tc ? vport[k].dwrr : 0;
+			ret = hclge_tm_qs_weight_cfg(
+				hdev, vport[k].qs_offset + i,
+				dwrr);
+			if (ret)
+				return ret;
+		}
+	}
+
+	return 0;
+}
+
+static int hclge_tm_ets_tc_dwrr_cfg(struct hclge_dev *hdev)
+{
+#define DEFAULT_TC_OFFSET	14
+
+	struct hclge_ets_tc_weight_cmd *ets_weight;
+	struct hclge_desc desc;
+	unsigned int i;
+
+	hclge_cmd_setup_basic_desc(&desc, HCLGE_OPC_ETS_TC_WEIGHT, false);
+	ets_weight = (struct hclge_ets_tc_weight_cmd *)desc.data;
+
+	for (i = 0; i < HNAE3_MAX_TC; i++) {
+		struct hclge_pg_info *pg_info;
+
+		pg_info = &hdev->tm_info.pg_info[hdev->tm_info.tc_info[i].pgid];
+		ets_weight->tc_weight[i] = pg_info->tc_dwrr[i];
+	}
+
+	ets_weight->weight_offset = DEFAULT_TC_OFFSET;
+
+	return hclge_cmd_send(&hdev->hw, &desc, 1);
+}
+
+static int hclge_tm_pri_vnet_base_dwrr_pri_cfg(struct hclge_vport *vport)
+{
+	struct hnae3_knic_private_info *kinfo = &vport->nic.kinfo;
+	struct hclge_dev *hdev = vport->back;
+	int ret;
+	u8 i;
+
+	/* Vf dwrr */
+	ret = hclge_tm_pri_weight_cfg(hdev, vport->vport_id, vport->dwrr);
+	if (ret)
+		return ret;
+
+	/* Qset dwrr */
+	for (i = 0; i < kinfo->tc_info.num_tc; i++) {
+		ret = hclge_tm_qs_weight_cfg(
+			hdev, vport->qs_offset + i,
+			hdev->tm_info.pg_info[0].tc_dwrr[i]);
+		if (ret)
+			return ret;
+	}
+
+	return 0;
+}
+
+static int hclge_tm_pri_vnet_base_dwrr_cfg(struct hclge_dev *hdev)
+{
+	struct hclge_vport *vport = hdev->vport;
+	int ret;
+	u32 i;
+
+	for (i = 0; i < hdev->num_alloc_vport; i++) {
+		ret = hclge_tm_pri_vnet_base_dwrr_pri_cfg(vport);
+		if (ret)
+			return ret;
+
+		vport++;
+	}
+
+	return 0;
+}
+
+static int hclge_tm_pri_dwrr_cfg(struct hclge_dev *hdev)
+{
+	int ret;
+
+	if (hdev->tx_sch_mode == HCLGE_FLAG_TC_BASE_SCH_MODE) {
+		ret = hclge_tm_pri_tc_base_dwrr_cfg(hdev);
+		if (ret)
+			return ret;
+
+		if (!hnae3_dev_dcb_supported(hdev))
+			return 0;
+
+		ret = hclge_tm_ets_tc_dwrr_cfg(hdev);
+		if (ret == -EOPNOTSUPP) {
+			dev_warn(&hdev->pdev->dev,
+				 "fw %08x doesn't support ets tc weight cmd\n",
+				 hdev->fw_version);
+			ret = 0;
+		}
+
+		return ret;
+	} else {
+		ret = hclge_tm_pri_vnet_base_dwrr_cfg(hdev);
+		if (ret)
+			return ret;
+	}
+
+	return 0;
+}
+
+static int hclge_tm_map_cfg(struct hclge_dev *hdev)
+{
+	int ret;
+
+	ret = hclge_up_to_tc_map(hdev);
+	if (ret)
+		return ret;
+
+	if (hdev->vport[0].nic.kinfo.tc_map_mode == HNAE3_TC_MAP_MODE_DSCP) {
+		ret = hclge_dscp_to_tc_map(hdev);
+		if (ret)
+			return ret;
+	}
+
+	ret = hclge_tm_pg_to_pri_map(hdev);
+	if (ret)
+		return ret;
+
+	return hclge_tm_pri_q_qs_cfg(hdev);
+}
+
+static int hclge_tm_shaper_cfg(struct hclge_dev *hdev)
+{
+	int ret;
+
+	ret = hclge_tm_port_shaper_cfg(hdev);
+	if (ret)
+		return ret;
+
+	ret = hclge_tm_pg_shaper_cfg(hdev);
+	if (ret)
+		return ret;
+
+	return hclge_tm_pri_shaper_cfg(hdev);
+}
+
+int hclge_tm_dwrr_cfg(struct hclge_dev *hdev)
+{
+	int ret;
+
+	ret = hclge_tm_pg_dwrr_cfg(hdev);
+	if (ret)
+		return ret;
+
+	return hclge_tm_pri_dwrr_cfg(hdev);
+}
+
+static int hclge_tm_lvl2_schd_mode_cfg(struct hclge_dev *hdev)
+{
+	int ret;
+	u8 i;
+
+	/* Only being config on TC-Based scheduler mode */
+	if (hdev->tx_sch_mode == HCLGE_FLAG_VNET_BASE_SCH_MODE)
+		return 0;
+
+	for (i = 0; i < hdev->tm_info.num_pg; i++) {
+		ret = hclge_tm_pg_schd_mode_cfg(hdev, i);
+		if (ret)
+			return ret;
+	}
+
+	return 0;
+}
+
+static int hclge_tm_schd_mode_tc_base_cfg(struct hclge_dev *hdev, u8 pri_id)
+{
+	struct hclge_vport *vport = hdev->vport;
+	int ret;
+	u8 mode;
+	u16 i;
+
+	ret = hclge_tm_pri_schd_mode_cfg(hdev, pri_id);
+	if (ret)
+		return ret;
+
+	for (i = 0; i < hdev->num_alloc_vport; i++) {
+		struct hnae3_knic_private_info *kinfo = &vport[i].nic.kinfo;
+
+		if (pri_id >= kinfo->tc_info.max_tc)
+			continue;
+
+		mode = pri_id < kinfo->tc_info.num_tc ? HCLGE_SCH_MODE_DWRR :
+		       HCLGE_SCH_MODE_SP;
+		ret = hclge_tm_qs_schd_mode_cfg(hdev,
+						vport[i].qs_offset + pri_id,
+						mode);
+		if (ret)
+			return ret;
+	}
+
+	return 0;
+}
+
+static int hclge_tm_schd_mode_vnet_base_cfg(struct hclge_vport *vport)
+{
+	struct hnae3_knic_private_info *kinfo = &vport->nic.kinfo;
+	struct hclge_dev *hdev = vport->back;
+	int ret;
+	u8 i;
+
+	if (vport->vport_id >= HNAE3_MAX_TC)
+		return -EINVAL;
+
+	ret = hclge_tm_pri_schd_mode_cfg(hdev, vport->vport_id);
+	if (ret)
+		return ret;
+
+	for (i = 0; i < kinfo->tc_info.num_tc; i++) {
+		u8 sch_mode = hdev->tm_info.tc_info[i].tc_sch_mode;
+
+		ret = hclge_tm_qs_schd_mode_cfg(hdev, vport->qs_offset + i,
+						sch_mode);
+		if (ret)
+			return ret;
+	}
+
+	return 0;
+}
+
+static int hclge_tm_lvl34_schd_mode_cfg(struct hclge_dev *hdev)
+{
+	struct hclge_vport *vport = hdev->vport;
+	int ret;
+	u8 i;
+
+	if (hdev->tx_sch_mode == HCLGE_FLAG_TC_BASE_SCH_MODE) {
+		for (i = 0; i < hdev->tc_max; i++) {
+			ret = hclge_tm_schd_mode_tc_base_cfg(hdev, i);
+			if (ret)
+				return ret;
+		}
+	} else {
+		for (i = 0; i < hdev->num_alloc_vport; i++) {
+			ret = hclge_tm_schd_mode_vnet_base_cfg(vport);
+			if (ret)
+				return ret;
+
+			vport++;
+		}
+	}
+
+	return 0;
+}
+
+static int hclge_tm_schd_mode_hw(struct hclge_dev *hdev)
+{
+	int ret;
+
+	ret = hclge_tm_lvl2_schd_mode_cfg(hdev);
+	if (ret)
+		return ret;
+
+	return hclge_tm_lvl34_schd_mode_cfg(hdev);
+}
+
+int hclge_tm_schd_setup_hw(struct hclge_dev *hdev)
+{
+	int ret;
+
+	/* Cfg tm mapping  */
+	ret = hclge_tm_map_cfg(hdev);
+	if (ret)
+		return ret;
+
+	/* Cfg tm shaper */
+	ret = hclge_tm_shaper_cfg(hdev);
+	if (ret)
+		return ret;
+
+	/* Cfg dwrr */
+	ret = hclge_tm_dwrr_cfg(hdev);
+	if (ret)
+		return ret;
+
+	/* Cfg schd mode for each level schd */
+	ret = hclge_tm_schd_mode_hw(hdev);
+	if (ret)
+		return ret;
+
+	return hclge_tm_flush_cfg(hdev, false);
+}
+
+static int hclge_pause_param_setup_hw(struct hclge_dev *hdev)
+{
+	struct hclge_mac *mac = &hdev->hw.mac;
+
+	return hclge_pause_param_cfg(hdev, mac->mac_addr,
+				     HCLGE_DEFAULT_PAUSE_TRANS_GAP,
+				     HCLGE_DEFAULT_PAUSE_TRANS_TIME);
+}
+
+static int hclge_pfc_setup_hw(struct hclge_dev *hdev)
+{
+	u8 enable_bitmap = 0;
+
+	if (hdev->tm_info.fc_mode == HCLGE_FC_PFC)
+		enable_bitmap = HCLGE_TX_MAC_PAUSE_EN_MSK |
+				HCLGE_RX_MAC_PAUSE_EN_MSK;
+
+	return hclge_pfc_pause_en_cfg(hdev, enable_bitmap,
+				      hdev->tm_info.pfc_en);
+}
+
+/* for the queues that use for backpress, divides to several groups,
+ * each group contains 32 queue sets, which can be represented by u32 bitmap.
+ */
+static int hclge_bp_setup_hw(struct hclge_dev *hdev, u8 tc)
+{
+	u16 grp_id_shift = HCLGE_BP_GRP_ID_S;
+	u16 grp_id_mask = HCLGE_BP_GRP_ID_M;
+	u8 grp_num = HCLGE_BP_GRP_NUM;
+	int i;
+
+	if (hdev->num_tqps > HCLGE_TQP_MAX_SIZE_DEV_V2) {
+		grp_num = HCLGE_BP_EXT_GRP_NUM;
+		grp_id_mask = HCLGE_BP_EXT_GRP_ID_M;
+		grp_id_shift = HCLGE_BP_EXT_GRP_ID_S;
+	}
+
+	for (i = 0; i < grp_num; i++) {
+		u32 qs_bitmap = 0;
+		int k, ret;
+
+		for (k = 0; k < hdev->num_alloc_vport; k++) {
+			struct hclge_vport *vport = &hdev->vport[k];
+			u16 qs_id = vport->qs_offset + tc;
+			u8 grp, sub_grp;
+
+			grp = hnae3_get_field(qs_id, grp_id_mask, grp_id_shift);
+			sub_grp = hnae3_get_field(qs_id, HCLGE_BP_SUB_GRP_ID_M,
+						  HCLGE_BP_SUB_GRP_ID_S);
+			if (i == grp)
+				qs_bitmap |= (1 << sub_grp);
+		}
+
+		ret = hclge_tm_qs_bp_cfg(hdev, tc, i, qs_bitmap);
+		if (ret)
+			return ret;
+	}
+
+	return 0;
+}
+
+int hclge_mac_pause_setup_hw(struct hclge_dev *hdev)
+{
+	bool tx_en, rx_en;
+
+	switch (hdev->tm_info.fc_mode) {
+	case HCLGE_FC_NONE:
+		tx_en = false;
+		rx_en = false;
+		break;
+	case HCLGE_FC_RX_PAUSE:
+		tx_en = false;
+		rx_en = true;
+		break;
+	case HCLGE_FC_TX_PAUSE:
+		tx_en = true;
+		rx_en = false;
+		break;
+	case HCLGE_FC_FULL:
+		tx_en = true;
+		rx_en = true;
+		break;
+	case HCLGE_FC_PFC:
+		tx_en = false;
+		rx_en = false;
+		break;
+	default:
+		tx_en = true;
+		rx_en = true;
+	}
+
+	return hclge_mac_pause_en_cfg(hdev, tx_en, rx_en);
+}
+
+static int hclge_tm_bp_setup(struct hclge_dev *hdev)
+{
+	int ret;
+	int i;
+
+	for (i = 0; i < hdev->tm_info.num_tc; i++) {
+		ret = hclge_bp_setup_hw(hdev, i);
+		if (ret)
+			return ret;
+	}
+
+	return 0;
+}
+
+int hclge_pause_setup_hw(struct hclge_dev *hdev, bool init)
+{
+	int ret;
+
+	ret = hclge_pause_param_setup_hw(hdev);
+	if (ret)
+		return ret;
+
+	ret = hclge_mac_pause_setup_hw(hdev);
+	if (ret)
+		return ret;
+
+	/* Only DCB-supported dev supports qset back pressure and pfc cmd */
+	if (!hnae3_dev_dcb_supported(hdev))
+		return 0;
+
+	/* GE MAC does not support PFC, when driver is initializing and MAC
+	 * is in GE Mode, ignore the error here, otherwise initialization
+	 * will fail.
+	 */
+	ret = hclge_pfc_setup_hw(hdev);
+	if (init && ret == -EOPNOTSUPP)
+		dev_warn(&hdev->pdev->dev, "GE MAC does not support pfc\n");
+	else if (ret) {
+		dev_err(&hdev->pdev->dev, "config pfc failed! ret = %d\n",
+			ret);
+		return ret;
+	}
+
+	return hclge_tm_bp_setup(hdev);
+}
+
+void hclge_tm_prio_tc_info_update(struct hclge_dev *hdev, u8 *prio_tc)
+{
+	struct hclge_vport *vport = hdev->vport;
+	struct hnae3_knic_private_info *kinfo;
+	u32 i, k;
+
+	for (i = 0; i < HNAE3_MAX_USER_PRIO; i++) {
+		hdev->tm_info.prio_tc[i] = prio_tc[i];
+
+		for (k = 0;  k < hdev->num_alloc_vport; k++) {
+			kinfo = &vport[k].nic.kinfo;
+			kinfo->tc_info.prio_tc[i] = prio_tc[i];
+		}
+	}
+}
+
+void hclge_tm_schd_info_update(struct hclge_dev *hdev, u8 num_tc)
+{
+	u8 bit_map = 0;
+	u8 i;
+
+	hdev->tm_info.num_tc = num_tc;
+
+	for (i = 0; i < hdev->tm_info.num_tc; i++)
+		bit_map |= BIT(i);
+
+	if (!bit_map) {
+		bit_map = 1;
+		hdev->tm_info.num_tc = 1;
+	}
+
+	hdev->hw_tc_map = bit_map;
+
+	hclge_tm_schd_info_init(hdev);
+}
+
+int hclge_tm_init_hw(struct hclge_dev *hdev, bool init)
+{
+	int ret;
+
+	if ((hdev->tx_sch_mode != HCLGE_FLAG_TC_BASE_SCH_MODE) &&
+	    (hdev->tx_sch_mode != HCLGE_FLAG_VNET_BASE_SCH_MODE))
+		return -ENOTSUPP;
+
+	ret = hclge_tm_schd_setup_hw(hdev);
+	if (ret)
+		return ret;
+
+	ret = hclge_pause_setup_hw(hdev, init);
+	if (ret)
+		return ret;
+
+	return 0;
+}
+
+int hclge_tm_schd_init(struct hclge_dev *hdev)
+{
+	/* fc_mode is HCLGE_FC_FULL on reset */
+	hdev->tm_info.fc_mode = HCLGE_FC_FULL;
+	hdev->fc_mode_last_time = hdev->tm_info.fc_mode;
+
+	if (hdev->tx_sch_mode != HCLGE_FLAG_TC_BASE_SCH_MODE &&
+	    hdev->tm_info.num_pg != 1)
+		return -EINVAL;
+
+	hclge_tm_schd_info_init(hdev);
+	hclge_dscp_to_prio_map_init(hdev);
+
+	return hclge_tm_init_hw(hdev, true);
+}
+
+int hclge_tm_vport_map_update(struct hclge_dev *hdev)
+{
+	struct hclge_vport *vport = hdev->vport;
+	int ret;
+
+	hclge_tm_vport_tc_info_update(vport);
+
+	ret = hclge_vport_q_to_qs_map(hdev, vport);
+	if (ret)
+		return ret;
+
+	if (hdev->tm_info.num_tc == 1 && !hdev->tm_info.pfc_en)
+		return 0;
+
+	return hclge_tm_bp_setup(hdev);
+}
+
+int hclge_tm_get_qset_num(struct hclge_dev *hdev, u16 *qset_num)
+{
+	struct hclge_tm_nodes_cmd *nodes;
+	struct hclge_desc desc;
+	int ret;
+
+	if (hdev->ae_dev->dev_version <= HNAE3_DEVICE_VERSION_V2) {
+		/* Each PF has 8 qsets and each VF has 1 qset */
+		*qset_num = HCLGE_TM_PF_MAX_QSET_NUM + pci_num_vf(hdev->pdev);
+		return 0;
+	}
+
+	hclge_cmd_setup_basic_desc(&desc, HCLGE_OPC_TM_NODES, true);
+	ret = hclge_cmd_send(&hdev->hw, &desc, 1);
+	if (ret) {
+		dev_err(&hdev->pdev->dev,
+			"failed to get qset num, ret = %d\n", ret);
+		return ret;
+	}
+
+	nodes = (struct hclge_tm_nodes_cmd *)desc.data;
+	*qset_num = le16_to_cpu(nodes->qset_num);
+	return 0;
+}
+
+int hclge_tm_get_pri_num(struct hclge_dev *hdev, u8 *pri_num)
+{
+	struct hclge_tm_nodes_cmd *nodes;
+	struct hclge_desc desc;
+	int ret;
+
+	if (hdev->ae_dev->dev_version <= HNAE3_DEVICE_VERSION_V2) {
+		*pri_num = HCLGE_TM_PF_MAX_PRI_NUM;
+		return 0;
+	}
+
+	hclge_cmd_setup_basic_desc(&desc, HCLGE_OPC_TM_NODES, true);
+	ret = hclge_cmd_send(&hdev->hw, &desc, 1);
+	if (ret) {
+		dev_err(&hdev->pdev->dev,
+			"failed to get pri num, ret = %d\n", ret);
+		return ret;
+	}
+
+	nodes = (struct hclge_tm_nodes_cmd *)desc.data;
+	*pri_num = nodes->pri_num;
+	return 0;
+}
+
+int hclge_tm_get_qset_map_pri(struct hclge_dev *hdev, u16 qset_id, u8 *priority,
+			      u8 *link_vld)
+{
+	struct hclge_qs_to_pri_link_cmd *map;
+	struct hclge_desc desc;
+	int ret;
+
+	hclge_cmd_setup_basic_desc(&desc, HCLGE_OPC_TM_QS_TO_PRI_LINK, true);
+	map = (struct hclge_qs_to_pri_link_cmd *)desc.data;
+	map->qs_id = cpu_to_le16(qset_id);
+	ret = hclge_cmd_send(&hdev->hw, &desc, 1);
+	if (ret) {
+		dev_err(&hdev->pdev->dev,
+			"failed to get qset map priority, ret = %d\n", ret);
+		return ret;
+	}
+
+	*priority = map->priority;
+	*link_vld = map->link_vld;
+	return 0;
+}
+
+int hclge_tm_get_qset_sch_mode(struct hclge_dev *hdev, u16 qset_id, u8 *mode)
+{
+	struct hclge_qs_sch_mode_cfg_cmd *qs_sch_mode;
+	struct hclge_desc desc;
+	int ret;
+
+	hclge_cmd_setup_basic_desc(&desc, HCLGE_OPC_TM_QS_SCH_MODE_CFG, true);
+	qs_sch_mode = (struct hclge_qs_sch_mode_cfg_cmd *)desc.data;
+	qs_sch_mode->qs_id = cpu_to_le16(qset_id);
+	ret = hclge_cmd_send(&hdev->hw, &desc, 1);
+	if (ret) {
+		dev_err(&hdev->pdev->dev,
+			"failed to get qset sch mode, ret = %d\n", ret);
+		return ret;
+	}
+
+	*mode = qs_sch_mode->sch_mode;
+	return 0;
+}
+
+int hclge_tm_get_qset_weight(struct hclge_dev *hdev, u16 qset_id, u8 *weight)
+{
+	struct hclge_qs_weight_cmd *qs_weight;
+	struct hclge_desc desc;
+	int ret;
+
+	hclge_cmd_setup_basic_desc(&desc, HCLGE_OPC_TM_QS_WEIGHT, true);
+	qs_weight = (struct hclge_qs_weight_cmd *)desc.data;
+	qs_weight->qs_id = cpu_to_le16(qset_id);
+	ret = hclge_cmd_send(&hdev->hw, &desc, 1);
+	if (ret) {
+		dev_err(&hdev->pdev->dev,
+			"failed to get qset weight, ret = %d\n", ret);
+		return ret;
+	}
+
+	*weight = qs_weight->dwrr;
+	return 0;
+}
+
+int hclge_tm_get_qset_shaper(struct hclge_dev *hdev, u16 qset_id,
+			     struct hclge_tm_shaper_para *para)
+{
+	struct hclge_qs_shapping_cmd *shap_cfg_cmd;
+	struct hclge_desc desc;
+	u32 shapping_para;
+	int ret;
+
+	hclge_cmd_setup_basic_desc(&desc, HCLGE_OPC_QCN_SHAPPING_CFG, true);
+	shap_cfg_cmd = (struct hclge_qs_shapping_cmd *)desc.data;
+	shap_cfg_cmd->qs_id = cpu_to_le16(qset_id);
+	ret = hclge_cmd_send(&hdev->hw, &desc, 1);
+	if (ret) {
+		dev_err(&hdev->pdev->dev,
+			"failed to get qset %u shaper, ret = %d\n", qset_id,
+			ret);
+		return ret;
+	}
+
+	shapping_para = le32_to_cpu(shap_cfg_cmd->qs_shapping_para);
+	para->ir_b = hclge_tm_get_field(shapping_para, IR_B);
+	para->ir_u = hclge_tm_get_field(shapping_para, IR_U);
+	para->ir_s = hclge_tm_get_field(shapping_para, IR_S);
+	para->bs_b = hclge_tm_get_field(shapping_para, BS_B);
+	para->bs_s = hclge_tm_get_field(shapping_para, BS_S);
+	para->flag = shap_cfg_cmd->flag;
+	para->rate = le32_to_cpu(shap_cfg_cmd->qs_rate);
+	return 0;
+}
+
+int hclge_tm_get_pri_sch_mode(struct hclge_dev *hdev, u8 pri_id, u8 *mode)
+{
+	struct hclge_pri_sch_mode_cfg_cmd *pri_sch_mode;
+	struct hclge_desc desc;
+	int ret;
+
+	hclge_cmd_setup_basic_desc(&desc, HCLGE_OPC_TM_PRI_SCH_MODE_CFG, true);
+	pri_sch_mode = (struct hclge_pri_sch_mode_cfg_cmd *)desc.data;
+	pri_sch_mode->pri_id = pri_id;
+	ret = hclge_cmd_send(&hdev->hw, &desc, 1);
+	if (ret) {
+		dev_err(&hdev->pdev->dev,
+			"failed to get priority sch mode, ret = %d\n", ret);
+		return ret;
+	}
+
+	*mode = pri_sch_mode->sch_mode;
+	return 0;
+}
+
+int hclge_tm_get_pri_weight(struct hclge_dev *hdev, u8 pri_id, u8 *weight)
+{
+	struct hclge_priority_weight_cmd *priority_weight;
+	struct hclge_desc desc;
+	int ret;
+
+	hclge_cmd_setup_basic_desc(&desc, HCLGE_OPC_TM_PRI_WEIGHT, true);
+	priority_weight = (struct hclge_priority_weight_cmd *)desc.data;
+	priority_weight->pri_id = pri_id;
+	ret = hclge_cmd_send(&hdev->hw, &desc, 1);
+	if (ret) {
+		dev_err(&hdev->pdev->dev,
+			"failed to get priority weight, ret = %d\n", ret);
+		return ret;
+	}
+
+	*weight = priority_weight->dwrr;
+	return 0;
+}
+
+int hclge_tm_get_pri_shaper(struct hclge_dev *hdev, u8 pri_id,
+			    enum hclge_opcode_type cmd,
+			    struct hclge_tm_shaper_para *para)
+{
+	struct hclge_pri_shapping_cmd *shap_cfg_cmd;
+	struct hclge_desc desc;
+	u32 shapping_para;
+	int ret;
+
+	if (cmd != HCLGE_OPC_TM_PRI_C_SHAPPING &&
+	    cmd != HCLGE_OPC_TM_PRI_P_SHAPPING)
+		return -EINVAL;
+
+	hclge_cmd_setup_basic_desc(&desc, cmd, true);
+	shap_cfg_cmd = (struct hclge_pri_shapping_cmd *)desc.data;
+	shap_cfg_cmd->pri_id = pri_id;
+	ret = hclge_cmd_send(&hdev->hw, &desc, 1);
+	if (ret) {
+		dev_err(&hdev->pdev->dev,
+			"failed to get priority shaper(%#x), ret = %d\n",
+			cmd, ret);
+		return ret;
+	}
+
+	shapping_para = le32_to_cpu(shap_cfg_cmd->pri_shapping_para);
+	para->ir_b = hclge_tm_get_field(shapping_para, IR_B);
+	para->ir_u = hclge_tm_get_field(shapping_para, IR_U);
+	para->ir_s = hclge_tm_get_field(shapping_para, IR_S);
+	para->bs_b = hclge_tm_get_field(shapping_para, BS_B);
+	para->bs_s = hclge_tm_get_field(shapping_para, BS_S);
+	para->flag = shap_cfg_cmd->flag;
+	para->rate = le32_to_cpu(shap_cfg_cmd->pri_rate);
+	return 0;
+}
+
+int hclge_tm_get_q_to_qs_map(struct hclge_dev *hdev, u16 q_id, u16 *qset_id)
+{
+	struct hclge_nq_to_qs_link_cmd *map;
+	struct hclge_desc desc;
+	u16 qs_id_l;
+	u16 qs_id_h;
+	int ret;
+
+	map = (struct hclge_nq_to_qs_link_cmd *)desc.data;
+	hclge_cmd_setup_basic_desc(&desc, HCLGE_OPC_TM_NQ_TO_QS_LINK, true);
+	map->nq_id = cpu_to_le16(q_id);
+	ret = hclge_cmd_send(&hdev->hw, &desc, 1);
+	if (ret) {
+		dev_err(&hdev->pdev->dev,
+			"failed to get queue to qset map, ret = %d\n", ret);
+		return ret;
+	}
+	*qset_id = le16_to_cpu(map->qset_id);
+
+	/* convert qset_id to the following format, drop the vld bit
+	 *            | qs_id_h | vld | qs_id_l |
+	 * qset_id:   | 15 ~ 11 |  10 |  9 ~ 0  |
+	 *             \         \   /         /
+	 *              \         \ /         /
+	 * qset_id: | 15 | 14 ~ 10 |  9 ~ 0  |
+	 */
+	qs_id_l = hnae3_get_field(*qset_id, HCLGE_TM_QS_ID_L_MSK,
+				  HCLGE_TM_QS_ID_L_S);
+	qs_id_h = hnae3_get_field(*qset_id, HCLGE_TM_QS_ID_H_EXT_MSK,
+				  HCLGE_TM_QS_ID_H_EXT_S);
+	*qset_id = 0;
+	hnae3_set_field(*qset_id, HCLGE_TM_QS_ID_L_MSK, HCLGE_TM_QS_ID_L_S,
+			qs_id_l);
+	hnae3_set_field(*qset_id, HCLGE_TM_QS_ID_H_MSK, HCLGE_TM_QS_ID_H_S,
+			qs_id_h);
+	return 0;
+}
+
+int hclge_tm_get_q_to_tc(struct hclge_dev *hdev, u16 q_id, u8 *tc_id)
+{
+#define HCLGE_TM_TC_MASK		0x7
+
+	struct hclge_tqp_tx_queue_tc_cmd *tc;
+	struct hclge_desc desc;
+	int ret;
+
+	tc = (struct hclge_tqp_tx_queue_tc_cmd *)desc.data;
+	hclge_cmd_setup_basic_desc(&desc, HCLGE_OPC_TQP_TX_QUEUE_TC, true);
+	tc->queue_id = cpu_to_le16(q_id);
+	ret = hclge_cmd_send(&hdev->hw, &desc, 1);
+	if (ret) {
+		dev_err(&hdev->pdev->dev,
+			"failed to get queue to tc map, ret = %d\n", ret);
+		return ret;
+	}
+
+	*tc_id = tc->tc_id & HCLGE_TM_TC_MASK;
+	return 0;
+}
+
+int hclge_tm_get_pg_to_pri_map(struct hclge_dev *hdev, u8 pg_id,
+			       u8 *pri_bit_map)
+{
+	struct hclge_pg_to_pri_link_cmd *map;
+	struct hclge_desc desc;
+	int ret;
+
+	hclge_cmd_setup_basic_desc(&desc, HCLGE_OPC_TM_PG_TO_PRI_LINK, true);
+	map = (struct hclge_pg_to_pri_link_cmd *)desc.data;
+	map->pg_id = pg_id;
+	ret = hclge_cmd_send(&hdev->hw, &desc, 1);
+	if (ret) {
+		dev_err(&hdev->pdev->dev,
+			"failed to get pg to pri map, ret = %d\n", ret);
+		return ret;
+	}
+
+	*pri_bit_map = map->pri_bit_map;
+	return 0;
+}
+
+int hclge_tm_get_pg_weight(struct hclge_dev *hdev, u8 pg_id, u8 *weight)
+{
+	struct hclge_pg_weight_cmd *pg_weight_cmd;
+	struct hclge_desc desc;
+	int ret;
+
+	hclge_cmd_setup_basic_desc(&desc, HCLGE_OPC_TM_PG_WEIGHT, true);
+	pg_weight_cmd = (struct hclge_pg_weight_cmd *)desc.data;
+	pg_weight_cmd->pg_id = pg_id;
+	ret = hclge_cmd_send(&hdev->hw, &desc, 1);
+	if (ret) {
+		dev_err(&hdev->pdev->dev,
+			"failed to get pg weight, ret = %d\n", ret);
+		return ret;
+	}
+
+	*weight = pg_weight_cmd->dwrr;
+	return 0;
+}
+
+int hclge_tm_get_pg_sch_mode(struct hclge_dev *hdev, u8 pg_id, u8 *mode)
+{
+	struct hclge_desc desc;
+	int ret;
+
+	hclge_cmd_setup_basic_desc(&desc, HCLGE_OPC_TM_PG_SCH_MODE_CFG, true);
+	desc.data[0] = cpu_to_le32(pg_id);
+	ret = hclge_cmd_send(&hdev->hw, &desc, 1);
+	if (ret) {
+		dev_err(&hdev->pdev->dev,
+			"failed to get pg sch mode, ret = %d\n", ret);
+		return ret;
+	}
+
+	*mode = (u8)le32_to_cpu(desc.data[1]);
+	return 0;
+}
+
+int hclge_tm_get_pg_shaper(struct hclge_dev *hdev, u8 pg_id,
+			   enum hclge_opcode_type cmd,
+			   struct hclge_tm_shaper_para *para)
+{
+	struct hclge_pg_shapping_cmd *shap_cfg_cmd;
+	struct hclge_desc desc;
+	u32 shapping_para;
+	int ret;
+
+	if (cmd != HCLGE_OPC_TM_PG_C_SHAPPING &&
+	    cmd != HCLGE_OPC_TM_PG_P_SHAPPING)
+		return -EINVAL;
+
+	hclge_cmd_setup_basic_desc(&desc, cmd, true);
+	shap_cfg_cmd = (struct hclge_pg_shapping_cmd *)desc.data;
+	shap_cfg_cmd->pg_id = pg_id;
+	ret = hclge_cmd_send(&hdev->hw, &desc, 1);
+	if (ret) {
+		dev_err(&hdev->pdev->dev,
+			"failed to get pg shaper(%#x), ret = %d\n",
+			cmd, ret);
+		return ret;
+	}
+
+	shapping_para = le32_to_cpu(shap_cfg_cmd->pg_shapping_para);
+	para->ir_b = hclge_tm_get_field(shapping_para, IR_B);
+	para->ir_u = hclge_tm_get_field(shapping_para, IR_U);
+	para->ir_s = hclge_tm_get_field(shapping_para, IR_S);
+	para->bs_b = hclge_tm_get_field(shapping_para, BS_B);
+	para->bs_s = hclge_tm_get_field(shapping_para, BS_S);
+	para->flag = shap_cfg_cmd->flag;
+	para->rate = le32_to_cpu(shap_cfg_cmd->pg_rate);
+	return 0;
+}
+
+int hclge_tm_get_port_shaper(struct hclge_dev *hdev,
+			     struct hclge_tm_shaper_para *para)
+{
+	struct hclge_port_shapping_cmd *port_shap_cfg_cmd;
+	struct hclge_desc desc;
+	u32 shapping_para;
+	int ret;
+
+	hclge_cmd_setup_basic_desc(&desc, HCLGE_OPC_TM_PORT_SHAPPING, true);
+	ret = hclge_cmd_send(&hdev->hw, &desc, 1);
+	if (ret) {
+		dev_err(&hdev->pdev->dev,
+			"failed to get port shaper, ret = %d\n", ret);
+		return ret;
+	}
+
+	port_shap_cfg_cmd = (struct hclge_port_shapping_cmd *)desc.data;
+	shapping_para = le32_to_cpu(port_shap_cfg_cmd->port_shapping_para);
+	para->ir_b = hclge_tm_get_field(shapping_para, IR_B);
+	para->ir_u = hclge_tm_get_field(shapping_para, IR_U);
+	para->ir_s = hclge_tm_get_field(shapping_para, IR_S);
+	para->bs_b = hclge_tm_get_field(shapping_para, BS_B);
+	para->bs_s = hclge_tm_get_field(shapping_para, BS_S);
+	para->flag = port_shap_cfg_cmd->flag;
+	para->rate = le32_to_cpu(port_shap_cfg_cmd->port_rate);
+
+	return 0;
+}
+
+int hclge_tm_flush_cfg(struct hclge_dev *hdev, bool enable)
+{
+	struct hclge_desc desc;
+	int ret;
+
+	if (!hnae3_ae_dev_tm_flush_supported(hdev))
+		return 0;
+
+	hclge_cmd_setup_basic_desc(&desc, HCLGE_OPC_TM_FLUSH, false);
+
+	desc.data[0] = cpu_to_le32(enable ? HCLGE_TM_FLUSH_EN_MSK : 0);
+
+	ret = hclge_cmd_send(&hdev->hw, &desc, 1);
+	if (ret) {
+		dev_err(&hdev->pdev->dev,
+			"failed to config tm flush, ret = %d\n", ret);
+		return ret;
+	}
+
+	if (enable)
+		msleep(HCLGE_TM_FLUSH_TIME_MS);
+
+	return ret;
+}
+
+void hclge_reset_tc_config(struct hclge_dev *hdev)
+{
+	struct hclge_vport *vport = &hdev->vport[0];
+	struct hnae3_knic_private_info *kinfo;
+
+	kinfo = &vport->nic.kinfo;
+
+	if (!kinfo->tc_info.mqprio_destroy)
+		return;
+
+	/* clear tc info, including mqprio_destroy and mqprio_active */
+	memset(&kinfo->tc_info, 0, sizeof(kinfo->tc_info));
+	hclge_tm_schd_info_update(hdev, 0);
+	hclge_comm_rss_indir_init_cfg(hdev->ae_dev, &hdev->rss_cfg);
+}
diff --git a/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_tm.h b/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_tm.h
new file mode 100644
index 000000000000..0985916629d3
--- /dev/null
+++ b/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_tm.h
@@ -0,0 +1,281 @@
+/* SPDX-License-Identifier: GPL-2.0+ */
+// Copyright (c) 2016-2017 Hisilicon Limited.
+
+#ifndef __HCLGE_TM_H
+#define __HCLGE_TM_H
+
+#include <linux/types.h>
+
+#include "hnae3.h"
+
+struct hclge_dev;
+struct hclge_vport;
+enum hclge_opcode_type;
+
+/* MAC Pause */
+#define HCLGE_TX_MAC_PAUSE_EN_MSK	BIT(0)
+#define HCLGE_RX_MAC_PAUSE_EN_MSK	BIT(1)
+
+#define HCLGE_TM_PORT_BASE_MODE_MSK	BIT(0)
+
+#define HCLGE_DEFAULT_PAUSE_TRANS_GAP	0x7F
+#define HCLGE_DEFAULT_PAUSE_TRANS_TIME	0xFFFF
+
+/* SP or DWRR */
+#define HCLGE_TM_TX_SCHD_DWRR_MSK	BIT(0)
+#define HCLGE_TM_TX_SCHD_SP_MSK		0xFE
+
+#define HCLGE_ETHER_MAX_RATE	100000
+
+#define HCLGE_TM_PF_MAX_PRI_NUM		8
+#define HCLGE_TM_PF_MAX_QSET_NUM	8
+
+#define HCLGE_DSCP_MAP_TC_BD_NUM	2
+#define HCLGE_DSCP_TC_SHIFT(n)		(((n) & 1) * 4)
+
+#define HCLGE_TM_FLUSH_TIME_MS	10
+#define HCLGE_TM_FLUSH_EN_MSK	BIT(0)
+
+struct hclge_pg_to_pri_link_cmd {
+	u8 pg_id;
+	u8 rsvd1[3];
+	u8 pri_bit_map;
+};
+
+struct hclge_qs_to_pri_link_cmd {
+	__le16 qs_id;
+	__le16 rsvd;
+	u8 priority;
+#define HCLGE_TM_QS_PRI_LINK_VLD_MSK	BIT(0)
+	u8 link_vld;
+};
+
+struct hclge_nq_to_qs_link_cmd {
+	__le16 nq_id;
+	__le16 rsvd;
+#define HCLGE_TM_Q_QS_LINK_VLD_MSK	BIT(10)
+#define HCLGE_TM_QS_ID_L_MSK		GENMASK(9, 0)
+#define HCLGE_TM_QS_ID_L_S		0
+#define HCLGE_TM_QS_ID_H_MSK		GENMASK(14, 10)
+#define HCLGE_TM_QS_ID_H_S		10
+#define HCLGE_TM_QS_ID_H_EXT_S		11
+#define HCLGE_TM_QS_ID_H_EXT_MSK	GENMASK(15, 11)
+	__le16 qset_id;
+};
+
+struct hclge_tqp_tx_queue_tc_cmd {
+	__le16 queue_id;
+	__le16 rsvd;
+	u8 tc_id;
+	u8 rev[3];
+};
+
+struct hclge_pg_weight_cmd {
+	u8 pg_id;
+	u8 dwrr;
+};
+
+struct hclge_priority_weight_cmd {
+	u8 pri_id;
+	u8 dwrr;
+};
+
+struct hclge_pri_sch_mode_cfg_cmd {
+	u8 pri_id;
+	u8 rsvd[3];
+	u8 sch_mode;
+};
+
+struct hclge_qs_sch_mode_cfg_cmd {
+	__le16 qs_id;
+	u8 rsvd[2];
+	u8 sch_mode;
+};
+
+struct hclge_qs_weight_cmd {
+	__le16 qs_id;
+	u8 dwrr;
+};
+
+struct hclge_ets_tc_weight_cmd {
+	u8 tc_weight[HNAE3_MAX_TC];
+	u8 weight_offset;
+	u8 rsvd[15];
+};
+
+#define HCLGE_TM_SHAP_IR_B_MSK  GENMASK(7, 0)
+#define HCLGE_TM_SHAP_IR_B_LSH	0
+#define HCLGE_TM_SHAP_IR_U_MSK  GENMASK(11, 8)
+#define HCLGE_TM_SHAP_IR_U_LSH	8
+#define HCLGE_TM_SHAP_IR_S_MSK  GENMASK(15, 12)
+#define HCLGE_TM_SHAP_IR_S_LSH	12
+#define HCLGE_TM_SHAP_BS_B_MSK  GENMASK(20, 16)
+#define HCLGE_TM_SHAP_BS_B_LSH	16
+#define HCLGE_TM_SHAP_BS_S_MSK  GENMASK(25, 21)
+#define HCLGE_TM_SHAP_BS_S_LSH	21
+
+enum hclge_shap_bucket {
+	HCLGE_TM_SHAP_C_BUCKET = 0,
+	HCLGE_TM_SHAP_P_BUCKET,
+};
+
+/* set bit HCLGE_TM_RATE_VLD to 1 means use 'rate' to config shaping */
+#define HCLGE_TM_RATE_VLD	0
+
+struct hclge_pri_shapping_cmd {
+	u8 pri_id;
+	u8 rsvd[3];
+	__le32 pri_shapping_para;
+	u8 flag;
+	u8 rsvd1[3];
+	__le32 pri_rate;
+};
+
+struct hclge_pg_shapping_cmd {
+	u8 pg_id;
+	u8 rsvd[3];
+	__le32 pg_shapping_para;
+	u8 flag;
+	u8 rsvd1[3];
+	__le32 pg_rate;
+};
+
+struct hclge_qs_shapping_cmd {
+	__le16 qs_id;
+	u8 rsvd[2];
+	__le32 qs_shapping_para;
+	u8 flag;
+	u8 rsvd1[3];
+	__le32 qs_rate;
+};
+
+#define HCLGE_BP_GRP_NUM		32
+#define HCLGE_BP_SUB_GRP_ID_S		0
+#define HCLGE_BP_SUB_GRP_ID_M		GENMASK(4, 0)
+#define HCLGE_BP_GRP_ID_S		5
+#define HCLGE_BP_GRP_ID_M		GENMASK(9, 5)
+
+#define HCLGE_BP_EXT_GRP_NUM		40
+#define HCLGE_BP_EXT_GRP_ID_S		5
+#define HCLGE_BP_EXT_GRP_ID_M		GENMASK(10, 5)
+
+struct hclge_bp_to_qs_map_cmd {
+	u8 tc_id;
+	u8 rsvd[2];
+	u8 qs_group_id;
+	__le32 qs_bit_map;
+	u32 rsvd1;
+};
+
+#define HCLGE_PFC_DISABLE	0
+#define HCLGE_PFC_TX_RX_DISABLE	0
+
+struct hclge_pfc_en_cmd {
+	u8 tx_rx_en_bitmap;
+	u8 pri_en_bitmap;
+};
+
+struct hclge_cfg_pause_param_cmd {
+	u8 mac_addr[ETH_ALEN];
+	u8 pause_trans_gap;
+	u8 rsvd;
+	__le16 pause_trans_time;
+	u8 rsvd1[6];
+	/* extra mac address to do double check for pause frame */
+	u8 mac_addr_extra[ETH_ALEN];
+	u16 rsvd2;
+};
+
+struct hclge_pfc_stats_cmd {
+	__le64 pkt_num[3];
+};
+
+struct hclge_port_shapping_cmd {
+	__le32 port_shapping_para;
+	u8 flag;
+	u8 rsvd[3];
+	__le32 port_rate;
+};
+
+struct hclge_shaper_ir_para {
+	u8 ir_b; /* IR_B parameter of IR shaper */
+	u8 ir_u; /* IR_U parameter of IR shaper */
+	u8 ir_s; /* IR_S parameter of IR shaper */
+};
+
+struct hclge_tm_nodes_cmd {
+	u8 pg_base_id;
+	u8 pri_base_id;
+	__le16 qset_base_id;
+	__le16 queue_base_id;
+	u8 pg_num;
+	u8 pri_num;
+	__le16 qset_num;
+	__le16 queue_num;
+};
+
+struct hclge_tm_shaper_para {
+	u32 rate;
+	u8 ir_b;
+	u8 ir_u;
+	u8 ir_s;
+	u8 bs_b;
+	u8 bs_s;
+	u8 flag;
+};
+
+#define hclge_tm_set_field(dest, string, val) \
+			   hnae3_set_field((dest), \
+			   (HCLGE_TM_SHAP_##string##_MSK), \
+			   (HCLGE_TM_SHAP_##string##_LSH), val)
+#define hclge_tm_get_field(src, string) \
+			hnae3_get_field((src), HCLGE_TM_SHAP_##string##_MSK, \
+					HCLGE_TM_SHAP_##string##_LSH)
+
+int hclge_tm_schd_init(struct hclge_dev *hdev);
+int hclge_tm_vport_map_update(struct hclge_dev *hdev);
+int hclge_pause_setup_hw(struct hclge_dev *hdev, bool init);
+int hclge_tm_schd_setup_hw(struct hclge_dev *hdev);
+void hclge_tm_prio_tc_info_update(struct hclge_dev *hdev, u8 *prio_tc);
+void hclge_tm_schd_info_update(struct hclge_dev *hdev, u8 num_tc);
+void hclge_tm_pfc_info_update(struct hclge_dev *hdev);
+int hclge_tm_dwrr_cfg(struct hclge_dev *hdev);
+int hclge_tm_init_hw(struct hclge_dev *hdev, bool init);
+int hclge_pfc_pause_en_cfg(struct hclge_dev *hdev, u8 tx_rx_bitmap,
+			   u8 pfc_bitmap);
+int hclge_mac_pause_en_cfg(struct hclge_dev *hdev, bool tx, bool rx);
+int hclge_pause_addr_cfg(struct hclge_dev *hdev, const u8 *mac_addr);
+int hclge_mac_pause_setup_hw(struct hclge_dev *hdev);
+void hclge_pfc_rx_stats_get(struct hclge_dev *hdev, u64 *stats);
+void hclge_pfc_tx_stats_get(struct hclge_dev *hdev, u64 *stats);
+int hclge_tm_qs_shaper_cfg(struct hclge_vport *vport, int max_tx_rate);
+int hclge_tm_port_shaper_cfg(struct hclge_dev *hdev);
+int hclge_tm_get_qset_num(struct hclge_dev *hdev, u16 *qset_num);
+int hclge_tm_get_pri_num(struct hclge_dev *hdev, u8 *pri_num);
+int hclge_tm_get_qset_map_pri(struct hclge_dev *hdev, u16 qset_id, u8 *priority,
+			      u8 *link_vld);
+int hclge_tm_get_qset_sch_mode(struct hclge_dev *hdev, u16 qset_id, u8 *mode);
+int hclge_tm_get_qset_weight(struct hclge_dev *hdev, u16 qset_id, u8 *weight);
+int hclge_tm_get_qset_shaper(struct hclge_dev *hdev, u16 qset_id,
+			     struct hclge_tm_shaper_para *para);
+int hclge_tm_get_pri_sch_mode(struct hclge_dev *hdev, u8 pri_id, u8 *mode);
+int hclge_tm_get_pri_weight(struct hclge_dev *hdev, u8 pri_id, u8 *weight);
+int hclge_tm_get_pri_shaper(struct hclge_dev *hdev, u8 pri_id,
+			    enum hclge_opcode_type cmd,
+			    struct hclge_tm_shaper_para *para);
+int hclge_tm_get_q_to_qs_map(struct hclge_dev *hdev, u16 q_id, u16 *qset_id);
+int hclge_tm_get_q_to_tc(struct hclge_dev *hdev, u16 q_id, u8 *tc_id);
+int hclge_tm_get_pg_to_pri_map(struct hclge_dev *hdev, u8 pg_id,
+			       u8 *pri_bit_map);
+int hclge_tm_get_pg_weight(struct hclge_dev *hdev, u8 pg_id, u8 *weight);
+int hclge_tm_get_pg_sch_mode(struct hclge_dev *hdev, u8 pg_id, u8 *mode);
+int hclge_tm_get_pg_shaper(struct hclge_dev *hdev, u8 pg_id,
+			   enum hclge_opcode_type cmd,
+			   struct hclge_tm_shaper_para *para);
+int hclge_tm_get_port_shaper(struct hclge_dev *hdev,
+			     struct hclge_tm_shaper_para *para);
+int hclge_up_to_tc_map(struct hclge_dev *hdev);
+int hclge_dscp_to_tc_map(struct hclge_dev *hdev);
+int hclge_tm_flush_cfg(struct hclge_dev *hdev, bool enable);
+void hclge_reset_tc_config(struct hclge_dev *hdev);
+#endif
diff --git a/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_trace.h b/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_trace.h
new file mode 100644
index 000000000000..7103cf04bffc
--- /dev/null
+++ b/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_trace.h
@@ -0,0 +1,181 @@
+/* SPDX-License-Identifier: GPL-2.0+ */
+/* Copyright (c) 2018-2020 Hisilicon Limited. */
+
+/* This must be outside ifdef _HCLGE_TRACE_H */
+#undef TRACE_SYSTEM
+#define TRACE_SYSTEM hns3
+
+#if !defined(_HCLGE_TRACE_H_) || defined(TRACE_HEADER_MULTI_READ)
+#define _HCLGE_TRACE_H_
+
+#include <linux/tracepoint.h>
+
+#define PF_DESC_LEN	(sizeof(struct hclge_desc) / sizeof(u32))
+#define PF_GET_MBX_LEN	(sizeof(struct hclge_mbx_vf_to_pf_cmd) / sizeof(u32))
+#define PF_SEND_MBX_LEN	(sizeof(struct hclge_mbx_pf_to_vf_cmd) / sizeof(u32))
+
+TRACE_EVENT(hclge_pf_mbx_get,
+	TP_PROTO(
+		struct hclge_dev *hdev,
+		struct hclge_mbx_vf_to_pf_cmd *req),
+	TP_ARGS(hdev, req),
+
+	TP_STRUCT__entry(
+		__field(u8, vfid)
+		__field(u8, code)
+		__field(u8, subcode)
+		__string(pciname, pci_name(hdev->pdev))
+		__string(devname, hdev->vport[0].nic.kinfo.netdev->name)
+		__array(u32, mbx_data, PF_GET_MBX_LEN)
+	),
+
+	TP_fast_assign(
+		__entry->vfid = req->mbx_src_vfid;
+		__entry->code = req->msg.code;
+		__entry->subcode = req->msg.subcode;
+		__assign_str(pciname);
+		__assign_str(devname);
+		memcpy(__entry->mbx_data, req,
+		       sizeof(struct hclge_mbx_vf_to_pf_cmd));
+	),
+
+	TP_printk(
+		"%s %s vfid:%u code:%u subcode:%u data:%s",
+		__get_str(pciname), __get_str(devname), __entry->vfid,
+		__entry->code, __entry->subcode,
+		__print_array(__entry->mbx_data, PF_GET_MBX_LEN, sizeof(u32))
+	)
+);
+
+TRACE_EVENT(hclge_pf_mbx_send,
+	TP_PROTO(
+		struct hclge_dev *hdev,
+		struct hclge_mbx_pf_to_vf_cmd *req),
+	TP_ARGS(hdev, req),
+
+	TP_STRUCT__entry(
+		__field(u8, vfid)
+		__field(u16, code)
+		__string(pciname, pci_name(hdev->pdev))
+		__string(devname, hdev->vport[0].nic.kinfo.netdev->name)
+		__array(u32, mbx_data, PF_SEND_MBX_LEN)
+	),
+
+	TP_fast_assign(
+		__entry->vfid = req->dest_vfid;
+		__entry->code = le16_to_cpu(req->msg.code);
+		__assign_str(pciname);
+		__assign_str(devname);
+		memcpy(__entry->mbx_data, req,
+		       sizeof(struct hclge_mbx_pf_to_vf_cmd));
+	),
+
+	TP_printk(
+		"%s %s vfid:%u code:%u data:%s",
+		__get_str(pciname), __get_str(devname), __entry->vfid,
+		__entry->code,
+		__print_array(__entry->mbx_data, PF_SEND_MBX_LEN, sizeof(u32))
+	)
+);
+
+DECLARE_EVENT_CLASS(hclge_pf_cmd_template,
+		    TP_PROTO(struct hclge_comm_hw *hw,
+			     struct hclge_desc *desc,
+			     int index,
+			     int num),
+		    TP_ARGS(hw, desc, index, num),
+
+		    TP_STRUCT__entry(__field(u16, opcode)
+			__field(u16, flag)
+			__field(u16, retval)
+			__field(u16, rsv)
+			__field(int, index)
+			__field(int, num)
+			__string(pciname, pci_name(hw->cmq.csq.pdev))
+			__array(u32, data, HCLGE_DESC_DATA_LEN)),
+
+		    TP_fast_assign(int i;
+			__entry->opcode = le16_to_cpu(desc->opcode);
+			__entry->flag = le16_to_cpu(desc->flag);
+			__entry->retval = le16_to_cpu(desc->retval);
+			__entry->rsv = le16_to_cpu(desc->rsv);
+			__entry->index = index;
+			__entry->num = num;
+			__assign_str(pciname);
+			for (i = 0; i < HCLGE_DESC_DATA_LEN; i++)
+				__entry->data[i] = le32_to_cpu(desc->data[i]);),
+
+		    TP_printk("%s opcode:0x%04x %d-%d flag:0x%04x retval:0x%04x rsv:0x%04x data:%s",
+			      __get_str(pciname), __entry->opcode,
+			      __entry->index, __entry->num,
+			      __entry->flag, __entry->retval, __entry->rsv,
+			      __print_array(__entry->data,
+					    HCLGE_DESC_DATA_LEN, sizeof(u32)))
+);
+
+DEFINE_EVENT(hclge_pf_cmd_template, hclge_pf_cmd_send,
+	     TP_PROTO(struct hclge_comm_hw *hw,
+		      struct hclge_desc *desc,
+		      int index,
+		      int num),
+	     TP_ARGS(hw, desc, index, num)
+);
+
+DEFINE_EVENT(hclge_pf_cmd_template, hclge_pf_cmd_get,
+	     TP_PROTO(struct hclge_comm_hw *hw,
+		      struct hclge_desc *desc,
+		      int index,
+		      int num),
+	     TP_ARGS(hw, desc, index, num)
+);
+
+DECLARE_EVENT_CLASS(hclge_pf_special_cmd_template,
+		    TP_PROTO(struct hclge_comm_hw *hw,
+			     __le32 *data,
+			     int index,
+			     int num),
+		    TP_ARGS(hw, data, index, num),
+
+		    TP_STRUCT__entry(__field(int, index)
+			__field(int, num)
+			__string(pciname, pci_name(hw->cmq.csq.pdev))
+			__array(u32, data, PF_DESC_LEN)),
+
+		    TP_fast_assign(int i;
+			__entry->index = index;
+			__entry->num = num;
+			__assign_str(pciname);
+			for (i = 0; i < PF_DESC_LEN; i++)
+				__entry->data[i] = le32_to_cpu(data[i]);
+		),
+
+		    TP_printk("%s %d-%d data:%s",
+			      __get_str(pciname),
+			      __entry->index, __entry->num,
+			      __print_array(__entry->data,
+					    PF_DESC_LEN, sizeof(u32)))
+);
+
+DEFINE_EVENT(hclge_pf_special_cmd_template, hclge_pf_special_cmd_send,
+	     TP_PROTO(struct hclge_comm_hw *hw,
+		      __le32 *desc,
+		      int index,
+		      int num),
+	     TP_ARGS(hw, desc, index, num));
+
+DEFINE_EVENT(hclge_pf_special_cmd_template, hclge_pf_special_cmd_get,
+	     TP_PROTO(struct hclge_comm_hw *hw,
+		      __le32 *desc,
+		      int index,
+		      int num),
+	     TP_ARGS(hw, desc, index, num)
+);
+
+#endif /* _HCLGE_TRACE_H_ */
+
+/* This must be outside ifdef _HCLGE_TRACE_H */
+#undef TRACE_INCLUDE_PATH
+#define TRACE_INCLUDE_PATH .
+#undef TRACE_INCLUDE_FILE
+#define TRACE_INCLUDE_FILE hclge_trace
+#include <trace/define_trace.h>
diff --git a/drivers/net/ethernet/hisilicon/hns3/hns3vf/hclgevf_cmd.h b/drivers/net/ethernet/hisilicon/hns3/hns3vf/hclgevf_cmd.h
new file mode 100644
index 000000000000..537b887fa0a2
--- /dev/null
+++ b/drivers/net/ethernet/hisilicon/hns3/hns3vf/hclgevf_cmd.h
@@ -0,0 +1,136 @@
+/* SPDX-License-Identifier: GPL-2.0+ */
+/* Copyright (c) 2016-2017 Hisilicon Limited. */
+
+#ifndef __HCLGEVF_CMD_H
+#define __HCLGEVF_CMD_H
+#include <linux/io.h>
+#include <linux/types.h>
+#include "hnae3.h"
+#include "hclge_comm_cmd.h"
+
+#define HCLGEVF_CMDQ_RX_INVLD_B		0
+#define HCLGEVF_CMDQ_RX_OUTVLD_B	1
+
+struct hclgevf_hw;
+struct hclgevf_dev;
+
+#define HCLGEVF_SYNC_RX_RING_HEAD_EN_B	4
+
+#define HCLGEVF_TQP_REG_OFFSET		0x80000
+#define HCLGEVF_TQP_REG_SIZE		0x200
+
+#define HCLGEVF_TQP_MAX_SIZE_DEV_V2	1024
+#define HCLGEVF_TQP_EXT_REG_OFFSET	0x100
+
+struct hclgevf_tqp_map {
+	__le16 tqp_id;	/* Absolute tqp id for in this pf */
+	u8 tqp_vf; /* VF id */
+#define HCLGEVF_TQP_MAP_TYPE_PF		0
+#define HCLGEVF_TQP_MAP_TYPE_VF		1
+#define HCLGEVF_TQP_MAP_TYPE_B		0
+#define HCLGEVF_TQP_MAP_EN_B		1
+	u8 tqp_flag;	/* Indicate it's pf or vf tqp */
+	__le16 tqp_vid; /* Virtual id in this pf/vf */
+	u8 rsv[18];
+};
+
+#define HCLGEVF_VECTOR_ELEMENTS_PER_CMD	10
+
+enum hclgevf_int_type {
+	HCLGEVF_INT_TX = 0,
+	HCLGEVF_INT_RX,
+	HCLGEVF_INT_EVENT,
+};
+
+struct hclgevf_ctrl_vector_chain {
+	u8 int_vector_id;
+	u8 int_cause_num;
+#define HCLGEVF_INT_TYPE_S	0
+#define HCLGEVF_INT_TYPE_M	0x3
+#define HCLGEVF_TQP_ID_S	2
+#define HCLGEVF_TQP_ID_M	(0x3fff << HCLGEVF_TQP_ID_S)
+	__le16 tqp_type_and_id[HCLGEVF_VECTOR_ELEMENTS_PER_CMD];
+	u8 vfid;
+	u8 resv;
+};
+
+#define HCLGEVF_MSIX_OFT_ROCEE_S       0
+#define HCLGEVF_MSIX_OFT_ROCEE_M       (0xffff << HCLGEVF_MSIX_OFT_ROCEE_S)
+#define HCLGEVF_VEC_NUM_S              0
+#define HCLGEVF_VEC_NUM_M              (0xff << HCLGEVF_VEC_NUM_S)
+struct hclgevf_query_res_cmd {
+	__le16 tqp_num;
+	__le16 reserved;
+	__le16 msixcap_localid_ba_nic;
+	__le16 msixcap_localid_ba_rocee;
+	__le16 vf_intr_vector_number;
+	__le16 rsv[7];
+};
+
+#define HCLGEVF_GRO_EN_B               0
+struct hclgevf_cfg_gro_status_cmd {
+	u8 gro_en;
+	u8 rsv[23];
+};
+
+#define HCLGEVF_LINK_STS_B	0
+#define HCLGEVF_LINK_STATUS	BIT(HCLGEVF_LINK_STS_B)
+struct hclgevf_link_status_cmd {
+	u8 status;
+	u8 rsv[23];
+};
+
+#define HCLGEVF_RING_ID_MASK	0x3ff
+#define HCLGEVF_TQP_ENABLE_B	0
+
+struct hclgevf_cfg_com_tqp_queue_cmd {
+	__le16 tqp_id;
+	__le16 stream_id;
+	u8 enable;
+	u8 rsv[19];
+};
+
+struct hclgevf_cfg_tx_queue_pointer_cmd {
+	__le16 tqp_id;
+	__le16 tx_tail;
+	__le16 tx_head;
+	__le16 fbd_num;
+	__le16 ring_offset;
+	u8 rsv[14];
+};
+
+/* this bit indicates that the driver is ready for hardware reset */
+#define HCLGEVF_NIC_SW_RST_RDY_B	16
+#define HCLGEVF_NIC_SW_RST_RDY		BIT(HCLGEVF_NIC_SW_RST_RDY_B)
+
+#define HCLGEVF_NIC_CMQ_DESC_NUM	1024
+#define HCLGEVF_NIC_CMQ_DESC_NUM_S	3
+
+#define HCLGEVF_QUERY_DEV_SPECS_BD_NUM		4
+
+#define hclgevf_cmd_setup_basic_desc(desc, opcode, is_read) \
+	hclge_comm_cmd_setup_basic_desc(desc, opcode, is_read)
+
+struct hclgevf_dev_specs_0_cmd {
+	__le32 rsv0;
+	__le32 mac_entry_num;
+	__le32 mng_entry_num;
+	__le16 rss_ind_tbl_size;
+	__le16 rss_key_size;
+	__le16 int_ql_max;
+	u8 max_non_tso_bd_num;
+	u8 rsv1[5];
+};
+
+#define HCLGEVF_DEF_MAX_INT_GL		0x1FE0U
+
+struct hclgevf_dev_specs_1_cmd {
+	__le16 max_frm_size;
+	__le16 rsv0;
+	__le16 max_int_gl;
+	u8 rsv1[18];
+};
+
+int hclgevf_cmd_send(struct hclgevf_hw *hw, struct hclge_desc *desc, int num);
+void hclgevf_arq_init(struct hclgevf_dev *hdev);
+#endif
diff --git a/drivers/net/ethernet/hisilicon/hns3/hns3vf/hclgevf_devlink.c b/drivers/net/ethernet/hisilicon/hns3/hns3vf/hclgevf_devlink.c
new file mode 100644
index 000000000000..1b535142c65a
--- /dev/null
+++ b/drivers/net/ethernet/hisilicon/hns3/hns3vf/hclgevf_devlink.c
@@ -0,0 +1,130 @@
+// SPDX-License-Identifier: GPL-2.0+
+/* Copyright (c) 2021 Hisilicon Limited. */
+
+#include <net/devlink.h>
+
+#include "hclgevf_devlink.h"
+
+static int hclgevf_devlink_info_get(struct devlink *devlink,
+				    struct devlink_info_req *req,
+				    struct netlink_ext_ack *extack)
+{
+#define	HCLGEVF_DEVLINK_FW_STRING_LEN	32
+	struct hclgevf_devlink_priv *priv = devlink_priv(devlink);
+	char version_str[HCLGEVF_DEVLINK_FW_STRING_LEN];
+	struct hclgevf_dev *hdev = priv->hdev;
+
+	snprintf(version_str, sizeof(version_str), "%lu.%lu.%lu.%lu",
+		 hnae3_get_field(hdev->fw_version, HNAE3_FW_VERSION_BYTE3_MASK,
+				 HNAE3_FW_VERSION_BYTE3_SHIFT),
+		 hnae3_get_field(hdev->fw_version, HNAE3_FW_VERSION_BYTE2_MASK,
+				 HNAE3_FW_VERSION_BYTE2_SHIFT),
+		 hnae3_get_field(hdev->fw_version, HNAE3_FW_VERSION_BYTE1_MASK,
+				 HNAE3_FW_VERSION_BYTE1_SHIFT),
+		 hnae3_get_field(hdev->fw_version, HNAE3_FW_VERSION_BYTE0_MASK,
+				 HNAE3_FW_VERSION_BYTE0_SHIFT));
+
+	return devlink_info_version_running_put(req,
+						DEVLINK_INFO_VERSION_GENERIC_FW,
+						version_str);
+}
+
+static int hclgevf_devlink_reload_down(struct devlink *devlink,
+				       bool netns_change,
+				       enum devlink_reload_action action,
+				       enum devlink_reload_limit limit,
+				       struct netlink_ext_ack *extack)
+{
+	struct hclgevf_devlink_priv *priv = devlink_priv(devlink);
+	struct hclgevf_dev *hdev = priv->hdev;
+	struct hnae3_handle *h = &hdev->nic;
+	struct pci_dev *pdev = hdev->pdev;
+	int ret;
+
+	if (test_bit(HCLGEVF_STATE_RST_HANDLING, &hdev->state)) {
+		dev_err(&pdev->dev, "reset is handling\n");
+		return -EBUSY;
+	}
+
+	switch (action) {
+	case DEVLINK_RELOAD_ACTION_DRIVER_REINIT:
+		rtnl_lock();
+		ret = hdev->nic_client->ops->reset_notify(h, HNAE3_DOWN_CLIENT);
+		if (ret) {
+			rtnl_unlock();
+			return ret;
+		}
+
+		ret = hdev->nic_client->ops->reset_notify(h,
+							  HNAE3_UNINIT_CLIENT);
+		rtnl_unlock();
+		return ret;
+	default:
+		return -EOPNOTSUPP;
+	}
+}
+
+static int hclgevf_devlink_reload_up(struct devlink *devlink,
+				     enum devlink_reload_action action,
+				     enum devlink_reload_limit limit,
+				     u32 *actions_performed,
+				     struct netlink_ext_ack *extack)
+{
+	struct hclgevf_devlink_priv *priv = devlink_priv(devlink);
+	struct hclgevf_dev *hdev = priv->hdev;
+	struct hnae3_handle *h = &hdev->nic;
+	int ret;
+
+	*actions_performed = BIT(action);
+	switch (action) {
+	case DEVLINK_RELOAD_ACTION_DRIVER_REINIT:
+		rtnl_lock();
+		ret = hdev->nic_client->ops->reset_notify(h, HNAE3_INIT_CLIENT);
+		if (ret) {
+			rtnl_unlock();
+			return ret;
+		}
+
+		ret = hdev->nic_client->ops->reset_notify(h, HNAE3_UP_CLIENT);
+		rtnl_unlock();
+		return ret;
+	default:
+		return -EOPNOTSUPP;
+	}
+}
+
+static const struct devlink_ops hclgevf_devlink_ops = {
+	.info_get = hclgevf_devlink_info_get,
+	.reload_actions = BIT(DEVLINK_RELOAD_ACTION_DRIVER_REINIT),
+	.reload_down = hclgevf_devlink_reload_down,
+	.reload_up = hclgevf_devlink_reload_up,
+};
+
+int hclgevf_devlink_init(struct hclgevf_dev *hdev)
+{
+	struct pci_dev *pdev = hdev->pdev;
+	struct hclgevf_devlink_priv *priv;
+	struct devlink *devlink;
+
+	devlink =
+		devlink_alloc(&hclgevf_devlink_ops,
+			      sizeof(struct hclgevf_devlink_priv), &pdev->dev);
+	if (!devlink)
+		return -ENOMEM;
+
+	priv = devlink_priv(devlink);
+	priv->hdev = hdev;
+	hdev->devlink = devlink;
+
+	devlink_register(devlink);
+	return 0;
+}
+
+void hclgevf_devlink_uninit(struct hclgevf_dev *hdev)
+{
+	struct devlink *devlink = hdev->devlink;
+
+	devlink_unregister(devlink);
+
+	devlink_free(devlink);
+}
diff --git a/drivers/net/ethernet/hisilicon/hns3/hns3vf/hclgevf_devlink.h b/drivers/net/ethernet/hisilicon/hns3/hns3vf/hclgevf_devlink.h
new file mode 100644
index 000000000000..e09ea3d8a963
--- /dev/null
+++ b/drivers/net/ethernet/hisilicon/hns3/hns3vf/hclgevf_devlink.h
@@ -0,0 +1,15 @@
+/* SPDX-License-Identifier: GPL-2.0+ */
+/* Copyright (c) 2021 Hisilicon Limited. */
+
+#ifndef __HCLGEVF_DEVLINK_H
+#define __HCLGEVF_DEVLINK_H
+
+#include "hclgevf_main.h"
+
+struct hclgevf_devlink_priv {
+	struct hclgevf_dev *hdev;
+};
+
+int hclgevf_devlink_init(struct hclgevf_dev *hdev);
+void hclgevf_devlink_uninit(struct hclgevf_dev *hdev);
+#endif
diff --git a/drivers/net/ethernet/hisilicon/hns3/hns3vf/hclgevf_main.c b/drivers/net/ethernet/hisilicon/hns3/hns3vf/hclgevf_main.c
new file mode 100644
index 000000000000..8fcf220a120d
--- /dev/null
+++ b/drivers/net/ethernet/hisilicon/hns3/hns3vf/hclgevf_main.c
@@ -0,0 +1,3433 @@
+// SPDX-License-Identifier: GPL-2.0+
+// Copyright (c) 2016-2017 Hisilicon Limited.
+
+#include <linux/etherdevice.h>
+#include <linux/iopoll.h>
+#include <net/rtnetlink.h>
+#include "hclgevf_cmd.h"
+#include "hclgevf_main.h"
+#include "hclgevf_regs.h"
+#include "hclge_mbx.h"
+#include "hnae3.h"
+#include "hclgevf_devlink.h"
+#include "hclge_comm_rss.h"
+#include "hclgevf_trace.h"
+
+#define HCLGEVF_NAME	"hclgevf"
+
+#define HCLGEVF_RESET_MAX_FAIL_CNT	5
+
+static int hclgevf_reset_hdev(struct hclgevf_dev *hdev);
+static void hclgevf_task_schedule(struct hclgevf_dev *hdev,
+				  unsigned long delay);
+
+static struct hnae3_ae_algo ae_algovf;
+
+static struct workqueue_struct *hclgevf_wq;
+
+static const struct pci_device_id ae_algovf_pci_tbl[] = {
+	{PCI_VDEVICE(HUAWEI, HNAE3_DEV_ID_VF), 0},
+	{PCI_VDEVICE(HUAWEI, HNAE3_DEV_ID_RDMA_DCB_PFC_VF),
+	 HNAE3_DEV_SUPPORT_ROCE_DCB_BITS},
+	/* required last entry */
+	{0, }
+};
+
+MODULE_DEVICE_TABLE(pci, ae_algovf_pci_tbl);
+
+/* hclgevf_cmd_send - send command to command queue
+ * @hw: pointer to the hw struct
+ * @desc: prefilled descriptor for describing the command
+ * @num : the number of descriptors to be sent
+ *
+ * This is the main send command for command queue, it
+ * sends the queue, cleans the queue, etc
+ */
+int hclgevf_cmd_send(struct hclgevf_hw *hw, struct hclge_desc *desc, int num)
+{
+	return hclge_comm_cmd_send(&hw->hw, desc, num);
+}
+
+static void hclgevf_trace_cmd_send(struct hclge_comm_hw *hw, struct hclge_desc *desc,
+				   int num, bool is_special)
+{
+	int i;
+
+	trace_hclge_vf_cmd_send(hw, desc, 0, num);
+
+	if (is_special)
+		return;
+
+	for (i = 1; i < num; i++)
+		trace_hclge_vf_cmd_send(hw, &desc[i], i, num);
+}
+
+static void hclgevf_trace_cmd_get(struct hclge_comm_hw *hw, struct hclge_desc *desc,
+				  int num, bool is_special)
+{
+	int i;
+
+	if (!HCLGE_COMM_SEND_SYNC(le16_to_cpu(desc->flag)))
+		return;
+
+	trace_hclge_vf_cmd_get(hw, desc, 0, num);
+
+	if (is_special)
+		return;
+
+	for (i = 1; i < num; i++)
+		trace_hclge_vf_cmd_get(hw, &desc[i], i, num);
+}
+
+static const struct hclge_comm_cmq_ops hclgevf_cmq_ops = {
+	.trace_cmd_send = hclgevf_trace_cmd_send,
+	.trace_cmd_get = hclgevf_trace_cmd_get,
+};
+
+void hclgevf_arq_init(struct hclgevf_dev *hdev)
+{
+	struct hclge_comm_cmq *cmdq = &hdev->hw.hw.cmq;
+
+	spin_lock(&cmdq->crq.lock);
+	/* initialize the pointers of async rx queue of mailbox */
+	hdev->arq.hdev = hdev;
+	hdev->arq.head = 0;
+	hdev->arq.tail = 0;
+	atomic_set(&hdev->arq.count, 0);
+	spin_unlock(&cmdq->crq.lock);
+}
+
+struct hclgevf_dev *hclgevf_ae_get_hdev(struct hnae3_handle *handle)
+{
+	if (!handle->client)
+		return container_of(handle, struct hclgevf_dev, nic);
+	else if (handle->client->type == HNAE3_CLIENT_ROCE)
+		return container_of(handle, struct hclgevf_dev, roce);
+	else
+		return container_of(handle, struct hclgevf_dev, nic);
+}
+
+static void hclgevf_update_stats(struct hnae3_handle *handle)
+{
+	struct hclgevf_dev *hdev = hclgevf_ae_get_hdev(handle);
+	int status;
+
+	status = hclge_comm_tqps_update_stats(handle, &hdev->hw.hw);
+	if (status)
+		dev_err(&hdev->pdev->dev,
+			"VF update of TQPS stats fail, status = %d.\n",
+			status);
+}
+
+static int hclgevf_get_sset_count(struct hnae3_handle *handle, int strset)
+{
+	if (strset == ETH_SS_TEST)
+		return -EOPNOTSUPP;
+	else if (strset == ETH_SS_STATS)
+		return hclge_comm_tqps_get_sset_count(handle);
+
+	return 0;
+}
+
+static void hclgevf_get_strings(struct hnae3_handle *handle, u32 strset,
+				u8 **data)
+{
+	if (strset == ETH_SS_STATS)
+		hclge_comm_tqps_get_strings(handle, data);
+}
+
+static void hclgevf_get_stats(struct hnae3_handle *handle, u64 *data)
+{
+	hclge_comm_tqps_get_stats(handle, data);
+}
+
+static void hclgevf_build_send_msg(struct hclge_vf_to_pf_msg *msg, u8 code,
+				   u8 subcode)
+{
+	if (msg) {
+		memset(msg, 0, sizeof(struct hclge_vf_to_pf_msg));
+		msg->code = code;
+		msg->subcode = subcode;
+	}
+}
+
+static int hclgevf_get_basic_info(struct hclgevf_dev *hdev)
+{
+	struct hnae3_ae_dev *ae_dev = hdev->ae_dev;
+	u8 resp_msg[HCLGE_MBX_MAX_RESP_DATA_SIZE];
+	struct hclge_basic_info *basic_info;
+	struct hclge_vf_to_pf_msg send_msg;
+	unsigned long caps;
+	int status;
+
+	hclgevf_build_send_msg(&send_msg, HCLGE_MBX_GET_BASIC_INFO, 0);
+	status = hclgevf_send_mbx_msg(hdev, &send_msg, true, resp_msg,
+				      sizeof(resp_msg));
+	if (status) {
+		dev_err(&hdev->pdev->dev,
+			"failed to get basic info from pf, ret = %d", status);
+		return status;
+	}
+
+	basic_info = (struct hclge_basic_info *)resp_msg;
+
+	hdev->hw_tc_map = basic_info->hw_tc_map;
+	hdev->mbx_api_version = le16_to_cpu(basic_info->mbx_api_version);
+	caps = le32_to_cpu(basic_info->pf_caps);
+	if (test_bit(HNAE3_PF_SUPPORT_VLAN_FLTR_MDF_B, &caps))
+		set_bit(HNAE3_DEV_SUPPORT_VLAN_FLTR_MDF_B, ae_dev->caps);
+
+	return 0;
+}
+
+static int hclgevf_get_port_base_vlan_filter_state(struct hclgevf_dev *hdev)
+{
+	struct hnae3_handle *nic = &hdev->nic;
+	struct hclge_vf_to_pf_msg send_msg;
+	u8 resp_msg;
+	int ret;
+
+	hclgevf_build_send_msg(&send_msg, HCLGE_MBX_SET_VLAN,
+			       HCLGE_MBX_GET_PORT_BASE_VLAN_STATE);
+	ret = hclgevf_send_mbx_msg(hdev, &send_msg, true, &resp_msg,
+				   sizeof(u8));
+	if (ret) {
+		dev_err(&hdev->pdev->dev,
+			"VF request to get port based vlan state failed %d",
+			ret);
+		return ret;
+	}
+
+	nic->port_base_vlan_state = resp_msg;
+
+	return 0;
+}
+
+static int hclgevf_get_queue_info(struct hclgevf_dev *hdev)
+{
+#define HCLGEVF_TQPS_RSS_INFO_LEN	6
+
+	struct hclge_mbx_vf_queue_info *queue_info;
+	u8 resp_msg[HCLGEVF_TQPS_RSS_INFO_LEN];
+	struct hclge_vf_to_pf_msg send_msg;
+	int status;
+
+	hclgevf_build_send_msg(&send_msg, HCLGE_MBX_GET_QINFO, 0);
+	status = hclgevf_send_mbx_msg(hdev, &send_msg, true, resp_msg,
+				      HCLGEVF_TQPS_RSS_INFO_LEN);
+	if (status) {
+		dev_err(&hdev->pdev->dev,
+			"VF request to get tqp info from PF failed %d",
+			status);
+		return status;
+	}
+
+	queue_info = (struct hclge_mbx_vf_queue_info *)resp_msg;
+	hdev->num_tqps = le16_to_cpu(queue_info->num_tqps);
+	hdev->rss_size_max = le16_to_cpu(queue_info->rss_size);
+	hdev->rx_buf_len = le16_to_cpu(queue_info->rx_buf_len);
+
+	return 0;
+}
+
+static int hclgevf_get_queue_depth(struct hclgevf_dev *hdev)
+{
+#define HCLGEVF_TQPS_DEPTH_INFO_LEN	4
+
+	struct hclge_mbx_vf_queue_depth *queue_depth;
+	u8 resp_msg[HCLGEVF_TQPS_DEPTH_INFO_LEN];
+	struct hclge_vf_to_pf_msg send_msg;
+	int ret;
+
+	hclgevf_build_send_msg(&send_msg, HCLGE_MBX_GET_QDEPTH, 0);
+	ret = hclgevf_send_mbx_msg(hdev, &send_msg, true, resp_msg,
+				   HCLGEVF_TQPS_DEPTH_INFO_LEN);
+	if (ret) {
+		dev_err(&hdev->pdev->dev,
+			"VF request to get tqp depth info from PF failed %d",
+			ret);
+		return ret;
+	}
+
+	queue_depth = (struct hclge_mbx_vf_queue_depth *)resp_msg;
+	hdev->num_tx_desc = le16_to_cpu(queue_depth->num_tx_desc);
+	hdev->num_rx_desc = le16_to_cpu(queue_depth->num_rx_desc);
+
+	return 0;
+}
+
+static u16 hclgevf_get_qid_global(struct hnae3_handle *handle, u16 queue_id)
+{
+	struct hclgevf_dev *hdev = hclgevf_ae_get_hdev(handle);
+	struct hclge_vf_to_pf_msg send_msg;
+	u16 qid_in_pf = 0;
+	u8 resp_data[2];
+	int ret;
+
+	hclgevf_build_send_msg(&send_msg, HCLGE_MBX_GET_QID_IN_PF, 0);
+	*(__le16 *)send_msg.data = cpu_to_le16(queue_id);
+	ret = hclgevf_send_mbx_msg(hdev, &send_msg, true, resp_data,
+				   sizeof(resp_data));
+	if (!ret)
+		qid_in_pf = le16_to_cpu(*(__le16 *)resp_data);
+
+	return qid_in_pf;
+}
+
+static int hclgevf_get_pf_media_type(struct hclgevf_dev *hdev)
+{
+	struct hclge_vf_to_pf_msg send_msg;
+	u8 resp_msg[2];
+	int ret;
+
+	hclgevf_build_send_msg(&send_msg, HCLGE_MBX_GET_MEDIA_TYPE, 0);
+	ret = hclgevf_send_mbx_msg(hdev, &send_msg, true, resp_msg,
+				   sizeof(resp_msg));
+	if (ret) {
+		dev_err(&hdev->pdev->dev,
+			"VF request to get the pf port media type failed %d",
+			ret);
+		return ret;
+	}
+
+	hdev->hw.mac.media_type = resp_msg[0];
+	hdev->hw.mac.module_type = resp_msg[1];
+
+	return 0;
+}
+
+static int hclgevf_alloc_tqps(struct hclgevf_dev *hdev)
+{
+	struct hnae3_ae_dev *ae_dev = pci_get_drvdata(hdev->pdev);
+	struct hclge_comm_tqp *tqp;
+	int i;
+
+	hdev->htqp = devm_kcalloc(&hdev->pdev->dev, hdev->num_tqps,
+				  sizeof(struct hclge_comm_tqp), GFP_KERNEL);
+	if (!hdev->htqp)
+		return -ENOMEM;
+
+	tqp = hdev->htqp;
+
+	for (i = 0; i < hdev->num_tqps; i++) {
+		tqp->dev = &hdev->pdev->dev;
+		tqp->index = i;
+
+		tqp->q.ae_algo = &ae_algovf;
+		tqp->q.buf_size = hdev->rx_buf_len;
+		tqp->q.tx_desc_num = hdev->num_tx_desc;
+		tqp->q.rx_desc_num = hdev->num_rx_desc;
+
+		/* need an extended offset to configure queues >=
+		 * HCLGEVF_TQP_MAX_SIZE_DEV_V2.
+		 */
+		if (i < HCLGEVF_TQP_MAX_SIZE_DEV_V2)
+			tqp->q.io_base = hdev->hw.hw.io_base +
+					 HCLGEVF_TQP_REG_OFFSET +
+					 i * HCLGEVF_TQP_REG_SIZE;
+		else
+			tqp->q.io_base = hdev->hw.hw.io_base +
+					 HCLGEVF_TQP_REG_OFFSET +
+					 HCLGEVF_TQP_EXT_REG_OFFSET +
+					 (i - HCLGEVF_TQP_MAX_SIZE_DEV_V2) *
+					 HCLGEVF_TQP_REG_SIZE;
+
+		/* when device supports tx push and has device memory,
+		 * the queue can execute push mode or doorbell mode on
+		 * device memory.
+		 */
+		if (test_bit(HNAE3_DEV_SUPPORT_TX_PUSH_B, ae_dev->caps))
+			tqp->q.mem_base = hdev->hw.hw.mem_base +
+					  HCLGEVF_TQP_MEM_OFFSET(hdev, i);
+
+		tqp++;
+	}
+
+	return 0;
+}
+
+static int hclgevf_knic_setup(struct hclgevf_dev *hdev)
+{
+	struct hnae3_handle *nic = &hdev->nic;
+	struct hnae3_knic_private_info *kinfo;
+	u16 new_tqps = hdev->num_tqps;
+	unsigned int i;
+	u8 num_tc = 0;
+
+	kinfo = &nic->kinfo;
+	kinfo->num_tx_desc = hdev->num_tx_desc;
+	kinfo->num_rx_desc = hdev->num_rx_desc;
+	kinfo->rx_buf_len = hdev->rx_buf_len;
+	for (i = 0; i < HCLGE_COMM_MAX_TC_NUM; i++)
+		if (hdev->hw_tc_map & BIT(i))
+			num_tc++;
+
+	num_tc = num_tc ? num_tc : 1;
+	kinfo->tc_info.num_tc = num_tc;
+	kinfo->rss_size = min_t(u16, hdev->rss_size_max, new_tqps / num_tc);
+	new_tqps = kinfo->rss_size * num_tc;
+	kinfo->num_tqps = min(new_tqps, hdev->num_tqps);
+
+	kinfo->tqp = devm_kcalloc(&hdev->pdev->dev, kinfo->num_tqps,
+				  sizeof(struct hnae3_queue *), GFP_KERNEL);
+	if (!kinfo->tqp)
+		return -ENOMEM;
+
+	for (i = 0; i < kinfo->num_tqps; i++) {
+		hdev->htqp[i].q.handle = &hdev->nic;
+		hdev->htqp[i].q.tqp_index = i;
+		kinfo->tqp[i] = &hdev->htqp[i].q;
+	}
+
+	/* after init the max rss_size and tqps, adjust the default tqp numbers
+	 * and rss size with the actual vector numbers
+	 */
+	kinfo->num_tqps = min_t(u16, hdev->num_nic_msix - 1, kinfo->num_tqps);
+	kinfo->rss_size = min_t(u16, kinfo->num_tqps / num_tc,
+				kinfo->rss_size);
+
+	return 0;
+}
+
+static void hclgevf_request_link_info(struct hclgevf_dev *hdev)
+{
+	struct hclge_vf_to_pf_msg send_msg;
+	int status;
+
+	hclgevf_build_send_msg(&send_msg, HCLGE_MBX_GET_LINK_STATUS, 0);
+	status = hclgevf_send_mbx_msg(hdev, &send_msg, false, NULL, 0);
+	if (status)
+		dev_err(&hdev->pdev->dev,
+			"VF failed to fetch link status(%d) from PF", status);
+}
+
+void hclgevf_update_link_status(struct hclgevf_dev *hdev, int link_state)
+{
+	struct hnae3_handle *rhandle = &hdev->roce;
+	struct hnae3_handle *handle = &hdev->nic;
+	struct hnae3_client *rclient;
+	struct hnae3_client *client;
+
+	if (test_and_set_bit(HCLGEVF_STATE_LINK_UPDATING, &hdev->state))
+		return;
+
+	client = handle->client;
+	rclient = hdev->roce_client;
+
+	link_state =
+		test_bit(HCLGEVF_STATE_DOWN, &hdev->state) ? 0 : link_state;
+	if (link_state != hdev->hw.mac.link) {
+		hdev->hw.mac.link = link_state;
+		client->ops->link_status_change(handle, !!link_state);
+		if (rclient && rclient->ops->link_status_change)
+			rclient->ops->link_status_change(rhandle, !!link_state);
+	}
+
+	clear_bit(HCLGEVF_STATE_LINK_UPDATING, &hdev->state);
+}
+
+static void hclgevf_update_link_mode(struct hclgevf_dev *hdev)
+{
+#define HCLGEVF_ADVERTISING	0
+#define HCLGEVF_SUPPORTED	1
+
+	struct hclge_vf_to_pf_msg send_msg;
+
+	hclgevf_build_send_msg(&send_msg, HCLGE_MBX_GET_LINK_MODE, 0);
+	send_msg.data[0] = HCLGEVF_ADVERTISING;
+	hclgevf_send_mbx_msg(hdev, &send_msg, false, NULL, 0);
+	send_msg.data[0] = HCLGEVF_SUPPORTED;
+	hclgevf_send_mbx_msg(hdev, &send_msg, false, NULL, 0);
+}
+
+static int hclgevf_set_handle_info(struct hclgevf_dev *hdev)
+{
+	struct hnae3_handle *nic = &hdev->nic;
+	int ret;
+
+	nic->ae_algo = &ae_algovf;
+	nic->pdev = hdev->pdev;
+	bitmap_copy(nic->numa_node_mask.bits, hdev->numa_node_mask.bits,
+		    MAX_NUMNODES);
+	nic->flags |= HNAE3_SUPPORT_VF;
+	nic->kinfo.io_base = hdev->hw.hw.io_base;
+
+	ret = hclgevf_knic_setup(hdev);
+	if (ret)
+		dev_err(&hdev->pdev->dev, "VF knic setup failed %d\n",
+			ret);
+	return ret;
+}
+
+static void hclgevf_free_vector(struct hclgevf_dev *hdev, int vector_id)
+{
+	if (hdev->vector_status[vector_id] == HCLGEVF_INVALID_VPORT) {
+		dev_warn(&hdev->pdev->dev,
+			 "vector(vector_id %d) has been freed.\n", vector_id);
+		return;
+	}
+
+	hdev->vector_status[vector_id] = HCLGEVF_INVALID_VPORT;
+	hdev->num_msi_left += 1;
+	hdev->num_msi_used -= 1;
+}
+
+static int hclgevf_get_vector(struct hnae3_handle *handle, u16 vector_num,
+			      struct hnae3_vector_info *vector_info)
+{
+	struct hclgevf_dev *hdev = hclgevf_ae_get_hdev(handle);
+	struct hnae3_vector_info *vector = vector_info;
+	int alloc = 0;
+	int i, j;
+
+	vector_num = min_t(u16, hdev->num_nic_msix - 1, vector_num);
+	vector_num = min(hdev->num_msi_left, vector_num);
+
+	for (j = 0; j < vector_num; j++) {
+		for (i = HCLGEVF_MISC_VECTOR_NUM + 1; i < hdev->num_msi; i++) {
+			if (hdev->vector_status[i] == HCLGEVF_INVALID_VPORT) {
+				vector->vector = pci_irq_vector(hdev->pdev, i);
+				vector->io_addr = hdev->hw.hw.io_base +
+					HCLGEVF_VECTOR_REG_BASE +
+					(i - 1) * HCLGEVF_VECTOR_REG_OFFSET;
+				hdev->vector_status[i] = 0;
+				hdev->vector_irq[i] = vector->vector;
+
+				vector++;
+				alloc++;
+
+				break;
+			}
+		}
+	}
+	hdev->num_msi_left -= alloc;
+	hdev->num_msi_used += alloc;
+
+	return alloc;
+}
+
+static int hclgevf_get_vector_index(struct hclgevf_dev *hdev, int vector)
+{
+	int i;
+
+	for (i = 0; i < hdev->num_msi; i++)
+		if (vector == hdev->vector_irq[i])
+			return i;
+
+	return -EINVAL;
+}
+
+/* for revision 0x20, vf shared the same rss config with pf */
+static int hclgevf_get_rss_hash_key(struct hclgevf_dev *hdev)
+{
+#define HCLGEVF_RSS_MBX_RESP_LEN	8
+	struct hclge_comm_rss_cfg *rss_cfg = &hdev->rss_cfg;
+	u8 resp_msg[HCLGEVF_RSS_MBX_RESP_LEN];
+	struct hclge_vf_to_pf_msg send_msg;
+	u16 msg_num, hash_key_index;
+	u8 index;
+	int ret;
+
+	hclgevf_build_send_msg(&send_msg, HCLGE_MBX_GET_RSS_KEY, 0);
+	msg_num = (HCLGE_COMM_RSS_KEY_SIZE + HCLGEVF_RSS_MBX_RESP_LEN - 1) /
+			HCLGEVF_RSS_MBX_RESP_LEN;
+	for (index = 0; index < msg_num; index++) {
+		send_msg.data[0] = index;
+		ret = hclgevf_send_mbx_msg(hdev, &send_msg, true, resp_msg,
+					   HCLGEVF_RSS_MBX_RESP_LEN);
+		if (ret) {
+			dev_err(&hdev->pdev->dev,
+				"VF get rss hash key from PF failed, ret=%d",
+				ret);
+			return ret;
+		}
+
+		hash_key_index = HCLGEVF_RSS_MBX_RESP_LEN * index;
+		if (index == msg_num - 1)
+			memcpy(&rss_cfg->rss_hash_key[hash_key_index],
+			       &resp_msg[0],
+			       HCLGE_COMM_RSS_KEY_SIZE - hash_key_index);
+		else
+			memcpy(&rss_cfg->rss_hash_key[hash_key_index],
+			       &resp_msg[0], HCLGEVF_RSS_MBX_RESP_LEN);
+	}
+
+	return 0;
+}
+
+static int hclgevf_get_rss(struct hnae3_handle *handle, u32 *indir, u8 *key,
+			   u8 *hfunc)
+{
+	struct hclgevf_dev *hdev = hclgevf_ae_get_hdev(handle);
+	struct hclge_comm_rss_cfg *rss_cfg = &hdev->rss_cfg;
+	int ret;
+
+	if (hdev->ae_dev->dev_version >= HNAE3_DEVICE_VERSION_V2) {
+		hclge_comm_get_rss_hash_info(rss_cfg, key, hfunc);
+	} else {
+		if (hfunc)
+			*hfunc = ETH_RSS_HASH_TOP;
+		if (key) {
+			ret = hclgevf_get_rss_hash_key(hdev);
+			if (ret)
+				return ret;
+			memcpy(key, rss_cfg->rss_hash_key,
+			       HCLGE_COMM_RSS_KEY_SIZE);
+		}
+	}
+
+	hclge_comm_get_rss_indir_tbl(rss_cfg, indir,
+				     hdev->ae_dev->dev_specs.rss_ind_tbl_size);
+
+	return 0;
+}
+
+static int hclgevf_set_rss(struct hnae3_handle *handle, const u32 *indir,
+			   const u8 *key, const u8 hfunc)
+{
+	struct hclgevf_dev *hdev = hclgevf_ae_get_hdev(handle);
+	struct hclge_comm_rss_cfg *rss_cfg = &hdev->rss_cfg;
+	int ret, i;
+
+	if (hdev->ae_dev->dev_version >= HNAE3_DEVICE_VERSION_V2) {
+		ret = hclge_comm_set_rss_hash_key(rss_cfg, &hdev->hw.hw, key,
+						  hfunc);
+		if (ret)
+			return ret;
+	}
+
+	/* update the shadow RSS table with user specified qids */
+	for (i = 0; i < hdev->ae_dev->dev_specs.rss_ind_tbl_size; i++)
+		rss_cfg->rss_indirection_tbl[i] = indir[i];
+
+	/* update the hardware */
+	return hclge_comm_set_rss_indir_table(hdev->ae_dev, &hdev->hw.hw,
+					      rss_cfg->rss_indirection_tbl);
+}
+
+static int hclgevf_set_rss_tuple(struct hnae3_handle *handle,
+				 const struct ethtool_rxfh_fields *nfc)
+{
+	struct hclgevf_dev *hdev = hclgevf_ae_get_hdev(handle);
+	int ret;
+
+	if (hdev->ae_dev->dev_version < HNAE3_DEVICE_VERSION_V2)
+		return -EOPNOTSUPP;
+
+	ret = hclge_comm_set_rss_tuple(hdev->ae_dev, &hdev->hw.hw,
+				       &hdev->rss_cfg, nfc);
+	if (ret)
+		dev_err(&hdev->pdev->dev,
+		"failed to set rss tuple, ret = %d.\n", ret);
+
+	return ret;
+}
+
+static int hclgevf_get_rss_tuple(struct hnae3_handle *handle,
+				 struct ethtool_rxfh_fields *nfc)
+{
+	struct hclgevf_dev *hdev = hclgevf_ae_get_hdev(handle);
+	u8 tuple_sets;
+	int ret;
+
+	if (hdev->ae_dev->dev_version < HNAE3_DEVICE_VERSION_V2)
+		return -EOPNOTSUPP;
+
+	nfc->data = 0;
+
+	ret = hclge_comm_get_rss_tuple(&hdev->rss_cfg, nfc->flow_type,
+				       &tuple_sets);
+	if (ret || !tuple_sets)
+		return ret;
+
+	nfc->data = hclge_comm_convert_rss_tuple(tuple_sets);
+
+	return 0;
+}
+
+static int hclgevf_get_tc_size(struct hnae3_handle *handle)
+{
+	struct hclgevf_dev *hdev = hclgevf_ae_get_hdev(handle);
+	struct hclge_comm_rss_cfg *rss_cfg = &hdev->rss_cfg;
+
+	return rss_cfg->rss_size;
+}
+
+static int hclgevf_bind_ring_to_vector(struct hnae3_handle *handle, bool en,
+				       int vector_id,
+				       struct hnae3_ring_chain_node *ring_chain)
+{
+	struct hclgevf_dev *hdev = hclgevf_ae_get_hdev(handle);
+	struct hclge_vf_to_pf_msg send_msg;
+	struct hnae3_ring_chain_node *node;
+	int status;
+	int i = 0;
+
+	memset(&send_msg, 0, sizeof(send_msg));
+	send_msg.code = en ? HCLGE_MBX_MAP_RING_TO_VECTOR :
+		HCLGE_MBX_UNMAP_RING_TO_VECTOR;
+	send_msg.vector_id = vector_id;
+
+	for (node = ring_chain; node; node = node->next) {
+		send_msg.param[i].ring_type =
+				hnae3_get_bit(node->flag, HNAE3_RING_TYPE_B);
+
+		send_msg.param[i].tqp_index = node->tqp_index;
+		send_msg.param[i].int_gl_index =
+					hnae3_get_field(node->int_gl_idx,
+							HNAE3_RING_GL_IDX_M,
+							HNAE3_RING_GL_IDX_S);
+
+		i++;
+		if (i == HCLGE_MBX_MAX_RING_CHAIN_PARAM_NUM || !node->next) {
+			send_msg.ring_num = i;
+
+			status = hclgevf_send_mbx_msg(hdev, &send_msg, false,
+						      NULL, 0);
+			if (status) {
+				dev_err(&hdev->pdev->dev,
+					"Map TQP fail, status is %d.\n",
+					status);
+				return status;
+			}
+			i = 0;
+		}
+	}
+
+	return 0;
+}
+
+static int hclgevf_map_ring_to_vector(struct hnae3_handle *handle, int vector,
+				      struct hnae3_ring_chain_node *ring_chain)
+{
+	struct hclgevf_dev *hdev = hclgevf_ae_get_hdev(handle);
+	int vector_id;
+
+	vector_id = hclgevf_get_vector_index(hdev, vector);
+	if (vector_id < 0) {
+		dev_err(&handle->pdev->dev,
+			"Get vector index fail. ret =%d\n", vector_id);
+		return vector_id;
+	}
+
+	return hclgevf_bind_ring_to_vector(handle, true, vector_id, ring_chain);
+}
+
+static int hclgevf_unmap_ring_from_vector(
+				struct hnae3_handle *handle,
+				int vector,
+				struct hnae3_ring_chain_node *ring_chain)
+{
+	struct hclgevf_dev *hdev = hclgevf_ae_get_hdev(handle);
+	int ret, vector_id;
+
+	if (test_bit(HCLGEVF_STATE_RST_HANDLING, &hdev->state))
+		return 0;
+
+	vector_id = hclgevf_get_vector_index(hdev, vector);
+	if (vector_id < 0) {
+		dev_err(&handle->pdev->dev,
+			"Get vector index fail. ret =%d\n", vector_id);
+		return vector_id;
+	}
+
+	ret = hclgevf_bind_ring_to_vector(handle, false, vector_id, ring_chain);
+	if (ret)
+		dev_err(&handle->pdev->dev,
+			"Unmap ring from vector fail. vector=%d, ret =%d\n",
+			vector_id,
+			ret);
+
+	return ret;
+}
+
+static int hclgevf_put_vector(struct hnae3_handle *handle, int vector)
+{
+	struct hclgevf_dev *hdev = hclgevf_ae_get_hdev(handle);
+	int vector_id;
+
+	vector_id = hclgevf_get_vector_index(hdev, vector);
+	if (vector_id < 0) {
+		dev_err(&handle->pdev->dev,
+			"hclgevf_put_vector get vector index fail. ret =%d\n",
+			vector_id);
+		return vector_id;
+	}
+
+	hclgevf_free_vector(hdev, vector_id);
+
+	return 0;
+}
+
+static int hclgevf_cmd_set_promisc_mode(struct hclgevf_dev *hdev,
+					bool en_uc_pmc, bool en_mc_pmc,
+					bool en_bc_pmc)
+{
+	struct hnae3_handle *handle = &hdev->nic;
+	struct hclge_vf_to_pf_msg send_msg;
+	int ret;
+
+	memset(&send_msg, 0, sizeof(send_msg));
+	send_msg.code = HCLGE_MBX_SET_PROMISC_MODE;
+	send_msg.en_bc = en_bc_pmc ? 1 : 0;
+	send_msg.en_uc = en_uc_pmc ? 1 : 0;
+	send_msg.en_mc = en_mc_pmc ? 1 : 0;
+	send_msg.en_limit_promisc = test_bit(HNAE3_PFLAG_LIMIT_PROMISC,
+					     &handle->priv_flags) ? 1 : 0;
+
+	ret = hclgevf_send_mbx_msg(hdev, &send_msg, false, NULL, 0);
+	if (ret)
+		dev_err(&hdev->pdev->dev,
+			"Set promisc mode fail, status is %d.\n", ret);
+
+	return ret;
+}
+
+static int hclgevf_set_promisc_mode(struct hnae3_handle *handle, bool en_uc_pmc,
+				    bool en_mc_pmc)
+{
+	struct hclgevf_dev *hdev = hclgevf_ae_get_hdev(handle);
+	bool en_bc_pmc;
+
+	en_bc_pmc = hdev->ae_dev->dev_version >= HNAE3_DEVICE_VERSION_V2;
+
+	return hclgevf_cmd_set_promisc_mode(hdev, en_uc_pmc, en_mc_pmc,
+					    en_bc_pmc);
+}
+
+static void hclgevf_request_update_promisc_mode(struct hnae3_handle *handle)
+{
+	struct hclgevf_dev *hdev = hclgevf_ae_get_hdev(handle);
+
+	set_bit(HCLGEVF_STATE_PROMISC_CHANGED, &hdev->state);
+	hclgevf_task_schedule(hdev, 0);
+}
+
+static void hclgevf_sync_promisc_mode(struct hclgevf_dev *hdev)
+{
+	struct hnae3_handle *handle = &hdev->nic;
+	bool en_uc_pmc = handle->netdev_flags & HNAE3_UPE;
+	bool en_mc_pmc = handle->netdev_flags & HNAE3_MPE;
+	int ret;
+
+	if (test_bit(HCLGEVF_STATE_PROMISC_CHANGED, &hdev->state)) {
+		ret = hclgevf_set_promisc_mode(handle, en_uc_pmc, en_mc_pmc);
+		if (!ret)
+			clear_bit(HCLGEVF_STATE_PROMISC_CHANGED, &hdev->state);
+	}
+}
+
+static int hclgevf_tqp_enable_cmd_send(struct hclgevf_dev *hdev, u16 tqp_id,
+				       u16 stream_id, bool enable)
+{
+	struct hclgevf_cfg_com_tqp_queue_cmd *req;
+	struct hclge_desc desc;
+
+	req = (struct hclgevf_cfg_com_tqp_queue_cmd *)desc.data;
+
+	hclgevf_cmd_setup_basic_desc(&desc, HCLGE_OPC_CFG_COM_TQP_QUEUE, false);
+	req->tqp_id = cpu_to_le16(tqp_id & HCLGEVF_RING_ID_MASK);
+	req->stream_id = cpu_to_le16(stream_id);
+	if (enable)
+		req->enable |= 1U << HCLGEVF_TQP_ENABLE_B;
+
+	return hclgevf_cmd_send(&hdev->hw, &desc, 1);
+}
+
+static int hclgevf_tqp_enable(struct hnae3_handle *handle, bool enable)
+{
+	struct hclgevf_dev *hdev = hclgevf_ae_get_hdev(handle);
+	int ret;
+	u16 i;
+
+	for (i = 0; i < handle->kinfo.num_tqps; i++) {
+		ret = hclgevf_tqp_enable_cmd_send(hdev, i, 0, enable);
+		if (ret)
+			return ret;
+	}
+
+	return 0;
+}
+
+static int hclgevf_get_host_mac_addr(struct hclgevf_dev *hdev, u8 *p)
+{
+	struct hclge_vf_to_pf_msg send_msg;
+	u8 host_mac[ETH_ALEN];
+	int status;
+
+	hclgevf_build_send_msg(&send_msg, HCLGE_MBX_GET_MAC_ADDR, 0);
+	status = hclgevf_send_mbx_msg(hdev, &send_msg, true, host_mac,
+				      ETH_ALEN);
+	if (status) {
+		dev_err(&hdev->pdev->dev,
+			"fail to get VF MAC from host %d", status);
+		return status;
+	}
+
+	ether_addr_copy(p, host_mac);
+
+	return 0;
+}
+
+static void hclgevf_get_mac_addr(struct hnae3_handle *handle, u8 *p)
+{
+	struct hclgevf_dev *hdev = hclgevf_ae_get_hdev(handle);
+	u8 host_mac_addr[ETH_ALEN];
+
+	if (hclgevf_get_host_mac_addr(hdev, host_mac_addr))
+		return;
+
+	hdev->has_pf_mac = !is_zero_ether_addr(host_mac_addr);
+	if (hdev->has_pf_mac)
+		ether_addr_copy(p, host_mac_addr);
+	else
+		ether_addr_copy(p, hdev->hw.mac.mac_addr);
+}
+
+static int hclgevf_set_mac_addr(struct hnae3_handle *handle, const void *p,
+				bool is_first)
+{
+	struct hclgevf_dev *hdev = hclgevf_ae_get_hdev(handle);
+	u8 *old_mac_addr = (u8 *)hdev->hw.mac.mac_addr;
+	struct hclge_vf_to_pf_msg send_msg;
+	u8 *new_mac_addr = (u8 *)p;
+	int status;
+
+	hclgevf_build_send_msg(&send_msg, HCLGE_MBX_SET_UNICAST, 0);
+	send_msg.subcode = HCLGE_MBX_MAC_VLAN_UC_MODIFY;
+	ether_addr_copy(send_msg.data, new_mac_addr);
+	if (is_first && !hdev->has_pf_mac)
+		eth_zero_addr(&send_msg.data[ETH_ALEN]);
+	else
+		ether_addr_copy(&send_msg.data[ETH_ALEN], old_mac_addr);
+	status = hclgevf_send_mbx_msg(hdev, &send_msg, true, NULL, 0);
+	if (!status)
+		ether_addr_copy(hdev->hw.mac.mac_addr, new_mac_addr);
+
+	return status;
+}
+
+static struct hclgevf_mac_addr_node *
+hclgevf_find_mac_node(struct list_head *list, const u8 *mac_addr)
+{
+	struct hclgevf_mac_addr_node *mac_node, *tmp;
+
+	list_for_each_entry_safe(mac_node, tmp, list, node)
+		if (ether_addr_equal(mac_addr, mac_node->mac_addr))
+			return mac_node;
+
+	return NULL;
+}
+
+static void hclgevf_update_mac_node(struct hclgevf_mac_addr_node *mac_node,
+				    enum HCLGEVF_MAC_NODE_STATE state)
+{
+	switch (state) {
+	/* from set_rx_mode or tmp_add_list */
+	case HCLGEVF_MAC_TO_ADD:
+		if (mac_node->state == HCLGEVF_MAC_TO_DEL)
+			mac_node->state = HCLGEVF_MAC_ACTIVE;
+		break;
+	/* only from set_rx_mode */
+	case HCLGEVF_MAC_TO_DEL:
+		if (mac_node->state == HCLGEVF_MAC_TO_ADD) {
+			list_del(&mac_node->node);
+			kfree(mac_node);
+		} else {
+			mac_node->state = HCLGEVF_MAC_TO_DEL;
+		}
+		break;
+	/* only from tmp_add_list, the mac_node->state won't be
+	 * HCLGEVF_MAC_ACTIVE
+	 */
+	case HCLGEVF_MAC_ACTIVE:
+		if (mac_node->state == HCLGEVF_MAC_TO_ADD)
+			mac_node->state = HCLGEVF_MAC_ACTIVE;
+		break;
+	}
+}
+
+static int hclgevf_update_mac_list(struct hnae3_handle *handle,
+				   enum HCLGEVF_MAC_NODE_STATE state,
+				   enum HCLGEVF_MAC_ADDR_TYPE mac_type,
+				   const unsigned char *addr)
+{
+	struct hclgevf_dev *hdev = hclgevf_ae_get_hdev(handle);
+	struct hclgevf_mac_addr_node *mac_node;
+	struct list_head *list;
+
+	list = (mac_type == HCLGEVF_MAC_ADDR_UC) ?
+	       &hdev->mac_table.uc_mac_list : &hdev->mac_table.mc_mac_list;
+
+	spin_lock_bh(&hdev->mac_table.mac_list_lock);
+
+	/* if the mac addr is already in the mac list, no need to add a new
+	 * one into it, just check the mac addr state, convert it to a new
+	 * state, or just remove it, or do nothing.
+	 */
+	mac_node = hclgevf_find_mac_node(list, addr);
+	if (mac_node) {
+		hclgevf_update_mac_node(mac_node, state);
+		spin_unlock_bh(&hdev->mac_table.mac_list_lock);
+		return 0;
+	}
+	/* if this address is never added, unnecessary to delete */
+	if (state == HCLGEVF_MAC_TO_DEL) {
+		spin_unlock_bh(&hdev->mac_table.mac_list_lock);
+		return -ENOENT;
+	}
+
+	mac_node = kzalloc(sizeof(*mac_node), GFP_ATOMIC);
+	if (!mac_node) {
+		spin_unlock_bh(&hdev->mac_table.mac_list_lock);
+		return -ENOMEM;
+	}
+
+	mac_node->state = state;
+	ether_addr_copy(mac_node->mac_addr, addr);
+	list_add_tail(&mac_node->node, list);
+
+	spin_unlock_bh(&hdev->mac_table.mac_list_lock);
+	return 0;
+}
+
+static int hclgevf_add_uc_addr(struct hnae3_handle *handle,
+			       const unsigned char *addr)
+{
+	return hclgevf_update_mac_list(handle, HCLGEVF_MAC_TO_ADD,
+				       HCLGEVF_MAC_ADDR_UC, addr);
+}
+
+static int hclgevf_rm_uc_addr(struct hnae3_handle *handle,
+			      const unsigned char *addr)
+{
+	return hclgevf_update_mac_list(handle, HCLGEVF_MAC_TO_DEL,
+				       HCLGEVF_MAC_ADDR_UC, addr);
+}
+
+static int hclgevf_add_mc_addr(struct hnae3_handle *handle,
+			       const unsigned char *addr)
+{
+	return hclgevf_update_mac_list(handle, HCLGEVF_MAC_TO_ADD,
+				       HCLGEVF_MAC_ADDR_MC, addr);
+}
+
+static int hclgevf_rm_mc_addr(struct hnae3_handle *handle,
+			      const unsigned char *addr)
+{
+	return hclgevf_update_mac_list(handle, HCLGEVF_MAC_TO_DEL,
+				       HCLGEVF_MAC_ADDR_MC, addr);
+}
+
+static int hclgevf_add_del_mac_addr(struct hclgevf_dev *hdev,
+				    struct hclgevf_mac_addr_node *mac_node,
+				    enum HCLGEVF_MAC_ADDR_TYPE mac_type)
+{
+	struct hclge_vf_to_pf_msg send_msg;
+	u8 code, subcode;
+
+	if (mac_type == HCLGEVF_MAC_ADDR_UC) {
+		code = HCLGE_MBX_SET_UNICAST;
+		if (mac_node->state == HCLGEVF_MAC_TO_ADD)
+			subcode = HCLGE_MBX_MAC_VLAN_UC_ADD;
+		else
+			subcode = HCLGE_MBX_MAC_VLAN_UC_REMOVE;
+	} else {
+		code = HCLGE_MBX_SET_MULTICAST;
+		if (mac_node->state == HCLGEVF_MAC_TO_ADD)
+			subcode = HCLGE_MBX_MAC_VLAN_MC_ADD;
+		else
+			subcode = HCLGE_MBX_MAC_VLAN_MC_REMOVE;
+	}
+
+	hclgevf_build_send_msg(&send_msg, code, subcode);
+	ether_addr_copy(send_msg.data, mac_node->mac_addr);
+	return hclgevf_send_mbx_msg(hdev, &send_msg, false, NULL, 0);
+}
+
+static void hclgevf_config_mac_list(struct hclgevf_dev *hdev,
+				    struct list_head *list,
+				    enum HCLGEVF_MAC_ADDR_TYPE mac_type)
+{
+	char format_mac_addr[HNAE3_FORMAT_MAC_ADDR_LEN];
+	struct hclgevf_mac_addr_node *mac_node, *tmp;
+	int ret;
+
+	list_for_each_entry_safe(mac_node, tmp, list, node) {
+		ret = hclgevf_add_del_mac_addr(hdev, mac_node, mac_type);
+		if  (ret) {
+			hnae3_format_mac_addr(format_mac_addr,
+					      mac_node->mac_addr);
+			dev_err(&hdev->pdev->dev,
+				"failed to configure mac %s, state = %d, ret = %d\n",
+				format_mac_addr, mac_node->state, ret);
+			return;
+		}
+		if (mac_node->state == HCLGEVF_MAC_TO_ADD) {
+			mac_node->state = HCLGEVF_MAC_ACTIVE;
+		} else {
+			list_del(&mac_node->node);
+			kfree(mac_node);
+		}
+	}
+}
+
+static void hclgevf_sync_from_add_list(struct list_head *add_list,
+				       struct list_head *mac_list)
+{
+	struct hclgevf_mac_addr_node *mac_node, *tmp, *new_node;
+
+	list_for_each_entry_safe(mac_node, tmp, add_list, node) {
+		/* if the mac address from tmp_add_list is not in the
+		 * uc/mc_mac_list, it means have received a TO_DEL request
+		 * during the time window of sending mac config request to PF
+		 * If mac_node state is ACTIVE, then change its state to TO_DEL,
+		 * then it will be removed at next time. If is TO_ADD, it means
+		 * send TO_ADD request failed, so just remove the mac node.
+		 */
+		new_node = hclgevf_find_mac_node(mac_list, mac_node->mac_addr);
+		if (new_node) {
+			hclgevf_update_mac_node(new_node, mac_node->state);
+			list_del(&mac_node->node);
+			kfree(mac_node);
+		} else if (mac_node->state == HCLGEVF_MAC_ACTIVE) {
+			mac_node->state = HCLGEVF_MAC_TO_DEL;
+			list_move_tail(&mac_node->node, mac_list);
+		} else {
+			list_del(&mac_node->node);
+			kfree(mac_node);
+		}
+	}
+}
+
+static void hclgevf_sync_from_del_list(struct list_head *del_list,
+				       struct list_head *mac_list)
+{
+	struct hclgevf_mac_addr_node *mac_node, *tmp, *new_node;
+
+	list_for_each_entry_safe(mac_node, tmp, del_list, node) {
+		new_node = hclgevf_find_mac_node(mac_list, mac_node->mac_addr);
+		if (new_node) {
+			/* If the mac addr is exist in the mac list, it means
+			 * received a new request TO_ADD during the time window
+			 * of sending mac addr configurrequest to PF, so just
+			 * change the mac state to ACTIVE.
+			 */
+			new_node->state = HCLGEVF_MAC_ACTIVE;
+			list_del(&mac_node->node);
+			kfree(mac_node);
+		} else {
+			list_move_tail(&mac_node->node, mac_list);
+		}
+	}
+}
+
+static void hclgevf_clear_list(struct list_head *list)
+{
+	struct hclgevf_mac_addr_node *mac_node, *tmp;
+
+	list_for_each_entry_safe(mac_node, tmp, list, node) {
+		list_del(&mac_node->node);
+		kfree(mac_node);
+	}
+}
+
+static void hclgevf_sync_mac_list(struct hclgevf_dev *hdev,
+				  enum HCLGEVF_MAC_ADDR_TYPE mac_type)
+{
+	struct hclgevf_mac_addr_node *mac_node, *tmp, *new_node;
+	struct list_head tmp_add_list, tmp_del_list;
+	struct list_head *list;
+
+	INIT_LIST_HEAD(&tmp_add_list);
+	INIT_LIST_HEAD(&tmp_del_list);
+
+	/* move the mac addr to the tmp_add_list and tmp_del_list, then
+	 * we can add/delete these mac addr outside the spin lock
+	 */
+	list = (mac_type == HCLGEVF_MAC_ADDR_UC) ?
+		&hdev->mac_table.uc_mac_list : &hdev->mac_table.mc_mac_list;
+
+	spin_lock_bh(&hdev->mac_table.mac_list_lock);
+
+	list_for_each_entry_safe(mac_node, tmp, list, node) {
+		switch (mac_node->state) {
+		case HCLGEVF_MAC_TO_DEL:
+			list_move_tail(&mac_node->node, &tmp_del_list);
+			break;
+		case HCLGEVF_MAC_TO_ADD:
+			new_node = kzalloc(sizeof(*new_node), GFP_ATOMIC);
+			if (!new_node)
+				goto stop_traverse;
+
+			ether_addr_copy(new_node->mac_addr, mac_node->mac_addr);
+			new_node->state = mac_node->state;
+			list_add_tail(&new_node->node, &tmp_add_list);
+			break;
+		default:
+			break;
+		}
+	}
+
+stop_traverse:
+	spin_unlock_bh(&hdev->mac_table.mac_list_lock);
+
+	/* delete first, in order to get max mac table space for adding */
+	hclgevf_config_mac_list(hdev, &tmp_del_list, mac_type);
+	hclgevf_config_mac_list(hdev, &tmp_add_list, mac_type);
+
+	/* if some mac addresses were added/deleted fail, move back to the
+	 * mac_list, and retry at next time.
+	 */
+	spin_lock_bh(&hdev->mac_table.mac_list_lock);
+
+	hclgevf_sync_from_del_list(&tmp_del_list, list);
+	hclgevf_sync_from_add_list(&tmp_add_list, list);
+
+	spin_unlock_bh(&hdev->mac_table.mac_list_lock);
+}
+
+static void hclgevf_sync_mac_table(struct hclgevf_dev *hdev)
+{
+	hclgevf_sync_mac_list(hdev, HCLGEVF_MAC_ADDR_UC);
+	hclgevf_sync_mac_list(hdev, HCLGEVF_MAC_ADDR_MC);
+}
+
+static void hclgevf_uninit_mac_list(struct hclgevf_dev *hdev)
+{
+	spin_lock_bh(&hdev->mac_table.mac_list_lock);
+
+	hclgevf_clear_list(&hdev->mac_table.uc_mac_list);
+	hclgevf_clear_list(&hdev->mac_table.mc_mac_list);
+
+	spin_unlock_bh(&hdev->mac_table.mac_list_lock);
+}
+
+static int hclgevf_enable_vlan_filter(struct hnae3_handle *handle, bool enable)
+{
+	struct hclgevf_dev *hdev = hclgevf_ae_get_hdev(handle);
+	struct hnae3_ae_dev *ae_dev = hdev->ae_dev;
+	struct hclge_vf_to_pf_msg send_msg;
+
+	if (!test_bit(HNAE3_DEV_SUPPORT_VLAN_FLTR_MDF_B, ae_dev->caps))
+		return -EOPNOTSUPP;
+
+	hclgevf_build_send_msg(&send_msg, HCLGE_MBX_SET_VLAN,
+			       HCLGE_MBX_ENABLE_VLAN_FILTER);
+	send_msg.data[0] = enable ? 1 : 0;
+
+	return hclgevf_send_mbx_msg(hdev, &send_msg, true, NULL, 0);
+}
+
+static int hclgevf_set_vlan_filter(struct hnae3_handle *handle,
+				   __be16 proto, u16 vlan_id,
+				   bool is_kill)
+{
+	struct hclgevf_dev *hdev = hclgevf_ae_get_hdev(handle);
+	struct hclge_mbx_vlan_filter *vlan_filter;
+	struct hclge_vf_to_pf_msg send_msg;
+	int ret;
+
+	if (vlan_id > HCLGEVF_MAX_VLAN_ID)
+		return -EINVAL;
+
+	if (proto != htons(ETH_P_8021Q))
+		return -EPROTONOSUPPORT;
+
+	/* When device is resetting or reset failed, firmware is unable to
+	 * handle mailbox. Just record the vlan id, and remove it after
+	 * reset finished.
+	 */
+	if ((test_bit(HCLGEVF_STATE_RST_HANDLING, &hdev->state) ||
+	     test_bit(HCLGEVF_STATE_RST_FAIL, &hdev->state)) && is_kill) {
+		set_bit(vlan_id, hdev->vlan_del_fail_bmap);
+		return -EBUSY;
+	} else if (!is_kill && test_bit(vlan_id, hdev->vlan_del_fail_bmap)) {
+		clear_bit(vlan_id, hdev->vlan_del_fail_bmap);
+	}
+
+	hclgevf_build_send_msg(&send_msg, HCLGE_MBX_SET_VLAN,
+			       HCLGE_MBX_VLAN_FILTER);
+	vlan_filter = (struct hclge_mbx_vlan_filter *)send_msg.data;
+	vlan_filter->is_kill = is_kill;
+	vlan_filter->vlan_id = cpu_to_le16(vlan_id);
+	vlan_filter->proto = cpu_to_le16(be16_to_cpu(proto));
+
+	/* when remove hw vlan filter failed, record the vlan id,
+	 * and try to remove it from hw later, to be consistence
+	 * with stack.
+	 */
+	ret = hclgevf_send_mbx_msg(hdev, &send_msg, true, NULL, 0);
+	if (is_kill && ret)
+		set_bit(vlan_id, hdev->vlan_del_fail_bmap);
+
+	return ret;
+}
+
+static void hclgevf_sync_vlan_filter(struct hclgevf_dev *hdev)
+{
+#define HCLGEVF_MAX_SYNC_COUNT	60
+	struct hnae3_handle *handle = &hdev->nic;
+	int ret, sync_cnt = 0;
+	u16 vlan_id;
+
+	if (bitmap_empty(hdev->vlan_del_fail_bmap, VLAN_N_VID))
+		return;
+
+	rtnl_lock();
+	vlan_id = find_first_bit(hdev->vlan_del_fail_bmap, VLAN_N_VID);
+	while (vlan_id != VLAN_N_VID) {
+		ret = hclgevf_set_vlan_filter(handle, htons(ETH_P_8021Q),
+					      vlan_id, true);
+		if (ret)
+			break;
+
+		clear_bit(vlan_id, hdev->vlan_del_fail_bmap);
+		sync_cnt++;
+		if (sync_cnt >= HCLGEVF_MAX_SYNC_COUNT)
+			break;
+
+		vlan_id = find_first_bit(hdev->vlan_del_fail_bmap, VLAN_N_VID);
+	}
+	rtnl_unlock();
+}
+
+static int hclgevf_en_hw_strip_rxvtag_cmd(struct hclgevf_dev *hdev, bool enable)
+{
+	struct hclge_vf_to_pf_msg send_msg;
+
+	hclgevf_build_send_msg(&send_msg, HCLGE_MBX_SET_VLAN,
+			       HCLGE_MBX_VLAN_RX_OFF_CFG);
+	send_msg.data[0] = enable ? 1 : 0;
+	return hclgevf_send_mbx_msg(hdev, &send_msg, false, NULL, 0);
+}
+
+static int hclgevf_en_hw_strip_rxvtag(struct hnae3_handle *handle, bool enable)
+{
+	struct hclgevf_dev *hdev = hclgevf_ae_get_hdev(handle);
+	int ret;
+
+	ret = hclgevf_en_hw_strip_rxvtag_cmd(hdev, enable);
+	if (ret)
+		return ret;
+
+	hdev->rxvtag_strip_en = enable;
+	return 0;
+}
+
+static int hclgevf_reset_tqp(struct hnae3_handle *handle)
+{
+#define HCLGEVF_RESET_ALL_QUEUE_DONE	1U
+	struct hclgevf_dev *hdev = hclgevf_ae_get_hdev(handle);
+	struct hclge_vf_to_pf_msg send_msg;
+	u8 return_status = 0;
+	int ret;
+	u16 i;
+
+	/* disable vf queue before send queue reset msg to PF */
+	ret = hclgevf_tqp_enable(handle, false);
+	if (ret) {
+		dev_err(&hdev->pdev->dev, "failed to disable tqp, ret = %d\n",
+			ret);
+		return ret;
+	}
+
+	hclgevf_build_send_msg(&send_msg, HCLGE_MBX_QUEUE_RESET, 0);
+
+	ret = hclgevf_send_mbx_msg(hdev, &send_msg, true, &return_status,
+				   sizeof(return_status));
+	if (ret || return_status == HCLGEVF_RESET_ALL_QUEUE_DONE)
+		return ret;
+
+	for (i = 1; i < handle->kinfo.num_tqps; i++) {
+		hclgevf_build_send_msg(&send_msg, HCLGE_MBX_QUEUE_RESET, 0);
+		*(__le16 *)send_msg.data = cpu_to_le16(i);
+		ret = hclgevf_send_mbx_msg(hdev, &send_msg, true, NULL, 0);
+		if (ret)
+			return ret;
+	}
+
+	return 0;
+}
+
+static int hclgevf_set_mtu(struct hnae3_handle *handle, int new_mtu)
+{
+	struct hclgevf_dev *hdev = hclgevf_ae_get_hdev(handle);
+	struct hclge_mbx_mtu_info *mtu_info;
+	struct hclge_vf_to_pf_msg send_msg;
+
+	hclgevf_build_send_msg(&send_msg, HCLGE_MBX_SET_MTU, 0);
+	mtu_info = (struct hclge_mbx_mtu_info *)send_msg.data;
+	mtu_info->mtu = cpu_to_le32(new_mtu);
+
+	return hclgevf_send_mbx_msg(hdev, &send_msg, true, NULL, 0);
+}
+
+static int hclgevf_notify_client(struct hclgevf_dev *hdev,
+				 enum hnae3_reset_notify_type type)
+{
+	struct hnae3_client *client = hdev->nic_client;
+	struct hnae3_handle *handle = &hdev->nic;
+	int ret;
+
+	if (!test_bit(HCLGEVF_STATE_NIC_REGISTERED, &hdev->state) ||
+	    !client)
+		return 0;
+
+	if (!client->ops->reset_notify)
+		return -EOPNOTSUPP;
+
+	ret = client->ops->reset_notify(handle, type);
+	if (ret)
+		dev_err(&hdev->pdev->dev, "notify nic client failed %d(%d)\n",
+			type, ret);
+
+	return ret;
+}
+
+static int hclgevf_notify_roce_client(struct hclgevf_dev *hdev,
+				      enum hnae3_reset_notify_type type)
+{
+	struct hnae3_client *client = hdev->roce_client;
+	struct hnae3_handle *handle = &hdev->roce;
+	int ret;
+
+	if (!test_bit(HCLGEVF_STATE_ROCE_REGISTERED, &hdev->state) || !client)
+		return 0;
+
+	if (!client->ops->reset_notify)
+		return -EOPNOTSUPP;
+
+	ret = client->ops->reset_notify(handle, type);
+	if (ret)
+		dev_err(&hdev->pdev->dev, "notify roce client failed %d(%d)",
+			type, ret);
+	return ret;
+}
+
+static void hclgevf_set_reset_pending(struct hclgevf_dev *hdev,
+				      enum hnae3_reset_type reset_type)
+{
+	/* When an incorrect reset type is executed, the get_reset_level
+	 * function generates the HNAE3_NONE_RESET flag. As a result, this
+	 * type do not need to pending.
+	 */
+	if (reset_type != HNAE3_NONE_RESET)
+		set_bit(reset_type, &hdev->reset_pending);
+}
+
+static int hclgevf_reset_wait(struct hclgevf_dev *hdev)
+{
+#define HCLGEVF_RESET_WAIT_US	20000
+#define HCLGEVF_RESET_WAIT_CNT	2000
+#define HCLGEVF_RESET_WAIT_TIMEOUT_US	\
+	(HCLGEVF_RESET_WAIT_US * HCLGEVF_RESET_WAIT_CNT)
+
+	u32 val;
+	int ret;
+
+	if (hdev->reset_type == HNAE3_VF_RESET)
+		ret = readl_poll_timeout(hdev->hw.hw.io_base +
+					 HCLGEVF_VF_RST_ING, val,
+					 !(val & HCLGEVF_VF_RST_ING_BIT),
+					 HCLGEVF_RESET_WAIT_US,
+					 HCLGEVF_RESET_WAIT_TIMEOUT_US);
+	else
+		ret = readl_poll_timeout(hdev->hw.hw.io_base +
+					 HCLGEVF_RST_ING, val,
+					 !(val & HCLGEVF_RST_ING_BITS),
+					 HCLGEVF_RESET_WAIT_US,
+					 HCLGEVF_RESET_WAIT_TIMEOUT_US);
+
+	/* hardware completion status should be available by this time */
+	if (ret) {
+		dev_err(&hdev->pdev->dev,
+			"couldn't get reset done status from h/w, timeout!\n");
+		return ret;
+	}
+
+	/* we will wait a bit more to let reset of the stack to complete. This
+	 * might happen in case reset assertion was made by PF. Yes, this also
+	 * means we might end up waiting bit more even for VF reset.
+	 */
+	if (hdev->reset_type == HNAE3_VF_FULL_RESET)
+		msleep(5000);
+	else
+		msleep(500);
+
+	return 0;
+}
+
+static void hclgevf_reset_handshake(struct hclgevf_dev *hdev, bool enable)
+{
+	u32 reg_val;
+
+	reg_val = hclgevf_read_dev(&hdev->hw, HCLGE_COMM_NIC_CSQ_DEPTH_REG);
+	if (enable)
+		reg_val |= HCLGEVF_NIC_SW_RST_RDY;
+	else
+		reg_val &= ~HCLGEVF_NIC_SW_RST_RDY;
+
+	hclgevf_write_dev(&hdev->hw, HCLGE_COMM_NIC_CSQ_DEPTH_REG,
+			  reg_val);
+}
+
+static int hclgevf_reset_stack(struct hclgevf_dev *hdev)
+{
+	int ret;
+
+	/* uninitialize the nic client */
+	ret = hclgevf_notify_client(hdev, HNAE3_UNINIT_CLIENT);
+	if (ret)
+		return ret;
+
+	/* re-initialize the hclge device */
+	ret = hclgevf_reset_hdev(hdev);
+	if (ret) {
+		dev_err(&hdev->pdev->dev,
+			"hclge device re-init failed, VF is disabled!\n");
+		return ret;
+	}
+
+	/* bring up the nic client again */
+	ret = hclgevf_notify_client(hdev, HNAE3_INIT_CLIENT);
+	if (ret)
+		return ret;
+
+	/* clear handshake status with IMP */
+	hclgevf_reset_handshake(hdev, false);
+
+	/* bring up the nic to enable TX/RX again */
+	return hclgevf_notify_client(hdev, HNAE3_UP_CLIENT);
+}
+
+static int hclgevf_reset_prepare_wait(struct hclgevf_dev *hdev)
+{
+#define HCLGEVF_RESET_SYNC_TIME 100
+
+	if (hdev->reset_type == HNAE3_VF_FUNC_RESET) {
+		struct hclge_vf_to_pf_msg send_msg;
+		int ret;
+
+		hclgevf_build_send_msg(&send_msg, HCLGE_MBX_RESET, 0);
+		ret = hclgevf_send_mbx_msg(hdev, &send_msg, true, NULL, 0);
+		if (ret) {
+			dev_err(&hdev->pdev->dev,
+				"failed to assert VF reset, ret = %d\n", ret);
+			return ret;
+		}
+		hdev->rst_stats.vf_func_rst_cnt++;
+	}
+
+	set_bit(HCLGE_COMM_STATE_CMD_DISABLE, &hdev->hw.hw.comm_state);
+	/* inform hardware that preparatory work is done */
+	msleep(HCLGEVF_RESET_SYNC_TIME);
+	hclgevf_reset_handshake(hdev, true);
+	dev_info(&hdev->pdev->dev, "prepare reset(%d) wait done\n",
+		 hdev->reset_type);
+
+	return 0;
+}
+
+static void hclgevf_dump_rst_info(struct hclgevf_dev *hdev)
+{
+	dev_info(&hdev->pdev->dev, "VF function reset count: %u\n",
+		 hdev->rst_stats.vf_func_rst_cnt);
+	dev_info(&hdev->pdev->dev, "FLR reset count: %u\n",
+		 hdev->rst_stats.flr_rst_cnt);
+	dev_info(&hdev->pdev->dev, "VF reset count: %u\n",
+		 hdev->rst_stats.vf_rst_cnt);
+	dev_info(&hdev->pdev->dev, "reset done count: %u\n",
+		 hdev->rst_stats.rst_done_cnt);
+	dev_info(&hdev->pdev->dev, "HW reset done count: %u\n",
+		 hdev->rst_stats.hw_rst_done_cnt);
+	dev_info(&hdev->pdev->dev, "reset count: %u\n",
+		 hdev->rst_stats.rst_cnt);
+	dev_info(&hdev->pdev->dev, "reset fail count: %u\n",
+		 hdev->rst_stats.rst_fail_cnt);
+	dev_info(&hdev->pdev->dev, "vector0 interrupt enable status: 0x%x\n",
+		 hclgevf_read_dev(&hdev->hw, HCLGEVF_MISC_VECTOR_REG_BASE));
+	dev_info(&hdev->pdev->dev, "vector0 interrupt status: 0x%x\n",
+		 hclgevf_read_dev(&hdev->hw, HCLGE_COMM_VECTOR0_CMDQ_STATE_REG));
+	dev_info(&hdev->pdev->dev, "handshake status: 0x%x\n",
+		 hclgevf_read_dev(&hdev->hw, HCLGE_COMM_NIC_CSQ_DEPTH_REG));
+	dev_info(&hdev->pdev->dev, "function reset status: 0x%x\n",
+		 hclgevf_read_dev(&hdev->hw, HCLGEVF_RST_ING));
+	dev_info(&hdev->pdev->dev, "hdev state: 0x%lx\n", hdev->state);
+}
+
+static void hclgevf_reset_err_handle(struct hclgevf_dev *hdev)
+{
+	/* recover handshake status with IMP when reset fail */
+	hclgevf_reset_handshake(hdev, true);
+	hdev->rst_stats.rst_fail_cnt++;
+	dev_err(&hdev->pdev->dev, "failed to reset VF(%u)\n",
+		hdev->rst_stats.rst_fail_cnt);
+
+	if (hdev->rst_stats.rst_fail_cnt < HCLGEVF_RESET_MAX_FAIL_CNT)
+		hclgevf_set_reset_pending(hdev, hdev->reset_type);
+
+	if (hclgevf_is_reset_pending(hdev)) {
+		set_bit(HCLGEVF_RESET_PENDING, &hdev->reset_state);
+		hclgevf_reset_task_schedule(hdev);
+	} else {
+		set_bit(HCLGEVF_STATE_RST_FAIL, &hdev->state);
+		hclgevf_dump_rst_info(hdev);
+	}
+}
+
+static int hclgevf_reset_prepare(struct hclgevf_dev *hdev)
+{
+	int ret;
+
+	hdev->rst_stats.rst_cnt++;
+
+	/* perform reset of the stack & ae device for a client */
+	ret = hclgevf_notify_roce_client(hdev, HNAE3_DOWN_CLIENT);
+	if (ret)
+		return ret;
+
+	rtnl_lock();
+	/* bring down the nic to stop any ongoing TX/RX */
+	ret = hclgevf_notify_client(hdev, HNAE3_DOWN_CLIENT);
+	rtnl_unlock();
+	if (ret)
+		return ret;
+
+	return hclgevf_reset_prepare_wait(hdev);
+}
+
+static int hclgevf_reset_rebuild(struct hclgevf_dev *hdev)
+{
+	int ret;
+
+	hdev->rst_stats.hw_rst_done_cnt++;
+	ret = hclgevf_notify_roce_client(hdev, HNAE3_UNINIT_CLIENT);
+	if (ret)
+		return ret;
+
+	rtnl_lock();
+	/* now, re-initialize the nic client and ae device */
+	ret = hclgevf_reset_stack(hdev);
+	rtnl_unlock();
+	if (ret) {
+		dev_err(&hdev->pdev->dev, "failed to reset VF stack\n");
+		return ret;
+	}
+
+	ret = hclgevf_notify_roce_client(hdev, HNAE3_INIT_CLIENT);
+	/* ignore RoCE notify error if it fails HCLGEVF_RESET_MAX_FAIL_CNT - 1
+	 * times
+	 */
+	if (ret &&
+	    hdev->rst_stats.rst_fail_cnt < HCLGEVF_RESET_MAX_FAIL_CNT - 1)
+		return ret;
+
+	ret = hclgevf_notify_roce_client(hdev, HNAE3_UP_CLIENT);
+	if (ret)
+		return ret;
+
+	hdev->last_reset_time = jiffies;
+	hdev->rst_stats.rst_done_cnt++;
+	hdev->rst_stats.rst_fail_cnt = 0;
+	clear_bit(HCLGEVF_STATE_RST_FAIL, &hdev->state);
+
+	return 0;
+}
+
+static void hclgevf_reset(struct hclgevf_dev *hdev)
+{
+	if (hclgevf_reset_prepare(hdev))
+		goto err_reset;
+
+	/* check if VF could successfully fetch the hardware reset completion
+	 * status from the hardware
+	 */
+	if (hclgevf_reset_wait(hdev)) {
+		/* can't do much in this situation, will disable VF */
+		dev_err(&hdev->pdev->dev,
+			"failed to fetch H/W reset completion status\n");
+		goto err_reset;
+	}
+
+	if (hclgevf_reset_rebuild(hdev))
+		goto err_reset;
+
+	return;
+
+err_reset:
+	hclgevf_reset_err_handle(hdev);
+}
+
+static enum hnae3_reset_type hclgevf_get_reset_level(unsigned long *addr)
+{
+	enum hnae3_reset_type rst_level = HNAE3_NONE_RESET;
+
+	/* return the highest priority reset level amongst all */
+	if (test_bit(HNAE3_VF_RESET, addr)) {
+		rst_level = HNAE3_VF_RESET;
+		clear_bit(HNAE3_VF_RESET, addr);
+		clear_bit(HNAE3_VF_PF_FUNC_RESET, addr);
+		clear_bit(HNAE3_VF_FUNC_RESET, addr);
+	} else if (test_bit(HNAE3_VF_FULL_RESET, addr)) {
+		rst_level = HNAE3_VF_FULL_RESET;
+		clear_bit(HNAE3_VF_FULL_RESET, addr);
+		clear_bit(HNAE3_VF_FUNC_RESET, addr);
+	} else if (test_bit(HNAE3_VF_PF_FUNC_RESET, addr)) {
+		rst_level = HNAE3_VF_PF_FUNC_RESET;
+		clear_bit(HNAE3_VF_PF_FUNC_RESET, addr);
+		clear_bit(HNAE3_VF_FUNC_RESET, addr);
+	} else if (test_bit(HNAE3_VF_FUNC_RESET, addr)) {
+		rst_level = HNAE3_VF_FUNC_RESET;
+		clear_bit(HNAE3_VF_FUNC_RESET, addr);
+	} else if (test_bit(HNAE3_FLR_RESET, addr)) {
+		rst_level = HNAE3_FLR_RESET;
+		clear_bit(HNAE3_FLR_RESET, addr);
+	}
+
+	clear_bit(HNAE3_NONE_RESET, addr);
+
+	return rst_level;
+}
+
+static void hclgevf_reset_event(struct pci_dev *pdev,
+				struct hnae3_handle *handle)
+{
+	struct hnae3_ae_dev *ae_dev = pci_get_drvdata(pdev);
+	struct hclgevf_dev *hdev = ae_dev->priv;
+
+	if (hdev->default_reset_request)
+		hdev->reset_level =
+			hclgevf_get_reset_level(&hdev->default_reset_request);
+	else
+		hdev->reset_level = HNAE3_VF_FUNC_RESET;
+
+	dev_info(&hdev->pdev->dev, "received reset request from VF enet, reset level is %d\n",
+		 hdev->reset_level);
+
+	/* reset of this VF requested */
+	set_bit(HCLGEVF_RESET_REQUESTED, &hdev->reset_state);
+	hclgevf_reset_task_schedule(hdev);
+
+	hdev->last_reset_time = jiffies;
+}
+
+static void hclgevf_set_def_reset_request(struct hnae3_ae_dev *ae_dev,
+					  enum hnae3_reset_type rst_type)
+{
+#define HCLGEVF_SUPPORT_RESET_TYPE \
+	(BIT(HNAE3_VF_RESET) | BIT(HNAE3_VF_FUNC_RESET) | \
+	BIT(HNAE3_VF_PF_FUNC_RESET) | BIT(HNAE3_VF_FULL_RESET) | \
+	BIT(HNAE3_FLR_RESET) | BIT(HNAE3_VF_EXP_RESET))
+
+	struct hclgevf_dev *hdev = ae_dev->priv;
+
+	if (!(BIT(rst_type) & HCLGEVF_SUPPORT_RESET_TYPE)) {
+		/* To prevent reset triggered by hclge_reset_event */
+		set_bit(HNAE3_NONE_RESET, &hdev->default_reset_request);
+		dev_info(&hdev->pdev->dev, "unsupported reset type %d\n",
+			 rst_type);
+		return;
+	}
+	set_bit(rst_type, &hdev->default_reset_request);
+}
+
+static void hclgevf_enable_vector(struct hclgevf_misc_vector *vector, bool en)
+{
+	writel(en ? 1 : 0, vector->addr);
+}
+
+static void hclgevf_reset_prepare_general(struct hnae3_ae_dev *ae_dev,
+					  enum hnae3_reset_type rst_type)
+{
+#define HCLGEVF_RESET_RETRY_WAIT_MS	500
+#define HCLGEVF_RESET_RETRY_CNT		5
+
+	struct hclgevf_dev *hdev = ae_dev->priv;
+	int retry_cnt = 0;
+	int ret;
+
+	while (retry_cnt++ < HCLGEVF_RESET_RETRY_CNT) {
+		down(&hdev->reset_sem);
+		set_bit(HCLGEVF_STATE_RST_HANDLING, &hdev->state);
+		hdev->reset_type = rst_type;
+		ret = hclgevf_reset_prepare(hdev);
+		if (!ret && !hdev->reset_pending)
+			break;
+
+		dev_err(&hdev->pdev->dev,
+			"failed to prepare to reset, ret=%d, reset_pending:0x%lx, retry_cnt:%d\n",
+			ret, hdev->reset_pending, retry_cnt);
+		clear_bit(HCLGEVF_STATE_RST_HANDLING, &hdev->state);
+		up(&hdev->reset_sem);
+		msleep(HCLGEVF_RESET_RETRY_WAIT_MS);
+	}
+
+	/* disable misc vector before reset done */
+	hclgevf_enable_vector(&hdev->misc_vector, false);
+
+	if (hdev->reset_type == HNAE3_FLR_RESET)
+		hdev->rst_stats.flr_rst_cnt++;
+}
+
+static void hclgevf_reset_done(struct hnae3_ae_dev *ae_dev)
+{
+	struct hclgevf_dev *hdev = ae_dev->priv;
+	int ret;
+
+	hclgevf_enable_vector(&hdev->misc_vector, true);
+
+	ret = hclgevf_reset_rebuild(hdev);
+	if (ret)
+		dev_warn(&hdev->pdev->dev, "fail to rebuild, ret=%d\n",
+			 ret);
+
+	hdev->reset_type = HNAE3_NONE_RESET;
+	if (test_and_clear_bit(HCLGEVF_STATE_RST_HANDLING, &hdev->state))
+		up(&hdev->reset_sem);
+}
+
+static u32 hclgevf_get_fw_version(struct hnae3_handle *handle)
+{
+	struct hclgevf_dev *hdev = hclgevf_ae_get_hdev(handle);
+
+	return hdev->fw_version;
+}
+
+static void hclgevf_get_misc_vector(struct hclgevf_dev *hdev)
+{
+	struct hclgevf_misc_vector *vector = &hdev->misc_vector;
+
+	vector->vector_irq = pci_irq_vector(hdev->pdev,
+					    HCLGEVF_MISC_VECTOR_NUM);
+	vector->addr = hdev->hw.hw.io_base + HCLGEVF_MISC_VECTOR_REG_BASE;
+	/* vector status always valid for Vector 0 */
+	hdev->vector_status[HCLGEVF_MISC_VECTOR_NUM] = 0;
+	hdev->vector_irq[HCLGEVF_MISC_VECTOR_NUM] = vector->vector_irq;
+
+	hdev->num_msi_left -= 1;
+	hdev->num_msi_used += 1;
+}
+
+void hclgevf_reset_task_schedule(struct hclgevf_dev *hdev)
+{
+	if (!test_bit(HCLGEVF_STATE_REMOVING, &hdev->state) &&
+	    test_bit(HCLGEVF_STATE_SERVICE_INITED, &hdev->state) &&
+	    !test_and_set_bit(HCLGEVF_STATE_RST_SERVICE_SCHED,
+			      &hdev->state))
+		mod_delayed_work(hclgevf_wq, &hdev->service_task, 0);
+}
+
+void hclgevf_mbx_task_schedule(struct hclgevf_dev *hdev)
+{
+	if (!test_bit(HCLGEVF_STATE_REMOVING, &hdev->state) &&
+	    !test_and_set_bit(HCLGEVF_STATE_MBX_SERVICE_SCHED,
+			      &hdev->state))
+		mod_delayed_work(hclgevf_wq, &hdev->service_task, 0);
+}
+
+static void hclgevf_task_schedule(struct hclgevf_dev *hdev,
+				  unsigned long delay)
+{
+	if (!test_bit(HCLGEVF_STATE_REMOVING, &hdev->state) &&
+	    !test_bit(HCLGEVF_STATE_RST_FAIL, &hdev->state))
+		mod_delayed_work(hclgevf_wq, &hdev->service_task, delay);
+}
+
+static void hclgevf_reset_service_task(struct hclgevf_dev *hdev)
+{
+#define	HCLGEVF_MAX_RESET_ATTEMPTS_CNT	3
+
+	if (!test_and_clear_bit(HCLGEVF_STATE_RST_SERVICE_SCHED, &hdev->state))
+		return;
+
+	down(&hdev->reset_sem);
+	set_bit(HCLGEVF_STATE_RST_HANDLING, &hdev->state);
+
+	if (test_and_clear_bit(HCLGEVF_RESET_PENDING,
+			       &hdev->reset_state)) {
+		/* PF has intimated that it is about to reset the hardware.
+		 * We now have to poll & check if hardware has actually
+		 * completed the reset sequence. On hardware reset completion,
+		 * VF needs to reset the client and ae device.
+		 */
+		hdev->reset_attempts = 0;
+
+		hdev->last_reset_time = jiffies;
+		hdev->reset_type =
+			hclgevf_get_reset_level(&hdev->reset_pending);
+		if (hdev->reset_type != HNAE3_NONE_RESET)
+			hclgevf_reset(hdev);
+	} else if (test_and_clear_bit(HCLGEVF_RESET_REQUESTED,
+				      &hdev->reset_state)) {
+		/* we could be here when either of below happens:
+		 * 1. reset was initiated due to watchdog timeout caused by
+		 *    a. IMP was earlier reset and our TX got choked down and
+		 *       which resulted in watchdog reacting and inducing VF
+		 *       reset. This also means our cmdq would be unreliable.
+		 *    b. problem in TX due to other lower layer(example link
+		 *       layer not functioning properly etc.)
+		 * 2. VF reset might have been initiated due to some config
+		 *    change.
+		 *
+		 * NOTE: Theres no clear way to detect above cases than to react
+		 * to the response of PF for this reset request. PF will ack the
+		 * 1b and 2. cases but we will not get any intimation about 1a
+		 * from PF as cmdq would be in unreliable state i.e. mailbox
+		 * communication between PF and VF would be broken.
+		 *
+		 * if we are never geting into pending state it means either:
+		 * 1. PF is not receiving our request which could be due to IMP
+		 *    reset
+		 * 2. PF is screwed
+		 * We cannot do much for 2. but to check first we can try reset
+		 * our PCIe + stack and see if it alleviates the problem.
+		 */
+		if (hdev->reset_attempts > HCLGEVF_MAX_RESET_ATTEMPTS_CNT) {
+			/* prepare for full reset of stack + pcie interface */
+			hclgevf_set_reset_pending(hdev, HNAE3_VF_FULL_RESET);
+
+			/* "defer" schedule the reset task again */
+			set_bit(HCLGEVF_RESET_PENDING, &hdev->reset_state);
+		} else {
+			hdev->reset_attempts++;
+
+			hclgevf_set_reset_pending(hdev, hdev->reset_level);
+			set_bit(HCLGEVF_RESET_PENDING, &hdev->reset_state);
+		}
+		hclgevf_reset_task_schedule(hdev);
+	}
+
+	hdev->reset_type = HNAE3_NONE_RESET;
+	clear_bit(HCLGEVF_STATE_RST_HANDLING, &hdev->state);
+	up(&hdev->reset_sem);
+}
+
+static void hclgevf_mailbox_service_task(struct hclgevf_dev *hdev)
+{
+	if (!test_and_clear_bit(HCLGEVF_STATE_MBX_SERVICE_SCHED, &hdev->state))
+		return;
+
+	if (test_and_set_bit(HCLGEVF_STATE_MBX_HANDLING, &hdev->state))
+		return;
+
+	hclgevf_mbx_async_handler(hdev);
+
+	clear_bit(HCLGEVF_STATE_MBX_HANDLING, &hdev->state);
+}
+
+static void hclgevf_keep_alive(struct hclgevf_dev *hdev)
+{
+	struct hclge_vf_to_pf_msg send_msg;
+	int ret;
+
+	if (test_bit(HCLGE_COMM_STATE_CMD_DISABLE, &hdev->hw.hw.comm_state))
+		return;
+
+	hclgevf_build_send_msg(&send_msg, HCLGE_MBX_KEEP_ALIVE, 0);
+	ret = hclgevf_send_mbx_msg(hdev, &send_msg, false, NULL, 0);
+	if (ret)
+		dev_err(&hdev->pdev->dev,
+			"VF sends keep alive cmd failed(=%d)\n", ret);
+}
+
+static void hclgevf_periodic_service_task(struct hclgevf_dev *hdev)
+{
+	unsigned long delta = round_jiffies_relative(HZ);
+	struct hnae3_handle *handle = &hdev->nic;
+
+	if (test_bit(HCLGEVF_STATE_RST_FAIL, &hdev->state) ||
+	    test_bit(HCLGE_COMM_STATE_CMD_DISABLE, &hdev->hw.hw.comm_state))
+		return;
+
+	if (time_is_after_jiffies(hdev->last_serv_processed + HZ)) {
+		delta = jiffies - hdev->last_serv_processed;
+
+		if (delta < round_jiffies_relative(HZ)) {
+			delta = round_jiffies_relative(HZ) - delta;
+			goto out;
+		}
+	}
+
+	hdev->serv_processed_cnt++;
+	if (!(hdev->serv_processed_cnt % HCLGEVF_KEEP_ALIVE_TASK_INTERVAL))
+		hclgevf_keep_alive(hdev);
+
+	if (test_bit(HCLGEVF_STATE_DOWN, &hdev->state)) {
+		hdev->last_serv_processed = jiffies;
+		goto out;
+	}
+
+	if (!(hdev->serv_processed_cnt % HCLGEVF_STATS_TIMER_INTERVAL))
+		hclge_comm_tqps_update_stats(handle, &hdev->hw.hw);
+
+	/* VF does not need to request link status when this bit is set, because
+	 * PF will push its link status to VFs when link status changed.
+	 */
+	if (!test_bit(HCLGEVF_STATE_PF_PUSH_LINK_STATUS, &hdev->state))
+		hclgevf_request_link_info(hdev);
+
+	hclgevf_update_link_mode(hdev);
+
+	hclgevf_sync_vlan_filter(hdev);
+
+	hclgevf_sync_mac_table(hdev);
+
+	hclgevf_sync_promisc_mode(hdev);
+
+	hdev->last_serv_processed = jiffies;
+
+out:
+	hclgevf_task_schedule(hdev, delta);
+}
+
+static void hclgevf_service_task(struct work_struct *work)
+{
+	struct hclgevf_dev *hdev = container_of(work, struct hclgevf_dev,
+						service_task.work);
+
+	hclgevf_reset_service_task(hdev);
+	hclgevf_mailbox_service_task(hdev);
+	hclgevf_periodic_service_task(hdev);
+
+	/* Handle reset and mbx again in case periodical task delays the
+	 * handling by calling hclgevf_task_schedule() in
+	 * hclgevf_periodic_service_task()
+	 */
+	hclgevf_reset_service_task(hdev);
+	hclgevf_mailbox_service_task(hdev);
+}
+
+static void hclgevf_clear_event_cause(struct hclgevf_dev *hdev, u32 regclr)
+{
+	hclgevf_write_dev(&hdev->hw, HCLGE_COMM_VECTOR0_CMDQ_SRC_REG, regclr);
+}
+
+static enum hclgevf_evt_cause hclgevf_check_evt_cause(struct hclgevf_dev *hdev,
+						      u32 *clearval)
+{
+	u32 val, cmdq_stat_reg, rst_ing_reg;
+
+	/* fetch the events from their corresponding regs */
+	cmdq_stat_reg = hclgevf_read_dev(&hdev->hw,
+					 HCLGE_COMM_VECTOR0_CMDQ_STATE_REG);
+	if (BIT(HCLGEVF_VECTOR0_RST_INT_B) & cmdq_stat_reg) {
+		rst_ing_reg = hclgevf_read_dev(&hdev->hw, HCLGEVF_RST_ING);
+		dev_info(&hdev->pdev->dev,
+			 "receive reset interrupt 0x%x!\n", rst_ing_reg);
+		hclgevf_set_reset_pending(hdev, HNAE3_VF_RESET);
+		set_bit(HCLGEVF_RESET_PENDING, &hdev->reset_state);
+		set_bit(HCLGE_COMM_STATE_CMD_DISABLE, &hdev->hw.hw.comm_state);
+		*clearval = ~(1U << HCLGEVF_VECTOR0_RST_INT_B);
+		hdev->rst_stats.vf_rst_cnt++;
+		/* set up VF hardware reset status, its PF will clear
+		 * this status when PF has initialized done.
+		 */
+		val = hclgevf_read_dev(&hdev->hw, HCLGEVF_VF_RST_ING);
+		hclgevf_write_dev(&hdev->hw, HCLGEVF_VF_RST_ING,
+				  val | HCLGEVF_VF_RST_ING_BIT);
+		return HCLGEVF_VECTOR0_EVENT_RST;
+	}
+
+	/* check for vector0 mailbox(=CMDQ RX) event source */
+	if (BIT(HCLGEVF_VECTOR0_RX_CMDQ_INT_B) & cmdq_stat_reg) {
+		/* for revision 0x21, clearing interrupt is writing bit 0
+		 * to the clear register, writing bit 1 means to keep the
+		 * old value.
+		 * for revision 0x20, the clear register is a read & write
+		 * register, so we should just write 0 to the bit we are
+		 * handling, and keep other bits as cmdq_stat_reg.
+		 */
+		if (hdev->ae_dev->dev_version >= HNAE3_DEVICE_VERSION_V2)
+			*clearval = ~(1U << HCLGEVF_VECTOR0_RX_CMDQ_INT_B);
+		else
+			*clearval = cmdq_stat_reg &
+				    ~BIT(HCLGEVF_VECTOR0_RX_CMDQ_INT_B);
+
+		return HCLGEVF_VECTOR0_EVENT_MBX;
+	}
+
+	/* print other vector0 event source */
+	dev_info(&hdev->pdev->dev,
+		 "vector 0 interrupt from unknown source, cmdq_src = %#x\n",
+		 cmdq_stat_reg);
+
+	return HCLGEVF_VECTOR0_EVENT_OTHER;
+}
+
+static void hclgevf_reset_timer(struct timer_list *t)
+{
+	struct hclgevf_dev *hdev = timer_container_of(hdev, t, reset_timer);
+
+	hclgevf_clear_event_cause(hdev, HCLGEVF_VECTOR0_EVENT_RST);
+	hclgevf_reset_task_schedule(hdev);
+}
+
+static irqreturn_t hclgevf_misc_irq_handle(int irq, void *data)
+{
+#define HCLGEVF_RESET_DELAY	5
+
+	enum hclgevf_evt_cause event_cause;
+	struct hclgevf_dev *hdev = data;
+	u32 clearval;
+
+	hclgevf_enable_vector(&hdev->misc_vector, false);
+	event_cause = hclgevf_check_evt_cause(hdev, &clearval);
+	if (event_cause != HCLGEVF_VECTOR0_EVENT_OTHER)
+		hclgevf_clear_event_cause(hdev, clearval);
+
+	switch (event_cause) {
+	case HCLGEVF_VECTOR0_EVENT_RST:
+		mod_timer(&hdev->reset_timer,
+			  jiffies + msecs_to_jiffies(HCLGEVF_RESET_DELAY));
+		break;
+	case HCLGEVF_VECTOR0_EVENT_MBX:
+		hclgevf_mbx_handler(hdev);
+		break;
+	default:
+		break;
+	}
+
+	hclgevf_enable_vector(&hdev->misc_vector, true);
+
+	return IRQ_HANDLED;
+}
+
+static int hclgevf_configure(struct hclgevf_dev *hdev)
+{
+	int ret;
+
+	hdev->gro_en = true;
+
+	ret = hclgevf_get_basic_info(hdev);
+	if (ret)
+		return ret;
+
+	/* get current port based vlan state from PF */
+	ret = hclgevf_get_port_base_vlan_filter_state(hdev);
+	if (ret)
+		return ret;
+
+	/* get queue configuration from PF */
+	ret = hclgevf_get_queue_info(hdev);
+	if (ret)
+		return ret;
+
+	/* get queue depth info from PF */
+	ret = hclgevf_get_queue_depth(hdev);
+	if (ret)
+		return ret;
+
+	return hclgevf_get_pf_media_type(hdev);
+}
+
+static int hclgevf_alloc_hdev(struct hnae3_ae_dev *ae_dev)
+{
+	struct pci_dev *pdev = ae_dev->pdev;
+	struct hclgevf_dev *hdev;
+
+	hdev = devm_kzalloc(&pdev->dev, sizeof(*hdev), GFP_KERNEL);
+	if (!hdev)
+		return -ENOMEM;
+
+	hdev->pdev = pdev;
+	hdev->ae_dev = ae_dev;
+	ae_dev->priv = hdev;
+
+	return 0;
+}
+
+static int hclgevf_init_roce_base_info(struct hclgevf_dev *hdev)
+{
+	struct hnae3_handle *roce = &hdev->roce;
+	struct hnae3_handle *nic = &hdev->nic;
+
+	roce->rinfo.num_vectors = hdev->num_roce_msix;
+
+	if (hdev->num_msi_left < roce->rinfo.num_vectors ||
+	    hdev->num_msi_left == 0)
+		return -EINVAL;
+
+	roce->rinfo.base_vector = hdev->roce_base_msix_offset;
+
+	roce->rinfo.netdev = nic->kinfo.netdev;
+	roce->rinfo.roce_io_base = hdev->hw.hw.io_base;
+	roce->rinfo.roce_mem_base = hdev->hw.hw.mem_base;
+
+	roce->pdev = nic->pdev;
+	roce->ae_algo = nic->ae_algo;
+	bitmap_copy(roce->numa_node_mask.bits, nic->numa_node_mask.bits,
+		    MAX_NUMNODES);
+	return 0;
+}
+
+static int hclgevf_config_gro(struct hclgevf_dev *hdev)
+{
+	struct hclgevf_cfg_gro_status_cmd *req;
+	struct hclge_desc desc;
+	int ret;
+
+	if (!hnae3_ae_dev_gro_supported(hdev->ae_dev))
+		return 0;
+
+	hclgevf_cmd_setup_basic_desc(&desc, HCLGE_OPC_GRO_GENERIC_CONFIG,
+				     false);
+	req = (struct hclgevf_cfg_gro_status_cmd *)desc.data;
+
+	req->gro_en = hdev->gro_en ? 1 : 0;
+
+	ret = hclgevf_cmd_send(&hdev->hw, &desc, 1);
+	if (ret)
+		dev_err(&hdev->pdev->dev,
+			"VF GRO hardware config cmd failed, ret = %d.\n", ret);
+
+	return ret;
+}
+
+static int hclgevf_rss_init_hw(struct hclgevf_dev *hdev)
+{
+	struct hclge_comm_rss_cfg *rss_cfg = &hdev->rss_cfg;
+	u16 tc_offset[HCLGE_COMM_MAX_TC_NUM];
+	u16 tc_valid[HCLGE_COMM_MAX_TC_NUM];
+	u16 tc_size[HCLGE_COMM_MAX_TC_NUM];
+	int ret;
+
+	if (hdev->ae_dev->dev_version >= HNAE3_DEVICE_VERSION_V2) {
+		ret = hclge_comm_set_rss_algo_key(&hdev->hw.hw,
+						  rss_cfg->rss_algo,
+						  rss_cfg->rss_hash_key);
+		if (ret)
+			return ret;
+
+		ret = hclge_comm_set_rss_input_tuple(&hdev->hw.hw, rss_cfg);
+		if (ret)
+			return ret;
+	}
+
+	ret = hclge_comm_set_rss_indir_table(hdev->ae_dev, &hdev->hw.hw,
+					     rss_cfg->rss_indirection_tbl);
+	if (ret)
+		return ret;
+
+	hclge_comm_get_rss_tc_info(rss_cfg->rss_size, hdev->hw_tc_map,
+				   tc_offset, tc_valid, tc_size);
+
+	return hclge_comm_set_rss_tc_mode(&hdev->hw.hw, tc_offset,
+					  tc_valid, tc_size);
+}
+
+static int hclgevf_init_vlan_config(struct hclgevf_dev *hdev,
+				    bool rxvtag_strip_en)
+{
+	struct hnae3_handle *nic = &hdev->nic;
+	int ret;
+
+	ret = hclgevf_en_hw_strip_rxvtag(nic, rxvtag_strip_en);
+	if (ret) {
+		dev_err(&hdev->pdev->dev,
+			"failed to enable rx vlan offload, ret = %d\n", ret);
+		return ret;
+	}
+
+	return hclgevf_set_vlan_filter(&hdev->nic, htons(ETH_P_8021Q), 0,
+				       false);
+}
+
+static void hclgevf_flush_link_update(struct hclgevf_dev *hdev)
+{
+#define HCLGEVF_FLUSH_LINK_TIMEOUT	100000
+
+	unsigned long last = hdev->serv_processed_cnt;
+	int i = 0;
+
+	while (test_bit(HCLGEVF_STATE_LINK_UPDATING, &hdev->state) &&
+	       i++ < HCLGEVF_FLUSH_LINK_TIMEOUT &&
+	       last == hdev->serv_processed_cnt)
+		usleep_range(1, 1);
+}
+
+static void hclgevf_set_timer_task(struct hnae3_handle *handle, bool enable)
+{
+	struct hclgevf_dev *hdev = hclgevf_ae_get_hdev(handle);
+
+	if (enable) {
+		hclgevf_task_schedule(hdev, 0);
+	} else {
+		set_bit(HCLGEVF_STATE_DOWN, &hdev->state);
+
+		smp_mb__after_atomic(); /* flush memory to make sure DOWN is seen by service task */
+		hclgevf_flush_link_update(hdev);
+	}
+}
+
+static int hclgevf_ae_start(struct hnae3_handle *handle)
+{
+	struct hclgevf_dev *hdev = hclgevf_ae_get_hdev(handle);
+
+	clear_bit(HCLGEVF_STATE_DOWN, &hdev->state);
+	clear_bit(HCLGEVF_STATE_PF_PUSH_LINK_STATUS, &hdev->state);
+
+	hclge_comm_reset_tqp_stats(handle);
+
+	hclgevf_request_link_info(hdev);
+
+	hclgevf_update_link_mode(hdev);
+
+	return 0;
+}
+
+static void hclgevf_ae_stop(struct hnae3_handle *handle)
+{
+	struct hclgevf_dev *hdev = hclgevf_ae_get_hdev(handle);
+
+	set_bit(HCLGEVF_STATE_DOWN, &hdev->state);
+
+	if (hdev->reset_type != HNAE3_VF_RESET)
+		hclgevf_reset_tqp(handle);
+
+	hclge_comm_reset_tqp_stats(handle);
+	hclgevf_update_link_status(hdev, 0);
+}
+
+static int hclgevf_set_alive(struct hnae3_handle *handle, bool alive)
+{
+#define HCLGEVF_STATE_ALIVE	1
+#define HCLGEVF_STATE_NOT_ALIVE	0
+
+	struct hclgevf_dev *hdev = hclgevf_ae_get_hdev(handle);
+	struct hclge_vf_to_pf_msg send_msg;
+
+	hclgevf_build_send_msg(&send_msg, HCLGE_MBX_SET_ALIVE, 0);
+	send_msg.data[0] = alive ? HCLGEVF_STATE_ALIVE :
+				HCLGEVF_STATE_NOT_ALIVE;
+	return hclgevf_send_mbx_msg(hdev, &send_msg, false, NULL, 0);
+}
+
+static int hclgevf_client_start(struct hnae3_handle *handle)
+{
+	return hclgevf_set_alive(handle, true);
+}
+
+static void hclgevf_client_stop(struct hnae3_handle *handle)
+{
+	struct hclgevf_dev *hdev = hclgevf_ae_get_hdev(handle);
+	int ret;
+
+	ret = hclgevf_set_alive(handle, false);
+	if (ret)
+		dev_warn(&hdev->pdev->dev,
+			 "%s failed %d\n", __func__, ret);
+}
+
+static void hclgevf_state_init(struct hclgevf_dev *hdev)
+{
+	clear_bit(HCLGEVF_STATE_MBX_SERVICE_SCHED, &hdev->state);
+	clear_bit(HCLGEVF_STATE_MBX_HANDLING, &hdev->state);
+	clear_bit(HCLGEVF_STATE_RST_FAIL, &hdev->state);
+
+	INIT_DELAYED_WORK(&hdev->service_task, hclgevf_service_task);
+	/* timer needs to be initialized before misc irq */
+	timer_setup(&hdev->reset_timer, hclgevf_reset_timer, 0);
+
+	mutex_init(&hdev->mbx_resp.mbx_mutex);
+	sema_init(&hdev->reset_sem, 1);
+
+	spin_lock_init(&hdev->mac_table.mac_list_lock);
+	INIT_LIST_HEAD(&hdev->mac_table.uc_mac_list);
+	INIT_LIST_HEAD(&hdev->mac_table.mc_mac_list);
+
+	/* bring the device down */
+	set_bit(HCLGEVF_STATE_DOWN, &hdev->state);
+}
+
+static void hclgevf_state_uninit(struct hclgevf_dev *hdev)
+{
+	set_bit(HCLGEVF_STATE_DOWN, &hdev->state);
+	set_bit(HCLGEVF_STATE_REMOVING, &hdev->state);
+
+	if (hdev->service_task.work.func)
+		cancel_delayed_work_sync(&hdev->service_task);
+
+	mutex_destroy(&hdev->mbx_resp.mbx_mutex);
+}
+
+static int hclgevf_init_msi(struct hclgevf_dev *hdev)
+{
+	struct pci_dev *pdev = hdev->pdev;
+	int vectors;
+	int i;
+
+	if (hnae3_dev_roce_supported(hdev))
+		vectors = pci_alloc_irq_vectors(pdev,
+						hdev->roce_base_msix_offset + 1,
+						hdev->num_msi,
+						PCI_IRQ_MSIX);
+	else
+		vectors = pci_alloc_irq_vectors(pdev, HNAE3_MIN_VECTOR_NUM,
+						hdev->num_msi,
+						PCI_IRQ_MSI | PCI_IRQ_MSIX);
+
+	if (vectors < 0) {
+		dev_err(&pdev->dev,
+			"failed(%d) to allocate MSI/MSI-X vectors\n",
+			vectors);
+		return vectors;
+	}
+	if (vectors < hdev->num_msi)
+		dev_warn(&hdev->pdev->dev,
+			 "requested %u MSI/MSI-X, but allocated %d MSI/MSI-X\n",
+			 hdev->num_msi, vectors);
+
+	hdev->num_msi = vectors;
+	hdev->num_msi_left = vectors;
+
+	hdev->vector_status = devm_kcalloc(&pdev->dev, hdev->num_msi,
+					   sizeof(u16), GFP_KERNEL);
+	if (!hdev->vector_status) {
+		pci_free_irq_vectors(pdev);
+		return -ENOMEM;
+	}
+
+	for (i = 0; i < hdev->num_msi; i++)
+		hdev->vector_status[i] = HCLGEVF_INVALID_VPORT;
+
+	hdev->vector_irq = devm_kcalloc(&pdev->dev, hdev->num_msi,
+					sizeof(int), GFP_KERNEL);
+	if (!hdev->vector_irq) {
+		devm_kfree(&pdev->dev, hdev->vector_status);
+		pci_free_irq_vectors(pdev);
+		return -ENOMEM;
+	}
+
+	return 0;
+}
+
+static void hclgevf_uninit_msi(struct hclgevf_dev *hdev)
+{
+	struct pci_dev *pdev = hdev->pdev;
+
+	devm_kfree(&pdev->dev, hdev->vector_status);
+	devm_kfree(&pdev->dev, hdev->vector_irq);
+	pci_free_irq_vectors(pdev);
+}
+
+static int hclgevf_misc_irq_init(struct hclgevf_dev *hdev)
+{
+	int ret;
+
+	hclgevf_get_misc_vector(hdev);
+
+	snprintf(hdev->misc_vector.name, HNAE3_INT_NAME_LEN, "%s-misc-%s",
+		 HCLGEVF_NAME, pci_name(hdev->pdev));
+	ret = request_irq(hdev->misc_vector.vector_irq, hclgevf_misc_irq_handle,
+			  0, hdev->misc_vector.name, hdev);
+	if (ret) {
+		dev_err(&hdev->pdev->dev, "VF failed to request misc irq(%d)\n",
+			hdev->misc_vector.vector_irq);
+		return ret;
+	}
+
+	hclgevf_clear_event_cause(hdev, 0);
+
+	/* enable misc. vector(vector 0) */
+	hclgevf_enable_vector(&hdev->misc_vector, true);
+
+	return ret;
+}
+
+static void hclgevf_misc_irq_uninit(struct hclgevf_dev *hdev)
+{
+	/* disable misc vector(vector 0) */
+	hclgevf_enable_vector(&hdev->misc_vector, false);
+	synchronize_irq(hdev->misc_vector.vector_irq);
+	free_irq(hdev->misc_vector.vector_irq, hdev);
+	hclgevf_free_vector(hdev, 0);
+}
+
+static void hclgevf_info_show(struct hclgevf_dev *hdev)
+{
+	struct device *dev = &hdev->pdev->dev;
+
+	dev_info(dev, "VF info begin:\n");
+
+	dev_info(dev, "Task queue pairs numbers: %u\n", hdev->num_tqps);
+	dev_info(dev, "Desc num per TX queue: %u\n", hdev->num_tx_desc);
+	dev_info(dev, "Desc num per RX queue: %u\n", hdev->num_rx_desc);
+	dev_info(dev, "Numbers of vports: %u\n", hdev->num_alloc_vport);
+	dev_info(dev, "HW tc map: 0x%x\n", hdev->hw_tc_map);
+	dev_info(dev, "PF media type of this VF: %u\n",
+		 hdev->hw.mac.media_type);
+
+	dev_info(dev, "VF info end.\n");
+}
+
+static int hclgevf_init_nic_client_instance(struct hnae3_ae_dev *ae_dev,
+					    struct hnae3_client *client)
+{
+	struct hclgevf_dev *hdev = ae_dev->priv;
+	u32 rst_cnt = hdev->rst_stats.rst_cnt;
+	int ret;
+
+	ret = client->ops->init_instance(&hdev->nic);
+	if (ret)
+		return ret;
+
+	set_bit(HCLGEVF_STATE_NIC_REGISTERED, &hdev->state);
+	if (test_bit(HCLGEVF_STATE_RST_HANDLING, &hdev->state) ||
+	    rst_cnt != hdev->rst_stats.rst_cnt) {
+		clear_bit(HCLGEVF_STATE_NIC_REGISTERED, &hdev->state);
+
+		client->ops->uninit_instance(&hdev->nic, 0);
+		return -EBUSY;
+	}
+
+	hnae3_set_client_init_flag(client, ae_dev, 1);
+
+	if (netif_msg_drv(&hdev->nic))
+		hclgevf_info_show(hdev);
+
+	return 0;
+}
+
+static int hclgevf_init_roce_client_instance(struct hnae3_ae_dev *ae_dev,
+					     struct hnae3_client *client)
+{
+	struct hclgevf_dev *hdev = ae_dev->priv;
+	int ret;
+
+	if (!hnae3_dev_roce_supported(hdev) || !hdev->roce_client ||
+	    !hdev->nic_client)
+		return 0;
+
+	ret = hclgevf_init_roce_base_info(hdev);
+	if (ret)
+		return ret;
+
+	ret = client->ops->init_instance(&hdev->roce);
+	if (ret)
+		return ret;
+
+	set_bit(HCLGEVF_STATE_ROCE_REGISTERED, &hdev->state);
+	hnae3_set_client_init_flag(client, ae_dev, 1);
+
+	return 0;
+}
+
+static int hclgevf_init_client_instance(struct hnae3_client *client,
+					struct hnae3_ae_dev *ae_dev)
+{
+	struct hclgevf_dev *hdev = ae_dev->priv;
+	int ret;
+
+	switch (client->type) {
+	case HNAE3_CLIENT_KNIC:
+		hdev->nic_client = client;
+		hdev->nic.client = client;
+
+		ret = hclgevf_init_nic_client_instance(ae_dev, client);
+		if (ret)
+			goto clear_nic;
+
+		ret = hclgevf_init_roce_client_instance(ae_dev,
+							hdev->roce_client);
+		if (ret)
+			goto clear_roce;
+
+		break;
+	case HNAE3_CLIENT_ROCE:
+		if (hnae3_dev_roce_supported(hdev)) {
+			hdev->roce_client = client;
+			hdev->roce.client = client;
+		}
+
+		ret = hclgevf_init_roce_client_instance(ae_dev, client);
+		if (ret)
+			goto clear_roce;
+
+		break;
+	default:
+		return -EINVAL;
+	}
+
+	return 0;
+
+clear_nic:
+	hdev->nic_client = NULL;
+	hdev->nic.client = NULL;
+	return ret;
+clear_roce:
+	hdev->roce_client = NULL;
+	hdev->roce.client = NULL;
+	return ret;
+}
+
+static void hclgevf_uninit_client_instance(struct hnae3_client *client,
+					   struct hnae3_ae_dev *ae_dev)
+{
+	struct hclgevf_dev *hdev = ae_dev->priv;
+
+	/* un-init roce, if it exists */
+	if (hdev->roce_client) {
+		while (test_bit(HCLGEVF_STATE_RST_HANDLING, &hdev->state))
+			msleep(HCLGEVF_WAIT_RESET_DONE);
+		clear_bit(HCLGEVF_STATE_ROCE_REGISTERED, &hdev->state);
+
+		hdev->roce_client->ops->uninit_instance(&hdev->roce, 0);
+		hdev->roce_client = NULL;
+		hdev->roce.client = NULL;
+	}
+
+	/* un-init nic/unic, if this was not called by roce client */
+	if (client->ops->uninit_instance && hdev->nic_client &&
+	    client->type != HNAE3_CLIENT_ROCE) {
+		while (test_bit(HCLGEVF_STATE_RST_HANDLING, &hdev->state))
+			msleep(HCLGEVF_WAIT_RESET_DONE);
+		clear_bit(HCLGEVF_STATE_NIC_REGISTERED, &hdev->state);
+
+		client->ops->uninit_instance(&hdev->nic, 0);
+		hdev->nic_client = NULL;
+		hdev->nic.client = NULL;
+	}
+}
+
+static int hclgevf_dev_mem_map(struct hclgevf_dev *hdev)
+{
+	struct pci_dev *pdev = hdev->pdev;
+	struct hclgevf_hw *hw = &hdev->hw;
+
+	/* for device does not have device memory, return directly */
+	if (!(pci_select_bars(pdev, IORESOURCE_MEM) & BIT(HCLGEVF_MEM_BAR)))
+		return 0;
+
+	hw->hw.mem_base =
+		devm_ioremap_wc(&pdev->dev,
+				pci_resource_start(pdev, HCLGEVF_MEM_BAR),
+				pci_resource_len(pdev, HCLGEVF_MEM_BAR));
+	if (!hw->hw.mem_base) {
+		dev_err(&pdev->dev, "failed to map device memory\n");
+		return -EFAULT;
+	}
+
+	return 0;
+}
+
+static int hclgevf_pci_init(struct hclgevf_dev *hdev)
+{
+	struct pci_dev *pdev = hdev->pdev;
+	struct hclgevf_hw *hw;
+	int ret;
+
+	ret = pci_enable_device(pdev);
+	if (ret) {
+		dev_err(&pdev->dev, "failed to enable PCI device\n");
+		return ret;
+	}
+
+	ret = dma_set_mask_and_coherent(&pdev->dev, DMA_BIT_MASK(64));
+	if (ret) {
+		dev_err(&pdev->dev, "can't set consistent PCI DMA, exiting\n");
+		goto err_disable_device;
+	}
+
+	ret = pci_request_regions(pdev, HCLGEVF_DRIVER_NAME);
+	if (ret) {
+		dev_err(&pdev->dev, "PCI request regions failed %d\n", ret);
+		goto err_disable_device;
+	}
+
+	pci_set_master(pdev);
+	hw = &hdev->hw;
+	hw->hw.io_base = pci_iomap(pdev, 2, 0);
+	if (!hw->hw.io_base) {
+		dev_err(&pdev->dev, "can't map configuration register space\n");
+		ret = -ENOMEM;
+		goto err_release_regions;
+	}
+
+	ret = hclgevf_dev_mem_map(hdev);
+	if (ret)
+		goto err_unmap_io_base;
+
+	return 0;
+
+err_unmap_io_base:
+	pci_iounmap(pdev, hdev->hw.hw.io_base);
+err_release_regions:
+	pci_release_regions(pdev);
+err_disable_device:
+	pci_disable_device(pdev);
+
+	return ret;
+}
+
+static void hclgevf_pci_uninit(struct hclgevf_dev *hdev)
+{
+	struct pci_dev *pdev = hdev->pdev;
+
+	if (hdev->hw.hw.mem_base)
+		devm_iounmap(&pdev->dev, hdev->hw.hw.mem_base);
+
+	pci_iounmap(pdev, hdev->hw.hw.io_base);
+	pci_release_regions(pdev);
+	pci_disable_device(pdev);
+}
+
+static int hclgevf_query_vf_resource(struct hclgevf_dev *hdev)
+{
+	struct hclgevf_query_res_cmd *req;
+	struct hclge_desc desc;
+	int ret;
+
+	hclgevf_cmd_setup_basic_desc(&desc, HCLGE_OPC_QUERY_VF_RSRC, true);
+	ret = hclgevf_cmd_send(&hdev->hw, &desc, 1);
+	if (ret) {
+		dev_err(&hdev->pdev->dev,
+			"query vf resource failed, ret = %d.\n", ret);
+		return ret;
+	}
+
+	req = (struct hclgevf_query_res_cmd *)desc.data;
+
+	if (hnae3_dev_roce_supported(hdev)) {
+		hdev->roce_base_msix_offset =
+		hnae3_get_field(le16_to_cpu(req->msixcap_localid_ba_rocee),
+				HCLGEVF_MSIX_OFT_ROCEE_M,
+				HCLGEVF_MSIX_OFT_ROCEE_S);
+		hdev->num_roce_msix =
+		hnae3_get_field(le16_to_cpu(req->vf_intr_vector_number),
+				HCLGEVF_VEC_NUM_M, HCLGEVF_VEC_NUM_S);
+
+		/* nic's msix numbers is always equals to the roce's. */
+		hdev->num_nic_msix = hdev->num_roce_msix;
+
+		/* VF should have NIC vectors and Roce vectors, NIC vectors
+		 * are queued before Roce vectors. The offset is fixed to 64.
+		 */
+		hdev->num_msi = hdev->num_roce_msix +
+				hdev->roce_base_msix_offset;
+	} else {
+		hdev->num_msi =
+		hnae3_get_field(le16_to_cpu(req->vf_intr_vector_number),
+				HCLGEVF_VEC_NUM_M, HCLGEVF_VEC_NUM_S);
+
+		hdev->num_nic_msix = hdev->num_msi;
+	}
+
+	if (hdev->num_nic_msix < HNAE3_MIN_VECTOR_NUM) {
+		dev_err(&hdev->pdev->dev,
+			"Just %u msi resources, not enough for vf(min:2).\n",
+			hdev->num_nic_msix);
+		return -EINVAL;
+	}
+
+	return 0;
+}
+
+static void hclgevf_set_default_dev_specs(struct hclgevf_dev *hdev)
+{
+#define HCLGEVF_MAX_NON_TSO_BD_NUM			8U
+
+	struct hnae3_ae_dev *ae_dev = pci_get_drvdata(hdev->pdev);
+
+	ae_dev->dev_specs.max_non_tso_bd_num =
+					HCLGEVF_MAX_NON_TSO_BD_NUM;
+	ae_dev->dev_specs.rss_ind_tbl_size = HCLGEVF_RSS_IND_TBL_SIZE;
+	ae_dev->dev_specs.rss_key_size = HCLGE_COMM_RSS_KEY_SIZE;
+	ae_dev->dev_specs.max_int_gl = HCLGEVF_DEF_MAX_INT_GL;
+	ae_dev->dev_specs.max_frm_size = HCLGEVF_MAC_MAX_FRAME;
+}
+
+static void hclgevf_parse_dev_specs(struct hclgevf_dev *hdev,
+				    struct hclge_desc *desc)
+{
+	struct hnae3_ae_dev *ae_dev = pci_get_drvdata(hdev->pdev);
+	struct hclgevf_dev_specs_0_cmd *req0;
+	struct hclgevf_dev_specs_1_cmd *req1;
+
+	req0 = (struct hclgevf_dev_specs_0_cmd *)desc[0].data;
+	req1 = (struct hclgevf_dev_specs_1_cmd *)desc[1].data;
+
+	ae_dev->dev_specs.max_non_tso_bd_num = req0->max_non_tso_bd_num;
+	ae_dev->dev_specs.rss_ind_tbl_size =
+					le16_to_cpu(req0->rss_ind_tbl_size);
+	ae_dev->dev_specs.int_ql_max = le16_to_cpu(req0->int_ql_max);
+	ae_dev->dev_specs.rss_key_size = le16_to_cpu(req0->rss_key_size);
+	ae_dev->dev_specs.max_int_gl = le16_to_cpu(req1->max_int_gl);
+	ae_dev->dev_specs.max_frm_size = le16_to_cpu(req1->max_frm_size);
+}
+
+static void hclgevf_check_dev_specs(struct hclgevf_dev *hdev)
+{
+	struct hnae3_dev_specs *dev_specs = &hdev->ae_dev->dev_specs;
+
+	if (!dev_specs->max_non_tso_bd_num)
+		dev_specs->max_non_tso_bd_num = HCLGEVF_MAX_NON_TSO_BD_NUM;
+	if (!dev_specs->rss_ind_tbl_size)
+		dev_specs->rss_ind_tbl_size = HCLGEVF_RSS_IND_TBL_SIZE;
+	if (!dev_specs->rss_key_size)
+		dev_specs->rss_key_size = HCLGE_COMM_RSS_KEY_SIZE;
+	if (!dev_specs->max_int_gl)
+		dev_specs->max_int_gl = HCLGEVF_DEF_MAX_INT_GL;
+	if (!dev_specs->max_frm_size)
+		dev_specs->max_frm_size = HCLGEVF_MAC_MAX_FRAME;
+}
+
+static int hclgevf_query_dev_specs(struct hclgevf_dev *hdev)
+{
+	struct hclge_desc desc[HCLGEVF_QUERY_DEV_SPECS_BD_NUM];
+	int ret;
+	int i;
+
+	/* set default specifications as devices lower than version V3 do not
+	 * support querying specifications from firmware.
+	 */
+	if (hdev->ae_dev->dev_version < HNAE3_DEVICE_VERSION_V3) {
+		hclgevf_set_default_dev_specs(hdev);
+		return 0;
+	}
+
+	for (i = 0; i < HCLGEVF_QUERY_DEV_SPECS_BD_NUM - 1; i++) {
+		hclgevf_cmd_setup_basic_desc(&desc[i],
+					     HCLGE_OPC_QUERY_DEV_SPECS, true);
+		desc[i].flag |= cpu_to_le16(HCLGE_COMM_CMD_FLAG_NEXT);
+	}
+	hclgevf_cmd_setup_basic_desc(&desc[i], HCLGE_OPC_QUERY_DEV_SPECS, true);
+
+	ret = hclgevf_cmd_send(&hdev->hw, desc, HCLGEVF_QUERY_DEV_SPECS_BD_NUM);
+	if (ret)
+		return ret;
+
+	hclgevf_parse_dev_specs(hdev, desc);
+	hclgevf_check_dev_specs(hdev);
+
+	return 0;
+}
+
+static int hclgevf_pci_reset(struct hclgevf_dev *hdev)
+{
+	struct pci_dev *pdev = hdev->pdev;
+	int ret = 0;
+
+	if ((hdev->reset_type == HNAE3_VF_FULL_RESET ||
+	     hdev->reset_type == HNAE3_FLR_RESET) &&
+	    test_bit(HCLGEVF_STATE_IRQ_INITED, &hdev->state)) {
+		hclgevf_misc_irq_uninit(hdev);
+		hclgevf_uninit_msi(hdev);
+		clear_bit(HCLGEVF_STATE_IRQ_INITED, &hdev->state);
+	}
+
+	if (!test_bit(HCLGEVF_STATE_IRQ_INITED, &hdev->state)) {
+		pci_set_master(pdev);
+		ret = hclgevf_init_msi(hdev);
+		if (ret) {
+			dev_err(&pdev->dev,
+				"failed(%d) to init MSI/MSI-X\n", ret);
+			return ret;
+		}
+
+		ret = hclgevf_misc_irq_init(hdev);
+		if (ret) {
+			hclgevf_uninit_msi(hdev);
+			dev_err(&pdev->dev, "failed(%d) to init Misc IRQ(vector0)\n",
+				ret);
+			return ret;
+		}
+
+		set_bit(HCLGEVF_STATE_IRQ_INITED, &hdev->state);
+	}
+
+	return ret;
+}
+
+static int hclgevf_clear_vport_list(struct hclgevf_dev *hdev)
+{
+	struct hclge_vf_to_pf_msg send_msg;
+
+	hclgevf_build_send_msg(&send_msg, HCLGE_MBX_HANDLE_VF_TBL,
+			       HCLGE_MBX_VPORT_LIST_CLEAR);
+	return hclgevf_send_mbx_msg(hdev, &send_msg, false, NULL, 0);
+}
+
+static void hclgevf_init_rxd_adv_layout(struct hclgevf_dev *hdev)
+{
+	if (hnae3_ae_dev_rxd_adv_layout_supported(hdev->ae_dev))
+		hclgevf_write_dev(&hdev->hw, HCLGEVF_RXD_ADV_LAYOUT_EN_REG, 1);
+}
+
+static void hclgevf_uninit_rxd_adv_layout(struct hclgevf_dev *hdev)
+{
+	if (hnae3_ae_dev_rxd_adv_layout_supported(hdev->ae_dev))
+		hclgevf_write_dev(&hdev->hw, HCLGEVF_RXD_ADV_LAYOUT_EN_REG, 0);
+}
+
+static int hclgevf_reset_hdev(struct hclgevf_dev *hdev)
+{
+	struct pci_dev *pdev = hdev->pdev;
+	int ret;
+
+	ret = hclgevf_pci_reset(hdev);
+	if (ret) {
+		dev_err(&pdev->dev, "pci reset failed %d\n", ret);
+		return ret;
+	}
+
+	hclgevf_arq_init(hdev);
+
+	ret = hclge_comm_cmd_init(hdev->ae_dev, &hdev->hw.hw,
+				  &hdev->fw_version, false,
+				  hdev->reset_pending);
+	if (ret) {
+		dev_err(&pdev->dev, "cmd failed %d\n", ret);
+		return ret;
+	}
+
+	ret = hclgevf_rss_init_hw(hdev);
+	if (ret) {
+		dev_err(&hdev->pdev->dev,
+			"failed(%d) to initialize RSS\n", ret);
+		return ret;
+	}
+
+	ret = hclgevf_config_gro(hdev);
+	if (ret)
+		return ret;
+
+	ret = hclgevf_init_vlan_config(hdev, hdev->rxvtag_strip_en);
+	if (ret) {
+		dev_err(&hdev->pdev->dev,
+			"failed(%d) to initialize VLAN config\n", ret);
+		return ret;
+	}
+
+	/* get current port based vlan state from PF */
+	ret = hclgevf_get_port_base_vlan_filter_state(hdev);
+	if (ret)
+		return ret;
+
+	set_bit(HCLGEVF_STATE_PROMISC_CHANGED, &hdev->state);
+
+	hclgevf_init_rxd_adv_layout(hdev);
+
+	dev_info(&hdev->pdev->dev, "Reset done\n");
+
+	return 0;
+}
+
+static int hclgevf_init_hdev(struct hclgevf_dev *hdev)
+{
+	struct pci_dev *pdev = hdev->pdev;
+	int ret;
+
+	ret = hclgevf_pci_init(hdev);
+	if (ret)
+		return ret;
+
+	ret = hclge_comm_cmd_queue_init(hdev->pdev, &hdev->hw.hw);
+	if (ret)
+		goto err_cmd_queue_init;
+
+	hclgevf_arq_init(hdev);
+
+	hclge_comm_cmd_init_ops(&hdev->hw.hw, &hclgevf_cmq_ops);
+	ret = hclge_comm_cmd_init(hdev->ae_dev, &hdev->hw.hw,
+				  &hdev->fw_version, false,
+				  hdev->reset_pending);
+	if (ret)
+		goto err_cmd_init;
+
+	/* Get vf resource */
+	ret = hclgevf_query_vf_resource(hdev);
+	if (ret)
+		goto err_cmd_init;
+
+	ret = hclgevf_query_dev_specs(hdev);
+	if (ret) {
+		dev_err(&pdev->dev,
+			"failed to query dev specifications, ret = %d\n", ret);
+		goto err_cmd_init;
+	}
+
+	ret = hclgevf_init_msi(hdev);
+	if (ret) {
+		dev_err(&pdev->dev, "failed(%d) to init MSI/MSI-X\n", ret);
+		goto err_cmd_init;
+	}
+
+	hclgevf_state_init(hdev);
+	hdev->reset_level = HNAE3_VF_FUNC_RESET;
+	hdev->reset_type = HNAE3_NONE_RESET;
+
+	ret = hclgevf_misc_irq_init(hdev);
+	if (ret)
+		goto err_misc_irq_init;
+
+	set_bit(HCLGEVF_STATE_IRQ_INITED, &hdev->state);
+
+	ret = hclgevf_configure(hdev);
+	if (ret) {
+		dev_err(&pdev->dev, "failed(%d) to fetch configuration\n", ret);
+		goto err_config;
+	}
+
+	ret = hclgevf_alloc_tqps(hdev);
+	if (ret) {
+		dev_err(&pdev->dev, "failed(%d) to allocate TQPs\n", ret);
+		goto err_config;
+	}
+
+	ret = hclgevf_set_handle_info(hdev);
+	if (ret)
+		goto err_config;
+
+	ret = hclgevf_config_gro(hdev);
+	if (ret)
+		goto err_config;
+
+	/* Initialize RSS for this VF */
+	ret = hclge_comm_rss_init_cfg(&hdev->nic, hdev->ae_dev,
+				      &hdev->rss_cfg);
+	if (ret) {
+		dev_err(&pdev->dev, "failed to init rss cfg, ret = %d\n", ret);
+		goto err_config;
+	}
+
+	ret = hclgevf_rss_init_hw(hdev);
+	if (ret) {
+		dev_err(&hdev->pdev->dev,
+			"failed(%d) to initialize RSS\n", ret);
+		goto err_config;
+	}
+
+	/* ensure vf tbl list as empty before init */
+	ret = hclgevf_clear_vport_list(hdev);
+	if (ret) {
+		dev_err(&pdev->dev,
+			"failed to clear tbl list configuration, ret = %d.\n",
+			ret);
+		goto err_config;
+	}
+
+	ret = hclgevf_init_vlan_config(hdev, true);
+	if (ret) {
+		dev_err(&hdev->pdev->dev,
+			"failed(%d) to initialize VLAN config\n", ret);
+		goto err_config;
+	}
+
+	hclgevf_init_rxd_adv_layout(hdev);
+
+	ret = hclgevf_devlink_init(hdev);
+	if (ret)
+		goto err_config;
+
+	set_bit(HCLGEVF_STATE_SERVICE_INITED, &hdev->state);
+
+	hdev->last_reset_time = jiffies;
+	dev_info(&hdev->pdev->dev, "finished initializing %s driver\n",
+		 HCLGEVF_DRIVER_NAME);
+
+	hclgevf_task_schedule(hdev, round_jiffies_relative(HZ));
+
+	return 0;
+
+err_config:
+	hclgevf_misc_irq_uninit(hdev);
+err_misc_irq_init:
+	hclgevf_state_uninit(hdev);
+	hclgevf_uninit_msi(hdev);
+err_cmd_init:
+	hclge_comm_cmd_uninit(hdev->ae_dev, &hdev->hw.hw);
+err_cmd_queue_init:
+	hclgevf_pci_uninit(hdev);
+	clear_bit(HCLGEVF_STATE_IRQ_INITED, &hdev->state);
+	return ret;
+}
+
+static void hclgevf_uninit_hdev(struct hclgevf_dev *hdev)
+{
+	struct hclge_vf_to_pf_msg send_msg;
+
+	hclgevf_state_uninit(hdev);
+	hclgevf_uninit_rxd_adv_layout(hdev);
+
+	hclgevf_build_send_msg(&send_msg, HCLGE_MBX_VF_UNINIT, 0);
+	hclgevf_send_mbx_msg(hdev, &send_msg, false, NULL, 0);
+
+	if (test_bit(HCLGEVF_STATE_IRQ_INITED, &hdev->state)) {
+		hclgevf_misc_irq_uninit(hdev);
+		hclgevf_uninit_msi(hdev);
+	}
+
+	hclge_comm_cmd_uninit(hdev->ae_dev, &hdev->hw.hw);
+	hclgevf_devlink_uninit(hdev);
+	hclgevf_pci_uninit(hdev);
+	hclgevf_uninit_mac_list(hdev);
+}
+
+static int hclgevf_init_ae_dev(struct hnae3_ae_dev *ae_dev)
+{
+	struct pci_dev *pdev = ae_dev->pdev;
+	int ret;
+
+	ret = hclgevf_alloc_hdev(ae_dev);
+	if (ret) {
+		dev_err(&pdev->dev, "hclge device allocation failed\n");
+		return ret;
+	}
+
+	ret = hclgevf_init_hdev(ae_dev->priv);
+	if (ret) {
+		dev_err(&pdev->dev, "hclge device initialization failed\n");
+		return ret;
+	}
+
+	return 0;
+}
+
+static void hclgevf_uninit_ae_dev(struct hnae3_ae_dev *ae_dev)
+{
+	struct hclgevf_dev *hdev = ae_dev->priv;
+
+	hclgevf_uninit_hdev(hdev);
+	ae_dev->priv = NULL;
+}
+
+static u32 hclgevf_get_max_channels(struct hclgevf_dev *hdev)
+{
+	return min(hdev->rss_size_max, hdev->num_tqps);
+}
+
+/**
+ * hclgevf_get_channels - Get the current channels enabled and max supported.
+ * @handle: hardware information for network interface
+ * @ch: ethtool channels structure
+ *
+ * We don't support separate tx and rx queues as channels. The other count
+ * represents how many queues are being used for control. max_combined counts
+ * how many queue pairs we can support. They may not be mapped 1 to 1 with
+ * q_vectors since we support a lot more queue pairs than q_vectors.
+ **/
+static void hclgevf_get_channels(struct hnae3_handle *handle,
+				 struct ethtool_channels *ch)
+{
+	struct hclgevf_dev *hdev = hclgevf_ae_get_hdev(handle);
+
+	ch->max_combined = hclgevf_get_max_channels(hdev);
+	ch->other_count = 0;
+	ch->max_other = 0;
+	ch->combined_count = handle->kinfo.rss_size;
+}
+
+static void hclgevf_get_tqps_and_rss_info(struct hnae3_handle *handle,
+					  u16 *alloc_tqps, u16 *max_rss_size)
+{
+	struct hclgevf_dev *hdev = hclgevf_ae_get_hdev(handle);
+
+	*alloc_tqps = hdev->num_tqps;
+	*max_rss_size = hdev->rss_size_max;
+}
+
+static void hclgevf_update_rss_size(struct hnae3_handle *handle,
+				    u32 new_tqps_num)
+{
+	struct hnae3_knic_private_info *kinfo = &handle->kinfo;
+	struct hclgevf_dev *hdev = hclgevf_ae_get_hdev(handle);
+	u16 max_rss_size;
+
+	kinfo->req_rss_size = new_tqps_num;
+
+	max_rss_size = min_t(u16, hdev->rss_size_max,
+			     hdev->num_tqps / kinfo->tc_info.num_tc);
+
+	/* Use the user's configuration when it is not larger than
+	 * max_rss_size, otherwise, use the maximum specification value.
+	 */
+	if (kinfo->req_rss_size != kinfo->rss_size && kinfo->req_rss_size &&
+	    kinfo->req_rss_size <= max_rss_size)
+		kinfo->rss_size = kinfo->req_rss_size;
+	else if (kinfo->rss_size > max_rss_size ||
+		 (!kinfo->req_rss_size && kinfo->rss_size < max_rss_size))
+		kinfo->rss_size = max_rss_size;
+
+	kinfo->num_tqps = kinfo->tc_info.num_tc * kinfo->rss_size;
+}
+
+static int hclgevf_set_channels(struct hnae3_handle *handle, u32 new_tqps_num,
+				bool rxfh_configured)
+{
+	struct hclgevf_dev *hdev = hclgevf_ae_get_hdev(handle);
+	struct hnae3_knic_private_info *kinfo = &handle->kinfo;
+	u16 tc_offset[HCLGE_COMM_MAX_TC_NUM];
+	u16 tc_valid[HCLGE_COMM_MAX_TC_NUM];
+	u16 tc_size[HCLGE_COMM_MAX_TC_NUM];
+	u16 cur_rss_size = kinfo->rss_size;
+	u16 cur_tqps = kinfo->num_tqps;
+	u32 *rss_indir;
+	unsigned int i;
+	int ret;
+
+	hclgevf_update_rss_size(handle, new_tqps_num);
+
+	hclge_comm_get_rss_tc_info(kinfo->rss_size, hdev->hw_tc_map,
+				   tc_offset, tc_valid, tc_size);
+	ret = hclge_comm_set_rss_tc_mode(&hdev->hw.hw, tc_offset,
+					 tc_valid, tc_size);
+	if (ret)
+		return ret;
+
+	/* RSS indirection table has been configured by user */
+	if (rxfh_configured)
+		goto out;
+
+	/* Reinitializes the rss indirect table according to the new RSS size */
+	rss_indir = kcalloc(hdev->ae_dev->dev_specs.rss_ind_tbl_size,
+			    sizeof(u32), GFP_KERNEL);
+	if (!rss_indir)
+		return -ENOMEM;
+
+	for (i = 0; i < hdev->ae_dev->dev_specs.rss_ind_tbl_size; i++)
+		rss_indir[i] = i % kinfo->rss_size;
+
+	hdev->rss_cfg.rss_size = kinfo->rss_size;
+
+	ret = hclgevf_set_rss(handle, rss_indir, NULL, 0);
+	if (ret)
+		dev_err(&hdev->pdev->dev, "set rss indir table fail, ret=%d\n",
+			ret);
+
+	kfree(rss_indir);
+
+out:
+	if (!ret)
+		dev_info(&hdev->pdev->dev,
+			 "Channels changed, rss_size from %u to %u, tqps from %u to %u",
+			 cur_rss_size, kinfo->rss_size,
+			 cur_tqps, kinfo->rss_size * kinfo->tc_info.num_tc);
+
+	return ret;
+}
+
+static int hclgevf_get_status(struct hnae3_handle *handle)
+{
+	struct hclgevf_dev *hdev = hclgevf_ae_get_hdev(handle);
+
+	return hdev->hw.mac.link;
+}
+
+static void hclgevf_get_ksettings_an_result(struct hnae3_handle *handle,
+					    u8 *auto_neg, u32 *speed,
+					    u8 *duplex, u32 *lane_num)
+{
+	struct hclgevf_dev *hdev = hclgevf_ae_get_hdev(handle);
+
+	if (speed)
+		*speed = hdev->hw.mac.speed;
+	if (duplex)
+		*duplex = hdev->hw.mac.duplex;
+	if (auto_neg)
+		*auto_neg = AUTONEG_DISABLE;
+}
+
+void hclgevf_update_speed_duplex(struct hclgevf_dev *hdev, u32 speed,
+				 u8 duplex)
+{
+	hdev->hw.mac.speed = speed;
+	hdev->hw.mac.duplex = duplex;
+}
+
+static int hclgevf_gro_en(struct hnae3_handle *handle, bool enable)
+{
+	struct hclgevf_dev *hdev = hclgevf_ae_get_hdev(handle);
+	bool gro_en_old = hdev->gro_en;
+	int ret;
+
+	hdev->gro_en = enable;
+	ret = hclgevf_config_gro(hdev);
+	if (ret)
+		hdev->gro_en = gro_en_old;
+
+	return ret;
+}
+
+static void hclgevf_get_media_type(struct hnae3_handle *handle, u8 *media_type,
+				   u8 *module_type)
+{
+	struct hclgevf_dev *hdev = hclgevf_ae_get_hdev(handle);
+
+	if (media_type)
+		*media_type = hdev->hw.mac.media_type;
+
+	if (module_type)
+		*module_type = hdev->hw.mac.module_type;
+}
+
+static bool hclgevf_get_hw_reset_stat(struct hnae3_handle *handle)
+{
+	struct hclgevf_dev *hdev = hclgevf_ae_get_hdev(handle);
+
+	return !!hclgevf_read_dev(&hdev->hw, HCLGEVF_RST_ING);
+}
+
+static bool hclgevf_get_cmdq_stat(struct hnae3_handle *handle)
+{
+	struct hclgevf_dev *hdev = hclgevf_ae_get_hdev(handle);
+
+	return test_bit(HCLGE_COMM_STATE_CMD_DISABLE, &hdev->hw.hw.comm_state);
+}
+
+static bool hclgevf_ae_dev_resetting(struct hnae3_handle *handle)
+{
+	struct hclgevf_dev *hdev = hclgevf_ae_get_hdev(handle);
+
+	return test_bit(HCLGEVF_STATE_RST_HANDLING, &hdev->state);
+}
+
+static unsigned long hclgevf_ae_dev_reset_cnt(struct hnae3_handle *handle)
+{
+	struct hclgevf_dev *hdev = hclgevf_ae_get_hdev(handle);
+
+	return hdev->rst_stats.hw_rst_done_cnt;
+}
+
+static void hclgevf_get_link_mode(struct hnae3_handle *handle,
+				  unsigned long *supported,
+				  unsigned long *advertising)
+{
+	struct hclgevf_dev *hdev = hclgevf_ae_get_hdev(handle);
+
+	*supported = hdev->hw.mac.supported;
+	*advertising = hdev->hw.mac.advertising;
+}
+
+void hclgevf_update_port_base_vlan_info(struct hclgevf_dev *hdev, u16 state,
+				struct hclge_mbx_port_base_vlan *port_base_vlan)
+{
+	struct hnae3_handle *nic = &hdev->nic;
+	struct hclge_vf_to_pf_msg send_msg;
+	int ret;
+
+	rtnl_lock();
+
+	if (test_bit(HCLGEVF_STATE_RST_HANDLING, &hdev->state) ||
+	    test_bit(HCLGEVF_STATE_RST_FAIL, &hdev->state)) {
+		dev_warn(&hdev->pdev->dev,
+			 "is resetting when updating port based vlan info\n");
+		rtnl_unlock();
+		return;
+	}
+
+	ret = hclgevf_notify_client(hdev, HNAE3_DOWN_CLIENT);
+	if (ret) {
+		rtnl_unlock();
+		return;
+	}
+
+	/* send msg to PF and wait update port based vlan info */
+	hclgevf_build_send_msg(&send_msg, HCLGE_MBX_SET_VLAN,
+			       HCLGE_MBX_PORT_BASE_VLAN_CFG);
+	memcpy(send_msg.data, port_base_vlan, sizeof(*port_base_vlan));
+	ret = hclgevf_send_mbx_msg(hdev, &send_msg, false, NULL, 0);
+	if (!ret) {
+		if (state == HNAE3_PORT_BASE_VLAN_DISABLE)
+			nic->port_base_vlan_state = state;
+		else
+			nic->port_base_vlan_state = HNAE3_PORT_BASE_VLAN_ENABLE;
+	}
+
+	hclgevf_notify_client(hdev, HNAE3_UP_CLIENT);
+	rtnl_unlock();
+}
+
+static const struct hnae3_ae_ops hclgevf_ops = {
+	.init_ae_dev = hclgevf_init_ae_dev,
+	.uninit_ae_dev = hclgevf_uninit_ae_dev,
+	.reset_prepare = hclgevf_reset_prepare_general,
+	.reset_done = hclgevf_reset_done,
+	.init_client_instance = hclgevf_init_client_instance,
+	.uninit_client_instance = hclgevf_uninit_client_instance,
+	.start = hclgevf_ae_start,
+	.stop = hclgevf_ae_stop,
+	.client_start = hclgevf_client_start,
+	.client_stop = hclgevf_client_stop,
+	.map_ring_to_vector = hclgevf_map_ring_to_vector,
+	.unmap_ring_from_vector = hclgevf_unmap_ring_from_vector,
+	.get_vector = hclgevf_get_vector,
+	.put_vector = hclgevf_put_vector,
+	.reset_queue = hclgevf_reset_tqp,
+	.get_mac_addr = hclgevf_get_mac_addr,
+	.set_mac_addr = hclgevf_set_mac_addr,
+	.add_uc_addr = hclgevf_add_uc_addr,
+	.rm_uc_addr = hclgevf_rm_uc_addr,
+	.add_mc_addr = hclgevf_add_mc_addr,
+	.rm_mc_addr = hclgevf_rm_mc_addr,
+	.get_stats = hclgevf_get_stats,
+	.update_stats = hclgevf_update_stats,
+	.get_strings = hclgevf_get_strings,
+	.get_sset_count = hclgevf_get_sset_count,
+	.get_rss_key_size = hclge_comm_get_rss_key_size,
+	.get_rss = hclgevf_get_rss,
+	.set_rss = hclgevf_set_rss,
+	.get_rss_tuple = hclgevf_get_rss_tuple,
+	.set_rss_tuple = hclgevf_set_rss_tuple,
+	.get_tc_size = hclgevf_get_tc_size,
+	.get_fw_version = hclgevf_get_fw_version,
+	.set_vlan_filter = hclgevf_set_vlan_filter,
+	.enable_vlan_filter = hclgevf_enable_vlan_filter,
+	.enable_hw_strip_rxvtag = hclgevf_en_hw_strip_rxvtag,
+	.reset_event = hclgevf_reset_event,
+	.set_default_reset_request = hclgevf_set_def_reset_request,
+	.set_channels = hclgevf_set_channels,
+	.get_channels = hclgevf_get_channels,
+	.get_tqps_and_rss_info = hclgevf_get_tqps_and_rss_info,
+	.get_regs_len = hclgevf_get_regs_len,
+	.get_regs = hclgevf_get_regs,
+	.get_status = hclgevf_get_status,
+	.get_ksettings_an_result = hclgevf_get_ksettings_an_result,
+	.get_media_type = hclgevf_get_media_type,
+	.get_hw_reset_stat = hclgevf_get_hw_reset_stat,
+	.ae_dev_resetting = hclgevf_ae_dev_resetting,
+	.ae_dev_reset_cnt = hclgevf_ae_dev_reset_cnt,
+	.set_gro_en = hclgevf_gro_en,
+	.set_mtu = hclgevf_set_mtu,
+	.get_global_queue_id = hclgevf_get_qid_global,
+	.set_timer_task = hclgevf_set_timer_task,
+	.get_link_mode = hclgevf_get_link_mode,
+	.set_promisc_mode = hclgevf_set_promisc_mode,
+	.request_update_promisc_mode = hclgevf_request_update_promisc_mode,
+	.get_cmdq_stat = hclgevf_get_cmdq_stat,
+};
+
+static struct hnae3_ae_algo ae_algovf = {
+	.ops = &hclgevf_ops,
+	.pdev_id_table = ae_algovf_pci_tbl,
+};
+
+static int __init hclgevf_init(void)
+{
+	pr_info("%s is initializing\n", HCLGEVF_NAME);
+
+	hclgevf_wq = alloc_workqueue("%s", WQ_UNBOUND, 0, HCLGEVF_NAME);
+	if (!hclgevf_wq) {
+		pr_err("%s: failed to create workqueue\n", HCLGEVF_NAME);
+		return -ENOMEM;
+	}
+
+	hnae3_register_ae_algo(&ae_algovf);
+
+	return 0;
+}
+
+static void __exit hclgevf_exit(void)
+{
+	hnae3_acquire_unload_lock();
+	hnae3_unregister_ae_algo(&ae_algovf);
+	destroy_workqueue(hclgevf_wq);
+	hnae3_release_unload_lock();
+}
+module_init(hclgevf_init);
+module_exit(hclgevf_exit);
+
+MODULE_LICENSE("GPL");
+MODULE_AUTHOR("Huawei Tech. Co., Ltd.");
+MODULE_DESCRIPTION("HCLGEVF Driver");
+MODULE_VERSION(HCLGEVF_MOD_VERSION);
diff --git a/drivers/net/ethernet/hisilicon/hns3/hns3vf/hclgevf_main.h b/drivers/net/ethernet/hisilicon/hns3/hns3vf/hclgevf_main.h
new file mode 100644
index 000000000000..0208425ab594
--- /dev/null
+++ b/drivers/net/ethernet/hisilicon/hns3/hns3vf/hclgevf_main.h
@@ -0,0 +1,300 @@
+/* SPDX-License-Identifier: GPL-2.0+ */
+/* Copyright (c) 2016-2017 Hisilicon Limited. */
+
+#ifndef __HCLGEVF_MAIN_H
+#define __HCLGEVF_MAIN_H
+#include <linux/fs.h>
+#include <linux/if_vlan.h>
+#include <linux/types.h>
+#include <net/devlink.h>
+#include "hclge_mbx.h"
+#include "hclgevf_cmd.h"
+#include "hnae3.h"
+#include "hclge_comm_rss.h"
+#include "hclge_comm_tqp_stats.h"
+
+#define HCLGEVF_MOD_VERSION "1.0"
+#define HCLGEVF_DRIVER_NAME "hclgevf"
+
+#define HCLGEVF_MAX_VLAN_ID	4095
+#define HCLGEVF_MISC_VECTOR_NUM		0
+
+#define HCLGEVF_INVALID_VPORT		0xffff
+#define HCLGEVF_GENERAL_TASK_INTERVAL	  5
+#define HCLGEVF_KEEP_ALIVE_TASK_INTERVAL  2
+
+/* This number in actual depends upon the total number of VFs
+ * created by physical function. But the maximum number of
+ * possible vector-per-VF is {VFn(1-32), VECTn(32 + 1)}.
+ */
+#define HCLGEVF_MAX_VF_VECTOR_NUM	(32 + 1)
+
+#define HCLGEVF_VECTOR_REG_BASE		0x20000
+#define HCLGEVF_MISC_VECTOR_REG_BASE	0x20400
+#define HCLGEVF_VECTOR_REG_OFFSET	0x4
+#define HCLGEVF_VECTOR_VF_OFFSET		0x100000
+
+/* bar registers for common func */
+#define HCLGEVF_GRO_EN_REG			0x28000
+#define HCLGEVF_RXD_ADV_LAYOUT_EN_REG		0x28008
+
+/* bar registers for rcb */
+#define HCLGEVF_RING_RX_ADDR_L_REG		0x80000
+#define HCLGEVF_RING_RX_ADDR_H_REG		0x80004
+#define HCLGEVF_RING_RX_BD_NUM_REG		0x80008
+#define HCLGEVF_RING_RX_BD_LENGTH_REG		0x8000C
+#define HCLGEVF_RING_RX_MERGE_EN_REG		0x80014
+#define HCLGEVF_RING_RX_TAIL_REG		0x80018
+#define HCLGEVF_RING_RX_HEAD_REG		0x8001C
+#define HCLGEVF_RING_RX_FBD_NUM_REG		0x80020
+#define HCLGEVF_RING_RX_OFFSET_REG		0x80024
+#define HCLGEVF_RING_RX_FBD_OFFSET_REG		0x80028
+#define HCLGEVF_RING_RX_STASH_REG		0x80030
+#define HCLGEVF_RING_RX_BD_ERR_REG		0x80034
+#define HCLGEVF_RING_TX_ADDR_L_REG		0x80040
+#define HCLGEVF_RING_TX_ADDR_H_REG		0x80044
+#define HCLGEVF_RING_TX_BD_NUM_REG		0x80048
+#define HCLGEVF_RING_TX_PRIORITY_REG		0x8004C
+#define HCLGEVF_RING_TX_TC_REG			0x80050
+#define HCLGEVF_RING_TX_MERGE_EN_REG		0x80054
+#define HCLGEVF_RING_TX_TAIL_REG		0x80058
+#define HCLGEVF_RING_TX_HEAD_REG		0x8005C
+#define HCLGEVF_RING_TX_FBD_NUM_REG		0x80060
+#define HCLGEVF_RING_TX_OFFSET_REG		0x80064
+#define HCLGEVF_RING_TX_EBD_NUM_REG		0x80068
+#define HCLGEVF_RING_TX_EBD_OFFSET_REG		0x80070
+#define HCLGEVF_RING_TX_BD_ERR_REG		0x80074
+#define HCLGEVF_RING_EN_REG			0x80090
+
+/* bar registers for tqp interrupt */
+#define HCLGEVF_TQP_INTR_CTRL_REG		0x20000
+#define HCLGEVF_TQP_INTR_GL0_REG		0x20100
+#define HCLGEVF_TQP_INTR_GL1_REG		0x20200
+#define HCLGEVF_TQP_INTR_GL2_REG		0x20300
+#define HCLGEVF_TQP_INTR_RL_REG			0x20900
+
+/* CMDQ register bits for RX event(=MBX event) */
+#define HCLGEVF_VECTOR0_RX_CMDQ_INT_B	1
+/* RST register bits for RESET event */
+#define HCLGEVF_VECTOR0_RST_INT_B	2
+
+#define HCLGEVF_TQP_RESET_TRY_TIMES	10
+/* Reset related Registers */
+#define HCLGEVF_RST_ING			0x20C00
+#define HCLGEVF_FUN_RST_ING_BIT		BIT(0)
+#define HCLGEVF_GLOBAL_RST_ING_BIT	BIT(5)
+#define HCLGEVF_CORE_RST_ING_BIT	BIT(6)
+#define HCLGEVF_IMP_RST_ING_BIT		BIT(7)
+#define HCLGEVF_RST_ING_BITS \
+	(HCLGEVF_FUN_RST_ING_BIT | HCLGEVF_GLOBAL_RST_ING_BIT | \
+	 HCLGEVF_CORE_RST_ING_BIT | HCLGEVF_IMP_RST_ING_BIT)
+
+#define HCLGEVF_VF_RST_ING		0x07008
+#define HCLGEVF_VF_RST_ING_BIT		BIT(16)
+
+#define HCLGEVF_WAIT_RESET_DONE		100
+
+#define HCLGEVF_RSS_IND_TBL_SIZE		512
+
+#define HCLGEVF_TQP_MEM_SIZE		0x10000
+#define HCLGEVF_MEM_BAR			4
+/* in the bar4, the first half is for roce, and the second half is for nic */
+#define HCLGEVF_NIC_MEM_OFFSET(hdev)	\
+	(pci_resource_len((hdev)->pdev, HCLGEVF_MEM_BAR) >> 1)
+#define HCLGEVF_TQP_MEM_OFFSET(hdev, i)		\
+	(HCLGEVF_NIC_MEM_OFFSET(hdev) + HCLGEVF_TQP_MEM_SIZE * (i))
+
+#define HCLGEVF_MAC_MAX_FRAME		9728
+
+#define HCLGEVF_STATS_TIMER_INTERVAL	36U
+
+#define hclgevf_read_dev(a, reg) \
+	hclge_comm_read_reg((a)->hw.io_base, reg)
+#define hclgevf_write_dev(a, reg, value) \
+	hclge_comm_write_reg((a)->hw.io_base, reg, value)
+
+enum hclgevf_evt_cause {
+	HCLGEVF_VECTOR0_EVENT_RST,
+	HCLGEVF_VECTOR0_EVENT_MBX,
+	HCLGEVF_VECTOR0_EVENT_OTHER,
+};
+
+/* states of hclgevf device & tasks */
+enum hclgevf_states {
+	/* device states */
+	HCLGEVF_STATE_DOWN,
+	HCLGEVF_STATE_DISABLED,
+	HCLGEVF_STATE_IRQ_INITED,
+	HCLGEVF_STATE_REMOVING,
+	HCLGEVF_STATE_NIC_REGISTERED,
+	HCLGEVF_STATE_ROCE_REGISTERED,
+	HCLGEVF_STATE_SERVICE_INITED,
+	/* task states */
+	HCLGEVF_STATE_RST_SERVICE_SCHED,
+	HCLGEVF_STATE_RST_HANDLING,
+	HCLGEVF_STATE_MBX_SERVICE_SCHED,
+	HCLGEVF_STATE_MBX_HANDLING,
+	HCLGEVF_STATE_LINK_UPDATING,
+	HCLGEVF_STATE_PROMISC_CHANGED,
+	HCLGEVF_STATE_RST_FAIL,
+	HCLGEVF_STATE_PF_PUSH_LINK_STATUS,
+};
+
+struct hclgevf_mac {
+	u8 media_type;
+	u8 module_type;
+	u8 mac_addr[ETH_ALEN];
+	int link;
+	u8 duplex;
+	u32 speed;
+	u64 supported;
+	u64 advertising;
+};
+
+struct hclgevf_hw {
+	struct hclge_comm_hw hw;
+	int num_vec;
+	struct hclgevf_mac mac;
+};
+
+struct hclgevf_cfg {
+	u8 tc_num;
+	u16 tqp_desc_num;
+	u16 rx_buf_len;
+	u8 phy_addr;
+	u8 media_type;
+	u8 mac_addr[ETH_ALEN];
+	u32 numa_node_map;
+};
+
+struct hclgevf_misc_vector {
+	u8 __iomem *addr;
+	int vector_irq;
+	char name[HNAE3_INT_NAME_LEN];
+};
+
+struct hclgevf_rst_stats {
+	u32 rst_cnt;			/* the number of reset */
+	u32 vf_func_rst_cnt;		/* the number of VF function reset */
+	u32 flr_rst_cnt;		/* the number of FLR */
+	u32 vf_rst_cnt;			/* the number of VF reset */
+	u32 rst_done_cnt;		/* the number of reset completed */
+	u32 hw_rst_done_cnt;		/* the number of HW reset completed */
+	u32 rst_fail_cnt;		/* the number of VF reset fail */
+};
+
+enum HCLGEVF_MAC_ADDR_TYPE {
+	HCLGEVF_MAC_ADDR_UC,
+	HCLGEVF_MAC_ADDR_MC
+};
+
+enum HCLGEVF_MAC_NODE_STATE {
+	HCLGEVF_MAC_TO_ADD,
+	HCLGEVF_MAC_TO_DEL,
+	HCLGEVF_MAC_ACTIVE
+};
+
+struct hclgevf_mac_addr_node {
+	struct list_head node;
+	enum HCLGEVF_MAC_NODE_STATE state;
+	u8 mac_addr[ETH_ALEN];
+};
+
+struct hclgevf_mac_table_cfg {
+	spinlock_t mac_list_lock; /* protect mac address need to add/detele */
+	struct list_head uc_mac_list;
+	struct list_head mc_mac_list;
+};
+
+struct hclgevf_dev {
+	struct pci_dev *pdev;
+	struct hnae3_ae_dev *ae_dev;
+	struct hclgevf_hw hw;
+	struct hclgevf_misc_vector misc_vector;
+	struct hclge_comm_rss_cfg rss_cfg;
+	unsigned long state;
+	unsigned long flr_state;
+	unsigned long default_reset_request;
+	unsigned long last_reset_time;
+	enum hnae3_reset_type reset_level;
+	unsigned long reset_pending;
+	enum hnae3_reset_type reset_type;
+	struct timer_list reset_timer;
+
+#define HCLGEVF_RESET_REQUESTED		0
+#define HCLGEVF_RESET_PENDING		1
+	unsigned long reset_state;	/* requested, pending */
+	struct hclgevf_rst_stats rst_stats;
+	u32 reset_attempts;
+	struct semaphore reset_sem;	/* protect reset process */
+
+	u32 fw_version;
+	u16 mbx_api_version;
+	u16 num_tqps;		/* num task queue pairs of this VF */
+
+	u16 alloc_rss_size;	/* allocated RSS task queue */
+	u16 rss_size_max;	/* HW defined max RSS task queue */
+
+	u16 num_alloc_vport;	/* num vports this driver supports */
+	nodemask_t numa_node_mask;
+	u16 rx_buf_len;
+	u16 num_tx_desc;	/* desc num of per tx queue */
+	u16 num_rx_desc;	/* desc num of per rx queue */
+	u8 hw_tc_map;
+	u8 has_pf_mac;
+
+	u16 num_msi;
+	u16 num_msi_left;
+	u16 num_msi_used;
+	u16 num_nic_msix;	/* Num of nic vectors for this VF */
+	u16 num_roce_msix;	/* Num of roce vectors for this VF */
+	u16 roce_base_msix_offset;
+	u16 *vector_status;
+	int *vector_irq;
+
+	bool gro_en;
+	bool rxvtag_strip_en;
+
+	unsigned long vlan_del_fail_bmap[BITS_TO_LONGS(VLAN_N_VID)];
+
+	struct hclgevf_mac_table_cfg mac_table;
+
+	struct hclgevf_mbx_resp_status mbx_resp; /* mailbox response */
+	struct hclgevf_mbx_arq_ring arq; /* mailbox async rx queue */
+
+	struct delayed_work service_task;
+
+	struct hclge_comm_tqp *htqp;
+
+	struct hnae3_handle nic;
+	struct hnae3_handle roce;
+
+	struct hnae3_client *nic_client;
+	struct hnae3_client *roce_client;
+	u32 flag;
+	unsigned long serv_processed_cnt;
+	unsigned long last_serv_processed;
+
+	struct devlink *devlink;
+};
+
+static inline bool hclgevf_is_reset_pending(struct hclgevf_dev *hdev)
+{
+	return !!hdev->reset_pending;
+}
+
+int hclgevf_send_mbx_msg(struct hclgevf_dev *hdev,
+			 struct hclge_vf_to_pf_msg *send_msg, bool need_resp,
+			 u8 *resp_data, u16 resp_len);
+void hclgevf_mbx_handler(struct hclgevf_dev *hdev);
+void hclgevf_mbx_async_handler(struct hclgevf_dev *hdev);
+
+void hclgevf_update_link_status(struct hclgevf_dev *hdev, int link_state);
+void hclgevf_update_speed_duplex(struct hclgevf_dev *hdev, u32 speed,
+				 u8 duplex);
+void hclgevf_reset_task_schedule(struct hclgevf_dev *hdev);
+void hclgevf_mbx_task_schedule(struct hclgevf_dev *hdev);
+void hclgevf_update_port_base_vlan_info(struct hclgevf_dev *hdev, u16 state,
+			struct hclge_mbx_port_base_vlan *port_base_vlan);
+struct hclgevf_dev *hclgevf_ae_get_hdev(struct hnae3_handle *handle);
+#endif
diff --git a/drivers/net/ethernet/hisilicon/hns3/hns3vf/hclgevf_mbx.c b/drivers/net/ethernet/hisilicon/hns3/hns3vf/hclgevf_mbx.c
new file mode 100644
index 000000000000..f5c99ca54369
--- /dev/null
+++ b/drivers/net/ethernet/hisilicon/hns3/hns3vf/hclgevf_mbx.c
@@ -0,0 +1,388 @@
+// SPDX-License-Identifier: GPL-2.0+
+// Copyright (c) 2016-2017 Hisilicon Limited.
+
+#include "hclge_mbx.h"
+#include "hclgevf_main.h"
+#include "hnae3.h"
+
+#define CREATE_TRACE_POINTS
+#include "hclgevf_trace.h"
+
+static int hclgevf_resp_to_errno(u16 resp_code)
+{
+	return resp_code ? -resp_code : 0;
+}
+
+#define HCLGEVF_MBX_MATCH_ID_START	1
+static void hclgevf_reset_mbx_resp_status(struct hclgevf_dev *hdev)
+{
+	/* this function should be called with mbx_resp.mbx_mutex held
+	 * to protect the received_response from race condition
+	 */
+	hdev->mbx_resp.received_resp  = false;
+	hdev->mbx_resp.origin_mbx_msg = 0;
+	hdev->mbx_resp.resp_status    = 0;
+	hdev->mbx_resp.match_id++;
+	/* Update match_id and ensure the value of match_id is not zero */
+	if (hdev->mbx_resp.match_id == 0)
+		hdev->mbx_resp.match_id = HCLGEVF_MBX_MATCH_ID_START;
+	memset(hdev->mbx_resp.additional_info, 0, HCLGE_MBX_MAX_RESP_DATA_SIZE);
+}
+
+/* hclgevf_get_mbx_resp: used to get a response from PF after VF sends a mailbox
+ * message to PF.
+ * @hdev: pointer to struct hclgevf_dev
+ * @code0: the message opcode VF send to PF.
+ * @code1: the message sub-opcode VF send to PF.
+ * @resp_data: pointer to store response data from PF to VF.
+ * @resp_len: the length of resp_data from PF to VF.
+ */
+static int hclgevf_get_mbx_resp(struct hclgevf_dev *hdev, u16 code0, u16 code1,
+				u8 *resp_data, u16 resp_len)
+{
+#define HCLGEVF_MAX_TRY_TIMES	500
+#define HCLGEVF_SLEEP_USECOND	1000
+	struct hclgevf_mbx_resp_status *mbx_resp;
+	u16 r_code0, r_code1;
+	int i = 0;
+
+	if (resp_len > HCLGE_MBX_MAX_RESP_DATA_SIZE) {
+		dev_err(&hdev->pdev->dev,
+			"VF mbx response len(=%u) exceeds maximum(=%u)\n",
+			resp_len,
+			HCLGE_MBX_MAX_RESP_DATA_SIZE);
+		return -EINVAL;
+	}
+
+	while ((!hdev->mbx_resp.received_resp) && (i < HCLGEVF_MAX_TRY_TIMES)) {
+		if (test_bit(HCLGE_COMM_STATE_CMD_DISABLE,
+			     &hdev->hw.hw.comm_state))
+			return -EIO;
+
+		usleep_range(HCLGEVF_SLEEP_USECOND, HCLGEVF_SLEEP_USECOND * 2);
+		i++;
+	}
+
+	/* ensure additional_info will be seen after received_resp */
+	smp_rmb();
+
+	if (i >= HCLGEVF_MAX_TRY_TIMES) {
+		dev_err(&hdev->pdev->dev,
+			"VF could not get mbx(%u,%u) resp(=%d) from PF in %d tries\n",
+			code0, code1, hdev->mbx_resp.received_resp, i);
+		return -EIO;
+	}
+
+	mbx_resp = &hdev->mbx_resp;
+	r_code0 = (u16)(mbx_resp->origin_mbx_msg >> 16);
+	r_code1 = (u16)(mbx_resp->origin_mbx_msg & 0xff);
+
+	if (mbx_resp->resp_status)
+		return mbx_resp->resp_status;
+
+	if (resp_data)
+		memcpy(resp_data, &mbx_resp->additional_info[0], resp_len);
+
+	hclgevf_reset_mbx_resp_status(hdev);
+
+	if (!(r_code0 == code0 && r_code1 == code1 && !mbx_resp->resp_status)) {
+		dev_err(&hdev->pdev->dev,
+			"VF could not match resp code(code0=%u,code1=%u), %d\n",
+			code0, code1, mbx_resp->resp_status);
+		dev_err(&hdev->pdev->dev,
+			"VF could not match resp r_code(r_code0=%u,r_code1=%u)\n",
+			r_code0, r_code1);
+		return -EIO;
+	}
+
+	return 0;
+}
+
+int hclgevf_send_mbx_msg(struct hclgevf_dev *hdev,
+			 struct hclge_vf_to_pf_msg *send_msg, bool need_resp,
+			 u8 *resp_data, u16 resp_len)
+{
+	struct hclge_mbx_vf_to_pf_cmd *req;
+	struct hclge_desc desc;
+	int status;
+
+	req = (struct hclge_mbx_vf_to_pf_cmd *)desc.data;
+
+	if (!send_msg) {
+		dev_err(&hdev->pdev->dev,
+			"failed to send mbx, msg is NULL\n");
+		return -EINVAL;
+	}
+
+	hclgevf_cmd_setup_basic_desc(&desc, HCLGEVF_OPC_MBX_VF_TO_PF, false);
+	if (need_resp)
+		hnae3_set_bit(req->mbx_need_resp, HCLGE_MBX_NEED_RESP_B, 1);
+
+	memcpy(&req->msg, send_msg, sizeof(struct hclge_vf_to_pf_msg));
+
+	if (test_bit(HCLGEVF_STATE_NIC_REGISTERED, &hdev->state))
+		trace_hclge_vf_mbx_send(hdev, req);
+
+	/* synchronous send */
+	if (need_resp) {
+		mutex_lock(&hdev->mbx_resp.mbx_mutex);
+		hclgevf_reset_mbx_resp_status(hdev);
+		req->match_id = cpu_to_le16(hdev->mbx_resp.match_id);
+		status = hclgevf_cmd_send(&hdev->hw, &desc, 1);
+		if (status) {
+			dev_err(&hdev->pdev->dev,
+				"VF failed(=%d) to send mbx message to PF\n",
+				status);
+			mutex_unlock(&hdev->mbx_resp.mbx_mutex);
+			return status;
+		}
+
+		status = hclgevf_get_mbx_resp(hdev, send_msg->code,
+					      send_msg->subcode, resp_data,
+					      resp_len);
+		mutex_unlock(&hdev->mbx_resp.mbx_mutex);
+	} else {
+		/* asynchronous send */
+		status = hclgevf_cmd_send(&hdev->hw, &desc, 1);
+		if (status) {
+			dev_err(&hdev->pdev->dev,
+				"VF failed(=%d) to send mbx message to PF\n",
+				status);
+			return status;
+		}
+	}
+
+	return status;
+}
+
+static bool hclgevf_cmd_crq_empty(struct hclgevf_hw *hw)
+{
+	u32 tail = hclgevf_read_dev(hw, HCLGE_COMM_NIC_CRQ_TAIL_REG);
+
+	return tail == (u32)hw->hw.cmq.crq.next_to_use;
+}
+
+static void hclgevf_handle_mbx_response(struct hclgevf_dev *hdev,
+					struct hclge_mbx_pf_to_vf_cmd *req)
+{
+	u16 vf_mbx_msg_subcode = le16_to_cpu(req->msg.vf_mbx_msg_subcode);
+	u16 vf_mbx_msg_code = le16_to_cpu(req->msg.vf_mbx_msg_code);
+	struct hclgevf_mbx_resp_status *resp = &hdev->mbx_resp;
+	u16 resp_status = le16_to_cpu(req->msg.resp_status);
+	u16 match_id = le16_to_cpu(req->match_id);
+
+	if (resp->received_resp)
+		dev_warn(&hdev->pdev->dev,
+			"VF mbx resp flag not clear(%u)\n",
+			 vf_mbx_msg_code);
+
+	resp->origin_mbx_msg = (vf_mbx_msg_code << 16);
+	resp->origin_mbx_msg |= vf_mbx_msg_subcode;
+	resp->resp_status = hclgevf_resp_to_errno(resp_status);
+	memcpy(resp->additional_info, req->msg.resp_data,
+	       HCLGE_MBX_MAX_RESP_DATA_SIZE * sizeof(u8));
+
+	/* ensure additional_info will be seen before setting received_resp */
+	smp_wmb();
+
+	if (match_id) {
+		/* If match_id is not zero, it means PF support match_id.
+		 * if the match_id is right, VF get the right response, or
+		 * ignore the response. and driver will clear hdev->mbx_resp
+		 * when send next message which need response.
+		 */
+		if (match_id == resp->match_id)
+			resp->received_resp = true;
+	} else {
+		resp->received_resp = true;
+	}
+}
+
+static void hclgevf_handle_mbx_msg(struct hclgevf_dev *hdev,
+				   struct hclge_mbx_pf_to_vf_cmd *req)
+{
+	/* we will drop the async msg if we find ARQ as full
+	 * and continue with next message
+	 */
+	if (atomic_read(&hdev->arq.count) >=
+	    HCLGE_MBX_MAX_ARQ_MSG_NUM) {
+		dev_warn(&hdev->pdev->dev,
+			 "Async Q full, dropping msg(%u)\n",
+			 le16_to_cpu(req->msg.code));
+		return;
+	}
+
+	/* tail the async message in arq */
+	memcpy(hdev->arq.msg_q[hdev->arq.tail], &req->msg,
+	       HCLGE_MBX_MAX_ARQ_MSG_SIZE * sizeof(u16));
+	hclge_mbx_tail_ptr_move_arq(hdev->arq);
+	atomic_inc(&hdev->arq.count);
+
+	hclgevf_mbx_task_schedule(hdev);
+}
+
+void hclgevf_mbx_handler(struct hclgevf_dev *hdev)
+{
+	struct hclge_mbx_pf_to_vf_cmd *req;
+	struct hclge_comm_cmq_ring *crq;
+	struct hclge_desc *desc;
+	u16 flag;
+	u16 code;
+
+	crq = &hdev->hw.hw.cmq.crq;
+
+	while (!hclgevf_cmd_crq_empty(&hdev->hw)) {
+		if (test_bit(HCLGE_COMM_STATE_CMD_DISABLE,
+			     &hdev->hw.hw.comm_state)) {
+			dev_info(&hdev->pdev->dev, "vf crq need init\n");
+			return;
+		}
+
+		desc = &crq->desc[crq->next_to_use];
+		req = (struct hclge_mbx_pf_to_vf_cmd *)desc->data;
+
+		flag = le16_to_cpu(crq->desc[crq->next_to_use].flag);
+		code = le16_to_cpu(req->msg.code);
+		if (unlikely(!hnae3_get_bit(flag, HCLGEVF_CMDQ_RX_OUTVLD_B))) {
+			dev_warn(&hdev->pdev->dev,
+				 "dropped invalid mailbox message, code = %u\n",
+				 code);
+
+			/* dropping/not processing this invalid message */
+			crq->desc[crq->next_to_use].flag = 0;
+			hclge_mbx_ring_ptr_move_crq(crq);
+			continue;
+		}
+
+		trace_hclge_vf_mbx_get(hdev, req);
+
+		/* synchronous messages are time critical and need preferential
+		 * treatment. Therefore, we need to acknowledge all the sync
+		 * responses as quickly as possible so that waiting tasks do not
+		 * timeout and simultaneously queue the async messages for later
+		 * prcessing in context of mailbox task i.e. the slow path.
+		 */
+		switch (code) {
+		case HCLGE_MBX_PF_VF_RESP:
+			hclgevf_handle_mbx_response(hdev, req);
+			break;
+		case HCLGE_MBX_LINK_STAT_CHANGE:
+		case HCLGE_MBX_ASSERTING_RESET:
+		case HCLGE_MBX_LINK_STAT_MODE:
+		case HCLGE_MBX_PUSH_VLAN_INFO:
+		case HCLGE_MBX_PUSH_PROMISC_INFO:
+			hclgevf_handle_mbx_msg(hdev, req);
+			break;
+		default:
+			dev_err(&hdev->pdev->dev,
+				"VF received unsupported(%u) mbx msg from PF\n",
+				code);
+			break;
+		}
+		crq->desc[crq->next_to_use].flag = 0;
+		hclge_mbx_ring_ptr_move_crq(crq);
+	}
+
+	/* Write back CMDQ_RQ header pointer, M7 need this pointer */
+	hclgevf_write_dev(&hdev->hw, HCLGE_COMM_NIC_CRQ_HEAD_REG,
+			  crq->next_to_use);
+}
+
+static void hclgevf_parse_promisc_info(struct hclgevf_dev *hdev,
+				       u16 promisc_info)
+{
+	if (!promisc_info)
+		dev_info(&hdev->pdev->dev,
+			 "Promisc mode is closed by host for being untrusted.\n");
+}
+
+void hclgevf_mbx_async_handler(struct hclgevf_dev *hdev)
+{
+	struct hclge_mbx_port_base_vlan *vlan_info;
+	struct hclge_mbx_link_status *link_info;
+	struct hclge_mbx_link_mode *link_mode;
+	enum hnae3_reset_type reset_type;
+	u16 link_status, state;
+	__le16 *msg_q;
+	u16 opcode;
+	u8 duplex;
+	u32 speed;
+	u32 tail;
+	u8 flag;
+	u16 idx;
+
+	tail = hdev->arq.tail;
+
+	/* process all the async queue messages */
+	while (tail != hdev->arq.head) {
+		if (test_bit(HCLGE_COMM_STATE_CMD_DISABLE,
+			     &hdev->hw.hw.comm_state)) {
+			dev_info(&hdev->pdev->dev,
+				 "vf crq need init in async\n");
+			return;
+		}
+
+		msg_q = hdev->arq.msg_q[hdev->arq.head];
+		opcode = le16_to_cpu(msg_q[0]);
+		switch (opcode) {
+		case HCLGE_MBX_LINK_STAT_CHANGE:
+			link_info = (struct hclge_mbx_link_status *)(msg_q + 1);
+			link_status = le16_to_cpu(link_info->link_status);
+			speed = le32_to_cpu(link_info->speed);
+			duplex = (u8)le16_to_cpu(link_info->duplex);
+			flag = link_info->flag;
+
+			/* update upper layer with new link link status */
+			hclgevf_update_speed_duplex(hdev, speed, duplex);
+			hclgevf_update_link_status(hdev, link_status);
+
+			if (flag & HCLGE_MBX_PUSH_LINK_STATUS_EN)
+				set_bit(HCLGEVF_STATE_PF_PUSH_LINK_STATUS,
+					&hdev->state);
+
+			break;
+		case HCLGE_MBX_LINK_STAT_MODE:
+			link_mode = (struct hclge_mbx_link_mode *)(msg_q + 1);
+			idx = le16_to_cpu(link_mode->idx);
+			if (idx)
+				hdev->hw.mac.supported =
+					le64_to_cpu(link_mode->link_mode);
+			else
+				hdev->hw.mac.advertising =
+					le64_to_cpu(link_mode->link_mode);
+			break;
+		case HCLGE_MBX_ASSERTING_RESET:
+			/* PF has asserted reset hence VF should go in pending
+			 * state and poll for the hardware reset status till it
+			 * has been completely reset. After this stack should
+			 * eventually be re-initialized.
+			 */
+			reset_type =
+				(enum hnae3_reset_type)le16_to_cpu(msg_q[1]);
+			set_bit(reset_type, &hdev->reset_pending);
+			set_bit(HCLGEVF_RESET_PENDING, &hdev->reset_state);
+			hclgevf_reset_task_schedule(hdev);
+
+			break;
+		case HCLGE_MBX_PUSH_VLAN_INFO:
+			vlan_info =
+				(struct hclge_mbx_port_base_vlan *)(msg_q + 1);
+			state = le16_to_cpu(vlan_info->state);
+			hclgevf_update_port_base_vlan_info(hdev, state,
+							   vlan_info);
+			break;
+		case HCLGE_MBX_PUSH_PROMISC_INFO:
+			hclgevf_parse_promisc_info(hdev, le16_to_cpu(msg_q[1]));
+			break;
+		default:
+			dev_err(&hdev->pdev->dev,
+				"fetched unsupported(%u) message from arq\n",
+				opcode);
+			break;
+		}
+
+		hclge_mbx_head_ptr_move_arq(hdev->arq);
+		atomic_dec(&hdev->arq.count);
+		msg_q = hdev->arq.msg_q[hdev->arq.head];
+	}
+}
diff --git a/drivers/net/ethernet/hisilicon/hns3/hns3vf/hclgevf_regs.c b/drivers/net/ethernet/hisilicon/hns3/hns3vf/hclgevf_regs.c
new file mode 100644
index 000000000000..9de01e344e27
--- /dev/null
+++ b/drivers/net/ethernet/hisilicon/hns3/hns3vf/hclgevf_regs.c
@@ -0,0 +1,166 @@
+// SPDX-License-Identifier: GPL-2.0+
+// Copyright (c) 2023 Hisilicon Limited.
+
+#include "hclgevf_main.h"
+#include "hclgevf_regs.h"
+#include "hnae3.h"
+
+static const u32 cmdq_reg_addr_list[] = {HCLGE_COMM_NIC_CSQ_BASEADDR_L_REG,
+					 HCLGE_COMM_NIC_CSQ_BASEADDR_H_REG,
+					 HCLGE_COMM_NIC_CSQ_DEPTH_REG,
+					 HCLGE_COMM_NIC_CSQ_TAIL_REG,
+					 HCLGE_COMM_NIC_CSQ_HEAD_REG,
+					 HCLGE_COMM_NIC_CRQ_BASEADDR_L_REG,
+					 HCLGE_COMM_NIC_CRQ_BASEADDR_H_REG,
+					 HCLGE_COMM_NIC_CRQ_DEPTH_REG,
+					 HCLGE_COMM_NIC_CRQ_TAIL_REG,
+					 HCLGE_COMM_NIC_CRQ_HEAD_REG,
+					 HCLGE_COMM_VECTOR0_CMDQ_SRC_REG,
+					 HCLGE_COMM_VECTOR0_CMDQ_STATE_REG,
+					 HCLGE_COMM_CMDQ_INTR_EN_REG,
+					 HCLGE_COMM_CMDQ_INTR_GEN_REG};
+
+static const u32 common_reg_addr_list[] = {HCLGEVF_MISC_VECTOR_REG_BASE,
+					   HCLGEVF_RST_ING,
+					   HCLGEVF_GRO_EN_REG};
+
+static const u32 ring_reg_addr_list[] = {HCLGEVF_RING_RX_ADDR_L_REG,
+					 HCLGEVF_RING_RX_ADDR_H_REG,
+					 HCLGEVF_RING_RX_BD_NUM_REG,
+					 HCLGEVF_RING_RX_BD_LENGTH_REG,
+					 HCLGEVF_RING_RX_MERGE_EN_REG,
+					 HCLGEVF_RING_RX_TAIL_REG,
+					 HCLGEVF_RING_RX_HEAD_REG,
+					 HCLGEVF_RING_RX_FBD_NUM_REG,
+					 HCLGEVF_RING_RX_OFFSET_REG,
+					 HCLGEVF_RING_RX_FBD_OFFSET_REG,
+					 HCLGEVF_RING_RX_STASH_REG,
+					 HCLGEVF_RING_RX_BD_ERR_REG,
+					 HCLGEVF_RING_TX_ADDR_L_REG,
+					 HCLGEVF_RING_TX_ADDR_H_REG,
+					 HCLGEVF_RING_TX_BD_NUM_REG,
+					 HCLGEVF_RING_TX_PRIORITY_REG,
+					 HCLGEVF_RING_TX_TC_REG,
+					 HCLGEVF_RING_TX_MERGE_EN_REG,
+					 HCLGEVF_RING_TX_TAIL_REG,
+					 HCLGEVF_RING_TX_HEAD_REG,
+					 HCLGEVF_RING_TX_FBD_NUM_REG,
+					 HCLGEVF_RING_TX_OFFSET_REG,
+					 HCLGEVF_RING_TX_EBD_NUM_REG,
+					 HCLGEVF_RING_TX_EBD_OFFSET_REG,
+					 HCLGEVF_RING_TX_BD_ERR_REG,
+					 HCLGEVF_RING_EN_REG};
+
+static const u32 tqp_intr_reg_addr_list[] = {HCLGEVF_TQP_INTR_CTRL_REG,
+					     HCLGEVF_TQP_INTR_GL0_REG,
+					     HCLGEVF_TQP_INTR_GL1_REG,
+					     HCLGEVF_TQP_INTR_GL2_REG,
+					     HCLGEVF_TQP_INTR_RL_REG};
+
+enum hclgevf_reg_tag {
+	HCLGEVF_REG_TAG_CMDQ = 0,
+	HCLGEVF_REG_TAG_COMMON,
+	HCLGEVF_REG_TAG_RING,
+	HCLGEVF_REG_TAG_TQP_INTR,
+};
+
+#pragma pack(4)
+struct hclgevf_reg_tlv {
+	u16 tag;
+	u16 len;
+};
+
+struct hclgevf_reg_header {
+	u64 magic_number;
+	u8 is_vf;
+	u8 rsv[7];
+};
+
+#pragma pack()
+
+#define HCLGEVF_REG_TLV_SIZE		sizeof(struct hclgevf_reg_tlv)
+#define HCLGEVF_REG_HEADER_SIZE		sizeof(struct hclgevf_reg_header)
+#define HCLGEVF_REG_TLV_SPACE		(sizeof(struct hclgevf_reg_tlv) / sizeof(u32))
+#define HCLGEVF_REG_HEADER_SPACE	(sizeof(struct hclgevf_reg_header) / sizeof(u32))
+#define HCLGEVF_REG_MAGIC_NUMBER	0x686e733372656773 /* meaning is hns3regs */
+
+static u32 hclgevf_reg_get_header(void *data)
+{
+	struct hclgevf_reg_header *header = data;
+
+	header->magic_number = HCLGEVF_REG_MAGIC_NUMBER;
+	header->is_vf = 0x1;
+
+	return HCLGEVF_REG_HEADER_SPACE;
+}
+
+static u32 hclgevf_reg_get_tlv(u32 tag, u32 regs_num, void *data)
+{
+	struct hclgevf_reg_tlv *tlv = data;
+
+	tlv->tag = tag;
+	tlv->len = regs_num * sizeof(u32) + HCLGEVF_REG_TLV_SIZE;
+
+	return HCLGEVF_REG_TLV_SPACE;
+}
+
+int hclgevf_get_regs_len(struct hnae3_handle *handle)
+{
+	struct hclgevf_dev *hdev = hclgevf_ae_get_hdev(handle);
+	int cmdq_len, common_len, ring_len, tqp_intr_len;
+
+	cmdq_len = HCLGEVF_REG_TLV_SIZE + sizeof(cmdq_reg_addr_list);
+	common_len = HCLGEVF_REG_TLV_SIZE + sizeof(common_reg_addr_list);
+	ring_len = HCLGEVF_REG_TLV_SIZE + sizeof(ring_reg_addr_list);
+	tqp_intr_len = HCLGEVF_REG_TLV_SIZE + sizeof(tqp_intr_reg_addr_list);
+
+	/* return the total length of all register values */
+	return HCLGEVF_REG_HEADER_SIZE + cmdq_len + common_len +
+	       tqp_intr_len * (hdev->num_msi_used - 1) +
+	       ring_len * hdev->num_tqps;
+}
+
+void hclgevf_get_regs(struct hnae3_handle *handle, u32 *version,
+		      void *data)
+{
+#define HCLGEVF_RING_INT_REG_OFFSET	0x4
+
+	struct hclgevf_dev *hdev = hclgevf_ae_get_hdev(handle);
+	struct hnae3_queue *tqp;
+	int i, j, reg_num;
+	u32 *reg = data;
+
+	*version = hdev->fw_version;
+	reg += hclgevf_reg_get_header(reg);
+
+	/* fetching per-VF registers values from VF PCIe register space */
+	reg_num = ARRAY_SIZE(cmdq_reg_addr_list);
+	reg += hclgevf_reg_get_tlv(HCLGEVF_REG_TAG_CMDQ, reg_num, reg);
+	for (i = 0; i < reg_num; i++)
+		*reg++ = hclgevf_read_dev(&hdev->hw, cmdq_reg_addr_list[i]);
+
+	reg_num = ARRAY_SIZE(common_reg_addr_list);
+	reg += hclgevf_reg_get_tlv(HCLGEVF_REG_TAG_COMMON, reg_num, reg);
+	for (i = 0; i < reg_num; i++)
+		*reg++ = hclgevf_read_dev(&hdev->hw, common_reg_addr_list[i]);
+
+	reg_num = ARRAY_SIZE(ring_reg_addr_list);
+	for (j = 0; j < hdev->num_tqps; j++) {
+		reg += hclgevf_reg_get_tlv(HCLGEVF_REG_TAG_RING, reg_num, reg);
+		tqp = &hdev->htqp[j].q;
+		for (i = 0; i < reg_num; i++)
+			*reg++ = readl_relaxed(tqp->io_base -
+					       HCLGEVF_TQP_REG_OFFSET +
+					       ring_reg_addr_list[i]);
+	}
+
+	reg_num = ARRAY_SIZE(tqp_intr_reg_addr_list);
+	for (j = 0; j < hdev->num_msi_used - 1; j++) {
+		reg += hclgevf_reg_get_tlv(HCLGEVF_REG_TAG_TQP_INTR,
+					   reg_num, reg);
+		for (i = 0; i < reg_num; i++)
+			*reg++ = hclgevf_read_dev(&hdev->hw,
+						  tqp_intr_reg_addr_list[i] +
+						  HCLGEVF_RING_INT_REG_OFFSET * j);
+	}
+}
diff --git a/drivers/net/ethernet/hisilicon/hns3/hns3vf/hclgevf_regs.h b/drivers/net/ethernet/hisilicon/hns3/hns3vf/hclgevf_regs.h
new file mode 100644
index 000000000000..77bdcf60a1af
--- /dev/null
+++ b/drivers/net/ethernet/hisilicon/hns3/hns3vf/hclgevf_regs.h
@@ -0,0 +1,13 @@
+/* SPDX-License-Identifier: GPL-2.0+ */
+/* Copyright (c) 2023 Hisilicon Limited. */
+
+#ifndef __HCLGEVF_REGS_H
+#define __HCLGEVF_REGS_H
+#include <linux/types.h>
+
+struct hnae3_handle;
+
+int hclgevf_get_regs_len(struct hnae3_handle *handle);
+void hclgevf_get_regs(struct hnae3_handle *handle, u32 *version,
+		      void *data);
+#endif
diff --git a/drivers/net/ethernet/hisilicon/hns3/hns3vf/hclgevf_trace.h b/drivers/net/ethernet/hisilicon/hns3/hns3vf/hclgevf_trace.h
new file mode 100644
index 000000000000..66b084309c91
--- /dev/null
+++ b/drivers/net/ethernet/hisilicon/hns3/hns3vf/hclgevf_trace.h
@@ -0,0 +1,137 @@
+/* SPDX-License-Identifier: GPL-2.0+ */
+/* Copyright (c) 2018-2019 Hisilicon Limited. */
+
+/* This must be outside ifdef _HCLGEVF_TRACE_H */
+#undef TRACE_SYSTEM
+#define TRACE_SYSTEM hns3
+
+#if !defined(_HCLGEVF_TRACE_H_) || defined(TRACE_HEADER_MULTI_READ)
+#define _HCLGEVF_TRACE_H_
+
+#include <linux/tracepoint.h>
+
+#define VF_GET_MBX_LEN	(sizeof(struct hclge_mbx_pf_to_vf_cmd) / sizeof(u32))
+#define VF_SEND_MBX_LEN	(sizeof(struct hclge_mbx_vf_to_pf_cmd) / sizeof(u32))
+
+TRACE_EVENT(hclge_vf_mbx_get,
+	TP_PROTO(
+		struct hclgevf_dev *hdev,
+		struct hclge_mbx_pf_to_vf_cmd *req),
+	TP_ARGS(hdev, req),
+
+	TP_STRUCT__entry(
+		__field(u8, vfid)
+		__field(u16, code)
+		__string(pciname, pci_name(hdev->pdev))
+		__string(devname, hdev->nic.kinfo.netdev->name)
+		__array(u32, mbx_data, VF_GET_MBX_LEN)
+	),
+
+	TP_fast_assign(
+		__entry->vfid = req->dest_vfid;
+		__entry->code = le16_to_cpu(req->msg.code);
+		__assign_str(pciname);
+		__assign_str(devname);
+		memcpy(__entry->mbx_data, req,
+		       sizeof(struct hclge_mbx_pf_to_vf_cmd));
+	),
+
+	TP_printk(
+		"%s %s vfid:%u code:%u data:%s",
+		__get_str(pciname), __get_str(devname), __entry->vfid,
+		__entry->code,
+		__print_array(__entry->mbx_data, VF_GET_MBX_LEN, sizeof(u32))
+	)
+);
+
+TRACE_EVENT(hclge_vf_mbx_send,
+	TP_PROTO(
+		struct hclgevf_dev *hdev,
+		struct hclge_mbx_vf_to_pf_cmd *req),
+	TP_ARGS(hdev, req),
+
+	TP_STRUCT__entry(
+		__field(u8, vfid)
+		__field(u8, code)
+		__field(u8, subcode)
+		__string(pciname, pci_name(hdev->pdev))
+		__string(devname, hdev->nic.kinfo.netdev->name)
+		__array(u32, mbx_data, VF_SEND_MBX_LEN)
+	),
+
+	TP_fast_assign(
+		__entry->vfid = req->mbx_src_vfid;
+		__entry->code = req->msg.code;
+		__entry->subcode = req->msg.subcode;
+		__assign_str(pciname);
+		__assign_str(devname);
+		memcpy(__entry->mbx_data, req,
+		       sizeof(struct hclge_mbx_vf_to_pf_cmd));
+	),
+
+	TP_printk(
+		"%s %s vfid:%u code:%u subcode:%u data:%s",
+		__get_str(pciname), __get_str(devname), __entry->vfid,
+		__entry->code, __entry->subcode,
+		__print_array(__entry->mbx_data, VF_SEND_MBX_LEN, sizeof(u32))
+	)
+);
+
+DECLARE_EVENT_CLASS(hclge_vf_cmd_template,
+		    TP_PROTO(struct hclge_comm_hw *hw,
+			     struct hclge_desc *desc,
+			     int index,
+			     int num),
+
+		    TP_ARGS(hw, desc, index, num),
+
+		    TP_STRUCT__entry(__field(u16, opcode)
+			__field(u16, flag)
+			__field(u16, retval)
+			__field(u16, rsv)
+			__field(int, index)
+			__field(int, num)
+			__string(pciname, pci_name(hw->cmq.csq.pdev))
+			__array(u32, data, HCLGE_DESC_DATA_LEN)),
+
+		    TP_fast_assign(int i;
+			__entry->opcode = le16_to_cpu(desc->opcode);
+			__entry->flag = le16_to_cpu(desc->flag);
+			__entry->retval = le16_to_cpu(desc->retval);
+			__entry->rsv = le16_to_cpu(desc->rsv);
+			__entry->index = index;
+			__entry->num = num;
+			__assign_str(pciname);
+			for (i = 0; i < HCLGE_DESC_DATA_LEN; i++)
+				__entry->data[i] = le32_to_cpu(desc->data[i]);),
+
+		    TP_printk("%s opcode:0x%04x %d-%d flag:0x%04x retval:0x%04x rsv:0x%04x data:%s",
+			      __get_str(pciname), __entry->opcode,
+			      __entry->index, __entry->num,
+			      __entry->flag, __entry->retval, __entry->rsv,
+			      __print_array(__entry->data,
+					    HCLGE_DESC_DATA_LEN, sizeof(u32)))
+);
+
+DEFINE_EVENT(hclge_vf_cmd_template, hclge_vf_cmd_send,
+	     TP_PROTO(struct hclge_comm_hw *hw,
+		      struct hclge_desc *desc,
+		      int index,
+		      int num),
+	     TP_ARGS(hw, desc, index, num));
+
+DEFINE_EVENT(hclge_vf_cmd_template, hclge_vf_cmd_get,
+	     TP_PROTO(struct hclge_comm_hw *hw,
+		      struct hclge_desc *desc,
+		      int index,
+		      int num),
+	     TP_ARGS(hw, desc, index, num));
+
+#endif /* _HCLGEVF_TRACE_H_ */
+
+/* This must be outside ifdef _HCLGEVF_TRACE_H */
+#undef TRACE_INCLUDE_PATH
+#define TRACE_INCLUDE_PATH .
+#undef TRACE_INCLUDE_FILE
+#define TRACE_INCLUDE_FILE hclgevf_trace
+#include <trace/define_trace.h>
diff --git a/drivers/net/ethernet/hisilicon/hns_mdio.c b/drivers/net/ethernet/hisilicon/hns_mdio.c
index e5221d95afe1..6812be8dc64f 100644
--- a/drivers/net/ethernet/hisilicon/hns_mdio.c
+++ b/drivers/net/ethernet/hisilicon/hns_mdio.c
@@ -1,10 +1,6 @@
+// SPDX-License-Identifier: GPL-2.0-or-later
 /*
  * Copyright (c) 2014-2015 Hisilicon Limited.
- *
- * This program is free software; you can redistribute it and/or modify
- * it under the terms of the GNU General Public License as published by
- * the Free Software Foundation; either version 2 of the License, or
- * (at your option) any later version.
  */
 
 #include <linux/acpi.h>
@@ -39,7 +35,7 @@ struct hns_mdio_sc_reg {
 };
 
 struct hns_mdio_device {
-	void *vbase;		/* mdio reg base address */
+	u8 __iomem *vbase;		/* mdio reg base address */
 	struct regmap *subctrl_vbase;
 	struct hns_mdio_sc_reg sc_reg;
 };
@@ -96,21 +92,17 @@ enum mdio_c45_op_seq {
 #define MDIO_SC_CLK_ST		0x531C
 #define MDIO_SC_RESET_ST	0x5A1C
 
-static void mdio_write_reg(void *base, u32 reg, u32 value)
+static void mdio_write_reg(u8 __iomem *base, u32 reg, u32 value)
 {
-	u8 __iomem *reg_addr = (u8 __iomem *)base;
-
-	writel_relaxed(value, reg_addr + reg);
+	writel_relaxed(value, base + reg);
 }
 
 #define MDIO_WRITE_REG(a, reg, value) \
 	mdio_write_reg((a)->vbase, (reg), (value))
 
-static u32 mdio_read_reg(void *base, u32 reg)
+static u32 mdio_read_reg(u8 __iomem *base, u32 reg)
 {
-	u8 __iomem *reg_addr = (u8 __iomem *)base;
-
-	return readl_relaxed(reg_addr + reg);
+	return readl_relaxed(base + reg);
 }
 
 #define mdio_set_field(origin, mask, shift, val) \
@@ -121,7 +113,7 @@ static u32 mdio_read_reg(void *base, u32 reg)
 
 #define mdio_get_field(origin, mask, shift) (((origin) >> (shift)) & (mask))
 
-static void mdio_set_reg_field(void *base, u32 reg, u32 mask, u32 shift,
+static void mdio_set_reg_field(u8 __iomem *base, u32 reg, u32 mask, u32 shift,
 			       u32 val)
 {
 	u32 origin = mdio_read_reg(base, reg);
@@ -133,7 +125,7 @@ static void mdio_set_reg_field(void *base, u32 reg, u32 mask, u32 shift,
 #define MDIO_SET_REG_FIELD(dev, reg, mask, shift, val) \
 	mdio_set_reg_field((dev)->vbase, (reg), (mask), (shift), (val))
 
-static u32 mdio_get_reg_field(void *base, u32 reg, u32 mask, u32 shift)
+static u32 mdio_get_reg_field(u8 __iomem *base, u32 reg, u32 mask, u32 shift)
 {
 	u32 origin;
 
@@ -156,11 +148,15 @@ static int mdio_sc_cfg_reg_write(struct hns_mdio_device *mdio_dev,
 {
 	u32 time_cnt;
 	u32 reg_value;
+	int ret;
 
 	regmap_write(mdio_dev->subctrl_vbase, cfg_reg, set_val);
 
 	for (time_cnt = MDIO_TIMEOUT; time_cnt; time_cnt--) {
-		regmap_read(mdio_dev->subctrl_vbase, st_reg, &reg_value);
+		ret = regmap_read(mdio_dev->subctrl_vbase, st_reg, &reg_value);
+		if (ret)
+			return ret;
+
 		reg_value &= st_msk;
 		if ((!!check_st) == (!!reg_value))
 			break;
@@ -178,7 +174,7 @@ static int hns_mdio_wait_ready(struct mii_bus *bus)
 	u32 cmd_reg_value;
 	int i;
 
-	/* waitting for MDIO_COMMAND_REG 's mdio_start==0 */
+	/* waiting for MDIO_COMMAND_REG's mdio_start==0 */
 	/* after that can do read or write*/
 	for (i = 0; i < MDIO_TIMEOUT; i++) {
 		cmd_reg_value = MDIO_GET_REG_BIT(mdio_dev,
@@ -210,29 +206,27 @@ static void hns_mdio_cmd_write(struct hns_mdio_device *mdio_dev,
 }
 
 /**
- * hns_mdio_write - access phy register
+ * hns_mdio_write_c22 - access phy register
  * @bus: mdio bus
  * @phy_id: phy id
  * @regnum: register num
- * @value: register value
+ * @data: register value
  *
  * Return 0 on success, negative on failure
  */
-static int hns_mdio_write(struct mii_bus *bus,
-			  int phy_id, int regnum, u16 data)
+static int hns_mdio_write_c22(struct mii_bus *bus,
+			      int phy_id, int regnum, u16 data)
 {
-	int ret;
-	struct hns_mdio_device *mdio_dev = (struct hns_mdio_device *)bus->priv;
-	u8 devad = ((regnum >> 16) & 0x1f);
-	u8 is_c45 = !!(regnum & MII_ADDR_C45);
+	struct hns_mdio_device *mdio_dev = bus->priv;
 	u16 reg = (u16)(regnum & 0xffff);
-	u8 op;
 	u16 cmd_reg_cfg;
+	int ret;
+	u8 op;
 
 	dev_dbg(&bus->dev, "mdio write %s,base is %p\n",
 		bus->id, mdio_dev->vbase);
-	dev_dbg(&bus->dev, "phy id=%d, is_c45=%d, devad=%d, reg=%#x, write data=%d\n",
-		phy_id, is_c45, devad, reg, data);
+	dev_dbg(&bus->dev, "phy id=%d, reg=%#x, write data=%d\n",
+		phy_id, reg, data);
 
 	/* wait for ready */
 	ret = hns_mdio_wait_ready(bus);
@@ -241,59 +235,91 @@ static int hns_mdio_write(struct mii_bus *bus,
 		return ret;
 	}
 
-	if (!is_c45) {
-		cmd_reg_cfg = reg;
-		op = MDIO_C22_WRITE;
-	} else {
-		/* config the cmd-reg to write addr*/
-		MDIO_SET_REG_FIELD(mdio_dev, MDIO_ADDR_REG, MDIO_ADDR_DATA_M,
-				   MDIO_ADDR_DATA_S, reg);
+	cmd_reg_cfg = reg;
+	op = MDIO_C22_WRITE;
+
+	MDIO_SET_REG_FIELD(mdio_dev, MDIO_WDATA_REG, MDIO_WDATA_DATA_M,
+			   MDIO_WDATA_DATA_S, data);
 
-		hns_mdio_cmd_write(mdio_dev, is_c45,
-				   MDIO_C45_WRITE_ADDR, phy_id, devad);
+	hns_mdio_cmd_write(mdio_dev, false, op, phy_id, cmd_reg_cfg);
 
-		/* check for read or write opt is finished */
-		ret = hns_mdio_wait_ready(bus);
-		if (ret) {
-			dev_err(&bus->dev, "MDIO bus is busy\n");
-			return ret;
-		}
+	return 0;
+}
 
-		/* config the data needed writing */
-		cmd_reg_cfg = devad;
-		op = MDIO_C45_WRITE_ADDR;
+/**
+ * hns_mdio_write_c45 - access phy register
+ * @bus: mdio bus
+ * @phy_id: phy id
+ * @devad: device address to read
+ * @regnum: register num
+ * @data: register value
+ *
+ * Return 0 on success, negative on failure
+ */
+static int hns_mdio_write_c45(struct mii_bus *bus, int phy_id, int devad,
+			      int regnum, u16 data)
+{
+	struct hns_mdio_device *mdio_dev = bus->priv;
+	u16 reg = (u16)(regnum & 0xffff);
+	u16 cmd_reg_cfg;
+	int ret;
+	u8 op;
+
+	dev_dbg(&bus->dev, "mdio write %s,base is %p\n",
+		bus->id, mdio_dev->vbase);
+	dev_dbg(&bus->dev, "phy id=%d, devad=%d, reg=%#x, write data=%d\n",
+		phy_id, devad, reg, data);
+
+	/* wait for ready */
+	ret = hns_mdio_wait_ready(bus);
+	if (ret) {
+		dev_err(&bus->dev, "MDIO bus is busy\n");
+		return ret;
 	}
 
+	/* config the cmd-reg to write addr*/
+	MDIO_SET_REG_FIELD(mdio_dev, MDIO_ADDR_REG, MDIO_ADDR_DATA_M,
+			   MDIO_ADDR_DATA_S, reg);
+
+	hns_mdio_cmd_write(mdio_dev, true, MDIO_C45_WRITE_ADDR, phy_id, devad);
+
+	/* check for read or write opt is finished */
+	ret = hns_mdio_wait_ready(bus);
+	if (ret) {
+		dev_err(&bus->dev, "MDIO bus is busy\n");
+		return ret;
+	}
+
+	/* config the data needed writing */
+	cmd_reg_cfg = devad;
+	op = MDIO_C45_WRITE_DATA;
+
 	MDIO_SET_REG_FIELD(mdio_dev, MDIO_WDATA_REG, MDIO_WDATA_DATA_M,
 			   MDIO_WDATA_DATA_S, data);
 
-	hns_mdio_cmd_write(mdio_dev, is_c45, op, phy_id, cmd_reg_cfg);
+	hns_mdio_cmd_write(mdio_dev, true, op, phy_id, cmd_reg_cfg);
 
 	return 0;
 }
 
 /**
- * hns_mdio_read - access phy register
+ * hns_mdio_read_c22 - access phy register
  * @bus: mdio bus
  * @phy_id: phy id
  * @regnum: register num
- * @value: register value
  *
  * Return phy register value
  */
-static int hns_mdio_read(struct mii_bus *bus, int phy_id, int regnum)
+static int hns_mdio_read_c22(struct mii_bus *bus, int phy_id, int regnum)
 {
-	int ret;
-	u16 reg_val = 0;
-	u8 devad = ((regnum >> 16) & 0x1f);
-	u8 is_c45 = !!(regnum & MII_ADDR_C45);
+	struct hns_mdio_device *mdio_dev = bus->priv;
 	u16 reg = (u16)(regnum & 0xffff);
-	struct hns_mdio_device *mdio_dev = (struct hns_mdio_device *)bus->priv;
+	u16 reg_val;
+	int ret;
 
 	dev_dbg(&bus->dev, "mdio read %s,base is %p\n",
 		bus->id, mdio_dev->vbase);
-	dev_dbg(&bus->dev, "phy id=%d, is_c45=%d, devad=%d, reg=%#x!\n",
-		phy_id, is_c45, devad, reg);
+	dev_dbg(&bus->dev, "phy id=%d, reg=%#x!\n", phy_id, reg);
 
 	/* Step 1: wait for ready */
 	ret = hns_mdio_wait_ready(bus);
@@ -302,29 +328,74 @@ static int hns_mdio_read(struct mii_bus *bus, int phy_id, int regnum)
 		return ret;
 	}
 
-	if (!is_c45) {
-		hns_mdio_cmd_write(mdio_dev, is_c45,
-				   MDIO_C22_READ, phy_id, reg);
-	} else {
-		MDIO_SET_REG_FIELD(mdio_dev, MDIO_ADDR_REG, MDIO_ADDR_DATA_M,
-				   MDIO_ADDR_DATA_S, reg);
+	hns_mdio_cmd_write(mdio_dev, false, MDIO_C22_READ, phy_id, reg);
 
-		/* Step 2; config the cmd-reg to write addr*/
-		hns_mdio_cmd_write(mdio_dev, is_c45,
-				   MDIO_C45_WRITE_ADDR, phy_id, devad);
+	/* Step 2: waiting for MDIO_COMMAND_REG 's mdio_start==0,*/
+	/* check for read or write opt is finished */
+	ret = hns_mdio_wait_ready(bus);
+	if (ret) {
+		dev_err(&bus->dev, "MDIO bus is busy\n");
+		return ret;
+	}
 
-		/* Step 3: check for read or write opt is finished */
-		ret = hns_mdio_wait_ready(bus);
-		if (ret) {
-			dev_err(&bus->dev, "MDIO bus is busy\n");
-			return ret;
-		}
+	reg_val = MDIO_GET_REG_BIT(mdio_dev, MDIO_STA_REG, MDIO_STATE_STA_B);
+	if (reg_val) {
+		dev_err(&bus->dev, " ERROR! MDIO Read failed!\n");
+		return -EBUSY;
+	}
+
+	/* Step 3; get out data*/
+	reg_val = (u16)MDIO_GET_REG_FIELD(mdio_dev, MDIO_RDATA_REG,
+					  MDIO_RDATA_DATA_M, MDIO_RDATA_DATA_S);
+
+	return reg_val;
+}
+
+/**
+ * hns_mdio_read_c45 - access phy register
+ * @bus: mdio bus
+ * @phy_id: phy id
+ * @devad: device address to read
+ * @regnum: register num
+ *
+ * Return phy register value
+ */
+static int hns_mdio_read_c45(struct mii_bus *bus, int phy_id, int devad,
+			     int regnum)
+{
+	struct hns_mdio_device *mdio_dev = bus->priv;
+	u16 reg = (u16)(regnum & 0xffff);
+	u16 reg_val;
+	int ret;
 
-		hns_mdio_cmd_write(mdio_dev, is_c45,
-				   MDIO_C45_WRITE_ADDR, phy_id, devad);
+	dev_dbg(&bus->dev, "mdio read %s,base is %p\n",
+		bus->id, mdio_dev->vbase);
+	dev_dbg(&bus->dev, "phy id=%d, devad=%d, reg=%#x!\n",
+		phy_id, devad, reg);
+
+	/* Step 1: wait for ready */
+	ret = hns_mdio_wait_ready(bus);
+	if (ret) {
+		dev_err(&bus->dev, "MDIO bus is busy\n");
+		return ret;
 	}
 
-	/* Step 5: waitting for MDIO_COMMAND_REG 's mdio_start==0,*/
+	MDIO_SET_REG_FIELD(mdio_dev, MDIO_ADDR_REG, MDIO_ADDR_DATA_M,
+			   MDIO_ADDR_DATA_S, reg);
+
+	/* Step 2; config the cmd-reg to write addr*/
+	hns_mdio_cmd_write(mdio_dev, true, MDIO_C45_WRITE_ADDR, phy_id, devad);
+
+	/* Step 3: check for read or write opt is finished */
+	ret = hns_mdio_wait_ready(bus);
+	if (ret) {
+		dev_err(&bus->dev, "MDIO bus is busy\n");
+		return ret;
+	}
+
+	hns_mdio_cmd_write(mdio_dev, true, MDIO_C45_READ, phy_id, devad);
+
+	/* Step 5: waiting for MDIO_COMMAND_REG 's mdio_start==0,*/
 	/* check for read or write opt is finished */
 	ret = hns_mdio_wait_ready(bus);
 	if (ret) {
@@ -353,13 +424,13 @@ static int hns_mdio_read(struct mii_bus *bus, int phy_id, int regnum)
  */
 static int hns_mdio_reset(struct mii_bus *bus)
 {
-	struct hns_mdio_device *mdio_dev = (struct hns_mdio_device *)bus->priv;
+	struct hns_mdio_device *mdio_dev = bus->priv;
 	const struct hns_mdio_sc_reg *sc_reg;
 	int ret;
 
 	if (dev_of_node(bus->parent)) {
 		if (!mdio_dev->subctrl_vbase) {
-			dev_err(&bus->dev, "mdio sys ctl reg has not maped\n");
+			dev_err(&bus->dev, "mdio sys ctl reg has not mapped\n");
 			return -ENODEV;
 		}
 
@@ -425,8 +496,7 @@ static int hns_mdio_probe(struct platform_device *pdev)
 {
 	struct hns_mdio_device *mdio_dev;
 	struct mii_bus *new_bus;
-	struct resource *res;
-	int ret = -ENODEV;
+	int ret;
 
 	if (!pdev) {
 		dev_err(NULL, "pdev is NULL!\r\n");
@@ -444,14 +514,15 @@ static int hns_mdio_probe(struct platform_device *pdev)
 	}
 
 	new_bus->name = MDIO_BUS_NAME;
-	new_bus->read = hns_mdio_read;
-	new_bus->write = hns_mdio_write;
+	new_bus->read = hns_mdio_read_c22;
+	new_bus->write = hns_mdio_write_c22;
+	new_bus->read_c45 = hns_mdio_read_c45;
+	new_bus->write_c45 = hns_mdio_write_c45;
 	new_bus->reset = hns_mdio_reset;
 	new_bus->priv = mdio_dev;
 	new_bus->parent = &pdev->dev;
 
-	res = platform_get_resource(pdev, IORESOURCE_MEM, 0);
-	mdio_dev->vbase = devm_ioremap_resource(&pdev->dev, res);
+	mdio_dev->vbase = devm_platform_ioremap_resource(pdev, 0);
 	if (IS_ERR(mdio_dev->vbase)) {
 		ret = PTR_ERR(mdio_dev->vbase);
 		return ret;
@@ -504,6 +575,7 @@ static int hns_mdio_probe(struct platform_device *pdev)
 						MDIO_SC_RESET_ST;
 				}
 			}
+			of_node_put(reg_args.np);
 		} else {
 			dev_warn(&pdev->dev, "find syscon ret = %#x\n", ret);
 			mdio_dev->subctrl_vbase = NULL;
@@ -539,7 +611,7 @@ static int hns_mdio_probe(struct platform_device *pdev)
  *
  * Return 0 on success, negative on failure
  */
-static int hns_mdio_remove(struct platform_device *pdev)
+static void hns_mdio_remove(struct platform_device *pdev)
 {
 	struct mii_bus *bus;
 
@@ -547,7 +619,6 @@ static int hns_mdio_remove(struct platform_device *pdev)
 
 	mdiobus_unregister(bus);
 	platform_set_drvdata(pdev, NULL);
-	return 0;
 }
 
 static const struct of_device_id hns_mdio_match[] = {
@@ -569,7 +640,7 @@ static struct platform_driver hns_mdio_driver = {
 	.driver = {
 		   .name = MDIO_DRV_NAME,
 		   .of_match_table = hns_mdio_match,
-		   .acpi_match_table = ACPI_PTR(hns_mdio_acpi_match),
+		   .acpi_match_table = hns_mdio_acpi_match,
 		   },
 };
 
diff --git a/drivers/net/ethernet/hp/Kconfig b/drivers/net/ethernet/hp/Kconfig
deleted file mode 100644
index d4df78c2abce..000000000000
--- a/drivers/net/ethernet/hp/Kconfig
+++ /dev/null
@@ -1,28 +0,0 @@
-#
-# HP network device configuration
-#
-
-config NET_VENDOR_HP
-	bool "HP devices"
-	default y
-	depends on ISA || EISA || PCI
-	---help---
-	  If you have a network (Ethernet) card belonging to this class, say Y.
-
-	  Note that the answer to this question doesn't directly affect the
-	  kernel: saying N will just cause the configurator to skip all
-	  the questions about HP cards. If you say Y, you will be asked for
-	  your specific card in the following questions.
-
-if NET_VENDOR_HP
-
-config HP100
-	tristate "HP 10/100VG PCLAN (ISA, EISA, PCI) support"
-	depends on (ISA || EISA || PCI)
-	---help---
-	  If you have a network (Ethernet) card of this type, say Y here.
-
-	  To compile this driver as a module, choose M here. The module
-	  will be called hp100.
-
-endif # NET_VENDOR_HP
diff --git a/drivers/net/ethernet/hp/Makefile b/drivers/net/ethernet/hp/Makefile
deleted file mode 100644
index 20b6918b52bd..000000000000
--- a/drivers/net/ethernet/hp/Makefile
+++ /dev/null
@@ -1,5 +0,0 @@
-#
-# Makefile for the HP network device drivers.
-#
-
-obj-$(CONFIG_HP100) += hp100.o
diff --git a/drivers/net/ethernet/hp/hp100.c b/drivers/net/ethernet/hp/hp100.c
deleted file mode 100644
index 5673b071e39d..000000000000
--- a/drivers/net/ethernet/hp/hp100.c
+++ /dev/null
@@ -1,3049 +0,0 @@
-/*
-** hp100.c
-** HP CASCADE Architecture Driver for 100VG-AnyLan Network Adapters
-**
-** $Id: hp100.c,v 1.58 2001/09/24 18:03:01 perex Exp perex $
-**
-** Based on the HP100 driver written by Jaroslav Kysela <perex@jcu.cz>
-** Extended for new busmaster capable chipsets by
-** Siegfried "Frieder" Loeffler (dg1sek) <floeff@mathematik.uni-stuttgart.de>
-**
-** Maintained by: Jaroslav Kysela <perex@perex.cz>
-**
-** This driver has only been tested with
-** -- HP J2585B 10/100 Mbit/s PCI Busmaster
-** -- HP J2585A 10/100 Mbit/s PCI
-** -- HP J2970A 10 Mbit/s PCI Combo 10base-T/BNC
-** -- HP J2973A 10 Mbit/s PCI 10base-T
-** -- HP J2573  10/100 ISA
-** -- Compex ReadyLink ENET100-VG4  10/100 Mbit/s PCI / EISA
-** -- Compex FreedomLine 100/VG  10/100 Mbit/s ISA / EISA / PCI
-**
-** but it should also work with the other CASCADE based adapters.
-**
-** TODO:
-**       -  J2573 seems to hang sometimes when in shared memory mode.
-**       -  Mode for Priority TX
-**       -  Check PCI registers, performance might be improved?
-**       -  To reduce interrupt load in busmaster, one could switch off
-**          the interrupts that are used to refill the queues whenever the
-**          queues are filled up to more than a certain threshold.
-**       -  some updates for EISA version of card
-**
-**
-**   This code is free software; you can redistribute it and/or modify
-**   it under the terms of the GNU General Public License as published by
-**   the Free Software Foundation; either version 2 of the License, or
-**   (at your option) any later version.
-**
-**   This code is distributed in the hope that it will be useful,
-**   but WITHOUT ANY WARRANTY; without even the implied warranty of
-**   MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
-**   GNU General Public License for more details.
-**
-**   You should have received a copy of the GNU General Public License
-**   along with this program; if not, write to the Free Software
-**   Foundation, Inc., 675 Mass Ave, Cambridge, MA 02139, USA.
-**
-** 1.57c -> 1.58
-**   - used indent to change coding-style
-**   - added KTI DP-200 EISA ID
-**   - ioremap is also used for low (<1MB) memory (multi-architecture support)
-**
-** 1.57b -> 1.57c - Arnaldo Carvalho de Melo <acme@conectiva.com.br>
-**   - release resources on failure in init_module
-**
-** 1.57 -> 1.57b - Jean II
-**   - fix spinlocks, SMP is now working !
-**
-** 1.56 -> 1.57
-**   - updates for new PCI interface for 2.1 kernels
-**
-** 1.55 -> 1.56
-**   - removed printk in misc. interrupt and update statistics to allow
-**     monitoring of card status
-**   - timing changes in xmit routines, relogin to 100VG hub added when
-**     driver does reset
-**   - included fix for Compex FreedomLine PCI adapter
-**
-** 1.54 -> 1.55
-**   - fixed bad initialization in init_module
-**   - added Compex FreedomLine adapter
-**   - some fixes in card initialization
-**
-** 1.53 -> 1.54
-**   - added hardware multicast filter support (doesn't work)
-**   - little changes in hp100_sense_lan routine
-**     - added support for Coax and AUI (J2970)
-**   - fix for multiple cards and hp100_mode parameter (insmod)
-**   - fix for shared IRQ
-**
-** 1.52 -> 1.53
-**   - fixed bug in multicast support
-**
-*/
-
-#define HP100_DEFAULT_PRIORITY_TX 0
-
-#undef HP100_DEBUG
-#undef HP100_DEBUG_B		/* Trace  */
-#undef HP100_DEBUG_BM		/* Debug busmaster code (PDL stuff) */
-
-#undef HP100_DEBUG_TRAINING	/* Debug login-to-hub procedure */
-#undef HP100_DEBUG_TX
-#undef HP100_DEBUG_IRQ
-#undef HP100_DEBUG_RX
-
-#undef HP100_MULTICAST_FILTER	/* Need to be debugged... */
-
-#include <linux/module.h>
-#include <linux/kernel.h>
-#include <linux/sched.h>
-#include <linux/string.h>
-#include <linux/errno.h>
-#include <linux/ioport.h>
-#include <linux/interrupt.h>
-#include <linux/eisa.h>
-#include <linux/pci.h>
-#include <linux/dma-mapping.h>
-#include <linux/spinlock.h>
-#include <linux/netdevice.h>
-#include <linux/etherdevice.h>
-#include <linux/skbuff.h>
-#include <linux/types.h>
-#include <linux/delay.h>
-#include <linux/init.h>
-#include <linux/bitops.h>
-#include <linux/jiffies.h>
-
-#include <asm/io.h>
-
-#include "hp100.h"
-
-/*
- *  defines
- */
-
-#define HP100_BUS_ISA     0
-#define HP100_BUS_EISA    1
-#define HP100_BUS_PCI     2
-
-#define HP100_REGION_SIZE	0x20	/* for ioports */
-#define HP100_SIG_LEN		8	/* same as EISA_SIG_LEN */
-
-#define HP100_MAX_PACKET_SIZE	(1536+4)
-#define HP100_MIN_PACKET_SIZE	60
-
-#ifndef HP100_DEFAULT_RX_RATIO
-/* default - 75% onboard memory on the card are used for RX packets */
-#define HP100_DEFAULT_RX_RATIO	75
-#endif
-
-#ifndef HP100_DEFAULT_PRIORITY_TX
-/* default - don't enable transmit outgoing packets as priority */
-#define HP100_DEFAULT_PRIORITY_TX 0
-#endif
-
-/*
- *  structures
- */
-
-struct hp100_private {
-	spinlock_t lock;
-	char id[HP100_SIG_LEN];
-	u_short chip;
-	u_short soft_model;
-	u_int memory_size;
-	u_int virt_memory_size;
-	u_short rx_ratio;	/* 1 - 99 */
-	u_short priority_tx;	/* != 0 - priority tx */
-	u_short mode;		/* PIO, Shared Mem or Busmaster */
-	u_char bus;
-	struct pci_dev *pci_dev;
-	short mem_mapped;	/* memory mapped access */
-	void __iomem *mem_ptr_virt;	/* virtual memory mapped area, maybe NULL */
-	unsigned long mem_ptr_phys;	/* physical memory mapped area */
-	short lan_type;		/* 10Mb/s, 100Mb/s or -1 (error) */
-	int hub_status;		/* was login to hub successful? */
-	u_char mac1_mode;
-	u_char mac2_mode;
-	u_char hash_bytes[8];
-
-	/* Rings for busmaster mode: */
-	hp100_ring_t *rxrhead;	/* Head (oldest) index into rxring */
-	hp100_ring_t *rxrtail;	/* Tail (newest) index into rxring */
-	hp100_ring_t *txrhead;	/* Head (oldest) index into txring */
-	hp100_ring_t *txrtail;	/* Tail (newest) index into txring */
-
-	hp100_ring_t rxring[MAX_RX_PDL];
-	hp100_ring_t txring[MAX_TX_PDL];
-
-	u_int *page_vaddr_algn;	/* Aligned virtual address of allocated page */
-	u_long whatever_offset;	/* Offset to bus/phys/dma address */
-	int rxrcommit;		/* # Rx PDLs committed to adapter */
-	int txrcommit;		/* # Tx PDLs committed to adapter */
-};
-
-/*
- *  variables
- */
-#ifdef CONFIG_ISA
-static const char *hp100_isa_tbl[] = {
-	"HWPF150", /* HP J2573 rev A */
-	"HWP1950", /* HP J2573 */
-};
-#endif
-
-static struct eisa_device_id hp100_eisa_tbl[] = {
-	{ "HWPF180" }, /* HP J2577 rev A */
-	{ "HWP1920" }, /* HP 27248B */
-	{ "HWP1940" }, /* HP J2577 */
-	{ "HWP1990" }, /* HP J2577 */
-	{ "CPX0301" }, /* ReadyLink ENET100-VG4 */
-	{ "CPX0401" }, /* FreedomLine 100/VG */
-	{ "" }	       /* Mandatory final entry ! */
-};
-MODULE_DEVICE_TABLE(eisa, hp100_eisa_tbl);
-
-static const struct pci_device_id hp100_pci_tbl[] = {
-	{PCI_VENDOR_ID_HP, PCI_DEVICE_ID_HP_J2585A, PCI_ANY_ID, PCI_ANY_ID,},
-	{PCI_VENDOR_ID_HP, PCI_DEVICE_ID_HP_J2585B, PCI_ANY_ID, PCI_ANY_ID,},
-	{PCI_VENDOR_ID_HP, PCI_DEVICE_ID_HP_J2970A, PCI_ANY_ID, PCI_ANY_ID,},
-	{PCI_VENDOR_ID_HP, PCI_DEVICE_ID_HP_J2973A, PCI_ANY_ID, PCI_ANY_ID,},
-	{PCI_VENDOR_ID_COMPEX, PCI_DEVICE_ID_COMPEX_ENET100VG4, PCI_ANY_ID, PCI_ANY_ID,},
-	{PCI_VENDOR_ID_COMPEX2, PCI_DEVICE_ID_COMPEX2_100VG, PCI_ANY_ID, PCI_ANY_ID,},
-/*	{PCI_VENDOR_ID_KTI, PCI_DEVICE_ID_KTI_DP200, PCI_ANY_ID, PCI_ANY_ID }, */
-	{}			/* Terminating entry */
-};
-MODULE_DEVICE_TABLE(pci, hp100_pci_tbl);
-
-static int hp100_rx_ratio = HP100_DEFAULT_RX_RATIO;
-static int hp100_priority_tx = HP100_DEFAULT_PRIORITY_TX;
-static int hp100_mode = 1;
-
-module_param(hp100_rx_ratio, int, 0);
-module_param(hp100_priority_tx, int, 0);
-module_param(hp100_mode, int, 0);
-
-/*
- *  prototypes
- */
-
-static int hp100_probe1(struct net_device *dev, int ioaddr, u_char bus,
-			struct pci_dev *pci_dev);
-
-
-static int hp100_open(struct net_device *dev);
-static int hp100_close(struct net_device *dev);
-static netdev_tx_t hp100_start_xmit(struct sk_buff *skb,
-				    struct net_device *dev);
-static netdev_tx_t hp100_start_xmit_bm(struct sk_buff *skb,
-				       struct net_device *dev);
-static void hp100_rx(struct net_device *dev);
-static struct net_device_stats *hp100_get_stats(struct net_device *dev);
-static void hp100_misc_interrupt(struct net_device *dev);
-static void hp100_update_stats(struct net_device *dev);
-static void hp100_clear_stats(struct hp100_private *lp, int ioaddr);
-static void hp100_set_multicast_list(struct net_device *dev);
-static irqreturn_t hp100_interrupt(int irq, void *dev_id);
-static void hp100_start_interface(struct net_device *dev);
-static void hp100_stop_interface(struct net_device *dev);
-static void hp100_load_eeprom(struct net_device *dev, u_short ioaddr);
-static int hp100_sense_lan(struct net_device *dev);
-static int hp100_login_to_vg_hub(struct net_device *dev,
-				 u_short force_relogin);
-static int hp100_down_vg_link(struct net_device *dev);
-static void hp100_cascade_reset(struct net_device *dev, u_short enable);
-static void hp100_BM_shutdown(struct net_device *dev);
-static void hp100_mmuinit(struct net_device *dev);
-static void hp100_init_pdls(struct net_device *dev);
-static int hp100_init_rxpdl(struct net_device *dev,
-			    register hp100_ring_t * ringptr,
-			    register u_int * pdlptr);
-static int hp100_init_txpdl(struct net_device *dev,
-			    register hp100_ring_t * ringptr,
-			    register u_int * pdlptr);
-static void hp100_rxfill(struct net_device *dev);
-static void hp100_hwinit(struct net_device *dev);
-static void hp100_clean_txring(struct net_device *dev);
-#ifdef HP100_DEBUG
-static void hp100_RegisterDump(struct net_device *dev);
-#endif
-
-/* Conversion to new PCI API :
- * Convert an address in a kernel buffer to a bus/phys/dma address.
- * This work *only* for memory fragments part of lp->page_vaddr,
- * because it was properly DMA allocated via pci_alloc_consistent(),
- * so we just need to "retrieve" the original mapping to bus/phys/dma
- * address - Jean II */
-static inline dma_addr_t virt_to_whatever(struct net_device *dev, u32 * ptr)
-{
-	struct hp100_private *lp = netdev_priv(dev);
-	return ((u_long) ptr) + lp->whatever_offset;
-}
-
-static inline u_int pdl_map_data(struct hp100_private *lp, void *data)
-{
-	return pci_map_single(lp->pci_dev, data,
-			      MAX_ETHER_SIZE, PCI_DMA_FROMDEVICE);
-}
-
-/* TODO: This function should not really be needed in a good design... */
-static void wait(void)
-{
-	mdelay(1);
-}
-
-/*
- *  probe functions
- *  These functions should - if possible - avoid doing write operations
- *  since this could cause problems when the card is not installed.
- */
-
-/*
- * Read board id and convert to string.
- * Effectively same code as decode_eisa_sig
- */
-static const char *hp100_read_id(int ioaddr)
-{
-	int i;
-	static char str[HP100_SIG_LEN];
-	unsigned char sig[4], sum;
-        unsigned short rev;
-
-	hp100_page(ID_MAC_ADDR);
-	sum = 0;
-	for (i = 0; i < 4; i++) {
-		sig[i] = hp100_inb(BOARD_ID + i);
-		sum += sig[i];
-	}
-
-	sum += hp100_inb(BOARD_ID + i);
-	if (sum != 0xff)
-		return NULL;	/* bad checksum */
-
-        str[0] = ((sig[0] >> 2) & 0x1f) + ('A' - 1);
-        str[1] = (((sig[0] & 3) << 3) | (sig[1] >> 5)) + ('A' - 1);
-        str[2] = (sig[1] & 0x1f) + ('A' - 1);
-        rev = (sig[2] << 8) | sig[3];
-        sprintf(str + 3, "%04X", rev);
-
-	return str;
-}
-
-#ifdef CONFIG_ISA
-static __init int hp100_isa_probe1(struct net_device *dev, int ioaddr)
-{
-	const char *sig;
-	int i;
-
-	if (!request_region(ioaddr, HP100_REGION_SIZE, "hp100"))
-		goto err;
-
-	if (hp100_inw(HW_ID) != HP100_HW_ID_CASCADE) {
-		release_region(ioaddr, HP100_REGION_SIZE);
-		goto err;
-	}
-
-	sig = hp100_read_id(ioaddr);
-	release_region(ioaddr, HP100_REGION_SIZE);
-
-	if (sig == NULL)
-		goto err;
-
-	for (i = 0; i < ARRAY_SIZE(hp100_isa_tbl); i++) {
-		if (!strcmp(hp100_isa_tbl[i], sig))
-			break;
-
-	}
-
-	if (i < ARRAY_SIZE(hp100_isa_tbl))
-		return hp100_probe1(dev, ioaddr, HP100_BUS_ISA, NULL);
- err:
-	return -ENODEV;
-
-}
-/*
- * Probe for ISA board.
- * EISA and PCI are handled by device infrastructure.
- */
-
-static int  __init hp100_isa_probe(struct net_device *dev, int addr)
-{
-	int err = -ENODEV;
-
-	/* Probe for a specific ISA address */
-	if (addr > 0xff && addr < 0x400)
-		err = hp100_isa_probe1(dev, addr);
-
-	else if (addr != 0)
-		err = -ENXIO;
-
-	else {
-		/* Probe all ISA possible port regions */
-		for (addr = 0x100; addr < 0x400; addr += 0x20) {
-			err = hp100_isa_probe1(dev, addr);
-			if (!err)
-				break;
-		}
-	}
-	return err;
-}
-#endif /* CONFIG_ISA */
-
-#if !defined(MODULE) && defined(CONFIG_ISA)
-struct net_device * __init hp100_probe(int unit)
-{
-	struct net_device *dev = alloc_etherdev(sizeof(struct hp100_private));
-	int err;
-
-	if (!dev)
-		return ERR_PTR(-ENODEV);
-
-#ifdef HP100_DEBUG_B
-	hp100_outw(0x4200, TRACE);
-	printk("hp100: %s: probe\n", dev->name);
-#endif
-
-	if (unit >= 0) {
-		sprintf(dev->name, "eth%d", unit);
-		netdev_boot_setup_check(dev);
-	}
-
-	err = hp100_isa_probe(dev, dev->base_addr);
-	if (err)
-		goto out;
-
-	return dev;
- out:
-	free_netdev(dev);
-	return ERR_PTR(err);
-}
-#endif /* !MODULE && CONFIG_ISA */
-
-static const struct net_device_ops hp100_bm_netdev_ops = {
-	.ndo_open		= hp100_open,
-	.ndo_stop		= hp100_close,
-	.ndo_start_xmit		= hp100_start_xmit_bm,
-	.ndo_get_stats 		= hp100_get_stats,
-	.ndo_set_rx_mode	= hp100_set_multicast_list,
-	.ndo_set_mac_address 	= eth_mac_addr,
-	.ndo_validate_addr	= eth_validate_addr,
-};
-
-static const struct net_device_ops hp100_netdev_ops = {
-	.ndo_open		= hp100_open,
-	.ndo_stop		= hp100_close,
-	.ndo_start_xmit		= hp100_start_xmit,
-	.ndo_get_stats 		= hp100_get_stats,
-	.ndo_set_rx_mode	= hp100_set_multicast_list,
-	.ndo_set_mac_address 	= eth_mac_addr,
-	.ndo_validate_addr	= eth_validate_addr,
-};
-
-static int hp100_probe1(struct net_device *dev, int ioaddr, u_char bus,
-			struct pci_dev *pci_dev)
-{
-	int i;
-	int err = -ENODEV;
-	const char *eid;
-	u_int chip;
-	u_char uc;
-	u_int memory_size = 0, virt_memory_size = 0;
-	u_short local_mode, lsw;
-	short mem_mapped;
-	unsigned long mem_ptr_phys;
-	void __iomem *mem_ptr_virt;
-	struct hp100_private *lp;
-
-#ifdef HP100_DEBUG_B
-	hp100_outw(0x4201, TRACE);
-	printk("hp100: %s: probe1\n", dev->name);
-#endif
-
-	/* memory region for programmed i/o */
-	if (!request_region(ioaddr, HP100_REGION_SIZE, "hp100"))
-		goto out1;
-
-	if (hp100_inw(HW_ID) != HP100_HW_ID_CASCADE)
-		goto out2;
-
-	chip = hp100_inw(PAGING) & HP100_CHIPID_MASK;
-#ifdef HP100_DEBUG
-	if (chip == HP100_CHIPID_SHASTA)
-		printk("hp100: %s: Shasta Chip detected. (This is a pre 802.12 chip)\n", dev->name);
-	else if (chip == HP100_CHIPID_RAINIER)
-		printk("hp100: %s: Rainier Chip detected. (This is a pre 802.12 chip)\n", dev->name);
-	else if (chip == HP100_CHIPID_LASSEN)
-		printk("hp100: %s: Lassen Chip detected.\n", dev->name);
-	else
-		printk("hp100: %s: Warning: Unknown CASCADE chip (id=0x%.4x).\n", dev->name, chip);
-#endif
-
-	dev->base_addr = ioaddr;
-
-	eid = hp100_read_id(ioaddr);
-	if (eid == NULL) {	/* bad checksum? */
-		printk(KERN_WARNING "%s: bad ID checksum at base port 0x%x\n",
-		       __func__, ioaddr);
-		goto out2;
-	}
-
-	hp100_page(ID_MAC_ADDR);
-	for (i = uc = 0; i < 7; i++)
-		uc += hp100_inb(LAN_ADDR + i);
-	if (uc != 0xff) {
-		printk(KERN_WARNING
-		       "%s: bad lan address checksum at port 0x%x)\n",
-		       __func__, ioaddr);
-		err = -EIO;
-		goto out2;
-	}
-
-	/* Make sure, that all registers are correctly updated... */
-
-	hp100_load_eeprom(dev, ioaddr);
-	wait();
-
-	/*
-	 * Determine driver operation mode
-	 *
-	 * Use the variable "hp100_mode" upon insmod or as kernel parameter to
-	 * force driver modes:
-	 * hp100_mode=1 -> default, use busmaster mode if configured.
-	 * hp100_mode=2 -> enable shared memory mode
-	 * hp100_mode=3 -> force use of i/o mapped mode.
-	 * hp100_mode=4 -> same as 1, but re-set the enable bit on the card.
-	 */
-
-	/*
-	 * LSW values:
-	 *   0x2278 -> J2585B, PnP shared memory mode
-	 *   0x2270 -> J2585B, shared memory mode, 0xdc000
-	 *   0xa23c -> J2585B, I/O mapped mode
-	 *   0x2240 -> EISA COMPEX, BusMaster (Shasta Chip)
-	 *   0x2220 -> EISA HP, I/O (Shasta Chip)
-	 *   0x2260 -> EISA HP, BusMaster (Shasta Chip)
-	 */
-
-#if 0
-	local_mode = 0x2270;
-	hp100_outw(0xfefe, OPTION_LSW);
-	hp100_outw(local_mode | HP100_SET_LB | HP100_SET_HB, OPTION_LSW);
-#endif
-
-	/* hp100_mode value maybe used in future by another card */
-	local_mode = hp100_mode;
-	if (local_mode < 1 || local_mode > 4)
-		local_mode = 1;	/* default */
-#ifdef HP100_DEBUG
-	printk("hp100: %s: original LSW = 0x%x\n", dev->name,
-	       hp100_inw(OPTION_LSW));
-#endif
-
-	if (local_mode == 3) {
-		hp100_outw(HP100_MEM_EN | HP100_RESET_LB, OPTION_LSW);
-		hp100_outw(HP100_IO_EN | HP100_SET_LB, OPTION_LSW);
-		hp100_outw(HP100_BM_WRITE | HP100_BM_READ | HP100_RESET_HB, OPTION_LSW);
-		printk("hp100: IO mapped mode forced.\n");
-	} else if (local_mode == 2) {
-		hp100_outw(HP100_MEM_EN | HP100_SET_LB, OPTION_LSW);
-		hp100_outw(HP100_IO_EN | HP100_SET_LB, OPTION_LSW);
-		hp100_outw(HP100_BM_WRITE | HP100_BM_READ | HP100_RESET_HB, OPTION_LSW);
-		printk("hp100: Shared memory mode requested.\n");
-	} else if (local_mode == 4) {
-		if (chip == HP100_CHIPID_LASSEN) {
-			hp100_outw(HP100_BM_WRITE | HP100_BM_READ | HP100_SET_HB, OPTION_LSW);
-			hp100_outw(HP100_IO_EN | HP100_MEM_EN | HP100_RESET_LB, OPTION_LSW);
-			printk("hp100: Busmaster mode requested.\n");
-		}
-		local_mode = 1;
-	}
-
-	if (local_mode == 1) {	/* default behaviour */
-		lsw = hp100_inw(OPTION_LSW);
-
-		if ((lsw & HP100_IO_EN) && (~lsw & HP100_MEM_EN) &&
-		    (~lsw & (HP100_BM_WRITE | HP100_BM_READ))) {
-#ifdef HP100_DEBUG
-			printk("hp100: %s: IO_EN bit is set on card.\n", dev->name);
-#endif
-			local_mode = 3;
-		} else if (chip == HP100_CHIPID_LASSEN &&
-			   (lsw & (HP100_BM_WRITE | HP100_BM_READ)) == (HP100_BM_WRITE | HP100_BM_READ)) {
-			/* Conversion to new PCI API :
-			 * I don't have the doc, but I assume that the card
-			 * can map the full 32bit address space.
-			 * Also, we can have EISA Busmaster cards (not tested),
-			 * so beware !!! - Jean II */
-			if((bus == HP100_BUS_PCI) &&
-			   (pci_set_dma_mask(pci_dev, DMA_BIT_MASK(32)))) {
-				/* Gracefully fallback to shared memory */
-				goto busmasterfail;
-			}
-			printk("hp100: Busmaster mode enabled.\n");
-			hp100_outw(HP100_MEM_EN | HP100_IO_EN | HP100_RESET_LB, OPTION_LSW);
-		} else {
-		busmasterfail:
-#ifdef HP100_DEBUG
-			printk("hp100: %s: Card not configured for BM or BM not supported with this card.\n", dev->name);
-			printk("hp100: %s: Trying shared memory mode.\n", dev->name);
-#endif
-			/* In this case, try shared memory mode */
-			local_mode = 2;
-			hp100_outw(HP100_MEM_EN | HP100_SET_LB, OPTION_LSW);
-			/* hp100_outw(HP100_IO_EN|HP100_RESET_LB, OPTION_LSW); */
-		}
-	}
-#ifdef HP100_DEBUG
-	printk("hp100: %s: new LSW = 0x%x\n", dev->name, hp100_inw(OPTION_LSW));
-#endif
-
-	/* Check for shared memory on the card, eventually remap it */
-	hp100_page(HW_MAP);
-	mem_mapped = ((hp100_inw(OPTION_LSW) & (HP100_MEM_EN)) != 0);
-	mem_ptr_phys = 0UL;
-	mem_ptr_virt = NULL;
-	memory_size = (8192 << ((hp100_inb(SRAM) >> 5) & 0x07));
-	virt_memory_size = 0;
-
-	/* For memory mapped or busmaster mode, we want the memory address */
-	if (mem_mapped || (local_mode == 1)) {
-		mem_ptr_phys = (hp100_inw(MEM_MAP_LSW) | (hp100_inw(MEM_MAP_MSW) << 16));
-		mem_ptr_phys &= ~0x1fff;	/* 8k alignment */
-
-		if (bus == HP100_BUS_ISA && (mem_ptr_phys & ~0xfffff) != 0) {
-			printk("hp100: Can only use programmed i/o mode.\n");
-			mem_ptr_phys = 0;
-			mem_mapped = 0;
-			local_mode = 3;	/* Use programmed i/o */
-		}
-
-		/* We do not need access to shared memory in busmaster mode */
-		/* However in slave mode we need to remap high (>1GB) card memory  */
-		if (local_mode != 1) {	/* = not busmaster */
-			/* We try with smaller memory sizes, if ioremap fails */
-			for (virt_memory_size = memory_size; virt_memory_size > 16383; virt_memory_size >>= 1) {
-				if ((mem_ptr_virt = ioremap((u_long) mem_ptr_phys, virt_memory_size)) == NULL) {
-#ifdef HP100_DEBUG
-					printk("hp100: %s: ioremap for 0x%x bytes high PCI memory at 0x%lx failed\n", dev->name, virt_memory_size, mem_ptr_phys);
-#endif
-				} else {
-#ifdef HP100_DEBUG
-					printk("hp100: %s: remapped 0x%x bytes high PCI memory at 0x%lx to %p.\n", dev->name, virt_memory_size, mem_ptr_phys, mem_ptr_virt);
-#endif
-					break;
-				}
-			}
-
-			if (mem_ptr_virt == NULL) {	/* all ioremap tries failed */
-				printk("hp100: Failed to ioremap the PCI card memory. Will have to use i/o mapped mode.\n");
-				local_mode = 3;
-				virt_memory_size = 0;
-			}
-		}
-	}
-
-	if (local_mode == 3) {	/* io mapped forced */
-		mem_mapped = 0;
-		mem_ptr_phys = 0;
-		mem_ptr_virt = NULL;
-		printk("hp100: Using (slow) programmed i/o mode.\n");
-	}
-
-	/* Initialise the "private" data structure for this card. */
-	lp = netdev_priv(dev);
-
-	spin_lock_init(&lp->lock);
-	strlcpy(lp->id, eid, HP100_SIG_LEN);
-	lp->chip = chip;
-	lp->mode = local_mode;
-	lp->bus = bus;
-	lp->pci_dev = pci_dev;
-	lp->priority_tx = hp100_priority_tx;
-	lp->rx_ratio = hp100_rx_ratio;
-	lp->mem_ptr_phys = mem_ptr_phys;
-	lp->mem_ptr_virt = mem_ptr_virt;
-	hp100_page(ID_MAC_ADDR);
-	lp->soft_model = hp100_inb(SOFT_MODEL);
-	lp->mac1_mode = HP100_MAC1MODE3;
-	lp->mac2_mode = HP100_MAC2MODE3;
-	memset(&lp->hash_bytes, 0x00, 8);
-
-	dev->base_addr = ioaddr;
-
-	lp->memory_size = memory_size;
-	lp->virt_memory_size = virt_memory_size;
-	lp->rx_ratio = hp100_rx_ratio;	/* can be conf'd with insmod */
-
-	if (lp->mode == 1)	/* busmaster */
-		dev->netdev_ops = &hp100_bm_netdev_ops;
-	else
-		dev->netdev_ops = &hp100_netdev_ops;
-
-	/* Ask the card for which IRQ line it is configured */
-	if (bus == HP100_BUS_PCI) {
-		dev->irq = pci_dev->irq;
-	} else {
-		hp100_page(HW_MAP);
-		dev->irq = hp100_inb(IRQ_CHANNEL) & HP100_IRQMASK;
-		if (dev->irq == 2)
-			dev->irq = 9;
-	}
-
-	if (lp->mode == 1)	/* busmaster */
-		dev->dma = 4;
-
-	/* Ask the card for its MAC address and store it for later use. */
-	hp100_page(ID_MAC_ADDR);
-	for (i = uc = 0; i < 6; i++)
-		dev->dev_addr[i] = hp100_inb(LAN_ADDR + i);
-
-	/* Reset statistics (counters) */
-	hp100_clear_stats(lp, ioaddr);
-
-	/* If busmaster mode is wanted, a dma-capable memory area is needed for
-	 * the rx and tx PDLs
-	 * PCI cards can access the whole PC memory. Therefore GFP_DMA is not
-	 * needed for the allocation of the memory area.
-	 */
-
-	/* TODO: We do not need this with old cards, where PDLs are stored
-	 * in the cards shared memory area. But currently, busmaster has been
-	 * implemented/tested only with the lassen chip anyway... */
-	if (lp->mode == 1) {	/* busmaster */
-		dma_addr_t page_baddr;
-		/* Get physically continuous memory for TX & RX PDLs    */
-		/* Conversion to new PCI API :
-		 * Pages are always aligned and zeroed, no need to it ourself.
-		 * Doc says should be OK for EISA bus as well - Jean II */
-		lp->page_vaddr_algn = pci_alloc_consistent(lp->pci_dev, MAX_RINGSIZE, &page_baddr);
-		if (!lp->page_vaddr_algn) {
-			err = -ENOMEM;
-			goto out_mem_ptr;
-		}
-		lp->whatever_offset = ((u_long) page_baddr) - ((u_long) lp->page_vaddr_algn);
-
-#ifdef HP100_DEBUG_BM
-		printk("hp100: %s: Reserved DMA memory from 0x%x to 0x%x\n", dev->name, (u_int) lp->page_vaddr_algn, (u_int) lp->page_vaddr_algn + MAX_RINGSIZE);
-#endif
-		lp->rxrcommit = lp->txrcommit = 0;
-		lp->rxrhead = lp->rxrtail = &(lp->rxring[0]);
-		lp->txrhead = lp->txrtail = &(lp->txring[0]);
-	}
-
-	/* Initialise the card. */
-	/* (I'm not really sure if it's a good idea to do this during probing, but
-	 * like this it's assured that the lan connection type can be sensed
-	 * correctly)
-	 */
-	hp100_hwinit(dev);
-
-	/* Try to find out which kind of LAN the card is connected to. */
-	lp->lan_type = hp100_sense_lan(dev);
-
-	/* Print out a message what about what we think we have probed. */
-	printk("hp100: at 0x%x, IRQ %d, ", ioaddr, dev->irq);
-	switch (bus) {
-	case HP100_BUS_EISA:
-		printk("EISA");
-		break;
-	case HP100_BUS_PCI:
-		printk("PCI");
-		break;
-	default:
-		printk("ISA");
-		break;
-	}
-	printk(" bus, %dk SRAM (rx/tx %d%%).\n", lp->memory_size >> 10, lp->rx_ratio);
-
-	if (lp->mode == 2) {	/* memory mapped */
-		printk("hp100: Memory area at 0x%lx-0x%lx", mem_ptr_phys,
-				(mem_ptr_phys + (mem_ptr_phys > 0x100000 ? (u_long) lp->memory_size : 16 * 1024)) - 1);
-		if (mem_ptr_virt)
-			printk(" (virtual base %p)", mem_ptr_virt);
-		printk(".\n");
-
-		/* Set for info when doing ifconfig */
-		dev->mem_start = mem_ptr_phys;
-		dev->mem_end = mem_ptr_phys + lp->memory_size;
-	}
-
-	printk("hp100: ");
-	if (lp->lan_type != HP100_LAN_ERR)
-		printk("Adapter is attached to ");
-	switch (lp->lan_type) {
-	case HP100_LAN_100:
-		printk("100Mb/s Voice Grade AnyLAN network.\n");
-		break;
-	case HP100_LAN_10:
-		printk("10Mb/s network (10baseT).\n");
-		break;
-	case HP100_LAN_COAX:
-		printk("10Mb/s network (coax).\n");
-		break;
-	default:
-		printk("Warning! Link down.\n");
-	}
-
-	err = register_netdev(dev);
-	if (err)
-		goto out3;
-
-	return 0;
-out3:
-	if (local_mode == 1)
-		pci_free_consistent(lp->pci_dev, MAX_RINGSIZE + 0x0f,
-				    lp->page_vaddr_algn,
-				    virt_to_whatever(dev, lp->page_vaddr_algn));
-out_mem_ptr:
-	if (mem_ptr_virt)
-		iounmap(mem_ptr_virt);
-out2:
-	release_region(ioaddr, HP100_REGION_SIZE);
-out1:
-	return err;
-}
-
-/* This procedure puts the card into a stable init state */
-static void hp100_hwinit(struct net_device *dev)
-{
-	int ioaddr = dev->base_addr;
-	struct hp100_private *lp = netdev_priv(dev);
-
-#ifdef HP100_DEBUG_B
-	hp100_outw(0x4202, TRACE);
-	printk("hp100: %s: hwinit\n", dev->name);
-#endif
-
-	/* Initialise the card. -------------------------------------------- */
-
-	/* Clear all pending Ints and disable Ints */
-	hp100_page(PERFORMANCE);
-	hp100_outw(0xfefe, IRQ_MASK);	/* mask off all ints */
-	hp100_outw(0xffff, IRQ_STATUS);	/* clear all pending ints */
-
-	hp100_outw(HP100_INT_EN | HP100_RESET_LB, OPTION_LSW);
-	hp100_outw(HP100_TRI_INT | HP100_SET_HB, OPTION_LSW);
-
-	if (lp->mode == 1) {
-		hp100_BM_shutdown(dev);	/* disables BM, puts cascade in reset */
-		wait();
-	} else {
-		hp100_outw(HP100_INT_EN | HP100_RESET_LB, OPTION_LSW);
-		hp100_cascade_reset(dev, 1);
-		hp100_page(MAC_CTRL);
-		hp100_andb(~(HP100_RX_EN | HP100_TX_EN), MAC_CFG_1);
-	}
-
-	/* Initiate EEPROM reload */
-	hp100_load_eeprom(dev, 0);
-
-	wait();
-
-	/* Go into reset again. */
-	hp100_cascade_reset(dev, 1);
-
-	/* Set Option Registers to a safe state  */
-	hp100_outw(HP100_DEBUG_EN |
-		   HP100_RX_HDR |
-		   HP100_EE_EN |
-		   HP100_BM_WRITE |
-		   HP100_BM_READ | HP100_RESET_HB |
-		   HP100_FAKE_INT |
-		   HP100_INT_EN |
-		   HP100_MEM_EN |
-		   HP100_IO_EN | HP100_RESET_LB, OPTION_LSW);
-
-	hp100_outw(HP100_TRI_INT |
-		   HP100_MMAP_DIS | HP100_SET_HB, OPTION_LSW);
-
-	hp100_outb(HP100_PRIORITY_TX |
-		   HP100_ADV_NXT_PKT |
-		   HP100_TX_CMD | HP100_RESET_LB, OPTION_MSW);
-
-	/* TODO: Configure MMU for Ram Test. */
-	/* TODO: Ram Test. */
-
-	/* Re-check if adapter is still at same i/o location      */
-	/* (If the base i/o in eeprom has been changed but the    */
-	/* registers had not been changed, a reload of the eeprom */
-	/* would move the adapter to the address stored in eeprom */
-
-	/* TODO: Code to implement. */
-
-	/* Until here it was code from HWdiscover procedure. */
-	/* Next comes code from mmuinit procedure of SCO BM driver which is
-	 * called from HWconfigure in the SCO driver.  */
-
-	/* Initialise MMU, eventually switch on Busmaster Mode, initialise
-	 * multicast filter...
-	 */
-	hp100_mmuinit(dev);
-
-	/* We don't turn the interrupts on here - this is done by start_interface. */
-	wait();			/* TODO: Do we really need this? */
-
-	/* Enable Hardware (e.g. unreset) */
-	hp100_cascade_reset(dev, 0);
-
-	/* ------- initialisation complete ----------- */
-
-	/* Finally try to log in the Hub if there may be a VG connection. */
-	if ((lp->lan_type == HP100_LAN_100) || (lp->lan_type == HP100_LAN_ERR))
-		hp100_login_to_vg_hub(dev, 0);	/* relogin */
-
-}
-
-
-/*
- * mmuinit - Reinitialise Cascade MMU and MAC settings.
- * Note: Must already be in reset and leaves card in reset.
- */
-static void hp100_mmuinit(struct net_device *dev)
-{
-	int ioaddr = dev->base_addr;
-	struct hp100_private *lp = netdev_priv(dev);
-	int i;
-
-#ifdef HP100_DEBUG_B
-	hp100_outw(0x4203, TRACE);
-	printk("hp100: %s: mmuinit\n", dev->name);
-#endif
-
-#ifdef HP100_DEBUG
-	if (0 != (hp100_inw(OPTION_LSW) & HP100_HW_RST)) {
-		printk("hp100: %s: Not in reset when entering mmuinit. Fix me.\n", dev->name);
-		return;
-	}
-#endif
-
-	/* Make sure IRQs are masked off and ack'ed. */
-	hp100_page(PERFORMANCE);
-	hp100_outw(0xfefe, IRQ_MASK);	/* mask off all ints */
-	hp100_outw(0xffff, IRQ_STATUS);	/* ack IRQ */
-
-	/*
-	 * Enable Hardware
-	 * - Clear Debug En, Rx Hdr Pipe, EE En, I/O En, Fake Int and Intr En
-	 * - Set Tri-State Int, Bus Master Rd/Wr, and Mem Map Disable
-	 * - Clear Priority, Advance Pkt and Xmit Cmd
-	 */
-
-	hp100_outw(HP100_DEBUG_EN |
-		   HP100_RX_HDR |
-		   HP100_EE_EN | HP100_RESET_HB |
-		   HP100_IO_EN |
-		   HP100_FAKE_INT |
-		   HP100_INT_EN | HP100_RESET_LB, OPTION_LSW);
-
-	hp100_outw(HP100_TRI_INT | HP100_SET_HB, OPTION_LSW);
-
-	if (lp->mode == 1) {	/* busmaster */
-		hp100_outw(HP100_BM_WRITE |
-			   HP100_BM_READ |
-			   HP100_MMAP_DIS | HP100_SET_HB, OPTION_LSW);
-	} else if (lp->mode == 2) {	/* memory mapped */
-		hp100_outw(HP100_BM_WRITE |
-			   HP100_BM_READ | HP100_RESET_HB, OPTION_LSW);
-		hp100_outw(HP100_MMAP_DIS | HP100_RESET_HB, OPTION_LSW);
-		hp100_outw(HP100_MEM_EN | HP100_SET_LB, OPTION_LSW);
-		hp100_outw(HP100_IO_EN | HP100_SET_LB, OPTION_LSW);
-	} else if (lp->mode == 3) {	/* i/o mapped mode */
-		hp100_outw(HP100_MMAP_DIS | HP100_SET_HB |
-			   HP100_IO_EN | HP100_SET_LB, OPTION_LSW);
-	}
-
-	hp100_page(HW_MAP);
-	hp100_outb(0, EARLYRXCFG);
-	hp100_outw(0, EARLYTXCFG);
-
-	/*
-	 * Enable Bus Master mode
-	 */
-	if (lp->mode == 1) {	/* busmaster */
-		/* Experimental: Set some PCI configuration bits */
-		hp100_page(HW_MAP);
-		hp100_andb(~HP100_PDL_USE3, MODECTRL1);	/* BM engine read maximum */
-		hp100_andb(~HP100_TX_DUALQ, MODECTRL1);	/* No Queue for Priority TX */
-
-		/* PCI Bus failures should result in a Misc. Interrupt */
-		hp100_orb(HP100_EN_BUS_FAIL, MODECTRL2);
-
-		hp100_outw(HP100_BM_READ | HP100_BM_WRITE | HP100_SET_HB, OPTION_LSW);
-		hp100_page(HW_MAP);
-		/* Use Burst Mode and switch on PAGE_CK */
-		hp100_orb(HP100_BM_BURST_RD | HP100_BM_BURST_WR, BM);
-		if ((lp->chip == HP100_CHIPID_RAINIER) || (lp->chip == HP100_CHIPID_SHASTA))
-			hp100_orb(HP100_BM_PAGE_CK, BM);
-		hp100_orb(HP100_BM_MASTER, BM);
-	} else {		/* not busmaster */
-
-		hp100_page(HW_MAP);
-		hp100_andb(~HP100_BM_MASTER, BM);
-	}
-
-	/*
-	 * Divide card memory into regions for Rx, Tx and, if non-ETR chip, PDLs
-	 */
-	hp100_page(MMU_CFG);
-	if (lp->mode == 1) {	/* only needed for Busmaster */
-		int xmit_stop, recv_stop;
-
-		if ((lp->chip == HP100_CHIPID_RAINIER) ||
-		    (lp->chip == HP100_CHIPID_SHASTA)) {
-			int pdl_stop;
-
-			/*
-			 * Each pdl is 508 bytes long. (63 frags * 4 bytes for address and
-			 * 4 bytes for header). We will leave NUM_RXPDLS * 508 (rounded
-			 * to the next higher 1k boundary) bytes for the rx-pdl's
-			 * Note: For non-etr chips the transmit stop register must be
-			 * programmed on a 1k boundary, i.e. bits 9:0 must be zero.
-			 */
-			pdl_stop = lp->memory_size;
-			xmit_stop = (pdl_stop - 508 * (MAX_RX_PDL) - 16) & ~(0x03ff);
-			recv_stop = (xmit_stop * (lp->rx_ratio) / 100) & ~(0x03ff);
-			hp100_outw((pdl_stop >> 4) - 1, PDL_MEM_STOP);
-#ifdef HP100_DEBUG_BM
-			printk("hp100: %s: PDL_STOP = 0x%x\n", dev->name, pdl_stop);
-#endif
-		} else {
-			/* ETR chip (Lassen) in busmaster mode */
-			xmit_stop = (lp->memory_size) - 1;
-			recv_stop = ((lp->memory_size * lp->rx_ratio) / 100) & ~(0x03ff);
-		}
-
-		hp100_outw(xmit_stop >> 4, TX_MEM_STOP);
-		hp100_outw(recv_stop >> 4, RX_MEM_STOP);
-#ifdef HP100_DEBUG_BM
-		printk("hp100: %s: TX_STOP  = 0x%x\n", dev->name, xmit_stop >> 4);
-		printk("hp100: %s: RX_STOP  = 0x%x\n", dev->name, recv_stop >> 4);
-#endif
-	} else {
-		/* Slave modes (memory mapped and programmed io)  */
-		hp100_outw((((lp->memory_size * lp->rx_ratio) / 100) >> 4), RX_MEM_STOP);
-		hp100_outw(((lp->memory_size - 1) >> 4), TX_MEM_STOP);
-#ifdef HP100_DEBUG
-		printk("hp100: %s: TX_MEM_STOP: 0x%x\n", dev->name, hp100_inw(TX_MEM_STOP));
-		printk("hp100: %s: RX_MEM_STOP: 0x%x\n", dev->name, hp100_inw(RX_MEM_STOP));
-#endif
-	}
-
-	/* Write MAC address into page 1 */
-	hp100_page(MAC_ADDRESS);
-	for (i = 0; i < 6; i++)
-		hp100_outb(dev->dev_addr[i], MAC_ADDR + i);
-
-	/* Zero the multicast hash registers */
-	for (i = 0; i < 8; i++)
-		hp100_outb(0x0, HASH_BYTE0 + i);
-
-	/* Set up MAC defaults */
-	hp100_page(MAC_CTRL);
-
-	/* Go to LAN Page and zero all filter bits */
-	/* Zero accept error, accept multicast, accept broadcast and accept */
-	/* all directed packet bits */
-	hp100_andb(~(HP100_RX_EN |
-		     HP100_TX_EN |
-		     HP100_ACC_ERRORED |
-		     HP100_ACC_MC |
-		     HP100_ACC_BC | HP100_ACC_PHY), MAC_CFG_1);
-
-	hp100_outb(0x00, MAC_CFG_2);
-
-	/* Zero the frame format bit. This works around a training bug in the */
-	/* new hubs. */
-	hp100_outb(0x00, VG_LAN_CFG_2);	/* (use 802.3) */
-
-	if (lp->priority_tx)
-		hp100_outb(HP100_PRIORITY_TX | HP100_SET_LB, OPTION_MSW);
-	else
-		hp100_outb(HP100_PRIORITY_TX | HP100_RESET_LB, OPTION_MSW);
-
-	hp100_outb(HP100_ADV_NXT_PKT |
-		   HP100_TX_CMD | HP100_RESET_LB, OPTION_MSW);
-
-	/* If busmaster, initialize the PDLs */
-	if (lp->mode == 1)
-		hp100_init_pdls(dev);
-
-	/* Go to performance page and initialize isr and imr registers */
-	hp100_page(PERFORMANCE);
-	hp100_outw(0xfefe, IRQ_MASK);	/* mask off all ints */
-	hp100_outw(0xffff, IRQ_STATUS);	/* ack IRQ */
-}
-
-/*
- *  open/close functions
- */
-
-static int hp100_open(struct net_device *dev)
-{
-	struct hp100_private *lp = netdev_priv(dev);
-#ifdef HP100_DEBUG_B
-	int ioaddr = dev->base_addr;
-#endif
-
-#ifdef HP100_DEBUG_B
-	hp100_outw(0x4204, TRACE);
-	printk("hp100: %s: open\n", dev->name);
-#endif
-
-	/* New: if bus is PCI or EISA, interrupts might be shared interrupts */
-	if (request_irq(dev->irq, hp100_interrupt,
-			lp->bus == HP100_BUS_PCI || lp->bus ==
-			HP100_BUS_EISA ? IRQF_SHARED : 0,
-			dev->name, dev)) {
-		printk("hp100: %s: unable to get IRQ %d\n", dev->name, dev->irq);
-		return -EAGAIN;
-	}
-
-	netif_trans_update(dev); /* prevent tx timeout */
-	netif_start_queue(dev);
-
-	lp->lan_type = hp100_sense_lan(dev);
-	lp->mac1_mode = HP100_MAC1MODE3;
-	lp->mac2_mode = HP100_MAC2MODE3;
-	memset(&lp->hash_bytes, 0x00, 8);
-
-	hp100_stop_interface(dev);
-
-	hp100_hwinit(dev);
-
-	hp100_start_interface(dev);	/* sets mac modes, enables interrupts */
-
-	return 0;
-}
-
-/* The close function is called when the interface is to be brought down */
-static int hp100_close(struct net_device *dev)
-{
-	int ioaddr = dev->base_addr;
-	struct hp100_private *lp = netdev_priv(dev);
-
-#ifdef HP100_DEBUG_B
-	hp100_outw(0x4205, TRACE);
-	printk("hp100: %s: close\n", dev->name);
-#endif
-
-	hp100_page(PERFORMANCE);
-	hp100_outw(0xfefe, IRQ_MASK);	/* mask off all IRQs */
-
-	hp100_stop_interface(dev);
-
-	if (lp->lan_type == HP100_LAN_100)
-		lp->hub_status = hp100_login_to_vg_hub(dev, 0);
-
-	netif_stop_queue(dev);
-
-	free_irq(dev->irq, dev);
-
-#ifdef HP100_DEBUG
-	printk("hp100: %s: close LSW = 0x%x\n", dev->name,
-	       hp100_inw(OPTION_LSW));
-#endif
-
-	return 0;
-}
-
-
-/*
- * Configure the PDL Rx rings and LAN
- */
-static void hp100_init_pdls(struct net_device *dev)
-{
-	struct hp100_private *lp = netdev_priv(dev);
-	hp100_ring_t *ringptr;
-	u_int *pageptr;		/* Warning : increment by 4 - Jean II */
-	int i;
-
-#ifdef HP100_DEBUG_B
-	int ioaddr = dev->base_addr;
-#endif
-
-#ifdef HP100_DEBUG_B
-	hp100_outw(0x4206, TRACE);
-	printk("hp100: %s: init pdls\n", dev->name);
-#endif
-
-	if (!lp->page_vaddr_algn)
-		printk("hp100: %s: Warning: lp->page_vaddr_algn not initialised!\n", dev->name);
-	else {
-		/* pageptr shall point into the DMA accessible memory region  */
-		/* we use this pointer to status the upper limit of allocated */
-		/* memory in the allocated page. */
-		/* note: align the pointers to the pci cache line size */
-		memset(lp->page_vaddr_algn, 0, MAX_RINGSIZE);	/* Zero  Rx/Tx ring page */
-		pageptr = lp->page_vaddr_algn;
-
-		lp->rxrcommit = 0;
-		ringptr = lp->rxrhead = lp->rxrtail = &(lp->rxring[0]);
-
-		/* Initialise Rx Ring */
-		for (i = MAX_RX_PDL - 1; i >= 0; i--) {
-			lp->rxring[i].next = ringptr;
-			ringptr = &(lp->rxring[i]);
-			pageptr += hp100_init_rxpdl(dev, ringptr, pageptr);
-		}
-
-		/* Initialise Tx Ring */
-		lp->txrcommit = 0;
-		ringptr = lp->txrhead = lp->txrtail = &(lp->txring[0]);
-		for (i = MAX_TX_PDL - 1; i >= 0; i--) {
-			lp->txring[i].next = ringptr;
-			ringptr = &(lp->txring[i]);
-			pageptr += hp100_init_txpdl(dev, ringptr, pageptr);
-		}
-	}
-}
-
-
-/* These functions "format" the entries in the pdl structure   */
-/* They return how much memory the fragments need.            */
-static int hp100_init_rxpdl(struct net_device *dev,
-			    register hp100_ring_t * ringptr,
-			    register u32 * pdlptr)
-{
-	/* pdlptr is starting address for this pdl */
-
-	if (0 != (((unsigned long) pdlptr) & 0xf))
-		printk("hp100: %s: Init rxpdl: Unaligned pdlptr 0x%lx.\n",
-		       dev->name, (unsigned long) pdlptr);
-
-	ringptr->pdl = pdlptr + 1;
-	ringptr->pdl_paddr = virt_to_whatever(dev, pdlptr + 1);
-	ringptr->skb = NULL;
-
-	/*
-	 * Write address and length of first PDL Fragment (which is used for
-	 * storing the RX-Header
-	 * We use the 4 bytes _before_ the PDH in the pdl memory area to
-	 * store this information. (PDH is at offset 0x04)
-	 */
-	/* Note that pdlptr+1 and not pdlptr is the pointer to the PDH */
-
-	*(pdlptr + 2) = (u_int) virt_to_whatever(dev, pdlptr);	/* Address Frag 1 */
-	*(pdlptr + 3) = 4;	/* Length  Frag 1 */
-
-	return roundup(MAX_RX_FRAG * 2 + 2, 4);
-}
-
-
-static int hp100_init_txpdl(struct net_device *dev,
-			    register hp100_ring_t * ringptr,
-			    register u32 * pdlptr)
-{
-	if (0 != (((unsigned long) pdlptr) & 0xf))
-		printk("hp100: %s: Init txpdl: Unaligned pdlptr 0x%lx.\n", dev->name, (unsigned long) pdlptr);
-
-	ringptr->pdl = pdlptr;	/* +1; */
-	ringptr->pdl_paddr = virt_to_whatever(dev, pdlptr);	/* +1 */
-	ringptr->skb = NULL;
-
-	return roundup(MAX_TX_FRAG * 2 + 2, 4);
-}
-
-/*
- * hp100_build_rx_pdl allocates an skb_buff of maximum size plus two bytes
- * for possible odd word alignment rounding up to next dword and set PDL
- * address for fragment#2
- * Returns: 0 if unable to allocate skb_buff
- *          1 if successful
- */
-static int hp100_build_rx_pdl(hp100_ring_t * ringptr,
-			      struct net_device *dev)
-{
-#ifdef HP100_DEBUG_B
-	int ioaddr = dev->base_addr;
-#endif
-#ifdef HP100_DEBUG_BM
-	u_int *p;
-#endif
-
-#ifdef HP100_DEBUG_B
-	hp100_outw(0x4207, TRACE);
-	printk("hp100: %s: build rx pdl\n", dev->name);
-#endif
-
-	/* Allocate skb buffer of maximum size */
-	/* Note: This depends on the alloc_skb functions allocating more
-	 * space than requested, i.e. aligning to 16bytes */
-
-	ringptr->skb = netdev_alloc_skb(dev, roundup(MAX_ETHER_SIZE + 2, 4));
-
-	if (NULL != ringptr->skb) {
-		/*
-		 * Reserve 2 bytes at the head of the buffer to land the IP header
-		 * on a long word boundary (According to the Network Driver section
-		 * in the Linux KHG, this should help to increase performance.)
-		 */
-		skb_reserve(ringptr->skb, 2);
-
-		ringptr->skb->data = (u_char *) skb_put(ringptr->skb, MAX_ETHER_SIZE);
-
-		/* ringptr->pdl points to the beginning of the PDL, i.e. the PDH */
-		/* Note: 1st Fragment is used for the 4 byte packet status
-		 * (receive header). Its PDL entries are set up by init_rxpdl. So
-		 * here we only have to set up the PDL fragment entries for the data
-		 * part. Those 4 bytes will be stored in the DMA memory region
-		 * directly before the PDL.
-		 */
-#ifdef HP100_DEBUG_BM
-		printk("hp100: %s: build_rx_pdl: PDH@0x%x, skb->data (len %d) at 0x%x\n",
-				     dev->name, (u_int) ringptr->pdl,
-				     roundup(MAX_ETHER_SIZE + 2, 4),
-				     (unsigned int) ringptr->skb->data);
-#endif
-
-		/* Conversion to new PCI API : map skbuf data to PCI bus.
-		 * Doc says it's OK for EISA as well - Jean II */
-		ringptr->pdl[0] = 0x00020000;	/* Write PDH */
-		ringptr->pdl[3] = pdl_map_data(netdev_priv(dev),
-					       ringptr->skb->data);
-		ringptr->pdl[4] = MAX_ETHER_SIZE;	/* Length of Data */
-
-#ifdef HP100_DEBUG_BM
-		for (p = (ringptr->pdl); p < (ringptr->pdl + 5); p++)
-			printk("hp100: %s: Adr 0x%.8x = 0x%.8x\n", dev->name, (u_int) p, (u_int) * p);
-#endif
-		return 1;
-	}
-	/* else: */
-	/* alloc_skb failed (no memory) -> still can receive the header
-	 * fragment into PDL memory. make PDL safe by clearing msgptr and
-	 * making the PDL only 1 fragment (i.e. the 4 byte packet status)
-	 */
-#ifdef HP100_DEBUG_BM
-	printk("hp100: %s: build_rx_pdl: PDH@0x%x, No space for skb.\n", dev->name, (u_int) ringptr->pdl);
-#endif
-
-	ringptr->pdl[0] = 0x00010000;	/* PDH: Count=1 Fragment */
-
-	return 0;
-}
-
-/*
- *  hp100_rxfill - attempt to fill the Rx Ring will empty skb's
- *
- * Makes assumption that skb's are always contiguous memory areas and
- * therefore PDLs contain only 2 physical fragments.
- * -  While the number of Rx PDLs with buffers is less than maximum
- *      a.  Get a maximum packet size skb
- *      b.  Put the physical address of the buffer into the PDL.
- *      c.  Output physical address of PDL to adapter.
- */
-static void hp100_rxfill(struct net_device *dev)
-{
-	int ioaddr = dev->base_addr;
-
-	struct hp100_private *lp = netdev_priv(dev);
-	hp100_ring_t *ringptr;
-
-#ifdef HP100_DEBUG_B
-	hp100_outw(0x4208, TRACE);
-	printk("hp100: %s: rxfill\n", dev->name);
-#endif
-
-	hp100_page(PERFORMANCE);
-
-	while (lp->rxrcommit < MAX_RX_PDL) {
-		/*
-		   ** Attempt to get a buffer and build a Rx PDL.
-		 */
-		ringptr = lp->rxrtail;
-		if (0 == hp100_build_rx_pdl(ringptr, dev)) {
-			return;	/* None available, return */
-		}
-
-		/* Hand this PDL over to the card */
-		/* Note: This needs performance page selected! */
-#ifdef HP100_DEBUG_BM
-		printk("hp100: %s: rxfill: Hand to card: pdl #%d @0x%x phys:0x%x, buffer: 0x%x\n",
-				     dev->name, lp->rxrcommit, (u_int) ringptr->pdl,
-				     (u_int) ringptr->pdl_paddr, (u_int) ringptr->pdl[3]);
-#endif
-
-		hp100_outl((u32) ringptr->pdl_paddr, RX_PDA);
-
-		lp->rxrcommit += 1;
-		lp->rxrtail = ringptr->next;
-	}
-}
-
-/*
- * BM_shutdown - shutdown bus mastering and leave chip in reset state
- */
-
-static void hp100_BM_shutdown(struct net_device *dev)
-{
-	int ioaddr = dev->base_addr;
-	struct hp100_private *lp = netdev_priv(dev);
-	unsigned long time;
-
-#ifdef HP100_DEBUG_B
-	hp100_outw(0x4209, TRACE);
-	printk("hp100: %s: bm shutdown\n", dev->name);
-#endif
-
-	hp100_page(PERFORMANCE);
-	hp100_outw(0xfefe, IRQ_MASK);	/* mask off all ints */
-	hp100_outw(0xffff, IRQ_STATUS);	/* Ack all ints */
-
-	/* Ensure Interrupts are off */
-	hp100_outw(HP100_INT_EN | HP100_RESET_LB, OPTION_LSW);
-
-	/* Disable all MAC activity */
-	hp100_page(MAC_CTRL);
-	hp100_andb(~(HP100_RX_EN | HP100_TX_EN), MAC_CFG_1);	/* stop rx/tx */
-
-	/* If cascade MMU is not already in reset */
-	if (0 != (hp100_inw(OPTION_LSW) & HP100_HW_RST)) {
-		/* Wait 1.3ms (10Mb max packet time) to ensure MAC is idle so
-		 * MMU pointers will not be reset out from underneath
-		 */
-		hp100_page(MAC_CTRL);
-		for (time = 0; time < 5000; time++) {
-			if ((hp100_inb(MAC_CFG_1) & (HP100_TX_IDLE | HP100_RX_IDLE)) == (HP100_TX_IDLE | HP100_RX_IDLE))
-				break;
-		}
-
-		/* Shutdown algorithm depends on the generation of Cascade */
-		if (lp->chip == HP100_CHIPID_LASSEN) {	/* ETR shutdown/reset */
-			/* Disable Busmaster mode and wait for bit to go to zero. */
-			hp100_page(HW_MAP);
-			hp100_andb(~HP100_BM_MASTER, BM);
-			/* 100 ms timeout */
-			for (time = 0; time < 32000; time++) {
-				if (0 == (hp100_inb(BM) & HP100_BM_MASTER))
-					break;
-			}
-		} else {	/* Shasta or Rainier Shutdown/Reset */
-			/* To ensure all bus master inloading activity has ceased,
-			 * wait for no Rx PDAs or no Rx packets on card.
-			 */
-			hp100_page(PERFORMANCE);
-			/* 100 ms timeout */
-			for (time = 0; time < 10000; time++) {
-				/* RX_PDL: PDLs not executed. */
-				/* RX_PKT_CNT: RX'd packets on card. */
-				if ((hp100_inb(RX_PDL) == 0) && (hp100_inb(RX_PKT_CNT) == 0))
-					break;
-			}
-
-			if (time >= 10000)
-				printk("hp100: %s: BM shutdown error.\n", dev->name);
-
-			/* To ensure all bus master outloading activity has ceased,
-			 * wait until the Tx PDA count goes to zero or no more Tx space
-			 * available in the Tx region of the card.
-			 */
-			/* 100 ms timeout */
-			for (time = 0; time < 10000; time++) {
-				if ((0 == hp100_inb(TX_PKT_CNT)) &&
-				    (0 != (hp100_inb(TX_MEM_FREE) & HP100_AUTO_COMPARE)))
-					break;
-			}
-
-			/* Disable Busmaster mode */
-			hp100_page(HW_MAP);
-			hp100_andb(~HP100_BM_MASTER, BM);
-		}	/* end of shutdown procedure for non-etr parts */
-
-		hp100_cascade_reset(dev, 1);
-	}
-	hp100_page(PERFORMANCE);
-	/* hp100_outw( HP100_BM_READ | HP100_BM_WRITE | HP100_RESET_HB, OPTION_LSW ); */
-	/* Busmaster mode should be shut down now. */
-}
-
-static int hp100_check_lan(struct net_device *dev)
-{
-	struct hp100_private *lp = netdev_priv(dev);
-
-	if (lp->lan_type < 0) {	/* no LAN type detected yet? */
-		hp100_stop_interface(dev);
-		if ((lp->lan_type = hp100_sense_lan(dev)) < 0) {
-			printk("hp100: %s: no connection found - check wire\n", dev->name);
-			hp100_start_interface(dev);	/* 10Mb/s RX packets maybe handled */
-			return -EIO;
-		}
-		if (lp->lan_type == HP100_LAN_100)
-			lp->hub_status = hp100_login_to_vg_hub(dev, 0);	/* relogin */
-		hp100_start_interface(dev);
-	}
-	return 0;
-}
-
-/*
- *  transmit functions
- */
-
-/* tx function for busmaster mode */
-static netdev_tx_t hp100_start_xmit_bm(struct sk_buff *skb,
-				       struct net_device *dev)
-{
-	unsigned long flags;
-	int i, ok_flag;
-	int ioaddr = dev->base_addr;
-	struct hp100_private *lp = netdev_priv(dev);
-	hp100_ring_t *ringptr;
-
-#ifdef HP100_DEBUG_B
-	hp100_outw(0x4210, TRACE);
-	printk("hp100: %s: start_xmit_bm\n", dev->name);
-#endif
-	if (skb->len <= 0)
-		goto drop;
-
-	if (lp->chip == HP100_CHIPID_SHASTA && skb_padto(skb, ETH_ZLEN))
-		return NETDEV_TX_OK;
-
-	/* Get Tx ring tail pointer */
-	if (lp->txrtail->next == lp->txrhead) {
-		/* No memory. */
-#ifdef HP100_DEBUG
-		printk("hp100: %s: start_xmit_bm: No TX PDL available.\n", dev->name);
-#endif
-		/* not waited long enough since last tx? */
-		if (time_before(jiffies, dev_trans_start(dev) + HZ))
-			goto drop;
-
-		if (hp100_check_lan(dev))
-			goto drop;
-
-		if (lp->lan_type == HP100_LAN_100 && lp->hub_status < 0) {
-			/* we have a 100Mb/s adapter but it isn't connected to hub */
-			printk("hp100: %s: login to 100Mb/s hub retry\n", dev->name);
-			hp100_stop_interface(dev);
-			lp->hub_status = hp100_login_to_vg_hub(dev, 0);
-			hp100_start_interface(dev);
-		} else {
-			spin_lock_irqsave(&lp->lock, flags);
-			hp100_ints_off();	/* Useful ? Jean II */
-			i = hp100_sense_lan(dev);
-			hp100_ints_on();
-			spin_unlock_irqrestore(&lp->lock, flags);
-			if (i == HP100_LAN_ERR)
-				printk("hp100: %s: link down detected\n", dev->name);
-			else if (lp->lan_type != i) {	/* cable change! */
-				/* it's very hard - all network settings must be changed!!! */
-				printk("hp100: %s: cable change 10Mb/s <-> 100Mb/s detected\n", dev->name);
-				lp->lan_type = i;
-				hp100_stop_interface(dev);
-				if (lp->lan_type == HP100_LAN_100)
-					lp->hub_status = hp100_login_to_vg_hub(dev, 0);
-				hp100_start_interface(dev);
-			} else {
-				printk("hp100: %s: interface reset\n", dev->name);
-				hp100_stop_interface(dev);
-				if (lp->lan_type == HP100_LAN_100)
-					lp->hub_status = hp100_login_to_vg_hub(dev, 0);
-				hp100_start_interface(dev);
-			}
-		}
-
-		goto drop;
-	}
-
-	/*
-	 * we have to turn int's off before modifying this, otherwise
-	 * a tx_pdl_cleanup could occur at the same time
-	 */
-	spin_lock_irqsave(&lp->lock, flags);
-	ringptr = lp->txrtail;
-	lp->txrtail = ringptr->next;
-
-	/* Check whether packet has minimal packet size */
-	ok_flag = skb->len >= HP100_MIN_PACKET_SIZE;
-	i = ok_flag ? skb->len : HP100_MIN_PACKET_SIZE;
-
-	ringptr->skb = skb;
-	ringptr->pdl[0] = ((1 << 16) | i);	/* PDH: 1 Fragment & length */
-	if (lp->chip == HP100_CHIPID_SHASTA) {
-		/* TODO:Could someone who has the EISA card please check if this works? */
-		ringptr->pdl[2] = i;
-	} else {		/* Lassen */
-		/* In the PDL, don't use the padded size but the real packet size: */
-		ringptr->pdl[2] = skb->len;	/* 1st Frag: Length of frag */
-	}
-	/* Conversion to new PCI API : map skbuf data to PCI bus.
-	 * Doc says it's OK for EISA as well - Jean II */
-	ringptr->pdl[1] = ((u32) pci_map_single(lp->pci_dev, skb->data, ringptr->pdl[2], PCI_DMA_TODEVICE));	/* 1st Frag: Adr. of data */
-
-	/* Hand this PDL to the card. */
-	hp100_outl(ringptr->pdl_paddr, TX_PDA_L);	/* Low Prio. Queue */
-
-	lp->txrcommit++;
-
-	dev->stats.tx_packets++;
-	dev->stats.tx_bytes += skb->len;
-
-	spin_unlock_irqrestore(&lp->lock, flags);
-
-	return NETDEV_TX_OK;
-
-drop:
-	dev_kfree_skb(skb);
-	return NETDEV_TX_OK;
-}
-
-
-/* clean_txring checks if packets have been sent by the card by reading
- * the TX_PDL register from the performance page and comparing it to the
- * number of committed packets. It then frees the skb's of the packets that
- * obviously have been sent to the network.
- *
- * Needs the PERFORMANCE page selected.
- */
-static void hp100_clean_txring(struct net_device *dev)
-{
-	struct hp100_private *lp = netdev_priv(dev);
-	int ioaddr = dev->base_addr;
-	int donecount;
-
-#ifdef HP100_DEBUG_B
-	hp100_outw(0x4211, TRACE);
-	printk("hp100: %s: clean txring\n", dev->name);
-#endif
-
-	/* How many PDLs have been transmitted? */
-	donecount = (lp->txrcommit) - hp100_inb(TX_PDL);
-
-#ifdef HP100_DEBUG
-	if (donecount > MAX_TX_PDL)
-		printk("hp100: %s: Warning: More PDLs transmitted than committed to card???\n", dev->name);
-#endif
-
-	for (; 0 != donecount; donecount--) {
-#ifdef HP100_DEBUG_BM
-		printk("hp100: %s: Free skb: data @0x%.8x txrcommit=0x%x TXPDL=0x%x, done=0x%x\n",
-				dev->name, (u_int) lp->txrhead->skb->data,
-				lp->txrcommit, hp100_inb(TX_PDL), donecount);
-#endif
-		/* Conversion to new PCI API : NOP */
-		pci_unmap_single(lp->pci_dev, (dma_addr_t) lp->txrhead->pdl[1], lp->txrhead->pdl[2], PCI_DMA_TODEVICE);
-		dev_consume_skb_any(lp->txrhead->skb);
-		lp->txrhead->skb = NULL;
-		lp->txrhead = lp->txrhead->next;
-		lp->txrcommit--;
-	}
-}
-
-/* tx function for slave modes */
-static netdev_tx_t hp100_start_xmit(struct sk_buff *skb,
-				    struct net_device *dev)
-{
-	unsigned long flags;
-	int i, ok_flag;
-	int ioaddr = dev->base_addr;
-	u_short val;
-	struct hp100_private *lp = netdev_priv(dev);
-
-#ifdef HP100_DEBUG_B
-	hp100_outw(0x4212, TRACE);
-	printk("hp100: %s: start_xmit\n", dev->name);
-#endif
-	if (skb->len <= 0)
-		goto drop;
-
-	if (hp100_check_lan(dev))
-		goto drop;
-
-	/* If there is not enough free memory on the card... */
-	i = hp100_inl(TX_MEM_FREE) & 0x7fffffff;
-	if (!(((i / 2) - 539) > (skb->len + 16) && (hp100_inb(TX_PKT_CNT) < 255))) {
-#ifdef HP100_DEBUG
-		printk("hp100: %s: start_xmit: tx free mem = 0x%x\n", dev->name, i);
-#endif
-		/* not waited long enough since last failed tx try? */
-		if (time_before(jiffies, dev_trans_start(dev) + HZ)) {
-#ifdef HP100_DEBUG
-			printk("hp100: %s: trans_start timing problem\n",
-			       dev->name);
-#endif
-			goto drop;
-		}
-		if (lp->lan_type == HP100_LAN_100 && lp->hub_status < 0) {
-			/* we have a 100Mb/s adapter but it isn't connected to hub */
-			printk("hp100: %s: login to 100Mb/s hub retry\n", dev->name);
-			hp100_stop_interface(dev);
-			lp->hub_status = hp100_login_to_vg_hub(dev, 0);
-			hp100_start_interface(dev);
-		} else {
-			spin_lock_irqsave(&lp->lock, flags);
-			hp100_ints_off();	/* Useful ? Jean II */
-			i = hp100_sense_lan(dev);
-			hp100_ints_on();
-			spin_unlock_irqrestore(&lp->lock, flags);
-			if (i == HP100_LAN_ERR)
-				printk("hp100: %s: link down detected\n", dev->name);
-			else if (lp->lan_type != i) {	/* cable change! */
-				/* it's very hard - all network setting must be changed!!! */
-				printk("hp100: %s: cable change 10Mb/s <-> 100Mb/s detected\n", dev->name);
-				lp->lan_type = i;
-				hp100_stop_interface(dev);
-				if (lp->lan_type == HP100_LAN_100)
-					lp->hub_status = hp100_login_to_vg_hub(dev, 0);
-				hp100_start_interface(dev);
-			} else {
-				printk("hp100: %s: interface reset\n", dev->name);
-				hp100_stop_interface(dev);
-				if (lp->lan_type == HP100_LAN_100)
-					lp->hub_status = hp100_login_to_vg_hub(dev, 0);
-				hp100_start_interface(dev);
-				mdelay(1);
-			}
-		}
-		goto drop;
-	}
-
-	for (i = 0; i < 6000 && (hp100_inb(OPTION_MSW) & HP100_TX_CMD); i++) {
-#ifdef HP100_DEBUG_TX
-		printk("hp100: %s: start_xmit: busy\n", dev->name);
-#endif
-	}
-
-	spin_lock_irqsave(&lp->lock, flags);
-	hp100_ints_off();
-	val = hp100_inw(IRQ_STATUS);
-	/* Ack / clear the interrupt TX_COMPLETE interrupt - this interrupt is set
-	 * when the current packet being transmitted on the wire is completed. */
-	hp100_outw(HP100_TX_COMPLETE, IRQ_STATUS);
-#ifdef HP100_DEBUG_TX
-	printk("hp100: %s: start_xmit: irq_status=0x%.4x, irqmask=0x%.4x, len=%d\n",
-			dev->name, val, hp100_inw(IRQ_MASK), (int) skb->len);
-#endif
-
-	ok_flag = skb->len >= HP100_MIN_PACKET_SIZE;
-	i = ok_flag ? skb->len : HP100_MIN_PACKET_SIZE;
-
-	hp100_outw(i, DATA32);	/* tell card the total packet length */
-	hp100_outw(i, FRAGMENT_LEN);	/* and first/only fragment length    */
-
-	if (lp->mode == 2) {	/* memory mapped */
-		/* Note: The J2585B needs alignment to 32bits here!  */
-		memcpy_toio(lp->mem_ptr_virt, skb->data, (skb->len + 3) & ~3);
-		if (!ok_flag)
-			memset_io(lp->mem_ptr_virt, 0, HP100_MIN_PACKET_SIZE - skb->len);
-	} else {		/* programmed i/o */
-		outsl(ioaddr + HP100_REG_DATA32, skb->data,
-		      (skb->len + 3) >> 2);
-		if (!ok_flag)
-			for (i = (skb->len + 3) & ~3; i < HP100_MIN_PACKET_SIZE; i += 4)
-				hp100_outl(0, DATA32);
-	}
-
-	hp100_outb(HP100_TX_CMD | HP100_SET_LB, OPTION_MSW);	/* send packet */
-
-	dev->stats.tx_packets++;
-	dev->stats.tx_bytes += skb->len;
-	hp100_ints_on();
-	spin_unlock_irqrestore(&lp->lock, flags);
-
-	dev_consume_skb_any(skb);
-
-#ifdef HP100_DEBUG_TX
-	printk("hp100: %s: start_xmit: end\n", dev->name);
-#endif
-
-	return NETDEV_TX_OK;
-
-drop:
-	dev_kfree_skb(skb);
-	return NETDEV_TX_OK;
-
-}
-
-
-/*
- * Receive Function (Non-Busmaster mode)
- * Called when an "Receive Packet" interrupt occurs, i.e. the receive
- * packet counter is non-zero.
- * For non-busmaster, this function does the whole work of transferring
- * the packet to the host memory and then up to higher layers via skb
- * and netif_rx.
- */
-
-static void hp100_rx(struct net_device *dev)
-{
-	int packets, pkt_len;
-	int ioaddr = dev->base_addr;
-	struct hp100_private *lp = netdev_priv(dev);
-	u_int header;
-	struct sk_buff *skb;
-
-#ifdef DEBUG_B
-	hp100_outw(0x4213, TRACE);
-	printk("hp100: %s: rx\n", dev->name);
-#endif
-
-	/* First get indication of received lan packet */
-	/* RX_PKT_CND indicates the number of packets which have been fully */
-	/* received onto the card but have not been fully transferred of the card */
-	packets = hp100_inb(RX_PKT_CNT);
-#ifdef HP100_DEBUG_RX
-	if (packets > 1)
-		printk("hp100: %s: rx: waiting packets = %d\n", dev->name, packets);
-#endif
-
-	while (packets-- > 0) {
-		/* If ADV_NXT_PKT is still set, we have to wait until the card has */
-		/* really advanced to the next packet. */
-		for (pkt_len = 0; pkt_len < 6000 && (hp100_inb(OPTION_MSW) & HP100_ADV_NXT_PKT); pkt_len++) {
-#ifdef HP100_DEBUG_RX
-			printk ("hp100: %s: rx: busy, remaining packets = %d\n", dev->name, packets);
-#endif
-		}
-
-		/* First we get the header, which contains information about the */
-		/* actual length of the received packet. */
-		if (lp->mode == 2) {	/* memory mapped mode */
-			header = readl(lp->mem_ptr_virt);
-		} else		/* programmed i/o */
-			header = hp100_inl(DATA32);
-
-		pkt_len = ((header & HP100_PKT_LEN_MASK) + 3) & ~3;
-
-#ifdef HP100_DEBUG_RX
-		printk("hp100: %s: rx: new packet - length=%d, errors=0x%x, dest=0x%x\n",
-				     dev->name, header & HP100_PKT_LEN_MASK,
-				     (header >> 16) & 0xfff8, (header >> 16) & 7);
-#endif
-
-		/* Now we allocate the skb and transfer the data into it. */
-		skb = netdev_alloc_skb(dev, pkt_len + 2);
-		if (skb == NULL) {	/* Not enough memory->drop packet */
-#ifdef HP100_DEBUG
-			printk("hp100: %s: rx: couldn't allocate a sk_buff of size %d\n",
-					     dev->name, pkt_len);
-#endif
-			dev->stats.rx_dropped++;
-		} else {	/* skb successfully allocated */
-
-			u_char *ptr;
-
-			skb_reserve(skb,2);
-
-			/* ptr to start of the sk_buff data area */
-			skb_put(skb, pkt_len);
-			ptr = skb->data;
-
-			/* Now transfer the data from the card into that area */
-			if (lp->mode == 2)
-				memcpy_fromio(ptr, lp->mem_ptr_virt,pkt_len);
-			else	/* io mapped */
-				insl(ioaddr + HP100_REG_DATA32, ptr, pkt_len >> 2);
-
-			skb->protocol = eth_type_trans(skb, dev);
-
-#ifdef HP100_DEBUG_RX
-			printk("hp100: %s: rx: %02x %02x %02x %02x %02x %02x %02x %02x %02x %02x %02x %02x\n",
-					dev->name, ptr[0], ptr[1], ptr[2], ptr[3],
-		 			ptr[4], ptr[5], ptr[6], ptr[7], ptr[8],
-					ptr[9], ptr[10], ptr[11]);
-#endif
-			netif_rx(skb);
-			dev->stats.rx_packets++;
-			dev->stats.rx_bytes += pkt_len;
-		}
-
-		/* Indicate the card that we have got the packet */
-		hp100_outb(HP100_ADV_NXT_PKT | HP100_SET_LB, OPTION_MSW);
-
-		switch (header & 0x00070000) {
-		case (HP100_MULTI_ADDR_HASH << 16):
-		case (HP100_MULTI_ADDR_NO_HASH << 16):
-			dev->stats.multicast++;
-			break;
-		}
-	}			/* end of while(there are packets) loop */
-#ifdef HP100_DEBUG_RX
-	printk("hp100_rx: %s: end\n", dev->name);
-#endif
-}
-
-/*
- * Receive Function for Busmaster Mode
- */
-static void hp100_rx_bm(struct net_device *dev)
-{
-	int ioaddr = dev->base_addr;
-	struct hp100_private *lp = netdev_priv(dev);
-	hp100_ring_t *ptr;
-	u_int header;
-	int pkt_len;
-
-#ifdef HP100_DEBUG_B
-	hp100_outw(0x4214, TRACE);
-	printk("hp100: %s: rx_bm\n", dev->name);
-#endif
-
-#ifdef HP100_DEBUG
-	if (0 == lp->rxrcommit) {
-		printk("hp100: %s: rx_bm called although no PDLs were committed to adapter?\n", dev->name);
-		return;
-	} else
-		/* RX_PKT_CNT states how many PDLs are currently formatted and available to
-		 * the cards BM engine */
-	if ((hp100_inw(RX_PKT_CNT) & 0x00ff) >= lp->rxrcommit) {
-		printk("hp100: %s: More packets received than committed? RX_PKT_CNT=0x%x, commit=0x%x\n",
-				     dev->name, hp100_inw(RX_PKT_CNT) & 0x00ff,
-				     lp->rxrcommit);
-		return;
-	}
-#endif
-
-	while ((lp->rxrcommit > hp100_inb(RX_PDL))) {
-		/*
-		 * The packet was received into the pdl pointed to by lp->rxrhead (
-		 * the oldest pdl in the ring
-		 */
-
-		/* First we get the header, which contains information about the */
-		/* actual length of the received packet. */
-
-		ptr = lp->rxrhead;
-
-		header = *(ptr->pdl - 1);
-		pkt_len = (header & HP100_PKT_LEN_MASK);
-
-		/* Conversion to new PCI API : NOP */
-		pci_unmap_single(lp->pci_dev, (dma_addr_t) ptr->pdl[3], MAX_ETHER_SIZE, PCI_DMA_FROMDEVICE);
-
-#ifdef HP100_DEBUG_BM
-		printk("hp100: %s: rx_bm: header@0x%x=0x%x length=%d, errors=0x%x, dest=0x%x\n",
-				dev->name, (u_int) (ptr->pdl - 1), (u_int) header,
-				pkt_len, (header >> 16) & 0xfff8, (header >> 16) & 7);
-		printk("hp100: %s: RX_PDL_COUNT:0x%x TX_PDL_COUNT:0x%x, RX_PKT_CNT=0x%x PDH=0x%x, Data@0x%x len=0x%x\n",
-		   		dev->name, hp100_inb(RX_PDL), hp100_inb(TX_PDL),
-				hp100_inb(RX_PKT_CNT), (u_int) * (ptr->pdl),
-				(u_int) * (ptr->pdl + 3), (u_int) * (ptr->pdl + 4));
-#endif
-
-		if ((pkt_len >= MIN_ETHER_SIZE) &&
-		    (pkt_len <= MAX_ETHER_SIZE)) {
-			if (ptr->skb == NULL) {
-				printk("hp100: %s: rx_bm: skb null\n", dev->name);
-				/* can happen if we only allocated room for the pdh due to memory shortage. */
-				dev->stats.rx_dropped++;
-			} else {
-				skb_trim(ptr->skb, pkt_len);	/* Shorten it */
-				ptr->skb->protocol =
-				    eth_type_trans(ptr->skb, dev);
-
-				netif_rx(ptr->skb);	/* Up and away... */
-
-				dev->stats.rx_packets++;
-				dev->stats.rx_bytes += pkt_len;
-			}
-
-			switch (header & 0x00070000) {
-			case (HP100_MULTI_ADDR_HASH << 16):
-			case (HP100_MULTI_ADDR_NO_HASH << 16):
-				dev->stats.multicast++;
-				break;
-			}
-		} else {
-#ifdef HP100_DEBUG
-			printk("hp100: %s: rx_bm: Received bad packet (length=%d)\n", dev->name, pkt_len);
-#endif
-			if (ptr->skb != NULL)
-				dev_kfree_skb_any(ptr->skb);
-			dev->stats.rx_errors++;
-		}
-
-		lp->rxrhead = lp->rxrhead->next;
-
-		/* Allocate a new rx PDL (so lp->rxrcommit stays the same) */
-		if (0 == hp100_build_rx_pdl(lp->rxrtail, dev)) {
-			/* No space for skb, header can still be received. */
-#ifdef HP100_DEBUG
-			printk("hp100: %s: rx_bm: No space for new PDL.\n", dev->name);
-#endif
-			return;
-		} else {	/* successfully allocated new PDL - put it in ringlist at tail. */
-			hp100_outl((u32) lp->rxrtail->pdl_paddr, RX_PDA);
-			lp->rxrtail = lp->rxrtail->next;
-		}
-
-	}
-}
-
-/*
- *  statistics
- */
-static struct net_device_stats *hp100_get_stats(struct net_device *dev)
-{
-	unsigned long flags;
-	int ioaddr = dev->base_addr;
-	struct hp100_private *lp = netdev_priv(dev);
-
-#ifdef HP100_DEBUG_B
-	hp100_outw(0x4215, TRACE);
-#endif
-
-	spin_lock_irqsave(&lp->lock, flags);
-	hp100_ints_off();	/* Useful ? Jean II */
-	hp100_update_stats(dev);
-	hp100_ints_on();
-	spin_unlock_irqrestore(&lp->lock, flags);
-	return &(dev->stats);
-}
-
-static void hp100_update_stats(struct net_device *dev)
-{
-	int ioaddr = dev->base_addr;
-	u_short val;
-
-#ifdef HP100_DEBUG_B
-	hp100_outw(0x4216, TRACE);
-	printk("hp100: %s: update-stats\n", dev->name);
-#endif
-
-	/* Note: Statistics counters clear when read. */
-	hp100_page(MAC_CTRL);
-	val = hp100_inw(DROPPED) & 0x0fff;
-	dev->stats.rx_errors += val;
-	dev->stats.rx_over_errors += val;
-	val = hp100_inb(CRC);
-	dev->stats.rx_errors += val;
-	dev->stats.rx_crc_errors += val;
-	val = hp100_inb(ABORT);
-	dev->stats.tx_errors += val;
-	dev->stats.tx_aborted_errors += val;
-	hp100_page(PERFORMANCE);
-}
-
-static void hp100_misc_interrupt(struct net_device *dev)
-{
-#ifdef HP100_DEBUG_B
-	int ioaddr = dev->base_addr;
-#endif
-
-#ifdef HP100_DEBUG_B
-	int ioaddr = dev->base_addr;
-	hp100_outw(0x4216, TRACE);
-	printk("hp100: %s: misc_interrupt\n", dev->name);
-#endif
-
-	/* Note: Statistics counters clear when read. */
-	dev->stats.rx_errors++;
-	dev->stats.tx_errors++;
-}
-
-static void hp100_clear_stats(struct hp100_private *lp, int ioaddr)
-{
-	unsigned long flags;
-
-#ifdef HP100_DEBUG_B
-	hp100_outw(0x4217, TRACE);
-	printk("hp100: %s: clear_stats\n", dev->name);
-#endif
-
-	spin_lock_irqsave(&lp->lock, flags);
-	hp100_page(MAC_CTRL);	/* get all statistics bytes */
-	hp100_inw(DROPPED);
-	hp100_inb(CRC);
-	hp100_inb(ABORT);
-	hp100_page(PERFORMANCE);
-	spin_unlock_irqrestore(&lp->lock, flags);
-}
-
-
-/*
- *  multicast setup
- */
-
-/*
- *  Set or clear the multicast filter for this adapter.
- */
-
-static void hp100_set_multicast_list(struct net_device *dev)
-{
-	unsigned long flags;
-	int ioaddr = dev->base_addr;
-	struct hp100_private *lp = netdev_priv(dev);
-
-#ifdef HP100_DEBUG_B
-	hp100_outw(0x4218, TRACE);
-	printk("hp100: %s: set_mc_list\n", dev->name);
-#endif
-
-	spin_lock_irqsave(&lp->lock, flags);
-	hp100_ints_off();
-	hp100_page(MAC_CTRL);
-	hp100_andb(~(HP100_RX_EN | HP100_TX_EN), MAC_CFG_1);	/* stop rx/tx */
-
-	if (dev->flags & IFF_PROMISC) {
-		lp->mac2_mode = HP100_MAC2MODE6;	/* promiscuous mode = get all good */
-		lp->mac1_mode = HP100_MAC1MODE6;	/* packets on the net */
-		memset(&lp->hash_bytes, 0xff, 8);
-	} else if (!netdev_mc_empty(dev) || (dev->flags & IFF_ALLMULTI)) {
-		lp->mac2_mode = HP100_MAC2MODE5;	/* multicast mode = get packets for */
-		lp->mac1_mode = HP100_MAC1MODE5;	/* me, broadcasts and all multicasts */
-#ifdef HP100_MULTICAST_FILTER	/* doesn't work!!! */
-		if (dev->flags & IFF_ALLMULTI) {
-			/* set hash filter to receive all multicast packets */
-			memset(&lp->hash_bytes, 0xff, 8);
-		} else {
-			int i, idx;
-			u_char *addrs;
-			struct netdev_hw_addr *ha;
-
-			memset(&lp->hash_bytes, 0x00, 8);
-#ifdef HP100_DEBUG
-			printk("hp100: %s: computing hash filter - mc_count = %i\n",
-			       dev->name, netdev_mc_count(dev));
-#endif
-			netdev_for_each_mc_addr(ha, dev) {
-				addrs = ha->addr;
-#ifdef HP100_DEBUG
-				printk("hp100: %s: multicast = %pM, ",
-					     dev->name, addrs);
-#endif
-				for (i = idx = 0; i < 6; i++) {
-					idx ^= *addrs++ & 0x3f;
-					printk(":%02x:", idx);
-				}
-#ifdef HP100_DEBUG
-				printk("idx = %i\n", idx);
-#endif
-				lp->hash_bytes[idx >> 3] |= (1 << (idx & 7));
-			}
-		}
-#else
-		memset(&lp->hash_bytes, 0xff, 8);
-#endif
-	} else {
-		lp->mac2_mode = HP100_MAC2MODE3;	/* normal mode = get packets for me */
-		lp->mac1_mode = HP100_MAC1MODE3;	/* and broadcasts */
-		memset(&lp->hash_bytes, 0x00, 8);
-	}
-
-	if (((hp100_inb(MAC_CFG_1) & 0x0f) != lp->mac1_mode) ||
-	    (hp100_inb(MAC_CFG_2) != lp->mac2_mode)) {
-		int i;
-
-		hp100_outb(lp->mac2_mode, MAC_CFG_2);
-		hp100_andb(HP100_MAC1MODEMASK, MAC_CFG_1);	/* clear mac1 mode bits */
-		hp100_orb(lp->mac1_mode, MAC_CFG_1);	/* and set the new mode */
-
-		hp100_page(MAC_ADDRESS);
-		for (i = 0; i < 8; i++)
-			hp100_outb(lp->hash_bytes[i], HASH_BYTE0 + i);
-#ifdef HP100_DEBUG
-		printk("hp100: %s: mac1 = 0x%x, mac2 = 0x%x, multicast hash = %02x:%02x:%02x:%02x:%02x:%02x:%02x:%02x\n",
-				     dev->name, lp->mac1_mode, lp->mac2_mode,
-				     lp->hash_bytes[0], lp->hash_bytes[1],
-				     lp->hash_bytes[2], lp->hash_bytes[3],
-				     lp->hash_bytes[4], lp->hash_bytes[5],
-				     lp->hash_bytes[6], lp->hash_bytes[7]);
-#endif
-
-		if (lp->lan_type == HP100_LAN_100) {
-#ifdef HP100_DEBUG
-			printk("hp100: %s: 100VG MAC settings have changed - relogin.\n", dev->name);
-#endif
-			lp->hub_status = hp100_login_to_vg_hub(dev, 1);	/* force a relogin to the hub */
-		}
-	} else {
-		int i;
-		u_char old_hash_bytes[8];
-
-		hp100_page(MAC_ADDRESS);
-		for (i = 0; i < 8; i++)
-			old_hash_bytes[i] = hp100_inb(HASH_BYTE0 + i);
-		if (memcmp(old_hash_bytes, &lp->hash_bytes, 8)) {
-			for (i = 0; i < 8; i++)
-				hp100_outb(lp->hash_bytes[i], HASH_BYTE0 + i);
-#ifdef HP100_DEBUG
-			printk("hp100: %s: multicast hash = %02x:%02x:%02x:%02x:%02x:%02x:%02x:%02x\n",
-					dev->name, lp->hash_bytes[0],
-					lp->hash_bytes[1], lp->hash_bytes[2],
-					lp->hash_bytes[3], lp->hash_bytes[4],
-					lp->hash_bytes[5], lp->hash_bytes[6],
-					lp->hash_bytes[7]);
-#endif
-
-			if (lp->lan_type == HP100_LAN_100) {
-#ifdef HP100_DEBUG
-				printk("hp100: %s: 100VG MAC settings have changed - relogin.\n", dev->name);
-#endif
-				lp->hub_status = hp100_login_to_vg_hub(dev, 1);	/* force a relogin to the hub */
-			}
-		}
-	}
-
-	hp100_page(MAC_CTRL);
-	hp100_orb(HP100_RX_EN | HP100_RX_IDLE |	/* enable rx */
-		  HP100_TX_EN | HP100_TX_IDLE, MAC_CFG_1);	/* enable tx */
-
-	hp100_page(PERFORMANCE);
-	hp100_ints_on();
-	spin_unlock_irqrestore(&lp->lock, flags);
-}
-
-/*
- *  hardware interrupt handling
- */
-
-static irqreturn_t hp100_interrupt(int irq, void *dev_id)
-{
-	struct net_device *dev = (struct net_device *) dev_id;
-	struct hp100_private *lp = netdev_priv(dev);
-
-	int ioaddr;
-	u_int val;
-
-	if (dev == NULL)
-		return IRQ_NONE;
-	ioaddr = dev->base_addr;
-
-	spin_lock(&lp->lock);
-
-	hp100_ints_off();
-
-#ifdef HP100_DEBUG_B
-	hp100_outw(0x4219, TRACE);
-#endif
-
-	/*  hp100_page( PERFORMANCE ); */
-	val = hp100_inw(IRQ_STATUS);
-#ifdef HP100_DEBUG_IRQ
-	printk("hp100: %s: mode=%x,IRQ_STAT=0x%.4x,RXPKTCNT=0x%.2x RXPDL=0x%.2x TXPKTCNT=0x%.2x TXPDL=0x%.2x\n",
-			     dev->name, lp->mode, (u_int) val, hp100_inb(RX_PKT_CNT),
-			     hp100_inb(RX_PDL), hp100_inb(TX_PKT_CNT), hp100_inb(TX_PDL));
-#endif
-
-	if (val == 0) {		/* might be a shared interrupt */
-		spin_unlock(&lp->lock);
-		hp100_ints_on();
-		return IRQ_NONE;
-	}
-	/* We're only interested in those interrupts we really enabled. */
-	/* val &= hp100_inw( IRQ_MASK ); */
-
-	/*
-	 * RX_PDL_FILL_COMPL is set whenever a RX_PDL has been executed. A RX_PDL
-	 * is considered executed whenever the RX_PDL data structure is no longer
-	 * needed.
-	 */
-	if (val & HP100_RX_PDL_FILL_COMPL) {
-		if (lp->mode == 1)
-			hp100_rx_bm(dev);
-		else {
-			printk("hp100: %s: rx_pdl_fill_compl interrupt although not busmaster?\n", dev->name);
-		}
-	}
-
-	/*
-	 * The RX_PACKET interrupt is set, when the receive packet counter is
-	 * non zero. We use this interrupt for receiving in slave mode. In
-	 * busmaster mode, we use it to make sure we did not miss any rx_pdl_fill
-	 * interrupts. If rx_pdl_fill_compl is not set and rx_packet is set, then
-	 * we somehow have missed a rx_pdl_fill_compl interrupt.
-	 */
-
-	if (val & HP100_RX_PACKET) {	/* Receive Packet Counter is non zero */
-		if (lp->mode != 1)	/* non busmaster */
-			hp100_rx(dev);
-		else if (!(val & HP100_RX_PDL_FILL_COMPL)) {
-			/* Shouldn't happen - maybe we missed a RX_PDL_FILL Interrupt?  */
-			hp100_rx_bm(dev);
-		}
-	}
-
-	/*
-	 * Ack. that we have noticed the interrupt and thereby allow next one.
-	 * Note that this is now done after the slave rx function, since first
-	 * acknowledging and then setting ADV_NXT_PKT caused an extra interrupt
-	 * on the J2573.
-	 */
-	hp100_outw(val, IRQ_STATUS);
-
-	/*
-	 * RX_ERROR is set when a packet is dropped due to no memory resources on
-	 * the card or when a RCV_ERR occurs.
-	 * TX_ERROR is set when a TX_ABORT condition occurs in the MAC->exists
-	 * only in the 802.3 MAC and happens when 16 collisions occur during a TX
-	 */
-	if (val & (HP100_TX_ERROR | HP100_RX_ERROR)) {
-#ifdef HP100_DEBUG_IRQ
-		printk("hp100: %s: TX/RX Error IRQ\n", dev->name);
-#endif
-		hp100_update_stats(dev);
-		if (lp->mode == 1) {
-			hp100_rxfill(dev);
-			hp100_clean_txring(dev);
-		}
-	}
-
-	/*
-	 * RX_PDA_ZERO is set when the PDA count goes from non-zero to zero.
-	 */
-	if ((lp->mode == 1) && (val & (HP100_RX_PDA_ZERO)))
-		hp100_rxfill(dev);
-
-	/*
-	 * HP100_TX_COMPLETE interrupt occurs when packet transmitted on wire
-	 * is completed
-	 */
-	if ((lp->mode == 1) && (val & (HP100_TX_COMPLETE)))
-		hp100_clean_txring(dev);
-
-	/*
-	 * MISC_ERROR is set when either the LAN link goes down or a detected
-	 * bus error occurs.
-	 */
-	if (val & HP100_MISC_ERROR) {	/* New for J2585B */
-#ifdef HP100_DEBUG_IRQ
-		printk
-		    ("hp100: %s: Misc. Error Interrupt - Check cabling.\n",
-		     dev->name);
-#endif
-		if (lp->mode == 1) {
-			hp100_clean_txring(dev);
-			hp100_rxfill(dev);
-		}
-		hp100_misc_interrupt(dev);
-	}
-
-	spin_unlock(&lp->lock);
-	hp100_ints_on();
-	return IRQ_HANDLED;
-}
-
-/*
- *  some misc functions
- */
-
-static void hp100_start_interface(struct net_device *dev)
-{
-	unsigned long flags;
-	int ioaddr = dev->base_addr;
-	struct hp100_private *lp = netdev_priv(dev);
-
-#ifdef HP100_DEBUG_B
-	hp100_outw(0x4220, TRACE);
-	printk("hp100: %s: hp100_start_interface\n", dev->name);
-#endif
-
-	spin_lock_irqsave(&lp->lock, flags);
-
-	/* Ensure the adapter does not want to request an interrupt when */
-	/* enabling the IRQ line to be active on the bus (i.e. not tri-stated) */
-	hp100_page(PERFORMANCE);
-	hp100_outw(0xfefe, IRQ_MASK);	/* mask off all ints */
-	hp100_outw(0xffff, IRQ_STATUS);	/* ack all IRQs */
-	hp100_outw(HP100_FAKE_INT | HP100_INT_EN | HP100_RESET_LB,
-		   OPTION_LSW);
-	/* Un Tri-state int. TODO: Check if shared interrupts can be realised? */
-	hp100_outw(HP100_TRI_INT | HP100_RESET_HB, OPTION_LSW);
-
-	if (lp->mode == 1) {
-		/* Make sure BM bit is set... */
-		hp100_page(HW_MAP);
-		hp100_orb(HP100_BM_MASTER, BM);
-		hp100_rxfill(dev);
-	} else if (lp->mode == 2) {
-		/* Enable memory mapping. Note: Don't do this when busmaster. */
-		hp100_outw(HP100_MMAP_DIS | HP100_RESET_HB, OPTION_LSW);
-	}
-
-	hp100_page(PERFORMANCE);
-	hp100_outw(0xfefe, IRQ_MASK);	/* mask off all ints */
-	hp100_outw(0xffff, IRQ_STATUS);	/* ack IRQ */
-
-	/* enable a few interrupts: */
-	if (lp->mode == 1) {	/* busmaster mode */
-		hp100_outw(HP100_RX_PDL_FILL_COMPL |
-			   HP100_RX_PDA_ZERO | HP100_RX_ERROR |
-			   /* HP100_RX_PACKET    | */
-			   /* HP100_RX_EARLY_INT |  */ HP100_SET_HB |
-			   /* HP100_TX_PDA_ZERO  |  */
-			   HP100_TX_COMPLETE |
-			   /* HP100_MISC_ERROR   |  */
-			   HP100_TX_ERROR | HP100_SET_LB, IRQ_MASK);
-	} else {
-		hp100_outw(HP100_RX_PACKET |
-			   HP100_RX_ERROR | HP100_SET_HB |
-			   HP100_TX_ERROR | HP100_SET_LB, IRQ_MASK);
-	}
-
-	/* Note : before hp100_set_multicast_list(), because it will play with
-	 * spinlock itself... Jean II */
-	spin_unlock_irqrestore(&lp->lock, flags);
-
-	/* Enable MAC Tx and RX, set MAC modes, ... */
-	hp100_set_multicast_list(dev);
-}
-
-static void hp100_stop_interface(struct net_device *dev)
-{
-	struct hp100_private *lp = netdev_priv(dev);
-	int ioaddr = dev->base_addr;
-	u_int val;
-
-#ifdef HP100_DEBUG_B
-	printk("hp100: %s: hp100_stop_interface\n", dev->name);
-	hp100_outw(0x4221, TRACE);
-#endif
-
-	if (lp->mode == 1)
-		hp100_BM_shutdown(dev);
-	else {
-		/* Note: MMAP_DIS will be reenabled by start_interface */
-		hp100_outw(HP100_INT_EN | HP100_RESET_LB |
-			   HP100_TRI_INT | HP100_MMAP_DIS | HP100_SET_HB,
-			   OPTION_LSW);
-		val = hp100_inw(OPTION_LSW);
-
-		hp100_page(MAC_CTRL);
-		hp100_andb(~(HP100_RX_EN | HP100_TX_EN), MAC_CFG_1);
-
-		if (!(val & HP100_HW_RST))
-			return;	/* If reset, imm. return ... */
-		/* ... else: busy wait until idle */
-		for (val = 0; val < 6000; val++)
-			if ((hp100_inb(MAC_CFG_1) & (HP100_TX_IDLE | HP100_RX_IDLE)) == (HP100_TX_IDLE | HP100_RX_IDLE)) {
-				hp100_page(PERFORMANCE);
-				return;
-			}
-		printk("hp100: %s: hp100_stop_interface - timeout\n", dev->name);
-		hp100_page(PERFORMANCE);
-	}
-}
-
-static void hp100_load_eeprom(struct net_device *dev, u_short probe_ioaddr)
-{
-	int i;
-	int ioaddr = probe_ioaddr > 0 ? probe_ioaddr : dev->base_addr;
-
-#ifdef HP100_DEBUG_B
-	hp100_outw(0x4222, TRACE);
-#endif
-
-	hp100_page(EEPROM_CTRL);
-	hp100_andw(~HP100_EEPROM_LOAD, EEPROM_CTRL);
-	hp100_orw(HP100_EEPROM_LOAD, EEPROM_CTRL);
-	for (i = 0; i < 10000; i++)
-		if (!(hp100_inb(OPTION_MSW) & HP100_EE_LOAD))
-			return;
-	printk("hp100: %s: hp100_load_eeprom - timeout\n", dev->name);
-}
-
-/*  Sense connection status.
- *  return values: LAN_10  - Connected to 10Mbit/s network
- *                 LAN_100 - Connected to 100Mbit/s network
- *                 LAN_ERR - not connected or 100Mbit/s Hub down
- */
-static int hp100_sense_lan(struct net_device *dev)
-{
-	int ioaddr = dev->base_addr;
-	u_short val_VG, val_10;
-	struct hp100_private *lp = netdev_priv(dev);
-
-#ifdef HP100_DEBUG_B
-	hp100_outw(0x4223, TRACE);
-#endif
-
-	hp100_page(MAC_CTRL);
-	val_10 = hp100_inb(10_LAN_CFG_1);
-	val_VG = hp100_inb(VG_LAN_CFG_1);
-	hp100_page(PERFORMANCE);
-#ifdef HP100_DEBUG
-	printk("hp100: %s: sense_lan: val_VG = 0x%04x, val_10 = 0x%04x\n",
-	       dev->name, val_VG, val_10);
-#endif
-
-	if (val_10 & HP100_LINK_BEAT_ST)	/* 10Mb connection is active */
-		return HP100_LAN_10;
-
-	if (val_10 & HP100_AUI_ST) {	/* have we BNC or AUI onboard? */
-		/*
-		 * This can be overriden by dos utility, so if this has no effect,
-		 * perhaps you need to download that utility from HP and set card
-		 * back to "auto detect".
-		 */
-		val_10 |= HP100_AUI_SEL | HP100_LOW_TH;
-		hp100_page(MAC_CTRL);
-		hp100_outb(val_10, 10_LAN_CFG_1);
-		hp100_page(PERFORMANCE);
-		return HP100_LAN_COAX;
-	}
-
-	/* Those cards don't have a 100 Mbit connector */
-	if ( !strcmp(lp->id, "HWP1920")  ||
-	     (lp->pci_dev &&
-	      lp->pci_dev->vendor == PCI_VENDOR_ID &&
-	      (lp->pci_dev->device == PCI_DEVICE_ID_HP_J2970A ||
-	       lp->pci_dev->device == PCI_DEVICE_ID_HP_J2973A)))
-		return HP100_LAN_ERR;
-
-	if (val_VG & HP100_LINK_CABLE_ST)	/* Can hear the HUBs tone. */
-		return HP100_LAN_100;
-	return HP100_LAN_ERR;
-}
-
-static int hp100_down_vg_link(struct net_device *dev)
-{
-	struct hp100_private *lp = netdev_priv(dev);
-	int ioaddr = dev->base_addr;
-	unsigned long time;
-	long savelan, newlan;
-
-#ifdef HP100_DEBUG_B
-	hp100_outw(0x4224, TRACE);
-	printk("hp100: %s: down_vg_link\n", dev->name);
-#endif
-
-	hp100_page(MAC_CTRL);
-	time = jiffies + (HZ / 4);
-	do {
-		if (hp100_inb(VG_LAN_CFG_1) & HP100_LINK_CABLE_ST)
-			break;
-		if (!in_interrupt())
-			schedule_timeout_interruptible(1);
-	} while (time_after(time, jiffies));
-
-	if (time_after_eq(jiffies, time))	/* no signal->no logout */
-		return 0;
-
-	/* Drop the VG Link by clearing the link up cmd and load addr. */
-
-	hp100_andb(~(HP100_LOAD_ADDR | HP100_LINK_CMD), VG_LAN_CFG_1);
-	hp100_orb(HP100_VG_SEL, VG_LAN_CFG_1);
-
-	/* Conditionally stall for >250ms on Link-Up Status (to go down) */
-	time = jiffies + (HZ / 2);
-	do {
-		if (!(hp100_inb(VG_LAN_CFG_1) & HP100_LINK_UP_ST))
-			break;
-		if (!in_interrupt())
-			schedule_timeout_interruptible(1);
-	} while (time_after(time, jiffies));
-
-#ifdef HP100_DEBUG
-	if (time_after_eq(jiffies, time))
-		printk("hp100: %s: down_vg_link: Link does not go down?\n", dev->name);
-#endif
-
-	/* To prevent condition where Rev 1 VG MAC and old hubs do not complete */
-	/* logout under traffic (even though all the status bits are cleared),  */
-	/* do this workaround to get the Rev 1 MAC in its idle state */
-	if (lp->chip == HP100_CHIPID_LASSEN) {
-		/* Reset VG MAC to insure it leaves the logoff state even if */
-		/* the Hub is still emitting tones */
-		hp100_andb(~HP100_VG_RESET, VG_LAN_CFG_1);
-		udelay(1500);	/* wait for >1ms */
-		hp100_orb(HP100_VG_RESET, VG_LAN_CFG_1);	/* Release Reset */
-		udelay(1500);
-	}
-
-	/* New: For lassen, switch to 10 Mbps mac briefly to clear training ACK */
-	/* to get the VG mac to full reset. This is not req.d with later chips */
-	/* Note: It will take the between 1 and 2 seconds for the VG mac to be */
-	/* selected again! This will be left to the connect hub function to */
-	/* perform if desired.  */
-	if (lp->chip == HP100_CHIPID_LASSEN) {
-		/* Have to write to 10 and 100VG control registers simultaneously */
-		savelan = newlan = hp100_inl(10_LAN_CFG_1);	/* read 10+100 LAN_CFG regs */
-		newlan &= ~(HP100_VG_SEL << 16);
-		newlan |= (HP100_DOT3_MAC) << 8;
-		hp100_andb(~HP100_AUTO_MODE, MAC_CFG_3);	/* Autosel off */
-		hp100_outl(newlan, 10_LAN_CFG_1);
-
-		/* Conditionally stall for 5sec on VG selected. */
-		time = jiffies + (HZ * 5);
-		do {
-			if (!(hp100_inb(MAC_CFG_4) & HP100_MAC_SEL_ST))
-				break;
-			if (!in_interrupt())
-				schedule_timeout_interruptible(1);
-		} while (time_after(time, jiffies));
-
-		hp100_orb(HP100_AUTO_MODE, MAC_CFG_3);	/* Autosel back on */
-		hp100_outl(savelan, 10_LAN_CFG_1);
-	}
-
-	time = jiffies + (3 * HZ);	/* Timeout 3s */
-	do {
-		if ((hp100_inb(VG_LAN_CFG_1) & HP100_LINK_CABLE_ST) == 0)
-			break;
-		if (!in_interrupt())
-			schedule_timeout_interruptible(1);
-	} while (time_after(time, jiffies));
-
-	if (time_before_eq(time, jiffies)) {
-#ifdef HP100_DEBUG
-		printk("hp100: %s: down_vg_link: timeout\n", dev->name);
-#endif
-		return -EIO;
-	}
-
-	time = jiffies + (2 * HZ);	/* This seems to take a while.... */
-	do {
-		if (!in_interrupt())
-			schedule_timeout_interruptible(1);
-	} while (time_after(time, jiffies));
-
-	return 0;
-}
-
-static int hp100_login_to_vg_hub(struct net_device *dev, u_short force_relogin)
-{
-	int ioaddr = dev->base_addr;
-	struct hp100_private *lp = netdev_priv(dev);
-	u_short val = 0;
-	unsigned long time;
-	int startst;
-
-#ifdef HP100_DEBUG_B
-	hp100_outw(0x4225, TRACE);
-	printk("hp100: %s: login_to_vg_hub\n", dev->name);
-#endif
-
-	/* Initiate a login sequence iff VG MAC is enabled and either Load Address
-	 * bit is zero or the force relogin flag is set (e.g. due to MAC address or
-	 * promiscuous mode change)
-	 */
-	hp100_page(MAC_CTRL);
-	startst = hp100_inb(VG_LAN_CFG_1);
-	if ((force_relogin == 1) || (hp100_inb(MAC_CFG_4) & HP100_MAC_SEL_ST)) {
-#ifdef HP100_DEBUG_TRAINING
-		printk("hp100: %s: Start training\n", dev->name);
-#endif
-
-		/* Ensure VG Reset bit is 1 (i.e., do not reset) */
-		hp100_orb(HP100_VG_RESET, VG_LAN_CFG_1);
-
-		/* If Lassen AND auto-select-mode AND VG tones were sensed on */
-		/* entry then temporarily put them into force 100Mbit mode */
-		if ((lp->chip == HP100_CHIPID_LASSEN) && (startst & HP100_LINK_CABLE_ST))
-			hp100_andb(~HP100_DOT3_MAC, 10_LAN_CFG_2);
-
-		/* Drop the VG link by zeroing Link Up Command and Load Address  */
-		hp100_andb(~(HP100_LINK_CMD /* |HP100_LOAD_ADDR */ ), VG_LAN_CFG_1);
-
-#ifdef HP100_DEBUG_TRAINING
-		printk("hp100: %s: Bring down the link\n", dev->name);
-#endif
-
-		/* Wait for link to drop */
-		time = jiffies + (HZ / 10);
-		do {
-			if (~(hp100_inb(VG_LAN_CFG_1) & HP100_LINK_UP_ST))
-				break;
-			if (!in_interrupt())
-				schedule_timeout_interruptible(1);
-		} while (time_after(time, jiffies));
-
-		/* Start an addressed training and optionally request promiscuous port */
-		if ((dev->flags) & IFF_PROMISC) {
-			hp100_orb(HP100_PROM_MODE, VG_LAN_CFG_2);
-			if (lp->chip == HP100_CHIPID_LASSEN)
-				hp100_orw(HP100_MACRQ_PROMSC, TRAIN_REQUEST);
-		} else {
-			hp100_andb(~HP100_PROM_MODE, VG_LAN_CFG_2);
-			/* For ETR parts we need to reset the prom. bit in the training
-			 * register, otherwise promiscious mode won't be disabled.
-			 */
-			if (lp->chip == HP100_CHIPID_LASSEN) {
-				hp100_andw(~HP100_MACRQ_PROMSC, TRAIN_REQUEST);
-			}
-		}
-
-		/* With ETR parts, frame format request bits can be set. */
-		if (lp->chip == HP100_CHIPID_LASSEN)
-			hp100_orb(HP100_MACRQ_FRAMEFMT_EITHER, TRAIN_REQUEST);
-
-		hp100_orb(HP100_LINK_CMD | HP100_LOAD_ADDR | HP100_VG_RESET, VG_LAN_CFG_1);
-
-		/* Note: Next wait could be omitted for Hood and earlier chips under */
-		/* certain circumstances */
-		/* TODO: check if hood/earlier and skip wait. */
-
-		/* Wait for either short timeout for VG tones or long for login    */
-		/* Wait for the card hardware to signalise link cable status ok... */
-		hp100_page(MAC_CTRL);
-		time = jiffies + (1 * HZ);	/* 1 sec timeout for cable st */
-		do {
-			if (hp100_inb(VG_LAN_CFG_1) & HP100_LINK_CABLE_ST)
-				break;
-			if (!in_interrupt())
-				schedule_timeout_interruptible(1);
-		} while (time_before(jiffies, time));
-
-		if (time_after_eq(jiffies, time)) {
-#ifdef HP100_DEBUG_TRAINING
-			printk("hp100: %s: Link cable status not ok? Training aborted.\n", dev->name);
-#endif
-		} else {
-#ifdef HP100_DEBUG_TRAINING
-			printk
-			    ("hp100: %s: HUB tones detected. Trying to train.\n",
-			     dev->name);
-#endif
-
-			time = jiffies + (2 * HZ);	/* again a timeout */
-			do {
-				val = hp100_inb(VG_LAN_CFG_1);
-				if ((val & (HP100_LINK_UP_ST))) {
-#ifdef HP100_DEBUG_TRAINING
-					printk("hp100: %s: Passed training.\n", dev->name);
-#endif
-					break;
-				}
-				if (!in_interrupt())
-					schedule_timeout_interruptible(1);
-			} while (time_after(time, jiffies));
-		}
-
-		/* If LINK_UP_ST is set, then we are logged into the hub. */
-		if (time_before_eq(jiffies, time) && (val & HP100_LINK_UP_ST)) {
-#ifdef HP100_DEBUG_TRAINING
-			printk("hp100: %s: Successfully logged into the HUB.\n", dev->name);
-			if (lp->chip == HP100_CHIPID_LASSEN) {
-				val = hp100_inw(TRAIN_ALLOW);
-				printk("hp100: %s: Card supports 100VG MAC Version \"%s\" ",
-					     dev->name, (hp100_inw(TRAIN_REQUEST) & HP100_CARD_MACVER) ? "802.12" : "Pre");
-				printk("Driver will use MAC Version \"%s\"\n", (val & HP100_HUB_MACVER) ? "802.12" : "Pre");
-				printk("hp100: %s: Frame format is %s.\n", dev->name, (val & HP100_MALLOW_FRAMEFMT) ? "802.5" : "802.3");
-			}
-#endif
-		} else {
-			/* If LINK_UP_ST is not set, login was not successful */
-			printk("hp100: %s: Problem logging into the HUB.\n", dev->name);
-			if (lp->chip == HP100_CHIPID_LASSEN) {
-				/* Check allowed Register to find out why there is a problem. */
-				val = hp100_inw(TRAIN_ALLOW);	/* won't work on non-ETR card */
-#ifdef HP100_DEBUG_TRAINING
-				printk("hp100: %s: MAC Configuration requested: 0x%04x, HUB allowed: 0x%04x\n", dev->name, hp100_inw(TRAIN_REQUEST), val);
-#endif
-				if (val & HP100_MALLOW_ACCDENIED)
-					printk("hp100: %s: HUB access denied.\n", dev->name);
-				if (val & HP100_MALLOW_CONFIGURE)
-					printk("hp100: %s: MAC Configuration is incompatible with the Network.\n", dev->name);
-				if (val & HP100_MALLOW_DUPADDR)
-					printk("hp100: %s: Duplicate MAC Address on the Network.\n", dev->name);
-			}
-		}
-
-		/* If we have put the chip into forced 100 Mbit mode earlier, go back */
-		/* to auto-select mode */
-
-		if ((lp->chip == HP100_CHIPID_LASSEN) && (startst & HP100_LINK_CABLE_ST)) {
-			hp100_page(MAC_CTRL);
-			hp100_orb(HP100_DOT3_MAC, 10_LAN_CFG_2);
-		}
-
-		val = hp100_inb(VG_LAN_CFG_1);
-
-		/* Clear the MISC_ERROR Interrupt, which might be generated when doing the relogin */
-		hp100_page(PERFORMANCE);
-		hp100_outw(HP100_MISC_ERROR, IRQ_STATUS);
-
-		if (val & HP100_LINK_UP_ST)
-			return 0;	/* login was ok */
-		else {
-			printk("hp100: %s: Training failed.\n", dev->name);
-			hp100_down_vg_link(dev);
-			return -EIO;
-		}
-	}
-	/* no forced relogin & already link there->no training. */
-	return -EIO;
-}
-
-static void hp100_cascade_reset(struct net_device *dev, u_short enable)
-{
-	int ioaddr = dev->base_addr;
-	struct hp100_private *lp = netdev_priv(dev);
-
-#ifdef HP100_DEBUG_B
-	hp100_outw(0x4226, TRACE);
-	printk("hp100: %s: cascade_reset\n", dev->name);
-#endif
-
-	if (enable) {
-		hp100_outw(HP100_HW_RST | HP100_RESET_LB, OPTION_LSW);
-		if (lp->chip == HP100_CHIPID_LASSEN) {
-			/* Lassen requires a PCI transmit fifo reset */
-			hp100_page(HW_MAP);
-			hp100_andb(~HP100_PCI_RESET, PCICTRL2);
-			hp100_orb(HP100_PCI_RESET, PCICTRL2);
-			/* Wait for min. 300 ns */
-			/* we can't use jiffies here, because it may be */
-			/* that we have disabled the timer... */
-			udelay(400);
-			hp100_andb(~HP100_PCI_RESET, PCICTRL2);
-			hp100_page(PERFORMANCE);
-		}
-	} else {		/* bring out of reset */
-		hp100_outw(HP100_HW_RST | HP100_SET_LB, OPTION_LSW);
-		udelay(400);
-		hp100_page(PERFORMANCE);
-	}
-}
-
-#ifdef HP100_DEBUG
-void hp100_RegisterDump(struct net_device *dev)
-{
-	int ioaddr = dev->base_addr;
-	int Page;
-	int Register;
-
-	/* Dump common registers */
-	printk("hp100: %s: Cascade Register Dump\n", dev->name);
-	printk("hardware id #1: 0x%.2x\n", hp100_inb(HW_ID));
-	printk("hardware id #2/paging: 0x%.2x\n", hp100_inb(PAGING));
-	printk("option #1: 0x%.4x\n", hp100_inw(OPTION_LSW));
-	printk("option #2: 0x%.4x\n", hp100_inw(OPTION_MSW));
-
-	/* Dump paged registers */
-	for (Page = 0; Page < 8; Page++) {
-		/* Dump registers */
-		printk("page: 0x%.2x\n", Page);
-		outw(Page, ioaddr + 0x02);
-		for (Register = 0x8; Register < 0x22; Register += 2) {
-			/* Display Register contents except data port */
-			if (((Register != 0x10) && (Register != 0x12)) || (Page > 0)) {
-				printk("0x%.2x = 0x%.4x\n", Register, inw(ioaddr + Register));
-			}
-		}
-	}
-	hp100_page(PERFORMANCE);
-}
-#endif
-
-
-static void cleanup_dev(struct net_device *d)
-{
-	struct hp100_private *p = netdev_priv(d);
-
-	unregister_netdev(d);
-	release_region(d->base_addr, HP100_REGION_SIZE);
-
-	if (p->mode == 1)	/* busmaster */
-		pci_free_consistent(p->pci_dev, MAX_RINGSIZE + 0x0f,
-				    p->page_vaddr_algn,
-				    virt_to_whatever(d, p->page_vaddr_algn));
-	if (p->mem_ptr_virt)
-		iounmap(p->mem_ptr_virt);
-
-	free_netdev(d);
-}
-
-static int hp100_eisa_probe(struct device *gendev)
-{
-	struct net_device *dev = alloc_etherdev(sizeof(struct hp100_private));
-	struct eisa_device *edev = to_eisa_device(gendev);
-	int err;
-
-	if (!dev)
-		return -ENOMEM;
-
-	SET_NETDEV_DEV(dev, &edev->dev);
-
-	err = hp100_probe1(dev, edev->base_addr + 0xC38, HP100_BUS_EISA, NULL);
-	if (err)
-		goto out1;
-
-#ifdef HP100_DEBUG
-	printk("hp100: %s: EISA adapter found at 0x%x\n", dev->name,
-	       dev->base_addr);
-#endif
-	dev_set_drvdata(gendev, dev);
-	return 0;
- out1:
-	free_netdev(dev);
-	return err;
-}
-
-static int hp100_eisa_remove(struct device *gendev)
-{
-	struct net_device *dev = dev_get_drvdata(gendev);
-	cleanup_dev(dev);
-	return 0;
-}
-
-static struct eisa_driver hp100_eisa_driver = {
-        .id_table = hp100_eisa_tbl,
-        .driver   = {
-                .name    = "hp100",
-                .probe   = hp100_eisa_probe,
-		.remove  = hp100_eisa_remove,
-        }
-};
-
-static int hp100_pci_probe(struct pci_dev *pdev,
-			   const struct pci_device_id *ent)
-{
-	struct net_device *dev;
-	int ioaddr;
-	u_short pci_command;
-	int err;
-
-	if (pci_enable_device(pdev))
-		return -ENODEV;
-
-	dev = alloc_etherdev(sizeof(struct hp100_private));
-	if (!dev) {
-		err = -ENOMEM;
-		goto out0;
-	}
-
-	SET_NETDEV_DEV(dev, &pdev->dev);
-
-	pci_read_config_word(pdev, PCI_COMMAND, &pci_command);
-	if (!(pci_command & PCI_COMMAND_IO)) {
-#ifdef HP100_DEBUG
-		printk("hp100: %s: PCI I/O Bit has not been set. Setting...\n", dev->name);
-#endif
-		pci_command |= PCI_COMMAND_IO;
-		pci_write_config_word(pdev, PCI_COMMAND, pci_command);
-	}
-
-	if (!(pci_command & PCI_COMMAND_MASTER)) {
-#ifdef HP100_DEBUG
-		printk("hp100: %s: PCI Master Bit has not been set. Setting...\n", dev->name);
-#endif
-		pci_command |= PCI_COMMAND_MASTER;
-		pci_write_config_word(pdev, PCI_COMMAND, pci_command);
-	}
-
-	ioaddr = pci_resource_start(pdev, 0);
-	err = hp100_probe1(dev, ioaddr, HP100_BUS_PCI, pdev);
-	if (err)
-		goto out1;
-
-#ifdef HP100_DEBUG
-	printk("hp100: %s: PCI adapter found at 0x%x\n", dev->name, ioaddr);
-#endif
-	pci_set_drvdata(pdev, dev);
-	return 0;
- out1:
-	free_netdev(dev);
- out0:
-	pci_disable_device(pdev);
-	return err;
-}
-
-static void hp100_pci_remove(struct pci_dev *pdev)
-{
-	struct net_device *dev = pci_get_drvdata(pdev);
-
-	cleanup_dev(dev);
-	pci_disable_device(pdev);
-}
-
-
-static struct pci_driver hp100_pci_driver = {
-	.name		= "hp100",
-	.id_table	= hp100_pci_tbl,
-	.probe		= hp100_pci_probe,
-	.remove		= hp100_pci_remove,
-};
-
-/*
- *  module section
- */
-
-MODULE_LICENSE("GPL");
-MODULE_AUTHOR("Jaroslav Kysela <perex@perex.cz>, "
-              "Siegfried \"Frieder\" Loeffler (dg1sek) <floeff@mathematik.uni-stuttgart.de>");
-MODULE_DESCRIPTION("HP CASCADE Architecture Driver for 100VG-AnyLan Network Adapters");
-
-/*
- * Note: to register three isa devices, use:
- * option hp100 hp100_port=0,0,0
- *        to register one card at io 0x280 as eth239, use:
- * option hp100 hp100_port=0x280
- */
-#if defined(MODULE) && defined(CONFIG_ISA)
-#define HP100_DEVICES 5
-/* Parameters set by insmod */
-static int hp100_port[HP100_DEVICES] = { 0, [1 ... (HP100_DEVICES-1)] = -1 };
-module_param_hw_array(hp100_port, int, ioport, NULL, 0);
-
-/* List of devices */
-static struct net_device *hp100_devlist[HP100_DEVICES];
-
-static int __init hp100_isa_init(void)
-{
-	struct net_device *dev;
-	int i, err, cards = 0;
-
-	/* Don't autoprobe ISA bus */
-	if (hp100_port[0] == 0)
-		return -ENODEV;
-
-	/* Loop on all possible base addresses */
-	for (i = 0; i < HP100_DEVICES && hp100_port[i] != -1; ++i) {
-		dev = alloc_etherdev(sizeof(struct hp100_private));
-		if (!dev) {
-			while (cards > 0)
-				cleanup_dev(hp100_devlist[--cards]);
-
-			return -ENOMEM;
-		}
-
-		err = hp100_isa_probe(dev, hp100_port[i]);
-		if (!err)
-			hp100_devlist[cards++] = dev;
-		else
-			free_netdev(dev);
-	}
-
-	return cards > 0 ? 0 : -ENODEV;
-}
-
-static void hp100_isa_cleanup(void)
-{
-	int i;
-
-	for (i = 0; i < HP100_DEVICES; i++) {
-		struct net_device *dev = hp100_devlist[i];
-		if (dev)
-			cleanup_dev(dev);
-	}
-}
-#else
-#define hp100_isa_init()	(0)
-#define hp100_isa_cleanup()	do { } while(0)
-#endif
-
-static int __init hp100_module_init(void)
-{
-	int err;
-
-	err = hp100_isa_init();
-	if (err && err != -ENODEV)
-		goto out;
-	err = eisa_driver_register(&hp100_eisa_driver);
-	if (err && err != -ENODEV)
-		goto out2;
-	err = pci_register_driver(&hp100_pci_driver);
-	if (err && err != -ENODEV)
-		goto out3;
- out:
-	return err;
- out3:
-	eisa_driver_unregister (&hp100_eisa_driver);
- out2:
-	hp100_isa_cleanup();
-	goto out;
-}
-
-
-static void __exit hp100_module_exit(void)
-{
-	hp100_isa_cleanup();
-	eisa_driver_unregister (&hp100_eisa_driver);
-	pci_unregister_driver (&hp100_pci_driver);
-}
-
-module_init(hp100_module_init)
-module_exit(hp100_module_exit)
diff --git a/drivers/net/ethernet/hp/hp100.h b/drivers/net/ethernet/hp/hp100.h
deleted file mode 100644
index b60e96fe38b4..000000000000
--- a/drivers/net/ethernet/hp/hp100.h
+++ /dev/null
@@ -1,615 +0,0 @@
-/*
- * hp100.h: Hewlett Packard HP10/100VG ANY LAN ethernet driver for Linux.
- *
- * $Id: hp100.h,v 1.51 1997/04/08 14:26:42 floeff Exp floeff $
- *
- * Authors:  Jaroslav Kysela, <perex@pf.jcu.cz>
- *           Siegfried Loeffler <floeff@tunix.mathematik.uni-stuttgart.de>
- *
- * This driver is based on the 'hpfepkt' crynwr packet driver.
- *
- * This source/code is public free; you can distribute it and/or modify
- * it under terms of the GNU General Public License (published by the
- * Free Software Foundation) either version two of this License, or any
- * later version.
- */
-
-/****************************************************************************
- *  Hardware Constants
- ****************************************************************************/
-
-/*
- * Page Identifiers
- * (Swap Paging Register, PAGING, bits 3:0, Offset 0x02)
- */
-
-#define HP100_PAGE_PERFORMANCE	0x0	/* Page 0 */
-#define HP100_PAGE_MAC_ADDRESS	0x1	/* Page 1 */
-#define HP100_PAGE_HW_MAP	0x2	/* Page 2 */
-#define HP100_PAGE_EEPROM_CTRL	0x3	/* Page 3 */
-#define HP100_PAGE_MAC_CTRL	0x4	/* Page 4 */
-#define HP100_PAGE_MMU_CFG	0x5	/* Page 5 */
-#define HP100_PAGE_ID_MAC_ADDR	0x6	/* Page 6 */
-#define HP100_PAGE_MMU_POINTER	0x7	/* Page 7 */
-
-
-/* Registers that are present on all pages  */
-
-#define HP100_REG_HW_ID		0x00	/* R:  (16) Unique card ID           */
-#define HP100_REG_TRACE		0x00	/* W:  (16) Used for debug output    */
-#define HP100_REG_PAGING	0x02	/* R:  (16),15:4 Card ID             */
-					/* W:  (16),3:0 Switch pages         */
-#define HP100_REG_OPTION_LSW	0x04	/* RW: (16) Select card functions    */
-#define HP100_REG_OPTION_MSW	0x06	/* RW: (16) Select card functions    */
-
-/*  Page 0 - Performance  */
-
-#define HP100_REG_IRQ_STATUS	0x08	/* RW: (16) Which ints are pending   */
-#define HP100_REG_IRQ_MASK	0x0a	/* RW: (16) Select ints to allow     */
-#define HP100_REG_FRAGMENT_LEN	0x0c	/* W: (16)12:0 Current fragment len */
-/* Note: For 32 bit systems, fragment len and offset registers are available */
-/*       at offset 0x28 and 0x2c, where they can be written as 32bit values. */
-#define HP100_REG_OFFSET	0x0e	/* RW: (16)12:0 Offset to start read */
-#define HP100_REG_DATA32	0x10	/* RW: (32) I/O mode data port       */
-#define HP100_REG_DATA16	0x12	/* RW: WORDs must be read from here  */
-#define HP100_REG_TX_MEM_FREE	0x14	/* RD: (32) Amount of free Tx mem    */
-#define HP100_REG_TX_PDA_L      0x14	/* W: (32) BM: Ptr to PDL, Low Pri  */
-#define HP100_REG_TX_PDA_H      0x1c	/* W: (32) BM: Ptr to PDL, High Pri */
-#define HP100_REG_RX_PKT_CNT	0x18	/* RD: (8) Rx count of pkts on card  */
-#define HP100_REG_TX_PKT_CNT	0x19	/* RD: (8) Tx count of pkts on card  */
-#define HP100_REG_RX_PDL        0x1a	/* R: (8) BM: # rx pdl not executed */
-#define HP100_REG_TX_PDL        0x1b	/* R: (8) BM: # tx pdl not executed */
-#define HP100_REG_RX_PDA        0x18	/* W: (32) BM: Up to 31 addresses */
-					/*             which point to a PDL */
-#define HP100_REG_SL_EARLY      0x1c	/*    (32) Enhanced Slave Early Rx */
-#define HP100_REG_STAT_DROPPED  0x20	/* R (12) Dropped Packet Counter */
-#define HP100_REG_STAT_ERRORED  0x22	/* R (8) Errored Packet Counter */
-#define HP100_REG_STAT_ABORT    0x23	/* R (8) Abort Counter/OW Coll. Flag */
-#define HP100_REG_RX_RING       0x24	/* W (32) Slave: RX Ring Pointers */
-#define HP100_REG_32_FRAGMENT_LEN 0x28	/* W (13) Slave: Fragment Length Reg */
-#define HP100_REG_32_OFFSET     0x2c	/* W (16) Slave: Offset Register */
-
-/*  Page 1 - MAC Address/Hash Table  */
-
-#define HP100_REG_MAC_ADDR	0x08	/* RW: (8) Cards MAC address         */
-#define HP100_REG_HASH_BYTE0	0x10	/* RW: (8) Cards multicast filter    */
-
-/*  Page 2 - Hardware Mapping  */
-
-#define HP100_REG_MEM_MAP_LSW	0x08	/* RW: (16) LSW of cards mem addr    */
-#define HP100_REG_MEM_MAP_MSW	0x0a	/* RW: (16) MSW of cards mem addr    */
-#define HP100_REG_IO_MAP	0x0c	/* RW: (8) Cards I/O address         */
-#define HP100_REG_IRQ_CHANNEL	0x0d	/* RW: (8) IRQ and edge/level int    */
-#define HP100_REG_SRAM		0x0e	/* RW: (8) How much RAM on card      */
-#define HP100_REG_BM		0x0f	/* RW: (8) Controls BM functions     */
-
-/* New on Page 2 for ETR chips: */
-#define HP100_REG_MODECTRL1     0x10	/* RW: (8) Mode Control 1 */
-#define HP100_REG_MODECTRL2     0x11	/* RW: (8) Mode Control 2 */
-#define HP100_REG_PCICTRL1      0x12	/* RW: (8) PCI Cfg 1 */
-#define HP100_REG_PCICTRL2      0x13	/* RW: (8) PCI Cfg 2 */
-#define HP100_REG_PCIBUSMLAT    0x15	/* RW: (8) PCI Bus Master Latency */
-#define HP100_REG_EARLYTXCFG    0x16	/* RW: (16) Early TX Cfg/Cntrl Reg */
-#define HP100_REG_EARLYRXCFG    0x18	/* RW: (8) Early RX Cfg/Cntrl Reg */
-#define HP100_REG_ISAPNPCFG1    0x1a	/* RW: (8) ISA PnP Cfg/Cntrl Reg 1 */
-#define HP100_REG_ISAPNPCFG2    0x1b	/* RW: (8) ISA PnP Cfg/Cntrl Reg 2 */
-
-/*  Page 3 - EEPROM/Boot ROM  */
-
-#define HP100_REG_EEPROM_CTRL	0x08	/* RW: (16) Used to load EEPROM      */
-#define HP100_REG_BOOTROM_CTRL  0x0a
-
-/*  Page 4 - LAN Configuration  (MAC_CTRL) */
-
-#define HP100_REG_10_LAN_CFG_1	0x08	/* RW: (8) Set 10M XCVR functions   */
-#define HP100_REG_10_LAN_CFG_2  0x09	/* RW: (8)     10M XCVR functions   */
-#define HP100_REG_VG_LAN_CFG_1	0x0a	/* RW: (8) Set 100M XCVR functions  */
-#define HP100_REG_VG_LAN_CFG_2  0x0b	/* RW: (8) 100M LAN Training cfgregs */
-#define HP100_REG_MAC_CFG_1	0x0c	/* RW: (8) Types of pkts to accept   */
-#define HP100_REG_MAC_CFG_2	0x0d	/* RW: (8) Misc MAC functions        */
-#define HP100_REG_MAC_CFG_3     0x0e	/* RW: (8) Misc MAC functions */
-#define HP100_REG_MAC_CFG_4     0x0f	/* R:  (8) Misc MAC states */
-#define HP100_REG_DROPPED	0x10	/* R:  (16),11:0 Pkts can't fit in mem */
-#define HP100_REG_CRC		0x12	/* R:  (8) Pkts with CRC             */
-#define HP100_REG_ABORT		0x13	/* R:  (8) Aborted Tx pkts           */
-#define HP100_REG_TRAIN_REQUEST 0x14	/* RW: (16) Endnode MAC register. */
-#define HP100_REG_TRAIN_ALLOW   0x16	/* R:  (16) Hub allowed register */
-
-/*  Page 5 - MMU  */
-
-#define HP100_REG_RX_MEM_STOP	0x0c	/* RW: (16) End of Rx ring addr      */
-#define HP100_REG_TX_MEM_STOP	0x0e	/* RW: (16) End of Tx ring addr      */
-#define HP100_REG_PDL_MEM_STOP  0x10	/* Not used by 802.12 devices */
-#define HP100_REG_ECB_MEM_STOP  0x14	/* I've no idea what this is */
-
-/*  Page 6 - Card ID/Physical LAN Address  */
-
-#define HP100_REG_BOARD_ID	0x08	/* R:  (8) EISA/ISA card ID          */
-#define HP100_REG_BOARD_IO_CHCK 0x0c	/* R:  (8) Added to ID to get FFh    */
-#define HP100_REG_SOFT_MODEL	0x0d	/* R:  (8) Config program defined    */
-#define HP100_REG_LAN_ADDR	0x10	/* R:  (8) MAC addr of card          */
-#define HP100_REG_LAN_ADDR_CHCK 0x16	/* R:  (8) Added to addr to get FFh  */
-
-/*  Page 7 - MMU Current Pointers  */
-
-#define HP100_REG_PTR_RXSTART	0x08	/* R:  (16) Current begin of Rx ring */
-#define HP100_REG_PTR_RXEND	0x0a	/* R:  (16) Current end of Rx ring   */
-#define HP100_REG_PTR_TXSTART	0x0c	/* R:  (16) Current begin of Tx ring */
-#define HP100_REG_PTR_TXEND	0x0e	/* R:  (16) Current end of Rx ring   */
-#define HP100_REG_PTR_RPDLSTART 0x10
-#define HP100_REG_PTR_RPDLEND   0x12
-#define HP100_REG_PTR_RINGPTRS  0x14
-#define HP100_REG_PTR_MEMDEBUG  0x1a
-/* ------------------------------------------------------------------------ */
-
-
-/*
- * Hardware ID Register I (Always available, HW_ID, Offset 0x00)
- */
-#define HP100_HW_ID_CASCADE     0x4850	/* Identifies Cascade Chip */
-
-/*
- * Hardware ID Register 2 & Paging Register
- * (Always available, PAGING, Offset 0x02)
- * Bits 15:4 are for the Chip ID
- */
-#define HP100_CHIPID_MASK        0xFFF0
-#define HP100_CHIPID_SHASTA      0x5350	/* Not 802.12 compliant */
-					 /* EISA BM/SL, MCA16/32 SL, ISA SL */
-#define HP100_CHIPID_RAINIER     0x5360	/* Not 802.12 compliant EISA BM, */
-					 /* PCI SL, MCA16/32 SL, ISA SL */
-#define HP100_CHIPID_LASSEN      0x5370	/* 802.12 compliant PCI BM, PCI SL */
-					 /* LRF supported */
-
-/*
- *  Option Registers I and II
- * (Always available, OPTION_LSW, Offset 0x04-0x05)
- */
-#define HP100_DEBUG_EN		0x8000	/* 0:Dis., 1:Enable Debug Dump Ptr. */
-#define HP100_RX_HDR		0x4000	/* 0:Dis., 1:Enable putting pkt into */
-					/*   system mem. before Rx interrupt */
-#define HP100_MMAP_DIS		0x2000	/* 0:Enable, 1:Disable mem.mapping. */
-					/*   MMAP_DIS must be 0 and MEM_EN */
-					/*   must be 1 for memory-mapped */
-					/*   mode to be enabled */
-#define HP100_EE_EN		0x1000	/* 0:Disable,1:Enable EEPROM writing */
-#define HP100_BM_WRITE		0x0800	/* 0:Slave, 1:Bus Master for Tx data */
-#define HP100_BM_READ		0x0400	/* 0:Slave, 1:Bus Master for Rx data */
-#define HP100_TRI_INT		0x0200	/* 0:Don't, 1:Do tri-state the int */
-#define HP100_MEM_EN		0x0040	/* Config program set this to */
-					/*   0:Disable, 1:Enable mem map. */
-					/*   See MMAP_DIS. */
-#define HP100_IO_EN		0x0020	/* 1:Enable I/O transfers */
-#define HP100_BOOT_EN		0x0010	/* 1:Enable boot ROM access */
-#define HP100_FAKE_INT		0x0008	/* 1:int */
-#define HP100_INT_EN		0x0004	/* 1:Enable ints from card */
-#define HP100_HW_RST		0x0002	/* 0:Reset, 1:Out of reset */
-					/* NIC reset on 0 to 1 transition */
-
-/*
- *  Option Register III
- * (Always available, OPTION_MSW, Offset 0x06)
- */
-#define HP100_PRIORITY_TX	0x0080	/* 1:Do all Tx pkts as priority */
-#define HP100_EE_LOAD		0x0040	/* 1:EEPROM loading, 0 when done */
-#define HP100_ADV_NXT_PKT	0x0004	/* 1:Advance to next pkt in Rx queue */
-					/*   h/w will set to 0 when done */
-#define HP100_TX_CMD		0x0002	/* 1:Tell h/w download done, h/w */
-					/*   will set to 0 when done */
-
-/*
- * Interrupt Status Registers I and II
- * (Page PERFORMANCE, IRQ_STATUS, Offset 0x08-0x09)
- * Note: With old chips, these Registers will clear when 1 is written to them
- *       with new chips this depends on setting of CLR_ISMODE
- */
-#define HP100_RX_EARLY_INT      0x2000
-#define HP100_RX_PDA_ZERO       0x1000
-#define HP100_RX_PDL_FILL_COMPL 0x0800
-#define HP100_RX_PACKET		0x0400	/* 0:No, 1:Yes pkt has been Rx */
-#define HP100_RX_ERROR		0x0200	/* 0:No, 1:Yes Rx pkt had error */
-#define HP100_TX_PDA_ZERO       0x0020	/* 1 when PDA count goes to zero */
-#define HP100_TX_SPACE_AVAIL	0x0010	/* 0:<8192, 1:>=8192 Tx free bytes */
-#define HP100_TX_COMPLETE	0x0008	/* 0:No, 1:Yes a Tx has completed */
-#define HP100_MISC_ERROR        0x0004	/* 0:No, 1:Lan Link down or bus error */
-#define HP100_TX_ERROR		0x0002	/* 0:No, 1:Yes Tx pkt had error */
-
-/*
- * Xmit Memory Free Count
- * (Page PERFORMANCE, TX_MEM_FREE, Offset 0x14) (Read only, 32bit)
- */
-#define HP100_AUTO_COMPARE	0x80000000	/* Tx Space avail & pkts<255 */
-#define HP100_FREE_SPACE	0x7fffffe0	/* Tx free memory */
-
-/*
- *  IRQ Channel
- * (Page HW_MAP, IRQ_CHANNEL, Offset 0x0d)
- */
-#define HP100_ZERO_WAIT_EN	0x80	/* 0:No, 1:Yes asserts NOWS signal */
-#define HP100_IRQ_SCRAMBLE      0x40
-#define HP100_BOND_HP           0x20
-#define HP100_LEVEL_IRQ		0x10	/* 0:Edge, 1:Level type interrupts. */
-					/* (Only valid on EISA cards) */
-#define HP100_IRQMASK		0x0F	/* Isolate the IRQ bits */
-
-/*
- * SRAM Parameters
- * (Page HW_MAP, SRAM, Offset 0x0e)
- */
-#define HP100_RAM_SIZE_MASK	0xe0	/* AND to get SRAM size index */
-#define HP100_RAM_SIZE_SHIFT	0x05	/* Shift count(put index in lwr bits) */
-
-/*
- * Bus Master Register
- * (Page HW_MAP, BM, Offset 0x0f)
- */
-#define HP100_BM_BURST_RD       0x01	/* EISA only: 1=Use burst trans. fm system */
-					/* memory to chip (tx) */
-#define HP100_BM_BURST_WR       0x02	/* EISA only: 1=Use burst trans. fm system */
-					/* memory to chip (rx) */
-#define HP100_BM_MASTER		0x04	/* 0:Slave, 1:BM mode */
-#define HP100_BM_PAGE_CK        0x08	/* This bit should be set whenever in */
-					/* an EISA system */
-#define HP100_BM_PCI_8CLK       0x40	/* ... cycles 8 clocks apart */
-
-
-/*
- * Mode Control Register I
- * (Page HW_MAP, MODECTRL1, Offset0x10)
- */
-#define HP100_TX_DUALQ          0x10
-   /* If set and BM -> dual tx pda queues */
-#define HP100_ISR_CLRMODE       0x02	/* If set ISR will clear all pending */
-				       /* interrupts on read (etr only?) */
-#define HP100_EE_NOLOAD         0x04	/* Status whether res will be loaded */
-				       /* from the eeprom */
-#define HP100_TX_CNT_FLG        0x08	/* Controls Early TX Reg Cnt Field */
-#define HP100_PDL_USE3          0x10	/* If set BM engine will read only */
-				       /* first three data elements of a PDL */
-				       /* on the first access. */
-#define HP100_BUSTYPE_MASK      0xe0	/* Three bit bus type info */
-
-/*
- * Mode Control Register II
- * (Page HW_MAP, MODECTRL2, Offset0x11)
- */
-#define HP100_EE_MASK           0x0f	/* Tell EEPROM circuit not to load */
-				       /* certain resources */
-#define HP100_DIS_CANCEL        0x20	/* For tx dualq mode operation */
-#define HP100_EN_PDL_WB         0x40	/* 1: Status of PDL completion may be */
-				       /* written back to system mem */
-#define HP100_EN_BUS_FAIL       0x80	/* Enables bus-fail portion of misc */
-				       /* interrupt */
-
-/*
- * PCI Configuration and Control Register I
- * (Page HW_MAP, PCICTRL1, Offset 0x12)
- */
-#define HP100_LO_MEM            0x01	/* 1: Mapped Mem requested below 1MB */
-#define HP100_NO_MEM            0x02	/* 1: Disables Req for sysmem to PCI */
-				       /* bios */
-#define HP100_USE_ISA           0x04	/* 1: isa type decodes will occur */
-				       /* simultaneously with PCI decodes */
-#define HP100_IRQ_HI_MASK       0xf0	/* pgmed by pci bios */
-#define HP100_PCI_IRQ_HI_MASK   0x78	/* Isolate 4 bits for PCI IRQ  */
-
-/*
- * PCI Configuration and Control Register II
- * (Page HW_MAP, PCICTRL2, Offset 0x13)
- */
-#define HP100_RD_LINE_PDL       0x01	/* 1: PCI command Memory Read Line en */
-#define HP100_RD_TX_DATA_MASK   0x06	/* choose PCI memread cmds for TX */
-#define HP100_MWI               0x08	/* 1: en. PCI memory write invalidate */
-#define HP100_ARB_MODE          0x10	/* Select PCI arbitor type */
-#define HP100_STOP_EN           0x20	/* Enables PCI state machine to issue */
-				       /* pci stop if cascade not ready */
-#define HP100_IGNORE_PAR        0x40	/* 1: PCI state machine ignores parity */
-#define HP100_PCI_RESET         0x80	/* 0->1: Reset PCI block */
-
-/*
- * Early TX Configuration and Control Register
- * (Page HW_MAP, EARLYTXCFG, Offset 0x16)
- */
-#define HP100_EN_EARLY_TX       0x8000	/* 1=Enable Early TX */
-#define HP100_EN_ADAPTIVE       0x4000	/* 1=Enable adaptive mode */
-#define HP100_EN_TX_UR_IRQ      0x2000	/* reserved, must be 0 */
-#define HP100_EN_LOW_TX         0x1000	/* reserved, must be 0 */
-#define HP100_ET_CNT_MASK       0x0fff	/* bits 11..0: ET counters */
-
-/*
- * Early RX Configuration and Control Register
- * (Page HW_MAP, EARLYRXCFG, Offset 0x18)
- */
-#define HP100_EN_EARLY_RX       0x80	/* 1=Enable Early RX */
-#define HP100_EN_LOW_RX         0x40	/* reserved, must be 0 */
-#define HP100_RX_TRIP_MASK      0x1f	/* bits 4..0: threshold at which the
-					 * early rx circuit will start the
-					 * dma of received packet into system
-					 * memory for BM */
-
-/*
- *  Serial Devices Control Register
- * (Page EEPROM_CTRL, EEPROM_CTRL, Offset 0x08)
- */
-#define HP100_EEPROM_LOAD	0x0001	/* 0->1 loads EEPROM into registers. */
-					/* When it goes back to 0, load is   */
-					/* complete. This should take ~600us. */
-
-/*
- * 10MB LAN Control and Configuration Register I
- * (Page MAC_CTRL, 10_LAN_CFG_1, Offset 0x08)
- */
-#define HP100_MAC10_SEL		0xc0	/* Get bits to indicate MAC */
-#define HP100_AUI_SEL		0x20	/* Status of AUI selection */
-#define HP100_LOW_TH		0x10	/* 0:No, 1:Yes allow better cabling */
-#define HP100_LINK_BEAT_DIS	0x08	/* 0:Enable, 1:Disable link beat */
-#define HP100_LINK_BEAT_ST	0x04	/* 0:No, 1:Yes link beat being Rx */
-#define HP100_R_ROL_ST		0x02	/* 0:No, 1:Yes Rx twisted pair has */
-					/*             been reversed */
-#define HP100_AUI_ST		0x01	/* 0:No, 1:Yes use AUI on TP card */
-
-/*
- * 10 MB LAN Control and Configuration Register II
- * (Page MAC_CTRL, 10_LAN_CFG_2, Offset 0x09)
- */
-#define HP100_SQU_ST		0x01	/* 0:No, 1:Yes collision signal sent */
-					/*       after Tx.Only used for AUI. */
-#define HP100_FULLDUP           0x02	/* 1: LXT901 XCVR fullduplx enabled */
-#define HP100_DOT3_MAC          0x04	/* 1: DOT 3 Mac sel. unless Autosel */
-
-/*
- * MAC Selection, use with MAC10_SEL bits
- */
-#define HP100_AUTO_SEL_10	0x0	/* Auto select */
-#define HP100_XCVR_LXT901_10	0x1	/* LXT901 10BaseT transceiver */
-#define HP100_XCVR_7213		0x2	/* 7213 transceiver */
-#define HP100_XCVR_82503	0x3	/* 82503 transceiver */
-
-/*
- *  100MB LAN Training Register
- * (Page MAC_CTRL, VG_LAN_CFG_2, Offset 0x0b) (old, pre 802.12)
- */
-#define HP100_FRAME_FORMAT	0x08	/* 0:802.3, 1:802.5 frames */
-#define HP100_BRIDGE		0x04	/* 0:No, 1:Yes tell hub i am a bridge */
-#define HP100_PROM_MODE		0x02	/* 0:No, 1:Yes tell hub card is */
-					/*         promiscuous */
-#define HP100_REPEATER		0x01	/* 0:No, 1:Yes tell hub MAC wants to */
-					/*         be a cascaded repeater */
-
-/*
- * 100MB LAN Control and Configuration Register
- * (Page MAC_CTRL, VG_LAN_CFG_1, Offset 0x0a)
- */
-#define HP100_VG_SEL	        0x80	/* 0:No, 1:Yes use 100 Mbit MAC */
-#define HP100_LINK_UP_ST	0x40	/* 0:No, 1:Yes endnode logged in */
-#define HP100_LINK_CABLE_ST	0x20	/* 0:No, 1:Yes cable can hear tones */
-					/*         from  hub */
-#define HP100_LOAD_ADDR		0x10	/* 0->1 card addr will be sent  */
-					/* 100ms later the link status  */
-					/* bits are valid */
-#define HP100_LINK_CMD		0x08	/* 0->1 link will attempt to log in. */
-					/* 100ms later the link status */
-					/* bits are valid */
-#define HP100_TRN_DONE          0x04	/* NEW ETR-Chips only: Will be reset */
-					/* after LinkUp Cmd is given and set */
-					/* when training has completed. */
-#define HP100_LINK_GOOD_ST	0x02	/* 0:No, 1:Yes cable passed training */
-#define HP100_VG_RESET		0x01	/* 0:Yes, 1:No reset the 100VG MAC */
-
-
-/*
- *  MAC Configuration Register I
- * (Page MAC_CTRL, MAC_CFG_1, Offset 0x0c)
- */
-#define HP100_RX_IDLE		0x80	/* 0:Yes, 1:No currently receiving pkts */
-#define HP100_TX_IDLE		0x40	/* 0:Yes, 1:No currently Txing pkts */
-#define HP100_RX_EN		0x20	/* 1: allow receiving of pkts */
-#define HP100_TX_EN		0x10	/* 1: allow transmitting of pkts */
-#define HP100_ACC_ERRORED	0x08	/* 0:No, 1:Yes allow Rx of errored pkts */
-#define HP100_ACC_MC		0x04	/* 0:No, 1:Yes allow Rx of multicast pkts */
-#define HP100_ACC_BC		0x02	/* 0:No, 1:Yes allow Rx of broadcast pkts */
-#define HP100_ACC_PHY		0x01	/* 0:No, 1:Yes allow Rx of ALL phys. pkts */
-#define HP100_MAC1MODEMASK	0xf0	/* Hide ACC bits */
-#define HP100_MAC1MODE1		0x00	/* Receive nothing, must also disable RX */
-#define HP100_MAC1MODE2		0x00
-#define HP100_MAC1MODE3		HP100_MAC1MODE2 | HP100_ACC_BC
-#define HP100_MAC1MODE4		HP100_MAC1MODE3 | HP100_ACC_MC
-#define HP100_MAC1MODE5		HP100_MAC1MODE4	/* set mc hash to all ones also */
-#define HP100_MAC1MODE6		HP100_MAC1MODE5 | HP100_ACC_PHY	/* Promiscuous */
-/* Note MODE6 will receive all GOOD packets on the LAN. This really needs
-   a mode 7 defined to be LAN Analyzer mode, which will receive errored and
-   runt packets, and keep the CRC bytes. */
-#define HP100_MAC1MODE7		HP100_MAC1MODE6 | HP100_ACC_ERRORED
-
-/*
- *  MAC Configuration Register II
- * (Page MAC_CTRL, MAC_CFG_2, Offset 0x0d)
- */
-#define HP100_TR_MODE		0x80	/* 0:No, 1:Yes support Token Ring formats */
-#define HP100_TX_SAME		0x40	/* 0:No, 1:Yes Tx same packet continuous */
-#define HP100_LBK_XCVR		0x20	/* 0:No, 1:Yes loopback through MAC & */
-					/*   transceiver */
-#define HP100_LBK_MAC		0x10	/* 0:No, 1:Yes loopback through MAC */
-#define HP100_CRC_I		0x08	/* 0:No, 1:Yes inhibit CRC on Tx packets */
-#define HP100_ACCNA             0x04	/* 1: For 802.5: Accept only token ring
-					 * group addr that maches NA mask */
-#define HP100_KEEP_CRC		0x02	/* 0:No, 1:Yes keep CRC on Rx packets. */
-					/*   The length will reflect this. */
-#define HP100_ACCFA             0x01	/* 1: For 802.5: Accept only functional
-					 * addrs that match FA mask (page1) */
-#define HP100_MAC2MODEMASK	0x02
-#define HP100_MAC2MODE1		0x00
-#define HP100_MAC2MODE2		0x00
-#define HP100_MAC2MODE3		0x00
-#define HP100_MAC2MODE4		0x00
-#define HP100_MAC2MODE5		0x00
-#define HP100_MAC2MODE6		0x00
-#define HP100_MAC2MODE7		KEEP_CRC
-
-/*
- * MAC Configuration Register III
- * (Page MAC_CTRL, MAC_CFG_3, Offset 0x0e)
- */
-#define HP100_PACKET_PACE       0x03	/* Packet Pacing:
-					 * 00: No packet pacing
-					 * 01: 8 to 16 uS delay
-					 * 10: 16 to 32 uS delay
-					 * 11: 32 to 64 uS delay
-					 */
-#define HP100_LRF_EN            0x04	/* 1: External LAN Rcv Filter and
-					 * TCP/IP Checksumming enabled. */
-#define HP100_AUTO_MODE         0x10	/* 1: AutoSelect between 10/100 */
-
-/*
- * MAC Configuration Register IV
- * (Page MAC_CTRL, MAC_CFG_4, Offset 0x0f)
- */
-#define HP100_MAC_SEL_ST        0x01	/* (R): Status of external VGSEL
-					 * Signal, 1=100VG, 0=10Mbit sel. */
-#define HP100_LINK_FAIL_ST      0x02	/* (R): Status of Link Fail portion
-					 * of the Misc. Interrupt */
-
-/*
- *  100 MB LAN Training Request/Allowed Registers
- * (Page MAC_CTRL, TRAIN_REQUEST and TRAIN_ALLOW, Offset 0x14-0x16)(ETR parts only)
- */
-#define HP100_MACRQ_REPEATER         0x0001	/* 1: MAC tells HUB it wants to be
-						 *    a cascaded repeater
-						 * 0: ... wants to be a DTE */
-#define HP100_MACRQ_PROMSC           0x0006	/* 2 bits: Promiscious mode
-						 * 00: Rcv only unicast packets
-						 *     specifically addr to this
-						 *     endnode
-						 * 10: Rcv all pckts fwded by
-						 *     the local repeater */
-#define HP100_MACRQ_FRAMEFMT_EITHER  0x0018	/* 11: either format allowed */
-#define HP100_MACRQ_FRAMEFMT_802_3   0x0000	/* 00: 802.3 is requested */
-#define HP100_MACRQ_FRAMEFMT_802_5   0x0010	/* 10: 802.5 format is requested */
-#define HP100_CARD_MACVER            0xe000	/* R: 3 bit Cards 100VG MAC version */
-#define HP100_MALLOW_REPEATER        0x0001	/* If reset, requested access as an
-						 * end node is allowed */
-#define HP100_MALLOW_PROMSC          0x0004	/* 2 bits: Promiscious mode
-						 * 00: Rcv only unicast packets
-						 *     specifically addr to this
-						 *     endnode
-						 * 10: Rcv all pckts fwded by
-						 *     the local repeater */
-#define HP100_MALLOW_FRAMEFMT        0x00e0	/* 2 bits: Frame Format
-						 * 00: 802.3 format will be used
-						 * 10: 802.5 format will be used */
-#define HP100_MALLOW_ACCDENIED       0x0400	/* N bit */
-#define HP100_MALLOW_CONFIGURE       0x0f00	/* C bit */
-#define HP100_MALLOW_DUPADDR         0x1000	/* D bit */
-#define HP100_HUB_MACVER             0xe000	/* R: 3 bit 802.12 MAC/RMAC training */
-					     /*    protocol of repeater */
-
-/* ****************************************************************************** */
-
-/*
- *  Set/Reset bits
- */
-#define HP100_SET_HB		0x0100	/* 0:Set fields to 0 whose mask is 1 */
-#define HP100_SET_LB		0x0001	/* HB sets upper byte, LB sets lower byte */
-#define HP100_RESET_HB		0x0000	/* For readability when resetting bits */
-#define HP100_RESET_LB		0x0000	/* For readability when resetting bits */
-
-/*
- *  Misc. Constants
- */
-#define HP100_LAN_100		100	/* lan_type value for VG */
-#define HP100_LAN_10		10	/* lan_type value for 10BaseT */
-#define HP100_LAN_COAX		9	/* lan_type value for Coax */
-#define HP100_LAN_ERR		(-1)	/* lan_type value for link down */
-
-/*
- * Bus Master Data Structures  ----------------------------------------------
- */
-
-#define MAX_RX_PDL              30	/* Card limit = 31 */
-#define MAX_RX_FRAG             2	/* Don't need more... */
-#define MAX_TX_PDL              29
-#define MAX_TX_FRAG             2	/* Limit = 31 */
-
-/* Define total PDL area size in bytes (should be 4096) */
-/* This is the size of kernel (dma) memory that will be allocated. */
-#define MAX_RINGSIZE ((MAX_RX_FRAG*8+4+4)*MAX_RX_PDL+(MAX_TX_FRAG*8+4+4)*MAX_TX_PDL)+16
-
-/* Ethernet Packet Sizes */
-#define MIN_ETHER_SIZE          60
-#define MAX_ETHER_SIZE          1514	/* Needed for preallocation of */
-					/* skb buffer when busmastering */
-
-/* Tx or Rx Ring Entry */
-typedef struct hp100_ring {
-	u_int *pdl;		/* Address of PDLs PDH, dword before
-				 * this address is used for rx hdr */
-	u_int pdl_paddr;	/* Physical address of PDL */
-	struct sk_buff *skb;
-	struct hp100_ring *next;
-} hp100_ring_t;
-
-
-
-/* Mask for Header Descriptor */
-#define HP100_PKT_LEN_MASK	0x1FFF	/* AND with RxLength to get length */
-
-
-/* Receive Packet Status.  Note, the error bits are only valid if ACC_ERRORED
-   bit in the MAC Configuration Register 1 is set. */
-#define HP100_RX_PRI		0x8000	/* 0:No, 1:Yes packet is priority */
-#define HP100_SDF_ERR		0x4000	/* 0:No, 1:Yes start of frame error */
-#define HP100_SKEW_ERR		0x2000	/* 0:No, 1:Yes skew out of range */
-#define HP100_BAD_SYMBOL_ERR	0x1000	/* 0:No, 1:Yes invalid symbol received */
-#define HP100_RCV_IPM_ERR	0x0800	/* 0:No, 1:Yes pkt had an invalid packet */
-					/*   marker */
-#define HP100_SYMBOL_BAL_ERR	0x0400	/* 0:No, 1:Yes symbol balance error */
-#define HP100_VG_ALN_ERR	0x0200	/* 0:No, 1:Yes non-octet received */
-#define HP100_TRUNC_ERR		0x0100	/* 0:No, 1:Yes the packet was truncated */
-#define HP100_RUNT_ERR		0x0040	/* 0:No, 1:Yes pkt length < Min Pkt */
-					/*   Length Reg. */
-#define HP100_ALN_ERR		0x0010	/* 0:No, 1:Yes align error. */
-#define HP100_CRC_ERR		0x0008	/* 0:No, 1:Yes CRC occurred. */
-
-/* The last three bits indicate the type of destination address */
-
-#define HP100_MULTI_ADDR_HASH	0x0006	/* 110: Addr multicast, matched hash */
-#define HP100_BROADCAST_ADDR	0x0003	/* x11: Addr broadcast */
-#define HP100_MULTI_ADDR_NO_HASH 0x0002	/* 010: Addr multicast, didn't match hash */
-#define HP100_PHYS_ADDR_MATCH	0x0001	/* x01: Addr was physical and mine */
-#define HP100_PHYS_ADDR_NO_MATCH 0x0000	/* x00: Addr was physical but not mine */
-
-/*
- *  macros
- */
-
-#define hp100_inb( reg ) \
-        inb( ioaddr + HP100_REG_##reg )
-#define hp100_inw( reg ) \
-	inw( ioaddr + HP100_REG_##reg )
-#define hp100_inl( reg ) \
-	inl( ioaddr + HP100_REG_##reg )
-#define hp100_outb( data, reg ) \
-	outb( data, ioaddr + HP100_REG_##reg )
-#define hp100_outw( data, reg ) \
-	outw( data, ioaddr + HP100_REG_##reg )
-#define hp100_outl( data, reg ) \
-	outl( data, ioaddr + HP100_REG_##reg )
-#define hp100_orb( data, reg ) \
-	outb( inb( ioaddr + HP100_REG_##reg ) | (data), ioaddr + HP100_REG_##reg )
-#define hp100_orw( data, reg ) \
-	outw( inw( ioaddr + HP100_REG_##reg ) | (data), ioaddr + HP100_REG_##reg )
-#define hp100_andb( data, reg ) \
-	outb( inb( ioaddr + HP100_REG_##reg ) & (data), ioaddr + HP100_REG_##reg )
-#define hp100_andw( data, reg ) \
-	outw( inw( ioaddr + HP100_REG_##reg ) & (data), ioaddr + HP100_REG_##reg )
-
-#define hp100_page( page ) \
-	outw( HP100_PAGE_##page, ioaddr + HP100_REG_PAGING )
-#define hp100_ints_off() \
-	outw( HP100_INT_EN | HP100_RESET_LB, ioaddr + HP100_REG_OPTION_LSW )
-#define hp100_ints_on() \
-	outw( HP100_INT_EN | HP100_SET_LB, ioaddr + HP100_REG_OPTION_LSW )
-#define hp100_mem_map_enable() \
-	outw( HP100_MMAP_DIS | HP100_RESET_HB, ioaddr + HP100_REG_OPTION_LSW )
-#define hp100_mem_map_disable() \
-	outw( HP100_MMAP_DIS | HP100_SET_HB, ioaddr + HP100_REG_OPTION_LSW )
diff --git a/drivers/net/ethernet/huawei/Kconfig b/drivers/net/ethernet/huawei/Kconfig
new file mode 100644
index 000000000000..7d0feb1da158
--- /dev/null
+++ b/drivers/net/ethernet/huawei/Kconfig
@@ -0,0 +1,21 @@
+# SPDX-License-Identifier: GPL-2.0-only
+#
+# Huawei driver configuration
+#
+
+config NET_VENDOR_HUAWEI
+	bool "Huawei devices"
+	default y
+	help
+	  If you have a network (Ethernet) card belonging to this class, say Y.
+	  Note that the answer to this question doesn't directly affect the
+	  kernel: saying N will just cause the configurator to skip all
+	  the questions about Huawei cards. If you say Y, you will be asked
+	  for your specific card in the following questions.
+
+if NET_VENDOR_HUAWEI
+
+source "drivers/net/ethernet/huawei/hinic/Kconfig"
+source "drivers/net/ethernet/huawei/hinic3/Kconfig"
+
+endif # NET_VENDOR_HUAWEI
diff --git a/drivers/net/ethernet/huawei/Makefile b/drivers/net/ethernet/huawei/Makefile
new file mode 100644
index 000000000000..59865b882879
--- /dev/null
+++ b/drivers/net/ethernet/huawei/Makefile
@@ -0,0 +1,7 @@
+# SPDX-License-Identifier: GPL-2.0-only
+#
+# Makefile for the Huawei device drivers.
+#
+
+obj-$(CONFIG_HINIC) += hinic/
+obj-$(CONFIG_HINIC3) += hinic3/
diff --git a/drivers/net/ethernet/huawei/hinic/Kconfig b/drivers/net/ethernet/huawei/hinic/Kconfig
new file mode 100644
index 000000000000..b47bd5440c5f
--- /dev/null
+++ b/drivers/net/ethernet/huawei/hinic/Kconfig
@@ -0,0 +1,14 @@
+# SPDX-License-Identifier: GPL-2.0-only
+#
+# Huawei driver configuration
+#
+
+config HINIC
+	tristate "Huawei Intelligent PCIE Network Interface Card"
+	depends on (PCI_MSI && (X86 || ARM64))
+	select NET_DEVLINK
+	help
+	  This driver supports HiNIC PCIE Ethernet cards.
+	  To compile this driver as part of the kernel, choose Y here.
+	  If unsure, choose N.
+	  The default is compiled as module.
diff --git a/drivers/net/ethernet/huawei/hinic/Makefile b/drivers/net/ethernet/huawei/hinic/Makefile
new file mode 100644
index 000000000000..2f89119c9b69
--- /dev/null
+++ b/drivers/net/ethernet/huawei/hinic/Makefile
@@ -0,0 +1,8 @@
+# SPDX-License-Identifier: GPL-2.0-only
+obj-$(CONFIG_HINIC) += hinic.o
+
+hinic-y := hinic_main.o hinic_tx.o hinic_rx.o hinic_port.o hinic_hw_dev.o \
+	   hinic_hw_io.o hinic_hw_qp.o hinic_hw_cmdq.o hinic_hw_wq.o \
+	   hinic_hw_mgmt.o hinic_hw_api_cmd.o hinic_hw_eqs.o hinic_hw_if.o \
+	   hinic_common.o hinic_ethtool.o hinic_devlink.o hinic_hw_mbox.o \
+	   hinic_sriov.o hinic_debugfs.o
diff --git a/drivers/net/ethernet/huawei/hinic/hinic_common.c b/drivers/net/ethernet/huawei/hinic/hinic_common.c
new file mode 100644
index 000000000000..8e9b4a6c88c2
--- /dev/null
+++ b/drivers/net/ethernet/huawei/hinic/hinic_common.c
@@ -0,0 +1,71 @@
+// SPDX-License-Identifier: GPL-2.0-only
+/*
+ * Huawei HiNIC PCI Express Linux driver
+ * Copyright(c) 2017 Huawei Technologies Co., Ltd
+ */
+
+#include <linux/kernel.h>
+#include <linux/types.h>
+#include <asm/byteorder.h>
+
+#include "hinic_common.h"
+
+/**
+ * hinic_cpu_to_be32 - convert data to big endian 32 bit format
+ * @data: the data to convert
+ * @len: length of data to convert
+ **/
+void hinic_cpu_to_be32(void *data, int len)
+{
+	u32 *mem = data;
+	int i;
+
+	len = len / sizeof(u32);
+
+	for (i = 0; i < len; i++) {
+		*mem = cpu_to_be32(*mem);
+		mem++;
+	}
+}
+
+/**
+ * hinic_be32_to_cpu - convert data from big endian 32 bit format
+ * @data: the data to convert
+ * @len: length of data to convert
+ **/
+void hinic_be32_to_cpu(void *data, int len)
+{
+	u32 *mem = data;
+	int i;
+
+	len = len / sizeof(u32);
+
+	for (i = 0; i < len; i++) {
+		*mem = be32_to_cpu(*mem);
+		mem++;
+	}
+}
+
+/**
+ * hinic_set_sge - set dma area in scatter gather entry
+ * @sge: scatter gather entry
+ * @addr: dma address
+ * @len: length of relevant data in the dma address
+ **/
+void hinic_set_sge(struct hinic_sge *sge, dma_addr_t addr, int len)
+{
+	sge->hi_addr = upper_32_bits(addr);
+	sge->lo_addr = lower_32_bits(addr);
+	sge->len  = len;
+}
+
+/**
+ * hinic_sge_to_dma - get dma address from scatter gather entry
+ * @sge: scatter gather entry
+ *
+ * Return dma address of sg entry
+ **/
+dma_addr_t hinic_sge_to_dma(struct hinic_sge *sge)
+{
+	return (dma_addr_t)((((u64)sge->hi_addr) << 32) | sge->lo_addr);
+}
diff --git a/drivers/net/ethernet/huawei/hinic/hinic_common.h b/drivers/net/ethernet/huawei/hinic/hinic_common.h
new file mode 100644
index 000000000000..a0de9d9644c6
--- /dev/null
+++ b/drivers/net/ethernet/huawei/hinic/hinic_common.h
@@ -0,0 +1,29 @@
+/* SPDX-License-Identifier: GPL-2.0-only */
+/*
+ * Huawei HiNIC PCI Express Linux driver
+ * Copyright(c) 2017 Huawei Technologies Co., Ltd
+ */
+
+#ifndef HINIC_COMMON_H
+#define HINIC_COMMON_H
+
+#include <linux/types.h>
+
+#define UPPER_8_BITS(data)      (((data) >> 8) & 0xFF)
+#define LOWER_8_BITS(data)      ((data) & 0xFF)
+
+struct hinic_sge {
+	u32             hi_addr;
+	u32             lo_addr;
+	u32             len;
+};
+
+void hinic_cpu_to_be32(void *data, int len);
+
+void hinic_be32_to_cpu(void *data, int len);
+
+void hinic_set_sge(struct hinic_sge *sge, dma_addr_t addr, int len);
+
+dma_addr_t hinic_sge_to_dma(struct hinic_sge *sge);
+
+#endif
diff --git a/drivers/net/ethernet/huawei/hinic/hinic_debugfs.c b/drivers/net/ethernet/huawei/hinic/hinic_debugfs.c
new file mode 100644
index 000000000000..061952c6c21a
--- /dev/null
+++ b/drivers/net/ethernet/huawei/hinic/hinic_debugfs.c
@@ -0,0 +1,324 @@
+// SPDX-License-Identifier: GPL-2.0-only
+/* Huawei HiNIC PCI Express Linux driver
+ * Copyright(c) 2017 Huawei Technologies Co., Ltd
+ */
+
+#include <linux/debugfs.h>
+#include <linux/device.h>
+
+#include "hinic_debugfs.h"
+
+static struct dentry *hinic_dbgfs_root;
+
+enum sq_dbg_info {
+	GLB_SQ_ID,
+	SQ_PI,
+	SQ_CI,
+	SQ_FI,
+	SQ_MSIX_ENTRY,
+};
+
+static char *sq_fields[] = {"glb_sq_id", "sq_pi", "sq_ci", "sq_fi", "sq_msix_entry"};
+
+static u64 hinic_dbg_get_sq_info(struct hinic_dev *nic_dev, struct hinic_sq *sq, int idx)
+{
+	struct hinic_wq *wq = sq->wq;
+
+	switch (idx) {
+	case GLB_SQ_ID:
+		return nic_dev->hwdev->func_to_io.global_qpn + sq->qid;
+	case SQ_PI:
+		return atomic_read(&wq->prod_idx) & wq->mask;
+	case SQ_CI:
+		return atomic_read(&wq->cons_idx) & wq->mask;
+	case SQ_FI:
+		return be16_to_cpu(*(__be16 *)(sq->hw_ci_addr)) & wq->mask;
+	case SQ_MSIX_ENTRY:
+		return sq->msix_entry;
+	}
+
+	return 0;
+}
+
+enum rq_dbg_info {
+	GLB_RQ_ID,
+	RQ_HW_PI,
+	RQ_SW_CI,
+	RQ_SW_PI,
+	RQ_MSIX_ENTRY,
+};
+
+static char *rq_fields[] = {"glb_rq_id", "rq_hw_pi", "rq_sw_ci", "rq_sw_pi", "rq_msix_entry"};
+
+static u64 hinic_dbg_get_rq_info(struct hinic_dev *nic_dev, struct hinic_rq *rq, int idx)
+{
+	struct hinic_wq *wq = rq->wq;
+
+	switch (idx) {
+	case GLB_RQ_ID:
+		return nic_dev->hwdev->func_to_io.global_qpn + rq->qid;
+	case RQ_HW_PI:
+		return be16_to_cpu(*(__be16 *)(rq->pi_virt_addr)) & wq->mask;
+	case RQ_SW_CI:
+		return atomic_read(&wq->cons_idx) & wq->mask;
+	case RQ_SW_PI:
+		return atomic_read(&wq->prod_idx) & wq->mask;
+	case RQ_MSIX_ENTRY:
+		return rq->msix_entry;
+	}
+
+	return 0;
+}
+
+enum func_tbl_info {
+	VALID,
+	RX_MODE,
+	MTU,
+	RQ_DEPTH,
+	QUEUE_NUM,
+};
+
+static char *func_table_fields[] = {"valid", "rx_mode", "mtu", "rq_depth", "cfg_q_num"};
+
+static int hinic_dbg_get_func_table(struct hinic_dev *nic_dev, int idx)
+{
+	struct tag_sml_funcfg_tbl *funcfg_table_elem;
+	struct hinic_cmd_lt_rd *read_data;
+	u16 out_size = sizeof(*read_data);
+	int ret = ~0;
+	int err;
+
+	read_data = kzalloc(sizeof(*read_data), GFP_KERNEL);
+	if (!read_data)
+		return ~0;
+
+	read_data->node = TBL_ID_FUNC_CFG_SM_NODE;
+	read_data->inst = TBL_ID_FUNC_CFG_SM_INST;
+	read_data->entry_size = HINIC_FUNCTION_CONFIGURE_TABLE_SIZE;
+	read_data->lt_index = HINIC_HWIF_FUNC_IDX(nic_dev->hwdev->hwif);
+	read_data->len = HINIC_FUNCTION_CONFIGURE_TABLE_SIZE;
+
+	err = hinic_port_msg_cmd(nic_dev->hwdev, HINIC_PORT_CMD_RD_LINE_TBL, read_data,
+				 sizeof(*read_data), read_data, &out_size);
+	if (err || out_size != sizeof(*read_data) || read_data->status) {
+		netif_err(nic_dev, drv, nic_dev->netdev,
+			  "Failed to get func table, err: %d, status: 0x%x, out size: 0x%x\n",
+			  err, read_data->status, out_size);
+		kfree(read_data);
+		return ~0;
+	}
+
+	funcfg_table_elem = (struct tag_sml_funcfg_tbl *)read_data->data;
+
+	switch (idx) {
+	case VALID:
+		ret = funcfg_table_elem->dw0.bs.valid;
+		break;
+	case RX_MODE:
+		ret = funcfg_table_elem->dw0.bs.nic_rx_mode;
+		break;
+	case MTU:
+		ret = funcfg_table_elem->dw1.bs.mtu;
+		break;
+	case RQ_DEPTH:
+		ret = funcfg_table_elem->dw13.bs.cfg_rq_depth;
+		break;
+	case QUEUE_NUM:
+		ret = funcfg_table_elem->dw13.bs.cfg_q_num;
+		break;
+	}
+
+	kfree(read_data);
+
+	return ret;
+}
+
+static ssize_t hinic_dbg_cmd_read(struct file *filp, char __user *buffer, size_t count,
+				  loff_t *ppos)
+{
+	struct hinic_debug_priv *dbg;
+	char ret_buf[20];
+	int *desc;
+	u64 out;
+	int ret;
+
+	desc = filp->private_data;
+	dbg = container_of(desc, struct hinic_debug_priv, field_id[*desc]);
+
+	switch (dbg->type) {
+	case HINIC_DBG_SQ_INFO:
+		out = hinic_dbg_get_sq_info(dbg->dev, dbg->object, *desc);
+		break;
+
+	case HINIC_DBG_RQ_INFO:
+		out = hinic_dbg_get_rq_info(dbg->dev, dbg->object, *desc);
+		break;
+
+	case HINIC_DBG_FUNC_TABLE:
+		out = hinic_dbg_get_func_table(dbg->dev, *desc);
+		break;
+
+	default:
+		netif_warn(dbg->dev, drv, dbg->dev->netdev, "Invalid hinic debug cmd: %d\n",
+			   dbg->type);
+		return -EINVAL;
+	}
+
+	ret = snprintf(ret_buf, sizeof(ret_buf), "0x%llx\n", out);
+
+	return simple_read_from_buffer(buffer, count, ppos, ret_buf, ret);
+}
+
+static const struct file_operations hinic_dbg_cmd_fops = {
+	.owner = THIS_MODULE,
+	.open  = simple_open,
+	.read  = hinic_dbg_cmd_read,
+};
+
+static int create_dbg_files(struct hinic_dev *dev, enum hinic_dbg_type type, void *data,
+			    struct dentry *root, struct hinic_debug_priv **dbg, char **field,
+			    int nfile)
+{
+	struct hinic_debug_priv *tmp;
+	int i;
+
+	tmp = kzalloc(sizeof(*tmp), GFP_KERNEL);
+	if (!tmp)
+		return -ENOMEM;
+
+	tmp->dev = dev;
+	tmp->object = data;
+	tmp->type = type;
+	tmp->root = root;
+
+	for (i = 0; i < nfile; i++) {
+		tmp->field_id[i] = i;
+		debugfs_create_file(field[i], 0400, root, &tmp->field_id[i], &hinic_dbg_cmd_fops);
+	}
+
+	*dbg = tmp;
+
+	return 0;
+}
+
+static void rem_dbg_files(struct hinic_debug_priv *dbg)
+{
+	if (dbg->type != HINIC_DBG_FUNC_TABLE)
+		debugfs_remove_recursive(dbg->root);
+
+	kfree(dbg);
+}
+
+int hinic_sq_debug_add(struct hinic_dev *dev, u16 sq_id)
+{
+	struct hinic_sq *sq;
+	struct dentry *root;
+	char sub_dir[16];
+
+	sq = dev->txqs[sq_id].sq;
+
+	sprintf(sub_dir, "0x%x", sq_id);
+
+	root = debugfs_create_dir(sub_dir, dev->sq_dbgfs);
+
+	return create_dbg_files(dev, HINIC_DBG_SQ_INFO, sq, root, &sq->dbg, sq_fields,
+				ARRAY_SIZE(sq_fields));
+}
+
+void hinic_sq_debug_rem(struct hinic_sq *sq)
+{
+	if (sq->dbg)
+		rem_dbg_files(sq->dbg);
+}
+
+int hinic_rq_debug_add(struct hinic_dev *dev, u16 rq_id)
+{
+	struct hinic_rq *rq;
+	struct dentry *root;
+	char sub_dir[16];
+
+	rq = dev->rxqs[rq_id].rq;
+
+	sprintf(sub_dir, "0x%x", rq_id);
+
+	root = debugfs_create_dir(sub_dir, dev->rq_dbgfs);
+
+	return create_dbg_files(dev, HINIC_DBG_RQ_INFO, rq, root, &rq->dbg, rq_fields,
+				ARRAY_SIZE(rq_fields));
+}
+
+void hinic_rq_debug_rem(struct hinic_rq *rq)
+{
+	if (rq->dbg)
+		rem_dbg_files(rq->dbg);
+}
+
+int hinic_func_table_debug_add(struct hinic_dev *dev)
+{
+	if (HINIC_IS_VF(dev->hwdev->hwif))
+		return 0;
+
+	return create_dbg_files(dev, HINIC_DBG_FUNC_TABLE, dev, dev->func_tbl_dbgfs, &dev->dbg,
+				func_table_fields, ARRAY_SIZE(func_table_fields));
+}
+
+void hinic_func_table_debug_rem(struct hinic_dev *dev)
+{
+	if (!HINIC_IS_VF(dev->hwdev->hwif) && dev->dbg)
+		rem_dbg_files(dev->dbg);
+}
+
+void hinic_sq_dbgfs_init(struct hinic_dev *nic_dev)
+{
+	nic_dev->sq_dbgfs = debugfs_create_dir("SQs", nic_dev->dbgfs_root);
+}
+
+void hinic_sq_dbgfs_uninit(struct hinic_dev *nic_dev)
+{
+	debugfs_remove_recursive(nic_dev->sq_dbgfs);
+}
+
+void hinic_rq_dbgfs_init(struct hinic_dev *nic_dev)
+{
+	nic_dev->rq_dbgfs = debugfs_create_dir("RQs", nic_dev->dbgfs_root);
+}
+
+void hinic_rq_dbgfs_uninit(struct hinic_dev *nic_dev)
+{
+	debugfs_remove_recursive(nic_dev->rq_dbgfs);
+}
+
+void hinic_func_tbl_dbgfs_init(struct hinic_dev *nic_dev)
+{
+	if (!HINIC_IS_VF(nic_dev->hwdev->hwif))
+		nic_dev->func_tbl_dbgfs = debugfs_create_dir("func_table", nic_dev->dbgfs_root);
+}
+
+void hinic_func_tbl_dbgfs_uninit(struct hinic_dev *nic_dev)
+{
+	if (!HINIC_IS_VF(nic_dev->hwdev->hwif))
+		debugfs_remove_recursive(nic_dev->func_tbl_dbgfs);
+}
+
+void hinic_dbg_init(struct hinic_dev *nic_dev)
+{
+	nic_dev->dbgfs_root = debugfs_create_dir(pci_name(nic_dev->hwdev->hwif->pdev),
+						 hinic_dbgfs_root);
+}
+
+void hinic_dbg_uninit(struct hinic_dev *nic_dev)
+{
+	debugfs_remove_recursive(nic_dev->dbgfs_root);
+	nic_dev->dbgfs_root = NULL;
+}
+
+void hinic_dbg_register_debugfs(const char *debugfs_dir_name)
+{
+	hinic_dbgfs_root = debugfs_create_dir(debugfs_dir_name, NULL);
+}
+
+void hinic_dbg_unregister_debugfs(void)
+{
+	debugfs_remove_recursive(hinic_dbgfs_root);
+	hinic_dbgfs_root = NULL;
+}
diff --git a/drivers/net/ethernet/huawei/hinic/hinic_debugfs.h b/drivers/net/ethernet/huawei/hinic/hinic_debugfs.h
new file mode 100644
index 000000000000..e10f739d8339
--- /dev/null
+++ b/drivers/net/ethernet/huawei/hinic/hinic_debugfs.h
@@ -0,0 +1,113 @@
+/* SPDX-License-Identifier: GPL-2.0-only */
+/* Huawei HiNIC PCI Express Linux driver
+ * Copyright(c) 2017 Huawei Technologies Co., Ltd
+ */
+
+#ifndef HINIC_DEBUGFS_H
+#define HINIC_DEBUGFS_H
+
+#include "hinic_dev.h"
+
+#define    TBL_ID_FUNC_CFG_SM_NODE                      11
+#define    TBL_ID_FUNC_CFG_SM_INST                      1
+
+#define HINIC_FUNCTION_CONFIGURE_TABLE_SIZE             64
+
+struct hinic_cmd_lt_rd {
+	u8	status;
+	u8	version;
+	u8	rsvd0[6];
+
+	unsigned char node;
+	unsigned char inst;
+	unsigned char entry_size;
+	unsigned char rsvd;
+	unsigned int lt_index;
+	unsigned int offset;
+	unsigned int len;
+	unsigned char data[100];
+};
+
+struct tag_sml_funcfg_tbl {
+	union {
+		struct {
+			u32 rsvd0            :8;
+			u32 nic_rx_mode      :5;
+			u32 rsvd1            :18;
+			u32 valid            :1;
+		} bs;
+
+		u32 value;
+	} dw0;
+
+	union {
+		struct {
+			u32 vlan_id             :12;
+			u32 vlan_mode           :3;
+			u32 fast_recycled_mode  :1;
+			u32 mtu                 :16;
+		} bs;
+
+		u32 value;
+	} dw1;
+
+	u32 dw2;
+	u32 dw3;
+	u32 dw4;
+	u32 dw5;
+	u32 dw6;
+	u32 dw7;
+	u32 dw8;
+	u32 dw9;
+	u32 dw10;
+	u32 dw11;
+	u32 dw12;
+
+	union {
+		struct {
+			u32 rsvd2               :15;
+			u32 cfg_q_num           :9;
+			u32 cfg_rq_depth        :6;
+			u32 vhd_type            :2;
+		} bs;
+
+		u32 value;
+	} dw13;
+
+	u32 dw14;
+	u32 dw15;
+};
+
+int hinic_sq_debug_add(struct hinic_dev *dev, u16 sq_id);
+
+void hinic_sq_debug_rem(struct hinic_sq *sq);
+
+int hinic_rq_debug_add(struct hinic_dev *dev, u16 rq_id);
+
+void hinic_rq_debug_rem(struct hinic_rq *rq);
+
+int hinic_func_table_debug_add(struct hinic_dev *dev);
+
+void hinic_func_table_debug_rem(struct hinic_dev *dev);
+
+void hinic_sq_dbgfs_init(struct hinic_dev *nic_dev);
+
+void hinic_sq_dbgfs_uninit(struct hinic_dev *nic_dev);
+
+void hinic_rq_dbgfs_init(struct hinic_dev *nic_dev);
+
+void hinic_rq_dbgfs_uninit(struct hinic_dev *nic_dev);
+
+void hinic_func_tbl_dbgfs_init(struct hinic_dev *nic_dev);
+
+void hinic_func_tbl_dbgfs_uninit(struct hinic_dev *nic_dev);
+
+void hinic_dbg_init(struct hinic_dev *nic_dev);
+
+void hinic_dbg_uninit(struct hinic_dev *nic_dev);
+
+void hinic_dbg_register_debugfs(const char *debugfs_dir_name);
+
+void hinic_dbg_unregister_debugfs(void);
+
+#endif
diff --git a/drivers/net/ethernet/huawei/hinic/hinic_dev.h b/drivers/net/ethernet/huawei/hinic/hinic_dev.h
new file mode 100644
index 000000000000..52ea97c818b8
--- /dev/null
+++ b/drivers/net/ethernet/huawei/hinic/hinic_dev.h
@@ -0,0 +1,132 @@
+/* SPDX-License-Identifier: GPL-2.0-only */
+/*
+ * Huawei HiNIC PCI Express Linux driver
+ * Copyright(c) 2017 Huawei Technologies Co., Ltd
+ */
+
+#ifndef HINIC_DEV_H
+#define HINIC_DEV_H
+
+#include <linux/netdevice.h>
+#include <linux/types.h>
+#include <linux/semaphore.h>
+#include <linux/workqueue.h>
+#include <linux/bitops.h>
+
+#include "hinic_hw_dev.h"
+#include "hinic_tx.h"
+#include "hinic_rx.h"
+#include "hinic_sriov.h"
+
+#define HINIC_DRV_NAME          "hinic"
+
+#define LP_PKT_CNT		64
+
+#define HINIC_MAX_JUMBO_FRAME_SIZE      15872
+#define HINIC_MAX_MTU_SIZE      (HINIC_MAX_JUMBO_FRAME_SIZE - ETH_HLEN - ETH_FCS_LEN)
+#define HINIC_MIN_MTU_SIZE      256
+
+enum hinic_flags {
+	HINIC_LINK_UP = BIT(0),
+	HINIC_INTF_UP = BIT(1),
+	HINIC_RSS_ENABLE = BIT(2),
+	HINIC_LINK_DOWN = BIT(3),
+	HINIC_LP_TEST = BIT(4),
+};
+
+struct hinic_rx_mode_work {
+	struct work_struct      work;
+	u32                     rx_mode;
+};
+
+struct hinic_rss_type {
+	u8 tcp_ipv6_ext;
+	u8 ipv6_ext;
+	u8 tcp_ipv6;
+	u8 ipv6;
+	u8 tcp_ipv4;
+	u8 ipv4;
+	u8 udp_ipv6;
+	u8 udp_ipv4;
+};
+
+enum hinic_rss_hash_type {
+	HINIC_RSS_HASH_ENGINE_TYPE_XOR,
+	HINIC_RSS_HASH_ENGINE_TYPE_TOEP,
+	HINIC_RSS_HASH_ENGINE_TYPE_MAX,
+};
+
+struct hinic_intr_coal_info {
+	u8	pending_limt;
+	u8	coalesce_timer_cfg;
+	u8	resend_timer_cfg;
+};
+
+enum hinic_dbg_type {
+	HINIC_DBG_SQ_INFO,
+	HINIC_DBG_RQ_INFO,
+	HINIC_DBG_FUNC_TABLE,
+};
+
+struct hinic_debug_priv {
+	struct hinic_dev	*dev;
+	void			*object;
+	enum hinic_dbg_type	type;
+	struct dentry		*root;
+	int			field_id[64];
+};
+
+struct hinic_dev {
+	struct net_device               *netdev;
+	struct hinic_hwdev              *hwdev;
+
+	u32                             msg_enable;
+	unsigned int                    tx_weight;
+	unsigned int                    rx_weight;
+	u16				num_qps;
+	u16				max_qps;
+
+	unsigned int                    flags;
+
+	struct semaphore                mgmt_lock;
+	unsigned long                   *vlan_bitmap;
+
+	struct hinic_rx_mode_work       rx_mode_work;
+	struct workqueue_struct         *workq;
+
+	struct hinic_txq                *txqs;
+	struct hinic_rxq                *rxqs;
+	u16				sq_depth;
+	u16				rq_depth;
+
+	u8				rss_tmpl_idx;
+	u8				rss_hash_engine;
+	u16				num_rss;
+	u16				rss_limit;
+	struct hinic_rss_type		rss_type;
+	u8				*rss_hkey_user;
+	s32				*rss_indir_user;
+	struct hinic_intr_coal_info	*rx_intr_coalesce;
+	struct hinic_intr_coal_info	*tx_intr_coalesce;
+	struct hinic_sriov_info sriov_info;
+	int				lb_test_rx_idx;
+	int				lb_pkt_len;
+	u8				*lb_test_rx_buf;
+
+	struct dentry			*dbgfs_root;
+	struct dentry			*sq_dbgfs;
+	struct dentry			*rq_dbgfs;
+	struct dentry			*func_tbl_dbgfs;
+	struct hinic_debug_priv		*dbg;
+	struct devlink			*devlink;
+	bool				cable_unplugged;
+	bool				module_unrecognized;
+};
+
+struct hinic_devlink_priv {
+	struct hinic_hwdev		*hwdev;
+	struct devlink_health_reporter  *hw_fault_reporter;
+	struct devlink_health_reporter  *fw_fault_reporter;
+};
+
+#endif
diff --git a/drivers/net/ethernet/huawei/hinic/hinic_devlink.c b/drivers/net/ethernet/huawei/hinic/hinic_devlink.c
new file mode 100644
index 000000000000..300bc267a259
--- /dev/null
+++ b/drivers/net/ethernet/huawei/hinic/hinic_devlink.c
@@ -0,0 +1,481 @@
+// SPDX-License-Identifier: GPL-2.0
+/* Huawei HiNIC PCI Express Linux driver
+ * Copyright(c) 2017 Huawei Technologies Co., Ltd
+ *
+ * This program is free software; you can redistribute it and/or modify it
+ * under the terms and conditions of the GNU General Public License,
+ * version 2, as published by the Free Software Foundation.
+ *
+ * This program is distributed in the hope it will be useful, but WITHOUT
+ * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
+ * FITNESS FOR A PARTICULAR PURPOSE.  See the GNU General Public License
+ * for more details.
+ *
+ */
+#include <linux/netlink.h>
+#include <net/devlink.h>
+#include <linux/firmware.h>
+
+#include "hinic_port.h"
+#include "hinic_devlink.h"
+#include "hinic_hw_dev.h"
+
+static bool check_image_valid(struct hinic_devlink_priv *priv, const u8 *buf,
+			      u32 image_size, struct host_image_st *host_image)
+{
+	struct fw_image_st *fw_image = NULL;
+	u32 len = 0;
+	u32 i;
+
+	fw_image = (struct fw_image_st *)buf;
+
+	if (fw_image->fw_magic != HINIC_MAGIC_NUM) {
+		dev_err(&priv->hwdev->hwif->pdev->dev, "Wrong fw_magic read from file, fw_magic: 0x%x\n",
+			fw_image->fw_magic);
+		return false;
+	}
+
+	if (fw_image->fw_info.fw_section_cnt > MAX_FW_TYPE_NUM) {
+		dev_err(&priv->hwdev->hwif->pdev->dev, "Wrong fw_type_num read from file, fw_type_num: 0x%x\n",
+			fw_image->fw_info.fw_section_cnt);
+		return false;
+	}
+
+	for (i = 0; i < fw_image->fw_info.fw_section_cnt; i++) {
+		len += fw_image->fw_section_info[i].fw_section_len;
+		host_image->image_section_info[i] = fw_image->fw_section_info[i];
+	}
+
+	if (len != fw_image->fw_len ||
+	    (fw_image->fw_len + UPDATEFW_IMAGE_HEAD_SIZE) != image_size) {
+		dev_err(&priv->hwdev->hwif->pdev->dev, "Wrong data size read from file\n");
+		return false;
+	}
+
+	host_image->image_info.up_total_len = fw_image->fw_len;
+	host_image->image_info.fw_version = fw_image->fw_version;
+	host_image->section_type_num = fw_image->fw_info.fw_section_cnt;
+	host_image->device_id = fw_image->device_id;
+
+	return true;
+}
+
+static bool check_image_integrity(struct hinic_devlink_priv *priv,
+				  struct host_image_st *host_image,
+				  u32 update_type)
+{
+	u32 collect_section_type = 0;
+	u32 i, type;
+
+	for (i = 0; i < host_image->section_type_num; i++) {
+		type = host_image->image_section_info[i].fw_section_type;
+		if (collect_section_type & (1U << type)) {
+			dev_err(&priv->hwdev->hwif->pdev->dev, "Duplicate section type: %u\n",
+				type);
+			return false;
+		}
+		collect_section_type |= (1U << type);
+	}
+
+	if (update_type == FW_UPDATE_COLD &&
+	    (((collect_section_type & _IMAGE_COLD_SUB_MODULES_MUST_IN) ==
+	       _IMAGE_COLD_SUB_MODULES_MUST_IN) ||
+	      collect_section_type == _IMAGE_CFG_SUB_MODULES_MUST_IN))
+		return true;
+
+	if (update_type == FW_UPDATE_HOT &&
+	    (collect_section_type & _IMAGE_HOT_SUB_MODULES_MUST_IN) ==
+	    _IMAGE_HOT_SUB_MODULES_MUST_IN)
+		return true;
+
+	if (update_type == FW_UPDATE_COLD)
+		dev_err(&priv->hwdev->hwif->pdev->dev, "Check file integrity failed, valid: 0x%x or 0x%lx, current: 0x%x\n",
+			_IMAGE_COLD_SUB_MODULES_MUST_IN,
+			_IMAGE_CFG_SUB_MODULES_MUST_IN, collect_section_type);
+	else
+		dev_err(&priv->hwdev->hwif->pdev->dev, "Check file integrity failed, valid:0x%x, current: 0x%x\n",
+			_IMAGE_HOT_SUB_MODULES_MUST_IN, collect_section_type);
+
+	return false;
+}
+
+static int check_image_device_type(struct hinic_devlink_priv *priv,
+				   u32 image_device_type)
+{
+	struct hinic_comm_board_info board_info = {0};
+
+	if (hinic_get_board_info(priv->hwdev, &board_info)) {
+		dev_err(&priv->hwdev->hwif->pdev->dev, "Get board info failed\n");
+		return false;
+	}
+
+	if (image_device_type == board_info.info.board_type)
+		return true;
+
+	dev_err(&priv->hwdev->hwif->pdev->dev, "The device type of upgrade file doesn't match the device type of current firmware, please check the upgrade file\n");
+	dev_err(&priv->hwdev->hwif->pdev->dev, "The image device type: 0x%x, firmware device type: 0x%x\n",
+		image_device_type, board_info.info.board_type);
+
+	return false;
+}
+
+static int hinic_flash_fw(struct hinic_devlink_priv *priv, const u8 *data,
+			  struct host_image_st *host_image)
+{
+	u32 section_remain_send_len, send_fragment_len, send_pos, up_total_len;
+	struct hinic_cmd_update_fw *fw_update_msg = NULL;
+	u32 section_type, section_crc, section_version;
+	u32 i, len, section_len, section_offset;
+	u16 out_size = sizeof(*fw_update_msg);
+	int total_len_flag = 0;
+	int err;
+
+	fw_update_msg = kzalloc(sizeof(*fw_update_msg), GFP_KERNEL);
+	if (!fw_update_msg)
+		return -ENOMEM;
+
+	up_total_len = host_image->image_info.up_total_len;
+
+	for (i = 0; i < host_image->section_type_num; i++) {
+		len = host_image->image_section_info[i].fw_section_len;
+		if (host_image->image_section_info[i].fw_section_type ==
+		    UP_FW_UPDATE_BOOT) {
+			up_total_len = up_total_len - len;
+			break;
+		}
+	}
+
+	for (i = 0; i < host_image->section_type_num; i++) {
+		section_len =
+			host_image->image_section_info[i].fw_section_len;
+		section_offset =
+			host_image->image_section_info[i].fw_section_offset;
+		section_remain_send_len = section_len;
+		section_type =
+			host_image->image_section_info[i].fw_section_type;
+		section_crc = host_image->image_section_info[i].fw_section_crc;
+		section_version =
+			host_image->image_section_info[i].fw_section_version;
+
+		if (section_type == UP_FW_UPDATE_BOOT)
+			continue;
+
+		send_fragment_len = 0;
+		send_pos = 0;
+
+		while (section_remain_send_len > 0) {
+			if (!total_len_flag) {
+				fw_update_msg->total_len = up_total_len;
+				total_len_flag = 1;
+			} else {
+				fw_update_msg->total_len = 0;
+			}
+
+			memset(fw_update_msg->data, 0, MAX_FW_FRAGMENT_LEN);
+
+			fw_update_msg->ctl_info.SF =
+				(section_remain_send_len == section_len) ?
+				true : false;
+			fw_update_msg->section_info.FW_section_CRC = section_crc;
+			fw_update_msg->fw_section_version = section_version;
+			fw_update_msg->ctl_info.flag = UP_TYPE_A;
+
+			if (section_type <= UP_FW_UPDATE_UP_DATA_B) {
+				fw_update_msg->section_info.FW_section_type =
+					(section_type % 2) ?
+					UP_FW_UPDATE_UP_DATA :
+					UP_FW_UPDATE_UP_TEXT;
+
+				fw_update_msg->ctl_info.flag = UP_TYPE_B;
+				if (section_type <= UP_FW_UPDATE_UP_DATA_A)
+					fw_update_msg->ctl_info.flag = UP_TYPE_A;
+			} else {
+				fw_update_msg->section_info.FW_section_type =
+					section_type - 0x2;
+			}
+
+			fw_update_msg->setion_total_len = section_len;
+			fw_update_msg->section_offset = send_pos;
+
+			if (section_remain_send_len <= MAX_FW_FRAGMENT_LEN) {
+				fw_update_msg->ctl_info.SL = true;
+				fw_update_msg->ctl_info.fragment_len =
+					section_remain_send_len;
+				send_fragment_len += section_remain_send_len;
+			} else {
+				fw_update_msg->ctl_info.SL = false;
+				fw_update_msg->ctl_info.fragment_len =
+					MAX_FW_FRAGMENT_LEN;
+				send_fragment_len += MAX_FW_FRAGMENT_LEN;
+			}
+
+			memcpy(fw_update_msg->data,
+			       data + UPDATEFW_IMAGE_HEAD_SIZE +
+			       section_offset + send_pos,
+			       fw_update_msg->ctl_info.fragment_len);
+
+			err = hinic_port_msg_cmd(priv->hwdev,
+						 HINIC_PORT_CMD_UPDATE_FW,
+						 fw_update_msg,
+						 sizeof(*fw_update_msg),
+						 fw_update_msg, &out_size);
+			if (err || !out_size || fw_update_msg->status) {
+				dev_err(&priv->hwdev->hwif->pdev->dev, "Failed to update firmware, err: %d, status: 0x%x, out size: 0x%x\n",
+					err, fw_update_msg->status, out_size);
+				err = fw_update_msg->status ?
+					fw_update_msg->status : -EIO;
+				kfree(fw_update_msg);
+				return err;
+			}
+
+			send_pos = send_fragment_len;
+			section_remain_send_len = section_len -
+						  send_fragment_len;
+		}
+	}
+
+	kfree(fw_update_msg);
+
+	return 0;
+}
+
+static int hinic_firmware_update(struct hinic_devlink_priv *priv,
+				 const struct firmware *fw,
+				 struct netlink_ext_ack *extack)
+{
+	struct host_image_st host_image;
+	int err;
+
+	memset(&host_image, 0, sizeof(struct host_image_st));
+
+	if (!check_image_valid(priv, fw->data, fw->size, &host_image) ||
+	    !check_image_integrity(priv, &host_image, FW_UPDATE_COLD) ||
+	    !check_image_device_type(priv, host_image.device_id)) {
+		NL_SET_ERR_MSG_MOD(extack, "Check image failed");
+		return -EINVAL;
+	}
+
+	dev_info(&priv->hwdev->hwif->pdev->dev, "Flash firmware begin\n");
+
+	err = hinic_flash_fw(priv, fw->data, &host_image);
+	if (err) {
+		if (err == HINIC_FW_DISMATCH_ERROR) {
+			dev_err(&priv->hwdev->hwif->pdev->dev, "Firmware image doesn't match this card, please use newer image, err: %d\n",
+				err);
+			NL_SET_ERR_MSG_MOD(extack,
+					   "Firmware image doesn't match this card, please use newer image");
+		} else {
+			dev_err(&priv->hwdev->hwif->pdev->dev, "Send firmware image data failed, err: %d\n",
+				err);
+			NL_SET_ERR_MSG_MOD(extack, "Send firmware image data failed");
+		}
+
+		return err;
+	}
+
+	dev_info(&priv->hwdev->hwif->pdev->dev, "Flash firmware end\n");
+
+	return 0;
+}
+
+static int hinic_devlink_flash_update(struct devlink *devlink,
+				      struct devlink_flash_update_params *params,
+				      struct netlink_ext_ack *extack)
+{
+	struct hinic_devlink_priv *priv = devlink_priv(devlink);
+
+	return hinic_firmware_update(priv, params->fw, extack);
+}
+
+static const struct devlink_ops hinic_devlink_ops = {
+	.flash_update = hinic_devlink_flash_update,
+};
+
+struct devlink *hinic_devlink_alloc(struct device *dev)
+{
+	return devlink_alloc(&hinic_devlink_ops, sizeof(struct hinic_dev), dev);
+}
+
+void hinic_devlink_free(struct devlink *devlink)
+{
+	devlink_free(devlink);
+}
+
+void hinic_devlink_register(struct hinic_devlink_priv *priv)
+{
+	struct devlink *devlink = priv_to_devlink(priv);
+
+	devlink_register(devlink);
+}
+
+void hinic_devlink_unregister(struct hinic_devlink_priv *priv)
+{
+	struct devlink *devlink = priv_to_devlink(priv);
+
+	devlink_unregister(devlink);
+}
+
+static void chip_fault_show(struct devlink_fmsg *fmsg,
+			    struct hinic_fault_event *event)
+{
+	const char * const level_str[FAULT_LEVEL_MAX + 1] = {
+		"fatal", "reset", "flr", "general", "suggestion", "Unknown"};
+	u8 fault_level;
+
+	fault_level = (event->event.chip.err_level < FAULT_LEVEL_MAX) ?
+		event->event.chip.err_level : FAULT_LEVEL_MAX;
+	if (fault_level == FAULT_LEVEL_SERIOUS_FLR)
+		devlink_fmsg_u32_pair_put(fmsg, "Function level err func_id",
+					  (u32)event->event.chip.func_id);
+	devlink_fmsg_u8_pair_put(fmsg, "module_id", event->event.chip.node_id);
+	devlink_fmsg_u32_pair_put(fmsg, "err_type", (u32)event->event.chip.err_type);
+	devlink_fmsg_string_pair_put(fmsg, "err_level", level_str[fault_level]);
+	devlink_fmsg_u32_pair_put(fmsg, "err_csr_addr",
+				  event->event.chip.err_csr_addr);
+	devlink_fmsg_u32_pair_put(fmsg, "err_csr_value",
+				  event->event.chip.err_csr_value);
+}
+
+static void fault_report_show(struct devlink_fmsg *fmsg,
+			      struct hinic_fault_event *event)
+{
+	const char * const type_str[FAULT_TYPE_MAX + 1] = {
+		"chip", "ucode", "mem rd timeout", "mem wr timeout",
+		"reg rd timeout", "reg wr timeout", "phy fault", "Unknown"};
+	u8 fault_type;
+
+	fault_type = (event->type < FAULT_TYPE_MAX) ? event->type : FAULT_TYPE_MAX;
+
+	devlink_fmsg_string_pair_put(fmsg, "Fault type", type_str[fault_type]);
+	devlink_fmsg_binary_pair_put(fmsg, "Fault raw data", event->event.val,
+				     sizeof(event->event.val));
+
+	switch (event->type) {
+	case FAULT_TYPE_CHIP:
+		chip_fault_show(fmsg, event);
+		break;
+	case FAULT_TYPE_UCODE:
+		devlink_fmsg_u8_pair_put(fmsg, "Cause_id", event->event.ucode.cause_id);
+		devlink_fmsg_u8_pair_put(fmsg, "core_id", event->event.ucode.core_id);
+		devlink_fmsg_u8_pair_put(fmsg, "c_id", event->event.ucode.c_id);
+		devlink_fmsg_u8_pair_put(fmsg, "epc", event->event.ucode.epc);
+		break;
+	case FAULT_TYPE_MEM_RD_TIMEOUT:
+	case FAULT_TYPE_MEM_WR_TIMEOUT:
+		devlink_fmsg_u32_pair_put(fmsg, "Err_csr_ctrl",
+					  event->event.mem_timeout.err_csr_ctrl);
+		devlink_fmsg_u32_pair_put(fmsg, "err_csr_data",
+					  event->event.mem_timeout.err_csr_data);
+		devlink_fmsg_u32_pair_put(fmsg, "ctrl_tab",
+					  event->event.mem_timeout.ctrl_tab);
+		devlink_fmsg_u32_pair_put(fmsg, "mem_index",
+					  event->event.mem_timeout.mem_index);
+		break;
+	case FAULT_TYPE_REG_RD_TIMEOUT:
+	case FAULT_TYPE_REG_WR_TIMEOUT:
+		devlink_fmsg_u32_pair_put(fmsg, "Err_csr", event->event.reg_timeout.err_csr);
+		break;
+	case FAULT_TYPE_PHY_FAULT:
+		devlink_fmsg_u8_pair_put(fmsg, "Op_type", event->event.phy_fault.op_type);
+		devlink_fmsg_u8_pair_put(fmsg, "port_id", event->event.phy_fault.port_id);
+		devlink_fmsg_u8_pair_put(fmsg, "dev_ad", event->event.phy_fault.dev_ad);
+		devlink_fmsg_u32_pair_put(fmsg, "csr_addr", event->event.phy_fault.csr_addr);
+		devlink_fmsg_u32_pair_put(fmsg, "op_data", event->event.phy_fault.op_data);
+		break;
+	default:
+		break;
+	}
+}
+
+static int hinic_hw_reporter_dump(struct devlink_health_reporter *reporter,
+				  struct devlink_fmsg *fmsg, void *priv_ctx,
+				  struct netlink_ext_ack *extack)
+{
+	if (priv_ctx)
+		fault_report_show(fmsg, priv_ctx);
+
+	return 0;
+}
+
+static void mgmt_watchdog_report_show(struct devlink_fmsg *fmsg,
+				      struct hinic_mgmt_watchdog_info *winfo)
+{
+	devlink_fmsg_u32_pair_put(fmsg, "Mgmt deadloop time_h", winfo->curr_time_h);
+	devlink_fmsg_u32_pair_put(fmsg, "time_l", winfo->curr_time_l);
+	devlink_fmsg_u32_pair_put(fmsg, "task_id", winfo->task_id);
+	devlink_fmsg_u32_pair_put(fmsg, "sp", winfo->sp);
+	devlink_fmsg_u32_pair_put(fmsg, "stack_current_used", winfo->curr_used);
+	devlink_fmsg_u32_pair_put(fmsg, "peak_used", winfo->peak_used);
+	devlink_fmsg_u32_pair_put(fmsg, "\n Overflow_flag", winfo->is_overflow);
+	devlink_fmsg_u32_pair_put(fmsg, "stack_top", winfo->stack_top);
+	devlink_fmsg_u32_pair_put(fmsg, "stack_bottom", winfo->stack_bottom);
+	devlink_fmsg_u32_pair_put(fmsg, "mgmt_pc", winfo->pc);
+	devlink_fmsg_u32_pair_put(fmsg, "lr", winfo->lr);
+	devlink_fmsg_u32_pair_put(fmsg, "cpsr", winfo->cpsr);
+	devlink_fmsg_binary_pair_put(fmsg, "Mgmt register info", winfo->reg,
+				     sizeof(winfo->reg));
+	devlink_fmsg_binary_pair_put(fmsg, "Mgmt dump stack(start from sp)",
+				     winfo->data, sizeof(winfo->data));
+}
+
+static int hinic_fw_reporter_dump(struct devlink_health_reporter *reporter,
+				  struct devlink_fmsg *fmsg, void *priv_ctx,
+				  struct netlink_ext_ack *extack)
+{
+	if (priv_ctx)
+		mgmt_watchdog_report_show(fmsg, priv_ctx);
+
+	return 0;
+}
+
+static const struct devlink_health_reporter_ops hinic_hw_fault_reporter_ops = {
+	.name = "hw",
+	.dump = hinic_hw_reporter_dump,
+};
+
+static const struct devlink_health_reporter_ops hinic_fw_fault_reporter_ops = {
+	.name = "fw",
+	.dump = hinic_fw_reporter_dump,
+};
+
+int hinic_health_reporters_create(struct hinic_devlink_priv *priv)
+{
+	struct devlink *devlink = priv_to_devlink(priv);
+
+	priv->hw_fault_reporter =
+		devlink_health_reporter_create(devlink,
+					       &hinic_hw_fault_reporter_ops,
+					       priv);
+	if (IS_ERR(priv->hw_fault_reporter)) {
+		dev_warn(&priv->hwdev->hwif->pdev->dev, "Failed to create hw fault reporter, err: %ld\n",
+			 PTR_ERR(priv->hw_fault_reporter));
+		return PTR_ERR(priv->hw_fault_reporter);
+	}
+
+	priv->fw_fault_reporter =
+		devlink_health_reporter_create(devlink,
+					       &hinic_fw_fault_reporter_ops,
+					       priv);
+	if (IS_ERR(priv->fw_fault_reporter)) {
+		dev_warn(&priv->hwdev->hwif->pdev->dev, "Failed to create fw fault reporter, err: %ld\n",
+			 PTR_ERR(priv->fw_fault_reporter));
+		devlink_health_reporter_destroy(priv->hw_fault_reporter);
+		priv->hw_fault_reporter = NULL;
+		return PTR_ERR(priv->fw_fault_reporter);
+	}
+
+	return 0;
+}
+
+void hinic_health_reporters_destroy(struct hinic_devlink_priv *priv)
+{
+	if (!IS_ERR_OR_NULL(priv->fw_fault_reporter)) {
+		devlink_health_reporter_destroy(priv->fw_fault_reporter);
+		priv->fw_fault_reporter = NULL;
+	}
+
+	if (!IS_ERR_OR_NULL(priv->hw_fault_reporter)) {
+		devlink_health_reporter_destroy(priv->hw_fault_reporter);
+		priv->hw_fault_reporter = NULL;
+	}
+}
diff --git a/drivers/net/ethernet/huawei/hinic/hinic_devlink.h b/drivers/net/ethernet/huawei/hinic/hinic_devlink.h
new file mode 100644
index 000000000000..46760d607b9b
--- /dev/null
+++ b/drivers/net/ethernet/huawei/hinic/hinic_devlink.h
@@ -0,0 +1,119 @@
+/* SPDX-License-Identifier: GPL-2.0-only */
+/* Huawei HiNIC PCI Express Linux driver
+ * Copyright(c) 2017 Huawei Technologies Co., Ltd
+ */
+
+#ifndef __HINIC_DEVLINK_H__
+#define __HINIC_DEVLINK_H__
+
+#include <net/devlink.h>
+#include "hinic_dev.h"
+
+#define MAX_FW_TYPE_NUM 30
+#define HINIC_MAGIC_NUM 0x18221100
+#define UPDATEFW_IMAGE_HEAD_SIZE 1024
+#define FW_UPDATE_COLD 0
+#define FW_UPDATE_HOT  1
+
+#define UP_TYPE_A 0x0
+#define UP_TYPE_B 0x1
+
+#define MAX_FW_FRAGMENT_LEN 1536
+#define HINIC_FW_DISMATCH_ERROR 10
+
+enum hinic_fw_type {
+	UP_FW_UPDATE_UP_TEXT_A = 0x0,
+	UP_FW_UPDATE_UP_DATA_A,
+	UP_FW_UPDATE_UP_TEXT_B,
+	UP_FW_UPDATE_UP_DATA_B,
+	UP_FW_UPDATE_UP_DICT,
+
+	UP_FW_UPDATE_HLINK_ONE = 0x5,
+	UP_FW_UPDATE_HLINK_TWO,
+	UP_FW_UPDATE_HLINK_THR,
+	UP_FW_UPDATE_PHY,
+	UP_FW_UPDATE_TILE_TEXT,
+
+	UP_FW_UPDATE_TILE_DATA = 0xa,
+	UP_FW_UPDATE_TILE_DICT,
+	UP_FW_UPDATE_PPE_STATE,
+	UP_FW_UPDATE_PPE_BRANCH,
+	UP_FW_UPDATE_PPE_EXTACT,
+
+	UP_FW_UPDATE_CLP_LEGACY = 0xf,
+	UP_FW_UPDATE_PXE_LEGACY,
+	UP_FW_UPDATE_ISCSI_LEGACY,
+	UP_FW_UPDATE_CLP_EFI,
+	UP_FW_UPDATE_PXE_EFI,
+
+	UP_FW_UPDATE_ISCSI_EFI = 0x14,
+	UP_FW_UPDATE_CFG,
+	UP_FW_UPDATE_BOOT,
+	UP_FW_UPDATE_VPD,
+	FILE_TYPE_TOTAL_NUM
+};
+
+#define _IMAGE_UP_ALL_IN ((1 << UP_FW_UPDATE_UP_TEXT_A) | \
+			  (1 << UP_FW_UPDATE_UP_DATA_A) | \
+			  (1 << UP_FW_UPDATE_UP_TEXT_B) | \
+			  (1 << UP_FW_UPDATE_UP_DATA_B) | \
+			  (1 << UP_FW_UPDATE_UP_DICT) | \
+			  (1 << UP_FW_UPDATE_BOOT) | \
+			  (1 << UP_FW_UPDATE_HLINK_ONE) | \
+			  (1 << UP_FW_UPDATE_HLINK_TWO) | \
+			  (1 << UP_FW_UPDATE_HLINK_THR))
+
+#define _IMAGE_UCODE_ALL_IN ((1 << UP_FW_UPDATE_TILE_TEXT) | \
+			     (1 << UP_FW_UPDATE_TILE_DICT) | \
+			     (1 << UP_FW_UPDATE_PPE_STATE) | \
+			     (1 << UP_FW_UPDATE_PPE_BRANCH) | \
+			     (1 << UP_FW_UPDATE_PPE_EXTACT))
+
+#define _IMAGE_COLD_SUB_MODULES_MUST_IN (_IMAGE_UP_ALL_IN | _IMAGE_UCODE_ALL_IN)
+#define _IMAGE_HOT_SUB_MODULES_MUST_IN (_IMAGE_UP_ALL_IN | _IMAGE_UCODE_ALL_IN)
+#define _IMAGE_CFG_SUB_MODULES_MUST_IN BIT(UP_FW_UPDATE_CFG)
+#define UP_FW_UPDATE_UP_TEXT  0x0
+#define UP_FW_UPDATE_UP_DATA  0x1
+#define UP_FW_UPDATE_VPD_B    0x15
+
+struct fw_section_info_st {
+	u32 fw_section_len;
+	u32 fw_section_offset;
+	u32 fw_section_version;
+	u32 fw_section_type;
+	u32 fw_section_crc;
+};
+
+struct fw_image_st {
+	u32 fw_version;
+	u32 fw_len;
+	u32 fw_magic;
+	struct {
+		u32 fw_section_cnt:16;
+		u32 resd:16;
+	} fw_info;
+	struct fw_section_info_st fw_section_info[MAX_FW_TYPE_NUM];
+	u32 device_id;
+	u32 res[101];
+	void *bin_data;
+};
+
+struct host_image_st {
+	struct fw_section_info_st image_section_info[MAX_FW_TYPE_NUM];
+	struct {
+		u32 up_total_len;
+		u32 fw_version;
+	} image_info;
+	u32 section_type_num;
+	u32 device_id;
+};
+
+struct devlink *hinic_devlink_alloc(struct device *dev);
+void hinic_devlink_free(struct devlink *devlink);
+void hinic_devlink_register(struct hinic_devlink_priv *priv);
+void hinic_devlink_unregister(struct hinic_devlink_priv *priv);
+
+int hinic_health_reporters_create(struct hinic_devlink_priv *priv);
+void hinic_health_reporters_destroy(struct hinic_devlink_priv *priv);
+
+#endif /* __HINIC_DEVLINK_H__ */
diff --git a/drivers/net/ethernet/huawei/hinic/hinic_ethtool.c b/drivers/net/ethernet/huawei/hinic/hinic_ethtool.c
new file mode 100644
index 000000000000..e9f338e9dbe7
--- /dev/null
+++ b/drivers/net/ethernet/huawei/hinic/hinic_ethtool.c
@@ -0,0 +1,1835 @@
+// SPDX-License-Identifier: GPL-2.0
+/* Huawei HiNIC PCI Express Linux driver
+ * Copyright(c) 2017 Huawei Technologies Co., Ltd
+ *
+ * This program is free software; you can redistribute it and/or modify it
+ * under the terms and conditions of the GNU General Public License,
+ * version 2, as published by the Free Software Foundation.
+ *
+ * This program is distributed in the hope it will be useful, but WITHOUT
+ * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
+ * FITNESS FOR A PARTICULAR PURPOSE.  See the GNU General Public License
+ * for more details.
+ *
+ */
+
+#include <linux/kernel.h>
+#include <linux/pci.h>
+#include <linux/device.h>
+#include <linux/module.h>
+#include <linux/types.h>
+#include <linux/errno.h>
+#include <linux/interrupt.h>
+#include <linux/etherdevice.h>
+#include <linux/netdevice.h>
+#include <linux/if_vlan.h>
+#include <linux/ethtool.h>
+#include <linux/vmalloc.h>
+#include <linux/sfp.h>
+
+#include "hinic_hw_qp.h"
+#include "hinic_hw_dev.h"
+#include "hinic_port.h"
+#include "hinic_tx.h"
+#include "hinic_rx.h"
+#include "hinic_dev.h"
+
+#define SET_LINK_STR_MAX_LEN	16
+
+#define GET_SUPPORTED_MODE	0
+#define GET_ADVERTISED_MODE	1
+
+#define ETHTOOL_ADD_SUPPORTED_SPEED_LINK_MODE(ecmd, mode)	\
+		((ecmd)->supported |=	\
+		(1UL << hw_to_ethtool_link_mode_table[mode].link_mode_bit))
+#define ETHTOOL_ADD_ADVERTISED_SPEED_LINK_MODE(ecmd, mode)	\
+		((ecmd)->advertising |=	\
+		(1UL << hw_to_ethtool_link_mode_table[mode].link_mode_bit))
+#define ETHTOOL_ADD_SUPPORTED_LINK_MODE(ecmd, mode)	\
+				((ecmd)->supported |= SUPPORTED_##mode)
+#define ETHTOOL_ADD_ADVERTISED_LINK_MODE(ecmd, mode)	\
+				((ecmd)->advertising |= ADVERTISED_##mode)
+
+#define COALESCE_PENDING_LIMIT_UNIT	8
+#define	COALESCE_TIMER_CFG_UNIT		9
+#define COALESCE_ALL_QUEUE		0xFFFF
+#define COALESCE_MAX_PENDING_LIMIT	(255 * COALESCE_PENDING_LIMIT_UNIT)
+#define COALESCE_MAX_TIMER_CFG		(255 * COALESCE_TIMER_CFG_UNIT)
+
+struct hw2ethtool_link_mode {
+	enum ethtool_link_mode_bit_indices link_mode_bit;
+	u32 speed;
+	enum hinic_link_mode hw_link_mode;
+};
+
+struct cmd_link_settings {
+	u64	supported;
+	u64	advertising;
+
+	u32	speed;
+	u8	duplex;
+	u8	port;
+	u8	autoneg;
+};
+
+static u32 hw_to_ethtool_speed[LINK_SPEED_LEVELS] = {
+	SPEED_10, SPEED_100,
+	SPEED_1000, SPEED_10000,
+	SPEED_25000, SPEED_40000,
+	SPEED_100000
+};
+
+static struct hw2ethtool_link_mode
+	hw_to_ethtool_link_mode_table[HINIC_LINK_MODE_NUMBERS] = {
+	{
+		.link_mode_bit = ETHTOOL_LINK_MODE_10000baseKR_Full_BIT,
+		.speed = SPEED_10000,
+		.hw_link_mode = HINIC_10GE_BASE_KR,
+	},
+	{
+		.link_mode_bit = ETHTOOL_LINK_MODE_40000baseKR4_Full_BIT,
+		.speed = SPEED_40000,
+		.hw_link_mode = HINIC_40GE_BASE_KR4,
+	},
+	{
+		.link_mode_bit = ETHTOOL_LINK_MODE_40000baseCR4_Full_BIT,
+		.speed = SPEED_40000,
+		.hw_link_mode = HINIC_40GE_BASE_CR4,
+	},
+	{
+		.link_mode_bit = ETHTOOL_LINK_MODE_100000baseKR4_Full_BIT,
+		.speed = SPEED_100000,
+		.hw_link_mode = HINIC_100GE_BASE_KR4,
+	},
+	{
+		.link_mode_bit = ETHTOOL_LINK_MODE_100000baseCR4_Full_BIT,
+		.speed = SPEED_100000,
+		.hw_link_mode = HINIC_100GE_BASE_CR4,
+	},
+	{
+		.link_mode_bit = ETHTOOL_LINK_MODE_25000baseKR_Full_BIT,
+		.speed = SPEED_25000,
+		.hw_link_mode = HINIC_25GE_BASE_KR_S,
+	},
+	{
+		.link_mode_bit = ETHTOOL_LINK_MODE_25000baseCR_Full_BIT,
+		.speed = SPEED_25000,
+		.hw_link_mode = HINIC_25GE_BASE_CR_S,
+	},
+	{
+		.link_mode_bit = ETHTOOL_LINK_MODE_25000baseKR_Full_BIT,
+		.speed = SPEED_25000,
+		.hw_link_mode = HINIC_25GE_BASE_KR,
+	},
+	{
+		.link_mode_bit = ETHTOOL_LINK_MODE_25000baseCR_Full_BIT,
+		.speed = SPEED_25000,
+		.hw_link_mode = HINIC_25GE_BASE_CR,
+	},
+	{
+		.link_mode_bit = ETHTOOL_LINK_MODE_1000baseKX_Full_BIT,
+		.speed = SPEED_1000,
+		.hw_link_mode = HINIC_GE_BASE_KX,
+	},
+};
+
+#define LP_DEFAULT_TIME                 5 /* seconds */
+#define LP_PKT_LEN                      1514
+
+#define PORT_DOWN_ERR_IDX		0
+enum diag_test_index {
+	INTERNAL_LP_TEST = 0,
+	EXTERNAL_LP_TEST = 1,
+	DIAG_TEST_MAX = 2,
+};
+
+static void set_link_speed(struct ethtool_link_ksettings *link_ksettings,
+			   enum hinic_speed speed)
+{
+	switch (speed) {
+	case HINIC_SPEED_10MB_LINK:
+		link_ksettings->base.speed = SPEED_10;
+		break;
+
+	case HINIC_SPEED_100MB_LINK:
+		link_ksettings->base.speed = SPEED_100;
+		break;
+
+	case HINIC_SPEED_1000MB_LINK:
+		link_ksettings->base.speed = SPEED_1000;
+		break;
+
+	case HINIC_SPEED_10GB_LINK:
+		link_ksettings->base.speed = SPEED_10000;
+		break;
+
+	case HINIC_SPEED_25GB_LINK:
+		link_ksettings->base.speed = SPEED_25000;
+		break;
+
+	case HINIC_SPEED_40GB_LINK:
+		link_ksettings->base.speed = SPEED_40000;
+		break;
+
+	case HINIC_SPEED_100GB_LINK:
+		link_ksettings->base.speed = SPEED_100000;
+		break;
+
+	default:
+		link_ksettings->base.speed = SPEED_UNKNOWN;
+		break;
+	}
+}
+
+static int hinic_get_link_mode_index(enum hinic_link_mode link_mode)
+{
+	int i = 0;
+
+	for (i = 0; i < HINIC_LINK_MODE_NUMBERS; i++) {
+		if (link_mode == hw_to_ethtool_link_mode_table[i].hw_link_mode)
+			break;
+	}
+
+	return i;
+}
+
+static void hinic_add_ethtool_link_mode(struct cmd_link_settings *link_settings,
+					enum hinic_link_mode hw_link_mode,
+					u32 name)
+{
+	enum hinic_link_mode link_mode;
+	int idx = 0;
+
+	for (link_mode = 0; link_mode < HINIC_LINK_MODE_NUMBERS; link_mode++) {
+		if (hw_link_mode & ((u32)1 << link_mode)) {
+			idx = hinic_get_link_mode_index(link_mode);
+			if (idx >= HINIC_LINK_MODE_NUMBERS)
+				continue;
+
+			if (name == GET_SUPPORTED_MODE)
+				ETHTOOL_ADD_SUPPORTED_SPEED_LINK_MODE
+					(link_settings, idx);
+			else
+				ETHTOOL_ADD_ADVERTISED_SPEED_LINK_MODE
+					(link_settings, idx);
+		}
+	}
+}
+
+static void hinic_link_port_type(struct cmd_link_settings *link_settings,
+				 enum hinic_port_type port_type)
+{
+	switch (port_type) {
+	case HINIC_PORT_ELEC:
+	case HINIC_PORT_TP:
+		ETHTOOL_ADD_SUPPORTED_LINK_MODE(link_settings, TP);
+		ETHTOOL_ADD_ADVERTISED_LINK_MODE(link_settings, TP);
+		link_settings->port = PORT_TP;
+		break;
+
+	case HINIC_PORT_AOC:
+	case HINIC_PORT_FIBRE:
+		ETHTOOL_ADD_SUPPORTED_LINK_MODE(link_settings, FIBRE);
+		ETHTOOL_ADD_ADVERTISED_LINK_MODE(link_settings, FIBRE);
+		link_settings->port = PORT_FIBRE;
+		break;
+
+	case HINIC_PORT_COPPER:
+		ETHTOOL_ADD_SUPPORTED_LINK_MODE(link_settings, FIBRE);
+		ETHTOOL_ADD_ADVERTISED_LINK_MODE(link_settings, FIBRE);
+		link_settings->port = PORT_DA;
+		break;
+
+	case HINIC_PORT_BACKPLANE:
+		ETHTOOL_ADD_SUPPORTED_LINK_MODE(link_settings, Backplane);
+		ETHTOOL_ADD_ADVERTISED_LINK_MODE(link_settings, Backplane);
+		link_settings->port = PORT_NONE;
+		break;
+
+	default:
+		link_settings->port = PORT_OTHER;
+		break;
+	}
+}
+
+static int hinic_get_link_ksettings(struct net_device *netdev,
+				    struct ethtool_link_ksettings
+				    *link_ksettings)
+{
+	struct hinic_dev *nic_dev = netdev_priv(netdev);
+	struct hinic_link_mode_cmd link_mode = { 0 };
+	struct hinic_pause_config pause_info = { 0 };
+	struct cmd_link_settings settings = { 0 };
+	enum hinic_port_link_state link_state;
+	struct hinic_port_cap port_cap;
+	int err;
+
+	ethtool_link_ksettings_zero_link_mode(link_ksettings, supported);
+	ethtool_link_ksettings_zero_link_mode(link_ksettings, advertising);
+
+	link_ksettings->base.speed = SPEED_UNKNOWN;
+	link_ksettings->base.autoneg = AUTONEG_DISABLE;
+	link_ksettings->base.duplex = DUPLEX_UNKNOWN;
+
+	err = hinic_port_get_cap(nic_dev, &port_cap);
+	if (err)
+		return err;
+
+	hinic_link_port_type(&settings, port_cap.port_type);
+	link_ksettings->base.port = settings.port;
+
+	err = hinic_port_link_state(nic_dev, &link_state);
+	if (err)
+		return err;
+
+	if (link_state == HINIC_LINK_STATE_UP) {
+		set_link_speed(link_ksettings, port_cap.speed);
+		link_ksettings->base.duplex =
+			(port_cap.duplex == HINIC_DUPLEX_FULL) ?
+			DUPLEX_FULL : DUPLEX_HALF;
+	}
+
+	if (!!(port_cap.autoneg_cap & HINIC_AUTONEG_SUPPORTED))
+		ethtool_link_ksettings_add_link_mode(link_ksettings,
+						     advertising, Autoneg);
+
+	if (port_cap.autoneg_state == HINIC_AUTONEG_ACTIVE)
+		link_ksettings->base.autoneg = AUTONEG_ENABLE;
+
+	err = hinic_get_link_mode(nic_dev->hwdev, &link_mode);
+	if (err || link_mode.supported == HINIC_SUPPORTED_UNKNOWN ||
+	    link_mode.advertised == HINIC_SUPPORTED_UNKNOWN)
+		return -EIO;
+
+	hinic_add_ethtool_link_mode(&settings, link_mode.supported,
+				    GET_SUPPORTED_MODE);
+	hinic_add_ethtool_link_mode(&settings, link_mode.advertised,
+				    GET_ADVERTISED_MODE);
+
+	if (!HINIC_IS_VF(nic_dev->hwdev->hwif)) {
+		err = hinic_get_hw_pause_info(nic_dev->hwdev, &pause_info);
+		if (err)
+			return err;
+		ETHTOOL_ADD_SUPPORTED_LINK_MODE(&settings, Pause);
+		if (pause_info.rx_pause && pause_info.tx_pause) {
+			ETHTOOL_ADD_ADVERTISED_LINK_MODE(&settings, Pause);
+		} else if (pause_info.tx_pause) {
+			ETHTOOL_ADD_ADVERTISED_LINK_MODE(&settings, Asym_Pause);
+		} else if (pause_info.rx_pause) {
+			ETHTOOL_ADD_ADVERTISED_LINK_MODE(&settings, Pause);
+			ETHTOOL_ADD_ADVERTISED_LINK_MODE(&settings, Asym_Pause);
+		}
+	}
+
+	linkmode_copy(link_ksettings->link_modes.supported,
+		      (unsigned long *)&settings.supported);
+	linkmode_copy(link_ksettings->link_modes.advertising,
+		      (unsigned long *)&settings.advertising);
+
+	return 0;
+}
+
+static int hinic_ethtool_to_hw_speed_level(u32 speed)
+{
+	int i;
+
+	for (i = 0; i < LINK_SPEED_LEVELS; i++) {
+		if (hw_to_ethtool_speed[i] == speed)
+			break;
+	}
+
+	return i;
+}
+
+static bool hinic_is_support_speed(enum hinic_link_mode supported_link,
+				   u32 speed)
+{
+	enum hinic_link_mode link_mode;
+	int idx;
+
+	for (link_mode = 0; link_mode < HINIC_LINK_MODE_NUMBERS; link_mode++) {
+		if (!(supported_link & ((u32)1 << link_mode)))
+			continue;
+
+		idx = hinic_get_link_mode_index(link_mode);
+		if (idx >= HINIC_LINK_MODE_NUMBERS)
+			continue;
+
+		if (hw_to_ethtool_link_mode_table[idx].speed == speed)
+			return true;
+	}
+
+	return false;
+}
+
+static bool hinic_is_speed_legal(struct hinic_dev *nic_dev, u32 speed)
+{
+	struct hinic_link_mode_cmd link_mode = { 0 };
+	struct net_device *netdev = nic_dev->netdev;
+	enum nic_speed_level speed_level = 0;
+	int err;
+
+	err = hinic_get_link_mode(nic_dev->hwdev, &link_mode);
+	if (err)
+		return false;
+
+	if (link_mode.supported == HINIC_SUPPORTED_UNKNOWN ||
+	    link_mode.advertised == HINIC_SUPPORTED_UNKNOWN)
+		return false;
+
+	speed_level = hinic_ethtool_to_hw_speed_level(speed);
+	if (speed_level >= LINK_SPEED_LEVELS ||
+	    !hinic_is_support_speed(link_mode.supported, speed)) {
+		netif_err(nic_dev, drv, netdev,
+			  "Unsupported speed: %d\n", speed);
+		return false;
+	}
+
+	return true;
+}
+
+static int get_link_settings_type(struct hinic_dev *nic_dev,
+				  u8 autoneg, u32 speed, u32 *set_settings)
+{
+	struct hinic_port_cap port_cap = { 0 };
+	int err;
+
+	err = hinic_port_get_cap(nic_dev, &port_cap);
+	if (err)
+		return err;
+
+	/* always set autonegotiation */
+	if (port_cap.autoneg_cap)
+		*set_settings |= HILINK_LINK_SET_AUTONEG;
+
+	if (autoneg == AUTONEG_ENABLE) {
+		if (!port_cap.autoneg_cap) {
+			netif_err(nic_dev, drv, nic_dev->netdev, "Not support autoneg\n");
+			return -EOPNOTSUPP;
+		}
+	} else if (speed != (u32)SPEED_UNKNOWN) {
+		/* set speed only when autoneg is disabled */
+		if (!hinic_is_speed_legal(nic_dev, speed))
+			return -EINVAL;
+		*set_settings |= HILINK_LINK_SET_SPEED;
+	} else {
+		netif_err(nic_dev, drv, nic_dev->netdev, "Need to set speed when autoneg is off\n");
+		return -EOPNOTSUPP;
+	}
+
+	return 0;
+}
+
+static int set_link_settings_separate_cmd(struct hinic_dev *nic_dev,
+					  u32 set_settings, u8 autoneg,
+					  u32 speed)
+{
+	enum nic_speed_level speed_level = 0;
+	int err = 0;
+
+	if (set_settings & HILINK_LINK_SET_AUTONEG) {
+		err = hinic_set_autoneg(nic_dev->hwdev,
+					(autoneg == AUTONEG_ENABLE));
+		if (err)
+			netif_err(nic_dev, drv, nic_dev->netdev, "%s autoneg failed\n",
+				  (autoneg == AUTONEG_ENABLE) ?
+				  "Enable" : "Disable");
+		else
+			netif_info(nic_dev, drv, nic_dev->netdev, "%s autoneg successfully\n",
+				   (autoneg == AUTONEG_ENABLE) ?
+				   "Enable" : "Disable");
+	}
+
+	if (!err && (set_settings & HILINK_LINK_SET_SPEED)) {
+		speed_level = hinic_ethtool_to_hw_speed_level(speed);
+		err = hinic_set_speed(nic_dev->hwdev, speed_level);
+		if (err)
+			netif_err(nic_dev, drv, nic_dev->netdev, "Set speed %d failed\n",
+				  speed);
+		else
+			netif_info(nic_dev, drv, nic_dev->netdev, "Set speed %d successfully\n",
+				   speed);
+	}
+
+	return err;
+}
+
+static int hinic_set_settings_to_hw(struct hinic_dev *nic_dev,
+				    u32 set_settings, u8 autoneg, u32 speed)
+{
+	struct hinic_link_ksettings_info settings = {0};
+	char set_link_str[SET_LINK_STR_MAX_LEN] = {0};
+	const char *autoneg_str;
+	struct net_device *netdev = nic_dev->netdev;
+	enum nic_speed_level speed_level = 0;
+	int err;
+
+	autoneg_str = (set_settings & HILINK_LINK_SET_AUTONEG) ?
+		      (autoneg ? "autong enable " : "autong disable ") : "";
+
+	if (set_settings & HILINK_LINK_SET_SPEED) {
+		speed_level = hinic_ethtool_to_hw_speed_level(speed);
+		err = snprintf(set_link_str, SET_LINK_STR_MAX_LEN,
+			       "speed %d ", speed);
+		if (err >= SET_LINK_STR_MAX_LEN) {
+			netif_err(nic_dev, drv, netdev, "Failed to snprintf link speed, function return(%d) and dest_len(%d)\n",
+				  err, SET_LINK_STR_MAX_LEN);
+			return -EFAULT;
+		}
+	}
+
+	settings.func_id = HINIC_HWIF_FUNC_IDX(nic_dev->hwdev->hwif);
+	settings.valid_bitmap = set_settings;
+	settings.autoneg = autoneg;
+	settings.speed = speed_level;
+
+	err = hinic_set_link_settings(nic_dev->hwdev, &settings);
+	if (err != HINIC_MGMT_CMD_UNSUPPORTED) {
+		if (err)
+			netif_err(nic_dev, drv, netdev, "Set %s%sfailed\n",
+				  autoneg_str, set_link_str);
+		else
+			netif_info(nic_dev, drv, netdev, "Set %s%ssuccessfully\n",
+				   autoneg_str, set_link_str);
+
+		return err;
+	}
+
+	return set_link_settings_separate_cmd(nic_dev, set_settings, autoneg,
+					      speed);
+}
+
+static int set_link_settings(struct net_device *netdev, u8 autoneg, u32 speed)
+{
+	struct hinic_dev *nic_dev = netdev_priv(netdev);
+	u32 set_settings = 0;
+	int err;
+
+	err = get_link_settings_type(nic_dev, autoneg, speed, &set_settings);
+	if (err)
+		return err;
+
+	if (set_settings)
+		err = hinic_set_settings_to_hw(nic_dev, set_settings,
+					       autoneg, speed);
+	else
+		netif_info(nic_dev, drv, netdev, "Nothing changed, exit without setting anything\n");
+
+	return err;
+}
+
+static int hinic_set_link_ksettings(struct net_device *netdev, const struct
+				    ethtool_link_ksettings *link_settings)
+{
+	/* only support to set autoneg and speed */
+	return set_link_settings(netdev, link_settings->base.autoneg,
+				 link_settings->base.speed);
+}
+
+static void hinic_get_drvinfo(struct net_device *netdev,
+			      struct ethtool_drvinfo *info)
+{
+	struct hinic_dev *nic_dev = netdev_priv(netdev);
+	u8 mgmt_ver[HINIC_MGMT_VERSION_MAX_LEN] = {0};
+	struct hinic_hwdev *hwdev = nic_dev->hwdev;
+	struct hinic_hwif *hwif = hwdev->hwif;
+	int err;
+
+	strscpy(info->driver, HINIC_DRV_NAME, sizeof(info->driver));
+	strscpy(info->bus_info, pci_name(hwif->pdev), sizeof(info->bus_info));
+
+	err = hinic_get_mgmt_version(nic_dev, mgmt_ver);
+	if (err)
+		return;
+
+	snprintf(info->fw_version, sizeof(info->fw_version), "%s", mgmt_ver);
+}
+
+static void hinic_get_ringparam(struct net_device *netdev,
+				struct ethtool_ringparam *ring,
+				struct kernel_ethtool_ringparam *kernel_ring,
+				struct netlink_ext_ack *extack)
+{
+	struct hinic_dev *nic_dev = netdev_priv(netdev);
+
+	ring->rx_max_pending = HINIC_MAX_QUEUE_DEPTH;
+	ring->tx_max_pending = HINIC_MAX_QUEUE_DEPTH;
+	ring->rx_pending = nic_dev->rq_depth;
+	ring->tx_pending = nic_dev->sq_depth;
+}
+
+static int check_ringparam_valid(struct hinic_dev *nic_dev,
+				 struct ethtool_ringparam *ring)
+{
+	if (ring->rx_jumbo_pending || ring->rx_mini_pending) {
+		netif_err(nic_dev, drv, nic_dev->netdev,
+			  "Unsupported rx_jumbo_pending/rx_mini_pending\n");
+		return -EINVAL;
+	}
+
+	if (ring->tx_pending > HINIC_MAX_QUEUE_DEPTH ||
+	    ring->tx_pending < HINIC_MIN_QUEUE_DEPTH ||
+	    ring->rx_pending > HINIC_MAX_QUEUE_DEPTH ||
+	    ring->rx_pending < HINIC_MIN_QUEUE_DEPTH) {
+		netif_err(nic_dev, drv, nic_dev->netdev,
+			  "Queue depth out of range [%d-%d]\n",
+			  HINIC_MIN_QUEUE_DEPTH, HINIC_MAX_QUEUE_DEPTH);
+		return -EINVAL;
+	}
+
+	return 0;
+}
+
+static int hinic_set_ringparam(struct net_device *netdev,
+			       struct ethtool_ringparam *ring,
+			       struct kernel_ethtool_ringparam *kernel_ring,
+			       struct netlink_ext_ack *extack)
+{
+	struct hinic_dev *nic_dev = netdev_priv(netdev);
+	u16 new_sq_depth, new_rq_depth;
+	int err;
+
+	err = check_ringparam_valid(nic_dev, ring);
+	if (err)
+		return err;
+
+	new_sq_depth = (u16)(1U << (u16)ilog2(ring->tx_pending));
+	new_rq_depth = (u16)(1U << (u16)ilog2(ring->rx_pending));
+
+	if (new_sq_depth == nic_dev->sq_depth &&
+	    new_rq_depth == nic_dev->rq_depth)
+		return 0;
+
+	netif_info(nic_dev, drv, netdev,
+		   "Change Tx/Rx ring depth from %d/%d to %d/%d\n",
+		   nic_dev->sq_depth, nic_dev->rq_depth,
+		   new_sq_depth, new_rq_depth);
+
+	nic_dev->sq_depth = new_sq_depth;
+	nic_dev->rq_depth = new_rq_depth;
+
+	if (netif_running(netdev)) {
+		netif_info(nic_dev, drv, netdev, "Restarting netdev\n");
+		err = hinic_close(netdev);
+		if (err) {
+			netif_err(nic_dev, drv, netdev,
+				  "Failed to close netdev\n");
+			return -EFAULT;
+		}
+
+		err = hinic_open(netdev);
+		if (err) {
+			netif_err(nic_dev, drv, netdev,
+				  "Failed to open netdev\n");
+			return -EFAULT;
+		}
+	}
+
+	return 0;
+}
+
+static int __hinic_get_coalesce(struct net_device *netdev,
+				struct ethtool_coalesce *coal, u16 queue)
+{
+	struct hinic_dev *nic_dev = netdev_priv(netdev);
+	struct hinic_intr_coal_info *rx_intr_coal_info;
+	struct hinic_intr_coal_info *tx_intr_coal_info;
+
+	if (queue == COALESCE_ALL_QUEUE) {
+		/* get tx/rx irq0 as default parameters */
+		rx_intr_coal_info = &nic_dev->rx_intr_coalesce[0];
+		tx_intr_coal_info = &nic_dev->tx_intr_coalesce[0];
+	} else {
+		if (queue >= nic_dev->num_qps) {
+			netif_err(nic_dev, drv, netdev,
+				  "Invalid queue_id: %d\n", queue);
+			return -EINVAL;
+		}
+		rx_intr_coal_info = &nic_dev->rx_intr_coalesce[queue];
+		tx_intr_coal_info = &nic_dev->tx_intr_coalesce[queue];
+	}
+
+	/* coalesce_timer is in unit of 9us */
+	coal->rx_coalesce_usecs = rx_intr_coal_info->coalesce_timer_cfg *
+			COALESCE_TIMER_CFG_UNIT;
+	/* coalesced_frames is in unit of 8 */
+	coal->rx_max_coalesced_frames = rx_intr_coal_info->pending_limt *
+			COALESCE_PENDING_LIMIT_UNIT;
+	coal->tx_coalesce_usecs = tx_intr_coal_info->coalesce_timer_cfg *
+			COALESCE_TIMER_CFG_UNIT;
+	coal->tx_max_coalesced_frames = tx_intr_coal_info->pending_limt *
+			COALESCE_PENDING_LIMIT_UNIT;
+
+	return 0;
+}
+
+static int is_coalesce_exceed_limit(const struct ethtool_coalesce *coal)
+{
+	if (coal->rx_coalesce_usecs > COALESCE_MAX_TIMER_CFG ||
+	    coal->rx_max_coalesced_frames > COALESCE_MAX_PENDING_LIMIT ||
+	    coal->tx_coalesce_usecs > COALESCE_MAX_TIMER_CFG ||
+	    coal->tx_max_coalesced_frames > COALESCE_MAX_PENDING_LIMIT)
+		return -ERANGE;
+
+	return 0;
+}
+
+static int set_queue_coalesce(struct hinic_dev *nic_dev, u16 q_id,
+			      struct hinic_intr_coal_info *coal,
+			      bool set_rx_coal)
+{
+	struct hinic_intr_coal_info *intr_coal = NULL;
+	struct hinic_msix_config interrupt_info = {0};
+	struct net_device *netdev = nic_dev->netdev;
+	u16 msix_idx;
+	int err;
+
+	intr_coal = set_rx_coal ? &nic_dev->rx_intr_coalesce[q_id] :
+		    &nic_dev->tx_intr_coalesce[q_id];
+
+	intr_coal->coalesce_timer_cfg = coal->coalesce_timer_cfg;
+	intr_coal->pending_limt = coal->pending_limt;
+
+	/* netdev not running or qp not in using,
+	 * don't need to set coalesce to hw
+	 */
+	if (!(nic_dev->flags & HINIC_INTF_UP) ||
+	    q_id >= nic_dev->num_qps)
+		return 0;
+
+	msix_idx = set_rx_coal ? nic_dev->rxqs[q_id].rq->msix_entry :
+		   nic_dev->txqs[q_id].sq->msix_entry;
+	interrupt_info.msix_index = msix_idx;
+	interrupt_info.coalesce_timer_cnt = intr_coal->coalesce_timer_cfg;
+	interrupt_info.pending_cnt = intr_coal->pending_limt;
+	interrupt_info.resend_timer_cnt = intr_coal->resend_timer_cfg;
+
+	err = hinic_set_interrupt_cfg(nic_dev->hwdev, &interrupt_info);
+	if (err)
+		netif_warn(nic_dev, drv, netdev,
+			   "Failed to set %s queue%d coalesce",
+			   set_rx_coal ? "rx" : "tx", q_id);
+
+	return err;
+}
+
+static int __set_hw_coal_param(struct hinic_dev *nic_dev,
+			       struct hinic_intr_coal_info *intr_coal,
+			       u16 queue, bool set_rx_coal)
+{
+	int err;
+	u16 i;
+
+	if (queue == COALESCE_ALL_QUEUE) {
+		for (i = 0; i < nic_dev->max_qps; i++) {
+			err = set_queue_coalesce(nic_dev, i, intr_coal,
+						 set_rx_coal);
+			if (err)
+				return err;
+		}
+	} else {
+		if (queue >= nic_dev->num_qps) {
+			netif_err(nic_dev, drv, nic_dev->netdev,
+				  "Invalid queue_id: %d\n", queue);
+			return -EINVAL;
+		}
+		err = set_queue_coalesce(nic_dev, queue, intr_coal,
+					 set_rx_coal);
+		if (err)
+			return err;
+	}
+
+	return 0;
+}
+
+static int __hinic_set_coalesce(struct net_device *netdev,
+				struct ethtool_coalesce *coal, u16 queue)
+{
+	struct hinic_dev *nic_dev = netdev_priv(netdev);
+	struct hinic_intr_coal_info rx_intr_coal = {0};
+	struct hinic_intr_coal_info tx_intr_coal = {0};
+	bool set_rx_coal = false;
+	bool set_tx_coal = false;
+	int err;
+
+	err = is_coalesce_exceed_limit(coal);
+	if (err)
+		return err;
+
+	if (coal->rx_coalesce_usecs || coal->rx_max_coalesced_frames) {
+		rx_intr_coal.coalesce_timer_cfg =
+		(u8)(coal->rx_coalesce_usecs / COALESCE_TIMER_CFG_UNIT);
+		rx_intr_coal.pending_limt = (u8)(coal->rx_max_coalesced_frames /
+				COALESCE_PENDING_LIMIT_UNIT);
+		set_rx_coal = true;
+	}
+
+	if (coal->tx_coalesce_usecs || coal->tx_max_coalesced_frames) {
+		tx_intr_coal.coalesce_timer_cfg =
+		(u8)(coal->tx_coalesce_usecs / COALESCE_TIMER_CFG_UNIT);
+		tx_intr_coal.pending_limt = (u8)(coal->tx_max_coalesced_frames /
+		COALESCE_PENDING_LIMIT_UNIT);
+		set_tx_coal = true;
+	}
+
+	/* setting coalesce timer or pending limit to zero will disable
+	 * coalesce
+	 */
+	if (set_rx_coal && (!rx_intr_coal.coalesce_timer_cfg ||
+			    !rx_intr_coal.pending_limt))
+		netif_warn(nic_dev, drv, netdev, "RX coalesce will be disabled\n");
+	if (set_tx_coal && (!tx_intr_coal.coalesce_timer_cfg ||
+			    !tx_intr_coal.pending_limt))
+		netif_warn(nic_dev, drv, netdev, "TX coalesce will be disabled\n");
+
+	if (set_rx_coal) {
+		err = __set_hw_coal_param(nic_dev, &rx_intr_coal, queue, true);
+		if (err)
+			return err;
+	}
+	if (set_tx_coal) {
+		err = __set_hw_coal_param(nic_dev, &tx_intr_coal, queue, false);
+		if (err)
+			return err;
+	}
+	return 0;
+}
+
+static int hinic_get_coalesce(struct net_device *netdev,
+			      struct ethtool_coalesce *coal,
+			      struct kernel_ethtool_coalesce *kernel_coal,
+			      struct netlink_ext_ack *extack)
+{
+	return __hinic_get_coalesce(netdev, coal, COALESCE_ALL_QUEUE);
+}
+
+static int hinic_set_coalesce(struct net_device *netdev,
+			      struct ethtool_coalesce *coal,
+			      struct kernel_ethtool_coalesce *kernel_coal,
+			      struct netlink_ext_ack *extack)
+{
+	return __hinic_set_coalesce(netdev, coal, COALESCE_ALL_QUEUE);
+}
+
+static int hinic_get_per_queue_coalesce(struct net_device *netdev, u32 queue,
+					struct ethtool_coalesce *coal)
+{
+	return __hinic_get_coalesce(netdev, coal, queue);
+}
+
+static int hinic_set_per_queue_coalesce(struct net_device *netdev, u32 queue,
+					struct ethtool_coalesce *coal)
+{
+	return __hinic_set_coalesce(netdev, coal, queue);
+}
+
+static void hinic_get_pauseparam(struct net_device *netdev,
+				 struct ethtool_pauseparam *pause)
+{
+	struct hinic_dev *nic_dev = netdev_priv(netdev);
+	struct hinic_pause_config pause_info = {0};
+	struct hinic_nic_cfg *nic_cfg;
+	int err;
+
+	nic_cfg = &nic_dev->hwdev->func_to_io.nic_cfg;
+
+	err = hinic_get_hw_pause_info(nic_dev->hwdev, &pause_info);
+	if (!err) {
+		pause->autoneg = pause_info.auto_neg;
+		if (nic_cfg->pause_set || !pause_info.auto_neg) {
+			pause->rx_pause = nic_cfg->rx_pause;
+			pause->tx_pause = nic_cfg->tx_pause;
+		} else {
+			pause->rx_pause = pause_info.rx_pause;
+			pause->tx_pause = pause_info.tx_pause;
+		}
+	}
+}
+
+static int hinic_set_pauseparam(struct net_device *netdev,
+				struct ethtool_pauseparam *pause)
+{
+	struct hinic_dev *nic_dev = netdev_priv(netdev);
+	struct hinic_pause_config pause_info = {0};
+	struct hinic_port_cap port_cap = {0};
+	int err;
+
+	err = hinic_port_get_cap(nic_dev, &port_cap);
+	if (err)
+		return -EIO;
+
+	if (pause->autoneg != port_cap.autoneg_state)
+		return -EOPNOTSUPP;
+
+	pause_info.auto_neg = pause->autoneg;
+	pause_info.rx_pause = pause->rx_pause;
+	pause_info.tx_pause = pause->tx_pause;
+
+	mutex_lock(&nic_dev->hwdev->func_to_io.nic_cfg.cfg_mutex);
+	err = hinic_set_hw_pause_info(nic_dev->hwdev, &pause_info);
+	if (err) {
+		mutex_unlock(&nic_dev->hwdev->func_to_io.nic_cfg.cfg_mutex);
+		return err;
+	}
+	nic_dev->hwdev->func_to_io.nic_cfg.pause_set = true;
+	nic_dev->hwdev->func_to_io.nic_cfg.auto_neg = pause->autoneg;
+	nic_dev->hwdev->func_to_io.nic_cfg.rx_pause = pause->rx_pause;
+	nic_dev->hwdev->func_to_io.nic_cfg.tx_pause = pause->tx_pause;
+	mutex_unlock(&nic_dev->hwdev->func_to_io.nic_cfg.cfg_mutex);
+
+	return 0;
+}
+
+static void hinic_get_channels(struct net_device *netdev,
+			       struct ethtool_channels *channels)
+{
+	struct hinic_dev *nic_dev = netdev_priv(netdev);
+	struct hinic_hwdev *hwdev = nic_dev->hwdev;
+
+	channels->max_combined = nic_dev->max_qps;
+	channels->combined_count = hinic_hwdev_num_qps(hwdev);
+}
+
+static int hinic_set_channels(struct net_device *netdev,
+			      struct ethtool_channels *channels)
+{
+	struct hinic_dev *nic_dev = netdev_priv(netdev);
+	unsigned int count = channels->combined_count;
+	int err;
+
+	netif_info(nic_dev, drv, netdev, "Set max combined queue number from %d to %d\n",
+		   hinic_hwdev_num_qps(nic_dev->hwdev), count);
+
+	if (netif_running(netdev)) {
+		netif_info(nic_dev, drv, netdev, "Restarting netdev\n");
+		hinic_close(netdev);
+
+		nic_dev->hwdev->nic_cap.num_qps = count;
+
+		err = hinic_open(netdev);
+		if (err) {
+			netif_err(nic_dev, drv, netdev,
+				  "Failed to open netdev\n");
+			return -EFAULT;
+		}
+	} else {
+		nic_dev->hwdev->nic_cap.num_qps = count;
+	}
+
+	return 0;
+}
+
+static int hinic_get_rxfh_fields(struct net_device *netdev,
+				 struct ethtool_rxfh_fields *cmd)
+{
+	struct hinic_dev *nic_dev = netdev_priv(netdev);
+	struct hinic_rss_type rss_type = { 0 };
+	int err;
+
+	cmd->data = 0;
+
+	if (!(nic_dev->flags & HINIC_RSS_ENABLE))
+		return 0;
+
+	err = hinic_get_rss_type(nic_dev, nic_dev->rss_tmpl_idx,
+				 &rss_type);
+	if (err)
+		return err;
+
+	cmd->data = RXH_IP_SRC | RXH_IP_DST;
+	switch (cmd->flow_type) {
+	case TCP_V4_FLOW:
+		if (rss_type.tcp_ipv4)
+			cmd->data |= RXH_L4_B_0_1 | RXH_L4_B_2_3;
+		break;
+	case TCP_V6_FLOW:
+		if (rss_type.tcp_ipv6)
+			cmd->data |= RXH_L4_B_0_1 | RXH_L4_B_2_3;
+		break;
+	case UDP_V4_FLOW:
+		if (rss_type.udp_ipv4)
+			cmd->data |= RXH_L4_B_0_1 | RXH_L4_B_2_3;
+		break;
+	case UDP_V6_FLOW:
+		if (rss_type.udp_ipv6)
+			cmd->data |= RXH_L4_B_0_1 | RXH_L4_B_2_3;
+		break;
+	case IPV4_FLOW:
+	case IPV6_FLOW:
+		break;
+	default:
+		cmd->data = 0;
+		return -EINVAL;
+	}
+
+	return 0;
+}
+
+static int set_l4_rss_hash_ops(const struct ethtool_rxfh_fields *cmd,
+			       struct hinic_rss_type *rss_type)
+{
+	u8 rss_l4_en = 0;
+
+	switch (cmd->data & (RXH_L4_B_0_1 | RXH_L4_B_2_3)) {
+	case 0:
+		rss_l4_en = 0;
+		break;
+	case (RXH_L4_B_0_1 | RXH_L4_B_2_3):
+		rss_l4_en = 1;
+		break;
+	default:
+		return -EINVAL;
+	}
+
+	switch (cmd->flow_type) {
+	case TCP_V4_FLOW:
+		rss_type->tcp_ipv4 = rss_l4_en;
+		break;
+	case TCP_V6_FLOW:
+		rss_type->tcp_ipv6 = rss_l4_en;
+		break;
+	case UDP_V4_FLOW:
+		rss_type->udp_ipv4 = rss_l4_en;
+		break;
+	case UDP_V6_FLOW:
+		rss_type->udp_ipv6 = rss_l4_en;
+		break;
+	default:
+		return -EINVAL;
+	}
+
+	return 0;
+}
+
+static int hinic_set_rxfh_fields(struct net_device *dev,
+				 const struct ethtool_rxfh_fields *cmd,
+				 struct netlink_ext_ack *extack)
+{
+	struct hinic_dev *nic_dev = netdev_priv(dev);
+	struct hinic_rss_type *rss_type;
+	int err;
+
+	rss_type = &nic_dev->rss_type;
+
+	if (!(nic_dev->flags & HINIC_RSS_ENABLE))
+		return -EOPNOTSUPP;
+
+	/* RSS does not support anything other than hashing
+	 * to queues on src and dst IPs and ports
+	 */
+	if (cmd->data & ~(RXH_IP_SRC | RXH_IP_DST | RXH_L4_B_0_1 |
+		RXH_L4_B_2_3))
+		return -EINVAL;
+
+	/* We need at least the IP SRC and DEST fields for hashing */
+	if (!(cmd->data & RXH_IP_SRC) || !(cmd->data & RXH_IP_DST))
+		return -EINVAL;
+
+	err = hinic_get_rss_type(nic_dev,
+				 nic_dev->rss_tmpl_idx, rss_type);
+	if (err)
+		return -EFAULT;
+
+	switch (cmd->flow_type) {
+	case TCP_V4_FLOW:
+	case TCP_V6_FLOW:
+	case UDP_V4_FLOW:
+	case UDP_V6_FLOW:
+		err = set_l4_rss_hash_ops(cmd, rss_type);
+		if (err)
+			return err;
+		break;
+	case IPV4_FLOW:
+		rss_type->ipv4 = 1;
+		break;
+	case IPV6_FLOW:
+		rss_type->ipv6 = 1;
+		break;
+	default:
+		return -EINVAL;
+	}
+
+	err = hinic_set_rss_type(nic_dev, nic_dev->rss_tmpl_idx,
+				 *rss_type);
+	if (err)
+		return -EFAULT;
+
+	return 0;
+}
+
+static int __set_rss_rxfh(struct net_device *netdev,
+			  const u32 *indir, const u8 *key)
+{
+	struct hinic_dev *nic_dev = netdev_priv(netdev);
+	int err;
+
+	if (indir) {
+		if (!nic_dev->rss_indir_user) {
+			nic_dev->rss_indir_user =
+				kzalloc(sizeof(u32) * HINIC_RSS_INDIR_SIZE,
+					GFP_KERNEL);
+			if (!nic_dev->rss_indir_user)
+				return -ENOMEM;
+		}
+
+		memcpy(nic_dev->rss_indir_user, indir,
+		       sizeof(u32) * HINIC_RSS_INDIR_SIZE);
+
+		err = hinic_rss_set_indir_tbl(nic_dev,
+					      nic_dev->rss_tmpl_idx, indir);
+		if (err)
+			return -EFAULT;
+	}
+
+	if (key) {
+		if (!nic_dev->rss_hkey_user) {
+			nic_dev->rss_hkey_user =
+				kzalloc(HINIC_RSS_KEY_SIZE * 2, GFP_KERNEL);
+
+			if (!nic_dev->rss_hkey_user)
+				return -ENOMEM;
+		}
+
+		memcpy(nic_dev->rss_hkey_user, key, HINIC_RSS_KEY_SIZE);
+
+		err = hinic_rss_set_template_tbl(nic_dev,
+						 nic_dev->rss_tmpl_idx, key);
+		if (err)
+			return -EFAULT;
+	}
+
+	return 0;
+}
+
+static int hinic_get_rxnfc(struct net_device *netdev,
+			   struct ethtool_rxnfc *cmd, u32 *rule_locs)
+{
+	struct hinic_dev *nic_dev = netdev_priv(netdev);
+	int err = 0;
+
+	switch (cmd->cmd) {
+	case ETHTOOL_GRXRINGS:
+		cmd->data = nic_dev->num_qps;
+		break;
+	default:
+		err = -EOPNOTSUPP;
+		break;
+	}
+
+	return err;
+}
+
+static int hinic_get_rxfh(struct net_device *netdev,
+			  struct ethtool_rxfh_param *rxfh)
+{
+	struct hinic_dev *nic_dev = netdev_priv(netdev);
+	u8 hash_engine_type = 0;
+	int err = 0;
+
+	if (!(nic_dev->flags & HINIC_RSS_ENABLE))
+		return -EOPNOTSUPP;
+
+	err = hinic_rss_get_hash_engine(nic_dev,
+					nic_dev->rss_tmpl_idx,
+					&hash_engine_type);
+	if (err)
+		return -EFAULT;
+
+	rxfh->hfunc = hash_engine_type ? ETH_RSS_HASH_TOP : ETH_RSS_HASH_XOR;
+
+	if (rxfh->indir) {
+		err = hinic_rss_get_indir_tbl(nic_dev,
+					      nic_dev->rss_tmpl_idx,
+					      rxfh->indir);
+		if (err)
+			return -EFAULT;
+	}
+
+	if (rxfh->key)
+		err = hinic_rss_get_template_tbl(nic_dev,
+						 nic_dev->rss_tmpl_idx,
+						 rxfh->key);
+
+	return err;
+}
+
+static int hinic_set_rxfh(struct net_device *netdev,
+			  struct ethtool_rxfh_param *rxfh,
+			  struct netlink_ext_ack *extack)
+{
+	struct hinic_dev *nic_dev = netdev_priv(netdev);
+	int err = 0;
+
+	if (!(nic_dev->flags & HINIC_RSS_ENABLE))
+		return -EOPNOTSUPP;
+
+	if (rxfh->hfunc != ETH_RSS_HASH_NO_CHANGE) {
+		if (rxfh->hfunc != ETH_RSS_HASH_TOP &&
+		    rxfh->hfunc != ETH_RSS_HASH_XOR)
+			return -EOPNOTSUPP;
+
+		nic_dev->rss_hash_engine = (rxfh->hfunc == ETH_RSS_HASH_XOR) ?
+			HINIC_RSS_HASH_ENGINE_TYPE_XOR :
+			HINIC_RSS_HASH_ENGINE_TYPE_TOEP;
+		err = hinic_rss_set_hash_engine
+			(nic_dev, nic_dev->rss_tmpl_idx,
+			nic_dev->rss_hash_engine);
+		if (err)
+			return -EFAULT;
+	}
+
+	err = __set_rss_rxfh(netdev, rxfh->indir, rxfh->key);
+
+	return err;
+}
+
+static u32 hinic_get_rxfh_key_size(struct net_device *netdev)
+{
+	return HINIC_RSS_KEY_SIZE;
+}
+
+static u32 hinic_get_rxfh_indir_size(struct net_device *netdev)
+{
+	return HINIC_RSS_INDIR_SIZE;
+}
+
+#define HINIC_FUNC_STAT(_stat_item) {	\
+	.name = #_stat_item, \
+	.size = sizeof_field(struct hinic_vport_stats, _stat_item), \
+	.offset = offsetof(struct hinic_vport_stats, _stat_item) \
+}
+
+static struct hinic_stats hinic_function_stats[] = {
+	HINIC_FUNC_STAT(tx_unicast_pkts_vport),
+	HINIC_FUNC_STAT(tx_unicast_bytes_vport),
+	HINIC_FUNC_STAT(tx_multicast_pkts_vport),
+	HINIC_FUNC_STAT(tx_multicast_bytes_vport),
+	HINIC_FUNC_STAT(tx_broadcast_pkts_vport),
+	HINIC_FUNC_STAT(tx_broadcast_bytes_vport),
+
+	HINIC_FUNC_STAT(rx_unicast_pkts_vport),
+	HINIC_FUNC_STAT(rx_unicast_bytes_vport),
+	HINIC_FUNC_STAT(rx_multicast_pkts_vport),
+	HINIC_FUNC_STAT(rx_multicast_bytes_vport),
+	HINIC_FUNC_STAT(rx_broadcast_pkts_vport),
+	HINIC_FUNC_STAT(rx_broadcast_bytes_vport),
+
+	HINIC_FUNC_STAT(tx_discard_vport),
+	HINIC_FUNC_STAT(rx_discard_vport),
+	HINIC_FUNC_STAT(tx_err_vport),
+	HINIC_FUNC_STAT(rx_err_vport),
+};
+
+static char hinic_test_strings[][ETH_GSTRING_LEN] = {
+	"Internal lb test  (on/offline)",
+	"External lb test (external_lb)",
+};
+
+#define HINIC_PORT_STAT(_stat_item) { \
+	.name = #_stat_item, \
+	.size = sizeof_field(struct hinic_phy_port_stats, _stat_item), \
+	.offset = offsetof(struct hinic_phy_port_stats, _stat_item) \
+}
+
+static struct hinic_stats hinic_port_stats[] = {
+	HINIC_PORT_STAT(mac_rx_total_pkt_num),
+	HINIC_PORT_STAT(mac_rx_total_oct_num),
+	HINIC_PORT_STAT(mac_rx_bad_pkt_num),
+	HINIC_PORT_STAT(mac_rx_bad_oct_num),
+	HINIC_PORT_STAT(mac_rx_good_pkt_num),
+	HINIC_PORT_STAT(mac_rx_good_oct_num),
+	HINIC_PORT_STAT(mac_rx_uni_pkt_num),
+	HINIC_PORT_STAT(mac_rx_multi_pkt_num),
+	HINIC_PORT_STAT(mac_rx_broad_pkt_num),
+	HINIC_PORT_STAT(mac_tx_total_pkt_num),
+	HINIC_PORT_STAT(mac_tx_total_oct_num),
+	HINIC_PORT_STAT(mac_tx_bad_pkt_num),
+	HINIC_PORT_STAT(mac_tx_bad_oct_num),
+	HINIC_PORT_STAT(mac_tx_good_pkt_num),
+	HINIC_PORT_STAT(mac_tx_good_oct_num),
+	HINIC_PORT_STAT(mac_tx_uni_pkt_num),
+	HINIC_PORT_STAT(mac_tx_multi_pkt_num),
+	HINIC_PORT_STAT(mac_tx_broad_pkt_num),
+	HINIC_PORT_STAT(mac_rx_fragment_pkt_num),
+	HINIC_PORT_STAT(mac_rx_undersize_pkt_num),
+	HINIC_PORT_STAT(mac_rx_undermin_pkt_num),
+	HINIC_PORT_STAT(mac_rx_64_oct_pkt_num),
+	HINIC_PORT_STAT(mac_rx_65_127_oct_pkt_num),
+	HINIC_PORT_STAT(mac_rx_128_255_oct_pkt_num),
+	HINIC_PORT_STAT(mac_rx_256_511_oct_pkt_num),
+	HINIC_PORT_STAT(mac_rx_512_1023_oct_pkt_num),
+	HINIC_PORT_STAT(mac_rx_1024_1518_oct_pkt_num),
+	HINIC_PORT_STAT(mac_rx_1519_2047_oct_pkt_num),
+	HINIC_PORT_STAT(mac_rx_2048_4095_oct_pkt_num),
+	HINIC_PORT_STAT(mac_rx_4096_8191_oct_pkt_num),
+	HINIC_PORT_STAT(mac_rx_8192_9216_oct_pkt_num),
+	HINIC_PORT_STAT(mac_rx_9217_12287_oct_pkt_num),
+	HINIC_PORT_STAT(mac_rx_12288_16383_oct_pkt_num),
+	HINIC_PORT_STAT(mac_rx_1519_max_good_pkt_num),
+	HINIC_PORT_STAT(mac_rx_1519_max_bad_pkt_num),
+	HINIC_PORT_STAT(mac_rx_oversize_pkt_num),
+	HINIC_PORT_STAT(mac_rx_jabber_pkt_num),
+	HINIC_PORT_STAT(mac_rx_pause_num),
+	HINIC_PORT_STAT(mac_rx_pfc_pkt_num),
+	HINIC_PORT_STAT(mac_rx_pfc_pri0_pkt_num),
+	HINIC_PORT_STAT(mac_rx_pfc_pri1_pkt_num),
+	HINIC_PORT_STAT(mac_rx_pfc_pri2_pkt_num),
+	HINIC_PORT_STAT(mac_rx_pfc_pri3_pkt_num),
+	HINIC_PORT_STAT(mac_rx_pfc_pri4_pkt_num),
+	HINIC_PORT_STAT(mac_rx_pfc_pri5_pkt_num),
+	HINIC_PORT_STAT(mac_rx_pfc_pri6_pkt_num),
+	HINIC_PORT_STAT(mac_rx_pfc_pri7_pkt_num),
+	HINIC_PORT_STAT(mac_rx_control_pkt_num),
+	HINIC_PORT_STAT(mac_rx_sym_err_pkt_num),
+	HINIC_PORT_STAT(mac_rx_fcs_err_pkt_num),
+	HINIC_PORT_STAT(mac_rx_send_app_good_pkt_num),
+	HINIC_PORT_STAT(mac_rx_send_app_bad_pkt_num),
+	HINIC_PORT_STAT(mac_tx_fragment_pkt_num),
+	HINIC_PORT_STAT(mac_tx_undersize_pkt_num),
+	HINIC_PORT_STAT(mac_tx_undermin_pkt_num),
+	HINIC_PORT_STAT(mac_tx_64_oct_pkt_num),
+	HINIC_PORT_STAT(mac_tx_65_127_oct_pkt_num),
+	HINIC_PORT_STAT(mac_tx_128_255_oct_pkt_num),
+	HINIC_PORT_STAT(mac_tx_256_511_oct_pkt_num),
+	HINIC_PORT_STAT(mac_tx_512_1023_oct_pkt_num),
+	HINIC_PORT_STAT(mac_tx_1024_1518_oct_pkt_num),
+	HINIC_PORT_STAT(mac_tx_1519_2047_oct_pkt_num),
+	HINIC_PORT_STAT(mac_tx_2048_4095_oct_pkt_num),
+	HINIC_PORT_STAT(mac_tx_4096_8191_oct_pkt_num),
+	HINIC_PORT_STAT(mac_tx_8192_9216_oct_pkt_num),
+	HINIC_PORT_STAT(mac_tx_9217_12287_oct_pkt_num),
+	HINIC_PORT_STAT(mac_tx_12288_16383_oct_pkt_num),
+	HINIC_PORT_STAT(mac_tx_1519_max_good_pkt_num),
+	HINIC_PORT_STAT(mac_tx_1519_max_bad_pkt_num),
+	HINIC_PORT_STAT(mac_tx_oversize_pkt_num),
+	HINIC_PORT_STAT(mac_tx_jabber_pkt_num),
+	HINIC_PORT_STAT(mac_tx_pause_num),
+	HINIC_PORT_STAT(mac_tx_pfc_pkt_num),
+	HINIC_PORT_STAT(mac_tx_pfc_pri0_pkt_num),
+	HINIC_PORT_STAT(mac_tx_pfc_pri1_pkt_num),
+	HINIC_PORT_STAT(mac_tx_pfc_pri2_pkt_num),
+	HINIC_PORT_STAT(mac_tx_pfc_pri3_pkt_num),
+	HINIC_PORT_STAT(mac_tx_pfc_pri4_pkt_num),
+	HINIC_PORT_STAT(mac_tx_pfc_pri5_pkt_num),
+	HINIC_PORT_STAT(mac_tx_pfc_pri6_pkt_num),
+	HINIC_PORT_STAT(mac_tx_pfc_pri7_pkt_num),
+	HINIC_PORT_STAT(mac_tx_control_pkt_num),
+	HINIC_PORT_STAT(mac_tx_err_all_pkt_num),
+	HINIC_PORT_STAT(mac_tx_from_app_good_pkt_num),
+	HINIC_PORT_STAT(mac_tx_from_app_bad_pkt_num),
+};
+
+#define HINIC_TXQ_STAT(_stat_item) { \
+	.name = "txq%d_"#_stat_item, \
+	.size = sizeof_field(struct hinic_txq_stats, _stat_item), \
+	.offset = offsetof(struct hinic_txq_stats, _stat_item) \
+}
+
+static struct hinic_stats hinic_tx_queue_stats[] = {
+	HINIC_TXQ_STAT(pkts),
+	HINIC_TXQ_STAT(bytes),
+	HINIC_TXQ_STAT(tx_busy),
+	HINIC_TXQ_STAT(tx_wake),
+	HINIC_TXQ_STAT(tx_dropped),
+	HINIC_TXQ_STAT(big_frags_pkts),
+};
+
+#define HINIC_RXQ_STAT(_stat_item) { \
+	.name = "rxq%d_"#_stat_item, \
+	.size = sizeof_field(struct hinic_rxq_stats, _stat_item), \
+	.offset = offsetof(struct hinic_rxq_stats, _stat_item) \
+}
+
+static struct hinic_stats hinic_rx_queue_stats[] = {
+	HINIC_RXQ_STAT(pkts),
+	HINIC_RXQ_STAT(bytes),
+	HINIC_RXQ_STAT(errors),
+	HINIC_RXQ_STAT(csum_errors),
+	HINIC_RXQ_STAT(other_errors),
+};
+
+static void get_drv_queue_stats(struct hinic_dev *nic_dev, u64 *data)
+{
+	struct hinic_txq_stats txq_stats;
+	struct hinic_rxq_stats rxq_stats;
+	u16 i = 0, j = 0, qid = 0;
+	char *p;
+
+	for (qid = 0; qid < nic_dev->num_qps; qid++) {
+		if (!nic_dev->txqs)
+			break;
+
+		hinic_txq_get_stats(&nic_dev->txqs[qid], &txq_stats);
+		for (j = 0; j < ARRAY_SIZE(hinic_tx_queue_stats); j++, i++) {
+			p = (char *)&txq_stats +
+				hinic_tx_queue_stats[j].offset;
+			data[i] = (hinic_tx_queue_stats[j].size ==
+					sizeof(u64)) ? *(u64 *)p : *(u32 *)p;
+		}
+	}
+
+	for (qid = 0; qid < nic_dev->num_qps; qid++) {
+		if (!nic_dev->rxqs)
+			break;
+
+		hinic_rxq_get_stats(&nic_dev->rxqs[qid], &rxq_stats);
+		for (j = 0; j < ARRAY_SIZE(hinic_rx_queue_stats); j++, i++) {
+			p = (char *)&rxq_stats +
+				hinic_rx_queue_stats[j].offset;
+			data[i] = (hinic_rx_queue_stats[j].size ==
+					sizeof(u64)) ? *(u64 *)p : *(u32 *)p;
+		}
+	}
+}
+
+static void hinic_get_ethtool_stats(struct net_device *netdev,
+				    struct ethtool_stats *stats, u64 *data)
+{
+	struct hinic_dev *nic_dev = netdev_priv(netdev);
+	struct hinic_vport_stats vport_stats = {0};
+	struct hinic_phy_port_stats *port_stats;
+	u16 i = 0, j = 0;
+	char *p;
+	int err;
+
+	err = hinic_get_vport_stats(nic_dev, &vport_stats);
+	if (err)
+		netif_err(nic_dev, drv, netdev,
+			  "Failed to get vport stats from firmware\n");
+
+	for (j = 0; j < ARRAY_SIZE(hinic_function_stats); j++, i++) {
+		p = (char *)&vport_stats + hinic_function_stats[j].offset;
+		data[i] = (hinic_function_stats[j].size ==
+				sizeof(u64)) ? *(u64 *)p : *(u32 *)p;
+	}
+
+	port_stats = kzalloc(sizeof(*port_stats), GFP_KERNEL);
+	if (!port_stats) {
+		memset(&data[i], 0,
+		       ARRAY_SIZE(hinic_port_stats) * sizeof(*data));
+		i += ARRAY_SIZE(hinic_port_stats);
+		goto get_drv_stats;
+	}
+
+	err = hinic_get_phy_port_stats(nic_dev, port_stats);
+	if (err)
+		netif_err(nic_dev, drv, netdev,
+			  "Failed to get port stats from firmware\n");
+
+	for (j = 0; j < ARRAY_SIZE(hinic_port_stats); j++, i++) {
+		p = (char *)port_stats + hinic_port_stats[j].offset;
+		data[i] = (hinic_port_stats[j].size ==
+				sizeof(u64)) ? *(u64 *)p : *(u32 *)p;
+	}
+
+	kfree(port_stats);
+
+get_drv_stats:
+	get_drv_queue_stats(nic_dev, data + i);
+}
+
+static int hinic_get_sset_count(struct net_device *netdev, int sset)
+{
+	struct hinic_dev *nic_dev = netdev_priv(netdev);
+	int count, q_num;
+
+	switch (sset) {
+	case ETH_SS_TEST:
+		return ARRAY_SIZE(hinic_test_strings);
+	case ETH_SS_STATS:
+		q_num = nic_dev->num_qps;
+		count = ARRAY_SIZE(hinic_function_stats) +
+			(ARRAY_SIZE(hinic_tx_queue_stats) +
+			ARRAY_SIZE(hinic_rx_queue_stats)) * q_num;
+
+		count += ARRAY_SIZE(hinic_port_stats);
+
+		return count;
+	default:
+		return -EOPNOTSUPP;
+	}
+}
+
+static void hinic_get_strings(struct net_device *netdev,
+			      u32 stringset, u8 *data)
+{
+	struct hinic_dev *nic_dev = netdev_priv(netdev);
+	u16 i, j;
+
+	switch (stringset) {
+	case ETH_SS_TEST:
+		memcpy(data, *hinic_test_strings, sizeof(hinic_test_strings));
+		return;
+	case ETH_SS_STATS:
+		for (i = 0; i < ARRAY_SIZE(hinic_function_stats); i++)
+			ethtool_puts(&data, hinic_function_stats[i].name);
+
+		for (i = 0; i < ARRAY_SIZE(hinic_port_stats); i++)
+			ethtool_puts(&data, hinic_port_stats[i].name);
+
+		for (i = 0; i < nic_dev->num_qps; i++)
+			for (j = 0; j < ARRAY_SIZE(hinic_tx_queue_stats); j++)
+				ethtool_sprintf(&data, hinic_tx_queue_stats[j].name, i);
+
+		for (i = 0; i < nic_dev->num_qps; i++)
+			for (j = 0; j < ARRAY_SIZE(hinic_rx_queue_stats); j++)
+				ethtool_sprintf(&data, hinic_rx_queue_stats[j].name, i);
+
+		return;
+	default:
+		return;
+	}
+}
+
+static int hinic_run_lp_test(struct hinic_dev *nic_dev, u32 test_time)
+{
+	u8 *lb_test_rx_buf = nic_dev->lb_test_rx_buf;
+	struct net_device *netdev = nic_dev->netdev;
+	struct sk_buff *skb_tmp = NULL;
+	struct sk_buff *skb = NULL;
+	u32 cnt = test_time * 5;
+	u8 *test_data = NULL;
+	u32 i;
+	u8 j;
+
+	skb_tmp = alloc_skb(LP_PKT_LEN, GFP_ATOMIC);
+	if (!skb_tmp)
+		return -ENOMEM;
+
+	test_data = __skb_put(skb_tmp, LP_PKT_LEN);
+
+	memset(test_data, 0xFF, 2 * ETH_ALEN);
+	test_data[ETH_ALEN] = 0xFE;
+	test_data[2 * ETH_ALEN] = 0x08;
+	test_data[2 * ETH_ALEN + 1] = 0x0;
+
+	for (i = ETH_HLEN; i < LP_PKT_LEN; i++)
+		test_data[i] = i & 0xFF;
+
+	skb_tmp->queue_mapping = 0;
+	skb_tmp->ip_summed = CHECKSUM_COMPLETE;
+	skb_tmp->dev = netdev;
+
+	for (i = 0; i < cnt; i++) {
+		nic_dev->lb_test_rx_idx = 0;
+		memset(lb_test_rx_buf, 0, LP_PKT_CNT * LP_PKT_LEN);
+
+		for (j = 0; j < LP_PKT_CNT; j++) {
+			skb = pskb_copy(skb_tmp, GFP_ATOMIC);
+			if (!skb) {
+				dev_kfree_skb_any(skb_tmp);
+				netif_err(nic_dev, drv, netdev,
+					  "Copy skb failed for loopback test\n");
+				return -ENOMEM;
+			}
+
+			/* mark index for every pkt */
+			skb->data[LP_PKT_LEN - 1] = j;
+
+			if (hinic_lb_xmit_frame(skb, netdev)) {
+				dev_kfree_skb_any(skb);
+				dev_kfree_skb_any(skb_tmp);
+				netif_err(nic_dev, drv, netdev,
+					  "Xmit pkt failed for loopback test\n");
+				return -EBUSY;
+			}
+		}
+
+		/* wait till all pkts received to RX buffer */
+		msleep(200);
+
+		for (j = 0; j < LP_PKT_CNT; j++) {
+			if (memcmp(lb_test_rx_buf + j * LP_PKT_LEN,
+				   skb_tmp->data, LP_PKT_LEN - 1) ||
+			    (*(lb_test_rx_buf + j * LP_PKT_LEN +
+			     LP_PKT_LEN - 1) != j)) {
+				dev_kfree_skb_any(skb_tmp);
+				netif_err(nic_dev, drv, netdev,
+					  "Compare pkt failed in loopback test(index=0x%02x, data[%d]=0x%02x)\n",
+					  j + i * LP_PKT_CNT,
+					  LP_PKT_LEN - 1,
+					  *(lb_test_rx_buf + j * LP_PKT_LEN +
+					    LP_PKT_LEN - 1));
+				return -EIO;
+			}
+		}
+	}
+
+	dev_kfree_skb_any(skb_tmp);
+	return 0;
+}
+
+static int do_lp_test(struct hinic_dev *nic_dev, u32 flags, u32 test_time,
+		      enum diag_test_index *test_index)
+{
+	struct net_device *netdev = nic_dev->netdev;
+	u8 *lb_test_rx_buf = NULL;
+	int err = 0;
+
+	if (!(flags & ETH_TEST_FL_EXTERNAL_LB)) {
+		*test_index = INTERNAL_LP_TEST;
+		if (hinic_set_loopback_mode(nic_dev->hwdev,
+					    HINIC_INTERNAL_LP_MODE, true)) {
+			netif_err(nic_dev, drv, netdev,
+				  "Failed to set port loopback mode before loopback test\n");
+			return -EIO;
+		}
+	} else {
+		*test_index = EXTERNAL_LP_TEST;
+	}
+
+	lb_test_rx_buf = vmalloc(LP_PKT_CNT * LP_PKT_LEN);
+	if (!lb_test_rx_buf) {
+		err = -ENOMEM;
+	} else {
+		nic_dev->lb_test_rx_buf = lb_test_rx_buf;
+		nic_dev->lb_pkt_len = LP_PKT_LEN;
+		nic_dev->flags |= HINIC_LP_TEST;
+		err = hinic_run_lp_test(nic_dev, test_time);
+		nic_dev->flags &= ~HINIC_LP_TEST;
+		msleep(100);
+		vfree(lb_test_rx_buf);
+		nic_dev->lb_test_rx_buf = NULL;
+	}
+
+	if (!(flags & ETH_TEST_FL_EXTERNAL_LB)) {
+		if (hinic_set_loopback_mode(nic_dev->hwdev,
+					    HINIC_INTERNAL_LP_MODE, false)) {
+			netif_err(nic_dev, drv, netdev,
+				  "Failed to cancel port loopback mode after loopback test\n");
+			err = -EIO;
+		}
+	}
+
+	return err;
+}
+
+static void hinic_diag_test(struct net_device *netdev,
+			    struct ethtool_test *eth_test, u64 *data)
+{
+	struct hinic_dev *nic_dev = netdev_priv(netdev);
+	enum hinic_port_link_state link_state;
+	enum diag_test_index test_index = 0;
+	int err = 0;
+
+	memset(data, 0, DIAG_TEST_MAX * sizeof(u64));
+
+	/* don't support loopback test when netdev is closed. */
+	if (!(nic_dev->flags & HINIC_INTF_UP)) {
+		netif_err(nic_dev, drv, netdev,
+			  "Do not support loopback test when netdev is closed\n");
+		eth_test->flags |= ETH_TEST_FL_FAILED;
+		data[PORT_DOWN_ERR_IDX] = 1;
+		return;
+	}
+
+	netif_carrier_off(netdev);
+	netif_tx_disable(netdev);
+
+	err = do_lp_test(nic_dev, eth_test->flags, LP_DEFAULT_TIME,
+			 &test_index);
+	if (err) {
+		eth_test->flags |= ETH_TEST_FL_FAILED;
+		data[test_index] = 1;
+	}
+
+	netif_tx_wake_all_queues(netdev);
+
+	err = hinic_port_link_state(nic_dev, &link_state);
+	if (!err && link_state == HINIC_LINK_STATE_UP)
+		netif_carrier_on(netdev);
+}
+
+static int hinic_set_phys_id(struct net_device *netdev,
+			     enum ethtool_phys_id_state state)
+{
+	struct hinic_dev *nic_dev = netdev_priv(netdev);
+	int err = 0;
+	u8 port;
+
+	port = nic_dev->hwdev->port_id;
+
+	switch (state) {
+	case ETHTOOL_ID_ACTIVE:
+		err = hinic_set_led_status(nic_dev->hwdev, port,
+					   HINIC_LED_TYPE_LINK,
+					   HINIC_LED_MODE_FORCE_2HZ);
+		if (err)
+			netif_err(nic_dev, drv, netdev,
+				  "Set LED blinking in 2HZ failed\n");
+		break;
+
+	case ETHTOOL_ID_INACTIVE:
+		err = hinic_reset_led_status(nic_dev->hwdev, port);
+		if (err)
+			netif_err(nic_dev, drv, netdev,
+				  "Reset LED to original status failed\n");
+		break;
+
+	default:
+		return -EOPNOTSUPP;
+	}
+
+	return err;
+}
+
+static int hinic_get_module_info(struct net_device *netdev,
+				 struct ethtool_modinfo *modinfo)
+{
+	struct hinic_dev *nic_dev = netdev_priv(netdev);
+	u8 sfp_type_ext;
+	u8 sfp_type;
+	int err;
+
+	err = hinic_get_sfp_type(nic_dev->hwdev, &sfp_type, &sfp_type_ext);
+	if (err)
+		return err;
+
+	switch (sfp_type) {
+	case SFF8024_ID_SFP:
+		modinfo->type = ETH_MODULE_SFF_8472;
+		modinfo->eeprom_len = ETH_MODULE_SFF_8472_LEN;
+		break;
+	case SFF8024_ID_QSFP_8438:
+		modinfo->type = ETH_MODULE_SFF_8436;
+		modinfo->eeprom_len = ETH_MODULE_SFF_8436_MAX_LEN;
+		break;
+	case SFF8024_ID_QSFP_8436_8636:
+		if (sfp_type_ext >= 0x3) {
+			modinfo->type = ETH_MODULE_SFF_8636;
+			modinfo->eeprom_len = ETH_MODULE_SFF_8636_MAX_LEN;
+
+		} else {
+			modinfo->type = ETH_MODULE_SFF_8436;
+			modinfo->eeprom_len = ETH_MODULE_SFF_8436_MAX_LEN;
+		}
+		break;
+	case SFF8024_ID_QSFP28_8636:
+		modinfo->type = ETH_MODULE_SFF_8636;
+		modinfo->eeprom_len = ETH_MODULE_SFF_8636_MAX_LEN;
+		break;
+	default:
+		netif_warn(nic_dev, drv, netdev,
+			   "Optical module unknown: 0x%x\n", sfp_type);
+		return -EINVAL;
+	}
+
+	return 0;
+}
+
+static int hinic_get_module_eeprom(struct net_device *netdev,
+				   struct ethtool_eeprom *ee, u8 *data)
+{
+	struct hinic_dev *nic_dev = netdev_priv(netdev);
+	u8 sfp_data[STD_SFP_INFO_MAX_SIZE];
+	u16 len;
+	int err;
+
+	if (!ee->len || ((ee->len + ee->offset) > STD_SFP_INFO_MAX_SIZE))
+		return -EINVAL;
+
+	memset(data, 0, ee->len);
+
+	err = hinic_get_sfp_eeprom(nic_dev->hwdev, sfp_data, &len);
+	if (err)
+		return err;
+
+	memcpy(data, sfp_data + ee->offset, ee->len);
+
+	return 0;
+}
+
+static int
+hinic_get_link_ext_state(struct net_device *netdev,
+			 struct ethtool_link_ext_state_info *link_ext_state_info)
+{
+	struct hinic_dev *nic_dev = netdev_priv(netdev);
+
+	if (netif_carrier_ok(netdev))
+		return -ENODATA;
+
+	if (nic_dev->cable_unplugged)
+		link_ext_state_info->link_ext_state =
+			ETHTOOL_LINK_EXT_STATE_NO_CABLE;
+	else if (nic_dev->module_unrecognized)
+		link_ext_state_info->link_ext_state =
+			ETHTOOL_LINK_EXT_STATE_LINK_LOGICAL_MISMATCH;
+
+	return 0;
+}
+
+static const struct ethtool_ops hinic_ethtool_ops = {
+	.supported_coalesce_params = ETHTOOL_COALESCE_RX_USECS |
+				     ETHTOOL_COALESCE_RX_MAX_FRAMES |
+				     ETHTOOL_COALESCE_TX_USECS |
+				     ETHTOOL_COALESCE_TX_MAX_FRAMES,
+
+	.get_link_ksettings = hinic_get_link_ksettings,
+	.set_link_ksettings = hinic_set_link_ksettings,
+	.get_drvinfo = hinic_get_drvinfo,
+	.get_link = ethtool_op_get_link,
+	.get_link_ext_state = hinic_get_link_ext_state,
+	.get_ringparam = hinic_get_ringparam,
+	.set_ringparam = hinic_set_ringparam,
+	.get_coalesce = hinic_get_coalesce,
+	.set_coalesce = hinic_set_coalesce,
+	.get_per_queue_coalesce = hinic_get_per_queue_coalesce,
+	.set_per_queue_coalesce = hinic_set_per_queue_coalesce,
+	.get_pauseparam = hinic_get_pauseparam,
+	.set_pauseparam = hinic_set_pauseparam,
+	.get_channels = hinic_get_channels,
+	.set_channels = hinic_set_channels,
+	.get_rxnfc = hinic_get_rxnfc,
+	.get_rxfh_key_size = hinic_get_rxfh_key_size,
+	.get_rxfh_indir_size = hinic_get_rxfh_indir_size,
+	.get_rxfh = hinic_get_rxfh,
+	.set_rxfh = hinic_set_rxfh,
+	.get_rxfh_fields = hinic_get_rxfh_fields,
+	.set_rxfh_fields = hinic_set_rxfh_fields,
+	.get_sset_count = hinic_get_sset_count,
+	.get_ethtool_stats = hinic_get_ethtool_stats,
+	.get_strings = hinic_get_strings,
+	.self_test = hinic_diag_test,
+	.set_phys_id = hinic_set_phys_id,
+	.get_module_info = hinic_get_module_info,
+	.get_module_eeprom = hinic_get_module_eeprom,
+};
+
+static const struct ethtool_ops hinicvf_ethtool_ops = {
+	.supported_coalesce_params = ETHTOOL_COALESCE_RX_USECS |
+				     ETHTOOL_COALESCE_RX_MAX_FRAMES |
+				     ETHTOOL_COALESCE_TX_USECS |
+				     ETHTOOL_COALESCE_TX_MAX_FRAMES,
+
+	.get_link_ksettings = hinic_get_link_ksettings,
+	.get_drvinfo = hinic_get_drvinfo,
+	.get_link = ethtool_op_get_link,
+	.get_ringparam = hinic_get_ringparam,
+	.set_ringparam = hinic_set_ringparam,
+	.get_coalesce = hinic_get_coalesce,
+	.set_coalesce = hinic_set_coalesce,
+	.get_per_queue_coalesce = hinic_get_per_queue_coalesce,
+	.set_per_queue_coalesce = hinic_set_per_queue_coalesce,
+	.get_channels = hinic_get_channels,
+	.set_channels = hinic_set_channels,
+	.get_rxnfc = hinic_get_rxnfc,
+	.get_rxfh_key_size = hinic_get_rxfh_key_size,
+	.get_rxfh_indir_size = hinic_get_rxfh_indir_size,
+	.get_rxfh = hinic_get_rxfh,
+	.set_rxfh = hinic_set_rxfh,
+	.get_rxfh_fields = hinic_get_rxfh_fields,
+	.set_rxfh_fields = hinic_set_rxfh_fields,
+	.get_sset_count = hinic_get_sset_count,
+	.get_ethtool_stats = hinic_get_ethtool_stats,
+	.get_strings = hinic_get_strings,
+};
+
+void hinic_set_ethtool_ops(struct net_device *netdev)
+{
+	struct hinic_dev *nic_dev = netdev_priv(netdev);
+
+	if (!HINIC_IS_VF(nic_dev->hwdev->hwif))
+		netdev->ethtool_ops = &hinic_ethtool_ops;
+	else
+		netdev->ethtool_ops = &hinicvf_ethtool_ops;
+}
diff --git a/drivers/net/ethernet/huawei/hinic/hinic_hw_api_cmd.c b/drivers/net/ethernet/huawei/hinic/hinic_hw_api_cmd.c
new file mode 100644
index 000000000000..998717f02136
--- /dev/null
+++ b/drivers/net/ethernet/huawei/hinic/hinic_hw_api_cmd.c
@@ -0,0 +1,987 @@
+// SPDX-License-Identifier: GPL-2.0-only
+/*
+ * Huawei HiNIC PCI Express Linux driver
+ * Copyright(c) 2017 Huawei Technologies Co., Ltd
+ */
+
+#include <linux/kernel.h>
+#include <linux/types.h>
+#include <linux/errno.h>
+#include <linux/pci.h>
+#include <linux/device.h>
+#include <linux/slab.h>
+#include <linux/dma-mapping.h>
+#include <linux/bitops.h>
+#include <linux/err.h>
+#include <linux/jiffies.h>
+#include <linux/delay.h>
+#include <linux/log2.h>
+#include <linux/semaphore.h>
+#include <asm/byteorder.h>
+#include <asm/barrier.h>
+
+#include "hinic_hw_csr.h"
+#include "hinic_hw_if.h"
+#include "hinic_hw_api_cmd.h"
+
+#define API_CHAIN_NUM_CELLS                     32
+
+#define API_CMD_CELL_SIZE_SHIFT                 6
+#define API_CMD_CELL_SIZE_MIN                   (BIT(API_CMD_CELL_SIZE_SHIFT))
+
+#define API_CMD_CELL_SIZE(cell_size)            \
+		(((cell_size) >= API_CMD_CELL_SIZE_MIN) ? \
+		 (1 << (fls(cell_size - 1))) : API_CMD_CELL_SIZE_MIN)
+
+#define API_CMD_CELL_SIZE_VAL(size)             \
+		ilog2((size) >> API_CMD_CELL_SIZE_SHIFT)
+
+#define API_CMD_BUF_SIZE                        2048
+
+/* Sizes of the members in hinic_api_cmd_cell */
+#define API_CMD_CELL_DESC_SIZE          8
+#define API_CMD_CELL_DATA_ADDR_SIZE     8
+
+#define API_CMD_CELL_ALIGNMENT          8
+
+#define API_CMD_TIMEOUT                 1000
+
+#define MASKED_IDX(chain, idx)          ((idx) & ((chain)->num_cells - 1))
+
+#define SIZE_8BYTES(size)               (ALIGN((size), 8) >> 3)
+#define SIZE_4BYTES(size)               (ALIGN((size), 4) >> 2)
+
+#define RD_DMA_ATTR_DEFAULT             0
+#define WR_DMA_ATTR_DEFAULT             0
+
+enum api_cmd_data_format {
+	SGE_DATA = 1,           /* cell data is passed by hw address */
+};
+
+enum api_cmd_type {
+	API_CMD_WRITE = 0,
+};
+
+enum api_cmd_bypass {
+	NO_BYPASS       = 0,
+	BYPASS          = 1,
+};
+
+enum api_cmd_xor_chk_level {
+	XOR_CHK_DIS = 0,
+
+	XOR_CHK_ALL = 3,
+};
+
+static u8 xor_chksum_set(void *data)
+{
+	int idx;
+	u8 *val, checksum = 0;
+
+	val = data;
+
+	for (idx = 0; idx < 7; idx++)
+		checksum ^= val[idx];
+
+	return checksum;
+}
+
+static void set_prod_idx(struct hinic_api_cmd_chain *chain)
+{
+	enum hinic_api_cmd_chain_type chain_type = chain->chain_type;
+	struct hinic_hwif *hwif = chain->hwif;
+	u32 addr, prod_idx;
+
+	addr = HINIC_CSR_API_CMD_CHAIN_PI_ADDR(chain_type);
+	prod_idx = hinic_hwif_read_reg(hwif, addr);
+
+	prod_idx = HINIC_API_CMD_PI_CLEAR(prod_idx, IDX);
+
+	prod_idx |= HINIC_API_CMD_PI_SET(chain->prod_idx, IDX);
+
+	hinic_hwif_write_reg(hwif, addr, prod_idx);
+}
+
+static u32 get_hw_cons_idx(struct hinic_api_cmd_chain *chain)
+{
+	u32 addr, val;
+
+	addr = HINIC_CSR_API_CMD_STATUS_ADDR(chain->chain_type);
+	val  = hinic_hwif_read_reg(chain->hwif, addr);
+
+	return HINIC_API_CMD_STATUS_GET(val, CONS_IDX);
+}
+
+static void dump_api_chain_reg(struct hinic_api_cmd_chain *chain)
+{
+	u32 addr, val;
+
+	addr = HINIC_CSR_API_CMD_STATUS_ADDR(chain->chain_type);
+	val  = hinic_hwif_read_reg(chain->hwif, addr);
+
+	dev_err(&chain->hwif->pdev->dev, "Chain type: 0x%x, cpld error: 0x%x, check error: 0x%x, current fsm: 0x%x\n",
+		chain->chain_type, HINIC_API_CMD_STATUS_GET(val, CPLD_ERR),
+		HINIC_API_CMD_STATUS_GET(val, CHKSUM_ERR),
+		HINIC_API_CMD_STATUS_GET(val, FSM));
+
+	dev_err(&chain->hwif->pdev->dev, "Chain hw current ci: 0x%x\n",
+		HINIC_API_CMD_STATUS_GET(val, CONS_IDX));
+
+	addr = HINIC_CSR_API_CMD_CHAIN_PI_ADDR(chain->chain_type);
+	val  = hinic_hwif_read_reg(chain->hwif, addr);
+	dev_err(&chain->hwif->pdev->dev, "Chain hw current pi: 0x%x\n", val);
+}
+
+/**
+ * chain_busy - check if the chain is still processing last requests
+ * @chain: chain to check
+ *
+ * Return 0 - Success, negative - Failure
+ **/
+static int chain_busy(struct hinic_api_cmd_chain *chain)
+{
+	struct hinic_hwif *hwif = chain->hwif;
+	struct pci_dev *pdev = hwif->pdev;
+	u32 prod_idx;
+
+	switch (chain->chain_type) {
+	case HINIC_API_CMD_WRITE_TO_MGMT_CPU:
+		chain->cons_idx = get_hw_cons_idx(chain);
+		prod_idx = chain->prod_idx;
+
+		/* check for a space for a new command */
+		if (chain->cons_idx == MASKED_IDX(chain, prod_idx + 1)) {
+			dev_err(&pdev->dev, "API CMD chain %d is busy, cons_idx: %d, prod_idx: %d\n",
+				chain->chain_type, chain->cons_idx,
+				chain->prod_idx);
+			dump_api_chain_reg(chain);
+			return -EBUSY;
+		}
+		break;
+
+	default:
+		dev_err(&pdev->dev, "Unknown API CMD Chain type\n");
+		break;
+	}
+
+	return 0;
+}
+
+/**
+ * get_cell_data_size - get the data size of a specific cell type
+ * @type: chain type
+ *
+ * Return the data(Desc + Address) size in the cell
+ **/
+static u8 get_cell_data_size(enum hinic_api_cmd_chain_type type)
+{
+	u8 cell_data_size = 0;
+
+	switch (type) {
+	case HINIC_API_CMD_WRITE_TO_MGMT_CPU:
+		cell_data_size = ALIGN(API_CMD_CELL_DESC_SIZE +
+				       API_CMD_CELL_DATA_ADDR_SIZE,
+				       API_CMD_CELL_ALIGNMENT);
+		break;
+	default:
+		break;
+	}
+
+	return cell_data_size;
+}
+
+/**
+ * prepare_cell_ctrl - prepare the ctrl of the cell for the command
+ * @cell_ctrl: the control of the cell to set the control value into it
+ * @data_size: the size of the data in the cell
+ **/
+static void prepare_cell_ctrl(u64 *cell_ctrl, u16 data_size)
+{
+	u8 chksum;
+	u64 ctrl;
+
+	ctrl =  HINIC_API_CMD_CELL_CTRL_SET(SIZE_8BYTES(data_size), DATA_SZ)  |
+		HINIC_API_CMD_CELL_CTRL_SET(RD_DMA_ATTR_DEFAULT, RD_DMA_ATTR) |
+		HINIC_API_CMD_CELL_CTRL_SET(WR_DMA_ATTR_DEFAULT, WR_DMA_ATTR);
+
+	chksum = xor_chksum_set(&ctrl);
+
+	ctrl |= HINIC_API_CMD_CELL_CTRL_SET(chksum, XOR_CHKSUM);
+
+	/* The data in the HW should be in Big Endian Format */
+	*cell_ctrl = cpu_to_be64(ctrl);
+}
+
+/**
+ * prepare_api_cmd - prepare API CMD command
+ * @chain: chain for the command
+ * @dest: destination node on the card that will receive the command
+ * @cmd: command data
+ * @cmd_size: the command size
+ **/
+static void prepare_api_cmd(struct hinic_api_cmd_chain *chain,
+			    enum hinic_node_id dest,
+			    void *cmd, u16 cmd_size)
+{
+	struct hinic_api_cmd_cell *cell = chain->curr_node;
+	struct hinic_api_cmd_cell_ctxt *cell_ctxt;
+	struct hinic_hwif *hwif = chain->hwif;
+	struct pci_dev *pdev = hwif->pdev;
+
+	cell_ctxt = &chain->cell_ctxt[chain->prod_idx];
+
+	switch (chain->chain_type) {
+	case HINIC_API_CMD_WRITE_TO_MGMT_CPU:
+		cell->desc = HINIC_API_CMD_DESC_SET(SGE_DATA, API_TYPE)   |
+			     HINIC_API_CMD_DESC_SET(API_CMD_WRITE, RD_WR) |
+			     HINIC_API_CMD_DESC_SET(NO_BYPASS, MGMT_BYPASS);
+		break;
+
+	default:
+		dev_err(&pdev->dev, "unknown Chain type\n");
+		return;
+	}
+
+	cell->desc |= HINIC_API_CMD_DESC_SET(dest, DEST)        |
+		      HINIC_API_CMD_DESC_SET(SIZE_4BYTES(cmd_size), SIZE);
+
+	cell->desc |= HINIC_API_CMD_DESC_SET(xor_chksum_set(&cell->desc),
+					     XOR_CHKSUM);
+
+	/* The data in the HW should be in Big Endian Format */
+	cell->desc = cpu_to_be64(cell->desc);
+
+	memcpy(cell_ctxt->api_cmd_vaddr, cmd, cmd_size);
+}
+
+/**
+ * prepare_cell - prepare cell ctrl and cmd in the current cell
+ * @chain: chain for the command
+ * @dest: destination node on the card that will receive the command
+ * @cmd: command data
+ * @cmd_size: the command size
+ *
+ * Return 0 - Success, negative - Failure
+ **/
+static void prepare_cell(struct hinic_api_cmd_chain *chain,
+			 enum  hinic_node_id dest,
+			 void *cmd, u16 cmd_size)
+{
+	struct hinic_api_cmd_cell *curr_node = chain->curr_node;
+	u16 data_size = get_cell_data_size(chain->chain_type);
+
+	prepare_cell_ctrl(&curr_node->ctrl, data_size);
+	prepare_api_cmd(chain, dest, cmd, cmd_size);
+}
+
+static inline void cmd_chain_prod_idx_inc(struct hinic_api_cmd_chain *chain)
+{
+	chain->prod_idx = MASKED_IDX(chain, chain->prod_idx + 1);
+}
+
+/**
+ * api_cmd_status_update - update the status in the chain struct
+ * @chain: chain to update
+ **/
+static void api_cmd_status_update(struct hinic_api_cmd_chain *chain)
+{
+	enum hinic_api_cmd_chain_type chain_type;
+	struct hinic_api_cmd_status *wb_status;
+	struct hinic_hwif *hwif = chain->hwif;
+	struct pci_dev *pdev = hwif->pdev;
+	u64 status_header;
+	u32 status;
+
+	wb_status = chain->wb_status;
+	status_header = be64_to_cpu(wb_status->header);
+
+	status = be32_to_cpu(wb_status->status);
+	if (HINIC_API_CMD_STATUS_GET(status, CHKSUM_ERR)) {
+		dev_err(&pdev->dev, "API CMD status: Xor check error\n");
+		return;
+	}
+
+	chain_type = HINIC_API_CMD_STATUS_HEADER_GET(status_header, CHAIN_ID);
+	if (chain_type >= HINIC_API_CMD_MAX) {
+		dev_err(&pdev->dev, "unknown API CMD Chain %d\n", chain_type);
+		return;
+	}
+
+	chain->cons_idx = HINIC_API_CMD_STATUS_GET(status, CONS_IDX);
+}
+
+/**
+ * wait_for_status_poll - wait for write to api cmd command to complete
+ * @chain: the chain of the command
+ *
+ * Return 0 - Success, negative - Failure
+ **/
+static int wait_for_status_poll(struct hinic_api_cmd_chain *chain)
+{
+	int err = -ETIMEDOUT;
+	unsigned long end;
+
+	end = jiffies + msecs_to_jiffies(API_CMD_TIMEOUT);
+	do {
+		api_cmd_status_update(chain);
+
+		/* wait for CI to be updated - sign for completion */
+		if (chain->cons_idx == chain->prod_idx) {
+			err = 0;
+			break;
+		}
+
+		msleep(20);
+	} while (time_before(jiffies, end));
+
+	return err;
+}
+
+/**
+ * wait_for_api_cmd_completion - wait for command to complete
+ * @chain: chain for the command
+ *
+ * Return 0 - Success, negative - Failure
+ **/
+static int wait_for_api_cmd_completion(struct hinic_api_cmd_chain *chain)
+{
+	struct hinic_hwif *hwif = chain->hwif;
+	struct pci_dev *pdev = hwif->pdev;
+	int err;
+
+	switch (chain->chain_type) {
+	case HINIC_API_CMD_WRITE_TO_MGMT_CPU:
+		err = wait_for_status_poll(chain);
+		if (err) {
+			dev_err(&pdev->dev, "API CMD Poll status timeout\n");
+			dump_api_chain_reg(chain);
+			break;
+		}
+		break;
+
+	default:
+		dev_err(&pdev->dev, "unknown API CMD Chain type\n");
+		err = -EINVAL;
+		break;
+	}
+
+	return err;
+}
+
+/**
+ * api_cmd - API CMD command
+ * @chain: chain for the command
+ * @dest: destination node on the card that will receive the command
+ * @cmd: command data
+ * @cmd_size: the command size
+ *
+ * Return 0 - Success, negative - Failure
+ **/
+static int api_cmd(struct hinic_api_cmd_chain *chain,
+		   enum hinic_node_id dest, u8 *cmd, u16 cmd_size)
+{
+	struct hinic_api_cmd_cell_ctxt *ctxt;
+	int err;
+
+	down(&chain->sem);
+	if (chain_busy(chain)) {
+		up(&chain->sem);
+		return -EBUSY;
+	}
+
+	prepare_cell(chain, dest, cmd, cmd_size);
+	cmd_chain_prod_idx_inc(chain);
+
+	wmb();  /* inc pi before issue the command */
+
+	set_prod_idx(chain);    /* issue the command */
+
+	ctxt = &chain->cell_ctxt[chain->prod_idx];
+
+	chain->curr_node = ctxt->cell_vaddr;
+
+	err = wait_for_api_cmd_completion(chain);
+
+	up(&chain->sem);
+	return err;
+}
+
+/**
+ * hinic_api_cmd_write - Write API CMD command
+ * @chain: chain for write command
+ * @dest: destination node on the card that will receive the command
+ * @cmd: command data
+ * @size: the command size
+ *
+ * Return 0 - Success, negative - Failure
+ **/
+int hinic_api_cmd_write(struct hinic_api_cmd_chain *chain,
+			enum hinic_node_id dest, u8 *cmd, u16 size)
+{
+	/* Verify the chain type */
+	if (chain->chain_type == HINIC_API_CMD_WRITE_TO_MGMT_CPU)
+		return api_cmd(chain, dest, cmd, size);
+
+	return -EINVAL;
+}
+
+/**
+ * api_cmd_hw_restart - restart the chain in the HW
+ * @chain: the API CMD specific chain to restart
+ *
+ * Return 0 - Success, negative - Failure
+ **/
+static int api_cmd_hw_restart(struct hinic_api_cmd_chain *chain)
+{
+	struct hinic_hwif *hwif = chain->hwif;
+	int err = -ETIMEDOUT;
+	unsigned long end;
+	u32 reg_addr, val;
+
+	/* Read Modify Write */
+	reg_addr = HINIC_CSR_API_CMD_CHAIN_REQ_ADDR(chain->chain_type);
+	val = hinic_hwif_read_reg(hwif, reg_addr);
+
+	val = HINIC_API_CMD_CHAIN_REQ_CLEAR(val, RESTART);
+	val |= HINIC_API_CMD_CHAIN_REQ_SET(1, RESTART);
+
+	hinic_hwif_write_reg(hwif, reg_addr, val);
+
+	end = jiffies + msecs_to_jiffies(API_CMD_TIMEOUT);
+	do {
+		val = hinic_hwif_read_reg(hwif, reg_addr);
+
+		if (!HINIC_API_CMD_CHAIN_REQ_GET(val, RESTART)) {
+			err = 0;
+			break;
+		}
+
+		msleep(20);
+	} while (time_before(jiffies, end));
+
+	return err;
+}
+
+/**
+ * api_cmd_ctrl_init - set the control register of a chain
+ * @chain: the API CMD specific chain to set control register for
+ **/
+static void api_cmd_ctrl_init(struct hinic_api_cmd_chain *chain)
+{
+	struct hinic_hwif *hwif = chain->hwif;
+	u32 addr, ctrl;
+	u16 cell_size;
+
+	/* Read Modify Write */
+	addr = HINIC_CSR_API_CMD_CHAIN_CTRL_ADDR(chain->chain_type);
+
+	cell_size = API_CMD_CELL_SIZE_VAL(chain->cell_size);
+
+	ctrl = hinic_hwif_read_reg(hwif, addr);
+
+	ctrl =  HINIC_API_CMD_CHAIN_CTRL_CLEAR(ctrl, RESTART_WB_STAT) &
+		HINIC_API_CMD_CHAIN_CTRL_CLEAR(ctrl, XOR_ERR)         &
+		HINIC_API_CMD_CHAIN_CTRL_CLEAR(ctrl, AEQE_EN)         &
+		HINIC_API_CMD_CHAIN_CTRL_CLEAR(ctrl, XOR_CHK_EN)      &
+		HINIC_API_CMD_CHAIN_CTRL_CLEAR(ctrl, CELL_SIZE);
+
+	ctrl |= HINIC_API_CMD_CHAIN_CTRL_SET(1, XOR_ERR)              |
+		HINIC_API_CMD_CHAIN_CTRL_SET(XOR_CHK_ALL, XOR_CHK_EN) |
+		HINIC_API_CMD_CHAIN_CTRL_SET(cell_size, CELL_SIZE);
+
+	hinic_hwif_write_reg(hwif, addr, ctrl);
+}
+
+/**
+ * api_cmd_set_status_addr - set the status address of a chain in the HW
+ * @chain: the API CMD specific chain to set in HW status address for
+ **/
+static void api_cmd_set_status_addr(struct hinic_api_cmd_chain *chain)
+{
+	struct hinic_hwif *hwif = chain->hwif;
+	u32 addr, val;
+
+	addr = HINIC_CSR_API_CMD_STATUS_HI_ADDR(chain->chain_type);
+	val = upper_32_bits(chain->wb_status_paddr);
+	hinic_hwif_write_reg(hwif, addr, val);
+
+	addr = HINIC_CSR_API_CMD_STATUS_LO_ADDR(chain->chain_type);
+	val = lower_32_bits(chain->wb_status_paddr);
+	hinic_hwif_write_reg(hwif, addr, val);
+}
+
+/**
+ * api_cmd_set_num_cells - set the number cells of a chain in the HW
+ * @chain: the API CMD specific chain to set in HW the number of cells for
+ **/
+static void api_cmd_set_num_cells(struct hinic_api_cmd_chain *chain)
+{
+	struct hinic_hwif *hwif = chain->hwif;
+	u32 addr, val;
+
+	addr = HINIC_CSR_API_CMD_CHAIN_NUM_CELLS_ADDR(chain->chain_type);
+	val  = chain->num_cells;
+	hinic_hwif_write_reg(hwif, addr, val);
+}
+
+/**
+ * api_cmd_head_init - set the head of a chain in the HW
+ * @chain: the API CMD specific chain to set in HW the head for
+ **/
+static void api_cmd_head_init(struct hinic_api_cmd_chain *chain)
+{
+	struct hinic_hwif *hwif = chain->hwif;
+	u32 addr, val;
+
+	addr = HINIC_CSR_API_CMD_CHAIN_HEAD_HI_ADDR(chain->chain_type);
+	val = upper_32_bits(chain->head_cell_paddr);
+	hinic_hwif_write_reg(hwif, addr, val);
+
+	addr = HINIC_CSR_API_CMD_CHAIN_HEAD_LO_ADDR(chain->chain_type);
+	val = lower_32_bits(chain->head_cell_paddr);
+	hinic_hwif_write_reg(hwif, addr, val);
+}
+
+/**
+ * api_cmd_chain_hw_clean - clean the HW
+ * @chain: the API CMD specific chain
+ **/
+static void api_cmd_chain_hw_clean(struct hinic_api_cmd_chain *chain)
+{
+	struct hinic_hwif *hwif = chain->hwif;
+	u32 addr, ctrl;
+
+	addr = HINIC_CSR_API_CMD_CHAIN_CTRL_ADDR(chain->chain_type);
+
+	ctrl = hinic_hwif_read_reg(hwif, addr);
+	ctrl = HINIC_API_CMD_CHAIN_CTRL_CLEAR(ctrl, RESTART_WB_STAT) &
+	       HINIC_API_CMD_CHAIN_CTRL_CLEAR(ctrl, XOR_ERR)         &
+	       HINIC_API_CMD_CHAIN_CTRL_CLEAR(ctrl, AEQE_EN)         &
+	       HINIC_API_CMD_CHAIN_CTRL_CLEAR(ctrl, XOR_CHK_EN)      &
+	       HINIC_API_CMD_CHAIN_CTRL_CLEAR(ctrl, CELL_SIZE);
+
+	hinic_hwif_write_reg(hwif, addr, ctrl);
+}
+
+/**
+ * api_cmd_chain_hw_init - initialize the chain in the HW
+ * @chain: the API CMD specific chain to initialize in HW
+ *
+ * Return 0 - Success, negative - Failure
+ **/
+static int api_cmd_chain_hw_init(struct hinic_api_cmd_chain *chain)
+{
+	struct hinic_hwif *hwif = chain->hwif;
+	struct pci_dev *pdev = hwif->pdev;
+	int err;
+
+	api_cmd_chain_hw_clean(chain);
+
+	api_cmd_set_status_addr(chain);
+
+	err = api_cmd_hw_restart(chain);
+	if (err) {
+		dev_err(&pdev->dev, "Failed to restart API CMD HW\n");
+		return err;
+	}
+
+	api_cmd_ctrl_init(chain);
+	api_cmd_set_num_cells(chain);
+	api_cmd_head_init(chain);
+	return 0;
+}
+
+/**
+ * free_cmd_buf - free the dma buffer of API CMD command
+ * @chain: the API CMD specific chain of the cmd
+ * @cell_idx: the cell index of the cmd
+ **/
+static void free_cmd_buf(struct hinic_api_cmd_chain *chain, int cell_idx)
+{
+	struct hinic_api_cmd_cell_ctxt *cell_ctxt;
+	struct hinic_hwif *hwif = chain->hwif;
+	struct pci_dev *pdev = hwif->pdev;
+
+	cell_ctxt = &chain->cell_ctxt[cell_idx];
+
+	dma_free_coherent(&pdev->dev, API_CMD_BUF_SIZE,
+			  cell_ctxt->api_cmd_vaddr,
+			  cell_ctxt->api_cmd_paddr);
+}
+
+/**
+ * alloc_cmd_buf - allocate a dma buffer for API CMD command
+ * @chain: the API CMD specific chain for the cmd
+ * @cell: the cell in the HW for the cmd
+ * @cell_idx: the index of the cell
+ *
+ * Return 0 - Success, negative - Failure
+ **/
+static int alloc_cmd_buf(struct hinic_api_cmd_chain *chain,
+			 struct hinic_api_cmd_cell *cell, int cell_idx)
+{
+	struct hinic_api_cmd_cell_ctxt *cell_ctxt;
+	struct hinic_hwif *hwif = chain->hwif;
+	struct pci_dev *pdev = hwif->pdev;
+	dma_addr_t cmd_paddr;
+	u8 *cmd_vaddr;
+	int err = 0;
+
+	cmd_vaddr = dma_alloc_coherent(&pdev->dev, API_CMD_BUF_SIZE,
+				       &cmd_paddr, GFP_KERNEL);
+	if (!cmd_vaddr)
+		return -ENOMEM;
+
+	cell_ctxt = &chain->cell_ctxt[cell_idx];
+
+	cell_ctxt->api_cmd_vaddr = cmd_vaddr;
+	cell_ctxt->api_cmd_paddr = cmd_paddr;
+
+	/* set the cmd DMA address in the cell */
+	switch (chain->chain_type) {
+	case HINIC_API_CMD_WRITE_TO_MGMT_CPU:
+		/* The data in the HW should be in Big Endian Format */
+		cell->write.hw_cmd_paddr = cpu_to_be64(cmd_paddr);
+		break;
+
+	default:
+		dev_err(&pdev->dev, "Unsupported API CMD chain type\n");
+		free_cmd_buf(chain, cell_idx);
+		err = -EINVAL;
+		break;
+	}
+
+	return err;
+}
+
+/**
+ * api_cmd_create_cell - create API CMD cell for specific chain
+ * @chain: the API CMD specific chain to create its cell
+ * @cell_idx: the index of the cell to create
+ * @pre_node: previous cell
+ * @node_vaddr: the returned virt addr of the cell
+ *
+ * Return 0 - Success, negative - Failure
+ **/
+static int api_cmd_create_cell(struct hinic_api_cmd_chain *chain,
+			       int cell_idx,
+			       struct hinic_api_cmd_cell *pre_node,
+			       struct hinic_api_cmd_cell **node_vaddr)
+{
+	struct hinic_api_cmd_cell_ctxt *cell_ctxt;
+	struct hinic_hwif *hwif = chain->hwif;
+	struct pci_dev *pdev = hwif->pdev;
+	struct hinic_api_cmd_cell *node;
+	dma_addr_t node_paddr;
+	int err;
+
+	node = dma_alloc_coherent(&pdev->dev, chain->cell_size, &node_paddr,
+				  GFP_KERNEL);
+	if (!node)
+		return -ENOMEM;
+
+	node->read.hw_wb_resp_paddr = 0;
+
+	cell_ctxt = &chain->cell_ctxt[cell_idx];
+	cell_ctxt->cell_vaddr = node;
+	cell_ctxt->cell_paddr = node_paddr;
+
+	if (!pre_node) {
+		chain->head_cell_paddr = node_paddr;
+		chain->head_node = node;
+	} else {
+		/* The data in the HW should be in Big Endian Format */
+		pre_node->next_cell_paddr = cpu_to_be64(node_paddr);
+	}
+
+	switch (chain->chain_type) {
+	case HINIC_API_CMD_WRITE_TO_MGMT_CPU:
+		err = alloc_cmd_buf(chain, node, cell_idx);
+		if (err) {
+			dev_err(&pdev->dev, "Failed to allocate cmd buffer\n");
+			goto err_alloc_cmd_buf;
+		}
+		break;
+
+	default:
+		dev_err(&pdev->dev, "Unsupported API CMD chain type\n");
+		err = -EINVAL;
+		goto err_alloc_cmd_buf;
+	}
+
+	*node_vaddr = node;
+	return 0;
+
+err_alloc_cmd_buf:
+	dma_free_coherent(&pdev->dev, chain->cell_size, node, node_paddr);
+	return err;
+}
+
+/**
+ * api_cmd_destroy_cell - destroy API CMD cell of specific chain
+ * @chain: the API CMD specific chain to destroy its cell
+ * @cell_idx: the cell to destroy
+ **/
+static void api_cmd_destroy_cell(struct hinic_api_cmd_chain *chain,
+				 int cell_idx)
+{
+	struct hinic_api_cmd_cell_ctxt *cell_ctxt;
+	struct hinic_hwif *hwif = chain->hwif;
+	struct pci_dev *pdev = hwif->pdev;
+	struct hinic_api_cmd_cell *node;
+	dma_addr_t node_paddr;
+	size_t node_size;
+
+	cell_ctxt = &chain->cell_ctxt[cell_idx];
+
+	node = cell_ctxt->cell_vaddr;
+	node_paddr = cell_ctxt->cell_paddr;
+	node_size = chain->cell_size;
+
+	if (cell_ctxt->api_cmd_vaddr) {
+		switch (chain->chain_type) {
+		case HINIC_API_CMD_WRITE_TO_MGMT_CPU:
+			free_cmd_buf(chain, cell_idx);
+			break;
+		default:
+			dev_err(&pdev->dev, "Unsupported API CMD chain type\n");
+			break;
+		}
+
+		dma_free_coherent(&pdev->dev, node_size, node,
+				  node_paddr);
+	}
+}
+
+/**
+ * api_cmd_destroy_cells - destroy API CMD cells of specific chain
+ * @chain: the API CMD specific chain to destroy its cells
+ * @num_cells: number of cells to destroy
+ **/
+static void api_cmd_destroy_cells(struct hinic_api_cmd_chain *chain,
+				  int num_cells)
+{
+	int cell_idx;
+
+	for (cell_idx = 0; cell_idx < num_cells; cell_idx++)
+		api_cmd_destroy_cell(chain, cell_idx);
+}
+
+/**
+ * api_cmd_create_cells - create API CMD cells for specific chain
+ * @chain: the API CMD specific chain
+ *
+ * Return 0 - Success, negative - Failure
+ **/
+static int api_cmd_create_cells(struct hinic_api_cmd_chain *chain)
+{
+	struct hinic_api_cmd_cell *node = NULL, *pre_node = NULL;
+	struct hinic_hwif *hwif = chain->hwif;
+	struct pci_dev *pdev = hwif->pdev;
+	int err, cell_idx;
+
+	for (cell_idx = 0; cell_idx < chain->num_cells; cell_idx++) {
+		err = api_cmd_create_cell(chain, cell_idx, pre_node, &node);
+		if (err) {
+			dev_err(&pdev->dev, "Failed to create API CMD cell\n");
+			goto err_create_cell;
+		}
+
+		pre_node = node;
+	}
+
+	/* set the Final node to point on the start */
+	node->next_cell_paddr = cpu_to_be64(chain->head_cell_paddr);
+
+	/* set the current node to be the head */
+	chain->curr_node = chain->head_node;
+	return 0;
+
+err_create_cell:
+	api_cmd_destroy_cells(chain, cell_idx);
+	return err;
+}
+
+/**
+ * api_chain_init - initialize API CMD specific chain
+ * @chain: the API CMD specific chain to initialize
+ * @attr: attributes to set in the chain
+ *
+ * Return 0 - Success, negative - Failure
+ **/
+static int api_chain_init(struct hinic_api_cmd_chain *chain,
+			  struct hinic_api_cmd_chain_attr *attr)
+{
+	struct hinic_hwif *hwif = attr->hwif;
+	struct pci_dev *pdev = hwif->pdev;
+
+	chain->hwif = hwif;
+	chain->chain_type  = attr->chain_type;
+	chain->num_cells = attr->num_cells;
+	chain->cell_size = attr->cell_size;
+
+	chain->prod_idx  = 0;
+	chain->cons_idx  = 0;
+
+	sema_init(&chain->sem, 1);
+
+	chain->cell_ctxt = devm_kcalloc(&pdev->dev, chain->num_cells,
+					sizeof(*chain->cell_ctxt), GFP_KERNEL);
+	if (!chain->cell_ctxt)
+		return -ENOMEM;
+
+	chain->wb_status = dma_alloc_coherent(&pdev->dev,
+					      sizeof(*chain->wb_status),
+					      &chain->wb_status_paddr,
+					      GFP_KERNEL);
+	if (!chain->wb_status) {
+		dev_err(&pdev->dev, "Failed to allocate DMA wb status\n");
+		return -ENOMEM;
+	}
+
+	return 0;
+}
+
+/**
+ * api_chain_free - free API CMD specific chain
+ * @chain: the API CMD specific chain to free
+ **/
+static void api_chain_free(struct hinic_api_cmd_chain *chain)
+{
+	struct hinic_hwif *hwif = chain->hwif;
+	struct pci_dev *pdev = hwif->pdev;
+
+	dma_free_coherent(&pdev->dev, sizeof(*chain->wb_status),
+			  chain->wb_status, chain->wb_status_paddr);
+}
+
+/**
+ * api_cmd_create_chain - create API CMD specific chain
+ * @attr: attributes to set the chain
+ *
+ * Return the created chain
+ **/
+static struct hinic_api_cmd_chain *
+	api_cmd_create_chain(struct hinic_api_cmd_chain_attr *attr)
+{
+	struct hinic_hwif *hwif = attr->hwif;
+	struct pci_dev *pdev = hwif->pdev;
+	struct hinic_api_cmd_chain *chain;
+	int err;
+
+	if (attr->num_cells & (attr->num_cells - 1)) {
+		dev_err(&pdev->dev, "Invalid number of cells, must be power of 2\n");
+		return ERR_PTR(-EINVAL);
+	}
+
+	chain = devm_kzalloc(&pdev->dev, sizeof(*chain), GFP_KERNEL);
+	if (!chain)
+		return ERR_PTR(-ENOMEM);
+
+	err = api_chain_init(chain, attr);
+	if (err) {
+		dev_err(&pdev->dev, "Failed to initialize chain\n");
+		return ERR_PTR(err);
+	}
+
+	err = api_cmd_create_cells(chain);
+	if (err) {
+		dev_err(&pdev->dev, "Failed to create cells for API CMD chain\n");
+		goto err_create_cells;
+	}
+
+	err = api_cmd_chain_hw_init(chain);
+	if (err) {
+		dev_err(&pdev->dev, "Failed to initialize chain HW\n");
+		goto err_chain_hw_init;
+	}
+
+	return chain;
+
+err_chain_hw_init:
+	api_cmd_destroy_cells(chain, chain->num_cells);
+
+err_create_cells:
+	api_chain_free(chain);
+	return ERR_PTR(err);
+}
+
+/**
+ * api_cmd_destroy_chain - destroy API CMD specific chain
+ * @chain: the API CMD specific chain to destroy
+ **/
+static void api_cmd_destroy_chain(struct hinic_api_cmd_chain *chain)
+{
+	api_cmd_chain_hw_clean(chain);
+	api_cmd_destroy_cells(chain, chain->num_cells);
+	api_chain_free(chain);
+}
+
+/**
+ * hinic_api_cmd_init - Initialize all the API CMD chains
+ * @chain: the API CMD chains that are initialized
+ * @hwif: the hardware interface of a pci function device
+ *
+ * Return 0 - Success, negative - Failure
+ **/
+int hinic_api_cmd_init(struct hinic_api_cmd_chain **chain,
+		       struct hinic_hwif *hwif)
+{
+	enum hinic_api_cmd_chain_type type, chain_type;
+	struct hinic_api_cmd_chain_attr attr;
+	struct pci_dev *pdev = hwif->pdev;
+	size_t hw_cell_sz;
+	int err;
+
+	hw_cell_sz = sizeof(struct hinic_api_cmd_cell);
+
+	attr.hwif = hwif;
+	attr.num_cells  = API_CHAIN_NUM_CELLS;
+	attr.cell_size  = API_CMD_CELL_SIZE(hw_cell_sz);
+
+	chain_type = HINIC_API_CMD_WRITE_TO_MGMT_CPU;
+	for ( ; chain_type < HINIC_API_CMD_MAX; chain_type++) {
+		attr.chain_type = chain_type;
+
+		if (chain_type != HINIC_API_CMD_WRITE_TO_MGMT_CPU)
+			continue;
+
+		chain[chain_type] = api_cmd_create_chain(&attr);
+		if (IS_ERR(chain[chain_type])) {
+			dev_err(&pdev->dev, "Failed to create chain %d\n",
+				chain_type);
+			err = PTR_ERR(chain[chain_type]);
+			goto err_create_chain;
+		}
+	}
+
+	return 0;
+
+err_create_chain:
+	type = HINIC_API_CMD_WRITE_TO_MGMT_CPU;
+	for ( ; type < chain_type; type++) {
+		if (type != HINIC_API_CMD_WRITE_TO_MGMT_CPU)
+			continue;
+
+		api_cmd_destroy_chain(chain[type]);
+	}
+
+	return err;
+}
+
+/**
+ * hinic_api_cmd_free - free the API CMD chains
+ * @chain: the API CMD chains that are freed
+ **/
+void hinic_api_cmd_free(struct hinic_api_cmd_chain **chain)
+{
+	enum hinic_api_cmd_chain_type chain_type;
+
+	chain_type = HINIC_API_CMD_WRITE_TO_MGMT_CPU;
+	for ( ; chain_type < HINIC_API_CMD_MAX; chain_type++) {
+		if (chain_type != HINIC_API_CMD_WRITE_TO_MGMT_CPU)
+			continue;
+
+		api_cmd_destroy_chain(chain[chain_type]);
+	}
+}
diff --git a/drivers/net/ethernet/huawei/hinic/hinic_hw_api_cmd.h b/drivers/net/ethernet/huawei/hinic/hinic_hw_api_cmd.h
new file mode 100644
index 000000000000..6d1654b050ad
--- /dev/null
+++ b/drivers/net/ethernet/huawei/hinic/hinic_hw_api_cmd.h
@@ -0,0 +1,203 @@
+/* SPDX-License-Identifier: GPL-2.0-only */
+/*
+ * Huawei HiNIC PCI Express Linux driver
+ * Copyright(c) 2017 Huawei Technologies Co., Ltd
+ */
+
+#ifndef HINIC_HW_API_CMD_H
+#define HINIC_HW_API_CMD_H
+
+#include <linux/types.h>
+#include <linux/semaphore.h>
+
+#include "hinic_hw_if.h"
+
+#define HINIC_API_CMD_PI_IDX_SHIFT                              0
+
+#define HINIC_API_CMD_PI_IDX_MASK                               0xFFFFFF
+
+#define HINIC_API_CMD_PI_SET(val, member)                       \
+	(((u32)(val) & HINIC_API_CMD_PI_##member##_MASK) <<     \
+	 HINIC_API_CMD_PI_##member##_SHIFT)
+
+#define HINIC_API_CMD_PI_CLEAR(val, member)                     \
+	((val) & (~(HINIC_API_CMD_PI_##member##_MASK            \
+	 << HINIC_API_CMD_PI_##member##_SHIFT)))
+
+#define HINIC_API_CMD_CHAIN_REQ_RESTART_SHIFT                   1
+
+#define HINIC_API_CMD_CHAIN_REQ_RESTART_MASK                    0x1
+
+#define HINIC_API_CMD_CHAIN_REQ_SET(val, member)                \
+	(((u32)(val) & HINIC_API_CMD_CHAIN_REQ_##member##_MASK) << \
+	 HINIC_API_CMD_CHAIN_REQ_##member##_SHIFT)
+
+#define HINIC_API_CMD_CHAIN_REQ_GET(val, member)                \
+	(((val) >> HINIC_API_CMD_CHAIN_REQ_##member##_SHIFT) &  \
+	 HINIC_API_CMD_CHAIN_REQ_##member##_MASK)
+
+#define HINIC_API_CMD_CHAIN_REQ_CLEAR(val, member)              \
+	((val) & (~(HINIC_API_CMD_CHAIN_REQ_##member##_MASK     \
+	 << HINIC_API_CMD_CHAIN_REQ_##member##_SHIFT)))
+
+#define HINIC_API_CMD_CHAIN_CTRL_RESTART_WB_STAT_SHIFT          1
+#define HINIC_API_CMD_CHAIN_CTRL_XOR_ERR_SHIFT                  2
+#define HINIC_API_CMD_CHAIN_CTRL_AEQE_EN_SHIFT                  4
+#define HINIC_API_CMD_CHAIN_CTRL_AEQ_ID_SHIFT                   8
+#define HINIC_API_CMD_CHAIN_CTRL_XOR_CHK_EN_SHIFT               28
+#define HINIC_API_CMD_CHAIN_CTRL_CELL_SIZE_SHIFT                30
+
+#define HINIC_API_CMD_CHAIN_CTRL_RESTART_WB_STAT_MASK           0x1
+#define HINIC_API_CMD_CHAIN_CTRL_XOR_ERR_MASK                   0x1
+#define HINIC_API_CMD_CHAIN_CTRL_AEQE_EN_MASK                   0x1
+#define HINIC_API_CMD_CHAIN_CTRL_AEQ_ID_MASK                    0x3
+#define HINIC_API_CMD_CHAIN_CTRL_XOR_CHK_EN_MASK                0x3
+#define HINIC_API_CMD_CHAIN_CTRL_CELL_SIZE_MASK                 0x3
+
+#define HINIC_API_CMD_CHAIN_CTRL_SET(val, member)               \
+	(((u32)(val) & HINIC_API_CMD_CHAIN_CTRL_##member##_MASK) << \
+	 HINIC_API_CMD_CHAIN_CTRL_##member##_SHIFT)
+
+#define HINIC_API_CMD_CHAIN_CTRL_CLEAR(val, member)             \
+	((val) & (~(HINIC_API_CMD_CHAIN_CTRL_##member##_MASK    \
+	 << HINIC_API_CMD_CHAIN_CTRL_##member##_SHIFT)))
+
+#define HINIC_API_CMD_CELL_CTRL_DATA_SZ_SHIFT                   0
+#define HINIC_API_CMD_CELL_CTRL_RD_DMA_ATTR_SHIFT               16
+#define HINIC_API_CMD_CELL_CTRL_WR_DMA_ATTR_SHIFT               24
+#define HINIC_API_CMD_CELL_CTRL_XOR_CHKSUM_SHIFT                56
+
+#define HINIC_API_CMD_CELL_CTRL_DATA_SZ_MASK                    0x3F
+#define HINIC_API_CMD_CELL_CTRL_RD_DMA_ATTR_MASK                0x3F
+#define HINIC_API_CMD_CELL_CTRL_WR_DMA_ATTR_MASK                0x3F
+#define HINIC_API_CMD_CELL_CTRL_XOR_CHKSUM_MASK                 0xFF
+
+#define HINIC_API_CMD_CELL_CTRL_SET(val, member)                \
+	((((u64)val) & HINIC_API_CMD_CELL_CTRL_##member##_MASK) << \
+	 HINIC_API_CMD_CELL_CTRL_##member##_SHIFT)
+
+#define HINIC_API_CMD_DESC_API_TYPE_SHIFT                       0
+#define HINIC_API_CMD_DESC_RD_WR_SHIFT                          1
+#define HINIC_API_CMD_DESC_MGMT_BYPASS_SHIFT                    2
+#define HINIC_API_CMD_DESC_DEST_SHIFT                           32
+#define HINIC_API_CMD_DESC_SIZE_SHIFT                           40
+#define HINIC_API_CMD_DESC_XOR_CHKSUM_SHIFT                     56
+
+#define HINIC_API_CMD_DESC_API_TYPE_MASK                        0x1
+#define HINIC_API_CMD_DESC_RD_WR_MASK                           0x1
+#define HINIC_API_CMD_DESC_MGMT_BYPASS_MASK                     0x1
+#define HINIC_API_CMD_DESC_DEST_MASK                            0x1F
+#define HINIC_API_CMD_DESC_SIZE_MASK                            0x7FF
+#define HINIC_API_CMD_DESC_XOR_CHKSUM_MASK                      0xFF
+
+#define HINIC_API_CMD_DESC_SET(val, member)                     \
+	((((u64)val) & HINIC_API_CMD_DESC_##member##_MASK) <<   \
+	 HINIC_API_CMD_DESC_##member##_SHIFT)
+
+#define HINIC_API_CMD_STATUS_HEADER_CHAIN_ID_SHIFT              16
+
+#define HINIC_API_CMD_STATUS_HEADER_CHAIN_ID_MASK               0xFF
+
+#define HINIC_API_CMD_STATUS_HEADER_GET(val, member)            \
+	(((val) >> HINIC_API_CMD_STATUS_HEADER_##member##_SHIFT) & \
+	 HINIC_API_CMD_STATUS_HEADER_##member##_MASK)
+
+#define HINIC_API_CMD_STATUS_CONS_IDX_SHIFT                     0
+#define HINIC_API_CMD_STATUS_FSM_SHIFT				24
+#define HINIC_API_CMD_STATUS_CHKSUM_ERR_SHIFT                   28
+#define HINIC_API_CMD_STATUS_CPLD_ERR_SHIFT			30
+
+#define HINIC_API_CMD_STATUS_CONS_IDX_MASK                      0xFFFFFF
+#define HINIC_API_CMD_STATUS_FSM_MASK				0xFU
+#define HINIC_API_CMD_STATUS_CHKSUM_ERR_MASK                    0x3
+#define HINIC_API_CMD_STATUS_CPLD_ERR_MASK			0x1U
+
+#define HINIC_API_CMD_STATUS_GET(val, member)                   \
+	(((val) >> HINIC_API_CMD_STATUS_##member##_SHIFT) &     \
+	 HINIC_API_CMD_STATUS_##member##_MASK)
+
+enum hinic_api_cmd_chain_type {
+	HINIC_API_CMD_WRITE_TO_MGMT_CPU = 2,
+
+	HINIC_API_CMD_MAX,
+};
+
+struct hinic_api_cmd_chain_attr {
+	struct hinic_hwif               *hwif;
+	enum hinic_api_cmd_chain_type   chain_type;
+
+	u32                             num_cells;
+	u16                             cell_size;
+};
+
+struct hinic_api_cmd_status {
+	u64     header;
+	u32     status;
+	u32     rsvd0;
+	u32     rsvd1;
+	u32     rsvd2;
+	u64     rsvd3;
+};
+
+/* HW struct */
+struct hinic_api_cmd_cell {
+	u64 ctrl;
+
+	/* address is 64 bit in HW struct */
+	u64 next_cell_paddr;
+
+	u64 desc;
+
+	/* HW struct */
+	union {
+		struct {
+			u64 hw_cmd_paddr;
+		} write;
+
+		struct {
+			u64 hw_wb_resp_paddr;
+			u64 hw_cmd_paddr;
+		} read;
+	};
+};
+
+struct hinic_api_cmd_cell_ctxt {
+	dma_addr_t                      cell_paddr;
+	struct hinic_api_cmd_cell       *cell_vaddr;
+
+	dma_addr_t                      api_cmd_paddr;
+	u8                              *api_cmd_vaddr;
+};
+
+struct hinic_api_cmd_chain {
+	struct hinic_hwif               *hwif;
+	enum hinic_api_cmd_chain_type   chain_type;
+
+	u32                             num_cells;
+	u16                             cell_size;
+
+	/* HW members in 24 bit format */
+	u32                             prod_idx;
+	u32                             cons_idx;
+
+	struct semaphore                sem;
+
+	struct hinic_api_cmd_cell_ctxt  *cell_ctxt;
+
+	dma_addr_t                      wb_status_paddr;
+	struct hinic_api_cmd_status     *wb_status;
+
+	dma_addr_t                      head_cell_paddr;
+	struct hinic_api_cmd_cell       *head_node;
+	struct hinic_api_cmd_cell       *curr_node;
+};
+
+int hinic_api_cmd_write(struct hinic_api_cmd_chain *chain,
+			enum hinic_node_id dest, u8 *cmd, u16 size);
+
+int hinic_api_cmd_init(struct hinic_api_cmd_chain **chain,
+		       struct hinic_hwif *hwif);
+
+void hinic_api_cmd_free(struct hinic_api_cmd_chain **chain);
+
+#endif
diff --git a/drivers/net/ethernet/huawei/hinic/hinic_hw_cmdq.c b/drivers/net/ethernet/huawei/hinic/hinic_hw_cmdq.c
new file mode 100644
index 000000000000..d39eec9c62bf
--- /dev/null
+++ b/drivers/net/ethernet/huawei/hinic/hinic_hw_cmdq.c
@@ -0,0 +1,963 @@
+// SPDX-License-Identifier: GPL-2.0-only
+/*
+ * Huawei HiNIC PCI Express Linux driver
+ * Copyright(c) 2017 Huawei Technologies Co., Ltd
+ */
+
+#include <linux/kernel.h>
+#include <linux/types.h>
+#include <linux/errno.h>
+#include <linux/pci.h>
+#include <linux/device.h>
+#include <linux/slab.h>
+#include <linux/vmalloc.h>
+#include <linux/spinlock.h>
+#include <linux/sizes.h>
+#include <linux/atomic.h>
+#include <linux/log2.h>
+#include <linux/io.h>
+#include <linux/completion.h>
+#include <linux/err.h>
+#include <asm/byteorder.h>
+#include <asm/barrier.h>
+
+#include "hinic_common.h"
+#include "hinic_hw_if.h"
+#include "hinic_hw_eqs.h"
+#include "hinic_hw_mgmt.h"
+#include "hinic_hw_wqe.h"
+#include "hinic_hw_wq.h"
+#include "hinic_hw_cmdq.h"
+#include "hinic_hw_io.h"
+#include "hinic_hw_dev.h"
+
+#define CMDQ_CEQE_TYPE_SHIFT                    0
+
+#define CMDQ_CEQE_TYPE_MASK                     0x7
+
+#define CMDQ_CEQE_GET(val, member)              \
+			(((val) >> CMDQ_CEQE_##member##_SHIFT) \
+			 & CMDQ_CEQE_##member##_MASK)
+
+#define CMDQ_WQE_ERRCODE_VAL_SHIFT              20
+
+#define CMDQ_WQE_ERRCODE_VAL_MASK               0xF
+
+#define CMDQ_WQE_ERRCODE_GET(val, member)       \
+			(((val) >> CMDQ_WQE_ERRCODE_##member##_SHIFT) \
+			 & CMDQ_WQE_ERRCODE_##member##_MASK)
+
+#define CMDQ_DB_PI_OFF(pi)              (((u16)LOWER_8_BITS(pi)) << 3)
+
+#define CMDQ_DB_ADDR(db_base, pi)       ((db_base) + CMDQ_DB_PI_OFF(pi))
+
+#define CMDQ_WQE_HEADER(wqe)            ((struct hinic_cmdq_header *)(wqe))
+
+#define CMDQ_WQE_COMPLETED(ctrl_info)   \
+			HINIC_CMDQ_CTRL_GET(ctrl_info, HW_BUSY_BIT)
+
+#define FIRST_DATA_TO_WRITE_LAST        sizeof(u64)
+
+#define CMDQ_DB_OFF                     SZ_2K
+
+#define CMDQ_WQEBB_SIZE                 64
+#define CMDQ_WQE_SIZE                   64
+#define CMDQ_DEPTH                      SZ_4K
+
+#define CMDQ_WQ_PAGE_SIZE               SZ_256K
+
+#define WQE_LCMD_SIZE                   64
+#define WQE_SCMD_SIZE                   64
+
+#define COMPLETE_LEN                    3
+
+#define CMDQ_TIMEOUT                    1000
+
+#define CMDQ_PFN(addr, page_size)       ((addr) >> (ilog2(page_size)))
+
+#define cmdq_to_cmdqs(cmdq)     container_of((cmdq) - (cmdq)->cmdq_type, \
+					     struct hinic_cmdqs, cmdq[0])
+
+#define cmdqs_to_func_to_io(cmdqs)      container_of(cmdqs, \
+						     struct hinic_func_to_io, \
+						     cmdqs)
+
+enum completion_format {
+	COMPLETE_DIRECT = 0,
+	COMPLETE_SGE    = 1,
+};
+
+enum data_format {
+	DATA_SGE        = 0,
+	DATA_DIRECT     = 1,
+};
+
+enum bufdesc_len {
+	BUFDESC_LCMD_LEN = 2,   /* 16 bytes - 2(8 byte unit) */
+	BUFDESC_SCMD_LEN = 3,   /* 24 bytes - 3(8 byte unit) */
+};
+
+enum ctrl_sect_len {
+	CTRL_SECT_LEN        = 1, /* 4 bytes (ctrl) - 1(8 byte unit) */
+	CTRL_DIRECT_SECT_LEN = 2, /* 12 bytes (ctrl + rsvd) - 2(8 byte unit) */
+};
+
+enum cmdq_scmd_type {
+	CMDQ_SET_ARM_CMD = 2,
+};
+
+enum cmdq_cmd_type {
+	CMDQ_CMD_SYNC_DIRECT_RESP = 0,
+	CMDQ_CMD_SYNC_SGE_RESP    = 1,
+};
+
+enum completion_request {
+	NO_CEQ  = 0,
+	CEQ_SET = 1,
+};
+
+/**
+ * hinic_alloc_cmdq_buf - alloc buffer for sending command
+ * @cmdqs: the cmdqs
+ * @cmdq_buf: the buffer returned in this struct
+ *
+ * Return 0 - Success, negative - Failure
+ **/
+int hinic_alloc_cmdq_buf(struct hinic_cmdqs *cmdqs,
+			 struct hinic_cmdq_buf *cmdq_buf)
+{
+	struct hinic_hwif *hwif = cmdqs->hwif;
+	struct pci_dev *pdev = hwif->pdev;
+
+	cmdq_buf->buf = dma_pool_alloc(cmdqs->cmdq_buf_pool, GFP_KERNEL,
+				       &cmdq_buf->dma_addr);
+	if (!cmdq_buf->buf) {
+		dev_err(&pdev->dev, "Failed to allocate cmd from the pool\n");
+		return -ENOMEM;
+	}
+
+	return 0;
+}
+
+/**
+ * hinic_free_cmdq_buf - free buffer
+ * @cmdqs: the cmdqs
+ * @cmdq_buf: the buffer to free that is in this struct
+ **/
+void hinic_free_cmdq_buf(struct hinic_cmdqs *cmdqs,
+			 struct hinic_cmdq_buf *cmdq_buf)
+{
+	dma_pool_free(cmdqs->cmdq_buf_pool, cmdq_buf->buf, cmdq_buf->dma_addr);
+}
+
+static unsigned int cmdq_wqe_size_from_bdlen(enum bufdesc_len len)
+{
+	unsigned int wqe_size = 0;
+
+	switch (len) {
+	case BUFDESC_LCMD_LEN:
+		wqe_size = WQE_LCMD_SIZE;
+		break;
+	case BUFDESC_SCMD_LEN:
+		wqe_size = WQE_SCMD_SIZE;
+		break;
+	}
+
+	return wqe_size;
+}
+
+static void cmdq_set_sge_completion(struct hinic_cmdq_completion *completion,
+				    struct hinic_cmdq_buf *buf_out)
+{
+	struct hinic_sge_resp *sge_resp = &completion->sge_resp;
+
+	hinic_set_sge(&sge_resp->sge, buf_out->dma_addr, buf_out->size);
+}
+
+static void cmdq_prepare_wqe_ctrl(struct hinic_cmdq_wqe *wqe, int wrapped,
+				  enum hinic_cmd_ack_type ack_type,
+				  enum hinic_mod_type mod, u8 cmd, u16 prod_idx,
+				  enum completion_format complete_format,
+				  enum data_format data_format,
+				  enum bufdesc_len buf_len)
+{
+	struct hinic_cmdq_wqe_lcmd *wqe_lcmd;
+	struct hinic_cmdq_wqe_scmd *wqe_scmd;
+	enum ctrl_sect_len ctrl_len;
+	struct hinic_ctrl *ctrl;
+	u32 saved_data;
+
+	if (data_format == DATA_SGE) {
+		wqe_lcmd = &wqe->wqe_lcmd;
+
+		wqe_lcmd->status.status_info = 0;
+		ctrl = &wqe_lcmd->ctrl;
+		ctrl_len = CTRL_SECT_LEN;
+	} else {
+		wqe_scmd = &wqe->direct_wqe.wqe_scmd;
+
+		wqe_scmd->status.status_info = 0;
+		ctrl = &wqe_scmd->ctrl;
+		ctrl_len = CTRL_DIRECT_SECT_LEN;
+	}
+
+	ctrl->ctrl_info = HINIC_CMDQ_CTRL_SET(prod_idx, PI)             |
+			  HINIC_CMDQ_CTRL_SET(cmd, CMD)                 |
+			  HINIC_CMDQ_CTRL_SET(mod, MOD)                 |
+			  HINIC_CMDQ_CTRL_SET(ack_type, ACK_TYPE);
+
+	CMDQ_WQE_HEADER(wqe)->header_info =
+		HINIC_CMDQ_WQE_HEADER_SET(buf_len, BUFDESC_LEN)            |
+		HINIC_CMDQ_WQE_HEADER_SET(complete_format, COMPLETE_FMT)   |
+		HINIC_CMDQ_WQE_HEADER_SET(data_format, DATA_FMT)           |
+		HINIC_CMDQ_WQE_HEADER_SET(CEQ_SET, COMPLETE_REQ)           |
+		HINIC_CMDQ_WQE_HEADER_SET(COMPLETE_LEN, COMPLETE_SECT_LEN) |
+		HINIC_CMDQ_WQE_HEADER_SET(ctrl_len, CTRL_LEN)              |
+		HINIC_CMDQ_WQE_HEADER_SET(wrapped, TOGGLED_WRAPPED);
+
+	saved_data = CMDQ_WQE_HEADER(wqe)->saved_data;
+	saved_data = HINIC_SAVED_DATA_CLEAR(saved_data, ARM);
+
+	if (cmd == CMDQ_SET_ARM_CMD && mod == HINIC_MOD_COMM)
+		CMDQ_WQE_HEADER(wqe)->saved_data |=
+						HINIC_SAVED_DATA_SET(1, ARM);
+	else
+		CMDQ_WQE_HEADER(wqe)->saved_data = saved_data;
+}
+
+static void cmdq_set_lcmd_bufdesc(struct hinic_cmdq_wqe_lcmd *wqe_lcmd,
+				  struct hinic_cmdq_buf *buf_in)
+{
+	hinic_set_sge(&wqe_lcmd->buf_desc.sge, buf_in->dma_addr, buf_in->size);
+}
+
+static void cmdq_set_direct_wqe_data(struct hinic_cmdq_direct_wqe *wqe,
+				     void *buf_in, u32 in_size)
+{
+	struct hinic_cmdq_wqe_scmd *wqe_scmd = &wqe->wqe_scmd;
+
+	wqe_scmd->buf_desc.buf_len = in_size;
+	memcpy(wqe_scmd->buf_desc.data, buf_in, in_size);
+}
+
+static void cmdq_set_lcmd_wqe(struct hinic_cmdq_wqe *wqe,
+			      enum cmdq_cmd_type cmd_type,
+			      struct hinic_cmdq_buf *buf_in,
+			      struct hinic_cmdq_buf *buf_out, int wrapped,
+			      enum hinic_cmd_ack_type ack_type,
+			      enum hinic_mod_type mod, u8 cmd, u16 prod_idx)
+{
+	struct hinic_cmdq_wqe_lcmd *wqe_lcmd = &wqe->wqe_lcmd;
+	enum completion_format complete_format;
+
+	switch (cmd_type) {
+	case CMDQ_CMD_SYNC_SGE_RESP:
+		complete_format = COMPLETE_SGE;
+		cmdq_set_sge_completion(&wqe_lcmd->completion, buf_out);
+		break;
+	case CMDQ_CMD_SYNC_DIRECT_RESP:
+		complete_format = COMPLETE_DIRECT;
+		wqe_lcmd->completion.direct_resp = 0;
+		break;
+	}
+
+	cmdq_prepare_wqe_ctrl(wqe, wrapped, ack_type, mod, cmd,
+			      prod_idx, complete_format, DATA_SGE,
+			      BUFDESC_LCMD_LEN);
+
+	cmdq_set_lcmd_bufdesc(wqe_lcmd, buf_in);
+}
+
+static void cmdq_set_direct_wqe(struct hinic_cmdq_wqe *wqe,
+				enum cmdq_cmd_type cmd_type,
+				void *buf_in, u16 in_size,
+				struct hinic_cmdq_buf *buf_out, int wrapped,
+				enum hinic_cmd_ack_type ack_type,
+				enum hinic_mod_type mod, u8 cmd, u16 prod_idx)
+{
+	struct hinic_cmdq_direct_wqe *direct_wqe = &wqe->direct_wqe;
+	enum completion_format complete_format;
+	struct hinic_cmdq_wqe_scmd *wqe_scmd;
+
+	wqe_scmd = &direct_wqe->wqe_scmd;
+
+	switch (cmd_type) {
+	case CMDQ_CMD_SYNC_SGE_RESP:
+		complete_format = COMPLETE_SGE;
+		cmdq_set_sge_completion(&wqe_scmd->completion, buf_out);
+		break;
+	case CMDQ_CMD_SYNC_DIRECT_RESP:
+		complete_format = COMPLETE_DIRECT;
+		wqe_scmd->completion.direct_resp = 0;
+		break;
+	}
+
+	cmdq_prepare_wqe_ctrl(wqe, wrapped, ack_type, mod, cmd, prod_idx,
+			      complete_format, DATA_DIRECT, BUFDESC_SCMD_LEN);
+
+	cmdq_set_direct_wqe_data(direct_wqe, buf_in, in_size);
+}
+
+static void cmdq_wqe_fill(void *dst, void *src)
+{
+	memcpy(dst + FIRST_DATA_TO_WRITE_LAST, src + FIRST_DATA_TO_WRITE_LAST,
+	       CMDQ_WQE_SIZE - FIRST_DATA_TO_WRITE_LAST);
+
+	wmb();          /* The first 8 bytes should be written last */
+
+	*(u64 *)dst = *(u64 *)src;
+}
+
+static void cmdq_fill_db(u32 *db_info,
+			 enum hinic_cmdq_type cmdq_type, u16 prod_idx)
+{
+	*db_info = HINIC_CMDQ_DB_INFO_SET(UPPER_8_BITS(prod_idx), HI_PROD_IDX) |
+		   HINIC_CMDQ_DB_INFO_SET(HINIC_CTRL_PATH, PATH)               |
+		   HINIC_CMDQ_DB_INFO_SET(cmdq_type, CMDQ_TYPE)                |
+		   HINIC_CMDQ_DB_INFO_SET(HINIC_DB_CMDQ_TYPE, DB_TYPE);
+}
+
+static void cmdq_set_db(struct hinic_cmdq *cmdq,
+			enum hinic_cmdq_type cmdq_type, u16 prod_idx)
+{
+	u32 db_info;
+
+	cmdq_fill_db(&db_info, cmdq_type, prod_idx);
+
+	/* The data that is written to HW should be in Big Endian Format */
+	db_info = cpu_to_be32(db_info);
+
+	wmb();  /* write all before the doorbell */
+
+	writel(db_info, CMDQ_DB_ADDR(cmdq->db_base, prod_idx));
+}
+
+static int cmdq_sync_cmd_direct_resp(struct hinic_cmdq *cmdq,
+				     enum hinic_mod_type mod, u8 cmd,
+				     struct hinic_cmdq_buf *buf_in,
+				     u64 *resp)
+{
+	struct hinic_cmdq_wqe *curr_cmdq_wqe, cmdq_wqe;
+	u16 curr_prod_idx, next_prod_idx;
+	int errcode, wrapped, num_wqebbs;
+	struct hinic_wq *wq = cmdq->wq;
+	struct hinic_hw_wqe *hw_wqe;
+	struct completion done;
+
+	/* Keep doorbell index correct. bh - for tasklet(ceq). */
+	spin_lock_bh(&cmdq->cmdq_lock);
+
+	/* WQE_SIZE = WQEBB_SIZE, we will get the wq element and not shadow*/
+	hw_wqe = hinic_get_wqe(wq, WQE_LCMD_SIZE, &curr_prod_idx);
+	if (IS_ERR(hw_wqe)) {
+		spin_unlock_bh(&cmdq->cmdq_lock);
+		return -EBUSY;
+	}
+
+	curr_cmdq_wqe = &hw_wqe->cmdq_wqe;
+
+	wrapped = cmdq->wrapped;
+
+	num_wqebbs = ALIGN(WQE_LCMD_SIZE, wq->wqebb_size) / wq->wqebb_size;
+	next_prod_idx = curr_prod_idx + num_wqebbs;
+	if (next_prod_idx >= wq->q_depth) {
+		cmdq->wrapped = !cmdq->wrapped;
+		next_prod_idx -= wq->q_depth;
+	}
+
+	cmdq->errcode[curr_prod_idx] = &errcode;
+
+	init_completion(&done);
+	cmdq->done[curr_prod_idx] = &done;
+
+	cmdq_set_lcmd_wqe(&cmdq_wqe, CMDQ_CMD_SYNC_DIRECT_RESP, buf_in, NULL,
+			  wrapped, HINIC_CMD_ACK_TYPE_CMDQ, mod, cmd,
+			  curr_prod_idx);
+
+	/* The data that is written to HW should be in Big Endian Format */
+	hinic_cpu_to_be32(&cmdq_wqe, WQE_LCMD_SIZE);
+
+	/* CMDQ WQE is not shadow, therefore wqe will be written to wq */
+	cmdq_wqe_fill(curr_cmdq_wqe, &cmdq_wqe);
+
+	cmdq_set_db(cmdq, HINIC_CMDQ_SYNC, next_prod_idx);
+
+	spin_unlock_bh(&cmdq->cmdq_lock);
+
+	if (!wait_for_completion_timeout(&done,
+					 msecs_to_jiffies(CMDQ_TIMEOUT))) {
+		spin_lock_bh(&cmdq->cmdq_lock);
+
+		if (cmdq->errcode[curr_prod_idx] == &errcode)
+			cmdq->errcode[curr_prod_idx] = NULL;
+
+		if (cmdq->done[curr_prod_idx] == &done)
+			cmdq->done[curr_prod_idx] = NULL;
+
+		spin_unlock_bh(&cmdq->cmdq_lock);
+
+		hinic_dump_ceq_info(cmdq->hwdev);
+		return -ETIMEDOUT;
+	}
+
+	smp_rmb();      /* read error code after completion */
+
+	if (resp) {
+		struct hinic_cmdq_wqe_lcmd *wqe_lcmd = &curr_cmdq_wqe->wqe_lcmd;
+
+		*resp = cpu_to_be64(wqe_lcmd->completion.direct_resp);
+	}
+
+	if (errcode != 0)
+		return -EFAULT;
+
+	return 0;
+}
+
+static int cmdq_set_arm_bit(struct hinic_cmdq *cmdq, void *buf_in,
+			    u16 in_size)
+{
+	struct hinic_cmdq_wqe *curr_cmdq_wqe, cmdq_wqe;
+	u16 curr_prod_idx, next_prod_idx;
+	struct hinic_wq *wq = cmdq->wq;
+	struct hinic_hw_wqe *hw_wqe;
+	int wrapped, num_wqebbs;
+
+	/* Keep doorbell index correct */
+	spin_lock(&cmdq->cmdq_lock);
+
+	/* WQE_SIZE = WQEBB_SIZE, we will get the wq element and not shadow*/
+	hw_wqe = hinic_get_wqe(wq, WQE_SCMD_SIZE, &curr_prod_idx);
+	if (IS_ERR(hw_wqe)) {
+		spin_unlock(&cmdq->cmdq_lock);
+		return -EBUSY;
+	}
+
+	curr_cmdq_wqe = &hw_wqe->cmdq_wqe;
+
+	wrapped = cmdq->wrapped;
+
+	num_wqebbs = ALIGN(WQE_SCMD_SIZE, wq->wqebb_size) / wq->wqebb_size;
+	next_prod_idx = curr_prod_idx + num_wqebbs;
+	if (next_prod_idx >= wq->q_depth) {
+		cmdq->wrapped = !cmdq->wrapped;
+		next_prod_idx -= wq->q_depth;
+	}
+
+	cmdq_set_direct_wqe(&cmdq_wqe, CMDQ_CMD_SYNC_DIRECT_RESP, buf_in,
+			    in_size, NULL, wrapped, HINIC_CMD_ACK_TYPE_CMDQ,
+			    HINIC_MOD_COMM, CMDQ_SET_ARM_CMD, curr_prod_idx);
+
+	/* The data that is written to HW should be in Big Endian Format */
+	hinic_cpu_to_be32(&cmdq_wqe, WQE_SCMD_SIZE);
+
+	/* cmdq wqe is not shadow, therefore wqe will be written to wq */
+	cmdq_wqe_fill(curr_cmdq_wqe, &cmdq_wqe);
+
+	cmdq_set_db(cmdq, HINIC_CMDQ_SYNC, next_prod_idx);
+
+	spin_unlock(&cmdq->cmdq_lock);
+	return 0;
+}
+
+static int cmdq_params_valid(struct hinic_cmdq_buf *buf_in)
+{
+	if (buf_in->size > HINIC_CMDQ_MAX_DATA_SIZE)
+		return -EINVAL;
+
+	return 0;
+}
+
+/**
+ * hinic_cmdq_direct_resp - send command with direct data as resp
+ * @cmdqs: the cmdqs
+ * @mod: module on the card that will handle the command
+ * @cmd: the command
+ * @buf_in: the buffer for the command
+ * @resp: the response to return
+ *
+ * Return 0 - Success, negative - Failure
+ **/
+int hinic_cmdq_direct_resp(struct hinic_cmdqs *cmdqs,
+			   enum hinic_mod_type mod, u8 cmd,
+			   struct hinic_cmdq_buf *buf_in, u64 *resp)
+{
+	struct hinic_hwif *hwif = cmdqs->hwif;
+	struct pci_dev *pdev = hwif->pdev;
+	int err;
+
+	err = cmdq_params_valid(buf_in);
+	if (err) {
+		dev_err(&pdev->dev, "Invalid CMDQ parameters\n");
+		return err;
+	}
+
+	return cmdq_sync_cmd_direct_resp(&cmdqs->cmdq[HINIC_CMDQ_SYNC],
+					 mod, cmd, buf_in, resp);
+}
+
+/**
+ * hinic_set_arm_bit - set arm bit for enable interrupt again
+ * @cmdqs: the cmdqs
+ * @q_type: type of queue to set the arm bit for
+ * @q_id: the queue number
+ *
+ * Return 0 - Success, negative - Failure
+ **/
+static int hinic_set_arm_bit(struct hinic_cmdqs *cmdqs,
+			     enum hinic_set_arm_qtype q_type, u32 q_id)
+{
+	struct hinic_cmdq *cmdq = &cmdqs->cmdq[HINIC_CMDQ_SYNC];
+	struct hinic_hwif *hwif = cmdqs->hwif;
+	struct pci_dev *pdev = hwif->pdev;
+	struct hinic_cmdq_arm_bit arm_bit;
+	int err;
+
+	arm_bit.q_type = q_type;
+	arm_bit.q_id   = q_id;
+
+	err = cmdq_set_arm_bit(cmdq, &arm_bit, sizeof(arm_bit));
+	if (err) {
+		dev_err(&pdev->dev, "Failed to set arm for qid %d\n", q_id);
+		return err;
+	}
+
+	return 0;
+}
+
+static void clear_wqe_complete_bit(struct hinic_cmdq *cmdq,
+				   struct hinic_cmdq_wqe *wqe)
+{
+	u32 header_info = be32_to_cpu(CMDQ_WQE_HEADER(wqe)->header_info);
+	unsigned int bufdesc_len, wqe_size;
+	struct hinic_ctrl *ctrl;
+
+	bufdesc_len = HINIC_CMDQ_WQE_HEADER_GET(header_info, BUFDESC_LEN);
+	wqe_size = cmdq_wqe_size_from_bdlen(bufdesc_len);
+	if (wqe_size == WQE_LCMD_SIZE) {
+		struct hinic_cmdq_wqe_lcmd *wqe_lcmd = &wqe->wqe_lcmd;
+
+		ctrl = &wqe_lcmd->ctrl;
+	} else {
+		struct hinic_cmdq_direct_wqe *direct_wqe = &wqe->direct_wqe;
+		struct hinic_cmdq_wqe_scmd *wqe_scmd;
+
+		wqe_scmd = &direct_wqe->wqe_scmd;
+		ctrl = &wqe_scmd->ctrl;
+	}
+
+	/* clear HW busy bit */
+	ctrl->ctrl_info = 0;
+
+	wmb();  /* verify wqe is clear */
+}
+
+/**
+ * cmdq_arm_ceq_handler - cmdq completion event handler for arm command
+ * @cmdq: the cmdq of the arm command
+ * @wqe: the wqe of the arm command
+ *
+ * Return 0 - Success, negative - Failure
+ **/
+static int cmdq_arm_ceq_handler(struct hinic_cmdq *cmdq,
+				struct hinic_cmdq_wqe *wqe)
+{
+	struct hinic_cmdq_direct_wqe *direct_wqe = &wqe->direct_wqe;
+	struct hinic_cmdq_wqe_scmd *wqe_scmd;
+	struct hinic_ctrl *ctrl;
+	u32 ctrl_info;
+
+	wqe_scmd = &direct_wqe->wqe_scmd;
+	ctrl = &wqe_scmd->ctrl;
+	ctrl_info = be32_to_cpu(ctrl->ctrl_info);
+
+	/* HW should toggle the HW BUSY BIT */
+	if (!CMDQ_WQE_COMPLETED(ctrl_info))
+		return -EBUSY;
+
+	clear_wqe_complete_bit(cmdq, wqe);
+
+	hinic_put_wqe(cmdq->wq, WQE_SCMD_SIZE);
+	return 0;
+}
+
+static void cmdq_update_errcode(struct hinic_cmdq *cmdq, u16 prod_idx,
+				int errcode)
+{
+	if (cmdq->errcode[prod_idx])
+		*cmdq->errcode[prod_idx] = errcode;
+}
+
+/**
+ * cmdq_sync_cmd_handler - cmdq completion event handler for sync command
+ * @cmdq: the cmdq of the command
+ * @cons_idx: the consumer index to update the error code for
+ * @errcode: the error code
+ **/
+static void cmdq_sync_cmd_handler(struct hinic_cmdq *cmdq, u16 cons_idx,
+				  int errcode)
+{
+	u16 prod_idx = cons_idx;
+
+	spin_lock(&cmdq->cmdq_lock);
+	cmdq_update_errcode(cmdq, prod_idx, errcode);
+
+	wmb();  /* write all before update for the command request */
+
+	if (cmdq->done[prod_idx])
+		complete(cmdq->done[prod_idx]);
+	spin_unlock(&cmdq->cmdq_lock);
+}
+
+static int cmdq_cmd_ceq_handler(struct hinic_cmdq *cmdq, u16 ci,
+				struct hinic_cmdq_wqe *cmdq_wqe)
+{
+	struct hinic_cmdq_wqe_lcmd *wqe_lcmd = &cmdq_wqe->wqe_lcmd;
+	struct hinic_status *status = &wqe_lcmd->status;
+	struct hinic_ctrl *ctrl = &wqe_lcmd->ctrl;
+	int errcode;
+
+	if (!CMDQ_WQE_COMPLETED(be32_to_cpu(ctrl->ctrl_info)))
+		return -EBUSY;
+
+	dma_rmb();
+
+	errcode = CMDQ_WQE_ERRCODE_GET(be32_to_cpu(status->status_info), VAL);
+
+	cmdq_sync_cmd_handler(cmdq, ci, errcode);
+
+	clear_wqe_complete_bit(cmdq, cmdq_wqe);
+	hinic_put_wqe(cmdq->wq, WQE_LCMD_SIZE);
+	return 0;
+}
+
+/**
+ * cmdq_ceq_handler - cmdq completion event handler
+ * @handle: private data for the handler(cmdqs)
+ * @ceqe_data: ceq element data
+ **/
+static void cmdq_ceq_handler(void *handle, u32 ceqe_data)
+{
+	enum hinic_cmdq_type cmdq_type = CMDQ_CEQE_GET(ceqe_data, TYPE);
+	struct hinic_cmdqs *cmdqs = (struct hinic_cmdqs *)handle;
+	struct hinic_cmdq *cmdq = &cmdqs->cmdq[cmdq_type];
+	struct hinic_cmdq_header *header;
+	struct hinic_hw_wqe *hw_wqe;
+	int err, set_arm = 0;
+	u32 saved_data;
+	u16 ci;
+
+	/* Read the smallest wqe size for getting wqe size */
+	while ((hw_wqe = hinic_read_wqe(cmdq->wq, WQE_SCMD_SIZE, &ci))) {
+		if (IS_ERR(hw_wqe))
+			break;
+
+		header = CMDQ_WQE_HEADER(&hw_wqe->cmdq_wqe);
+		saved_data = be32_to_cpu(header->saved_data);
+
+		if (HINIC_SAVED_DATA_GET(saved_data, ARM)) {
+			/* arm_bit was set until here */
+			set_arm = 0;
+
+			if (cmdq_arm_ceq_handler(cmdq, &hw_wqe->cmdq_wqe))
+				break;
+		} else {
+			set_arm = 1;
+
+			hw_wqe = hinic_read_wqe(cmdq->wq, WQE_LCMD_SIZE, &ci);
+			if (IS_ERR(hw_wqe))
+				break;
+
+			if (cmdq_cmd_ceq_handler(cmdq, ci, &hw_wqe->cmdq_wqe))
+				break;
+		}
+	}
+
+	if (set_arm) {
+		struct hinic_hwif *hwif = cmdqs->hwif;
+		struct pci_dev *pdev = hwif->pdev;
+
+		err = hinic_set_arm_bit(cmdqs, HINIC_SET_ARM_CMDQ, cmdq_type);
+		if (err)
+			dev_err(&pdev->dev, "Failed to set arm for CMDQ\n");
+	}
+}
+
+/**
+ * cmdq_init_queue_ctxt - init the queue ctxt of a cmdq
+ * @cmdq_ctxt: cmdq ctxt to initialize
+ * @cmdq: the cmdq
+ * @cmdq_pages: the memory of the queue
+ **/
+static void cmdq_init_queue_ctxt(struct hinic_cmdq_ctxt *cmdq_ctxt,
+				 struct hinic_cmdq *cmdq,
+				 struct hinic_cmdq_pages *cmdq_pages)
+{
+	struct hinic_cmdq_ctxt_info *ctxt_info = &cmdq_ctxt->ctxt_info;
+	u64 wq_first_page_paddr, cmdq_first_block_paddr, pfn;
+	struct hinic_cmdqs *cmdqs = cmdq_to_cmdqs(cmdq);
+	struct hinic_wq *wq = cmdq->wq;
+
+	/* The data in the HW is in Big Endian Format */
+	wq_first_page_paddr = be64_to_cpu(*wq->block_vaddr);
+
+	pfn = CMDQ_PFN(wq_first_page_paddr, SZ_4K);
+
+	ctxt_info->curr_wqe_page_pfn =
+		HINIC_CMDQ_CTXT_PAGE_INFO_SET(pfn, CURR_WQE_PAGE_PFN)   |
+		HINIC_CMDQ_CTXT_PAGE_INFO_SET(HINIC_CEQ_ID_CMDQ, EQ_ID) |
+		HINIC_CMDQ_CTXT_PAGE_INFO_SET(1, CEQ_ARM)               |
+		HINIC_CMDQ_CTXT_PAGE_INFO_SET(1, CEQ_EN)                |
+		HINIC_CMDQ_CTXT_PAGE_INFO_SET(cmdq->wrapped, WRAPPED);
+
+	if (wq->num_q_pages != 1) {
+		/* block PFN - Read Modify Write */
+		cmdq_first_block_paddr = cmdq_pages->page_paddr;
+
+		pfn = CMDQ_PFN(cmdq_first_block_paddr, wq->wq_page_size);
+	}
+
+	ctxt_info->wq_block_pfn =
+		HINIC_CMDQ_CTXT_BLOCK_INFO_SET(pfn, WQ_BLOCK_PFN) |
+		HINIC_CMDQ_CTXT_BLOCK_INFO_SET(atomic_read(&wq->cons_idx), CI);
+
+	cmdq_ctxt->func_idx = HINIC_HWIF_FUNC_IDX(cmdqs->hwif);
+	cmdq_ctxt->ppf_idx = HINIC_HWIF_PPF_IDX(cmdqs->hwif);
+	cmdq_ctxt->cmdq_type  = cmdq->cmdq_type;
+}
+
+/**
+ * init_cmdq - initialize cmdq
+ * @cmdq: the cmdq
+ * @wq: the wq attaced to the cmdq
+ * @q_type: the cmdq type of the cmdq
+ * @db_area: doorbell area for the cmdq
+ *
+ * Return 0 - Success, negative - Failure
+ **/
+static int init_cmdq(struct hinic_cmdq *cmdq, struct hinic_wq *wq,
+		     enum hinic_cmdq_type q_type, void __iomem *db_area)
+{
+	int err;
+
+	cmdq->wq = wq;
+	cmdq->cmdq_type = q_type;
+	cmdq->wrapped = 1;
+
+	spin_lock_init(&cmdq->cmdq_lock);
+
+	cmdq->done = vzalloc(array_size(sizeof(*cmdq->done), wq->q_depth));
+	if (!cmdq->done)
+		return -ENOMEM;
+
+	cmdq->errcode = vzalloc(array_size(sizeof(*cmdq->errcode),
+					   wq->q_depth));
+	if (!cmdq->errcode) {
+		err = -ENOMEM;
+		goto err_errcode;
+	}
+
+	cmdq->db_base = db_area + CMDQ_DB_OFF;
+	return 0;
+
+err_errcode:
+	vfree(cmdq->done);
+	return err;
+}
+
+/**
+ * free_cmdq - Free cmdq
+ * @cmdq: the cmdq to free
+ **/
+static void free_cmdq(struct hinic_cmdq *cmdq)
+{
+	vfree(cmdq->errcode);
+	vfree(cmdq->done);
+}
+
+/**
+ * init_cmdqs_ctxt - write the cmdq ctxt to HW after init all cmdq
+ * @hwdev: the NIC HW device
+ * @cmdqs: cmdqs to write the ctxts for
+ * @db_area: db_area for all the cmdqs
+ *
+ * Return 0 - Success, negative - Failure
+ **/
+static int init_cmdqs_ctxt(struct hinic_hwdev *hwdev,
+			   struct hinic_cmdqs *cmdqs, void __iomem **db_area)
+{
+	struct hinic_hwif *hwif = hwdev->hwif;
+	enum hinic_cmdq_type type, cmdq_type;
+	struct hinic_cmdq_ctxt *cmdq_ctxts;
+	struct pci_dev *pdev = hwif->pdev;
+	struct hinic_pfhwdev *pfhwdev;
+	int err;
+
+	cmdq_ctxts = devm_kcalloc(&pdev->dev, HINIC_MAX_CMDQ_TYPES,
+				  sizeof(*cmdq_ctxts), GFP_KERNEL);
+	if (!cmdq_ctxts)
+		return -ENOMEM;
+
+	pfhwdev = container_of(hwdev, struct hinic_pfhwdev, hwdev);
+
+	cmdq_type = HINIC_CMDQ_SYNC;
+	for (; cmdq_type < HINIC_MAX_CMDQ_TYPES; cmdq_type++) {
+		cmdqs->cmdq[cmdq_type].hwdev = hwdev;
+		err = init_cmdq(&cmdqs->cmdq[cmdq_type],
+				&cmdqs->saved_wqs[cmdq_type], cmdq_type,
+				db_area[cmdq_type]);
+		if (err) {
+			dev_err(&pdev->dev, "Failed to initialize cmdq\n");
+			goto err_init_cmdq;
+		}
+
+		cmdq_init_queue_ctxt(&cmdq_ctxts[cmdq_type],
+				     &cmdqs->cmdq[cmdq_type],
+				     &cmdqs->cmdq_pages);
+	}
+
+	/* Write the CMDQ ctxts */
+	cmdq_type = HINIC_CMDQ_SYNC;
+	for (; cmdq_type < HINIC_MAX_CMDQ_TYPES; cmdq_type++) {
+		err = hinic_msg_to_mgmt(&pfhwdev->pf_to_mgmt, HINIC_MOD_COMM,
+					HINIC_COMM_CMD_CMDQ_CTXT_SET,
+					&cmdq_ctxts[cmdq_type],
+					sizeof(cmdq_ctxts[cmdq_type]),
+					NULL, NULL, HINIC_MGMT_MSG_SYNC);
+		if (err) {
+			dev_err(&pdev->dev, "Failed to set CMDQ CTXT type = %d\n",
+				cmdq_type);
+			goto err_write_cmdq_ctxt;
+		}
+	}
+
+	devm_kfree(&pdev->dev, cmdq_ctxts);
+	return 0;
+
+err_write_cmdq_ctxt:
+	cmdq_type = HINIC_MAX_CMDQ_TYPES;
+
+err_init_cmdq:
+	for (type = HINIC_CMDQ_SYNC; type < cmdq_type; type++)
+		free_cmdq(&cmdqs->cmdq[type]);
+
+	devm_kfree(&pdev->dev, cmdq_ctxts);
+	return err;
+}
+
+static int hinic_set_cmdq_depth(struct hinic_hwdev *hwdev, u16 cmdq_depth)
+{
+	struct hinic_cmd_hw_ioctxt hw_ioctxt = { 0 };
+	struct hinic_pfhwdev *pfhwdev;
+
+	pfhwdev = container_of(hwdev, struct hinic_pfhwdev, hwdev);
+
+	hw_ioctxt.func_idx = HINIC_HWIF_FUNC_IDX(hwdev->hwif);
+	hw_ioctxt.ppf_idx = HINIC_HWIF_PPF_IDX(hwdev->hwif);
+
+	hw_ioctxt.set_cmdq_depth = HW_IOCTXT_SET_CMDQ_DEPTH_ENABLE;
+	hw_ioctxt.cmdq_depth = (u8)ilog2(cmdq_depth);
+
+	return hinic_msg_to_mgmt(&pfhwdev->pf_to_mgmt, HINIC_MOD_COMM,
+				 HINIC_COMM_CMD_HWCTXT_SET,
+				 &hw_ioctxt, sizeof(hw_ioctxt), NULL,
+				 NULL, HINIC_MGMT_MSG_SYNC);
+}
+
+/**
+ * hinic_init_cmdqs - init all cmdqs
+ * @cmdqs: cmdqs to init
+ * @hwif: HW interface for accessing cmdqs
+ * @db_area: doorbell areas for all the cmdqs
+ *
+ * Return 0 - Success, negative - Failure
+ **/
+int hinic_init_cmdqs(struct hinic_cmdqs *cmdqs, struct hinic_hwif *hwif,
+		     void __iomem **db_area)
+{
+	struct hinic_func_to_io *func_to_io = cmdqs_to_func_to_io(cmdqs);
+	struct pci_dev *pdev = hwif->pdev;
+	struct hinic_hwdev *hwdev;
+	u16 max_wqe_size;
+	int err;
+
+	cmdqs->hwif = hwif;
+	cmdqs->cmdq_buf_pool = dma_pool_create("hinic_cmdq", &pdev->dev,
+					       HINIC_CMDQ_BUF_SIZE,
+					       HINIC_CMDQ_BUF_SIZE, 0);
+	if (!cmdqs->cmdq_buf_pool)
+		return -ENOMEM;
+
+	cmdqs->saved_wqs = devm_kcalloc(&pdev->dev, HINIC_MAX_CMDQ_TYPES,
+					sizeof(*cmdqs->saved_wqs), GFP_KERNEL);
+	if (!cmdqs->saved_wqs) {
+		err = -ENOMEM;
+		goto err_saved_wqs;
+	}
+
+	max_wqe_size = WQE_LCMD_SIZE;
+	err = hinic_wqs_cmdq_alloc(&cmdqs->cmdq_pages, cmdqs->saved_wqs, hwif,
+				   HINIC_MAX_CMDQ_TYPES, CMDQ_WQEBB_SIZE,
+				   CMDQ_WQ_PAGE_SIZE, CMDQ_DEPTH, max_wqe_size);
+	if (err) {
+		dev_err(&pdev->dev, "Failed to allocate CMDQ wqs\n");
+		goto err_cmdq_wqs;
+	}
+
+	hwdev = container_of(func_to_io, struct hinic_hwdev, func_to_io);
+	err = init_cmdqs_ctxt(hwdev, cmdqs, db_area);
+	if (err) {
+		dev_err(&pdev->dev, "Failed to write cmdq ctxt\n");
+		goto err_cmdq_ctxt;
+	}
+
+	hinic_ceq_register_cb(&func_to_io->ceqs, HINIC_CEQ_CMDQ, cmdqs,
+			      cmdq_ceq_handler);
+
+	err = hinic_set_cmdq_depth(hwdev, CMDQ_DEPTH);
+	if (err) {
+		dev_err(&hwif->pdev->dev, "Failed to set cmdq depth\n");
+		goto err_set_cmdq_depth;
+	}
+
+	return 0;
+
+err_set_cmdq_depth:
+	hinic_ceq_unregister_cb(&func_to_io->ceqs, HINIC_CEQ_CMDQ);
+	free_cmdq(&cmdqs->cmdq[HINIC_CMDQ_SYNC]);
+err_cmdq_ctxt:
+	hinic_wqs_cmdq_free(&cmdqs->cmdq_pages, cmdqs->saved_wqs,
+			    HINIC_MAX_CMDQ_TYPES);
+
+err_cmdq_wqs:
+	devm_kfree(&pdev->dev, cmdqs->saved_wqs);
+
+err_saved_wqs:
+	dma_pool_destroy(cmdqs->cmdq_buf_pool);
+	return err;
+}
+
+/**
+ * hinic_free_cmdqs - free all cmdqs
+ * @cmdqs: cmdqs to free
+ **/
+void hinic_free_cmdqs(struct hinic_cmdqs *cmdqs)
+{
+	struct hinic_func_to_io *func_to_io = cmdqs_to_func_to_io(cmdqs);
+	struct hinic_hwif *hwif = cmdqs->hwif;
+	struct pci_dev *pdev = hwif->pdev;
+	enum hinic_cmdq_type cmdq_type;
+
+	hinic_ceq_unregister_cb(&func_to_io->ceqs, HINIC_CEQ_CMDQ);
+
+	cmdq_type = HINIC_CMDQ_SYNC;
+	for (; cmdq_type < HINIC_MAX_CMDQ_TYPES; cmdq_type++)
+		free_cmdq(&cmdqs->cmdq[cmdq_type]);
+
+	hinic_wqs_cmdq_free(&cmdqs->cmdq_pages, cmdqs->saved_wqs,
+			    HINIC_MAX_CMDQ_TYPES);
+
+	devm_kfree(&pdev->dev, cmdqs->saved_wqs);
+
+	dma_pool_destroy(cmdqs->cmdq_buf_pool);
+}
diff --git a/drivers/net/ethernet/huawei/hinic/hinic_hw_cmdq.h b/drivers/net/ethernet/huawei/hinic/hinic_hw_cmdq.h
new file mode 100644
index 000000000000..ff09cf0ed52b
--- /dev/null
+++ b/drivers/net/ethernet/huawei/hinic/hinic_hw_cmdq.h
@@ -0,0 +1,185 @@
+/* SPDX-License-Identifier: GPL-2.0-only */
+/*
+ * Huawei HiNIC PCI Express Linux driver
+ * Copyright(c) 2017 Huawei Technologies Co., Ltd
+ */
+
+#ifndef HINIC_CMDQ_H
+#define HINIC_CMDQ_H
+
+#include <linux/types.h>
+#include <linux/spinlock.h>
+#include <linux/completion.h>
+#include <linux/pci.h>
+
+#include "hinic_hw_if.h"
+#include "hinic_hw_wq.h"
+
+#define HINIC_CMDQ_CTXT_CURR_WQE_PAGE_PFN_SHIFT         0
+#define HINIC_CMDQ_CTXT_EQ_ID_SHIFT                     56
+#define HINIC_CMDQ_CTXT_CEQ_ARM_SHIFT                   61
+#define HINIC_CMDQ_CTXT_CEQ_EN_SHIFT                    62
+#define HINIC_CMDQ_CTXT_WRAPPED_SHIFT                   63
+
+#define HINIC_CMDQ_CTXT_CURR_WQE_PAGE_PFN_MASK          0xFFFFFFFFFFFFF
+#define HINIC_CMDQ_CTXT_EQ_ID_MASK                      0x1F
+#define HINIC_CMDQ_CTXT_CEQ_ARM_MASK                    0x1
+#define HINIC_CMDQ_CTXT_CEQ_EN_MASK                     0x1
+#define HINIC_CMDQ_CTXT_WRAPPED_MASK                    0x1
+
+#define HINIC_CMDQ_CTXT_PAGE_INFO_SET(val, member)      \
+			(((u64)(val) & HINIC_CMDQ_CTXT_##member##_MASK) \
+			 << HINIC_CMDQ_CTXT_##member##_SHIFT)
+
+#define HINIC_CMDQ_CTXT_PAGE_INFO_GET(val, member)	\
+			(((u64)(val) >> HINIC_CMDQ_CTXT_##member##_SHIFT) \
+			 & HINIC_CMDQ_CTXT_##member##_MASK)
+
+#define HINIC_CMDQ_CTXT_PAGE_INFO_CLEAR(val, member)    \
+			((val) & (~((u64)HINIC_CMDQ_CTXT_##member##_MASK \
+			 << HINIC_CMDQ_CTXT_##member##_SHIFT)))
+
+#define HINIC_CMDQ_CTXT_WQ_BLOCK_PFN_SHIFT              0
+#define HINIC_CMDQ_CTXT_CI_SHIFT                        52
+
+#define HINIC_CMDQ_CTXT_WQ_BLOCK_PFN_MASK               0xFFFFFFFFFFFFF
+#define HINIC_CMDQ_CTXT_CI_MASK                         0xFFF
+
+#define HINIC_CMDQ_CTXT_BLOCK_INFO_SET(val, member)     \
+			(((u64)(val) & HINIC_CMDQ_CTXT_##member##_MASK) \
+			 << HINIC_CMDQ_CTXT_##member##_SHIFT)
+
+#define HINIC_CMDQ_CTXT_BLOCK_INFO_GET(val, member)	\
+			(((u64)(val) >> HINIC_CMDQ_CTXT_##member##_SHIFT) \
+			& HINIC_CMDQ_CTXT_##member##_MASK)
+
+#define HINIC_CMDQ_CTXT_BLOCK_INFO_CLEAR(val, member)   \
+			((val) & (~((u64)HINIC_CMDQ_CTXT_##member##_MASK \
+			 << HINIC_CMDQ_CTXT_##member##_SHIFT)))
+
+#define HINIC_SAVED_DATA_ARM_SHIFT                      31
+
+#define HINIC_SAVED_DATA_ARM_MASK                       0x1
+
+#define HINIC_SAVED_DATA_SET(val, member)               \
+			(((u32)(val) & HINIC_SAVED_DATA_##member##_MASK) \
+			 << HINIC_SAVED_DATA_##member##_SHIFT)
+
+#define HINIC_SAVED_DATA_GET(val, member)               \
+			(((val) >> HINIC_SAVED_DATA_##member##_SHIFT) \
+			 & HINIC_SAVED_DATA_##member##_MASK)
+
+#define HINIC_SAVED_DATA_CLEAR(val, member)             \
+			((val) & (~(HINIC_SAVED_DATA_##member##_MASK \
+			 << HINIC_SAVED_DATA_##member##_SHIFT)))
+
+#define HINIC_CMDQ_DB_INFO_HI_PROD_IDX_SHIFT            0
+#define HINIC_CMDQ_DB_INFO_PATH_SHIFT                   23
+#define HINIC_CMDQ_DB_INFO_CMDQ_TYPE_SHIFT              24
+#define HINIC_CMDQ_DB_INFO_DB_TYPE_SHIFT                27
+
+#define HINIC_CMDQ_DB_INFO_HI_PROD_IDX_MASK             0xFF
+#define HINIC_CMDQ_DB_INFO_PATH_MASK                    0x1
+#define HINIC_CMDQ_DB_INFO_CMDQ_TYPE_MASK               0x7
+#define HINIC_CMDQ_DB_INFO_DB_TYPE_MASK                 0x1F
+
+#define HINIC_CMDQ_DB_INFO_SET(val, member)             \
+			(((u32)(val) & HINIC_CMDQ_DB_INFO_##member##_MASK) \
+			 << HINIC_CMDQ_DB_INFO_##member##_SHIFT)
+
+#define HINIC_CMDQ_BUF_SIZE             2048
+
+#define HINIC_CMDQ_BUF_HW_RSVD          8
+#define HINIC_CMDQ_MAX_DATA_SIZE        (HINIC_CMDQ_BUF_SIZE - \
+					 HINIC_CMDQ_BUF_HW_RSVD)
+
+enum hinic_cmdq_type {
+	HINIC_CMDQ_SYNC,
+
+	HINIC_MAX_CMDQ_TYPES,
+};
+
+enum hinic_set_arm_qtype {
+	HINIC_SET_ARM_CMDQ,
+};
+
+enum hinic_cmd_ack_type {
+	HINIC_CMD_ACK_TYPE_CMDQ,
+};
+
+struct hinic_cmdq_buf {
+	void            *buf;
+	dma_addr_t      dma_addr;
+	size_t          size;
+};
+
+struct hinic_cmdq_arm_bit {
+	u32     q_type;
+	u32     q_id;
+};
+
+struct hinic_cmdq_ctxt_info {
+	u64     curr_wqe_page_pfn;
+	u64     wq_block_pfn;
+};
+
+struct hinic_cmdq_ctxt {
+	u8      status;
+	u8      version;
+	u8      rsvd0[6];
+
+	u16     func_idx;
+	u8      cmdq_type;
+	u8      ppf_idx;
+
+	u8      rsvd2[4];
+
+	struct hinic_cmdq_ctxt_info ctxt_info;
+};
+
+struct hinic_cmdq {
+	struct hinic_hwdev      *hwdev;
+
+	struct hinic_wq         *wq;
+
+	enum hinic_cmdq_type    cmdq_type;
+	int                     wrapped;
+
+	/* Lock for keeping the doorbell order */
+	spinlock_t              cmdq_lock;
+
+	struct completion       **done;
+	int                     **errcode;
+
+	/* doorbell area */
+	void __iomem            *db_base;
+};
+
+struct hinic_cmdqs {
+	struct hinic_hwif       *hwif;
+
+	struct dma_pool         *cmdq_buf_pool;
+
+	struct hinic_wq         *saved_wqs;
+
+	struct hinic_cmdq_pages cmdq_pages;
+
+	struct hinic_cmdq       cmdq[HINIC_MAX_CMDQ_TYPES];
+};
+
+int hinic_alloc_cmdq_buf(struct hinic_cmdqs *cmdqs,
+			 struct hinic_cmdq_buf *cmdq_buf);
+
+void hinic_free_cmdq_buf(struct hinic_cmdqs *cmdqs,
+			 struct hinic_cmdq_buf *cmdq_buf);
+
+int hinic_cmdq_direct_resp(struct hinic_cmdqs *cmdqs,
+			   enum hinic_mod_type mod, u8 cmd,
+			   struct hinic_cmdq_buf *buf_in, u64 *out_param);
+
+int hinic_init_cmdqs(struct hinic_cmdqs *cmdqs, struct hinic_hwif *hwif,
+		     void __iomem **db_area);
+
+void hinic_free_cmdqs(struct hinic_cmdqs *cmdqs);
+
+#endif
diff --git a/drivers/net/ethernet/huawei/hinic/hinic_hw_csr.h b/drivers/net/ethernet/huawei/hinic/hinic_hw_csr.h
new file mode 100644
index 000000000000..d56e7413ace0
--- /dev/null
+++ b/drivers/net/ethernet/huawei/hinic/hinic_hw_csr.h
@@ -0,0 +1,139 @@
+/* SPDX-License-Identifier: GPL-2.0-only */
+/*
+ * Huawei HiNIC PCI Express Linux driver
+ * Copyright(c) 2017 Huawei Technologies Co., Ltd
+ */
+
+#ifndef HINIC_HW_CSR_H
+#define HINIC_HW_CSR_H
+
+/* HW interface registers */
+#define HINIC_CSR_FUNC_ATTR0_ADDR                       0x0
+#define HINIC_CSR_FUNC_ATTR1_ADDR                       0x4
+#define HINIC_CSR_FUNC_ATTR2_ADDR			0x8
+#define HINIC_CSR_FUNC_ATTR4_ADDR                       0x10
+#define HINIC_CSR_FUNC_ATTR5_ADDR                       0x14
+
+#define HINIC_DMA_ATTR_BASE                             0xC80
+#define HINIC_ELECTION_BASE                             0x4200
+
+#define HINIC_DMA_ATTR_STRIDE                           0x4
+#define HINIC_CSR_DMA_ATTR_ADDR(idx)                    \
+	(HINIC_DMA_ATTR_BASE + (idx) * HINIC_DMA_ATTR_STRIDE)
+
+#define HINIC_PPF_ELECTION_STRIDE                       0x4
+
+#define HINIC_CSR_PPF_ELECTION_ADDR(idx)                \
+	(HINIC_ELECTION_BASE +  (idx) * HINIC_PPF_ELECTION_STRIDE)
+
+/* API CMD registers */
+#define HINIC_CSR_API_CMD_BASE                          0xF000
+
+#define HINIC_CSR_API_CMD_STRIDE                        0x100
+
+#define HINIC_CSR_API_CMD_CHAIN_HEAD_HI_ADDR(idx)       \
+	(HINIC_CSR_API_CMD_BASE + 0x0 + (idx) * HINIC_CSR_API_CMD_STRIDE)
+
+#define HINIC_CSR_API_CMD_CHAIN_HEAD_LO_ADDR(idx)       \
+	(HINIC_CSR_API_CMD_BASE + 0x4 + (idx) * HINIC_CSR_API_CMD_STRIDE)
+
+#define HINIC_CSR_API_CMD_STATUS_HI_ADDR(idx)           \
+	(HINIC_CSR_API_CMD_BASE + 0x8 + (idx) * HINIC_CSR_API_CMD_STRIDE)
+
+#define HINIC_CSR_API_CMD_STATUS_LO_ADDR(idx)           \
+	(HINIC_CSR_API_CMD_BASE + 0xC + (idx) * HINIC_CSR_API_CMD_STRIDE)
+
+#define HINIC_CSR_API_CMD_CHAIN_NUM_CELLS_ADDR(idx)     \
+	(HINIC_CSR_API_CMD_BASE + 0x10 + (idx) * HINIC_CSR_API_CMD_STRIDE)
+
+#define HINIC_CSR_API_CMD_CHAIN_CTRL_ADDR(idx)          \
+	(HINIC_CSR_API_CMD_BASE + 0x14 + (idx) * HINIC_CSR_API_CMD_STRIDE)
+
+#define HINIC_CSR_API_CMD_CHAIN_PI_ADDR(idx)            \
+	(HINIC_CSR_API_CMD_BASE + 0x1C + (idx) * HINIC_CSR_API_CMD_STRIDE)
+
+#define HINIC_CSR_API_CMD_CHAIN_REQ_ADDR(idx)           \
+	(HINIC_CSR_API_CMD_BASE + 0x20 + (idx) * HINIC_CSR_API_CMD_STRIDE)
+
+#define HINIC_CSR_API_CMD_STATUS_ADDR(idx)              \
+	(HINIC_CSR_API_CMD_BASE + 0x30 + (idx) * HINIC_CSR_API_CMD_STRIDE)
+
+/* MSI-X registers */
+#define HINIC_CSR_MSIX_CTRL_BASE                        0x2000
+#define HINIC_CSR_MSIX_CNT_BASE                         0x2004
+
+#define HINIC_CSR_MSIX_STRIDE                           0x8
+
+#define HINIC_CSR_MSIX_CTRL_ADDR(idx)                   \
+	(HINIC_CSR_MSIX_CTRL_BASE + (idx) * HINIC_CSR_MSIX_STRIDE)
+
+#define HINIC_CSR_MSIX_CNT_ADDR(idx)                    \
+	(HINIC_CSR_MSIX_CNT_BASE + (idx) * HINIC_CSR_MSIX_STRIDE)
+
+/* EQ registers */
+#define HINIC_AEQ_MTT_OFF_BASE_ADDR                     0x200
+#define HINIC_CEQ_MTT_OFF_BASE_ADDR                     0x400
+
+#define HINIC_EQ_MTT_OFF_STRIDE                         0x40
+
+#define HINIC_CSR_AEQ_MTT_OFF(id)                       \
+	(HINIC_AEQ_MTT_OFF_BASE_ADDR + (id) * HINIC_EQ_MTT_OFF_STRIDE)
+
+#define HINIC_CSR_CEQ_MTT_OFF(id)                       \
+	(HINIC_CEQ_MTT_OFF_BASE_ADDR + (id) * HINIC_EQ_MTT_OFF_STRIDE)
+
+#define HINIC_CSR_EQ_PAGE_OFF_STRIDE                    8
+
+#define HINIC_CSR_AEQ_HI_PHYS_ADDR_REG(q_id, pg_num)    \
+	(HINIC_CSR_AEQ_MTT_OFF(q_id) + \
+	 (pg_num) * HINIC_CSR_EQ_PAGE_OFF_STRIDE)
+
+#define HINIC_CSR_CEQ_HI_PHYS_ADDR_REG(q_id, pg_num)    \
+	(HINIC_CSR_CEQ_MTT_OFF(q_id) +          \
+	 (pg_num) * HINIC_CSR_EQ_PAGE_OFF_STRIDE)
+
+#define HINIC_CSR_AEQ_LO_PHYS_ADDR_REG(q_id, pg_num)    \
+	(HINIC_CSR_AEQ_MTT_OFF(q_id) + \
+	 (pg_num) * HINIC_CSR_EQ_PAGE_OFF_STRIDE + 4)
+
+#define HINIC_CSR_CEQ_LO_PHYS_ADDR_REG(q_id, pg_num)    \
+	(HINIC_CSR_CEQ_MTT_OFF(q_id) +  \
+	 (pg_num) * HINIC_CSR_EQ_PAGE_OFF_STRIDE + 4)
+
+#define HINIC_AEQ_CTRL_0_ADDR_BASE                      0xE00
+#define HINIC_AEQ_CTRL_1_ADDR_BASE                      0xE04
+#define HINIC_AEQ_CONS_IDX_ADDR_BASE                    0xE08
+#define HINIC_AEQ_PROD_IDX_ADDR_BASE                    0xE0C
+
+#define HINIC_CEQ_CTRL_0_ADDR_BASE                      0x1000
+#define HINIC_CEQ_CTRL_1_ADDR_BASE                      0x1004
+#define HINIC_CEQ_CONS_IDX_ADDR_BASE                    0x1008
+#define HINIC_CEQ_PROD_IDX_ADDR_BASE                    0x100C
+
+#define HINIC_EQ_OFF_STRIDE                             0x80
+
+#define HINIC_CSR_AEQ_CTRL_0_ADDR(idx)                  \
+	(HINIC_AEQ_CTRL_0_ADDR_BASE + (idx) * HINIC_EQ_OFF_STRIDE)
+
+#define HINIC_CSR_AEQ_CTRL_1_ADDR(idx)                  \
+	(HINIC_AEQ_CTRL_1_ADDR_BASE + (idx) * HINIC_EQ_OFF_STRIDE)
+
+#define HINIC_CSR_AEQ_CONS_IDX_ADDR(idx)                \
+	(HINIC_AEQ_CONS_IDX_ADDR_BASE + (idx) * HINIC_EQ_OFF_STRIDE)
+
+#define HINIC_CSR_AEQ_PROD_IDX_ADDR(idx)                \
+	(HINIC_AEQ_PROD_IDX_ADDR_BASE + (idx) * HINIC_EQ_OFF_STRIDE)
+
+#define HINIC_CSR_CEQ_CTRL_0_ADDR(idx)                  \
+	(HINIC_CEQ_CTRL_0_ADDR_BASE + (idx) * HINIC_EQ_OFF_STRIDE)
+
+#define HINIC_CSR_CEQ_CTRL_1_ADDR(idx)                  \
+	(HINIC_CEQ_CTRL_1_ADDR_BASE + (idx) * HINIC_EQ_OFF_STRIDE)
+
+#define HINIC_CSR_CEQ_CONS_IDX_ADDR(idx)                \
+	(HINIC_CEQ_CONS_IDX_ADDR_BASE + (idx) * HINIC_EQ_OFF_STRIDE)
+
+#define HINIC_CSR_CEQ_PROD_IDX_ADDR(idx)                \
+	(HINIC_CEQ_PROD_IDX_ADDR_BASE + (idx) * HINIC_EQ_OFF_STRIDE)
+
+#endif
diff --git a/drivers/net/ethernet/huawei/hinic/hinic_hw_dev.c b/drivers/net/ethernet/huawei/hinic/hinic_hw_dev.c
new file mode 100644
index 000000000000..27795288c586
--- /dev/null
+++ b/drivers/net/ethernet/huawei/hinic/hinic_hw_dev.c
@@ -0,0 +1,1197 @@
+// SPDX-License-Identifier: GPL-2.0-only
+/*
+ * Huawei HiNIC PCI Express Linux driver
+ * Copyright(c) 2017 Huawei Technologies Co., Ltd
+ */
+
+#include <linux/kernel.h>
+#include <linux/types.h>
+#include <linux/pci.h>
+#include <linux/device.h>
+#include <linux/errno.h>
+#include <linux/slab.h>
+#include <linux/bitops.h>
+#include <linux/delay.h>
+#include <linux/jiffies.h>
+#include <linux/log2.h>
+#include <linux/err.h>
+#include <linux/netdevice.h>
+#include <net/devlink.h>
+
+#include "hinic_devlink.h"
+#include "hinic_sriov.h"
+#include "hinic_dev.h"
+#include "hinic_hw_if.h"
+#include "hinic_hw_eqs.h"
+#include "hinic_hw_mgmt.h"
+#include "hinic_hw_qp_ctxt.h"
+#include "hinic_hw_qp.h"
+#include "hinic_hw_io.h"
+#include "hinic_hw_dev.h"
+
+#define OUTBOUND_STATE_TIMEOUT          100
+#define DB_STATE_TIMEOUT                100
+
+#define MAX_IRQS(max_qps, num_aeqs, num_ceqs)   \
+		 (2 * (max_qps) + (num_aeqs) + (num_ceqs))
+
+#define ADDR_IN_4BYTES(addr)            ((addr) >> 2)
+
+enum intr_type {
+	INTR_MSIX_TYPE,
+};
+
+/**
+ * parse_capability - convert device capabilities to NIC capabilities
+ * @hwdev: the HW device to set and convert device capabilities for
+ * @dev_cap: device capabilities from FW
+ *
+ * Return 0 - Success, negative - Failure
+ **/
+static int parse_capability(struct hinic_hwdev *hwdev,
+			    struct hinic_dev_cap *dev_cap)
+{
+	struct hinic_cap *nic_cap = &hwdev->nic_cap;
+	int num_aeqs, num_ceqs, num_irqs;
+
+	if (!HINIC_IS_VF(hwdev->hwif) && dev_cap->intr_type != INTR_MSIX_TYPE)
+		return -EFAULT;
+
+	num_aeqs = HINIC_HWIF_NUM_AEQS(hwdev->hwif);
+	num_ceqs = HINIC_HWIF_NUM_CEQS(hwdev->hwif);
+	num_irqs = HINIC_HWIF_NUM_IRQS(hwdev->hwif);
+
+	/* Each QP has its own (SQ + RQ) interrupts */
+	nic_cap->num_qps = (num_irqs - (num_aeqs + num_ceqs)) / 2;
+
+	if (nic_cap->num_qps > HINIC_Q_CTXT_MAX)
+		nic_cap->num_qps = HINIC_Q_CTXT_MAX;
+
+	if (!HINIC_IS_VF(hwdev->hwif))
+		nic_cap->max_qps = dev_cap->max_sqs + 1;
+	else
+		nic_cap->max_qps = dev_cap->max_sqs;
+
+	if (nic_cap->num_qps > nic_cap->max_qps)
+		nic_cap->num_qps = nic_cap->max_qps;
+
+	if (!HINIC_IS_VF(hwdev->hwif)) {
+		nic_cap->max_vf = dev_cap->max_vf;
+		nic_cap->max_vf_qps = dev_cap->max_vf_sqs + 1;
+	}
+
+	hwdev->port_id = dev_cap->port_id;
+
+	return 0;
+}
+
+/**
+ * get_capability - get device capabilities from FW
+ * @pfhwdev: the PF HW device to get capabilities for
+ *
+ * Return 0 - Success, negative - Failure
+ **/
+static int get_capability(struct hinic_pfhwdev *pfhwdev)
+{
+	struct hinic_hwdev *hwdev = &pfhwdev->hwdev;
+	struct hinic_hwif *hwif = hwdev->hwif;
+	struct pci_dev *pdev = hwif->pdev;
+	struct hinic_dev_cap dev_cap;
+	u16 out_len;
+	int err;
+
+	out_len = sizeof(dev_cap);
+
+	err = hinic_msg_to_mgmt(&pfhwdev->pf_to_mgmt, HINIC_MOD_CFGM,
+				HINIC_CFG_NIC_CAP, &dev_cap, sizeof(dev_cap),
+				&dev_cap, &out_len, HINIC_MGMT_MSG_SYNC);
+	if (err) {
+		dev_err(&pdev->dev, "Failed to get capability from FW\n");
+		return err;
+	}
+
+	return parse_capability(hwdev, &dev_cap);
+}
+
+/**
+ * get_dev_cap - get device capabilities
+ * @hwdev: the NIC HW device to get capabilities for
+ *
+ * Return 0 - Success, negative - Failure
+ **/
+static int get_dev_cap(struct hinic_hwdev *hwdev)
+{
+	struct hinic_hwif *hwif = hwdev->hwif;
+	struct pci_dev *pdev = hwif->pdev;
+	struct hinic_pfhwdev *pfhwdev;
+	int err;
+
+	switch (HINIC_FUNC_TYPE(hwif)) {
+	case HINIC_PPF:
+	case HINIC_PF:
+	case HINIC_VF:
+		pfhwdev = container_of(hwdev, struct hinic_pfhwdev, hwdev);
+		err = get_capability(pfhwdev);
+		if (err) {
+			dev_err(&pdev->dev, "Failed to get capability\n");
+			return err;
+		}
+		break;
+	default:
+		dev_err(&pdev->dev, "Unsupported PCI Function type\n");
+		return -EINVAL;
+	}
+
+	return 0;
+}
+
+/**
+ * init_msix - enable the msix and save the entries
+ * @hwdev: the NIC HW device
+ *
+ * Return 0 - Success, negative - Failure
+ **/
+static int init_msix(struct hinic_hwdev *hwdev)
+{
+	struct hinic_hwif *hwif = hwdev->hwif;
+	struct pci_dev *pdev = hwif->pdev;
+	int nr_irqs, num_aeqs, num_ceqs;
+	int i, err;
+
+	num_aeqs = HINIC_HWIF_NUM_AEQS(hwif);
+	num_ceqs = HINIC_HWIF_NUM_CEQS(hwif);
+	nr_irqs = MAX_IRQS(HINIC_MAX_QPS, num_aeqs, num_ceqs);
+	if (nr_irqs > HINIC_HWIF_NUM_IRQS(hwif))
+		nr_irqs = HINIC_HWIF_NUM_IRQS(hwif);
+
+	hwdev->msix_entries = devm_kcalloc(&pdev->dev, nr_irqs,
+					   sizeof(*hwdev->msix_entries),
+					   GFP_KERNEL);
+	if (!hwdev->msix_entries)
+		return -ENOMEM;
+
+	for (i = 0; i < nr_irqs; i++)
+		hwdev->msix_entries[i].entry = i;
+
+	err = pci_enable_msix_exact(pdev, hwdev->msix_entries, nr_irqs);
+	if (err) {
+		dev_err(&pdev->dev, "Failed to enable pci msix\n");
+		return err;
+	}
+
+	return 0;
+}
+
+/**
+ * disable_msix - disable the msix
+ * @hwdev: the NIC HW device
+ **/
+static void disable_msix(struct hinic_hwdev *hwdev)
+{
+	struct hinic_hwif *hwif = hwdev->hwif;
+	struct pci_dev *pdev = hwif->pdev;
+
+	pci_disable_msix(pdev);
+}
+
+/**
+ * hinic_port_msg_cmd - send port msg to mgmt
+ * @hwdev: the NIC HW device
+ * @cmd: the port command
+ * @buf_in: input buffer
+ * @in_size: input size
+ * @buf_out: output buffer
+ * @out_size: returned output size
+ *
+ * Return 0 - Success, negative - Failure
+ **/
+int hinic_port_msg_cmd(struct hinic_hwdev *hwdev, enum hinic_port_cmd cmd,
+		       void *buf_in, u16 in_size, void *buf_out, u16 *out_size)
+{
+	struct hinic_pfhwdev *pfhwdev;
+
+	pfhwdev = container_of(hwdev, struct hinic_pfhwdev, hwdev);
+
+	return hinic_msg_to_mgmt(&pfhwdev->pf_to_mgmt, HINIC_MOD_L2NIC, cmd,
+				 buf_in, in_size, buf_out, out_size,
+				 HINIC_MGMT_MSG_SYNC);
+}
+
+int hinic_hilink_msg_cmd(struct hinic_hwdev *hwdev, enum hinic_hilink_cmd cmd,
+			 void *buf_in, u16 in_size, void *buf_out,
+			 u16 *out_size)
+{
+	struct hinic_pfhwdev *pfhwdev;
+
+	pfhwdev = container_of(hwdev, struct hinic_pfhwdev, hwdev);
+
+	return hinic_msg_to_mgmt(&pfhwdev->pf_to_mgmt, HINIC_MOD_HILINK, cmd,
+				 buf_in, in_size, buf_out, out_size,
+				 HINIC_MGMT_MSG_SYNC);
+}
+
+/**
+ * init_fw_ctxt- Init Firmware tables before network mgmt and io operations
+ * @hwdev: the NIC HW device
+ *
+ * Return 0 - Success, negative - Failure
+ **/
+static int init_fw_ctxt(struct hinic_hwdev *hwdev)
+{
+	struct hinic_hwif *hwif = hwdev->hwif;
+	struct pci_dev *pdev = hwif->pdev;
+	struct hinic_cmd_fw_ctxt fw_ctxt;
+	u16 out_size = sizeof(fw_ctxt);
+	int err;
+
+	fw_ctxt.func_idx = HINIC_HWIF_FUNC_IDX(hwif);
+	fw_ctxt.rx_buf_sz = HINIC_RX_BUF_SZ;
+
+	err = hinic_port_msg_cmd(hwdev, HINIC_PORT_CMD_FWCTXT_INIT,
+				 &fw_ctxt, sizeof(fw_ctxt),
+				 &fw_ctxt, &out_size);
+	if (err || out_size != sizeof(fw_ctxt) || fw_ctxt.status) {
+		dev_err(&pdev->dev, "Failed to init FW ctxt, err: %d, status: 0x%x, out size: 0x%x\n",
+			err, fw_ctxt.status, out_size);
+		return -EIO;
+	}
+
+	return 0;
+}
+
+/**
+ * set_hw_ioctxt - set the shape of the IO queues in FW
+ * @hwdev: the NIC HW device
+ * @rq_depth: rq depth
+ * @sq_depth: sq depth
+ *
+ * Return 0 - Success, negative - Failure
+ **/
+static int set_hw_ioctxt(struct hinic_hwdev *hwdev, unsigned int sq_depth,
+			 unsigned int rq_depth)
+{
+	struct hinic_hwif *hwif = hwdev->hwif;
+	struct hinic_cmd_hw_ioctxt hw_ioctxt;
+	struct hinic_pfhwdev *pfhwdev;
+
+	hw_ioctxt.func_idx = HINIC_HWIF_FUNC_IDX(hwif);
+	hw_ioctxt.ppf_idx = HINIC_HWIF_PPF_IDX(hwif);
+
+	hw_ioctxt.set_cmdq_depth = HW_IOCTXT_SET_CMDQ_DEPTH_DEFAULT;
+	hw_ioctxt.cmdq_depth = 0;
+
+	hw_ioctxt.lro_en = 1;
+
+	hw_ioctxt.rq_depth  = ilog2(rq_depth);
+
+	hw_ioctxt.rx_buf_sz_idx = HINIC_RX_BUF_SZ_IDX;
+
+	hw_ioctxt.sq_depth  = ilog2(sq_depth);
+
+	pfhwdev = container_of(hwdev, struct hinic_pfhwdev, hwdev);
+
+	return hinic_msg_to_mgmt(&pfhwdev->pf_to_mgmt, HINIC_MOD_COMM,
+				 HINIC_COMM_CMD_HWCTXT_SET,
+				 &hw_ioctxt, sizeof(hw_ioctxt), NULL,
+				 NULL, HINIC_MGMT_MSG_SYNC);
+}
+
+static int wait_for_outbound_state(struct hinic_hwdev *hwdev)
+{
+	enum hinic_outbound_state outbound_state;
+	struct hinic_hwif *hwif = hwdev->hwif;
+	struct pci_dev *pdev = hwif->pdev;
+	unsigned long end;
+
+	end = jiffies + msecs_to_jiffies(OUTBOUND_STATE_TIMEOUT);
+	do {
+		outbound_state = hinic_outbound_state_get(hwif);
+
+		if (outbound_state == HINIC_OUTBOUND_ENABLE)
+			return 0;
+
+		msleep(20);
+	} while (time_before(jiffies, end));
+
+	dev_err(&pdev->dev, "Wait for OUTBOUND - Timeout\n");
+	return -EFAULT;
+}
+
+static int wait_for_db_state(struct hinic_hwdev *hwdev)
+{
+	struct hinic_hwif *hwif = hwdev->hwif;
+	struct pci_dev *pdev = hwif->pdev;
+	enum hinic_db_state db_state;
+	unsigned long end;
+
+	end = jiffies + msecs_to_jiffies(DB_STATE_TIMEOUT);
+	do {
+		db_state = hinic_db_state_get(hwif);
+
+		if (db_state == HINIC_DB_ENABLE)
+			return 0;
+
+		msleep(20);
+	} while (time_before(jiffies, end));
+
+	dev_err(&pdev->dev, "Wait for DB - Timeout\n");
+	return -EFAULT;
+}
+
+/**
+ * clear_io_resources - set the IO resources as not active in the NIC
+ * @hwdev: the NIC HW device
+ *
+ * Return 0 - Success, negative - Failure
+ **/
+static int clear_io_resources(struct hinic_hwdev *hwdev)
+{
+	struct hinic_cmd_clear_io_res cmd_clear_io_res;
+	struct hinic_hwif *hwif = hwdev->hwif;
+	struct pci_dev *pdev = hwif->pdev;
+	struct hinic_pfhwdev *pfhwdev;
+	int err;
+
+	/* sleep 100ms to wait for firmware stopping I/O */
+	msleep(100);
+
+	cmd_clear_io_res.func_idx = HINIC_HWIF_FUNC_IDX(hwif);
+
+	pfhwdev = container_of(hwdev, struct hinic_pfhwdev, hwdev);
+
+	err = hinic_msg_to_mgmt(&pfhwdev->pf_to_mgmt, HINIC_MOD_COMM,
+				HINIC_COMM_CMD_IO_RES_CLEAR, &cmd_clear_io_res,
+				sizeof(cmd_clear_io_res), NULL, NULL,
+				HINIC_MGMT_MSG_SYNC);
+	if (err) {
+		dev_err(&pdev->dev, "Failed to clear IO resources\n");
+		return err;
+	}
+
+	return 0;
+}
+
+/**
+ * set_resources_state - set the state of the resources in the NIC
+ * @hwdev: the NIC HW device
+ * @state: the state to set
+ *
+ * Return 0 - Success, negative - Failure
+ **/
+static int set_resources_state(struct hinic_hwdev *hwdev,
+			       enum hinic_res_state state)
+{
+	struct hinic_cmd_set_res_state res_state;
+	struct hinic_hwif *hwif = hwdev->hwif;
+	struct hinic_pfhwdev *pfhwdev;
+
+	res_state.func_idx = HINIC_HWIF_FUNC_IDX(hwif);
+	res_state.state = state;
+
+	pfhwdev = container_of(hwdev, struct hinic_pfhwdev, hwdev);
+
+	return hinic_msg_to_mgmt(&pfhwdev->pf_to_mgmt,
+				 HINIC_MOD_COMM,
+				 HINIC_COMM_CMD_RES_STATE_SET,
+				 &res_state, sizeof(res_state), NULL,
+				 NULL, HINIC_MGMT_MSG_SYNC);
+}
+
+/**
+ * get_base_qpn - get the first qp number
+ * @hwdev: the NIC HW device
+ * @base_qpn: returned qp number
+ *
+ * Return 0 - Success, negative - Failure
+ **/
+static int get_base_qpn(struct hinic_hwdev *hwdev, u16 *base_qpn)
+{
+	struct hinic_cmd_base_qpn cmd_base_qpn;
+	struct hinic_hwif *hwif = hwdev->hwif;
+	u16 out_size = sizeof(cmd_base_qpn);
+	struct pci_dev *pdev = hwif->pdev;
+	int err;
+
+	cmd_base_qpn.func_idx = HINIC_HWIF_FUNC_IDX(hwif);
+
+	err = hinic_port_msg_cmd(hwdev, HINIC_PORT_CMD_GET_GLOBAL_QPN,
+				 &cmd_base_qpn, sizeof(cmd_base_qpn),
+				 &cmd_base_qpn, &out_size);
+	if (err || out_size != sizeof(cmd_base_qpn) || cmd_base_qpn.status) {
+		dev_err(&pdev->dev, "Failed to get base qpn, err: %d, status: 0x%x, out size: 0x%x\n",
+			err, cmd_base_qpn.status, out_size);
+		return -EIO;
+	}
+
+	*base_qpn = cmd_base_qpn.qpn;
+	return 0;
+}
+
+/**
+ * hinic_hwdev_ifup - Preparing the HW for passing IO
+ * @hwdev: the NIC HW device
+ * @sq_depth: the send queue depth
+ * @rq_depth: the receive queue depth
+ *
+ * Return 0 - Success, negative - Failure
+ **/
+int hinic_hwdev_ifup(struct hinic_hwdev *hwdev, u16 sq_depth, u16 rq_depth)
+{
+	struct hinic_func_to_io *func_to_io = &hwdev->func_to_io;
+	struct hinic_cap *nic_cap = &hwdev->nic_cap;
+	struct hinic_hwif *hwif = hwdev->hwif;
+	int err, num_aeqs, num_ceqs, num_qps;
+	struct msix_entry *ceq_msix_entries;
+	struct msix_entry *sq_msix_entries;
+	struct msix_entry *rq_msix_entries;
+	struct pci_dev *pdev = hwif->pdev;
+	u16 base_qpn;
+
+	err = get_base_qpn(hwdev, &base_qpn);
+	if (err) {
+		dev_err(&pdev->dev, "Failed to get global base qp number\n");
+		return err;
+	}
+
+	num_aeqs = HINIC_HWIF_NUM_AEQS(hwif);
+	num_ceqs = HINIC_HWIF_NUM_CEQS(hwif);
+
+	ceq_msix_entries = &hwdev->msix_entries[num_aeqs];
+	func_to_io->hwdev = hwdev;
+	func_to_io->sq_depth = sq_depth;
+	func_to_io->rq_depth = rq_depth;
+	func_to_io->global_qpn = base_qpn;
+
+	err = hinic_io_init(func_to_io, hwif, nic_cap->max_qps, num_ceqs,
+			    ceq_msix_entries);
+	if (err) {
+		dev_err(&pdev->dev, "Failed to init IO channel\n");
+		return err;
+	}
+
+	num_qps = nic_cap->num_qps;
+	sq_msix_entries = &hwdev->msix_entries[num_aeqs + num_ceqs];
+	rq_msix_entries = &hwdev->msix_entries[num_aeqs + num_ceqs + num_qps];
+
+	err = hinic_io_create_qps(func_to_io, base_qpn, num_qps,
+				  sq_msix_entries, rq_msix_entries);
+	if (err) {
+		dev_err(&pdev->dev, "Failed to create QPs\n");
+		goto err_create_qps;
+	}
+
+	err = wait_for_db_state(hwdev);
+	if (err) {
+		dev_warn(&pdev->dev, "db - disabled, try again\n");
+		hinic_db_state_set(hwif, HINIC_DB_ENABLE);
+	}
+
+	err = set_hw_ioctxt(hwdev, sq_depth, rq_depth);
+	if (err) {
+		dev_err(&pdev->dev, "Failed to set HW IO ctxt\n");
+		goto err_hw_ioctxt;
+	}
+
+	return 0;
+
+err_hw_ioctxt:
+	hinic_io_destroy_qps(func_to_io, num_qps);
+
+err_create_qps:
+	hinic_io_free(func_to_io);
+	return err;
+}
+
+/**
+ * hinic_hwdev_ifdown - Closing the HW for passing IO
+ * @hwdev: the NIC HW device
+ *
+ **/
+void hinic_hwdev_ifdown(struct hinic_hwdev *hwdev)
+{
+	struct hinic_func_to_io *func_to_io = &hwdev->func_to_io;
+	struct hinic_cap *nic_cap = &hwdev->nic_cap;
+
+	clear_io_resources(hwdev);
+
+	hinic_io_destroy_qps(func_to_io, nic_cap->num_qps);
+	hinic_io_free(func_to_io);
+}
+
+/**
+ * hinic_hwdev_cb_register - register callback handler for MGMT events
+ * @hwdev: the NIC HW device
+ * @cmd: the mgmt event
+ * @handle: private data for the handler
+ * @handler: event handler
+ **/
+void hinic_hwdev_cb_register(struct hinic_hwdev *hwdev,
+			     enum hinic_mgmt_msg_cmd cmd, void *handle,
+			     void (*handler)(void *handle, void *buf_in,
+					     u16 in_size, void *buf_out,
+					     u16 *out_size))
+{
+	struct hinic_pfhwdev *pfhwdev;
+	struct hinic_nic_cb *nic_cb;
+	u8 cmd_cb;
+
+	pfhwdev = container_of(hwdev, struct hinic_pfhwdev, hwdev);
+
+	cmd_cb = cmd - HINIC_MGMT_MSG_CMD_BASE;
+	nic_cb = &pfhwdev->nic_cb[cmd_cb];
+
+	nic_cb->handler = handler;
+	nic_cb->handle = handle;
+	nic_cb->cb_state = HINIC_CB_ENABLED;
+}
+
+/**
+ * hinic_hwdev_cb_unregister - unregister callback handler for MGMT events
+ * @hwdev: the NIC HW device
+ * @cmd: the mgmt event
+ **/
+void hinic_hwdev_cb_unregister(struct hinic_hwdev *hwdev,
+			       enum hinic_mgmt_msg_cmd cmd)
+{
+	struct hinic_hwif *hwif = hwdev->hwif;
+	struct hinic_pfhwdev *pfhwdev;
+	struct hinic_nic_cb *nic_cb;
+	u8 cmd_cb;
+
+	if (!HINIC_IS_PF(hwif) && !HINIC_IS_PPF(hwif))
+		return;
+
+	pfhwdev = container_of(hwdev, struct hinic_pfhwdev, hwdev);
+
+	cmd_cb = cmd - HINIC_MGMT_MSG_CMD_BASE;
+	nic_cb = &pfhwdev->nic_cb[cmd_cb];
+
+	nic_cb->cb_state &= ~HINIC_CB_ENABLED;
+
+	while (nic_cb->cb_state & HINIC_CB_RUNNING)
+		schedule();
+
+	nic_cb->handler = NULL;
+}
+
+/**
+ * nic_mgmt_msg_handler - nic mgmt event handler
+ * @handle: private data for the handler
+ * @cmd: message command
+ * @buf_in: input buffer
+ * @in_size: input size
+ * @buf_out: output buffer
+ * @out_size: returned output size
+ **/
+static void nic_mgmt_msg_handler(void *handle, u8 cmd, void *buf_in,
+				 u16 in_size, void *buf_out, u16 *out_size)
+{
+	struct hinic_pfhwdev *pfhwdev = handle;
+	enum hinic_cb_state cb_state;
+	struct hinic_nic_cb *nic_cb;
+	struct hinic_hwdev *hwdev;
+	struct hinic_hwif *hwif;
+	struct pci_dev *pdev;
+	u8 cmd_cb;
+
+	hwdev = &pfhwdev->hwdev;
+	hwif = hwdev->hwif;
+	pdev = hwif->pdev;
+
+	if (cmd < HINIC_MGMT_MSG_CMD_BASE ||
+	    cmd >= HINIC_MGMT_MSG_CMD_MAX) {
+		dev_err(&pdev->dev, "unknown L2NIC event, cmd = %d\n", cmd);
+		return;
+	}
+
+	cmd_cb = cmd - HINIC_MGMT_MSG_CMD_BASE;
+
+	nic_cb = &pfhwdev->nic_cb[cmd_cb];
+
+	cb_state = cmpxchg(&nic_cb->cb_state,
+			   HINIC_CB_ENABLED,
+			   HINIC_CB_ENABLED | HINIC_CB_RUNNING);
+
+	if (cb_state == HINIC_CB_ENABLED && nic_cb->handler)
+		nic_cb->handler(nic_cb->handle, buf_in,
+				in_size, buf_out, out_size);
+	else
+		dev_err(&pdev->dev, "Unhandled NIC Event %d\n", cmd);
+
+	nic_cb->cb_state &= ~HINIC_CB_RUNNING;
+}
+
+static void hinic_comm_recv_mgmt_self_cmd_reg(struct hinic_pfhwdev *pfhwdev,
+					      u8 cmd,
+					      comm_mgmt_self_msg_proc proc)
+{
+	u8 cmd_idx;
+
+	cmd_idx = pfhwdev->proc.cmd_num;
+	if (cmd_idx >= HINIC_COMM_SELF_CMD_MAX) {
+		dev_err(&pfhwdev->hwdev.hwif->pdev->dev,
+			"Register recv mgmt process failed, cmd: 0x%x\n", cmd);
+		return;
+	}
+
+	pfhwdev->proc.info[cmd_idx].cmd = cmd;
+	pfhwdev->proc.info[cmd_idx].proc = proc;
+	pfhwdev->proc.cmd_num++;
+}
+
+static void hinic_comm_recv_mgmt_self_cmd_unreg(struct hinic_pfhwdev *pfhwdev,
+						u8 cmd)
+{
+	u8 cmd_idx;
+
+	cmd_idx = pfhwdev->proc.cmd_num;
+	if (cmd_idx >= HINIC_COMM_SELF_CMD_MAX) {
+		dev_err(&pfhwdev->hwdev.hwif->pdev->dev, "Unregister recv mgmt process failed, cmd: 0x%x\n",
+			cmd);
+		return;
+	}
+
+	for (cmd_idx = 0; cmd_idx < HINIC_COMM_SELF_CMD_MAX; cmd_idx++) {
+		if (cmd == pfhwdev->proc.info[cmd_idx].cmd) {
+			pfhwdev->proc.info[cmd_idx].cmd = 0;
+			pfhwdev->proc.info[cmd_idx].proc = NULL;
+			pfhwdev->proc.cmd_num--;
+		}
+	}
+}
+
+static void comm_mgmt_msg_handler(void *handle, u8 cmd, void *buf_in,
+				  u16 in_size, void *buf_out, u16 *out_size)
+{
+	struct hinic_pfhwdev *pfhwdev = handle;
+	u8 cmd_idx;
+
+	for (cmd_idx = 0; cmd_idx < pfhwdev->proc.cmd_num; cmd_idx++) {
+		if (cmd == pfhwdev->proc.info[cmd_idx].cmd) {
+			if (!pfhwdev->proc.info[cmd_idx].proc) {
+				dev_warn(&pfhwdev->hwdev.hwif->pdev->dev,
+					 "PF recv mgmt comm msg handle null, cmd: 0x%x\n",
+					 cmd);
+			} else {
+				pfhwdev->proc.info[cmd_idx].proc
+					(&pfhwdev->hwdev, buf_in, in_size,
+					 buf_out, out_size);
+			}
+
+			return;
+		}
+	}
+
+	dev_warn(&pfhwdev->hwdev.hwif->pdev->dev, "Received unknown mgmt cpu event: 0x%x\n",
+		 cmd);
+
+	*out_size = 0;
+}
+
+/* pf fault report event */
+static void pf_fault_event_handler(void *dev, void *buf_in, u16 in_size,
+				   void *buf_out, u16 *out_size)
+{
+	struct hinic_cmd_fault_event *fault_event = buf_in;
+	struct hinic_hwdev *hwdev = dev;
+
+	if (in_size != sizeof(*fault_event)) {
+		dev_err(&hwdev->hwif->pdev->dev, "Invalid fault event report, length: %d, should be %zu\n",
+			in_size, sizeof(*fault_event));
+		return;
+	}
+
+	if (!hwdev->devlink_dev || IS_ERR_OR_NULL(hwdev->devlink_dev->hw_fault_reporter))
+		return;
+
+	devlink_health_report(hwdev->devlink_dev->hw_fault_reporter,
+			      "HW fatal error reported", &fault_event->event);
+}
+
+static void mgmt_watchdog_timeout_event_handler(void *dev,
+						void *buf_in, u16 in_size,
+						void *buf_out, u16 *out_size)
+{
+	struct hinic_mgmt_watchdog_info *watchdog_info = buf_in;
+	struct hinic_hwdev *hwdev = dev;
+
+	if (in_size != sizeof(*watchdog_info)) {
+		dev_err(&hwdev->hwif->pdev->dev, "Invalid mgmt watchdog report, length: %d, should be %zu\n",
+			in_size, sizeof(*watchdog_info));
+		return;
+	}
+
+	if (!hwdev->devlink_dev || IS_ERR_OR_NULL(hwdev->devlink_dev->fw_fault_reporter))
+		return;
+
+	devlink_health_report(hwdev->devlink_dev->fw_fault_reporter,
+			      "FW fatal error reported", watchdog_info);
+}
+
+/**
+ * init_pfhwdev - Initialize the extended components of PF
+ * @pfhwdev: the HW device for PF
+ *
+ * Return 0 - success, negative - failure
+ **/
+static int init_pfhwdev(struct hinic_pfhwdev *pfhwdev)
+{
+	struct hinic_hwdev *hwdev = &pfhwdev->hwdev;
+	struct hinic_hwif *hwif = hwdev->hwif;
+	struct pci_dev *pdev = hwif->pdev;
+	int err;
+
+	err = hinic_pf_to_mgmt_init(&pfhwdev->pf_to_mgmt, hwif);
+	if (err) {
+		dev_err(&pdev->dev, "Failed to initialize PF to MGMT channel\n");
+		return err;
+	}
+
+	err = hinic_func_to_func_init(hwdev);
+	if (err) {
+		dev_err(&hwif->pdev->dev, "Failed to init mailbox\n");
+		hinic_pf_to_mgmt_free(&pfhwdev->pf_to_mgmt);
+		return err;
+	}
+
+	if (!HINIC_IS_VF(hwif)) {
+		hinic_register_mgmt_msg_cb(&pfhwdev->pf_to_mgmt,
+					   HINIC_MOD_L2NIC, pfhwdev,
+					   nic_mgmt_msg_handler);
+		hinic_register_mgmt_msg_cb(&pfhwdev->pf_to_mgmt, HINIC_MOD_COMM,
+					   pfhwdev, comm_mgmt_msg_handler);
+		hinic_comm_recv_mgmt_self_cmd_reg(pfhwdev,
+						  HINIC_COMM_CMD_FAULT_REPORT,
+						  pf_fault_event_handler);
+		hinic_comm_recv_mgmt_self_cmd_reg
+			(pfhwdev, HINIC_COMM_CMD_WATCHDOG_INFO,
+			 mgmt_watchdog_timeout_event_handler);
+	} else {
+		hinic_register_vf_mbox_cb(hwdev, HINIC_MOD_L2NIC,
+					  nic_mgmt_msg_handler);
+	}
+
+	hinic_set_pf_action(hwif, HINIC_PF_MGMT_ACTIVE);
+	hinic_devlink_register(hwdev->devlink_dev);
+	return 0;
+}
+
+/**
+ * free_pfhwdev - Free the extended components of PF
+ * @pfhwdev: the HW device for PF
+ **/
+static void free_pfhwdev(struct hinic_pfhwdev *pfhwdev)
+{
+	struct hinic_hwdev *hwdev = &pfhwdev->hwdev;
+
+	hinic_devlink_unregister(hwdev->devlink_dev);
+	hinic_set_pf_action(hwdev->hwif, HINIC_PF_MGMT_INIT);
+
+	if (!HINIC_IS_VF(hwdev->hwif)) {
+		hinic_comm_recv_mgmt_self_cmd_unreg(pfhwdev,
+						    HINIC_COMM_CMD_WATCHDOG_INFO);
+		hinic_comm_recv_mgmt_self_cmd_unreg(pfhwdev,
+						    HINIC_COMM_CMD_FAULT_REPORT);
+		hinic_unregister_mgmt_msg_cb(&pfhwdev->pf_to_mgmt,
+					     HINIC_MOD_COMM);
+		hinic_unregister_mgmt_msg_cb(&pfhwdev->pf_to_mgmt,
+					     HINIC_MOD_L2NIC);
+	} else {
+		hinic_unregister_vf_mbox_cb(hwdev, HINIC_MOD_L2NIC);
+	}
+
+	hinic_func_to_func_free(hwdev);
+
+	hinic_pf_to_mgmt_free(&pfhwdev->pf_to_mgmt);
+}
+
+static int hinic_l2nic_reset(struct hinic_hwdev *hwdev)
+{
+	struct hinic_cmd_l2nic_reset l2nic_reset = {0};
+	u16 out_size = sizeof(l2nic_reset);
+	struct hinic_pfhwdev *pfhwdev;
+	int err;
+
+	pfhwdev = container_of(hwdev, struct hinic_pfhwdev, hwdev);
+
+	l2nic_reset.func_id = HINIC_HWIF_FUNC_IDX(hwdev->hwif);
+	/* 0 represents standard l2nic reset flow */
+	l2nic_reset.reset_flag = 0;
+
+	err = hinic_msg_to_mgmt(&pfhwdev->pf_to_mgmt, HINIC_MOD_COMM,
+				HINIC_COMM_CMD_L2NIC_RESET, &l2nic_reset,
+				sizeof(l2nic_reset), &l2nic_reset,
+				&out_size, HINIC_MGMT_MSG_SYNC);
+	if (err || !out_size || l2nic_reset.status) {
+		dev_err(&hwdev->hwif->pdev->dev, "Failed to reset L2NIC resources, err: %d, status: 0x%x, out_size: 0x%x\n",
+			err, l2nic_reset.status, out_size);
+		return -EIO;
+	}
+
+	return 0;
+}
+
+static int hinic_get_interrupt_cfg(struct hinic_hwdev *hwdev,
+				   struct hinic_msix_config *interrupt_info)
+{
+	u16 out_size = sizeof(*interrupt_info);
+	struct hinic_pfhwdev *pfhwdev;
+	int err;
+
+	if (!hwdev || !interrupt_info)
+		return -EINVAL;
+
+	pfhwdev = container_of(hwdev, struct hinic_pfhwdev, hwdev);
+
+	interrupt_info->func_id = HINIC_HWIF_FUNC_IDX(hwdev->hwif);
+
+	err = hinic_msg_to_mgmt(&pfhwdev->pf_to_mgmt, HINIC_MOD_COMM,
+				HINIC_COMM_CMD_MSI_CTRL_REG_RD_BY_UP,
+				interrupt_info, sizeof(*interrupt_info),
+				interrupt_info, &out_size, HINIC_MGMT_MSG_SYNC);
+	if (err || !out_size || interrupt_info->status) {
+		dev_err(&hwdev->hwif->pdev->dev, "Failed to get interrupt config, err: %d, status: 0x%x, out size: 0x%x\n",
+			err, interrupt_info->status, out_size);
+		return -EIO;
+	}
+
+	return 0;
+}
+
+int hinic_set_interrupt_cfg(struct hinic_hwdev *hwdev,
+			    struct hinic_msix_config *interrupt_info)
+{
+	u16 out_size = sizeof(*interrupt_info);
+	struct hinic_msix_config temp_info;
+	struct hinic_pfhwdev *pfhwdev;
+	int err;
+
+	if (!hwdev)
+		return -EINVAL;
+
+	pfhwdev = container_of(hwdev, struct hinic_pfhwdev, hwdev);
+
+	interrupt_info->func_id = HINIC_HWIF_FUNC_IDX(hwdev->hwif);
+
+	err = hinic_get_interrupt_cfg(hwdev, &temp_info);
+	if (err)
+		return -EINVAL;
+
+	interrupt_info->lli_credit_cnt = temp_info.lli_credit_cnt;
+	interrupt_info->lli_timer_cnt = temp_info.lli_timer_cnt;
+
+	err = hinic_msg_to_mgmt(&pfhwdev->pf_to_mgmt, HINIC_MOD_COMM,
+				HINIC_COMM_CMD_MSI_CTRL_REG_WR_BY_UP,
+				interrupt_info, sizeof(*interrupt_info),
+				interrupt_info, &out_size, HINIC_MGMT_MSG_SYNC);
+	if (err || !out_size || interrupt_info->status) {
+		dev_err(&hwdev->hwif->pdev->dev, "Failed to get interrupt config, err: %d, status: 0x%x, out size: 0x%x\n",
+			err, interrupt_info->status, out_size);
+		return -EIO;
+	}
+
+	return 0;
+}
+
+/**
+ * hinic_init_hwdev - Initialize the NIC HW
+ * @pdev: the NIC pci device
+ * @devlink: the poniter of hinic devlink
+ *
+ * Return initialized NIC HW device
+ *
+ * Initialize the NIC HW device and return a pointer to it
+ **/
+struct hinic_hwdev *hinic_init_hwdev(struct pci_dev *pdev, struct devlink *devlink)
+{
+	struct hinic_pfhwdev *pfhwdev;
+	struct hinic_hwdev *hwdev;
+	struct hinic_hwif *hwif;
+	int err, num_aeqs;
+
+	hwif = devm_kzalloc(&pdev->dev, sizeof(*hwif), GFP_KERNEL);
+	if (!hwif)
+		return ERR_PTR(-ENOMEM);
+
+	err = hinic_init_hwif(hwif, pdev);
+	if (err) {
+		dev_err(&pdev->dev, "Failed to init HW interface\n");
+		return ERR_PTR(err);
+	}
+
+	pfhwdev = devm_kzalloc(&pdev->dev, sizeof(*pfhwdev), GFP_KERNEL);
+	if (!pfhwdev) {
+		err = -ENOMEM;
+		goto err_pfhwdev_alloc;
+	}
+
+	hwdev = &pfhwdev->hwdev;
+	hwdev->hwif = hwif;
+	hwdev->devlink_dev = devlink_priv(devlink);
+	hwdev->devlink_dev->hwdev = hwdev;
+
+	err = init_msix(hwdev);
+	if (err) {
+		dev_err(&pdev->dev, "Failed to init msix\n");
+		goto err_init_msix;
+	}
+
+	err = wait_for_outbound_state(hwdev);
+	if (err) {
+		dev_warn(&pdev->dev, "outbound - disabled, try again\n");
+		hinic_outbound_state_set(hwif, HINIC_OUTBOUND_ENABLE);
+	}
+
+	num_aeqs = HINIC_HWIF_NUM_AEQS(hwif);
+
+	err = hinic_aeqs_init(&hwdev->aeqs, hwif, num_aeqs,
+			      HINIC_DEFAULT_AEQ_LEN, HINIC_EQ_PAGE_SIZE,
+			      hwdev->msix_entries);
+	if (err) {
+		dev_err(&pdev->dev, "Failed to init async event queues\n");
+		goto err_aeqs_init;
+	}
+
+	err = init_pfhwdev(pfhwdev);
+	if (err) {
+		dev_err(&pdev->dev, "Failed to init PF HW device\n");
+		goto err_init_pfhwdev;
+	}
+
+	err = hinic_l2nic_reset(hwdev);
+	if (err)
+		goto err_l2nic_reset;
+
+	err = get_dev_cap(hwdev);
+	if (err) {
+		dev_err(&pdev->dev, "Failed to get device capabilities\n");
+		goto err_dev_cap;
+	}
+
+	mutex_init(&hwdev->func_to_io.nic_cfg.cfg_mutex);
+
+	err = hinic_vf_func_init(hwdev);
+	if (err) {
+		dev_err(&pdev->dev, "Failed to init nic mbox\n");
+		goto err_vf_func_init;
+	}
+
+	err = init_fw_ctxt(hwdev);
+	if (err) {
+		dev_err(&pdev->dev, "Failed to init function table\n");
+		goto err_init_fw_ctxt;
+	}
+
+	err = set_resources_state(hwdev, HINIC_RES_ACTIVE);
+	if (err) {
+		dev_err(&pdev->dev, "Failed to set resources state\n");
+		goto err_resources_state;
+	}
+
+	return hwdev;
+
+err_resources_state:
+err_init_fw_ctxt:
+	hinic_vf_func_free(hwdev);
+err_vf_func_init:
+err_l2nic_reset:
+err_dev_cap:
+	free_pfhwdev(pfhwdev);
+
+err_init_pfhwdev:
+	hinic_aeqs_free(&hwdev->aeqs);
+
+err_aeqs_init:
+	disable_msix(hwdev);
+
+err_init_msix:
+err_pfhwdev_alloc:
+	hinic_free_hwif(hwif);
+	if (err > 0)
+		err = -EIO;
+	return ERR_PTR(err);
+}
+
+/**
+ * hinic_free_hwdev - Free the NIC HW device
+ * @hwdev: the NIC HW device
+ **/
+void hinic_free_hwdev(struct hinic_hwdev *hwdev)
+{
+	struct hinic_pfhwdev *pfhwdev = container_of(hwdev,
+						     struct hinic_pfhwdev,
+						     hwdev);
+
+	set_resources_state(hwdev, HINIC_RES_CLEAN);
+
+	hinic_vf_func_free(hwdev);
+
+	free_pfhwdev(pfhwdev);
+
+	hinic_aeqs_free(&hwdev->aeqs);
+
+	disable_msix(hwdev);
+
+	hinic_free_hwif(hwdev->hwif);
+}
+
+/**
+ * hinic_hwdev_num_qps - return the number QPs available for use
+ * @hwdev: the NIC HW device
+ *
+ * Return number QPs available for use
+ **/
+int hinic_hwdev_num_qps(struct hinic_hwdev *hwdev)
+{
+	struct hinic_cap *nic_cap = &hwdev->nic_cap;
+
+	return nic_cap->num_qps;
+}
+
+/**
+ * hinic_hwdev_get_sq - get SQ
+ * @hwdev: the NIC HW device
+ * @i: the position of the SQ
+ *
+ * Return: the SQ in the i position
+ **/
+struct hinic_sq *hinic_hwdev_get_sq(struct hinic_hwdev *hwdev, int i)
+{
+	struct hinic_func_to_io *func_to_io = &hwdev->func_to_io;
+	struct hinic_qp *qp = &func_to_io->qps[i];
+
+	if (i >= hinic_hwdev_num_qps(hwdev))
+		return NULL;
+
+	return &qp->sq;
+}
+
+/**
+ * hinic_hwdev_get_rq - get RQ
+ * @hwdev: the NIC HW device
+ * @i: the position of the RQ
+ *
+ * Return: the RQ in the i position
+ **/
+struct hinic_rq *hinic_hwdev_get_rq(struct hinic_hwdev *hwdev, int i)
+{
+	struct hinic_func_to_io *func_to_io = &hwdev->func_to_io;
+	struct hinic_qp *qp = &func_to_io->qps[i];
+
+	if (i >= hinic_hwdev_num_qps(hwdev))
+		return NULL;
+
+	return &qp->rq;
+}
+
+/**
+ * hinic_hwdev_msix_cnt_set - clear message attribute counters for msix entry
+ * @hwdev: the NIC HW device
+ * @msix_index: msix_index
+ *
+ * Return 0 - Success, negative - Failure
+ **/
+int hinic_hwdev_msix_cnt_set(struct hinic_hwdev *hwdev, u16 msix_index)
+{
+	return hinic_msix_attr_cnt_clear(hwdev->hwif, msix_index);
+}
+
+/**
+ * hinic_hwdev_msix_set - set message attribute for msix entry
+ * @hwdev: the NIC HW device
+ * @msix_index: msix_index
+ * @pending_limit: the maximum pending interrupt events (unit 8)
+ * @coalesc_timer: coalesc period for interrupt (unit 8 us)
+ * @lli_timer_cfg: replenishing period for low latency credit (unit 8 us)
+ * @lli_credit_limit: maximum credits for low latency msix messages (unit 8)
+ * @resend_timer: maximum wait for resending msix (unit coalesc period)
+ *
+ * Return 0 - Success, negative - Failure
+ **/
+int hinic_hwdev_msix_set(struct hinic_hwdev *hwdev, u16 msix_index,
+			 u8 pending_limit, u8 coalesc_timer,
+			 u8 lli_timer_cfg, u8 lli_credit_limit,
+			 u8 resend_timer)
+{
+	return hinic_msix_attr_set(hwdev->hwif, msix_index,
+				   pending_limit, coalesc_timer,
+				   lli_timer_cfg, lli_credit_limit,
+				   resend_timer);
+}
+
+/**
+ * hinic_hwdev_hw_ci_addr_set - set cons idx addr and attributes in HW for sq
+ * @hwdev: the NIC HW device
+ * @sq: send queue
+ * @pending_limit: the maximum pending update ci events (unit 8)
+ * @coalesc_timer: coalesc period for update ci (unit 8 us)
+ *
+ * Return 0 - Success, negative - Failure
+ **/
+int hinic_hwdev_hw_ci_addr_set(struct hinic_hwdev *hwdev, struct hinic_sq *sq,
+			       u8 pending_limit, u8 coalesc_timer)
+{
+	struct hinic_qp *qp = container_of(sq, struct hinic_qp, sq);
+	struct hinic_hwif *hwif = hwdev->hwif;
+	struct hinic_pfhwdev *pfhwdev;
+	struct hinic_cmd_hw_ci hw_ci;
+
+	hw_ci.dma_attr_off  = 0;
+	hw_ci.pending_limit = pending_limit;
+	hw_ci.coalesc_timer = coalesc_timer;
+
+	hw_ci.msix_en = 1;
+	hw_ci.msix_entry_idx = sq->msix_entry;
+
+	hw_ci.func_idx = HINIC_HWIF_FUNC_IDX(hwif);
+
+	hw_ci.sq_id = qp->q_id;
+
+	hw_ci.ci_addr = ADDR_IN_4BYTES(sq->hw_ci_dma_addr);
+
+	pfhwdev = container_of(hwdev, struct hinic_pfhwdev, hwdev);
+	return hinic_msg_to_mgmt(&pfhwdev->pf_to_mgmt,
+				 HINIC_MOD_COMM,
+				 HINIC_COMM_CMD_SQ_HI_CI_SET,
+				 &hw_ci, sizeof(hw_ci), NULL,
+				 NULL, HINIC_MGMT_MSG_SYNC);
+}
+
+/**
+ * hinic_hwdev_set_msix_state- set msix state
+ * @hwdev: the NIC HW device
+ * @msix_index: IRQ corresponding index number
+ * @flag: msix state
+ *
+ **/
+void hinic_hwdev_set_msix_state(struct hinic_hwdev *hwdev, u16 msix_index,
+				enum hinic_msix_state flag)
+{
+	hinic_set_msix_state(hwdev->hwif, msix_index, flag);
+}
+
+int hinic_get_board_info(struct hinic_hwdev *hwdev,
+			 struct hinic_comm_board_info *board_info)
+{
+	u16 out_size = sizeof(*board_info);
+	struct hinic_pfhwdev *pfhwdev;
+	int err;
+
+	if (!hwdev || !board_info)
+		return -EINVAL;
+
+	pfhwdev = container_of(hwdev, struct hinic_pfhwdev, hwdev);
+
+	err = hinic_msg_to_mgmt(&pfhwdev->pf_to_mgmt, HINIC_MOD_COMM,
+				HINIC_COMM_CMD_GET_BOARD_INFO,
+				board_info, sizeof(*board_info),
+				board_info, &out_size, HINIC_MGMT_MSG_SYNC);
+	if (err || board_info->status || !out_size) {
+		dev_err(&hwdev->hwif->pdev->dev,
+			"Failed to get board info, err: %d, status: 0x%x, out size: 0x%x\n",
+			err, board_info->status, out_size);
+		return -EIO;
+	}
+
+	return 0;
+}
diff --git a/drivers/net/ethernet/huawei/hinic/hinic_hw_dev.h b/drivers/net/ethernet/huawei/hinic/hinic_hw_dev.h
new file mode 100644
index 000000000000..6b5797e69781
--- /dev/null
+++ b/drivers/net/ethernet/huawei/hinic/hinic_hw_dev.h
@@ -0,0 +1,660 @@
+/* SPDX-License-Identifier: GPL-2.0-only */
+/*
+ * Huawei HiNIC PCI Express Linux driver
+ * Copyright(c) 2017 Huawei Technologies Co., Ltd
+ */
+
+#ifndef HINIC_HW_DEV_H
+#define HINIC_HW_DEV_H
+
+#include <linux/pci.h>
+#include <linux/types.h>
+#include <linux/bitops.h>
+#include <net/devlink.h>
+
+#include "hinic_hw_if.h"
+#include "hinic_hw_eqs.h"
+#include "hinic_hw_mgmt.h"
+#include "hinic_hw_qp.h"
+#include "hinic_hw_io.h"
+#include "hinic_hw_mbox.h"
+
+#define HINIC_MAX_QPS   32
+
+#define HINIC_MGMT_NUM_MSG_CMD  (HINIC_MGMT_MSG_CMD_MAX - \
+				 HINIC_MGMT_MSG_CMD_BASE)
+
+#define HINIC_PF_SET_VF_ALREADY				0x4
+#define HINIC_MGMT_STATUS_EXIST				0x6
+#define HINIC_MGMT_CMD_UNSUPPORTED			0xFF
+
+#define HINIC_CMD_VER_FUNC_ID				2
+
+struct hinic_cap {
+	u16     max_qps;
+	u16     num_qps;
+	u8		max_vf;
+	u16     max_vf_qps;
+};
+
+enum hw_ioctxt_set_cmdq_depth {
+	HW_IOCTXT_SET_CMDQ_DEPTH_DEFAULT,
+	HW_IOCTXT_SET_CMDQ_DEPTH_ENABLE,
+};
+
+enum hinic_port_cmd {
+	HINIC_PORT_CMD_VF_REGISTER = 0x0,
+	HINIC_PORT_CMD_VF_UNREGISTER = 0x1,
+
+	HINIC_PORT_CMD_CHANGE_MTU = 0x2,
+
+	HINIC_PORT_CMD_ADD_VLAN = 0x3,
+	HINIC_PORT_CMD_DEL_VLAN = 0x4,
+
+	HINIC_PORT_CMD_SET_ETS = 0x7,
+	HINIC_PORT_CMD_GET_ETS = 0x8,
+
+	HINIC_PORT_CMD_SET_PFC = 0x5,
+
+	HINIC_PORT_CMD_SET_MAC = 0x9,
+	HINIC_PORT_CMD_GET_MAC = 0xA,
+	HINIC_PORT_CMD_DEL_MAC = 0xB,
+
+	HINIC_PORT_CMD_SET_RX_MODE = 0xC,
+
+	HINIC_PORT_CMD_SET_ANTI_ATTACK_RATE = 0xD,
+
+	HINIC_PORT_CMD_GET_PAUSE_INFO = 0x14,
+	HINIC_PORT_CMD_SET_PAUSE_INFO = 0x15,
+
+	HINIC_PORT_CMD_GET_LINK_STATE = 0x18,
+
+	HINIC_PORT_CMD_SET_LRO = 0x19,
+
+	HINIC_PORT_CMD_SET_RX_CSUM = 0x1A,
+
+	HINIC_PORT_CMD_SET_RX_VLAN_OFFLOAD = 0x1B,
+
+	HINIC_PORT_CMD_GET_PORT_STATISTICS = 0x1C,
+
+	HINIC_PORT_CMD_CLEAR_PORT_STATISTICS = 0x1D,
+
+	HINIC_PORT_CMD_GET_VPORT_STAT = 0x1E,
+
+	HINIC_PORT_CMD_CLEAN_VPORT_STAT	= 0x1F,
+
+	HINIC_PORT_CMD_GET_RSS_TEMPLATE_INDIR_TBL = 0x25,
+
+	HINIC_PORT_CMD_SET_PORT_STATE = 0x29,
+	HINIC_PORT_CMD_GET_PORT_STATE = 0x30,
+
+	HINIC_PORT_CMD_SET_RSS_TEMPLATE_TBL = 0x2B,
+
+	HINIC_PORT_CMD_GET_RSS_TEMPLATE_TBL = 0x2C,
+
+	HINIC_PORT_CMD_SET_RSS_HASH_ENGINE = 0x2D,
+
+	HINIC_PORT_CMD_GET_RSS_HASH_ENGINE = 0x2E,
+
+	HINIC_PORT_CMD_GET_RSS_CTX_TBL = 0x2F,
+
+	HINIC_PORT_CMD_SET_RSS_CTX_TBL = 0x30,
+
+	HINIC_PORT_CMD_RSS_TEMP_MGR	= 0x31,
+
+	HINIC_PORT_CMD_RD_LINE_TBL = 0x39,
+
+	HINIC_PORT_CMD_RSS_CFG = 0x42,
+
+	HINIC_PORT_CMD_GET_PHY_TYPE = 0x44,
+
+	HINIC_PORT_CMD_FWCTXT_INIT = 0x45,
+
+	HINIC_PORT_CMD_GET_LOOPBACK_MODE = 0x48,
+	HINIC_PORT_CMD_SET_LOOPBACK_MODE = 0x49,
+
+	HINIC_PORT_CMD_GET_JUMBO_FRAME_SIZE = 0x4A,
+	HINIC_PORT_CMD_SET_JUMBO_FRAME_SIZE = 0x4B,
+
+	HINIC_PORT_CMD_ENABLE_SPOOFCHK = 0x4E,
+
+	HINIC_PORT_CMD_GET_MGMT_VERSION = 0x58,
+
+	HINIC_PORT_CMD_GET_PORT_TYPE = 0x5B,
+
+	HINIC_PORT_CMD_SET_FUNC_STATE = 0x5D,
+
+	HINIC_PORT_CMD_GET_PORT_ID_BY_FUNC_ID = 0x5E,
+
+	HINIC_PORT_CMD_GET_DMA_CS = 0x64,
+	HINIC_PORT_CMD_SET_DMA_CS = 0x65,
+
+	HINIC_PORT_CMD_GET_GLOBAL_QPN = 0x66,
+
+	HINIC_PORT_CMD_SET_VF_RATE = 0x69,
+
+	HINIC_PORT_CMD_SET_VF_VLAN = 0x6A,
+
+	HINIC_PORT_CMD_CLR_VF_VLAN = 0x6B,
+
+	HINIC_PORT_CMD_SET_TSO = 0x70,
+
+	HINIC_PORT_CMD_UPDATE_FW = 0x72,
+
+	HINIC_PORT_CMD_SET_RQ_IQ_MAP = 0x73,
+
+	HINIC_PORT_CMD_SET_PFC_THD = 0x75,
+
+	HINIC_PORT_CMD_LINK_STATUS_REPORT = 0xA0,
+
+	HINIC_PORT_CMD_SET_LOSSLESS_ETH	= 0xA3,
+
+	HINIC_PORT_CMD_UPDATE_MAC = 0xA4,
+
+	HINIC_PORT_CMD_GET_CAP = 0xAA,
+
+	HINIC_PORT_CMD_UP_TC_ADD_FLOW = 0xAF,
+	HINIC_PORT_CMD_UP_TC_DEL_FLOW = 0xB0,
+	HINIC_PORT_CMD_UP_TC_GET_FLOW = 0xB1,
+
+	HINIC_PORT_CMD_UP_TC_FLUSH_TCAM = 0xB2,
+
+	HINIC_PORT_CMD_UP_TC_CTRL_TCAM_BLOCK = 0xB3,
+
+	HINIC_PORT_CMD_UP_TC_ENABLE = 0xB4,
+
+	HINIC_PORT_CMD_UP_TC_GET_TCAM_BLOCK = 0xB5,
+
+	HINIC_PORT_CMD_SET_IPSU_MAC = 0xCB,
+	HINIC_PORT_CMD_GET_IPSU_MAC = 0xCC,
+
+	HINIC_PORT_CMD_SET_XSFP_STATUS = 0xD4,
+
+	HINIC_PORT_CMD_GET_LINK_MODE = 0xD9,
+
+	HINIC_PORT_CMD_SET_SPEED = 0xDA,
+
+	HINIC_PORT_CMD_SET_AUTONEG = 0xDB,
+
+	HINIC_PORT_CMD_CLEAR_QP_RES = 0xDD,
+
+	HINIC_PORT_CMD_SET_SUPER_CQE = 0xDE,
+
+	HINIC_PORT_CMD_SET_VF_COS = 0xDF,
+	HINIC_PORT_CMD_GET_VF_COS = 0xE1,
+
+	HINIC_PORT_CMD_CABLE_PLUG_EVENT	= 0xE5,
+
+	HINIC_PORT_CMD_LINK_ERR_EVENT = 0xE6,
+
+	HINIC_PORT_CMD_SET_COS_UP_MAP = 0xE8,
+
+	HINIC_PORT_CMD_RESET_LINK_CFG = 0xEB,
+
+	HINIC_PORT_CMD_GET_STD_SFP_INFO = 0xF0,
+
+	HINIC_PORT_CMD_FORCE_PKT_DROP = 0xF3,
+
+	HINIC_PORT_CMD_SET_LRO_TIMER = 0xF4,
+
+	HINIC_PORT_CMD_SET_VHD_CFG = 0xF7,
+
+	HINIC_PORT_CMD_SET_LINK_FOLLOW = 0xF8,
+
+	HINIC_PORT_CMD_SET_VF_MAX_MIN_RATE = 0xF9,
+
+	HINIC_PORT_CMD_GET_SFP_ABS = 0xFB,
+
+	HINIC_PORT_CMD_Q_FILTER	= 0xFC,
+
+	HINIC_PORT_CMD_TCAM_FILTER = 0xFE,
+
+	HINIC_PORT_CMD_SET_VLAN_FILTER = 0xFF,
+};
+
+/* cmd of mgmt CPU message for HILINK module */
+enum hinic_hilink_cmd {
+	HINIC_HILINK_CMD_GET_LINK_INFO		= 0x3,
+	HINIC_HILINK_CMD_SET_LINK_SETTINGS	= 0x8,
+};
+
+enum hinic_ucode_cmd {
+	HINIC_UCODE_CMD_MODIFY_QUEUE_CONTEXT    = 0,
+	HINIC_UCODE_CMD_CLEAN_QUEUE_CONTEXT,
+	HINIC_UCODE_CMD_ARM_SQ,
+	HINIC_UCODE_CMD_ARM_RQ,
+	HINIC_UCODE_CMD_SET_RSS_INDIR_TABLE,
+	HINIC_UCODE_CMD_SET_RSS_CONTEXT_TABLE,
+	HINIC_UCODE_CMD_GET_RSS_INDIR_TABLE,
+	HINIC_UCODE_CMD_GET_RSS_CONTEXT_TABLE,
+	HINIC_UCODE_CMD_SET_IQ_ENABLE,
+	HINIC_UCODE_CMD_SET_RQ_FLUSH            = 10
+};
+
+#define NIC_RSS_CMD_TEMP_ALLOC  0x01
+#define NIC_RSS_CMD_TEMP_FREE   0x02
+
+enum hinic_mgmt_msg_cmd {
+	HINIC_MGMT_MSG_CMD_BASE         = 0xA0,
+
+	HINIC_MGMT_MSG_CMD_LINK_STATUS  = 0xA0,
+
+	HINIC_MGMT_MSG_CMD_CABLE_PLUG_EVENT	= 0xE5,
+	HINIC_MGMT_MSG_CMD_LINK_ERR_EVENT	= 0xE6,
+
+	HINIC_MGMT_MSG_CMD_MAX,
+};
+
+enum hinic_cb_state {
+	HINIC_CB_ENABLED = BIT(0),
+	HINIC_CB_RUNNING = BIT(1),
+};
+
+enum hinic_res_state {
+	HINIC_RES_CLEAN         = 0,
+	HINIC_RES_ACTIVE        = 1,
+};
+
+struct hinic_cmd_fw_ctxt {
+	u8      status;
+	u8      version;
+	u8      rsvd0[6];
+
+	u16     func_idx;
+	u16     rx_buf_sz;
+
+	u32     rsvd1;
+};
+
+struct hinic_cmd_hw_ioctxt {
+	u8      status;
+	u8      version;
+	u8      rsvd0[6];
+
+	u16     func_idx;
+
+	u16     rsvd1;
+
+	u8      set_cmdq_depth;
+	u8      cmdq_depth;
+
+	u8      lro_en;
+	u8      rsvd3;
+	u8      ppf_idx;
+	u8      rsvd4;
+
+	u16     rq_depth;
+	u16     rx_buf_sz_idx;
+	u16     sq_depth;
+};
+
+struct hinic_cmd_io_status {
+	u8      status;
+	u8      version;
+	u8      rsvd0[6];
+
+	u16     func_idx;
+	u8      rsvd1;
+	u8      rsvd2;
+	u32     io_status;
+};
+
+struct hinic_cmd_clear_io_res {
+	u8      status;
+	u8      version;
+	u8      rsvd0[6];
+
+	u16     func_idx;
+	u8      rsvd1;
+	u8      rsvd2;
+};
+
+struct hinic_cmd_set_res_state {
+	u8      status;
+	u8      version;
+	u8      rsvd0[6];
+
+	u16     func_idx;
+	u8      state;
+	u8      rsvd1;
+	u32     rsvd2;
+};
+
+struct hinic_ceq_ctrl_reg {
+	u8 status;
+	u8 version;
+	u8 rsvd0[6];
+
+	u16 func_id;
+	u16 q_id;
+	u32 ctrl0;
+	u32 ctrl1;
+};
+
+struct hinic_cmd_base_qpn {
+	u8      status;
+	u8      version;
+	u8      rsvd0[6];
+
+	u16     func_idx;
+	u16     qpn;
+};
+
+struct hinic_cmd_hw_ci {
+	u8      status;
+	u8      version;
+	u8      rsvd0[6];
+
+	u16     func_idx;
+
+	u8      dma_attr_off;
+	u8      pending_limit;
+	u8      coalesc_timer;
+
+	u8      msix_en;
+	u16     msix_entry_idx;
+
+	u32     sq_id;
+	u32     rsvd1;
+	u64     ci_addr;
+};
+
+struct hinic_cmd_l2nic_reset {
+	u8	status;
+	u8	version;
+	u8	rsvd0[6];
+
+	u16	func_id;
+	u16	reset_flag;
+};
+
+struct hinic_msix_config {
+	u8	status;
+	u8	version;
+	u8	rsvd0[6];
+
+	u16	func_id;
+	u16	msix_index;
+	u8	pending_cnt;
+	u8	coalesce_timer_cnt;
+	u8	lli_timer_cnt;
+	u8	lli_credit_cnt;
+	u8	resend_timer_cnt;
+	u8	rsvd1[3];
+};
+
+struct hinic_set_random_id {
+	u8    status;
+	u8    version;
+	u8    rsvd0[6];
+
+	u8    vf_in_pf;
+	u8    rsvd1;
+	u16   func_idx;
+	u32   random_id;
+};
+
+struct hinic_board_info {
+	u32	board_type;
+	u32	port_num;
+	u32	port_speed;
+	u32	pcie_width;
+	u32	host_num;
+	u32	pf_num;
+	u32	vf_total_num;
+	u32	tile_num;
+	u32	qcm_num;
+	u32	core_num;
+	u32	work_mode;
+	u32	service_mode;
+	u32	pcie_mode;
+	u32	cfg_addr;
+	u32	boot_sel;
+	u32	board_id;
+};
+
+struct hinic_comm_board_info {
+	u8	status;
+	u8	version;
+	u8	rsvd0[6];
+
+	struct hinic_board_info info;
+
+	u32	rsvd1[4];
+};
+
+struct hinic_hwdev {
+	struct hinic_hwif               *hwif;
+	struct msix_entry               *msix_entries;
+
+	struct hinic_aeqs               aeqs;
+	struct hinic_func_to_io         func_to_io;
+	struct hinic_mbox_func_to_func  *func_to_func;
+
+	struct hinic_cap                nic_cap;
+	u8				port_id;
+	struct hinic_devlink_priv	*devlink_dev;
+};
+
+struct hinic_nic_cb {
+	void    (*handler)(void *handle, void *buf_in,
+			   u16 in_size, void *buf_out,
+			   u16 *out_size);
+
+	void            *handle;
+	unsigned long   cb_state;
+};
+
+#define HINIC_COMM_SELF_CMD_MAX 4
+
+typedef void (*comm_mgmt_self_msg_proc)(void *handle, void *buf_in, u16 in_size,
+					void *buf_out, u16 *out_size);
+
+struct comm_mgmt_self_msg_sub_info {
+	u8 cmd;
+	comm_mgmt_self_msg_proc proc;
+};
+
+struct comm_mgmt_self_msg_info {
+	u8 cmd_num;
+	struct comm_mgmt_self_msg_sub_info info[HINIC_COMM_SELF_CMD_MAX];
+};
+
+struct hinic_pfhwdev {
+	struct hinic_hwdev              hwdev;
+
+	struct hinic_pf_to_mgmt         pf_to_mgmt;
+
+	struct hinic_nic_cb             nic_cb[HINIC_MGMT_NUM_MSG_CMD];
+
+	struct comm_mgmt_self_msg_info	proc;
+};
+
+struct hinic_dev_cap {
+	u8      status;
+	u8      version;
+	u8      rsvd0[6];
+
+	u8      rsvd1[5];
+	u8      intr_type;
+	u8	max_cos_id;
+	u8	er_id;
+	u8	port_id;
+	u8      max_vf;
+	u8      rsvd2[62];
+	u16     max_sqs;
+	u16	max_rqs;
+	u16	max_vf_sqs;
+	u16     max_vf_rqs;
+	u8      rsvd3[204];
+};
+
+union hinic_fault_hw_mgmt {
+	u32 val[4];
+	/* valid only type == FAULT_TYPE_CHIP */
+	struct {
+		u8 node_id;
+		u8 err_level;
+		u16 err_type;
+		u32 err_csr_addr;
+		u32 err_csr_value;
+		/* func_id valid only if err_level == FAULT_LEVEL_SERIOUS_FLR */
+		u16 func_id;
+		u16 rsvd2;
+	} chip;
+
+	/* valid only if type == FAULT_TYPE_UCODE */
+	struct {
+		u8 cause_id;
+		u8 core_id;
+		u8 c_id;
+		u8 rsvd3;
+		u32 epc;
+		u32 rsvd4;
+		u32 rsvd5;
+	} ucode;
+
+	/* valid only if type == FAULT_TYPE_MEM_RD_TIMEOUT ||
+	 * FAULT_TYPE_MEM_WR_TIMEOUT
+	 */
+	struct {
+		u32 err_csr_ctrl;
+		u32 err_csr_data;
+		u32 ctrl_tab;
+		u32 mem_index;
+	} mem_timeout;
+
+	/* valid only if type == FAULT_TYPE_REG_RD_TIMEOUT ||
+	 * FAULT_TYPE_REG_WR_TIMEOUT
+	 */
+	struct {
+		u32 err_csr;
+		u32 rsvd6;
+		u32 rsvd7;
+		u32 rsvd8;
+	} reg_timeout;
+
+	struct {
+		/* 0: read; 1: write */
+		u8 op_type;
+		u8 port_id;
+		u8 dev_ad;
+		u8 rsvd9;
+		u32 csr_addr;
+		u32 op_data;
+		u32 rsvd10;
+	} phy_fault;
+};
+
+struct hinic_fault_event {
+	u8 type;
+	u8 fault_level;
+	u8 rsvd0[2];
+	union hinic_fault_hw_mgmt event;
+};
+
+struct hinic_cmd_fault_event {
+	u8	status;
+	u8	version;
+	u8	rsvd0[6];
+
+	struct hinic_fault_event event;
+};
+
+enum hinic_fault_type {
+	FAULT_TYPE_CHIP,
+	FAULT_TYPE_UCODE,
+	FAULT_TYPE_MEM_RD_TIMEOUT,
+	FAULT_TYPE_MEM_WR_TIMEOUT,
+	FAULT_TYPE_REG_RD_TIMEOUT,
+	FAULT_TYPE_REG_WR_TIMEOUT,
+	FAULT_TYPE_PHY_FAULT,
+	FAULT_TYPE_MAX,
+};
+
+enum hinic_fault_err_level {
+	FAULT_LEVEL_FATAL,
+	FAULT_LEVEL_SERIOUS_RESET,
+	FAULT_LEVEL_SERIOUS_FLR,
+	FAULT_LEVEL_GENERAL,
+	FAULT_LEVEL_SUGGESTION,
+	FAULT_LEVEL_MAX
+};
+
+struct hinic_mgmt_watchdog_info {
+	u8 status;
+	u8 version;
+	u8 rsvd0[6];
+
+	u32 curr_time_h;
+	u32 curr_time_l;
+	u32 task_id;
+	u32 rsv;
+
+	u32 reg[13];
+	u32 pc;
+	u32 lr;
+	u32 cpsr;
+
+	u32 stack_top;
+	u32 stack_bottom;
+	u32 sp;
+	u32 curr_used;
+	u32 peak_used;
+	u32 is_overflow;
+
+	u32 stack_actlen;
+	u8 data[1024];
+};
+
+void hinic_hwdev_cb_register(struct hinic_hwdev *hwdev,
+			     enum hinic_mgmt_msg_cmd cmd, void *handle,
+			     void (*handler)(void *handle, void *buf_in,
+					     u16 in_size, void *buf_out,
+					     u16 *out_size));
+
+void hinic_hwdev_cb_unregister(struct hinic_hwdev *hwdev,
+			       enum hinic_mgmt_msg_cmd cmd);
+
+int hinic_port_msg_cmd(struct hinic_hwdev *hwdev, enum hinic_port_cmd cmd,
+		       void *buf_in, u16 in_size, void *buf_out,
+		       u16 *out_size);
+
+int hinic_hilink_msg_cmd(struct hinic_hwdev *hwdev, enum hinic_hilink_cmd cmd,
+			 void *buf_in, u16 in_size, void *buf_out,
+			 u16 *out_size);
+
+int hinic_hwdev_ifup(struct hinic_hwdev *hwdev, u16 sq_depth, u16 rq_depth);
+
+void hinic_hwdev_ifdown(struct hinic_hwdev *hwdev);
+
+struct hinic_hwdev *hinic_init_hwdev(struct pci_dev *pdev, struct devlink *devlink);
+
+void hinic_free_hwdev(struct hinic_hwdev *hwdev);
+
+int hinic_hwdev_num_qps(struct hinic_hwdev *hwdev);
+
+struct hinic_sq *hinic_hwdev_get_sq(struct hinic_hwdev *hwdev, int i);
+
+struct hinic_rq *hinic_hwdev_get_rq(struct hinic_hwdev *hwdev, int i);
+
+int hinic_hwdev_msix_cnt_set(struct hinic_hwdev *hwdev, u16 msix_index);
+
+int hinic_hwdev_msix_set(struct hinic_hwdev *hwdev, u16 msix_index,
+			 u8 pending_limit, u8 coalesc_timer,
+			 u8 lli_timer_cfg, u8 lli_credit_limit,
+			 u8 resend_timer);
+
+int hinic_hwdev_hw_ci_addr_set(struct hinic_hwdev *hwdev, struct hinic_sq *sq,
+			       u8 pending_limit, u8 coalesc_timer);
+
+void hinic_hwdev_set_msix_state(struct hinic_hwdev *hwdev, u16 msix_index,
+				enum hinic_msix_state flag);
+
+int hinic_set_interrupt_cfg(struct hinic_hwdev *hwdev,
+			    struct hinic_msix_config *interrupt_info);
+
+int hinic_get_board_info(struct hinic_hwdev *hwdev,
+			 struct hinic_comm_board_info *board_info);
+
+#endif
diff --git a/drivers/net/ethernet/huawei/hinic/hinic_hw_eqs.c b/drivers/net/ethernet/huawei/hinic/hinic_hw_eqs.c
new file mode 100644
index 000000000000..28114a59347e
--- /dev/null
+++ b/drivers/net/ethernet/huawei/hinic/hinic_hw_eqs.c
@@ -0,0 +1,996 @@
+// SPDX-License-Identifier: GPL-2.0-only
+/*
+ * Huawei HiNIC PCI Express Linux driver
+ * Copyright(c) 2017 Huawei Technologies Co., Ltd
+ */
+
+#include <linux/kernel.h>
+#include <linux/types.h>
+#include <linux/errno.h>
+#include <linux/pci.h>
+#include <linux/device.h>
+#include <linux/workqueue.h>
+#include <linux/interrupt.h>
+#include <linux/slab.h>
+#include <linux/dma-mapping.h>
+#include <linux/log2.h>
+#include <asm/byteorder.h>
+#include <asm/barrier.h>
+
+#include "hinic_hw_dev.h"
+#include "hinic_hw_csr.h"
+#include "hinic_hw_if.h"
+#include "hinic_hw_eqs.h"
+
+#define HINIC_EQS_WQ_NAME                       "hinic_eqs"
+
+#define GET_EQ_NUM_PAGES(eq, pg_size)           \
+		(ALIGN((eq)->q_len * (eq)->elem_size, pg_size) / (pg_size))
+
+#define GET_EQ_NUM_ELEMS_IN_PG(eq, pg_size)     ((pg_size) / (eq)->elem_size)
+
+#define EQ_CONS_IDX_REG_ADDR(eq)        (((eq)->type == HINIC_AEQ) ? \
+			HINIC_CSR_AEQ_CONS_IDX_ADDR((eq)->q_id) : \
+			HINIC_CSR_CEQ_CONS_IDX_ADDR((eq)->q_id))
+
+#define EQ_PROD_IDX_REG_ADDR(eq)        (((eq)->type == HINIC_AEQ) ? \
+			HINIC_CSR_AEQ_PROD_IDX_ADDR((eq)->q_id) : \
+			HINIC_CSR_CEQ_PROD_IDX_ADDR((eq)->q_id))
+
+#define EQ_HI_PHYS_ADDR_REG(eq, pg_num) (((eq)->type == HINIC_AEQ) ? \
+			HINIC_CSR_AEQ_HI_PHYS_ADDR_REG((eq)->q_id, pg_num) : \
+			HINIC_CSR_CEQ_HI_PHYS_ADDR_REG((eq)->q_id, pg_num))
+
+#define EQ_LO_PHYS_ADDR_REG(eq, pg_num) (((eq)->type == HINIC_AEQ) ? \
+			HINIC_CSR_AEQ_LO_PHYS_ADDR_REG((eq)->q_id, pg_num) : \
+			HINIC_CSR_CEQ_LO_PHYS_ADDR_REG((eq)->q_id, pg_num))
+
+#define GET_EQ_ELEMENT(eq, idx)         \
+		((eq)->virt_addr[(idx) / (eq)->num_elem_in_pg] + \
+		 (((idx) & ((eq)->num_elem_in_pg - 1)) * (eq)->elem_size))
+
+#define GET_AEQ_ELEM(eq, idx)           ((struct hinic_aeq_elem *) \
+					GET_EQ_ELEMENT(eq, idx))
+
+#define GET_CEQ_ELEM(eq, idx)           ((u32 *) \
+					 GET_EQ_ELEMENT(eq, idx))
+
+#define GET_CURR_AEQ_ELEM(eq)           GET_AEQ_ELEM(eq, (eq)->cons_idx)
+
+#define GET_CURR_CEQ_ELEM(eq)           GET_CEQ_ELEM(eq, (eq)->cons_idx)
+
+#define PAGE_IN_4K(page_size)           ((page_size) >> 12)
+#define EQ_SET_HW_PAGE_SIZE_VAL(eq)     (ilog2(PAGE_IN_4K((eq)->page_size)))
+
+#define ELEMENT_SIZE_IN_32B(eq)         (((eq)->elem_size) >> 5)
+#define EQ_SET_HW_ELEM_SIZE_VAL(eq)     (ilog2(ELEMENT_SIZE_IN_32B(eq)))
+
+#define EQ_MAX_PAGES                    8
+
+#define CEQE_TYPE_SHIFT                 23
+#define CEQE_TYPE_MASK                  0x7
+
+#define CEQE_TYPE(ceqe)                 (((ceqe) >> CEQE_TYPE_SHIFT) &  \
+					 CEQE_TYPE_MASK)
+
+#define CEQE_DATA_MASK                  0x3FFFFFF
+#define CEQE_DATA(ceqe)                 ((ceqe) & CEQE_DATA_MASK)
+
+#define aeq_to_aeqs(eq)                 \
+		container_of((eq) - (eq)->q_id, struct hinic_aeqs, aeq[0])
+
+#define ceq_to_ceqs(eq)                 \
+		container_of((eq) - (eq)->q_id, struct hinic_ceqs, ceq[0])
+
+#define work_to_aeq_work(work)          \
+		container_of(work, struct hinic_eq_work, work)
+
+#define DMA_ATTR_AEQ_DEFAULT            0
+#define DMA_ATTR_CEQ_DEFAULT            0
+
+/* No coalescence */
+#define THRESH_CEQ_DEFAULT              0
+
+enum eq_int_mode {
+	EQ_INT_MODE_ARMED,
+	EQ_INT_MODE_ALWAYS
+};
+
+enum eq_arm_state {
+	EQ_NOT_ARMED,
+	EQ_ARMED
+};
+
+/**
+ * hinic_aeq_register_hw_cb - register AEQ callback for specific event
+ * @aeqs: pointer to Async eqs of the chip
+ * @event: aeq event to register callback for it
+ * @handle: private data will be used by the callback
+ * @hwe_handler: callback function
+ **/
+void hinic_aeq_register_hw_cb(struct hinic_aeqs *aeqs,
+			      enum hinic_aeq_type event, void *handle,
+			      void (*hwe_handler)(void *handle, void *data,
+						  u8 size))
+{
+	struct hinic_hw_event_cb *hwe_cb = &aeqs->hwe_cb[event];
+
+	hwe_cb->hwe_handler = hwe_handler;
+	hwe_cb->handle = handle;
+	hwe_cb->hwe_state = HINIC_EQE_ENABLED;
+}
+
+/**
+ * hinic_aeq_unregister_hw_cb - unregister the AEQ callback for specific event
+ * @aeqs: pointer to Async eqs of the chip
+ * @event: aeq event to unregister callback for it
+ **/
+void hinic_aeq_unregister_hw_cb(struct hinic_aeqs *aeqs,
+				enum hinic_aeq_type event)
+{
+	struct hinic_hw_event_cb *hwe_cb = &aeqs->hwe_cb[event];
+
+	hwe_cb->hwe_state &= ~HINIC_EQE_ENABLED;
+
+	while (hwe_cb->hwe_state & HINIC_EQE_RUNNING)
+		schedule();
+
+	hwe_cb->hwe_handler = NULL;
+}
+
+/**
+ * hinic_ceq_register_cb - register CEQ callback for specific event
+ * @ceqs: pointer to Completion eqs part of the chip
+ * @event: ceq event to register callback for it
+ * @handle: private data will be used by the callback
+ * @handler: callback function
+ **/
+void hinic_ceq_register_cb(struct hinic_ceqs *ceqs,
+			   enum hinic_ceq_type event, void *handle,
+			   void (*handler)(void *handle, u32 ceqe_data))
+{
+	struct hinic_ceq_cb *ceq_cb = &ceqs->ceq_cb[event];
+
+	ceq_cb->handler = handler;
+	ceq_cb->handle = handle;
+	ceq_cb->ceqe_state = HINIC_EQE_ENABLED;
+}
+
+/**
+ * hinic_ceq_unregister_cb - unregister the CEQ callback for specific event
+ * @ceqs: pointer to Completion eqs part of the chip
+ * @event: ceq event to unregister callback for it
+ **/
+void hinic_ceq_unregister_cb(struct hinic_ceqs *ceqs,
+			     enum hinic_ceq_type event)
+{
+	struct hinic_ceq_cb *ceq_cb = &ceqs->ceq_cb[event];
+
+	ceq_cb->ceqe_state &= ~HINIC_EQE_ENABLED;
+
+	while (ceq_cb->ceqe_state & HINIC_EQE_RUNNING)
+		schedule();
+
+	ceq_cb->handler = NULL;
+}
+
+static u8 eq_cons_idx_checksum_set(u32 val)
+{
+	u8 checksum = 0;
+	int idx;
+
+	for (idx = 0; idx < 32; idx += 4)
+		checksum ^= ((val >> idx) & 0xF);
+
+	return (checksum & 0xF);
+}
+
+/**
+ * eq_update_ci - update the HW cons idx of event queue
+ * @eq: the event queue to update the cons idx for
+ * @arm_state: the arm bit value of eq's interrupt
+ **/
+static void eq_update_ci(struct hinic_eq *eq, u32 arm_state)
+{
+	u32 val, addr = EQ_CONS_IDX_REG_ADDR(eq);
+
+	/* Read Modify Write */
+	val = hinic_hwif_read_reg(eq->hwif, addr);
+
+	val = HINIC_EQ_CI_CLEAR(val, IDX)       &
+	      HINIC_EQ_CI_CLEAR(val, WRAPPED)   &
+	      HINIC_EQ_CI_CLEAR(val, INT_ARMED) &
+	      HINIC_EQ_CI_CLEAR(val, XOR_CHKSUM);
+
+	val |= HINIC_EQ_CI_SET(eq->cons_idx, IDX)    |
+	       HINIC_EQ_CI_SET(eq->wrapped, WRAPPED) |
+	       HINIC_EQ_CI_SET(arm_state, INT_ARMED);
+
+	val |= HINIC_EQ_CI_SET(eq_cons_idx_checksum_set(val), XOR_CHKSUM);
+
+	hinic_hwif_write_reg(eq->hwif, addr, val);
+}
+
+/**
+ * aeq_irq_handler - handler for the AEQ event
+ * @eq: the Async Event Queue that received the event
+ **/
+static void aeq_irq_handler(struct hinic_eq *eq)
+{
+	struct hinic_aeqs *aeqs = aeq_to_aeqs(eq);
+	struct hinic_hwif *hwif = aeqs->hwif;
+	struct pci_dev *pdev = hwif->pdev;
+	struct hinic_aeq_elem *aeqe_curr;
+	struct hinic_hw_event_cb *hwe_cb;
+	enum hinic_aeq_type event;
+	unsigned long eqe_state;
+	u32 aeqe_desc;
+	int i, size;
+
+	for (i = 0; i < eq->q_len; i++) {
+		aeqe_curr = GET_CURR_AEQ_ELEM(eq);
+
+		/* Data in HW is in Big endian Format */
+		aeqe_desc = be32_to_cpu(aeqe_curr->desc);
+
+		/* HW toggles the wrapped bit, when it adds eq element */
+		if (HINIC_EQ_ELEM_DESC_GET(aeqe_desc, WRAPPED) == eq->wrapped)
+			break;
+
+		dma_rmb();
+
+		event = HINIC_EQ_ELEM_DESC_GET(aeqe_desc, TYPE);
+		if (event >= HINIC_MAX_AEQ_EVENTS) {
+			dev_err(&pdev->dev, "Unknown AEQ Event %d\n", event);
+			return;
+		}
+
+		if (!HINIC_EQ_ELEM_DESC_GET(aeqe_desc, SRC)) {
+			hwe_cb = &aeqs->hwe_cb[event];
+
+			size = HINIC_EQ_ELEM_DESC_GET(aeqe_desc, SIZE);
+
+			eqe_state = cmpxchg(&hwe_cb->hwe_state,
+					    HINIC_EQE_ENABLED,
+					    HINIC_EQE_ENABLED |
+					    HINIC_EQE_RUNNING);
+			if (eqe_state == HINIC_EQE_ENABLED &&
+			    hwe_cb->hwe_handler)
+				hwe_cb->hwe_handler(hwe_cb->handle,
+						    aeqe_curr->data, size);
+			else
+				dev_err(&pdev->dev, "Unhandled AEQ Event %d\n",
+					event);
+
+			hwe_cb->hwe_state &= ~HINIC_EQE_RUNNING;
+		}
+
+		eq->cons_idx++;
+
+		if (eq->cons_idx == eq->q_len) {
+			eq->cons_idx = 0;
+			eq->wrapped = !eq->wrapped;
+		}
+	}
+}
+
+/**
+ * ceq_event_handler - handler for the ceq events
+ * @ceqs: ceqs part of the chip
+ * @ceqe: ceq element that describes the event
+ **/
+static void ceq_event_handler(struct hinic_ceqs *ceqs, u32 ceqe)
+{
+	struct hinic_hwif *hwif = ceqs->hwif;
+	struct pci_dev *pdev = hwif->pdev;
+	struct hinic_ceq_cb *ceq_cb;
+	enum hinic_ceq_type event;
+	unsigned long eqe_state;
+
+	event = CEQE_TYPE(ceqe);
+	if (event >= HINIC_MAX_CEQ_EVENTS) {
+		dev_err(&pdev->dev, "Unknown CEQ event, event = %d\n", event);
+		return;
+	}
+
+	ceq_cb = &ceqs->ceq_cb[event];
+
+	eqe_state = cmpxchg(&ceq_cb->ceqe_state,
+			    HINIC_EQE_ENABLED,
+			    HINIC_EQE_ENABLED | HINIC_EQE_RUNNING);
+
+	if (eqe_state == HINIC_EQE_ENABLED && ceq_cb->handler)
+		ceq_cb->handler(ceq_cb->handle, CEQE_DATA(ceqe));
+	else
+		dev_err(&pdev->dev, "Unhandled CEQ Event %d\n", event);
+
+	ceq_cb->ceqe_state &= ~HINIC_EQE_RUNNING;
+}
+
+/**
+ * ceq_irq_handler - handler for the CEQ event
+ * @eq: the Completion Event Queue that received the event
+ **/
+static void ceq_irq_handler(struct hinic_eq *eq)
+{
+	struct hinic_ceqs *ceqs = ceq_to_ceqs(eq);
+	u32 ceqe;
+	int i;
+
+	for (i = 0; i < eq->q_len; i++) {
+		ceqe = *(GET_CURR_CEQ_ELEM(eq));
+
+		/* Data in HW is in Big endian Format */
+		ceqe = be32_to_cpu(ceqe);
+
+		/* HW toggles the wrapped bit, when it adds eq element event */
+		if (HINIC_EQ_ELEM_DESC_GET(ceqe, WRAPPED) == eq->wrapped)
+			break;
+
+		ceq_event_handler(ceqs, ceqe);
+
+		eq->cons_idx++;
+
+		if (eq->cons_idx == eq->q_len) {
+			eq->cons_idx = 0;
+			eq->wrapped = !eq->wrapped;
+		}
+	}
+}
+
+/**
+ * eq_irq_handler - handler for the EQ event
+ * @data: the Event Queue that received the event
+ **/
+static void eq_irq_handler(void *data)
+{
+	struct hinic_eq *eq = data;
+
+	if (eq->type == HINIC_AEQ)
+		aeq_irq_handler(eq);
+	else if (eq->type == HINIC_CEQ)
+		ceq_irq_handler(eq);
+
+	eq_update_ci(eq, EQ_ARMED);
+}
+
+/**
+ * eq_irq_work - the work of the EQ that received the event
+ * @work: the work struct that is associated with the EQ
+ **/
+static void eq_irq_work(struct work_struct *work)
+{
+	struct hinic_eq_work *aeq_work = work_to_aeq_work(work);
+	struct hinic_eq *aeq;
+
+	aeq = aeq_work->data;
+	eq_irq_handler(aeq);
+}
+
+/**
+ * ceq_tasklet - the tasklet of the EQ that received the event
+ * @t: the tasklet struct pointer
+ **/
+static void ceq_tasklet(struct tasklet_struct *t)
+{
+	struct hinic_eq *ceq = from_tasklet(ceq, t, ceq_tasklet);
+
+	eq_irq_handler(ceq);
+}
+
+/**
+ * aeq_interrupt - aeq interrupt handler
+ * @irq: irq number
+ * @data: the Async Event Queue that collected the event
+ **/
+static irqreturn_t aeq_interrupt(int irq, void *data)
+{
+	struct hinic_eq_work *aeq_work;
+	struct hinic_eq *aeq = data;
+	struct hinic_aeqs *aeqs;
+
+	/* clear resend timer cnt register */
+	hinic_msix_attr_cnt_clear(aeq->hwif, aeq->msix_entry.entry);
+
+	aeq_work = &aeq->aeq_work;
+	aeq_work->data = aeq;
+
+	aeqs = aeq_to_aeqs(aeq);
+	queue_work(aeqs->workq, &aeq_work->work);
+
+	return IRQ_HANDLED;
+}
+
+/**
+ * ceq_interrupt - ceq interrupt handler
+ * @irq: irq number
+ * @data: the Completion Event Queue that collected the event
+ **/
+static irqreturn_t ceq_interrupt(int irq, void *data)
+{
+	struct hinic_eq *ceq = data;
+
+	/* clear resend timer cnt register */
+	hinic_msix_attr_cnt_clear(ceq->hwif, ceq->msix_entry.entry);
+
+	tasklet_schedule(&ceq->ceq_tasklet);
+
+	return IRQ_HANDLED;
+}
+
+static u32 get_ctrl0_val(struct hinic_eq *eq, u32 addr)
+{
+	struct msix_entry *msix_entry = &eq->msix_entry;
+	enum hinic_eq_type type = eq->type;
+	u32 val, ctrl0;
+
+	if (type == HINIC_AEQ) {
+		/* RMW Ctrl0 */
+		addr = HINIC_CSR_AEQ_CTRL_0_ADDR(eq->q_id);
+
+		val = hinic_hwif_read_reg(eq->hwif, addr);
+
+		val = HINIC_AEQ_CTRL_0_CLEAR(val, INT_IDX)      &
+		      HINIC_AEQ_CTRL_0_CLEAR(val, DMA_ATTR)     &
+		      HINIC_AEQ_CTRL_0_CLEAR(val, PCI_INTF_IDX) &
+		      HINIC_AEQ_CTRL_0_CLEAR(val, INT_MODE);
+
+		ctrl0 = HINIC_AEQ_CTRL_0_SET(msix_entry->entry, INT_IDX)     |
+			HINIC_AEQ_CTRL_0_SET(DMA_ATTR_AEQ_DEFAULT, DMA_ATTR) |
+			HINIC_AEQ_CTRL_0_SET(HINIC_HWIF_PCI_INTF(eq->hwif),
+					     PCI_INTF_IDX)                   |
+			HINIC_AEQ_CTRL_0_SET(EQ_INT_MODE_ARMED, INT_MODE);
+
+		val |= ctrl0;
+	} else {
+		/* RMW Ctrl0 */
+		addr = HINIC_CSR_CEQ_CTRL_0_ADDR(eq->q_id);
+
+		val = hinic_hwif_read_reg(eq->hwif, addr);
+
+		val = HINIC_CEQ_CTRL_0_CLEAR(val, INTR_IDX)     &
+		      HINIC_CEQ_CTRL_0_CLEAR(val, DMA_ATTR)     &
+		      HINIC_CEQ_CTRL_0_CLEAR(val, KICK_THRESH)  &
+		      HINIC_CEQ_CTRL_0_CLEAR(val, PCI_INTF_IDX) &
+		      HINIC_CEQ_CTRL_0_CLEAR(val, INTR_MODE);
+
+		ctrl0 = HINIC_CEQ_CTRL_0_SET(msix_entry->entry, INTR_IDX)     |
+			HINIC_CEQ_CTRL_0_SET(DMA_ATTR_CEQ_DEFAULT, DMA_ATTR)  |
+			HINIC_CEQ_CTRL_0_SET(THRESH_CEQ_DEFAULT, KICK_THRESH) |
+			HINIC_CEQ_CTRL_0_SET(HINIC_HWIF_PCI_INTF(eq->hwif),
+					     PCI_INTF_IDX)                    |
+			HINIC_CEQ_CTRL_0_SET(EQ_INT_MODE_ARMED, INTR_MODE);
+
+		val |= ctrl0;
+	}
+	return val;
+}
+
+static void set_ctrl0(struct hinic_eq *eq)
+{
+	u32 val, addr;
+
+	if (eq->type == HINIC_AEQ)
+		addr = HINIC_CSR_AEQ_CTRL_0_ADDR(eq->q_id);
+	else
+		addr = HINIC_CSR_CEQ_CTRL_0_ADDR(eq->q_id);
+
+	val = get_ctrl0_val(eq, addr);
+
+	hinic_hwif_write_reg(eq->hwif, addr, val);
+}
+
+static u32 get_ctrl1_val(struct hinic_eq *eq, u32 addr)
+{
+	u32 page_size_val, elem_size, val, ctrl1;
+	enum hinic_eq_type type = eq->type;
+
+	if (type == HINIC_AEQ) {
+		/* RMW Ctrl1 */
+		addr = HINIC_CSR_AEQ_CTRL_1_ADDR(eq->q_id);
+
+		page_size_val = EQ_SET_HW_PAGE_SIZE_VAL(eq);
+		elem_size = EQ_SET_HW_ELEM_SIZE_VAL(eq);
+
+		val = hinic_hwif_read_reg(eq->hwif, addr);
+
+		val = HINIC_AEQ_CTRL_1_CLEAR(val, LEN)          &
+		      HINIC_AEQ_CTRL_1_CLEAR(val, ELEM_SIZE)    &
+		      HINIC_AEQ_CTRL_1_CLEAR(val, PAGE_SIZE);
+
+		ctrl1 = HINIC_AEQ_CTRL_1_SET(eq->q_len, LEN)            |
+			HINIC_AEQ_CTRL_1_SET(elem_size, ELEM_SIZE)      |
+			HINIC_AEQ_CTRL_1_SET(page_size_val, PAGE_SIZE);
+
+		val |= ctrl1;
+	} else {
+		/* RMW Ctrl1 */
+		addr = HINIC_CSR_CEQ_CTRL_1_ADDR(eq->q_id);
+
+		page_size_val = EQ_SET_HW_PAGE_SIZE_VAL(eq);
+
+		val = hinic_hwif_read_reg(eq->hwif, addr);
+
+		val = HINIC_CEQ_CTRL_1_CLEAR(val, LEN) &
+		      HINIC_CEQ_CTRL_1_CLEAR(val, PAGE_SIZE);
+
+		ctrl1 = HINIC_CEQ_CTRL_1_SET(eq->q_len, LEN) |
+			HINIC_CEQ_CTRL_1_SET(page_size_val, PAGE_SIZE);
+
+		val |= ctrl1;
+	}
+	return val;
+}
+
+static void set_ctrl1(struct hinic_eq *eq)
+{
+	u32 addr, val;
+
+	if (eq->type == HINIC_AEQ)
+		addr = HINIC_CSR_AEQ_CTRL_1_ADDR(eq->q_id);
+	else
+		addr = HINIC_CSR_CEQ_CTRL_1_ADDR(eq->q_id);
+
+	val = get_ctrl1_val(eq, addr);
+
+	hinic_hwif_write_reg(eq->hwif, addr, val);
+}
+
+static int set_ceq_ctrl_reg(struct hinic_eq *eq)
+{
+	struct hinic_ceq_ctrl_reg ceq_ctrl = {0};
+	struct hinic_hwdev *hwdev = eq->hwdev;
+	u16 out_size = sizeof(ceq_ctrl);
+	u16 in_size = sizeof(ceq_ctrl);
+	struct hinic_pfhwdev *pfhwdev;
+	u32 addr;
+	int err;
+
+	pfhwdev = container_of(hwdev, struct hinic_pfhwdev, hwdev);
+
+	addr = HINIC_CSR_CEQ_CTRL_0_ADDR(eq->q_id);
+	ceq_ctrl.ctrl0 = get_ctrl0_val(eq, addr);
+	addr = HINIC_CSR_CEQ_CTRL_1_ADDR(eq->q_id);
+	ceq_ctrl.ctrl1 = get_ctrl1_val(eq, addr);
+
+	ceq_ctrl.func_id = HINIC_HWIF_FUNC_IDX(hwdev->hwif);
+	ceq_ctrl.q_id = eq->q_id;
+
+	err = hinic_msg_to_mgmt(&pfhwdev->pf_to_mgmt, HINIC_MOD_COMM,
+				HINIC_COMM_CMD_CEQ_CTRL_REG_WR_BY_UP,
+				&ceq_ctrl, in_size,
+				&ceq_ctrl, &out_size, HINIC_MGMT_MSG_SYNC);
+	if (err || !out_size || ceq_ctrl.status) {
+		dev_err(&hwdev->hwif->pdev->dev,
+			"Failed to set ceq %d ctrl reg, err: %d status: 0x%x, out_size: 0x%x\n",
+			eq->q_id, err, ceq_ctrl.status, out_size);
+		return -EFAULT;
+	}
+
+	return 0;
+}
+
+/**
+ * set_eq_ctrls - setting eq's ctrl registers
+ * @eq: the Event Queue for setting
+ **/
+static int set_eq_ctrls(struct hinic_eq *eq)
+{
+	if (HINIC_IS_VF(eq->hwif) && eq->type == HINIC_CEQ)
+		return set_ceq_ctrl_reg(eq);
+
+	set_ctrl0(eq);
+	set_ctrl1(eq);
+	return 0;
+}
+
+/**
+ * aeq_elements_init - initialize all the elements in the aeq
+ * @eq: the Async Event Queue
+ * @init_val: value to initialize the elements with it
+ **/
+static void aeq_elements_init(struct hinic_eq *eq, u32 init_val)
+{
+	struct hinic_aeq_elem *aeqe;
+	int i;
+
+	for (i = 0; i < eq->q_len; i++) {
+		aeqe = GET_AEQ_ELEM(eq, i);
+		aeqe->desc = cpu_to_be32(init_val);
+	}
+
+	wmb();  /* Write the initilzation values */
+}
+
+/**
+ * ceq_elements_init - Initialize all the elements in the ceq
+ * @eq: the event queue
+ * @init_val: value to init the elements with
+ **/
+static void ceq_elements_init(struct hinic_eq *eq, u32 init_val)
+{
+	u32 *ceqe;
+	int i;
+
+	for (i = 0; i < eq->q_len; i++) {
+		ceqe = GET_CEQ_ELEM(eq, i);
+		*(ceqe) = cpu_to_be32(init_val);
+	}
+
+	wmb();  /* Write the initilzation values */
+}
+
+/**
+ * alloc_eq_pages - allocate the pages for the queue
+ * @eq: the event queue
+ *
+ * Return 0 - Success, Negative - Failure
+ **/
+static int alloc_eq_pages(struct hinic_eq *eq)
+{
+	struct hinic_hwif *hwif = eq->hwif;
+	struct pci_dev *pdev = hwif->pdev;
+	u32 init_val, addr, val;
+	int err, pg;
+
+	eq->dma_addr = devm_kcalloc(&pdev->dev, eq->num_pages,
+				    sizeof(*eq->dma_addr), GFP_KERNEL);
+	if (!eq->dma_addr)
+		return -ENOMEM;
+
+	eq->virt_addr = devm_kcalloc(&pdev->dev, eq->num_pages,
+				     sizeof(*eq->virt_addr), GFP_KERNEL);
+	if (!eq->virt_addr) {
+		err = -ENOMEM;
+		goto err_virt_addr_alloc;
+	}
+
+	for (pg = 0; pg < eq->num_pages; pg++) {
+		eq->virt_addr[pg] = dma_alloc_coherent(&pdev->dev,
+						       eq->page_size,
+						       &eq->dma_addr[pg],
+						       GFP_KERNEL);
+		if (!eq->virt_addr[pg]) {
+			err = -ENOMEM;
+			goto err_dma_alloc;
+		}
+
+		addr = EQ_HI_PHYS_ADDR_REG(eq, pg);
+		val = upper_32_bits(eq->dma_addr[pg]);
+
+		hinic_hwif_write_reg(hwif, addr, val);
+
+		addr = EQ_LO_PHYS_ADDR_REG(eq, pg);
+		val = lower_32_bits(eq->dma_addr[pg]);
+
+		hinic_hwif_write_reg(hwif, addr, val);
+	}
+
+	init_val = HINIC_EQ_ELEM_DESC_SET(eq->wrapped, WRAPPED);
+
+	if (eq->type == HINIC_AEQ)
+		aeq_elements_init(eq, init_val);
+	else if (eq->type == HINIC_CEQ)
+		ceq_elements_init(eq, init_val);
+
+	return 0;
+
+err_dma_alloc:
+	while (--pg >= 0)
+		dma_free_coherent(&pdev->dev, eq->page_size,
+				  eq->virt_addr[pg],
+				  eq->dma_addr[pg]);
+
+	devm_kfree(&pdev->dev, eq->virt_addr);
+
+err_virt_addr_alloc:
+	devm_kfree(&pdev->dev, eq->dma_addr);
+	return err;
+}
+
+/**
+ * free_eq_pages - free the pages of the queue
+ * @eq: the Event Queue
+ **/
+static void free_eq_pages(struct hinic_eq *eq)
+{
+	struct hinic_hwif *hwif = eq->hwif;
+	struct pci_dev *pdev = hwif->pdev;
+	int pg;
+
+	for (pg = 0; pg < eq->num_pages; pg++)
+		dma_free_coherent(&pdev->dev, eq->page_size,
+				  eq->virt_addr[pg],
+				  eq->dma_addr[pg]);
+
+	devm_kfree(&pdev->dev, eq->virt_addr);
+	devm_kfree(&pdev->dev, eq->dma_addr);
+}
+
+/**
+ * init_eq - initialize Event Queue
+ * @eq: the event queue
+ * @hwif: the HW interface of a PCI function device
+ * @type: the type of the event queue, aeq or ceq
+ * @q_id: Queue id number
+ * @q_len: the number of EQ elements
+ * @page_size: the page size of the pages in the event queue
+ * @entry: msix entry associated with the event queue
+ *
+ * Return 0 - Success, Negative - Failure
+ **/
+static int init_eq(struct hinic_eq *eq, struct hinic_hwif *hwif,
+		   enum hinic_eq_type type, int q_id, u32 q_len, u32 page_size,
+		   struct msix_entry entry)
+{
+	struct pci_dev *pdev = hwif->pdev;
+	int err;
+
+	eq->hwif = hwif;
+	eq->type = type;
+	eq->q_id = q_id;
+	eq->q_len = q_len;
+	eq->page_size = page_size;
+
+	/* Clear PI and CI, also clear the ARM bit */
+	hinic_hwif_write_reg(eq->hwif, EQ_CONS_IDX_REG_ADDR(eq), 0);
+	hinic_hwif_write_reg(eq->hwif, EQ_PROD_IDX_REG_ADDR(eq), 0);
+
+	eq->cons_idx = 0;
+	eq->wrapped = 0;
+
+	if (type == HINIC_AEQ) {
+		eq->elem_size = HINIC_AEQE_SIZE;
+	} else if (type == HINIC_CEQ) {
+		eq->elem_size = HINIC_CEQE_SIZE;
+	} else {
+		dev_err(&pdev->dev, "Invalid EQ type\n");
+		return -EINVAL;
+	}
+
+	eq->num_pages = GET_EQ_NUM_PAGES(eq, page_size);
+	eq->num_elem_in_pg = GET_EQ_NUM_ELEMS_IN_PG(eq, page_size);
+
+	eq->msix_entry = entry;
+
+	if (eq->num_elem_in_pg & (eq->num_elem_in_pg - 1)) {
+		dev_err(&pdev->dev, "num elements in eq page != power of 2\n");
+		return -EINVAL;
+	}
+
+	if (eq->num_pages > EQ_MAX_PAGES) {
+		dev_err(&pdev->dev, "too many pages for eq\n");
+		return -EINVAL;
+	}
+
+	err = set_eq_ctrls(eq);
+	if (err) {
+		dev_err(&pdev->dev, "Failed to set eq ctrls\n");
+		return err;
+	}
+
+	eq_update_ci(eq, EQ_ARMED);
+
+	err = alloc_eq_pages(eq);
+	if (err) {
+		dev_err(&pdev->dev, "Failed to allocate pages for eq\n");
+		return err;
+	}
+
+	if (type == HINIC_AEQ) {
+		struct hinic_eq_work *aeq_work = &eq->aeq_work;
+
+		INIT_WORK(&aeq_work->work, eq_irq_work);
+	} else if (type == HINIC_CEQ) {
+		tasklet_setup(&eq->ceq_tasklet, ceq_tasklet);
+	}
+
+	/* set the attributes of the msix entry */
+	hinic_msix_attr_set(eq->hwif, eq->msix_entry.entry,
+			    HINIC_EQ_MSIX_PENDING_LIMIT_DEFAULT,
+			    HINIC_EQ_MSIX_COALESC_TIMER_DEFAULT,
+			    HINIC_EQ_MSIX_LLI_TIMER_DEFAULT,
+			    HINIC_EQ_MSIX_LLI_CREDIT_LIMIT_DEFAULT,
+			    HINIC_EQ_MSIX_RESEND_TIMER_DEFAULT);
+
+	if (type == HINIC_AEQ) {
+		snprintf(eq->irq_name, sizeof(eq->irq_name), "hinic_aeq%d@pci:%s", eq->q_id,
+			 pci_name(pdev));
+		err = request_irq(entry.vector, aeq_interrupt, 0, eq->irq_name, eq);
+	} else if (type == HINIC_CEQ) {
+		snprintf(eq->irq_name, sizeof(eq->irq_name), "hinic_ceq%d@pci:%s", eq->q_id,
+			 pci_name(pdev));
+		err = request_irq(entry.vector, ceq_interrupt, 0, eq->irq_name, eq);
+	}
+
+	if (err) {
+		dev_err(&pdev->dev, "Failed to request irq for the EQ\n");
+		goto err_req_irq;
+	}
+
+	return 0;
+
+err_req_irq:
+	free_eq_pages(eq);
+	return err;
+}
+
+/**
+ * remove_eq - remove Event Queue
+ * @eq: the event queue
+ **/
+static void remove_eq(struct hinic_eq *eq)
+{
+	hinic_set_msix_state(eq->hwif, eq->msix_entry.entry,
+			     HINIC_MSIX_DISABLE);
+	free_irq(eq->msix_entry.vector, eq);
+
+	if (eq->type == HINIC_AEQ) {
+		struct hinic_eq_work *aeq_work = &eq->aeq_work;
+
+		cancel_work_sync(&aeq_work->work);
+		/* clear aeq_len to avoid hw access host memory */
+		hinic_hwif_write_reg(eq->hwif,
+				     HINIC_CSR_AEQ_CTRL_1_ADDR(eq->q_id), 0);
+	} else if (eq->type == HINIC_CEQ) {
+		tasklet_kill(&eq->ceq_tasklet);
+		/* clear ceq_len to avoid hw access host memory */
+		hinic_hwif_write_reg(eq->hwif,
+				     HINIC_CSR_CEQ_CTRL_1_ADDR(eq->q_id), 0);
+	}
+
+	/* update cons_idx to avoid invalid interrupt */
+	eq->cons_idx = hinic_hwif_read_reg(eq->hwif, EQ_PROD_IDX_REG_ADDR(eq));
+	eq_update_ci(eq, EQ_NOT_ARMED);
+
+	free_eq_pages(eq);
+}
+
+/**
+ * hinic_aeqs_init - initialize all the aeqs
+ * @aeqs: pointer to Async eqs of the chip
+ * @hwif: the HW interface of a PCI function device
+ * @num_aeqs: number of AEQs
+ * @q_len: number of EQ elements
+ * @page_size: the page size of the pages in the event queue
+ * @msix_entries: msix entries associated with the event queues
+ *
+ * Return 0 - Success, negative - Failure
+ **/
+int hinic_aeqs_init(struct hinic_aeqs *aeqs, struct hinic_hwif *hwif,
+		    int num_aeqs, u32 q_len, u32 page_size,
+		    struct msix_entry *msix_entries)
+{
+	struct pci_dev *pdev = hwif->pdev;
+	int err, i, q_id;
+
+	aeqs->workq = create_singlethread_workqueue(HINIC_EQS_WQ_NAME);
+	if (!aeqs->workq)
+		return -ENOMEM;
+
+	aeqs->hwif = hwif;
+	aeqs->num_aeqs = num_aeqs;
+
+	for (q_id = 0; q_id < num_aeqs; q_id++) {
+		err = init_eq(&aeqs->aeq[q_id], hwif, HINIC_AEQ, q_id, q_len,
+			      page_size, msix_entries[q_id]);
+		if (err) {
+			dev_err(&pdev->dev, "Failed to init aeq %d\n", q_id);
+			goto err_init_aeq;
+		}
+	}
+
+	return 0;
+
+err_init_aeq:
+	for (i = 0; i < q_id; i++)
+		remove_eq(&aeqs->aeq[i]);
+
+	destroy_workqueue(aeqs->workq);
+	return err;
+}
+
+/**
+ * hinic_aeqs_free - free all the aeqs
+ * @aeqs: pointer to Async eqs of the chip
+ **/
+void hinic_aeqs_free(struct hinic_aeqs *aeqs)
+{
+	int q_id;
+
+	for (q_id = 0; q_id < aeqs->num_aeqs ; q_id++)
+		remove_eq(&aeqs->aeq[q_id]);
+
+	destroy_workqueue(aeqs->workq);
+}
+
+/**
+ * hinic_ceqs_init - init all the ceqs
+ * @ceqs: ceqs part of the chip
+ * @hwif: the hardware interface of a pci function device
+ * @num_ceqs: number of CEQs
+ * @q_len: number of EQ elements
+ * @page_size: the page size of the event queue
+ * @msix_entries: msix entries associated with the event queues
+ *
+ * Return 0 - Success, Negative - Failure
+ **/
+int hinic_ceqs_init(struct hinic_ceqs *ceqs, struct hinic_hwif *hwif,
+		    int num_ceqs, u32 q_len, u32 page_size,
+		    struct msix_entry *msix_entries)
+{
+	struct pci_dev *pdev = hwif->pdev;
+	int i, q_id, err;
+
+	ceqs->hwif = hwif;
+	ceqs->num_ceqs = num_ceqs;
+
+	for (q_id = 0; q_id < num_ceqs; q_id++) {
+		ceqs->ceq[q_id].hwdev = ceqs->hwdev;
+		err = init_eq(&ceqs->ceq[q_id], hwif, HINIC_CEQ, q_id, q_len,
+			      page_size, msix_entries[q_id]);
+		if (err) {
+			dev_err(&pdev->dev, "Failed to init ceq %d\n", q_id);
+			goto err_init_ceq;
+		}
+	}
+
+	return 0;
+
+err_init_ceq:
+	for (i = 0; i < q_id; i++)
+		remove_eq(&ceqs->ceq[i]);
+
+	return err;
+}
+
+/**
+ * hinic_ceqs_free - free all the ceqs
+ * @ceqs: ceqs part of the chip
+ **/
+void hinic_ceqs_free(struct hinic_ceqs *ceqs)
+{
+	int q_id;
+
+	for (q_id = 0; q_id < ceqs->num_ceqs; q_id++)
+		remove_eq(&ceqs->ceq[q_id]);
+}
+
+void hinic_dump_ceq_info(struct hinic_hwdev *hwdev)
+{
+	struct hinic_eq *eq = NULL;
+	u32 addr, ci, pi;
+	int q_id;
+
+	for (q_id = 0; q_id < hwdev->func_to_io.ceqs.num_ceqs; q_id++) {
+		eq = &hwdev->func_to_io.ceqs.ceq[q_id];
+		addr = EQ_CONS_IDX_REG_ADDR(eq);
+		ci = hinic_hwif_read_reg(hwdev->hwif, addr);
+		addr = EQ_PROD_IDX_REG_ADDR(eq);
+		pi = hinic_hwif_read_reg(hwdev->hwif, addr);
+		dev_err(&hwdev->hwif->pdev->dev, "Ceq id: %d, ci: 0x%08x, sw_ci: 0x%08x, pi: 0x%x, tasklet_state: 0x%lx, wrap: %d, ceqe: 0x%x\n",
+			q_id, ci, eq->cons_idx, pi,
+			eq->ceq_tasklet.state,
+			eq->wrapped, be32_to_cpu(*(__be32 *)(GET_CURR_CEQ_ELEM(eq))));
+	}
+}
+
+void hinic_dump_aeq_info(struct hinic_hwdev *hwdev)
+{
+	struct hinic_aeq_elem *aeqe_pos = NULL;
+	struct hinic_eq *eq = NULL;
+	u32 addr, ci, pi;
+	int q_id;
+
+	for (q_id = 0; q_id < hwdev->aeqs.num_aeqs; q_id++) {
+		eq = &hwdev->aeqs.aeq[q_id];
+		addr = EQ_CONS_IDX_REG_ADDR(eq);
+		ci = hinic_hwif_read_reg(hwdev->hwif, addr);
+		addr = EQ_PROD_IDX_REG_ADDR(eq);
+		pi = hinic_hwif_read_reg(hwdev->hwif, addr);
+		aeqe_pos = GET_CURR_AEQ_ELEM(eq);
+		dev_err(&hwdev->hwif->pdev->dev, "Aeq id: %d, ci: 0x%08x, pi: 0x%x, work_state: 0x%x, wrap: %d, desc: 0x%x\n",
+			q_id, ci, pi, work_busy(&eq->aeq_work.work),
+			eq->wrapped, be32_to_cpu(aeqe_pos->desc));
+	}
+}
diff --git a/drivers/net/ethernet/huawei/hinic/hinic_hw_eqs.h b/drivers/net/ethernet/huawei/hinic/hinic_hw_eqs.h
new file mode 100644
index 000000000000..2f3222174fc7
--- /dev/null
+++ b/drivers/net/ethernet/huawei/hinic/hinic_hw_eqs.h
@@ -0,0 +1,262 @@
+/* SPDX-License-Identifier: GPL-2.0-only */
+/*
+ * Huawei HiNIC PCI Express Linux driver
+ * Copyright(c) 2017 Huawei Technologies Co., Ltd
+ */
+
+#ifndef HINIC_HW_EQS_H
+#define HINIC_HW_EQS_H
+
+#include <linux/types.h>
+#include <linux/workqueue.h>
+#include <linux/pci.h>
+#include <linux/sizes.h>
+#include <linux/bitops.h>
+#include <linux/interrupt.h>
+
+#include "hinic_hw_if.h"
+
+#define HINIC_AEQ_CTRL_0_INT_IDX_SHIFT          0
+#define HINIC_AEQ_CTRL_0_DMA_ATTR_SHIFT         12
+#define HINIC_AEQ_CTRL_0_PCI_INTF_IDX_SHIFT     20
+#define HINIC_AEQ_CTRL_0_INT_MODE_SHIFT         31
+
+#define HINIC_AEQ_CTRL_0_INT_IDX_MASK           0x3FF
+#define HINIC_AEQ_CTRL_0_DMA_ATTR_MASK          0x3F
+#define HINIC_AEQ_CTRL_0_PCI_INTF_IDX_MASK      0x3
+#define HINIC_AEQ_CTRL_0_INT_MODE_MASK          0x1
+
+#define HINIC_AEQ_CTRL_0_SET(val, member)       \
+			(((u32)(val) & HINIC_AEQ_CTRL_0_##member##_MASK) << \
+			 HINIC_AEQ_CTRL_0_##member##_SHIFT)
+
+#define HINIC_AEQ_CTRL_0_CLEAR(val, member)     \
+			((val) & (~(HINIC_AEQ_CTRL_0_##member##_MASK \
+			 << HINIC_AEQ_CTRL_0_##member##_SHIFT)))
+
+#define HINIC_AEQ_CTRL_1_LEN_SHIFT              0
+#define HINIC_AEQ_CTRL_1_ELEM_SIZE_SHIFT        24
+#define HINIC_AEQ_CTRL_1_PAGE_SIZE_SHIFT        28
+
+#define HINIC_AEQ_CTRL_1_LEN_MASK               0x1FFFFF
+#define HINIC_AEQ_CTRL_1_ELEM_SIZE_MASK         0x3
+#define HINIC_AEQ_CTRL_1_PAGE_SIZE_MASK         0xF
+
+#define HINIC_AEQ_CTRL_1_SET(val, member)       \
+			(((u32)(val) & HINIC_AEQ_CTRL_1_##member##_MASK) << \
+			 HINIC_AEQ_CTRL_1_##member##_SHIFT)
+
+#define HINIC_AEQ_CTRL_1_CLEAR(val, member)     \
+			((val) & (~(HINIC_AEQ_CTRL_1_##member##_MASK \
+			 << HINIC_AEQ_CTRL_1_##member##_SHIFT)))
+
+#define HINIC_CEQ_CTRL_0_INTR_IDX_SHIFT         0
+#define HINIC_CEQ_CTRL_0_DMA_ATTR_SHIFT         12
+#define HINIC_CEQ_CTRL_0_KICK_THRESH_SHIFT      20
+#define HINIC_CEQ_CTRL_0_PCI_INTF_IDX_SHIFT     24
+#define HINIC_CEQ_CTRL_0_INTR_MODE_SHIFT        31
+
+#define HINIC_CEQ_CTRL_0_INTR_IDX_MASK          0x3FF
+#define HINIC_CEQ_CTRL_0_DMA_ATTR_MASK          0x3F
+#define HINIC_CEQ_CTRL_0_KICK_THRESH_MASK       0xF
+#define HINIC_CEQ_CTRL_0_PCI_INTF_IDX_MASK      0x3
+#define HINIC_CEQ_CTRL_0_INTR_MODE_MASK         0x1
+
+#define HINIC_CEQ_CTRL_0_SET(val, member)       \
+			(((u32)(val) & HINIC_CEQ_CTRL_0_##member##_MASK) << \
+			 HINIC_CEQ_CTRL_0_##member##_SHIFT)
+
+#define HINIC_CEQ_CTRL_0_CLEAR(val, member)     \
+			((val) & (~(HINIC_CEQ_CTRL_0_##member##_MASK \
+			 << HINIC_CEQ_CTRL_0_##member##_SHIFT)))
+
+#define HINIC_CEQ_CTRL_1_LEN_SHIFT              0
+#define HINIC_CEQ_CTRL_1_PAGE_SIZE_SHIFT        28
+
+#define HINIC_CEQ_CTRL_1_LEN_MASK               0x1FFFFF
+#define HINIC_CEQ_CTRL_1_PAGE_SIZE_MASK         0xF
+
+#define HINIC_CEQ_CTRL_1_SET(val, member)       \
+			(((u32)(val) & HINIC_CEQ_CTRL_1_##member##_MASK) << \
+			 HINIC_CEQ_CTRL_1_##member##_SHIFT)
+
+#define HINIC_CEQ_CTRL_1_CLEAR(val, member)     \
+			((val) & (~(HINIC_CEQ_CTRL_1_##member##_MASK \
+			 << HINIC_CEQ_CTRL_1_##member##_SHIFT)))
+
+#define HINIC_EQ_ELEM_DESC_TYPE_SHIFT           0
+#define HINIC_EQ_ELEM_DESC_SRC_SHIFT            7
+#define HINIC_EQ_ELEM_DESC_SIZE_SHIFT           8
+#define HINIC_EQ_ELEM_DESC_WRAPPED_SHIFT        31
+
+#define HINIC_EQ_ELEM_DESC_TYPE_MASK            0x7F
+#define HINIC_EQ_ELEM_DESC_SRC_MASK             0x1
+#define HINIC_EQ_ELEM_DESC_SIZE_MASK            0xFF
+#define HINIC_EQ_ELEM_DESC_WRAPPED_MASK         0x1
+
+#define HINIC_EQ_ELEM_DESC_SET(val, member)     \
+			(((u32)(val) & HINIC_EQ_ELEM_DESC_##member##_MASK) << \
+			 HINIC_EQ_ELEM_DESC_##member##_SHIFT)
+
+#define HINIC_EQ_ELEM_DESC_GET(val, member)     \
+			(((val) >> HINIC_EQ_ELEM_DESC_##member##_SHIFT) & \
+			 HINIC_EQ_ELEM_DESC_##member##_MASK)
+
+#define HINIC_EQ_CI_IDX_SHIFT                   0
+#define HINIC_EQ_CI_WRAPPED_SHIFT               20
+#define HINIC_EQ_CI_XOR_CHKSUM_SHIFT            24
+#define HINIC_EQ_CI_INT_ARMED_SHIFT             31
+
+#define HINIC_EQ_CI_IDX_MASK                    0xFFFFF
+#define HINIC_EQ_CI_WRAPPED_MASK                0x1
+#define HINIC_EQ_CI_XOR_CHKSUM_MASK             0xF
+#define HINIC_EQ_CI_INT_ARMED_MASK              0x1
+
+#define HINIC_EQ_CI_SET(val, member)            \
+			(((u32)(val) & HINIC_EQ_CI_##member##_MASK) << \
+			 HINIC_EQ_CI_##member##_SHIFT)
+
+#define HINIC_EQ_CI_CLEAR(val, member)          \
+			((val) & (~(HINIC_EQ_CI_##member##_MASK \
+			 << HINIC_EQ_CI_##member##_SHIFT)))
+
+#define HINIC_MAX_AEQS                  4
+#define HINIC_MAX_CEQS                  32
+
+#define HINIC_AEQE_SIZE                 64
+#define HINIC_CEQE_SIZE                 4
+
+#define HINIC_AEQE_DESC_SIZE            4
+#define HINIC_AEQE_DATA_SIZE            \
+			(HINIC_AEQE_SIZE - HINIC_AEQE_DESC_SIZE)
+
+#define HINIC_DEFAULT_AEQ_LEN           64
+#define HINIC_DEFAULT_CEQ_LEN           1024
+
+#define HINIC_EQ_PAGE_SIZE              SZ_4K
+
+#define HINIC_CEQ_ID_CMDQ               0
+
+enum hinic_eq_type {
+	HINIC_AEQ,
+	HINIC_CEQ,
+};
+
+enum hinic_aeq_type {
+	HINIC_MBX_FROM_FUNC = 1,
+	HINIC_MSG_FROM_MGMT_CPU = 2,
+	HINIC_MBX_SEND_RSLT = 5,
+	HINIC_MAX_AEQ_EVENTS,
+};
+
+enum hinic_ceq_type {
+	HINIC_CEQ_CMDQ = 3,
+
+	HINIC_MAX_CEQ_EVENTS,
+};
+
+enum hinic_eqe_state {
+	HINIC_EQE_ENABLED = BIT(0),
+	HINIC_EQE_RUNNING = BIT(1),
+};
+
+struct hinic_aeq_elem {
+	u8      data[HINIC_AEQE_DATA_SIZE];
+	__be32  desc;
+};
+
+struct hinic_eq_work {
+	struct work_struct      work;
+	void                    *data;
+};
+
+struct hinic_eq {
+	struct hinic_hwif       *hwif;
+	struct hinic_hwdev      *hwdev;
+	enum hinic_eq_type      type;
+	int                     q_id;
+	u32                     q_len;
+	u32                     page_size;
+
+	u32                     cons_idx;
+	int                     wrapped;
+
+	size_t                  elem_size;
+	int                     num_pages;
+	int                     num_elem_in_pg;
+
+	struct msix_entry       msix_entry;
+	char			irq_name[64];
+
+	dma_addr_t              *dma_addr;
+	void                    **virt_addr;
+
+	struct hinic_eq_work    aeq_work;
+
+	struct tasklet_struct   ceq_tasklet;
+};
+
+struct hinic_hw_event_cb {
+	void    (*hwe_handler)(void *handle, void *data, u8 size);
+	void                    *handle;
+	unsigned long           hwe_state;
+};
+
+struct hinic_aeqs {
+	struct hinic_hwif       *hwif;
+
+	struct hinic_eq         aeq[HINIC_MAX_AEQS];
+	int                     num_aeqs;
+
+	struct hinic_hw_event_cb hwe_cb[HINIC_MAX_AEQ_EVENTS];
+
+	struct workqueue_struct *workq;
+};
+
+struct hinic_ceq_cb {
+	void    (*handler)(void *handle, u32 ceqe_data);
+	void                    *handle;
+	enum hinic_eqe_state    ceqe_state;
+};
+
+struct hinic_ceqs {
+	struct hinic_hwif       *hwif;
+	struct hinic_hwdev		*hwdev;
+	struct hinic_eq         ceq[HINIC_MAX_CEQS];
+	int                     num_ceqs;
+
+	struct hinic_ceq_cb     ceq_cb[HINIC_MAX_CEQ_EVENTS];
+};
+
+void hinic_aeq_register_hw_cb(struct hinic_aeqs *aeqs,
+			      enum hinic_aeq_type event, void *handle,
+			      void (*hwe_handler)(void *handle, void *data,
+						  u8 size));
+
+void hinic_aeq_unregister_hw_cb(struct hinic_aeqs *aeqs,
+				enum hinic_aeq_type event);
+
+void hinic_ceq_register_cb(struct hinic_ceqs *ceqs,
+			   enum hinic_ceq_type event, void *handle,
+			   void (*ceq_cb)(void *handle, u32 ceqe_data));
+
+void hinic_ceq_unregister_cb(struct hinic_ceqs *ceqs,
+			     enum hinic_ceq_type event);
+
+int hinic_aeqs_init(struct hinic_aeqs *aeqs, struct hinic_hwif *hwif,
+		    int num_aeqs, u32 q_len, u32 page_size,
+		    struct msix_entry *msix_entries);
+
+void hinic_aeqs_free(struct hinic_aeqs *aeqs);
+
+int hinic_ceqs_init(struct hinic_ceqs *ceqs, struct hinic_hwif *hwif,
+		    int num_ceqs, u32 q_len, u32 page_size,
+		    struct msix_entry *msix_entries);
+
+void hinic_ceqs_free(struct hinic_ceqs *ceqs);
+
+void hinic_dump_ceq_info(struct hinic_hwdev *hwdev);
+
+void hinic_dump_aeq_info(struct hinic_hwdev *hwdev);
+
+#endif
diff --git a/drivers/net/ethernet/huawei/hinic/hinic_hw_if.c b/drivers/net/ethernet/huawei/hinic/hinic_hw_if.c
new file mode 100644
index 000000000000..88567305d06e
--- /dev/null
+++ b/drivers/net/ethernet/huawei/hinic/hinic_hw_if.c
@@ -0,0 +1,419 @@
+// SPDX-License-Identifier: GPL-2.0-only
+/*
+ * Huawei HiNIC PCI Express Linux driver
+ * Copyright(c) 2017 Huawei Technologies Co., Ltd
+ */
+
+#include <linux/pci.h>
+#include <linux/device.h>
+#include <linux/errno.h>
+#include <linux/io.h>
+#include <linux/types.h>
+#include <linux/bitops.h>
+#include <linux/delay.h>
+
+#include "hinic_hw_csr.h"
+#include "hinic_hw_if.h"
+
+#define PCIE_ATTR_ENTRY         0
+
+#define VALID_MSIX_IDX(attr, msix_index) ((msix_index) < (attr)->num_irqs)
+
+#define WAIT_HWIF_READY_TIMEOUT	10000
+
+#define HINIC_SELFTEST_RESULT 0x883C
+
+/**
+ * hinic_msix_attr_set - set message attribute for msix entry
+ * @hwif: the HW interface of a pci function device
+ * @msix_index: msix_index
+ * @pending_limit: the maximum pending interrupt events (unit 8)
+ * @coalesc_timer: coalesc period for interrupt (unit 8 us)
+ * @lli_timer: replenishing period for low latency credit (unit 8 us)
+ * @lli_credit_limit: maximum credits for low latency msix messages (unit 8)
+ * @resend_timer: maximum wait for resending msix (unit coalesc period)
+ *
+ * Return 0 - Success, negative - Failure
+ **/
+int hinic_msix_attr_set(struct hinic_hwif *hwif, u16 msix_index,
+			u8 pending_limit, u8 coalesc_timer,
+			u8 lli_timer, u8 lli_credit_limit,
+			u8 resend_timer)
+{
+	u32 msix_ctrl, addr;
+
+	if (!VALID_MSIX_IDX(&hwif->attr, msix_index))
+		return -EINVAL;
+
+	msix_ctrl = HINIC_MSIX_ATTR_SET(pending_limit, PENDING_LIMIT)   |
+		    HINIC_MSIX_ATTR_SET(coalesc_timer, COALESC_TIMER)   |
+		    HINIC_MSIX_ATTR_SET(lli_timer, LLI_TIMER)           |
+		    HINIC_MSIX_ATTR_SET(lli_credit_limit, LLI_CREDIT)   |
+		    HINIC_MSIX_ATTR_SET(resend_timer, RESEND_TIMER);
+
+	addr = HINIC_CSR_MSIX_CTRL_ADDR(msix_index);
+
+	hinic_hwif_write_reg(hwif, addr, msix_ctrl);
+	return 0;
+}
+
+/**
+ * hinic_msix_attr_cnt_clear - clear message attribute counters for msix entry
+ * @hwif: the HW interface of a pci function device
+ * @msix_index: msix_index
+ *
+ * Return 0 - Success, negative - Failure
+ **/
+int hinic_msix_attr_cnt_clear(struct hinic_hwif *hwif, u16 msix_index)
+{
+	u32 msix_ctrl, addr;
+
+	if (!VALID_MSIX_IDX(&hwif->attr, msix_index))
+		return -EINVAL;
+
+	msix_ctrl = HINIC_MSIX_CNT_SET(1, RESEND_TIMER);
+	addr = HINIC_CSR_MSIX_CNT_ADDR(msix_index);
+
+	hinic_hwif_write_reg(hwif, addr, msix_ctrl);
+	return 0;
+}
+
+/**
+ * hinic_set_pf_action - set action on pf channel
+ * @hwif: the HW interface of a pci function device
+ * @action: action on pf channel
+ **/
+void hinic_set_pf_action(struct hinic_hwif *hwif, enum hinic_pf_action action)
+{
+	u32 attr5;
+
+	if (HINIC_IS_VF(hwif))
+		return;
+
+	attr5 = hinic_hwif_read_reg(hwif, HINIC_CSR_FUNC_ATTR5_ADDR);
+	attr5 = HINIC_FA5_CLEAR(attr5, PF_ACTION);
+	attr5 |= HINIC_FA5_SET(action, PF_ACTION);
+
+	hinic_hwif_write_reg(hwif, HINIC_CSR_FUNC_ATTR5_ADDR, attr5);
+}
+
+enum hinic_outbound_state hinic_outbound_state_get(struct hinic_hwif *hwif)
+{
+	u32 attr4 = hinic_hwif_read_reg(hwif, HINIC_CSR_FUNC_ATTR4_ADDR);
+
+	return HINIC_FA4_GET(attr4, OUTBOUND_STATE);
+}
+
+void hinic_outbound_state_set(struct hinic_hwif *hwif,
+			      enum hinic_outbound_state outbound_state)
+{
+	u32 attr4 = hinic_hwif_read_reg(hwif, HINIC_CSR_FUNC_ATTR4_ADDR);
+
+	attr4 = HINIC_FA4_CLEAR(attr4, OUTBOUND_STATE);
+	attr4 |= HINIC_FA4_SET(outbound_state, OUTBOUND_STATE);
+
+	hinic_hwif_write_reg(hwif, HINIC_CSR_FUNC_ATTR4_ADDR, attr4);
+}
+
+enum hinic_db_state hinic_db_state_get(struct hinic_hwif *hwif)
+{
+	u32 attr4 = hinic_hwif_read_reg(hwif, HINIC_CSR_FUNC_ATTR4_ADDR);
+
+	return HINIC_FA4_GET(attr4, DB_STATE);
+}
+
+void hinic_db_state_set(struct hinic_hwif *hwif,
+			enum hinic_db_state db_state)
+{
+	u32 attr4 = hinic_hwif_read_reg(hwif, HINIC_CSR_FUNC_ATTR4_ADDR);
+
+	attr4 = HINIC_FA4_CLEAR(attr4, DB_STATE);
+	attr4 |= HINIC_FA4_SET(db_state, DB_STATE);
+
+	hinic_hwif_write_reg(hwif, HINIC_CSR_FUNC_ATTR4_ADDR, attr4);
+}
+
+void hinic_set_msix_state(struct hinic_hwif *hwif, u16 msix_idx,
+			  enum hinic_msix_state flag)
+{
+	u32 offset = msix_idx * HINIC_PCI_MSIX_ENTRY_SIZE +
+			HINIC_PCI_MSIX_ENTRY_VECTOR_CTRL;
+	u32 mask_bits;
+
+	mask_bits = readl(hwif->intr_regs_base + offset);
+	mask_bits &= ~HINIC_PCI_MSIX_ENTRY_CTRL_MASKBIT;
+
+	if (flag)
+		mask_bits |= HINIC_PCI_MSIX_ENTRY_CTRL_MASKBIT;
+
+	writel(mask_bits, hwif->intr_regs_base + offset);
+}
+
+/**
+ * hwif_ready - test if the HW is ready for use
+ * @hwif: the HW interface of a pci function device
+ *
+ * Return 0 - Success, negative - Failure
+ **/
+static int hwif_ready(struct hinic_hwif *hwif)
+{
+	u32 addr, attr1;
+
+	addr   = HINIC_CSR_FUNC_ATTR1_ADDR;
+	attr1  = hinic_hwif_read_reg(hwif, addr);
+
+	if (!HINIC_FA1_GET(attr1, MGMT_INIT_STATUS))
+		return -EBUSY;
+
+	if (HINIC_IS_VF(hwif)) {
+		if (!HINIC_FA1_GET(attr1, PF_INIT_STATUS))
+			return -EBUSY;
+	}
+
+	return 0;
+}
+
+static int wait_hwif_ready(struct hinic_hwif *hwif)
+{
+	unsigned long timeout = 0;
+
+	do {
+		if (!hwif_ready(hwif))
+			return 0;
+
+		usleep_range(999, 1000);
+		timeout++;
+	} while (timeout <= WAIT_HWIF_READY_TIMEOUT);
+
+	dev_err(&hwif->pdev->dev, "Wait for hwif timeout\n");
+
+	return -EBUSY;
+}
+
+/**
+ * set_hwif_attr - set the attributes in the relevant members in hwif
+ * @hwif: the HW interface of a pci function device
+ * @attr0: the first attribute that was read from the hw
+ * @attr1: the second attribute that was read from the hw
+ * @attr2: the third attribute that was read from the hw
+ **/
+static void set_hwif_attr(struct hinic_hwif *hwif, u32 attr0, u32 attr1,
+			  u32 attr2)
+{
+	hwif->attr.func_idx     = HINIC_FA0_GET(attr0, FUNC_IDX);
+	hwif->attr.pf_idx       = HINIC_FA0_GET(attr0, PF_IDX);
+	hwif->attr.pci_intf_idx = HINIC_FA0_GET(attr0, PCI_INTF_IDX);
+	hwif->attr.func_type    = HINIC_FA0_GET(attr0, FUNC_TYPE);
+
+	hwif->attr.num_aeqs = BIT(HINIC_FA1_GET(attr1, AEQS_PER_FUNC));
+	hwif->attr.num_ceqs = BIT(HINIC_FA1_GET(attr1, CEQS_PER_FUNC));
+	hwif->attr.num_irqs = BIT(HINIC_FA1_GET(attr1, IRQS_PER_FUNC));
+	hwif->attr.num_dma_attr = BIT(HINIC_FA1_GET(attr1, DMA_ATTR_PER_FUNC));
+	hwif->attr.global_vf_id_of_pf = HINIC_FA2_GET(attr2,
+						      GLOBAL_VF_ID_OF_PF);
+}
+
+/**
+ * read_hwif_attr - read the attributes and set members in hwif
+ * @hwif: the HW interface of a pci function device
+ **/
+static void read_hwif_attr(struct hinic_hwif *hwif)
+{
+	u32 addr, attr0, attr1, attr2;
+
+	addr   = HINIC_CSR_FUNC_ATTR0_ADDR;
+	attr0  = hinic_hwif_read_reg(hwif, addr);
+
+	addr   = HINIC_CSR_FUNC_ATTR1_ADDR;
+	attr1  = hinic_hwif_read_reg(hwif, addr);
+
+	addr   = HINIC_CSR_FUNC_ATTR2_ADDR;
+	attr2  = hinic_hwif_read_reg(hwif, addr);
+
+	set_hwif_attr(hwif, attr0, attr1, attr2);
+}
+
+/**
+ * set_ppf - try to set hwif as ppf and set the type of hwif in this case
+ * @hwif: the HW interface of a pci function device
+ **/
+static void set_ppf(struct hinic_hwif *hwif)
+{
+	struct hinic_func_attr *attr = &hwif->attr;
+	u32 addr, val, ppf_election;
+
+	/* Read Modify Write */
+	addr = HINIC_CSR_PPF_ELECTION_ADDR(HINIC_HWIF_PCI_INTF(hwif));
+
+	val = hinic_hwif_read_reg(hwif, addr);
+	val = HINIC_PPF_ELECTION_CLEAR(val, IDX);
+
+	ppf_election = HINIC_PPF_ELECTION_SET(HINIC_HWIF_FUNC_IDX(hwif), IDX);
+
+	val |= ppf_election;
+	hinic_hwif_write_reg(hwif, addr, val);
+
+	/* check PPF */
+	val = hinic_hwif_read_reg(hwif, addr);
+
+	attr->ppf_idx = HINIC_PPF_ELECTION_GET(val, IDX);
+	if (attr->ppf_idx == HINIC_HWIF_FUNC_IDX(hwif))
+		attr->func_type = HINIC_PPF;
+}
+
+/**
+ * set_dma_attr - set the dma attributes in the HW
+ * @hwif: the HW interface of a pci function device
+ * @entry_idx: the entry index in the dma table
+ * @st: PCIE TLP steering tag
+ * @at: PCIE TLP AT field
+ * @ph: PCIE TLP Processing Hint field
+ * @no_snooping: PCIE TLP No snooping
+ * @tph_en: PCIE TLP Processing Hint Enable
+ **/
+static void set_dma_attr(struct hinic_hwif *hwif, u32 entry_idx,
+			 u8 st, u8 at, u8 ph,
+			 enum hinic_pcie_nosnoop no_snooping,
+			 enum hinic_pcie_tph tph_en)
+{
+	u32 addr, val, dma_attr_entry;
+
+	/* Read Modify Write */
+	addr = HINIC_CSR_DMA_ATTR_ADDR(entry_idx);
+
+	val = hinic_hwif_read_reg(hwif, addr);
+	val = HINIC_DMA_ATTR_CLEAR(val, ST)             &
+	      HINIC_DMA_ATTR_CLEAR(val, AT)             &
+	      HINIC_DMA_ATTR_CLEAR(val, PH)             &
+	      HINIC_DMA_ATTR_CLEAR(val, NO_SNOOPING)    &
+	      HINIC_DMA_ATTR_CLEAR(val, TPH_EN);
+
+	dma_attr_entry = HINIC_DMA_ATTR_SET(st, ST)                     |
+			 HINIC_DMA_ATTR_SET(at, AT)                     |
+			 HINIC_DMA_ATTR_SET(ph, PH)                     |
+			 HINIC_DMA_ATTR_SET(no_snooping, NO_SNOOPING)   |
+			 HINIC_DMA_ATTR_SET(tph_en, TPH_EN);
+
+	val |= dma_attr_entry;
+	hinic_hwif_write_reg(hwif, addr, val);
+}
+
+/**
+ * dma_attr_init - initialize the default dma attributes
+ * @hwif: the HW interface of a pci function device
+ **/
+static void dma_attr_init(struct hinic_hwif *hwif)
+{
+	set_dma_attr(hwif, PCIE_ATTR_ENTRY, HINIC_PCIE_ST_DISABLE,
+		     HINIC_PCIE_AT_DISABLE, HINIC_PCIE_PH_DISABLE,
+		     HINIC_PCIE_SNOOP, HINIC_PCIE_TPH_DISABLE);
+}
+
+u16 hinic_glb_pf_vf_offset(struct hinic_hwif *hwif)
+{
+	if (!hwif)
+		return 0;
+
+	return hwif->attr.global_vf_id_of_pf;
+}
+
+u16 hinic_global_func_id_hw(struct hinic_hwif *hwif)
+{
+	u32 addr, attr0;
+
+	addr   = HINIC_CSR_FUNC_ATTR0_ADDR;
+	attr0  = hinic_hwif_read_reg(hwif, addr);
+
+	return HINIC_FA0_GET(attr0, FUNC_IDX);
+}
+
+u16 hinic_pf_id_of_vf_hw(struct hinic_hwif *hwif)
+{
+	u32 addr, attr0;
+
+	addr   = HINIC_CSR_FUNC_ATTR0_ADDR;
+	attr0  = hinic_hwif_read_reg(hwif, addr);
+
+	return HINIC_FA0_GET(attr0, PF_IDX);
+}
+
+static void __print_selftest_reg(struct hinic_hwif *hwif)
+{
+	u32 addr, attr0, attr1;
+
+	addr   = HINIC_CSR_FUNC_ATTR1_ADDR;
+	attr1  = hinic_hwif_read_reg(hwif, addr);
+
+	if (attr1 == HINIC_PCIE_LINK_DOWN) {
+		dev_err(&hwif->pdev->dev, "PCIE is link down\n");
+		return;
+	}
+
+	addr   = HINIC_CSR_FUNC_ATTR0_ADDR;
+	attr0  = hinic_hwif_read_reg(hwif, addr);
+	if (HINIC_FA0_GET(attr0, FUNC_TYPE) != HINIC_VF &&
+	    !HINIC_FA0_GET(attr0, PCI_INTF_IDX))
+		dev_err(&hwif->pdev->dev, "Selftest reg: 0x%08x\n",
+			hinic_hwif_read_reg(hwif, HINIC_SELFTEST_RESULT));
+}
+
+/**
+ * hinic_init_hwif - initialize the hw interface
+ * @hwif: the HW interface of a pci function device
+ * @pdev: the pci device for accessing PCI resources
+ *
+ * Return 0 - Success, negative - Failure
+ **/
+int hinic_init_hwif(struct hinic_hwif *hwif, struct pci_dev *pdev)
+{
+	int err;
+
+	hwif->pdev = pdev;
+
+	hwif->cfg_regs_bar = pci_ioremap_bar(pdev, HINIC_PCI_CFG_REGS_BAR);
+	if (!hwif->cfg_regs_bar) {
+		dev_err(&pdev->dev, "Failed to map configuration regs\n");
+		return -ENOMEM;
+	}
+
+	hwif->intr_regs_base = pci_ioremap_bar(pdev, HINIC_PCI_INTR_REGS_BAR);
+	if (!hwif->intr_regs_base) {
+		dev_err(&pdev->dev, "Failed to map configuration regs\n");
+		err = -ENOMEM;
+		goto err_map_intr_bar;
+	}
+
+	err = wait_hwif_ready(hwif);
+	if (err) {
+		dev_err(&pdev->dev, "HW interface is not ready\n");
+		__print_selftest_reg(hwif);
+		goto err_hwif_ready;
+	}
+
+	read_hwif_attr(hwif);
+
+	if (HINIC_IS_PF(hwif))
+		set_ppf(hwif);
+
+	/* No transactionss before DMA is initialized */
+	dma_attr_init(hwif);
+	return 0;
+
+err_hwif_ready:
+	iounmap(hwif->intr_regs_base);
+
+err_map_intr_bar:
+	iounmap(hwif->cfg_regs_bar);
+
+	return err;
+}
+
+/**
+ * hinic_free_hwif - free the HW interface
+ * @hwif: the HW interface of a pci function device
+ **/
+void hinic_free_hwif(struct hinic_hwif *hwif)
+{
+	iounmap(hwif->intr_regs_base);
+	iounmap(hwif->cfg_regs_bar);
+}
diff --git a/drivers/net/ethernet/huawei/hinic/hinic_hw_if.h b/drivers/net/ethernet/huawei/hinic/hinic_hw_if.h
new file mode 100644
index 000000000000..3d588896a367
--- /dev/null
+++ b/drivers/net/ethernet/huawei/hinic/hinic_hw_if.h
@@ -0,0 +1,295 @@
+/* SPDX-License-Identifier: GPL-2.0-only */
+/*
+ * Huawei HiNIC PCI Express Linux driver
+ * Copyright(c) 2017 Huawei Technologies Co., Ltd
+ */
+
+#ifndef HINIC_HW_IF_H
+#define HINIC_HW_IF_H
+
+#include <linux/pci.h>
+#include <linux/io.h>
+#include <linux/types.h>
+#include <asm/byteorder.h>
+
+#define HINIC_PCIE_LINK_DOWN					0xFFFFFFFF
+
+#define HINIC_DMA_ATTR_ST_SHIFT                                 0
+#define HINIC_DMA_ATTR_AT_SHIFT                                 8
+#define HINIC_DMA_ATTR_PH_SHIFT                                 10
+#define HINIC_DMA_ATTR_NO_SNOOPING_SHIFT                        12
+#define HINIC_DMA_ATTR_TPH_EN_SHIFT                             13
+
+#define HINIC_DMA_ATTR_ST_MASK                                  0xFF
+#define HINIC_DMA_ATTR_AT_MASK                                  0x3
+#define HINIC_DMA_ATTR_PH_MASK                                  0x3
+#define HINIC_DMA_ATTR_NO_SNOOPING_MASK                         0x1
+#define HINIC_DMA_ATTR_TPH_EN_MASK                              0x1
+
+#define HINIC_DMA_ATTR_SET(val, member)                         \
+	(((u32)(val) & HINIC_DMA_ATTR_##member##_MASK) <<       \
+	 HINIC_DMA_ATTR_##member##_SHIFT)
+
+#define HINIC_DMA_ATTR_CLEAR(val, member)                       \
+	((val) & (~(HINIC_DMA_ATTR_##member##_MASK              \
+	 << HINIC_DMA_ATTR_##member##_SHIFT)))
+
+#define HINIC_FA0_FUNC_IDX_SHIFT                                0
+#define HINIC_FA0_PF_IDX_SHIFT                                  10
+#define HINIC_FA0_PCI_INTF_IDX_SHIFT                            14
+#define HINIC_FA0_VF_IN_PF_SHIFT				16
+/* reserved members - off 16 */
+#define HINIC_FA0_FUNC_TYPE_SHIFT                               24
+
+#define HINIC_FA0_FUNC_IDX_MASK                                 0x3FF
+#define HINIC_FA0_PF_IDX_MASK                                   0xF
+#define HINIC_FA0_PCI_INTF_IDX_MASK                             0x3
+#define HINIC_FA0_FUNC_TYPE_MASK                                0x1
+#define HINIC_FA0_VF_IN_PF_MASK					0xFF
+
+#define HINIC_FA0_GET(val, member)                              \
+	(((val) >> HINIC_FA0_##member##_SHIFT) & HINIC_FA0_##member##_MASK)
+
+#define HINIC_FA1_AEQS_PER_FUNC_SHIFT                           8
+/* reserved members - off 10 */
+#define HINIC_FA1_CEQS_PER_FUNC_SHIFT                           12
+/* reserved members - off 15 */
+#define HINIC_FA1_IRQS_PER_FUNC_SHIFT                           20
+#define HINIC_FA1_DMA_ATTR_PER_FUNC_SHIFT                       24
+/* reserved members - off 27 */
+#define HINIC_FA1_MGMT_INIT_STATUS_SHIFT			30
+#define HINIC_FA1_PF_INIT_STATUS_SHIFT				31
+
+#define HINIC_FA1_AEQS_PER_FUNC_MASK                            0x3
+#define HINIC_FA1_CEQS_PER_FUNC_MASK                            0x7
+#define HINIC_FA1_IRQS_PER_FUNC_MASK                            0xF
+#define HINIC_FA1_DMA_ATTR_PER_FUNC_MASK                        0x7
+#define HINIC_FA1_MGMT_INIT_STATUS_MASK                         0x1
+#define HINIC_FA1_PF_INIT_STATUS_MASK				0x1
+
+#define HINIC_FA1_GET(val, member)                              \
+	(((val) >> HINIC_FA1_##member##_SHIFT) & HINIC_FA1_##member##_MASK)
+
+#define HINIC_FA2_GLOBAL_VF_ID_OF_PF_SHIFT	16
+#define HINIC_FA2_GLOBAL_VF_ID_OF_PF_MASK	0x3FF
+
+#define HINIC_FA2_GET(val, member)				\
+	(((val) >> HINIC_FA2_##member##_SHIFT) & HINIC_FA2_##member##_MASK)
+
+#define HINIC_FA4_OUTBOUND_STATE_SHIFT                          0
+#define HINIC_FA4_DB_STATE_SHIFT                                1
+
+#define HINIC_FA4_OUTBOUND_STATE_MASK                           0x1
+#define HINIC_FA4_DB_STATE_MASK                                 0x1
+
+#define HINIC_FA4_GET(val, member)                              \
+	(((val) >> HINIC_FA4_##member##_SHIFT) & HINIC_FA4_##member##_MASK)
+
+#define HINIC_FA4_SET(val, member)                              \
+	((((u32)val) & HINIC_FA4_##member##_MASK) << HINIC_FA4_##member##_SHIFT)
+
+#define HINIC_FA4_CLEAR(val, member)                            \
+	((val) & (~(HINIC_FA4_##member##_MASK << HINIC_FA4_##member##_SHIFT)))
+
+#define HINIC_FA5_PF_ACTION_SHIFT                               0
+#define HINIC_FA5_PF_ACTION_MASK                                0xFFFF
+
+#define HINIC_FA5_SET(val, member)                              \
+	(((u32)(val) & HINIC_FA5_##member##_MASK) << HINIC_FA5_##member##_SHIFT)
+
+#define HINIC_FA5_CLEAR(val, member)                            \
+	((val) & (~(HINIC_FA5_##member##_MASK << HINIC_FA5_##member##_SHIFT)))
+
+#define HINIC_PPF_ELECTION_IDX_SHIFT                            0
+#define HINIC_PPF_ELECTION_IDX_MASK                             0x1F
+
+#define HINIC_PPF_ELECTION_SET(val, member)                     \
+	(((u32)(val) & HINIC_PPF_ELECTION_##member##_MASK) <<   \
+	 HINIC_PPF_ELECTION_##member##_SHIFT)
+
+#define HINIC_PPF_ELECTION_GET(val, member)                     \
+	(((val) >> HINIC_PPF_ELECTION_##member##_SHIFT) &       \
+	 HINIC_PPF_ELECTION_##member##_MASK)
+
+#define HINIC_PPF_ELECTION_CLEAR(val, member)                   \
+	((val) & (~(HINIC_PPF_ELECTION_##member##_MASK          \
+	 << HINIC_PPF_ELECTION_##member##_SHIFT)))
+
+#define HINIC_MSIX_PENDING_LIMIT_SHIFT                          0
+#define HINIC_MSIX_COALESC_TIMER_SHIFT                          8
+#define HINIC_MSIX_LLI_TIMER_SHIFT                              16
+#define HINIC_MSIX_LLI_CREDIT_SHIFT                             24
+#define HINIC_MSIX_RESEND_TIMER_SHIFT                           29
+
+#define HINIC_MSIX_PENDING_LIMIT_MASK                           0xFF
+#define HINIC_MSIX_COALESC_TIMER_MASK                           0xFF
+#define HINIC_MSIX_LLI_TIMER_MASK                               0xFF
+#define HINIC_MSIX_LLI_CREDIT_MASK                              0x1F
+#define HINIC_MSIX_RESEND_TIMER_MASK                            0x7
+
+#define HINIC_MSIX_ATTR_SET(val, member)                        \
+	(((u32)(val) & HINIC_MSIX_##member##_MASK) <<           \
+	 HINIC_MSIX_##member##_SHIFT)
+
+#define HINIC_MSIX_CNT_RESEND_TIMER_SHIFT                       29
+
+#define HINIC_MSIX_CNT_RESEND_TIMER_MASK                        0x1
+
+#define HINIC_MSIX_CNT_SET(val, member)                         \
+	(((u32)(val) & HINIC_MSIX_CNT_##member##_MASK) <<       \
+	 HINIC_MSIX_CNT_##member##_SHIFT)
+
+#define HINIC_HWIF_NUM_AEQS(hwif)       ((hwif)->attr.num_aeqs)
+#define HINIC_HWIF_NUM_CEQS(hwif)       ((hwif)->attr.num_ceqs)
+#define HINIC_HWIF_NUM_IRQS(hwif)       ((hwif)->attr.num_irqs)
+#define HINIC_HWIF_FUNC_IDX(hwif)       ((hwif)->attr.func_idx)
+#define HINIC_HWIF_PCI_INTF(hwif)       ((hwif)->attr.pci_intf_idx)
+#define HINIC_HWIF_PF_IDX(hwif)         ((hwif)->attr.pf_idx)
+#define HINIC_HWIF_PPF_IDX(hwif)        ((hwif)->attr.ppf_idx)
+
+#define HINIC_FUNC_TYPE(hwif)           ((hwif)->attr.func_type)
+#define HINIC_IS_VF(hwif)               (HINIC_FUNC_TYPE(hwif) == HINIC_VF)
+#define HINIC_IS_PF(hwif)               (HINIC_FUNC_TYPE(hwif) == HINIC_PF)
+#define HINIC_IS_PPF(hwif)              (HINIC_FUNC_TYPE(hwif) == HINIC_PPF)
+
+#define HINIC_PCI_CFG_REGS_BAR          0
+#define HINIC_PCI_INTR_REGS_BAR         2
+#define HINIC_PCI_DB_BAR                4
+
+#define HINIC_PCIE_ST_DISABLE           0
+#define HINIC_PCIE_AT_DISABLE           0
+#define HINIC_PCIE_PH_DISABLE           0
+
+#define HINIC_EQ_MSIX_PENDING_LIMIT_DEFAULT     0       /* Disabled */
+#define HINIC_EQ_MSIX_COALESC_TIMER_DEFAULT     0xFF    /* max */
+#define HINIC_EQ_MSIX_LLI_TIMER_DEFAULT         0       /* Disabled */
+#define HINIC_EQ_MSIX_LLI_CREDIT_LIMIT_DEFAULT  0       /* Disabled */
+#define HINIC_EQ_MSIX_RESEND_TIMER_DEFAULT      7       /* max */
+
+#define HINIC_PCI_MSIX_ENTRY_SIZE               16
+#define HINIC_PCI_MSIX_ENTRY_VECTOR_CTRL        12
+#define HINIC_PCI_MSIX_ENTRY_CTRL_MASKBIT       1
+
+enum hinic_pcie_nosnoop {
+	HINIC_PCIE_SNOOP        = 0,
+	HINIC_PCIE_NO_SNOOP     = 1,
+};
+
+enum hinic_pcie_tph {
+	HINIC_PCIE_TPH_DISABLE  = 0,
+	HINIC_PCIE_TPH_ENABLE   = 1,
+};
+
+enum hinic_func_type {
+	HINIC_PF        = 0,
+	HINIC_VF	    = 1,
+	HINIC_PPF       = 2,
+};
+
+enum hinic_mod_type {
+	HINIC_MOD_COMM  = 0,    /* HW communication module */
+	HINIC_MOD_L2NIC = 1,    /* L2NIC module */
+	HINIC_MOD_CFGM  = 7,    /* Configuration module */
+	HINIC_MOD_HILINK = 14,  /* Hilink module */
+	HINIC_MOD_MAX   = 15
+};
+
+enum hinic_node_id {
+	HINIC_NODE_ID_MGMT = 21,
+};
+
+enum hinic_pf_action {
+	HINIC_PF_MGMT_INIT = 0x0,
+
+	HINIC_PF_MGMT_ACTIVE = 0x11,
+};
+
+enum hinic_outbound_state {
+	HINIC_OUTBOUND_ENABLE  = 0,
+	HINIC_OUTBOUND_DISABLE = 1,
+};
+
+enum hinic_db_state {
+	HINIC_DB_ENABLE  = 0,
+	HINIC_DB_DISABLE = 1,
+};
+
+enum hinic_msix_state {
+	HINIC_MSIX_ENABLE,
+	HINIC_MSIX_DISABLE,
+};
+
+struct hinic_func_attr {
+	u16                     func_idx;
+	u8                      pf_idx;
+	u8                      pci_intf_idx;
+
+	enum hinic_func_type    func_type;
+
+	u8                      ppf_idx;
+
+	u16                     num_irqs;
+	u8                      num_aeqs;
+	u8                      num_ceqs;
+
+	u8                      num_dma_attr;
+
+	u16						global_vf_id_of_pf;
+};
+
+struct hinic_hwif {
+	struct pci_dev          *pdev;
+	void __iomem            *cfg_regs_bar;
+	void __iomem		*intr_regs_base;
+
+	struct hinic_func_attr  attr;
+};
+
+static inline u32 hinic_hwif_read_reg(struct hinic_hwif *hwif, u32 reg)
+{
+	u32 out = readl(hwif->cfg_regs_bar + reg);
+
+	return be32_to_cpu(*(__be32 *)&out);
+}
+
+static inline void hinic_hwif_write_reg(struct hinic_hwif *hwif, u32 reg,
+					u32 val)
+{
+	__be32 in = cpu_to_be32(val);
+
+	writel(*(u32 *)&in, hwif->cfg_regs_bar + reg);
+}
+
+int hinic_msix_attr_set(struct hinic_hwif *hwif, u16 msix_index,
+			u8 pending_limit, u8 coalesc_timer,
+			u8 lli_timer_cfg, u8 lli_credit_limit,
+			u8 resend_timer);
+
+void hinic_set_msix_state(struct hinic_hwif *hwif, u16 msix_idx,
+			  enum hinic_msix_state flag);
+
+int hinic_msix_attr_cnt_clear(struct hinic_hwif *hwif, u16 msix_index);
+
+void hinic_set_pf_action(struct hinic_hwif *hwif, enum hinic_pf_action action);
+
+enum hinic_outbound_state hinic_outbound_state_get(struct hinic_hwif *hwif);
+
+void hinic_outbound_state_set(struct hinic_hwif *hwif,
+			      enum hinic_outbound_state outbound_state);
+
+enum hinic_db_state hinic_db_state_get(struct hinic_hwif *hwif);
+
+void hinic_db_state_set(struct hinic_hwif *hwif,
+			enum hinic_db_state db_state);
+
+u16 hinic_glb_pf_vf_offset(struct hinic_hwif *hwif);
+
+u16 hinic_global_func_id_hw(struct hinic_hwif *hwif);
+
+u16 hinic_pf_id_of_vf_hw(struct hinic_hwif *hwif);
+
+int hinic_init_hwif(struct hinic_hwif *hwif, struct pci_dev *pdev);
+
+void hinic_free_hwif(struct hinic_hwif *hwif);
+
+#endif
diff --git a/drivers/net/ethernet/huawei/hinic/hinic_hw_io.c b/drivers/net/ethernet/huawei/hinic/hinic_hw_io.c
new file mode 100644
index 000000000000..c4a0ba6e183a
--- /dev/null
+++ b/drivers/net/ethernet/huawei/hinic/hinic_hw_io.c
@@ -0,0 +1,634 @@
+// SPDX-License-Identifier: GPL-2.0-only
+/*
+ * Huawei HiNIC PCI Express Linux driver
+ * Copyright(c) 2017 Huawei Technologies Co., Ltd
+ */
+
+#include <linux/kernel.h>
+#include <linux/types.h>
+#include <linux/pci.h>
+#include <linux/device.h>
+#include <linux/errno.h>
+#include <linux/slab.h>
+#include <linux/semaphore.h>
+#include <linux/dma-mapping.h>
+#include <linux/io.h>
+#include <linux/err.h>
+
+#include "hinic_hw_dev.h"
+#include "hinic_hw_if.h"
+#include "hinic_hw_eqs.h"
+#include "hinic_hw_wqe.h"
+#include "hinic_hw_wq.h"
+#include "hinic_hw_cmdq.h"
+#include "hinic_hw_qp_ctxt.h"
+#include "hinic_hw_qp.h"
+#include "hinic_hw_io.h"
+
+#define CI_Q_ADDR_SIZE                  sizeof(u32)
+
+#define CI_ADDR(base_addr, q_id)        ((base_addr) + \
+					 (q_id) * CI_Q_ADDR_SIZE)
+
+#define CI_TABLE_SIZE(num_qps)          ((num_qps) * CI_Q_ADDR_SIZE)
+
+#define DB_IDX(db, db_base)             \
+	(((unsigned long)(db) - (unsigned long)(db_base)) / HINIC_DB_PAGE_SIZE)
+
+#define HINIC_PAGE_SIZE_HW(pg_size)	((u8)ilog2((u32)((pg_size) >> 12)))
+
+enum io_cmd {
+	IO_CMD_MODIFY_QUEUE_CTXT = 0,
+	IO_CMD_CLEAN_QUEUE_CTXT,
+};
+
+static void init_db_area_idx(struct hinic_free_db_area *free_db_area)
+{
+	int i;
+
+	for (i = 0; i < HINIC_DB_MAX_AREAS; i++)
+		free_db_area->db_idx[i] = i;
+
+	free_db_area->alloc_pos = 0;
+	free_db_area->return_pos = HINIC_DB_MAX_AREAS;
+
+	free_db_area->num_free = HINIC_DB_MAX_AREAS;
+
+	sema_init(&free_db_area->idx_lock, 1);
+}
+
+static void __iomem *get_db_area(struct hinic_func_to_io *func_to_io)
+{
+	struct hinic_free_db_area *free_db_area = &func_to_io->free_db_area;
+	int pos, idx;
+
+	down(&free_db_area->idx_lock);
+
+	free_db_area->num_free--;
+
+	if (free_db_area->num_free < 0) {
+		free_db_area->num_free++;
+		up(&free_db_area->idx_lock);
+		return ERR_PTR(-ENOMEM);
+	}
+
+	pos = free_db_area->alloc_pos++;
+	pos &= HINIC_DB_MAX_AREAS - 1;
+
+	idx = free_db_area->db_idx[pos];
+
+	free_db_area->db_idx[pos] = -1;
+
+	up(&free_db_area->idx_lock);
+
+	return func_to_io->db_base + idx * HINIC_DB_PAGE_SIZE;
+}
+
+static void return_db_area(struct hinic_func_to_io *func_to_io,
+			   void __iomem *db_base)
+{
+	struct hinic_free_db_area *free_db_area = &func_to_io->free_db_area;
+	int pos, idx = DB_IDX(db_base, func_to_io->db_base);
+
+	down(&free_db_area->idx_lock);
+
+	pos = free_db_area->return_pos++;
+	pos &= HINIC_DB_MAX_AREAS - 1;
+
+	free_db_area->db_idx[pos] = idx;
+
+	free_db_area->num_free++;
+
+	up(&free_db_area->idx_lock);
+}
+
+static int write_sq_ctxts(struct hinic_func_to_io *func_to_io, u16 base_qpn,
+			  u16 num_sqs)
+{
+	struct hinic_hwif *hwif = func_to_io->hwif;
+	struct hinic_sq_ctxt_block *sq_ctxt_block;
+	struct pci_dev *pdev = hwif->pdev;
+	struct hinic_cmdq_buf cmdq_buf;
+	struct hinic_sq_ctxt *sq_ctxt;
+	struct hinic_qp *qp;
+	u64 out_param;
+	int err, i;
+
+	err = hinic_alloc_cmdq_buf(&func_to_io->cmdqs, &cmdq_buf);
+	if (err) {
+		dev_err(&pdev->dev, "Failed to allocate cmdq buf\n");
+		return err;
+	}
+
+	sq_ctxt_block = cmdq_buf.buf;
+	sq_ctxt = sq_ctxt_block->sq_ctxt;
+
+	hinic_qp_prepare_header(&sq_ctxt_block->hdr, HINIC_QP_CTXT_TYPE_SQ,
+				num_sqs, func_to_io->max_qps);
+	for (i = 0; i < num_sqs; i++) {
+		qp = &func_to_io->qps[i];
+
+		hinic_sq_prepare_ctxt(&sq_ctxt[i], &qp->sq,
+				      base_qpn + qp->q_id);
+	}
+
+	cmdq_buf.size = HINIC_SQ_CTXT_SIZE(num_sqs);
+
+	err = hinic_cmdq_direct_resp(&func_to_io->cmdqs, HINIC_MOD_L2NIC,
+				     IO_CMD_MODIFY_QUEUE_CTXT, &cmdq_buf,
+				     &out_param);
+	if (err || out_param != 0) {
+		dev_err(&pdev->dev, "Failed to set SQ ctxts\n");
+		err = -EFAULT;
+	}
+
+	hinic_free_cmdq_buf(&func_to_io->cmdqs, &cmdq_buf);
+	return err;
+}
+
+static int write_rq_ctxts(struct hinic_func_to_io *func_to_io, u16 base_qpn,
+			  u16 num_rqs)
+{
+	struct hinic_hwif *hwif = func_to_io->hwif;
+	struct hinic_rq_ctxt_block *rq_ctxt_block;
+	struct pci_dev *pdev = hwif->pdev;
+	struct hinic_cmdq_buf cmdq_buf;
+	struct hinic_rq_ctxt *rq_ctxt;
+	struct hinic_qp *qp;
+	u64 out_param;
+	int err, i;
+
+	err = hinic_alloc_cmdq_buf(&func_to_io->cmdqs, &cmdq_buf);
+	if (err) {
+		dev_err(&pdev->dev, "Failed to allocate cmdq buf\n");
+		return err;
+	}
+
+	rq_ctxt_block = cmdq_buf.buf;
+	rq_ctxt = rq_ctxt_block->rq_ctxt;
+
+	hinic_qp_prepare_header(&rq_ctxt_block->hdr, HINIC_QP_CTXT_TYPE_RQ,
+				num_rqs, func_to_io->max_qps);
+	for (i = 0; i < num_rqs; i++) {
+		qp = &func_to_io->qps[i];
+
+		hinic_rq_prepare_ctxt(&rq_ctxt[i], &qp->rq,
+				      base_qpn + qp->q_id);
+	}
+
+	cmdq_buf.size = HINIC_RQ_CTXT_SIZE(num_rqs);
+
+	err = hinic_cmdq_direct_resp(&func_to_io->cmdqs, HINIC_MOD_L2NIC,
+				     IO_CMD_MODIFY_QUEUE_CTXT, &cmdq_buf,
+				     &out_param);
+	if (err || out_param != 0) {
+		dev_err(&pdev->dev, "Failed to set RQ ctxts\n");
+		err = -EFAULT;
+	}
+
+	hinic_free_cmdq_buf(&func_to_io->cmdqs, &cmdq_buf);
+	return err;
+}
+
+/**
+ * write_qp_ctxts - write the qp ctxt to HW
+ * @func_to_io: func to io channel that holds the IO components
+ * @base_qpn: first qp number
+ * @num_qps: number of qps to write
+ *
+ * Return 0 - Success, negative - Failure
+ **/
+static int write_qp_ctxts(struct hinic_func_to_io *func_to_io, u16 base_qpn,
+			  u16 num_qps)
+{
+	return (write_sq_ctxts(func_to_io, base_qpn, num_qps) ||
+		write_rq_ctxts(func_to_io, base_qpn, num_qps));
+}
+
+static int hinic_clean_queue_offload_ctxt(struct hinic_func_to_io *func_to_io,
+					  enum hinic_qp_ctxt_type ctxt_type)
+{
+	struct hinic_hwif *hwif = func_to_io->hwif;
+	struct hinic_clean_queue_ctxt *ctxt_block;
+	struct pci_dev *pdev = hwif->pdev;
+	struct hinic_cmdq_buf cmdq_buf;
+	u64 out_param = 0;
+	int err;
+
+	err = hinic_alloc_cmdq_buf(&func_to_io->cmdqs, &cmdq_buf);
+	if (err) {
+		dev_err(&pdev->dev, "Failed to allocate cmdq buf\n");
+		return err;
+	}
+
+	ctxt_block = cmdq_buf.buf;
+	ctxt_block->cmdq_hdr.num_queues = func_to_io->max_qps;
+	ctxt_block->cmdq_hdr.queue_type = ctxt_type;
+	ctxt_block->cmdq_hdr.addr_offset = 0;
+
+	/* TSO/LRO ctxt size: 0x0:0B; 0x1:160B; 0x2:200B; 0x3:240B */
+	ctxt_block->ctxt_size = 0x3;
+
+	hinic_cpu_to_be32(ctxt_block, sizeof(*ctxt_block));
+
+	cmdq_buf.size = sizeof(*ctxt_block);
+
+	err = hinic_cmdq_direct_resp(&func_to_io->cmdqs, HINIC_MOD_L2NIC,
+				     IO_CMD_CLEAN_QUEUE_CTXT,
+				     &cmdq_buf, &out_param);
+
+	if (err || out_param) {
+		dev_err(&pdev->dev, "Failed to clean offload ctxts, err: %d, out_param: 0x%llx\n",
+			err, out_param);
+
+		err = -EFAULT;
+	}
+
+	hinic_free_cmdq_buf(&func_to_io->cmdqs, &cmdq_buf);
+
+	return err;
+}
+
+static int hinic_clean_qp_offload_ctxt(struct hinic_func_to_io *func_to_io)
+{
+	/* clean LRO/TSO context space */
+	return (hinic_clean_queue_offload_ctxt(func_to_io,
+					       HINIC_QP_CTXT_TYPE_SQ) ||
+		hinic_clean_queue_offload_ctxt(func_to_io,
+					       HINIC_QP_CTXT_TYPE_RQ));
+}
+
+/**
+ * init_qp - Initialize a Queue Pair
+ * @func_to_io: func to io channel that holds the IO components
+ * @qp: pointer to the qp to initialize
+ * @q_id: the id of the qp
+ * @sq_msix_entry: msix entry for sq
+ * @rq_msix_entry: msix entry for rq
+ *
+ * Return 0 - Success, negative - Failure
+ **/
+static int init_qp(struct hinic_func_to_io *func_to_io,
+		   struct hinic_qp *qp, int q_id,
+		   struct msix_entry *sq_msix_entry,
+		   struct msix_entry *rq_msix_entry)
+{
+	struct hinic_hwif *hwif = func_to_io->hwif;
+	struct pci_dev *pdev = hwif->pdev;
+	void __iomem *db_base;
+	int err;
+
+	qp->q_id = q_id;
+
+	err = hinic_wq_allocate(&func_to_io->wqs, &func_to_io->sq_wq[q_id],
+				HINIC_SQ_WQEBB_SIZE, HINIC_SQ_PAGE_SIZE,
+				func_to_io->sq_depth, HINIC_SQ_WQE_MAX_SIZE);
+	if (err) {
+		dev_err(&pdev->dev, "Failed to allocate WQ for SQ\n");
+		return err;
+	}
+
+	err = hinic_wq_allocate(&func_to_io->wqs, &func_to_io->rq_wq[q_id],
+				HINIC_RQ_WQEBB_SIZE, HINIC_RQ_PAGE_SIZE,
+				func_to_io->rq_depth, HINIC_RQ_WQE_SIZE);
+	if (err) {
+		dev_err(&pdev->dev, "Failed to allocate WQ for RQ\n");
+		goto err_rq_alloc;
+	}
+
+	db_base = get_db_area(func_to_io);
+	if (IS_ERR(db_base)) {
+		dev_err(&pdev->dev, "Failed to get DB area for SQ\n");
+		err = PTR_ERR(db_base);
+		goto err_get_db;
+	}
+
+	func_to_io->sq_db[q_id] = db_base;
+
+	qp->sq.qid = q_id;
+	err = hinic_init_sq(&qp->sq, hwif, &func_to_io->sq_wq[q_id],
+			    sq_msix_entry,
+			    CI_ADDR(func_to_io->ci_addr_base, q_id),
+			    CI_ADDR(func_to_io->ci_dma_base, q_id), db_base);
+	if (err) {
+		dev_err(&pdev->dev, "Failed to init SQ\n");
+		goto err_sq_init;
+	}
+
+	qp->rq.qid = q_id;
+	err = hinic_init_rq(&qp->rq, hwif, &func_to_io->rq_wq[q_id],
+			    rq_msix_entry);
+	if (err) {
+		dev_err(&pdev->dev, "Failed to init RQ\n");
+		goto err_rq_init;
+	}
+
+	return 0;
+
+err_rq_init:
+	hinic_clean_sq(&qp->sq);
+
+err_sq_init:
+	return_db_area(func_to_io, db_base);
+
+err_get_db:
+	hinic_wq_free(&func_to_io->wqs, &func_to_io->rq_wq[q_id]);
+
+err_rq_alloc:
+	hinic_wq_free(&func_to_io->wqs, &func_to_io->sq_wq[q_id]);
+	return err;
+}
+
+/**
+ * destroy_qp - Clean the resources of a Queue Pair
+ * @func_to_io: func to io channel that holds the IO components
+ * @qp: pointer to the qp to clean
+ **/
+static void destroy_qp(struct hinic_func_to_io *func_to_io,
+		       struct hinic_qp *qp)
+{
+	int q_id = qp->q_id;
+
+	hinic_clean_rq(&qp->rq);
+	hinic_clean_sq(&qp->sq);
+
+	return_db_area(func_to_io, func_to_io->sq_db[q_id]);
+
+	hinic_wq_free(&func_to_io->wqs, &func_to_io->rq_wq[q_id]);
+	hinic_wq_free(&func_to_io->wqs, &func_to_io->sq_wq[q_id]);
+}
+
+/**
+ * hinic_io_create_qps - Create Queue Pairs
+ * @func_to_io: func to io channel that holds the IO components
+ * @base_qpn: base qp number
+ * @num_qps: number queue pairs to create
+ * @sq_msix_entries: msix entries for sq
+ * @rq_msix_entries: msix entries for rq
+ *
+ * Return 0 - Success, negative - Failure
+ **/
+int hinic_io_create_qps(struct hinic_func_to_io *func_to_io,
+			u16 base_qpn, int num_qps,
+			struct msix_entry *sq_msix_entries,
+			struct msix_entry *rq_msix_entries)
+{
+	struct hinic_hwif *hwif = func_to_io->hwif;
+	struct pci_dev *pdev = hwif->pdev;
+	void *ci_addr_base;
+	int i, j, err;
+
+	func_to_io->qps = devm_kcalloc(&pdev->dev, num_qps,
+				       sizeof(*func_to_io->qps), GFP_KERNEL);
+	if (!func_to_io->qps)
+		return -ENOMEM;
+
+	func_to_io->sq_wq = devm_kcalloc(&pdev->dev, num_qps,
+					 sizeof(*func_to_io->sq_wq), GFP_KERNEL);
+	if (!func_to_io->sq_wq) {
+		err = -ENOMEM;
+		goto err_sq_wq;
+	}
+
+	func_to_io->rq_wq = devm_kcalloc(&pdev->dev, num_qps,
+					 sizeof(*func_to_io->rq_wq), GFP_KERNEL);
+	if (!func_to_io->rq_wq) {
+		err = -ENOMEM;
+		goto err_rq_wq;
+	}
+
+	func_to_io->sq_db = devm_kcalloc(&pdev->dev, num_qps,
+					 sizeof(*func_to_io->sq_db), GFP_KERNEL);
+	if (!func_to_io->sq_db) {
+		err = -ENOMEM;
+		goto err_sq_db;
+	}
+
+	ci_addr_base = dma_alloc_coherent(&pdev->dev, CI_TABLE_SIZE(num_qps),
+					  &func_to_io->ci_dma_base,
+					  GFP_KERNEL);
+	if (!ci_addr_base) {
+		dev_err(&pdev->dev, "Failed to allocate CI area\n");
+		err = -ENOMEM;
+		goto err_ci_base;
+	}
+
+	func_to_io->ci_addr_base = ci_addr_base;
+
+	for (i = 0; i < num_qps; i++) {
+		err = init_qp(func_to_io, &func_to_io->qps[i], i,
+			      &sq_msix_entries[i], &rq_msix_entries[i]);
+		if (err) {
+			dev_err(&pdev->dev, "Failed to create QP %d\n", i);
+			goto err_init_qp;
+		}
+	}
+
+	err = write_qp_ctxts(func_to_io, base_qpn, num_qps);
+	if (err) {
+		dev_err(&pdev->dev, "Failed to init QP ctxts\n");
+		goto err_write_qp_ctxts;
+	}
+
+	err = hinic_clean_qp_offload_ctxt(func_to_io);
+	if (err) {
+		dev_err(&pdev->dev, "Failed to clean QP contexts space\n");
+		goto err_write_qp_ctxts;
+	}
+
+	return 0;
+
+err_write_qp_ctxts:
+err_init_qp:
+	for (j = 0; j < i; j++)
+		destroy_qp(func_to_io, &func_to_io->qps[j]);
+
+	dma_free_coherent(&pdev->dev, CI_TABLE_SIZE(num_qps),
+			  func_to_io->ci_addr_base, func_to_io->ci_dma_base);
+
+err_ci_base:
+	devm_kfree(&pdev->dev, func_to_io->sq_db);
+
+err_sq_db:
+	devm_kfree(&pdev->dev, func_to_io->rq_wq);
+
+err_rq_wq:
+	devm_kfree(&pdev->dev, func_to_io->sq_wq);
+
+err_sq_wq:
+	devm_kfree(&pdev->dev, func_to_io->qps);
+	return err;
+}
+
+/**
+ * hinic_io_destroy_qps - Destroy the IO Queue Pairs
+ * @func_to_io: func to io channel that holds the IO components
+ * @num_qps: number queue pairs to destroy
+ **/
+void hinic_io_destroy_qps(struct hinic_func_to_io *func_to_io, int num_qps)
+{
+	struct hinic_hwif *hwif = func_to_io->hwif;
+	struct pci_dev *pdev = hwif->pdev;
+	size_t ci_table_size;
+	int i;
+
+	ci_table_size = CI_TABLE_SIZE(num_qps);
+
+	for (i = 0; i < num_qps; i++)
+		destroy_qp(func_to_io, &func_to_io->qps[i]);
+
+	dma_free_coherent(&pdev->dev, ci_table_size, func_to_io->ci_addr_base,
+			  func_to_io->ci_dma_base);
+
+	devm_kfree(&pdev->dev, func_to_io->sq_db);
+
+	devm_kfree(&pdev->dev, func_to_io->rq_wq);
+	devm_kfree(&pdev->dev, func_to_io->sq_wq);
+
+	devm_kfree(&pdev->dev, func_to_io->qps);
+}
+
+int hinic_set_wq_page_size(struct hinic_hwdev *hwdev, u16 func_idx,
+			   u32 page_size)
+{
+	struct hinic_wq_page_size page_size_info = {0};
+	u16 out_size = sizeof(page_size_info);
+	struct hinic_pfhwdev *pfhwdev;
+	int err;
+
+	pfhwdev = container_of(hwdev, struct hinic_pfhwdev, hwdev);
+
+	page_size_info.func_idx = func_idx;
+	page_size_info.ppf_idx = HINIC_HWIF_PPF_IDX(hwdev->hwif);
+	page_size_info.page_size = HINIC_PAGE_SIZE_HW(page_size);
+
+	err = hinic_msg_to_mgmt(&pfhwdev->pf_to_mgmt, HINIC_MOD_COMM,
+				HINIC_COMM_CMD_PAGESIZE_SET, &page_size_info,
+				sizeof(page_size_info), &page_size_info,
+				&out_size, HINIC_MGMT_MSG_SYNC);
+	if (err || !out_size || page_size_info.status) {
+		dev_err(&hwdev->hwif->pdev->dev, "Failed to set wq page size, err: %d, status: 0x%x, out_size: 0x%0x\n",
+			err, page_size_info.status, out_size);
+		return -EFAULT;
+	}
+
+	return 0;
+}
+
+/**
+ * hinic_io_init - Initialize the IO components
+ * @func_to_io: func to io channel that holds the IO components
+ * @hwif: HW interface for accessing IO
+ * @max_qps: maximum QPs in HW
+ * @num_ceqs: number completion event queues
+ * @ceq_msix_entries: msix entries for ceqs
+ *
+ * Return 0 - Success, negative - Failure
+ **/
+int hinic_io_init(struct hinic_func_to_io *func_to_io,
+		  struct hinic_hwif *hwif, u16 max_qps, int num_ceqs,
+		  struct msix_entry *ceq_msix_entries)
+{
+	struct pci_dev *pdev = hwif->pdev;
+	enum hinic_cmdq_type cmdq, type;
+	void __iomem *db_area;
+	int err;
+
+	func_to_io->hwif = hwif;
+	func_to_io->qps = NULL;
+	func_to_io->max_qps = max_qps;
+	func_to_io->ceqs.hwdev = func_to_io->hwdev;
+
+	err = hinic_ceqs_init(&func_to_io->ceqs, hwif, num_ceqs,
+			      HINIC_DEFAULT_CEQ_LEN, HINIC_EQ_PAGE_SIZE,
+			      ceq_msix_entries);
+	if (err) {
+		dev_err(&pdev->dev, "Failed to init CEQs\n");
+		return err;
+	}
+
+	err = hinic_wqs_alloc(&func_to_io->wqs, 2 * max_qps, hwif);
+	if (err) {
+		dev_err(&pdev->dev, "Failed to allocate WQS for IO\n");
+		goto err_wqs_alloc;
+	}
+
+	func_to_io->db_base = pci_ioremap_bar(pdev, HINIC_PCI_DB_BAR);
+	if (!func_to_io->db_base) {
+		dev_err(&pdev->dev, "Failed to remap IO DB area\n");
+		err = -ENOMEM;
+		goto err_db_ioremap;
+	}
+
+	init_db_area_idx(&func_to_io->free_db_area);
+
+	for (cmdq = HINIC_CMDQ_SYNC; cmdq < HINIC_MAX_CMDQ_TYPES; cmdq++) {
+		db_area = get_db_area(func_to_io);
+		if (IS_ERR(db_area)) {
+			dev_err(&pdev->dev, "Failed to get cmdq db area\n");
+			err = PTR_ERR(db_area);
+			goto err_db_area;
+		}
+
+		func_to_io->cmdq_db_area[cmdq] = db_area;
+	}
+
+	err = hinic_set_wq_page_size(func_to_io->hwdev,
+				     HINIC_HWIF_FUNC_IDX(hwif),
+				     HINIC_DEFAULT_WQ_PAGE_SIZE);
+	if (err) {
+		dev_err(&func_to_io->hwif->pdev->dev, "Failed to set wq page size\n");
+		goto init_wq_pg_size_err;
+	}
+
+	err = hinic_init_cmdqs(&func_to_io->cmdqs, hwif,
+			       func_to_io->cmdq_db_area);
+	if (err) {
+		dev_err(&pdev->dev, "Failed to initialize cmdqs\n");
+		goto err_init_cmdqs;
+	}
+
+	return 0;
+
+err_init_cmdqs:
+	if (!HINIC_IS_VF(func_to_io->hwif))
+		hinic_set_wq_page_size(func_to_io->hwdev,
+				       HINIC_HWIF_FUNC_IDX(hwif),
+				       HINIC_HW_WQ_PAGE_SIZE);
+init_wq_pg_size_err:
+err_db_area:
+	for (type = HINIC_CMDQ_SYNC; type < cmdq; type++)
+		return_db_area(func_to_io, func_to_io->cmdq_db_area[type]);
+
+	iounmap(func_to_io->db_base);
+
+err_db_ioremap:
+	hinic_wqs_free(&func_to_io->wqs);
+
+err_wqs_alloc:
+	hinic_ceqs_free(&func_to_io->ceqs);
+	return err;
+}
+
+/**
+ * hinic_io_free - Free the IO components
+ * @func_to_io: func to io channel that holds the IO components
+ **/
+void hinic_io_free(struct hinic_func_to_io *func_to_io)
+{
+	enum hinic_cmdq_type cmdq;
+
+	hinic_free_cmdqs(&func_to_io->cmdqs);
+
+	if (!HINIC_IS_VF(func_to_io->hwif))
+		hinic_set_wq_page_size(func_to_io->hwdev,
+				       HINIC_HWIF_FUNC_IDX(func_to_io->hwif),
+				       HINIC_HW_WQ_PAGE_SIZE);
+
+	for (cmdq = HINIC_CMDQ_SYNC; cmdq < HINIC_MAX_CMDQ_TYPES; cmdq++)
+		return_db_area(func_to_io, func_to_io->cmdq_db_area[cmdq]);
+
+	iounmap(func_to_io->db_base);
+	hinic_wqs_free(&func_to_io->wqs);
+	hinic_ceqs_free(&func_to_io->ceqs);
+}
diff --git a/drivers/net/ethernet/huawei/hinic/hinic_hw_io.h b/drivers/net/ethernet/huawei/hinic/hinic_hw_io.h
new file mode 100644
index 000000000000..52159a90278a
--- /dev/null
+++ b/drivers/net/ethernet/huawei/hinic/hinic_hw_io.h
@@ -0,0 +1,123 @@
+/* SPDX-License-Identifier: GPL-2.0-only */
+/*
+ * Huawei HiNIC PCI Express Linux driver
+ * Copyright(c) 2017 Huawei Technologies Co., Ltd
+ */
+
+#ifndef HINIC_HW_IO_H
+#define HINIC_HW_IO_H
+
+#include <linux/types.h>
+#include <linux/pci.h>
+#include <linux/semaphore.h>
+#include <linux/sizes.h>
+
+#include "hinic_hw_if.h"
+#include "hinic_hw_eqs.h"
+#include "hinic_hw_wq.h"
+#include "hinic_hw_cmdq.h"
+#include "hinic_hw_qp.h"
+
+#define HINIC_DB_PAGE_SIZE      SZ_4K
+#define HINIC_DB_SIZE           SZ_4M
+#define HINIC_HW_WQ_PAGE_SIZE	SZ_4K
+#define HINIC_DEFAULT_WQ_PAGE_SIZE SZ_256K
+
+#define HINIC_DB_MAX_AREAS      (HINIC_DB_SIZE / HINIC_DB_PAGE_SIZE)
+
+enum hinic_db_type {
+	HINIC_DB_CMDQ_TYPE,
+	HINIC_DB_SQ_TYPE,
+};
+
+enum hinic_io_path {
+	HINIC_CTRL_PATH,
+	HINIC_DATA_PATH,
+};
+
+struct hinic_free_db_area {
+	int             db_idx[HINIC_DB_MAX_AREAS];
+
+	int             alloc_pos;
+	int             return_pos;
+
+	int             num_free;
+
+	/* Lock for getting db area */
+	struct semaphore        idx_lock;
+};
+
+struct hinic_nic_cfg {
+	/* lock for getting nic cfg */
+	struct mutex		cfg_mutex;
+	bool			pause_set;
+	u32			auto_neg;
+	u32			rx_pause;
+	u32			tx_pause;
+};
+
+struct hinic_func_to_io {
+	struct hinic_hwif       *hwif;
+	struct hinic_hwdev      *hwdev;
+	u16			global_qpn;
+	struct hinic_ceqs       ceqs;
+
+	struct hinic_wqs        wqs;
+
+	struct hinic_wq         *sq_wq;
+	struct hinic_wq         *rq_wq;
+
+	struct hinic_qp         *qps;
+	u16                     max_qps;
+
+	u16			sq_depth;
+	u16			rq_depth;
+
+	void __iomem            **sq_db;
+	void __iomem            *db_base;
+
+	void                    *ci_addr_base;
+	dma_addr_t              ci_dma_base;
+
+	struct hinic_free_db_area       free_db_area;
+
+	void __iomem                    *cmdq_db_area[HINIC_MAX_CMDQ_TYPES];
+
+	struct hinic_cmdqs              cmdqs;
+
+	u16			max_vfs;
+	struct vf_data_storage	*vf_infos;
+	u8			link_status;
+	struct hinic_nic_cfg	nic_cfg;
+};
+
+struct hinic_wq_page_size {
+	u8	status;
+	u8	version;
+	u8	rsvd0[6];
+
+	u16	func_idx;
+	u8	ppf_idx;
+	u8	page_size;
+
+	u32	rsvd1;
+};
+
+int hinic_set_wq_page_size(struct hinic_hwdev *hwdev, u16 func_idx,
+			   u32 page_size);
+
+int hinic_io_create_qps(struct hinic_func_to_io *func_to_io,
+			u16 base_qpn, int num_qps,
+			struct msix_entry *sq_msix_entries,
+			struct msix_entry *rq_msix_entries);
+
+void hinic_io_destroy_qps(struct hinic_func_to_io *func_to_io,
+			  int num_qps);
+
+int hinic_io_init(struct hinic_func_to_io *func_to_io,
+		  struct hinic_hwif *hwif, u16 max_qps, int num_ceqs,
+		  struct msix_entry *ceq_msix_entries);
+
+void hinic_io_free(struct hinic_func_to_io *func_to_io);
+
+#endif
diff --git a/drivers/net/ethernet/huawei/hinic/hinic_hw_mbox.c b/drivers/net/ethernet/huawei/hinic/hinic_hw_mbox.c
new file mode 100644
index 000000000000..97c1584dc05b
--- /dev/null
+++ b/drivers/net/ethernet/huawei/hinic/hinic_hw_mbox.c
@@ -0,0 +1,1515 @@
+// SPDX-License-Identifier: GPL-2.0-only
+/* Huawei HiNIC PCI Express Linux driver
+ * Copyright(c) 2017 Huawei Technologies Co., Ltd
+ */
+#include <linux/pci.h>
+#include <linux/delay.h>
+#include <linux/types.h>
+#include <linux/completion.h>
+#include <linux/semaphore.h>
+#include <linux/spinlock.h>
+#include <linux/workqueue.h>
+
+#include "hinic_hw_if.h"
+#include "hinic_hw_mgmt.h"
+#include "hinic_hw_csr.h"
+#include "hinic_hw_dev.h"
+#include "hinic_hw_mbox.h"
+
+#define HINIC_MBOX_INT_DST_FUNC_SHIFT				0
+#define HINIC_MBOX_INT_DST_AEQN_SHIFT				10
+#define HINIC_MBOX_INT_SRC_RESP_AEQN_SHIFT			12
+#define HINIC_MBOX_INT_STAT_DMA_SHIFT				14
+/* The size of data to be sended (unit of 4 bytes) */
+#define HINIC_MBOX_INT_TX_SIZE_SHIFT				20
+/* SO_RO(strong order, relax order) */
+#define HINIC_MBOX_INT_STAT_DMA_SO_RO_SHIFT			25
+#define HINIC_MBOX_INT_WB_EN_SHIFT				28
+
+#define HINIC_MBOX_INT_DST_FUNC_MASK				0x3FF
+#define HINIC_MBOX_INT_DST_AEQN_MASK				0x3
+#define HINIC_MBOX_INT_SRC_RESP_AEQN_MASK			0x3
+#define HINIC_MBOX_INT_STAT_DMA_MASK				0x3F
+#define HINIC_MBOX_INT_TX_SIZE_MASK				0x1F
+#define HINIC_MBOX_INT_STAT_DMA_SO_RO_MASK			0x3
+#define HINIC_MBOX_INT_WB_EN_MASK				0x1
+
+#define HINIC_MBOX_INT_SET(val, field)	\
+			(((val) & HINIC_MBOX_INT_##field##_MASK) << \
+			HINIC_MBOX_INT_##field##_SHIFT)
+
+enum hinic_mbox_tx_status {
+	TX_NOT_DONE = 1,
+};
+
+#define HINIC_MBOX_CTRL_TRIGGER_AEQE_SHIFT			0
+
+/* specifies the issue request for the message data.
+ * 0 - Tx request is done;
+ * 1 - Tx request is in process.
+ */
+#define HINIC_MBOX_CTRL_TX_STATUS_SHIFT				1
+
+#define HINIC_MBOX_CTRL_TRIGGER_AEQE_MASK			0x1
+#define HINIC_MBOX_CTRL_TX_STATUS_MASK				0x1
+
+#define HINIC_MBOX_CTRL_SET(val, field)	\
+			(((val) & HINIC_MBOX_CTRL_##field##_MASK) << \
+			HINIC_MBOX_CTRL_##field##_SHIFT)
+
+#define HINIC_MBOX_HEADER_MSG_LEN_SHIFT				0
+#define HINIC_MBOX_HEADER_MODULE_SHIFT				11
+#define HINIC_MBOX_HEADER_SEG_LEN_SHIFT				16
+#define HINIC_MBOX_HEADER_NO_ACK_SHIFT				22
+#define HINIC_MBOX_HEADER_SEQID_SHIFT				24
+#define HINIC_MBOX_HEADER_LAST_SHIFT				30
+
+/* specifies the mailbox message direction
+ * 0 - send
+ * 1 - receive
+ */
+#define HINIC_MBOX_HEADER_DIRECTION_SHIFT			31
+#define HINIC_MBOX_HEADER_CMD_SHIFT				32
+#define HINIC_MBOX_HEADER_MSG_ID_SHIFT				40
+#define HINIC_MBOX_HEADER_STATUS_SHIFT				48
+#define HINIC_MBOX_HEADER_SRC_GLB_FUNC_IDX_SHIFT		54
+
+#define HINIC_MBOX_HEADER_MSG_LEN_MASK				0x7FF
+#define HINIC_MBOX_HEADER_MODULE_MASK				0x1F
+#define HINIC_MBOX_HEADER_SEG_LEN_MASK				0x3F
+#define HINIC_MBOX_HEADER_NO_ACK_MASK				0x1
+#define HINIC_MBOX_HEADER_SEQID_MASK				0x3F
+#define HINIC_MBOX_HEADER_LAST_MASK				0x1
+#define HINIC_MBOX_HEADER_DIRECTION_MASK			0x1
+#define HINIC_MBOX_HEADER_CMD_MASK				0xFF
+#define HINIC_MBOX_HEADER_MSG_ID_MASK				0xFF
+#define HINIC_MBOX_HEADER_STATUS_MASK				0x3F
+#define HINIC_MBOX_HEADER_SRC_GLB_FUNC_IDX_MASK			0x3FF
+
+#define HINIC_MBOX_HEADER_GET(val, field)	\
+			(((val) >> HINIC_MBOX_HEADER_##field##_SHIFT) & \
+			HINIC_MBOX_HEADER_##field##_MASK)
+#define HINIC_MBOX_HEADER_SET(val, field)	\
+			((u64)((val) & HINIC_MBOX_HEADER_##field##_MASK) << \
+			HINIC_MBOX_HEADER_##field##_SHIFT)
+
+#define MBOX_SEGLEN_MASK			\
+		HINIC_MBOX_HEADER_SET(HINIC_MBOX_HEADER_SEG_LEN_MASK, SEG_LEN)
+
+#define HINIC_MBOX_SEG_LEN			48
+#define HINIC_MBOX_COMP_TIME			8000U
+#define MBOX_MSG_POLLING_TIMEOUT		8000
+
+#define HINIC_MBOX_DATA_SIZE			2040
+
+#define MBOX_MAX_BUF_SZ				2048UL
+#define MBOX_HEADER_SZ				8
+
+#define MBOX_INFO_SZ				4
+
+/* MBOX size is 64B, 8B for mbox_header, 4B reserved */
+#define MBOX_SEG_LEN				48
+#define MBOX_SEG_LEN_ALIGN			4
+#define MBOX_WB_STATUS_LEN			16UL
+
+/* mbox write back status is 16B, only first 4B is used */
+#define MBOX_WB_STATUS_ERRCODE_MASK		0xFFFF
+#define MBOX_WB_STATUS_MASK			0xFF
+#define MBOX_WB_ERROR_CODE_MASK			0xFF00
+#define MBOX_WB_STATUS_FINISHED_SUCCESS		0xFF
+#define MBOX_WB_STATUS_NOT_FINISHED		0x00
+
+#define MBOX_STATUS_FINISHED(wb)	\
+	(((wb) & MBOX_WB_STATUS_MASK) != MBOX_WB_STATUS_NOT_FINISHED)
+#define MBOX_STATUS_SUCCESS(wb)		\
+	(((wb) & MBOX_WB_STATUS_MASK) == MBOX_WB_STATUS_FINISHED_SUCCESS)
+#define MBOX_STATUS_ERRCODE(wb)		\
+	((wb) & MBOX_WB_ERROR_CODE_MASK)
+
+#define SEQ_ID_START_VAL			0
+#define SEQ_ID_MAX_VAL				42
+
+#define NO_DMA_ATTRIBUTE_VAL			0
+
+#define HINIC_MBOX_RSP_AEQN			2
+#define HINIC_MBOX_RECV_AEQN			0
+
+#define MBOX_MSG_NO_DATA_LEN			1
+
+#define MBOX_BODY_FROM_HDR(header)	((u8 *)(header) + MBOX_HEADER_SZ)
+#define MBOX_AREA(hwif)			\
+	((hwif)->cfg_regs_bar + HINIC_FUNC_CSR_MAILBOX_DATA_OFF)
+
+#define IS_PF_OR_PPF_SRC(src_func_idx)	((src_func_idx) < HINIC_MAX_PF_FUNCS)
+
+#define MBOX_MSG_ID_MASK		0xFF
+#define MBOX_MSG_ID(func_to_func)	((func_to_func)->send_msg_id)
+#define MBOX_MSG_ID_INC(func_to_func_mbox) (MBOX_MSG_ID(func_to_func_mbox) = \
+			(MBOX_MSG_ID(func_to_func_mbox) + 1) & MBOX_MSG_ID_MASK)
+
+#define FUNC_ID_OFF_SET_8B		8
+
+/* max message counter wait to process for one function */
+#define HINIC_MAX_MSG_CNT_TO_PROCESS	10
+
+#define HINIC_QUEUE_MIN_DEPTH		6
+#define HINIC_QUEUE_MAX_DEPTH		12
+#define HINIC_MAX_RX_BUFFER_SIZE		15
+
+enum hinic_hwif_direction_type {
+	HINIC_HWIF_DIRECT_SEND	= 0,
+	HINIC_HWIF_RESPONSE	= 1,
+};
+
+enum mbox_send_mod {
+	MBOX_SEND_MSG_INT,
+};
+
+enum mbox_seg_type {
+	NOT_LAST_SEG,
+	LAST_SEG,
+};
+
+enum mbox_ordering_type {
+	STRONG_ORDER,
+};
+
+enum mbox_write_back_type {
+	WRITE_BACK = 1,
+};
+
+enum mbox_aeq_trig_type {
+	NOT_TRIGGER,
+	TRIGGER,
+};
+
+static bool check_func_id(struct hinic_hwdev *hwdev, u16 src_func_idx,
+			  const void *buf_in, u16 in_size, u16 offset)
+{
+	u16 func_idx;
+
+	if (in_size < offset + sizeof(func_idx)) {
+		dev_warn(&hwdev->hwif->pdev->dev,
+			 "Receive mailbox msg len: %d less than %d Bytes is invalid\n",
+			 in_size, offset);
+		return false;
+	}
+
+	func_idx = *((u16 *)((u8 *)buf_in + offset));
+
+	if (src_func_idx != func_idx) {
+		dev_warn(&hwdev->hwif->pdev->dev,
+			 "Receive mailbox function id: 0x%x not equal to msg function id: 0x%x\n",
+			 src_func_idx, func_idx);
+		return false;
+	}
+
+	return true;
+}
+
+bool hinic_mbox_check_func_id_8B(struct hinic_hwdev *hwdev, u16 func_idx,
+				 void *buf_in, u16 in_size)
+{
+	return check_func_id(hwdev, func_idx, buf_in, in_size,
+			     FUNC_ID_OFF_SET_8B);
+}
+
+/**
+ * hinic_register_pf_mbox_cb - register mbox callback for pf
+ * @hwdev: the pointer to hw device
+ * @mod:	specific mod that the callback will handle
+ * @callback:	callback function
+ * Return: 0 - success, negative - failure
+ */
+int hinic_register_pf_mbox_cb(struct hinic_hwdev *hwdev,
+			      enum hinic_mod_type mod,
+			      hinic_pf_mbox_cb callback)
+{
+	struct hinic_mbox_func_to_func *func_to_func = hwdev->func_to_func;
+
+	if (mod >= HINIC_MOD_MAX)
+		return -EFAULT;
+
+	func_to_func->pf_mbox_cb[mod] = callback;
+
+	set_bit(HINIC_PF_MBOX_CB_REG, &func_to_func->pf_mbox_cb_state[mod]);
+
+	return 0;
+}
+
+/**
+ * hinic_register_vf_mbox_cb - register mbox callback for vf
+ * @hwdev: the pointer to hw device
+ * @mod:	specific mod that the callback will handle
+ * @callback:	callback function
+ * Return: 0 - success, negative - failure
+ */
+int hinic_register_vf_mbox_cb(struct hinic_hwdev *hwdev,
+			      enum hinic_mod_type mod,
+			      hinic_vf_mbox_cb callback)
+{
+	struct hinic_mbox_func_to_func *func_to_func = hwdev->func_to_func;
+
+	if (mod >= HINIC_MOD_MAX)
+		return -EFAULT;
+
+	func_to_func->vf_mbox_cb[mod] = callback;
+
+	set_bit(HINIC_VF_MBOX_CB_REG, &func_to_func->vf_mbox_cb_state[mod]);
+
+	return 0;
+}
+
+/**
+ * hinic_unregister_pf_mbox_cb - unregister the mbox callback for pf
+ * @hwdev:	the pointer to hw device
+ * @mod:	specific mod that the callback will handle
+ */
+void hinic_unregister_pf_mbox_cb(struct hinic_hwdev *hwdev,
+				 enum hinic_mod_type mod)
+{
+	struct hinic_mbox_func_to_func *func_to_func = hwdev->func_to_func;
+
+	clear_bit(HINIC_PF_MBOX_CB_REG, &func_to_func->pf_mbox_cb_state[mod]);
+
+	while (test_bit(HINIC_PF_MBOX_CB_RUNNING,
+			&func_to_func->pf_mbox_cb_state[mod]))
+		usleep_range(900, 1000);
+
+	func_to_func->pf_mbox_cb[mod] = NULL;
+}
+
+/**
+ * hinic_unregister_vf_mbox_cb - unregister the mbox callback for vf
+ * @hwdev:	the pointer to hw device
+ * @mod:	specific mod that the callback will handle
+ */
+void hinic_unregister_vf_mbox_cb(struct hinic_hwdev *hwdev,
+				 enum hinic_mod_type mod)
+{
+	struct hinic_mbox_func_to_func *func_to_func = hwdev->func_to_func;
+
+	clear_bit(HINIC_VF_MBOX_CB_REG, &func_to_func->vf_mbox_cb_state[mod]);
+
+	while (test_bit(HINIC_VF_MBOX_CB_RUNNING,
+			&func_to_func->vf_mbox_cb_state[mod]))
+		usleep_range(900, 1000);
+
+	func_to_func->vf_mbox_cb[mod] = NULL;
+}
+
+static int recv_vf_mbox_handler(struct hinic_mbox_func_to_func *func_to_func,
+				struct hinic_recv_mbox *recv_mbox,
+				void *buf_out, u16 *out_size)
+{
+	hinic_vf_mbox_cb cb;
+	int ret = 0;
+
+	if (recv_mbox->mod >= HINIC_MOD_MAX) {
+		dev_err(&func_to_func->hwif->pdev->dev, "Receive illegal mbox message, mod = %d\n",
+			recv_mbox->mod);
+		return -EINVAL;
+	}
+
+	set_bit(HINIC_VF_MBOX_CB_RUNNING,
+		&func_to_func->vf_mbox_cb_state[recv_mbox->mod]);
+
+	cb = func_to_func->vf_mbox_cb[recv_mbox->mod];
+	if (cb && test_bit(HINIC_VF_MBOX_CB_REG,
+			   &func_to_func->vf_mbox_cb_state[recv_mbox->mod])) {
+		cb(func_to_func->hwdev, recv_mbox->cmd, recv_mbox->mbox,
+		   recv_mbox->mbox_len, buf_out, out_size);
+	} else {
+		dev_err(&func_to_func->hwif->pdev->dev, "VF mbox cb is not registered\n");
+		ret = -EINVAL;
+	}
+
+	clear_bit(HINIC_VF_MBOX_CB_RUNNING,
+		  &func_to_func->vf_mbox_cb_state[recv_mbox->mod]);
+
+	return ret;
+}
+
+static int
+recv_pf_from_vf_mbox_handler(struct hinic_mbox_func_to_func *func_to_func,
+			     struct hinic_recv_mbox *recv_mbox,
+			     u16 src_func_idx, void *buf_out,
+			     u16 *out_size)
+{
+	hinic_pf_mbox_cb cb;
+	u16 vf_id = 0;
+	int ret;
+
+	if (recv_mbox->mod >= HINIC_MOD_MAX) {
+		dev_err(&func_to_func->hwif->pdev->dev, "Receive illegal mbox message, mod = %d\n",
+			recv_mbox->mod);
+		return -EINVAL;
+	}
+
+	set_bit(HINIC_PF_MBOX_CB_RUNNING,
+		&func_to_func->pf_mbox_cb_state[recv_mbox->mod]);
+
+	cb = func_to_func->pf_mbox_cb[recv_mbox->mod];
+	if (cb && test_bit(HINIC_PF_MBOX_CB_REG,
+			   &func_to_func->pf_mbox_cb_state[recv_mbox->mod])) {
+		vf_id = src_func_idx -
+			hinic_glb_pf_vf_offset(func_to_func->hwif);
+		ret = cb(func_to_func->hwdev, vf_id, recv_mbox->cmd,
+			 recv_mbox->mbox, recv_mbox->mbox_len,
+			 buf_out, out_size);
+	} else {
+		dev_err(&func_to_func->hwif->pdev->dev, "PF mbox mod(0x%x) cb is not registered\n",
+			recv_mbox->mod);
+		ret = -EINVAL;
+	}
+
+	clear_bit(HINIC_PF_MBOX_CB_RUNNING,
+		  &func_to_func->pf_mbox_cb_state[recv_mbox->mod]);
+
+	return ret;
+}
+
+static bool check_mbox_seq_id_and_seg_len(struct hinic_recv_mbox *recv_mbox,
+					  u8 seq_id, u8 seg_len)
+{
+	if (seq_id > SEQ_ID_MAX_VAL || seg_len > MBOX_SEG_LEN)
+		return false;
+
+	if (seq_id == 0) {
+		recv_mbox->seq_id = seq_id;
+	} else {
+		if (seq_id != recv_mbox->seq_id + 1)
+			return false;
+
+		recv_mbox->seq_id = seq_id;
+	}
+
+	return true;
+}
+
+static void resp_mbox_handler(struct hinic_mbox_func_to_func *func_to_func,
+			      struct hinic_recv_mbox *recv_mbox)
+{
+	spin_lock(&func_to_func->mbox_lock);
+	if (recv_mbox->msg_info.msg_id == func_to_func->send_msg_id &&
+	    func_to_func->event_flag == EVENT_START)
+		complete(&recv_mbox->recv_done);
+	else
+		dev_err(&func_to_func->hwif->pdev->dev,
+			"Mbox response timeout, current send msg id(0x%x), recv msg id(0x%x), status(0x%x)\n",
+			func_to_func->send_msg_id, recv_mbox->msg_info.msg_id,
+			recv_mbox->msg_info.status);
+	spin_unlock(&func_to_func->mbox_lock);
+}
+
+static void recv_func_mbox_handler(struct hinic_mbox_func_to_func *func_to_func,
+				   struct hinic_recv_mbox *recv_mbox,
+				   u16 src_func_idx);
+
+static void recv_func_mbox_work_handler(struct work_struct *work)
+{
+	struct hinic_mbox_work *mbox_work =
+			container_of(work, struct hinic_mbox_work, work);
+	struct hinic_recv_mbox *recv_mbox;
+
+	recv_func_mbox_handler(mbox_work->func_to_func, mbox_work->recv_mbox,
+			       mbox_work->src_func_idx);
+
+	recv_mbox =
+		&mbox_work->func_to_func->mbox_send[mbox_work->src_func_idx];
+
+	atomic_dec(&recv_mbox->msg_cnt);
+
+	kfree(mbox_work);
+}
+
+static void recv_mbox_handler(struct hinic_mbox_func_to_func *func_to_func,
+			      void *header, struct hinic_recv_mbox *recv_mbox)
+{
+	void *mbox_body = MBOX_BODY_FROM_HDR(header);
+	struct hinic_recv_mbox *rcv_mbox_temp = NULL;
+	u64 mbox_header = *((u64 *)header);
+	struct hinic_mbox_work *mbox_work;
+	u8 seq_id, seg_len;
+	u16 src_func_idx;
+	int pos;
+
+	seq_id = HINIC_MBOX_HEADER_GET(mbox_header, SEQID);
+	seg_len = HINIC_MBOX_HEADER_GET(mbox_header, SEG_LEN);
+	src_func_idx = HINIC_MBOX_HEADER_GET(mbox_header, SRC_GLB_FUNC_IDX);
+
+	if (!check_mbox_seq_id_and_seg_len(recv_mbox, seq_id, seg_len)) {
+		dev_err(&func_to_func->hwif->pdev->dev,
+			"Mailbox sequence and segment check fail, src func id: 0x%x, front id: 0x%x, current id: 0x%x, seg len: 0x%x\n",
+			src_func_idx, recv_mbox->seq_id, seq_id, seg_len);
+		recv_mbox->seq_id = SEQ_ID_MAX_VAL;
+		return;
+	}
+
+	pos = seq_id * MBOX_SEG_LEN;
+	memcpy((u8 *)recv_mbox->mbox + pos, mbox_body,
+	       HINIC_MBOX_HEADER_GET(mbox_header, SEG_LEN));
+
+	if (!HINIC_MBOX_HEADER_GET(mbox_header, LAST))
+		return;
+
+	recv_mbox->cmd = HINIC_MBOX_HEADER_GET(mbox_header, CMD);
+	recv_mbox->mod = HINIC_MBOX_HEADER_GET(mbox_header, MODULE);
+	recv_mbox->mbox_len = HINIC_MBOX_HEADER_GET(mbox_header, MSG_LEN);
+	recv_mbox->ack_type = HINIC_MBOX_HEADER_GET(mbox_header, NO_ACK);
+	recv_mbox->msg_info.msg_id = HINIC_MBOX_HEADER_GET(mbox_header, MSG_ID);
+	recv_mbox->msg_info.status = HINIC_MBOX_HEADER_GET(mbox_header, STATUS);
+	recv_mbox->seq_id = SEQ_ID_MAX_VAL;
+
+	if (HINIC_MBOX_HEADER_GET(mbox_header, DIRECTION) ==
+	    HINIC_HWIF_RESPONSE) {
+		resp_mbox_handler(func_to_func, recv_mbox);
+		return;
+	}
+
+	if (atomic_read(&recv_mbox->msg_cnt) > HINIC_MAX_MSG_CNT_TO_PROCESS) {
+		dev_warn(&func_to_func->hwif->pdev->dev,
+			 "This function(%u) have %d message wait to process,can't add to work queue\n",
+			 src_func_idx, atomic_read(&recv_mbox->msg_cnt));
+		return;
+	}
+
+	rcv_mbox_temp = kmemdup(recv_mbox, sizeof(*rcv_mbox_temp), GFP_KERNEL);
+	if (!rcv_mbox_temp)
+		return;
+
+	rcv_mbox_temp->mbox = kmemdup(recv_mbox->mbox, MBOX_MAX_BUF_SZ,
+				      GFP_KERNEL);
+	if (!rcv_mbox_temp->mbox)
+		goto err_alloc_rcv_mbox_msg;
+
+	rcv_mbox_temp->buf_out = kzalloc(MBOX_MAX_BUF_SZ, GFP_KERNEL);
+	if (!rcv_mbox_temp->buf_out)
+		goto err_alloc_rcv_mbox_buf;
+
+	mbox_work = kzalloc(sizeof(*mbox_work), GFP_KERNEL);
+	if (!mbox_work)
+		goto err_alloc_mbox_work;
+
+	mbox_work->func_to_func = func_to_func;
+	mbox_work->recv_mbox = rcv_mbox_temp;
+	mbox_work->src_func_idx = src_func_idx;
+
+	atomic_inc(&recv_mbox->msg_cnt);
+	INIT_WORK(&mbox_work->work, recv_func_mbox_work_handler);
+	queue_work(func_to_func->workq, &mbox_work->work);
+
+	return;
+
+err_alloc_mbox_work:
+	kfree(rcv_mbox_temp->buf_out);
+
+err_alloc_rcv_mbox_buf:
+	kfree(rcv_mbox_temp->mbox);
+
+err_alloc_rcv_mbox_msg:
+	kfree(rcv_mbox_temp);
+}
+
+static int set_vf_mbox_random_id(struct hinic_hwdev *hwdev, u16 func_id)
+{
+	struct hinic_mbox_func_to_func *func_to_func = hwdev->func_to_func;
+	struct hinic_set_random_id rand_info = {0};
+	u16 out_size = sizeof(rand_info);
+	struct hinic_pfhwdev *pfhwdev;
+	int ret;
+
+	pfhwdev = container_of(hwdev, struct hinic_pfhwdev, hwdev);
+
+	rand_info.version = HINIC_CMD_VER_FUNC_ID;
+	rand_info.func_idx = func_id;
+	rand_info.vf_in_pf = func_id - hinic_glb_pf_vf_offset(hwdev->hwif);
+	rand_info.random_id = get_random_u32();
+
+	func_to_func->vf_mbx_rand_id[func_id] = rand_info.random_id;
+
+	ret = hinic_msg_to_mgmt(&pfhwdev->pf_to_mgmt, HINIC_MOD_COMM,
+				HINIC_MGMT_CMD_SET_VF_RANDOM_ID,
+				&rand_info, sizeof(rand_info),
+				&rand_info, &out_size, HINIC_MGMT_MSG_SYNC);
+	if ((rand_info.status != HINIC_MGMT_CMD_UNSUPPORTED &&
+	     rand_info.status) || !out_size || ret) {
+		dev_err(&hwdev->hwif->pdev->dev, "Set VF random id failed, err: %d, status: 0x%x, out size: 0x%x\n",
+			ret, rand_info.status, out_size);
+		return -EIO;
+	}
+
+	if (rand_info.status == HINIC_MGMT_CMD_UNSUPPORTED)
+		return rand_info.status;
+
+	func_to_func->vf_mbx_old_rand_id[func_id] =
+				func_to_func->vf_mbx_rand_id[func_id];
+
+	return 0;
+}
+
+static void update_random_id_work_handler(struct work_struct *work)
+{
+	struct hinic_mbox_work *mbox_work =
+			container_of(work, struct hinic_mbox_work, work);
+	struct hinic_mbox_func_to_func *func_to_func;
+	u16 src = mbox_work->src_func_idx;
+
+	func_to_func = mbox_work->func_to_func;
+
+	if (set_vf_mbox_random_id(func_to_func->hwdev, src))
+		dev_warn(&func_to_func->hwdev->hwif->pdev->dev, "Update VF id: 0x%x random id failed\n",
+			 mbox_work->src_func_idx);
+
+	kfree(mbox_work);
+}
+
+static bool check_vf_mbox_random_id(struct hinic_mbox_func_to_func *func_to_func,
+				    u8 *header)
+{
+	struct hinic_hwdev *hwdev = func_to_func->hwdev;
+	struct hinic_mbox_work *mbox_work = NULL;
+	u64 mbox_header = *((u64 *)header);
+	u16 offset, src;
+	u32 random_id;
+	int vf_in_pf;
+
+	src = HINIC_MBOX_HEADER_GET(mbox_header, SRC_GLB_FUNC_IDX);
+
+	if (IS_PF_OR_PPF_SRC(src) || !func_to_func->support_vf_random)
+		return true;
+
+	if (!HINIC_IS_PPF(hwdev->hwif)) {
+		offset = hinic_glb_pf_vf_offset(hwdev->hwif);
+		vf_in_pf = src - offset;
+
+		if (vf_in_pf < 1 || vf_in_pf > hwdev->nic_cap.max_vf) {
+			dev_warn(&hwdev->hwif->pdev->dev,
+				 "Receive vf id(0x%x) is invalid, vf id should be from 0x%x to 0x%x\n",
+				 src, offset + 1,
+				 hwdev->nic_cap.max_vf + offset);
+			return false;
+		}
+	}
+
+	random_id = be32_to_cpu(*(u32 *)(header + MBOX_SEG_LEN +
+					 MBOX_HEADER_SZ));
+
+	if (random_id == func_to_func->vf_mbx_rand_id[src] ||
+	    random_id == func_to_func->vf_mbx_old_rand_id[src])
+		return true;
+
+	dev_warn(&hwdev->hwif->pdev->dev,
+		 "The mailbox random id(0x%x) of func_id(0x%x) doesn't match with pf reservation(0x%x)\n",
+		 random_id, src, func_to_func->vf_mbx_rand_id[src]);
+
+	mbox_work = kzalloc(sizeof(*mbox_work), GFP_KERNEL);
+	if (!mbox_work)
+		return false;
+
+	mbox_work->func_to_func = func_to_func;
+	mbox_work->src_func_idx = src;
+
+	INIT_WORK(&mbox_work->work, update_random_id_work_handler);
+	queue_work(func_to_func->workq, &mbox_work->work);
+
+	return false;
+}
+
+static void hinic_mbox_func_aeqe_handler(void *handle, void *header, u8 size)
+{
+	struct hinic_mbox_func_to_func *func_to_func;
+	u64 mbox_header = *((u64 *)header);
+	struct hinic_recv_mbox *recv_mbox;
+	u64 src, dir;
+
+	func_to_func = ((struct hinic_hwdev *)handle)->func_to_func;
+
+	dir = HINIC_MBOX_HEADER_GET(mbox_header, DIRECTION);
+	src = HINIC_MBOX_HEADER_GET(mbox_header, SRC_GLB_FUNC_IDX);
+
+	if (src >= HINIC_MAX_FUNCTIONS) {
+		dev_err(&func_to_func->hwif->pdev->dev,
+			"Mailbox source function id:%u is invalid\n", (u32)src);
+		return;
+	}
+
+	if (!check_vf_mbox_random_id(func_to_func, header))
+		return;
+
+	recv_mbox = (dir == HINIC_HWIF_DIRECT_SEND) ?
+		    &func_to_func->mbox_send[src] :
+		    &func_to_func->mbox_resp[src];
+
+	recv_mbox_handler(func_to_func, (u64 *)header, recv_mbox);
+}
+
+static void hinic_mbox_self_aeqe_handler(void *handle, void *header, u8 size)
+{
+	struct hinic_mbox_func_to_func *func_to_func;
+	struct hinic_send_mbox *send_mbox;
+
+	func_to_func = ((struct hinic_hwdev *)handle)->func_to_func;
+	send_mbox = &func_to_func->send_mbox;
+
+	complete(&send_mbox->send_done);
+}
+
+static void clear_mbox_status(struct hinic_send_mbox *mbox)
+{
+	*mbox->wb_status = 0;
+
+	/* clear mailbox write back status */
+	wmb();
+}
+
+static void mbox_copy_header(struct hinic_hwdev *hwdev,
+			     struct hinic_send_mbox *mbox, u64 *header)
+{
+	u32 i, idx_max = MBOX_HEADER_SZ / sizeof(u32);
+	u32 *data = (u32 *)header;
+
+	for (i = 0; i < idx_max; i++)
+		__raw_writel(*(data + i), mbox->data + i * sizeof(u32));
+}
+
+static void mbox_copy_send_data(struct hinic_hwdev *hwdev,
+				struct hinic_send_mbox *mbox, void *seg,
+				u16 seg_len)
+{
+	u8 mbox_max_buf[MBOX_SEG_LEN] = {0};
+	u32 data_len, chk_sz = sizeof(u32);
+	u32 *data = seg;
+	u32 i, idx_max;
+
+	/* The mbox message should be aligned in 4 bytes. */
+	if (seg_len % chk_sz) {
+		memcpy(mbox_max_buf, seg, seg_len);
+		data = (u32 *)mbox_max_buf;
+	}
+
+	data_len = seg_len;
+	idx_max = ALIGN(data_len, chk_sz) / chk_sz;
+
+	for (i = 0; i < idx_max; i++)
+		__raw_writel(*(data + i),
+			     mbox->data + MBOX_HEADER_SZ + i * sizeof(u32));
+}
+
+static void write_mbox_msg_attr(struct hinic_mbox_func_to_func *func_to_func,
+				u16 dst_func, u16 dst_aeqn, u16 seg_len,
+				int poll)
+{
+	u16 rsp_aeq = (dst_aeqn == 0) ? 0 : HINIC_MBOX_RSP_AEQN;
+	u32 mbox_int, mbox_ctrl;
+
+	mbox_int = HINIC_MBOX_INT_SET(dst_func, DST_FUNC) |
+		   HINIC_MBOX_INT_SET(dst_aeqn, DST_AEQN) |
+		   HINIC_MBOX_INT_SET(rsp_aeq, SRC_RESP_AEQN) |
+		   HINIC_MBOX_INT_SET(NO_DMA_ATTRIBUTE_VAL, STAT_DMA) |
+		   HINIC_MBOX_INT_SET(ALIGN(MBOX_SEG_LEN + MBOX_HEADER_SZ +
+				      MBOX_INFO_SZ, MBOX_SEG_LEN_ALIGN) >> 2,
+				      TX_SIZE) |
+		   HINIC_MBOX_INT_SET(STRONG_ORDER, STAT_DMA_SO_RO) |
+		   HINIC_MBOX_INT_SET(WRITE_BACK, WB_EN);
+
+	hinic_hwif_write_reg(func_to_func->hwif,
+			     HINIC_FUNC_CSR_MAILBOX_INT_OFFSET_OFF, mbox_int);
+
+	wmb(); /* writing the mbox int attributes */
+	mbox_ctrl = HINIC_MBOX_CTRL_SET(TX_NOT_DONE, TX_STATUS);
+
+	if (poll)
+		mbox_ctrl |= HINIC_MBOX_CTRL_SET(NOT_TRIGGER, TRIGGER_AEQE);
+	else
+		mbox_ctrl |= HINIC_MBOX_CTRL_SET(TRIGGER, TRIGGER_AEQE);
+
+	hinic_hwif_write_reg(func_to_func->hwif,
+			     HINIC_FUNC_CSR_MAILBOX_CONTROL_OFF, mbox_ctrl);
+}
+
+static void dump_mox_reg(struct hinic_hwdev *hwdev)
+{
+	u32 val;
+
+	val = hinic_hwif_read_reg(hwdev->hwif,
+				  HINIC_FUNC_CSR_MAILBOX_CONTROL_OFF);
+	dev_err(&hwdev->hwif->pdev->dev, "Mailbox control reg: 0x%x\n", val);
+
+	val = hinic_hwif_read_reg(hwdev->hwif,
+				  HINIC_FUNC_CSR_MAILBOX_INT_OFFSET_OFF);
+	dev_err(&hwdev->hwif->pdev->dev, "Mailbox interrupt offset: 0x%x\n",
+		val);
+}
+
+static u16 get_mbox_status(struct hinic_send_mbox *mbox)
+{
+	/* write back is 16B, but only use first 4B */
+	u64 wb_val = be64_to_cpu(*mbox->wb_status);
+
+	rmb(); /* verify reading before check */
+
+	return (u16)(wb_val & MBOX_WB_STATUS_ERRCODE_MASK);
+}
+
+static int
+wait_for_mbox_seg_completion(struct hinic_mbox_func_to_func *func_to_func,
+			     int poll, u16 *wb_status)
+{
+	struct hinic_send_mbox *send_mbox = &func_to_func->send_mbox;
+	struct hinic_hwdev *hwdev = func_to_func->hwdev;
+	struct completion *done = &send_mbox->send_done;
+	u32 cnt = 0;
+	unsigned long jif;
+
+	if (poll) {
+		while (cnt < MBOX_MSG_POLLING_TIMEOUT) {
+			*wb_status = get_mbox_status(send_mbox);
+			if (MBOX_STATUS_FINISHED(*wb_status))
+				break;
+
+			usleep_range(900, 1000);
+			cnt++;
+		}
+
+		if (cnt == MBOX_MSG_POLLING_TIMEOUT) {
+			dev_err(&hwdev->hwif->pdev->dev, "Send mailbox segment timeout, wb status: 0x%x\n",
+				*wb_status);
+			dump_mox_reg(hwdev);
+			return -ETIMEDOUT;
+		}
+	} else {
+		jif = msecs_to_jiffies(HINIC_MBOX_COMP_TIME);
+		if (!wait_for_completion_timeout(done, jif)) {
+			dev_err(&hwdev->hwif->pdev->dev, "Send mailbox segment timeout\n");
+			dump_mox_reg(hwdev);
+			hinic_dump_aeq_info(hwdev);
+			return -ETIMEDOUT;
+		}
+
+		*wb_status = get_mbox_status(send_mbox);
+	}
+
+	return 0;
+}
+
+static int send_mbox_seg(struct hinic_mbox_func_to_func *func_to_func,
+			 u64 header, u16 dst_func, void *seg, u16 seg_len,
+			 int poll, void *msg_info)
+{
+	struct hinic_send_mbox *send_mbox = &func_to_func->send_mbox;
+	u16 seq_dir = HINIC_MBOX_HEADER_GET(header, DIRECTION);
+	struct hinic_hwdev *hwdev = func_to_func->hwdev;
+	struct completion *done = &send_mbox->send_done;
+	u8 num_aeqs = hwdev->hwif->attr.num_aeqs;
+	u16 dst_aeqn, wb_status = 0, errcode;
+
+	if (num_aeqs >= 4)
+		dst_aeqn = (seq_dir == HINIC_HWIF_DIRECT_SEND) ?
+			   HINIC_MBOX_RECV_AEQN : HINIC_MBOX_RSP_AEQN;
+	else
+		dst_aeqn = 0;
+
+	if (!poll)
+		init_completion(done);
+
+	clear_mbox_status(send_mbox);
+
+	mbox_copy_header(hwdev, send_mbox, &header);
+
+	mbox_copy_send_data(hwdev, send_mbox, seg, seg_len);
+
+	write_mbox_msg_attr(func_to_func, dst_func, dst_aeqn, seg_len, poll);
+
+	wmb(); /* writing the mbox msg attributes */
+
+	if (wait_for_mbox_seg_completion(func_to_func, poll, &wb_status))
+		return -ETIMEDOUT;
+
+	if (!MBOX_STATUS_SUCCESS(wb_status)) {
+		dev_err(&hwdev->hwif->pdev->dev, "Send mailbox segment to function %d error, wb status: 0x%x\n",
+			dst_func, wb_status);
+		errcode = MBOX_STATUS_ERRCODE(wb_status);
+		return errcode ? errcode : -EFAULT;
+	}
+
+	return 0;
+}
+
+static int send_mbox_to_func(struct hinic_mbox_func_to_func *func_to_func,
+			     enum hinic_mod_type mod, u16 cmd, void *msg,
+			     u16 msg_len, u16 dst_func,
+			     enum hinic_hwif_direction_type direction,
+			     enum hinic_mbox_ack_type ack_type,
+			     struct mbox_msg_info *msg_info)
+{
+	struct hinic_hwdev *hwdev = func_to_func->hwdev;
+	u16 seg_len = MBOX_SEG_LEN;
+	u8 *msg_seg = (u8 *)msg;
+	u16 left = msg_len;
+	u32 seq_id = 0;
+	u64 header = 0;
+	int err = 0;
+
+	down(&func_to_func->msg_send_sem);
+
+	header = HINIC_MBOX_HEADER_SET(msg_len, MSG_LEN) |
+		 HINIC_MBOX_HEADER_SET(mod, MODULE) |
+		 HINIC_MBOX_HEADER_SET(seg_len, SEG_LEN) |
+		 HINIC_MBOX_HEADER_SET(ack_type, NO_ACK) |
+		 HINIC_MBOX_HEADER_SET(SEQ_ID_START_VAL, SEQID) |
+		 HINIC_MBOX_HEADER_SET(NOT_LAST_SEG, LAST) |
+		 HINIC_MBOX_HEADER_SET(direction, DIRECTION) |
+		 HINIC_MBOX_HEADER_SET(cmd, CMD) |
+		 /* The vf's offset to its associated pf */
+		 HINIC_MBOX_HEADER_SET(msg_info->msg_id, MSG_ID) |
+		 HINIC_MBOX_HEADER_SET(msg_info->status, STATUS) |
+		 HINIC_MBOX_HEADER_SET(hinic_global_func_id_hw(hwdev->hwif),
+				       SRC_GLB_FUNC_IDX);
+
+	while (!(HINIC_MBOX_HEADER_GET(header, LAST))) {
+		if (left <= HINIC_MBOX_SEG_LEN) {
+			header &= ~MBOX_SEGLEN_MASK;
+			header |= HINIC_MBOX_HEADER_SET(left, SEG_LEN);
+			header |= HINIC_MBOX_HEADER_SET(LAST_SEG, LAST);
+
+			seg_len = left;
+		}
+
+		err = send_mbox_seg(func_to_func, header, dst_func, msg_seg,
+				    seg_len, MBOX_SEND_MSG_INT, msg_info);
+		if (err) {
+			dev_err(&hwdev->hwif->pdev->dev, "Failed to send mbox seg, seq_id=0x%llx\n",
+				HINIC_MBOX_HEADER_GET(header, SEQID));
+			goto err_send_mbox_seg;
+		}
+
+		left -= HINIC_MBOX_SEG_LEN;
+		msg_seg += HINIC_MBOX_SEG_LEN;
+
+		seq_id++;
+		header &= ~(HINIC_MBOX_HEADER_SET(HINIC_MBOX_HEADER_SEQID_MASK,
+						  SEQID));
+		header |= HINIC_MBOX_HEADER_SET(seq_id, SEQID);
+	}
+
+err_send_mbox_seg:
+	up(&func_to_func->msg_send_sem);
+
+	return err;
+}
+
+static void
+response_for_recv_func_mbox(struct hinic_mbox_func_to_func *func_to_func,
+			    struct hinic_recv_mbox *recv_mbox, int err,
+			    u16 out_size, u16 src_func_idx)
+{
+	struct mbox_msg_info msg_info = {0};
+
+	if (recv_mbox->ack_type == MBOX_ACK) {
+		msg_info.msg_id = recv_mbox->msg_info.msg_id;
+		if (err == HINIC_MBOX_PF_BUSY_ACTIVE_FW)
+			msg_info.status = HINIC_MBOX_PF_BUSY_ACTIVE_FW;
+		else if (err == HINIC_MBOX_VF_CMD_ERROR)
+			msg_info.status = HINIC_MBOX_VF_CMD_ERROR;
+		else if (err)
+			msg_info.status = HINIC_MBOX_PF_SEND_ERR;
+
+		/* if no data needs to response, set out_size to 1 */
+		if (!out_size || err)
+			out_size = MBOX_MSG_NO_DATA_LEN;
+
+		send_mbox_to_func(func_to_func, recv_mbox->mod, recv_mbox->cmd,
+				  recv_mbox->buf_out, out_size, src_func_idx,
+				  HINIC_HWIF_RESPONSE, MBOX_ACK,
+				  &msg_info);
+	}
+}
+
+static void recv_func_mbox_handler(struct hinic_mbox_func_to_func *func_to_func,
+				   struct hinic_recv_mbox *recv_mbox,
+				   u16 src_func_idx)
+{
+	void *buf_out = recv_mbox->buf_out;
+	u16 out_size = MBOX_MAX_BUF_SZ;
+	int err = 0;
+
+	if (HINIC_IS_VF(func_to_func->hwif)) {
+		err = recv_vf_mbox_handler(func_to_func, recv_mbox, buf_out,
+					   &out_size);
+	} else {
+		if (IS_PF_OR_PPF_SRC(src_func_idx))
+			dev_warn(&func_to_func->hwif->pdev->dev,
+				 "Unsupported pf2pf mbox msg\n");
+		else
+			err = recv_pf_from_vf_mbox_handler(func_to_func,
+							   recv_mbox,
+							   src_func_idx,
+							   buf_out, &out_size);
+	}
+
+	response_for_recv_func_mbox(func_to_func, recv_mbox, err, out_size,
+				    src_func_idx);
+	kfree(recv_mbox->buf_out);
+	kfree(recv_mbox->mbox);
+	kfree(recv_mbox);
+}
+
+static void set_mbox_to_func_event(struct hinic_mbox_func_to_func *func_to_func,
+				   enum mbox_event_state event_flag)
+{
+	spin_lock(&func_to_func->mbox_lock);
+	func_to_func->event_flag = event_flag;
+	spin_unlock(&func_to_func->mbox_lock);
+}
+
+static int mbox_resp_info_handler(struct hinic_mbox_func_to_func *func_to_func,
+				  struct hinic_recv_mbox *mbox_for_resp,
+				  enum hinic_mod_type mod, u16 cmd,
+				  void *buf_out, u16 *out_size)
+{
+	int err;
+
+	if (mbox_for_resp->msg_info.status) {
+		err = mbox_for_resp->msg_info.status;
+		if (err != HINIC_MBOX_PF_BUSY_ACTIVE_FW)
+			dev_err(&func_to_func->hwif->pdev->dev, "Mbox response error(0x%x)\n",
+				mbox_for_resp->msg_info.status);
+		return err;
+	}
+
+	if (buf_out && out_size) {
+		if (*out_size < mbox_for_resp->mbox_len) {
+			dev_err(&func_to_func->hwif->pdev->dev,
+				"Invalid response mbox message length: %d for mod %d cmd %d, should less than: %d\n",
+				mbox_for_resp->mbox_len, mod, cmd, *out_size);
+			return -EFAULT;
+		}
+
+		if (mbox_for_resp->mbox_len)
+			memcpy(buf_out, mbox_for_resp->mbox,
+			       mbox_for_resp->mbox_len);
+
+		*out_size = mbox_for_resp->mbox_len;
+	}
+
+	return 0;
+}
+
+int hinic_mbox_to_func(struct hinic_mbox_func_to_func *func_to_func,
+		       enum hinic_mod_type mod, u16 cmd, u16 dst_func,
+		       void *buf_in, u16 in_size, void *buf_out,
+		       u16 *out_size, u32 timeout)
+{
+	struct hinic_recv_mbox *mbox_for_resp;
+	struct mbox_msg_info msg_info = {0};
+	unsigned long timeo;
+	int err;
+
+	mbox_for_resp = &func_to_func->mbox_resp[dst_func];
+
+	down(&func_to_func->mbox_send_sem);
+
+	init_completion(&mbox_for_resp->recv_done);
+
+	msg_info.msg_id = MBOX_MSG_ID_INC(func_to_func);
+
+	set_mbox_to_func_event(func_to_func, EVENT_START);
+
+	err = send_mbox_to_func(func_to_func, mod, cmd, buf_in, in_size,
+				dst_func, HINIC_HWIF_DIRECT_SEND, MBOX_ACK,
+				&msg_info);
+	if (err) {
+		dev_err(&func_to_func->hwif->pdev->dev, "Send mailbox failed, msg_id: %d\n",
+			msg_info.msg_id);
+		set_mbox_to_func_event(func_to_func, EVENT_FAIL);
+		goto err_send_mbox;
+	}
+
+	timeo = msecs_to_jiffies(timeout ? timeout : HINIC_MBOX_COMP_TIME);
+	if (!wait_for_completion_timeout(&mbox_for_resp->recv_done, timeo)) {
+		set_mbox_to_func_event(func_to_func, EVENT_TIMEOUT);
+		dev_err(&func_to_func->hwif->pdev->dev,
+			"Send mbox msg timeout, msg_id: %d\n", msg_info.msg_id);
+		hinic_dump_aeq_info(func_to_func->hwdev);
+		err = -ETIMEDOUT;
+		goto err_send_mbox;
+	}
+
+	set_mbox_to_func_event(func_to_func, EVENT_END);
+
+	err = mbox_resp_info_handler(func_to_func, mbox_for_resp, mod, cmd,
+				     buf_out, out_size);
+
+err_send_mbox:
+	up(&func_to_func->mbox_send_sem);
+
+	return err;
+}
+
+static int mbox_func_params_valid(struct hinic_mbox_func_to_func *func_to_func,
+				  void *buf_in, u16 in_size)
+{
+	if (in_size > HINIC_MBOX_DATA_SIZE) {
+		dev_err(&func_to_func->hwif->pdev->dev,
+			"Mbox msg len(%d) exceed limit(%d)\n",
+			in_size, HINIC_MBOX_DATA_SIZE);
+		return -EINVAL;
+	}
+
+	return 0;
+}
+
+int hinic_mbox_to_pf(struct hinic_hwdev *hwdev,
+		     enum hinic_mod_type mod, u8 cmd, void *buf_in,
+		     u16 in_size, void *buf_out, u16 *out_size, u32 timeout)
+{
+	struct hinic_mbox_func_to_func *func_to_func = hwdev->func_to_func;
+	int err = mbox_func_params_valid(func_to_func, buf_in, in_size);
+
+	if (err)
+		return err;
+
+	if (!HINIC_IS_VF(hwdev->hwif)) {
+		dev_err(&hwdev->hwif->pdev->dev, "Params error, func_type: %d\n",
+			HINIC_FUNC_TYPE(hwdev->hwif));
+		return -EINVAL;
+	}
+
+	return hinic_mbox_to_func(func_to_func, mod, cmd,
+				  hinic_pf_id_of_vf_hw(hwdev->hwif), buf_in,
+				  in_size, buf_out, out_size, timeout);
+}
+
+int hinic_mbox_to_vf(struct hinic_hwdev *hwdev,
+		     enum hinic_mod_type mod, u16 vf_id, u8 cmd, void *buf_in,
+		     u16 in_size, void *buf_out, u16 *out_size, u32 timeout)
+{
+	struct hinic_mbox_func_to_func *func_to_func;
+	u16 dst_func_idx;
+	int err;
+
+	if (!hwdev)
+		return -EINVAL;
+
+	func_to_func = hwdev->func_to_func;
+	err = mbox_func_params_valid(func_to_func, buf_in, in_size);
+	if (err)
+		return err;
+
+	if (HINIC_IS_VF(hwdev->hwif)) {
+		dev_err(&hwdev->hwif->pdev->dev, "Params error, func_type: %d\n",
+			HINIC_FUNC_TYPE(hwdev->hwif));
+		return -EINVAL;
+	}
+
+	if (!vf_id) {
+		dev_err(&hwdev->hwif->pdev->dev,
+			"VF id(%d) error!\n", vf_id);
+		return -EINVAL;
+	}
+
+	/* vf_offset_to_pf + vf_id is the vf's global function id of vf in
+	 * this pf
+	 */
+	dst_func_idx = hinic_glb_pf_vf_offset(hwdev->hwif) + vf_id;
+
+	return hinic_mbox_to_func(func_to_func, mod, cmd, dst_func_idx, buf_in,
+				  in_size, buf_out, out_size, timeout);
+}
+
+static int init_mbox_info(struct hinic_recv_mbox *mbox_info)
+{
+	int err;
+
+	mbox_info->seq_id = SEQ_ID_MAX_VAL;
+
+	mbox_info->mbox = kzalloc(MBOX_MAX_BUF_SZ, GFP_KERNEL);
+	if (!mbox_info->mbox)
+		return -ENOMEM;
+
+	mbox_info->buf_out = kzalloc(MBOX_MAX_BUF_SZ, GFP_KERNEL);
+	if (!mbox_info->buf_out) {
+		err = -ENOMEM;
+		goto err_alloc_buf_out;
+	}
+
+	atomic_set(&mbox_info->msg_cnt, 0);
+
+	return 0;
+
+err_alloc_buf_out:
+	kfree(mbox_info->mbox);
+
+	return err;
+}
+
+static void clean_mbox_info(struct hinic_recv_mbox *mbox_info)
+{
+	kfree(mbox_info->buf_out);
+	kfree(mbox_info->mbox);
+}
+
+static int alloc_mbox_info(struct hinic_hwdev *hwdev,
+			   struct hinic_recv_mbox *mbox_info)
+{
+	u16 func_idx, i;
+	int err;
+
+	for (func_idx = 0; func_idx < HINIC_MAX_FUNCTIONS; func_idx++) {
+		err = init_mbox_info(&mbox_info[func_idx]);
+		if (err) {
+			dev_err(&hwdev->hwif->pdev->dev, "Failed to init function %d mbox info\n",
+				func_idx);
+			goto err_init_mbox_info;
+		}
+	}
+
+	return 0;
+
+err_init_mbox_info:
+	for (i = 0; i < func_idx; i++)
+		clean_mbox_info(&mbox_info[i]);
+
+	return err;
+}
+
+static void free_mbox_info(struct hinic_recv_mbox *mbox_info)
+{
+	u16 func_idx;
+
+	for (func_idx = 0; func_idx < HINIC_MAX_FUNCTIONS; func_idx++)
+		clean_mbox_info(&mbox_info[func_idx]);
+}
+
+static void prepare_send_mbox(struct hinic_mbox_func_to_func *func_to_func)
+{
+	struct hinic_send_mbox *send_mbox = &func_to_func->send_mbox;
+
+	send_mbox->data = MBOX_AREA(func_to_func->hwif);
+}
+
+static int alloc_mbox_wb_status(struct hinic_mbox_func_to_func *func_to_func)
+{
+	struct hinic_send_mbox *send_mbox = &func_to_func->send_mbox;
+	struct hinic_hwdev *hwdev = func_to_func->hwdev;
+	u32 addr_h, addr_l;
+
+	send_mbox->wb_vaddr = dma_alloc_coherent(&hwdev->hwif->pdev->dev,
+						 MBOX_WB_STATUS_LEN,
+						 &send_mbox->wb_paddr,
+						 GFP_KERNEL);
+	if (!send_mbox->wb_vaddr)
+		return -ENOMEM;
+
+	send_mbox->wb_status = send_mbox->wb_vaddr;
+
+	addr_h = upper_32_bits(send_mbox->wb_paddr);
+	addr_l = lower_32_bits(send_mbox->wb_paddr);
+
+	hinic_hwif_write_reg(hwdev->hwif, HINIC_FUNC_CSR_MAILBOX_RESULT_H_OFF,
+			     addr_h);
+	hinic_hwif_write_reg(hwdev->hwif, HINIC_FUNC_CSR_MAILBOX_RESULT_L_OFF,
+			     addr_l);
+
+	return 0;
+}
+
+static void free_mbox_wb_status(struct hinic_mbox_func_to_func *func_to_func)
+{
+	struct hinic_send_mbox *send_mbox = &func_to_func->send_mbox;
+	struct hinic_hwdev *hwdev = func_to_func->hwdev;
+
+	hinic_hwif_write_reg(hwdev->hwif, HINIC_FUNC_CSR_MAILBOX_RESULT_H_OFF,
+			     0);
+	hinic_hwif_write_reg(hwdev->hwif, HINIC_FUNC_CSR_MAILBOX_RESULT_L_OFF,
+			     0);
+
+	dma_free_coherent(&hwdev->hwif->pdev->dev, MBOX_WB_STATUS_LEN,
+			  send_mbox->wb_vaddr,
+			  send_mbox->wb_paddr);
+}
+
+bool hinic_mbox_check_cmd_valid(struct hinic_hwdev *hwdev,
+				struct vf_cmd_check_handle *cmd_handle,
+				u16 vf_id, u8 cmd, void *buf_in,
+				u16 in_size, u8 size)
+{
+	u16 src_idx = vf_id + hinic_glb_pf_vf_offset(hwdev->hwif);
+	int i;
+
+	for (i = 0; i < size; i++) {
+		if (cmd == cmd_handle[i].cmd) {
+			if (cmd_handle[i].check_cmd)
+				return cmd_handle[i].check_cmd(hwdev, src_idx,
+							       buf_in, in_size);
+			else
+				return true;
+		}
+	}
+
+	dev_err(&hwdev->hwif->pdev->dev,
+		"PF Receive VF(%d) unsupported cmd(0x%x)\n",
+		vf_id + hinic_glb_pf_vf_offset(hwdev->hwif), cmd);
+
+	return false;
+}
+
+static bool hinic_cmdq_check_vf_ctxt(struct hinic_hwdev *hwdev,
+				     struct hinic_cmdq_ctxt *cmdq_ctxt)
+{
+	struct hinic_cmdq_ctxt_info *ctxt_info = &cmdq_ctxt->ctxt_info;
+	u64 curr_pg_pfn, wq_block_pfn;
+
+	if (cmdq_ctxt->ppf_idx != HINIC_HWIF_PPF_IDX(hwdev->hwif) ||
+	    cmdq_ctxt->cmdq_type > HINIC_MAX_CMDQ_TYPES)
+		return false;
+
+	curr_pg_pfn = HINIC_CMDQ_CTXT_PAGE_INFO_GET
+		(ctxt_info->curr_wqe_page_pfn, CURR_WQE_PAGE_PFN);
+	wq_block_pfn = HINIC_CMDQ_CTXT_BLOCK_INFO_GET
+		(ctxt_info->wq_block_pfn, WQ_BLOCK_PFN);
+	/* VF must use 0-level CLA */
+	if (curr_pg_pfn != wq_block_pfn)
+		return false;
+
+	return true;
+}
+
+static bool check_cmdq_ctxt(struct hinic_hwdev *hwdev, u16 func_idx,
+			    void *buf_in, u16 in_size)
+{
+	if (!hinic_mbox_check_func_id_8B(hwdev, func_idx, buf_in, in_size))
+		return false;
+
+	return hinic_cmdq_check_vf_ctxt(hwdev, buf_in);
+}
+
+#define HW_CTX_QPS_VALID(hw_ctxt)   \
+		((hw_ctxt)->rq_depth >= HINIC_QUEUE_MIN_DEPTH &&	\
+		(hw_ctxt)->rq_depth <= HINIC_QUEUE_MAX_DEPTH &&	\
+		(hw_ctxt)->sq_depth >= HINIC_QUEUE_MIN_DEPTH &&	\
+		(hw_ctxt)->sq_depth <= HINIC_QUEUE_MAX_DEPTH &&	\
+		(hw_ctxt)->rx_buf_sz_idx <= HINIC_MAX_RX_BUFFER_SIZE)
+
+static bool hw_ctxt_qps_param_valid(struct hinic_cmd_hw_ioctxt *hw_ctxt)
+{
+	if (HW_CTX_QPS_VALID(hw_ctxt))
+		return true;
+
+	if (!hw_ctxt->rq_depth && !hw_ctxt->sq_depth &&
+	    !hw_ctxt->rx_buf_sz_idx)
+		return true;
+
+	return false;
+}
+
+static bool check_hwctxt(struct hinic_hwdev *hwdev, u16 func_idx,
+			 void *buf_in, u16 in_size)
+{
+	struct hinic_cmd_hw_ioctxt *hw_ctxt = buf_in;
+
+	if (!hinic_mbox_check_func_id_8B(hwdev, func_idx, buf_in, in_size))
+		return false;
+
+	if (hw_ctxt->ppf_idx != HINIC_HWIF_PPF_IDX(hwdev->hwif))
+		return false;
+
+	if (hw_ctxt->set_cmdq_depth) {
+		if (hw_ctxt->cmdq_depth >= HINIC_QUEUE_MIN_DEPTH &&
+		    hw_ctxt->cmdq_depth <= HINIC_QUEUE_MAX_DEPTH)
+			return true;
+
+		return false;
+	}
+
+	return hw_ctxt_qps_param_valid(hw_ctxt);
+}
+
+static bool check_set_wq_page_size(struct hinic_hwdev *hwdev, u16 func_idx,
+				   void *buf_in, u16 in_size)
+{
+	struct hinic_wq_page_size *page_size_info = buf_in;
+
+	if (!hinic_mbox_check_func_id_8B(hwdev, func_idx, buf_in, in_size))
+		return false;
+
+	if (page_size_info->ppf_idx != HINIC_HWIF_PPF_IDX(hwdev->hwif))
+		return false;
+
+	if (((1U << page_size_info->page_size) * SZ_4K) !=
+	    HINIC_DEFAULT_WQ_PAGE_SIZE)
+		return false;
+
+	return true;
+}
+
+static struct vf_cmd_check_handle hw_cmd_support_vf[] = {
+	{HINIC_COMM_CMD_START_FLR, hinic_mbox_check_func_id_8B},
+	{HINIC_COMM_CMD_DMA_ATTR_SET, hinic_mbox_check_func_id_8B},
+	{HINIC_COMM_CMD_CMDQ_CTXT_SET, check_cmdq_ctxt},
+	{HINIC_COMM_CMD_CMDQ_CTXT_GET, check_cmdq_ctxt},
+	{HINIC_COMM_CMD_HWCTXT_SET, check_hwctxt},
+	{HINIC_COMM_CMD_HWCTXT_GET, check_hwctxt},
+	{HINIC_COMM_CMD_SQ_HI_CI_SET, hinic_mbox_check_func_id_8B},
+	{HINIC_COMM_CMD_RES_STATE_SET, hinic_mbox_check_func_id_8B},
+	{HINIC_COMM_CMD_IO_RES_CLEAR, hinic_mbox_check_func_id_8B},
+	{HINIC_COMM_CMD_CEQ_CTRL_REG_WR_BY_UP, hinic_mbox_check_func_id_8B},
+	{HINIC_COMM_CMD_MSI_CTRL_REG_WR_BY_UP, hinic_mbox_check_func_id_8B},
+	{HINIC_COMM_CMD_MSI_CTRL_REG_RD_BY_UP, hinic_mbox_check_func_id_8B},
+	{HINIC_COMM_CMD_L2NIC_RESET, hinic_mbox_check_func_id_8B},
+	{HINIC_COMM_CMD_PAGESIZE_SET, check_set_wq_page_size},
+};
+
+static int comm_pf_mbox_handler(void *handle, u16 vf_id, u8 cmd, void *buf_in,
+				u16 in_size, void *buf_out, u16 *out_size)
+{
+	u8 size = ARRAY_SIZE(hw_cmd_support_vf);
+	struct hinic_hwdev *hwdev = handle;
+	struct hinic_pfhwdev *pfhwdev;
+	int err = 0;
+
+	pfhwdev = container_of(hwdev, struct hinic_pfhwdev, hwdev);
+
+	if (!hinic_mbox_check_cmd_valid(handle, hw_cmd_support_vf, vf_id, cmd,
+					buf_in, in_size, size)) {
+		dev_err(&hwdev->hwif->pdev->dev,
+			"PF Receive VF: %d common cmd: 0x%x or mbox len: 0x%x is invalid\n",
+			vf_id + hinic_glb_pf_vf_offset(hwdev->hwif), cmd,
+			in_size);
+		return HINIC_MBOX_VF_CMD_ERROR;
+	}
+
+	if (cmd == HINIC_COMM_CMD_START_FLR) {
+		*out_size = 0;
+	} else {
+		err = hinic_msg_to_mgmt(&pfhwdev->pf_to_mgmt, HINIC_MOD_COMM,
+					cmd, buf_in, in_size, buf_out, out_size,
+					HINIC_MGMT_MSG_SYNC);
+		if (err && err != HINIC_MBOX_PF_BUSY_ACTIVE_FW)
+			dev_err(&hwdev->hwif->pdev->dev,
+				"PF mbox common callback handler err: %d\n",
+				err);
+	}
+
+	return err;
+}
+
+int hinic_func_to_func_init(struct hinic_hwdev *hwdev)
+{
+	struct hinic_mbox_func_to_func *func_to_func;
+	struct hinic_pfhwdev *pfhwdev;
+	int err;
+
+	pfhwdev =  container_of(hwdev, struct hinic_pfhwdev, hwdev);
+	func_to_func = kzalloc(sizeof(*func_to_func), GFP_KERNEL);
+	if (!func_to_func)
+		return -ENOMEM;
+
+	hwdev->func_to_func = func_to_func;
+	func_to_func->hwdev = hwdev;
+	func_to_func->hwif = hwdev->hwif;
+	sema_init(&func_to_func->mbox_send_sem, 1);
+	sema_init(&func_to_func->msg_send_sem, 1);
+	spin_lock_init(&func_to_func->mbox_lock);
+	func_to_func->workq = create_singlethread_workqueue(HINIC_MBOX_WQ_NAME);
+	if (!func_to_func->workq) {
+		dev_err(&hwdev->hwif->pdev->dev, "Failed to initialize MBOX workqueue\n");
+		err = -ENOMEM;
+		goto err_create_mbox_workq;
+	}
+
+	err = alloc_mbox_info(hwdev, func_to_func->mbox_send);
+	if (err) {
+		dev_err(&hwdev->hwif->pdev->dev, "Failed to alloc mem for mbox_active\n");
+		goto err_alloc_mbox_for_send;
+	}
+
+	err = alloc_mbox_info(hwdev, func_to_func->mbox_resp);
+	if (err) {
+		dev_err(&hwdev->hwif->pdev->dev, "Failed to alloc mem for mbox_passive\n");
+		goto err_alloc_mbox_for_resp;
+	}
+
+	err = alloc_mbox_wb_status(func_to_func);
+	if (err) {
+		dev_err(&hwdev->hwif->pdev->dev, "Failed to alloc mbox write back status\n");
+		goto err_alloc_wb_status;
+	}
+
+	prepare_send_mbox(func_to_func);
+
+	hinic_aeq_register_hw_cb(&hwdev->aeqs, HINIC_MBX_FROM_FUNC,
+				 &pfhwdev->hwdev, hinic_mbox_func_aeqe_handler);
+	hinic_aeq_register_hw_cb(&hwdev->aeqs, HINIC_MBX_SEND_RSLT,
+				 &pfhwdev->hwdev, hinic_mbox_self_aeqe_handler);
+
+	if (!HINIC_IS_VF(hwdev->hwif))
+		hinic_register_pf_mbox_cb(hwdev, HINIC_MOD_COMM,
+					  comm_pf_mbox_handler);
+
+	return 0;
+
+err_alloc_wb_status:
+	free_mbox_info(func_to_func->mbox_resp);
+
+err_alloc_mbox_for_resp:
+	free_mbox_info(func_to_func->mbox_send);
+
+err_alloc_mbox_for_send:
+	destroy_workqueue(func_to_func->workq);
+
+err_create_mbox_workq:
+	kfree(func_to_func);
+
+	return err;
+}
+
+void hinic_func_to_func_free(struct hinic_hwdev *hwdev)
+{
+	struct hinic_mbox_func_to_func *func_to_func = hwdev->func_to_func;
+
+	hinic_aeq_unregister_hw_cb(&hwdev->aeqs, HINIC_MBX_FROM_FUNC);
+	hinic_aeq_unregister_hw_cb(&hwdev->aeqs, HINIC_MBX_SEND_RSLT);
+
+	hinic_unregister_pf_mbox_cb(hwdev, HINIC_MOD_COMM);
+	/* destroy workqueue before free related mbox resources in case of
+	 * illegal resource access
+	 */
+	destroy_workqueue(func_to_func->workq);
+
+	free_mbox_wb_status(func_to_func);
+	free_mbox_info(func_to_func->mbox_resp);
+	free_mbox_info(func_to_func->mbox_send);
+
+	kfree(func_to_func);
+}
+
+int hinic_vf_mbox_random_id_init(struct hinic_hwdev *hwdev)
+{
+	u16 vf_offset;
+	u8 vf_in_pf;
+	int err = 0;
+
+	if (HINIC_IS_VF(hwdev->hwif))
+		return 0;
+
+	vf_offset = hinic_glb_pf_vf_offset(hwdev->hwif);
+
+	for (vf_in_pf = 1; vf_in_pf <= hwdev->nic_cap.max_vf; vf_in_pf++) {
+		err = set_vf_mbox_random_id(hwdev, vf_offset + vf_in_pf);
+		if (err)
+			break;
+	}
+
+	if (err == HINIC_MGMT_CMD_UNSUPPORTED) {
+		hwdev->func_to_func->support_vf_random = false;
+		err = 0;
+		dev_warn(&hwdev->hwif->pdev->dev, "Mgmt is unsupported to set VF%d random id\n",
+			 vf_in_pf - 1);
+	} else if (!err) {
+		hwdev->func_to_func->support_vf_random = true;
+	}
+
+	return err;
+}
diff --git a/drivers/net/ethernet/huawei/hinic/hinic_hw_mbox.h b/drivers/net/ethernet/huawei/hinic/hinic_hw_mbox.h
new file mode 100644
index 000000000000..33ac7814d3b3
--- /dev/null
+++ b/drivers/net/ethernet/huawei/hinic/hinic_hw_mbox.h
@@ -0,0 +1,172 @@
+/* SPDX-License-Identifier: GPL-2.0-only */
+/* Huawei HiNIC PCI Express Linux driver
+ * Copyright(c) 2017 Huawei Technologies Co., Ltd
+ */
+
+#ifndef HINIC_MBOX_H_
+#define HINIC_MBOX_H_
+
+#define HINIC_MBOX_PF_SEND_ERR		0x1
+#define HINIC_MBOX_PF_BUSY_ACTIVE_FW	0x2
+#define HINIC_MBOX_VF_CMD_ERROR		0x3
+
+#define HINIC_MAX_FUNCTIONS		512
+
+#define HINIC_MAX_PF_FUNCS		16
+
+#define HINIC_MBOX_WQ_NAME		"hinic_mbox"
+
+#define HINIC_FUNC_CSR_MAILBOX_DATA_OFF			0x80
+#define HINIC_FUNC_CSR_MAILBOX_CONTROL_OFF		0x0100
+#define HINIC_FUNC_CSR_MAILBOX_INT_OFFSET_OFF		0x0104
+#define HINIC_FUNC_CSR_MAILBOX_RESULT_H_OFF		0x0108
+#define HINIC_FUNC_CSR_MAILBOX_RESULT_L_OFF		0x010C
+
+#define MAX_FUNCTION_NUM		512
+
+struct vf_cmd_check_handle {
+	u8 cmd;
+	bool (*check_cmd)(struct hinic_hwdev *hwdev, u16 src_func_idx,
+			  void *buf_in, u16 in_size);
+};
+
+enum hinic_mbox_ack_type {
+	MBOX_ACK,
+	MBOX_NO_ACK,
+};
+
+struct mbox_msg_info {
+	u8 msg_id;
+	u8 status;
+};
+
+struct hinic_recv_mbox {
+	struct completion	recv_done;
+	void			*mbox;
+	u8			cmd;
+	enum hinic_mod_type	mod;
+	u16			mbox_len;
+	void			*buf_out;
+	enum hinic_mbox_ack_type ack_type;
+	struct mbox_msg_info	msg_info;
+	u8			seq_id;
+	atomic_t		msg_cnt;
+};
+
+struct hinic_send_mbox {
+	struct completion	send_done;
+	u8			*data;
+
+	u64			*wb_status;
+	void			*wb_vaddr;
+	dma_addr_t		wb_paddr;
+};
+
+typedef void (*hinic_vf_mbox_cb)(void *handle, u8 cmd, void *buf_in,
+				u16 in_size, void *buf_out, u16 *out_size);
+typedef int (*hinic_pf_mbox_cb)(void *handle, u16 vf_id, u8 cmd, void *buf_in,
+				u16 in_size, void *buf_out, u16 *out_size);
+
+enum mbox_event_state {
+	EVENT_START = 0,
+	EVENT_FAIL,
+	EVENT_TIMEOUT,
+	EVENT_END,
+};
+
+enum hinic_mbox_cb_state {
+	HINIC_VF_MBOX_CB_REG = 0,
+	HINIC_VF_MBOX_CB_RUNNING,
+	HINIC_PF_MBOX_CB_REG,
+	HINIC_PF_MBOX_CB_RUNNING,
+	HINIC_PPF_MBOX_CB_REG,
+	HINIC_PPF_MBOX_CB_RUNNING,
+	HINIC_PPF_TO_PF_MBOX_CB_REG,
+	HINIC_PPF_TO_PF_MBOX_CB_RUNNIG,
+};
+
+struct hinic_mbox_func_to_func {
+	struct hinic_hwdev	*hwdev;
+	struct hinic_hwif		*hwif;
+
+	struct semaphore	mbox_send_sem;
+	struct semaphore	msg_send_sem;
+	struct hinic_send_mbox	send_mbox;
+
+	struct workqueue_struct *workq;
+
+	struct hinic_recv_mbox	mbox_resp[HINIC_MAX_FUNCTIONS];
+	struct hinic_recv_mbox	mbox_send[HINIC_MAX_FUNCTIONS];
+
+	hinic_vf_mbox_cb	vf_mbox_cb[HINIC_MOD_MAX];
+	hinic_pf_mbox_cb	pf_mbox_cb[HINIC_MOD_MAX];
+	unsigned long		pf_mbox_cb_state[HINIC_MOD_MAX];
+	unsigned long		vf_mbox_cb_state[HINIC_MOD_MAX];
+
+	u8 send_msg_id;
+	enum mbox_event_state event_flag;
+
+	/* lock for mbox event flag */
+	spinlock_t mbox_lock;
+
+	u32 vf_mbx_old_rand_id[MAX_FUNCTION_NUM];
+	u32 vf_mbx_rand_id[MAX_FUNCTION_NUM];
+	bool support_vf_random;
+};
+
+struct hinic_mbox_work {
+	struct work_struct work;
+	u16 src_func_idx;
+	struct hinic_mbox_func_to_func *func_to_func;
+	struct hinic_recv_mbox *recv_mbox;
+};
+
+struct vf_cmd_msg_handle {
+	u8 cmd;
+	int (*cmd_msg_handler)(void *hwdev, u16 vf_id,
+			       void *buf_in, u16 in_size,
+			       void *buf_out, u16 *out_size);
+};
+
+bool hinic_mbox_check_func_id_8B(struct hinic_hwdev *hwdev, u16 func_idx,
+				 void *buf_in, u16 in_size);
+
+bool hinic_mbox_check_cmd_valid(struct hinic_hwdev *hwdev,
+				struct vf_cmd_check_handle *cmd_handle,
+				u16 vf_id, u8 cmd, void *buf_in,
+				u16 in_size, u8 size);
+
+int hinic_register_pf_mbox_cb(struct hinic_hwdev *hwdev,
+			      enum hinic_mod_type mod,
+			      hinic_pf_mbox_cb callback);
+
+int hinic_register_vf_mbox_cb(struct hinic_hwdev *hwdev,
+			      enum hinic_mod_type mod,
+			      hinic_vf_mbox_cb callback);
+
+void hinic_unregister_pf_mbox_cb(struct hinic_hwdev *hwdev,
+				 enum hinic_mod_type mod);
+
+void hinic_unregister_vf_mbox_cb(struct hinic_hwdev *hwdev,
+				 enum hinic_mod_type mod);
+
+int hinic_func_to_func_init(struct hinic_hwdev *hwdev);
+
+void hinic_func_to_func_free(struct hinic_hwdev *hwdev);
+
+int hinic_mbox_to_pf(struct hinic_hwdev *hwdev, enum hinic_mod_type mod,
+		     u8 cmd, void *buf_in, u16 in_size, void *buf_out,
+		     u16 *out_size, u32 timeout);
+
+int hinic_mbox_to_func(struct hinic_mbox_func_to_func *func_to_func,
+		       enum hinic_mod_type mod, u16 cmd, u16 dst_func,
+		       void *buf_in, u16 in_size, void *buf_out,
+		       u16 *out_size, u32 timeout);
+
+int hinic_mbox_to_vf(struct hinic_hwdev *hwdev,
+		     enum hinic_mod_type mod, u16 vf_id, u8 cmd, void *buf_in,
+		     u16 in_size, void *buf_out, u16 *out_size, u32 timeout);
+
+int hinic_vf_mbox_random_id_init(struct hinic_hwdev *hwdev);
+
+#endif
diff --git a/drivers/net/ethernet/huawei/hinic/hinic_hw_mgmt.c b/drivers/net/ethernet/huawei/hinic/hinic_hw_mgmt.c
new file mode 100644
index 000000000000..4aa1f433ed24
--- /dev/null
+++ b/drivers/net/ethernet/huawei/hinic/hinic_hw_mgmt.c
@@ -0,0 +1,681 @@
+// SPDX-License-Identifier: GPL-2.0-only
+/*
+ * Huawei HiNIC PCI Express Linux driver
+ * Copyright(c) 2017 Huawei Technologies Co., Ltd
+ */
+
+#include <linux/kernel.h>
+#include <linux/types.h>
+#include <linux/errno.h>
+#include <linux/pci.h>
+#include <linux/device.h>
+#include <linux/semaphore.h>
+#include <linux/completion.h>
+#include <linux/slab.h>
+#include <net/devlink.h>
+#include <asm/barrier.h>
+
+#include "hinic_devlink.h"
+#include "hinic_hw_if.h"
+#include "hinic_hw_eqs.h"
+#include "hinic_hw_api_cmd.h"
+#include "hinic_hw_mgmt.h"
+#include "hinic_hw_dev.h"
+
+#define SYNC_MSG_ID_MASK                0x1FF
+
+#define SYNC_MSG_ID(pf_to_mgmt)         ((pf_to_mgmt)->sync_msg_id)
+
+#define SYNC_MSG_ID_INC(pf_to_mgmt)     (SYNC_MSG_ID(pf_to_mgmt) = \
+					((SYNC_MSG_ID(pf_to_mgmt) + 1) & \
+					 SYNC_MSG_ID_MASK))
+
+#define MSG_SZ_IS_VALID(in_size)        ((in_size) <= MAX_MSG_LEN)
+
+#define MGMT_MSG_LEN_MIN                20
+#define MGMT_MSG_LEN_STEP               16
+#define MGMT_MSG_RSVD_FOR_DEV           8
+
+#define SEGMENT_LEN                     48
+
+#define MAX_PF_MGMT_BUF_SIZE            2048
+
+/* Data should be SEG LEN size aligned */
+#define MAX_MSG_LEN                     2016
+
+#define MSG_NOT_RESP                    0xFFFF
+
+#define MGMT_MSG_TIMEOUT                5000
+
+#define SET_FUNC_PORT_MBOX_TIMEOUT	30000
+
+#define SET_FUNC_PORT_MGMT_TIMEOUT	25000
+
+#define UPDATE_FW_MGMT_TIMEOUT		20000
+
+#define mgmt_to_pfhwdev(pf_mgmt)        \
+		container_of(pf_mgmt, struct hinic_pfhwdev, pf_to_mgmt)
+
+enum msg_segment_type {
+	NOT_LAST_SEGMENT = 0,
+	LAST_SEGMENT     = 1,
+};
+
+enum mgmt_direction_type {
+	MGMT_DIRECT_SEND = 0,
+	MGMT_RESP        = 1,
+};
+
+enum msg_ack_type {
+	MSG_ACK         = 0,
+	MSG_NO_ACK      = 1,
+};
+
+/**
+ * hinic_register_mgmt_msg_cb - register msg handler for a msg from a module
+ * @pf_to_mgmt: PF to MGMT channel
+ * @mod: module in the chip that this handler will handle its messages
+ * @handle: private data for the callback
+ * @callback: the handler that will handle messages
+ **/
+void hinic_register_mgmt_msg_cb(struct hinic_pf_to_mgmt *pf_to_mgmt,
+				enum hinic_mod_type mod,
+				void *handle,
+				void (*callback)(void *handle,
+						 u8 cmd, void *buf_in,
+						 u16 in_size, void *buf_out,
+						 u16 *out_size))
+{
+	struct hinic_mgmt_cb *mgmt_cb = &pf_to_mgmt->mgmt_cb[mod];
+
+	mgmt_cb->cb = callback;
+	mgmt_cb->handle = handle;
+	mgmt_cb->state = HINIC_MGMT_CB_ENABLED;
+}
+
+/**
+ * hinic_unregister_mgmt_msg_cb - unregister msg handler for a msg from a module
+ * @pf_to_mgmt: PF to MGMT channel
+ * @mod: module in the chip that this handler handles its messages
+ **/
+void hinic_unregister_mgmt_msg_cb(struct hinic_pf_to_mgmt *pf_to_mgmt,
+				  enum hinic_mod_type mod)
+{
+	struct hinic_mgmt_cb *mgmt_cb = &pf_to_mgmt->mgmt_cb[mod];
+
+	mgmt_cb->state &= ~HINIC_MGMT_CB_ENABLED;
+
+	while (mgmt_cb->state & HINIC_MGMT_CB_RUNNING)
+		schedule();
+
+	mgmt_cb->cb = NULL;
+}
+
+/**
+ * prepare_header - prepare the header of the message
+ * @pf_to_mgmt: PF to MGMT channel
+ * @msg_len: the length of the message
+ * @mod: module in the chip that will get the message
+ * @ack_type: ask for response
+ * @direction: the direction of the message
+ * @cmd: command of the message
+ * @msg_id: message id
+ *
+ * Return the prepared header value
+ **/
+static u64 prepare_header(struct hinic_pf_to_mgmt *pf_to_mgmt,
+			  u16 msg_len, enum hinic_mod_type mod,
+			  enum msg_ack_type ack_type,
+			  enum mgmt_direction_type direction,
+			  u16 cmd, u16 msg_id)
+{
+	struct hinic_hwif *hwif = pf_to_mgmt->hwif;
+
+	return HINIC_MSG_HEADER_SET(msg_len, MSG_LEN)           |
+	       HINIC_MSG_HEADER_SET(mod, MODULE)                |
+	       HINIC_MSG_HEADER_SET(SEGMENT_LEN, SEG_LEN)       |
+	       HINIC_MSG_HEADER_SET(ack_type, NO_ACK)           |
+	       HINIC_MSG_HEADER_SET(0, ASYNC_MGMT_TO_PF)        |
+	       HINIC_MSG_HEADER_SET(0, SEQID)                   |
+	       HINIC_MSG_HEADER_SET(LAST_SEGMENT, LAST)         |
+	       HINIC_MSG_HEADER_SET(direction, DIRECTION)       |
+	       HINIC_MSG_HEADER_SET(cmd, CMD)                   |
+	       HINIC_MSG_HEADER_SET(HINIC_HWIF_PCI_INTF(hwif), PCI_INTF) |
+	       HINIC_MSG_HEADER_SET(HINIC_HWIF_PF_IDX(hwif), PF_IDX)     |
+	       HINIC_MSG_HEADER_SET(msg_id, MSG_ID);
+}
+
+/**
+ * prepare_mgmt_cmd - prepare the mgmt command
+ * @mgmt_cmd: pointer to the command to prepare
+ * @header: pointer of the header for the message
+ * @msg: the data of the message
+ * @msg_len: the length of the message
+ **/
+static void prepare_mgmt_cmd(u8 *mgmt_cmd, u64 *header, u8 *msg, u16 msg_len)
+{
+	memset(mgmt_cmd, 0, MGMT_MSG_RSVD_FOR_DEV);
+
+	mgmt_cmd += MGMT_MSG_RSVD_FOR_DEV;
+	memcpy(mgmt_cmd, header, sizeof(*header));
+
+	mgmt_cmd += sizeof(*header);
+	memcpy(mgmt_cmd, msg, msg_len);
+}
+
+/**
+ * mgmt_msg_len - calculate the total message length
+ * @msg_data_len: the length of the message data
+ *
+ * Return the total message length
+ **/
+static u16 mgmt_msg_len(u16 msg_data_len)
+{
+	/* RSVD + HEADER_SIZE + DATA_LEN */
+	u16 msg_len = MGMT_MSG_RSVD_FOR_DEV + sizeof(u64) + msg_data_len;
+
+	if (msg_len > MGMT_MSG_LEN_MIN)
+		msg_len = MGMT_MSG_LEN_MIN +
+			   ALIGN((msg_len - MGMT_MSG_LEN_MIN),
+				 MGMT_MSG_LEN_STEP);
+	else
+		msg_len = MGMT_MSG_LEN_MIN;
+
+	return msg_len;
+}
+
+/**
+ * send_msg_to_mgmt - send message to mgmt by API CMD
+ * @pf_to_mgmt: PF to MGMT channel
+ * @mod: module in the chip that will get the message
+ * @cmd: command of the message
+ * @data: the msg data
+ * @data_len: the msg data length
+ * @ack_type: ask for response
+ * @direction: the direction of the original message
+ * @resp_msg_id: msg id to response for
+ *
+ * Return 0 - Success, negative - Failure
+ **/
+static int send_msg_to_mgmt(struct hinic_pf_to_mgmt *pf_to_mgmt,
+			    enum hinic_mod_type mod, u8 cmd,
+			    u8 *data, u16 data_len,
+			    enum msg_ack_type ack_type,
+			    enum mgmt_direction_type direction,
+			    u16 resp_msg_id)
+{
+	struct hinic_api_cmd_chain *chain;
+	u64 header;
+	u16 msg_id;
+
+	msg_id = SYNC_MSG_ID(pf_to_mgmt);
+
+	if (direction == MGMT_RESP) {
+		header = prepare_header(pf_to_mgmt, data_len, mod, ack_type,
+					direction, cmd, resp_msg_id);
+	} else {
+		SYNC_MSG_ID_INC(pf_to_mgmt);
+		header = prepare_header(pf_to_mgmt, data_len, mod, ack_type,
+					direction, cmd, msg_id);
+	}
+
+	prepare_mgmt_cmd(pf_to_mgmt->sync_msg_buf, &header, data, data_len);
+
+	chain = pf_to_mgmt->cmd_chain[HINIC_API_CMD_WRITE_TO_MGMT_CPU];
+	return hinic_api_cmd_write(chain, HINIC_NODE_ID_MGMT,
+				   pf_to_mgmt->sync_msg_buf,
+				   mgmt_msg_len(data_len));
+}
+
+/**
+ * msg_to_mgmt_sync - send sync message to mgmt
+ * @pf_to_mgmt: PF to MGMT channel
+ * @mod: module in the chip that will get the message
+ * @cmd: command of the message
+ * @buf_in: the msg data
+ * @in_size: the msg data length
+ * @buf_out: response
+ * @out_size: response length
+ * @direction: the direction of the original message
+ * @resp_msg_id: msg id to response for
+ * @timeout: time-out period of waiting for response
+ *
+ * Return 0 - Success, negative - Failure
+ **/
+static int msg_to_mgmt_sync(struct hinic_pf_to_mgmt *pf_to_mgmt,
+			    enum hinic_mod_type mod, u8 cmd,
+			    u8 *buf_in, u16 in_size,
+			    u8 *buf_out, u16 *out_size,
+			    enum mgmt_direction_type direction,
+			    u16 resp_msg_id, u32 timeout)
+{
+	struct hinic_hwif *hwif = pf_to_mgmt->hwif;
+	struct pci_dev *pdev = hwif->pdev;
+	struct hinic_recv_msg *recv_msg;
+	struct completion *recv_done;
+	unsigned long timeo;
+	u16 msg_id;
+	int err;
+
+	/* Lock the sync_msg_buf */
+	down(&pf_to_mgmt->sync_msg_lock);
+
+	recv_msg = &pf_to_mgmt->recv_resp_msg_from_mgmt;
+	recv_done = &recv_msg->recv_done;
+
+	if (resp_msg_id == MSG_NOT_RESP)
+		msg_id = SYNC_MSG_ID(pf_to_mgmt);
+	else
+		msg_id = resp_msg_id;
+
+	init_completion(recv_done);
+
+	err = send_msg_to_mgmt(pf_to_mgmt, mod, cmd, buf_in, in_size,
+			       MSG_ACK, direction, resp_msg_id);
+	if (err) {
+		dev_err(&pdev->dev, "Failed to send sync msg to mgmt\n");
+		goto unlock_sync_msg;
+	}
+
+	timeo = msecs_to_jiffies(timeout ? timeout : MGMT_MSG_TIMEOUT);
+
+	if (!wait_for_completion_timeout(recv_done, timeo)) {
+		dev_err(&pdev->dev, "MGMT timeout, MSG id = %d\n", msg_id);
+		hinic_dump_aeq_info(pf_to_mgmt->hwdev);
+		err = -ETIMEDOUT;
+		goto unlock_sync_msg;
+	}
+
+	smp_rmb();      /* verify reading after completion */
+
+	if (recv_msg->msg_id != msg_id) {
+		dev_err(&pdev->dev, "incorrect MSG for id = %d\n", msg_id);
+		err = -EFAULT;
+		goto unlock_sync_msg;
+	}
+
+	if (buf_out && recv_msg->msg_len <= MAX_PF_MGMT_BUF_SIZE) {
+		memcpy(buf_out, recv_msg->msg, recv_msg->msg_len);
+		*out_size = recv_msg->msg_len;
+	}
+
+unlock_sync_msg:
+	up(&pf_to_mgmt->sync_msg_lock);
+	return err;
+}
+
+/**
+ * msg_to_mgmt_async - send message to mgmt without response
+ * @pf_to_mgmt: PF to MGMT channel
+ * @mod: module in the chip that will get the message
+ * @cmd: command of the message
+ * @buf_in: the msg data
+ * @in_size: the msg data length
+ * @direction: the direction of the original message
+ * @resp_msg_id: msg id to response for
+ *
+ * Return 0 - Success, negative - Failure
+ **/
+static int msg_to_mgmt_async(struct hinic_pf_to_mgmt *pf_to_mgmt,
+			     enum hinic_mod_type mod, u8 cmd,
+			     u8 *buf_in, u16 in_size,
+			     enum mgmt_direction_type direction,
+			     u16 resp_msg_id)
+{
+	int err;
+
+	/* Lock the sync_msg_buf */
+	down(&pf_to_mgmt->sync_msg_lock);
+
+	err = send_msg_to_mgmt(pf_to_mgmt, mod, cmd, buf_in, in_size,
+			       MSG_NO_ACK, direction, resp_msg_id);
+
+	up(&pf_to_mgmt->sync_msg_lock);
+	return err;
+}
+
+/**
+ * hinic_msg_to_mgmt - send message to mgmt
+ * @pf_to_mgmt: PF to MGMT channel
+ * @mod: module in the chip that will get the message
+ * @cmd: command of the message
+ * @buf_in: the msg data
+ * @in_size: the msg data length
+ * @buf_out: response
+ * @out_size: returned response length
+ * @sync: sync msg or async msg
+ *
+ * Return 0 - Success, negative - Failure
+ **/
+int hinic_msg_to_mgmt(struct hinic_pf_to_mgmt *pf_to_mgmt,
+		      enum hinic_mod_type mod, u8 cmd,
+		      void *buf_in, u16 in_size, void *buf_out, u16 *out_size,
+		      enum hinic_mgmt_msg_type sync)
+{
+	struct hinic_hwif *hwif = pf_to_mgmt->hwif;
+	struct pci_dev *pdev = hwif->pdev;
+	u32 timeout = 0;
+
+	if (sync != HINIC_MGMT_MSG_SYNC) {
+		dev_err(&pdev->dev, "Invalid MGMT msg type\n");
+		return -EINVAL;
+	}
+
+	if (!MSG_SZ_IS_VALID(in_size)) {
+		dev_err(&pdev->dev, "Invalid MGMT msg buffer size\n");
+		return -EINVAL;
+	}
+
+	if (HINIC_IS_VF(hwif)) {
+		if (cmd == HINIC_PORT_CMD_SET_FUNC_STATE)
+			timeout = SET_FUNC_PORT_MBOX_TIMEOUT;
+
+		return hinic_mbox_to_pf(pf_to_mgmt->hwdev, mod, cmd, buf_in,
+					in_size, buf_out, out_size, timeout);
+	} else {
+		if (cmd == HINIC_PORT_CMD_SET_FUNC_STATE)
+			timeout = SET_FUNC_PORT_MGMT_TIMEOUT;
+		else if (cmd == HINIC_PORT_CMD_UPDATE_FW)
+			timeout = UPDATE_FW_MGMT_TIMEOUT;
+
+		return msg_to_mgmt_sync(pf_to_mgmt, mod, cmd, buf_in, in_size,
+				buf_out, out_size, MGMT_DIRECT_SEND,
+				MSG_NOT_RESP, timeout);
+	}
+}
+
+static void recv_mgmt_msg_work_handler(struct work_struct *work)
+{
+	struct hinic_mgmt_msg_handle_work *mgmt_work =
+		container_of(work, struct hinic_mgmt_msg_handle_work, work);
+	struct hinic_pf_to_mgmt *pf_to_mgmt = mgmt_work->pf_to_mgmt;
+	struct pci_dev *pdev = pf_to_mgmt->hwif->pdev;
+	u8 *buf_out = pf_to_mgmt->mgmt_ack_buf;
+	struct hinic_mgmt_cb *mgmt_cb;
+	unsigned long cb_state;
+	u16 out_size = 0;
+
+	memset(buf_out, 0, MAX_PF_MGMT_BUF_SIZE);
+
+	if (mgmt_work->mod >= HINIC_MOD_MAX) {
+		dev_err(&pdev->dev, "Unknown MGMT MSG module = %d\n",
+			mgmt_work->mod);
+		kfree(mgmt_work->msg);
+		kfree(mgmt_work);
+		return;
+	}
+
+	mgmt_cb = &pf_to_mgmt->mgmt_cb[mgmt_work->mod];
+
+	cb_state = cmpxchg(&mgmt_cb->state,
+			   HINIC_MGMT_CB_ENABLED,
+			   HINIC_MGMT_CB_ENABLED | HINIC_MGMT_CB_RUNNING);
+
+	if (cb_state == HINIC_MGMT_CB_ENABLED && mgmt_cb->cb)
+		mgmt_cb->cb(mgmt_cb->handle, mgmt_work->cmd,
+			    mgmt_work->msg, mgmt_work->msg_len,
+			    buf_out, &out_size);
+	else
+		dev_err(&pdev->dev, "No MGMT msg handler, mod: %d, cmd: %d\n",
+			mgmt_work->mod, mgmt_work->cmd);
+
+	mgmt_cb->state &= ~HINIC_MGMT_CB_RUNNING;
+
+	if (!mgmt_work->async_mgmt_to_pf)
+		/* MGMT sent sync msg, send the response */
+		msg_to_mgmt_async(pf_to_mgmt, mgmt_work->mod, mgmt_work->cmd,
+				  buf_out, out_size, MGMT_RESP,
+				  mgmt_work->msg_id);
+
+	kfree(mgmt_work->msg);
+	kfree(mgmt_work);
+}
+
+/**
+ * mgmt_recv_msg_handler - handler for message from mgmt cpu
+ * @pf_to_mgmt: PF to MGMT channel
+ * @recv_msg: received message details
+ **/
+static void mgmt_recv_msg_handler(struct hinic_pf_to_mgmt *pf_to_mgmt,
+				  struct hinic_recv_msg *recv_msg)
+{
+	struct hinic_mgmt_msg_handle_work *mgmt_work = NULL;
+
+	mgmt_work = kzalloc(sizeof(*mgmt_work), GFP_KERNEL);
+	if (!mgmt_work)
+		return;
+
+	if (recv_msg->msg_len) {
+		mgmt_work->msg = kzalloc(recv_msg->msg_len, GFP_KERNEL);
+		if (!mgmt_work->msg) {
+			kfree(mgmt_work);
+			return;
+		}
+	}
+
+	mgmt_work->pf_to_mgmt = pf_to_mgmt;
+	mgmt_work->msg_len = recv_msg->msg_len;
+	memcpy(mgmt_work->msg, recv_msg->msg, recv_msg->msg_len);
+	mgmt_work->msg_id = recv_msg->msg_id;
+	mgmt_work->mod = recv_msg->mod;
+	mgmt_work->cmd = recv_msg->cmd;
+	mgmt_work->async_mgmt_to_pf = recv_msg->async_mgmt_to_pf;
+
+	INIT_WORK(&mgmt_work->work, recv_mgmt_msg_work_handler);
+	queue_work(pf_to_mgmt->workq, &mgmt_work->work);
+}
+
+/**
+ * mgmt_resp_msg_handler - handler for a response message from mgmt cpu
+ * @pf_to_mgmt: PF to MGMT channel
+ * @recv_msg: received message details
+ **/
+static void mgmt_resp_msg_handler(struct hinic_pf_to_mgmt *pf_to_mgmt,
+				  struct hinic_recv_msg *recv_msg)
+{
+	wmb();  /* verify writing all, before reading */
+
+	complete(&recv_msg->recv_done);
+}
+
+/**
+ * recv_mgmt_msg_handler - handler for a message from mgmt cpu
+ * @pf_to_mgmt: PF to MGMT channel
+ * @header: the header of the message
+ * @recv_msg: received message details
+ **/
+static void recv_mgmt_msg_handler(struct hinic_pf_to_mgmt *pf_to_mgmt,
+				  u64 *header, struct hinic_recv_msg *recv_msg)
+{
+	struct hinic_hwif *hwif = pf_to_mgmt->hwif;
+	struct pci_dev *pdev = hwif->pdev;
+	int seq_id, seg_len;
+	u8 *msg_body;
+
+	seq_id = HINIC_MSG_HEADER_GET(*header, SEQID);
+	seg_len = HINIC_MSG_HEADER_GET(*header, SEG_LEN);
+
+	if (seq_id >= (MAX_MSG_LEN / SEGMENT_LEN)) {
+		dev_err(&pdev->dev, "recv big mgmt msg\n");
+		return;
+	}
+
+	msg_body = (u8 *)header + sizeof(*header);
+	memcpy(recv_msg->msg + seq_id * SEGMENT_LEN, msg_body, seg_len);
+
+	if (!HINIC_MSG_HEADER_GET(*header, LAST))
+		return;
+
+	recv_msg->cmd = HINIC_MSG_HEADER_GET(*header, CMD);
+	recv_msg->mod = HINIC_MSG_HEADER_GET(*header, MODULE);
+	recv_msg->async_mgmt_to_pf = HINIC_MSG_HEADER_GET(*header,
+							  ASYNC_MGMT_TO_PF);
+	recv_msg->msg_len = HINIC_MSG_HEADER_GET(*header, MSG_LEN);
+	recv_msg->msg_id = HINIC_MSG_HEADER_GET(*header, MSG_ID);
+
+	if (HINIC_MSG_HEADER_GET(*header, DIRECTION) == MGMT_RESP)
+		mgmt_resp_msg_handler(pf_to_mgmt, recv_msg);
+	else
+		mgmt_recv_msg_handler(pf_to_mgmt, recv_msg);
+}
+
+/**
+ * mgmt_msg_aeqe_handler - handler for a mgmt message event
+ * @handle: PF to MGMT channel
+ * @data: the header of the message
+ * @size: unused
+ **/
+static void mgmt_msg_aeqe_handler(void *handle, void *data, u8 size)
+{
+	struct hinic_pf_to_mgmt *pf_to_mgmt = handle;
+	struct hinic_recv_msg *recv_msg;
+	u64 *header = (u64 *)data;
+
+	recv_msg = HINIC_MSG_HEADER_GET(*header, DIRECTION) ==
+		   MGMT_DIRECT_SEND ?
+		   &pf_to_mgmt->recv_msg_from_mgmt :
+		   &pf_to_mgmt->recv_resp_msg_from_mgmt;
+
+	recv_mgmt_msg_handler(pf_to_mgmt, header, recv_msg);
+}
+
+/**
+ * alloc_recv_msg - allocate receive message memory
+ * @pf_to_mgmt: PF to MGMT channel
+ * @recv_msg: pointer that will hold the allocated data
+ *
+ * Return 0 - Success, negative - Failure
+ **/
+static int alloc_recv_msg(struct hinic_pf_to_mgmt *pf_to_mgmt,
+			  struct hinic_recv_msg *recv_msg)
+{
+	struct hinic_hwif *hwif = pf_to_mgmt->hwif;
+	struct pci_dev *pdev = hwif->pdev;
+
+	recv_msg->msg = devm_kzalloc(&pdev->dev, MAX_PF_MGMT_BUF_SIZE,
+				     GFP_KERNEL);
+	if (!recv_msg->msg)
+		return -ENOMEM;
+
+	recv_msg->buf_out = devm_kzalloc(&pdev->dev, MAX_PF_MGMT_BUF_SIZE,
+					 GFP_KERNEL);
+	if (!recv_msg->buf_out)
+		return -ENOMEM;
+
+	return 0;
+}
+
+/**
+ * alloc_msg_buf - allocate all the message buffers of PF to MGMT channel
+ * @pf_to_mgmt: PF to MGMT channel
+ *
+ * Return 0 - Success, negative - Failure
+ **/
+static int alloc_msg_buf(struct hinic_pf_to_mgmt *pf_to_mgmt)
+{
+	struct hinic_hwif *hwif = pf_to_mgmt->hwif;
+	struct pci_dev *pdev = hwif->pdev;
+	int err;
+
+	err = alloc_recv_msg(pf_to_mgmt,
+			     &pf_to_mgmt->recv_msg_from_mgmt);
+	if (err) {
+		dev_err(&pdev->dev, "Failed to allocate recv msg\n");
+		return err;
+	}
+
+	err = alloc_recv_msg(pf_to_mgmt,
+			     &pf_to_mgmt->recv_resp_msg_from_mgmt);
+	if (err) {
+		dev_err(&pdev->dev, "Failed to allocate resp recv msg\n");
+		return err;
+	}
+
+	pf_to_mgmt->sync_msg_buf = devm_kzalloc(&pdev->dev,
+						MAX_PF_MGMT_BUF_SIZE,
+						GFP_KERNEL);
+	if (!pf_to_mgmt->sync_msg_buf)
+		return -ENOMEM;
+
+	pf_to_mgmt->mgmt_ack_buf = devm_kzalloc(&pdev->dev,
+						MAX_PF_MGMT_BUF_SIZE,
+						GFP_KERNEL);
+	if (!pf_to_mgmt->mgmt_ack_buf)
+		return -ENOMEM;
+
+	return 0;
+}
+
+/**
+ * hinic_pf_to_mgmt_init - initialize PF to MGMT channel
+ * @pf_to_mgmt: PF to MGMT channel
+ * @hwif: HW interface the PF to MGMT will use for accessing HW
+ *
+ * Return 0 - Success, negative - Failure
+ **/
+int hinic_pf_to_mgmt_init(struct hinic_pf_to_mgmt *pf_to_mgmt,
+			  struct hinic_hwif *hwif)
+{
+	struct hinic_pfhwdev *pfhwdev = mgmt_to_pfhwdev(pf_to_mgmt);
+	struct hinic_hwdev *hwdev = &pfhwdev->hwdev;
+	struct pci_dev *pdev = hwif->pdev;
+	int err;
+
+	pf_to_mgmt->hwif = hwif;
+	pf_to_mgmt->hwdev = hwdev;
+
+	if (HINIC_IS_VF(hwif))
+		return 0;
+
+	err = hinic_health_reporters_create(hwdev->devlink_dev);
+	if (err)
+		return err;
+
+	sema_init(&pf_to_mgmt->sync_msg_lock, 1);
+	pf_to_mgmt->workq = create_singlethread_workqueue("hinic_mgmt");
+	if (!pf_to_mgmt->workq) {
+		dev_err(&pdev->dev, "Failed to initialize MGMT workqueue\n");
+		hinic_health_reporters_destroy(hwdev->devlink_dev);
+		return -ENOMEM;
+	}
+	pf_to_mgmt->sync_msg_id = 0;
+
+	err = alloc_msg_buf(pf_to_mgmt);
+	if (err) {
+		dev_err(&pdev->dev, "Failed to allocate msg buffers\n");
+		destroy_workqueue(pf_to_mgmt->workq);
+		hinic_health_reporters_destroy(hwdev->devlink_dev);
+		return err;
+	}
+
+	err = hinic_api_cmd_init(pf_to_mgmt->cmd_chain, hwif);
+	if (err) {
+		dev_err(&pdev->dev, "Failed to initialize cmd chains\n");
+		destroy_workqueue(pf_to_mgmt->workq);
+		hinic_health_reporters_destroy(hwdev->devlink_dev);
+		return err;
+	}
+
+	hinic_aeq_register_hw_cb(&hwdev->aeqs, HINIC_MSG_FROM_MGMT_CPU,
+				 pf_to_mgmt,
+				 mgmt_msg_aeqe_handler);
+	return 0;
+}
+
+/**
+ * hinic_pf_to_mgmt_free - free PF to MGMT channel
+ * @pf_to_mgmt: PF to MGMT channel
+ **/
+void hinic_pf_to_mgmt_free(struct hinic_pf_to_mgmt *pf_to_mgmt)
+{
+	struct hinic_pfhwdev *pfhwdev = mgmt_to_pfhwdev(pf_to_mgmt);
+	struct hinic_hwdev *hwdev = &pfhwdev->hwdev;
+
+	if (HINIC_IS_VF(hwdev->hwif))
+		return;
+
+	hinic_aeq_unregister_hw_cb(&hwdev->aeqs, HINIC_MSG_FROM_MGMT_CPU);
+	hinic_api_cmd_free(pf_to_mgmt->cmd_chain);
+	destroy_workqueue(pf_to_mgmt->workq);
+	hinic_health_reporters_destroy(hwdev->devlink_dev);
+}
diff --git a/drivers/net/ethernet/huawei/hinic/hinic_hw_mgmt.h b/drivers/net/ethernet/huawei/hinic/hinic_hw_mgmt.h
new file mode 100644
index 000000000000..4ca81cc838db
--- /dev/null
+++ b/drivers/net/ethernet/huawei/hinic/hinic_hw_mgmt.h
@@ -0,0 +1,181 @@
+/* SPDX-License-Identifier: GPL-2.0-only */
+/*
+ * Huawei HiNIC PCI Express Linux driver
+ * Copyright(c) 2017 Huawei Technologies Co., Ltd
+ */
+
+#ifndef HINIC_HW_MGMT_H
+#define HINIC_HW_MGMT_H
+
+#include <linux/types.h>
+#include <linux/semaphore.h>
+#include <linux/completion.h>
+#include <linux/bitops.h>
+
+#include "hinic_hw_if.h"
+#include "hinic_hw_api_cmd.h"
+
+#define HINIC_MSG_HEADER_MSG_LEN_SHIFT                          0
+#define HINIC_MSG_HEADER_MODULE_SHIFT                           11
+#define HINIC_MSG_HEADER_SEG_LEN_SHIFT                          16
+#define HINIC_MSG_HEADER_NO_ACK_SHIFT                           22
+#define HINIC_MSG_HEADER_ASYNC_MGMT_TO_PF_SHIFT                 23
+#define HINIC_MSG_HEADER_SEQID_SHIFT                            24
+#define HINIC_MSG_HEADER_LAST_SHIFT                             30
+#define HINIC_MSG_HEADER_DIRECTION_SHIFT                        31
+#define HINIC_MSG_HEADER_CMD_SHIFT                              32
+#define HINIC_MSG_HEADER_ZEROS_SHIFT                            40
+#define HINIC_MSG_HEADER_PCI_INTF_SHIFT                         48
+#define HINIC_MSG_HEADER_PF_IDX_SHIFT                           50
+#define HINIC_MSG_HEADER_MSG_ID_SHIFT                           54
+
+#define HINIC_MSG_HEADER_MSG_LEN_MASK                           0x7FF
+#define HINIC_MSG_HEADER_MODULE_MASK                            0x1F
+#define HINIC_MSG_HEADER_SEG_LEN_MASK                           0x3F
+#define HINIC_MSG_HEADER_NO_ACK_MASK                            0x1
+#define HINIC_MSG_HEADER_ASYNC_MGMT_TO_PF_MASK                  0x1
+#define HINIC_MSG_HEADER_SEQID_MASK                             0x3F
+#define HINIC_MSG_HEADER_LAST_MASK                              0x1
+#define HINIC_MSG_HEADER_DIRECTION_MASK                         0x1
+#define HINIC_MSG_HEADER_CMD_MASK                               0xFF
+#define HINIC_MSG_HEADER_ZEROS_MASK                             0xFF
+#define HINIC_MSG_HEADER_PCI_INTF_MASK                          0x3
+#define HINIC_MSG_HEADER_PF_IDX_MASK                            0xF
+#define HINIC_MSG_HEADER_MSG_ID_MASK                            0x3FF
+
+#define HINIC_MSG_HEADER_SET(val, member)                       \
+		((u64)((val) & HINIC_MSG_HEADER_##member##_MASK) << \
+		 HINIC_MSG_HEADER_##member##_SHIFT)
+
+#define HINIC_MSG_HEADER_GET(val, member)                       \
+		(((val) >> HINIC_MSG_HEADER_##member##_SHIFT) & \
+		 HINIC_MSG_HEADER_##member##_MASK)
+
+enum hinic_mgmt_msg_type {
+	HINIC_MGMT_MSG_SYNC = 1,
+};
+
+enum hinic_cfg_cmd {
+	HINIC_CFG_NIC_CAP = 0,
+};
+
+enum hinic_comm_cmd {
+	HINIC_COMM_CMD_START_FLR          = 0x1,
+	HINIC_COMM_CMD_IO_STATUS_GET    = 0x3,
+	HINIC_COMM_CMD_DMA_ATTR_SET	    = 0x4,
+
+	HINIC_COMM_CMD_CMDQ_CTXT_SET    = 0x10,
+	HINIC_COMM_CMD_CMDQ_CTXT_GET    = 0x11,
+
+	HINIC_COMM_CMD_HWCTXT_SET       = 0x12,
+	HINIC_COMM_CMD_HWCTXT_GET       = 0x13,
+
+	HINIC_COMM_CMD_SQ_HI_CI_SET     = 0x14,
+
+	HINIC_COMM_CMD_RES_STATE_SET    = 0x24,
+
+	HINIC_COMM_CMD_IO_RES_CLEAR     = 0x29,
+
+	HINIC_COMM_CMD_CEQ_CTRL_REG_WR_BY_UP = 0x33,
+
+	HINIC_COMM_CMD_MSI_CTRL_REG_WR_BY_UP,
+	HINIC_COMM_CMD_MSI_CTRL_REG_RD_BY_UP,
+
+	HINIC_COMM_CMD_FAULT_REPORT	= 0x37,
+
+	HINIC_COMM_CMD_SET_LED_STATUS	= 0x4a,
+
+	HINIC_COMM_CMD_L2NIC_RESET	= 0x4b,
+
+	HINIC_COMM_CMD_PAGESIZE_SET	= 0x50,
+
+	HINIC_COMM_CMD_GET_BOARD_INFO	= 0x52,
+
+	HINIC_COMM_CMD_WATCHDOG_INFO	= 0x56,
+
+	HINIC_MGMT_CMD_SET_VF_RANDOM_ID = 0x61,
+
+	HINIC_COMM_CMD_MAX,
+};
+
+enum hinic_mgmt_cb_state {
+	HINIC_MGMT_CB_ENABLED = BIT(0),
+	HINIC_MGMT_CB_RUNNING = BIT(1),
+};
+
+struct hinic_recv_msg {
+	u8                      *msg;
+	u8                      *buf_out;
+
+	struct completion       recv_done;
+
+	u16                     cmd;
+	enum hinic_mod_type     mod;
+	int                     async_mgmt_to_pf;
+
+	u16                     msg_len;
+	u16                     msg_id;
+};
+
+struct hinic_mgmt_cb {
+	void    (*cb)(void *handle, u8 cmd,
+		      void *buf_in, u16 in_size,
+		      void *buf_out, u16 *out_size);
+
+	void            *handle;
+	unsigned long   state;
+};
+
+struct hinic_pf_to_mgmt {
+	struct hinic_hwif               *hwif;
+	struct hinic_hwdev		*hwdev;
+	struct semaphore                sync_msg_lock;
+	u16                             sync_msg_id;
+	u8                              *sync_msg_buf;
+	void				*mgmt_ack_buf;
+
+	struct hinic_recv_msg           recv_resp_msg_from_mgmt;
+	struct hinic_recv_msg           recv_msg_from_mgmt;
+
+	struct hinic_api_cmd_chain      *cmd_chain[HINIC_API_CMD_MAX];
+
+	struct hinic_mgmt_cb            mgmt_cb[HINIC_MOD_MAX];
+
+	struct workqueue_struct		*workq;
+};
+
+struct hinic_mgmt_msg_handle_work {
+	struct work_struct work;
+	struct hinic_pf_to_mgmt *pf_to_mgmt;
+
+	void			*msg;
+	u16			msg_len;
+
+	enum hinic_mod_type	mod;
+	u8			cmd;
+	u16			msg_id;
+	int			async_mgmt_to_pf;
+};
+
+void hinic_register_mgmt_msg_cb(struct hinic_pf_to_mgmt *pf_to_mgmt,
+				enum hinic_mod_type mod,
+				void *handle,
+				void (*callback)(void *handle,
+						 u8 cmd, void *buf_in,
+						 u16 in_size, void *buf_out,
+						 u16 *out_size));
+
+void hinic_unregister_mgmt_msg_cb(struct hinic_pf_to_mgmt *pf_to_mgmt,
+				  enum hinic_mod_type mod);
+
+int hinic_msg_to_mgmt(struct hinic_pf_to_mgmt *pf_to_mgmt,
+		      enum hinic_mod_type mod, u8 cmd,
+		      void *buf_in, u16 in_size, void *buf_out, u16 *out_size,
+		      enum hinic_mgmt_msg_type sync);
+
+int hinic_pf_to_mgmt_init(struct hinic_pf_to_mgmt *pf_to_mgmt,
+			  struct hinic_hwif *hwif);
+
+void hinic_pf_to_mgmt_free(struct hinic_pf_to_mgmt *pf_to_mgmt);
+
+#endif
diff --git a/drivers/net/ethernet/huawei/hinic/hinic_hw_qp.c b/drivers/net/ethernet/huawei/hinic/hinic_hw_qp.c
new file mode 100644
index 000000000000..537a8098bc4e
--- /dev/null
+++ b/drivers/net/ethernet/huawei/hinic/hinic_hw_qp.c
@@ -0,0 +1,971 @@
+// SPDX-License-Identifier: GPL-2.0-only
+/*
+ * Huawei HiNIC PCI Express Linux driver
+ * Copyright(c) 2017 Huawei Technologies Co., Ltd
+ */
+
+#include <linux/kernel.h>
+#include <linux/types.h>
+#include <linux/pci.h>
+#include <linux/device.h>
+#include <linux/dma-mapping.h>
+#include <linux/vmalloc.h>
+#include <linux/errno.h>
+#include <linux/sizes.h>
+#include <linux/atomic.h>
+#include <linux/skbuff.h>
+#include <linux/io.h>
+#include <asm/barrier.h>
+#include <asm/byteorder.h>
+
+#include "hinic_common.h"
+#include "hinic_hw_if.h"
+#include "hinic_hw_wqe.h"
+#include "hinic_hw_wq.h"
+#include "hinic_hw_qp_ctxt.h"
+#include "hinic_hw_qp.h"
+#include "hinic_hw_io.h"
+
+#define SQ_DB_OFF               SZ_2K
+
+/* The number of cache line to prefetch Until threshold state */
+#define WQ_PREFETCH_MAX         2
+/* The number of cache line to prefetch After threshold state */
+#define WQ_PREFETCH_MIN         1
+/* Threshold state */
+#define WQ_PREFETCH_THRESHOLD   256
+
+/* sizes of the SQ/RQ ctxt */
+#define Q_CTXT_SIZE             48
+#define CTXT_RSVD               240
+
+#define SQ_CTXT_OFFSET(max_sqs, max_rqs, q_id)  \
+		(((max_rqs) + (max_sqs)) * CTXT_RSVD + (q_id) * Q_CTXT_SIZE)
+
+#define RQ_CTXT_OFFSET(max_sqs, max_rqs, q_id)  \
+		(((max_rqs) + (max_sqs)) * CTXT_RSVD + \
+		 (max_sqs + (q_id)) * Q_CTXT_SIZE)
+
+#define SIZE_16BYTES(size)              (ALIGN(size, 16) >> 4)
+#define SIZE_8BYTES(size)               (ALIGN(size, 8) >> 3)
+#define SECT_SIZE_FROM_8BYTES(size)     ((size) << 3)
+
+#define SQ_DB_PI_HI_SHIFT       8
+#define SQ_DB_PI_HI(prod_idx)   ((prod_idx) >> SQ_DB_PI_HI_SHIFT)
+
+#define SQ_DB_PI_LOW_MASK       0xFF
+#define SQ_DB_PI_LOW(prod_idx)  ((prod_idx) & SQ_DB_PI_LOW_MASK)
+
+#define SQ_DB_ADDR(sq, pi)      ((u64 *)((sq)->db_base) + SQ_DB_PI_LOW(pi))
+
+#define SQ_MASKED_IDX(sq, idx)  ((idx) & (sq)->wq->mask)
+#define RQ_MASKED_IDX(rq, idx)  ((idx) & (rq)->wq->mask)
+
+enum sq_wqe_type {
+	SQ_NORMAL_WQE = 0,
+};
+
+enum rq_completion_fmt {
+	RQ_COMPLETE_SGE = 1
+};
+
+void hinic_qp_prepare_header(struct hinic_qp_ctxt_header *qp_ctxt_hdr,
+			     enum hinic_qp_ctxt_type ctxt_type,
+			     u16 num_queues, u16 max_queues)
+{
+	u16 max_sqs = max_queues;
+	u16 max_rqs = max_queues;
+
+	qp_ctxt_hdr->num_queues = num_queues;
+	qp_ctxt_hdr->queue_type = ctxt_type;
+
+	if (ctxt_type == HINIC_QP_CTXT_TYPE_SQ)
+		qp_ctxt_hdr->addr_offset = SQ_CTXT_OFFSET(max_sqs, max_rqs, 0);
+	else
+		qp_ctxt_hdr->addr_offset = RQ_CTXT_OFFSET(max_sqs, max_rqs, 0);
+
+	qp_ctxt_hdr->addr_offset = SIZE_16BYTES(qp_ctxt_hdr->addr_offset);
+
+	hinic_cpu_to_be32(qp_ctxt_hdr, sizeof(*qp_ctxt_hdr));
+}
+
+void hinic_sq_prepare_ctxt(struct hinic_sq_ctxt *sq_ctxt,
+			   struct hinic_sq *sq, u16 global_qid)
+{
+	u32 wq_page_pfn_hi, wq_page_pfn_lo, wq_block_pfn_hi, wq_block_pfn_lo;
+	u64 wq_page_addr, wq_page_pfn, wq_block_pfn;
+	u16 pi_start, ci_start;
+	struct hinic_wq *wq;
+
+	wq = sq->wq;
+	ci_start = atomic_read(&wq->cons_idx);
+	pi_start = atomic_read(&wq->prod_idx);
+
+	/* Read the first page paddr from the WQ page paddr ptrs */
+	wq_page_addr = be64_to_cpu(*wq->block_vaddr);
+
+	wq_page_pfn = HINIC_WQ_PAGE_PFN(wq_page_addr);
+	wq_page_pfn_hi = upper_32_bits(wq_page_pfn);
+	wq_page_pfn_lo = lower_32_bits(wq_page_pfn);
+
+	/* If only one page, use 0-level CLA */
+	if (wq->num_q_pages == 1)
+		wq_block_pfn = HINIC_WQ_BLOCK_PFN(wq_page_addr);
+	else
+		wq_block_pfn = HINIC_WQ_BLOCK_PFN(wq->block_paddr);
+
+	wq_block_pfn_hi = upper_32_bits(wq_block_pfn);
+	wq_block_pfn_lo = lower_32_bits(wq_block_pfn);
+
+	sq_ctxt->ceq_attr = HINIC_SQ_CTXT_CEQ_ATTR_SET(global_qid,
+						       GLOBAL_SQ_ID) |
+			    HINIC_SQ_CTXT_CEQ_ATTR_SET(0, EN);
+
+	sq_ctxt->ci_wrapped = HINIC_SQ_CTXT_CI_SET(ci_start, IDX) |
+			      HINIC_SQ_CTXT_CI_SET(1, WRAPPED);
+
+	sq_ctxt->wq_hi_pfn_pi =
+			HINIC_SQ_CTXT_WQ_PAGE_SET(wq_page_pfn_hi, HI_PFN) |
+			HINIC_SQ_CTXT_WQ_PAGE_SET(pi_start, PI);
+
+	sq_ctxt->wq_lo_pfn = wq_page_pfn_lo;
+
+	sq_ctxt->pref_cache =
+		HINIC_SQ_CTXT_PREF_SET(WQ_PREFETCH_MIN, CACHE_MIN) |
+		HINIC_SQ_CTXT_PREF_SET(WQ_PREFETCH_MAX, CACHE_MAX) |
+		HINIC_SQ_CTXT_PREF_SET(WQ_PREFETCH_THRESHOLD, CACHE_THRESHOLD);
+
+	sq_ctxt->pref_wrapped = 1;
+
+	sq_ctxt->pref_wq_hi_pfn_ci =
+		HINIC_SQ_CTXT_PREF_SET(ci_start, CI) |
+		HINIC_SQ_CTXT_PREF_SET(wq_page_pfn_hi, WQ_HI_PFN);
+
+	sq_ctxt->pref_wq_lo_pfn = wq_page_pfn_lo;
+
+	sq_ctxt->wq_block_hi_pfn =
+		HINIC_SQ_CTXT_WQ_BLOCK_SET(wq_block_pfn_hi, HI_PFN);
+
+	sq_ctxt->wq_block_lo_pfn = wq_block_pfn_lo;
+
+	hinic_cpu_to_be32(sq_ctxt, sizeof(*sq_ctxt));
+}
+
+void hinic_rq_prepare_ctxt(struct hinic_rq_ctxt *rq_ctxt,
+			   struct hinic_rq *rq, u16 global_qid)
+{
+	u32 wq_page_pfn_hi, wq_page_pfn_lo, wq_block_pfn_hi, wq_block_pfn_lo;
+	u64 wq_page_addr, wq_page_pfn, wq_block_pfn;
+	u16 pi_start, ci_start;
+	struct hinic_wq *wq;
+
+	wq = rq->wq;
+	ci_start = atomic_read(&wq->cons_idx);
+	pi_start = atomic_read(&wq->prod_idx);
+
+	/* Read the first page paddr from the WQ page paddr ptrs */
+	wq_page_addr = be64_to_cpu(*wq->block_vaddr);
+
+	wq_page_pfn = HINIC_WQ_PAGE_PFN(wq_page_addr);
+	wq_page_pfn_hi = upper_32_bits(wq_page_pfn);
+	wq_page_pfn_lo = lower_32_bits(wq_page_pfn);
+
+	wq_block_pfn = HINIC_WQ_BLOCK_PFN(wq->block_paddr);
+	wq_block_pfn_hi = upper_32_bits(wq_block_pfn);
+	wq_block_pfn_lo = lower_32_bits(wq_block_pfn);
+
+	rq_ctxt->ceq_attr = HINIC_RQ_CTXT_CEQ_ATTR_SET(0, EN) |
+			    HINIC_RQ_CTXT_CEQ_ATTR_SET(1, WRAPPED);
+
+	rq_ctxt->pi_intr_attr = HINIC_RQ_CTXT_PI_SET(pi_start, IDX) |
+				HINIC_RQ_CTXT_PI_SET(rq->msix_entry, INTR);
+
+	rq_ctxt->wq_hi_pfn_ci = HINIC_RQ_CTXT_WQ_PAGE_SET(wq_page_pfn_hi,
+							  HI_PFN) |
+				HINIC_RQ_CTXT_WQ_PAGE_SET(ci_start, CI);
+
+	rq_ctxt->wq_lo_pfn = wq_page_pfn_lo;
+
+	rq_ctxt->pref_cache =
+		HINIC_RQ_CTXT_PREF_SET(WQ_PREFETCH_MIN, CACHE_MIN) |
+		HINIC_RQ_CTXT_PREF_SET(WQ_PREFETCH_MAX, CACHE_MAX) |
+		HINIC_RQ_CTXT_PREF_SET(WQ_PREFETCH_THRESHOLD, CACHE_THRESHOLD);
+
+	rq_ctxt->pref_wrapped = 1;
+
+	rq_ctxt->pref_wq_hi_pfn_ci =
+		HINIC_RQ_CTXT_PREF_SET(wq_page_pfn_hi, WQ_HI_PFN) |
+		HINIC_RQ_CTXT_PREF_SET(ci_start, CI);
+
+	rq_ctxt->pref_wq_lo_pfn = wq_page_pfn_lo;
+
+	rq_ctxt->pi_paddr_hi = upper_32_bits(rq->pi_dma_addr);
+	rq_ctxt->pi_paddr_lo = lower_32_bits(rq->pi_dma_addr);
+
+	rq_ctxt->wq_block_hi_pfn =
+		HINIC_RQ_CTXT_WQ_BLOCK_SET(wq_block_pfn_hi, HI_PFN);
+
+	rq_ctxt->wq_block_lo_pfn = wq_block_pfn_lo;
+
+	hinic_cpu_to_be32(rq_ctxt, sizeof(*rq_ctxt));
+}
+
+/**
+ * alloc_sq_skb_arr - allocate sq array for saved skb
+ * @sq: HW Send Queue
+ *
+ * Return 0 - Success, negative - Failure
+ **/
+static int alloc_sq_skb_arr(struct hinic_sq *sq)
+{
+	struct hinic_wq *wq = sq->wq;
+	size_t skb_arr_size;
+
+	skb_arr_size = wq->q_depth * sizeof(*sq->saved_skb);
+	sq->saved_skb = vzalloc(skb_arr_size);
+	if (!sq->saved_skb)
+		return -ENOMEM;
+
+	return 0;
+}
+
+/**
+ * free_sq_skb_arr - free sq array for saved skb
+ * @sq: HW Send Queue
+ **/
+static void free_sq_skb_arr(struct hinic_sq *sq)
+{
+	vfree(sq->saved_skb);
+}
+
+/**
+ * alloc_rq_skb_arr - allocate rq array for saved skb
+ * @rq: HW Receive Queue
+ *
+ * Return 0 - Success, negative - Failure
+ **/
+static int alloc_rq_skb_arr(struct hinic_rq *rq)
+{
+	struct hinic_wq *wq = rq->wq;
+	size_t skb_arr_size;
+
+	skb_arr_size = wq->q_depth * sizeof(*rq->saved_skb);
+	rq->saved_skb = vzalloc(skb_arr_size);
+	if (!rq->saved_skb)
+		return -ENOMEM;
+
+	return 0;
+}
+
+/**
+ * free_rq_skb_arr - free rq array for saved skb
+ * @rq: HW Receive Queue
+ **/
+static void free_rq_skb_arr(struct hinic_rq *rq)
+{
+	vfree(rq->saved_skb);
+}
+
+/**
+ * hinic_init_sq - Initialize HW Send Queue
+ * @sq: HW Send Queue
+ * @hwif: HW Interface for accessing HW
+ * @wq: Work Queue for the data of the SQ
+ * @entry: msix entry for sq
+ * @ci_addr: address for reading the current HW consumer index
+ * @ci_dma_addr: dma address for reading the current HW consumer index
+ * @db_base: doorbell base address
+ *
+ * Return 0 - Success, negative - Failure
+ **/
+int hinic_init_sq(struct hinic_sq *sq, struct hinic_hwif *hwif,
+		  struct hinic_wq *wq, struct msix_entry *entry,
+		  void *ci_addr, dma_addr_t ci_dma_addr,
+		  void __iomem *db_base)
+{
+	sq->hwif = hwif;
+
+	sq->wq = wq;
+
+	sq->irq = entry->vector;
+	sq->msix_entry = entry->entry;
+
+	sq->hw_ci_addr = ci_addr;
+	sq->hw_ci_dma_addr = ci_dma_addr;
+
+	sq->db_base = db_base + SQ_DB_OFF;
+
+	return alloc_sq_skb_arr(sq);
+}
+
+/**
+ * hinic_clean_sq - Clean HW Send Queue's Resources
+ * @sq: Send Queue
+ **/
+void hinic_clean_sq(struct hinic_sq *sq)
+{
+	free_sq_skb_arr(sq);
+}
+
+/**
+ * alloc_rq_cqe - allocate rq completion queue elements
+ * @rq: HW Receive Queue
+ *
+ * Return 0 - Success, negative - Failure
+ **/
+static int alloc_rq_cqe(struct hinic_rq *rq)
+{
+	struct hinic_hwif *hwif = rq->hwif;
+	struct pci_dev *pdev = hwif->pdev;
+	size_t cqe_dma_size, cqe_size;
+	struct hinic_wq *wq = rq->wq;
+	int j, i;
+
+	cqe_size = wq->q_depth * sizeof(*rq->cqe);
+	rq->cqe = vzalloc(cqe_size);
+	if (!rq->cqe)
+		return -ENOMEM;
+
+	cqe_dma_size = wq->q_depth * sizeof(*rq->cqe_dma);
+	rq->cqe_dma = vzalloc(cqe_dma_size);
+	if (!rq->cqe_dma)
+		goto err_cqe_dma_arr_alloc;
+
+	for (i = 0; i < wq->q_depth; i++) {
+		rq->cqe[i] = dma_alloc_coherent(&pdev->dev,
+						sizeof(*rq->cqe[i]),
+						&rq->cqe_dma[i], GFP_KERNEL);
+		if (!rq->cqe[i])
+			goto err_cqe_alloc;
+	}
+
+	return 0;
+
+err_cqe_alloc:
+	for (j = 0; j < i; j++)
+		dma_free_coherent(&pdev->dev, sizeof(*rq->cqe[j]), rq->cqe[j],
+				  rq->cqe_dma[j]);
+
+	vfree(rq->cqe_dma);
+
+err_cqe_dma_arr_alloc:
+	vfree(rq->cqe);
+	return -ENOMEM;
+}
+
+/**
+ * free_rq_cqe - free rq completion queue elements
+ * @rq: HW Receive Queue
+ **/
+static void free_rq_cqe(struct hinic_rq *rq)
+{
+	struct hinic_hwif *hwif = rq->hwif;
+	struct pci_dev *pdev = hwif->pdev;
+	struct hinic_wq *wq = rq->wq;
+	int i;
+
+	for (i = 0; i < wq->q_depth; i++)
+		dma_free_coherent(&pdev->dev, sizeof(*rq->cqe[i]), rq->cqe[i],
+				  rq->cqe_dma[i]);
+
+	vfree(rq->cqe_dma);
+	vfree(rq->cqe);
+}
+
+/**
+ * hinic_init_rq - Initialize HW Receive Queue
+ * @rq: HW Receive Queue
+ * @hwif: HW Interface for accessing HW
+ * @wq: Work Queue for the data of the RQ
+ * @entry: msix entry for rq
+ *
+ * Return 0 - Success, negative - Failure
+ **/
+int hinic_init_rq(struct hinic_rq *rq, struct hinic_hwif *hwif,
+		  struct hinic_wq *wq, struct msix_entry *entry)
+{
+	struct pci_dev *pdev = hwif->pdev;
+	size_t pi_size;
+	int err;
+
+	rq->hwif = hwif;
+
+	rq->wq = wq;
+
+	rq->irq = entry->vector;
+	rq->msix_entry = entry->entry;
+
+	rq->buf_sz = HINIC_RX_BUF_SZ;
+
+	err = alloc_rq_skb_arr(rq);
+	if (err) {
+		dev_err(&pdev->dev, "Failed to allocate rq priv data\n");
+		return err;
+	}
+
+	err = alloc_rq_cqe(rq);
+	if (err) {
+		dev_err(&pdev->dev, "Failed to allocate rq cqe\n");
+		goto err_alloc_rq_cqe;
+	}
+
+	/* HW requirements: Must be at least 32 bit */
+	pi_size = ALIGN(sizeof(*rq->pi_virt_addr), sizeof(u32));
+	rq->pi_virt_addr = dma_alloc_coherent(&pdev->dev, pi_size,
+					      &rq->pi_dma_addr, GFP_KERNEL);
+	if (!rq->pi_virt_addr) {
+		err = -ENOMEM;
+		goto err_pi_virt;
+	}
+
+	return 0;
+
+err_pi_virt:
+	free_rq_cqe(rq);
+
+err_alloc_rq_cqe:
+	free_rq_skb_arr(rq);
+	return err;
+}
+
+/**
+ * hinic_clean_rq - Clean HW Receive Queue's Resources
+ * @rq: HW Receive Queue
+ **/
+void hinic_clean_rq(struct hinic_rq *rq)
+{
+	struct hinic_hwif *hwif = rq->hwif;
+	struct pci_dev *pdev = hwif->pdev;
+	size_t pi_size;
+
+	pi_size = ALIGN(sizeof(*rq->pi_virt_addr), sizeof(u32));
+	dma_free_coherent(&pdev->dev, pi_size, rq->pi_virt_addr,
+			  rq->pi_dma_addr);
+
+	free_rq_cqe(rq);
+	free_rq_skb_arr(rq);
+}
+
+/**
+ * hinic_get_sq_free_wqebbs - return number of free wqebbs for use
+ * @sq: send queue
+ *
+ * Return number of free wqebbs
+ **/
+int hinic_get_sq_free_wqebbs(struct hinic_sq *sq)
+{
+	struct hinic_wq *wq = sq->wq;
+
+	return atomic_read(&wq->delta) - 1;
+}
+
+/**
+ * hinic_get_rq_free_wqebbs - return number of free wqebbs for use
+ * @rq: recv queue
+ *
+ * Return number of free wqebbs
+ **/
+int hinic_get_rq_free_wqebbs(struct hinic_rq *rq)
+{
+	struct hinic_wq *wq = rq->wq;
+
+	return atomic_read(&wq->delta) - 1;
+}
+
+static void sq_prepare_ctrl(struct hinic_sq_ctrl *ctrl, int nr_descs)
+{
+	u32 ctrl_size, task_size, bufdesc_size;
+
+	ctrl_size = SIZE_8BYTES(sizeof(struct hinic_sq_ctrl));
+	task_size = SIZE_8BYTES(sizeof(struct hinic_sq_task));
+	bufdesc_size = nr_descs * sizeof(struct hinic_sq_bufdesc);
+	bufdesc_size = SIZE_8BYTES(bufdesc_size);
+
+	ctrl->ctrl_info = HINIC_SQ_CTRL_SET(bufdesc_size, BUFDESC_SECT_LEN) |
+			  HINIC_SQ_CTRL_SET(task_size, TASKSECT_LEN)        |
+			  HINIC_SQ_CTRL_SET(SQ_NORMAL_WQE, DATA_FORMAT)     |
+			  HINIC_SQ_CTRL_SET(ctrl_size, LEN);
+
+	ctrl->queue_info = HINIC_SQ_CTRL_SET(HINIC_MSS_DEFAULT,
+					     QUEUE_INFO_MSS) |
+			   HINIC_SQ_CTRL_SET(1, QUEUE_INFO_UC);
+}
+
+static void sq_prepare_task(struct hinic_sq_task *task)
+{
+	task->pkt_info0 = 0;
+	task->pkt_info1 = 0;
+	task->pkt_info2 = 0;
+
+	task->ufo_v6_identify = 0;
+
+	task->pkt_info4 = HINIC_SQ_TASK_INFO4_SET(HINIC_L2TYPE_ETH, L2TYPE);
+
+	task->zero_pad = 0;
+}
+
+void hinic_task_set_l2hdr(struct hinic_sq_task *task, u32 len)
+{
+	task->pkt_info0 |= HINIC_SQ_TASK_INFO0_SET(len, L2HDR_LEN);
+}
+
+void hinic_task_set_outter_l3(struct hinic_sq_task *task,
+			      enum hinic_l3_offload_type l3_type,
+			      u32 network_len)
+{
+	task->pkt_info2 |= HINIC_SQ_TASK_INFO2_SET(l3_type, OUTER_L3TYPE) |
+			   HINIC_SQ_TASK_INFO2_SET(network_len, OUTER_L3LEN);
+}
+
+void hinic_task_set_inner_l3(struct hinic_sq_task *task,
+			     enum hinic_l3_offload_type l3_type,
+			     u32 network_len)
+{
+	task->pkt_info0 |= HINIC_SQ_TASK_INFO0_SET(l3_type, INNER_L3TYPE);
+	task->pkt_info1 |= HINIC_SQ_TASK_INFO1_SET(network_len, INNER_L3LEN);
+}
+
+void hinic_task_set_tunnel_l4(struct hinic_sq_task *task,
+			      enum hinic_l4_tunnel_type l4_type,
+			      u32 tunnel_len)
+{
+	task->pkt_info2 |= HINIC_SQ_TASK_INFO2_SET(l4_type, TUNNEL_L4TYPE) |
+			   HINIC_SQ_TASK_INFO2_SET(tunnel_len, TUNNEL_L4LEN);
+}
+
+void hinic_set_cs_inner_l4(struct hinic_sq_task *task, u32 *queue_info,
+			   enum hinic_l4_offload_type l4_offload,
+			   u32 l4_len, u32 offset)
+{
+	u32 tcp_udp_cs = 0, sctp = 0;
+	u32 mss = HINIC_MSS_DEFAULT;
+
+	if (l4_offload == TCP_OFFLOAD_ENABLE ||
+	    l4_offload == UDP_OFFLOAD_ENABLE)
+		tcp_udp_cs = 1;
+	else if (l4_offload == SCTP_OFFLOAD_ENABLE)
+		sctp = 1;
+
+	task->pkt_info0 |= HINIC_SQ_TASK_INFO0_SET(l4_offload, L4_OFFLOAD);
+	task->pkt_info1 |= HINIC_SQ_TASK_INFO1_SET(l4_len, INNER_L4LEN);
+
+	*queue_info |= HINIC_SQ_CTRL_SET(offset, QUEUE_INFO_PLDOFF) |
+		       HINIC_SQ_CTRL_SET(tcp_udp_cs, QUEUE_INFO_TCPUDP_CS) |
+		       HINIC_SQ_CTRL_SET(sctp, QUEUE_INFO_SCTP);
+
+	*queue_info = HINIC_SQ_CTRL_CLEAR(*queue_info, QUEUE_INFO_MSS);
+	*queue_info |= HINIC_SQ_CTRL_SET(mss, QUEUE_INFO_MSS);
+}
+
+void hinic_set_tso_inner_l4(struct hinic_sq_task *task, u32 *queue_info,
+			    enum hinic_l4_offload_type l4_offload,
+			    u32 l4_len, u32 offset, u32 ip_ident, u32 mss)
+{
+	u32 tso = 0, ufo = 0;
+
+	if (l4_offload == TCP_OFFLOAD_ENABLE)
+		tso = 1;
+	else if (l4_offload == UDP_OFFLOAD_ENABLE)
+		ufo = 1;
+
+	task->ufo_v6_identify = ip_ident;
+
+	task->pkt_info0 |= HINIC_SQ_TASK_INFO0_SET(l4_offload, L4_OFFLOAD);
+	task->pkt_info0 |= HINIC_SQ_TASK_INFO0_SET(tso || ufo, TSO_FLAG);
+	task->pkt_info1 |= HINIC_SQ_TASK_INFO1_SET(l4_len, INNER_L4LEN);
+
+	*queue_info |= HINIC_SQ_CTRL_SET(offset, QUEUE_INFO_PLDOFF) |
+		       HINIC_SQ_CTRL_SET(tso, QUEUE_INFO_TSO) |
+		       HINIC_SQ_CTRL_SET(ufo, QUEUE_INFO_UFO) |
+		       HINIC_SQ_CTRL_SET(!!l4_offload, QUEUE_INFO_TCPUDP_CS);
+
+	/* set MSS value */
+	*queue_info = HINIC_SQ_CTRL_CLEAR(*queue_info, QUEUE_INFO_MSS);
+	*queue_info |= HINIC_SQ_CTRL_SET(mss, QUEUE_INFO_MSS);
+}
+
+/**
+ * hinic_sq_prepare_wqe - prepare wqe before insert to the queue
+ * @sq: send queue
+ * @sq_wqe: wqe to prepare
+ * @sges: sges for use by the wqe for send for buf addresses
+ * @nr_sges: number of sges
+ **/
+void hinic_sq_prepare_wqe(struct hinic_sq *sq, struct hinic_sq_wqe *sq_wqe,
+			  struct hinic_sge *sges, int nr_sges)
+{
+	int i;
+
+	sq_prepare_ctrl(&sq_wqe->ctrl, nr_sges);
+
+	sq_prepare_task(&sq_wqe->task);
+
+	for (i = 0; i < nr_sges; i++)
+		sq_wqe->buf_descs[i].sge = sges[i];
+}
+
+/**
+ * sq_prepare_db - prepare doorbell to write
+ * @sq: send queue
+ * @prod_idx: pi value for the doorbell
+ * @cos: cos of the doorbell
+ *
+ * Return db value
+ **/
+static u32 sq_prepare_db(struct hinic_sq *sq, u16 prod_idx, unsigned int cos)
+{
+	struct hinic_qp *qp = container_of(sq, struct hinic_qp, sq);
+	u8 hi_prod_idx = SQ_DB_PI_HI(SQ_MASKED_IDX(sq, prod_idx));
+
+	/* Data should be written to HW in Big Endian Format */
+	return cpu_to_be32(HINIC_SQ_DB_INFO_SET(hi_prod_idx, PI_HI)     |
+			   HINIC_SQ_DB_INFO_SET(HINIC_DB_SQ_TYPE, TYPE) |
+			   HINIC_SQ_DB_INFO_SET(HINIC_DATA_PATH, PATH)  |
+			   HINIC_SQ_DB_INFO_SET(cos, COS)               |
+			   HINIC_SQ_DB_INFO_SET(qp->q_id, QID));
+}
+
+/**
+ * hinic_sq_write_db- write doorbell
+ * @sq: send queue
+ * @prod_idx: pi value for the doorbell
+ * @wqe_size: wqe size
+ * @cos: cos of the wqe
+ **/
+void hinic_sq_write_db(struct hinic_sq *sq, u16 prod_idx, unsigned int wqe_size,
+		       unsigned int cos)
+{
+	struct hinic_wq *wq = sq->wq;
+
+	/* increment prod_idx to the next */
+	prod_idx += ALIGN(wqe_size, wq->wqebb_size) / wq->wqebb_size;
+	prod_idx = SQ_MASKED_IDX(sq, prod_idx);
+
+	wmb();  /* Write all before the doorbell */
+
+	writel(sq_prepare_db(sq, prod_idx, cos), SQ_DB_ADDR(sq, prod_idx));
+}
+
+/**
+ * hinic_sq_get_wqe - get wqe ptr in the current pi and update the pi
+ * @sq: sq to get wqe from
+ * @wqe_size: wqe size
+ * @prod_idx: returned pi
+ *
+ * Return wqe pointer
+ **/
+struct hinic_sq_wqe *hinic_sq_get_wqe(struct hinic_sq *sq,
+				      unsigned int wqe_size, u16 *prod_idx)
+{
+	struct hinic_hw_wqe *hw_wqe = hinic_get_wqe(sq->wq, wqe_size,
+						    prod_idx);
+
+	if (IS_ERR(hw_wqe))
+		return NULL;
+
+	return &hw_wqe->sq_wqe;
+}
+
+/**
+ * hinic_sq_return_wqe - return the wqe to the sq
+ * @sq: send queue
+ * @wqe_size: the size of the wqe
+ **/
+void hinic_sq_return_wqe(struct hinic_sq *sq, unsigned int wqe_size)
+{
+	hinic_return_wqe(sq->wq, wqe_size);
+}
+
+/**
+ * hinic_sq_write_wqe - write the wqe to the sq
+ * @sq: send queue
+ * @prod_idx: pi of the wqe
+ * @sq_wqe: the wqe to write
+ * @skb: skb to save
+ * @wqe_size: the size of the wqe
+ **/
+void hinic_sq_write_wqe(struct hinic_sq *sq, u16 prod_idx,
+			struct hinic_sq_wqe *sq_wqe,
+			struct sk_buff *skb, unsigned int wqe_size)
+{
+	struct hinic_hw_wqe *hw_wqe = (struct hinic_hw_wqe *)sq_wqe;
+
+	sq->saved_skb[prod_idx] = skb;
+
+	/* The data in the HW should be in Big Endian Format */
+	hinic_cpu_to_be32(sq_wqe, wqe_size);
+
+	hinic_write_wqe(sq->wq, hw_wqe, wqe_size);
+}
+
+/**
+ * hinic_sq_read_wqebb - read wqe ptr in the current ci and update the ci, the
+ * wqe only have one wqebb
+ * @sq: send queue
+ * @skb: return skb that was saved
+ * @wqe_size: the wqe size ptr
+ * @cons_idx: consumer index of the wqe
+ *
+ * Return wqe in ci position
+ **/
+struct hinic_sq_wqe *hinic_sq_read_wqebb(struct hinic_sq *sq,
+					 struct sk_buff **skb,
+					 unsigned int *wqe_size, u16 *cons_idx)
+{
+	struct hinic_hw_wqe *hw_wqe;
+	struct hinic_sq_wqe *sq_wqe;
+	struct hinic_sq_ctrl *ctrl;
+	unsigned int buf_sect_len;
+	u32 ctrl_info;
+
+	/* read the ctrl section for getting wqe size */
+	hw_wqe = hinic_read_wqe(sq->wq, sizeof(*ctrl), cons_idx);
+	if (IS_ERR(hw_wqe))
+		return NULL;
+
+	*skb = sq->saved_skb[*cons_idx];
+
+	sq_wqe = &hw_wqe->sq_wqe;
+	ctrl = &sq_wqe->ctrl;
+	ctrl_info = be32_to_cpu(ctrl->ctrl_info);
+	buf_sect_len = HINIC_SQ_CTRL_GET(ctrl_info, BUFDESC_SECT_LEN);
+
+	*wqe_size = sizeof(*ctrl) + sizeof(sq_wqe->task);
+	*wqe_size += SECT_SIZE_FROM_8BYTES(buf_sect_len);
+	*wqe_size = ALIGN(*wqe_size, sq->wq->wqebb_size);
+
+	return &hw_wqe->sq_wqe;
+}
+
+/**
+ * hinic_sq_read_wqe - read wqe ptr in the current ci and update the ci
+ * @sq: send queue
+ * @skb: return skb that was saved
+ * @wqe_size: the size of the wqe
+ * @cons_idx: consumer index of the wqe
+ *
+ * Return wqe in ci position
+ **/
+struct hinic_sq_wqe *hinic_sq_read_wqe(struct hinic_sq *sq,
+				       struct sk_buff **skb,
+				       unsigned int wqe_size, u16 *cons_idx)
+{
+	struct hinic_hw_wqe *hw_wqe;
+
+	hw_wqe = hinic_read_wqe(sq->wq, wqe_size, cons_idx);
+	*skb = sq->saved_skb[*cons_idx];
+
+	return &hw_wqe->sq_wqe;
+}
+
+/**
+ * hinic_sq_put_wqe - release the ci for new wqes
+ * @sq: send queue
+ * @wqe_size: the size of the wqe
+ **/
+void hinic_sq_put_wqe(struct hinic_sq *sq, unsigned int wqe_size)
+{
+	hinic_put_wqe(sq->wq, wqe_size);
+}
+
+/**
+ * hinic_sq_get_sges - get sges from the wqe
+ * @sq_wqe: wqe to get the sges from its buffer addresses
+ * @sges: returned sges
+ * @nr_sges: number sges to return
+ **/
+void hinic_sq_get_sges(struct hinic_sq_wqe *sq_wqe, struct hinic_sge *sges,
+		       int nr_sges)
+{
+	int i;
+
+	for (i = 0; i < nr_sges && i < HINIC_MAX_SQ_BUFDESCS; i++) {
+		sges[i] = sq_wqe->buf_descs[i].sge;
+		hinic_be32_to_cpu(&sges[i], sizeof(sges[i]));
+	}
+}
+
+/**
+ * hinic_rq_get_wqe - get wqe ptr in the current pi and update the pi
+ * @rq: rq to get wqe from
+ * @wqe_size: wqe size
+ * @prod_idx: returned pi
+ *
+ * Return wqe pointer
+ **/
+struct hinic_rq_wqe *hinic_rq_get_wqe(struct hinic_rq *rq,
+				      unsigned int wqe_size, u16 *prod_idx)
+{
+	struct hinic_hw_wqe *hw_wqe = hinic_get_wqe(rq->wq, wqe_size,
+						    prod_idx);
+
+	if (IS_ERR(hw_wqe))
+		return NULL;
+
+	return &hw_wqe->rq_wqe;
+}
+
+/**
+ * hinic_rq_write_wqe - write the wqe to the rq
+ * @rq: recv queue
+ * @prod_idx: pi of the wqe
+ * @rq_wqe: the wqe to write
+ * @skb: skb to save
+ **/
+void hinic_rq_write_wqe(struct hinic_rq *rq, u16 prod_idx,
+			struct hinic_rq_wqe *rq_wqe, struct sk_buff *skb)
+{
+	struct hinic_hw_wqe *hw_wqe = (struct hinic_hw_wqe *)rq_wqe;
+
+	rq->saved_skb[prod_idx] = skb;
+
+	/* The data in the HW should be in Big Endian Format */
+	hinic_cpu_to_be32(rq_wqe, sizeof(*rq_wqe));
+
+	hinic_write_wqe(rq->wq, hw_wqe, sizeof(*rq_wqe));
+}
+
+/**
+ * hinic_rq_read_wqe - read wqe ptr in the current ci and update the ci
+ * @rq: recv queue
+ * @wqe_size: the size of the wqe
+ * @skb: return saved skb
+ * @cons_idx: consumer index of the wqe
+ *
+ * Return wqe in ci position
+ **/
+struct hinic_rq_wqe *hinic_rq_read_wqe(struct hinic_rq *rq,
+				       unsigned int wqe_size,
+				       struct sk_buff **skb, u16 *cons_idx)
+{
+	struct hinic_hw_wqe *hw_wqe;
+	struct hinic_rq_cqe *cqe;
+	int rx_done;
+	u32 status;
+
+	hw_wqe = hinic_read_wqe(rq->wq, wqe_size, cons_idx);
+	if (IS_ERR(hw_wqe))
+		return NULL;
+
+	cqe = rq->cqe[*cons_idx];
+
+	status = be32_to_cpu(cqe->status);
+
+	rx_done = HINIC_RQ_CQE_STATUS_GET(status, RXDONE);
+	if (!rx_done)
+		return NULL;
+
+	*skb = rq->saved_skb[*cons_idx];
+
+	return &hw_wqe->rq_wqe;
+}
+
+/**
+ * hinic_rq_read_next_wqe - increment ci and read the wqe in ci position
+ * @rq: recv queue
+ * @wqe_size: the size of the wqe
+ * @skb: return saved skb
+ * @cons_idx: consumer index in the wq
+ *
+ * Return wqe in incremented ci position
+ **/
+struct hinic_rq_wqe *hinic_rq_read_next_wqe(struct hinic_rq *rq,
+					    unsigned int wqe_size,
+					    struct sk_buff **skb,
+					    u16 *cons_idx)
+{
+	struct hinic_wq *wq = rq->wq;
+	struct hinic_hw_wqe *hw_wqe;
+	unsigned int num_wqebbs;
+
+	wqe_size = ALIGN(wqe_size, wq->wqebb_size);
+	num_wqebbs = wqe_size / wq->wqebb_size;
+
+	*cons_idx = RQ_MASKED_IDX(rq, *cons_idx + num_wqebbs);
+
+	*skb = rq->saved_skb[*cons_idx];
+
+	hw_wqe = hinic_read_wqe_direct(wq, *cons_idx);
+
+	return &hw_wqe->rq_wqe;
+}
+
+/**
+ * hinic_rq_put_wqe - release the ci for new wqes
+ * @rq: recv queue
+ * @cons_idx: consumer index of the wqe
+ * @wqe_size: the size of the wqe
+ **/
+void hinic_rq_put_wqe(struct hinic_rq *rq, u16 cons_idx,
+		      unsigned int wqe_size)
+{
+	struct hinic_rq_cqe *cqe = rq->cqe[cons_idx];
+	u32 status = be32_to_cpu(cqe->status);
+
+	status = HINIC_RQ_CQE_STATUS_CLEAR(status, RXDONE);
+
+	/* Rx WQE size is 1 WQEBB, no wq shadow*/
+	cqe->status = cpu_to_be32(status);
+
+	wmb();          /* clear done flag */
+
+	hinic_put_wqe(rq->wq, wqe_size);
+}
+
+/**
+ * hinic_rq_get_sge - get sge from the wqe
+ * @rq: recv queue
+ * @rq_wqe: wqe to get the sge from its buf address
+ * @cons_idx: consumer index
+ * @sge: returned sge
+ **/
+void hinic_rq_get_sge(struct hinic_rq *rq, struct hinic_rq_wqe *rq_wqe,
+		      u16 cons_idx, struct hinic_sge *sge)
+{
+	struct hinic_rq_cqe *cqe = rq->cqe[cons_idx];
+	u32 len = be32_to_cpu(cqe->len);
+
+	sge->hi_addr = be32_to_cpu(rq_wqe->buf_desc.hi_addr);
+	sge->lo_addr = be32_to_cpu(rq_wqe->buf_desc.lo_addr);
+	sge->len = HINIC_RQ_CQE_SGE_GET(len, LEN);
+}
+
+/**
+ * hinic_rq_prepare_wqe - prepare wqe before insert to the queue
+ * @rq: recv queue
+ * @prod_idx: pi value
+ * @rq_wqe: the wqe
+ * @sge: sge for use by the wqe for recv buf address
+ **/
+void hinic_rq_prepare_wqe(struct hinic_rq *rq, u16 prod_idx,
+			  struct hinic_rq_wqe *rq_wqe, struct hinic_sge *sge)
+{
+	struct hinic_rq_cqe_sect *cqe_sect = &rq_wqe->cqe_sect;
+	struct hinic_rq_bufdesc *buf_desc = &rq_wqe->buf_desc;
+	struct hinic_rq_cqe *cqe = rq->cqe[prod_idx];
+	struct hinic_rq_ctrl *ctrl = &rq_wqe->ctrl;
+	dma_addr_t cqe_dma = rq->cqe_dma[prod_idx];
+
+	ctrl->ctrl_info =
+		HINIC_RQ_CTRL_SET(SIZE_8BYTES(sizeof(*ctrl)), LEN) |
+		HINIC_RQ_CTRL_SET(SIZE_8BYTES(sizeof(*cqe_sect)),
+				  COMPLETE_LEN)                    |
+		HINIC_RQ_CTRL_SET(SIZE_8BYTES(sizeof(*buf_desc)),
+				  BUFDESC_SECT_LEN)                |
+		HINIC_RQ_CTRL_SET(RQ_COMPLETE_SGE, COMPLETE_FORMAT);
+
+	hinic_set_sge(&cqe_sect->sge, cqe_dma, sizeof(*cqe));
+
+	buf_desc->hi_addr = sge->hi_addr;
+	buf_desc->lo_addr = sge->lo_addr;
+}
+
+/**
+ * hinic_rq_update - update pi of the rq
+ * @rq: recv queue
+ * @prod_idx: pi value
+ **/
+void hinic_rq_update(struct hinic_rq *rq, u16 prod_idx)
+{
+	*rq->pi_virt_addr = cpu_to_be16(RQ_MASKED_IDX(rq, prod_idx + 1));
+}
diff --git a/drivers/net/ethernet/huawei/hinic/hinic_hw_qp.h b/drivers/net/ethernet/huawei/hinic/hinic_hw_qp.h
new file mode 100644
index 000000000000..178dcc874370
--- /dev/null
+++ b/drivers/net/ethernet/huawei/hinic/hinic_hw_qp.h
@@ -0,0 +1,232 @@
+/* SPDX-License-Identifier: GPL-2.0-only */
+/*
+ * Huawei HiNIC PCI Express Linux driver
+ * Copyright(c) 2017 Huawei Technologies Co., Ltd
+ */
+
+#ifndef HINIC_HW_QP_H
+#define HINIC_HW_QP_H
+
+#include <linux/kernel.h>
+#include <linux/types.h>
+#include <linux/sizes.h>
+#include <linux/pci.h>
+#include <linux/skbuff.h>
+
+#include "hinic_common.h"
+#include "hinic_hw_if.h"
+#include "hinic_hw_wqe.h"
+#include "hinic_hw_wq.h"
+#include "hinic_hw_qp_ctxt.h"
+
+#define HINIC_SQ_DB_INFO_PI_HI_SHIFT            0
+#define HINIC_SQ_DB_INFO_QID_SHIFT              8
+#define HINIC_SQ_DB_INFO_PATH_SHIFT             23
+#define HINIC_SQ_DB_INFO_COS_SHIFT              24
+#define HINIC_SQ_DB_INFO_TYPE_SHIFT             27
+
+#define HINIC_SQ_DB_INFO_PI_HI_MASK             0xFF
+#define HINIC_SQ_DB_INFO_QID_MASK               0x3FF
+#define HINIC_SQ_DB_INFO_PATH_MASK              0x1
+#define HINIC_SQ_DB_INFO_COS_MASK               0x7
+#define HINIC_SQ_DB_INFO_TYPE_MASK              0x1F
+
+#define HINIC_SQ_DB_INFO_SET(val, member)       \
+		(((u32)(val) & HINIC_SQ_DB_INFO_##member##_MASK) \
+		 << HINIC_SQ_DB_INFO_##member##_SHIFT)
+
+#define HINIC_SQ_WQEBB_SIZE                     64
+#define HINIC_RQ_WQEBB_SIZE                     32
+
+#define HINIC_SQ_PAGE_SIZE                      SZ_256K
+#define HINIC_RQ_PAGE_SIZE                      SZ_256K
+
+#define HINIC_SQ_DEPTH                          SZ_4K
+#define HINIC_RQ_DEPTH                          SZ_4K
+
+#define HINIC_MAX_QUEUE_DEPTH			SZ_4K
+#define HINIC_MIN_QUEUE_DEPTH			128
+
+/* In any change to HINIC_RX_BUF_SZ, HINIC_RX_BUF_SZ_IDX must be changed */
+#define HINIC_RX_BUF_SZ                         2048
+#define HINIC_RX_BUF_SZ_IDX			HINIC_RX_BUF_SZ_2048_IDX
+
+#define HINIC_MIN_TX_WQE_SIZE(wq)               \
+		ALIGN(HINIC_SQ_WQE_SIZE(1), (wq)->wqebb_size)
+
+#define HINIC_MIN_TX_NUM_WQEBBS(sq)             \
+		(HINIC_MIN_TX_WQE_SIZE((sq)->wq) / (sq)->wq->wqebb_size)
+
+enum hinic_rx_buf_sz_idx {
+	HINIC_RX_BUF_SZ_32_IDX,
+	HINIC_RX_BUF_SZ_64_IDX,
+	HINIC_RX_BUF_SZ_96_IDX,
+	HINIC_RX_BUF_SZ_128_IDX,
+	HINIC_RX_BUF_SZ_192_IDX,
+	HINIC_RX_BUF_SZ_256_IDX,
+	HINIC_RX_BUF_SZ_384_IDX,
+	HINIC_RX_BUF_SZ_512_IDX,
+	HINIC_RX_BUF_SZ_768_IDX,
+	HINIC_RX_BUF_SZ_1024_IDX,
+	HINIC_RX_BUF_SZ_1536_IDX,
+	HINIC_RX_BUF_SZ_2048_IDX,
+	HINIC_RX_BUF_SZ_3072_IDX,
+	HINIC_RX_BUF_SZ_4096_IDX,
+	HINIC_RX_BUF_SZ_8192_IDX,
+	HINIC_RX_BUF_SZ_16384_IDX,
+};
+
+struct hinic_sq {
+	struct hinic_hwif       *hwif;
+
+	struct hinic_wq         *wq;
+
+	u16			qid;
+
+	u32                     irq;
+	u16                     msix_entry;
+
+	void                    *hw_ci_addr;
+	dma_addr_t              hw_ci_dma_addr;
+
+	void __iomem            *db_base;
+
+	struct sk_buff          **saved_skb;
+	struct hinic_debug_priv	*dbg;
+};
+
+struct hinic_rq {
+	struct hinic_hwif       *hwif;
+
+	struct hinic_wq         *wq;
+
+	u16			qid;
+
+	struct cpumask		affinity_mask;
+	u32                     irq;
+	u16                     msix_entry;
+
+	size_t                  buf_sz;
+
+	struct sk_buff          **saved_skb;
+
+	struct hinic_rq_cqe     **cqe;
+	dma_addr_t              *cqe_dma;
+
+	u16                     *pi_virt_addr;
+	dma_addr_t              pi_dma_addr;
+	struct hinic_debug_priv	*dbg;
+};
+
+struct hinic_qp {
+	struct hinic_sq         sq;
+	struct hinic_rq         rq;
+
+	u16     q_id;
+};
+
+void hinic_qp_prepare_header(struct hinic_qp_ctxt_header *qp_ctxt_hdr,
+			     enum hinic_qp_ctxt_type ctxt_type,
+			     u16 num_queues, u16 max_queues);
+
+void hinic_sq_prepare_ctxt(struct hinic_sq_ctxt *sq_ctxt,
+			   struct hinic_sq *sq, u16 global_qid);
+
+void hinic_rq_prepare_ctxt(struct hinic_rq_ctxt *rq_ctxt,
+			   struct hinic_rq *rq, u16 global_qid);
+
+int hinic_init_sq(struct hinic_sq *sq, struct hinic_hwif *hwif,
+		  struct hinic_wq *wq, struct msix_entry *entry, void *ci_addr,
+		  dma_addr_t ci_dma_addr, void __iomem *db_base);
+
+void hinic_clean_sq(struct hinic_sq *sq);
+
+int hinic_init_rq(struct hinic_rq *rq, struct hinic_hwif *hwif,
+		  struct hinic_wq *wq, struct msix_entry *entry);
+
+void hinic_clean_rq(struct hinic_rq *rq);
+
+int hinic_get_sq_free_wqebbs(struct hinic_sq *sq);
+
+int hinic_get_rq_free_wqebbs(struct hinic_rq *rq);
+
+void hinic_task_set_l2hdr(struct hinic_sq_task *task, u32 len);
+
+void hinic_task_set_outter_l3(struct hinic_sq_task *task,
+			      enum hinic_l3_offload_type l3_type,
+			      u32 network_len);
+
+void hinic_task_set_inner_l3(struct hinic_sq_task *task,
+			     enum hinic_l3_offload_type l3_type,
+			     u32 network_len);
+
+void hinic_task_set_tunnel_l4(struct hinic_sq_task *task,
+			      enum hinic_l4_tunnel_type l4_type,
+			      u32 tunnel_len);
+
+void hinic_set_cs_inner_l4(struct hinic_sq_task *task,
+			   u32 *queue_info,
+			   enum hinic_l4_offload_type l4_offload,
+			   u32 l4_len, u32 offset);
+
+void hinic_set_tso_inner_l4(struct hinic_sq_task *task,
+			    u32 *queue_info,
+			    enum hinic_l4_offload_type l4_offload,
+			    u32 l4_len,
+			    u32 offset, u32 ip_ident, u32 mss);
+
+void hinic_sq_prepare_wqe(struct hinic_sq *sq, struct hinic_sq_wqe *wqe,
+			  struct hinic_sge *sges, int nr_sges);
+
+void hinic_sq_write_db(struct hinic_sq *sq, u16 prod_idx, unsigned int wqe_size,
+		       unsigned int cos);
+
+struct hinic_sq_wqe *hinic_sq_get_wqe(struct hinic_sq *sq,
+				      unsigned int wqe_size, u16 *prod_idx);
+
+void hinic_sq_return_wqe(struct hinic_sq *sq, unsigned int wqe_size);
+
+void hinic_sq_write_wqe(struct hinic_sq *sq, u16 prod_idx,
+			struct hinic_sq_wqe *wqe, struct sk_buff *skb,
+			unsigned int wqe_size);
+
+struct hinic_sq_wqe *hinic_sq_read_wqe(struct hinic_sq *sq,
+				       struct sk_buff **skb,
+				       unsigned int wqe_size, u16 *cons_idx);
+
+struct hinic_sq_wqe *hinic_sq_read_wqebb(struct hinic_sq *sq,
+					 struct sk_buff **skb,
+					 unsigned int *wqe_size, u16 *cons_idx);
+
+void hinic_sq_put_wqe(struct hinic_sq *sq, unsigned int wqe_size);
+
+void hinic_sq_get_sges(struct hinic_sq_wqe *wqe, struct hinic_sge *sges,
+		       int nr_sges);
+
+struct hinic_rq_wqe *hinic_rq_get_wqe(struct hinic_rq *rq,
+				      unsigned int wqe_size, u16 *prod_idx);
+
+void hinic_rq_write_wqe(struct hinic_rq *rq, u16 prod_idx,
+			struct hinic_rq_wqe *wqe, struct sk_buff *skb);
+
+struct hinic_rq_wqe *hinic_rq_read_wqe(struct hinic_rq *rq,
+				       unsigned int wqe_size,
+				       struct sk_buff **skb, u16 *cons_idx);
+
+struct hinic_rq_wqe *hinic_rq_read_next_wqe(struct hinic_rq *rq,
+					    unsigned int wqe_size,
+					    struct sk_buff **skb,
+					    u16 *cons_idx);
+
+void hinic_rq_put_wqe(struct hinic_rq *rq, u16 cons_idx,
+		      unsigned int wqe_size);
+
+void hinic_rq_get_sge(struct hinic_rq *rq, struct hinic_rq_wqe *wqe,
+		      u16 cons_idx, struct hinic_sge *sge);
+
+void hinic_rq_prepare_wqe(struct hinic_rq *rq, u16 prod_idx,
+			  struct hinic_rq_wqe *wqe, struct hinic_sge *sge);
+
+void hinic_rq_update(struct hinic_rq *rq, u16 prod_idx);
+
+#endif
diff --git a/drivers/net/ethernet/huawei/hinic/hinic_hw_qp_ctxt.h b/drivers/net/ethernet/huawei/hinic/hinic_hw_qp_ctxt.h
new file mode 100644
index 000000000000..00900a6640ad
--- /dev/null
+++ b/drivers/net/ethernet/huawei/hinic/hinic_hw_qp_ctxt.h
@@ -0,0 +1,210 @@
+/* SPDX-License-Identifier: GPL-2.0-only */
+/*
+ * Huawei HiNIC PCI Express Linux driver
+ * Copyright(c) 2017 Huawei Technologies Co., Ltd
+ */
+
+#ifndef HINIC_HW_QP_CTXT_H
+#define HINIC_HW_QP_CTXT_H
+
+#include <linux/types.h>
+
+#include "hinic_hw_cmdq.h"
+
+#define HINIC_SQ_CTXT_CEQ_ATTR_GLOBAL_SQ_ID_SHIFT       13
+#define HINIC_SQ_CTXT_CEQ_ATTR_EN_SHIFT                 23
+
+#define HINIC_SQ_CTXT_CEQ_ATTR_GLOBAL_SQ_ID_MASK        0x3FF
+#define HINIC_SQ_CTXT_CEQ_ATTR_EN_MASK                  0x1
+
+#define HINIC_SQ_CTXT_CEQ_ATTR_SET(val, member)         \
+	(((u32)(val) & HINIC_SQ_CTXT_CEQ_ATTR_##member##_MASK) \
+	 << HINIC_SQ_CTXT_CEQ_ATTR_##member##_SHIFT)
+
+#define HINIC_SQ_CTXT_CI_IDX_SHIFT                      11
+#define HINIC_SQ_CTXT_CI_WRAPPED_SHIFT                  23
+
+#define HINIC_SQ_CTXT_CI_IDX_MASK                       0xFFF
+#define HINIC_SQ_CTXT_CI_WRAPPED_MASK                   0x1
+
+#define HINIC_SQ_CTXT_CI_SET(val, member)               \
+	(((u32)(val) & HINIC_SQ_CTXT_CI_##member##_MASK) \
+	 << HINIC_SQ_CTXT_CI_##member##_SHIFT)
+
+#define HINIC_SQ_CTXT_WQ_PAGE_HI_PFN_SHIFT              0
+#define HINIC_SQ_CTXT_WQ_PAGE_PI_SHIFT                  20
+
+#define HINIC_SQ_CTXT_WQ_PAGE_HI_PFN_MASK               0xFFFFF
+#define HINIC_SQ_CTXT_WQ_PAGE_PI_MASK                   0xFFF
+
+#define HINIC_SQ_CTXT_WQ_PAGE_SET(val, member)          \
+	(((u32)(val) & HINIC_SQ_CTXT_WQ_PAGE_##member##_MASK) \
+	 << HINIC_SQ_CTXT_WQ_PAGE_##member##_SHIFT)
+
+#define HINIC_SQ_CTXT_PREF_CACHE_THRESHOLD_SHIFT        0
+#define HINIC_SQ_CTXT_PREF_CACHE_MAX_SHIFT              14
+#define HINIC_SQ_CTXT_PREF_CACHE_MIN_SHIFT              25
+
+#define HINIC_SQ_CTXT_PREF_CACHE_THRESHOLD_MASK         0x3FFF
+#define HINIC_SQ_CTXT_PREF_CACHE_MAX_MASK               0x7FF
+#define HINIC_SQ_CTXT_PREF_CACHE_MIN_MASK               0x7F
+
+#define HINIC_SQ_CTXT_PREF_WQ_HI_PFN_SHIFT              0
+#define HINIC_SQ_CTXT_PREF_CI_SHIFT                     20
+
+#define HINIC_SQ_CTXT_PREF_WQ_HI_PFN_MASK               0xFFFFF
+#define HINIC_SQ_CTXT_PREF_CI_MASK                      0xFFF
+
+#define HINIC_SQ_CTXT_PREF_SET(val, member)             \
+	(((u32)(val) & HINIC_SQ_CTXT_PREF_##member##_MASK) \
+	 << HINIC_SQ_CTXT_PREF_##member##_SHIFT)
+
+#define HINIC_SQ_CTXT_WQ_BLOCK_HI_PFN_SHIFT             0
+
+#define HINIC_SQ_CTXT_WQ_BLOCK_HI_PFN_MASK              0x7FFFFF
+
+#define HINIC_SQ_CTXT_WQ_BLOCK_SET(val, member)         \
+	(((u32)(val) & HINIC_SQ_CTXT_WQ_BLOCK_##member##_MASK) \
+	 << HINIC_SQ_CTXT_WQ_BLOCK_##member##_SHIFT)
+
+#define HINIC_RQ_CTXT_CEQ_ATTR_EN_SHIFT                 0
+#define HINIC_RQ_CTXT_CEQ_ATTR_WRAPPED_SHIFT            1
+
+#define HINIC_RQ_CTXT_CEQ_ATTR_EN_MASK                  0x1
+#define HINIC_RQ_CTXT_CEQ_ATTR_WRAPPED_MASK             0x1
+
+#define HINIC_RQ_CTXT_CEQ_ATTR_SET(val, member)         \
+	(((u32)(val) & HINIC_RQ_CTXT_CEQ_ATTR_##member##_MASK) \
+	 << HINIC_RQ_CTXT_CEQ_ATTR_##member##_SHIFT)
+
+#define HINIC_RQ_CTXT_PI_IDX_SHIFT                      0
+#define HINIC_RQ_CTXT_PI_INTR_SHIFT                     22
+
+#define HINIC_RQ_CTXT_PI_IDX_MASK                       0xFFF
+#define HINIC_RQ_CTXT_PI_INTR_MASK                      0x3FF
+
+#define HINIC_RQ_CTXT_PI_SET(val, member)               \
+	(((u32)(val) & HINIC_RQ_CTXT_PI_##member##_MASK) << \
+	 HINIC_RQ_CTXT_PI_##member##_SHIFT)
+
+#define HINIC_RQ_CTXT_WQ_PAGE_HI_PFN_SHIFT              0
+#define HINIC_RQ_CTXT_WQ_PAGE_CI_SHIFT                  20
+
+#define HINIC_RQ_CTXT_WQ_PAGE_HI_PFN_MASK               0xFFFFF
+#define HINIC_RQ_CTXT_WQ_PAGE_CI_MASK                   0xFFF
+
+#define HINIC_RQ_CTXT_WQ_PAGE_SET(val, member)          \
+	(((u32)(val) & HINIC_RQ_CTXT_WQ_PAGE_##member##_MASK) << \
+	 HINIC_RQ_CTXT_WQ_PAGE_##member##_SHIFT)
+
+#define HINIC_RQ_CTXT_PREF_CACHE_THRESHOLD_SHIFT        0
+#define HINIC_RQ_CTXT_PREF_CACHE_MAX_SHIFT              14
+#define HINIC_RQ_CTXT_PREF_CACHE_MIN_SHIFT              25
+
+#define HINIC_RQ_CTXT_PREF_CACHE_THRESHOLD_MASK         0x3FFF
+#define HINIC_RQ_CTXT_PREF_CACHE_MAX_MASK               0x7FF
+#define HINIC_RQ_CTXT_PREF_CACHE_MIN_MASK               0x7F
+
+#define HINIC_RQ_CTXT_PREF_WQ_HI_PFN_SHIFT              0
+#define HINIC_RQ_CTXT_PREF_CI_SHIFT                     20
+
+#define HINIC_RQ_CTXT_PREF_WQ_HI_PFN_MASK               0xFFFFF
+#define HINIC_RQ_CTXT_PREF_CI_MASK                      0xFFF
+
+#define HINIC_RQ_CTXT_PREF_SET(val, member)             \
+	(((u32)(val) & HINIC_RQ_CTXT_PREF_##member##_MASK) << \
+	 HINIC_RQ_CTXT_PREF_##member##_SHIFT)
+
+#define HINIC_RQ_CTXT_WQ_BLOCK_HI_PFN_SHIFT             0
+
+#define HINIC_RQ_CTXT_WQ_BLOCK_HI_PFN_MASK              0x7FFFFF
+
+#define HINIC_RQ_CTXT_WQ_BLOCK_SET(val, member)         \
+	(((u32)(val) & HINIC_RQ_CTXT_WQ_BLOCK_##member##_MASK) << \
+	 HINIC_RQ_CTXT_WQ_BLOCK_##member##_SHIFT)
+
+#define HINIC_SQ_CTXT_SIZE(num_sqs) (sizeof(struct hinic_qp_ctxt_header) \
+				     + (num_sqs) * sizeof(struct hinic_sq_ctxt))
+
+#define HINIC_RQ_CTXT_SIZE(num_rqs) (sizeof(struct hinic_qp_ctxt_header) \
+				     + (num_rqs) * sizeof(struct hinic_rq_ctxt))
+
+#define HINIC_WQ_PAGE_PFN_SHIFT         12
+#define HINIC_WQ_BLOCK_PFN_SHIFT        9
+
+#define HINIC_WQ_PAGE_PFN(page_addr)    ((page_addr) >> HINIC_WQ_PAGE_PFN_SHIFT)
+#define HINIC_WQ_BLOCK_PFN(page_addr)   ((page_addr) >> \
+					 HINIC_WQ_BLOCK_PFN_SHIFT)
+
+#define HINIC_Q_CTXT_MAX                \
+		((HINIC_CMDQ_BUF_SIZE - sizeof(struct hinic_qp_ctxt_header)) \
+		 / sizeof(struct hinic_sq_ctxt))
+
+enum hinic_qp_ctxt_type {
+	HINIC_QP_CTXT_TYPE_SQ,
+	HINIC_QP_CTXT_TYPE_RQ
+};
+
+struct hinic_qp_ctxt_header {
+	u16     num_queues;
+	u16     queue_type;
+	u32     addr_offset;
+};
+
+struct hinic_sq_ctxt {
+	u32     ceq_attr;
+
+	u32     ci_wrapped;
+
+	u32     wq_hi_pfn_pi;
+	u32     wq_lo_pfn;
+
+	u32     pref_cache;
+	u32     pref_wrapped;
+	u32     pref_wq_hi_pfn_ci;
+	u32     pref_wq_lo_pfn;
+
+	u32     rsvd0;
+	u32     rsvd1;
+
+	u32     wq_block_hi_pfn;
+	u32     wq_block_lo_pfn;
+};
+
+struct hinic_rq_ctxt {
+	u32     ceq_attr;
+
+	u32     pi_intr_attr;
+
+	u32     wq_hi_pfn_ci;
+	u32     wq_lo_pfn;
+
+	u32     pref_cache;
+	u32     pref_wrapped;
+
+	u32     pref_wq_hi_pfn_ci;
+	u32     pref_wq_lo_pfn;
+
+	u32     pi_paddr_hi;
+	u32     pi_paddr_lo;
+
+	u32     wq_block_hi_pfn;
+	u32     wq_block_lo_pfn;
+};
+
+struct hinic_clean_queue_ctxt {
+	struct hinic_qp_ctxt_header	cmdq_hdr;
+	u32				ctxt_size;
+};
+
+struct hinic_sq_ctxt_block {
+	struct hinic_qp_ctxt_header hdr;
+	struct hinic_sq_ctxt sq_ctxt[HINIC_Q_CTXT_MAX];
+};
+
+struct hinic_rq_ctxt_block {
+	struct hinic_qp_ctxt_header hdr;
+	struct hinic_rq_ctxt rq_ctxt[HINIC_Q_CTXT_MAX];
+};
+
+#endif
diff --git a/drivers/net/ethernet/huawei/hinic/hinic_hw_wq.c b/drivers/net/ethernet/huawei/hinic/hinic_hw_wq.c
new file mode 100644
index 000000000000..e1a1735c00c1
--- /dev/null
+++ b/drivers/net/ethernet/huawei/hinic/hinic_hw_wq.c
@@ -0,0 +1,904 @@
+// SPDX-License-Identifier: GPL-2.0-only
+/*
+ * Huawei HiNIC PCI Express Linux driver
+ * Copyright(c) 2017 Huawei Technologies Co., Ltd
+ */
+
+#include <linux/kernel.h>
+#include <linux/types.h>
+#include <linux/pci.h>
+#include <linux/device.h>
+#include <linux/dma-mapping.h>
+#include <linux/slab.h>
+#include <linux/atomic.h>
+#include <linux/semaphore.h>
+#include <linux/errno.h>
+#include <linux/vmalloc.h>
+#include <linux/err.h>
+#include <asm/byteorder.h>
+
+#include "hinic_hw_if.h"
+#include "hinic_hw_wqe.h"
+#include "hinic_hw_wq.h"
+#include "hinic_hw_cmdq.h"
+
+#define WQS_BLOCKS_PER_PAGE             4
+
+#define WQ_BLOCK_SIZE                   4096
+#define WQS_PAGE_SIZE                   (WQS_BLOCKS_PER_PAGE * WQ_BLOCK_SIZE)
+
+#define WQS_MAX_NUM_BLOCKS              128
+#define WQS_FREE_BLOCKS_SIZE(wqs)       (WQS_MAX_NUM_BLOCKS * \
+					 sizeof((wqs)->free_blocks[0]))
+
+#define WQ_SIZE(wq)                     ((wq)->q_depth * (wq)->wqebb_size)
+
+#define WQ_PAGE_ADDR_SIZE               sizeof(u64)
+#define WQ_MAX_PAGES                    (WQ_BLOCK_SIZE / WQ_PAGE_ADDR_SIZE)
+
+#define CMDQ_BLOCK_SIZE                 512
+#define CMDQ_PAGE_SIZE                  4096
+
+#define CMDQ_WQ_MAX_PAGES               (CMDQ_BLOCK_SIZE / WQ_PAGE_ADDR_SIZE)
+
+#define WQ_BASE_VADDR(wqs, wq)          \
+			((void *)((wqs)->page_vaddr[(wq)->page_idx]) \
+				+ (wq)->block_idx * WQ_BLOCK_SIZE)
+
+#define WQ_BASE_PADDR(wqs, wq)          \
+			((wqs)->page_paddr[(wq)->page_idx] \
+				+ (wq)->block_idx * WQ_BLOCK_SIZE)
+
+#define WQ_BASE_ADDR(wqs, wq)           \
+			((void *)((wqs)->shadow_page_vaddr[(wq)->page_idx]) \
+				+ (wq)->block_idx * WQ_BLOCK_SIZE)
+
+#define CMDQ_BASE_VADDR(cmdq_pages, wq) \
+			((void *)((cmdq_pages)->page_vaddr) \
+				+ (wq)->block_idx * CMDQ_BLOCK_SIZE)
+
+#define CMDQ_BASE_PADDR(cmdq_pages, wq) \
+			((cmdq_pages)->page_paddr \
+				+ (wq)->block_idx * CMDQ_BLOCK_SIZE)
+
+#define CMDQ_BASE_ADDR(cmdq_pages, wq)  \
+			((void *)((cmdq_pages)->shadow_page_vaddr) \
+				+ (wq)->block_idx * CMDQ_BLOCK_SIZE)
+
+#define WQ_PAGE_ADDR(wq, idx)           \
+			((wq)->shadow_block_vaddr[WQE_PAGE_NUM(wq, idx)])
+
+#define MASKED_WQE_IDX(wq, idx)         ((idx) & (wq)->mask)
+
+#define WQE_IN_RANGE(wqe, start, end)   \
+		(((unsigned long)(wqe) >= (unsigned long)(start)) && \
+		 ((unsigned long)(wqe) < (unsigned long)(end)))
+
+#define WQE_SHADOW_PAGE(wq, wqe)        \
+		(((unsigned long)(wqe) - (unsigned long)(wq)->shadow_wqe) \
+			/ (wq)->max_wqe_size)
+
+static inline int WQE_PAGE_OFF(struct hinic_wq *wq, u16 idx)
+{
+	return (((idx) & ((wq)->num_wqebbs_per_page - 1))
+		<< (wq)->wqebb_size_shift);
+}
+
+static inline int WQE_PAGE_NUM(struct hinic_wq *wq, u16 idx)
+{
+	return (((idx) >> ((wq)->wqebbs_per_page_shift))
+		& ((wq)->num_q_pages - 1));
+}
+
+/**
+ * queue_alloc_page - allocate page for Queue
+ * @hwif: HW interface for allocating DMA
+ * @vaddr: virtual address will be returned in this address
+ * @paddr: physical address will be returned in this address
+ * @shadow_vaddr: VM area will be return here for holding WQ page addresses
+ * @page_sz: page size of each WQ page
+ *
+ * Return 0 - Success, negative - Failure
+ **/
+static int queue_alloc_page(struct hinic_hwif *hwif, u64 **vaddr, u64 *paddr,
+			    void ***shadow_vaddr, size_t page_sz)
+{
+	struct pci_dev *pdev = hwif->pdev;
+	dma_addr_t dma_addr;
+
+	*vaddr = dma_alloc_coherent(&pdev->dev, page_sz, &dma_addr,
+				    GFP_KERNEL);
+	if (!*vaddr) {
+		dev_err(&pdev->dev, "Failed to allocate dma for wqs page\n");
+		return -ENOMEM;
+	}
+
+	*paddr = (u64)dma_addr;
+
+	/* use vzalloc for big mem */
+	*shadow_vaddr = vzalloc(page_sz);
+	if (!*shadow_vaddr)
+		goto err_shadow_vaddr;
+
+	return 0;
+
+err_shadow_vaddr:
+	dma_free_coherent(&pdev->dev, page_sz, *vaddr, dma_addr);
+	return -ENOMEM;
+}
+
+/**
+ * wqs_allocate_page - allocate page for WQ set
+ * @wqs: Work Queue Set
+ * @page_idx: the page index of the page will be allocated
+ *
+ * Return 0 - Success, negative - Failure
+ **/
+static int wqs_allocate_page(struct hinic_wqs *wqs, int page_idx)
+{
+	return queue_alloc_page(wqs->hwif, &wqs->page_vaddr[page_idx],
+				&wqs->page_paddr[page_idx],
+				&wqs->shadow_page_vaddr[page_idx],
+				WQS_PAGE_SIZE);
+}
+
+/**
+ * wqs_free_page - free page of WQ set
+ * @wqs: Work Queue Set
+ * @page_idx: the page index of the page will be freed
+ **/
+static void wqs_free_page(struct hinic_wqs *wqs, int page_idx)
+{
+	struct hinic_hwif *hwif = wqs->hwif;
+	struct pci_dev *pdev = hwif->pdev;
+
+	dma_free_coherent(&pdev->dev, WQS_PAGE_SIZE,
+			  wqs->page_vaddr[page_idx],
+			  (dma_addr_t)wqs->page_paddr[page_idx]);
+	vfree(wqs->shadow_page_vaddr[page_idx]);
+}
+
+/**
+ * cmdq_allocate_page - allocate page for cmdq
+ * @cmdq_pages: the pages of the cmdq queue struct to hold the page
+ *
+ * Return 0 - Success, negative - Failure
+ **/
+static int cmdq_allocate_page(struct hinic_cmdq_pages *cmdq_pages)
+{
+	return queue_alloc_page(cmdq_pages->hwif, &cmdq_pages->page_vaddr,
+				&cmdq_pages->page_paddr,
+				&cmdq_pages->shadow_page_vaddr,
+				CMDQ_PAGE_SIZE);
+}
+
+/**
+ * cmdq_free_page - free page from cmdq
+ * @cmdq_pages: the pages of the cmdq queue struct that hold the page
+ **/
+static void cmdq_free_page(struct hinic_cmdq_pages *cmdq_pages)
+{
+	struct hinic_hwif *hwif = cmdq_pages->hwif;
+	struct pci_dev *pdev = hwif->pdev;
+
+	dma_free_coherent(&pdev->dev, CMDQ_PAGE_SIZE,
+			  cmdq_pages->page_vaddr,
+			  (dma_addr_t)cmdq_pages->page_paddr);
+	vfree(cmdq_pages->shadow_page_vaddr);
+}
+
+static int alloc_page_arrays(struct hinic_wqs *wqs)
+{
+	struct hinic_hwif *hwif = wqs->hwif;
+	struct pci_dev *pdev = hwif->pdev;
+
+	wqs->page_paddr = devm_kcalloc(&pdev->dev, wqs->num_pages,
+				       sizeof(*wqs->page_paddr), GFP_KERNEL);
+	if (!wqs->page_paddr)
+		return -ENOMEM;
+
+	wqs->page_vaddr = devm_kcalloc(&pdev->dev, wqs->num_pages,
+				       sizeof(*wqs->page_vaddr), GFP_KERNEL);
+	if (!wqs->page_vaddr)
+		goto err_page_vaddr;
+
+	wqs->shadow_page_vaddr = devm_kcalloc(&pdev->dev, wqs->num_pages,
+					      sizeof(*wqs->shadow_page_vaddr),
+					      GFP_KERNEL);
+	if (!wqs->shadow_page_vaddr)
+		goto err_page_shadow_vaddr;
+
+	return 0;
+
+err_page_shadow_vaddr:
+	devm_kfree(&pdev->dev, wqs->page_vaddr);
+
+err_page_vaddr:
+	devm_kfree(&pdev->dev, wqs->page_paddr);
+	return -ENOMEM;
+}
+
+static void free_page_arrays(struct hinic_wqs *wqs)
+{
+	struct hinic_hwif *hwif = wqs->hwif;
+	struct pci_dev *pdev = hwif->pdev;
+
+	devm_kfree(&pdev->dev, wqs->shadow_page_vaddr);
+	devm_kfree(&pdev->dev, wqs->page_vaddr);
+	devm_kfree(&pdev->dev, wqs->page_paddr);
+}
+
+static int wqs_next_block(struct hinic_wqs *wqs, int *page_idx,
+			  int *block_idx)
+{
+	int pos;
+
+	down(&wqs->alloc_blocks_lock);
+
+	wqs->num_free_blks--;
+
+	if (wqs->num_free_blks < 0) {
+		wqs->num_free_blks++;
+		up(&wqs->alloc_blocks_lock);
+		return -ENOMEM;
+	}
+
+	pos = wqs->alloc_blk_pos++;
+	pos &= WQS_MAX_NUM_BLOCKS - 1;
+
+	*page_idx = wqs->free_blocks[pos].page_idx;
+	*block_idx = wqs->free_blocks[pos].block_idx;
+
+	wqs->free_blocks[pos].page_idx = -1;
+	wqs->free_blocks[pos].block_idx = -1;
+
+	up(&wqs->alloc_blocks_lock);
+	return 0;
+}
+
+static void wqs_return_block(struct hinic_wqs *wqs, int page_idx,
+			     int block_idx)
+{
+	int pos;
+
+	down(&wqs->alloc_blocks_lock);
+
+	pos = wqs->return_blk_pos++;
+	pos &= WQS_MAX_NUM_BLOCKS - 1;
+
+	wqs->free_blocks[pos].page_idx = page_idx;
+	wqs->free_blocks[pos].block_idx = block_idx;
+
+	wqs->num_free_blks++;
+
+	up(&wqs->alloc_blocks_lock);
+}
+
+static void init_wqs_blocks_arr(struct hinic_wqs *wqs)
+{
+	int page_idx, blk_idx, pos = 0;
+
+	for (page_idx = 0; page_idx < wqs->num_pages; page_idx++) {
+		for (blk_idx = 0; blk_idx < WQS_BLOCKS_PER_PAGE; blk_idx++) {
+			wqs->free_blocks[pos].page_idx = page_idx;
+			wqs->free_blocks[pos].block_idx = blk_idx;
+			pos++;
+		}
+	}
+
+	wqs->alloc_blk_pos = 0;
+	wqs->return_blk_pos = pos;
+	wqs->num_free_blks = pos;
+
+	sema_init(&wqs->alloc_blocks_lock, 1);
+}
+
+/**
+ * hinic_wqs_alloc - allocate Work Queues set
+ * @wqs: Work Queue Set
+ * @max_wqs: maximum wqs to allocate
+ * @hwif: HW interface for use for the allocation
+ *
+ * Return 0 - Success, negative - Failure
+ **/
+int hinic_wqs_alloc(struct hinic_wqs *wqs, int max_wqs,
+		    struct hinic_hwif *hwif)
+{
+	struct pci_dev *pdev = hwif->pdev;
+	int err, i, page_idx;
+
+	max_wqs = ALIGN(max_wqs, WQS_BLOCKS_PER_PAGE);
+	if (max_wqs > WQS_MAX_NUM_BLOCKS)  {
+		dev_err(&pdev->dev, "Invalid max_wqs = %d\n", max_wqs);
+		return -EINVAL;
+	}
+
+	wqs->hwif = hwif;
+	wqs->num_pages = max_wqs / WQS_BLOCKS_PER_PAGE;
+
+	if (alloc_page_arrays(wqs)) {
+		dev_err(&pdev->dev,
+			"Failed to allocate mem for page addresses\n");
+		return -ENOMEM;
+	}
+
+	for (page_idx = 0; page_idx < wqs->num_pages; page_idx++) {
+		err = wqs_allocate_page(wqs, page_idx);
+		if (err) {
+			dev_err(&pdev->dev, "Failed wq page allocation\n");
+			goto err_wq_allocate_page;
+		}
+	}
+
+	wqs->free_blocks = devm_kzalloc(&pdev->dev, WQS_FREE_BLOCKS_SIZE(wqs),
+					GFP_KERNEL);
+	if (!wqs->free_blocks) {
+		err = -ENOMEM;
+		goto err_alloc_blocks;
+	}
+
+	init_wqs_blocks_arr(wqs);
+	return 0;
+
+err_alloc_blocks:
+err_wq_allocate_page:
+	for (i = 0; i < page_idx; i++)
+		wqs_free_page(wqs, i);
+
+	free_page_arrays(wqs);
+	return err;
+}
+
+/**
+ * hinic_wqs_free - free Work Queues set
+ * @wqs: Work Queue Set
+ **/
+void hinic_wqs_free(struct hinic_wqs *wqs)
+{
+	struct hinic_hwif *hwif = wqs->hwif;
+	struct pci_dev *pdev = hwif->pdev;
+	int page_idx;
+
+	devm_kfree(&pdev->dev, wqs->free_blocks);
+
+	for (page_idx = 0; page_idx < wqs->num_pages; page_idx++)
+		wqs_free_page(wqs, page_idx);
+
+	free_page_arrays(wqs);
+}
+
+/**
+ * alloc_wqes_shadow - allocate WQE shadows for WQ
+ * @wq: WQ to allocate shadows for
+ *
+ * Return 0 - Success, negative - Failure
+ **/
+static int alloc_wqes_shadow(struct hinic_wq *wq)
+{
+	struct hinic_hwif *hwif = wq->hwif;
+	struct pci_dev *pdev = hwif->pdev;
+
+	wq->shadow_wqe = devm_kcalloc(&pdev->dev, wq->num_q_pages,
+				      wq->max_wqe_size, GFP_KERNEL);
+	if (!wq->shadow_wqe)
+		return -ENOMEM;
+
+	wq->shadow_idx = devm_kcalloc(&pdev->dev, wq->num_q_pages,
+				      sizeof(*wq->shadow_idx), GFP_KERNEL);
+	if (!wq->shadow_idx)
+		goto err_shadow_idx;
+
+	return 0;
+
+err_shadow_idx:
+	devm_kfree(&pdev->dev, wq->shadow_wqe);
+	return -ENOMEM;
+}
+
+/**
+ * free_wqes_shadow - free WQE shadows of WQ
+ * @wq: WQ to free shadows from
+ **/
+static void free_wqes_shadow(struct hinic_wq *wq)
+{
+	struct hinic_hwif *hwif = wq->hwif;
+	struct pci_dev *pdev = hwif->pdev;
+
+	devm_kfree(&pdev->dev, wq->shadow_idx);
+	devm_kfree(&pdev->dev, wq->shadow_wqe);
+}
+
+/**
+ * free_wq_pages - free pages of WQ
+ * @hwif: HW interface for releasing dma addresses
+ * @wq: WQ to free pages from
+ * @num_q_pages: number pages to free
+ **/
+static void free_wq_pages(struct hinic_wq *wq, struct hinic_hwif *hwif,
+			  int num_q_pages)
+{
+	struct pci_dev *pdev = hwif->pdev;
+	int i;
+
+	for (i = 0; i < num_q_pages; i++) {
+		void **vaddr = &wq->shadow_block_vaddr[i];
+		u64 *paddr = &wq->block_vaddr[i];
+		dma_addr_t dma_addr;
+
+		dma_addr = (dma_addr_t)be64_to_cpu(*paddr);
+		dma_free_coherent(&pdev->dev, wq->wq_page_size, *vaddr,
+				  dma_addr);
+	}
+
+	free_wqes_shadow(wq);
+}
+
+/**
+ * alloc_wq_pages - alloc pages for WQ
+ * @hwif: HW interface for allocating dma addresses
+ * @wq: WQ to allocate pages for
+ * @max_pages: maximum pages allowed
+ *
+ * Return 0 - Success, negative - Failure
+ **/
+static int alloc_wq_pages(struct hinic_wq *wq, struct hinic_hwif *hwif,
+			  int max_pages)
+{
+	struct pci_dev *pdev = hwif->pdev;
+	int i, err, num_q_pages;
+
+	num_q_pages = ALIGN(WQ_SIZE(wq), wq->wq_page_size) / wq->wq_page_size;
+	if (num_q_pages > max_pages) {
+		dev_err(&pdev->dev, "Number wq pages exceeds the limit\n");
+		return -EINVAL;
+	}
+
+	if (num_q_pages & (num_q_pages - 1)) {
+		dev_err(&pdev->dev, "Number wq pages must be power of 2\n");
+		return -EINVAL;
+	}
+
+	wq->num_q_pages = num_q_pages;
+
+	err = alloc_wqes_shadow(wq);
+	if (err) {
+		dev_err(&pdev->dev, "Failed to allocate wqe shadow\n");
+		return err;
+	}
+
+	for (i = 0; i < num_q_pages; i++) {
+		void **vaddr = &wq->shadow_block_vaddr[i];
+		u64 *paddr = &wq->block_vaddr[i];
+		dma_addr_t dma_addr;
+
+		*vaddr = dma_alloc_coherent(&pdev->dev, wq->wq_page_size,
+					    &dma_addr, GFP_KERNEL);
+		if (!*vaddr) {
+			dev_err(&pdev->dev, "Failed to allocate wq page\n");
+			goto err_alloc_wq_pages;
+		}
+
+		/* HW uses Big Endian Format */
+		*paddr = cpu_to_be64(dma_addr);
+	}
+
+	return 0;
+
+err_alloc_wq_pages:
+	free_wq_pages(wq, hwif, i);
+	return -ENOMEM;
+}
+
+/**
+ * hinic_wq_allocate - Allocate the WQ resources from the WQS
+ * @wqs: WQ set from which to allocate the WQ resources
+ * @wq: WQ to allocate resources for it from the WQ set
+ * @wqebb_size: Work Queue Block Byte Size
+ * @wq_page_size: the page size in the Work Queue
+ * @q_depth: number of wqebbs in WQ
+ * @max_wqe_size: maximum WQE size that will be used in the WQ
+ *
+ * Return 0 - Success, negative - Failure
+ **/
+int hinic_wq_allocate(struct hinic_wqs *wqs, struct hinic_wq *wq,
+		      u16 wqebb_size, u32 wq_page_size, u16 q_depth,
+		      u16 max_wqe_size)
+{
+	struct hinic_hwif *hwif = wqs->hwif;
+	struct pci_dev *pdev = hwif->pdev;
+	u16 num_wqebbs_per_page;
+	u16 wqebb_size_shift;
+	int err;
+
+	if (!is_power_of_2(wqebb_size)) {
+		dev_err(&pdev->dev, "wqebb_size must be power of 2\n");
+		return -EINVAL;
+	}
+
+	if (wq_page_size == 0) {
+		dev_err(&pdev->dev, "wq_page_size must be > 0\n");
+		return -EINVAL;
+	}
+
+	if (q_depth & (q_depth - 1)) {
+		dev_err(&pdev->dev, "WQ q_depth must be power of 2\n");
+		return -EINVAL;
+	}
+
+	wqebb_size_shift = ilog2(wqebb_size);
+	num_wqebbs_per_page = ALIGN(wq_page_size, wqebb_size)
+				>> wqebb_size_shift;
+
+	if (!is_power_of_2(num_wqebbs_per_page)) {
+		dev_err(&pdev->dev, "num wqebbs per page must be power of 2\n");
+		return -EINVAL;
+	}
+
+	wq->hwif = hwif;
+
+	err = wqs_next_block(wqs, &wq->page_idx, &wq->block_idx);
+	if (err) {
+		dev_err(&pdev->dev, "Failed to get free wqs next block\n");
+		return err;
+	}
+
+	wq->wqebb_size = wqebb_size;
+	wq->wq_page_size = wq_page_size;
+	wq->q_depth = q_depth;
+	wq->max_wqe_size = max_wqe_size;
+	wq->num_wqebbs_per_page = num_wqebbs_per_page;
+	wq->wqebbs_per_page_shift = ilog2(num_wqebbs_per_page);
+	wq->wqebb_size_shift = wqebb_size_shift;
+	wq->block_vaddr = WQ_BASE_VADDR(wqs, wq);
+	wq->shadow_block_vaddr = WQ_BASE_ADDR(wqs, wq);
+	wq->block_paddr = WQ_BASE_PADDR(wqs, wq);
+
+	err = alloc_wq_pages(wq, wqs->hwif, WQ_MAX_PAGES);
+	if (err) {
+		dev_err(&pdev->dev, "Failed to allocate wq pages\n");
+		goto err_alloc_wq_pages;
+	}
+
+	atomic_set(&wq->cons_idx, 0);
+	atomic_set(&wq->prod_idx, 0);
+	atomic_set(&wq->delta, q_depth);
+	wq->mask = q_depth - 1;
+
+	return 0;
+
+err_alloc_wq_pages:
+	wqs_return_block(wqs, wq->page_idx, wq->block_idx);
+	return err;
+}
+
+/**
+ * hinic_wq_free - Free the WQ resources to the WQS
+ * @wqs: WQ set to free the WQ resources to it
+ * @wq: WQ to free its resources to the WQ set resources
+ **/
+void hinic_wq_free(struct hinic_wqs *wqs, struct hinic_wq *wq)
+{
+	free_wq_pages(wq, wqs->hwif, wq->num_q_pages);
+
+	wqs_return_block(wqs, wq->page_idx, wq->block_idx);
+}
+
+/**
+ * hinic_wqs_cmdq_alloc - Allocate wqs for cmdqs
+ * @cmdq_pages: will hold the pages of the cmdq
+ * @wq: returned wqs
+ * @hwif: HW interface
+ * @cmdq_blocks: number of cmdq blocks/wq to allocate
+ * @wqebb_size: Work Queue Block Byte Size
+ * @wq_page_size: the page size in the Work Queue
+ * @q_depth: number of wqebbs in WQ
+ * @max_wqe_size: maximum WQE size that will be used in the WQ
+ *
+ * Return 0 - Success, negative - Failure
+ **/
+int hinic_wqs_cmdq_alloc(struct hinic_cmdq_pages *cmdq_pages,
+			 struct hinic_wq *wq, struct hinic_hwif *hwif,
+			 int cmdq_blocks, u16 wqebb_size, u32 wq_page_size,
+			 u16 q_depth, u16 max_wqe_size)
+{
+	struct pci_dev *pdev = hwif->pdev;
+	u16 num_wqebbs_per_page_shift;
+	u16 num_wqebbs_per_page;
+	u16 wqebb_size_shift;
+	int i, j, err = -ENOMEM;
+
+	if (!is_power_of_2(wqebb_size)) {
+		dev_err(&pdev->dev, "wqebb_size must be power of 2\n");
+		return -EINVAL;
+	}
+
+	if (wq_page_size == 0) {
+		dev_err(&pdev->dev, "wq_page_size must be > 0\n");
+		return -EINVAL;
+	}
+
+	if (q_depth & (q_depth - 1)) {
+		dev_err(&pdev->dev, "WQ q_depth must be power of 2\n");
+		return -EINVAL;
+	}
+
+	wqebb_size_shift = ilog2(wqebb_size);
+	num_wqebbs_per_page = ALIGN(wq_page_size, wqebb_size)
+				>> wqebb_size_shift;
+
+	if (!is_power_of_2(num_wqebbs_per_page)) {
+		dev_err(&pdev->dev, "num wqebbs per page must be power of 2\n");
+		return -EINVAL;
+	}
+
+	cmdq_pages->hwif = hwif;
+
+	err = cmdq_allocate_page(cmdq_pages);
+	if (err) {
+		dev_err(&pdev->dev, "Failed to allocate CMDQ page\n");
+		return err;
+	}
+	num_wqebbs_per_page_shift = ilog2(num_wqebbs_per_page);
+
+	for (i = 0; i < cmdq_blocks; i++) {
+		wq[i].hwif = hwif;
+		wq[i].page_idx = 0;
+		wq[i].block_idx = i;
+
+		wq[i].wqebb_size = wqebb_size;
+		wq[i].wq_page_size = wq_page_size;
+		wq[i].q_depth = q_depth;
+		wq[i].max_wqe_size = max_wqe_size;
+		wq[i].num_wqebbs_per_page = num_wqebbs_per_page;
+		wq[i].wqebbs_per_page_shift = num_wqebbs_per_page_shift;
+		wq[i].wqebb_size_shift = wqebb_size_shift;
+		wq[i].block_vaddr = CMDQ_BASE_VADDR(cmdq_pages, &wq[i]);
+		wq[i].shadow_block_vaddr = CMDQ_BASE_ADDR(cmdq_pages, &wq[i]);
+		wq[i].block_paddr = CMDQ_BASE_PADDR(cmdq_pages, &wq[i]);
+
+		err = alloc_wq_pages(&wq[i], cmdq_pages->hwif,
+				     CMDQ_WQ_MAX_PAGES);
+		if (err) {
+			dev_err(&pdev->dev, "Failed to alloc CMDQ blocks\n");
+			goto err_cmdq_block;
+		}
+
+		atomic_set(&wq[i].cons_idx, 0);
+		atomic_set(&wq[i].prod_idx, 0);
+		atomic_set(&wq[i].delta, q_depth);
+		wq[i].mask = q_depth - 1;
+	}
+
+	return 0;
+
+err_cmdq_block:
+	for (j = 0; j < i; j++)
+		free_wq_pages(&wq[j], cmdq_pages->hwif, wq[j].num_q_pages);
+
+	cmdq_free_page(cmdq_pages);
+	return err;
+}
+
+/**
+ * hinic_wqs_cmdq_free - Free wqs from cmdqs
+ * @cmdq_pages: hold the pages of the cmdq
+ * @wq: wqs to free
+ * @cmdq_blocks: number of wqs to free
+ **/
+void hinic_wqs_cmdq_free(struct hinic_cmdq_pages *cmdq_pages,
+			 struct hinic_wq *wq, int cmdq_blocks)
+{
+	int i;
+
+	for (i = 0; i < cmdq_blocks; i++)
+		free_wq_pages(&wq[i], cmdq_pages->hwif, wq[i].num_q_pages);
+
+	cmdq_free_page(cmdq_pages);
+}
+
+static void copy_wqe_to_shadow(struct hinic_wq *wq, void *shadow_addr,
+			       int num_wqebbs, u16 idx)
+{
+	void *wqebb_addr;
+	int i;
+
+	for (i = 0; i < num_wqebbs; i++, idx++) {
+		idx = MASKED_WQE_IDX(wq, idx);
+		wqebb_addr = WQ_PAGE_ADDR(wq, idx) +
+			     WQE_PAGE_OFF(wq, idx);
+
+		memcpy(shadow_addr, wqebb_addr, wq->wqebb_size);
+
+		shadow_addr += wq->wqebb_size;
+	}
+}
+
+static void copy_wqe_from_shadow(struct hinic_wq *wq, void *shadow_addr,
+				 int num_wqebbs, u16 idx)
+{
+	void *wqebb_addr;
+	int i;
+
+	for (i = 0; i < num_wqebbs; i++, idx++) {
+		idx = MASKED_WQE_IDX(wq, idx);
+		wqebb_addr = WQ_PAGE_ADDR(wq, idx) +
+			     WQE_PAGE_OFF(wq, idx);
+
+		memcpy(wqebb_addr, shadow_addr, wq->wqebb_size);
+		shadow_addr += wq->wqebb_size;
+	}
+}
+
+/**
+ * hinic_get_wqe - get wqe ptr in the current pi and update the pi
+ * @wq: wq to get wqe from
+ * @wqe_size: wqe size
+ * @prod_idx: returned pi
+ *
+ * Return wqe pointer
+ **/
+struct hinic_hw_wqe *hinic_get_wqe(struct hinic_wq *wq, unsigned int wqe_size,
+				   u16 *prod_idx)
+{
+	int curr_pg, end_pg, num_wqebbs;
+	u16 curr_prod_idx, end_prod_idx;
+
+	*prod_idx = MASKED_WQE_IDX(wq, atomic_read(&wq->prod_idx));
+
+	num_wqebbs = ALIGN(wqe_size, wq->wqebb_size) >> wq->wqebb_size_shift;
+
+	if (atomic_sub_return(num_wqebbs, &wq->delta) <= 0) {
+		atomic_add(num_wqebbs, &wq->delta);
+		return ERR_PTR(-EBUSY);
+	}
+
+	end_prod_idx = atomic_add_return(num_wqebbs, &wq->prod_idx);
+
+	end_prod_idx = MASKED_WQE_IDX(wq, end_prod_idx);
+	curr_prod_idx = end_prod_idx - num_wqebbs;
+	curr_prod_idx = MASKED_WQE_IDX(wq, curr_prod_idx);
+
+	/* end prod index points to the next wqebb, therefore minus 1 */
+	end_prod_idx = MASKED_WQE_IDX(wq, end_prod_idx - 1);
+
+	curr_pg = WQE_PAGE_NUM(wq, curr_prod_idx);
+	end_pg = WQE_PAGE_NUM(wq, end_prod_idx);
+
+	*prod_idx = curr_prod_idx;
+
+	/* If we only have one page, still need to get shadown wqe when
+	 * wqe rolling-over page
+	 */
+	if (curr_pg != end_pg || end_prod_idx < *prod_idx) {
+		void *shadow_addr = &wq->shadow_wqe[curr_pg * wq->max_wqe_size];
+
+		copy_wqe_to_shadow(wq, shadow_addr, num_wqebbs, *prod_idx);
+
+		wq->shadow_idx[curr_pg] = *prod_idx;
+		return shadow_addr;
+	}
+
+	return WQ_PAGE_ADDR(wq, *prod_idx) + WQE_PAGE_OFF(wq, *prod_idx);
+}
+
+/**
+ * hinic_return_wqe - return the wqe when transmit failed
+ * @wq: wq to return wqe
+ * @wqe_size: wqe size
+ **/
+void hinic_return_wqe(struct hinic_wq *wq, unsigned int wqe_size)
+{
+	int num_wqebbs = ALIGN(wqe_size, wq->wqebb_size) / wq->wqebb_size;
+
+	atomic_sub(num_wqebbs, &wq->prod_idx);
+
+	atomic_add(num_wqebbs, &wq->delta);
+}
+
+/**
+ * hinic_put_wqe - return the wqe place to use for a new wqe
+ * @wq: wq to return wqe
+ * @wqe_size: wqe size
+ **/
+void hinic_put_wqe(struct hinic_wq *wq, unsigned int wqe_size)
+{
+	int num_wqebbs = ALIGN(wqe_size, wq->wqebb_size)
+			>> wq->wqebb_size_shift;
+
+	atomic_add(num_wqebbs, &wq->cons_idx);
+
+	atomic_add(num_wqebbs, &wq->delta);
+}
+
+/**
+ * hinic_read_wqe - read wqe ptr in the current ci
+ * @wq: wq to get read from
+ * @wqe_size: wqe size
+ * @cons_idx: returned ci
+ *
+ * Return wqe pointer
+ **/
+struct hinic_hw_wqe *hinic_read_wqe(struct hinic_wq *wq, unsigned int wqe_size,
+				    u16 *cons_idx)
+{
+	int num_wqebbs = ALIGN(wqe_size, wq->wqebb_size)
+			>> wq->wqebb_size_shift;
+	u16 curr_cons_idx, end_cons_idx;
+	int curr_pg, end_pg;
+
+	if ((atomic_read(&wq->delta) + num_wqebbs) > wq->q_depth)
+		return ERR_PTR(-EBUSY);
+
+	curr_cons_idx = atomic_read(&wq->cons_idx);
+
+	curr_cons_idx = MASKED_WQE_IDX(wq, curr_cons_idx);
+	end_cons_idx = MASKED_WQE_IDX(wq, curr_cons_idx + num_wqebbs - 1);
+
+	curr_pg = WQE_PAGE_NUM(wq, curr_cons_idx);
+	end_pg = WQE_PAGE_NUM(wq, end_cons_idx);
+
+	*cons_idx = curr_cons_idx;
+
+	/* If we only have one page, still need to get shadown wqe when
+	 * wqe rolling-over page
+	 */
+	if (curr_pg != end_pg || end_cons_idx < curr_cons_idx) {
+		void *shadow_addr = &wq->shadow_wqe[curr_pg * wq->max_wqe_size];
+
+		copy_wqe_to_shadow(wq, shadow_addr, num_wqebbs, *cons_idx);
+		return shadow_addr;
+	}
+
+	return WQ_PAGE_ADDR(wq, *cons_idx) + WQE_PAGE_OFF(wq, *cons_idx);
+}
+
+/**
+ * hinic_read_wqe_direct - read wqe directly from ci position
+ * @wq: wq
+ * @cons_idx: ci position
+ *
+ * Return wqe
+ **/
+struct hinic_hw_wqe *hinic_read_wqe_direct(struct hinic_wq *wq, u16 cons_idx)
+{
+	return WQ_PAGE_ADDR(wq, cons_idx) + WQE_PAGE_OFF(wq, cons_idx);
+}
+
+/**
+ * wqe_shadow - check if a wqe is shadow
+ * @wq: wq of the wqe
+ * @wqe: the wqe for shadow checking
+ *
+ * Return true - shadow, false - Not shadow
+ **/
+static inline bool wqe_shadow(struct hinic_wq *wq, struct hinic_hw_wqe *wqe)
+{
+	size_t wqe_shadow_size = wq->num_q_pages * wq->max_wqe_size;
+
+	return WQE_IN_RANGE(wqe, wq->shadow_wqe,
+			    &wq->shadow_wqe[wqe_shadow_size]);
+}
+
+/**
+ * hinic_write_wqe - write the wqe to the wq
+ * @wq: wq to write wqe to
+ * @wqe: wqe to write
+ * @wqe_size: wqe size
+ **/
+void hinic_write_wqe(struct hinic_wq *wq, struct hinic_hw_wqe *wqe,
+		     unsigned int wqe_size)
+{
+	int curr_pg, num_wqebbs;
+	void *shadow_addr;
+	u16 prod_idx;
+
+	if (wqe_shadow(wq, wqe)) {
+		curr_pg = WQE_SHADOW_PAGE(wq, wqe);
+
+		prod_idx = wq->shadow_idx[curr_pg];
+		num_wqebbs = ALIGN(wqe_size, wq->wqebb_size) / wq->wqebb_size;
+		shadow_addr = &wq->shadow_wqe[curr_pg * wq->max_wqe_size];
+
+		copy_wqe_from_shadow(wq, shadow_addr, num_wqebbs, prod_idx);
+	}
+}
diff --git a/drivers/net/ethernet/huawei/hinic/hinic_hw_wq.h b/drivers/net/ethernet/huawei/hinic/hinic_hw_wq.h
new file mode 100644
index 000000000000..b06f8c0255de
--- /dev/null
+++ b/drivers/net/ethernet/huawei/hinic/hinic_hw_wq.h
@@ -0,0 +1,111 @@
+/* SPDX-License-Identifier: GPL-2.0-only */
+/*
+ * Huawei HiNIC PCI Express Linux driver
+ * Copyright(c) 2017 Huawei Technologies Co., Ltd
+ */
+
+#ifndef HINIC_HW_WQ_H
+#define HINIC_HW_WQ_H
+
+#include <linux/types.h>
+#include <linux/semaphore.h>
+#include <linux/atomic.h>
+
+#include "hinic_hw_if.h"
+#include "hinic_hw_wqe.h"
+
+struct hinic_free_block {
+	int     page_idx;
+	int     block_idx;
+};
+
+struct hinic_wq {
+	struct hinic_hwif       *hwif;
+
+	int             page_idx;
+	int             block_idx;
+
+	u16             wqebb_size;
+	u32             wq_page_size;
+	u16             q_depth;
+	u16             max_wqe_size;
+	u16             num_wqebbs_per_page;
+	u16		wqebbs_per_page_shift;
+	u16		wqebb_size_shift;
+	/* The addresses are 64 bit in the HW */
+	u64             block_paddr;
+	void            **shadow_block_vaddr;
+	u64             *block_vaddr;
+
+	int             num_q_pages;
+	u8              *shadow_wqe;
+	u16             *shadow_idx;
+
+	atomic_t        cons_idx;
+	atomic_t        prod_idx;
+	atomic_t        delta;
+	u16             mask;
+};
+
+struct hinic_wqs {
+	struct hinic_hwif       *hwif;
+	int                     num_pages;
+
+	/* The addresses are 64 bit in the HW */
+	u64                     *page_paddr;
+	u64                     **page_vaddr;
+	void                    ***shadow_page_vaddr;
+
+	struct hinic_free_block *free_blocks;
+	int                     alloc_blk_pos;
+	int                     return_blk_pos;
+	int                     num_free_blks;
+
+	/* Lock for getting a free block from the WQ set */
+	struct semaphore        alloc_blocks_lock;
+};
+
+struct hinic_cmdq_pages {
+	/* The addresses are 64 bit in the HW */
+	u64                     page_paddr;
+	u64                     *page_vaddr;
+	void                    **shadow_page_vaddr;
+
+	struct hinic_hwif       *hwif;
+};
+
+int hinic_wqs_cmdq_alloc(struct hinic_cmdq_pages *cmdq_pages,
+			 struct hinic_wq *wq, struct hinic_hwif *hwif,
+			 int cmdq_blocks, u16 wqebb_size, u32 wq_page_size,
+			 u16 q_depth, u16 max_wqe_size);
+
+void hinic_wqs_cmdq_free(struct hinic_cmdq_pages *cmdq_pages,
+			 struct hinic_wq *wq, int cmdq_blocks);
+
+int hinic_wqs_alloc(struct hinic_wqs *wqs, int num_wqs,
+		    struct hinic_hwif *hwif);
+
+void hinic_wqs_free(struct hinic_wqs *wqs);
+
+int hinic_wq_allocate(struct hinic_wqs *wqs, struct hinic_wq *wq,
+		      u16 wqebb_size, u32 wq_page_size, u16 q_depth,
+		      u16 max_wqe_size);
+
+void hinic_wq_free(struct hinic_wqs *wqs, struct hinic_wq *wq);
+
+struct hinic_hw_wqe *hinic_get_wqe(struct hinic_wq *wq, unsigned int wqe_size,
+				   u16 *prod_idx);
+
+void hinic_return_wqe(struct hinic_wq *wq, unsigned int wqe_size);
+
+void hinic_put_wqe(struct hinic_wq *wq, unsigned int wqe_size);
+
+struct hinic_hw_wqe *hinic_read_wqe(struct hinic_wq *wq, unsigned int wqe_size,
+				    u16 *cons_idx);
+
+struct hinic_hw_wqe *hinic_read_wqe_direct(struct hinic_wq *wq, u16 cons_idx);
+
+void hinic_write_wqe(struct hinic_wq *wq, struct hinic_hw_wqe *wqe,
+		     unsigned int wqe_size);
+
+#endif
diff --git a/drivers/net/ethernet/huawei/hinic/hinic_hw_wqe.h b/drivers/net/ethernet/huawei/hinic/hinic_hw_wqe.h
new file mode 100644
index 000000000000..c6bdeed5606e
--- /dev/null
+++ b/drivers/net/ethernet/huawei/hinic/hinic_hw_wqe.h
@@ -0,0 +1,430 @@
+/* SPDX-License-Identifier: GPL-2.0-only */
+/*
+ * Huawei HiNIC PCI Express Linux driver
+ * Copyright(c) 2017 Huawei Technologies Co., Ltd
+ */
+
+#ifndef HINIC_HW_WQE_H
+#define HINIC_HW_WQE_H
+
+#include "hinic_common.h"
+
+#define HINIC_CMDQ_CTRL_PI_SHIFT                        0
+#define HINIC_CMDQ_CTRL_CMD_SHIFT                       16
+#define HINIC_CMDQ_CTRL_MOD_SHIFT                       24
+#define HINIC_CMDQ_CTRL_ACK_TYPE_SHIFT                  29
+#define HINIC_CMDQ_CTRL_HW_BUSY_BIT_SHIFT               31
+
+#define HINIC_CMDQ_CTRL_PI_MASK                         0xFFFF
+#define HINIC_CMDQ_CTRL_CMD_MASK                        0xFF
+#define HINIC_CMDQ_CTRL_MOD_MASK                        0x1F
+#define HINIC_CMDQ_CTRL_ACK_TYPE_MASK                   0x3
+#define HINIC_CMDQ_CTRL_HW_BUSY_BIT_MASK                0x1
+
+#define HINIC_CMDQ_CTRL_SET(val, member)                        \
+			(((u32)(val) & HINIC_CMDQ_CTRL_##member##_MASK) \
+			 << HINIC_CMDQ_CTRL_##member##_SHIFT)
+
+#define HINIC_CMDQ_CTRL_GET(val, member)                        \
+			(((val) >> HINIC_CMDQ_CTRL_##member##_SHIFT) \
+			 & HINIC_CMDQ_CTRL_##member##_MASK)
+
+#define HINIC_CMDQ_WQE_HEADER_BUFDESC_LEN_SHIFT         0
+#define HINIC_CMDQ_WQE_HEADER_COMPLETE_FMT_SHIFT        15
+#define HINIC_CMDQ_WQE_HEADER_DATA_FMT_SHIFT            22
+#define HINIC_CMDQ_WQE_HEADER_COMPLETE_REQ_SHIFT        23
+#define HINIC_CMDQ_WQE_HEADER_COMPLETE_SECT_LEN_SHIFT   27
+#define HINIC_CMDQ_WQE_HEADER_CTRL_LEN_SHIFT            29
+#define HINIC_CMDQ_WQE_HEADER_TOGGLED_WRAPPED_SHIFT     31
+
+#define HINIC_CMDQ_WQE_HEADER_BUFDESC_LEN_MASK          0xFF
+#define HINIC_CMDQ_WQE_HEADER_COMPLETE_FMT_MASK         0x1
+#define HINIC_CMDQ_WQE_HEADER_DATA_FMT_MASK             0x1
+#define HINIC_CMDQ_WQE_HEADER_COMPLETE_REQ_MASK         0x1
+#define HINIC_CMDQ_WQE_HEADER_COMPLETE_SECT_LEN_MASK    0x3
+#define HINIC_CMDQ_WQE_HEADER_CTRL_LEN_MASK             0x3
+#define HINIC_CMDQ_WQE_HEADER_TOGGLED_WRAPPED_MASK      0x1
+
+#define HINIC_CMDQ_WQE_HEADER_SET(val, member)                  \
+			(((u32)(val) & HINIC_CMDQ_WQE_HEADER_##member##_MASK) \
+			 << HINIC_CMDQ_WQE_HEADER_##member##_SHIFT)
+
+#define HINIC_CMDQ_WQE_HEADER_GET(val, member)                  \
+			(((val) >> HINIC_CMDQ_WQE_HEADER_##member##_SHIFT) \
+			 & HINIC_CMDQ_WQE_HEADER_##member##_MASK)
+
+#define HINIC_SQ_CTRL_BUFDESC_SECT_LEN_SHIFT           0
+#define HINIC_SQ_CTRL_TASKSECT_LEN_SHIFT               16
+#define HINIC_SQ_CTRL_DATA_FORMAT_SHIFT                22
+#define HINIC_SQ_CTRL_LEN_SHIFT                        29
+
+#define HINIC_SQ_CTRL_BUFDESC_SECT_LEN_MASK            0xFF
+#define HINIC_SQ_CTRL_TASKSECT_LEN_MASK                0x1F
+#define HINIC_SQ_CTRL_DATA_FORMAT_MASK                 0x1
+#define HINIC_SQ_CTRL_LEN_MASK                         0x3
+
+#define HINIC_SQ_CTRL_QUEUE_INFO_PLDOFF_SHIFT          2
+#define HINIC_SQ_CTRL_QUEUE_INFO_UFO_SHIFT             10
+#define HINIC_SQ_CTRL_QUEUE_INFO_TSO_SHIFT             11
+#define HINIC_SQ_CTRL_QUEUE_INFO_TCPUDP_CS_SHIFT       12
+#define HINIC_SQ_CTRL_QUEUE_INFO_MSS_SHIFT             13
+#define HINIC_SQ_CTRL_QUEUE_INFO_SCTP_SHIFT            27
+#define HINIC_SQ_CTRL_QUEUE_INFO_UC_SHIFT              28
+#define HINIC_SQ_CTRL_QUEUE_INFO_PRI_SHIFT             29
+
+#define HINIC_SQ_CTRL_QUEUE_INFO_PLDOFF_MASK           0xFF
+#define HINIC_SQ_CTRL_QUEUE_INFO_UFO_MASK              0x1
+#define HINIC_SQ_CTRL_QUEUE_INFO_TSO_MASK              0x1
+#define HINIC_SQ_CTRL_QUEUE_INFO_TCPUDP_CS_MASK	       0x1
+#define HINIC_SQ_CTRL_QUEUE_INFO_MSS_MASK              0x3FFF
+#define HINIC_SQ_CTRL_QUEUE_INFO_SCTP_MASK             0x1
+#define HINIC_SQ_CTRL_QUEUE_INFO_UC_MASK               0x1
+#define HINIC_SQ_CTRL_QUEUE_INFO_PRI_MASK              0x7
+
+#define HINIC_SQ_CTRL_SET(val, member)          \
+		(((u32)(val) & HINIC_SQ_CTRL_##member##_MASK) \
+		 << HINIC_SQ_CTRL_##member##_SHIFT)
+
+#define HINIC_SQ_CTRL_GET(val, member)          \
+		(((val) >> HINIC_SQ_CTRL_##member##_SHIFT) \
+		 & HINIC_SQ_CTRL_##member##_MASK)
+
+#define HINIC_SQ_CTRL_CLEAR(val, member)	\
+		((u32)(val) & (~(HINIC_SQ_CTRL_##member##_MASK \
+		 << HINIC_SQ_CTRL_##member##_SHIFT)))
+
+#define HINIC_SQ_TASK_INFO0_L2HDR_LEN_SHIFT     0
+#define HINIC_SQ_TASK_INFO0_L4_OFFLOAD_SHIFT    8
+#define HINIC_SQ_TASK_INFO0_INNER_L3TYPE_SHIFT  10
+#define HINIC_SQ_TASK_INFO0_VLAN_OFFLOAD_SHIFT  12
+#define HINIC_SQ_TASK_INFO0_PARSE_FLAG_SHIFT    13
+/* 1 bit reserved */
+#define HINIC_SQ_TASK_INFO0_TSO_FLAG_SHIFT      15
+#define HINIC_SQ_TASK_INFO0_VLAN_TAG_SHIFT      16
+
+#define HINIC_SQ_TASK_INFO0_L2HDR_LEN_MASK      0xFF
+#define HINIC_SQ_TASK_INFO0_L4_OFFLOAD_MASK     0x3
+#define HINIC_SQ_TASK_INFO0_INNER_L3TYPE_MASK   0x3
+#define HINIC_SQ_TASK_INFO0_VLAN_OFFLOAD_MASK   0x1
+#define HINIC_SQ_TASK_INFO0_PARSE_FLAG_MASK     0x1
+/* 1 bit reserved */
+#define HINIC_SQ_TASK_INFO0_TSO_FLAG_MASK       0x1
+#define HINIC_SQ_TASK_INFO0_VLAN_TAG_MASK       0xFFFF
+
+#define HINIC_SQ_TASK_INFO0_SET(val, member)    \
+		(((u32)(val) & HINIC_SQ_TASK_INFO0_##member##_MASK) <<  \
+		 HINIC_SQ_TASK_INFO0_##member##_SHIFT)
+
+/* 8 bits reserved */
+#define HINIC_SQ_TASK_INFO1_MEDIA_TYPE_SHIFT    8
+#define HINIC_SQ_TASK_INFO1_INNER_L4LEN_SHIFT   16
+#define HINIC_SQ_TASK_INFO1_INNER_L3LEN_SHIFT   24
+
+/* 8 bits reserved */
+#define HINIC_SQ_TASK_INFO1_MEDIA_TYPE_MASK     0xFF
+#define HINIC_SQ_TASK_INFO1_INNER_L4LEN_MASK    0xFF
+#define HINIC_SQ_TASK_INFO1_INNER_L3LEN_MASK    0xFF
+
+#define HINIC_SQ_TASK_INFO1_SET(val, member)    \
+		(((u32)(val) & HINIC_SQ_TASK_INFO1_##member##_MASK) <<  \
+		 HINIC_SQ_TASK_INFO1_##member##_SHIFT)
+
+#define HINIC_SQ_TASK_INFO2_TUNNEL_L4LEN_SHIFT  0
+#define HINIC_SQ_TASK_INFO2_OUTER_L3LEN_SHIFT   8
+#define HINIC_SQ_TASK_INFO2_TUNNEL_L4TYPE_SHIFT 16
+/* 1 bit reserved */
+#define HINIC_SQ_TASK_INFO2_OUTER_L3TYPE_SHIFT  24
+/* 8 bits reserved */
+
+#define HINIC_SQ_TASK_INFO2_TUNNEL_L4LEN_MASK   0xFF
+#define HINIC_SQ_TASK_INFO2_OUTER_L3LEN_MASK    0xFF
+#define HINIC_SQ_TASK_INFO2_TUNNEL_L4TYPE_MASK  0x7
+/* 1 bit reserved */
+#define HINIC_SQ_TASK_INFO2_OUTER_L3TYPE_MASK   0x3
+/* 8 bits reserved */
+
+#define HINIC_SQ_TASK_INFO2_SET(val, member)    \
+		(((u32)(val) & HINIC_SQ_TASK_INFO2_##member##_MASK) <<  \
+		 HINIC_SQ_TASK_INFO2_##member##_SHIFT)
+
+/* 31 bits reserved */
+#define HINIC_SQ_TASK_INFO4_L2TYPE_SHIFT        31
+
+/* 31 bits reserved */
+#define HINIC_SQ_TASK_INFO4_L2TYPE_MASK         0x1
+
+#define HINIC_SQ_TASK_INFO4_SET(val, member)    \
+		(((u32)(val) & HINIC_SQ_TASK_INFO4_##member##_MASK) << \
+		 HINIC_SQ_TASK_INFO4_##member##_SHIFT)
+
+#define HINIC_RQ_CQE_STATUS_RXDONE_SHIFT        31
+
+#define HINIC_RQ_CQE_STATUS_RXDONE_MASK         0x1
+
+#define HINIC_RQ_CQE_STATUS_CSUM_ERR_SHIFT	0
+
+#define HINIC_RQ_CQE_STATUS_CSUM_ERR_MASK	0xFFFFU
+
+#define HINIC_RQ_CQE_STATUS_GET(val, member)    \
+		(((val) >> HINIC_RQ_CQE_STATUS_##member##_SHIFT) & \
+		 HINIC_RQ_CQE_STATUS_##member##_MASK)
+
+#define HINIC_RQ_CQE_STATUS_CLEAR(val, member)  \
+		((val) & (~(HINIC_RQ_CQE_STATUS_##member##_MASK << \
+		 HINIC_RQ_CQE_STATUS_##member##_SHIFT)))
+
+#define HINIC_RQ_CQE_SGE_LEN_SHIFT              16
+
+#define HINIC_RQ_CQE_SGE_LEN_MASK               0xFFFF
+
+#define HINIC_RQ_CQE_SGE_GET(val, member)       \
+		(((val) >> HINIC_RQ_CQE_SGE_##member##_SHIFT) & \
+		 HINIC_RQ_CQE_SGE_##member##_MASK)
+
+#define HINIC_RQ_CTRL_BUFDESC_SECT_LEN_SHIFT    0
+#define HINIC_RQ_CTRL_COMPLETE_FORMAT_SHIFT     15
+#define HINIC_RQ_CTRL_COMPLETE_LEN_SHIFT        27
+#define HINIC_RQ_CTRL_LEN_SHIFT                 29
+
+#define HINIC_RQ_CTRL_BUFDESC_SECT_LEN_MASK     0xFF
+#define HINIC_RQ_CTRL_COMPLETE_FORMAT_MASK      0x1
+#define HINIC_RQ_CTRL_COMPLETE_LEN_MASK         0x3
+#define HINIC_RQ_CTRL_LEN_MASK                  0x3
+
+#define HINIC_RQ_CTRL_SET(val, member)          \
+		(((u32)(val) & HINIC_RQ_CTRL_##member##_MASK) << \
+		 HINIC_RQ_CTRL_##member##_SHIFT)
+
+#define HINIC_SQ_WQE_SIZE(nr_sges)              \
+		(sizeof(struct hinic_sq_ctrl) + \
+		 sizeof(struct hinic_sq_task) + \
+		 (nr_sges) * sizeof(struct hinic_sq_bufdesc))
+
+#define HINIC_SCMD_DATA_LEN                     16
+
+#define HINIC_MAX_SQ_BUFDESCS                   17
+
+#define HINIC_SQ_WQE_MAX_SIZE                   320
+#define HINIC_RQ_WQE_SIZE                       32
+
+#define HINIC_MSS_DEFAULT		        0x3E00
+#define HINIC_MSS_MIN		                0x50
+
+#define RQ_CQE_STATUS_NUM_LRO_SHIFT		16
+#define RQ_CQE_STATUS_NUM_LRO_MASK		0xFFU
+
+#define RQ_CQE_STATUS_GET(val, member)		(((val) >> \
+			RQ_CQE_STATUS_##member##_SHIFT) & \
+			RQ_CQE_STATUS_##member##_MASK)
+
+#define HINIC_GET_RX_NUM_LRO(status)	\
+		RQ_CQE_STATUS_GET(status, NUM_LRO)
+
+#define RQ_CQE_OFFOLAD_TYPE_PKT_TYPE_SHIFT		0
+#define RQ_CQE_OFFOLAD_TYPE_PKT_TYPE_MASK		0xFFFU
+#define RQ_CQE_OFFOLAD_TYPE_VLAN_EN_SHIFT		21
+#define RQ_CQE_OFFOLAD_TYPE_VLAN_EN_MASK		0x1U
+
+#define RQ_CQE_OFFOLAD_TYPE_GET(val, member)		(((val) >> \
+				RQ_CQE_OFFOLAD_TYPE_##member##_SHIFT) & \
+				RQ_CQE_OFFOLAD_TYPE_##member##_MASK)
+
+#define HINIC_GET_RX_PKT_TYPE(offload_type)	\
+			RQ_CQE_OFFOLAD_TYPE_GET(offload_type, PKT_TYPE)
+
+#define HINIC_GET_RX_VLAN_OFFLOAD_EN(offload_type)	\
+			RQ_CQE_OFFOLAD_TYPE_GET(offload_type, VLAN_EN)
+
+#define RQ_CQE_SGE_VLAN_MASK				0xFFFFU
+#define RQ_CQE_SGE_VLAN_SHIFT				0
+
+#define RQ_CQE_SGE_GET(val, member)			(((val) >> \
+					RQ_CQE_SGE_##member##_SHIFT) & \
+					RQ_CQE_SGE_##member##_MASK)
+
+#define HINIC_GET_RX_VLAN_TAG(vlan_len)	\
+		RQ_CQE_SGE_GET(vlan_len, VLAN)
+
+#define HINIC_RSS_TYPE_VALID_SHIFT			23
+#define HINIC_RSS_TYPE_TCP_IPV6_EXT_SHIFT		24
+#define HINIC_RSS_TYPE_IPV6_EXT_SHIFT			25
+#define HINIC_RSS_TYPE_TCP_IPV6_SHIFT			26
+#define HINIC_RSS_TYPE_IPV6_SHIFT			27
+#define HINIC_RSS_TYPE_TCP_IPV4_SHIFT			28
+#define HINIC_RSS_TYPE_IPV4_SHIFT			29
+#define HINIC_RSS_TYPE_UDP_IPV6_SHIFT			30
+#define HINIC_RSS_TYPE_UDP_IPV4_SHIFT			31
+
+#define HINIC_RSS_TYPE_SET(val, member)                        \
+		(((u32)(val) & 0x1) << HINIC_RSS_TYPE_##member##_SHIFT)
+
+#define HINIC_RSS_TYPE_GET(val, member)                        \
+		(((u32)(val) >> HINIC_RSS_TYPE_##member##_SHIFT) & 0x1)
+
+enum hinic_l3_offload_type {
+	L3TYPE_UNKNOWN = 0,
+	IPV6_PKT = 1,
+	IPV4_PKT_NO_CHKSUM_OFFLOAD = 2,
+	IPV4_PKT_WITH_CHKSUM_OFFLOAD = 3,
+};
+
+enum hinic_l4_offload_type {
+	OFFLOAD_DISABLE     = 0,
+	TCP_OFFLOAD_ENABLE  = 1,
+	SCTP_OFFLOAD_ENABLE = 2,
+	UDP_OFFLOAD_ENABLE  = 3,
+};
+
+enum hinic_l4_tunnel_type {
+	NOT_TUNNEL,
+	TUNNEL_UDP_NO_CSUM,
+	TUNNEL_UDP_CSUM,
+};
+
+enum hinic_outer_l3type {
+	HINIC_OUTER_L3TYPE_UNKNOWN              = 0,
+	HINIC_OUTER_L3TYPE_IPV6                 = 1,
+	HINIC_OUTER_L3TYPE_IPV4_NO_CHKSUM       = 2,
+	HINIC_OUTER_L3TYPE_IPV4_CHKSUM          = 3,
+};
+
+enum hinic_l2type {
+	HINIC_L2TYPE_ETH = 0,
+};
+
+struct hinic_cmdq_header {
+	u32     header_info;
+	u32     saved_data;
+};
+
+struct hinic_status {
+	u32 status_info;
+};
+
+struct hinic_ctrl {
+	u32 ctrl_info;
+};
+
+struct hinic_sge_resp {
+	struct hinic_sge        sge;
+	u32                     rsvd;
+};
+
+struct hinic_cmdq_completion {
+	/* HW Format */
+	union {
+		struct hinic_sge_resp   sge_resp;
+		u64                     direct_resp;
+	};
+};
+
+struct hinic_scmd_bufdesc {
+	u32     buf_len;
+	u32     rsvd;
+	u8      data[HINIC_SCMD_DATA_LEN];
+};
+
+struct hinic_lcmd_bufdesc {
+	struct hinic_sge        sge;
+	u32                     rsvd1;
+	u64                     rsvd2;
+	u64                     rsvd3;
+};
+
+struct hinic_cmdq_wqe_scmd {
+	struct hinic_cmdq_header        header;
+	u64                             rsvd;
+	struct hinic_status             status;
+	struct hinic_ctrl               ctrl;
+	struct hinic_cmdq_completion    completion;
+	struct hinic_scmd_bufdesc       buf_desc;
+};
+
+struct hinic_cmdq_wqe_lcmd {
+	struct hinic_cmdq_header        header;
+	struct hinic_status             status;
+	struct hinic_ctrl               ctrl;
+	struct hinic_cmdq_completion    completion;
+	struct hinic_lcmd_bufdesc       buf_desc;
+};
+
+struct hinic_cmdq_direct_wqe {
+	struct hinic_cmdq_wqe_scmd      wqe_scmd;
+};
+
+struct hinic_cmdq_wqe {
+	/* HW Format */
+	union {
+		struct hinic_cmdq_direct_wqe    direct_wqe;
+		struct hinic_cmdq_wqe_lcmd      wqe_lcmd;
+	};
+};
+
+struct hinic_sq_ctrl {
+	u32     ctrl_info;
+	u32     queue_info;
+};
+
+struct hinic_sq_task {
+	u32     pkt_info0;
+	u32     pkt_info1;
+	u32     pkt_info2;
+	u32     ufo_v6_identify;
+	u32     pkt_info4;
+	u32     zero_pad;
+};
+
+struct hinic_sq_bufdesc {
+	struct hinic_sge sge;
+	u32     rsvd;
+};
+
+struct hinic_sq_wqe {
+	struct hinic_sq_ctrl            ctrl;
+	struct hinic_sq_task            task;
+	struct hinic_sq_bufdesc         buf_descs[HINIC_MAX_SQ_BUFDESCS];
+};
+
+struct hinic_rq_cqe {
+	u32     status;
+	u32     len;
+
+	u32     offload_type;
+	u32     rsvd3;
+	u32     rsvd4;
+	u32     rsvd5;
+	u32     rsvd6;
+	u32     rsvd7;
+};
+
+struct hinic_rq_ctrl {
+	u32     ctrl_info;
+};
+
+struct hinic_rq_cqe_sect {
+	struct hinic_sge        sge;
+	u32                     rsvd;
+};
+
+struct hinic_rq_bufdesc {
+	u32     hi_addr;
+	u32     lo_addr;
+};
+
+struct hinic_rq_wqe {
+	struct hinic_rq_ctrl            ctrl;
+	u32                             rsvd;
+	struct hinic_rq_cqe_sect        cqe_sect;
+	struct hinic_rq_bufdesc         buf_desc;
+};
+
+struct hinic_hw_wqe {
+	/* HW Format */
+	union {
+		struct hinic_cmdq_wqe   cmdq_wqe;
+		struct hinic_sq_wqe     sq_wqe;
+		struct hinic_rq_wqe     rq_wqe;
+	};
+};
+
+#endif
diff --git a/drivers/net/ethernet/huawei/hinic/hinic_main.c b/drivers/net/ethernet/huawei/hinic/hinic_main.c
new file mode 100644
index 000000000000..ae1f523d6841
--- /dev/null
+++ b/drivers/net/ethernet/huawei/hinic/hinic_main.c
@@ -0,0 +1,1508 @@
+// SPDX-License-Identifier: GPL-2.0-only
+/*
+ * Huawei HiNIC PCI Express Linux driver
+ * Copyright(c) 2017 Huawei Technologies Co., Ltd
+ */
+
+#include <linux/kernel.h>
+#include <linux/module.h>
+#include <linux/moduleparam.h>
+#include <linux/pci.h>
+#include <linux/device.h>
+#include <linux/errno.h>
+#include <linux/types.h>
+#include <linux/etherdevice.h>
+#include <linux/netdevice.h>
+#include <linux/slab.h>
+#include <linux/if_vlan.h>
+#include <linux/semaphore.h>
+#include <linux/workqueue.h>
+#include <net/ip.h>
+#include <net/devlink.h>
+#include <linux/bitops.h>
+#include <linux/bitmap.h>
+#include <linux/delay.h>
+#include <linux/err.h>
+
+#include "hinic_debugfs.h"
+#include "hinic_hw_qp.h"
+#include "hinic_hw_dev.h"
+#include "hinic_devlink.h"
+#include "hinic_port.h"
+#include "hinic_tx.h"
+#include "hinic_rx.h"
+#include "hinic_dev.h"
+#include "hinic_sriov.h"
+
+MODULE_AUTHOR("Huawei Technologies CO., Ltd");
+MODULE_DESCRIPTION("Huawei Intelligent NIC driver");
+MODULE_LICENSE("GPL");
+
+static unsigned int tx_weight = 64;
+module_param(tx_weight, uint, 0644);
+MODULE_PARM_DESC(tx_weight, "Number Tx packets for NAPI budget (default=64)");
+
+static unsigned int rx_weight = 64;
+module_param(rx_weight, uint, 0644);
+MODULE_PARM_DESC(rx_weight, "Number Rx packets for NAPI budget (default=64)");
+
+#define HINIC_DEV_ID_QUAD_PORT_25GE         0x1822
+#define HINIC_DEV_ID_DUAL_PORT_100GE        0x0200
+#define HINIC_DEV_ID_DUAL_PORT_100GE_MEZZ   0x0205
+#define HINIC_DEV_ID_QUAD_PORT_25GE_MEZZ    0x0210
+#define HINIC_DEV_ID_VF    0x375e
+
+#define HINIC_WQ_NAME                   "hinic_dev"
+
+#define MSG_ENABLE_DEFAULT              (NETIF_MSG_DRV | NETIF_MSG_PROBE | \
+					 NETIF_MSG_IFUP |                  \
+					 NETIF_MSG_TX_ERR | NETIF_MSG_RX_ERR)
+
+#define HINIC_LRO_MAX_WQE_NUM_DEFAULT	8
+
+#define HINIC_LRO_RX_TIMER_DEFAULT	16
+
+#define work_to_rx_mode_work(work)      \
+		container_of(work, struct hinic_rx_mode_work, work)
+
+#define rx_mode_work_to_nic_dev(rx_mode_work) \
+		container_of(rx_mode_work, struct hinic_dev, rx_mode_work)
+
+#define HINIC_WAIT_SRIOV_CFG_TIMEOUT	15000
+
+#define HINIC_DEAULT_TXRX_MSIX_PENDING_LIMIT		2
+#define HINIC_DEAULT_TXRX_MSIX_COALESC_TIMER_CFG	32
+#define HINIC_DEAULT_TXRX_MSIX_RESEND_TIMER_CFG		7
+
+static int change_mac_addr(struct net_device *netdev, const u8 *addr);
+
+static int set_features(struct hinic_dev *nic_dev,
+			netdev_features_t pre_features,
+			netdev_features_t features, bool force_change);
+
+static void gather_rx_stats(struct hinic_rxq_stats *nic_rx_stats, struct hinic_rxq *rxq)
+{
+	struct hinic_rxq_stats rx_stats;
+
+	hinic_rxq_get_stats(rxq, &rx_stats);
+
+	nic_rx_stats->bytes += rx_stats.bytes;
+	nic_rx_stats->pkts  += rx_stats.pkts;
+	nic_rx_stats->errors += rx_stats.errors;
+	nic_rx_stats->csum_errors += rx_stats.csum_errors;
+	nic_rx_stats->other_errors += rx_stats.other_errors;
+}
+
+static void gather_tx_stats(struct hinic_txq_stats *nic_tx_stats, struct hinic_txq *txq)
+{
+	struct hinic_txq_stats tx_stats;
+
+	hinic_txq_get_stats(txq, &tx_stats);
+
+	nic_tx_stats->bytes += tx_stats.bytes;
+	nic_tx_stats->pkts += tx_stats.pkts;
+	nic_tx_stats->tx_busy += tx_stats.tx_busy;
+	nic_tx_stats->tx_wake += tx_stats.tx_wake;
+	nic_tx_stats->tx_dropped += tx_stats.tx_dropped;
+	nic_tx_stats->big_frags_pkts += tx_stats.big_frags_pkts;
+}
+
+static void gather_nic_stats(struct hinic_dev *nic_dev,
+			     struct hinic_rxq_stats *nic_rx_stats,
+			     struct hinic_txq_stats *nic_tx_stats)
+{
+	int i, num_qps = hinic_hwdev_num_qps(nic_dev->hwdev);
+
+	for (i = 0; i < num_qps; i++)
+		gather_rx_stats(nic_rx_stats, &nic_dev->rxqs[i]);
+
+	for (i = 0; i < num_qps; i++)
+		gather_tx_stats(nic_tx_stats, &nic_dev->txqs[i]);
+}
+
+/**
+ * create_txqs - Create the Logical Tx Queues of specific NIC device
+ * @nic_dev: the specific NIC device
+ *
+ * Return 0 - Success, negative - Failure
+ **/
+static int create_txqs(struct hinic_dev *nic_dev)
+{
+	int err, i, j, num_txqs = hinic_hwdev_num_qps(nic_dev->hwdev);
+	struct net_device *netdev = nic_dev->netdev;
+
+	if (nic_dev->txqs)
+		return -EINVAL;
+
+	nic_dev->txqs = devm_kcalloc(&netdev->dev, num_txqs,
+				     sizeof(*nic_dev->txqs), GFP_KERNEL);
+	if (!nic_dev->txqs)
+		return -ENOMEM;
+
+	hinic_sq_dbgfs_init(nic_dev);
+
+	for (i = 0; i < num_txqs; i++) {
+		struct hinic_sq *sq = hinic_hwdev_get_sq(nic_dev->hwdev, i);
+
+		err = hinic_init_txq(&nic_dev->txqs[i], sq, netdev);
+		if (err) {
+			netif_err(nic_dev, drv, netdev,
+				  "Failed to init Txq\n");
+			goto err_init_txq;
+		}
+
+		err = hinic_sq_debug_add(nic_dev, i);
+		if (err) {
+			netif_err(nic_dev, drv, netdev,
+				  "Failed to add SQ%d debug\n", i);
+			goto err_add_sq_dbg;
+		}
+	}
+
+	return 0;
+
+err_add_sq_dbg:
+	hinic_clean_txq(&nic_dev->txqs[i]);
+err_init_txq:
+	for (j = 0; j < i; j++) {
+		hinic_sq_debug_rem(nic_dev->txqs[j].sq);
+		hinic_clean_txq(&nic_dev->txqs[j]);
+	}
+
+	hinic_sq_dbgfs_uninit(nic_dev);
+
+	devm_kfree(&netdev->dev, nic_dev->txqs);
+	nic_dev->txqs = NULL;
+	return err;
+}
+
+static void enable_txqs_napi(struct hinic_dev *nic_dev)
+{
+	int num_txqs = hinic_hwdev_num_qps(nic_dev->hwdev);
+	int i;
+
+	for (i = 0; i < num_txqs; i++)
+		napi_enable(&nic_dev->txqs[i].napi);
+}
+
+static void disable_txqs_napi(struct hinic_dev *nic_dev)
+{
+	int num_txqs = hinic_hwdev_num_qps(nic_dev->hwdev);
+	int i;
+
+	for (i = 0; i < num_txqs; i++)
+		napi_disable(&nic_dev->txqs[i].napi);
+}
+
+/**
+ * free_txqs - Free the Logical Tx Queues of specific NIC device
+ * @nic_dev: the specific NIC device
+ **/
+static void free_txqs(struct hinic_dev *nic_dev)
+{
+	int i, num_txqs = hinic_hwdev_num_qps(nic_dev->hwdev);
+	struct net_device *netdev = nic_dev->netdev;
+
+	if (!nic_dev->txqs)
+		return;
+
+	for (i = 0; i < num_txqs; i++) {
+		hinic_sq_debug_rem(nic_dev->txqs[i].sq);
+		hinic_clean_txq(&nic_dev->txqs[i]);
+	}
+
+	hinic_sq_dbgfs_uninit(nic_dev);
+
+	devm_kfree(&netdev->dev, nic_dev->txqs);
+	nic_dev->txqs = NULL;
+}
+
+/**
+ * create_rxqs - Create the Logical Rx Queues of specific NIC device
+ * @nic_dev: the specific NIC device
+ *
+ * Return 0 - Success, negative - Failure
+ **/
+static int create_rxqs(struct hinic_dev *nic_dev)
+{
+	int err, i, j, num_rxqs = hinic_hwdev_num_qps(nic_dev->hwdev);
+	struct net_device *netdev = nic_dev->netdev;
+
+	if (nic_dev->rxqs)
+		return -EINVAL;
+
+	nic_dev->rxqs = devm_kcalloc(&netdev->dev, num_rxqs,
+				     sizeof(*nic_dev->rxqs), GFP_KERNEL);
+	if (!nic_dev->rxqs)
+		return -ENOMEM;
+
+	hinic_rq_dbgfs_init(nic_dev);
+
+	for (i = 0; i < num_rxqs; i++) {
+		struct hinic_rq *rq = hinic_hwdev_get_rq(nic_dev->hwdev, i);
+
+		err = hinic_init_rxq(&nic_dev->rxqs[i], rq, netdev);
+		if (err) {
+			netif_err(nic_dev, drv, netdev,
+				  "Failed to init rxq\n");
+			goto err_init_rxq;
+		}
+
+		err = hinic_rq_debug_add(nic_dev, i);
+		if (err) {
+			netif_err(nic_dev, drv, netdev,
+				  "Failed to add RQ%d debug\n", i);
+			goto err_add_rq_dbg;
+		}
+	}
+
+	return 0;
+
+err_add_rq_dbg:
+	hinic_clean_rxq(&nic_dev->rxqs[i]);
+err_init_rxq:
+	for (j = 0; j < i; j++) {
+		hinic_rq_debug_rem(nic_dev->rxqs[j].rq);
+		hinic_clean_rxq(&nic_dev->rxqs[j]);
+	}
+
+	hinic_rq_dbgfs_uninit(nic_dev);
+
+	devm_kfree(&netdev->dev, nic_dev->rxqs);
+	nic_dev->rxqs = NULL;
+	return err;
+}
+
+/**
+ * free_rxqs - Free the Logical Rx Queues of specific NIC device
+ * @nic_dev: the specific NIC device
+ **/
+static void free_rxqs(struct hinic_dev *nic_dev)
+{
+	int i, num_rxqs = hinic_hwdev_num_qps(nic_dev->hwdev);
+	struct net_device *netdev = nic_dev->netdev;
+
+	if (!nic_dev->rxqs)
+		return;
+
+	for (i = 0; i < num_rxqs; i++) {
+		hinic_rq_debug_rem(nic_dev->rxqs[i].rq);
+		hinic_clean_rxq(&nic_dev->rxqs[i]);
+	}
+
+	hinic_rq_dbgfs_uninit(nic_dev);
+
+	devm_kfree(&netdev->dev, nic_dev->rxqs);
+	nic_dev->rxqs = NULL;
+}
+
+static int hinic_configure_max_qnum(struct hinic_dev *nic_dev)
+{
+	return hinic_set_max_qnum(nic_dev, nic_dev->hwdev->nic_cap.max_qps);
+}
+
+static int hinic_rss_init(struct hinic_dev *nic_dev)
+{
+	u8 default_rss_key[HINIC_RSS_KEY_SIZE];
+	u8 tmpl_idx = nic_dev->rss_tmpl_idx;
+	u32 *indir_tbl;
+	int err, i;
+
+	indir_tbl = kcalloc(HINIC_RSS_INDIR_SIZE, sizeof(u32), GFP_KERNEL);
+	if (!indir_tbl)
+		return -ENOMEM;
+
+	netdev_rss_key_fill(default_rss_key, sizeof(default_rss_key));
+	for (i = 0; i < HINIC_RSS_INDIR_SIZE; i++)
+		indir_tbl[i] = ethtool_rxfh_indir_default(i, nic_dev->num_rss);
+
+	err = hinic_rss_set_template_tbl(nic_dev, tmpl_idx, default_rss_key);
+	if (err)
+		goto out;
+
+	err = hinic_rss_set_indir_tbl(nic_dev, tmpl_idx, indir_tbl);
+	if (err)
+		goto out;
+
+	err = hinic_set_rss_type(nic_dev, tmpl_idx, nic_dev->rss_type);
+	if (err)
+		goto out;
+
+	err = hinic_rss_set_hash_engine(nic_dev, tmpl_idx,
+					nic_dev->rss_hash_engine);
+	if (err)
+		goto out;
+
+	err = hinic_rss_cfg(nic_dev, 1, tmpl_idx);
+	if (err)
+		goto out;
+
+out:
+	kfree(indir_tbl);
+	return err;
+}
+
+static void hinic_rss_deinit(struct hinic_dev *nic_dev)
+{
+	hinic_rss_cfg(nic_dev, 0, nic_dev->rss_tmpl_idx);
+}
+
+static void hinic_init_rss_parameters(struct hinic_dev *nic_dev)
+{
+	nic_dev->rss_hash_engine = HINIC_RSS_HASH_ENGINE_TYPE_XOR;
+	nic_dev->rss_type.tcp_ipv6_ext = 1;
+	nic_dev->rss_type.ipv6_ext = 1;
+	nic_dev->rss_type.tcp_ipv6 = 1;
+	nic_dev->rss_type.ipv6 = 1;
+	nic_dev->rss_type.tcp_ipv4 = 1;
+	nic_dev->rss_type.ipv4 = 1;
+	nic_dev->rss_type.udp_ipv6 = 1;
+	nic_dev->rss_type.udp_ipv4 = 1;
+}
+
+static void hinic_enable_rss(struct hinic_dev *nic_dev)
+{
+	struct net_device *netdev = nic_dev->netdev;
+	struct hinic_hwdev *hwdev = nic_dev->hwdev;
+	struct hinic_hwif *hwif = hwdev->hwif;
+	struct pci_dev *pdev = hwif->pdev;
+	int i, node, err = 0;
+	u16 num_cpus = 0;
+
+	if (nic_dev->max_qps <= 1) {
+		nic_dev->flags &= ~HINIC_RSS_ENABLE;
+		nic_dev->rss_limit = nic_dev->max_qps;
+		nic_dev->num_qps = nic_dev->max_qps;
+		nic_dev->num_rss = nic_dev->max_qps;
+
+		return;
+	}
+
+	err = hinic_rss_template_alloc(nic_dev, &nic_dev->rss_tmpl_idx);
+	if (err) {
+		netif_err(nic_dev, drv, netdev,
+			  "Failed to alloc tmpl_idx for rss, can't enable rss for this function\n");
+		nic_dev->flags &= ~HINIC_RSS_ENABLE;
+		nic_dev->max_qps = 1;
+		nic_dev->rss_limit = nic_dev->max_qps;
+		nic_dev->num_qps = nic_dev->max_qps;
+		nic_dev->num_rss = nic_dev->max_qps;
+
+		return;
+	}
+
+	nic_dev->flags |= HINIC_RSS_ENABLE;
+
+	for (i = 0; i < num_online_cpus(); i++) {
+		node = cpu_to_node(i);
+		if (node == dev_to_node(&pdev->dev))
+			num_cpus++;
+	}
+
+	if (!num_cpus)
+		num_cpus = num_online_cpus();
+
+	nic_dev->num_qps = hinic_hwdev_num_qps(hwdev);
+	nic_dev->num_qps = min_t(u16, nic_dev->num_qps, num_cpus);
+
+	nic_dev->rss_limit = nic_dev->num_qps;
+	nic_dev->num_rss = nic_dev->num_qps;
+
+	hinic_init_rss_parameters(nic_dev);
+	err = hinic_rss_init(nic_dev);
+	if (err)
+		netif_err(nic_dev, drv, netdev, "Failed to init rss\n");
+}
+
+int hinic_open(struct net_device *netdev)
+{
+	struct hinic_dev *nic_dev = netdev_priv(netdev);
+	enum hinic_port_link_state link_state;
+	int err, ret;
+
+	if (!(nic_dev->flags & HINIC_INTF_UP)) {
+		err = hinic_hwdev_ifup(nic_dev->hwdev, nic_dev->sq_depth,
+				       nic_dev->rq_depth);
+		if (err) {
+			netif_err(nic_dev, drv, netdev,
+				  "Failed - HW interface up\n");
+			return err;
+		}
+	}
+
+	err = create_txqs(nic_dev);
+	if (err) {
+		netif_err(nic_dev, drv, netdev,
+			  "Failed to create Tx queues\n");
+		goto err_create_txqs;
+	}
+
+	enable_txqs_napi(nic_dev);
+
+	err = create_rxqs(nic_dev);
+	if (err) {
+		netif_err(nic_dev, drv, netdev,
+			  "Failed to create Rx queues\n");
+		goto err_create_rxqs;
+	}
+
+	hinic_enable_rss(nic_dev);
+
+	err = hinic_configure_max_qnum(nic_dev);
+	if (err) {
+		netif_err(nic_dev, drv, nic_dev->netdev,
+			  "Failed to configure the maximum number of queues\n");
+		goto err_port_state;
+	}
+
+	netif_set_real_num_tx_queues(netdev, nic_dev->num_qps);
+	netif_set_real_num_rx_queues(netdev, nic_dev->num_qps);
+
+	err = hinic_port_set_state(nic_dev, HINIC_PORT_ENABLE);
+	if (err) {
+		netif_err(nic_dev, drv, netdev,
+			  "Failed to set port state\n");
+		goto err_port_state;
+	}
+
+	err = hinic_port_set_func_state(nic_dev, HINIC_FUNC_PORT_ENABLE);
+	if (err) {
+		netif_err(nic_dev, drv, netdev,
+			  "Failed to set func port state\n");
+		goto err_func_port_state;
+	}
+
+	down(&nic_dev->mgmt_lock);
+
+	err = hinic_port_link_state(nic_dev, &link_state);
+	if (err) {
+		netif_err(nic_dev, drv, netdev, "Failed to get link state\n");
+		goto err_port_link;
+	}
+
+	if (!HINIC_IS_VF(nic_dev->hwdev->hwif))
+		hinic_notify_all_vfs_link_changed(nic_dev->hwdev, link_state);
+
+	if (link_state == HINIC_LINK_STATE_UP) {
+		nic_dev->flags |= HINIC_LINK_UP;
+		nic_dev->cable_unplugged = false;
+		nic_dev->module_unrecognized = false;
+	}
+
+	nic_dev->flags |= HINIC_INTF_UP;
+
+	if ((nic_dev->flags & (HINIC_LINK_UP | HINIC_INTF_UP)) ==
+	    (HINIC_LINK_UP | HINIC_INTF_UP)) {
+		netif_info(nic_dev, drv, netdev, "link + intf UP\n");
+		netif_carrier_on(netdev);
+		netif_tx_wake_all_queues(netdev);
+	}
+
+	up(&nic_dev->mgmt_lock);
+
+	netif_info(nic_dev, drv, netdev, "HINIC_INTF is UP\n");
+	return 0;
+
+err_port_link:
+	up(&nic_dev->mgmt_lock);
+	ret = hinic_port_set_func_state(nic_dev, HINIC_FUNC_PORT_DISABLE);
+	if (ret)
+		netif_warn(nic_dev, drv, netdev,
+			   "Failed to revert func port state\n");
+
+err_func_port_state:
+	ret = hinic_port_set_state(nic_dev, HINIC_PORT_DISABLE);
+	if (ret)
+		netif_warn(nic_dev, drv, netdev,
+			   "Failed to revert port state\n");
+err_port_state:
+	free_rxqs(nic_dev);
+	if (nic_dev->flags & HINIC_RSS_ENABLE) {
+		hinic_rss_deinit(nic_dev);
+		hinic_rss_template_free(nic_dev, nic_dev->rss_tmpl_idx);
+	}
+
+err_create_rxqs:
+	disable_txqs_napi(nic_dev);
+	free_txqs(nic_dev);
+
+err_create_txqs:
+	if (!(nic_dev->flags & HINIC_INTF_UP))
+		hinic_hwdev_ifdown(nic_dev->hwdev);
+	return err;
+}
+
+int hinic_close(struct net_device *netdev)
+{
+	struct hinic_dev *nic_dev = netdev_priv(netdev);
+	unsigned int flags;
+
+	/* Disable txq napi firstly to aviod rewaking txq in free_tx_poll */
+	disable_txqs_napi(nic_dev);
+
+	down(&nic_dev->mgmt_lock);
+
+	flags = nic_dev->flags;
+	nic_dev->flags &= ~HINIC_INTF_UP;
+
+	netif_carrier_off(netdev);
+	netif_tx_disable(netdev);
+
+	up(&nic_dev->mgmt_lock);
+
+	if (!HINIC_IS_VF(nic_dev->hwdev->hwif))
+		hinic_notify_all_vfs_link_changed(nic_dev->hwdev, 0);
+
+	hinic_port_set_state(nic_dev, HINIC_PORT_DISABLE);
+
+	hinic_port_set_func_state(nic_dev, HINIC_FUNC_PORT_DISABLE);
+
+	if (nic_dev->flags & HINIC_RSS_ENABLE) {
+		hinic_rss_deinit(nic_dev);
+		hinic_rss_template_free(nic_dev, nic_dev->rss_tmpl_idx);
+	}
+
+	free_rxqs(nic_dev);
+	free_txqs(nic_dev);
+
+	if (flags & HINIC_INTF_UP)
+		hinic_hwdev_ifdown(nic_dev->hwdev);
+
+	netif_info(nic_dev, drv, netdev, "HINIC_INTF is DOWN\n");
+	return 0;
+}
+
+static int hinic_change_mtu(struct net_device *netdev, int new_mtu)
+{
+	struct hinic_dev *nic_dev = netdev_priv(netdev);
+	int err;
+
+	netif_info(nic_dev, drv, netdev, "set_mtu = %d\n", new_mtu);
+
+	err = hinic_port_set_mtu(nic_dev, new_mtu);
+	if (err)
+		netif_err(nic_dev, drv, netdev, "Failed to set port mtu\n");
+	else
+		WRITE_ONCE(netdev->mtu, new_mtu);
+
+	return err;
+}
+
+/**
+ * change_mac_addr - change the main mac address of network device
+ * @netdev: network device
+ * @addr: mac address to set
+ *
+ * Return 0 - Success, negative - Failure
+ **/
+static int change_mac_addr(struct net_device *netdev, const u8 *addr)
+{
+	struct hinic_dev *nic_dev = netdev_priv(netdev);
+	u16 vid = 0;
+	int err;
+
+	if (!is_valid_ether_addr(addr))
+		return -EADDRNOTAVAIL;
+
+	netif_info(nic_dev, drv, netdev, "change mac addr = %02x %02x %02x %02x %02x %02x\n",
+		   addr[0], addr[1], addr[2], addr[3], addr[4], addr[5]);
+
+	down(&nic_dev->mgmt_lock);
+
+	do {
+		err = hinic_port_del_mac(nic_dev, netdev->dev_addr, vid);
+		if (err) {
+			netif_err(nic_dev, drv, netdev,
+				  "Failed to delete mac\n");
+			break;
+		}
+
+		err = hinic_port_add_mac(nic_dev, addr, vid);
+		if (err) {
+			netif_err(nic_dev, drv, netdev, "Failed to add mac\n");
+			break;
+		}
+
+		vid = find_next_bit(nic_dev->vlan_bitmap, VLAN_N_VID, vid + 1);
+	} while (vid != VLAN_N_VID);
+
+	up(&nic_dev->mgmt_lock);
+	return err;
+}
+
+static int hinic_set_mac_addr(struct net_device *netdev, void *addr)
+{
+	unsigned char new_mac[ETH_ALEN];
+	struct sockaddr *saddr = addr;
+	int err;
+
+	memcpy(new_mac, saddr->sa_data, ETH_ALEN);
+
+	err = change_mac_addr(netdev, new_mac);
+	if (!err)
+		eth_hw_addr_set(netdev, new_mac);
+
+	return err;
+}
+
+/**
+ * add_mac_addr - add mac address to network device
+ * @netdev: network device
+ * @addr: mac address to add
+ *
+ * Return 0 - Success, negative - Failure
+ **/
+static int add_mac_addr(struct net_device *netdev, const u8 *addr)
+{
+	struct hinic_dev *nic_dev = netdev_priv(netdev);
+	u16 vid = 0;
+	int err;
+
+	netif_info(nic_dev, drv, netdev, "set mac addr = %02x %02x %02x %02x %02x %02x\n",
+		   addr[0], addr[1], addr[2], addr[3], addr[4], addr[5]);
+
+	down(&nic_dev->mgmt_lock);
+
+	do {
+		err = hinic_port_add_mac(nic_dev, addr, vid);
+		if (err) {
+			netif_err(nic_dev, drv, netdev, "Failed to add mac\n");
+			break;
+		}
+
+		vid = find_next_bit(nic_dev->vlan_bitmap, VLAN_N_VID, vid + 1);
+	} while (vid != VLAN_N_VID);
+
+	up(&nic_dev->mgmt_lock);
+	return err;
+}
+
+/**
+ * remove_mac_addr - remove mac address from network device
+ * @netdev: network device
+ * @addr: mac address to remove
+ *
+ * Return 0 - Success, negative - Failure
+ **/
+static int remove_mac_addr(struct net_device *netdev, const u8 *addr)
+{
+	struct hinic_dev *nic_dev = netdev_priv(netdev);
+	u16 vid = 0;
+	int err;
+
+	if (!is_valid_ether_addr(addr))
+		return -EADDRNOTAVAIL;
+
+	netif_info(nic_dev, drv, netdev, "remove mac addr = %02x %02x %02x %02x %02x %02x\n",
+		   addr[0], addr[1], addr[2], addr[3], addr[4], addr[5]);
+
+	down(&nic_dev->mgmt_lock);
+
+	do {
+		err = hinic_port_del_mac(nic_dev, addr, vid);
+		if (err) {
+			netif_err(nic_dev, drv, netdev,
+				  "Failed to delete mac\n");
+			break;
+		}
+
+		vid = find_next_bit(nic_dev->vlan_bitmap, VLAN_N_VID, vid + 1);
+	} while (vid != VLAN_N_VID);
+
+	up(&nic_dev->mgmt_lock);
+	return err;
+}
+
+static int hinic_vlan_rx_add_vid(struct net_device *netdev,
+				 __always_unused __be16 proto, u16 vid)
+{
+	struct hinic_dev *nic_dev = netdev_priv(netdev);
+	int ret, err;
+
+	netif_info(nic_dev, drv, netdev, "add vid = %d\n", vid);
+
+	down(&nic_dev->mgmt_lock);
+
+	err = hinic_port_add_vlan(nic_dev, vid);
+	if (err) {
+		netif_err(nic_dev, drv, netdev, "Failed to add vlan\n");
+		goto err_vlan_add;
+	}
+
+	err = hinic_port_add_mac(nic_dev, netdev->dev_addr, vid);
+	if (err && err != HINIC_PF_SET_VF_ALREADY) {
+		netif_err(nic_dev, drv, netdev, "Failed to set mac\n");
+		goto err_add_mac;
+	}
+
+	bitmap_set(nic_dev->vlan_bitmap, vid, 1);
+
+	up(&nic_dev->mgmt_lock);
+	return 0;
+
+err_add_mac:
+	ret = hinic_port_del_vlan(nic_dev, vid);
+	if (ret)
+		netif_err(nic_dev, drv, netdev,
+			  "Failed to revert by removing vlan\n");
+
+err_vlan_add:
+	up(&nic_dev->mgmt_lock);
+	return err;
+}
+
+static int hinic_vlan_rx_kill_vid(struct net_device *netdev,
+				  __always_unused __be16 proto, u16 vid)
+{
+	struct hinic_dev *nic_dev = netdev_priv(netdev);
+	int err;
+
+	netif_info(nic_dev, drv, netdev, "remove vid = %d\n", vid);
+
+	down(&nic_dev->mgmt_lock);
+
+	err = hinic_port_del_vlan(nic_dev, vid);
+	if (err) {
+		netif_err(nic_dev, drv, netdev, "Failed to delete vlan\n");
+		goto err_del_vlan;
+	}
+
+	bitmap_clear(nic_dev->vlan_bitmap, vid, 1);
+
+	up(&nic_dev->mgmt_lock);
+	return 0;
+
+err_del_vlan:
+	up(&nic_dev->mgmt_lock);
+	return err;
+}
+
+static void set_rx_mode(struct work_struct *work)
+{
+	struct hinic_rx_mode_work *rx_mode_work = work_to_rx_mode_work(work);
+	struct hinic_dev *nic_dev = rx_mode_work_to_nic_dev(rx_mode_work);
+
+	hinic_port_set_rx_mode(nic_dev, rx_mode_work->rx_mode);
+
+	__dev_uc_sync(nic_dev->netdev, add_mac_addr, remove_mac_addr);
+	__dev_mc_sync(nic_dev->netdev, add_mac_addr, remove_mac_addr);
+}
+
+static void hinic_set_rx_mode(struct net_device *netdev)
+{
+	struct hinic_dev *nic_dev = netdev_priv(netdev);
+	struct hinic_rx_mode_work *rx_mode_work;
+	u32 rx_mode;
+
+	rx_mode_work = &nic_dev->rx_mode_work;
+
+	rx_mode = HINIC_RX_MODE_UC |
+		  HINIC_RX_MODE_MC |
+		  HINIC_RX_MODE_BC;
+
+	if (netdev->flags & IFF_PROMISC) {
+		if (!HINIC_IS_VF(nic_dev->hwdev->hwif))
+			rx_mode |= HINIC_RX_MODE_PROMISC;
+	} else if (netdev->flags & IFF_ALLMULTI) {
+		rx_mode |= HINIC_RX_MODE_MC_ALL;
+	}
+
+	rx_mode_work->rx_mode = rx_mode;
+
+	queue_work(nic_dev->workq, &rx_mode_work->work);
+}
+
+static void hinic_tx_timeout(struct net_device *netdev, unsigned int txqueue)
+{
+	struct hinic_dev *nic_dev = netdev_priv(netdev);
+	u16 sw_pi, hw_ci, sw_ci;
+	struct hinic_sq *sq;
+	u16 num_sqs, q_id;
+
+	num_sqs = hinic_hwdev_num_qps(nic_dev->hwdev);
+
+	netif_err(nic_dev, drv, netdev, "Tx timeout\n");
+
+	for (q_id = 0; q_id < num_sqs; q_id++) {
+		if (!netif_xmit_stopped(netdev_get_tx_queue(netdev, q_id)))
+			continue;
+
+		sq = hinic_hwdev_get_sq(nic_dev->hwdev, q_id);
+		sw_pi = atomic_read(&sq->wq->prod_idx) & sq->wq->mask;
+		hw_ci = be16_to_cpu(*(u16 *)(sq->hw_ci_addr)) & sq->wq->mask;
+		sw_ci = atomic_read(&sq->wq->cons_idx) & sq->wq->mask;
+		netif_err(nic_dev, drv, netdev, "Txq%d: sw_pi: %d, hw_ci: %d, sw_ci: %d, napi->state: 0x%lx\n",
+			  q_id, sw_pi, hw_ci, sw_ci,
+			  nic_dev->txqs[q_id].napi.state);
+	}
+}
+
+static void hinic_get_stats64(struct net_device *netdev,
+			      struct rtnl_link_stats64 *stats)
+{
+	struct hinic_dev *nic_dev = netdev_priv(netdev);
+	struct hinic_rxq_stats nic_rx_stats = {};
+	struct hinic_txq_stats nic_tx_stats = {};
+
+	if (nic_dev->flags & HINIC_INTF_UP)
+		gather_nic_stats(nic_dev, &nic_rx_stats, &nic_tx_stats);
+
+	stats->rx_bytes   = nic_rx_stats.bytes;
+	stats->rx_packets = nic_rx_stats.pkts;
+	stats->rx_errors  = nic_rx_stats.errors;
+
+	stats->tx_bytes   = nic_tx_stats.bytes;
+	stats->tx_packets = nic_tx_stats.pkts;
+	stats->tx_errors  = nic_tx_stats.tx_dropped;
+}
+
+static int hinic_set_features(struct net_device *netdev,
+			      netdev_features_t features)
+{
+	struct hinic_dev *nic_dev = netdev_priv(netdev);
+
+	return set_features(nic_dev, nic_dev->netdev->features,
+			    features, false);
+}
+
+static netdev_features_t hinic_fix_features(struct net_device *netdev,
+					    netdev_features_t features)
+{
+	struct hinic_dev *nic_dev = netdev_priv(netdev);
+
+	/* If Rx checksum is disabled, then LRO should also be disabled */
+	if (!(features & NETIF_F_RXCSUM)) {
+		netif_info(nic_dev, drv, netdev, "disabling LRO as RXCSUM is off\n");
+		features &= ~NETIF_F_LRO;
+	}
+
+	return features;
+}
+
+static const struct net_device_ops hinic_netdev_ops = {
+	.ndo_open = hinic_open,
+	.ndo_stop = hinic_close,
+	.ndo_change_mtu = hinic_change_mtu,
+	.ndo_set_mac_address = hinic_set_mac_addr,
+	.ndo_validate_addr = eth_validate_addr,
+	.ndo_vlan_rx_add_vid = hinic_vlan_rx_add_vid,
+	.ndo_vlan_rx_kill_vid = hinic_vlan_rx_kill_vid,
+	.ndo_set_rx_mode = hinic_set_rx_mode,
+	.ndo_start_xmit = hinic_xmit_frame,
+	.ndo_tx_timeout = hinic_tx_timeout,
+	.ndo_get_stats64 = hinic_get_stats64,
+	.ndo_fix_features = hinic_fix_features,
+	.ndo_set_features = hinic_set_features,
+	.ndo_set_vf_mac	= hinic_ndo_set_vf_mac,
+	.ndo_set_vf_vlan = hinic_ndo_set_vf_vlan,
+	.ndo_get_vf_config = hinic_ndo_get_vf_config,
+	.ndo_set_vf_trust = hinic_ndo_set_vf_trust,
+	.ndo_set_vf_rate = hinic_ndo_set_vf_bw,
+	.ndo_set_vf_spoofchk = hinic_ndo_set_vf_spoofchk,
+	.ndo_set_vf_link_state = hinic_ndo_set_vf_link_state,
+};
+
+static const struct net_device_ops hinicvf_netdev_ops = {
+	.ndo_open = hinic_open,
+	.ndo_stop = hinic_close,
+	.ndo_change_mtu = hinic_change_mtu,
+	.ndo_set_mac_address = hinic_set_mac_addr,
+	.ndo_validate_addr = eth_validate_addr,
+	.ndo_vlan_rx_add_vid = hinic_vlan_rx_add_vid,
+	.ndo_vlan_rx_kill_vid = hinic_vlan_rx_kill_vid,
+	.ndo_set_rx_mode = hinic_set_rx_mode,
+	.ndo_start_xmit = hinic_xmit_frame,
+	.ndo_tx_timeout = hinic_tx_timeout,
+	.ndo_get_stats64 = hinic_get_stats64,
+	.ndo_fix_features = hinic_fix_features,
+	.ndo_set_features = hinic_set_features,
+};
+
+static void netdev_features_init(struct net_device *netdev)
+{
+	netdev->hw_features = NETIF_F_SG | NETIF_F_HIGHDMA | NETIF_F_IP_CSUM |
+			      NETIF_F_IPV6_CSUM | NETIF_F_TSO | NETIF_F_TSO6 |
+			      NETIF_F_RXCSUM | NETIF_F_LRO |
+			      NETIF_F_HW_VLAN_CTAG_TX | NETIF_F_HW_VLAN_CTAG_RX |
+			      NETIF_F_GSO_UDP_TUNNEL | NETIF_F_GSO_UDP_TUNNEL_CSUM;
+
+	netdev->vlan_features = netdev->hw_features;
+
+	netdev->features = netdev->hw_features | NETIF_F_HW_VLAN_CTAG_FILTER;
+
+	netdev->hw_enc_features = NETIF_F_IP_CSUM | NETIF_F_IPV6_CSUM | NETIF_F_SCTP_CRC |
+				  NETIF_F_SG | NETIF_F_TSO | NETIF_F_TSO6 | NETIF_F_TSO_ECN |
+				  NETIF_F_GSO_UDP_TUNNEL_CSUM | NETIF_F_GSO_UDP_TUNNEL;
+}
+
+static void hinic_refresh_nic_cfg(struct hinic_dev *nic_dev)
+{
+	struct hinic_nic_cfg *nic_cfg = &nic_dev->hwdev->func_to_io.nic_cfg;
+	struct hinic_pause_config pause_info = {0};
+	struct hinic_port_cap port_cap = {0};
+
+	if (hinic_port_get_cap(nic_dev, &port_cap))
+		return;
+
+	mutex_lock(&nic_cfg->cfg_mutex);
+	if (nic_cfg->pause_set || !port_cap.autoneg_state) {
+		nic_cfg->auto_neg = port_cap.autoneg_state;
+		pause_info.auto_neg = nic_cfg->auto_neg;
+		pause_info.rx_pause = nic_cfg->rx_pause;
+		pause_info.tx_pause = nic_cfg->tx_pause;
+		hinic_set_hw_pause_info(nic_dev->hwdev, &pause_info);
+	}
+	mutex_unlock(&nic_cfg->cfg_mutex);
+}
+
+/**
+ * link_status_event_handler - link event handler
+ * @handle: nic device for the handler
+ * @buf_in: input buffer
+ * @in_size: input size
+ * @buf_out: output buffer
+ * @out_size: returned output size
+ **/
+static void link_status_event_handler(void *handle, void *buf_in, u16 in_size,
+				      void *buf_out, u16 *out_size)
+{
+	struct hinic_port_link_status *link_status, *ret_link_status;
+	struct hinic_dev *nic_dev = handle;
+
+	link_status = buf_in;
+
+	if (link_status->link == HINIC_LINK_STATE_UP) {
+		down(&nic_dev->mgmt_lock);
+
+		nic_dev->flags |= HINIC_LINK_UP;
+		nic_dev->cable_unplugged = false;
+		nic_dev->module_unrecognized = false;
+
+		if ((nic_dev->flags & (HINIC_LINK_UP | HINIC_INTF_UP)) ==
+		    (HINIC_LINK_UP | HINIC_INTF_UP)) {
+			netif_carrier_on(nic_dev->netdev);
+			netif_tx_wake_all_queues(nic_dev->netdev);
+		}
+
+		up(&nic_dev->mgmt_lock);
+
+		if (!HINIC_IS_VF(nic_dev->hwdev->hwif))
+			hinic_refresh_nic_cfg(nic_dev);
+
+		netif_info(nic_dev, drv, nic_dev->netdev, "HINIC_Link is UP\n");
+	} else {
+		down(&nic_dev->mgmt_lock);
+
+		nic_dev->flags &= ~HINIC_LINK_UP;
+
+		netif_carrier_off(nic_dev->netdev);
+		netif_tx_disable(nic_dev->netdev);
+
+		up(&nic_dev->mgmt_lock);
+
+		netif_info(nic_dev, drv, nic_dev->netdev, "HINIC_Link is DOWN\n");
+	}
+
+	if (!HINIC_IS_VF(nic_dev->hwdev->hwif))
+		hinic_notify_all_vfs_link_changed(nic_dev->hwdev,
+						  link_status->link);
+
+	ret_link_status = buf_out;
+	ret_link_status->status = 0;
+
+	*out_size = sizeof(*ret_link_status);
+}
+
+static void cable_plug_event(void *handle,
+			     void *buf_in, u16 in_size,
+			     void *buf_out, u16 *out_size)
+{
+	struct hinic_cable_plug_event *plug_event = buf_in;
+	struct hinic_dev *nic_dev = handle;
+
+	nic_dev->cable_unplugged = plug_event->plugged ? false : true;
+
+	*out_size = sizeof(*plug_event);
+	plug_event = buf_out;
+	plug_event->status = 0;
+}
+
+static void link_err_event(void *handle,
+			   void *buf_in, u16 in_size,
+			   void *buf_out, u16 *out_size)
+{
+	struct hinic_link_err_event *link_err = buf_in;
+	struct hinic_dev *nic_dev = handle;
+
+	if (link_err->err_type >= LINK_ERR_NUM)
+		netif_info(nic_dev, link, nic_dev->netdev,
+			   "Link failed, Unknown error type: 0x%x\n",
+			   link_err->err_type);
+	else
+		nic_dev->module_unrecognized = true;
+
+	*out_size = sizeof(*link_err);
+	link_err = buf_out;
+	link_err->status = 0;
+}
+
+static int set_features(struct hinic_dev *nic_dev,
+			netdev_features_t pre_features,
+			netdev_features_t features, bool force_change)
+{
+	netdev_features_t changed = force_change ? ~0 : pre_features ^ features;
+	u32 csum_en = HINIC_RX_CSUM_OFFLOAD_EN;
+	netdev_features_t failed_features = 0;
+	int ret = 0;
+	int err = 0;
+
+	if (changed & NETIF_F_TSO) {
+		ret = hinic_port_set_tso(nic_dev, (features & NETIF_F_TSO) ?
+					 HINIC_TSO_ENABLE : HINIC_TSO_DISABLE);
+		if (ret) {
+			err = ret;
+			failed_features |= NETIF_F_TSO;
+		}
+	}
+
+	if (changed & NETIF_F_RXCSUM) {
+		ret = hinic_set_rx_csum_offload(nic_dev, csum_en);
+		if (ret) {
+			err = ret;
+			failed_features |= NETIF_F_RXCSUM;
+		}
+	}
+
+	if (changed & NETIF_F_LRO) {
+		ret = hinic_set_rx_lro_state(nic_dev,
+					     !!(features & NETIF_F_LRO),
+					     HINIC_LRO_RX_TIMER_DEFAULT,
+					     HINIC_LRO_MAX_WQE_NUM_DEFAULT);
+		if (ret) {
+			err = ret;
+			failed_features |= NETIF_F_LRO;
+		}
+	}
+
+	if (changed & NETIF_F_HW_VLAN_CTAG_RX) {
+		ret = hinic_set_rx_vlan_offload(nic_dev,
+						!!(features &
+						   NETIF_F_HW_VLAN_CTAG_RX));
+		if (ret) {
+			err = ret;
+			failed_features |= NETIF_F_HW_VLAN_CTAG_RX;
+		}
+	}
+
+	if (changed & NETIF_F_HW_VLAN_CTAG_FILTER) {
+		ret = hinic_set_vlan_fliter(nic_dev,
+					    !!(features &
+					       NETIF_F_HW_VLAN_CTAG_FILTER));
+		if (ret) {
+			err = ret;
+			failed_features |= NETIF_F_HW_VLAN_CTAG_FILTER;
+		}
+	}
+
+	if (err) {
+		nic_dev->netdev->features = features ^ failed_features;
+		return -EIO;
+	}
+
+	return 0;
+}
+
+static int hinic_init_intr_coalesce(struct hinic_dev *nic_dev)
+{
+	u64 size;
+	u16 i;
+
+	size = sizeof(struct hinic_intr_coal_info) * nic_dev->max_qps;
+	nic_dev->rx_intr_coalesce = kzalloc(size, GFP_KERNEL);
+	if (!nic_dev->rx_intr_coalesce)
+		return -ENOMEM;
+	nic_dev->tx_intr_coalesce = kzalloc(size, GFP_KERNEL);
+	if (!nic_dev->tx_intr_coalesce) {
+		kfree(nic_dev->rx_intr_coalesce);
+		return -ENOMEM;
+	}
+
+	for (i = 0; i < nic_dev->max_qps; i++) {
+		nic_dev->rx_intr_coalesce[i].pending_limt =
+			HINIC_DEAULT_TXRX_MSIX_PENDING_LIMIT;
+		nic_dev->rx_intr_coalesce[i].coalesce_timer_cfg =
+			HINIC_DEAULT_TXRX_MSIX_COALESC_TIMER_CFG;
+		nic_dev->rx_intr_coalesce[i].resend_timer_cfg =
+			HINIC_DEAULT_TXRX_MSIX_RESEND_TIMER_CFG;
+		nic_dev->tx_intr_coalesce[i].pending_limt =
+			HINIC_DEAULT_TXRX_MSIX_PENDING_LIMIT;
+		nic_dev->tx_intr_coalesce[i].coalesce_timer_cfg =
+			HINIC_DEAULT_TXRX_MSIX_COALESC_TIMER_CFG;
+		nic_dev->tx_intr_coalesce[i].resend_timer_cfg =
+			HINIC_DEAULT_TXRX_MSIX_RESEND_TIMER_CFG;
+	}
+
+	return 0;
+}
+
+static void hinic_free_intr_coalesce(struct hinic_dev *nic_dev)
+{
+	kfree(nic_dev->tx_intr_coalesce);
+	kfree(nic_dev->rx_intr_coalesce);
+}
+
+/**
+ * nic_dev_init - Initialize the NIC device
+ * @pdev: the NIC pci device
+ *
+ * Return 0 - Success, negative - Failure
+ **/
+static int nic_dev_init(struct pci_dev *pdev)
+{
+	struct hinic_rx_mode_work *rx_mode_work;
+	struct hinic_dev *nic_dev;
+	struct net_device *netdev;
+	struct hinic_hwdev *hwdev;
+	struct devlink *devlink;
+	u8 addr[ETH_ALEN];
+	int err, num_qps;
+
+	devlink = hinic_devlink_alloc(&pdev->dev);
+	if (!devlink) {
+		dev_err(&pdev->dev, "Hinic devlink alloc failed\n");
+		return -ENOMEM;
+	}
+
+	hwdev = hinic_init_hwdev(pdev, devlink);
+	if (IS_ERR(hwdev)) {
+		dev_err(&pdev->dev, "Failed to initialize HW device\n");
+		hinic_devlink_free(devlink);
+		return PTR_ERR(hwdev);
+	}
+
+	num_qps = hinic_hwdev_num_qps(hwdev);
+	if (num_qps <= 0) {
+		dev_err(&pdev->dev, "Invalid number of QPS\n");
+		err = -EINVAL;
+		goto err_num_qps;
+	}
+
+	netdev = alloc_etherdev_mq(sizeof(*nic_dev), num_qps);
+	if (!netdev) {
+		dev_err(&pdev->dev, "Failed to allocate Ethernet device\n");
+		err = -ENOMEM;
+		goto err_alloc_etherdev;
+	}
+
+	if (!HINIC_IS_VF(hwdev->hwif))
+		netdev->netdev_ops = &hinic_netdev_ops;
+	else
+		netdev->netdev_ops = &hinicvf_netdev_ops;
+
+	netdev->max_mtu = HINIC_MAX_MTU_SIZE;
+	netdev->min_mtu = HINIC_MIN_MTU_SIZE;
+
+	nic_dev = netdev_priv(netdev);
+	nic_dev->netdev = netdev;
+	nic_dev->hwdev  = hwdev;
+	nic_dev->msg_enable = MSG_ENABLE_DEFAULT;
+	nic_dev->flags = 0;
+	nic_dev->txqs = NULL;
+	nic_dev->rxqs = NULL;
+	nic_dev->tx_weight = tx_weight;
+	nic_dev->rx_weight = rx_weight;
+	nic_dev->sq_depth = HINIC_SQ_DEPTH;
+	nic_dev->rq_depth = HINIC_RQ_DEPTH;
+	nic_dev->sriov_info.hwdev = hwdev;
+	nic_dev->sriov_info.pdev = pdev;
+	nic_dev->max_qps = num_qps;
+	nic_dev->devlink = devlink;
+
+	hinic_set_ethtool_ops(netdev);
+
+	sema_init(&nic_dev->mgmt_lock, 1);
+
+	nic_dev->vlan_bitmap = devm_bitmap_zalloc(&pdev->dev, VLAN_N_VID,
+						  GFP_KERNEL);
+	if (!nic_dev->vlan_bitmap) {
+		err = -ENOMEM;
+		goto err_vlan_bitmap;
+	}
+
+	nic_dev->workq = create_singlethread_workqueue(HINIC_WQ_NAME);
+	if (!nic_dev->workq) {
+		err = -ENOMEM;
+		goto err_workq;
+	}
+
+	pci_set_drvdata(pdev, netdev);
+
+	err = hinic_port_get_mac(nic_dev, addr);
+	if (err) {
+		dev_err(&pdev->dev, "Failed to get mac address\n");
+		goto err_get_mac;
+	}
+	eth_hw_addr_set(netdev, addr);
+
+	if (!is_valid_ether_addr(netdev->dev_addr)) {
+		if (!HINIC_IS_VF(nic_dev->hwdev->hwif)) {
+			dev_err(&pdev->dev, "Invalid MAC address\n");
+			err = -EIO;
+			goto err_add_mac;
+		}
+
+		dev_info(&pdev->dev, "Invalid MAC address %pM, using random\n",
+			 netdev->dev_addr);
+		eth_hw_addr_random(netdev);
+	}
+
+	err = hinic_port_add_mac(nic_dev, netdev->dev_addr, 0);
+	if (err && err != HINIC_PF_SET_VF_ALREADY) {
+		dev_err(&pdev->dev, "Failed to add mac\n");
+		goto err_add_mac;
+	}
+
+	err = hinic_port_set_mtu(nic_dev, netdev->mtu);
+	if (err) {
+		dev_err(&pdev->dev, "Failed to set mtu\n");
+		goto err_set_mtu;
+	}
+
+	rx_mode_work = &nic_dev->rx_mode_work;
+	INIT_WORK(&rx_mode_work->work, set_rx_mode);
+
+	netdev_features_init(netdev);
+
+	netif_carrier_off(netdev);
+
+	hinic_hwdev_cb_register(nic_dev->hwdev, HINIC_MGMT_MSG_CMD_LINK_STATUS,
+				nic_dev, link_status_event_handler);
+	hinic_hwdev_cb_register(nic_dev->hwdev,
+				HINIC_MGMT_MSG_CMD_CABLE_PLUG_EVENT,
+				nic_dev, cable_plug_event);
+	hinic_hwdev_cb_register(nic_dev->hwdev,
+				HINIC_MGMT_MSG_CMD_LINK_ERR_EVENT,
+				nic_dev, link_err_event);
+
+	err = set_features(nic_dev, 0, nic_dev->netdev->features, true);
+	if (err)
+		goto err_set_features;
+
+	/* enable pause and disable pfc by default */
+	err = hinic_dcb_set_pfc(nic_dev->hwdev, 0, 0);
+	if (err)
+		goto err_set_pfc;
+
+	SET_NETDEV_DEV(netdev, &pdev->dev);
+
+	err = hinic_init_intr_coalesce(nic_dev);
+	if (err) {
+		dev_err(&pdev->dev, "Failed to init_intr_coalesce\n");
+		goto err_init_intr;
+	}
+
+	hinic_dbg_init(nic_dev);
+
+	hinic_func_tbl_dbgfs_init(nic_dev);
+
+	err = hinic_func_table_debug_add(nic_dev);
+	if (err) {
+		dev_err(&pdev->dev, "Failed to add func_table debug\n");
+		goto err_add_func_table_dbg;
+	}
+
+	err = register_netdev(netdev);
+	if (err) {
+		dev_err(&pdev->dev, "Failed to register netdev\n");
+		goto err_reg_netdev;
+	}
+
+	return 0;
+
+err_reg_netdev:
+	hinic_func_table_debug_rem(nic_dev);
+err_add_func_table_dbg:
+	hinic_func_tbl_dbgfs_uninit(nic_dev);
+	hinic_dbg_uninit(nic_dev);
+	hinic_free_intr_coalesce(nic_dev);
+err_init_intr:
+err_set_pfc:
+err_set_features:
+	hinic_hwdev_cb_unregister(nic_dev->hwdev,
+				  HINIC_MGMT_MSG_CMD_LINK_ERR_EVENT);
+	hinic_hwdev_cb_unregister(nic_dev->hwdev,
+				  HINIC_MGMT_MSG_CMD_CABLE_PLUG_EVENT);
+	hinic_hwdev_cb_unregister(nic_dev->hwdev,
+				  HINIC_MGMT_MSG_CMD_LINK_STATUS);
+	cancel_work_sync(&rx_mode_work->work);
+
+err_set_mtu:
+	hinic_port_del_mac(nic_dev, netdev->dev_addr, 0);
+err_add_mac:
+err_get_mac:
+	pci_set_drvdata(pdev, NULL);
+	destroy_workqueue(nic_dev->workq);
+err_workq:
+err_vlan_bitmap:
+	free_netdev(netdev);
+
+err_alloc_etherdev:
+err_num_qps:
+	hinic_free_hwdev(hwdev);
+	hinic_devlink_free(devlink);
+	return err;
+}
+
+static int hinic_probe(struct pci_dev *pdev,
+		       const struct pci_device_id *id)
+{
+	int err = pci_enable_device(pdev);
+
+	if (err)
+		return dev_err_probe(&pdev->dev, err, "Failed to enable PCI device\n");
+
+	err = pci_request_regions(pdev, HINIC_DRV_NAME);
+	if (err) {
+		dev_err(&pdev->dev, "Failed to request PCI regions\n");
+		goto err_pci_regions;
+	}
+
+	pci_set_master(pdev);
+
+	err = dma_set_mask_and_coherent(&pdev->dev, DMA_BIT_MASK(64));
+	if (err) {
+		dev_err(&pdev->dev, "Failed to set DMA mask\n");
+		goto err_dma_mask;
+	}
+
+	err = nic_dev_init(pdev);
+	if (err) {
+		dev_err(&pdev->dev, "Failed to initialize NIC device\n");
+		goto err_nic_dev_init;
+	}
+
+	dev_info(&pdev->dev, "HiNIC driver - probed\n");
+	return 0;
+
+err_nic_dev_init:
+err_dma_mask:
+	pci_release_regions(pdev);
+
+err_pci_regions:
+	pci_disable_device(pdev);
+	return err;
+}
+
+static void wait_sriov_cfg_complete(struct hinic_dev *nic_dev)
+{
+	struct hinic_sriov_info *sriov_info = &nic_dev->sriov_info;
+	u32 loop_cnt = 0;
+
+	set_bit(HINIC_FUNC_REMOVE, &sriov_info->state);
+	usleep_range(9900, 10000);
+
+	while (loop_cnt < HINIC_WAIT_SRIOV_CFG_TIMEOUT) {
+		if (!test_bit(HINIC_SRIOV_ENABLE, &sriov_info->state) &&
+		    !test_bit(HINIC_SRIOV_DISABLE, &sriov_info->state))
+			return;
+
+		usleep_range(9900, 10000);
+		loop_cnt++;
+	}
+}
+
+static void hinic_remove(struct pci_dev *pdev)
+{
+	struct net_device *netdev = pci_get_drvdata(pdev);
+	struct hinic_dev *nic_dev = netdev_priv(netdev);
+	struct devlink *devlink = nic_dev->devlink;
+	struct hinic_rx_mode_work *rx_mode_work;
+
+	if (!HINIC_IS_VF(nic_dev->hwdev->hwif)) {
+		wait_sriov_cfg_complete(nic_dev);
+		hinic_pci_sriov_disable(pdev);
+	}
+
+	unregister_netdev(netdev);
+
+	hinic_func_table_debug_rem(nic_dev);
+
+	hinic_func_tbl_dbgfs_uninit(nic_dev);
+
+	hinic_dbg_uninit(nic_dev);
+
+	hinic_free_intr_coalesce(nic_dev);
+
+	hinic_port_del_mac(nic_dev, netdev->dev_addr, 0);
+
+	hinic_hwdev_cb_unregister(nic_dev->hwdev,
+				  HINIC_MGMT_MSG_CMD_LINK_ERR_EVENT);
+	hinic_hwdev_cb_unregister(nic_dev->hwdev,
+				  HINIC_MGMT_MSG_CMD_CABLE_PLUG_EVENT);
+	hinic_hwdev_cb_unregister(nic_dev->hwdev,
+				  HINIC_MGMT_MSG_CMD_LINK_STATUS);
+
+	rx_mode_work = &nic_dev->rx_mode_work;
+	cancel_work_sync(&rx_mode_work->work);
+
+	pci_set_drvdata(pdev, NULL);
+
+	destroy_workqueue(nic_dev->workq);
+
+	hinic_free_hwdev(nic_dev->hwdev);
+
+	free_netdev(netdev);
+
+	hinic_devlink_free(devlink);
+
+	pci_release_regions(pdev);
+	pci_disable_device(pdev);
+
+	dev_info(&pdev->dev, "HiNIC driver - removed\n");
+}
+
+static void hinic_shutdown(struct pci_dev *pdev)
+{
+	pci_disable_device(pdev);
+}
+
+static const struct pci_device_id hinic_pci_table[] = {
+	{ PCI_VDEVICE(HUAWEI, HINIC_DEV_ID_QUAD_PORT_25GE), 0},
+	{ PCI_VDEVICE(HUAWEI, HINIC_DEV_ID_DUAL_PORT_100GE), 0},
+	{ PCI_VDEVICE(HUAWEI, HINIC_DEV_ID_DUAL_PORT_100GE_MEZZ), 0},
+	{ PCI_VDEVICE(HUAWEI, HINIC_DEV_ID_QUAD_PORT_25GE_MEZZ), 0},
+	{ PCI_VDEVICE(HUAWEI, HINIC_DEV_ID_VF), 0},
+	{ 0, 0}
+};
+MODULE_DEVICE_TABLE(pci, hinic_pci_table);
+
+static struct pci_driver hinic_driver = {
+	.name           = HINIC_DRV_NAME,
+	.id_table       = hinic_pci_table,
+	.probe          = hinic_probe,
+	.remove         = hinic_remove,
+	.shutdown       = hinic_shutdown,
+	.sriov_configure = hinic_pci_sriov_configure,
+};
+
+static int __init hinic_module_init(void)
+{
+	int ret;
+
+	hinic_dbg_register_debugfs(HINIC_DRV_NAME);
+
+	ret = pci_register_driver(&hinic_driver);
+	if (ret)
+		hinic_dbg_unregister_debugfs();
+
+	return ret;
+}
+
+static void __exit hinic_module_exit(void)
+{
+	pci_unregister_driver(&hinic_driver);
+	hinic_dbg_unregister_debugfs();
+}
+
+module_init(hinic_module_init);
+module_exit(hinic_module_exit);
diff --git a/drivers/net/ethernet/huawei/hinic/hinic_port.c b/drivers/net/ethernet/huawei/hinic/hinic_port.c
new file mode 100644
index 000000000000..486fb0e20bef
--- /dev/null
+++ b/drivers/net/ethernet/huawei/hinic/hinic_port.c
@@ -0,0 +1,1412 @@
+// SPDX-License-Identifier: GPL-2.0-only
+/*
+ * Huawei HiNIC PCI Express Linux driver
+ * Copyright(c) 2017 Huawei Technologies Co., Ltd
+ */
+
+#include <linux/types.h>
+#include <linux/netdevice.h>
+#include <linux/etherdevice.h>
+#include <linux/if_vlan.h>
+#include <linux/pci.h>
+#include <linux/device.h>
+#include <linux/errno.h>
+
+#include "hinic_hw_if.h"
+#include "hinic_hw_dev.h"
+#include "hinic_port.h"
+#include "hinic_dev.h"
+
+enum mac_op {
+	MAC_DEL,
+	MAC_SET,
+};
+
+/**
+ * change_mac - change(add or delete) mac address
+ * @nic_dev: nic device
+ * @addr: mac address
+ * @vlan_id: vlan number to set with the mac
+ * @op: add or delete the mac
+ *
+ * Return 0 - Success, negative - Failure
+ **/
+static int change_mac(struct hinic_dev *nic_dev, const u8 *addr,
+		      u16 vlan_id, enum mac_op op)
+{
+	struct hinic_hwdev *hwdev = nic_dev->hwdev;
+	struct hinic_port_mac_cmd port_mac_cmd;
+	struct hinic_hwif *hwif = hwdev->hwif;
+	u16 out_size = sizeof(port_mac_cmd);
+	struct pci_dev *pdev = hwif->pdev;
+	enum hinic_port_cmd cmd;
+	int err;
+
+	if (op == MAC_SET)
+		cmd = HINIC_PORT_CMD_SET_MAC;
+	else
+		cmd = HINIC_PORT_CMD_DEL_MAC;
+
+	port_mac_cmd.func_idx = HINIC_HWIF_FUNC_IDX(hwif);
+	port_mac_cmd.vlan_id = vlan_id;
+	memcpy(port_mac_cmd.mac, addr, ETH_ALEN);
+
+	err = hinic_port_msg_cmd(hwdev, cmd, &port_mac_cmd,
+				 sizeof(port_mac_cmd),
+				 &port_mac_cmd, &out_size);
+	if (err || out_size != sizeof(port_mac_cmd) ||
+	    (port_mac_cmd.status &&
+	     (port_mac_cmd.status != HINIC_PF_SET_VF_ALREADY || !HINIC_IS_VF(hwif)) &&
+	     port_mac_cmd.status != HINIC_MGMT_STATUS_EXIST)) {
+		dev_err(&pdev->dev, "Failed to change MAC, err: %d, status: 0x%x, out size: 0x%x\n",
+			err, port_mac_cmd.status, out_size);
+		return -EFAULT;
+	}
+
+	if (port_mac_cmd.status == HINIC_PF_SET_VF_ALREADY) {
+		dev_warn(&pdev->dev, "PF has already set VF mac, ignore %s operation\n",
+			 (op == MAC_SET) ? "set" : "del");
+		return HINIC_PF_SET_VF_ALREADY;
+	}
+
+	if (cmd == HINIC_PORT_CMD_SET_MAC && port_mac_cmd.status ==
+	    HINIC_MGMT_STATUS_EXIST)
+		dev_warn(&pdev->dev, "MAC is repeated, ignore set operation\n");
+
+	return 0;
+}
+
+/**
+ * hinic_port_add_mac - add mac address
+ * @nic_dev: nic device
+ * @addr: mac address
+ * @vlan_id: vlan number to set with the mac
+ *
+ * Return 0 - Success, negative - Failure
+ **/
+int hinic_port_add_mac(struct hinic_dev *nic_dev,
+		       const u8 *addr, u16 vlan_id)
+{
+	return change_mac(nic_dev, addr, vlan_id, MAC_SET);
+}
+
+/**
+ * hinic_port_del_mac - remove mac address
+ * @nic_dev: nic device
+ * @addr: mac address
+ * @vlan_id: vlan number that is connected to the mac
+ *
+ * Return 0 - Success, negative - Failure
+ **/
+int hinic_port_del_mac(struct hinic_dev *nic_dev, const u8 *addr,
+		       u16 vlan_id)
+{
+	return change_mac(nic_dev, addr, vlan_id, MAC_DEL);
+}
+
+/**
+ * hinic_port_get_mac - get the mac address of the nic device
+ * @nic_dev: nic device
+ * @addr: returned mac address
+ *
+ * Return 0 - Success, negative - Failure
+ **/
+int hinic_port_get_mac(struct hinic_dev *nic_dev, u8 *addr)
+{
+	struct hinic_hwdev *hwdev = nic_dev->hwdev;
+	struct hinic_port_mac_cmd port_mac_cmd;
+	struct hinic_hwif *hwif = hwdev->hwif;
+	u16 out_size = sizeof(port_mac_cmd);
+	struct pci_dev *pdev = hwif->pdev;
+	int err;
+
+	port_mac_cmd.func_idx = HINIC_HWIF_FUNC_IDX(hwif);
+
+	err = hinic_port_msg_cmd(hwdev, HINIC_PORT_CMD_GET_MAC,
+				 &port_mac_cmd, sizeof(port_mac_cmd),
+				 &port_mac_cmd, &out_size);
+	if (err || out_size != sizeof(port_mac_cmd) || port_mac_cmd.status) {
+		dev_err(&pdev->dev, "Failed to get mac, err: %d, status: 0x%x, out size: 0x%x\n",
+			err, port_mac_cmd.status, out_size);
+		return -EFAULT;
+	}
+
+	memcpy(addr, port_mac_cmd.mac, ETH_ALEN);
+	return 0;
+}
+
+/**
+ * hinic_port_set_mtu - set mtu
+ * @nic_dev: nic device
+ * @new_mtu: new mtu
+ *
+ * Return 0 - Success, negative - Failure
+ **/
+int hinic_port_set_mtu(struct hinic_dev *nic_dev, int new_mtu)
+{
+	struct hinic_hwdev *hwdev = nic_dev->hwdev;
+	struct hinic_port_mtu_cmd port_mtu_cmd;
+	struct hinic_hwif *hwif = hwdev->hwif;
+	u16 out_size = sizeof(port_mtu_cmd);
+	struct pci_dev *pdev = hwif->pdev;
+	int err;
+
+	port_mtu_cmd.func_idx = HINIC_HWIF_FUNC_IDX(hwif);
+	port_mtu_cmd.mtu = new_mtu;
+
+	err = hinic_port_msg_cmd(hwdev, HINIC_PORT_CMD_CHANGE_MTU,
+				 &port_mtu_cmd, sizeof(port_mtu_cmd),
+				 &port_mtu_cmd, &out_size);
+	if (err || out_size != sizeof(port_mtu_cmd) || port_mtu_cmd.status) {
+		dev_err(&pdev->dev, "Failed to set mtu, err: %d, status: 0x%x, out size: 0x%x\n",
+			err, port_mtu_cmd.status, out_size);
+		return -EFAULT;
+	}
+
+	return 0;
+}
+
+/**
+ * hinic_port_add_vlan - add vlan to the nic device
+ * @nic_dev: nic device
+ * @vlan_id: the vlan number to add
+ *
+ * Return 0 - Success, negative - Failure
+ **/
+int hinic_port_add_vlan(struct hinic_dev *nic_dev, u16 vlan_id)
+{
+	struct hinic_hwdev *hwdev = nic_dev->hwdev;
+	struct hinic_port_vlan_cmd port_vlan_cmd;
+
+	port_vlan_cmd.func_idx = HINIC_HWIF_FUNC_IDX(hwdev->hwif);
+	port_vlan_cmd.vlan_id = vlan_id;
+
+	return hinic_port_msg_cmd(hwdev, HINIC_PORT_CMD_ADD_VLAN,
+				  &port_vlan_cmd, sizeof(port_vlan_cmd),
+				  NULL, NULL);
+}
+
+/**
+ * hinic_port_del_vlan - delete vlan from the nic device
+ * @nic_dev: nic device
+ * @vlan_id: the vlan number to delete
+ *
+ * Return 0 - Success, negative - Failure
+ **/
+int hinic_port_del_vlan(struct hinic_dev *nic_dev, u16 vlan_id)
+{
+	struct hinic_hwdev *hwdev = nic_dev->hwdev;
+	struct hinic_port_vlan_cmd port_vlan_cmd;
+
+	port_vlan_cmd.func_idx = HINIC_HWIF_FUNC_IDX(hwdev->hwif);
+	port_vlan_cmd.vlan_id = vlan_id;
+
+	return hinic_port_msg_cmd(hwdev, HINIC_PORT_CMD_DEL_VLAN,
+				 &port_vlan_cmd, sizeof(port_vlan_cmd),
+				 NULL, NULL);
+}
+
+/**
+ * hinic_port_set_rx_mode - set rx mode in the nic device
+ * @nic_dev: nic device
+ * @rx_mode: the rx mode to set
+ *
+ * Return 0 - Success, negative - Failure
+ **/
+int hinic_port_set_rx_mode(struct hinic_dev *nic_dev, u32 rx_mode)
+{
+	struct hinic_hwdev *hwdev = nic_dev->hwdev;
+	struct hinic_port_rx_mode_cmd rx_mode_cmd;
+
+	rx_mode_cmd.func_idx = HINIC_HWIF_FUNC_IDX(hwdev->hwif);
+	rx_mode_cmd.rx_mode = rx_mode;
+
+	return hinic_port_msg_cmd(hwdev, HINIC_PORT_CMD_SET_RX_MODE,
+				  &rx_mode_cmd, sizeof(rx_mode_cmd),
+				  NULL, NULL);
+}
+
+/**
+ * hinic_port_link_state - get the link state
+ * @nic_dev: nic device
+ * @link_state: the returned link state
+ *
+ * Return 0 - Success, negative - Failure
+ **/
+int hinic_port_link_state(struct hinic_dev *nic_dev,
+			  enum hinic_port_link_state *link_state)
+{
+	struct hinic_hwdev *hwdev = nic_dev->hwdev;
+	struct hinic_hwif *hwif = hwdev->hwif;
+	struct hinic_port_link_cmd link_cmd;
+	struct pci_dev *pdev = hwif->pdev;
+	u16 out_size = sizeof(link_cmd);
+	int err;
+
+	link_cmd.func_idx = HINIC_HWIF_FUNC_IDX(hwif);
+
+	err = hinic_port_msg_cmd(hwdev, HINIC_PORT_CMD_GET_LINK_STATE,
+				 &link_cmd, sizeof(link_cmd),
+				 &link_cmd, &out_size);
+	if (err || out_size != sizeof(link_cmd) || link_cmd.status) {
+		dev_err(&pdev->dev, "Failed to get link state, err: %d, status: 0x%x, out size: 0x%x\n",
+			err, link_cmd.status, out_size);
+		return -EINVAL;
+	}
+
+	*link_state = link_cmd.state;
+	return 0;
+}
+
+/**
+ * hinic_port_set_state - set port state
+ * @nic_dev: nic device
+ * @state: the state to set
+ *
+ * Return 0 - Success, negative - Failure
+ **/
+int hinic_port_set_state(struct hinic_dev *nic_dev, enum hinic_port_state state)
+{
+	struct hinic_hwdev *hwdev = nic_dev->hwdev;
+	struct hinic_port_state_cmd port_state;
+	struct hinic_hwif *hwif = hwdev->hwif;
+	struct pci_dev *pdev = hwif->pdev;
+	u16 out_size = sizeof(port_state);
+	int err;
+
+	if (HINIC_IS_VF(hwdev->hwif))
+		return 0;
+
+	port_state.state = state;
+
+	err = hinic_port_msg_cmd(hwdev, HINIC_PORT_CMD_SET_PORT_STATE,
+				 &port_state, sizeof(port_state),
+				 &port_state, &out_size);
+	if (err || out_size != sizeof(port_state) || port_state.status) {
+		dev_err(&pdev->dev, "Failed to set port state, err: %d, status: 0x%x, out size: 0x%x\n",
+			err, port_state.status, out_size);
+		return -EFAULT;
+	}
+
+	return 0;
+}
+
+/**
+ * hinic_port_set_func_state- set func device state
+ * @nic_dev: nic device
+ * @state: the state to set
+ *
+ * Return 0 - Success, negative - Failure
+ **/
+int hinic_port_set_func_state(struct hinic_dev *nic_dev,
+			      enum hinic_func_port_state state)
+{
+	struct hinic_port_func_state_cmd func_state;
+	struct hinic_hwdev *hwdev = nic_dev->hwdev;
+	struct hinic_hwif *hwif = hwdev->hwif;
+	struct pci_dev *pdev = hwif->pdev;
+	u16 out_size = sizeof(func_state);
+	int err;
+
+	func_state.func_idx = HINIC_HWIF_FUNC_IDX(hwif);
+	func_state.state = state;
+
+	err = hinic_port_msg_cmd(hwdev, HINIC_PORT_CMD_SET_FUNC_STATE,
+				 &func_state, sizeof(func_state),
+				 &func_state, &out_size);
+	if (err || out_size != sizeof(func_state) || func_state.status) {
+		dev_err(&pdev->dev, "Failed to set port func state, err: %d, status: 0x%x, out size: 0x%x\n",
+			err, func_state.status, out_size);
+		return -EFAULT;
+	}
+
+	return 0;
+}
+
+/**
+ * hinic_port_get_cap - get port capabilities
+ * @nic_dev: nic device
+ * @port_cap: returned port capabilities
+ *
+ * Return 0 - Success, negative - Failure
+ **/
+int hinic_port_get_cap(struct hinic_dev *nic_dev,
+		       struct hinic_port_cap *port_cap)
+{
+	struct hinic_hwdev *hwdev = nic_dev->hwdev;
+	struct hinic_hwif *hwif = hwdev->hwif;
+	struct pci_dev *pdev = hwif->pdev;
+	u16 out_size = sizeof(*port_cap);
+	int err;
+
+	port_cap->func_idx = HINIC_HWIF_FUNC_IDX(hwif);
+
+	err = hinic_port_msg_cmd(hwdev, HINIC_PORT_CMD_GET_CAP,
+				 port_cap, sizeof(*port_cap),
+				 port_cap, &out_size);
+	if (err || out_size != sizeof(*port_cap) || port_cap->status) {
+		dev_err(&pdev->dev,
+			"Failed to get port capabilities, err: %d, status: 0x%x, out size: 0x%x\n",
+			err, port_cap->status, out_size);
+		return -EIO;
+	}
+
+	return 0;
+}
+
+/**
+ * hinic_port_set_tso - set port tso configuration
+ * @nic_dev: nic device
+ * @state: the tso state to set
+ *
+ * Return 0 - Success, negative - Failure
+ **/
+int hinic_port_set_tso(struct hinic_dev *nic_dev, enum hinic_tso_state state)
+{
+	struct hinic_hwdev *hwdev = nic_dev->hwdev;
+	struct hinic_hwif *hwif = hwdev->hwif;
+	struct hinic_tso_config tso_cfg = {0};
+	struct pci_dev *pdev = hwif->pdev;
+	u16 out_size = sizeof(tso_cfg);
+	int err;
+
+	tso_cfg.func_id = HINIC_HWIF_FUNC_IDX(hwif);
+	tso_cfg.tso_en = state;
+
+	err = hinic_port_msg_cmd(hwdev, HINIC_PORT_CMD_SET_TSO,
+				 &tso_cfg, sizeof(tso_cfg),
+				 &tso_cfg, &out_size);
+	if (err || out_size != sizeof(tso_cfg) || tso_cfg.status) {
+		dev_err(&pdev->dev,
+			"Failed to set port tso, err: %d, status: 0x%x, out size: 0x%x\n",
+			err, tso_cfg.status, out_size);
+		return -EIO;
+	}
+
+	return 0;
+}
+
+int hinic_set_rx_csum_offload(struct hinic_dev *nic_dev, u32 en)
+{
+	struct hinic_checksum_offload rx_csum_cfg = {0};
+	struct hinic_hwdev *hwdev = nic_dev->hwdev;
+	u16 out_size = sizeof(rx_csum_cfg);
+	struct hinic_hwif *hwif;
+	struct pci_dev *pdev;
+	int err;
+
+	if (!hwdev)
+		return -EINVAL;
+
+	hwif = hwdev->hwif;
+	pdev = hwif->pdev;
+	rx_csum_cfg.func_id = HINIC_HWIF_FUNC_IDX(hwif);
+	rx_csum_cfg.rx_csum_offload = en;
+
+	err = hinic_port_msg_cmd(hwdev, HINIC_PORT_CMD_SET_RX_CSUM,
+				 &rx_csum_cfg, sizeof(rx_csum_cfg),
+				 &rx_csum_cfg, &out_size);
+	if (err || !out_size || rx_csum_cfg.status) {
+		dev_err(&pdev->dev,
+			"Failed to set rx csum offload, err: %d, status: 0x%x, out size: 0x%x\n",
+			err, rx_csum_cfg.status, out_size);
+		return -EIO;
+	}
+
+	return 0;
+}
+
+int hinic_set_rx_vlan_offload(struct hinic_dev *nic_dev, u8 en)
+{
+	struct hinic_hwdev *hwdev = nic_dev->hwdev;
+	struct hinic_vlan_cfg vlan_cfg;
+	struct hinic_hwif *hwif;
+	struct pci_dev *pdev;
+	u16 out_size;
+	int err;
+
+	if (!hwdev)
+		return -EINVAL;
+
+	out_size = sizeof(vlan_cfg);
+	hwif = hwdev->hwif;
+	pdev = hwif->pdev;
+	vlan_cfg.func_id = HINIC_HWIF_FUNC_IDX(hwif);
+	vlan_cfg.vlan_rx_offload = en;
+
+	err = hinic_port_msg_cmd(hwdev, HINIC_PORT_CMD_SET_RX_VLAN_OFFLOAD,
+				 &vlan_cfg, sizeof(vlan_cfg),
+				 &vlan_cfg, &out_size);
+	if (err || !out_size || vlan_cfg.status) {
+		dev_err(&pdev->dev,
+			"Failed to set rx vlan offload, err: %d, status: 0x%x, out size: 0x%x\n",
+			err, vlan_cfg.status, out_size);
+		return -EINVAL;
+	}
+
+	return 0;
+}
+
+int hinic_set_vlan_fliter(struct hinic_dev *nic_dev, u32 en)
+{
+	struct hinic_hwdev *hwdev = nic_dev->hwdev;
+	struct hinic_hwif *hwif = hwdev->hwif;
+	struct pci_dev *pdev = hwif->pdev;
+	struct hinic_vlan_filter vlan_filter;
+	u16 out_size = sizeof(vlan_filter);
+	int err;
+
+	vlan_filter.func_idx = HINIC_HWIF_FUNC_IDX(hwif);
+	vlan_filter.enable = en;
+
+	err = hinic_port_msg_cmd(hwdev, HINIC_PORT_CMD_SET_VLAN_FILTER,
+				 &vlan_filter, sizeof(vlan_filter),
+				 &vlan_filter, &out_size);
+	if (vlan_filter.status == HINIC_MGMT_CMD_UNSUPPORTED) {
+		err = HINIC_MGMT_CMD_UNSUPPORTED;
+	} else if ((err == HINIC_MBOX_VF_CMD_ERROR) &&
+			   HINIC_IS_VF(hwif)) {
+		err = HINIC_MGMT_CMD_UNSUPPORTED;
+	} else if (err || !out_size || vlan_filter.status) {
+		dev_err(&pdev->dev,
+			"Failed to set vlan filter, err: %d, status: 0x%x, out size: 0x%x\n",
+			err, vlan_filter.status, out_size);
+		err = -EINVAL;
+	}
+
+	return err;
+}
+
+int hinic_set_max_qnum(struct hinic_dev *nic_dev, u8 num_rqs)
+{
+	struct hinic_hwdev *hwdev = nic_dev->hwdev;
+	struct hinic_hwif *hwif = hwdev->hwif;
+	struct hinic_rq_num rq_num = { 0 };
+	struct pci_dev *pdev = hwif->pdev;
+	u16 out_size = sizeof(rq_num);
+	int err;
+
+	rq_num.func_id = HINIC_HWIF_FUNC_IDX(hwif);
+	rq_num.num_rqs = num_rqs;
+	rq_num.rq_depth = ilog2(nic_dev->rq_depth);
+
+	err = hinic_port_msg_cmd(hwdev, HINIC_PORT_CMD_SET_RQ_IQ_MAP,
+				 &rq_num, sizeof(rq_num),
+				 &rq_num, &out_size);
+	if (err || !out_size || rq_num.status) {
+		dev_err(&pdev->dev,
+			"Failed to set rxq number, err: %d, status: 0x%x, out size: 0x%x\n",
+			err, rq_num.status, out_size);
+		return -EIO;
+	}
+
+	return 0;
+}
+
+static int hinic_set_rx_lro(struct hinic_dev *nic_dev, u8 ipv4_en, u8 ipv6_en,
+			    u8 max_wqe_num)
+{
+	struct hinic_hwdev *hwdev = nic_dev->hwdev;
+	struct hinic_lro_config lro_cfg = { 0 };
+	struct hinic_hwif *hwif = hwdev->hwif;
+	struct pci_dev *pdev = hwif->pdev;
+	u16 out_size = sizeof(lro_cfg);
+	int err;
+
+	lro_cfg.func_id = HINIC_HWIF_FUNC_IDX(hwif);
+	lro_cfg.lro_ipv4_en = ipv4_en;
+	lro_cfg.lro_ipv6_en = ipv6_en;
+	lro_cfg.lro_max_wqe_num = max_wqe_num;
+
+	err = hinic_port_msg_cmd(hwdev, HINIC_PORT_CMD_SET_LRO,
+				 &lro_cfg, sizeof(lro_cfg),
+				 &lro_cfg, &out_size);
+	if (err || !out_size || lro_cfg.status) {
+		dev_err(&pdev->dev,
+			"Failed to set lro offload, err: %d, status: 0x%x, out size: 0x%x\n",
+			err, lro_cfg.status, out_size);
+		return -EIO;
+	}
+
+	return 0;
+}
+
+static int hinic_set_rx_lro_timer(struct hinic_dev *nic_dev, u32 timer_value)
+{
+	struct hinic_hwdev *hwdev = nic_dev->hwdev;
+	struct hinic_lro_timer lro_timer = { 0 };
+	struct hinic_hwif *hwif = hwdev->hwif;
+	struct pci_dev *pdev = hwif->pdev;
+	u16 out_size = sizeof(lro_timer);
+	int err;
+
+	lro_timer.status = 0;
+	lro_timer.type = 0;
+	lro_timer.enable = 1;
+	lro_timer.timer = timer_value;
+
+	err = hinic_port_msg_cmd(hwdev, HINIC_PORT_CMD_SET_LRO_TIMER,
+				 &lro_timer, sizeof(lro_timer),
+				 &lro_timer, &out_size);
+	if (lro_timer.status == 0xFF) {
+		/* For this case, we think status (0xFF) is OK */
+		lro_timer.status = 0;
+		dev_dbg(&pdev->dev,
+			"Set lro timer not supported by the current FW version, it will be 1ms default\n");
+	}
+
+	if (err || !out_size || lro_timer.status) {
+		dev_err(&pdev->dev,
+			"Failed to set lro timer, err: %d, status: 0x%x, out size: 0x%x\n",
+			err, lro_timer.status, out_size);
+
+		return -EIO;
+	}
+
+	return 0;
+}
+
+int hinic_set_rx_lro_state(struct hinic_dev *nic_dev, u8 lro_en,
+			   u32 lro_timer, u32 wqe_num)
+{
+	struct hinic_hwdev *hwdev = nic_dev->hwdev;
+	u8 ipv4_en;
+	u8 ipv6_en;
+	int err;
+
+	if (!hwdev)
+		return -EINVAL;
+
+	ipv4_en = lro_en ? 1 : 0;
+	ipv6_en = lro_en ? 1 : 0;
+
+	err = hinic_set_rx_lro(nic_dev, ipv4_en, ipv6_en, (u8)wqe_num);
+	if (err)
+		return err;
+
+	if (HINIC_IS_VF(nic_dev->hwdev->hwif))
+		return 0;
+
+	err = hinic_set_rx_lro_timer(nic_dev, lro_timer);
+	if (err)
+		return err;
+
+	return 0;
+}
+
+int hinic_rss_set_indir_tbl(struct hinic_dev *nic_dev, u32 tmpl_idx,
+			    const u32 *indir_table)
+{
+	struct hinic_rss_indirect_tbl *indir_tbl;
+	struct hinic_func_to_io *func_to_io;
+	struct hinic_cmdq_buf cmd_buf;
+	struct hinic_hwdev *hwdev;
+	struct hinic_hwif *hwif;
+	struct pci_dev *pdev;
+	u32 indir_size;
+	u64 out_param;
+	int err, i;
+	u32 *temp;
+
+	hwdev = nic_dev->hwdev;
+	func_to_io = &hwdev->func_to_io;
+	hwif = hwdev->hwif;
+	pdev = hwif->pdev;
+
+	err = hinic_alloc_cmdq_buf(&func_to_io->cmdqs, &cmd_buf);
+	if (err) {
+		dev_err(&pdev->dev, "Failed to allocate cmdq buf\n");
+		return err;
+	}
+
+	cmd_buf.size = sizeof(*indir_tbl);
+
+	indir_tbl = cmd_buf.buf;
+	indir_tbl->group_index = cpu_to_be32(tmpl_idx);
+
+	for (i = 0; i < HINIC_RSS_INDIR_SIZE; i++) {
+		indir_tbl->entry[i] = indir_table[i];
+
+		if (0x3 == (i & 0x3)) {
+			temp = (u32 *)&indir_tbl->entry[i - 3];
+			*temp = cpu_to_be32(*temp);
+		}
+	}
+
+	/* cfg the rss indirect table by command queue */
+	indir_size = HINIC_RSS_INDIR_SIZE / 2;
+	indir_tbl->offset = 0;
+	indir_tbl->size = cpu_to_be32(indir_size);
+
+	err = hinic_cmdq_direct_resp(&func_to_io->cmdqs, HINIC_MOD_L2NIC,
+				     HINIC_UCODE_CMD_SET_RSS_INDIR_TABLE,
+				     &cmd_buf, &out_param);
+	if (err || out_param != 0) {
+		dev_err(&pdev->dev, "Failed to set rss indir table\n");
+		err = -EFAULT;
+		goto free_buf;
+	}
+
+	indir_tbl->offset = cpu_to_be32(indir_size);
+	indir_tbl->size = cpu_to_be32(indir_size);
+	memcpy(&indir_tbl->entry[0], &indir_tbl->entry[indir_size], indir_size);
+
+	err = hinic_cmdq_direct_resp(&func_to_io->cmdqs, HINIC_MOD_L2NIC,
+				     HINIC_UCODE_CMD_SET_RSS_INDIR_TABLE,
+				     &cmd_buf, &out_param);
+	if (err || out_param != 0) {
+		dev_err(&pdev->dev, "Failed to set rss indir table\n");
+		err = -EFAULT;
+	}
+
+free_buf:
+	hinic_free_cmdq_buf(&func_to_io->cmdqs, &cmd_buf);
+
+	return err;
+}
+
+int hinic_rss_get_indir_tbl(struct hinic_dev *nic_dev, u32 tmpl_idx,
+			    u32 *indir_table)
+{
+	struct hinic_rss_indir_table rss_cfg = { 0 };
+	struct hinic_hwdev *hwdev = nic_dev->hwdev;
+	struct hinic_hwif *hwif = hwdev->hwif;
+	struct pci_dev *pdev = hwif->pdev;
+	u16 out_size = sizeof(rss_cfg);
+	int err = 0, i;
+
+	rss_cfg.func_id = HINIC_HWIF_FUNC_IDX(hwif);
+	rss_cfg.template_id = tmpl_idx;
+
+	err = hinic_port_msg_cmd(hwdev,
+				 HINIC_PORT_CMD_GET_RSS_TEMPLATE_INDIR_TBL,
+				 &rss_cfg, sizeof(rss_cfg), &rss_cfg,
+				 &out_size);
+	if (err || !out_size || rss_cfg.status) {
+		dev_err(&pdev->dev, "Failed to get indir table, err: %d, status: 0x%x, out size: 0x%x\n",
+			err, rss_cfg.status, out_size);
+		return -EINVAL;
+	}
+
+	hinic_be32_to_cpu(rss_cfg.indir, HINIC_RSS_INDIR_SIZE);
+	for (i = 0; i < HINIC_RSS_INDIR_SIZE; i++)
+		indir_table[i] = rss_cfg.indir[i];
+
+	return 0;
+}
+
+int hinic_set_rss_type(struct hinic_dev *nic_dev, u32 tmpl_idx,
+		       struct hinic_rss_type rss_type)
+{
+	struct hinic_rss_context_tbl *ctx_tbl;
+	struct hinic_func_to_io *func_to_io;
+	struct hinic_cmdq_buf cmd_buf;
+	struct hinic_hwdev *hwdev;
+	struct hinic_hwif *hwif;
+	struct pci_dev *pdev;
+	u64 out_param;
+	u32 ctx = 0;
+	int err;
+
+	hwdev = nic_dev->hwdev;
+	func_to_io = &hwdev->func_to_io;
+	hwif = hwdev->hwif;
+	pdev = hwif->pdev;
+
+	err = hinic_alloc_cmdq_buf(&func_to_io->cmdqs, &cmd_buf);
+	if (err) {
+		dev_err(&pdev->dev, "Failed to allocate cmd buf\n");
+		return -ENOMEM;
+	}
+
+	ctx |=  HINIC_RSS_TYPE_SET(1, VALID) |
+		HINIC_RSS_TYPE_SET(rss_type.ipv4, IPV4) |
+		HINIC_RSS_TYPE_SET(rss_type.ipv6, IPV6) |
+		HINIC_RSS_TYPE_SET(rss_type.ipv6_ext, IPV6_EXT) |
+		HINIC_RSS_TYPE_SET(rss_type.tcp_ipv4, TCP_IPV4) |
+		HINIC_RSS_TYPE_SET(rss_type.tcp_ipv6, TCP_IPV6) |
+		HINIC_RSS_TYPE_SET(rss_type.tcp_ipv6_ext, TCP_IPV6_EXT) |
+		HINIC_RSS_TYPE_SET(rss_type.udp_ipv4, UDP_IPV4) |
+		HINIC_RSS_TYPE_SET(rss_type.udp_ipv6, UDP_IPV6);
+
+	cmd_buf.size = sizeof(struct hinic_rss_context_tbl);
+
+	ctx_tbl = (struct hinic_rss_context_tbl *)cmd_buf.buf;
+	ctx_tbl->group_index = cpu_to_be32(tmpl_idx);
+	ctx_tbl->offset = 0;
+	ctx_tbl->size = sizeof(u32);
+	ctx_tbl->size = cpu_to_be32(ctx_tbl->size);
+	ctx_tbl->rsvd = 0;
+	ctx_tbl->ctx = cpu_to_be32(ctx);
+
+	/* cfg the rss context table by command queue */
+	err = hinic_cmdq_direct_resp(&func_to_io->cmdqs, HINIC_MOD_L2NIC,
+				     HINIC_UCODE_CMD_SET_RSS_CONTEXT_TABLE,
+				     &cmd_buf, &out_param);
+
+	hinic_free_cmdq_buf(&func_to_io->cmdqs, &cmd_buf);
+
+	if (err || out_param != 0) {
+		dev_err(&pdev->dev, "Failed to set rss context table, err: %d\n",
+			err);
+		return -EFAULT;
+	}
+
+	return 0;
+}
+
+int hinic_get_rss_type(struct hinic_dev *nic_dev, u32 tmpl_idx,
+		       struct hinic_rss_type *rss_type)
+{
+	struct hinic_rss_context_table ctx_tbl = { 0 };
+	struct hinic_hwdev *hwdev = nic_dev->hwdev;
+	u16 out_size = sizeof(ctx_tbl);
+	struct hinic_hwif *hwif;
+	struct pci_dev *pdev;
+	int err;
+
+	if (!hwdev || !rss_type)
+		return -EINVAL;
+
+	hwif = hwdev->hwif;
+	pdev = hwif->pdev;
+
+	ctx_tbl.func_id = HINIC_HWIF_FUNC_IDX(hwif);
+	ctx_tbl.template_id = tmpl_idx;
+
+	err = hinic_port_msg_cmd(hwdev, HINIC_PORT_CMD_GET_RSS_CTX_TBL,
+				 &ctx_tbl, sizeof(ctx_tbl),
+				 &ctx_tbl, &out_size);
+	if (err || !out_size || ctx_tbl.status) {
+		dev_err(&pdev->dev, "Failed to get hash type, err: %d, status: 0x%x, out size: 0x%x\n",
+			err, ctx_tbl.status, out_size);
+		return -EINVAL;
+	}
+
+	rss_type->ipv4          = HINIC_RSS_TYPE_GET(ctx_tbl.context, IPV4);
+	rss_type->ipv6          = HINIC_RSS_TYPE_GET(ctx_tbl.context, IPV6);
+	rss_type->ipv6_ext      = HINIC_RSS_TYPE_GET(ctx_tbl.context, IPV6_EXT);
+	rss_type->tcp_ipv4      = HINIC_RSS_TYPE_GET(ctx_tbl.context, TCP_IPV4);
+	rss_type->tcp_ipv6      = HINIC_RSS_TYPE_GET(ctx_tbl.context, TCP_IPV6);
+	rss_type->tcp_ipv6_ext  = HINIC_RSS_TYPE_GET(ctx_tbl.context,
+						     TCP_IPV6_EXT);
+	rss_type->udp_ipv4      = HINIC_RSS_TYPE_GET(ctx_tbl.context, UDP_IPV4);
+	rss_type->udp_ipv6      = HINIC_RSS_TYPE_GET(ctx_tbl.context, UDP_IPV6);
+
+	return 0;
+}
+
+int hinic_rss_set_template_tbl(struct hinic_dev *nic_dev, u32 template_id,
+			       const u8 *temp)
+{
+	struct hinic_hwdev *hwdev = nic_dev->hwdev;
+	struct hinic_hwif *hwif = hwdev->hwif;
+	struct hinic_rss_key rss_key = { 0 };
+	struct pci_dev *pdev = hwif->pdev;
+	u16 out_size = sizeof(rss_key);
+	int err;
+
+	rss_key.func_id = HINIC_HWIF_FUNC_IDX(hwif);
+	rss_key.template_id = template_id;
+	memcpy(rss_key.key, temp, HINIC_RSS_KEY_SIZE);
+
+	err = hinic_port_msg_cmd(hwdev, HINIC_PORT_CMD_SET_RSS_TEMPLATE_TBL,
+				 &rss_key, sizeof(rss_key),
+				 &rss_key, &out_size);
+	if (err || !out_size || rss_key.status) {
+		dev_err(&pdev->dev,
+			"Failed to set rss hash key, err: %d, status: 0x%x, out size: 0x%x\n",
+			err, rss_key.status, out_size);
+		return -EINVAL;
+	}
+
+	return 0;
+}
+
+int hinic_rss_get_template_tbl(struct hinic_dev *nic_dev, u32 tmpl_idx,
+			       u8 *temp)
+{
+	struct hinic_rss_template_key temp_key = { 0 };
+	struct hinic_hwdev *hwdev = nic_dev->hwdev;
+	u16 out_size = sizeof(temp_key);
+	struct hinic_hwif *hwif;
+	struct pci_dev *pdev;
+	int err;
+
+	if (!hwdev || !temp)
+		return -EINVAL;
+
+	hwif = hwdev->hwif;
+	pdev = hwif->pdev;
+
+	temp_key.func_id = HINIC_HWIF_FUNC_IDX(hwif);
+	temp_key.template_id = tmpl_idx;
+
+	err = hinic_port_msg_cmd(hwdev, HINIC_PORT_CMD_GET_RSS_TEMPLATE_TBL,
+				 &temp_key, sizeof(temp_key),
+				 &temp_key, &out_size);
+	if (err || !out_size || temp_key.status) {
+		dev_err(&pdev->dev, "Failed to set hash key, err: %d, status: 0x%x, out size: 0x%x\n",
+			err, temp_key.status, out_size);
+		return -EINVAL;
+	}
+
+	memcpy(temp, temp_key.key, HINIC_RSS_KEY_SIZE);
+
+	return 0;
+}
+
+int hinic_rss_set_hash_engine(struct hinic_dev *nic_dev, u8 template_id,
+			      u8 type)
+{
+	struct hinic_rss_engine_type rss_engine = { 0 };
+	struct hinic_hwdev *hwdev = nic_dev->hwdev;
+	struct hinic_hwif *hwif = hwdev->hwif;
+	struct pci_dev *pdev = hwif->pdev;
+	u16 out_size = sizeof(rss_engine);
+	int err;
+
+	rss_engine.func_id = HINIC_HWIF_FUNC_IDX(hwif);
+	rss_engine.hash_engine = type;
+	rss_engine.template_id = template_id;
+
+	err = hinic_port_msg_cmd(hwdev, HINIC_PORT_CMD_SET_RSS_HASH_ENGINE,
+				 &rss_engine, sizeof(rss_engine),
+				 &rss_engine, &out_size);
+	if (err || !out_size || rss_engine.status) {
+		dev_err(&pdev->dev,
+			"Failed to set hash engine, err: %d, status: 0x%x, out size: 0x%x\n",
+			err, rss_engine.status, out_size);
+		return -EINVAL;
+	}
+
+	return 0;
+}
+
+int hinic_rss_get_hash_engine(struct hinic_dev *nic_dev, u8 tmpl_idx, u8 *type)
+{
+	struct hinic_rss_engine_type hash_type = { 0 };
+	struct hinic_hwdev *hwdev = nic_dev->hwdev;
+	u16 out_size = sizeof(hash_type);
+	struct hinic_hwif *hwif;
+	struct pci_dev *pdev;
+	int err;
+
+	if (!hwdev || !type)
+		return -EINVAL;
+
+	hwif = hwdev->hwif;
+	pdev = hwif->pdev;
+
+	hash_type.func_id = HINIC_HWIF_FUNC_IDX(hwif);
+	hash_type.template_id = tmpl_idx;
+
+	err = hinic_port_msg_cmd(hwdev, HINIC_PORT_CMD_GET_RSS_HASH_ENGINE,
+				 &hash_type, sizeof(hash_type),
+				 &hash_type, &out_size);
+	if (err || !out_size || hash_type.status) {
+		dev_err(&pdev->dev, "Failed to get hash engine, err: %d, status: 0x%x, out size: 0x%x\n",
+			err, hash_type.status, out_size);
+		return -EINVAL;
+	}
+
+	*type = hash_type.hash_engine;
+	return 0;
+}
+
+int hinic_rss_cfg(struct hinic_dev *nic_dev, u8 rss_en, u8 template_id)
+{
+	struct hinic_hwdev *hwdev = nic_dev->hwdev;
+	struct hinic_rss_config rss_cfg = { 0 };
+	struct hinic_hwif *hwif = hwdev->hwif;
+	struct pci_dev *pdev = hwif->pdev;
+	u16 out_size = sizeof(rss_cfg);
+	int err;
+
+	rss_cfg.func_id = HINIC_HWIF_FUNC_IDX(hwif);
+	rss_cfg.rss_en = rss_en;
+	rss_cfg.template_id = template_id;
+	rss_cfg.rq_priority_number = 0;
+
+	err = hinic_port_msg_cmd(hwdev, HINIC_PORT_CMD_RSS_CFG,
+				 &rss_cfg, sizeof(rss_cfg),
+				 &rss_cfg, &out_size);
+	if (err || !out_size || rss_cfg.status) {
+		dev_err(&pdev->dev,
+			"Failed to set rss cfg, err: %d, status: 0x%x, out size: 0x%x\n",
+			err, rss_cfg.status, out_size);
+		return -EINVAL;
+	}
+
+	return 0;
+}
+
+int hinic_rss_template_alloc(struct hinic_dev *nic_dev, u8 *tmpl_idx)
+{
+	struct hinic_rss_template_mgmt template_mgmt = { 0 };
+	struct hinic_hwdev *hwdev = nic_dev->hwdev;
+	struct hinic_hwif *hwif = hwdev->hwif;
+	u16 out_size = sizeof(template_mgmt);
+	struct pci_dev *pdev = hwif->pdev;
+	int err;
+
+	template_mgmt.func_id = HINIC_HWIF_FUNC_IDX(hwif);
+	template_mgmt.cmd = NIC_RSS_CMD_TEMP_ALLOC;
+
+	err = hinic_port_msg_cmd(hwdev, HINIC_PORT_CMD_RSS_TEMP_MGR,
+				 &template_mgmt, sizeof(template_mgmt),
+				 &template_mgmt, &out_size);
+	if (err || !out_size || template_mgmt.status) {
+		dev_err(&pdev->dev, "Failed to alloc rss template, err: %d, status: 0x%x, out size: 0x%x\n",
+			err, template_mgmt.status, out_size);
+		return -EINVAL;
+	}
+
+	*tmpl_idx = template_mgmt.template_id;
+
+	return 0;
+}
+
+int hinic_rss_template_free(struct hinic_dev *nic_dev, u8 tmpl_idx)
+{
+	struct hinic_rss_template_mgmt template_mgmt = { 0 };
+	struct hinic_hwdev *hwdev = nic_dev->hwdev;
+	struct hinic_hwif *hwif = hwdev->hwif;
+	u16 out_size = sizeof(template_mgmt);
+	struct pci_dev *pdev = hwif->pdev;
+	int err;
+
+	template_mgmt.func_id = HINIC_HWIF_FUNC_IDX(hwif);
+	template_mgmt.template_id = tmpl_idx;
+	template_mgmt.cmd = NIC_RSS_CMD_TEMP_FREE;
+
+	err = hinic_port_msg_cmd(hwdev, HINIC_PORT_CMD_RSS_TEMP_MGR,
+				 &template_mgmt, sizeof(template_mgmt),
+				 &template_mgmt, &out_size);
+	if (err || !out_size || template_mgmt.status) {
+		dev_err(&pdev->dev, "Failed to free rss template, err: %d, status: 0x%x, out size: 0x%x\n",
+			err, template_mgmt.status, out_size);
+		return -EINVAL;
+	}
+
+	return 0;
+}
+
+int hinic_get_vport_stats(struct hinic_dev *nic_dev,
+			  struct hinic_vport_stats *stats)
+{
+	struct hinic_cmd_vport_stats vport_stats = { 0 };
+	struct hinic_port_stats_info stats_info = { 0 };
+	struct hinic_hwdev *hwdev = nic_dev->hwdev;
+	struct hinic_hwif *hwif = hwdev->hwif;
+	u16 out_size = sizeof(vport_stats);
+	struct pci_dev *pdev = hwif->pdev;
+	int err;
+
+	stats_info.stats_version = HINIC_PORT_STATS_VERSION;
+	stats_info.func_id = HINIC_HWIF_FUNC_IDX(hwif);
+	stats_info.stats_size = sizeof(vport_stats);
+
+	err = hinic_port_msg_cmd(hwdev, HINIC_PORT_CMD_GET_VPORT_STAT,
+				 &stats_info, sizeof(stats_info),
+				 &vport_stats, &out_size);
+	if (err || !out_size || vport_stats.status) {
+		dev_err(&pdev->dev,
+			"Failed to get function statistics, err: %d, status: 0x%x, out size: 0x%x\n",
+			err, vport_stats.status, out_size);
+		return -EFAULT;
+	}
+
+	memcpy(stats, &vport_stats.stats, sizeof(*stats));
+	return 0;
+}
+
+int hinic_get_phy_port_stats(struct hinic_dev *nic_dev,
+			     struct hinic_phy_port_stats *stats)
+{
+	struct hinic_port_stats_info stats_info = { 0 };
+	struct hinic_hwdev *hwdev = nic_dev->hwdev;
+	struct hinic_hwif *hwif = hwdev->hwif;
+	struct hinic_port_stats *port_stats;
+	u16 out_size = sizeof(*port_stats);
+	struct pci_dev *pdev = hwif->pdev;
+	int err;
+
+	port_stats = kzalloc(sizeof(*port_stats), GFP_KERNEL);
+	if (!port_stats)
+		return -ENOMEM;
+
+	stats_info.stats_version = HINIC_PORT_STATS_VERSION;
+	stats_info.stats_size = sizeof(*port_stats);
+
+	err = hinic_port_msg_cmd(hwdev, HINIC_PORT_CMD_GET_PORT_STATISTICS,
+				 &stats_info, sizeof(stats_info),
+				 port_stats, &out_size);
+	if (err || !out_size || port_stats->status) {
+		dev_err(&pdev->dev,
+			"Failed to get port statistics, err: %d, status: 0x%x, out size: 0x%x\n",
+			err, port_stats->status, out_size);
+		err = -EINVAL;
+		goto out;
+	}
+
+	memcpy(stats, &port_stats->stats, sizeof(*stats));
+
+out:
+	kfree(port_stats);
+
+	return err;
+}
+
+int hinic_get_mgmt_version(struct hinic_dev *nic_dev, u8 *mgmt_ver)
+{
+	struct hinic_hwdev *hwdev = nic_dev->hwdev;
+	struct hinic_version_info up_ver = {0};
+	u16 out_size = sizeof(up_ver);
+	struct hinic_hwif *hwif;
+	struct pci_dev *pdev;
+	int err;
+
+	if (!hwdev)
+		return -EINVAL;
+
+	hwif = hwdev->hwif;
+	pdev = hwif->pdev;
+
+	err = hinic_port_msg_cmd(hwdev, HINIC_PORT_CMD_GET_MGMT_VERSION,
+				 &up_ver, sizeof(up_ver), &up_ver,
+				 &out_size);
+	if (err || !out_size || up_ver.status) {
+		dev_err(&pdev->dev,
+			"Failed to get mgmt version, err: %d, status: 0x%x, out size: 0x%x\n",
+			err, up_ver.status, out_size);
+		return -EINVAL;
+	}
+
+	snprintf(mgmt_ver, HINIC_MGMT_VERSION_MAX_LEN, "%s", up_ver.ver);
+
+	return 0;
+}
+
+int hinic_get_link_mode(struct hinic_hwdev *hwdev,
+			struct hinic_link_mode_cmd *link_mode)
+{
+	u16 out_size;
+	int err;
+
+	if (!hwdev || !link_mode)
+		return -EINVAL;
+
+	link_mode->func_id = HINIC_HWIF_FUNC_IDX(hwdev->hwif);
+	out_size = sizeof(*link_mode);
+
+	err = hinic_port_msg_cmd(hwdev, HINIC_PORT_CMD_GET_LINK_MODE,
+				 link_mode, sizeof(*link_mode),
+				 link_mode, &out_size);
+	if (err || !out_size || link_mode->status) {
+		dev_err(&hwdev->hwif->pdev->dev,
+			"Failed to get link mode, err: %d, status: 0x%x, out size: 0x%x\n",
+			err, link_mode->status, out_size);
+		return -EIO;
+	}
+
+	return 0;
+}
+
+int hinic_set_autoneg(struct hinic_hwdev *hwdev, bool enable)
+{
+	struct hinic_set_autoneg_cmd autoneg = {0};
+	u16 out_size = sizeof(autoneg);
+	int err;
+
+	if (!hwdev)
+		return -EINVAL;
+
+	autoneg.func_id = HINIC_HWIF_FUNC_IDX(hwdev->hwif);
+	autoneg.enable = enable;
+
+	err = hinic_port_msg_cmd(hwdev, HINIC_PORT_CMD_SET_AUTONEG,
+				 &autoneg, sizeof(autoneg),
+				 &autoneg, &out_size);
+	if (err || !out_size || autoneg.status) {
+		dev_err(&hwdev->hwif->pdev->dev, "Failed to %s autoneg, err: %d, status: 0x%x, out size: 0x%x\n",
+			enable ? "enable" : "disable", err, autoneg.status,
+			out_size);
+		return -EIO;
+	}
+
+	return 0;
+}
+
+int hinic_set_speed(struct hinic_hwdev *hwdev, enum nic_speed_level speed)
+{
+	struct hinic_speed_cmd speed_info = {0};
+	u16 out_size = sizeof(speed_info);
+	int err;
+
+	if (!hwdev)
+		return -EINVAL;
+
+	speed_info.func_id = HINIC_HWIF_FUNC_IDX(hwdev->hwif);
+	speed_info.speed = speed;
+
+	err = hinic_port_msg_cmd(hwdev, HINIC_PORT_CMD_SET_SPEED,
+				 &speed_info, sizeof(speed_info),
+				 &speed_info, &out_size);
+	if (err || !out_size || speed_info.status) {
+		dev_err(&hwdev->hwif->pdev->dev,
+			"Failed to set speed, err: %d, status: 0x%x, out size: 0x%x\n",
+			err, speed_info.status, out_size);
+		return -EIO;
+	}
+
+	return 0;
+}
+
+int hinic_set_link_settings(struct hinic_hwdev *hwdev,
+			    struct hinic_link_ksettings_info *info)
+{
+	u16 out_size = sizeof(*info);
+	int err;
+
+	err = hinic_hilink_msg_cmd(hwdev, HINIC_HILINK_CMD_SET_LINK_SETTINGS,
+				   info, sizeof(*info), info, &out_size);
+	if ((info->status != HINIC_MGMT_CMD_UNSUPPORTED &&
+	     info->status) || err || !out_size) {
+		dev_err(&hwdev->hwif->pdev->dev,
+			"Failed to set link settings, err: %d, status: 0x%x, out size: 0x%x\n",
+			err, info->status, out_size);
+		return -EFAULT;
+	}
+
+	return info->status;
+}
+
+int hinic_get_hw_pause_info(struct hinic_hwdev *hwdev,
+			    struct hinic_pause_config *pause_info)
+{
+	u16 out_size = sizeof(*pause_info);
+	int err;
+
+	pause_info->func_id = HINIC_HWIF_FUNC_IDX(hwdev->hwif);
+
+	err = hinic_port_msg_cmd(hwdev, HINIC_PORT_CMD_GET_PAUSE_INFO,
+				 pause_info, sizeof(*pause_info),
+				 pause_info, &out_size);
+	if (err || !out_size || pause_info->status) {
+		dev_err(&hwdev->hwif->pdev->dev, "Failed to get pause info, err: %d, status: 0x%x, out size: 0x%x\n",
+			err, pause_info->status, out_size);
+		return -EIO;
+	}
+
+	return 0;
+}
+
+int hinic_set_hw_pause_info(struct hinic_hwdev *hwdev,
+			    struct hinic_pause_config *pause_info)
+{
+	u16 out_size = sizeof(*pause_info);
+	int err;
+
+	pause_info->func_id = HINIC_HWIF_FUNC_IDX(hwdev->hwif);
+
+	err = hinic_port_msg_cmd(hwdev, HINIC_PORT_CMD_SET_PAUSE_INFO,
+				 pause_info, sizeof(*pause_info),
+				 pause_info, &out_size);
+	if (err || !out_size || pause_info->status) {
+		dev_err(&hwdev->hwif->pdev->dev, "Failed to set pause info, err: %d, status: 0x%x, out size: 0x%x\n",
+			err, pause_info->status, out_size);
+		return -EIO;
+	}
+
+	return 0;
+}
+
+int hinic_dcb_set_pfc(struct hinic_hwdev *hwdev, u8 pfc_en, u8 pfc_bitmap)
+{
+	struct hinic_nic_cfg *nic_cfg = &hwdev->func_to_io.nic_cfg;
+	struct hinic_set_pfc pfc = {0};
+	u16 out_size = sizeof(pfc);
+	int err;
+
+	if (HINIC_IS_VF(hwdev->hwif))
+		return 0;
+
+	mutex_lock(&nic_cfg->cfg_mutex);
+
+	pfc.func_id = HINIC_HWIF_FUNC_IDX(hwdev->hwif);
+	pfc.pfc_bitmap = pfc_bitmap;
+	pfc.pfc_en = pfc_en;
+
+	err = hinic_port_msg_cmd(hwdev, HINIC_PORT_CMD_SET_PFC,
+				 &pfc, sizeof(pfc), &pfc, &out_size);
+	if (err || pfc.status || !out_size) {
+		dev_err(&hwdev->hwif->pdev->dev, "Failed to %s pfc, err: %d, status: 0x%x, out size: 0x%x\n",
+			pfc_en ? "enable" : "disable", err, pfc.status,
+			out_size);
+		mutex_unlock(&nic_cfg->cfg_mutex);
+		return -EIO;
+	}
+
+	/* pause settings is opposite from pfc */
+	nic_cfg->rx_pause = pfc_en ? 0 : 1;
+	nic_cfg->tx_pause = pfc_en ? 0 : 1;
+
+	mutex_unlock(&nic_cfg->cfg_mutex);
+
+	return 0;
+}
+
+int hinic_set_loopback_mode(struct hinic_hwdev *hwdev, u32 mode, u32 enable)
+{
+	struct hinic_port_loopback lb = {0};
+	u16 out_size = sizeof(lb);
+	int err;
+
+	lb.mode = mode;
+	lb.en = enable;
+
+	if (mode < LOOP_MODE_MIN || mode > LOOP_MODE_MAX) {
+		dev_err(&hwdev->hwif->pdev->dev,
+			"Invalid loopback mode %d to set\n", mode);
+		return -EINVAL;
+	}
+
+	err = hinic_port_msg_cmd(hwdev, HINIC_PORT_CMD_SET_LOOPBACK_MODE,
+				 &lb, sizeof(lb), &lb, &out_size);
+	if (err || !out_size || lb.status) {
+		dev_err(&hwdev->hwif->pdev->dev,
+			"Failed to set loopback mode %d en %d, err: %d, status: 0x%x, out size: 0x%x\n",
+			mode, enable, err, lb.status, out_size);
+		return -EIO;
+	}
+
+	return 0;
+}
+
+static int _set_led_status(struct hinic_hwdev *hwdev, u8 port,
+			   enum hinic_led_type type,
+			   enum hinic_led_mode mode, u8 reset)
+{
+	struct hinic_led_info led_info = {0};
+	u16 out_size = sizeof(led_info);
+	struct hinic_pfhwdev *pfhwdev;
+	int err;
+
+	pfhwdev = container_of(hwdev, struct hinic_pfhwdev, hwdev);
+
+	led_info.port = port;
+	led_info.reset = reset;
+
+	led_info.type = type;
+	led_info.mode = mode;
+
+	err = hinic_msg_to_mgmt(&pfhwdev->pf_to_mgmt, HINIC_MOD_COMM,
+				HINIC_COMM_CMD_SET_LED_STATUS,
+				&led_info, sizeof(led_info),
+				&led_info, &out_size, HINIC_MGMT_MSG_SYNC);
+	if (err || led_info.status || !out_size) {
+		dev_err(&hwdev->hwif->pdev->dev, "Failed to set led status, err: %d, status: 0x%x, out size: 0x%x\n",
+			err, led_info.status, out_size);
+		return -EIO;
+	}
+
+	return 0;
+}
+
+int hinic_set_led_status(struct hinic_hwdev *hwdev, u8 port,
+			 enum hinic_led_type type, enum hinic_led_mode mode)
+{
+	if (!hwdev)
+		return -EINVAL;
+
+	return _set_led_status(hwdev, port, type, mode, 0);
+}
+
+int hinic_reset_led_status(struct hinic_hwdev *hwdev, u8 port)
+{
+	int err;
+
+	if (!hwdev)
+		return -EINVAL;
+
+	err = _set_led_status(hwdev, port, HINIC_LED_TYPE_INVALID,
+			      HINIC_LED_MODE_INVALID, 1);
+	if (err)
+		dev_err(&hwdev->hwif->pdev->dev,
+			"Failed to reset led status\n");
+
+	return err;
+}
+
+static bool hinic_if_sfp_absent(struct hinic_hwdev *hwdev)
+{
+	struct hinic_cmd_get_light_module_abs sfp_abs = {0};
+	u16 out_size = sizeof(sfp_abs);
+	u8 port_id = hwdev->port_id;
+	int err;
+
+	sfp_abs.port_id = port_id;
+	err = hinic_port_msg_cmd(hwdev, HINIC_PORT_CMD_GET_SFP_ABS,
+				 &sfp_abs, sizeof(sfp_abs), &sfp_abs,
+				 &out_size);
+	if (sfp_abs.status || err || !out_size) {
+		dev_err(&hwdev->hwif->pdev->dev,
+			"Failed to get port%d sfp absent status, err: %d, status: 0x%x, out size: 0x%x\n",
+			port_id, err, sfp_abs.status, out_size);
+		return true;
+	}
+
+	return ((sfp_abs.abs_status == 0) ? false : true);
+}
+
+int hinic_get_sfp_eeprom(struct hinic_hwdev *hwdev, u8 *data, u16 *len)
+{
+	struct hinic_cmd_get_std_sfp_info sfp_info = {0};
+	u16 out_size = sizeof(sfp_info);
+	u8 port_id;
+	int err;
+
+	if (!hwdev || !data || !len)
+		return -EINVAL;
+
+	port_id = hwdev->port_id;
+
+	if (hinic_if_sfp_absent(hwdev))
+		return -ENXIO;
+
+	sfp_info.port_id = port_id;
+	err = hinic_port_msg_cmd(hwdev, HINIC_PORT_CMD_GET_STD_SFP_INFO,
+				 &sfp_info, sizeof(sfp_info), &sfp_info,
+				 &out_size);
+	if (sfp_info.status || err || !out_size) {
+		dev_err(&hwdev->hwif->pdev->dev,
+			"Failed to get port%d sfp eeprom information, err: %d, status: 0x%x, out size: 0x%x\n",
+			port_id, err, sfp_info.status, out_size);
+		return -EIO;
+	}
+
+	*len = min_t(u16, sfp_info.eeprom_len, STD_SFP_INFO_MAX_SIZE);
+	memcpy(data, sfp_info.sfp_info, STD_SFP_INFO_MAX_SIZE);
+
+	return 0;
+}
+
+int hinic_get_sfp_type(struct hinic_hwdev *hwdev, u8 *data0, u8 *data1)
+{
+	u8 sfp_data[STD_SFP_INFO_MAX_SIZE];
+	u16 len;
+	int err;
+
+	if (hinic_if_sfp_absent(hwdev))
+		return -ENXIO;
+
+	err = hinic_get_sfp_eeprom(hwdev, sfp_data, &len);
+	if (err)
+		return err;
+
+	*data0 = sfp_data[0];
+	*data1 = sfp_data[1];
+
+	return 0;
+}
diff --git a/drivers/net/ethernet/huawei/hinic/hinic_port.h b/drivers/net/ethernet/huawei/hinic/hinic_port.h
new file mode 100644
index 000000000000..c8694ac7c702
--- /dev/null
+++ b/drivers/net/ethernet/huawei/hinic/hinic_port.h
@@ -0,0 +1,900 @@
+/* SPDX-License-Identifier: GPL-2.0-only */
+/*
+ * Huawei HiNIC PCI Express Linux driver
+ * Copyright(c) 2017 Huawei Technologies Co., Ltd
+ */
+
+#ifndef HINIC_PORT_H
+#define HINIC_PORT_H
+
+#include <linux/types.h>
+#include <linux/ethtool.h>
+#include <linux/etherdevice.h>
+#include <linux/bitops.h>
+
+#include "hinic_dev.h"
+
+#define HINIC_RSS_KEY_SIZE	40
+#define HINIC_RSS_INDIR_SIZE	256
+#define HINIC_PORT_STATS_VERSION	0
+#define HINIC_FW_VERSION_NAME	16
+#define HINIC_COMPILE_TIME_LEN	20
+#define HINIC_MGMT_VERSION_MAX_LEN	32
+
+struct hinic_version_info {
+	u8 status;
+	u8 version;
+	u8 rsvd[6];
+
+	u8 ver[HINIC_FW_VERSION_NAME];
+	u8 time[HINIC_COMPILE_TIME_LEN];
+};
+
+enum hinic_rx_mode {
+	HINIC_RX_MODE_UC        = BIT(0),
+	HINIC_RX_MODE_MC        = BIT(1),
+	HINIC_RX_MODE_BC        = BIT(2),
+	HINIC_RX_MODE_MC_ALL    = BIT(3),
+	HINIC_RX_MODE_PROMISC   = BIT(4),
+};
+
+enum hinic_port_link_state {
+	HINIC_LINK_STATE_DOWN,
+	HINIC_LINK_STATE_UP,
+};
+
+enum hinic_port_state {
+	HINIC_PORT_DISABLE      = 0,
+	HINIC_PORT_ENABLE       = 3,
+};
+
+enum hinic_func_port_state {
+	HINIC_FUNC_PORT_DISABLE = 0,
+	HINIC_FUNC_PORT_ENABLE  = 2,
+};
+
+enum hinic_autoneg_cap {
+	HINIC_AUTONEG_UNSUPPORTED,
+	HINIC_AUTONEG_SUPPORTED,
+};
+
+enum hinic_autoneg_state {
+	HINIC_AUTONEG_DISABLED,
+	HINIC_AUTONEG_ACTIVE,
+};
+
+enum hinic_duplex {
+	HINIC_DUPLEX_HALF,
+	HINIC_DUPLEX_FULL,
+};
+
+enum hinic_speed {
+	HINIC_SPEED_10MB_LINK = 0,
+	HINIC_SPEED_100MB_LINK,
+	HINIC_SPEED_1000MB_LINK,
+	HINIC_SPEED_10GB_LINK,
+	HINIC_SPEED_25GB_LINK,
+	HINIC_SPEED_40GB_LINK,
+	HINIC_SPEED_100GB_LINK,
+
+	HINIC_SPEED_UNKNOWN = 0xFF,
+};
+
+enum hinic_link_mode {
+	HINIC_10GE_BASE_KR = 0,
+	HINIC_40GE_BASE_KR4 = 1,
+	HINIC_40GE_BASE_CR4 = 2,
+	HINIC_100GE_BASE_KR4 = 3,
+	HINIC_100GE_BASE_CR4 = 4,
+	HINIC_25GE_BASE_KR_S = 5,
+	HINIC_25GE_BASE_CR_S = 6,
+	HINIC_25GE_BASE_KR = 7,
+	HINIC_25GE_BASE_CR = 8,
+	HINIC_GE_BASE_KX = 9,
+	HINIC_LINK_MODE_NUMBERS,
+
+	HINIC_SUPPORTED_UNKNOWN = 0xFFFF,
+};
+
+enum hinic_port_type {
+	HINIC_PORT_TP,		/* BASET */
+	HINIC_PORT_AUI,
+	HINIC_PORT_MII,
+	HINIC_PORT_FIBRE,	/* OPTICAL */
+	HINIC_PORT_BNC,
+	HINIC_PORT_ELEC,
+	HINIC_PORT_COPPER,	/* PORT_DA */
+	HINIC_PORT_AOC,
+	HINIC_PORT_BACKPLANE,
+	HINIC_PORT_NONE = 0xEF,
+	HINIC_PORT_OTHER = 0xFF,
+};
+
+enum hinic_valid_link_settings {
+	HILINK_LINK_SET_SPEED = 0x1,
+	HILINK_LINK_SET_AUTONEG = 0x2,
+	HILINK_LINK_SET_FEC = 0x4,
+};
+
+enum hinic_tso_state {
+	HINIC_TSO_DISABLE = 0,
+	HINIC_TSO_ENABLE  = 1,
+};
+
+struct hinic_port_mac_cmd {
+	u8              status;
+	u8              version;
+	u8              rsvd0[6];
+
+	u16             func_idx;
+	u16             vlan_id;
+	u16             rsvd1;
+	unsigned char   mac[ETH_ALEN];
+};
+
+struct hinic_port_mtu_cmd {
+	u8      status;
+	u8      version;
+	u8      rsvd0[6];
+
+	u16     func_idx;
+	u16     rsvd1;
+	u32     mtu;
+};
+
+struct hinic_port_vlan_cmd {
+	u8      status;
+	u8      version;
+	u8      rsvd0[6];
+
+	u16     func_idx;
+	u16     vlan_id;
+};
+
+struct hinic_port_rx_mode_cmd {
+	u8      status;
+	u8      version;
+	u8      rsvd0[6];
+
+	u16     func_idx;
+	u16     rsvd;
+	u32     rx_mode;
+};
+
+struct hinic_port_link_cmd {
+	u8      status;
+	u8      version;
+	u8      rsvd0[6];
+
+	u16     func_idx;
+	u8      state;
+	u8      rsvd1;
+};
+
+struct hinic_port_state_cmd {
+	u8      status;
+	u8      version;
+	u8      rsvd0[6];
+
+	u8      state;
+	u8      rsvd1[3];
+};
+
+struct hinic_port_link_status {
+	u8      status;
+	u8      version;
+	u8      rsvd0[6];
+
+	u16     func_id;
+	u8      link;
+	u8      port_id;
+};
+
+struct hinic_cable_plug_event {
+	u8	status;
+	u8	version;
+	u8	rsvd0[6];
+
+	u16	func_id;
+	u8	plugged; /* 0: unplugged, 1: plugged */
+	u8	port_id;
+};
+
+enum link_err_type {
+	LINK_ERR_MODULE_UNRECOGENIZED,
+	LINK_ERR_NUM,
+};
+
+struct hinic_link_err_event {
+	u8	status;
+	u8	version;
+	u8	rsvd0[6];
+
+	u16	func_id;
+	u8	err_type;
+	u8	port_id;
+};
+
+struct hinic_port_func_state_cmd {
+	u8      status;
+	u8      version;
+	u8      rsvd0[6];
+
+	u16     func_idx;
+	u16     rsvd1;
+	u8      state;
+	u8      rsvd2[3];
+};
+
+struct hinic_port_cap {
+	u8      status;
+	u8      version;
+	u8      rsvd0[6];
+
+	u16     func_idx;
+	u16     rsvd1;
+	u8      port_type;
+	u8      autoneg_cap;
+	u8      autoneg_state;
+	u8      duplex;
+	u8      speed;
+	u8      rsvd2[3];
+};
+
+struct hinic_link_mode_cmd {
+	u8	status;
+	u8	version;
+	u8	rsvd0[6];
+
+	u16	func_id;
+	u16	rsvd1;
+	u16	supported;	/* 0xFFFF represents invalid value */
+	u16	advertised;
+};
+
+struct hinic_speed_cmd {
+	u8	status;
+	u8	version;
+	u8	rsvd0[6];
+
+	u16	func_id;
+	u16	speed;
+};
+
+struct hinic_set_autoneg_cmd {
+	u8	status;
+	u8	version;
+	u8	rsvd0[6];
+
+	u16	func_id;
+	u16	enable;	/* 1: enable , 0: disable */
+};
+
+struct hinic_link_ksettings_info {
+	u8	status;
+	u8	version;
+	u8	rsvd0[6];
+
+	u16	func_id;
+	u16	rsvd1;
+
+	u32	valid_bitmap;
+	u32	speed;		/* enum nic_speed_level */
+	u8	autoneg;	/* 0 - off; 1 - on */
+	u8	fec;		/* 0 - RSFEC; 1 - BASEFEC; 2 - NOFEC */
+	u8	rsvd2[18];	/* reserved for duplex, port, etc. */
+};
+
+struct hinic_tso_config {
+	u8	status;
+	u8	version;
+	u8	rsvd0[6];
+
+	u16	func_id;
+	u16	rsvd1;
+	u8	tso_en;
+	u8	resv2[3];
+};
+
+struct hinic_checksum_offload {
+	u8	status;
+	u8	version;
+	u8	rsvd0[6];
+
+	u16	func_id;
+	u16	rsvd1;
+	u32	rx_csum_offload;
+};
+
+struct hinic_rq_num {
+	u8	status;
+	u8	version;
+	u8	rsvd0[6];
+
+	u16	func_id;
+	u16	rsvd1[33];
+	u32	num_rqs;
+	u32	rq_depth;
+};
+
+struct hinic_lro_config {
+	u8	status;
+	u8	version;
+	u8	rsvd0[6];
+
+	u16	func_id;
+	u16	rsvd1;
+	u8	lro_ipv4_en;
+	u8	lro_ipv6_en;
+	u8	lro_max_wqe_num;
+	u8	resv2[13];
+};
+
+struct hinic_lro_timer {
+	u8	status;
+	u8	version;
+	u8	rsvd0[6];
+
+	u8	type;   /* 0: set timer value, 1: get timer value */
+	u8	enable; /* when set lro time, enable should be 1 */
+	u16	rsvd1;
+	u32	timer;
+};
+
+struct hinic_vlan_cfg {
+	u8      status;
+	u8      version;
+	u8      rsvd0[6];
+
+	u16     func_id;
+	u8      vlan_rx_offload;
+	u8      rsvd1[5];
+};
+
+struct hinic_vlan_filter {
+	u8	status;
+	u8	version;
+	u8	rsvd0[6];
+
+	u16	func_idx;
+	u8	rsvd1[2];
+	u32	enable;
+};
+
+struct hinic_rss_template_mgmt {
+	u8	status;
+	u8	version;
+	u8	rsvd0[6];
+
+	u16	func_id;
+	u8	cmd;
+	u8	template_id;
+	u8	rsvd1[4];
+};
+
+struct hinic_rss_template_key {
+	u8	status;
+	u8	version;
+	u8	rsvd0[6];
+
+	u16	func_id;
+	u8	template_id;
+	u8	rsvd1;
+	u8	key[HINIC_RSS_KEY_SIZE];
+};
+
+struct hinic_rss_context_tbl {
+	u32 group_index;
+	u32 offset;
+	u32 size;
+	u32 rsvd;
+	u32 ctx;
+};
+
+struct hinic_rss_context_table {
+	u8      status;
+	u8      version;
+	u8      rsvd0[6];
+
+	u16     func_id;
+	u8      template_id;
+	u8      rsvd1;
+	u32     context;
+};
+
+struct hinic_rss_indirect_tbl {
+	u32 group_index;
+	u32 offset;
+	u32 size;
+	u32 rsvd;
+	u8 entry[HINIC_RSS_INDIR_SIZE];
+};
+
+struct hinic_rss_indir_table {
+	u8      status;
+	u8      version;
+	u8      rsvd0[6];
+
+	u16     func_id;
+	u8      template_id;
+	u8      rsvd1;
+	u8      indir[HINIC_RSS_INDIR_SIZE];
+};
+
+struct hinic_rss_key {
+	u8	status;
+	u8	version;
+	u8	rsvd0[6];
+
+	u16	func_id;
+	u8	template_id;
+	u8	rsvd1;
+	u8	key[HINIC_RSS_KEY_SIZE];
+};
+
+struct hinic_rss_engine_type {
+	u8	status;
+	u8	version;
+	u8	rsvd0[6];
+
+	u16	func_id;
+	u8	template_id;
+	u8	hash_engine;
+	u8	rsvd1[4];
+};
+
+struct hinic_rss_config {
+	u8	status;
+	u8	version;
+	u8	rsvd0[6];
+
+	u16	func_id;
+	u8	rss_en;
+	u8	template_id;
+	u8	rq_priority_number;
+	u8	rsvd1[11];
+};
+
+struct hinic_stats {
+	char name[ETH_GSTRING_LEN];
+	u32 size;
+	int offset;
+};
+
+struct hinic_vport_stats {
+	u64 tx_unicast_pkts_vport;
+	u64 tx_unicast_bytes_vport;
+	u64 tx_multicast_pkts_vport;
+	u64 tx_multicast_bytes_vport;
+	u64 tx_broadcast_pkts_vport;
+	u64 tx_broadcast_bytes_vport;
+
+	u64 rx_unicast_pkts_vport;
+	u64 rx_unicast_bytes_vport;
+	u64 rx_multicast_pkts_vport;
+	u64 rx_multicast_bytes_vport;
+	u64 rx_broadcast_pkts_vport;
+	u64 rx_broadcast_bytes_vport;
+
+	u64 tx_discard_vport;
+	u64 rx_discard_vport;
+	u64 tx_err_vport;
+	u64 rx_err_vport;
+};
+
+struct hinic_phy_port_stats {
+	u64 mac_rx_total_pkt_num;
+	u64 mac_rx_total_oct_num;
+	u64 mac_rx_bad_pkt_num;
+	u64 mac_rx_bad_oct_num;
+	u64 mac_rx_good_pkt_num;
+	u64 mac_rx_good_oct_num;
+	u64 mac_rx_uni_pkt_num;
+	u64 mac_rx_multi_pkt_num;
+	u64 mac_rx_broad_pkt_num;
+
+	u64 mac_tx_total_pkt_num;
+	u64 mac_tx_total_oct_num;
+	u64 mac_tx_bad_pkt_num;
+	u64 mac_tx_bad_oct_num;
+	u64 mac_tx_good_pkt_num;
+	u64 mac_tx_good_oct_num;
+	u64 mac_tx_uni_pkt_num;
+	u64 mac_tx_multi_pkt_num;
+	u64 mac_tx_broad_pkt_num;
+
+	u64 mac_rx_fragment_pkt_num;
+	u64 mac_rx_undersize_pkt_num;
+	u64 mac_rx_undermin_pkt_num;
+	u64 mac_rx_64_oct_pkt_num;
+	u64 mac_rx_65_127_oct_pkt_num;
+	u64 mac_rx_128_255_oct_pkt_num;
+	u64 mac_rx_256_511_oct_pkt_num;
+	u64 mac_rx_512_1023_oct_pkt_num;
+	u64 mac_rx_1024_1518_oct_pkt_num;
+	u64 mac_rx_1519_2047_oct_pkt_num;
+	u64 mac_rx_2048_4095_oct_pkt_num;
+	u64 mac_rx_4096_8191_oct_pkt_num;
+	u64 mac_rx_8192_9216_oct_pkt_num;
+	u64 mac_rx_9217_12287_oct_pkt_num;
+	u64 mac_rx_12288_16383_oct_pkt_num;
+	u64 mac_rx_1519_max_bad_pkt_num;
+	u64 mac_rx_1519_max_good_pkt_num;
+	u64 mac_rx_oversize_pkt_num;
+	u64 mac_rx_jabber_pkt_num;
+
+	u64 mac_rx_pause_num;
+	u64 mac_rx_pfc_pkt_num;
+	u64 mac_rx_pfc_pri0_pkt_num;
+	u64 mac_rx_pfc_pri1_pkt_num;
+	u64 mac_rx_pfc_pri2_pkt_num;
+	u64 mac_rx_pfc_pri3_pkt_num;
+	u64 mac_rx_pfc_pri4_pkt_num;
+	u64 mac_rx_pfc_pri5_pkt_num;
+	u64 mac_rx_pfc_pri6_pkt_num;
+	u64 mac_rx_pfc_pri7_pkt_num;
+	u64 mac_rx_control_pkt_num;
+	u64 mac_rx_y1731_pkt_num;
+	u64 mac_rx_sym_err_pkt_num;
+	u64 mac_rx_fcs_err_pkt_num;
+	u64 mac_rx_send_app_good_pkt_num;
+	u64 mac_rx_send_app_bad_pkt_num;
+
+	u64 mac_tx_fragment_pkt_num;
+	u64 mac_tx_undersize_pkt_num;
+	u64 mac_tx_undermin_pkt_num;
+	u64 mac_tx_64_oct_pkt_num;
+	u64 mac_tx_65_127_oct_pkt_num;
+	u64 mac_tx_128_255_oct_pkt_num;
+	u64 mac_tx_256_511_oct_pkt_num;
+	u64 mac_tx_512_1023_oct_pkt_num;
+	u64 mac_tx_1024_1518_oct_pkt_num;
+	u64 mac_tx_1519_2047_oct_pkt_num;
+	u64 mac_tx_2048_4095_oct_pkt_num;
+	u64 mac_tx_4096_8191_oct_pkt_num;
+	u64 mac_tx_8192_9216_oct_pkt_num;
+	u64 mac_tx_9217_12287_oct_pkt_num;
+	u64 mac_tx_12288_16383_oct_pkt_num;
+	u64 mac_tx_1519_max_bad_pkt_num;
+	u64 mac_tx_1519_max_good_pkt_num;
+	u64 mac_tx_oversize_pkt_num;
+	u64 mac_tx_jabber_pkt_num;
+
+	u64 mac_tx_pause_num;
+	u64 mac_tx_pfc_pkt_num;
+	u64 mac_tx_pfc_pri0_pkt_num;
+	u64 mac_tx_pfc_pri1_pkt_num;
+	u64 mac_tx_pfc_pri2_pkt_num;
+	u64 mac_tx_pfc_pri3_pkt_num;
+	u64 mac_tx_pfc_pri4_pkt_num;
+	u64 mac_tx_pfc_pri5_pkt_num;
+	u64 mac_tx_pfc_pri6_pkt_num;
+	u64 mac_tx_pfc_pri7_pkt_num;
+	u64 mac_tx_control_pkt_num;
+	u64 mac_tx_y1731_pkt_num;
+	u64 mac_tx_1588_pkt_num;
+	u64 mac_tx_err_all_pkt_num;
+	u64 mac_tx_from_app_good_pkt_num;
+	u64 mac_tx_from_app_bad_pkt_num;
+
+	u64 mac_rx_higig2_ext_pkt_num;
+	u64 mac_rx_higig2_message_pkt_num;
+	u64 mac_rx_higig2_error_pkt_num;
+	u64 mac_rx_higig2_cpu_ctrl_pkt_num;
+	u64 mac_rx_higig2_unicast_pkt_num;
+	u64 mac_rx_higig2_broadcast_pkt_num;
+	u64 mac_rx_higig2_l2_multicast_pkt_num;
+	u64 mac_rx_higig2_l3_multicast_pkt_num;
+
+	u64 mac_tx_higig2_message_pkt_num;
+	u64 mac_tx_higig2_ext_pkt_num;
+	u64 mac_tx_higig2_cpu_ctrl_pkt_num;
+	u64 mac_tx_higig2_unicast_pkt_num;
+	u64 mac_tx_higig2_broadcast_pkt_num;
+	u64 mac_tx_higig2_l2_multicast_pkt_num;
+	u64 mac_tx_higig2_l3_multicast_pkt_num;
+};
+
+struct hinic_port_stats_info {
+	u8	status;
+	u8	version;
+	u8	rsvd0[6];
+
+	u16	func_id;
+	u16	rsvd1;
+	u32	stats_version;
+	u32	stats_size;
+};
+
+struct hinic_port_stats {
+	u8 status;
+	u8 version;
+	u8 rsvd[6];
+
+	struct hinic_phy_port_stats stats;
+};
+
+struct hinic_cmd_vport_stats {
+	u8 status;
+	u8 version;
+	u8 rsvd0[6];
+
+	struct hinic_vport_stats stats;
+};
+
+struct hinic_tx_rate_cfg_max_min {
+	u8	status;
+	u8	version;
+	u8	rsvd0[6];
+
+	u16	func_id;
+	u16	rsvd1;
+	u32	min_rate;
+	u32	max_rate;
+	u8	rsvd2[8];
+};
+
+struct hinic_tx_rate_cfg {
+	u8	status;
+	u8	version;
+	u8	rsvd0[6];
+
+	u16	func_id;
+	u16	rsvd1;
+	u32	tx_rate;
+};
+
+enum nic_speed_level {
+	LINK_SPEED_10MB = 0,
+	LINK_SPEED_100MB,
+	LINK_SPEED_1GB,
+	LINK_SPEED_10GB,
+	LINK_SPEED_25GB,
+	LINK_SPEED_40GB,
+	LINK_SPEED_100GB,
+	LINK_SPEED_LEVELS,
+};
+
+struct hinic_spoofchk_set {
+	u8	status;
+	u8	version;
+	u8	rsvd0[6];
+
+	u8	state;
+	u8	rsvd1;
+	u16	func_id;
+};
+
+struct hinic_pause_config {
+	u8	status;
+	u8	version;
+	u8	rsvd0[6];
+
+	u16	func_id;
+	u16	rsvd1;
+	u32	auto_neg;
+	u32	rx_pause;
+	u32	tx_pause;
+};
+
+struct hinic_set_pfc {
+	u8	status;
+	u8	version;
+	u8	rsvd0[6];
+
+	u16	func_id;
+	u8	pfc_en;
+	u8	pfc_bitmap;
+	u8	rsvd1[4];
+};
+
+/* get or set loopback mode, need to modify by base API */
+#define HINIC_INTERNAL_LP_MODE			5
+#define LOOP_MODE_MIN				1
+#define LOOP_MODE_MAX				6
+
+struct hinic_port_loopback {
+	u8	status;
+	u8	version;
+	u8	rsvd[6];
+
+	u32	mode;
+	u32	en;
+};
+
+struct hinic_led_info {
+	u8	status;
+	u8	version;
+	u8	rsvd0[6];
+
+	u8	port;
+	u8	type;
+	u8	mode;
+	u8	reset;
+};
+
+#define STD_SFP_INFO_MAX_SIZE	640
+
+struct hinic_cmd_get_light_module_abs {
+	u8 status;
+	u8 version;
+	u8 rsvd0[6];
+
+	u8 port_id;
+	u8 abs_status; /* 0:present, 1:absent */
+	u8 rsv[2];
+};
+
+#define STD_SFP_INFO_MAX_SIZE	640
+
+struct hinic_cmd_get_std_sfp_info {
+	u8 status;
+	u8 version;
+	u8 rsvd0[6];
+
+	u8 port_id;
+	u8 wire_type;
+	u16 eeprom_len;
+	u32 rsvd;
+	u8 sfp_info[STD_SFP_INFO_MAX_SIZE];
+};
+
+struct hinic_cmd_update_fw {
+	u8 status;
+	u8 version;
+	u8 rsvd0[6];
+
+	struct {
+		u32 SL:1;
+		u32 SF:1;
+		u32 flag:1;
+		u32 reserved:13;
+		u32 fragment_len:16;
+	} ctl_info;
+
+	struct {
+		u32 FW_section_CRC;
+		u32 FW_section_type;
+	} section_info;
+
+	u32 total_len;
+	u32 setion_total_len;
+	u32 fw_section_version;
+	u32 section_offset;
+	u32 data[384];
+};
+
+int hinic_port_add_mac(struct hinic_dev *nic_dev, const u8 *addr,
+		       u16 vlan_id);
+
+int hinic_port_del_mac(struct hinic_dev *nic_dev, const u8 *addr,
+		       u16 vlan_id);
+
+int hinic_port_get_mac(struct hinic_dev *nic_dev, u8 *addr);
+
+int hinic_port_set_mtu(struct hinic_dev *nic_dev, int new_mtu);
+
+int hinic_port_add_vlan(struct hinic_dev *nic_dev, u16 vlan_id);
+
+int hinic_port_del_vlan(struct hinic_dev *nic_dev, u16 vlan_id);
+
+int hinic_port_set_rx_mode(struct hinic_dev *nic_dev, u32 rx_mode);
+
+int hinic_port_link_state(struct hinic_dev *nic_dev,
+			  enum hinic_port_link_state *link_state);
+
+int hinic_port_set_state(struct hinic_dev *nic_dev,
+			 enum hinic_port_state state);
+
+int hinic_port_set_func_state(struct hinic_dev *nic_dev,
+			      enum hinic_func_port_state state);
+
+int hinic_port_get_cap(struct hinic_dev *nic_dev,
+		       struct hinic_port_cap *port_cap);
+
+int hinic_set_max_qnum(struct hinic_dev *nic_dev, u8 num_rqs);
+
+int hinic_port_set_tso(struct hinic_dev *nic_dev, enum hinic_tso_state state);
+
+int hinic_set_rx_csum_offload(struct hinic_dev *nic_dev, u32 en);
+
+int hinic_set_rx_lro_state(struct hinic_dev *nic_dev, u8 lro_en,
+			   u32 lro_timer, u32 wqe_num);
+
+int hinic_set_rss_type(struct hinic_dev *nic_dev, u32 tmpl_idx,
+		       struct hinic_rss_type rss_type);
+
+int hinic_rss_set_indir_tbl(struct hinic_dev *nic_dev, u32 tmpl_idx,
+			    const u32 *indir_table);
+
+int hinic_rss_set_template_tbl(struct hinic_dev *nic_dev, u32 template_id,
+			       const u8 *temp);
+
+int hinic_rss_set_hash_engine(struct hinic_dev *nic_dev, u8 template_id,
+			      u8 type);
+
+int hinic_rss_cfg(struct hinic_dev *nic_dev, u8 rss_en, u8 template_id);
+
+int hinic_rss_template_alloc(struct hinic_dev *nic_dev, u8 *tmpl_idx);
+
+int hinic_rss_template_free(struct hinic_dev *nic_dev, u8 tmpl_idx);
+
+void hinic_set_ethtool_ops(struct net_device *netdev);
+
+int hinic_get_rss_type(struct hinic_dev *nic_dev, u32 tmpl_idx,
+		       struct hinic_rss_type *rss_type);
+
+int hinic_rss_get_indir_tbl(struct hinic_dev *nic_dev, u32 tmpl_idx,
+			    u32 *indir_table);
+
+int hinic_rss_get_template_tbl(struct hinic_dev *nic_dev, u32 tmpl_idx,
+			       u8 *temp);
+
+int hinic_rss_get_hash_engine(struct hinic_dev *nic_dev, u8 tmpl_idx,
+			      u8 *type);
+
+int hinic_get_phy_port_stats(struct hinic_dev *nic_dev,
+			     struct hinic_phy_port_stats *stats);
+
+int hinic_get_vport_stats(struct hinic_dev *nic_dev,
+			  struct hinic_vport_stats *stats);
+
+int hinic_set_rx_vlan_offload(struct hinic_dev *nic_dev, u8 en);
+
+int hinic_set_vlan_fliter(struct hinic_dev *nic_dev, u32 en);
+
+int hinic_get_mgmt_version(struct hinic_dev *nic_dev, u8 *mgmt_ver);
+
+int hinic_set_link_settings(struct hinic_hwdev *hwdev,
+			    struct hinic_link_ksettings_info *info);
+
+int hinic_get_link_mode(struct hinic_hwdev *hwdev,
+			struct hinic_link_mode_cmd *link_mode);
+
+int hinic_set_autoneg(struct hinic_hwdev *hwdev, bool enable);
+
+int hinic_set_speed(struct hinic_hwdev *hwdev, enum nic_speed_level speed);
+
+int hinic_get_hw_pause_info(struct hinic_hwdev *hwdev,
+			    struct hinic_pause_config *pause_info);
+
+int hinic_set_hw_pause_info(struct hinic_hwdev *hwdev,
+			    struct hinic_pause_config *pause_info);
+
+int hinic_dcb_set_pfc(struct hinic_hwdev *hwdev, u8 pfc_en, u8 pfc_bitmap);
+
+int hinic_set_loopback_mode(struct hinic_hwdev *hwdev, u32 mode, u32 enable);
+
+enum hinic_led_mode {
+	HINIC_LED_MODE_ON,
+	HINIC_LED_MODE_OFF,
+	HINIC_LED_MODE_FORCE_1HZ,
+	HINIC_LED_MODE_FORCE_2HZ,
+	HINIC_LED_MODE_FORCE_4HZ,
+	HINIC_LED_MODE_1HZ,
+	HINIC_LED_MODE_2HZ,
+	HINIC_LED_MODE_4HZ,
+	HINIC_LED_MODE_INVALID,
+};
+
+enum hinic_led_type {
+	HINIC_LED_TYPE_LINK,
+	HINIC_LED_TYPE_LOW_SPEED,
+	HINIC_LED_TYPE_HIGH_SPEED,
+	HINIC_LED_TYPE_INVALID,
+};
+
+int hinic_reset_led_status(struct hinic_hwdev *hwdev, u8 port);
+
+int hinic_set_led_status(struct hinic_hwdev *hwdev, u8 port,
+			 enum hinic_led_type type, enum hinic_led_mode mode);
+
+int hinic_get_sfp_type(struct hinic_hwdev *hwdev, u8 *data0, u8 *data1);
+
+int hinic_get_sfp_eeprom(struct hinic_hwdev *hwdev, u8 *data, u16 *len);
+
+int hinic_open(struct net_device *netdev);
+
+int hinic_close(struct net_device *netdev);
+
+#endif
diff --git a/drivers/net/ethernet/huawei/hinic/hinic_rx.c b/drivers/net/ethernet/huawei/hinic/hinic_rx.c
new file mode 100644
index 000000000000..ceec8be2a73b
--- /dev/null
+++ b/drivers/net/ethernet/huawei/hinic/hinic_rx.c
@@ -0,0 +1,631 @@
+// SPDX-License-Identifier: GPL-2.0-only
+/*
+ * Huawei HiNIC PCI Express Linux driver
+ * Copyright(c) 2017 Huawei Technologies Co., Ltd
+ */
+
+#include <linux/kernel.h>
+#include <linux/types.h>
+#include <linux/errno.h>
+#include <linux/pci.h>
+#include <linux/device.h>
+#include <linux/netdevice.h>
+#include <linux/etherdevice.h>
+#include <linux/u64_stats_sync.h>
+#include <linux/slab.h>
+#include <linux/interrupt.h>
+#include <linux/skbuff.h>
+#include <linux/dma-mapping.h>
+#include <linux/prefetch.h>
+#include <linux/cpumask.h>
+#include <linux/if_vlan.h>
+#include <asm/barrier.h>
+
+#include "hinic_common.h"
+#include "hinic_hw_if.h"
+#include "hinic_hw_wqe.h"
+#include "hinic_hw_wq.h"
+#include "hinic_hw_qp.h"
+#include "hinic_hw_dev.h"
+#include "hinic_rx.h"
+#include "hinic_dev.h"
+
+#define RX_IRQ_NO_PENDING               0
+#define RX_IRQ_NO_COALESC               0
+#define RX_IRQ_NO_LLI_TIMER             0
+#define RX_IRQ_NO_CREDIT                0
+#define RX_IRQ_NO_RESEND_TIMER          0
+#define HINIC_RX_BUFFER_WRITE           16
+
+#define HINIC_RX_IPV6_PKT		7
+#define LRO_PKT_HDR_LEN_IPV4		66
+#define LRO_PKT_HDR_LEN_IPV6		86
+#define LRO_REPLENISH_THLD		256
+
+#define LRO_PKT_HDR_LEN(cqe)		\
+	(HINIC_GET_RX_PKT_TYPE(be32_to_cpu((cqe)->offload_type)) == \
+	 HINIC_RX_IPV6_PKT ? LRO_PKT_HDR_LEN_IPV6 : LRO_PKT_HDR_LEN_IPV4)
+
+/**
+ * hinic_rxq_clean_stats - Clean the statistics of specific queue
+ * @rxq: Logical Rx Queue
+ **/
+static void hinic_rxq_clean_stats(struct hinic_rxq *rxq)
+{
+	struct hinic_rxq_stats *rxq_stats = &rxq->rxq_stats;
+
+	u64_stats_update_begin(&rxq_stats->syncp);
+	rxq_stats->pkts  = 0;
+	rxq_stats->bytes = 0;
+	rxq_stats->errors = 0;
+	rxq_stats->csum_errors = 0;
+	rxq_stats->other_errors = 0;
+	u64_stats_update_end(&rxq_stats->syncp);
+}
+
+/**
+ * hinic_rxq_get_stats - get statistics of Rx Queue
+ * @rxq: Logical Rx Queue
+ * @stats: return updated stats here
+ **/
+void hinic_rxq_get_stats(struct hinic_rxq *rxq, struct hinic_rxq_stats *stats)
+{
+	struct hinic_rxq_stats *rxq_stats = &rxq->rxq_stats;
+	unsigned int start;
+
+	do {
+		start = u64_stats_fetch_begin(&rxq_stats->syncp);
+		stats->pkts = rxq_stats->pkts;
+		stats->bytes = rxq_stats->bytes;
+		stats->errors = rxq_stats->csum_errors +
+				rxq_stats->other_errors;
+		stats->csum_errors = rxq_stats->csum_errors;
+		stats->other_errors = rxq_stats->other_errors;
+	} while (u64_stats_fetch_retry(&rxq_stats->syncp, start));
+}
+
+/**
+ * rxq_stats_init - Initialize the statistics of specific queue
+ * @rxq: Logical Rx Queue
+ **/
+static void rxq_stats_init(struct hinic_rxq *rxq)
+{
+	struct hinic_rxq_stats *rxq_stats = &rxq->rxq_stats;
+
+	u64_stats_init(&rxq_stats->syncp);
+	hinic_rxq_clean_stats(rxq);
+}
+
+static void rx_csum(struct hinic_rxq *rxq, u32 status,
+		    struct sk_buff *skb)
+{
+	struct net_device *netdev = rxq->netdev;
+	u32 csum_err;
+
+	csum_err = HINIC_RQ_CQE_STATUS_GET(status, CSUM_ERR);
+
+	if (!(netdev->features & NETIF_F_RXCSUM))
+		return;
+
+	if (!csum_err) {
+		skb->ip_summed = CHECKSUM_UNNECESSARY;
+	} else {
+		if (!(csum_err & (HINIC_RX_CSUM_HW_CHECK_NONE |
+			HINIC_RX_CSUM_IPSU_OTHER_ERR)))
+			rxq->rxq_stats.csum_errors++;
+		skb->ip_summed = CHECKSUM_NONE;
+	}
+}
+
+/**
+ * rx_alloc_skb - allocate skb and map it to dma address
+ * @rxq: rx queue
+ * @dma_addr: returned dma address for the skb
+ *
+ * Return skb
+ **/
+static struct sk_buff *rx_alloc_skb(struct hinic_rxq *rxq,
+				    dma_addr_t *dma_addr)
+{
+	struct hinic_dev *nic_dev = netdev_priv(rxq->netdev);
+	struct hinic_hwdev *hwdev = nic_dev->hwdev;
+	struct hinic_hwif *hwif = hwdev->hwif;
+	struct pci_dev *pdev = hwif->pdev;
+	struct sk_buff *skb;
+	dma_addr_t addr;
+	int err;
+
+	skb = netdev_alloc_skb_ip_align(rxq->netdev, rxq->rq->buf_sz);
+	if (!skb)
+		return NULL;
+
+	addr = dma_map_single(&pdev->dev, skb->data, rxq->rq->buf_sz,
+			      DMA_FROM_DEVICE);
+	err = dma_mapping_error(&pdev->dev, addr);
+	if (err) {
+		dev_err(&pdev->dev, "Failed to map Rx DMA, err = %d\n", err);
+		goto err_rx_map;
+	}
+
+	*dma_addr = addr;
+	return skb;
+
+err_rx_map:
+	dev_kfree_skb_any(skb);
+	return NULL;
+}
+
+/**
+ * rx_unmap_skb - unmap the dma address of the skb
+ * @rxq: rx queue
+ * @dma_addr: dma address of the skb
+ **/
+static void rx_unmap_skb(struct hinic_rxq *rxq, dma_addr_t dma_addr)
+{
+	struct hinic_dev *nic_dev = netdev_priv(rxq->netdev);
+	struct hinic_hwdev *hwdev = nic_dev->hwdev;
+	struct hinic_hwif *hwif = hwdev->hwif;
+	struct pci_dev *pdev = hwif->pdev;
+
+	dma_unmap_single(&pdev->dev, dma_addr, rxq->rq->buf_sz,
+			 DMA_FROM_DEVICE);
+}
+
+/**
+ * rx_free_skb - unmap and free skb
+ * @rxq: rx queue
+ * @skb: skb to free
+ * @dma_addr: dma address of the skb
+ **/
+static void rx_free_skb(struct hinic_rxq *rxq, struct sk_buff *skb,
+			dma_addr_t dma_addr)
+{
+	rx_unmap_skb(rxq, dma_addr);
+	dev_kfree_skb_any(skb);
+}
+
+/**
+ * rx_alloc_pkts - allocate pkts in rx queue
+ * @rxq: rx queue
+ *
+ * Return number of skbs allocated
+ **/
+static int rx_alloc_pkts(struct hinic_rxq *rxq)
+{
+	struct hinic_dev *nic_dev = netdev_priv(rxq->netdev);
+	struct hinic_rq_wqe *rq_wqe;
+	unsigned int free_wqebbs;
+	struct hinic_sge sge;
+	dma_addr_t dma_addr;
+	struct sk_buff *skb;
+	u16 prod_idx;
+	int i;
+
+	free_wqebbs = hinic_get_rq_free_wqebbs(rxq->rq);
+
+	/* Limit the allocation chunks */
+	if (free_wqebbs > nic_dev->rx_weight)
+		free_wqebbs = nic_dev->rx_weight;
+
+	for (i = 0; i < free_wqebbs; i++) {
+		skb = rx_alloc_skb(rxq, &dma_addr);
+		if (!skb)
+			goto skb_out;
+
+		hinic_set_sge(&sge, dma_addr, skb->len);
+
+		rq_wqe = hinic_rq_get_wqe(rxq->rq, HINIC_RQ_WQE_SIZE,
+					  &prod_idx);
+		if (!rq_wqe) {
+			rx_free_skb(rxq, skb, dma_addr);
+			goto skb_out;
+		}
+
+		hinic_rq_prepare_wqe(rxq->rq, prod_idx, rq_wqe, &sge);
+
+		hinic_rq_write_wqe(rxq->rq, prod_idx, rq_wqe, skb);
+	}
+
+skb_out:
+	if (i) {
+		wmb();  /* write all the wqes before update PI */
+
+		hinic_rq_update(rxq->rq, prod_idx);
+	}
+
+	return i;
+}
+
+/**
+ * free_all_rx_skbs - free all skbs in rx queue
+ * @rxq: rx queue
+ **/
+static void free_all_rx_skbs(struct hinic_rxq *rxq)
+{
+	struct hinic_rq *rq = rxq->rq;
+	struct hinic_hw_wqe *hw_wqe;
+	struct hinic_sge sge;
+	u16 ci;
+
+	while ((hw_wqe = hinic_read_wqe(rq->wq, HINIC_RQ_WQE_SIZE, &ci))) {
+		if (IS_ERR(hw_wqe))
+			break;
+
+		hinic_rq_get_sge(rq, &hw_wqe->rq_wqe, ci, &sge);
+
+		hinic_put_wqe(rq->wq, HINIC_RQ_WQE_SIZE);
+
+		rx_free_skb(rxq, rq->saved_skb[ci], hinic_sge_to_dma(&sge));
+	}
+}
+
+/**
+ * rx_recv_jumbo_pkt - Rx handler for jumbo pkt
+ * @rxq: rx queue
+ * @head_skb: the first skb in the list
+ * @left_pkt_len: left size of the pkt exclude head skb
+ * @ci: consumer index
+ *
+ * Return number of wqes that used for the left of the pkt
+ **/
+static int rx_recv_jumbo_pkt(struct hinic_rxq *rxq, struct sk_buff *head_skb,
+			     unsigned int left_pkt_len, u16 ci)
+{
+	struct sk_buff *skb, *curr_skb = head_skb;
+	struct hinic_rq_wqe *rq_wqe;
+	unsigned int curr_len;
+	struct hinic_sge sge;
+	int num_wqes = 0;
+
+	while (left_pkt_len > 0) {
+		rq_wqe = hinic_rq_read_next_wqe(rxq->rq, HINIC_RQ_WQE_SIZE,
+						&skb, &ci);
+
+		num_wqes++;
+
+		hinic_rq_get_sge(rxq->rq, rq_wqe, ci, &sge);
+
+		rx_unmap_skb(rxq, hinic_sge_to_dma(&sge));
+
+		prefetch(skb->data);
+
+		curr_len = (left_pkt_len > HINIC_RX_BUF_SZ) ? HINIC_RX_BUF_SZ :
+			    left_pkt_len;
+
+		left_pkt_len -= curr_len;
+
+		__skb_put(skb, curr_len);
+
+		if (curr_skb == head_skb)
+			skb_shinfo(head_skb)->frag_list = skb;
+		else
+			curr_skb->next = skb;
+
+		head_skb->len += skb->len;
+		head_skb->data_len += skb->len;
+		head_skb->truesize += skb->truesize;
+
+		curr_skb = skb;
+	}
+
+	return num_wqes;
+}
+
+static void hinic_copy_lp_data(struct hinic_dev *nic_dev,
+			       struct sk_buff *skb)
+{
+	struct net_device *netdev = nic_dev->netdev;
+	u8 *lb_buf = nic_dev->lb_test_rx_buf;
+	int lb_len = nic_dev->lb_pkt_len;
+	int pkt_offset, frag_len, i;
+	void *frag_data = NULL;
+
+	if (nic_dev->lb_test_rx_idx == LP_PKT_CNT) {
+		nic_dev->lb_test_rx_idx = 0;
+		netif_warn(nic_dev, drv, netdev, "Loopback test warning, receive too more test pkts\n");
+	}
+
+	if (skb->len != nic_dev->lb_pkt_len) {
+		netif_warn(nic_dev, drv, netdev, "Wrong packet length\n");
+		nic_dev->lb_test_rx_idx++;
+		return;
+	}
+
+	pkt_offset = nic_dev->lb_test_rx_idx * lb_len;
+	frag_len = (int)skb_headlen(skb);
+	memcpy(lb_buf + pkt_offset, skb->data, frag_len);
+	pkt_offset += frag_len;
+	for (i = 0; i < skb_shinfo(skb)->nr_frags; i++) {
+		frag_data = skb_frag_address(&skb_shinfo(skb)->frags[i]);
+		frag_len = (int)skb_frag_size(&skb_shinfo(skb)->frags[i]);
+		memcpy((lb_buf + pkt_offset), frag_data, frag_len);
+		pkt_offset += frag_len;
+	}
+	nic_dev->lb_test_rx_idx++;
+}
+
+/**
+ * rxq_recv - Rx handler
+ * @rxq: rx queue
+ * @budget: maximum pkts to process
+ *
+ * Return number of pkts received
+ **/
+static int rxq_recv(struct hinic_rxq *rxq, int budget)
+{
+	struct hinic_qp *qp = container_of(rxq->rq, struct hinic_qp, rq);
+	struct net_device *netdev = rxq->netdev;
+	u64 pkt_len = 0, rx_bytes = 0;
+	struct hinic_rq *rq = rxq->rq;
+	struct hinic_rq_wqe *rq_wqe;
+	struct hinic_dev *nic_dev;
+	unsigned int free_wqebbs;
+	struct hinic_rq_cqe *cqe;
+	int num_wqes, pkts = 0;
+	struct hinic_sge sge;
+	unsigned int status;
+	struct sk_buff *skb;
+	u32 offload_type;
+	u16 ci, num_lro;
+	u16 num_wqe = 0;
+	u32 vlan_len;
+	u16 vid;
+
+	nic_dev = netdev_priv(netdev);
+
+	while (pkts < budget) {
+		num_wqes = 0;
+
+		rq_wqe = hinic_rq_read_wqe(rxq->rq, HINIC_RQ_WQE_SIZE, &skb,
+					   &ci);
+		if (!rq_wqe)
+			break;
+
+		/* make sure we read rx_done before packet length */
+		dma_rmb();
+
+		cqe = rq->cqe[ci];
+		status =  be32_to_cpu(cqe->status);
+		hinic_rq_get_sge(rxq->rq, rq_wqe, ci, &sge);
+
+		rx_unmap_skb(rxq, hinic_sge_to_dma(&sge));
+
+		rx_csum(rxq, status, skb);
+
+		prefetch(skb->data);
+
+		pkt_len = sge.len;
+
+		if (pkt_len <= HINIC_RX_BUF_SZ) {
+			__skb_put(skb, pkt_len);
+		} else {
+			__skb_put(skb, HINIC_RX_BUF_SZ);
+			num_wqes = rx_recv_jumbo_pkt(rxq, skb, pkt_len -
+						     HINIC_RX_BUF_SZ, ci);
+		}
+
+		hinic_rq_put_wqe(rq, ci,
+				 (num_wqes + 1) * HINIC_RQ_WQE_SIZE);
+
+		offload_type = be32_to_cpu(cqe->offload_type);
+		vlan_len = be32_to_cpu(cqe->len);
+		if ((netdev->features & NETIF_F_HW_VLAN_CTAG_RX) &&
+		    HINIC_GET_RX_VLAN_OFFLOAD_EN(offload_type)) {
+			vid = HINIC_GET_RX_VLAN_TAG(vlan_len);
+			__vlan_hwaccel_put_tag(skb, htons(ETH_P_8021Q), vid);
+		}
+
+		if (unlikely(nic_dev->flags & HINIC_LP_TEST))
+			hinic_copy_lp_data(nic_dev, skb);
+
+		skb_record_rx_queue(skb, qp->q_id);
+		skb->protocol = eth_type_trans(skb, rxq->netdev);
+
+		napi_gro_receive(&rxq->napi, skb);
+
+		pkts++;
+		rx_bytes += pkt_len;
+
+		num_lro = HINIC_GET_RX_NUM_LRO(status);
+		if (num_lro) {
+			rx_bytes += ((num_lro - 1) *
+				     LRO_PKT_HDR_LEN(cqe));
+
+			num_wqe +=
+			(u16)(pkt_len >> rxq->rx_buff_shift) +
+			((pkt_len & (rxq->buf_len - 1)) ? 1 : 0);
+		}
+
+		cqe->status = 0;
+
+		if (num_wqe >= LRO_REPLENISH_THLD)
+			break;
+	}
+
+	free_wqebbs = hinic_get_rq_free_wqebbs(rxq->rq);
+	if (free_wqebbs > HINIC_RX_BUFFER_WRITE)
+		rx_alloc_pkts(rxq);
+
+	u64_stats_update_begin(&rxq->rxq_stats.syncp);
+	rxq->rxq_stats.pkts += pkts;
+	rxq->rxq_stats.bytes += rx_bytes;
+	u64_stats_update_end(&rxq->rxq_stats.syncp);
+
+	return pkts;
+}
+
+static int rx_poll(struct napi_struct *napi, int budget)
+{
+	struct hinic_rxq *rxq = container_of(napi, struct hinic_rxq, napi);
+	struct hinic_dev *nic_dev = netdev_priv(rxq->netdev);
+	struct hinic_rq *rq = rxq->rq;
+	int pkts;
+
+	pkts = rxq_recv(rxq, budget);
+	if (pkts >= budget)
+		return budget;
+
+	napi_complete(napi);
+
+	if (!HINIC_IS_VF(nic_dev->hwdev->hwif))
+		hinic_hwdev_set_msix_state(nic_dev->hwdev,
+					   rq->msix_entry,
+					   HINIC_MSIX_ENABLE);
+
+	return pkts;
+}
+
+static void rx_add_napi(struct hinic_rxq *rxq)
+{
+	struct hinic_dev *nic_dev = netdev_priv(rxq->netdev);
+
+	netif_napi_add_weight(rxq->netdev, &rxq->napi, rx_poll,
+			      nic_dev->rx_weight);
+	napi_enable(&rxq->napi);
+}
+
+static void rx_del_napi(struct hinic_rxq *rxq)
+{
+	napi_disable(&rxq->napi);
+	netif_napi_del(&rxq->napi);
+}
+
+static irqreturn_t rx_irq(int irq, void *data)
+{
+	struct hinic_rxq *rxq = (struct hinic_rxq *)data;
+	struct hinic_rq *rq = rxq->rq;
+	struct hinic_dev *nic_dev;
+
+	/* Disable the interrupt until napi will be completed */
+	nic_dev = netdev_priv(rxq->netdev);
+	if (!HINIC_IS_VF(nic_dev->hwdev->hwif))
+		hinic_hwdev_set_msix_state(nic_dev->hwdev,
+					   rq->msix_entry,
+					   HINIC_MSIX_DISABLE);
+
+	nic_dev = netdev_priv(rxq->netdev);
+	hinic_hwdev_msix_cnt_set(nic_dev->hwdev, rq->msix_entry);
+
+	napi_schedule(&rxq->napi);
+	return IRQ_HANDLED;
+}
+
+static int rx_request_irq(struct hinic_rxq *rxq)
+{
+	struct hinic_dev *nic_dev = netdev_priv(rxq->netdev);
+	struct hinic_msix_config interrupt_info = {0};
+	struct hinic_intr_coal_info *intr_coal = NULL;
+	struct hinic_hwdev *hwdev = nic_dev->hwdev;
+	struct hinic_rq *rq = rxq->rq;
+	struct hinic_qp *qp;
+	int err;
+
+	qp = container_of(rq, struct hinic_qp, rq);
+
+	rx_add_napi(rxq);
+
+	hinic_hwdev_msix_set(hwdev, rq->msix_entry,
+			     RX_IRQ_NO_PENDING, RX_IRQ_NO_COALESC,
+			     RX_IRQ_NO_LLI_TIMER, RX_IRQ_NO_CREDIT,
+			     RX_IRQ_NO_RESEND_TIMER);
+
+	intr_coal = &nic_dev->rx_intr_coalesce[qp->q_id];
+	interrupt_info.msix_index = rq->msix_entry;
+	interrupt_info.coalesce_timer_cnt = intr_coal->coalesce_timer_cfg;
+	interrupt_info.pending_cnt = intr_coal->pending_limt;
+	interrupt_info.resend_timer_cnt = intr_coal->resend_timer_cfg;
+
+	err = hinic_set_interrupt_cfg(hwdev, &interrupt_info);
+	if (err) {
+		netif_err(nic_dev, drv, rxq->netdev,
+			  "Failed to set RX interrupt coalescing attribute\n");
+		goto err_req_irq;
+	}
+
+	err = request_irq(rq->irq, rx_irq, 0, rxq->irq_name, rxq);
+	if (err)
+		goto err_req_irq;
+
+	cpumask_set_cpu(qp->q_id % num_online_cpus(), &rq->affinity_mask);
+	err = irq_set_affinity_and_hint(rq->irq, &rq->affinity_mask);
+	if (err)
+		goto err_irq_affinity;
+
+	return 0;
+
+err_irq_affinity:
+	free_irq(rq->irq, rxq);
+err_req_irq:
+	rx_del_napi(rxq);
+	return err;
+}
+
+static void rx_free_irq(struct hinic_rxq *rxq)
+{
+	struct hinic_rq *rq = rxq->rq;
+
+	irq_update_affinity_hint(rq->irq, NULL);
+	free_irq(rq->irq, rxq);
+	rx_del_napi(rxq);
+}
+
+/**
+ * hinic_init_rxq - Initialize the Rx Queue
+ * @rxq: Logical Rx Queue
+ * @rq: Hardware Rx Queue to connect the Logical queue with
+ * @netdev: network device to connect the Logical queue with
+ *
+ * Return 0 - Success, negative - Failure
+ **/
+int hinic_init_rxq(struct hinic_rxq *rxq, struct hinic_rq *rq,
+		   struct net_device *netdev)
+{
+	struct hinic_qp *qp = container_of(rq, struct hinic_qp, rq);
+	int err, pkts;
+
+	rxq->netdev = netdev;
+	rxq->rq = rq;
+	rxq->buf_len = HINIC_RX_BUF_SZ;
+	rxq->rx_buff_shift = ilog2(HINIC_RX_BUF_SZ);
+
+	rxq_stats_init(rxq);
+
+	rxq->irq_name = devm_kasprintf(&netdev->dev, GFP_KERNEL,
+				       "%s_rxq%d", netdev->name, qp->q_id);
+	if (!rxq->irq_name)
+		return -ENOMEM;
+
+	pkts = rx_alloc_pkts(rxq);
+	if (!pkts) {
+		err = -ENOMEM;
+		goto err_rx_pkts;
+	}
+
+	err = rx_request_irq(rxq);
+	if (err) {
+		netdev_err(netdev, "Failed to request Rx irq\n");
+		goto err_req_rx_irq;
+	}
+
+	return 0;
+
+err_req_rx_irq:
+err_rx_pkts:
+	free_all_rx_skbs(rxq);
+	devm_kfree(&netdev->dev, rxq->irq_name);
+	return err;
+}
+
+/**
+ * hinic_clean_rxq - Clean the Rx Queue
+ * @rxq: Logical Rx Queue
+ **/
+void hinic_clean_rxq(struct hinic_rxq *rxq)
+{
+	struct net_device *netdev = rxq->netdev;
+
+	rx_free_irq(rxq);
+
+	free_all_rx_skbs(rxq);
+	devm_kfree(&netdev->dev, rxq->irq_name);
+}
diff --git a/drivers/net/ethernet/huawei/hinic/hinic_rx.h b/drivers/net/ethernet/huawei/hinic/hinic_rx.h
new file mode 100644
index 000000000000..8f7bd6a049bd
--- /dev/null
+++ b/drivers/net/ethernet/huawei/hinic/hinic_rx.h
@@ -0,0 +1,51 @@
+/* SPDX-License-Identifier: GPL-2.0-only */
+/*
+ * Huawei HiNIC PCI Express Linux driver
+ * Copyright(c) 2017 Huawei Technologies Co., Ltd
+ */
+
+#ifndef HINIC_RX_H
+#define HINIC_RX_H
+
+#include <linux/types.h>
+#include <linux/netdevice.h>
+#include <linux/u64_stats_sync.h>
+#include <linux/interrupt.h>
+
+#include "hinic_hw_qp.h"
+
+#define HINIC_RX_CSUM_OFFLOAD_EN	0xFFF
+#define HINIC_RX_CSUM_HW_CHECK_NONE	BIT(7)
+#define HINIC_RX_CSUM_IPSU_OTHER_ERR	BIT(8)
+
+struct hinic_rxq_stats {
+	u64                     pkts;
+	u64                     bytes;
+	u64			errors;
+	u64			csum_errors;
+	u64			other_errors;
+	u64			alloc_skb_err;
+	struct u64_stats_sync   syncp;
+};
+
+struct hinic_rxq {
+	struct net_device       *netdev;
+	struct hinic_rq         *rq;
+
+	struct hinic_rxq_stats  rxq_stats;
+
+	char                    *irq_name;
+	u16			buf_len;
+	u32			rx_buff_shift;
+
+	struct napi_struct      napi;
+};
+
+void hinic_rxq_get_stats(struct hinic_rxq *rxq, struct hinic_rxq_stats *stats);
+
+int hinic_init_rxq(struct hinic_rxq *rxq, struct hinic_rq *rq,
+		   struct net_device *netdev);
+
+void hinic_clean_rxq(struct hinic_rxq *rxq);
+
+#endif
diff --git a/drivers/net/ethernet/huawei/hinic/hinic_sriov.c b/drivers/net/ethernet/huawei/hinic/hinic_sriov.c
new file mode 100644
index 000000000000..ee357088d021
--- /dev/null
+++ b/drivers/net/ethernet/huawei/hinic/hinic_sriov.c
@@ -0,0 +1,1364 @@
+// SPDX-License-Identifier: GPL-2.0-only
+/* Huawei HiNIC PCI Express Linux driver
+ * Copyright(c) 2017 Huawei Technologies Co., Ltd
+ */
+
+#include <linux/pci.h>
+#include <linux/if_vlan.h>
+#include <linux/interrupt.h>
+#include <linux/etherdevice.h>
+#include <linux/netdevice.h>
+#include <linux/module.h>
+
+#include "hinic_hw_dev.h"
+#include "hinic_dev.h"
+#include "hinic_hw_mbox.h"
+#include "hinic_hw_cmdq.h"
+#include "hinic_port.h"
+#include "hinic_sriov.h"
+
+static unsigned char set_vf_link_state;
+module_param(set_vf_link_state, byte, 0444);
+MODULE_PARM_DESC(set_vf_link_state, "Set vf link state, 0 represents link auto, 1 represents link always up, 2 represents link always down. - default is 0.");
+
+#define HINIC_VLAN_PRIORITY_SHIFT 13
+#define HINIC_ADD_VLAN_IN_MAC 0x8000
+#define HINIC_TX_RATE_TABLE_FULL 12
+#define HINIC_MAX_QOS 7
+
+static int hinic_set_mac(struct hinic_hwdev *hwdev, const u8 *mac_addr,
+			 u16 vlan_id, u16 func_id)
+{
+	struct hinic_port_mac_cmd mac_info = {0};
+	u16 out_size = sizeof(mac_info);
+	int err;
+
+	mac_info.func_idx = func_id;
+	mac_info.vlan_id = vlan_id;
+	memcpy(mac_info.mac, mac_addr, ETH_ALEN);
+
+	err = hinic_port_msg_cmd(hwdev, HINIC_PORT_CMD_SET_MAC, &mac_info,
+				 sizeof(mac_info), &mac_info, &out_size);
+	if (err || out_size != sizeof(mac_info) ||
+	    (mac_info.status && mac_info.status != HINIC_MGMT_STATUS_EXIST)) {
+		dev_err(&hwdev->func_to_io.hwif->pdev->dev, "Failed to set MAC, err: %d, status: 0x%x, out size: 0x%x\n",
+			err, mac_info.status, out_size);
+		return -EIO;
+	}
+
+	return 0;
+}
+
+static void hinic_notify_vf_link_status(struct hinic_hwdev *hwdev, u16 vf_id,
+					u8 link_status)
+{
+	struct vf_data_storage *vf_infos = hwdev->func_to_io.vf_infos;
+	struct hinic_port_link_status link = {0};
+	u16 out_size = sizeof(link);
+	int err;
+
+	if (vf_infos[HW_VF_ID_TO_OS(vf_id)].registered) {
+		link.link = link_status;
+		link.func_id = hinic_glb_pf_vf_offset(hwdev->hwif) + vf_id;
+		err = hinic_mbox_to_vf(hwdev, HINIC_MOD_L2NIC,
+				       vf_id, HINIC_PORT_CMD_LINK_STATUS_REPORT,
+				       &link, sizeof(link),
+				       &link, &out_size, 0);
+		if (err || !out_size || link.status)
+			dev_err(&hwdev->hwif->pdev->dev,
+				"Send link change event to VF %d failed, err: %d, status: 0x%x, out_size: 0x%x\n",
+				HW_VF_ID_TO_OS(vf_id), err,
+				link.status, out_size);
+	}
+}
+
+/* send link change event mbox msg to active vfs under the pf */
+void hinic_notify_all_vfs_link_changed(struct hinic_hwdev *hwdev,
+				       u8 link_status)
+{
+	struct hinic_func_to_io *nic_io = &hwdev->func_to_io;
+	u16 i;
+
+	nic_io->link_status = link_status;
+	for (i = 1; i <= nic_io->max_vfs; i++) {
+		if (!nic_io->vf_infos[HW_VF_ID_TO_OS(i)].link_forced)
+			hinic_notify_vf_link_status(hwdev, i,  link_status);
+	}
+}
+
+static u16 hinic_vf_info_vlanprio(struct hinic_hwdev *hwdev, int vf_id)
+{
+	struct hinic_func_to_io *nic_io = &hwdev->func_to_io;
+	u16 pf_vlan, vlanprio;
+	u8 pf_qos;
+
+	pf_vlan = nic_io->vf_infos[HW_VF_ID_TO_OS(vf_id)].pf_vlan;
+	pf_qos = nic_io->vf_infos[HW_VF_ID_TO_OS(vf_id)].pf_qos;
+	vlanprio = pf_vlan | pf_qos << HINIC_VLAN_PRIORITY_SHIFT;
+
+	return vlanprio;
+}
+
+static int hinic_set_vf_vlan(struct hinic_hwdev *hwdev, bool add, u16 vid,
+			     u8 qos, int vf_id)
+{
+	struct hinic_vf_vlan_config vf_vlan = {0};
+	u16 out_size = sizeof(vf_vlan);
+	int err;
+	u8 cmd;
+
+	/* VLAN 0 is a special case, don't allow it to be removed */
+	if (!vid && !add)
+		return 0;
+
+	vf_vlan.func_id = hinic_glb_pf_vf_offset(hwdev->hwif) + vf_id;
+	vf_vlan.vlan_id = vid;
+	vf_vlan.qos = qos;
+
+	if (add)
+		cmd = HINIC_PORT_CMD_SET_VF_VLAN;
+	else
+		cmd = HINIC_PORT_CMD_CLR_VF_VLAN;
+
+	err = hinic_port_msg_cmd(hwdev, cmd, &vf_vlan,
+				 sizeof(vf_vlan), &vf_vlan, &out_size);
+	if (err || !out_size || vf_vlan.status) {
+		dev_err(&hwdev->hwif->pdev->dev, "Failed to set VF %d vlan, err: %d, status: 0x%x, out size: 0x%x\n",
+			HW_VF_ID_TO_OS(vf_id), err, vf_vlan.status, out_size);
+		return -EFAULT;
+	}
+
+	return 0;
+}
+
+static int hinic_set_vf_tx_rate_max_min(struct hinic_hwdev *hwdev, u16 vf_id,
+					u32 max_rate, u32 min_rate)
+{
+	struct hinic_func_to_io *nic_io = &hwdev->func_to_io;
+	struct hinic_tx_rate_cfg_max_min rate_cfg = {0};
+	u16 out_size = sizeof(rate_cfg);
+	int err;
+
+	rate_cfg.func_id = hinic_glb_pf_vf_offset(hwdev->hwif) + vf_id;
+	rate_cfg.max_rate = max_rate;
+	rate_cfg.min_rate = min_rate;
+	err = hinic_port_msg_cmd(hwdev, HINIC_PORT_CMD_SET_VF_MAX_MIN_RATE,
+				 &rate_cfg, sizeof(rate_cfg), &rate_cfg,
+				 &out_size);
+	if ((rate_cfg.status != HINIC_MGMT_CMD_UNSUPPORTED &&
+	     rate_cfg.status) || err || !out_size) {
+		dev_err(&hwdev->hwif->pdev->dev, "Failed to set VF(%d) max rate(%d), min rate(%d), err: %d, status: 0x%x, out size: 0x%x\n",
+			HW_VF_ID_TO_OS(vf_id), max_rate, min_rate, err,
+			rate_cfg.status, out_size);
+		return -EIO;
+	}
+
+	if (!rate_cfg.status) {
+		nic_io->vf_infos[HW_VF_ID_TO_OS(vf_id)].max_rate = max_rate;
+		nic_io->vf_infos[HW_VF_ID_TO_OS(vf_id)].min_rate = min_rate;
+	}
+
+	return rate_cfg.status;
+}
+
+static int hinic_set_vf_rate_limit(struct hinic_hwdev *hwdev, u16 vf_id,
+				   u32 tx_rate)
+{
+	struct hinic_func_to_io *nic_io = &hwdev->func_to_io;
+	struct hinic_tx_rate_cfg rate_cfg = {0};
+	u16 out_size = sizeof(rate_cfg);
+	int err;
+
+	rate_cfg.func_id = hinic_glb_pf_vf_offset(hwdev->hwif) + vf_id;
+	rate_cfg.tx_rate = tx_rate;
+	err = hinic_port_msg_cmd(hwdev, HINIC_PORT_CMD_SET_VF_RATE,
+				 &rate_cfg, sizeof(rate_cfg), &rate_cfg,
+				 &out_size);
+	if (err || !out_size || rate_cfg.status) {
+		dev_err(&hwdev->hwif->pdev->dev, "Failed to set VF(%d) rate(%d), err: %d, status: 0x%x, out size: 0x%x\n",
+			HW_VF_ID_TO_OS(vf_id), tx_rate, err, rate_cfg.status,
+			out_size);
+		if (rate_cfg.status)
+			return rate_cfg.status;
+
+		return -EIO;
+	}
+
+	nic_io->vf_infos[HW_VF_ID_TO_OS(vf_id)].max_rate = tx_rate;
+	nic_io->vf_infos[HW_VF_ID_TO_OS(vf_id)].min_rate = 0;
+
+	return 0;
+}
+
+static int hinic_set_vf_tx_rate(struct hinic_hwdev *hwdev, u16 vf_id,
+				u32 max_rate, u32 min_rate)
+{
+	int err;
+
+	err = hinic_set_vf_tx_rate_max_min(hwdev, vf_id, max_rate, min_rate);
+	if (err != HINIC_MGMT_CMD_UNSUPPORTED)
+		return err;
+
+	if (min_rate) {
+		dev_err(&hwdev->hwif->pdev->dev, "Current firmware doesn't support to set min tx rate\n");
+		return -EOPNOTSUPP;
+	}
+
+	dev_info(&hwdev->hwif->pdev->dev, "Current firmware doesn't support to set min tx rate, force min_tx_rate = max_tx_rate\n");
+
+	return hinic_set_vf_rate_limit(hwdev, vf_id, max_rate);
+}
+
+static int hinic_init_vf_config(struct hinic_hwdev *hwdev, u16 vf_id)
+{
+	struct vf_data_storage *vf_info;
+	u16 func_id, vlan_id;
+	int err = 0;
+
+	vf_info = hwdev->func_to_io.vf_infos + HW_VF_ID_TO_OS(vf_id);
+	if (vf_info->pf_set_mac) {
+		func_id = hinic_glb_pf_vf_offset(hwdev->hwif) + vf_id;
+
+		vlan_id = 0;
+
+		err = hinic_set_mac(hwdev, vf_info->vf_mac_addr, vlan_id,
+				    func_id);
+		if (err) {
+			dev_err(&hwdev->func_to_io.hwif->pdev->dev, "Failed to set VF %d MAC\n",
+				HW_VF_ID_TO_OS(vf_id));
+			return err;
+		}
+	}
+
+	if (hinic_vf_info_vlanprio(hwdev, vf_id)) {
+		err = hinic_set_vf_vlan(hwdev, true, vf_info->pf_vlan,
+					vf_info->pf_qos, vf_id);
+		if (err) {
+			dev_err(&hwdev->hwif->pdev->dev, "Failed to add VF %d VLAN_QOS\n",
+				HW_VF_ID_TO_OS(vf_id));
+			return err;
+		}
+	}
+
+	if (vf_info->max_rate) {
+		err = hinic_set_vf_tx_rate(hwdev, vf_id, vf_info->max_rate,
+					   vf_info->min_rate);
+		if (err) {
+			dev_err(&hwdev->hwif->pdev->dev, "Failed to set VF %d max rate: %d, min rate: %d\n",
+				HW_VF_ID_TO_OS(vf_id), vf_info->max_rate,
+				vf_info->min_rate);
+			return err;
+		}
+	}
+
+	return 0;
+}
+
+static int hinic_register_vf_msg_handler(void *hwdev, u16 vf_id,
+					 void *buf_in, u16 in_size,
+					 void *buf_out, u16 *out_size)
+{
+	struct hinic_register_vf *register_info = buf_out;
+	struct hinic_hwdev *hw_dev = hwdev;
+	struct hinic_func_to_io *nic_io;
+	int err;
+
+	nic_io = &hw_dev->func_to_io;
+	if (vf_id > nic_io->max_vfs) {
+		dev_err(&hw_dev->hwif->pdev->dev, "Register VF id %d exceed limit[0-%d]\n",
+			HW_VF_ID_TO_OS(vf_id), HW_VF_ID_TO_OS(nic_io->max_vfs));
+		register_info->status = EFAULT;
+		return -EFAULT;
+	}
+
+	*out_size = sizeof(*register_info);
+	err = hinic_init_vf_config(hw_dev, vf_id);
+	if (err) {
+		register_info->status = EFAULT;
+		return err;
+	}
+
+	nic_io->vf_infos[HW_VF_ID_TO_OS(vf_id)].registered = true;
+
+	return 0;
+}
+
+static int hinic_unregister_vf_msg_handler(void *hwdev, u16 vf_id,
+					   void *buf_in, u16 in_size,
+					   void *buf_out, u16 *out_size)
+{
+	struct hinic_hwdev *hw_dev = hwdev;
+	struct hinic_func_to_io *nic_io;
+
+	nic_io = &hw_dev->func_to_io;
+	*out_size = 0;
+	if (vf_id > nic_io->max_vfs)
+		return 0;
+
+	nic_io->vf_infos[HW_VF_ID_TO_OS(vf_id)].registered = false;
+
+	return 0;
+}
+
+static int hinic_change_vf_mtu_msg_handler(void *hwdev, u16 vf_id,
+					   void *buf_in, u16 in_size,
+					   void *buf_out, u16 *out_size)
+{
+	struct hinic_hwdev *hw_dev = hwdev;
+	int err;
+
+	err = hinic_port_msg_cmd(hwdev, HINIC_PORT_CMD_CHANGE_MTU, buf_in,
+				 in_size, buf_out, out_size);
+	if (err) {
+		dev_err(&hw_dev->hwif->pdev->dev, "Failed to set VF %u mtu\n",
+			vf_id);
+		return err;
+	}
+
+	return 0;
+}
+
+static int hinic_get_vf_mac_msg_handler(void *hwdev, u16 vf_id,
+					void *buf_in, u16 in_size,
+					void *buf_out, u16 *out_size)
+{
+	struct hinic_port_mac_cmd *mac_info = buf_out;
+	struct hinic_hwdev *dev = hwdev;
+	struct hinic_func_to_io *nic_io;
+	struct vf_data_storage *vf_info;
+
+	nic_io = &dev->func_to_io;
+	vf_info = nic_io->vf_infos + HW_VF_ID_TO_OS(vf_id);
+
+	memcpy(mac_info->mac, vf_info->vf_mac_addr, ETH_ALEN);
+	mac_info->status = 0;
+	*out_size = sizeof(*mac_info);
+
+	return 0;
+}
+
+static int hinic_set_vf_mac_msg_handler(void *hwdev, u16 vf_id,
+					void *buf_in, u16 in_size,
+					void *buf_out, u16 *out_size)
+{
+	struct hinic_port_mac_cmd *mac_out = buf_out;
+	struct hinic_port_mac_cmd *mac_in = buf_in;
+	struct hinic_hwdev *hw_dev = hwdev;
+	struct hinic_func_to_io *nic_io;
+	struct vf_data_storage *vf_info;
+	int err;
+
+	nic_io =  &hw_dev->func_to_io;
+	vf_info = nic_io->vf_infos + HW_VF_ID_TO_OS(vf_id);
+	if (vf_info->pf_set_mac && !(vf_info->trust) &&
+	    is_valid_ether_addr(mac_in->mac)) {
+		dev_warn(&hw_dev->hwif->pdev->dev, "PF has already set VF %d MAC address\n",
+			 HW_VF_ID_TO_OS(vf_id));
+		mac_out->status = HINIC_PF_SET_VF_ALREADY;
+		*out_size = sizeof(*mac_out);
+		return 0;
+	}
+
+	err = hinic_port_msg_cmd(hw_dev, HINIC_PORT_CMD_SET_MAC, buf_in,
+				 in_size, buf_out, out_size);
+	if ((err &&  err != HINIC_MBOX_PF_BUSY_ACTIVE_FW) || !(*out_size)) {
+		dev_err(&hw_dev->hwif->pdev->dev,
+			"Failed to set VF %d MAC address, err: %d, status: 0x%x, out size: 0x%x\n",
+			HW_VF_ID_TO_OS(vf_id), err, mac_out->status, *out_size);
+		return -EFAULT;
+	}
+
+	return err;
+}
+
+static int hinic_del_vf_mac_msg_handler(void *hwdev, u16 vf_id,
+					void *buf_in, u16 in_size,
+					void *buf_out, u16 *out_size)
+{
+	struct hinic_port_mac_cmd *mac_out = buf_out;
+	struct hinic_port_mac_cmd *mac_in = buf_in;
+	struct hinic_hwdev *hw_dev = hwdev;
+	struct hinic_func_to_io *nic_io;
+	struct vf_data_storage *vf_info;
+	int err;
+
+	nic_io = &hw_dev->func_to_io;
+	vf_info = nic_io->vf_infos + HW_VF_ID_TO_OS(vf_id);
+	if (vf_info->pf_set_mac && is_valid_ether_addr(mac_in->mac) &&
+	    !memcmp(vf_info->vf_mac_addr, mac_in->mac, ETH_ALEN)) {
+		dev_warn(&hw_dev->hwif->pdev->dev, "PF has already set VF mac.\n");
+		mac_out->status = HINIC_PF_SET_VF_ALREADY;
+		*out_size = sizeof(*mac_out);
+		return 0;
+	}
+
+	err = hinic_port_msg_cmd(hw_dev, HINIC_PORT_CMD_DEL_MAC, buf_in,
+				 in_size, buf_out, out_size);
+	if ((err && err != HINIC_MBOX_PF_BUSY_ACTIVE_FW) || !(*out_size)) {
+		dev_err(&hw_dev->hwif->pdev->dev, "Failed to delete VF %d MAC, err: %d, status: 0x%x, out size: 0x%x\n",
+			HW_VF_ID_TO_OS(vf_id), err, mac_out->status, *out_size);
+		return -EFAULT;
+	}
+
+	return err;
+}
+
+static int hinic_get_vf_link_status_msg_handler(void *hwdev, u16 vf_id,
+						void *buf_in, u16 in_size,
+						void *buf_out, u16 *out_size)
+{
+	struct hinic_port_link_cmd *get_link = buf_out;
+	struct hinic_hwdev *hw_dev = hwdev;
+	struct vf_data_storage *vf_infos;
+	struct hinic_func_to_io *nic_io;
+	bool link_forced, link_up;
+
+	nic_io = &hw_dev->func_to_io;
+	vf_infos = nic_io->vf_infos;
+	link_forced = vf_infos[HW_VF_ID_TO_OS(vf_id)].link_forced;
+	link_up = vf_infos[HW_VF_ID_TO_OS(vf_id)].link_up;
+
+	if (link_forced)
+		get_link->state = link_up ?
+			HINIC_LINK_STATE_UP : HINIC_LINK_STATE_DOWN;
+	else
+		get_link->state = nic_io->link_status;
+
+	get_link->status = 0;
+	*out_size = sizeof(*get_link);
+
+	return 0;
+}
+
+static bool check_func_table(struct hinic_hwdev *hwdev, u16 func_idx,
+			     void *buf_in, u16 in_size)
+{
+	struct hinic_cmd_fw_ctxt *function_table = buf_in;
+
+	if (!hinic_mbox_check_func_id_8B(hwdev, func_idx, buf_in, in_size) ||
+	    !function_table->rx_buf_sz)
+		return false;
+
+	return true;
+}
+
+static struct vf_cmd_msg_handle nic_vf_cmd_msg_handler[] = {
+	{HINIC_PORT_CMD_VF_REGISTER, hinic_register_vf_msg_handler},
+	{HINIC_PORT_CMD_VF_UNREGISTER, hinic_unregister_vf_msg_handler},
+	{HINIC_PORT_CMD_CHANGE_MTU, hinic_change_vf_mtu_msg_handler},
+	{HINIC_PORT_CMD_GET_MAC, hinic_get_vf_mac_msg_handler},
+	{HINIC_PORT_CMD_SET_MAC, hinic_set_vf_mac_msg_handler},
+	{HINIC_PORT_CMD_DEL_MAC, hinic_del_vf_mac_msg_handler},
+	{HINIC_PORT_CMD_GET_LINK_STATE, hinic_get_vf_link_status_msg_handler},
+};
+
+static struct vf_cmd_check_handle nic_cmd_support_vf[] = {
+	{HINIC_PORT_CMD_VF_REGISTER, NULL},
+	{HINIC_PORT_CMD_VF_UNREGISTER, NULL},
+	{HINIC_PORT_CMD_CHANGE_MTU, hinic_mbox_check_func_id_8B},
+	{HINIC_PORT_CMD_ADD_VLAN, hinic_mbox_check_func_id_8B},
+	{HINIC_PORT_CMD_DEL_VLAN, hinic_mbox_check_func_id_8B},
+	{HINIC_PORT_CMD_SET_MAC, hinic_mbox_check_func_id_8B},
+	{HINIC_PORT_CMD_GET_MAC, hinic_mbox_check_func_id_8B},
+	{HINIC_PORT_CMD_DEL_MAC, hinic_mbox_check_func_id_8B},
+	{HINIC_PORT_CMD_SET_RX_MODE, hinic_mbox_check_func_id_8B},
+	{HINIC_PORT_CMD_GET_PAUSE_INFO, hinic_mbox_check_func_id_8B},
+	{HINIC_PORT_CMD_GET_LINK_STATE, hinic_mbox_check_func_id_8B},
+	{HINIC_PORT_CMD_SET_LRO, hinic_mbox_check_func_id_8B},
+	{HINIC_PORT_CMD_SET_RX_CSUM, hinic_mbox_check_func_id_8B},
+	{HINIC_PORT_CMD_SET_RX_VLAN_OFFLOAD, hinic_mbox_check_func_id_8B},
+	{HINIC_PORT_CMD_GET_VPORT_STAT, hinic_mbox_check_func_id_8B},
+	{HINIC_PORT_CMD_CLEAN_VPORT_STAT, hinic_mbox_check_func_id_8B},
+	{HINIC_PORT_CMD_GET_RSS_TEMPLATE_INDIR_TBL,
+	 hinic_mbox_check_func_id_8B},
+	{HINIC_PORT_CMD_SET_RSS_TEMPLATE_TBL, hinic_mbox_check_func_id_8B},
+	{HINIC_PORT_CMD_GET_RSS_TEMPLATE_TBL, hinic_mbox_check_func_id_8B},
+	{HINIC_PORT_CMD_SET_RSS_HASH_ENGINE, hinic_mbox_check_func_id_8B},
+	{HINIC_PORT_CMD_GET_RSS_HASH_ENGINE, hinic_mbox_check_func_id_8B},
+	{HINIC_PORT_CMD_GET_RSS_CTX_TBL, hinic_mbox_check_func_id_8B},
+	{HINIC_PORT_CMD_SET_RSS_CTX_TBL, hinic_mbox_check_func_id_8B},
+	{HINIC_PORT_CMD_RSS_TEMP_MGR, hinic_mbox_check_func_id_8B},
+	{HINIC_PORT_CMD_RSS_CFG, hinic_mbox_check_func_id_8B},
+	{HINIC_PORT_CMD_FWCTXT_INIT, check_func_table},
+	{HINIC_PORT_CMD_GET_MGMT_VERSION, NULL},
+	{HINIC_PORT_CMD_SET_FUNC_STATE, hinic_mbox_check_func_id_8B},
+	{HINIC_PORT_CMD_GET_GLOBAL_QPN, hinic_mbox_check_func_id_8B},
+	{HINIC_PORT_CMD_SET_TSO, hinic_mbox_check_func_id_8B},
+	{HINIC_PORT_CMD_SET_RQ_IQ_MAP, hinic_mbox_check_func_id_8B},
+	{HINIC_PORT_CMD_LINK_STATUS_REPORT, hinic_mbox_check_func_id_8B},
+	{HINIC_PORT_CMD_UPDATE_MAC, hinic_mbox_check_func_id_8B},
+	{HINIC_PORT_CMD_GET_CAP, hinic_mbox_check_func_id_8B},
+	{HINIC_PORT_CMD_GET_LINK_MODE, hinic_mbox_check_func_id_8B},
+	{HINIC_PORT_CMD_GET_VF_COS, NULL},
+	{HINIC_PORT_CMD_SET_VHD_CFG, hinic_mbox_check_func_id_8B},
+	{HINIC_PORT_CMD_SET_VLAN_FILTER, hinic_mbox_check_func_id_8B},
+	{HINIC_PORT_CMD_Q_FILTER, hinic_mbox_check_func_id_8B},
+	{HINIC_PORT_CMD_TCAM_FILTER, NULL},
+	{HINIC_PORT_CMD_UP_TC_ADD_FLOW, NULL},
+	{HINIC_PORT_CMD_UP_TC_DEL_FLOW, NULL},
+	{HINIC_PORT_CMD_UP_TC_FLUSH_TCAM, hinic_mbox_check_func_id_8B},
+	{HINIC_PORT_CMD_UP_TC_CTRL_TCAM_BLOCK, hinic_mbox_check_func_id_8B},
+	{HINIC_PORT_CMD_UP_TC_ENABLE, hinic_mbox_check_func_id_8B},
+	{HINIC_PORT_CMD_CABLE_PLUG_EVENT, NULL},
+	{HINIC_PORT_CMD_LINK_ERR_EVENT, NULL},
+	{HINIC_PORT_CMD_SET_PORT_STATE, hinic_mbox_check_func_id_8B},
+	{HINIC_PORT_CMD_SET_ETS, NULL},
+	{HINIC_PORT_CMD_SET_ANTI_ATTACK_RATE, NULL},
+	{HINIC_PORT_CMD_RESET_LINK_CFG, hinic_mbox_check_func_id_8B},
+	{HINIC_PORT_CMD_SET_LINK_FOLLOW, NULL},
+	{HINIC_PORT_CMD_CLEAR_QP_RES, NULL},
+};
+
+#define CHECK_IPSU_15BIT	0X8000
+
+static
+struct hinic_sriov_info *hinic_get_sriov_info_by_pcidev(struct pci_dev *pdev)
+{
+	struct net_device *netdev = pci_get_drvdata(pdev);
+	struct hinic_dev *nic_dev = netdev_priv(netdev);
+
+	return &nic_dev->sriov_info;
+}
+
+static int hinic_check_mac_info(u8 status, u16 vlan_id)
+{
+	if ((status && status != HINIC_MGMT_STATUS_EXIST) ||
+	    (vlan_id & CHECK_IPSU_15BIT &&
+	     status == HINIC_MGMT_STATUS_EXIST))
+		return -EINVAL;
+
+	return 0;
+}
+
+#define HINIC_VLAN_ID_MASK	0x7FFF
+
+static int hinic_update_mac(struct hinic_hwdev *hwdev, u8 *old_mac,
+			    u8 *new_mac, u16 vlan_id, u16 func_id)
+{
+	struct hinic_port_mac_update mac_info = {0};
+	u16 out_size = sizeof(mac_info);
+	int err;
+
+	if (!hwdev || !old_mac || !new_mac)
+		return -EINVAL;
+
+	if ((vlan_id & HINIC_VLAN_ID_MASK) >= VLAN_N_VID) {
+		dev_err(&hwdev->hwif->pdev->dev, "Invalid VLAN number: %d\n",
+			(vlan_id & HINIC_VLAN_ID_MASK));
+		return -EINVAL;
+	}
+
+	mac_info.func_id = func_id;
+	mac_info.vlan_id = vlan_id;
+	memcpy(mac_info.old_mac, old_mac, ETH_ALEN);
+	memcpy(mac_info.new_mac, new_mac, ETH_ALEN);
+
+	err = hinic_port_msg_cmd(hwdev, HINIC_PORT_CMD_UPDATE_MAC, &mac_info,
+				 sizeof(mac_info), &mac_info, &out_size);
+
+	if (err || !out_size ||
+	    hinic_check_mac_info(mac_info.status, mac_info.vlan_id)) {
+		dev_err(&hwdev->hwif->pdev->dev,
+			"Failed to update MAC, err: %d, status: 0x%x, out size: 0x%x\n",
+			err, mac_info.status, out_size);
+		return -EINVAL;
+	}
+
+	if (mac_info.status == HINIC_MGMT_STATUS_EXIST)
+		dev_warn(&hwdev->hwif->pdev->dev, "MAC is repeated. Ignore update operation\n");
+
+	return 0;
+}
+
+static void hinic_get_vf_config(struct hinic_hwdev *hwdev, u16 vf_id,
+				struct ifla_vf_info *ivi)
+{
+	struct vf_data_storage *vfinfo;
+
+	vfinfo = hwdev->func_to_io.vf_infos + HW_VF_ID_TO_OS(vf_id);
+
+	ivi->vf = HW_VF_ID_TO_OS(vf_id);
+	memcpy(ivi->mac, vfinfo->vf_mac_addr, ETH_ALEN);
+	ivi->vlan = vfinfo->pf_vlan;
+	ivi->qos = vfinfo->pf_qos;
+	ivi->spoofchk = vfinfo->spoofchk;
+	ivi->trusted = vfinfo->trust;
+	ivi->max_tx_rate = vfinfo->max_rate;
+	ivi->min_tx_rate = vfinfo->min_rate;
+
+	if (!vfinfo->link_forced)
+		ivi->linkstate = IFLA_VF_LINK_STATE_AUTO;
+	else if (vfinfo->link_up)
+		ivi->linkstate = IFLA_VF_LINK_STATE_ENABLE;
+	else
+		ivi->linkstate = IFLA_VF_LINK_STATE_DISABLE;
+}
+
+int hinic_ndo_get_vf_config(struct net_device *netdev,
+			    int vf, struct ifla_vf_info *ivi)
+{
+	struct hinic_dev *nic_dev = netdev_priv(netdev);
+	struct hinic_sriov_info *sriov_info;
+
+	sriov_info = &nic_dev->sriov_info;
+	if (vf >= sriov_info->num_vfs)
+		return -EINVAL;
+
+	hinic_get_vf_config(sriov_info->hwdev, OS_VF_ID_TO_HW(vf), ivi);
+
+	return 0;
+}
+
+static int hinic_set_vf_mac(struct hinic_hwdev *hwdev, int vf,
+			    unsigned char *mac_addr)
+{
+	struct hinic_func_to_io *nic_io = &hwdev->func_to_io;
+	struct vf_data_storage *vf_info;
+	u16 func_id;
+	int err;
+
+	vf_info = nic_io->vf_infos + HW_VF_ID_TO_OS(vf);
+
+	/* duplicate request, so just return success */
+	if (vf_info->pf_set_mac &&
+	    !memcmp(vf_info->vf_mac_addr, mac_addr, ETH_ALEN))
+		return 0;
+
+	vf_info->pf_set_mac = true;
+
+	func_id = hinic_glb_pf_vf_offset(hwdev->hwif) + vf;
+	err = hinic_update_mac(hwdev, vf_info->vf_mac_addr,
+			       mac_addr, 0, func_id);
+	if (err) {
+		vf_info->pf_set_mac = false;
+		return err;
+	}
+
+	memcpy(vf_info->vf_mac_addr, mac_addr, ETH_ALEN);
+
+	return 0;
+}
+
+int hinic_ndo_set_vf_mac(struct net_device *netdev, int vf, u8 *mac)
+{
+	struct hinic_dev *nic_dev = netdev_priv(netdev);
+	struct hinic_sriov_info *sriov_info;
+	int err;
+
+	sriov_info = &nic_dev->sriov_info;
+	if (!is_valid_ether_addr(mac) || vf >= sriov_info->num_vfs)
+		return -EINVAL;
+
+	err = hinic_set_vf_mac(sriov_info->hwdev, OS_VF_ID_TO_HW(vf), mac);
+	if (err)
+		return err;
+
+	netif_info(nic_dev, drv, netdev, "Setting MAC %pM on VF %d\n", mac, vf);
+	netif_info(nic_dev, drv, netdev, "Reload the VF driver to make this change effective.");
+
+	return 0;
+}
+
+static int hinic_add_vf_vlan(struct hinic_hwdev *hwdev, int vf_id,
+			     u16 vlan, u8 qos)
+{
+	struct hinic_func_to_io *nic_io = &hwdev->func_to_io;
+	int err;
+
+	err = hinic_set_vf_vlan(hwdev, true, vlan, qos, vf_id);
+	if (err)
+		return err;
+
+	nic_io->vf_infos[HW_VF_ID_TO_OS(vf_id)].pf_vlan = vlan;
+	nic_io->vf_infos[HW_VF_ID_TO_OS(vf_id)].pf_qos = qos;
+
+	dev_info(&hwdev->hwif->pdev->dev, "Setting VLAN %d, QOS 0x%x on VF %d\n",
+		 vlan, qos, HW_VF_ID_TO_OS(vf_id));
+	return 0;
+}
+
+static int hinic_kill_vf_vlan(struct hinic_hwdev *hwdev, int vf_id)
+{
+	struct hinic_func_to_io *nic_io = &hwdev->func_to_io;
+	int err;
+
+	err = hinic_set_vf_vlan(hwdev, false,
+				nic_io->vf_infos[HW_VF_ID_TO_OS(vf_id)].pf_vlan,
+				nic_io->vf_infos[HW_VF_ID_TO_OS(vf_id)].pf_qos,
+				vf_id);
+	if (err)
+		return err;
+
+	dev_info(&hwdev->hwif->pdev->dev, "Remove VLAN %d on VF %d\n",
+		 nic_io->vf_infos[HW_VF_ID_TO_OS(vf_id)].pf_vlan,
+		 HW_VF_ID_TO_OS(vf_id));
+
+	nic_io->vf_infos[HW_VF_ID_TO_OS(vf_id)].pf_vlan = 0;
+	nic_io->vf_infos[HW_VF_ID_TO_OS(vf_id)].pf_qos = 0;
+
+	return 0;
+}
+
+static int hinic_update_mac_vlan(struct hinic_dev *nic_dev, u16 old_vlan,
+				 u16 new_vlan, int vf_id)
+{
+	struct vf_data_storage *vf_info;
+	u16 vlan_id;
+	int err;
+
+	if (!nic_dev || old_vlan >= VLAN_N_VID || new_vlan >= VLAN_N_VID)
+		return -EINVAL;
+
+	vf_info = nic_dev->hwdev->func_to_io.vf_infos + HW_VF_ID_TO_OS(vf_id);
+	if (!vf_info->pf_set_mac)
+		return 0;
+
+	vlan_id = old_vlan;
+	if (vlan_id)
+		vlan_id |= HINIC_ADD_VLAN_IN_MAC;
+
+	err = hinic_port_del_mac(nic_dev, vf_info->vf_mac_addr, vlan_id);
+	if (err) {
+		dev_err(&nic_dev->hwdev->hwif->pdev->dev, "Failed to delete VF %d MAC %pM vlan %d\n",
+			HW_VF_ID_TO_OS(vf_id), vf_info->vf_mac_addr, old_vlan);
+		return err;
+	}
+
+	vlan_id = new_vlan;
+	if (vlan_id)
+		vlan_id |= HINIC_ADD_VLAN_IN_MAC;
+
+	err = hinic_port_add_mac(nic_dev, vf_info->vf_mac_addr, vlan_id);
+	if (err) {
+		dev_err(&nic_dev->hwdev->hwif->pdev->dev, "Failed to add VF %d MAC %pM vlan %d\n",
+			HW_VF_ID_TO_OS(vf_id), vf_info->vf_mac_addr, new_vlan);
+		goto out;
+	}
+
+	return 0;
+
+out:
+	vlan_id = old_vlan;
+	if (vlan_id)
+		vlan_id |= HINIC_ADD_VLAN_IN_MAC;
+	hinic_port_add_mac(nic_dev, vf_info->vf_mac_addr, vlan_id);
+
+	return err;
+}
+
+static int set_hw_vf_vlan(struct hinic_dev *nic_dev,
+			  u16 cur_vlanprio, int vf, u16 vlan, u8 qos)
+{
+	u16 old_vlan = cur_vlanprio & VLAN_VID_MASK;
+	int err = 0;
+
+	if (vlan || qos) {
+		if (cur_vlanprio) {
+			err = hinic_kill_vf_vlan(nic_dev->hwdev,
+						 OS_VF_ID_TO_HW(vf));
+			if (err) {
+				dev_err(&nic_dev->sriov_info.pdev->dev, "Failed to delete vf %d old vlan %d\n",
+					vf, old_vlan);
+				goto out;
+			}
+		}
+		err = hinic_add_vf_vlan(nic_dev->hwdev,
+					OS_VF_ID_TO_HW(vf), vlan, qos);
+		if (err) {
+			dev_err(&nic_dev->sriov_info.pdev->dev, "Failed to add vf %d new vlan %d\n",
+				vf, vlan);
+			goto out;
+		}
+	} else {
+		err = hinic_kill_vf_vlan(nic_dev->hwdev, OS_VF_ID_TO_HW(vf));
+		if (err) {
+			dev_err(&nic_dev->sriov_info.pdev->dev, "Failed to delete vf %d vlan %d\n",
+				vf, old_vlan);
+			goto out;
+		}
+	}
+
+	err = hinic_update_mac_vlan(nic_dev, old_vlan, vlan,
+				    OS_VF_ID_TO_HW(vf));
+
+out:
+	return err;
+}
+
+int hinic_ndo_set_vf_vlan(struct net_device *netdev, int vf, u16 vlan, u8 qos,
+			  __be16 vlan_proto)
+{
+	struct hinic_dev *nic_dev = netdev_priv(netdev);
+	struct hinic_sriov_info *sriov_info;
+	u16 vlanprio, cur_vlanprio;
+
+	sriov_info = &nic_dev->sriov_info;
+	if (vf >= sriov_info->num_vfs || vlan >= VLAN_N_VID || qos > HINIC_MAX_QOS)
+		return -EINVAL;
+	if (vlan_proto != htons(ETH_P_8021Q))
+		return -EPROTONOSUPPORT;
+	vlanprio = vlan | qos << HINIC_VLAN_PRIORITY_SHIFT;
+	cur_vlanprio = hinic_vf_info_vlanprio(nic_dev->hwdev,
+					      OS_VF_ID_TO_HW(vf));
+	/* duplicate request, so just return success */
+	if (vlanprio == cur_vlanprio)
+		return 0;
+
+	return set_hw_vf_vlan(nic_dev, cur_vlanprio, vf, vlan, qos);
+}
+
+static int hinic_set_vf_trust(struct hinic_hwdev *hwdev, u16 vf_id,
+			      bool trust)
+{
+	struct vf_data_storage *vf_infos;
+	struct hinic_func_to_io *nic_io;
+
+	if (!hwdev)
+		return -EINVAL;
+
+	nic_io = &hwdev->func_to_io;
+	vf_infos = nic_io->vf_infos;
+	vf_infos[vf_id].trust = trust;
+
+	return 0;
+}
+
+int hinic_ndo_set_vf_trust(struct net_device *netdev, int vf, bool setting)
+{
+	struct hinic_dev *adapter = netdev_priv(netdev);
+	struct hinic_sriov_info *sriov_info;
+	struct hinic_func_to_io *nic_io;
+	bool cur_trust;
+	int err;
+
+	sriov_info = &adapter->sriov_info;
+	nic_io = &adapter->hwdev->func_to_io;
+
+	if (vf >= sriov_info->num_vfs)
+		return -EINVAL;
+
+	cur_trust = nic_io->vf_infos[vf].trust;
+	/* same request, so just return success */
+	if (setting == cur_trust)
+		return 0;
+
+	err = hinic_set_vf_trust(adapter->hwdev, vf, setting);
+	if (!err)
+		dev_info(&sriov_info->pdev->dev, "Set VF %d trusted %s succeed\n",
+			 vf, setting ? "on" : "off");
+	else
+		dev_err(&sriov_info->pdev->dev, "Failed set VF %d trusted %s\n",
+			vf, setting ? "on" : "off");
+
+	return err;
+}
+
+int hinic_ndo_set_vf_bw(struct net_device *netdev,
+			int vf, int min_tx_rate, int max_tx_rate)
+{
+	static const u32 speeds[] = {
+		SPEED_10, SPEED_100, SPEED_1000, SPEED_10000,
+		SPEED_25000, SPEED_40000, SPEED_100000
+	};
+	struct hinic_dev *nic_dev = netdev_priv(netdev);
+	struct hinic_port_cap port_cap = { 0 };
+	enum hinic_port_link_state link_state;
+	int err;
+
+	if (vf >= nic_dev->sriov_info.num_vfs) {
+		netif_err(nic_dev, drv, netdev, "VF number must be less than %d\n",
+			  nic_dev->sriov_info.num_vfs);
+		return -EINVAL;
+	}
+
+	err = hinic_port_link_state(nic_dev, &link_state);
+	if (err) {
+		netif_err(nic_dev, drv, netdev,
+			  "Get link status failed when setting vf tx rate\n");
+		return -EIO;
+	}
+
+	if (link_state == HINIC_LINK_STATE_DOWN) {
+		netif_err(nic_dev, drv, netdev,
+			  "Link status must be up when setting vf tx rate\n");
+		return -EPERM;
+	}
+
+	err = hinic_port_get_cap(nic_dev, &port_cap);
+	if (err || port_cap.speed > LINK_SPEED_100GB)
+		return -EIO;
+
+	/* rate limit cannot be less than 0 and greater than link speed */
+	if (max_tx_rate < 0 || max_tx_rate > speeds[port_cap.speed]) {
+		netif_err(nic_dev, drv, netdev, "Max tx rate must be in [0 - %d]\n",
+			  speeds[port_cap.speed]);
+		return -EINVAL;
+	}
+
+	err = hinic_set_vf_tx_rate(nic_dev->hwdev, OS_VF_ID_TO_HW(vf),
+				   max_tx_rate, min_tx_rate);
+	if (err) {
+		netif_err(nic_dev, drv, netdev,
+			  "Unable to set VF %d max rate %d min rate %d%s\n",
+			  vf, max_tx_rate, min_tx_rate,
+			  err == HINIC_TX_RATE_TABLE_FULL ?
+			  ", tx rate profile is full" : "");
+		return -EIO;
+	}
+
+	netif_info(nic_dev, drv, netdev,
+		   "Set VF %d max tx rate %d min tx rate %d successfully\n",
+		   vf, max_tx_rate, min_tx_rate);
+
+	return 0;
+}
+
+static int hinic_set_vf_spoofchk(struct hinic_hwdev *hwdev, u16 vf_id,
+				 bool spoofchk)
+{
+	struct hinic_spoofchk_set spoofchk_cfg = {0};
+	struct vf_data_storage *vf_infos = NULL;
+	u16 out_size = sizeof(spoofchk_cfg);
+	int err;
+
+	if (!hwdev)
+		return -EINVAL;
+
+	vf_infos = hwdev->func_to_io.vf_infos;
+
+	spoofchk_cfg.func_id = hinic_glb_pf_vf_offset(hwdev->hwif) + vf_id;
+	spoofchk_cfg.state = spoofchk ? 1 : 0;
+	err = hinic_port_msg_cmd(hwdev, HINIC_PORT_CMD_ENABLE_SPOOFCHK,
+				 &spoofchk_cfg, sizeof(spoofchk_cfg),
+				 &spoofchk_cfg, &out_size);
+	if (spoofchk_cfg.status == HINIC_MGMT_CMD_UNSUPPORTED) {
+		err = HINIC_MGMT_CMD_UNSUPPORTED;
+	} else if (err || !out_size || spoofchk_cfg.status) {
+		dev_err(&hwdev->hwif->pdev->dev, "Failed to set VF(%d) spoofchk, err: %d, status: 0x%x, out size: 0x%x\n",
+			HW_VF_ID_TO_OS(vf_id), err, spoofchk_cfg.status,
+			out_size);
+		err = -EIO;
+	}
+
+	vf_infos[HW_VF_ID_TO_OS(vf_id)].spoofchk = spoofchk;
+
+	return err;
+}
+
+int hinic_ndo_set_vf_spoofchk(struct net_device *netdev, int vf, bool setting)
+{
+	struct hinic_dev *nic_dev = netdev_priv(netdev);
+	struct hinic_sriov_info *sriov_info;
+	bool cur_spoofchk;
+	int err;
+
+	sriov_info = &nic_dev->sriov_info;
+	if (vf >= sriov_info->num_vfs)
+		return -EINVAL;
+
+	cur_spoofchk = nic_dev->hwdev->func_to_io.vf_infos[vf].spoofchk;
+
+	/* same request, so just return success */
+	if (setting == cur_spoofchk)
+		return 0;
+
+	err = hinic_set_vf_spoofchk(sriov_info->hwdev,
+				    OS_VF_ID_TO_HW(vf), setting);
+	if (!err) {
+		netif_info(nic_dev, drv, netdev, "Set VF %d spoofchk %s successfully\n",
+			   vf, setting ? "on" : "off");
+	} else if (err == HINIC_MGMT_CMD_UNSUPPORTED) {
+		netif_err(nic_dev, drv, netdev,
+			  "Current firmware doesn't support to set vf spoofchk, need to upgrade latest firmware version\n");
+		err = -EOPNOTSUPP;
+	}
+
+	return err;
+}
+
+static int hinic_set_vf_link_state(struct hinic_hwdev *hwdev, u16 vf_id,
+				   int link)
+{
+	struct hinic_func_to_io *nic_io = &hwdev->func_to_io;
+	struct vf_data_storage *vf_infos = nic_io->vf_infos;
+	u8 link_status = 0;
+
+	switch (link) {
+	case HINIC_IFLA_VF_LINK_STATE_AUTO:
+		vf_infos[HW_VF_ID_TO_OS(vf_id)].link_forced = false;
+		vf_infos[HW_VF_ID_TO_OS(vf_id)].link_up = nic_io->link_status ?
+			true : false;
+		link_status = nic_io->link_status;
+		break;
+	case HINIC_IFLA_VF_LINK_STATE_ENABLE:
+		vf_infos[HW_VF_ID_TO_OS(vf_id)].link_forced = true;
+		vf_infos[HW_VF_ID_TO_OS(vf_id)].link_up = true;
+		link_status = HINIC_LINK_UP;
+		break;
+	case HINIC_IFLA_VF_LINK_STATE_DISABLE:
+		vf_infos[HW_VF_ID_TO_OS(vf_id)].link_forced = true;
+		vf_infos[HW_VF_ID_TO_OS(vf_id)].link_up = false;
+		link_status = HINIC_LINK_DOWN;
+		break;
+	default:
+		return -EINVAL;
+	}
+
+	/* Notify the VF of its new link state */
+	hinic_notify_vf_link_status(hwdev, vf_id, link_status);
+
+	return 0;
+}
+
+int hinic_ndo_set_vf_link_state(struct net_device *netdev, int vf_id, int link)
+{
+	struct hinic_dev *nic_dev = netdev_priv(netdev);
+	struct hinic_sriov_info *sriov_info;
+
+	sriov_info = &nic_dev->sriov_info;
+
+	if (vf_id >= sriov_info->num_vfs) {
+		netif_err(nic_dev, drv, netdev,
+			  "Invalid VF Identifier %d\n", vf_id);
+		return -EINVAL;
+	}
+
+	return hinic_set_vf_link_state(sriov_info->hwdev,
+				      OS_VF_ID_TO_HW(vf_id), link);
+}
+
+/* pf receive message from vf */
+static int nic_pf_mbox_handler(void *hwdev, u16 vf_id, u8 cmd, void *buf_in,
+			       u16 in_size, void *buf_out, u16 *out_size)
+{
+	u8 size = ARRAY_SIZE(nic_cmd_support_vf);
+	struct vf_cmd_msg_handle *vf_msg_handle;
+	struct hinic_hwdev *dev = hwdev;
+	struct hinic_func_to_io *nic_io;
+	struct hinic_pfhwdev *pfhwdev;
+	int err = 0;
+	u32 i;
+
+	if (!hwdev)
+		return -EINVAL;
+
+	if (!hinic_mbox_check_cmd_valid(hwdev, nic_cmd_support_vf, vf_id, cmd,
+					buf_in, in_size, size)) {
+		dev_err(&dev->hwif->pdev->dev,
+			"PF Receive VF nic cmd: 0x%x, mbox len: 0x%x is invalid\n",
+			cmd, in_size);
+		return HINIC_MBOX_VF_CMD_ERROR;
+	}
+
+	pfhwdev = container_of(dev, struct hinic_pfhwdev, hwdev);
+	nic_io = &dev->func_to_io;
+	for (i = 0; i < ARRAY_SIZE(nic_vf_cmd_msg_handler); i++) {
+		vf_msg_handle = &nic_vf_cmd_msg_handler[i];
+		if (cmd == vf_msg_handle->cmd &&
+		    vf_msg_handle->cmd_msg_handler) {
+			err = vf_msg_handle->cmd_msg_handler(hwdev, vf_id,
+							     buf_in, in_size,
+							     buf_out,
+							     out_size);
+			break;
+		}
+	}
+	if (i == ARRAY_SIZE(nic_vf_cmd_msg_handler))
+		err = hinic_msg_to_mgmt(&pfhwdev->pf_to_mgmt, HINIC_MOD_L2NIC,
+					cmd, buf_in, in_size, buf_out,
+					out_size, HINIC_MGMT_MSG_SYNC);
+
+	if (err &&  err != HINIC_MBOX_PF_BUSY_ACTIVE_FW)
+		dev_err(&nic_io->hwif->pdev->dev, "PF receive VF L2NIC cmd: %d process error, err:%d\n",
+			cmd, err);
+	return err;
+}
+
+static int cfg_mbx_pf_proc_vf_msg(void *hwdev, u16 vf_id, u8 cmd, void *buf_in,
+				  u16 in_size, void *buf_out, u16 *out_size)
+{
+	struct hinic_dev_cap *dev_cap = buf_out;
+	struct hinic_hwdev *dev = hwdev;
+	struct hinic_cap *cap;
+
+	cap = &dev->nic_cap;
+	memset(dev_cap, 0, sizeof(*dev_cap));
+
+	dev_cap->max_vf = cap->max_vf;
+	dev_cap->max_sqs = cap->max_vf_qps;
+	dev_cap->max_rqs = cap->max_vf_qps;
+	dev_cap->port_id = dev->port_id;
+
+	*out_size = sizeof(*dev_cap);
+
+	return 0;
+}
+
+static int hinic_init_vf_infos(struct hinic_func_to_io *nic_io, u16 vf_id)
+{
+	struct vf_data_storage *vf_infos = nic_io->vf_infos;
+
+	if (set_vf_link_state > HINIC_IFLA_VF_LINK_STATE_DISABLE) {
+		dev_warn(&nic_io->hwif->pdev->dev, "Module Parameter set_vf_link_state value %d is out of range, resetting to %d\n",
+			 set_vf_link_state, HINIC_IFLA_VF_LINK_STATE_AUTO);
+		set_vf_link_state = HINIC_IFLA_VF_LINK_STATE_AUTO;
+	}
+
+	switch (set_vf_link_state) {
+	case HINIC_IFLA_VF_LINK_STATE_AUTO:
+		vf_infos[vf_id].link_forced = false;
+		break;
+	case HINIC_IFLA_VF_LINK_STATE_ENABLE:
+		vf_infos[vf_id].link_forced = true;
+		vf_infos[vf_id].link_up = true;
+		break;
+	case HINIC_IFLA_VF_LINK_STATE_DISABLE:
+		vf_infos[vf_id].link_forced = true;
+		vf_infos[vf_id].link_up = false;
+		break;
+	default:
+		dev_err(&nic_io->hwif->pdev->dev, "Invalid input parameter set_vf_link_state: %d\n",
+			set_vf_link_state);
+		return -EINVAL;
+	}
+
+	return 0;
+}
+
+static void hinic_clear_vf_infos(struct hinic_dev *nic_dev, u16 vf_id)
+{
+	struct vf_data_storage *vf_infos;
+
+	vf_infos = nic_dev->hwdev->func_to_io.vf_infos + HW_VF_ID_TO_OS(vf_id);
+	if (vf_infos->pf_set_mac)
+		hinic_port_del_mac(nic_dev, vf_infos->vf_mac_addr, 0);
+
+	if (hinic_vf_info_vlanprio(nic_dev->hwdev, vf_id))
+		hinic_kill_vf_vlan(nic_dev->hwdev, vf_id);
+
+	if (vf_infos->max_rate)
+		hinic_set_vf_tx_rate(nic_dev->hwdev, vf_id, 0, 0);
+
+	if (vf_infos->spoofchk)
+		hinic_set_vf_spoofchk(nic_dev->hwdev, vf_id, false);
+
+	if (vf_infos->trust)
+		hinic_set_vf_trust(nic_dev->hwdev, vf_id, false);
+
+	memset(vf_infos, 0, sizeof(*vf_infos));
+	/* set vf_infos to default */
+	hinic_init_vf_infos(&nic_dev->hwdev->func_to_io, HW_VF_ID_TO_OS(vf_id));
+}
+
+static void hinic_deinit_vf_hw(struct hinic_sriov_info *sriov_info,
+			       u16 start_vf_id, u16 end_vf_id)
+{
+	struct hinic_dev *nic_dev;
+	u16 func_idx, idx;
+
+	nic_dev = container_of(sriov_info, struct hinic_dev, sriov_info);
+
+	for (idx = start_vf_id; idx <= end_vf_id; idx++) {
+		func_idx = hinic_glb_pf_vf_offset(nic_dev->hwdev->hwif) + idx;
+		hinic_set_wq_page_size(nic_dev->hwdev, func_idx,
+				       HINIC_HW_WQ_PAGE_SIZE);
+		hinic_clear_vf_infos(nic_dev, idx);
+	}
+}
+
+int hinic_vf_func_init(struct hinic_hwdev *hwdev)
+{
+	struct hinic_register_vf register_info = {0};
+	u16 out_size = sizeof(register_info);
+	struct hinic_func_to_io *nic_io;
+	int err = 0;
+	u32 size, i;
+
+	err = hinic_vf_mbox_random_id_init(hwdev);
+	if (err) {
+		dev_err(&hwdev->hwif->pdev->dev, "Failed to init vf mbox random id, err: %d\n",
+			err);
+		return err;
+	}
+
+	nic_io = &hwdev->func_to_io;
+
+	if (HINIC_IS_VF(hwdev->hwif)) {
+		err = hinic_mbox_to_pf(hwdev, HINIC_MOD_L2NIC,
+				       HINIC_PORT_CMD_VF_REGISTER,
+				       &register_info, sizeof(register_info),
+				       &register_info, &out_size, 0);
+		if (err || register_info.status || !out_size) {
+			dev_err(&hwdev->hwif->pdev->dev,
+				"Failed to register VF, err: %d, status: 0x%x, out size: 0x%x\n",
+				err, register_info.status, out_size);
+			return -EIO;
+		}
+	} else {
+		err = hinic_register_pf_mbox_cb(hwdev, HINIC_MOD_CFGM,
+						cfg_mbx_pf_proc_vf_msg);
+		if (err) {
+			dev_err(&hwdev->hwif->pdev->dev,
+				"Register PF mailbox callback failed\n");
+			return err;
+		}
+		nic_io->max_vfs = hwdev->nic_cap.max_vf;
+		size = sizeof(*nic_io->vf_infos) * nic_io->max_vfs;
+		if (size != 0) {
+			nic_io->vf_infos = kzalloc(size, GFP_KERNEL);
+			if (!nic_io->vf_infos) {
+				err = -ENOMEM;
+				goto out_free_nic_io;
+			}
+
+			for (i = 0; i < nic_io->max_vfs; i++) {
+				err = hinic_init_vf_infos(nic_io, i);
+				if (err)
+					goto err_init_vf_infos;
+			}
+
+			err = hinic_register_pf_mbox_cb(hwdev, HINIC_MOD_L2NIC,
+							nic_pf_mbox_handler);
+			if (err)
+				goto err_register_pf_mbox_cb;
+		}
+	}
+
+	return 0;
+
+err_register_pf_mbox_cb:
+err_init_vf_infos:
+	kfree(nic_io->vf_infos);
+out_free_nic_io:
+	return err;
+}
+
+void hinic_vf_func_free(struct hinic_hwdev *hwdev)
+{
+	struct hinic_register_vf unregister = {0};
+	u16 out_size = sizeof(unregister);
+	int err;
+
+	if (HINIC_IS_VF(hwdev->hwif)) {
+		err = hinic_mbox_to_pf(hwdev, HINIC_MOD_L2NIC,
+				       HINIC_PORT_CMD_VF_UNREGISTER,
+				       &unregister, sizeof(unregister),
+				       &unregister, &out_size, 0);
+		if (err || !out_size || unregister.status)
+			dev_err(&hwdev->hwif->pdev->dev, "Failed to unregister VF, err: %d, status: 0x%x, out_size: 0x%x\n",
+				err, unregister.status, out_size);
+	} else {
+		if (hwdev->func_to_io.vf_infos) {
+			hinic_unregister_pf_mbox_cb(hwdev, HINIC_MOD_L2NIC);
+			kfree(hwdev->func_to_io.vf_infos);
+		}
+	}
+}
+
+static int hinic_init_vf_hw(struct hinic_hwdev *hwdev, u16 start_vf_id,
+			    u16 end_vf_id)
+{
+	u16 i, func_idx;
+	int err;
+
+	/* vf use 256K as default wq page size, and can't change it */
+	for (i = start_vf_id; i <= end_vf_id; i++) {
+		func_idx = hinic_glb_pf_vf_offset(hwdev->hwif) + i;
+		err = hinic_set_wq_page_size(hwdev, func_idx,
+					     HINIC_DEFAULT_WQ_PAGE_SIZE);
+		if (err)
+			return err;
+	}
+
+	return 0;
+}
+
+int hinic_pci_sriov_disable(struct pci_dev *pdev)
+{
+	struct hinic_sriov_info *sriov_info;
+	u16 tmp_vfs;
+
+	sriov_info = hinic_get_sriov_info_by_pcidev(pdev);
+	/* if SR-IOV is already disabled then nothing will be done */
+	if (!sriov_info->sriov_enabled)
+		return 0;
+
+	set_bit(HINIC_SRIOV_DISABLE, &sriov_info->state);
+
+	/* If our VFs are assigned we cannot shut down SR-IOV
+	 * without causing issues, so just leave the hardware
+	 * available but disabled
+	 */
+	if (pci_vfs_assigned(sriov_info->pdev)) {
+		clear_bit(HINIC_SRIOV_DISABLE, &sriov_info->state);
+		dev_warn(&pdev->dev, "Unloading driver while VFs are assigned - VFs will not be deallocated\n");
+		return -EPERM;
+	}
+	sriov_info->sriov_enabled = false;
+
+	/* disable iov and allow time for transactions to clear */
+	pci_disable_sriov(sriov_info->pdev);
+
+	tmp_vfs = (u16)sriov_info->num_vfs;
+	sriov_info->num_vfs = 0;
+	hinic_deinit_vf_hw(sriov_info, OS_VF_ID_TO_HW(0),
+			   OS_VF_ID_TO_HW(tmp_vfs - 1));
+
+	clear_bit(HINIC_SRIOV_DISABLE, &sriov_info->state);
+
+	return 0;
+}
+
+static int hinic_pci_sriov_enable(struct pci_dev *pdev, int num_vfs)
+{
+	struct hinic_sriov_info *sriov_info;
+	int err;
+
+	sriov_info = hinic_get_sriov_info_by_pcidev(pdev);
+
+	if (test_and_set_bit(HINIC_SRIOV_ENABLE, &sriov_info->state)) {
+		dev_err(&pdev->dev,
+			"SR-IOV enable in process, please wait, num_vfs %d\n",
+			num_vfs);
+		return -EPERM;
+	}
+
+	err = hinic_init_vf_hw(sriov_info->hwdev, OS_VF_ID_TO_HW(0),
+			       OS_VF_ID_TO_HW((u16)num_vfs - 1));
+	if (err) {
+		dev_err(&sriov_info->pdev->dev,
+			"Failed to init vf in hardware before enable sriov, error %d\n",
+			err);
+		clear_bit(HINIC_SRIOV_ENABLE, &sriov_info->state);
+		return err;
+	}
+
+	err = pci_enable_sriov(sriov_info->pdev, num_vfs);
+	if (err) {
+		dev_err(&pdev->dev,
+			"Failed to enable SR-IOV, error %d\n", err);
+		clear_bit(HINIC_SRIOV_ENABLE, &sriov_info->state);
+		return err;
+	}
+
+	sriov_info->sriov_enabled = true;
+	sriov_info->num_vfs = num_vfs;
+	clear_bit(HINIC_SRIOV_ENABLE, &sriov_info->state);
+
+	return num_vfs;
+}
+
+int hinic_pci_sriov_configure(struct pci_dev *dev, int num_vfs)
+{
+	struct hinic_sriov_info *sriov_info;
+
+	sriov_info = hinic_get_sriov_info_by_pcidev(dev);
+
+	if (test_bit(HINIC_FUNC_REMOVE, &sriov_info->state))
+		return -EBUSY;
+
+	if (!num_vfs)
+		return hinic_pci_sriov_disable(dev);
+	else
+		return hinic_pci_sriov_enable(dev, num_vfs);
+}
diff --git a/drivers/net/ethernet/huawei/hinic/hinic_sriov.h b/drivers/net/ethernet/huawei/hinic/hinic_sriov.h
new file mode 100644
index 000000000000..d4d4e63d31ea
--- /dev/null
+++ b/drivers/net/ethernet/huawei/hinic/hinic_sriov.h
@@ -0,0 +1,107 @@
+/* SPDX-License-Identifier: GPL-2.0-only */
+/* Huawei HiNIC PCI Express Linux driver
+ * Copyright(c) 2017 Huawei Technologies Co., Ltd
+ */
+
+#ifndef HINIC_SRIOV_H
+#define HINIC_SRIOV_H
+
+#include "hinic_hw_dev.h"
+
+#define OS_VF_ID_TO_HW(os_vf_id) ((os_vf_id) + 1)
+#define HW_VF_ID_TO_OS(hw_vf_id) ((hw_vf_id) - 1)
+
+enum hinic_sriov_state {
+	HINIC_SRIOV_DISABLE,
+	HINIC_SRIOV_ENABLE,
+	HINIC_FUNC_REMOVE,
+};
+
+enum {
+	HINIC_IFLA_VF_LINK_STATE_AUTO,	/* link state of the uplink */
+	HINIC_IFLA_VF_LINK_STATE_ENABLE,	/* link always up */
+	HINIC_IFLA_VF_LINK_STATE_DISABLE,	/* link always down */
+};
+
+struct hinic_sriov_info {
+	struct pci_dev *pdev;
+	struct hinic_hwdev *hwdev;
+	bool sriov_enabled;
+	unsigned int num_vfs;
+	unsigned long state;
+};
+
+struct vf_data_storage {
+	u8 vf_mac_addr[ETH_ALEN];
+	bool registered;
+	bool pf_set_mac;
+	u16 pf_vlan;
+	u8 pf_qos;
+	u32 max_rate;
+	u32 min_rate;
+
+	bool link_forced;
+	bool link_up;		/* only valid if VF link is forced */
+	bool spoofchk;
+	bool trust;
+};
+
+struct hinic_register_vf {
+	u8	status;
+	u8	version;
+	u8	rsvd0[6];
+};
+
+struct hinic_port_mac_update {
+	u8	status;
+	u8	version;
+	u8	rsvd0[6];
+
+	u16	func_id;
+	u16	vlan_id;
+	u16	rsvd1;
+	u8	old_mac[ETH_ALEN];
+	u16	rsvd2;
+	u8	new_mac[ETH_ALEN];
+};
+
+struct hinic_vf_vlan_config {
+	u8 status;
+	u8 version;
+	u8 rsvd0[6];
+
+	u16 func_id;
+	u16 vlan_id;
+	u8  qos;
+	u8  rsvd1[7];
+};
+
+int hinic_ndo_set_vf_mac(struct net_device *netdev, int vf, u8 *mac);
+
+int hinic_ndo_set_vf_vlan(struct net_device *netdev, int vf, u16 vlan, u8 qos,
+			  __be16 vlan_proto);
+
+int hinic_ndo_get_vf_config(struct net_device *netdev,
+			    int vf, struct ifla_vf_info *ivi);
+
+int hinic_ndo_set_vf_trust(struct net_device *netdev, int vf, bool setting);
+
+int hinic_ndo_set_vf_bw(struct net_device *netdev,
+			int vf, int min_tx_rate, int max_tx_rate);
+
+int hinic_ndo_set_vf_spoofchk(struct net_device *netdev, int vf, bool setting);
+
+int hinic_ndo_set_vf_link_state(struct net_device *netdev, int vf_id, int link);
+
+void hinic_notify_all_vfs_link_changed(struct hinic_hwdev *hwdev,
+				       u8 link_status);
+
+int hinic_pci_sriov_disable(struct pci_dev *dev);
+
+int hinic_vf_func_init(struct hinic_hwdev *hwdev);
+
+void hinic_vf_func_free(struct hinic_hwdev *hwdev);
+
+int hinic_pci_sriov_configure(struct pci_dev *dev, int num_vfs);
+
+#endif
diff --git a/drivers/net/ethernet/huawei/hinic/hinic_tx.c b/drivers/net/ethernet/huawei/hinic/hinic_tx.c
new file mode 100644
index 000000000000..9b60966736db
--- /dev/null
+++ b/drivers/net/ethernet/huawei/hinic/hinic_tx.c
@@ -0,0 +1,932 @@
+// SPDX-License-Identifier: GPL-2.0-only
+/*
+ * Huawei HiNIC PCI Express Linux driver
+ * Copyright(c) 2017 Huawei Technologies Co., Ltd
+ */
+
+#include <linux/if_vlan.h>
+#include <linux/kernel.h>
+#include <linux/netdevice.h>
+#include <linux/u64_stats_sync.h>
+#include <linux/errno.h>
+#include <linux/types.h>
+#include <linux/pci.h>
+#include <linux/device.h>
+#include <linux/dma-mapping.h>
+#include <linux/slab.h>
+#include <linux/interrupt.h>
+#include <linux/skbuff.h>
+#include <linux/smp.h>
+#include <asm/byteorder.h>
+#include <linux/ip.h>
+#include <linux/tcp.h>
+#include <linux/sctp.h>
+#include <linux/ipv6.h>
+#include <net/ipv6.h>
+#include <net/checksum.h>
+#include <net/ip6_checksum.h>
+
+#include "hinic_common.h"
+#include "hinic_hw_if.h"
+#include "hinic_hw_wqe.h"
+#include "hinic_hw_wq.h"
+#include "hinic_hw_qp.h"
+#include "hinic_hw_dev.h"
+#include "hinic_dev.h"
+#include "hinic_tx.h"
+
+#define TX_IRQ_NO_PENDING               0
+#define TX_IRQ_NO_COALESC               0
+#define TX_IRQ_NO_LLI_TIMER             0
+#define TX_IRQ_NO_CREDIT                0
+#define TX_IRQ_NO_RESEND_TIMER          0
+
+#define CI_UPDATE_NO_PENDING            0
+#define CI_UPDATE_NO_COALESC            0
+
+#define HW_CONS_IDX(sq)                 be16_to_cpu(*(u16 *)((sq)->hw_ci_addr))
+
+#define MIN_SKB_LEN			32
+
+#define	MAX_PAYLOAD_OFFSET	        221
+#define TRANSPORT_OFFSET(l4_hdr, skb)	((u32)((l4_hdr) - (skb)->data))
+
+union hinic_l3 {
+	struct iphdr *v4;
+	struct ipv6hdr *v6;
+	unsigned char *hdr;
+};
+
+union hinic_l4 {
+	struct tcphdr *tcp;
+	struct udphdr *udp;
+	unsigned char *hdr;
+};
+
+enum hinic_offload_type {
+	TX_OFFLOAD_TSO     = BIT(0),
+	TX_OFFLOAD_CSUM    = BIT(1),
+	TX_OFFLOAD_VLAN    = BIT(2),
+	TX_OFFLOAD_INVALID = BIT(3),
+};
+
+/**
+ * hinic_txq_clean_stats - Clean the statistics of specific queue
+ * @txq: Logical Tx Queue
+ **/
+static void hinic_txq_clean_stats(struct hinic_txq *txq)
+{
+	struct hinic_txq_stats *txq_stats = &txq->txq_stats;
+
+	u64_stats_update_begin(&txq_stats->syncp);
+	txq_stats->pkts    = 0;
+	txq_stats->bytes   = 0;
+	txq_stats->tx_busy = 0;
+	txq_stats->tx_wake = 0;
+	txq_stats->tx_dropped = 0;
+	txq_stats->big_frags_pkts = 0;
+	u64_stats_update_end(&txq_stats->syncp);
+}
+
+/**
+ * hinic_txq_get_stats - get statistics of Tx Queue
+ * @txq: Logical Tx Queue
+ * @stats: return updated stats here
+ **/
+void hinic_txq_get_stats(struct hinic_txq *txq, struct hinic_txq_stats *stats)
+{
+	struct hinic_txq_stats *txq_stats = &txq->txq_stats;
+	unsigned int start;
+
+	do {
+		start = u64_stats_fetch_begin(&txq_stats->syncp);
+		stats->pkts    = txq_stats->pkts;
+		stats->bytes   = txq_stats->bytes;
+		stats->tx_busy = txq_stats->tx_busy;
+		stats->tx_wake = txq_stats->tx_wake;
+		stats->tx_dropped = txq_stats->tx_dropped;
+		stats->big_frags_pkts = txq_stats->big_frags_pkts;
+	} while (u64_stats_fetch_retry(&txq_stats->syncp, start));
+}
+
+/**
+ * txq_stats_init - Initialize the statistics of specific queue
+ * @txq: Logical Tx Queue
+ **/
+static void txq_stats_init(struct hinic_txq *txq)
+{
+	struct hinic_txq_stats *txq_stats = &txq->txq_stats;
+
+	u64_stats_init(&txq_stats->syncp);
+	hinic_txq_clean_stats(txq);
+}
+
+/**
+ * tx_map_skb - dma mapping for skb and return sges
+ * @nic_dev: nic device
+ * @skb: the skb
+ * @sges: returned sges
+ *
+ * Return 0 - Success, negative - Failure
+ **/
+static int tx_map_skb(struct hinic_dev *nic_dev, struct sk_buff *skb,
+		      struct hinic_sge *sges)
+{
+	struct hinic_hwdev *hwdev = nic_dev->hwdev;
+	struct hinic_hwif *hwif = hwdev->hwif;
+	struct pci_dev *pdev = hwif->pdev;
+	skb_frag_t *frag;
+	dma_addr_t dma_addr;
+	int i, j;
+
+	dma_addr = dma_map_single(&pdev->dev, skb->data, skb_headlen(skb),
+				  DMA_TO_DEVICE);
+	if (dma_mapping_error(&pdev->dev, dma_addr)) {
+		dev_err(&pdev->dev, "Failed to map Tx skb data\n");
+		return -EFAULT;
+	}
+
+	hinic_set_sge(&sges[0], dma_addr, skb_headlen(skb));
+
+	for (i = 0 ; i < skb_shinfo(skb)->nr_frags; i++) {
+		frag = &skb_shinfo(skb)->frags[i];
+
+		dma_addr = skb_frag_dma_map(&pdev->dev, frag, 0,
+					    skb_frag_size(frag),
+					    DMA_TO_DEVICE);
+		if (dma_mapping_error(&pdev->dev, dma_addr)) {
+			dev_err(&pdev->dev, "Failed to map Tx skb frag\n");
+			goto err_tx_map;
+		}
+
+		hinic_set_sge(&sges[i + 1], dma_addr, skb_frag_size(frag));
+	}
+
+	return 0;
+
+err_tx_map:
+	for (j = 0; j < i; j++)
+		dma_unmap_page(&pdev->dev, hinic_sge_to_dma(&sges[j + 1]),
+			       sges[j + 1].len, DMA_TO_DEVICE);
+
+	dma_unmap_single(&pdev->dev, hinic_sge_to_dma(&sges[0]), sges[0].len,
+			 DMA_TO_DEVICE);
+	return -EFAULT;
+}
+
+/**
+ * tx_unmap_skb - unmap the dma address of the skb
+ * @nic_dev: nic device
+ * @skb: the skb
+ * @sges: the sges that are connected to the skb
+ **/
+static void tx_unmap_skb(struct hinic_dev *nic_dev, struct sk_buff *skb,
+			 struct hinic_sge *sges)
+{
+	struct hinic_hwdev *hwdev = nic_dev->hwdev;
+	struct hinic_hwif *hwif = hwdev->hwif;
+	struct pci_dev *pdev = hwif->pdev;
+	int i;
+
+	for (i = 0; i < skb_shinfo(skb)->nr_frags ; i++)
+		dma_unmap_page(&pdev->dev, hinic_sge_to_dma(&sges[i + 1]),
+			       sges[i + 1].len, DMA_TO_DEVICE);
+
+	dma_unmap_single(&pdev->dev, hinic_sge_to_dma(&sges[0]), sges[0].len,
+			 DMA_TO_DEVICE);
+}
+
+static void get_inner_l3_l4_type(struct sk_buff *skb, union hinic_l3 *ip,
+				 union hinic_l4 *l4,
+				 enum hinic_offload_type offload_type,
+				 enum hinic_l3_offload_type *l3_type,
+				 u8 *l4_proto)
+{
+	u8 *exthdr;
+
+	if (ip->v4->version == 4) {
+		*l3_type = (offload_type == TX_OFFLOAD_CSUM) ?
+			   IPV4_PKT_NO_CHKSUM_OFFLOAD :
+			   IPV4_PKT_WITH_CHKSUM_OFFLOAD;
+		*l4_proto = ip->v4->protocol;
+	} else if (ip->v4->version == 6) {
+		*l3_type = IPV6_PKT;
+		exthdr = ip->hdr + sizeof(*ip->v6);
+		*l4_proto = ip->v6->nexthdr;
+		if (exthdr != l4->hdr) {
+			int start = exthdr - skb->data;
+			__be16 frag_off;
+
+			ipv6_skip_exthdr(skb, start, l4_proto, &frag_off);
+		}
+	} else {
+		*l3_type = L3TYPE_UNKNOWN;
+		*l4_proto = 0;
+	}
+}
+
+static void get_inner_l4_info(struct sk_buff *skb, union hinic_l4 *l4,
+			      enum hinic_offload_type offload_type, u8 l4_proto,
+			      enum hinic_l4_offload_type *l4_offload,
+			      u32 *l4_len, u32 *offset)
+{
+	*l4_offload = OFFLOAD_DISABLE;
+	*offset = 0;
+	*l4_len = 0;
+
+	switch (l4_proto) {
+	case IPPROTO_TCP:
+		*l4_offload = TCP_OFFLOAD_ENABLE;
+		/* doff in unit of 4B */
+		*l4_len = l4->tcp->doff * 4;
+		*offset = *l4_len + TRANSPORT_OFFSET(l4->hdr, skb);
+		break;
+
+	case IPPROTO_UDP:
+		*l4_offload = UDP_OFFLOAD_ENABLE;
+		*l4_len = sizeof(struct udphdr);
+		*offset = TRANSPORT_OFFSET(l4->hdr, skb);
+		break;
+
+	case IPPROTO_SCTP:
+		/* only csum offload support sctp */
+		if (offload_type != TX_OFFLOAD_CSUM)
+			break;
+
+		*l4_offload = SCTP_OFFLOAD_ENABLE;
+		*l4_len = sizeof(struct sctphdr);
+		*offset = TRANSPORT_OFFSET(l4->hdr, skb);
+		break;
+
+	default:
+		break;
+	}
+}
+
+static __sum16 csum_magic(union hinic_l3 *ip, unsigned short proto)
+{
+	return (ip->v4->version == 4) ?
+		csum_tcpudp_magic(ip->v4->saddr, ip->v4->daddr, 0, proto, 0) :
+		csum_ipv6_magic(&ip->v6->saddr, &ip->v6->daddr, 0, proto, 0);
+}
+
+static int offload_tso(struct hinic_sq_task *task, u32 *queue_info,
+		       struct sk_buff *skb)
+{
+	u32 offset, l4_len, ip_identify, network_hdr_len;
+	enum hinic_l3_offload_type l3_offload;
+	enum hinic_l4_offload_type l4_offload;
+	union hinic_l3 ip;
+	union hinic_l4 l4;
+	u8 l4_proto;
+
+	if (!skb_is_gso(skb))
+		return 0;
+
+	if (skb_cow_head(skb, 0) < 0)
+		return -EPROTONOSUPPORT;
+
+	if (skb->encapsulation) {
+		u32 gso_type = skb_shinfo(skb)->gso_type;
+		u32 tunnel_type = 0;
+		u32 l4_tunnel_len;
+
+		ip.hdr = skb_network_header(skb);
+		l4.hdr = skb_transport_header(skb);
+		network_hdr_len = skb_inner_network_header_len(skb);
+
+		if (ip.v4->version == 4) {
+			ip.v4->tot_len = 0;
+			l3_offload = IPV4_PKT_WITH_CHKSUM_OFFLOAD;
+		} else if (ip.v4->version == 6) {
+			l3_offload = IPV6_PKT;
+		} else {
+			l3_offload = 0;
+		}
+
+		hinic_task_set_outter_l3(task, l3_offload,
+					 skb_network_header_len(skb));
+
+		if (gso_type & SKB_GSO_UDP_TUNNEL_CSUM) {
+			l4.udp->check = ~csum_magic(&ip, IPPROTO_UDP);
+			tunnel_type = TUNNEL_UDP_CSUM;
+		} else if (gso_type & SKB_GSO_UDP_TUNNEL) {
+			tunnel_type = TUNNEL_UDP_NO_CSUM;
+		}
+
+		l4_tunnel_len = skb_inner_network_offset(skb) -
+				skb_transport_offset(skb);
+		hinic_task_set_tunnel_l4(task, tunnel_type, l4_tunnel_len);
+
+		ip.hdr = skb_inner_network_header(skb);
+		l4.hdr = skb_inner_transport_header(skb);
+	} else {
+		ip.hdr = skb_network_header(skb);
+		l4.hdr = skb_transport_header(skb);
+		network_hdr_len = skb_network_header_len(skb);
+	}
+
+	/* initialize inner IP header fields */
+	if (ip.v4->version == 4)
+		ip.v4->tot_len = 0;
+	else
+		ip.v6->payload_len = 0;
+
+	get_inner_l3_l4_type(skb, &ip, &l4, TX_OFFLOAD_TSO, &l3_offload,
+			     &l4_proto);
+
+	hinic_task_set_inner_l3(task, l3_offload, network_hdr_len);
+
+	ip_identify = 0;
+	if (l4_proto == IPPROTO_TCP)
+		l4.tcp->check = ~csum_magic(&ip, IPPROTO_TCP);
+
+	get_inner_l4_info(skb, &l4, TX_OFFLOAD_TSO, l4_proto, &l4_offload,
+			  &l4_len, &offset);
+
+	hinic_set_tso_inner_l4(task, queue_info, l4_offload, l4_len, offset,
+			       ip_identify, skb_shinfo(skb)->gso_size);
+
+	return 1;
+}
+
+static int offload_csum(struct hinic_sq_task *task, u32 *queue_info,
+			struct sk_buff *skb)
+{
+	enum hinic_l4_offload_type l4_offload;
+	u32 offset, l4_len, network_hdr_len;
+	enum hinic_l3_offload_type l3_type;
+	u32 tunnel_type = NOT_TUNNEL;
+	union hinic_l3 ip;
+	union hinic_l4 l4;
+	u8 l4_proto;
+
+	if (skb->ip_summed != CHECKSUM_PARTIAL)
+		return 0;
+
+	if (skb->encapsulation) {
+		u32 l4_tunnel_len;
+
+		tunnel_type = TUNNEL_UDP_NO_CSUM;
+		ip.hdr = skb_network_header(skb);
+
+		if (ip.v4->version == 4) {
+			l3_type = IPV4_PKT_NO_CHKSUM_OFFLOAD;
+			l4_proto = ip.v4->protocol;
+		} else if (ip.v4->version == 6) {
+			unsigned char *exthdr;
+			__be16 frag_off;
+
+			l3_type = IPV6_PKT;
+			tunnel_type = TUNNEL_UDP_CSUM;
+			exthdr = ip.hdr + sizeof(*ip.v6);
+			l4_proto = ip.v6->nexthdr;
+			l4.hdr = skb_transport_header(skb);
+			if (l4.hdr != exthdr)
+				ipv6_skip_exthdr(skb, exthdr - skb->data,
+						 &l4_proto, &frag_off);
+		} else {
+			l3_type = L3TYPE_UNKNOWN;
+			l4_proto = IPPROTO_RAW;
+		}
+
+		hinic_task_set_outter_l3(task, l3_type,
+					 skb_network_header_len(skb));
+
+		switch (l4_proto) {
+		case IPPROTO_UDP:
+			l4_tunnel_len = skb_inner_network_offset(skb) -
+					skb_transport_offset(skb);
+			ip.hdr = skb_inner_network_header(skb);
+			l4.hdr = skb_inner_transport_header(skb);
+			network_hdr_len = skb_inner_network_header_len(skb);
+			break;
+		case IPPROTO_IPIP:
+		case IPPROTO_IPV6:
+			tunnel_type = NOT_TUNNEL;
+			l4_tunnel_len = 0;
+
+			ip.hdr = skb_inner_network_header(skb);
+			l4.hdr = skb_transport_header(skb);
+			network_hdr_len = skb_network_header_len(skb);
+			break;
+		default:
+			/* Unsupported tunnel packet, disable csum offload */
+			skb_checksum_help(skb);
+			return 0;
+		}
+
+		hinic_task_set_tunnel_l4(task, tunnel_type, l4_tunnel_len);
+	} else {
+		ip.hdr = skb_network_header(skb);
+		l4.hdr = skb_transport_header(skb);
+		network_hdr_len = skb_network_header_len(skb);
+	}
+
+	get_inner_l3_l4_type(skb, &ip, &l4, TX_OFFLOAD_CSUM, &l3_type,
+			     &l4_proto);
+
+	hinic_task_set_inner_l3(task, l3_type, network_hdr_len);
+
+	get_inner_l4_info(skb, &l4, TX_OFFLOAD_CSUM, l4_proto, &l4_offload,
+			  &l4_len, &offset);
+
+	hinic_set_cs_inner_l4(task, queue_info, l4_offload, l4_len, offset);
+
+	return 1;
+}
+
+static void offload_vlan(struct hinic_sq_task *task, u32 *queue_info,
+			 u16 vlan_tag, u16 vlan_pri)
+{
+	task->pkt_info0 |= HINIC_SQ_TASK_INFO0_SET(vlan_tag, VLAN_TAG) |
+				HINIC_SQ_TASK_INFO0_SET(1U, VLAN_OFFLOAD);
+
+	*queue_info |= HINIC_SQ_CTRL_SET(vlan_pri, QUEUE_INFO_PRI);
+}
+
+static int hinic_tx_offload(struct sk_buff *skb, struct hinic_sq_task *task,
+			    u32 *queue_info)
+{
+	enum hinic_offload_type offload = 0;
+	u16 vlan_tag;
+	int enabled;
+
+	enabled = offload_tso(task, queue_info, skb);
+	if (enabled > 0) {
+		offload |= TX_OFFLOAD_TSO;
+	} else if (enabled == 0) {
+		enabled = offload_csum(task, queue_info, skb);
+		if (enabled)
+			offload |= TX_OFFLOAD_CSUM;
+	} else {
+		return -EPROTONOSUPPORT;
+	}
+
+	if (unlikely(skb_vlan_tag_present(skb))) {
+		vlan_tag = skb_vlan_tag_get(skb);
+		offload_vlan(task, queue_info, vlan_tag,
+			     vlan_tag >> VLAN_PRIO_SHIFT);
+		offload |= TX_OFFLOAD_VLAN;
+	}
+
+	if (offload)
+		hinic_task_set_l2hdr(task, skb_network_offset(skb));
+
+	/* payload offset should not more than 221 */
+	if (HINIC_SQ_CTRL_GET(*queue_info, QUEUE_INFO_PLDOFF) >
+	    MAX_PAYLOAD_OFFSET) {
+		return -EPROTONOSUPPORT;
+	}
+
+	/* mss should not less than 80 */
+	if (HINIC_SQ_CTRL_GET(*queue_info, QUEUE_INFO_MSS) < HINIC_MSS_MIN) {
+		*queue_info = HINIC_SQ_CTRL_CLEAR(*queue_info, QUEUE_INFO_MSS);
+		*queue_info |= HINIC_SQ_CTRL_SET(HINIC_MSS_MIN, QUEUE_INFO_MSS);
+	}
+
+	return 0;
+}
+
+netdev_tx_t hinic_lb_xmit_frame(struct sk_buff *skb, struct net_device *netdev)
+{
+	struct hinic_dev *nic_dev = netdev_priv(netdev);
+	u16 prod_idx, q_id = skb->queue_mapping;
+	struct netdev_queue *netdev_txq;
+	int nr_sges, err = NETDEV_TX_OK;
+	struct hinic_sq_wqe *sq_wqe;
+	unsigned int wqe_size;
+	struct hinic_txq *txq;
+	struct hinic_qp *qp;
+
+	txq = &nic_dev->txqs[q_id];
+	qp = container_of(txq->sq, struct hinic_qp, sq);
+	nr_sges = skb_shinfo(skb)->nr_frags + 1;
+
+	err = tx_map_skb(nic_dev, skb, txq->sges);
+	if (err)
+		goto skb_error;
+
+	wqe_size = HINIC_SQ_WQE_SIZE(nr_sges);
+
+	sq_wqe = hinic_sq_get_wqe(txq->sq, wqe_size, &prod_idx);
+	if (!sq_wqe) {
+		netif_stop_subqueue(netdev, qp->q_id);
+
+		sq_wqe = hinic_sq_get_wqe(txq->sq, wqe_size, &prod_idx);
+		if (sq_wqe) {
+			netif_wake_subqueue(nic_dev->netdev, qp->q_id);
+			goto process_sq_wqe;
+		}
+
+		tx_unmap_skb(nic_dev, skb, txq->sges);
+
+		u64_stats_update_begin(&txq->txq_stats.syncp);
+		txq->txq_stats.tx_busy++;
+		u64_stats_update_end(&txq->txq_stats.syncp);
+		err = NETDEV_TX_BUSY;
+		wqe_size = 0;
+		goto flush_skbs;
+	}
+
+process_sq_wqe:
+	hinic_sq_prepare_wqe(txq->sq, sq_wqe, txq->sges, nr_sges);
+	hinic_sq_write_wqe(txq->sq, prod_idx, sq_wqe, skb, wqe_size);
+
+flush_skbs:
+	netdev_txq = netdev_get_tx_queue(netdev, q_id);
+	if ((!netdev_xmit_more()) || (netif_xmit_stopped(netdev_txq)))
+		hinic_sq_write_db(txq->sq, prod_idx, wqe_size, 0);
+
+	return err;
+
+skb_error:
+	dev_kfree_skb_any(skb);
+	u64_stats_update_begin(&txq->txq_stats.syncp);
+	txq->txq_stats.tx_dropped++;
+	u64_stats_update_end(&txq->txq_stats.syncp);
+
+	return NETDEV_TX_OK;
+}
+
+netdev_tx_t hinic_xmit_frame(struct sk_buff *skb, struct net_device *netdev)
+{
+	struct hinic_dev *nic_dev = netdev_priv(netdev);
+	u16 prod_idx, q_id = skb->queue_mapping;
+	struct netdev_queue *netdev_txq;
+	int nr_sges, err = NETDEV_TX_OK;
+	struct hinic_sq_wqe *sq_wqe;
+	unsigned int wqe_size;
+	struct hinic_txq *txq;
+	struct hinic_qp *qp;
+
+	txq = &nic_dev->txqs[q_id];
+	qp = container_of(txq->sq, struct hinic_qp, sq);
+
+	if (skb->len < MIN_SKB_LEN) {
+		if (skb_pad(skb, MIN_SKB_LEN - skb->len)) {
+			netdev_err(netdev, "Failed to pad skb\n");
+			goto update_error_stats;
+		}
+
+		skb->len = MIN_SKB_LEN;
+	}
+
+	nr_sges = skb_shinfo(skb)->nr_frags + 1;
+	if (nr_sges > 17) {
+		u64_stats_update_begin(&txq->txq_stats.syncp);
+		txq->txq_stats.big_frags_pkts++;
+		u64_stats_update_end(&txq->txq_stats.syncp);
+	}
+
+	if (nr_sges > txq->max_sges) {
+		netdev_err(netdev, "Too many Tx sges\n");
+		goto skb_error;
+	}
+
+	err = tx_map_skb(nic_dev, skb, txq->sges);
+	if (err)
+		goto skb_error;
+
+	wqe_size = HINIC_SQ_WQE_SIZE(nr_sges);
+
+	sq_wqe = hinic_sq_get_wqe(txq->sq, wqe_size, &prod_idx);
+	if (!sq_wqe) {
+		netif_stop_subqueue(netdev, qp->q_id);
+
+		/* Check for the case free_tx_poll is called in another cpu
+		 * and we stopped the subqueue after free_tx_poll check.
+		 */
+		sq_wqe = hinic_sq_get_wqe(txq->sq, wqe_size, &prod_idx);
+		if (sq_wqe) {
+			netif_wake_subqueue(nic_dev->netdev, qp->q_id);
+			goto process_sq_wqe;
+		}
+
+		tx_unmap_skb(nic_dev, skb, txq->sges);
+
+		u64_stats_update_begin(&txq->txq_stats.syncp);
+		txq->txq_stats.tx_busy++;
+		u64_stats_update_end(&txq->txq_stats.syncp);
+		err = NETDEV_TX_BUSY;
+		wqe_size = 0;
+		goto flush_skbs;
+	}
+
+process_sq_wqe:
+	hinic_sq_prepare_wqe(txq->sq, sq_wqe, txq->sges, nr_sges);
+
+	err = hinic_tx_offload(skb, &sq_wqe->task, &sq_wqe->ctrl.queue_info);
+	if (err)
+		goto offload_error;
+
+	hinic_sq_write_wqe(txq->sq, prod_idx, sq_wqe, skb, wqe_size);
+
+flush_skbs:
+	netdev_txq = netdev_get_tx_queue(netdev, q_id);
+	if ((!netdev_xmit_more()) || (netif_xmit_stopped(netdev_txq)))
+		hinic_sq_write_db(txq->sq, prod_idx, wqe_size, 0);
+
+	return err;
+
+offload_error:
+	hinic_sq_return_wqe(txq->sq, wqe_size);
+	tx_unmap_skb(nic_dev, skb, txq->sges);
+
+skb_error:
+	dev_kfree_skb_any(skb);
+
+update_error_stats:
+	u64_stats_update_begin(&txq->txq_stats.syncp);
+	txq->txq_stats.tx_dropped++;
+	u64_stats_update_end(&txq->txq_stats.syncp);
+
+	return NETDEV_TX_OK;
+}
+
+/**
+ * tx_free_skb - unmap and free skb
+ * @nic_dev: nic device
+ * @skb: the skb
+ * @sges: the sges that are connected to the skb
+ **/
+static void tx_free_skb(struct hinic_dev *nic_dev, struct sk_buff *skb,
+			struct hinic_sge *sges)
+{
+	tx_unmap_skb(nic_dev, skb, sges);
+
+	dev_kfree_skb_any(skb);
+}
+
+/**
+ * free_all_tx_skbs - free all skbs in tx queue
+ * @txq: tx queue
+ **/
+static void free_all_tx_skbs(struct hinic_txq *txq)
+{
+	struct hinic_dev *nic_dev = netdev_priv(txq->netdev);
+	struct hinic_sq *sq = txq->sq;
+	struct hinic_sq_wqe *sq_wqe;
+	unsigned int wqe_size;
+	struct sk_buff *skb;
+	int nr_sges;
+	u16 ci;
+
+	while ((sq_wqe = hinic_sq_read_wqebb(sq, &skb, &wqe_size, &ci))) {
+		sq_wqe = hinic_sq_read_wqe(sq, &skb, wqe_size, &ci);
+		if (!sq_wqe)
+			break;
+
+		nr_sges = skb_shinfo(skb)->nr_frags + 1;
+
+		hinic_sq_get_sges(sq_wqe, txq->free_sges, nr_sges);
+
+		hinic_sq_put_wqe(sq, wqe_size);
+
+		tx_free_skb(nic_dev, skb, txq->free_sges);
+	}
+}
+
+/**
+ * free_tx_poll - free finished tx skbs in tx queue that connected to napi
+ * @napi: napi
+ * @budget: number of tx
+ *
+ * Return 0 - Success, negative - Failure
+ **/
+static int free_tx_poll(struct napi_struct *napi, int budget)
+{
+	struct hinic_txq *txq = container_of(napi, struct hinic_txq, napi);
+	struct hinic_qp *qp = container_of(txq->sq, struct hinic_qp, sq);
+	struct hinic_dev *nic_dev = netdev_priv(txq->netdev);
+	struct netdev_queue *netdev_txq;
+	struct hinic_sq *sq = txq->sq;
+	struct hinic_wq *wq = sq->wq;
+	struct hinic_sq_wqe *sq_wqe;
+	unsigned int wqe_size;
+	int nr_sges, pkts = 0;
+	struct sk_buff *skb;
+	u64 tx_bytes = 0;
+	u16 hw_ci, sw_ci;
+
+	do {
+		hw_ci = HW_CONS_IDX(sq) & wq->mask;
+
+		dma_rmb();
+
+		/* Reading a WQEBB to get real WQE size and consumer index. */
+		sq_wqe = hinic_sq_read_wqebb(sq, &skb, &wqe_size, &sw_ci);
+		if (!sq_wqe ||
+		    (((hw_ci - sw_ci) & wq->mask) * wq->wqebb_size < wqe_size))
+			break;
+
+		/* If this WQE have multiple WQEBBs, we will read again to get
+		 * full size WQE.
+		 */
+		if (wqe_size > wq->wqebb_size) {
+			sq_wqe = hinic_sq_read_wqe(sq, &skb, wqe_size, &sw_ci);
+			if (unlikely(!sq_wqe))
+				break;
+		}
+
+		tx_bytes += skb->len;
+		pkts++;
+
+		nr_sges = skb_shinfo(skb)->nr_frags + 1;
+
+		hinic_sq_get_sges(sq_wqe, txq->free_sges, nr_sges);
+
+		hinic_sq_put_wqe(sq, wqe_size);
+
+		tx_free_skb(nic_dev, skb, txq->free_sges);
+	} while (pkts < budget);
+
+	if (__netif_subqueue_stopped(nic_dev->netdev, qp->q_id) &&
+	    hinic_get_sq_free_wqebbs(sq) >= HINIC_MIN_TX_NUM_WQEBBS(sq)) {
+		netdev_txq = netdev_get_tx_queue(txq->netdev, qp->q_id);
+
+		__netif_tx_lock(netdev_txq, smp_processor_id());
+		if (!netif_testing(nic_dev->netdev))
+			netif_wake_subqueue(nic_dev->netdev, qp->q_id);
+
+		__netif_tx_unlock(netdev_txq);
+
+		u64_stats_update_begin(&txq->txq_stats.syncp);
+		txq->txq_stats.tx_wake++;
+		u64_stats_update_end(&txq->txq_stats.syncp);
+	}
+
+	u64_stats_update_begin(&txq->txq_stats.syncp);
+	txq->txq_stats.bytes += tx_bytes;
+	txq->txq_stats.pkts += pkts;
+	u64_stats_update_end(&txq->txq_stats.syncp);
+
+	if (pkts < budget) {
+		napi_complete(napi);
+		if (!HINIC_IS_VF(nic_dev->hwdev->hwif))
+			hinic_hwdev_set_msix_state(nic_dev->hwdev,
+						   sq->msix_entry,
+						   HINIC_MSIX_ENABLE);
+
+		return pkts;
+	}
+
+	return budget;
+}
+
+static irqreturn_t tx_irq(int irq, void *data)
+{
+	struct hinic_txq *txq = data;
+	struct hinic_dev *nic_dev;
+
+	nic_dev = netdev_priv(txq->netdev);
+
+	if (!HINIC_IS_VF(nic_dev->hwdev->hwif))
+		/* Disable the interrupt until napi will be completed */
+		hinic_hwdev_set_msix_state(nic_dev->hwdev,
+					   txq->sq->msix_entry,
+					   HINIC_MSIX_DISABLE);
+
+	hinic_hwdev_msix_cnt_set(nic_dev->hwdev, txq->sq->msix_entry);
+
+	napi_schedule(&txq->napi);
+	return IRQ_HANDLED;
+}
+
+static int tx_request_irq(struct hinic_txq *txq)
+{
+	struct hinic_dev *nic_dev = netdev_priv(txq->netdev);
+	struct hinic_msix_config interrupt_info = {0};
+	struct hinic_intr_coal_info *intr_coal = NULL;
+	struct hinic_hwdev *hwdev = nic_dev->hwdev;
+	struct hinic_hwif *hwif = hwdev->hwif;
+	struct pci_dev *pdev = hwif->pdev;
+	struct hinic_sq *sq = txq->sq;
+	struct hinic_qp *qp;
+	int err;
+
+	qp = container_of(sq, struct hinic_qp, sq);
+
+	netif_napi_add_weight(txq->netdev, &txq->napi, free_tx_poll,
+			      nic_dev->tx_weight);
+
+	hinic_hwdev_msix_set(nic_dev->hwdev, sq->msix_entry,
+			     TX_IRQ_NO_PENDING, TX_IRQ_NO_COALESC,
+			     TX_IRQ_NO_LLI_TIMER, TX_IRQ_NO_CREDIT,
+			     TX_IRQ_NO_RESEND_TIMER);
+
+	intr_coal = &nic_dev->tx_intr_coalesce[qp->q_id];
+	interrupt_info.msix_index = sq->msix_entry;
+	interrupt_info.coalesce_timer_cnt = intr_coal->coalesce_timer_cfg;
+	interrupt_info.pending_cnt = intr_coal->pending_limt;
+	interrupt_info.resend_timer_cnt = intr_coal->resend_timer_cfg;
+
+	err = hinic_set_interrupt_cfg(hwdev, &interrupt_info);
+	if (err) {
+		netif_err(nic_dev, drv, txq->netdev,
+			  "Failed to set TX interrupt coalescing attribute\n");
+		netif_napi_del(&txq->napi);
+		return err;
+	}
+
+	err = request_irq(sq->irq, tx_irq, 0, txq->irq_name, txq);
+	if (err) {
+		dev_err(&pdev->dev, "Failed to request Tx irq\n");
+		netif_napi_del(&txq->napi);
+		return err;
+	}
+
+	return 0;
+}
+
+static void tx_free_irq(struct hinic_txq *txq)
+{
+	struct hinic_sq *sq = txq->sq;
+
+	free_irq(sq->irq, txq);
+	netif_napi_del(&txq->napi);
+}
+
+/**
+ * hinic_init_txq - Initialize the Tx Queue
+ * @txq: Logical Tx Queue
+ * @sq: Hardware Tx Queue to connect the Logical queue with
+ * @netdev: network device to connect the Logical queue with
+ *
+ * Return 0 - Success, negative - Failure
+ **/
+int hinic_init_txq(struct hinic_txq *txq, struct hinic_sq *sq,
+		   struct net_device *netdev)
+{
+	struct hinic_qp *qp = container_of(sq, struct hinic_qp, sq);
+	struct hinic_dev *nic_dev = netdev_priv(netdev);
+	struct hinic_hwdev *hwdev = nic_dev->hwdev;
+	int err;
+
+	txq->netdev = netdev;
+	txq->sq = sq;
+
+	txq_stats_init(txq);
+
+	txq->max_sges = HINIC_MAX_SQ_BUFDESCS;
+
+	txq->sges = devm_kcalloc(&netdev->dev, txq->max_sges,
+				 sizeof(*txq->sges), GFP_KERNEL);
+	if (!txq->sges)
+		return -ENOMEM;
+
+	txq->free_sges = devm_kcalloc(&netdev->dev, txq->max_sges,
+				      sizeof(*txq->free_sges), GFP_KERNEL);
+	if (!txq->free_sges) {
+		err = -ENOMEM;
+		goto err_alloc_free_sges;
+	}
+
+	txq->irq_name = devm_kasprintf(&netdev->dev, GFP_KERNEL, "%s_txq%d",
+				       netdev->name, qp->q_id);
+	if (!txq->irq_name) {
+		err = -ENOMEM;
+		goto err_alloc_irqname;
+	}
+
+	err = hinic_hwdev_hw_ci_addr_set(hwdev, sq, CI_UPDATE_NO_PENDING,
+					 CI_UPDATE_NO_COALESC);
+	if (err)
+		goto err_hw_ci;
+
+	err = tx_request_irq(txq);
+	if (err) {
+		netdev_err(netdev, "Failed to request Tx irq\n");
+		goto err_req_tx_irq;
+	}
+
+	return 0;
+
+err_req_tx_irq:
+err_hw_ci:
+	devm_kfree(&netdev->dev, txq->irq_name);
+
+err_alloc_irqname:
+	devm_kfree(&netdev->dev, txq->free_sges);
+
+err_alloc_free_sges:
+	devm_kfree(&netdev->dev, txq->sges);
+	return err;
+}
+
+/**
+ * hinic_clean_txq - Clean the Tx Queue
+ * @txq: Logical Tx Queue
+ **/
+void hinic_clean_txq(struct hinic_txq *txq)
+{
+	struct net_device *netdev = txq->netdev;
+
+	tx_free_irq(txq);
+
+	free_all_tx_skbs(txq);
+
+	devm_kfree(&netdev->dev, txq->irq_name);
+	devm_kfree(&netdev->dev, txq->free_sges);
+	devm_kfree(&netdev->dev, txq->sges);
+}
diff --git a/drivers/net/ethernet/huawei/hinic/hinic_tx.h b/drivers/net/ethernet/huawei/hinic/hinic_tx.h
new file mode 100644
index 000000000000..91dc778362f3
--- /dev/null
+++ b/drivers/net/ethernet/huawei/hinic/hinic_tx.h
@@ -0,0 +1,54 @@
+/* SPDX-License-Identifier: GPL-2.0-only */
+/*
+ * Huawei HiNIC PCI Express Linux driver
+ * Copyright(c) 2017 Huawei Technologies Co., Ltd
+ */
+
+#ifndef HINIC_TX_H
+#define HINIC_TX_H
+
+#include <linux/types.h>
+#include <linux/netdevice.h>
+#include <linux/skbuff.h>
+#include <linux/u64_stats_sync.h>
+
+#include "hinic_common.h"
+#include "hinic_hw_qp.h"
+
+struct hinic_txq_stats {
+	u64     pkts;
+	u64     bytes;
+	u64     tx_busy;
+	u64     tx_wake;
+	u64     tx_dropped;
+	u64	big_frags_pkts;
+
+	struct u64_stats_sync   syncp;
+};
+
+struct hinic_txq {
+	struct net_device       *netdev;
+	struct hinic_sq         *sq;
+
+	struct hinic_txq_stats  txq_stats;
+
+	int                     max_sges;
+	struct hinic_sge        *sges;
+	struct hinic_sge        *free_sges;
+
+	char                    *irq_name;
+	struct napi_struct      napi;
+};
+
+void hinic_txq_get_stats(struct hinic_txq *txq, struct hinic_txq_stats *stats);
+
+netdev_tx_t hinic_lb_xmit_frame(struct sk_buff *skb, struct net_device *netdev);
+
+netdev_tx_t hinic_xmit_frame(struct sk_buff *skb, struct net_device *netdev);
+
+int hinic_init_txq(struct hinic_txq *txq, struct hinic_sq *sq,
+		   struct net_device *netdev);
+
+void hinic_clean_txq(struct hinic_txq *txq);
+
+#endif
diff --git a/drivers/net/ethernet/huawei/hinic3/Kconfig b/drivers/net/ethernet/huawei/hinic3/Kconfig
new file mode 100644
index 000000000000..ce4331d1387b
--- /dev/null
+++ b/drivers/net/ethernet/huawei/hinic3/Kconfig
@@ -0,0 +1,20 @@
+# SPDX-License-Identifier: GPL-2.0-only
+#
+# Huawei driver configuration
+#
+
+config HINIC3
+	tristate "Huawei 3rd generation network adapters (HINIC3) support"
+	# Fields of HW and management structures are little endian and are
+	# currently not converted
+	depends on !CPU_BIG_ENDIAN
+	depends on X86 || ARM64 || COMPILE_TEST
+	depends on PCI_MSI && 64BIT
+	select AUXILIARY_BUS
+	select PAGE_POOL
+	help
+	  This driver supports HiNIC 3rd gen Network Adapter (HINIC3).
+	  The driver is supported on X86_64 and ARM64 little endian.
+
+	  To compile this driver as a module, choose M here.
+	  The module will be called hinic3.
diff --git a/drivers/net/ethernet/huawei/hinic3/Makefile b/drivers/net/ethernet/huawei/hinic3/Makefile
new file mode 100644
index 000000000000..c3efa45a6a42
--- /dev/null
+++ b/drivers/net/ethernet/huawei/hinic3/Makefile
@@ -0,0 +1,25 @@
+# SPDX-License-Identifier: GPL-2.0
+# Copyright (c) Huawei Technologies Co., Ltd. 2025. All rights reserved.
+
+obj-$(CONFIG_HINIC3) += hinic3.o
+
+hinic3-objs := hinic3_cmdq.o \
+	       hinic3_common.o \
+	       hinic3_eqs.o \
+	       hinic3_hw_cfg.o \
+	       hinic3_hw_comm.o \
+	       hinic3_hwdev.o \
+	       hinic3_hwif.o \
+	       hinic3_irq.o \
+	       hinic3_lld.o \
+	       hinic3_main.o \
+	       hinic3_mbox.o \
+	       hinic3_mgmt.o \
+	       hinic3_netdev_ops.o \
+	       hinic3_nic_cfg.o \
+	       hinic3_nic_io.o \
+	       hinic3_queue_common.o \
+	       hinic3_rss.o \
+	       hinic3_rx.o \
+	       hinic3_tx.o \
+	       hinic3_wq.o
diff --git a/drivers/net/ethernet/huawei/hinic3/hinic3_cmdq.c b/drivers/net/ethernet/huawei/hinic3/hinic3_cmdq.c
new file mode 100644
index 000000000000..ef539d1b69a3
--- /dev/null
+++ b/drivers/net/ethernet/huawei/hinic3/hinic3_cmdq.c
@@ -0,0 +1,915 @@
+// SPDX-License-Identifier: GPL-2.0
+// Copyright (c) Huawei Technologies Co., Ltd. 2025. All rights reserved.
+
+#include <linux/bitfield.h>
+#include <linux/delay.h>
+#include <linux/dma-mapping.h>
+
+#include "hinic3_cmdq.h"
+#include "hinic3_hwdev.h"
+#include "hinic3_hwif.h"
+#include "hinic3_mbox.h"
+
+#define CMDQ_BUF_SIZE             2048
+#define CMDQ_WQEBB_SIZE           64
+
+#define CMDQ_CMD_TIMEOUT          5000
+#define CMDQ_ENABLE_WAIT_TIMEOUT  300
+
+#define CMDQ_CTXT_CURR_WQE_PAGE_PFN_MASK  GENMASK_ULL(51, 0)
+#define CMDQ_CTXT_EQ_ID_MASK              GENMASK_ULL(60, 53)
+#define CMDQ_CTXT_CEQ_ARM_MASK            BIT_ULL(61)
+#define CMDQ_CTXT_CEQ_EN_MASK             BIT_ULL(62)
+#define CMDQ_CTXT_HW_BUSY_BIT_MASK        BIT_ULL(63)
+
+#define CMDQ_CTXT_WQ_BLOCK_PFN_MASK       GENMASK_ULL(51, 0)
+#define CMDQ_CTXT_CI_MASK                 GENMASK_ULL(63, 52)
+#define CMDQ_CTXT_SET(val, member)  \
+	FIELD_PREP(CMDQ_CTXT_##member##_MASK, val)
+
+#define CMDQ_WQE_HDR_BUFDESC_LEN_MASK        GENMASK(7, 0)
+#define CMDQ_WQE_HDR_COMPLETE_FMT_MASK       BIT(15)
+#define CMDQ_WQE_HDR_DATA_FMT_MASK           BIT(22)
+#define CMDQ_WQE_HDR_COMPLETE_REQ_MASK       BIT(23)
+#define CMDQ_WQE_HDR_COMPLETE_SECT_LEN_MASK  GENMASK(28, 27)
+#define CMDQ_WQE_HDR_CTRL_LEN_MASK           GENMASK(30, 29)
+#define CMDQ_WQE_HDR_HW_BUSY_BIT_MASK        BIT(31)
+#define CMDQ_WQE_HDR_SET(val, member)  \
+	FIELD_PREP(CMDQ_WQE_HDR_##member##_MASK, val)
+#define CMDQ_WQE_HDR_GET(val, member)  \
+	FIELD_GET(CMDQ_WQE_HDR_##member##_MASK, le32_to_cpu(val))
+
+#define CMDQ_CTRL_PI_MASK              GENMASK(15, 0)
+#define CMDQ_CTRL_CMD_MASK             GENMASK(23, 16)
+#define CMDQ_CTRL_MOD_MASK             GENMASK(28, 24)
+#define CMDQ_CTRL_HW_BUSY_BIT_MASK     BIT(31)
+#define CMDQ_CTRL_SET(val, member)  \
+	FIELD_PREP(CMDQ_CTRL_##member##_MASK, val)
+#define CMDQ_CTRL_GET(val, member)  \
+	FIELD_GET(CMDQ_CTRL_##member##_MASK, val)
+
+#define CMDQ_WQE_ERRCODE_VAL_MASK      GENMASK(30, 0)
+#define CMDQ_WQE_ERRCODE_GET(val, member)  \
+	FIELD_GET(CMDQ_WQE_ERRCODE_##member##_MASK, le32_to_cpu(val))
+
+#define CMDQ_DB_INFO_HI_PROD_IDX_MASK  GENMASK(7, 0)
+#define CMDQ_DB_INFO_SET(val, member)  \
+	FIELD_PREP(CMDQ_DB_INFO_##member##_MASK, val)
+
+#define CMDQ_DB_HEAD_QUEUE_TYPE_MASK   BIT(23)
+#define CMDQ_DB_HEAD_CMDQ_TYPE_MASK    GENMASK(26, 24)
+#define CMDQ_DB_HEAD_SET(val, member)  \
+	FIELD_PREP(CMDQ_DB_HEAD_##member##_MASK, val)
+
+#define CMDQ_CEQE_TYPE_MASK            GENMASK(2, 0)
+#define CMDQ_CEQE_GET(val, member)  \
+	FIELD_GET(CMDQ_CEQE_##member##_MASK, le32_to_cpu(val))
+
+#define CMDQ_WQE_HEADER(wqe)           ((struct cmdq_header *)(wqe))
+#define CMDQ_WQE_COMPLETED(ctrl_info)  \
+	CMDQ_CTRL_GET(le32_to_cpu(ctrl_info), HW_BUSY_BIT)
+
+#define CMDQ_PFN(addr)  ((addr) >> 12)
+
+/* cmdq work queue's chip logical address table is up to 512B */
+#define CMDQ_WQ_CLA_SIZE  512
+
+/* Completion codes: send, direct sync, force stop */
+#define CMDQ_SEND_CMPT_CODE         10
+#define CMDQ_DIRECT_SYNC_CMPT_CODE  11
+#define CMDQ_FORCE_STOP_CMPT_CODE   12
+
+enum cmdq_data_format {
+	CMDQ_DATA_SGE    = 0,
+	CMDQ_DATA_DIRECT = 1,
+};
+
+enum cmdq_ctrl_sect_len {
+	CMDQ_CTRL_SECT_LEN        = 1,
+	CMDQ_CTRL_DIRECT_SECT_LEN = 2,
+};
+
+enum cmdq_bufdesc_len {
+	CMDQ_BUFDESC_LCMD_LEN = 2,
+	CMDQ_BUFDESC_SCMD_LEN = 3,
+};
+
+enum cmdq_completion_format {
+	CMDQ_COMPLETE_DIRECT = 0,
+	CMDQ_COMPLETE_SGE    = 1,
+};
+
+enum cmdq_cmd_type {
+	CMDQ_CMD_DIRECT_RESP,
+	CMDQ_CMD_SGE_RESP,
+};
+
+#define CMDQ_WQE_NUM_WQEBBS  1
+
+static struct cmdq_wqe *cmdq_read_wqe(struct hinic3_wq *wq, u16 *ci)
+{
+	if (hinic3_wq_get_used(wq) == 0)
+		return NULL;
+
+	*ci = wq->cons_idx & wq->idx_mask;
+
+	return get_q_element(&wq->qpages, wq->cons_idx, NULL);
+}
+
+struct hinic3_cmd_buf *hinic3_alloc_cmd_buf(struct hinic3_hwdev *hwdev)
+{
+	struct hinic3_cmd_buf *cmd_buf;
+	struct hinic3_cmdqs *cmdqs;
+
+	cmdqs = hwdev->cmdqs;
+
+	cmd_buf = kmalloc(sizeof(*cmd_buf), GFP_ATOMIC);
+	if (!cmd_buf)
+		return NULL;
+
+	cmd_buf->buf = dma_pool_alloc(cmdqs->cmd_buf_pool, GFP_ATOMIC,
+				      &cmd_buf->dma_addr);
+	if (!cmd_buf->buf) {
+		dev_err(hwdev->dev, "Failed to allocate cmdq cmd buf from the pool\n");
+		goto err_free_cmd_buf;
+	}
+
+	cmd_buf->size = cpu_to_le16(CMDQ_BUF_SIZE);
+	refcount_set(&cmd_buf->ref_cnt, 1);
+
+	return cmd_buf;
+
+err_free_cmd_buf:
+	kfree(cmd_buf);
+
+	return NULL;
+}
+
+void hinic3_free_cmd_buf(struct hinic3_hwdev *hwdev,
+			 struct hinic3_cmd_buf *cmd_buf)
+{
+	struct hinic3_cmdqs *cmdqs;
+
+	if (!refcount_dec_and_test(&cmd_buf->ref_cnt))
+		return;
+
+	cmdqs = hwdev->cmdqs;
+
+	dma_pool_free(cmdqs->cmd_buf_pool, cmd_buf->buf, cmd_buf->dma_addr);
+	kfree(cmd_buf);
+}
+
+static void cmdq_clear_cmd_buf(struct hinic3_cmdq_cmd_info *cmd_info,
+			       struct hinic3_hwdev *hwdev)
+{
+	if (cmd_info->buf_in) {
+		hinic3_free_cmd_buf(hwdev, cmd_info->buf_in);
+		cmd_info->buf_in = NULL;
+	}
+}
+
+static void clear_wqe_complete_bit(struct hinic3_cmdq *cmdq,
+				   struct cmdq_wqe *wqe, u16 ci)
+{
+	struct cmdq_header *hdr = CMDQ_WQE_HEADER(wqe);
+	__le32 header_info = hdr->header_info;
+	enum cmdq_data_format df;
+	struct cmdq_ctrl *ctrl;
+
+	df = CMDQ_WQE_HDR_GET(header_info, DATA_FMT);
+	if (df == CMDQ_DATA_SGE)
+		ctrl = &wqe->wqe_lcmd.ctrl;
+	else
+		ctrl = &wqe->wqe_scmd.ctrl;
+
+	/* clear HW busy bit */
+	ctrl->ctrl_info = 0;
+	cmdq->cmd_infos[ci].cmd_type = HINIC3_CMD_TYPE_NONE;
+	wmb(); /* verify wqe is clear before updating ci */
+	hinic3_wq_put_wqebbs(&cmdq->wq, CMDQ_WQE_NUM_WQEBBS);
+}
+
+static void cmdq_update_cmd_status(struct hinic3_cmdq *cmdq, u16 prod_idx,
+				   struct cmdq_wqe *wqe)
+{
+	struct hinic3_cmdq_cmd_info *cmd_info;
+	struct cmdq_wqe_lcmd *wqe_lcmd;
+	__le32 status_info;
+
+	wqe_lcmd = &wqe->wqe_lcmd;
+	cmd_info = &cmdq->cmd_infos[prod_idx];
+	if (cmd_info->errcode) {
+		status_info = wqe_lcmd->status.status_info;
+		*cmd_info->errcode = CMDQ_WQE_ERRCODE_GET(status_info, VAL);
+	}
+
+	if (cmd_info->direct_resp)
+		*cmd_info->direct_resp = wqe_lcmd->completion.resp.direct.val;
+}
+
+static void cmdq_sync_cmd_handler(struct hinic3_cmdq *cmdq,
+				  struct cmdq_wqe *wqe, u16 ci)
+{
+	spin_lock(&cmdq->cmdq_lock);
+	cmdq_update_cmd_status(cmdq, ci, wqe);
+	if (cmdq->cmd_infos[ci].cmpt_code) {
+		*cmdq->cmd_infos[ci].cmpt_code = CMDQ_DIRECT_SYNC_CMPT_CODE;
+		cmdq->cmd_infos[ci].cmpt_code = NULL;
+	}
+
+	/* Ensure that completion code has been updated before updating done */
+	smp_wmb();
+	if (cmdq->cmd_infos[ci].done) {
+		complete(cmdq->cmd_infos[ci].done);
+		cmdq->cmd_infos[ci].done = NULL;
+	}
+	spin_unlock(&cmdq->cmdq_lock);
+
+	cmdq_clear_cmd_buf(&cmdq->cmd_infos[ci], cmdq->hwdev);
+	clear_wqe_complete_bit(cmdq, wqe, ci);
+}
+
+void hinic3_cmdq_ceq_handler(struct hinic3_hwdev *hwdev, __le32 ceqe_data)
+{
+	enum hinic3_cmdq_type cmdq_type = CMDQ_CEQE_GET(ceqe_data, TYPE);
+	struct hinic3_cmdqs *cmdqs = hwdev->cmdqs;
+	struct hinic3_cmdq_cmd_info *cmd_info;
+	struct cmdq_wqe_lcmd *wqe_lcmd;
+	struct hinic3_cmdq *cmdq;
+	struct cmdq_wqe *wqe;
+	__le32 ctrl_info;
+	u16 ci;
+
+	if (unlikely(cmdq_type >= ARRAY_SIZE(cmdqs->cmdq)))
+		return;
+
+	cmdq = &cmdqs->cmdq[cmdq_type];
+	while ((wqe = cmdq_read_wqe(&cmdq->wq, &ci)) != NULL) {
+		cmd_info = &cmdq->cmd_infos[ci];
+		switch (cmd_info->cmd_type) {
+		case HINIC3_CMD_TYPE_NONE:
+			return;
+		case HINIC3_CMD_TYPE_TIMEOUT:
+			dev_warn(hwdev->dev, "Cmdq timeout, q_id: %u, ci: %u\n",
+				 cmdq_type, ci);
+			fallthrough;
+		case HINIC3_CMD_TYPE_FAKE_TIMEOUT:
+			cmdq_clear_cmd_buf(cmd_info, hwdev);
+			clear_wqe_complete_bit(cmdq, wqe, ci);
+			break;
+		default:
+			/* only arm bit is using scmd wqe,
+			 * the other wqe is lcmd
+			 */
+			wqe_lcmd = &wqe->wqe_lcmd;
+			ctrl_info = wqe_lcmd->ctrl.ctrl_info;
+			if (!CMDQ_WQE_COMPLETED(ctrl_info))
+				return;
+
+			dma_rmb();
+			/* For FORCE_STOP cmd_type, we also need to wait for
+			 * the firmware processing to complete to prevent the
+			 * firmware from accessing the released cmd_buf
+			 */
+			if (cmd_info->cmd_type == HINIC3_CMD_TYPE_FORCE_STOP) {
+				cmdq_clear_cmd_buf(cmd_info, hwdev);
+				clear_wqe_complete_bit(cmdq, wqe, ci);
+			} else {
+				cmdq_sync_cmd_handler(cmdq, wqe, ci);
+			}
+
+			break;
+		}
+	}
+}
+
+static int wait_cmdqs_enable(struct hinic3_cmdqs *cmdqs)
+{
+	unsigned long end;
+
+	end = jiffies + msecs_to_jiffies(CMDQ_ENABLE_WAIT_TIMEOUT);
+	do {
+		if (cmdqs->status & HINIC3_CMDQ_ENABLE)
+			return 0;
+		usleep_range(1000, 2000);
+	} while (time_before(jiffies, end) && !cmdqs->disable_flag);
+
+	cmdqs->disable_flag = 1;
+
+	return -EBUSY;
+}
+
+static void cmdq_set_completion(struct cmdq_completion *complete,
+				struct hinic3_cmd_buf *buf_out)
+{
+	struct hinic3_sge *sge = &complete->resp.sge;
+
+	hinic3_set_sge(sge, buf_out->dma_addr, cpu_to_le32(CMDQ_BUF_SIZE));
+}
+
+static struct cmdq_wqe *cmdq_get_wqe(struct hinic3_wq *wq, u16 *pi)
+{
+	if (!hinic3_wq_free_wqebbs(wq))
+		return NULL;
+
+	return hinic3_wq_get_one_wqebb(wq, pi);
+}
+
+static void cmdq_set_lcmd_bufdesc(struct cmdq_wqe_lcmd *wqe,
+				  struct hinic3_cmd_buf *buf_in)
+{
+	hinic3_set_sge(&wqe->buf_desc.sge, buf_in->dma_addr,
+		       (__force __le32)buf_in->size);
+}
+
+static void cmdq_set_db(struct hinic3_cmdq *cmdq,
+			enum hinic3_cmdq_type cmdq_type, u16 prod_idx)
+{
+	u8 __iomem *db_base = cmdq->hwdev->cmdqs->cmdqs_db_base;
+	u16 db_ofs = (prod_idx & 0xFF) << 3;
+	struct cmdq_db db;
+
+	db.db_info = cpu_to_le32(CMDQ_DB_INFO_SET(prod_idx >> 8, HI_PROD_IDX));
+	db.db_head = cpu_to_le32(CMDQ_DB_HEAD_SET(1, QUEUE_TYPE) |
+				 CMDQ_DB_HEAD_SET(cmdq_type, CMDQ_TYPE));
+	writeq(*(u64 *)&db, db_base + db_ofs);
+}
+
+static void cmdq_wqe_fill(struct cmdq_wqe *hw_wqe,
+			  const struct cmdq_wqe *shadow_wqe)
+{
+	const struct cmdq_header *src = (struct cmdq_header *)shadow_wqe;
+	struct cmdq_header *dst = (struct cmdq_header *)hw_wqe;
+	size_t len;
+
+	len = sizeof(struct cmdq_wqe) - sizeof(struct cmdq_header);
+	memcpy(dst + 1, src + 1, len);
+	/* Ensure buffer len before updating header */
+	wmb();
+	WRITE_ONCE(*dst, *src);
+}
+
+static void cmdq_prepare_wqe_ctrl(struct cmdq_wqe *wqe, u8 wrapped,
+				  u8 mod, u8 cmd, u16 prod_idx,
+				  enum cmdq_completion_format complete_format,
+				  enum cmdq_data_format data_format,
+				  enum cmdq_bufdesc_len buf_len)
+{
+	struct cmdq_header *hdr = CMDQ_WQE_HEADER(wqe);
+	enum cmdq_ctrl_sect_len ctrl_len;
+	struct cmdq_wqe_lcmd *wqe_lcmd;
+	struct cmdq_wqe_scmd *wqe_scmd;
+	struct cmdq_ctrl *ctrl;
+
+	if (data_format == CMDQ_DATA_SGE) {
+		wqe_lcmd = &wqe->wqe_lcmd;
+		wqe_lcmd->status.status_info = 0;
+		ctrl = &wqe_lcmd->ctrl;
+		ctrl_len = CMDQ_CTRL_SECT_LEN;
+	} else {
+		wqe_scmd = &wqe->wqe_scmd;
+		wqe_scmd->status.status_info = 0;
+		ctrl = &wqe_scmd->ctrl;
+		ctrl_len = CMDQ_CTRL_DIRECT_SECT_LEN;
+	}
+
+	ctrl->ctrl_info =
+		cpu_to_le32(CMDQ_CTRL_SET(prod_idx, PI) |
+			    CMDQ_CTRL_SET(cmd, CMD) |
+			    CMDQ_CTRL_SET(mod, MOD));
+
+	hdr->header_info =
+		cpu_to_le32(CMDQ_WQE_HDR_SET(buf_len, BUFDESC_LEN) |
+			    CMDQ_WQE_HDR_SET(complete_format, COMPLETE_FMT) |
+			    CMDQ_WQE_HDR_SET(data_format, DATA_FMT) |
+			    CMDQ_WQE_HDR_SET(1, COMPLETE_REQ) |
+			    CMDQ_WQE_HDR_SET(3, COMPLETE_SECT_LEN) |
+			    CMDQ_WQE_HDR_SET(ctrl_len, CTRL_LEN) |
+			    CMDQ_WQE_HDR_SET(wrapped, HW_BUSY_BIT));
+}
+
+static void cmdq_set_lcmd_wqe(struct cmdq_wqe *wqe,
+			      enum cmdq_cmd_type cmd_type,
+			      struct hinic3_cmd_buf *buf_in,
+			      struct hinic3_cmd_buf *buf_out,
+			      u8 wrapped, u8 mod, u8 cmd, u16 prod_idx)
+{
+	enum cmdq_completion_format complete_format = CMDQ_COMPLETE_DIRECT;
+	struct cmdq_wqe_lcmd *wqe_lcmd = &wqe->wqe_lcmd;
+
+	switch (cmd_type) {
+	case CMDQ_CMD_DIRECT_RESP:
+		wqe_lcmd->completion.resp.direct.val = 0;
+		break;
+	case CMDQ_CMD_SGE_RESP:
+		if (buf_out) {
+			complete_format = CMDQ_COMPLETE_SGE;
+			cmdq_set_completion(&wqe_lcmd->completion, buf_out);
+		}
+		break;
+	}
+
+	cmdq_prepare_wqe_ctrl(wqe, wrapped, mod, cmd, prod_idx, complete_format,
+			      CMDQ_DATA_SGE, CMDQ_BUFDESC_LCMD_LEN);
+	cmdq_set_lcmd_bufdesc(wqe_lcmd, buf_in);
+}
+
+static int hinic3_cmdq_sync_timeout_check(struct hinic3_cmdq *cmdq,
+					  struct cmdq_wqe *wqe, u16 pi)
+{
+	struct cmdq_wqe_lcmd *wqe_lcmd;
+	struct cmdq_ctrl *ctrl;
+	__le32 ctrl_info;
+
+	wqe_lcmd = &wqe->wqe_lcmd;
+	ctrl = &wqe_lcmd->ctrl;
+	ctrl_info = ctrl->ctrl_info;
+	if (!CMDQ_WQE_COMPLETED(ctrl_info)) {
+		dev_dbg(cmdq->hwdev->dev, "Cmdq sync command check busy bit not set\n");
+		return -EFAULT;
+	}
+	cmdq_update_cmd_status(cmdq, pi, wqe);
+
+	return 0;
+}
+
+static void clear_cmd_info(struct hinic3_cmdq_cmd_info *cmd_info,
+			   const struct hinic3_cmdq_cmd_info *saved_cmd_info)
+{
+	if (cmd_info->errcode == saved_cmd_info->errcode)
+		cmd_info->errcode = NULL;
+
+	if (cmd_info->done == saved_cmd_info->done)
+		cmd_info->done = NULL;
+
+	if (cmd_info->direct_resp == saved_cmd_info->direct_resp)
+		cmd_info->direct_resp = NULL;
+}
+
+static int wait_cmdq_sync_cmd_completion(struct hinic3_cmdq *cmdq,
+					 struct hinic3_cmdq_cmd_info *cmd_info,
+					 struct hinic3_cmdq_cmd_info *saved_cmd_info,
+					 u64 curr_msg_id, u16 curr_prod_idx,
+					 struct cmdq_wqe *curr_wqe,
+					 u32 timeout)
+{
+	ulong timeo = msecs_to_jiffies(timeout);
+	int err;
+
+	if (wait_for_completion_timeout(saved_cmd_info->done, timeo))
+		return 0;
+
+	spin_lock_bh(&cmdq->cmdq_lock);
+	if (cmd_info->cmpt_code == saved_cmd_info->cmpt_code)
+		cmd_info->cmpt_code = NULL;
+
+	if (*saved_cmd_info->cmpt_code == CMDQ_DIRECT_SYNC_CMPT_CODE) {
+		dev_dbg(cmdq->hwdev->dev, "Cmdq direct sync command has been completed\n");
+		spin_unlock_bh(&cmdq->cmdq_lock);
+		return 0;
+	}
+
+	if (curr_msg_id == cmd_info->cmdq_msg_id) {
+		err = hinic3_cmdq_sync_timeout_check(cmdq, curr_wqe,
+						     curr_prod_idx);
+		if (err)
+			cmd_info->cmd_type = HINIC3_CMD_TYPE_TIMEOUT;
+		else
+			cmd_info->cmd_type = HINIC3_CMD_TYPE_FAKE_TIMEOUT;
+	} else {
+		err = -ETIMEDOUT;
+		dev_err(cmdq->hwdev->dev,
+			"Cmdq sync command current msg id mismatch cmd_info msg id\n");
+	}
+
+	clear_cmd_info(cmd_info, saved_cmd_info);
+	spin_unlock_bh(&cmdq->cmdq_lock);
+
+	return err;
+}
+
+static int cmdq_sync_cmd_direct_resp(struct hinic3_cmdq *cmdq, u8 mod, u8 cmd,
+				     struct hinic3_cmd_buf *buf_in,
+				     __le64 *out_param)
+{
+	struct hinic3_cmdq_cmd_info *cmd_info, saved_cmd_info;
+	int cmpt_code = CMDQ_SEND_CMPT_CODE;
+	struct cmdq_wqe *curr_wqe, wqe = {};
+	struct hinic3_wq *wq = &cmdq->wq;
+	u16 curr_prod_idx, next_prod_idx;
+	struct completion done;
+	u64 curr_msg_id;
+	int errcode;
+	u8 wrapped;
+	int err;
+
+	spin_lock_bh(&cmdq->cmdq_lock);
+	curr_wqe = cmdq_get_wqe(wq, &curr_prod_idx);
+	if (!curr_wqe) {
+		spin_unlock_bh(&cmdq->cmdq_lock);
+		return -EBUSY;
+	}
+
+	wrapped = cmdq->wrapped;
+	next_prod_idx = curr_prod_idx + CMDQ_WQE_NUM_WQEBBS;
+	if (next_prod_idx >= wq->q_depth) {
+		cmdq->wrapped ^= 1;
+		next_prod_idx -= wq->q_depth;
+	}
+
+	cmd_info = &cmdq->cmd_infos[curr_prod_idx];
+	init_completion(&done);
+	refcount_inc(&buf_in->ref_cnt);
+	cmd_info->cmd_type = HINIC3_CMD_TYPE_DIRECT_RESP;
+	cmd_info->done = &done;
+	cmd_info->errcode = &errcode;
+	cmd_info->direct_resp = out_param;
+	cmd_info->cmpt_code = &cmpt_code;
+	cmd_info->buf_in = buf_in;
+	saved_cmd_info = *cmd_info;
+	cmdq_set_lcmd_wqe(&wqe, CMDQ_CMD_DIRECT_RESP, buf_in, NULL,
+			  wrapped, mod, cmd, curr_prod_idx);
+
+	cmdq_wqe_fill(curr_wqe, &wqe);
+	(cmd_info->cmdq_msg_id)++;
+	curr_msg_id = cmd_info->cmdq_msg_id;
+	cmdq_set_db(cmdq, HINIC3_CMDQ_SYNC, next_prod_idx);
+	spin_unlock_bh(&cmdq->cmdq_lock);
+
+	err = wait_cmdq_sync_cmd_completion(cmdq, cmd_info, &saved_cmd_info,
+					    curr_msg_id, curr_prod_idx,
+					    curr_wqe, CMDQ_CMD_TIMEOUT);
+	if (err) {
+		dev_err(cmdq->hwdev->dev,
+			"Cmdq sync command timeout, mod: %u, cmd: %u, prod idx: 0x%x\n",
+			mod, cmd, curr_prod_idx);
+		err = -ETIMEDOUT;
+	}
+
+	if (cmpt_code == CMDQ_FORCE_STOP_CMPT_CODE) {
+		dev_dbg(cmdq->hwdev->dev,
+			"Force stop cmdq cmd, mod: %u, cmd: %u\n", mod, cmd);
+		err = -EAGAIN;
+	}
+
+	smp_rmb(); /* read error code after completion */
+
+	return err ? err : errcode;
+}
+
+int hinic3_cmdq_direct_resp(struct hinic3_hwdev *hwdev, u8 mod, u8 cmd,
+			    struct hinic3_cmd_buf *buf_in, __le64 *out_param)
+{
+	struct hinic3_cmdqs *cmdqs;
+	int err;
+
+	cmdqs = hwdev->cmdqs;
+	err = wait_cmdqs_enable(cmdqs);
+	if (err) {
+		dev_err(hwdev->dev, "Cmdq is disabled\n");
+		return err;
+	}
+
+	err = cmdq_sync_cmd_direct_resp(&cmdqs->cmdq[HINIC3_CMDQ_SYNC],
+					mod, cmd, buf_in, out_param);
+
+	return err;
+}
+
+static void cmdq_init_queue_ctxt(struct hinic3_hwdev *hwdev, u8 cmdq_id,
+				 struct comm_cmdq_ctxt_info *ctxt_info)
+{
+	const struct hinic3_cmdqs *cmdqs;
+	u64 cmdq_first_block_paddr, pfn;
+	const struct hinic3_wq *wq;
+
+	cmdqs = hwdev->cmdqs;
+	wq = &cmdqs->cmdq[cmdq_id].wq;
+	pfn = CMDQ_PFN(hinic3_wq_get_first_wqe_page_addr(wq));
+
+	ctxt_info->curr_wqe_page_pfn =
+		cpu_to_le64(CMDQ_CTXT_SET(1, HW_BUSY_BIT) |
+			    CMDQ_CTXT_SET(1, CEQ_EN)	|
+			    CMDQ_CTXT_SET(1, CEQ_ARM)	|
+			    CMDQ_CTXT_SET(0, EQ_ID) |
+			    CMDQ_CTXT_SET(pfn, CURR_WQE_PAGE_PFN));
+
+	if (!hinic3_wq_is_0_level_cla(wq)) {
+		cmdq_first_block_paddr = cmdqs->wq_block_paddr;
+		pfn = CMDQ_PFN(cmdq_first_block_paddr);
+	}
+
+	ctxt_info->wq_block_pfn = cpu_to_le64(CMDQ_CTXT_SET(wq->cons_idx, CI) |
+					      CMDQ_CTXT_SET(pfn, WQ_BLOCK_PFN));
+}
+
+static int init_cmdq(struct hinic3_cmdq *cmdq, struct hinic3_hwdev *hwdev,
+		     enum hinic3_cmdq_type q_type)
+{
+	int err;
+
+	cmdq->cmdq_type = q_type;
+	cmdq->wrapped = 1;
+	cmdq->hwdev = hwdev;
+
+	spin_lock_init(&cmdq->cmdq_lock);
+
+	cmdq->cmd_infos = kcalloc(cmdq->wq.q_depth, sizeof(*cmdq->cmd_infos),
+				  GFP_KERNEL);
+	if (!cmdq->cmd_infos) {
+		err = -ENOMEM;
+		return err;
+	}
+
+	return 0;
+}
+
+static int hinic3_set_cmdq_ctxt(struct hinic3_hwdev *hwdev, u8 cmdq_id)
+{
+	struct comm_cmd_set_cmdq_ctxt cmdq_ctxt = {};
+	struct mgmt_msg_params msg_params = {};
+	int err;
+
+	cmdq_init_queue_ctxt(hwdev, cmdq_id, &cmdq_ctxt.ctxt);
+	cmdq_ctxt.func_id = hinic3_global_func_id(hwdev);
+	cmdq_ctxt.cmdq_id = cmdq_id;
+
+	mgmt_msg_params_init_default(&msg_params, &cmdq_ctxt,
+				     sizeof(cmdq_ctxt));
+
+	err = hinic3_send_mbox_to_mgmt(hwdev, MGMT_MOD_COMM,
+				       COMM_CMD_SET_CMDQ_CTXT, &msg_params);
+	if (err || cmdq_ctxt.head.status) {
+		dev_err(hwdev->dev, "Failed to set cmdq ctxt, err: %d, status: 0x%x\n",
+			err, cmdq_ctxt.head.status);
+		return -EFAULT;
+	}
+
+	return 0;
+}
+
+static int hinic3_set_cmdq_ctxts(struct hinic3_hwdev *hwdev)
+{
+	struct hinic3_cmdqs *cmdqs = hwdev->cmdqs;
+	u8 cmdq_type;
+	int err;
+
+	for (cmdq_type = 0; cmdq_type < cmdqs->cmdq_num; cmdq_type++) {
+		err = hinic3_set_cmdq_ctxt(hwdev, cmdq_type);
+		if (err)
+			return err;
+	}
+
+	cmdqs->status |= HINIC3_CMDQ_ENABLE;
+	cmdqs->disable_flag = 0;
+
+	return 0;
+}
+
+static int create_cmdq_wq(struct hinic3_hwdev *hwdev,
+			  struct hinic3_cmdqs *cmdqs)
+{
+	u8 cmdq_type;
+	int err;
+
+	for (cmdq_type = 0; cmdq_type < cmdqs->cmdq_num; cmdq_type++) {
+		err = hinic3_wq_create(hwdev, &cmdqs->cmdq[cmdq_type].wq,
+				       CMDQ_DEPTH, CMDQ_WQEBB_SIZE);
+		if (err) {
+			dev_err(hwdev->dev, "Failed to create cmdq wq\n");
+			goto err_destroy_wq;
+		}
+	}
+
+	/* 1-level Chip Logical Address (CLA) must put all
+	 * cmdq's wq page addr in one wq block
+	 */
+	if (!hinic3_wq_is_0_level_cla(&cmdqs->cmdq[HINIC3_CMDQ_SYNC].wq)) {
+		if (cmdqs->cmdq[HINIC3_CMDQ_SYNC].wq.qpages.num_pages >
+		    CMDQ_WQ_CLA_SIZE / sizeof(u64)) {
+			err = -EINVAL;
+			dev_err(hwdev->dev,
+				"Cmdq number of wq pages exceeds limit: %lu\n",
+				CMDQ_WQ_CLA_SIZE / sizeof(u64));
+			goto err_destroy_wq;
+		}
+
+		cmdqs->wq_block_vaddr =
+			dma_alloc_coherent(hwdev->dev, HINIC3_MIN_PAGE_SIZE,
+					   &cmdqs->wq_block_paddr, GFP_KERNEL);
+		if (!cmdqs->wq_block_vaddr) {
+			err = -ENOMEM;
+			goto err_destroy_wq;
+		}
+
+		for (cmdq_type = 0; cmdq_type < cmdqs->cmdq_num; cmdq_type++)
+			memcpy((u8 *)cmdqs->wq_block_vaddr +
+			       CMDQ_WQ_CLA_SIZE * cmdq_type,
+			       cmdqs->cmdq[cmdq_type].wq.wq_block_vaddr,
+			       cmdqs->cmdq[cmdq_type].wq.qpages.num_pages *
+			       sizeof(__be64));
+	}
+
+	return 0;
+
+err_destroy_wq:
+	while (cmdq_type > 0) {
+		cmdq_type--;
+		hinic3_wq_destroy(hwdev, &cmdqs->cmdq[cmdq_type].wq);
+	}
+
+	return err;
+}
+
+static void destroy_cmdq_wq(struct hinic3_hwdev *hwdev,
+			    struct hinic3_cmdqs *cmdqs)
+{
+	u8 cmdq_type;
+
+	if (cmdqs->wq_block_vaddr)
+		dma_free_coherent(hwdev->dev, HINIC3_MIN_PAGE_SIZE,
+				  cmdqs->wq_block_vaddr, cmdqs->wq_block_paddr);
+
+	for (cmdq_type = 0; cmdq_type < cmdqs->cmdq_num; cmdq_type++)
+		hinic3_wq_destroy(hwdev, &cmdqs->cmdq[cmdq_type].wq);
+}
+
+static int init_cmdqs(struct hinic3_hwdev *hwdev)
+{
+	struct hinic3_cmdqs *cmdqs;
+
+	cmdqs = kzalloc(sizeof(*cmdqs), GFP_KERNEL);
+	if (!cmdqs)
+		return -ENOMEM;
+
+	hwdev->cmdqs = cmdqs;
+	cmdqs->hwdev = hwdev;
+	cmdqs->cmdq_num = hwdev->max_cmdq;
+
+	cmdqs->cmd_buf_pool = dma_pool_create("hinic3_cmdq", hwdev->dev,
+					      CMDQ_BUF_SIZE, CMDQ_BUF_SIZE, 0);
+	if (!cmdqs->cmd_buf_pool) {
+		dev_err(hwdev->dev, "Failed to create cmdq buffer pool\n");
+		kfree(cmdqs);
+		return -ENOMEM;
+	}
+
+	return 0;
+}
+
+static void cmdq_flush_sync_cmd(struct hinic3_cmdq_cmd_info *cmd_info)
+{
+	if (cmd_info->cmd_type != HINIC3_CMD_TYPE_DIRECT_RESP)
+		return;
+
+	cmd_info->cmd_type = HINIC3_CMD_TYPE_FORCE_STOP;
+
+	if (cmd_info->cmpt_code &&
+	    *cmd_info->cmpt_code == CMDQ_SEND_CMPT_CODE)
+		*cmd_info->cmpt_code = CMDQ_FORCE_STOP_CMPT_CODE;
+
+	if (cmd_info->done) {
+		complete(cmd_info->done);
+		cmd_info->done = NULL;
+		cmd_info->cmpt_code = NULL;
+		cmd_info->direct_resp = NULL;
+		cmd_info->errcode = NULL;
+	}
+}
+
+static void hinic3_cmdq_flush_cmd(struct hinic3_cmdq *cmdq)
+{
+	struct hinic3_cmdq_cmd_info *cmd_info;
+	u16 ci;
+
+	spin_lock_bh(&cmdq->cmdq_lock);
+	while (cmdq_read_wqe(&cmdq->wq, &ci)) {
+		hinic3_wq_put_wqebbs(&cmdq->wq, CMDQ_WQE_NUM_WQEBBS);
+		cmd_info = &cmdq->cmd_infos[ci];
+		if (cmd_info->cmd_type == HINIC3_CMD_TYPE_DIRECT_RESP)
+			cmdq_flush_sync_cmd(cmd_info);
+	}
+	spin_unlock_bh(&cmdq->cmdq_lock);
+}
+
+void hinic3_cmdq_flush_sync_cmd(struct hinic3_hwdev *hwdev)
+{
+	struct hinic3_cmdq *cmdq;
+	u16 wqe_cnt, wqe_idx, i;
+	struct hinic3_wq *wq;
+
+	cmdq = &hwdev->cmdqs->cmdq[HINIC3_CMDQ_SYNC];
+	spin_lock_bh(&cmdq->cmdq_lock);
+	wq = &cmdq->wq;
+	wqe_cnt = hinic3_wq_get_used(wq);
+	for (i = 0; i < wqe_cnt; i++) {
+		wqe_idx = (wq->cons_idx + i) & wq->idx_mask;
+		cmdq_flush_sync_cmd(cmdq->cmd_infos + wqe_idx);
+	}
+	spin_unlock_bh(&cmdq->cmdq_lock);
+}
+
+static void hinic3_cmdq_reset_all_cmd_buf(struct hinic3_cmdq *cmdq)
+{
+	u16 i;
+
+	for (i = 0; i < cmdq->wq.q_depth; i++)
+		cmdq_clear_cmd_buf(&cmdq->cmd_infos[i], cmdq->hwdev);
+}
+
+int hinic3_reinit_cmdq_ctxts(struct hinic3_hwdev *hwdev)
+{
+	struct hinic3_cmdqs *cmdqs = hwdev->cmdqs;
+	u8 cmdq_type;
+
+	for (cmdq_type = 0; cmdq_type < cmdqs->cmdq_num; cmdq_type++) {
+		hinic3_cmdq_flush_cmd(&cmdqs->cmdq[cmdq_type]);
+		hinic3_cmdq_reset_all_cmd_buf(&cmdqs->cmdq[cmdq_type]);
+		cmdqs->cmdq[cmdq_type].wrapped = 1;
+		hinic3_wq_reset(&cmdqs->cmdq[cmdq_type].wq);
+	}
+
+	return hinic3_set_cmdq_ctxts(hwdev);
+}
+
+int hinic3_cmdqs_init(struct hinic3_hwdev *hwdev)
+{
+	struct hinic3_cmdqs *cmdqs;
+	void __iomem *db_base;
+	u8 cmdq_type;
+	int err;
+
+	err = init_cmdqs(hwdev);
+	if (err)
+		goto err_out;
+
+	cmdqs = hwdev->cmdqs;
+	err = create_cmdq_wq(hwdev, cmdqs);
+	if (err)
+		goto err_free_cmdqs;
+
+	err = hinic3_alloc_db_addr(hwdev, &db_base, NULL);
+	if (err) {
+		dev_err(hwdev->dev, "Failed to allocate doorbell address\n");
+		goto err_destroy_cmdq_wq;
+	}
+	cmdqs->cmdqs_db_base = db_base;
+
+	for (cmdq_type = 0; cmdq_type < cmdqs->cmdq_num; cmdq_type++) {
+		err = init_cmdq(&cmdqs->cmdq[cmdq_type], hwdev, cmdq_type);
+		if (err) {
+			dev_err(hwdev->dev,
+				"Failed to initialize cmdq type : %d\n",
+				cmdq_type);
+			goto err_free_cmd_infos;
+		}
+	}
+
+	err = hinic3_set_cmdq_ctxts(hwdev);
+	if (err)
+		goto err_free_cmd_infos;
+
+	return 0;
+
+err_free_cmd_infos:
+	while (cmdq_type > 0) {
+		cmdq_type--;
+		kfree(cmdqs->cmdq[cmdq_type].cmd_infos);
+	}
+
+	hinic3_free_db_addr(hwdev, cmdqs->cmdqs_db_base);
+
+err_destroy_cmdq_wq:
+	destroy_cmdq_wq(hwdev, cmdqs);
+
+err_free_cmdqs:
+	dma_pool_destroy(cmdqs->cmd_buf_pool);
+	kfree(cmdqs);
+
+err_out:
+	return err;
+}
+
+void hinic3_cmdqs_free(struct hinic3_hwdev *hwdev)
+{
+	struct hinic3_cmdqs *cmdqs = hwdev->cmdqs;
+	u8 cmdq_type;
+
+	cmdqs->status &= ~HINIC3_CMDQ_ENABLE;
+
+	for (cmdq_type = 0; cmdq_type < cmdqs->cmdq_num; cmdq_type++) {
+		hinic3_cmdq_flush_cmd(&cmdqs->cmdq[cmdq_type]);
+		hinic3_cmdq_reset_all_cmd_buf(&cmdqs->cmdq[cmdq_type]);
+		kfree(cmdqs->cmdq[cmdq_type].cmd_infos);
+	}
+
+	hinic3_free_db_addr(hwdev, cmdqs->cmdqs_db_base);
+	destroy_cmdq_wq(hwdev, cmdqs);
+	dma_pool_destroy(cmdqs->cmd_buf_pool);
+	kfree(cmdqs);
+}
+
+bool hinic3_cmdq_idle(struct hinic3_cmdq *cmdq)
+{
+	return hinic3_wq_get_used(&cmdq->wq) == 0;
+}
diff --git a/drivers/net/ethernet/huawei/hinic3/hinic3_cmdq.h b/drivers/net/ethernet/huawei/hinic3/hinic3_cmdq.h
new file mode 100644
index 000000000000..f99c386a2780
--- /dev/null
+++ b/drivers/net/ethernet/huawei/hinic3/hinic3_cmdq.h
@@ -0,0 +1,156 @@
+/* SPDX-License-Identifier: GPL-2.0 */
+/* Copyright (c) Huawei Technologies Co., Ltd. 2025. All rights reserved. */
+
+#ifndef _HINIC3_CMDQ_H_
+#define _HINIC3_CMDQ_H_
+
+#include <linux/dmapool.h>
+
+#include "hinic3_hw_intf.h"
+#include "hinic3_wq.h"
+
+#define CMDQ_DEPTH  4096
+
+struct cmdq_db {
+	__le32 db_head;
+	__le32 db_info;
+};
+
+/* hw defined cmdq wqe header */
+struct cmdq_header {
+	__le32 header_info;
+	__le32 saved_data;
+};
+
+struct cmdq_lcmd_bufdesc {
+	struct hinic3_sge sge;
+	__le64            rsvd2;
+	__le64            rsvd3;
+};
+
+struct cmdq_status {
+	__le32 status_info;
+};
+
+struct cmdq_ctrl {
+	__le32 ctrl_info;
+};
+
+struct cmdq_direct_resp {
+	__le64 val;
+	__le64 rsvd;
+};
+
+struct cmdq_completion {
+	union {
+		struct hinic3_sge       sge;
+		struct cmdq_direct_resp direct;
+	} resp;
+};
+
+struct cmdq_wqe_scmd {
+	struct cmdq_header     header;
+	__le64                 rsvd3;
+	struct cmdq_status     status;
+	struct cmdq_ctrl       ctrl;
+	struct cmdq_completion completion;
+	__le32                 rsvd10[6];
+};
+
+struct cmdq_wqe_lcmd {
+	struct cmdq_header       header;
+	struct cmdq_status       status;
+	struct cmdq_ctrl         ctrl;
+	struct cmdq_completion   completion;
+	struct cmdq_lcmd_bufdesc buf_desc;
+};
+
+struct cmdq_wqe {
+	union {
+		struct cmdq_wqe_scmd wqe_scmd;
+		struct cmdq_wqe_lcmd wqe_lcmd;
+	};
+};
+
+static_assert(sizeof(struct cmdq_wqe) == 64);
+
+enum hinic3_cmdq_type {
+	HINIC3_CMDQ_SYNC      = 0,
+	HINIC3_MAX_CMDQ_TYPES = 4
+};
+
+enum hinic3_cmdq_status {
+	HINIC3_CMDQ_ENABLE = BIT(0),
+};
+
+enum hinic3_cmdq_cmd_type {
+	HINIC3_CMD_TYPE_NONE,
+	HINIC3_CMD_TYPE_DIRECT_RESP,
+	HINIC3_CMD_TYPE_FAKE_TIMEOUT,
+	HINIC3_CMD_TYPE_TIMEOUT,
+	HINIC3_CMD_TYPE_FORCE_STOP,
+};
+
+struct hinic3_cmd_buf {
+	void       *buf;
+	dma_addr_t dma_addr;
+	__le16     size;
+	refcount_t ref_cnt;
+};
+
+struct hinic3_cmdq_cmd_info {
+	enum hinic3_cmdq_cmd_type cmd_type;
+	struct completion         *done;
+	int                       *errcode;
+	/* completion code */
+	int                       *cmpt_code;
+	__le64                    *direct_resp;
+	u64                       cmdq_msg_id;
+	struct hinic3_cmd_buf     *buf_in;
+};
+
+struct hinic3_cmdq {
+	struct hinic3_wq            wq;
+	enum hinic3_cmdq_type       cmdq_type;
+	u8                          wrapped;
+	/* synchronize command submission with completions via event queue */
+	spinlock_t                  cmdq_lock;
+	struct hinic3_cmdq_cmd_info *cmd_infos;
+	struct hinic3_hwdev         *hwdev;
+};
+
+struct hinic3_cmdqs {
+	struct hinic3_hwdev *hwdev;
+	struct hinic3_cmdq  cmdq[HINIC3_MAX_CMDQ_TYPES];
+	struct dma_pool     *cmd_buf_pool;
+	/* doorbell area */
+	u8 __iomem          *cmdqs_db_base;
+
+	/* When command queue uses multiple memory pages (1-level CLA), this
+	 * block will hold aggregated indirection table for all command queues
+	 * of cmdqs. Not used for small cmdq (0-level CLA).
+	 */
+	dma_addr_t          wq_block_paddr;
+	void                *wq_block_vaddr;
+
+	u32                 status;
+	u32                 disable_flag;
+	u8                  cmdq_num;
+};
+
+int hinic3_cmdqs_init(struct hinic3_hwdev *hwdev);
+void hinic3_cmdqs_free(struct hinic3_hwdev *hwdev);
+
+struct hinic3_cmd_buf *hinic3_alloc_cmd_buf(struct hinic3_hwdev *hwdev);
+void hinic3_free_cmd_buf(struct hinic3_hwdev *hwdev,
+			 struct hinic3_cmd_buf *cmd_buf);
+void hinic3_cmdq_ceq_handler(struct hinic3_hwdev *hwdev, __le32 ceqe_data);
+
+int hinic3_cmdq_direct_resp(struct hinic3_hwdev *hwdev, u8 mod, u8 cmd,
+			    struct hinic3_cmd_buf *buf_in, __le64 *out_param);
+
+void hinic3_cmdq_flush_sync_cmd(struct hinic3_hwdev *hwdev);
+int hinic3_reinit_cmdq_ctxts(struct hinic3_hwdev *hwdev);
+bool hinic3_cmdq_idle(struct hinic3_cmdq *cmdq);
+
+#endif
diff --git a/drivers/net/ethernet/huawei/hinic3/hinic3_common.c b/drivers/net/ethernet/huawei/hinic3/hinic3_common.c
new file mode 100644
index 000000000000..fe4778d152cf
--- /dev/null
+++ b/drivers/net/ethernet/huawei/hinic3/hinic3_common.c
@@ -0,0 +1,76 @@
+// SPDX-License-Identifier: GPL-2.0
+// Copyright (c) Huawei Technologies Co., Ltd. 2025. All rights reserved.
+
+#include <linux/delay.h>
+#include <linux/dma-mapping.h>
+#include <linux/iopoll.h>
+
+#include "hinic3_common.h"
+
+int hinic3_dma_zalloc_coherent_align(struct device *dev, u32 size, u32 align,
+				     gfp_t flag,
+				     struct hinic3_dma_addr_align *mem_align)
+{
+	dma_addr_t paddr, align_paddr;
+	void *vaddr, *align_vaddr;
+	u32 real_size = size;
+
+	vaddr = dma_alloc_coherent(dev, real_size, &paddr, flag);
+	if (!vaddr)
+		return -ENOMEM;
+
+	align_paddr = ALIGN(paddr, align);
+	if (align_paddr == paddr) {
+		align_vaddr = vaddr;
+		goto out;
+	}
+
+	dma_free_coherent(dev, real_size, vaddr, paddr);
+
+	/* realloc memory for align */
+	real_size = size + align;
+	vaddr = dma_alloc_coherent(dev, real_size, &paddr, flag);
+	if (!vaddr)
+		return -ENOMEM;
+
+	align_paddr = ALIGN(paddr, align);
+	align_vaddr = vaddr + (align_paddr - paddr);
+
+out:
+	mem_align->real_size = real_size;
+	mem_align->ori_vaddr = vaddr;
+	mem_align->ori_paddr = paddr;
+	mem_align->align_vaddr = align_vaddr;
+	mem_align->align_paddr = align_paddr;
+
+	return 0;
+}
+
+void hinic3_dma_free_coherent_align(struct device *dev,
+				    struct hinic3_dma_addr_align *mem_align)
+{
+	dma_free_coherent(dev, mem_align->real_size,
+			  mem_align->ori_vaddr, mem_align->ori_paddr);
+}
+
+int hinic3_wait_for_timeout(void *priv_data, wait_cpl_handler handler,
+			    u32 wait_total_ms, u32 wait_once_us)
+{
+	enum hinic3_wait_return ret;
+	int err;
+
+	err = read_poll_timeout(handler, ret, ret == HINIC3_WAIT_PROCESS_CPL,
+				wait_once_us, wait_total_ms * USEC_PER_MSEC,
+				false, priv_data);
+
+	return err;
+}
+
+/* Data provided to/by cmdq is arranged in structs with little endian fields but
+ * every dword (32bits) should be swapped since HW swaps it again when it
+ * copies it from/to host memory.
+ */
+void hinic3_cmdq_buf_swab32(void *data, int len)
+{
+	swab32_array(data, len / sizeof(u32));
+}
diff --git a/drivers/net/ethernet/huawei/hinic3/hinic3_common.h b/drivers/net/ethernet/huawei/hinic3/hinic3_common.h
new file mode 100644
index 000000000000..a8fabfae90fb
--- /dev/null
+++ b/drivers/net/ethernet/huawei/hinic3/hinic3_common.h
@@ -0,0 +1,54 @@
+/* SPDX-License-Identifier: GPL-2.0 */
+/* Copyright (c) Huawei Technologies Co., Ltd. 2025. All rights reserved. */
+
+#ifndef _HINIC3_COMMON_H_
+#define _HINIC3_COMMON_H_
+
+#include <linux/device.h>
+
+#define HINIC3_MIN_PAGE_SIZE  0x1000
+
+struct hinic3_dma_addr_align {
+	u32        real_size;
+
+	void       *ori_vaddr;
+	dma_addr_t ori_paddr;
+
+	void       *align_vaddr;
+	dma_addr_t align_paddr;
+};
+
+enum hinic3_wait_return {
+	HINIC3_WAIT_PROCESS_CPL     = 0,
+	HINIC3_WAIT_PROCESS_WAITING = 1,
+};
+
+struct hinic3_sge {
+	__le32 hi_addr;
+	__le32 lo_addr;
+	__le32 len;
+	__le32 rsvd;
+};
+
+static inline void hinic3_set_sge(struct hinic3_sge *sge, dma_addr_t addr,
+				  __le32 len)
+{
+	sge->hi_addr = cpu_to_le32(upper_32_bits(addr));
+	sge->lo_addr = cpu_to_le32(lower_32_bits(addr));
+	sge->len = len;
+	sge->rsvd = 0;
+}
+
+int hinic3_dma_zalloc_coherent_align(struct device *dev, u32 size, u32 align,
+				     gfp_t flag,
+				     struct hinic3_dma_addr_align *mem_align);
+void hinic3_dma_free_coherent_align(struct device *dev,
+				    struct hinic3_dma_addr_align *mem_align);
+
+typedef enum hinic3_wait_return (*wait_cpl_handler)(void *priv_data);
+int hinic3_wait_for_timeout(void *priv_data, wait_cpl_handler handler,
+			    u32 wait_total_ms, u32 wait_once_us);
+
+void hinic3_cmdq_buf_swab32(void *data, int len);
+
+#endif
diff --git a/drivers/net/ethernet/huawei/hinic3/hinic3_csr.h b/drivers/net/ethernet/huawei/hinic3/hinic3_csr.h
new file mode 100644
index 000000000000..e7417e8efa99
--- /dev/null
+++ b/drivers/net/ethernet/huawei/hinic3/hinic3_csr.h
@@ -0,0 +1,79 @@
+/* SPDX-License-Identifier: GPL-2.0 */
+/* Copyright (c) Huawei Technologies Co., Ltd. 2025. All rights reserved. */
+
+#ifndef _HINIC3_CSR_H_
+#define _HINIC3_CSR_H_
+
+#define HINIC3_CFG_REGS_FLAG                  0x40000000
+#define HINIC3_REGS_FLAG_MASK                 0x3FFFFFFF
+
+#define HINIC3_VF_CFG_REG_OFFSET              0x2000
+
+/* HW interface registers */
+#define HINIC3_CSR_FUNC_ATTR0_ADDR            (HINIC3_CFG_REGS_FLAG + 0x0)
+#define HINIC3_CSR_FUNC_ATTR1_ADDR            (HINIC3_CFG_REGS_FLAG + 0x4)
+#define HINIC3_CSR_FUNC_ATTR2_ADDR            (HINIC3_CFG_REGS_FLAG + 0x8)
+#define HINIC3_CSR_FUNC_ATTR3_ADDR            (HINIC3_CFG_REGS_FLAG + 0xC)
+#define HINIC3_CSR_FUNC_ATTR4_ADDR            (HINIC3_CFG_REGS_FLAG + 0x10)
+#define HINIC3_CSR_FUNC_ATTR5_ADDR            (HINIC3_CFG_REGS_FLAG + 0x14)
+#define HINIC3_CSR_FUNC_ATTR6_ADDR            (HINIC3_CFG_REGS_FLAG + 0x18)
+
+#define HINIC3_FUNC_CSR_MAILBOX_DATA_OFF      0x80
+#define HINIC3_FUNC_CSR_MAILBOX_CONTROL_OFF   (HINIC3_CFG_REGS_FLAG + 0x0100)
+#define HINIC3_FUNC_CSR_MAILBOX_INT_OFF       (HINIC3_CFG_REGS_FLAG + 0x0104)
+#define HINIC3_FUNC_CSR_MAILBOX_RESULT_H_OFF  (HINIC3_CFG_REGS_FLAG + 0x0108)
+#define HINIC3_FUNC_CSR_MAILBOX_RESULT_L_OFF  (HINIC3_CFG_REGS_FLAG + 0x010C)
+
+#define HINIC3_CSR_DMA_ATTR_TBL_ADDR          (HINIC3_CFG_REGS_FLAG + 0x380)
+#define HINIC3_CSR_DMA_ATTR_INDIR_IDX_ADDR    (HINIC3_CFG_REGS_FLAG + 0x390)
+
+/* MSI-X registers */
+#define HINIC3_CSR_FUNC_MSI_CLR_WR_ADDR       (HINIC3_CFG_REGS_FLAG + 0x58)
+
+#define HINIC3_MSI_CLR_INDIR_RESEND_TIMER_CLR_MASK  BIT(0)
+#define HINIC3_MSI_CLR_INDIR_INT_MSK_SET_MASK       BIT(1)
+#define HINIC3_MSI_CLR_INDIR_INT_MSK_CLR_MASK       BIT(2)
+#define HINIC3_MSI_CLR_INDIR_AUTO_MSK_SET_MASK      BIT(3)
+#define HINIC3_MSI_CLR_INDIR_AUTO_MSK_CLR_MASK      BIT(4)
+#define HINIC3_MSI_CLR_INDIR_SIMPLE_INDIR_IDX_MASK  GENMASK(31, 22)
+#define HINIC3_MSI_CLR_INDIR_SET(val, member)  \
+	FIELD_PREP(HINIC3_MSI_CLR_INDIR_##member##_MASK, val)
+
+/* EQ registers */
+#define HINIC3_AEQ_INDIR_IDX_ADDR      (HINIC3_CFG_REGS_FLAG + 0x210)
+#define HINIC3_CEQ_INDIR_IDX_ADDR      (HINIC3_CFG_REGS_FLAG + 0x290)
+
+#define HINIC3_EQ_INDIR_IDX_ADDR(type)  \
+	((type == HINIC3_AEQ) ? HINIC3_AEQ_INDIR_IDX_ADDR :  \
+	 HINIC3_CEQ_INDIR_IDX_ADDR)
+
+#define HINIC3_AEQ_MTT_OFF_BASE_ADDR   (HINIC3_CFG_REGS_FLAG + 0x240)
+#define HINIC3_CEQ_MTT_OFF_BASE_ADDR   (HINIC3_CFG_REGS_FLAG + 0x2C0)
+
+#define HINIC3_CSR_EQ_PAGE_OFF_STRIDE  8
+
+#define HINIC3_AEQ_HI_PHYS_ADDR_REG(pg_num)  \
+	(HINIC3_AEQ_MTT_OFF_BASE_ADDR + (pg_num) *  \
+	 HINIC3_CSR_EQ_PAGE_OFF_STRIDE)
+
+#define HINIC3_AEQ_LO_PHYS_ADDR_REG(pg_num)  \
+	(HINIC3_AEQ_MTT_OFF_BASE_ADDR + (pg_num) *  \
+	 HINIC3_CSR_EQ_PAGE_OFF_STRIDE + 4)
+
+#define HINIC3_CEQ_HI_PHYS_ADDR_REG(pg_num)  \
+	(HINIC3_CEQ_MTT_OFF_BASE_ADDR + (pg_num) *  \
+	 HINIC3_CSR_EQ_PAGE_OFF_STRIDE)
+
+#define HINIC3_CEQ_LO_PHYS_ADDR_REG(pg_num)  \
+	(HINIC3_CEQ_MTT_OFF_BASE_ADDR + (pg_num) *  \
+	 HINIC3_CSR_EQ_PAGE_OFF_STRIDE + 4)
+
+#define HINIC3_CSR_AEQ_CTRL_0_ADDR           (HINIC3_CFG_REGS_FLAG + 0x200)
+#define HINIC3_CSR_AEQ_CTRL_1_ADDR           (HINIC3_CFG_REGS_FLAG + 0x204)
+#define HINIC3_CSR_AEQ_PROD_IDX_ADDR         (HINIC3_CFG_REGS_FLAG + 0x20C)
+#define HINIC3_CSR_AEQ_CI_SIMPLE_INDIR_ADDR  (HINIC3_CFG_REGS_FLAG + 0x50)
+
+#define HINIC3_CSR_CEQ_PROD_IDX_ADDR         (HINIC3_CFG_REGS_FLAG + 0x28c)
+#define HINIC3_CSR_CEQ_CI_SIMPLE_INDIR_ADDR  (HINIC3_CFG_REGS_FLAG + 0x54)
+
+#endif
diff --git a/drivers/net/ethernet/huawei/hinic3/hinic3_eqs.c b/drivers/net/ethernet/huawei/hinic3/hinic3_eqs.c
new file mode 100644
index 000000000000..01686472985b
--- /dev/null
+++ b/drivers/net/ethernet/huawei/hinic3/hinic3_eqs.c
@@ -0,0 +1,776 @@
+// SPDX-License-Identifier: GPL-2.0
+// Copyright (c) Huawei Technologies Co., Ltd. 2025. All rights reserved.
+
+#include <linux/delay.h>
+
+#include "hinic3_csr.h"
+#include "hinic3_eqs.h"
+#include "hinic3_hwdev.h"
+#include "hinic3_hwif.h"
+#include "hinic3_mbox.h"
+
+#define AEQ_CTRL_0_INTR_IDX_MASK      GENMASK(9, 0)
+#define AEQ_CTRL_0_DMA_ATTR_MASK      GENMASK(17, 12)
+#define AEQ_CTRL_0_PCI_INTF_IDX_MASK  GENMASK(22, 20)
+#define AEQ_CTRL_0_INTR_MODE_MASK     BIT(31)
+#define AEQ_CTRL_0_SET(val, member)  \
+	FIELD_PREP(AEQ_CTRL_0_##member##_MASK, val)
+
+#define AEQ_CTRL_1_LEN_MASK           GENMASK(20, 0)
+#define AEQ_CTRL_1_ELEM_SIZE_MASK     GENMASK(25, 24)
+#define AEQ_CTRL_1_PAGE_SIZE_MASK     GENMASK(31, 28)
+#define AEQ_CTRL_1_SET(val, member)  \
+	FIELD_PREP(AEQ_CTRL_1_##member##_MASK, val)
+
+#define CEQ_CTRL_0_INTR_IDX_MASK      GENMASK(9, 0)
+#define CEQ_CTRL_0_DMA_ATTR_MASK      GENMASK(17, 12)
+#define CEQ_CTRL_0_LIMIT_KICK_MASK    GENMASK(23, 20)
+#define CEQ_CTRL_0_PCI_INTF_IDX_MASK  GENMASK(25, 24)
+#define CEQ_CTRL_0_PAGE_SIZE_MASK     GENMASK(30, 27)
+#define CEQ_CTRL_0_INTR_MODE_MASK     BIT(31)
+#define CEQ_CTRL_0_SET(val, member)  \
+	FIELD_PREP(CEQ_CTRL_0_##member##_MASK, val)
+
+#define CEQ_CTRL_1_LEN_MASK           GENMASK(19, 0)
+#define CEQ_CTRL_1_SET(val, member)  \
+	FIELD_PREP(CEQ_CTRL_1_##member##_MASK, val)
+
+#define CEQE_TYPE_MASK                GENMASK(25, 23)
+#define CEQE_TYPE(type)  \
+	FIELD_GET(CEQE_TYPE_MASK, le32_to_cpu(type))
+
+#define CEQE_DATA_MASK                GENMASK(25, 0)
+#define CEQE_DATA(data)               ((data) & cpu_to_le32(CEQE_DATA_MASK))
+
+#define EQ_ELEM_DESC_TYPE_MASK        GENMASK(6, 0)
+#define EQ_ELEM_DESC_SRC_MASK         BIT(7)
+#define EQ_ELEM_DESC_SIZE_MASK        GENMASK(15, 8)
+#define EQ_ELEM_DESC_WRAPPED_MASK     BIT(31)
+#define EQ_ELEM_DESC_GET(val, member)  \
+	FIELD_GET(EQ_ELEM_DESC_##member##_MASK, le32_to_cpu(val))
+
+#define EQ_CI_SIMPLE_INDIR_CI_MASK       GENMASK(20, 0)
+#define EQ_CI_SIMPLE_INDIR_ARMED_MASK    BIT(21)
+#define EQ_CI_SIMPLE_INDIR_AEQ_IDX_MASK  GENMASK(31, 30)
+#define EQ_CI_SIMPLE_INDIR_CEQ_IDX_MASK  GENMASK(31, 24)
+#define EQ_CI_SIMPLE_INDIR_SET(val, member)  \
+	FIELD_PREP(EQ_CI_SIMPLE_INDIR_##member##_MASK, val)
+
+#define EQ_CI_SIMPLE_INDIR_REG_ADDR(eq)  \
+	(((eq)->type == HINIC3_AEQ) ?  \
+	 HINIC3_CSR_AEQ_CI_SIMPLE_INDIR_ADDR :  \
+	 HINIC3_CSR_CEQ_CI_SIMPLE_INDIR_ADDR)
+
+#define EQ_PROD_IDX_REG_ADDR(eq)  \
+	(((eq)->type == HINIC3_AEQ) ?  \
+	 HINIC3_CSR_AEQ_PROD_IDX_ADDR : HINIC3_CSR_CEQ_PROD_IDX_ADDR)
+
+#define EQ_HI_PHYS_ADDR_REG(type, pg_num)  \
+	(((type) == HINIC3_AEQ) ?  \
+	       HINIC3_AEQ_HI_PHYS_ADDR_REG(pg_num) :  \
+	       HINIC3_CEQ_HI_PHYS_ADDR_REG(pg_num))
+
+#define EQ_LO_PHYS_ADDR_REG(type, pg_num)  \
+	(((type) == HINIC3_AEQ) ?  \
+	       HINIC3_AEQ_LO_PHYS_ADDR_REG(pg_num) :  \
+	       HINIC3_CEQ_LO_PHYS_ADDR_REG(pg_num))
+
+#define EQ_MSIX_RESEND_TIMER_CLEAR  1
+
+#define HINIC3_EQ_MAX_PAGES(eq)  \
+	((eq)->type == HINIC3_AEQ ?  \
+	 HINIC3_AEQ_MAX_PAGES : HINIC3_CEQ_MAX_PAGES)
+
+#define HINIC3_TASK_PROCESS_EQE_LIMIT  1024
+#define HINIC3_EQ_UPDATE_CI_STEP       64
+#define HINIC3_EQS_WQ_NAME             "hinic3_eqs"
+
+#define HINIC3_EQ_VALID_SHIFT          31
+#define HINIC3_EQ_WRAPPED(eq)  \
+	((eq)->wrapped << HINIC3_EQ_VALID_SHIFT)
+
+#define HINIC3_EQ_WRAPPED_SHIFT        20
+#define HINIC3_EQ_CONS_IDX(eq)  \
+	((eq)->cons_idx | ((eq)->wrapped << HINIC3_EQ_WRAPPED_SHIFT))
+
+static const struct hinic3_aeq_elem *get_curr_aeq_elem(const struct hinic3_eq *eq)
+{
+	return get_q_element(&eq->qpages, eq->cons_idx, NULL);
+}
+
+static const __be32 *get_curr_ceq_elem(const struct hinic3_eq *eq)
+{
+	return get_q_element(&eq->qpages, eq->cons_idx, NULL);
+}
+
+int hinic3_aeq_register_cb(struct hinic3_hwdev *hwdev,
+			   enum hinic3_aeq_type event,
+			   hinic3_aeq_event_cb hwe_cb)
+{
+	struct hinic3_aeqs *aeqs;
+
+	aeqs = hwdev->aeqs;
+	aeqs->aeq_cb[event] = hwe_cb;
+	spin_lock_init(&aeqs->aeq_lock);
+
+	return 0;
+}
+
+void hinic3_aeq_unregister_cb(struct hinic3_hwdev *hwdev,
+			      enum hinic3_aeq_type event)
+{
+	struct hinic3_aeqs *aeqs;
+
+	aeqs = hwdev->aeqs;
+
+	spin_lock_bh(&aeqs->aeq_lock);
+	aeqs->aeq_cb[event] = NULL;
+	spin_unlock_bh(&aeqs->aeq_lock);
+}
+
+int hinic3_ceq_register_cb(struct hinic3_hwdev *hwdev,
+			   enum hinic3_ceq_event event,
+			   hinic3_ceq_event_cb callback)
+{
+	struct hinic3_ceqs *ceqs;
+
+	ceqs = hwdev->ceqs;
+	ceqs->ceq_cb[event] = callback;
+	spin_lock_init(&ceqs->ceq_lock);
+
+	return 0;
+}
+
+void hinic3_ceq_unregister_cb(struct hinic3_hwdev *hwdev,
+			      enum hinic3_ceq_event event)
+{
+	struct hinic3_ceqs *ceqs;
+
+	ceqs = hwdev->ceqs;
+
+	spin_lock_bh(&ceqs->ceq_lock);
+	ceqs->ceq_cb[event] = NULL;
+	spin_unlock_bh(&ceqs->ceq_lock);
+}
+
+/* Set consumer index in the hw. */
+static void set_eq_cons_idx(struct hinic3_eq *eq, u32 arm_state)
+{
+	u32 addr = EQ_CI_SIMPLE_INDIR_REG_ADDR(eq);
+	u32 eq_wrap_ci, val;
+
+	eq_wrap_ci = HINIC3_EQ_CONS_IDX(eq);
+	val = EQ_CI_SIMPLE_INDIR_SET(arm_state, ARMED);
+	if (eq->type == HINIC3_AEQ) {
+		val = val |
+			EQ_CI_SIMPLE_INDIR_SET(eq_wrap_ci, CI) |
+			EQ_CI_SIMPLE_INDIR_SET(eq->q_id, AEQ_IDX);
+	} else {
+		val = val |
+			EQ_CI_SIMPLE_INDIR_SET(eq_wrap_ci, CI) |
+			EQ_CI_SIMPLE_INDIR_SET(eq->q_id, CEQ_IDX);
+	}
+
+	hinic3_hwif_write_reg(eq->hwdev->hwif, addr, val);
+}
+
+static struct hinic3_ceqs *ceq_to_ceqs(const struct hinic3_eq *eq)
+{
+	return container_of(eq, struct hinic3_ceqs, ceq[eq->q_id]);
+}
+
+static void ceq_event_handler(struct hinic3_ceqs *ceqs, __le32 ceqe)
+{
+	enum hinic3_ceq_event event = CEQE_TYPE(ceqe);
+	struct hinic3_hwdev *hwdev = ceqs->hwdev;
+	__le32 ceqe_data = CEQE_DATA(ceqe);
+
+	if (event >= HINIC3_MAX_CEQ_EVENTS) {
+		dev_warn(hwdev->dev, "Ceq unknown event:%d, ceqe data: 0x%x\n",
+			 event, ceqe_data);
+		return;
+	}
+
+	spin_lock_bh(&ceqs->ceq_lock);
+	if (ceqs->ceq_cb[event])
+		ceqs->ceq_cb[event](hwdev, ceqe_data);
+
+	spin_unlock_bh(&ceqs->ceq_lock);
+}
+
+static struct hinic3_aeqs *aeq_to_aeqs(const struct hinic3_eq *eq)
+{
+	return container_of(eq, struct hinic3_aeqs, aeq[eq->q_id]);
+}
+
+static void aeq_event_handler(struct hinic3_aeqs *aeqs, __le32 aeqe,
+			      const struct hinic3_aeq_elem *aeqe_pos)
+{
+	struct hinic3_hwdev *hwdev = aeqs->hwdev;
+	u8 data[HINIC3_AEQE_DATA_SIZE], size;
+	enum hinic3_aeq_type event;
+	hinic3_aeq_event_cb hwe_cb;
+
+	if (EQ_ELEM_DESC_GET(aeqe, SRC))
+		return;
+
+	event = EQ_ELEM_DESC_GET(aeqe, TYPE);
+	if (event >= HINIC3_MAX_AEQ_EVENTS) {
+		dev_warn(hwdev->dev, "Aeq unknown event:%d\n", event);
+		return;
+	}
+
+	memcpy(data, aeqe_pos->aeqe_data, HINIC3_AEQE_DATA_SIZE);
+	swab32_array((u32 *)data, HINIC3_AEQE_DATA_SIZE / sizeof(u32));
+	size = EQ_ELEM_DESC_GET(aeqe, SIZE);
+
+	spin_lock_bh(&aeqs->aeq_lock);
+	hwe_cb = aeqs->aeq_cb[event];
+	if (hwe_cb)
+		hwe_cb(aeqs->hwdev, data, size);
+	spin_unlock_bh(&aeqs->aeq_lock);
+}
+
+static int aeq_irq_handler(struct hinic3_eq *eq)
+{
+	const struct hinic3_aeq_elem *aeqe_pos;
+	struct hinic3_aeqs *aeqs;
+	u32 i, eqe_cnt = 0;
+	__le32 aeqe;
+
+	aeqs = aeq_to_aeqs(eq);
+	for (i = 0; i < HINIC3_TASK_PROCESS_EQE_LIMIT; i++) {
+		aeqe_pos = get_curr_aeq_elem(eq);
+		aeqe = (__force __le32)swab32((__force __u32)aeqe_pos->desc);
+		/* HW updates wrapped bit, when it adds eq element event */
+		if (EQ_ELEM_DESC_GET(aeqe, WRAPPED) == eq->wrapped)
+			return 0;
+
+		/* Prevent speculative reads from element */
+		dma_rmb();
+		aeq_event_handler(aeqs, aeqe, aeqe_pos);
+		eq->cons_idx++;
+		if (eq->cons_idx == eq->eq_len) {
+			eq->cons_idx = 0;
+			eq->wrapped = !eq->wrapped;
+		}
+
+		if (++eqe_cnt >= HINIC3_EQ_UPDATE_CI_STEP) {
+			eqe_cnt = 0;
+			set_eq_cons_idx(eq, HINIC3_EQ_NOT_ARMED);
+		}
+	}
+
+	return -EAGAIN;
+}
+
+static int ceq_irq_handler(struct hinic3_eq *eq)
+{
+	struct hinic3_ceqs *ceqs;
+	u32 eqe_cnt = 0;
+	__be32 ceqe_raw;
+	__le32 ceqe;
+	u32 i;
+
+	ceqs = ceq_to_ceqs(eq);
+	for (i = 0; i < HINIC3_TASK_PROCESS_EQE_LIMIT; i++) {
+		ceqe_raw = *get_curr_ceq_elem(eq);
+		ceqe = (__force __le32)swab32((__force __u32)ceqe_raw);
+
+		/* HW updates wrapped bit, when it adds eq element event */
+		if (EQ_ELEM_DESC_GET(ceqe, WRAPPED) == eq->wrapped)
+			return 0;
+
+		ceq_event_handler(ceqs, ceqe);
+		eq->cons_idx++;
+		if (eq->cons_idx == eq->eq_len) {
+			eq->cons_idx = 0;
+			eq->wrapped = !eq->wrapped;
+		}
+
+		if (++eqe_cnt >= HINIC3_EQ_UPDATE_CI_STEP) {
+			eqe_cnt = 0;
+			set_eq_cons_idx(eq, HINIC3_EQ_NOT_ARMED);
+		}
+	}
+
+	return -EAGAIN;
+}
+
+static void reschedule_aeq_handler(struct hinic3_eq *eq)
+{
+	struct hinic3_aeqs *aeqs = aeq_to_aeqs(eq);
+
+	queue_work(aeqs->workq, &eq->aeq_work);
+}
+
+static int eq_irq_handler(struct hinic3_eq *eq)
+{
+	int err;
+
+	if (eq->type == HINIC3_AEQ)
+		err = aeq_irq_handler(eq);
+	else
+		err = ceq_irq_handler(eq);
+
+	set_eq_cons_idx(eq, err ? HINIC3_EQ_NOT_ARMED :
+			HINIC3_EQ_ARMED);
+
+	return err;
+}
+
+static void aeq_irq_work(struct work_struct *work)
+{
+	struct hinic3_eq *eq = container_of(work, struct hinic3_eq, aeq_work);
+	int err;
+
+	err = eq_irq_handler(eq);
+	if (err)
+		reschedule_aeq_handler(eq);
+}
+
+static irqreturn_t aeq_interrupt(int irq, void *data)
+{
+	struct workqueue_struct *workq;
+	struct hinic3_eq *aeq = data;
+	struct hinic3_hwdev *hwdev;
+	struct hinic3_aeqs *aeqs;
+
+	aeqs = aeq_to_aeqs(aeq);
+	hwdev = aeq->hwdev;
+
+	/* clear resend timer cnt register */
+	workq = aeqs->workq;
+	hinic3_msix_intr_clear_resend_bit(hwdev, aeq->msix_entry_idx,
+					  EQ_MSIX_RESEND_TIMER_CLEAR);
+	queue_work(workq, &aeq->aeq_work);
+
+	return IRQ_HANDLED;
+}
+
+static irqreturn_t ceq_interrupt(int irq, void *data)
+{
+	struct hinic3_eq *ceq = data;
+	int err;
+
+	/* clear resend timer counters */
+	hinic3_msix_intr_clear_resend_bit(ceq->hwdev, ceq->msix_entry_idx,
+					  EQ_MSIX_RESEND_TIMER_CLEAR);
+	err = eq_irq_handler(ceq);
+	if (err)
+		return IRQ_NONE;
+
+	return IRQ_HANDLED;
+}
+
+static int hinic3_set_ceq_ctrl_reg(struct hinic3_hwdev *hwdev, u16 q_id,
+				   u32 ctrl0, u32 ctrl1)
+{
+	struct comm_cmd_set_ceq_ctrl_reg ceq_ctrl = {};
+	struct mgmt_msg_params msg_params = {};
+	int err;
+
+	ceq_ctrl.func_id = hinic3_global_func_id(hwdev);
+	ceq_ctrl.q_id = q_id;
+	ceq_ctrl.ctrl0 = ctrl0;
+	ceq_ctrl.ctrl1 = ctrl1;
+
+	mgmt_msg_params_init_default(&msg_params, &ceq_ctrl, sizeof(ceq_ctrl));
+
+	err = hinic3_send_mbox_to_mgmt(hwdev, MGMT_MOD_COMM,
+				       COMM_CMD_SET_CEQ_CTRL_REG, &msg_params);
+	if (err || ceq_ctrl.head.status) {
+		dev_err(hwdev->dev, "Failed to set ceq %u ctrl reg, err: %d status: 0x%x\n",
+			q_id, err, ceq_ctrl.head.status);
+		return -EFAULT;
+	}
+
+	return 0;
+}
+
+static int set_eq_ctrls(struct hinic3_eq *eq)
+{
+	struct hinic3_hwif *hwif = eq->hwdev->hwif;
+	struct hinic3_queue_pages *qpages;
+	u8 pci_intf_idx, elem_size;
+	u32 mask, ctrl0, ctrl1;
+	u32 page_size_val;
+	int err;
+
+	qpages = &eq->qpages;
+	page_size_val = ilog2(qpages->page_size / HINIC3_MIN_PAGE_SIZE);
+	pci_intf_idx = hwif->attr.pci_intf_idx;
+
+	if (eq->type == HINIC3_AEQ) {
+		/* set ctrl0 using read-modify-write */
+		mask = AEQ_CTRL_0_INTR_IDX_MASK |
+		       AEQ_CTRL_0_DMA_ATTR_MASK |
+		       AEQ_CTRL_0_PCI_INTF_IDX_MASK |
+		       AEQ_CTRL_0_INTR_MODE_MASK;
+		ctrl0 = hinic3_hwif_read_reg(hwif, HINIC3_CSR_AEQ_CTRL_0_ADDR);
+		ctrl0 = (ctrl0 & ~mask) |
+			AEQ_CTRL_0_SET(eq->msix_entry_idx, INTR_IDX) |
+			AEQ_CTRL_0_SET(0, DMA_ATTR) |
+			AEQ_CTRL_0_SET(pci_intf_idx, PCI_INTF_IDX) |
+			AEQ_CTRL_0_SET(HINIC3_INTR_MODE_ARMED, INTR_MODE);
+		hinic3_hwif_write_reg(hwif, HINIC3_CSR_AEQ_CTRL_0_ADDR, ctrl0);
+
+		/* HW expects log2(number of 32 byte units). */
+		elem_size = qpages->elem_size_shift - 5;
+		ctrl1 = AEQ_CTRL_1_SET(eq->eq_len, LEN) |
+			AEQ_CTRL_1_SET(elem_size, ELEM_SIZE) |
+			AEQ_CTRL_1_SET(page_size_val, PAGE_SIZE);
+		hinic3_hwif_write_reg(hwif, HINIC3_CSR_AEQ_CTRL_1_ADDR, ctrl1);
+	} else {
+		ctrl0 = CEQ_CTRL_0_SET(eq->msix_entry_idx, INTR_IDX) |
+			CEQ_CTRL_0_SET(0, DMA_ATTR) |
+			CEQ_CTRL_0_SET(0, LIMIT_KICK) |
+			CEQ_CTRL_0_SET(pci_intf_idx, PCI_INTF_IDX) |
+			CEQ_CTRL_0_SET(page_size_val, PAGE_SIZE) |
+			CEQ_CTRL_0_SET(HINIC3_INTR_MODE_ARMED, INTR_MODE);
+
+		ctrl1 = CEQ_CTRL_1_SET(eq->eq_len, LEN);
+
+		/* set ceq ctrl reg through mgmt cpu */
+		err = hinic3_set_ceq_ctrl_reg(eq->hwdev, eq->q_id, ctrl0,
+					      ctrl1);
+		if (err)
+			return err;
+	}
+
+	return 0;
+}
+
+static void ceq_elements_init(struct hinic3_eq *eq, u32 init_val)
+{
+	__be32 *ceqe;
+	u32 i;
+
+	for (i = 0; i < eq->eq_len; i++) {
+		ceqe = get_q_element(&eq->qpages, i, NULL);
+		*ceqe = cpu_to_be32(init_val);
+	}
+
+	wmb();    /* Clear ceq elements bit */
+}
+
+static void aeq_elements_init(struct hinic3_eq *eq, u32 init_val)
+{
+	struct hinic3_aeq_elem *aeqe;
+	u32 i;
+
+	for (i = 0; i < eq->eq_len; i++) {
+		aeqe = get_q_element(&eq->qpages, i, NULL);
+		aeqe->desc = cpu_to_be32(init_val);
+	}
+
+	wmb();    /* Clear aeq elements bit */
+}
+
+static void eq_elements_init(struct hinic3_eq *eq, u32 init_val)
+{
+	if (eq->type == HINIC3_AEQ)
+		aeq_elements_init(eq, init_val);
+	else
+		ceq_elements_init(eq, init_val);
+}
+
+static int alloc_eq_pages(struct hinic3_eq *eq)
+{
+	struct hinic3_hwif *hwif = eq->hwdev->hwif;
+	struct hinic3_queue_pages *qpages;
+	dma_addr_t page_paddr;
+	u32 reg, init_val;
+	u16 pg_idx;
+	int err;
+
+	qpages = &eq->qpages;
+	err = hinic3_queue_pages_alloc(eq->hwdev, qpages, HINIC3_MIN_PAGE_SIZE);
+	if (err)
+		return err;
+
+	for (pg_idx = 0; pg_idx < qpages->num_pages; pg_idx++) {
+		page_paddr = qpages->pages[pg_idx].align_paddr;
+		reg = EQ_HI_PHYS_ADDR_REG(eq->type, pg_idx);
+		hinic3_hwif_write_reg(hwif, reg, upper_32_bits(page_paddr));
+		reg = EQ_LO_PHYS_ADDR_REG(eq->type, pg_idx);
+		hinic3_hwif_write_reg(hwif, reg, lower_32_bits(page_paddr));
+	}
+
+	init_val = HINIC3_EQ_WRAPPED(eq);
+	eq_elements_init(eq, init_val);
+
+	return 0;
+}
+
+static void eq_calc_page_size_and_num(struct hinic3_eq *eq, u32 elem_size)
+{
+	u32 max_pages, min_page_size, page_size, total_size;
+
+	/* No need for complicated arithmetic. All values must be power of 2.
+	 * Multiplications give power of 2 and divisions give power of 2 without
+	 * remainder.
+	 */
+	max_pages = HINIC3_EQ_MAX_PAGES(eq);
+	min_page_size = HINIC3_MIN_PAGE_SIZE;
+	total_size = eq->eq_len * elem_size;
+
+	if (total_size <= max_pages * min_page_size)
+		page_size = min_page_size;
+	else
+		page_size = total_size / max_pages;
+
+	hinic3_queue_pages_init(&eq->qpages, eq->eq_len, page_size, elem_size);
+}
+
+static int request_eq_irq(struct hinic3_eq *eq)
+{
+	int err;
+
+	if (eq->type == HINIC3_AEQ) {
+		INIT_WORK(&eq->aeq_work, aeq_irq_work);
+		snprintf(eq->irq_name, sizeof(eq->irq_name),
+			 "hinic3_aeq%u@pci:%s", eq->q_id,
+			 pci_name(eq->hwdev->pdev));
+		err = request_irq(eq->irq_id, aeq_interrupt, 0,
+				  eq->irq_name, eq);
+	} else {
+		snprintf(eq->irq_name, sizeof(eq->irq_name),
+			 "hinic3_ceq%u@pci:%s", eq->q_id,
+			 pci_name(eq->hwdev->pdev));
+		err = request_threaded_irq(eq->irq_id, NULL, ceq_interrupt,
+					   IRQF_ONESHOT, eq->irq_name, eq);
+	}
+
+	return err;
+}
+
+static void reset_eq(struct hinic3_eq *eq)
+{
+	/* clear eq_len to force eqe drop in hardware */
+	if (eq->type == HINIC3_AEQ)
+		hinic3_hwif_write_reg(eq->hwdev->hwif,
+				      HINIC3_CSR_AEQ_CTRL_1_ADDR, 0);
+	else
+		hinic3_set_ceq_ctrl_reg(eq->hwdev, eq->q_id, 0, 0);
+
+	hinic3_hwif_write_reg(eq->hwdev->hwif, EQ_PROD_IDX_REG_ADDR(eq), 0);
+}
+
+static int init_eq(struct hinic3_eq *eq, struct hinic3_hwdev *hwdev, u16 q_id,
+		   u32 q_len, enum hinic3_eq_type type,
+		   struct msix_entry *msix_entry)
+{
+	u32 elem_size;
+	int err;
+
+	eq->hwdev = hwdev;
+	eq->q_id = q_id;
+	eq->type = type;
+	eq->eq_len = q_len;
+
+	/* Indirect access should set q_id first */
+	hinic3_hwif_write_reg(hwdev->hwif, HINIC3_EQ_INDIR_IDX_ADDR(eq->type),
+			      eq->q_id);
+
+	reset_eq(eq);
+
+	eq->cons_idx = 0;
+	eq->wrapped = 0;
+
+	elem_size = (type == HINIC3_AEQ) ? HINIC3_AEQE_SIZE : HINIC3_CEQE_SIZE;
+	eq_calc_page_size_and_num(eq, elem_size);
+
+	err = alloc_eq_pages(eq);
+	if (err) {
+		dev_err(hwdev->dev, "Failed to allocate pages for eq\n");
+		return err;
+	}
+
+	eq->msix_entry_idx = msix_entry->entry;
+	eq->irq_id = msix_entry->vector;
+
+	err = set_eq_ctrls(eq);
+	if (err) {
+		dev_err(hwdev->dev, "Failed to set ctrls for eq\n");
+		goto err_free_queue_pages;
+	}
+
+	set_eq_cons_idx(eq, HINIC3_EQ_ARMED);
+
+	err = request_eq_irq(eq);
+	if (err) {
+		dev_err(hwdev->dev,
+			"Failed to request irq for the eq, err: %d\n", err);
+		goto err_free_queue_pages;
+	}
+
+	hinic3_set_msix_state(hwdev, eq->msix_entry_idx, HINIC3_MSIX_DISABLE);
+
+	return 0;
+
+err_free_queue_pages:
+	hinic3_queue_pages_free(hwdev, &eq->qpages);
+
+	return err;
+}
+
+static void remove_eq(struct hinic3_eq *eq)
+{
+	hinic3_set_msix_state(eq->hwdev, eq->msix_entry_idx,
+			      HINIC3_MSIX_DISABLE);
+	free_irq(eq->irq_id, eq);
+	/* Indirect access should set q_id first */
+	hinic3_hwif_write_reg(eq->hwdev->hwif,
+			      HINIC3_EQ_INDIR_IDX_ADDR(eq->type),
+			      eq->q_id);
+
+	if (eq->type == HINIC3_AEQ) {
+		disable_work_sync(&eq->aeq_work);
+		/* clear eq_len to avoid hw access host memory */
+		hinic3_hwif_write_reg(eq->hwdev->hwif,
+				      HINIC3_CSR_AEQ_CTRL_1_ADDR, 0);
+	} else {
+		hinic3_set_ceq_ctrl_reg(eq->hwdev, eq->q_id, 0, 0);
+	}
+
+	/* update consumer index to avoid invalid interrupt */
+	eq->cons_idx = hinic3_hwif_read_reg(eq->hwdev->hwif,
+					    EQ_PROD_IDX_REG_ADDR(eq));
+	set_eq_cons_idx(eq, HINIC3_EQ_NOT_ARMED);
+	hinic3_queue_pages_free(eq->hwdev, &eq->qpages);
+}
+
+int hinic3_aeqs_init(struct hinic3_hwdev *hwdev, u16 num_aeqs,
+		     struct msix_entry *msix_entries)
+{
+	struct hinic3_aeqs *aeqs;
+	u16 q_id;
+	int err;
+
+	aeqs = kzalloc(sizeof(*aeqs), GFP_KERNEL);
+	if (!aeqs)
+		return -ENOMEM;
+
+	hwdev->aeqs = aeqs;
+	aeqs->hwdev = hwdev;
+	aeqs->num_aeqs = num_aeqs;
+	aeqs->workq = alloc_workqueue(HINIC3_EQS_WQ_NAME, WQ_MEM_RECLAIM,
+				      HINIC3_MAX_AEQS);
+	if (!aeqs->workq) {
+		dev_err(hwdev->dev, "Failed to initialize aeq workqueue\n");
+		err = -ENOMEM;
+		goto err_free_aeqs;
+	}
+
+	for (q_id = 0; q_id < num_aeqs; q_id++) {
+		err = init_eq(&aeqs->aeq[q_id], hwdev, q_id,
+			      HINIC3_DEFAULT_AEQ_LEN, HINIC3_AEQ,
+			      &msix_entries[q_id]);
+		if (err) {
+			dev_err(hwdev->dev, "Failed to init aeq %u\n",
+				q_id);
+			goto err_remove_eqs;
+		}
+	}
+	for (q_id = 0; q_id < num_aeqs; q_id++)
+		hinic3_set_msix_state(hwdev, aeqs->aeq[q_id].msix_entry_idx,
+				      HINIC3_MSIX_ENABLE);
+
+	return 0;
+
+err_remove_eqs:
+	while (q_id > 0) {
+		q_id--;
+		remove_eq(&aeqs->aeq[q_id]);
+	}
+
+	destroy_workqueue(aeqs->workq);
+
+err_free_aeqs:
+	kfree(aeqs);
+
+	return err;
+}
+
+void hinic3_aeqs_free(struct hinic3_hwdev *hwdev)
+{
+	struct hinic3_aeqs *aeqs = hwdev->aeqs;
+	enum hinic3_aeq_type aeq_event;
+	struct hinic3_eq *eq;
+	u16 q_id;
+
+	for (q_id = 0; q_id < aeqs->num_aeqs; q_id++) {
+		eq = aeqs->aeq + q_id;
+		remove_eq(eq);
+		hinic3_free_irq(hwdev, eq->irq_id);
+	}
+
+	for (aeq_event = 0; aeq_event < HINIC3_MAX_AEQ_EVENTS; aeq_event++)
+		hinic3_aeq_unregister_cb(hwdev, aeq_event);
+
+	destroy_workqueue(aeqs->workq);
+
+	kfree(aeqs);
+}
+
+int hinic3_ceqs_init(struct hinic3_hwdev *hwdev, u16 num_ceqs,
+		     struct msix_entry *msix_entries)
+{
+	struct hinic3_ceqs *ceqs;
+	u16 q_id;
+	int err;
+
+	ceqs = kzalloc(sizeof(*ceqs), GFP_KERNEL);
+	if (!ceqs)
+		return -ENOMEM;
+
+	hwdev->ceqs = ceqs;
+	ceqs->hwdev = hwdev;
+	ceqs->num_ceqs = num_ceqs;
+
+	for (q_id = 0; q_id < num_ceqs; q_id++) {
+		err = init_eq(&ceqs->ceq[q_id], hwdev, q_id,
+			      HINIC3_DEFAULT_CEQ_LEN, HINIC3_CEQ,
+			      &msix_entries[q_id]);
+		if (err) {
+			dev_err(hwdev->dev, "Failed to init ceq %u\n",
+				q_id);
+			goto err_free_ceqs;
+		}
+	}
+	for (q_id = 0; q_id < num_ceqs; q_id++)
+		hinic3_set_msix_state(hwdev, ceqs->ceq[q_id].msix_entry_idx,
+				      HINIC3_MSIX_ENABLE);
+
+	return 0;
+
+err_free_ceqs:
+	while (q_id > 0) {
+		q_id--;
+		remove_eq(&ceqs->ceq[q_id]);
+	}
+
+	kfree(ceqs);
+
+	return err;
+}
+
+void hinic3_ceqs_free(struct hinic3_hwdev *hwdev)
+{
+	struct hinic3_ceqs *ceqs = hwdev->ceqs;
+	enum hinic3_ceq_event ceq_event;
+	struct hinic3_eq *eq;
+	u16 q_id;
+
+	for (q_id = 0; q_id < ceqs->num_ceqs; q_id++) {
+		eq = ceqs->ceq + q_id;
+		remove_eq(eq);
+		hinic3_free_irq(hwdev, eq->irq_id);
+	}
+
+	for (ceq_event = 0; ceq_event < HINIC3_MAX_CEQ_EVENTS; ceq_event++)
+		hinic3_ceq_unregister_cb(hwdev, ceq_event);
+
+	kfree(ceqs);
+}
diff --git a/drivers/net/ethernet/huawei/hinic3/hinic3_eqs.h b/drivers/net/ethernet/huawei/hinic3/hinic3_eqs.h
new file mode 100644
index 000000000000..005a6e0745b3
--- /dev/null
+++ b/drivers/net/ethernet/huawei/hinic3/hinic3_eqs.h
@@ -0,0 +1,122 @@
+/* SPDX-License-Identifier: GPL-2.0 */
+/* Copyright (c) Huawei Technologies Co., Ltd. 2025. All rights reserved. */
+
+#ifndef _HINIC3_EQS_H_
+#define _HINIC3_EQS_H_
+
+#include <linux/interrupt.h>
+
+#include "hinic3_hw_cfg.h"
+#include "hinic3_queue_common.h"
+
+#define HINIC3_MAX_AEQS              4
+#define HINIC3_MAX_CEQS              32
+
+#define HINIC3_AEQ_MAX_PAGES         4
+#define HINIC3_CEQ_MAX_PAGES         8
+
+#define HINIC3_AEQE_SIZE             64
+#define HINIC3_CEQE_SIZE             4
+
+#define HINIC3_AEQE_DESC_SIZE        4
+#define HINIC3_AEQE_DATA_SIZE        (HINIC3_AEQE_SIZE - HINIC3_AEQE_DESC_SIZE)
+
+#define HINIC3_DEFAULT_AEQ_LEN       0x10000
+#define HINIC3_DEFAULT_CEQ_LEN       0x10000
+
+#define HINIC3_EQ_IRQ_NAME_LEN       64
+
+#define HINIC3_EQ_USLEEP_LOW_BOUND   900
+#define HINIC3_EQ_USLEEP_HIGH_BOUND  1000
+
+enum hinic3_eq_type {
+	HINIC3_AEQ = 0,
+	HINIC3_CEQ = 1,
+};
+
+enum hinic3_eq_intr_mode {
+	HINIC3_INTR_MODE_ARMED  = 0,
+	HINIC3_INTR_MODE_ALWAYS = 1,
+};
+
+enum hinic3_eq_ci_arm_state {
+	HINIC3_EQ_NOT_ARMED = 0,
+	HINIC3_EQ_ARMED     = 1,
+};
+
+struct hinic3_eq {
+	struct hinic3_hwdev       *hwdev;
+	struct hinic3_queue_pages qpages;
+	u16                       q_id;
+	enum hinic3_eq_type       type;
+	u32                       eq_len;
+	u32                       cons_idx;
+	u8                        wrapped;
+	u32                       irq_id;
+	u16                       msix_entry_idx;
+	char                      irq_name[HINIC3_EQ_IRQ_NAME_LEN];
+	struct work_struct        aeq_work;
+};
+
+struct hinic3_aeq_elem {
+	u8     aeqe_data[HINIC3_AEQE_DATA_SIZE];
+	__be32 desc;
+};
+
+enum hinic3_aeq_type {
+	HINIC3_HW_INTER_INT   = 0,
+	HINIC3_MBX_FROM_FUNC  = 1,
+	HINIC3_MSG_FROM_FW    = 2,
+	HINIC3_MAX_AEQ_EVENTS = 6,
+};
+
+typedef void (*hinic3_aeq_event_cb)(struct hinic3_hwdev *hwdev, u8 *data,
+				    u8 size);
+
+struct hinic3_aeqs {
+	struct hinic3_hwdev     *hwdev;
+	hinic3_aeq_event_cb     aeq_cb[HINIC3_MAX_AEQ_EVENTS];
+	struct hinic3_eq        aeq[HINIC3_MAX_AEQS];
+	u16                     num_aeqs;
+	struct workqueue_struct *workq;
+	/* lock for aeq event flag */
+	spinlock_t              aeq_lock;
+};
+
+enum hinic3_ceq_event {
+	HINIC3_CMDQ           = 3,
+	HINIC3_MAX_CEQ_EVENTS = 6,
+};
+
+typedef void (*hinic3_ceq_event_cb)(struct hinic3_hwdev *hwdev,
+				    __le32 ceqe_data);
+
+struct hinic3_ceqs {
+	struct hinic3_hwdev *hwdev;
+
+	hinic3_ceq_event_cb ceq_cb[HINIC3_MAX_CEQ_EVENTS];
+
+	struct hinic3_eq    ceq[HINIC3_MAX_CEQS];
+	u16                 num_ceqs;
+	/* lock for ceq event flag */
+	spinlock_t          ceq_lock;
+};
+
+int hinic3_aeqs_init(struct hinic3_hwdev *hwdev, u16 num_aeqs,
+		     struct msix_entry *msix_entries);
+void hinic3_aeqs_free(struct hinic3_hwdev *hwdev);
+int hinic3_aeq_register_cb(struct hinic3_hwdev *hwdev,
+			   enum hinic3_aeq_type event,
+			   hinic3_aeq_event_cb hwe_cb);
+void hinic3_aeq_unregister_cb(struct hinic3_hwdev *hwdev,
+			      enum hinic3_aeq_type event);
+int hinic3_ceqs_init(struct hinic3_hwdev *hwdev, u16 num_ceqs,
+		     struct msix_entry *msix_entries);
+void hinic3_ceqs_free(struct hinic3_hwdev *hwdev);
+int hinic3_ceq_register_cb(struct hinic3_hwdev *hwdev,
+			   enum hinic3_ceq_event event,
+			   hinic3_ceq_event_cb callback);
+void hinic3_ceq_unregister_cb(struct hinic3_hwdev *hwdev,
+			      enum hinic3_ceq_event event);
+
+#endif
diff --git a/drivers/net/ethernet/huawei/hinic3/hinic3_hw_cfg.c b/drivers/net/ethernet/huawei/hinic3/hinic3_hw_cfg.c
new file mode 100644
index 000000000000..7827c1f626db
--- /dev/null
+++ b/drivers/net/ethernet/huawei/hinic3/hinic3_hw_cfg.c
@@ -0,0 +1,236 @@
+// SPDX-License-Identifier: GPL-2.0
+// Copyright (c) Huawei Technologies Co., Ltd. 2025. All rights reserved.
+
+#include <linux/device.h>
+
+#include "hinic3_hw_cfg.h"
+#include "hinic3_hwdev.h"
+#include "hinic3_hwif.h"
+#include "hinic3_mbox.h"
+
+#define HINIC3_CFG_MAX_QP  256
+
+static void hinic3_parse_pub_res_cap(struct hinic3_hwdev *hwdev,
+				     struct hinic3_dev_cap *cap,
+				     const struct cfg_cmd_dev_cap *dev_cap,
+				     enum hinic3_func_type type)
+{
+	cap->port_id = dev_cap->port_id;
+	cap->supp_svcs_bitmap = dev_cap->svc_cap_en;
+}
+
+static void hinic3_parse_l2nic_res_cap(struct hinic3_hwdev *hwdev,
+				       struct hinic3_dev_cap *cap,
+				       const struct cfg_cmd_dev_cap *dev_cap,
+				       enum hinic3_func_type type)
+{
+	struct hinic3_nic_service_cap *nic_svc_cap = &cap->nic_svc_cap;
+
+	nic_svc_cap->max_sqs = min(dev_cap->nic_max_sq_id + 1,
+				   HINIC3_CFG_MAX_QP);
+}
+
+static void hinic3_parse_dev_cap(struct hinic3_hwdev *hwdev,
+				 const struct cfg_cmd_dev_cap *dev_cap,
+				 enum hinic3_func_type type)
+{
+	struct hinic3_dev_cap *cap = &hwdev->cfg_mgmt->cap;
+
+	/* Public resource */
+	hinic3_parse_pub_res_cap(hwdev, cap, dev_cap, type);
+
+	/* L2 NIC resource */
+	if (hinic3_support_nic(hwdev))
+		hinic3_parse_l2nic_res_cap(hwdev, cap, dev_cap, type);
+}
+
+static int get_cap_from_fw(struct hinic3_hwdev *hwdev,
+			   enum hinic3_func_type type)
+{
+	struct mgmt_msg_params msg_params = {};
+	struct cfg_cmd_dev_cap dev_cap = {};
+	int err;
+
+	dev_cap.func_id = hinic3_global_func_id(hwdev);
+
+	mgmt_msg_params_init_default(&msg_params, &dev_cap, sizeof(dev_cap));
+
+	err = hinic3_send_mbox_to_mgmt(hwdev, MGMT_MOD_CFGM,
+				       CFG_CMD_GET_DEV_CAP, &msg_params);
+	if (err || dev_cap.head.status) {
+		dev_err(hwdev->dev,
+			"Failed to get capability from FW, err: %d, status: 0x%x\n",
+			err, dev_cap.head.status);
+		return -EIO;
+	}
+
+	hinic3_parse_dev_cap(hwdev, &dev_cap, type);
+
+	return 0;
+}
+
+static int hinic3_init_irq_info(struct hinic3_hwdev *hwdev)
+{
+	struct hinic3_cfg_mgmt_info *cfg_mgmt = hwdev->cfg_mgmt;
+	struct hinic3_hwif *hwif = hwdev->hwif;
+	u16 intr_num = hwif->attr.num_irqs;
+	struct hinic3_irq_info *irq_info;
+	u16 intr_needed;
+
+	intr_needed = hwif->attr.msix_flex_en ? (hwif->attr.num_aeqs +
+		      hwif->attr.num_ceqs + hwif->attr.num_sq) : intr_num;
+	if (intr_needed > intr_num) {
+		dev_warn(hwdev->dev, "Irq num cfg %d is less than the needed irq num %d msix_flex_en %d\n",
+			 intr_num, intr_needed, hwdev->hwif->attr.msix_flex_en);
+		intr_needed = intr_num;
+	}
+
+	irq_info = &cfg_mgmt->irq_info;
+	irq_info->irq = kcalloc(intr_num, sizeof(struct hinic3_irq),
+				GFP_KERNEL);
+	if (!irq_info->irq)
+		return -ENOMEM;
+
+	irq_info->num_irq_hw = intr_needed;
+	mutex_init(&irq_info->irq_mutex);
+
+	return 0;
+}
+
+static int hinic3_init_irq_alloc_info(struct hinic3_hwdev *hwdev)
+{
+	struct hinic3_cfg_mgmt_info *cfg_mgmt = hwdev->cfg_mgmt;
+	struct hinic3_irq *irq = cfg_mgmt->irq_info.irq;
+	u16 nreq = cfg_mgmt->irq_info.num_irq_hw;
+	struct pci_dev *pdev = hwdev->pdev;
+	int actual_irq;
+	u16 i;
+
+	actual_irq = pci_alloc_irq_vectors(pdev, 2, nreq, PCI_IRQ_MSIX);
+	if (actual_irq < 0) {
+		dev_err(hwdev->dev, "Alloc msix entries with threshold 2 failed. actual_irq: %d\n",
+			actual_irq);
+		return -ENOMEM;
+	}
+
+	nreq = actual_irq;
+	cfg_mgmt->irq_info.num_irq = nreq;
+
+	for (i = 0; i < nreq; ++i) {
+		irq[i].msix_entry_idx = i;
+		irq[i].irq_id = pci_irq_vector(pdev, i);
+		irq[i].allocated = false;
+	}
+
+	return 0;
+}
+
+int hinic3_init_cfg_mgmt(struct hinic3_hwdev *hwdev)
+{
+	struct hinic3_cfg_mgmt_info *cfg_mgmt;
+	int err;
+
+	cfg_mgmt = kzalloc(sizeof(*cfg_mgmt), GFP_KERNEL);
+	if (!cfg_mgmt)
+		return -ENOMEM;
+
+	hwdev->cfg_mgmt = cfg_mgmt;
+
+	err = hinic3_init_irq_info(hwdev);
+	if (err) {
+		dev_err(hwdev->dev, "Failed to init hinic3_irq_mgmt_info, err: %d\n",
+			err);
+		goto err_free_cfg_mgmt;
+	}
+
+	err = hinic3_init_irq_alloc_info(hwdev);
+	if (err) {
+		dev_err(hwdev->dev, "Failed to init hinic3_irq_info, err: %d\n",
+			err);
+		goto err_free_irq_info;
+	}
+
+	return 0;
+
+err_free_irq_info:
+	kfree(cfg_mgmt->irq_info.irq);
+	cfg_mgmt->irq_info.irq = NULL;
+err_free_cfg_mgmt:
+	kfree(cfg_mgmt);
+
+	return err;
+}
+
+void hinic3_free_cfg_mgmt(struct hinic3_hwdev *hwdev)
+{
+	struct hinic3_cfg_mgmt_info *cfg_mgmt = hwdev->cfg_mgmt;
+
+	pci_free_irq_vectors(hwdev->pdev);
+	kfree(cfg_mgmt->irq_info.irq);
+	cfg_mgmt->irq_info.irq = NULL;
+	kfree(cfg_mgmt);
+}
+
+int hinic3_alloc_irqs(struct hinic3_hwdev *hwdev, u16 num,
+		      struct msix_entry *alloc_arr, u16 *act_num)
+{
+	struct hinic3_irq_info *irq_info;
+	struct hinic3_irq *curr;
+	u16 i, found = 0;
+
+	irq_info = &hwdev->cfg_mgmt->irq_info;
+	mutex_lock(&irq_info->irq_mutex);
+	for (i = 0; i < irq_info->num_irq && found < num; i++) {
+		curr = irq_info->irq + i;
+		if (curr->allocated)
+			continue;
+		curr->allocated = true;
+		alloc_arr[found].vector = curr->irq_id;
+		alloc_arr[found].entry = curr->msix_entry_idx;
+		found++;
+	}
+	mutex_unlock(&irq_info->irq_mutex);
+
+	*act_num = found;
+
+	return found == 0 ? -ENOMEM : 0;
+}
+
+void hinic3_free_irq(struct hinic3_hwdev *hwdev, u32 irq_id)
+{
+	struct hinic3_irq_info *irq_info;
+	struct hinic3_irq *curr;
+	u16 i;
+
+	irq_info = &hwdev->cfg_mgmt->irq_info;
+	mutex_lock(&irq_info->irq_mutex);
+	for (i = 0; i < irq_info->num_irq; i++) {
+		curr = irq_info->irq + i;
+		if (curr->irq_id == irq_id) {
+			curr->allocated = false;
+			break;
+		}
+	}
+	mutex_unlock(&irq_info->irq_mutex);
+}
+
+int hinic3_init_capability(struct hinic3_hwdev *hwdev)
+{
+	return get_cap_from_fw(hwdev, HINIC3_FUNC_TYPE_VF);
+}
+
+bool hinic3_support_nic(struct hinic3_hwdev *hwdev)
+{
+	return hwdev->cfg_mgmt->cap.supp_svcs_bitmap &
+	       BIT(HINIC3_SERVICE_T_NIC);
+}
+
+u16 hinic3_func_max_qnum(struct hinic3_hwdev *hwdev)
+{
+	return hwdev->cfg_mgmt->cap.nic_svc_cap.max_sqs;
+}
+
+u8 hinic3_physical_port_id(struct hinic3_hwdev *hwdev)
+{
+	return hwdev->cfg_mgmt->cap.port_id;
+}
diff --git a/drivers/net/ethernet/huawei/hinic3/hinic3_hw_cfg.h b/drivers/net/ethernet/huawei/hinic3/hinic3_hw_cfg.h
new file mode 100644
index 000000000000..58806199bf54
--- /dev/null
+++ b/drivers/net/ethernet/huawei/hinic3/hinic3_hw_cfg.h
@@ -0,0 +1,57 @@
+/* SPDX-License-Identifier: GPL-2.0 */
+/* Copyright (c) Huawei Technologies Co., Ltd. 2025. All rights reserved. */
+
+#ifndef _HINIC3_HW_CFG_H_
+#define _HINIC3_HW_CFG_H_
+
+#include <linux/mutex.h>
+#include <linux/pci.h>
+
+struct hinic3_hwdev;
+
+struct hinic3_irq {
+	u32  irq_id;
+	u16  msix_entry_idx;
+	bool allocated;
+};
+
+struct hinic3_irq_info {
+	struct hinic3_irq *irq;
+	u16               num_irq;
+	/* device max irq number */
+	u16               num_irq_hw;
+	/* protect irq alloc and free */
+	struct mutex      irq_mutex;
+};
+
+struct hinic3_nic_service_cap {
+	u16 max_sqs;
+};
+
+/* Device capabilities */
+struct hinic3_dev_cap {
+	/* Bitmasks of services supported by device */
+	u16                           supp_svcs_bitmap;
+	/* Physical port */
+	u8                            port_id;
+	struct hinic3_nic_service_cap nic_svc_cap;
+};
+
+struct hinic3_cfg_mgmt_info {
+	struct hinic3_irq_info irq_info;
+	struct hinic3_dev_cap  cap;
+};
+
+int hinic3_init_cfg_mgmt(struct hinic3_hwdev *hwdev);
+void hinic3_free_cfg_mgmt(struct hinic3_hwdev *hwdev);
+
+int hinic3_alloc_irqs(struct hinic3_hwdev *hwdev, u16 num,
+		      struct msix_entry *alloc_arr, u16 *act_num);
+void hinic3_free_irq(struct hinic3_hwdev *hwdev, u32 irq_id);
+
+int hinic3_init_capability(struct hinic3_hwdev *hwdev);
+bool hinic3_support_nic(struct hinic3_hwdev *hwdev);
+u16 hinic3_func_max_qnum(struct hinic3_hwdev *hwdev);
+u8 hinic3_physical_port_id(struct hinic3_hwdev *hwdev);
+
+#endif
diff --git a/drivers/net/ethernet/huawei/hinic3/hinic3_hw_comm.c b/drivers/net/ethernet/huawei/hinic3/hinic3_hw_comm.c
new file mode 100644
index 000000000000..89638813df40
--- /dev/null
+++ b/drivers/net/ethernet/huawei/hinic3/hinic3_hw_comm.c
@@ -0,0 +1,426 @@
+// SPDX-License-Identifier: GPL-2.0
+// Copyright (c) Huawei Technologies Co., Ltd. 2025. All rights reserved.
+
+#include <linux/delay.h>
+
+#include "hinic3_cmdq.h"
+#include "hinic3_hw_comm.h"
+#include "hinic3_hwdev.h"
+#include "hinic3_hwif.h"
+#include "hinic3_mbox.h"
+
+int hinic3_set_interrupt_cfg_direct(struct hinic3_hwdev *hwdev,
+				    const struct hinic3_interrupt_info *info)
+{
+	struct comm_cmd_cfg_msix_ctrl_reg msix_cfg = {};
+	struct mgmt_msg_params msg_params = {};
+	int err;
+
+	msix_cfg.func_id = hinic3_global_func_id(hwdev);
+	msix_cfg.msix_index = info->msix_index;
+	msix_cfg.opcode = MGMT_MSG_CMD_OP_SET;
+
+	msix_cfg.lli_credit_cnt = info->lli_credit_limit;
+	msix_cfg.lli_timer_cnt = info->lli_timer_cfg;
+	msix_cfg.pending_cnt = info->pending_limit;
+	msix_cfg.coalesce_timer_cnt = info->coalesc_timer_cfg;
+	msix_cfg.resend_timer_cnt = info->resend_timer_cfg;
+
+	mgmt_msg_params_init_default(&msg_params, &msix_cfg, sizeof(msix_cfg));
+
+	err = hinic3_send_mbox_to_mgmt(hwdev, MGMT_MOD_COMM,
+				       COMM_CMD_CFG_MSIX_CTRL_REG, &msg_params);
+	if (err || msix_cfg.head.status) {
+		dev_err(hwdev->dev,
+			"Failed to set interrupt config, err: %d, status: 0x%x\n",
+			err, msix_cfg.head.status);
+		return -EINVAL;
+	}
+
+	return 0;
+}
+
+int hinic3_func_reset(struct hinic3_hwdev *hwdev, u16 func_id, u64 reset_flag)
+{
+	struct comm_cmd_func_reset func_reset = {};
+	struct mgmt_msg_params msg_params = {};
+	int err;
+
+	func_reset.func_id = func_id;
+	func_reset.reset_flag = reset_flag;
+
+	mgmt_msg_params_init_default(&msg_params, &func_reset,
+				     sizeof(func_reset));
+
+	err = hinic3_send_mbox_to_mgmt(hwdev, MGMT_MOD_COMM,
+				       COMM_CMD_FUNC_RESET, &msg_params);
+	if (err || func_reset.head.status) {
+		dev_err(hwdev->dev, "Failed to reset func resources, reset_flag 0x%llx, err: %d, status: 0x%x\n",
+			reset_flag, err, func_reset.head.status);
+		return -EIO;
+	}
+
+	return 0;
+}
+
+static int hinic3_comm_features_nego(struct hinic3_hwdev *hwdev, u8 opcode,
+				     u64 *s_feature, u16 size)
+{
+	struct comm_cmd_feature_nego feature_nego = {};
+	struct mgmt_msg_params msg_params = {};
+	int err;
+
+	feature_nego.func_id = hinic3_global_func_id(hwdev);
+	feature_nego.opcode = opcode;
+	if (opcode == MGMT_MSG_CMD_OP_SET)
+		memcpy(feature_nego.s_feature, s_feature,
+		       array_size(size, sizeof(u64)));
+
+	mgmt_msg_params_init_default(&msg_params, &feature_nego,
+				     sizeof(feature_nego));
+
+	err = hinic3_send_mbox_to_mgmt(hwdev, MGMT_MOD_COMM,
+				       COMM_CMD_FEATURE_NEGO, &msg_params);
+	if (err || feature_nego.head.status) {
+		dev_err(hwdev->dev, "Failed to negotiate feature, err: %d, status: 0x%x\n",
+			err, feature_nego.head.status);
+		return -EINVAL;
+	}
+
+	if (opcode == MGMT_MSG_CMD_OP_GET)
+		memcpy(s_feature, feature_nego.s_feature,
+		       array_size(size, sizeof(u64)));
+
+	return 0;
+}
+
+int hinic3_get_comm_features(struct hinic3_hwdev *hwdev, u64 *s_feature,
+			     u16 size)
+{
+	return hinic3_comm_features_nego(hwdev, MGMT_MSG_CMD_OP_GET, s_feature,
+					 size);
+}
+
+int hinic3_set_comm_features(struct hinic3_hwdev *hwdev, u64 *s_feature,
+			     u16 size)
+{
+	return hinic3_comm_features_nego(hwdev, MGMT_MSG_CMD_OP_SET, s_feature,
+					 size);
+}
+
+int hinic3_get_global_attr(struct hinic3_hwdev *hwdev,
+			   struct comm_global_attr *attr)
+{
+	struct comm_cmd_get_glb_attr get_attr = {};
+	struct mgmt_msg_params msg_params = {};
+	int err;
+
+	mgmt_msg_params_init_default(&msg_params, &get_attr, sizeof(get_attr));
+
+	err = hinic3_send_mbox_to_mgmt(hwdev, MGMT_MOD_COMM,
+				       COMM_CMD_GET_GLOBAL_ATTR, &msg_params);
+	if (err || get_attr.head.status) {
+		dev_err(hwdev->dev,
+			"Failed to get global attribute, err: %d, status: 0x%x\n",
+			err, get_attr.head.status);
+		return -EIO;
+	}
+
+	memcpy(attr, &get_attr.attr, sizeof(*attr));
+
+	return 0;
+}
+
+int hinic3_set_func_svc_used_state(struct hinic3_hwdev *hwdev, u16 svc_type,
+				   u8 state)
+{
+	struct comm_cmd_set_func_svc_used_state used_state = {};
+	struct mgmt_msg_params msg_params = {};
+	int err;
+
+	used_state.func_id = hinic3_global_func_id(hwdev);
+	used_state.svc_type = svc_type;
+	used_state.used_state = state;
+
+	mgmt_msg_params_init_default(&msg_params, &used_state,
+				     sizeof(used_state));
+
+	err = hinic3_send_mbox_to_mgmt(hwdev, MGMT_MOD_COMM,
+				       COMM_CMD_SET_FUNC_SVC_USED_STATE,
+				       &msg_params);
+	if (err || used_state.head.status) {
+		dev_err(hwdev->dev,
+			"Failed to set func service used state, err: %d, status: 0x%x\n",
+			err, used_state.head.status);
+		return -EIO;
+	}
+
+	return 0;
+}
+
+int hinic3_set_dma_attr_tbl(struct hinic3_hwdev *hwdev, u8 entry_idx, u8 st,
+			    u8 at, u8 ph, u8 no_snooping, u8 tph_en)
+{
+	struct comm_cmd_set_dma_attr dma_attr = {};
+	struct mgmt_msg_params msg_params = {};
+	int err;
+
+	dma_attr.func_id = hinic3_global_func_id(hwdev);
+	dma_attr.entry_idx = entry_idx;
+	dma_attr.st = st;
+	dma_attr.at = at;
+	dma_attr.ph = ph;
+	dma_attr.no_snooping = no_snooping;
+	dma_attr.tph_en = tph_en;
+
+	mgmt_msg_params_init_default(&msg_params, &dma_attr, sizeof(dma_attr));
+
+	err = hinic3_send_mbox_to_mgmt(hwdev, MGMT_MOD_COMM,
+				       COMM_CMD_SET_DMA_ATTR, &msg_params);
+	if (err || dma_attr.head.status) {
+		dev_err(hwdev->dev, "Failed to set dma attr, err: %d, status: 0x%x\n",
+			err, dma_attr.head.status);
+		return -EIO;
+	}
+
+	return 0;
+}
+
+int hinic3_set_wq_page_size(struct hinic3_hwdev *hwdev, u16 func_idx,
+			    u32 page_size)
+{
+	struct comm_cmd_cfg_wq_page_size page_size_info = {};
+	struct mgmt_msg_params msg_params = {};
+	int err;
+
+	page_size_info.func_id = func_idx;
+	page_size_info.page_size = ilog2(page_size / HINIC3_MIN_PAGE_SIZE);
+	page_size_info.opcode = MGMT_MSG_CMD_OP_SET;
+
+	mgmt_msg_params_init_default(&msg_params, &page_size_info,
+				     sizeof(page_size_info));
+
+	err = hinic3_send_mbox_to_mgmt(hwdev, MGMT_MOD_COMM,
+				       COMM_CMD_CFG_PAGESIZE, &msg_params);
+	if (err || page_size_info.head.status) {
+		dev_err(hwdev->dev,
+			"Failed to set wq page size, err: %d, status: 0x%x\n",
+			err, page_size_info.head.status);
+		return -EFAULT;
+	}
+
+	return 0;
+}
+
+int hinic3_set_cmdq_depth(struct hinic3_hwdev *hwdev, u16 cmdq_depth)
+{
+	struct comm_cmd_set_root_ctxt root_ctxt = {};
+	struct mgmt_msg_params msg_params = {};
+	int err;
+
+	root_ctxt.func_id = hinic3_global_func_id(hwdev);
+
+	root_ctxt.set_cmdq_depth = 1;
+	root_ctxt.cmdq_depth = ilog2(cmdq_depth);
+
+	mgmt_msg_params_init_default(&msg_params, &root_ctxt,
+				     sizeof(root_ctxt));
+
+	err = hinic3_send_mbox_to_mgmt(hwdev, MGMT_MOD_COMM,
+				       COMM_CMD_SET_VAT, &msg_params);
+	if (err || root_ctxt.head.status) {
+		dev_err(hwdev->dev,
+			"Failed to set cmdq depth, err: %d, status: 0x%x\n",
+			err, root_ctxt.head.status);
+		return -EFAULT;
+	}
+
+	return 0;
+}
+
+#define HINIC3_WAIT_CMDQ_IDLE_TIMEOUT    5000
+
+static enum hinic3_wait_return check_cmdq_stop_handler(void *priv_data)
+{
+	struct hinic3_hwdev *hwdev = priv_data;
+	enum hinic3_cmdq_type cmdq_type;
+	struct hinic3_cmdqs *cmdqs;
+
+	cmdqs = hwdev->cmdqs;
+	for (cmdq_type = 0; cmdq_type < cmdqs->cmdq_num; cmdq_type++) {
+		if (!hinic3_cmdq_idle(&cmdqs->cmdq[cmdq_type]))
+			return HINIC3_WAIT_PROCESS_WAITING;
+	}
+
+	return HINIC3_WAIT_PROCESS_CPL;
+}
+
+static int wait_cmdq_stop(struct hinic3_hwdev *hwdev)
+{
+	struct hinic3_cmdqs *cmdqs = hwdev->cmdqs;
+	enum hinic3_cmdq_type cmdq_type;
+	int err;
+
+	if (!(cmdqs->status & HINIC3_CMDQ_ENABLE))
+		return 0;
+
+	cmdqs->status &= ~HINIC3_CMDQ_ENABLE;
+	err = hinic3_wait_for_timeout(hwdev, check_cmdq_stop_handler,
+				      HINIC3_WAIT_CMDQ_IDLE_TIMEOUT,
+				      USEC_PER_MSEC);
+
+	if (err)
+		goto err_reenable_cmdq;
+
+	return 0;
+
+err_reenable_cmdq:
+	for (cmdq_type = 0; cmdq_type < cmdqs->cmdq_num; cmdq_type++) {
+		if (!hinic3_cmdq_idle(&cmdqs->cmdq[cmdq_type]))
+			dev_err(hwdev->dev, "Cmdq %d is busy\n", cmdq_type);
+	}
+	cmdqs->status |= HINIC3_CMDQ_ENABLE;
+
+	return err;
+}
+
+int hinic3_func_rx_tx_flush(struct hinic3_hwdev *hwdev)
+{
+	struct comm_cmd_clear_resource clear_db = {};
+	struct comm_cmd_clear_resource clr_res = {};
+	struct hinic3_hwif *hwif = hwdev->hwif;
+	struct mgmt_msg_params msg_params = {};
+	int ret = 0;
+	int err;
+
+	err = wait_cmdq_stop(hwdev);
+	if (err) {
+		dev_warn(hwdev->dev, "CMDQ is still working, CMDQ timeout value is unreasonable\n");
+		ret = err;
+	}
+
+	hinic3_toggle_doorbell(hwif, DISABLE_DOORBELL);
+
+	clear_db.func_id = hwif->attr.func_global_idx;
+	mgmt_msg_params_init_default(&msg_params, &clear_db, sizeof(clear_db));
+	err = hinic3_send_mbox_to_mgmt(hwdev, MGMT_MOD_COMM,
+				       COMM_CMD_FLUSH_DOORBELL, &msg_params);
+	if (err || clear_db.head.status) {
+		dev_warn(hwdev->dev, "Failed to flush doorbell, err: %d, status: 0x%x\n",
+			 err, clear_db.head.status);
+		if (err)
+			ret = err;
+		else
+			ret = -EFAULT;
+	}
+
+	clr_res.func_id = hwif->attr.func_global_idx;
+	msg_params.buf_in = &clr_res;
+	msg_params.in_size = sizeof(clr_res);
+	err = hinic3_send_mbox_to_mgmt_no_ack(hwdev, MGMT_MOD_COMM,
+					      COMM_CMD_START_FLUSH,
+					      &msg_params);
+	if (err) {
+		dev_warn(hwdev->dev, "Failed to notice flush message, err: %d\n",
+			 err);
+		ret = err;
+	}
+
+	hinic3_toggle_doorbell(hwif, ENABLE_DOORBELL);
+
+	err = hinic3_reinit_cmdq_ctxts(hwdev);
+	if (err) {
+		dev_warn(hwdev->dev, "Failed to reinit cmdq\n");
+		ret = err;
+	}
+
+	return ret;
+}
+
+static int get_hw_rx_buf_size_idx(int rx_buf_sz, u16 *buf_sz_idx)
+{
+	/* Supported RX buffer sizes in bytes. Configured by array index. */
+	static const int supported_sizes[16] = {
+		[0] = 32,     [1] = 64,     [2] = 96,     [3] = 128,
+		[4] = 192,    [5] = 256,    [6] = 384,    [7] = 512,
+		[8] = 768,    [9] = 1024,   [10] = 1536,  [11] = 2048,
+		[12] = 3072,  [13] = 4096,  [14] = 8192,  [15] = 16384,
+	};
+	u16 idx;
+
+	/* Scan from biggest to smallest. Choose supported size that is equal or
+	 * smaller. For smaller value HW will under-utilize posted buffers. For
+	 * bigger value HW may overrun posted buffers.
+	 */
+	idx = ARRAY_SIZE(supported_sizes);
+	while (idx > 0) {
+		idx--;
+		if (supported_sizes[idx] <= rx_buf_sz) {
+			*buf_sz_idx = idx;
+			return 0;
+		}
+	}
+
+	return -EINVAL;
+}
+
+int hinic3_set_root_ctxt(struct hinic3_hwdev *hwdev, u32 rq_depth, u32 sq_depth,
+			 int rx_buf_sz)
+{
+	struct comm_cmd_set_root_ctxt root_ctxt = {};
+	struct mgmt_msg_params msg_params = {};
+	u16 buf_sz_idx;
+	int err;
+
+	err = get_hw_rx_buf_size_idx(rx_buf_sz, &buf_sz_idx);
+	if (err)
+		return err;
+
+	root_ctxt.func_id = hinic3_global_func_id(hwdev);
+
+	root_ctxt.set_cmdq_depth = 0;
+	root_ctxt.cmdq_depth = 0;
+
+	root_ctxt.lro_en = 1;
+
+	root_ctxt.rq_depth  = ilog2(rq_depth);
+	root_ctxt.rx_buf_sz = buf_sz_idx;
+	root_ctxt.sq_depth  = ilog2(sq_depth);
+
+	mgmt_msg_params_init_default(&msg_params, &root_ctxt,
+				     sizeof(root_ctxt));
+
+	err = hinic3_send_mbox_to_mgmt(hwdev, MGMT_MOD_COMM,
+				       COMM_CMD_SET_VAT, &msg_params);
+	if (err || root_ctxt.head.status) {
+		dev_err(hwdev->dev,
+			"Failed to set root context, err: %d, status: 0x%x\n",
+			err, root_ctxt.head.status);
+		return -EFAULT;
+	}
+
+	return 0;
+}
+
+int hinic3_clean_root_ctxt(struct hinic3_hwdev *hwdev)
+{
+	struct comm_cmd_set_root_ctxt root_ctxt = {};
+	struct mgmt_msg_params msg_params = {};
+	int err;
+
+	root_ctxt.func_id = hinic3_global_func_id(hwdev);
+
+	mgmt_msg_params_init_default(&msg_params, &root_ctxt,
+				     sizeof(root_ctxt));
+
+	err = hinic3_send_mbox_to_mgmt(hwdev, MGMT_MOD_COMM,
+				       COMM_CMD_SET_VAT, &msg_params);
+	if (err || root_ctxt.head.status) {
+		dev_err(hwdev->dev,
+			"Failed to set root context, err: %d, status: 0x%x\n",
+			err, root_ctxt.head.status);
+		return -EFAULT;
+	}
+
+	return 0;
+}
diff --git a/drivers/net/ethernet/huawei/hinic3/hinic3_hw_comm.h b/drivers/net/ethernet/huawei/hinic3/hinic3_hw_comm.h
new file mode 100644
index 000000000000..304f5691f0c2
--- /dev/null
+++ b/drivers/net/ethernet/huawei/hinic3/hinic3_hw_comm.h
@@ -0,0 +1,47 @@
+/* SPDX-License-Identifier: GPL-2.0 */
+/* Copyright (c) Huawei Technologies Co., Ltd. 2025. All rights reserved. */
+
+#ifndef _HINIC3_HW_COMM_H_
+#define _HINIC3_HW_COMM_H_
+
+#include "hinic3_hw_intf.h"
+
+struct hinic3_hwdev;
+
+#define HINIC3_WQ_PAGE_SIZE_ORDER  8
+
+struct hinic3_interrupt_info {
+	u32 lli_set;
+	u32 interrupt_coalesc_set;
+	u16 msix_index;
+	u8  lli_credit_limit;
+	u8  lli_timer_cfg;
+	u8  pending_limit;
+	u8  coalesc_timer_cfg;
+	u8  resend_timer_cfg;
+};
+
+int hinic3_set_interrupt_cfg_direct(struct hinic3_hwdev *hwdev,
+				    const struct hinic3_interrupt_info *info);
+int hinic3_func_reset(struct hinic3_hwdev *hwdev, u16 func_id, u64 reset_flag);
+
+int hinic3_get_comm_features(struct hinic3_hwdev *hwdev, u64 *s_feature,
+			     u16 size);
+int hinic3_set_comm_features(struct hinic3_hwdev *hwdev, u64 *s_feature,
+			     u16 size);
+int hinic3_get_global_attr(struct hinic3_hwdev *hwdev,
+			   struct comm_global_attr *attr);
+int hinic3_set_func_svc_used_state(struct hinic3_hwdev *hwdev, u16 svc_type,
+				   u8 state);
+int hinic3_set_dma_attr_tbl(struct hinic3_hwdev *hwdev, u8 entry_idx, u8 st,
+			    u8 at, u8 ph, u8 no_snooping, u8 tph_en);
+
+int hinic3_set_wq_page_size(struct hinic3_hwdev *hwdev, u16 func_idx,
+			    u32 page_size);
+int hinic3_set_cmdq_depth(struct hinic3_hwdev *hwdev, u16 cmdq_depth);
+int hinic3_func_rx_tx_flush(struct hinic3_hwdev *hwdev);
+int hinic3_set_root_ctxt(struct hinic3_hwdev *hwdev, u32 rq_depth, u32 sq_depth,
+			 int rx_buf_sz);
+int hinic3_clean_root_ctxt(struct hinic3_hwdev *hwdev);
+
+#endif
diff --git a/drivers/net/ethernet/huawei/hinic3/hinic3_hw_intf.h b/drivers/net/ethernet/huawei/hinic3/hinic3_hw_intf.h
new file mode 100644
index 000000000000..623cf2d14cbc
--- /dev/null
+++ b/drivers/net/ethernet/huawei/hinic3/hinic3_hw_intf.h
@@ -0,0 +1,264 @@
+/* SPDX-License-Identifier: GPL-2.0 */
+/* Copyright (c) Huawei Technologies Co., Ltd. 2025. All rights reserved. */
+
+#ifndef _HINIC3_HW_INTF_H_
+#define _HINIC3_HW_INTF_H_
+
+#include <linux/bits.h>
+#include <linux/types.h>
+
+#define MGMT_MSG_CMD_OP_SET   1
+#define MGMT_MSG_CMD_OP_GET   0
+
+#define MGMT_STATUS_PF_SET_VF_ALREADY  0x4
+#define MGMT_STATUS_EXIST              0x6
+#define MGMT_STATUS_CMD_UNSUPPORTED    0xFF
+
+#define MGMT_MSG_POLLING_TIMEOUT 0
+
+struct mgmt_msg_head {
+	u8 status;
+	u8 version;
+	u8 rsvd0[6];
+};
+
+struct mgmt_msg_params {
+	const void  *buf_in;
+	u32         in_size;
+	void        *buf_out;
+	u32         expected_out_size;
+	u32         timeout_ms;
+};
+
+/* CMDQ MODULE_TYPE */
+enum mgmt_mod_type {
+	/* HW communication module */
+	MGMT_MOD_COMM   = 0,
+	/* L2NIC module */
+	MGMT_MOD_L2NIC  = 1,
+	/* Configuration module */
+	MGMT_MOD_CFGM   = 7,
+	MGMT_MOD_HILINK = 14,
+};
+
+static inline void mgmt_msg_params_init_default(struct mgmt_msg_params *msg_params,
+						void *inout_buf, u32 buf_size)
+{
+	msg_params->buf_in = inout_buf;
+	msg_params->buf_out = inout_buf;
+	msg_params->in_size = buf_size;
+	msg_params->expected_out_size = buf_size;
+	msg_params->timeout_ms = 0;
+}
+
+enum cfg_cmd {
+	CFG_CMD_GET_DEV_CAP = 0,
+};
+
+/* Device capabilities, defined by hw */
+struct cfg_cmd_dev_cap {
+	struct mgmt_msg_head head;
+
+	u16                  func_id;
+	u16                  rsvd1;
+
+	/* Public resources */
+	u8                   host_id;
+	u8                   ep_id;
+	u8                   er_id;
+	u8                   port_id;
+
+	u16                  host_total_func;
+	u8                   host_pf_num;
+	u8                   pf_id_start;
+	u16                  host_vf_num;
+	u16                  vf_id_start;
+	u8                   host_oq_id_mask_val;
+	u8                   timer_en;
+	u8                   host_valid_bitmap;
+	u8                   rsvd_host;
+
+	u16                  svc_cap_en;
+	u16                  max_vf;
+	u8                   flexq_en;
+	u8                   valid_cos_bitmap;
+	u8                   port_cos_valid_bitmap;
+	u8                   rsvd2[45];
+
+	/* l2nic */
+	u16                  nic_max_sq_id;
+	u16                  nic_max_rq_id;
+	u16                  nic_default_num_queues;
+
+	u8                   rsvd3[250];
+};
+
+/* COMM Commands between Driver to fw */
+enum comm_cmd {
+	/* Commands for clearing FLR and resources */
+	COMM_CMD_FUNC_RESET              = 0,
+	COMM_CMD_FEATURE_NEGO            = 1,
+	COMM_CMD_FLUSH_DOORBELL          = 2,
+	COMM_CMD_START_FLUSH             = 3,
+	COMM_CMD_GET_GLOBAL_ATTR         = 5,
+	COMM_CMD_SET_FUNC_SVC_USED_STATE = 7,
+
+	/* Driver Configuration Commands */
+	COMM_CMD_SET_CMDQ_CTXT           = 20,
+	COMM_CMD_SET_VAT                 = 21,
+	COMM_CMD_CFG_PAGESIZE            = 22,
+	COMM_CMD_CFG_MSIX_CTRL_REG       = 23,
+	COMM_CMD_SET_CEQ_CTRL_REG        = 24,
+	COMM_CMD_SET_DMA_ATTR            = 25,
+};
+
+struct comm_cmd_cfg_msix_ctrl_reg {
+	struct mgmt_msg_head head;
+	u16                  func_id;
+	u8                   opcode;
+	u8                   rsvd1;
+	u16                  msix_index;
+	u8                   pending_cnt;
+	u8                   coalesce_timer_cnt;
+	u8                   resend_timer_cnt;
+	u8                   lli_timer_cnt;
+	u8                   lli_credit_cnt;
+	u8                   rsvd2[5];
+};
+
+enum comm_func_reset_bits {
+	COMM_FUNC_RESET_BIT_FLUSH        = BIT(0),
+	COMM_FUNC_RESET_BIT_MQM          = BIT(1),
+	COMM_FUNC_RESET_BIT_SMF          = BIT(2),
+	COMM_FUNC_RESET_BIT_PF_BW_CFG    = BIT(3),
+
+	COMM_FUNC_RESET_BIT_COMM         = BIT(10),
+	/* clear mbox and aeq, The COMM_FUNC_RESET_BIT_COMM bit must be set */
+	COMM_FUNC_RESET_BIT_COMM_MGMT_CH = BIT(11),
+	/* clear cmdq and ceq, The COMM_FUNC_RESET_BIT_COMM bit must be set */
+	COMM_FUNC_RESET_BIT_COMM_CMD_CH  = BIT(12),
+	COMM_FUNC_RESET_BIT_NIC          = BIT(13),
+};
+
+#define COMM_FUNC_RESET_FLAG \
+	(COMM_FUNC_RESET_BIT_COMM | COMM_FUNC_RESET_BIT_COMM_CMD_CH | \
+	 COMM_FUNC_RESET_BIT_FLUSH | COMM_FUNC_RESET_BIT_MQM | \
+	 COMM_FUNC_RESET_BIT_SMF | COMM_FUNC_RESET_BIT_PF_BW_CFG)
+
+struct comm_cmd_func_reset {
+	struct mgmt_msg_head head;
+	u16                  func_id;
+	u16                  rsvd1[3];
+	u64                  reset_flag;
+};
+
+#define COMM_MAX_FEATURE_QWORD  4
+struct comm_cmd_feature_nego {
+	struct mgmt_msg_head head;
+	u16                  func_id;
+	u8                   opcode;
+	u8                   rsvd;
+	u64                  s_feature[COMM_MAX_FEATURE_QWORD];
+};
+
+struct comm_global_attr {
+	u8  max_host_num;
+	u8  max_pf_num;
+	u16 vf_id_start;
+	/* for api cmd to mgmt cpu */
+	u8  mgmt_host_node_id;
+	u8  cmdq_num;
+	u8  rsvd1[34];
+};
+
+struct comm_cmd_get_glb_attr {
+	struct mgmt_msg_head    head;
+	struct comm_global_attr attr;
+};
+
+enum comm_func_svc_type {
+	COMM_FUNC_SVC_T_COMM = 0,
+	COMM_FUNC_SVC_T_NIC  = 1,
+};
+
+struct comm_cmd_set_func_svc_used_state {
+	struct mgmt_msg_head head;
+	u16                  func_id;
+	u16                  svc_type;
+	u8                   used_state;
+	u8                   rsvd[35];
+};
+
+struct comm_cmd_set_dma_attr {
+	struct mgmt_msg_head head;
+	u16                  func_id;
+	u8                   entry_idx;
+	u8                   st;
+	u8                   at;
+	u8                   ph;
+	u8                   no_snooping;
+	u8                   tph_en;
+	u32                  resv1;
+};
+
+struct comm_cmd_set_ceq_ctrl_reg {
+	struct mgmt_msg_head head;
+	u16                  func_id;
+	u16                  q_id;
+	u32                  ctrl0;
+	u32                  ctrl1;
+	u32                  rsvd1;
+};
+
+struct comm_cmd_cfg_wq_page_size {
+	struct mgmt_msg_head head;
+	u16                  func_id;
+	u8                   opcode;
+	/* real_size=4KB*2^page_size, range(0~20) must be checked by driver */
+	u8                   page_size;
+	u32                  rsvd1;
+};
+
+struct comm_cmd_set_root_ctxt {
+	struct mgmt_msg_head head;
+	u16                  func_id;
+	u8                   set_cmdq_depth;
+	u8                   cmdq_depth;
+	u16                  rx_buf_sz;
+	u8                   lro_en;
+	u8                   rsvd1;
+	u16                  sq_depth;
+	u16                  rq_depth;
+	u64                  rsvd2;
+};
+
+struct comm_cmdq_ctxt_info {
+	__le64 curr_wqe_page_pfn;
+	__le64 wq_block_pfn;
+};
+
+struct comm_cmd_set_cmdq_ctxt {
+	struct mgmt_msg_head       head;
+	u16                        func_id;
+	u8                         cmdq_id;
+	u8                         rsvd1[5];
+	struct comm_cmdq_ctxt_info ctxt;
+};
+
+struct comm_cmd_clear_resource {
+	struct mgmt_msg_head head;
+	u16                  func_id;
+	u16                  rsvd1[3];
+};
+
+/* Services supported by HW. HW uses these values when delivering events.
+ * HW supports multiple services that are not yet supported by driver
+ * (e.g. RoCE).
+ */
+enum hinic3_service_type {
+	HINIC3_SERVICE_T_NIC = 0,
+	/* MAX is only used by SW for array sizes. */
+	HINIC3_SERVICE_T_MAX = 1,
+};
+
+#endif
diff --git a/drivers/net/ethernet/huawei/hinic3/hinic3_hwdev.c b/drivers/net/ethernet/huawei/hinic3/hinic3_hwdev.c
new file mode 100644
index 000000000000..95a213133be9
--- /dev/null
+++ b/drivers/net/ethernet/huawei/hinic3/hinic3_hwdev.c
@@ -0,0 +1,557 @@
+// SPDX-License-Identifier: GPL-2.0
+// Copyright (c) Huawei Technologies Co., Ltd. 2025. All rights reserved.
+
+#include "hinic3_cmdq.h"
+#include "hinic3_csr.h"
+#include "hinic3_eqs.h"
+#include "hinic3_hw_comm.h"
+#include "hinic3_hwdev.h"
+#include "hinic3_hwif.h"
+#include "hinic3_mbox.h"
+#include "hinic3_mgmt.h"
+
+#define HINIC3_PCIE_SNOOP        0
+#define HINIC3_PCIE_TPH_DISABLE  0
+
+#define HINIC3_DMA_ATTR_INDIR_IDX_MASK          GENMASK(9, 0)
+#define HINIC3_DMA_ATTR_INDIR_IDX_SET(val, member)  \
+	FIELD_PREP(HINIC3_DMA_ATTR_INDIR_##member##_MASK, val)
+
+#define HINIC3_DMA_ATTR_ENTRY_ST_MASK           GENMASK(7, 0)
+#define HINIC3_DMA_ATTR_ENTRY_AT_MASK           GENMASK(9, 8)
+#define HINIC3_DMA_ATTR_ENTRY_PH_MASK           GENMASK(11, 10)
+#define HINIC3_DMA_ATTR_ENTRY_NO_SNOOPING_MASK  BIT(12)
+#define HINIC3_DMA_ATTR_ENTRY_TPH_EN_MASK       BIT(13)
+#define HINIC3_DMA_ATTR_ENTRY_SET(val, member)  \
+	FIELD_PREP(HINIC3_DMA_ATTR_ENTRY_##member##_MASK, val)
+
+#define HINIC3_PCIE_ST_DISABLE       0
+#define HINIC3_PCIE_AT_DISABLE       0
+#define HINIC3_PCIE_PH_DISABLE       0
+#define HINIC3_PCIE_MSIX_ATTR_ENTRY  0
+
+#define HINIC3_DEFAULT_EQ_MSIX_PENDING_LIMIT      0
+#define HINIC3_DEFAULT_EQ_MSIX_COALESC_TIMER_CFG  0xFF
+#define HINIC3_DEFAULT_EQ_MSIX_RESEND_TIMER_CFG   7
+
+#define HINIC3_HWDEV_WQ_NAME    "hinic3_hardware"
+#define HINIC3_WQ_MAX_REQ       10
+
+enum hinic3_hwdev_init_state {
+	HINIC3_HWDEV_MBOX_INITED = 2,
+	HINIC3_HWDEV_CMDQ_INITED = 3,
+};
+
+static int hinic3_comm_aeqs_init(struct hinic3_hwdev *hwdev)
+{
+	struct msix_entry aeq_msix_entries[HINIC3_MAX_AEQS];
+	u16 num_aeqs, resp_num_irq, i;
+	int err;
+
+	num_aeqs = hwdev->hwif->attr.num_aeqs;
+	if (num_aeqs > HINIC3_MAX_AEQS) {
+		dev_warn(hwdev->dev, "Adjust aeq num to %d\n",
+			 HINIC3_MAX_AEQS);
+		num_aeqs = HINIC3_MAX_AEQS;
+	}
+	err = hinic3_alloc_irqs(hwdev, num_aeqs, aeq_msix_entries,
+				&resp_num_irq);
+	if (err) {
+		dev_err(hwdev->dev, "Failed to alloc aeq irqs, num_aeqs: %u\n",
+			num_aeqs);
+		return err;
+	}
+
+	if (resp_num_irq < num_aeqs) {
+		dev_warn(hwdev->dev, "Adjust aeq num to %u\n",
+			 resp_num_irq);
+		num_aeqs = resp_num_irq;
+	}
+
+	err = hinic3_aeqs_init(hwdev, num_aeqs, aeq_msix_entries);
+	if (err) {
+		dev_err(hwdev->dev, "Failed to init aeqs\n");
+		goto err_free_irqs;
+	}
+
+	return 0;
+
+err_free_irqs:
+	for (i = 0; i < num_aeqs; i++)
+		hinic3_free_irq(hwdev, aeq_msix_entries[i].vector);
+
+	return err;
+}
+
+static int hinic3_comm_ceqs_init(struct hinic3_hwdev *hwdev)
+{
+	struct msix_entry ceq_msix_entries[HINIC3_MAX_CEQS];
+	u16 num_ceqs, resp_num_irq, i;
+	int err;
+
+	num_ceqs = hwdev->hwif->attr.num_ceqs;
+	if (num_ceqs > HINIC3_MAX_CEQS) {
+		dev_warn(hwdev->dev, "Adjust ceq num to %d\n",
+			 HINIC3_MAX_CEQS);
+		num_ceqs = HINIC3_MAX_CEQS;
+	}
+
+	err = hinic3_alloc_irqs(hwdev, num_ceqs, ceq_msix_entries,
+				&resp_num_irq);
+	if (err) {
+		dev_err(hwdev->dev, "Failed to alloc ceq irqs, num_ceqs: %u\n",
+			num_ceqs);
+		return err;
+	}
+
+	if (resp_num_irq < num_ceqs) {
+		dev_warn(hwdev->dev, "Adjust ceq num to %u\n",
+			 resp_num_irq);
+		num_ceqs = resp_num_irq;
+	}
+
+	err = hinic3_ceqs_init(hwdev, num_ceqs, ceq_msix_entries);
+	if (err) {
+		dev_err(hwdev->dev,
+			"Failed to init ceqs, err:%d\n", err);
+		goto err_free_irqs;
+	}
+
+	return 0;
+
+err_free_irqs:
+	for (i = 0; i < num_ceqs; i++)
+		hinic3_free_irq(hwdev, ceq_msix_entries[i].vector);
+
+	return err;
+}
+
+static int hinic3_comm_mbox_init(struct hinic3_hwdev *hwdev)
+{
+	int err;
+
+	err = hinic3_init_mbox(hwdev);
+	if (err)
+		return err;
+
+	hinic3_aeq_register_cb(hwdev, HINIC3_MBX_FROM_FUNC,
+			       hinic3_mbox_func_aeqe_handler);
+	hinic3_aeq_register_cb(hwdev, HINIC3_MSG_FROM_FW,
+			       hinic3_mgmt_msg_aeqe_handler);
+
+	set_bit(HINIC3_HWDEV_MBOX_INITED, &hwdev->func_state);
+
+	return 0;
+}
+
+static void hinic3_comm_mbox_free(struct hinic3_hwdev *hwdev)
+{
+	spin_lock_bh(&hwdev->channel_lock);
+	clear_bit(HINIC3_HWDEV_MBOX_INITED, &hwdev->func_state);
+	spin_unlock_bh(&hwdev->channel_lock);
+	hinic3_aeq_unregister_cb(hwdev, HINIC3_MBX_FROM_FUNC);
+	hinic3_aeq_unregister_cb(hwdev, HINIC3_MSG_FROM_FW);
+	hinic3_free_mbox(hwdev);
+}
+
+static int init_aeqs_msix_attr(struct hinic3_hwdev *hwdev)
+{
+	struct hinic3_aeqs *aeqs = hwdev->aeqs;
+	struct hinic3_interrupt_info info = {};
+	struct hinic3_eq *eq;
+	u16 q_id;
+	int err;
+
+	info.interrupt_coalesc_set = 1;
+	info.pending_limit = HINIC3_DEFAULT_EQ_MSIX_PENDING_LIMIT;
+	info.coalesc_timer_cfg = HINIC3_DEFAULT_EQ_MSIX_COALESC_TIMER_CFG;
+	info.resend_timer_cfg = HINIC3_DEFAULT_EQ_MSIX_RESEND_TIMER_CFG;
+
+	for (q_id = 0; q_id < aeqs->num_aeqs; q_id++) {
+		eq = &aeqs->aeq[q_id];
+		info.msix_index = eq->msix_entry_idx;
+		err = hinic3_set_interrupt_cfg_direct(hwdev, &info);
+		if (err) {
+			dev_err(hwdev->dev, "Set msix attr for aeq %d failed\n",
+				q_id);
+			return err;
+		}
+	}
+
+	return 0;
+}
+
+static int init_ceqs_msix_attr(struct hinic3_hwdev *hwdev)
+{
+	struct hinic3_ceqs *ceqs = hwdev->ceqs;
+	struct hinic3_interrupt_info info = {};
+	struct hinic3_eq *eq;
+	u16 q_id;
+	int err;
+
+	info.interrupt_coalesc_set = 1;
+	info.pending_limit = HINIC3_DEFAULT_EQ_MSIX_PENDING_LIMIT;
+	info.coalesc_timer_cfg = HINIC3_DEFAULT_EQ_MSIX_COALESC_TIMER_CFG;
+	info.resend_timer_cfg = HINIC3_DEFAULT_EQ_MSIX_RESEND_TIMER_CFG;
+
+	for (q_id = 0; q_id < ceqs->num_ceqs; q_id++) {
+		eq = &ceqs->ceq[q_id];
+		info.msix_index = eq->msix_entry_idx;
+		err = hinic3_set_interrupt_cfg_direct(hwdev, &info);
+		if (err) {
+			dev_err(hwdev->dev, "Set msix attr for ceq %u failed\n",
+				q_id);
+			return err;
+		}
+	}
+
+	return 0;
+}
+
+static int init_basic_mgmt_channel(struct hinic3_hwdev *hwdev)
+{
+	int err;
+
+	err = hinic3_comm_aeqs_init(hwdev);
+	if (err) {
+		dev_err(hwdev->dev, "Failed to init async event queues\n");
+		return err;
+	}
+
+	err = hinic3_comm_mbox_init(hwdev);
+	if (err) {
+		dev_err(hwdev->dev, "Failed to init mailbox\n");
+		goto err_free_comm_aeqs;
+	}
+
+	err = init_aeqs_msix_attr(hwdev);
+	if (err) {
+		dev_err(hwdev->dev, "Failed to init aeqs msix attr\n");
+		goto err_free_comm_mbox;
+	}
+
+	return 0;
+
+err_free_comm_mbox:
+	hinic3_comm_mbox_free(hwdev);
+err_free_comm_aeqs:
+	hinic3_aeqs_free(hwdev);
+
+	return err;
+}
+
+static void free_base_mgmt_channel(struct hinic3_hwdev *hwdev)
+{
+	hinic3_comm_mbox_free(hwdev);
+	hinic3_aeqs_free(hwdev);
+}
+
+static int dma_attr_table_init(struct hinic3_hwdev *hwdev)
+{
+	u32 addr, val, dst_attr;
+
+	/* Indirect access, set entry_idx first */
+	addr = HINIC3_CSR_DMA_ATTR_INDIR_IDX_ADDR;
+	val = hinic3_hwif_read_reg(hwdev->hwif, addr);
+	val &= ~HINIC3_DMA_ATTR_ENTRY_AT_MASK;
+	val |= HINIC3_DMA_ATTR_INDIR_IDX_SET(HINIC3_PCIE_MSIX_ATTR_ENTRY, IDX);
+	hinic3_hwif_write_reg(hwdev->hwif, addr, val);
+
+	addr = HINIC3_CSR_DMA_ATTR_TBL_ADDR;
+	val = hinic3_hwif_read_reg(hwdev->hwif, addr);
+
+	dst_attr = HINIC3_DMA_ATTR_ENTRY_SET(HINIC3_PCIE_ST_DISABLE, ST) |
+		   HINIC3_DMA_ATTR_ENTRY_SET(HINIC3_PCIE_AT_DISABLE, AT) |
+		   HINIC3_DMA_ATTR_ENTRY_SET(HINIC3_PCIE_PH_DISABLE, PH) |
+		   HINIC3_DMA_ATTR_ENTRY_SET(HINIC3_PCIE_SNOOP, NO_SNOOPING) |
+		   HINIC3_DMA_ATTR_ENTRY_SET(HINIC3_PCIE_TPH_DISABLE, TPH_EN);
+	if (val == dst_attr)
+		return 0;
+
+	return hinic3_set_dma_attr_tbl(hwdev,
+				       HINIC3_PCIE_MSIX_ATTR_ENTRY,
+				       HINIC3_PCIE_ST_DISABLE,
+				       HINIC3_PCIE_AT_DISABLE,
+				       HINIC3_PCIE_PH_DISABLE,
+				       HINIC3_PCIE_SNOOP,
+				       HINIC3_PCIE_TPH_DISABLE);
+}
+
+static int init_basic_attributes(struct hinic3_hwdev *hwdev)
+{
+	struct comm_global_attr glb_attr;
+	int err;
+
+	err = hinic3_func_reset(hwdev, hinic3_global_func_id(hwdev),
+				COMM_FUNC_RESET_FLAG);
+	if (err)
+		return err;
+
+	err = hinic3_get_comm_features(hwdev, hwdev->features,
+				       COMM_MAX_FEATURE_QWORD);
+	if (err)
+		return err;
+
+	dev_dbg(hwdev->dev, "Comm hw features: 0x%llx\n", hwdev->features[0]);
+
+	err = hinic3_get_global_attr(hwdev, &glb_attr);
+	if (err)
+		return err;
+
+	err = hinic3_set_func_svc_used_state(hwdev, COMM_FUNC_SVC_T_COMM, 1);
+	if (err)
+		return err;
+
+	err = dma_attr_table_init(hwdev);
+	if (err)
+		return err;
+
+	hwdev->max_cmdq = min(glb_attr.cmdq_num, HINIC3_MAX_CMDQ_TYPES);
+	dev_dbg(hwdev->dev,
+		"global attribute: max_host: 0x%x, max_pf: 0x%x, vf_id_start: 0x%x, mgmt node id: 0x%x, cmdq_num: 0x%x\n",
+		glb_attr.max_host_num, glb_attr.max_pf_num,
+		glb_attr.vf_id_start, glb_attr.mgmt_host_node_id,
+		glb_attr.cmdq_num);
+
+	return 0;
+}
+
+static int hinic3_comm_cmdqs_init(struct hinic3_hwdev *hwdev)
+{
+	int err;
+
+	err = hinic3_cmdqs_init(hwdev);
+	if (err) {
+		dev_err(hwdev->dev, "Failed to init cmd queues\n");
+		return err;
+	}
+
+	hinic3_ceq_register_cb(hwdev, HINIC3_CMDQ, hinic3_cmdq_ceq_handler);
+
+	err = hinic3_set_cmdq_depth(hwdev, CMDQ_DEPTH);
+	if (err) {
+		dev_err(hwdev->dev, "Failed to set cmdq depth\n");
+		goto err_free_cmdqs;
+	}
+
+	set_bit(HINIC3_HWDEV_CMDQ_INITED, &hwdev->func_state);
+
+	return 0;
+
+err_free_cmdqs:
+	hinic3_cmdqs_free(hwdev);
+
+	return err;
+}
+
+static void hinic3_comm_cmdqs_free(struct hinic3_hwdev *hwdev)
+{
+	spin_lock_bh(&hwdev->channel_lock);
+	clear_bit(HINIC3_HWDEV_CMDQ_INITED, &hwdev->func_state);
+	spin_unlock_bh(&hwdev->channel_lock);
+
+	hinic3_ceq_unregister_cb(hwdev, HINIC3_CMDQ);
+	hinic3_cmdqs_free(hwdev);
+}
+
+static int init_cmdqs_channel(struct hinic3_hwdev *hwdev)
+{
+	int err;
+
+	err = hinic3_comm_ceqs_init(hwdev);
+	if (err) {
+		dev_err(hwdev->dev, "Failed to init completion event queues\n");
+		return err;
+	}
+
+	err = init_ceqs_msix_attr(hwdev);
+	if (err) {
+		dev_err(hwdev->dev, "Failed to init ceqs msix attr\n");
+		goto err_free_ceqs;
+	}
+
+	hwdev->wq_page_size = HINIC3_MIN_PAGE_SIZE << HINIC3_WQ_PAGE_SIZE_ORDER;
+	err = hinic3_set_wq_page_size(hwdev, hinic3_global_func_id(hwdev),
+				      hwdev->wq_page_size);
+	if (err) {
+		dev_err(hwdev->dev, "Failed to set wq page size\n");
+		goto err_free_ceqs;
+	}
+
+	err = hinic3_comm_cmdqs_init(hwdev);
+	if (err) {
+		dev_err(hwdev->dev, "Failed to init cmd queues\n");
+		goto err_reset_wq_page_size;
+	}
+
+	return 0;
+
+err_reset_wq_page_size:
+	hinic3_set_wq_page_size(hwdev, hinic3_global_func_id(hwdev),
+				HINIC3_MIN_PAGE_SIZE);
+err_free_ceqs:
+	hinic3_ceqs_free(hwdev);
+
+	return err;
+}
+
+static void hinic3_free_cmdqs_channel(struct hinic3_hwdev *hwdev)
+{
+	hinic3_comm_cmdqs_free(hwdev);
+	hinic3_ceqs_free(hwdev);
+}
+
+static int hinic3_init_comm_ch(struct hinic3_hwdev *hwdev)
+{
+	int err;
+
+	err = init_basic_mgmt_channel(hwdev);
+	if (err)
+		return err;
+
+	err = init_basic_attributes(hwdev);
+	if (err)
+		goto err_free_basic_mgmt_ch;
+
+	err = init_cmdqs_channel(hwdev);
+	if (err) {
+		dev_err(hwdev->dev, "Failed to init cmdq channel\n");
+		goto err_clear_func_svc_used_state;
+	}
+
+	return 0;
+
+err_clear_func_svc_used_state:
+	hinic3_set_func_svc_used_state(hwdev, COMM_FUNC_SVC_T_COMM, 0);
+err_free_basic_mgmt_ch:
+	free_base_mgmt_channel(hwdev);
+
+	return err;
+}
+
+static void hinic3_uninit_comm_ch(struct hinic3_hwdev *hwdev)
+{
+	hinic3_free_cmdqs_channel(hwdev);
+	hinic3_set_func_svc_used_state(hwdev, COMM_FUNC_SVC_T_COMM, 0);
+	free_base_mgmt_channel(hwdev);
+}
+
+static DEFINE_IDA(hinic3_adev_ida);
+
+static int hinic3_adev_idx_alloc(void)
+{
+	return ida_alloc(&hinic3_adev_ida, GFP_KERNEL);
+}
+
+static void hinic3_adev_idx_free(int id)
+{
+	ida_free(&hinic3_adev_ida, id);
+}
+
+int hinic3_init_hwdev(struct pci_dev *pdev)
+{
+	struct hinic3_pcidev *pci_adapter = pci_get_drvdata(pdev);
+	struct hinic3_hwdev *hwdev;
+	int err;
+
+	hwdev = kzalloc(sizeof(*hwdev), GFP_KERNEL);
+	if (!hwdev)
+		return -ENOMEM;
+
+	pci_adapter->hwdev = hwdev;
+	hwdev->adapter = pci_adapter;
+	hwdev->pdev = pci_adapter->pdev;
+	hwdev->dev = &pci_adapter->pdev->dev;
+	hwdev->func_state = 0;
+	hwdev->dev_id = hinic3_adev_idx_alloc();
+	spin_lock_init(&hwdev->channel_lock);
+
+	err = hinic3_init_hwif(hwdev);
+	if (err) {
+		dev_err(hwdev->dev, "Failed to init hwif\n");
+		goto err_free_hwdev;
+	}
+
+	hwdev->workq = alloc_workqueue(HINIC3_HWDEV_WQ_NAME, WQ_MEM_RECLAIM,
+				       HINIC3_WQ_MAX_REQ);
+	if (!hwdev->workq) {
+		dev_err(hwdev->dev, "Failed to alloc hardware workq\n");
+		err = -ENOMEM;
+		goto err_free_hwif;
+	}
+
+	err = hinic3_init_cfg_mgmt(hwdev);
+	if (err) {
+		dev_err(hwdev->dev, "Failed to init config mgmt\n");
+		goto err_destroy_workqueue;
+	}
+
+	err = hinic3_init_comm_ch(hwdev);
+	if (err) {
+		dev_err(hwdev->dev, "Failed to init communication channel\n");
+		goto err_free_cfg_mgmt;
+	}
+
+	err = hinic3_init_capability(hwdev);
+	if (err) {
+		dev_err(hwdev->dev, "Failed to init capability\n");
+		goto err_uninit_comm_ch;
+	}
+
+	err = hinic3_set_comm_features(hwdev, hwdev->features,
+				       COMM_MAX_FEATURE_QWORD);
+	if (err) {
+		dev_err(hwdev->dev, "Failed to set comm features\n");
+		goto err_uninit_comm_ch;
+	}
+
+	return 0;
+
+err_uninit_comm_ch:
+	hinic3_uninit_comm_ch(hwdev);
+err_free_cfg_mgmt:
+	hinic3_free_cfg_mgmt(hwdev);
+err_destroy_workqueue:
+	destroy_workqueue(hwdev->workq);
+err_free_hwif:
+	hinic3_free_hwif(hwdev);
+err_free_hwdev:
+	pci_adapter->hwdev = NULL;
+	hinic3_adev_idx_free(hwdev->dev_id);
+	kfree(hwdev);
+
+	return err;
+}
+
+void hinic3_free_hwdev(struct hinic3_hwdev *hwdev)
+{
+	u64 drv_features[COMM_MAX_FEATURE_QWORD] = {};
+
+	hinic3_set_comm_features(hwdev, drv_features, COMM_MAX_FEATURE_QWORD);
+	hinic3_func_rx_tx_flush(hwdev);
+	hinic3_uninit_comm_ch(hwdev);
+	hinic3_free_cfg_mgmt(hwdev);
+	destroy_workqueue(hwdev->workq);
+	hinic3_free_hwif(hwdev);
+	hinic3_adev_idx_free(hwdev->dev_id);
+	kfree(hwdev);
+}
+
+void hinic3_set_api_stop(struct hinic3_hwdev *hwdev)
+{
+	struct hinic3_mbox *mbox;
+
+	spin_lock_bh(&hwdev->channel_lock);
+	if (test_bit(HINIC3_HWDEV_MBOX_INITED, &hwdev->func_state)) {
+		mbox = hwdev->mbox;
+		spin_lock(&mbox->mbox_lock);
+		if (mbox->event_flag == MBOX_EVENT_START)
+			mbox->event_flag = MBOX_EVENT_TIMEOUT;
+		spin_unlock(&mbox->mbox_lock);
+	}
+
+	if (test_bit(HINIC3_HWDEV_CMDQ_INITED, &hwdev->func_state))
+		hinic3_cmdq_flush_sync_cmd(hwdev);
+
+	spin_unlock_bh(&hwdev->channel_lock);
+}
diff --git a/drivers/net/ethernet/huawei/hinic3/hinic3_hwdev.h b/drivers/net/ethernet/huawei/hinic3/hinic3_hwdev.h
new file mode 100644
index 000000000000..62e2745e9316
--- /dev/null
+++ b/drivers/net/ethernet/huawei/hinic3/hinic3_hwdev.h
@@ -0,0 +1,81 @@
+/* SPDX-License-Identifier: GPL-2.0 */
+/* Copyright (c) Huawei Technologies Co., Ltd. 2025. All rights reserved. */
+
+#ifndef _HINIC3_HWDEV_H_
+#define _HINIC3_HWDEV_H_
+
+#include <linux/auxiliary_bus.h>
+#include <linux/pci.h>
+
+#include "hinic3_hw_intf.h"
+
+struct hinic3_cmdqs;
+struct hinic3_hwif;
+
+enum hinic3_event_service_type {
+	HINIC3_EVENT_SRV_COMM = 0,
+	HINIC3_EVENT_SRV_NIC  = 1
+};
+
+#define HINIC3_SRV_EVENT_TYPE(svc, type)    (((svc) << 16) | (type))
+
+/* driver-specific data of pci_dev */
+struct hinic3_pcidev {
+	struct pci_dev       *pdev;
+	struct hinic3_hwdev  *hwdev;
+	/* Auxiliary devices */
+	struct hinic3_adev   *hadev[HINIC3_SERVICE_T_MAX];
+
+	void __iomem         *cfg_reg_base;
+	void __iomem         *intr_reg_base;
+	void __iomem         *db_base;
+	u64                  db_dwqe_len;
+	u64                  db_base_phy;
+
+	/* lock for attach/detach uld */
+	struct mutex         pdev_mutex;
+	unsigned long        state;
+};
+
+struct hinic3_hwdev {
+	struct hinic3_pcidev        *adapter;
+	struct pci_dev              *pdev;
+	struct device               *dev;
+	int                         dev_id;
+	struct hinic3_hwif          *hwif;
+	struct hinic3_cfg_mgmt_info *cfg_mgmt;
+	struct hinic3_aeqs          *aeqs;
+	struct hinic3_ceqs          *ceqs;
+	struct hinic3_mbox          *mbox;
+	struct hinic3_cmdqs         *cmdqs;
+	struct workqueue_struct     *workq;
+	/* protect channel init and uninit */
+	spinlock_t                  channel_lock;
+	u64                         features[COMM_MAX_FEATURE_QWORD];
+	u32                         wq_page_size;
+	u8                          max_cmdq;
+	ulong                       func_state;
+};
+
+struct hinic3_event_info {
+	/* enum hinic3_event_service_type */
+	u16 service;
+	u16 type;
+	u8  event_data[104];
+};
+
+struct hinic3_adev {
+	struct auxiliary_device  adev;
+	struct hinic3_hwdev      *hwdev;
+	enum hinic3_service_type svc_type;
+
+	void (*event)(struct auxiliary_device *adev,
+		      struct hinic3_event_info *event);
+};
+
+int hinic3_init_hwdev(struct pci_dev *pdev);
+void hinic3_free_hwdev(struct hinic3_hwdev *hwdev);
+
+void hinic3_set_api_stop(struct hinic3_hwdev *hwdev);
+
+#endif
diff --git a/drivers/net/ethernet/huawei/hinic3/hinic3_hwif.c b/drivers/net/ethernet/huawei/hinic3/hinic3_hwif.c
new file mode 100644
index 000000000000..f76f140fb6f7
--- /dev/null
+++ b/drivers/net/ethernet/huawei/hinic3/hinic3_hwif.c
@@ -0,0 +1,436 @@
+// SPDX-License-Identifier: GPL-2.0
+// Copyright (c) Huawei Technologies Co., Ltd. 2025. All rights reserved.
+
+#include <linux/bitfield.h>
+#include <linux/device.h>
+#include <linux/io.h>
+
+#include "hinic3_common.h"
+#include "hinic3_csr.h"
+#include "hinic3_hwdev.h"
+#include "hinic3_hwif.h"
+
+#define HINIC3_HWIF_READY_TIMEOUT          10000
+#define HINIC3_DB_AND_OUTBOUND_EN_TIMEOUT  60000
+#define HINIC3_PCIE_LINK_DOWN              0xFFFFFFFF
+
+/* config BAR4/5 4MB, DB & DWQE both 2MB */
+#define HINIC3_DB_DWQE_SIZE    0x00400000
+
+/* db/dwqe page size: 4K */
+#define HINIC3_DB_PAGE_SIZE    0x00001000
+#define HINIC3_DWQE_OFFSET     0x00000800
+#define HINIC3_DB_MAX_AREAS    (HINIC3_DB_DWQE_SIZE / HINIC3_DB_PAGE_SIZE)
+
+#define HINIC3_MAX_MSIX_ENTRY  2048
+
+#define HINIC3_AF0_FUNC_GLOBAL_IDX_MASK  GENMASK(11, 0)
+#define HINIC3_AF0_P2P_IDX_MASK          GENMASK(16, 12)
+#define HINIC3_AF0_PCI_INTF_IDX_MASK     GENMASK(19, 17)
+#define HINIC3_AF0_FUNC_TYPE_MASK        BIT(28)
+#define HINIC3_AF0_GET(val, member) \
+	FIELD_GET(HINIC3_AF0_##member##_MASK, val)
+
+#define HINIC3_AF1_AEQS_PER_FUNC_MASK     GENMASK(9, 8)
+#define HINIC3_AF1_MGMT_INIT_STATUS_MASK  BIT(30)
+#define HINIC3_AF1_GET(val, member) \
+	FIELD_GET(HINIC3_AF1_##member##_MASK, val)
+
+#define HINIC3_AF2_CEQS_PER_FUNC_MASK      GENMASK(8, 0)
+#define HINIC3_AF2_IRQS_PER_FUNC_MASK      GENMASK(26, 16)
+#define HINIC3_AF2_GET(val, member) \
+	FIELD_GET(HINIC3_AF2_##member##_MASK, val)
+
+#define HINIC3_AF4_DOORBELL_CTRL_MASK  BIT(0)
+#define HINIC3_AF4_GET(val, member) \
+	FIELD_GET(HINIC3_AF4_##member##_MASK, val)
+#define HINIC3_AF4_SET(val, member) \
+	FIELD_PREP(HINIC3_AF4_##member##_MASK, val)
+
+#define HINIC3_AF5_OUTBOUND_CTRL_MASK  BIT(0)
+#define HINIC3_AF5_GET(val, member) \
+	FIELD_GET(HINIC3_AF5_##member##_MASK, val)
+
+#define HINIC3_AF6_PF_STATUS_MASK     GENMASK(15, 0)
+#define HINIC3_AF6_FUNC_MAX_SQ_MASK   GENMASK(31, 23)
+#define HINIC3_AF6_MSIX_FLEX_EN_MASK  BIT(22)
+#define HINIC3_AF6_GET(val, member) \
+	FIELD_GET(HINIC3_AF6_##member##_MASK, val)
+
+#define HINIC3_GET_REG_ADDR(reg)  ((reg) & (HINIC3_REGS_FLAG_MASK))
+
+static void __iomem *hinic3_reg_addr(struct hinic3_hwif *hwif, u32 reg)
+{
+	return hwif->cfg_regs_base + HINIC3_GET_REG_ADDR(reg);
+}
+
+u32 hinic3_hwif_read_reg(struct hinic3_hwif *hwif, u32 reg)
+{
+	void __iomem *addr = hinic3_reg_addr(hwif, reg);
+
+	return ioread32be(addr);
+}
+
+void hinic3_hwif_write_reg(struct hinic3_hwif *hwif, u32 reg, u32 val)
+{
+	void __iomem *addr = hinic3_reg_addr(hwif, reg);
+
+	iowrite32be(val, addr);
+}
+
+static enum hinic3_wait_return check_hwif_ready_handler(void *priv_data)
+{
+	struct hinic3_hwdev *hwdev = priv_data;
+	u32 attr1;
+
+	attr1 = hinic3_hwif_read_reg(hwdev->hwif, HINIC3_CSR_FUNC_ATTR1_ADDR);
+
+	return HINIC3_AF1_GET(attr1, MGMT_INIT_STATUS) ?
+	       HINIC3_WAIT_PROCESS_CPL : HINIC3_WAIT_PROCESS_WAITING;
+}
+
+static int wait_hwif_ready(struct hinic3_hwdev *hwdev)
+{
+	return hinic3_wait_for_timeout(hwdev, check_hwif_ready_handler,
+				       HINIC3_HWIF_READY_TIMEOUT,
+				       USEC_PER_MSEC);
+}
+
+/* Set attr struct from HW attr values. */
+static void set_hwif_attr(struct hinic3_func_attr *attr, u32 attr0, u32 attr1,
+			  u32 attr2, u32 attr3, u32 attr6)
+{
+	attr->func_global_idx = HINIC3_AF0_GET(attr0, FUNC_GLOBAL_IDX);
+	attr->port_to_port_idx = HINIC3_AF0_GET(attr0, P2P_IDX);
+	attr->pci_intf_idx = HINIC3_AF0_GET(attr0, PCI_INTF_IDX);
+	attr->func_type = HINIC3_AF0_GET(attr0, FUNC_TYPE);
+
+	attr->num_aeqs = BIT(HINIC3_AF1_GET(attr1, AEQS_PER_FUNC));
+	attr->num_ceqs = HINIC3_AF2_GET(attr2, CEQS_PER_FUNC);
+	attr->num_irqs = HINIC3_AF2_GET(attr2, IRQS_PER_FUNC);
+	if (attr->num_irqs > HINIC3_MAX_MSIX_ENTRY)
+		attr->num_irqs = HINIC3_MAX_MSIX_ENTRY;
+
+	attr->num_sq = HINIC3_AF6_GET(attr6, FUNC_MAX_SQ);
+	attr->msix_flex_en = HINIC3_AF6_GET(attr6, MSIX_FLEX_EN);
+}
+
+/* Read attributes from HW and set attribute struct. */
+static int init_hwif_attr(struct hinic3_hwdev *hwdev)
+{
+	u32 attr0, attr1, attr2, attr3, attr6;
+	struct hinic3_hwif *hwif;
+
+	hwif = hwdev->hwif;
+	attr0 = hinic3_hwif_read_reg(hwif, HINIC3_CSR_FUNC_ATTR0_ADDR);
+	if (attr0 == HINIC3_PCIE_LINK_DOWN)
+		return -EFAULT;
+
+	attr1 = hinic3_hwif_read_reg(hwif, HINIC3_CSR_FUNC_ATTR1_ADDR);
+	if (attr1 == HINIC3_PCIE_LINK_DOWN)
+		return -EFAULT;
+
+	attr2 = hinic3_hwif_read_reg(hwif, HINIC3_CSR_FUNC_ATTR2_ADDR);
+	if (attr2 == HINIC3_PCIE_LINK_DOWN)
+		return -EFAULT;
+
+	attr3 = hinic3_hwif_read_reg(hwif, HINIC3_CSR_FUNC_ATTR3_ADDR);
+	if (attr3 == HINIC3_PCIE_LINK_DOWN)
+		return -EFAULT;
+
+	attr6 = hinic3_hwif_read_reg(hwif, HINIC3_CSR_FUNC_ATTR6_ADDR);
+	if (attr6 == HINIC3_PCIE_LINK_DOWN)
+		return -EFAULT;
+
+	set_hwif_attr(&hwif->attr, attr0, attr1, attr2, attr3, attr6);
+
+	if (!hwif->attr.num_ceqs) {
+		dev_err(hwdev->dev, "Ceq num cfg in fw is zero\n");
+		return -EFAULT;
+	}
+
+	if (!hwif->attr.num_irqs) {
+		dev_err(hwdev->dev,
+			"Irq num cfg in fw is zero, msix_flex_en %d\n",
+			hwif->attr.msix_flex_en);
+		return -EFAULT;
+	}
+
+	return 0;
+}
+
+static enum hinic3_doorbell_ctrl hinic3_get_doorbell_ctrl_status(struct hinic3_hwif *hwif)
+{
+	u32 attr4 = hinic3_hwif_read_reg(hwif, HINIC3_CSR_FUNC_ATTR4_ADDR);
+
+	return HINIC3_AF4_GET(attr4, DOORBELL_CTRL);
+}
+
+static enum hinic3_outbound_ctrl hinic3_get_outbound_ctrl_status(struct hinic3_hwif *hwif)
+{
+	u32 attr5 = hinic3_hwif_read_reg(hwif, HINIC3_CSR_FUNC_ATTR5_ADDR);
+
+	return HINIC3_AF5_GET(attr5, OUTBOUND_CTRL);
+}
+
+void hinic3_toggle_doorbell(struct hinic3_hwif *hwif,
+			    enum hinic3_doorbell_ctrl flag)
+{
+	u32 addr, attr4;
+
+	addr = HINIC3_CSR_FUNC_ATTR4_ADDR;
+	attr4 = hinic3_hwif_read_reg(hwif, addr);
+
+	attr4 &= ~HINIC3_AF4_DOORBELL_CTRL_MASK;
+	attr4 |= HINIC3_AF4_SET(flag, DOORBELL_CTRL);
+
+	hinic3_hwif_write_reg(hwif, addr, attr4);
+}
+
+static int db_area_idx_init(struct hinic3_hwif *hwif, u64 db_base_phy,
+			    u8 __iomem *db_base, u64 db_dwqe_len)
+{
+	struct hinic3_db_area *db_area = &hwif->db_area;
+	u32 db_max_areas;
+
+	hwif->db_base_phy = db_base_phy;
+	hwif->db_base = db_base;
+	hwif->db_dwqe_len = db_dwqe_len;
+
+	db_max_areas = db_dwqe_len > HINIC3_DB_DWQE_SIZE ?
+		       HINIC3_DB_MAX_AREAS : db_dwqe_len / HINIC3_DB_PAGE_SIZE;
+	db_area->db_bitmap_array = bitmap_zalloc(db_max_areas, GFP_KERNEL);
+	if (!db_area->db_bitmap_array)
+		return -ENOMEM;
+
+	db_area->db_max_areas = db_max_areas;
+	spin_lock_init(&db_area->idx_lock);
+
+	return 0;
+}
+
+static void db_area_idx_free(struct hinic3_db_area *db_area)
+{
+	bitmap_free(db_area->db_bitmap_array);
+}
+
+static int get_db_idx(struct hinic3_hwif *hwif, u32 *idx)
+{
+	struct hinic3_db_area *db_area = &hwif->db_area;
+	u32 pg_idx;
+
+	spin_lock(&db_area->idx_lock);
+	pg_idx = find_first_zero_bit(db_area->db_bitmap_array,
+				     db_area->db_max_areas);
+	if (pg_idx == db_area->db_max_areas) {
+		spin_unlock(&db_area->idx_lock);
+		return -ENOMEM;
+	}
+	set_bit(pg_idx, db_area->db_bitmap_array);
+	spin_unlock(&db_area->idx_lock);
+
+	*idx = pg_idx;
+
+	return 0;
+}
+
+static void free_db_idx(struct hinic3_hwif *hwif, u32 idx)
+{
+	struct hinic3_db_area *db_area = &hwif->db_area;
+
+	spin_lock(&db_area->idx_lock);
+	clear_bit(idx, db_area->db_bitmap_array);
+	spin_unlock(&db_area->idx_lock);
+}
+
+void hinic3_free_db_addr(struct hinic3_hwdev *hwdev, const u8 __iomem *db_base)
+{
+	struct hinic3_hwif *hwif;
+	uintptr_t distance;
+	u32 idx;
+
+	hwif = hwdev->hwif;
+	distance = db_base - hwif->db_base;
+	idx = distance / HINIC3_DB_PAGE_SIZE;
+
+	free_db_idx(hwif, idx);
+}
+
+int hinic3_alloc_db_addr(struct hinic3_hwdev *hwdev, void __iomem **db_base,
+			 void __iomem **dwqe_base)
+{
+	struct hinic3_hwif *hwif;
+	u8 __iomem *addr;
+	u32 idx;
+	int err;
+
+	hwif = hwdev->hwif;
+
+	err = get_db_idx(hwif, &idx);
+	if (err)
+		return err;
+
+	addr = hwif->db_base + idx * HINIC3_DB_PAGE_SIZE;
+	*db_base = addr;
+
+	if (dwqe_base)
+		*dwqe_base = addr + HINIC3_DWQE_OFFSET;
+
+	return 0;
+}
+
+void hinic3_set_msix_state(struct hinic3_hwdev *hwdev, u16 msix_idx,
+			   enum hinic3_msix_state flag)
+{
+	struct hinic3_hwif *hwif;
+	u8 int_msk = 1;
+	u32 mask_bits;
+	u32 addr;
+
+	hwif = hwdev->hwif;
+
+	if (flag)
+		mask_bits = HINIC3_MSI_CLR_INDIR_SET(int_msk, INT_MSK_SET);
+	else
+		mask_bits = HINIC3_MSI_CLR_INDIR_SET(int_msk, INT_MSK_CLR);
+	mask_bits = mask_bits |
+		    HINIC3_MSI_CLR_INDIR_SET(msix_idx, SIMPLE_INDIR_IDX);
+
+	addr = HINIC3_CSR_FUNC_MSI_CLR_WR_ADDR;
+	hinic3_hwif_write_reg(hwif, addr, mask_bits);
+}
+
+static void disable_all_msix(struct hinic3_hwdev *hwdev)
+{
+	u16 num_irqs = hwdev->hwif->attr.num_irqs;
+	u16 i;
+
+	for (i = 0; i < num_irqs; i++)
+		hinic3_set_msix_state(hwdev, i, HINIC3_MSIX_DISABLE);
+}
+
+void hinic3_msix_intr_clear_resend_bit(struct hinic3_hwdev *hwdev, u16 msix_idx,
+				       u8 clear_resend_en)
+{
+	struct hinic3_hwif *hwif;
+	u32 msix_ctrl, addr;
+
+	hwif = hwdev->hwif;
+
+	msix_ctrl = HINIC3_MSI_CLR_INDIR_SET(msix_idx, SIMPLE_INDIR_IDX) |
+		    HINIC3_MSI_CLR_INDIR_SET(clear_resend_en, RESEND_TIMER_CLR);
+
+	addr = HINIC3_CSR_FUNC_MSI_CLR_WR_ADDR;
+	hinic3_hwif_write_reg(hwif, addr, msix_ctrl);
+}
+
+void hinic3_set_msix_auto_mask_state(struct hinic3_hwdev *hwdev, u16 msix_idx,
+				     enum hinic3_msix_auto_mask flag)
+{
+	struct hinic3_hwif *hwif;
+	u32 mask_bits;
+	u32 addr;
+
+	hwif = hwdev->hwif;
+
+	if (flag)
+		mask_bits = HINIC3_MSI_CLR_INDIR_SET(1, AUTO_MSK_SET);
+	else
+		mask_bits = HINIC3_MSI_CLR_INDIR_SET(1, AUTO_MSK_CLR);
+
+	mask_bits = mask_bits |
+		    HINIC3_MSI_CLR_INDIR_SET(msix_idx, SIMPLE_INDIR_IDX);
+
+	addr = HINIC3_CSR_FUNC_MSI_CLR_WR_ADDR;
+	hinic3_hwif_write_reg(hwif, addr, mask_bits);
+}
+
+static enum hinic3_wait_return check_db_outbound_enable_handler(void *priv_data)
+{
+	enum hinic3_outbound_ctrl outbound_ctrl;
+	struct hinic3_hwif *hwif = priv_data;
+	enum hinic3_doorbell_ctrl db_ctrl;
+
+	db_ctrl = hinic3_get_doorbell_ctrl_status(hwif);
+	outbound_ctrl = hinic3_get_outbound_ctrl_status(hwif);
+	if (outbound_ctrl == ENABLE_OUTBOUND && db_ctrl == ENABLE_DOORBELL)
+		return HINIC3_WAIT_PROCESS_CPL;
+
+	return HINIC3_WAIT_PROCESS_WAITING;
+}
+
+static int wait_until_doorbell_and_outbound_enabled(struct hinic3_hwif *hwif)
+{
+	return hinic3_wait_for_timeout(hwif, check_db_outbound_enable_handler,
+				       HINIC3_DB_AND_OUTBOUND_EN_TIMEOUT,
+				       USEC_PER_MSEC);
+}
+
+int hinic3_init_hwif(struct hinic3_hwdev *hwdev)
+{
+	struct hinic3_pcidev *pci_adapter = hwdev->adapter;
+	struct hinic3_hwif *hwif;
+	u32 attr1, attr4, attr5;
+	int err;
+
+	hwif = kzalloc(sizeof(*hwif), GFP_KERNEL);
+	if (!hwif)
+		return -ENOMEM;
+
+	hwdev->hwif = hwif;
+	hwif->cfg_regs_base = (u8 __iomem *)pci_adapter->cfg_reg_base +
+			      HINIC3_VF_CFG_REG_OFFSET;
+
+	err = db_area_idx_init(hwif, pci_adapter->db_base_phy,
+			       pci_adapter->db_base,
+			       pci_adapter->db_dwqe_len);
+	if (err) {
+		dev_err(hwdev->dev, "Failed to init db area.\n");
+		goto err_free_hwif;
+	}
+
+	err = wait_hwif_ready(hwdev);
+	if (err) {
+		attr1 = hinic3_hwif_read_reg(hwif, HINIC3_CSR_FUNC_ATTR1_ADDR);
+		dev_err(hwdev->dev, "Chip status is not ready, attr1:0x%x\n",
+			attr1);
+		goto err_free_db_area_idx;
+	}
+
+	err = init_hwif_attr(hwdev);
+	if (err) {
+		dev_err(hwdev->dev, "Init hwif attr failed\n");
+		goto err_free_db_area_idx;
+	}
+
+	err = wait_until_doorbell_and_outbound_enabled(hwif);
+	if (err) {
+		attr4 = hinic3_hwif_read_reg(hwif, HINIC3_CSR_FUNC_ATTR4_ADDR);
+		attr5 = hinic3_hwif_read_reg(hwif, HINIC3_CSR_FUNC_ATTR5_ADDR);
+		dev_err(hwdev->dev, "HW doorbell/outbound is disabled, attr4 0x%x attr5 0x%x\n",
+			attr4, attr5);
+		goto err_free_db_area_idx;
+	}
+
+	disable_all_msix(hwdev);
+
+	return 0;
+
+err_free_db_area_idx:
+	db_area_idx_free(&hwif->db_area);
+err_free_hwif:
+	kfree(hwif);
+
+	return err;
+}
+
+void hinic3_free_hwif(struct hinic3_hwdev *hwdev)
+{
+	db_area_idx_free(&hwdev->hwif->db_area);
+	kfree(hwdev->hwif);
+}
+
+u16 hinic3_global_func_id(struct hinic3_hwdev *hwdev)
+{
+	return hwdev->hwif->attr.func_global_idx;
+}
diff --git a/drivers/net/ethernet/huawei/hinic3/hinic3_hwif.h b/drivers/net/ethernet/huawei/hinic3/hinic3_hwif.h
new file mode 100644
index 000000000000..c02904e861cc
--- /dev/null
+++ b/drivers/net/ethernet/huawei/hinic3/hinic3_hwif.h
@@ -0,0 +1,90 @@
+/* SPDX-License-Identifier: GPL-2.0 */
+/* Copyright (c) Huawei Technologies Co., Ltd. 2025. All rights reserved. */
+
+#ifndef _HINIC3_HWIF_H_
+#define _HINIC3_HWIF_H_
+
+#include <linux/build_bug.h>
+#include <linux/spinlock_types.h>
+
+struct hinic3_hwdev;
+
+enum hinic3_func_type {
+	HINIC3_FUNC_TYPE_VF = 1,
+};
+
+struct hinic3_db_area {
+	unsigned long *db_bitmap_array;
+	u32           db_max_areas;
+	/* protect doorbell area alloc and free */
+	spinlock_t    idx_lock;
+};
+
+struct hinic3_func_attr {
+	enum hinic3_func_type func_type;
+	u16                   func_global_idx;
+	u16                   global_vf_id_of_pf;
+	u16                   num_irqs;
+	u16                   num_sq;
+	u8                    port_to_port_idx;
+	u8                    pci_intf_idx;
+	u8                    ppf_idx;
+	u8                    num_aeqs;
+	u8                    num_ceqs;
+	u8                    msix_flex_en;
+};
+
+static_assert(sizeof(struct hinic3_func_attr) == 20);
+
+struct hinic3_hwif {
+	u8 __iomem              *cfg_regs_base;
+	u64                     db_base_phy;
+	u64                     db_dwqe_len;
+	u8 __iomem              *db_base;
+	struct hinic3_db_area   db_area;
+	struct hinic3_func_attr attr;
+};
+
+enum hinic3_outbound_ctrl {
+	ENABLE_OUTBOUND  = 0x0,
+	DISABLE_OUTBOUND = 0x1,
+};
+
+enum hinic3_doorbell_ctrl {
+	ENABLE_DOORBELL  = 0,
+	DISABLE_DOORBELL = 1,
+};
+
+enum hinic3_msix_state {
+	HINIC3_MSIX_ENABLE,
+	HINIC3_MSIX_DISABLE,
+};
+
+enum hinic3_msix_auto_mask {
+	HINIC3_CLR_MSIX_AUTO_MASK,
+	HINIC3_SET_MSIX_AUTO_MASK,
+};
+
+u32 hinic3_hwif_read_reg(struct hinic3_hwif *hwif, u32 reg);
+void hinic3_hwif_write_reg(struct hinic3_hwif *hwif, u32 reg, u32 val);
+
+void hinic3_toggle_doorbell(struct hinic3_hwif *hwif,
+			    enum hinic3_doorbell_ctrl flag);
+
+int hinic3_alloc_db_addr(struct hinic3_hwdev *hwdev, void __iomem **db_base,
+			 void __iomem **dwqe_base);
+void hinic3_free_db_addr(struct hinic3_hwdev *hwdev, const u8 __iomem *db_base);
+
+int hinic3_init_hwif(struct hinic3_hwdev *hwdev);
+void hinic3_free_hwif(struct hinic3_hwdev *hwdev);
+
+void hinic3_set_msix_state(struct hinic3_hwdev *hwdev, u16 msix_idx,
+			   enum hinic3_msix_state flag);
+void hinic3_msix_intr_clear_resend_bit(struct hinic3_hwdev *hwdev, u16 msix_idx,
+				       u8 clear_resend_en);
+void hinic3_set_msix_auto_mask_state(struct hinic3_hwdev *hwdev, u16 msix_idx,
+				     enum hinic3_msix_auto_mask flag);
+
+u16 hinic3_global_func_id(struct hinic3_hwdev *hwdev);
+
+#endif
diff --git a/drivers/net/ethernet/huawei/hinic3/hinic3_irq.c b/drivers/net/ethernet/huawei/hinic3/hinic3_irq.c
new file mode 100644
index 000000000000..a69b361225e9
--- /dev/null
+++ b/drivers/net/ethernet/huawei/hinic3/hinic3_irq.c
@@ -0,0 +1,194 @@
+// SPDX-License-Identifier: GPL-2.0
+// Copyright (c) Huawei Technologies Co., Ltd. 2025. All rights reserved.
+
+#include <linux/netdevice.h>
+
+#include "hinic3_hw_comm.h"
+#include "hinic3_hwdev.h"
+#include "hinic3_hwif.h"
+#include "hinic3_nic_dev.h"
+#include "hinic3_rx.h"
+#include "hinic3_tx.h"
+
+static int hinic3_poll(struct napi_struct *napi, int budget)
+{
+	struct hinic3_irq_cfg *irq_cfg =
+		container_of(napi, struct hinic3_irq_cfg, napi);
+	struct hinic3_nic_dev *nic_dev;
+	bool busy = false;
+	int work_done;
+
+	nic_dev = netdev_priv(irq_cfg->netdev);
+
+	busy |= hinic3_tx_poll(irq_cfg->txq, budget);
+
+	if (unlikely(!budget))
+		return 0;
+
+	work_done = hinic3_rx_poll(irq_cfg->rxq, budget);
+	busy |= work_done >= budget;
+
+	if (busy)
+		return budget;
+
+	if (likely(napi_complete_done(napi, work_done)))
+		hinic3_set_msix_state(nic_dev->hwdev, irq_cfg->msix_entry_idx,
+				      HINIC3_MSIX_ENABLE);
+
+	return work_done;
+}
+
+static void qp_add_napi(struct hinic3_irq_cfg *irq_cfg)
+{
+	struct hinic3_nic_dev *nic_dev = netdev_priv(irq_cfg->netdev);
+
+	netif_napi_add(nic_dev->netdev, &irq_cfg->napi, hinic3_poll);
+	netif_queue_set_napi(irq_cfg->netdev, irq_cfg->irq_id,
+			     NETDEV_QUEUE_TYPE_RX, &irq_cfg->napi);
+	netif_queue_set_napi(irq_cfg->netdev, irq_cfg->irq_id,
+			     NETDEV_QUEUE_TYPE_TX, &irq_cfg->napi);
+	napi_enable(&irq_cfg->napi);
+}
+
+static void qp_del_napi(struct hinic3_irq_cfg *irq_cfg)
+{
+	napi_disable(&irq_cfg->napi);
+	netif_queue_set_napi(irq_cfg->netdev, irq_cfg->irq_id,
+			     NETDEV_QUEUE_TYPE_RX, NULL);
+	netif_queue_set_napi(irq_cfg->netdev, irq_cfg->irq_id,
+			     NETDEV_QUEUE_TYPE_TX, NULL);
+	netif_stop_subqueue(irq_cfg->netdev, irq_cfg->irq_id);
+	netif_napi_del(&irq_cfg->napi);
+}
+
+static irqreturn_t qp_irq(int irq, void *data)
+{
+	struct hinic3_irq_cfg *irq_cfg = data;
+	struct hinic3_nic_dev *nic_dev;
+
+	nic_dev = netdev_priv(irq_cfg->netdev);
+	hinic3_msix_intr_clear_resend_bit(nic_dev->hwdev,
+					  irq_cfg->msix_entry_idx, 1);
+
+	napi_schedule(&irq_cfg->napi);
+
+	return IRQ_HANDLED;
+}
+
+static int hinic3_request_irq(struct hinic3_irq_cfg *irq_cfg, u16 q_id)
+{
+	struct hinic3_interrupt_info info = {};
+	struct hinic3_nic_dev *nic_dev;
+	struct net_device *netdev;
+	int err;
+
+	netdev = irq_cfg->netdev;
+	nic_dev = netdev_priv(netdev);
+	qp_add_napi(irq_cfg);
+
+	info.msix_index = irq_cfg->msix_entry_idx;
+	info.interrupt_coalesc_set = 1;
+	info.pending_limit = nic_dev->intr_coalesce[q_id].pending_limit;
+	info.coalesc_timer_cfg =
+		nic_dev->intr_coalesce[q_id].coalesce_timer_cfg;
+	info.resend_timer_cfg = nic_dev->intr_coalesce[q_id].resend_timer_cfg;
+	err = hinic3_set_interrupt_cfg_direct(nic_dev->hwdev, &info);
+	if (err) {
+		netdev_err(netdev, "Failed to set RX interrupt coalescing attribute.\n");
+		qp_del_napi(irq_cfg);
+		return err;
+	}
+
+	err = request_irq(irq_cfg->irq_id, qp_irq, 0, irq_cfg->irq_name,
+			  irq_cfg);
+	if (err) {
+		qp_del_napi(irq_cfg);
+		return err;
+	}
+
+	irq_set_affinity_hint(irq_cfg->irq_id, &irq_cfg->affinity_mask);
+
+	return 0;
+}
+
+static void hinic3_release_irq(struct hinic3_irq_cfg *irq_cfg)
+{
+	irq_set_affinity_hint(irq_cfg->irq_id, NULL);
+	free_irq(irq_cfg->irq_id, irq_cfg);
+}
+
+int hinic3_qps_irq_init(struct net_device *netdev)
+{
+	struct hinic3_nic_dev *nic_dev = netdev_priv(netdev);
+	struct pci_dev *pdev = nic_dev->pdev;
+	struct hinic3_irq_cfg *irq_cfg;
+	struct msix_entry *msix_entry;
+	u32 local_cpu;
+	u16 q_id;
+	int err;
+
+	for (q_id = 0; q_id < nic_dev->q_params.num_qps; q_id++) {
+		msix_entry = &nic_dev->qps_msix_entries[q_id];
+		irq_cfg = &nic_dev->q_params.irq_cfg[q_id];
+
+		irq_cfg->irq_id = msix_entry->vector;
+		irq_cfg->msix_entry_idx = msix_entry->entry;
+		irq_cfg->netdev = netdev;
+		irq_cfg->txq = &nic_dev->txqs[q_id];
+		irq_cfg->rxq = &nic_dev->rxqs[q_id];
+		nic_dev->rxqs[q_id].irq_cfg = irq_cfg;
+
+		local_cpu = cpumask_local_spread(q_id, dev_to_node(&pdev->dev));
+		cpumask_set_cpu(local_cpu, &irq_cfg->affinity_mask);
+
+		snprintf(irq_cfg->irq_name, sizeof(irq_cfg->irq_name),
+			 "%s_qp%u", netdev->name, q_id);
+
+		err = hinic3_request_irq(irq_cfg, q_id);
+		if (err) {
+			netdev_err(netdev, "Failed to request Rx irq\n");
+			goto err_release_irqs;
+		}
+
+		hinic3_set_msix_auto_mask_state(nic_dev->hwdev,
+						irq_cfg->msix_entry_idx,
+						HINIC3_SET_MSIX_AUTO_MASK);
+		hinic3_set_msix_state(nic_dev->hwdev, irq_cfg->msix_entry_idx,
+				      HINIC3_MSIX_ENABLE);
+	}
+
+	return 0;
+
+err_release_irqs:
+	while (q_id > 0) {
+		q_id--;
+		irq_cfg = &nic_dev->q_params.irq_cfg[q_id];
+		qp_del_napi(irq_cfg);
+		hinic3_set_msix_state(nic_dev->hwdev, irq_cfg->msix_entry_idx,
+				      HINIC3_MSIX_DISABLE);
+		hinic3_set_msix_auto_mask_state(nic_dev->hwdev,
+						irq_cfg->msix_entry_idx,
+						HINIC3_CLR_MSIX_AUTO_MASK);
+		hinic3_release_irq(irq_cfg);
+	}
+
+	return err;
+}
+
+void hinic3_qps_irq_uninit(struct net_device *netdev)
+{
+	struct hinic3_nic_dev *nic_dev = netdev_priv(netdev);
+	struct hinic3_irq_cfg *irq_cfg;
+	u16 q_id;
+
+	for (q_id = 0; q_id < nic_dev->q_params.num_qps; q_id++) {
+		irq_cfg = &nic_dev->q_params.irq_cfg[q_id];
+		qp_del_napi(irq_cfg);
+		hinic3_set_msix_state(nic_dev->hwdev, irq_cfg->msix_entry_idx,
+				      HINIC3_MSIX_DISABLE);
+		hinic3_set_msix_auto_mask_state(nic_dev->hwdev,
+						irq_cfg->msix_entry_idx,
+						HINIC3_CLR_MSIX_AUTO_MASK);
+		hinic3_release_irq(irq_cfg);
+	}
+}
diff --git a/drivers/net/ethernet/huawei/hinic3/hinic3_lld.c b/drivers/net/ethernet/huawei/hinic3/hinic3_lld.c
new file mode 100644
index 000000000000..3db8241a3b0c
--- /dev/null
+++ b/drivers/net/ethernet/huawei/hinic3/hinic3_lld.c
@@ -0,0 +1,421 @@
+// SPDX-License-Identifier: GPL-2.0
+// Copyright (c) Huawei Technologies Co., Ltd. 2025. All rights reserved.
+
+#include <linux/delay.h>
+#include <linux/iopoll.h>
+
+#include "hinic3_hw_cfg.h"
+#include "hinic3_hwdev.h"
+#include "hinic3_lld.h"
+#include "hinic3_mgmt.h"
+#include "hinic3_pci_id_tbl.h"
+
+#define HINIC3_VF_PCI_CFG_REG_BAR  0
+#define HINIC3_PCI_INTR_REG_BAR    2
+#define HINIC3_PCI_DB_BAR          4
+
+#define HINIC3_EVENT_POLL_SLEEP_US   1000
+#define HINIC3_EVENT_POLL_TIMEOUT_US 10000000
+
+static struct hinic3_adev_device {
+	const char *name;
+} hinic3_adev_devices[HINIC3_SERVICE_T_MAX] = {
+	[HINIC3_SERVICE_T_NIC] = {
+		.name = "nic",
+	},
+};
+
+static bool hinic3_adev_svc_supported(struct hinic3_hwdev *hwdev,
+				      enum hinic3_service_type svc_type)
+{
+	switch (svc_type) {
+	case HINIC3_SERVICE_T_NIC:
+		return hinic3_support_nic(hwdev);
+	default:
+		break;
+	}
+
+	return false;
+}
+
+static void hinic3_comm_adev_release(struct device *dev)
+{
+	struct hinic3_adev *hadev = container_of(dev, struct hinic3_adev,
+						 adev.dev);
+
+	kfree(hadev);
+}
+
+static struct hinic3_adev *hinic3_add_one_adev(struct hinic3_hwdev *hwdev,
+					       enum hinic3_service_type svc_type)
+{
+	struct hinic3_adev *hadev;
+	const char *svc_name;
+	int ret;
+
+	hadev = kzalloc(sizeof(*hadev), GFP_KERNEL);
+	if (!hadev)
+		return NULL;
+
+	svc_name = hinic3_adev_devices[svc_type].name;
+	hadev->adev.name = svc_name;
+	hadev->adev.id = hwdev->dev_id;
+	hadev->adev.dev.parent = hwdev->dev;
+	hadev->adev.dev.release = hinic3_comm_adev_release;
+	hadev->svc_type = svc_type;
+	hadev->hwdev = hwdev;
+
+	ret = auxiliary_device_init(&hadev->adev);
+	if (ret) {
+		dev_err(hwdev->dev, "failed init adev %s %u\n",
+			svc_name, hwdev->dev_id);
+		kfree(hadev);
+		return NULL;
+	}
+
+	ret = auxiliary_device_add(&hadev->adev);
+	if (ret) {
+		dev_err(hwdev->dev, "failed to add adev %s %u\n",
+			svc_name, hwdev->dev_id);
+		auxiliary_device_uninit(&hadev->adev);
+		return NULL;
+	}
+
+	return hadev;
+}
+
+static void hinic3_del_one_adev(struct hinic3_hwdev *hwdev,
+				enum hinic3_service_type svc_type)
+{
+	struct hinic3_pcidev *pci_adapter = hwdev->adapter;
+	struct hinic3_adev *hadev;
+	int timeout;
+	bool state;
+
+	timeout = read_poll_timeout(test_and_set_bit, state, !state,
+				    HINIC3_EVENT_POLL_SLEEP_US,
+				    HINIC3_EVENT_POLL_TIMEOUT_US,
+				    false, svc_type, &pci_adapter->state);
+
+	hadev = pci_adapter->hadev[svc_type];
+	auxiliary_device_delete(&hadev->adev);
+	auxiliary_device_uninit(&hadev->adev);
+	pci_adapter->hadev[svc_type] = NULL;
+	if (!timeout)
+		clear_bit(svc_type, &pci_adapter->state);
+}
+
+static int hinic3_attach_aux_devices(struct hinic3_hwdev *hwdev)
+{
+	struct hinic3_pcidev *pci_adapter = hwdev->adapter;
+	enum hinic3_service_type svc_type;
+
+	mutex_lock(&pci_adapter->pdev_mutex);
+
+	for (svc_type = 0; svc_type < HINIC3_SERVICE_T_MAX; svc_type++) {
+		if (!hinic3_adev_svc_supported(hwdev, svc_type))
+			continue;
+
+		pci_adapter->hadev[svc_type] = hinic3_add_one_adev(hwdev,
+								   svc_type);
+		if (!pci_adapter->hadev[svc_type])
+			goto err_del_adevs;
+	}
+	mutex_unlock(&pci_adapter->pdev_mutex);
+
+	return 0;
+
+err_del_adevs:
+	while (svc_type > 0) {
+		svc_type--;
+		if (pci_adapter->hadev[svc_type]) {
+			hinic3_del_one_adev(hwdev, svc_type);
+			pci_adapter->hadev[svc_type] = NULL;
+		}
+	}
+	mutex_unlock(&pci_adapter->pdev_mutex);
+
+	return -ENOMEM;
+}
+
+static void hinic3_detach_aux_devices(struct hinic3_hwdev *hwdev)
+{
+	struct hinic3_pcidev *pci_adapter = hwdev->adapter;
+	int i;
+
+	mutex_lock(&pci_adapter->pdev_mutex);
+	for (i = 0; i < ARRAY_SIZE(hinic3_adev_devices); i++) {
+		if (pci_adapter->hadev[i])
+			hinic3_del_one_adev(hwdev, i);
+	}
+	mutex_unlock(&pci_adapter->pdev_mutex);
+}
+
+struct hinic3_hwdev *hinic3_adev_get_hwdev(struct auxiliary_device *adev)
+{
+	struct hinic3_adev *hadev;
+
+	hadev = container_of(adev, struct hinic3_adev, adev);
+
+	return hadev->hwdev;
+}
+
+void hinic3_adev_event_register(struct auxiliary_device *adev,
+				void (*event_handler)(struct auxiliary_device *adev,
+						      struct hinic3_event_info *event))
+{
+	struct hinic3_adev *hadev;
+
+	hadev = container_of(adev, struct hinic3_adev, adev);
+	hadev->event = event_handler;
+}
+
+void hinic3_adev_event_unregister(struct auxiliary_device *adev)
+{
+	struct hinic3_adev *hadev;
+
+	hadev = container_of(adev, struct hinic3_adev, adev);
+	hadev->event = NULL;
+}
+
+static int hinic3_mapping_bar(struct pci_dev *pdev,
+			      struct hinic3_pcidev *pci_adapter)
+{
+	pci_adapter->cfg_reg_base = pci_ioremap_bar(pdev,
+						    HINIC3_VF_PCI_CFG_REG_BAR);
+	if (!pci_adapter->cfg_reg_base) {
+		dev_err(&pdev->dev, "Failed to map configuration regs\n");
+		return -ENOMEM;
+	}
+
+	pci_adapter->intr_reg_base = pci_ioremap_bar(pdev,
+						     HINIC3_PCI_INTR_REG_BAR);
+	if (!pci_adapter->intr_reg_base) {
+		dev_err(&pdev->dev, "Failed to map interrupt regs\n");
+		goto err_unmap_cfg_reg_base;
+	}
+
+	pci_adapter->db_base_phy = pci_resource_start(pdev, HINIC3_PCI_DB_BAR);
+	pci_adapter->db_dwqe_len = pci_resource_len(pdev, HINIC3_PCI_DB_BAR);
+	pci_adapter->db_base = pci_ioremap_bar(pdev, HINIC3_PCI_DB_BAR);
+	if (!pci_adapter->db_base) {
+		dev_err(&pdev->dev, "Failed to map doorbell regs\n");
+		goto err_unmap_intr_reg_base;
+	}
+
+	return 0;
+
+err_unmap_intr_reg_base:
+	iounmap(pci_adapter->intr_reg_base);
+
+err_unmap_cfg_reg_base:
+	iounmap(pci_adapter->cfg_reg_base);
+
+	return -ENOMEM;
+}
+
+static void hinic3_unmapping_bar(struct hinic3_pcidev *pci_adapter)
+{
+	iounmap(pci_adapter->db_base);
+	iounmap(pci_adapter->intr_reg_base);
+	iounmap(pci_adapter->cfg_reg_base);
+}
+
+static int hinic3_pci_init(struct pci_dev *pdev)
+{
+	struct hinic3_pcidev *pci_adapter;
+	int err;
+
+	pci_adapter = kzalloc(sizeof(*pci_adapter), GFP_KERNEL);
+	if (!pci_adapter)
+		return -ENOMEM;
+
+	pci_adapter->pdev = pdev;
+	mutex_init(&pci_adapter->pdev_mutex);
+
+	pci_set_drvdata(pdev, pci_adapter);
+
+	err = pci_enable_device(pdev);
+	if (err) {
+		dev_err(&pdev->dev, "Failed to enable PCI device\n");
+		goto err_free_pci_adapter;
+	}
+
+	err = pci_request_regions(pdev, HINIC3_NIC_DRV_NAME);
+	if (err) {
+		dev_err(&pdev->dev, "Failed to request regions\n");
+		goto err_disable_device;
+	}
+
+	pci_set_master(pdev);
+
+	err = dma_set_mask_and_coherent(&pdev->dev, DMA_BIT_MASK(64));
+	if (err) {
+		dev_err(&pdev->dev, "Failed to set DMA mask\n");
+		goto err_release_regions;
+	}
+
+	return 0;
+
+err_release_regions:
+	pci_clear_master(pdev);
+	pci_release_regions(pdev);
+
+err_disable_device:
+	pci_disable_device(pdev);
+
+err_free_pci_adapter:
+	pci_set_drvdata(pdev, NULL);
+	mutex_destroy(&pci_adapter->pdev_mutex);
+	kfree(pci_adapter);
+
+	return err;
+}
+
+static void hinic3_pci_uninit(struct pci_dev *pdev)
+{
+	struct hinic3_pcidev *pci_adapter = pci_get_drvdata(pdev);
+
+	pci_clear_master(pdev);
+	pci_release_regions(pdev);
+	pci_disable_device(pdev);
+	pci_set_drvdata(pdev, NULL);
+	mutex_destroy(&pci_adapter->pdev_mutex);
+	kfree(pci_adapter);
+}
+
+static int hinic3_func_init(struct pci_dev *pdev,
+			    struct hinic3_pcidev *pci_adapter)
+{
+	int err;
+
+	err = hinic3_init_hwdev(pdev);
+	if (err) {
+		dev_err(&pdev->dev, "Failed to initialize hardware device\n");
+		return err;
+	}
+
+	err = hinic3_attach_aux_devices(pci_adapter->hwdev);
+	if (err)
+		goto err_free_hwdev;
+
+	return 0;
+
+err_free_hwdev:
+	hinic3_free_hwdev(pci_adapter->hwdev);
+
+	return err;
+}
+
+static void hinic3_func_uninit(struct pci_dev *pdev)
+{
+	struct hinic3_pcidev *pci_adapter = pci_get_drvdata(pdev);
+
+	hinic3_flush_mgmt_workq(pci_adapter->hwdev);
+	hinic3_detach_aux_devices(pci_adapter->hwdev);
+	hinic3_free_hwdev(pci_adapter->hwdev);
+}
+
+static int hinic3_probe_func(struct hinic3_pcidev *pci_adapter)
+{
+	struct pci_dev *pdev = pci_adapter->pdev;
+	int err;
+
+	err = hinic3_mapping_bar(pdev, pci_adapter);
+	if (err) {
+		dev_err(&pdev->dev, "Failed to map bar\n");
+		goto err_out;
+	}
+
+	err = hinic3_func_init(pdev, pci_adapter);
+	if (err)
+		goto err_unmap_bar;
+
+	return 0;
+
+err_unmap_bar:
+	hinic3_unmapping_bar(pci_adapter);
+
+err_out:
+	dev_err(&pdev->dev, "PCIe device probe function failed\n");
+
+	return err;
+}
+
+static void hinic3_remove_func(struct hinic3_pcidev *pci_adapter)
+{
+	struct pci_dev *pdev = pci_adapter->pdev;
+
+	hinic3_func_uninit(pdev);
+	hinic3_unmapping_bar(pci_adapter);
+}
+
+static int hinic3_probe(struct pci_dev *pdev, const struct pci_device_id *id)
+{
+	struct hinic3_pcidev *pci_adapter;
+	int err;
+
+	err = hinic3_pci_init(pdev);
+	if (err)
+		goto err_out;
+
+	pci_adapter = pci_get_drvdata(pdev);
+	err = hinic3_probe_func(pci_adapter);
+	if (err)
+		goto err_uninit_pci;
+
+	return 0;
+
+err_uninit_pci:
+	hinic3_pci_uninit(pdev);
+
+err_out:
+	dev_err(&pdev->dev, "PCIe device probe failed\n");
+
+	return err;
+}
+
+static void hinic3_remove(struct pci_dev *pdev)
+{
+	struct hinic3_pcidev *pci_adapter = pci_get_drvdata(pdev);
+
+	hinic3_remove_func(pci_adapter);
+	hinic3_pci_uninit(pdev);
+}
+
+static const struct pci_device_id hinic3_pci_table[] = {
+	{PCI_VDEVICE(HUAWEI, PCI_DEV_ID_HINIC3_VF), 0},
+	{0, 0}
+
+};
+
+MODULE_DEVICE_TABLE(pci, hinic3_pci_table);
+
+static void hinic3_shutdown(struct pci_dev *pdev)
+{
+	struct hinic3_pcidev *pci_adapter = pci_get_drvdata(pdev);
+
+	pci_disable_device(pdev);
+
+	if (pci_adapter)
+		hinic3_set_api_stop(pci_adapter->hwdev);
+}
+
+static struct pci_driver hinic3_driver = {
+	.name            = HINIC3_NIC_DRV_NAME,
+	.id_table        = hinic3_pci_table,
+	.probe           = hinic3_probe,
+	.remove          = hinic3_remove,
+	.shutdown        = hinic3_shutdown,
+	.sriov_configure = pci_sriov_configure_simple
+};
+
+int hinic3_lld_init(void)
+{
+	return pci_register_driver(&hinic3_driver);
+}
+
+void hinic3_lld_exit(void)
+{
+	pci_unregister_driver(&hinic3_driver);
+}
diff --git a/drivers/net/ethernet/huawei/hinic3/hinic3_lld.h b/drivers/net/ethernet/huawei/hinic3/hinic3_lld.h
new file mode 100644
index 000000000000..322b44803476
--- /dev/null
+++ b/drivers/net/ethernet/huawei/hinic3/hinic3_lld.h
@@ -0,0 +1,21 @@
+/* SPDX-License-Identifier: GPL-2.0 */
+/* Copyright (c) Huawei Technologies Co., Ltd. 2025. All rights reserved. */
+
+#ifndef _HINIC3_LLD_H_
+#define _HINIC3_LLD_H_
+
+#include <linux/auxiliary_bus.h>
+
+struct hinic3_event_info;
+
+#define HINIC3_NIC_DRV_NAME "hinic3"
+
+int hinic3_lld_init(void);
+void hinic3_lld_exit(void);
+void hinic3_adev_event_register(struct auxiliary_device *adev,
+				void (*event_handler)(struct auxiliary_device *adev,
+						      struct hinic3_event_info *event));
+void hinic3_adev_event_unregister(struct auxiliary_device *adev);
+struct hinic3_hwdev *hinic3_adev_get_hwdev(struct auxiliary_device *adev);
+
+#endif
diff --git a/drivers/net/ethernet/huawei/hinic3/hinic3_main.c b/drivers/net/ethernet/huawei/hinic3/hinic3_main.c
new file mode 100644
index 000000000000..6d87d4d895ba
--- /dev/null
+++ b/drivers/net/ethernet/huawei/hinic3/hinic3_main.c
@@ -0,0 +1,409 @@
+// SPDX-License-Identifier: GPL-2.0
+// Copyright (c) Huawei Technologies Co., Ltd. 2025. All rights reserved.
+
+#include <linux/etherdevice.h>
+#include <linux/netdevice.h>
+
+#include "hinic3_common.h"
+#include "hinic3_hw_comm.h"
+#include "hinic3_hwdev.h"
+#include "hinic3_hwif.h"
+#include "hinic3_lld.h"
+#include "hinic3_nic_cfg.h"
+#include "hinic3_nic_dev.h"
+#include "hinic3_nic_io.h"
+#include "hinic3_rss.h"
+#include "hinic3_rx.h"
+#include "hinic3_tx.h"
+
+#define HINIC3_NIC_DRV_DESC  "Intelligent Network Interface Card Driver"
+
+#define HINIC3_RX_BUF_LEN          2048
+#define HINIC3_LRO_REPLENISH_THLD  256
+#define HINIC3_NIC_DEV_WQ_NAME     "hinic3_nic_dev_wq"
+
+#define HINIC3_SQ_DEPTH            1024
+#define HINIC3_RQ_DEPTH            1024
+
+#define HINIC3_DEFAULT_TXRX_MSIX_PENDING_LIMIT      2
+#define HINIC3_DEFAULT_TXRX_MSIX_COALESC_TIMER_CFG  25
+#define HINIC3_DEFAULT_TXRX_MSIX_RESEND_TIMER_CFG   7
+
+static void init_intr_coal_param(struct net_device *netdev)
+{
+	struct hinic3_nic_dev *nic_dev = netdev_priv(netdev);
+	struct hinic3_intr_coal_info *info;
+	u16 i;
+
+	for (i = 0; i < nic_dev->max_qps; i++) {
+		info = &nic_dev->intr_coalesce[i];
+		info->pending_limit = HINIC3_DEFAULT_TXRX_MSIX_PENDING_LIMIT;
+		info->coalesce_timer_cfg = HINIC3_DEFAULT_TXRX_MSIX_COALESC_TIMER_CFG;
+		info->resend_timer_cfg = HINIC3_DEFAULT_TXRX_MSIX_RESEND_TIMER_CFG;
+	}
+}
+
+static int hinic3_init_intr_coalesce(struct net_device *netdev)
+{
+	struct hinic3_nic_dev *nic_dev = netdev_priv(netdev);
+
+	nic_dev->intr_coalesce = kcalloc(nic_dev->max_qps,
+					 sizeof(*nic_dev->intr_coalesce),
+					 GFP_KERNEL);
+
+	if (!nic_dev->intr_coalesce)
+		return -ENOMEM;
+
+	init_intr_coal_param(netdev);
+
+	return 0;
+}
+
+static void hinic3_free_intr_coalesce(struct net_device *netdev)
+{
+	struct hinic3_nic_dev *nic_dev = netdev_priv(netdev);
+
+	kfree(nic_dev->intr_coalesce);
+}
+
+static int hinic3_alloc_txrxqs(struct net_device *netdev)
+{
+	struct hinic3_nic_dev *nic_dev = netdev_priv(netdev);
+	struct hinic3_hwdev *hwdev = nic_dev->hwdev;
+	int err;
+
+	err = hinic3_alloc_txqs(netdev);
+	if (err) {
+		dev_err(hwdev->dev, "Failed to alloc txqs\n");
+		return err;
+	}
+
+	err = hinic3_alloc_rxqs(netdev);
+	if (err) {
+		dev_err(hwdev->dev, "Failed to alloc rxqs\n");
+		goto err_free_txqs;
+	}
+
+	err = hinic3_init_intr_coalesce(netdev);
+	if (err) {
+		dev_err(hwdev->dev, "Failed to init_intr_coalesce\n");
+		goto err_free_rxqs;
+	}
+
+	return 0;
+
+err_free_rxqs:
+	hinic3_free_rxqs(netdev);
+
+err_free_txqs:
+	hinic3_free_txqs(netdev);
+
+	return err;
+}
+
+static void hinic3_free_txrxqs(struct net_device *netdev)
+{
+	hinic3_free_intr_coalesce(netdev);
+	hinic3_free_rxqs(netdev);
+	hinic3_free_txqs(netdev);
+}
+
+static int hinic3_init_nic_dev(struct net_device *netdev,
+			       struct hinic3_hwdev *hwdev)
+{
+	struct hinic3_nic_dev *nic_dev = netdev_priv(netdev);
+	struct pci_dev *pdev = hwdev->pdev;
+
+	nic_dev->netdev = netdev;
+	SET_NETDEV_DEV(netdev, &pdev->dev);
+	nic_dev->hwdev = hwdev;
+	nic_dev->pdev = pdev;
+
+	nic_dev->rx_buf_len = HINIC3_RX_BUF_LEN;
+	nic_dev->lro_replenish_thld = HINIC3_LRO_REPLENISH_THLD;
+	nic_dev->nic_svc_cap = hwdev->cfg_mgmt->cap.nic_svc_cap;
+
+	return 0;
+}
+
+static int hinic3_sw_init(struct net_device *netdev)
+{
+	struct hinic3_nic_dev *nic_dev = netdev_priv(netdev);
+	struct hinic3_hwdev *hwdev = nic_dev->hwdev;
+	int err;
+
+	nic_dev->q_params.sq_depth = HINIC3_SQ_DEPTH;
+	nic_dev->q_params.rq_depth = HINIC3_RQ_DEPTH;
+
+	hinic3_try_to_enable_rss(netdev);
+
+	/* VF driver always uses random MAC address. During VM migration to a
+	 * new device, the new device should learn the VMs old MAC rather than
+	 * provide its own MAC. The product design assumes that every VF is
+	 * suspectable to migration so the device avoids offering MAC address
+	 * to VFs.
+	 */
+	eth_hw_addr_random(netdev);
+	err = hinic3_set_mac(hwdev, netdev->dev_addr, 0,
+			     hinic3_global_func_id(hwdev));
+	if (err) {
+		dev_err(hwdev->dev, "Failed to set default MAC\n");
+		goto err_clear_rss_config;
+	}
+
+	err = hinic3_alloc_txrxqs(netdev);
+	if (err) {
+		dev_err(hwdev->dev, "Failed to alloc qps\n");
+		goto err_del_mac;
+	}
+
+	return 0;
+
+err_del_mac:
+	hinic3_del_mac(hwdev, netdev->dev_addr, 0,
+		       hinic3_global_func_id(hwdev));
+err_clear_rss_config:
+	hinic3_clear_rss_config(netdev);
+
+	return err;
+}
+
+static void hinic3_sw_uninit(struct net_device *netdev)
+{
+	struct hinic3_nic_dev *nic_dev = netdev_priv(netdev);
+
+	hinic3_free_txrxqs(netdev);
+	hinic3_del_mac(nic_dev->hwdev, netdev->dev_addr, 0,
+		       hinic3_global_func_id(nic_dev->hwdev));
+	hinic3_clear_rss_config(netdev);
+}
+
+static void hinic3_assign_netdev_ops(struct net_device *netdev)
+{
+	hinic3_set_netdev_ops(netdev);
+}
+
+static void netdev_feature_init(struct net_device *netdev)
+{
+	struct hinic3_nic_dev *nic_dev = netdev_priv(netdev);
+	netdev_features_t cso_fts = 0;
+	netdev_features_t tso_fts = 0;
+	netdev_features_t dft_fts;
+
+	dft_fts = NETIF_F_SG | NETIF_F_HIGHDMA;
+	if (hinic3_test_support(nic_dev, HINIC3_NIC_F_CSUM))
+		cso_fts |= NETIF_F_IP_CSUM | NETIF_F_IPV6_CSUM | NETIF_F_RXCSUM;
+	if (hinic3_test_support(nic_dev, HINIC3_NIC_F_SCTP_CRC))
+		cso_fts |= NETIF_F_SCTP_CRC;
+	if (hinic3_test_support(nic_dev, HINIC3_NIC_F_TSO))
+		tso_fts |= NETIF_F_TSO | NETIF_F_TSO6;
+
+	netdev->features |= dft_fts | cso_fts | tso_fts;
+}
+
+static int hinic3_set_default_hw_feature(struct net_device *netdev)
+{
+	struct hinic3_nic_dev *nic_dev = netdev_priv(netdev);
+	struct hinic3_hwdev *hwdev = nic_dev->hwdev;
+	int err;
+
+	err = hinic3_set_nic_feature_to_hw(nic_dev);
+	if (err) {
+		dev_err(hwdev->dev, "Failed to set nic features\n");
+		return err;
+	}
+
+	return 0;
+}
+
+static void hinic3_link_status_change(struct net_device *netdev,
+				      bool link_status_up)
+{
+	struct hinic3_nic_dev *nic_dev = netdev_priv(netdev);
+
+	if (link_status_up) {
+		if (netif_carrier_ok(netdev))
+			return;
+
+		nic_dev->link_status_up = true;
+		netif_carrier_on(netdev);
+		netdev_dbg(netdev, "Link is up\n");
+	} else {
+		if (!netif_carrier_ok(netdev))
+			return;
+
+		nic_dev->link_status_up = false;
+		netif_carrier_off(netdev);
+		netdev_dbg(netdev, "Link is down\n");
+	}
+}
+
+static void hinic3_nic_event(struct auxiliary_device *adev,
+			     struct hinic3_event_info *event)
+{
+	struct hinic3_nic_dev *nic_dev = dev_get_drvdata(&adev->dev);
+	struct net_device *netdev;
+
+	netdev = nic_dev->netdev;
+
+	switch (HINIC3_SRV_EVENT_TYPE(event->service, event->type)) {
+	case HINIC3_SRV_EVENT_TYPE(HINIC3_EVENT_SRV_NIC,
+				   HINIC3_NIC_EVENT_LINK_UP):
+		hinic3_link_status_change(netdev, true);
+		break;
+	case HINIC3_SRV_EVENT_TYPE(HINIC3_EVENT_SRV_NIC,
+				   HINIC3_NIC_EVENT_LINK_DOWN):
+		hinic3_link_status_change(netdev, false);
+		break;
+	default:
+		break;
+	}
+}
+
+static int hinic3_nic_probe(struct auxiliary_device *adev,
+			    const struct auxiliary_device_id *id)
+{
+	struct hinic3_hwdev *hwdev = hinic3_adev_get_hwdev(adev);
+	struct pci_dev *pdev = hwdev->pdev;
+	struct hinic3_nic_dev *nic_dev;
+	struct net_device *netdev;
+	u16 max_qps, glb_func_id;
+	int err;
+
+	if (!hinic3_support_nic(hwdev)) {
+		dev_dbg(&adev->dev, "HW doesn't support nic\n");
+		return 0;
+	}
+
+	hinic3_adev_event_register(adev, hinic3_nic_event);
+
+	glb_func_id = hinic3_global_func_id(hwdev);
+	err = hinic3_func_reset(hwdev, glb_func_id, COMM_FUNC_RESET_BIT_NIC);
+	if (err) {
+		dev_err(&adev->dev, "Failed to reset function\n");
+		goto err_unregister_adev_event;
+	}
+
+	max_qps = hinic3_func_max_qnum(hwdev);
+	netdev = alloc_etherdev_mq(sizeof(*nic_dev), max_qps);
+	if (!netdev) {
+		dev_err(&adev->dev, "Failed to allocate netdev\n");
+		err = -ENOMEM;
+		goto err_unregister_adev_event;
+	}
+
+	nic_dev = netdev_priv(netdev);
+	dev_set_drvdata(&adev->dev, nic_dev);
+	err = hinic3_init_nic_dev(netdev, hwdev);
+	if (err)
+		goto err_free_netdev;
+
+	err = hinic3_init_nic_io(nic_dev);
+	if (err)
+		goto err_free_netdev;
+
+	err = hinic3_sw_init(netdev);
+	if (err)
+		goto err_free_nic_io;
+
+	hinic3_assign_netdev_ops(netdev);
+
+	netdev_feature_init(netdev);
+	err = hinic3_set_default_hw_feature(netdev);
+	if (err)
+		goto err_uninit_sw;
+
+	netif_carrier_off(netdev);
+
+	err = register_netdev(netdev);
+	if (err)
+		goto err_uninit_nic_feature;
+
+	return 0;
+
+err_uninit_nic_feature:
+	hinic3_update_nic_feature(nic_dev, 0);
+	hinic3_set_nic_feature_to_hw(nic_dev);
+
+err_uninit_sw:
+	hinic3_sw_uninit(netdev);
+
+err_free_nic_io:
+	hinic3_free_nic_io(nic_dev);
+
+err_free_netdev:
+	free_netdev(netdev);
+
+err_unregister_adev_event:
+	hinic3_adev_event_unregister(adev);
+	dev_err(&pdev->dev, "NIC service probe failed\n");
+
+	return err;
+}
+
+static void hinic3_nic_remove(struct auxiliary_device *adev)
+{
+	struct hinic3_nic_dev *nic_dev = dev_get_drvdata(&adev->dev);
+	struct net_device *netdev;
+
+	if (!hinic3_support_nic(nic_dev->hwdev))
+		return;
+
+	netdev = nic_dev->netdev;
+	unregister_netdev(netdev);
+
+	hinic3_update_nic_feature(nic_dev, 0);
+	hinic3_set_nic_feature_to_hw(nic_dev);
+	hinic3_sw_uninit(netdev);
+
+	hinic3_free_nic_io(nic_dev);
+
+	free_netdev(netdev);
+}
+
+static const struct auxiliary_device_id hinic3_nic_id_table[] = {
+	{
+		.name = HINIC3_NIC_DRV_NAME ".nic",
+	},
+	{}
+};
+
+static struct auxiliary_driver hinic3_nic_driver = {
+	.probe    = hinic3_nic_probe,
+	.remove   = hinic3_nic_remove,
+	.suspend  = NULL,
+	.resume   = NULL,
+	.name     = "nic",
+	.id_table = hinic3_nic_id_table,
+};
+
+static __init int hinic3_nic_lld_init(void)
+{
+	int err;
+
+	err = hinic3_lld_init();
+	if (err)
+		return err;
+
+	err = auxiliary_driver_register(&hinic3_nic_driver);
+	if (err) {
+		hinic3_lld_exit();
+		return err;
+	}
+
+	return 0;
+}
+
+static __exit void hinic3_nic_lld_exit(void)
+{
+	auxiliary_driver_unregister(&hinic3_nic_driver);
+
+	hinic3_lld_exit();
+}
+
+module_init(hinic3_nic_lld_init);
+module_exit(hinic3_nic_lld_exit);
+
+MODULE_AUTHOR("Huawei Technologies CO., Ltd");
+MODULE_DESCRIPTION(HINIC3_NIC_DRV_DESC);
+MODULE_LICENSE("GPL");
diff --git a/drivers/net/ethernet/huawei/hinic3/hinic3_mbox.c b/drivers/net/ethernet/huawei/hinic3/hinic3_mbox.c
new file mode 100644
index 000000000000..cf67e26acece
--- /dev/null
+++ b/drivers/net/ethernet/huawei/hinic3/hinic3_mbox.c
@@ -0,0 +1,860 @@
+// SPDX-License-Identifier: GPL-2.0
+// Copyright (c) Huawei Technologies Co., Ltd. 2025. All rights reserved.
+
+#include <linux/dma-mapping.h>
+
+#include "hinic3_common.h"
+#include "hinic3_csr.h"
+#include "hinic3_hwdev.h"
+#include "hinic3_hwif.h"
+#include "hinic3_mbox.h"
+
+#define MBOX_INT_DST_AEQN_MASK        GENMASK(11, 10)
+#define MBOX_INT_SRC_RESP_AEQN_MASK   GENMASK(13, 12)
+#define MBOX_INT_STAT_DMA_MASK        GENMASK(19, 14)
+/* TX size, expressed in 4 bytes units */
+#define MBOX_INT_TX_SIZE_MASK         GENMASK(24, 20)
+/* SO_RO == strong order, relaxed order */
+#define MBOX_INT_STAT_DMA_SO_RO_MASK  GENMASK(26, 25)
+#define MBOX_INT_WB_EN_MASK           BIT(28)
+#define MBOX_INT_SET(val, field)  \
+	FIELD_PREP(MBOX_INT_##field##_MASK, val)
+
+#define MBOX_CTRL_TRIGGER_AEQE_MASK   BIT(0)
+#define MBOX_CTRL_TX_STATUS_MASK      BIT(1)
+#define MBOX_CTRL_DST_FUNC_MASK       GENMASK(28, 16)
+#define MBOX_CTRL_SET(val, field)  \
+	FIELD_PREP(MBOX_CTRL_##field##_MASK, val)
+
+#define MBOX_MSG_POLLING_TIMEOUT_MS  8000 // send msg seg timeout
+#define MBOX_COMP_POLLING_TIMEOUT_MS 40000 // response
+
+#define MBOX_MAX_BUF_SZ           2048
+#define MBOX_HEADER_SZ            8
+
+/* MBOX size is 64B, 8B for mbox_header, 8B reserved */
+#define MBOX_SEG_LEN              48
+#define MBOX_SEG_LEN_ALIGN        4
+#define MBOX_WB_STATUS_LEN        16
+
+#define MBOX_SEQ_ID_START_VAL     0
+#define MBOX_SEQ_ID_MAX_VAL       42
+#define MBOX_LAST_SEG_MAX_LEN  \
+	(MBOX_MAX_BUF_SZ - MBOX_SEQ_ID_MAX_VAL * MBOX_SEG_LEN)
+
+/* mbox write back status is 16B, only first 4B is used */
+#define MBOX_WB_STATUS_ERRCODE_MASK      0xFFFF
+#define MBOX_WB_STATUS_MASK              0xFF
+#define MBOX_WB_ERROR_CODE_MASK          0xFF00
+#define MBOX_WB_STATUS_FINISHED_SUCCESS  0xFF
+#define MBOX_WB_STATUS_NOT_FINISHED      0x00
+
+#define MBOX_STATUS_FINISHED(wb)  \
+	((FIELD_PREP(MBOX_WB_STATUS_MASK, (wb))) != MBOX_WB_STATUS_NOT_FINISHED)
+#define MBOX_STATUS_SUCCESS(wb)  \
+	((FIELD_PREP(MBOX_WB_STATUS_MASK, (wb))) ==  \
+	MBOX_WB_STATUS_FINISHED_SUCCESS)
+#define MBOX_STATUS_ERRCODE(wb)  \
+	((wb) & MBOX_WB_ERROR_CODE_MASK)
+
+#define MBOX_DMA_MSG_QUEUE_DEPTH    32
+#define MBOX_AREA(hwif)  \
+	((hwif)->cfg_regs_base + HINIC3_FUNC_CSR_MAILBOX_DATA_OFF)
+
+#define MBOX_MQ_CI_OFFSET  \
+	(HINIC3_CFG_REGS_FLAG + HINIC3_FUNC_CSR_MAILBOX_DATA_OFF + \
+	 MBOX_HEADER_SZ + MBOX_SEG_LEN)
+
+#define MBOX_MQ_SYNC_CI_MASK   GENMASK(7, 0)
+#define MBOX_MQ_ASYNC_CI_MASK  GENMASK(15, 8)
+#define MBOX_MQ_CI_GET(val, field)  \
+	FIELD_GET(MBOX_MQ_##field##_CI_MASK, val)
+
+#define MBOX_MGMT_FUNC_ID         0x1FFF
+#define MBOX_COMM_F_MBOX_SEGMENT  BIT(3)
+
+static u8 *get_mobx_body_from_hdr(u8 *header)
+{
+	return header + MBOX_HEADER_SZ;
+}
+
+static struct hinic3_msg_desc *get_mbox_msg_desc(struct hinic3_mbox *mbox,
+						 enum mbox_msg_direction_type dir,
+						 u16 src_func_id)
+{
+	struct hinic3_msg_channel *msg_ch;
+
+	msg_ch = (src_func_id == MBOX_MGMT_FUNC_ID) ?
+		&mbox->mgmt_msg : mbox->func_msg;
+
+	return (dir == MBOX_MSG_SEND) ?
+		&msg_ch->recv_msg : &msg_ch->resp_msg;
+}
+
+static void resp_mbox_handler(struct hinic3_mbox *mbox,
+			      const struct hinic3_msg_desc *msg_desc)
+{
+	spin_lock(&mbox->mbox_lock);
+	if (msg_desc->msg_info.msg_id == mbox->send_msg_id &&
+	    mbox->event_flag == MBOX_EVENT_START)
+		mbox->event_flag = MBOX_EVENT_SUCCESS;
+	spin_unlock(&mbox->mbox_lock);
+}
+
+static bool mbox_segment_valid(struct hinic3_mbox *mbox,
+			       struct hinic3_msg_desc *msg_desc,
+			       __le64 mbox_header)
+{
+	u8 seq_id, seg_len, msg_id, mod;
+	__le16 src_func_idx, cmd;
+
+	seq_id = MBOX_MSG_HEADER_GET(mbox_header, SEQID);
+	seg_len = MBOX_MSG_HEADER_GET(mbox_header, SEG_LEN);
+	msg_id = MBOX_MSG_HEADER_GET(mbox_header, MSG_ID);
+	mod = MBOX_MSG_HEADER_GET(mbox_header, MODULE);
+	cmd = cpu_to_le16(MBOX_MSG_HEADER_GET(mbox_header, CMD));
+	src_func_idx = cpu_to_le16(MBOX_MSG_HEADER_GET(mbox_header,
+						       SRC_GLB_FUNC_IDX));
+
+	if (seq_id > MBOX_SEQ_ID_MAX_VAL || seg_len > MBOX_SEG_LEN ||
+	    (seq_id == MBOX_SEQ_ID_MAX_VAL && seg_len > MBOX_LAST_SEG_MAX_LEN))
+		goto err_seg;
+
+	if (seq_id == 0) {
+		msg_desc->seq_id = seq_id;
+		msg_desc->msg_info.msg_id = msg_id;
+		msg_desc->mod = mod;
+		msg_desc->cmd = cmd;
+	} else {
+		if (seq_id != msg_desc->seq_id + 1 ||
+		    msg_id != msg_desc->msg_info.msg_id ||
+		    mod != msg_desc->mod || cmd != msg_desc->cmd)
+			goto err_seg;
+
+		msg_desc->seq_id = seq_id;
+	}
+
+	return true;
+
+err_seg:
+	dev_err(mbox->hwdev->dev,
+		"Mailbox segment check failed, src func id: 0x%x, front seg info: seq id: 0x%x, msg id: 0x%x, mod: 0x%x, cmd: 0x%x\n",
+		src_func_idx, msg_desc->seq_id, msg_desc->msg_info.msg_id,
+		msg_desc->mod, msg_desc->cmd);
+	dev_err(mbox->hwdev->dev,
+		"Current seg info: seg len: 0x%x, seq id: 0x%x, msg id: 0x%x, mod: 0x%x, cmd: 0x%x\n",
+		seg_len, seq_id, msg_id, mod, cmd);
+
+	return false;
+}
+
+static void recv_mbox_handler(struct hinic3_mbox *mbox,
+			      u8 *header, struct hinic3_msg_desc *msg_desc)
+{
+	__le64 mbox_header = *((__force __le64 *)header);
+	u8 *mbox_body = get_mobx_body_from_hdr(header);
+	u8 seq_id, seg_len;
+	int pos;
+
+	if (!mbox_segment_valid(mbox, msg_desc, mbox_header)) {
+		msg_desc->seq_id = MBOX_SEQ_ID_MAX_VAL;
+		return;
+	}
+
+	seq_id = MBOX_MSG_HEADER_GET(mbox_header, SEQID);
+	seg_len = MBOX_MSG_HEADER_GET(mbox_header, SEG_LEN);
+
+	pos = seq_id * MBOX_SEG_LEN;
+	memcpy(msg_desc->msg + pos, mbox_body, seg_len);
+
+	if (!MBOX_MSG_HEADER_GET(mbox_header, LAST))
+		return;
+
+	msg_desc->msg_len = cpu_to_le16(MBOX_MSG_HEADER_GET(mbox_header,
+							    MSG_LEN));
+	msg_desc->msg_info.status = MBOX_MSG_HEADER_GET(mbox_header, STATUS);
+
+	if (MBOX_MSG_HEADER_GET(mbox_header, DIRECTION) == MBOX_MSG_RESP)
+		resp_mbox_handler(mbox, msg_desc);
+}
+
+void hinic3_mbox_func_aeqe_handler(struct hinic3_hwdev *hwdev, u8 *header,
+				   u8 size)
+{
+	__le64 mbox_header = *((__force __le64 *)header);
+	enum mbox_msg_direction_type dir;
+	struct hinic3_msg_desc *msg_desc;
+	struct hinic3_mbox *mbox;
+	u16 src_func_id;
+
+	mbox = hwdev->mbox;
+	dir = MBOX_MSG_HEADER_GET(mbox_header, DIRECTION);
+	src_func_id = MBOX_MSG_HEADER_GET(mbox_header, SRC_GLB_FUNC_IDX);
+	msg_desc = get_mbox_msg_desc(mbox, dir, src_func_id);
+	recv_mbox_handler(mbox, header, msg_desc);
+}
+
+static int init_mbox_dma_queue(struct hinic3_hwdev *hwdev,
+			       struct mbox_dma_queue *mq)
+{
+	u32 size;
+
+	mq->depth = MBOX_DMA_MSG_QUEUE_DEPTH;
+	mq->prod_idx = 0;
+	mq->cons_idx = 0;
+
+	size = mq->depth * MBOX_MAX_BUF_SZ;
+	mq->dma_buf_vaddr = dma_alloc_coherent(hwdev->dev, size,
+					       &mq->dma_buf_paddr,
+					       GFP_KERNEL);
+	if (!mq->dma_buf_vaddr)
+		return -ENOMEM;
+
+	return 0;
+}
+
+static void uninit_mbox_dma_queue(struct hinic3_hwdev *hwdev,
+				  struct mbox_dma_queue *mq)
+{
+	dma_free_coherent(hwdev->dev, mq->depth * MBOX_MAX_BUF_SZ,
+			  mq->dma_buf_vaddr, mq->dma_buf_paddr);
+}
+
+static int hinic3_init_mbox_dma_queue(struct hinic3_mbox *mbox)
+{
+	u32 val;
+	int err;
+
+	err = init_mbox_dma_queue(mbox->hwdev, &mbox->sync_msg_queue);
+	if (err)
+		return err;
+
+	err = init_mbox_dma_queue(mbox->hwdev, &mbox->async_msg_queue);
+	if (err) {
+		uninit_mbox_dma_queue(mbox->hwdev, &mbox->sync_msg_queue);
+		return err;
+	}
+
+	val = hinic3_hwif_read_reg(mbox->hwdev->hwif, MBOX_MQ_CI_OFFSET);
+	val &= ~MBOX_MQ_SYNC_CI_MASK;
+	val &= ~MBOX_MQ_ASYNC_CI_MASK;
+	hinic3_hwif_write_reg(mbox->hwdev->hwif, MBOX_MQ_CI_OFFSET, val);
+
+	return 0;
+}
+
+static void hinic3_uninit_mbox_dma_queue(struct hinic3_mbox *mbox)
+{
+	uninit_mbox_dma_queue(mbox->hwdev, &mbox->sync_msg_queue);
+	uninit_mbox_dma_queue(mbox->hwdev, &mbox->async_msg_queue);
+}
+
+static int alloc_mbox_msg_channel(struct hinic3_msg_channel *msg_ch)
+{
+	msg_ch->resp_msg.msg = kzalloc(MBOX_MAX_BUF_SZ, GFP_KERNEL);
+	if (!msg_ch->resp_msg.msg)
+		return -ENOMEM;
+
+	msg_ch->recv_msg.msg = kzalloc(MBOX_MAX_BUF_SZ, GFP_KERNEL);
+	if (!msg_ch->recv_msg.msg) {
+		kfree(msg_ch->resp_msg.msg);
+		return -ENOMEM;
+	}
+
+	msg_ch->resp_msg.seq_id = MBOX_SEQ_ID_MAX_VAL;
+	msg_ch->recv_msg.seq_id = MBOX_SEQ_ID_MAX_VAL;
+
+	return 0;
+}
+
+static void free_mbox_msg_channel(struct hinic3_msg_channel *msg_ch)
+{
+	kfree(msg_ch->recv_msg.msg);
+	kfree(msg_ch->resp_msg.msg);
+}
+
+static int init_mgmt_msg_channel(struct hinic3_mbox *mbox)
+{
+	int err;
+
+	err = alloc_mbox_msg_channel(&mbox->mgmt_msg);
+	if (err) {
+		dev_err(mbox->hwdev->dev, "Failed to alloc mgmt message channel\n");
+		return err;
+	}
+
+	err = hinic3_init_mbox_dma_queue(mbox);
+	if (err) {
+		dev_err(mbox->hwdev->dev, "Failed to init mbox dma queue\n");
+		free_mbox_msg_channel(&mbox->mgmt_msg);
+		return err;
+	}
+
+	return 0;
+}
+
+static void uninit_mgmt_msg_channel(struct hinic3_mbox *mbox)
+{
+	hinic3_uninit_mbox_dma_queue(mbox);
+	free_mbox_msg_channel(&mbox->mgmt_msg);
+}
+
+static int hinic3_init_func_mbox_msg_channel(struct hinic3_hwdev *hwdev)
+{
+	struct hinic3_mbox *mbox;
+	int err;
+
+	mbox = hwdev->mbox;
+	mbox->func_msg = kzalloc(sizeof(*mbox->func_msg), GFP_KERNEL);
+	if (!mbox->func_msg)
+		return -ENOMEM;
+
+	err = alloc_mbox_msg_channel(mbox->func_msg);
+	if (err)
+		goto err_free_func_msg;
+
+	return 0;
+
+err_free_func_msg:
+	kfree(mbox->func_msg);
+	mbox->func_msg = NULL;
+
+	return err;
+}
+
+static void hinic3_uninit_func_mbox_msg_channel(struct hinic3_hwdev *hwdev)
+{
+	struct hinic3_mbox *mbox = hwdev->mbox;
+
+	free_mbox_msg_channel(mbox->func_msg);
+	kfree(mbox->func_msg);
+	mbox->func_msg = NULL;
+}
+
+static void prepare_send_mbox(struct hinic3_mbox *mbox)
+{
+	struct hinic3_send_mbox *send_mbox = &mbox->send_mbox;
+
+	send_mbox->data = MBOX_AREA(mbox->hwdev->hwif);
+}
+
+static int alloc_mbox_wb_status(struct hinic3_mbox *mbox)
+{
+	struct hinic3_send_mbox *send_mbox = &mbox->send_mbox;
+	struct hinic3_hwdev *hwdev = mbox->hwdev;
+	u32 addr_h, addr_l;
+
+	send_mbox->wb_vaddr = dma_alloc_coherent(hwdev->dev,
+						 MBOX_WB_STATUS_LEN,
+						 &send_mbox->wb_paddr,
+						 GFP_KERNEL);
+	if (!send_mbox->wb_vaddr)
+		return -ENOMEM;
+
+	addr_h = upper_32_bits(send_mbox->wb_paddr);
+	addr_l = lower_32_bits(send_mbox->wb_paddr);
+	hinic3_hwif_write_reg(hwdev->hwif, HINIC3_FUNC_CSR_MAILBOX_RESULT_H_OFF,
+			      addr_h);
+	hinic3_hwif_write_reg(hwdev->hwif, HINIC3_FUNC_CSR_MAILBOX_RESULT_L_OFF,
+			      addr_l);
+
+	return 0;
+}
+
+static void free_mbox_wb_status(struct hinic3_mbox *mbox)
+{
+	struct hinic3_send_mbox *send_mbox = &mbox->send_mbox;
+	struct hinic3_hwdev *hwdev = mbox->hwdev;
+
+	hinic3_hwif_write_reg(hwdev->hwif, HINIC3_FUNC_CSR_MAILBOX_RESULT_H_OFF,
+			      0);
+	hinic3_hwif_write_reg(hwdev->hwif, HINIC3_FUNC_CSR_MAILBOX_RESULT_L_OFF,
+			      0);
+
+	dma_free_coherent(hwdev->dev, MBOX_WB_STATUS_LEN,
+			  send_mbox->wb_vaddr, send_mbox->wb_paddr);
+}
+
+static int hinic3_mbox_pre_init(struct hinic3_hwdev *hwdev,
+				struct hinic3_mbox *mbox)
+{
+	mbox->hwdev = hwdev;
+	mutex_init(&mbox->mbox_send_lock);
+	spin_lock_init(&mbox->mbox_lock);
+
+	mbox->workq = create_singlethread_workqueue(HINIC3_MBOX_WQ_NAME);
+	if (!mbox->workq) {
+		dev_err(hwdev->dev, "Failed to initialize MBOX workqueue\n");
+		return -ENOMEM;
+	}
+	hwdev->mbox = mbox;
+
+	return 0;
+}
+
+int hinic3_init_mbox(struct hinic3_hwdev *hwdev)
+{
+	struct hinic3_mbox *mbox;
+	int err;
+
+	mbox = kzalloc(sizeof(*mbox), GFP_KERNEL);
+	if (!mbox)
+		return -ENOMEM;
+
+	err = hinic3_mbox_pre_init(hwdev, mbox);
+	if (err)
+		goto err_free_mbox;
+
+	err = init_mgmt_msg_channel(mbox);
+	if (err)
+		goto err_destroy_workqueue;
+
+	err = hinic3_init_func_mbox_msg_channel(hwdev);
+	if (err)
+		goto err_uninit_mgmt_msg_ch;
+
+	err = alloc_mbox_wb_status(mbox);
+	if (err) {
+		dev_err(hwdev->dev, "Failed to alloc mbox write back status\n");
+		goto err_uninit_func_mbox_msg_ch;
+	}
+
+	prepare_send_mbox(mbox);
+
+	return 0;
+
+err_uninit_func_mbox_msg_ch:
+	hinic3_uninit_func_mbox_msg_channel(hwdev);
+
+err_uninit_mgmt_msg_ch:
+	uninit_mgmt_msg_channel(mbox);
+
+err_destroy_workqueue:
+	destroy_workqueue(mbox->workq);
+
+err_free_mbox:
+	kfree(mbox);
+
+	return err;
+}
+
+void hinic3_free_mbox(struct hinic3_hwdev *hwdev)
+{
+	struct hinic3_mbox *mbox = hwdev->mbox;
+
+	destroy_workqueue(mbox->workq);
+	free_mbox_wb_status(mbox);
+	hinic3_uninit_func_mbox_msg_channel(hwdev);
+	uninit_mgmt_msg_channel(mbox);
+	kfree(mbox);
+}
+
+#define MBOX_DMA_MSG_INIT_XOR_VAL    0x5a5a5a5a
+#define MBOX_XOR_DATA_ALIGN          4
+static u32 mbox_dma_msg_xor(u32 *data, u32 msg_len)
+{
+	u32 xor = MBOX_DMA_MSG_INIT_XOR_VAL;
+	u32 dw_len = msg_len / sizeof(u32);
+	u32 i;
+
+	for (i = 0; i < dw_len; i++)
+		xor ^= data[i];
+
+	return xor;
+}
+
+#define MBOX_MQ_ID_MASK(mq, idx)  ((idx) & ((mq)->depth - 1))
+
+static bool is_msg_queue_full(struct mbox_dma_queue *mq)
+{
+	return MBOX_MQ_ID_MASK(mq, (mq)->prod_idx + 1) ==
+	       MBOX_MQ_ID_MASK(mq, (mq)->cons_idx);
+}
+
+static int mbox_prepare_dma_entry(struct hinic3_mbox *mbox,
+				  struct mbox_dma_queue *mq,
+				  struct mbox_dma_msg *dma_msg,
+				  const void *msg, u32 msg_len)
+{
+	u64 dma_addr, offset;
+	void *dma_vaddr;
+
+	if (is_msg_queue_full(mq)) {
+		dev_err(mbox->hwdev->dev, "Mbox sync message queue is busy, pi: %u, ci: %u\n",
+			mq->prod_idx, MBOX_MQ_ID_MASK(mq, mq->cons_idx));
+		return -EBUSY;
+	}
+
+	/* copy data to DMA buffer */
+	offset = mq->prod_idx * MBOX_MAX_BUF_SZ;
+	dma_vaddr = (u8 *)mq->dma_buf_vaddr + offset;
+	memcpy(dma_vaddr, msg, msg_len);
+	dma_addr = mq->dma_buf_paddr + offset;
+	dma_msg->dma_addr_high = cpu_to_le32(upper_32_bits(dma_addr));
+	dma_msg->dma_addr_low = cpu_to_le32(lower_32_bits(dma_addr));
+	dma_msg->msg_len = cpu_to_le32(msg_len);
+	/* The firmware obtains message based on 4B alignment. */
+	dma_msg->xor = cpu_to_le32(mbox_dma_msg_xor(dma_vaddr,
+						    ALIGN(msg_len, MBOX_XOR_DATA_ALIGN)));
+	mq->prod_idx++;
+	mq->prod_idx = MBOX_MQ_ID_MASK(mq, mq->prod_idx);
+
+	return 0;
+}
+
+static int mbox_prepare_dma_msg(struct hinic3_mbox *mbox,
+				enum mbox_msg_ack_type ack_type,
+				struct mbox_dma_msg *dma_msg, const void *msg,
+				u32 msg_len)
+{
+	struct mbox_dma_queue *mq;
+	u32 val;
+
+	val = hinic3_hwif_read_reg(mbox->hwdev->hwif, MBOX_MQ_CI_OFFSET);
+	if (ack_type == MBOX_MSG_ACK) {
+		mq = &mbox->sync_msg_queue;
+		mq->cons_idx = MBOX_MQ_CI_GET(val, SYNC);
+	} else {
+		mq = &mbox->async_msg_queue;
+		mq->cons_idx = MBOX_MQ_CI_GET(val, ASYNC);
+	}
+
+	return mbox_prepare_dma_entry(mbox, mq, dma_msg, msg, msg_len);
+}
+
+static void clear_mbox_status(struct hinic3_send_mbox *mbox)
+{
+	__be64 *wb_status = mbox->wb_vaddr;
+
+	*wb_status = 0;
+	/* clear mailbox write back status */
+	wmb();
+}
+
+static void mbox_dword_write(const void *src, void __iomem *dst, u32 count)
+{
+	const __le32 *src32 = src;
+	u32 __iomem *dst32 = dst;
+	u32 i;
+
+	/* Data written to mbox is arranged in structs with little endian fields
+	 * but when written to HW every dword (32bits) should be swapped since
+	 * the HW will swap it again.
+	 */
+	for (i = 0; i < count; i++)
+		__raw_writel(swab32((__force __u32)src32[i]), dst32 + i);
+}
+
+static void mbox_copy_header(struct hinic3_hwdev *hwdev,
+			     struct hinic3_send_mbox *mbox, __le64 *header)
+{
+	mbox_dword_write(header, mbox->data, MBOX_HEADER_SZ / sizeof(__le32));
+}
+
+static void mbox_copy_send_data(struct hinic3_hwdev *hwdev,
+				struct hinic3_send_mbox *mbox, void *seg,
+				u32 seg_len)
+{
+	u32 __iomem *dst = (u32 __iomem *)(mbox->data + MBOX_HEADER_SZ);
+	u32 count, leftover, last_dword;
+	const __le32 *src = seg;
+
+	count = seg_len / sizeof(u32);
+	leftover = seg_len % sizeof(u32);
+	if (count > 0)
+		mbox_dword_write(src, dst, count);
+
+	if (leftover > 0) {
+		last_dword = 0;
+		memcpy(&last_dword, src + count, leftover);
+		mbox_dword_write(&last_dword, dst + count, 1);
+	}
+}
+
+static void write_mbox_msg_attr(struct hinic3_mbox *mbox,
+				u16 dst_func, u16 dst_aeqn, u32 seg_len)
+{
+	struct hinic3_hwif *hwif = mbox->hwdev->hwif;
+	u32 mbox_int, mbox_ctrl, tx_size;
+
+	tx_size = ALIGN(seg_len + MBOX_HEADER_SZ, MBOX_SEG_LEN_ALIGN) >> 2;
+
+	mbox_int = MBOX_INT_SET(dst_aeqn, DST_AEQN) |
+		   MBOX_INT_SET(0, STAT_DMA) |
+		   MBOX_INT_SET(tx_size, TX_SIZE) |
+		   MBOX_INT_SET(0, STAT_DMA_SO_RO) |
+		   MBOX_INT_SET(1, WB_EN);
+
+	mbox_ctrl = MBOX_CTRL_SET(1, TX_STATUS) |
+		    MBOX_CTRL_SET(0, TRIGGER_AEQE) |
+		    MBOX_CTRL_SET(dst_func, DST_FUNC);
+
+	hinic3_hwif_write_reg(hwif, HINIC3_FUNC_CSR_MAILBOX_INT_OFF, mbox_int);
+	hinic3_hwif_write_reg(hwif, HINIC3_FUNC_CSR_MAILBOX_CONTROL_OFF,
+			      mbox_ctrl);
+}
+
+static u16 get_mbox_status(const struct hinic3_send_mbox *mbox)
+{
+	__be64 *wb_status = mbox->wb_vaddr;
+	u64 wb_val;
+
+	wb_val = be64_to_cpu(*wb_status);
+	/* verify reading before check */
+	rmb();
+
+	return wb_val & MBOX_WB_STATUS_ERRCODE_MASK;
+}
+
+static enum hinic3_wait_return check_mbox_wb_status(void *priv_data)
+{
+	struct hinic3_mbox *mbox = priv_data;
+	u16 wb_status;
+
+	wb_status = get_mbox_status(&mbox->send_mbox);
+
+	return MBOX_STATUS_FINISHED(wb_status) ?
+	       HINIC3_WAIT_PROCESS_CPL : HINIC3_WAIT_PROCESS_WAITING;
+}
+
+static int send_mbox_seg(struct hinic3_mbox *mbox, __le64 header,
+			 u16 dst_func, void *seg, u32 seg_len, void *msg_info)
+{
+	struct hinic3_send_mbox *send_mbox = &mbox->send_mbox;
+	struct hinic3_hwdev *hwdev = mbox->hwdev;
+	u8 num_aeqs = hwdev->hwif->attr.num_aeqs;
+	enum mbox_msg_direction_type dir;
+	u16 dst_aeqn, wb_status, errcode;
+	int err;
+
+	/* mbox to mgmt cpu, hardware doesn't care about dst aeq id */
+	if (num_aeqs > MBOX_MSG_AEQ_FOR_MBOX) {
+		dir = MBOX_MSG_HEADER_GET(header, DIRECTION);
+		dst_aeqn = (dir == MBOX_MSG_SEND) ?
+			   MBOX_MSG_AEQ_FOR_EVENT : MBOX_MSG_AEQ_FOR_MBOX;
+	} else {
+		dst_aeqn = 0;
+	}
+
+	clear_mbox_status(send_mbox);
+	mbox_copy_header(hwdev, send_mbox, &header);
+	mbox_copy_send_data(hwdev, send_mbox, seg, seg_len);
+	write_mbox_msg_attr(mbox, dst_func, dst_aeqn, seg_len);
+
+	err = hinic3_wait_for_timeout(mbox, check_mbox_wb_status,
+				      MBOX_MSG_POLLING_TIMEOUT_MS,
+				      USEC_PER_MSEC);
+	wb_status = get_mbox_status(send_mbox);
+	if (err) {
+		dev_err(hwdev->dev, "Send mailbox segment timeout, wb status: 0x%x\n",
+			wb_status);
+		return err;
+	}
+
+	if (!MBOX_STATUS_SUCCESS(wb_status)) {
+		dev_err(hwdev->dev,
+			"Send mailbox segment to function %u error, wb status: 0x%x\n",
+			dst_func, wb_status);
+		errcode = MBOX_STATUS_ERRCODE(wb_status);
+		return errcode ? errcode : -EFAULT;
+	}
+
+	return 0;
+}
+
+static int send_mbox_msg(struct hinic3_mbox *mbox, u8 mod, u16 cmd,
+			 const void *msg, u32 msg_len, u16 dst_func,
+			 enum mbox_msg_direction_type direction,
+			 enum mbox_msg_ack_type ack_type,
+			 struct mbox_msg_info *msg_info)
+{
+	enum mbox_msg_data_type data_type = MBOX_MSG_DATA_INLINE;
+	struct hinic3_hwdev *hwdev = mbox->hwdev;
+	struct mbox_dma_msg dma_msg;
+	u32 seg_len = MBOX_SEG_LEN;
+	__le64 header = 0;
+	u32 seq_id = 0;
+	u16 rsp_aeq_id;
+	u8 *msg_seg;
+	int err = 0;
+	u32 left;
+
+	if (hwdev->hwif->attr.num_aeqs > MBOX_MSG_AEQ_FOR_MBOX)
+		rsp_aeq_id = MBOX_MSG_AEQ_FOR_MBOX;
+	else
+		rsp_aeq_id = 0;
+
+	if (dst_func == MBOX_MGMT_FUNC_ID &&
+	    !(hwdev->features[0] & MBOX_COMM_F_MBOX_SEGMENT)) {
+		err = mbox_prepare_dma_msg(mbox, ack_type, &dma_msg,
+					   msg, msg_len);
+		if (err)
+			goto err_send;
+
+		msg = &dma_msg;
+		msg_len = sizeof(dma_msg);
+		data_type = MBOX_MSG_DATA_DMA;
+	}
+
+	msg_seg = (u8 *)msg;
+	left = msg_len;
+
+	header = cpu_to_le64(MBOX_MSG_HEADER_SET(msg_len, MSG_LEN) |
+			     MBOX_MSG_HEADER_SET(mod, MODULE) |
+			     MBOX_MSG_HEADER_SET(seg_len, SEG_LEN) |
+			     MBOX_MSG_HEADER_SET(ack_type, NO_ACK) |
+			     MBOX_MSG_HEADER_SET(data_type, DATA_TYPE) |
+			     MBOX_MSG_HEADER_SET(MBOX_SEQ_ID_START_VAL, SEQID) |
+			     MBOX_MSG_HEADER_SET(direction, DIRECTION) |
+			     MBOX_MSG_HEADER_SET(cmd, CMD) |
+			     MBOX_MSG_HEADER_SET(msg_info->msg_id, MSG_ID) |
+			     MBOX_MSG_HEADER_SET(rsp_aeq_id, AEQ_ID) |
+			     MBOX_MSG_HEADER_SET(MBOX_MSG_FROM_MBOX, SOURCE) |
+			     MBOX_MSG_HEADER_SET(!!msg_info->status, STATUS));
+
+	while (!(MBOX_MSG_HEADER_GET(header, LAST))) {
+		if (left <= MBOX_SEG_LEN) {
+			header &= cpu_to_le64(~MBOX_MSG_HEADER_SEG_LEN_MASK);
+			header |=
+				cpu_to_le64(MBOX_MSG_HEADER_SET(left, SEG_LEN) |
+					    MBOX_MSG_HEADER_SET(1, LAST));
+			seg_len = left;
+		}
+
+		err = send_mbox_seg(mbox, header, dst_func, msg_seg,
+				    seg_len, msg_info);
+		if (err) {
+			dev_err(hwdev->dev, "Failed to send mbox seg, seq_id=0x%llx\n",
+				MBOX_MSG_HEADER_GET(header, SEQID));
+			goto err_send;
+		}
+
+		left -= MBOX_SEG_LEN;
+		msg_seg += MBOX_SEG_LEN;
+		seq_id++;
+		header &= cpu_to_le64(~MBOX_MSG_HEADER_SEG_LEN_MASK);
+		header |= cpu_to_le64(MBOX_MSG_HEADER_SET(seq_id, SEQID));
+	}
+
+err_send:
+	return err;
+}
+
+static void set_mbox_to_func_event(struct hinic3_mbox *mbox,
+				   enum mbox_event_state event_flag)
+{
+	spin_lock(&mbox->mbox_lock);
+	mbox->event_flag = event_flag;
+	spin_unlock(&mbox->mbox_lock);
+}
+
+static enum hinic3_wait_return check_mbox_msg_finish(void *priv_data)
+{
+	struct hinic3_mbox *mbox = priv_data;
+
+	return (mbox->event_flag == MBOX_EVENT_SUCCESS) ?
+		HINIC3_WAIT_PROCESS_CPL : HINIC3_WAIT_PROCESS_WAITING;
+}
+
+static int wait_mbox_msg_completion(struct hinic3_mbox *mbox,
+				    u32 timeout)
+{
+	u32 wait_time;
+	int err;
+
+	wait_time = (timeout != 0) ? timeout : MBOX_COMP_POLLING_TIMEOUT_MS;
+	err = hinic3_wait_for_timeout(mbox, check_mbox_msg_finish,
+				      wait_time, USEC_PER_MSEC);
+	if (err) {
+		set_mbox_to_func_event(mbox, MBOX_EVENT_TIMEOUT);
+		return err;
+	}
+	set_mbox_to_func_event(mbox, MBOX_EVENT_END);
+
+	return 0;
+}
+
+int hinic3_send_mbox_to_mgmt(struct hinic3_hwdev *hwdev, u8 mod, u16 cmd,
+			     const struct mgmt_msg_params *msg_params)
+{
+	struct hinic3_mbox *mbox = hwdev->mbox;
+	struct mbox_msg_info msg_info = {};
+	struct hinic3_msg_desc *msg_desc;
+	u32 msg_len;
+	int err;
+
+	/* expect response message */
+	msg_desc = get_mbox_msg_desc(mbox, MBOX_MSG_RESP, MBOX_MGMT_FUNC_ID);
+	mutex_lock(&mbox->mbox_send_lock);
+	msg_info.msg_id = (mbox->send_msg_id + 1) & 0xF;
+	mbox->send_msg_id = msg_info.msg_id;
+	set_mbox_to_func_event(mbox, MBOX_EVENT_START);
+
+	err = send_mbox_msg(mbox, mod, cmd, msg_params->buf_in,
+			    msg_params->in_size, MBOX_MGMT_FUNC_ID,
+			    MBOX_MSG_SEND, MBOX_MSG_ACK, &msg_info);
+	if (err) {
+		dev_err(hwdev->dev, "Send mailbox mod %u, cmd %u failed, msg_id: %u, err: %d\n",
+			mod, cmd, msg_info.msg_id, err);
+		set_mbox_to_func_event(mbox, MBOX_EVENT_FAIL);
+		goto err_send;
+	}
+
+	if (wait_mbox_msg_completion(mbox, msg_params->timeout_ms)) {
+		dev_err(hwdev->dev,
+			"Send mbox msg timeout, msg_id: %u\n", msg_info.msg_id);
+		err = -ETIMEDOUT;
+		goto err_send;
+	}
+
+	if (mod != msg_desc->mod || cmd != le16_to_cpu(msg_desc->cmd)) {
+		dev_err(hwdev->dev,
+			"Invalid response mbox message, mod: 0x%x, cmd: 0x%x, expect mod: 0x%x, cmd: 0x%x\n",
+			msg_desc->mod, msg_desc->cmd, mod, cmd);
+		err = -EFAULT;
+		goto err_send;
+	}
+
+	if (msg_desc->msg_info.status) {
+		err = msg_desc->msg_info.status;
+		goto err_send;
+	}
+
+	if (msg_params->buf_out) {
+		msg_len = le16_to_cpu(msg_desc->msg_len);
+		if (msg_len != msg_params->expected_out_size) {
+			dev_err(hwdev->dev,
+				"Invalid response mbox message length: %u for mod %d cmd %u, expected length: %u\n",
+				msg_desc->msg_len, mod, cmd,
+				msg_params->expected_out_size);
+			err = -EFAULT;
+			goto err_send;
+		}
+
+		memcpy(msg_params->buf_out, msg_desc->msg, msg_len);
+	}
+
+err_send:
+	mutex_unlock(&mbox->mbox_send_lock);
+
+	return err;
+}
+
+int hinic3_send_mbox_to_mgmt_no_ack(struct hinic3_hwdev *hwdev, u8 mod, u16 cmd,
+				    const struct mgmt_msg_params *msg_params)
+{
+	struct hinic3_mbox *mbox = hwdev->mbox;
+	struct mbox_msg_info msg_info = {};
+	int err;
+
+	mutex_lock(&mbox->mbox_send_lock);
+	err = send_mbox_msg(mbox, mod, cmd, msg_params->buf_in,
+			    msg_params->in_size, MBOX_MGMT_FUNC_ID,
+			    MBOX_MSG_SEND, MBOX_MSG_NO_ACK, &msg_info);
+	if (err)
+		dev_err(hwdev->dev, "Send mailbox no ack failed\n");
+
+	mutex_unlock(&mbox->mbox_send_lock);
+
+	return err;
+}
diff --git a/drivers/net/ethernet/huawei/hinic3/hinic3_mbox.h b/drivers/net/ethernet/huawei/hinic3/hinic3_mbox.h
new file mode 100644
index 000000000000..e71629e95086
--- /dev/null
+++ b/drivers/net/ethernet/huawei/hinic3/hinic3_mbox.h
@@ -0,0 +1,141 @@
+/* SPDX-License-Identifier: GPL-2.0 */
+/* Copyright (c) Huawei Technologies Co., Ltd. 2025. All rights reserved. */
+
+#ifndef _HINIC3_MBOX_H_
+#define _HINIC3_MBOX_H_
+
+#include <linux/bitfield.h>
+#include <linux/mutex.h>
+
+struct hinic3_hwdev;
+struct mgmt_msg_params;
+
+#define MBOX_MSG_HEADER_SRC_GLB_FUNC_IDX_MASK  GENMASK_ULL(12, 0)
+#define MBOX_MSG_HEADER_STATUS_MASK            BIT_ULL(13)
+#define MBOX_MSG_HEADER_SOURCE_MASK            BIT_ULL(15)
+#define MBOX_MSG_HEADER_AEQ_ID_MASK            GENMASK_ULL(17, 16)
+#define MBOX_MSG_HEADER_MSG_ID_MASK            GENMASK_ULL(21, 18)
+#define MBOX_MSG_HEADER_CMD_MASK               GENMASK_ULL(31, 22)
+#define MBOX_MSG_HEADER_MSG_LEN_MASK           GENMASK_ULL(42, 32)
+#define MBOX_MSG_HEADER_MODULE_MASK            GENMASK_ULL(47, 43)
+#define MBOX_MSG_HEADER_SEG_LEN_MASK           GENMASK_ULL(53, 48)
+#define MBOX_MSG_HEADER_NO_ACK_MASK            BIT_ULL(54)
+#define MBOX_MSG_HEADER_DATA_TYPE_MASK         BIT_ULL(55)
+#define MBOX_MSG_HEADER_SEQID_MASK             GENMASK_ULL(61, 56)
+#define MBOX_MSG_HEADER_LAST_MASK              BIT_ULL(62)
+#define MBOX_MSG_HEADER_DIRECTION_MASK         BIT_ULL(63)
+
+#define MBOX_MSG_HEADER_SET(val, member) \
+	FIELD_PREP(MBOX_MSG_HEADER_##member##_MASK, val)
+#define MBOX_MSG_HEADER_GET(val, member) \
+	FIELD_GET(MBOX_MSG_HEADER_##member##_MASK, le64_to_cpu(val))
+
+/* identifies if a segment belongs to a message or to a response. A VF is only
+ * expected to send messages and receive responses. PF driver could receive
+ * messages and send responses.
+ */
+enum mbox_msg_direction_type {
+	MBOX_MSG_SEND = 0,
+	MBOX_MSG_RESP = 1,
+};
+
+/* Indicates if mbox message expects a response (ack) or not */
+enum mbox_msg_ack_type {
+	MBOX_MSG_ACK    = 0,
+	MBOX_MSG_NO_ACK = 1,
+};
+
+enum mbox_msg_data_type {
+	MBOX_MSG_DATA_INLINE = 0,
+	MBOX_MSG_DATA_DMA    = 1,
+};
+
+enum mbox_msg_src_type {
+	MBOX_MSG_FROM_MBOX = 1,
+};
+
+enum mbox_msg_aeq_type {
+	MBOX_MSG_AEQ_FOR_EVENT = 0,
+	MBOX_MSG_AEQ_FOR_MBOX  = 1,
+};
+
+#define HINIC3_MBOX_WQ_NAME  "hinic3_mbox"
+
+struct mbox_msg_info {
+	u8 msg_id;
+	u8 status;
+};
+
+struct hinic3_msg_desc {
+	u8                   *msg;
+	__le16               msg_len;
+	u8                   seq_id;
+	u8                   mod;
+	__le16               cmd;
+	struct mbox_msg_info msg_info;
+};
+
+struct hinic3_msg_channel {
+	struct   hinic3_msg_desc resp_msg;
+	struct   hinic3_msg_desc recv_msg;
+};
+
+struct hinic3_send_mbox {
+	u8 __iomem *data;
+	void       *wb_vaddr;
+	dma_addr_t wb_paddr;
+};
+
+enum mbox_event_state {
+	MBOX_EVENT_START   = 0,
+	MBOX_EVENT_FAIL    = 1,
+	MBOX_EVENT_SUCCESS = 2,
+	MBOX_EVENT_TIMEOUT = 3,
+	MBOX_EVENT_END     = 4,
+};
+
+struct mbox_dma_msg {
+	__le32 xor;
+	__le32 dma_addr_high;
+	__le32 dma_addr_low;
+	__le32 msg_len;
+	__le64 rsvd;
+};
+
+struct mbox_dma_queue {
+	void       *dma_buf_vaddr;
+	dma_addr_t dma_buf_paddr;
+	u16        depth;
+	u16        prod_idx;
+	u16        cons_idx;
+};
+
+struct hinic3_mbox {
+	struct hinic3_hwdev       *hwdev;
+	/* lock for send mbox message and ack message */
+	struct mutex              mbox_send_lock;
+	struct hinic3_send_mbox   send_mbox;
+	struct mbox_dma_queue     sync_msg_queue;
+	struct mbox_dma_queue     async_msg_queue;
+	struct workqueue_struct   *workq;
+	/* driver and MGMT CPU */
+	struct hinic3_msg_channel mgmt_msg;
+	/* VF to PF */
+	struct hinic3_msg_channel *func_msg;
+	u8                        send_msg_id;
+	enum mbox_event_state     event_flag;
+	/* lock for mbox event flag */
+	spinlock_t                mbox_lock;
+};
+
+void hinic3_mbox_func_aeqe_handler(struct hinic3_hwdev *hwdev, u8 *header,
+				   u8 size);
+int hinic3_init_mbox(struct hinic3_hwdev *hwdev);
+void hinic3_free_mbox(struct hinic3_hwdev *hwdev);
+
+int hinic3_send_mbox_to_mgmt(struct hinic3_hwdev *hwdev, u8 mod, u16 cmd,
+			     const struct mgmt_msg_params *msg_params);
+int hinic3_send_mbox_to_mgmt_no_ack(struct hinic3_hwdev *hwdev, u8 mod, u16 cmd,
+				    const struct mgmt_msg_params *msg_params);
+
+#endif
diff --git a/drivers/net/ethernet/huawei/hinic3/hinic3_mgmt.c b/drivers/net/ethernet/huawei/hinic3/hinic3_mgmt.c
new file mode 100644
index 000000000000..c38d10cd7fac
--- /dev/null
+++ b/drivers/net/ethernet/huawei/hinic3/hinic3_mgmt.c
@@ -0,0 +1,21 @@
+// SPDX-License-Identifier: GPL-2.0
+// Copyright (c) Huawei Technologies Co., Ltd. 2025. All rights reserved.
+
+#include "hinic3_eqs.h"
+#include "hinic3_hwdev.h"
+#include "hinic3_mbox.h"
+#include "hinic3_mgmt.h"
+
+void hinic3_flush_mgmt_workq(struct hinic3_hwdev *hwdev)
+{
+	if (hwdev->aeqs)
+		flush_workqueue(hwdev->aeqs->workq);
+}
+
+void hinic3_mgmt_msg_aeqe_handler(struct hinic3_hwdev *hwdev, u8 *header,
+				  u8 size)
+{
+	if (MBOX_MSG_HEADER_GET(*(__force __le64 *)header, SOURCE) ==
+				MBOX_MSG_FROM_MBOX)
+		hinic3_mbox_func_aeqe_handler(hwdev, header, size);
+}
diff --git a/drivers/net/ethernet/huawei/hinic3/hinic3_mgmt.h b/drivers/net/ethernet/huawei/hinic3/hinic3_mgmt.h
new file mode 100644
index 000000000000..bbef3b32a6ec
--- /dev/null
+++ b/drivers/net/ethernet/huawei/hinic3/hinic3_mgmt.h
@@ -0,0 +1,15 @@
+/* SPDX-License-Identifier: GPL-2.0 */
+/* Copyright (c) Huawei Technologies Co., Ltd. 2025. All rights reserved. */
+
+#ifndef _HINIC3_MGMT_H_
+#define _HINIC3_MGMT_H_
+
+#include <linux/types.h>
+
+struct hinic3_hwdev;
+
+void hinic3_flush_mgmt_workq(struct hinic3_hwdev *hwdev);
+void hinic3_mgmt_msg_aeqe_handler(struct hinic3_hwdev *hwdev,
+				  u8 *header, u8 size);
+
+#endif
diff --git a/drivers/net/ethernet/huawei/hinic3/hinic3_mgmt_interface.h b/drivers/net/ethernet/huawei/hinic3/hinic3_mgmt_interface.h
new file mode 100644
index 000000000000..6cc0345c39e4
--- /dev/null
+++ b/drivers/net/ethernet/huawei/hinic3/hinic3_mgmt_interface.h
@@ -0,0 +1,224 @@
+/* SPDX-License-Identifier: GPL-2.0 */
+/* Copyright (c) Huawei Technologies Co., Ltd. 2025. All rights reserved. */
+
+#ifndef _HINIC3_MGMT_INTERFACE_H_
+#define _HINIC3_MGMT_INTERFACE_H_
+
+#include <linux/bitfield.h>
+#include <linux/bits.h>
+#include <linux/if_ether.h>
+
+#include "hinic3_hw_intf.h"
+
+struct l2nic_cmd_feature_nego {
+	struct mgmt_msg_head msg_head;
+	u16                  func_id;
+	u8                   opcode;
+	u8                   rsvd;
+	u64                  s_feature[4];
+};
+
+enum l2nic_func_tbl_cfg_bitmap {
+	L2NIC_FUNC_TBL_CFG_INIT        = 0,
+	L2NIC_FUNC_TBL_CFG_RX_BUF_SIZE = 1,
+	L2NIC_FUNC_TBL_CFG_MTU         = 2,
+};
+
+struct l2nic_func_tbl_cfg {
+	u16 rx_wqe_buf_size;
+	u16 mtu;
+	u32 rsvd[9];
+};
+
+struct l2nic_cmd_set_func_tbl {
+	struct mgmt_msg_head      msg_head;
+	u16                       func_id;
+	u16                       rsvd;
+	u32                       cfg_bitmap;
+	struct l2nic_func_tbl_cfg tbl_cfg;
+};
+
+struct l2nic_cmd_set_mac {
+	struct mgmt_msg_head msg_head;
+	u16                  func_id;
+	u16                  vlan_id;
+	u16                  rsvd1;
+	u8                   mac[ETH_ALEN];
+};
+
+struct l2nic_cmd_update_mac {
+	struct mgmt_msg_head msg_head;
+	u16                  func_id;
+	u16                  vlan_id;
+	u16                  rsvd1;
+	u8                   old_mac[ETH_ALEN];
+	u16                  rsvd2;
+	u8                   new_mac[ETH_ALEN];
+};
+
+struct l2nic_cmd_set_ci_attr {
+	struct mgmt_msg_head msg_head;
+	u16                  func_idx;
+	u8                   dma_attr_off;
+	u8                   pending_limit;
+	u8                   coalescing_time;
+	u8                   intr_en;
+	u16                  intr_idx;
+	u32                  l2nic_sqn;
+	u32                  rsvd;
+	u64                  ci_addr;
+};
+
+struct l2nic_cmd_clear_qp_resource {
+	struct mgmt_msg_head msg_head;
+	u16                  func_id;
+	u16                  rsvd1;
+};
+
+struct l2nic_cmd_force_pkt_drop {
+	struct mgmt_msg_head msg_head;
+	u8                   port;
+	u8                   rsvd1[3];
+};
+
+struct l2nic_cmd_set_vport_state {
+	struct mgmt_msg_head msg_head;
+	u16                  func_id;
+	u16                  rsvd1;
+	/* 0--disable, 1--enable */
+	u8                   state;
+	u8                   rsvd2[3];
+};
+
+struct l2nic_cmd_set_dcb_state {
+	struct mgmt_msg_head head;
+	u16                  func_id;
+	/* 0 - get dcb state, 1 - set dcb state */
+	u8                   op_code;
+	/* 0 - disable, 1 - enable dcb */
+	u8                   state;
+	/* 0 - disable, 1 - enable dcb */
+	u8                   port_state;
+	u8                   rsvd[7];
+};
+
+#define L2NIC_RSS_TYPE_VALID_MASK         BIT(23)
+#define L2NIC_RSS_TYPE_TCP_IPV6_EXT_MASK  BIT(24)
+#define L2NIC_RSS_TYPE_IPV6_EXT_MASK      BIT(25)
+#define L2NIC_RSS_TYPE_TCP_IPV6_MASK      BIT(26)
+#define L2NIC_RSS_TYPE_IPV6_MASK          BIT(27)
+#define L2NIC_RSS_TYPE_TCP_IPV4_MASK      BIT(28)
+#define L2NIC_RSS_TYPE_IPV4_MASK          BIT(29)
+#define L2NIC_RSS_TYPE_UDP_IPV6_MASK      BIT(30)
+#define L2NIC_RSS_TYPE_UDP_IPV4_MASK      BIT(31)
+#define L2NIC_RSS_TYPE_SET(val, member)  \
+	FIELD_PREP(L2NIC_RSS_TYPE_##member##_MASK, val)
+#define L2NIC_RSS_TYPE_GET(val, member)  \
+	FIELD_GET(L2NIC_RSS_TYPE_##member##_MASK, val)
+
+#define L2NIC_RSS_INDIR_SIZE  256
+#define L2NIC_RSS_KEY_SIZE    40
+
+/* IEEE 802.1Qaz std */
+#define L2NIC_DCB_COS_MAX     0x8
+
+struct l2nic_cmd_set_rss_ctx_tbl {
+	struct mgmt_msg_head msg_head;
+	u16                  func_id;
+	u16                  rsvd1;
+	u32                  context;
+};
+
+struct l2nic_cmd_cfg_rss_engine {
+	struct mgmt_msg_head msg_head;
+	u16                  func_id;
+	u8                   opcode;
+	u8                   hash_engine;
+	u8                   rsvd1[4];
+};
+
+struct l2nic_cmd_cfg_rss_hash_key {
+	struct mgmt_msg_head msg_head;
+	u16                  func_id;
+	u8                   opcode;
+	u8                   rsvd1;
+	u8                   key[L2NIC_RSS_KEY_SIZE];
+};
+
+struct l2nic_cmd_cfg_rss {
+	struct mgmt_msg_head msg_head;
+	u16                  func_id;
+	u8                   rss_en;
+	u8                   rq_priority_number;
+	u8                   prio_tc[L2NIC_DCB_COS_MAX];
+	u16                  num_qps;
+	u16                  rsvd1;
+};
+
+/* Commands between NIC to fw */
+enum l2nic_cmd {
+	/* FUNC CFG */
+	L2NIC_CMD_SET_FUNC_TBL        = 5,
+	L2NIC_CMD_SET_VPORT_ENABLE    = 6,
+	L2NIC_CMD_SET_SQ_CI_ATTR      = 8,
+	L2NIC_CMD_CLEAR_QP_RESOURCE   = 11,
+	L2NIC_CMD_FEATURE_NEGO        = 15,
+	L2NIC_CMD_SET_MAC             = 21,
+	L2NIC_CMD_DEL_MAC             = 22,
+	L2NIC_CMD_UPDATE_MAC          = 23,
+	L2NIC_CMD_CFG_RSS             = 60,
+	L2NIC_CMD_CFG_RSS_HASH_KEY    = 63,
+	L2NIC_CMD_CFG_RSS_HASH_ENGINE = 64,
+	L2NIC_CMD_SET_RSS_CTX_TBL     = 65,
+	L2NIC_CMD_QOS_DCB_STATE       = 110,
+	L2NIC_CMD_FORCE_PKT_DROP      = 113,
+	L2NIC_CMD_MAX                 = 256,
+};
+
+struct l2nic_cmd_rss_set_indir_tbl {
+	__le32 rsvd[4];
+	__le16 entry[L2NIC_RSS_INDIR_SIZE];
+};
+
+/* NIC CMDQ MODE */
+enum l2nic_ucode_cmd {
+	L2NIC_UCODE_CMD_MODIFY_QUEUE_CTX  = 0,
+	L2NIC_UCODE_CMD_CLEAN_QUEUE_CTX   = 1,
+	L2NIC_UCODE_CMD_SET_RSS_INDIR_TBL = 4,
+};
+
+/* hilink mac group command */
+enum mag_cmd {
+	MAG_CMD_GET_LINK_STATUS = 7,
+};
+
+/* firmware also use this cmd report link event to driver */
+struct mag_cmd_get_link_status {
+	struct mgmt_msg_head head;
+	u8                   port_id;
+	/* 0:link down  1:link up */
+	u8                   status;
+	u8                   rsvd0[2];
+};
+
+enum hinic3_nic_feature_cap {
+	HINIC3_NIC_F_CSUM           = BIT(0),
+	HINIC3_NIC_F_SCTP_CRC       = BIT(1),
+	HINIC3_NIC_F_TSO            = BIT(2),
+	HINIC3_NIC_F_LRO            = BIT(3),
+	HINIC3_NIC_F_UFO            = BIT(4),
+	HINIC3_NIC_F_RSS            = BIT(5),
+	HINIC3_NIC_F_RX_VLAN_FILTER = BIT(6),
+	HINIC3_NIC_F_RX_VLAN_STRIP  = BIT(7),
+	HINIC3_NIC_F_TX_VLAN_INSERT = BIT(8),
+	HINIC3_NIC_F_VXLAN_OFFLOAD  = BIT(9),
+	HINIC3_NIC_F_FDIR           = BIT(11),
+	HINIC3_NIC_F_PROMISC        = BIT(12),
+	HINIC3_NIC_F_ALLMULTI       = BIT(13),
+	HINIC3_NIC_F_RATE_LIMIT     = BIT(16),
+};
+
+#define HINIC3_NIC_F_ALL_MASK           0x33bff
+#define HINIC3_NIC_DRV_DEFAULT_FEATURE  0x3f03f
+
+#endif
diff --git a/drivers/net/ethernet/huawei/hinic3/hinic3_netdev_ops.c b/drivers/net/ethernet/huawei/hinic3/hinic3_netdev_ops.c
new file mode 100644
index 000000000000..0fa3c7900225
--- /dev/null
+++ b/drivers/net/ethernet/huawei/hinic3/hinic3_netdev_ops.c
@@ -0,0 +1,496 @@
+// SPDX-License-Identifier: GPL-2.0
+// Copyright (c) Huawei Technologies Co., Ltd. 2025. All rights reserved.
+
+#include <linux/etherdevice.h>
+#include <linux/netdevice.h>
+
+#include "hinic3_hwif.h"
+#include "hinic3_nic_cfg.h"
+#include "hinic3_nic_dev.h"
+#include "hinic3_nic_io.h"
+#include "hinic3_rss.h"
+#include "hinic3_rx.h"
+#include "hinic3_tx.h"
+
+/* try to modify the number of irq to the target number,
+ * and return the actual number of irq.
+ */
+static u16 hinic3_qp_irq_change(struct net_device *netdev,
+				u16 dst_num_qp_irq)
+{
+	struct hinic3_nic_dev *nic_dev = netdev_priv(netdev);
+	struct msix_entry *qps_msix_entries;
+	u16 resp_irq_num, irq_num_gap, i;
+	u16 idx;
+	int err;
+
+	qps_msix_entries = nic_dev->qps_msix_entries;
+	if (dst_num_qp_irq > nic_dev->num_qp_irq) {
+		irq_num_gap = dst_num_qp_irq - nic_dev->num_qp_irq;
+		err = hinic3_alloc_irqs(nic_dev->hwdev, irq_num_gap,
+					&qps_msix_entries[nic_dev->num_qp_irq],
+					&resp_irq_num);
+		if (err) {
+			netdev_err(netdev, "Failed to alloc irqs\n");
+			return nic_dev->num_qp_irq;
+		}
+
+		nic_dev->num_qp_irq += resp_irq_num;
+	} else if (dst_num_qp_irq < nic_dev->num_qp_irq) {
+		irq_num_gap = nic_dev->num_qp_irq - dst_num_qp_irq;
+		for (i = 0; i < irq_num_gap; i++) {
+			idx = (nic_dev->num_qp_irq - i) - 1;
+			hinic3_free_irq(nic_dev->hwdev,
+					qps_msix_entries[idx].vector);
+			qps_msix_entries[idx].vector = 0;
+			qps_msix_entries[idx].entry = 0;
+		}
+		nic_dev->num_qp_irq = dst_num_qp_irq;
+	}
+
+	return nic_dev->num_qp_irq;
+}
+
+static void hinic3_config_num_qps(struct net_device *netdev,
+				  struct hinic3_dyna_txrxq_params *q_params)
+{
+	struct hinic3_nic_dev *nic_dev = netdev_priv(netdev);
+	u16 alloc_num_irq, cur_num_irq;
+	u16 dst_num_irq;
+
+	if (!test_bit(HINIC3_RSS_ENABLE, &nic_dev->flags))
+		q_params->num_qps = 1;
+
+	if (nic_dev->num_qp_irq >= q_params->num_qps)
+		goto out;
+
+	cur_num_irq = nic_dev->num_qp_irq;
+
+	alloc_num_irq = hinic3_qp_irq_change(netdev, q_params->num_qps);
+	if (alloc_num_irq < q_params->num_qps) {
+		q_params->num_qps = alloc_num_irq;
+		netdev_warn(netdev, "Can not get enough irqs, adjust num_qps to %u\n",
+			    q_params->num_qps);
+
+		/* The current irq may be in use, we must keep it */
+		dst_num_irq = max_t(u16, cur_num_irq, q_params->num_qps);
+		hinic3_qp_irq_change(netdev, dst_num_irq);
+	}
+
+out:
+	netdev_dbg(netdev, "No need to change irqs, num_qps is %u\n",
+		   q_params->num_qps);
+}
+
+static int hinic3_setup_num_qps(struct net_device *netdev)
+{
+	struct hinic3_nic_dev *nic_dev = netdev_priv(netdev);
+
+	nic_dev->num_qp_irq = 0;
+
+	nic_dev->qps_msix_entries = kcalloc(nic_dev->max_qps,
+					    sizeof(struct msix_entry),
+					    GFP_KERNEL);
+	if (!nic_dev->qps_msix_entries)
+		return -ENOMEM;
+
+	hinic3_config_num_qps(netdev, &nic_dev->q_params);
+
+	return 0;
+}
+
+static void hinic3_destroy_num_qps(struct net_device *netdev)
+{
+	struct hinic3_nic_dev *nic_dev = netdev_priv(netdev);
+	u16 i;
+
+	for (i = 0; i < nic_dev->num_qp_irq; i++)
+		hinic3_free_irq(nic_dev->hwdev,
+				nic_dev->qps_msix_entries[i].vector);
+
+	kfree(nic_dev->qps_msix_entries);
+}
+
+static int hinic3_alloc_txrxq_resources(struct net_device *netdev,
+					struct hinic3_dyna_txrxq_params *q_params)
+{
+	int err;
+
+	q_params->txqs_res = kcalloc(q_params->num_qps,
+				     sizeof(*q_params->txqs_res), GFP_KERNEL);
+	if (!q_params->txqs_res)
+		return -ENOMEM;
+
+	q_params->rxqs_res = kcalloc(q_params->num_qps,
+				     sizeof(*q_params->rxqs_res), GFP_KERNEL);
+	if (!q_params->rxqs_res) {
+		err = -ENOMEM;
+		goto err_free_txqs_res_arr;
+	}
+
+	q_params->irq_cfg = kcalloc(q_params->num_qps,
+				    sizeof(*q_params->irq_cfg), GFP_KERNEL);
+	if (!q_params->irq_cfg) {
+		err = -ENOMEM;
+		goto err_free_rxqs_res_arr;
+	}
+
+	err = hinic3_alloc_txqs_res(netdev, q_params->num_qps,
+				    q_params->sq_depth, q_params->txqs_res);
+	if (err) {
+		netdev_err(netdev, "Failed to alloc txqs resource\n");
+		goto err_free_irq_cfg;
+	}
+
+	err = hinic3_alloc_rxqs_res(netdev, q_params->num_qps,
+				    q_params->rq_depth, q_params->rxqs_res);
+	if (err) {
+		netdev_err(netdev, "Failed to alloc rxqs resource\n");
+		goto err_free_txqs_res;
+	}
+
+	return 0;
+
+err_free_txqs_res:
+	hinic3_free_txqs_res(netdev, q_params->num_qps, q_params->sq_depth,
+			     q_params->txqs_res);
+err_free_irq_cfg:
+	kfree(q_params->irq_cfg);
+	q_params->irq_cfg = NULL;
+err_free_rxqs_res_arr:
+	kfree(q_params->rxqs_res);
+	q_params->rxqs_res = NULL;
+err_free_txqs_res_arr:
+	kfree(q_params->txqs_res);
+	q_params->txqs_res = NULL;
+
+	return err;
+}
+
+static void hinic3_free_txrxq_resources(struct net_device *netdev,
+					struct hinic3_dyna_txrxq_params *q_params)
+{
+	hinic3_free_rxqs_res(netdev, q_params->num_qps, q_params->rq_depth,
+			     q_params->rxqs_res);
+	hinic3_free_txqs_res(netdev, q_params->num_qps, q_params->sq_depth,
+			     q_params->txqs_res);
+
+	kfree(q_params->irq_cfg);
+	q_params->irq_cfg = NULL;
+
+	kfree(q_params->rxqs_res);
+	q_params->rxqs_res = NULL;
+
+	kfree(q_params->txqs_res);
+	q_params->txqs_res = NULL;
+}
+
+static int hinic3_configure_txrxqs(struct net_device *netdev,
+				   struct hinic3_dyna_txrxq_params *q_params)
+{
+	int err;
+
+	err = hinic3_configure_txqs(netdev, q_params->num_qps,
+				    q_params->sq_depth, q_params->txqs_res);
+	if (err) {
+		netdev_err(netdev, "Failed to configure txqs\n");
+		return err;
+	}
+
+	err = hinic3_configure_rxqs(netdev, q_params->num_qps,
+				    q_params->rq_depth, q_params->rxqs_res);
+	if (err) {
+		netdev_err(netdev, "Failed to configure rxqs\n");
+		return err;
+	}
+
+	return 0;
+}
+
+static int hinic3_configure(struct net_device *netdev)
+{
+	struct hinic3_nic_dev *nic_dev = netdev_priv(netdev);
+	int err;
+
+	netdev->min_mtu = HINIC3_MIN_MTU_SIZE;
+	netdev->max_mtu = HINIC3_MAX_JUMBO_FRAME_SIZE;
+	err = hinic3_set_port_mtu(netdev, netdev->mtu);
+	if (err) {
+		netdev_err(netdev, "Failed to set mtu\n");
+		return err;
+	}
+
+	/* Ensure DCB is disabled */
+	hinic3_sync_dcb_state(nic_dev->hwdev, 1, 0);
+
+	if (test_bit(HINIC3_RSS_ENABLE, &nic_dev->flags)) {
+		err = hinic3_rss_init(netdev);
+		if (err) {
+			netdev_err(netdev, "Failed to init rss\n");
+			return err;
+		}
+	}
+
+	return 0;
+}
+
+static void hinic3_remove_configure(struct net_device *netdev)
+{
+	struct hinic3_nic_dev *nic_dev = netdev_priv(netdev);
+
+	if (test_bit(HINIC3_RSS_ENABLE, &nic_dev->flags))
+		hinic3_rss_uninit(netdev);
+}
+
+static int hinic3_alloc_channel_resources(struct net_device *netdev,
+					  struct hinic3_dyna_qp_params *qp_params,
+					  struct hinic3_dyna_txrxq_params *trxq_params)
+{
+	struct hinic3_nic_dev *nic_dev = netdev_priv(netdev);
+	int err;
+
+	qp_params->num_qps = trxq_params->num_qps;
+	qp_params->sq_depth = trxq_params->sq_depth;
+	qp_params->rq_depth = trxq_params->rq_depth;
+
+	err = hinic3_alloc_qps(nic_dev, qp_params);
+	if (err) {
+		netdev_err(netdev, "Failed to alloc qps\n");
+		return err;
+	}
+
+	err = hinic3_alloc_txrxq_resources(netdev, trxq_params);
+	if (err) {
+		netdev_err(netdev, "Failed to alloc txrxq resources\n");
+		hinic3_free_qps(nic_dev, qp_params);
+		return err;
+	}
+
+	return 0;
+}
+
+static void hinic3_free_channel_resources(struct net_device *netdev,
+					  struct hinic3_dyna_qp_params *qp_params,
+					  struct hinic3_dyna_txrxq_params *trxq_params)
+{
+	struct hinic3_nic_dev *nic_dev = netdev_priv(netdev);
+
+	hinic3_free_txrxq_resources(netdev, trxq_params);
+	hinic3_free_qps(nic_dev, qp_params);
+}
+
+static int hinic3_open_channel(struct net_device *netdev)
+{
+	struct hinic3_nic_dev *nic_dev = netdev_priv(netdev);
+	int err;
+
+	err = hinic3_init_qp_ctxts(nic_dev);
+	if (err) {
+		netdev_err(netdev, "Failed to init qps\n");
+		return err;
+	}
+
+	err = hinic3_configure_txrxqs(netdev, &nic_dev->q_params);
+	if (err) {
+		netdev_err(netdev, "Failed to configure txrxqs\n");
+		goto err_free_qp_ctxts;
+	}
+
+	err = hinic3_qps_irq_init(netdev);
+	if (err) {
+		netdev_err(netdev, "Failed to init txrxq irq\n");
+		goto err_free_qp_ctxts;
+	}
+
+	err = hinic3_configure(netdev);
+	if (err) {
+		netdev_err(netdev, "Failed to init txrxq irq\n");
+		goto err_uninit_qps_irq;
+	}
+
+	return 0;
+
+err_uninit_qps_irq:
+	hinic3_qps_irq_uninit(netdev);
+err_free_qp_ctxts:
+	hinic3_free_qp_ctxts(nic_dev);
+
+	return err;
+}
+
+static void hinic3_close_channel(struct net_device *netdev)
+{
+	struct hinic3_nic_dev *nic_dev = netdev_priv(netdev);
+
+	hinic3_remove_configure(netdev);
+	hinic3_qps_irq_uninit(netdev);
+	hinic3_free_qp_ctxts(nic_dev);
+}
+
+static int hinic3_vport_up(struct net_device *netdev)
+{
+	struct hinic3_nic_dev *nic_dev = netdev_priv(netdev);
+	bool link_status_up;
+	u16 glb_func_id;
+	int err;
+
+	glb_func_id = hinic3_global_func_id(nic_dev->hwdev);
+	err = hinic3_set_vport_enable(nic_dev->hwdev, glb_func_id, true);
+	if (err) {
+		netdev_err(netdev, "Failed to enable vport\n");
+		goto err_flush_qps_res;
+	}
+
+	err = netif_set_real_num_queues(netdev, nic_dev->q_params.num_qps,
+					nic_dev->q_params.num_qps);
+	if (err) {
+		netdev_err(netdev, "Failed to set real number of queues\n");
+		goto err_flush_qps_res;
+	}
+	netif_tx_start_all_queues(netdev);
+
+	err = hinic3_get_link_status(nic_dev->hwdev, &link_status_up);
+	if (!err && link_status_up)
+		netif_carrier_on(netdev);
+
+	return 0;
+
+err_flush_qps_res:
+	hinic3_flush_qps_res(nic_dev->hwdev);
+	/* wait to guarantee that no packets will be sent to host */
+	msleep(100);
+
+	return err;
+}
+
+static void hinic3_vport_down(struct net_device *netdev)
+{
+	struct hinic3_nic_dev *nic_dev = netdev_priv(netdev);
+	u16 glb_func_id;
+
+	netif_carrier_off(netdev);
+	netif_tx_disable(netdev);
+
+	glb_func_id = hinic3_global_func_id(nic_dev->hwdev);
+	hinic3_set_vport_enable(nic_dev->hwdev, glb_func_id, false);
+
+	hinic3_flush_txqs(netdev);
+	/* wait to guarantee that no packets will be sent to host */
+	msleep(100);
+	hinic3_flush_qps_res(nic_dev->hwdev);
+}
+
+static int hinic3_open(struct net_device *netdev)
+{
+	struct hinic3_nic_dev *nic_dev = netdev_priv(netdev);
+	struct hinic3_dyna_qp_params qp_params;
+	int err;
+
+	err = hinic3_init_nicio_res(nic_dev);
+	if (err) {
+		netdev_err(netdev, "Failed to init nicio resources\n");
+		return err;
+	}
+
+	err = hinic3_setup_num_qps(netdev);
+	if (err) {
+		netdev_err(netdev, "Failed to setup num_qps\n");
+		goto err_free_nicio_res;
+	}
+
+	err = hinic3_alloc_channel_resources(netdev, &qp_params,
+					     &nic_dev->q_params);
+	if (err)
+		goto err_destroy_num_qps;
+
+	hinic3_init_qps(nic_dev, &qp_params);
+
+	err = hinic3_open_channel(netdev);
+	if (err)
+		goto err_uninit_qps;
+
+	err = hinic3_vport_up(netdev);
+	if (err)
+		goto err_close_channel;
+
+	return 0;
+
+err_close_channel:
+	hinic3_close_channel(netdev);
+err_uninit_qps:
+	hinic3_uninit_qps(nic_dev, &qp_params);
+	hinic3_free_channel_resources(netdev, &qp_params, &nic_dev->q_params);
+err_destroy_num_qps:
+	hinic3_destroy_num_qps(netdev);
+err_free_nicio_res:
+	hinic3_free_nicio_res(nic_dev);
+
+	return err;
+}
+
+static int hinic3_close(struct net_device *netdev)
+{
+	struct hinic3_nic_dev *nic_dev = netdev_priv(netdev);
+	struct hinic3_dyna_qp_params qp_params;
+
+	hinic3_vport_down(netdev);
+	hinic3_close_channel(netdev);
+	hinic3_uninit_qps(nic_dev, &qp_params);
+	hinic3_free_channel_resources(netdev, &qp_params, &nic_dev->q_params);
+
+	return 0;
+}
+
+static int hinic3_change_mtu(struct net_device *netdev, int new_mtu)
+{
+	int err;
+
+	err = hinic3_set_port_mtu(netdev, new_mtu);
+	if (err) {
+		netdev_err(netdev, "Failed to change port mtu to %d\n",
+			   new_mtu);
+		return err;
+	}
+
+	netdev_dbg(netdev, "Change mtu from %u to %d\n", netdev->mtu, new_mtu);
+	WRITE_ONCE(netdev->mtu, new_mtu);
+
+	return 0;
+}
+
+static int hinic3_set_mac_addr(struct net_device *netdev, void *addr)
+{
+	struct hinic3_nic_dev *nic_dev = netdev_priv(netdev);
+	struct sockaddr *saddr = addr;
+	int err;
+
+	if (!is_valid_ether_addr(saddr->sa_data))
+		return -EADDRNOTAVAIL;
+
+	if (ether_addr_equal(netdev->dev_addr, saddr->sa_data))
+		return 0;
+
+	err = hinic3_update_mac(nic_dev->hwdev, netdev->dev_addr,
+				saddr->sa_data, 0,
+				hinic3_global_func_id(nic_dev->hwdev));
+
+	if (err)
+		return err;
+
+	eth_hw_addr_set(netdev, saddr->sa_data);
+
+	return 0;
+}
+
+static const struct net_device_ops hinic3_netdev_ops = {
+	.ndo_open             = hinic3_open,
+	.ndo_stop             = hinic3_close,
+	.ndo_change_mtu       = hinic3_change_mtu,
+	.ndo_set_mac_address  = hinic3_set_mac_addr,
+	.ndo_start_xmit       = hinic3_xmit_frame,
+};
+
+void hinic3_set_netdev_ops(struct net_device *netdev)
+{
+	netdev->netdev_ops = &hinic3_netdev_ops;
+}
diff --git a/drivers/net/ethernet/huawei/hinic3/hinic3_nic_cfg.c b/drivers/net/ethernet/huawei/hinic3/hinic3_nic_cfg.c
new file mode 100644
index 000000000000..979f47ca77f9
--- /dev/null
+++ b/drivers/net/ethernet/huawei/hinic3/hinic3_nic_cfg.c
@@ -0,0 +1,385 @@
+// SPDX-License-Identifier: GPL-2.0
+// Copyright (c) Huawei Technologies Co., Ltd. 2025. All rights reserved.
+
+#include <linux/if_vlan.h>
+
+#include "hinic3_hwdev.h"
+#include "hinic3_hwif.h"
+#include "hinic3_mbox.h"
+#include "hinic3_nic_cfg.h"
+#include "hinic3_nic_dev.h"
+#include "hinic3_nic_io.h"
+
+static int hinic3_feature_nego(struct hinic3_hwdev *hwdev, u8 opcode,
+			       u64 *s_feature, u16 size)
+{
+	struct l2nic_cmd_feature_nego feature_nego = {};
+	struct mgmt_msg_params msg_params = {};
+	int err;
+
+	feature_nego.func_id = hinic3_global_func_id(hwdev);
+	feature_nego.opcode = opcode;
+	if (opcode == MGMT_MSG_CMD_OP_SET)
+		memcpy(feature_nego.s_feature, s_feature, size * sizeof(u64));
+
+	mgmt_msg_params_init_default(&msg_params, &feature_nego,
+				     sizeof(feature_nego));
+
+	err = hinic3_send_mbox_to_mgmt(hwdev, MGMT_MOD_L2NIC,
+				       L2NIC_CMD_FEATURE_NEGO, &msg_params);
+	if (err || feature_nego.msg_head.status) {
+		dev_err(hwdev->dev, "Failed to negotiate nic feature, err:%d, status: 0x%x\n",
+			err, feature_nego.msg_head.status);
+		return -EIO;
+	}
+
+	if (opcode == MGMT_MSG_CMD_OP_GET)
+		memcpy(s_feature, feature_nego.s_feature, size * sizeof(u64));
+
+	return 0;
+}
+
+int hinic3_get_nic_feature_from_hw(struct hinic3_nic_dev *nic_dev)
+{
+	return hinic3_feature_nego(nic_dev->hwdev, MGMT_MSG_CMD_OP_GET,
+				   &nic_dev->nic_io->feature_cap, 1);
+}
+
+int hinic3_set_nic_feature_to_hw(struct hinic3_nic_dev *nic_dev)
+{
+	return hinic3_feature_nego(nic_dev->hwdev, MGMT_MSG_CMD_OP_SET,
+				   &nic_dev->nic_io->feature_cap, 1);
+}
+
+bool hinic3_test_support(struct hinic3_nic_dev *nic_dev,
+			 enum hinic3_nic_feature_cap feature_bits)
+{
+	return (nic_dev->nic_io->feature_cap & feature_bits) == feature_bits;
+}
+
+void hinic3_update_nic_feature(struct hinic3_nic_dev *nic_dev, u64 feature_cap)
+{
+	nic_dev->nic_io->feature_cap = feature_cap;
+}
+
+static int hinic3_set_function_table(struct hinic3_hwdev *hwdev, u32 cfg_bitmap,
+				     const struct l2nic_func_tbl_cfg *cfg)
+{
+	struct l2nic_cmd_set_func_tbl cmd_func_tbl = {};
+	struct mgmt_msg_params msg_params = {};
+	int err;
+
+	cmd_func_tbl.func_id = hinic3_global_func_id(hwdev);
+	cmd_func_tbl.cfg_bitmap = cfg_bitmap;
+	cmd_func_tbl.tbl_cfg = *cfg;
+
+	mgmt_msg_params_init_default(&msg_params, &cmd_func_tbl,
+				     sizeof(cmd_func_tbl));
+
+	err = hinic3_send_mbox_to_mgmt(hwdev, MGMT_MOD_L2NIC,
+				       L2NIC_CMD_SET_FUNC_TBL, &msg_params);
+	if (err || cmd_func_tbl.msg_head.status) {
+		dev_err(hwdev->dev,
+			"Failed to set func table, bitmap: 0x%x, err: %d, status: 0x%x\n",
+			cfg_bitmap, err, cmd_func_tbl.msg_head.status);
+		return -EFAULT;
+	}
+
+	return 0;
+}
+
+int hinic3_init_function_table(struct hinic3_nic_dev *nic_dev)
+{
+	struct hinic3_nic_io *nic_io = nic_dev->nic_io;
+	struct l2nic_func_tbl_cfg func_tbl_cfg = {};
+	u32 cfg_bitmap;
+
+	func_tbl_cfg.mtu = 0x3FFF; /* default, max mtu */
+	func_tbl_cfg.rx_wqe_buf_size = nic_io->rx_buf_len;
+
+	cfg_bitmap = BIT(L2NIC_FUNC_TBL_CFG_INIT) |
+		     BIT(L2NIC_FUNC_TBL_CFG_MTU) |
+		     BIT(L2NIC_FUNC_TBL_CFG_RX_BUF_SIZE);
+
+	return hinic3_set_function_table(nic_dev->hwdev, cfg_bitmap,
+					 &func_tbl_cfg);
+}
+
+int hinic3_set_port_mtu(struct net_device *netdev, u16 new_mtu)
+{
+	struct hinic3_nic_dev *nic_dev = netdev_priv(netdev);
+	struct l2nic_func_tbl_cfg func_tbl_cfg = {};
+	struct hinic3_hwdev *hwdev = nic_dev->hwdev;
+
+	func_tbl_cfg.mtu = new_mtu;
+
+	return hinic3_set_function_table(hwdev, BIT(L2NIC_FUNC_TBL_CFG_MTU),
+					 &func_tbl_cfg);
+}
+
+static int hinic3_check_mac_info(struct hinic3_hwdev *hwdev, u8 status,
+				 u16 vlan_id)
+{
+	if ((status && status != MGMT_STATUS_EXIST) ||
+	    ((vlan_id & BIT(15)) && status == MGMT_STATUS_EXIST)) {
+		return -EINVAL;
+	}
+
+	return 0;
+}
+
+int hinic3_set_mac(struct hinic3_hwdev *hwdev, const u8 *mac_addr, u16 vlan_id,
+		   u16 func_id)
+{
+	struct l2nic_cmd_set_mac mac_info = {};
+	struct mgmt_msg_params msg_params = {};
+	int err;
+
+	if ((vlan_id & HINIC3_VLAN_ID_MASK) >= VLAN_N_VID) {
+		dev_err(hwdev->dev, "Invalid VLAN number: %d\n",
+			(vlan_id & HINIC3_VLAN_ID_MASK));
+		return -EINVAL;
+	}
+
+	mac_info.func_id = func_id;
+	mac_info.vlan_id = vlan_id;
+	ether_addr_copy(mac_info.mac, mac_addr);
+
+	mgmt_msg_params_init_default(&msg_params, &mac_info, sizeof(mac_info));
+
+	err = hinic3_send_mbox_to_mgmt(hwdev, MGMT_MOD_L2NIC,
+				       L2NIC_CMD_SET_MAC, &msg_params);
+	if (err || hinic3_check_mac_info(hwdev, mac_info.msg_head.status,
+					 mac_info.vlan_id)) {
+		dev_err(hwdev->dev,
+			"Failed to update MAC, err: %d, status: 0x%x\n",
+			err, mac_info.msg_head.status);
+		return -EIO;
+	}
+
+	if (mac_info.msg_head.status == MGMT_STATUS_PF_SET_VF_ALREADY) {
+		dev_warn(hwdev->dev, "PF has already set VF mac, Ignore set operation\n");
+		return 0;
+	}
+
+	if (mac_info.msg_head.status == MGMT_STATUS_EXIST) {
+		dev_warn(hwdev->dev, "MAC is repeated. Ignore update operation\n");
+		return 0;
+	}
+
+	return 0;
+}
+
+int hinic3_del_mac(struct hinic3_hwdev *hwdev, const u8 *mac_addr, u16 vlan_id,
+		   u16 func_id)
+{
+	struct l2nic_cmd_set_mac mac_info = {};
+	struct mgmt_msg_params msg_params = {};
+	int err;
+
+	if ((vlan_id & HINIC3_VLAN_ID_MASK) >= VLAN_N_VID) {
+		dev_err(hwdev->dev, "Invalid VLAN number: %d\n",
+			(vlan_id & HINIC3_VLAN_ID_MASK));
+		return -EINVAL;
+	}
+
+	mac_info.func_id = func_id;
+	mac_info.vlan_id = vlan_id;
+	ether_addr_copy(mac_info.mac, mac_addr);
+
+	mgmt_msg_params_init_default(&msg_params, &mac_info, sizeof(mac_info));
+
+	err = hinic3_send_mbox_to_mgmt(hwdev, MGMT_MOD_L2NIC,
+				       L2NIC_CMD_DEL_MAC, &msg_params);
+	if (err) {
+		dev_err(hwdev->dev,
+			"Failed to delete MAC, err: %d, status: 0x%x\n",
+			err, mac_info.msg_head.status);
+		return err;
+	}
+
+	return 0;
+}
+
+int hinic3_update_mac(struct hinic3_hwdev *hwdev, const u8 *old_mac,
+		      u8 *new_mac, u16 vlan_id, u16 func_id)
+{
+	struct l2nic_cmd_update_mac mac_info = {};
+	struct mgmt_msg_params msg_params = {};
+	int err;
+
+	if ((vlan_id & HINIC3_VLAN_ID_MASK) >= VLAN_N_VID) {
+		dev_err(hwdev->dev, "Invalid VLAN number: %d\n",
+			(vlan_id & HINIC3_VLAN_ID_MASK));
+		return -EINVAL;
+	}
+
+	mac_info.func_id = func_id;
+	mac_info.vlan_id = vlan_id;
+	ether_addr_copy(mac_info.old_mac, old_mac);
+	ether_addr_copy(mac_info.new_mac, new_mac);
+
+	mgmt_msg_params_init_default(&msg_params, &mac_info, sizeof(mac_info));
+
+	err = hinic3_send_mbox_to_mgmt(hwdev, MGMT_MOD_L2NIC,
+				       L2NIC_CMD_UPDATE_MAC, &msg_params);
+	if (err || hinic3_check_mac_info(hwdev, mac_info.msg_head.status,
+					 mac_info.vlan_id)) {
+		dev_err(hwdev->dev,
+			"Failed to update MAC, err: %d, status: 0x%x\n",
+			err, mac_info.msg_head.status);
+		return -EIO;
+	}
+
+	return 0;
+}
+
+int hinic3_set_ci_table(struct hinic3_hwdev *hwdev, struct hinic3_sq_attr *attr)
+{
+	struct l2nic_cmd_set_ci_attr cons_idx_attr = {};
+	struct mgmt_msg_params msg_params = {};
+	int err;
+
+	cons_idx_attr.func_idx = hinic3_global_func_id(hwdev);
+	cons_idx_attr.dma_attr_off  = attr->dma_attr_off;
+	cons_idx_attr.pending_limit = attr->pending_limit;
+	cons_idx_attr.coalescing_time  = attr->coalescing_time;
+
+	if (attr->intr_en) {
+		cons_idx_attr.intr_en = attr->intr_en;
+		cons_idx_attr.intr_idx = attr->intr_idx;
+	}
+
+	cons_idx_attr.l2nic_sqn = attr->l2nic_sqn;
+	cons_idx_attr.ci_addr = attr->ci_dma_base;
+
+	mgmt_msg_params_init_default(&msg_params, &cons_idx_attr,
+				     sizeof(cons_idx_attr));
+
+	err = hinic3_send_mbox_to_mgmt(hwdev, MGMT_MOD_L2NIC,
+				       L2NIC_CMD_SET_SQ_CI_ATTR, &msg_params);
+	if (err || cons_idx_attr.msg_head.status) {
+		dev_err(hwdev->dev,
+			"Failed to set ci attribute table, err: %d, status: 0x%x\n",
+			err, cons_idx_attr.msg_head.status);
+		return -EFAULT;
+	}
+
+	return 0;
+}
+
+int hinic3_flush_qps_res(struct hinic3_hwdev *hwdev)
+{
+	struct l2nic_cmd_clear_qp_resource sq_res = {};
+	struct mgmt_msg_params msg_params = {};
+	int err;
+
+	sq_res.func_id = hinic3_global_func_id(hwdev);
+
+	mgmt_msg_params_init_default(&msg_params, &sq_res, sizeof(sq_res));
+
+	err = hinic3_send_mbox_to_mgmt(hwdev, MGMT_MOD_L2NIC,
+				       L2NIC_CMD_CLEAR_QP_RESOURCE,
+				       &msg_params);
+	if (err || sq_res.msg_head.status) {
+		dev_err(hwdev->dev, "Failed to clear sq resources, err: %d, status: 0x%x\n",
+			err, sq_res.msg_head.status);
+		return -EINVAL;
+	}
+
+	return 0;
+}
+
+int hinic3_force_drop_tx_pkt(struct hinic3_hwdev *hwdev)
+{
+	struct l2nic_cmd_force_pkt_drop pkt_drop = {};
+	struct mgmt_msg_params msg_params = {};
+	int err;
+
+	pkt_drop.port = hinic3_physical_port_id(hwdev);
+
+	mgmt_msg_params_init_default(&msg_params, &pkt_drop, sizeof(pkt_drop));
+
+	err = hinic3_send_mbox_to_mgmt(hwdev, MGMT_MOD_L2NIC,
+				       L2NIC_CMD_FORCE_PKT_DROP, &msg_params);
+	if ((pkt_drop.msg_head.status != MGMT_STATUS_CMD_UNSUPPORTED &&
+	     pkt_drop.msg_head.status) || err) {
+		dev_err(hwdev->dev,
+			"Failed to set force tx packets drop, err: %d, status: 0x%x\n",
+			err, pkt_drop.msg_head.status);
+		return -EFAULT;
+	}
+
+	return pkt_drop.msg_head.status;
+}
+
+int hinic3_sync_dcb_state(struct hinic3_hwdev *hwdev, u8 op_code, u8 state)
+{
+	struct l2nic_cmd_set_dcb_state dcb_state = {};
+	struct mgmt_msg_params msg_params = {};
+	int err;
+
+	dcb_state.op_code = op_code;
+	dcb_state.state = state;
+	dcb_state.func_id = hinic3_global_func_id(hwdev);
+
+	mgmt_msg_params_init_default(&msg_params, &dcb_state,
+				     sizeof(dcb_state));
+
+	err = hinic3_send_mbox_to_mgmt(hwdev, MGMT_MOD_L2NIC,
+				       L2NIC_CMD_QOS_DCB_STATE, &msg_params);
+	if (err || dcb_state.head.status) {
+		dev_err(hwdev->dev,
+			"Failed to set dcb state, err: %d, status: 0x%x\n",
+			err, dcb_state.head.status);
+		return -EFAULT;
+	}
+
+	return 0;
+}
+
+int hinic3_get_link_status(struct hinic3_hwdev *hwdev, bool *link_status_up)
+{
+	struct mag_cmd_get_link_status get_link = {};
+	struct mgmt_msg_params msg_params = {};
+	int err;
+
+	get_link.port_id = hinic3_physical_port_id(hwdev);
+
+	mgmt_msg_params_init_default(&msg_params, &get_link, sizeof(get_link));
+
+	err = hinic3_send_mbox_to_mgmt(hwdev, MGMT_MOD_HILINK,
+				       MAG_CMD_GET_LINK_STATUS, &msg_params);
+	if (err || get_link.head.status) {
+		dev_err(hwdev->dev, "Failed to get link state, err: %d, status: 0x%x\n",
+			err, get_link.head.status);
+		return -EIO;
+	}
+
+	*link_status_up = !!get_link.status;
+
+	return 0;
+}
+
+int hinic3_set_vport_enable(struct hinic3_hwdev *hwdev, u16 func_id,
+			    bool enable)
+{
+	struct l2nic_cmd_set_vport_state en_state = {};
+	struct mgmt_msg_params msg_params = {};
+	int err;
+
+	en_state.func_id = func_id;
+	en_state.state = enable ? 1 : 0;
+
+	mgmt_msg_params_init_default(&msg_params, &en_state, sizeof(en_state));
+
+	err = hinic3_send_mbox_to_mgmt(hwdev, MGMT_MOD_L2NIC,
+				       L2NIC_CMD_SET_VPORT_ENABLE, &msg_params);
+	if (err || en_state.msg_head.status) {
+		dev_err(hwdev->dev, "Failed to set vport state, err: %d, status: 0x%x\n",
+			err, en_state.msg_head.status);
+		return -EINVAL;
+	}
+
+	return 0;
+}
diff --git a/drivers/net/ethernet/huawei/hinic3/hinic3_nic_cfg.h b/drivers/net/ethernet/huawei/hinic3/hinic3_nic_cfg.h
new file mode 100644
index 000000000000..b83b567fa542
--- /dev/null
+++ b/drivers/net/ethernet/huawei/hinic3/hinic3_nic_cfg.h
@@ -0,0 +1,61 @@
+/* SPDX-License-Identifier: GPL-2.0 */
+/* Copyright (c) Huawei Technologies Co., Ltd. 2025. All rights reserved. */
+
+#ifndef _HINIC3_NIC_CFG_H_
+#define _HINIC3_NIC_CFG_H_
+
+#include <linux/types.h>
+
+#include "hinic3_hw_intf.h"
+#include "hinic3_mgmt_interface.h"
+
+struct hinic3_hwdev;
+struct hinic3_nic_dev;
+
+#define HINIC3_MIN_MTU_SIZE          256
+#define HINIC3_MAX_JUMBO_FRAME_SIZE  9600
+
+#define HINIC3_VLAN_ID_MASK          0x7FFF
+
+enum hinic3_nic_event_type {
+	HINIC3_NIC_EVENT_LINK_DOWN = 0,
+	HINIC3_NIC_EVENT_LINK_UP   = 1,
+};
+
+struct hinic3_sq_attr {
+	u8  dma_attr_off;
+	u8  pending_limit;
+	u8  coalescing_time;
+	u8  intr_en;
+	u16 intr_idx;
+	u32 l2nic_sqn;
+	u64 ci_dma_base;
+};
+
+int hinic3_get_nic_feature_from_hw(struct hinic3_nic_dev *nic_dev);
+int hinic3_set_nic_feature_to_hw(struct hinic3_nic_dev *nic_dev);
+bool hinic3_test_support(struct hinic3_nic_dev *nic_dev,
+			 enum hinic3_nic_feature_cap feature_bits);
+void hinic3_update_nic_feature(struct hinic3_nic_dev *nic_dev, u64 feature_cap);
+
+int hinic3_init_function_table(struct hinic3_nic_dev *nic_dev);
+int hinic3_set_port_mtu(struct net_device *netdev, u16 new_mtu);
+
+int hinic3_set_mac(struct hinic3_hwdev *hwdev, const u8 *mac_addr, u16 vlan_id,
+		   u16 func_id);
+int hinic3_del_mac(struct hinic3_hwdev *hwdev, const u8 *mac_addr, u16 vlan_id,
+		   u16 func_id);
+int hinic3_update_mac(struct hinic3_hwdev *hwdev, const u8 *old_mac,
+		      u8 *new_mac, u16 vlan_id, u16 func_id);
+
+int hinic3_set_ci_table(struct hinic3_hwdev *hwdev,
+			struct hinic3_sq_attr *attr);
+int hinic3_flush_qps_res(struct hinic3_hwdev *hwdev);
+int hinic3_force_drop_tx_pkt(struct hinic3_hwdev *hwdev);
+
+int hinic3_sync_dcb_state(struct hinic3_hwdev *hwdev, u8 op_code, u8 state);
+int hinic3_get_link_status(struct hinic3_hwdev *hwdev, bool *link_status_up);
+int hinic3_set_vport_enable(struct hinic3_hwdev *hwdev, u16 func_id,
+			    bool enable);
+
+#endif
diff --git a/drivers/net/ethernet/huawei/hinic3/hinic3_nic_dev.h b/drivers/net/ethernet/huawei/hinic3/hinic3_nic_dev.h
new file mode 100644
index 000000000000..5ba83261616c
--- /dev/null
+++ b/drivers/net/ethernet/huawei/hinic3/hinic3_nic_dev.h
@@ -0,0 +1,93 @@
+/* SPDX-License-Identifier: GPL-2.0 */
+/* Copyright (c) Huawei Technologies Co., Ltd. 2025. All rights reserved. */
+
+#ifndef _HINIC3_NIC_DEV_H_
+#define _HINIC3_NIC_DEV_H_
+
+#include <linux/netdevice.h>
+
+#include "hinic3_hw_cfg.h"
+#include "hinic3_mgmt_interface.h"
+
+enum hinic3_flags {
+	HINIC3_RSS_ENABLE,
+};
+
+enum hinic3_rss_hash_type {
+	HINIC3_RSS_HASH_ENGINE_TYPE_XOR  = 0,
+	HINIC3_RSS_HASH_ENGINE_TYPE_TOEP = 1,
+};
+
+struct hinic3_rss_type {
+	u8 tcp_ipv6_ext;
+	u8 ipv6_ext;
+	u8 tcp_ipv6;
+	u8 ipv6;
+	u8 tcp_ipv4;
+	u8 ipv4;
+	u8 udp_ipv6;
+	u8 udp_ipv4;
+};
+
+struct hinic3_irq_cfg {
+	struct net_device  *netdev;
+	u16                msix_entry_idx;
+	/* provided by OS */
+	u32                irq_id;
+	char               irq_name[IFNAMSIZ + 16];
+	struct napi_struct napi;
+	cpumask_t          affinity_mask;
+	struct hinic3_txq  *txq;
+	struct hinic3_rxq  *rxq;
+};
+
+struct hinic3_dyna_txrxq_params {
+	u16                        num_qps;
+	u32                        sq_depth;
+	u32                        rq_depth;
+
+	struct hinic3_dyna_txq_res *txqs_res;
+	struct hinic3_dyna_rxq_res *rxqs_res;
+	struct hinic3_irq_cfg      *irq_cfg;
+};
+
+struct hinic3_intr_coal_info {
+	u8 pending_limit;
+	u8 coalesce_timer_cfg;
+	u8 resend_timer_cfg;
+};
+
+struct hinic3_nic_dev {
+	struct pci_dev                  *pdev;
+	struct net_device               *netdev;
+	struct hinic3_hwdev             *hwdev;
+	struct hinic3_nic_io            *nic_io;
+
+	u16                             max_qps;
+	u16                             rx_buf_len;
+	u32                             lro_replenish_thld;
+	unsigned long                   flags;
+	struct hinic3_nic_service_cap   nic_svc_cap;
+
+	struct hinic3_dyna_txrxq_params q_params;
+	struct hinic3_txq               *txqs;
+	struct hinic3_rxq               *rxqs;
+
+	enum hinic3_rss_hash_type       rss_hash_type;
+	struct hinic3_rss_type          rss_type;
+	u8                              *rss_hkey;
+	u16                             *rss_indir;
+
+	u16                             num_qp_irq;
+	struct msix_entry               *qps_msix_entries;
+
+	struct hinic3_intr_coal_info    *intr_coalesce;
+
+	bool                            link_status_up;
+};
+
+void hinic3_set_netdev_ops(struct net_device *netdev);
+int hinic3_qps_irq_init(struct net_device *netdev);
+void hinic3_qps_irq_uninit(struct net_device *netdev);
+
+#endif
diff --git a/drivers/net/ethernet/huawei/hinic3/hinic3_nic_io.c b/drivers/net/ethernet/huawei/hinic3/hinic3_nic_io.c
new file mode 100644
index 000000000000..d86cd1ba4605
--- /dev/null
+++ b/drivers/net/ethernet/huawei/hinic3/hinic3_nic_io.c
@@ -0,0 +1,885 @@
+// SPDX-License-Identifier: GPL-2.0
+// Copyright (c) Huawei Technologies Co., Ltd. 2025. All rights reserved.
+
+#include "hinic3_cmdq.h"
+#include "hinic3_hw_comm.h"
+#include "hinic3_hw_intf.h"
+#include "hinic3_hwdev.h"
+#include "hinic3_hwif.h"
+#include "hinic3_nic_cfg.h"
+#include "hinic3_nic_dev.h"
+#include "hinic3_nic_io.h"
+
+#define HINIC3_DEFAULT_TX_CI_PENDING_LIMIT    1
+#define HINIC3_DEFAULT_TX_CI_COALESCING_TIME  1
+#define HINIC3_DEFAULT_DROP_THD_ON            (0xFFFF)
+#define HINIC3_DEFAULT_DROP_THD_OFF           0
+
+#define HINIC3_CI_Q_ADDR_SIZE                (64)
+
+#define HINIC3_CI_TABLE_SIZE(num_qps)  \
+	(ALIGN((num_qps) * HINIC3_CI_Q_ADDR_SIZE, HINIC3_MIN_PAGE_SIZE))
+
+#define HINIC3_CI_VADDR(base_addr, q_id)  \
+	((u8 *)(base_addr) + (q_id) * HINIC3_CI_Q_ADDR_SIZE)
+
+#define HINIC3_CI_PADDR(base_paddr, q_id)  \
+	((base_paddr) + (q_id) * HINIC3_CI_Q_ADDR_SIZE)
+
+#define SQ_WQ_PREFETCH_MAX        1
+#define SQ_WQ_PREFETCH_MIN        1
+#define SQ_WQ_PREFETCH_THRESHOLD  16
+
+#define RQ_WQ_PREFETCH_MAX        4
+#define RQ_WQ_PREFETCH_MIN        1
+#define RQ_WQ_PREFETCH_THRESHOLD  256
+
+/* (2048 - 8) / 64 */
+#define HINIC3_Q_CTXT_MAX         31
+
+enum hinic3_qp_ctxt_type {
+	HINIC3_QP_CTXT_TYPE_SQ = 0,
+	HINIC3_QP_CTXT_TYPE_RQ = 1,
+};
+
+struct hinic3_qp_ctxt_hdr {
+	__le16 num_queues;
+	__le16 queue_type;
+	__le16 start_qid;
+	__le16 rsvd;
+};
+
+struct hinic3_sq_ctxt {
+	__le32 ci_pi;
+	__le32 drop_mode_sp;
+	__le32 wq_pfn_hi_owner;
+	__le32 wq_pfn_lo;
+
+	__le32 rsvd0;
+	__le32 pkt_drop_thd;
+	__le32 global_sq_id;
+	__le32 vlan_ceq_attr;
+
+	__le32 pref_cache;
+	__le32 pref_ci_owner;
+	__le32 pref_wq_pfn_hi_ci;
+	__le32 pref_wq_pfn_lo;
+
+	__le32 rsvd8;
+	__le32 rsvd9;
+	__le32 wq_block_pfn_hi;
+	__le32 wq_block_pfn_lo;
+};
+
+struct hinic3_rq_ctxt {
+	__le32 ci_pi;
+	__le32 ceq_attr;
+	__le32 wq_pfn_hi_type_owner;
+	__le32 wq_pfn_lo;
+
+	__le32 rsvd[3];
+	__le32 cqe_sge_len;
+
+	__le32 pref_cache;
+	__le32 pref_ci_owner;
+	__le32 pref_wq_pfn_hi_ci;
+	__le32 pref_wq_pfn_lo;
+
+	__le32 pi_paddr_hi;
+	__le32 pi_paddr_lo;
+	__le32 wq_block_pfn_hi;
+	__le32 wq_block_pfn_lo;
+};
+
+struct hinic3_sq_ctxt_block {
+	struct hinic3_qp_ctxt_hdr cmdq_hdr;
+	struct hinic3_sq_ctxt     sq_ctxt[HINIC3_Q_CTXT_MAX];
+};
+
+struct hinic3_rq_ctxt_block {
+	struct hinic3_qp_ctxt_hdr cmdq_hdr;
+	struct hinic3_rq_ctxt     rq_ctxt[HINIC3_Q_CTXT_MAX];
+};
+
+struct hinic3_clean_queue_ctxt {
+	struct hinic3_qp_ctxt_hdr cmdq_hdr;
+	__le32                    rsvd;
+};
+
+#define SQ_CTXT_SIZE(num_sqs)  \
+	(sizeof(struct hinic3_qp_ctxt_hdr) +  \
+	(num_sqs) * sizeof(struct hinic3_sq_ctxt))
+
+#define RQ_CTXT_SIZE(num_rqs)  \
+	(sizeof(struct hinic3_qp_ctxt_hdr) +  \
+	(num_rqs) * sizeof(struct hinic3_rq_ctxt))
+
+#define SQ_CTXT_PREF_CI_HI_SHIFT           12
+#define SQ_CTXT_PREF_CI_HI(val)            ((val) >> SQ_CTXT_PREF_CI_HI_SHIFT)
+
+#define SQ_CTXT_PI_IDX_MASK                GENMASK(15, 0)
+#define SQ_CTXT_CI_IDX_MASK                GENMASK(31, 16)
+#define SQ_CTXT_CI_PI_SET(val, member)  \
+	FIELD_PREP(SQ_CTXT_##member##_MASK, val)
+
+#define SQ_CTXT_MODE_SP_FLAG_MASK          BIT(0)
+#define SQ_CTXT_MODE_PKT_DROP_MASK         BIT(1)
+#define SQ_CTXT_MODE_SET(val, member)  \
+	FIELD_PREP(SQ_CTXT_MODE_##member##_MASK, val)
+
+#define SQ_CTXT_WQ_PAGE_HI_PFN_MASK        GENMASK(19, 0)
+#define SQ_CTXT_WQ_PAGE_OWNER_MASK         BIT(23)
+#define SQ_CTXT_WQ_PAGE_SET(val, member)  \
+	FIELD_PREP(SQ_CTXT_WQ_PAGE_##member##_MASK, val)
+
+#define SQ_CTXT_PKT_DROP_THD_ON_MASK       GENMASK(15, 0)
+#define SQ_CTXT_PKT_DROP_THD_OFF_MASK      GENMASK(31, 16)
+#define SQ_CTXT_PKT_DROP_THD_SET(val, member)  \
+	FIELD_PREP(SQ_CTXT_PKT_DROP_##member##_MASK, val)
+
+#define SQ_CTXT_GLOBAL_SQ_ID_MASK          GENMASK(12, 0)
+#define SQ_CTXT_GLOBAL_QUEUE_ID_SET(val, member)  \
+	FIELD_PREP(SQ_CTXT_##member##_MASK, val)
+
+#define SQ_CTXT_VLAN_INSERT_MODE_MASK      GENMASK(20, 19)
+#define SQ_CTXT_VLAN_CEQ_EN_MASK           BIT(23)
+#define SQ_CTXT_VLAN_CEQ_SET(val, member)  \
+	FIELD_PREP(SQ_CTXT_VLAN_##member##_MASK, val)
+
+#define SQ_CTXT_PREF_CACHE_THRESHOLD_MASK  GENMASK(13, 0)
+#define SQ_CTXT_PREF_CACHE_MAX_MASK        GENMASK(24, 14)
+#define SQ_CTXT_PREF_CACHE_MIN_MASK        GENMASK(31, 25)
+
+#define SQ_CTXT_PREF_CI_HI_MASK            GENMASK(3, 0)
+#define SQ_CTXT_PREF_OWNER_MASK            BIT(4)
+
+#define SQ_CTXT_PREF_WQ_PFN_HI_MASK        GENMASK(19, 0)
+#define SQ_CTXT_PREF_CI_LOW_MASK           GENMASK(31, 20)
+#define SQ_CTXT_PREF_SET(val, member)  \
+	FIELD_PREP(SQ_CTXT_PREF_##member##_MASK, val)
+
+#define SQ_CTXT_WQ_BLOCK_PFN_HI_MASK       GENMASK(22, 0)
+#define SQ_CTXT_WQ_BLOCK_SET(val, member)  \
+	FIELD_PREP(SQ_CTXT_WQ_BLOCK_##member##_MASK, val)
+
+#define RQ_CTXT_PI_IDX_MASK                GENMASK(15, 0)
+#define RQ_CTXT_CI_IDX_MASK                GENMASK(31, 16)
+#define RQ_CTXT_CI_PI_SET(val, member)  \
+	FIELD_PREP(RQ_CTXT_##member##_MASK, val)
+
+#define RQ_CTXT_CEQ_ATTR_INTR_MASK         GENMASK(30, 21)
+#define RQ_CTXT_CEQ_ATTR_EN_MASK           BIT(31)
+#define RQ_CTXT_CEQ_ATTR_SET(val, member)  \
+	FIELD_PREP(RQ_CTXT_CEQ_ATTR_##member##_MASK, val)
+
+#define RQ_CTXT_WQ_PAGE_HI_PFN_MASK        GENMASK(19, 0)
+#define RQ_CTXT_WQ_PAGE_WQE_TYPE_MASK      GENMASK(29, 28)
+#define RQ_CTXT_WQ_PAGE_OWNER_MASK         BIT(31)
+#define RQ_CTXT_WQ_PAGE_SET(val, member)  \
+	FIELD_PREP(RQ_CTXT_WQ_PAGE_##member##_MASK, val)
+
+#define RQ_CTXT_CQE_LEN_MASK               GENMASK(29, 28)
+#define RQ_CTXT_CQE_LEN_SET(val, member)  \
+	FIELD_PREP(RQ_CTXT_##member##_MASK, val)
+
+#define RQ_CTXT_PREF_CACHE_THRESHOLD_MASK  GENMASK(13, 0)
+#define RQ_CTXT_PREF_CACHE_MAX_MASK        GENMASK(24, 14)
+#define RQ_CTXT_PREF_CACHE_MIN_MASK        GENMASK(31, 25)
+
+#define RQ_CTXT_PREF_CI_HI_MASK            GENMASK(3, 0)
+#define RQ_CTXT_PREF_OWNER_MASK            BIT(4)
+
+#define RQ_CTXT_PREF_WQ_PFN_HI_MASK        GENMASK(19, 0)
+#define RQ_CTXT_PREF_CI_LOW_MASK           GENMASK(31, 20)
+#define RQ_CTXT_PREF_SET(val, member)  \
+	FIELD_PREP(RQ_CTXT_PREF_##member##_MASK, val)
+
+#define RQ_CTXT_WQ_BLOCK_PFN_HI_MASK       GENMASK(22, 0)
+#define RQ_CTXT_WQ_BLOCK_SET(val, member)  \
+	FIELD_PREP(RQ_CTXT_WQ_BLOCK_##member##_MASK, val)
+
+#define WQ_PAGE_PFN_SHIFT       12
+#define WQ_BLOCK_PFN_SHIFT      9
+#define WQ_PAGE_PFN(page_addr)  ((page_addr) >> WQ_PAGE_PFN_SHIFT)
+#define WQ_BLOCK_PFN(page_addr) ((page_addr) >> WQ_BLOCK_PFN_SHIFT)
+
+int hinic3_init_nic_io(struct hinic3_nic_dev *nic_dev)
+{
+	struct hinic3_hwdev *hwdev = nic_dev->hwdev;
+	struct hinic3_nic_io *nic_io;
+	int err;
+
+	nic_io = kzalloc(sizeof(*nic_io), GFP_KERNEL);
+	if (!nic_io)
+		return -ENOMEM;
+
+	nic_dev->nic_io = nic_io;
+
+	err = hinic3_set_func_svc_used_state(hwdev, COMM_FUNC_SVC_T_NIC, 1);
+	if (err) {
+		dev_err(hwdev->dev, "Failed to set function svc used state\n");
+		goto err_free_nicio;
+	}
+
+	err = hinic3_init_function_table(nic_dev);
+	if (err) {
+		dev_err(hwdev->dev, "Failed to init function table\n");
+		goto err_clear_func_svc_used_state;
+	}
+
+	nic_io->rx_buf_len = nic_dev->rx_buf_len;
+
+	err = hinic3_get_nic_feature_from_hw(nic_dev);
+	if (err) {
+		dev_err(hwdev->dev, "Failed to get nic features\n");
+		goto err_clear_func_svc_used_state;
+	}
+
+	nic_io->feature_cap &= HINIC3_NIC_F_ALL_MASK;
+	nic_io->feature_cap &= HINIC3_NIC_DRV_DEFAULT_FEATURE;
+	dev_dbg(hwdev->dev, "nic features: 0x%llx\n\n", nic_io->feature_cap);
+
+	return 0;
+
+err_clear_func_svc_used_state:
+	hinic3_set_func_svc_used_state(hwdev, COMM_FUNC_SVC_T_NIC, 0);
+err_free_nicio:
+	nic_dev->nic_io = NULL;
+	kfree(nic_io);
+
+	return err;
+}
+
+void hinic3_free_nic_io(struct hinic3_nic_dev *nic_dev)
+{
+	struct hinic3_nic_io *nic_io = nic_dev->nic_io;
+
+	hinic3_set_func_svc_used_state(nic_dev->hwdev, COMM_FUNC_SVC_T_NIC, 0);
+	nic_dev->nic_io = NULL;
+	kfree(nic_io);
+}
+
+int hinic3_init_nicio_res(struct hinic3_nic_dev *nic_dev)
+{
+	struct hinic3_nic_io *nic_io = nic_dev->nic_io;
+	struct hinic3_hwdev *hwdev = nic_dev->hwdev;
+	void __iomem *db_base;
+	int err;
+
+	nic_io->max_qps = hinic3_func_max_qnum(hwdev);
+
+	err = hinic3_alloc_db_addr(hwdev, &db_base, NULL);
+	if (err) {
+		dev_err(hwdev->dev, "Failed to allocate doorbell for sqs\n");
+		return err;
+	}
+	nic_io->sqs_db_addr = db_base;
+
+	err = hinic3_alloc_db_addr(hwdev, &db_base, NULL);
+	if (err) {
+		hinic3_free_db_addr(hwdev, nic_io->sqs_db_addr);
+		dev_err(hwdev->dev, "Failed to allocate doorbell for rqs\n");
+		return err;
+	}
+	nic_io->rqs_db_addr = db_base;
+
+	nic_io->ci_vaddr_base =
+		dma_alloc_coherent(hwdev->dev,
+				   HINIC3_CI_TABLE_SIZE(nic_io->max_qps),
+				   &nic_io->ci_dma_base,
+				   GFP_KERNEL);
+	if (!nic_io->ci_vaddr_base) {
+		hinic3_free_db_addr(hwdev, nic_io->sqs_db_addr);
+		hinic3_free_db_addr(hwdev, nic_io->rqs_db_addr);
+		return -ENOMEM;
+	}
+
+	return 0;
+}
+
+void hinic3_free_nicio_res(struct hinic3_nic_dev *nic_dev)
+{
+	struct hinic3_nic_io *nic_io = nic_dev->nic_io;
+	struct hinic3_hwdev *hwdev = nic_dev->hwdev;
+
+	dma_free_coherent(hwdev->dev,
+			  HINIC3_CI_TABLE_SIZE(nic_io->max_qps),
+			  nic_io->ci_vaddr_base, nic_io->ci_dma_base);
+
+	hinic3_free_db_addr(hwdev, nic_io->sqs_db_addr);
+	hinic3_free_db_addr(hwdev, nic_io->rqs_db_addr);
+}
+
+static int hinic3_create_sq(struct hinic3_hwdev *hwdev,
+			    struct hinic3_io_queue *sq,
+			    u16 q_id, u32 sq_depth, u16 sq_msix_idx)
+{
+	int err;
+
+	/* sq used & hardware request init 1 */
+	sq->owner = 1;
+
+	sq->q_id = q_id;
+	sq->msix_entry_idx = sq_msix_idx;
+
+	err = hinic3_wq_create(hwdev, &sq->wq, sq_depth,
+			       BIT(HINIC3_SQ_WQEBB_SHIFT));
+	if (err) {
+		dev_err(hwdev->dev, "Failed to create tx queue %u wq\n",
+			q_id);
+		return err;
+	}
+
+	return 0;
+}
+
+static int hinic3_create_rq(struct hinic3_hwdev *hwdev,
+			    struct hinic3_io_queue *rq,
+			    u16 q_id, u32 rq_depth, u16 rq_msix_idx)
+{
+	int err;
+
+	rq->q_id = q_id;
+	rq->msix_entry_idx = rq_msix_idx;
+
+	err = hinic3_wq_create(hwdev, &rq->wq, rq_depth,
+			       BIT(HINIC3_RQ_WQEBB_SHIFT +
+				   HINIC3_NORMAL_RQ_WQE));
+	if (err) {
+		dev_err(hwdev->dev, "Failed to create rx queue %u wq\n",
+			q_id);
+		return err;
+	}
+
+	return 0;
+}
+
+static int hinic3_create_qp(struct hinic3_hwdev *hwdev,
+			    struct hinic3_io_queue *sq,
+			    struct hinic3_io_queue *rq, u16 q_id, u32 sq_depth,
+			    u32 rq_depth, u16 qp_msix_idx)
+{
+	int err;
+
+	err = hinic3_create_sq(hwdev, sq, q_id, sq_depth, qp_msix_idx);
+	if (err) {
+		dev_err(hwdev->dev, "Failed to create sq, qid: %u\n",
+			q_id);
+		return err;
+	}
+
+	err = hinic3_create_rq(hwdev, rq, q_id, rq_depth, qp_msix_idx);
+	if (err) {
+		dev_err(hwdev->dev, "Failed to create rq, qid: %u\n",
+			q_id);
+		goto err_destroy_sq_wq;
+	}
+
+	return 0;
+
+err_destroy_sq_wq:
+	hinic3_wq_destroy(hwdev, &sq->wq);
+
+	return err;
+}
+
+static void hinic3_destroy_qp(struct hinic3_hwdev *hwdev,
+			      struct hinic3_io_queue *sq,
+			      struct hinic3_io_queue *rq)
+{
+	hinic3_wq_destroy(hwdev, &sq->wq);
+	hinic3_wq_destroy(hwdev, &rq->wq);
+}
+
+int hinic3_alloc_qps(struct hinic3_nic_dev *nic_dev,
+		     struct hinic3_dyna_qp_params *qp_params)
+{
+	struct msix_entry *qps_msix_entries = nic_dev->qps_msix_entries;
+	struct hinic3_nic_io *nic_io = nic_dev->nic_io;
+	struct hinic3_hwdev *hwdev = nic_dev->hwdev;
+	struct hinic3_io_queue *sqs;
+	struct hinic3_io_queue *rqs;
+	u16 q_id;
+	int err;
+
+	if (qp_params->num_qps > nic_io->max_qps || !qp_params->num_qps)
+		return -EINVAL;
+
+	sqs = kcalloc(qp_params->num_qps, sizeof(*sqs), GFP_KERNEL);
+	if (!sqs) {
+		err = -ENOMEM;
+		goto err_out;
+	}
+
+	rqs = kcalloc(qp_params->num_qps, sizeof(*rqs), GFP_KERNEL);
+	if (!rqs) {
+		err = -ENOMEM;
+		goto err_free_sqs;
+	}
+
+	for (q_id = 0; q_id < qp_params->num_qps; q_id++) {
+		err = hinic3_create_qp(hwdev, &sqs[q_id], &rqs[q_id], q_id,
+				       qp_params->sq_depth, qp_params->rq_depth,
+				       qps_msix_entries[q_id].entry);
+		if (err) {
+			dev_err(hwdev->dev, "Failed to allocate qp %u, err: %d\n",
+				q_id, err);
+			goto err_destroy_qp;
+		}
+	}
+
+	qp_params->sqs = sqs;
+	qp_params->rqs = rqs;
+
+	return 0;
+
+err_destroy_qp:
+	while (q_id > 0) {
+		q_id--;
+		hinic3_destroy_qp(hwdev, &sqs[q_id], &rqs[q_id]);
+	}
+	kfree(rqs);
+err_free_sqs:
+	kfree(sqs);
+err_out:
+	return err;
+}
+
+void hinic3_free_qps(struct hinic3_nic_dev *nic_dev,
+		     struct hinic3_dyna_qp_params *qp_params)
+{
+	struct hinic3_hwdev *hwdev = nic_dev->hwdev;
+	u16 q_id;
+
+	for (q_id = 0; q_id < qp_params->num_qps; q_id++)
+		hinic3_destroy_qp(hwdev, &qp_params->sqs[q_id],
+				  &qp_params->rqs[q_id]);
+
+	kfree(qp_params->sqs);
+	kfree(qp_params->rqs);
+}
+
+void hinic3_init_qps(struct hinic3_nic_dev *nic_dev,
+		     struct hinic3_dyna_qp_params *qp_params)
+{
+	struct hinic3_nic_io *nic_io = nic_dev->nic_io;
+	struct hinic3_io_queue *sqs = qp_params->sqs;
+	struct hinic3_io_queue *rqs = qp_params->rqs;
+	u16 q_id;
+
+	nic_io->num_qps = qp_params->num_qps;
+	nic_io->sq = qp_params->sqs;
+	nic_io->rq = qp_params->rqs;
+	for (q_id = 0; q_id < nic_io->num_qps; q_id++) {
+		sqs[q_id].cons_idx_addr =
+			(u16 *)HINIC3_CI_VADDR(nic_io->ci_vaddr_base, q_id);
+		/* clear ci value */
+		WRITE_ONCE(*sqs[q_id].cons_idx_addr, 0);
+
+		sqs[q_id].db_addr = nic_io->sqs_db_addr;
+		rqs[q_id].db_addr = nic_io->rqs_db_addr;
+	}
+}
+
+void hinic3_uninit_qps(struct hinic3_nic_dev *nic_dev,
+		       struct hinic3_dyna_qp_params *qp_params)
+{
+	struct hinic3_nic_io *nic_io = nic_dev->nic_io;
+
+	qp_params->sqs = nic_io->sq;
+	qp_params->rqs = nic_io->rq;
+	qp_params->num_qps = nic_io->num_qps;
+}
+
+static void hinic3_qp_prepare_cmdq_header(struct hinic3_qp_ctxt_hdr *qp_ctxt_hdr,
+					  enum hinic3_qp_ctxt_type ctxt_type,
+					  u16 num_queues, u16 q_id)
+{
+	qp_ctxt_hdr->queue_type = cpu_to_le16(ctxt_type);
+	qp_ctxt_hdr->num_queues = cpu_to_le16(num_queues);
+	qp_ctxt_hdr->start_qid = cpu_to_le16(q_id);
+	qp_ctxt_hdr->rsvd = 0;
+}
+
+static void hinic3_sq_prepare_ctxt(struct hinic3_io_queue *sq, u16 sq_id,
+				   struct hinic3_sq_ctxt *sq_ctxt)
+{
+	u64 wq_page_addr, wq_page_pfn, wq_block_pfn;
+	u32 wq_block_pfn_hi, wq_block_pfn_lo;
+	u32 wq_page_pfn_hi, wq_page_pfn_lo;
+	u16 pi_start, ci_start;
+
+	ci_start = hinic3_get_sq_local_ci(sq);
+	pi_start = hinic3_get_sq_local_pi(sq);
+
+	wq_page_addr = hinic3_wq_get_first_wqe_page_addr(&sq->wq);
+
+	wq_page_pfn = WQ_PAGE_PFN(wq_page_addr);
+	wq_page_pfn_hi = upper_32_bits(wq_page_pfn);
+	wq_page_pfn_lo = lower_32_bits(wq_page_pfn);
+
+	wq_block_pfn = WQ_BLOCK_PFN(sq->wq.wq_block_paddr);
+	wq_block_pfn_hi = upper_32_bits(wq_block_pfn);
+	wq_block_pfn_lo = lower_32_bits(wq_block_pfn);
+
+	sq_ctxt->ci_pi =
+		cpu_to_le32(SQ_CTXT_CI_PI_SET(ci_start, CI_IDX) |
+			    SQ_CTXT_CI_PI_SET(pi_start, PI_IDX));
+
+	sq_ctxt->drop_mode_sp =
+		cpu_to_le32(SQ_CTXT_MODE_SET(0, SP_FLAG) |
+			    SQ_CTXT_MODE_SET(0, PKT_DROP));
+
+	sq_ctxt->wq_pfn_hi_owner =
+		cpu_to_le32(SQ_CTXT_WQ_PAGE_SET(wq_page_pfn_hi, HI_PFN) |
+			    SQ_CTXT_WQ_PAGE_SET(1, OWNER));
+
+	sq_ctxt->wq_pfn_lo = cpu_to_le32(wq_page_pfn_lo);
+
+	sq_ctxt->pkt_drop_thd =
+		cpu_to_le32(SQ_CTXT_PKT_DROP_THD_SET(HINIC3_DEFAULT_DROP_THD_ON, THD_ON) |
+			    SQ_CTXT_PKT_DROP_THD_SET(HINIC3_DEFAULT_DROP_THD_OFF, THD_OFF));
+
+	sq_ctxt->global_sq_id =
+		cpu_to_le32(SQ_CTXT_GLOBAL_QUEUE_ID_SET((u32)sq_id,
+							GLOBAL_SQ_ID));
+
+	/* enable insert c-vlan by default */
+	sq_ctxt->vlan_ceq_attr =
+		cpu_to_le32(SQ_CTXT_VLAN_CEQ_SET(0, CEQ_EN) |
+			    SQ_CTXT_VLAN_CEQ_SET(1, INSERT_MODE));
+
+	sq_ctxt->rsvd0 = 0;
+
+	sq_ctxt->pref_cache =
+		cpu_to_le32(SQ_CTXT_PREF_SET(SQ_WQ_PREFETCH_MIN, CACHE_MIN) |
+			    SQ_CTXT_PREF_SET(SQ_WQ_PREFETCH_MAX, CACHE_MAX) |
+			    SQ_CTXT_PREF_SET(SQ_WQ_PREFETCH_THRESHOLD, CACHE_THRESHOLD));
+
+	sq_ctxt->pref_ci_owner =
+		cpu_to_le32(SQ_CTXT_PREF_SET(SQ_CTXT_PREF_CI_HI(ci_start), CI_HI) |
+			    SQ_CTXT_PREF_SET(1, OWNER));
+
+	sq_ctxt->pref_wq_pfn_hi_ci =
+		cpu_to_le32(SQ_CTXT_PREF_SET(ci_start, CI_LOW) |
+			    SQ_CTXT_PREF_SET(wq_page_pfn_hi, WQ_PFN_HI));
+
+	sq_ctxt->pref_wq_pfn_lo = cpu_to_le32(wq_page_pfn_lo);
+
+	sq_ctxt->wq_block_pfn_hi =
+		cpu_to_le32(SQ_CTXT_WQ_BLOCK_SET(wq_block_pfn_hi, PFN_HI));
+
+	sq_ctxt->wq_block_pfn_lo = cpu_to_le32(wq_block_pfn_lo);
+}
+
+static void hinic3_rq_prepare_ctxt_get_wq_info(struct hinic3_io_queue *rq,
+					       u32 *wq_page_pfn_hi,
+					       u32 *wq_page_pfn_lo,
+					       u32 *wq_block_pfn_hi,
+					       u32 *wq_block_pfn_lo)
+{
+	u64 wq_page_addr, wq_page_pfn, wq_block_pfn;
+
+	wq_page_addr = hinic3_wq_get_first_wqe_page_addr(&rq->wq);
+
+	wq_page_pfn = WQ_PAGE_PFN(wq_page_addr);
+	*wq_page_pfn_hi = upper_32_bits(wq_page_pfn);
+	*wq_page_pfn_lo = lower_32_bits(wq_page_pfn);
+
+	wq_block_pfn = WQ_BLOCK_PFN(rq->wq.wq_block_paddr);
+	*wq_block_pfn_hi = upper_32_bits(wq_block_pfn);
+	*wq_block_pfn_lo = lower_32_bits(wq_block_pfn);
+}
+
+static void hinic3_rq_prepare_ctxt(struct hinic3_io_queue *rq,
+				   struct hinic3_rq_ctxt *rq_ctxt)
+{
+	u32 wq_block_pfn_hi, wq_block_pfn_lo;
+	u32 wq_page_pfn_hi, wq_page_pfn_lo;
+	u16 pi_start, ci_start;
+
+	ci_start = (rq->wq.cons_idx & rq->wq.idx_mask) << HINIC3_NORMAL_RQ_WQE;
+	pi_start = (rq->wq.prod_idx & rq->wq.idx_mask) << HINIC3_NORMAL_RQ_WQE;
+
+	hinic3_rq_prepare_ctxt_get_wq_info(rq, &wq_page_pfn_hi, &wq_page_pfn_lo,
+					   &wq_block_pfn_hi, &wq_block_pfn_lo);
+
+	rq_ctxt->ci_pi =
+		cpu_to_le32(RQ_CTXT_CI_PI_SET(ci_start, CI_IDX) |
+			    RQ_CTXT_CI_PI_SET(pi_start, PI_IDX));
+
+	rq_ctxt->ceq_attr =
+		cpu_to_le32(RQ_CTXT_CEQ_ATTR_SET(0, EN) |
+			    RQ_CTXT_CEQ_ATTR_SET(rq->msix_entry_idx, INTR));
+
+	rq_ctxt->wq_pfn_hi_type_owner =
+		cpu_to_le32(RQ_CTXT_WQ_PAGE_SET(wq_page_pfn_hi, HI_PFN) |
+			    RQ_CTXT_WQ_PAGE_SET(1, OWNER));
+
+	/* use 16Byte WQE */
+	rq_ctxt->wq_pfn_hi_type_owner |=
+		cpu_to_le32(RQ_CTXT_WQ_PAGE_SET(2, WQE_TYPE));
+	rq_ctxt->cqe_sge_len = cpu_to_le32(RQ_CTXT_CQE_LEN_SET(1, CQE_LEN));
+
+	rq_ctxt->wq_pfn_lo = cpu_to_le32(wq_page_pfn_lo);
+
+	rq_ctxt->pref_cache =
+		cpu_to_le32(RQ_CTXT_PREF_SET(RQ_WQ_PREFETCH_MIN, CACHE_MIN) |
+			    RQ_CTXT_PREF_SET(RQ_WQ_PREFETCH_MAX, CACHE_MAX) |
+			    RQ_CTXT_PREF_SET(RQ_WQ_PREFETCH_THRESHOLD, CACHE_THRESHOLD));
+
+	rq_ctxt->pref_ci_owner =
+		cpu_to_le32(RQ_CTXT_PREF_SET(SQ_CTXT_PREF_CI_HI(ci_start), CI_HI) |
+			    RQ_CTXT_PREF_SET(1, OWNER));
+
+	rq_ctxt->pref_wq_pfn_hi_ci =
+		cpu_to_le32(RQ_CTXT_PREF_SET(wq_page_pfn_hi, WQ_PFN_HI) |
+			    RQ_CTXT_PREF_SET(ci_start, CI_LOW));
+
+	rq_ctxt->pref_wq_pfn_lo = cpu_to_le32(wq_page_pfn_lo);
+
+	rq_ctxt->wq_block_pfn_hi =
+		cpu_to_le32(RQ_CTXT_WQ_BLOCK_SET(wq_block_pfn_hi, PFN_HI));
+
+	rq_ctxt->wq_block_pfn_lo = cpu_to_le32(wq_block_pfn_lo);
+}
+
+static int init_sq_ctxts(struct hinic3_nic_dev *nic_dev)
+{
+	struct hinic3_nic_io *nic_io = nic_dev->nic_io;
+	struct hinic3_hwdev *hwdev = nic_dev->hwdev;
+	struct hinic3_sq_ctxt_block *sq_ctxt_block;
+	u16 q_id, curr_id, max_ctxts, i;
+	struct hinic3_sq_ctxt *sq_ctxt;
+	struct hinic3_cmd_buf *cmd_buf;
+	struct hinic3_io_queue *sq;
+	__le64 out_param;
+	int err = 0;
+
+	cmd_buf = hinic3_alloc_cmd_buf(hwdev);
+	if (!cmd_buf) {
+		dev_err(hwdev->dev, "Failed to allocate cmd buf\n");
+		return -ENOMEM;
+	}
+
+	q_id = 0;
+	while (q_id < nic_io->num_qps) {
+		sq_ctxt_block = cmd_buf->buf;
+		sq_ctxt = sq_ctxt_block->sq_ctxt;
+
+		max_ctxts = (nic_io->num_qps - q_id) > HINIC3_Q_CTXT_MAX ?
+			     HINIC3_Q_CTXT_MAX : (nic_io->num_qps - q_id);
+
+		hinic3_qp_prepare_cmdq_header(&sq_ctxt_block->cmdq_hdr,
+					      HINIC3_QP_CTXT_TYPE_SQ, max_ctxts,
+					      q_id);
+
+		for (i = 0; i < max_ctxts; i++) {
+			curr_id = q_id + i;
+			sq = &nic_io->sq[curr_id];
+			hinic3_sq_prepare_ctxt(sq, curr_id, &sq_ctxt[i]);
+		}
+
+		hinic3_cmdq_buf_swab32(sq_ctxt_block, sizeof(*sq_ctxt_block));
+
+		cmd_buf->size = cpu_to_le16(SQ_CTXT_SIZE(max_ctxts));
+		err = hinic3_cmdq_direct_resp(hwdev, MGMT_MOD_L2NIC,
+					      L2NIC_UCODE_CMD_MODIFY_QUEUE_CTX,
+					      cmd_buf, &out_param);
+		if (err || out_param) {
+			dev_err(hwdev->dev, "Failed to set SQ ctxts, err: %d, out_param: 0x%llx\n",
+				err, out_param);
+			err = -EFAULT;
+			break;
+		}
+
+		q_id += max_ctxts;
+	}
+
+	hinic3_free_cmd_buf(hwdev, cmd_buf);
+
+	return err;
+}
+
+static int init_rq_ctxts(struct hinic3_nic_dev *nic_dev)
+{
+	struct hinic3_nic_io *nic_io = nic_dev->nic_io;
+	struct hinic3_hwdev *hwdev = nic_dev->hwdev;
+	struct hinic3_rq_ctxt_block *rq_ctxt_block;
+	u16 q_id, curr_id, max_ctxts, i;
+	struct hinic3_rq_ctxt *rq_ctxt;
+	struct hinic3_cmd_buf *cmd_buf;
+	struct hinic3_io_queue *rq;
+	__le64 out_param;
+	int err = 0;
+
+	cmd_buf = hinic3_alloc_cmd_buf(hwdev);
+	if (!cmd_buf) {
+		dev_err(hwdev->dev, "Failed to allocate cmd buf\n");
+		return -ENOMEM;
+	}
+
+	q_id = 0;
+	while (q_id < nic_io->num_qps) {
+		rq_ctxt_block = cmd_buf->buf;
+		rq_ctxt = rq_ctxt_block->rq_ctxt;
+
+		max_ctxts = (nic_io->num_qps - q_id) > HINIC3_Q_CTXT_MAX ?
+				HINIC3_Q_CTXT_MAX : (nic_io->num_qps - q_id);
+
+		hinic3_qp_prepare_cmdq_header(&rq_ctxt_block->cmdq_hdr,
+					      HINIC3_QP_CTXT_TYPE_RQ, max_ctxts,
+					      q_id);
+
+		for (i = 0; i < max_ctxts; i++) {
+			curr_id = q_id + i;
+			rq = &nic_io->rq[curr_id];
+			hinic3_rq_prepare_ctxt(rq, &rq_ctxt[i]);
+		}
+
+		hinic3_cmdq_buf_swab32(rq_ctxt_block, sizeof(*rq_ctxt_block));
+
+		cmd_buf->size = cpu_to_le16(RQ_CTXT_SIZE(max_ctxts));
+
+		err = hinic3_cmdq_direct_resp(hwdev, MGMT_MOD_L2NIC,
+					      L2NIC_UCODE_CMD_MODIFY_QUEUE_CTX,
+					      cmd_buf, &out_param);
+		if (err || out_param) {
+			dev_err(hwdev->dev, "Failed to set RQ ctxts, err: %d, out_param: 0x%llx\n",
+				err, out_param);
+			err = -EFAULT;
+			break;
+		}
+
+		q_id += max_ctxts;
+	}
+
+	hinic3_free_cmd_buf(hwdev, cmd_buf);
+
+	return err;
+}
+
+static int init_qp_ctxts(struct hinic3_nic_dev *nic_dev)
+{
+	int err;
+
+	err = init_sq_ctxts(nic_dev);
+	if (err)
+		return err;
+
+	err = init_rq_ctxts(nic_dev);
+	if (err)
+		return err;
+
+	return 0;
+}
+
+static int clean_queue_offload_ctxt(struct hinic3_nic_dev *nic_dev,
+				    enum hinic3_qp_ctxt_type ctxt_type)
+{
+	struct hinic3_nic_io *nic_io = nic_dev->nic_io;
+	struct hinic3_hwdev *hwdev = nic_dev->hwdev;
+	struct hinic3_clean_queue_ctxt *ctxt_block;
+	struct hinic3_cmd_buf *cmd_buf;
+	__le64 out_param;
+	int err;
+
+	cmd_buf = hinic3_alloc_cmd_buf(hwdev);
+	if (!cmd_buf) {
+		dev_err(hwdev->dev, "Failed to allocate cmd buf\n");
+		return -ENOMEM;
+	}
+
+	ctxt_block = cmd_buf->buf;
+	ctxt_block->cmdq_hdr.num_queues = cpu_to_le16(nic_io->max_qps);
+	ctxt_block->cmdq_hdr.queue_type = cpu_to_le16(ctxt_type);
+	ctxt_block->cmdq_hdr.start_qid = 0;
+	ctxt_block->cmdq_hdr.rsvd = 0;
+	ctxt_block->rsvd = 0;
+
+	hinic3_cmdq_buf_swab32(ctxt_block, sizeof(*ctxt_block));
+
+	cmd_buf->size = cpu_to_le16(sizeof(*ctxt_block));
+
+	err = hinic3_cmdq_direct_resp(hwdev, MGMT_MOD_L2NIC,
+				      L2NIC_UCODE_CMD_CLEAN_QUEUE_CTX,
+				      cmd_buf, &out_param);
+	if (err || out_param) {
+		dev_err(hwdev->dev, "Failed to clean queue offload ctxts, err: %d,out_param: 0x%llx\n",
+			err, out_param);
+
+		err = -EFAULT;
+	}
+
+	hinic3_free_cmd_buf(hwdev, cmd_buf);
+
+	return err;
+}
+
+static int clean_qp_offload_ctxt(struct hinic3_nic_dev *nic_dev)
+{
+	/* clean LRO/TSO context space */
+	return clean_queue_offload_ctxt(nic_dev, HINIC3_QP_CTXT_TYPE_SQ) ||
+	       clean_queue_offload_ctxt(nic_dev, HINIC3_QP_CTXT_TYPE_RQ);
+}
+
+/* init qps ctxt and set sq ci attr and arm all sq */
+int hinic3_init_qp_ctxts(struct hinic3_nic_dev *nic_dev)
+{
+	struct hinic3_nic_io *nic_io = nic_dev->nic_io;
+	struct hinic3_hwdev *hwdev = nic_dev->hwdev;
+	struct hinic3_sq_attr sq_attr;
+	u32 rq_depth;
+	u16 q_id;
+	int err;
+
+	err = init_qp_ctxts(nic_dev);
+	if (err) {
+		dev_err(hwdev->dev, "Failed to init QP ctxts\n");
+		return err;
+	}
+
+	/* clean LRO/TSO context space */
+	err = clean_qp_offload_ctxt(nic_dev);
+	if (err) {
+		dev_err(hwdev->dev, "Failed to clean qp offload ctxts\n");
+		return err;
+	}
+
+	rq_depth = nic_io->rq[0].wq.q_depth << HINIC3_NORMAL_RQ_WQE;
+
+	err = hinic3_set_root_ctxt(hwdev, rq_depth, nic_io->sq[0].wq.q_depth,
+				   nic_io->rx_buf_len);
+	if (err) {
+		dev_err(hwdev->dev, "Failed to set root context\n");
+		return err;
+	}
+
+	for (q_id = 0; q_id < nic_io->num_qps; q_id++) {
+		sq_attr.ci_dma_base =
+			HINIC3_CI_PADDR(nic_io->ci_dma_base, q_id) >> 0x2;
+		sq_attr.pending_limit = HINIC3_DEFAULT_TX_CI_PENDING_LIMIT;
+		sq_attr.coalescing_time = HINIC3_DEFAULT_TX_CI_COALESCING_TIME;
+		sq_attr.intr_en = 1;
+		sq_attr.intr_idx = nic_io->sq[q_id].msix_entry_idx;
+		sq_attr.l2nic_sqn = q_id;
+		sq_attr.dma_attr_off = 0;
+		err = hinic3_set_ci_table(hwdev, &sq_attr);
+		if (err) {
+			dev_err(hwdev->dev, "Failed to set ci table\n");
+			goto err_clean_root_ctxt;
+		}
+	}
+
+	return 0;
+
+err_clean_root_ctxt:
+	hinic3_clean_root_ctxt(hwdev);
+
+	return err;
+}
+
+void hinic3_free_qp_ctxts(struct hinic3_nic_dev *nic_dev)
+{
+	hinic3_clean_root_ctxt(nic_dev->hwdev);
+}
diff --git a/drivers/net/ethernet/huawei/hinic3/hinic3_nic_io.h b/drivers/net/ethernet/huawei/hinic3/hinic3_nic_io.h
new file mode 100644
index 000000000000..12eefabcf1db
--- /dev/null
+++ b/drivers/net/ethernet/huawei/hinic3/hinic3_nic_io.h
@@ -0,0 +1,145 @@
+/* SPDX-License-Identifier: GPL-2.0 */
+/* Copyright (c) Huawei Technologies Co., Ltd. 2025. All rights reserved. */
+
+#ifndef _HINIC3_NIC_IO_H_
+#define _HINIC3_NIC_IO_H_
+
+#include <linux/bitfield.h>
+
+#include "hinic3_wq.h"
+
+struct hinic3_nic_dev;
+
+#define HINIC3_SQ_WQEBB_SHIFT      4
+#define HINIC3_RQ_WQEBB_SHIFT      3
+#define HINIC3_SQ_WQEBB_SIZE       BIT(HINIC3_SQ_WQEBB_SHIFT)
+
+/* ******************** RQ_CTRL ******************** */
+enum hinic3_rq_wqe_type {
+	HINIC3_NORMAL_RQ_WQE = 1,
+};
+
+/* ******************** SQ_CTRL ******************** */
+#define HINIC3_TX_MSS_DEFAULT  0x3E00
+#define HINIC3_TX_MSS_MIN      0x50
+#define HINIC3_MAX_SQ_SGE      18
+
+struct hinic3_io_queue {
+	struct hinic3_wq  wq;
+	u8                owner;
+	u16               q_id;
+	u16               msix_entry_idx;
+	u8 __iomem        *db_addr;
+	u16               *cons_idx_addr;
+} ____cacheline_aligned;
+
+static inline u16 hinic3_get_sq_local_ci(const struct hinic3_io_queue *sq)
+{
+	const struct hinic3_wq *wq = &sq->wq;
+
+	return wq->cons_idx & wq->idx_mask;
+}
+
+static inline u16 hinic3_get_sq_local_pi(const struct hinic3_io_queue *sq)
+{
+	const struct hinic3_wq *wq = &sq->wq;
+
+	return wq->prod_idx & wq->idx_mask;
+}
+
+static inline u16 hinic3_get_sq_hw_ci(const struct hinic3_io_queue *sq)
+{
+	const struct hinic3_wq *wq = &sq->wq;
+
+	return READ_ONCE(*sq->cons_idx_addr) & wq->idx_mask;
+}
+
+/* ******************** DB INFO ******************** */
+#define DB_INFO_QID_MASK    GENMASK(12, 0)
+#define DB_INFO_CFLAG_MASK  BIT(23)
+#define DB_INFO_COS_MASK    GENMASK(26, 24)
+#define DB_INFO_TYPE_MASK   GENMASK(31, 27)
+#define DB_INFO_SET(val, member)  \
+	FIELD_PREP(DB_INFO_##member##_MASK, val)
+
+#define DB_PI_LOW_MASK   0xFFU
+#define DB_PI_HIGH_MASK  0xFFU
+#define DB_PI_HI_SHIFT   8
+#define DB_PI_LOW(pi)    ((pi) & DB_PI_LOW_MASK)
+#define DB_PI_HIGH(pi)   (((pi) >> DB_PI_HI_SHIFT) & DB_PI_HIGH_MASK)
+#define DB_ADDR(q, pi)   ((u64 __iomem *)((q)->db_addr) + DB_PI_LOW(pi))
+#define DB_SRC_TYPE      1
+
+/* CFLAG_DATA_PATH */
+#define DB_CFLAG_DP_SQ   0
+#define DB_CFLAG_DP_RQ   1
+
+struct hinic3_nic_db {
+	__le32 db_info;
+	__le32 pi_hi;
+};
+
+static inline void hinic3_write_db(struct hinic3_io_queue *queue, int cos,
+				   u8 cflag, u16 pi)
+{
+	struct hinic3_nic_db db;
+
+	db.db_info =
+		cpu_to_le32(DB_INFO_SET(DB_SRC_TYPE, TYPE) |
+			    DB_INFO_SET(cflag, CFLAG) |
+			    DB_INFO_SET(cos, COS) |
+			    DB_INFO_SET(queue->q_id, QID));
+	db.pi_hi = cpu_to_le32(DB_PI_HIGH(pi));
+
+	writeq(*((u64 *)&db), DB_ADDR(queue, pi));
+}
+
+struct hinic3_dyna_qp_params {
+	u16                    num_qps;
+	u32                    sq_depth;
+	u32                    rq_depth;
+
+	struct hinic3_io_queue *sqs;
+	struct hinic3_io_queue *rqs;
+};
+
+struct hinic3_nic_io {
+	struct hinic3_io_queue *sq;
+	struct hinic3_io_queue *rq;
+
+	u16                    num_qps;
+	u16                    max_qps;
+
+	/* Base address for consumer index of all tx queues. Each queue is
+	 * given a full cache line to hold its consumer index. HW updates
+	 * current consumer index as it consumes tx WQEs.
+	 */
+	void                   *ci_vaddr_base;
+	dma_addr_t             ci_dma_base;
+
+	u8 __iomem             *sqs_db_addr;
+	u8 __iomem             *rqs_db_addr;
+
+	u16                    rx_buf_len;
+	u64                    feature_cap;
+};
+
+int hinic3_init_nic_io(struct hinic3_nic_dev *nic_dev);
+void hinic3_free_nic_io(struct hinic3_nic_dev *nic_dev);
+
+int hinic3_init_nicio_res(struct hinic3_nic_dev *nic_dev);
+void hinic3_free_nicio_res(struct hinic3_nic_dev *nic_dev);
+
+int hinic3_alloc_qps(struct hinic3_nic_dev *nic_dev,
+		     struct hinic3_dyna_qp_params *qp_params);
+void hinic3_free_qps(struct hinic3_nic_dev *nic_dev,
+		     struct hinic3_dyna_qp_params *qp_params);
+void hinic3_init_qps(struct hinic3_nic_dev *nic_dev,
+		     struct hinic3_dyna_qp_params *qp_params);
+void hinic3_uninit_qps(struct hinic3_nic_dev *nic_dev,
+		       struct hinic3_dyna_qp_params *qp_params);
+
+int hinic3_init_qp_ctxts(struct hinic3_nic_dev *nic_dev);
+void hinic3_free_qp_ctxts(struct hinic3_nic_dev *nic_dev);
+
+#endif
diff --git a/drivers/net/ethernet/huawei/hinic3/hinic3_pci_id_tbl.h b/drivers/net/ethernet/huawei/hinic3/hinic3_pci_id_tbl.h
new file mode 100644
index 000000000000..86c88d0bb4bd
--- /dev/null
+++ b/drivers/net/ethernet/huawei/hinic3/hinic3_pci_id_tbl.h
@@ -0,0 +1,9 @@
+/* SPDX-License-Identifier: GPL-2.0 */
+/* Copyright (c) Huawei Technologies Co., Ltd. 2025. All rights reserved. */
+
+#ifndef _HINIC3_PCI_ID_TBL_H_
+#define _HINIC3_PCI_ID_TBL_H_
+
+#define PCI_DEV_ID_HINIC3_VF    0x375F
+
+#endif
diff --git a/drivers/net/ethernet/huawei/hinic3/hinic3_queue_common.c b/drivers/net/ethernet/huawei/hinic3/hinic3_queue_common.c
new file mode 100644
index 000000000000..fab9011de9ad
--- /dev/null
+++ b/drivers/net/ethernet/huawei/hinic3/hinic3_queue_common.c
@@ -0,0 +1,68 @@
+// SPDX-License-Identifier: GPL-2.0
+// Copyright (c) Huawei Technologies Co., Ltd. 2025. All rights reserved.
+
+#include <linux/device.h>
+
+#include "hinic3_hwdev.h"
+#include "hinic3_queue_common.h"
+
+void hinic3_queue_pages_init(struct hinic3_queue_pages *qpages, u32 q_depth,
+			     u32 page_size, u32 elem_size)
+{
+	u32 elem_per_page;
+
+	elem_per_page = min(page_size / elem_size, q_depth);
+
+	qpages->pages = NULL;
+	qpages->page_size = page_size;
+	qpages->num_pages = max(q_depth / elem_per_page, 1);
+	qpages->elem_size_shift = ilog2(elem_size);
+	qpages->elem_per_pg_shift = ilog2(elem_per_page);
+}
+
+static void __queue_pages_free(struct hinic3_hwdev *hwdev,
+			       struct hinic3_queue_pages *qpages, u32 pg_cnt)
+{
+	while (pg_cnt > 0) {
+		pg_cnt--;
+		hinic3_dma_free_coherent_align(hwdev->dev,
+					       qpages->pages + pg_cnt);
+	}
+	kfree(qpages->pages);
+	qpages->pages = NULL;
+}
+
+void hinic3_queue_pages_free(struct hinic3_hwdev *hwdev,
+			     struct hinic3_queue_pages *qpages)
+{
+	__queue_pages_free(hwdev, qpages, qpages->num_pages);
+}
+
+int hinic3_queue_pages_alloc(struct hinic3_hwdev *hwdev,
+			     struct hinic3_queue_pages *qpages, u32 align)
+{
+	u32 pg_idx;
+	int err;
+
+	qpages->pages = kcalloc(qpages->num_pages, sizeof(qpages->pages[0]),
+				GFP_KERNEL);
+	if (!qpages->pages)
+		return -ENOMEM;
+
+	if (align == 0)
+		align = qpages->page_size;
+
+	for (pg_idx = 0; pg_idx < qpages->num_pages; pg_idx++) {
+		err = hinic3_dma_zalloc_coherent_align(hwdev->dev,
+						       qpages->page_size,
+						       align,
+						       GFP_KERNEL,
+						       qpages->pages + pg_idx);
+		if (err) {
+			__queue_pages_free(hwdev, qpages, pg_idx);
+			return err;
+		}
+	}
+
+	return 0;
+}
diff --git a/drivers/net/ethernet/huawei/hinic3/hinic3_queue_common.h b/drivers/net/ethernet/huawei/hinic3/hinic3_queue_common.h
new file mode 100644
index 000000000000..ec4cae0a0929
--- /dev/null
+++ b/drivers/net/ethernet/huawei/hinic3/hinic3_queue_common.h
@@ -0,0 +1,54 @@
+/* SPDX-License-Identifier: GPL-2.0 */
+/* Copyright (c) Huawei Technologies Co., Ltd. 2025. All rights reserved. */
+
+#ifndef _HINIC3_QUEUE_COMMON_H_
+#define _HINIC3_QUEUE_COMMON_H_
+
+#include <linux/types.h>
+
+#include "hinic3_common.h"
+
+struct hinic3_hwdev;
+
+struct hinic3_queue_pages {
+	/* Array of DMA-able pages that actually holds the queue entries. */
+	struct hinic3_dma_addr_align  *pages;
+	/* Page size in bytes. */
+	u32                           page_size;
+	/* Number of pages, must be power of 2. */
+	u16                           num_pages;
+	u8                            elem_size_shift;
+	u8                            elem_per_pg_shift;
+};
+
+void hinic3_queue_pages_init(struct hinic3_queue_pages *qpages, u32 q_depth,
+			     u32 page_size, u32 elem_size);
+int hinic3_queue_pages_alloc(struct hinic3_hwdev *hwdev,
+			     struct hinic3_queue_pages *qpages, u32 align);
+void hinic3_queue_pages_free(struct hinic3_hwdev *hwdev,
+			     struct hinic3_queue_pages *qpages);
+
+/* Get pointer to queue entry at the specified index. Index does not have to be
+ * masked to queue depth, only least significant bits will be used. Also
+ * provides remaining elements in same page (including the first one) in case
+ * caller needs multiple entries.
+ */
+static inline void *get_q_element(const struct hinic3_queue_pages *qpages,
+				  u32 idx, u32 *remaining_in_page)
+{
+	const struct hinic3_dma_addr_align *page;
+	u32 page_idx, elem_idx, elem_per_pg, ofs;
+	u8 shift;
+
+	shift = qpages->elem_per_pg_shift;
+	page_idx = (idx >> shift) & (qpages->num_pages - 1);
+	elem_per_pg = 1 << shift;
+	elem_idx = idx & (elem_per_pg - 1);
+	if (remaining_in_page)
+		*remaining_in_page = elem_per_pg - elem_idx;
+	ofs = elem_idx << qpages->elem_size_shift;
+	page = qpages->pages + page_idx;
+	return (char *)page->align_vaddr + ofs;
+}
+
+#endif
diff --git a/drivers/net/ethernet/huawei/hinic3/hinic3_rss.c b/drivers/net/ethernet/huawei/hinic3/hinic3_rss.c
new file mode 100644
index 000000000000..4ff1b2f79838
--- /dev/null
+++ b/drivers/net/ethernet/huawei/hinic3/hinic3_rss.c
@@ -0,0 +1,336 @@
+// SPDX-License-Identifier: GPL-2.0
+// Copyright (c) Huawei Technologies Co., Ltd. 2025. All rights reserved.
+
+#include <linux/ethtool.h>
+
+#include "hinic3_cmdq.h"
+#include "hinic3_hwdev.h"
+#include "hinic3_hwif.h"
+#include "hinic3_mbox.h"
+#include "hinic3_nic_cfg.h"
+#include "hinic3_nic_dev.h"
+#include "hinic3_rss.h"
+
+static void hinic3_fillout_indir_tbl(struct net_device *netdev, u16 *indir)
+{
+	struct hinic3_nic_dev *nic_dev = netdev_priv(netdev);
+	u16 i, num_qps;
+
+	num_qps = nic_dev->q_params.num_qps;
+	for (i = 0; i < L2NIC_RSS_INDIR_SIZE; i++)
+		indir[i] = ethtool_rxfh_indir_default(i, num_qps);
+}
+
+static int hinic3_rss_cfg(struct hinic3_hwdev *hwdev, u8 rss_en, u16 num_qps)
+{
+	struct mgmt_msg_params msg_params = {};
+	struct l2nic_cmd_cfg_rss rss_cfg = {};
+	int err;
+
+	rss_cfg.func_id = hinic3_global_func_id(hwdev);
+	rss_cfg.rss_en = rss_en;
+	rss_cfg.rq_priority_number = 0;
+	rss_cfg.num_qps = num_qps;
+
+	mgmt_msg_params_init_default(&msg_params, &rss_cfg, sizeof(rss_cfg));
+
+	err = hinic3_send_mbox_to_mgmt(hwdev, MGMT_MOD_L2NIC,
+				       L2NIC_CMD_CFG_RSS, &msg_params);
+	if (err || rss_cfg.msg_head.status) {
+		dev_err(hwdev->dev, "Failed to set rss cfg, err: %d, status: 0x%x\n",
+			err, rss_cfg.msg_head.status);
+		return -EINVAL;
+	}
+
+	return 0;
+}
+
+static void hinic3_init_rss_parameters(struct net_device *netdev)
+{
+	struct hinic3_nic_dev *nic_dev = netdev_priv(netdev);
+
+	nic_dev->rss_hash_type = HINIC3_RSS_HASH_ENGINE_TYPE_XOR;
+	nic_dev->rss_type.tcp_ipv6_ext = 1;
+	nic_dev->rss_type.ipv6_ext = 1;
+	nic_dev->rss_type.tcp_ipv6 = 1;
+	nic_dev->rss_type.ipv6 = 1;
+	nic_dev->rss_type.tcp_ipv4 = 1;
+	nic_dev->rss_type.ipv4 = 1;
+	nic_dev->rss_type.udp_ipv6 = 1;
+	nic_dev->rss_type.udp_ipv4 = 1;
+}
+
+static void decide_num_qps(struct net_device *netdev)
+{
+	struct hinic3_nic_dev *nic_dev = netdev_priv(netdev);
+	unsigned int dev_cpus;
+
+	dev_cpus = netif_get_num_default_rss_queues();
+	nic_dev->q_params.num_qps = min(dev_cpus, nic_dev->max_qps);
+}
+
+static int alloc_rss_resource(struct net_device *netdev)
+{
+	struct hinic3_nic_dev *nic_dev = netdev_priv(netdev);
+
+	nic_dev->rss_hkey = kmalloc_array(L2NIC_RSS_KEY_SIZE,
+					  sizeof(nic_dev->rss_hkey[0]),
+					  GFP_KERNEL);
+	if (!nic_dev->rss_hkey)
+		return -ENOMEM;
+
+	netdev_rss_key_fill(nic_dev->rss_hkey, L2NIC_RSS_KEY_SIZE);
+
+	nic_dev->rss_indir = kcalloc(L2NIC_RSS_INDIR_SIZE, sizeof(u16),
+				     GFP_KERNEL);
+	if (!nic_dev->rss_indir) {
+		kfree(nic_dev->rss_hkey);
+		nic_dev->rss_hkey = NULL;
+		return -ENOMEM;
+	}
+
+	return 0;
+}
+
+static int hinic3_rss_set_indir_tbl(struct hinic3_hwdev *hwdev,
+				    const u16 *indir_table)
+{
+	struct l2nic_cmd_rss_set_indir_tbl *indir_tbl;
+	struct hinic3_cmd_buf *cmd_buf;
+	__le64 out_param;
+	int err;
+	u32 i;
+
+	cmd_buf = hinic3_alloc_cmd_buf(hwdev);
+	if (!cmd_buf) {
+		dev_err(hwdev->dev, "Failed to allocate cmd buf\n");
+		return -ENOMEM;
+	}
+
+	cmd_buf->size = cpu_to_le16(sizeof(struct l2nic_cmd_rss_set_indir_tbl));
+	indir_tbl = cmd_buf->buf;
+	memset(indir_tbl, 0, sizeof(*indir_tbl));
+
+	for (i = 0; i < L2NIC_RSS_INDIR_SIZE; i++)
+		indir_tbl->entry[i] = cpu_to_le16(indir_table[i]);
+
+	hinic3_cmdq_buf_swab32(indir_tbl, sizeof(*indir_tbl));
+
+	err = hinic3_cmdq_direct_resp(hwdev, MGMT_MOD_L2NIC,
+				      L2NIC_UCODE_CMD_SET_RSS_INDIR_TBL,
+				      cmd_buf, &out_param);
+	if (err || out_param) {
+		dev_err(hwdev->dev, "Failed to set rss indir table\n");
+		err = -EFAULT;
+	}
+
+	hinic3_free_cmd_buf(hwdev, cmd_buf);
+
+	return err;
+}
+
+static int hinic3_set_rss_type(struct hinic3_hwdev *hwdev,
+			       struct hinic3_rss_type rss_type)
+{
+	struct l2nic_cmd_set_rss_ctx_tbl ctx_tbl = {};
+	struct mgmt_msg_params msg_params = {};
+	u32 ctx;
+	int err;
+
+	ctx_tbl.func_id = hinic3_global_func_id(hwdev);
+	ctx = L2NIC_RSS_TYPE_SET(1, VALID) |
+	      L2NIC_RSS_TYPE_SET(rss_type.ipv4, IPV4) |
+	      L2NIC_RSS_TYPE_SET(rss_type.ipv6, IPV6) |
+	      L2NIC_RSS_TYPE_SET(rss_type.ipv6_ext, IPV6_EXT) |
+	      L2NIC_RSS_TYPE_SET(rss_type.tcp_ipv4, TCP_IPV4) |
+	      L2NIC_RSS_TYPE_SET(rss_type.tcp_ipv6, TCP_IPV6) |
+	      L2NIC_RSS_TYPE_SET(rss_type.tcp_ipv6_ext, TCP_IPV6_EXT) |
+	      L2NIC_RSS_TYPE_SET(rss_type.udp_ipv4, UDP_IPV4) |
+	      L2NIC_RSS_TYPE_SET(rss_type.udp_ipv6, UDP_IPV6);
+	ctx_tbl.context = ctx;
+
+	mgmt_msg_params_init_default(&msg_params, &ctx_tbl, sizeof(ctx_tbl));
+
+	err = hinic3_send_mbox_to_mgmt(hwdev, MGMT_MOD_L2NIC,
+				       L2NIC_CMD_SET_RSS_CTX_TBL, &msg_params);
+
+	if (ctx_tbl.msg_head.status == MGMT_STATUS_CMD_UNSUPPORTED) {
+		return MGMT_STATUS_CMD_UNSUPPORTED;
+	} else if (err || ctx_tbl.msg_head.status) {
+		dev_err(hwdev->dev, "mgmt Failed to set rss context offload, err: %d, status: 0x%x\n",
+			err, ctx_tbl.msg_head.status);
+		return -EINVAL;
+	}
+
+	return 0;
+}
+
+static int hinic3_rss_cfg_hash_type(struct hinic3_hwdev *hwdev, u8 opcode,
+				    enum hinic3_rss_hash_type *type)
+{
+	struct l2nic_cmd_cfg_rss_engine hash_type_cmd = {};
+	struct mgmt_msg_params msg_params = {};
+	int err;
+
+	hash_type_cmd.func_id = hinic3_global_func_id(hwdev);
+	hash_type_cmd.opcode = opcode;
+
+	if (opcode == MGMT_MSG_CMD_OP_SET)
+		hash_type_cmd.hash_engine = *type;
+
+	mgmt_msg_params_init_default(&msg_params, &hash_type_cmd,
+				     sizeof(hash_type_cmd));
+
+	err = hinic3_send_mbox_to_mgmt(hwdev, MGMT_MOD_L2NIC,
+				       L2NIC_CMD_CFG_RSS_HASH_ENGINE,
+				       &msg_params);
+	if (err || hash_type_cmd.msg_head.status) {
+		dev_err(hwdev->dev, "Failed to %s hash engine, err: %d, status: 0x%x\n",
+			opcode == MGMT_MSG_CMD_OP_SET ? "set" : "get",
+			err, hash_type_cmd.msg_head.status);
+		return -EIO;
+	}
+
+	if (opcode == MGMT_MSG_CMD_OP_GET)
+		*type = hash_type_cmd.hash_engine;
+
+	return 0;
+}
+
+static int hinic3_rss_set_hash_type(struct hinic3_hwdev *hwdev,
+				    enum hinic3_rss_hash_type type)
+{
+	return hinic3_rss_cfg_hash_type(hwdev, MGMT_MSG_CMD_OP_SET, &type);
+}
+
+static int hinic3_config_rss_hw_resource(struct net_device *netdev,
+					 u16 *indir_tbl)
+{
+	struct hinic3_nic_dev *nic_dev = netdev_priv(netdev);
+	int err;
+
+	err = hinic3_rss_set_indir_tbl(nic_dev->hwdev, indir_tbl);
+	if (err)
+		return err;
+
+	err = hinic3_set_rss_type(nic_dev->hwdev, nic_dev->rss_type);
+	if (err)
+		return err;
+
+	return hinic3_rss_set_hash_type(nic_dev->hwdev, nic_dev->rss_hash_type);
+}
+
+static int hinic3_rss_cfg_hash_key(struct hinic3_hwdev *hwdev, u8 opcode,
+				   u8 *key)
+{
+	struct l2nic_cmd_cfg_rss_hash_key hash_key = {};
+	struct mgmt_msg_params msg_params = {};
+	int err;
+
+	hash_key.func_id = hinic3_global_func_id(hwdev);
+	hash_key.opcode = opcode;
+
+	if (opcode == MGMT_MSG_CMD_OP_SET)
+		memcpy(hash_key.key, key, L2NIC_RSS_KEY_SIZE);
+
+	mgmt_msg_params_init_default(&msg_params, &hash_key, sizeof(hash_key));
+
+	err = hinic3_send_mbox_to_mgmt(hwdev, MGMT_MOD_L2NIC,
+				       L2NIC_CMD_CFG_RSS_HASH_KEY, &msg_params);
+	if (err || hash_key.msg_head.status) {
+		dev_err(hwdev->dev, "Failed to %s hash key, err: %d, status: 0x%x\n",
+			opcode == MGMT_MSG_CMD_OP_SET ? "set" : "get",
+			err, hash_key.msg_head.status);
+		return -EINVAL;
+	}
+
+	if (opcode == MGMT_MSG_CMD_OP_GET)
+		memcpy(key, hash_key.key, L2NIC_RSS_KEY_SIZE);
+
+	return 0;
+}
+
+static int hinic3_rss_set_hash_key(struct hinic3_hwdev *hwdev, u8 *key)
+{
+	return hinic3_rss_cfg_hash_key(hwdev, MGMT_MSG_CMD_OP_SET, key);
+}
+
+static int hinic3_set_hw_rss_parameters(struct net_device *netdev, u8 rss_en)
+{
+	struct hinic3_nic_dev *nic_dev = netdev_priv(netdev);
+	int err;
+
+	err = hinic3_rss_set_hash_key(nic_dev->hwdev, nic_dev->rss_hkey);
+	if (err)
+		return err;
+
+	hinic3_fillout_indir_tbl(netdev, nic_dev->rss_indir);
+
+	err = hinic3_config_rss_hw_resource(netdev, nic_dev->rss_indir);
+	if (err)
+		return err;
+
+	err = hinic3_rss_cfg(nic_dev->hwdev, rss_en, nic_dev->q_params.num_qps);
+	if (err)
+		return err;
+
+	return 0;
+}
+
+int hinic3_rss_init(struct net_device *netdev)
+{
+	return hinic3_set_hw_rss_parameters(netdev, 1);
+}
+
+void hinic3_rss_uninit(struct net_device *netdev)
+{
+	struct hinic3_nic_dev *nic_dev = netdev_priv(netdev);
+
+	hinic3_rss_cfg(nic_dev->hwdev, 0, 0);
+}
+
+void hinic3_clear_rss_config(struct net_device *netdev)
+{
+	struct hinic3_nic_dev *nic_dev = netdev_priv(netdev);
+
+	kfree(nic_dev->rss_hkey);
+	nic_dev->rss_hkey = NULL;
+
+	kfree(nic_dev->rss_indir);
+	nic_dev->rss_indir = NULL;
+}
+
+void hinic3_try_to_enable_rss(struct net_device *netdev)
+{
+	struct hinic3_nic_dev *nic_dev = netdev_priv(netdev);
+	struct hinic3_hwdev *hwdev = nic_dev->hwdev;
+	int err;
+
+	nic_dev->max_qps = hinic3_func_max_qnum(hwdev);
+	if (nic_dev->max_qps <= 1 ||
+	    !hinic3_test_support(nic_dev, HINIC3_NIC_F_RSS))
+		goto err_reset_q_params;
+
+	err = alloc_rss_resource(netdev);
+	if (err) {
+		nic_dev->max_qps = 1;
+		goto err_reset_q_params;
+	}
+
+	set_bit(HINIC3_RSS_ENABLE, &nic_dev->flags);
+	decide_num_qps(netdev);
+	hinic3_init_rss_parameters(netdev);
+	err = hinic3_set_hw_rss_parameters(netdev, 0);
+	if (err) {
+		dev_err(hwdev->dev, "Failed to set hardware rss parameters\n");
+		hinic3_clear_rss_config(netdev);
+		nic_dev->max_qps = 1;
+		goto err_reset_q_params;
+	}
+
+	return;
+
+err_reset_q_params:
+	clear_bit(HINIC3_RSS_ENABLE, &nic_dev->flags);
+	nic_dev->q_params.num_qps = nic_dev->max_qps;
+}
diff --git a/drivers/net/ethernet/huawei/hinic3/hinic3_rss.h b/drivers/net/ethernet/huawei/hinic3/hinic3_rss.h
new file mode 100644
index 000000000000..78d82c2aca06
--- /dev/null
+++ b/drivers/net/ethernet/huawei/hinic3/hinic3_rss.h
@@ -0,0 +1,14 @@
+/* SPDX-License-Identifier: GPL-2.0 */
+/* Copyright (c) Huawei Technologies Co., Ltd. 2025. All rights reserved. */
+
+#ifndef _HINIC3_RSS_H_
+#define _HINIC3_RSS_H_
+
+#include <linux/netdevice.h>
+
+int hinic3_rss_init(struct net_device *netdev);
+void hinic3_rss_uninit(struct net_device *netdev);
+void hinic3_try_to_enable_rss(struct net_device *netdev);
+void hinic3_clear_rss_config(struct net_device *netdev);
+
+#endif
diff --git a/drivers/net/ethernet/huawei/hinic3/hinic3_rx.c b/drivers/net/ethernet/huawei/hinic3/hinic3_rx.c
new file mode 100644
index 000000000000..16c00c3bb1ed
--- /dev/null
+++ b/drivers/net/ethernet/huawei/hinic3/hinic3_rx.c
@@ -0,0 +1,551 @@
+// SPDX-License-Identifier: GPL-2.0
+// Copyright (c) Huawei Technologies Co., Ltd. 2025. All rights reserved.
+
+#include <linux/etherdevice.h>
+#include <linux/if_vlan.h>
+#include <linux/netdevice.h>
+#include <net/gro.h>
+#include <net/page_pool/helpers.h>
+
+#include "hinic3_hwdev.h"
+#include "hinic3_nic_dev.h"
+#include "hinic3_nic_io.h"
+#include "hinic3_rx.h"
+
+#define HINIC3_RX_HDR_SIZE              256
+#define HINIC3_RX_BUFFER_WRITE          16
+
+#define HINIC3_RX_TCP_PKT               0x3
+#define HINIC3_RX_UDP_PKT               0x4
+#define HINIC3_RX_SCTP_PKT              0x7
+
+#define HINIC3_RX_IPV4_PKT              0
+#define HINIC3_RX_IPV6_PKT              1
+#define HINIC3_RX_INVALID_IP_TYPE       2
+
+#define HINIC3_RX_PKT_FORMAT_NON_TUNNEL 0
+#define HINIC3_RX_PKT_FORMAT_VXLAN      1
+
+#define HINIC3_LRO_PKT_HDR_LEN_IPV4     66
+#define HINIC3_LRO_PKT_HDR_LEN_IPV6     86
+#define HINIC3_LRO_PKT_HDR_LEN(cqe) \
+	(RQ_CQE_OFFOLAD_TYPE_GET((cqe)->offload_type, IP_TYPE) == \
+	 HINIC3_RX_IPV6_PKT ? HINIC3_LRO_PKT_HDR_LEN_IPV6 : \
+	 HINIC3_LRO_PKT_HDR_LEN_IPV4)
+
+int hinic3_alloc_rxqs(struct net_device *netdev)
+{
+	struct hinic3_nic_dev *nic_dev = netdev_priv(netdev);
+	struct pci_dev *pdev = nic_dev->pdev;
+	u16 num_rxqs = nic_dev->max_qps;
+	struct hinic3_rxq *rxq;
+	u16 q_id;
+
+	nic_dev->rxqs = kcalloc(num_rxqs, sizeof(*nic_dev->rxqs), GFP_KERNEL);
+	if (!nic_dev->rxqs)
+		return -ENOMEM;
+
+	for (q_id = 0; q_id < num_rxqs; q_id++) {
+		rxq = &nic_dev->rxqs[q_id];
+		rxq->netdev = netdev;
+		rxq->dev = &pdev->dev;
+		rxq->q_id = q_id;
+		rxq->buf_len = nic_dev->rx_buf_len;
+		rxq->buf_len_shift = ilog2(nic_dev->rx_buf_len);
+		rxq->q_depth = nic_dev->q_params.rq_depth;
+		rxq->q_mask = nic_dev->q_params.rq_depth - 1;
+	}
+
+	return 0;
+}
+
+void hinic3_free_rxqs(struct net_device *netdev)
+{
+	struct hinic3_nic_dev *nic_dev = netdev_priv(netdev);
+
+	kfree(nic_dev->rxqs);
+}
+
+static int rx_alloc_mapped_page(struct page_pool *page_pool,
+				struct hinic3_rx_info *rx_info, u16 buf_len)
+{
+	struct page *page;
+	u32 page_offset;
+
+	if (likely(rx_info->page))
+		return 0;
+
+	page = page_pool_dev_alloc_frag(page_pool, &page_offset, buf_len);
+	if (unlikely(!page))
+		return -ENOMEM;
+
+	rx_info->page = page;
+	rx_info->page_offset = page_offset;
+
+	return 0;
+}
+
+/* Associate fixed completion element to every wqe in the rq. Every rq wqe will
+ * always post completion to the same place.
+ */
+static void rq_associate_cqes(struct hinic3_rxq *rxq)
+{
+	struct hinic3_queue_pages *qpages;
+	struct hinic3_rq_wqe *rq_wqe;
+	dma_addr_t cqe_dma;
+	u32 i;
+
+	qpages = &rxq->rq->wq.qpages;
+
+	for (i = 0; i < rxq->q_depth; i++) {
+		rq_wqe = get_q_element(qpages, i, NULL);
+		cqe_dma = rxq->cqe_start_paddr +
+			  i * sizeof(struct hinic3_rq_cqe);
+		rq_wqe->cqe_hi_addr = cpu_to_le32(upper_32_bits(cqe_dma));
+		rq_wqe->cqe_lo_addr = cpu_to_le32(lower_32_bits(cqe_dma));
+	}
+}
+
+static void rq_wqe_buf_set(struct hinic3_io_queue *rq, uint32_t wqe_idx,
+			   dma_addr_t dma_addr, u16 len)
+{
+	struct hinic3_rq_wqe *rq_wqe;
+
+	rq_wqe = get_q_element(&rq->wq.qpages, wqe_idx, NULL);
+	rq_wqe->buf_hi_addr = cpu_to_le32(upper_32_bits(dma_addr));
+	rq_wqe->buf_lo_addr = cpu_to_le32(lower_32_bits(dma_addr));
+}
+
+static u32 hinic3_rx_fill_buffers(struct hinic3_rxq *rxq)
+{
+	u32 i, free_wqebbs = rxq->delta - 1;
+	struct hinic3_rx_info *rx_info;
+	dma_addr_t dma_addr;
+	int err;
+
+	for (i = 0; i < free_wqebbs; i++) {
+		rx_info = &rxq->rx_info[rxq->next_to_update];
+
+		err = rx_alloc_mapped_page(rxq->page_pool, rx_info,
+					   rxq->buf_len);
+		if (unlikely(err))
+			break;
+
+		dma_addr = page_pool_get_dma_addr(rx_info->page) +
+			rx_info->page_offset;
+		rq_wqe_buf_set(rxq->rq, rxq->next_to_update, dma_addr,
+			       rxq->buf_len);
+		rxq->next_to_update = (rxq->next_to_update + 1) & rxq->q_mask;
+	}
+
+	if (likely(i)) {
+		hinic3_write_db(rxq->rq, rxq->q_id & 3, DB_CFLAG_DP_RQ,
+				rxq->next_to_update << HINIC3_NORMAL_RQ_WQE);
+		rxq->delta -= i;
+		rxq->next_to_alloc = rxq->next_to_update;
+	}
+
+	return i;
+}
+
+static u32 hinic3_alloc_rx_buffers(struct hinic3_dyna_rxq_res *rqres,
+				   u32 rq_depth, u16 buf_len)
+{
+	u32 free_wqebbs = rq_depth - 1;
+	u32 idx;
+	int err;
+
+	for (idx = 0; idx < free_wqebbs; idx++) {
+		err = rx_alloc_mapped_page(rqres->page_pool,
+					   &rqres->rx_info[idx], buf_len);
+		if (err)
+			break;
+	}
+
+	return idx;
+}
+
+static void hinic3_free_rx_buffers(struct hinic3_dyna_rxq_res *rqres,
+				   u32 q_depth)
+{
+	struct hinic3_rx_info *rx_info;
+	u32 i;
+
+	/* Free all the Rx ring sk_buffs */
+	for (i = 0; i < q_depth; i++) {
+		rx_info = &rqres->rx_info[i];
+
+		if (rx_info->page) {
+			page_pool_put_full_page(rqres->page_pool,
+						rx_info->page, false);
+			rx_info->page = NULL;
+		}
+	}
+}
+
+static void hinic3_add_rx_frag(struct hinic3_rxq *rxq,
+			       struct hinic3_rx_info *rx_info,
+			       struct sk_buff *skb, u32 size)
+{
+	struct page *page;
+	u8 *va;
+
+	page = rx_info->page;
+	va = (u8 *)page_address(page) + rx_info->page_offset;
+	net_prefetch(va);
+
+	page_pool_dma_sync_for_cpu(rxq->page_pool, page, rx_info->page_offset,
+				   rxq->buf_len);
+
+	if (size <= HINIC3_RX_HDR_SIZE && !skb_is_nonlinear(skb)) {
+		memcpy(__skb_put(skb, size), va,
+		       ALIGN(size, sizeof(long)));
+		page_pool_put_full_page(rxq->page_pool, page, false);
+
+		return;
+	}
+
+	skb_add_rx_frag(skb, skb_shinfo(skb)->nr_frags, page,
+			rx_info->page_offset, size, rxq->buf_len);
+	skb_mark_for_recycle(skb);
+}
+
+static void packaging_skb(struct hinic3_rxq *rxq, struct sk_buff *skb,
+			  u32 sge_num, u32 pkt_len)
+{
+	struct hinic3_rx_info *rx_info;
+	u32 temp_pkt_len = pkt_len;
+	u32 temp_sge_num = sge_num;
+	u32 sw_ci;
+	u32 size;
+
+	sw_ci = rxq->cons_idx & rxq->q_mask;
+	while (temp_sge_num) {
+		rx_info = &rxq->rx_info[sw_ci];
+		sw_ci = (sw_ci + 1) & rxq->q_mask;
+		if (unlikely(temp_pkt_len > rxq->buf_len)) {
+			size = rxq->buf_len;
+			temp_pkt_len -= rxq->buf_len;
+		} else {
+			size = temp_pkt_len;
+		}
+
+		hinic3_add_rx_frag(rxq, rx_info, skb, size);
+
+		/* clear contents of buffer_info */
+		rx_info->page = NULL;
+		temp_sge_num--;
+	}
+}
+
+static u32 hinic3_get_sge_num(struct hinic3_rxq *rxq, u32 pkt_len)
+{
+	u32 sge_num;
+
+	sge_num = pkt_len >> rxq->buf_len_shift;
+	sge_num += (pkt_len & (rxq->buf_len - 1)) ? 1 : 0;
+
+	return sge_num;
+}
+
+static struct sk_buff *hinic3_fetch_rx_buffer(struct hinic3_rxq *rxq,
+					      u32 pkt_len)
+{
+	struct sk_buff *skb;
+	u32 sge_num;
+
+	skb = napi_alloc_skb(&rxq->irq_cfg->napi, HINIC3_RX_HDR_SIZE);
+	if (unlikely(!skb))
+		return NULL;
+
+	sge_num = hinic3_get_sge_num(rxq, pkt_len);
+
+	net_prefetchw(skb->data);
+	packaging_skb(rxq, skb, sge_num, pkt_len);
+
+	rxq->cons_idx += sge_num;
+	rxq->delta += sge_num;
+
+	return skb;
+}
+
+static void hinic3_pull_tail(struct sk_buff *skb)
+{
+	skb_frag_t *frag = &skb_shinfo(skb)->frags[0];
+	unsigned int pull_len;
+	unsigned char *va;
+
+	va = skb_frag_address(frag);
+
+	/* we need the header to contain the greater of either ETH_HLEN or
+	 * 60 bytes if the skb->len is less than 60 for skb_pad.
+	 */
+	pull_len = eth_get_headlen(skb->dev, va, HINIC3_RX_HDR_SIZE);
+
+	/* align pull length to size of long to optimize memcpy performance */
+	skb_copy_to_linear_data(skb, va, ALIGN(pull_len, sizeof(long)));
+
+	/* update all of the pointers */
+	skb_frag_size_sub(frag, pull_len);
+	skb_frag_off_add(frag, pull_len);
+
+	skb->data_len -= pull_len;
+	skb->tail += pull_len;
+}
+
+static void hinic3_rx_csum(struct hinic3_rxq *rxq, u32 offload_type,
+			   u32 status, struct sk_buff *skb)
+{
+	u32 pkt_fmt = RQ_CQE_OFFOLAD_TYPE_GET(offload_type, TUNNEL_PKT_FORMAT);
+	u32 pkt_type = RQ_CQE_OFFOLAD_TYPE_GET(offload_type, PKT_TYPE);
+	u32 ip_type = RQ_CQE_OFFOLAD_TYPE_GET(offload_type, IP_TYPE);
+	u32 csum_err = RQ_CQE_STATUS_GET(status, CSUM_ERR);
+	struct net_device *netdev = rxq->netdev;
+
+	if (!(netdev->features & NETIF_F_RXCSUM))
+		return;
+
+	if (unlikely(csum_err)) {
+		/* pkt type is recognized by HW, and csum is wrong */
+		skb->ip_summed = CHECKSUM_NONE;
+		return;
+	}
+
+	if (ip_type == HINIC3_RX_INVALID_IP_TYPE ||
+	    !(pkt_fmt == HINIC3_RX_PKT_FORMAT_NON_TUNNEL ||
+	      pkt_fmt == HINIC3_RX_PKT_FORMAT_VXLAN)) {
+		skb->ip_summed = CHECKSUM_NONE;
+		return;
+	}
+
+	switch (pkt_type) {
+	case HINIC3_RX_TCP_PKT:
+	case HINIC3_RX_UDP_PKT:
+	case HINIC3_RX_SCTP_PKT:
+		skb->ip_summed = CHECKSUM_UNNECESSARY;
+		break;
+	default:
+		skb->ip_summed = CHECKSUM_NONE;
+		break;
+	}
+}
+
+static void hinic3_lro_set_gso_params(struct sk_buff *skb, u16 num_lro)
+{
+	struct ethhdr *eth = (struct ethhdr *)(skb->data);
+	__be16 proto;
+
+	proto = __vlan_get_protocol(skb, eth->h_proto, NULL);
+
+	skb_shinfo(skb)->gso_size = DIV_ROUND_UP(skb->len - skb_headlen(skb),
+						 num_lro);
+	skb_shinfo(skb)->gso_type = proto == htons(ETH_P_IP) ?
+				    SKB_GSO_TCPV4 : SKB_GSO_TCPV6;
+	skb_shinfo(skb)->gso_segs = num_lro;
+}
+
+static int recv_one_pkt(struct hinic3_rxq *rxq, struct hinic3_rq_cqe *rx_cqe,
+			u32 pkt_len, u32 vlan_len, u32 status)
+{
+	struct net_device *netdev = rxq->netdev;
+	struct sk_buff *skb;
+	u32 offload_type;
+	u16 num_lro;
+
+	skb = hinic3_fetch_rx_buffer(rxq, pkt_len);
+	if (unlikely(!skb))
+		return -ENOMEM;
+
+	/* place header in linear portion of buffer */
+	if (skb_is_nonlinear(skb))
+		hinic3_pull_tail(skb);
+
+	offload_type = le32_to_cpu(rx_cqe->offload_type);
+	hinic3_rx_csum(rxq, offload_type, status, skb);
+
+	num_lro = RQ_CQE_STATUS_GET(status, NUM_LRO);
+	if (num_lro)
+		hinic3_lro_set_gso_params(skb, num_lro);
+
+	skb_record_rx_queue(skb, rxq->q_id);
+	skb->protocol = eth_type_trans(skb, netdev);
+
+	if (skb_has_frag_list(skb)) {
+		napi_gro_flush(&rxq->irq_cfg->napi, false);
+		netif_receive_skb(skb);
+	} else {
+		napi_gro_receive(&rxq->irq_cfg->napi, skb);
+	}
+
+	return 0;
+}
+
+int hinic3_alloc_rxqs_res(struct net_device *netdev, u16 num_rq,
+			  u32 rq_depth, struct hinic3_dyna_rxq_res *rxqs_res)
+{
+	u64 cqe_mem_size = sizeof(struct hinic3_rq_cqe) * rq_depth;
+	struct hinic3_nic_dev *nic_dev = netdev_priv(netdev);
+	struct page_pool_params pp_params = {};
+	struct hinic3_dyna_rxq_res *rqres;
+	u32 pkt_idx;
+	int idx;
+
+	for (idx = 0; idx < num_rq; idx++) {
+		rqres = &rxqs_res[idx];
+		rqres->rx_info = kcalloc(rq_depth, sizeof(*rqres->rx_info),
+					 GFP_KERNEL);
+		if (!rqres->rx_info)
+			goto err_free_rqres;
+
+		rqres->cqe_start_vaddr =
+			dma_alloc_coherent(&nic_dev->pdev->dev, cqe_mem_size,
+					   &rqres->cqe_start_paddr, GFP_KERNEL);
+		if (!rqres->cqe_start_vaddr) {
+			netdev_err(netdev, "Failed to alloc rxq%d rx cqe\n",
+				   idx);
+			goto err_free_rx_info;
+		}
+
+		pp_params.flags = PP_FLAG_DMA_MAP | PP_FLAG_DMA_SYNC_DEV;
+		pp_params.pool_size = rq_depth * nic_dev->rx_buf_len /
+				      PAGE_SIZE;
+		pp_params.nid = dev_to_node(&nic_dev->pdev->dev);
+		pp_params.dev = &nic_dev->pdev->dev;
+		pp_params.dma_dir = DMA_FROM_DEVICE;
+		pp_params.max_len = PAGE_SIZE;
+		rqres->page_pool = page_pool_create(&pp_params);
+		if (IS_ERR(rqres->page_pool)) {
+			netdev_err(netdev, "Failed to create rxq%d page pool\n",
+				   idx);
+			goto err_free_cqe;
+		}
+
+		pkt_idx = hinic3_alloc_rx_buffers(rqres, rq_depth,
+						  nic_dev->rx_buf_len);
+		if (!pkt_idx) {
+			netdev_err(netdev, "Failed to alloc rxq%d rx buffers\n",
+				   idx);
+			goto err_destroy_page_pool;
+		}
+		rqres->next_to_alloc = pkt_idx;
+	}
+
+	return 0;
+
+err_destroy_page_pool:
+	page_pool_destroy(rqres->page_pool);
+err_free_cqe:
+	dma_free_coherent(&nic_dev->pdev->dev, cqe_mem_size,
+			  rqres->cqe_start_vaddr,
+			  rqres->cqe_start_paddr);
+err_free_rx_info:
+	kfree(rqres->rx_info);
+err_free_rqres:
+	hinic3_free_rxqs_res(netdev, idx, rq_depth, rxqs_res);
+
+	return -ENOMEM;
+}
+
+void hinic3_free_rxqs_res(struct net_device *netdev, u16 num_rq,
+			  u32 rq_depth, struct hinic3_dyna_rxq_res *rxqs_res)
+{
+	u64 cqe_mem_size = sizeof(struct hinic3_rq_cqe) * rq_depth;
+	struct hinic3_nic_dev *nic_dev = netdev_priv(netdev);
+	struct hinic3_dyna_rxq_res *rqres;
+	int idx;
+
+	for (idx = 0; idx < num_rq; idx++) {
+		rqres = &rxqs_res[idx];
+
+		hinic3_free_rx_buffers(rqres, rq_depth);
+		page_pool_destroy(rqres->page_pool);
+		dma_free_coherent(&nic_dev->pdev->dev, cqe_mem_size,
+				  rqres->cqe_start_vaddr,
+				  rqres->cqe_start_paddr);
+		kfree(rqres->rx_info);
+	}
+}
+
+int hinic3_configure_rxqs(struct net_device *netdev, u16 num_rq,
+			  u32 rq_depth, struct hinic3_dyna_rxq_res *rxqs_res)
+{
+	struct hinic3_nic_dev *nic_dev = netdev_priv(netdev);
+	struct hinic3_dyna_rxq_res *rqres;
+	struct msix_entry *msix_entry;
+	struct hinic3_rxq *rxq;
+	u16 q_id;
+	u32 pkts;
+
+	for (q_id = 0; q_id < num_rq; q_id++) {
+		rxq = &nic_dev->rxqs[q_id];
+		rqres = &rxqs_res[q_id];
+		msix_entry = &nic_dev->qps_msix_entries[q_id];
+
+		rxq->irq_id = msix_entry->vector;
+		rxq->msix_entry_idx = msix_entry->entry;
+		rxq->next_to_update = 0;
+		rxq->next_to_alloc = rqres->next_to_alloc;
+		rxq->q_depth = rq_depth;
+		rxq->delta = rxq->q_depth;
+		rxq->q_mask = rxq->q_depth - 1;
+		rxq->cons_idx = 0;
+
+		rxq->cqe_arr = rqres->cqe_start_vaddr;
+		rxq->cqe_start_paddr = rqres->cqe_start_paddr;
+		rxq->rx_info = rqres->rx_info;
+		rxq->page_pool = rqres->page_pool;
+
+		rxq->rq = &nic_dev->nic_io->rq[rxq->q_id];
+
+		rq_associate_cqes(rxq);
+
+		pkts = hinic3_rx_fill_buffers(rxq);
+		if (!pkts) {
+			netdev_err(netdev, "Failed to fill Rx buffer\n");
+			return -ENOMEM;
+		}
+	}
+
+	return 0;
+}
+
+int hinic3_rx_poll(struct hinic3_rxq *rxq, int budget)
+{
+	struct hinic3_nic_dev *nic_dev = netdev_priv(rxq->netdev);
+	u32 sw_ci, status, pkt_len, vlan_len;
+	struct hinic3_rq_cqe *rx_cqe;
+	u32 num_wqe = 0;
+	int nr_pkts = 0;
+	u16 num_lro;
+
+	while (likely(nr_pkts < budget)) {
+		sw_ci = rxq->cons_idx & rxq->q_mask;
+		rx_cqe = rxq->cqe_arr + sw_ci;
+		status = le32_to_cpu(rx_cqe->status);
+		if (!RQ_CQE_STATUS_GET(status, RXDONE))
+			break;
+
+		/* make sure we read rx_done before packet length */
+		rmb();
+
+		vlan_len = le32_to_cpu(rx_cqe->vlan_len);
+		pkt_len = RQ_CQE_SGE_GET(vlan_len, LEN);
+		if (recv_one_pkt(rxq, rx_cqe, pkt_len, vlan_len, status))
+			break;
+
+		nr_pkts++;
+		num_lro = RQ_CQE_STATUS_GET(status, NUM_LRO);
+		if (num_lro)
+			num_wqe += hinic3_get_sge_num(rxq, pkt_len);
+
+		rx_cqe->status = 0;
+
+		if (num_wqe >= nic_dev->lro_replenish_thld)
+			break;
+	}
+
+	if (rxq->delta >= HINIC3_RX_BUFFER_WRITE)
+		hinic3_rx_fill_buffers(rxq);
+
+	return nr_pkts;
+}
diff --git a/drivers/net/ethernet/huawei/hinic3/hinic3_rx.h b/drivers/net/ethernet/huawei/hinic3/hinic3_rx.h
new file mode 100644
index 000000000000..44ae841a3648
--- /dev/null
+++ b/drivers/net/ethernet/huawei/hinic3/hinic3_rx.h
@@ -0,0 +1,104 @@
+/* SPDX-License-Identifier: GPL-2.0 */
+/* Copyright (c) Huawei Technologies Co., Ltd. 2025. All rights reserved. */
+
+#ifndef _HINIC3_RX_H_
+#define _HINIC3_RX_H_
+
+#include <linux/bitfield.h>
+#include <linux/netdevice.h>
+
+#define RQ_CQE_OFFOLAD_TYPE_PKT_TYPE_MASK           GENMASK(4, 0)
+#define RQ_CQE_OFFOLAD_TYPE_IP_TYPE_MASK            GENMASK(6, 5)
+#define RQ_CQE_OFFOLAD_TYPE_TUNNEL_PKT_FORMAT_MASK  GENMASK(11, 8)
+#define RQ_CQE_OFFOLAD_TYPE_VLAN_EN_MASK            BIT(21)
+#define RQ_CQE_OFFOLAD_TYPE_GET(val, member) \
+	FIELD_GET(RQ_CQE_OFFOLAD_TYPE_##member##_MASK, val)
+
+#define RQ_CQE_SGE_VLAN_MASK  GENMASK(15, 0)
+#define RQ_CQE_SGE_LEN_MASK   GENMASK(31, 16)
+#define RQ_CQE_SGE_GET(val, member) \
+	FIELD_GET(RQ_CQE_SGE_##member##_MASK, val)
+
+#define RQ_CQE_STATUS_CSUM_ERR_MASK  GENMASK(15, 0)
+#define RQ_CQE_STATUS_NUM_LRO_MASK   GENMASK(23, 16)
+#define RQ_CQE_STATUS_RXDONE_MASK    BIT(31)
+#define RQ_CQE_STATUS_GET(val, member) \
+	FIELD_GET(RQ_CQE_STATUS_##member##_MASK, val)
+
+/* RX Completion information that is provided by HW for a specific RX WQE */
+struct hinic3_rq_cqe {
+	__le32 status;
+	__le32 vlan_len;
+	__le32 offload_type;
+	__le32 rsvd3;
+	__le32 rsvd4;
+	__le32 rsvd5;
+	__le32 rsvd6;
+	__le32 pkt_info;
+};
+
+struct hinic3_rq_wqe {
+	__le32 buf_hi_addr;
+	__le32 buf_lo_addr;
+	__le32 cqe_hi_addr;
+	__le32 cqe_lo_addr;
+};
+
+struct hinic3_rx_info {
+	struct page      *page;
+	u32              page_offset;
+};
+
+struct hinic3_rxq {
+	struct net_device       *netdev;
+
+	u16                     q_id;
+	u32                     q_depth;
+	u32                     q_mask;
+
+	u16                     buf_len;
+	u32                     buf_len_shift;
+
+	u32                     cons_idx;
+	u32                     delta;
+
+	u32                     irq_id;
+	u16                     msix_entry_idx;
+
+	/* cqe_arr and rx_info are arrays of rq_depth elements. Each element is
+	 * statically associated (by index) to a specific rq_wqe.
+	 */
+	struct hinic3_rq_cqe   *cqe_arr;
+	struct hinic3_rx_info  *rx_info;
+	struct page_pool       *page_pool;
+
+	struct hinic3_io_queue *rq;
+
+	struct hinic3_irq_cfg  *irq_cfg;
+	u16                    next_to_alloc;
+	u16                    next_to_update;
+	struct device          *dev; /* device for DMA mapping */
+
+	dma_addr_t             cqe_start_paddr;
+} ____cacheline_aligned;
+
+struct hinic3_dyna_rxq_res {
+	u16                   next_to_alloc;
+	struct hinic3_rx_info *rx_info;
+	dma_addr_t            cqe_start_paddr;
+	void                  *cqe_start_vaddr;
+	struct page_pool      *page_pool;
+};
+
+int hinic3_alloc_rxqs(struct net_device *netdev);
+void hinic3_free_rxqs(struct net_device *netdev);
+
+int hinic3_alloc_rxqs_res(struct net_device *netdev, u16 num_rq,
+			  u32 rq_depth, struct hinic3_dyna_rxq_res *rxqs_res);
+void hinic3_free_rxqs_res(struct net_device *netdev, u16 num_rq,
+			  u32 rq_depth, struct hinic3_dyna_rxq_res *rxqs_res);
+int hinic3_configure_rxqs(struct net_device *netdev, u16 num_rq,
+			  u32 rq_depth, struct hinic3_dyna_rxq_res *rxqs_res);
+int hinic3_rx_poll(struct hinic3_rxq *rxq, int budget);
+
+#endif
diff --git a/drivers/net/ethernet/huawei/hinic3/hinic3_tx.c b/drivers/net/ethernet/huawei/hinic3/hinic3_tx.c
new file mode 100644
index 000000000000..92c43c05e3f2
--- /dev/null
+++ b/drivers/net/ethernet/huawei/hinic3/hinic3_tx.c
@@ -0,0 +1,779 @@
+// SPDX-License-Identifier: GPL-2.0
+// Copyright (c) Huawei Technologies Co., Ltd. 2025. All rights reserved.
+
+#include <linux/if_vlan.h>
+#include <linux/iopoll.h>
+#include <net/ip6_checksum.h>
+#include <net/ipv6.h>
+#include <net/netdev_queues.h>
+
+#include "hinic3_hwdev.h"
+#include "hinic3_nic_cfg.h"
+#include "hinic3_nic_dev.h"
+#include "hinic3_nic_io.h"
+#include "hinic3_tx.h"
+#include "hinic3_wq.h"
+
+#define MIN_SKB_LEN                32
+
+int hinic3_alloc_txqs(struct net_device *netdev)
+{
+	struct hinic3_nic_dev *nic_dev = netdev_priv(netdev);
+	struct hinic3_hwdev *hwdev = nic_dev->hwdev;
+	u16 q_id, num_txqs = nic_dev->max_qps;
+	struct pci_dev *pdev = nic_dev->pdev;
+	struct hinic3_txq *txq;
+
+	if (!num_txqs) {
+		dev_err(hwdev->dev, "Cannot allocate zero size txqs\n");
+		return -EINVAL;
+	}
+
+	nic_dev->txqs = kcalloc(num_txqs, sizeof(*nic_dev->txqs),  GFP_KERNEL);
+	if (!nic_dev->txqs)
+		return -ENOMEM;
+
+	for (q_id = 0; q_id < num_txqs; q_id++) {
+		txq = &nic_dev->txqs[q_id];
+		txq->netdev = netdev;
+		txq->q_id = q_id;
+		txq->q_depth = nic_dev->q_params.sq_depth;
+		txq->q_mask = nic_dev->q_params.sq_depth - 1;
+		txq->dev = &pdev->dev;
+	}
+
+	return 0;
+}
+
+void hinic3_free_txqs(struct net_device *netdev)
+{
+	struct hinic3_nic_dev *nic_dev = netdev_priv(netdev);
+
+	kfree(nic_dev->txqs);
+}
+
+static void hinic3_set_buf_desc(struct hinic3_sq_bufdesc *buf_descs,
+				dma_addr_t addr, u32 len)
+{
+	buf_descs->hi_addr = cpu_to_le32(upper_32_bits(addr));
+	buf_descs->lo_addr = cpu_to_le32(lower_32_bits(addr));
+	buf_descs->len = cpu_to_le32(len);
+}
+
+static int hinic3_tx_map_skb(struct net_device *netdev, struct sk_buff *skb,
+			     struct hinic3_txq *txq,
+			     struct hinic3_tx_info *tx_info,
+			     struct hinic3_sq_wqe_combo *wqe_combo)
+{
+	struct hinic3_sq_wqe_desc *wqe_desc = wqe_combo->ctrl_bd0;
+	struct hinic3_sq_bufdesc *buf_desc = wqe_combo->bds_head;
+	struct hinic3_nic_dev *nic_dev = netdev_priv(netdev);
+	struct hinic3_dma_info *dma_info = tx_info->dma_info;
+	struct pci_dev *pdev = nic_dev->pdev;
+	skb_frag_t *frag;
+	u32 i, idx;
+	int err;
+
+	dma_info[0].dma = dma_map_single(&pdev->dev, skb->data,
+					 skb_headlen(skb), DMA_TO_DEVICE);
+	if (dma_mapping_error(&pdev->dev, dma_info[0].dma))
+		return -EFAULT;
+
+	dma_info[0].len = skb_headlen(skb);
+
+	wqe_desc->hi_addr = cpu_to_le32(upper_32_bits(dma_info[0].dma));
+	wqe_desc->lo_addr = cpu_to_le32(lower_32_bits(dma_info[0].dma));
+
+	wqe_desc->ctrl_len = cpu_to_le32(dma_info[0].len);
+
+	for (i = 0; i < skb_shinfo(skb)->nr_frags; i++) {
+		frag = &(skb_shinfo(skb)->frags[i]);
+		if (unlikely(i == wqe_combo->first_bds_num))
+			buf_desc = wqe_combo->bds_sec2;
+
+		idx = i + 1;
+		dma_info[idx].dma = skb_frag_dma_map(&pdev->dev, frag, 0,
+						     skb_frag_size(frag),
+						     DMA_TO_DEVICE);
+		if (dma_mapping_error(&pdev->dev, dma_info[idx].dma)) {
+			err = -EFAULT;
+			goto err_unmap_page;
+		}
+		dma_info[idx].len = skb_frag_size(frag);
+
+		hinic3_set_buf_desc(buf_desc, dma_info[idx].dma,
+				    dma_info[idx].len);
+		buf_desc++;
+	}
+
+	return 0;
+
+err_unmap_page:
+	while (idx > 1) {
+		idx--;
+		dma_unmap_page(&pdev->dev, dma_info[idx].dma,
+			       dma_info[idx].len, DMA_TO_DEVICE);
+	}
+	dma_unmap_single(&pdev->dev, dma_info[0].dma, dma_info[0].len,
+			 DMA_TO_DEVICE);
+
+	return err;
+}
+
+static void hinic3_tx_unmap_skb(struct net_device *netdev,
+				struct sk_buff *skb,
+				struct hinic3_dma_info *dma_info)
+{
+	struct hinic3_nic_dev *nic_dev = netdev_priv(netdev);
+	struct pci_dev *pdev = nic_dev->pdev;
+	int i;
+
+	for (i = 0; i < skb_shinfo(skb)->nr_frags;) {
+		i++;
+		dma_unmap_page(&pdev->dev,
+			       dma_info[i].dma,
+			       dma_info[i].len, DMA_TO_DEVICE);
+	}
+
+	dma_unmap_single(&pdev->dev, dma_info[0].dma,
+			 dma_info[0].len, DMA_TO_DEVICE);
+}
+
+static void free_all_tx_skbs(struct net_device *netdev, u32 sq_depth,
+			     struct hinic3_tx_info *tx_info_arr)
+{
+	struct hinic3_tx_info *tx_info;
+	u32 idx;
+
+	for (idx = 0; idx < sq_depth; idx++) {
+		tx_info = &tx_info_arr[idx];
+		if (tx_info->skb) {
+			hinic3_tx_unmap_skb(netdev, tx_info->skb,
+					    tx_info->dma_info);
+			dev_kfree_skb_any(tx_info->skb);
+			tx_info->skb = NULL;
+		}
+	}
+}
+
+union hinic3_ip {
+	struct iphdr   *v4;
+	struct ipv6hdr *v6;
+	unsigned char  *hdr;
+};
+
+union hinic3_l4 {
+	struct tcphdr *tcp;
+	struct udphdr *udp;
+	unsigned char *hdr;
+};
+
+enum hinic3_l3_type {
+	HINIC3_L3_UNKNOWN         = 0,
+	HINIC3_L3_IP6_PKT         = 1,
+	HINIC3_L3_IP4_PKT_NO_CSUM = 2,
+	HINIC3_L3_IP4_PKT_CSUM    = 3,
+};
+
+enum hinic3_l4_offload_type {
+	HINIC3_L4_OFFLOAD_DISABLE = 0,
+	HINIC3_L4_OFFLOAD_TCP     = 1,
+	HINIC3_L4_OFFLOAD_STCP    = 2,
+	HINIC3_L4_OFFLOAD_UDP     = 3,
+};
+
+/* initialize l4 offset and offload */
+static void get_inner_l4_info(struct sk_buff *skb, union hinic3_l4 *l4,
+			      u8 l4_proto, u32 *offset,
+			      enum hinic3_l4_offload_type *l4_offload)
+{
+	switch (l4_proto) {
+	case IPPROTO_TCP:
+		*l4_offload = HINIC3_L4_OFFLOAD_TCP;
+		/* To be same with TSO, payload offset begins from payload */
+		*offset = (l4->tcp->doff << TCP_HDR_DATA_OFF_UNIT_SHIFT) +
+			   TRANSPORT_OFFSET(l4->hdr, skb);
+		break;
+
+	case IPPROTO_UDP:
+		*l4_offload = HINIC3_L4_OFFLOAD_UDP;
+		*offset = TRANSPORT_OFFSET(l4->hdr, skb);
+		break;
+	default:
+		*l4_offload = HINIC3_L4_OFFLOAD_DISABLE;
+		*offset = 0;
+	}
+}
+
+static int hinic3_tx_csum(struct hinic3_txq *txq, struct hinic3_sq_task *task,
+			  struct sk_buff *skb)
+{
+	if (skb->ip_summed != CHECKSUM_PARTIAL)
+		return 0;
+
+	if (skb->encapsulation) {
+		union hinic3_ip ip;
+		u8 l4_proto;
+
+		task->pkt_info0 |= cpu_to_le32(SQ_TASK_INFO0_SET(1,
+								 TUNNEL_FLAG));
+
+		ip.hdr = skb_network_header(skb);
+		if (ip.v4->version == 4) {
+			l4_proto = ip.v4->protocol;
+		} else if (ip.v4->version == 6) {
+			union hinic3_l4 l4;
+			unsigned char *exthdr;
+			__be16 frag_off;
+
+			exthdr = ip.hdr + sizeof(*ip.v6);
+			l4_proto = ip.v6->nexthdr;
+			l4.hdr = skb_transport_header(skb);
+			if (l4.hdr != exthdr)
+				ipv6_skip_exthdr(skb, exthdr - skb->data,
+						 &l4_proto, &frag_off);
+		} else {
+			l4_proto = IPPROTO_RAW;
+		}
+
+		if (l4_proto != IPPROTO_UDP ||
+		    ((struct udphdr *)skb_transport_header(skb))->dest !=
+		    VXLAN_OFFLOAD_PORT_LE) {
+			/* Unsupported tunnel packet, disable csum offload */
+			skb_checksum_help(skb);
+			return 0;
+		}
+	}
+
+	task->pkt_info0 |= cpu_to_le32(SQ_TASK_INFO0_SET(1, INNER_L4_EN));
+
+	return 1;
+}
+
+static void get_inner_l3_l4_type(struct sk_buff *skb, union hinic3_ip *ip,
+				 union hinic3_l4 *l4,
+				 enum hinic3_l3_type *l3_type, u8 *l4_proto)
+{
+	unsigned char *exthdr;
+	__be16 frag_off;
+
+	if (ip->v4->version == 4) {
+		*l3_type = HINIC3_L3_IP4_PKT_CSUM;
+		*l4_proto = ip->v4->protocol;
+	} else if (ip->v4->version == 6) {
+		*l3_type = HINIC3_L3_IP6_PKT;
+		exthdr = ip->hdr + sizeof(*ip->v6);
+		*l4_proto = ip->v6->nexthdr;
+		if (exthdr != l4->hdr) {
+			ipv6_skip_exthdr(skb, exthdr - skb->data,
+					 l4_proto, &frag_off);
+		}
+	} else {
+		*l3_type = HINIC3_L3_UNKNOWN;
+		*l4_proto = 0;
+	}
+}
+
+static void hinic3_set_tso_info(struct hinic3_sq_task *task, __le32 *queue_info,
+				enum hinic3_l4_offload_type l4_offload,
+				u32 offset, u32 mss)
+{
+	if (l4_offload == HINIC3_L4_OFFLOAD_TCP) {
+		*queue_info |= cpu_to_le32(SQ_CTRL_QUEUE_INFO_SET(1, TSO));
+		task->pkt_info0 |= cpu_to_le32(SQ_TASK_INFO0_SET(1,
+								 INNER_L4_EN));
+	} else if (l4_offload == HINIC3_L4_OFFLOAD_UDP) {
+		*queue_info |= cpu_to_le32(SQ_CTRL_QUEUE_INFO_SET(1, UFO));
+		task->pkt_info0 |= cpu_to_le32(SQ_TASK_INFO0_SET(1,
+								 INNER_L4_EN));
+	}
+
+	/* enable L3 calculation */
+	task->pkt_info0 |= cpu_to_le32(SQ_TASK_INFO0_SET(1, INNER_L3_EN));
+
+	*queue_info |= cpu_to_le32(SQ_CTRL_QUEUE_INFO_SET(offset >> 1, PLDOFF));
+
+	/* set MSS value */
+	*queue_info &= cpu_to_le32(~SQ_CTRL_QUEUE_INFO_MSS_MASK);
+	*queue_info |= cpu_to_le32(SQ_CTRL_QUEUE_INFO_SET(mss, MSS));
+}
+
+static __sum16 csum_magic(union hinic3_ip *ip, unsigned short proto)
+{
+	return (ip->v4->version == 4) ?
+		csum_tcpudp_magic(ip->v4->saddr, ip->v4->daddr, 0, proto, 0) :
+		csum_ipv6_magic(&ip->v6->saddr, &ip->v6->daddr, 0, proto, 0);
+}
+
+static int hinic3_tso(struct hinic3_sq_task *task, __le32 *queue_info,
+		      struct sk_buff *skb)
+{
+	enum hinic3_l4_offload_type l4_offload;
+	enum hinic3_l3_type l3_type;
+	union hinic3_ip ip;
+	union hinic3_l4 l4;
+	u8 l4_proto;
+	u32 offset;
+	int err;
+
+	if (!skb_is_gso(skb))
+		return 0;
+
+	err = skb_cow_head(skb, 0);
+	if (err < 0)
+		return err;
+
+	if (skb->encapsulation) {
+		u32 gso_type = skb_shinfo(skb)->gso_type;
+		/* L3 checksum is always enabled */
+		task->pkt_info0 |= cpu_to_le32(SQ_TASK_INFO0_SET(1, OUT_L3_EN));
+		task->pkt_info0 |= cpu_to_le32(SQ_TASK_INFO0_SET(1,
+								 TUNNEL_FLAG));
+
+		l4.hdr = skb_transport_header(skb);
+		ip.hdr = skb_network_header(skb);
+
+		if (gso_type & SKB_GSO_UDP_TUNNEL_CSUM) {
+			l4.udp->check = ~csum_magic(&ip, IPPROTO_UDP);
+			task->pkt_info0 |=
+				cpu_to_le32(SQ_TASK_INFO0_SET(1, OUT_L4_EN));
+		}
+
+		ip.hdr = skb_inner_network_header(skb);
+		l4.hdr = skb_inner_transport_header(skb);
+	} else {
+		ip.hdr = skb_network_header(skb);
+		l4.hdr = skb_transport_header(skb);
+	}
+
+	get_inner_l3_l4_type(skb, &ip, &l4, &l3_type, &l4_proto);
+
+	if (l4_proto == IPPROTO_TCP)
+		l4.tcp->check = ~csum_magic(&ip, IPPROTO_TCP);
+
+	get_inner_l4_info(skb, &l4, l4_proto, &offset, &l4_offload);
+
+	hinic3_set_tso_info(task, queue_info, l4_offload, offset,
+			    skb_shinfo(skb)->gso_size);
+
+	return 1;
+}
+
+static void hinic3_set_vlan_tx_offload(struct hinic3_sq_task *task,
+				       u16 vlan_tag, u8 vlan_tpid)
+{
+	/* vlan_tpid: 0=select TPID0 in IPSU, 1=select TPID1 in IPSU
+	 * 2=select TPID2 in IPSU, 3=select TPID3 in IPSU,
+	 * 4=select TPID4 in IPSU
+	 */
+	task->vlan_offload =
+		cpu_to_le32(SQ_TASK_INFO3_SET(vlan_tag, VLAN_TAG) |
+			    SQ_TASK_INFO3_SET(vlan_tpid, VLAN_TPID) |
+			    SQ_TASK_INFO3_SET(1, VLAN_TAG_VALID));
+}
+
+static u32 hinic3_tx_offload(struct sk_buff *skb, struct hinic3_sq_task *task,
+			     __le32 *queue_info, struct hinic3_txq *txq)
+{
+	u32 offload = 0;
+	int tso_cs_en;
+
+	task->pkt_info0 = 0;
+	task->ip_identify = 0;
+	task->rsvd = 0;
+	task->vlan_offload = 0;
+
+	tso_cs_en = hinic3_tso(task, queue_info, skb);
+	if (tso_cs_en < 0) {
+		offload = HINIC3_TX_OFFLOAD_INVALID;
+		return offload;
+	} else if (tso_cs_en) {
+		offload |= HINIC3_TX_OFFLOAD_TSO;
+	} else {
+		tso_cs_en = hinic3_tx_csum(txq, task, skb);
+		if (tso_cs_en)
+			offload |= HINIC3_TX_OFFLOAD_CSUM;
+	}
+
+#define VLAN_INSERT_MODE_MAX 5
+	if (unlikely(skb_vlan_tag_present(skb))) {
+		/* select vlan insert mode by qid, default 802.1Q Tag type */
+		hinic3_set_vlan_tx_offload(task, skb_vlan_tag_get(skb),
+					   txq->q_id % VLAN_INSERT_MODE_MAX);
+		offload |= HINIC3_TX_OFFLOAD_VLAN;
+	}
+
+	if (unlikely(SQ_CTRL_QUEUE_INFO_GET(*queue_info, PLDOFF) >
+		     SQ_CTRL_MAX_PLDOFF)) {
+		offload = HINIC3_TX_OFFLOAD_INVALID;
+		return offload;
+	}
+
+	return offload;
+}
+
+static u16 hinic3_get_and_update_sq_owner(struct hinic3_io_queue *sq,
+					  u16 curr_pi, u16 wqebb_cnt)
+{
+	u16 owner = sq->owner;
+
+	if (unlikely(curr_pi + wqebb_cnt >= sq->wq.q_depth))
+		sq->owner = !sq->owner;
+
+	return owner;
+}
+
+static u16 hinic3_set_wqe_combo(struct hinic3_txq *txq,
+				struct hinic3_sq_wqe_combo *wqe_combo,
+				u32 offload, u16 num_sge, u16 *curr_pi)
+{
+	struct hinic3_sq_bufdesc *first_part_wqebbs, *second_part_wqebbs;
+	u16 first_part_wqebbs_num, tmp_pi;
+
+	wqe_combo->ctrl_bd0 = hinic3_wq_get_one_wqebb(&txq->sq->wq, curr_pi);
+	if (!offload && num_sge == 1) {
+		wqe_combo->wqe_type = SQ_WQE_COMPACT_TYPE;
+		return hinic3_get_and_update_sq_owner(txq->sq, *curr_pi, 1);
+	}
+
+	wqe_combo->wqe_type = SQ_WQE_EXTENDED_TYPE;
+
+	if (offload) {
+		wqe_combo->task = hinic3_wq_get_one_wqebb(&txq->sq->wq,
+							  &tmp_pi);
+		wqe_combo->task_type = SQ_WQE_TASKSECT_16BYTES;
+	} else {
+		wqe_combo->task_type = SQ_WQE_TASKSECT_46BITS;
+	}
+
+	if (num_sge > 1) {
+		/* first wqebb contain bd0, and bd size is equal to sq wqebb
+		 * size, so we use (num_sge - 1) as wanted weqbb_cnt
+		 */
+		hinic3_wq_get_multi_wqebbs(&txq->sq->wq, num_sge - 1, &tmp_pi,
+					   &first_part_wqebbs,
+					   &second_part_wqebbs,
+					   &first_part_wqebbs_num);
+		wqe_combo->bds_head = first_part_wqebbs;
+		wqe_combo->bds_sec2 = second_part_wqebbs;
+		wqe_combo->first_bds_num = first_part_wqebbs_num;
+	}
+
+	return hinic3_get_and_update_sq_owner(txq->sq, *curr_pi,
+					      num_sge + !!offload);
+}
+
+static void hinic3_prepare_sq_ctrl(struct hinic3_sq_wqe_combo *wqe_combo,
+				   __le32 queue_info, int nr_descs, u16 owner)
+{
+	struct hinic3_sq_wqe_desc *wqe_desc = wqe_combo->ctrl_bd0;
+
+	if (wqe_combo->wqe_type == SQ_WQE_COMPACT_TYPE) {
+		wqe_desc->ctrl_len |=
+			cpu_to_le32(SQ_CTRL_SET(SQ_NORMAL_WQE, DATA_FORMAT) |
+				    SQ_CTRL_SET(wqe_combo->wqe_type, EXTENDED) |
+				    SQ_CTRL_SET(owner, OWNER));
+
+		/* compact wqe queue_info will transfer to chip */
+		wqe_desc->queue_info = 0;
+		return;
+	}
+
+	wqe_desc->ctrl_len |=
+		cpu_to_le32(SQ_CTRL_SET(nr_descs, BUFDESC_NUM) |
+			    SQ_CTRL_SET(wqe_combo->task_type, TASKSECT_LEN) |
+			    SQ_CTRL_SET(SQ_NORMAL_WQE, DATA_FORMAT) |
+			    SQ_CTRL_SET(wqe_combo->wqe_type, EXTENDED) |
+			    SQ_CTRL_SET(owner, OWNER));
+
+	wqe_desc->queue_info = queue_info;
+	wqe_desc->queue_info |= cpu_to_le32(SQ_CTRL_QUEUE_INFO_SET(1, UC));
+
+	if (!SQ_CTRL_QUEUE_INFO_GET(wqe_desc->queue_info, MSS)) {
+		wqe_desc->queue_info |=
+		    cpu_to_le32(SQ_CTRL_QUEUE_INFO_SET(HINIC3_TX_MSS_DEFAULT, MSS));
+	} else if (SQ_CTRL_QUEUE_INFO_GET(wqe_desc->queue_info, MSS) <
+		   HINIC3_TX_MSS_MIN) {
+		/* mss should not be less than 80 */
+		wqe_desc->queue_info &=
+		    cpu_to_le32(~SQ_CTRL_QUEUE_INFO_MSS_MASK);
+		wqe_desc->queue_info |=
+		    cpu_to_le32(SQ_CTRL_QUEUE_INFO_SET(HINIC3_TX_MSS_MIN, MSS));
+	}
+}
+
+static netdev_tx_t hinic3_send_one_skb(struct sk_buff *skb,
+				       struct net_device *netdev,
+				       struct hinic3_txq *txq)
+{
+	struct hinic3_sq_wqe_combo wqe_combo = {};
+	struct hinic3_tx_info *tx_info;
+	struct hinic3_sq_task task;
+	u16 wqebb_cnt, num_sge;
+	__le32 queue_info = 0;
+	u16 saved_wq_prod_idx;
+	u16 owner, pi = 0;
+	u8 saved_sq_owner;
+	u32 offload;
+	int err;
+
+	if (unlikely(skb->len < MIN_SKB_LEN)) {
+		if (skb_pad(skb, MIN_SKB_LEN - skb->len))
+			goto err_out;
+
+		skb->len = MIN_SKB_LEN;
+	}
+
+	num_sge = skb_shinfo(skb)->nr_frags + 1;
+	/* assume normal wqe format + 1 wqebb for task info */
+	wqebb_cnt = num_sge + 1;
+
+	if (unlikely(hinic3_wq_free_wqebbs(&txq->sq->wq) < wqebb_cnt)) {
+		if (likely(wqebb_cnt > txq->tx_stop_thrs))
+			txq->tx_stop_thrs = min(wqebb_cnt, txq->tx_start_thrs);
+
+		netif_subqueue_try_stop(netdev, txq->sq->q_id,
+					hinic3_wq_free_wqebbs(&txq->sq->wq),
+					txq->tx_start_thrs);
+
+		return NETDEV_TX_BUSY;
+	}
+
+	offload = hinic3_tx_offload(skb, &task, &queue_info, txq);
+	if (unlikely(offload == HINIC3_TX_OFFLOAD_INVALID)) {
+		goto err_drop_pkt;
+	} else if (!offload) {
+		wqebb_cnt -= 1;
+		if (unlikely(num_sge == 1 &&
+			     skb->len > HINIC3_COMPACT_WQEE_SKB_MAX_LEN))
+			goto err_drop_pkt;
+	}
+
+	saved_wq_prod_idx = txq->sq->wq.prod_idx;
+	saved_sq_owner = txq->sq->owner;
+
+	owner = hinic3_set_wqe_combo(txq, &wqe_combo, offload, num_sge, &pi);
+	if (offload)
+		*wqe_combo.task = task;
+
+	tx_info = &txq->tx_info[pi];
+	tx_info->skb = skb;
+	tx_info->wqebb_cnt = wqebb_cnt;
+
+	err = hinic3_tx_map_skb(netdev, skb, txq, tx_info, &wqe_combo);
+	if (err) {
+		/* Rollback work queue to reclaim the wqebb we did not use */
+		txq->sq->wq.prod_idx = saved_wq_prod_idx;
+		txq->sq->owner = saved_sq_owner;
+		goto err_drop_pkt;
+	}
+
+	netif_subqueue_sent(netdev, txq->sq->q_id, skb->len);
+	netif_subqueue_maybe_stop(netdev, txq->sq->q_id,
+				  hinic3_wq_free_wqebbs(&txq->sq->wq),
+				  txq->tx_stop_thrs,
+				  txq->tx_start_thrs);
+
+	hinic3_prepare_sq_ctrl(&wqe_combo, queue_info, num_sge, owner);
+	hinic3_write_db(txq->sq, 0, DB_CFLAG_DP_SQ,
+			hinic3_get_sq_local_pi(txq->sq));
+
+	return NETDEV_TX_OK;
+
+err_drop_pkt:
+	dev_kfree_skb_any(skb);
+
+err_out:
+	return NETDEV_TX_OK;
+}
+
+netdev_tx_t hinic3_xmit_frame(struct sk_buff *skb, struct net_device *netdev)
+{
+	struct hinic3_nic_dev *nic_dev = netdev_priv(netdev);
+	u16 q_id = skb_get_queue_mapping(skb);
+
+	if (unlikely(!netif_carrier_ok(netdev)))
+		goto err_drop_pkt;
+
+	if (unlikely(q_id >= nic_dev->q_params.num_qps))
+		goto err_drop_pkt;
+
+	return hinic3_send_one_skb(skb, netdev, &nic_dev->txqs[q_id]);
+
+err_drop_pkt:
+	dev_kfree_skb_any(skb);
+
+	return NETDEV_TX_OK;
+}
+
+static bool is_hw_complete_sq_process(struct hinic3_io_queue *sq)
+{
+	u16 sw_pi, hw_ci;
+
+	sw_pi = hinic3_get_sq_local_pi(sq);
+	hw_ci = hinic3_get_sq_hw_ci(sq);
+
+	return sw_pi == hw_ci;
+}
+
+#define HINIC3_FLUSH_QUEUE_POLL_SLEEP_US   10000
+#define HINIC3_FLUSH_QUEUE_POLL_TIMEOUT_US 10000000
+static int hinic3_stop_sq(struct hinic3_txq *txq)
+{
+	struct hinic3_nic_dev *nic_dev = netdev_priv(txq->netdev);
+	int err, rc;
+
+	err = read_poll_timeout(hinic3_force_drop_tx_pkt, rc,
+				is_hw_complete_sq_process(txq->sq) || rc,
+				HINIC3_FLUSH_QUEUE_POLL_SLEEP_US,
+				HINIC3_FLUSH_QUEUE_POLL_TIMEOUT_US,
+				true, nic_dev->hwdev);
+	if (rc)
+		return rc;
+	else
+		return err;
+}
+
+/* packet transmission should be stopped before calling this function */
+void hinic3_flush_txqs(struct net_device *netdev)
+{
+	struct hinic3_nic_dev *nic_dev = netdev_priv(netdev);
+	u16 qid;
+	int err;
+
+	for (qid = 0; qid < nic_dev->q_params.num_qps; qid++) {
+		err = hinic3_stop_sq(&nic_dev->txqs[qid]);
+		netdev_tx_reset_subqueue(netdev, qid);
+		if (err)
+			netdev_err(netdev, "Failed to stop sq%u\n", qid);
+	}
+}
+
+#define HINIC3_BDS_PER_SQ_WQEBB \
+	(HINIC3_SQ_WQEBB_SIZE / sizeof(struct hinic3_sq_bufdesc))
+
+int hinic3_alloc_txqs_res(struct net_device *netdev, u16 num_sq,
+			  u32 sq_depth, struct hinic3_dyna_txq_res *txqs_res)
+{
+	struct hinic3_dyna_txq_res *tqres;
+	int idx;
+
+	for (idx = 0; idx < num_sq; idx++) {
+		tqres = &txqs_res[idx];
+
+		tqres->tx_info = kcalloc(sq_depth, sizeof(*tqres->tx_info),
+					 GFP_KERNEL);
+		if (!tqres->tx_info)
+			goto err_free_tqres;
+
+		tqres->bds = kcalloc(sq_depth * HINIC3_BDS_PER_SQ_WQEBB +
+				     HINIC3_MAX_SQ_SGE, sizeof(*tqres->bds),
+				     GFP_KERNEL);
+		if (!tqres->bds) {
+			kfree(tqres->tx_info);
+			goto err_free_tqres;
+		}
+	}
+
+	return 0;
+
+err_free_tqres:
+	while (idx > 0) {
+		idx--;
+		tqres = &txqs_res[idx];
+
+		kfree(tqres->bds);
+		kfree(tqres->tx_info);
+	}
+
+	return -ENOMEM;
+}
+
+void hinic3_free_txqs_res(struct net_device *netdev, u16 num_sq,
+			  u32 sq_depth, struct hinic3_dyna_txq_res *txqs_res)
+{
+	struct hinic3_dyna_txq_res *tqres;
+	int idx;
+
+	for (idx = 0; idx < num_sq; idx++) {
+		tqres = &txqs_res[idx];
+
+		free_all_tx_skbs(netdev, sq_depth, tqres->tx_info);
+		kfree(tqres->bds);
+		kfree(tqres->tx_info);
+	}
+}
+
+int hinic3_configure_txqs(struct net_device *netdev, u16 num_sq,
+			  u32 sq_depth, struct hinic3_dyna_txq_res *txqs_res)
+{
+	struct hinic3_nic_dev *nic_dev = netdev_priv(netdev);
+	struct hinic3_dyna_txq_res *tqres;
+	struct hinic3_txq *txq;
+	u16 q_id;
+	u32 idx;
+
+	for (q_id = 0; q_id < num_sq; q_id++) {
+		txq = &nic_dev->txqs[q_id];
+		tqres = &txqs_res[q_id];
+
+		txq->q_depth = sq_depth;
+		txq->q_mask = sq_depth - 1;
+
+		txq->tx_stop_thrs = min(HINIC3_DEFAULT_STOP_THRS,
+					sq_depth / 20);
+		txq->tx_start_thrs = min(HINIC3_DEFAULT_START_THRS,
+					 sq_depth / 10);
+
+		txq->tx_info = tqres->tx_info;
+		for (idx = 0; idx < sq_depth; idx++)
+			txq->tx_info[idx].dma_info =
+				&tqres->bds[idx * HINIC3_BDS_PER_SQ_WQEBB];
+
+		txq->sq = &nic_dev->nic_io->sq[q_id];
+	}
+
+	return 0;
+}
+
+bool hinic3_tx_poll(struct hinic3_txq *txq, int budget)
+{
+	struct net_device *netdev = txq->netdev;
+	u16 hw_ci, sw_ci, q_id = txq->sq->q_id;
+	struct hinic3_tx_info *tx_info;
+	unsigned int bytes_compl = 0;
+	unsigned int pkts = 0;
+	u16 wqebb_cnt = 0;
+
+	hw_ci = hinic3_get_sq_hw_ci(txq->sq);
+	dma_rmb();
+	sw_ci = hinic3_get_sq_local_ci(txq->sq);
+
+	do {
+		tx_info = &txq->tx_info[sw_ci];
+
+		/* Did all wqebb of this wqe complete? */
+		if (hw_ci == sw_ci ||
+		    ((hw_ci - sw_ci) & txq->q_mask) < tx_info->wqebb_cnt)
+			break;
+
+		sw_ci = (sw_ci + tx_info->wqebb_cnt) & txq->q_mask;
+		net_prefetch(&txq->tx_info[sw_ci]);
+
+		wqebb_cnt += tx_info->wqebb_cnt;
+		bytes_compl += tx_info->skb->len;
+		pkts++;
+
+		hinic3_tx_unmap_skb(netdev, tx_info->skb, tx_info->dma_info);
+		napi_consume_skb(tx_info->skb, budget);
+		tx_info->skb = NULL;
+	} while (likely(pkts < HINIC3_TX_POLL_WEIGHT));
+
+	hinic3_wq_put_wqebbs(&txq->sq->wq, wqebb_cnt);
+
+	netif_subqueue_completed_wake(netdev, q_id, pkts, bytes_compl,
+				      hinic3_wq_free_wqebbs(&txq->sq->wq),
+				      txq->tx_start_thrs);
+
+	return pkts == HINIC3_TX_POLL_WEIGHT;
+}
diff --git a/drivers/net/ethernet/huawei/hinic3/hinic3_tx.h b/drivers/net/ethernet/huawei/hinic3/hinic3_tx.h
new file mode 100644
index 000000000000..7e1b872ba752
--- /dev/null
+++ b/drivers/net/ethernet/huawei/hinic3/hinic3_tx.h
@@ -0,0 +1,147 @@
+/* SPDX-License-Identifier: GPL-2.0 */
+/* Copyright (c) Huawei Technologies Co., Ltd. 2025. All rights reserved. */
+
+#ifndef _HINIC3_TX_H_
+#define _HINIC3_TX_H_
+
+#include <linux/bitops.h>
+#include <linux/ip.h>
+#include <linux/ipv6.h>
+#include <linux/netdevice.h>
+#include <net/checksum.h>
+
+#define VXLAN_OFFLOAD_PORT_LE            cpu_to_be16(4789)
+#define TCP_HDR_DATA_OFF_UNIT_SHIFT      2
+#define TRANSPORT_OFFSET(l4_hdr, skb)    ((l4_hdr) - (skb)->data)
+
+#define HINIC3_COMPACT_WQEE_SKB_MAX_LEN  16383
+#define HINIC3_TX_POLL_WEIGHT		 64
+#define HINIC3_DEFAULT_STOP_THRS	 6
+#define HINIC3_DEFAULT_START_THRS	 24
+
+enum sq_wqe_data_format {
+	SQ_NORMAL_WQE = 0,
+};
+
+enum sq_wqe_ec_type {
+	SQ_WQE_COMPACT_TYPE  = 0,
+	SQ_WQE_EXTENDED_TYPE = 1,
+};
+
+enum sq_wqe_tasksect_len_type {
+	SQ_WQE_TASKSECT_46BITS  = 0,
+	SQ_WQE_TASKSECT_16BYTES = 1,
+};
+
+enum hinic3_tx_offload_type {
+	HINIC3_TX_OFFLOAD_TSO     = BIT(0),
+	HINIC3_TX_OFFLOAD_CSUM    = BIT(1),
+	HINIC3_TX_OFFLOAD_VLAN    = BIT(2),
+	HINIC3_TX_OFFLOAD_INVALID = BIT(3),
+	HINIC3_TX_OFFLOAD_ESP     = BIT(4),
+};
+
+#define SQ_CTRL_BUFDESC_NUM_MASK   GENMASK(26, 19)
+#define SQ_CTRL_TASKSECT_LEN_MASK  BIT(27)
+#define SQ_CTRL_DATA_FORMAT_MASK   BIT(28)
+#define SQ_CTRL_EXTENDED_MASK      BIT(30)
+#define SQ_CTRL_OWNER_MASK         BIT(31)
+#define SQ_CTRL_SET(val, member) \
+	FIELD_PREP(SQ_CTRL_##member##_MASK, val)
+
+#define SQ_CTRL_QUEUE_INFO_PLDOFF_MASK  GENMASK(9, 2)
+#define SQ_CTRL_QUEUE_INFO_UFO_MASK     BIT(10)
+#define SQ_CTRL_QUEUE_INFO_TSO_MASK     BIT(11)
+#define SQ_CTRL_QUEUE_INFO_MSS_MASK     GENMASK(26, 13)
+#define SQ_CTRL_QUEUE_INFO_UC_MASK      BIT(28)
+
+#define SQ_CTRL_QUEUE_INFO_SET(val, member) \
+	FIELD_PREP(SQ_CTRL_QUEUE_INFO_##member##_MASK, val)
+#define SQ_CTRL_QUEUE_INFO_GET(val, member) \
+	FIELD_GET(SQ_CTRL_QUEUE_INFO_##member##_MASK, le32_to_cpu(val))
+
+#define SQ_CTRL_MAX_PLDOFF  221
+
+#define SQ_TASK_INFO0_TUNNEL_FLAG_MASK  BIT(19)
+#define SQ_TASK_INFO0_INNER_L4_EN_MASK  BIT(24)
+#define SQ_TASK_INFO0_INNER_L3_EN_MASK  BIT(25)
+#define SQ_TASK_INFO0_OUT_L4_EN_MASK    BIT(27)
+#define SQ_TASK_INFO0_OUT_L3_EN_MASK    BIT(28)
+#define SQ_TASK_INFO0_SET(val, member) \
+	FIELD_PREP(SQ_TASK_INFO0_##member##_MASK, val)
+
+#define SQ_TASK_INFO3_VLAN_TAG_MASK        GENMASK(15, 0)
+#define SQ_TASK_INFO3_VLAN_TPID_MASK       GENMASK(18, 16)
+#define SQ_TASK_INFO3_VLAN_TAG_VALID_MASK  BIT(19)
+#define SQ_TASK_INFO3_SET(val, member) \
+	FIELD_PREP(SQ_TASK_INFO3_##member##_MASK, val)
+
+struct hinic3_sq_wqe_desc {
+	__le32 ctrl_len;
+	__le32 queue_info;
+	__le32 hi_addr;
+	__le32 lo_addr;
+};
+
+struct hinic3_sq_task {
+	__le32 pkt_info0;
+	__le32 ip_identify;
+	__le32 rsvd;
+	__le32 vlan_offload;
+};
+
+struct hinic3_sq_wqe_combo {
+	struct hinic3_sq_wqe_desc *ctrl_bd0;
+	struct hinic3_sq_task     *task;
+	struct hinic3_sq_bufdesc  *bds_head;
+	struct hinic3_sq_bufdesc  *bds_sec2;
+	u16                       first_bds_num;
+	u32                       wqe_type;
+	u32                       task_type;
+};
+
+struct hinic3_dma_info {
+	dma_addr_t dma;
+	u32        len;
+};
+
+struct hinic3_tx_info {
+	struct sk_buff         *skb;
+	u16                    wqebb_cnt;
+	struct hinic3_dma_info *dma_info;
+};
+
+struct hinic3_txq {
+	struct net_device       *netdev;
+	struct device           *dev;
+
+	u16                     q_id;
+	u16                     tx_stop_thrs;
+	u16                     tx_start_thrs;
+	u32                     q_mask;
+	u32                     q_depth;
+
+	struct hinic3_tx_info   *tx_info;
+	struct hinic3_io_queue  *sq;
+} ____cacheline_aligned;
+
+struct hinic3_dyna_txq_res {
+	struct hinic3_tx_info  *tx_info;
+	struct hinic3_dma_info *bds;
+};
+
+int hinic3_alloc_txqs(struct net_device *netdev);
+void hinic3_free_txqs(struct net_device *netdev);
+
+int hinic3_alloc_txqs_res(struct net_device *netdev, u16 num_sq,
+			  u32 sq_depth, struct hinic3_dyna_txq_res *txqs_res);
+void hinic3_free_txqs_res(struct net_device *netdev, u16 num_sq,
+			  u32 sq_depth, struct hinic3_dyna_txq_res *txqs_res);
+int hinic3_configure_txqs(struct net_device *netdev, u16 num_sq,
+			  u32 sq_depth, struct hinic3_dyna_txq_res *txqs_res);
+
+netdev_tx_t hinic3_xmit_frame(struct sk_buff *skb, struct net_device *netdev);
+bool hinic3_tx_poll(struct hinic3_txq *txq, int budget);
+void hinic3_flush_txqs(struct net_device *netdev);
+
+#endif
diff --git a/drivers/net/ethernet/huawei/hinic3/hinic3_wq.c b/drivers/net/ethernet/huawei/hinic3/hinic3_wq.c
new file mode 100644
index 000000000000..bc3ffdc25cf6
--- /dev/null
+++ b/drivers/net/ethernet/huawei/hinic3/hinic3_wq.c
@@ -0,0 +1,138 @@
+// SPDX-License-Identifier: GPL-2.0
+// Copyright (c) Huawei Technologies Co., Ltd. 2025. All rights reserved.
+
+#include <linux/dma-mapping.h>
+
+#include "hinic3_hwdev.h"
+#include "hinic3_wq.h"
+
+#define WQ_MIN_DEPTH            64
+#define WQ_MAX_DEPTH            65536
+#define WQ_PAGE_ADDR_SIZE       sizeof(u64)
+#define WQ_MAX_NUM_PAGES        (HINIC3_MIN_PAGE_SIZE / WQ_PAGE_ADDR_SIZE)
+
+static int wq_init_wq_block(struct hinic3_hwdev *hwdev, struct hinic3_wq *wq)
+{
+	struct hinic3_queue_pages *qpages = &wq->qpages;
+	int i;
+
+	if (hinic3_wq_is_0_level_cla(wq)) {
+		wq->wq_block_paddr = qpages->pages[0].align_paddr;
+		wq->wq_block_vaddr = qpages->pages[0].align_vaddr;
+
+		return 0;
+	}
+
+	if (wq->qpages.num_pages > WQ_MAX_NUM_PAGES) {
+		dev_err(hwdev->dev, "wq num_pages exceed limit: %lu\n",
+			WQ_MAX_NUM_PAGES);
+		return -EFAULT;
+	}
+
+	wq->wq_block_vaddr = dma_alloc_coherent(hwdev->dev,
+						HINIC3_MIN_PAGE_SIZE,
+						&wq->wq_block_paddr,
+						GFP_KERNEL);
+	if (!wq->wq_block_vaddr)
+		return -ENOMEM;
+
+	for (i = 0; i < qpages->num_pages; i++)
+		wq->wq_block_vaddr[i] = cpu_to_be64(qpages->pages[i].align_paddr);
+
+	return 0;
+}
+
+static int wq_alloc_pages(struct hinic3_hwdev *hwdev, struct hinic3_wq *wq)
+{
+	int err;
+
+	err = hinic3_queue_pages_alloc(hwdev, &wq->qpages, 0);
+	if (err)
+		return err;
+
+	err = wq_init_wq_block(hwdev, wq);
+	if (err) {
+		hinic3_queue_pages_free(hwdev, &wq->qpages);
+		return err;
+	}
+
+	return 0;
+}
+
+static void wq_free_pages(struct hinic3_hwdev *hwdev, struct hinic3_wq *wq)
+{
+	if (!hinic3_wq_is_0_level_cla(wq))
+		dma_free_coherent(hwdev->dev,
+				  HINIC3_MIN_PAGE_SIZE,
+				  wq->wq_block_vaddr,
+				  wq->wq_block_paddr);
+
+	hinic3_queue_pages_free(hwdev, &wq->qpages);
+}
+
+int hinic3_wq_create(struct hinic3_hwdev *hwdev, struct hinic3_wq *wq,
+		     u32 q_depth, u16 wqebb_size)
+{
+	u32 wq_page_size;
+
+	if (q_depth < WQ_MIN_DEPTH || q_depth > WQ_MAX_DEPTH ||
+	    !is_power_of_2(q_depth) || !is_power_of_2(wqebb_size)) {
+		dev_err(hwdev->dev, "Invalid WQ: q_depth %u, wqebb_size %u\n",
+			q_depth, wqebb_size);
+		return -EINVAL;
+	}
+
+	wq_page_size = ALIGN(hwdev->wq_page_size, HINIC3_MIN_PAGE_SIZE);
+
+	memset(wq, 0, sizeof(*wq));
+	wq->q_depth = q_depth;
+	wq->idx_mask = q_depth - 1;
+
+	hinic3_queue_pages_init(&wq->qpages, q_depth, wq_page_size, wqebb_size);
+
+	return wq_alloc_pages(hwdev, wq);
+}
+
+void hinic3_wq_destroy(struct hinic3_hwdev *hwdev, struct hinic3_wq *wq)
+{
+	wq_free_pages(hwdev, wq);
+}
+
+void hinic3_wq_reset(struct hinic3_wq *wq)
+{
+	struct hinic3_queue_pages *qpages = &wq->qpages;
+	u16 pg_idx;
+
+	wq->cons_idx = 0;
+	wq->prod_idx = 0;
+
+	for (pg_idx = 0; pg_idx < qpages->num_pages; pg_idx++)
+		memset(qpages->pages[pg_idx].align_vaddr, 0, qpages->page_size);
+}
+
+void hinic3_wq_get_multi_wqebbs(struct hinic3_wq *wq,
+				u16 num_wqebbs, u16 *prod_idx,
+				struct hinic3_sq_bufdesc **first_part_wqebbs,
+				struct hinic3_sq_bufdesc **second_part_wqebbs,
+				u16 *first_part_wqebbs_num)
+{
+	u32 idx, remaining;
+
+	idx = wq->prod_idx & wq->idx_mask;
+	wq->prod_idx += num_wqebbs;
+	*prod_idx = idx;
+	*first_part_wqebbs = get_q_element(&wq->qpages, idx, &remaining);
+	if (likely(remaining >= num_wqebbs)) {
+		*first_part_wqebbs_num = num_wqebbs;
+		*second_part_wqebbs = NULL;
+	} else {
+		*first_part_wqebbs_num = remaining;
+		idx += remaining;
+		*second_part_wqebbs = get_q_element(&wq->qpages, idx, NULL);
+	}
+}
+
+bool hinic3_wq_is_0_level_cla(const struct hinic3_wq *wq)
+{
+	return wq->qpages.num_pages == 1;
+}
diff --git a/drivers/net/ethernet/huawei/hinic3/hinic3_wq.h b/drivers/net/ethernet/huawei/hinic3/hinic3_wq.h
new file mode 100644
index 000000000000..9b3f012bec80
--- /dev/null
+++ b/drivers/net/ethernet/huawei/hinic3/hinic3_wq.h
@@ -0,0 +1,87 @@
+/* SPDX-License-Identifier: GPL-2.0 */
+/* Copyright (c) Huawei Technologies Co., Ltd. 2025. All rights reserved. */
+
+#ifndef _HINIC3_WQ_H_
+#define _HINIC3_WQ_H_
+
+#include <linux/io.h>
+
+#include "hinic3_queue_common.h"
+
+struct hinic3_sq_bufdesc {
+	/* 31-bits Length, L2NIC only uses length[17:0] */
+	__le32 len;
+	__le32 rsvd;
+	__le32 hi_addr;
+	__le32 lo_addr;
+};
+
+/* Work queue is used to submit elements (tx, rx, cmd) to hw.
+ * Driver is the producer that advances prod_idx. cons_idx is advanced when
+ * HW reports completions of previously submitted elements.
+ */
+struct hinic3_wq {
+	struct hinic3_queue_pages qpages;
+	/* Unmasked producer/consumer indices that are advanced to natural
+	 * integer overflow regardless of queue depth.
+	 */
+	u16                       cons_idx;
+	u16                       prod_idx;
+
+	u32                       q_depth;
+	u16                       idx_mask;
+
+	/* Work Queue (logical WQEBB array) is mapped to hw via Chip Logical
+	 * Address (CLA) using 1 of 2 levels:
+	 *     level 0 - direct mapping of single wq page
+	 *     level 1 - indirect mapping of multiple pages via additional page
+	 *               table.
+	 * When wq uses level 1, wq_block will hold the allocated indirection
+	 * table.
+	 */
+	dma_addr_t                wq_block_paddr;
+	__be64                    *wq_block_vaddr;
+} ____cacheline_aligned;
+
+/* Get number of elements in work queue that are in-use. */
+static inline u16 hinic3_wq_get_used(const struct hinic3_wq *wq)
+{
+	return READ_ONCE(wq->prod_idx) - READ_ONCE(wq->cons_idx);
+}
+
+static inline u16 hinic3_wq_free_wqebbs(struct hinic3_wq *wq)
+{
+	/* Don't allow queue to become completely full, report (free - 1). */
+	return wq->q_depth - hinic3_wq_get_used(wq) - 1;
+}
+
+static inline void *hinic3_wq_get_one_wqebb(struct hinic3_wq *wq, u16 *pi)
+{
+	*pi = wq->prod_idx & wq->idx_mask;
+	wq->prod_idx++;
+
+	return get_q_element(&wq->qpages, *pi, NULL);
+}
+
+static inline void hinic3_wq_put_wqebbs(struct hinic3_wq *wq, u16 num_wqebbs)
+{
+	wq->cons_idx += num_wqebbs;
+}
+
+static inline u64 hinic3_wq_get_first_wqe_page_addr(const struct hinic3_wq *wq)
+{
+	return wq->qpages.pages[0].align_paddr;
+}
+
+int hinic3_wq_create(struct hinic3_hwdev *hwdev, struct hinic3_wq *wq,
+		     u32 q_depth, u16 wqebb_size);
+void hinic3_wq_destroy(struct hinic3_hwdev *hwdev, struct hinic3_wq *wq);
+void hinic3_wq_reset(struct hinic3_wq *wq);
+void hinic3_wq_get_multi_wqebbs(struct hinic3_wq *wq,
+				u16 num_wqebbs, u16 *prod_idx,
+				struct hinic3_sq_bufdesc **first_part_wqebbs,
+				struct hinic3_sq_bufdesc **second_part_wqebbs,
+				u16 *first_part_wqebbs_num);
+bool hinic3_wq_is_0_level_cla(const struct hinic3_wq *wq);
+
+#endif
diff --git a/drivers/net/ethernet/i825xx/82596.c b/drivers/net/ethernet/i825xx/82596.c
index 945883842533..773d7aa29ef5 100644
--- a/drivers/net/ethernet/i825xx/82596.c
+++ b/drivers/net/ethernet/i825xx/82596.c
@@ -1,3 +1,4 @@
+// SPDX-License-Identifier: GPL-1.0+
 /* 82596.c: A generic 82596 ethernet driver for linux. */
 /*
    Based on Apricot.c
@@ -31,9 +32,7 @@
    Driver skeleton
    Written 1993 by Donald Becker.
    Copyright 1993 United States Government as represented by the Director,
-   National Security Agency. This software may only be used and distributed
-   according to the terms of the GNU General Public License as modified by SRC,
-   incorporated herein by reference.
+   National Security Agency.
 
    The author may be reached as becker@scyld.com, or C/O
    Scyld Computing Corporation, 410 Severn Ave., Suite 210, Annapolis MD 21403
@@ -53,10 +52,10 @@
 #include <linux/init.h>
 #include <linux/bitops.h>
 #include <linux/gfp.h>
+#include <linux/pgtable.h>
 
 #include <asm/io.h>
 #include <asm/dma.h>
-#include <asm/pgtable.h>
 #include <asm/cacheflush.h>
 
 static char version[] __initdata =
@@ -363,7 +362,7 @@ static netdev_tx_t i596_start_xmit(struct sk_buff *skb, struct net_device *dev);
 static irqreturn_t i596_interrupt(int irq, void *dev_id);
 static int i596_close(struct net_device *dev);
 static void i596_add_cmd(struct net_device *dev, struct i596_cmd *cmd);
-static void i596_tx_timeout (struct net_device *dev);
+static void i596_tx_timeout (struct net_device *dev, unsigned int txqueue);
 static void print_eth(unsigned char *buf, char *str);
 static void set_multicast_list(struct net_device *dev);
 
@@ -809,7 +808,8 @@ memory_squeeze:
 				if (!rx_in_place) {
 					/* 16 byte align the data fields */
 					skb_reserve(skb, 2);
-					memcpy(skb_put(skb,pkt_len), rbd->v_data, pkt_len);
+					skb_put_data(skb, rbd->v_data,
+						     pkt_len);
 				}
 				skb->protocol=eth_type_trans(skb,dev);
 				skb->len = pkt_len;
@@ -1018,7 +1018,7 @@ err_irq_dev:
 	return res;
 }
 
-static void i596_tx_timeout (struct net_device *dev)
+static void i596_tx_timeout (struct net_device *dev, unsigned int txqueue)
 {
 	struct i596_private *lp = dev->ml_priv;
 	int ioaddr = dev->base_addr;
@@ -1109,9 +1109,6 @@ static void print_eth(unsigned char *add, char *str)
 	       add, add + 6, add, add[12], add[13], str);
 }
 
-static int io = 0x300;
-static int irq = 10;
-
 static const struct net_device_ops i596_netdev_ops = {
 	.ndo_open 		= i596_open,
 	.ndo_stop		= i596_close,
@@ -1122,7 +1119,7 @@ static const struct net_device_ops i596_netdev_ops = {
 	.ndo_validate_addr	= eth_validate_addr,
 };
 
-struct net_device * __init i82596_probe(int unit)
+static struct net_device * __init i82596_probe(void)
 {
 	struct net_device *dev;
 	int i;
@@ -1139,14 +1136,6 @@ struct net_device * __init i82596_probe(int unit)
 	if (!dev)
 		return ERR_PTR(-ENOMEM);
 
-	if (unit >= 0) {
-		sprintf(dev->name, "eth%d", unit);
-		netdev_boot_setup_check(dev);
-	} else {
-		dev->base_addr = io;
-		dev->irq = irq;
-	}
-
 #ifdef ENABLE_MVME16x_NET
 	if (MACH_IS_MVME16x) {
 		if (mvme16x_config & MVME16x_CONFIG_NO_ETHERNET) {
@@ -1154,7 +1143,7 @@ struct net_device * __init i82596_probe(int unit)
 			err = -ENODEV;
 			goto out;
 		}
-		memcpy(eth_addr, (void *) 0xfffc1f2c, ETH_ALEN);	/* YUCK! Get addr from NOVRAM */
+		memcpy(eth_addr, absolute_pointer(0xfffc1f2c), ETH_ALEN); /* YUCK! Get addr from NOVRAM */
 		dev->base_addr = MVME_I596_BASE;
 		dev->irq = (unsigned) MVME16x_IRQ_I596;
 		goto found;
@@ -1188,7 +1177,8 @@ found:
 	DEB(DEB_PROBE,printk(KERN_INFO "%s: 82596 at %#3lx,", dev->name, dev->base_addr));
 
 	for (i = 0; i < 6; i++)
-		DEB(DEB_PROBE,printk(" %2.2X", dev->dev_addr[i] = eth_addr[i]));
+		DEB(DEB_PROBE,printk(" %2.2X", eth_addr[i]));
+	eth_hw_addr_set(dev, eth_addr);
 
 	DEB(DEB_PROBE,printk(" IRQ %d.\n", dev->irq));
 
@@ -1309,7 +1299,7 @@ static irqreturn_t i596_interrupt(int irq, void *dev_id)
 						dev->stats.tx_aborted_errors++;
 				}
 
-				dev_kfree_skb_irq(skb);
+				dev_consume_skb_irq(skb);
 
 				tx_cmd->cmd.command = 0; /* Mark free */
 				break;
@@ -1514,22 +1504,22 @@ static void set_multicast_list(struct net_device *dev)
 	}
 }
 
-#ifdef MODULE
 static struct net_device *dev_82596;
 
 static int debug = -1;
 module_param(debug, int, 0);
 MODULE_PARM_DESC(debug, "i82596 debug mask");
 
-int __init init_module(void)
+static int __init i82596_init(void)
 {
 	if (debug >= 0)
 		i596_debug = debug;
-	dev_82596 = i82596_probe(-1);
+	dev_82596 = i82596_probe();
 	return PTR_ERR_OR_ZERO(dev_82596);
 }
+module_init(i82596_init);
 
-void __exit cleanup_module(void)
+static void __exit i82596_cleanup(void)
 {
 	unregister_netdev(dev_82596);
 #ifdef __mc68000__
@@ -1543,5 +1533,4 @@ void __exit cleanup_module(void)
 	free_page ((u32)(dev_82596->mem_start));
 	free_netdev(dev_82596);
 }
-
-#endif				/* MODULE */
+module_exit(i82596_cleanup);
diff --git a/drivers/net/ethernet/i825xx/Kconfig b/drivers/net/ethernet/i825xx/Kconfig
index e8d61f670479..3b5fab123824 100644
--- a/drivers/net/ethernet/i825xx/Kconfig
+++ b/drivers/net/ethernet/i825xx/Kconfig
@@ -1,3 +1,4 @@
+# SPDX-License-Identifier: GPL-2.0-only
 #
 # Intel 82596/82593/82596 network device configuration
 #
@@ -6,7 +7,7 @@ config NET_VENDOR_I825XX
 	bool "Intel (82586/82593/82596) devices"
 	default y
 	depends on NET_VENDOR_INTEL
-	---help---
+	help
 	  If you have a network (Ethernet) card belonging to this class, say Y.
 
 	  Note that the answer to this question does not directly affect the
@@ -19,14 +20,14 @@ if NET_VENDOR_I825XX
 config ARM_ETHER1
 	tristate "Acorn Ether1 support"
 	depends on ARM && ARCH_ACORN
-	---help---
+	help
 	  If you have an Acorn system with one of these (AKA25) network cards,
 	  you should say Y to this option if you wish to use it with Linux.
 
 config BVME6000_NET
 	tristate "BVME6000 Ethernet support"
 	depends on BVME6000
-	---help---
+	help
 	  This is the driver for the Ethernet interface on BVME4000 and
 	  BVME6000 VME boards.  Say Y here to include the driver for this chip
 	  in your kernel.
@@ -35,14 +36,14 @@ config BVME6000_NET
 config LASI_82596
 	tristate "Lasi ethernet"
 	depends on GSC
-	---help---
+	help
 	  Say Y here to support the builtin Intel 82596 ethernet controller
 	  found in Hewlett-Packard PA-RISC machines with 10Mbit ethernet.
 
 config MVME16x_NET
 	tristate "MVME16x Ethernet support"
 	depends on MVME16x
-	---help---
+	help
 	  This is the driver for the Ethernet interface on the Motorola
 	  MVME162, 166, 167, 172 and 177 boards.  Say Y here to include the
 	  driver for this chip in your kernel.
@@ -51,14 +52,14 @@ config MVME16x_NET
 config SNI_82596
 	tristate "SNI RM ethernet"
 	depends on SNI_RM
-	---help---
+	help
 	  Say Y here to support the on-board Intel 82596 ethernet controller
 	  built into SNI RM machines.
 
 config SUN3_82586
 	bool "Sun3 on-board Intel 82586 support"
 	depends on SUN3
-	---help---
+	help
 	  This driver enables support for the on-board Intel 82586 based
 	  Ethernet adapter found on Sun 3/1xx and 3/2xx motherboards.  Note
 	  that this driver does not support 82586-based adapters on additional
diff --git a/drivers/net/ethernet/i825xx/Makefile b/drivers/net/ethernet/i825xx/Makefile
index 8c8dcd29c40d..422a19a5d94f 100644
--- a/drivers/net/ethernet/i825xx/Makefile
+++ b/drivers/net/ethernet/i825xx/Makefile
@@ -1,3 +1,4 @@
+# SPDX-License-Identifier: GPL-2.0
 #
 # Makefile for the Intel 82586/82593/82596 chipset device drivers.
 #
diff --git a/drivers/net/ethernet/i825xx/ether1.c b/drivers/net/ethernet/i825xx/ether1.c
index dc983450354b..3e7d7c4bafdc 100644
--- a/drivers/net/ethernet/i825xx/ether1.c
+++ b/drivers/net/ethernet/i825xx/ether1.c
@@ -1,12 +1,9 @@
+// SPDX-License-Identifier: GPL-2.0-only
 /*
  *  linux/drivers/acorn/net/ether1.c
  *
  *  Copyright (C) 1996-2000 Russell King
  *
- * This program is free software; you can redistribute it and/or modify
- * it under the terms of the GNU General Public License version 2 as
- * published by the Free Software Foundation.
- *
  *  Acorn ether1 driver (82586 chip) for Acorn machines
  *
  * We basically keep two queues in the cards memory - one for transmit
@@ -23,7 +20,7 @@
  * 1.02	RMK	25/05/1997	Added code to restart RU if it goes not ready
  * 1.03	RMK	14/09/1997	Cleaned up the handling of a reset during the TX interrupt.
  *				Should prevent lockup.
- * 1.04 RMK	17/09/1997	Added more info when initialsation of chip goes wrong.
+ * 1.04 RMK	17/09/1997	Added more info when initialisation of chip goes wrong.
  *				TDR now only reports failure when chip reports non-zero
  *				TDR time-distance.
  * 1.05	RMK	31/12/1997	Removed calls to dev_tint for 2.1
@@ -64,11 +61,12 @@ static unsigned int net_debug = NET_DEBUG;
 #define RX_AREA_END	0x0fc00
 
 static int ether1_open(struct net_device *dev);
-static int ether1_sendpacket(struct sk_buff *skb, struct net_device *dev);
+static netdev_tx_t ether1_sendpacket(struct sk_buff *skb,
+				     struct net_device *dev);
 static irqreturn_t ether1_interrupt(int irq, void *dev_id);
 static int ether1_close(struct net_device *dev);
 static void ether1_setmulticastlist(struct net_device *dev);
-static void ether1_timeout(struct net_device *dev);
+static void ether1_timeout(struct net_device *dev, unsigned int txqueue);
 
 /* ------------------------------------------------------------------------- */
 
@@ -119,7 +117,7 @@ ether1_outw_p (struct net_device *dev, unsigned short val, int addr, int svflgs)
  * Some inline assembler to allow fast transfers on to/off of the card.
  * Since this driver depends on some features presented by the ARM
  * specific architecture, and that you can't configure this driver
- * without specifiing ARM mode, this is not a problem.
+ * without specifying ARM mode, this is not a problem.
  *
  * This routine is essentially an optimised memcpy from the card's
  * onboard RAM to kernel memory.
@@ -652,7 +650,7 @@ ether1_open (struct net_device *dev)
 }
 
 static void
-ether1_timeout(struct net_device *dev)
+ether1_timeout(struct net_device *dev, unsigned int txqueue)
 {
 	printk(KERN_WARNING "%s: transmit timeout, network cable problem?\n",
 		dev->name);
@@ -667,7 +665,7 @@ ether1_timeout(struct net_device *dev)
 	netif_wake_queue(dev);
 }
 
-static int
+static netdev_tx_t
 ether1_sendpacket (struct sk_buff *skb, struct net_device *dev)
 {
 	int tmp, tst, nopaddr, txaddr, tbdaddr, dataddr;
@@ -988,6 +986,7 @@ static int
 ether1_probe(struct expansion_card *ec, const struct ecard_id *id)
 {
 	struct net_device *dev;
+	u8 addr[ETH_ALEN];
 	int i, ret = 0;
 
 	ether1_banner();
@@ -1017,7 +1016,8 @@ ether1_probe(struct expansion_card *ec, const struct ecard_id *id)
 	}
 
 	for (i = 0; i < 6; i++)
-		dev->dev_addr[i] = readb(IDPROM_ADDRESS + (i << 2));
+		addr[i] = readb(IDPROM_ADDRESS + (i << 2));
+	eth_hw_addr_set(dev, addr);
 
 	if (ether1_init_2(dev)) {
 		ret = -ENODEV;
diff --git a/drivers/net/ethernet/i825xx/ether1.h b/drivers/net/ethernet/i825xx/ether1.h
index 3a5830ab3dc7..3926e042fe2e 100644
--- a/drivers/net/ethernet/i825xx/ether1.h
+++ b/drivers/net/ethernet/i825xx/ether1.h
@@ -1,12 +1,9 @@
+/* SPDX-License-Identifier: GPL-2.0-only */
 /*
  *  linux/drivers/acorn/net/ether1.h
  *
  *  Copyright (C) 1996 Russell King
  *
- * This program is free software; you can redistribute it and/or modify
- * it under the terms of the GNU General Public License version 2 as
- * published by the Free Software Foundation.
- *
  *  Network driver for Acorn Ether1 cards.
  */
 
diff --git a/drivers/net/ethernet/i825xx/lasi_82596.c b/drivers/net/ethernet/i825xx/lasi_82596.c
index d787fdd5db7b..3e53e0c243ba 100644
--- a/drivers/net/ethernet/i825xx/lasi_82596.c
+++ b/drivers/net/ethernet/i825xx/lasi_82596.c
@@ -1,3 +1,4 @@
+// SPDX-License-Identifier: GPL-1.0+
 /* lasi_82596.c -- driver for the intel 82596 ethernet controller, as
    munged into HPPA boxen .
 
@@ -59,9 +60,7 @@
    Driver skeleton
    Written 1993 by Donald Becker.
    Copyright 1993 United States Government as represented by the Director,
-   National Security Agency. This software may only be used and distributed
-   according to the terms of the GNU General Public License as modified by SRC,
-   incorporated herein by reference.
+   National Security Agency.
 
    The author may be reached as becker@scyld.com, or C/O
    Scyld Computing Corporation, 410 Severn Ave., Suite 210, Annapolis MD 21403
@@ -96,23 +95,14 @@
 
 #define OPT_SWAP_PORT	0x0001	/* Need to wordswp on the MPU port */
 
-#define DMA_ALLOC                        dma_alloc_noncoherent
-#define DMA_FREE                         dma_free_noncoherent
-#define DMA_WBACK(ndev, addr, len) \
-	do { dma_cache_sync((ndev)->dev.parent, (void *)addr, len, DMA_TO_DEVICE); } while (0)
-
-#define DMA_INV(ndev, addr, len) \
-	do { dma_cache_sync((ndev)->dev.parent, (void *)addr, len, DMA_FROM_DEVICE); } while (0)
-
-#define DMA_WBACK_INV(ndev, addr, len) \
-	do { dma_cache_sync((ndev)->dev.parent, (void *)addr, len, DMA_BIDIRECTIONAL); } while (0)
-
-#define SYSBUS      0x0000006c;
+#define SYSBUS      0x0000006c
 
 /* big endian CPU, 82596 "big" endian mode */
 #define SWAP32(x)   (((u32)(x)<<16) | ((((u32)(x)))>>16))
 #define SWAP16(x)   (x)
 
+#define NONCOHERENT_DMA 1
+
 #include "lib82596.c"
 
 MODULE_AUTHOR("Richard Hirst");
@@ -143,18 +133,20 @@ static void mpu_port(struct net_device *dev, int c, dma_addr_t x)
 	}
 
 	gsc_writel(a, dev->base_addr + PA_CPU_PORT_L_ACCESS);
-	udelay(1);
+	if (!running_on_qemu)
+		udelay(1);
 	gsc_writel(b, dev->base_addr + PA_CPU_PORT_L_ACCESS);
 }
 
 #define LAN_PROM_ADDR	0xF0810000
 
-static int
+static int __init
 lan_init_chip(struct parisc_device *dev)
 {
 	struct	net_device *netdevice;
 	struct i596_private *lp;
-	int	retval;
+	int retval = -ENOMEM;
+	u8 addr[ETH_ALEN];
 	int i;
 
 	if (!dev->irq) {
@@ -175,38 +167,48 @@ lan_init_chip(struct parisc_device *dev)
 	netdevice->base_addr = dev->hpa.start;
 	netdevice->irq = dev->irq;
 
-	if (pdc_lan_station_id(netdevice->dev_addr, netdevice->base_addr)) {
+	if (pdc_lan_station_id(addr, netdevice->base_addr)) {
 		for (i = 0; i < 6; i++) {
-			netdevice->dev_addr[i] = gsc_readb(LAN_PROM_ADDR + i);
+			addr[i] = gsc_readb(LAN_PROM_ADDR + i);
 		}
 		printk(KERN_INFO
 		       "%s: MAC of HP700 LAN read from EEPROM\n", __FILE__);
 	}
+	eth_hw_addr_set(netdevice, addr);
 
 	lp = netdev_priv(netdevice);
 	lp->options = dev->id.sversion == 0x72 ? OPT_SWAP_PORT : 0;
+	lp->dma = dma_alloc_noncoherent(&dev->dev,
+			sizeof(struct i596_dma), &lp->dma_addr,
+			DMA_BIDIRECTIONAL, GFP_KERNEL);
+	if (!lp->dma)
+		goto out_free_netdev;
 
 	retval = i82596_probe(netdevice);
-	if (retval) {
-		free_netdev(netdevice);
-		return -ENODEV;
-	}
+	if (retval)
+		goto out_free_dma;
+	return 0;
+
+out_free_dma:
+	dma_free_noncoherent(&dev->dev, sizeof(struct i596_dma),
+		       lp->dma, lp->dma_addr, DMA_BIDIRECTIONAL);
+out_free_netdev:
+	free_netdev(netdevice);
 	return retval;
 }
 
-static int lan_remove_chip(struct parisc_device *pdev)
+static void __exit lan_remove_chip(struct parisc_device *pdev)
 {
 	struct net_device *dev = parisc_get_drvdata(pdev);
 	struct i596_private *lp = netdev_priv(dev);
 
 	unregister_netdev (dev);
-	DMA_FREE(&pdev->dev, sizeof(struct i596_private),
-		 (void *)lp->dma, lp->dma_addr);
+	dma_free_noncoherent(&pdev->dev, sizeof(struct i596_private), lp->dma,
+		       lp->dma_addr, DMA_BIDIRECTIONAL);
 	free_netdev (dev);
-	return 0;
 }
 
-static struct parisc_device_id lan_tbl[] = {
+static const struct parisc_device_id lan_tbl[] __initconst = {
 	{ HPHW_FIO, HVERSION_REV_ANY_ID, HVERSION_ANY_ID, 0x0008a },
 	{ HPHW_FIO, HVERSION_REV_ANY_ID, HVERSION_ANY_ID, 0x00072 },
 	{ 0, }
@@ -214,11 +216,11 @@ static struct parisc_device_id lan_tbl[] = {
 
 MODULE_DEVICE_TABLE(parisc, lan_tbl);
 
-static struct parisc_driver lan_driver = {
+static struct parisc_driver lan_driver __refdata = {
 	.name		= "lasi_82596",
 	.id_table	= lan_tbl,
 	.probe		= lan_init_chip,
-	.remove         = lan_remove_chip,
+	.remove         = __exit_p(lan_remove_chip),
 };
 
 static int lasi_82596_init(void)
diff --git a/drivers/net/ethernet/i825xx/lib82596.c b/drivers/net/ethernet/i825xx/lib82596.c
index e86773325cbe..67d248a7a6f4 100644
--- a/drivers/net/ethernet/i825xx/lib82596.c
+++ b/drivers/net/ethernet/i825xx/lib82596.c
@@ -1,3 +1,4 @@
+// SPDX-License-Identifier: GPL-1.0+
 /* lasi_82596.c -- driver for the intel 82596 ethernet controller, as
    munged into HPPA boxen .
 
@@ -59,9 +60,7 @@
    Driver skeleton
    Written 1993 by Donald Becker.
    Copyright 1993 United States Government as represented by the Director,
-   National Security Agency. This software may only be used and distributed
-   according to the terms of the GNU General Public License as modified by SRC,
-   incorporated herein by reference.
+   National Security Agency.
 
    The author may be reached as becker@scyld.com, or C/O
    Scyld Computing Corporation, 410 Severn Ave., Suite 210, Annapolis MD 21403
@@ -347,11 +346,11 @@ static const char init_setup[] =
 	0x7f /*  *multi IA */ };
 
 static int i596_open(struct net_device *dev);
-static int i596_start_xmit(struct sk_buff *skb, struct net_device *dev);
+static netdev_tx_t i596_start_xmit(struct sk_buff *skb, struct net_device *dev);
 static irqreturn_t i596_interrupt(int irq, void *dev_id);
 static int i596_close(struct net_device *dev);
 static void i596_add_cmd(struct net_device *dev, struct i596_cmd *cmd);
-static void i596_tx_timeout (struct net_device *dev);
+static void i596_tx_timeout (struct net_device *dev, unsigned int txqueue);
 static void print_eth(unsigned char *buf, char *str);
 static void set_multicast_list(struct net_device *dev);
 static inline void ca(struct net_device *dev);
@@ -365,13 +364,44 @@ static int max_cmd_backlog = TX_RING_SIZE-1;
 static void i596_poll_controller(struct net_device *dev);
 #endif
 
+static inline dma_addr_t virt_to_dma(struct i596_private *lp, volatile void *v)
+{
+	return lp->dma_addr + ((unsigned long)v - (unsigned long)lp->dma);
+}
+
+#ifdef NONCOHERENT_DMA
+static inline void dma_sync_dev(struct net_device *ndev, volatile void *addr,
+		size_t len)
+{
+	dma_sync_single_for_device(ndev->dev.parent,
+			virt_to_dma(netdev_priv(ndev), addr), len,
+			DMA_BIDIRECTIONAL);
+}
+
+static inline void dma_sync_cpu(struct net_device *ndev, volatile void *addr,
+		size_t len)
+{
+	dma_sync_single_for_cpu(ndev->dev.parent,
+			virt_to_dma(netdev_priv(ndev), addr), len,
+			DMA_BIDIRECTIONAL);
+}
+#else
+static inline void dma_sync_dev(struct net_device *ndev, volatile void *addr,
+		size_t len)
+{
+}
+static inline void dma_sync_cpu(struct net_device *ndev, volatile void *addr,
+		size_t len)
+{
+}
+#endif /* NONCOHERENT_DMA */
 
 static inline int wait_istat(struct net_device *dev, struct i596_dma *dma, int delcnt, char *str)
 {
-	DMA_INV(dev, &(dma->iscp), sizeof(struct i596_iscp));
+	dma_sync_cpu(dev, &(dma->iscp), sizeof(struct i596_iscp));
 	while (--delcnt && dma->iscp.stat) {
 		udelay(10);
-		DMA_INV(dev, &(dma->iscp), sizeof(struct i596_iscp));
+		dma_sync_cpu(dev, &(dma->iscp), sizeof(struct i596_iscp));
 	}
 	if (!delcnt) {
 		printk(KERN_ERR "%s: %s, iscp.stat %04x, didn't clear\n",
@@ -384,10 +414,10 @@ static inline int wait_istat(struct net_device *dev, struct i596_dma *dma, int d
 
 static inline int wait_cmd(struct net_device *dev, struct i596_dma *dma, int delcnt, char *str)
 {
-	DMA_INV(dev, &(dma->scb), sizeof(struct i596_scb));
+	dma_sync_cpu(dev, &(dma->scb), sizeof(struct i596_scb));
 	while (--delcnt && dma->scb.command) {
 		udelay(10);
-		DMA_INV(dev, &(dma->scb), sizeof(struct i596_scb));
+		dma_sync_cpu(dev, &(dma->scb), sizeof(struct i596_scb));
 	}
 	if (!delcnt) {
 		printk(KERN_ERR "%s: %s, status %4.4x, cmd %4.4x.\n",
@@ -451,12 +481,9 @@ static void i596_display_data(struct net_device *dev)
 		       SWAP32(rbd->b_data), SWAP16(rbd->size));
 		rbd = rbd->v_next;
 	} while (rbd != lp->rbd_head);
-	DMA_INV(dev, dma, sizeof(struct i596_dma));
+	dma_sync_cpu(dev, dma, sizeof(struct i596_dma));
 }
 
-
-#define virt_to_dma(lp, v) ((lp)->dma_addr + (dma_addr_t)((unsigned long)(v)-(unsigned long)((lp)->dma)))
-
 static inline int init_rx_bufs(struct net_device *dev)
 {
 	struct i596_private *lp = netdev_priv(dev);
@@ -508,7 +535,7 @@ static inline int init_rx_bufs(struct net_device *dev)
 	rfd->b_next = SWAP32(virt_to_dma(lp, dma->rfds));
 	rfd->cmd = SWAP16(CMD_EOL|CMD_FLEX);
 
-	DMA_WBACK_INV(dev, dma, sizeof(struct i596_dma));
+	dma_sync_dev(dev, dma, sizeof(struct i596_dma));
 	return 0;
 }
 
@@ -547,7 +574,7 @@ static void rebuild_rx_bufs(struct net_device *dev)
 	lp->rbd_head = dma->rbds;
 	dma->rfds[0].rbd = SWAP32(virt_to_dma(lp, dma->rbds));
 
-	DMA_WBACK_INV(dev, dma, sizeof(struct i596_dma));
+	dma_sync_dev(dev, dma, sizeof(struct i596_dma));
 }
 
 
@@ -575,9 +602,9 @@ static int init_i596_mem(struct net_device *dev)
 
 	DEB(DEB_INIT, printk(KERN_DEBUG "%s: starting i82596.\n", dev->name));
 
-	DMA_WBACK(dev, &(dma->scp), sizeof(struct i596_scp));
-	DMA_WBACK(dev, &(dma->iscp), sizeof(struct i596_iscp));
-	DMA_WBACK(dev, &(dma->scb), sizeof(struct i596_scb));
+	dma_sync_dev(dev, &(dma->scp), sizeof(struct i596_scp));
+	dma_sync_dev(dev, &(dma->iscp), sizeof(struct i596_iscp));
+	dma_sync_dev(dev, &(dma->scb), sizeof(struct i596_scb));
 
 	mpu_port(dev, PORT_ALTSCP, virt_to_dma(lp, &dma->scp));
 	ca(dev);
@@ -596,24 +623,24 @@ static int init_i596_mem(struct net_device *dev)
 	rebuild_rx_bufs(dev);
 
 	dma->scb.command = 0;
-	DMA_WBACK(dev, &(dma->scb), sizeof(struct i596_scb));
+	dma_sync_dev(dev, &(dma->scb), sizeof(struct i596_scb));
 
 	DEB(DEB_INIT, printk(KERN_DEBUG
 			     "%s: queuing CmdConfigure\n", dev->name));
 	memcpy(dma->cf_cmd.i596_config, init_setup, 14);
 	dma->cf_cmd.cmd.command = SWAP16(CmdConfigure);
-	DMA_WBACK(dev, &(dma->cf_cmd), sizeof(struct cf_cmd));
+	dma_sync_dev(dev, &(dma->cf_cmd), sizeof(struct cf_cmd));
 	i596_add_cmd(dev, &dma->cf_cmd.cmd);
 
 	DEB(DEB_INIT, printk(KERN_DEBUG "%s: queuing CmdSASetup\n", dev->name));
 	memcpy(dma->sa_cmd.eth_addr, dev->dev_addr, ETH_ALEN);
 	dma->sa_cmd.cmd.command = SWAP16(CmdSASetup);
-	DMA_WBACK(dev, &(dma->sa_cmd), sizeof(struct sa_cmd));
+	dma_sync_dev(dev, &(dma->sa_cmd), sizeof(struct sa_cmd));
 	i596_add_cmd(dev, &dma->sa_cmd.cmd);
 
 	DEB(DEB_INIT, printk(KERN_DEBUG "%s: queuing CmdTDR\n", dev->name));
 	dma->tdr_cmd.cmd.command = SWAP16(CmdTDR);
-	DMA_WBACK(dev, &(dma->tdr_cmd), sizeof(struct tdr_cmd));
+	dma_sync_dev(dev, &(dma->tdr_cmd), sizeof(struct tdr_cmd));
 	i596_add_cmd(dev, &dma->tdr_cmd.cmd);
 
 	spin_lock_irqsave (&lp->lock, flags);
@@ -625,7 +652,7 @@ static int init_i596_mem(struct net_device *dev)
 	DEB(DEB_INIT, printk(KERN_DEBUG "%s: Issuing RX_START\n", dev->name));
 	dma->scb.command = SWAP16(RX_START);
 	dma->scb.rfd = SWAP32(virt_to_dma(lp, dma->rfds));
-	DMA_WBACK(dev, &(dma->scb), sizeof(struct i596_scb));
+	dma_sync_dev(dev, &(dma->scb), sizeof(struct i596_scb));
 
 	ca(dev);
 
@@ -659,13 +686,13 @@ static inline int i596_rx(struct net_device *dev)
 
 	rfd = lp->rfd_head;		/* Ref next frame to check */
 
-	DMA_INV(dev, rfd, sizeof(struct i596_rfd));
+	dma_sync_cpu(dev, rfd, sizeof(struct i596_rfd));
 	while (rfd->stat & SWAP16(STAT_C)) {	/* Loop while complete frames */
 		if (rfd->rbd == I596_NULL)
 			rbd = NULL;
 		else if (rfd->rbd == lp->rbd_head->b_addr) {
 			rbd = lp->rbd_head;
-			DMA_INV(dev, rbd, sizeof(struct i596_rbd));
+			dma_sync_cpu(dev, rbd, sizeof(struct i596_rbd));
 		} else {
 			printk(KERN_ERR "%s: rbd chain broken!\n", dev->name);
 			/* XXX Now what? */
@@ -713,7 +740,7 @@ static inline int i596_rx(struct net_device *dev)
 							  DMA_FROM_DEVICE);
 				rbd->v_data = newskb->data;
 				rbd->b_data = SWAP32(dma_addr);
-				DMA_WBACK_INV(dev, rbd, sizeof(struct i596_rbd));
+				dma_sync_dev(dev, rbd, sizeof(struct i596_rbd));
 			} else {
 				skb = netdev_alloc_skb_ip_align(dev, pkt_len);
 			}
@@ -727,7 +754,8 @@ memory_squeeze:
 					dma_sync_single_for_cpu(dev->dev.parent,
 								(dma_addr_t)SWAP32(rbd->b_data),
 								PKT_BUF_SZ, DMA_FROM_DEVICE);
-					memcpy(skb_put(skb, pkt_len), rbd->v_data, pkt_len);
+					skb_put_data(skb, rbd->v_data,
+						     pkt_len);
 					dma_sync_single_for_device(dev->dev.parent,
 								   (dma_addr_t)SWAP32(rbd->b_data),
 								   PKT_BUF_SZ, DMA_FROM_DEVICE);
@@ -764,7 +792,7 @@ memory_squeeze:
 		if (rbd != NULL && (rbd->count & SWAP16(0x4000))) {
 			rbd->count = 0;
 			lp->rbd_head = rbd->v_next;
-			DMA_WBACK_INV(dev, rbd, sizeof(struct i596_rbd));
+			dma_sync_dev(dev, rbd, sizeof(struct i596_rbd));
 		}
 
 		/* Tidy the frame descriptor, marking it as end of list */
@@ -778,14 +806,14 @@ memory_squeeze:
 
 		lp->dma->scb.rfd = rfd->b_next;
 		lp->rfd_head = rfd->v_next;
-		DMA_WBACK_INV(dev, rfd, sizeof(struct i596_rfd));
+		dma_sync_dev(dev, rfd, sizeof(struct i596_rfd));
 
 		/* Remove end-of-list from old end descriptor */
 
 		rfd->v_prev->cmd = SWAP16(CMD_FLEX);
-		DMA_WBACK_INV(dev, rfd->v_prev, sizeof(struct i596_rfd));
+		dma_sync_dev(dev, rfd->v_prev, sizeof(struct i596_rfd));
 		rfd = lp->rfd_head;
-		DMA_INV(dev, rfd, sizeof(struct i596_rfd));
+		dma_sync_cpu(dev, rfd, sizeof(struct i596_rfd));
 	}
 
 	DEB(DEB_RXFRAME, printk(KERN_DEBUG "frames %d\n", frames));
@@ -826,12 +854,12 @@ static inline void i596_cleanup_cmd(struct net_device *dev, struct i596_private
 			ptr->v_next = NULL;
 			ptr->b_next = I596_NULL;
 		}
-		DMA_WBACK_INV(dev, ptr, sizeof(struct i596_cmd));
+		dma_sync_dev(dev, ptr, sizeof(struct i596_cmd));
 	}
 
 	wait_cmd(dev, lp->dma, 100, "i596_cleanup_cmd timed out");
 	lp->dma->scb.cmd = I596_NULL;
-	DMA_WBACK(dev, &(lp->dma->scb), sizeof(struct i596_scb));
+	dma_sync_dev(dev, &(lp->dma->scb), sizeof(struct i596_scb));
 }
 
 
@@ -849,7 +877,7 @@ static inline void i596_reset(struct net_device *dev, struct i596_private *lp)
 
 	/* FIXME: this command might cause an lpmc */
 	lp->dma->scb.command = SWAP16(CUC_ABORT | RX_ABORT);
-	DMA_WBACK(dev, &(lp->dma->scb), sizeof(struct i596_scb));
+	dma_sync_dev(dev, &(lp->dma->scb), sizeof(struct i596_scb));
 	ca(dev);
 
 	/* wait for shutdown */
@@ -877,20 +905,20 @@ static void i596_add_cmd(struct net_device *dev, struct i596_cmd *cmd)
 	cmd->command |= SWAP16(CMD_EOL | CMD_INTR);
 	cmd->v_next = NULL;
 	cmd->b_next = I596_NULL;
-	DMA_WBACK(dev, cmd, sizeof(struct i596_cmd));
+	dma_sync_dev(dev, cmd, sizeof(struct i596_cmd));
 
 	spin_lock_irqsave (&lp->lock, flags);
 
 	if (lp->cmd_head != NULL) {
 		lp->cmd_tail->v_next = cmd;
 		lp->cmd_tail->b_next = SWAP32(virt_to_dma(lp, &cmd->status));
-		DMA_WBACK(dev, lp->cmd_tail, sizeof(struct i596_cmd));
+		dma_sync_dev(dev, lp->cmd_tail, sizeof(struct i596_cmd));
 	} else {
 		lp->cmd_head = cmd;
 		wait_cmd(dev, dma, 100, "i596_add_cmd timed out");
 		dma->scb.cmd = SWAP32(virt_to_dma(lp, &cmd->status));
 		dma->scb.command = SWAP16(CUC_START);
-		DMA_WBACK(dev, &(dma->scb), sizeof(struct i596_scb));
+		dma_sync_dev(dev, &(dma->scb), sizeof(struct i596_scb));
 		ca(dev);
 	}
 	lp->cmd_tail = cmd;
@@ -935,7 +963,7 @@ out_remove_rx_bufs:
 	return -EAGAIN;
 }
 
-static void i596_tx_timeout (struct net_device *dev)
+static void i596_tx_timeout (struct net_device *dev, unsigned int txqueue)
 {
 	struct i596_private *lp = netdev_priv(dev);
 
@@ -955,7 +983,7 @@ static void i596_tx_timeout (struct net_device *dev)
 		/* Issue a channel attention signal */
 		DEB(DEB_ERRORS, printk(KERN_DEBUG "Kicking board.\n"));
 		lp->dma->scb.command = SWAP16(CUC_START | RX_START);
-		DMA_WBACK_INV(dev, &(lp->dma->scb), sizeof(struct i596_scb));
+		dma_sync_dev(dev, &(lp->dma->scb), sizeof(struct i596_scb));
 		ca (dev);
 		lp->last_restart = dev->stats.tx_packets;
 	}
@@ -965,7 +993,7 @@ static void i596_tx_timeout (struct net_device *dev)
 }
 
 
-static int i596_start_xmit(struct sk_buff *skb, struct net_device *dev)
+static netdev_tx_t i596_start_xmit(struct sk_buff *skb, struct net_device *dev)
 {
 	struct i596_private *lp = netdev_priv(dev);
 	struct tx_cmd *tx_cmd;
@@ -1013,8 +1041,8 @@ static int i596_start_xmit(struct sk_buff *skb, struct net_device *dev)
 		tbd->data = SWAP32(tx_cmd->dma_addr);
 
 		DEB(DEB_TXADDR, print_eth(skb->data, "tx-queued"));
-		DMA_WBACK_INV(dev, tx_cmd, sizeof(struct tx_cmd));
-		DMA_WBACK_INV(dev, tbd, sizeof(struct i596_tbd));
+		dma_sync_dev(dev, tx_cmd, sizeof(struct tx_cmd));
+		dma_sync_dev(dev, tbd, sizeof(struct i596_tbd));
 		i596_add_cmd(dev, &tx_cmd->cmd);
 
 		dev->stats.tx_packets++;
@@ -1046,9 +1074,8 @@ static const struct net_device_ops i596_netdev_ops = {
 
 static int i82596_probe(struct net_device *dev)
 {
-	int i;
 	struct i596_private *lp = netdev_priv(dev);
-	struct i596_dma *dma;
+	int ret;
 
 	/* This lot is ensure things have been cache line aligned. */
 	BUILD_BUG_ON(sizeof(struct i596_rfd) != 32);
@@ -1062,40 +1089,28 @@ static int i82596_probe(struct net_device *dev)
 	if (!dev->base_addr || !dev->irq)
 		return -ENODEV;
 
-	dma = (struct i596_dma *) DMA_ALLOC(dev->dev.parent,
-		sizeof(struct i596_dma), &lp->dma_addr, GFP_KERNEL);
-	if (!dma) {
-		printk(KERN_ERR "%s: Couldn't get shared memory\n", __FILE__);
-		return -ENOMEM;
-	}
-
 	dev->netdev_ops = &i596_netdev_ops;
 	dev->watchdog_timeo = TX_TIMEOUT;
 
-	memset(dma, 0, sizeof(struct i596_dma));
-	lp->dma = dma;
-
-	dma->scb.command = 0;
-	dma->scb.cmd = I596_NULL;
-	dma->scb.rfd = I596_NULL;
+	memset(lp->dma, 0, sizeof(struct i596_dma));
+	lp->dma->scb.command = 0;
+	lp->dma->scb.cmd = I596_NULL;
+	lp->dma->scb.rfd = I596_NULL;
 	spin_lock_init(&lp->lock);
 
-	DMA_WBACK_INV(dev, dma, sizeof(struct i596_dma));
+	dma_sync_dev(dev, lp->dma, sizeof(struct i596_dma));
 
-	i = register_netdev(dev);
-	if (i) {
-		DMA_FREE(dev->dev.parent, sizeof(struct i596_dma),
-				    (void *)dma, lp->dma_addr);
-		return i;
-	}
+	ret = register_netdev(dev);
+	if (ret)
+		return ret;
 
 	DEB(DEB_PROBE, printk(KERN_INFO "%s: 82596 at %#3lx, %pM IRQ %d.\n",
 			      dev->name, dev->base_addr, dev->dev_addr,
 			      dev->irq));
 	DEB(DEB_INIT, printk(KERN_INFO
 			     "%s: dma at 0x%p (%d bytes), lp->scb at 0x%p\n",
-			     dev->name, dma, (int)sizeof(struct i596_dma),
-			     &dma->scb));
+			     dev->name, lp->dma, (int)sizeof(struct i596_dma),
+			     &lp->dma->scb));
 
 	return 0;
 }
@@ -1153,7 +1168,7 @@ static irqreturn_t i596_interrupt(int irq, void *dev_id)
 				   dev->name, status & 0x0700));
 
 		while (lp->cmd_head != NULL) {
-			DMA_INV(dev, lp->cmd_head, sizeof(struct i596_cmd));
+			dma_sync_cpu(dev, lp->cmd_head, sizeof(struct i596_cmd));
 			if (!(lp->cmd_head->status & SWAP16(STAT_C)))
 				break;
 
@@ -1192,7 +1207,7 @@ static irqreturn_t i596_interrupt(int irq, void *dev_id)
 				dma_unmap_single(dev->dev.parent,
 						 tx_cmd->dma_addr,
 						 skb->len, DMA_TO_DEVICE);
-				dev_kfree_skb_irq(skb);
+				dev_consume_skb_irq(skb);
 
 				tx_cmd->cmd.command = 0; /* Mark free */
 				break;
@@ -1235,7 +1250,7 @@ static irqreturn_t i596_interrupt(int irq, void *dev_id)
 			}
 			ptr->v_next = NULL;
 			ptr->b_next = I596_NULL;
-			DMA_WBACK(dev, ptr, sizeof(struct i596_cmd));
+			dma_sync_dev(dev, ptr, sizeof(struct i596_cmd));
 			lp->last_cmd = jiffies;
 		}
 
@@ -1249,13 +1264,13 @@ static irqreturn_t i596_interrupt(int irq, void *dev_id)
 
 			ptr->command &= SWAP16(0x1fff);
 			ptr = ptr->v_next;
-			DMA_WBACK_INV(dev, prev, sizeof(struct i596_cmd));
+			dma_sync_dev(dev, prev, sizeof(struct i596_cmd));
 		}
 
 		if (lp->cmd_head != NULL)
 			ack_cmd |= CUC_START;
 		dma->scb.cmd = SWAP32(virt_to_dma(lp, &lp->cmd_head->status));
-		DMA_WBACK_INV(dev, &dma->scb, sizeof(struct i596_scb));
+		dma_sync_dev(dev, &dma->scb, sizeof(struct i596_scb));
 	}
 	if ((status & 0x1000) || (status & 0x4000)) {
 		if ((status & 0x4000))
@@ -1280,7 +1295,7 @@ static irqreturn_t i596_interrupt(int irq, void *dev_id)
 	}
 	wait_cmd(dev, dma, 100, "i596 interrupt, timeout");
 	dma->scb.command = SWAP16(ack_cmd);
-	DMA_WBACK(dev, &dma->scb, sizeof(struct i596_scb));
+	dma_sync_dev(dev, &dma->scb, sizeof(struct i596_scb));
 
 	/* DANGER: I suspect that some kind of interrupt
 	 acknowledgement aside from acking the 82596 might be needed
@@ -1311,7 +1326,7 @@ static int i596_close(struct net_device *dev)
 
 	wait_cmd(dev, lp->dma, 100, "close1 timed out");
 	lp->dma->scb.command = SWAP16(CUC_ABORT | RX_ABORT);
-	DMA_WBACK(dev, &lp->dma->scb, sizeof(struct i596_scb));
+	dma_sync_dev(dev, &lp->dma->scb, sizeof(struct i596_scb));
 
 	ca(dev);
 
@@ -1370,7 +1385,7 @@ static void set_multicast_list(struct net_device *dev)
 			       dev->name);
 		else {
 			dma->cf_cmd.cmd.command = SWAP16(CmdConfigure);
-			DMA_WBACK_INV(dev, &dma->cf_cmd, sizeof(struct cf_cmd));
+			dma_sync_dev(dev, &dma->cf_cmd, sizeof(struct cf_cmd));
 			i596_add_cmd(dev, &dma->cf_cmd.cmd);
 		}
 	}
@@ -1402,7 +1417,7 @@ static void set_multicast_list(struct net_device *dev)
 					   dev->name, cp));
 			cp += ETH_ALEN;
 		}
-		DMA_WBACK_INV(dev, &dma->mc_cmd, sizeof(struct mc_cmd));
+		dma_sync_dev(dev, &dma->mc_cmd, sizeof(struct mc_cmd));
 		i596_add_cmd(dev, &cmd->cmd);
 	}
 }
diff --git a/drivers/net/ethernet/i825xx/sni_82596.c b/drivers/net/ethernet/i825xx/sni_82596.c
index 2af7f77345fb..baa598988f47 100644
--- a/drivers/net/ethernet/i825xx/sni_82596.c
+++ b/drivers/net/ethernet/i825xx/sni_82596.c
@@ -1,3 +1,4 @@
+// SPDX-License-Identifier: GPL-2.0-only
 /*
  * sni_82596.c -- driver for intel 82596 ethernet controller, as
  *  		  used in older SNI RM machines
@@ -23,12 +24,6 @@
 
 static const char sni_82596_string[] = "snirm_82596";
 
-#define DMA_ALLOC                      dma_alloc_coherent
-#define DMA_FREE                       dma_free_coherent
-#define DMA_WBACK(priv, addr, len)     do { } while (0)
-#define DMA_INV(priv, addr, len)       do { } while (0)
-#define DMA_WBACK_INV(priv, addr, len) do { } while (0)
-
 #define SYSBUS      0x00004400
 
 /* big endian CPU, 82596 little endian */
@@ -83,6 +78,7 @@ static int sni_82596_probe(struct platform_device *dev)
 	void __iomem *mpu_addr;
 	void __iomem *ca_addr;
 	u8 __iomem *eth_addr;
+	u8 mac[ETH_ALEN];
 
 	res = platform_get_resource(dev, IORESOURCE_MEM, 0);
 	ca = platform_get_resource(dev, IORESOURCE_MEM, 1);
@@ -90,10 +86,10 @@ static int sni_82596_probe(struct platform_device *dev)
 	idprom = platform_get_resource(dev, IORESOURCE_MEM, 2);
 	if (!res || !ca || !options || !idprom)
 		return -ENODEV;
-	mpu_addr = ioremap_nocache(res->start, 4);
+	mpu_addr = ioremap(res->start, 4);
 	if (!mpu_addr)
 		return -ENOMEM;
-	ca_addr = ioremap_nocache(ca->start, 4);
+	ca_addr = ioremap(ca->start, 4);
 	if (!ca_addr)
 		goto probe_failed_free_mpu;
 
@@ -109,22 +105,24 @@ static int sni_82596_probe(struct platform_device *dev)
 	netdevice->base_addr = res->start;
 	netdevice->irq = platform_get_irq(dev, 0);
 
-	eth_addr = ioremap_nocache(idprom->start, 0x10);
+	eth_addr = ioremap(idprom->start, 0x10);
 	if (!eth_addr)
 		goto probe_failed;
 
 	/* someone seems to like messed up stuff */
-	netdevice->dev_addr[0] = readb(eth_addr + 0x0b);
-	netdevice->dev_addr[1] = readb(eth_addr + 0x0a);
-	netdevice->dev_addr[2] = readb(eth_addr + 0x09);
-	netdevice->dev_addr[3] = readb(eth_addr + 0x08);
-	netdevice->dev_addr[4] = readb(eth_addr + 0x07);
-	netdevice->dev_addr[5] = readb(eth_addr + 0x06);
+	mac[0] = readb(eth_addr + 0x0b);
+	mac[1] = readb(eth_addr + 0x0a);
+	mac[2] = readb(eth_addr + 0x09);
+	mac[3] = readb(eth_addr + 0x08);
+	mac[4] = readb(eth_addr + 0x07);
+	mac[5] = readb(eth_addr + 0x06);
+	eth_hw_addr_set(netdevice, mac);
 	iounmap(eth_addr);
 
-	if (!netdevice->irq) {
+	if (netdevice->irq < 0) {
 		printk(KERN_ERR "%s: IRQ not found for i82596 at 0x%lx\n",
 			__FILE__, netdevice->base_addr);
+		retval = netdevice->irq;
 		goto probe_failed;
 	}
 
@@ -133,10 +131,19 @@ static int sni_82596_probe(struct platform_device *dev)
 	lp->ca = ca_addr;
 	lp->mpu_port = mpu_addr;
 
+	lp->dma = dma_alloc_coherent(&dev->dev, sizeof(struct i596_dma),
+				     &lp->dma_addr, GFP_KERNEL);
+	if (!lp->dma)
+		goto probe_failed;
+
 	retval = i82596_probe(netdevice);
-	if (retval == 0)
-		return 0;
+	if (retval)
+		goto probe_failed_free_dma;
+	return 0;
 
+probe_failed_free_dma:
+	dma_free_coherent(&dev->dev, sizeof(struct i596_dma), lp->dma,
+			  lp->dma_addr);
 probe_failed:
 	free_netdev(netdevice);
 probe_failed_free_ca:
@@ -146,23 +153,22 @@ probe_failed_free_mpu:
 	return retval;
 }
 
-static int sni_82596_driver_remove(struct platform_device *pdev)
+static void sni_82596_driver_remove(struct platform_device *pdev)
 {
 	struct net_device *dev = platform_get_drvdata(pdev);
 	struct i596_private *lp = netdev_priv(dev);
 
 	unregister_netdev(dev);
-	DMA_FREE(dev->dev.parent, sizeof(struct i596_private),
-		 lp->dma, lp->dma_addr);
+	dma_free_coherent(&pdev->dev, sizeof(struct i596_private), lp->dma,
+			  lp->dma_addr);
 	iounmap(lp->ca);
 	iounmap(lp->mpu_port);
 	free_netdev (dev);
-	return 0;
 }
 
 static struct platform_driver sni_82596_driver = {
 	.probe	= sni_82596_probe,
-	.remove	= sni_82596_driver_remove,
+	.remove = sni_82596_driver_remove,
 	.driver	= {
 		.name	= sni_82596_string,
 	},
diff --git a/drivers/net/ethernet/i825xx/sun3_82586.c b/drivers/net/ethernet/i825xx/sun3_82586.c
index 8bb15a8c2a40..58a3d28d938c 100644
--- a/drivers/net/ethernet/i825xx/sun3_82586.c
+++ b/drivers/net/ethernet/i825xx/sun3_82586.c
@@ -1,3 +1,4 @@
+// SPDX-License-Identifier: GPL-2.0-only
 /*
  * Sun3 i82586 Ethernet driver
  *
@@ -29,6 +30,7 @@ static int rfdadd = 0; /* rfdadd=1 may be better for 8K MEM cards */
 static int fifo=0x8;	/* don't change */
 
 #include <linux/kernel.h>
+#include <linux/module.h>
 #include <linux/string.h>
 #include <linux/errno.h>
 #include <linux/ioport.h>
@@ -121,10 +123,11 @@ static int     sun3_82586_probe1(struct net_device *dev,int ioaddr);
 static irqreturn_t sun3_82586_interrupt(int irq,void *dev_id);
 static int     sun3_82586_open(struct net_device *dev);
 static int     sun3_82586_close(struct net_device *dev);
-static int     sun3_82586_send_packet(struct sk_buff *,struct net_device *);
+static netdev_tx_t     sun3_82586_send_packet(struct sk_buff *,
+					      struct net_device *);
 static struct  net_device_stats *sun3_82586_get_stats(struct net_device *dev);
 static void    set_multicast_list(struct net_device *dev);
-static void    sun3_82586_timeout(struct net_device *dev);
+static void    sun3_82586_timeout(struct net_device *dev, unsigned int txqueue);
 #if 0
 static void    sun3_82586_dump(struct net_device *,void *);
 #endif
@@ -275,7 +278,7 @@ static void alloc586(struct net_device *dev)
 	memset((char *)p->scb,0,sizeof(struct scb_struct));
 }
 
-struct net_device * __init sun3_82586_probe(int unit)
+static int __init sun3_82586_probe(void)
 {
 	struct net_device *dev;
 	unsigned long ioaddr;
@@ -290,25 +293,20 @@ struct net_device * __init sun3_82586_probe(int unit)
 		break;
 
 	default:
-		return ERR_PTR(-ENODEV);
+		return -ENODEV;
 	}
 
 	if (found)
-		return ERR_PTR(-ENODEV);
+		return -ENODEV;
 
 	ioaddr = (unsigned long)ioremap(IE_OBIO, SUN3_82586_TOTAL_SIZE);
 	if (!ioaddr)
-		return ERR_PTR(-ENOMEM);
+		return -ENOMEM;
 	found = 1;
 
 	dev = alloc_etherdev(sizeof(struct priv));
 	if (!dev)
 		goto out;
-	if (unit >= 0) {
-		sprintf(dev->name, "eth%d", unit);
-		netdev_boot_setup_check(dev);
-	}
-
 	dev->irq = IE_IRQ;
 	dev->base_addr = ioaddr;
 	err = sun3_82586_probe1(dev, ioaddr);
@@ -317,7 +315,7 @@ struct net_device * __init sun3_82586_probe(int unit)
 	err = register_netdev(dev);
 	if (err)
 		goto out2;
-	return dev;
+	return 0;
 
 out2:
 	release_region(ioaddr, SUN3_82586_TOTAL_SIZE);
@@ -325,8 +323,9 @@ out1:
 	free_netdev(dev);
 out:
 	iounmap((void __iomem *)ioaddr);
-	return ERR_PTR(err);
+	return err;
 }
+module_init(sun3_82586_probe);
 
 static const struct net_device_ops sun3_82586_netdev_ops = {
 	.ndo_open		= sun3_82586_open,
@@ -341,14 +340,13 @@ static const struct net_device_ops sun3_82586_netdev_ops = {
 
 static int __init sun3_82586_probe1(struct net_device *dev,int ioaddr)
 {
-	int i, size, retval;
+	int size, retval;
 
 	if (!request_region(ioaddr, SUN3_82586_TOTAL_SIZE, DRV_NAME))
 		return -EBUSY;
 
 	/* copy in the ethernet address from the prom */
-	for(i = 0; i < 6 ; i++)
-	     dev->dev_addr[i] = idprom->id_ethaddr[i];
+	eth_hw_addr_set(dev, idprom->id_ethaddr);
 
 	printk("%s: SUN3 Intel 82586 found at %lx, ",dev->name,dev->base_addr);
 
@@ -463,7 +461,7 @@ static int init586(struct net_device *dev)
 	ias_cmd->cmd_cmd	= swab16(CMD_IASETUP | CMD_LAST);
 	ias_cmd->cmd_link	= 0xffff;
 
-	memcpy((char *)&ias_cmd->iaddr,(char *) dev->dev_addr,ETH_ALEN);
+	memcpy((char *)&ias_cmd->iaddr,(const char *) dev->dev_addr,ETH_ALEN);
 
 	p->scb->cbl_offset = make16(ias_cmd);
 
@@ -964,7 +962,7 @@ static void startrecv586(struct net_device *dev)
 	WAIT_4_SCB_CMD_RUC();	/* wait for accept cmd. (no timeout!!) */
 }
 
-static void sun3_82586_timeout(struct net_device *dev)
+static void sun3_82586_timeout(struct net_device *dev, unsigned int txqueue)
 {
 	struct priv *p = netdev_priv(dev);
 #ifndef NO_NOPCOMMANDS
@@ -989,7 +987,7 @@ static void sun3_82586_timeout(struct net_device *dev)
 	{
 #ifdef DEBUG
 		printk("%s: xmitter timed out, try to restart! stat: %02x\n",dev->name,p->scb->cus);
-		printk("%s: command-stats: %04x %04x\n",dev->name,swab16(p->xmit_cmds[0]->cmd_status),swab16(p->xmit_cmds[1]->cmd_status));
+		printk("%s: command-stats: %04x\n", dev->name, swab16(p->xmit_cmds[0]->cmd_status));
 		printk("%s: check, whether you set the right interrupt number!\n",dev->name);
 #endif
 		sun3_82586_close(dev);
@@ -1002,7 +1000,8 @@ static void sun3_82586_timeout(struct net_device *dev)
  * send frame
  */
 
-static int sun3_82586_send_packet(struct sk_buff *skb, struct net_device *dev)
+static netdev_tx_t
+sun3_82586_send_packet(struct sk_buff *skb, struct net_device *dev)
 {
 	int len,i;
 #ifndef NO_NOPCOMMANDS
@@ -1013,6 +1012,7 @@ static int sun3_82586_send_packet(struct sk_buff *skb, struct net_device *dev)
 	if(skb->len > XMIT_BUFF_SIZE)
 	{
 		printk("%s: Sorry, max. framelength is %d bytes. The length of your frame is %d bytes.\n",dev->name,XMIT_BUFF_SIZE,skb->len);
+		dev_kfree_skb(skb);
 		return NETDEV_TX_OK;
 	}
 
diff --git a/drivers/net/ethernet/i825xx/sun3_82586.h b/drivers/net/ethernet/i825xx/sun3_82586.h
index 79aef681ac85..d8e249d704a7 100644
--- a/drivers/net/ethernet/i825xx/sun3_82586.h
+++ b/drivers/net/ethernet/i825xx/sun3_82586.h
@@ -1,3 +1,4 @@
+/* SPDX-License-Identifier: GPL-2.0-only */
 /*
  * Intel i82586 Ethernet definitions
  *
@@ -150,7 +151,7 @@ struct rfd_struct
 #define RFD_ERR_RNR  0x02     /* status: receiver out of resources */
 #define RFD_ERR_OVR  0x01     /* DMA Overrun! */
 
-#define RFD_ERR_FTS  0x0080	/* Frame to short */
+#define RFD_ERR_FTS  0x0080	/* Frame too short */
 #define RFD_ERR_NEOP 0x0040	/* No EOP flag (for bitstuffing only) */
 #define RFD_ERR_TRUN 0x0020	/* (82596 only/SF mode) indicates truncated frame */
 #define RFD_MATCHADD 0x0002     /* status: Destinationaddress !matches IA (only 82596) */
@@ -250,7 +251,7 @@ struct mcsetup_cmd_struct
   unsigned short cmd_cmd;
   unsigned short cmd_link;
   unsigned short mc_cnt;		/* number of bytes in the MC-List */
-  unsigned char  mc_list[0][6];  	/* pointer to 6 bytes entries */
+  unsigned char  mc_list[][6];  	/* pointer to 6 bytes entries */
 };
 
 /*
diff --git a/drivers/net/ethernet/ibm/Kconfig b/drivers/net/ethernet/ibm/Kconfig
index 37dceabf8861..4f4b23465c47 100644
--- a/drivers/net/ethernet/ibm/Kconfig
+++ b/drivers/net/ethernet/ibm/Kconfig
@@ -1,3 +1,4 @@
+# SPDX-License-Identifier: GPL-2.0-only
 #
 # IBM device configuration.
 #
@@ -6,7 +7,7 @@ config NET_VENDOR_IBM
 	bool "IBM devices"
 	default y
 	depends on PPC_PSERIES || PPC_DCR || (IBMEBUS && SPARSEMEM)
-	---help---
+	help
 	  If you have a network (Ethernet) card belonging to this class, say Y.
 
 	  Note that the answer to this question doesn't directly affect the
@@ -19,19 +20,32 @@ if NET_VENDOR_IBM
 config IBMVETH
 	tristate "IBM LAN Virtual Ethernet support"
 	depends on PPC_PSERIES
-	---help---
+	help
 	  This driver supports virtual ethernet adapters on newer IBM iSeries
 	  and pSeries systems.
 
 	  To compile this driver as a module, choose M here. The module will
 	  be called ibmveth.
 
+config IBMVETH_KUNIT_TEST
+	bool "KUnit test for IBM LAN Virtual Ethernet support" if !KUNIT_ALL_TESTS
+	depends on KUNIT
+	depends on KUNIT=y && IBMVETH=y
+	default KUNIT_ALL_TESTS
+	help
+	  This builds unit tests for the IBM LAN Virtual Ethernet driver.
+
+	  For more information on KUnit and unit tests in general, please refer
+	  to the KUnit documentation in Documentation/dev-tools/kunit/.
+
+	  If unsure, say N.
+
 source "drivers/net/ethernet/ibm/emac/Kconfig"
 
 config EHEA
 	tristate "eHEA Ethernet support"
 	depends on IBMEBUS && SPARSEMEM
-	---help---
+	help
 	  This driver supports the IBM pSeries eHEA ethernet adapter.
 
 	  To compile the driver as a module, choose M here. The module
@@ -40,7 +54,7 @@ config EHEA
 config IBMVNIC
 	tristate "IBM Virtual NIC support"
 	depends on PPC_PSERIES
-	---help---
+	help
 	  This driver supports Virtual NIC adapters on IBM i and IBM System p
 	  systems.
 
diff --git a/drivers/net/ethernet/ibm/Makefile b/drivers/net/ethernet/ibm/Makefile
index 447865c8b632..1d17d0c33d4d 100644
--- a/drivers/net/ethernet/ibm/Makefile
+++ b/drivers/net/ethernet/ibm/Makefile
@@ -1,3 +1,4 @@
+# SPDX-License-Identifier: GPL-2.0-only
 #
 # Makefile for th IBM network device drivers.
 #
diff --git a/drivers/net/ethernet/ibm/ehea/Makefile b/drivers/net/ethernet/ibm/ehea/Makefile
index cd473e295242..9e1e5c7aafe2 100644
--- a/drivers/net/ethernet/ibm/ehea/Makefile
+++ b/drivers/net/ethernet/ibm/ehea/Makefile
@@ -1,3 +1,4 @@
+# SPDX-License-Identifier: GPL-2.0-only
 #
 # Makefile for the eHEA ethernet device driver for IBM eServer System p
 #
diff --git a/drivers/net/ethernet/ibm/ehea/ehea.h b/drivers/net/ethernet/ibm/ehea/ehea.h
index 6be7b9839f35..208c440a602b 100644
--- a/drivers/net/ethernet/ibm/ehea/ehea.h
+++ b/drivers/net/ethernet/ibm/ehea/ehea.h
@@ -1,3 +1,4 @@
+/* SPDX-License-Identifier: GPL-2.0-or-later */
 /*
  *  linux/drivers/net/ethernet/ibm/ehea/ehea.h
  *
@@ -9,21 +10,6 @@
  *       Christoph Raisch <raisch@de.ibm.com>
  *       Jan-Bernd Themann <themann@de.ibm.com>
  *       Thomas Klein <tklein@de.ibm.com>
- *
- *
- * This program is free software; you can redistribute it and/or modify
- * it under the terms of the GNU General Public License as published by
- * the Free Software Foundation; either version 2, or (at your option)
- * any later version.
- *
- * This program is distributed in the hope that it will be useful,
- * but WITHOUT ANY WARRANTY; without even the implied warranty of
- * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.	 See the
- * GNU General Public License for more details.
- *
- * You should have received a copy of the GNU General Public License
- * along with this program; if not, write to the Free Software
- * Foundation, Inc., 675 Mass Ave, Cambridge, MA 02139, USA.
  */
 
 #ifndef __EHEA_H__
@@ -33,6 +19,7 @@
 #include <linux/ethtool.h>
 #include <linux/vmalloc.h>
 #include <linux/if_vlan.h>
+#include <linux/platform_device.h>
 
 #include <asm/ibmebus.h>
 #include <asm/io.h>
diff --git a/drivers/net/ethernet/ibm/ehea/ehea_ethtool.c b/drivers/net/ethernet/ibm/ehea/ehea_ethtool.c
index 4f58d338d739..1db5b6790a41 100644
--- a/drivers/net/ethernet/ibm/ehea/ehea_ethtool.c
+++ b/drivers/net/ethernet/ibm/ehea/ehea_ethtool.c
@@ -1,3 +1,4 @@
+// SPDX-License-Identifier: GPL-2.0-or-later
 /*
  *  linux/drivers/net/ethernet/ibm/ehea/ehea_ethtool.c
  *
@@ -9,21 +10,6 @@
  *       Christoph Raisch <raisch@de.ibm.com>
  *       Jan-Bernd Themann <themann@de.ibm.com>
  *       Thomas Klein <tklein@de.ibm.com>
- *
- *
- * This program is free software; you can redistribute it and/or modify
- * it under the terms of the GNU General Public License as published by
- * the Free Software Foundation; either version 2, or (at your option)
- * any later version.
- *
- * This program is distributed in the hope that it will be useful,
- * but WITHOUT ANY WARRANTY; without even the implied warranty of
- * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.	 See the
- * GNU General Public License for more details.
- *
- * You should have received a copy of the GNU General Public License
- * along with this program; if not, write to the Free Software
- * Foundation, Inc., 675 Mass Ave, Cambridge, MA 02139, USA.
  */
 
 #define pr_fmt(fmt) KBUILD_MODNAME ": " fmt
@@ -173,8 +159,8 @@ static int ehea_nway_reset(struct net_device *dev)
 static void ehea_get_drvinfo(struct net_device *dev,
 			       struct ethtool_drvinfo *info)
 {
-	strlcpy(info->driver, DRV_NAME, sizeof(info->driver));
-	strlcpy(info->version, DRV_VERSION, sizeof(info->version));
+	strscpy(info->driver, DRV_NAME, sizeof(info->driver));
+	strscpy(info->version, DRV_VERSION, sizeof(info->version));
 }
 
 static u32 ehea_get_msglevel(struct net_device *dev)
diff --git a/drivers/net/ethernet/ibm/ehea/ehea_hw.h b/drivers/net/ethernet/ibm/ehea/ehea_hw.h
index 180d4128a711..590933a45d65 100644
--- a/drivers/net/ethernet/ibm/ehea/ehea_hw.h
+++ b/drivers/net/ethernet/ibm/ehea/ehea_hw.h
@@ -1,3 +1,4 @@
+/* SPDX-License-Identifier: GPL-2.0-or-later */
 /*
  *  linux/drivers/net/ethernet/ibm/ehea/ehea_hw.h
  *
@@ -9,21 +10,6 @@
  *       Christoph Raisch <raisch@de.ibm.com>
  *       Jan-Bernd Themann <themann@de.ibm.com>
  *       Thomas Klein <tklein@de.ibm.com>
- *
- *
- * This program is free software; you can redistribute it and/or modify
- * it under the terms of the GNU General Public License as published by
- * the Free Software Foundation; either version 2, or (at your option)
- * any later version.
- *
- * This program is distributed in the hope that it will be useful,
- * but WITHOUT ANY WARRANTY; without even the implied warranty of
- * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.	 See the
- * GNU General Public License for more details.
- *
- * You should have received a copy of the GNU General Public License
- * along with this program; if not, write to the Free Software
- * Foundation, Inc., 675 Mass Ave, Cambridge, MA 02139, USA.
  */
 
 #ifndef __EHEA_HW_H__
diff --git a/drivers/net/ethernet/ibm/ehea/ehea_main.c b/drivers/net/ethernet/ibm/ehea/ehea_main.c
index 1e53d7a82675..9b006bc353a1 100644
--- a/drivers/net/ethernet/ibm/ehea/ehea_main.c
+++ b/drivers/net/ethernet/ibm/ehea/ehea_main.c
@@ -1,3 +1,4 @@
+// SPDX-License-Identifier: GPL-2.0-or-later
 /*
  *  linux/drivers/net/ethernet/ibm/ehea/ehea_main.c
  *
@@ -9,21 +10,6 @@
  *	 Christoph Raisch <raisch@de.ibm.com>
  *	 Jan-Bernd Themann <themann@de.ibm.com>
  *	 Thomas Klein <tklein@de.ibm.com>
- *
- *
- * This program is free software; you can redistribute it and/or modify
- * it under the terms of the GNU General Public License as published by
- * the Free Software Foundation; either version 2, or (at your option)
- * any later version.
- *
- * This program is distributed in the hope that it will be useful,
- * but WITHOUT ANY WARRANTY; without even the implied warranty of
- * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.	 See the
- * GNU General Public License for more details.
- *
- * You should have received a copy of the GNU General Public License
- * along with this program; if not, write to the Free Software
- * Foundation, Inc., 675 Mass Ave, Cambridge, MA 02139, USA.
  */
 
 #define pr_fmt(fmt) KBUILD_MODNAME ": " fmt
@@ -43,6 +29,9 @@
 #include <asm/kexec.h>
 #include <linux/mutex.h>
 #include <linux/prefetch.h>
+#include <linux/of.h>
+#include <linux/of_device.h>
+#include <linux/platform_device.h>
 
 #include <net/ip.h>
 
@@ -101,7 +90,7 @@ static struct ehea_bcmc_reg_array ehea_bcmc_regs;
 
 static int ehea_probe_adapter(struct platform_device *dev);
 
-static int ehea_remove(struct platform_device *dev);
+static void ehea_remove(struct platform_device *dev);
 
 static const struct of_device_id ehea_module_device_table[] = {
 	{
@@ -123,6 +112,7 @@ static const struct of_device_id ehea_device_table[] = {
 	},
 	{},
 };
+MODULE_DEVICE_TABLE(of, ehea_device_table);
 
 static struct platform_driver ehea_driver = {
 	.driver = {
@@ -778,12 +768,11 @@ static void check_sqs(struct ehea_port *port)
 {
 	struct ehea_swqe *swqe;
 	int swqe_index;
-	int i, k;
+	int i;
 
 	for (i = 0; i < port->num_def_qps; i++) {
 		struct ehea_port_res *pr = &port->port_res[i];
 		int ret;
-		k = 0;
 		swqe = ehea_get_swqe(pr->qp, &swqe_index);
 		memset(swqe, 0, SWQE_HEADER_SIZE);
 		atomic_dec(&pr->swqe_avail);
@@ -911,7 +900,7 @@ static int ehea_poll(struct napi_struct *napi, int budget)
 		if (!cqe && !cqe_skb)
 			return rx;
 
-		if (!napi_reschedule(napi))
+		if (!napi_schedule(napi))
 			return rx;
 
 		cqe_skb = ehea_proc_cqes(pr, EHEA_POLL_MAX_CQES);
@@ -921,17 +910,6 @@ static int ehea_poll(struct napi_struct *napi, int budget)
 	return rx;
 }
 
-#ifdef CONFIG_NET_POLL_CONTROLLER
-static void ehea_netpoll(struct net_device *dev)
-{
-	struct ehea_port *port = netdev_priv(dev);
-	int i;
-
-	for (i = 0; i < port->num_def_qps; i++)
-		napi_schedule(&port->port_res[i].napi);
-}
-#endif
-
 static irqreturn_t ehea_recv_irq_handler(int irq, void *param)
 {
 	struct ehea_port_res *pr = param;
@@ -1238,9 +1216,9 @@ static void ehea_parse_eqe(struct ehea_adapter *adapter, u64 eqe)
 	}
 }
 
-static void ehea_neq_tasklet(unsigned long data)
+static void ehea_neq_tasklet(struct tasklet_struct *t)
 {
-	struct ehea_adapter *adapter = (struct ehea_adapter *)data;
+	struct ehea_adapter *adapter = from_tasklet(adapter, t, neq_tasklet);
 	struct ehea_eqe *eqe;
 	u64 event_mask;
 
@@ -1475,7 +1453,7 @@ static int ehea_init_port_res(struct ehea_port *port, struct ehea_port_res *pr,
 
 	memset(pr, 0, sizeof(struct ehea_port_res));
 
-	pr->tx_bytes = rx_bytes;
+	pr->tx_bytes = tx_bytes;
 	pr->tx_packets = tx_packets;
 	pr->rx_bytes = rx_bytes;
 	pr->rx_packets = rx_packets;
@@ -1569,7 +1547,7 @@ static int ehea_init_port_res(struct ehea_port *port, struct ehea_port_res *pr,
 
 	kfree(init_attr);
 
-	netif_napi_add(pr->port->netdev, &pr->napi, ehea_poll, 64);
+	netif_napi_add(pr->port->netdev, &pr->napi, ehea_poll);
 
 	ret = 0;
 	goto out;
@@ -1603,20 +1581,16 @@ static int ehea_clean_portres(struct ehea_port *port, struct ehea_port_res *pr)
 		ehea_destroy_eq(pr->eq);
 
 		for (i = 0; i < pr->rq1_skba.len; i++)
-			if (pr->rq1_skba.arr[i])
-				dev_kfree_skb(pr->rq1_skba.arr[i]);
+			dev_kfree_skb(pr->rq1_skba.arr[i]);
 
 		for (i = 0; i < pr->rq2_skba.len; i++)
-			if (pr->rq2_skba.arr[i])
-				dev_kfree_skb(pr->rq2_skba.arr[i]);
+			dev_kfree_skb(pr->rq2_skba.arr[i]);
 
 		for (i = 0; i < pr->rq3_skba.len; i++)
-			if (pr->rq3_skba.arr[i])
-				dev_kfree_skb(pr->rq3_skba.arr[i]);
+			dev_kfree_skb(pr->rq3_skba.arr[i]);
 
 		for (i = 0; i < pr->sq_skba.len; i++)
-			if (pr->sq_skba.arr[i])
-				dev_kfree_skb(pr->sq_skba.arr[i]);
+			dev_kfree_skb(pr->sq_skba.arr[i]);
 
 		vfree(pr->rq1_skba.arr);
 		vfree(pr->rq2_skba.arr);
@@ -1644,7 +1618,7 @@ static void write_swqe2_immediate(struct sk_buff *skb, struct ehea_swqe *swqe,
 		 * For TSO packets we only copy the headers into the
 		 * immediate area.
 		 */
-		immediate_len = ETH_HLEN + ip_hdrlen(skb) + tcp_hdrlen(skb);
+		immediate_len = skb_tcp_all_headers(skb);
 	}
 
 	if (skb_is_gso(skb) || skb_data_size >= SWQE2_MAX_IMM) {
@@ -1770,7 +1744,7 @@ static int ehea_set_mac_addr(struct net_device *dev, void *sa)
 		goto out_free;
 	}
 
-	memcpy(dev->dev_addr, mac_addr->sa_data, dev->addr_len);
+	eth_hw_addr_set(dev, mac_addr->sa_data);
 
 	/* Deregister old MAC in pHYP */
 	if (port->state == EHEA_PORT_UP) {
@@ -2038,7 +2012,7 @@ static void ehea_xmit3(struct sk_buff *skb, struct net_device *dev,
 	dev_consume_skb_any(skb);
 }
 
-static int ehea_start_xmit(struct sk_buff *skb, struct net_device *dev)
+static netdev_tx_t ehea_start_xmit(struct sk_buff *skb, struct net_device *dev)
 {
 	struct ehea_port *port = netdev_priv(dev);
 	struct ehea_swqe *swqe;
@@ -2647,10 +2621,8 @@ static int ehea_restart_qps(struct net_device *dev)
 	u16 dummy16 = 0;
 
 	cb0 = (void *)get_zeroed_page(GFP_KERNEL);
-	if (!cb0) {
-		ret = -ENOMEM;
-		goto out;
-	}
+	if (!cb0)
+		return -ENOMEM;
 
 	for (i = 0; i < (port->num_def_qps); i++) {
 		struct ehea_port_res *pr =  &port->port_res[i];
@@ -2670,6 +2642,7 @@ static int ehea_restart_qps(struct net_device *dev)
 					    cb0);
 		if (hret != H_SUCCESS) {
 			netdev_err(dev, "query_ehea_qp failed (1)\n");
+			ret = -EFAULT;
 			goto out;
 		}
 
@@ -2682,6 +2655,7 @@ static int ehea_restart_qps(struct net_device *dev)
 					     &dummy64, &dummy16, &dummy16);
 		if (hret != H_SUCCESS) {
 			netdev_err(dev, "modify_ehea_qp failed (1)\n");
+			ret = -EFAULT;
 			goto out;
 		}
 
@@ -2690,6 +2664,7 @@ static int ehea_restart_qps(struct net_device *dev)
 					    cb0);
 		if (hret != H_SUCCESS) {
 			netdev_err(dev, "query_ehea_qp failed (2)\n");
+			ret = -EFAULT;
 			goto out;
 		}
 
@@ -2816,7 +2791,7 @@ out:
 	return;
 }
 
-static void ehea_tx_watchdog(struct net_device *dev)
+static void ehea_tx_watchdog(struct net_device *dev, unsigned int txqueue)
 {
 	struct ehea_port *port = netdev_priv(dev);
 
@@ -2896,15 +2871,14 @@ out:
 	return ret;
 }
 
-static ssize_t ehea_show_port_id(struct device *dev,
-				 struct device_attribute *attr, char *buf)
+static ssize_t log_port_id_show(struct device *dev,
+				struct device_attribute *attr, char *buf)
 {
 	struct ehea_port *port = container_of(dev, struct ehea_port, ofdev.dev);
 	return sprintf(buf, "%d", port->logical_port_id);
 }
 
-static DEVICE_ATTR(log_port_id, S_IRUSR | S_IRGRP | S_IROTH, ehea_show_port_id,
-		   NULL);
+static DEVICE_ATTR_RO(log_port_id);
 
 static void logical_port_release(struct device *dev)
 {
@@ -2927,6 +2901,7 @@ static struct device *ehea_register_port(struct ehea_port *port,
 	ret = of_device_register(&port->ofdev);
 	if (ret) {
 		pr_err("failed to register device. ret=%d\n", ret);
+		put_device(&port->ofdev.dev);
 		goto out;
 	}
 
@@ -2954,9 +2929,6 @@ static const struct net_device_ops ehea_netdev_ops = {
 	.ndo_open		= ehea_open,
 	.ndo_stop		= ehea_stop,
 	.ndo_start_xmit		= ehea_start_xmit,
-#ifdef CONFIG_NET_POLL_CONTROLLER
-	.ndo_poll_controller	= ehea_netpoll,
-#endif
 	.ndo_get_stats64	= ehea_get_stats64,
 	.ndo_set_mac_address	= ehea_set_mac_addr,
 	.ndo_validate_addr	= eth_validate_addr,
@@ -3018,7 +2990,7 @@ static struct ehea_port *ehea_setup_single_port(struct ehea_adapter *adapter,
 	SET_NETDEV_DEV(dev, port_dev);
 
 	/* initialize net_device structure */
-	memcpy(dev->dev_addr, &port->mac_addr, ETH_ALEN);
+	eth_hw_addr_set(dev, (u8 *)&port->mac_addr);
 
 	dev->netdev_ops = &ehea_netdev_ops;
 	ehea_set_ethtool_ops(dev);
@@ -3091,19 +3063,17 @@ static void ehea_shutdown_single_port(struct ehea_port *port)
 static int ehea_setup_ports(struct ehea_adapter *adapter)
 {
 	struct device_node *lhea_dn;
-	struct device_node *eth_dn = NULL;
+	struct device_node *eth_dn;
 
 	const u32 *dn_log_port_id;
 	int i = 0;
 
 	lhea_dn = adapter->ofdev->dev.of_node;
-	while ((eth_dn = of_get_next_child(lhea_dn, eth_dn))) {
-
+	for_each_child_of_node(lhea_dn, eth_dn) {
 		dn_log_port_id = of_get_property(eth_dn, "ibm,hea-port-no",
 						 NULL);
 		if (!dn_log_port_id) {
-			pr_err("bad device node: eth_dn name=%s\n",
-			       eth_dn->full_name);
+			pr_err("bad device node: eth_dn name=%pOF\n", eth_dn);
 			continue;
 		}
 
@@ -3131,12 +3101,11 @@ static struct device_node *ehea_get_eth_dn(struct ehea_adapter *adapter,
 					   u32 logical_port_id)
 {
 	struct device_node *lhea_dn;
-	struct device_node *eth_dn = NULL;
+	struct device_node *eth_dn;
 	const u32 *dn_log_port_id;
 
 	lhea_dn = adapter->ofdev->dev.of_node;
-	while ((eth_dn = of_get_next_child(lhea_dn, eth_dn))) {
-
+	for_each_child_of_node(lhea_dn, eth_dn) {
 		dn_log_port_id = of_get_property(eth_dn, "ibm,hea-port-no",
 						 NULL);
 		if (dn_log_port_id)
@@ -3147,7 +3116,7 @@ static struct device_node *ehea_get_eth_dn(struct ehea_adapter *adapter,
 	return NULL;
 }
 
-static ssize_t ehea_probe_port(struct device *dev,
+static ssize_t probe_port_store(struct device *dev,
 			       struct device_attribute *attr,
 			       const char *buf, size_t count)
 {
@@ -3177,6 +3146,7 @@ static ssize_t ehea_probe_port(struct device *dev,
 
 	if (ehea_add_adapter_mr(adapter)) {
 		pr_err("creating MR failed\n");
+		of_node_put(eth_dn);
 		return -EIO;
 	}
 
@@ -3201,9 +3171,9 @@ static ssize_t ehea_probe_port(struct device *dev,
 	return (ssize_t) count;
 }
 
-static ssize_t ehea_remove_port(struct device *dev,
-				struct device_attribute *attr,
-				const char *buf, size_t count)
+static ssize_t remove_port_store(struct device *dev,
+				 struct device_attribute *attr,
+				 const char *buf, size_t count)
 {
 	struct ehea_adapter *adapter = dev_get_drvdata(dev);
 	struct ehea_port *port;
@@ -3236,8 +3206,8 @@ static ssize_t ehea_remove_port(struct device *dev,
 	return (ssize_t) count;
 }
 
-static DEVICE_ATTR(probe_port, S_IWUSR, NULL, ehea_probe_port);
-static DEVICE_ATTR(remove_port, S_IWUSR, NULL, ehea_remove_port);
+static DEVICE_ATTR_WO(probe_port);
+static DEVICE_ATTR_WO(remove_port);
 
 static int ehea_create_device_sysfs(struct platform_device *dev)
 {
@@ -3281,7 +3251,7 @@ static int ehea_mem_notifier(struct notifier_block *nb,
 	switch (action) {
 	case MEM_CANCEL_OFFLINE:
 		pr_info("memory offlining canceled");
-		/* Fall through: re-add canceled memory block */
+		fallthrough;	/* re-add canceled memory block */
 
 	case MEM_ONLINE:
 		pr_info("memory is going online");
@@ -3425,7 +3395,7 @@ static int ehea_probe_adapter(struct platform_device *dev)
 
 	if (!adapter->handle) {
 		dev_err(&dev->dev, "failed getting handle for adapter"
-			" '%s'\n", dev->dev.of_node->full_name);
+			" '%pOF'\n", dev->dev.of_node);
 		ret = -ENODEV;
 		goto out_free_ad;
 	}
@@ -3451,8 +3421,7 @@ static int ehea_probe_adapter(struct platform_device *dev)
 		goto out_free_ad;
 	}
 
-	tasklet_init(&adapter->neq_tasklet, ehea_neq_tasklet,
-		     (unsigned long)adapter);
+	tasklet_setup(&adapter->neq_tasklet, ehea_neq_tasklet);
 
 	ret = ehea_create_device_sysfs(dev);
 	if (ret)
@@ -3500,7 +3469,7 @@ out:
 	return ret;
 }
 
-static int ehea_remove(struct platform_device *dev)
+static void ehea_remove(struct platform_device *dev)
 {
 	struct ehea_adapter *adapter = platform_get_drvdata(dev);
 	int i;
@@ -3521,8 +3490,6 @@ static int ehea_remove(struct platform_device *dev)
 	list_del(&adapter->list);
 
 	ehea_update_firmware_handles();
-
-	return 0;
 }
 
 static int check_module_parm(void)
@@ -3553,14 +3520,12 @@ static int check_module_parm(void)
 	return ret;
 }
 
-static ssize_t ehea_show_capabilities(struct device_driver *drv,
-				      char *buf)
+static ssize_t capabilities_show(struct device_driver *drv, char *buf)
 {
 	return sprintf(buf, "%d", EHEA_CAPABILITIES);
 }
 
-static DRIVER_ATTR(capabilities, S_IRUSR | S_IRGRP | S_IROTH,
-		   ehea_show_capabilities, NULL);
+static DRIVER_ATTR_RO(capabilities);
 
 static int __init ehea_module_init(void)
 {
diff --git a/drivers/net/ethernet/ibm/ehea/ehea_phyp.c b/drivers/net/ethernet/ibm/ehea/ehea_phyp.c
index d3a130ccdcc8..e63716e139f5 100644
--- a/drivers/net/ethernet/ibm/ehea/ehea_phyp.c
+++ b/drivers/net/ethernet/ibm/ehea/ehea_phyp.c
@@ -1,3 +1,4 @@
+// SPDX-License-Identifier: GPL-2.0-or-later
 /*
  *  linux/drivers/net/ethernet/ibm/ehea/ehea_phyp.c
  *
@@ -9,21 +10,6 @@
  *	 Christoph Raisch <raisch@de.ibm.com>
  *	 Jan-Bernd Themann <themann@de.ibm.com>
  *	 Thomas Klein <tklein@de.ibm.com>
- *
- *
- * This program is free software; you can redistribute it and/or modify
- * it under the terms of the GNU General Public License as published by
- * the Free Software Foundation; either version 2, or (at your option)
- * any later version.
- *
- * This program is distributed in the hope that it will be useful,
- * but WITHOUT ANY WARRANTY; without even the implied warranty of
- * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.	 See the
- * GNU General Public License for more details.
- *
- * You should have received a copy of the GNU General Public License
- * along with this program; if not, write to the Free Software
- * Foundation, Inc., 675 Mass Ave, Cambridge, MA 02139, USA.
  */
 
 #define pr_fmt(fmt) KBUILD_MODNAME ": " fmt
diff --git a/drivers/net/ethernet/ibm/ehea/ehea_phyp.h b/drivers/net/ethernet/ibm/ehea/ehea_phyp.h
index 99b6c2a38dbf..e8b56c103410 100644
--- a/drivers/net/ethernet/ibm/ehea/ehea_phyp.h
+++ b/drivers/net/ethernet/ibm/ehea/ehea_phyp.h
@@ -1,3 +1,4 @@
+/* SPDX-License-Identifier: GPL-2.0-or-later */
 /*
  *  linux/drivers/net/ethernet/ibm/ehea/ehea_phyp.h
  *
@@ -9,21 +10,6 @@
  *       Christoph Raisch <raisch@de.ibm.com>
  *       Jan-Bernd Themann <themann@de.ibm.com>
  *       Thomas Klein <tklein@de.ibm.com>
- *
- *
- * This program is free software; you can redistribute it and/or modify
- * it under the terms of the GNU General Public License as published by
- * the Free Software Foundation; either version 2, or (at your option)
- * any later version.
- *
- * This program is distributed in the hope that it will be useful,
- * but WITHOUT ANY WARRANTY; without even the implied warranty of
- * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.	 See the
- * GNU General Public License for more details.
- *
- * You should have received a copy of the GNU General Public License
- * along with this program; if not, write to the Free Software
- * Foundation, Inc., 675 Mass Ave, Cambridge, MA 02139, USA.
  */
 
 #ifndef __EHEA_PHYP_H__
diff --git a/drivers/net/ethernet/ibm/ehea/ehea_qmr.c b/drivers/net/ethernet/ibm/ehea/ehea_qmr.c
index a0820f72b25c..db45373ea31c 100644
--- a/drivers/net/ethernet/ibm/ehea/ehea_qmr.c
+++ b/drivers/net/ethernet/ibm/ehea/ehea_qmr.c
@@ -1,3 +1,4 @@
+// SPDX-License-Identifier: GPL-2.0-or-later
 /*
  *  linux/drivers/net/ethernet/ibm/ehea/ehea_qmr.c
  *
@@ -9,21 +10,6 @@
  *       Christoph Raisch <raisch@de.ibm.com>
  *       Jan-Bernd Themann <themann@de.ibm.com>
  *       Thomas Klein <tklein@de.ibm.com>
- *
- *
- * This program is free software; you can redistribute it and/or modify
- * it under the terms of the GNU General Public License as published by
- * the Free Software Foundation; either version 2, or (at your option)
- * any later version.
- *
- * This program is distributed in the hope that it will be useful,
- * but WITHOUT ANY WARRANTY; without even the implied warranty of
- * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.	 See the
- * GNU General Public License for more details.
- *
- * You should have received a copy of the GNU General Public License
- * along with this program; if not, write to the Free Software
- * Foundation, Inc., 675 Mass Ave, Cambridge, MA 02139, USA.
  */
 
 #define pr_fmt(fmt) KBUILD_MODNAME ": " fmt
@@ -123,9 +109,8 @@ struct ehea_cq *ehea_create_cq(struct ehea_adapter *adapter,
 			       int nr_of_cqe, u64 eq_handle, u32 cq_token)
 {
 	struct ehea_cq *cq;
-	struct h_epa epa;
-	u64 *cq_handle_ref, hret, rpage;
-	u32 act_nr_of_entries, act_pages, counter;
+	u64 hret, rpage;
+	u32 counter;
 	int ret;
 	void *vpage;
 
@@ -139,10 +124,6 @@ struct ehea_cq *ehea_create_cq(struct ehea_adapter *adapter,
 
 	cq->adapter = adapter;
 
-	cq_handle_ref = &cq->fw_handle;
-	act_nr_of_entries = 0;
-	act_pages = 0;
-
 	hret = ehea_h_alloc_resource_cq(adapter->handle, &cq->attr,
 					&cq->fw_handle, &cq->epas);
 	if (hret != H_SUCCESS) {
@@ -190,7 +171,6 @@ struct ehea_cq *ehea_create_cq(struct ehea_adapter *adapter,
 	}
 
 	hw_qeit_reset(&cq->hw_queue);
-	epa = cq->epas.kernel;
 	ehea_reset_cq_ep(cq);
 	ehea_reset_cq_n1(cq);
 
@@ -690,13 +670,10 @@ int ehea_rem_sect_bmap(unsigned long pfn, unsigned long nr_pages)
 
 static int ehea_is_hugepage(unsigned long pfn)
 {
-	int page_order;
-
 	if (pfn & EHEA_HUGEPAGE_PFN_MASK)
 		return 0;
 
-	page_order = compound_order(pfn_to_page(pfn));
-	if (page_order + PAGE_SHIFT != EHEA_HUGEPAGESHIFT)
+	if (page_shift(pfn_to_page(pfn)) != EHEA_HUGEPAGESHIFT)
 		return 0;
 
 	return 1;
diff --git a/drivers/net/ethernet/ibm/ehea/ehea_qmr.h b/drivers/net/ethernet/ibm/ehea/ehea_qmr.h
index 8e4a70c20ab7..7c7cccd820f7 100644
--- a/drivers/net/ethernet/ibm/ehea/ehea_qmr.h
+++ b/drivers/net/ethernet/ibm/ehea/ehea_qmr.h
@@ -1,3 +1,4 @@
+/* SPDX-License-Identifier: GPL-2.0-or-later */
 /*
  *  linux/drivers/net/ethernet/ibm/ehea/ehea_qmr.h
  *
@@ -9,21 +10,6 @@
  *       Christoph Raisch <raisch@de.ibm.com>
  *       Jan-Bernd Themann <themann@de.ibm.com>
  *       Thomas Klein <tklein@de.ibm.com>
- *
- *
- * This program is free software; you can redistribute it and/or modify
- * it under the terms of the GNU General Public License as published by
- * the Free Software Foundation; either version 2, or (at your option)
- * any later version.
- *
- * This program is distributed in the hope that it will be useful,
- * but WITHOUT ANY WARRANTY; without even the implied warranty of
- * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.	 See the
- * GNU General Public License for more details.
- *
- * You should have received a copy of the GNU General Public License
- * along with this program; if not, write to the Free Software
- * Foundation, Inc., 675 Mass Ave, Cambridge, MA 02139, USA.
  */
 
 #ifndef __EHEA_QMR_H__
diff --git a/drivers/net/ethernet/ibm/emac/Kconfig b/drivers/net/ethernet/ibm/emac/Kconfig
index 90d49191beb3..c8e5de5987ac 100644
--- a/drivers/net/ethernet/ibm/emac/Kconfig
+++ b/drivers/net/ethernet/ibm/emac/Kconfig
@@ -1,3 +1,4 @@
+# SPDX-License-Identifier: GPL-2.0-only
 config IBM_EMAC
 	tristate "IBM EMAC Ethernet support"
 	depends on PPC_DCR
@@ -28,18 +29,6 @@ config IBM_EMAC_RX_COPY_THRESHOLD
 	depends on IBM_EMAC
 	default "256"
 
-config IBM_EMAC_RX_SKB_HEADROOM
-	int "Additional RX skb headroom (bytes)"
-	depends on IBM_EMAC
-	default "0"
-	help
-	  Additional receive skb headroom. Note, that driver
-	  will always reserve at least 2 bytes to make IP header
-	  aligned, so usually there is no need to add any additional
-	  headroom.
-
-	  If unsure, set to 0.
-
 config IBM_EMAC_DEBUG
 	bool "Debugging"
 	depends on IBM_EMAC
diff --git a/drivers/net/ethernet/ibm/emac/Makefile b/drivers/net/ethernet/ibm/emac/Makefile
index 98768ba0955a..ddf1ce3c8cca 100644
--- a/drivers/net/ethernet/ibm/emac/Makefile
+++ b/drivers/net/ethernet/ibm/emac/Makefile
@@ -1,3 +1,4 @@
+# SPDX-License-Identifier: GPL-2.0
 #
 # Makefile for the PowerPC 4xx on-chip ethernet driver
 #
diff --git a/drivers/net/ethernet/ibm/emac/core.c b/drivers/net/ethernet/ibm/emac/core.c
index 508923f39ccf..417dfa18daae 100644
--- a/drivers/net/ethernet/ibm/emac/core.c
+++ b/drivers/net/ethernet/ibm/emac/core.c
@@ -1,3 +1,4 @@
+// SPDX-License-Identifier: GPL-2.0-or-later
 /*
  * drivers/net/ethernet/ibm/emac/core.c
  *
@@ -16,12 +17,6 @@
  *	(c) 2003 Benjamin Herrenschmidt <benh@kernel.crashing.org>
  *      Armin Kuster <akuster@mvista.com>
  * 	Johnnie Peters <jpeters@mvista.com>
- *
- * This program is free software; you can redistribute  it and/or modify it
- * under  the terms of  the GNU General  Public License as published by the
- * Free Software Foundation;  either version 2 of the  License, or (at your
- * option) any later version.
- *
  */
 
 #include <linux/module.h>
@@ -37,12 +32,13 @@
 #include <linux/ethtool.h>
 #include <linux/mii.h>
 #include <linux/bitops.h>
-#include <linux/workqueue.h>
 #include <linux/of.h>
 #include <linux/of_address.h>
 #include <linux/of_irq.h>
 #include <linux/of_net.h>
 #include <linux/of_mdio.h>
+#include <linux/of_platform.h>
+#include <linux/platform_device.h>
 #include <linux/slab.h>
 
 #include <asm/processor.h>
@@ -99,11 +95,6 @@ MODULE_LICENSE("GPL");
 static u32 busy_phy_map;
 static DEFINE_MUTEX(emac_phy_map_lock);
 
-/* This is the wait queue used to wait on any event related to probe, that
- * is discovery of MALs, other EMACs, ZMII/RGMIIs, etc...
- */
-static DECLARE_WAIT_QUEUE_HEAD(emac_probe_wait);
-
 /* Having stable interface names is a doomed idea. However, it would be nice
  * if we didn't have completely random interface names at boot too :-) It's
  * just a matter of making everybody's life easier. Since we are doing
@@ -119,9 +110,6 @@ static DECLARE_WAIT_QUEUE_HEAD(emac_probe_wait);
 #define EMAC_BOOT_LIST_SIZE	4
 static struct device_node *emac_boot_list[EMAC_BOOT_LIST_SIZE];
 
-/* How long should I wait for dependent devices ? */
-#define EMAC_PROBE_DEP_TIMEOUT	(HZ * 5)
-
 /* I don't want to litter system log with timeout errors
  * when we have brain-damaged PHY.
  */
@@ -133,8 +121,7 @@ static inline void emac_report_timeout_error(struct emac_instance *dev,
 				  EMAC_FTR_440EP_PHY_CLK_FIX))
 		DBG(dev, "%s" NL, error);
 	else if (net_ratelimit())
-		printk(KERN_ERR "%s: %s\n", dev->ofdev->dev.of_node->full_name,
-			error);
+		printk(KERN_ERR "%pOF: %s\n", dev->ofdev->dev.of_node, error);
 }
 
 /* EMAC PHY clock workaround:
@@ -200,18 +187,18 @@ static void __emac_set_multicast_list(struct emac_instance *dev);
 
 static inline int emac_phy_supports_gige(int phy_mode)
 {
-	return  phy_mode == PHY_MODE_GMII ||
-		phy_mode == PHY_MODE_RGMII ||
-		phy_mode == PHY_MODE_SGMII ||
-		phy_mode == PHY_MODE_TBI ||
-		phy_mode == PHY_MODE_RTBI;
+	return  phy_interface_mode_is_rgmii(phy_mode) ||
+		phy_mode == PHY_INTERFACE_MODE_GMII ||
+		phy_mode == PHY_INTERFACE_MODE_SGMII ||
+		phy_mode == PHY_INTERFACE_MODE_TBI ||
+		phy_mode == PHY_INTERFACE_MODE_RTBI;
 }
 
 static inline int emac_phy_gpcs(int phy_mode)
 {
-	return  phy_mode == PHY_MODE_SGMII ||
-		phy_mode == PHY_MODE_TBI ||
-		phy_mode == PHY_MODE_RTBI;
+	return  phy_mode == PHY_INTERFACE_MODE_SGMII ||
+		phy_mode == PHY_INTERFACE_MODE_TBI ||
+		phy_mode == PHY_INTERFACE_MODE_RTBI;
 }
 
 static inline void emac_tx_enable(struct emac_instance *dev)
@@ -343,6 +330,7 @@ static int emac_reset(struct emac_instance *dev)
 {
 	struct emac_regs __iomem *p = dev->emacp;
 	int n = 20;
+	bool __maybe_unused try_internal_clock = false;
 
 	DBG(dev, "reset" NL);
 
@@ -355,6 +343,7 @@ static int emac_reset(struct emac_instance *dev)
 	}
 
 #ifdef CONFIG_PPC_DCR_NATIVE
+do_retry:
 	/*
 	 * PPC460EX/GT Embedded Processor Advanced User's Manual
 	 * section 28.10.1 Mode Register 0 (EMACx_MR0) states:
@@ -362,10 +351,19 @@ static int emac_reset(struct emac_instance *dev)
 	 * of the EMAC. If none is present, select the internal clock
 	 * (SDR0_ETH_CFG[EMACx_PHY_CLK] = 1).
 	 * After a soft reset, select the external clock.
+	 *
+	 * The AR8035-A PHY Meraki MR24 does not provide a TX Clk if the
+	 * ethernet cable is not attached. This causes the reset to timeout
+	 * and the PHY detection code in emac_init_phy() is unable to
+	 * communicate and detect the AR8035-A PHY. As a result, the emac
+	 * driver bails out early and the user has no ethernet.
+	 * In order to stay compatible with existing configurations, the
+	 * driver will temporarily switch to the internal clock, after
+	 * the first reset fails.
 	 */
 	if (emac_has_feature(dev, EMAC_FTR_460EX_PHY_CLK_FIX)) {
-		if (dev->phy_address == 0xffffffff &&
-		    dev->phy_map == 0xffffffff) {
+		if (try_internal_clock || (dev->phy_address == 0xffffffff &&
+					   dev->phy_map == 0xffffffff)) {
 			/* No PHY: select internal loop clock before reset */
 			dcri_clrset(SDR0, SDR0_ETH_CFG,
 				    0, SDR0_ETH_CFG_ECS << dev->cell_index);
@@ -383,8 +381,15 @@ static int emac_reset(struct emac_instance *dev)
 
 #ifdef CONFIG_PPC_DCR_NATIVE
 	if (emac_has_feature(dev, EMAC_FTR_460EX_PHY_CLK_FIX)) {
-		if (dev->phy_address == 0xffffffff &&
-		    dev->phy_map == 0xffffffff) {
+		if (!n && !try_internal_clock) {
+			/* first attempt has timed out. */
+			n = 20;
+			try_internal_clock = true;
+			goto do_retry;
+		}
+
+		if (try_internal_clock || (dev->phy_address == 0xffffffff &&
+					   dev->phy_map == 0xffffffff)) {
 			/* No PHY: restore external clock source after reset */
 			dcri_clrset(SDR0, SDR0_ETH_CFG,
 				    SDR0_ETH_CFG_ECS << dev->cell_index, 0);
@@ -404,9 +409,9 @@ static int emac_reset(struct emac_instance *dev)
 
 static void emac_hash_mc(struct emac_instance *dev)
 {
+	u32 __iomem *gaht_base = emac_gaht_base(dev);
 	const int regs = EMAC_XAHT_REGS(dev);
-	u32 *gaht_base = emac_gaht_base(dev);
-	u32 gaht_temp[regs];
+	u32 gaht_temp[EMAC_XAHT_MAX_REGS];
 	struct netdev_hw_addr *ha;
 	int i;
 
@@ -499,6 +504,9 @@ static u32 __emac4_calc_base_mr1(struct emac_instance *dev, int tx_size, int rx_
 	case 16384:
 		ret |= EMAC4_MR1_TFS_16K;
 		break;
+	case 8192:
+		ret |= EMAC4_MR1_TFS_8K;
+		break;
 	case 4096:
 		ret |= EMAC4_MR1_TFS_4K;
 		break;
@@ -514,6 +522,9 @@ static u32 __emac4_calc_base_mr1(struct emac_instance *dev, int tx_size, int rx_
 	case 16384:
 		ret |= EMAC4_MR1_RFS_16K;
 		break;
+	case 8192:
+		ret |= EMAC4_MR1_RFS_8K;
+		break;
 	case 4096:
 		ret |= EMAC4_MR1_RFS_4K;
 		break;
@@ -758,7 +769,7 @@ static void emac_reset_work(struct work_struct *work)
 	mutex_unlock(&dev->link_lock);
 }
 
-static void emac_tx_timeout(struct net_device *ndev)
+static void emac_tx_timeout(struct net_device *ndev, unsigned int txqueue)
 {
 	struct emac_instance *dev = netdev_priv(ndev);
 
@@ -854,7 +865,7 @@ static void __emac_mdio_write(struct emac_instance *dev, u8 id, u8 reg,
 {
 	struct emac_regs __iomem *p = dev->emacp;
 	u32 r = 0;
-	int n, err = -ETIMEDOUT;
+	int n;
 
 	mutex_lock(&dev->mdio_lock);
 
@@ -901,7 +912,6 @@ static void __emac_mdio_write(struct emac_instance *dev, u8 id, u8 reg,
 			goto bail;
 		}
 	}
-	err = 0;
  bail:
 	if (emac_has_feature(dev, EMAC_FTR_HAS_RGMII))
 		rgmii_put_mdio(dev->rgmii_dev, dev->rgmii_port);
@@ -954,8 +964,6 @@ static void __emac_set_multicast_list(struct emac_instance *dev)
 	 * we need is just to stop RX channel. This seems to work on all
 	 * tested SoCs.                                                --ebs
 	 *
-	 * If we need the full reset, we might just trigger the workqueue
-	 * and do it async... a bit nasty but should work --BenH
 	 */
 	dev->mcast_pending = 0;
 	emac_rx_disable(dev);
@@ -995,7 +1003,7 @@ static int emac_set_mac_address(struct net_device *ndev, void *sa)
 
 	mutex_lock(&dev->link_lock);
 
-	memcpy(ndev->dev_addr, addr->sa_data, ndev->addr_len);
+	eth_hw_addr_set(ndev, addr->sa_data);
 
 	emac_rx_disable(dev);
 	emac_tx_disable(dev);
@@ -1048,7 +1056,9 @@ static int emac_resize_rx_ring(struct emac_instance *dev, int new_mtu)
 
 	/* Second pass, allocate new skbs */
 	for (i = 0; i < NUM_RX_BUFF; ++i) {
-		struct sk_buff *skb = alloc_skb(rx_skb_size, GFP_ATOMIC);
+		struct sk_buff *skb;
+
+		skb = netdev_alloc_skb_ip_align(dev->ndev, rx_skb_size);
 		if (!skb) {
 			ret = -ENOMEM;
 			goto oom;
@@ -1057,10 +1067,10 @@ static int emac_resize_rx_ring(struct emac_instance *dev, int new_mtu)
 		BUG_ON(!dev->rx_skb[i]);
 		dev_kfree_skb(dev->rx_skb[i]);
 
-		skb_reserve(skb, EMAC_RX_SKB_HEADROOM + 2);
 		dev->rx_desc[i].data_ptr =
-		    dma_map_single(&dev->ofdev->dev, skb->data - 2, rx_sync_size,
-				   DMA_FROM_DEVICE) + 2;
+		    dma_map_single(&dev->ofdev->dev, skb->data - NET_IP_ALIGN,
+				   rx_sync_size, DMA_FROM_DEVICE)
+				   + NET_IP_ALIGN;
 		dev->rx_skb[i] = skb;
 	}
  skip:
@@ -1077,7 +1087,7 @@ static int emac_resize_rx_ring(struct emac_instance *dev, int new_mtu)
 		/* This is to prevent starting RX channel in emac_rx_enable() */
 		set_bit(MAL_COMMAC_RX_STOPPED, &dev->commac.flags);
 
-		dev->ndev->mtu = new_mtu;
+		WRITE_ONCE(dev->ndev->mtu, new_mtu);
 		emac_full_tx_reset(dev);
 	}
 
@@ -1109,7 +1119,7 @@ static int emac_change_mtu(struct net_device *ndev, int new_mtu)
 	}
 
 	if (!ret) {
-		ndev->mtu = new_mtu;
+		WRITE_ONCE(ndev->mtu, new_mtu);
 		dev->rx_skb_size = emac_rx_skb_size(new_mtu);
 		dev->rx_sync_size = emac_rx_sync_size(new_mtu);
 	}
@@ -1151,20 +1161,18 @@ static void emac_clean_rx_ring(struct emac_instance *dev)
 	}
 }
 
-static inline int emac_alloc_rx_skb(struct emac_instance *dev, int slot,
-				    gfp_t flags)
+static int
+__emac_prepare_rx_skb(struct sk_buff *skb, struct emac_instance *dev, int slot)
 {
-	struct sk_buff *skb = alloc_skb(dev->rx_skb_size, flags);
 	if (unlikely(!skb))
 		return -ENOMEM;
 
 	dev->rx_skb[slot] = skb;
 	dev->rx_desc[slot].data_len = 0;
 
-	skb_reserve(skb, EMAC_RX_SKB_HEADROOM + 2);
 	dev->rx_desc[slot].data_ptr =
-	    dma_map_single(&dev->ofdev->dev, skb->data - 2, dev->rx_sync_size,
-			   DMA_FROM_DEVICE) + 2;
+	    dma_map_single(&dev->ofdev->dev, skb->data - NET_IP_ALIGN,
+			   dev->rx_sync_size, DMA_FROM_DEVICE) + NET_IP_ALIGN;
 	wmb();
 	dev->rx_desc[slot].ctrl = MAL_RX_CTRL_EMPTY |
 	    (slot == (NUM_RX_BUFF - 1) ? MAL_RX_CTRL_WRAP : 0);
@@ -1172,6 +1180,27 @@ static inline int emac_alloc_rx_skb(struct emac_instance *dev, int slot,
 	return 0;
 }
 
+static int
+emac_alloc_rx_skb(struct emac_instance *dev, int slot)
+{
+	struct sk_buff *skb;
+
+	skb = __netdev_alloc_skb_ip_align(dev->ndev, dev->rx_skb_size,
+					  GFP_KERNEL);
+
+	return __emac_prepare_rx_skb(skb, dev, slot);
+}
+
+static int
+emac_alloc_rx_skb_napi(struct emac_instance *dev, int slot)
+{
+	struct sk_buff *skb;
+
+	skb = napi_alloc_skb(&dev->mal->napi, dev->rx_skb_size);
+
+	return __emac_prepare_rx_skb(skb, dev, slot);
+}
+
 static void emac_print_link_status(struct emac_instance *dev)
 {
 	if (netif_carrier_ok(dev->ndev))
@@ -1188,21 +1217,13 @@ static void emac_print_link_status(struct emac_instance *dev)
 static int emac_open(struct net_device *ndev)
 {
 	struct emac_instance *dev = netdev_priv(ndev);
-	int err, i;
+	int i;
 
 	DBG(dev, "open" NL);
 
-	/* Setup error IRQ handler */
-	err = request_irq(dev->emac_irq, emac_irq, 0, "EMAC", dev);
-	if (err) {
-		printk(KERN_ERR "%s: failed to request IRQ %d\n",
-		       ndev->name, dev->emac_irq);
-		return err;
-	}
-
 	/* Allocate RX ring */
 	for (i = 0; i < NUM_RX_BUFF; ++i)
-		if (emac_alloc_rx_skb(dev, i, GFP_KERNEL)) {
+		if (emac_alloc_rx_skb(dev, i)) {
 			printk(KERN_ERR "%s: failed to allocate RX ring\n",
 			       ndev->name);
 			goto oom;
@@ -1253,8 +1274,6 @@ static int emac_open(struct net_device *ndev)
 	return 0;
  oom:
 	emac_clean_rx_ring(dev);
-	free_irq(dev->emac_irq, dev);
-
 	return -ENOMEM;
 }
 
@@ -1368,8 +1387,6 @@ static int emac_close(struct net_device *ndev)
 	emac_clean_tx_ring(dev);
 	emac_clean_rx_ring(dev);
 
-	free_irq(dev->emac_irq, dev);
-
 	netif_carrier_off(ndev);
 
 	return 0;
@@ -1386,7 +1403,7 @@ static inline u16 emac_tx_csum(struct emac_instance *dev,
 	return 0;
 }
 
-static inline int emac_xmit_finish(struct emac_instance *dev, int len)
+static inline netdev_tx_t emac_xmit_finish(struct emac_instance *dev, int len)
 {
 	struct emac_regs __iomem *p = dev->emacp;
 	struct net_device *ndev = dev->ndev;
@@ -1413,7 +1430,7 @@ static inline int emac_xmit_finish(struct emac_instance *dev, int len)
 }
 
 /* Tx lock BH */
-static int emac_start_xmit(struct sk_buff *skb, struct net_device *ndev)
+static netdev_tx_t emac_start_xmit(struct sk_buff *skb, struct net_device *ndev)
 {
 	struct emac_instance *dev = netdev_priv(ndev);
 	unsigned int len = skb->len;
@@ -1471,7 +1488,8 @@ static inline int emac_xmit_split(struct emac_instance *dev, int slot,
 }
 
 /* Tx lock BH disabled (SG version for TAH equipped EMACs) */
-static int emac_start_xmit_sg(struct sk_buff *skb, struct net_device *ndev)
+static netdev_tx_t
+emac_start_xmit_sg(struct sk_buff *skb, struct net_device *ndev)
 {
 	struct emac_instance *dev = netdev_priv(ndev);
 	int nr_frags = skb_shinfo(skb)->nr_frags;
@@ -1509,7 +1527,7 @@ static int emac_start_xmit_sg(struct sk_buff *skb, struct net_device *ndev)
 				       ctrl);
 	/* skb fragments */
 	for (i = 0; i < nr_frags; ++i) {
-		struct skb_frag_struct *frag = &skb_shinfo(skb)->frags[i];
+		skb_frag_t *frag = &skb_shinfo(skb)->frags[i];
 		len = skb_frag_size(frag);
 
 		if (unlikely(dev->tx_cnt + mal_tx_chunks(len) >= NUM_TX_BUFF))
@@ -1636,8 +1654,9 @@ static inline void emac_recycle_rx_skb(struct emac_instance *dev, int slot,
 	DBG2(dev, "recycle %d %d" NL, slot, len);
 
 	if (len)
-		dma_map_single(&dev->ofdev->dev, skb->data - 2,
-			       EMAC_DMA_ALIGN(len + 2), DMA_FROM_DEVICE);
+		dma_map_single(&dev->ofdev->dev, skb->data - NET_IP_ALIGN,
+			       SKB_DATA_ALIGN(len + NET_IP_ALIGN),
+			       DMA_FROM_DEVICE);
 
 	dev->rx_desc[slot].data_len = 0;
 	wmb();
@@ -1689,7 +1708,7 @@ static inline int emac_rx_sg_append(struct emac_instance *dev, int slot)
 		int len = dev->rx_desc[slot].data_len;
 		int tot_len = dev->rx_sg_skb->len + len;
 
-		if (unlikely(tot_len + 2 > dev->rx_skb_size)) {
+		if (unlikely(tot_len + NET_IP_ALIGN > dev->rx_skb_size)) {
 			++dev->estats.rx_dropped_mtu;
 			dev_kfree_skb(dev->rx_sg_skb);
 			dev->rx_sg_skb = NULL;
@@ -1708,6 +1727,7 @@ static inline int emac_rx_sg_append(struct emac_instance *dev, int slot)
 /* NAPI poll context */
 static int emac_poll_rx(void *param, int budget)
 {
+	LIST_HEAD(rx_list);
 	struct emac_instance *dev = param;
 	int slot = dev->rx_slot, received = 0;
 
@@ -1745,16 +1765,18 @@ static int emac_poll_rx(void *param, int budget)
 		}
 
 		if (len && len < EMAC_RX_COPY_THRESH) {
-			struct sk_buff *copy_skb =
-			    alloc_skb(len + EMAC_RX_SKB_HEADROOM + 2, GFP_ATOMIC);
+			struct sk_buff *copy_skb;
+
+			copy_skb = napi_alloc_skb(&dev->mal->napi, len);
 			if (unlikely(!copy_skb))
 				goto oom;
 
-			skb_reserve(copy_skb, EMAC_RX_SKB_HEADROOM + 2);
-			memcpy(copy_skb->data - 2, skb->data - 2, len + 2);
+			memcpy(copy_skb->data - NET_IP_ALIGN,
+			       skb->data - NET_IP_ALIGN,
+			       len + NET_IP_ALIGN);
 			emac_recycle_rx_skb(dev, slot, len);
 			skb = copy_skb;
-		} else if (unlikely(emac_alloc_rx_skb(dev, slot, GFP_ATOMIC)))
+		} else if (unlikely(emac_alloc_rx_skb_napi(dev, slot)))
 			goto oom;
 
 		skb_put(skb, len);
@@ -1762,8 +1784,7 @@ static int emac_poll_rx(void *param, int budget)
 		skb->protocol = eth_type_trans(skb, dev->ndev);
 		emac_rx_csum(dev, skb, ctrl);
 
-		if (unlikely(netif_receive_skb(skb) == NET_RX_DROP))
-			++dev->estats.rx_dropped_stack;
+		list_add_tail(&skb->list, &rx_list);
 	next:
 		++dev->stats.rx_packets;
 	skip:
@@ -1775,7 +1796,7 @@ static int emac_poll_rx(void *param, int budget)
 	sg:
 		if (ctrl & MAL_RX_CTRL_FIRST) {
 			BUG_ON(dev->rx_sg_skb);
-			if (unlikely(emac_alloc_rx_skb(dev, slot, GFP_ATOMIC))) {
+			if (unlikely(emac_alloc_rx_skb_napi(dev, slot))) {
 				DBG(dev, "rx OOM %d" NL, slot);
 				++dev->estats.rx_dropped_oom;
 				emac_recycle_rx_skb(dev, slot, 0);
@@ -1807,6 +1828,8 @@ static int emac_poll_rx(void *param, int budget)
 		goto next;
 	}
 
+	netif_receive_skb_list(&rx_list);
+
 	if (received) {
 		DBG2(dev, "rx %d BDs" NL, received);
 		dev->rx_slot = slot;
@@ -2094,8 +2117,11 @@ emac_ethtool_set_link_ksettings(struct net_device *ndev,
 	return 0;
 }
 
-static void emac_ethtool_get_ringparam(struct net_device *ndev,
-				       struct ethtool_ringparam *rp)
+static void
+emac_ethtool_get_ringparam(struct net_device *ndev,
+			   struct ethtool_ringparam *rp,
+			   struct kernel_ethtool_ringparam *kernel_rp,
+			   struct netlink_ext_ack *extack)
 {
 	rp->rx_max_pending = rp->rx_pending = NUM_RX_BUFF;
 	rp->tx_max_pending = rp->tx_pending = NUM_TX_BUFF;
@@ -2238,10 +2264,10 @@ static void emac_ethtool_get_drvinfo(struct net_device *ndev,
 {
 	struct emac_instance *dev = netdev_priv(ndev);
 
-	strlcpy(info->driver, "ibm_emac", sizeof(info->driver));
-	strlcpy(info->version, DRV_VERSION, sizeof(info->version));
-	snprintf(info->bus_info, sizeof(info->bus_info), "PPC 4xx EMAC-%d %s",
-		 dev->cell_index, dev->ofdev->dev.of_node->full_name);
+	strscpy(info->driver, "ibm_emac", sizeof(info->driver));
+	strscpy(info->version, DRV_VERSION, sizeof(info->version));
+	snprintf(info->bus_info, sizeof(info->bus_info), "PPC 4xx EMAC-%d %pOF",
+		 dev->cell_index, dev->ofdev->dev.of_node);
 }
 
 static const struct ethtool_ops emac_ethtool_ops = {
@@ -2277,7 +2303,7 @@ static int emac_ioctl(struct net_device *ndev, struct ifreq *rq, int cmd)
 	switch (cmd) {
 	case SIOCGMIIPHY:
 		data->phy_id = dev->phy.address;
-		/* Fall through */
+		fallthrough;
 	case SIOCGMIIREG:
 		data->val_out = emac_mdio_read(ndev, dev->phy.address,
 					       data->reg_num);
@@ -2343,31 +2369,20 @@ static int emac_check_deps(struct emac_instance *dev,
 		if (deps[i].drvdata != NULL)
 			there++;
 	}
-	return there == EMAC_DEP_COUNT;
+	if (there != EMAC_DEP_COUNT)
+		return -EPROBE_DEFER;
+	return 0;
 }
 
 static void emac_put_deps(struct emac_instance *dev)
 {
-	of_dev_put(dev->mal_dev);
-	of_dev_put(dev->zmii_dev);
-	of_dev_put(dev->rgmii_dev);
-	of_dev_put(dev->mdio_dev);
-	of_dev_put(dev->tah_dev);
+	platform_device_put(dev->mal_dev);
+	platform_device_put(dev->zmii_dev);
+	platform_device_put(dev->rgmii_dev);
+	platform_device_put(dev->mdio_dev);
+	platform_device_put(dev->tah_dev);
 }
 
-static int emac_of_bus_notify(struct notifier_block *nb, unsigned long action,
-			      void *data)
-{
-	/* We are only intereted in device addition */
-	if (action == BUS_NOTIFY_BOUND_DRIVER)
-		wake_up_all(&emac_probe_wait);
-	return 0;
-}
-
-static struct notifier_block emac_of_bus_notifier = {
-	.notifier_call = emac_of_bus_notify
-};
-
 static int emac_wait_deps(struct emac_instance *dev)
 {
 	struct emac_depentry deps[EMAC_DEP_COUNT];
@@ -2384,54 +2399,49 @@ static int emac_wait_deps(struct emac_instance *dev)
 		deps[EMAC_DEP_MDIO_IDX].phandle = dev->mdio_ph;
 	if (dev->blist && dev->blist > emac_boot_list)
 		deps[EMAC_DEP_PREV_IDX].phandle = 0xffffffffu;
-	bus_register_notifier(&platform_bus_type, &emac_of_bus_notifier);
-	wait_event_timeout(emac_probe_wait,
-			   emac_check_deps(dev, deps),
-			   EMAC_PROBE_DEP_TIMEOUT);
-	bus_unregister_notifier(&platform_bus_type, &emac_of_bus_notifier);
-	err = emac_check_deps(dev, deps) ? 0 : -ENODEV;
+	err = emac_check_deps(dev, deps);
 	for (i = 0; i < EMAC_DEP_COUNT; i++) {
 		of_node_put(deps[i].node);
 		if (err)
-			of_dev_put(deps[i].ofdev);
+			platform_device_put(deps[i].ofdev);
 	}
-	if (err == 0) {
+	if (!err) {
 		dev->mal_dev = deps[EMAC_DEP_MAL_IDX].ofdev;
 		dev->zmii_dev = deps[EMAC_DEP_ZMII_IDX].ofdev;
 		dev->rgmii_dev = deps[EMAC_DEP_RGMII_IDX].ofdev;
 		dev->tah_dev = deps[EMAC_DEP_TAH_IDX].ofdev;
 		dev->mdio_dev = deps[EMAC_DEP_MDIO_IDX].ofdev;
 	}
-	of_dev_put(deps[EMAC_DEP_PREV_IDX].ofdev);
+	platform_device_put(deps[EMAC_DEP_PREV_IDX].ofdev);
 	return err;
 }
 
 static int emac_read_uint_prop(struct device_node *np, const char *name,
 			       u32 *val, int fatal)
 {
-	int len;
-	const u32 *prop = of_get_property(np, name, &len);
-	if (prop == NULL || len < sizeof(u32)) {
+	int err;
+
+	err = of_property_read_u32(np, name, val);
+	if (err) {
 		if (fatal)
-			printk(KERN_ERR "%s: missing %s property\n",
-			       np->full_name, name);
-		return -ENODEV;
+			pr_err("%pOF: missing %s property", np, name);
+		return err;
 	}
-	*val = *prop;
 	return 0;
 }
 
 static void emac_adjust_link(struct net_device *ndev)
 {
 	struct emac_instance *dev = netdev_priv(ndev);
-	struct phy_device *phy = dev->phy_dev;
+	struct phy_device *phy = ndev->phydev;
 
 	dev->phy.autoneg = phy->autoneg;
 	dev->phy.speed = phy->speed;
 	dev->phy.duplex = phy->duplex;
 	dev->phy.pause = phy->pause;
 	dev->phy.asym_pause = phy->asym_pause;
-	dev->phy.advertising = phy->advertising;
+	ethtool_convert_link_mode_to_legacy_u32(&dev->phy.advertising,
+						phy->advertising);
 }
 
 static int emac_mii_bus_read(struct mii_bus *bus, int addr, int regnum)
@@ -2460,34 +2470,34 @@ static int emac_mii_bus_reset(struct mii_bus *bus)
 	return emac_reset(dev);
 }
 
+static int emac_mdio_phy_start_aneg(struct mii_phy *phy,
+				    struct phy_device *phy_dev)
+{
+	phy_dev->autoneg = phy->autoneg;
+	phy_dev->speed = phy->speed;
+	phy_dev->duplex = phy->duplex;
+	ethtool_convert_legacy_u32_to_link_mode(phy_dev->advertising,
+						phy->advertising);
+	return phy_start_aneg(phy_dev);
+}
+
 static int emac_mdio_setup_aneg(struct mii_phy *phy, u32 advertise)
 {
 	struct net_device *ndev = phy->dev;
-	struct emac_instance *dev = netdev_priv(ndev);
 
-	dev->phy.autoneg = AUTONEG_ENABLE;
-	dev->phy.speed = SPEED_1000;
-	dev->phy.duplex = DUPLEX_FULL;
-	dev->phy.advertising = advertise;
 	phy->autoneg = AUTONEG_ENABLE;
-	phy->speed = dev->phy.speed;
-	phy->duplex = dev->phy.duplex;
 	phy->advertising = advertise;
-	return phy_start_aneg(dev->phy_dev);
+	return emac_mdio_phy_start_aneg(phy, ndev->phydev);
 }
 
 static int emac_mdio_setup_forced(struct mii_phy *phy, int speed, int fd)
 {
 	struct net_device *ndev = phy->dev;
-	struct emac_instance *dev = netdev_priv(ndev);
 
-	dev->phy.autoneg =  AUTONEG_DISABLE;
-	dev->phy.speed = speed;
-	dev->phy.duplex = fd;
 	phy->autoneg = AUTONEG_DISABLE;
 	phy->speed = speed;
 	phy->duplex = fd;
-	return phy_start_aneg(dev->phy_dev);
+	return emac_mdio_phy_start_aneg(phy, ndev->phydev);
 }
 
 static int emac_mdio_poll_link(struct mii_phy *phy)
@@ -2496,46 +2506,38 @@ static int emac_mdio_poll_link(struct mii_phy *phy)
 	struct emac_instance *dev = netdev_priv(ndev);
 	int res;
 
-	res = phy_read_status(dev->phy_dev);
+	res = phy_read_status(ndev->phydev);
 	if (res) {
 		dev_err(&dev->ofdev->dev, "link update failed (%d).", res);
 		return ethtool_op_get_link(ndev);
 	}
 
-	return dev->phy_dev->link;
+	return ndev->phydev->link;
 }
 
 static int emac_mdio_read_link(struct mii_phy *phy)
 {
 	struct net_device *ndev = phy->dev;
-	struct emac_instance *dev = netdev_priv(ndev);
+	struct phy_device *phy_dev = ndev->phydev;
 	int res;
 
-	res = phy_read_status(dev->phy_dev);
+	res = phy_read_status(phy_dev);
 	if (res)
 		return res;
 
-	dev->phy.speed = phy->speed;
-	dev->phy.duplex = phy->duplex;
-	dev->phy.pause = phy->pause;
-	dev->phy.asym_pause = phy->asym_pause;
+	phy->speed = phy_dev->speed;
+	phy->duplex = phy_dev->duplex;
+	phy->pause = phy_dev->pause;
+	phy->asym_pause = phy_dev->asym_pause;
 	return 0;
 }
 
 static int emac_mdio_init_phy(struct mii_phy *phy)
 {
 	struct net_device *ndev = phy->dev;
-	struct emac_instance *dev = netdev_priv(ndev);
-
-	phy_start(dev->phy_dev);
-	dev->phy.autoneg = phy->autoneg;
-	dev->phy.speed = phy->speed;
-	dev->phy.duplex = phy->duplex;
-	dev->phy.advertising = phy->advertising;
-	dev->phy.pause = phy->pause;
-	dev->phy.asym_pause = phy->asym_pause;
 
-	return phy_init_hw(dev->phy_dev);
+	phy_start(ndev->phydev);
+	return phy_init_hw(ndev->phydev);
 }
 
 static const struct mii_phy_ops emac_dt_mdio_phy_ops = {
@@ -2549,36 +2551,32 @@ static const struct mii_phy_ops emac_dt_mdio_phy_ops = {
 static int emac_dt_mdio_probe(struct emac_instance *dev)
 {
 	struct device_node *mii_np;
+	struct mii_bus *bus;
 	int res;
 
-	mii_np = of_get_child_by_name(dev->ofdev->dev.of_node, "mdio");
+	mii_np = of_get_available_child_by_name(dev->ofdev->dev.of_node, "mdio");
 	if (!mii_np) {
 		dev_err(&dev->ofdev->dev, "no mdio definition found.");
 		return -ENODEV;
 	}
 
-	if (!of_device_is_available(mii_np)) {
-		res = -ENODEV;
-		goto put_node;
-	}
-
-	dev->mii_bus = devm_mdiobus_alloc(&dev->ofdev->dev);
-	if (!dev->mii_bus) {
+	bus = devm_mdiobus_alloc(&dev->ofdev->dev);
+	if (!bus) {
 		res = -ENOMEM;
 		goto put_node;
 	}
 
-	dev->mii_bus->priv = dev->ndev;
-	dev->mii_bus->parent = dev->ndev->dev.parent;
-	dev->mii_bus->name = "emac_mdio";
-	dev->mii_bus->read = &emac_mii_bus_read;
-	dev->mii_bus->write = &emac_mii_bus_write;
-	dev->mii_bus->reset = &emac_mii_bus_reset;
-	snprintf(dev->mii_bus->id, MII_BUS_ID_SIZE, "%s", dev->ofdev->name);
-	res = of_mdiobus_register(dev->mii_bus, mii_np);
+	bus->priv = dev->ndev;
+	bus->parent = dev->ndev->dev.parent;
+	bus->name = "emac_mdio";
+	bus->read = &emac_mii_bus_read;
+	bus->write = &emac_mii_bus_write;
+	bus->reset = &emac_mii_bus_reset;
+	snprintf(bus->id, MII_BUS_ID_SIZE, "%s", dev->ofdev->name);
+	res = devm_of_mdiobus_register(&dev->ofdev->dev, bus, mii_np);
 	if (res) {
 		dev_err(&dev->ofdev->dev, "cannot register MDIO bus %s (%d)",
-			dev->mii_bus->name, res);
+			bus->name, res);
 	}
 
  put_node:
@@ -2589,25 +2587,28 @@ static int emac_dt_mdio_probe(struct emac_instance *dev)
 static int emac_dt_phy_connect(struct emac_instance *dev,
 			       struct device_node *phy_handle)
 {
+	struct phy_device *phy_dev;
+
 	dev->phy.def = devm_kzalloc(&dev->ofdev->dev, sizeof(*dev->phy.def),
 				    GFP_KERNEL);
 	if (!dev->phy.def)
 		return -ENOMEM;
 
-	dev->phy_dev = of_phy_connect(dev->ndev, phy_handle, &emac_adjust_link,
-				      0, dev->phy_mode);
-	if (!dev->phy_dev) {
+	phy_dev = of_phy_connect(dev->ndev, phy_handle, &emac_adjust_link, 0,
+				 dev->phy_mode);
+	if (!phy_dev) {
 		dev_err(&dev->ofdev->dev, "failed to connect to PHY.\n");
 		return -ENODEV;
 	}
 
-	dev->phy.def->phy_id = dev->phy_dev->drv->phy_id;
-	dev->phy.def->phy_id_mask = dev->phy_dev->drv->phy_id_mask;
-	dev->phy.def->name = dev->phy_dev->drv->name;
+	dev->phy.def->phy_id = phy_dev->drv->phy_id;
+	dev->phy.def->phy_id_mask = phy_dev->drv->phy_id_mask;
+	dev->phy.def->name = phy_dev->drv->name;
 	dev->phy.def->ops = &emac_dt_mdio_phy_ops;
-	dev->phy.features = dev->phy_dev->supported;
-	dev->phy.address = dev->phy_dev->mdio.addr;
-	dev->phy.mode = dev->phy_dev->interface;
+	ethtool_convert_link_mode_to_legacy_u32(&dev->phy.features,
+						phy_dev->supported);
+	dev->phy.address = phy_dev->mdio.addr;
+	dev->phy.mode = phy_dev->interface;
 	return 0;
 }
 
@@ -2623,8 +2624,6 @@ static int emac_dt_phy_probe(struct emac_instance *dev)
 		res = emac_dt_mdio_probe(dev);
 		if (!res) {
 			res = emac_dt_phy_connect(dev, phy_handle);
-			if (res)
-				mdiobus_unregister(dev->mii_bus);
 		}
 	}
 
@@ -2659,12 +2658,15 @@ static int emac_init_phy(struct emac_instance *dev)
 		if (of_phy_is_fixed_link(np)) {
 			int res = emac_dt_mdio_probe(dev);
 
-			if (!res) {
-				res = of_phy_register_fixed_link(np);
-				if (res)
-					mdiobus_unregister(dev->mii_bus);
-			}
-			return res;
+			if (res)
+				return res;
+
+			res = of_phy_register_fixed_link(np);
+			ndev->phydev = of_phy_find_device(np);
+			if (res || !ndev->phydev)
+				return res ? res : -EINVAL;
+			emac_adjust_link(dev->ndev);
+			put_device(&ndev->phydev->mdio.dev);
 		}
 		return 0;
 	}
@@ -2755,7 +2757,7 @@ static int emac_init_phy(struct emac_instance *dev)
 #endif
 	mutex_unlock(&emac_phy_map_lock);
 	if (i == 0x20) {
-		printk(KERN_WARNING "%s: can't find PHY!\n", np->full_name);
+		printk(KERN_WARNING "%pOF: can't find PHY!\n", np);
 		return -ENXIO;
 	}
 
@@ -2802,7 +2804,7 @@ static int emac_init_phy(struct emac_instance *dev)
 static int emac_init_config(struct emac_instance *dev)
 {
 	struct device_node *np = dev->ofdev->dev.of_node;
-	const void *p;
+	int err;
 
 	/* Read config from device-tree */
 	if (emac_read_uint_prop(np, "mal-device", &dev->mal_ph, 1))
@@ -2851,9 +2853,9 @@ static int emac_init_config(struct emac_instance *dev)
 		dev->mal_burst_size = 256;
 
 	/* PHY mode needs some decoding */
-	dev->phy_mode = of_get_phy_mode(np);
-	if (dev->phy_mode < 0)
-		dev->phy_mode = PHY_MODE_NA;
+	err = of_get_phy_mode(np, &dev->phy_mode);
+	if (err)
+		dev->phy_mode = PHY_INTERFACE_MODE_NA;
 
 	/* Check EMAC version */
 	if (of_device_is_compatible(np, "ibm,emac4sync")) {
@@ -2881,8 +2883,8 @@ static int emac_init_config(struct emac_instance *dev)
 #ifdef CONFIG_IBM_EMAC_NO_FLOW_CTRL
 			dev->features |= EMAC_FTR_NO_FLOW_CONTROL_40x;
 #else
-			printk(KERN_ERR "%s: Flow control not disabled!\n",
-					np->full_name);
+			printk(KERN_ERR "%pOF: Flow control not disabled!\n",
+					np);
 			return -ENXIO;
 #endif
 		}
@@ -2890,9 +2892,9 @@ static int emac_init_config(struct emac_instance *dev)
 	}
 
 	/* Fixup some feature bits based on the device tree */
-	if (of_get_property(np, "has-inverted-stacr-oc", NULL))
+	if (of_property_read_bool(np, "has-inverted-stacr-oc"))
 		dev->features |= EMAC_FTR_STACR_OC_INVERT;
-	if (of_get_property(np, "has-new-stacr-staopc", NULL))
+	if (of_property_read_bool(np, "has-new-stacr-staopc"))
 		dev->features |= EMAC_FTR_HAS_NEW_STACR;
 
 	/* CAB lacks the appropriate properties */
@@ -2905,8 +2907,7 @@ static int emac_init_config(struct emac_instance *dev)
 #ifdef CONFIG_IBM_EMAC_TAH
 		dev->features |= EMAC_FTR_HAS_TAH;
 #else
-		printk(KERN_ERR "%s: TAH support not enabled !\n",
-		       np->full_name);
+		printk(KERN_ERR "%pOF: TAH support not enabled !\n", np);
 		return -ENXIO;
 #endif
 	}
@@ -2915,8 +2916,7 @@ static int emac_init_config(struct emac_instance *dev)
 #ifdef CONFIG_IBM_EMAC_ZMII
 		dev->features |= EMAC_FTR_HAS_ZMII;
 #else
-		printk(KERN_ERR "%s: ZMII support not enabled !\n",
-		       np->full_name);
+		printk(KERN_ERR "%pOF: ZMII support not enabled !\n", np);
 		return -ENXIO;
 #endif
 	}
@@ -2925,20 +2925,19 @@ static int emac_init_config(struct emac_instance *dev)
 #ifdef CONFIG_IBM_EMAC_RGMII
 		dev->features |= EMAC_FTR_HAS_RGMII;
 #else
-		printk(KERN_ERR "%s: RGMII support not enabled !\n",
-		       np->full_name);
+		printk(KERN_ERR "%pOF: RGMII support not enabled !\n", np);
 		return -ENXIO;
 #endif
 	}
 
 	/* Read MAC-address */
-	p = of_get_property(np, "local-mac-address", NULL);
-	if (p == NULL) {
-		printk(KERN_ERR "%s: Can't find local-mac-address property\n",
-		       np->full_name);
-		return -ENXIO;
+	err = of_get_ethdev_address(np, dev->ndev);
+	if (err == -EPROBE_DEFER)
+		return err;
+	if (err) {
+		dev_warn(&dev->ofdev->dev, "Can't get valid mac-address. Generating random.");
+		eth_hw_addr_random(dev->ndev);
 	}
-	memcpy(dev->ndev->dev_addr, p, ETH_ALEN);
 
 	/* IAHT and GAHT filter parameterization */
 	if (emac_has_feature(dev, EMAC_FTR_EMAC4SYNC)) {
@@ -2949,6 +2948,10 @@ static int emac_init_config(struct emac_instance *dev)
 		dev->xaht_width_shift = EMAC4_XAHT_WIDTH_SHIFT;
 	}
 
+	/* This should never happen */
+	if (WARN_ON(EMAC_XAHT_REGS(dev) > EMAC_XAHT_MAX_REGS))
+		return -ENXIO;
+
 	DBG(dev, "features     : 0x%08x / 0x%08x\n", dev->features, EMAC_FTRS_POSSIBLE);
 	DBG(dev, "tx_fifo_size : %d (%d gige)\n", dev->tx_fifo_size, dev->tx_fifo_size_gige);
 	DBG(dev, "rx_fifo_size : %d (%d gige)\n", dev->rx_fifo_size, dev->rx_fifo_size_gige);
@@ -2963,7 +2966,7 @@ static const struct net_device_ops emac_netdev_ops = {
 	.ndo_stop		= emac_close,
 	.ndo_get_stats		= emac_stats,
 	.ndo_set_rx_mode	= emac_set_multicast_list,
-	.ndo_do_ioctl		= emac_ioctl,
+	.ndo_eth_ioctl		= emac_ioctl,
 	.ndo_tx_timeout		= emac_tx_timeout,
 	.ndo_validate_addr	= eth_validate_addr,
 	.ndo_set_mac_address	= emac_set_mac_address,
@@ -2975,7 +2978,7 @@ static const struct net_device_ops emac_gige_netdev_ops = {
 	.ndo_stop		= emac_close,
 	.ndo_get_stats		= emac_stats,
 	.ndo_set_rx_mode	= emac_set_multicast_list,
-	.ndo_do_ioctl		= emac_ioctl,
+	.ndo_eth_ioctl		= emac_ioctl,
 	.ndo_tx_timeout		= emac_tx_timeout,
 	.ndo_validate_addr	= eth_validate_addr,
 	.ndo_set_mac_address	= emac_set_mac_address,
@@ -2995,7 +2998,7 @@ static int emac_probe(struct platform_device *ofdev)
 	 * property here for now, but new flat device trees should set a
 	 * status property to "disabled" instead.
 	 */
-	if (of_get_property(np, "unused", NULL) || !of_device_is_available(np))
+	if (of_property_read_bool(np, "unused") || !of_device_is_available(np))
 		return -ENODEV;
 
 	/* Find ourselves in the bootlist if we are there */
@@ -3005,7 +3008,7 @@ static int emac_probe(struct platform_device *ofdev)
 
 	/* Allocate our net_device structure */
 	err = -ENOMEM;
-	ndev = alloc_etherdev(sizeof(struct emac_instance));
+	ndev = devm_alloc_etherdev(&ofdev->dev, sizeof(struct emac_instance));
 	if (!ndev)
 		goto err_gone;
 
@@ -3016,50 +3019,45 @@ static int emac_probe(struct platform_device *ofdev)
 	SET_NETDEV_DEV(ndev, &ofdev->dev);
 
 	/* Initialize some embedded data structures */
-	mutex_init(&dev->mdio_lock);
-	mutex_init(&dev->link_lock);
+	err = devm_mutex_init(&ofdev->dev, &dev->mdio_lock);
+	if (err)
+		goto err_gone;
+
+	err = devm_mutex_init(&ofdev->dev, &dev->link_lock);
+	if (err)
+		goto err_gone;
+
 	spin_lock_init(&dev->lock);
 	INIT_WORK(&dev->reset_work, emac_reset_work);
 
 	/* Init various config data based on device-tree */
 	err = emac_init_config(dev);
-	if (err != 0)
-		goto err_free;
+	if (err)
+		goto err_gone;
 
-	/* Get interrupts. EMAC irq is mandatory, WOL irq is optional */
-	dev->emac_irq = irq_of_parse_and_map(np, 0);
-	dev->wol_irq = irq_of_parse_and_map(np, 1);
-	if (!dev->emac_irq) {
-		printk(KERN_ERR "%s: Can't map main interrupt\n", np->full_name);
-		goto err_free;
+	/* Setup error IRQ handler */
+	dev->emac_irq = platform_get_irq(ofdev, 0);
+	err = devm_request_irq(&ofdev->dev, dev->emac_irq, emac_irq, 0, "EMAC",
+			       dev);
+	if (err) {
+		dev_err_probe(&ofdev->dev, err, "failed to request IRQ %d",
+			      dev->emac_irq);
+		goto err_gone;
 	}
+
 	ndev->irq = dev->emac_irq;
 
-	/* Map EMAC regs */
-	if (of_address_to_resource(np, 0, &dev->rsrc_regs)) {
-		printk(KERN_ERR "%s: Can't get registers address\n",
-		       np->full_name);
-		goto err_irq_unmap;
-	}
-	// TODO : request_mem_region
-	dev->emacp = ioremap(dev->rsrc_regs.start,
-			     resource_size(&dev->rsrc_regs));
-	if (dev->emacp == NULL) {
-		printk(KERN_ERR "%s: Can't map device registers!\n",
-		       np->full_name);
-		err = -ENOMEM;
-		goto err_irq_unmap;
+	dev->emacp = devm_platform_ioremap_resource(ofdev, 0);
+	if (IS_ERR(dev->emacp)) {
+		dev_err(&ofdev->dev, "can't map device registers");
+		err = PTR_ERR(dev->emacp);
+		goto err_gone;
 	}
 
 	/* Wait for dependent devices */
 	err = emac_wait_deps(dev);
-	if (err) {
-		printk(KERN_ERR
-		       "%s: Timeout waiting for dependent devices\n",
-		       np->full_name);
-		/*  display more info about what's missing ? */
-		goto err_reg_unmap;
-	}
+	if (err)
+		goto err_gone;
 	dev->mal = platform_get_drvdata(dev->mal_dev);
 	if (dev->mdio_dev != NULL)
 		dev->mdio_instance = platform_get_drvdata(dev->mdio_dev);
@@ -3071,8 +3069,8 @@ static int emac_probe(struct platform_device *ofdev)
 	dev->commac.rx_chan_mask = MAL_CHAN_MASK(dev->mal_rx_chan);
 	err = mal_register_commac(dev->mal, &dev->commac);
 	if (err) {
-		printk(KERN_ERR "%s: failed to register with mal %s!\n",
-		       np->full_name, dev->mal_dev->dev.of_node->full_name);
+		printk(KERN_ERR "%pOF: failed to register with mal %pOF!\n",
+		       np, dev->mal_dev->dev.of_node);
 		goto err_rel_deps;
 	}
 	dev->rx_skb_size = emac_rx_skb_size(ndev->mtu);
@@ -3146,10 +3144,10 @@ static int emac_probe(struct platform_device *ofdev)
 
 	netif_carrier_off(ndev);
 
-	err = register_netdev(ndev);
+	err = devm_register_netdev(&ofdev->dev, ndev);
 	if (err) {
-		printk(KERN_ERR "%s: failed to register net device (%d)!\n",
-		       np->full_name, err);
+		printk(KERN_ERR "%pOF: failed to register net device (%d)!\n",
+		       np, err);
 		goto err_detach_tah;
 	}
 
@@ -3159,14 +3157,10 @@ static int emac_probe(struct platform_device *ofdev)
 	wmb();
 	platform_set_drvdata(ofdev, dev);
 
-	/* There's a new kid in town ! Let's tell everybody */
-	wake_up_all(&emac_probe_wait);
-
+	printk(KERN_INFO "%s: EMAC-%d %pOF, MAC %pM\n",
+	       ndev->name, dev->cell_index, np, ndev->dev_addr);
 
-	printk(KERN_INFO "%s: EMAC-%d %s, MAC %pM\n",
-	       ndev->name, dev->cell_index, np->full_name, ndev->dev_addr);
-
-	if (dev->phy_mode == PHY_MODE_SGMII)
+	if (dev->phy_mode == PHY_INTERFACE_MODE_SGMII)
 		printk(KERN_NOTICE "%s: in SGMII mode\n", ndev->name);
 
 	if (dev->phy.address >= 0)
@@ -3191,35 +3185,18 @@ static int emac_probe(struct platform_device *ofdev)
 	mal_unregister_commac(dev->mal, &dev->commac);
  err_rel_deps:
 	emac_put_deps(dev);
- err_reg_unmap:
-	iounmap(dev->emacp);
- err_irq_unmap:
-	if (dev->wol_irq)
-		irq_dispose_mapping(dev->wol_irq);
-	if (dev->emac_irq)
-		irq_dispose_mapping(dev->emac_irq);
- err_free:
-	free_netdev(ndev);
  err_gone:
-	/* if we were on the bootlist, remove us as we won't show up and
-	 * wake up all waiters to notify them in case they were waiting
-	 * on us
-	 */
-	if (blist) {
+	if (blist)
 		*blist = NULL;
-		wake_up_all(&emac_probe_wait);
-	}
 	return err;
 }
 
-static int emac_remove(struct platform_device *ofdev)
+static void emac_remove(struct platform_device *ofdev)
 {
 	struct emac_instance *dev = platform_get_drvdata(ofdev);
 
 	DBG(dev, "remove" NL);
 
-	unregister_netdev(dev->ndev);
-
 	cancel_work_sync(&dev->reset_work);
 
 	if (emac_has_feature(dev, EMAC_FTR_HAS_TAH))
@@ -3229,28 +3206,11 @@ static int emac_remove(struct platform_device *ofdev)
 	if (emac_has_feature(dev, EMAC_FTR_HAS_ZMII))
 		zmii_detach(dev->zmii_dev, dev->zmii_port);
 
-	if (dev->phy_dev)
-		phy_disconnect(dev->phy_dev);
-
-	if (dev->mii_bus)
-		mdiobus_unregister(dev->mii_bus);
-
 	busy_phy_map &= ~(1 << dev->phy.address);
 	DBG(dev, "busy_phy_map now %#x" NL, busy_phy_map);
 
 	mal_unregister_commac(dev->mal, &dev->commac);
 	emac_put_deps(dev);
-
-	iounmap(dev->emacp);
-
-	if (dev->wol_irq)
-		irq_dispose_mapping(dev->wol_irq);
-	if (dev->emac_irq)
-		irq_dispose_mapping(dev->emac_irq);
-
-	free_netdev(dev->ndev);
-
-	return 0;
 }
 
 /* XXX Features in here should be replaced by properties... */
@@ -3289,16 +3249,15 @@ static void __init emac_make_bootlist(void)
 
 	/* Collect EMACs */
 	while((np = of_find_all_nodes(np)) != NULL) {
-		const u32 *idx;
+		u32 idx;
 
 		if (of_match_node(emac_match, np) == NULL)
 			continue;
-		if (of_get_property(np, "unused", NULL))
+		if (of_property_read_bool(np, "unused"))
 			continue;
-		idx = of_get_property(np, "cell-index", NULL);
-		if (idx == NULL)
+		if (of_property_read_u32(np, "cell-index", &idx))
 			continue;
-		cell_indices[i] = *idx;
+		cell_indices[i] = idx;
 		emac_boot_list[i++] = of_node_get(np);
 		if (i >= EMAC_BOOT_LIST_SIZE) {
 			of_node_put(np);
diff --git a/drivers/net/ethernet/ibm/emac/core.h b/drivers/net/ethernet/ibm/emac/core.h
index f10e156641d5..89fa1683ec3c 100644
--- a/drivers/net/ethernet/ibm/emac/core.h
+++ b/drivers/net/ethernet/ibm/emac/core.h
@@ -1,3 +1,4 @@
+/* SPDX-License-Identifier: GPL-2.0-or-later */
 /*
  * drivers/net/ethernet/ibm/emac/core.h
  *
@@ -15,12 +16,6 @@
  *      Armin Kuster <akuster@mvista.com>
  * 	Johnnie Peters <jpeters@mvista.com>
  *      Copyright 2000, 2001 MontaVista Softare Inc.
- *
- * This program is free software; you can redistribute  it and/or modify it
- * under  the terms of  the GNU General  Public License as published by the
- * Free Software Foundation;  either version 2 of the  License, or (at your
- * option) any later version.
- *
  */
 #ifndef __IBM_NEWEMAC_CORE_H
 #define __IBM_NEWEMAC_CORE_H
@@ -32,7 +27,6 @@
 #include <linux/netdevice.h>
 #include <linux/dma-mapping.h>
 #include <linux/spinlock.h>
-#include <linux/of_platform.h>
 #include <linux/slab.h>
 
 #include <asm/io.h>
@@ -68,22 +62,18 @@ static inline int emac_rx_size(int mtu)
 		return mal_rx_size(ETH_DATA_LEN + EMAC_MTU_OVERHEAD);
 }
 
-#define EMAC_DMA_ALIGN(x)		ALIGN((x), dma_get_cache_alignment())
-
-#define EMAC_RX_SKB_HEADROOM		\
-	EMAC_DMA_ALIGN(CONFIG_IBM_EMAC_RX_SKB_HEADROOM)
-
 /* Size of RX skb for the given MTU */
 static inline int emac_rx_skb_size(int mtu)
 {
 	int size = max(mtu + EMAC_MTU_OVERHEAD, emac_rx_size(mtu));
-	return EMAC_DMA_ALIGN(size + 2) + EMAC_RX_SKB_HEADROOM;
+
+	return SKB_DATA_ALIGN(size + NET_IP_ALIGN) + NET_SKB_PAD;
 }
 
 /* RX DMA sync size */
 static inline int emac_rx_sync_size(int mtu)
 {
-	return EMAC_DMA_ALIGN(emac_rx_size(mtu) + 2);
+	return SKB_DATA_ALIGN(emac_rx_size(mtu) + NET_IP_ALIGN);
 }
 
 /* Driver statistcs is split into two parts to make it more cache friendly:
@@ -167,7 +157,6 @@ struct emac_error_stats {
 
 struct emac_instance {
 	struct net_device		*ndev;
-	struct resource			rsrc_regs;
 	struct emac_regs		__iomem *emacp;
 	struct platform_device		*ofdev;
 	struct device_node		**blist; /* bootlist entry */
@@ -181,7 +170,7 @@ struct emac_instance {
 	struct mal_commac		commac;
 
 	/* PHY infos */
-	int				phy_mode;
+	phy_interface_t			phy_mode;
 	u32				phy_map;
 	u32				phy_address;
 	u32				phy_feat_exc;
@@ -199,10 +188,6 @@ struct emac_instance {
 	struct emac_instance		*mdio_instance;
 	struct mutex			mdio_lock;
 
-	/* Device-tree based phy configuration */
-	struct mii_bus			*mii_bus;
-	struct phy_device		*phy_dev;
-
 	/* ZMII infos if any */
 	u32				zmii_ph;
 	u32				zmii_port;
@@ -391,6 +376,9 @@ static inline int emac_has_feature(struct emac_instance *dev,
 #define	EMAC4SYNC_XAHT_SLOTS_SHIFT	8
 #define	EMAC4SYNC_XAHT_WIDTH_SHIFT	5
 
+/* The largest span between slots and widths above is 3 */
+#define	EMAC_XAHT_MAX_REGS		(1 << 3)
+
 #define	EMAC_XAHT_SLOTS(dev)         	(1 << (dev)->xaht_slots_shift)
 #define	EMAC_XAHT_WIDTH(dev)         	(1 << (dev)->xaht_width_shift)
 #define	EMAC_XAHT_REGS(dev)          	(1 << ((dev)->xaht_slots_shift - \
@@ -408,7 +396,7 @@ static inline int emac_has_feature(struct emac_instance *dev,
 	((u32)(1 << (EMAC_XAHT_WIDTH(dev) - 1)) >>	\
 	 ((slot) & (u32)(EMAC_XAHT_WIDTH(dev) - 1)))
 
-static inline u32 *emac_xaht_base(struct emac_instance *dev)
+static inline u32 __iomem *emac_xaht_base(struct emac_instance *dev)
 {
 	struct emac_regs __iomem *p = dev->emacp;
 	int offset;
@@ -421,10 +409,10 @@ static inline u32 *emac_xaht_base(struct emac_instance *dev)
 	else
 		offset = offsetof(struct emac_regs, u0.emac4.iaht1);
 
-	return (u32 *)((ptrdiff_t)p + offset);
+	return (u32 __iomem *)((__force ptrdiff_t)p + offset);
 }
 
-static inline u32 *emac_gaht_base(struct emac_instance *dev)
+static inline u32 __iomem *emac_gaht_base(struct emac_instance *dev)
 {
 	/* GAHT registers always come after an identical number of
 	 * IAHT registers.
diff --git a/drivers/net/ethernet/ibm/emac/debug.h b/drivers/net/ethernet/ibm/emac/debug.h
index 5bdfc174a07e..c09a46a329d9 100644
--- a/drivers/net/ethernet/ibm/emac/debug.h
+++ b/drivers/net/ethernet/ibm/emac/debug.h
@@ -1,3 +1,4 @@
+/* SPDX-License-Identifier: GPL-2.0-or-later */
 /*
  * drivers/net/ethernet/ibm/emac/debug.h
  *
@@ -10,12 +11,6 @@
  *
  * Copyright (c) 2004, 2005 Zultys Technologies
  * Eugene Surovegin <eugene.surovegin@zultys.com> or <ebs@ebshome.net>
- *
- * This program is free software; you can redistribute  it and/or modify it
- * under  the terms of  the GNU General  Public License as published by the
- * Free Software Foundation;  either version 2 of the  License, or (at your
- * option) any later version.
- *
  */
 #ifndef __IBM_NEWEMAC_DEBUG_H
 #define __IBM_NEWEMAC_DEBUG_H
@@ -31,7 +26,7 @@
 #endif
 
 #define EMAC_DBG(d, name, fmt, arg...) \
-	printk(KERN_DEBUG #name "%s: " fmt, d->ofdev->dev.of_node->full_name, ## arg)
+	printk(KERN_DEBUG #name "%pOF: " fmt, d->ofdev->dev.of_node, ## arg)
 
 #if DBG_LEVEL > 0
 #  define DBG(d,f,x...)		EMAC_DBG(d, emac, f, ##x)
diff --git a/drivers/net/ethernet/ibm/emac/emac.h b/drivers/net/ethernet/ibm/emac/emac.h
index 5afcc27ceebb..09d3ac374b2d 100644
--- a/drivers/net/ethernet/ibm/emac/emac.h
+++ b/drivers/net/ethernet/ibm/emac/emac.h
@@ -1,3 +1,4 @@
+/* SPDX-License-Identifier: GPL-2.0-or-later */
 /*
  * drivers/net/ethernet/ibm/emac/emac.h
  *
@@ -15,12 +16,6 @@
  *      Matt Porter <mporter@kernel.crashing.org>
  *      Armin Kuster <akuster@mvista.com>
  * 	Copyright 2002-2004 MontaVista Software Inc.
- *
- * This program is free software; you can redistribute  it and/or modify it
- * under  the terms of  the GNU General  Public License as published by the
- * Free Software Foundation;  either version 2 of the  License, or (at your
- * option) any later version.
- *
  */
 #ifndef __IBM_NEWEMAC_H
 #define __IBM_NEWEMAC_H
@@ -82,7 +77,7 @@ struct emac_regs {
 		struct {
 			u32 rsvd1;
 			u32 revid;
- 			u32 rsvd2[2];
+			u32 rsvd2[2];
 			u32 iaht1;	/* Reset, R     */
 			u32 iaht2;	/* Reset, R     */
 			u32 iaht3;	/* Reset, R     */
@@ -104,19 +99,6 @@ struct emac_regs {
 	} u1;
 };
 
-/*
- * PHY mode settings (EMAC <-> ZMII/RGMII bridge <-> PHY)
- */
-#define PHY_MODE_NA	PHY_INTERFACE_MODE_NA
-#define PHY_MODE_MII	PHY_INTERFACE_MODE_MII
-#define PHY_MODE_RMII	PHY_INTERFACE_MODE_RMII
-#define PHY_MODE_SMII	PHY_INTERFACE_MODE_SMII
-#define PHY_MODE_RGMII	PHY_INTERFACE_MODE_RGMII
-#define PHY_MODE_TBI	PHY_INTERFACE_MODE_TBI
-#define PHY_MODE_GMII	PHY_INTERFACE_MODE_GMII
-#define PHY_MODE_RTBI	PHY_INTERFACE_MODE_RTBI
-#define PHY_MODE_SGMII	PHY_INTERFACE_MODE_SGMII
-
 /* EMACx_MR0 */
 #define EMAC_MR0_RXI			0x80000000
 #define EMAC_MR0_TXI			0x40000000
@@ -151,9 +133,11 @@ struct emac_regs {
 
 #define EMAC4_MR1_RFS_2K		0x00100000
 #define EMAC4_MR1_RFS_4K		0x00180000
+#define EMAC4_MR1_RFS_8K		0x00200000
 #define EMAC4_MR1_RFS_16K		0x00280000
 #define EMAC4_MR1_TFS_2K       		0x00020000
 #define EMAC4_MR1_TFS_4K		0x00030000
+#define EMAC4_MR1_TFS_8K		0x00040000
 #define EMAC4_MR1_TFS_16K		0x00050000
 #define EMAC4_MR1_TR			0x00008000
 #define EMAC4_MR1_MWSW_001		0x00001000
diff --git a/drivers/net/ethernet/ibm/emac/mal.c b/drivers/net/ethernet/ibm/emac/mal.c
index 91b1a558f37d..7d70056e9008 100644
--- a/drivers/net/ethernet/ibm/emac/mal.c
+++ b/drivers/net/ethernet/ibm/emac/mal.c
@@ -1,3 +1,4 @@
+// SPDX-License-Identifier: GPL-2.0-or-later
 /*
  * drivers/net/ethernet/ibm/emac/mal.c
  *
@@ -17,17 +18,13 @@
  *
  *      Armin Kuster <akuster@mvista.com>
  *      Copyright 2002 MontaVista Softare Inc.
- *
- * This program is free software; you can redistribute  it and/or modify it
- * under  the terms of  the GNU General  Public License as published by the
- * Free Software Foundation;  either version 2 of the  License, or (at your
- * option) any later version.
- *
  */
 
 #include <linux/delay.h>
 #include <linux/slab.h>
+#include <linux/of.h>
 #include <linux/of_irq.h>
+#include <linux/platform_device.h>
 
 #include "core.h"
 #include <asm/dcr-regs.h>
@@ -402,7 +399,7 @@ static int mal_poll(struct napi_struct *napi, int budget)
 	unsigned long flags;
 
 	MAL_DBG2(mal, "poll(%d)" NL, budget);
- again:
+
 	/* Process TX skbs */
 	list_for_each(l, &mal->poll_list) {
 		struct mal_commac *mc =
@@ -445,13 +442,12 @@ static int mal_poll(struct napi_struct *napi, int budget)
 		if (unlikely(mc->ops->peek_rx(mc->dev) ||
 			     test_bit(MAL_COMMAC_RX_STOPPED, &mc->flags))) {
 			MAL_DBG2(mal, "rotting packet" NL);
-			if (!napi_reschedule(napi))
+			if (!napi_schedule(napi))
 				goto more_work;
 
 			spin_lock_irqsave(&mal->lock, flags);
 			mal_disable_eob_irq(mal);
 			spin_unlock_irqrestore(&mal->lock, flags);
-			goto again;
 		}
 		mc->ops->poll_tx(mc->dev);
 	}
@@ -528,7 +524,8 @@ static int mal_probe(struct platform_device *ofdev)
 	unsigned long irqflags;
 	irq_handler_t hdlr_serr, hdlr_txde, hdlr_rxde;
 
-	mal = kzalloc(sizeof(struct mal_instance), GFP_KERNEL);
+	mal = devm_kzalloc(&ofdev->dev, sizeof(struct mal_instance),
+			   GFP_KERNEL);
 	if (!mal)
 		return -ENOMEM;
 
@@ -543,8 +540,7 @@ static int mal_probe(struct platform_device *ofdev)
 		printk(KERN_ERR
 		       "mal%d: can't find MAL num-tx-chans property!\n",
 		       index);
-		err = -ENODEV;
-		goto fail;
+		return -ENODEV;
 	}
 	mal->num_tx_chans = prop[0];
 
@@ -553,8 +549,7 @@ static int mal_probe(struct platform_device *ofdev)
 		printk(KERN_ERR
 		       "mal%d: can't find MAL num-rx-chans property!\n",
 		       index);
-		err = -ENODEV;
-		goto fail;
+		return -ENODEV;
 	}
 	mal->num_rx_chans = prop[0];
 
@@ -562,15 +557,13 @@ static int mal_probe(struct platform_device *ofdev)
 	if (dcr_base == 0) {
 		printk(KERN_ERR
 		       "mal%d: can't find DCR resource!\n", index);
-		err = -ENODEV;
-		goto fail;
+		return -ENODEV;
 	}
 	mal->dcr_host = dcr_map(ofdev->dev.of_node, dcr_base, 0x100);
 	if (!DCR_MAP_OK(mal->dcr_host)) {
 		printk(KERN_ERR
 		       "mal%d: failed to map DCRs !\n", index);
-		err = -ENODEV;
-		goto fail;
+		return -ENODEV;
 	}
 
 	if (of_device_is_compatible(ofdev->dev.of_node, "ibm,mcmal-405ez")) {
@@ -579,40 +572,25 @@ static int mal_probe(struct platform_device *ofdev)
 		mal->features |= (MAL_FTR_CLEAR_ICINTSTAT |
 				MAL_FTR_COMMON_ERR_INT);
 #else
-		printk(KERN_ERR "%s: Support for 405EZ not enabled!\n",
-				ofdev->dev.of_node->full_name);
-		err = -ENODEV;
-		goto fail;
-#endif
-	}
-
-	mal->txeob_irq = irq_of_parse_and_map(ofdev->dev.of_node, 0);
-	mal->rxeob_irq = irq_of_parse_and_map(ofdev->dev.of_node, 1);
-	mal->serr_irq = irq_of_parse_and_map(ofdev->dev.of_node, 2);
-
-	if (mal_has_feature(mal, MAL_FTR_COMMON_ERR_INT)) {
-		mal->txde_irq = mal->rxde_irq = mal->serr_irq;
-	} else {
-		mal->txde_irq = irq_of_parse_and_map(ofdev->dev.of_node, 3);
-		mal->rxde_irq = irq_of_parse_and_map(ofdev->dev.of_node, 4);
-	}
-
-	if (!mal->txeob_irq || !mal->rxeob_irq || !mal->serr_irq ||
-	    !mal->txde_irq  || !mal->rxde_irq) {
-		printk(KERN_ERR
-		       "mal%d: failed to map interrupts !\n", index);
+		printk(KERN_ERR "%pOF: Support for 405EZ not enabled!\n",
+				ofdev->dev.of_node);
 		err = -ENODEV;
 		goto fail_unmap;
+#endif
 	}
 
 	INIT_LIST_HEAD(&mal->poll_list);
 	INIT_LIST_HEAD(&mal->list);
 	spin_lock_init(&mal->lock);
 
-	init_dummy_netdev(&mal->dummy_dev);
+	mal->dummy_dev = alloc_netdev_dummy(0);
+	if (!mal->dummy_dev) {
+		err = -ENOMEM;
+		goto fail_unmap;
+	}
 
-	netif_napi_add(&mal->dummy_dev, &mal->napi, mal_poll,
-		       CONFIG_IBM_EMAC_POLL_WEIGHT);
+	netif_napi_add_weight(mal->dummy_dev, &mal->napi, mal_poll,
+			      CONFIG_IBM_EMAC_POLL_WEIGHT);
 
 	/* Load power-on reset defaults */
 	mal_reset(mal);
@@ -637,11 +615,11 @@ static int mal_probe(struct platform_device *ofdev)
 	bd_size = sizeof(struct mal_descriptor) *
 		(NUM_TX_BUFF * mal->num_tx_chans +
 		 NUM_RX_BUFF * mal->num_rx_chans);
-	mal->bd_virt = dma_zalloc_coherent(&ofdev->dev, bd_size, &mal->bd_dma,
-					   GFP_KERNEL);
+	mal->bd_virt = dma_alloc_coherent(&ofdev->dev, bd_size, &mal->bd_dma,
+					  GFP_KERNEL);
 	if (mal->bd_virt == NULL) {
 		err = -ENOMEM;
-		goto fail_unmap;
+		goto fail_dummy;
 	}
 
 	for (i = 0; i < mal->num_tx_chans; ++i)
@@ -654,31 +632,43 @@ static int mal_probe(struct platform_device *ofdev)
 			     sizeof(struct mal_descriptor) *
 			     mal_rx_bd_offset(mal, i));
 
+	mal->txeob_irq = platform_get_irq(ofdev, 0);
+	mal->rxeob_irq = platform_get_irq(ofdev, 1);
+	mal->serr_irq = platform_get_irq(ofdev, 2);
+
 	if (mal_has_feature(mal, MAL_FTR_COMMON_ERR_INT)) {
+		mal->txde_irq = mal->rxde_irq = mal->serr_irq;
 		irqflags = IRQF_SHARED;
 		hdlr_serr = hdlr_txde = hdlr_rxde = mal_int;
 	} else {
+		mal->txde_irq = platform_get_irq(ofdev, 3);
+		mal->rxde_irq = platform_get_irq(ofdev, 4);
 		irqflags = 0;
 		hdlr_serr = mal_serr;
 		hdlr_txde = mal_txde;
 		hdlr_rxde = mal_rxde;
 	}
 
-	err = request_irq(mal->serr_irq, hdlr_serr, irqflags, "MAL SERR", mal);
+	err = devm_request_irq(&ofdev->dev, mal->serr_irq, hdlr_serr, irqflags,
+			       "MAL SERR", mal);
 	if (err)
 		goto fail2;
-	err = request_irq(mal->txde_irq, hdlr_txde, irqflags, "MAL TX DE", mal);
+	err = devm_request_irq(&ofdev->dev, mal->txde_irq, hdlr_txde, irqflags,
+			       "MAL TX DE", mal);
 	if (err)
-		goto fail3;
-	err = request_irq(mal->txeob_irq, mal_txeob, 0, "MAL TX EOB", mal);
+		goto fail2;
+	err = devm_request_irq(&ofdev->dev, mal->txeob_irq, mal_txeob, 0,
+			       "MAL TX EOB", mal);
 	if (err)
-		goto fail4;
-	err = request_irq(mal->rxde_irq, hdlr_rxde, irqflags, "MAL RX DE", mal);
+		goto fail2;
+	err = devm_request_irq(&ofdev->dev, mal->rxde_irq, hdlr_rxde, irqflags,
+			       "MAL RX DE", mal);
 	if (err)
-		goto fail5;
-	err = request_irq(mal->rxeob_irq, mal_rxeob, 0, "MAL RX EOB", mal);
+		goto fail2;
+	err = devm_request_irq(&ofdev->dev, mal->rxeob_irq, mal_rxeob, 0,
+			       "MAL RX EOB", mal);
 	if (err)
-		goto fail6;
+		goto fail2;
 
 	/* Enable all MAL SERR interrupt sources */
 	set_mal_dcrn(mal, MAL_IER, MAL_IER_EVENTS);
@@ -687,8 +677,8 @@ static int mal_probe(struct platform_device *ofdev)
 	mal_enable_eob_irq(mal);
 
 	printk(KERN_INFO
-	       "MAL v%d %s, %d TX channels, %d RX channels\n",
-	       mal->version, ofdev->dev.of_node->full_name,
+	       "MAL v%d %pOF, %d TX channels, %d RX channels\n",
+	       mal->version, ofdev->dev.of_node,
 	       mal->num_tx_chans, mal->num_rx_chans);
 
 	/* Advertise this instance to the rest of the world */
@@ -697,25 +687,16 @@ static int mal_probe(struct platform_device *ofdev)
 
 	return 0;
 
- fail6:
-	free_irq(mal->rxde_irq, mal);
- fail5:
-	free_irq(mal->txeob_irq, mal);
- fail4:
-	free_irq(mal->txde_irq, mal);
- fail3:
-	free_irq(mal->serr_irq, mal);
  fail2:
 	dma_free_coherent(&ofdev->dev, bd_size, mal->bd_virt, mal->bd_dma);
+ fail_dummy:
+	free_netdev(mal->dummy_dev);
  fail_unmap:
 	dcr_unmap(mal->dcr_host, 0x100);
- fail:
-	kfree(mal);
-
 	return err;
 }
 
-static int mal_remove(struct platform_device *ofdev)
+static void mal_remove(struct platform_device *ofdev)
 {
 	struct mal_instance *mal = platform_get_drvdata(ofdev);
 
@@ -730,22 +711,17 @@ static int mal_remove(struct platform_device *ofdev)
 		       "mal%d: commac list is not empty on remove!\n",
 		       mal->index);
 
-	free_irq(mal->serr_irq, mal);
-	free_irq(mal->txde_irq, mal);
-	free_irq(mal->txeob_irq, mal);
-	free_irq(mal->rxde_irq, mal);
-	free_irq(mal->rxeob_irq, mal);
-
 	mal_reset(mal);
 
+	free_netdev(mal->dummy_dev);
+
+	dcr_unmap(mal->dcr_host, 0x100);
+
 	dma_free_coherent(&ofdev->dev,
 			  sizeof(struct mal_descriptor) *
-			  (NUM_TX_BUFF * mal->num_tx_chans +
-			   NUM_RX_BUFF * mal->num_rx_chans), mal->bd_virt,
-			  mal->bd_dma);
-	kfree(mal);
-
-	return 0;
+				  (NUM_TX_BUFF * mal->num_tx_chans +
+				   NUM_RX_BUFF * mal->num_rx_chans),
+			  mal->bd_virt, mal->bd_dma);
 }
 
 static const struct of_device_id mal_platform_match[] =
diff --git a/drivers/net/ethernet/ibm/emac/mal.h b/drivers/net/ethernet/ibm/emac/mal.h
index eeade2ea8334..e0ddc41186a2 100644
--- a/drivers/net/ethernet/ibm/emac/mal.h
+++ b/drivers/net/ethernet/ibm/emac/mal.h
@@ -1,3 +1,4 @@
+/* SPDX-License-Identifier: GPL-2.0-or-later */
 /*
  * drivers/net/ethernet/ibm/emac/mal.h
  *
@@ -14,12 +15,6 @@
  * Based on original work by
  *      Armin Kuster <akuster@mvista.com>
  *      Copyright 2002 MontaVista Softare Inc.
- *
- * This program is free software; you can redistribute  it and/or modify it
- * under  the terms of  the GNU General  Public License as published by the
- * Free Software Foundation;  either version 2 of the  License, or (at your
- * option) any later version.
- *
  */
 #ifndef __IBM_NEWEMAC_MAL_H
 #define __IBM_NEWEMAC_MAL_H
@@ -136,7 +131,7 @@ static inline int mal_rx_size(int len)
 
 static inline int mal_tx_chunks(int len)
 {
-	return (len + MAL_MAX_TX_SIZE - 1) / MAL_MAX_TX_SIZE;
+	return DIV_ROUND_UP(len, MAL_MAX_TX_SIZE);
 }
 
 #define MAL_CHAN_MASK(n)	(0x80000000 >> (n))
@@ -210,7 +205,7 @@ struct mal_instance {
 	int			index;
 	spinlock_t		lock;
 
-	struct net_device	dummy_dev;
+	struct net_device	*dummy_dev;
 
 	unsigned int features;
 };
diff --git a/drivers/net/ethernet/ibm/emac/phy.c b/drivers/net/ethernet/ibm/emac/phy.c
index 5b88cc690c22..1e798cc9b6b8 100644
--- a/drivers/net/ethernet/ibm/emac/phy.c
+++ b/drivers/net/ethernet/ibm/emac/phy.c
@@ -1,3 +1,4 @@
+// SPDX-License-Identifier: GPL-2.0-only
 /*
  * drivers/net/ethernet/ibm/emac/phy.c
  *
@@ -96,7 +97,7 @@ int emac_mii_reset_gpcs(struct mii_phy *phy)
 	if ((val & BMCR_ISOLATE) && limit > 0)
 		gpcs_phy_write(phy, MII_BMCR, val & ~BMCR_ISOLATE);
 
-	if (limit > 0 && phy->mode == PHY_MODE_SGMII) {
+	if (limit > 0 && phy->mode == PHY_INTERFACE_MODE_SGMII) {
 		/* Configure GPCS interface to recommended setting for SGMII */
 		gpcs_phy_write(phy, 0x04, 0x8120); /* AsymPause, FDX */
 		gpcs_phy_write(phy, 0x07, 0x2801); /* msg_pg, toggle */
@@ -276,7 +277,7 @@ static int genmii_read_link(struct mii_phy *phy)
 }
 
 /* Generic implementation for most 10/100/1000 PHYs */
-static struct mii_phy_ops generic_phy_ops = {
+static const struct mii_phy_ops generic_phy_ops = {
 	.setup_aneg	= genmii_setup_aneg,
 	.setup_forced	= genmii_setup_forced,
 	.poll_link	= genmii_poll_link,
@@ -313,16 +314,16 @@ static int cis8201_init(struct mii_phy *phy)
 	epcr &= ~EPCR_MODE_MASK;
 
 	switch (phy->mode) {
-	case PHY_MODE_TBI:
+	case PHY_INTERFACE_MODE_TBI:
 		epcr |= EPCR_TBI_MODE;
 		break;
-	case PHY_MODE_RTBI:
+	case PHY_INTERFACE_MODE_RTBI:
 		epcr |= EPCR_RTBI_MODE;
 		break;
-	case PHY_MODE_GMII:
+	case PHY_INTERFACE_MODE_GMII:
 		epcr |= EPCR_GMII_MODE;
 		break;
-	case PHY_MODE_RGMII:
+	case PHY_INTERFACE_MODE_RGMII:
 	default:
 		epcr |= EPCR_RGMII_MODE;
 	}
@@ -340,7 +341,7 @@ static int cis8201_init(struct mii_phy *phy)
 	return 0;
 }
 
-static struct mii_phy_ops cis8201_phy_ops = {
+static const struct mii_phy_ops cis8201_phy_ops = {
 	.init		= cis8201_init,
 	.setup_aneg	= genmii_setup_aneg,
 	.setup_forced	= genmii_setup_forced,
@@ -420,7 +421,7 @@ static int et1011c_init(struct mii_phy *phy)
 	return 0;
 }
 
-static struct mii_phy_ops et1011c_phy_ops = {
+static const struct mii_phy_ops et1011c_phy_ops = {
 	.init		= et1011c_init,
 	.setup_aneg	= genmii_setup_aneg,
 	.setup_forced	= genmii_setup_forced,
@@ -439,7 +440,7 @@ static struct mii_phy_def et1011c_phy_def = {
 
 
 
-static struct mii_phy_ops m88e1111_phy_ops = {
+static const struct mii_phy_ops m88e1111_phy_ops = {
 	.init		= m88e1111_init,
 	.setup_aneg	= genmii_setup_aneg,
 	.setup_forced	= genmii_setup_forced,
@@ -455,7 +456,7 @@ static struct mii_phy_def m88e1111_phy_def = {
 	.ops		= &m88e1111_phy_ops,
 };
 
-static struct mii_phy_ops m88e1112_phy_ops = {
+static const struct mii_phy_ops m88e1112_phy_ops = {
 	.init		= m88e1112_init,
 	.setup_aneg	= genmii_setup_aneg,
 	.setup_forced	= genmii_setup_forced,
@@ -480,7 +481,7 @@ static int ar8035_init(struct mii_phy *phy)
 	return 0;
 }
 
-static struct mii_phy_ops ar8035_phy_ops = {
+static const struct mii_phy_ops ar8035_phy_ops = {
 	.init		= ar8035_init,
 	.setup_aneg	= genmii_setup_aneg,
 	.setup_forced	= genmii_setup_forced,
diff --git a/drivers/net/ethernet/ibm/emac/phy.h b/drivers/net/ethernet/ibm/emac/phy.h
index d7e41ec37467..2184e8373ee5 100644
--- a/drivers/net/ethernet/ibm/emac/phy.h
+++ b/drivers/net/ethernet/ibm/emac/phy.h
@@ -1,3 +1,4 @@
+/* SPDX-License-Identifier: GPL-2.0-or-later */
 /*
  * drivers/net/ethernet/ibm/emac/phy.h
  *
@@ -13,11 +14,6 @@
  *
  * Minor additions by Eugene Surovegin <ebs@ebshome.net>, 2004
  *
- * This program is free software; you can redistribute  it and/or modify it
- * under  the terms of  the GNU General  Public License as published by the
- * Free Software Foundation;  either version 2 of the  License, or (at your
- * option) any later version.
- *
  * This file basically duplicates sungem_phy.{c,h} with different PHYs
  * supported. I'm looking into merging that in a single mii layer more
  * flexible than mii.c
diff --git a/drivers/net/ethernet/ibm/emac/rgmii.c b/drivers/net/ethernet/ibm/emac/rgmii.c
index 206ccbbae7bb..b544dd8633b7 100644
--- a/drivers/net/ethernet/ibm/emac/rgmii.c
+++ b/drivers/net/ethernet/ibm/emac/rgmii.c
@@ -1,3 +1,4 @@
+// SPDX-License-Identifier: GPL-2.0-or-later
 /*
  * drivers/net/ethernet/ibm/emac/rgmii.c
  *
@@ -14,17 +15,13 @@
  * Based on original work by
  * 	Matt Porter <mporter@kernel.crashing.org>
  * 	Copyright 2004 MontaVista Software, Inc.
- *
- * This program is free software; you can redistribute  it and/or modify it
- * under  the terms of  the GNU General  Public License as published by the
- * Free Software Foundation;  either version 2 of the  License, or (at your
- * option) any later version.
- *
  */
 #include <linux/slab.h>
 #include <linux/kernel.h>
 #include <linux/ethtool.h>
+#include <linux/of.h>
 #include <linux/of_address.h>
+#include <linux/platform_device.h>
 #include <asm/io.h>
 
 #include "emac.h"
@@ -52,43 +49,28 @@
 /* RGMII bridge supports only GMII/TBI and RGMII/RTBI PHYs */
 static inline int rgmii_valid_mode(int phy_mode)
 {
-	return  phy_mode == PHY_MODE_GMII ||
-		phy_mode == PHY_MODE_MII ||
-		phy_mode == PHY_MODE_RGMII ||
-		phy_mode == PHY_MODE_TBI ||
-		phy_mode == PHY_MODE_RTBI;
-}
-
-static inline const char *rgmii_mode_name(int mode)
-{
-	switch (mode) {
-	case PHY_MODE_RGMII:
-		return "RGMII";
-	case PHY_MODE_TBI:
-		return "TBI";
-	case PHY_MODE_GMII:
-		return "GMII";
-	case PHY_MODE_MII:
-		return "MII";
-	case PHY_MODE_RTBI:
-		return "RTBI";
-	default:
-		BUG();
-	}
+	return  phy_interface_mode_is_rgmii(phy_mode) ||
+		phy_mode == PHY_INTERFACE_MODE_GMII ||
+		phy_mode == PHY_INTERFACE_MODE_MII ||
+		phy_mode == PHY_INTERFACE_MODE_TBI ||
+		phy_mode == PHY_INTERFACE_MODE_RTBI;
 }
 
 static inline u32 rgmii_mode_mask(int mode, int input)
 {
 	switch (mode) {
-	case PHY_MODE_RGMII:
+	case PHY_INTERFACE_MODE_RGMII:
+	case PHY_INTERFACE_MODE_RGMII_ID:
+	case PHY_INTERFACE_MODE_RGMII_RXID:
+	case PHY_INTERFACE_MODE_RGMII_TXID:
 		return RGMII_FER_RGMII(input);
-	case PHY_MODE_TBI:
+	case PHY_INTERFACE_MODE_TBI:
 		return RGMII_FER_TBI(input);
-	case PHY_MODE_GMII:
+	case PHY_INTERFACE_MODE_GMII:
 		return RGMII_FER_GMII(input);
-	case PHY_MODE_MII:
+	case PHY_INTERFACE_MODE_MII:
 		return RGMII_FER_MII(input);
-	case PHY_MODE_RTBI:
+	case PHY_INTERFACE_MODE_RTBI:
 		return RGMII_FER_RTBI(input);
 	default:
 		BUG();
@@ -104,8 +86,8 @@ int rgmii_attach(struct platform_device *ofdev, int input, int mode)
 
 	/* Check if we need to attach to a RGMII */
 	if (input < 0 || !rgmii_valid_mode(mode)) {
-		printk(KERN_ERR "%s: unsupported settings !\n",
-		       ofdev->dev.of_node->full_name);
+		printk(KERN_ERR "%pOF: unsupported settings !\n",
+		       ofdev->dev.of_node);
 		return -ENODEV;
 	}
 
@@ -114,8 +96,8 @@ int rgmii_attach(struct platform_device *ofdev, int input, int mode)
 	/* Enable this input */
 	out_be32(&p->fer, in_be32(&p->fer) | rgmii_mode_mask(mode, input));
 
-	printk(KERN_NOTICE "%s: input %d in %s mode\n",
-	       ofdev->dev.of_node->full_name, input, rgmii_mode_name(mode));
+	printk(KERN_NOTICE "%pOF: input %d in %s mode\n",
+	       ofdev->dev.of_node, input, phy_modes(mode));
 
 	++dev->users;
 
@@ -234,37 +216,28 @@ void *rgmii_dump_regs(struct platform_device *ofdev, void *buf)
 
 static int rgmii_probe(struct platform_device *ofdev)
 {
-	struct device_node *np = ofdev->dev.of_node;
 	struct rgmii_instance *dev;
-	struct resource regs;
-	int rc;
+	int err;
 
-	rc = -ENOMEM;
-	dev = kzalloc(sizeof(struct rgmii_instance), GFP_KERNEL);
-	if (dev == NULL)
-		goto err_gone;
+	dev = devm_kzalloc(&ofdev->dev, sizeof(struct rgmii_instance),
+			   GFP_KERNEL);
+	if (!dev)
+		return -ENOMEM;
 
-	mutex_init(&dev->lock);
-	dev->ofdev = ofdev;
+	err = devm_mutex_init(&ofdev->dev, &dev->lock);
+	if (err)
+		return err;
 
-	rc = -ENXIO;
-	if (of_address_to_resource(np, 0, &regs)) {
-		printk(KERN_ERR "%s: Can't get registers address\n",
-		       np->full_name);
-		goto err_free;
-	}
+	dev->ofdev = ofdev;
 
-	rc = -ENOMEM;
-	dev->base = (struct rgmii_regs __iomem *)ioremap(regs.start,
-						 sizeof(struct rgmii_regs));
-	if (dev->base == NULL) {
-		printk(KERN_ERR "%s: Can't map device registers!\n",
-		       np->full_name);
-		goto err_free;
+	dev->base = devm_platform_ioremap_resource(ofdev, 0);
+	if (IS_ERR(dev->base)) {
+		dev_err(&ofdev->dev, "can't map device registers");
+		return PTR_ERR(dev->base);
 	}
 
 	/* Check for RGMII flags */
-	if (of_get_property(ofdev->dev.of_node, "has-mdio", NULL))
+	if (of_property_read_bool(ofdev->dev.of_node, "has-mdio"))
 		dev->flags |= EMAC_RGMII_FLAG_HAS_MDIO;
 
 	/* CAB lacks the right properties, fix this up */
@@ -278,31 +251,14 @@ static int rgmii_probe(struct platform_device *ofdev)
 	out_be32(&dev->base->fer, 0);
 
 	printk(KERN_INFO
-	       "RGMII %s initialized with%s MDIO support\n",
-	       ofdev->dev.of_node->full_name,
+	       "RGMII %pOF initialized with%s MDIO support\n",
+	       ofdev->dev.of_node,
 	       (dev->flags & EMAC_RGMII_FLAG_HAS_MDIO) ? "" : "out");
 
 	wmb();
 	platform_set_drvdata(ofdev, dev);
 
 	return 0;
-
- err_free:
-	kfree(dev);
- err_gone:
-	return rc;
-}
-
-static int rgmii_remove(struct platform_device *ofdev)
-{
-	struct rgmii_instance *dev = platform_get_drvdata(ofdev);
-
-	WARN_ON(dev->users != 0);
-
-	iounmap(dev->base);
-	kfree(dev);
-
-	return 0;
 }
 
 static const struct of_device_id rgmii_match[] =
@@ -322,7 +278,6 @@ static struct platform_driver rgmii_driver = {
 		.of_match_table = rgmii_match,
 	},
 	.probe = rgmii_probe,
-	.remove = rgmii_remove,
 };
 
 int __init rgmii_init(void)
diff --git a/drivers/net/ethernet/ibm/emac/rgmii.h b/drivers/net/ethernet/ibm/emac/rgmii.h
index d4f1374d1900..8e4e36eed172 100644
--- a/drivers/net/ethernet/ibm/emac/rgmii.h
+++ b/drivers/net/ethernet/ibm/emac/rgmii.h
@@ -1,3 +1,4 @@
+/* SPDX-License-Identifier: GPL-2.0-or-later */
 /*
  * drivers/net/ethernet/ibm/emac/rgmii.h
  *
@@ -16,11 +17,6 @@
  *
  * Copyright (c) 2004, 2005 Zultys Technologies.
  * Eugene Surovegin <eugene.surovegin@zultys.com> or <ebs@ebshome.net>
- *
- * This program is free software; you can redistribute  it and/or modify it
- * under  the terms of  the GNU General  Public License as published by the
- * Free Software Foundation;  either version 2 of the  License, or (at your
- * option) any later version.
  */
 
 #ifndef __IBM_NEWEMAC_RGMII_H
diff --git a/drivers/net/ethernet/ibm/emac/tah.c b/drivers/net/ethernet/ibm/emac/tah.c
index 32cb6c9007c5..09f6373ed2f9 100644
--- a/drivers/net/ethernet/ibm/emac/tah.c
+++ b/drivers/net/ethernet/ibm/emac/tah.c
@@ -1,3 +1,4 @@
+// SPDX-License-Identifier: GPL-2.0-or-later
 /*
  * drivers/net/ethernet/ibm/emac/tah.c
  *
@@ -12,13 +13,10 @@
  * Matt Porter <mporter@kernel.crashing.org>
  *
  * Copyright (c) 2005 Eugene Surovegin <ebs@ebshome.net>
- *
- * This program is free software; you can redistribute  it and/or modify it
- * under  the terms of  the GNU General  Public License as published by the
- * Free Software Foundation;  either version 2 of the  License, or (at your
- * option) any later version.
  */
+#include <linux/mod_devicetable.h>
 #include <linux/of_address.h>
+#include <linux/platform_device.h>
 #include <asm/io.h>
 
 #include "emac.h"
@@ -58,8 +56,7 @@ void tah_reset(struct platform_device *ofdev)
 		--n;
 
 	if (unlikely(!n))
-		printk(KERN_ERR "%s: reset timeout\n",
-			ofdev->dev.of_node->full_name);
+		printk(KERN_ERR "%pOF: reset timeout\n", ofdev->dev.of_node);
 
 	/* 10KB TAH TX FIFO accommodates the max MTU of 9000 */
 	out_be32(&p->mr,
@@ -90,33 +87,24 @@ void *tah_dump_regs(struct platform_device *ofdev, void *buf)
 
 static int tah_probe(struct platform_device *ofdev)
 {
-	struct device_node *np = ofdev->dev.of_node;
 	struct tah_instance *dev;
-	struct resource regs;
-	int rc;
+	int err;
 
-	rc = -ENOMEM;
-	dev = kzalloc(sizeof(struct tah_instance), GFP_KERNEL);
-	if (dev == NULL)
-		goto err_gone;
+	dev = devm_kzalloc(&ofdev->dev, sizeof(struct tah_instance),
+			   GFP_KERNEL);
+	if (!dev)
+		return -ENOMEM;
 
-	mutex_init(&dev->lock);
-	dev->ofdev = ofdev;
+	err = devm_mutex_init(&ofdev->dev, &dev->lock);
+	if (err)
+		return err;
 
-	rc = -ENXIO;
-	if (of_address_to_resource(np, 0, &regs)) {
-		printk(KERN_ERR "%s: Can't get registers address\n",
-		       np->full_name);
-		goto err_free;
-	}
+	dev->ofdev = ofdev;
 
-	rc = -ENOMEM;
-	dev->base = (struct tah_regs __iomem *)ioremap(regs.start,
-					       sizeof(struct tah_regs));
-	if (dev->base == NULL) {
-		printk(KERN_ERR "%s: Can't map device registers!\n",
-		       np->full_name);
-		goto err_free;
+	dev->base = devm_platform_ioremap_resource(ofdev, 0);
+	if (IS_ERR(dev->base)) {
+		dev_err(&ofdev->dev, "can't map device registers");
+		return PTR_ERR(dev->base);
 	}
 
 	platform_set_drvdata(ofdev, dev);
@@ -124,28 +112,10 @@ static int tah_probe(struct platform_device *ofdev)
 	/* Initialize TAH and enable IPv4 checksum verification, no TSO yet */
 	tah_reset(ofdev);
 
-	printk(KERN_INFO
-	       "TAH %s initialized\n", ofdev->dev.of_node->full_name);
+	printk(KERN_INFO "TAH %pOF initialized\n", ofdev->dev.of_node);
 	wmb();
 
 	return 0;
-
- err_free:
-	kfree(dev);
- err_gone:
-	return rc;
-}
-
-static int tah_remove(struct platform_device *ofdev)
-{
-	struct tah_instance *dev = platform_get_drvdata(ofdev);
-
-	WARN_ON(dev->users != 0);
-
-	iounmap(dev->base);
-	kfree(dev);
-
-	return 0;
 }
 
 static const struct of_device_id tah_match[] =
@@ -166,7 +136,6 @@ static struct platform_driver tah_driver = {
 		.of_match_table = tah_match,
 	},
 	.probe = tah_probe,
-	.remove = tah_remove,
 };
 
 int __init tah_init(void)
diff --git a/drivers/net/ethernet/ibm/emac/tah.h b/drivers/net/ethernet/ibm/emac/tah.h
index 4d5f336f07b3..86c2b6b9d460 100644
--- a/drivers/net/ethernet/ibm/emac/tah.h
+++ b/drivers/net/ethernet/ibm/emac/tah.h
@@ -1,3 +1,4 @@
+/* SPDX-License-Identifier: GPL-2.0-or-later */
 /*
  * drivers/net/ethernet/ibm/emac/tah.h
  *
@@ -12,11 +13,6 @@
  * Matt Porter <mporter@kernel.crashing.org>
  *
  * Copyright (c) 2005 Eugene Surovegin <ebs@ebshome.net>
- *
- * This program is free software; you can redistribute  it and/or modify it
- * under  the terms of  the GNU General  Public License as published by the
- * Free Software Foundation;  either version 2 of the  License, or (at your
- * option) any later version.
  */
 
 #ifndef __IBM_NEWEMAC_TAH_H
diff --git a/drivers/net/ethernet/ibm/emac/zmii.c b/drivers/net/ethernet/ibm/emac/zmii.c
index 8727b865ea02..69ca6065de1c 100644
--- a/drivers/net/ethernet/ibm/emac/zmii.c
+++ b/drivers/net/ethernet/ibm/emac/zmii.c
@@ -1,3 +1,4 @@
+// SPDX-License-Identifier: GPL-2.0-or-later
 /*
  * drivers/net/ethernet/ibm/emac/zmii.c
  *
@@ -14,17 +15,13 @@
  * Based on original work by
  *      Armin Kuster <akuster@mvista.com>
  * 	Copyright 2001 MontaVista Softare Inc.
- *
- * This program is free software; you can redistribute  it and/or modify it
- * under  the terms of  the GNU General  Public License as published by the
- * Free Software Foundation;  either version 2 of the  License, or (at your
- * option) any later version.
- *
  */
 #include <linux/slab.h>
 #include <linux/kernel.h>
 #include <linux/ethtool.h>
+#include <linux/mod_devicetable.h>
 #include <linux/of_address.h>
+#include <linux/platform_device.h>
 #include <asm/io.h>
 
 #include "emac.h"
@@ -49,20 +46,20 @@
  */
 static inline int zmii_valid_mode(int mode)
 {
-	return  mode == PHY_MODE_MII ||
-		mode == PHY_MODE_RMII ||
-		mode == PHY_MODE_SMII ||
-		mode == PHY_MODE_NA;
+	return  mode == PHY_INTERFACE_MODE_MII ||
+		mode == PHY_INTERFACE_MODE_RMII ||
+		mode == PHY_INTERFACE_MODE_SMII ||
+		mode == PHY_INTERFACE_MODE_NA;
 }
 
 static inline const char *zmii_mode_name(int mode)
 {
 	switch (mode) {
-	case PHY_MODE_MII:
+	case PHY_INTERFACE_MODE_MII:
 		return "MII";
-	case PHY_MODE_RMII:
+	case PHY_INTERFACE_MODE_RMII:
 		return "RMII";
-	case PHY_MODE_SMII:
+	case PHY_INTERFACE_MODE_SMII:
 		return "SMII";
 	default:
 		BUG();
@@ -72,18 +69,19 @@ static inline const char *zmii_mode_name(int mode)
 static inline u32 zmii_mode_mask(int mode, int input)
 {
 	switch (mode) {
-	case PHY_MODE_MII:
+	case PHY_INTERFACE_MODE_MII:
 		return ZMII_FER_MII(input);
-	case PHY_MODE_RMII:
+	case PHY_INTERFACE_MODE_RMII:
 		return ZMII_FER_RMII(input);
-	case PHY_MODE_SMII:
+	case PHY_INTERFACE_MODE_SMII:
 		return ZMII_FER_SMII(input);
 	default:
 		return 0;
 	}
 }
 
-int zmii_attach(struct platform_device *ofdev, int input, int *mode)
+int zmii_attach(struct platform_device *ofdev, int input,
+		phy_interface_t *mode)
 {
 	struct zmii_instance *dev = platform_get_drvdata(ofdev);
 	struct zmii_regs __iomem *p = dev->base;
@@ -106,30 +104,30 @@ int zmii_attach(struct platform_device *ofdev, int input, int *mode)
 	 * Please, always specify PHY mode in your board port to avoid
 	 * any surprises.
 	 */
-	if (dev->mode == PHY_MODE_NA) {
-		if (*mode == PHY_MODE_NA) {
+	if (dev->mode == PHY_INTERFACE_MODE_NA) {
+		if (*mode == PHY_INTERFACE_MODE_NA) {
 			u32 r = dev->fer_save;
 
 			ZMII_DBG(dev, "autodetecting mode, FER = 0x%08x" NL, r);
 
 			if (r & (ZMII_FER_MII(0) | ZMII_FER_MII(1)))
-				dev->mode = PHY_MODE_MII;
+				dev->mode = PHY_INTERFACE_MODE_MII;
 			else if (r & (ZMII_FER_RMII(0) | ZMII_FER_RMII(1)))
-				dev->mode = PHY_MODE_RMII;
+				dev->mode = PHY_INTERFACE_MODE_RMII;
 			else
-				dev->mode = PHY_MODE_SMII;
-		} else
+				dev->mode = PHY_INTERFACE_MODE_SMII;
+		} else {
 			dev->mode = *mode;
-
-		printk(KERN_NOTICE "%s: bridge in %s mode\n",
-		       ofdev->dev.of_node->full_name,
+		}
+		printk(KERN_NOTICE "%pOF: bridge in %s mode\n",
+		       ofdev->dev.of_node,
 		       zmii_mode_name(dev->mode));
 	} else {
 		/* All inputs must use the same mode */
-		if (*mode != PHY_MODE_NA && *mode != dev->mode) {
+		if (*mode != PHY_INTERFACE_MODE_NA && *mode != dev->mode) {
 			printk(KERN_ERR
-			       "%s: invalid mode %d specified for input %d\n",
-			       ofdev->dev.of_node->full_name, *mode, input);
+			       "%pOF: invalid mode %d specified for input %d\n",
+			       ofdev->dev.of_node, *mode, input);
 			mutex_unlock(&dev->lock);
 			return -EINVAL;
 		}
@@ -234,34 +232,25 @@ void *zmii_dump_regs(struct platform_device *ofdev, void *buf)
 
 static int zmii_probe(struct platform_device *ofdev)
 {
-	struct device_node *np = ofdev->dev.of_node;
 	struct zmii_instance *dev;
-	struct resource regs;
-	int rc;
+	int err;
 
-	rc = -ENOMEM;
-	dev = kzalloc(sizeof(struct zmii_instance), GFP_KERNEL);
-	if (dev == NULL)
-		goto err_gone;
+	dev = devm_kzalloc(&ofdev->dev, sizeof(struct zmii_instance),
+			   GFP_KERNEL);
+	if (!dev)
+		return -ENOMEM;
 
-	mutex_init(&dev->lock);
-	dev->ofdev = ofdev;
-	dev->mode = PHY_MODE_NA;
+	err = devm_mutex_init(&ofdev->dev, &dev->lock);
+	if (err)
+		return err;
 
-	rc = -ENXIO;
-	if (of_address_to_resource(np, 0, &regs)) {
-		printk(KERN_ERR "%s: Can't get registers address\n",
-		       np->full_name);
-		goto err_free;
-	}
+	dev->ofdev = ofdev;
+	dev->mode = PHY_INTERFACE_MODE_NA;
 
-	rc = -ENOMEM;
-	dev->base = (struct zmii_regs __iomem *)ioremap(regs.start,
-						sizeof(struct zmii_regs));
-	if (dev->base == NULL) {
-		printk(KERN_ERR "%s: Can't map device registers!\n",
-		       np->full_name);
-		goto err_free;
+	dev->base = devm_platform_ioremap_resource(ofdev, 0);
+	if (IS_ERR(dev->base)) {
+		dev_err(&ofdev->dev, "can't map device registers");
+		return PTR_ERR(dev->base);
 	}
 
 	/* We may need FER value for autodetection later */
@@ -270,29 +259,11 @@ static int zmii_probe(struct platform_device *ofdev)
 	/* Disable all inputs by default */
 	out_be32(&dev->base->fer, 0);
 
-	printk(KERN_INFO
-	       "ZMII %s initialized\n", ofdev->dev.of_node->full_name);
+	printk(KERN_INFO "ZMII %pOF initialized\n", ofdev->dev.of_node);
 	wmb();
 	platform_set_drvdata(ofdev, dev);
 
 	return 0;
-
- err_free:
-	kfree(dev);
- err_gone:
-	return rc;
-}
-
-static int zmii_remove(struct platform_device *ofdev)
-{
-	struct zmii_instance *dev = platform_get_drvdata(ofdev);
-
-	WARN_ON(dev->users != 0);
-
-	iounmap(dev->base);
-	kfree(dev);
-
-	return 0;
 }
 
 static const struct of_device_id zmii_match[] =
@@ -313,7 +284,6 @@ static struct platform_driver zmii_driver = {
 		.of_match_table = zmii_match,
 	},
 	.probe = zmii_probe,
-	.remove = zmii_remove,
 };
 
 int __init zmii_init(void)
diff --git a/drivers/net/ethernet/ibm/emac/zmii.h b/drivers/net/ethernet/ibm/emac/zmii.h
index 0959c55b1459..65daedc78594 100644
--- a/drivers/net/ethernet/ibm/emac/zmii.h
+++ b/drivers/net/ethernet/ibm/emac/zmii.h
@@ -1,3 +1,4 @@
+/* SPDX-License-Identifier: GPL-2.0-or-later */
 /*
  * drivers/net/ethernet/ibm/emac/zmii.h
  *
@@ -14,12 +15,6 @@
  * Based on original work by
  *      Armin Kuster <akuster@mvista.com>
  * 	Copyright 2001 MontaVista Softare Inc.
- *
- * This program is free software; you can redistribute  it and/or modify it
- * under  the terms of  the GNU General  Public License as published by the
- * Free Software Foundation;  either version 2 of the  License, or (at your
- * option) any later version.
- *
  */
 #ifndef __IBM_NEWEMAC_ZMII_H
 #define __IBM_NEWEMAC_ZMII_H
@@ -55,7 +50,8 @@ struct zmii_instance {
 
 int zmii_init(void);
 void zmii_exit(void);
-int zmii_attach(struct platform_device *ofdev, int input, int *mode);
+int zmii_attach(struct platform_device *ofdev, int input,
+		phy_interface_t *mode);
 void zmii_detach(struct platform_device *ofdev, int input);
 void zmii_get_mdio(struct platform_device *ofdev, int input);
 void zmii_put_mdio(struct platform_device *ofdev, int input);
diff --git a/drivers/net/ethernet/ibm/ibmveth.c b/drivers/net/ethernet/ibm/ibmveth.c
index 9a74c4e2e193..6f0821f1e798 100644
--- a/drivers/net/ethernet/ibm/ibmveth.c
+++ b/drivers/net/ethernet/ibm/ibmveth.c
@@ -1,19 +1,7 @@
+// SPDX-License-Identifier: GPL-2.0-or-later
 /*
  * IBM Power Virtual Ethernet Device Driver
  *
- * This program is free software; you can redistribute it and/or modify
- * it under the terms of the GNU General Public License as published by
- * the Free Software Foundation; either version 2 of the License, or
- * (at your option) any later version.
- *
- * This program is distributed in the hope that it will be useful,
- * but WITHOUT ANY WARRANTY; without even the implied warranty of
- * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
- * GNU General Public License for more details.
- *
- * You should have received a copy of the GNU General Public License
- * along with this program; if not, see <http://www.gnu.org/licenses/>.
- *
  * Copyright (C) IBM Corporation, 2003, 2010
  *
  * Authors: Dave Larson <larson1@us.ibm.com>
@@ -24,7 +12,6 @@
  */
 
 #include <linux/module.h>
-#include <linux/moduleparam.h>
 #include <linux/types.h>
 #include <linux/errno.h>
 #include <linux/dma-mapping.h>
@@ -52,7 +39,6 @@
 #include "ibmveth.h"
 
 static irqreturn_t ibmveth_interrupt(int irq, void *dev_instance);
-static void ibmveth_rxq_harvest_buffer(struct ibmveth_adapter *adapter);
 static unsigned long ibmveth_get_desired_dma(struct vio_dev *vdev);
 
 static struct kobj_type ktype_veth_pool;
@@ -82,7 +68,7 @@ module_param(rx_flush, uint, 0644);
 MODULE_PARM_DESC(rx_flush, "Flush receive buffers before use");
 
 static bool old_large_send __read_mostly;
-module_param(old_large_send, bool, S_IRUGO);
+module_param(old_large_send, bool, 0444);
 MODULE_PARM_DESC(old_large_send,
 	"Use old large send method on firmware that supports the new method");
 
@@ -94,7 +80,7 @@ struct ibmveth_stat {
 #define IBMVETH_STAT_OFF(stat) offsetof(struct ibmveth_adapter, stat)
 #define IBMVETH_GET_STAT(a, off) *((u64 *)(((unsigned long)(a)) + off))
 
-struct ibmveth_stat ibmveth_stats[] = {
+static struct ibmveth_stat ibmveth_stats[] = {
 	{ "replenish_task_cycles", IBMVETH_STAT_OFF(replenish_task_cycles) },
 	{ "replenish_no_mem", IBMVETH_STAT_OFF(replenish_no_mem) },
 	{ "replenish_add_buff_failure",
@@ -154,6 +140,13 @@ static inline int ibmveth_rxq_csum_good(struct ibmveth_adapter *adapter)
 	return ibmveth_rxq_flags(adapter) & IBMVETH_RXQ_CSUM_GOOD;
 }
 
+static unsigned int ibmveth_real_max_tx_queues(void)
+{
+	unsigned int n_cpu = num_online_cpus();
+
+	return min(n_cpu, IBMVETH_MAX_QUEUES);
+}
+
 /* setup the initial settings for a buffer pool */
 static void ibmveth_init_buffer_pool(struct ibmveth_buff_pool *pool,
 				     u32 pool_index, u32 pool_size,
@@ -171,7 +164,7 @@ static int ibmveth_alloc_buffer_pool(struct ibmveth_buff_pool *pool)
 {
 	int i;
 
-	pool->free_map = kmalloc(sizeof(u16) * pool->size, GFP_KERNEL);
+	pool->free_map = kmalloc_array(pool->size, sizeof(u16), GFP_KERNEL);
 
 	if (!pool->free_map)
 		return -1;
@@ -209,7 +202,7 @@ static inline void ibmveth_flush_buffer(void *addr, unsigned long length)
 	unsigned long offset;
 
 	for (offset = 0; offset < length; offset += SMP_CACHE_BYTES)
-		asm("dcbfl %0,%1" :: "b" (addr), "r" (offset));
+		asm("dcbf %0,%1,1" :: "b" (addr), "r" (offset));
 }
 
 /* replenish the buffers for a pool.  note that we don't need to
@@ -218,88 +211,169 @@ static inline void ibmveth_flush_buffer(void *addr, unsigned long length)
 static void ibmveth_replenish_buffer_pool(struct ibmveth_adapter *adapter,
 					  struct ibmveth_buff_pool *pool)
 {
-	u32 i;
-	u32 count = pool->size - atomic_read(&pool->available);
-	u32 buffers_added = 0;
-	struct sk_buff *skb;
-	unsigned int free_index, index;
-	u64 correlator;
+	union ibmveth_buf_desc descs[IBMVETH_MAX_RX_PER_HCALL] = {0};
+	u32 remaining = pool->size - atomic_read(&pool->available);
+	u64 correlators[IBMVETH_MAX_RX_PER_HCALL] = {0};
 	unsigned long lpar_rc;
+	u32 buffers_added = 0;
+	u32 i, filled, batch;
+	struct vio_dev *vdev;
 	dma_addr_t dma_addr;
+	struct device *dev;
+	u32 index;
 
-	mb();
+	vdev = adapter->vdev;
+	dev = &vdev->dev;
 
-	for (i = 0; i < count; ++i) {
-		union ibmveth_buf_desc desc;
+	mb();
 
-		skb = netdev_alloc_skb(adapter->netdev, pool->buff_size);
+	batch = adapter->rx_buffers_per_hcall;
+
+	while (remaining > 0) {
+		unsigned int free_index = pool->consumer_index;
+
+		/* Fill a batch of descriptors */
+		for (filled = 0; filled < min(remaining, batch); filled++) {
+			index = pool->free_map[free_index];
+			if (WARN_ON(index == IBM_VETH_INVALID_MAP)) {
+				adapter->replenish_add_buff_failure++;
+				netdev_info(adapter->netdev,
+					    "Invalid map index %u, reset\n",
+					    index);
+				schedule_work(&adapter->work);
+				break;
+			}
 
-		if (!skb) {
-			netdev_dbg(adapter->netdev,
-				   "replenish: unable to allocate skb\n");
-			adapter->replenish_no_mem++;
-			break;
-		}
+			if (!pool->skbuff[index]) {
+				struct sk_buff *skb = NULL;
 
-		free_index = pool->consumer_index;
-		pool->consumer_index++;
-		if (pool->consumer_index >= pool->size)
-			pool->consumer_index = 0;
-		index = pool->free_map[free_index];
+				skb = netdev_alloc_skb(adapter->netdev,
+						       pool->buff_size);
+				if (!skb) {
+					adapter->replenish_no_mem++;
+					adapter->replenish_add_buff_failure++;
+					break;
+				}
 
-		BUG_ON(index == IBM_VETH_INVALID_MAP);
-		BUG_ON(pool->skbuff[index] != NULL);
+				dma_addr = dma_map_single(dev, skb->data,
+							  pool->buff_size,
+							  DMA_FROM_DEVICE);
+				if (dma_mapping_error(dev, dma_addr)) {
+					dev_kfree_skb_any(skb);
+					adapter->replenish_add_buff_failure++;
+					break;
+				}
 
-		dma_addr = dma_map_single(&adapter->vdev->dev, skb->data,
-				pool->buff_size, DMA_FROM_DEVICE);
+				pool->dma_addr[index] = dma_addr;
+				pool->skbuff[index] = skb;
+			} else {
+				/* re-use case */
+				dma_addr = pool->dma_addr[index];
+			}
 
-		if (dma_mapping_error(&adapter->vdev->dev, dma_addr))
-			goto failure;
+			if (rx_flush) {
+				unsigned int len;
 
-		pool->free_map[free_index] = IBM_VETH_INVALID_MAP;
-		pool->dma_addr[index] = dma_addr;
-		pool->skbuff[index] = skb;
+				len = adapter->netdev->mtu + IBMVETH_BUFF_OH;
+				len = min(pool->buff_size, len);
+				ibmveth_flush_buffer(pool->skbuff[index]->data,
+						     len);
+			}
 
-		correlator = ((u64)pool->index << 32) | index;
-		*(u64 *)skb->data = correlator;
+			descs[filled].fields.flags_len = IBMVETH_BUF_VALID |
+							  pool->buff_size;
+			descs[filled].fields.address = dma_addr;
 
-		desc.fields.flags_len = IBMVETH_BUF_VALID | pool->buff_size;
-		desc.fields.address = dma_addr;
+			correlators[filled] = ((u64)pool->index << 32) | index;
+			*(u64 *)pool->skbuff[index]->data = correlators[filled];
 
-		if (rx_flush) {
-			unsigned int len = min(pool->buff_size,
-						adapter->netdev->mtu +
-						IBMVETH_BUFF_OH);
-			ibmveth_flush_buffer(skb->data, len);
+			free_index++;
+			if (free_index >= pool->size)
+				free_index = 0;
 		}
-		lpar_rc = h_add_logical_lan_buffer(adapter->vdev->unit_address,
-						   desc.desc);
 
+		if (!filled)
+			break;
+
+		/* single buffer case*/
+		if (filled == 1)
+			lpar_rc = h_add_logical_lan_buffer(vdev->unit_address,
+							   descs[0].desc);
+		else
+			/* Multi-buffer hcall */
+			lpar_rc = h_add_logical_lan_buffers(vdev->unit_address,
+							    descs[0].desc,
+							    descs[1].desc,
+							    descs[2].desc,
+							    descs[3].desc,
+							    descs[4].desc,
+							    descs[5].desc,
+							    descs[6].desc,
+							    descs[7].desc);
 		if (lpar_rc != H_SUCCESS) {
-			goto failure;
-		} else {
-			buffers_added++;
-			adapter->replenish_add_buff_success++;
+			dev_warn_ratelimited(dev,
+					     "RX h_add_logical_lan failed: filled=%u, rc=%lu, batch=%u\n",
+					     filled, lpar_rc, batch);
+			goto hcall_failure;
 		}
-	}
 
-	mb();
-	atomic_add(buffers_added, &(pool->available));
-	return;
+		/* Only update pool state after hcall succeeds */
+		for (i = 0; i < filled; i++) {
+			free_index = pool->consumer_index;
+			pool->free_map[free_index] = IBM_VETH_INVALID_MAP;
 
-failure:
-	pool->free_map[free_index] = index;
-	pool->skbuff[index] = NULL;
-	if (pool->consumer_index == 0)
-		pool->consumer_index = pool->size - 1;
-	else
-		pool->consumer_index--;
-	if (!dma_mapping_error(&adapter->vdev->dev, dma_addr))
-		dma_unmap_single(&adapter->vdev->dev,
-		                 pool->dma_addr[index], pool->buff_size,
-		                 DMA_FROM_DEVICE);
-	dev_kfree_skb_any(skb);
-	adapter->replenish_add_buff_failure++;
+			pool->consumer_index++;
+			if (pool->consumer_index >= pool->size)
+				pool->consumer_index = 0;
+		}
+
+		buffers_added += filled;
+		adapter->replenish_add_buff_success += filled;
+		remaining -= filled;
+
+		memset(&descs, 0, sizeof(descs));
+		memset(&correlators, 0, sizeof(correlators));
+		continue;
+
+hcall_failure:
+		for (i = 0; i < filled; i++) {
+			index = correlators[i] & 0xffffffffUL;
+			dma_addr =  pool->dma_addr[index];
+
+			if (pool->skbuff[index]) {
+				if (dma_addr &&
+				    !dma_mapping_error(dev, dma_addr))
+					dma_unmap_single(dev, dma_addr,
+							 pool->buff_size,
+							 DMA_FROM_DEVICE);
+
+				dev_kfree_skb_any(pool->skbuff[index]);
+				pool->skbuff[index] = NULL;
+			}
+		}
+		adapter->replenish_add_buff_failure += filled;
+
+		/*
+		 * If multi rx buffers hcall is no longer supported by FW
+		 * e.g. in the case of Live Parttion Migration
+		 */
+		if (batch > 1 && lpar_rc == H_FUNCTION) {
+			/*
+			 * Instead of retry submit single buffer individually
+			 * here just set the max rx buffer per hcall to 1
+			 * buffers will be respleshed next time
+			 * when ibmveth_replenish_buffer_pool() is called again
+			 * with single-buffer case
+			 */
+			netdev_info(adapter->netdev,
+				    "RX Multi buffers not supported by FW, rc=%lu\n",
+				    lpar_rc);
+			adapter->rx_buffers_per_hcall = 1;
+			netdev_info(adapter->netdev,
+				    "Next rx replesh will fall back to single-buffer hcall\n");
+		}
+		break;
+	}
 
 	mb();
 	atomic_add(buffers_added, &(pool->available));
@@ -369,28 +443,52 @@ static void ibmveth_free_buffer_pool(struct ibmveth_adapter *adapter,
 	}
 }
 
-/* remove a buffer from a pool */
-static void ibmveth_remove_buffer_from_pool(struct ibmveth_adapter *adapter,
-					    u64 correlator)
+/**
+ * ibmveth_remove_buffer_from_pool - remove a buffer from a pool
+ * @adapter: adapter instance
+ * @correlator: identifies pool and index
+ * @reuse: whether to reuse buffer
+ *
+ * Return:
+ * * %0       - success
+ * * %-EINVAL - correlator maps to pool or index out of range
+ * * %-EFAULT - pool and index map to null skb
+ */
+static int ibmveth_remove_buffer_from_pool(struct ibmveth_adapter *adapter,
+					   u64 correlator, bool reuse)
 {
 	unsigned int pool  = correlator >> 32;
 	unsigned int index = correlator & 0xffffffffUL;
 	unsigned int free_index;
 	struct sk_buff *skb;
 
-	BUG_ON(pool >= IBMVETH_NUM_BUFF_POOLS);
-	BUG_ON(index >= adapter->rx_buff_pool[pool].size);
+	if (WARN_ON(pool >= IBMVETH_NUM_BUFF_POOLS) ||
+	    WARN_ON(index >= adapter->rx_buff_pool[pool].size)) {
+		schedule_work(&adapter->work);
+		return -EINVAL;
+	}
 
 	skb = adapter->rx_buff_pool[pool].skbuff[index];
+	if (WARN_ON(!skb)) {
+		schedule_work(&adapter->work);
+		return -EFAULT;
+	}
 
-	BUG_ON(skb == NULL);
-
-	adapter->rx_buff_pool[pool].skbuff[index] = NULL;
+	/* if we are going to reuse the buffer then keep the pointers around
+	 * but mark index as available. replenish will see the skb pointer and
+	 * assume it is to be recycled.
+	 */
+	if (!reuse) {
+		/* remove the skb pointer to mark free. actual freeing is done
+		 * by upper level networking after gro_recieve
+		 */
+		adapter->rx_buff_pool[pool].skbuff[index] = NULL;
 
-	dma_unmap_single(&adapter->vdev->dev,
-			 adapter->rx_buff_pool[pool].dma_addr[index],
-			 adapter->rx_buff_pool[pool].buff_size,
-			 DMA_FROM_DEVICE);
+		dma_unmap_single(&adapter->vdev->dev,
+				 adapter->rx_buff_pool[pool].dma_addr[index],
+				 adapter->rx_buff_pool[pool].buff_size,
+				 DMA_FROM_DEVICE);
+	}
 
 	free_index = adapter->rx_buff_pool[pool].producer_index;
 	adapter->rx_buff_pool[pool].producer_index++;
@@ -402,6 +500,8 @@ static void ibmveth_remove_buffer_from_pool(struct ibmveth_adapter *adapter,
 	mb();
 
 	atomic_dec(&(adapter->rx_buff_pool[pool].available));
+
+	return 0;
 }
 
 /* get the current buffer on the rx queue */
@@ -411,112 +511,76 @@ static inline struct sk_buff *ibmveth_rxq_get_buffer(struct ibmveth_adapter *ada
 	unsigned int pool = correlator >> 32;
 	unsigned int index = correlator & 0xffffffffUL;
 
-	BUG_ON(pool >= IBMVETH_NUM_BUFF_POOLS);
-	BUG_ON(index >= adapter->rx_buff_pool[pool].size);
+	if (WARN_ON(pool >= IBMVETH_NUM_BUFF_POOLS) ||
+	    WARN_ON(index >= adapter->rx_buff_pool[pool].size)) {
+		schedule_work(&adapter->work);
+		return NULL;
+	}
 
 	return adapter->rx_buff_pool[pool].skbuff[index];
 }
 
-/* recycle the current buffer on the rx queue */
-static int ibmveth_rxq_recycle_buffer(struct ibmveth_adapter *adapter)
+/**
+ * ibmveth_rxq_harvest_buffer - Harvest buffer from pool
+ *
+ * @adapter: pointer to adapter
+ * @reuse:   whether to reuse buffer
+ *
+ * Context: called from ibmveth_poll
+ *
+ * Return:
+ * * %0    - success
+ * * other - non-zero return from ibmveth_remove_buffer_from_pool
+ */
+static int ibmveth_rxq_harvest_buffer(struct ibmveth_adapter *adapter,
+				      bool reuse)
 {
-	u32 q_index = adapter->rx_queue.index;
-	u64 correlator = adapter->rx_queue.queue_addr[q_index].correlator;
-	unsigned int pool = correlator >> 32;
-	unsigned int index = correlator & 0xffffffffUL;
-	union ibmveth_buf_desc desc;
-	unsigned long lpar_rc;
-	int ret = 1;
-
-	BUG_ON(pool >= IBMVETH_NUM_BUFF_POOLS);
-	BUG_ON(index >= adapter->rx_buff_pool[pool].size);
-
-	if (!adapter->rx_buff_pool[pool].active) {
-		ibmveth_rxq_harvest_buffer(adapter);
-		ibmveth_free_buffer_pool(adapter, &adapter->rx_buff_pool[pool]);
-		goto out;
-	}
-
-	desc.fields.flags_len = IBMVETH_BUF_VALID |
-		adapter->rx_buff_pool[pool].buff_size;
-	desc.fields.address = adapter->rx_buff_pool[pool].dma_addr[index];
-
-	lpar_rc = h_add_logical_lan_buffer(adapter->vdev->unit_address, desc.desc);
+	u64 cor;
+	int rc;
 
-	if (lpar_rc != H_SUCCESS) {
-		netdev_dbg(adapter->netdev, "h_add_logical_lan_buffer failed "
-			   "during recycle rc=%ld", lpar_rc);
-		ibmveth_remove_buffer_from_pool(adapter, adapter->rx_queue.queue_addr[adapter->rx_queue.index].correlator);
-		ret = 0;
-	}
+	cor = adapter->rx_queue.queue_addr[adapter->rx_queue.index].correlator;
+	rc = ibmveth_remove_buffer_from_pool(adapter, cor, reuse);
+	if (unlikely(rc))
+		return rc;
 
 	if (++adapter->rx_queue.index == adapter->rx_queue.num_slots) {
 		adapter->rx_queue.index = 0;
 		adapter->rx_queue.toggle = !adapter->rx_queue.toggle;
 	}
 
-out:
-	return ret;
+	return 0;
 }
 
-static void ibmveth_rxq_harvest_buffer(struct ibmveth_adapter *adapter)
+static void ibmveth_free_tx_ltb(struct ibmveth_adapter *adapter, int idx)
 {
-	ibmveth_remove_buffer_from_pool(adapter, adapter->rx_queue.queue_addr[adapter->rx_queue.index].correlator);
-
-	if (++adapter->rx_queue.index == adapter->rx_queue.num_slots) {
-		adapter->rx_queue.index = 0;
-		adapter->rx_queue.toggle = !adapter->rx_queue.toggle;
-	}
+	dma_unmap_single(&adapter->vdev->dev, adapter->tx_ltb_dma[idx],
+			 adapter->tx_ltb_size, DMA_TO_DEVICE);
+	kfree(adapter->tx_ltb_ptr[idx]);
+	adapter->tx_ltb_ptr[idx] = NULL;
 }
 
-static void ibmveth_cleanup(struct ibmveth_adapter *adapter)
+static int ibmveth_allocate_tx_ltb(struct ibmveth_adapter *adapter, int idx)
 {
-	int i;
-	struct device *dev = &adapter->vdev->dev;
-
-	if (adapter->buffer_list_addr != NULL) {
-		if (!dma_mapping_error(dev, adapter->buffer_list_dma)) {
-			dma_unmap_single(dev, adapter->buffer_list_dma, 4096,
-					DMA_BIDIRECTIONAL);
-			adapter->buffer_list_dma = DMA_ERROR_CODE;
-		}
-		free_page((unsigned long)adapter->buffer_list_addr);
-		adapter->buffer_list_addr = NULL;
-	}
-
-	if (adapter->filter_list_addr != NULL) {
-		if (!dma_mapping_error(dev, adapter->filter_list_dma)) {
-			dma_unmap_single(dev, adapter->filter_list_dma, 4096,
-					DMA_BIDIRECTIONAL);
-			adapter->filter_list_dma = DMA_ERROR_CODE;
-		}
-		free_page((unsigned long)adapter->filter_list_addr);
-		adapter->filter_list_addr = NULL;
+	adapter->tx_ltb_ptr[idx] = kzalloc(adapter->tx_ltb_size,
+					   GFP_KERNEL);
+	if (!adapter->tx_ltb_ptr[idx]) {
+		netdev_err(adapter->netdev,
+			   "unable to allocate tx long term buffer\n");
+		return -ENOMEM;
 	}
-
-	if (adapter->rx_queue.queue_addr != NULL) {
-		dma_free_coherent(dev, adapter->rx_queue.queue_len,
-				  adapter->rx_queue.queue_addr,
-				  adapter->rx_queue.queue_dma);
-		adapter->rx_queue.queue_addr = NULL;
+	adapter->tx_ltb_dma[idx] = dma_map_single(&adapter->vdev->dev,
+						  adapter->tx_ltb_ptr[idx],
+						  adapter->tx_ltb_size,
+						  DMA_TO_DEVICE);
+	if (dma_mapping_error(&adapter->vdev->dev, adapter->tx_ltb_dma[idx])) {
+		netdev_err(adapter->netdev,
+			   "unable to DMA map tx long term buffer\n");
+		kfree(adapter->tx_ltb_ptr[idx]);
+		adapter->tx_ltb_ptr[idx] = NULL;
+		return -ENOMEM;
 	}
 
-	for (i = 0; i < IBMVETH_NUM_BUFF_POOLS; i++)
-		if (adapter->rx_buff_pool[i].active)
-			ibmveth_free_buffer_pool(adapter,
-						 &adapter->rx_buff_pool[i]);
-
-	if (adapter->bounce_buffer != NULL) {
-		if (!dma_mapping_error(dev, adapter->bounce_buffer_dma)) {
-			dma_unmap_single(&adapter->vdev->dev,
-					adapter->bounce_buffer_dma,
-					adapter->netdev->mtu + IBMVETH_BUFF_OH,
-					DMA_BIDIRECTIONAL);
-			adapter->bounce_buffer_dma = DMA_ERROR_CODE;
-		}
-		kfree(adapter->bounce_buffer);
-		adapter->bounce_buffer = NULL;
-	}
+	return 0;
 }
 
 static int ibmveth_register_logical_lan(struct ibmveth_adapter *adapter,
@@ -546,17 +610,6 @@ retry:
 	return rc;
 }
 
-static u64 ibmveth_encode_mac_addr(u8 *mac)
-{
-	int i;
-	u64 encoded = 0;
-
-	for (i = 0; i < ETH_ALEN; i++)
-		encoded = (encoded << 8) | mac[i];
-
-	return encoded;
-}
-
 static int ibmveth_open(struct net_device *netdev)
 {
 	struct ibmveth_adapter *adapter = netdev_priv(netdev);
@@ -575,14 +628,17 @@ static int ibmveth_open(struct net_device *netdev)
 	for(i = 0; i < IBMVETH_NUM_BUFF_POOLS; i++)
 		rxq_entries += adapter->rx_buff_pool[i].size;
 
+	rc = -ENOMEM;
 	adapter->buffer_list_addr = (void*) get_zeroed_page(GFP_KERNEL);
-	adapter->filter_list_addr = (void*) get_zeroed_page(GFP_KERNEL);
+	if (!adapter->buffer_list_addr) {
+		netdev_err(netdev, "unable to allocate list pages\n");
+		goto out;
+	}
 
-	if (!adapter->buffer_list_addr || !adapter->filter_list_addr) {
-		netdev_err(netdev, "unable to allocate filter or buffer list "
-			   "pages\n");
-		rc = -ENOMEM;
-		goto err_out;
+	adapter->filter_list_addr = (void*) get_zeroed_page(GFP_KERNEL);
+	if (!adapter->filter_list_addr) {
+		netdev_err(netdev, "unable to allocate filter pages\n");
+		goto out_free_buffer_list;
 	}
 
 	dev = &adapter->vdev->dev;
@@ -592,29 +648,33 @@ static int ibmveth_open(struct net_device *netdev)
 	adapter->rx_queue.queue_addr =
 		dma_alloc_coherent(dev, adapter->rx_queue.queue_len,
 				   &adapter->rx_queue.queue_dma, GFP_KERNEL);
-	if (!adapter->rx_queue.queue_addr) {
-		rc = -ENOMEM;
-		goto err_out;
-	}
+	if (!adapter->rx_queue.queue_addr)
+		goto out_free_filter_list;
 
 	adapter->buffer_list_dma = dma_map_single(dev,
 			adapter->buffer_list_addr, 4096, DMA_BIDIRECTIONAL);
+	if (dma_mapping_error(dev, adapter->buffer_list_dma)) {
+		netdev_err(netdev, "unable to map buffer list pages\n");
+		goto out_free_queue_mem;
+	}
+
 	adapter->filter_list_dma = dma_map_single(dev,
 			adapter->filter_list_addr, 4096, DMA_BIDIRECTIONAL);
+	if (dma_mapping_error(dev, adapter->filter_list_dma)) {
+		netdev_err(netdev, "unable to map filter list pages\n");
+		goto out_unmap_buffer_list;
+	}
 
-	if ((dma_mapping_error(dev, adapter->buffer_list_dma)) ||
-	    (dma_mapping_error(dev, adapter->filter_list_dma))) {
-		netdev_err(netdev, "unable to map filter or buffer list "
-			   "pages\n");
-		rc = -ENOMEM;
-		goto err_out;
+	for (i = 0; i < netdev->real_num_tx_queues; i++) {
+		if (ibmveth_allocate_tx_ltb(adapter, i))
+			goto out_free_tx_ltb;
 	}
 
 	adapter->rx_queue.index = 0;
 	adapter->rx_queue.num_slots = rxq_entries;
 	adapter->rx_queue.toggle = 1;
 
-	mac_address = ibmveth_encode_mac_addr(netdev->dev_addr);
+	mac_address = ether_addr_to_u64(netdev->dev_addr);
 
 	rxq_desc.fields.flags_len = IBMVETH_BUF_VALID |
 					adapter->rx_queue.queue_len;
@@ -638,7 +698,7 @@ static int ibmveth_open(struct net_device *netdev)
 				     rxq_desc.desc,
 				     mac_address);
 		rc = -ENONET;
-		goto err_out;
+		goto out_unmap_filter_list;
 	}
 
 	for (i = 0; i < IBMVETH_NUM_BUFF_POOLS; i++) {
@@ -648,7 +708,7 @@ static int ibmveth_open(struct net_device *netdev)
 			netdev_err(netdev, "unable to alloc pool\n");
 			adapter->rx_buff_pool[i].active = 0;
 			rc = -ENOMEM;
-			goto err_out;
+			goto out_free_buffer_pools;
 		}
 	}
 
@@ -662,37 +722,47 @@ static int ibmveth_open(struct net_device *netdev)
 			lpar_rc = h_free_logical_lan(adapter->vdev->unit_address);
 		} while (H_IS_LONG_BUSY(lpar_rc) || (lpar_rc == H_BUSY));
 
-		goto err_out;
+		goto out_free_buffer_pools;
 	}
 
-	adapter->bounce_buffer =
-	    kmalloc(netdev->mtu + IBMVETH_BUFF_OH, GFP_KERNEL);
-	if (!adapter->bounce_buffer) {
-		rc = -ENOMEM;
-		goto err_out_free_irq;
-	}
-	adapter->bounce_buffer_dma =
-	    dma_map_single(&adapter->vdev->dev, adapter->bounce_buffer,
-			   netdev->mtu + IBMVETH_BUFF_OH, DMA_BIDIRECTIONAL);
-	if (dma_mapping_error(dev, adapter->bounce_buffer_dma)) {
-		netdev_err(netdev, "unable to map bounce buffer\n");
-		rc = -ENOMEM;
-		goto err_out_free_irq;
-	}
+	rc = -ENOMEM;
 
 	netdev_dbg(netdev, "initial replenish cycle\n");
 	ibmveth_interrupt(netdev->irq, netdev);
 
-	netif_start_queue(netdev);
+	netif_tx_start_all_queues(netdev);
 
 	netdev_dbg(netdev, "open complete\n");
 
 	return 0;
 
-err_out_free_irq:
-	free_irq(netdev->irq, netdev);
-err_out:
-	ibmveth_cleanup(adapter);
+out_free_buffer_pools:
+	while (--i >= 0) {
+		if (adapter->rx_buff_pool[i].active)
+			ibmveth_free_buffer_pool(adapter,
+						 &adapter->rx_buff_pool[i]);
+	}
+out_unmap_filter_list:
+	dma_unmap_single(dev, adapter->filter_list_dma, 4096,
+			 DMA_BIDIRECTIONAL);
+
+out_free_tx_ltb:
+	while (--i >= 0) {
+		ibmveth_free_tx_ltb(adapter, i);
+	}
+
+out_unmap_buffer_list:
+	dma_unmap_single(dev, adapter->buffer_list_dma, 4096,
+			 DMA_BIDIRECTIONAL);
+out_free_queue_mem:
+	dma_free_coherent(dev, adapter->rx_queue.queue_len,
+			  adapter->rx_queue.queue_addr,
+			  adapter->rx_queue.queue_dma);
+out_free_filter_list:
+	free_page((unsigned long)adapter->filter_list_addr);
+out_free_buffer_list:
+	free_page((unsigned long)adapter->buffer_list_addr);
+out:
 	napi_disable(&adapter->napi);
 	return rc;
 }
@@ -700,14 +770,15 @@ err_out:
 static int ibmveth_close(struct net_device *netdev)
 {
 	struct ibmveth_adapter *adapter = netdev_priv(netdev);
+	struct device *dev = &adapter->vdev->dev;
 	long lpar_rc;
+	int i;
 
 	netdev_dbg(netdev, "close starting\n");
 
 	napi_disable(&adapter->napi);
 
-	if (!adapter->pool_config)
-		netif_stop_queue(netdev);
+	netif_tx_stop_all_queues(netdev);
 
 	h_vio_signal(adapter->vdev->unit_address, VIO_IRQ_DISABLE);
 
@@ -724,41 +795,95 @@ static int ibmveth_close(struct net_device *netdev)
 
 	ibmveth_update_rx_no_buffer(adapter);
 
-	ibmveth_cleanup(adapter);
+	dma_unmap_single(dev, adapter->buffer_list_dma, 4096,
+			 DMA_BIDIRECTIONAL);
+	free_page((unsigned long)adapter->buffer_list_addr);
+
+	dma_unmap_single(dev, adapter->filter_list_dma, 4096,
+			 DMA_BIDIRECTIONAL);
+	free_page((unsigned long)adapter->filter_list_addr);
+
+	dma_free_coherent(dev, adapter->rx_queue.queue_len,
+			  adapter->rx_queue.queue_addr,
+			  adapter->rx_queue.queue_dma);
+
+	for (i = 0; i < IBMVETH_NUM_BUFF_POOLS; i++)
+		if (adapter->rx_buff_pool[i].active)
+			ibmveth_free_buffer_pool(adapter,
+						 &adapter->rx_buff_pool[i]);
+
+	for (i = 0; i < netdev->real_num_tx_queues; i++)
+		ibmveth_free_tx_ltb(adapter, i);
 
 	netdev_dbg(netdev, "close complete\n");
 
 	return 0;
 }
 
-static int netdev_get_link_ksettings(struct net_device *dev,
-				     struct ethtool_link_ksettings *cmd)
+/**
+ * ibmveth_reset - Handle scheduled reset work
+ *
+ * @w: pointer to work_struct embedded in adapter structure
+ *
+ * Context: This routine acquires rtnl_mutex and disables its NAPI through
+ *          ibmveth_close. It can't be called directly in a context that has
+ *          already acquired rtnl_mutex or disabled its NAPI, or directly from
+ *          a poll routine.
+ *
+ * Return: void
+ */
+static void ibmveth_reset(struct work_struct *w)
+{
+	struct ibmveth_adapter *adapter = container_of(w, struct ibmveth_adapter, work);
+	struct net_device *netdev = adapter->netdev;
+
+	netdev_dbg(netdev, "reset starting\n");
+
+	rtnl_lock();
+
+	dev_close(adapter->netdev);
+	dev_open(adapter->netdev, NULL);
+
+	rtnl_unlock();
+
+	netdev_dbg(netdev, "reset complete\n");
+}
+
+static int ibmveth_set_link_ksettings(struct net_device *dev,
+				      const struct ethtool_link_ksettings *cmd)
 {
-	u32 supported, advertising;
-
-	supported = (SUPPORTED_1000baseT_Full | SUPPORTED_Autoneg |
-				SUPPORTED_FIBRE);
-	advertising = (ADVERTISED_1000baseT_Full | ADVERTISED_Autoneg |
-				ADVERTISED_FIBRE);
-	cmd->base.speed = SPEED_1000;
-	cmd->base.duplex = DUPLEX_FULL;
-	cmd->base.port = PORT_FIBRE;
-	cmd->base.phy_address = 0;
-	cmd->base.autoneg = AUTONEG_ENABLE;
-
-	ethtool_convert_legacy_u32_to_link_mode(cmd->link_modes.supported,
-						supported);
-	ethtool_convert_legacy_u32_to_link_mode(cmd->link_modes.advertising,
-						advertising);
+	struct ibmveth_adapter *adapter = netdev_priv(dev);
+
+	return ethtool_virtdev_set_link_ksettings(dev, cmd,
+						  &adapter->speed,
+						  &adapter->duplex);
+}
+
+static int ibmveth_get_link_ksettings(struct net_device *dev,
+				      struct ethtool_link_ksettings *cmd)
+{
+	struct ibmveth_adapter *adapter = netdev_priv(dev);
+
+	cmd->base.speed = adapter->speed;
+	cmd->base.duplex = adapter->duplex;
+	cmd->base.port = PORT_OTHER;
 
 	return 0;
 }
 
+static void ibmveth_init_link_settings(struct net_device *dev)
+{
+	struct ibmveth_adapter *adapter = netdev_priv(dev);
+
+	adapter->speed = SPEED_1000;
+	adapter->duplex = DUPLEX_FULL;
+}
+
 static void netdev_get_drvinfo(struct net_device *dev,
 			       struct ethtool_drvinfo *info)
 {
-	strlcpy(info->driver, ibmveth_driver_name, sizeof(info->driver));
-	strlcpy(info->version, ibmveth_driver_version, sizeof(info->version));
+	strscpy(info->driver, ibmveth_driver_name, sizeof(info->driver));
+	strscpy(info->version, ibmveth_driver_version, sizeof(info->version));
 }
 
 static netdev_features_t ibmveth_fix_features(struct net_device *dev,
@@ -790,9 +915,7 @@ static int ibmveth_set_csum_offload(struct net_device *dev, u32 data)
 
 	if (netif_running(dev)) {
 		restart = 1;
-		adapter->pool_config = 1;
 		ibmveth_close(dev);
-		adapter->pool_config = 0;
 	}
 
 	set_attr = 0;
@@ -874,9 +997,7 @@ static int ibmveth_set_tso(struct net_device *dev, u32 data)
 
 	if (netif_running(dev)) {
 		restart = 1;
-		adapter->pool_config = 1;
 		ibmveth_close(dev);
-		adapter->pool_config = 0;
 	}
 
 	set_attr = 0;
@@ -983,13 +1104,79 @@ static void ibmveth_get_ethtool_stats(struct net_device *dev,
 		data[i] = IBMVETH_GET_STAT(adapter, ibmveth_stats[i].offset);
 }
 
+static void ibmveth_get_channels(struct net_device *netdev,
+				 struct ethtool_channels *channels)
+{
+	channels->max_tx = ibmveth_real_max_tx_queues();
+	channels->tx_count = netdev->real_num_tx_queues;
+
+	channels->max_rx = netdev->real_num_rx_queues;
+	channels->rx_count = netdev->real_num_rx_queues;
+}
+
+static int ibmveth_set_channels(struct net_device *netdev,
+				struct ethtool_channels *channels)
+{
+	struct ibmveth_adapter *adapter = netdev_priv(netdev);
+	unsigned int old = netdev->real_num_tx_queues,
+		     goal = channels->tx_count;
+	int rc, i;
+
+	/* If ndo_open has not been called yet then don't allocate, just set
+	 * desired netdev_queue's and return
+	 */
+	if (!(netdev->flags & IFF_UP))
+		return netif_set_real_num_tx_queues(netdev, goal);
+
+	/* We have IBMVETH_MAX_QUEUES netdev_queue's allocated
+	 * but we may need to alloc/free the ltb's.
+	 */
+	netif_tx_stop_all_queues(netdev);
+
+	/* Allocate any queue that we need */
+	for (i = old; i < goal; i++) {
+		if (adapter->tx_ltb_ptr[i])
+			continue;
+
+		rc = ibmveth_allocate_tx_ltb(adapter, i);
+		if (!rc)
+			continue;
+
+		/* if something goes wrong, free everything we just allocated */
+		netdev_err(netdev, "Failed to allocate more tx queues, returning to %d queues\n",
+			   old);
+		goal = old;
+		old = i;
+		break;
+	}
+	rc = netif_set_real_num_tx_queues(netdev, goal);
+	if (rc) {
+		netdev_err(netdev, "Failed to set real tx queues, returning to %d queues\n",
+			   old);
+		goal = old;
+		old = i;
+	}
+	/* Free any that are no longer needed */
+	for (i = old; i > goal; i--) {
+		if (adapter->tx_ltb_ptr[i - 1])
+			ibmveth_free_tx_ltb(adapter, i - 1);
+	}
+
+	netif_tx_wake_all_queues(netdev);
+
+	return rc;
+}
+
 static const struct ethtool_ops netdev_ethtool_ops = {
-	.get_drvinfo		= netdev_get_drvinfo,
-	.get_link		= ethtool_op_get_link,
-	.get_strings		= ibmveth_get_strings,
-	.get_sset_count		= ibmveth_get_sset_count,
-	.get_ethtool_stats	= ibmveth_get_ethtool_stats,
-	.get_link_ksettings	= netdev_get_link_ksettings,
+	.get_drvinfo		         = netdev_get_drvinfo,
+	.get_link		         = ethtool_op_get_link,
+	.get_strings		         = ibmveth_get_strings,
+	.get_sset_count		         = ibmveth_get_sset_count,
+	.get_ethtool_stats	         = ibmveth_get_ethtool_stats,
+	.get_link_ksettings	         = ibmveth_get_link_ksettings,
+	.set_link_ksettings              = ibmveth_set_link_ksettings,
+	.get_channels			 = ibmveth_get_channels,
+	.set_channels			 = ibmveth_set_channels
 };
 
 static int ibmveth_ioctl(struct net_device *dev, struct ifreq *ifr, int cmd)
@@ -997,10 +1184,8 @@ static int ibmveth_ioctl(struct net_device *dev, struct ifreq *ifr, int cmd)
 	return -EOPNOTSUPP;
 }
 
-#define page_offset(v) ((unsigned long)(v) & ((1 << 12) - 1))
-
 static int ibmveth_send(struct ibmveth_adapter *adapter,
-			union ibmveth_buf_desc *descs, unsigned long mss)
+			unsigned long desc, unsigned long mss)
 {
 	unsigned long correlator;
 	unsigned int retry_count;
@@ -1013,12 +1198,9 @@ static int ibmveth_send(struct ibmveth_adapter *adapter,
 	retry_count = 1024;
 	correlator = 0;
 	do {
-		ret = h_send_logical_lan(adapter->vdev->unit_address,
-					     descs[0].desc, descs[1].desc,
-					     descs[2].desc, descs[3].desc,
-					     descs[4].desc, descs[5].desc,
-					     correlator, &correlator, mss,
-					     adapter->fw_large_send_support);
+		ret = h_send_logical_lan(adapter->vdev->unit_address, desc,
+					 correlator, &correlator, mss,
+					 adapter->fw_large_send_support);
 	} while ((ret == H_BUSY) && (retry_count--));
 
 	if (ret != H_SUCCESS && ret != H_DROPPED) {
@@ -1030,35 +1212,34 @@ static int ibmveth_send(struct ibmveth_adapter *adapter,
 	return 0;
 }
 
+static int ibmveth_is_packet_unsupported(struct sk_buff *skb,
+					 struct net_device *netdev)
+{
+	struct ethhdr *ether_header;
+	int ret = 0;
+
+	ether_header = eth_hdr(skb);
+
+	if (ether_addr_equal(ether_header->h_dest, netdev->dev_addr)) {
+		netdev_dbg(netdev, "veth doesn't support loopback packets, dropping packet.\n");
+		netdev->stats.tx_dropped++;
+		ret = -EOPNOTSUPP;
+	}
+
+	return ret;
+}
+
 static netdev_tx_t ibmveth_start_xmit(struct sk_buff *skb,
 				      struct net_device *netdev)
 {
 	struct ibmveth_adapter *adapter = netdev_priv(netdev);
-	unsigned int desc_flags;
-	union ibmveth_buf_desc descs[6];
-	int last, i;
-	int force_bounce = 0;
-	dma_addr_t dma_addr;
+	unsigned int desc_flags, total_bytes;
+	union ibmveth_buf_desc desc;
+	int i, queue_num = skb_get_queue_mapping(skb);
 	unsigned long mss = 0;
 
-	/* veth doesn't handle frag_list, so linearize the skb.
-	 * When GRO is enabled SKB's can have frag_list.
-	 */
-	if (adapter->is_active_trunk &&
-	    skb_has_frag_list(skb) && __skb_linearize(skb)) {
-		netdev->stats.tx_dropped++;
-		goto out;
-	}
-
-	/*
-	 * veth handles a maximum of 6 segments including the header, so
-	 * we have to linearize the skb if there are more than this.
-	 */
-	if (skb_shinfo(skb)->nr_frags > 5 && __skb_linearize(skb)) {
-		netdev->stats.tx_dropped++;
+	if (ibmveth_is_packet_unsupported(skb, netdev))
 		goto out;
-	}
-
 	/* veth can't checksum offload UDP */
 	if (skb->ip_summed == CHECKSUM_PARTIAL &&
 	    ((skb->protocol == htons(ETH_P_IP) &&
@@ -1088,56 +1269,6 @@ static netdev_tx_t ibmveth_start_xmit(struct sk_buff *skb,
 			desc_flags |= IBMVETH_BUF_LRG_SND;
 	}
 
-retry_bounce:
-	memset(descs, 0, sizeof(descs));
-
-	/*
-	 * If a linear packet is below the rx threshold then
-	 * copy it into the static bounce buffer. This avoids the
-	 * cost of a TCE insert and remove.
-	 */
-	if (force_bounce || (!skb_is_nonlinear(skb) &&
-				(skb->len < tx_copybreak))) {
-		skb_copy_from_linear_data(skb, adapter->bounce_buffer,
-					  skb->len);
-
-		descs[0].fields.flags_len = desc_flags | skb->len;
-		descs[0].fields.address = adapter->bounce_buffer_dma;
-
-		if (ibmveth_send(adapter, descs, 0)) {
-			adapter->tx_send_failed++;
-			netdev->stats.tx_dropped++;
-		} else {
-			netdev->stats.tx_packets++;
-			netdev->stats.tx_bytes += skb->len;
-		}
-
-		goto out;
-	}
-
-	/* Map the header */
-	dma_addr = dma_map_single(&adapter->vdev->dev, skb->data,
-				  skb_headlen(skb), DMA_TO_DEVICE);
-	if (dma_mapping_error(&adapter->vdev->dev, dma_addr))
-		goto map_failed;
-
-	descs[0].fields.flags_len = desc_flags | skb_headlen(skb);
-	descs[0].fields.address = dma_addr;
-
-	/* Map the frags */
-	for (i = 0; i < skb_shinfo(skb)->nr_frags; i++) {
-		const skb_frag_t *frag = &skb_shinfo(skb)->frags[i];
-
-		dma_addr = skb_frag_dma_map(&adapter->vdev->dev, frag, 0,
-					    skb_frag_size(frag), DMA_TO_DEVICE);
-
-		if (dma_mapping_error(&adapter->vdev->dev, dma_addr))
-			goto map_failed_frags;
-
-		descs[i+1].fields.flags_len = desc_flags | skb_frag_size(frag);
-		descs[i+1].fields.address = dma_addr;
-	}
-
 	if (skb->ip_summed == CHECKSUM_PARTIAL && skb_is_gso(skb)) {
 		if (adapter->fw_large_send_support) {
 			mss = (unsigned long)skb_shinfo(skb)->gso_size;
@@ -1154,7 +1285,36 @@ retry_bounce:
 		}
 	}
 
-	if (ibmveth_send(adapter, descs, mss)) {
+	/* Copy header into mapped buffer */
+	if (unlikely(skb->len > adapter->tx_ltb_size)) {
+		netdev_err(adapter->netdev, "tx: packet size (%u) exceeds ltb (%u)\n",
+			   skb->len, adapter->tx_ltb_size);
+		netdev->stats.tx_dropped++;
+		goto out;
+	}
+	memcpy(adapter->tx_ltb_ptr[queue_num], skb->data, skb_headlen(skb));
+	total_bytes = skb_headlen(skb);
+	/* Copy frags into mapped buffers */
+	for (i = 0; i < skb_shinfo(skb)->nr_frags; i++) {
+		const skb_frag_t *frag = &skb_shinfo(skb)->frags[i];
+
+		memcpy(adapter->tx_ltb_ptr[queue_num] + total_bytes,
+		       skb_frag_address_safe(frag), skb_frag_size(frag));
+		total_bytes += skb_frag_size(frag);
+	}
+
+	if (unlikely(total_bytes != skb->len)) {
+		netdev_err(adapter->netdev, "tx: incorrect packet len copied into ltb (%u != %u)\n",
+			   skb->len, total_bytes);
+		netdev->stats.tx_dropped++;
+		goto out;
+	}
+	desc.fields.flags_len = desc_flags | skb->len;
+	desc.fields.address = adapter->tx_ltb_dma[queue_num];
+	/* finish writing to long_term_buff before VIOS accessing it */
+	dma_wmb();
+
+	if (ibmveth_send(adapter, desc.desc, mss)) {
 		adapter->tx_send_failed++;
 		netdev->stats.tx_dropped++;
 	} else {
@@ -1162,37 +1322,11 @@ retry_bounce:
 		netdev->stats.tx_bytes += skb->len;
 	}
 
-	dma_unmap_single(&adapter->vdev->dev,
-			 descs[0].fields.address,
-			 descs[0].fields.flags_len & IBMVETH_BUF_LEN_MASK,
-			 DMA_TO_DEVICE);
-
-	for (i = 1; i < skb_shinfo(skb)->nr_frags + 1; i++)
-		dma_unmap_page(&adapter->vdev->dev, descs[i].fields.address,
-			       descs[i].fields.flags_len & IBMVETH_BUF_LEN_MASK,
-			       DMA_TO_DEVICE);
-
 out:
 	dev_consume_skb_any(skb);
 	return NETDEV_TX_OK;
 
-map_failed_frags:
-	last = i+1;
-	for (i = 0; i < last; i++)
-		dma_unmap_page(&adapter->vdev->dev, descs[i].fields.address,
-			       descs[i].fields.flags_len & IBMVETH_BUF_LEN_MASK,
-			       DMA_TO_DEVICE);
-
-map_failed:
-	if (!firmware_has_feature(FW_FEATURE_CMO))
-		netdev_err(netdev, "tx: unable to map xmit buffer\n");
-	adapter->tx_map_failed++;
-	if (skb_linearize(skb)) {
-		netdev->stats.tx_dropped++;
-		goto out;
-	}
-	force_bounce = 1;
-	goto retry_bounce;
+
 }
 
 static void ibmveth_rx_mss_helper(struct sk_buff *skb, u16 mss, int lrg_pkt)
@@ -1274,36 +1408,40 @@ static void ibmveth_rx_csum_helper(struct sk_buff *skb,
 		iph_proto = iph6->nexthdr;
 	}
 
-	/* In OVS environment, when a flow is not cached, specifically for a
-	 * new TCP connection, the first packet information is passed up
+	/* When CSO is enabled the TCP checksum may have be set to NULL by
+	 * the sender given that we zeroed out TCP checksum field in
+	 * transmit path (refer ibmveth_start_xmit routine). In this case set
+	 * up CHECKSUM_PARTIAL. If the packet is forwarded, the checksum will
+	 * then be recalculated by the destination NIC (CSO must be enabled
+	 * on the destination NIC).
+	 *
+	 * In an OVS environment, when a flow is not cached, specifically for a
+	 * new TCP connection, the first packet information is passed up to
 	 * the user space for finding a flow. During this process, OVS computes
 	 * checksum on the first packet when CHECKSUM_PARTIAL flag is set.
 	 *
-	 * Given that we zeroed out TCP checksum field in transmit path
-	 * (refer ibmveth_start_xmit routine) as we set "no checksum bit",
-	 * OVS computed checksum will be incorrect w/o TCP pseudo checksum
-	 * in the packet. This leads to OVS dropping the packet and hence
-	 * TCP retransmissions are seen.
-	 *
 	 * So, re-compute TCP pseudo header checksum.
 	 */
-	if (iph_proto == IPPROTO_TCP && adapter->is_active_trunk) {
-		struct tcphdr *tcph = (struct tcphdr *)(skb->data + iphlen);
 
-		tcphdrlen = skb->len - iphlen;
-
-		/* Recompute TCP pseudo header checksum */
-		if (skb_proto == ETH_P_IP)
-			tcph->check = ~csum_tcpudp_magic(iph->saddr,
-					iph->daddr, tcphdrlen, iph_proto, 0);
-		else if (skb_proto == ETH_P_IPV6)
-			tcph->check = ~csum_ipv6_magic(&iph6->saddr,
-					&iph6->daddr, tcphdrlen, iph_proto, 0);
+	if (iph_proto == IPPROTO_TCP) {
+		struct tcphdr *tcph = (struct tcphdr *)(skb->data + iphlen);
 
-		/* Setup SKB fields for checksum offload */
-		skb_partial_csum_set(skb, iphlen,
-				     offsetof(struct tcphdr, check));
-		skb_reset_network_header(skb);
+		if (tcph->check == 0x0000) {
+			/* Recompute TCP pseudo header checksum  */
+			tcphdrlen = skb->len - iphlen;
+			if (skb_proto == ETH_P_IP)
+				tcph->check =
+				 ~csum_tcpudp_magic(iph->saddr,
+				iph->daddr, tcphdrlen, iph_proto, 0);
+			else if (skb_proto == ETH_P_IPV6)
+				tcph->check =
+				 ~csum_ipv6_magic(&iph6->saddr,
+				&iph6->daddr, tcphdrlen, iph_proto, 0);
+			/* Setup SKB fields for checksum offload */
+			skb_partial_csum_set(skb, iphlen,
+					     offsetof(struct tcphdr, check));
+			skb_reset_network_header(skb);
+		}
 	}
 }
 
@@ -1326,15 +1464,19 @@ restart_poll:
 			wmb(); /* suggested by larson1 */
 			adapter->rx_invalid_buffer++;
 			netdev_dbg(netdev, "recycling invalid buffer\n");
-			ibmveth_rxq_recycle_buffer(adapter);
+			if (unlikely(ibmveth_rxq_harvest_buffer(adapter, true)))
+				break;
 		} else {
 			struct sk_buff *skb, *new_skb;
 			int length = ibmveth_rxq_frame_length(adapter);
 			int offset = ibmveth_rxq_frame_offset(adapter);
 			int csum_good = ibmveth_rxq_csum_good(adapter);
 			int lrg_pkt = ibmveth_rxq_large_packet(adapter);
+			__sum16 iph_check = 0;
 
 			skb = ibmveth_rxq_get_buffer(adapter);
+			if (unlikely(!skb))
+				break;
 
 			/* if the large packet bit is set in the rx queue
 			 * descriptor, the mss will be written by PHYP eight
@@ -1358,27 +1500,38 @@ restart_poll:
 				if (rx_flush)
 					ibmveth_flush_buffer(skb->data,
 						length + offset);
-				if (!ibmveth_rxq_recycle_buffer(adapter))
-					kfree_skb(skb);
+				if (unlikely(ibmveth_rxq_harvest_buffer(adapter, true)))
+					break;
 				skb = new_skb;
 			} else {
-				ibmveth_rxq_harvest_buffer(adapter);
+				if (unlikely(ibmveth_rxq_harvest_buffer(adapter, false)))
+					break;
 				skb_reserve(skb, offset);
 			}
 
 			skb_put(skb, length);
 			skb->protocol = eth_type_trans(skb, netdev);
 
-			if (csum_good) {
-				skb->ip_summed = CHECKSUM_UNNECESSARY;
-				ibmveth_rx_csum_helper(skb, adapter);
+			/* PHYP without PLSO support places a -1 in the ip
+			 * checksum for large send frames.
+			 */
+			if (skb->protocol == cpu_to_be16(ETH_P_IP)) {
+				struct iphdr *iph = (struct iphdr *)skb->data;
+
+				iph_check = iph->check;
 			}
 
-			if (length > netdev->mtu + ETH_HLEN) {
+			if ((length > netdev->mtu + ETH_HLEN) ||
+			    lrg_pkt || iph_check == 0xffff) {
 				ibmveth_rx_mss_helper(skb, mss, lrg_pkt);
 				adapter->rx_large_packets++;
 			}
 
+			if (csum_good) {
+				skb->ip_summed = CHECKSUM_UNNECESSARY;
+				ibmveth_rx_csum_helper(skb, adapter);
+			}
+
 			napi_gro_receive(napi, skb);	/* send it up */
 
 			netdev->stats.rx_packets++;
@@ -1389,25 +1542,28 @@ restart_poll:
 
 	ibmveth_replenish_task(adapter);
 
-	if (frames_processed < budget) {
-		napi_complete_done(napi, frames_processed);
+	if (frames_processed == budget)
+		goto out;
 
-		/* We think we are done - reenable interrupts,
-		 * then check once more to make sure we are done.
-		 */
-		lpar_rc = h_vio_signal(adapter->vdev->unit_address,
-				       VIO_IRQ_ENABLE);
+	if (!napi_complete_done(napi, frames_processed))
+		goto out;
 
-		BUG_ON(lpar_rc != H_SUCCESS);
+	/* We think we are done - reenable interrupts,
+	 * then check once more to make sure we are done.
+	 */
+	lpar_rc = h_vio_signal(adapter->vdev->unit_address, VIO_IRQ_ENABLE);
+	if (WARN_ON(lpar_rc != H_SUCCESS)) {
+		schedule_work(&adapter->work);
+		goto out;
+	}
 
-		if (ibmveth_rxq_pending_buffer(adapter) &&
-		    napi_reschedule(napi)) {
-			lpar_rc = h_vio_signal(adapter->vdev->unit_address,
-					       VIO_IRQ_DISABLE);
-			goto restart_poll;
-		}
+	if (ibmveth_rxq_pending_buffer(adapter) && napi_schedule(napi)) {
+		lpar_rc = h_vio_signal(adapter->vdev->unit_address,
+				       VIO_IRQ_DISABLE);
+		goto restart_poll;
 	}
 
+out:
 	return frames_processed;
 }
 
@@ -1420,7 +1576,7 @@ static irqreturn_t ibmveth_interrupt(int irq, void *dev_instance)
 	if (napi_schedule_prep(&adapter->napi)) {
 		lpar_rc = h_vio_signal(adapter->vdev->unit_address,
 				       VIO_IRQ_DISABLE);
-		BUG_ON(lpar_rc != H_SUCCESS);
+		WARN_ON(lpar_rc != H_SUCCESS);
 		__napi_schedule(&adapter->napi);
 	}
 	return IRQ_HANDLED;
@@ -1458,7 +1614,7 @@ static void ibmveth_set_multicast_list(struct net_device *netdev)
 		netdev_for_each_mc_addr(ha, netdev) {
 			/* add the multicast address to the filter table */
 			u64 mcast_addr;
-			mcast_addr = ibmveth_encode_mac_addr(ha->addr);
+			mcast_addr = ether_addr_to_u64(ha->addr);
 			lpar_rc = h_multicast_ctrl(adapter->vdev->unit_address,
 						   IbmVethMcastAddFilter,
 						   mcast_addr);
@@ -1499,9 +1655,7 @@ static int ibmveth_change_mtu(struct net_device *dev, int new_mtu)
 	   only the buffer pools necessary to hold the new MTU */
 	if (netif_running(adapter->netdev)) {
 		need_restart = 1;
-		adapter->pool_config = 1;
 		ibmveth_close(adapter->netdev);
-		adapter->pool_config = 0;
 	}
 
 	/* Look for an active buffer pool that can hold the new MTU */
@@ -1509,7 +1663,7 @@ static int ibmveth_change_mtu(struct net_device *dev, int new_mtu)
 		adapter->rx_buff_pool[i].active = 1;
 
 		if (new_mtu_oh <= adapter->rx_buff_pool[i].buff_size) {
-			dev->mtu = new_mtu;
+			WRITE_ONCE(dev->mtu, new_mtu);
 			vio_cmo_set_dev_desired(viodev,
 						ibmveth_get_desired_dma
 						(viodev));
@@ -1561,6 +1715,8 @@ static unsigned long ibmveth_get_desired_dma(struct vio_dev *vdev)
 
 	ret = IBMVETH_BUFF_LIST_SIZE + IBMVETH_FILT_LIST_SIZE;
 	ret += IOMMU_PAGE_ALIGN(netdev->mtu, tbl);
+	/* add size of mapped tx buffers */
+	ret += IOMMU_PAGE_ALIGN(IBMVETH_MAX_TX_BUF_SIZE, tbl);
 
 	for (i = 0; i < IBMVETH_NUM_BUFF_POOLS; i++) {
 		/* add the size of the active receive buffers */
@@ -1588,14 +1744,14 @@ static int ibmveth_set_mac_addr(struct net_device *dev, void *p)
 	if (!is_valid_ether_addr(addr->sa_data))
 		return -EADDRNOTAVAIL;
 
-	mac_address = ibmveth_encode_mac_addr(addr->sa_data);
+	mac_address = ether_addr_to_u64(addr->sa_data);
 	rc = h_change_logical_lan_mac(adapter->vdev->unit_address, mac_address);
 	if (rc) {
 		netdev_err(adapter->netdev, "h_change_logical_lan_mac failed with rc=%d\n", rc);
 		return rc;
 	}
 
-	ether_addr_copy(dev->dev_addr, addr->sa_data);
+	eth_hw_addr_set(dev, addr->sa_data);
 
 	return 0;
 }
@@ -1605,7 +1761,7 @@ static const struct net_device_ops ibmveth_netdev_ops = {
 	.ndo_stop		= ibmveth_close,
 	.ndo_start_xmit		= ibmveth_start_xmit,
 	.ndo_set_rx_mode	= ibmveth_set_multicast_list,
-	.ndo_do_ioctl		= ibmveth_ioctl,
+	.ndo_eth_ioctl		= ibmveth_ioctl,
 	.ndo_change_mtu		= ibmveth_change_mtu,
 	.ndo_fix_features	= ibmveth_fix_features,
 	.ndo_set_features	= ibmveth_set_features,
@@ -1622,7 +1778,7 @@ static int ibmveth_probe(struct vio_dev *dev, const struct vio_device_id *id)
 	struct net_device *netdev;
 	struct ibmveth_adapter *adapter;
 	unsigned char *mac_addr_p;
-	unsigned int *mcastFilterSize_p;
+	__be32 *mcastFilterSize_p;
 	long ret;
 	unsigned long ret_attr;
 
@@ -1644,16 +1800,16 @@ static int ibmveth_probe(struct vio_dev *dev, const struct vio_device_id *id)
 		return -EINVAL;
 	}
 
-	mcastFilterSize_p = (unsigned int *)vio_get_attribute(dev,
-						VETH_MCAST_FILTER_SIZE, NULL);
+	mcastFilterSize_p = (__be32 *)vio_get_attribute(dev,
+							VETH_MCAST_FILTER_SIZE,
+							NULL);
 	if (!mcastFilterSize_p) {
 		dev_err(&dev->dev, "Can't find VETH_MCAST_FILTER_SIZE "
 			"attribute\n");
 		return -EINVAL;
 	}
 
-	netdev = alloc_etherdev(sizeof(struct ibmveth_adapter));
-
+	netdev = alloc_etherdev_mqs(sizeof(struct ibmveth_adapter), IBMVETH_MAX_QUEUES, 1);
 	if (!netdev)
 		return -ENOMEM;
 
@@ -1662,10 +1818,11 @@ static int ibmveth_probe(struct vio_dev *dev, const struct vio_device_id *id)
 
 	adapter->vdev = dev;
 	adapter->netdev = netdev;
-	adapter->mcastFilterSize = *mcastFilterSize_p;
-	adapter->pool_config = 0;
+	INIT_WORK(&adapter->work, ibmveth_reset);
+	adapter->mcastFilterSize = be32_to_cpu(*mcastFilterSize_p);
+	ibmveth_init_link_settings(netdev);
 
-	netif_napi_add(netdev, &adapter->napi, ibmveth_poll, 16);
+	netif_napi_add_weight(netdev, &adapter->napi, ibmveth_poll, 16);
 
 	netdev->irq = dev->irq;
 	netdev->netdev_ops = &ibmveth_netdev_ops;
@@ -1697,10 +1854,23 @@ static int ibmveth_probe(struct vio_dev *dev, const struct vio_device_id *id)
 		netdev->features |= NETIF_F_FRAGLIST;
 	}
 
+	if (ret == H_SUCCESS &&
+	    (ret_attr & IBMVETH_ILLAN_RX_MULTI_BUFF_SUPPORT)) {
+		adapter->rx_buffers_per_hcall = IBMVETH_MAX_RX_PER_HCALL;
+		netdev_dbg(netdev,
+			   "RX Multi-buffer hcall supported by FW, batch set to %u\n",
+			    adapter->rx_buffers_per_hcall);
+	} else {
+		adapter->rx_buffers_per_hcall = 1;
+		netdev_dbg(netdev,
+			   "RX Single-buffer hcall mode, batch set to %u\n",
+			   adapter->rx_buffers_per_hcall);
+	}
+
 	netdev->min_mtu = IBMVETH_MIN_MTU;
-	netdev->max_mtu = ETH_MAX_MTU;
+	netdev->max_mtu = ETH_MAX_MTU - IBMVETH_BUFF_OH;
 
-	memcpy(netdev->dev_addr, mac_addr_p, ETH_ALEN);
+	eth_hw_addr_set(netdev, mac_addr_p);
 
 	if (firmware_has_feature(FW_FEATURE_CMO))
 		memcpy(pool_count, pool_count_cmo, sizeof(pool_count));
@@ -1718,12 +1888,19 @@ static int ibmveth_probe(struct vio_dev *dev, const struct vio_device_id *id)
 			kobject_uevent(kobj, KOBJ_ADD);
 	}
 
-	netdev_dbg(netdev, "adapter @ 0x%p\n", adapter);
-
-	adapter->buffer_list_dma = DMA_ERROR_CODE;
-	adapter->filter_list_dma = DMA_ERROR_CODE;
-	adapter->rx_queue.queue_dma = DMA_ERROR_CODE;
+	rc = netif_set_real_num_tx_queues(netdev, min(num_online_cpus(),
+						      IBMVETH_DEFAULT_QUEUES));
+	if (rc) {
+		netdev_dbg(netdev, "failed to set number of tx queues rc=%d\n",
+			   rc);
+		free_netdev(netdev);
+		return rc;
+	}
+	adapter->tx_ltb_size = PAGE_ALIGN(IBMVETH_MAX_TX_BUF_SIZE);
+	for (i = 0; i < IBMVETH_MAX_QUEUES; i++)
+		adapter->tx_ltb_ptr[i] = NULL;
 
+	netdev_dbg(netdev, "adapter @ 0x%p\n", adapter);
 	netdev_dbg(netdev, "registering netdev...\n");
 
 	ibmveth_set_features(netdev, netdev->features);
@@ -1741,12 +1918,14 @@ static int ibmveth_probe(struct vio_dev *dev, const struct vio_device_id *id)
 	return 0;
 }
 
-static int ibmveth_remove(struct vio_dev *dev)
+static void ibmveth_remove(struct vio_dev *dev)
 {
 	struct net_device *netdev = dev_get_drvdata(&dev->dev);
 	struct ibmveth_adapter *adapter = netdev_priv(netdev);
 	int i;
 
+	cancel_work_sync(&adapter->work);
+
 	for (i = 0; i < IBMVETH_NUM_BUFF_POOLS; i++)
 		kobject_put(&adapter->rx_buff_pool[i].kobj);
 
@@ -1754,8 +1933,6 @@ static int ibmveth_remove(struct vio_dev *dev)
 
 	free_netdev(netdev);
 	dev_set_drvdata(&dev->dev, NULL);
-
-	return 0;
 }
 
 static struct attribute veth_active_attr;
@@ -1778,36 +1955,59 @@ static ssize_t veth_pool_show(struct kobject *kobj,
 	return 0;
 }
 
+/**
+ * veth_pool_store - sysfs store handler for pool attributes
+ * @kobj: kobject embedded in pool
+ * @attr: attribute being changed
+ * @buf: value being stored
+ * @count: length of @buf in bytes
+ *
+ * Stores new value in pool attribute. Verifies the range of the new value for
+ * size and buff_size. Verifies that at least one pool remains available to
+ * receive MTU-sized packets.
+ *
+ * Context: Process context.
+ *          Takes and releases rtnl_mutex to ensure correct ordering of close
+ *	    and open calls.
+ * Return:
+ * * %-EPERM  - Not allowed to disabled all MTU-sized buffer pools
+ * * %-EINVAL - New pool size or buffer size is out of range
+ * * count    - Return count for success
+ * * other    - Return value from a failed ibmveth_open call
+ */
 static ssize_t veth_pool_store(struct kobject *kobj, struct attribute *attr,
 			       const char *buf, size_t count)
 {
 	struct ibmveth_buff_pool *pool = container_of(kobj,
 						      struct ibmveth_buff_pool,
 						      kobj);
-	struct net_device *netdev = dev_get_drvdata(
-	    container_of(kobj->parent, struct device, kobj));
+	struct net_device *netdev = dev_get_drvdata(kobj_to_dev(kobj->parent));
 	struct ibmveth_adapter *adapter = netdev_priv(netdev);
 	long value = simple_strtol(buf, NULL, 10);
+	bool change = false;
+	u32 newbuff_size;
+	u32 oldbuff_size;
+	int newactive;
+	int oldactive;
+	u32 newsize;
+	u32 oldsize;
 	long rc;
 
+	rtnl_lock();
+
+	oldbuff_size = pool->buff_size;
+	oldactive = pool->active;
+	oldsize = pool->size;
+
+	newbuff_size = oldbuff_size;
+	newactive = oldactive;
+	newsize = oldsize;
+
 	if (attr == &veth_active_attr) {
-		if (value && !pool->active) {
-			if (netif_running(netdev)) {
-				if (ibmveth_alloc_buffer_pool(pool)) {
-					netdev_err(netdev,
-						   "unable to alloc pool\n");
-					return -ENOMEM;
-				}
-				pool->active = 1;
-				adapter->pool_config = 1;
-				ibmveth_close(netdev);
-				adapter->pool_config = 0;
-				if ((rc = ibmveth_open(netdev)))
-					return rc;
-			} else {
-				pool->active = 1;
-			}
-		} else if (!value && pool->active) {
+		if (value && !oldactive) {
+			newactive = 1;
+			change = true;
+		} else if (!value && oldactive) {
 			int mtu = netdev->mtu + IBMVETH_BUFF_OH;
 			int i;
 			/* Make sure there is a buffer pool with buffers that
@@ -1823,54 +2023,60 @@ static ssize_t veth_pool_store(struct kobject *kobj, struct attribute *attr,
 
 			if (i == IBMVETH_NUM_BUFF_POOLS) {
 				netdev_err(netdev, "no active pool >= MTU\n");
-				return -EPERM;
+				rc = -EPERM;
+				goto unlock_err;
 			}
 
-			if (netif_running(netdev)) {
-				adapter->pool_config = 1;
-				ibmveth_close(netdev);
-				pool->active = 0;
-				adapter->pool_config = 0;
-				if ((rc = ibmveth_open(netdev)))
-					return rc;
-			}
-			pool->active = 0;
+			newactive = 0;
+			change = true;
 		}
 	} else if (attr == &veth_num_attr) {
 		if (value <= 0 || value > IBMVETH_MAX_POOL_COUNT) {
-			return -EINVAL;
-		} else {
-			if (netif_running(netdev)) {
-				adapter->pool_config = 1;
-				ibmveth_close(netdev);
-				adapter->pool_config = 0;
-				pool->size = value;
-				if ((rc = ibmveth_open(netdev)))
-					return rc;
-			} else {
-				pool->size = value;
-			}
+			rc = -EINVAL;
+			goto unlock_err;
+		}
+		if (value != oldsize) {
+			newsize = value;
+			change = true;
 		}
 	} else if (attr == &veth_size_attr) {
 		if (value <= IBMVETH_BUFF_OH || value > IBMVETH_MAX_BUF_SIZE) {
-			return -EINVAL;
-		} else {
-			if (netif_running(netdev)) {
-				adapter->pool_config = 1;
-				ibmveth_close(netdev);
-				adapter->pool_config = 0;
-				pool->buff_size = value;
-				if ((rc = ibmveth_open(netdev)))
-					return rc;
-			} else {
-				pool->buff_size = value;
+			rc = -EINVAL;
+			goto unlock_err;
+		}
+		if (value != oldbuff_size) {
+			newbuff_size = value;
+			change = true;
+		}
+	}
+
+	if (change) {
+		if (netif_running(netdev))
+			ibmveth_close(netdev);
+
+		pool->active = newactive;
+		pool->buff_size = newbuff_size;
+		pool->size = newsize;
+
+		if (netif_running(netdev)) {
+			rc = ibmveth_open(netdev);
+			if (rc) {
+				pool->active = oldactive;
+				pool->buff_size = oldbuff_size;
+				pool->size = oldsize;
+				goto unlock_err;
 			}
 		}
 	}
+	rtnl_unlock();
 
 	/* kick the interrupt handler to allocate/deallocate pools */
 	ibmveth_interrupt(netdev->irq, netdev);
 	return count;
+
+unlock_err:
+	rtnl_unlock();
+	return rc;
 }
 
 
@@ -1889,6 +2095,7 @@ static struct attribute *veth_pool_attrs[] = {
 	&veth_size_attr,
 	NULL,
 };
+ATTRIBUTE_GROUPS(veth_pool);
 
 static const struct sysfs_ops veth_pool_ops = {
 	.show   = veth_pool_show,
@@ -1898,7 +2105,7 @@ static const struct sysfs_ops veth_pool_ops = {
 static struct kobj_type ktype_veth_pool = {
 	.release        = NULL,
 	.sysfs_ops      = &veth_pool_ops,
-	.default_attrs  = veth_pool_attrs,
+	.default_groups = veth_pool_groups,
 };
 
 static int ibmveth_resume(struct device *dev)
@@ -1908,13 +2115,13 @@ static int ibmveth_resume(struct device *dev)
 	return 0;
 }
 
-static struct vio_device_id ibmveth_device_table[] = {
+static const struct vio_device_id ibmveth_device_table[] = {
 	{ "network", "IBM,l-lan"},
 	{ "", "" }
 };
 MODULE_DEVICE_TABLE(vio, ibmveth_device_table);
 
-static struct dev_pm_ops ibmveth_pm_ops = {
+static const struct dev_pm_ops ibmveth_pm_ops = {
 	.resume = ibmveth_resume
 };
 
@@ -1942,3 +2149,132 @@ static void __exit ibmveth_module_exit(void)
 
 module_init(ibmveth_module_init);
 module_exit(ibmveth_module_exit);
+
+#ifdef CONFIG_IBMVETH_KUNIT_TEST
+#include <kunit/test.h>
+
+/**
+ * ibmveth_reset_kunit - reset routine for running in KUnit environment
+ *
+ * @w: pointer to work_struct embedded in adapter structure
+ *
+ * Context: Called in the KUnit environment. Does nothing.
+ *
+ * Return: void
+ */
+static void ibmveth_reset_kunit(struct work_struct *w)
+{
+	netdev_dbg(NULL, "reset_kunit starting\n");
+	netdev_dbg(NULL, "reset_kunit complete\n");
+}
+
+/**
+ * ibmveth_remove_buffer_from_pool_test - unit test for some of
+ *                                        ibmveth_remove_buffer_from_pool
+ * @test: pointer to kunit structure
+ *
+ * Tests the error returns from ibmveth_remove_buffer_from_pool.
+ * ibmveth_remove_buffer_from_pool also calls WARN_ON, so dmesg should be
+ * checked to see that these warnings happened.
+ *
+ * Return: void
+ */
+static void ibmveth_remove_buffer_from_pool_test(struct kunit *test)
+{
+	struct ibmveth_adapter *adapter = kunit_kzalloc(test, sizeof(*adapter), GFP_KERNEL);
+	struct ibmveth_buff_pool *pool;
+	u64 correlator;
+
+	KUNIT_ASSERT_NOT_ERR_OR_NULL(test, adapter);
+
+	INIT_WORK(&adapter->work, ibmveth_reset_kunit);
+
+	/* Set sane values for buffer pools */
+	for (int i = 0; i < IBMVETH_NUM_BUFF_POOLS; i++)
+		ibmveth_init_buffer_pool(&adapter->rx_buff_pool[i], i,
+					 pool_count[i], pool_size[i],
+					 pool_active[i]);
+
+	pool = &adapter->rx_buff_pool[0];
+	pool->skbuff = kunit_kcalloc(test, pool->size, sizeof(void *), GFP_KERNEL);
+	KUNIT_ASSERT_NOT_ERR_OR_NULL(test, pool->skbuff);
+
+	correlator = ((u64)IBMVETH_NUM_BUFF_POOLS << 32) | 0;
+	KUNIT_EXPECT_EQ(test, -EINVAL, ibmveth_remove_buffer_from_pool(adapter, correlator, false));
+	KUNIT_EXPECT_EQ(test, -EINVAL, ibmveth_remove_buffer_from_pool(adapter, correlator, true));
+
+	correlator = ((u64)0 << 32) | adapter->rx_buff_pool[0].size;
+	KUNIT_EXPECT_EQ(test, -EINVAL, ibmveth_remove_buffer_from_pool(adapter, correlator, false));
+	KUNIT_EXPECT_EQ(test, -EINVAL, ibmveth_remove_buffer_from_pool(adapter, correlator, true));
+
+	correlator = (u64)0 | 0;
+	pool->skbuff[0] = NULL;
+	KUNIT_EXPECT_EQ(test, -EFAULT, ibmveth_remove_buffer_from_pool(adapter, correlator, false));
+	KUNIT_EXPECT_EQ(test, -EFAULT, ibmveth_remove_buffer_from_pool(adapter, correlator, true));
+
+	flush_work(&adapter->work);
+}
+
+/**
+ * ibmveth_rxq_get_buffer_test - unit test for ibmveth_rxq_get_buffer
+ * @test: pointer to kunit structure
+ *
+ * Tests ibmveth_rxq_get_buffer. ibmveth_rxq_get_buffer also calls WARN_ON for
+ * the NULL returns, so dmesg should be checked to see that these warnings
+ * happened.
+ *
+ * Return: void
+ */
+static void ibmveth_rxq_get_buffer_test(struct kunit *test)
+{
+	struct ibmveth_adapter *adapter = kunit_kzalloc(test, sizeof(*adapter), GFP_KERNEL);
+	struct sk_buff *skb = kunit_kzalloc(test, sizeof(*skb), GFP_KERNEL);
+	struct ibmveth_buff_pool *pool;
+
+	KUNIT_ASSERT_NOT_ERR_OR_NULL(test, adapter);
+	KUNIT_ASSERT_NOT_ERR_OR_NULL(test, skb);
+
+	INIT_WORK(&adapter->work, ibmveth_reset_kunit);
+
+	adapter->rx_queue.queue_len = 1;
+	adapter->rx_queue.index = 0;
+	adapter->rx_queue.queue_addr = kunit_kzalloc(test, sizeof(struct ibmveth_rx_q_entry),
+						     GFP_KERNEL);
+	KUNIT_ASSERT_NOT_ERR_OR_NULL(test, adapter->rx_queue.queue_addr);
+
+	/* Set sane values for buffer pools */
+	for (int i = 0; i < IBMVETH_NUM_BUFF_POOLS; i++)
+		ibmveth_init_buffer_pool(&adapter->rx_buff_pool[i], i,
+					 pool_count[i], pool_size[i],
+					 pool_active[i]);
+
+	pool = &adapter->rx_buff_pool[0];
+	pool->skbuff = kunit_kcalloc(test, pool->size, sizeof(void *), GFP_KERNEL);
+	KUNIT_ASSERT_NOT_ERR_OR_NULL(test, pool->skbuff);
+
+	adapter->rx_queue.queue_addr[0].correlator = (u64)IBMVETH_NUM_BUFF_POOLS << 32 | 0;
+	KUNIT_EXPECT_PTR_EQ(test, NULL, ibmveth_rxq_get_buffer(adapter));
+
+	adapter->rx_queue.queue_addr[0].correlator = (u64)0 << 32 | adapter->rx_buff_pool[0].size;
+	KUNIT_EXPECT_PTR_EQ(test, NULL, ibmveth_rxq_get_buffer(adapter));
+
+	pool->skbuff[0] = skb;
+	adapter->rx_queue.queue_addr[0].correlator = (u64)0 << 32 | 0;
+	KUNIT_EXPECT_PTR_EQ(test, skb, ibmveth_rxq_get_buffer(adapter));
+
+	flush_work(&adapter->work);
+}
+
+static struct kunit_case ibmveth_test_cases[] = {
+	KUNIT_CASE(ibmveth_remove_buffer_from_pool_test),
+	KUNIT_CASE(ibmveth_rxq_get_buffer_test),
+	{}
+};
+
+static struct kunit_suite ibmveth_test_suite = {
+	.name = "ibmveth-kunit-test",
+	.test_cases = ibmveth_test_cases,
+};
+
+kunit_test_suite(ibmveth_test_suite);
+#endif
diff --git a/drivers/net/ethernet/ibm/ibmveth.h b/drivers/net/ethernet/ibm/ibmveth.h
index 01c587fc02c7..068f99df133e 100644
--- a/drivers/net/ethernet/ibm/ibmveth.h
+++ b/drivers/net/ethernet/ibm/ibmveth.h
@@ -1,19 +1,7 @@
+/* SPDX-License-Identifier: GPL-2.0-or-later */
 /*
  * IBM Power Virtual Ethernet Device Driver
  *
- * This program is free software; you can redistribute it and/or modify
- * it under the terms of the GNU General Public License as published by
- * the Free Software Foundation; either version 2 of the License, or
- * (at your option) any later version.
- *
- * This program is distributed in the hope that it will be useful,
- * but WITHOUT ANY WARRANTY; without even the implied warranty of
- * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
- * GNU General Public License for more details.
- *
- * You should have received a copy of the GNU General Public License
- * along with this program; if not, see <http://www.gnu.org/licenses/>.
- *
  * Copyright (C) IBM Corporation, 2003, 2010
  *
  * Authors: Dave Larson <larson1@us.ibm.com>
@@ -40,6 +28,7 @@
 #define IbmVethMcastRemoveFilter     0x2UL
 #define IbmVethMcastClearFilterTable 0x3UL
 
+#define IBMVETH_ILLAN_RX_MULTI_BUFF_SUPPORT	0x0000000000040000UL
 #define IBMVETH_ILLAN_LRG_SR_ENABLED	0x0000000000010000UL
 #define IBMVETH_ILLAN_LRG_SND_SUPPORT	0x0000000000008000UL
 #define IBMVETH_ILLAN_PADDED_PKT_CSUM	0x0000000000002000UL
@@ -58,23 +47,41 @@
 #define h_add_logical_lan_buffer(ua, buf) \
   plpar_hcall_norets(H_ADD_LOGICAL_LAN_BUFFER, ua, buf)
 
+static inline long h_add_logical_lan_buffers(unsigned long unit_address,
+					     unsigned long desc1,
+					     unsigned long desc2,
+					     unsigned long desc3,
+					     unsigned long desc4,
+					     unsigned long desc5,
+					     unsigned long desc6,
+					     unsigned long desc7,
+					     unsigned long desc8)
+{
+	unsigned long retbuf[PLPAR_HCALL9_BUFSIZE];
+
+	return plpar_hcall9(H_ADD_LOGICAL_LAN_BUFFERS,
+			    retbuf, unit_address,
+			    desc1, desc2, desc3, desc4,
+			    desc5, desc6, desc7, desc8);
+}
+
+/* FW allows us to send 6 descriptors but we only use one so mark
+ * the other 5 as unused (0)
+ */
 static inline long h_send_logical_lan(unsigned long unit_address,
-		unsigned long desc1, unsigned long desc2, unsigned long desc3,
-		unsigned long desc4, unsigned long desc5, unsigned long desc6,
-		unsigned long corellator_in, unsigned long *corellator_out,
-		unsigned long mss, unsigned long large_send_support)
+		unsigned long desc, unsigned long corellator_in,
+		unsigned long *corellator_out, unsigned long mss,
+		unsigned long large_send_support)
 {
 	long rc;
 	unsigned long retbuf[PLPAR_HCALL9_BUFSIZE];
 
 	if (large_send_support)
 		rc = plpar_hcall9(H_SEND_LOGICAL_LAN, retbuf, unit_address,
-				  desc1, desc2, desc3, desc4, desc5, desc6,
-				  corellator_in, mss);
+				  desc, 0, 0, 0, 0, 0, corellator_in, mss);
 	else
 		rc = plpar_hcall9(H_SEND_LOGICAL_LAN, retbuf, unit_address,
-				  desc1, desc2, desc3, desc4, desc5, desc6,
-				  corellator_in);
+				  desc, 0, 0, 0, 0, 0, corellator_in);
 
 	*corellator_out = retbuf[0];
 
@@ -110,6 +117,10 @@ static inline long h_illan_attributes(unsigned long unit_address,
 #define IBMVETH_BUFF_LIST_SIZE 4096
 #define IBMVETH_FILT_LIST_SIZE 4096
 #define IBMVETH_MAX_BUF_SIZE (1024 * 128)
+#define IBMVETH_MAX_TX_BUF_SIZE (1024 * 64)
+#define IBMVETH_MAX_QUEUES 16U
+#define IBMVETH_DEFAULT_QUEUES 8U
+#define IBMVETH_MAX_RX_PER_HCALL 8U
 
 static int pool_size[] = { 512, 1024 * 2, 1024 * 16, 1024 * 32, 1024 * 64 };
 static int pool_count[] = { 256, 512, 256, 256, 256 };
@@ -143,37 +154,42 @@ struct ibmveth_rx_q {
 };
 
 struct ibmveth_adapter {
-    struct vio_dev *vdev;
-    struct net_device *netdev;
-    struct napi_struct napi;
-    unsigned int mcastFilterSize;
-    void * buffer_list_addr;
-    void * filter_list_addr;
-    dma_addr_t buffer_list_dma;
-    dma_addr_t filter_list_dma;
-    struct ibmveth_buff_pool rx_buff_pool[IBMVETH_NUM_BUFF_POOLS];
-    struct ibmveth_rx_q rx_queue;
-    int pool_config;
-    int rx_csum;
-    int large_send;
-    bool is_active_trunk;
-    void *bounce_buffer;
-    dma_addr_t bounce_buffer_dma;
-
-    u64 fw_ipv6_csum_support;
-    u64 fw_ipv4_csum_support;
-    u64 fw_large_send_support;
-    /* adapter specific stats */
-    u64 replenish_task_cycles;
-    u64 replenish_no_mem;
-    u64 replenish_add_buff_failure;
-    u64 replenish_add_buff_success;
-    u64 rx_invalid_buffer;
-    u64 rx_no_buffer;
-    u64 tx_map_failed;
-    u64 tx_send_failed;
-    u64 tx_large_packets;
-    u64 rx_large_packets;
+	struct vio_dev *vdev;
+	struct net_device *netdev;
+	struct napi_struct napi;
+	struct work_struct work;
+	unsigned int mcastFilterSize;
+	void *buffer_list_addr;
+	void *filter_list_addr;
+	void *tx_ltb_ptr[IBMVETH_MAX_QUEUES];
+	unsigned int tx_ltb_size;
+	dma_addr_t tx_ltb_dma[IBMVETH_MAX_QUEUES];
+	dma_addr_t buffer_list_dma;
+	dma_addr_t filter_list_dma;
+	struct ibmveth_buff_pool rx_buff_pool[IBMVETH_NUM_BUFF_POOLS];
+	struct ibmveth_rx_q rx_queue;
+	int rx_csum;
+	int large_send;
+	bool is_active_trunk;
+	unsigned int rx_buffers_per_hcall;
+
+	u64 fw_ipv6_csum_support;
+	u64 fw_ipv4_csum_support;
+	u64 fw_large_send_support;
+	/* adapter specific stats */
+	u64 replenish_task_cycles;
+	u64 replenish_no_mem;
+	u64 replenish_add_buff_failure;
+	u64 replenish_add_buff_success;
+	u64 rx_invalid_buffer;
+	u64 rx_no_buffer;
+	u64 tx_map_failed;
+	u64 tx_send_failed;
+	u64 tx_large_packets;
+	u64 rx_large_packets;
+	/* Ethtool settings */
+	u8 duplex;
+	u32 speed;
 };
 
 /*
diff --git a/drivers/net/ethernet/ibm/ibmvnic.c b/drivers/net/ethernet/ibm/ibmvnic.c
index 8dcf58088178..3808148c1fc7 100644
--- a/drivers/net/ethernet/ibm/ibmvnic.c
+++ b/drivers/net/ethernet/ibm/ibmvnic.c
@@ -1,3 +1,4 @@
+// SPDX-License-Identifier: GPL-2.0-or-later
 /**************************************************************************/
 /*                                                                        */
 /*  IBM System i and System p Virtual NIC Device Driver                   */
@@ -6,18 +7,6 @@
 /*  Thomas Falcon (tlfalcon@linux.vnet.ibm.com)                           */
 /*  John Allen (jallen@linux.vnet.ibm.com)                                */
 /*                                                                        */
-/*  This program is free software; you can redistribute it and/or modify  */
-/*  it under the terms of the GNU General Public License as published by  */
-/*  the Free Software Foundation; either version 2 of the License, or     */
-/*  (at your option) any later version.                                   */
-/*                                                                        */
-/*  This program is distributed in the hope that it will be useful,       */
-/*  but WITHOUT ANY WARRANTY; without even the implied warranty of        */
-/*  MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the         */
-/*  GNU General Public License for more details.                          */
-/*                                                                        */
-/*  You should have received a copy of the GNU General Public License     */
-/*  along with this program.                                              */
 /*                                                                        */
 /* This module contains the implementation of a virtual ethernet device   */
 /* for use with IBM i/p Series LPAR Linux. It utilizes the logical LAN    */
@@ -59,10 +48,12 @@
 #include <linux/mm.h>
 #include <linux/ethtool.h>
 #include <linux/proc_fs.h>
+#include <linux/if_arp.h>
 #include <linux/in.h>
 #include <linux/ip.h>
 #include <linux/ipv6.h>
 #include <linux/irq.h>
+#include <linux/irqdomain.h>
 #include <linux/kthread.h>
 #include <linux/seq_file.h>
 #include <linux/interrupt.h>
@@ -70,31 +61,31 @@
 #include <asm/hvcall.h>
 #include <linux/atomic.h>
 #include <asm/vio.h>
+#include <asm/xive.h>
 #include <asm/iommu.h>
 #include <linux/uaccess.h>
 #include <asm/firmware.h>
 #include <linux/workqueue.h>
 #include <linux/if_vlan.h>
+#include <linux/utsname.h>
+#include <linux/cpu.h>
 
 #include "ibmvnic.h"
 
 static const char ibmvnic_driver_name[] = "ibmvnic";
 static const char ibmvnic_driver_string[] = "IBM System i/p Virtual NIC Driver";
 
-MODULE_AUTHOR("Santiago Leon <santi_leon@yahoo.com>");
+MODULE_AUTHOR("Santiago Leon");
 MODULE_DESCRIPTION("IBM System i/p Virtual NIC Driver");
 MODULE_LICENSE("GPL");
 MODULE_VERSION(IBMVNIC_DRIVER_VERSION);
 
 static int ibmvnic_version = IBMVNIC_INITIAL_VERSION;
-static int ibmvnic_remove(struct vio_dev *);
-static void release_sub_crqs(struct ibmvnic_adapter *);
+static void release_sub_crqs(struct ibmvnic_adapter *, bool);
 static int ibmvnic_reset_crq(struct ibmvnic_adapter *);
 static int ibmvnic_send_crq_init(struct ibmvnic_adapter *);
 static int ibmvnic_reenable_crq_queue(struct ibmvnic_adapter *);
 static int ibmvnic_send_crq(struct ibmvnic_adapter *, union ibmvnic_crq *);
-static int send_subcrq(struct ibmvnic_adapter *adapter, u64 remote_handle,
-		       union sub_crq *sub_crq);
 static int send_subcrq_indirect(struct ibmvnic_adapter *, u64, u64, u64);
 static irqreturn_t ibmvnic_interrupt_rx(int irq, void *instance);
 static int enable_scrq_irq(struct ibmvnic_adapter *,
@@ -106,14 +97,27 @@ static int pending_scrq(struct ibmvnic_adapter *,
 static union sub_crq *ibmvnic_next_scrq(struct ibmvnic_adapter *,
 					struct ibmvnic_sub_crq_queue *);
 static int ibmvnic_poll(struct napi_struct *napi, int data);
-static void send_map_query(struct ibmvnic_adapter *adapter);
-static void send_request_map(struct ibmvnic_adapter *, dma_addr_t, __be32, u8);
-static void send_request_unmap(struct ibmvnic_adapter *, u8);
-static void send_login(struct ibmvnic_adapter *adapter);
-static void send_cap_queries(struct ibmvnic_adapter *adapter);
+static int reset_sub_crq_queues(struct ibmvnic_adapter *adapter);
+static inline void reinit_init_done(struct ibmvnic_adapter *adapter);
+static void send_query_map(struct ibmvnic_adapter *adapter);
+static int send_request_map(struct ibmvnic_adapter *, dma_addr_t, u32, u8);
+static int send_request_unmap(struct ibmvnic_adapter *, u8);
+static int send_login(struct ibmvnic_adapter *adapter);
+static void send_query_cap(struct ibmvnic_adapter *adapter);
+static int init_sub_crqs(struct ibmvnic_adapter *);
 static int init_sub_crq_irqs(struct ibmvnic_adapter *adapter);
-static int ibmvnic_init(struct ibmvnic_adapter *);
+static int ibmvnic_reset_init(struct ibmvnic_adapter *, bool reset);
 static void release_crq_queue(struct ibmvnic_adapter *);
+static int __ibmvnic_set_mac(struct net_device *, u8 *);
+static int init_crq_queue(struct ibmvnic_adapter *adapter);
+static int send_query_phys_parms(struct ibmvnic_adapter *adapter);
+static void ibmvnic_tx_scrq_clean_buffer(struct ibmvnic_adapter *adapter,
+					 struct ibmvnic_sub_crq_queue *tx_scrq);
+static void free_long_term_buff(struct ibmvnic_adapter *adapter,
+				struct ibmvnic_long_term_buff *ltb);
+static void ibmvnic_disable_irqs(struct ibmvnic_adapter *adapter);
+static void flush_reset_queue(struct ibmvnic_adapter *adapter);
+static void print_subcrq_error(struct device *dev, int rc, const char *func);
 
 struct ibmvnic_stat {
 	char name[ETH_GSTRING_LEN];
@@ -122,7 +126,7 @@ struct ibmvnic_stat {
 
 #define IBMVNIC_STAT_OFF(stat) (offsetof(struct ibmvnic_adapter, stats) + \
 			     offsetof(struct ibmvnic_statistics, stat))
-#define IBMVNIC_GET_STAT(a, off) (*((u64 *)(((unsigned long)(a)) + off)))
+#define IBMVNIC_GET_STAT(a, off) (*((u64 *)(((unsigned long)(a)) + (off))))
 
 static const struct ibmvnic_stat ibmvnic_stats[] = {
 	{"rx_packets", IBMVNIC_STAT_OFF(rx_packets)},
@@ -149,6 +153,222 @@ static const struct ibmvnic_stat ibmvnic_stats[] = {
 	{"internal_mac_rx_errors", IBMVNIC_STAT_OFF(internal_mac_rx_errors)},
 };
 
+static int send_crq_init_complete(struct ibmvnic_adapter *adapter)
+{
+	union ibmvnic_crq crq;
+
+	memset(&crq, 0, sizeof(crq));
+	crq.generic.first = IBMVNIC_CRQ_INIT_CMD;
+	crq.generic.cmd = IBMVNIC_CRQ_INIT_COMPLETE;
+
+	return ibmvnic_send_crq(adapter, &crq);
+}
+
+static int send_version_xchg(struct ibmvnic_adapter *adapter)
+{
+	union ibmvnic_crq crq;
+
+	memset(&crq, 0, sizeof(crq));
+	crq.version_exchange.first = IBMVNIC_CRQ_CMD;
+	crq.version_exchange.cmd = VERSION_EXCHANGE;
+	crq.version_exchange.version = cpu_to_be16(ibmvnic_version);
+
+	return ibmvnic_send_crq(adapter, &crq);
+}
+
+static void ibmvnic_clean_queue_affinity(struct ibmvnic_adapter *adapter,
+					 struct ibmvnic_sub_crq_queue *queue)
+{
+	if (!(queue && queue->irq))
+		return;
+
+	cpumask_clear(queue->affinity_mask);
+
+	if (irq_set_affinity_and_hint(queue->irq, NULL))
+		netdev_warn(adapter->netdev,
+			    "%s: Clear affinity failed, queue addr = %p, IRQ = %d\n",
+			    __func__, queue, queue->irq);
+}
+
+static void ibmvnic_clean_affinity(struct ibmvnic_adapter *adapter)
+{
+	struct ibmvnic_sub_crq_queue **rxqs;
+	struct ibmvnic_sub_crq_queue **txqs;
+	int num_rxqs, num_txqs;
+	int i;
+
+	rxqs = adapter->rx_scrq;
+	txqs = adapter->tx_scrq;
+	num_txqs = adapter->num_active_tx_scrqs;
+	num_rxqs = adapter->num_active_rx_scrqs;
+
+	netdev_dbg(adapter->netdev, "%s: Cleaning irq affinity hints", __func__);
+	if (txqs) {
+		for (i = 0; i < num_txqs; i++)
+			ibmvnic_clean_queue_affinity(adapter, txqs[i]);
+	}
+	if (rxqs) {
+		for (i = 0; i < num_rxqs; i++)
+			ibmvnic_clean_queue_affinity(adapter, rxqs[i]);
+	}
+}
+
+static int ibmvnic_set_queue_affinity(struct ibmvnic_sub_crq_queue *queue,
+				      unsigned int *cpu, int *stragglers,
+				      int stride)
+{
+	cpumask_var_t mask;
+	int i;
+	int rc = 0;
+
+	if (!(queue && queue->irq))
+		return rc;
+
+	/* cpumask_var_t is either a pointer or array, allocation works here */
+	if (!zalloc_cpumask_var(&mask, GFP_KERNEL))
+		return -ENOMEM;
+
+	/* while we have extra cpu give one extra to this irq */
+	if (*stragglers) {
+		stride++;
+		(*stragglers)--;
+	}
+	/* atomic write is safer than writing bit by bit directly */
+	for_each_online_cpu_wrap(i, *cpu) {
+		if (!stride--) {
+			/* For the next queue we start from the first
+			 * unused CPU in this queue
+			 */
+			*cpu = i;
+			break;
+		}
+		cpumask_set_cpu(i, mask);
+	}
+
+	/* set queue affinity mask */
+	cpumask_copy(queue->affinity_mask, mask);
+	rc = irq_set_affinity_and_hint(queue->irq, queue->affinity_mask);
+	free_cpumask_var(mask);
+
+	return rc;
+}
+
+/* assumes cpu read lock is held */
+static void ibmvnic_set_affinity(struct ibmvnic_adapter *adapter)
+{
+	struct ibmvnic_sub_crq_queue **rxqs = adapter->rx_scrq;
+	struct ibmvnic_sub_crq_queue **txqs = adapter->tx_scrq;
+	struct ibmvnic_sub_crq_queue *queue;
+	int num_rxqs = adapter->num_active_rx_scrqs, i_rxqs = 0;
+	int num_txqs = adapter->num_active_tx_scrqs, i_txqs = 0;
+	int total_queues, stride, stragglers, i;
+	unsigned int num_cpu, cpu = 0;
+	bool is_rx_queue;
+	int rc = 0;
+
+	netdev_dbg(adapter->netdev, "%s: Setting irq affinity hints", __func__);
+	if (!(adapter->rx_scrq && adapter->tx_scrq)) {
+		netdev_warn(adapter->netdev,
+			    "%s: Set affinity failed, queues not allocated\n",
+			    __func__);
+		return;
+	}
+
+	total_queues = num_rxqs + num_txqs;
+	num_cpu = num_online_cpus();
+	/* number of cpu's assigned per irq */
+	stride = max_t(int, num_cpu / total_queues, 1);
+	/* number of leftover cpu's */
+	stragglers = num_cpu >= total_queues ? num_cpu % total_queues : 0;
+
+	for (i = 0; i < total_queues; i++) {
+		is_rx_queue = false;
+		/* balance core load by alternating rx and tx assignments
+		 * ex: TX0 -> RX0 -> TX1 -> RX1 etc.
+		 */
+		if ((i % 2 == 1 && i_rxqs < num_rxqs) || i_txqs == num_txqs) {
+			queue = rxqs[i_rxqs++];
+			is_rx_queue = true;
+		} else {
+			queue = txqs[i_txqs++];
+		}
+
+		rc = ibmvnic_set_queue_affinity(queue, &cpu, &stragglers,
+						stride);
+		if (rc)
+			goto out;
+
+		if (!queue || is_rx_queue)
+			continue;
+
+		rc = __netif_set_xps_queue(adapter->netdev,
+					   cpumask_bits(queue->affinity_mask),
+					   i_txqs - 1, XPS_CPUS);
+		if (rc)
+			netdev_warn(adapter->netdev, "%s: Set XPS on queue %d failed, rc = %d.\n",
+				    __func__, i_txqs - 1, rc);
+	}
+
+out:
+	if (rc) {
+		netdev_warn(adapter->netdev,
+			    "%s: Set affinity failed, queue addr = %p, IRQ = %d, rc = %d.\n",
+			    __func__, queue, queue->irq, rc);
+		ibmvnic_clean_affinity(adapter);
+	}
+}
+
+static int ibmvnic_cpu_online(unsigned int cpu, struct hlist_node *node)
+{
+	struct ibmvnic_adapter *adapter;
+
+	adapter = hlist_entry_safe(node, struct ibmvnic_adapter, node);
+	ibmvnic_set_affinity(adapter);
+	return 0;
+}
+
+static int ibmvnic_cpu_dead(unsigned int cpu, struct hlist_node *node)
+{
+	struct ibmvnic_adapter *adapter;
+
+	adapter = hlist_entry_safe(node, struct ibmvnic_adapter, node_dead);
+	ibmvnic_set_affinity(adapter);
+	return 0;
+}
+
+static int ibmvnic_cpu_down_prep(unsigned int cpu, struct hlist_node *node)
+{
+	struct ibmvnic_adapter *adapter;
+
+	adapter = hlist_entry_safe(node, struct ibmvnic_adapter, node);
+	ibmvnic_clean_affinity(adapter);
+	return 0;
+}
+
+static enum cpuhp_state ibmvnic_online;
+
+static int ibmvnic_cpu_notif_add(struct ibmvnic_adapter *adapter)
+{
+	int ret;
+
+	ret = cpuhp_state_add_instance_nocalls(ibmvnic_online, &adapter->node);
+	if (ret)
+		return ret;
+	ret = cpuhp_state_add_instance_nocalls(CPUHP_IBMVNIC_DEAD,
+					       &adapter->node_dead);
+	if (!ret)
+		return ret;
+	cpuhp_state_remove_instance_nocalls(ibmvnic_online, &adapter->node);
+	return ret;
+}
+
+static void ibmvnic_cpu_notif_remove(struct ibmvnic_adapter *adapter)
+{
+	cpuhp_state_remove_instance_nocalls(ibmvnic_online, &adapter->node);
+	cpuhp_state_remove_instance_nocalls(CPUHP_IBMVNIC_DEAD,
+					    &adapter->node_dead);
+}
+
 static long h_reg_sub_crq(unsigned long unit_address, unsigned long token,
 			  unsigned long length, unsigned long *number,
 			  unsigned long *irq)
@@ -163,37 +383,142 @@ static long h_reg_sub_crq(unsigned long unit_address, unsigned long token,
 	return rc;
 }
 
-static void reset_long_term_buff(struct ibmvnic_adapter *adapter,
-				 struct ibmvnic_long_term_buff *ltb)
+/**
+ * ibmvnic_wait_for_completion - Check device state and wait for completion
+ * @adapter: private device data
+ * @comp_done: completion structure to wait for
+ * @timeout: time to wait in milliseconds
+ *
+ * Wait for a completion signal or until the timeout limit is reached
+ * while checking that the device is still active.
+ */
+static int ibmvnic_wait_for_completion(struct ibmvnic_adapter *adapter,
+				       struct completion *comp_done,
+				       unsigned long timeout)
 {
-	memset(ltb->buff, 0, ltb->size);
+	struct net_device *netdev;
+	unsigned long div_timeout;
+	u8 retry;
 
-	init_completion(&adapter->fw_done);
-	send_request_map(adapter, ltb->addr, ltb->size, ltb->map_id);
-	wait_for_completion(&adapter->fw_done);
+	netdev = adapter->netdev;
+	retry = 5;
+	div_timeout = msecs_to_jiffies(timeout / retry);
+	while (true) {
+		if (!adapter->crq.active) {
+			netdev_err(netdev, "Device down!\n");
+			return -ENODEV;
+		}
+		if (!retry--)
+			break;
+		if (wait_for_completion_timeout(comp_done, div_timeout))
+			return 0;
+	}
+	netdev_err(netdev, "Operation timed out.\n");
+	return -ETIMEDOUT;
 }
 
+/**
+ * reuse_ltb() - Check if a long term buffer can be reused
+ * @ltb:  The long term buffer to be checked
+ * @size: The size of the long term buffer.
+ *
+ * An LTB can be reused unless its size has changed.
+ *
+ * Return: Return true if the LTB can be reused, false otherwise.
+ */
+static bool reuse_ltb(struct ibmvnic_long_term_buff *ltb, int size)
+{
+	return (ltb->buff && ltb->size == size);
+}
+
+/**
+ * alloc_long_term_buff() - Allocate a long term buffer (LTB)
+ *
+ * @adapter: ibmvnic adapter associated to the LTB
+ * @ltb:     container object for the LTB
+ * @size:    size of the LTB
+ *
+ * Allocate an LTB of the specified size and notify VIOS.
+ *
+ * If the given @ltb already has the correct size, reuse it. Otherwise if
+ * its non-NULL, free it. Then allocate a new one of the correct size.
+ * Notify the VIOS either way since we may now be working with a new VIOS.
+ *
+ * Allocating larger chunks of memory during resets, specially LPM or under
+ * low memory situations can cause resets to fail/timeout and for LPAR to
+ * lose connectivity. So hold onto the LTB even if we fail to communicate
+ * with the VIOS and reuse it on next open. Free LTB when adapter is closed.
+ *
+ * Return: 0 if we were able to allocate the LTB and notify the VIOS and
+ *	   a negative value otherwise.
+ */
 static int alloc_long_term_buff(struct ibmvnic_adapter *adapter,
 				struct ibmvnic_long_term_buff *ltb, int size)
 {
 	struct device *dev = &adapter->vdev->dev;
+	u64 prev = 0;
+	int rc;
 
-	ltb->size = size;
-	ltb->buff = dma_alloc_coherent(dev, ltb->size, &ltb->addr,
-				       GFP_KERNEL);
+	if (!reuse_ltb(ltb, size)) {
+		dev_dbg(dev,
+			"LTB size changed from 0x%llx to 0x%x, reallocating\n",
+			 ltb->size, size);
+		prev = ltb->size;
+		free_long_term_buff(adapter, ltb);
+	}
 
-	if (!ltb->buff) {
-		dev_err(dev, "Couldn't alloc long term buffer\n");
-		return -ENOMEM;
+	if (ltb->buff) {
+		dev_dbg(dev, "Reusing LTB [map %d, size 0x%llx]\n",
+			ltb->map_id, ltb->size);
+	} else {
+		ltb->buff = dma_alloc_coherent(dev, size, &ltb->addr,
+					       GFP_KERNEL);
+		if (!ltb->buff) {
+			dev_err(dev, "Couldn't alloc long term buffer\n");
+			return -ENOMEM;
+		}
+		ltb->size = size;
+
+		ltb->map_id = find_first_zero_bit(adapter->map_ids,
+						  MAX_MAP_ID);
+		bitmap_set(adapter->map_ids, ltb->map_id, 1);
+
+		dev_dbg(dev,
+			"Allocated new LTB [map %d, size 0x%llx was 0x%llx]\n",
+			 ltb->map_id, ltb->size, prev);
 	}
-	ltb->map_id = adapter->map_id;
-	adapter->map_id++;
 
-	init_completion(&adapter->fw_done);
-	send_request_map(adapter, ltb->addr,
-			 ltb->size, ltb->map_id);
-	wait_for_completion(&adapter->fw_done);
-	return 0;
+	/* Ensure ltb is zeroed - specially when reusing it. */
+	memset(ltb->buff, 0, ltb->size);
+
+	mutex_lock(&adapter->fw_lock);
+	adapter->fw_done_rc = 0;
+	reinit_completion(&adapter->fw_done);
+
+	rc = send_request_map(adapter, ltb->addr, ltb->size, ltb->map_id);
+	if (rc) {
+		dev_err(dev, "send_request_map failed, rc = %d\n", rc);
+		goto out;
+	}
+
+	rc = ibmvnic_wait_for_completion(adapter, &adapter->fw_done, 10000);
+	if (rc) {
+		dev_err(dev, "LTB map request aborted or timed out, rc = %d\n",
+			rc);
+		goto out;
+	}
+
+	if (adapter->fw_done_rc) {
+		dev_err(dev, "Couldn't map LTB, rc = %d\n",
+			adapter->fw_done_rc);
+		rc = -EIO;
+		goto out;
+	}
+	rc = 0;
+out:
+	/* don't free LTB on communication error - see function header */
+	mutex_unlock(&adapter->fw_lock);
+	return rc;
 }
 
 static void free_long_term_buff(struct ibmvnic_adapter *adapter,
@@ -204,77 +529,319 @@ static void free_long_term_buff(struct ibmvnic_adapter *adapter,
 	if (!ltb->buff)
 		return;
 
+	/* VIOS automatically unmaps the long term buffer at remote
+	 * end for the following resets:
+	 * FAILOVER, MOBILITY, TIMEOUT.
+	 */
 	if (adapter->reset_reason != VNIC_RESET_FAILOVER &&
-	    adapter->reset_reason != VNIC_RESET_MOBILITY)
+	    adapter->reset_reason != VNIC_RESET_MOBILITY &&
+	    adapter->reset_reason != VNIC_RESET_TIMEOUT)
 		send_request_unmap(adapter, ltb->map_id);
+
 	dma_free_coherent(dev, ltb->size, ltb->buff, ltb->addr);
+
+	ltb->buff = NULL;
+	/* mark this map_id free */
+	bitmap_clear(adapter->map_ids, ltb->map_id, 1);
+	ltb->map_id = 0;
+}
+
+/**
+ * free_ltb_set - free the given set of long term buffers (LTBS)
+ * @adapter: The ibmvnic adapter containing this ltb set
+ * @ltb_set: The ltb_set to be freed
+ *
+ * Free the set of LTBs in the given set.
+ */
+
+static void free_ltb_set(struct ibmvnic_adapter *adapter,
+			 struct ibmvnic_ltb_set *ltb_set)
+{
+	int i;
+
+	for (i = 0; i < ltb_set->num_ltbs; i++)
+		free_long_term_buff(adapter, &ltb_set->ltbs[i]);
+
+	kfree(ltb_set->ltbs);
+	ltb_set->ltbs = NULL;
+	ltb_set->num_ltbs = 0;
+}
+
+/**
+ * alloc_ltb_set() - Allocate a set of long term buffers (LTBs)
+ *
+ * @adapter: ibmvnic adapter associated to the LTB
+ * @ltb_set: container object for the set of LTBs
+ * @num_buffs: Number of buffers in the LTB
+ * @buff_size: Size of each buffer in the LTB
+ *
+ * Allocate a set of LTBs to accommodate @num_buffs buffers of @buff_size
+ * each. We currently cap size each LTB to IBMVNIC_ONE_LTB_SIZE. If the
+ * new set of LTBs have fewer LTBs than the old set, free the excess LTBs.
+ * If new set needs more than in old set, allocate the remaining ones.
+ * Try and reuse as many LTBs as possible and avoid reallocation.
+ *
+ * Any changes to this allocation strategy must be reflected in
+ * map_rxpool_buff_to_ltb() and map_txpool_buff_to_ltb().
+ */
+static int alloc_ltb_set(struct ibmvnic_adapter *adapter,
+			 struct ibmvnic_ltb_set *ltb_set, int num_buffs,
+			 int buff_size)
+{
+	struct device *dev = &adapter->vdev->dev;
+	struct ibmvnic_ltb_set old_set;
+	struct ibmvnic_ltb_set new_set;
+	int rem_size;
+	int tot_size;		/* size of all ltbs */
+	int ltb_size;		/* size of one ltb */
+	int nltbs;
+	int rc;
+	int n;
+	int i;
+
+	dev_dbg(dev, "%s() num_buffs %d, buff_size %d\n", __func__, num_buffs,
+		buff_size);
+
+	ltb_size = rounddown(IBMVNIC_ONE_LTB_SIZE, buff_size);
+	tot_size = num_buffs * buff_size;
+
+	if (ltb_size > tot_size)
+		ltb_size = tot_size;
+
+	nltbs = tot_size / ltb_size;
+	if (tot_size % ltb_size)
+		nltbs++;
+
+	old_set = *ltb_set;
+
+	if (old_set.num_ltbs == nltbs) {
+		new_set = old_set;
+	} else {
+		int tmp = nltbs * sizeof(struct ibmvnic_long_term_buff);
+
+		new_set.ltbs = kzalloc(tmp, GFP_KERNEL);
+		if (!new_set.ltbs)
+			return -ENOMEM;
+
+		new_set.num_ltbs = nltbs;
+
+		/* Free any excess ltbs in old set */
+		for (i = new_set.num_ltbs; i < old_set.num_ltbs; i++)
+			free_long_term_buff(adapter, &old_set.ltbs[i]);
+
+		/* Copy remaining ltbs to new set. All LTBs except the
+		 * last one are of the same size. alloc_long_term_buff()
+		 * will realloc if the size changes.
+		 */
+		n = min(old_set.num_ltbs, new_set.num_ltbs);
+		for (i = 0; i < n; i++)
+			new_set.ltbs[i] = old_set.ltbs[i];
+
+		/* Any additional ltbs in new set will have NULL ltbs for
+		 * now and will be allocated in alloc_long_term_buff().
+		 */
+
+		/* We no longer need the old_set so free it. Note that we
+		 * may have reused some ltbs from old set and freed excess
+		 * ltbs above. So we only need to free the container now
+		 * not the LTBs themselves. (i.e. dont free_ltb_set()!)
+		 */
+		kfree(old_set.ltbs);
+		old_set.ltbs = NULL;
+		old_set.num_ltbs = 0;
+
+		/* Install the new set. If allocations fail below, we will
+		 * retry later and know what size LTBs we need.
+		 */
+		*ltb_set = new_set;
+	}
+
+	i = 0;
+	rem_size = tot_size;
+	while (rem_size) {
+		if (ltb_size > rem_size)
+			ltb_size = rem_size;
+
+		rem_size -= ltb_size;
+
+		rc = alloc_long_term_buff(adapter, &new_set.ltbs[i], ltb_size);
+		if (rc)
+			goto out;
+		i++;
+	}
+
+	WARN_ON(i != new_set.num_ltbs);
+
+	return 0;
+out:
+	/* We may have allocated one/more LTBs before failing and we
+	 * want to try and reuse on next reset. So don't free ltb set.
+	 */
+	return rc;
+}
+
+/**
+ * map_rxpool_buf_to_ltb - Map given rxpool buffer to offset in an LTB.
+ * @rxpool: The receive buffer pool containing buffer
+ * @bufidx: Index of buffer in rxpool
+ * @ltbp: (Output) pointer to the long term buffer containing the buffer
+ * @offset: (Output) offset of buffer in the LTB from @ltbp
+ *
+ * Map the given buffer identified by [rxpool, bufidx] to an LTB in the
+ * pool and its corresponding offset. Assume for now that each LTB is of
+ * different size but could possibly be optimized based on the allocation
+ * strategy in alloc_ltb_set().
+ */
+static void map_rxpool_buf_to_ltb(struct ibmvnic_rx_pool *rxpool,
+				  unsigned int bufidx,
+				  struct ibmvnic_long_term_buff **ltbp,
+				  unsigned int *offset)
+{
+	struct ibmvnic_long_term_buff *ltb;
+	int nbufs;	/* # of buffers in one ltb */
+	int i;
+
+	WARN_ON(bufidx >= rxpool->size);
+
+	for (i = 0; i < rxpool->ltb_set.num_ltbs; i++) {
+		ltb = &rxpool->ltb_set.ltbs[i];
+		nbufs = ltb->size / rxpool->buff_size;
+		if (bufidx < nbufs)
+			break;
+		bufidx -= nbufs;
+	}
+
+	*ltbp = ltb;
+	*offset = bufidx * rxpool->buff_size;
+}
+
+/**
+ * map_txpool_buf_to_ltb - Map given txpool buffer to offset in an LTB.
+ * @txpool: The transmit buffer pool containing buffer
+ * @bufidx: Index of buffer in txpool
+ * @ltbp: (Output) pointer to the long term buffer (LTB) containing the buffer
+ * @offset: (Output) offset of buffer in the LTB from @ltbp
+ *
+ * Map the given buffer identified by [txpool, bufidx] to an LTB in the
+ * pool and its corresponding offset.
+ */
+static void map_txpool_buf_to_ltb(struct ibmvnic_tx_pool *txpool,
+				  unsigned int bufidx,
+				  struct ibmvnic_long_term_buff **ltbp,
+				  unsigned int *offset)
+{
+	struct ibmvnic_long_term_buff *ltb;
+	int nbufs;	/* # of buffers in one ltb */
+	int i;
+
+	WARN_ON_ONCE(bufidx >= txpool->num_buffers);
+
+	for (i = 0; i < txpool->ltb_set.num_ltbs; i++) {
+		ltb = &txpool->ltb_set.ltbs[i];
+		nbufs = ltb->size / txpool->buf_size;
+		if (bufidx < nbufs)
+			break;
+		bufidx -= nbufs;
+	}
+
+	*ltbp = ltb;
+	*offset = bufidx * txpool->buf_size;
 }
 
 static void deactivate_rx_pools(struct ibmvnic_adapter *adapter)
 {
 	int i;
 
-	for (i = 0; i < be32_to_cpu(adapter->login_rsp_buf->num_rxadd_subcrqs);
-	     i++)
+	for (i = 0; i < adapter->num_active_rx_pools; i++)
 		adapter->rx_pool[i].active = 0;
 }
 
+static void ibmvnic_set_safe_max_ind_descs(struct ibmvnic_adapter *adapter)
+{
+	if (adapter->cur_max_ind_descs > IBMVNIC_SAFE_IND_DESC) {
+		netdev_info(adapter->netdev,
+			    "set max ind descs from %u to safe limit %u\n",
+			    adapter->cur_max_ind_descs,
+			    IBMVNIC_SAFE_IND_DESC);
+		adapter->cur_max_ind_descs = IBMVNIC_SAFE_IND_DESC;
+	}
+}
+
 static void replenish_rx_pool(struct ibmvnic_adapter *adapter,
 			      struct ibmvnic_rx_pool *pool)
 {
 	int count = pool->size - atomic_read(&pool->available);
+	u64 handle = adapter->rx_scrq[pool->index]->handle;
 	struct device *dev = &adapter->vdev->dev;
+	struct ibmvnic_ind_xmit_queue *ind_bufp;
+	struct ibmvnic_sub_crq_queue *rx_scrq;
+	struct ibmvnic_long_term_buff *ltb;
+	union sub_crq *sub_crq;
 	int buffers_added = 0;
 	unsigned long lpar_rc;
-	union sub_crq sub_crq;
 	struct sk_buff *skb;
 	unsigned int offset;
 	dma_addr_t dma_addr;
 	unsigned char *dst;
-	u64 *handle_array;
 	int shift = 0;
-	int index;
+	int bufidx;
 	int i;
 
 	if (!pool->active)
 		return;
 
-	handle_array = (u64 *)((u8 *)(adapter->login_rsp_buf) +
-				      be32_to_cpu(adapter->login_rsp_buf->
-				      off_rxadd_subcrqs));
+	rx_scrq = adapter->rx_scrq[pool->index];
+	ind_bufp = &rx_scrq->ind_buf;
 
-	for (i = 0; i < count; ++i) {
-		skb = alloc_skb(pool->buff_size, GFP_ATOMIC);
+	/* netdev_skb_alloc() could have failed after we saved a few skbs
+	 * in the indir_buf and we would not have sent them to VIOS yet.
+	 * To account for them, start the loop at ind_bufp->index rather
+	 * than 0. If we pushed all the skbs to VIOS, ind_bufp->index will
+	 * be 0.
+	 */
+	for (i = ind_bufp->index; i < count; ++i) {
+		bufidx = pool->free_map[pool->next_free];
+
+		/* We maybe reusing the skb from earlier resets. Allocate
+		 * only if necessary. But since the LTB may have changed
+		 * during reset (see init_rx_pools()), update LTB below
+		 * even if reusing skb.
+		 */
+		skb = pool->rx_buff[bufidx].skb;
 		if (!skb) {
-			dev_err(dev, "Couldn't replenish rx buff\n");
-			adapter->replenish_no_mem++;
-			break;
+			skb = netdev_alloc_skb(adapter->netdev,
+					       pool->buff_size);
+			if (!skb) {
+				dev_err(dev, "Couldn't replenish rx buff\n");
+				adapter->replenish_no_mem++;
+				break;
+			}
 		}
 
-		index = pool->free_map[pool->next_free];
-
-		if (pool->rx_buff[index].skb)
-			dev_err(dev, "Inconsistent free_map!\n");
+		pool->free_map[pool->next_free] = IBMVNIC_INVALID_MAP;
+		pool->next_free = (pool->next_free + 1) % pool->size;
 
 		/* Copy the skb to the long term mapped DMA buffer */
-		offset = index * pool->buff_size;
-		dst = pool->long_term_buff.buff + offset;
+		map_rxpool_buf_to_ltb(pool, bufidx, &ltb, &offset);
+		dst = ltb->buff + offset;
 		memset(dst, 0, pool->buff_size);
-		dma_addr = pool->long_term_buff.addr + offset;
-		pool->rx_buff[index].data = dst;
-
-		pool->free_map[pool->next_free] = IBMVNIC_INVALID_MAP;
-		pool->rx_buff[index].dma = dma_addr;
-		pool->rx_buff[index].skb = skb;
-		pool->rx_buff[index].pool_index = pool->index;
-		pool->rx_buff[index].size = pool->buff_size;
-
-		memset(&sub_crq, 0, sizeof(sub_crq));
-		sub_crq.rx_add.first = IBMVNIC_CRQ_CMD;
-		sub_crq.rx_add.correlator =
-		    cpu_to_be64((u64)&pool->rx_buff[index]);
-		sub_crq.rx_add.ioba = cpu_to_be32(dma_addr);
-		sub_crq.rx_add.map_id = pool->long_term_buff.map_id;
+		dma_addr = ltb->addr + offset;
+
+		/* add the skb to an rx_buff in the pool */
+		pool->rx_buff[bufidx].data = dst;
+		pool->rx_buff[bufidx].dma = dma_addr;
+		pool->rx_buff[bufidx].skb = skb;
+		pool->rx_buff[bufidx].pool_index = pool->index;
+		pool->rx_buff[bufidx].size = pool->buff_size;
+
+		/* queue the rx_buff for the next send_subcrq_indirect */
+		sub_crq = &ind_bufp->indir_arr[ind_bufp->index++];
+		memset(sub_crq, 0, sizeof(*sub_crq));
+		sub_crq->rx_add.first = IBMVNIC_CRQ_CMD;
+		sub_crq->rx_add.correlator =
+		    cpu_to_be64((u64)&pool->rx_buff[bufidx]);
+		sub_crq->rx_add.ioba = cpu_to_be32(dma_addr);
+		sub_crq->rx_add.map_id = ltb->map_id;
 
 		/* The length field of the sCRQ is defined to be 24 bits so the
 		 * buffer size needs to be left shifted by a byte before it is
@@ -284,36 +851,57 @@ static void replenish_rx_pool(struct ibmvnic_adapter *adapter,
 #ifdef __LITTLE_ENDIAN__
 		shift = 8;
 #endif
-		sub_crq.rx_add.len = cpu_to_be32(pool->buff_size << shift);
-
-		lpar_rc = send_subcrq(adapter, handle_array[pool->index],
-				      &sub_crq);
-		if (lpar_rc != H_SUCCESS)
-			goto failure;
-
-		buffers_added++;
-		adapter->replenish_add_buff_success++;
-		pool->next_free = (pool->next_free + 1) % pool->size;
+		sub_crq->rx_add.len = cpu_to_be32(pool->buff_size << shift);
+
+		/* if send_subcrq_indirect queue is full, flush to VIOS */
+		if (ind_bufp->index == adapter->cur_max_ind_descs ||
+		    i == count - 1) {
+			lpar_rc =
+				send_subcrq_indirect(adapter, handle,
+						     (u64)ind_bufp->indir_dma,
+						     (u64)ind_bufp->index);
+			if (lpar_rc != H_SUCCESS)
+				goto failure;
+			buffers_added += ind_bufp->index;
+			adapter->replenish_add_buff_success += ind_bufp->index;
+			ind_bufp->index = 0;
+		}
 	}
 	atomic_add(buffers_added, &pool->available);
 	return;
 
 failure:
-	dev_info(dev, "replenish pools failure\n");
-	pool->free_map[pool->next_free] = index;
-	pool->rx_buff[index].skb = NULL;
-	if (!dma_mapping_error(dev, dma_addr))
-		dma_unmap_single(dev, dma_addr, pool->buff_size,
-				 DMA_FROM_DEVICE);
+	if (lpar_rc != H_PARAMETER && lpar_rc != H_CLOSED)
+		dev_err_ratelimited(dev, "rx: replenish packet buffer failed\n");
 
-	dev_kfree_skb_any(skb);
-	adapter->replenish_add_buff_failure++;
-	atomic_add(buffers_added, &pool->available);
+	/* Detect platform limit H_PARAMETER */
+	if (lpar_rc == H_PARAMETER)
+		ibmvnic_set_safe_max_ind_descs(adapter);
 
-	if (lpar_rc == H_CLOSED) {
+	/* For all error case, temporarily drop only this batch
+	 * Rely on TCP/IP retransmissions to retry and recover
+	 */
+	for (i = ind_bufp->index - 1; i >= 0; --i) {
+		struct ibmvnic_rx_buff *rx_buff;
+
+		pool->next_free = pool->next_free == 0 ?
+				  pool->size - 1 : pool->next_free - 1;
+		sub_crq = &ind_bufp->indir_arr[i];
+		rx_buff = (struct ibmvnic_rx_buff *)
+				be64_to_cpu(sub_crq->rx_add.correlator);
+		bufidx = (int)(rx_buff - pool->rx_buff);
+		pool->free_map[pool->next_free] = bufidx;
+		dev_kfree_skb_any(pool->rx_buff[bufidx].skb);
+		pool->rx_buff[bufidx].skb = NULL;
+	}
+	adapter->replenish_add_buff_failure += ind_bufp->index;
+	atomic_add(buffers_added, &pool->available);
+	ind_bufp->index = 0;
+	if (lpar_rc == H_CLOSED || adapter->failover_pending) {
 		/* Disable buffer pool replenishment and report carrier off if
-		 * queue is closed. Firmware guarantees that a signal will
-		 * be sent to the driver, triggering a reset.
+		 * queue is closed or pending failover.
+		 * Firmware guarantees that a signal will be sent to the
+		 * driver, triggering a reset.
 		 */
 		deactivate_rx_pools(adapter);
 		netif_carrier_off(adapter->netdev);
@@ -325,11 +913,39 @@ static void replenish_pools(struct ibmvnic_adapter *adapter)
 	int i;
 
 	adapter->replenish_task_cycles++;
-	for (i = 0; i < be32_to_cpu(adapter->login_rsp_buf->num_rxadd_subcrqs);
-	     i++) {
+	for (i = 0; i < adapter->num_active_rx_pools; i++) {
 		if (adapter->rx_pool[i].active)
 			replenish_rx_pool(adapter, &adapter->rx_pool[i]);
 	}
+
+	netdev_dbg(adapter->netdev, "Replenished %d pools\n", i);
+}
+
+static void release_stats_buffers(struct ibmvnic_adapter *adapter)
+{
+	kfree(adapter->tx_stats_buffers);
+	kfree(adapter->rx_stats_buffers);
+	adapter->tx_stats_buffers = NULL;
+	adapter->rx_stats_buffers = NULL;
+}
+
+static int init_stats_buffers(struct ibmvnic_adapter *adapter)
+{
+	adapter->tx_stats_buffers =
+				kcalloc(IBMVNIC_MAX_QUEUES,
+					sizeof(struct ibmvnic_tx_queue_stats),
+					GFP_KERNEL);
+	if (!adapter->tx_stats_buffers)
+		return -ENOMEM;
+
+	adapter->rx_stats_buffers =
+				kcalloc(IBMVNIC_MAX_QUEUES,
+					sizeof(struct ibmvnic_rx_queue_stats),
+					GFP_KERNEL);
+	if (!adapter->rx_stats_buffers)
+		return -ENOMEM;
+
+	return 0;
 }
 
 static void release_stats_token(struct ibmvnic_adapter *adapter)
@@ -349,68 +965,52 @@ static int init_stats_token(struct ibmvnic_adapter *adapter)
 {
 	struct device *dev = &adapter->vdev->dev;
 	dma_addr_t stok;
+	int rc;
 
 	stok = dma_map_single(dev, &adapter->stats,
 			      sizeof(struct ibmvnic_statistics),
 			      DMA_FROM_DEVICE);
-	if (dma_mapping_error(dev, stok)) {
-		dev_err(dev, "Couldn't map stats buffer\n");
-		return -1;
+	rc = dma_mapping_error(dev, stok);
+	if (rc) {
+		dev_err(dev, "Couldn't map stats buffer, rc = %d\n", rc);
+		return rc;
 	}
 
 	adapter->stats_token = stok;
+	netdev_dbg(adapter->netdev, "Stats token initialized (%llx)\n", stok);
 	return 0;
 }
 
-static int reset_rx_pools(struct ibmvnic_adapter *adapter)
-{
-	struct ibmvnic_rx_pool *rx_pool;
-	int rx_scrqs;
-	int i, j;
-
-	rx_scrqs = be32_to_cpu(adapter->login_rsp_buf->num_rxadd_subcrqs);
-	for (i = 0; i < rx_scrqs; i++) {
-		rx_pool = &adapter->rx_pool[i];
-
-		reset_long_term_buff(adapter, &rx_pool->long_term_buff);
-
-		for (j = 0; j < rx_pool->size; j++)
-			rx_pool->free_map[j] = j;
-
-		memset(rx_pool->rx_buff, 0,
-		       rx_pool->size * sizeof(struct ibmvnic_rx_buff));
-
-		atomic_set(&rx_pool->available, 0);
-		rx_pool->next_alloc = 0;
-		rx_pool->next_free = 0;
-	}
-
-	return 0;
-}
-
+/**
+ * release_rx_pools() - Release any rx pools attached to @adapter.
+ * @adapter: ibmvnic adapter
+ *
+ * Safe to call this multiple times - even if no pools are attached.
+ */
 static void release_rx_pools(struct ibmvnic_adapter *adapter)
 {
 	struct ibmvnic_rx_pool *rx_pool;
-	int rx_scrqs;
 	int i, j;
 
 	if (!adapter->rx_pool)
 		return;
 
-	rx_scrqs = be32_to_cpu(adapter->login_rsp_buf->num_rxadd_subcrqs);
-	for (i = 0; i < rx_scrqs; i++) {
+	for (i = 0; i < adapter->num_active_rx_pools; i++) {
 		rx_pool = &adapter->rx_pool[i];
 
+		netdev_dbg(adapter->netdev, "Releasing rx_pool[%d]\n", i);
+
 		kfree(rx_pool->free_map);
-		free_long_term_buff(adapter, &rx_pool->long_term_buff);
+
+		free_ltb_set(adapter, &rx_pool->ltb_set);
 
 		if (!rx_pool->rx_buff)
 			continue;
 
 		for (j = 0; j < rx_pool->size; j++) {
 			if (rx_pool->rx_buff[j].skb) {
-				dev_kfree_skb_any(rx_pool->rx_buff[i].skb);
-				rx_pool->rx_buff[i].skb = NULL;
+				dev_kfree_skb_any(rx_pool->rx_buff[j].skb);
+				rx_pool->rx_buff[j].skb = NULL;
 			}
 		}
 
@@ -419,48 +1019,113 @@ static void release_rx_pools(struct ibmvnic_adapter *adapter)
 
 	kfree(adapter->rx_pool);
 	adapter->rx_pool = NULL;
+	adapter->num_active_rx_pools = 0;
+	adapter->prev_rx_pool_size = 0;
 }
 
+/**
+ * reuse_rx_pools() - Check if the existing rx pools can be reused.
+ * @adapter: ibmvnic adapter
+ *
+ * Check if the existing rx pools in the adapter can be reused. The
+ * pools can be reused if the pool parameters (number of pools,
+ * number of buffers in the pool and size of each buffer) have not
+ * changed.
+ *
+ * NOTE: This assumes that all pools have the same number of buffers
+ *       which is the case currently. If that changes, we must fix this.
+ *
+ * Return: true if the rx pools can be reused, false otherwise.
+ */
+static bool reuse_rx_pools(struct ibmvnic_adapter *adapter)
+{
+	u64 old_num_pools, new_num_pools;
+	u64 old_pool_size, new_pool_size;
+	u64 old_buff_size, new_buff_size;
+
+	if (!adapter->rx_pool)
+		return false;
+
+	old_num_pools = adapter->num_active_rx_pools;
+	new_num_pools = adapter->req_rx_queues;
+
+	old_pool_size = adapter->prev_rx_pool_size;
+	new_pool_size = adapter->req_rx_add_entries_per_subcrq;
+
+	old_buff_size = adapter->prev_rx_buf_sz;
+	new_buff_size = adapter->cur_rx_buf_sz;
+
+	if (old_buff_size != new_buff_size ||
+	    old_num_pools != new_num_pools ||
+	    old_pool_size != new_pool_size)
+		return false;
+
+	return true;
+}
+
+/**
+ * init_rx_pools(): Initialize the set of receiver pools in the adapter.
+ * @netdev: net device associated with the vnic interface
+ *
+ * Initialize the set of receiver pools in the ibmvnic adapter associated
+ * with the net_device @netdev. If possible, reuse the existing rx pools.
+ * Otherwise free any existing pools and  allocate a new set of pools
+ * before initializing them.
+ *
+ * Return: 0 on success and negative value on error.
+ */
 static int init_rx_pools(struct net_device *netdev)
 {
 	struct ibmvnic_adapter *adapter = netdev_priv(netdev);
 	struct device *dev = &adapter->vdev->dev;
 	struct ibmvnic_rx_pool *rx_pool;
-	int rxadd_subcrqs;
-	u64 *size_array;
-	int i, j;
+	u64 num_pools;
+	u64 pool_size;		/* # of buffers in one pool */
+	u64 buff_size;
+	int i, j, rc;
+
+	pool_size = adapter->req_rx_add_entries_per_subcrq;
+	num_pools = adapter->req_rx_queues;
+	buff_size = adapter->cur_rx_buf_sz;
+
+	if (reuse_rx_pools(adapter)) {
+		dev_dbg(dev, "Reusing rx pools\n");
+		goto update_ltb;
+	}
 
-	rxadd_subcrqs =
-		be32_to_cpu(adapter->login_rsp_buf->num_rxadd_subcrqs);
-	size_array = (u64 *)((u8 *)(adapter->login_rsp_buf) +
-		be32_to_cpu(adapter->login_rsp_buf->off_rxadd_buff_size));
+	/* Allocate/populate the pools. */
+	release_rx_pools(adapter);
 
-	adapter->rx_pool = kcalloc(rxadd_subcrqs,
+	adapter->rx_pool = kcalloc(num_pools,
 				   sizeof(struct ibmvnic_rx_pool),
 				   GFP_KERNEL);
 	if (!adapter->rx_pool) {
 		dev_err(dev, "Failed to allocate rx pools\n");
-		return -1;
+		return -ENOMEM;
 	}
 
-	for (i = 0; i < rxadd_subcrqs; i++) {
+	/* Set num_active_rx_pools early. If we fail below after partial
+	 * allocation, release_rx_pools() will know how many to look for.
+	 */
+	adapter->num_active_rx_pools = num_pools;
+
+	for (i = 0; i < num_pools; i++) {
 		rx_pool = &adapter->rx_pool[i];
 
 		netdev_dbg(adapter->netdev,
-			   "Initializing rx_pool %d, %lld buffs, %lld bytes each\n",
-			   i, adapter->req_rx_add_entries_per_subcrq,
-			   be64_to_cpu(size_array[i]));
+			   "Initializing rx_pool[%d], %lld buffs, %lld bytes each\n",
+			   i, pool_size, buff_size);
 
-		rx_pool->size = adapter->req_rx_add_entries_per_subcrq;
+		rx_pool->size = pool_size;
 		rx_pool->index = i;
-		rx_pool->buff_size = be64_to_cpu(size_array[i]);
-		rx_pool->active = 1;
+		rx_pool->buff_size = ALIGN(buff_size, L1_CACHE_BYTES);
 
 		rx_pool->free_map = kcalloc(rx_pool->size, sizeof(int),
 					    GFP_KERNEL);
 		if (!rx_pool->free_map) {
-			release_rx_pools(adapter);
-			return -1;
+			dev_err(dev, "Couldn't alloc free_map %d\n", i);
+			rc = -ENOMEM;
+			goto out_release;
 		}
 
 		rx_pool->rx_buff = kcalloc(rx_pool->size,
@@ -468,137 +1133,310 @@ static int init_rx_pools(struct net_device *netdev)
 					   GFP_KERNEL);
 		if (!rx_pool->rx_buff) {
 			dev_err(dev, "Couldn't alloc rx buffers\n");
-			release_rx_pools(adapter);
-			return -1;
+			rc = -ENOMEM;
+			goto out_release;
 		}
+	}
 
-		if (alloc_long_term_buff(adapter, &rx_pool->long_term_buff,
-					 rx_pool->size * rx_pool->buff_size)) {
-			release_rx_pools(adapter);
-			return -1;
-		}
+	adapter->prev_rx_pool_size = pool_size;
+	adapter->prev_rx_buf_sz = adapter->cur_rx_buf_sz;
+
+update_ltb:
+	for (i = 0; i < num_pools; i++) {
+		rx_pool = &adapter->rx_pool[i];
+		dev_dbg(dev, "Updating LTB for rx pool %d [%d, %d]\n",
+			i, rx_pool->size, rx_pool->buff_size);
+
+		rc = alloc_ltb_set(adapter, &rx_pool->ltb_set,
+				   rx_pool->size, rx_pool->buff_size);
+		if (rc)
+			goto out;
+
+		for (j = 0; j < rx_pool->size; ++j) {
+			struct ibmvnic_rx_buff *rx_buff;
 
-		for (j = 0; j < rx_pool->size; ++j)
 			rx_pool->free_map[j] = j;
 
+			/* NOTE: Don't clear rx_buff->skb here - will leak
+			 * memory! replenish_rx_pool() will reuse skbs or
+			 * allocate as necessary.
+			 */
+			rx_buff = &rx_pool->rx_buff[j];
+			rx_buff->dma = 0;
+			rx_buff->data = 0;
+			rx_buff->size = 0;
+			rx_buff->pool_index = 0;
+		}
+
+		/* Mark pool "empty" so replenish_rx_pools() will
+		 * update the LTB info for each buffer
+		 */
 		atomic_set(&rx_pool->available, 0);
 		rx_pool->next_alloc = 0;
 		rx_pool->next_free = 0;
+		/* replenish_rx_pool() may have called deactivate_rx_pools()
+		 * on failover. Ensure pool is active now.
+		 */
+		rx_pool->active = 1;
 	}
-
 	return 0;
+out_release:
+	release_rx_pools(adapter);
+out:
+	/* We failed to allocate one or more LTBs or map them on the VIOS.
+	 * Hold onto the pools and any LTBs that we did allocate/map.
+	 */
+	return rc;
 }
 
-static int reset_tx_pools(struct ibmvnic_adapter *adapter)
+static void release_vpd_data(struct ibmvnic_adapter *adapter)
 {
-	struct ibmvnic_tx_pool *tx_pool;
-	int tx_scrqs;
-	int i, j;
-
-	tx_scrqs = be32_to_cpu(adapter->login_rsp_buf->num_txsubm_subcrqs);
-	for (i = 0; i < tx_scrqs; i++) {
-		tx_pool = &adapter->tx_pool[i];
-
-		reset_long_term_buff(adapter, &tx_pool->long_term_buff);
-
-		memset(tx_pool->tx_buff, 0,
-		       adapter->req_tx_entries_per_subcrq *
-		       sizeof(struct ibmvnic_tx_buff));
+	if (!adapter->vpd)
+		return;
 
-		for (j = 0; j < adapter->req_tx_entries_per_subcrq; j++)
-			tx_pool->free_map[j] = j;
+	kfree(adapter->vpd->buff);
+	kfree(adapter->vpd);
 
-		tx_pool->consumer_index = 0;
-		tx_pool->producer_index = 0;
-	}
+	adapter->vpd = NULL;
+}
 
-	return 0;
+static void release_one_tx_pool(struct ibmvnic_adapter *adapter,
+				struct ibmvnic_tx_pool *tx_pool)
+{
+	kfree(tx_pool->tx_buff);
+	kfree(tx_pool->free_map);
+	free_ltb_set(adapter, &tx_pool->ltb_set);
 }
 
+/**
+ * release_tx_pools() - Release any tx pools attached to @adapter.
+ * @adapter: ibmvnic adapter
+ *
+ * Safe to call this multiple times - even if no pools are attached.
+ */
 static void release_tx_pools(struct ibmvnic_adapter *adapter)
 {
-	struct ibmvnic_tx_pool *tx_pool;
-	int i, tx_scrqs;
+	int i;
 
+	/* init_tx_pools() ensures that ->tx_pool and ->tso_pool are
+	 * both NULL or both non-NULL. So we only need to check one.
+	 */
 	if (!adapter->tx_pool)
 		return;
 
-	tx_scrqs = be32_to_cpu(adapter->login_rsp_buf->num_txsubm_subcrqs);
-	for (i = 0; i < tx_scrqs; i++) {
-		tx_pool = &adapter->tx_pool[i];
-		kfree(tx_pool->tx_buff);
-		free_long_term_buff(adapter, &tx_pool->long_term_buff);
-		kfree(tx_pool->free_map);
+	for (i = 0; i < adapter->num_active_tx_pools; i++) {
+		release_one_tx_pool(adapter, &adapter->tx_pool[i]);
+		release_one_tx_pool(adapter, &adapter->tso_pool[i]);
 	}
 
 	kfree(adapter->tx_pool);
 	adapter->tx_pool = NULL;
+	kfree(adapter->tso_pool);
+	adapter->tso_pool = NULL;
+	adapter->num_active_tx_pools = 0;
+	adapter->prev_tx_pool_size = 0;
+}
+
+static int init_one_tx_pool(struct net_device *netdev,
+			    struct ibmvnic_tx_pool *tx_pool,
+			    int pool_size, int buf_size)
+{
+	int i;
+
+	tx_pool->tx_buff = kcalloc(pool_size,
+				   sizeof(struct ibmvnic_tx_buff),
+				   GFP_KERNEL);
+	if (!tx_pool->tx_buff)
+		return -ENOMEM;
+
+	tx_pool->free_map = kcalloc(pool_size, sizeof(int), GFP_KERNEL);
+	if (!tx_pool->free_map) {
+		kfree(tx_pool->tx_buff);
+		tx_pool->tx_buff = NULL;
+		return -ENOMEM;
+	}
+
+	for (i = 0; i < pool_size; i++)
+		tx_pool->free_map[i] = i;
+
+	tx_pool->consumer_index = 0;
+	tx_pool->producer_index = 0;
+	tx_pool->num_buffers = pool_size;
+	tx_pool->buf_size = buf_size;
+
+	return 0;
+}
+
+/**
+ * reuse_tx_pools() - Check if the existing tx pools can be reused.
+ * @adapter: ibmvnic adapter
+ *
+ * Check if the existing tx pools in the adapter can be reused. The
+ * pools can be reused if the pool parameters (number of pools,
+ * number of buffers in the pool and mtu) have not changed.
+ *
+ * NOTE: This assumes that all pools have the same number of buffers
+ *       which is the case currently. If that changes, we must fix this.
+ *
+ * Return: true if the tx pools can be reused, false otherwise.
+ */
+static bool reuse_tx_pools(struct ibmvnic_adapter *adapter)
+{
+	u64 old_num_pools, new_num_pools;
+	u64 old_pool_size, new_pool_size;
+	u64 old_mtu, new_mtu;
+
+	if (!adapter->tx_pool)
+		return false;
+
+	old_num_pools = adapter->num_active_tx_pools;
+	new_num_pools = adapter->num_active_tx_scrqs;
+	old_pool_size = adapter->prev_tx_pool_size;
+	new_pool_size = adapter->req_tx_entries_per_subcrq;
+	old_mtu = adapter->prev_mtu;
+	new_mtu = adapter->req_mtu;
+
+	if (old_mtu != new_mtu ||
+	    old_num_pools != new_num_pools ||
+	    old_pool_size != new_pool_size)
+		return false;
+
+	return true;
 }
 
+/**
+ * init_tx_pools(): Initialize the set of transmit pools in the adapter.
+ * @netdev: net device associated with the vnic interface
+ *
+ * Initialize the set of transmit pools in the ibmvnic adapter associated
+ * with the net_device @netdev. If possible, reuse the existing tx pools.
+ * Otherwise free any existing pools and  allocate a new set of pools
+ * before initializing them.
+ *
+ * Return: 0 on success and negative value on error.
+ */
 static int init_tx_pools(struct net_device *netdev)
 {
 	struct ibmvnic_adapter *adapter = netdev_priv(netdev);
 	struct device *dev = &adapter->vdev->dev;
-	struct ibmvnic_tx_pool *tx_pool;
-	int tx_subcrqs;
-	int i, j;
+	int num_pools;
+	u64 pool_size;		/* # of buffers in pool */
+	u64 buff_size;
+	int i, j, rc;
+
+	num_pools = adapter->req_tx_queues;
+
+	/* We must notify the VIOS about the LTB on all resets - but we only
+	 * need to alloc/populate pools if either the number of buffers or
+	 * size of each buffer in the pool has changed.
+	 */
+	if (reuse_tx_pools(adapter)) {
+		netdev_dbg(netdev, "Reusing tx pools\n");
+		goto update_ltb;
+	}
 
-	tx_subcrqs = be32_to_cpu(adapter->login_rsp_buf->num_txsubm_subcrqs);
-	adapter->tx_pool = kcalloc(tx_subcrqs,
+	/* Allocate/populate the pools. */
+	release_tx_pools(adapter);
+
+	pool_size = adapter->req_tx_entries_per_subcrq;
+	num_pools = adapter->num_active_tx_scrqs;
+
+	adapter->tx_pool = kcalloc(num_pools,
 				   sizeof(struct ibmvnic_tx_pool), GFP_KERNEL);
 	if (!adapter->tx_pool)
-		return -1;
+		return -ENOMEM;
 
-	for (i = 0; i < tx_subcrqs; i++) {
-		tx_pool = &adapter->tx_pool[i];
-		tx_pool->tx_buff = kcalloc(adapter->req_tx_entries_per_subcrq,
-					   sizeof(struct ibmvnic_tx_buff),
-					   GFP_KERNEL);
-		if (!tx_pool->tx_buff) {
-			dev_err(dev, "tx pool buffer allocation failed\n");
-			release_tx_pools(adapter);
-			return -1;
-		}
+	adapter->tso_pool = kcalloc(num_pools,
+				    sizeof(struct ibmvnic_tx_pool), GFP_KERNEL);
+	/* To simplify release_tx_pools() ensure that ->tx_pool and
+	 * ->tso_pool are either both NULL or both non-NULL.
+	 */
+	if (!adapter->tso_pool) {
+		kfree(adapter->tx_pool);
+		adapter->tx_pool = NULL;
+		return -ENOMEM;
+	}
 
-		if (alloc_long_term_buff(adapter, &tx_pool->long_term_buff,
-					 adapter->req_tx_entries_per_subcrq *
-					 adapter->req_mtu)) {
-			release_tx_pools(adapter);
-			return -1;
-		}
+	/* Set num_active_tx_pools early. If we fail below after partial
+	 * allocation, release_tx_pools() will know how many to look for.
+	 */
+	adapter->num_active_tx_pools = num_pools;
 
-		tx_pool->free_map = kcalloc(adapter->req_tx_entries_per_subcrq,
-					    sizeof(int), GFP_KERNEL);
-		if (!tx_pool->free_map) {
-			release_tx_pools(adapter);
-			return -1;
-		}
+	buff_size = adapter->req_mtu + VLAN_HLEN;
+	buff_size = ALIGN(buff_size, L1_CACHE_BYTES);
 
-		for (j = 0; j < adapter->req_tx_entries_per_subcrq; j++)
-			tx_pool->free_map[j] = j;
+	for (i = 0; i < num_pools; i++) {
+		dev_dbg(dev, "Init tx pool %d [%llu, %llu]\n",
+			i, adapter->req_tx_entries_per_subcrq, buff_size);
+
+		rc = init_one_tx_pool(netdev, &adapter->tx_pool[i],
+				      pool_size, buff_size);
+		if (rc)
+			goto out_release;
+
+		rc = init_one_tx_pool(netdev, &adapter->tso_pool[i],
+				      IBMVNIC_TSO_BUFS,
+				      IBMVNIC_TSO_BUF_SZ);
+		if (rc)
+			goto out_release;
+	}
+
+	adapter->prev_tx_pool_size = pool_size;
+	adapter->prev_mtu = adapter->req_mtu;
+
+update_ltb:
+	/* NOTE: All tx_pools have the same number of buffers (which is
+	 *       same as pool_size). All tso_pools have IBMVNIC_TSO_BUFS
+	 *       buffers (see calls init_one_tx_pool() for these).
+	 *       For consistency, we use tx_pool->num_buffers and
+	 *       tso_pool->num_buffers below.
+	 */
+	rc = -1;
+	for (i = 0; i < num_pools; i++) {
+		struct ibmvnic_tx_pool *tso_pool;
+		struct ibmvnic_tx_pool *tx_pool;
+
+		tx_pool = &adapter->tx_pool[i];
+
+		dev_dbg(dev, "Updating LTB for tx pool %d [%d, %d]\n",
+			i, tx_pool->num_buffers, tx_pool->buf_size);
+
+		rc = alloc_ltb_set(adapter, &tx_pool->ltb_set,
+				   tx_pool->num_buffers, tx_pool->buf_size);
+		if (rc)
+			goto out;
 
 		tx_pool->consumer_index = 0;
 		tx_pool->producer_index = 0;
-	}
 
-	return 0;
-}
+		for (j = 0; j < tx_pool->num_buffers; j++)
+			tx_pool->free_map[j] = j;
 
-static void release_error_buffers(struct ibmvnic_adapter *adapter)
-{
-	struct device *dev = &adapter->vdev->dev;
-	struct ibmvnic_error_buff *error_buff, *tmp;
-	unsigned long flags;
+		tso_pool = &adapter->tso_pool[i];
 
-	spin_lock_irqsave(&adapter->error_list_lock, flags);
-	list_for_each_entry_safe(error_buff, tmp, &adapter->errors, list) {
-		list_del(&error_buff->list);
-		dma_unmap_single(dev, error_buff->dma, error_buff->len,
-				 DMA_FROM_DEVICE);
-		kfree(error_buff->buff);
-		kfree(error_buff);
+		dev_dbg(dev, "Updating LTB for tso pool %d [%d, %d]\n",
+			i, tso_pool->num_buffers, tso_pool->buf_size);
+
+		rc = alloc_ltb_set(adapter, &tso_pool->ltb_set,
+				   tso_pool->num_buffers, tso_pool->buf_size);
+		if (rc)
+			goto out;
+
+		tso_pool->consumer_index = 0;
+		tso_pool->producer_index = 0;
+
+		for (j = 0; j < tso_pool->num_buffers; j++)
+			tso_pool->free_map[j] = j;
 	}
-	spin_unlock_irqrestore(&adapter->error_list_lock, flags);
+
+	return 0;
+out_release:
+	release_tx_pools(adapter);
+out:
+	/* We failed to allocate one or more LTBs or map them on the VIOS.
+	 * Hold onto the pools and any LTBs that we did allocate/map.
+	 */
+	return rc;
 }
 
 static void ibmvnic_napi_enable(struct ibmvnic_adapter *adapter)
@@ -621,71 +1459,257 @@ static void ibmvnic_napi_disable(struct ibmvnic_adapter *adapter)
 	if (!adapter->napi_enabled)
 		return;
 
-	for (i = 0; i < adapter->req_rx_queues; i++)
+	for (i = 0; i < adapter->req_rx_queues; i++) {
+		netdev_dbg(adapter->netdev, "Disabling napi[%d]\n", i);
 		napi_disable(&adapter->napi[i]);
+	}
 
 	adapter->napi_enabled = false;
 }
 
+static int init_napi(struct ibmvnic_adapter *adapter)
+{
+	int i;
+
+	adapter->napi = kcalloc(adapter->req_rx_queues,
+				sizeof(struct napi_struct), GFP_KERNEL);
+	if (!adapter->napi)
+		return -ENOMEM;
+
+	for (i = 0; i < adapter->req_rx_queues; i++) {
+		netdev_dbg(adapter->netdev, "Adding napi[%d]\n", i);
+		netif_napi_add(adapter->netdev, &adapter->napi[i],
+			       ibmvnic_poll);
+	}
+
+	adapter->num_active_rx_napi = adapter->req_rx_queues;
+	return 0;
+}
+
+static void release_napi(struct ibmvnic_adapter *adapter)
+{
+	int i;
+
+	if (!adapter->napi)
+		return;
+
+	for (i = 0; i < adapter->num_active_rx_napi; i++) {
+		netdev_dbg(adapter->netdev, "Releasing napi[%d]\n", i);
+		netif_napi_del(&adapter->napi[i]);
+	}
+
+	kfree(adapter->napi);
+	adapter->napi = NULL;
+	adapter->num_active_rx_napi = 0;
+	adapter->napi_enabled = false;
+}
+
+static const char *adapter_state_to_string(enum vnic_state state)
+{
+	switch (state) {
+	case VNIC_PROBING:
+		return "PROBING";
+	case VNIC_PROBED:
+		return "PROBED";
+	case VNIC_OPENING:
+		return "OPENING";
+	case VNIC_OPEN:
+		return "OPEN";
+	case VNIC_CLOSING:
+		return "CLOSING";
+	case VNIC_CLOSED:
+		return "CLOSED";
+	case VNIC_REMOVING:
+		return "REMOVING";
+	case VNIC_REMOVED:
+		return "REMOVED";
+	case VNIC_DOWN:
+		return "DOWN";
+	}
+	return "UNKNOWN";
+}
+
 static int ibmvnic_login(struct net_device *netdev)
 {
+	unsigned long flags, timeout = msecs_to_jiffies(20000);
 	struct ibmvnic_adapter *adapter = netdev_priv(netdev);
-	unsigned long timeout = msecs_to_jiffies(30000);
-	struct device *dev = &adapter->vdev->dev;
+	int retry_count = 0;
+	int retries = 10;
+	bool retry;
+	int rc;
 
 	do {
-		if (adapter->renegotiate) {
-			adapter->renegotiate = false;
-			release_sub_crqs(adapter);
+		retry = false;
+		if (retry_count > retries) {
+			netdev_warn(netdev, "Login attempts exceeded\n");
+			return -EACCES;
+		}
+
+		adapter->init_done_rc = 0;
+		reinit_completion(&adapter->init_done);
+		rc = send_login(adapter);
+		if (rc)
+			return rc;
 
+		if (!wait_for_completion_timeout(&adapter->init_done,
+						 timeout)) {
+			netdev_warn(netdev, "Login timed out\n");
+			adapter->login_pending = false;
+			goto partial_reset;
+		}
+
+		if (adapter->init_done_rc == ABORTED) {
+			netdev_warn(netdev, "Login aborted, retrying...\n");
+			retry = true;
+			adapter->init_done_rc = 0;
+			retry_count++;
+			/* FW or device may be busy, so
+			 * wait a bit before retrying login
+			 */
+			msleep(500);
+		} else if (adapter->init_done_rc == PARTIALSUCCESS) {
+			retry_count++;
+			release_sub_crqs(adapter, 1);
+
+			retry = true;
+			netdev_dbg(netdev,
+				   "Received partial success, retrying...\n");
+			adapter->init_done_rc = 0;
 			reinit_completion(&adapter->init_done);
-			send_cap_queries(adapter);
+			send_query_cap(adapter);
 			if (!wait_for_completion_timeout(&adapter->init_done,
 							 timeout)) {
-				dev_err(dev, "Capabilities query timeout\n");
-				return -1;
+				netdev_warn(netdev,
+					    "Capabilities query timed out\n");
+				return -ETIMEDOUT;
 			}
-		}
 
-		reinit_completion(&adapter->init_done);
-		send_login(adapter);
-		if (!wait_for_completion_timeout(&adapter->init_done,
-						 timeout)) {
-			dev_err(dev, "Login timeout\n");
-			return -1;
+			rc = init_sub_crqs(adapter);
+			if (rc) {
+				netdev_warn(netdev,
+					    "SCRQ initialization failed\n");
+				return rc;
+			}
+
+			rc = init_sub_crq_irqs(adapter);
+			if (rc) {
+				netdev_warn(netdev,
+					    "SCRQ irq initialization failed\n");
+				return rc;
+			}
+		/* Default/timeout error handling, reset and start fresh */
+		} else if (adapter->init_done_rc) {
+			netdev_warn(netdev, "Adapter login failed, init_done_rc = %d\n",
+				    adapter->init_done_rc);
+
+partial_reset:
+			/* adapter login failed, so free any CRQs or sub-CRQs
+			 * and register again before attempting to login again.
+			 * If we don't do this then the VIOS may think that
+			 * we are already logged in and reject any subsequent
+			 * attempts
+			 */
+			netdev_warn(netdev,
+				    "Freeing and re-registering CRQs before attempting to login again\n");
+			retry = true;
+			adapter->init_done_rc = 0;
+			release_sub_crqs(adapter, true);
+			/* Much of this is similar logic as ibmvnic_probe(),
+			 * we are essentially re-initializing communication
+			 * with the server. We really should not run any
+			 * resets/failovers here because this is already a form
+			 * of reset and we do not want parallel resets occurring
+			 */
+			do {
+				reinit_init_done(adapter);
+				/* Clear any failovers we got in the previous
+				 * pass since we are re-initializing the CRQ
+				 */
+				adapter->failover_pending = false;
+				release_crq_queue(adapter);
+				/* If we don't sleep here then we risk an
+				 * unnecessary failover event from the VIOS.
+				 * This is a known VIOS issue caused by a vnic
+				 * device freeing and registering a CRQ too
+				 * quickly.
+				 */
+				msleep(1500);
+				/* Avoid any resets, since we are currently
+				 * resetting.
+				 */
+				spin_lock_irqsave(&adapter->rwi_lock, flags);
+				flush_reset_queue(adapter);
+				spin_unlock_irqrestore(&adapter->rwi_lock,
+						       flags);
+
+				rc = init_crq_queue(adapter);
+				if (rc) {
+					netdev_err(netdev, "login recovery: init CRQ failed %d\n",
+						   rc);
+					return -EIO;
+				}
+
+				rc = ibmvnic_reset_init(adapter, false);
+				if (rc)
+					netdev_err(netdev, "login recovery: Reset init failed %d\n",
+						   rc);
+				/* IBMVNIC_CRQ_INIT will return EAGAIN if it
+				 * fails, since ibmvnic_reset_init will free
+				 * irq's in failure, we won't be able to receive
+				 * new CRQs so we need to keep trying. probe()
+				 * handles this similarly.
+				 */
+			} while (rc == -EAGAIN && retry_count++ < retries);
 		}
-	} while (adapter->renegotiate);
+	} while (retry);
+
+	__ibmvnic_set_mac(netdev, adapter->mac_addr);
 
+	netdev_dbg(netdev, "[S:%s] Login succeeded\n", adapter_state_to_string(adapter->state));
 	return 0;
 }
 
-static void release_resources(struct ibmvnic_adapter *adapter)
+static void release_login_buffer(struct ibmvnic_adapter *adapter)
 {
-	int i;
+	if (!adapter->login_buf)
+		return;
 
-	release_tx_pools(adapter);
-	release_rx_pools(adapter);
+	dma_unmap_single(&adapter->vdev->dev, adapter->login_buf_token,
+			 adapter->login_buf_sz, DMA_TO_DEVICE);
+	kfree(adapter->login_buf);
+	adapter->login_buf = NULL;
+}
 
-	release_stats_token(adapter);
-	release_error_buffers(adapter);
+static void release_login_rsp_buffer(struct ibmvnic_adapter *adapter)
+{
+	if (!adapter->login_rsp_buf)
+		return;
 
-	if (adapter->napi) {
-		for (i = 0; i < adapter->req_rx_queues; i++) {
-			if (&adapter->napi[i])
-				netif_napi_del(&adapter->napi[i]);
-		}
-	}
+	dma_unmap_single(&adapter->vdev->dev, adapter->login_rsp_buf_token,
+			 adapter->login_rsp_buf_sz, DMA_FROM_DEVICE);
+	kfree(adapter->login_rsp_buf);
+	adapter->login_rsp_buf = NULL;
+}
+
+static void release_resources(struct ibmvnic_adapter *adapter)
+{
+	release_vpd_data(adapter);
+
+	release_napi(adapter);
+	release_login_buffer(adapter);
+	release_login_rsp_buffer(adapter);
 }
 
 static int set_link_state(struct ibmvnic_adapter *adapter, u8 link_state)
 {
 	struct net_device *netdev = adapter->netdev;
-	unsigned long timeout = msecs_to_jiffies(30000);
+	unsigned long timeout = msecs_to_jiffies(20000);
 	union ibmvnic_crq crq;
 	bool resend;
 	int rc;
 
-	netdev_err(netdev, "setting link state %d\n", link_state);
+	netdev_dbg(netdev, "setting link state %d\n", link_state);
+
 	memset(&crq, 0, sizeof(crq));
 	crq.logical_link_state.first = IBMVNIC_CRQ_CMD;
 	crq.logical_link_state.cmd = LOGICAL_LINK_STATE;
@@ -704,13 +1728,17 @@ static int set_link_state(struct ibmvnic_adapter *adapter, u8 link_state)
 		if (!wait_for_completion_timeout(&adapter->init_done,
 						 timeout)) {
 			netdev_err(netdev, "timeout setting link state\n");
-			return -1;
+			return -ETIMEDOUT;
 		}
 
-		if (adapter->init_done_rc == 1) {
+		if (adapter->init_done_rc == PARTIALSUCCESS) {
 			/* Partuial success, delay and re-send */
 			mdelay(1000);
 			resend = true;
+		} else if (adapter->init_done_rc) {
+			netdev_warn(netdev, "Unable to set link state, rc=%d\n",
+				    adapter->init_done_rc);
+			return adapter->init_done_rc;
 		}
 	} while (resend);
 
@@ -722,6 +1750,9 @@ static int set_real_num_queues(struct net_device *netdev)
 	struct ibmvnic_adapter *adapter = netdev_priv(netdev);
 	int rc;
 
+	netdev_dbg(netdev, "Setting real tx/rx queues (%llx/%llx)\n",
+		   adapter->req_tx_queues, adapter->req_rx_queues);
+
 	rc = netif_set_real_num_tx_queues(netdev, adapter->req_tx_queues);
 	if (rc) {
 		netdev_err(netdev, "failed to set the number of tx queues\n");
@@ -735,37 +1766,115 @@ static int set_real_num_queues(struct net_device *netdev)
 	return rc;
 }
 
-static int init_resources(struct ibmvnic_adapter *adapter)
+static int ibmvnic_get_vpd(struct ibmvnic_adapter *adapter)
 {
-	struct net_device *netdev = adapter->netdev;
-	int i, rc;
+	struct device *dev = &adapter->vdev->dev;
+	union ibmvnic_crq crq;
+	int len = 0;
+	int rc;
 
-	rc = set_real_num_queues(netdev);
-	if (rc)
+	if (adapter->vpd->buff)
+		len = adapter->vpd->len;
+
+	mutex_lock(&adapter->fw_lock);
+	adapter->fw_done_rc = 0;
+	reinit_completion(&adapter->fw_done);
+
+	crq.get_vpd_size.first = IBMVNIC_CRQ_CMD;
+	crq.get_vpd_size.cmd = GET_VPD_SIZE;
+	rc = ibmvnic_send_crq(adapter, &crq);
+	if (rc) {
+		mutex_unlock(&adapter->fw_lock);
 		return rc;
+	}
 
-	rc = init_sub_crq_irqs(adapter);
+	rc = ibmvnic_wait_for_completion(adapter, &adapter->fw_done, 10000);
 	if (rc) {
-		netdev_err(netdev, "failed to initialize sub crq irqs\n");
-		return -1;
+		dev_err(dev, "Could not retrieve VPD size, rc = %d\n", rc);
+		mutex_unlock(&adapter->fw_lock);
+		return rc;
 	}
+	mutex_unlock(&adapter->fw_lock);
 
-	rc = init_stats_token(adapter);
+	if (!adapter->vpd->len)
+		return -ENODATA;
+
+	if (!adapter->vpd->buff)
+		adapter->vpd->buff = kzalloc(adapter->vpd->len, GFP_KERNEL);
+	else if (adapter->vpd->len != len)
+		adapter->vpd->buff =
+			krealloc(adapter->vpd->buff,
+				 adapter->vpd->len, GFP_KERNEL);
+
+	if (!adapter->vpd->buff) {
+		dev_err(dev, "Could allocate VPD buffer\n");
+		return -ENOMEM;
+	}
+
+	adapter->vpd->dma_addr =
+		dma_map_single(dev, adapter->vpd->buff, adapter->vpd->len,
+			       DMA_FROM_DEVICE);
+	if (dma_mapping_error(dev, adapter->vpd->dma_addr)) {
+		dev_err(dev, "Could not map VPD buffer\n");
+		kfree(adapter->vpd->buff);
+		adapter->vpd->buff = NULL;
+		return -ENOMEM;
+	}
+
+	mutex_lock(&adapter->fw_lock);
+	adapter->fw_done_rc = 0;
+	reinit_completion(&adapter->fw_done);
+
+	crq.get_vpd.first = IBMVNIC_CRQ_CMD;
+	crq.get_vpd.cmd = GET_VPD;
+	crq.get_vpd.ioba = cpu_to_be32(adapter->vpd->dma_addr);
+	crq.get_vpd.len = cpu_to_be32((u32)adapter->vpd->len);
+	rc = ibmvnic_send_crq(adapter, &crq);
+	if (rc) {
+		kfree(adapter->vpd->buff);
+		adapter->vpd->buff = NULL;
+		mutex_unlock(&adapter->fw_lock);
+		return rc;
+	}
+
+	rc = ibmvnic_wait_for_completion(adapter, &adapter->fw_done, 10000);
+	if (rc) {
+		dev_err(dev, "Unable to retrieve VPD, rc = %d\n", rc);
+		kfree(adapter->vpd->buff);
+		adapter->vpd->buff = NULL;
+		mutex_unlock(&adapter->fw_lock);
+		return rc;
+	}
+
+	mutex_unlock(&adapter->fw_lock);
+	return 0;
+}
+
+static int init_resources(struct ibmvnic_adapter *adapter)
+{
+	struct net_device *netdev = adapter->netdev;
+	int rc;
+
+	rc = set_real_num_queues(netdev);
 	if (rc)
 		return rc;
 
-	adapter->map_id = 1;
-	adapter->napi = kcalloc(adapter->req_rx_queues,
-				sizeof(struct napi_struct), GFP_KERNEL);
-	if (!adapter->napi)
+	adapter->vpd = kzalloc(sizeof(*adapter->vpd), GFP_KERNEL);
+	if (!adapter->vpd)
 		return -ENOMEM;
 
-	for (i = 0; i < adapter->req_rx_queues; i++) {
-		netif_napi_add(netdev, &adapter->napi[i], ibmvnic_poll,
-			       NAPI_POLL_WEIGHT);
+	/* Vital Product Data (VPD) */
+	rc = ibmvnic_get_vpd(adapter);
+	if (rc) {
+		netdev_err(netdev, "failed to initialize Vital Product Data (VPD)\n");
+		return rc;
 	}
 
-	send_map_query(adapter);
+	rc = init_napi(adapter);
+	if (rc)
+		return rc;
+
+	send_query_map(adapter);
 
 	rc = init_rx_pools(netdev);
 	if (rc)
@@ -789,27 +1898,43 @@ static int __ibmvnic_open(struct net_device *netdev)
 	 * set the logical link state to up
 	 */
 	for (i = 0; i < adapter->req_rx_queues; i++) {
+		netdev_dbg(netdev, "Enabling rx_scrq[%d] irq\n", i);
 		if (prev_state == VNIC_CLOSED)
 			enable_irq(adapter->rx_scrq[i]->irq);
-		else
-			enable_scrq_irq(adapter, adapter->rx_scrq[i]);
+		enable_scrq_irq(adapter, adapter->rx_scrq[i]);
 	}
 
 	for (i = 0; i < adapter->req_tx_queues; i++) {
+		netdev_dbg(netdev, "Enabling tx_scrq[%d] irq\n", i);
 		if (prev_state == VNIC_CLOSED)
 			enable_irq(adapter->tx_scrq[i]->irq);
-		else
-			enable_scrq_irq(adapter, adapter->tx_scrq[i]);
+		enable_scrq_irq(adapter, adapter->tx_scrq[i]);
+		/* netdev_tx_reset_queue will reset dql stats. During NON_FATAL
+		 * resets, don't reset the stats because there could be batched
+		 * skb's waiting to be sent. If we reset dql stats, we risk
+		 * num_completed being greater than num_queued. This will cause
+		 * a BUG_ON in dql_completed().
+		 */
+		if (adapter->reset_reason != VNIC_RESET_NON_FATAL)
+			netdev_tx_reset_queue(netdev_get_tx_queue(netdev, i));
 	}
 
 	rc = set_link_state(adapter, IBMVNIC_LOGICAL_LNK_UP);
 	if (rc) {
-		for (i = 0; i < adapter->req_rx_queues; i++)
-			napi_disable(&adapter->napi[i]);
-		release_resources(adapter);
+		ibmvnic_napi_disable(adapter);
+		ibmvnic_disable_irqs(adapter);
 		return rc;
 	}
 
+	adapter->tx_queues_active = true;
+
+	/* Since queues were stopped until now, there shouldn't be any
+	 * one in ibmvnic_complete_tx() or ibmvnic_xmit() so maybe we
+	 * don't need the synchronize_rcu()? Leaving it for consistency
+	 * with setting ->tx_queues_active = false.
+	 */
+	synchronize_rcu();
+
 	netif_tx_start_all_queues(netdev);
 
 	if (prev_state == VNIC_CLOSED) {
@@ -826,96 +1951,188 @@ static int ibmvnic_open(struct net_device *netdev)
 	struct ibmvnic_adapter *adapter = netdev_priv(netdev);
 	int rc;
 
-	mutex_lock(&adapter->reset_lock);
+	ASSERT_RTNL();
+
+	/* If device failover is pending or we are about to reset, just set
+	 * device state and return. Device operation will be handled by reset
+	 * routine.
+	 *
+	 * It should be safe to overwrite the adapter->state here. Since
+	 * we hold the rtnl, either the reset has not actually started or
+	 * the rtnl got dropped during the set_link_state() in do_reset().
+	 * In the former case, no one else is changing the state (again we
+	 * have the rtnl) and in the latter case, do_reset() will detect and
+	 * honor our setting below.
+	 */
+	if (adapter->failover_pending || (test_bit(0, &adapter->resetting))) {
+		netdev_dbg(netdev, "[S:%s FOP:%d] Resetting, deferring open\n",
+			   adapter_state_to_string(adapter->state),
+			   adapter->failover_pending);
+		adapter->state = VNIC_OPEN;
+		rc = 0;
+		goto out;
+	}
 
 	if (adapter->state != VNIC_CLOSED) {
 		rc = ibmvnic_login(netdev);
-		if (rc) {
-			mutex_unlock(&adapter->reset_lock);
-			return rc;
-		}
+		if (rc)
+			goto out;
 
 		rc = init_resources(adapter);
 		if (rc) {
 			netdev_err(netdev, "failed to initialize resources\n");
-			release_resources(adapter);
-			mutex_unlock(&adapter->reset_lock);
-			return rc;
+			goto out;
 		}
 	}
 
 	rc = __ibmvnic_open(netdev);
-	mutex_unlock(&adapter->reset_lock);
+
+out:
+	/* If open failed and there is a pending failover or in-progress reset,
+	 * set device state and return. Device operation will be handled by
+	 * reset routine. See also comments above regarding rtnl.
+	 */
+	if (rc &&
+	    (adapter->failover_pending || (test_bit(0, &adapter->resetting)))) {
+		adapter->state = VNIC_OPEN;
+		rc = 0;
+	}
+
+	if (rc) {
+		release_resources(adapter);
+		release_rx_pools(adapter);
+		release_tx_pools(adapter);
+	}
 
 	return rc;
 }
 
-static void clean_tx_pools(struct ibmvnic_adapter *adapter)
+static void clean_rx_pools(struct ibmvnic_adapter *adapter)
 {
-	struct ibmvnic_tx_pool *tx_pool;
-	u64 tx_entries;
-	int tx_scrqs;
+	struct ibmvnic_rx_pool *rx_pool;
+	struct ibmvnic_rx_buff *rx_buff;
+	u64 rx_entries;
+	int rx_scrqs;
 	int i, j;
 
-	if (!adapter->tx_pool)
+	if (!adapter->rx_pool)
 		return;
 
-	tx_scrqs = be32_to_cpu(adapter->login_rsp_buf->num_txsubm_subcrqs);
-	tx_entries = adapter->req_tx_entries_per_subcrq;
+	rx_scrqs = adapter->num_active_rx_pools;
+	rx_entries = adapter->req_rx_add_entries_per_subcrq;
 
-	/* Free any remaining skbs in the tx buffer pools */
-	for (i = 0; i < tx_scrqs; i++) {
-		tx_pool = &adapter->tx_pool[i];
-		if (!tx_pool)
+	/* Free any remaining skbs in the rx buffer pools */
+	for (i = 0; i < rx_scrqs; i++) {
+		rx_pool = &adapter->rx_pool[i];
+		if (!rx_pool || !rx_pool->rx_buff)
 			continue;
 
-		for (j = 0; j < tx_entries; j++) {
-			if (tx_pool->tx_buff[j].skb) {
-				dev_kfree_skb_any(tx_pool->tx_buff[j].skb);
-				tx_pool->tx_buff[j].skb = NULL;
+		netdev_dbg(adapter->netdev, "Cleaning rx_pool[%d]\n", i);
+		for (j = 0; j < rx_entries; j++) {
+			rx_buff = &rx_pool->rx_buff[j];
+			if (rx_buff && rx_buff->skb) {
+				dev_kfree_skb_any(rx_buff->skb);
+				rx_buff->skb = NULL;
 			}
 		}
 	}
 }
 
-static int __ibmvnic_close(struct net_device *netdev)
+static void clean_one_tx_pool(struct ibmvnic_adapter *adapter,
+			      struct ibmvnic_tx_pool *tx_pool)
 {
-	struct ibmvnic_adapter *adapter = netdev_priv(netdev);
-	int rc = 0;
+	struct ibmvnic_tx_buff *tx_buff;
+	u64 tx_entries;
 	int i;
 
-	adapter->state = VNIC_CLOSING;
-	netif_tx_stop_all_queues(netdev);
-	ibmvnic_napi_disable(adapter);
+	if (!tx_pool || !tx_pool->tx_buff)
+		return;
 
-	if (adapter->tx_scrq) {
-		for (i = 0; i < adapter->req_tx_queues; i++)
-			if (adapter->tx_scrq[i]->irq)
-				disable_irq(adapter->tx_scrq[i]->irq);
+	tx_entries = tx_pool->num_buffers;
+
+	for (i = 0; i < tx_entries; i++) {
+		tx_buff = &tx_pool->tx_buff[i];
+		if (tx_buff && tx_buff->skb) {
+			dev_kfree_skb_any(tx_buff->skb);
+			tx_buff->skb = NULL;
+		}
 	}
+}
 
-	rc = set_link_state(adapter, IBMVNIC_LOGICAL_LNK_DN);
-	if (rc)
-		return rc;
+static void clean_tx_pools(struct ibmvnic_adapter *adapter)
+{
+	int tx_scrqs;
+	int i;
 
-	if (adapter->rx_scrq) {
-		for (i = 0; i < adapter->req_rx_queues; i++) {
-			int retries = 10;
+	if (!adapter->tx_pool || !adapter->tso_pool)
+		return;
 
-			while (pending_scrq(adapter, adapter->rx_scrq[i])) {
-				retries--;
-				mdelay(100);
+	tx_scrqs = adapter->num_active_tx_pools;
 
-				if (retries == 0)
-					break;
+	/* Free any remaining skbs in the tx buffer pools */
+	for (i = 0; i < tx_scrqs; i++) {
+		netdev_dbg(adapter->netdev, "Cleaning tx_pool[%d]\n", i);
+		clean_one_tx_pool(adapter, &adapter->tx_pool[i]);
+		clean_one_tx_pool(adapter, &adapter->tso_pool[i]);
+	}
+}
+
+static void ibmvnic_disable_irqs(struct ibmvnic_adapter *adapter)
+{
+	struct net_device *netdev = adapter->netdev;
+	int i;
+
+	if (adapter->tx_scrq) {
+		for (i = 0; i < adapter->req_tx_queues; i++)
+			if (adapter->tx_scrq[i]->irq) {
+				netdev_dbg(netdev,
+					   "Disabling tx_scrq[%d] irq\n", i);
+				disable_scrq_irq(adapter, adapter->tx_scrq[i]);
+				disable_irq(adapter->tx_scrq[i]->irq);
 			}
+	}
 
-			if (adapter->rx_scrq[i]->irq)
+	if (adapter->rx_scrq) {
+		for (i = 0; i < adapter->req_rx_queues; i++) {
+			if (adapter->rx_scrq[i]->irq) {
+				netdev_dbg(netdev,
+					   "Disabling rx_scrq[%d] irq\n", i);
+				disable_scrq_irq(adapter, adapter->rx_scrq[i]);
 				disable_irq(adapter->rx_scrq[i]->irq);
+			}
 		}
 	}
+}
 
-	clean_tx_pools(adapter);
+static void ibmvnic_cleanup(struct net_device *netdev)
+{
+	struct ibmvnic_adapter *adapter = netdev_priv(netdev);
+
+	/* ensure that transmissions are stopped if called by do_reset */
+
+	adapter->tx_queues_active = false;
+
+	/* Ensure complete_tx() and ibmvnic_xmit() see ->tx_queues_active
+	 * update so they don't restart a queue after we stop it below.
+	 */
+	synchronize_rcu();
+
+	if (test_bit(0, &adapter->resetting))
+		netif_tx_disable(netdev);
+	else
+		netif_tx_stop_all_queues(netdev);
+
+	ibmvnic_napi_disable(adapter);
+	ibmvnic_disable_irqs(adapter);
+}
+
+static int __ibmvnic_close(struct net_device *netdev)
+{
+	struct ibmvnic_adapter *adapter = netdev_priv(netdev);
+	int rc = 0;
+
+	adapter->state = VNIC_CLOSING;
+	rc = set_link_state(adapter, IBMVNIC_LOGICAL_LNK_DN);
 	adapter->state = VNIC_CLOSED;
 	return rc;
 }
@@ -925,171 +2142,320 @@ static int ibmvnic_close(struct net_device *netdev)
 	struct ibmvnic_adapter *adapter = netdev_priv(netdev);
 	int rc;
 
-	mutex_lock(&adapter->reset_lock);
+	netdev_dbg(netdev, "[S:%s FOP:%d FRR:%d] Closing\n",
+		   adapter_state_to_string(adapter->state),
+		   adapter->failover_pending,
+		   adapter->force_reset_recovery);
+
+	/* If device failover is pending, just set device state and return.
+	 * Device operation will be handled by reset routine.
+	 */
+	if (adapter->failover_pending) {
+		adapter->state = VNIC_CLOSED;
+		return 0;
+	}
+
 	rc = __ibmvnic_close(netdev);
-	mutex_unlock(&adapter->reset_lock);
+	ibmvnic_cleanup(netdev);
+	clean_rx_pools(adapter);
+	clean_tx_pools(adapter);
 
 	return rc;
 }
 
 /**
- * build_hdr_data - creates L2/L3/L4 header data buffer
- * @hdr_field - bitfield determining needed headers
- * @skb - socket buffer
- * @hdr_len - array of header lengths
- * @tot_len - total length of data
+ * get_hdr_lens - fills list of L2/L3/L4 hdr lens
+ * @hdr_field: bitfield determining needed headers
+ * @skb: socket buffer
+ * @hdr_len: array of header lengths to be filled
  *
  * Reads hdr_field to determine which headers are needed by firmware.
  * Builds a buffer containing these headers.  Saves individual header
  * lengths and total buffer length to be used to build descriptors.
+ *
+ * Return: total len of all headers
  */
-static int build_hdr_data(u8 hdr_field, struct sk_buff *skb,
-			  int *hdr_len, u8 *hdr_data)
+static int get_hdr_lens(u8 hdr_field, struct sk_buff *skb,
+			int *hdr_len)
 {
 	int len = 0;
-	u8 *hdr;
 
-	hdr_len[0] = sizeof(struct ethhdr);
+
+	if ((hdr_field >> 6) & 1) {
+		hdr_len[0] = skb_mac_header_len(skb);
+		len += hdr_len[0];
+	}
+
+	if ((hdr_field >> 5) & 1) {
+		hdr_len[1] = skb_network_header_len(skb);
+		len += hdr_len[1];
+	}
+
+	if (!((hdr_field >> 4) & 1))
+		return len;
 
 	if (skb->protocol == htons(ETH_P_IP)) {
-		hdr_len[1] = ip_hdr(skb)->ihl * 4;
 		if (ip_hdr(skb)->protocol == IPPROTO_TCP)
 			hdr_len[2] = tcp_hdrlen(skb);
 		else if (ip_hdr(skb)->protocol == IPPROTO_UDP)
 			hdr_len[2] = sizeof(struct udphdr);
 	} else if (skb->protocol == htons(ETH_P_IPV6)) {
-		hdr_len[1] = sizeof(struct ipv6hdr);
 		if (ipv6_hdr(skb)->nexthdr == IPPROTO_TCP)
 			hdr_len[2] = tcp_hdrlen(skb);
 		else if (ipv6_hdr(skb)->nexthdr == IPPROTO_UDP)
 			hdr_len[2] = sizeof(struct udphdr);
 	}
 
-	memset(hdr_data, 0, 120);
-	if ((hdr_field >> 6) & 1) {
-		hdr = skb_mac_header(skb);
-		memcpy(hdr_data, hdr, hdr_len[0]);
-		len += hdr_len[0];
-	}
-
-	if ((hdr_field >> 5) & 1) {
-		hdr = skb_network_header(skb);
-		memcpy(hdr_data + len, hdr, hdr_len[1]);
-		len += hdr_len[1];
-	}
-
-	if ((hdr_field >> 4) & 1) {
-		hdr = skb_transport_header(skb);
-		memcpy(hdr_data + len, hdr, hdr_len[2]);
-		len += hdr_len[2];
-	}
-	return len;
+	return len + hdr_len[2];
 }
 
 /**
  * create_hdr_descs - create header and header extension descriptors
- * @hdr_field - bitfield determining needed headers
- * @data - buffer containing header data
- * @len - length of data buffer
- * @hdr_len - array of individual header lengths
- * @scrq_arr - descriptor array
+ * @hdr_field: bitfield determining needed headers
+ * @hdr_data: buffer containing header data
+ * @len: length of data buffer
+ * @hdr_len: array of individual header lengths
+ * @scrq_arr: descriptor array
  *
  * Creates header and, if needed, header extension descriptors and
  * places them in a descriptor array, scrq_arr
+ *
+ * Return: Number of header descs
  */
 
-static void create_hdr_descs(u8 hdr_field, u8 *hdr_data, int len, int *hdr_len,
-			     union sub_crq *scrq_arr)
+static int create_hdr_descs(u8 hdr_field, u8 *hdr_data, int len, int *hdr_len,
+			    union sub_crq *scrq_arr)
 {
-	union sub_crq hdr_desc;
+	union sub_crq *hdr_desc;
 	int tmp_len = len;
+	int num_descs = 0;
 	u8 *data, *cur;
 	int tmp;
 
 	while (tmp_len > 0) {
 		cur = hdr_data + len - tmp_len;
 
-		memset(&hdr_desc, 0, sizeof(hdr_desc));
-		if (cur != hdr_data) {
-			data = hdr_desc.hdr_ext.data;
+		hdr_desc = &scrq_arr[num_descs];
+		if (num_descs) {
+			data = hdr_desc->hdr_ext.data;
 			tmp = tmp_len > 29 ? 29 : tmp_len;
-			hdr_desc.hdr_ext.first = IBMVNIC_CRQ_CMD;
-			hdr_desc.hdr_ext.type = IBMVNIC_HDR_EXT_DESC;
-			hdr_desc.hdr_ext.len = tmp;
+			hdr_desc->hdr_ext.first = IBMVNIC_CRQ_CMD;
+			hdr_desc->hdr_ext.type = IBMVNIC_HDR_EXT_DESC;
+			hdr_desc->hdr_ext.len = tmp;
 		} else {
-			data = hdr_desc.hdr.data;
+			data = hdr_desc->hdr.data;
 			tmp = tmp_len > 24 ? 24 : tmp_len;
-			hdr_desc.hdr.first = IBMVNIC_CRQ_CMD;
-			hdr_desc.hdr.type = IBMVNIC_HDR_DESC;
-			hdr_desc.hdr.len = tmp;
-			hdr_desc.hdr.l2_len = (u8)hdr_len[0];
-			hdr_desc.hdr.l3_len = cpu_to_be16((u16)hdr_len[1]);
-			hdr_desc.hdr.l4_len = (u8)hdr_len[2];
-			hdr_desc.hdr.flag = hdr_field << 1;
+			hdr_desc->hdr.first = IBMVNIC_CRQ_CMD;
+			hdr_desc->hdr.type = IBMVNIC_HDR_DESC;
+			hdr_desc->hdr.len = tmp;
+			hdr_desc->hdr.l2_len = (u8)hdr_len[0];
+			hdr_desc->hdr.l3_len = cpu_to_be16((u16)hdr_len[1]);
+			hdr_desc->hdr.l4_len = (u8)hdr_len[2];
+			hdr_desc->hdr.flag = hdr_field << 1;
 		}
 		memcpy(data, cur, tmp);
 		tmp_len -= tmp;
-		*scrq_arr = hdr_desc;
-		scrq_arr++;
+		num_descs++;
 	}
+
+	return num_descs;
 }
 
 /**
  * build_hdr_descs_arr - build a header descriptor array
- * @skb - socket buffer
- * @num_entries - number of descriptors to be sent
- * @subcrq - first TX descriptor
- * @hdr_field - bit field determining which headers will be sent
+ * @skb: tx socket buffer
+ * @indir_arr: indirect array
+ * @num_entries: number of descriptors to be sent
+ * @hdr_field: bit field determining which headers will be sent
  *
  * This function will build a TX descriptor array with applicable
  * L2/L3/L4 packet header descriptors to be sent by send_subcrq_indirect.
  */
 
-static void build_hdr_descs_arr(struct ibmvnic_tx_buff *txbuff,
+static void build_hdr_descs_arr(struct sk_buff *skb,
+				union sub_crq *indir_arr,
 				int *num_entries, u8 hdr_field)
 {
 	int hdr_len[3] = {0, 0, 0};
-	int tot_len, len;
-	u8 *hdr_data = txbuff->hdr_data;
+	int tot_len;
+
+	tot_len = get_hdr_lens(hdr_field, skb, hdr_len);
+	*num_entries += create_hdr_descs(hdr_field, skb_mac_header(skb),
+					 tot_len, hdr_len, indir_arr + 1);
+}
+
+static int ibmvnic_xmit_workarounds(struct sk_buff *skb,
+				    struct net_device *netdev)
+{
+	/* For some backing devices, mishandling of small packets
+	 * can result in a loss of connection or TX stall. Device
+	 * architects recommend that no packet should be smaller
+	 * than the minimum MTU value provided to the driver, so
+	 * pad any packets to that length
+	 */
+	if (skb->len < netdev->min_mtu)
+		return skb_put_padto(skb, netdev->min_mtu);
+
+	return 0;
+}
+
+static void ibmvnic_tx_scrq_clean_buffer(struct ibmvnic_adapter *adapter,
+					 struct ibmvnic_sub_crq_queue *tx_scrq)
+{
+	struct ibmvnic_ind_xmit_queue *ind_bufp;
+	struct ibmvnic_tx_buff *tx_buff;
+	struct ibmvnic_tx_pool *tx_pool;
+	union sub_crq tx_scrq_entry;
+	int queue_num;
+	int entries;
+	int index;
+	int i;
+
+	ind_bufp = &tx_scrq->ind_buf;
+	entries = (u64)ind_bufp->index;
+	queue_num = tx_scrq->pool_index;
+
+	for (i = entries - 1; i >= 0; --i) {
+		tx_scrq_entry = ind_bufp->indir_arr[i];
+		if (tx_scrq_entry.v1.type != IBMVNIC_TX_DESC)
+			continue;
+		index = be32_to_cpu(tx_scrq_entry.v1.correlator);
+		if (index & IBMVNIC_TSO_POOL_MASK) {
+			tx_pool = &adapter->tso_pool[queue_num];
+			index &= ~IBMVNIC_TSO_POOL_MASK;
+		} else {
+			tx_pool = &adapter->tx_pool[queue_num];
+		}
+		tx_pool->free_map[tx_pool->consumer_index] = index;
+		tx_pool->consumer_index = tx_pool->consumer_index == 0 ?
+					  tx_pool->num_buffers - 1 :
+					  tx_pool->consumer_index - 1;
+		tx_buff = &tx_pool->tx_buff[index];
+		adapter->tx_stats_buffers[queue_num].batched_packets--;
+		adapter->tx_stats_buffers[queue_num].bytes -=
+						tx_buff->skb->len;
+		dev_kfree_skb_any(tx_buff->skb);
+		tx_buff->skb = NULL;
+		adapter->netdev->stats.tx_dropped++;
+	}
+
+	ind_bufp->index = 0;
 
-	tot_len = build_hdr_data(hdr_field, txbuff->skb, hdr_len,
-				 txbuff->hdr_data);
-	len = tot_len;
-	len -= 24;
-	if (len > 0)
-		num_entries += len % 29 ? len / 29 + 1 : len / 29;
-	create_hdr_descs(hdr_field, hdr_data, tot_len, hdr_len,
-			 txbuff->indir_arr + 1);
+	if (atomic_sub_return(entries, &tx_scrq->used) <=
+	    (adapter->req_tx_entries_per_subcrq / 2) &&
+	    __netif_subqueue_stopped(adapter->netdev, queue_num)) {
+		rcu_read_lock();
+
+		if (adapter->tx_queues_active) {
+			netif_wake_subqueue(adapter->netdev, queue_num);
+			netdev_dbg(adapter->netdev, "Started queue %d\n",
+				   queue_num);
+		}
+
+		rcu_read_unlock();
+	}
 }
 
-static int ibmvnic_xmit(struct sk_buff *skb, struct net_device *netdev)
+static int send_subcrq_direct(struct ibmvnic_adapter *adapter,
+			      u64 remote_handle, u64 *entry)
+{
+	unsigned int ua = adapter->vdev->unit_address;
+	struct device *dev = &adapter->vdev->dev;
+	int rc;
+
+	/* Make sure the hypervisor sees the complete request */
+	dma_wmb();
+	rc = plpar_hcall_norets(H_SEND_SUB_CRQ, ua,
+				cpu_to_be64(remote_handle),
+				cpu_to_be64(entry[0]), cpu_to_be64(entry[1]),
+				cpu_to_be64(entry[2]), cpu_to_be64(entry[3]));
+
+	if (rc)
+		print_subcrq_error(dev, rc, __func__);
+
+	return rc;
+}
+
+static int ibmvnic_tx_scrq_flush(struct ibmvnic_adapter *adapter,
+				 struct ibmvnic_sub_crq_queue *tx_scrq,
+				 bool indirect)
+{
+	struct ibmvnic_ind_xmit_queue *ind_bufp;
+	u64 dma_addr;
+	u64 entries;
+	u64 handle;
+	int rc;
+
+	ind_bufp = &tx_scrq->ind_buf;
+	dma_addr = (u64)ind_bufp->indir_dma;
+	entries = (u64)ind_bufp->index;
+	handle = tx_scrq->handle;
+
+	if (!entries)
+		return 0;
+
+	if (indirect)
+		rc = send_subcrq_indirect(adapter, handle, dma_addr, entries);
+	else
+		rc = send_subcrq_direct(adapter, handle,
+					(u64 *)ind_bufp->indir_arr);
+
+	if (rc) {
+		dev_err_ratelimited(&adapter->vdev->dev,
+				    "tx_flush failed, rc=%u (%llu entries dma=%pad handle=%llx)\n",
+				    rc, entries, &dma_addr, handle);
+		/* Detect platform limit H_PARAMETER */
+		if (rc == H_PARAMETER)
+			ibmvnic_set_safe_max_ind_descs(adapter);
+
+		/* For all error case, temporarily drop only this batch
+		 * Rely on TCP/IP retransmissions to retry and recover
+		 */
+		ibmvnic_tx_scrq_clean_buffer(adapter, tx_scrq);
+	} else {
+		ind_bufp->index = 0;
+	}
+	return rc;
+}
+
+static netdev_tx_t ibmvnic_xmit(struct sk_buff *skb, struct net_device *netdev)
 {
 	struct ibmvnic_adapter *adapter = netdev_priv(netdev);
+	u32 cur_max_ind_descs = adapter->cur_max_ind_descs;
 	int queue_num = skb_get_queue_mapping(skb);
 	u8 *hdrs = (u8 *)&adapter->tx_rx_desc_req;
 	struct device *dev = &adapter->vdev->dev;
+	struct ibmvnic_ind_xmit_queue *ind_bufp;
 	struct ibmvnic_tx_buff *tx_buff = NULL;
 	struct ibmvnic_sub_crq_queue *tx_scrq;
+	struct ibmvnic_long_term_buff *ltb;
 	struct ibmvnic_tx_pool *tx_pool;
 	unsigned int tx_send_failed = 0;
+	netdev_tx_t ret = NETDEV_TX_OK;
 	unsigned int tx_map_failed = 0;
+	union sub_crq indir_arr[16];
 	unsigned int tx_dropped = 0;
-	unsigned int tx_packets = 0;
+	unsigned int tx_dpackets = 0;
+	unsigned int tx_bpackets = 0;
 	unsigned int tx_bytes = 0;
 	dma_addr_t data_dma_addr;
 	struct netdev_queue *txq;
 	unsigned long lpar_rc;
+	unsigned int skblen;
 	union sub_crq tx_crq;
 	unsigned int offset;
+	bool use_scrq_send_direct = false;
 	int num_entries = 1;
 	unsigned char *dst;
-	u64 *handle_array;
-	int index = 0;
-	int ret = 0;
+	int bufidx = 0;
+	u8 proto = 0;
 
-	if (adapter->resetting) {
-		if (!netif_subqueue_stopped(netdev, skb))
-			netif_stop_subqueue(netdev, queue_num);
+	/* If a reset is in progress, drop the packet since
+	 * the scrqs may get torn down. Otherwise use the
+	 * rcu to ensure reset waits for us to complete.
+	 */
+	rcu_read_lock();
+	if (!adapter->tx_queues_active) {
 		dev_kfree_skb_any(skb);
 
 		tx_send_failed++;
@@ -1098,30 +2464,99 @@ static int ibmvnic_xmit(struct sk_buff *skb, struct net_device *netdev)
 		goto out;
 	}
 
-	tx_pool = &adapter->tx_pool[queue_num];
 	tx_scrq = adapter->tx_scrq[queue_num];
-	txq = netdev_get_tx_queue(netdev, skb_get_queue_mapping(skb));
-	handle_array = (u64 *)((u8 *)(adapter->login_rsp_buf) +
-		be32_to_cpu(adapter->login_rsp_buf->off_txsubm_subcrqs));
+	txq = netdev_get_tx_queue(netdev, queue_num);
+	ind_bufp = &tx_scrq->ind_buf;
+
+	if (ibmvnic_xmit_workarounds(skb, netdev)) {
+		tx_dropped++;
+		tx_send_failed++;
+		ret = NETDEV_TX_OK;
+		lpar_rc = ibmvnic_tx_scrq_flush(adapter, tx_scrq, true);
+		if (lpar_rc != H_SUCCESS)
+			goto tx_err;
+		goto out;
+	}
+
+	if (skb_is_gso(skb))
+		tx_pool = &adapter->tso_pool[queue_num];
+	else
+		tx_pool = &adapter->tx_pool[queue_num];
+
+	bufidx = tx_pool->free_map[tx_pool->consumer_index];
+
+	if (bufidx == IBMVNIC_INVALID_MAP) {
+		dev_kfree_skb_any(skb);
+		tx_send_failed++;
+		tx_dropped++;
+		ret = NETDEV_TX_OK;
+		lpar_rc = ibmvnic_tx_scrq_flush(adapter, tx_scrq, true);
+		if (lpar_rc != H_SUCCESS)
+			goto tx_err;
+		goto out;
+	}
+
+	tx_pool->free_map[tx_pool->consumer_index] = IBMVNIC_INVALID_MAP;
+
+	map_txpool_buf_to_ltb(tx_pool, bufidx, &ltb, &offset);
+
+	dst = ltb->buff + offset;
+	memset(dst, 0, tx_pool->buf_size);
+	data_dma_addr = ltb->addr + offset;
+
+	/* if we are going to send_subcrq_direct this then we need to
+	 * update the checksum before copying the data into ltb. Essentially
+	 * these packets force disable CSO so that we can guarantee that
+	 * FW does not need header info and we can send direct. Also, vnic
+	 * server must be able to xmit standard packets without header data
+	 */
+	if (*hdrs == 0 && !skb_is_gso(skb) &&
+	    !ind_bufp->index && !netdev_xmit_more()) {
+		use_scrq_send_direct = true;
+		if (skb->ip_summed == CHECKSUM_PARTIAL &&
+		    skb_checksum_help(skb))
+			use_scrq_send_direct = false;
+	}
+
+	if (skb_shinfo(skb)->nr_frags) {
+		int cur, i;
+
+		/* Copy the head */
+		skb_copy_from_linear_data(skb, dst, skb_headlen(skb));
+		cur = skb_headlen(skb);
+
+		/* Copy the frags */
+		for (i = 0; i < skb_shinfo(skb)->nr_frags; i++) {
+			const skb_frag_t *frag = &skb_shinfo(skb)->frags[i];
 
-	index = tx_pool->free_map[tx_pool->consumer_index];
-	offset = index * adapter->req_mtu;
-	dst = tx_pool->long_term_buff.buff + offset;
-	memset(dst, 0, adapter->req_mtu);
-	skb_copy_from_linear_data(skb, dst, skb->len);
-	data_dma_addr = tx_pool->long_term_buff.addr + offset;
+			memcpy(dst + cur, skb_frag_address(frag),
+			       skb_frag_size(frag));
+			cur += skb_frag_size(frag);
+		}
+	} else {
+		skb_copy_from_linear_data(skb, dst, skb->len);
+	}
 
 	tx_pool->consumer_index =
-	    (tx_pool->consumer_index + 1) %
-		adapter->req_tx_entries_per_subcrq;
+	    (tx_pool->consumer_index + 1) % tx_pool->num_buffers;
+
+	tx_buff = &tx_pool->tx_buff[bufidx];
+
+	/* Sanity checks on our free map to make sure it points to an index
+	 * that is not being occupied by another skb. If skb memory is
+	 * not freed then we see congestion control kick in and halt tx.
+	 */
+	if (unlikely(tx_buff->skb)) {
+		dev_warn_ratelimited(dev, "TX free map points to untracked skb (%s %d idx=%d)\n",
+				     skb_is_gso(skb) ? "tso_pool" : "tx_pool",
+				     queue_num, bufidx);
+		dev_kfree_skb_any(tx_buff->skb);
+	}
 
-	tx_buff = &tx_pool->tx_buff[index];
 	tx_buff->skb = skb;
-	tx_buff->data_dma[0] = data_dma_addr;
-	tx_buff->data_len[0] = skb->len;
-	tx_buff->index = index;
+	tx_buff->index = bufidx;
 	tx_buff->pool_index = queue_num;
-	tx_buff->last_frag = true;
+	skblen = skb->len;
 
 	memset(&tx_crq, 0, sizeof(tx_crq));
 	tx_crq.v1.first = IBMVNIC_CRQ_CMD;
@@ -1129,103 +2564,124 @@ static int ibmvnic_xmit(struct sk_buff *skb, struct net_device *netdev)
 	tx_crq.v1.n_crq_elem = 1;
 	tx_crq.v1.n_sge = 1;
 	tx_crq.v1.flags1 = IBMVNIC_TX_COMP_NEEDED;
-	tx_crq.v1.correlator = cpu_to_be32(index);
-	tx_crq.v1.dma_reg = cpu_to_be16(tx_pool->long_term_buff.map_id);
+
+	if (skb_is_gso(skb))
+		tx_crq.v1.correlator =
+			cpu_to_be32(bufidx | IBMVNIC_TSO_POOL_MASK);
+	else
+		tx_crq.v1.correlator = cpu_to_be32(bufidx);
+	tx_crq.v1.dma_reg = cpu_to_be16(ltb->map_id);
 	tx_crq.v1.sge_len = cpu_to_be32(skb->len);
 	tx_crq.v1.ioba = cpu_to_be64(data_dma_addr);
 
-	if (adapter->vlan_header_insertion) {
+	if (adapter->vlan_header_insertion && skb_vlan_tag_present(skb)) {
 		tx_crq.v1.flags2 |= IBMVNIC_TX_VLAN_INSERT;
 		tx_crq.v1.vlan_id = cpu_to_be16(skb->vlan_tci);
 	}
 
 	if (skb->protocol == htons(ETH_P_IP)) {
-		if (ip_hdr(skb)->version == 4)
-			tx_crq.v1.flags1 |= IBMVNIC_TX_PROT_IPV4;
-		else if (ip_hdr(skb)->version == 6)
-			tx_crq.v1.flags1 |= IBMVNIC_TX_PROT_IPV6;
-
-		if (ip_hdr(skb)->protocol == IPPROTO_TCP)
-			tx_crq.v1.flags1 |= IBMVNIC_TX_PROT_TCP;
-		else if (ip_hdr(skb)->protocol != IPPROTO_TCP)
-			tx_crq.v1.flags1 |= IBMVNIC_TX_PROT_UDP;
+		tx_crq.v1.flags1 |= IBMVNIC_TX_PROT_IPV4;
+		proto = ip_hdr(skb)->protocol;
+	} else if (skb->protocol == htons(ETH_P_IPV6)) {
+		tx_crq.v1.flags1 |= IBMVNIC_TX_PROT_IPV6;
+		proto = ipv6_hdr(skb)->nexthdr;
 	}
 
+	if (proto == IPPROTO_TCP)
+		tx_crq.v1.flags1 |= IBMVNIC_TX_PROT_TCP;
+	else if (proto == IPPROTO_UDP)
+		tx_crq.v1.flags1 |= IBMVNIC_TX_PROT_UDP;
+
 	if (skb->ip_summed == CHECKSUM_PARTIAL) {
 		tx_crq.v1.flags1 |= IBMVNIC_TX_CHKSUM_OFFLOAD;
 		hdrs += 2;
 	}
-	/* determine if l2/3/4 headers are sent to firmware */
-	if ((*hdrs >> 7) & 1 &&
-	    (skb->protocol == htons(ETH_P_IP) ||
-	     skb->protocol == htons(ETH_P_IPV6))) {
-		build_hdr_descs_arr(tx_buff, &num_entries, *hdrs);
-		tx_crq.v1.n_crq_elem = num_entries;
-		tx_buff->indir_arr[0] = tx_crq;
-		tx_buff->indir_dma = dma_map_single(dev, tx_buff->indir_arr,
-						    sizeof(tx_buff->indir_arr),
-						    DMA_TO_DEVICE);
-		if (dma_mapping_error(dev, tx_buff->indir_dma)) {
-			dev_kfree_skb_any(skb);
-			tx_buff->skb = NULL;
-			if (!firmware_has_feature(FW_FEATURE_CMO))
-				dev_err(dev, "tx: unable to map descriptor array\n");
-			tx_map_failed++;
-			tx_dropped++;
-			ret = NETDEV_TX_OK;
-			goto out;
-		}
-		lpar_rc = send_subcrq_indirect(adapter, handle_array[queue_num],
-					       (u64)tx_buff->indir_dma,
-					       (u64)num_entries);
-	} else {
-		lpar_rc = send_subcrq(adapter, handle_array[queue_num],
-				      &tx_crq);
+	if (skb_is_gso(skb)) {
+		tx_crq.v1.flags1 |= IBMVNIC_TX_LSO;
+		tx_crq.v1.mss = cpu_to_be16(skb_shinfo(skb)->gso_size);
+		hdrs += 2;
+	} else if (use_scrq_send_direct) {
+		/* See above comment, CSO disabled with direct xmit */
+		tx_crq.v1.flags1 &= ~(IBMVNIC_TX_CHKSUM_OFFLOAD);
+		ind_bufp->index = 1;
+		tx_buff->num_entries = 1;
+		netdev_tx_sent_queue(txq, skb->len);
+		ind_bufp->indir_arr[0] = tx_crq;
+		lpar_rc = ibmvnic_tx_scrq_flush(adapter, tx_scrq, false);
+		if (lpar_rc != H_SUCCESS)
+			goto tx_err;
+
+		tx_dpackets++;
+		goto early_exit;
 	}
-	if (lpar_rc != H_SUCCESS) {
-		dev_err(dev, "tx failed with code %ld\n", lpar_rc);
 
-		if (tx_pool->consumer_index == 0)
-			tx_pool->consumer_index =
-				adapter->req_tx_entries_per_subcrq - 1;
-		else
-			tx_pool->consumer_index--;
+	if ((*hdrs >> 7) & 1)
+		build_hdr_descs_arr(skb, indir_arr, &num_entries, *hdrs);
 
-		dev_kfree_skb_any(skb);
-		tx_buff->skb = NULL;
+	tx_crq.v1.n_crq_elem = num_entries;
+	tx_buff->num_entries = num_entries;
+	/* flush buffer if current entry can not fit */
+	if (num_entries + ind_bufp->index > cur_max_ind_descs) {
+		lpar_rc = ibmvnic_tx_scrq_flush(adapter, tx_scrq, true);
+		if (lpar_rc != H_SUCCESS)
+			goto tx_flush_err;
+	}
 
-		if (lpar_rc == H_CLOSED) {
-			/* Disable TX and report carrier off if queue is closed.
-			 * Firmware guarantees that a signal will be sent to the
-			 * driver, triggering a reset or some other action.
-			 */
-			netif_tx_stop_all_queues(netdev);
-			netif_carrier_off(netdev);
-		}
+	indir_arr[0] = tx_crq;
+	memcpy(&ind_bufp->indir_arr[ind_bufp->index], &indir_arr[0],
+	       num_entries * sizeof(struct ibmvnic_generic_scrq));
 
-		tx_send_failed++;
-		tx_dropped++;
-		ret = NETDEV_TX_OK;
-		goto out;
+	ind_bufp->index += num_entries;
+	if (__netdev_tx_sent_queue(txq, skb->len,
+				   netdev_xmit_more() &&
+				   ind_bufp->index < cur_max_ind_descs)) {
+		lpar_rc = ibmvnic_tx_scrq_flush(adapter, tx_scrq, true);
+		if (lpar_rc != H_SUCCESS)
+			goto tx_err;
 	}
 
-	if (atomic_inc_return(&tx_scrq->used)
+	tx_bpackets++;
+
+early_exit:
+	if (atomic_add_return(num_entries, &tx_scrq->used)
 					>= adapter->req_tx_entries_per_subcrq) {
-		netdev_info(netdev, "Stopping queue %d\n", queue_num);
+		netdev_dbg(netdev, "Stopping queue %d\n", queue_num);
 		netif_stop_subqueue(netdev, queue_num);
 	}
 
-	tx_packets++;
-	tx_bytes += skb->len;
-	txq->trans_start = jiffies;
+	tx_bytes += skblen;
+	txq_trans_cond_update(txq);
 	ret = NETDEV_TX_OK;
+	goto out;
 
+tx_flush_err:
+	dev_kfree_skb_any(skb);
+	tx_buff->skb = NULL;
+	tx_pool->consumer_index = tx_pool->consumer_index == 0 ?
+				  tx_pool->num_buffers - 1 :
+				  tx_pool->consumer_index - 1;
+	tx_dropped++;
+tx_err:
+	if (lpar_rc != H_CLOSED && lpar_rc != H_PARAMETER)
+		dev_err_ratelimited(dev, "tx: send failed\n");
+
+	if (lpar_rc == H_CLOSED || adapter->failover_pending) {
+		/* Disable TX and report carrier off if queue is closed
+		 * or pending failover.
+		 * Firmware guarantees that a signal will be sent to the
+		 * driver, triggering a reset or some other action.
+		 */
+		netif_tx_stop_all_queues(netdev);
+		netif_carrier_off(netdev);
+	}
 out:
-	netdev->stats.tx_dropped += tx_dropped;
-	netdev->stats.tx_bytes += tx_bytes;
-	netdev->stats.tx_packets += tx_packets;
+	rcu_read_unlock();
 	adapter->tx_send_failed += tx_send_failed;
 	adapter->tx_map_failed += tx_map_failed;
+	adapter->tx_stats_buffers[queue_num].batched_packets += tx_bpackets;
+	adapter->tx_stats_buffers[queue_num].direct_packets += tx_dpackets;
+	adapter->tx_stats_buffers[queue_num].bytes += tx_bytes;
+	adapter->tx_stats_buffers[queue_num].dropped_packets += tx_dropped;
 
 	return ret;
 }
@@ -1273,25 +2729,99 @@ static void ibmvnic_set_multi(struct net_device *netdev)
 	}
 }
 
-static int ibmvnic_set_mac(struct net_device *netdev, void *p)
+static int __ibmvnic_set_mac(struct net_device *netdev, u8 *dev_addr)
 {
 	struct ibmvnic_adapter *adapter = netdev_priv(netdev);
-	struct sockaddr *addr = p;
 	union ibmvnic_crq crq;
+	int rc;
 
-	if (!is_valid_ether_addr(addr->sa_data))
-		return -EADDRNOTAVAIL;
+	if (!is_valid_ether_addr(dev_addr)) {
+		rc = -EADDRNOTAVAIL;
+		goto err;
+	}
 
 	memset(&crq, 0, sizeof(crq));
 	crq.change_mac_addr.first = IBMVNIC_CRQ_CMD;
 	crq.change_mac_addr.cmd = CHANGE_MAC_ADDR;
-	ether_addr_copy(&crq.change_mac_addr.mac_addr[0], addr->sa_data);
-	ibmvnic_send_crq(adapter, &crq);
+	ether_addr_copy(&crq.change_mac_addr.mac_addr[0], dev_addr);
+
+	mutex_lock(&adapter->fw_lock);
+	adapter->fw_done_rc = 0;
+	reinit_completion(&adapter->fw_done);
+
+	rc = ibmvnic_send_crq(adapter, &crq);
+	if (rc) {
+		rc = -EIO;
+		mutex_unlock(&adapter->fw_lock);
+		goto err;
+	}
+
+	rc = ibmvnic_wait_for_completion(adapter, &adapter->fw_done, 10000);
 	/* netdev->dev_addr is changed in handle_change_mac_rsp function */
+	if (rc || adapter->fw_done_rc) {
+		rc = -EIO;
+		mutex_unlock(&adapter->fw_lock);
+		goto err;
+	}
+	mutex_unlock(&adapter->fw_lock);
 	return 0;
+err:
+	ether_addr_copy(adapter->mac_addr, netdev->dev_addr);
+	return rc;
 }
 
-/**
+static int ibmvnic_set_mac(struct net_device *netdev, void *p)
+{
+	struct ibmvnic_adapter *adapter = netdev_priv(netdev);
+	struct sockaddr *addr = p;
+	int rc;
+
+	rc = 0;
+	if (!is_valid_ether_addr(addr->sa_data))
+		return -EADDRNOTAVAIL;
+
+	ether_addr_copy(adapter->mac_addr, addr->sa_data);
+	if (adapter->state != VNIC_PROBED)
+		rc = __ibmvnic_set_mac(netdev, addr->sa_data);
+
+	return rc;
+}
+
+static const char *reset_reason_to_string(enum ibmvnic_reset_reason reason)
+{
+	switch (reason) {
+	case VNIC_RESET_FAILOVER:
+		return "FAILOVER";
+	case VNIC_RESET_MOBILITY:
+		return "MOBILITY";
+	case VNIC_RESET_FATAL:
+		return "FATAL";
+	case VNIC_RESET_NON_FATAL:
+		return "NON_FATAL";
+	case VNIC_RESET_TIMEOUT:
+		return "TIMEOUT";
+	case VNIC_RESET_CHANGE_PARAM:
+		return "CHANGE_PARAM";
+	case VNIC_RESET_PASSIVE_INIT:
+		return "PASSIVE_INIT";
+	}
+	return "UNKNOWN";
+}
+
+/*
+ * Initialize the init_done completion and return code values. We
+ * can get a transport event just after registering the CRQ and the
+ * tasklet will use this to communicate the transport event. To ensure
+ * we don't miss the notification/error, initialize these _before_
+ * regisering the CRQ.
+ */
+static inline void reinit_init_done(struct ibmvnic_adapter *adapter)
+{
+	reinit_completion(&adapter->init_done);
+	adapter->init_done_rc = 0;
+}
+
+/*
  * do_reset returns zero if we are able to keep processing reset events, or
  * non-zero if we hit a fatal error and must halt.
  */
@@ -1299,20 +2829,97 @@ static int do_reset(struct ibmvnic_adapter *adapter,
 		    struct ibmvnic_rwi *rwi, u32 reset_state)
 {
 	struct net_device *netdev = adapter->netdev;
-	int i, rc;
+	u64 old_num_rx_queues, old_num_tx_queues;
+	u64 old_num_rx_slots, old_num_tx_slots;
+	int rc;
+
+	netdev_dbg(adapter->netdev,
+		   "[S:%s FOP:%d] Reset reason: %s, reset_state: %s\n",
+		   adapter_state_to_string(adapter->state),
+		   adapter->failover_pending,
+		   reset_reason_to_string(rwi->reset_reason),
+		   adapter_state_to_string(reset_state));
 
-	netif_carrier_off(netdev);
 	adapter->reset_reason = rwi->reset_reason;
+	/* requestor of VNIC_RESET_CHANGE_PARAM already has the rtnl lock */
+	if (!(adapter->reset_reason == VNIC_RESET_CHANGE_PARAM))
+		rtnl_lock();
 
-	if (rwi->reset_reason == VNIC_RESET_MOBILITY) {
-		rc = ibmvnic_reenable_crq_queue(adapter);
-		if (rc)
-			return 0;
+	/* Now that we have the rtnl lock, clear any pending failover.
+	 * This will ensure ibmvnic_open() has either completed or will
+	 * block until failover is complete.
+	 */
+	if (rwi->reset_reason == VNIC_RESET_FAILOVER)
+		adapter->failover_pending = false;
+
+	/* read the state and check (again) after getting rtnl */
+	reset_state = adapter->state;
+
+	if (reset_state == VNIC_REMOVING || reset_state == VNIC_REMOVED) {
+		rc = -EBUSY;
+		goto out;
 	}
 
-	rc = __ibmvnic_close(netdev);
-	if (rc)
-		return rc;
+	netif_carrier_off(netdev);
+
+	old_num_rx_queues = adapter->req_rx_queues;
+	old_num_tx_queues = adapter->req_tx_queues;
+	old_num_rx_slots = adapter->req_rx_add_entries_per_subcrq;
+	old_num_tx_slots = adapter->req_tx_entries_per_subcrq;
+
+	ibmvnic_cleanup(netdev);
+
+	if (reset_state == VNIC_OPEN &&
+	    adapter->reset_reason != VNIC_RESET_MOBILITY &&
+	    adapter->reset_reason != VNIC_RESET_FAILOVER) {
+		if (adapter->reset_reason == VNIC_RESET_CHANGE_PARAM) {
+			rc = __ibmvnic_close(netdev);
+			if (rc)
+				goto out;
+		} else {
+			adapter->state = VNIC_CLOSING;
+
+			/* Release the RTNL lock before link state change and
+			 * re-acquire after the link state change to allow
+			 * linkwatch_event to grab the RTNL lock and run during
+			 * a reset.
+			 */
+			rtnl_unlock();
+			rc = set_link_state(adapter, IBMVNIC_LOGICAL_LNK_DN);
+			rtnl_lock();
+			if (rc)
+				goto out;
+
+			if (adapter->state == VNIC_OPEN) {
+				/* When we dropped rtnl, ibmvnic_open() got
+				 * it and noticed that we are resetting and
+				 * set the adapter state to OPEN. Update our
+				 * new "target" state, and resume the reset
+				 * from VNIC_CLOSING state.
+				 */
+				netdev_dbg(netdev,
+					   "Open changed state from %s, updating.\n",
+					   adapter_state_to_string(reset_state));
+				reset_state = VNIC_OPEN;
+				adapter->state = VNIC_CLOSING;
+			}
+
+			if (adapter->state != VNIC_CLOSING) {
+				/* If someone else changed the adapter state
+				 * when we dropped the rtnl, fail the reset
+				 */
+				rc = -EAGAIN;
+				goto out;
+			}
+			adapter->state = VNIC_CLOSED;
+		}
+	}
+
+	if (adapter->reset_reason == VNIC_RESET_CHANGE_PARAM) {
+		release_resources(adapter);
+		release_sub_crqs(adapter, 1);
+		release_crq_queue(adapter);
+	}
 
 	if (adapter->reset_reason != VNIC_RESET_NON_FATAL) {
 		/* remove the closed state so when we call open it appears
@@ -1320,59 +2927,207 @@ static int do_reset(struct ibmvnic_adapter *adapter,
 		 */
 		adapter->state = VNIC_PROBED;
 
-		rc = ibmvnic_init(adapter);
-		if (rc)
-			return 0;
+		reinit_init_done(adapter);
 
-		/* If the adapter was in PROBE state prior to the reset,
-		 * exit here.
-		 */
-		if (reset_state == VNIC_PROBED)
-			return 0;
+		if (adapter->reset_reason == VNIC_RESET_CHANGE_PARAM) {
+			rc = init_crq_queue(adapter);
+		} else if (adapter->reset_reason == VNIC_RESET_MOBILITY) {
+			rc = ibmvnic_reenable_crq_queue(adapter);
+			release_sub_crqs(adapter, 1);
+		} else {
+			rc = ibmvnic_reset_crq(adapter);
+			if (rc == H_CLOSED || rc == H_SUCCESS) {
+				rc = vio_enable_interrupts(adapter->vdev);
+				if (rc)
+					netdev_err(adapter->netdev,
+						   "Reset failed to enable interrupts. rc=%d\n",
+						   rc);
+			}
+		}
 
-		rc = ibmvnic_login(netdev);
 		if (rc) {
-			adapter->state = VNIC_PROBED;
-			return 0;
+			netdev_err(adapter->netdev,
+				   "Reset couldn't initialize crq. rc=%d\n", rc);
+			goto out;
 		}
 
-		rc = reset_tx_pools(adapter);
+		rc = ibmvnic_reset_init(adapter, true);
 		if (rc)
-			return rc;
+			goto out;
+
+		/* If the adapter was in PROBE or DOWN state prior to the reset,
+		 * exit here.
+		 */
+		if (reset_state == VNIC_PROBED || reset_state == VNIC_DOWN) {
+			rc = 0;
+			goto out;
+		}
 
-		rc = reset_rx_pools(adapter);
+		rc = ibmvnic_login(netdev);
 		if (rc)
-			return rc;
+			goto out;
 
-		if (reset_state == VNIC_CLOSED)
-			return 0;
+		if (adapter->reset_reason == VNIC_RESET_CHANGE_PARAM) {
+			rc = init_resources(adapter);
+			if (rc)
+				goto out;
+		} else if (adapter->req_rx_queues != old_num_rx_queues ||
+		    adapter->req_tx_queues != old_num_tx_queues ||
+		    adapter->req_rx_add_entries_per_subcrq !=
+		    old_num_rx_slots ||
+		    adapter->req_tx_entries_per_subcrq !=
+		    old_num_tx_slots ||
+		    !adapter->rx_pool ||
+		    !adapter->tso_pool ||
+		    !adapter->tx_pool) {
+			release_napi(adapter);
+			release_vpd_data(adapter);
+
+			rc = init_resources(adapter);
+			if (rc)
+				goto out;
+
+		} else {
+			rc = init_tx_pools(netdev);
+			if (rc) {
+				netdev_dbg(netdev,
+					   "init tx pools failed (%d)\n",
+					   rc);
+				goto out;
+			}
+
+			rc = init_rx_pools(netdev);
+			if (rc) {
+				netdev_dbg(netdev,
+					   "init rx pools failed (%d)\n",
+					   rc);
+				goto out;
+			}
+		}
+		ibmvnic_disable_irqs(adapter);
+	}
+	adapter->state = VNIC_CLOSED;
+
+	if (reset_state == VNIC_CLOSED) {
+		rc = 0;
+		goto out;
 	}
 
 	rc = __ibmvnic_open(netdev);
 	if (rc) {
-		if (list_empty(&adapter->rwi_list))
-			adapter->state = VNIC_CLOSED;
-		else
-			adapter->state = reset_state;
+		rc = IBMVNIC_OPEN_FAILED;
+		goto out;
+	}
 
-		return 0;
+	/* refresh device's multicast list */
+	ibmvnic_set_multi(netdev);
+
+	if (adapter->reset_reason == VNIC_RESET_FAILOVER ||
+	    adapter->reset_reason == VNIC_RESET_MOBILITY)
+		__netdev_notify_peers(netdev);
+
+	rc = 0;
+
+out:
+	/* restore the adapter state if reset failed */
+	if (rc)
+		adapter->state = reset_state;
+	/* requestor of VNIC_RESET_CHANGE_PARAM should still hold the rtnl lock */
+	if (!(adapter->reset_reason == VNIC_RESET_CHANGE_PARAM))
+		rtnl_unlock();
+
+	netdev_dbg(adapter->netdev, "[S:%s FOP:%d] Reset done, rc %d\n",
+		   adapter_state_to_string(adapter->state),
+		   adapter->failover_pending, rc);
+	return rc;
+}
+
+static int do_hard_reset(struct ibmvnic_adapter *adapter,
+			 struct ibmvnic_rwi *rwi, u32 reset_state)
+{
+	struct net_device *netdev = adapter->netdev;
+	int rc;
+
+	netdev_dbg(adapter->netdev, "Hard resetting driver (%s)\n",
+		   reset_reason_to_string(rwi->reset_reason));
+
+	/* read the state and check (again) after getting rtnl */
+	reset_state = adapter->state;
+
+	if (reset_state == VNIC_REMOVING || reset_state == VNIC_REMOVED) {
+		rc = -EBUSY;
+		goto out;
 	}
 
-	netif_carrier_on(netdev);
+	netif_carrier_off(netdev);
+	adapter->reset_reason = rwi->reset_reason;
 
-	/* kick napi */
-	for (i = 0; i < adapter->req_rx_queues; i++)
-		napi_schedule(&adapter->napi[i]);
+	ibmvnic_cleanup(netdev);
+	release_resources(adapter);
+	release_sub_crqs(adapter, 0);
+	release_crq_queue(adapter);
 
-	netdev_notify_peers(netdev);
-	return 0;
+	/* remove the closed state so when we call open it appears
+	 * we are coming from the probed state.
+	 */
+	adapter->state = VNIC_PROBED;
+
+	reinit_init_done(adapter);
+
+	rc = init_crq_queue(adapter);
+	if (rc) {
+		netdev_err(adapter->netdev,
+			   "Couldn't initialize crq. rc=%d\n", rc);
+		goto out;
+	}
+
+	rc = ibmvnic_reset_init(adapter, false);
+	if (rc)
+		goto out;
+
+	/* If the adapter was in PROBE or DOWN state prior to the reset,
+	 * exit here.
+	 */
+	if (reset_state == VNIC_PROBED || reset_state == VNIC_DOWN)
+		goto out;
+
+	rc = ibmvnic_login(netdev);
+	if (rc)
+		goto out;
+
+	rc = init_resources(adapter);
+	if (rc)
+		goto out;
+
+	ibmvnic_disable_irqs(adapter);
+	adapter->state = VNIC_CLOSED;
+
+	if (reset_state == VNIC_CLOSED)
+		goto out;
+
+	rc = __ibmvnic_open(netdev);
+	if (rc) {
+		rc = IBMVNIC_OPEN_FAILED;
+		goto out;
+	}
+
+	__netdev_notify_peers(netdev);
+out:
+	/* restore adapter state if reset failed */
+	if (rc)
+		adapter->state = reset_state;
+	netdev_dbg(adapter->netdev, "[S:%s FOP:%d] Hard reset done, rc %d\n",
+		   adapter_state_to_string(adapter->state),
+		   adapter->failover_pending, rc);
+	return rc;
 }
 
 static struct ibmvnic_rwi *get_next_rwi(struct ibmvnic_adapter *adapter)
 {
 	struct ibmvnic_rwi *rwi;
+	unsigned long flags;
 
-	mutex_lock(&adapter->rwi_lock);
+	spin_lock_irqsave(&adapter->rwi_lock, flags);
 
 	if (!list_empty(&adapter->rwi_list)) {
 		rwi = list_first_entry(&adapter->rwi_list, struct ibmvnic_rwi,
@@ -1382,97 +3137,382 @@ static struct ibmvnic_rwi *get_next_rwi(struct ibmvnic_adapter *adapter)
 		rwi = NULL;
 	}
 
-	mutex_unlock(&adapter->rwi_lock);
+	spin_unlock_irqrestore(&adapter->rwi_lock, flags);
 	return rwi;
 }
 
-static void free_all_rwi(struct ibmvnic_adapter *adapter)
+/**
+ * do_passive_init - complete probing when partner device is detected.
+ * @adapter: ibmvnic_adapter struct
+ *
+ * If the ibmvnic device does not have a partner device to communicate with at boot
+ * and that partner device comes online at a later time, this function is called
+ * to complete the initialization process of ibmvnic device.
+ * Caller is expected to hold rtnl_lock().
+ *
+ * Returns non-zero if sub-CRQs are not initialized properly leaving the device
+ * in the down state.
+ * Returns 0 upon success and the device is in PROBED state.
+ */
+
+static int do_passive_init(struct ibmvnic_adapter *adapter)
 {
-	struct ibmvnic_rwi *rwi;
+	unsigned long timeout = msecs_to_jiffies(30000);
+	struct net_device *netdev = adapter->netdev;
+	struct device *dev = &adapter->vdev->dev;
+	int rc;
 
-	rwi = get_next_rwi(adapter);
-	while (rwi) {
-		kfree(rwi);
-		rwi = get_next_rwi(adapter);
+	netdev_dbg(netdev, "Partner device found, probing.\n");
+
+	adapter->state = VNIC_PROBING;
+	reinit_completion(&adapter->init_done);
+	adapter->init_done_rc = 0;
+	adapter->crq.active = true;
+
+	rc = send_crq_init_complete(adapter);
+	if (rc)
+		goto out;
+
+	rc = send_version_xchg(adapter);
+	if (rc)
+		netdev_dbg(adapter->netdev, "send_version_xchg failed, rc=%d\n", rc);
+
+	if (!wait_for_completion_timeout(&adapter->init_done, timeout)) {
+		dev_err(dev, "Initialization sequence timed out\n");
+		rc = -ETIMEDOUT;
+		goto out;
 	}
+
+	rc = init_sub_crqs(adapter);
+	if (rc) {
+		dev_err(dev, "Initialization of sub crqs failed, rc=%d\n", rc);
+		goto out;
+	}
+
+	rc = init_sub_crq_irqs(adapter);
+	if (rc) {
+		dev_err(dev, "Failed to initialize sub crq irqs\n, rc=%d", rc);
+		goto init_failed;
+	}
+
+	netdev->mtu = adapter->req_mtu - ETH_HLEN;
+	netdev->min_mtu = adapter->min_mtu - ETH_HLEN;
+	netdev->max_mtu = adapter->max_mtu - ETH_HLEN;
+
+	adapter->state = VNIC_PROBED;
+	netdev_dbg(netdev, "Probed successfully. Waiting for signal from partner device.\n");
+
+	return 0;
+
+init_failed:
+	release_sub_crqs(adapter, 1);
+out:
+	adapter->state = VNIC_DOWN;
+	return rc;
 }
 
 static void __ibmvnic_reset(struct work_struct *work)
 {
-	struct ibmvnic_rwi *rwi;
 	struct ibmvnic_adapter *adapter;
-	struct net_device *netdev;
+	unsigned int timeout = 5000;
+	struct ibmvnic_rwi *tmprwi;
+	bool saved_state = false;
+	struct ibmvnic_rwi *rwi;
+	unsigned long flags;
+	struct device *dev;
+	bool need_reset;
+	int num_fails = 0;
 	u32 reset_state;
-	int rc;
+	int rc = 0;
 
 	adapter = container_of(work, struct ibmvnic_adapter, ibmvnic_reset);
-	netdev = adapter->netdev;
+		dev = &adapter->vdev->dev;
+
+	/* Wait for ibmvnic_probe() to complete. If probe is taking too long
+	 * or if another reset is in progress, defer work for now. If probe
+	 * eventually fails it will flush and terminate our work.
+	 *
+	 * Three possibilities here:
+	 * 1. Adpater being removed  - just return
+	 * 2. Timed out on probe or another reset in progress - delay the work
+	 * 3. Completed probe - perform any resets in queue
+	 */
+	if (adapter->state == VNIC_PROBING &&
+	    !wait_for_completion_timeout(&adapter->probe_done, timeout)) {
+		dev_err(dev, "Reset thread timed out on probe");
+		queue_delayed_work(system_long_wq,
+				   &adapter->ibmvnic_delayed_reset,
+				   IBMVNIC_RESET_DELAY);
+		return;
+	}
 
-	mutex_lock(&adapter->reset_lock);
-	adapter->resetting = true;
-	reset_state = adapter->state;
+	/* adapter is done with probe (i.e state is never VNIC_PROBING now) */
+	if (adapter->state == VNIC_REMOVING)
+		return;
+
+	/* ->rwi_list is stable now (no one else is removing entries) */
+
+	/* ibmvnic_probe() may have purged the reset queue after we were
+	 * scheduled to process a reset so there maybe no resets to process.
+	 * Before setting the ->resetting bit though, we have to make sure
+	 * that there is infact a reset to process. Otherwise we may race
+	 * with ibmvnic_open() and end up leaving the vnic down:
+	 *
+	 *	__ibmvnic_reset()	    ibmvnic_open()
+	 *	-----------------	    --------------
+	 *
+	 *  set ->resetting bit
+	 *  				find ->resetting bit is set
+	 *  				set ->state to IBMVNIC_OPEN (i.e
+	 *  				assume reset will open device)
+	 *  				return
+	 *  find reset queue empty
+	 *  return
+	 *
+	 *  	Neither performed vnic login/open and vnic stays down
+	 *
+	 * If we hold the lock and conditionally set the bit, either we
+	 * or ibmvnic_open() will complete the open.
+	 */
+	need_reset = false;
+	spin_lock(&adapter->rwi_lock);
+	if (!list_empty(&adapter->rwi_list)) {
+		if (test_and_set_bit_lock(0, &adapter->resetting)) {
+			queue_delayed_work(system_long_wq,
+					   &adapter->ibmvnic_delayed_reset,
+					   IBMVNIC_RESET_DELAY);
+		} else {
+			need_reset = true;
+		}
+	}
+	spin_unlock(&adapter->rwi_lock);
+
+	if (!need_reset)
+		return;
 
 	rwi = get_next_rwi(adapter);
 	while (rwi) {
-		rc = do_reset(adapter, rwi, reset_state);
-		kfree(rwi);
-		if (rc)
+		spin_lock_irqsave(&adapter->state_lock, flags);
+
+		if (adapter->state == VNIC_REMOVING ||
+		    adapter->state == VNIC_REMOVED) {
+			spin_unlock_irqrestore(&adapter->state_lock, flags);
+			kfree(rwi);
+			rc = EBUSY;
 			break;
+		}
+
+		if (!saved_state) {
+			reset_state = adapter->state;
+			saved_state = true;
+		}
+		spin_unlock_irqrestore(&adapter->state_lock, flags);
+
+		if (rwi->reset_reason == VNIC_RESET_PASSIVE_INIT) {
+			rtnl_lock();
+			rc = do_passive_init(adapter);
+			rtnl_unlock();
+			if (!rc)
+				netif_carrier_on(adapter->netdev);
+		} else if (adapter->force_reset_recovery) {
+			/* Since we are doing a hard reset now, clear the
+			 * failover_pending flag so we don't ignore any
+			 * future MOBILITY or other resets.
+			 */
+			adapter->failover_pending = false;
+
+			/* Transport event occurred during previous reset */
+			if (adapter->wait_for_reset) {
+				/* Previous was CHANGE_PARAM; caller locked */
+				adapter->force_reset_recovery = false;
+				rc = do_hard_reset(adapter, rwi, reset_state);
+			} else {
+				rtnl_lock();
+				adapter->force_reset_recovery = false;
+				rc = do_hard_reset(adapter, rwi, reset_state);
+				rtnl_unlock();
+			}
+			if (rc)
+				num_fails++;
+			else
+				num_fails = 0;
+
+			/* If auto-priority-failover is enabled we can get
+			 * back to back failovers during resets, resulting
+			 * in at least two failed resets (from high-priority
+			 * backing device to low-priority one and then back)
+			 * If resets continue to fail beyond that, give the
+			 * adapter some time to settle down before retrying.
+			 */
+			if (num_fails >= 3) {
+				netdev_dbg(adapter->netdev,
+					   "[S:%s] Hard reset failed %d times, waiting 60 secs\n",
+					   adapter_state_to_string(adapter->state),
+					   num_fails);
+				set_current_state(TASK_UNINTERRUPTIBLE);
+				schedule_timeout(60 * HZ);
+			}
+		} else {
+			rc = do_reset(adapter, rwi, reset_state);
+		}
+		tmprwi = rwi;
+		adapter->last_reset_time = jiffies;
+
+		if (rc)
+			netdev_dbg(adapter->netdev, "Reset failed, rc=%d\n", rc);
 
 		rwi = get_next_rwi(adapter);
+
+		/*
+		 * If there are no resets queued and the previous reset failed,
+		 * the adapter would be in an undefined state. So retry the
+		 * previous reset as a hard reset.
+		 *
+		 * Else, free the previous rwi and, if there is another reset
+		 * queued, process the new reset even if previous reset failed
+		 * (the previous reset could have failed because of a fail
+		 * over for instance, so process the fail over).
+		 */
+		if (!rwi && rc)
+			rwi = tmprwi;
+		else
+			kfree(tmprwi);
+
+		if (rwi && (rwi->reset_reason == VNIC_RESET_FAILOVER ||
+			    rwi->reset_reason == VNIC_RESET_MOBILITY || rc))
+			adapter->force_reset_recovery = true;
 	}
 
-	if (rc) {
-		free_all_rwi(adapter);
-		mutex_unlock(&adapter->reset_lock);
-		return;
+	if (adapter->wait_for_reset) {
+		adapter->reset_done_rc = rc;
+		complete(&adapter->reset_done);
 	}
 
-	adapter->resetting = false;
-	mutex_unlock(&adapter->reset_lock);
+	clear_bit_unlock(0, &adapter->resetting);
+
+	netdev_dbg(adapter->netdev,
+		   "[S:%s FRR:%d WFR:%d] Done processing resets\n",
+		   adapter_state_to_string(adapter->state),
+		   adapter->force_reset_recovery,
+		   adapter->wait_for_reset);
 }
 
-static void ibmvnic_reset(struct ibmvnic_adapter *adapter,
-			  enum ibmvnic_reset_reason reason)
+static void __ibmvnic_delayed_reset(struct work_struct *work)
+{
+	struct ibmvnic_adapter *adapter;
+
+	adapter = container_of(work, struct ibmvnic_adapter,
+			       ibmvnic_delayed_reset.work);
+	__ibmvnic_reset(&adapter->ibmvnic_reset);
+}
+
+static void flush_reset_queue(struct ibmvnic_adapter *adapter)
+{
+	struct list_head *entry, *tmp_entry;
+
+	if (!list_empty(&adapter->rwi_list)) {
+		list_for_each_safe(entry, tmp_entry, &adapter->rwi_list) {
+			list_del(entry);
+			kfree(list_entry(entry, struct ibmvnic_rwi, list));
+		}
+	}
+}
+
+static int ibmvnic_reset(struct ibmvnic_adapter *adapter,
+			 enum ibmvnic_reset_reason reason)
 {
-	struct ibmvnic_rwi *rwi, *tmp;
 	struct net_device *netdev = adapter->netdev;
-	struct list_head *entry;
+	struct ibmvnic_rwi *rwi, *tmp;
+	unsigned long flags;
+	int ret;
+
+	spin_lock_irqsave(&adapter->rwi_lock, flags);
 
+	/* If failover is pending don't schedule any other reset.
+	 * Instead let the failover complete. If there is already a
+	 * a failover reset scheduled, we will detect and drop the
+	 * duplicate reset when walking the ->rwi_list below.
+	 */
 	if (adapter->state == VNIC_REMOVING ||
-	    adapter->state == VNIC_REMOVED) {
-		netdev_dbg(netdev, "Adapter removing, skipping reset\n");
-		return;
+	    adapter->state == VNIC_REMOVED ||
+	    (adapter->failover_pending && reason != VNIC_RESET_FAILOVER)) {
+		ret = EBUSY;
+		netdev_dbg(netdev, "Adapter removing or pending failover, skipping reset\n");
+		goto err;
 	}
 
-	mutex_lock(&adapter->rwi_lock);
-
-	list_for_each(entry, &adapter->rwi_list) {
-		tmp = list_entry(entry, struct ibmvnic_rwi, list);
+	list_for_each_entry(tmp, &adapter->rwi_list, list) {
 		if (tmp->reset_reason == reason) {
-			netdev_err(netdev, "Matching reset found, skipping\n");
-			mutex_unlock(&adapter->rwi_lock);
-			return;
+			netdev_dbg(netdev, "Skipping matching reset, reason=%s\n",
+				   reset_reason_to_string(reason));
+			ret = EBUSY;
+			goto err;
 		}
 	}
 
-	rwi = kzalloc(sizeof(*rwi), GFP_KERNEL);
+	rwi = kzalloc(sizeof(*rwi), GFP_ATOMIC);
 	if (!rwi) {
-		mutex_unlock(&adapter->rwi_lock);
-		ibmvnic_close(netdev);
-		return;
+		ret = ENOMEM;
+		goto err;
 	}
+	/* if we just received a transport event,
+	 * flush reset queue and process this reset
+	 */
+	if (adapter->force_reset_recovery)
+		flush_reset_queue(adapter);
 
 	rwi->reset_reason = reason;
 	list_add_tail(&rwi->list, &adapter->rwi_list);
-	mutex_unlock(&adapter->rwi_lock);
-	schedule_work(&adapter->ibmvnic_reset);
+	netdev_dbg(adapter->netdev, "Scheduling reset (reason %s)\n",
+		   reset_reason_to_string(reason));
+	queue_work(system_long_wq, &adapter->ibmvnic_reset);
+
+	ret = 0;
+err:
+	/* ibmvnic_close() below can block, so drop the lock first */
+	spin_unlock_irqrestore(&adapter->rwi_lock, flags);
+
+	if (ret == ENOMEM)
+		ibmvnic_close(netdev);
+
+	return -ret;
 }
 
-static void ibmvnic_tx_timeout(struct net_device *dev)
+static void ibmvnic_get_stats64(struct net_device *netdev,
+				struct rtnl_link_stats64 *stats)
+{
+	struct ibmvnic_adapter *adapter = netdev_priv(netdev);
+	int i;
+
+	for (i = 0; i < adapter->req_rx_queues; i++) {
+		stats->rx_packets += adapter->rx_stats_buffers[i].packets;
+		stats->rx_bytes   += adapter->rx_stats_buffers[i].bytes;
+	}
+
+	for (i = 0; i < adapter->req_tx_queues; i++) {
+		stats->tx_packets += adapter->tx_stats_buffers[i].batched_packets;
+		stats->tx_packets += adapter->tx_stats_buffers[i].direct_packets;
+		stats->tx_bytes   += adapter->tx_stats_buffers[i].bytes;
+		stats->tx_dropped += adapter->tx_stats_buffers[i].dropped_packets;
+	}
+}
+
+static void ibmvnic_tx_timeout(struct net_device *dev, unsigned int txqueue)
 {
 	struct ibmvnic_adapter *adapter = netdev_priv(dev);
 
+	if (test_bit(0, &adapter->resetting)) {
+		netdev_err(adapter->netdev,
+			   "Adapter is resetting, skip timeout reset\n");
+		return;
+	}
+	/* No queuing up reset until at least 5 seconds (default watchdog val)
+	 * after last reset
+	 */
+	if (time_before(jiffies, (adapter->last_reset_time + dev->watchdog_timeo))) {
+		netdev_dbg(dev, "Not yet time to tx timeout.\n");
+		return;
+	}
 	ibmvnic_reset(adapter, VNIC_RESET_TIMEOUT);
 }
 
@@ -1491,13 +3531,17 @@ static void remove_buff_from_pool(struct ibmvnic_adapter *adapter,
 
 static int ibmvnic_poll(struct napi_struct *napi, int budget)
 {
-	struct net_device *netdev = napi->dev;
-	struct ibmvnic_adapter *adapter = netdev_priv(netdev);
-	int scrq_num = (int)(napi - adapter->napi);
-	int frames_processed = 0;
+	struct ibmvnic_sub_crq_queue *rx_scrq;
+	struct ibmvnic_adapter *adapter;
+	struct net_device *netdev;
+	int frames_processed;
+	int scrq_num;
 
-	if (adapter->resetting)
-		return 0;
+	netdev = napi->dev;
+	adapter = netdev_priv(netdev);
+	scrq_num = (int)(napi - adapter->napi);
+	frames_processed = 0;
+	rx_scrq = adapter->rx_scrq[scrq_num];
 
 restart_poll:
 	while (frames_processed < budget) {
@@ -1508,15 +3552,28 @@ restart_poll:
 		u16 offset;
 		u8 flags = 0;
 
-		if (!pending_scrq(adapter, adapter->rx_scrq[scrq_num]))
+		if (unlikely(test_bit(0, &adapter->resetting) &&
+			     adapter->reset_reason != VNIC_RESET_NON_FATAL)) {
+			enable_scrq_irq(adapter, rx_scrq);
+			napi_complete_done(napi, frames_processed);
+			return frames_processed;
+		}
+
+		if (!pending_scrq(adapter, rx_scrq))
 			break;
-		next = ibmvnic_next_scrq(adapter, adapter->rx_scrq[scrq_num]);
-		rx_buff =
-		    (struct ibmvnic_rx_buff *)be64_to_cpu(next->
-							  rx_comp.correlator);
+		next = ibmvnic_next_scrq(adapter, rx_scrq);
+		rx_buff = (struct ibmvnic_rx_buff *)
+			  be64_to_cpu(next->rx_comp.correlator);
 		/* do error checking */
 		if (next->rx_comp.rc) {
-			netdev_err(netdev, "rx error %x\n", next->rx_comp.rc);
+			netdev_dbg(netdev, "rx buffer returned with rc %x\n",
+				   be16_to_cpu(next->rx_comp.rc));
+			/* free the entry */
+			next->rx_comp.first = 0;
+			dev_kfree_skb_any(rx_buff->skb);
+			remove_buff_from_pool(adapter, rx_buff);
+			continue;
+		} else if (!rx_buff->skb) {
 			/* free the entry */
 			next->rx_comp.first = 0;
 			remove_buff_from_pool(adapter, rx_buff);
@@ -1527,6 +3584,8 @@ restart_poll:
 		offset = be16_to_cpu(next->rx_comp.off_frame_data);
 		flags = next->rx_comp.flags;
 		skb = rx_buff->skb;
+		/* load long_term_buff before copying to skb */
+		dma_rmb();
 		skb_copy_to_linear_data(skb, rx_buff->data + offset,
 					length);
 
@@ -1553,38 +3612,107 @@ restart_poll:
 
 		length = skb->len;
 		napi_gro_receive(napi, skb); /* send it up */
-		netdev->stats.rx_packets++;
-		netdev->stats.rx_bytes += length;
+		adapter->rx_stats_buffers[scrq_num].packets++;
+		adapter->rx_stats_buffers[scrq_num].bytes += length;
 		frames_processed++;
 	}
 
-	if (adapter->state != VNIC_CLOSING)
+	if (adapter->state != VNIC_CLOSING &&
+	    (atomic_read(&adapter->rx_pool[scrq_num].available) <
+	      adapter->req_rx_add_entries_per_subcrq / 2))
 		replenish_rx_pool(adapter, &adapter->rx_pool[scrq_num]);
-
 	if (frames_processed < budget) {
-		enable_scrq_irq(adapter, adapter->rx_scrq[scrq_num]);
-		napi_complete_done(napi, frames_processed);
-		if (pending_scrq(adapter, adapter->rx_scrq[scrq_num]) &&
-		    napi_reschedule(napi)) {
-			disable_scrq_irq(adapter, adapter->rx_scrq[scrq_num]);
-			goto restart_poll;
+		if (napi_complete_done(napi, frames_processed)) {
+			enable_scrq_irq(adapter, rx_scrq);
+			if (pending_scrq(adapter, rx_scrq)) {
+				if (napi_schedule(napi)) {
+					disable_scrq_irq(adapter, rx_scrq);
+					goto restart_poll;
+				}
+			}
 		}
 	}
 	return frames_processed;
 }
 
-#ifdef CONFIG_NET_POLL_CONTROLLER
-static void ibmvnic_netpoll_controller(struct net_device *dev)
+static int wait_for_reset(struct ibmvnic_adapter *adapter)
 {
-	struct ibmvnic_adapter *adapter = netdev_priv(dev);
-	int i;
+	int rc, ret;
 
-	replenish_pools(netdev_priv(dev));
-	for (i = 0; i < adapter->req_rx_queues; i++)
-		ibmvnic_interrupt_rx(adapter->rx_scrq[i]->irq,
-				     adapter->rx_scrq[i]);
+	adapter->fallback.mtu = adapter->req_mtu;
+	adapter->fallback.rx_queues = adapter->req_rx_queues;
+	adapter->fallback.tx_queues = adapter->req_tx_queues;
+	adapter->fallback.rx_entries = adapter->req_rx_add_entries_per_subcrq;
+	adapter->fallback.tx_entries = adapter->req_tx_entries_per_subcrq;
+
+	reinit_completion(&adapter->reset_done);
+	adapter->wait_for_reset = true;
+	rc = ibmvnic_reset(adapter, VNIC_RESET_CHANGE_PARAM);
+
+	if (rc) {
+		ret = rc;
+		goto out;
+	}
+	rc = ibmvnic_wait_for_completion(adapter, &adapter->reset_done, 60000);
+	if (rc) {
+		ret = -ENODEV;
+		goto out;
+	}
+
+	ret = 0;
+	if (adapter->reset_done_rc) {
+		ret = -EIO;
+		adapter->desired.mtu = adapter->fallback.mtu;
+		adapter->desired.rx_queues = adapter->fallback.rx_queues;
+		adapter->desired.tx_queues = adapter->fallback.tx_queues;
+		adapter->desired.rx_entries = adapter->fallback.rx_entries;
+		adapter->desired.tx_entries = adapter->fallback.tx_entries;
+
+		reinit_completion(&adapter->reset_done);
+		adapter->wait_for_reset = true;
+		rc = ibmvnic_reset(adapter, VNIC_RESET_CHANGE_PARAM);
+		if (rc) {
+			ret = rc;
+			goto out;
+		}
+		rc = ibmvnic_wait_for_completion(adapter, &adapter->reset_done,
+						 60000);
+		if (rc) {
+			ret = -ENODEV;
+			goto out;
+		}
+	}
+out:
+	adapter->wait_for_reset = false;
+
+	return ret;
+}
+
+static int ibmvnic_change_mtu(struct net_device *netdev, int new_mtu)
+{
+	struct ibmvnic_adapter *adapter = netdev_priv(netdev);
+
+	adapter->desired.mtu = new_mtu + ETH_HLEN;
+
+	return wait_for_reset(adapter);
+}
+
+static netdev_features_t ibmvnic_features_check(struct sk_buff *skb,
+						struct net_device *dev,
+						netdev_features_t features)
+{
+	/* Some backing hardware adapters can not
+	 * handle packets with a MSS less than 224
+	 * or with only one segment.
+	 */
+	if (skb_is_gso(skb)) {
+		if (skb_shinfo(skb)->gso_size < 224 ||
+		    skb_shinfo(skb)->gso_segs == 1)
+			features &= ~NETIF_F_GSO_MASK;
+	}
+
+	return features;
 }
-#endif
 
 static const struct net_device_ops ibmvnic_netdev_ops = {
 	.ndo_open		= ibmvnic_open,
@@ -1593,10 +3721,10 @@ static const struct net_device_ops ibmvnic_netdev_ops = {
 	.ndo_set_rx_mode	= ibmvnic_set_multi,
 	.ndo_set_mac_address	= ibmvnic_set_mac,
 	.ndo_validate_addr	= eth_validate_addr,
+	.ndo_get_stats64	= ibmvnic_get_stats64,
 	.ndo_tx_timeout		= ibmvnic_tx_timeout,
-#ifdef CONFIG_NET_POLL_CONTROLLER
-	.ndo_poll_controller	= ibmvnic_netpoll_controller,
-#endif
+	.ndo_change_mtu		= ibmvnic_change_mtu,
+	.ndo_features_check     = ibmvnic_features_check,
 };
 
 /* ethtool functions */
@@ -1604,31 +3732,32 @@ static const struct net_device_ops ibmvnic_netdev_ops = {
 static int ibmvnic_get_link_ksettings(struct net_device *netdev,
 				      struct ethtool_link_ksettings *cmd)
 {
-	u32 supported, advertising;
+	struct ibmvnic_adapter *adapter = netdev_priv(netdev);
+	int rc;
 
-	supported = (SUPPORTED_1000baseT_Full | SUPPORTED_Autoneg |
-			  SUPPORTED_FIBRE);
-	advertising = (ADVERTISED_1000baseT_Full | ADVERTISED_Autoneg |
-			    ADVERTISED_FIBRE);
-	cmd->base.speed = SPEED_1000;
-	cmd->base.duplex = DUPLEX_FULL;
+	rc = send_query_phys_parms(adapter);
+	if (rc) {
+		adapter->speed = SPEED_UNKNOWN;
+		adapter->duplex = DUPLEX_UNKNOWN;
+	}
+	cmd->base.speed = adapter->speed;
+	cmd->base.duplex = adapter->duplex;
 	cmd->base.port = PORT_FIBRE;
 	cmd->base.phy_address = 0;
 	cmd->base.autoneg = AUTONEG_ENABLE;
 
-	ethtool_convert_legacy_u32_to_link_mode(cmd->link_modes.supported,
-						supported);
-	ethtool_convert_legacy_u32_to_link_mode(cmd->link_modes.advertising,
-						advertising);
-
 	return 0;
 }
 
-static void ibmvnic_get_drvinfo(struct net_device *dev,
+static void ibmvnic_get_drvinfo(struct net_device *netdev,
 				struct ethtool_drvinfo *info)
 {
-	strlcpy(info->driver, ibmvnic_driver_name, sizeof(info->driver));
-	strlcpy(info->version, IBMVNIC_DRIVER_VERSION, sizeof(info->version));
+	struct ibmvnic_adapter *adapter = netdev_priv(netdev);
+
+	strscpy(info->driver, ibmvnic_driver_name, sizeof(info->driver));
+	strscpy(info->version, IBMVNIC_DRIVER_VERSION, sizeof(info->version));
+	strscpy(info->fw_version, adapter->fw_version,
+		sizeof(info->fw_version));
 }
 
 static u32 ibmvnic_get_msglevel(struct net_device *netdev)
@@ -1656,34 +3785,105 @@ static u32 ibmvnic_get_link(struct net_device *netdev)
 }
 
 static void ibmvnic_get_ringparam(struct net_device *netdev,
-				  struct ethtool_ringparam *ring)
+				  struct ethtool_ringparam *ring,
+				  struct kernel_ethtool_ringparam *kernel_ring,
+				  struct netlink_ext_ack *extack)
 {
-	ring->rx_max_pending = 0;
-	ring->tx_max_pending = 0;
+	struct ibmvnic_adapter *adapter = netdev_priv(netdev);
+
+	ring->rx_max_pending = adapter->max_rx_add_entries_per_subcrq;
+	ring->tx_max_pending = adapter->max_tx_entries_per_subcrq;
 	ring->rx_mini_max_pending = 0;
 	ring->rx_jumbo_max_pending = 0;
-	ring->rx_pending = 0;
-	ring->tx_pending = 0;
+	ring->rx_pending = adapter->req_rx_add_entries_per_subcrq;
+	ring->tx_pending = adapter->req_tx_entries_per_subcrq;
 	ring->rx_mini_pending = 0;
 	ring->rx_jumbo_pending = 0;
 }
 
+static int ibmvnic_set_ringparam(struct net_device *netdev,
+				 struct ethtool_ringparam *ring,
+				 struct kernel_ethtool_ringparam *kernel_ring,
+				 struct netlink_ext_ack *extack)
+{
+	struct ibmvnic_adapter *adapter = netdev_priv(netdev);
+
+	if (ring->rx_pending > adapter->max_rx_add_entries_per_subcrq  ||
+	    ring->tx_pending > adapter->max_tx_entries_per_subcrq) {
+		netdev_err(netdev, "Invalid request.\n");
+		netdev_err(netdev, "Max tx buffers = %llu\n",
+			   adapter->max_rx_add_entries_per_subcrq);
+		netdev_err(netdev, "Max rx buffers = %llu\n",
+			   adapter->max_tx_entries_per_subcrq);
+		return -EINVAL;
+	}
+
+	adapter->desired.rx_entries = ring->rx_pending;
+	adapter->desired.tx_entries = ring->tx_pending;
+
+	return wait_for_reset(adapter);
+}
+
+static void ibmvnic_get_channels(struct net_device *netdev,
+				 struct ethtool_channels *channels)
+{
+	struct ibmvnic_adapter *adapter = netdev_priv(netdev);
+
+	channels->max_rx = adapter->max_rx_queues;
+	channels->max_tx = adapter->max_tx_queues;
+	channels->max_other = 0;
+	channels->max_combined = 0;
+	channels->rx_count = adapter->req_rx_queues;
+	channels->tx_count = adapter->req_tx_queues;
+	channels->other_count = 0;
+	channels->combined_count = 0;
+}
+
+static int ibmvnic_set_channels(struct net_device *netdev,
+				struct ethtool_channels *channels)
+{
+	struct ibmvnic_adapter *adapter = netdev_priv(netdev);
+
+	adapter->desired.rx_queues = channels->rx_count;
+	adapter->desired.tx_queues = channels->tx_count;
+
+	return wait_for_reset(adapter);
+}
+
 static void ibmvnic_get_strings(struct net_device *dev, u32 stringset, u8 *data)
 {
+	struct ibmvnic_adapter *adapter = netdev_priv(dev);
 	int i;
 
 	if (stringset != ETH_SS_STATS)
 		return;
 
-	for (i = 0; i < ARRAY_SIZE(ibmvnic_stats); i++, data += ETH_GSTRING_LEN)
-		memcpy(data, ibmvnic_stats[i].name, ETH_GSTRING_LEN);
+	for (i = 0; i < ARRAY_SIZE(ibmvnic_stats); i++)
+		ethtool_puts(&data, ibmvnic_stats[i].name);
+
+	for (i = 0; i < adapter->req_tx_queues; i++) {
+		ethtool_sprintf(&data, "tx%d_batched_packets", i);
+		ethtool_sprintf(&data, "tx%d_direct_packets", i);
+		ethtool_sprintf(&data, "tx%d_bytes", i);
+		ethtool_sprintf(&data, "tx%d_dropped_packets", i);
+	}
+
+	for (i = 0; i < adapter->req_rx_queues; i++) {
+		ethtool_sprintf(&data, "rx%d_packets", i);
+		ethtool_sprintf(&data, "rx%d_bytes", i);
+		ethtool_sprintf(&data, "rx%d_interrupts", i);
+	}
 }
 
 static int ibmvnic_get_sset_count(struct net_device *dev, int sset)
 {
+	struct ibmvnic_adapter *adapter = netdev_priv(dev);
+
 	switch (sset) {
 	case ETH_SS_STATS:
-		return ARRAY_SIZE(ibmvnic_stats);
+		return ARRAY_SIZE(ibmvnic_stats) +
+		       adapter->req_tx_queues * NUM_TX_STATS +
+		       adapter->req_rx_queues * NUM_RX_STATS;
 	default:
 		return -EOPNOTSUPP;
 	}
@@ -1694,7 +3894,8 @@ static void ibmvnic_get_ethtool_stats(struct net_device *dev,
 {
 	struct ibmvnic_adapter *adapter = netdev_priv(dev);
 	union ibmvnic_crq crq;
-	int i;
+	int i, j;
+	int rc;
 
 	memset(&crq, 0, sizeof(crq));
 	crq.request_statistics.first = IBMVNIC_CRQ_CMD;
@@ -1704,12 +3905,37 @@ static void ibmvnic_get_ethtool_stats(struct net_device *dev,
 	    cpu_to_be32(sizeof(struct ibmvnic_statistics));
 
 	/* Wait for data to be written */
-	init_completion(&adapter->stats_done);
-	ibmvnic_send_crq(adapter, &crq);
-	wait_for_completion(&adapter->stats_done);
+	reinit_completion(&adapter->stats_done);
+	rc = ibmvnic_send_crq(adapter, &crq);
+	if (rc)
+		return;
+	rc = ibmvnic_wait_for_completion(adapter, &adapter->stats_done, 10000);
+	if (rc)
+		return;
 
 	for (i = 0; i < ARRAY_SIZE(ibmvnic_stats); i++)
-		data[i] = IBMVNIC_GET_STAT(adapter, ibmvnic_stats[i].offset);
+		data[i] = be64_to_cpu(IBMVNIC_GET_STAT
+				      (adapter, ibmvnic_stats[i].offset));
+
+	for (j = 0; j < adapter->req_tx_queues; j++) {
+		data[i] = adapter->tx_stats_buffers[j].batched_packets;
+		i++;
+		data[i] = adapter->tx_stats_buffers[j].direct_packets;
+		i++;
+		data[i] = adapter->tx_stats_buffers[j].bytes;
+		i++;
+		data[i] = adapter->tx_stats_buffers[j].dropped_packets;
+		i++;
+	}
+
+	for (j = 0; j < adapter->req_rx_queues; j++) {
+		data[i] = adapter->rx_stats_buffers[j].packets;
+		i++;
+		data[i] = adapter->rx_stats_buffers[j].bytes;
+		i++;
+		data[i] = adapter->rx_stats_buffers[j].interrupts;
+		i++;
+	}
 }
 
 static const struct ethtool_ops ibmvnic_ethtool_ops = {
@@ -1718,6 +3944,9 @@ static const struct ethtool_ops ibmvnic_ethtool_ops = {
 	.set_msglevel		= ibmvnic_set_msglevel,
 	.get_link		= ibmvnic_get_link,
 	.get_ringparam		= ibmvnic_get_ringparam,
+	.set_ringparam		= ibmvnic_set_ringparam,
+	.get_channels		= ibmvnic_get_channels,
+	.set_channels		= ibmvnic_set_channels,
 	.get_strings            = ibmvnic_get_strings,
 	.get_sset_count         = ibmvnic_get_sset_count,
 	.get_ethtool_stats	= ibmvnic_get_ethtool_stats,
@@ -1731,14 +3960,26 @@ static int reset_one_sub_crq_queue(struct ibmvnic_adapter *adapter,
 {
 	int rc;
 
+	if (!scrq) {
+		netdev_dbg(adapter->netdev, "Invalid scrq reset.\n");
+		return -EINVAL;
+	}
+
 	if (scrq->irq) {
 		free_irq(scrq->irq, scrq);
 		irq_dispose_mapping(scrq->irq);
 		scrq->irq = 0;
 	}
 
-	memset(scrq->msgs, 0, 2 * PAGE_SIZE);
-	scrq->cur = 0;
+	if (scrq->msgs) {
+		memset(scrq->msgs, 0, 4 * PAGE_SIZE);
+		atomic_set(&scrq->used, 0);
+		scrq->cur = 0;
+		scrq->ind_buf.index = 0;
+	} else {
+		netdev_dbg(adapter->netdev, "Invalid scrq reset\n");
+		return -EINVAL;
+	}
 
 	rc = h_reg_sub_crq(adapter->vdev->unit_address, scrq->msg_token,
 			   4 * PAGE_SIZE, &scrq->crq_num, &scrq->hw_irq);
@@ -1749,46 +3990,61 @@ static int reset_sub_crq_queues(struct ibmvnic_adapter *adapter)
 {
 	int i, rc;
 
+	if (!adapter->tx_scrq || !adapter->rx_scrq)
+		return -EINVAL;
+
+	ibmvnic_clean_affinity(adapter);
+
 	for (i = 0; i < adapter->req_tx_queues; i++) {
+		netdev_dbg(adapter->netdev, "Re-setting tx_scrq[%d]\n", i);
 		rc = reset_one_sub_crq_queue(adapter, adapter->tx_scrq[i]);
 		if (rc)
 			return rc;
 	}
 
 	for (i = 0; i < adapter->req_rx_queues; i++) {
+		netdev_dbg(adapter->netdev, "Re-setting rx_scrq[%d]\n", i);
 		rc = reset_one_sub_crq_queue(adapter, adapter->rx_scrq[i]);
 		if (rc)
 			return rc;
 	}
 
-	rc = init_sub_crq_irqs(adapter);
 	return rc;
 }
 
 static void release_sub_crq_queue(struct ibmvnic_adapter *adapter,
-				  struct ibmvnic_sub_crq_queue *scrq)
+				  struct ibmvnic_sub_crq_queue *scrq,
+				  bool do_h_free)
 {
 	struct device *dev = &adapter->vdev->dev;
 	long rc;
 
 	netdev_dbg(adapter->netdev, "Releasing sub-CRQ\n");
 
-	/* Close the sub-crqs */
-	do {
-		rc = plpar_hcall_norets(H_FREE_SUB_CRQ,
-					adapter->vdev->unit_address,
-					scrq->crq_num);
-	} while (rc == H_BUSY || H_IS_LONG_BUSY(rc));
+	if (do_h_free) {
+		/* Close the sub-crqs */
+		do {
+			rc = plpar_hcall_norets(H_FREE_SUB_CRQ,
+						adapter->vdev->unit_address,
+						scrq->crq_num);
+		} while (rc == H_BUSY || H_IS_LONG_BUSY(rc));
 
-	if (rc) {
-		netdev_err(adapter->netdev,
-			   "Failed to release sub-CRQ %16lx, rc = %ld\n",
-			   scrq->crq_num, rc);
+		if (rc) {
+			netdev_err(adapter->netdev,
+				   "Failed to release sub-CRQ %16lx, rc = %ld\n",
+				   scrq->crq_num, rc);
+		}
 	}
 
+	dma_free_coherent(dev,
+			  IBMVNIC_IND_MAX_ARR_SZ,
+			  scrq->ind_buf.indir_arr,
+			  scrq->ind_buf.indir_dma);
+
 	dma_unmap_single(dev, scrq->msg_token, 4 * PAGE_SIZE,
 			 DMA_BIDIRECTIONAL);
 	free_pages((unsigned long)scrq->msgs, 2);
+	free_cpumask_var(scrq->affinity_mask);
 	kfree(scrq);
 }
 
@@ -1809,6 +4065,8 @@ static struct ibmvnic_sub_crq_queue *init_sub_crq_queue(struct ibmvnic_adapter
 		dev_warn(dev, "Couldn't allocate crq queue messages page\n");
 		goto zero_page_failed;
 	}
+	if (!zalloc_cpumask_var(&scrq->affinity_mask, GFP_KERNEL))
+		goto cpumask_alloc_failed;
 
 	scrq->msg_token = dma_map_single(dev, scrq->msgs, 4 * PAGE_SIZE,
 					 DMA_BIDIRECTIONAL);
@@ -1832,6 +4090,17 @@ static struct ibmvnic_sub_crq_queue *init_sub_crq_queue(struct ibmvnic_adapter
 
 	scrq->adapter = adapter;
 	scrq->size = 4 * PAGE_SIZE / sizeof(*scrq->msgs);
+	scrq->ind_buf.index = 0;
+
+	scrq->ind_buf.indir_arr =
+		dma_alloc_coherent(dev,
+				   IBMVNIC_IND_MAX_ARR_SZ,
+				   &scrq->ind_buf.indir_dma,
+				   GFP_KERNEL);
+
+	if (!scrq->ind_buf.indir_arr)
+		goto indir_failed;
+
 	spin_lock_init(&scrq->lock);
 
 	netdev_dbg(adapter->netdev,
@@ -1840,10 +4109,18 @@ static struct ibmvnic_sub_crq_queue *init_sub_crq_queue(struct ibmvnic_adapter
 
 	return scrq;
 
+indir_failed:
+	do {
+		rc = plpar_hcall_norets(H_FREE_SUB_CRQ,
+					adapter->vdev->unit_address,
+					scrq->crq_num);
+	} while (rc == H_BUSY || rc == H_IS_LONG_BUSY(rc));
 reg_failed:
 	dma_unmap_single(dev, scrq->msg_token, 4 * PAGE_SIZE,
 			 DMA_BIDIRECTIONAL);
 map_failed:
+	free_cpumask_var(scrq->affinity_mask);
+cpumask_alloc_failed:
 	free_pages((unsigned long)scrq->msgs, 2);
 zero_page_failed:
 	kfree(scrq);
@@ -1851,15 +4128,19 @@ zero_page_failed:
 	return NULL;
 }
 
-static void release_sub_crqs(struct ibmvnic_adapter *adapter)
+static void release_sub_crqs(struct ibmvnic_adapter *adapter, bool do_h_free)
 {
 	int i;
 
+	ibmvnic_clean_affinity(adapter);
 	if (adapter->tx_scrq) {
-		for (i = 0; i < adapter->req_tx_queues; i++) {
+		for (i = 0; i < adapter->num_active_tx_scrqs; i++) {
 			if (!adapter->tx_scrq[i])
 				continue;
 
+			netdev_dbg(adapter->netdev, "Releasing tx_scrq[%d]\n",
+				   i);
+			ibmvnic_tx_scrq_clean_buffer(adapter, adapter->tx_scrq[i]);
 			if (adapter->tx_scrq[i]->irq) {
 				free_irq(adapter->tx_scrq[i]->irq,
 					 adapter->tx_scrq[i]);
@@ -1867,18 +4148,28 @@ static void release_sub_crqs(struct ibmvnic_adapter *adapter)
 				adapter->tx_scrq[i]->irq = 0;
 			}
 
-			release_sub_crq_queue(adapter, adapter->tx_scrq[i]);
+			release_sub_crq_queue(adapter, adapter->tx_scrq[i],
+					      do_h_free);
 		}
 
 		kfree(adapter->tx_scrq);
 		adapter->tx_scrq = NULL;
+		adapter->num_active_tx_scrqs = 0;
 	}
 
+	/* Clean any remaining outstanding SKBs
+	 * we freed the irq so we won't be hearing
+	 * from them
+	 */
+	clean_tx_pools(adapter);
+
 	if (adapter->rx_scrq) {
-		for (i = 0; i < adapter->req_rx_queues; i++) {
+		for (i = 0; i < adapter->num_active_rx_scrqs; i++) {
 			if (!adapter->rx_scrq[i])
 				continue;
 
+			netdev_dbg(adapter->netdev, "Releasing rx_scrq[%d]\n",
+				   i);
 			if (adapter->rx_scrq[i]->irq) {
 				free_irq(adapter->rx_scrq[i]->irq,
 					 adapter->rx_scrq[i]);
@@ -1886,11 +4177,13 @@ static void release_sub_crqs(struct ibmvnic_adapter *adapter)
 				adapter->rx_scrq[i]->irq = 0;
 			}
 
-			release_sub_crq_queue(adapter, adapter->rx_scrq[i]);
+			release_sub_crq_queue(adapter, adapter->rx_scrq[i],
+					      do_h_free);
 		}
 
 		kfree(adapter->rx_scrq);
 		adapter->rx_scrq = NULL;
+		adapter->num_active_rx_scrqs = 0;
 	}
 }
 
@@ -1908,6 +4201,30 @@ static int disable_scrq_irq(struct ibmvnic_adapter *adapter,
 	return rc;
 }
 
+/* We can not use the IRQ chip EOI handler because that has the
+ * unintended effect of changing the interrupt priority.
+ */
+static void ibmvnic_xics_eoi(struct device *dev, struct ibmvnic_sub_crq_queue *scrq)
+{
+	u64 val = 0xff000000 | scrq->hw_irq;
+	unsigned long rc;
+
+	rc = plpar_hcall_norets(H_EOI, val);
+	if (rc)
+		dev_err(dev, "H_EOI FAILED irq 0x%llx. rc=%ld\n", val, rc);
+}
+
+/* Due to a firmware bug, the hypervisor can send an interrupt to a
+ * transmit or receive queue just prior to a partition migration.
+ * Force an EOI after migration.
+ */
+static void ibmvnic_clear_pending_interrupt(struct device *dev,
+					    struct ibmvnic_sub_crq_queue *scrq)
+{
+	if (!xive_enabled())
+		ibmvnic_xics_eoi(dev, scrq);
+}
+
 static int enable_scrq_irq(struct ibmvnic_adapter *adapter,
 			   struct ibmvnic_sub_crq_queue *scrq)
 {
@@ -1919,6 +4236,11 @@ static int enable_scrq_irq(struct ibmvnic_adapter *adapter,
 		return 1;
 	}
 
+	if (test_bit(0, &adapter->resetting) &&
+	    adapter->reset_reason == VNIC_RESET_MOBILITY) {
+		ibmvnic_clear_pending_interrupt(dev, scrq);
+	}
+
 	rc = plpar_hcall_norets(H_VIOCTL, adapter->vdev->unit_address,
 				H_ENABLE_VIO_INTERRUPT, scrq->hw_irq, 0, 0);
 	if (rc)
@@ -1931,61 +4253,66 @@ static int ibmvnic_complete_tx(struct ibmvnic_adapter *adapter,
 			       struct ibmvnic_sub_crq_queue *scrq)
 {
 	struct device *dev = &adapter->vdev->dev;
+	int num_packets = 0, total_bytes = 0;
+	struct ibmvnic_tx_pool *tx_pool;
 	struct ibmvnic_tx_buff *txbuff;
+	struct netdev_queue *txq;
 	union sub_crq *next;
-	int index;
-	int i, j;
-	u8 first;
+	int index, i;
 
 restart_loop:
 	while (pending_scrq(adapter, scrq)) {
 		unsigned int pool = scrq->pool_index;
-
+		int num_entries = 0;
 		next = ibmvnic_next_scrq(adapter, scrq);
 		for (i = 0; i < next->tx_comp.num_comps; i++) {
-			if (next->tx_comp.rcs[i]) {
-				dev_err(dev, "tx error %x\n",
-					next->tx_comp.rcs[i]);
-				continue;
-			}
 			index = be32_to_cpu(next->tx_comp.correlators[i]);
-			txbuff = &adapter->tx_pool[pool].tx_buff[index];
-
-			for (j = 0; j < IBMVNIC_MAX_FRAGS_PER_CRQ; j++) {
-				if (!txbuff->data_dma[j])
-					continue;
-
-				txbuff->data_dma[j] = 0;
-			}
-			/* if sub_crq was sent indirectly */
-			first = txbuff->indir_arr[0].generic.first;
-			if (first == IBMVNIC_CRQ_CMD) {
-				dma_unmap_single(dev, txbuff->indir_dma,
-						 sizeof(txbuff->indir_arr),
-						 DMA_TO_DEVICE);
+			if (index & IBMVNIC_TSO_POOL_MASK) {
+				tx_pool = &adapter->tso_pool[pool];
+				index &= ~IBMVNIC_TSO_POOL_MASK;
+			} else {
+				tx_pool = &adapter->tx_pool[pool];
 			}
 
-			if (txbuff->last_frag) {
-				dev_kfree_skb_any(txbuff->skb);
+			txbuff = &tx_pool->tx_buff[index];
+			num_packets++;
+			num_entries += txbuff->num_entries;
+			if (txbuff->skb) {
+				total_bytes += txbuff->skb->len;
+				if (next->tx_comp.rcs[i]) {
+					dev_err(dev, "tx error %x\n",
+						next->tx_comp.rcs[i]);
+					dev_kfree_skb_irq(txbuff->skb);
+				} else {
+					dev_consume_skb_irq(txbuff->skb);
+				}
 				txbuff->skb = NULL;
+			} else {
+				netdev_warn(adapter->netdev,
+					    "TX completion received with NULL socket buffer\n");
 			}
-
-			adapter->tx_pool[pool].free_map[adapter->tx_pool[pool].
-						     producer_index] = index;
-			adapter->tx_pool[pool].producer_index =
-			    (adapter->tx_pool[pool].producer_index + 1) %
-			    adapter->req_tx_entries_per_subcrq;
+			tx_pool->free_map[tx_pool->producer_index] = index;
+			tx_pool->producer_index =
+				(tx_pool->producer_index + 1) %
+					tx_pool->num_buffers;
 		}
 		/* remove tx_comp scrq*/
 		next->tx_comp.first = 0;
 
-		if (atomic_sub_return(next->tx_comp.num_comps, &scrq->used) <=
+
+		if (atomic_sub_return(num_entries, &scrq->used) <=
 		    (adapter->req_tx_entries_per_subcrq / 2) &&
 		    __netif_subqueue_stopped(adapter->netdev,
 					     scrq->pool_index)) {
-			netif_wake_subqueue(adapter->netdev, scrq->pool_index);
-			netdev_info(adapter->netdev, "Started queue %d\n",
-				    scrq->pool_index);
+			rcu_read_lock();
+			if (adapter->tx_queues_active) {
+				netif_wake_subqueue(adapter->netdev,
+						    scrq->pool_index);
+				netdev_dbg(adapter->netdev,
+					   "Started queue %d\n",
+					   scrq->pool_index);
+			}
+			rcu_read_unlock();
 		}
 	}
 
@@ -1996,6 +4323,9 @@ restart_loop:
 		goto restart_loop;
 	}
 
+	txq = netdev_get_tx_queue(adapter->netdev, scrq->pool_index);
+	netdev_tx_completed_queue(txq, num_packets, total_bytes);
+
 	return 0;
 }
 
@@ -2015,6 +4345,14 @@ static irqreturn_t ibmvnic_interrupt_rx(int irq, void *instance)
 	struct ibmvnic_sub_crq_queue *scrq = instance;
 	struct ibmvnic_adapter *adapter = scrq->adapter;
 
+	/* When booting a kdump kernel we can hit pending interrupts
+	 * prior to completing driver initialization.
+	 */
+	if (unlikely(adapter->state != VNIC_OPEN))
+		return IRQ_NONE;
+
+	adapter->rx_stats_buffers[scrq->scrq_num].interrupts++;
+
 	if (napi_schedule_prep(&adapter->napi[scrq->scrq_num])) {
 		disable_scrq_irq(adapter, scrq);
 		__napi_schedule(&adapter->napi[scrq->scrq_num]);
@@ -2031,6 +4369,8 @@ static int init_sub_crq_irqs(struct ibmvnic_adapter *adapter)
 	int rc = 0;
 
 	for (i = 0; i < adapter->req_tx_queues; i++) {
+		netdev_dbg(adapter->netdev, "Initializing tx_scrq[%d] irq\n",
+			   i);
 		scrq = adapter->tx_scrq[i];
 		scrq->irq = irq_create_mapping(NULL, scrq->hw_irq);
 
@@ -2040,18 +4380,22 @@ static int init_sub_crq_irqs(struct ibmvnic_adapter *adapter)
 			goto req_tx_irq_failed;
 		}
 
+		snprintf(scrq->name, sizeof(scrq->name), "ibmvnic-%x-tx%d",
+			 adapter->vdev->unit_address, i);
 		rc = request_irq(scrq->irq, ibmvnic_interrupt_tx,
-				 0, "ibmvnic_tx", scrq);
+				 0, scrq->name, scrq);
 
 		if (rc) {
 			dev_err(dev, "Couldn't register tx irq 0x%x. rc=%d\n",
 				scrq->irq, rc);
 			irq_dispose_mapping(scrq->irq);
-			goto req_rx_irq_failed;
+			goto req_tx_irq_failed;
 		}
 	}
 
 	for (i = 0; i < adapter->req_rx_queues; i++) {
+		netdev_dbg(adapter->netdev, "Initializing rx_scrq[%d] irq\n",
+			   i);
 		scrq = adapter->rx_scrq[i];
 		scrq->irq = irq_create_mapping(NULL, scrq->hw_irq);
 		if (!scrq->irq) {
@@ -2059,8 +4403,10 @@ static int init_sub_crq_irqs(struct ibmvnic_adapter *adapter)
 			dev_err(dev, "Error mapping irq\n");
 			goto req_rx_irq_failed;
 		}
+		snprintf(scrq->name, sizeof(scrq->name), "ibmvnic-%x-rx%d",
+			 adapter->vdev->unit_address, i);
 		rc = request_irq(scrq->irq, ibmvnic_interrupt_rx,
-				 0, "ibmvnic_rx", scrq);
+				 0, scrq->name, scrq);
 		if (rc) {
 			dev_err(dev, "Couldn't register rx irq 0x%x. rc=%d\n",
 				scrq->irq, rc);
@@ -2068,6 +4414,11 @@ static int init_sub_crq_irqs(struct ibmvnic_adapter *adapter)
 			goto req_rx_irq_failed;
 		}
 	}
+
+	cpus_read_lock();
+	ibmvnic_set_affinity(adapter);
+	cpus_read_unlock();
+
 	return rc;
 
 req_rx_irq_failed:
@@ -2079,9 +4430,9 @@ req_rx_irq_failed:
 req_tx_irq_failed:
 	for (j = 0; j < i; j++) {
 		free_irq(adapter->tx_scrq[j]->irq, adapter->tx_scrq[j]);
-		irq_dispose_mapping(adapter->rx_scrq[j]->irq);
+		irq_dispose_mapping(adapter->tx_scrq[j]->irq);
 	}
-	release_sub_crqs(adapter);
+	release_sub_crqs(adapter, 1);
 	return rc;
 }
 
@@ -2098,7 +4449,7 @@ static int init_sub_crqs(struct ibmvnic_adapter *adapter)
 
 	allqueues = kcalloc(total_queues, sizeof(*allqueues), GFP_KERNEL);
 	if (!allqueues)
-		return -1;
+		return -ENOMEM;
 
 	for (i = 0; i < total_queues; i++) {
 		allqueues[i] = init_sub_crq_queue(adapter);
@@ -2143,6 +4494,7 @@ static int init_sub_crqs(struct ibmvnic_adapter *adapter)
 	for (i = 0; i < adapter->req_tx_queues; i++) {
 		adapter->tx_scrq[i] = allqueues[i];
 		adapter->tx_scrq[i]->pool_index = i;
+		adapter->num_active_tx_scrqs++;
 	}
 
 	adapter->rx_scrq = kcalloc(adapter->req_rx_queues,
@@ -2153,6 +4505,7 @@ static int init_sub_crqs(struct ibmvnic_adapter *adapter)
 	for (i = 0; i < adapter->req_rx_queues; i++) {
 		adapter->rx_scrq[i] = allqueues[i + adapter->req_tx_queues];
 		adapter->rx_scrq[i]->scrq_num = i;
+		adapter->num_active_rx_scrqs++;
 	}
 
 	kfree(allqueues);
@@ -2163,79 +4516,134 @@ rx_failed:
 	adapter->tx_scrq = NULL;
 tx_failed:
 	for (i = 0; i < registered_queues; i++)
-		release_sub_crq_queue(adapter, allqueues[i]);
+		release_sub_crq_queue(adapter, allqueues[i], 1);
 	kfree(allqueues);
-	return -1;
+	return -ENOMEM;
 }
 
-static void ibmvnic_send_req_caps(struct ibmvnic_adapter *adapter, int retry)
+static void send_request_cap(struct ibmvnic_adapter *adapter, int retry)
 {
 	struct device *dev = &adapter->vdev->dev;
 	union ibmvnic_crq crq;
+	int max_entries;
+	int cap_reqs;
+
+	/* We send out 6 or 7 REQUEST_CAPABILITY CRQs below (depending on
+	 * the PROMISC flag). Initialize this count upfront. When the tasklet
+	 * receives a response to all of these, it will send the next protocol
+	 * message (QUERY_IP_OFFLOAD).
+	 */
+	if (!(adapter->netdev->flags & IFF_PROMISC) ||
+	    adapter->promisc_supported)
+		cap_reqs = 7;
+	else
+		cap_reqs = 6;
 
 	if (!retry) {
 		/* Sub-CRQ entries are 32 byte long */
 		int entries_page = 4 * PAGE_SIZE / (sizeof(u64) * 4);
 
+		atomic_set(&adapter->running_cap_crqs, cap_reqs);
+
 		if (adapter->min_tx_entries_per_subcrq > entries_page ||
 		    adapter->min_rx_add_entries_per_subcrq > entries_page) {
 			dev_err(dev, "Fatal, invalid entries per sub-crq\n");
 			return;
 		}
 
-		/* Get the minimum between the queried max and the entries
-		 * that fit in our PAGE_SIZE
-		 */
-		adapter->req_tx_entries_per_subcrq =
-		    adapter->max_tx_entries_per_subcrq > entries_page ?
-		    entries_page : adapter->max_tx_entries_per_subcrq;
-		adapter->req_rx_add_entries_per_subcrq =
-		    adapter->max_rx_add_entries_per_subcrq > entries_page ?
-		    entries_page : adapter->max_rx_add_entries_per_subcrq;
-
-		adapter->req_tx_queues = adapter->opt_tx_comp_sub_queues;
-		adapter->req_rx_queues = adapter->opt_rx_comp_queues;
-		adapter->req_rx_add_queues = adapter->max_rx_add_queues;
+		if (adapter->desired.mtu)
+			adapter->req_mtu = adapter->desired.mtu;
+		else
+			adapter->req_mtu = adapter->netdev->mtu + ETH_HLEN;
 
-		adapter->req_mtu = adapter->netdev->mtu + ETH_HLEN;
-	}
+		if (!adapter->desired.tx_entries)
+			adapter->desired.tx_entries =
+					adapter->max_tx_entries_per_subcrq;
+		if (!adapter->desired.rx_entries)
+			adapter->desired.rx_entries =
+					adapter->max_rx_add_entries_per_subcrq;
+
+		max_entries = IBMVNIC_LTB_SET_SIZE /
+			      (adapter->req_mtu + IBMVNIC_BUFFER_HLEN);
+
+		if ((adapter->req_mtu + IBMVNIC_BUFFER_HLEN) *
+			adapter->desired.tx_entries > IBMVNIC_LTB_SET_SIZE) {
+			adapter->desired.tx_entries = max_entries;
+		}
+
+		if ((adapter->req_mtu + IBMVNIC_BUFFER_HLEN) *
+			adapter->desired.rx_entries > IBMVNIC_LTB_SET_SIZE) {
+			adapter->desired.rx_entries = max_entries;
+		}
+
+		if (adapter->desired.tx_entries)
+			adapter->req_tx_entries_per_subcrq =
+					adapter->desired.tx_entries;
+		else
+			adapter->req_tx_entries_per_subcrq =
+					adapter->max_tx_entries_per_subcrq;
+
+		if (adapter->desired.rx_entries)
+			adapter->req_rx_add_entries_per_subcrq =
+					adapter->desired.rx_entries;
+		else
+			adapter->req_rx_add_entries_per_subcrq =
+					adapter->max_rx_add_entries_per_subcrq;
+
+		if (adapter->desired.tx_queues)
+			adapter->req_tx_queues =
+					adapter->desired.tx_queues;
+		else
+			adapter->req_tx_queues =
+					adapter->opt_tx_comp_sub_queues;
+
+		if (adapter->desired.rx_queues)
+			adapter->req_rx_queues =
+					adapter->desired.rx_queues;
+		else
+			adapter->req_rx_queues =
+					adapter->opt_rx_comp_queues;
 
+		adapter->req_rx_add_queues = adapter->max_rx_add_queues;
+	} else {
+		atomic_add(cap_reqs, &adapter->running_cap_crqs);
+	}
 	memset(&crq, 0, sizeof(crq));
 	crq.request_capability.first = IBMVNIC_CRQ_CMD;
 	crq.request_capability.cmd = REQUEST_CAPABILITY;
 
 	crq.request_capability.capability = cpu_to_be16(REQ_TX_QUEUES);
 	crq.request_capability.number = cpu_to_be64(adapter->req_tx_queues);
-	atomic_inc(&adapter->running_cap_crqs);
+	cap_reqs--;
 	ibmvnic_send_crq(adapter, &crq);
 
 	crq.request_capability.capability = cpu_to_be16(REQ_RX_QUEUES);
 	crq.request_capability.number = cpu_to_be64(adapter->req_rx_queues);
-	atomic_inc(&adapter->running_cap_crqs);
+	cap_reqs--;
 	ibmvnic_send_crq(adapter, &crq);
 
 	crq.request_capability.capability = cpu_to_be16(REQ_RX_ADD_QUEUES);
 	crq.request_capability.number = cpu_to_be64(adapter->req_rx_add_queues);
-	atomic_inc(&adapter->running_cap_crqs);
+	cap_reqs--;
 	ibmvnic_send_crq(adapter, &crq);
 
 	crq.request_capability.capability =
 	    cpu_to_be16(REQ_TX_ENTRIES_PER_SUBCRQ);
 	crq.request_capability.number =
 	    cpu_to_be64(adapter->req_tx_entries_per_subcrq);
-	atomic_inc(&adapter->running_cap_crqs);
+	cap_reqs--;
 	ibmvnic_send_crq(adapter, &crq);
 
 	crq.request_capability.capability =
 	    cpu_to_be16(REQ_RX_ADD_ENTRIES_PER_SUBCRQ);
 	crq.request_capability.number =
 	    cpu_to_be64(adapter->req_rx_add_entries_per_subcrq);
-	atomic_inc(&adapter->running_cap_crqs);
+	cap_reqs--;
 	ibmvnic_send_crq(adapter, &crq);
 
 	crq.request_capability.capability = cpu_to_be16(REQ_MTU);
 	crq.request_capability.number = cpu_to_be64(adapter->req_mtu);
-	atomic_inc(&adapter->running_cap_crqs);
+	cap_reqs--;
 	ibmvnic_send_crq(adapter, &crq);
 
 	if (adapter->netdev->flags & IFF_PROMISC) {
@@ -2243,28 +4651,37 @@ static void ibmvnic_send_req_caps(struct ibmvnic_adapter *adapter, int retry)
 			crq.request_capability.capability =
 			    cpu_to_be16(PROMISC_REQUESTED);
 			crq.request_capability.number = cpu_to_be64(1);
-			atomic_inc(&adapter->running_cap_crqs);
+			cap_reqs--;
 			ibmvnic_send_crq(adapter, &crq);
 		}
 	} else {
 		crq.request_capability.capability =
 		    cpu_to_be16(PROMISC_REQUESTED);
 		crq.request_capability.number = cpu_to_be64(0);
-		atomic_inc(&adapter->running_cap_crqs);
+		cap_reqs--;
 		ibmvnic_send_crq(adapter, &crq);
 	}
+
+	/* Keep at end to catch any discrepancy between expected and actual
+	 * CRQs sent.
+	 */
+	WARN_ON(cap_reqs != 0);
 }
 
 static int pending_scrq(struct ibmvnic_adapter *adapter,
 			struct ibmvnic_sub_crq_queue *scrq)
 {
 	union sub_crq *entry = &scrq->msgs[scrq->cur];
+	int rc;
 
-	if (entry->generic.first & IBMVNIC_CRQ_CMD_RSP ||
-	    adapter->state == VNIC_CLOSING)
-		return 1;
-	else
-		return 0;
+	rc = !!(entry->generic.first & IBMVNIC_CRQ_CMD_RSP);
+
+	/* Ensure that the SCRQ valid flag is loaded prior to loading the
+	 * contents of the SCRQ descriptor
+	 */
+	dma_rmb();
+
+	return rc;
 }
 
 static union sub_crq *ibmvnic_next_scrq(struct ibmvnic_adapter *adapter,
@@ -2283,6 +4700,11 @@ static union sub_crq *ibmvnic_next_scrq(struct ibmvnic_adapter *adapter,
 	}
 	spin_unlock_irqrestore(&scrq->lock, flags);
 
+	/* Ensure that the SCRQ valid flag is loaded prior to loading the
+	 * contents of the SCRQ descriptor
+	 */
+	dma_rmb();
+
 	return entry;
 }
 
@@ -2302,39 +4724,23 @@ static union ibmvnic_crq *ibmvnic_next_crq(struct ibmvnic_adapter *adapter)
 	return crq;
 }
 
-static int send_subcrq(struct ibmvnic_adapter *adapter, u64 remote_handle,
-		       union sub_crq *sub_crq)
+static void print_subcrq_error(struct device *dev, int rc, const char *func)
 {
-	unsigned int ua = adapter->vdev->unit_address;
-	struct device *dev = &adapter->vdev->dev;
-	u64 *u64_crq = (u64 *)sub_crq;
-	int rc;
-
-	netdev_dbg(adapter->netdev,
-		   "Sending sCRQ %016lx: %016lx %016lx %016lx %016lx\n",
-		   (unsigned long int)cpu_to_be64(remote_handle),
-		   (unsigned long int)cpu_to_be64(u64_crq[0]),
-		   (unsigned long int)cpu_to_be64(u64_crq[1]),
-		   (unsigned long int)cpu_to_be64(u64_crq[2]),
-		   (unsigned long int)cpu_to_be64(u64_crq[3]));
-
-	/* Make sure the hypervisor sees the complete request */
-	mb();
-
-	rc = plpar_hcall_norets(H_SEND_SUB_CRQ, ua,
-				cpu_to_be64(remote_handle),
-				cpu_to_be64(u64_crq[0]),
-				cpu_to_be64(u64_crq[1]),
-				cpu_to_be64(u64_crq[2]),
-				cpu_to_be64(u64_crq[3]));
-
-	if (rc) {
-		if (rc == H_CLOSED)
-			dev_warn(dev, "CRQ Queue closed\n");
-		dev_err(dev, "Send error (rc=%d)\n", rc);
+	switch (rc) {
+	case H_PARAMETER:
+		dev_warn_ratelimited(dev,
+				     "%s failed: Send request is malformed or adapter failover pending. (rc=%d)\n",
+				     func, rc);
+		break;
+	case H_CLOSED:
+		dev_warn_ratelimited(dev,
+				     "%s failed: Backing queue closed. Adapter is down or failover pending. (rc=%d)\n",
+				     func, rc);
+		break;
+	default:
+		dev_err_ratelimited(dev, "%s failed: (rc=%d)\n", func, rc);
+		break;
 	}
-
-	return rc;
 }
 
 static int send_subcrq_indirect(struct ibmvnic_adapter *adapter,
@@ -2345,16 +4751,13 @@ static int send_subcrq_indirect(struct ibmvnic_adapter *adapter,
 	int rc;
 
 	/* Make sure the hypervisor sees the complete request */
-	mb();
+	dma_wmb();
 	rc = plpar_hcall_norets(H_SEND_SUB_CRQ_INDIRECT, ua,
 				cpu_to_be64(remote_handle),
 				ioba, num_entries);
 
-	if (rc) {
-		if (rc == H_CLOSED)
-			dev_warn(dev, "CRQ Queue closed\n");
-		dev_err(dev, "Send (indirect) error (rc=%d)\n", rc);
-	}
+	if (rc)
+		print_subcrq_error(dev, rc, __func__);
 
 	return rc;
 }
@@ -2368,19 +4771,28 @@ static int ibmvnic_send_crq(struct ibmvnic_adapter *adapter,
 	int rc;
 
 	netdev_dbg(adapter->netdev, "Sending CRQ: %016lx %016lx\n",
-		   (unsigned long int)cpu_to_be64(u64_crq[0]),
-		   (unsigned long int)cpu_to_be64(u64_crq[1]));
+		   (unsigned long)cpu_to_be64(u64_crq[0]),
+		   (unsigned long)cpu_to_be64(u64_crq[1]));
+
+	if (!adapter->crq.active &&
+	    crq->generic.first != IBMVNIC_CRQ_INIT_CMD) {
+		dev_warn(dev, "Invalid request detected while CRQ is inactive, possible device state change during reset\n");
+		return -EINVAL;
+	}
 
 	/* Make sure the hypervisor sees the complete request */
-	mb();
+	dma_wmb();
 
 	rc = plpar_hcall_norets(H_SEND_CRQ, ua,
 				cpu_to_be64(u64_crq[0]),
 				cpu_to_be64(u64_crq[1]));
 
 	if (rc) {
-		if (rc == H_CLOSED)
+		if (rc == H_CLOSED) {
 			dev_warn(dev, "CRQ Queue closed\n");
+			/* do not reset, report the fail, wait for passive init from server */
+		}
+
 		dev_warn(dev, "Send error (rc=%d)\n", rc);
 	}
 
@@ -2389,47 +4801,128 @@ static int ibmvnic_send_crq(struct ibmvnic_adapter *adapter,
 
 static int ibmvnic_send_crq_init(struct ibmvnic_adapter *adapter)
 {
+	struct device *dev = &adapter->vdev->dev;
 	union ibmvnic_crq crq;
+	int retries = 100;
+	int rc;
 
 	memset(&crq, 0, sizeof(crq));
 	crq.generic.first = IBMVNIC_CRQ_INIT_CMD;
 	crq.generic.cmd = IBMVNIC_CRQ_INIT;
 	netdev_dbg(adapter->netdev, "Sending CRQ init\n");
 
-	return ibmvnic_send_crq(adapter, &crq);
+	do {
+		rc = ibmvnic_send_crq(adapter, &crq);
+		if (rc != H_CLOSED)
+			break;
+		retries--;
+		msleep(50);
+
+	} while (retries > 0);
+
+	if (rc) {
+		dev_err(dev, "Failed to send init request, rc = %d\n", rc);
+		return rc;
+	}
+
+	return 0;
 }
 
-static int send_version_xchg(struct ibmvnic_adapter *adapter)
+struct vnic_login_client_data {
+	u8	type;
+	__be16	len;
+	char	name[];
+} __packed;
+
+static int vnic_client_data_len(struct ibmvnic_adapter *adapter)
 {
-	union ibmvnic_crq crq;
+	int len;
 
-	memset(&crq, 0, sizeof(crq));
-	crq.version_exchange.first = IBMVNIC_CRQ_CMD;
-	crq.version_exchange.cmd = VERSION_EXCHANGE;
-	crq.version_exchange.version = cpu_to_be16(ibmvnic_version);
+	/* Calculate the amount of buffer space needed for the
+	 * vnic client data in the login buffer. There are four entries,
+	 * OS name, LPAR name, device name, and a null last entry.
+	 */
+	len = 4 * sizeof(struct vnic_login_client_data);
+	len += 6; /* "Linux" plus NULL */
+	len += strlen(utsname()->nodename) + 1;
+	len += strlen(adapter->netdev->name) + 1;
 
-	return ibmvnic_send_crq(adapter, &crq);
+	return len;
 }
 
-static void send_login(struct ibmvnic_adapter *adapter)
+static void vnic_add_client_data(struct ibmvnic_adapter *adapter,
+				 struct vnic_login_client_data *vlcd)
+{
+	const char *os_name = "Linux";
+	int len;
+
+	/* Type 1 - LPAR OS */
+	vlcd->type = 1;
+	len = strlen(os_name) + 1;
+	vlcd->len = cpu_to_be16(len);
+	strscpy(vlcd->name, os_name, len);
+	vlcd = (struct vnic_login_client_data *)(vlcd->name + len);
+
+	/* Type 2 - LPAR name */
+	vlcd->type = 2;
+	len = strlen(utsname()->nodename) + 1;
+	vlcd->len = cpu_to_be16(len);
+	strscpy(vlcd->name, utsname()->nodename, len);
+	vlcd = (struct vnic_login_client_data *)(vlcd->name + len);
+
+	/* Type 3 - device name */
+	vlcd->type = 3;
+	len = strlen(adapter->netdev->name) + 1;
+	vlcd->len = cpu_to_be16(len);
+	strscpy(vlcd->name, adapter->netdev->name, len);
+}
+
+static void ibmvnic_print_hex_dump(struct net_device *dev, void *buf,
+				   size_t len)
+{
+	unsigned char hex_str[16 * 3];
+
+	for (size_t i = 0; i < len; i += 16) {
+		hex_dump_to_buffer((unsigned char *)buf + i, len - i, 16, 8,
+				   hex_str, sizeof(hex_str), false);
+		netdev_dbg(dev, "%s\n", hex_str);
+	}
+}
+
+static int send_login(struct ibmvnic_adapter *adapter)
 {
 	struct ibmvnic_login_rsp_buffer *login_rsp_buffer;
 	struct ibmvnic_login_buffer *login_buffer;
 	struct device *dev = &adapter->vdev->dev;
+	struct vnic_login_client_data *vlcd;
 	dma_addr_t rsp_buffer_token;
 	dma_addr_t buffer_token;
 	size_t rsp_buffer_size;
 	union ibmvnic_crq crq;
+	int client_data_len;
 	size_t buffer_size;
 	__be64 *tx_list_p;
 	__be64 *rx_list_p;
+	int rc;
 	int i;
 
+	if (!adapter->tx_scrq || !adapter->rx_scrq) {
+		netdev_err(adapter->netdev,
+			   "RX or TX queues are not allocated, device login failed\n");
+		return -ENOMEM;
+	}
+
+	release_login_buffer(adapter);
+	release_login_rsp_buffer(adapter);
+
+	client_data_len = vnic_client_data_len(adapter);
+
 	buffer_size =
 	    sizeof(struct ibmvnic_login_buffer) +
-	    sizeof(u64) * (adapter->req_tx_queues + adapter->req_rx_queues);
+	    sizeof(u64) * (adapter->req_tx_queues + adapter->req_rx_queues) +
+	    client_data_len;
 
-	login_buffer = kmalloc(buffer_size, GFP_ATOMIC);
+	login_buffer = kzalloc(buffer_size, GFP_ATOMIC);
 	if (!login_buffer)
 		goto buf_alloc_failed;
 
@@ -2484,45 +4977,64 @@ static void send_login(struct ibmvnic_adapter *adapter)
 
 	for (i = 0; i < adapter->req_tx_queues; i++) {
 		if (adapter->tx_scrq[i]) {
-			tx_list_p[i] = cpu_to_be64(adapter->tx_scrq[i]->
-						   crq_num);
+			tx_list_p[i] =
+				cpu_to_be64(adapter->tx_scrq[i]->crq_num);
 		}
 	}
 
 	for (i = 0; i < adapter->req_rx_queues; i++) {
 		if (adapter->rx_scrq[i]) {
-			rx_list_p[i] = cpu_to_be64(adapter->rx_scrq[i]->
-						   crq_num);
+			rx_list_p[i] =
+				cpu_to_be64(adapter->rx_scrq[i]->crq_num);
 		}
 	}
 
+	/* Insert vNIC login client data */
+	vlcd = (struct vnic_login_client_data *)
+		((char *)rx_list_p + (sizeof(u64) * adapter->req_rx_queues));
+	login_buffer->client_data_offset =
+			cpu_to_be32((char *)vlcd - (char *)login_buffer);
+	login_buffer->client_data_len = cpu_to_be32(client_data_len);
+
+	vnic_add_client_data(adapter, vlcd);
+
 	netdev_dbg(adapter->netdev, "Login Buffer:\n");
-	for (i = 0; i < (adapter->login_buf_sz - 1) / 8 + 1; i++) {
-		netdev_dbg(adapter->netdev, "%016lx\n",
-			   ((unsigned long int *)(adapter->login_buf))[i]);
-	}
+	ibmvnic_print_hex_dump(adapter->netdev, adapter->login_buf,
+			       adapter->login_buf_sz);
 
 	memset(&crq, 0, sizeof(crq));
 	crq.login.first = IBMVNIC_CRQ_CMD;
 	crq.login.cmd = LOGIN;
 	crq.login.ioba = cpu_to_be32(buffer_token);
 	crq.login.len = cpu_to_be32(buffer_size);
-	ibmvnic_send_crq(adapter, &crq);
 
-	return;
+	adapter->login_pending = true;
+	rc = ibmvnic_send_crq(adapter, &crq);
+	if (rc) {
+		adapter->login_pending = false;
+		netdev_err(adapter->netdev, "Failed to send login, rc=%d\n", rc);
+		goto buf_send_failed;
+	}
+
+	return 0;
 
+buf_send_failed:
+	dma_unmap_single(dev, rsp_buffer_token, rsp_buffer_size,
+			 DMA_FROM_DEVICE);
 buf_rsp_map_failed:
 	kfree(login_rsp_buffer);
+	adapter->login_rsp_buf = NULL;
 buf_rsp_alloc_failed:
 	dma_unmap_single(dev, buffer_token, buffer_size, DMA_TO_DEVICE);
 buf_map_failed:
 	kfree(login_buffer);
+	adapter->login_buf = NULL;
 buf_alloc_failed:
-	return;
+	return -ENOMEM;
 }
 
-static void send_request_map(struct ibmvnic_adapter *adapter, dma_addr_t addr,
-			     u32 len, u8 map_id)
+static int send_request_map(struct ibmvnic_adapter *adapter, dma_addr_t addr,
+			    u32 len, u8 map_id)
 {
 	union ibmvnic_crq crq;
 
@@ -2532,10 +5044,10 @@ static void send_request_map(struct ibmvnic_adapter *adapter, dma_addr_t addr,
 	crq.request_map.map_id = map_id;
 	crq.request_map.ioba = cpu_to_be32(addr);
 	crq.request_map.len = cpu_to_be32(len);
-	ibmvnic_send_crq(adapter, &crq);
+	return ibmvnic_send_crq(adapter, &crq);
 }
 
-static void send_request_unmap(struct ibmvnic_adapter *adapter, u8 map_id)
+static int send_request_unmap(struct ibmvnic_adapter *adapter, u8 map_id)
 {
 	union ibmvnic_crq crq;
 
@@ -2543,10 +5055,10 @@ static void send_request_unmap(struct ibmvnic_adapter *adapter, u8 map_id)
 	crq.request_unmap.first = IBMVNIC_CRQ_CMD;
 	crq.request_unmap.cmd = REQUEST_UNMAP;
 	crq.request_unmap.map_id = map_id;
-	ibmvnic_send_crq(adapter, &crq);
+	return ibmvnic_send_crq(adapter, &crq);
 }
 
-static void send_map_query(struct ibmvnic_adapter *adapter)
+static void send_query_map(struct ibmvnic_adapter *adapter)
 {
 	union ibmvnic_crq crq;
 
@@ -2557,209 +5069,234 @@ static void send_map_query(struct ibmvnic_adapter *adapter)
 }
 
 /* Send a series of CRQs requesting various capabilities of the VNIC server */
-static void send_cap_queries(struct ibmvnic_adapter *adapter)
+static void send_query_cap(struct ibmvnic_adapter *adapter)
 {
 	union ibmvnic_crq crq;
+	int cap_reqs;
+
+	/* We send out 25 QUERY_CAPABILITY CRQs below.  Initialize this count
+	 * upfront. When the tasklet receives a response to all of these, it
+	 * can send out the next protocol messaage (REQUEST_CAPABILITY).
+	 */
+	cap_reqs = 25;
+
+	atomic_set(&adapter->running_cap_crqs, cap_reqs);
 
-	atomic_set(&adapter->running_cap_crqs, 0);
 	memset(&crq, 0, sizeof(crq));
 	crq.query_capability.first = IBMVNIC_CRQ_CMD;
 	crq.query_capability.cmd = QUERY_CAPABILITY;
 
 	crq.query_capability.capability = cpu_to_be16(MIN_TX_QUEUES);
-	atomic_inc(&adapter->running_cap_crqs);
 	ibmvnic_send_crq(adapter, &crq);
+	cap_reqs--;
 
 	crq.query_capability.capability = cpu_to_be16(MIN_RX_QUEUES);
-	atomic_inc(&adapter->running_cap_crqs);
 	ibmvnic_send_crq(adapter, &crq);
+	cap_reqs--;
 
 	crq.query_capability.capability = cpu_to_be16(MIN_RX_ADD_QUEUES);
-	atomic_inc(&adapter->running_cap_crqs);
 	ibmvnic_send_crq(adapter, &crq);
+	cap_reqs--;
 
 	crq.query_capability.capability = cpu_to_be16(MAX_TX_QUEUES);
-	atomic_inc(&adapter->running_cap_crqs);
 	ibmvnic_send_crq(adapter, &crq);
+	cap_reqs--;
 
 	crq.query_capability.capability = cpu_to_be16(MAX_RX_QUEUES);
-	atomic_inc(&adapter->running_cap_crqs);
 	ibmvnic_send_crq(adapter, &crq);
+	cap_reqs--;
 
 	crq.query_capability.capability = cpu_to_be16(MAX_RX_ADD_QUEUES);
-	atomic_inc(&adapter->running_cap_crqs);
 	ibmvnic_send_crq(adapter, &crq);
+	cap_reqs--;
 
 	crq.query_capability.capability =
 	    cpu_to_be16(MIN_TX_ENTRIES_PER_SUBCRQ);
-	atomic_inc(&adapter->running_cap_crqs);
 	ibmvnic_send_crq(adapter, &crq);
+	cap_reqs--;
 
 	crq.query_capability.capability =
 	    cpu_to_be16(MIN_RX_ADD_ENTRIES_PER_SUBCRQ);
-	atomic_inc(&adapter->running_cap_crqs);
 	ibmvnic_send_crq(adapter, &crq);
+	cap_reqs--;
 
 	crq.query_capability.capability =
 	    cpu_to_be16(MAX_TX_ENTRIES_PER_SUBCRQ);
-	atomic_inc(&adapter->running_cap_crqs);
 	ibmvnic_send_crq(adapter, &crq);
+	cap_reqs--;
 
 	crq.query_capability.capability =
 	    cpu_to_be16(MAX_RX_ADD_ENTRIES_PER_SUBCRQ);
-	atomic_inc(&adapter->running_cap_crqs);
 	ibmvnic_send_crq(adapter, &crq);
+	cap_reqs--;
 
 	crq.query_capability.capability = cpu_to_be16(TCP_IP_OFFLOAD);
-	atomic_inc(&adapter->running_cap_crqs);
 	ibmvnic_send_crq(adapter, &crq);
+	cap_reqs--;
 
 	crq.query_capability.capability = cpu_to_be16(PROMISC_SUPPORTED);
-	atomic_inc(&adapter->running_cap_crqs);
 	ibmvnic_send_crq(adapter, &crq);
+	cap_reqs--;
 
 	crq.query_capability.capability = cpu_to_be16(MIN_MTU);
-	atomic_inc(&adapter->running_cap_crqs);
 	ibmvnic_send_crq(adapter, &crq);
+	cap_reqs--;
 
 	crq.query_capability.capability = cpu_to_be16(MAX_MTU);
-	atomic_inc(&adapter->running_cap_crqs);
 	ibmvnic_send_crq(adapter, &crq);
+	cap_reqs--;
 
 	crq.query_capability.capability = cpu_to_be16(MAX_MULTICAST_FILTERS);
-	atomic_inc(&adapter->running_cap_crqs);
 	ibmvnic_send_crq(adapter, &crq);
+	cap_reqs--;
 
 	crq.query_capability.capability = cpu_to_be16(VLAN_HEADER_INSERTION);
-	atomic_inc(&adapter->running_cap_crqs);
 	ibmvnic_send_crq(adapter, &crq);
+	cap_reqs--;
 
 	crq.query_capability.capability = cpu_to_be16(RX_VLAN_HEADER_INSERTION);
-	atomic_inc(&adapter->running_cap_crqs);
 	ibmvnic_send_crq(adapter, &crq);
+	cap_reqs--;
 
 	crq.query_capability.capability = cpu_to_be16(MAX_TX_SG_ENTRIES);
-	atomic_inc(&adapter->running_cap_crqs);
 	ibmvnic_send_crq(adapter, &crq);
+	cap_reqs--;
 
 	crq.query_capability.capability = cpu_to_be16(RX_SG_SUPPORTED);
-	atomic_inc(&adapter->running_cap_crqs);
 	ibmvnic_send_crq(adapter, &crq);
+	cap_reqs--;
 
 	crq.query_capability.capability = cpu_to_be16(OPT_TX_COMP_SUB_QUEUES);
-	atomic_inc(&adapter->running_cap_crqs);
 	ibmvnic_send_crq(adapter, &crq);
+	cap_reqs--;
 
 	crq.query_capability.capability = cpu_to_be16(OPT_RX_COMP_QUEUES);
-	atomic_inc(&adapter->running_cap_crqs);
 	ibmvnic_send_crq(adapter, &crq);
+	cap_reqs--;
 
 	crq.query_capability.capability =
 			cpu_to_be16(OPT_RX_BUFADD_Q_PER_RX_COMP_Q);
-	atomic_inc(&adapter->running_cap_crqs);
 	ibmvnic_send_crq(adapter, &crq);
+	cap_reqs--;
 
 	crq.query_capability.capability =
 			cpu_to_be16(OPT_TX_ENTRIES_PER_SUBCRQ);
-	atomic_inc(&adapter->running_cap_crqs);
 	ibmvnic_send_crq(adapter, &crq);
+	cap_reqs--;
 
 	crq.query_capability.capability =
 			cpu_to_be16(OPT_RXBA_ENTRIES_PER_SUBCRQ);
-	atomic_inc(&adapter->running_cap_crqs);
 	ibmvnic_send_crq(adapter, &crq);
+	cap_reqs--;
 
 	crq.query_capability.capability = cpu_to_be16(TX_RX_DESC_REQ);
-	atomic_inc(&adapter->running_cap_crqs);
+
 	ibmvnic_send_crq(adapter, &crq);
+	cap_reqs--;
+
+	/* Keep at end to catch any discrepancy between expected and actual
+	 * CRQs sent.
+	 */
+	WARN_ON(cap_reqs != 0);
 }
 
-static void handle_query_ip_offload_rsp(struct ibmvnic_adapter *adapter)
+static void send_query_ip_offload(struct ibmvnic_adapter *adapter)
 {
+	int buf_sz = sizeof(struct ibmvnic_query_ip_offload_buffer);
 	struct device *dev = &adapter->vdev->dev;
-	struct ibmvnic_query_ip_offload_buffer *buf = &adapter->ip_offload_buf;
 	union ibmvnic_crq crq;
-	int i;
 
-	dma_unmap_single(dev, adapter->ip_offload_tok,
-			 sizeof(adapter->ip_offload_buf), DMA_FROM_DEVICE);
+	adapter->ip_offload_tok =
+		dma_map_single(dev,
+			       &adapter->ip_offload_buf,
+			       buf_sz,
+			       DMA_FROM_DEVICE);
 
-	netdev_dbg(adapter->netdev, "Query IP Offload Buffer:\n");
-	for (i = 0; i < (sizeof(adapter->ip_offload_buf) - 1) / 8 + 1; i++)
-		netdev_dbg(adapter->netdev, "%016lx\n",
-			   ((unsigned long int *)(buf))[i]);
+	if (dma_mapping_error(dev, adapter->ip_offload_tok)) {
+		if (!firmware_has_feature(FW_FEATURE_CMO))
+			dev_err(dev, "Couldn't map offload buffer\n");
+		return;
+	}
 
-	netdev_dbg(adapter->netdev, "ipv4_chksum = %d\n", buf->ipv4_chksum);
-	netdev_dbg(adapter->netdev, "ipv6_chksum = %d\n", buf->ipv6_chksum);
-	netdev_dbg(adapter->netdev, "tcp_ipv4_chksum = %d\n",
-		   buf->tcp_ipv4_chksum);
-	netdev_dbg(adapter->netdev, "tcp_ipv6_chksum = %d\n",
-		   buf->tcp_ipv6_chksum);
-	netdev_dbg(adapter->netdev, "udp_ipv4_chksum = %d\n",
-		   buf->udp_ipv4_chksum);
-	netdev_dbg(adapter->netdev, "udp_ipv6_chksum = %d\n",
-		   buf->udp_ipv6_chksum);
-	netdev_dbg(adapter->netdev, "large_tx_ipv4 = %d\n",
-		   buf->large_tx_ipv4);
-	netdev_dbg(adapter->netdev, "large_tx_ipv6 = %d\n",
-		   buf->large_tx_ipv6);
-	netdev_dbg(adapter->netdev, "large_rx_ipv4 = %d\n",
-		   buf->large_rx_ipv4);
-	netdev_dbg(adapter->netdev, "large_rx_ipv6 = %d\n",
-		   buf->large_rx_ipv6);
-	netdev_dbg(adapter->netdev, "max_ipv4_hdr_sz = %d\n",
-		   buf->max_ipv4_header_size);
-	netdev_dbg(adapter->netdev, "max_ipv6_hdr_sz = %d\n",
-		   buf->max_ipv6_header_size);
-	netdev_dbg(adapter->netdev, "max_tcp_hdr_size = %d\n",
-		   buf->max_tcp_header_size);
-	netdev_dbg(adapter->netdev, "max_udp_hdr_size = %d\n",
-		   buf->max_udp_header_size);
-	netdev_dbg(adapter->netdev, "max_large_tx_size = %d\n",
-		   buf->max_large_tx_size);
-	netdev_dbg(adapter->netdev, "max_large_rx_size = %d\n",
-		   buf->max_large_rx_size);
-	netdev_dbg(adapter->netdev, "ipv6_ext_hdr = %d\n",
-		   buf->ipv6_extension_header);
-	netdev_dbg(adapter->netdev, "tcp_pseudosum_req = %d\n",
-		   buf->tcp_pseudosum_req);
-	netdev_dbg(adapter->netdev, "num_ipv6_ext_hd = %d\n",
-		   buf->num_ipv6_ext_headers);
-	netdev_dbg(adapter->netdev, "off_ipv6_ext_hd = %d\n",
-		   buf->off_ipv6_ext_headers);
+	memset(&crq, 0, sizeof(crq));
+	crq.query_ip_offload.first = IBMVNIC_CRQ_CMD;
+	crq.query_ip_offload.cmd = QUERY_IP_OFFLOAD;
+	crq.query_ip_offload.len = cpu_to_be32(buf_sz);
+	crq.query_ip_offload.ioba =
+	    cpu_to_be32(adapter->ip_offload_tok);
+
+	ibmvnic_send_crq(adapter, &crq);
+}
+
+static void send_control_ip_offload(struct ibmvnic_adapter *adapter)
+{
+	struct ibmvnic_control_ip_offload_buffer *ctrl_buf = &adapter->ip_offload_ctrl;
+	struct ibmvnic_query_ip_offload_buffer *buf = &adapter->ip_offload_buf;
+	struct device *dev = &adapter->vdev->dev;
+	netdev_features_t old_hw_features = 0;
+	union ibmvnic_crq crq;
 
 	adapter->ip_offload_ctrl_tok =
-	    dma_map_single(dev, &adapter->ip_offload_ctrl,
-			   sizeof(adapter->ip_offload_ctrl), DMA_TO_DEVICE);
+		dma_map_single(dev,
+			       ctrl_buf,
+			       sizeof(adapter->ip_offload_ctrl),
+			       DMA_TO_DEVICE);
 
 	if (dma_mapping_error(dev, adapter->ip_offload_ctrl_tok)) {
 		dev_err(dev, "Couldn't map ip offload control buffer\n");
 		return;
 	}
 
-	adapter->ip_offload_ctrl.version = cpu_to_be32(INITIAL_VERSION_IOB);
-	adapter->ip_offload_ctrl.tcp_ipv4_chksum = buf->tcp_ipv4_chksum;
-	adapter->ip_offload_ctrl.udp_ipv4_chksum = buf->udp_ipv4_chksum;
-	adapter->ip_offload_ctrl.tcp_ipv6_chksum = buf->tcp_ipv6_chksum;
-	adapter->ip_offload_ctrl.udp_ipv6_chksum = buf->udp_ipv6_chksum;
-
-	/* large_tx/rx disabled for now, additional features needed */
-	adapter->ip_offload_ctrl.large_tx_ipv4 = 0;
-	adapter->ip_offload_ctrl.large_tx_ipv6 = 0;
-	adapter->ip_offload_ctrl.large_rx_ipv4 = 0;
-	adapter->ip_offload_ctrl.large_rx_ipv6 = 0;
+	ctrl_buf->len = cpu_to_be32(sizeof(adapter->ip_offload_ctrl));
+	ctrl_buf->version = cpu_to_be32(INITIAL_VERSION_IOB);
+	ctrl_buf->ipv4_chksum = buf->ipv4_chksum;
+	ctrl_buf->ipv6_chksum = buf->ipv6_chksum;
+	ctrl_buf->tcp_ipv4_chksum = buf->tcp_ipv4_chksum;
+	ctrl_buf->udp_ipv4_chksum = buf->udp_ipv4_chksum;
+	ctrl_buf->tcp_ipv6_chksum = buf->tcp_ipv6_chksum;
+	ctrl_buf->udp_ipv6_chksum = buf->udp_ipv6_chksum;
+	ctrl_buf->large_tx_ipv4 = buf->large_tx_ipv4;
+	ctrl_buf->large_tx_ipv6 = buf->large_tx_ipv6;
+
+	/* large_rx disabled for now, additional features needed */
+	ctrl_buf->large_rx_ipv4 = 0;
+	ctrl_buf->large_rx_ipv6 = 0;
+
+	if (adapter->state != VNIC_PROBING) {
+		old_hw_features = adapter->netdev->hw_features;
+		adapter->netdev->hw_features = 0;
+	}
 
-	adapter->netdev->features = NETIF_F_GSO;
+	adapter->netdev->hw_features = NETIF_F_SG | NETIF_F_GSO | NETIF_F_GRO;
 
 	if (buf->tcp_ipv4_chksum || buf->udp_ipv4_chksum)
-		adapter->netdev->features |= NETIF_F_IP_CSUM;
+		adapter->netdev->hw_features |= NETIF_F_IP_CSUM;
 
 	if (buf->tcp_ipv6_chksum || buf->udp_ipv6_chksum)
-		adapter->netdev->features |= NETIF_F_IPV6_CSUM;
+		adapter->netdev->hw_features |= NETIF_F_IPV6_CSUM;
 
 	if ((adapter->netdev->features &
 	    (NETIF_F_IP_CSUM | NETIF_F_IPV6_CSUM)))
-		adapter->netdev->features |= NETIF_F_RXCSUM;
+		adapter->netdev->hw_features |= NETIF_F_RXCSUM;
+
+	if (buf->large_tx_ipv4)
+		adapter->netdev->hw_features |= NETIF_F_TSO;
+	if (buf->large_tx_ipv6)
+		adapter->netdev->hw_features |= NETIF_F_TSO6;
+
+	if (adapter->state == VNIC_PROBING) {
+		adapter->netdev->features |= adapter->netdev->hw_features;
+	} else if (old_hw_features != adapter->netdev->hw_features) {
+		netdev_features_t tmp = 0;
+
+		/* disable features no longer supported */
+		adapter->netdev->features &= adapter->netdev->hw_features;
+		/* turn on features now supported if previously enabled */
+		tmp = (old_hw_features ^ adapter->netdev->hw_features) &
+			adapter->netdev->hw_features;
+		adapter->netdev->features |=
+				tmp & adapter->netdev->wanted_features;
+	}
 
 	memset(&crq, 0, sizeof(crq));
 	crq.control_ip_offload.first = IBMVNIC_CRQ_CMD;
@@ -2770,132 +5307,161 @@ static void handle_query_ip_offload_rsp(struct ibmvnic_adapter *adapter)
 	ibmvnic_send_crq(adapter, &crq);
 }
 
-static void handle_error_info_rsp(union ibmvnic_crq *crq,
-				  struct ibmvnic_adapter *adapter)
+static void handle_vpd_size_rsp(union ibmvnic_crq *crq,
+				struct ibmvnic_adapter *adapter)
 {
 	struct device *dev = &adapter->vdev->dev;
-	struct ibmvnic_error_buff *error_buff, *tmp;
-	unsigned long flags;
-	bool found = false;
-	int i;
-
-	if (!crq->request_error_rsp.rc.code) {
-		dev_info(dev, "Request Error Rsp returned with rc=%x\n",
-			 crq->request_error_rsp.rc.code);
-		return;
-	}
 
-	spin_lock_irqsave(&adapter->error_list_lock, flags);
-	list_for_each_entry_safe(error_buff, tmp, &adapter->errors, list)
-		if (error_buff->error_id == crq->request_error_rsp.error_id) {
-			found = true;
-			list_del(&error_buff->list);
-			break;
-		}
-	spin_unlock_irqrestore(&adapter->error_list_lock, flags);
-
-	if (!found) {
-		dev_err(dev, "Couldn't find error id %x\n",
-			be32_to_cpu(crq->request_error_rsp.error_id));
+	if (crq->get_vpd_size_rsp.rc.code) {
+		dev_err(dev, "Error retrieving VPD size, rc=%x\n",
+			crq->get_vpd_size_rsp.rc.code);
+		complete(&adapter->fw_done);
 		return;
 	}
 
-	dev_err(dev, "Detailed info for error id %x:",
-		be32_to_cpu(crq->request_error_rsp.error_id));
-
-	for (i = 0; i < error_buff->len; i++) {
-		pr_cont("%02x", (int)error_buff->buff[i]);
-		if (i % 8 == 7)
-			pr_cont(" ");
-	}
-	pr_cont("\n");
-
-	dma_unmap_single(dev, error_buff->dma, error_buff->len,
-			 DMA_FROM_DEVICE);
-	kfree(error_buff->buff);
-	kfree(error_buff);
+	adapter->vpd->len = be64_to_cpu(crq->get_vpd_size_rsp.len);
+	complete(&adapter->fw_done);
 }
 
-static void request_error_information(struct ibmvnic_adapter *adapter,
-				      union ibmvnic_crq *err_crq)
+static void handle_vpd_rsp(union ibmvnic_crq *crq,
+			   struct ibmvnic_adapter *adapter)
 {
 	struct device *dev = &adapter->vdev->dev;
-	struct net_device *netdev = adapter->netdev;
-	struct ibmvnic_error_buff *error_buff;
-	unsigned long timeout = msecs_to_jiffies(30000);
-	union ibmvnic_crq crq;
-	unsigned long flags;
-	int rc, detail_len;
+	unsigned char *substr = NULL;
+	u8 fw_level_len = 0;
 
-	error_buff = kmalloc(sizeof(*error_buff), GFP_ATOMIC);
-	if (!error_buff)
-		return;
+	memset(adapter->fw_version, 0, 32);
 
-	detail_len = be32_to_cpu(err_crq->error_indication.detail_error_sz);
-	error_buff->buff = kmalloc(detail_len, GFP_ATOMIC);
-	if (!error_buff->buff) {
-		kfree(error_buff);
-		return;
+	dma_unmap_single(dev, adapter->vpd->dma_addr, adapter->vpd->len,
+			 DMA_FROM_DEVICE);
+
+	if (crq->get_vpd_rsp.rc.code) {
+		dev_err(dev, "Error retrieving VPD from device, rc=%x\n",
+			crq->get_vpd_rsp.rc.code);
+		goto complete;
 	}
 
-	error_buff->dma = dma_map_single(dev, error_buff->buff, detail_len,
-					 DMA_FROM_DEVICE);
-	if (dma_mapping_error(dev, error_buff->dma)) {
-		netdev_err(netdev, "Couldn't map error buffer\n");
-		kfree(error_buff->buff);
-		kfree(error_buff);
-		return;
+	/* get the position of the firmware version info
+	 * located after the ASCII 'RM' substring in the buffer
+	 */
+	substr = strnstr(adapter->vpd->buff, "RM", adapter->vpd->len);
+	if (!substr) {
+		dev_info(dev, "Warning - No FW level has been provided in the VPD buffer by the VIOS Server\n");
+		goto complete;
 	}
 
-	error_buff->len = detail_len;
-	error_buff->error_id = err_crq->error_indication.error_id;
+	/* get length of firmware level ASCII substring */
+	if ((substr + 2) < (adapter->vpd->buff + adapter->vpd->len)) {
+		fw_level_len = *(substr + 2);
+	} else {
+		dev_info(dev, "Length of FW substr extrapolated VDP buff\n");
+		goto complete;
+	}
 
-	spin_lock_irqsave(&adapter->error_list_lock, flags);
-	list_add_tail(&error_buff->list, &adapter->errors);
-	spin_unlock_irqrestore(&adapter->error_list_lock, flags);
+	/* copy firmware version string from vpd into adapter */
+	if ((substr + 3 + fw_level_len) <
+	    (adapter->vpd->buff + adapter->vpd->len)) {
+		strscpy(adapter->fw_version, substr + 3,
+			sizeof(adapter->fw_version));
+	} else {
+		dev_info(dev, "FW substr extrapolated VPD buff\n");
+	}
 
-	memset(&crq, 0, sizeof(crq));
-	crq.request_error_info.first = IBMVNIC_CRQ_CMD;
-	crq.request_error_info.cmd = REQUEST_ERROR_INFO;
-	crq.request_error_info.ioba = cpu_to_be32(error_buff->dma);
-	crq.request_error_info.len = cpu_to_be32(detail_len);
-	crq.request_error_info.error_id = err_crq->error_indication.error_id;
+complete:
+	if (adapter->fw_version[0] == '\0')
+		strscpy((char *)adapter->fw_version, "N/A", sizeof(adapter->fw_version));
+	complete(&adapter->fw_done);
+}
 
-	rc = ibmvnic_send_crq(adapter, &crq);
-	if (rc) {
-		netdev_err(netdev, "failed to request error information\n");
-		goto err_info_fail;
-	}
+static void handle_query_ip_offload_rsp(struct ibmvnic_adapter *adapter)
+{
+	struct device *dev = &adapter->vdev->dev;
+	struct ibmvnic_query_ip_offload_buffer *buf = &adapter->ip_offload_buf;
 
-	if (!wait_for_completion_timeout(&adapter->init_done, timeout)) {
-		netdev_err(netdev, "timeout waiting for error information\n");
-		goto err_info_fail;
-	}
+	dma_unmap_single(dev, adapter->ip_offload_tok,
+			 sizeof(adapter->ip_offload_buf), DMA_FROM_DEVICE);
 
-	return;
+	netdev_dbg(adapter->netdev, "Query IP Offload Buffer:\n");
+	ibmvnic_print_hex_dump(adapter->netdev, buf,
+			       sizeof(adapter->ip_offload_buf));
+
+	netdev_dbg(adapter->netdev, "ipv4_chksum = %d\n", buf->ipv4_chksum);
+	netdev_dbg(adapter->netdev, "ipv6_chksum = %d\n", buf->ipv6_chksum);
+	netdev_dbg(adapter->netdev, "tcp_ipv4_chksum = %d\n",
+		   buf->tcp_ipv4_chksum);
+	netdev_dbg(adapter->netdev, "tcp_ipv6_chksum = %d\n",
+		   buf->tcp_ipv6_chksum);
+	netdev_dbg(adapter->netdev, "udp_ipv4_chksum = %d\n",
+		   buf->udp_ipv4_chksum);
+	netdev_dbg(adapter->netdev, "udp_ipv6_chksum = %d\n",
+		   buf->udp_ipv6_chksum);
+	netdev_dbg(adapter->netdev, "large_tx_ipv4 = %d\n",
+		   buf->large_tx_ipv4);
+	netdev_dbg(adapter->netdev, "large_tx_ipv6 = %d\n",
+		   buf->large_tx_ipv6);
+	netdev_dbg(adapter->netdev, "large_rx_ipv4 = %d\n",
+		   buf->large_rx_ipv4);
+	netdev_dbg(adapter->netdev, "large_rx_ipv6 = %d\n",
+		   buf->large_rx_ipv6);
+	netdev_dbg(adapter->netdev, "max_ipv4_hdr_sz = %d\n",
+		   buf->max_ipv4_header_size);
+	netdev_dbg(adapter->netdev, "max_ipv6_hdr_sz = %d\n",
+		   buf->max_ipv6_header_size);
+	netdev_dbg(adapter->netdev, "max_tcp_hdr_size = %d\n",
+		   buf->max_tcp_header_size);
+	netdev_dbg(adapter->netdev, "max_udp_hdr_size = %d\n",
+		   buf->max_udp_header_size);
+	netdev_dbg(adapter->netdev, "max_large_tx_size = %d\n",
+		   buf->max_large_tx_size);
+	netdev_dbg(adapter->netdev, "max_large_rx_size = %d\n",
+		   buf->max_large_rx_size);
+	netdev_dbg(adapter->netdev, "ipv6_ext_hdr = %d\n",
+		   buf->ipv6_extension_header);
+	netdev_dbg(adapter->netdev, "tcp_pseudosum_req = %d\n",
+		   buf->tcp_pseudosum_req);
+	netdev_dbg(adapter->netdev, "num_ipv6_ext_hd = %d\n",
+		   buf->num_ipv6_ext_headers);
+	netdev_dbg(adapter->netdev, "off_ipv6_ext_hd = %d\n",
+		   buf->off_ipv6_ext_headers);
 
-err_info_fail:
-	spin_lock_irqsave(&adapter->error_list_lock, flags);
-	list_del(&error_buff->list);
-	spin_unlock_irqrestore(&adapter->error_list_lock, flags);
+	send_control_ip_offload(adapter);
+}
 
-	kfree(error_buff->buff);
-	kfree(error_buff);
+static const char *ibmvnic_fw_err_cause(u16 cause)
+{
+	switch (cause) {
+	case ADAPTER_PROBLEM:
+		return "adapter problem";
+	case BUS_PROBLEM:
+		return "bus problem";
+	case FW_PROBLEM:
+		return "firmware problem";
+	case DD_PROBLEM:
+		return "device driver problem";
+	case EEH_RECOVERY:
+		return "EEH recovery";
+	case FW_UPDATED:
+		return "firmware updated";
+	case LOW_MEMORY:
+		return "low Memory";
+	default:
+		return "unknown";
+	}
 }
 
 static void handle_error_indication(union ibmvnic_crq *crq,
 				    struct ibmvnic_adapter *adapter)
 {
 	struct device *dev = &adapter->vdev->dev;
+	u16 cause;
 
-	dev_err(dev, "Firmware reports %serror id %x, cause %d\n",
-		crq->error_indication.flags
-			& IBMVNIC_FATAL_ERROR ? "FATAL " : "",
-		be32_to_cpu(crq->error_indication.error_id),
-		be16_to_cpu(crq->error_indication.error_cause));
+	cause = be16_to_cpu(crq->error_indication.error_cause);
 
-	if (be32_to_cpu(crq->error_indication.error_id))
-		request_error_information(adapter, crq);
+	dev_warn_ratelimited(dev,
+			     "Firmware reports %serror, cause: %s. Starting recovery...\n",
+			     crq->error_indication.flags
+				& IBMVNIC_FATAL_ERROR ? "FATAL " : "",
+			     ibmvnic_fw_err_cause(cause));
 
 	if (crq->error_indication.flags & IBMVNIC_FATAL_ERROR)
 		ibmvnic_reset(adapter, VNIC_RESET_FATAL);
@@ -2903,8 +5469,8 @@ static void handle_error_indication(union ibmvnic_crq *crq,
 		ibmvnic_reset(adapter, VNIC_RESET_NON_FATAL);
 }
 
-static void handle_change_mac_rsp(union ibmvnic_crq *crq,
-				  struct ibmvnic_adapter *adapter)
+static int handle_change_mac_rsp(union ibmvnic_crq *crq,
+				 struct ibmvnic_adapter *adapter)
 {
 	struct net_device *netdev = adapter->netdev;
 	struct device *dev = &adapter->vdev->dev;
@@ -2913,10 +5479,17 @@ static void handle_change_mac_rsp(union ibmvnic_crq *crq,
 	rc = crq->change_mac_addr_rsp.rc.code;
 	if (rc) {
 		dev_err(dev, "Error %ld in CHANGE_MAC_ADDR_RSP\n", rc);
-		return;
+		goto out;
 	}
-	memcpy(netdev->dev_addr, &crq->change_mac_addr_rsp.mac_addr[0],
-	       ETH_ALEN);
+	/* crq->change_mac_addr.mac_addr is the requested one
+	 * crq->change_mac_addr_rsp.mac_addr is the returned valid one.
+	 */
+	eth_hw_addr_set(netdev, &crq->change_mac_addr_rsp.mac_addr[0]);
+	ether_addr_copy(adapter->mac_addr,
+			&crq->change_mac_addr_rsp.mac_addr[0]);
+out:
+	complete(&adapter->fw_done);
+	return rc;
 }
 
 static void handle_request_cap_rsp(union ibmvnic_crq *crq,
@@ -2927,6 +5500,8 @@ static void handle_request_cap_rsp(union ibmvnic_crq *crq,
 	char *name;
 
 	atomic_dec(&adapter->running_cap_crqs);
+	netdev_dbg(adapter->netdev, "Outstanding request-caps: %d\n",
+		   atomic_read(&adapter->running_cap_crqs));
 	switch (be16_to_cpu(crq->request_capability_rsp.capability)) {
 	case REQ_TX_QUEUES:
 		req_value = &adapter->req_tx_queues;
@@ -2968,11 +5543,20 @@ static void handle_request_cap_rsp(union ibmvnic_crq *crq,
 	case PARTIALSUCCESS:
 		dev_info(dev, "req=%lld, rsp=%ld in %s queue, retrying.\n",
 			 *req_value,
-			 (long int)be64_to_cpu(crq->request_capability_rsp.
-					       number), name);
-		release_sub_crqs(adapter);
-		*req_value = be64_to_cpu(crq->request_capability_rsp.number);
-		ibmvnic_send_req_caps(adapter, 1);
+			 (long)be64_to_cpu(crq->request_capability_rsp.number),
+			 name);
+
+		if (be16_to_cpu(crq->request_capability_rsp.capability) ==
+		    REQ_MTU) {
+			pr_err("mtu of %llu is not supported. Reverting.\n",
+			       *req_value);
+			*req_value = adapter->fallback.mtu;
+		} else {
+			*req_value =
+				be64_to_cpu(crq->request_capability_rsp.number);
+		}
+
+		send_request_cap(adapter, 1);
 		return;
 	default:
 		dev_err(dev, "Error %d in request cap rsp\n",
@@ -2981,105 +5565,112 @@ static void handle_request_cap_rsp(union ibmvnic_crq *crq,
 	}
 
 	/* Done receiving requested capabilities, query IP offload support */
-	if (atomic_read(&adapter->running_cap_crqs) == 0) {
-		union ibmvnic_crq newcrq;
-		int buf_sz = sizeof(struct ibmvnic_query_ip_offload_buffer);
-		struct ibmvnic_query_ip_offload_buffer *ip_offload_buf =
-		    &adapter->ip_offload_buf;
-
-		adapter->wait_capability = false;
-		adapter->ip_offload_tok = dma_map_single(dev, ip_offload_buf,
-							 buf_sz,
-							 DMA_FROM_DEVICE);
-
-		if (dma_mapping_error(dev, adapter->ip_offload_tok)) {
-			if (!firmware_has_feature(FW_FEATURE_CMO))
-				dev_err(dev, "Couldn't map offload buffer\n");
-			return;
-		}
-
-		memset(&newcrq, 0, sizeof(newcrq));
-		newcrq.query_ip_offload.first = IBMVNIC_CRQ_CMD;
-		newcrq.query_ip_offload.cmd = QUERY_IP_OFFLOAD;
-		newcrq.query_ip_offload.len = cpu_to_be32(buf_sz);
-		newcrq.query_ip_offload.ioba =
-		    cpu_to_be32(adapter->ip_offload_tok);
-
-		ibmvnic_send_crq(adapter, &newcrq);
-	}
+	if (atomic_read(&adapter->running_cap_crqs) == 0)
+		send_query_ip_offload(adapter);
 }
 
 static int handle_login_rsp(union ibmvnic_crq *login_rsp_crq,
 			    struct ibmvnic_adapter *adapter)
 {
 	struct device *dev = &adapter->vdev->dev;
+	struct net_device *netdev = adapter->netdev;
 	struct ibmvnic_login_rsp_buffer *login_rsp = adapter->login_rsp_buf;
 	struct ibmvnic_login_buffer *login = adapter->login_buf;
+	u64 *tx_handle_array;
+	u64 *rx_handle_array;
+	int num_tx_pools;
+	int num_rx_pools;
+	u64 *size_array;
+	u32 rsp_len;
 	int i;
 
-	dma_unmap_single(dev, adapter->login_buf_token, adapter->login_buf_sz,
-			 DMA_BIDIRECTIONAL);
-	dma_unmap_single(dev, adapter->login_rsp_buf_token,
-			 adapter->login_rsp_buf_sz, DMA_BIDIRECTIONAL);
+	/* CHECK: Test/set of login_pending does not need to be atomic
+	 * because only ibmvnic_tasklet tests/clears this.
+	 */
+	if (!adapter->login_pending) {
+		netdev_warn(netdev, "Ignoring unexpected login response\n");
+		return 0;
+	}
+	adapter->login_pending = false;
 
 	/* If the number of queues requested can't be allocated by the
 	 * server, the login response will return with code 1. We will need
 	 * to resend the login buffer with fewer queues requested.
 	 */
 	if (login_rsp_crq->generic.rc.code) {
-		adapter->renegotiate = true;
+		adapter->init_done_rc = login_rsp_crq->generic.rc.code;
 		complete(&adapter->init_done);
 		return 0;
 	}
 
-	netdev_dbg(adapter->netdev, "Login Response Buffer:\n");
-	for (i = 0; i < (adapter->login_rsp_buf_sz - 1) / 8 + 1; i++) {
-		netdev_dbg(adapter->netdev, "%016lx\n",
-			   ((unsigned long int *)(adapter->login_rsp_buf))[i]);
+	if (adapter->failover_pending) {
+		adapter->init_done_rc = -EAGAIN;
+		netdev_dbg(netdev, "Failover pending, ignoring login response\n");
+		complete(&adapter->init_done);
+		/* login response buffer will be released on reset */
+		return 0;
 	}
 
+	netdev->mtu = adapter->req_mtu - ETH_HLEN;
+
+	netdev_dbg(adapter->netdev, "Login Response Buffer:\n");
+	ibmvnic_print_hex_dump(netdev, adapter->login_rsp_buf,
+			       adapter->login_rsp_buf_sz);
+
 	/* Sanity checks */
 	if (login->num_txcomp_subcrqs != login_rsp->num_txsubm_subcrqs ||
 	    (be32_to_cpu(login->num_rxcomp_subcrqs) *
 	     adapter->req_rx_add_queues !=
 	     be32_to_cpu(login_rsp->num_rxadd_subcrqs))) {
 		dev_err(dev, "FATAL: Inconsistent login and login rsp\n");
-		ibmvnic_remove(adapter->vdev);
+		ibmvnic_reset(adapter, VNIC_RESET_FATAL);
 		return -EIO;
 	}
-	complete(&adapter->init_done);
 
-	return 0;
-}
+	rsp_len = be32_to_cpu(login_rsp->len);
+	if (be32_to_cpu(login->login_rsp_len) < rsp_len ||
+	    rsp_len <= be32_to_cpu(login_rsp->off_txsubm_subcrqs) ||
+	    rsp_len <= be32_to_cpu(login_rsp->off_rxadd_subcrqs) ||
+	    rsp_len <= be32_to_cpu(login_rsp->off_rxadd_buff_size) ||
+	    rsp_len <= be32_to_cpu(login_rsp->off_supp_tx_desc)) {
+		/* This can happen if a login request times out and there are
+		 * 2 outstanding login requests sent, the LOGIN_RSP crq
+		 * could have been for the older login request. So we are
+		 * parsing the newer response buffer which may be incomplete
+		 */
+		dev_err(dev, "FATAL: Login rsp offsets/lengths invalid\n");
+		ibmvnic_reset(adapter, VNIC_RESET_FATAL);
+		return -EIO;
+	}
 
-static void handle_request_map_rsp(union ibmvnic_crq *crq,
-				   struct ibmvnic_adapter *adapter)
-{
-	struct device *dev = &adapter->vdev->dev;
-	u8 map_id = crq->request_map_rsp.map_id;
-	int tx_subcrqs;
-	int rx_subcrqs;
-	long rc;
-	int i;
+	size_array = (u64 *)((u8 *)(adapter->login_rsp_buf) +
+		be32_to_cpu(adapter->login_rsp_buf->off_rxadd_buff_size));
+	/* variable buffer sizes are not supported, so just read the
+	 * first entry.
+	 */
+	adapter->cur_rx_buf_sz = be64_to_cpu(size_array[0]);
 
-	tx_subcrqs = be32_to_cpu(adapter->login_rsp_buf->num_txsubm_subcrqs);
-	rx_subcrqs = be32_to_cpu(adapter->login_rsp_buf->num_rxadd_subcrqs);
+	num_tx_pools = be32_to_cpu(adapter->login_rsp_buf->num_txsubm_subcrqs);
+	num_rx_pools = be32_to_cpu(adapter->login_rsp_buf->num_rxadd_subcrqs);
 
-	rc = crq->request_map_rsp.rc.code;
-	if (rc) {
-		dev_err(dev, "Error %ld in REQUEST_MAP_RSP\n", rc);
-		adapter->map_id--;
-		/* need to find and zero tx/rx_pool map_id */
-		for (i = 0; i < tx_subcrqs; i++) {
-			if (adapter->tx_pool[i].long_term_buff.map_id == map_id)
-				adapter->tx_pool[i].long_term_buff.map_id = 0;
-		}
-		for (i = 0; i < rx_subcrqs; i++) {
-			if (adapter->rx_pool[i].long_term_buff.map_id == map_id)
-				adapter->rx_pool[i].long_term_buff.map_id = 0;
-		}
-	}
-	complete(&adapter->fw_done);
+	tx_handle_array = (u64 *)((u8 *)(adapter->login_rsp_buf) +
+				  be32_to_cpu(adapter->login_rsp_buf->off_txsubm_subcrqs));
+	rx_handle_array = (u64 *)((u8 *)(adapter->login_rsp_buf) +
+				  be32_to_cpu(adapter->login_rsp_buf->off_rxadd_subcrqs));
+
+	for (i = 0; i < num_tx_pools; i++)
+		adapter->tx_scrq[i]->handle = tx_handle_array[i];
+
+	for (i = 0; i < num_rx_pools; i++)
+		adapter->rx_scrq[i]->handle = rx_handle_array[i];
+
+	adapter->num_active_tx_scrqs = num_tx_pools;
+	adapter->num_active_rx_scrqs = num_rx_pools;
+	release_login_rsp_buffer(adapter);
+	release_login_buffer(adapter);
+	complete(&adapter->init_done);
+
+	return 0;
 }
 
 static void handle_request_unmap_rsp(union ibmvnic_crq *crq,
@@ -3105,9 +5696,10 @@ static void handle_query_map_rsp(union ibmvnic_crq *crq,
 		dev_err(dev, "Error %ld in QUERY_MAP_RSP\n", rc);
 		return;
 	}
-	netdev_dbg(netdev, "page_size = %d\ntot_pages = %d\nfree_pages = %d\n",
-		   crq->query_map_rsp.page_size, crq->query_map_rsp.tot_pages,
-		   crq->query_map_rsp.free_pages);
+	netdev_dbg(netdev, "page_size = %d\ntot_pages = %u\nfree_pages = %u\n",
+		   crq->query_map_rsp.page_size,
+		   __be32_to_cpu(crq->query_map_rsp.tot_pages),
+		   __be32_to_cpu(crq->query_map_rsp.free_pages));
 }
 
 static void handle_query_cap_rsp(union ibmvnic_crq *crq,
@@ -3283,10 +5875,92 @@ static void handle_query_cap_rsp(union ibmvnic_crq *crq,
 	}
 
 out:
-	if (atomic_read(&adapter->running_cap_crqs) == 0) {
-		adapter->wait_capability = false;
-		ibmvnic_send_req_caps(adapter, 0);
+	if (atomic_read(&adapter->running_cap_crqs) == 0)
+		send_request_cap(adapter, 0);
+}
+
+static int send_query_phys_parms(struct ibmvnic_adapter *adapter)
+{
+	union ibmvnic_crq crq;
+	int rc;
+
+	memset(&crq, 0, sizeof(crq));
+	crq.query_phys_parms.first = IBMVNIC_CRQ_CMD;
+	crq.query_phys_parms.cmd = QUERY_PHYS_PARMS;
+
+	mutex_lock(&adapter->fw_lock);
+	adapter->fw_done_rc = 0;
+	reinit_completion(&adapter->fw_done);
+
+	rc = ibmvnic_send_crq(adapter, &crq);
+	if (rc) {
+		mutex_unlock(&adapter->fw_lock);
+		return rc;
+	}
+
+	rc = ibmvnic_wait_for_completion(adapter, &adapter->fw_done, 10000);
+	if (rc) {
+		mutex_unlock(&adapter->fw_lock);
+		return rc;
+	}
+
+	mutex_unlock(&adapter->fw_lock);
+	return adapter->fw_done_rc ? -EIO : 0;
+}
+
+static int handle_query_phys_parms_rsp(union ibmvnic_crq *crq,
+				       struct ibmvnic_adapter *adapter)
+{
+	struct net_device *netdev = adapter->netdev;
+	int rc;
+	__be32 rspeed = cpu_to_be32(crq->query_phys_parms_rsp.speed);
+
+	rc = crq->query_phys_parms_rsp.rc.code;
+	if (rc) {
+		netdev_err(netdev, "Error %d in QUERY_PHYS_PARMS\n", rc);
+		return rc;
+	}
+	switch (rspeed) {
+	case IBMVNIC_10MBPS:
+		adapter->speed = SPEED_10;
+		break;
+	case IBMVNIC_100MBPS:
+		adapter->speed = SPEED_100;
+		break;
+	case IBMVNIC_1GBPS:
+		adapter->speed = SPEED_1000;
+		break;
+	case IBMVNIC_10GBPS:
+		adapter->speed = SPEED_10000;
+		break;
+	case IBMVNIC_25GBPS:
+		adapter->speed = SPEED_25000;
+		break;
+	case IBMVNIC_40GBPS:
+		adapter->speed = SPEED_40000;
+		break;
+	case IBMVNIC_50GBPS:
+		adapter->speed = SPEED_50000;
+		break;
+	case IBMVNIC_100GBPS:
+		adapter->speed = SPEED_100000;
+		break;
+	case IBMVNIC_200GBPS:
+		adapter->speed = SPEED_200000;
+		break;
+	default:
+		if (netif_carrier_ok(netdev))
+			netdev_warn(netdev, "Unknown speed 0x%08x\n", rspeed);
+		adapter->speed = SPEED_UNKNOWN;
 	}
+	if (crq->query_phys_parms_rsp.flags1 & IBMVNIC_FULL_DUPLEX)
+		adapter->duplex = DUPLEX_FULL;
+	else if (crq->query_phys_parms_rsp.flags1 & IBMVNIC_HALF_DUPLEX)
+		adapter->duplex = DUPLEX_HALF;
+	else
+		adapter->duplex = DUPLEX_UNKNOWN;
+
+	return rc;
 }
 
 static void ibmvnic_handle_crq(union ibmvnic_crq *crq,
@@ -3299,18 +5973,50 @@ static void ibmvnic_handle_crq(union ibmvnic_crq *crq,
 	long rc;
 
 	netdev_dbg(netdev, "Handling CRQ: %016lx %016lx\n",
-		   (unsigned long int)cpu_to_be64(u64_crq[0]),
-		   (unsigned long int)cpu_to_be64(u64_crq[1]));
+		   (unsigned long)cpu_to_be64(u64_crq[0]),
+		   (unsigned long)cpu_to_be64(u64_crq[1]));
 	switch (gen_crq->first) {
 	case IBMVNIC_CRQ_INIT_RSP:
 		switch (gen_crq->cmd) {
 		case IBMVNIC_CRQ_INIT:
 			dev_info(dev, "Partner initialized\n");
 			adapter->from_passive_init = true;
-			complete(&adapter->init_done);
+			/* Discard any stale login responses from prev reset.
+			 * CHECK: should we clear even on INIT_COMPLETE?
+			 */
+			adapter->login_pending = false;
+
+			if (adapter->state == VNIC_DOWN)
+				rc = ibmvnic_reset(adapter, VNIC_RESET_PASSIVE_INIT);
+			else
+				rc = ibmvnic_reset(adapter, VNIC_RESET_FAILOVER);
+
+			if (rc && rc != -EBUSY) {
+				/* We were unable to schedule the failover
+				 * reset either because the adapter was still
+				 * probing (eg: during kexec) or we could not
+				 * allocate memory. Clear the failover_pending
+				 * flag since no one else will. We ignore
+				 * EBUSY because it means either FAILOVER reset
+				 * is already scheduled or the adapter is
+				 * being removed.
+				 */
+				netdev_err(netdev,
+					   "Error %ld scheduling failover reset\n",
+					   rc);
+				adapter->failover_pending = false;
+			}
+
+			if (!completion_done(&adapter->init_done)) {
+				if (!adapter->init_done_rc)
+					adapter->init_done_rc = -EAGAIN;
+				complete(&adapter->init_done);
+			}
+
 			break;
 		case IBMVNIC_CRQ_INIT_COMPLETE:
 			dev_info(dev, "Partner initialization complete\n");
+			adapter->crq.active = true;
 			send_version_xchg(adapter);
 			break;
 		default:
@@ -3319,12 +6025,31 @@ static void ibmvnic_handle_crq(union ibmvnic_crq *crq,
 		return;
 	case IBMVNIC_CRQ_XPORT_EVENT:
 		netif_carrier_off(netdev);
+		adapter->crq.active = false;
+		/* terminate any thread waiting for a response
+		 * from the device
+		 */
+		if (!completion_done(&adapter->fw_done)) {
+			adapter->fw_done_rc = -EIO;
+			complete(&adapter->fw_done);
+		}
+
+		/* if we got here during crq-init, retry crq-init */
+		if (!completion_done(&adapter->init_done)) {
+			adapter->init_done_rc = -EAGAIN;
+			complete(&adapter->init_done);
+		}
+
+		if (!completion_done(&adapter->stats_done))
+			complete(&adapter->stats_done);
+		if (test_bit(0, &adapter->resetting))
+			adapter->force_reset_recovery = true;
 		if (gen_crq->cmd == IBMVNIC_PARTITION_MIGRATED) {
 			dev_info(dev, "Migrated, re-enabling adapter\n");
 			ibmvnic_reset(adapter, VNIC_RESET_MOBILITY);
 		} else if (gen_crq->cmd == IBMVNIC_DEVICE_FAILOVER) {
 			dev_info(dev, "Backing device failover detected\n");
-			ibmvnic_reset(adapter, VNIC_RESET_FAILOVER);
+			adapter->failover_pending = true;
 		} else {
 			/* The adapter lost the connection */
 			dev_err(dev, "Virtual Adapter failed (rc=%d)\n",
@@ -3347,13 +6072,11 @@ static void ibmvnic_handle_crq(union ibmvnic_crq *crq,
 			dev_err(dev, "Error %ld in VERSION_EXCHG_RSP\n", rc);
 			break;
 		}
-		dev_info(dev, "Partner protocol version is %d\n",
-			 crq->version_exchange_rsp.version);
-		if (be16_to_cpu(crq->version_exchange_rsp.version) <
-		    ibmvnic_version)
-			ibmvnic_version =
+		ibmvnic_version =
 			    be16_to_cpu(crq->version_exchange_rsp.version);
-		send_cap_queries(adapter);
+		dev_info(dev, "Partner protocol version is %d\n",
+			 ibmvnic_version);
+		send_query_cap(adapter);
 		break;
 	case QUERY_CAPABILITY_RSP:
 		handle_query_cap_rsp(crq, adapter);
@@ -3362,7 +6085,8 @@ static void ibmvnic_handle_crq(union ibmvnic_crq *crq,
 		handle_query_map_rsp(crq, adapter);
 		break;
 	case REQUEST_MAP_RSP:
-		handle_request_map_rsp(crq, adapter);
+		adapter->fw_done_rc = crq->request_map_rsp.rc.code;
+		complete(&adapter->fw_done);
 		break;
 	case REQUEST_UNMAP_RSP:
 		handle_request_unmap_rsp(crq, adapter);
@@ -3390,19 +6114,19 @@ static void ibmvnic_handle_crq(union ibmvnic_crq *crq,
 		    crq->link_state_indication.phys_link_state;
 		adapter->logical_link_state =
 		    crq->link_state_indication.logical_link_state;
+		if (adapter->phys_link_state && adapter->logical_link_state)
+			netif_carrier_on(netdev);
+		else
+			netif_carrier_off(netdev);
 		break;
 	case CHANGE_MAC_ADDR_RSP:
 		netdev_dbg(netdev, "Got MAC address change Response\n");
-		handle_change_mac_rsp(crq, adapter);
+		adapter->fw_done_rc = handle_change_mac_rsp(crq, adapter);
 		break;
 	case ERROR_INDICATION:
 		netdev_dbg(netdev, "Got Error Indication\n");
 		handle_error_indication(crq, adapter);
 		break;
-	case REQUEST_ERROR_RSP:
-		netdev_dbg(netdev, "Got Error Detail Response\n");
-		handle_error_info_rsp(crq, adapter);
-		break;
 	case REQUEST_STATISTICS_RSP:
 		netdev_dbg(netdev, "Got Statistics Response\n");
 		complete(&adapter->stats_done);
@@ -3425,6 +6149,16 @@ static void ibmvnic_handle_crq(union ibmvnic_crq *crq,
 		netdev_dbg(netdev, "Got Collect firmware trace Response\n");
 		complete(&adapter->fw_done);
 		break;
+	case GET_VPD_SIZE_RSP:
+		handle_vpd_size_rsp(crq, adapter);
+		break;
+	case GET_VPD_RSP:
+		handle_vpd_rsp(crq, adapter);
+		break;
+	case QUERY_PHYS_PARMS_RSP:
+		adapter->fw_done_rc = handle_query_phys_parms_rsp(crq, adapter);
+		complete(&adapter->fw_done);
+		break;
 	default:
 		netdev_err(netdev, "Got an invalid cmd type 0x%02x\n",
 			   gen_crq->cmd);
@@ -3439,33 +6173,27 @@ static irqreturn_t ibmvnic_interrupt(int irq, void *instance)
 	return IRQ_HANDLED;
 }
 
-static void ibmvnic_tasklet(void *data)
+static void ibmvnic_tasklet(struct tasklet_struct *t)
 {
-	struct ibmvnic_adapter *adapter = data;
+	struct ibmvnic_adapter *adapter = from_tasklet(adapter, t, tasklet);
 	struct ibmvnic_crq_queue *queue = &adapter->crq;
 	union ibmvnic_crq *crq;
 	unsigned long flags;
-	bool done = false;
 
 	spin_lock_irqsave(&queue->lock, flags);
-	while (!done) {
-		/* Pull all the valid messages off the CRQ */
-		while ((crq = ibmvnic_next_crq(adapter)) != NULL) {
-			ibmvnic_handle_crq(crq, adapter);
-			crq->generic.first = 0;
-		}
 
-		/* remain in tasklet until all
-		 * capabilities responses are received
+	/* Pull all the valid messages off the CRQ */
+	while ((crq = ibmvnic_next_crq(adapter)) != NULL) {
+		/* This barrier makes sure ibmvnic_next_crq()'s
+		 * crq->generic.first & IBMVNIC_CRQ_CMD_RSP is loaded
+		 * before ibmvnic_handle_crq()'s
+		 * switch(gen_crq->first) and switch(gen_crq->cmd).
 		 */
-		if (!adapter->wait_capability)
-			done = true;
+		dma_rmb();
+		ibmvnic_handle_crq(crq, adapter);
+		crq->generic.first = 0;
 	}
-	/* if capabilities CRQ's were sent in this tasklet, the following
-	 * tasklet must wait until all responses are received
-	 */
-	if (atomic_read(&adapter->running_cap_crqs) != 0)
-		adapter->wait_capability = true;
+
 	spin_unlock_irqrestore(&queue->lock, flags);
 }
 
@@ -3497,8 +6225,12 @@ static int ibmvnic_reset_crq(struct ibmvnic_adapter *adapter)
 	} while (rc == H_BUSY || H_IS_LONG_BUSY(rc));
 
 	/* Clean out the queue */
+	if (!crq->msgs)
+		return -EINVAL;
+
 	memset(crq->msgs, 0, PAGE_SIZE);
 	crq->cur = 0;
+	crq->active = false;
 
 	/* And re-open it again */
 	rc = plpar_hcall_norets(H_REG_CRQ, vdev->unit_address,
@@ -3533,6 +6265,7 @@ static void release_crq_queue(struct ibmvnic_adapter *adapter)
 			 DMA_BIDIRECTIONAL);
 	free_page((unsigned long)crq->msgs);
 	crq->msgs = NULL;
+	crq->active = false;
 }
 
 static int init_crq_queue(struct ibmvnic_adapter *adapter)
@@ -3574,12 +6307,12 @@ static int init_crq_queue(struct ibmvnic_adapter *adapter)
 
 	retrc = 0;
 
-	tasklet_init(&adapter->tasklet, (void *)ibmvnic_tasklet,
-		     (unsigned long)adapter);
+	tasklet_setup(&adapter->tasklet, (void *)ibmvnic_tasklet);
 
 	netdev_dbg(adapter->netdev, "registering irq 0x%x\n", vdev->irq);
-	rc = request_irq(vdev->irq, ibmvnic_interrupt, 0, IBMVNIC_NAME,
-			 adapter);
+	snprintf(crq->name, sizeof(crq->name), "ibmvnic-%x",
+		 adapter->vdev->unit_address);
+	rc = request_irq(vdev->irq, ibmvnic_interrupt, 0, crq->name, adapter);
 	if (rc) {
 		dev_err(dev, "Couldn't register irq 0x%x. rc=%d\n",
 			vdev->irq, rc);
@@ -3595,6 +6328,9 @@ static int init_crq_queue(struct ibmvnic_adapter *adapter)
 	crq->cur = 0;
 	spin_lock_init(&crq->lock);
 
+	/* process any CRQs that were queued before we enabled interrupts */
+	tasklet_schedule(&adapter->tasklet);
+
 	return retrc;
 
 req_irq_failed:
@@ -3610,57 +6346,100 @@ map_failed:
 	return retrc;
 }
 
-static int ibmvnic_init(struct ibmvnic_adapter *adapter)
+static int ibmvnic_reset_init(struct ibmvnic_adapter *adapter, bool reset)
 {
 	struct device *dev = &adapter->vdev->dev;
-	unsigned long timeout = msecs_to_jiffies(30000);
+	unsigned long timeout = msecs_to_jiffies(20000);
+	u64 old_num_rx_queues = adapter->req_rx_queues;
+	u64 old_num_tx_queues = adapter->req_tx_queues;
 	int rc;
 
-	if (adapter->resetting) {
-		rc = ibmvnic_reset_crq(adapter);
-		if (!rc)
-			rc = vio_enable_interrupts(adapter->vdev);
-	} else {
-		rc = init_crq_queue(adapter);
-	}
+	adapter->from_passive_init = false;
 
+	rc = ibmvnic_send_crq_init(adapter);
 	if (rc) {
-		dev_err(dev, "Couldn't initialize crq. rc=%d\n", rc);
+		dev_err(dev, "Send crq init failed with error %d\n", rc);
 		return rc;
 	}
 
-	adapter->from_passive_init = false;
-
-	init_completion(&adapter->init_done);
-	ibmvnic_send_crq_init(adapter);
 	if (!wait_for_completion_timeout(&adapter->init_done, timeout)) {
 		dev_err(dev, "Initialization sequence timed out\n");
-		return -1;
+		return -ETIMEDOUT;
+	}
+
+	if (adapter->init_done_rc) {
+		release_crq_queue(adapter);
+		dev_err(dev, "CRQ-init failed, %d\n", adapter->init_done_rc);
+		return adapter->init_done_rc;
 	}
 
 	if (adapter->from_passive_init) {
 		adapter->state = VNIC_OPEN;
 		adapter->from_passive_init = false;
-		return -1;
+		dev_err(dev, "CRQ-init failed, passive-init\n");
+		return -EINVAL;
 	}
 
-	if (adapter->resetting)
-		rc = reset_sub_crq_queues(adapter);
-	else
+	if (reset &&
+	    test_bit(0, &adapter->resetting) && !adapter->wait_for_reset &&
+	    adapter->reset_reason != VNIC_RESET_MOBILITY) {
+		if (adapter->req_rx_queues != old_num_rx_queues ||
+		    adapter->req_tx_queues != old_num_tx_queues) {
+			release_sub_crqs(adapter, 0);
+			rc = init_sub_crqs(adapter);
+		} else {
+			/* no need to reinitialize completely, but we do
+			 * need to clean up transmits that were in flight
+			 * when we processed the reset.  Failure to do so
+			 * will confound the upper layer, usually TCP, by
+			 * creating the illusion of transmits that are
+			 * awaiting completion.
+			 */
+			clean_tx_pools(adapter);
+
+			rc = reset_sub_crq_queues(adapter);
+		}
+	} else {
+		if (adapter->reset_reason == VNIC_RESET_MOBILITY) {
+			/* After an LPM, reset the max number of indirect
+			 * subcrq descriptors per H_SEND_SUB_CRQ_INDIRECT
+			 * hcall to the default max (e.g POWER8 -> POWER10)
+			 *
+			 * If the new destination platform does not support
+			 * the higher limit max (e.g. POWER10-> POWER8 LPM)
+			 * H_PARAMETER will trigger automatic fallback to the
+			 * safe minimum limit.
+			 */
+			adapter->cur_max_ind_descs = IBMVNIC_MAX_IND_DESCS;
+		}
+
 		rc = init_sub_crqs(adapter);
+	}
+
 	if (rc) {
 		dev_err(dev, "Initialization of sub crqs failed\n");
 		release_crq_queue(adapter);
+		return rc;
+	}
+
+	rc = init_sub_crq_irqs(adapter);
+	if (rc) {
+		dev_err(dev, "Failed to initialize sub crq irqs\n");
+		release_crq_queue(adapter);
 	}
 
 	return rc;
 }
 
+static struct device_attribute dev_attr_failover;
+
 static int ibmvnic_probe(struct vio_dev *dev, const struct vio_device_id *id)
 {
 	struct ibmvnic_adapter *adapter;
 	struct net_device *netdev;
 	unsigned char *mac_addr_p;
+	unsigned long flags;
+	bool init_success;
 	int rc;
 
 	dev_dbg(&dev->dev, "entering ibmvnic_probe for UA 0x%x\n",
@@ -3676,7 +6455,7 @@ static int ibmvnic_probe(struct vio_dev *dev, const struct vio_device_id *id)
 	}
 
 	netdev = alloc_etherdev_mq(sizeof(struct ibmvnic_adapter),
-				   IBMVNIC_MAX_TX_QUEUES);
+				   IBMVNIC_MAX_QUEUES);
 	if (!netdev)
 		return -ENOMEM;
 
@@ -3685,66 +6464,245 @@ static int ibmvnic_probe(struct vio_dev *dev, const struct vio_device_id *id)
 	dev_set_drvdata(&dev->dev, netdev);
 	adapter->vdev = dev;
 	adapter->netdev = netdev;
+	adapter->login_pending = false;
+	memset(&adapter->map_ids, 0, sizeof(adapter->map_ids));
+	/* map_ids start at 1, so ensure map_id 0 is always "in-use" */
+	bitmap_set(adapter->map_ids, 0, 1);
 
 	ether_addr_copy(adapter->mac_addr, mac_addr_p);
-	ether_addr_copy(netdev->dev_addr, adapter->mac_addr);
+	eth_hw_addr_set(netdev, adapter->mac_addr);
 	netdev->irq = dev->irq;
 	netdev->netdev_ops = &ibmvnic_netdev_ops;
 	netdev->ethtool_ops = &ibmvnic_ethtool_ops;
 	SET_NETDEV_DEV(netdev, &dev->dev);
 
-	spin_lock_init(&adapter->stats_lock);
-
-	INIT_LIST_HEAD(&adapter->errors);
-	spin_lock_init(&adapter->error_list_lock);
-
 	INIT_WORK(&adapter->ibmvnic_reset, __ibmvnic_reset);
+	INIT_DELAYED_WORK(&adapter->ibmvnic_delayed_reset,
+			  __ibmvnic_delayed_reset);
 	INIT_LIST_HEAD(&adapter->rwi_list);
-	mutex_init(&adapter->reset_lock);
-	mutex_init(&adapter->rwi_lock);
-	adapter->resetting = false;
+	spin_lock_init(&adapter->rwi_lock);
+	spin_lock_init(&adapter->state_lock);
+	mutex_init(&adapter->fw_lock);
+	init_completion(&adapter->probe_done);
+	init_completion(&adapter->init_done);
+	init_completion(&adapter->fw_done);
+	init_completion(&adapter->reset_done);
+	init_completion(&adapter->stats_done);
+	clear_bit(0, &adapter->resetting);
+	adapter->prev_rx_buf_sz = 0;
+	adapter->prev_mtu = 0;
 
-	rc = ibmvnic_init(adapter);
-	if (rc) {
-		free_netdev(netdev);
-		return rc;
+	init_success = false;
+	do {
+		reinit_init_done(adapter);
+
+		/* clear any failovers we got in the previous pass
+		 * since we are reinitializing the CRQ
+		 */
+		adapter->failover_pending = false;
+
+		/* If we had already initialized CRQ, we may have one or
+		 * more resets queued already. Discard those and release
+		 * the CRQ before initializing the CRQ again.
+		 */
+		release_crq_queue(adapter);
+
+		/* Since we are still in PROBING state, __ibmvnic_reset()
+		 * will not access the ->rwi_list and since we released CRQ,
+		 * we won't get _new_ transport events. But there maybe an
+		 * ongoing ibmvnic_reset() call. So serialize access to
+		 * rwi_list. If we win the race, ibvmnic_reset() could add
+		 * a reset after we purged but thats ok - we just may end
+		 * up with an extra reset (i.e similar to having two or more
+		 * resets in the queue at once).
+		 * CHECK.
+		 */
+		spin_lock_irqsave(&adapter->rwi_lock, flags);
+		flush_reset_queue(adapter);
+		spin_unlock_irqrestore(&adapter->rwi_lock, flags);
+
+		rc = init_crq_queue(adapter);
+		if (rc) {
+			dev_err(&dev->dev, "Couldn't initialize crq. rc=%d\n",
+				rc);
+			goto ibmvnic_init_fail;
+		}
+
+		rc = ibmvnic_reset_init(adapter, false);
+	} while (rc == -EAGAIN);
+
+	/* We are ignoring the error from ibmvnic_reset_init() assuming that the
+	 * partner is not ready. CRQ is not active. When the partner becomes
+	 * ready, we will do the passive init reset.
+	 */
+
+	if (!rc)
+		init_success = true;
+
+	rc = init_stats_buffers(adapter);
+	if (rc)
+		goto ibmvnic_init_fail;
+
+	rc = init_stats_token(adapter);
+	if (rc)
+		goto ibmvnic_stats_fail;
+
+	rc = device_create_file(&dev->dev, &dev_attr_failover);
+	if (rc)
+		goto ibmvnic_dev_file_err;
+
+	netif_carrier_off(netdev);
+
+	if (init_success) {
+		adapter->state = VNIC_PROBED;
+		netdev->mtu = adapter->req_mtu - ETH_HLEN;
+		netdev->min_mtu = adapter->min_mtu - ETH_HLEN;
+		netdev->max_mtu = adapter->max_mtu - ETH_HLEN;
+	} else {
+		adapter->state = VNIC_DOWN;
 	}
 
-	netdev->mtu = adapter->req_mtu - ETH_HLEN;
+	adapter->wait_for_reset = false;
+	adapter->last_reset_time = jiffies;
+	adapter->cur_max_ind_descs = IBMVNIC_MAX_IND_DESCS;
 
 	rc = register_netdev(netdev);
 	if (rc) {
 		dev_err(&dev->dev, "failed to register netdev rc=%d\n", rc);
-		free_netdev(netdev);
-		return rc;
+		goto ibmvnic_register_fail;
 	}
 	dev_info(&dev->dev, "ibmvnic registered\n");
 
-	adapter->state = VNIC_PROBED;
+	rc = ibmvnic_cpu_notif_add(adapter);
+	if (rc) {
+		netdev_err(netdev, "Registering cpu notifier failed\n");
+		goto cpu_notif_add_failed;
+	}
+
+	complete(&adapter->probe_done);
+
 	return 0;
+
+cpu_notif_add_failed:
+	unregister_netdev(netdev);
+
+ibmvnic_register_fail:
+	device_remove_file(&dev->dev, &dev_attr_failover);
+
+ibmvnic_dev_file_err:
+	release_stats_token(adapter);
+
+ibmvnic_stats_fail:
+	release_stats_buffers(adapter);
+
+ibmvnic_init_fail:
+	release_sub_crqs(adapter, 1);
+	release_crq_queue(adapter);
+
+	/* cleanup worker thread after releasing CRQ so we don't get
+	 * transport events (i.e new work items for the worker thread).
+	 */
+	adapter->state = VNIC_REMOVING;
+	complete(&adapter->probe_done);
+	flush_work(&adapter->ibmvnic_reset);
+	flush_delayed_work(&adapter->ibmvnic_delayed_reset);
+
+	flush_reset_queue(adapter);
+
+	mutex_destroy(&adapter->fw_lock);
+	free_netdev(netdev);
+
+	return rc;
 }
 
-static int ibmvnic_remove(struct vio_dev *dev)
+static void ibmvnic_remove(struct vio_dev *dev)
 {
 	struct net_device *netdev = dev_get_drvdata(&dev->dev);
 	struct ibmvnic_adapter *adapter = netdev_priv(netdev);
+	unsigned long flags;
+
+	spin_lock_irqsave(&adapter->state_lock, flags);
 
+	/* If ibmvnic_reset() is scheduling a reset, wait for it to
+	 * finish. Then, set the state to REMOVING to prevent it from
+	 * scheduling any more work and to have reset functions ignore
+	 * any resets that have already been scheduled. Drop the lock
+	 * after setting state, so __ibmvnic_reset() which is called
+	 * from the flush_work() below, can make progress.
+	 */
+	spin_lock(&adapter->rwi_lock);
 	adapter->state = VNIC_REMOVING;
-	unregister_netdev(netdev);
-	mutex_lock(&adapter->reset_lock);
+	spin_unlock(&adapter->rwi_lock);
+
+	spin_unlock_irqrestore(&adapter->state_lock, flags);
+
+	ibmvnic_cpu_notif_remove(adapter);
+
+	flush_work(&adapter->ibmvnic_reset);
+	flush_delayed_work(&adapter->ibmvnic_delayed_reset);
+
+	rtnl_lock();
+	unregister_netdevice(netdev);
 
 	release_resources(adapter);
-	release_sub_crqs(adapter);
+	release_rx_pools(adapter);
+	release_tx_pools(adapter);
+	release_sub_crqs(adapter, 1);
 	release_crq_queue(adapter);
 
+	release_stats_token(adapter);
+	release_stats_buffers(adapter);
+
 	adapter->state = VNIC_REMOVED;
 
-	mutex_unlock(&adapter->reset_lock);
+	rtnl_unlock();
+	mutex_destroy(&adapter->fw_lock);
+	device_remove_file(&dev->dev, &dev_attr_failover);
 	free_netdev(netdev);
 	dev_set_drvdata(&dev->dev, NULL);
+}
 
-	return 0;
+static ssize_t failover_store(struct device *dev, struct device_attribute *attr,
+			      const char *buf, size_t count)
+{
+	struct net_device *netdev = dev_get_drvdata(dev);
+	struct ibmvnic_adapter *adapter = netdev_priv(netdev);
+	unsigned long retbuf[PLPAR_HCALL_BUFSIZE];
+	__be64 session_token;
+	long rc;
+
+	if (!sysfs_streq(buf, "1"))
+		return -EINVAL;
+
+	rc = plpar_hcall(H_VIOCTL, retbuf, adapter->vdev->unit_address,
+			 H_GET_SESSION_TOKEN, 0, 0, 0);
+	if (rc) {
+		netdev_err(netdev, "Couldn't retrieve session token, rc %ld\n",
+			   rc);
+		goto last_resort;
+	}
+
+	session_token = (__be64)retbuf[0];
+	netdev_dbg(netdev, "Initiating client failover, session id %llx\n",
+		   be64_to_cpu(session_token));
+	rc = plpar_hcall_norets(H_VIOCTL, adapter->vdev->unit_address,
+				H_SESSION_ERR_DETECTED, session_token, 0, 0);
+	if (rc) {
+		netdev_err(netdev,
+			   "H_VIOCTL initiated failover failed, rc %ld\n",
+			   rc);
+		goto last_resort;
+	}
+
+	return count;
+
+last_resort:
+	netdev_dbg(netdev, "Trying to send CRQ_CMD, the last resort\n");
+	ibmvnic_reset(adapter, VNIC_RESET_FAILOVER);
+
+	return count;
 }
+static DEVICE_ATTR_WO(failover);
 
 static unsigned long ibmvnic_get_desired_dma(struct vio_dev *vdev)
 {
@@ -3768,8 +6726,7 @@ static unsigned long ibmvnic_get_desired_dma(struct vio_dev *vdev)
 	for (i = 0; i < adapter->req_tx_queues + adapter->req_rx_queues; i++)
 		ret += 4 * PAGE_SIZE; /* the scrq message queue */
 
-	for (i = 0; i < be32_to_cpu(adapter->login_rsp_buf->num_rxadd_subcrqs);
-	     i++)
+	for (i = 0; i < adapter->num_active_rx_pools; i++)
 		ret += adapter->rx_pool[i].size *
 		    IOMMU_PAGE_ALIGN(adapter->rx_pool[i].buff_size, tbl);
 
@@ -3780,17 +6737,16 @@ static int ibmvnic_resume(struct device *dev)
 {
 	struct net_device *netdev = dev_get_drvdata(dev);
 	struct ibmvnic_adapter *adapter = netdev_priv(netdev);
-	int i;
 
-	/* kick the interrupt handlers just in case we lost an interrupt */
-	for (i = 0; i < adapter->req_rx_queues; i++)
-		ibmvnic_interrupt_rx(adapter->rx_scrq[i]->irq,
-				     adapter->rx_scrq[i]);
+	if (adapter->state != VNIC_OPEN)
+		return 0;
+
+	tasklet_schedule(&adapter->tasklet);
 
 	return 0;
 }
 
-static struct vio_device_id ibmvnic_device_table[] = {
+static const struct vio_device_id ibmvnic_device_table[] = {
 	{"network", "IBM,vnic"},
 	{"", "" }
 };
@@ -3812,15 +6768,40 @@ static struct vio_driver ibmvnic_driver = {
 /* module functions */
 static int __init ibmvnic_module_init(void)
 {
+	int ret;
+
+	ret = cpuhp_setup_state_multi(CPUHP_AP_ONLINE_DYN, "net/ibmvnic:online",
+				      ibmvnic_cpu_online,
+				      ibmvnic_cpu_down_prep);
+	if (ret < 0)
+		goto out;
+	ibmvnic_online = ret;
+	ret = cpuhp_setup_state_multi(CPUHP_IBMVNIC_DEAD, "net/ibmvnic:dead",
+				      NULL, ibmvnic_cpu_dead);
+	if (ret)
+		goto err_dead;
+
+	ret = vio_register_driver(&ibmvnic_driver);
+	if (ret)
+		goto err_vio_register;
+
 	pr_info("%s: %s %s\n", ibmvnic_driver_name, ibmvnic_driver_string,
 		IBMVNIC_DRIVER_VERSION);
 
-	return vio_register_driver(&ibmvnic_driver);
+	return 0;
+err_vio_register:
+	cpuhp_remove_multi_state(CPUHP_IBMVNIC_DEAD);
+err_dead:
+	cpuhp_remove_multi_state(ibmvnic_online);
+out:
+	return ret;
 }
 
 static void __exit ibmvnic_module_exit(void)
 {
 	vio_unregister_driver(&ibmvnic_driver);
+	cpuhp_remove_multi_state(CPUHP_IBMVNIC_DEAD);
+	cpuhp_remove_multi_state(ibmvnic_online);
 }
 
 module_init(ibmvnic_module_init);
diff --git a/drivers/net/ethernet/ibm/ibmvnic.h b/drivers/net/ethernet/ibm/ibmvnic.h
index 7e2300e64a47..480dc587078f 100644
--- a/drivers/net/ethernet/ibm/ibmvnic.h
+++ b/drivers/net/ethernet/ibm/ibmvnic.h
@@ -1,3 +1,4 @@
+/* SPDX-License-Identifier: GPL-2.0-or-later */
 /**************************************************************************/
 /*                                                                        */
 /*  IBM System i and System p Virtual NIC Device Driver                   */
@@ -6,18 +7,6 @@
 /*  Thomas Falcon (tlfalcon@linux.vnet.ibm.com)                           */
 /*  John Allen (jallen@linux.vnet.ibm.com)                                */
 /*                                                                        */
-/*  This program is free software; you can redistribute it and/or modify  */
-/*  it under the terms of the GNU General Public License as published by  */
-/*  the Free Software Foundation; either version 2 of the License, or     */
-/*  (at your option) any later version.                                   */
-/*                                                                        */
-/*  This program is distributed in the hope that it will be useful,       */
-/*  but WITHOUT ANY WARRANTY; without even the implied warranty of        */
-/*  MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the         */
-/*  GNU General Public License for more details.                          */
-/*                                                                        */
-/*  You should have received a copy of the GNU General Public License     */
-/*  along with this program.                                              */
 /*                                                                        */
 /* This module contains the implementation of a virtual ethernet device   */
 /* for use with IBM i/pSeries LPAR Linux.  It utilizes the logical LAN    */
@@ -29,7 +18,8 @@
 #define IBMVNIC_NAME		"ibmvnic"
 #define IBMVNIC_DRIVER_VERSION	"1.0.1"
 #define IBMVNIC_INVALID_MAP	-1
-#define IBMVNIC_STATS_TIMEOUT	1
+#define IBMVNIC_OPEN_FAILED	3
+
 /* basic structures plus 100 2k buffers */
 #define IBMVNIC_IO_ENTITLEMENT_DEFAULT	610305
 
@@ -37,7 +27,61 @@
 #define IBMVNIC_RX_WEIGHT		16
 /* when changing this, update IBMVNIC_IO_ENTITLEMENT_DEFAULT */
 #define IBMVNIC_BUFFS_PER_POOL	100
-#define IBMVNIC_MAX_TX_QUEUES	5
+#define IBMVNIC_MAX_QUEUES	16
+#define IBMVNIC_MAX_QUEUE_SZ   4096
+#define IBMVNIC_MAX_IND_DESCS 128
+#define IBMVNIC_SAFE_IND_DESC 16
+#define IBMVNIC_IND_MAX_ARR_SZ (IBMVNIC_MAX_IND_DESCS * 32)
+
+#define IBMVNIC_TSO_BUF_SZ	65536
+#define IBMVNIC_TSO_BUFS	64
+#define IBMVNIC_TSO_POOL_MASK	0x80000000
+
+/* A VNIC adapter has set of Rx and Tx pools (aka queues). Each Rx/Tx pool
+ * has a set of buffers. The size of each buffer is determined by the MTU.
+ *
+ * Each Rx/Tx pool is also associated with a DMA region that is shared
+ * with the "hardware" (VIOS) and used to send/receive packets. The DMA
+ * region is also referred to as a Long Term Buffer or LTB.
+ *
+ * The size of the DMA region required for an Rx/Tx pool depends on the
+ * number and size (MTU) of the buffers in the pool. At the max levels
+ * of 4096 jumbo frames (MTU=9000) we will need about 9K*4K = 36MB plus
+ * some padding.
+ *
+ * But the size of a single DMA region is limited by MAX_PAGE_ORDER in the
+ * kernel (about 16MB currently).  To support say 4K Jumbo frames, we
+ * use a set of LTBs (struct ltb_set) per pool.
+ *
+ * IBMVNIC_ONE_LTB_MAX  - max size of each LTB supported by kernel
+ * IBMVNIC_ONE_LTB_SIZE - current max size of each LTB in an ltb_set
+ * (must be <= IBMVNIC_ONE_LTB_MAX)
+ * IBMVNIC_LTB_SET_SIZE - current size of all LTBs in an ltb_set
+ *
+ * Each VNIC can have upto 16 Rx, 16 Tx and 16 TSO pools. The TSO pools
+ * are of fixed length (IBMVNIC_TSO_BUF_SZ * IBMVNIC_TSO_BUFS) of 4MB.
+ *
+ * The Rx and Tx pools can have upto 4096 buffers. The max size of these
+ * buffers is about 9588 (for jumbo frames, including IBMVNIC_BUFFER_HLEN).
+ * So, setting the IBMVNIC_LTB_SET_SIZE for a pool to 4096 * 9588 ~= 38MB.
+ *
+ * There is a trade-off in setting IBMVNIC_ONE_LTB_SIZE. If it is large,
+ * the allocation of the LTB can fail when system is low in memory. If
+ * its too small, we would need several mappings for each of the Rx/
+ * Tx/TSO pools but there is a limit of 255 mappings per vnic in the
+ * VNIC protocol.
+ *
+ * So setting IBMVNIC_ONE_LTB_SIZE to 8MB. With IBMVNIC_LTB_SET_SIZE set
+ * to 38MB, we will need 5 LTBs per Rx and Tx pool and 1 LTB per TSO
+ * pool for the 4MB. Thus the 16 Rx and Tx queues require 32 * 5 = 160
+ * plus 16 for the TSO pools for a total of 176 LTB mappings per VNIC.
+ */
+#define IBMVNIC_ONE_LTB_MAX	((u32)((1 << MAX_PAGE_ORDER) * PAGE_SIZE))
+#define IBMVNIC_ONE_LTB_SIZE	min((u32)(8 << 20), IBMVNIC_ONE_LTB_MAX)
+#define IBMVNIC_LTB_SET_SIZE	(38 << 20)
+
+#define IBMVNIC_BUFFER_HLEN		500
+#define IBMVNIC_RESET_DELAY 100
 
 struct ibmvnic_login_buffer {
 	__be32 len;
@@ -49,6 +93,8 @@ struct ibmvnic_login_buffer {
 	__be32 off_rxcomp_subcrqs;
 	__be32 login_rsp_ioba;
 	__be32 login_rsp_len;
+	__be32 client_data_offset;
+	__be32 client_data_len;
 } __packed __aligned(8);
 
 struct ibmvnic_login_rsp_buffer {
@@ -166,6 +212,25 @@ struct ibmvnic_statistics {
 	u8 reserved[72];
 } __packed __aligned(8);
 
+struct ibmvnic_tx_queue_stats {
+	u64 batched_packets;
+	u64 direct_packets;
+	u64 bytes;
+	u64 dropped_packets;
+};
+
+#define NUM_TX_STATS \
+	(sizeof(struct ibmvnic_tx_queue_stats) / sizeof(u64))
+
+struct ibmvnic_rx_queue_stats {
+	u64 packets;
+	u64 bytes;
+	u64 interrupts;
+};
+
+#define NUM_RX_STATS \
+	(sizeof(struct ibmvnic_rx_queue_stats) / sizeof(u64))
+
 struct ibmvnic_acl_buffer {
 	__be32 len;
 	__be32 version;
@@ -201,8 +266,6 @@ struct ibmvnic_tx_comp_desc {
 #define IBMVNIC_TCP_CHKSUM		0x20
 #define IBMVNIC_UDP_CHKSUM		0x08
 
-#define IBMVNIC_MAX_FRAGS_PER_CRQ 3
-
 struct ibmvnic_tx_desc {
 	u8 first;
 	u8 type;
@@ -346,11 +409,16 @@ struct ibmvnic_phys_parms {
 	u8 flags2;
 #define IBMVNIC_LOGICAL_LNK_ACTIVE 0x80
 	__be32 speed;
-#define IBMVNIC_AUTONEG		0x80
-#define IBMVNIC_10MBPS		0x40
-#define IBMVNIC_100MBPS		0x20
-#define IBMVNIC_1GBPS		0x10
-#define IBMVNIC_10GBPS		0x08
+#define IBMVNIC_AUTONEG		0x80000000
+#define IBMVNIC_10MBPS		0x40000000
+#define IBMVNIC_100MBPS		0x20000000
+#define IBMVNIC_1GBPS		0x10000000
+#define IBMVNIC_10GBPS		0x08000000
+#define IBMVNIC_40GBPS		0x04000000
+#define IBMVNIC_100GBPS		0x02000000
+#define IBMVNIC_25GBPS		0x01000000
+#define IBMVNIC_50GBPS		0x00800000
+#define IBMVNIC_200GBPS		0x00400000
 	__be32 mtu;
 	struct ibmvnic_rc rc;
 } __packed __aligned(8);
@@ -384,77 +452,6 @@ struct ibmvnic_control_ip_offload {
 	struct ibmvnic_rc rc;
 } __packed __aligned(8);
 
-struct ibmvnic_request_dump_size {
-	u8 first;
-	u8 cmd;
-	u8 reserved[6];
-	__be32 len;
-	struct ibmvnic_rc rc;
-} __packed __aligned(8);
-
-struct ibmvnic_request_dump {
-	u8 first;
-	u8 cmd;
-	u8 reserved1[2];
-	__be32 ioba;
-	__be32 len;
-	u8 reserved2[4];
-} __packed __aligned(8);
-
-struct ibmvnic_request_dump_rsp {
-	u8 first;
-	u8 cmd;
-	u8 reserved[6];
-	__be32 dumped_len;
-	struct ibmvnic_rc rc;
-} __packed __aligned(8);
-
-struct ibmvnic_request_ras_comp_num {
-	u8 first;
-	u8 cmd;
-	u8 reserved1[2];
-	__be32 num_components;
-	u8 reserved2[4];
-	struct ibmvnic_rc rc;
-} __packed __aligned(8);
-
-struct ibmvnic_request_ras_comps {
-	u8 first;
-	u8 cmd;
-	u8 reserved[2];
-	__be32 ioba;
-	__be32 len;
-	struct ibmvnic_rc rc;
-} __packed __aligned(8);
-
-struct ibmvnic_control_ras {
-	u8 first;
-	u8 cmd;
-	u8 correlator;
-	u8 level;
-	u8 op;
-#define IBMVNIC_TRACE_LEVEL	1
-#define IBMVNIC_ERROR_LEVEL	2
-#define IBMVNIC_TRACE_PAUSE	3
-#define IBMVNIC_TRACE_RESUME	4
-#define IBMVNIC_TRACE_ON		5
-#define IBMVNIC_TRACE_OFF		6
-#define IBMVNIC_CHG_TRACE_BUFF_SZ	7
-	u8 trace_buff_sz[3];
-	u8 reserved[4];
-	struct ibmvnic_rc rc;
-} __packed __aligned(8);
-
-struct ibmvnic_collect_fw_trace {
-	u8 first;
-	u8 cmd;
-	u8 correlator;
-	u8 reserved;
-	__be32 ioba;
-	__be32 len;
-	struct ibmvnic_rc rc;
-} __packed __aligned(8);
-
 struct ibmvnic_request_statistics {
 	u8 first;
 	u8 cmd;
@@ -466,15 +463,6 @@ struct ibmvnic_request_statistics {
 	u8 reserved[4];
 } __packed __aligned(8);
 
-struct ibmvnic_request_debug_stats {
-	u8 first;
-	u8 cmd;
-	u8 reserved[2];
-	__be32 ioba;
-	__be32 len;
-	struct ibmvnic_rc rc;
-} __packed __aligned(8);
-
 struct ibmvnic_error_indication {
 	u8 first;
 	u8 cmd;
@@ -487,24 +475,6 @@ struct ibmvnic_error_indication {
 	u8 reserved2[2];
 } __packed __aligned(8);
 
-struct ibmvnic_request_error_info {
-	u8 first;
-	u8 cmd;
-	u8 reserved[2];
-	__be32 ioba;
-	__be32 len;
-	__be32 error_id;
-} __packed __aligned(8);
-
-struct ibmvnic_request_error_rsp {
-	u8 first;
-	u8 cmd;
-	u8 reserved[2];
-	__be32 error_id;
-	__be32 len;
-	struct ibmvnic_rc rc;
-} __packed __aligned(8);
-
 struct ibmvnic_link_state_indication {
 	u8 first;
 	u8 cmd;
@@ -536,6 +506,12 @@ struct ibmvnic_multicast_ctrl {
 	struct ibmvnic_rc rc;
 } __packed __aligned(8);
 
+struct ibmvnic_get_vpd_size {
+	u8 first;
+	u8 cmd;
+	u8 reserved[14];
+} __packed __aligned(8);
+
 struct ibmvnic_get_vpd_size_rsp {
 	u8 first;
 	u8 cmd;
@@ -553,6 +529,13 @@ struct ibmvnic_get_vpd {
 	u8 reserved[4];
 } __packed __aligned(8);
 
+struct ibmvnic_get_vpd_rsp {
+	u8 first;
+	u8 cmd;
+	u8 reserved[10];
+	struct ibmvnic_rc rc;
+} __packed __aligned(8);
+
 struct ibmvnic_acl_change_indication {
 	u8 first;
 	u8 cmd;
@@ -595,7 +578,7 @@ struct ibmvnic_request_map_rsp {
 	u8 cmd;
 	u8 reserved1;
 	u8 map_id;
-	u8 reserved2[4];
+	u8 reserved2[8];
 	struct ibmvnic_rc rc;
 } __packed __aligned(8);
 
@@ -654,34 +637,18 @@ union ibmvnic_crq {
 	struct ibmvnic_query_ip_offload query_ip_offload_rsp;
 	struct ibmvnic_control_ip_offload control_ip_offload;
 	struct ibmvnic_control_ip_offload control_ip_offload_rsp;
-	struct ibmvnic_request_dump_size request_dump_size;
-	struct ibmvnic_request_dump_size request_dump_size_rsp;
-	struct ibmvnic_request_dump request_dump;
-	struct ibmvnic_request_dump_rsp request_dump_rsp;
-	struct ibmvnic_request_ras_comp_num request_ras_comp_num;
-	struct ibmvnic_request_ras_comp_num request_ras_comp_num_rsp;
-	struct ibmvnic_request_ras_comps request_ras_comps;
-	struct ibmvnic_request_ras_comps request_ras_comps_rsp;
-	struct ibmvnic_control_ras control_ras;
-	struct ibmvnic_control_ras control_ras_rsp;
-	struct ibmvnic_collect_fw_trace collect_fw_trace;
-	struct ibmvnic_collect_fw_trace collect_fw_trace_rsp;
 	struct ibmvnic_request_statistics request_statistics;
 	struct ibmvnic_generic_crq request_statistics_rsp;
-	struct ibmvnic_request_debug_stats request_debug_stats;
-	struct ibmvnic_request_debug_stats request_debug_stats_rsp;
 	struct ibmvnic_error_indication error_indication;
-	struct ibmvnic_request_error_info request_error_info;
-	struct ibmvnic_request_error_rsp request_error_rsp;
 	struct ibmvnic_link_state_indication link_state_indication;
 	struct ibmvnic_change_mac_addr change_mac_addr;
 	struct ibmvnic_change_mac_addr change_mac_addr_rsp;
 	struct ibmvnic_multicast_ctrl multicast_ctrl;
 	struct ibmvnic_multicast_ctrl multicast_ctrl_rsp;
-	struct ibmvnic_generic_crq get_vpd_size;
+	struct ibmvnic_get_vpd_size get_vpd_size;
 	struct ibmvnic_get_vpd_size_rsp get_vpd_size_rsp;
 	struct ibmvnic_get_vpd get_vpd;
-	struct ibmvnic_generic_crq get_vpd_rsp;
+	struct ibmvnic_get_vpd_rsp get_vpd_rsp;
 	struct ibmvnic_acl_change_indication acl_change_indication;
 	struct ibmvnic_acl_query acl_query;
 	struct ibmvnic_generic_crq acl_query_rsp;
@@ -771,8 +738,6 @@ enum ibmvnic_commands {
 	SET_PHYS_PARMS = 0x07,
 	SET_PHYS_PARMS_RSP = 0x87,
 	ERROR_INDICATION = 0x08,
-	REQUEST_ERROR_INFO = 0x09,
-	REQUEST_ERROR_RSP = 0x89,
 	LOGICAL_LINK_STATE = 0x0C,
 	LOGICAL_LINK_STATE_RSP = 0x8C,
 	REQUEST_STATISTICS = 0x0D,
@@ -826,7 +791,10 @@ struct ibmvnic_crq_queue {
 	union ibmvnic_crq *msgs;
 	int size, cur;
 	dma_addr_t msg_token;
+	/* Used for serialization of msgs, cur */
 	spinlock_t lock;
+	bool active;
+	char name[32];
 };
 
 union sub_crq {
@@ -840,6 +808,12 @@ union sub_crq {
 	struct ibmvnic_rx_buff_add_desc rx_add;
 };
 
+struct ibmvnic_ind_xmit_queue {
+	union sub_crq *indir_arr;
+	dma_addr_t indir_dma;
+	int index;
+};
+
 struct ibmvnic_sub_crq_queue {
 	union sub_crq *msgs;
 	int size, cur;
@@ -849,11 +823,16 @@ struct ibmvnic_sub_crq_queue {
 	unsigned int irq;
 	unsigned int pool_index;
 	int scrq_num;
+	/* Used for serialization of msgs, cur */
 	spinlock_t lock;
 	struct sk_buff *rx_skb_top;
 	struct ibmvnic_adapter *adapter;
+	struct ibmvnic_ind_xmit_queue ind_buf;
 	atomic_t used;
-};
+	char name[32];
+	u64 handle;
+	cpumask_var_t affinity_mask;
+} ____cacheline_aligned;
 
 struct ibmvnic_long_term_buff {
 	unsigned char *buff;
@@ -862,16 +841,16 @@ struct ibmvnic_long_term_buff {
 	u8 map_id;
 };
 
+struct ibmvnic_ltb_set {
+	int num_ltbs;
+	struct ibmvnic_long_term_buff *ltbs;
+};
+
 struct ibmvnic_tx_buff {
 	struct sk_buff *skb;
-	dma_addr_t data_dma[IBMVNIC_MAX_FRAGS_PER_CRQ];
-	unsigned int data_len[IBMVNIC_MAX_FRAGS_PER_CRQ];
 	int index;
 	int pool_index;
-	bool last_frag;
-	union sub_crq indir_arr[6];
-	u8 hdr_data[140];
-	dma_addr_t indir_dma;
+	int num_entries;
 };
 
 struct ibmvnic_tx_pool {
@@ -879,10 +858,10 @@ struct ibmvnic_tx_pool {
 	int *free_map;
 	int consumer_index;
 	int producer_index;
-	wait_queue_head_t ibmvnic_tx_comp_q;
-	struct task_struct *work_thread;
-	struct ibmvnic_long_term_buff long_term_buff;
-};
+	struct ibmvnic_ltb_set ltb_set;
+	int num_buffers;
+	int buf_size;
+} ____cacheline_aligned;
 
 struct ibmvnic_rx_buff {
 	struct sk_buff *skb;
@@ -894,7 +873,7 @@ struct ibmvnic_rx_buff {
 
 struct ibmvnic_rx_pool {
 	struct ibmvnic_rx_buff *rx_buff;
-	int size;
+	int size;			/* # of buffers in the pool */
 	int index;
 	int buff_size;
 	atomic_t available;
@@ -902,15 +881,13 @@ struct ibmvnic_rx_pool {
 	int next_free;
 	int next_alloc;
 	int active;
-	struct ibmvnic_long_term_buff long_term_buff;
-};
+	struct ibmvnic_ltb_set ltb_set;
+} ____cacheline_aligned;
 
-struct ibmvnic_error_buff {
-	char *buff;
-	dma_addr_t dma;
-	int len;
-	struct list_head list;
-	__be32 error_id;
+struct ibmvnic_vpd {
+	unsigned char *buff;
+	dma_addr_t dma_addr;
+	u64 len;
 };
 
 enum vnic_state {VNIC_PROBING = 1,
@@ -920,19 +897,30 @@ enum vnic_state {VNIC_PROBING = 1,
 		 VNIC_CLOSING,
 		 VNIC_CLOSED,
 		 VNIC_REMOVING,
-		 VNIC_REMOVED};
+		 VNIC_REMOVED,
+		 VNIC_DOWN};
 
 enum ibmvnic_reset_reason {VNIC_RESET_FAILOVER = 1,
 			   VNIC_RESET_MOBILITY,
 			   VNIC_RESET_FATAL,
 			   VNIC_RESET_NON_FATAL,
-			   VNIC_RESET_TIMEOUT};
+			   VNIC_RESET_TIMEOUT,
+			   VNIC_RESET_CHANGE_PARAM,
+			   VNIC_RESET_PASSIVE_INIT};
 
 struct ibmvnic_rwi {
 	enum ibmvnic_reset_reason reset_reason;
 	struct list_head list;
 };
 
+struct ibmvnic_tunables {
+	u64 rx_queues;
+	u64 tx_queues;
+	u64 rx_entries;
+	u64 tx_entries;
+	u64 mtu;
+};
+
 struct ibmvnic_adapter {
 	struct vio_dev *vdev;
 	struct net_device *netdev;
@@ -943,12 +931,16 @@ struct ibmvnic_adapter {
 	struct ibmvnic_control_ip_offload_buffer ip_offload_ctrl;
 	dma_addr_t ip_offload_ctrl_tok;
 	u32 msg_enable;
+	u32 cur_max_ind_descs;
+
+	/* Vital Product Data (VPD) */
+	struct ibmvnic_vpd *vpd;
+	char fw_version[32];
 
 	/* Statistics */
 	struct ibmvnic_statistics stats;
 	dma_addr_t stats_token;
 	struct completion stats_done;
-	spinlock_t stats_lock;
 	int replenish_no_mem;
 	int replenish_add_buff_success;
 	int replenish_add_buff_failure;
@@ -956,9 +948,15 @@ struct ibmvnic_adapter {
 	int tx_send_failed;
 	int tx_map_failed;
 
+	struct ibmvnic_tx_queue_stats *tx_stats_buffers;
+	struct ibmvnic_rx_queue_stats *rx_stats_buffers;
+
 	int phys_link_state;
 	int logical_link_state;
 
+	u32 speed;
+	u8 duplex;
+
 	/* login data */
 	struct ibmvnic_login_buffer *login_buf;
 	dma_addr_t login_buf_token;
@@ -969,11 +967,9 @@ struct ibmvnic_adapter {
 	int login_rsp_buf_sz;
 
 	atomic_t running_cap_crqs;
-	bool wait_capability;
 
-	struct ibmvnic_sub_crq_queue **tx_scrq;
-	struct ibmvnic_sub_crq_queue **rx_scrq;
-	bool renegotiate;
+	struct ibmvnic_sub_crq_queue **tx_scrq ____cacheline_aligned;
+	struct ibmvnic_sub_crq_queue **rx_scrq ____cacheline_aligned;
 
 	/* rx structs */
 	struct napi_struct *napi;
@@ -981,13 +977,23 @@ struct ibmvnic_adapter {
 	u64 promisc;
 
 	struct ibmvnic_tx_pool *tx_pool;
+	struct ibmvnic_tx_pool *tso_pool;
+	struct completion probe_done;
 	struct completion init_done;
 	int init_done_rc;
 
-	struct list_head errors;
-	spinlock_t error_list_lock;
-
 	struct completion fw_done;
+	/* Used for serialization of device commands */
+	struct mutex fw_lock;
+	int fw_done_rc;
+
+	struct completion reset_done;
+	int reset_done_rc;
+	bool wait_for_reset;
+
+	/* CPU hotplug instances for online & dead */
+	struct hlist_node node;
+	struct hlist_node node_dead;
 
 	/* partner capabilities */
 	u64 min_tx_queues;
@@ -1011,6 +1017,7 @@ struct ibmvnic_adapter {
 	u64 min_mtu;
 	u64 max_mtu;
 	u64 req_mtu;
+	u64 prev_mtu;
 	u64 max_multicast_filters;
 	u64 vlan_header_insertion;
 	u64 rx_vlan_header_insertion;
@@ -1023,14 +1030,45 @@ struct ibmvnic_adapter {
 	u64 opt_tx_entries_per_subcrq;
 	u64 opt_rxba_entries_per_subcrq;
 	__be64 tx_rx_desc_req;
-	u8 map_id;
+#define MAX_MAP_ID	255
+	DECLARE_BITMAP(map_ids, MAX_MAP_ID);
+	u32 num_active_rx_scrqs;
+	u32 num_active_rx_pools;
+	u32 num_active_rx_napi;
+	u32 num_active_tx_scrqs;
+	u32 num_active_tx_pools;
+
+	u32 prev_rx_pool_size;
+	u32 prev_tx_pool_size;
+	u32 cur_rx_buf_sz;
+	u32 prev_rx_buf_sz;
 
 	struct tasklet_struct tasklet;
 	enum vnic_state state;
+	/* Used for serialization of state field. When taking both state
+	 * and rwi locks, take state lock first.
+	 */
+	spinlock_t state_lock;
 	enum ibmvnic_reset_reason reset_reason;
-	struct mutex reset_lock, rwi_lock;
 	struct list_head rwi_list;
+	/* Used for serialization of rwi_list. When taking both state
+	 * and rwi locks, take state lock first
+	 */
+	spinlock_t rwi_lock;
 	struct work_struct ibmvnic_reset;
-	bool resetting;
-	bool napi_enabled, from_passive_init;
+	struct delayed_work ibmvnic_delayed_reset;
+	unsigned long resetting;
+	/* last device reset time */
+	unsigned long last_reset_time;
+
+	bool napi_enabled;
+	bool from_passive_init;
+	bool login_pending;
+	/* protected by rcu */
+	bool tx_queues_active;
+	bool failover_pending;
+	bool force_reset_recovery;
+
+	struct ibmvnic_tunables desired;
+	struct ibmvnic_tunables fallback;
 };
diff --git a/drivers/net/ethernet/intel/Kconfig b/drivers/net/ethernet/intel/Kconfig
index 1542a2158e96..a563a94e2780 100644
--- a/drivers/net/ethernet/intel/Kconfig
+++ b/drivers/net/ethernet/intel/Kconfig
@@ -1,3 +1,4 @@
+# SPDX-License-Identifier: GPL-2.0-only
 #
 # Intel network device configuration
 #
@@ -5,7 +6,7 @@
 config NET_VENDOR_INTEL
 	bool "Intel devices"
 	default y
-	---help---
+	help
 	  If you have a network (Ethernet) card belonging to this class, say Y.
 
 	  Note that the answer to this question doesn't directly affect the
@@ -15,11 +16,14 @@ config NET_VENDOR_INTEL
 
 if NET_VENDOR_INTEL
 
+source "drivers/net/ethernet/intel/libeth/Kconfig"
+source "drivers/net/ethernet/intel/libie/Kconfig"
+
 config E100
 	tristate "Intel(R) PRO/100+ support"
 	depends on PCI
 	select MII
-	---help---
+	help
 	  This driver supports Intel(R) PRO/100 family of adapters.
 	  To verify that your adapter is supported, find the board ID number
 	  on the adapter. Look for a label that has a barcode and a number
@@ -33,15 +37,15 @@ config E100
 	  to identify the adapter.
 
 	  More specific information on configuring the driver is in
-	  <file:Documentation/networking/e100.txt>.
+	  <file:Documentation/networking/device_drivers/ethernet/intel/e100.rst>.
 
 	  To compile this driver as a module, choose M here. The module
 	  will be called e100.
 
 config E1000
 	tristate "Intel(R) PRO/1000 Gigabit Ethernet support"
-	depends on PCI
-	---help---
+	depends on PCI && HAS_IOPORT
+	help
 	  This driver supports Intel(R) PRO/1000 gigabit ethernet family of
 	  adapters.  For more information on how to identify your adapter, go
 	  to the Adapter & Driver ID Guide that can be located at:
@@ -49,7 +53,7 @@ config E1000
 	  <http://support.intel.com>
 
 	  More specific information on configuring the driver is in
-	  <file:Documentation/networking/e1000.txt>.
+	  <file:Documentation/networking/device_drivers/ethernet/intel/e1000.rst>.
 
 	  To compile this driver as a module, choose M here. The module
 	  will be called e1000.
@@ -57,9 +61,9 @@ config E1000
 config E1000E
 	tristate "Intel(R) PRO/1000 PCI-Express Gigabit Ethernet support"
 	depends on PCI && (!SPARC32 || BROKEN)
+	depends on PTP_1588_CLOCK_OPTIONAL
 	select CRC32
-	imply PTP_1588_CLOCK
-	---help---
+	help
 	  This driver supports the PCI-Express Intel(R) PRO/1000 gigabit
 	  ethernet family of adapters. For PCI or PCI-X e1000 adapters,
 	  use the regular e1000 driver For more information on how to
@@ -68,6 +72,9 @@ config E1000E
 
 	  <http://support.intel.com>
 
+	  More specific information on configuring the driver is in
+	  <file:Documentation/networking/device_drivers/ethernet/intel/e1000e.rst>.
+
 	  To compile this driver as a module, choose M here. The module
 	  will be called e1000e.
 
@@ -75,7 +82,7 @@ config E1000E_HWTS
 	bool "Support HW cross-timestamp on PCH devices"
 	default y
 	depends on E1000E && X86
-	---help---
+	help
 	 Say Y to enable hardware supported cross-timestamping on PCH
 	 devices. The cross-timestamp is available through the PTP clock
 	 driver precise cross-timestamp ioctl (PTP_SYS_OFFSET_PRECISE).
@@ -83,10 +90,10 @@ config E1000E_HWTS
 config IGB
 	tristate "Intel(R) 82575/82576 PCI-Express Gigabit Ethernet support"
 	depends on PCI
-	imply PTP_1588_CLOCK
+	depends on PTP_1588_CLOCK_OPTIONAL
 	select I2C
 	select I2C_ALGOBIT
-	---help---
+	help
 	  This driver supports Intel(R) 82575/82576 gigabit ethernet family of
 	  adapters.  For more information on how to identify your adapter, go
 	  to the Adapter & Driver ID Guide that can be located at:
@@ -94,7 +101,7 @@ config IGB
 	  <http://support.intel.com>
 
 	  More specific information on configuring the driver is in
-	  <file:Documentation/networking/e1000.txt>.
+	  <file:Documentation/networking/device_drivers/ethernet/intel/igb.rst>.
 
 	  To compile this driver as a module, choose M here. The module
 	  will be called igb.
@@ -103,7 +110,7 @@ config IGB_HWMON
 	bool "Intel(R) PCI-Express Gigabit adapters HWMON support"
 	default y
 	depends on IGB && HWMON && !(IGB=y && HWMON=m)
-	---help---
+	help
 	  Say Y if you want to expose thermal sensor data on Intel devices.
 
 	  Some of our devices contain thermal sensors, both external and internal.
@@ -114,7 +121,7 @@ config IGB_DCA
 	bool "Direct Cache Access (DCA) Support"
 	default y
 	depends on IGB && DCA && !(IGB=y && DCA=m)
-	---help---
+	help
 	  Say Y here if you want to use Direct Cache Access (DCA) in the
 	  driver.  DCA is a method for warming the CPU cache before data
 	  is used, with the intent of lessening the impact of cache misses.
@@ -122,7 +129,7 @@ config IGB_DCA
 config IGBVF
 	tristate "Intel(R) 82576 Virtual Function Ethernet support"
 	depends on PCI
-	---help---
+	help
 	  This driver supports Intel(R) 82576 virtual functions.  For more
 	  information on how to identify your adapter, go to the Adapter &
 	  Driver ID Guide that can be located at:
@@ -130,40 +137,30 @@ config IGBVF
 	  <http://support.intel.com>
 
 	  More specific information on configuring the driver is in
-	  <file:Documentation/networking/e1000.txt>.
+	  <file:Documentation/networking/device_drivers/ethernet/intel/igbvf.rst>.
 
 	  To compile this driver as a module, choose M here. The module
 	  will be called igbvf.
 
-config IXGB
-	tristate "Intel(R) PRO/10GbE support"
-	depends on PCI
-	---help---
-	  This driver supports Intel(R) PRO/10GbE family of adapters for
-	  PCI-X type cards. For PCI-E type cards, use the "ixgbe" driver
-	  instead. For more information on how to identify your adapter, go
-	  to the Adapter & Driver ID Guide that can be located at:
-
-	  <http://support.intel.com>
-
-	  More specific information on configuring the driver is in
-	  <file:Documentation/networking/ixgb.txt>.
-
-	  To compile this driver as a module, choose M here. The module
-	  will be called ixgb.
-
 config IXGBE
 	tristate "Intel(R) 10GbE PCI Express adapters support"
 	depends on PCI
+	depends on PTP_1588_CLOCK_OPTIONAL
+	select LIBIE_FWLOG
 	select MDIO
-	imply PTP_1588_CLOCK
-	---help---
+	select NET_DEVLINK
+	select PLDMFW
+	select PHYLIB
+	help
 	  This driver supports Intel(R) 10GbE PCI Express family of
 	  adapters.  For more information on how to identify your adapter, go
 	  to the Adapter & Driver ID Guide that can be located at:
 
 	  <http://support.intel.com>
 
+	  More specific information on configuring the driver is in
+	  <file:Documentation/networking/device_drivers/ethernet/intel/ixgbe.rst>.
+
 	  To compile this driver as a module, choose M here. The module
 	  will be called ixgbe.
 
@@ -171,7 +168,7 @@ config IXGBE_HWMON
 	bool "Intel(R) 10GbE PCI Express adapters HWMON support"
 	default y
 	depends on IXGBE && HWMON && !(IXGBE=y && HWMON=m)
-	---help---
+	help
 	  Say Y if you want to expose the thermal sensor data on some of
 	  our cards, via a hwmon sysfs interface.
 
@@ -179,7 +176,7 @@ config IXGBE_DCA
 	bool "Direct Cache Access (DCA) Support"
 	default y
 	depends on IXGBE && DCA && !(IXGBE=y && DCA=m)
-	---help---
+	help
 	  Say Y here if you want to use Direct Cache Access (DCA) in the
 	  driver.  DCA is a method for warming the CPU cache before data
 	  is used, with the intent of lessening the impact of cache misses.
@@ -188,16 +185,25 @@ config IXGBE_DCB
 	bool "Data Center Bridging (DCB) Support"
 	default n
 	depends on IXGBE && DCB
-	---help---
+	help
 	  Say Y here if you want to use Data Center Bridging (DCB) in the
 	  driver.
 
 	  If unsure, say N.
 
+config IXGBE_IPSEC
+	bool "IPSec XFRM cryptography-offload acceleration"
+	depends on IXGBE
+	depends on XFRM_OFFLOAD
+	default y
+	select XFRM_ALGO
+	help
+	  Enable support for IPSec offload in ixgbe.ko
+
 config IXGBEVF
 	tristate "Intel(R) 10GbE PCI Express Virtual Function Ethernet support"
 	depends on PCI_MSI
-	---help---
+	help
 	  This driver supports Intel(R) PCI Express virtual functions for the
 	  Intel(R) ixgbe driver.  For more information on how to identify your
 	  adapter, go to the Adapter & Driver ID Guide that can be located at:
@@ -205,23 +211,39 @@ config IXGBEVF
 	  <http://support.intel.com>
 
 	  More specific information on configuring the driver is in
-	  <file:Documentation/networking/ixgbevf.txt>.
+	  <file:Documentation/networking/device_drivers/ethernet/intel/ixgbevf.rst>.
 
 	  To compile this driver as a module, choose M here. The module
 	  will be called ixgbevf.  MSI-X interrupt support is required
 	  for this driver to work correctly.
 
+config IXGBEVF_IPSEC
+	bool "IPSec XFRM cryptography-offload acceleration"
+	depends on IXGBEVF
+	depends on XFRM_OFFLOAD
+	default y
+	select XFRM_ALGO
+	help
+	  Enable support for IPSec offload in ixgbevf.ko
+
 config I40E
 	tristate "Intel(R) Ethernet Controller XL710 Family support"
-	imply PTP_1588_CLOCK
+	depends on PTP_1588_CLOCK_OPTIONAL
 	depends on PCI
-	---help---
+	select AUXILIARY_BUS
+	select LIBIE
+	select LIBIE_ADMINQ
+	select NET_DEVLINK
+	help
 	  This driver supports Intel(R) Ethernet Controller XL710 Family of
 	  devices.  For more information on how to identify your adapter, go
 	  to the Adapter & Driver ID Guide that can be located at:
 
 	  <http://support.intel.com>
 
+	  More specific information on configuring the driver is in
+	  <file:Documentation/networking/device_drivers/ethernet/intel/i40e.rst>.
+
 	  To compile this driver as a module, choose M here. The module
 	  will be called i40e.
 
@@ -229,39 +251,150 @@ config I40E_DCB
 	bool "Data Center Bridging (DCB) Support"
 	default n
 	depends on I40E && DCB
-	---help---
+	help
 	  Say Y here if you want to use Data Center Bridging (DCB) in the
 	  driver.
 
 	  If unsure, say N.
 
+# this is here to allow seamless migration from I40EVF --> IAVF name
+# so that CONFIG_IAVF symbol will always mirror the state of CONFIG_I40EVF
+config IAVF
+	tristate
+	select LIBIE
+	select LIBIE_ADMINQ
+	select NET_SHAPER
+
 config I40EVF
-	tristate "Intel(R) XL710 X710 Virtual Function Ethernet support"
+	tristate "Intel(R) Ethernet Adaptive Virtual Function support"
+	select IAVF
 	depends on PCI_MSI
-	---help---
-	  This driver supports Intel(R) XL710 and X710 virtual functions.
-	  For more information on how to identify your adapter, go to the
-	  Adapter & Driver ID Guide that can be located at:
+	depends on PTP_1588_CLOCK_OPTIONAL
+	help
+	  This driver supports virtual functions for Intel XL710,
+	  X710, X722, XXV710, and all devices advertising support for
+	  Intel Ethernet Adaptive Virtual Function devices. For more
+	  information on how to identify your adapter, go to the Adapter
+	  & Driver ID Guide that can be located at:
 
-	  <http://support.intel.com>
+	  <https://support.intel.com>
+
+	  This driver was formerly named i40evf.
+
+	  More specific information on configuring the driver is in
+	  <file:Documentation/networking/device_drivers/ethernet/intel/iavf.rst>.
 
 	  To compile this driver as a module, choose M here. The module
-	  will be called i40evf.  MSI-X interrupt support is required
+	  will be called iavf.  MSI-X interrupt support is required
 	  for this driver to work correctly.
 
+config ICE
+	tristate "Intel(R) Ethernet Connection E800 Series Support"
+	default n
+	depends on PCI_MSI
+	depends on PTP_1588_CLOCK_OPTIONAL
+	depends on GNSS || GNSS = n
+	select AUXILIARY_BUS
+	select DIMLIB
+	select LIBIE
+	select LIBIE_ADMINQ
+	select LIBIE_FWLOG
+	select NET_DEVLINK
+	select PACKING
+	select PLDMFW
+	select DPLL
+	help
+	  This driver supports Intel(R) Ethernet Connection E800 Series of
+	  devices.  For more information on how to identify your adapter, go
+	  to the Adapter & Driver ID Guide that can be located at:
+
+	  <http://support.intel.com>
+
+	  More specific information on configuring the driver is in
+	  <file:Documentation/networking/device_drivers/ethernet/intel/ice.rst>.
+
+	  To compile this driver as a module, choose M here. The module
+	  will be called ice.
+
+config ICE_HWMON
+	bool "Intel(R) Ethernet Connection E800 Series Support HWMON support"
+	default y
+	depends on ICE && HWMON && !(ICE=y && HWMON=m)
+	help
+	  Say Y if you want to expose thermal sensor data on Intel devices.
+
+	  Some of our devices contain internal thermal sensors.
+	  This data is available via the hwmon sysfs interface and exposes
+	  the onboard sensors.
+
+config ICE_SWITCHDEV
+	bool "Switchdev Support"
+	default y
+	depends on ICE && NET_SWITCHDEV
+	help
+	  Switchdev support provides internal SRIOV packet steering and switching.
+
+	  To enable it on running kernel use devlink tool:
+	  #devlink dev eswitch set pci/0000:XX:XX.X mode switchdev
+
+	  Say Y here if you want to use Switchdev in the driver.
+
+	  If unsure, say N.
+
+config ICE_HWTS
+	bool "Support HW cross-timestamp on platforms with PTM support"
+	default y
+	depends on ICE && X86 && PCIE_PTM
+	help
+	  Say Y to enable hardware supported cross-timestamping on platforms
+	  with PCIe PTM support. The cross-timestamp is available through
+	  the PTP clock driver precise cross-timestamp ioctl
+	  (PTP_SYS_OFFSET_PRECISE).
+
 config FM10K
 	tristate "Intel(R) FM10000 Ethernet Switch Host Interface Support"
 	default n
 	depends on PCI_MSI
-	imply PTP_1588_CLOCK
-	---help---
+	depends on PTP_1588_CLOCK_OPTIONAL
+	help
 	  This driver supports Intel(R) FM10000 Ethernet Switch Host
 	  Interface.  For more information on how to identify your adapter,
 	  go to the Adapter & Driver ID Guide that can be located at:
 
 	  <http://support.intel.com>
 
+	  More specific information on configuring the driver is in
+	  <file:Documentation/networking/device_drivers/ethernet/intel/fm10k.rst>.
+
 	  To compile this driver as a module, choose M here. The module
 	  will be called fm10k.  MSI-X interrupt support is required
 
+config IGC
+	tristate "Intel(R) Ethernet Controller I225-LM/I225-V support"
+	default n
+	depends on PCI
+	depends on PTP_1588_CLOCK_OPTIONAL
+	depends on ETHTOOL_NETLINK
+	help
+	  This driver supports Intel(R) Ethernet Controller I225-LM/I225-V
+	  family of adapters.
+
+	  For more information on how to identify your adapter, go
+	  to the Adapter & Driver ID Guide that can be located at:
+
+	  <http://support.intel.com>
+
+	  To compile this driver as a module, choose M here. The module
+	  will be called igc.
+
+config IGC_LEDS
+	def_bool LEDS_TRIGGER_NETDEV
+	depends on IGC && LEDS_CLASS
+	depends on LEDS_CLASS=y || IGC=m
+	help
+	  Optional support for controlling the NIC LED's with the netdev
+	  LED trigger.
+
+source "drivers/net/ethernet/intel/idpf/Kconfig"
+
 endif # NET_VENDOR_INTEL
diff --git a/drivers/net/ethernet/intel/Makefile b/drivers/net/ethernet/intel/Makefile
index 5ea764d85ec3..9a37dc76aef0 100644
--- a/drivers/net/ethernet/intel/Makefile
+++ b/drivers/net/ethernet/intel/Makefile
@@ -1,15 +1,21 @@
+# SPDX-License-Identifier: GPL-2.0
 #
 # Makefile for the Intel network device drivers.
 #
 
+obj-$(CONFIG_LIBETH) += libeth/
+obj-y += libie/
+
 obj-$(CONFIG_E100) += e100.o
 obj-$(CONFIG_E1000) += e1000/
 obj-$(CONFIG_E1000E) += e1000e/
 obj-$(CONFIG_IGB) += igb/
+obj-$(CONFIG_IGC) += igc/
 obj-$(CONFIG_IGBVF) += igbvf/
 obj-$(CONFIG_IXGBE) += ixgbe/
 obj-$(CONFIG_IXGBEVF) += ixgbevf/
 obj-$(CONFIG_I40E) += i40e/
-obj-$(CONFIG_IXGB) += ixgb/
-obj-$(CONFIG_I40EVF) += i40evf/
+obj-$(CONFIG_IAVF) += iavf/
 obj-$(CONFIG_FM10K) += fm10k/
+obj-$(CONFIG_ICE) += ice/
+obj-$(CONFIG_IDPF) += idpf/
diff --git a/drivers/net/ethernet/intel/e100.c b/drivers/net/ethernet/intel/e100.c
index 2b7323d392dc..5c56c1edd492 100644
--- a/drivers/net/ethernet/intel/e100.c
+++ b/drivers/net/ethernet/intel/e100.c
@@ -1,30 +1,5 @@
-/*******************************************************************************
-
-  Intel PRO/100 Linux driver
-  Copyright(c) 1999 - 2006 Intel Corporation.
-
-  This program is free software; you can redistribute it and/or modify it
-  under the terms and conditions of the GNU General Public License,
-  version 2, as published by the Free Software Foundation.
-
-  This program is distributed in the hope it will be useful, but WITHOUT
-  ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
-  FITNESS FOR A PARTICULAR PURPOSE.  See the GNU General Public License for
-  more details.
-
-  You should have received a copy of the GNU General Public License along with
-  this program; if not, write to the Free Software Foundation, Inc.,
-  51 Franklin St - Fifth Floor, Boston, MA 02110-1301 USA.
-
-  The full GNU General Public License is included in this distribution in
-  the file called "COPYING".
-
-  Contact Information:
-  Linux NICS <linux.nics@intel.com>
-  e1000-devel Mailing List <e1000-devel@lists.sourceforge.net>
-  Intel Corporation, 5200 N.E. Elam Young Parkway, Hillsboro, OR 97124-6497
-
-*******************************************************************************/
+// SPDX-License-Identifier: GPL-2.0
+/* Copyright(c) 1999 - 2006 Intel Corporation. */
 
 /*
  *	e100.c: Intel(R) PRO/100 ethernet driver
@@ -171,12 +146,10 @@
 #include <linux/string.h>
 #include <linux/firmware.h>
 #include <linux/rtnetlink.h>
-#include <asm/unaligned.h>
+#include <linux/unaligned.h>
 
 
 #define DRV_NAME		"e100"
-#define DRV_EXT			"-NAPI"
-#define DRV_VERSION		"3.5.24-k2"DRV_EXT
 #define DRV_DESCRIPTION		"Intel(R) PRO/100 Network Driver"
 #define DRV_COPYRIGHT		"Copyright(c) 1999-2006 Intel Corporation"
 
@@ -188,9 +161,7 @@
 #define FIRMWARE_D102E		"e100/d102e_ucode.bin"
 
 MODULE_DESCRIPTION(DRV_DESCRIPTION);
-MODULE_AUTHOR(DRV_COPYRIGHT);
-MODULE_LICENSE("GPL");
-MODULE_VERSION(DRV_VERSION);
+MODULE_LICENSE("GPL v2");
 MODULE_FIRMWARE(FIRMWARE_D101M);
 MODULE_FIRMWARE(FIRMWARE_D101S);
 MODULE_FIRMWARE(FIRMWARE_D102E);
@@ -199,8 +170,8 @@ static int debug = 3;
 static int eeprom_bad_csum_allow = 0;
 static int use_io = 0;
 module_param(debug, int, 0);
-module_param(eeprom_bad_csum_allow, int, 0);
-module_param(use_io, int, 0);
+module_param(eeprom_bad_csum_allow, int, 0444);
+module_param(use_io, int, 0444);
 MODULE_PARM_DESC(debug, "Debug level (0=none,...,16=all)");
 MODULE_PARM_DESC(eeprom_bad_csum_allow, "Allow bad eeprom checksums");
 MODULE_PARM_DESC(use_io, "Force use of i/o access mode");
@@ -412,7 +383,7 @@ enum cb_status {
 	cb_ok       = 0x2000,
 };
 
-/**
+/*
  * cb_command - Command Block flags
  * @cb_tx_nc:  0: controller does CRC (normal),  1: CRC from skb memory
  */
@@ -607,7 +578,7 @@ struct nic {
 	struct mem *mem;
 	dma_addr_t dma_addr;
 
-	struct pci_pool *cbs_pool;
+	struct dma_pool *cbs_pool;
 	dma_addr_t cbs_dma_addr;
 	u8 adaptive_ifs;
 	u8 tx_threshold;
@@ -1370,8 +1341,8 @@ static inline int e100_load_ucode_wait(struct nic *nic)
 
 	fw = e100_request_firmware(nic);
 	/* If it's NULL, then no ucode is required */
-	if (!fw || IS_ERR(fw))
-		return PTR_ERR(fw);
+	if (IS_ERR_OR_NULL(fw))
+		return PTR_ERR_OR_ZERO(fw);
 
 	if ((err = e100_exec_cb(nic, (void *)fw, e100_setup_ucode)))
 		netif_err(nic, probe, nic->netdev,
@@ -1423,7 +1394,7 @@ static int e100_phy_check_without_mii(struct nic *nic)
 	u8 phy_type;
 	int without_mii;
 
-	phy_type = (nic->eeprom[eeprom_phy_iface] >> 8) & 0x0f;
+	phy_type = (le16_to_cpu(nic->eeprom[eeprom_phy_iface]) >> 8) & 0x0f;
 
 	switch (phy_type) {
 	case NoSuchPhy: /* Non-MII PHY; UNTESTED! */
@@ -1458,7 +1429,6 @@ static int e100_phy_check_without_mii(struct nic *nic)
 #define MII_NSC_CONG		MII_RESV1
 #define NSC_CONG_ENABLE		0x0100
 #define NSC_CONG_TXREADY	0x0400
-#define ADVERTISE_FC_SUPPORTED	0x0400
 static int e100_phy_init(struct nic *nic)
 {
 	struct net_device *netdev = nic->netdev;
@@ -1543,7 +1513,7 @@ static int e100_phy_init(struct nic *nic)
 		mdio_write(netdev, nic->mii.phy_id, MII_BMCR, bmcr);
 	} else if ((nic->mac >= mac_82550_D102) || ((nic->flags & ich) &&
 	   (mdio_read(netdev, nic->mii.phy_id, MII_TPISTATUS) & 0x8000) &&
-		(nic->eeprom[eeprom_cnfg_mdix] & eeprom_mdix_enabled))) {
+	   (le16_to_cpu(nic->eeprom[eeprom_cnfg_mdix]) & eeprom_mdix_enabled))) {
 		/* enable/disable MDI/MDI-X auto-switching. */
 		mdio_write(netdev, nic->mii.phy_id, MII_NCONFIG,
 				nic->mii.force_media ? 0 : NCONFIG_AUTO_SWITCH);
@@ -1559,7 +1529,7 @@ static int e100_hw_init(struct nic *nic)
 	e100_hw_reset(nic);
 
 	netif_err(nic, hw, nic->netdev, "e100_hw_init\n");
-	if (!in_interrupt() && (err = e100_self_test(nic)))
+	if ((err = e100_self_test(nic)))
 		return err;
 
 	if ((err = e100_phy_init(nic)))
@@ -1710,9 +1680,9 @@ static void e100_adjust_adaptive_ifs(struct nic *nic, int speed, int duplex)
 	}
 }
 
-static void e100_watchdog(unsigned long data)
+static void e100_watchdog(struct timer_list *t)
 {
-	struct nic *nic = (struct nic *)data;
+	struct nic *nic = timer_container_of(nic, t, watchdog);
 	struct ethtool_cmd cmd = { .cmd = ETHTOOL_GSET };
 	u32 speed;
 
@@ -1767,14 +1737,11 @@ static int e100_xmit_prepare(struct nic *nic, struct cb *cb,
 	dma_addr_t dma_addr;
 	cb->command = nic->tx_command;
 
-	dma_addr = pci_map_single(nic->pdev,
-				  skb->data, skb->len, PCI_DMA_TODEVICE);
+	dma_addr = dma_map_single(&nic->pdev->dev, skb->data, skb->len,
+				  DMA_TO_DEVICE);
 	/* If we can't map the skb, have the upper layer try later */
-	if (pci_dma_mapping_error(nic->pdev, dma_addr)) {
-		dev_kfree_skb_any(skb);
-		skb = NULL;
+	if (dma_mapping_error(&nic->pdev->dev, dma_addr))
 		return -ENOMEM;
-	}
 
 	/*
 	 * Use the last 4 bytes of the SKB payload packet as the CRC, used for
@@ -1856,10 +1823,10 @@ static int e100_tx_clean(struct nic *nic)
 			dev->stats.tx_packets++;
 			dev->stats.tx_bytes += cb->skb->len;
 
-			pci_unmap_single(nic->pdev,
-				le32_to_cpu(cb->u.tcb.tbd.buf_addr),
-				le16_to_cpu(cb->u.tcb.tbd.size),
-				PCI_DMA_TODEVICE);
+			dma_unmap_single(&nic->pdev->dev,
+					 le32_to_cpu(cb->u.tcb.tbd.buf_addr),
+					 le16_to_cpu(cb->u.tcb.tbd.size),
+					 DMA_TO_DEVICE);
 			dev_kfree_skb_any(cb->skb);
 			cb->skb = NULL;
 			tx_cleaned = 1;
@@ -1883,16 +1850,16 @@ static void e100_clean_cbs(struct nic *nic)
 		while (nic->cbs_avail != nic->params.cbs.count) {
 			struct cb *cb = nic->cb_to_clean;
 			if (cb->skb) {
-				pci_unmap_single(nic->pdev,
-					le32_to_cpu(cb->u.tcb.tbd.buf_addr),
-					le16_to_cpu(cb->u.tcb.tbd.size),
-					PCI_DMA_TODEVICE);
+				dma_unmap_single(&nic->pdev->dev,
+						 le32_to_cpu(cb->u.tcb.tbd.buf_addr),
+						 le16_to_cpu(cb->u.tcb.tbd.size),
+						 DMA_TO_DEVICE);
 				dev_kfree_skb(cb->skb);
 			}
 			nic->cb_to_clean = nic->cb_to_clean->next;
 			nic->cbs_avail++;
 		}
-		pci_pool_free(nic->cbs_pool, nic->cbs, nic->cbs_dma_addr);
+		dma_pool_free(nic->cbs_pool, nic->cbs, nic->cbs_dma_addr);
 		nic->cbs = NULL;
 		nic->cbs_avail = 0;
 	}
@@ -1910,11 +1877,10 @@ static int e100_alloc_cbs(struct nic *nic)
 	nic->cb_to_use = nic->cb_to_send = nic->cb_to_clean = NULL;
 	nic->cbs_avail = 0;
 
-	nic->cbs = pci_pool_alloc(nic->cbs_pool, GFP_KERNEL,
-				  &nic->cbs_dma_addr);
+	nic->cbs = dma_pool_zalloc(nic->cbs_pool, GFP_KERNEL,
+				   &nic->cbs_dma_addr);
 	if (!nic->cbs)
 		return -ENOMEM;
-	memset(nic->cbs, 0, count * sizeof(struct cb));
 
 	for (cb = nic->cbs, i = 0; i < count; cb++, i++) {
 		cb->next = (i + 1 < count) ? cb + 1 : nic->cbs;
@@ -1954,10 +1920,10 @@ static int e100_rx_alloc_skb(struct nic *nic, struct rx *rx)
 
 	/* Init, and map the RFD. */
 	skb_copy_to_linear_data(rx->skb, &nic->blank_rfd, sizeof(struct rfd));
-	rx->dma_addr = pci_map_single(nic->pdev, rx->skb->data,
-		RFD_BUF_LEN, PCI_DMA_BIDIRECTIONAL);
+	rx->dma_addr = dma_map_single(&nic->pdev->dev, rx->skb->data,
+				      RFD_BUF_LEN, DMA_BIDIRECTIONAL);
 
-	if (pci_dma_mapping_error(nic->pdev, rx->dma_addr)) {
+	if (dma_mapping_error(&nic->pdev->dev, rx->dma_addr)) {
 		dev_kfree_skb_any(rx->skb);
 		rx->skb = NULL;
 		rx->dma_addr = 0;
@@ -1970,8 +1936,10 @@ static int e100_rx_alloc_skb(struct nic *nic, struct rx *rx)
 	if (rx->prev->skb) {
 		struct rfd *prev_rfd = (struct rfd *)rx->prev->skb->data;
 		put_unaligned_le32(rx->dma_addr, &prev_rfd->link);
-		pci_dma_sync_single_for_device(nic->pdev, rx->prev->dma_addr,
-			sizeof(struct rfd), PCI_DMA_BIDIRECTIONAL);
+		dma_sync_single_for_device(&nic->pdev->dev,
+					   rx->prev->dma_addr,
+					   sizeof(struct rfd),
+					   DMA_BIDIRECTIONAL);
 	}
 
 	return 0;
@@ -1990,8 +1958,8 @@ static int e100_rx_indicate(struct nic *nic, struct rx *rx,
 		return -EAGAIN;
 
 	/* Need to sync before taking a peek at cb_complete bit */
-	pci_dma_sync_single_for_cpu(nic->pdev, rx->dma_addr,
-		sizeof(struct rfd), PCI_DMA_BIDIRECTIONAL);
+	dma_sync_single_for_cpu(&nic->pdev->dev, rx->dma_addr,
+				sizeof(struct rfd), DMA_BIDIRECTIONAL);
 	rfd_status = le16_to_cpu(rfd->status);
 
 	netif_printk(nic, rx_status, KERN_DEBUG, nic->netdev,
@@ -2010,9 +1978,9 @@ static int e100_rx_indicate(struct nic *nic, struct rx *rx,
 
 			if (ioread8(&nic->csr->scb.status) & rus_no_res)
 				nic->ru_running = RU_SUSPENDED;
-		pci_dma_sync_single_for_device(nic->pdev, rx->dma_addr,
-					       sizeof(struct rfd),
-					       PCI_DMA_FROMDEVICE);
+		dma_sync_single_for_device(&nic->pdev->dev, rx->dma_addr,
+					   sizeof(struct rfd),
+					   DMA_FROM_DEVICE);
 		return -ENODATA;
 	}
 
@@ -2024,8 +1992,8 @@ static int e100_rx_indicate(struct nic *nic, struct rx *rx,
 		actual_size = RFD_BUF_LEN - sizeof(struct rfd);
 
 	/* Get data */
-	pci_unmap_single(nic->pdev, rx->dma_addr,
-		RFD_BUF_LEN, PCI_DMA_BIDIRECTIONAL);
+	dma_unmap_single(&nic->pdev->dev, rx->dma_addr, RFD_BUF_LEN,
+			 DMA_BIDIRECTIONAL);
 
 	/* If this buffer has the el bit, but we think the receiver
 	 * is still running, check to see if it really stopped while
@@ -2126,22 +2094,25 @@ static void e100_rx_clean(struct nic *nic, unsigned int *work_done,
 			(struct rfd *)new_before_last_rx->skb->data;
 		new_before_last_rfd->size = 0;
 		new_before_last_rfd->command |= cpu_to_le16(cb_el);
-		pci_dma_sync_single_for_device(nic->pdev,
-			new_before_last_rx->dma_addr, sizeof(struct rfd),
-			PCI_DMA_BIDIRECTIONAL);
+		dma_sync_single_for_device(&nic->pdev->dev,
+					   new_before_last_rx->dma_addr,
+					   sizeof(struct rfd),
+					   DMA_BIDIRECTIONAL);
 
 		/* Now that we have a new stopping point, we can clear the old
 		 * stopping point.  We must sync twice to get the proper
 		 * ordering on the hardware side of things. */
 		old_before_last_rfd->command &= ~cpu_to_le16(cb_el);
-		pci_dma_sync_single_for_device(nic->pdev,
-			old_before_last_rx->dma_addr, sizeof(struct rfd),
-			PCI_DMA_BIDIRECTIONAL);
+		dma_sync_single_for_device(&nic->pdev->dev,
+					   old_before_last_rx->dma_addr,
+					   sizeof(struct rfd),
+					   DMA_BIDIRECTIONAL);
 		old_before_last_rfd->size = cpu_to_le16(VLAN_ETH_FRAME_LEN
 							+ ETH_FCS_LEN);
-		pci_dma_sync_single_for_device(nic->pdev,
-			old_before_last_rx->dma_addr, sizeof(struct rfd),
-			PCI_DMA_BIDIRECTIONAL);
+		dma_sync_single_for_device(&nic->pdev->dev,
+					   old_before_last_rx->dma_addr,
+					   sizeof(struct rfd),
+					   DMA_BIDIRECTIONAL);
 	}
 
 	if (restart_required) {
@@ -2163,8 +2134,9 @@ static void e100_rx_clean_list(struct nic *nic)
 	if (nic->rxs) {
 		for (rx = nic->rxs, i = 0; i < count; rx++, i++) {
 			if (rx->skb) {
-				pci_unmap_single(nic->pdev, rx->dma_addr,
-					RFD_BUF_LEN, PCI_DMA_BIDIRECTIONAL);
+				dma_unmap_single(&nic->pdev->dev,
+						 rx->dma_addr, RFD_BUF_LEN,
+						 DMA_BIDIRECTIONAL);
 				dev_kfree_skb(rx->skb);
 			}
 		}
@@ -2184,7 +2156,7 @@ static int e100_rx_alloc_list(struct nic *nic)
 	nic->rx_to_use = nic->rx_to_clean = NULL;
 	nic->ru_running = RU_UNINITIALIZED;
 
-	if (!(nic->rxs = kcalloc(count, sizeof(struct rx), GFP_ATOMIC)))
+	if (!(nic->rxs = kcalloc(count, sizeof(struct rx), GFP_KERNEL)))
 		return -ENOMEM;
 
 	for (rx = nic->rxs, i = 0; i < count; rx++, i++) {
@@ -2206,8 +2178,8 @@ static int e100_rx_alloc_list(struct nic *nic)
 	before_last = (struct rfd *)rx->skb->data;
 	before_last->command |= cpu_to_le16(cb_el);
 	before_last->size = 0;
-	pci_dma_sync_single_for_device(nic->pdev, rx->dma_addr,
-		sizeof(struct rfd), PCI_DMA_BIDIRECTIONAL);
+	dma_sync_single_for_device(&nic->pdev->dev, rx->dma_addr,
+				   sizeof(struct rfd), DMA_BIDIRECTIONAL);
 
 	nic->rx_to_use = nic->rx_to_clean = nic->rxs;
 	nic->ru_running = RU_SUSPENDED;
@@ -2251,11 +2223,13 @@ static int e100_poll(struct napi_struct *napi, int budget)
 	e100_rx_clean(nic, &work_done, budget);
 	e100_tx_clean(nic);
 
-	/* If budget not fully consumed, exit the polling mode */
-	if (work_done < budget) {
-		napi_complete_done(napi, work_done);
+	/* If budget fully consumed, continue polling */
+	if (work_done == budget)
+		return budget;
+
+	/* only re-enable interrupt if stack agrees polling is really done */
+	if (likely(napi_complete_done(napi, work_done)))
 		e100_enable_irq(nic);
-	}
 
 	return work_done;
 }
@@ -2280,7 +2254,7 @@ static int e100_set_mac_address(struct net_device *netdev, void *p)
 	if (!is_valid_ether_addr(addr->sa_data))
 		return -EADDRNOTAVAIL;
 
-	memcpy(netdev->dev_addr, addr->sa_data, netdev->addr_len);
+	eth_hw_addr_set(netdev, addr->sa_data);
 	e100_exec_cb(nic, NULL, e100_setup_iaaddr);
 
 	return 0;
@@ -2290,9 +2264,9 @@ static int e100_asf(struct nic *nic)
 {
 	/* ASF can be enabled from eeprom */
 	return (nic->pdev->device >= 0x1050) && (nic->pdev->device <= 0x1057) &&
-	   (nic->eeprom[eeprom_config_asf] & eeprom_asf) &&
-	   !(nic->eeprom[eeprom_config_asf] & eeprom_gcl) &&
-	   ((nic->eeprom[eeprom_smbus_addr] & 0xFF) != 0xFE);
+	   (le16_to_cpu(nic->eeprom[eeprom_config_asf]) & eeprom_asf) &&
+	   !(le16_to_cpu(nic->eeprom[eeprom_config_asf]) & eeprom_gcl) &&
+	   ((le16_to_cpu(nic->eeprom[eeprom_smbus_addr]) & 0xFF) != 0xFE);
 }
 
 static int e100_up(struct nic *nic)
@@ -2319,7 +2293,7 @@ static int e100_up(struct nic *nic)
 	return 0;
 
 err_no_irq:
-	del_timer_sync(&nic->watchdog);
+	timer_delete_sync(&nic->watchdog);
 err_clean_cbs:
 	e100_clean_cbs(nic);
 err_rx_clean_list:
@@ -2334,13 +2308,13 @@ static void e100_down(struct nic *nic)
 	netif_stop_queue(nic->netdev);
 	e100_hw_reset(nic);
 	free_irq(nic->pdev->irq, nic->netdev);
-	del_timer_sync(&nic->watchdog);
+	timer_delete_sync(&nic->watchdog);
 	netif_carrier_off(nic->netdev);
 	e100_clean_cbs(nic);
 	e100_rx_clean_list(nic);
 }
 
-static void e100_tx_timeout(struct net_device *netdev)
+static void e100_tx_timeout(struct net_device *netdev, unsigned int txqueue)
 {
 	struct nic *nic = netdev_priv(netdev);
 
@@ -2404,8 +2378,8 @@ static int e100_loopback_test(struct nic *nic, enum loopback loopback_mode)
 
 	msleep(10);
 
-	pci_dma_sync_single_for_cpu(nic->pdev, nic->rx_to_clean->dma_addr,
-			RFD_BUF_LEN, PCI_DMA_BIDIRECTIONAL);
+	dma_sync_single_for_cpu(&nic->pdev->dev, nic->rx_to_clean->dma_addr,
+				RFD_BUF_LEN, DMA_BIDIRECTIONAL);
 
 	if (memcmp(nic->rx_to_clean->skb->data + sizeof(struct rfd),
 	   skb->data, ETH_DATA_LEN))
@@ -2430,7 +2404,10 @@ static int e100_get_link_ksettings(struct net_device *netdev,
 				   struct ethtool_link_ksettings *cmd)
 {
 	struct nic *nic = netdev_priv(netdev);
-	return mii_ethtool_get_link_ksettings(&nic->mii, cmd);
+
+	mii_ethtool_get_link_ksettings(&nic->mii, cmd);
+
+	return 0;
 }
 
 static int e100_set_link_ksettings(struct net_device *netdev,
@@ -2450,17 +2427,20 @@ static void e100_get_drvinfo(struct net_device *netdev,
 	struct ethtool_drvinfo *info)
 {
 	struct nic *nic = netdev_priv(netdev);
-	strlcpy(info->driver, DRV_NAME, sizeof(info->driver));
-	strlcpy(info->version, DRV_VERSION, sizeof(info->version));
-	strlcpy(info->bus_info, pci_name(nic->pdev),
+	strscpy(info->driver, DRV_NAME, sizeof(info->driver));
+	strscpy(info->bus_info, pci_name(nic->pdev),
 		sizeof(info->bus_info));
 }
 
-#define E100_PHY_REGS 0x1C
+#define E100_PHY_REGS 0x1D
 static int e100_get_regs_len(struct net_device *netdev)
 {
 	struct nic *nic = netdev_priv(netdev);
-	return 1 + E100_PHY_REGS + sizeof(nic->mem->dump_buf);
+
+	/* We know the number of registers, and the size of the dump buffer.
+	 * Calculate the total size in bytes.
+	 */
+	return (1 + E100_PHY_REGS) * sizeof(u32) + sizeof(nic->mem->dump_buf);
 }
 
 static void e100_get_regs(struct net_device *netdev,
@@ -2474,14 +2454,18 @@ static void e100_get_regs(struct net_device *netdev,
 	buff[0] = ioread8(&nic->csr->scb.cmd_hi) << 24 |
 		ioread8(&nic->csr->scb.cmd_lo) << 16 |
 		ioread16(&nic->csr->scb.status);
-	for (i = E100_PHY_REGS; i >= 0; i--)
-		buff[1 + E100_PHY_REGS - i] =
-			mdio_read(netdev, nic->mii.phy_id, i);
+	for (i = 0; i < E100_PHY_REGS; i++)
+		/* Note that we read the registers in reverse order. This
+		 * ordering is the ABI apparently used by ethtool and other
+		 * applications.
+		 */
+		buff[1 + i] = mdio_read(netdev, nic->mii.phy_id,
+					E100_PHY_REGS - 1 - i);
 	memset(nic->mem->dump_buf, 0, sizeof(nic->mem->dump_buf));
 	e100_exec_cb(nic, NULL, e100_dump);
 	msleep(10);
-	memcpy(&buff[2 + E100_PHY_REGS], nic->mem->dump_buf,
-		sizeof(nic->mem->dump_buf));
+	memcpy(&buff[1 + E100_PHY_REGS], nic->mem->dump_buf,
+	       sizeof(nic->mem->dump_buf));
 }
 
 static void e100_get_wol(struct net_device *netdev, struct ethtool_wolinfo *wol)
@@ -2568,7 +2552,9 @@ static int e100_set_eeprom(struct net_device *netdev,
 }
 
 static void e100_get_ringparam(struct net_device *netdev,
-	struct ethtool_ringparam *ring)
+			       struct ethtool_ringparam *ring,
+			       struct kernel_ethtool_ringparam *kernel_ring,
+			       struct netlink_ext_ack *extack)
 {
 	struct nic *nic = netdev_priv(netdev);
 	struct param_range *rfds = &nic->params.rfds;
@@ -2581,7 +2567,9 @@ static void e100_get_ringparam(struct net_device *netdev,
 }
 
 static int e100_set_ringparam(struct net_device *netdev,
-	struct ethtool_ringparam *ring)
+			      struct ethtool_ringparam *ring,
+			      struct kernel_ethtool_ringparam *kernel_ring,
+			      struct netlink_ext_ack *extack)
 {
 	struct nic *nic = netdev_priv(netdev);
 	struct param_range *rfds = &nic->params.rfds;
@@ -2618,7 +2606,7 @@ static void e100_diag_test(struct net_device *netdev,
 {
 	struct ethtool_cmd cmd;
 	struct nic *nic = netdev_priv(netdev);
-	int i, err;
+	int i;
 
 	memset(data, 0, E100_TEST_LEN * sizeof(u64));
 	data[0] = !mii_link_ok(&nic->mii);
@@ -2626,7 +2614,7 @@ static void e100_diag_test(struct net_device *netdev,
 	if (test->flags & ETH_TEST_FL_OFFLINE) {
 
 		/* save speed, duplex & autoneg settings */
-		err = mii_ethtool_gset(&nic->mii, &cmd);
+		mii_ethtool_gset(&nic->mii, &cmd);
 
 		if (netif_running(netdev))
 			e100_down(nic);
@@ -2635,7 +2623,7 @@ static void e100_diag_test(struct net_device *netdev,
 		data[4] = e100_loopback_test(nic, lb_phy);
 
 		/* restore speed, duplex & autoneg settings */
-		err = mii_ethtool_sset(&nic->mii, &cmd);
+		mii_ethtool_sset(&nic->mii, &cmd);
 
 		if (netif_running(netdev))
 			e100_up(nic);
@@ -2734,10 +2722,10 @@ static void e100_get_strings(struct net_device *netdev, u32 stringset, u8 *data)
 {
 	switch (stringset) {
 	case ETH_SS_TEST:
-		memcpy(data, *e100_gstrings_test, sizeof(e100_gstrings_test));
+		memcpy(data, e100_gstrings_test, sizeof(e100_gstrings_test));
 		break;
 	case ETH_SS_STATS:
-		memcpy(data, *e100_gstrings_stats, sizeof(e100_gstrings_stats));
+		memcpy(data, e100_gstrings_stats, sizeof(e100_gstrings_stats));
 		break;
 	}
 }
@@ -2776,16 +2764,16 @@ static int e100_do_ioctl(struct net_device *netdev, struct ifreq *ifr, int cmd)
 
 static int e100_alloc(struct nic *nic)
 {
-	nic->mem = pci_alloc_consistent(nic->pdev, sizeof(struct mem),
-		&nic->dma_addr);
+	nic->mem = dma_alloc_coherent(&nic->pdev->dev, sizeof(struct mem),
+				      &nic->dma_addr, GFP_KERNEL);
 	return nic->mem ? 0 : -ENOMEM;
 }
 
 static void e100_free(struct nic *nic)
 {
 	if (nic->mem) {
-		pci_free_consistent(nic->pdev, sizeof(struct mem),
-			nic->mem, nic->dma_addr);
+		dma_free_coherent(&nic->pdev->dev, sizeof(struct mem),
+				  nic->mem, nic->dma_addr);
 		nic->mem = NULL;
 	}
 }
@@ -2818,7 +2806,7 @@ static int e100_set_features(struct net_device *netdev,
 
 	netdev->features = features;
 	e100_exec_cb(nic, NULL, e100_configure);
-	return 0;
+	return 1;
 }
 
 static const struct net_device_ops e100_netdev_ops = {
@@ -2828,7 +2816,7 @@ static const struct net_device_ops e100_netdev_ops = {
 	.ndo_validate_addr	= eth_validate_addr,
 	.ndo_set_rx_mode	= e100_set_multicast_list,
 	.ndo_set_mac_address	= e100_set_mac_address,
-	.ndo_do_ioctl		= e100_do_ioctl,
+	.ndo_eth_ioctl		= e100_do_ioctl,
 	.ndo_tx_timeout		= e100_tx_timeout,
 #ifdef CONFIG_NET_POLL_CONTROLLER
 	.ndo_poll_controller	= e100_netpoll,
@@ -2852,10 +2840,10 @@ static int e100_probe(struct pci_dev *pdev, const struct pci_device_id *ent)
 	netdev->netdev_ops = &e100_netdev_ops;
 	netdev->ethtool_ops = &e100_ethtool_ops;
 	netdev->watchdog_timeo = E100_WATCHDOG_PERIOD;
-	strncpy(netdev->name, pci_name(pdev), sizeof(netdev->name) - 1);
+	strscpy(netdev->name, pci_name(pdev), sizeof(netdev->name));
 
 	nic = netdev_priv(netdev);
-	netif_napi_add(netdev, &nic->napi, e100_poll, E100_NAPI_WEIGHT);
+	netif_napi_add_weight(netdev, &nic->napi, e100_poll, E100_NAPI_WEIGHT);
 	nic->netdev = netdev;
 	nic->pdev = pdev;
 	nic->msg_enable = (1 << debug) - 1;
@@ -2878,7 +2866,7 @@ static int e100_probe(struct pci_dev *pdev, const struct pci_device_id *ent)
 		goto err_out_disable_pdev;
 	}
 
-	if ((err = pci_set_dma_mask(pdev, DMA_BIT_MASK(32)))) {
+	if ((err = dma_set_mask(&pdev->dev, DMA_BIT_MASK(32)))) {
 		netif_err(nic, probe, nic->netdev, "No usable DMA configuration, aborting\n");
 		goto err_out_free_res;
 	}
@@ -2918,7 +2906,7 @@ static int e100_probe(struct pci_dev *pdev, const struct pci_device_id *ent)
 
 	pci_set_master(pdev);
 
-	setup_timer(&nic->watchdog, e100_watchdog, (unsigned long)nic);
+	timer_setup(&nic->watchdog, e100_watchdog, 0);
 
 	INIT_WORK(&nic->tx_timeout_task, e100_tx_timeout_task);
 
@@ -2932,7 +2920,7 @@ static int e100_probe(struct pci_dev *pdev, const struct pci_device_id *ent)
 
 	e100_phy_init(nic);
 
-	memcpy(netdev->dev_addr, nic->eeprom, ETH_ALEN);
+	eth_hw_addr_set(netdev, (u8 *)nic->eeprom);
 	if (!is_valid_ether_addr(netdev->dev_addr)) {
 		if (!eeprom_bad_csum_allow) {
 			netif_err(nic, probe, nic->netdev, "Invalid MAC address from EEPROM, aborting\n");
@@ -2945,7 +2933,7 @@ static int e100_probe(struct pci_dev *pdev, const struct pci_device_id *ent)
 
 	/* Wol magic packet can be enabled from eeprom */
 	if ((nic->mac >= mac_82558_D101_A4) &&
-	   (nic->eeprom[eeprom_id] & eeprom_id_wol)) {
+	   (le16_to_cpu(nic->eeprom[eeprom_id]) & eeprom_id_wol)) {
 		nic->flags |= wol_magic;
 		device_set_wakeup_enable(&pdev->dev, true);
 	}
@@ -2958,8 +2946,8 @@ static int e100_probe(struct pci_dev *pdev, const struct pci_device_id *ent)
 		netif_err(nic, probe, nic->netdev, "Cannot register net device, aborting\n");
 		goto err_out_free;
 	}
-	nic->cbs_pool = pci_pool_create(netdev->name,
-			   nic->pdev,
+	nic->cbs_pool = dma_pool_create(netdev->name,
+			   &nic->pdev->dev,
 			   nic->params.cbs.max * sizeof(struct cb),
 			   sizeof(u32),
 			   0);
@@ -2999,7 +2987,7 @@ static void e100_remove(struct pci_dev *pdev)
 		unregister_netdev(netdev);
 		e100_free(nic);
 		pci_iounmap(pdev, nic->csr);
-		pci_pool_destroy(nic->cbs_pool);
+		dma_pool_destroy(nic->cbs_pool);
 		free_netdev(netdev);
 		pci_release_regions(pdev);
 		pci_disable_device(pdev);
@@ -3014,11 +3002,10 @@ static void __e100_shutdown(struct pci_dev *pdev, bool *enable_wake)
 	struct net_device *netdev = pci_get_drvdata(pdev);
 	struct nic *nic = netdev_priv(netdev);
 
-	if (netif_running(netdev))
-		e100_down(nic);
 	netif_device_detach(netdev);
 
-	pci_save_state(pdev);
+	if (netif_running(netdev))
+		e100_down(nic);
 
 	if ((nic->flags & wol_magic) | e100_asf(nic)) {
 		/* enable reverse auto-negotiation */
@@ -3035,7 +3022,7 @@ static void __e100_shutdown(struct pci_dev *pdev, bool *enable_wake)
 		*enable_wake = false;
 	}
 
-	pci_clear_master(pdev);
+	pci_disable_device(pdev);
 }
 
 static int __e100_power_off(struct pci_dev *pdev, bool wake)
@@ -3049,23 +3036,27 @@ static int __e100_power_off(struct pci_dev *pdev, bool wake)
 	return 0;
 }
 
-#ifdef CONFIG_PM
-static int e100_suspend(struct pci_dev *pdev, pm_message_t state)
+static int e100_suspend(struct device *dev_d)
 {
 	bool wake;
-	__e100_shutdown(pdev, &wake);
-	return __e100_power_off(pdev, wake);
+
+	__e100_shutdown(to_pci_dev(dev_d), &wake);
+
+	return 0;
 }
 
-static int e100_resume(struct pci_dev *pdev)
+static int e100_resume(struct device *dev_d)
 {
-	struct net_device *netdev = pci_get_drvdata(pdev);
+	struct net_device *netdev = dev_get_drvdata(dev_d);
 	struct nic *nic = netdev_priv(netdev);
+	int err;
 
-	pci_set_power_state(pdev, PCI_D0);
-	pci_restore_state(pdev);
-	/* ack any pending wake events, disable PME */
-	pci_enable_wake(pdev, PCI_D0, 0);
+	err = pci_enable_device(to_pci_dev(dev_d));
+	if (err) {
+		netdev_err(netdev, "Resume cannot enable PCI device, aborting\n");
+		return err;
+	}
+	pci_set_master(to_pci_dev(dev_d));
 
 	/* disable reverse auto-negotiation */
 	if (nic->phy == phy_82552_v) {
@@ -3077,13 +3068,13 @@ static int e100_resume(struct pci_dev *pdev)
 		           smartspeed & ~(E100_82552_REV_ANEG));
 	}
 
-	netif_device_attach(netdev);
 	if (netif_running(netdev))
 		e100_up(nic);
 
+	netif_device_attach(netdev);
+
 	return 0;
 }
-#endif /* CONFIG_PM */
 
 static void e100_shutdown(struct pci_dev *pdev)
 {
@@ -3171,16 +3162,17 @@ static const struct pci_error_handlers e100_err_handler = {
 	.resume = e100_io_resume,
 };
 
+static DEFINE_SIMPLE_DEV_PM_OPS(e100_pm_ops, e100_suspend, e100_resume);
+
 static struct pci_driver e100_driver = {
 	.name =         DRV_NAME,
 	.id_table =     e100_id_table,
 	.probe =        e100_probe,
 	.remove =       e100_remove,
-#ifdef CONFIG_PM
+
 	/* Power Management hooks */
-	.suspend =      e100_suspend,
-	.resume =       e100_resume,
-#endif
+	.driver.pm =	pm_sleep_ptr(&e100_pm_ops),
+
 	.shutdown =     e100_shutdown,
 	.err_handler = &e100_err_handler,
 };
@@ -3188,7 +3180,7 @@ static struct pci_driver e100_driver = {
 static int __init e100_init_module(void)
 {
 	if (((1 << debug) - 1) & NETIF_MSG_DRV) {
-		pr_info("%s, %s\n", DRV_DESCRIPTION, DRV_VERSION);
+		pr_info("%s\n", DRV_DESCRIPTION);
 		pr_info("%s\n", DRV_COPYRIGHT);
 	}
 	return pci_register_driver(&e100_driver);
diff --git a/drivers/net/ethernet/intel/e1000/Makefile b/drivers/net/ethernet/intel/e1000/Makefile
index 4a6ab1522451..79491dec47e1 100644
--- a/drivers/net/ethernet/intel/e1000/Makefile
+++ b/drivers/net/ethernet/intel/e1000/Makefile
@@ -1,30 +1,5 @@
-################################################################################
-#
-# Intel PRO/1000 Linux driver
+# SPDX-License-Identifier: GPL-2.0
 # Copyright(c) 1999 - 2006 Intel Corporation.
-#
-# This program is free software; you can redistribute it and/or modify it
-# under the terms and conditions of the GNU General Public License,
-# version 2, as published by the Free Software Foundation.
-#
-# This program is distributed in the hope it will be useful, but WITHOUT
-# ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
-# FITNESS FOR A PARTICULAR PURPOSE.  See the GNU General Public License for
-# more details.
-#
-# You should have received a copy of the GNU General Public License along with
-# this program; if not, write to the Free Software Foundation, Inc.,
-# 51 Franklin St - Fifth Floor, Boston, MA 02110-1301 USA.
-#
-# The full GNU General Public License is included in this distribution in
-# the file called "COPYING".
-#
-# Contact Information:
-# Linux NICS <linux.nics@intel.com>
-# e1000-devel Mailing List <e1000-devel@lists.sourceforge.net>
-# Intel Corporation, 5200 N.E. Elam Young Parkway, Hillsboro, OR 97124-6497
-#
-################################################################################
 
 #
 # Makefile for the Intel(R) PRO/1000 ethernet driver
@@ -32,4 +7,4 @@
 
 obj-$(CONFIG_E1000) += e1000.o
 
-e1000-objs := e1000_main.o e1000_hw.o e1000_ethtool.o e1000_param.o
+e1000-y := e1000_main.o e1000_hw.o e1000_ethtool.o e1000_param.o
diff --git a/drivers/net/ethernet/intel/e1000/e1000.h b/drivers/net/ethernet/intel/e1000/e1000.h
index d7bdea79e9fa..ea6ccf4b728b 100644
--- a/drivers/net/ethernet/intel/e1000/e1000.h
+++ b/drivers/net/ethernet/intel/e1000/e1000.h
@@ -1,31 +1,5 @@
-/*******************************************************************************
-
-  Intel PRO/1000 Linux driver
-  Copyright(c) 1999 - 2006 Intel Corporation.
-
-  This program is free software; you can redistribute it and/or modify it
-  under the terms and conditions of the GNU General Public License,
-  version 2, as published by the Free Software Foundation.
-
-  This program is distributed in the hope it will be useful, but WITHOUT
-  ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
-  FITNESS FOR A PARTICULAR PURPOSE.  See the GNU General Public License for
-  more details.
-
-  You should have received a copy of the GNU General Public License along with
-  this program; if not, write to the Free Software Foundation, Inc.,
-  51 Franklin St - Fifth Floor, Boston, MA 02110-1301 USA.
-
-  The full GNU General Public License is included in this distribution in
-  the file called "COPYING".
-
-  Contact Information:
-  Linux NICS <linux.nics@intel.com>
-  e1000-devel Mailing List <e1000-devel@lists.sourceforge.net>
-  Intel Corporation, 5200 N.E. Elam Young Parkway, Hillsboro, OR 97124-6497
-
-*******************************************************************************/
-
+/* SPDX-License-Identifier: GPL-2.0 */
+/* Copyright(c) 1999 - 2006 Intel Corporation. */
 
 /* Linux PRO/1000 Ethernet Driver main header file */
 
@@ -71,7 +45,6 @@
 
 #define BAR_0		0
 #define BAR_1		1
-#define BAR_5		5
 
 #define INTEL_E1000_ETHERNET_DEVICE(device_id) {\
 	PCI_DEVICE(PCI_VENDOR_ID_INTEL, device_id)}
@@ -143,7 +116,7 @@ struct e1000_adapter;
 #define E1000_MASTER_SLAVE	e1000_ms_hw_default
 #endif
 
-#define E1000_MNG_VLAN_NONE	(-1)
+#define E1000_MNG_VLAN_NONE	0xFFFF
 
 /* wrapper around a pointer to a socket buffer,
  * so a DMA handle can be stored along with the buffer
@@ -331,7 +304,8 @@ struct e1000_adapter {
 enum e1000_state_t {
 	__E1000_TESTING,
 	__E1000_RESETTING,
-	__E1000_DOWN
+	__E1000_DOWN,
+	__E1000_DISABLED
 };
 
 #undef pr_fmt
@@ -356,7 +330,6 @@ struct net_device *e1000_get_hw_dev(struct e1000_hw *hw);
 	dev_err(&adapter->pdev->dev, format, ## arg)
 
 extern char e1000_driver_name[];
-extern const char e1000_driver_version[];
 
 int e1000_open(struct net_device *netdev);
 int e1000_close(struct net_device *netdev);
@@ -374,6 +347,5 @@ bool e1000_has_link(struct e1000_adapter *adapter);
 void e1000_power_up_phy(struct e1000_adapter *);
 void e1000_set_ethtool_ops(struct net_device *netdev);
 void e1000_check_options(struct e1000_adapter *adapter);
-char *e1000_get_hw_dev_name(struct e1000_hw *hw);
 
 #endif /* _E1000_H_ */
diff --git a/drivers/net/ethernet/intel/e1000/e1000_ethtool.c b/drivers/net/ethernet/intel/e1000/e1000_ethtool.c
index ec8aa4562cc9..726365c567ef 100644
--- a/drivers/net/ethernet/intel/e1000/e1000_ethtool.c
+++ b/drivers/net/ethernet/intel/e1000/e1000_ethtool.c
@@ -1,25 +1,5 @@
-/*******************************************************************************
- * Intel PRO/1000 Linux driver
- * Copyright(c) 1999 - 2006 Intel Corporation.
- *
- * This program is free software; you can redistribute it and/or modify it
- * under the terms and conditions of the GNU General Public License,
- * version 2, as published by the Free Software Foundation.
- *
- * This program is distributed in the hope it will be useful, but WITHOUT
- * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
- * FITNESS FOR A PARTICULAR PURPOSE.  See the GNU General Public License for
- * more details.
- *
- * The full GNU General Public License is included in this distribution in
- * the file called "COPYING".
- *
- * Contact Information:
- * Linux NICS <linux.nics@intel.com>
- * e1000-devel Mailing List <e1000-devel@lists.sourceforge.net>
- * Intel Corporation, 5200 N.E. Elam Young Parkway, Hillsboro, OR 97124-6497
- *
- ******************************************************************************/
+// SPDX-License-Identifier: GPL-2.0
+/* Copyright(c) 1999 - 2006 Intel Corporation. */
 
 /* ethtool support for e1000 */
 
@@ -455,8 +435,8 @@ static int e1000_get_eeprom(struct net_device *netdev,
 	first_word = eeprom->offset >> 1;
 	last_word = (eeprom->offset + eeprom->len - 1) >> 1;
 
-	eeprom_buff = kmalloc(sizeof(u16) *
-			(last_word - first_word + 1), GFP_KERNEL);
+	eeprom_buff = kmalloc_array(last_word - first_word + 1, sizeof(u16),
+				    GFP_KERNEL);
 	if (!eeprom_buff)
 		return -ENOMEM;
 
@@ -533,7 +513,7 @@ static int e1000_set_eeprom(struct net_device *netdev,
 	memcpy(ptr, bytes, eeprom->len);
 
 	for (i = 0; i < last_word - first_word + 1; i++)
-		eeprom_buff[i] = cpu_to_le16(eeprom_buff[i]);
+		cpu_to_le16s(&eeprom_buff[i]);
 
 	ret_val = e1000_write_eeprom(hw, first_word,
 				     last_word - first_word + 1, eeprom_buff);
@@ -551,17 +531,17 @@ static void e1000_get_drvinfo(struct net_device *netdev,
 {
 	struct e1000_adapter *adapter = netdev_priv(netdev);
 
-	strlcpy(drvinfo->driver,  e1000_driver_name,
+	strscpy(drvinfo->driver,  e1000_driver_name,
 		sizeof(drvinfo->driver));
-	strlcpy(drvinfo->version, e1000_driver_version,
-		sizeof(drvinfo->version));
 
-	strlcpy(drvinfo->bus_info, pci_name(adapter->pdev),
+	strscpy(drvinfo->bus_info, pci_name(adapter->pdev),
 		sizeof(drvinfo->bus_info));
 }
 
 static void e1000_get_ringparam(struct net_device *netdev,
-				struct ethtool_ringparam *ring)
+				struct ethtool_ringparam *ring,
+				struct kernel_ethtool_ringparam *kernel_ring,
+				struct netlink_ext_ack *extack)
 {
 	struct e1000_adapter *adapter = netdev_priv(netdev);
 	struct e1000_hw *hw = &adapter->hw;
@@ -578,7 +558,9 @@ static void e1000_get_ringparam(struct net_device *netdev,
 }
 
 static int e1000_set_ringparam(struct net_device *netdev,
-			       struct ethtool_ringparam *ring)
+			       struct ethtool_ringparam *ring,
+			       struct kernel_ethtool_ringparam *kernel_ring,
+			       struct netlink_ext_ack *extack)
 {
 	struct e1000_adapter *adapter = netdev_priv(netdev);
 	struct e1000_hw *hw = &adapter->hw;
@@ -627,6 +609,7 @@ static int e1000_set_ringparam(struct net_device *netdev,
 	for (i = 0; i < adapter->num_rx_queues; i++)
 		rxdr[i].count = rxdr->count;
 
+	err = 0;
 	if (netif_running(adapter->netdev)) {
 		/* Try to get new resources before deleting old */
 		err = e1000_setup_all_rx_resources(adapter);
@@ -644,17 +627,16 @@ static int e1000_set_ringparam(struct net_device *netdev,
 		adapter->tx_ring = tx_old;
 		e1000_free_all_rx_resources(adapter);
 		e1000_free_all_tx_resources(adapter);
-		kfree(tx_old);
-		kfree(rx_old);
 		adapter->rx_ring = rxdr;
 		adapter->tx_ring = txdr;
 		err = e1000_up(adapter);
-		if (err)
-			goto err_setup;
 	}
+	kfree(tx_old);
+	kfree(rx_old);
 
 	clear_bit(__E1000_RESETTING, &adapter->flags);
-	return 0;
+	return err;
+
 err_setup_tx:
 	e1000_free_all_rx_resources(adapter);
 err_setup_rx:
@@ -664,8 +646,8 @@ err_setup_rx:
 err_alloc_rx:
 	kfree(txdr);
 err_alloc_tx:
-	e1000_up(adapter);
-err_setup:
+	if (netif_running(adapter->netdev))
+		e1000_up(adapter);
 	clear_bit(__E1000_RESETTING, &adapter->flags);
 	return err;
 }
@@ -824,7 +806,7 @@ static int e1000_eeprom_test(struct e1000_adapter *adapter, u64 *data)
 	}
 
 	/* If Checksum is not Correct return error else test passed */
-	if ((checksum != (u16)EEPROM_SUM) && !(*data))
+	if (checksum != EEPROM_SUM && !(*data))
 		*data = 2;
 
 	return *data;
@@ -956,8 +938,7 @@ static void e1000_free_desc_rings(struct e1000_adapter *adapter)
 						 txdr->buffer_info[i].dma,
 						 txdr->buffer_info[i].length,
 						 DMA_TO_DEVICE);
-			if (txdr->buffer_info[i].skb)
-				dev_kfree_skb(txdr->buffer_info[i].skb);
+			dev_kfree_skb(txdr->buffer_info[i].skb);
 		}
 	}
 
@@ -1012,8 +993,8 @@ static int e1000_setup_desc_rings(struct e1000_adapter *adapter)
 
 	txdr->size = txdr->count * sizeof(struct e1000_tx_desc);
 	txdr->size = ALIGN(txdr->size, 4096);
-	txdr->desc = dma_zalloc_coherent(&pdev->dev, txdr->size, &txdr->dma,
-					 GFP_KERNEL);
+	txdr->desc = dma_alloc_coherent(&pdev->dev, txdr->size, &txdr->dma,
+					GFP_KERNEL);
 	if (!txdr->desc) {
 		ret_val = 2;
 		goto err_nomem;
@@ -1070,8 +1051,8 @@ static int e1000_setup_desc_rings(struct e1000_adapter *adapter)
 	}
 
 	rxdr->size = rxdr->count * sizeof(struct e1000_rx_desc);
-	rxdr->desc = dma_zalloc_coherent(&pdev->dev, rxdr->size, &rxdr->dma,
-					 GFP_KERNEL);
+	rxdr->desc = dma_alloc_coherent(&pdev->dev, rxdr->size, &rxdr->dma,
+					GFP_KERNEL);
 	if (!rxdr->desc) {
 		ret_val = 6;
 		goto err_nomem;
@@ -1379,8 +1360,8 @@ static void e1000_create_lbtest_frame(struct sk_buff *skb,
 	memset(skb->data, 0xFF, frame_size);
 	frame_size &= ~1;
 	memset(&skb->data[frame_size / 2], 0xAA, frame_size / 2 - 1);
-	memset(&skb->data[frame_size / 2 + 10], 0xBE, 1);
-	memset(&skb->data[frame_size / 2 + 12], 0xAF, 1);
+	skb->data[frame_size / 2 + 10] = 0xBE;
+	skb->data[frame_size / 2 + 12] = 0xAF;
 }
 
 static int e1000_check_lbtest_frame(const unsigned char *data,
@@ -1762,7 +1743,9 @@ static int e1000_set_phys_id(struct net_device *netdev,
 }
 
 static int e1000_get_coalesce(struct net_device *netdev,
-			      struct ethtool_coalesce *ec)
+			      struct ethtool_coalesce *ec,
+			      struct kernel_ethtool_coalesce *kernel_coal,
+			      struct netlink_ext_ack *extack)
 {
 	struct e1000_adapter *adapter = netdev_priv(netdev);
 
@@ -1778,7 +1761,9 @@ static int e1000_get_coalesce(struct net_device *netdev,
 }
 
 static int e1000_set_coalesce(struct net_device *netdev,
-			      struct ethtool_coalesce *ec)
+			      struct ethtool_coalesce *ec,
+			      struct kernel_ethtool_coalesce *kernel_coal,
+			      struct netlink_ext_ack *extack)
 {
 	struct e1000_adapter *adapter = netdev_priv(netdev);
 	struct e1000_hw *hw = &adapter->hw;
@@ -1824,11 +1809,12 @@ static void e1000_get_ethtool_stats(struct net_device *netdev,
 {
 	struct e1000_adapter *adapter = netdev_priv(netdev);
 	int i;
-	char *p = NULL;
 	const struct e1000_stats *stat = e1000_gstrings_stats;
 
 	e1000_update_stats(adapter);
-	for (i = 0; i < E1000_GLOBAL_STATS_LEN; i++) {
+	for (i = 0; i < E1000_GLOBAL_STATS_LEN; i++, stat++) {
+		char *p;
+
 		switch (stat->type) {
 		case NETDEV_STATS:
 			p = (char *)netdev + stat->stat_offset;
@@ -1837,17 +1823,15 @@ static void e1000_get_ethtool_stats(struct net_device *netdev,
 			p = (char *)adapter + stat->stat_offset;
 			break;
 		default:
-			WARN_ONCE(1, "Invalid E1000 stat type: %u index %d\n",
-				  stat->type, i);
-			break;
+			netdev_WARN_ONCE(netdev, "Invalid E1000 stat type: %u index %d\n",
+					 stat->type, i);
+			continue;
 		}
 
 		if (stat->sizeof_stat == sizeof(u64))
 			data[i] = *(u64 *)p;
 		else
 			data[i] = *(u32 *)p;
-
-		stat++;
 	}
 /* BUG_ON(i != E1000_STATS_LEN); */
 }
@@ -1874,6 +1858,7 @@ static void e1000_get_strings(struct net_device *netdev, u32 stringset,
 }
 
 static const struct ethtool_ops e1000_ethtool_ops = {
+	.supported_coalesce_params = ETHTOOL_COALESCE_RX_USECS,
 	.get_drvinfo		= e1000_get_drvinfo,
 	.get_regs_len		= e1000_get_regs_len,
 	.get_regs		= e1000_get_regs,
diff --git a/drivers/net/ethernet/intel/e1000/e1000_hw.c b/drivers/net/ethernet/intel/e1000/e1000_hw.c
index 8172cf08cc33..0e5de52b1067 100644
--- a/drivers/net/ethernet/intel/e1000/e1000_hw.c
+++ b/drivers/net/ethernet/intel/e1000/e1000_hw.c
@@ -1,35 +1,11 @@
-/*******************************************************************************
-*
-  Intel PRO/1000 Linux driver
-  Copyright(c) 1999 - 2006 Intel Corporation.
-
-  This program is free software; you can redistribute it and/or modify it
-  under the terms and conditions of the GNU General Public License,
-  version 2, as published by the Free Software Foundation.
-
-  This program is distributed in the hope it will be useful, but WITHOUT
-  ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
-  FITNESS FOR A PARTICULAR PURPOSE.  See the GNU General Public License for
-  more details.
-
-  You should have received a copy of the GNU General Public License along with
-  this program; if not, write to the Free Software Foundation, Inc.,
-  51 Franklin St - Fifth Floor, Boston, MA 02110-1301 USA.
-
-  The full GNU General Public License is included in this distribution in
-  the file called "COPYING".
-
-  Contact Information:
-  Linux NICS <linux.nics@intel.com>
-  e1000-devel Mailing List <e1000-devel@lists.sourceforge.net>
-  Intel Corporation, 5200 N.E. Elam Young Parkway, Hillsboro, OR 97124-6497
-
- */
+// SPDX-License-Identifier: GPL-2.0
+/* Copyright(c) 1999 - 2006 Intel Corporation. */
 
 /* e1000_hw.c
  * Shared functions for accessing and configuring the MAC
  */
 
+#include <linux/bitfield.h>
 #include "e1000.h"
 
 static s32 e1000_check_downshift(struct e1000_hw *hw);
@@ -154,7 +130,6 @@ static s32 e1000_set_phy_type(struct e1000_hw *hw)
  */
 static void e1000_phy_init_script(struct e1000_hw *hw)
 {
-	u32 ret_val;
 	u16 phy_saved_data;
 
 	if (hw->phy_init_script) {
@@ -163,7 +138,7 @@ static void e1000_phy_init_script(struct e1000_hw *hw)
 		/* Save off the current value of register 0x2F5B to be restored
 		 * at the end of this routine.
 		 */
-		ret_val = e1000_read_phy_reg(hw, 0x2F5B, &phy_saved_data);
+		e1000_read_phy_reg(hw, 0x2F5B, &phy_saved_data);
 
 		/* Disabled the PHY transmitter */
 		e1000_write_phy_reg(hw, 0x2F5B, 0x0003);
@@ -402,7 +377,6 @@ s32 e1000_reset_hw(struct e1000_hw *hw)
 {
 	u32 ctrl;
 	u32 ctrl_ext;
-	u32 icr;
 	u32 manc;
 	u32 led_ctrl;
 	s32 ret_val;
@@ -527,7 +501,7 @@ s32 e1000_reset_hw(struct e1000_hw *hw)
 	ew32(IMC, 0xffffffff);
 
 	/* Clear any pending interrupt events. */
-	icr = er32(ICR);
+	er32(ICR);
 
 	/* If MWI was previously enabled, reenable it. */
 	if (hw->mac_type == e1000_82542_rev2_0) {
@@ -1210,6 +1184,7 @@ static s32 e1000_copper_link_igp_setup(struct e1000_hw *hw)
 			break;
 		case e1000_ms_auto:
 			phy_data &= ~CR_1000T_MS_ENABLE;
+			break;
 		default:
 			break;
 		}
@@ -1922,7 +1897,6 @@ void e1000_config_collision_dist(struct e1000_hw *hw)
 /**
  * e1000_config_mac_to_phy - sync phy and mac settings
  * @hw: Struct containing variables accessed by shared code
- * @mii_reg: data to write to the MII control register
  *
  * Sets MAC speed and duplex settings to reflect the those in the PHY
  * The contents of the PHY register containing the needed information need to
@@ -2027,7 +2001,7 @@ s32 e1000_force_mac_fc(struct e1000_hw *hw)
 	 *      1:  Rx flow control is enabled (we can receive pause
 	 *          frames but not send pause frames).
 	 *      2:  Tx flow control is enabled (we can send pause frames
-	 *          frames but we do not receive pause frames).
+	 *          but we do not receive pause frames).
 	 *      3:  Both Rx and TX flow control (symmetric) is enabled.
 	 *  other:  No other values should be possible at this point.
 	 */
@@ -2395,16 +2369,13 @@ static s32 e1000_check_for_serdes_link_generic(struct e1000_hw *hw)
  */
 s32 e1000_check_for_link(struct e1000_hw *hw)
 {
-	u32 rxcw = 0;
-	u32 ctrl;
 	u32 status;
 	u32 rctl;
 	u32 icr;
-	u32 signal = 0;
 	s32 ret_val;
 	u16 phy_data;
 
-	ctrl = er32(CTRL);
+	er32(CTRL);
 	status = er32(STATUS);
 
 	/* On adapters with a MAC newer than 82544, SW Definable pin 1 will be
@@ -2413,12 +2384,9 @@ s32 e1000_check_for_link(struct e1000_hw *hw)
 	 */
 	if ((hw->media_type == e1000_media_type_fiber) ||
 	    (hw->media_type == e1000_media_type_internal_serdes)) {
-		rxcw = er32(RXCW);
+		er32(RXCW);
 
 		if (hw->media_type == e1000_media_type_fiber) {
-			signal =
-			    (hw->mac_type >
-			     e1000_82544) ? E1000_CTRL_SWDPIN1 : 0;
 			if (status & E1000_STATUS_LU)
 				hw->get_link_status = false;
 		}
@@ -2555,7 +2523,7 @@ s32 e1000_check_for_link(struct e1000_hw *hw)
 				 * turn it on. For compatibility with a TBI link
 				 * partner, we will store bad packets. Some
 				 * frames have an additional byte on the end and
-				 * will look like CRC errors to to the hardware.
+				 * will look like CRC errors to the hardware.
 				 */
 				if (!hw->tbi_compatibility_on) {
 					hw->tbi_compatibility_on = true;
@@ -2756,7 +2724,7 @@ static void e1000_shift_out_mdi_bits(struct e1000_hw *hw, u32 data, u16 count)
  * e1000_shift_in_mdi_bits - Shifts data bits in from the PHY
  * @hw: Struct containing variables accessed by shared code
  *
- * Bits are shifted in in MSB to LSB order.
+ * Bits are shifted in MSB to LSB order.
  */
 static u16 e1000_shift_in_mdi_bits(struct e1000_hw *hw)
 {
@@ -2947,7 +2915,7 @@ static s32 e1000_read_phy_reg_ex(struct e1000_hw *hw, u32 reg_addr,
  *
  * @hw: Struct containing variables accessed by shared code
  * @reg_addr: address of the PHY register to write
- * @data: data to write to the PHY
+ * @phy_data: data to write to the PHY
  *
  * Writes a value to a PHY register
  */
@@ -3293,8 +3261,7 @@ static s32 e1000_phy_igp_get_info(struct e1000_hw *hw,
 		return ret_val;
 
 	phy_info->mdix_mode =
-	    (e1000_auto_x_mode) ((phy_data & IGP01E1000_PSSR_MDIX) >>
-				 IGP01E1000_PSSR_MDIX_SHIFT);
+	    (e1000_auto_x_mode)FIELD_GET(IGP01E1000_PSSR_MDIX, phy_data);
 
 	if ((phy_data & IGP01E1000_PSSR_SPEED_MASK) ==
 	    IGP01E1000_PSSR_SPEED_1000MBPS) {
@@ -3305,11 +3272,11 @@ static s32 e1000_phy_igp_get_info(struct e1000_hw *hw,
 		if (ret_val)
 			return ret_val;
 
-		phy_info->local_rx = ((phy_data & SR_1000T_LOCAL_RX_STATUS) >>
-				      SR_1000T_LOCAL_RX_STATUS_SHIFT) ?
+		phy_info->local_rx = FIELD_GET(SR_1000T_LOCAL_RX_STATUS,
+					       phy_data) ?
 		    e1000_1000t_rx_status_ok : e1000_1000t_rx_status_not_ok;
-		phy_info->remote_rx = ((phy_data & SR_1000T_REMOTE_RX_STATUS) >>
-				       SR_1000T_REMOTE_RX_STATUS_SHIFT) ?
+		phy_info->remote_rx = FIELD_GET(SR_1000T_REMOTE_RX_STATUS,
+						phy_data) ?
 		    e1000_1000t_rx_status_ok : e1000_1000t_rx_status_not_ok;
 
 		/* Get cable length */
@@ -3359,14 +3326,12 @@ static s32 e1000_phy_m88_get_info(struct e1000_hw *hw,
 		return ret_val;
 
 	phy_info->extended_10bt_distance =
-	    ((phy_data & M88E1000_PSCR_10BT_EXT_DIST_ENABLE) >>
-	     M88E1000_PSCR_10BT_EXT_DIST_ENABLE_SHIFT) ?
+	    FIELD_GET(M88E1000_PSCR_10BT_EXT_DIST_ENABLE, phy_data) ?
 	    e1000_10bt_ext_dist_enable_lower :
 	    e1000_10bt_ext_dist_enable_normal;
 
 	phy_info->polarity_correction =
-	    ((phy_data & M88E1000_PSCR_POLARITY_REVERSAL) >>
-	     M88E1000_PSCR_POLARITY_REVERSAL_SHIFT) ?
+	    FIELD_GET(M88E1000_PSCR_POLARITY_REVERSAL, phy_data) ?
 	    e1000_polarity_reversal_disabled : e1000_polarity_reversal_enabled;
 
 	/* Check polarity status */
@@ -3380,27 +3345,25 @@ static s32 e1000_phy_m88_get_info(struct e1000_hw *hw,
 		return ret_val;
 
 	phy_info->mdix_mode =
-	    (e1000_auto_x_mode) ((phy_data & M88E1000_PSSR_MDIX) >>
-				 M88E1000_PSSR_MDIX_SHIFT);
+	    (e1000_auto_x_mode)FIELD_GET(M88E1000_PSSR_MDIX, phy_data);
 
 	if ((phy_data & M88E1000_PSSR_SPEED) == M88E1000_PSSR_1000MBS) {
 		/* Cable Length Estimation and Local/Remote Receiver Information
 		 * are only valid at 1000 Mbps.
 		 */
 		phy_info->cable_length =
-		    (e1000_cable_length) ((phy_data &
-					   M88E1000_PSSR_CABLE_LENGTH) >>
-					  M88E1000_PSSR_CABLE_LENGTH_SHIFT);
+		    (e1000_cable_length)FIELD_GET(M88E1000_PSSR_CABLE_LENGTH,
+						  phy_data);
 
 		ret_val = e1000_read_phy_reg(hw, PHY_1000T_STATUS, &phy_data);
 		if (ret_val)
 			return ret_val;
 
-		phy_info->local_rx = ((phy_data & SR_1000T_LOCAL_RX_STATUS) >>
-				      SR_1000T_LOCAL_RX_STATUS_SHIFT) ?
+		phy_info->local_rx = FIELD_GET(SR_1000T_LOCAL_RX_STATUS,
+					       phy_data) ?
 		    e1000_1000t_rx_status_ok : e1000_1000t_rx_status_not_ok;
-		phy_info->remote_rx = ((phy_data & SR_1000T_REMOTE_RX_STATUS) >>
-				       SR_1000T_REMOTE_RX_STATUS_SHIFT) ?
+		phy_info->remote_rx = FIELD_GET(SR_1000T_REMOTE_RX_STATUS,
+						phy_data) ?
 		    e1000_1000t_rx_status_ok : e1000_1000t_rx_status_not_ok;
 	}
 
@@ -3548,7 +3511,7 @@ s32 e1000_init_eeprom_params(struct e1000_hw *hw)
 		if (ret_val)
 			return ret_val;
 		eeprom_size =
-		    (eeprom_size & EEPROM_SIZE_MASK) >> EEPROM_SIZE_SHIFT;
+		    FIELD_GET(EEPROM_SIZE_MASK, eeprom_size);
 		/* 256B eeprom size was not supported in earlier hardware, so we
 		 * bump eeprom_size up one to ensure that "1" (which maps to
 		 * 256B) is never the result used in the shifting logic below.
@@ -3985,7 +3948,7 @@ static s32 e1000_do_read_eeprom(struct e1000_hw *hw, u16 offset, u16 words,
  * @hw: Struct containing variables accessed by shared code
  *
  * Reads the first 64 16 bit words of the EEPROM and sums the values read.
- * If the the sum of the 64 16 bit words is 0xBABA, the EEPROM's checksum is
+ * If the sum of the 64 16 bit words is 0xBABA, the EEPROM's checksum is
  * valid.
  */
 s32 e1000_validate_eeprom_checksum(struct e1000_hw *hw)
@@ -4007,7 +3970,7 @@ s32 e1000_validate_eeprom_checksum(struct e1000_hw *hw)
 		return E1000_SUCCESS;
 
 #endif
-	if (checksum == (u16)EEPROM_SUM)
+	if (checksum == EEPROM_SUM)
 		return E1000_SUCCESS;
 	else {
 		e_dbg("EEPROM Checksum Invalid\n");
@@ -4034,7 +3997,7 @@ s32 e1000_update_eeprom_checksum(struct e1000_hw *hw)
 		}
 		checksum += eeprom_data;
 	}
-	checksum = (u16)EEPROM_SUM - checksum;
+	checksum = EEPROM_SUM - checksum;
 	if (e1000_write_eeprom(hw, EEPROM_CHECKSUM_REG, 1, &checksum) < 0) {
 		e_dbg("EEPROM Write Error\n");
 		return -E1000_ERR_EEPROM;
@@ -4307,8 +4270,10 @@ static void e1000_init_rx_addrs(struct e1000_hw *hw)
 
 	rar_num = E1000_RAR_ENTRIES;
 
-	/* Zero out the other 15 receive addresses. */
-	e_dbg("Clearing RAR[1-15]\n");
+	/* Zero out the following 14 receive addresses. RAR[15] is for
+	 * manageability
+	 */
+	e_dbg("Clearing RAR[1-14]\n");
 	for (i = 1; i < rar_num; i++) {
 		E1000_WRITE_REG_ARRAY(hw, RA, (i << 1), 0);
 		E1000_WRITE_FLUSH();
@@ -4407,7 +4372,7 @@ void e1000_rar_set(struct e1000_hw *hw, u8 *addr, u32 index)
 /**
  * e1000_write_vfta - Writes a value to the specified offset in the VLAN filter table.
  * @hw: Struct containing variables accessed by shared code
- * @offset: Offset in VLAN filer table to write
+ * @offset: Offset in VLAN filter table to write
  * @value: Value to write into VLAN filter table
  */
 void e1000_write_vfta(struct e1000_hw *hw, u32 offset, u32 value)
@@ -4427,23 +4392,15 @@ void e1000_write_vfta(struct e1000_hw *hw, u32 offset, u32 value)
 }
 
 /**
- * e1000_clear_vfta - Clears the VLAN filer table
+ * e1000_clear_vfta - Clears the VLAN filter table
  * @hw: Struct containing variables accessed by shared code
  */
 static void e1000_clear_vfta(struct e1000_hw *hw)
 {
 	u32 offset;
-	u32 vfta_value = 0;
-	u32 vfta_offset = 0;
-	u32 vfta_bit_in_reg = 0;
 
 	for (offset = 0; offset < E1000_VLAN_FILTER_TBL_SIZE; offset++) {
-		/* If the offset we want to clear is the same offset of the
-		 * manageability VLAN ID, then clear all bits except that of the
-		 * manageability unit
-		 */
-		vfta_value = (offset == vfta_offset) ? vfta_bit_in_reg : 0;
-		E1000_WRITE_REG_ARRAY(hw, VFTA, offset, vfta_value);
+		E1000_WRITE_REG_ARRAY(hw, VFTA, offset, 0);
 		E1000_WRITE_FLUSH();
 	}
 }
@@ -4549,7 +4506,7 @@ s32 e1000_setup_led(struct e1000_hw *hw)
 						     ~IGP01E1000_GMII_SPD));
 		if (ret_val)
 			return ret_val;
-		/* Fall Through */
+		fallthrough;
 	default:
 		if (hw->media_type == e1000_media_type_fiber) {
 			ledctl = er32(LEDCTL);
@@ -4594,7 +4551,7 @@ s32 e1000_cleanup_led(struct e1000_hw *hw)
 					      hw->phy_spd_default);
 		if (ret_val)
 			return ret_val;
-		/* Fall Through */
+		fallthrough;
 	default:
 		/* Restore LEDCTL settings */
 		ew32(LEDCTL, hw->ledctl_default);
@@ -4698,78 +4655,76 @@ s32 e1000_led_off(struct e1000_hw *hw)
  */
 static void e1000_clear_hw_cntrs(struct e1000_hw *hw)
 {
-	volatile u32 temp;
-
-	temp = er32(CRCERRS);
-	temp = er32(SYMERRS);
-	temp = er32(MPC);
-	temp = er32(SCC);
-	temp = er32(ECOL);
-	temp = er32(MCC);
-	temp = er32(LATECOL);
-	temp = er32(COLC);
-	temp = er32(DC);
-	temp = er32(SEC);
-	temp = er32(RLEC);
-	temp = er32(XONRXC);
-	temp = er32(XONTXC);
-	temp = er32(XOFFRXC);
-	temp = er32(XOFFTXC);
-	temp = er32(FCRUC);
-
-	temp = er32(PRC64);
-	temp = er32(PRC127);
-	temp = er32(PRC255);
-	temp = er32(PRC511);
-	temp = er32(PRC1023);
-	temp = er32(PRC1522);
-
-	temp = er32(GPRC);
-	temp = er32(BPRC);
-	temp = er32(MPRC);
-	temp = er32(GPTC);
-	temp = er32(GORCL);
-	temp = er32(GORCH);
-	temp = er32(GOTCL);
-	temp = er32(GOTCH);
-	temp = er32(RNBC);
-	temp = er32(RUC);
-	temp = er32(RFC);
-	temp = er32(ROC);
-	temp = er32(RJC);
-	temp = er32(TORL);
-	temp = er32(TORH);
-	temp = er32(TOTL);
-	temp = er32(TOTH);
-	temp = er32(TPR);
-	temp = er32(TPT);
-
-	temp = er32(PTC64);
-	temp = er32(PTC127);
-	temp = er32(PTC255);
-	temp = er32(PTC511);
-	temp = er32(PTC1023);
-	temp = er32(PTC1522);
-
-	temp = er32(MPTC);
-	temp = er32(BPTC);
+	er32(CRCERRS);
+	er32(SYMERRS);
+	er32(MPC);
+	er32(SCC);
+	er32(ECOL);
+	er32(MCC);
+	er32(LATECOL);
+	er32(COLC);
+	er32(DC);
+	er32(SEC);
+	er32(RLEC);
+	er32(XONRXC);
+	er32(XONTXC);
+	er32(XOFFRXC);
+	er32(XOFFTXC);
+	er32(FCRUC);
+
+	er32(PRC64);
+	er32(PRC127);
+	er32(PRC255);
+	er32(PRC511);
+	er32(PRC1023);
+	er32(PRC1522);
+
+	er32(GPRC);
+	er32(BPRC);
+	er32(MPRC);
+	er32(GPTC);
+	er32(GORCL);
+	er32(GORCH);
+	er32(GOTCL);
+	er32(GOTCH);
+	er32(RNBC);
+	er32(RUC);
+	er32(RFC);
+	er32(ROC);
+	er32(RJC);
+	er32(TORL);
+	er32(TORH);
+	er32(TOTL);
+	er32(TOTH);
+	er32(TPR);
+	er32(TPT);
+
+	er32(PTC64);
+	er32(PTC127);
+	er32(PTC255);
+	er32(PTC511);
+	er32(PTC1023);
+	er32(PTC1522);
+
+	er32(MPTC);
+	er32(BPTC);
 
 	if (hw->mac_type < e1000_82543)
 		return;
 
-	temp = er32(ALGNERRC);
-	temp = er32(RXERRC);
-	temp = er32(TNCRS);
-	temp = er32(CEXTERR);
-	temp = er32(TSCTC);
-	temp = er32(TSCTFC);
+	er32(ALGNERRC);
+	er32(RXERRC);
+	er32(TNCRS);
+	er32(CEXTERR);
+	er32(TSCTC);
+	er32(TSCTFC);
 
 	if (hw->mac_type <= e1000_82544)
 		return;
 
-	temp = er32(MGTPRC);
-	temp = er32(MGTPDC);
-	temp = er32(MGTPTC);
+	er32(MGTPRC);
+	er32(MGTPDC);
+	er32(MGTPTC);
 }
 
 /**
@@ -4801,8 +4756,6 @@ void e1000_reset_adaptive(struct e1000_hw *hw)
 /**
  * e1000_update_adaptive - update adaptive IFS
  * @hw: Struct containing variables accessed by shared code
- * @tx_packets: Number of transmits since last callback
- * @total_collisions: Number of collisions since last callback
  *
  * Called during the callback/watchdog routine to update IFS value based on
  * the ratio of transmits to collisions.
@@ -4934,8 +4887,7 @@ static s32 e1000_get_cable_length(struct e1000_hw *hw, u16 *min_length,
 					     &phy_data);
 		if (ret_val)
 			return ret_val;
-		cable_length = (phy_data & M88E1000_PSSR_CABLE_LENGTH) >>
-		    M88E1000_PSSR_CABLE_LENGTH_SHIFT;
+		cable_length = FIELD_GET(M88E1000_PSSR_CABLE_LENGTH, phy_data);
 
 		/* Convert the enum value to ranged values */
 		switch (cable_length) {
@@ -5044,8 +4996,7 @@ static s32 e1000_check_polarity(struct e1000_hw *hw,
 					     &phy_data);
 		if (ret_val)
 			return ret_val;
-		*polarity = ((phy_data & M88E1000_PSSR_REV_POLARITY) >>
-			     M88E1000_PSSR_REV_POLARITY_SHIFT) ?
+		*polarity = FIELD_GET(M88E1000_PSSR_REV_POLARITY, phy_data) ?
 		    e1000_rev_polarity_reversed : e1000_rev_polarity_normal;
 
 	} else if (hw->phy_type == e1000_phy_igp) {
@@ -5087,8 +5038,6 @@ static s32 e1000_check_polarity(struct e1000_hw *hw,
 /**
  * e1000_check_downshift - Check if Downshift occurred
  * @hw: Struct containing variables accessed by shared code
- * @downshift: output parameter : 0 - No Downshift occurred.
- *                                1 - Downshift occurred.
  *
  * returns: - E1000_ERR_XXX
  *            E1000_SUCCESS
@@ -5117,8 +5066,8 @@ static s32 e1000_check_downshift(struct e1000_hw *hw)
 		if (ret_val)
 			return ret_val;
 
-		hw->speed_downgraded = (phy_data & M88E1000_PSSR_DOWNSHIFT) >>
-		    M88E1000_PSSR_DOWNSHIFT_SHIFT;
+		hw->speed_downgraded = FIELD_GET(M88E1000_PSSR_DOWNSHIFT,
+						 phy_data);
 	}
 
 	return E1000_SUCCESS;
diff --git a/drivers/net/ethernet/intel/e1000/e1000_hw.h b/drivers/net/ethernet/intel/e1000/e1000_hw.h
index 5cf7268cc4e1..95cdd17134e5 100644
--- a/drivers/net/ethernet/intel/e1000/e1000_hw.h
+++ b/drivers/net/ethernet/intel/e1000/e1000_hw.h
@@ -1,30 +1,5 @@
-/*******************************************************************************
-
-  Intel PRO/1000 Linux driver
-  Copyright(c) 1999 - 2006 Intel Corporation.
-
-  This program is free software; you can redistribute it and/or modify it
-  under the terms and conditions of the GNU General Public License,
-  version 2, as published by the Free Software Foundation.
-
-  This program is distributed in the hope it will be useful, but WITHOUT
-  ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
-  FITNESS FOR A PARTICULAR PURPOSE.  See the GNU General Public License for
-  more details.
-
-  You should have received a copy of the GNU General Public License along with
-  this program; if not, write to the Free Software Foundation, Inc.,
-  51 Franklin St - Fifth Floor, Boston, MA 02110-1301 USA.
-
-  The full GNU General Public License is included in this distribution in
-  the file called "COPYING".
-
-  Contact Information:
-  Linux NICS <linux.nics@intel.com>
-  e1000-devel Mailing List <e1000-devel@lists.sourceforge.net>
-  Intel Corporation, 5200 N.E. Elam Young Parkway, Hillsboro, OR 97124-6497
-
-*******************************************************************************/
+/* SPDX-License-Identifier: GPL-2.0 */
+/* Copyright(c) 1999 - 2006 Intel Corporation. */
 
 /* e1000_hw.h
  * Structures, enums, and macros for the MAC
@@ -368,7 +343,6 @@ struct e1000_host_mng_dhcp_cookie {
 };
 #endif
 
-bool e1000_check_mng_mode(struct e1000_hw *hw);
 s32 e1000_read_eeprom(struct e1000_hw *hw, u16 reg, u16 words, u16 * data);
 s32 e1000_validate_eeprom_checksum(struct e1000_hw *hw);
 s32 e1000_update_eeprom_checksum(struct e1000_hw *hw);
@@ -377,7 +351,6 @@ s32 e1000_read_mac_addr(struct e1000_hw *hw);
 
 /* Filters (multicast, vlan, receive) */
 u32 e1000_hash_mc_addr(struct e1000_hw *hw, u8 * mc_addr);
-void e1000_mta_set(struct e1000_hw *hw, u32 hash_value);
 void e1000_rar_set(struct e1000_hw *hw, u8 * mc_addr, u32 rar_index);
 void e1000_write_vfta(struct e1000_hw *hw, u32 offset, u32 value);
 
@@ -386,7 +359,6 @@ s32 e1000_setup_led(struct e1000_hw *hw);
 s32 e1000_cleanup_led(struct e1000_hw *hw);
 s32 e1000_led_on(struct e1000_hw *hw);
 s32 e1000_led_off(struct e1000_hw *hw);
-s32 e1000_blink_led_start(struct e1000_hw *hw);
 
 /* Adaptive IFS Functions */
 
diff --git a/drivers/net/ethernet/intel/e1000/e1000_main.c b/drivers/net/ethernet/intel/e1000/e1000_main.c
index bd8b05fe8258..292389aceb2d 100644
--- a/drivers/net/ethernet/intel/e1000/e1000_main.c
+++ b/drivers/net/ethernet/intel/e1000/e1000_main.c
@@ -1,30 +1,5 @@
-/*******************************************************************************
-
-  Intel PRO/1000 Linux driver
-  Copyright(c) 1999 - 2006 Intel Corporation.
-
-  This program is free software; you can redistribute it and/or modify it
-  under the terms and conditions of the GNU General Public License,
-  version 2, as published by the Free Software Foundation.
-
-  This program is distributed in the hope it will be useful, but WITHOUT
-  ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
-  FITNESS FOR A PARTICULAR PURPOSE.  See the GNU General Public License for
-  more details.
-
-  You should have received a copy of the GNU General Public License along with
-  this program; if not, write to the Free Software Foundation, Inc.,
-  51 Franklin St - Fifth Floor, Boston, MA 02110-1301 USA.
-
-  The full GNU General Public License is included in this distribution in
-  the file called "COPYING".
-
-  Contact Information:
-  Linux NICS <linux.nics@intel.com>
-  e1000-devel Mailing List <e1000-devel@lists.sourceforge.net>
-  Intel Corporation, 5200 N.E. Elam Young Parkway, Hillsboro, OR 97124-6497
-
-*******************************************************************************/
+// SPDX-License-Identifier: GPL-2.0
+/* Copyright(c) 1999 - 2006 Intel Corporation. */
 
 #include "e1000.h"
 #include <net/ip6_checksum.h>
@@ -35,8 +10,6 @@
 
 char e1000_driver_name[] = "e1000";
 static char e1000_driver_string[] = "Intel(R) PRO/1000 Network Driver";
-#define DRV_VERSION "7.3.21-k8-NAPI"
-const char e1000_driver_version[] = DRV_VERSION;
 static const char e1000_copyright[] = "Copyright (c) 1999-2006 Intel Corporation.";
 
 /* e1000_pci_tbl - PCI Device ID Table
@@ -159,7 +132,7 @@ static int e1000_mii_ioctl(struct net_device *netdev, struct ifreq *ifr,
 			   int cmd);
 static void e1000_enter_82542_rst(struct e1000_adapter *adapter);
 static void e1000_leave_82542_rst(struct e1000_adapter *adapter);
-static void e1000_tx_timeout(struct net_device *dev);
+static void e1000_tx_timeout(struct net_device *dev, unsigned int txqueue);
 static void e1000_reset_task(struct work_struct *work);
 static void e1000_smartspeed(struct e1000_adapter *adapter);
 static int e1000_82547_fifo_workaround(struct e1000_adapter *adapter,
@@ -176,10 +149,8 @@ static int e1000_vlan_rx_kill_vid(struct net_device *netdev,
 				  __be16 proto, u16 vid);
 static void e1000_restore_vlan(struct e1000_adapter *adapter);
 
-#ifdef CONFIG_PM
-static int e1000_suspend(struct pci_dev *pdev, pm_message_t state);
-static int e1000_resume(struct pci_dev *pdev);
-#endif
+static int e1000_suspend(struct device *dev);
+static int e1000_resume(struct device *dev);
 static void e1000_shutdown(struct pci_dev *pdev);
 
 #ifdef CONFIG_NET_POLL_CONTROLLER
@@ -204,24 +175,20 @@ static const struct pci_error_handlers e1000_err_handler = {
 	.resume = e1000_io_resume,
 };
 
+static DEFINE_SIMPLE_DEV_PM_OPS(e1000_pm_ops, e1000_suspend, e1000_resume);
+
 static struct pci_driver e1000_driver = {
 	.name     = e1000_driver_name,
 	.id_table = e1000_pci_tbl,
 	.probe    = e1000_probe,
 	.remove   = e1000_remove,
-#ifdef CONFIG_PM
-	/* Power Management Hooks */
-	.suspend  = e1000_suspend,
-	.resume   = e1000_resume,
-#endif
+	.driver.pm = pm_sleep_ptr(&e1000_pm_ops),
 	.shutdown = e1000_shutdown,
 	.err_handler = &e1000_err_handler
 };
 
-MODULE_AUTHOR("Intel Corporation, <linux.nics@intel.com>");
 MODULE_DESCRIPTION("Intel(R) PRO/1000 Network Driver");
-MODULE_LICENSE("GPL");
-MODULE_VERSION(DRV_VERSION);
+MODULE_LICENSE("GPL v2");
 
 #define DEFAULT_MSG_ENABLE (NETIF_MSG_DRV|NETIF_MSG_PROBE|NETIF_MSG_LINK)
 static int debug = -1;
@@ -229,8 +196,10 @@ module_param(debug, int, 0);
 MODULE_PARM_DESC(debug, "Debug level (0=none,...,16=all)");
 
 /**
- * e1000_get_hw_dev - return device
- * used by hardware layer to print debugging information
+ * e1000_get_hw_dev - helper function for getting netdev
+ * @hw: pointer to HW struct
+ *
+ * return device used by hardware layer to print debugging information
  *
  **/
 struct net_device *e1000_get_hw_dev(struct e1000_hw *hw)
@@ -248,7 +217,7 @@ struct net_device *e1000_get_hw_dev(struct e1000_hw *hw)
 static int __init e1000_init_module(void)
 {
 	int ret;
-	pr_info("%s - version %s\n", e1000_driver_string, e1000_driver_version);
+	pr_info("%s\n", e1000_driver_string);
 
 	pr_info("%s\n", e1000_copyright);
 
@@ -344,8 +313,7 @@ static void e1000_update_mng_vlan(struct e1000_adapter *adapter)
 		} else {
 			adapter->mng_vlan_id = E1000_MNG_VLAN_NONE;
 		}
-		if ((old_vid != (u16)E1000_MNG_VLAN_NONE) &&
-		    (vid != old_vid) &&
+		if (old_vid != E1000_MNG_VLAN_NONE && vid != old_vid &&
 		    !test_bit(old_vid, adapter->active_vlans))
 			e1000_vlan_rx_kill_vid(netdev, htons(ETH_P_8021Q),
 					       old_vid);
@@ -384,7 +352,7 @@ static void e1000_release_manageability(struct e1000_adapter *adapter)
 
 /**
  * e1000_configure - configure the hardware for RX and TX
- * @adapter = private board structure
+ * @adapter: private board structure
  **/
 static void e1000_configure(struct e1000_adapter *adapter)
 {
@@ -508,10 +476,6 @@ static void e1000_down_and_stop(struct e1000_adapter *adapter)
 
 	cancel_delayed_work_sync(&adapter->phy_info_task);
 	cancel_delayed_work_sync(&adapter->fifo_stall_task);
-
-	/* Only kill reset task if adapter is not resetting */
-	if (!test_bit(__E1000_RESETTING, &adapter->flags))
-		cancel_work_sync(&adapter->reset_task);
 }
 
 void e1000_down(struct e1000_adapter *adapter)
@@ -520,8 +484,6 @@ void e1000_down(struct e1000_adapter *adapter)
 	struct net_device *netdev = adapter->netdev;
 	u32 rctl, tctl;
 
-	netif_carrier_off(netdev);
-
 	/* disable receives in the hardware */
 	rctl = er32(RCTL);
 	ew32(RCTL, rctl & ~E1000_RCTL_EN);
@@ -537,6 +499,17 @@ void e1000_down(struct e1000_adapter *adapter)
 	E1000_WRITE_FLUSH();
 	msleep(10);
 
+	/* Set the carrier off after transmits have been disabled in the
+	 * hardware, to avoid race conditions with e1000_watchdog() (which
+	 * may be running concurrently to us, checking for the carrier
+	 * bit to decide whether it should enable transmits again). Such
+	 * a race condition would result into transmission being disabled
+	 * in the hardware until the next IFF_DOWN+IFF_UP cycle.
+	 */
+	netif_carrier_off(netdev);
+
+	netif_queue_set_napi(netdev, 0, NETDEV_QUEUE_TYPE_RX, NULL);
+	netif_queue_set_napi(netdev, 0, NETDEV_QUEUE_TYPE_TX, NULL);
 	napi_disable(&adapter->napi);
 
 	e1000_irq_disable(adapter);
@@ -557,11 +530,15 @@ void e1000_down(struct e1000_adapter *adapter)
 
 void e1000_reinit_locked(struct e1000_adapter *adapter)
 {
-	WARN_ON(in_interrupt());
 	while (test_and_set_bit(__E1000_RESETTING, &adapter->flags))
 		msleep(1);
-	e1000_down(adapter);
-	e1000_up(adapter);
+
+	/* only run the task if not already down */
+	if (!test_bit(__E1000_DOWN, &adapter->flags)) {
+		e1000_down(adapter);
+		e1000_up(adapter);
+	}
+
 	clear_bit(__E1000_RESETTING, &adapter->flags);
 }
 
@@ -838,7 +815,7 @@ static int e1000_set_features(struct net_device *netdev,
 	else
 		e1000_reset(adapter);
 
-	return 0;
+	return 1;
 }
 
 static const struct net_device_ops e1000_netdev_ops = {
@@ -849,7 +826,7 @@ static const struct net_device_ops e1000_netdev_ops = {
 	.ndo_set_mac_address	= e1000_set_mac,
 	.ndo_tx_timeout		= e1000_tx_timeout,
 	.ndo_change_mtu		= e1000_change_mtu,
-	.ndo_do_ioctl		= e1000_ioctl,
+	.ndo_eth_ioctl		= e1000_ioctl,
 	.ndo_validate_addr	= eth_validate_addr,
 	.ndo_vlan_rx_add_vid	= e1000_vlan_rx_add_vid,
 	.ndo_vlan_rx_kill_vid	= e1000_vlan_rx_kill_vid,
@@ -938,7 +915,7 @@ static int e1000_init_hw_struct(struct e1000_adapter *adapter,
 static int e1000_probe(struct pci_dev *pdev, const struct pci_device_id *ent)
 {
 	struct net_device *netdev;
-	struct e1000_adapter *adapter;
+	struct e1000_adapter *adapter = NULL;
 	struct e1000_hw *hw;
 
 	static int cards_found;
@@ -948,6 +925,7 @@ static int e1000_probe(struct pci_dev *pdev, const struct pci_device_id *ent)
 	u16 tmp = 0;
 	u16 eeprom_apme_mask = E1000_EEPROM_APME;
 	int bars, need_ioport;
+	bool disable_dev = false;
 
 	/* do not allocate ioport bars when not needed */
 	need_ioport = e1000_is_need_ioport(pdev);
@@ -994,7 +972,7 @@ static int e1000_probe(struct pci_dev *pdev, const struct pci_device_id *ent)
 		goto err_ioremap;
 
 	if (adapter->need_ioport) {
-		for (i = BAR_1; i <= BAR_5; i++) {
+		for (i = BAR_1; i < PCI_STD_NUM_BARS; i++) {
 			if (pci_resource_len(pdev, i) == 0)
 				continue;
 			if (pci_resource_flags(pdev, i) & IORESOURCE_IO) {
@@ -1028,9 +1006,9 @@ static int e1000_probe(struct pci_dev *pdev, const struct pci_device_id *ent)
 	netdev->netdev_ops = &e1000_netdev_ops;
 	e1000_set_ethtool_ops(netdev);
 	netdev->watchdog_timeo = 5 * HZ;
-	netif_napi_add(netdev, &adapter->napi, e1000_clean, 64);
+	netif_napi_add(netdev, &adapter->napi, e1000_clean);
 
-	strncpy(netdev->name, pci_name(pdev), sizeof(netdev->name) - 1);
+	strscpy(netdev->name, pci_name(pdev), sizeof(netdev->name));
 
 	adapter->bd_number = cards_found;
 
@@ -1119,7 +1097,7 @@ static int e1000_probe(struct pci_dev *pdev, const struct pci_device_id *ent)
 			e_err(probe, "EEPROM Read Error\n");
 	}
 	/* don't block initialization here due to bad MAC address */
-	memcpy(netdev->dev_addr, hw->mac_addr, netdev->addr_len);
+	eth_hw_addr_set(netdev, hw->mac_addr);
 
 	if (!is_valid_ether_addr(netdev->dev_addr))
 		e_err(probe, "Invalid MAC Address\n");
@@ -1155,7 +1133,7 @@ static int e1000_probe(struct pci_dev *pdev, const struct pci_device_id *ent)
 				EEPROM_INIT_CONTROL3_PORT_B, 1, &eeprom_data);
 			break;
 		}
-		/* Fall Through */
+		fallthrough;
 	default:
 		e1000_read_eeprom(hw,
 			EEPROM_INIT_CONTROL3_PORT_A, 1, &eeprom_data);
@@ -1252,11 +1230,13 @@ err_mdio_ioremap:
 	iounmap(hw->ce4100_gbe_mdio_base_virt);
 	iounmap(hw->hw_addr);
 err_ioremap:
+	disable_dev = !test_and_set_bit(__E1000_DISABLED, &adapter->flags);
 	free_netdev(netdev);
 err_alloc_etherdev:
 	pci_release_selected_regions(pdev, bars);
 err_pci_reg:
-	pci_disable_device(pdev);
+	if (!adapter || disable_dev)
+		pci_disable_device(pdev);
 	return err;
 }
 
@@ -1274,12 +1254,17 @@ static void e1000_remove(struct pci_dev *pdev)
 	struct net_device *netdev = pci_get_drvdata(pdev);
 	struct e1000_adapter *adapter = netdev_priv(netdev);
 	struct e1000_hw *hw = &adapter->hw;
+	bool disable_dev;
 
 	e1000_down_and_stop(adapter);
 	e1000_release_manageability(adapter);
 
 	unregister_netdev(netdev);
 
+	/* Only kill reset task if adapter is not resetting */
+	if (!test_bit(__E1000_RESETTING, &adapter->flags))
+		cancel_work_sync(&adapter->reset_task);
+
 	e1000_phy_hw_reset(hw);
 
 	kfree(adapter->tx_ring);
@@ -1292,9 +1277,11 @@ static void e1000_remove(struct pci_dev *pdev)
 		iounmap(hw->flash_address);
 	pci_release_selected_regions(pdev, adapter->bars);
 
+	disable_dev = !test_and_set_bit(__E1000_DISABLED, &adapter->flags);
 	free_netdev(netdev);
 
-	pci_disable_device(pdev);
+	if (disable_dev)
+		pci_disable_device(pdev);
 }
 
 /**
@@ -1406,7 +1393,10 @@ int e1000_open(struct net_device *netdev)
 	/* From here on the code is the same as e1000_up() */
 	clear_bit(__E1000_DOWN, &adapter->flags);
 
+	netif_napi_set_irq(&adapter->napi, adapter->pdev->irq);
 	napi_enable(&adapter->napi);
+	netif_queue_set_napi(netdev, 0, NETDEV_QUEUE_TYPE_RX, &adapter->napi);
+	netif_queue_set_napi(netdev, 0, NETDEV_QUEUE_TYPE_TX, &adapter->napi);
 
 	e1000_irq_enable(adapter);
 
@@ -1445,10 +1435,15 @@ int e1000_close(struct net_device *netdev)
 	struct e1000_hw *hw = &adapter->hw;
 	int count = E1000_CHECK_RESET_COUNT;
 
-	while (test_bit(__E1000_RESETTING, &adapter->flags) && count--)
+	while (test_and_set_bit(__E1000_RESETTING, &adapter->flags) && count--)
 		usleep_range(10000, 20000);
 
-	WARN_ON(test_bit(__E1000_RESETTING, &adapter->flags));
+	WARN_ON(count < 0);
+
+	/* signal that we're down so that the reset task will no longer run */
+	set_bit(__E1000_DOWN, &adapter->flags);
+	clear_bit(__E1000_RESETTING, &adapter->flags);
+
 	e1000_down(adapter);
 	e1000_power_down_phy(adapter);
 	e1000_free_irq(adapter);
@@ -1488,7 +1483,7 @@ static bool e1000_check_64k_bound(struct e1000_adapter *adapter, void *start,
 	if (hw->mac_type == e1000_82545 ||
 	    hw->mac_type == e1000_ce4100 ||
 	    hw->mac_type == e1000_82546) {
-		return ((begin ^ (end - 1)) >> 16) != 0 ? false : true;
+		return ((begin ^ (end - 1)) >> 16) == 0;
 	}
 
 	return true;
@@ -1959,7 +1954,8 @@ void e1000_free_all_tx_resources(struct e1000_adapter *adapter)
 
 static void
 e1000_unmap_and_free_tx_resource(struct e1000_adapter *adapter,
-				 struct e1000_tx_buffer *buffer_info)
+				 struct e1000_tx_buffer *buffer_info,
+				 int budget)
 {
 	if (buffer_info->dma) {
 		if (buffer_info->mapped_as_page)
@@ -1972,7 +1968,7 @@ e1000_unmap_and_free_tx_resource(struct e1000_adapter *adapter,
 		buffer_info->dma = 0;
 	}
 	if (buffer_info->skb) {
-		dev_kfree_skb_any(buffer_info->skb);
+		napi_consume_skb(buffer_info->skb, budget);
 		buffer_info->skb = NULL;
 	}
 	buffer_info->time_stamp = 0;
@@ -1996,7 +1992,7 @@ static void e1000_clean_tx_ring(struct e1000_adapter *adapter,
 
 	for (i = 0; i < tx_ring->count; i++) {
 		buffer_info = &tx_ring->buffer_info[i];
-		e1000_unmap_and_free_tx_resource(adapter, buffer_info);
+		e1000_unmap_and_free_tx_resource(adapter, buffer_info, 0);
 	}
 
 	netdev_reset_queue(adapter->netdev);
@@ -2215,7 +2211,7 @@ static int e1000_set_mac(struct net_device *netdev, void *p)
 	if (hw->mac_type == e1000_82542_rev2_0)
 		e1000_enter_82542_rst(adapter);
 
-	memcpy(netdev->dev_addr, addr->sa_data, netdev->addr_len);
+	eth_hw_addr_set(netdev, addr->sa_data);
 	memcpy(hw->mac_addr, addr->sa_data, netdev->addr_len);
 
 	e1000_rar_set(hw, hw->mac_addr, 0);
@@ -2445,7 +2441,6 @@ static void e1000_watchdog(struct work_struct *work)
 	if (link) {
 		if (!netif_carrier_ok(netdev)) {
 			u32 ctrl;
-			bool txb2b = true;
 			/* update snapshot of PHY registers on LSC */
 			e1000_get_speed_and_duplex(hw,
 						   &adapter->link_speed,
@@ -2467,11 +2462,9 @@ static void e1000_watchdog(struct work_struct *work)
 			adapter->tx_timeout_factor = 1;
 			switch (adapter->link_speed) {
 			case SPEED_10:
-				txb2b = false;
 				adapter->tx_timeout_factor = 16;
 				break;
 			case SPEED_100:
-				txb2b = false;
 				/* maybe add some timeout factor ? */
 				break;
 			}
@@ -2641,7 +2634,6 @@ static void e1000_set_itr(struct e1000_adapter *adapter)
 
 	/* for non-gigabit speeds, just fix the interrupt rate at 4000 */
 	if (unlikely(adapter->link_speed != SPEED_1000)) {
-		current_itr = 0;
 		new_itr = 4000;
 		goto set_itr_now;
 	}
@@ -2717,7 +2709,7 @@ static int e1000_tso(struct e1000_adapter *adapter,
 		if (err < 0)
 			return err;
 
-		hdr_len = skb_transport_offset(skb) + tcp_hdrlen(skb);
+		hdr_len = skb_tcp_all_headers(skb);
 		mss = skb_shinfo(skb)->gso_size;
 		if (protocol == htons(ETH_P_IP)) {
 			struct iphdr *iph = ip_hdr(skb);
@@ -2730,11 +2722,7 @@ static int e1000_tso(struct e1000_adapter *adapter,
 			cmd_length = E1000_TXD_CMD_IP;
 			ipcse = skb_transport_offset(skb) - 1;
 		} else if (skb_is_gso_v6(skb)) {
-			ipv6_hdr(skb)->payload_len = 0;
-			tcp_hdr(skb)->check =
-				~csum_ipv6_magic(&ipv6_hdr(skb)->saddr,
-						 &ipv6_hdr(skb)->daddr,
-						 0, IPPROTO_TCP, 0);
+			tcp_v6_gso_csum_prep(skb);
 			ipcse = 0;
 		}
 		ipcss = skb_network_offset(skb);
@@ -2904,9 +2892,8 @@ static int e1000_tx_map(struct e1000_adapter *adapter,
 	}
 
 	for (f = 0; f < nr_frags; f++) {
-		const struct skb_frag_struct *frag;
+		const skb_frag_t *frag = &skb_shinfo(skb)->frags[f];
 
-		frag = &skb_shinfo(skb)->frags[f];
 		len = skb_frag_size(frag);
 		offset = 0;
 
@@ -2973,7 +2960,7 @@ dma_error:
 			i += tx_ring->count;
 		i--;
 		buffer_info = &tx_ring->buffer_info[i];
-		e1000_unmap_and_free_tx_resource(adapter, buffer_info);
+		e1000_unmap_and_free_tx_resource(adapter, buffer_info, 0);
 	}
 
 	return 0;
@@ -3034,7 +3021,7 @@ static void e1000_tx_queue(struct e1000_adapter *adapter,
 	 * applicable for weak-ordered memory model archs,
 	 * such as IA-64).
 	 */
-	wmb();
+	dma_wmb();
 
 	tx_ring->next_to_use = i;
 }
@@ -3153,11 +3140,12 @@ static netdev_tx_t e1000_xmit_frame(struct sk_buff *skb,
 		max_per_txd = min(mss << 2, max_per_txd);
 		max_txd_pwr = fls(max_per_txd) - 1;
 
-		hdr_len = skb_transport_offset(skb) + tcp_hdrlen(skb);
+		hdr_len = skb_tcp_all_headers(skb);
 		if (skb->data_len && hdr_len == len) {
 			switch (hw->mac_type) {
+			case e1000_82544: {
 				unsigned int pull_size;
-			case e1000_82544:
+
 				/* Make sure we have room to chop off 4 bytes,
 				 * and that the end alignment will work out to
 				 * this hardware's requirements
@@ -3168,7 +3156,6 @@ static netdev_tx_t e1000_xmit_frame(struct sk_buff *skb,
 				if ((unsigned long)(skb_tail_pointer(skb) - 1)
 				    & 4)
 					break;
-				/* fall through */
 				pull_size = min((unsigned int)4, skb->data_len);
 				if (!__pskb_pull_tail(skb, pull_size)) {
 					e_err(drv, "__pskb_pull_tail "
@@ -3178,6 +3165,7 @@ static netdev_tx_t e1000_xmit_frame(struct sk_buff *skb,
 				}
 				len = skb_headlen(skb);
 				break;
+			}
 			default:
 				/* do nothing */
 				break;
@@ -3282,14 +3270,9 @@ static netdev_tx_t e1000_xmit_frame(struct sk_buff *skb,
 		/* Make sure there is space in the ring for the next send. */
 		e1000_maybe_stop_tx(netdev, tx_ring, desc_needed);
 
-		if (!skb->xmit_more ||
+		if (!netdev_xmit_more() ||
 		    netif_xmit_stopped(netdev_get_tx_queue(netdev, 0))) {
 			writel(tx_ring->next_to_use, hw->hw_addr + tx_ring->tdt);
-			/* we need this if more than one processor can write to
-			 * our tail at a time, it synchronizes IO on IA64/Altix
-			 * systems
-			 */
-			mmiowb();
 		}
 	} else {
 		dev_kfree_skb_any(skb);
@@ -3508,8 +3491,9 @@ exit:
 /**
  * e1000_tx_timeout - Respond to a Tx Hang
  * @netdev: network interface device structure
+ * @txqueue: number of the Tx queue that hung (unused)
  **/
-static void e1000_tx_timeout(struct net_device *netdev)
+static void e1000_tx_timeout(struct net_device *netdev, unsigned int __always_unused txqueue)
 {
 	struct e1000_adapter *adapter = netdev_priv(netdev);
 
@@ -3524,7 +3508,9 @@ static void e1000_reset_task(struct work_struct *work)
 		container_of(work, struct e1000_adapter, reset_task);
 
 	e_err(drv, "Reset adapter\n");
+	rtnl_lock();
 	e1000_reinit_locked(adapter);
+	rtnl_unlock();
 }
 
 /**
@@ -3586,9 +3572,9 @@ static int e1000_change_mtu(struct net_device *netdev, int new_mtu)
 	     (max_frame == MAXIMUM_ETHERNET_VLAN_SIZE)))
 		adapter->rx_buffer_len = MAXIMUM_ETHERNET_VLAN_SIZE;
 
-	pr_info("%s changing MTU from %d to %d\n",
-		netdev->name, netdev->mtu, new_mtu);
-	netdev->mtu = new_mtu;
+	netdev_dbg(netdev, "changing MTU from %d to %d\n",
+		   netdev->mtu, new_mtu);
+	WRITE_ONCE(netdev->mtu, new_mtu);
 
 	if (netif_running(netdev))
 		e1000_up(adapter);
@@ -3806,7 +3792,8 @@ static irqreturn_t e1000_intr(int irq, void *data)
 
 /**
  * e1000_clean - NAPI Rx polling callback
- * @adapter: board private structure
+ * @napi: napi struct containing references to driver info
+ * @budget: budget given to driver for receive packets
  **/
 static int e1000_clean(struct napi_struct *napi, int budget)
 {
@@ -3818,14 +3805,15 @@ static int e1000_clean(struct napi_struct *napi, int budget)
 
 	adapter->clean_rx(adapter, &adapter->rx_ring[0], &work_done, budget);
 
-	if (!tx_clean_complete)
-		work_done = budget;
+	if (!tx_clean_complete || work_done == budget)
+		return budget;
 
-	/* If budget not fully consumed, exit the polling mode */
-	if (work_done < budget) {
+	/* Exit the polling mode, but don't re-enable interrupts if stack might
+	 * poll us due to busy-polling
+	 */
+	if (likely(napi_complete_done(napi, work_done))) {
 		if (likely(adapter->itr_setting & 3))
 			e1000_set_itr(adapter);
-		napi_complete_done(napi, work_done);
 		if (!test_bit(__E1000_DOWN, &adapter->flags))
 			e1000_irq_enable(adapter);
 	}
@@ -3836,6 +3824,7 @@ static int e1000_clean(struct napi_struct *napi, int budget)
 /**
  * e1000_clean_tx_irq - Reclaim resources after transmit completes
  * @adapter: board private structure
+ * @tx_ring: ring to clean
  **/
 static bool e1000_clean_tx_irq(struct e1000_adapter *adapter,
 			       struct e1000_tx_ring *tx_ring)
@@ -3871,7 +3860,8 @@ static bool e1000_clean_tx_irq(struct e1000_adapter *adapter,
 				}
 
 			}
-			e1000_unmap_and_free_tx_resource(adapter, buffer_info);
+			e1000_unmap_and_free_tx_resource(adapter, buffer_info,
+							 64);
 			tx_desc->upper.data = 0;
 
 			if (unlikely(++i == tx_ring->count))
@@ -3951,7 +3941,7 @@ static bool e1000_clean_tx_irq(struct e1000_adapter *adapter,
  * @adapter:     board private structure
  * @status_err:  receive descriptor status and error fields
  * @csum:        receive descriptor csum field
- * @sk_buff:     socket buffer with received data
+ * @skb:         socket buffer with received data
  **/
 static void e1000_rx_checksum(struct e1000_adapter *adapter, u32 status_err,
 			      u32 csum, struct sk_buff *skb)
@@ -3988,6 +3978,9 @@ static void e1000_rx_checksum(struct e1000_adapter *adapter, u32 status_err,
 
 /**
  * e1000_consume_page - helper function for jumbo Rx path
+ * @bi: software descriptor shadow data
+ * @skb: skb being modified
+ * @length: length of data being added
  **/
 static void e1000_consume_page(struct e1000_rx_buffer *bi, struct sk_buff *skb,
 			       u16 length)
@@ -4021,6 +4014,7 @@ static void e1000_receive_skb(struct e1000_adapter *adapter, u8 status,
 /**
  * e1000_tbi_adjust_stats
  * @hw: Struct containing variables accessed by shared code
+ * @stats: point to stats struct
  * @frame_len: The length of the frame in question
  * @mac_addr: The Ethernet destination address of the frame in question
  *
@@ -4195,8 +4189,7 @@ static bool e1000_clean_jumbo_rx_irq(struct e1000_adapter *adapter,
 				/* an error means any chain goes out the window
 				 * too
 				 */
-				if (rx_ring->rx_skb_top)
-					dev_kfree_skb(rx_ring->rx_skb_top);
+				dev_kfree_skb(rx_ring->rx_skb_top);
 				rx_ring->rx_skb_top = NULL;
 				goto next_desc;
 			}
@@ -4239,8 +4232,6 @@ process_skb:
 				 */
 				p = buffer_info->rxbuf.page;
 				if (length <= copybreak) {
-					u8 *vaddr;
-
 					if (likely(!(netdev->features & NETIF_F_RXFCS)))
 						length -= 4;
 					skb = e1000_alloc_rx_skb(adapter,
@@ -4248,10 +4239,9 @@ process_skb:
 					if (!skb)
 						break;
 
-					vaddr = kmap_atomic(p);
-					memcpy(skb_tail_pointer(skb), vaddr,
-					       length);
-					kunmap_atomic(vaddr);
+					memcpy(skb_tail_pointer(skb),
+					       page_address(p), length);
+
 					/* re-use the page, so don't erase
 					 * buffer_info->rxbuf.page
 					 */
@@ -4345,7 +4335,7 @@ static struct sk_buff *e1000_copybreak(struct e1000_adapter *adapter,
 	dma_sync_single_for_cpu(&adapter->pdev->dev, buffer_info->dma,
 				length, DMA_FROM_DEVICE);
 
-	memcpy(skb_put(skb, length), data, length);
+	skb_put_data(skb, data, length);
 
 	return skb;
 }
@@ -4394,7 +4384,7 @@ static bool e1000_clean_rx_irq(struct e1000_adapter *adapter,
 		if (!skb) {
 			unsigned int frag_len = e1000_frag_len(adapter);
 
-			skb = build_skb(data - E1000_HEADROOM, frag_len);
+			skb = napi_build_skb(data - E1000_HEADROOM, frag_len);
 			if (!skb) {
 				adapter->alloc_rx_buff_failed++;
 				break;
@@ -4559,7 +4549,7 @@ e1000_alloc_jumbo_rx_buffers(struct e1000_adapter *adapter,
 		 * applicable for weak-ordered memory model archs,
 		 * such as IA-64).
 		 */
-		wmb();
+		dma_wmb();
 		writel(i, adapter->hw.hw_addr + rx_ring->rdt);
 	}
 }
@@ -4567,6 +4557,8 @@ e1000_alloc_jumbo_rx_buffers(struct e1000_adapter *adapter,
 /**
  * e1000_alloc_rx_buffers - Replace used receive buffers; legacy & extended
  * @adapter: address of board private structure
+ * @rx_ring: pointer to ring struct
+ * @cleaned_count: number of new Rx buffers to try to allocate
  **/
 static void e1000_alloc_rx_buffers(struct e1000_adapter *adapter,
 				   struct e1000_rx_ring *rx_ring,
@@ -4674,14 +4666,14 @@ static void e1000_alloc_rx_buffers(struct e1000_adapter *adapter,
 		 * applicable for weak-ordered memory model archs,
 		 * such as IA-64).
 		 */
-		wmb();
+		dma_wmb();
 		writel(i, hw->hw_addr + rx_ring->rdt);
 	}
 }
 
 /**
  * e1000_smartspeed - Workaround for SmartSpeed on 82541 and 82547 controllers.
- * @adapter:
+ * @adapter: address of board private structure
  **/
 static void e1000_smartspeed(struct e1000_adapter *adapter)
 {
@@ -4737,10 +4729,10 @@ static void e1000_smartspeed(struct e1000_adapter *adapter)
 }
 
 /**
- * e1000_ioctl -
- * @netdev:
- * @ifreq:
- * @cmd:
+ * e1000_ioctl - handle ioctl calls
+ * @netdev: pointer to our netdev
+ * @ifr: pointer to interface request structure
+ * @cmd: ioctl data
  **/
 static int e1000_ioctl(struct net_device *netdev, struct ifreq *ifr, int cmd)
 {
@@ -4756,9 +4748,9 @@ static int e1000_ioctl(struct net_device *netdev, struct ifreq *ifr, int cmd)
 
 /**
  * e1000_mii_ioctl -
- * @netdev:
- * @ifreq:
- * @cmd:
+ * @netdev: pointer to our netdev
+ * @ifr: pointer to interface request structure
+ * @cmd: ioctl data
  **/
 static int e1000_mii_ioctl(struct net_device *netdev, struct ifreq *ifr,
 			   int cmd)
@@ -5073,9 +5065,6 @@ static int __e1000_shutdown(struct pci_dev *pdev, bool *enable_wake)
 	struct e1000_hw *hw = &adapter->hw;
 	u32 ctrl, ctrl_ext, rctl, status;
 	u32 wufc = adapter->wol;
-#ifdef CONFIG_PM
-	int retval = 0;
-#endif
 
 	netif_device_detach(netdev);
 
@@ -5086,15 +5075,11 @@ static int __e1000_shutdown(struct pci_dev *pdev, bool *enable_wake)
 			usleep_range(10000, 20000);
 
 		WARN_ON(test_bit(__E1000_RESETTING, &adapter->flags));
+		rtnl_lock();
 		e1000_down(adapter);
+		rtnl_unlock();
 	}
 
-#ifdef CONFIG_PM
-	retval = pci_save_state(pdev);
-	if (retval)
-		return retval;
-#endif
-
 	status = er32(STATUS);
 	if (status & E1000_STATUS_LU)
 		wufc &= ~E1000_WUFC_LNKC;
@@ -5149,42 +5134,32 @@ static int __e1000_shutdown(struct pci_dev *pdev, bool *enable_wake)
 	if (netif_running(netdev))
 		e1000_free_irq(adapter);
 
-	pci_disable_device(pdev);
+	if (!test_and_set_bit(__E1000_DISABLED, &adapter->flags))
+		pci_disable_device(pdev);
 
 	return 0;
 }
 
-#ifdef CONFIG_PM
-static int e1000_suspend(struct pci_dev *pdev, pm_message_t state)
+static int e1000_suspend(struct device *dev)
 {
 	int retval;
+	struct pci_dev *pdev = to_pci_dev(dev);
 	bool wake;
 
 	retval = __e1000_shutdown(pdev, &wake);
-	if (retval)
-		return retval;
+	device_set_wakeup_enable(dev, wake);
 
-	if (wake) {
-		pci_prepare_to_sleep(pdev);
-	} else {
-		pci_wake_from_d3(pdev, false);
-		pci_set_power_state(pdev, PCI_D3hot);
-	}
-
-	return 0;
+	return retval;
 }
 
-static int e1000_resume(struct pci_dev *pdev)
+static int e1000_resume(struct device *dev)
 {
+	struct pci_dev *pdev = to_pci_dev(dev);
 	struct net_device *netdev = pci_get_drvdata(pdev);
 	struct e1000_adapter *adapter = netdev_priv(netdev);
 	struct e1000_hw *hw = &adapter->hw;
 	u32 err;
 
-	pci_set_power_state(pdev, PCI_D0);
-	pci_restore_state(pdev);
-	pci_save_state(pdev);
-
 	if (adapter->need_ioport)
 		err = pci_enable_device(pdev);
 	else
@@ -5193,6 +5168,10 @@ static int e1000_resume(struct pci_dev *pdev)
 		pr_err("Cannot enable PCI device from suspend\n");
 		return err;
 	}
+
+	/* flush memory to make sure state is correct */
+	smp_mb__before_atomic();
+	clear_bit(__E1000_DISABLED, &adapter->flags);
 	pci_set_master(pdev);
 
 	pci_enable_wake(pdev, PCI_D3hot, 0);
@@ -5217,7 +5196,6 @@ static int e1000_resume(struct pci_dev *pdev)
 
 	return 0;
 }
-#endif
 
 static void e1000_shutdown(struct pci_dev *pdev)
 {
@@ -5260,16 +5238,22 @@ static pci_ers_result_t e1000_io_error_detected(struct pci_dev *pdev,
 	struct net_device *netdev = pci_get_drvdata(pdev);
 	struct e1000_adapter *adapter = netdev_priv(netdev);
 
+	rtnl_lock();
 	netif_device_detach(netdev);
 
-	if (state == pci_channel_io_perm_failure)
+	if (state == pci_channel_io_perm_failure) {
+		rtnl_unlock();
 		return PCI_ERS_RESULT_DISCONNECT;
+	}
 
 	if (netif_running(netdev))
 		e1000_down(adapter);
-	pci_disable_device(pdev);
 
-	/* Request a slot slot reset. */
+	if (!test_and_set_bit(__E1000_DISABLED, &adapter->flags))
+		pci_disable_device(pdev);
+	rtnl_unlock();
+
+	/* Request a slot reset. */
 	return PCI_ERS_RESULT_NEED_RESET;
 }
 
@@ -5295,6 +5279,10 @@ static pci_ers_result_t e1000_io_slot_reset(struct pci_dev *pdev)
 		pr_err("Cannot re-enable PCI device after reset.\n");
 		return PCI_ERS_RESULT_DISCONNECT;
 	}
+
+	/* flush memory to make sure state is correct */
+	smp_mb__before_atomic();
+	clear_bit(__E1000_DISABLED, &adapter->flags);
 	pci_set_master(pdev);
 
 	pci_enable_wake(pdev, PCI_D3hot, 0);
diff --git a/drivers/net/ethernet/intel/e1000/e1000_osdep.h b/drivers/net/ethernet/intel/e1000/e1000_osdep.h
index 33e7c45a4fe4..e966bb290797 100644
--- a/drivers/net/ethernet/intel/e1000/e1000_osdep.h
+++ b/drivers/net/ethernet/intel/e1000/e1000_osdep.h
@@ -1,31 +1,5 @@
-/*******************************************************************************
-
-  Intel PRO/1000 Linux driver
-  Copyright(c) 1999 - 2006 Intel Corporation.
-
-  This program is free software; you can redistribute it and/or modify it
-  under the terms and conditions of the GNU General Public License,
-  version 2, as published by the Free Software Foundation.
-
-  This program is distributed in the hope it will be useful, but WITHOUT
-  ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
-  FITNESS FOR A PARTICULAR PURPOSE.  See the GNU General Public License for
-  more details.
-
-  You should have received a copy of the GNU General Public License along with
-  this program; if not, write to the Free Software Foundation, Inc.,
-  51 Franklin St - Fifth Floor, Boston, MA 02110-1301 USA.
-
-  The full GNU General Public License is included in this distribution in
-  the file called "COPYING".
-
-  Contact Information:
-  Linux NICS <linux.nics@intel.com>
-  e1000-devel Mailing List <e1000-devel@lists.sourceforge.net>
-  Intel Corporation, 5200 N.E. Elam Young Parkway, Hillsboro, OR 97124-6497
-
-*******************************************************************************/
-
+/* SPDX-License-Identifier: GPL-2.0 */
+/* Copyright(c) 1999 - 2006 Intel Corporation. */
 
 /* glue for the OS independent part of e1000
  * includes register access macros
diff --git a/drivers/net/ethernet/intel/e1000/e1000_param.c b/drivers/net/ethernet/intel/e1000/e1000_param.c
index c9cde352b1c8..f4154ca7fcb4 100644
--- a/drivers/net/ethernet/intel/e1000/e1000_param.c
+++ b/drivers/net/ethernet/intel/e1000/e1000_param.c
@@ -1,30 +1,5 @@
-/*******************************************************************************
-
-  Intel PRO/1000 Linux driver
-  Copyright(c) 1999 - 2006 Intel Corporation.
-
-  This program is free software; you can redistribute it and/or modify it
-  under the terms and conditions of the GNU General Public License,
-  version 2, as published by the Free Software Foundation.
-
-  This program is distributed in the hope it will be useful, but WITHOUT
-  ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
-  FITNESS FOR A PARTICULAR PURPOSE.  See the GNU General Public License for
-  more details.
-
-  You should have received a copy of the GNU General Public License along with
-  this program; if not, write to the Free Software Foundation, Inc.,
-  51 Franklin St - Fifth Floor, Boston, MA 02110-1301 USA.
-
-  The full GNU General Public License is included in this distribution in
-  the file called "COPYING".
-
-  Contact Information:
-  Linux NICS <linux.nics@intel.com>
-  e1000-devel Mailing List <e1000-devel@lists.sourceforge.net>
-  Intel Corporation, 5200 N.E. Elam Young Parkway, Hillsboro, OR 97124-6497
-
-*******************************************************************************/
+// SPDX-License-Identifier: GPL-2.0
+/* Copyright(c) 1999 - 2006 Intel Corporation. */
 
 #include "e1000.h"
 
@@ -107,7 +82,6 @@ E1000_PARAM(Duplex, "Duplex setting");
  */
 E1000_PARAM(AutoNeg, "Advertised auto-negotiation setting");
 #define AUTONEG_ADV_DEFAULT  0x2F
-#define AUTONEG_ADV_MASK     0x2F
 
 /* User Specified Flow Control Override
  *
@@ -120,7 +94,6 @@ E1000_PARAM(AutoNeg, "Advertised auto-negotiation setting");
  * Default Value: Read flow control settings from the EEPROM
  */
 E1000_PARAM(FlowControl, "Flow Control setting");
-#define FLOW_CONTROL_DEFAULT FLOW_CONTROL_FULL
 
 /* XsumRX - Receive Checksum Offload Enable/Disable
  *
@@ -733,7 +706,7 @@ static void e1000_check_copper_options(struct e1000_adapter *adapter)
 		goto full_duplex_only;
 	case SPEED_1000 + HALF_DUPLEX:
 		e_dev_info("Half Duplex is not supported at 1000 Mbps\n");
-		/* fall through */
+		fallthrough;
 	case SPEED_1000 + FULL_DUPLEX:
 full_duplex_only:
 		e_dev_info("Using Autonegotiation at 1000 Mbps Full Duplex "
diff --git a/drivers/net/ethernet/intel/e1000e/80003es2lan.c b/drivers/net/ethernet/intel/e1000e/80003es2lan.c
index cd391376036c..4eb1ceaf865a 100644
--- a/drivers/net/ethernet/intel/e1000e/80003es2lan.c
+++ b/drivers/net/ethernet/intel/e1000e/80003es2lan.c
@@ -1,23 +1,5 @@
-/* Intel PRO/1000 Linux driver
- * Copyright(c) 1999 - 2015 Intel Corporation.
- *
- * This program is free software; you can redistribute it and/or modify it
- * under the terms and conditions of the GNU General Public License,
- * version 2, as published by the Free Software Foundation.
- *
- * This program is distributed in the hope it will be useful, but WITHOUT
- * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
- * FITNESS FOR A PARTICULAR PURPOSE.  See the GNU General Public License for
- * more details.
- *
- * The full GNU General Public License is included in this distribution in
- * the file called "COPYING".
- *
- * Contact Information:
- * Linux NICS <linux.nics@intel.com>
- * e1000-devel Mailing List <e1000-devel@lists.sourceforge.net>
- * Intel Corporation, 5200 N.E. Elam Young Parkway, Hillsboro, OR 97124-6497
- */
+// SPDX-License-Identifier: GPL-2.0
+/* Copyright(c) 1999 - 2018 Intel Corporation. */
 
 /* 80003ES2LAN Gigabit Ethernet Controller (Copper)
  * 80003ES2LAN Gigabit Ethernet Controller (Serdes)
@@ -110,8 +92,7 @@ static s32 e1000_init_nvm_params_80003es2lan(struct e1000_hw *hw)
 
 	nvm->type = e1000_nvm_eeprom_spi;
 
-	size = (u16)((eecd & E1000_EECD_SIZE_EX_MASK) >>
-		     E1000_EECD_SIZE_EX_SHIFT);
+	size = (u16)FIELD_GET(E1000_EECD_SIZE_EX_MASK, eecd);
 
 	/* Added to a constant, "size" becomes the left-shift value
 	 * for setting word_size.
@@ -698,7 +679,7 @@ static s32 e1000_reset_hw_80003es2lan(struct e1000_hw *hw)
 	ew32(TCTL, E1000_TCTL_PSP);
 	e1e_flush();
 
-	usleep_range(10000, 20000);
+	usleep_range(10000, 11000);
 
 	ctrl = er32(CTRL);
 
@@ -714,11 +695,16 @@ static s32 e1000_reset_hw_80003es2lan(struct e1000_hw *hw)
 	ret_val =
 	    e1000_read_kmrn_reg_80003es2lan(hw, E1000_KMRNCTRLSTA_INBAND_PARAM,
 					    &kum_reg_data);
-	if (ret_val)
-		return ret_val;
-	kum_reg_data |= E1000_KMRNCTRLSTA_IBIST_DISABLE;
-	e1000_write_kmrn_reg_80003es2lan(hw, E1000_KMRNCTRLSTA_INBAND_PARAM,
-					 kum_reg_data);
+	if (!ret_val) {
+		kum_reg_data |= E1000_KMRNCTRLSTA_IBIST_DISABLE;
+		ret_val = e1000_write_kmrn_reg_80003es2lan(hw,
+						 E1000_KMRNCTRLSTA_INBAND_PARAM,
+						 kum_reg_data);
+		if (ret_val)
+			e_dbg("Error disabling far-end loopback\n");
+	} else {
+		e_dbg("Error disabling far-end loopback\n");
+	}
 
 	ret_val = e1000e_get_auto_rd_done(hw);
 	if (ret_val)
@@ -772,11 +758,19 @@ static s32 e1000_init_hw_80003es2lan(struct e1000_hw *hw)
 		return ret_val;
 
 	/* Disable IBIST slave mode (far-end loopback) */
-	e1000_read_kmrn_reg_80003es2lan(hw, E1000_KMRNCTRLSTA_INBAND_PARAM,
-					&kum_reg_data);
-	kum_reg_data |= E1000_KMRNCTRLSTA_IBIST_DISABLE;
-	e1000_write_kmrn_reg_80003es2lan(hw, E1000_KMRNCTRLSTA_INBAND_PARAM,
-					 kum_reg_data);
+	ret_val =
+	    e1000_read_kmrn_reg_80003es2lan(hw, E1000_KMRNCTRLSTA_INBAND_PARAM,
+					    &kum_reg_data);
+	if (!ret_val) {
+		kum_reg_data |= E1000_KMRNCTRLSTA_IBIST_DISABLE;
+		ret_val = e1000_write_kmrn_reg_80003es2lan(hw,
+						 E1000_KMRNCTRLSTA_INBAND_PARAM,
+						 kum_reg_data);
+		if (ret_val)
+			e_dbg("Error disabling far-end loopback\n");
+	} else {
+		e_dbg("Error disabling far-end loopback\n");
+	}
 
 	/* Set the transmit descriptor write-back policy */
 	reg_data = er32(TXDCTL(0));
@@ -1040,17 +1034,18 @@ static s32 e1000_setup_copper_link_80003es2lan(struct e1000_hw *hw)
 	 * iteration and increase the max iterations when
 	 * polling the phy; this fixes erroneous timeouts at 10Mbps.
 	 */
-	ret_val = e1000_write_kmrn_reg_80003es2lan(hw, GG82563_REG(0x34, 4),
-						   0xFFFF);
+	/* these next three accesses were always meant to use page 0x34 using
+	 * GG82563_REG(0x34, N) but never did, so we've just corrected the call
+	 * to not drop bits
+	 */
+	ret_val = e1000_write_kmrn_reg_80003es2lan(hw, 4, 0xFFFF);
 	if (ret_val)
 		return ret_val;
-	ret_val = e1000_read_kmrn_reg_80003es2lan(hw, GG82563_REG(0x34, 9),
-						  &reg_data);
+	ret_val = e1000_read_kmrn_reg_80003es2lan(hw, 9, &reg_data);
 	if (ret_val)
 		return ret_val;
 	reg_data |= 0x3F;
-	ret_val = e1000_write_kmrn_reg_80003es2lan(hw, GG82563_REG(0x34, 9),
-						   reg_data);
+	ret_val = e1000_write_kmrn_reg_80003es2lan(hw, 9, reg_data);
 	if (ret_val)
 		return ret_val;
 	ret_val =
@@ -1077,7 +1072,6 @@ static s32 e1000_setup_copper_link_80003es2lan(struct e1000_hw *hw)
 /**
  *  e1000_cfg_on_link_up_80003es2lan - es2 link configuration after link-up
  *  @hw: pointer to the HW structure
- *  @duplex: current duplex setting
  *
  *  Configure the KMRN interface by applying last minute quirks for
  *  10/100 operation.
@@ -1215,8 +1209,8 @@ static s32 e1000_read_kmrn_reg_80003es2lan(struct e1000_hw *hw, u32 offset,
 	if (ret_val)
 		return ret_val;
 
-	kmrnctrlsta = ((offset << E1000_KMRNCTRLSTA_OFFSET_SHIFT) &
-		       E1000_KMRNCTRLSTA_OFFSET) | E1000_KMRNCTRLSTA_REN;
+	kmrnctrlsta = FIELD_PREP(E1000_KMRNCTRLSTA_OFFSET, offset) |
+		      E1000_KMRNCTRLSTA_REN;
 	ew32(KMRNCTRLSTA, kmrnctrlsta);
 	e1e_flush();
 
@@ -1250,8 +1244,7 @@ static s32 e1000_write_kmrn_reg_80003es2lan(struct e1000_hw *hw, u32 offset,
 	if (ret_val)
 		return ret_val;
 
-	kmrnctrlsta = ((offset << E1000_KMRNCTRLSTA_OFFSET_SHIFT) &
-		       E1000_KMRNCTRLSTA_OFFSET) | data;
+	kmrnctrlsta = FIELD_PREP(E1000_KMRNCTRLSTA_OFFSET, offset) | data;
 	ew32(KMRNCTRLSTA, kmrnctrlsta);
 	e1e_flush();
 
diff --git a/drivers/net/ethernet/intel/e1000e/80003es2lan.h b/drivers/net/ethernet/intel/e1000e/80003es2lan.h
index a2162e11673e..aa9d639c6cbb 100644
--- a/drivers/net/ethernet/intel/e1000e/80003es2lan.h
+++ b/drivers/net/ethernet/intel/e1000e/80003es2lan.h
@@ -1,23 +1,5 @@
-/* Intel PRO/1000 Linux driver
- * Copyright(c) 1999 - 2015 Intel Corporation.
- *
- * This program is free software; you can redistribute it and/or modify it
- * under the terms and conditions of the GNU General Public License,
- * version 2, as published by the Free Software Foundation.
- *
- * This program is distributed in the hope it will be useful, but WITHOUT
- * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
- * FITNESS FOR A PARTICULAR PURPOSE.  See the GNU General Public License for
- * more details.
- *
- * The full GNU General Public License is included in this distribution in
- * the file called "COPYING".
- *
- * Contact Information:
- * Linux NICS <linux.nics@intel.com>
- * e1000-devel Mailing List <e1000-devel@lists.sourceforge.net>
- * Intel Corporation, 5200 N.E. Elam Young Parkway, Hillsboro, OR 97124-6497
- */
+/* SPDX-License-Identifier: GPL-2.0 */
+/* Copyright(c) 1999 - 2018 Intel Corporation. */
 
 #ifndef _E1000E_80003ES2LAN_H_
 #define _E1000E_80003ES2LAN_H_
diff --git a/drivers/net/ethernet/intel/e1000e/82571.c b/drivers/net/ethernet/intel/e1000e/82571.c
index 6b03c8553e59..969f855a79ee 100644
--- a/drivers/net/ethernet/intel/e1000e/82571.c
+++ b/drivers/net/ethernet/intel/e1000e/82571.c
@@ -1,23 +1,5 @@
-/* Intel PRO/1000 Linux driver
- * Copyright(c) 1999 - 2015 Intel Corporation.
- *
- * This program is free software; you can redistribute it and/or modify it
- * under the terms and conditions of the GNU General Public License,
- * version 2, as published by the Free Software Foundation.
- *
- * This program is distributed in the hope it will be useful, but WITHOUT
- * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
- * FITNESS FOR A PARTICULAR PURPOSE.  See the GNU General Public License for
- * more details.
- *
- * The full GNU General Public License is included in this distribution in
- * the file called "COPYING".
- *
- * Contact Information:
- * Linux NICS <linux.nics@intel.com>
- * e1000-devel Mailing List <e1000-devel@lists.sourceforge.net>
- * Intel Corporation, 5200 N.E. Elam Young Parkway, Hillsboro, OR 97124-6497
- */
+// SPDX-License-Identifier: GPL-2.0
+/* Copyright(c) 1999 - 2018 Intel Corporation. */
 
 /* 82571EB Gigabit Ethernet Controller
  * 82571EB Gigabit Ethernet Controller (Copper)
@@ -172,11 +154,10 @@ static s32 e1000_init_nvm_params_82571(struct e1000_hw *hw)
 			ew32(EECD, eecd);
 			break;
 		}
-		/* Fall Through */
+		fallthrough;
 	default:
 		nvm->type = e1000_nvm_eeprom_spi;
-		size = (u16)((eecd & E1000_EECD_SIZE_EX_MASK) >>
-			     E1000_EECD_SIZE_EX_SHIFT);
+		size = (u16)FIELD_GET(E1000_EECD_SIZE_EX_MASK, eecd);
 		/* Added to a constant, "size" becomes the left-shift value
 		 * for setting word_size.
 		 */
@@ -917,6 +898,8 @@ static s32 e1000_set_d0_lplu_state_82571(struct e1000_hw *hw, bool active)
 	} else {
 		data &= ~IGP02E1000_PM_D0_LPLU;
 		ret_val = e1e_wphy(hw, IGP02E1000_PHY_POWER_MGMT, data);
+		if (ret_val)
+			return ret_val;
 		/* LPLU and SmartSpeed are mutually exclusive.  LPLU is used
 		 * during Dx states where the power conservation is most
 		 * important.  During driver activity we should enable
@@ -977,7 +960,7 @@ static s32 e1000_reset_hw_82571(struct e1000_hw *hw)
 	ew32(TCTL, tctl);
 	e1e_flush();
 
-	usleep_range(10000, 20000);
+	usleep_range(10000, 11000);
 
 	/* Must acquire the MDIO ownership before MAC reset.
 	 * Ownership defaults to firmware after a reset.
@@ -1125,7 +1108,7 @@ static s32 e1000_init_hw_82571(struct e1000_hw *hw)
 	switch (mac->type) {
 	case e1000_82573:
 		e1000e_enable_tx_pkt_filtering(hw);
-		/* fall through */
+		fallthrough;
 	case e1000_82574:
 	case e1000_82583:
 		reg_data = er32(GCR);
diff --git a/drivers/net/ethernet/intel/e1000e/82571.h b/drivers/net/ethernet/intel/e1000e/82571.h
index abc6a9abff98..834c238d02db 100644
--- a/drivers/net/ethernet/intel/e1000e/82571.h
+++ b/drivers/net/ethernet/intel/e1000e/82571.h
@@ -1,23 +1,5 @@
-/* Intel PRO/1000 Linux driver
- * Copyright(c) 1999 - 2015 Intel Corporation.
- *
- * This program is free software; you can redistribute it and/or modify it
- * under the terms and conditions of the GNU General Public License,
- * version 2, as published by the Free Software Foundation.
- *
- * This program is distributed in the hope it will be useful, but WITHOUT
- * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
- * FITNESS FOR A PARTICULAR PURPOSE.  See the GNU General Public License for
- * more details.
- *
- * The full GNU General Public License is included in this distribution in
- * the file called "COPYING".
- *
- * Contact Information:
- * Linux NICS <linux.nics@intel.com>
- * e1000-devel Mailing List <e1000-devel@lists.sourceforge.net>
- * Intel Corporation, 5200 N.E. Elam Young Parkway, Hillsboro, OR 97124-6497
- */
+/* SPDX-License-Identifier: GPL-2.0 */
+/* Copyright(c) 1999 - 2018 Intel Corporation. */
 
 #ifndef _E1000E_82571_H_
 #define _E1000E_82571_H_
diff --git a/drivers/net/ethernet/intel/e1000e/Makefile b/drivers/net/ethernet/intel/e1000e/Makefile
index 106de493373c..18f22b6374d5 100644
--- a/drivers/net/ethernet/intel/e1000e/Makefile
+++ b/drivers/net/ethernet/intel/e1000e/Makefile
@@ -1,37 +1,15 @@
-################################################################################
-#
-# Intel PRO/1000 Linux driver
-# Copyright(c) 1999 - 2014 Intel Corporation.
-#
-# This program is free software; you can redistribute it and/or modify it
-# under the terms and conditions of the GNU General Public License,
-# version 2, as published by the Free Software Foundation.
-#
-# This program is distributed in the hope it will be useful, but WITHOUT
-# ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
-# FITNESS FOR A PARTICULAR PURPOSE.  See the GNU General Public License for
-# more details.
-#
-# You should have received a copy of the GNU General Public License
-# along with this program; if not, see <http://www.gnu.org/licenses/>.
-#
-# The full GNU General Public License is included in this distribution in
-# the file called "COPYING".
-#
-# Contact Information:
-# Linux NICS <linux.nics@intel.com>
-# e1000-devel Mailing List <e1000-devel@lists.sourceforge.net>
-# Intel Corporation, 5200 N.E. Elam Young Parkway, Hillsboro, OR 97124-6497
-#
-################################################################################
+# SPDX-License-Identifier: GPL-2.0
+# Copyright(c) 1999 - 2018 Intel Corporation.
 
 #
 # Makefile for the Intel(R) PRO/1000 ethernet driver
 #
 
-obj-$(CONFIG_E1000E) += e1000e.o
+ccflags-y += -I$(src)
+subdir-ccflags-y += -I$(src)
 
-e1000e-objs := 82571.o ich8lan.o 80003es2lan.o \
-	       mac.o manage.o nvm.o phy.o \
-	       param.o ethtool.o netdev.o ptp.o
+obj-$(CONFIG_E1000E) += e1000e.o
 
+e1000e-y := 82571.o ich8lan.o 80003es2lan.o \
+	    mac.o manage.o nvm.o phy.o \
+	    param.o ethtool.o netdev.o ptp.o
diff --git a/drivers/net/ethernet/intel/e1000e/defines.h b/drivers/net/ethernet/intel/e1000e/defines.h
index 0641c0098738..ba331899d186 100644
--- a/drivers/net/ethernet/intel/e1000e/defines.h
+++ b/drivers/net/ethernet/intel/e1000e/defines.h
@@ -1,23 +1,5 @@
-/* Intel PRO/1000 Linux driver
- * Copyright(c) 1999 - 2015 Intel Corporation.
- *
- * This program is free software; you can redistribute it and/or modify it
- * under the terms and conditions of the GNU General Public License,
- * version 2, as published by the Free Software Foundation.
- *
- * This program is distributed in the hope it will be useful, but WITHOUT
- * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
- * FITNESS FOR A PARTICULAR PURPOSE.  See the GNU General Public License for
- * more details.
- *
- * The full GNU General Public License is included in this distribution in
- * the file called "COPYING".
- *
- * Contact Information:
- * Linux NICS <linux.nics@intel.com>
- * e1000-devel Mailing List <e1000-devel@lists.sourceforge.net>
- * Intel Corporation, 5200 N.E. Elam Young Parkway, Hillsboro, OR 97124-6497
- */
+/* SPDX-License-Identifier: GPL-2.0 */
+/* Copyright(c) 1999 - 2018 Intel Corporation. */
 
 #ifndef _E1000_DEFINES_H_
 #define _E1000_DEFINES_H_
@@ -240,6 +222,9 @@
 #define E1000_STATUS_PHYRA      0x00000400      /* PHY Reset Asserted */
 #define E1000_STATUS_GIO_MASTER_ENABLE	0x00080000	/* Master Req status */
 
+/* PCIm function state */
+#define E1000_STATUS_PCIM_STATE	0x40000000
+
 #define HALF_DUPLEX 1
 #define FULL_DUPLEX 2
 
@@ -398,7 +383,12 @@
 #define E1000_ICR_LSC           0x00000004 /* Link Status Change */
 #define E1000_ICR_RXSEQ         0x00000008 /* Rx sequence error */
 #define E1000_ICR_RXDMT0        0x00000010 /* Rx desc min. threshold (0) */
+#define E1000_ICR_RXO           0x00000040 /* Receiver Overrun */
 #define E1000_ICR_RXT0          0x00000080 /* Rx timer intr (ring 0) */
+#define E1000_ICR_MDAC          0x00000200 /* MDIO Access Complete */
+#define E1000_ICR_SRPD          0x00010000 /* Small Receive Packet Detected */
+#define E1000_ICR_ACK           0x00020000 /* Receive ACK Frame Detected */
+#define E1000_ICR_MNG           0x00040000 /* Manageability Event Detected */
 #define E1000_ICR_ECCER         0x00400000 /* Uncorrectable ECC Error */
 /* If this bit asserted, the driver should claim the interrupt */
 #define E1000_ICR_INT_ASSERTED	0x80000000
@@ -406,7 +396,7 @@
 #define E1000_ICR_RXQ1          0x00200000 /* Rx Queue 1 Interrupt */
 #define E1000_ICR_TXQ0          0x00400000 /* Tx Queue 0 Interrupt */
 #define E1000_ICR_TXQ1          0x00800000 /* Tx Queue 1 Interrupt */
-#define E1000_ICR_OTHER         0x01000000 /* Other Interrupts */
+#define E1000_ICR_OTHER         0x01000000 /* Other Interrupt */
 
 /* PBA ECC Register */
 #define E1000_PBA_ECC_COUNTER_MASK  0xFFF00000 /* ECC counter mask */
@@ -430,12 +420,27 @@
 	E1000_IMS_RXSEQ  |    \
 	E1000_IMS_LSC)
 
+/* These are all of the events related to the OTHER interrupt.
+ */
+#define IMS_OTHER_MASK ( \
+	E1000_IMS_LSC  | \
+	E1000_IMS_RXO  | \
+	E1000_IMS_MDAC | \
+	E1000_IMS_SRPD | \
+	E1000_IMS_ACK  | \
+	E1000_IMS_MNG)
+
 /* Interrupt Mask Set */
 #define E1000_IMS_TXDW      E1000_ICR_TXDW      /* Transmit desc written back */
 #define E1000_IMS_LSC       E1000_ICR_LSC       /* Link Status Change */
 #define E1000_IMS_RXSEQ     E1000_ICR_RXSEQ     /* Rx sequence error */
 #define E1000_IMS_RXDMT0    E1000_ICR_RXDMT0    /* Rx desc min. threshold */
+#define E1000_IMS_RXO       E1000_ICR_RXO       /* Receiver Overrun */
 #define E1000_IMS_RXT0      E1000_ICR_RXT0      /* Rx timer intr */
+#define E1000_IMS_MDAC      E1000_ICR_MDAC      /* MDIO Access Complete */
+#define E1000_IMS_SRPD      E1000_ICR_SRPD      /* Small Receive Packet */
+#define E1000_IMS_ACK       E1000_ICR_ACK       /* Receive ACK Frame Detected */
+#define E1000_IMS_MNG       E1000_ICR_MNG       /* Manageability Event */
 #define E1000_IMS_ECCER     E1000_ICR_ECCER     /* Uncorrectable ECC Error */
 #define E1000_IMS_RXQ0      E1000_ICR_RXQ0      /* Rx Queue 0 Interrupt */
 #define E1000_IMS_RXQ1      E1000_ICR_RXQ1      /* Rx Queue 1 Interrupt */
@@ -633,6 +638,9 @@
 /* For checksumming, the sum of all words in the NVM should equal 0xBABA. */
 #define NVM_SUM                    0xBABA
 
+/* Uninitialized ("empty") checksum word value */
+#define NVM_CHECKSUM_UNINITIALIZED 0xFFFF
+
 /* PBA (printed board assembly) number words */
 #define NVM_PBA_OFFSET_0           8
 #define NVM_PBA_OFFSET_1           9
@@ -673,11 +681,6 @@
 
 /* PCI/PCI-X/PCI-EX Config space */
 #define PCI_HEADER_TYPE_REGISTER     0x0E
-#define PCIE_LINK_STATUS             0x12
-
-#define PCI_HEADER_TYPE_MULTIFUNC    0x80
-#define PCIE_LINK_WIDTH_MASK         0x3F0
-#define PCIE_LINK_WIDTH_SHIFT        4
 
 #define PHY_REVISION_MASK      0xFFFFFFF0
 #define MAX_PHY_REG_ADDRESS    0x1F  /* 5 bit address bus (0-0x1F) */
@@ -803,4 +806,7 @@
 /* SerDes Control */
 #define E1000_GEN_POLL_TIMEOUT          640
 
+#define E1000_FEXTNVM12_PHYPD_CTRL_MASK	0x00C00000
+#define E1000_FEXTNVM12_PHYPD_CTRL_P1	0x00800000
+
 #endif /* _E1000_DEFINES_H_ */
diff --git a/drivers/net/ethernet/intel/e1000e/e1000.h b/drivers/net/ethernet/intel/e1000e/e1000.h
index c7c994eb410e..018e61aea787 100644
--- a/drivers/net/ethernet/intel/e1000e/e1000.h
+++ b/drivers/net/ethernet/intel/e1000e/e1000.h
@@ -1,23 +1,5 @@
-/* Intel PRO/1000 Linux driver
- * Copyright(c) 1999 - 2015 Intel Corporation.
- *
- * This program is free software; you can redistribute it and/or modify it
- * under the terms and conditions of the GNU General Public License,
- * version 2, as published by the Free Software Foundation.
- *
- * This program is distributed in the hope it will be useful, but WITHOUT
- * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
- * FITNESS FOR A PARTICULAR PURPOSE.  See the GNU General Public License for
- * more details.
- *
- * The full GNU General Public License is included in this distribution in
- * the file called "COPYING".
- *
- * Contact Information:
- * Linux NICS <linux.nics@intel.com>
- * e1000-devel Mailing List <e1000-devel@lists.sourceforge.net>
- * Intel Corporation, 5200 N.E. Elam Young Parkway, Hillsboro, OR 97124-6497
- */
+/* SPDX-License-Identifier: GPL-2.0 */
+/* Copyright(c) 1999 - 2018 Intel Corporation. */
 
 /* Linux PRO/1000 Ethernet Driver main header file */
 
@@ -31,7 +13,6 @@
 #include <linux/io.h>
 #include <linux/netdevice.h>
 #include <linux/pci.h>
-#include <linux/pci-aspm.h>
 #include <linux/crc32.h>
 #include <linux/if_vlan.h>
 #include <linux/timecounter.h>
@@ -40,6 +21,7 @@
 #include <linux/ptp_classify.h>
 #include <linux/mii.h>
 #include <linux/mdio.h>
+#include <linux/mutex.h>
 #include <linux/pm_qos.h>
 #include "hw.h"
 
@@ -82,7 +64,7 @@ struct e1000_info;
 #define AUTO_ALL_MODES			0
 #define E1000_EEPROM_APME		0x0400
 
-#define E1000_MNG_VLAN_NONE		(-1)
+#define E1000_MNG_VLAN_NONE		0xFFFF
 
 #define DEFAULT_JUMBO			9234
 
@@ -94,10 +76,6 @@ struct e1000_info;
  */
 #define E1000_CHECK_RESET_COUNT		25
 
-#define DEFAULT_RDTR			0
-#define DEFAULT_RADV			8
-#define BURST_RDTR			0x20
-#define BURST_RADV			0x20
 #define PCICFG_DESC_RING_STATUS		0xe4
 #define FLUSH_DESC_REQUIRED		0x100
 
@@ -136,7 +114,10 @@ enum e1000_boards {
 	board_pch2lan,
 	board_pch_lpt,
 	board_pch_spt,
-	board_pch_cnp
+	board_pch_cnp,
+	board_pch_tgp,
+	board_pch_adp,
+	board_pch_mtp
 };
 
 struct e1000_ps_page {
@@ -268,6 +249,7 @@ struct e1000_adapter {
 	u32 tx_fifo_size;
 	u32 tx_dma_failed;
 	u32 tx_hwtstamp_timeouts;
+	u32 tx_hwtstamp_skipped;
 
 	/* Rx */
 	bool (*clean_rx)(struct e1000_ring *ring, int *work_done,
@@ -337,7 +319,7 @@ struct e1000_adapter {
 	u16 tx_ring_count;
 	u16 rx_ring_count;
 
-	struct hwtstamp_config hwtstamp_config;
+	struct kernel_hwtstamp_config hwtstamp_config;
 	struct delayed_work systim_overflow_work;
 	struct sk_buff *tx_hwtstamp_skb;
 	unsigned long tx_hwtstamp_start;
@@ -348,7 +330,7 @@ struct e1000_adapter {
 	struct ptp_clock *ptp_clock;
 	struct ptp_clock_info ptp_clock_info;
 	struct pm_qos_request pm_qos_req;
-	s32 ptp_delta;
+	long ptp_delta;
 
 	u16 eee_advert;
 };
@@ -378,23 +360,43 @@ s32 e1000e_get_base_timinca(struct e1000_adapter *adapter, u32 *timinca);
  * As a result, a shift of INCVALUE_SHIFT_n is used to fit a value of
  * INCVALUE_n into the TIMINCA register allowing 32+8+(24-INCVALUE_SHIFT_n)
  * bits to count nanoseconds leaving the rest for fractional nonseconds.
+ *
+ * Any given INCVALUE also has an associated maximum adjustment value. This
+ * maximum adjustment value is the largest increase (or decrease) which can be
+ * safely applied without overflowing the INCVALUE. Since INCVALUE has
+ * a maximum range of 24 bits, its largest value is 0xFFFFFF.
+ *
+ * To understand where the maximum value comes from, consider the following
+ * equation:
+ *
+ *   new_incval = base_incval + (base_incval * adjustment) / 1billion
+ *
+ * To avoid overflow that means:
+ *   max_incval = base_incval + (base_incval * max_adj) / billion
+ *
+ * Re-arranging:
+ *   max_adj = floor(((max_incval - base_incval) * 1billion) / 1billion)
  */
 #define INCVALUE_96MHZ		125
 #define INCVALUE_SHIFT_96MHZ	17
 #define INCPERIOD_SHIFT_96MHZ	2
 #define INCPERIOD_96MHZ		(12 >> INCPERIOD_SHIFT_96MHZ)
+#define MAX_PPB_96MHZ		23999900 /* 23,999,900 ppb */
 
 #define INCVALUE_25MHZ		40
 #define INCVALUE_SHIFT_25MHZ	18
 #define INCPERIOD_25MHZ		1
+#define MAX_PPB_25MHZ		599999900 /* 599,999,900 ppb */
 
 #define INCVALUE_24MHZ		125
 #define INCVALUE_SHIFT_24MHZ	14
 #define INCPERIOD_24MHZ		3
+#define MAX_PPB_24MHZ		999999999 /* 999,999,999 ppb */
 
 #define INCVALUE_38400KHZ	26
 #define INCVALUE_SHIFT_38400KHZ	19
 #define INCPERIOD_38400KHZ	1
+#define MAX_PPB_38400KHZ	230769100 /* 230,769,100 ppb */
 
 /* Another drawback of scaling the incvalue by a large factor is the
  * 64-bit SYSTIM register overflows more quickly.  This is dealt with
@@ -458,6 +460,7 @@ s32 e1000e_get_base_timinca(struct e1000_adapter *adapter, u32 *timinca);
 #define FLAG2_DFLT_CRC_STRIPPING          BIT(12)
 #define FLAG2_CHECK_RX_HWTSTAMP           BIT(13)
 #define FLAG2_CHECK_SYSTIM_OVERFLOW       BIT(14)
+#define FLAG2_ENABLE_S0IX_FLOWS           BIT(15)
 
 #define E1000_RX_DESC_PS(R, i)	    \
 	(&(((union e1000_rx_desc_packet_split *)((R).desc))[i]))
@@ -482,7 +485,6 @@ enum latency_range {
 };
 
 extern char e1000e_driver_name[];
-extern const char e1000e_driver_version[];
 
 void e1000e_check_options(struct e1000_adapter *adapter);
 void e1000e_set_ethtool_ops(struct net_device *netdev);
@@ -521,11 +523,17 @@ extern const struct e1000_info e1000_pch2_info;
 extern const struct e1000_info e1000_pch_lpt_info;
 extern const struct e1000_info e1000_pch_spt_info;
 extern const struct e1000_info e1000_pch_cnp_info;
+extern const struct e1000_info e1000_pch_tgp_info;
+extern const struct e1000_info e1000_pch_adp_info;
+extern const struct e1000_info e1000_pch_mtp_info;
 extern const struct e1000_info e1000_es2_info;
 
 void e1000e_ptp_init(struct e1000_adapter *adapter);
 void e1000e_ptp_remove(struct e1000_adapter *adapter);
 
+u64 e1000e_read_systim(struct e1000_adapter *adapter,
+		       struct ptp_system_timestamp *sts);
+
 static inline s32 e1000_phy_hw_reset(struct e1000_hw *hw)
 {
 	return hw->phy.ops.reset(hw);
@@ -595,7 +603,6 @@ static inline u32 __er32(struct e1000_hw *hw, unsigned long reg)
 
 #define er32(reg)	__er32(hw, E1000_##reg)
 
-s32 __ew32_prepare(struct e1000_hw *hw);
 void __ew32(struct e1000_hw *hw, unsigned long reg, u32 val);
 
 #define ew32(reg, val)	__ew32(hw, E1000_##reg, (val))
diff --git a/drivers/net/ethernet/intel/e1000e/e1000e_trace.h b/drivers/net/ethernet/intel/e1000e/e1000e_trace.h
new file mode 100644
index 000000000000..19d3cf4d924e
--- /dev/null
+++ b/drivers/net/ethernet/intel/e1000e/e1000e_trace.h
@@ -0,0 +1,42 @@
+/* SPDX-License-Identifier: GPL-2.0 */
+/* Copyright(c) 2022, Intel Corporation. */
+/* Modeled on trace-events-sample.h */
+/* The trace subsystem name for e1000e will be "e1000e_trace".
+ *
+ * This file is named e1000e_trace.h.
+ *
+ * Since this include file's name is different from the trace
+ * subsystem name, we'll have to define TRACE_INCLUDE_FILE at the end
+ * of this file.
+ */
+
+#undef TRACE_SYSTEM
+#define TRACE_SYSTEM e1000e_trace
+
+#if !defined(_TRACE_E1000E_TRACE_H) || defined(TRACE_HEADER_MULTI_READ)
+#define _TRACE_E1000E_TRACE_H
+
+#include <linux/tracepoint.h>
+
+TRACE_EVENT(e1000e_trace_mac_register,
+	    TP_PROTO(uint32_t reg),
+	    TP_ARGS(reg),
+	    TP_STRUCT__entry(__field(uint32_t,	reg)),
+	    TP_fast_assign(__entry->reg = reg;),
+	    TP_printk("event: TraceHub e1000e mac register: 0x%08x",
+		      __entry->reg)
+);
+
+#endif
+/* This must be outside ifdef _E1000E_TRACE_H */
+/* This trace include file is not located in the .../include/trace
+ * with the kernel tracepoint definitions, because we're a loadable
+ * module.
+ */
+
+#undef TRACE_INCLUDE_PATH
+#define TRACE_INCLUDE_PATH .
+#undef TRACE_INCLUDE_FILE
+#define TRACE_INCLUDE_FILE e1000e_trace
+
+#include <trace/define_trace.h>
diff --git a/drivers/net/ethernet/intel/e1000e/ethtool.c b/drivers/net/ethernet/intel/e1000e/ethtool.c
index e23dbd9190d6..8e40bb50a01e 100644
--- a/drivers/net/ethernet/intel/e1000e/ethtool.c
+++ b/drivers/net/ethernet/intel/e1000e/ethtool.c
@@ -1,23 +1,5 @@
-/* Intel PRO/1000 Linux driver
- * Copyright(c) 1999 - 2015 Intel Corporation.
- *
- * This program is free software; you can redistribute it and/or modify it
- * under the terms and conditions of the GNU General Public License,
- * version 2, as published by the Free Software Foundation.
- *
- * This program is distributed in the hope it will be useful, but WITHOUT
- * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
- * FITNESS FOR A PARTICULAR PURPOSE.  See the GNU General Public License for
- * more details.
- *
- * The full GNU General Public License is included in this distribution in
- * the file called "COPYING".
- *
- * Contact Information:
- * Linux NICS <linux.nics@intel.com>
- * e1000-devel Mailing List <e1000-devel@lists.sourceforge.net>
- * Intel Corporation, 5200 N.E. Elam Young Parkway, Hillsboro, OR 97124-6497
- */
+// SPDX-License-Identifier: GPL-2.0
+/* Copyright(c) 1999 - 2018 Intel Corporation. */
 
 /* ethtool support for e1000 */
 
@@ -41,6 +23,13 @@ struct e1000_stats {
 	int stat_offset;
 };
 
+static const char e1000e_priv_flags_strings[][ETH_GSTRING_LEN] = {
+#define E1000E_PRIV_FLAGS_S0IX_ENABLED	BIT(0)
+	"s0ix-enabled",
+};
+
+#define E1000E_PRIV_FLAGS_STR_LEN ARRAY_SIZE(e1000e_priv_flags_strings)
+
 #define E1000_STAT(str, m) { \
 		.stat_string = str, \
 		.type = E1000_STATS, \
@@ -105,6 +94,7 @@ static const struct e1000_stats e1000_gstrings_stats[] = {
 	E1000_STAT("uncorr_ecc_errors", uncorr_errors),
 	E1000_STAT("corr_ecc_errors", corr_errors),
 	E1000_STAT("tx_hwtstamp_timeouts", tx_hwtstamp_timeouts),
+	E1000_STAT("tx_hwtstamp_skipped", tx_hwtstamp_skipped),
 };
 
 #define E1000_GLOBAL_STATS_LEN	ARRAY_SIZE(e1000_gstrings_stats)
@@ -120,9 +110,9 @@ static const char e1000_gstrings_test[][ETH_GSTRING_LEN] = {
 static int e1000_get_link_ksettings(struct net_device *netdev,
 				    struct ethtool_link_ksettings *cmd)
 {
+	u32 speed, supported, advertising, lp_advertising, lpa_t;
 	struct e1000_adapter *adapter = netdev_priv(netdev);
 	struct e1000_hw *hw = &adapter->hw;
-	u32 speed, supported, advertising;
 
 	if (hw->phy.media_type == e1000_media_type_copper) {
 		supported = (SUPPORTED_10baseT_Half |
@@ -130,7 +120,9 @@ static int e1000_get_link_ksettings(struct net_device *netdev,
 			     SUPPORTED_100baseT_Half |
 			     SUPPORTED_100baseT_Full |
 			     SUPPORTED_1000baseT_Full |
+			     SUPPORTED_Asym_Pause |
 			     SUPPORTED_Autoneg |
+			     SUPPORTED_Pause |
 			     SUPPORTED_TP);
 		if (hw->phy.type == e1000_phy_ife)
 			supported &= ~SUPPORTED_1000baseT_Full;
@@ -164,7 +156,7 @@ static int e1000_get_link_ksettings(struct net_device *netdev,
 			speed = adapter->link_speed;
 			cmd->base.duplex = adapter->link_duplex - 1;
 		}
-	} else if (!pm_runtime_suspended(netdev->dev.parent)) {
+	} else {
 		u32 status = er32(STATUS);
 
 		if (status & E1000_STATUS_LU) {
@@ -202,10 +194,16 @@ static int e1000_get_link_ksettings(struct net_device *netdev,
 	if (hw->phy.media_type != e1000_media_type_copper)
 		cmd->base.eth_tp_mdix_ctrl = ETH_TP_MDI_INVALID;
 
+	lpa_t = mii_stat1000_to_ethtool_lpa_t(adapter->phy_regs.stat1000);
+	lp_advertising = lpa_t |
+	mii_lpa_to_ethtool_lpa_t(adapter->phy_regs.lpa);
+
 	ethtool_convert_legacy_u32_to_link_mode(cmd->link_modes.supported,
 						supported);
 	ethtool_convert_legacy_u32_to_link_mode(cmd->link_modes.advertising,
 						advertising);
+	ethtool_convert_legacy_u32_to_link_mode(cmd->link_modes.lp_advertising,
+						lp_advertising);
 
 	return 0;
 }
@@ -276,16 +274,13 @@ static int e1000_set_link_ksettings(struct net_device *netdev,
 	ethtool_convert_link_mode_to_legacy_u32(&advertising,
 						cmd->link_modes.advertising);
 
-	pm_runtime_get_sync(netdev->dev.parent);
-
 	/* When SoL/IDER sessions are active, autoneg/speed/duplex
 	 * cannot be changed
 	 */
 	if (hw->phy.ops.check_reset_block &&
 	    hw->phy.ops.check_reset_block(hw)) {
 		e_err("Cannot change link characteristics when SoL/IDER is active.\n");
-		ret_val = -EINVAL;
-		goto out;
+		return -EINVAL;
 	}
 
 	/* MDI setting is only allowed when autoneg enabled because
@@ -293,16 +288,13 @@ static int e1000_set_link_ksettings(struct net_device *netdev,
 	 * duplex is forced.
 	 */
 	if (cmd->base.eth_tp_mdix_ctrl) {
-		if (hw->phy.media_type != e1000_media_type_copper) {
-			ret_val = -EOPNOTSUPP;
-			goto out;
-		}
+		if (hw->phy.media_type != e1000_media_type_copper)
+			return -EOPNOTSUPP;
 
 		if ((cmd->base.eth_tp_mdix_ctrl != ETH_TP_MDI_AUTO) &&
 		    (cmd->base.autoneg != AUTONEG_ENABLE)) {
 			e_err("forcing MDI/MDI-X state is not supported when link speed and/or duplex are forced\n");
-			ret_val = -EINVAL;
-			goto out;
+			return -EINVAL;
 		}
 	}
 
@@ -349,7 +341,6 @@ static int e1000_set_link_ksettings(struct net_device *netdev,
 	}
 
 out:
-	pm_runtime_put_sync(netdev->dev.parent);
 	clear_bit(__E1000_RESETTING, &adapter->state);
 	return ret_val;
 }
@@ -385,8 +376,6 @@ static int e1000_set_pauseparam(struct net_device *netdev,
 	while (test_and_set_bit(__E1000_RESETTING, &adapter->state))
 		usleep_range(1000, 2000);
 
-	pm_runtime_get_sync(netdev->dev.parent);
-
 	if (adapter->fc_autoneg == AUTONEG_ENABLE) {
 		hw->fc.requested_mode = e1000_fc_default;
 		if (netif_running(adapter->netdev)) {
@@ -419,7 +408,6 @@ static int e1000_set_pauseparam(struct net_device *netdev,
 	}
 
 out:
-	pm_runtime_put_sync(netdev->dev.parent);
 	clear_bit(__E1000_RESETTING, &adapter->state);
 	return retval;
 }
@@ -450,8 +438,6 @@ static void e1000_get_regs(struct net_device *netdev,
 	u32 *regs_buff = p;
 	u16 phy_data;
 
-	pm_runtime_get_sync(netdev->dev.parent);
-
 	memset(p, 0, E1000_REGS_LEN * sizeof(u32));
 
 	regs->version = (1u << 24) |
@@ -497,8 +483,6 @@ static void e1000_get_regs(struct net_device *netdev,
 	e1e_rphy(hw, MII_STAT1000, &phy_data);
 	regs_buff[24] = (u32)phy_data;	/* phy local receiver status */
 	regs_buff[25] = regs_buff[24];	/* phy remote receiver status */
-
-	pm_runtime_put_sync(netdev->dev.parent);
 }
 
 static int e1000_get_eeprom_len(struct net_device *netdev)
@@ -526,13 +510,11 @@ static int e1000_get_eeprom(struct net_device *netdev,
 	first_word = eeprom->offset >> 1;
 	last_word = (eeprom->offset + eeprom->len - 1) >> 1;
 
-	eeprom_buff = kmalloc(sizeof(u16) * (last_word - first_word + 1),
-			      GFP_KERNEL);
+	eeprom_buff = kmalloc_array(last_word - first_word + 1, sizeof(u16),
+				    GFP_KERNEL);
 	if (!eeprom_buff)
 		return -ENOMEM;
 
-	pm_runtime_get_sync(netdev->dev.parent);
-
 	if (hw->nvm.type == e1000_nvm_eeprom_spi) {
 		ret_val = e1000_read_nvm(hw, first_word,
 					 last_word - first_word + 1,
@@ -546,8 +528,6 @@ static int e1000_get_eeprom(struct net_device *netdev,
 		}
 	}
 
-	pm_runtime_put_sync(netdev->dev.parent);
-
 	if (ret_val) {
 		/* a read error occurred, throw away the result */
 		memset(eeprom_buff, 0xff, sizeof(u16) *
@@ -569,12 +549,12 @@ static int e1000_set_eeprom(struct net_device *netdev,
 {
 	struct e1000_adapter *adapter = netdev_priv(netdev);
 	struct e1000_hw *hw = &adapter->hw;
+	size_t total_len, max_len;
 	u16 *eeprom_buff;
-	void *ptr;
-	int max_len;
+	int ret_val = 0;
 	int first_word;
 	int last_word;
-	int ret_val = 0;
+	void *ptr;
 	u16 i;
 
 	if (eeprom->len == 0)
@@ -589,6 +569,10 @@ static int e1000_set_eeprom(struct net_device *netdev,
 
 	max_len = hw->nvm.word_size * 2;
 
+	if (check_add_overflow(eeprom->offset, eeprom->len, &total_len) ||
+	    total_len > max_len)
+		return -EFBIG;
+
 	first_word = eeprom->offset >> 1;
 	last_word = (eeprom->offset + eeprom->len - 1) >> 1;
 	eeprom_buff = kmalloc(max_len, GFP_KERNEL);
@@ -597,8 +581,6 @@ static int e1000_set_eeprom(struct net_device *netdev,
 
 	ptr = (void *)eeprom_buff;
 
-	pm_runtime_get_sync(netdev->dev.parent);
-
 	if (eeprom->offset & 1) {
 		/* need read/modify/write of first changed EEPROM word */
 		/* only the second byte of the word is being modified */
@@ -639,7 +621,6 @@ static int e1000_set_eeprom(struct net_device *netdev,
 		ret_val = e1000e_update_nvm_checksum(hw);
 
 out:
-	pm_runtime_put_sync(netdev->dev.parent);
 	kfree(eeprom_buff);
 	return ret_val;
 }
@@ -649,25 +630,25 @@ static void e1000_get_drvinfo(struct net_device *netdev,
 {
 	struct e1000_adapter *adapter = netdev_priv(netdev);
 
-	strlcpy(drvinfo->driver, e1000e_driver_name, sizeof(drvinfo->driver));
-	strlcpy(drvinfo->version, e1000e_driver_version,
-		sizeof(drvinfo->version));
+	strscpy(drvinfo->driver, e1000e_driver_name, sizeof(drvinfo->driver));
 
 	/* EEPROM image version # is reported as firmware version # for
 	 * PCI-E controllers
 	 */
 	snprintf(drvinfo->fw_version, sizeof(drvinfo->fw_version),
 		 "%d.%d-%d",
-		 (adapter->eeprom_vers & 0xF000) >> 12,
-		 (adapter->eeprom_vers & 0x0FF0) >> 4,
+		 FIELD_GET(0xF000, adapter->eeprom_vers),
+		 FIELD_GET(0x0FF0, adapter->eeprom_vers),
 		 (adapter->eeprom_vers & 0x000F));
 
-	strlcpy(drvinfo->bus_info, pci_name(adapter->pdev),
+	strscpy(drvinfo->bus_info, pci_name(adapter->pdev),
 		sizeof(drvinfo->bus_info));
 }
 
 static void e1000_get_ringparam(struct net_device *netdev,
-				struct ethtool_ringparam *ring)
+				struct ethtool_ringparam *ring,
+				struct kernel_ethtool_ringparam *kernel_ring,
+				struct netlink_ext_ack *extack)
 {
 	struct e1000_adapter *adapter = netdev_priv(netdev);
 
@@ -678,7 +659,9 @@ static void e1000_get_ringparam(struct net_device *netdev,
 }
 
 static int e1000_set_ringparam(struct net_device *netdev,
-			       struct ethtool_ringparam *ring)
+			       struct ethtool_ringparam *ring,
+			       struct kernel_ethtool_ringparam *kernel_ring,
+			       struct netlink_ext_ack *extack)
 {
 	struct e1000_adapter *adapter = netdev_priv(netdev);
 	struct e1000_ring *temp_tx = NULL, *temp_rx = NULL;
@@ -733,8 +716,6 @@ static int e1000_set_ringparam(struct net_device *netdev,
 		}
 	}
 
-	pm_runtime_get_sync(netdev->dev.parent);
-
 	e1000e_down(adapter, true);
 
 	/* We can't just free everything and then setup again, because the
@@ -773,7 +754,6 @@ err_setup_rx:
 		e1000e_free_tx_resources(temp_tx);
 err_setup:
 	e1000e_up(adapter);
-	pm_runtime_put_sync(netdev->dev.parent);
 free_temp:
 	vfree(temp_tx);
 	vfree(temp_rx);
@@ -911,8 +891,13 @@ static int e1000_reg_test(struct e1000_adapter *adapter, u64 *data)
 	case e1000_pch2lan:
 	case e1000_pch_lpt:
 	case e1000_pch_spt:
-		/* fall through */
 	case e1000_pch_cnp:
+	case e1000_pch_tgp:
+	case e1000_pch_adp:
+	case e1000_pch_mtp:
+	case e1000_pch_lnp:
+	case e1000_pch_ptp:
+	case e1000_pch_nvp:
 		mask |= BIT(18);
 		break;
 	default:
@@ -920,8 +905,7 @@ static int e1000_reg_test(struct e1000_adapter *adapter, u64 *data)
 	}
 
 	if (mac->type >= e1000_pch_lpt)
-		wlock_mac = (er32(FWSM) & E1000_FWSM_WLOCK_MAC_MASK) >>
-		    E1000_FWSM_WLOCK_MAC_SHIFT;
+		wlock_mac = FIELD_GET(E1000_FWSM_WLOCK_MAC_MASK, er32(FWSM));
 
 	for (i = 0; i < mac->rar_entry_count; i++) {
 		if (mac->type >= e1000_pch_lpt) {
@@ -979,7 +963,7 @@ static int e1000_eeprom_test(struct e1000_adapter *adapter, u64 *data)
 	}
 
 	/* If Checksum is not Correct return error else test passed */
-	if ((checksum != (u16)NVM_SUM) && !(*data))
+	if (checksum != NVM_SUM && !(*data))
 		*data = 2;
 
 	return *data;
@@ -1031,7 +1015,7 @@ static int e1000_intr_test(struct e1000_adapter *adapter, u64 *data)
 	/* Disable all the interrupts */
 	ew32(IMC, 0xFFFFFFFF);
 	e1e_flush();
-	usleep_range(10000, 20000);
+	usleep_range(10000, 11000);
 
 	/* Test each interrupt */
 	for (i = 0; i < 10; i++) {
@@ -1063,7 +1047,7 @@ static int e1000_intr_test(struct e1000_adapter *adapter, u64 *data)
 			ew32(IMC, mask);
 			ew32(ICS, mask);
 			e1e_flush();
-			usleep_range(10000, 20000);
+			usleep_range(10000, 11000);
 
 			if (adapter->test_icr & mask) {
 				*data = 3;
@@ -1081,7 +1065,7 @@ static int e1000_intr_test(struct e1000_adapter *adapter, u64 *data)
 		ew32(IMS, mask);
 		ew32(ICS, mask);
 		e1e_flush();
-		usleep_range(10000, 20000);
+		usleep_range(10000, 11000);
 
 		if (!(adapter->test_icr & mask)) {
 			*data = 4;
@@ -1099,7 +1083,7 @@ static int e1000_intr_test(struct e1000_adapter *adapter, u64 *data)
 			ew32(IMC, ~mask & 0x00007FFF);
 			ew32(ICS, ~mask & 0x00007FFF);
 			e1e_flush();
-			usleep_range(10000, 20000);
+			usleep_range(10000, 11000);
 
 			if (adapter->test_icr) {
 				*data = 5;
@@ -1111,7 +1095,7 @@ static int e1000_intr_test(struct e1000_adapter *adapter, u64 *data)
 	/* Disable all the interrupts */
 	ew32(IMC, 0xFFFFFFFF);
 	e1e_flush();
-	usleep_range(10000, 20000);
+	usleep_range(10000, 11000);
 
 	/* Unhook test interrupt handler */
 	free_irq(irq, netdev);
@@ -1143,8 +1127,7 @@ static void e1000_free_desc_rings(struct e1000_adapter *adapter)
 						 buffer_info->dma,
 						 buffer_info->length,
 						 DMA_TO_DEVICE);
-			if (buffer_info->skb)
-				dev_kfree_skb(buffer_info->skb);
+			dev_kfree_skb(buffer_info->skb);
 		}
 	}
 
@@ -1156,8 +1139,7 @@ static void e1000_free_desc_rings(struct e1000_adapter *adapter)
 				dma_unmap_single(&pdev->dev,
 						 buffer_info->dma,
 						 2048, DMA_FROM_DEVICE);
-			if (buffer_info->skb)
-				dev_kfree_skb(buffer_info->skb);
+			dev_kfree_skb(buffer_info->skb);
 		}
 	}
 
@@ -1487,7 +1469,7 @@ static int e1000_set_82571_fiber_loopback(struct e1000_adapter *adapter)
 	 */
 	ew32(SCTL, E1000_SCTL_ENABLE_SERDES_LOOPBACK);
 	e1e_flush();
-	usleep_range(10000, 20000);
+	usleep_range(10000, 11000);
 
 	return 0;
 }
@@ -1578,6 +1560,12 @@ static void e1000_loopback_cleanup(struct e1000_adapter *adapter)
 	switch (hw->mac.type) {
 	case e1000_pch_spt:
 	case e1000_pch_cnp:
+	case e1000_pch_tgp:
+	case e1000_pch_adp:
+	case e1000_pch_mtp:
+	case e1000_pch_lnp:
+	case e1000_pch_ptp:
+	case e1000_pch_nvp:
 		fext_nvm11 = er32(FEXTNVM11);
 		fext_nvm11 &= ~E1000_FEXTNVM11_DISABLE_MULR_FIX;
 		ew32(FEXTNVM11, fext_nvm11);
@@ -1586,7 +1574,7 @@ static void e1000_loopback_cleanup(struct e1000_adapter *adapter)
 		/* set bit 29 (value of MULR requests is now 0) */
 		tarc0 &= 0xcfffffff;
 		ew32(TARC(0), tarc0);
-		/* fall through */
+		fallthrough;
 	case e1000_80003es2lan:
 		if (hw->phy.media_type == e1000_media_type_fiber ||
 		    hw->phy.media_type == e1000_media_type_internal_serdes) {
@@ -1594,17 +1582,17 @@ static void e1000_loopback_cleanup(struct e1000_adapter *adapter)
 			ew32(CTRL_EXT, adapter->tx_fifo_head);
 			adapter->tx_fifo_head = 0;
 		}
-		/* fall through */
+		fallthrough;
 	case e1000_82571:
 	case e1000_82572:
 		if (hw->phy.media_type == e1000_media_type_fiber ||
 		    hw->phy.media_type == e1000_media_type_internal_serdes) {
 			ew32(SCTL, E1000_SCTL_DISABLE_SERDES_LOOPBACK);
 			e1e_flush();
-			usleep_range(10000, 20000);
+			usleep_range(10000, 11000);
 			break;
 		}
-		/* Fall Through */
+		fallthrough;
 	default:
 		hw->mac.autoneg = 1;
 		if (hw->phy.type == e1000_phy_gg82563)
@@ -1626,8 +1614,8 @@ static void e1000_create_lbtest_frame(struct sk_buff *skb,
 	memset(skb->data, 0xFF, frame_size);
 	frame_size &= ~1;
 	memset(&skb->data[frame_size / 2], 0xAA, frame_size / 2 - 1);
-	memset(&skb->data[frame_size / 2 + 10], 0xBE, 1);
-	memset(&skb->data[frame_size / 2 + 12], 0xAF, 1);
+	skb->data[frame_size / 2 + 10] = 0xBE;
+	skb->data[frame_size / 2 + 12] = 0xAF;
 }
 
 static int e1000_check_lbtest_frame(struct sk_buff *skb,
@@ -1792,6 +1780,8 @@ static int e1000e_get_sset_count(struct net_device __always_unused *netdev,
 		return E1000_TEST_LEN;
 	case ETH_SS_STATS:
 		return E1000_STATS_LEN;
+	case ETH_SS_PRIV_FLAGS:
+		return E1000E_PRIV_FLAGS_STR_LEN;
 	default:
 		return -EOPNOTSUPP;
 	}
@@ -1806,8 +1796,6 @@ static void e1000_diag_test(struct net_device *netdev,
 	u8 autoneg;
 	bool if_running = netif_running(netdev);
 
-	pm_runtime_get_sync(netdev->dev.parent);
-
 	set_bit(__E1000_TESTING, &adapter->state);
 
 	if (!if_running) {
@@ -1893,8 +1881,6 @@ static void e1000_diag_test(struct net_device *netdev,
 	}
 
 	msleep_interruptible(4 * 1000);
-
-	pm_runtime_put_sync(netdev->dev.parent);
 }
 
 static void e1000_get_wol(struct net_device *netdev,
@@ -1998,7 +1984,9 @@ static int e1000_set_phys_id(struct net_device *netdev,
 }
 
 static int e1000_get_coalesce(struct net_device *netdev,
-			      struct ethtool_coalesce *ec)
+			      struct ethtool_coalesce *ec,
+			      struct kernel_ethtool_coalesce *kernel_coal,
+			      struct netlink_ext_ack *extack)
 {
 	struct e1000_adapter *adapter = netdev_priv(netdev);
 
@@ -2011,7 +1999,9 @@ static int e1000_get_coalesce(struct net_device *netdev,
 }
 
 static int e1000_set_coalesce(struct net_device *netdev,
-			      struct ethtool_coalesce *ec)
+			      struct ethtool_coalesce *ec,
+			      struct kernel_ethtool_coalesce *kernel_coal,
+			      struct netlink_ext_ack *extack)
 {
 	struct e1000_adapter *adapter = netdev_priv(netdev);
 
@@ -2032,15 +2022,11 @@ static int e1000_set_coalesce(struct net_device *netdev,
 		adapter->itr_setting = adapter->itr & ~3;
 	}
 
-	pm_runtime_get_sync(netdev->dev.parent);
-
 	if (adapter->itr_setting != 0)
 		e1000e_write_itr(adapter, adapter->itr);
 	else
 		e1000e_write_itr(adapter, 0);
 
-	pm_runtime_put_sync(netdev->dev.parent);
-
 	return 0;
 }
 
@@ -2054,9 +2040,7 @@ static int e1000_nway_reset(struct net_device *netdev)
 	if (!adapter->hw.mac.autoneg)
 		return -EINVAL;
 
-	pm_runtime_get_sync(netdev->dev.parent);
 	e1000e_reinit_locked(adapter);
-	pm_runtime_put_sync(netdev->dev.parent);
 
 	return 0;
 }
@@ -2070,11 +2054,7 @@ static void e1000_get_ethtool_stats(struct net_device *netdev,
 	int i;
 	char *p = NULL;
 
-	pm_runtime_get_sync(netdev->dev.parent);
-
-	e1000e_get_stats64(netdev, &net_stats);
-
-	pm_runtime_put_sync(netdev->dev.parent);
+	dev_get_stats(netdev, &net_stats);
 
 	for (i = 0; i < E1000_GLOBAL_STATS_LEN; i++) {
 		switch (e1000_gstrings_stats[i].type) {
@@ -2113,62 +2093,57 @@ static void e1000_get_strings(struct net_device __always_unused *netdev,
 			p += ETH_GSTRING_LEN;
 		}
 		break;
+	case ETH_SS_PRIV_FLAGS:
+		memcpy(data, e1000e_priv_flags_strings,
+		       E1000E_PRIV_FLAGS_STR_LEN * ETH_GSTRING_LEN);
+		break;
 	}
 }
 
-static int e1000_get_rxnfc(struct net_device *netdev,
-			   struct ethtool_rxnfc *info,
-			   u32 __always_unused *rule_locs)
+static int e1000_get_rxfh_fields(struct net_device *netdev,
+				 struct ethtool_rxfh_fields *info)
 {
-	info->data = 0;
-
-	switch (info->cmd) {
-	case ETHTOOL_GRXFH: {
-		struct e1000_adapter *adapter = netdev_priv(netdev);
-		struct e1000_hw *hw = &adapter->hw;
-		u32 mrqc;
+	struct e1000_adapter *adapter = netdev_priv(netdev);
+	struct e1000_hw *hw = &adapter->hw;
+	u32 mrqc;
 
-		pm_runtime_get_sync(netdev->dev.parent);
-		mrqc = er32(MRQC);
-		pm_runtime_put_sync(netdev->dev.parent);
+	info->data = 0;
 
-		if (!(mrqc & E1000_MRQC_RSS_FIELD_MASK))
-			return 0;
+	mrqc = er32(MRQC);
 
-		switch (info->flow_type) {
-		case TCP_V4_FLOW:
-			if (mrqc & E1000_MRQC_RSS_FIELD_IPV4_TCP)
-				info->data |= RXH_L4_B_0_1 | RXH_L4_B_2_3;
-			/* fall through */
-		case UDP_V4_FLOW:
-		case SCTP_V4_FLOW:
-		case AH_ESP_V4_FLOW:
-		case IPV4_FLOW:
-			if (mrqc & E1000_MRQC_RSS_FIELD_IPV4)
-				info->data |= RXH_IP_SRC | RXH_IP_DST;
-			break;
-		case TCP_V6_FLOW:
-			if (mrqc & E1000_MRQC_RSS_FIELD_IPV6_TCP)
-				info->data |= RXH_L4_B_0_1 | RXH_L4_B_2_3;
-			/* fall through */
-		case UDP_V6_FLOW:
-		case SCTP_V6_FLOW:
-		case AH_ESP_V6_FLOW:
-		case IPV6_FLOW:
-			if (mrqc & E1000_MRQC_RSS_FIELD_IPV6)
-				info->data |= RXH_IP_SRC | RXH_IP_DST;
-			break;
-		default:
-			break;
-		}
+	if (!(mrqc & E1000_MRQC_RSS_FIELD_MASK))
 		return 0;
-	}
+
+	switch (info->flow_type) {
+	case TCP_V4_FLOW:
+		if (mrqc & E1000_MRQC_RSS_FIELD_IPV4_TCP)
+			info->data |= RXH_L4_B_0_1 | RXH_L4_B_2_3;
+		fallthrough;
+	case UDP_V4_FLOW:
+	case SCTP_V4_FLOW:
+	case AH_ESP_V4_FLOW:
+	case IPV4_FLOW:
+		if (mrqc & E1000_MRQC_RSS_FIELD_IPV4)
+			info->data |= RXH_IP_SRC | RXH_IP_DST;
+		break;
+	case TCP_V6_FLOW:
+		if (mrqc & E1000_MRQC_RSS_FIELD_IPV6_TCP)
+			info->data |= RXH_L4_B_0_1 | RXH_L4_B_2_3;
+		fallthrough;
+	case UDP_V6_FLOW:
+	case SCTP_V6_FLOW:
+	case AH_ESP_V6_FLOW:
+	case IPV6_FLOW:
+		if (mrqc & E1000_MRQC_RSS_FIELD_IPV6)
+			info->data |= RXH_IP_SRC | RXH_IP_DST;
+		break;
 	default:
-		return -EOPNOTSUPP;
+		break;
 	}
+	return 0;
 }
 
-static int e1000e_get_eee(struct net_device *netdev, struct ethtool_eee *edata)
+static int e1000e_get_eee(struct net_device *netdev, struct ethtool_keee *edata)
 {
 	struct e1000_adapter *adapter = netdev_priv(netdev);
 	struct e1000_hw *hw = &adapter->hw;
@@ -2193,28 +2168,24 @@ static int e1000e_get_eee(struct net_device *netdev, struct ethtool_eee *edata)
 		return -EOPNOTSUPP;
 	}
 
-	pm_runtime_get_sync(netdev->dev.parent);
-
 	ret_val = hw->phy.ops.acquire(hw);
-	if (ret_val) {
-		pm_runtime_put_sync(netdev->dev.parent);
+	if (ret_val)
 		return -EBUSY;
-	}
 
 	/* EEE Capability */
 	ret_val = e1000_read_emi_reg_locked(hw, cap_addr, &phy_data);
 	if (ret_val)
 		goto release;
-	edata->supported = mmd_eee_cap_to_ethtool_sup_t(phy_data);
+	mii_eee_cap1_mod_linkmode_t(edata->supported, phy_data);
 
 	/* EEE Advertised */
-	edata->advertised = mmd_eee_adv_to_ethtool_adv_t(adapter->eee_advert);
+	mii_eee_cap1_mod_linkmode_t(edata->advertised, adapter->eee_advert);
 
 	/* EEE Link Partner Advertised */
 	ret_val = e1000_read_emi_reg_locked(hw, lpa_addr, &phy_data);
 	if (ret_val)
 		goto release;
-	edata->lp_advertised = mmd_eee_adv_to_ethtool_adv_t(phy_data);
+	mii_eee_cap1_mod_linkmode_t(edata->lp_advertised, phy_data);
 
 	/* EEE PCS Status */
 	ret_val = e1000_read_emi_reg_locked(hw, pcs_stat_addr, &phy_data);
@@ -2239,16 +2210,16 @@ release:
 	if (ret_val)
 		ret_val = -ENODATA;
 
-	pm_runtime_put_sync(netdev->dev.parent);
-
 	return ret_val;
 }
 
-static int e1000e_set_eee(struct net_device *netdev, struct ethtool_eee *edata)
+static int e1000e_set_eee(struct net_device *netdev, struct ethtool_keee *edata)
 {
 	struct e1000_adapter *adapter = netdev_priv(netdev);
+	__ETHTOOL_DECLARE_LINK_MODE_MASK(supported) = {};
+	__ETHTOOL_DECLARE_LINK_MODE_MASK(tmp) = {};
 	struct e1000_hw *hw = &adapter->hw;
-	struct ethtool_eee eee_curr;
+	struct ethtool_keee eee_curr;
 	s32 ret_val;
 
 	ret_val = e1000e_get_eee(netdev, &eee_curr);
@@ -2265,30 +2236,31 @@ static int e1000e_set_eee(struct net_device *netdev, struct ethtool_eee *edata)
 		return -EINVAL;
 	}
 
-	if (edata->advertised & ~(ADVERTISE_100_FULL | ADVERTISE_1000_FULL)) {
+	linkmode_set_bit(ETHTOOL_LINK_MODE_1000baseT_Full_BIT,
+			 supported);
+	linkmode_set_bit(ETHTOOL_LINK_MODE_100baseT_Full_BIT,
+			 supported);
+
+	if (linkmode_andnot(tmp, edata->advertised, supported)) {
 		e_err("EEE advertisement supports only 100TX and/or 1000T full-duplex\n");
 		return -EINVAL;
 	}
 
-	adapter->eee_advert = ethtool_adv_to_mmd_eee_adv_t(edata->advertised);
+	adapter->eee_advert = linkmode_to_mii_eee_cap1_t(edata->advertised);
 
 	hw->dev_spec.ich8lan.eee_disable = !edata->eee_enabled;
 
-	pm_runtime_get_sync(netdev->dev.parent);
-
 	/* reset the link */
 	if (netif_running(netdev))
 		e1000e_reinit_locked(adapter);
 	else
 		e1000e_reset(adapter);
 
-	pm_runtime_put_sync(netdev->dev.parent);
-
 	return 0;
 }
 
 static int e1000e_get_ts_info(struct net_device *netdev,
-			      struct ethtool_ts_info *info)
+			      struct kernel_ethtool_ts_info *info)
 {
 	struct e1000_adapter *adapter = netdev_priv(netdev);
 
@@ -2321,7 +2293,39 @@ static int e1000e_get_ts_info(struct net_device *netdev,
 	return 0;
 }
 
+static u32 e1000e_get_priv_flags(struct net_device *netdev)
+{
+	struct e1000_adapter *adapter = netdev_priv(netdev);
+	u32 priv_flags = 0;
+
+	if (adapter->flags2 & FLAG2_ENABLE_S0IX_FLOWS)
+		priv_flags |= E1000E_PRIV_FLAGS_S0IX_ENABLED;
+
+	return priv_flags;
+}
+
+static int e1000e_set_priv_flags(struct net_device *netdev, u32 priv_flags)
+{
+	struct e1000_adapter *adapter = netdev_priv(netdev);
+	unsigned int flags2 = adapter->flags2;
+
+	flags2 &= ~FLAG2_ENABLE_S0IX_FLOWS;
+	if (priv_flags & E1000E_PRIV_FLAGS_S0IX_ENABLED) {
+		struct e1000_hw *hw = &adapter->hw;
+
+		if (hw->mac.type < e1000_pch_cnp)
+			return -EINVAL;
+		flags2 |= FLAG2_ENABLE_S0IX_FLOWS;
+	}
+
+	if (flags2 != adapter->flags2)
+		adapter->flags2 = flags2;
+
+	return 0;
+}
+
 static const struct ethtool_ops e1000_ethtool_ops = {
+	.supported_coalesce_params = ETHTOOL_COALESCE_RX_USECS,
 	.get_drvinfo		= e1000_get_drvinfo,
 	.get_regs_len		= e1000_get_regs_len,
 	.get_regs		= e1000_get_regs,
@@ -2345,12 +2349,14 @@ static const struct ethtool_ops e1000_ethtool_ops = {
 	.get_sset_count		= e1000e_get_sset_count,
 	.get_coalesce		= e1000_get_coalesce,
 	.set_coalesce		= e1000_set_coalesce,
-	.get_rxnfc		= e1000_get_rxnfc,
+	.get_rxfh_fields	= e1000_get_rxfh_fields,
 	.get_ts_info		= e1000e_get_ts_info,
 	.get_eee		= e1000e_get_eee,
 	.set_eee		= e1000e_set_eee,
 	.get_link_ksettings	= e1000_get_link_ksettings,
 	.set_link_ksettings	= e1000_set_link_ksettings,
+	.get_priv_flags		= e1000e_get_priv_flags,
+	.set_priv_flags		= e1000e_set_priv_flags,
 };
 
 void e1000e_set_ethtool_ops(struct net_device *netdev)
diff --git a/drivers/net/ethernet/intel/e1000e/hw.h b/drivers/net/ethernet/intel/e1000e/hw.h
index 66bd5060a65b..fc8ed38aa095 100644
--- a/drivers/net/ethernet/intel/e1000e/hw.h
+++ b/drivers/net/ethernet/intel/e1000e/hw.h
@@ -1,26 +1,8 @@
-/* Intel PRO/1000 Linux driver
- * Copyright(c) 1999 - 2015 Intel Corporation.
- *
- * This program is free software; you can redistribute it and/or modify it
- * under the terms and conditions of the GNU General Public License,
- * version 2, as published by the Free Software Foundation.
- *
- * This program is distributed in the hope it will be useful, but WITHOUT
- * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
- * FITNESS FOR A PARTICULAR PURPOSE.  See the GNU General Public License for
- * more details.
- *
- * The full GNU General Public License is included in this distribution in
- * the file called "COPYING".
- *
- * Contact Information:
- * Linux NICS <linux.nics@intel.com>
- * e1000-devel Mailing List <e1000-devel@lists.sourceforge.net>
- * Intel Corporation, 5200 N.E. Elam Young Parkway, Hillsboro, OR 97124-6497
- */
+/* SPDX-License-Identifier: GPL-2.0 */
+/* Copyright(c) 1999 - 2018 Intel Corporation. */
 
-#ifndef _E1000_HW_H_
-#define _E1000_HW_H_
+#ifndef _E1000E_HW_H_
+#define _E1000E_HW_H_
 
 #include "regs.h"
 #include "defines.h"
@@ -100,6 +82,48 @@ struct e1000_hw;
 #define E1000_DEV_ID_PCH_CNP_I219_V6		0x15BE
 #define E1000_DEV_ID_PCH_CNP_I219_LM7		0x15BB
 #define E1000_DEV_ID_PCH_CNP_I219_V7		0x15BC
+#define E1000_DEV_ID_PCH_ICP_I219_LM8		0x15DF
+#define E1000_DEV_ID_PCH_ICP_I219_V8		0x15E0
+#define E1000_DEV_ID_PCH_ICP_I219_LM9		0x15E1
+#define E1000_DEV_ID_PCH_ICP_I219_V9		0x15E2
+#define E1000_DEV_ID_PCH_CMP_I219_LM10		0x0D4E
+#define E1000_DEV_ID_PCH_CMP_I219_V10		0x0D4F
+#define E1000_DEV_ID_PCH_CMP_I219_LM11		0x0D4C
+#define E1000_DEV_ID_PCH_CMP_I219_V11		0x0D4D
+#define E1000_DEV_ID_PCH_CMP_I219_LM12		0x0D53
+#define E1000_DEV_ID_PCH_CMP_I219_V12		0x0D55
+#define E1000_DEV_ID_PCH_TGP_I219_LM13		0x15FB
+#define E1000_DEV_ID_PCH_TGP_I219_V13		0x15FC
+#define E1000_DEV_ID_PCH_TGP_I219_LM14		0x15F9
+#define E1000_DEV_ID_PCH_TGP_I219_V14		0x15FA
+#define E1000_DEV_ID_PCH_TGP_I219_LM15		0x15F4
+#define E1000_DEV_ID_PCH_TGP_I219_V15		0x15F5
+#define E1000_DEV_ID_PCH_RPL_I219_LM23		0x0DC5
+#define E1000_DEV_ID_PCH_RPL_I219_V23		0x0DC6
+#define E1000_DEV_ID_PCH_ADP_I219_LM16		0x1A1E
+#define E1000_DEV_ID_PCH_ADP_I219_V16		0x1A1F
+#define E1000_DEV_ID_PCH_ADP_I219_LM17		0x1A1C
+#define E1000_DEV_ID_PCH_ADP_I219_V17		0x1A1D
+#define E1000_DEV_ID_PCH_RPL_I219_LM22		0x0DC7
+#define E1000_DEV_ID_PCH_RPL_I219_V22		0x0DC8
+#define E1000_DEV_ID_PCH_MTP_I219_LM18		0x550A
+#define E1000_DEV_ID_PCH_MTP_I219_V18		0x550B
+#define E1000_DEV_ID_PCH_ADP_I219_LM19		0x550C
+#define E1000_DEV_ID_PCH_ADP_I219_V19		0x550D
+#define E1000_DEV_ID_PCH_LNP_I219_LM20		0x550E
+#define E1000_DEV_ID_PCH_LNP_I219_V20		0x550F
+#define E1000_DEV_ID_PCH_LNP_I219_LM21		0x5510
+#define E1000_DEV_ID_PCH_LNP_I219_V21		0x5511
+#define E1000_DEV_ID_PCH_ARL_I219_LM24		0x57A0
+#define E1000_DEV_ID_PCH_ARL_I219_V24		0x57A1
+#define E1000_DEV_ID_PCH_PTP_I219_LM25		0x57B3
+#define E1000_DEV_ID_PCH_PTP_I219_V25		0x57B4
+#define E1000_DEV_ID_PCH_PTP_I219_LM26		0x57B5
+#define E1000_DEV_ID_PCH_PTP_I219_V26		0x57B6
+#define E1000_DEV_ID_PCH_PTP_I219_LM27		0x57B7
+#define E1000_DEV_ID_PCH_PTP_I219_V27		0x57B8
+#define E1000_DEV_ID_PCH_NVL_I219_LM29		0x57B9
+#define E1000_DEV_ID_PCH_NVL_I219_V29		0x57BA
 
 #define E1000_REVISION_4	4
 
@@ -123,6 +147,12 @@ enum e1000_mac_type {
 	e1000_pch_lpt,
 	e1000_pch_spt,
 	e1000_pch_cnp,
+	e1000_pch_tgp,
+	e1000_pch_adp,
+	e1000_pch_mtp,
+	e1000_pch_lnp,
+	e1000_pch_ptp,
+	e1000_pch_nvp,
 };
 
 enum e1000_media_type {
@@ -598,6 +628,7 @@ struct e1000_phy_info {
 	u32 id;
 	u32 reset_delay_us;	/* in usec */
 	u32 revision;
+	u32 retry_count;
 
 	enum e1000_media_type media_type;
 
@@ -614,6 +645,7 @@ struct e1000_phy_info {
 	bool polarity_correction;
 	bool speed_downgraded;
 	bool autoneg_wait_to_complete;
+	bool retry_enabled;
 };
 
 struct e1000_nvm_info {
@@ -705,4 +737,4 @@ struct e1000_hw {
 #include "80003es2lan.h"
 #include "ich8lan.h"
 
-#endif
+#endif /* _E1000E_HW_H_ */
diff --git a/drivers/net/ethernet/intel/e1000e/ich8lan.c b/drivers/net/ethernet/intel/e1000e/ich8lan.c
index 68ea8b4555ab..df4e7d781cb1 100644
--- a/drivers/net/ethernet/intel/e1000e/ich8lan.c
+++ b/drivers/net/ethernet/intel/e1000e/ich8lan.c
@@ -1,23 +1,5 @@
-/* Intel PRO/1000 Linux driver
- * Copyright(c) 1999 - 2015 Intel Corporation.
- *
- * This program is free software; you can redistribute it and/or modify it
- * under the terms and conditions of the GNU General Public License,
- * version 2, as published by the Free Software Foundation.
- *
- * This program is distributed in the hope it will be useful, but WITHOUT
- * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
- * FITNESS FOR A PARTICULAR PURPOSE.  See the GNU General Public License for
- * more details.
- *
- * The full GNU General Public License is included in this distribution in
- * the file called "COPYING".
- *
- * Contact Information:
- * Linux NICS <linux.nics@intel.com>
- * e1000-devel Mailing List <e1000-devel@lists.sourceforge.net>
- * Intel Corporation, 5200 N.E. Elam Young Parkway, Hillsboro, OR 97124-6497
- */
+// SPDX-License-Identifier: GPL-2.0
+/* Copyright(c) 1999 - 2018 Intel Corporation. */
 
 /* 82562G 10/100 Network Connection
  * 82562G-2 10/100 Network Connection
@@ -240,11 +222,18 @@ out:
 	if (hw->mac.type >= e1000_pch_lpt) {
 		/* Only unforce SMBus if ME is not active */
 		if (!(er32(FWSM) & E1000_ICH_FWSM_FW_VALID)) {
+			/* Switching PHY interface always returns MDI error
+			 * so disable retry mechanism to avoid wasting time
+			 */
+			e1000e_disable_phy_retry(hw);
+
 			/* Unforce SMBus mode in PHY */
 			e1e_rphy_locked(hw, CV_SMB_CTRL, &phy_reg);
 			phy_reg &= ~CV_SMB_CTRL_FORCE_SMBUS;
 			e1e_wphy_locked(hw, CV_SMB_CTRL, phy_reg);
 
+			e1000e_enable_phy_retry(hw);
+
 			/* Unforce SMBus mode in MAC */
 			mac_reg = er32(CTRL_EXT);
 			mac_reg &= ~E1000_CTRL_EXT_FORCE_SMBUS;
@@ -289,7 +278,7 @@ static void e1000_toggle_lanphypc_pch_lpt(struct e1000_hw *hw)
 		u16 count = 20;
 
 		do {
-			usleep_range(5000, 10000);
+			usleep_range(5000, 6000);
 		} while (!(er32(CTRL_EXT) & E1000_CTRL_EXT_LPCD) && count--);
 
 		msleep(30);
@@ -297,6 +286,45 @@ static void e1000_toggle_lanphypc_pch_lpt(struct e1000_hw *hw)
 }
 
 /**
+ * e1000_reconfigure_k1_exit_timeout - reconfigure K1 exit timeout to
+ * align to MTP and later platform requirements.
+ * @hw: pointer to the HW structure
+ *
+ * Context: PHY semaphore must be held by caller.
+ * Return: 0 on success, negative on failure
+ */
+static s32 e1000_reconfigure_k1_exit_timeout(struct e1000_hw *hw)
+{
+	u16 phy_timeout;
+	u32 fextnvm12;
+	s32 ret_val;
+
+	if (hw->mac.type < e1000_pch_mtp)
+		return 0;
+
+	/* Change Kumeran K1 power down state from P0s to P1 */
+	fextnvm12 = er32(FEXTNVM12);
+	fextnvm12 &= ~E1000_FEXTNVM12_PHYPD_CTRL_MASK;
+	fextnvm12 |= E1000_FEXTNVM12_PHYPD_CTRL_P1;
+	ew32(FEXTNVM12, fextnvm12);
+
+	/* Wait for the interface the settle */
+	usleep_range(1000, 1100);
+
+	/* Change K1 exit timeout */
+	ret_val = e1e_rphy_locked(hw, I217_PHY_TIMEOUTS_REG,
+				  &phy_timeout);
+	if (ret_val)
+		return ret_val;
+
+	phy_timeout &= ~I217_PHY_TIMEOUTS_K1_EXIT_TO_MASK;
+	phy_timeout |= 0xF00;
+
+	return e1e_wphy_locked(hw, I217_PHY_TIMEOUTS_REG,
+				  phy_timeout);
+}
+
+/**
  *  e1000_init_phy_workarounds_pchlan - PHY initialization workarounds
  *  @hw: pointer to the HW structure
  *
@@ -318,7 +346,9 @@ static s32 e1000_init_phy_workarounds_pchlan(struct e1000_hw *hw)
 	 * so forcibly disable it.
 	 */
 	hw->dev_spec.ich8lan.ulp_state = e1000_ulp_state_unknown;
-	e1000_disable_ulp_lpt_lp(hw, true);
+	ret_val = e1000_disable_ulp_lpt_lp(hw, true);
+	if (ret_val)
+		e_warn("Failed to disable ULP\n");
 
 	ret_val = hw->phy.ops.acquire(hw);
 	if (ret_val) {
@@ -326,14 +356,32 @@ static s32 e1000_init_phy_workarounds_pchlan(struct e1000_hw *hw)
 		goto out;
 	}
 
+	/* There is no guarantee that the PHY is accessible at this time
+	 * so disable retry mechanism to avoid wasting time
+	 */
+	e1000e_disable_phy_retry(hw);
+
 	/* The MAC-PHY interconnect may be in SMBus mode.  If the PHY is
 	 * inaccessible and resetting the PHY is not blocked, toggle the
 	 * LANPHYPC Value bit to force the interconnect to PCIe mode.
 	 */
 	switch (hw->mac.type) {
+	case e1000_pch_mtp:
+	case e1000_pch_lnp:
+	case e1000_pch_ptp:
+	case e1000_pch_nvp:
+		/* At this point the PHY might be inaccessible so don't
+		 * propagate the failure
+		 */
+		if (e1000_reconfigure_k1_exit_timeout(hw))
+			e_dbg("Failed to reconfigure K1 exit timeout\n");
+
+		fallthrough;
 	case e1000_pch_lpt:
 	case e1000_pch_spt:
 	case e1000_pch_cnp:
+	case e1000_pch_tgp:
+	case e1000_pch_adp:
 		if (e1000_phy_is_accessible_pchlan(hw))
 			break;
 
@@ -350,12 +398,12 @@ static s32 e1000_init_phy_workarounds_pchlan(struct e1000_hw *hw)
 		 */
 		msleep(50);
 
-		/* fall-through */
+		fallthrough;
 	case e1000_pch2lan:
 		if (e1000_phy_is_accessible_pchlan(hw))
 			break;
 
-		/* fall-through */
+		fallthrough;
 	case e1000_pchlan:
 		if ((hw->mac.type == e1000_pchlan) &&
 		    (fwsm & E1000_ICH_FWSM_FW_VALID))
@@ -390,6 +438,8 @@ static s32 e1000_init_phy_workarounds_pchlan(struct e1000_hw *hw)
 		break;
 	}
 
+	e1000e_enable_phy_retry(hw);
+
 	hw->phy.ops.release(hw);
 	if (!ret_val) {
 
@@ -415,15 +465,27 @@ static s32 e1000_init_phy_workarounds_pchlan(struct e1000_hw *hw)
 		 *  the PHY is in.
 		 */
 		ret_val = hw->phy.ops.check_reset_block(hw);
-		if (ret_val)
+		if (ret_val) {
 			e_err("ME blocked access to PHY after reset\n");
+			goto out;
+		}
+
+		if (hw->mac.type >= e1000_pch_mtp) {
+			ret_val = hw->phy.ops.acquire(hw);
+			if (ret_val) {
+				e_err("Failed to reconfigure K1 exit timeout\n");
+				goto out;
+			}
+			ret_val = e1000_reconfigure_k1_exit_timeout(hw);
+			hw->phy.ops.release(hw);
+		}
 	}
 
 out:
 	/* Ungate automatic PHY configuration on non-managed 82579 */
 	if ((hw->mac.type == e1000_pch2lan) &&
 	    !(fwsm & E1000_ICH_FWSM_FW_VALID)) {
-		usleep_range(10000, 20000);
+		usleep_range(10000, 11000);
 		e1000_gate_hw_phy_config_ich8lan(hw, false);
 	}
 
@@ -459,6 +521,11 @@ static s32 e1000_init_phy_params_pchlan(struct e1000_hw *hw)
 
 	phy->id = e1000_phy_unknown;
 
+	if (hw->mac.type == e1000_pch_mtp) {
+		phy->retry_count = 2;
+		e1000e_enable_phy_retry(hw);
+	}
+
 	ret_val = e1000_init_phy_workarounds_pchlan(hw);
 	if (ret_val)
 		return ret_val;
@@ -471,11 +538,17 @@ static s32 e1000_init_phy_params_pchlan(struct e1000_hw *hw)
 				return ret_val;
 			if ((phy->id != 0) && (phy->id != PHY_REVISION_MASK))
 				break;
-			/* fall-through */
+			fallthrough;
 		case e1000_pch2lan:
 		case e1000_pch_lpt:
 		case e1000_pch_spt:
 		case e1000_pch_cnp:
+		case e1000_pch_tgp:
+		case e1000_pch_adp:
+		case e1000_pch_mtp:
+		case e1000_pch_lnp:
+		case e1000_pch_ptp:
+		case e1000_pch_nvp:
 			/* In case the PHY needs to be in mdio slow mode,
 			 * set slow mode and try to get the PHY id again.
 			 */
@@ -549,7 +622,7 @@ static s32 e1000_init_phy_params_ich8lan(struct e1000_hw *hw)
 	phy->id = 0;
 	while ((e1000_phy_unknown == e1000e_get_phy_type_from_id(phy->id)) &&
 	       (i++ < 100)) {
-		usleep_range(1000, 2000);
+		usleep_range(1000, 1100);
 		ret_val = e1000e_get_phy_id(hw);
 		if (ret_val)
 			return ret_val;
@@ -714,10 +787,16 @@ static s32 e1000_init_mac_params_ich8lan(struct e1000_hw *hw)
 	case e1000_pch2lan:
 		mac->rar_entry_count = E1000_PCH2_RAR_ENTRIES;
 		mac->ops.rar_set = e1000_rar_set_pch2lan;
-		/* fall-through */
+		fallthrough;
 	case e1000_pch_lpt:
 	case e1000_pch_spt:
 	case e1000_pch_cnp:
+	case e1000_pch_tgp:
+	case e1000_pch_adp:
+	case e1000_pch_mtp:
+	case e1000_pch_lnp:
+	case e1000_pch_ptp:
+	case e1000_pch_nvp:
 	case e1000_pchlan:
 		/* check management mode */
 		mac->ops.check_mng_mode = e1000_check_mng_mode_pchlan;
@@ -753,7 +832,7 @@ static s32 e1000_init_mac_params_ich8lan(struct e1000_hw *hw)
 /**
  *  __e1000_access_emi_reg_locked - Read/write EMI register
  *  @hw: pointer to the HW structure
- *  @addr: EMI address to program
+ *  @address: EMI address to program
  *  @data: pointer to value to read/write from/to the EMI address
  *  @read: boolean flag to indicate read or write
  *
@@ -1013,6 +1092,8 @@ static s32 e1000_platform_pm_pch_lpt(struct e1000_hw *hw, bool link)
 {
 	u32 reg = link << (E1000_LTRV_REQ_SHIFT + E1000_LTRV_NOSNOOP_SHIFT) |
 	    link << E1000_LTRV_REQ_SHIFT | E1000_LTRV_SEND;
+	u32 max_ltr_enc_d = 0;	/* maximum LTR decoded by platform */
+	u32 lat_enc_d = 0;	/* latency decoded */
 	u16 lat_enc = 0;	/* latency encoded */
 
 	if (link) {
@@ -1066,7 +1147,15 @@ static s32 e1000_platform_pm_pch_lpt(struct e1000_hw *hw, bool link)
 				     E1000_PCI_LTR_CAP_LPT + 2, &max_nosnoop);
 		max_ltr_enc = max_t(u16, max_snoop, max_nosnoop);
 
-		if (lat_enc > max_ltr_enc)
+		lat_enc_d = (lat_enc & E1000_LTRV_VALUE_MASK) *
+			     (1U << (E1000_LTRV_SCALE_FACTOR *
+			     FIELD_GET(E1000_LTRV_SCALE_MASK, lat_enc)));
+
+		max_ltr_enc_d = (max_ltr_enc & E1000_LTRV_VALUE_MASK) *
+			(1U << (E1000_LTRV_SCALE_FACTOR *
+				FIELD_GET(E1000_LTRV_SCALE_MASK, max_ltr_enc)));
+
+		if (lat_enc_d > max_ltr_enc_d)
 			lat_enc = max_ltr_enc;
 	}
 
@@ -1078,6 +1167,46 @@ static s32 e1000_platform_pm_pch_lpt(struct e1000_hw *hw, bool link)
 }
 
 /**
+ *  e1000e_force_smbus - Force interfaces to transition to SMBUS mode.
+ *  @hw: pointer to the HW structure
+ *
+ *  Force the MAC and the PHY to SMBUS mode. Assumes semaphore already
+ *  acquired.
+ *
+ * Return: 0 on success, negative errno on failure.
+ **/
+static s32 e1000e_force_smbus(struct e1000_hw *hw)
+{
+	u16 smb_ctrl = 0;
+	u32 ctrl_ext;
+	s32 ret_val;
+
+	/* Switching PHY interface always returns MDI error
+	 * so disable retry mechanism to avoid wasting time
+	 */
+	e1000e_disable_phy_retry(hw);
+
+	/* Force SMBus mode in the PHY */
+	ret_val = e1000_read_phy_reg_hv_locked(hw, CV_SMB_CTRL, &smb_ctrl);
+	if (ret_val) {
+		e1000e_enable_phy_retry(hw);
+		return ret_val;
+	}
+
+	smb_ctrl |= CV_SMB_CTRL_FORCE_SMBUS;
+	e1000_write_phy_reg_hv_locked(hw, CV_SMB_CTRL, smb_ctrl);
+
+	e1000e_enable_phy_retry(hw);
+
+	/* Force SMBus mode in the MAC */
+	ctrl_ext = er32(CTRL_EXT);
+	ctrl_ext |= E1000_CTRL_EXT_FORCE_SMBUS;
+	ew32(CTRL_EXT, ctrl_ext);
+
+	return 0;
+}
+
+/**
  *  e1000_enable_ulp_lpt_lp - configure Ultra Low Power mode for LynxPoint-LP
  *  @hw: pointer to the HW structure
  *  @to_sx: boolean indicating a system power state transition to Sx
@@ -1134,17 +1263,11 @@ s32 e1000_enable_ulp_lpt_lp(struct e1000_hw *hw, bool to_sx)
 	if (ret_val)
 		goto out;
 
-	/* Force SMBus mode in PHY */
-	ret_val = e1000_read_phy_reg_hv_locked(hw, CV_SMB_CTRL, &phy_reg);
-	if (ret_val)
+	ret_val = e1000e_force_smbus(hw);
+	if (ret_val) {
+		e_dbg("Failed to force SMBUS: %d\n", ret_val);
 		goto release;
-	phy_reg |= CV_SMB_CTRL_FORCE_SMBUS;
-	e1000_write_phy_reg_hv_locked(hw, CV_SMB_CTRL, phy_reg);
-
-	/* Force SMBus mode in MAC */
-	mac_reg = er32(CTRL_EXT);
-	mac_reg |= E1000_CTRL_EXT_FORCE_SMBUS;
-	ew32(CTRL_EXT, mac_reg);
+	}
 
 	/* Si workaround for ULP entry flow on i127/rev6 h/w.  Enable
 	 * LPLU and disable Gig speed when entering ULP
@@ -1247,6 +1370,9 @@ static s32 e1000_disable_ulp_lpt_lp(struct e1000_hw *hw, bool force)
 		return 0;
 
 	if (er32(FWSM) & E1000_ICH_FWSM_FW_VALID) {
+		struct e1000_adapter *adapter = hw->adapter;
+		bool firmware_bug = false;
+
 		if (force) {
 			/* Request ME un-configure ULP mode in the PHY */
 			mac_reg = er32(H2ME);
@@ -1255,16 +1381,26 @@ static s32 e1000_disable_ulp_lpt_lp(struct e1000_hw *hw, bool force)
 			ew32(H2ME, mac_reg);
 		}
 
-		/* Poll up to 300msec for ME to clear ULP_CFG_DONE. */
+		/* Poll up to 2.5 seconds for ME to clear ULP_CFG_DONE.
+		 * If this takes more than 1 second, show a warning indicating a
+		 * firmware bug
+		 */
 		while (er32(FWSM) & E1000_FWSM_ULP_CFG_DONE) {
-			if (i++ == 30) {
+			if (i++ == 250) {
 				ret_val = -E1000_ERR_PHY;
 				goto out;
 			}
+			if (i > 100 && !firmware_bug)
+				firmware_bug = true;
 
-			usleep_range(10000, 20000);
+			usleep_range(10000, 11000);
 		}
-		e_dbg("ULP_CONFIG_DONE cleared after %dmsec\n", i * 10);
+		if (firmware_bug)
+			e_warn("ULP_CONFIG_DONE took %d msec. This is a firmware bug\n",
+			       i * 10);
+		else
+			e_dbg("ULP_CONFIG_DONE cleared after %d msec\n",
+			      i * 10);
 
 		if (force) {
 			mac_reg = er32(H2ME);
@@ -1288,6 +1424,11 @@ static s32 e1000_disable_ulp_lpt_lp(struct e1000_hw *hw, bool force)
 		/* Toggle LANPHYPC Value bit */
 		e1000_toggle_lanphypc_pch_lpt(hw);
 
+	/* Switching PHY interface always returns MDI error
+	 * so disable retry mechanism to avoid wasting time
+	 */
+	e1000e_disable_phy_retry(hw);
+
 	/* Unforce SMBus mode in PHY */
 	ret_val = e1000_read_phy_reg_hv_locked(hw, CV_SMB_CTRL, &phy_reg);
 	if (ret_val) {
@@ -1308,6 +1449,8 @@ static s32 e1000_disable_ulp_lpt_lp(struct e1000_hw *hw, bool force)
 	phy_reg &= ~CV_SMB_CTRL_FORCE_SMBUS;
 	e1000_write_phy_reg_hv_locked(hw, CV_SMB_CTRL, phy_reg);
 
+	e1000e_enable_phy_retry(hw);
+
 	/* Unforce SMBus mode in MAC */
 	mac_reg = er32(CTRL_EXT);
 	mac_reg &= ~E1000_CTRL_EXT_FORCE_SMBUS;
@@ -1383,6 +1526,7 @@ static s32 e1000_check_for_copper_link_ich8lan(struct e1000_hw *hw)
 	 */
 	if (!mac->get_link_status)
 		return 0;
+	mac->get_link_status = false;
 
 	/* First we want to see if the MII Status Register reports
 	 * link.  If so, then we want to get the current speed/duplex
@@ -1390,12 +1534,12 @@ static s32 e1000_check_for_copper_link_ich8lan(struct e1000_hw *hw)
 	 */
 	ret_val = e1000e_phy_has_link_generic(hw, 1, 0, &link);
 	if (ret_val)
-		return ret_val;
+		goto out;
 
 	if (hw->mac.type == e1000_pchlan) {
 		ret_val = e1000_k1_gig_workaround_hv(hw, link);
 		if (ret_val)
-			return ret_val;
+			goto out;
 	}
 
 	/* When connected at 10Mbps half-duplex, some parts are excessively
@@ -1428,7 +1572,7 @@ static s32 e1000_check_for_copper_link_ich8lan(struct e1000_hw *hw)
 
 		ret_val = hw->phy.ops.acquire(hw);
 		if (ret_val)
-			return ret_val;
+			goto out;
 
 		if (hw->mac.type == e1000_pch2lan)
 			emi_addr = I82579_RX_CONFIG;
@@ -1446,11 +1590,21 @@ static s32 e1000_check_for_copper_link_ich8lan(struct e1000_hw *hw)
 			else
 				phy_reg |= 0xFA;
 			e1e_wphy_locked(hw, I217_PLL_CLOCK_GATE_REG, phy_reg);
+
+			if (speed == SPEED_1000) {
+				hw->phy.ops.read_reg_locked(hw, HV_PM_CTRL,
+							    &phy_reg);
+
+				phy_reg |= HV_PM_CTRL_K1_CLK_REQ;
+
+				hw->phy.ops.write_reg_locked(hw, HV_PM_CTRL,
+							     phy_reg);
+			}
 		}
 		hw->phy.ops.release(hw);
 
 		if (ret_val)
-			return ret_val;
+			goto out;
 
 		if (hw->mac.type >= e1000_pch_spt) {
 			u16 data;
@@ -1459,14 +1613,14 @@ static s32 e1000_check_for_copper_link_ich8lan(struct e1000_hw *hw)
 			if (speed == SPEED_1000) {
 				ret_val = hw->phy.ops.acquire(hw);
 				if (ret_val)
-					return ret_val;
+					goto out;
 
 				ret_val = e1e_rphy_locked(hw,
 							  PHY_REG(776, 20),
 							  &data);
 				if (ret_val) {
 					hw->phy.ops.release(hw);
-					return ret_val;
+					goto out;
 				}
 
 				ptr_gap = (data & (0x3FF << 2)) >> 2;
@@ -1480,18 +1634,18 @@ static s32 e1000_check_for_copper_link_ich8lan(struct e1000_hw *hw)
 				}
 				hw->phy.ops.release(hw);
 				if (ret_val)
-					return ret_val;
+					goto out;
 			} else {
 				ret_val = hw->phy.ops.acquire(hw);
 				if (ret_val)
-					return ret_val;
+					goto out;
 
 				ret_val = e1e_wphy_locked(hw,
 							  PHY_REG(776, 20),
 							  0xC023);
 				hw->phy.ops.release(hw);
 				if (ret_val)
-					return ret_val;
+					goto out;
 
 			}
 		}
@@ -1518,7 +1672,7 @@ static s32 e1000_check_for_copper_link_ich8lan(struct e1000_hw *hw)
 	    (hw->adapter->pdev->device == E1000_DEV_ID_PCH_I218_V3)) {
 		ret_val = e1000_k1_workaround_lpt_lp(hw, link);
 		if (ret_val)
-			return ret_val;
+			goto out;
 	}
 	if (hw->mac.type >= e1000_pch_lpt) {
 		/* Set platform power management values for
@@ -1526,7 +1680,7 @@ static s32 e1000_check_for_copper_link_ich8lan(struct e1000_hw *hw)
 		 */
 		ret_val = e1000_platform_pm_pch_lpt(hw, link);
 		if (ret_val)
-			return ret_val;
+			goto out;
 	}
 
 	/* Clear link partner's EEE ability */
@@ -1549,16 +1703,14 @@ static s32 e1000_check_for_copper_link_ich8lan(struct e1000_hw *hw)
 	}
 
 	if (!link)
-		return 0;	/* No link detected */
-
-	mac->get_link_status = false;
+		goto out;
 
 	switch (hw->mac.type) {
 	case e1000_pch2lan:
 		ret_val = e1000_k1_workaround_lv(hw);
 		if (ret_val)
 			return ret_val;
-		/* fall-thru */
+		fallthrough;
 	case e1000_pchlan:
 		if (hw->phy.type == e1000_phy_82578) {
 			ret_val = e1000_link_stall_workaround_hv(hw);
@@ -1617,6 +1769,10 @@ static s32 e1000_check_for_copper_link_ich8lan(struct e1000_hw *hw)
 		e_dbg("Error configuring flow control\n");
 
 	return ret_val;
+
+out:
+	mac->get_link_status = true;
+	return ret_val;
 }
 
 static s32 e1000_get_variants_ich8lan(struct e1000_adapter *adapter)
@@ -1643,6 +1799,12 @@ static s32 e1000_get_variants_ich8lan(struct e1000_adapter *adapter)
 	case e1000_pch_lpt:
 	case e1000_pch_spt:
 	case e1000_pch_cnp:
+	case e1000_pch_tgp:
+	case e1000_pch_adp:
+	case e1000_pch_mtp:
+	case e1000_pch_lnp:
+	case e1000_pch_ptp:
+	case e1000_pch_nvp:
 		rc = e1000_init_phy_params_pchlan(hw);
 		break;
 	default:
@@ -2014,7 +2176,7 @@ static s32 e1000_check_reset_block_ich8lan(struct e1000_hw *hw)
 
 	while ((blocked = !(er32(FWSM) & E1000_ICH_FWSM_RSPCIPHY)) &&
 	       (i++ < 30))
-		usleep_range(10000, 20000);
+		usleep_range(10000, 11000);
 	return blocked ? E1000_BLK_PHY_RESET : 0;
 }
 
@@ -2029,8 +2191,7 @@ static s32 e1000_write_smbus_addr(struct e1000_hw *hw)
 {
 	u16 phy_data;
 	u32 strap = er32(STRAP);
-	u32 freq = (strap & E1000_STRAP_SMT_FREQ_MASK) >>
-	    E1000_STRAP_SMT_FREQ_SHIFT;
+	u32 freq = FIELD_GET(E1000_STRAP_SMT_FREQ_MASK, strap);
 	s32 ret_val;
 
 	strap &= E1000_STRAP_SMBUS_ADDRESS_MASK;
@@ -2089,12 +2250,18 @@ static s32 e1000_sw_lcd_config_ich8lan(struct e1000_hw *hw)
 			sw_cfg_mask = E1000_FEXTNVM_SW_CONFIG;
 			break;
 		}
-		/* Fall-thru */
+		fallthrough;
 	case e1000_pchlan:
 	case e1000_pch2lan:
 	case e1000_pch_lpt:
 	case e1000_pch_spt:
 	case e1000_pch_cnp:
+	case e1000_pch_tgp:
+	case e1000_pch_adp:
+	case e1000_pch_mtp:
+	case e1000_pch_lnp:
+	case e1000_pch_ptp:
+	case e1000_pch_nvp:
 		sw_cfg_mask = E1000_FEXTNVM_SW_CONFIG_ICH8M;
 		break;
 	default:
@@ -2259,7 +2426,7 @@ release:
 /**
  *  e1000_configure_k1_ich8lan - Configure K1 power state
  *  @hw: pointer to the HW structure
- *  @enable: K1 state to configure
+ *  @k1_enable: K1 state to configure
  *
  *  Configure the K1 power state based on the provided parameter.
  *  Assumes semaphore already acquired.
@@ -2398,8 +2565,10 @@ static s32 e1000_set_mdio_slow_mode_hv(struct e1000_hw *hw)
 }
 
 /**
- *  e1000_hv_phy_workarounds_ich8lan - A series of Phy workarounds to be
- *  done after every PHY reset.
+ *  e1000_hv_phy_workarounds_ich8lan - apply PHY workarounds
+ *  @hw: pointer to the HW structure
+ *
+ *  A series of PHY workarounds to be done after every PHY reset.
  **/
 static s32 e1000_hv_phy_workarounds_ich8lan(struct e1000_hw *hw)
 {
@@ -2437,6 +2606,8 @@ static s32 e1000_hv_phy_workarounds_ich8lan(struct e1000_hw *hw)
 		if (hw->phy.revision < 2) {
 			e1000e_phy_sw_reset(hw);
 			ret_val = e1e_wphy(hw, MII_BMCR, 0x3140);
+			if (ret_val)
+				return ret_val;
 		}
 	}
 
@@ -2506,8 +2677,7 @@ void e1000_copy_rx_addrs_to_phy_ich8lan(struct e1000_hw *hw)
 		hw->phy.ops.write_reg_page(hw, BM_RAR_H(i),
 					   (u16)(mac_reg & 0xFFFF));
 		hw->phy.ops.write_reg_page(hw, BM_RAR_CTRL(i),
-					   (u16)((mac_reg & E1000_RAH_AV)
-						 >> 16));
+					   (u16)((mac_reg & E1000_RAH_AV) >> 16));
 	}
 
 	e1000_disable_phy_wakeup_reg_access_bm(hw, &phy_reg);
@@ -2685,8 +2855,10 @@ s32 e1000_lv_jumbo_workaround_ich8lan(struct e1000_hw *hw, bool enable)
 }
 
 /**
- *  e1000_lv_phy_workarounds_ich8lan - A series of Phy workarounds to be
- *  done after every PHY reset.
+ *  e1000_lv_phy_workarounds_ich8lan - apply ich8 specific workarounds
+ *  @hw: pointer to the HW structure
+ *
+ *  A series of PHY workarounds to be done after every PHY reset.
  **/
 static s32 e1000_lv_phy_workarounds_ich8lan(struct e1000_hw *hw)
 {
@@ -2716,7 +2888,7 @@ release:
 }
 
 /**
- *  e1000_k1_gig_workaround_lv - K1 Si workaround
+ *  e1000_k1_workaround_lv - K1 Si workaround
  *  @hw:   pointer to the HW structure
  *
  *  Workaround to set the K1 beacon duration for 82579 parts in 10Mbps
@@ -2831,7 +3003,7 @@ static s32 e1000_post_phy_reset_ich8lan(struct e1000_hw *hw)
 		return 0;
 
 	/* Allow time for h/w to get to quiescent state after reset */
-	usleep_range(10000, 20000);
+	usleep_range(10000, 11000);
 
 	/* Perform any necessary post-reset workarounds */
 	switch (hw->mac.type) {
@@ -2867,7 +3039,7 @@ static s32 e1000_post_phy_reset_ich8lan(struct e1000_hw *hw)
 	if (hw->mac.type == e1000_pch2lan) {
 		/* Ungate automatic PHY configuration on non-managed 82579 */
 		if (!(er32(FWSM) & E1000_ICH_FWSM_FW_VALID)) {
-			usleep_range(10000, 20000);
+			usleep_range(10000, 11000);
 			e1000_gate_hw_phy_config_ich8lan(hw, false);
 		}
 
@@ -3130,6 +3302,12 @@ static s32 e1000_valid_nvm_bank_detect_ich8lan(struct e1000_hw *hw, u32 *bank)
 	switch (hw->mac.type) {
 	case e1000_pch_spt:
 	case e1000_pch_cnp:
+	case e1000_pch_tgp:
+	case e1000_pch_adp:
+	case e1000_pch_mtp:
+	case e1000_pch_lnp:
+	case e1000_pch_ptp:
+	case e1000_pch_nvp:
 		bank1_offset = nvm->flash_bank_size;
 		act_offset = E1000_ICH_NVM_SIG_WORD;
 
@@ -3141,7 +3319,7 @@ static s32 e1000_valid_nvm_bank_detect_ich8lan(struct e1000_hw *hw, u32 *bank)
 							 &nvm_dword);
 		if (ret_val)
 			return ret_val;
-		sig_byte = (u8)((nvm_dword & 0xFF00) >> 8);
+		sig_byte = FIELD_GET(0xFF00, nvm_dword);
 		if ((sig_byte & E1000_ICH_NVM_VALID_SIG_MASK) ==
 		    E1000_ICH_NVM_SIG_VALUE) {
 			*bank = 0;
@@ -3154,7 +3332,7 @@ static s32 e1000_valid_nvm_bank_detect_ich8lan(struct e1000_hw *hw, u32 *bank)
 							 &nvm_dword);
 		if (ret_val)
 			return ret_val;
-		sig_byte = (u8)((nvm_dword & 0xFF00) >> 8);
+		sig_byte = FIELD_GET(0xFF00, nvm_dword);
 		if ((sig_byte & E1000_ICH_NVM_VALID_SIG_MASK) ==
 		    E1000_ICH_NVM_SIG_VALUE) {
 			*bank = 1;
@@ -3176,7 +3354,7 @@ static s32 e1000_valid_nvm_bank_detect_ich8lan(struct e1000_hw *hw, u32 *bank)
 			return 0;
 		}
 		e_dbg("Unable to determine valid NVM bank via EEC - reading flash signature\n");
-		/* fall-thru */
+		fallthrough;
 	default:
 		/* set bank to 0 in case flash read fails */
 		*bank = 0;
@@ -3854,13 +4032,6 @@ static s32 e1000_update_nvm_checksum_spt(struct e1000_hw *hw)
 	if (ret_val)
 		goto release;
 
-	/* And invalidate the previously valid segment by setting
-	 * its signature word (0x13) high_byte to 0b. This can be
-	 * done without an erase because flash erase sets all bits
-	 * to 1's. We can write 1's to 0's without an erase
-	 */
-	act_offset = (old_bank_offset + E1000_ICH_NVM_SIG_WORD) * 2 + 1;
-
 	/* offset in words but we read dword */
 	act_offset = old_bank_offset + E1000_ICH_NVM_SIG_WORD - 1;
 	ret_val = e1000_read_flash_dword_ich8lan(hw, act_offset, &dword);
@@ -3888,7 +4059,7 @@ release:
 	 */
 	if (!ret_val) {
 		nvm->ops.reload(hw);
-		usleep_range(10000, 20000);
+		usleep_range(10000, 11000);
 	}
 
 out:
@@ -4039,7 +4210,7 @@ release:
 	 */
 	if (!ret_val) {
 		nvm->ops.reload(hw);
-		usleep_range(10000, 20000);
+		usleep_range(10000, 11000);
 	}
 
 out:
@@ -4073,6 +4244,12 @@ static s32 e1000_validate_nvm_checksum_ich8lan(struct e1000_hw *hw)
 	case e1000_pch_lpt:
 	case e1000_pch_spt:
 	case e1000_pch_cnp:
+	case e1000_pch_tgp:
+	case e1000_pch_adp:
+	case e1000_pch_mtp:
+	case e1000_pch_lnp:
+	case e1000_pch_ptp:
+	case e1000_pch_nvp:
 		word = NVM_COMPAT;
 		valid_csum_mask = NVM_COMPAT_VALID_CSUM;
 		break;
@@ -4087,13 +4264,19 @@ static s32 e1000_validate_nvm_checksum_ich8lan(struct e1000_hw *hw)
 		return ret_val;
 
 	if (!(data & valid_csum_mask)) {
-		data |= valid_csum_mask;
-		ret_val = e1000_write_nvm(hw, word, 1, &data);
-		if (ret_val)
-			return ret_val;
-		ret_val = e1000e_update_nvm_checksum(hw);
-		if (ret_val)
-			return ret_val;
+		e_dbg("NVM Checksum valid bit not set\n");
+
+		if (hw->mac.type < e1000_pch_tgp) {
+			data |= valid_csum_mask;
+			ret_val = e1000_write_nvm(hw, word, 1, &data);
+			if (ret_val)
+				return ret_val;
+			ret_val = e1000e_update_nvm_checksum(hw);
+			if (ret_val)
+				return ret_val;
+		} else if (hw->mac.type == e1000_pch_tgp) {
+			return 0;
+		}
 	}
 
 	return e1000e_validate_nvm_checksum_generic(hw);
@@ -4611,7 +4794,7 @@ static s32 e1000_id_led_init_pchlan(struct e1000_hw *hw)
  *  @hw: pointer to the HW structure
  *
  *  ICH8 use the PCI Express bus, but does not contain a PCI Express Capability
- *  register, so the the bus width is hard coded.
+ *  register, so the bus width is hard coded.
  **/
 static s32 e1000_get_bus_info_ich8lan(struct e1000_hw *hw)
 {
@@ -4663,7 +4846,7 @@ static s32 e1000_reset_hw_ich8lan(struct e1000_hw *hw)
 	ew32(TCTL, E1000_TCTL_PSP);
 	e1e_flush();
 
-	usleep_range(10000, 20000);
+	usleep_range(10000, 11000);
 
 	/* Workaround for ICH8 bit corruption issue in FIFO memory */
 	if (hw->mac.type == e1000_ich8lan) {
@@ -4760,11 +4943,23 @@ static s32 e1000_reset_hw_ich8lan(struct e1000_hw *hw)
 static s32 e1000_init_hw_ich8lan(struct e1000_hw *hw)
 {
 	struct e1000_mac_info *mac = &hw->mac;
-	u32 ctrl_ext, txdctl, snoop;
+	u32 ctrl_ext, txdctl, snoop, fflt_dbg;
 	s32 ret_val;
 	u16 i;
 
 	e1000_initialize_hw_bits_ich8lan(hw);
+	if (hw->mac.type >= e1000_pch_mtp) {
+		ret_val = hw->phy.ops.acquire(hw);
+		if (ret_val)
+			return ret_val;
+
+		ret_val = e1000_reconfigure_k1_exit_timeout(hw);
+		hw->phy.ops.release(hw);
+		if (ret_val) {
+			e_dbg("Error failed to reconfigure K1 exit timeout\n");
+			return ret_val;
+		}
+	}
 
 	/* Initialize identification LED */
 	ret_val = mac->ops.id_led_init(hw);
@@ -4819,6 +5014,15 @@ static s32 e1000_init_hw_ich8lan(struct e1000_hw *hw)
 		snoop = (u32)~(PCIE_NO_SNOOP_ALL);
 	e1000e_set_pcie_no_snoop(hw, snoop);
 
+	/* Enable workaround for packet loss issue on TGP PCH
+	 * Do not gate DMA clock from the modPHY block
+	 */
+	if (mac->type >= e1000_pch_tgp) {
+		fflt_dbg = er32(FFLT_DBG);
+		fflt_dbg |= E1000_FFLT_DBG_DONT_GATE_WAKE_DMA_CLK;
+		ew32(FFLT_DBG, fflt_dbg);
+	}
+
 	ctrl_ext = er32(CTRL_EXT);
 	ctrl_ext |= E1000_CTRL_EXT_RO_DIS;
 	ew32(CTRL_EXT, ctrl_ext);
@@ -5192,7 +5396,7 @@ void e1000e_set_kmrn_lock_loss_workaround_ich8lan(struct e1000_hw *hw,
 }
 
 /**
- *  e1000_ipg3_phy_powerdown_workaround_ich8lan - Power down workaround on D3
+ *  e1000e_igp3_phy_powerdown_workaround_ich8lan - Power down workaround on D3
  *  @hw: pointer to the HW structure
  *
  *  Workaround for 82566 power-down on D3 entry:
@@ -5939,3 +6143,63 @@ const struct e1000_info e1000_pch_cnp_info = {
 	.phy_ops		= &ich8_phy_ops,
 	.nvm_ops		= &spt_nvm_ops,
 };
+
+const struct e1000_info e1000_pch_tgp_info = {
+	.mac			= e1000_pch_tgp,
+	.flags			= FLAG_IS_ICH
+				  | FLAG_HAS_WOL
+				  | FLAG_HAS_HW_TIMESTAMP
+				  | FLAG_HAS_CTRLEXT_ON_LOAD
+				  | FLAG_HAS_AMT
+				  | FLAG_HAS_FLASH
+				  | FLAG_HAS_JUMBO_FRAMES
+				  | FLAG_APME_IN_WUC,
+	.flags2			= FLAG2_HAS_PHY_STATS
+				  | FLAG2_HAS_EEE,
+	.pba			= 26,
+	.max_hw_frame_size	= 9022,
+	.get_variants		= e1000_get_variants_ich8lan,
+	.mac_ops		= &ich8_mac_ops,
+	.phy_ops		= &ich8_phy_ops,
+	.nvm_ops		= &spt_nvm_ops,
+};
+
+const struct e1000_info e1000_pch_adp_info = {
+	.mac			= e1000_pch_adp,
+	.flags			= FLAG_IS_ICH
+				  | FLAG_HAS_WOL
+				  | FLAG_HAS_HW_TIMESTAMP
+				  | FLAG_HAS_CTRLEXT_ON_LOAD
+				  | FLAG_HAS_AMT
+				  | FLAG_HAS_FLASH
+				  | FLAG_HAS_JUMBO_FRAMES
+				  | FLAG_APME_IN_WUC,
+	.flags2			= FLAG2_HAS_PHY_STATS
+				  | FLAG2_HAS_EEE,
+	.pba			= 26,
+	.max_hw_frame_size	= 9022,
+	.get_variants		= e1000_get_variants_ich8lan,
+	.mac_ops		= &ich8_mac_ops,
+	.phy_ops		= &ich8_phy_ops,
+	.nvm_ops		= &spt_nvm_ops,
+};
+
+const struct e1000_info e1000_pch_mtp_info = {
+	.mac			= e1000_pch_mtp,
+	.flags			= FLAG_IS_ICH
+				  | FLAG_HAS_WOL
+				  | FLAG_HAS_HW_TIMESTAMP
+				  | FLAG_HAS_CTRLEXT_ON_LOAD
+				  | FLAG_HAS_AMT
+				  | FLAG_HAS_FLASH
+				  | FLAG_HAS_JUMBO_FRAMES
+				  | FLAG_APME_IN_WUC,
+	.flags2			= FLAG2_HAS_PHY_STATS
+				  | FLAG2_HAS_EEE,
+	.pba			= 26,
+	.max_hw_frame_size	= 9022,
+	.get_variants		= e1000_get_variants_ich8lan,
+	.mac_ops		= &ich8_mac_ops,
+	.phy_ops		= &ich8_phy_ops,
+	.nvm_ops		= &spt_nvm_ops,
+};
diff --git a/drivers/net/ethernet/intel/e1000e/ich8lan.h b/drivers/net/ethernet/intel/e1000e/ich8lan.h
index 67163ca898ba..5feb589a9b5f 100644
--- a/drivers/net/ethernet/intel/e1000e/ich8lan.h
+++ b/drivers/net/ethernet/intel/e1000e/ich8lan.h
@@ -1,23 +1,5 @@
-/* Intel PRO/1000 Linux driver
- * Copyright(c) 1999 - 2015 Intel Corporation.
- *
- * This program is free software; you can redistribute it and/or modify it
- * under the terms and conditions of the GNU General Public License,
- * version 2, as published by the Free Software Foundation.
- *
- * This program is distributed in the hope it will be useful, but WITHOUT
- * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
- * FITNESS FOR A PARTICULAR PURPOSE.  See the GNU General Public License for
- * more details.
- *
- * The full GNU General Public License is included in this distribution in
- * the file called "COPYING".
- *
- * Contact Information:
- * Linux NICS <linux.nics@intel.com>
- * e1000-devel Mailing List <e1000-devel@lists.sourceforge.net>
- * Intel Corporation, 5200 N.E. Elam Young Parkway, Hillsboro, OR 97124-6497
- */
+/* SPDX-License-Identifier: GPL-2.0 */
+/* Copyright(c) 1999 - 2018 Intel Corporation. */
 
 #ifndef _E1000E_ICH8LAN_H_
 #define _E1000E_ICH8LAN_H_
@@ -59,12 +41,15 @@
 #define E1000_FWSM_WLOCK_MAC_MASK	0x0380
 #define E1000_FWSM_WLOCK_MAC_SHIFT	7
 #define E1000_FWSM_ULP_CFG_DONE		0x00000400	/* Low power cfg done */
+#define E1000_EXFWSM_DPG_EXIT_DONE	0x00000001
 
 /* Shared Receive Address Registers */
 #define E1000_SHRAL_PCH_LPT(_i)		(0x05408 + ((_i) * 8))
 #define E1000_SHRAH_PCH_LPT(_i)		(0x0540C + ((_i) * 8))
 
 #define E1000_H2ME		0x05B50	/* Host to ME */
+#define E1000_H2ME_START_DPG	0x00000001	/* indicate the ME of DPG */
+#define E1000_H2ME_EXIT_DPG	0x00000002	/* indicate the ME exit DPG */
 #define E1000_H2ME_ULP		0x00000800	/* ULP Indication Bit */
 #define E1000_H2ME_ENFORCE_SETTINGS	0x00001000	/* Enforce Settings */
 
@@ -113,7 +98,8 @@
 #define NVM_SIZE_MULTIPLIER 4096	/*multiplier for NVMS field */
 #define E1000_FLASH_BASE_ADDR 0xE000	/*offset of NVM access regs */
 #define E1000_CTRL_EXT_NVMVS 0x3	/*NVM valid sector */
-#define E1000_TARC0_CB_MULTIQ_3_REQ	(1 << 28 | 1 << 29)
+#define E1000_TARC0_CB_MULTIQ_3_REQ	0x30000000
+#define E1000_TARC0_CB_MULTIQ_2_REQ	0x20000000
 #define PCIE_ICH8_SNOOP_ALL	PCIE_NO_SNOOP_ALL
 
 #define E1000_ICH_RAR_ENTRIES	7
@@ -227,12 +213,16 @@
 
 /* PHY Power Management Control */
 #define HV_PM_CTRL		PHY_REG(770, 17)
-#define HV_PM_CTRL_PLL_STOP_IN_K1_GIGA	0x100
+#define HV_PM_CTRL_K1_CLK_REQ		0x200
 #define HV_PM_CTRL_K1_ENABLE		0x4000
 
 #define I217_PLL_CLOCK_GATE_REG	PHY_REG(772, 28)
 #define I217_PLL_CLOCK_GATE_MASK	0x07FF
 
+/* PHY Timeouts */
+#define I217_PHY_TIMEOUTS_REG                   PHY_REG(770, 21)
+#define I217_PHY_TIMEOUTS_K1_EXIT_TO_MASK       0x0FC0
+
 #define SW_FLAG_TIMEOUT		1000	/* SW Semaphore flag timeout in ms */
 
 /* Inband Control */
@@ -291,8 +281,11 @@
 
 /* Latency Tolerance Reporting */
 #define E1000_LTRV			0x000F8
+#define E1000_LTRV_VALUE_MASK		0x000003FF
 #define E1000_LTRV_SCALE_MAX		5
 #define E1000_LTRV_SCALE_FACTOR		5
+#define E1000_LTRV_SCALE_SHIFT		10
+#define E1000_LTRV_SCALE_MASK		0x00001C00
 #define E1000_LTRV_REQ_SHIFT		15
 #define E1000_LTRV_NOSNOOP_SHIFT	16
 #define E1000_LTRV_SEND			(1 << 30)
@@ -300,6 +293,9 @@
 /* Proprietary Latency Tolerance Reporting PCI Capability */
 #define E1000_PCI_LTR_CAP_LPT		0xA8
 
+/* Don't gate wake DMA clock */
+#define E1000_FFLT_DBG_DONT_GATE_WAKE_DMA_CLK	0x1000
+
 void e1000e_write_protect_nvm_ich8lan(struct e1000_hw *hw);
 void e1000e_set_kmrn_lock_loss_workaround_ich8lan(struct e1000_hw *hw,
 						  bool state);
diff --git a/drivers/net/ethernet/intel/e1000e/mac.c b/drivers/net/ethernet/intel/e1000e/mac.c
index b322011ec282..44249dd91bd6 100644
--- a/drivers/net/ethernet/intel/e1000e/mac.c
+++ b/drivers/net/ethernet/intel/e1000e/mac.c
@@ -1,23 +1,7 @@
-/* Intel PRO/1000 Linux driver
- * Copyright(c) 1999 - 2015 Intel Corporation.
- *
- * This program is free software; you can redistribute it and/or modify it
- * under the terms and conditions of the GNU General Public License,
- * version 2, as published by the Free Software Foundation.
- *
- * This program is distributed in the hope it will be useful, but WITHOUT
- * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
- * FITNESS FOR A PARTICULAR PURPOSE.  See the GNU General Public License for
- * more details.
- *
- * The full GNU General Public License is included in this distribution in
- * the file called "COPYING".
- *
- * Contact Information:
- * Linux NICS <linux.nics@intel.com>
- * e1000-devel Mailing List <e1000-devel@lists.sourceforge.net>
- * Intel Corporation, 5200 N.E. Elam Young Parkway, Hillsboro, OR 97124-6497
- */
+// SPDX-License-Identifier: GPL-2.0
+/* Copyright(c) 1999 - 2018 Intel Corporation. */
+
+#include <linux/bitfield.h>
 
 #include "e1000.h"
 
@@ -31,21 +15,17 @@
  **/
 s32 e1000e_get_bus_info_pcie(struct e1000_hw *hw)
 {
+	struct pci_dev *pdev = hw->adapter->pdev;
 	struct e1000_mac_info *mac = &hw->mac;
 	struct e1000_bus_info *bus = &hw->bus;
-	struct e1000_adapter *adapter = hw->adapter;
-	u16 pcie_link_status, cap_offset;
+	u16 pcie_link_status;
 
-	cap_offset = adapter->pdev->pcie_cap;
-	if (!cap_offset) {
+	if (!pci_pcie_cap(pdev)) {
 		bus->width = e1000_bus_width_unknown;
 	} else {
-		pci_read_config_word(adapter->pdev,
-				     cap_offset + PCIE_LINK_STATUS,
-				     &pcie_link_status);
-		bus->width = (enum e1000_bus_width)((pcie_link_status &
-						     PCIE_LINK_WIDTH_MASK) >>
-						    PCIE_LINK_WIDTH_SHIFT);
+		pcie_capability_read_word(pdev, PCI_EXP_LNKSTA, &pcie_link_status);
+		bus->width = (enum e1000_bus_width)FIELD_GET(PCI_EXP_LNKSTA_NLW,
+							     pcie_link_status);
 	}
 
 	mac->ops.set_lan_id(hw);
@@ -70,7 +50,7 @@ void e1000_set_lan_id_multi_port_pcie(struct e1000_hw *hw)
 	 * for the device regardless of function swap state.
 	 */
 	reg = er32(STATUS);
-	bus->func = (reg & E1000_STATUS_FUNC_MASK) >> E1000_STATUS_FUNC_SHIFT;
+	bus->func = FIELD_GET(E1000_STATUS_FUNC_MASK, reg);
 }
 
 /**
@@ -351,8 +331,21 @@ void e1000e_update_mc_addr_list_generic(struct e1000_hw *hw,
 	}
 
 	/* replace the entire MTA table */
-	for (i = hw->mac.mta_reg_count - 1; i >= 0; i--)
+	for (i = hw->mac.mta_reg_count - 1; i >= 0; i--) {
 		E1000_WRITE_REG_ARRAY(hw, E1000_MTA, i, hw->mac.mta_shadow[i]);
+
+		if (IS_ENABLED(CONFIG_PREEMPT_RT)) {
+			/*
+			 * Do not queue up too many posted writes to prevent
+			 * increased latency for other devices on the
+			 * interconnect. Flush after each 8th posted write,
+			 * to keep additional execution time low while still
+			 * preventing increased latency.
+			 */
+			if (!(i % 8) && i)
+				e1e_flush();
+		}
+	}
 	e1e_flush();
 }
 
@@ -424,19 +417,15 @@ s32 e1000e_check_for_copper_link(struct e1000_hw *hw)
 	 */
 	if (!mac->get_link_status)
 		return 0;
+	mac->get_link_status = false;
 
 	/* First we want to see if the MII Status Register reports
 	 * link.  If so, then we want to get the current speed/duplex
 	 * of the PHY.
 	 */
 	ret_val = e1000e_phy_has_link_generic(hw, 1, 0, &link);
-	if (ret_val)
-		return ret_val;
-
-	if (!link)
-		return 0;	/* No link detected */
-
-	mac->get_link_status = false;
+	if (ret_val || !link)
+		goto out;
 
 	/* Check if there was DownShift, must be checked
 	 * immediately after link-up
@@ -465,6 +454,10 @@ s32 e1000e_check_for_copper_link(struct e1000_hw *hw)
 		e_dbg("Error configuring flow control\n");
 
 	return ret_val;
+
+out:
+	mac->get_link_status = true;
+	return ret_val;
 }
 
 /**
@@ -815,7 +808,7 @@ static s32 e1000_poll_fiber_serdes_link_generic(struct e1000_hw *hw)
 	 * milliseconds even if the other end is doing it in SW).
 	 */
 	for (i = 0; i < FIBER_LINK_UP_LIMIT; i++) {
-		usleep_range(10000, 20000);
+		usleep_range(10000, 11000);
 		status = er32(STATUS);
 		if (status & E1000_STATUS_LU)
 			break;
@@ -975,7 +968,7 @@ s32 e1000e_force_mac_fc(struct e1000_hw *hw)
 	 *      1:  Rx flow control is enabled (we can receive pause
 	 *          frames but not send pause frames).
 	 *      2:  Tx flow control is enabled (we can send pause frames
-	 *          frames but we do not receive pause frames).
+	 *          but we do not receive pause frames).
 	 *      3:  Both Rx and Tx flow control (symmetric) is enabled.
 	 *  other:  No other values should be possible at this point.
 	 */
@@ -1381,7 +1374,7 @@ s32 e1000e_get_hw_semaphore(struct e1000_hw *hw)
 		if (!(swsm & E1000_SWSM_SMBI))
 			break;
 
-		usleep_range(50, 100);
+		udelay(100);
 		i++;
 	}
 
@@ -1399,7 +1392,7 @@ s32 e1000e_get_hw_semaphore(struct e1000_hw *hw)
 		if (er32(SWSM) & E1000_SWSM_SWESMBI)
 			break;
 
-		usleep_range(50, 100);
+		udelay(100);
 	}
 
 	if (i == timeout) {
diff --git a/drivers/net/ethernet/intel/e1000e/mac.h b/drivers/net/ethernet/intel/e1000e/mac.h
index 8284618af9ff..563176fd436e 100644
--- a/drivers/net/ethernet/intel/e1000e/mac.h
+++ b/drivers/net/ethernet/intel/e1000e/mac.h
@@ -1,23 +1,5 @@
-/* Intel PRO/1000 Linux driver
- * Copyright(c) 1999 - 2015 Intel Corporation.
- *
- * This program is free software; you can redistribute it and/or modify it
- * under the terms and conditions of the GNU General Public License,
- * version 2, as published by the Free Software Foundation.
- *
- * This program is distributed in the hope it will be useful, but WITHOUT
- * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
- * FITNESS FOR A PARTICULAR PURPOSE.  See the GNU General Public License for
- * more details.
- *
- * The full GNU General Public License is included in this distribution in
- * the file called "COPYING".
- *
- * Contact Information:
- * Linux NICS <linux.nics@intel.com>
- * e1000-devel Mailing List <e1000-devel@lists.sourceforge.net>
- * Intel Corporation, 5200 N.E. Elam Young Parkway, Hillsboro, OR 97124-6497
- */
+/* SPDX-License-Identifier: GPL-2.0 */
+/* Copyright(c) 1999 - 2018 Intel Corporation. */
 
 #ifndef _E1000E_MAC_H_
 #define _E1000E_MAC_H_
@@ -47,8 +29,6 @@ s32 e1000e_set_fc_watermarks(struct e1000_hw *hw);
 s32 e1000e_setup_fiber_serdes_link(struct e1000_hw *hw);
 s32 e1000e_setup_led_generic(struct e1000_hw *hw);
 s32 e1000e_setup_link_generic(struct e1000_hw *hw);
-s32 e1000e_validate_mdi_setting_generic(struct e1000_hw *hw);
-s32 e1000e_validate_mdi_setting_crossover_generic(struct e1000_hw *hw);
 
 void e1000e_clear_hw_cntrs_base(struct e1000_hw *hw);
 void e1000_clear_vfta_generic(struct e1000_hw *hw);
diff --git a/drivers/net/ethernet/intel/e1000e/manage.c b/drivers/net/ethernet/intel/e1000e/manage.c
index cc9b3befc2bc..c4c9b20bc51f 100644
--- a/drivers/net/ethernet/intel/e1000e/manage.c
+++ b/drivers/net/ethernet/intel/e1000e/manage.c
@@ -1,23 +1,5 @@
-/* Intel PRO/1000 Linux driver
- * Copyright(c) 1999 - 2015 Intel Corporation.
- *
- * This program is free software; you can redistribute it and/or modify it
- * under the terms and conditions of the GNU General Public License,
- * version 2, as published by the Free Software Foundation.
- *
- * This program is distributed in the hope it will be useful, but WITHOUT
- * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
- * FITNESS FOR A PARTICULAR PURPOSE.  See the GNU General Public License for
- * more details.
- *
- * The full GNU General Public License is included in this distribution in
- * the file called "COPYING".
- *
- * Contact Information:
- * Linux NICS <linux.nics@intel.com>
- * e1000-devel Mailing List <e1000-devel@lists.sourceforge.net>
- * Intel Corporation, 5200 N.E. Elam Young Parkway, Hillsboro, OR 97124-6497
- */
+// SPDX-License-Identifier: GPL-2.0
+/* Copyright(c) 1999 - 2018 Intel Corporation. */
 
 #include "e1000.h"
 
diff --git a/drivers/net/ethernet/intel/e1000e/manage.h b/drivers/net/ethernet/intel/e1000e/manage.h
index 0b9ea5952b07..d868aad806d4 100644
--- a/drivers/net/ethernet/intel/e1000e/manage.h
+++ b/drivers/net/ethernet/intel/e1000e/manage.h
@@ -1,23 +1,5 @@
-/* Intel PRO/1000 Linux driver
- * Copyright(c) 1999 - 2015 Intel Corporation.
- *
- * This program is free software; you can redistribute it and/or modify it
- * under the terms and conditions of the GNU General Public License,
- * version 2, as published by the Free Software Foundation.
- *
- * This program is distributed in the hope it will be useful, but WITHOUT
- * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
- * FITNESS FOR A PARTICULAR PURPOSE.  See the GNU General Public License for
- * more details.
- *
- * The full GNU General Public License is included in this distribution in
- * the file called "COPYING".
- *
- * Contact Information:
- * Linux NICS <linux.nics@intel.com>
- * e1000-devel Mailing List <e1000-devel@lists.sourceforge.net>
- * Intel Corporation, 5200 N.E. Elam Young Parkway, Hillsboro, OR 97124-6497
- */
+/* SPDX-License-Identifier: GPL-2.0 */
+/* Copyright(c) 1999 - 2018 Intel Corporation. */
 
 #ifndef _E1000E_MANAGE_H_
 #define _E1000E_MANAGE_H_
diff --git a/drivers/net/ethernet/intel/e1000e/netdev.c b/drivers/net/ethernet/intel/e1000e/netdev.c
index 6ed3bc419b96..201322dac233 100644
--- a/drivers/net/ethernet/intel/e1000e/netdev.c
+++ b/drivers/net/ethernet/intel/e1000e/netdev.c
@@ -1,23 +1,5 @@
-/* Intel PRO/1000 Linux driver
- * Copyright(c) 1999 - 2015 Intel Corporation.
- *
- * This program is free software; you can redistribute it and/or modify it
- * under the terms and conditions of the GNU General Public License,
- * version 2, as published by the Free Software Foundation.
- *
- * This program is distributed in the hope it will be useful, but WITHOUT
- * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
- * FITNESS FOR A PARTICULAR PURPOSE.  See the GNU General Public License for
- * more details.
- *
- * The full GNU General Public License is included in this distribution in
- * the file called "COPYING".
- *
- * Contact Information:
- * Linux NICS <linux.nics@intel.com>
- * e1000-devel Mailing List <e1000-devel@lists.sourceforge.net>
- * Intel Corporation, 5200 N.E. Elam Young Parkway, Hillsboro, OR 97124-6497
- */
+// SPDX-License-Identifier: GPL-2.0
+/* Copyright(c) 1999 - 2018 Intel Corporation. */
 
 #define pr_fmt(fmt) KBUILD_MODNAME ": " fmt
 
@@ -41,16 +23,14 @@
 #include <linux/smp.h>
 #include <linux/pm_qos.h>
 #include <linux/pm_runtime.h>
-#include <linux/aer.h>
 #include <linux/prefetch.h>
+#include <linux/suspend.h>
 
 #include "e1000.h"
+#define CREATE_TRACE_POINTS
+#include "e1000e_trace.h"
 
-#define DRV_EXTRAVERSION "-k"
-
-#define DRV_VERSION "3.2.6" DRV_EXTRAVERSION
 char e1000e_driver_name[] = "e1000e";
-const char e1000e_driver_version[] = DRV_VERSION;
 
 #define DEFAULT_MSG_ENABLE (NETIF_MSG_DRV|NETIF_MSG_PROBE|NETIF_MSG_LINK)
 static int debug = -1;
@@ -72,6 +52,9 @@ static const struct e1000_info *e1000_info_tbl[] = {
 	[board_pch_lpt]		= &e1000_pch_lpt_info,
 	[board_pch_spt]		= &e1000_pch_spt_info,
 	[board_pch_cnp]		= &e1000_pch_cnp_info,
+	[board_pch_tgp]		= &e1000_pch_tgp_info,
+	[board_pch_adp]		= &e1000_pch_adp_info,
+	[board_pch_mtp]		= &e1000_pch_mtp_info,
 };
 
 struct e1000_reg_info {
@@ -137,14 +120,12 @@ static const struct e1000_reg_info e1000_reg_info_tbl[] = {
  * has bit 24 set while ME is accessing MAC CSR registers, wait if it is set
  * and try again a number of times.
  **/
-s32 __ew32_prepare(struct e1000_hw *hw)
+static void __ew32_prepare(struct e1000_hw *hw)
 {
 	s32 i = E1000_ICH_FWSM_PCIM2PCI_COUNT;
 
 	while ((er32(FWSM) & E1000_ICH_FWSM_PCIM2PCI) && --i)
 		udelay(50);
-
-	return i;
 }
 
 void __ew32(struct e1000_hw *hw, unsigned long reg, u32 val)
@@ -486,6 +467,7 @@ rx_ring_summary:
 
 /**
  * e1000_desc_unused - calculate if we have unused descriptors
+ * @ring: pointer to ring struct to perform calculation on
  **/
 static int e1000_desc_unused(struct e1000_ring *ring)
 {
@@ -562,6 +544,7 @@ static void e1000e_rx_hwtstamp(struct e1000_adapter *adapter, u32 status,
 /**
  * e1000_receive_skb - helper function to handle Rx indications
  * @adapter: board private structure
+ * @netdev: pointer to netdev struct
  * @staterr: descriptor extended error and status field as written by hardware
  * @vlan: descriptor vlan field as written by hardware (no le/be conversion)
  * @skb: pointer to sk_buff to be indicated to stack
@@ -586,8 +569,7 @@ static void e1000_receive_skb(struct e1000_adapter *adapter,
  * e1000_rx_checksum - Receive Checksum Offload
  * @adapter: board private structure
  * @status_err: receive descriptor status and error fields
- * @csum: receive descriptor csum field
- * @sk_buff: socket buffer with received data
+ * @skb: socket buffer with received data
  **/
 static void e1000_rx_checksum(struct e1000_adapter *adapter, u32 status_err,
 			      struct sk_buff *skb)
@@ -625,11 +607,11 @@ static void e1000e_update_rdt_wa(struct e1000_ring *rx_ring, unsigned int i)
 {
 	struct e1000_adapter *adapter = rx_ring->adapter;
 	struct e1000_hw *hw = &adapter->hw;
-	s32 ret_val = __ew32_prepare(hw);
 
+	__ew32_prepare(hw);
 	writel(i, rx_ring->tail);
 
-	if (unlikely(!ret_val && (i != readl(rx_ring->tail)))) {
+	if (unlikely(i != readl(rx_ring->tail))) {
 		u32 rctl = er32(RCTL);
 
 		ew32(RCTL, rctl & ~E1000_RCTL_EN);
@@ -642,11 +624,11 @@ static void e1000e_update_tdt_wa(struct e1000_ring *tx_ring, unsigned int i)
 {
 	struct e1000_adapter *adapter = tx_ring->adapter;
 	struct e1000_hw *hw = &adapter->hw;
-	s32 ret_val = __ew32_prepare(hw);
 
+	__ew32_prepare(hw);
 	writel(i, tx_ring->tail);
 
-	if (unlikely(!ret_val && (i != readl(tx_ring->tail)))) {
+	if (unlikely(i != readl(tx_ring->tail))) {
 		u32 tctl = er32(TCTL);
 
 		ew32(TCTL, tctl & ~E1000_TCTL_EN);
@@ -658,6 +640,8 @@ static void e1000e_update_tdt_wa(struct e1000_ring *tx_ring, unsigned int i)
 /**
  * e1000_alloc_rx_buffers - Replace used receive buffers
  * @rx_ring: Rx descriptor ring
+ * @cleaned_count: number to reallocate
+ * @gfp: flags for allocation
  **/
 static void e1000_alloc_rx_buffers(struct e1000_ring *rx_ring,
 				   int cleaned_count, gfp_t gfp)
@@ -726,6 +710,8 @@ map_skb:
 /**
  * e1000_alloc_rx_buffers_ps - Replace used receive buffers; packet split
  * @rx_ring: Rx descriptor ring
+ * @cleaned_count: number to reallocate
+ * @gfp: flags for allocation
  **/
 static void e1000_alloc_rx_buffers_ps(struct e1000_ring *rx_ring,
 				      int cleaned_count, gfp_t gfp)
@@ -829,6 +815,7 @@ no_buffers:
  * e1000_alloc_jumbo_rx_buffers - Replace used jumbo receive buffers
  * @rx_ring: Rx descriptor ring
  * @cleaned_count: number of buffers to allocate this pass
+ * @gfp: flags for allocation
  **/
 
 static void e1000_alloc_jumbo_rx_buffers(struct e1000_ring *rx_ring,
@@ -918,6 +905,8 @@ static inline void e1000_rx_hash(struct net_device *netdev, __le32 rss,
 /**
  * e1000_clean_rx_irq - Send received data up the network stack
  * @rx_ring: Rx descriptor ring
+ * @work_done: output parameter for indicating completed work
+ * @work_to_do: how many packets we can clean
  *
  * the return value indicates whether actual cleaning was done, there
  * is no guarantee that everything was cleaned
@@ -1071,7 +1060,8 @@ next_desc:
 }
 
 static void e1000_put_txbuf(struct e1000_ring *tx_ring,
-			    struct e1000_buffer *buffer_info)
+			    struct e1000_buffer *buffer_info,
+			    bool drop)
 {
 	struct e1000_adapter *adapter = tx_ring->adapter;
 
@@ -1085,7 +1075,10 @@ static void e1000_put_txbuf(struct e1000_ring *tx_ring,
 		buffer_info->dma = 0;
 	}
 	if (buffer_info->skb) {
-		dev_kfree_skb_any(buffer_info->skb);
+		if (drop)
+			dev_kfree_skb_any(buffer_info->skb);
+		else
+			dev_consume_skb_any(buffer_info->skb);
 		buffer_info->skb = NULL;
 	}
 	buffer_info->time_stamp = 0;
@@ -1183,6 +1176,7 @@ static void e1000e_tx_hwtstamp_work(struct work_struct *work)
 	struct e1000_hw *hw = &adapter->hw;
 
 	if (er32(TSYNCTXCTL) & E1000_TSYNCTXCTL_VALID) {
+		struct sk_buff *skb = adapter->tx_hwtstamp_skb;
 		struct skb_shared_hwtstamps shhwtstamps;
 		u64 txstmp;
 
@@ -1191,9 +1185,14 @@ static void e1000e_tx_hwtstamp_work(struct work_struct *work)
 
 		e1000e_systim_to_hwtstamp(adapter, &shhwtstamps, txstmp);
 
-		skb_tstamp_tx(adapter->tx_hwtstamp_skb, &shhwtstamps);
-		dev_kfree_skb_any(adapter->tx_hwtstamp_skb);
+		/* Clear the global tx_hwtstamp_skb pointer and force writes
+		 * prior to notifying the stack of a Tx timestamp.
+		 */
 		adapter->tx_hwtstamp_skb = NULL;
+		wmb(); /* force write prior to skb_tstamp_tx */
+
+		skb_tstamp_tx(skb, &shhwtstamps);
+		dev_consume_skb_any(skb);
 	} else if (time_after(jiffies, adapter->tx_hwtstamp_start
 			      + adapter->tx_timeout_factor * HZ)) {
 		dev_kfree_skb_any(adapter->tx_hwtstamp_skb);
@@ -1248,7 +1247,7 @@ static bool e1000_clean_tx_irq(struct e1000_ring *tx_ring)
 				}
 			}
 
-			e1000_put_txbuf(tx_ring, buffer_info);
+			e1000_put_txbuf(tx_ring, buffer_info, false);
 			tx_desc->upper.data = 0;
 
 			i++;
@@ -1302,6 +1301,8 @@ static bool e1000_clean_tx_irq(struct e1000_ring *tx_ring)
 /**
  * e1000_clean_rx_irq_ps - Send received data up the network stack; packet split
  * @rx_ring: Rx descriptor ring
+ * @work_done: output parameter for indicating completed work
+ * @work_to_do: how many packets we can clean
  *
  * the return value indicates whether actual cleaning was done, there
  * is no guarantee that everything was cleaned
@@ -1389,26 +1390,18 @@ static bool e1000_clean_rx_irq_ps(struct e1000_ring *rx_ring, int *work_done,
 
 			/* page alloc/put takes too long and effects small
 			 * packet throughput, so unsplit small packets and
-			 * save the alloc/put only valid in softirq (napi)
-			 * context to call kmap_*
+			 * save the alloc/put
 			 */
 			if (l1 && (l1 <= copybreak) &&
 			    ((length + l1) <= adapter->rx_ps_bsize0)) {
-				u8 *vaddr;
-
 				ps_page = &buffer_info->ps_pages[0];
 
-				/* there is no documentation about how to call
-				 * kmap_atomic, so we can't hold the mapping
-				 * very long
-				 */
 				dma_sync_single_for_cpu(&pdev->dev,
 							ps_page->dma,
 							PAGE_SIZE,
 							DMA_FROM_DEVICE);
-				vaddr = kmap_atomic(ps_page->page);
-				memcpy(skb_tail_pointer(skb), vaddr, l1);
-				kunmap_atomic(vaddr);
+				memcpy(skb_tail_pointer(skb),
+				       page_address(ps_page->page), l1);
 				dma_sync_single_for_device(&pdev->dev,
 							   ps_page->dma,
 							   PAGE_SIZE,
@@ -1492,9 +1485,6 @@ next_desc:
 	return cleaned;
 }
 
-/**
- * e1000_consume_page - helper function
- **/
 static void e1000_consume_page(struct e1000_buffer *bi, struct sk_buff *skb,
 			       u16 length)
 {
@@ -1506,7 +1496,9 @@ static void e1000_consume_page(struct e1000_buffer *bi, struct sk_buff *skb,
 
 /**
  * e1000_clean_jumbo_rx_irq - Send received data up the network stack; legacy
- * @adapter: board private structure
+ * @rx_ring: Rx descriptor ring
+ * @work_done: output parameter for indicating completed work
+ * @work_to_do: how many packets we can clean
  *
  * the return value indicates whether actual cleaning was done, there
  * is no guarantee that everything was cleaned
@@ -1609,11 +1601,9 @@ static bool e1000_clean_jumbo_rx_irq(struct e1000_ring *rx_ring, int *work_done,
 				 */
 				if (length <= copybreak &&
 				    skb_tailroom(skb) >= length) {
-					u8 *vaddr;
-					vaddr = kmap_atomic(buffer_info->page);
-					memcpy(skb_tail_pointer(skb), vaddr,
+					memcpy(skb_tail_pointer(skb),
+					       page_address(buffer_info->page),
 					       length);
-					kunmap_atomic(vaddr);
 					/* re-use the page, so don't erase
 					 * buffer_info->page
 					 */
@@ -1798,8 +1788,7 @@ static irqreturn_t e1000_intr_msi(int __always_unused irq, void *data)
 		adapter->corr_errors +=
 		    pbeccsts & E1000_PBECCSTS_CORR_ERR_CNT_MASK;
 		adapter->uncorr_errors +=
-		    (pbeccsts & E1000_PBECCSTS_UNCORR_ERR_CNT_MASK) >>
-		    E1000_PBECCSTS_UNCORR_ERR_CNT_SHIFT;
+		    FIELD_GET(E1000_PBECCSTS_UNCORR_ERR_CNT_MASK, pbeccsts);
 
 		/* Do the reset outside of interrupt context */
 		schedule_work(&adapter->reset_task);
@@ -1878,8 +1867,7 @@ static irqreturn_t e1000_intr(int __always_unused irq, void *data)
 		adapter->corr_errors +=
 		    pbeccsts & E1000_PBECCSTS_CORR_ERR_CNT_MASK;
 		adapter->uncorr_errors +=
-		    (pbeccsts & E1000_PBECCSTS_UNCORR_ERR_CNT_MASK) >>
-		    E1000_PBECCSTS_UNCORR_ERR_CNT_SHIFT;
+		    FIELD_GET(E1000_PBECCSTS_UNCORR_ERR_CNT_MASK, pbeccsts);
 
 		/* Do the reset outside of interrupt context */
 		schedule_work(&adapter->reset_task);
@@ -1904,15 +1892,21 @@ static irqreturn_t e1000_msix_other(int __always_unused irq, void *data)
 	struct net_device *netdev = data;
 	struct e1000_adapter *adapter = netdev_priv(netdev);
 	struct e1000_hw *hw = &adapter->hw;
+	u32 icr = er32(ICR);
 
-	hw->mac.get_link_status = true;
+	if (icr & adapter->eiac_mask)
+		ew32(ICS, (icr & adapter->eiac_mask));
 
-	/* guard against interrupt when we're going down */
-	if (!test_bit(__E1000_DOWN, &adapter->state)) {
-		mod_timer(&adapter->watchdog_timer, jiffies + 1);
-		ew32(IMS, E1000_IMS_OTHER);
+	if (icr & E1000_ICR_LSC) {
+		hw->mac.get_link_status = true;
+		/* guard against interrupt when we're going down */
+		if (!test_bit(__E1000_DOWN, &adapter->state))
+			mod_timer(&adapter->watchdog_timer, jiffies + 1);
 	}
 
+	if (!test_bit(__E1000_DOWN, &adapter->state))
+		ew32(IMS, E1000_IMS_OTHER | IMS_OTHER_MASK);
+
 	return IRQ_HANDLED;
 }
 
@@ -1963,6 +1957,7 @@ static irqreturn_t e1000_intr_msix_rx(int __always_unused irq, void *data)
 
 /**
  * e1000_configure_msix - Configure MSI-X hardware
+ * @adapter: board private structure
  *
  * e1000_configure_msix sets up the hardware to properly
  * generate MSI-X interrupts.
@@ -2014,7 +2009,6 @@ static void e1000_configure_msix(struct e1000_adapter *adapter)
 		       hw->hw_addr + E1000_EITR_82574(vector));
 	else
 		writel(1, hw->hw_addr + E1000_EITR_82574(vector));
-	adapter->eiac_mask |= E1000_IMS_OTHER;
 
 	/* Cause Tx interrupts on every write back */
 	ivar |= BIT(31);
@@ -2042,6 +2036,7 @@ void e1000e_reset_interrupt_capability(struct e1000_adapter *adapter)
 
 /**
  * e1000e_set_interrupt_capability - set MSI or MSI-X if supported
+ * @adapter: board private structure
  *
  * Attempt to configure interrupts using the best available
  * capabilities of the hardware and kernel.
@@ -2077,7 +2072,7 @@ void e1000e_set_interrupt_capability(struct e1000_adapter *adapter)
 			e1000e_reset_interrupt_capability(adapter);
 		}
 		adapter->int_mode = E1000E_INT_MODE_MSI;
-		/* Fall through */
+		fallthrough;
 	case E1000E_INT_MODE_MSI:
 		if (!pci_enable_msi(adapter->pdev)) {
 			adapter->flags |= FLAG_MSI_ENABLED;
@@ -2085,7 +2080,7 @@ void e1000e_set_interrupt_capability(struct e1000_adapter *adapter)
 			adapter->int_mode = E1000E_INT_MODE_LEGACY;
 			e_err("Failed to initialize MSI interrupts.  Falling back to legacy interrupts.\n");
 		}
-		/* Fall through */
+		fallthrough;
 	case E1000E_INT_MODE_LEGACY:
 		/* Don't do anything; this is the system default */
 		break;
@@ -2097,6 +2092,7 @@ void e1000e_set_interrupt_capability(struct e1000_adapter *adapter)
 
 /**
  * e1000_request_msix - Initialize MSI-X interrupts
+ * @adapter: board private structure
  *
  * e1000_request_msix allocates MSI-X vectors and requests interrupts from the
  * kernel.
@@ -2109,7 +2105,7 @@ static int e1000_request_msix(struct e1000_adapter *adapter)
 	if (strlen(netdev->name) < (IFNAMSIZ - 5))
 		snprintf(adapter->rx_ring->name,
 			 sizeof(adapter->rx_ring->name) - 1,
-			 "%s-rx-0", netdev->name);
+			 "%.14s-rx-0", netdev->name);
 	else
 		memcpy(adapter->rx_ring->name, netdev->name, IFNAMSIZ);
 	err = request_irq(adapter->msix_entries[vector].vector,
@@ -2125,7 +2121,7 @@ static int e1000_request_msix(struct e1000_adapter *adapter)
 	if (strlen(netdev->name) < (IFNAMSIZ - 5))
 		snprintf(adapter->tx_ring->name,
 			 sizeof(adapter->tx_ring->name) - 1,
-			 "%s-tx-0", netdev->name);
+			 "%.14s-tx-0", netdev->name);
 	else
 		memcpy(adapter->tx_ring->name, netdev->name, IFNAMSIZ);
 	err = request_irq(adapter->msix_entries[vector].vector,
@@ -2150,6 +2146,7 @@ static int e1000_request_msix(struct e1000_adapter *adapter)
 
 /**
  * e1000_request_irq - initialize interrupts
+ * @adapter: board private structure
  *
  * Attempts to configure interrupts using the best available
  * capabilities of the hardware and kernel.
@@ -2210,6 +2207,7 @@ static void e1000_free_irq(struct e1000_adapter *adapter)
 
 /**
  * e1000_irq_disable - Mask off interrupt generation on the NIC
+ * @adapter: board private structure
  **/
 static void e1000_irq_disable(struct e1000_adapter *adapter)
 {
@@ -2232,6 +2230,7 @@ static void e1000_irq_disable(struct e1000_adapter *adapter)
 
 /**
  * e1000_irq_enable - Enable default interrupt generation settings
+ * @adapter: board private structure
  **/
 static void e1000_irq_enable(struct e1000_adapter *adapter)
 {
@@ -2239,7 +2238,8 @@ static void e1000_irq_enable(struct e1000_adapter *adapter)
 
 	if (adapter->msix_entries) {
 		ew32(EIAC_82574, adapter->eiac_mask & E1000_EIAC_MASK_82574);
-		ew32(IMS, adapter->eiac_mask | E1000_IMS_LSC);
+		ew32(IMS, adapter->eiac_mask | E1000_IMS_OTHER |
+		     IMS_OTHER_MASK);
 	} else if (hw->mac.type >= e1000_pch_lpt) {
 		ew32(IMS, IMS_ENABLE_MASK | E1000_IMS_ECCER);
 	} else {
@@ -2301,6 +2301,8 @@ void e1000e_release_hw_control(struct e1000_adapter *adapter)
 
 /**
  * e1000_alloc_ring_dma - allocate memory for a ring structure
+ * @adapter: board private structure
+ * @ring: ring struct for which to allocate dma
  **/
 static int e1000_alloc_ring_dma(struct e1000_adapter *adapter,
 				struct e1000_ring *ring)
@@ -2415,7 +2417,7 @@ static void e1000_clean_tx_ring(struct e1000_ring *tx_ring)
 
 	for (i = 0; i < tx_ring->count; i++) {
 		buffer_info = &tx_ring->buffer_info[i];
-		e1000_put_txbuf(tx_ring, buffer_info);
+		e1000_put_txbuf(tx_ring, buffer_info, false);
 	}
 
 	netdev_reset_queue(adapter->netdev);
@@ -2476,7 +2478,6 @@ void e1000e_free_rx_resources(struct e1000_ring *rx_ring)
 
 /**
  * e1000_update_itr - update the dynamic ITR value based on statistics
- * @adapter: pointer to adapter
  * @itr_setting: current adapter->itr
  * @packets: the number of packets during this measurement interval
  * @bytes: the number of bytes during this measurement interval
@@ -2540,7 +2541,6 @@ static void e1000_set_itr(struct e1000_adapter *adapter)
 
 	/* for non-gigabit speeds, just fix the interrupt rate at 4000 */
 	if (adapter->link_speed != SPEED_1000) {
-		current_itr = 0;
 		new_itr = 4000;
 		goto set_itr_now;
 	}
@@ -2653,9 +2653,9 @@ err:
 /**
  * e1000e_poll - NAPI Rx polling callback
  * @napi: struct associated with this polling callback
- * @weight: number of packets driver is allowed to process this poll
+ * @budget: number of packets driver is allowed to process this poll
  **/
-static int e1000e_poll(struct napi_struct *napi, int weight)
+static int e1000e_poll(struct napi_struct *napi, int budget)
 {
 	struct e1000_adapter *adapter = container_of(napi, struct e1000_adapter,
 						     napi);
@@ -2669,16 +2669,17 @@ static int e1000e_poll(struct napi_struct *napi, int weight)
 	    (adapter->rx_ring->ims_val & adapter->tx_ring->ims_val))
 		tx_cleaned = e1000_clean_tx_irq(adapter->tx_ring);
 
-	adapter->clean_rx(adapter->rx_ring, &work_done, weight);
+	adapter->clean_rx(adapter->rx_ring, &work_done, budget);
 
-	if (!tx_cleaned)
-		work_done = weight;
+	if (!tx_cleaned || work_done == budget)
+		return budget;
 
-	/* If weight not fully consumed, exit the polling mode */
-	if (work_done < weight) {
+	/* Exit the polling mode, but don't re-enable interrupts if stack might
+	 * poll us due to busy-polling
+	 */
+	if (likely(napi_complete_done(napi, work_done))) {
 		if (adapter->itr_setting & 3)
 			e1000_set_itr(adapter);
-		napi_complete_done(napi, work_done);
 		if (!test_bit(__E1000_DOWN, &adapter->state)) {
 			if (adapter->msix_entries)
 				ew32(IMS, adapter->rx_ring->ims_val);
@@ -2760,7 +2761,7 @@ static void e1000e_vlan_filter_disable(struct e1000_adapter *adapter)
 		rctl &= ~(E1000_RCTL_VFE | E1000_RCTL_CFIEN);
 		ew32(RCTL, rctl);
 
-		if (adapter->mng_vlan_id != (u16)E1000_MNG_VLAN_NONE) {
+		if (adapter->mng_vlan_id != E1000_MNG_VLAN_NONE) {
 			e1000_vlan_rx_kill_vid(netdev, htons(ETH_P_8021Q),
 					       adapter->mng_vlan_id);
 			adapter->mng_vlan_id = E1000_MNG_VLAN_NONE;
@@ -2827,7 +2828,7 @@ static void e1000_update_mng_vlan(struct e1000_adapter *adapter)
 		adapter->mng_vlan_id = vid;
 	}
 
-	if ((old_vid != (u16)E1000_MNG_VLAN_NONE) && (vid != old_vid))
+	if (old_vid != E1000_MNG_VLAN_NONE && vid != old_vid)
 		e1000_vlan_rx_kill_vid(netdev, htons(ETH_P_8021Q), old_vid);
 }
 
@@ -2927,11 +2928,8 @@ static void e1000_configure_tx(struct e1000_adapter *adapter)
 	tx_ring->head = adapter->hw.hw_addr + E1000_TDH(0);
 	tx_ring->tail = adapter->hw.hw_addr + E1000_TDT(0);
 
-	writel(0, tx_ring->head);
 	if (adapter->flags2 & FLAG2_PCIM2PCI_ARBITER_WA)
 		e1000e_update_tdt_wa(tx_ring, 0);
-	else
-		writel(0, tx_ring->tail);
 
 	/* Set the Tx Interrupt Delay register */
 	ew32(TIDV, adapter->tx_int_delay);
@@ -2998,8 +2996,8 @@ static void e1000_configure_tx(struct e1000_adapter *adapter)
 
 	hw->mac.ops.config_collision_dist(hw);
 
-	/* SPT and CNP Si errata workaround to avoid data corruption */
-	if (hw->mac.type >= e1000_pch_spt) {
+	/* SPT and KBL Si errata workaround to avoid data corruption */
+	if (hw->mac.type == e1000_pch_spt) {
 		u32 reg_val;
 
 		reg_val = er32(IOSFPC);
@@ -3007,17 +3005,23 @@ static void e1000_configure_tx(struct e1000_adapter *adapter)
 		ew32(IOSFPC, reg_val);
 
 		reg_val = er32(TARC(0));
-		reg_val |= E1000_TARC0_CB_MULTIQ_3_REQ;
+		/* SPT and KBL Si errata workaround to avoid Tx hang.
+		 * Dropping the number of outstanding requests from
+		 * 3 to 2 in order to avoid a buffer overrun.
+		 */
+		reg_val &= ~E1000_TARC0_CB_MULTIQ_3_REQ;
+		reg_val |= E1000_TARC0_CB_MULTIQ_2_REQ;
 		ew32(TARC(0), reg_val);
 	}
 }
 
+#define PAGE_USE_COUNT(S) (((S) >> PAGE_SHIFT) + \
+			   (((S) & (PAGE_SIZE - 1)) ? 1 : 0))
+
 /**
  * e1000_setup_rctl - configure the receive control registers
  * @adapter: Board private structure
  **/
-#define PAGE_USE_COUNT(S) (((S) >> PAGE_SHIFT) + \
-			   (((S) & (PAGE_SIZE - 1)) ? 1 : 0))
 static void e1000_setup_rctl(struct e1000_adapter *adapter)
 {
 	struct e1000_hw *hw = &adapter->hw;
@@ -3136,10 +3140,10 @@ static void e1000_setup_rctl(struct e1000_adapter *adapter)
 		switch (adapter->rx_ps_pages) {
 		case 3:
 			psrctl |= PAGE_SIZE << E1000_PSRCTL_BSIZE3_SHIFT;
-			/* fall-through */
+			fallthrough;
 		case 2:
 			psrctl |= PAGE_SIZE << E1000_PSRCTL_BSIZE2_SHIFT;
-			/* fall-through */
+			fallthrough;
 		case 1:
 			psrctl |= PAGE_SIZE >> E1000_PSRCTL_BSIZE1_SHIFT;
 			break;
@@ -3204,7 +3208,7 @@ static void e1000_configure_rx(struct e1000_adapter *adapter)
 	if (!(adapter->flags2 & FLAG2_NO_DISABLE_RX))
 		ew32(RCTL, rctl & ~E1000_RCTL_EN);
 	e1e_flush();
-	usleep_range(10000, 20000);
+	usleep_range(10000, 11000);
 
 	if (adapter->flags2 & FLAG2_DMA_BURST) {
 		/* set the writeback threshold (only takes effect if the RDTR
@@ -3217,14 +3221,6 @@ static void e1000_configure_rx(struct e1000_adapter *adapter)
 		 */
 		ew32(RXDCTL(0), E1000_RXDCTL_DMA_BURST_ENABLE);
 		ew32(RXDCTL(1), E1000_RXDCTL_DMA_BURST_ENABLE);
-
-		/* override the delay timers for enabling bursting, only if
-		 * the value was not set by the user via module options
-		 */
-		if (adapter->rx_int_delay == DEFAULT_RDTR)
-			adapter->rx_int_delay = BURST_RDTR;
-		if (adapter->rx_abs_int_delay == DEFAULT_RADV)
-			adapter->rx_abs_int_delay = BURST_RADV;
 	}
 
 	/* set the Receive Delay Timer Register */
@@ -3254,11 +3250,8 @@ static void e1000_configure_rx(struct e1000_adapter *adapter)
 	rx_ring->head = adapter->hw.hw_addr + E1000_RDH(0);
 	rx_ring->tail = adapter->hw.hw_addr + E1000_RDT(0);
 
-	writel(0, rx_ring->head);
 	if (adapter->flags2 & FLAG2_PCIM2PCI_ARBITER_WA)
 		e1000e_update_rdt_wa(rx_ring, 0);
-	else
-		writel(0, rx_ring->tail);
 
 	/* Enable Receive Checksum Offload for TCP and UDP */
 	rxcsum = er32(RXCSUM);
@@ -3279,13 +3272,15 @@ static void e1000_configure_rx(struct e1000_adapter *adapter)
 		if (adapter->flags & FLAG_IS_ICH) {
 			u32 rxdctl = er32(RXDCTL(0));
 
-			ew32(RXDCTL(0), rxdctl | 0x3);
+			ew32(RXDCTL(0), rxdctl | 0x3 | BIT(8));
 		}
 
-		pm_qos_update_request(&adapter->pm_qos_req, lat);
+		dev_info(&adapter->pdev->dev,
+			 "Some CPU C-states have been disabled in order to enable jumbo frames\n");
+		cpu_latency_qos_update_request(&adapter->pm_qos_req, lat);
 	} else {
-		pm_qos_update_request(&adapter->pm_qos_req,
-				      PM_QOS_DEFAULT_VALUE);
+		cpu_latency_qos_update_request(&adapter->pm_qos_req,
+					       PM_QOS_DEFAULT_VALUE);
 	}
 
 	/* Enable Receives */
@@ -3315,7 +3310,7 @@ static int e1000e_write_mc_addr_list(struct net_device *netdev)
 		return 0;
 	}
 
-	mta_list = kzalloc(netdev_mc_count(netdev) * ETH_ALEN, GFP_ATOMIC);
+	mta_list = kcalloc(netdev_mc_count(netdev), ETH_ALEN, GFP_ATOMIC);
 	if (!mta_list)
 		return -ENOMEM;
 
@@ -3530,16 +3525,16 @@ s32 e1000e_get_base_timinca(struct e1000_adapter *adapter, u32 *timinca)
 		}
 		break;
 	case e1000_pch_spt:
-		if (er32(TSYNCRXCTL) & E1000_TSYNCRXCTL_SYSCFI) {
-			/* Stable 24MHz frequency */
-			incperiod = INCPERIOD_24MHZ;
-			incvalue = INCVALUE_24MHZ;
-			shift = INCVALUE_SHIFT_24MHZ;
-			adapter->cc.shift = shift;
-			break;
-		}
-		return -EINVAL;
+		/* Stable 24MHz frequency */
+		incperiod = INCPERIOD_24MHZ;
+		incvalue = INCVALUE_24MHZ;
+		shift = INCVALUE_SHIFT_24MHZ;
+		adapter->cc.shift = shift;
+		break;
 	case e1000_pch_cnp:
+	case e1000_pch_tgp:
+	case e1000_pch_adp:
+	case e1000_pch_nvp:
 		if (er32(TSYNCRXCTL) & E1000_TSYNCRXCTL_SYSCFI) {
 			/* Stable 24MHz frequency */
 			incperiod = INCPERIOD_24MHZ;
@@ -3554,6 +3549,17 @@ s32 e1000e_get_base_timinca(struct e1000_adapter *adapter, u32 *timinca)
 			adapter->cc.shift = shift;
 		}
 		break;
+	case e1000_pch_mtp:
+	case e1000_pch_lnp:
+	case e1000_pch_ptp:
+		/* System firmware can misreport this value, so set it to a
+		 * stable 38400KHz frequency.
+		 */
+		incperiod = INCPERIOD_38400KHZ;
+		incvalue = INCVALUE_38400KHZ;
+		shift = INCVALUE_SHIFT_38400KHZ;
+		adapter->cc.shift = shift;
+		break;
 	case e1000_82574:
 	case e1000_82583:
 		/* Stable 25MHz frequency */
@@ -3575,6 +3581,8 @@ s32 e1000e_get_base_timinca(struct e1000_adapter *adapter, u32 *timinca)
 /**
  * e1000e_config_hwtstamp - configure the hwtstamp registers and enable/disable
  * @adapter: board private structure
+ * @config: timestamp configuration
+ * @extack: netlink extended ACK for error report
  *
  * Outgoing time stamping can be enabled and disabled. Play nice and
  * disable it when requested, although it shouldn't cause any overhead
@@ -3588,7 +3596,8 @@ s32 e1000e_get_base_timinca(struct e1000_adapter *adapter, u32 *timinca)
  * exception of "all V2 events regardless of level 2 or 4".
  **/
 static int e1000e_config_hwtstamp(struct e1000_adapter *adapter,
-				  struct hwtstamp_config *config)
+				  struct kernel_hwtstamp_config *config,
+				  struct netlink_ext_ack *extack)
 {
 	struct e1000_hw *hw = &adapter->hw;
 	u32 tsync_tx_ctl = E1000_TSYNCTXCTL_ENABLED;
@@ -3599,12 +3608,10 @@ static int e1000e_config_hwtstamp(struct e1000_adapter *adapter,
 	bool is_l2 = false;
 	u32 regval;
 
-	if (!(adapter->flags & FLAG_HAS_HW_TIMESTAMP))
-		return -EINVAL;
-
-	/* flags reserved for future extensions - must be zero */
-	if (config->flags)
+	if (!(adapter->flags & FLAG_HAS_HW_TIMESTAMP)) {
+		NL_SET_ERR_MSG(extack, "No HW timestamp support");
 		return -EINVAL;
+	}
 
 	switch (config->tx_type) {
 	case HWTSTAMP_TX_OFF:
@@ -3613,6 +3620,7 @@ static int e1000e_config_hwtstamp(struct e1000_adapter *adapter,
 	case HWTSTAMP_TX_ON:
 		break;
 	default:
+		NL_SET_ERR_MSG(extack, "Unsupported TX HW timestamp type");
 		return -ERANGE;
 	}
 
@@ -3643,9 +3651,8 @@ static int e1000e_config_hwtstamp(struct e1000_adapter *adapter,
 		is_l2 = true;
 		break;
 	case HWTSTAMP_FILTER_PTP_V2_L4_SYNC:
-		/* Hardware cannot filter just V2 L4 Sync messages;
-		 * fall-through to V2 (both L2 and L4) Sync.
-		 */
+		/* Hardware cannot filter just V2 L4 Sync messages */
+		fallthrough;
 	case HWTSTAMP_FILTER_PTP_V2_SYNC:
 		/* Also time stamps V2 Path Delay Request/Response. */
 		tsync_rx_ctl |= E1000_TSYNCRXCTL_TYPE_L2_L4_V2;
@@ -3654,9 +3661,8 @@ static int e1000e_config_hwtstamp(struct e1000_adapter *adapter,
 		is_l4 = true;
 		break;
 	case HWTSTAMP_FILTER_PTP_V2_L4_DELAY_REQ:
-		/* Hardware cannot filter just V2 L4 Delay Request messages;
-		 * fall-through to V2 (both L2 and L4) Delay Request.
-		 */
+		/* Hardware cannot filter just V2 L4 Delay Request messages */
+		fallthrough;
 	case HWTSTAMP_FILTER_PTP_V2_DELAY_REQ:
 		/* Also time stamps V2 Path Delay Request/Response. */
 		tsync_rx_ctl |= E1000_TSYNCRXCTL_TYPE_L2_L4_V2;
@@ -3666,9 +3672,8 @@ static int e1000e_config_hwtstamp(struct e1000_adapter *adapter,
 		break;
 	case HWTSTAMP_FILTER_PTP_V2_L4_EVENT:
 	case HWTSTAMP_FILTER_PTP_V2_L2_EVENT:
-		/* Hardware cannot filter just V2 L4 or L2 Event messages;
-		 * fall-through to all V2 (both L2 and L4) Events.
-		 */
+		/* Hardware cannot filter just V2 L4 or L2 Event messages */
+		fallthrough;
 	case HWTSTAMP_FILTER_PTP_V2_EVENT:
 		tsync_rx_ctl |= E1000_TSYNCRXCTL_TYPE_EVENT_V2;
 		config->rx_filter = HWTSTAMP_FILTER_PTP_V2_EVENT;
@@ -3680,6 +3685,7 @@ static int e1000e_config_hwtstamp(struct e1000_adapter *adapter,
 		 * Delay Request messages but not both so fall-through to
 		 * time stamp all packets.
 		 */
+		fallthrough;
 	case HWTSTAMP_FILTER_NTP_ALL:
 	case HWTSTAMP_FILTER_ALL:
 		is_l2 = true;
@@ -3688,6 +3694,7 @@ static int e1000e_config_hwtstamp(struct e1000_adapter *adapter,
 		config->rx_filter = HWTSTAMP_FILTER_ALL;
 		break;
 	default:
+		NL_SET_ERR_MSG(extack, "Unsupported RX HW timestamp filter");
 		return -ERANGE;
 	}
 
@@ -3700,7 +3707,8 @@ static int e1000e_config_hwtstamp(struct e1000_adapter *adapter,
 	ew32(TSYNCTXCTL, regval);
 	if ((er32(TSYNCTXCTL) & E1000_TSYNCTXCTL_ENABLED) !=
 	    (regval & E1000_TSYNCTXCTL_ENABLED)) {
-		e_err("Timesync Tx Control register not set as expected\n");
+		NL_SET_ERR_MSG(extack,
+			       "Timesync Tx Control register not set as expected");
 		return -EAGAIN;
 	}
 
@@ -3713,7 +3721,8 @@ static int e1000e_config_hwtstamp(struct e1000_adapter *adapter,
 				 E1000_TSYNCRXCTL_TYPE_MASK)) !=
 	    (regval & (E1000_TSYNCRXCTL_ENABLED |
 		       E1000_TSYNCRXCTL_TYPE_MASK))) {
-		e_err("Timesync Rx Control register not set as expected\n");
+		NL_SET_ERR_MSG(extack,
+			       "Timesync Rx Control register not set as expected");
 		return -EAGAIN;
 	}
 
@@ -3780,6 +3789,7 @@ void e1000e_power_up_phy(struct e1000_adapter *adapter)
 
 /**
  * e1000_power_down_phy - Power down the PHY
+ * @adapter: board private structure
  *
  * Power down the PHY so no link is implied when interface is down.
  * The PHY cannot be powered down if management or WoL is active.
@@ -3792,6 +3802,7 @@ static void e1000_power_down_phy(struct e1000_adapter *adapter)
 
 /**
  * e1000_flush_tx_ring - remove all descriptors from the tx_ring
+ * @adapter: board private structure
  *
  * We want to clear all pending descriptors from the TX ring.
  * zeroing happens when the HW reads the regs. We  assign the ring itself as
@@ -3811,7 +3822,7 @@ static void e1000_flush_tx_ring(struct e1000_adapter *adapter)
 	tdt = er32(TDT(0));
 	BUG_ON(tdt != tx_ring->next_to_use);
 	tx_desc =  E1000_TX_DESC(*tx_ring, tx_ring->next_to_use);
-	tx_desc->buffer_addr = tx_ring->dma;
+	tx_desc->buffer_addr = cpu_to_le64(tx_ring->dma);
 
 	tx_desc->lower.data = cpu_to_le32(txd_lower | size);
 	tx_desc->upper.data = 0;
@@ -3821,12 +3832,12 @@ static void e1000_flush_tx_ring(struct e1000_adapter *adapter)
 	if (tx_ring->next_to_use == tx_ring->count)
 		tx_ring->next_to_use = 0;
 	ew32(TDT(0), tx_ring->next_to_use);
-	mmiowb();
 	usleep_range(200, 250);
 }
 
 /**
  * e1000_flush_rx_ring - remove all descriptors from the rx_ring
+ * @adapter: board private structure
  *
  * Mark all descriptors in the RX ring as consumed and disable the rx ring
  */
@@ -3859,6 +3870,7 @@ static void e1000_flush_rx_ring(struct e1000_adapter *adapter)
 
 /**
  * e1000_flush_desc_rings - remove all descriptors from the descriptor rings
+ * @adapter: board private structure
  *
  * In i219, the descriptor rings must be emptied before resetting the HW
  * or before changing the device state to D3 during runtime (runtime PM).
@@ -3905,6 +3917,7 @@ static void e1000e_systim_reset(struct e1000_adapter *adapter)
 {
 	struct ptp_clock_info *info = &adapter->ptp_clock_info;
 	struct e1000_hw *hw = &adapter->hw;
+	struct netlink_ext_ack extack = {};
 	unsigned long flags;
 	u32 timinca;
 	s32 ret_val;
@@ -3912,9 +3925,9 @@ static void e1000e_systim_reset(struct e1000_adapter *adapter)
 	if (!(adapter->flags & FLAG_HAS_HW_TIMESTAMP))
 		return;
 
-	if (info->adjfreq) {
+	if (info->adjfine) {
 		/* restore the previous ptp frequency delta */
-		ret_val = info->adjfreq(info, adapter->ptp_delta);
+		ret_val = info->adjfine(info, adapter->ptp_delta);
 	} else {
 		/* set the default base frequency if no adjustment possible */
 		ret_val = e1000e_get_base_timinca(adapter, &timinca);
@@ -3936,11 +3949,17 @@ static void e1000e_systim_reset(struct e1000_adapter *adapter)
 	spin_unlock_irqrestore(&adapter->systim_lock, flags);
 
 	/* restore the previous hwtstamp configuration settings */
-	e1000e_config_hwtstamp(adapter, &adapter->hwtstamp_config);
+	ret_val = e1000e_config_hwtstamp(adapter, &adapter->hwtstamp_config,
+					 &extack);
+	if (ret_val) {
+		if (extack._msg)
+			e_err("%s\n", extack._msg);
+	}
 }
 
 /**
  * e1000e_reset - bring the hardware into a known good state
+ * @adapter: board private structure
  *
  * This function boots the hardware and enables some settings that
  * require a configuration cycle of the hardware - those cannot be
@@ -4027,7 +4046,7 @@ void e1000e_reset(struct e1000_adapter *adapter)
 			fc->low_water = fc->high_water - 8;
 			break;
 		}
-		/* fall-through */
+		fallthrough;
 	default:
 		hwm = min(((pba << 10) * 9 / 10),
 			  ((pba << 10) - adapter->max_frame_size));
@@ -4052,12 +4071,18 @@ void e1000e_reset(struct e1000_adapter *adapter)
 	case e1000_pch_lpt:
 	case e1000_pch_spt:
 	case e1000_pch_cnp:
-		fc->refresh_time = 0x0400;
+	case e1000_pch_tgp:
+	case e1000_pch_adp:
+	case e1000_pch_mtp:
+	case e1000_pch_lnp:
+	case e1000_pch_ptp:
+	case e1000_pch_nvp:
+		fc->refresh_time = 0xFFFF;
+		fc->pause_time = 0xFFFF;
 
 		if (adapter->netdev->mtu <= ETH_DATA_LEN) {
 			fc->high_water = 0x05C20;
 			fc->low_water = 0x05048;
-			fc->pause_time = 0x0650;
 			break;
 		}
 
@@ -4189,7 +4214,7 @@ void e1000e_reset(struct e1000_adapter *adapter)
 
 /**
  * e1000e_trigger_lsc - trigger an LSC interrupt
- * @adapter: 
+ * @adapter: board private structure
  *
  * Fire a link status change interrupt to start the watchdog.
  **/
@@ -4198,7 +4223,7 @@ static void e1000e_trigger_lsc(struct e1000_adapter *adapter)
 	struct e1000_hw *hw = &adapter->hw;
 
 	if (adapter->msix_entries)
-		ew32(ICS, E1000_ICS_OTHER);
+		ew32(ICS, E1000_ICS_LSC | E1000_ICS_OTHER);
 	else
 		ew32(ICS, E1000_ICS_LSC);
 }
@@ -4214,7 +4239,7 @@ void e1000e_up(struct e1000_adapter *adapter)
 		e1000_configure_msix(adapter);
 	e1000_irq_enable(adapter);
 
-	netif_start_queue(adapter->netdev);
+	/* Tx queue started by watchdog timer when link is up */
 
 	e1000e_trigger_lsc(adapter);
 }
@@ -4278,14 +4303,14 @@ void e1000e_down(struct e1000_adapter *adapter, bool reset)
 
 	/* flush both disables and wait for them to finish */
 	e1e_flush();
-	usleep_range(10000, 20000);
+	usleep_range(10000, 11000);
 
 	e1000_irq_disable(adapter);
 
 	napi_synchronize(&adapter->napi);
 
-	del_timer_sync(&adapter->watchdog_timer);
-	del_timer_sync(&adapter->phy_info_timer);
+	timer_delete_sync(&adapter->watchdog_timer);
+	timer_delete_sync(&adapter->phy_info_timer);
 
 	spin_lock(&adapter->stats64_lock);
 	e1000e_update_stats(adapter);
@@ -4316,7 +4341,7 @@ void e1000e_reinit_locked(struct e1000_adapter *adapter)
 {
 	might_sleep();
 	while (test_and_set_bit(__E1000_RESETTING, &adapter->state))
-		usleep_range(1000, 2000);
+		usleep_range(1000, 1100);
 	e1000e_down(adapter, true);
 	e1000e_up(adapter);
 	clear_bit(__E1000_RESETTING, &adapter->state);
@@ -4325,13 +4350,16 @@ void e1000e_reinit_locked(struct e1000_adapter *adapter)
 /**
  * e1000e_sanitize_systim - sanitize raw cycle counter reads
  * @hw: pointer to the HW structure
- * @systim: time value read, sanitized and returned
+ * @systim: PHC time value read, sanitized and returned
+ * @sts: structure to hold system time before and after reading SYSTIML,
+ * may be NULL
  *
  * Errata for 82574/82583 possible bad bits read from SYSTIMH/L:
  * check to see that the time is incrementing at a reasonable
  * rate and is a multiple of incvalue.
  **/
-static u64 e1000e_sanitize_systim(struct e1000_hw *hw, u64 systim)
+static u64 e1000e_sanitize_systim(struct e1000_hw *hw, u64 systim,
+				  struct ptp_system_timestamp *sts)
 {
 	u64 time_delta, rem, temp;
 	u64 systim_next;
@@ -4341,7 +4369,9 @@ static u64 e1000e_sanitize_systim(struct e1000_hw *hw, u64 systim)
 	incvalue = er32(TIMINCA) & E1000_TIMINCA_INCVALUE_MASK;
 	for (i = 0; i < E1000_MAX_82574_SYSTIM_REREADS; i++) {
 		/* latch SYSTIMH on read of SYSTIML */
+		ptp_read_system_prets(sts);
 		systim_next = (u64)er32(SYSTIML);
+		ptp_read_system_postts(sts);
 		systim_next |= (u64)er32(SYSTIMH) << 32;
 
 		time_delta = systim_next - systim;
@@ -4359,15 +4389,16 @@ static u64 e1000e_sanitize_systim(struct e1000_hw *hw, u64 systim)
 }
 
 /**
- * e1000e_cyclecounter_read - read raw cycle counter (used by time counter)
- * @cc: cyclecounter structure
+ * e1000e_read_systim - read SYSTIM register
+ * @adapter: board private structure
+ * @sts: structure which will contain system time before and after reading
+ * SYSTIML, may be NULL
  **/
-static u64 e1000e_cyclecounter_read(const struct cyclecounter *cc)
+u64 e1000e_read_systim(struct e1000_adapter *adapter,
+		       struct ptp_system_timestamp *sts)
 {
-	struct e1000_adapter *adapter = container_of(cc, struct e1000_adapter,
-						     cc);
 	struct e1000_hw *hw = &adapter->hw;
-	u32 systimel, systimeh;
+	u32 systimel, systimel_2, systimeh;
 	u64 systim;
 	/* SYSTIMH latching upon SYSTIML read does not work well.
 	 * This means that if SYSTIML overflows after we read it but before
@@ -4375,11 +4406,15 @@ static u64 e1000e_cyclecounter_read(const struct cyclecounter *cc)
 	 * will experience a huge non linear increment in the systime value
 	 * to fix that we test for overflow and if true, we re-read systime.
 	 */
+	ptp_read_system_prets(sts);
 	systimel = er32(SYSTIML);
+	ptp_read_system_postts(sts);
 	systimeh = er32(SYSTIMH);
 	/* Is systimel is so large that overflow is possible? */
 	if (systimel >= (u32)0xffffffff - E1000_TIMINCA_INCVALUE_MASK) {
-		u32 systimel_2 = er32(SYSTIML);
+		ptp_read_system_prets(sts);
+		systimel_2 = er32(SYSTIML);
+		ptp_read_system_postts(sts);
 		if (systimel > systimel_2) {
 			/* There was an overflow, read again SYSTIMH, and use
 			 * systimel_2
@@ -4392,12 +4427,24 @@ static u64 e1000e_cyclecounter_read(const struct cyclecounter *cc)
 	systim |= (u64)systimeh << 32;
 
 	if (adapter->flags2 & FLAG2_CHECK_SYSTIM_OVERFLOW)
-		systim = e1000e_sanitize_systim(hw, systim);
+		systim = e1000e_sanitize_systim(hw, systim, sts);
 
 	return systim;
 }
 
 /**
+ * e1000e_cyclecounter_read - read raw cycle counter (used by time counter)
+ * @cc: cyclecounter structure
+ **/
+static u64 e1000e_cyclecounter_read(struct cyclecounter *cc)
+{
+	struct e1000_adapter *adapter = container_of(cc, struct e1000_adapter,
+						     cc);
+
+	return e1000e_read_systim(adapter, NULL);
+}
+
+/**
  * e1000_sw_init - Initialize general software structures (struct e1000_adapter)
  * @adapter: board private structure to initialize
  *
@@ -4582,6 +4629,7 @@ int e1000e_open(struct net_device *netdev)
 	struct e1000_hw *hw = &adapter->hw;
 	struct pci_dev *pdev = adapter->pdev;
 	int err;
+	int irq;
 
 	/* disallow open during test */
 	if (test_bit(__E1000_TESTING, &adapter->state))
@@ -4590,6 +4638,7 @@ int e1000e_open(struct net_device *netdev)
 	pm_runtime_get_sync(&pdev->dev);
 
 	netif_carrier_off(netdev);
+	netif_stop_queue(netdev);
 
 	/* allocate transmit descriptors */
 	err = e1000e_setup_tx_resources(adapter->tx_ring);
@@ -4616,8 +4665,7 @@ int e1000e_open(struct net_device *netdev)
 		e1000_update_mng_vlan(adapter);
 
 	/* DMA latency requirement to workaround jumbo issue */
-	pm_qos_add_request(&adapter->pm_qos_req, PM_QOS_CPU_DMA_LATENCY,
-			   PM_QOS_DEFAULT_VALUE);
+	cpu_latency_qos_add_request(&adapter->pm_qos_req, PM_QOS_DEFAULT_VALUE);
 
 	/* before we allocate an interrupt, we must be ready to handle it.
 	 * Setting DEBUG_SHIRQ in the kernel makes it fire an interrupt
@@ -4645,12 +4693,19 @@ int e1000e_open(struct net_device *netdev)
 	/* From here on the code is the same as e1000e_up() */
 	clear_bit(__E1000_DOWN, &adapter->state);
 
+	if (adapter->int_mode == E1000E_INT_MODE_MSIX)
+		irq = adapter->msix_entries[0].vector;
+	else
+		irq = adapter->pdev->irq;
+
+	netif_napi_set_irq(&adapter->napi, irq);
 	napi_enable(&adapter->napi);
+	netif_queue_set_napi(netdev, 0, NETDEV_QUEUE_TYPE_RX, &adapter->napi);
+	netif_queue_set_napi(netdev, 0, NETDEV_QUEUE_TYPE_TX, &adapter->napi);
 
 	e1000_irq_enable(adapter);
 
 	adapter->tx_hang_recheck = false;
-	netif_start_queue(netdev);
 
 	hw->mac.get_link_status = true;
 	pm_runtime_put(&pdev->dev);
@@ -4660,7 +4715,7 @@ int e1000e_open(struct net_device *netdev)
 	return 0;
 
 err_req_irq:
-	pm_qos_remove_request(&adapter->pm_qos_req);
+	cpu_latency_qos_remove_request(&adapter->pm_qos_req);
 	e1000e_release_hw_control(adapter);
 	e1000_power_down_phy(adapter);
 	e1000e_free_rx_resources(adapter->rx_ring);
@@ -4691,20 +4746,22 @@ int e1000e_close(struct net_device *netdev)
 	int count = E1000_CHECK_RESET_COUNT;
 
 	while (test_bit(__E1000_RESETTING, &adapter->state) && count--)
-		usleep_range(10000, 20000);
+		usleep_range(10000, 11000);
 
 	WARN_ON(test_bit(__E1000_RESETTING, &adapter->state));
 
 	pm_runtime_get_sync(&pdev->dev);
 
-	if (!test_bit(__E1000_DOWN, &adapter->state)) {
+	if (netif_device_present(netdev)) {
 		e1000e_down(adapter, true);
 		e1000_free_irq(adapter);
 
 		/* Link status message must follow this format */
-		pr_info("%s NIC Link is Down\n", adapter->netdev->name);
+		netdev_info(netdev, "NIC Link is Down\n");
 	}
 
+	netif_queue_set_napi(netdev, 0, NETDEV_QUEUE_TYPE_RX, NULL);
+	netif_queue_set_napi(netdev, 0, NETDEV_QUEUE_TYPE_TX, NULL);
 	napi_disable(&adapter->napi);
 
 	e1000e_free_tx_resources(adapter->tx_ring);
@@ -4724,7 +4781,7 @@ int e1000e_close(struct net_device *netdev)
 	    !test_bit(__E1000_TESTING, &adapter->state))
 		e1000e_release_hw_control(adapter);
 
-	pm_qos_remove_request(&adapter->pm_qos_req);
+	cpu_latency_qos_remove_request(&adapter->pm_qos_req);
 
 	pm_runtime_put_sync(&pdev->dev);
 
@@ -4747,7 +4804,7 @@ static int e1000_set_mac(struct net_device *netdev, void *p)
 	if (!is_valid_ether_addr(addr->sa_data))
 		return -EADDRNOTAVAIL;
 
-	memcpy(netdev->dev_addr, addr->sa_data, netdev->addr_len);
+	eth_hw_addr_set(netdev, addr->sa_data);
 	memcpy(adapter->hw.mac.addr, addr->sa_data, netdev->addr_len);
 
 	hw->mac.ops.rar_set(&adapter->hw, adapter->hw.mac.addr, 0);
@@ -4797,14 +4854,15 @@ static void e1000e_update_phy_task(struct work_struct *work)
 
 /**
  * e1000_update_phy_info - timre call-back to update PHY info
- * @data: pointer to adapter cast into an unsigned long
+ * @t: pointer to timer_list containing private info adapter
  *
  * Need to wait a few seconds after link up to get diagnostic information from
  * the phy
  **/
-static void e1000_update_phy_info(unsigned long data)
+static void e1000_update_phy_info(struct timer_list *t)
 {
-	struct e1000_adapter *adapter = (struct e1000_adapter *)data;
+	struct e1000_adapter *adapter = timer_container_of(adapter, t,
+							   phy_info_timer);
 
 	if (test_bit(__E1000_DOWN, &adapter->state))
 		return;
@@ -4999,8 +5057,7 @@ static void e1000e_update_stats(struct e1000_adapter *adapter)
 		adapter->corr_errors +=
 		    pbeccsts & E1000_PBECCSTS_CORR_ERR_CNT_MASK;
 		adapter->uncorr_errors +=
-		    (pbeccsts & E1000_PBECCSTS_UNCORR_ERR_CNT_MASK) >>
-		    E1000_PBECCSTS_UNCORR_ERR_CNT_SHIFT;
+		    FIELD_GET(E1000_PBECCSTS_UNCORR_ERR_CNT_MASK, pbeccsts);
 	}
 }
 
@@ -5052,12 +5109,13 @@ static void e1000_print_link_info(struct e1000_adapter *adapter)
 	u32 ctrl = er32(CTRL);
 
 	/* Link status message must follow this format for user tools */
-	pr_info("%s NIC Link is Up %d Mbps %s Duplex, Flow Control: %s\n",
-		adapter->netdev->name, adapter->link_speed,
-		adapter->link_duplex == FULL_DUPLEX ? "Full" : "Half",
-		(ctrl & E1000_CTRL_TFCE) && (ctrl & E1000_CTRL_RFCE) ? "Rx/Tx" :
-		(ctrl & E1000_CTRL_RFCE) ? "Rx" :
-		(ctrl & E1000_CTRL_TFCE) ? "Tx" : "None");
+	netdev_info(adapter->netdev,
+		    "NIC Link is Up %d Mbps %s Duplex, Flow Control: %s\n",
+		    adapter->link_speed,
+		    adapter->link_duplex == FULL_DUPLEX ? "Full" : "Half",
+		    (ctrl & E1000_CTRL_TFCE) && (ctrl & E1000_CTRL_RFCE) ? "Rx/Tx" :
+		    (ctrl & E1000_CTRL_RFCE) ? "Rx" :
+		    (ctrl & E1000_CTRL_TFCE) ? "Tx" : "None");
 }
 
 static bool e1000e_has_link(struct e1000_adapter *adapter)
@@ -5068,7 +5126,7 @@ static bool e1000e_has_link(struct e1000_adapter *adapter)
 
 	/* get_link_status is set on LSC (link status) interrupt or
 	 * Rx sequence error interrupt.  get_link_status will stay
-	 * false until the check_for_link establishes link
+	 * true until the check_for_link establishes link
 	 * for copper adapters ONLY
 	 */
 	switch (hw->phy.media_type) {
@@ -5086,14 +5144,14 @@ static bool e1000e_has_link(struct e1000_adapter *adapter)
 		break;
 	case e1000_media_type_internal_serdes:
 		ret_val = hw->mac.ops.check_for_link(hw);
-		link_active = adapter->hw.mac.serdes_has_link;
+		link_active = hw->mac.serdes_has_link;
 		break;
 	default:
 	case e1000_media_type_unknown:
 		break;
 	}
 
-	if ((ret_val == E1000_ERR_PHY) && (hw->phy.type == e1000_phy_igp_3) &&
+	if ((ret_val == -E1000_ERR_PHY) && (hw->phy.type == e1000_phy_igp_3) &&
 	    (er32(CTRL) & E1000_PHY_CTRL_GBE_DISABLE)) {
 		/* See e1000_kmrn_lock_loss_workaround_ich8lan() */
 		e_info("Gigabit has been disabled, downgrading speed\n");
@@ -5136,11 +5194,12 @@ static void e1000e_check_82574_phy_workaround(struct e1000_adapter *adapter)
 
 /**
  * e1000_watchdog - Timer Call-back
- * @data: pointer to adapter cast into an unsigned long
+ * @t: pointer to timer_list containing private info adapter
  **/
-static void e1000_watchdog(unsigned long data)
+static void e1000_watchdog(struct timer_list *t)
 {
-	struct e1000_adapter *adapter = (struct e1000_adapter *)data;
+	struct e1000_adapter *adapter = timer_container_of(adapter, t,
+							   watchdog_timer);
 
 	/* Do the rest outside of interrupt context */
 	schedule_work(&adapter->watchdog_task);
@@ -5157,8 +5216,9 @@ static void e1000_watchdog_task(struct work_struct *work)
 	struct e1000_mac_info *mac = &adapter->hw.mac;
 	struct e1000_phy_info *phy = &adapter->hw.phy;
 	struct e1000_ring *tx_ring = adapter->tx_ring;
+	u32 dmoff_exit_timeout = 100, tries = 0;
 	struct e1000_hw *hw = &adapter->hw;
-	u32 link, tctl;
+	u32 link, tctl, pcim_state;
 
 	if (test_bit(__E1000_DOWN, &adapter->state))
 		return;
@@ -5183,6 +5243,23 @@ static void e1000_watchdog_task(struct work_struct *work)
 			/* Cancel scheduled suspend requests. */
 			pm_runtime_resume(netdev->dev.parent);
 
+			/* Checking if MAC is in DMoff state*/
+			if (er32(FWSM) & E1000_ICH_FWSM_FW_VALID) {
+				pcim_state = er32(STATUS);
+				while (pcim_state & E1000_STATUS_PCIM_STATE) {
+					if (tries++ == dmoff_exit_timeout) {
+						e_dbg("Error in exiting dmoff\n");
+						break;
+					}
+					usleep_range(10000, 20000);
+					pcim_state = er32(STATUS);
+
+					/* Checking if MAC exited DMoff state */
+					if (!(pcim_state & E1000_STATUS_PCIM_STATE))
+						e1000_phy_hw_reset(&adapter->hw);
+				}
+			}
+
 			/* update snapshot of PHY registers on LSC */
 			e1000_phy_read_status(adapter);
 			mac->ops.get_link_up_info(&adapter->hw,
@@ -5238,27 +5315,6 @@ static void e1000_watchdog_task(struct work_struct *work)
 				ew32(TARC(0), tarc0);
 			}
 
-			/* disable TSO for pcie and 10/100 speeds, to avoid
-			 * some hardware issues
-			 */
-			if (!(adapter->flags & FLAG_TSO_FORCE)) {
-				switch (adapter->link_speed) {
-				case SPEED_10:
-				case SPEED_100:
-					e_info("10/100 speed: disabling TSO\n");
-					netdev->features &= ~NETIF_F_TSO;
-					netdev->features &= ~NETIF_F_TSO6;
-					break;
-				case SPEED_1000:
-					netdev->features |= NETIF_F_TSO;
-					netdev->features |= NETIF_F_TSO6;
-					break;
-				default:
-					/* oops */
-					break;
-				}
-			}
-
 			/* enable transmits in the hardware, need to do this
 			 * after setting TARC(0)
 			 */
@@ -5272,6 +5328,7 @@ static void e1000_watchdog_task(struct work_struct *work)
 			if (phy->ops.cfg_on_link_up)
 				phy->ops.cfg_on_link_up(hw);
 
+			netif_wake_queue(netdev);
 			netif_carrier_on(netdev);
 
 			if (!test_bit(__E1000_DOWN, &adapter->state))
@@ -5283,8 +5340,9 @@ static void e1000_watchdog_task(struct work_struct *work)
 			adapter->link_speed = 0;
 			adapter->link_duplex = 0;
 			/* Link status message must follow this format */
-			pr_info("%s NIC Link is Down\n", adapter->netdev->name);
+			netdev_info(netdev, "NIC Link is Down\n");
 			netif_carrier_off(netdev);
+			netif_stop_queue(netdev);
 			if (!test_bit(__E1000_DOWN, &adapter->state))
 				mod_timer(&adapter->phy_info_timer,
 					  round_jiffies(jiffies + 2 * HZ));
@@ -5413,7 +5471,7 @@ static int e1000_tso(struct e1000_ring *tx_ring, struct sk_buff *skb,
 	if (err < 0)
 		return err;
 
-	hdr_len = skb_transport_offset(skb) + tcp_hdrlen(skb);
+	hdr_len = skb_tcp_all_headers(skb);
 	mss = skb_shinfo(skb)->gso_size;
 	if (protocol == htons(ETH_P_IP)) {
 		struct iphdr *iph = ip_hdr(skb);
@@ -5424,10 +5482,7 @@ static int e1000_tso(struct e1000_ring *tx_ring, struct sk_buff *skb,
 		cmd_length = E1000_TXD_CMD_IP;
 		ipcse = skb_transport_offset(skb) - 1;
 	} else if (skb_is_gso_v6(skb)) {
-		ipv6_hdr(skb)->payload_len = 0;
-		tcp_hdr(skb)->check = ~csum_ipv6_magic(&ipv6_hdr(skb)->saddr,
-						       &ipv6_hdr(skb)->daddr,
-						       0, IPPROTO_TCP, 0);
+		tcp_v6_gso_csum_prep(skb);
 		ipcse = 0;
 	}
 	ipcss = skb_network_offset(skb);
@@ -5556,9 +5611,8 @@ static int e1000_tx_map(struct e1000_ring *tx_ring, struct sk_buff *skb,
 	}
 
 	for (f = 0; f < nr_frags; f++) {
-		const struct skb_frag_struct *frag;
+		const skb_frag_t *frag = &skb_shinfo(skb)->frags[f];
 
-		frag = &skb_shinfo(skb)->frags[f];
 		len = skb_frag_size(frag);
 		offset = 0;
 
@@ -5608,7 +5662,7 @@ dma_error:
 			i += tx_ring->count;
 		i--;
 		buffer_info = &tx_ring->buffer_info[i];
-		e1000_put_txbuf(tx_ring, buffer_info);
+		e1000_put_txbuf(tx_ring, buffer_info, true);
 	}
 
 	return 0;
@@ -5789,7 +5843,7 @@ static netdev_tx_t e1000_xmit_frame(struct sk_buff *skb,
 		 * points to just header, pull a few bytes of payload from
 		 * frags into skb->data
 		 */
-		hdr_len = skb_transport_offset(skb) + tcp_hdrlen(skb);
+		hdr_len = skb_tcp_all_headers(skb);
 		/* we do this workaround for ES2LAN, but it is un-necessary,
 		 * avoiding it could save a lot of cycles
 		 */
@@ -5861,13 +5915,16 @@ static netdev_tx_t e1000_xmit_frame(struct sk_buff *skb,
 			     nr_frags);
 	if (count) {
 		if (unlikely(skb_shinfo(skb)->tx_flags & SKBTX_HW_TSTAMP) &&
-		    (adapter->flags & FLAG_HAS_HW_TIMESTAMP) &&
-		    !adapter->tx_hwtstamp_skb) {
-			skb_shinfo(skb)->tx_flags |= SKBTX_IN_PROGRESS;
-			tx_flags |= E1000_TX_FLAGS_HWTSTAMP;
-			adapter->tx_hwtstamp_skb = skb_get(skb);
-			adapter->tx_hwtstamp_start = jiffies;
-			schedule_work(&adapter->tx_hwtstamp_work);
+		    (adapter->flags & FLAG_HAS_HW_TIMESTAMP)) {
+			if (!adapter->tx_hwtstamp_skb) {
+				skb_shinfo(skb)->tx_flags |= SKBTX_IN_PROGRESS;
+				tx_flags |= E1000_TX_FLAGS_HWTSTAMP;
+				adapter->tx_hwtstamp_skb = skb_get(skb);
+				adapter->tx_hwtstamp_start = jiffies;
+				schedule_work(&adapter->tx_hwtstamp_work);
+			} else {
+				adapter->tx_hwtstamp_skipped++;
+			}
 		}
 
 		skb_tx_timestamp(skb);
@@ -5876,23 +5933,17 @@ static netdev_tx_t e1000_xmit_frame(struct sk_buff *skb,
 		e1000_tx_queue(tx_ring, tx_flags, count);
 		/* Make sure there is space in the ring for the next send. */
 		e1000_maybe_stop_tx(tx_ring,
-				    (MAX_SKB_FRAGS *
+				    ((MAX_SKB_FRAGS + 1) *
 				     DIV_ROUND_UP(PAGE_SIZE,
-						  adapter->tx_fifo_limit) + 2));
+						  adapter->tx_fifo_limit) + 4));
 
-		if (!skb->xmit_more ||
+		if (!netdev_xmit_more() ||
 		    netif_xmit_stopped(netdev_get_tx_queue(netdev, 0))) {
 			if (adapter->flags2 & FLAG2_PCIM2PCI_ARBITER_WA)
 				e1000e_update_tdt_wa(tx_ring,
 						     tx_ring->next_to_use);
 			else
 				writel(tx_ring->next_to_use, tx_ring->tail);
-
-			/* we need this if more than one processor can write
-			 * to our tail at a time, it synchronizes IO on
-			 *IA64/Altix systems
-			 */
-			mmiowb();
 		}
 	} else {
 		dev_kfree_skb_any(skb);
@@ -5906,8 +5957,9 @@ static netdev_tx_t e1000_xmit_frame(struct sk_buff *skb,
 /**
  * e1000_tx_timeout - Respond to a Tx Hang
  * @netdev: network interface device structure
+ * @txqueue: index of the hung queue (unused)
  **/
-static void e1000_tx_timeout(struct net_device *netdev)
+static void e1000_tx_timeout(struct net_device *netdev, unsigned int __always_unused txqueue)
 {
 	struct e1000_adapter *adapter = netdev_priv(netdev);
 
@@ -5921,19 +5973,23 @@ static void e1000_reset_task(struct work_struct *work)
 	struct e1000_adapter *adapter;
 	adapter = container_of(work, struct e1000_adapter, reset_task);
 
+	rtnl_lock();
 	/* don't run the task if already down */
-	if (test_bit(__E1000_DOWN, &adapter->state))
+	if (test_bit(__E1000_DOWN, &adapter->state)) {
+		rtnl_unlock();
 		return;
+	}
 
 	if (!(adapter->flags & FLAG_RESTART_NOW)) {
 		e1000e_dump(adapter);
 		e_err("Reset adapter unexpectedly\n");
 	}
 	e1000e_reinit_locked(adapter);
+	rtnl_unlock();
 }
 
 /**
- * e1000_get_stats64 - Get System Network Statistics
+ * e1000e_get_stats64 - Get System Network Statistics
  * @netdev: network interface device structure
  * @stats: rtnl_link_stats64 pointer
  *
@@ -6006,11 +6062,12 @@ static int e1000_change_mtu(struct net_device *netdev, int new_mtu)
 	}
 
 	while (test_and_set_bit(__E1000_RESETTING, &adapter->state))
-		usleep_range(1000, 2000);
+		usleep_range(1000, 1100);
 	/* e1000e_down -> e1000e_reset dependent on max_frame_size & mtu */
 	adapter->max_frame_size = max_frame;
-	e_info("changing MTU from %d to %d\n", netdev->mtu, new_mtu);
-	netdev->mtu = new_mtu;
+	netdev_dbg(netdev, "changing MTU from %d to %d\n",
+		   netdev->mtu, new_mtu);
+	WRITE_ONCE(netdev->mtu, new_mtu);
 
 	pm_runtime_get_sync(netdev->dev.parent);
 
@@ -6046,8 +6103,7 @@ static int e1000_change_mtu(struct net_device *netdev, int new_mtu)
 	return 0;
 }
 
-static int e1000_mii_ioctl(struct net_device *netdev, struct ifreq *ifr,
-			   int cmd)
+static int e1000_ioctl(struct net_device *netdev, struct ifreq *ifr, int cmd)
 {
 	struct e1000_adapter *adapter = netdev_priv(netdev);
 	struct mii_ioctl_data *data = if_mii(ifr);
@@ -6105,9 +6161,10 @@ static int e1000_mii_ioctl(struct net_device *netdev, struct ifreq *ifr,
 }
 
 /**
- * e1000e_hwtstamp_ioctl - control hardware time stamping
+ * e1000e_hwtstamp_set - control hardware time stamping
  * @netdev: network interface device structure
- * @ifreq: interface request
+ * @config: timestamp configuration
+ * @extack: netlink extended ACK report
  *
  * Outgoing time stamping can be enabled and disabled. Play nice and
  * disable it when requested, although it shouldn't cause any overhead
@@ -6120,20 +6177,18 @@ static int e1000_mii_ioctl(struct net_device *netdev, struct ifreq *ifr,
  * specified. Matching the kind of event packet is not supported, with the
  * exception of "all V2 events regardless of level 2 or 4".
  **/
-static int e1000e_hwtstamp_set(struct net_device *netdev, struct ifreq *ifr)
+static int e1000e_hwtstamp_set(struct net_device *netdev,
+			       struct kernel_hwtstamp_config *config,
+			       struct netlink_ext_ack *extack)
 {
 	struct e1000_adapter *adapter = netdev_priv(netdev);
-	struct hwtstamp_config config;
 	int ret_val;
 
-	if (copy_from_user(&config, ifr->ifr_data, sizeof(config)))
-		return -EFAULT;
-
-	ret_val = e1000e_config_hwtstamp(adapter, &config);
+	ret_val = e1000e_config_hwtstamp(adapter, config, extack);
 	if (ret_val)
 		return ret_val;
 
-	switch (config.rx_filter) {
+	switch (config->rx_filter) {
 	case HWTSTAMP_FILTER_PTP_V2_L4_SYNC:
 	case HWTSTAMP_FILTER_PTP_V2_L2_SYNC:
 	case HWTSTAMP_FILTER_PTP_V2_SYNC:
@@ -6145,38 +6200,23 @@ static int e1000e_hwtstamp_set(struct net_device *netdev, struct ifreq *ifr)
 		 * by hardware so notify the caller the requested packets plus
 		 * some others are time stamped.
 		 */
-		config.rx_filter = HWTSTAMP_FILTER_SOME;
+		config->rx_filter = HWTSTAMP_FILTER_SOME;
 		break;
 	default:
 		break;
 	}
 
-	return copy_to_user(ifr->ifr_data, &config,
-			    sizeof(config)) ? -EFAULT : 0;
+	return 0;
 }
 
-static int e1000e_hwtstamp_get(struct net_device *netdev, struct ifreq *ifr)
+static int e1000e_hwtstamp_get(struct net_device *netdev,
+			       struct kernel_hwtstamp_config *kernel_config)
 {
 	struct e1000_adapter *adapter = netdev_priv(netdev);
 
-	return copy_to_user(ifr->ifr_data, &adapter->hwtstamp_config,
-			    sizeof(adapter->hwtstamp_config)) ? -EFAULT : 0;
-}
+	*kernel_config = adapter->hwtstamp_config;
 
-static int e1000_ioctl(struct net_device *netdev, struct ifreq *ifr, int cmd)
-{
-	switch (cmd) {
-	case SIOCGMIIPHY:
-	case SIOCGMIIREG:
-	case SIOCSMIIREG:
-		return e1000_mii_ioctl(netdev, ifr, cmd);
-	case SIOCSHWTSTAMP:
-		return e1000e_hwtstamp_set(netdev, ifr);
-	case SIOCGHWTSTAMP:
-		return e1000e_hwtstamp_get(netdev, ifr);
-	default:
-		return -EOPNOTSUPP;
-	}
+	return 0;
 }
 
 static int e1000_init_phy_wakeup(struct e1000_adapter *adapter, u32 wufc)
@@ -6218,7 +6258,7 @@ static int e1000_init_phy_wakeup(struct e1000_adapter *adapter, u32 wufc)
 		phy_reg |= BM_RCTL_MPE;
 	phy_reg &= ~(BM_RCTL_MO_MASK);
 	if (mac_reg & E1000_RCTL_MO_3)
-		phy_reg |= (((mac_reg & E1000_RCTL_MO_3) >> E1000_RCTL_MO_SHIFT)
+		phy_reg |= (FIELD_GET(E1000_RCTL_MO_3, mac_reg)
 			    << BM_RCTL_MO_SHIFT);
 	if (mac_reg & E1000_RCTL_BAM)
 		phy_reg |= BM_RCTL_BAM;
@@ -6275,18 +6315,302 @@ fl_out:
 	pm_runtime_put_sync(netdev->dev.parent);
 }
 
+/* S0ix implementation */
+static void e1000e_s0ix_entry_flow(struct e1000_adapter *adapter)
+{
+	struct e1000_hw *hw = &adapter->hw;
+	u32 mac_data;
+	u16 phy_data;
+
+	if (er32(FWSM) & E1000_ICH_FWSM_FW_VALID &&
+	    hw->mac.type >= e1000_pch_adp) {
+		/* Request ME configure the device for S0ix */
+		mac_data = er32(H2ME);
+		mac_data |= E1000_H2ME_START_DPG;
+		mac_data &= ~E1000_H2ME_EXIT_DPG;
+		trace_e1000e_trace_mac_register(mac_data);
+		ew32(H2ME, mac_data);
+	} else {
+		/* Request driver configure the device to S0ix */
+		/* Disable the periodic inband message,
+		 * don't request PCIe clock in K1 page770_17[10:9] = 10b
+		 */
+		e1e_rphy(hw, HV_PM_CTRL, &phy_data);
+		phy_data &= ~HV_PM_CTRL_K1_CLK_REQ;
+		phy_data |= BIT(10);
+		e1e_wphy(hw, HV_PM_CTRL, phy_data);
+
+		/* Make sure we don't exit K1 every time a new packet arrives
+		 * 772_29[5] = 1 CS_Mode_Stay_In_K1
+		 */
+		e1e_rphy(hw, I217_CGFREG, &phy_data);
+		phy_data |= BIT(5);
+		e1e_wphy(hw, I217_CGFREG, phy_data);
+
+		/* Change the MAC/PHY interface to SMBus
+		 * Force the SMBus in PHY page769_23[0] = 1
+		 * Force the SMBus in MAC CTRL_EXT[11] = 1
+		 */
+		e1e_rphy(hw, CV_SMB_CTRL, &phy_data);
+		phy_data |= CV_SMB_CTRL_FORCE_SMBUS;
+		e1e_wphy(hw, CV_SMB_CTRL, phy_data);
+		mac_data = er32(CTRL_EXT);
+		mac_data |= E1000_CTRL_EXT_FORCE_SMBUS;
+		ew32(CTRL_EXT, mac_data);
+
+		/* DFT control: PHY bit: page769_20[0] = 1
+		 * page769_20[7] - PHY PLL stop
+		 * page769_20[8] - PHY go to the electrical idle
+		 * page769_20[9] - PHY serdes disable
+		 * Gate PPW via EXTCNF_CTRL - set 0x0F00[7] = 1
+		 */
+		e1e_rphy(hw, I82579_DFT_CTRL, &phy_data);
+		phy_data |= BIT(0);
+		phy_data |= BIT(7);
+		phy_data |= BIT(8);
+		phy_data |= BIT(9);
+		e1e_wphy(hw, I82579_DFT_CTRL, phy_data);
+
+		mac_data = er32(EXTCNF_CTRL);
+		mac_data |= E1000_EXTCNF_CTRL_GATE_PHY_CFG;
+		ew32(EXTCNF_CTRL, mac_data);
+
+		/* Disable disconnected cable conditioning for Power Gating */
+		mac_data = er32(DPGFR);
+		mac_data |= BIT(2);
+		ew32(DPGFR, mac_data);
+
+		/* Enable the Dynamic Clock Gating in the DMA and MAC */
+		mac_data = er32(CTRL_EXT);
+		mac_data |= E1000_CTRL_EXT_DMA_DYN_CLK_EN;
+		ew32(CTRL_EXT, mac_data);
+	}
+
+	/* Enable the Dynamic Power Gating in the MAC */
+	mac_data = er32(FEXTNVM7);
+	mac_data |= BIT(22);
+	ew32(FEXTNVM7, mac_data);
+
+	/* Don't wake from dynamic Power Gating with clock request */
+	mac_data = er32(FEXTNVM12);
+	mac_data |= BIT(12);
+	ew32(FEXTNVM12, mac_data);
+
+	/* Ungate PGCB clock */
+	mac_data = er32(FEXTNVM9);
+	mac_data &= ~BIT(28);
+	ew32(FEXTNVM9, mac_data);
+
+	/* Enable K1 off to enable mPHY Power Gating */
+	mac_data = er32(FEXTNVM6);
+	mac_data |= BIT(31);
+	ew32(FEXTNVM6, mac_data);
+
+	/* Enable mPHY power gating for any link and speed */
+	mac_data = er32(FEXTNVM8);
+	mac_data |= BIT(9);
+	ew32(FEXTNVM8, mac_data);
+
+	/* No MAC DPG gating SLP_S0 in modern standby
+	 * Switch the logic of the lanphypc to use PMC counter
+	 */
+	mac_data = er32(FEXTNVM5);
+	mac_data |= BIT(7);
+	ew32(FEXTNVM5, mac_data);
+
+	/* Disable the time synchronization clock */
+	mac_data = er32(FEXTNVM7);
+	mac_data |= BIT(31);
+	mac_data &= ~BIT(0);
+	ew32(FEXTNVM7, mac_data);
+
+	/* Dynamic Power Gating Enable */
+	mac_data = er32(CTRL_EXT);
+	mac_data |= BIT(3);
+	ew32(CTRL_EXT, mac_data);
+
+	/* Check MAC Tx/Rx packet buffer pointers.
+	 * Reset MAC Tx/Rx packet buffer pointers to suppress any
+	 * pending traffic indication that would prevent power gating.
+	 */
+	mac_data = er32(TDFH);
+	if (mac_data)
+		ew32(TDFH, 0);
+	mac_data = er32(TDFT);
+	if (mac_data)
+		ew32(TDFT, 0);
+	mac_data = er32(TDFHS);
+	if (mac_data)
+		ew32(TDFHS, 0);
+	mac_data = er32(TDFTS);
+	if (mac_data)
+		ew32(TDFTS, 0);
+	mac_data = er32(TDFPC);
+	if (mac_data)
+		ew32(TDFPC, 0);
+	mac_data = er32(RDFH);
+	if (mac_data)
+		ew32(RDFH, 0);
+	mac_data = er32(RDFT);
+	if (mac_data)
+		ew32(RDFT, 0);
+	mac_data = er32(RDFHS);
+	if (mac_data)
+		ew32(RDFHS, 0);
+	mac_data = er32(RDFTS);
+	if (mac_data)
+		ew32(RDFTS, 0);
+	mac_data = er32(RDFPC);
+	if (mac_data)
+		ew32(RDFPC, 0);
+}
+
+static void e1000e_s0ix_exit_flow(struct e1000_adapter *adapter)
+{
+	struct e1000_hw *hw = &adapter->hw;
+	bool firmware_bug = false;
+	u32 mac_data;
+	u16 phy_data;
+	u32 i = 0;
+
+	if (er32(FWSM) & E1000_ICH_FWSM_FW_VALID &&
+	    hw->mac.type >= e1000_pch_adp) {
+		/* Keep the GPT clock enabled for CSME */
+		mac_data = er32(FEXTNVM);
+		mac_data |= BIT(3);
+		ew32(FEXTNVM, mac_data);
+		/* Request ME unconfigure the device from S0ix */
+		mac_data = er32(H2ME);
+		mac_data &= ~E1000_H2ME_START_DPG;
+		mac_data |= E1000_H2ME_EXIT_DPG;
+		trace_e1000e_trace_mac_register(mac_data);
+		ew32(H2ME, mac_data);
+
+		/* Poll up to 2.5 seconds for ME to unconfigure DPG.
+		 * If this takes more than 1 second, show a warning indicating a
+		 * firmware bug
+		 */
+		while (!(er32(EXFWSM) & E1000_EXFWSM_DPG_EXIT_DONE)) {
+			if (i > 100 && !firmware_bug)
+				firmware_bug = true;
+
+			if (i++ == 250) {
+				e_dbg("Timeout (firmware bug): %d msec\n",
+				      i * 10);
+				break;
+			}
+
+			usleep_range(10000, 11000);
+		}
+		if (firmware_bug)
+			e_warn("DPG_EXIT_DONE took %d msec. This is a firmware bug\n",
+			       i * 10);
+		else
+			e_dbg("DPG_EXIT_DONE cleared after %d msec\n", i * 10);
+	} else {
+		/* Request driver unconfigure the device from S0ix */
+
+		/* Cancel disable disconnected cable conditioning
+		 * for Power Gating
+		 */
+		mac_data = er32(DPGFR);
+		mac_data &= ~BIT(2);
+		ew32(DPGFR, mac_data);
+
+		/* Disable the Dynamic Clock Gating in the DMA and MAC */
+		mac_data = er32(CTRL_EXT);
+		mac_data &= 0xFFF7FFFF;
+		ew32(CTRL_EXT, mac_data);
+
+		/* Enable the periodic inband message,
+		 * Request PCIe clock in K1 page770_17[10:9] =01b
+		 */
+		e1e_rphy(hw, HV_PM_CTRL, &phy_data);
+		phy_data &= 0xFBFF;
+		phy_data |= HV_PM_CTRL_K1_CLK_REQ;
+		e1e_wphy(hw, HV_PM_CTRL, phy_data);
+
+		/* Return back configuration
+		 * 772_29[5] = 0 CS_Mode_Stay_In_K1
+		 */
+		e1e_rphy(hw, I217_CGFREG, &phy_data);
+		phy_data &= 0xFFDF;
+		e1e_wphy(hw, I217_CGFREG, phy_data);
+
+		/* Change the MAC/PHY interface to Kumeran
+		 * Unforce the SMBus in PHY page769_23[0] = 0
+		 * Unforce the SMBus in MAC CTRL_EXT[11] = 0
+		 */
+		e1e_rphy(hw, CV_SMB_CTRL, &phy_data);
+		phy_data &= ~CV_SMB_CTRL_FORCE_SMBUS;
+		e1e_wphy(hw, CV_SMB_CTRL, phy_data);
+		mac_data = er32(CTRL_EXT);
+		mac_data &= ~E1000_CTRL_EXT_FORCE_SMBUS;
+		ew32(CTRL_EXT, mac_data);
+	}
+
+	/* Disable Dynamic Power Gating */
+	mac_data = er32(CTRL_EXT);
+	mac_data &= 0xFFFFFFF7;
+	ew32(CTRL_EXT, mac_data);
+
+	/* Enable the time synchronization clock */
+	mac_data = er32(FEXTNVM7);
+	mac_data &= ~BIT(31);
+	mac_data |= BIT(0);
+	ew32(FEXTNVM7, mac_data);
+
+	/* Disable the Dynamic Power Gating in the MAC */
+	mac_data = er32(FEXTNVM7);
+	mac_data &= 0xFFBFFFFF;
+	ew32(FEXTNVM7, mac_data);
+
+	/* Disable mPHY power gating for any link and speed */
+	mac_data = er32(FEXTNVM8);
+	mac_data &= ~BIT(9);
+	ew32(FEXTNVM8, mac_data);
+
+	/* Disable K1 off */
+	mac_data = er32(FEXTNVM6);
+	mac_data &= ~BIT(31);
+	ew32(FEXTNVM6, mac_data);
+
+	/* Disable Ungate PGCB clock */
+	mac_data = er32(FEXTNVM9);
+	mac_data |= BIT(28);
+	ew32(FEXTNVM9, mac_data);
+
+	/* Cancel not waking from dynamic
+	 * Power Gating with clock request
+	 */
+	mac_data = er32(FEXTNVM12);
+	mac_data &= ~BIT(12);
+	ew32(FEXTNVM12, mac_data);
+
+	/* Revert the lanphypc logic to use the internal Gbe counter
+	 * and not the PMC counter
+	 */
+	mac_data = er32(FEXTNVM5);
+	mac_data &= 0xFFFFFF7F;
+	ew32(FEXTNVM5, mac_data);
+}
+
 static int e1000e_pm_freeze(struct device *dev)
 {
-	struct net_device *netdev = pci_get_drvdata(to_pci_dev(dev));
+	struct net_device *netdev = dev_get_drvdata(dev);
 	struct e1000_adapter *adapter = netdev_priv(netdev);
+	bool present;
 
+	rtnl_lock();
+
+	present = netif_device_present(netdev);
 	netif_device_detach(netdev);
 
-	if (netif_running(netdev)) {
+	if (present && netif_running(netdev)) {
 		int count = E1000_CHECK_RESET_COUNT;
 
 		while (test_bit(__E1000_RESETTING, &adapter->state) && count--)
-			usleep_range(10000, 20000);
+			usleep_range(10000, 11000);
 
 		WARN_ON(test_bit(__E1000_RESETTING, &adapter->state));
 
@@ -6294,6 +6618,8 @@ static int e1000e_pm_freeze(struct device *dev)
 		e1000e_down(adapter, false);
 		e1000_free_irq(adapter);
 	}
+	rtnl_unlock();
+
 	e1000e_reset_interrupt_capability(adapter);
 
 	/* Allow time for pending master requests to run */
@@ -6307,11 +6633,17 @@ static int __e1000_shutdown(struct pci_dev *pdev, bool runtime)
 	struct net_device *netdev = pci_get_drvdata(pdev);
 	struct e1000_adapter *adapter = netdev_priv(netdev);
 	struct e1000_hw *hw = &adapter->hw;
-	u32 ctrl, ctrl_ext, rctl, status;
-	/* Runtime suspend should only enable wakeup for link changes */
-	u32 wufc = runtime ? E1000_WUFC_LNKC : adapter->wol;
+	u32 ctrl, ctrl_ext, rctl, status, wufc;
 	int retval = 0;
 
+	/* Runtime suspend should only enable wakeup for link changes */
+	if (runtime)
+		wufc = E1000_WUFC_LNKC;
+	else if (device_may_wakeup(&pdev->dev))
+		wufc = adapter->wol;
+	else
+		wufc = 0;
+
 	status = er32(STATUS);
 	if (status & E1000_STATUS_LU)
 		wufc &= ~E1000_WUFC_LNKC;
@@ -6351,8 +6683,10 @@ static int __e1000_shutdown(struct pci_dev *pdev, bool runtime)
 		if (adapter->flags2 & FLAG2_HAS_PHY_WAKEUP) {
 			/* enable wakeup by the PHY */
 			retval = e1000_init_phy_wakeup(adapter, wufc);
-			if (retval)
-				return retval;
+			if (retval) {
+				e_err("Failed to enable wakeup\n");
+				goto skip_phy_configurations;
+			}
 		} else {
 			/* enable wakeup by the MAC */
 			ew32(WUFC, wufc);
@@ -6368,14 +6702,16 @@ static int __e1000_shutdown(struct pci_dev *pdev, bool runtime)
 	if (adapter->hw.phy.type == e1000_phy_igp_3) {
 		e1000e_igp3_phy_powerdown_workaround_ich8lan(&adapter->hw);
 	} else if (hw->mac.type >= e1000_pch_lpt) {
-		if (!(wufc & (E1000_WUFC_EX | E1000_WUFC_MC | E1000_WUFC_BC)))
+		if (wufc && !(wufc & (E1000_WUFC_EX | E1000_WUFC_MC | E1000_WUFC_BC))) {
 			/* ULP does not support wake from unicast, multicast
 			 * or broadcast.
 			 */
 			retval = e1000_enable_ulp_lpt_lp(hw, !runtime);
-
-		if (retval)
-			return retval;
+			if (retval) {
+				e_err("Failed to enable ULP\n");
+				goto skip_phy_configurations;
+			}
+		}
 	}
 
 	/* Ensure that the appropriate bits are set in LPI_CTRL
@@ -6406,6 +6742,7 @@ static int __e1000_shutdown(struct pci_dev *pdev, bool runtime)
 		hw->phy.ops.release(hw);
 	}
 
+skip_phy_configurations:
 	/* Release control of h/w to f/w.  If f/w is AMT enabled, this
 	 * would have already happened in close and is redundant.
 	 */
@@ -6460,7 +6797,7 @@ static void __e1000e_disable_aspm(struct pci_dev *pdev, u16 state, int locked)
 	case PCIE_LINK_STATE_L0S:
 	case PCIE_LINK_STATE_L0S | PCIE_LINK_STATE_L1:
 		aspm_dis_mask |= PCI_EXP_LNKCTL_ASPM_L0S;
-		/* fall-through - can't have L1 without L0s */
+		fallthrough; /* can't have L1 without L0s */
 	case PCIE_LINK_STATE_L1:
 		aspm_dis_mask |= PCI_EXP_LNKCTL_ASPM_L1;
 		break;
@@ -6529,7 +6866,7 @@ static void e1000e_disable_aspm(struct pci_dev *pdev, u16 state)
 }
 
 /**
- * e1000e_disable_aspm_locked   Disable ASPM states.
+ * e1000e_disable_aspm_locked - Disable ASPM states.
  * @pdev: pointer to PCI device struct
  * @state: bit-mask of ASPM states to disable
  *
@@ -6541,7 +6878,30 @@ static void e1000e_disable_aspm_locked(struct pci_dev *pdev, u16 state)
 	__e1000e_disable_aspm(pdev, state, 1);
 }
 
-#ifdef CONFIG_PM
+static int e1000e_pm_thaw(struct device *dev)
+{
+	struct net_device *netdev = dev_get_drvdata(dev);
+	struct e1000_adapter *adapter = netdev_priv(netdev);
+	int rc = 0;
+
+	e1000e_set_interrupt_capability(adapter);
+
+	rtnl_lock();
+	if (netif_running(netdev)) {
+		rc = e1000_request_irq(adapter);
+		if (rc)
+			goto err_irq;
+
+		e1000e_up(adapter);
+	}
+
+	netif_device_attach(netdev);
+err_irq:
+	rtnl_unlock();
+
+	return rc;
+}
+
 static int __e1000_resume(struct pci_dev *pdev)
 {
 	struct net_device *netdev = pci_get_drvdata(pdev);
@@ -6607,55 +6967,54 @@ static int __e1000_resume(struct pci_dev *pdev)
 	return 0;
 }
 
-#ifdef CONFIG_PM_SLEEP
-static int e1000e_pm_thaw(struct device *dev)
+static int e1000e_pm_prepare(struct device *dev)
 {
-	struct net_device *netdev = pci_get_drvdata(to_pci_dev(dev));
-	struct e1000_adapter *adapter = netdev_priv(netdev);
-
-	e1000e_set_interrupt_capability(adapter);
-	if (netif_running(netdev)) {
-		u32 err = e1000_request_irq(adapter);
-
-		if (err)
-			return err;
-
-		e1000e_up(adapter);
-	}
-
-	netif_device_attach(netdev);
-
-	return 0;
+	return pm_runtime_suspended(dev) &&
+		pm_suspend_via_firmware();
 }
 
 static int e1000e_pm_suspend(struct device *dev)
 {
+	struct net_device *netdev = pci_get_drvdata(to_pci_dev(dev));
+	struct e1000_adapter *adapter = netdev_priv(netdev);
 	struct pci_dev *pdev = to_pci_dev(dev);
+	int rc;
 
 	e1000e_flush_lpic(pdev);
 
 	e1000e_pm_freeze(dev);
 
-	return __e1000_shutdown(pdev, false);
+	rc = __e1000_shutdown(pdev, false);
+	if (!rc) {
+		/* Introduce S0ix implementation */
+		if (adapter->flags2 & FLAG2_ENABLE_S0IX_FLOWS)
+			e1000e_s0ix_entry_flow(adapter);
+	}
+
+	return 0;
 }
 
 static int e1000e_pm_resume(struct device *dev)
 {
+	struct net_device *netdev = pci_get_drvdata(to_pci_dev(dev));
+	struct e1000_adapter *adapter = netdev_priv(netdev);
 	struct pci_dev *pdev = to_pci_dev(dev);
 	int rc;
 
+	/* Introduce S0ix implementation */
+	if (adapter->flags2 & FLAG2_ENABLE_S0IX_FLOWS)
+		e1000e_s0ix_exit_flow(adapter);
+
 	rc = __e1000_resume(pdev);
 	if (rc)
 		return rc;
 
 	return e1000e_pm_thaw(dev);
 }
-#endif /* CONFIG_PM_SLEEP */
 
-static int e1000e_pm_runtime_idle(struct device *dev)
+static __maybe_unused int e1000e_pm_runtime_idle(struct device *dev)
 {
-	struct pci_dev *pdev = to_pci_dev(dev);
-	struct net_device *netdev = pci_get_drvdata(pdev);
+	struct net_device *netdev = dev_get_drvdata(dev);
 	struct e1000_adapter *adapter = netdev_priv(netdev);
 	u16 eee_lp;
 
@@ -6676,6 +7035,8 @@ static int e1000e_pm_runtime_resume(struct device *dev)
 	struct e1000_adapter *adapter = netdev_priv(netdev);
 	int rc;
 
+	pdev->pme_poll = true;
+
 	rc = __e1000_resume(pdev);
 	if (rc)
 		return rc;
@@ -6696,7 +7057,7 @@ static int e1000e_pm_runtime_suspend(struct device *dev)
 		int count = E1000_CHECK_RESET_COUNT;
 
 		while (test_bit(__E1000_RESETTING, &adapter->state) && count--)
-			usleep_range(10000, 20000);
+			usleep_range(10000, 11000);
 
 		WARN_ON(test_bit(__E1000_RESETTING, &adapter->state));
 
@@ -6711,7 +7072,6 @@ static int e1000e_pm_runtime_suspend(struct device *dev)
 
 	return 0;
 }
-#endif /* CONFIG_PM */
 
 static void e1000_shutdown(struct pci_dev *pdev)
 {
@@ -6734,20 +7094,20 @@ static irqreturn_t e1000_intr_msix(int __always_unused irq, void *data)
 
 		vector = 0;
 		msix_irq = adapter->msix_entries[vector].vector;
-		disable_irq(msix_irq);
-		e1000_intr_msix_rx(msix_irq, netdev);
+		if (disable_hardirq(msix_irq))
+			e1000_intr_msix_rx(msix_irq, netdev);
 		enable_irq(msix_irq);
 
 		vector++;
 		msix_irq = adapter->msix_entries[vector].vector;
-		disable_irq(msix_irq);
-		e1000_intr_msix_tx(msix_irq, netdev);
+		if (disable_hardirq(msix_irq))
+			e1000_intr_msix_tx(msix_irq, netdev);
 		enable_irq(msix_irq);
 
 		vector++;
 		msix_irq = adapter->msix_entries[vector].vector;
-		disable_irq(msix_irq);
-		e1000_msix_other(msix_irq, netdev);
+		if (disable_hardirq(msix_irq))
+			e1000_msix_other(msix_irq, netdev);
 		enable_irq(msix_irq);
 	}
 
@@ -6795,19 +7155,14 @@ static void e1000_netpoll(struct net_device *netdev)
 static pci_ers_result_t e1000_io_error_detected(struct pci_dev *pdev,
 						pci_channel_state_t state)
 {
-	struct net_device *netdev = pci_get_drvdata(pdev);
-	struct e1000_adapter *adapter = netdev_priv(netdev);
-
-	netif_device_detach(netdev);
+	e1000e_pm_freeze(&pdev->dev);
 
 	if (state == pci_channel_io_perm_failure)
 		return PCI_ERS_RESULT_DISCONNECT;
 
-	if (netif_running(netdev))
-		e1000e_down(adapter, true);
 	pci_disable_device(pdev);
 
-	/* Request a slot slot reset. */
+	/* Request a slot reset. */
 	return PCI_ERS_RESULT_NEED_RESET;
 }
 
@@ -6852,8 +7207,6 @@ static pci_ers_result_t e1000_io_slot_reset(struct pci_dev *pdev)
 		result = PCI_ERS_RESULT_RECOVERED;
 	}
 
-	pci_cleanup_aer_uncorrect_error_status(pdev);
-
 	return result;
 }
 
@@ -6872,10 +7225,7 @@ static void e1000_io_resume(struct pci_dev *pdev)
 
 	e1000_init_manageability_pt(adapter);
 
-	if (netif_running(netdev))
-		e1000e_up(adapter);
-
-	netif_device_attach(netdev);
+	e1000e_pm_thaw(&pdev->dev);
 
 	/* If the controller has AMT, do not set DRV_LOAD until the interface
 	 * is up.  For all other cases, let the f/w know that the h/w is now
@@ -6904,7 +7254,7 @@ static void e1000_print_device_info(struct e1000_adapter *adapter)
 	ret_val = e1000_read_pba_string_generic(hw, pba_str,
 						E1000_PBANUM_LENGTH);
 	if (ret_val)
-		strlcpy((char *)pba_str, "Unknown", sizeof(pba_str));
+		strscpy((char *)pba_str, "Unknown", sizeof(pba_str));
 	e_info("MAC: %d, PHY: %d, PBA No: %s\n",
 	       hw->mac.type, hw->phy.type, pba_str);
 }
@@ -6983,7 +7333,7 @@ static int e1000_set_features(struct net_device *netdev,
 	else
 		e1000e_reset(adapter);
 
-	return 0;
+	return 1;
 }
 
 static const struct net_device_ops e1000e_netdev_ops = {
@@ -6994,7 +7344,7 @@ static const struct net_device_ops e1000e_netdev_ops = {
 	.ndo_set_rx_mode	= e1000e_set_rx_mode,
 	.ndo_set_mac_address	= e1000_set_mac,
 	.ndo_change_mtu		= e1000_change_mtu,
-	.ndo_do_ioctl		= e1000_ioctl,
+	.ndo_eth_ioctl		= e1000_ioctl,
 	.ndo_tx_timeout		= e1000_tx_timeout,
 	.ndo_validate_addr	= eth_validate_addr,
 
@@ -7003,9 +7353,11 @@ static const struct net_device_ops e1000e_netdev_ops = {
 #ifdef CONFIG_NET_POLL_CONTROLLER
 	.ndo_poll_controller	= e1000_netpoll,
 #endif
-	.ndo_set_features = e1000_set_features,
-	.ndo_fix_features = e1000_fix_features,
+	.ndo_set_features	= e1000_set_features,
+	.ndo_fix_features	= e1000_fix_features,
 	.ndo_features_check	= passthru_features_check,
+	.ndo_hwtstamp_get	= e1000e_hwtstamp_get,
+	.ndo_hwtstamp_set	= e1000e_hwtstamp_set,
 };
 
 /**
@@ -7029,9 +7381,9 @@ static int e1000_probe(struct pci_dev *pdev, const struct pci_device_id *ent)
 	resource_size_t flash_start, flash_len;
 	static int cards_found;
 	u16 aspm_disable_flag = 0;
-	int bars, i, err, pci_using_dac;
 	u16 eeprom_data = 0;
 	u16 eeprom_apme_mask = E1000_EEPROM_APME;
+	int bars, i, err;
 	s32 ret_val = 0;
 
 	if (ei->flags2 & FLAG2_DISABLE_ASPM_L0S)
@@ -7045,17 +7397,11 @@ static int e1000_probe(struct pci_dev *pdev, const struct pci_device_id *ent)
 	if (err)
 		return err;
 
-	pci_using_dac = 0;
 	err = dma_set_mask_and_coherent(&pdev->dev, DMA_BIT_MASK(64));
-	if (!err) {
-		pci_using_dac = 1;
-	} else {
-		err = dma_set_mask_and_coherent(&pdev->dev, DMA_BIT_MASK(32));
-		if (err) {
-			dev_err(&pdev->dev,
-				"No usable DMA configuration, aborting\n");
-			goto err_dma;
-		}
+	if (err) {
+		dev_err(&pdev->dev,
+			"No usable DMA configuration, aborting\n");
+		goto err_dma;
 	}
 
 	bars = pci_select_bars(pdev, IORESOURCE_MEM);
@@ -7064,9 +7410,6 @@ static int e1000_probe(struct pci_dev *pdev, const struct pci_device_id *ent)
 	if (err)
 		goto err_pci_reg;
 
-	/* AER (Advanced Error Reporting) hooks */
-	pci_enable_pcie_error_reporting(pdev);
-
 	pci_set_master(pdev);
 	/* PCI config space info */
 	err = pci_save_state(pdev);
@@ -7122,8 +7465,8 @@ static int e1000_probe(struct pci_dev *pdev, const struct pci_device_id *ent)
 	netdev->netdev_ops = &e1000e_netdev_ops;
 	e1000e_set_ethtool_ops(netdev);
 	netdev->watchdog_timeo = 5 * HZ;
-	netif_napi_add(netdev, &adapter->napi, e1000e_poll, 64);
-	strlcpy(netdev->name, pci_name(pdev), sizeof(netdev->name));
+	netif_napi_add(netdev, &adapter->napi, e1000e_poll);
+	strscpy(netdev->name, pci_name(pdev), sizeof(netdev->name));
 
 	netdev->mem_start = mmio_start;
 	netdev->mem_end = mmio_start + mmio_len;
@@ -7175,6 +7518,32 @@ static int e1000_probe(struct pci_dev *pdev, const struct pci_device_id *ent)
 			    NETIF_F_RXCSUM |
 			    NETIF_F_HW_CSUM);
 
+	/* disable TSO for pcie and 10/100 speeds to avoid
+	 * some hardware issues and for i219 to fix transfer
+	 * speed being capped at 60%
+	 */
+	if (!(adapter->flags & FLAG_TSO_FORCE)) {
+		switch (adapter->link_speed) {
+		case SPEED_10:
+		case SPEED_100:
+			e_info("10/100 speed: disabling TSO\n");
+			netdev->features &= ~NETIF_F_TSO;
+			netdev->features &= ~NETIF_F_TSO6;
+			break;
+		case SPEED_1000:
+			netdev->features |= NETIF_F_TSO;
+			netdev->features |= NETIF_F_TSO6;
+			break;
+		default:
+			/* oops */
+			break;
+		}
+		if (hw->mac.type == e1000_pch_spt) {
+			netdev->features &= ~NETIF_F_TSO;
+			netdev->features &= ~NETIF_F_TSO6;
+		}
+	}
+
 	/* Set user-changeable features (subset of all device features) */
 	netdev->hw_features = netdev->features;
 	netdev->hw_features |= NETIF_F_RXFCS;
@@ -7191,10 +7560,8 @@ static int e1000_probe(struct pci_dev *pdev, const struct pci_device_id *ent)
 
 	netdev->priv_flags |= IFF_UNICAST_FLT;
 
-	if (pci_using_dac) {
-		netdev->features |= NETIF_F_HIGHDMA;
-		netdev->vlan_features |= NETIF_F_HIGHDMA;
-	}
+	netdev->features |= NETIF_F_HIGHDMA;
+	netdev->vlan_features |= NETIF_F_HIGHDMA;
 
 	/* MTU range: 68 - max_hw_frame_size */
 	netdev->min_mtu = ETH_MIN_MTU;
@@ -7229,7 +7596,7 @@ static int e1000_probe(struct pci_dev *pdev, const struct pci_device_id *ent)
 		dev_err(&pdev->dev,
 			"NVM Read Error while reading MAC address\n");
 
-	memcpy(netdev->dev_addr, adapter->hw.mac.addr, netdev->addr_len);
+	eth_hw_addr_set(netdev, adapter->hw.mac.addr);
 
 	if (!is_valid_ether_addr(netdev->dev_addr)) {
 		dev_err(&pdev->dev, "Invalid MAC Address: %pM\n",
@@ -7238,13 +7605,8 @@ static int e1000_probe(struct pci_dev *pdev, const struct pci_device_id *ent)
 		goto err_eeprom;
 	}
 
-	init_timer(&adapter->watchdog_timer);
-	adapter->watchdog_timer.function = e1000_watchdog;
-	adapter->watchdog_timer.data = (unsigned long)adapter;
-
-	init_timer(&adapter->phy_info_timer);
-	adapter->phy_info_timer.function = e1000_update_phy_info;
-	adapter->phy_info_timer.data = (unsigned long)adapter;
+	timer_setup(&adapter->watchdog_timer, e1000_watchdog, 0);
+	timer_setup(&adapter->phy_info_timer, e1000_update_phy_info, 0);
 
 	INIT_WORK(&adapter->reset_task, e1000_reset_task);
 	INIT_WORK(&adapter->watchdog_task, e1000_watchdog_task);
@@ -7323,7 +7685,10 @@ static int e1000_probe(struct pci_dev *pdev, const struct pci_device_id *ent)
 	if (!(adapter->flags & FLAG_HAS_AMT))
 		e1000e_get_hw_control(adapter);
 
-	strlcpy(netdev->name, "eth%d", sizeof(netdev->name));
+	if (hw->mac.type >= e1000_pch_cnp)
+		adapter->flags2 |= FLAG2_ENABLE_S0IX_FLOWS;
+
+	strscpy(netdev->name, "eth%d", sizeof(netdev->name));
 	err = register_netdev(netdev);
 	if (err)
 		goto err_register;
@@ -7333,6 +7698,8 @@ static int e1000_probe(struct pci_dev *pdev, const struct pci_device_id *ent)
 
 	e1000_print_device_info(adapter);
 
+	dev_pm_set_driver_flags(&pdev->dev, DPM_FLAG_SMART_PREPARE);
+
 	if (pci_dev_run_wake(pdev))
 		pm_runtime_put_noidle(&pdev->dev);
 
@@ -7368,7 +7735,7 @@ err_dma:
  * @pdev: PCI device information struct
  *
  * e1000_remove is called by the PCI subsystem to alert the driver
- * that it should release a PCI device.  The could be caused by a
+ * that it should release a PCI device.  This could be caused by a
  * Hot-Plug event, or because the driver is going to be removed from
  * memory.
  **/
@@ -7376,17 +7743,15 @@ static void e1000_remove(struct pci_dev *pdev)
 {
 	struct net_device *netdev = pci_get_drvdata(pdev);
 	struct e1000_adapter *adapter = netdev_priv(netdev);
-	bool down = test_bit(__E1000_DOWN, &adapter->state);
 
 	e1000e_ptp_remove(adapter);
 
 	/* The timers may be rescheduled, so explicitly disable them
 	 * from being rescheduled.
 	 */
-	if (!down)
-		set_bit(__E1000_DOWN, &adapter->state);
-	del_timer_sync(&adapter->watchdog_timer);
-	del_timer_sync(&adapter->phy_info_timer);
+	set_bit(__E1000_DOWN, &adapter->state);
+	timer_delete_sync(&adapter->watchdog_timer);
+	timer_delete_sync(&adapter->phy_info_timer);
 
 	cancel_work_sync(&adapter->reset_task);
 	cancel_work_sync(&adapter->watchdog_task);
@@ -7397,14 +7762,11 @@ static void e1000_remove(struct pci_dev *pdev)
 	if (adapter->flags & FLAG_HAS_HW_TIMESTAMP) {
 		cancel_work_sync(&adapter->tx_hwtstamp_work);
 		if (adapter->tx_hwtstamp_skb) {
-			dev_kfree_skb_any(adapter->tx_hwtstamp_skb);
+			dev_consume_skb_any(adapter->tx_hwtstamp_skb);
 			adapter->tx_hwtstamp_skb = NULL;
 		}
 	}
 
-	/* Don't lie to e1000_close() down the road. */
-	if (!down)
-		clear_bit(__E1000_DOWN, &adapter->state);
 	unregister_netdev(netdev);
 
 	if (pci_dev_run_wake(pdev))
@@ -7427,9 +7789,6 @@ static void e1000_remove(struct pci_dev *pdev)
 
 	free_netdev(netdev);
 
-	/* AER disable */
-	pci_disable_pcie_error_reporting(pdev);
-
 	pci_disable_device(pdev);
 }
 
@@ -7530,22 +7889,63 @@ static const struct pci_device_id e1000_pci_tbl[] = {
 	{ PCI_VDEVICE(INTEL, E1000_DEV_ID_PCH_CNP_I219_V6), board_pch_cnp },
 	{ PCI_VDEVICE(INTEL, E1000_DEV_ID_PCH_CNP_I219_LM7), board_pch_cnp },
 	{ PCI_VDEVICE(INTEL, E1000_DEV_ID_PCH_CNP_I219_V7), board_pch_cnp },
+	{ PCI_VDEVICE(INTEL, E1000_DEV_ID_PCH_ICP_I219_LM8), board_pch_cnp },
+	{ PCI_VDEVICE(INTEL, E1000_DEV_ID_PCH_ICP_I219_V8), board_pch_cnp },
+	{ PCI_VDEVICE(INTEL, E1000_DEV_ID_PCH_ICP_I219_LM9), board_pch_cnp },
+	{ PCI_VDEVICE(INTEL, E1000_DEV_ID_PCH_ICP_I219_V9), board_pch_cnp },
+	{ PCI_VDEVICE(INTEL, E1000_DEV_ID_PCH_CMP_I219_LM10), board_pch_cnp },
+	{ PCI_VDEVICE(INTEL, E1000_DEV_ID_PCH_CMP_I219_V10), board_pch_cnp },
+	{ PCI_VDEVICE(INTEL, E1000_DEV_ID_PCH_CMP_I219_LM11), board_pch_cnp },
+	{ PCI_VDEVICE(INTEL, E1000_DEV_ID_PCH_CMP_I219_V11), board_pch_cnp },
+	{ PCI_VDEVICE(INTEL, E1000_DEV_ID_PCH_CMP_I219_LM12), board_pch_spt },
+	{ PCI_VDEVICE(INTEL, E1000_DEV_ID_PCH_CMP_I219_V12), board_pch_spt },
+	{ PCI_VDEVICE(INTEL, E1000_DEV_ID_PCH_TGP_I219_LM13), board_pch_tgp },
+	{ PCI_VDEVICE(INTEL, E1000_DEV_ID_PCH_TGP_I219_V13), board_pch_tgp },
+	{ PCI_VDEVICE(INTEL, E1000_DEV_ID_PCH_TGP_I219_LM14), board_pch_tgp },
+	{ PCI_VDEVICE(INTEL, E1000_DEV_ID_PCH_TGP_I219_V14), board_pch_tgp },
+	{ PCI_VDEVICE(INTEL, E1000_DEV_ID_PCH_TGP_I219_LM15), board_pch_tgp },
+	{ PCI_VDEVICE(INTEL, E1000_DEV_ID_PCH_TGP_I219_V15), board_pch_tgp },
+	{ PCI_VDEVICE(INTEL, E1000_DEV_ID_PCH_RPL_I219_LM23), board_pch_adp },
+	{ PCI_VDEVICE(INTEL, E1000_DEV_ID_PCH_RPL_I219_V23), board_pch_adp },
+	{ PCI_VDEVICE(INTEL, E1000_DEV_ID_PCH_ADP_I219_LM16), board_pch_adp },
+	{ PCI_VDEVICE(INTEL, E1000_DEV_ID_PCH_ADP_I219_V16), board_pch_adp },
+	{ PCI_VDEVICE(INTEL, E1000_DEV_ID_PCH_ADP_I219_LM17), board_pch_adp },
+	{ PCI_VDEVICE(INTEL, E1000_DEV_ID_PCH_ADP_I219_V17), board_pch_adp },
+	{ PCI_VDEVICE(INTEL, E1000_DEV_ID_PCH_RPL_I219_LM22), board_pch_adp },
+	{ PCI_VDEVICE(INTEL, E1000_DEV_ID_PCH_RPL_I219_V22), board_pch_adp },
+	{ PCI_VDEVICE(INTEL, E1000_DEV_ID_PCH_ADP_I219_LM19), board_pch_adp },
+	{ PCI_VDEVICE(INTEL, E1000_DEV_ID_PCH_ADP_I219_V19), board_pch_adp },
+	{ PCI_VDEVICE(INTEL, E1000_DEV_ID_PCH_MTP_I219_LM18), board_pch_mtp },
+	{ PCI_VDEVICE(INTEL, E1000_DEV_ID_PCH_MTP_I219_V18), board_pch_mtp },
+	{ PCI_VDEVICE(INTEL, E1000_DEV_ID_PCH_LNP_I219_LM20), board_pch_mtp },
+	{ PCI_VDEVICE(INTEL, E1000_DEV_ID_PCH_LNP_I219_V20), board_pch_mtp },
+	{ PCI_VDEVICE(INTEL, E1000_DEV_ID_PCH_LNP_I219_LM21), board_pch_mtp },
+	{ PCI_VDEVICE(INTEL, E1000_DEV_ID_PCH_LNP_I219_V21), board_pch_mtp },
+	{ PCI_VDEVICE(INTEL, E1000_DEV_ID_PCH_ARL_I219_LM24), board_pch_mtp },
+	{ PCI_VDEVICE(INTEL, E1000_DEV_ID_PCH_ARL_I219_V24), board_pch_mtp },
+	{ PCI_VDEVICE(INTEL, E1000_DEV_ID_PCH_PTP_I219_LM25), board_pch_mtp },
+	{ PCI_VDEVICE(INTEL, E1000_DEV_ID_PCH_PTP_I219_V25), board_pch_mtp },
+	{ PCI_VDEVICE(INTEL, E1000_DEV_ID_PCH_PTP_I219_LM26), board_pch_mtp },
+	{ PCI_VDEVICE(INTEL, E1000_DEV_ID_PCH_PTP_I219_V26), board_pch_mtp },
+	{ PCI_VDEVICE(INTEL, E1000_DEV_ID_PCH_PTP_I219_LM27), board_pch_mtp },
+	{ PCI_VDEVICE(INTEL, E1000_DEV_ID_PCH_PTP_I219_V27), board_pch_mtp },
+	{ PCI_VDEVICE(INTEL, E1000_DEV_ID_PCH_NVL_I219_LM29), board_pch_mtp },
+	{ PCI_VDEVICE(INTEL, E1000_DEV_ID_PCH_NVL_I219_V29), board_pch_mtp },
 
 	{ 0, 0, 0, 0, 0, 0, 0 }	/* terminate list */
 };
 MODULE_DEVICE_TABLE(pci, e1000_pci_tbl);
 
-static const struct dev_pm_ops e1000_pm_ops = {
-#ifdef CONFIG_PM_SLEEP
+static const struct dev_pm_ops e1000e_pm_ops = {
+	.prepare	= e1000e_pm_prepare,
 	.suspend	= e1000e_pm_suspend,
 	.resume		= e1000e_pm_resume,
 	.freeze		= e1000e_pm_freeze,
 	.thaw		= e1000e_pm_thaw,
 	.poweroff	= e1000e_pm_suspend,
 	.restore	= e1000e_pm_resume,
-#endif
-	SET_RUNTIME_PM_OPS(e1000e_pm_runtime_suspend, e1000e_pm_runtime_resume,
-			   e1000e_pm_runtime_idle)
+	RUNTIME_PM_OPS(e1000e_pm_runtime_suspend, e1000e_pm_runtime_resume,
+		       e1000e_pm_runtime_idle)
 };
 
 /* PCI Device API Driver */
@@ -7554,9 +7954,7 @@ static struct pci_driver e1000_driver = {
 	.id_table = e1000_pci_tbl,
 	.probe    = e1000_probe,
 	.remove   = e1000_remove,
-	.driver   = {
-		.pm = &e1000_pm_ops,
-	},
+	.driver.pm = pm_ptr(&e1000e_pm_ops),
 	.shutdown = e1000_shutdown,
 	.err_handler = &e1000_err_handler
 };
@@ -7569,8 +7967,7 @@ static struct pci_driver e1000_driver = {
  **/
 static int __init e1000_init_module(void)
 {
-	pr_info("Intel(R) PRO/1000 Network Driver - %s\n",
-		e1000e_driver_version);
+	pr_info("Intel(R) PRO/1000 Network Driver\n");
 	pr_info("Copyright(c) 1999 - 2015 Intel Corporation.\n");
 
 	return pci_register_driver(&e1000_driver);
@@ -7589,9 +7986,7 @@ static void __exit e1000_exit_module(void)
 }
 module_exit(e1000_exit_module);
 
-MODULE_AUTHOR("Intel Corporation, <linux.nics@intel.com>");
 MODULE_DESCRIPTION("Intel(R) PRO/1000 Network Driver");
-MODULE_LICENSE("GPL");
-MODULE_VERSION(DRV_VERSION);
+MODULE_LICENSE("GPL v2");
 
 /* netdev.c */
diff --git a/drivers/net/ethernet/intel/e1000e/nvm.c b/drivers/net/ethernet/intel/e1000e/nvm.c
index 2efd80dfd88e..4bde1c9de1b9 100644
--- a/drivers/net/ethernet/intel/e1000e/nvm.c
+++ b/drivers/net/ethernet/intel/e1000e/nvm.c
@@ -1,23 +1,5 @@
-/* Intel PRO/1000 Linux driver
- * Copyright(c) 1999 - 2015 Intel Corporation.
- *
- * This program is free software; you can redistribute it and/or modify it
- * under the terms and conditions of the GNU General Public License,
- * version 2, as published by the Free Software Foundation.
- *
- * This program is distributed in the hope it will be useful, but WITHOUT
- * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
- * FITNESS FOR A PARTICULAR PURPOSE.  See the GNU General Public License for
- * more details.
- *
- * The full GNU General Public License is included in this distribution in
- * the file called "COPYING".
- *
- * Contact Information:
- * Linux NICS <linux.nics@intel.com>
- * e1000-devel Mailing List <e1000-devel@lists.sourceforge.net>
- * Intel Corporation, 5200 N.E. Elam Young Parkway, Hillsboro, OR 97124-6497
- */
+// SPDX-License-Identifier: GPL-2.0
+/* Copyright(c) 1999 - 2018 Intel Corporation. */
 
 #include "e1000.h"
 
@@ -410,7 +392,7 @@ s32 e1000e_write_nvm_spi(struct e1000_hw *hw, u16 offset, u16 words, u16 *data)
 				break;
 			}
 		}
-		usleep_range(10000, 20000);
+		usleep_range(10000, 11000);
 		nvm->ops.release(hw);
 	}
 
@@ -576,7 +558,13 @@ s32 e1000e_validate_nvm_checksum_generic(struct e1000_hw *hw)
 		checksum += nvm_data;
 	}
 
-	if (checksum != (u16)NVM_SUM) {
+	if (hw->mac.type == e1000_pch_tgp &&
+	    nvm_data == NVM_CHECKSUM_UNINITIALIZED) {
+		e_dbg("Uninitialized NVM Checksum on TGP platform - ignoring\n");
+		return 0;
+	}
+
+	if (checksum != NVM_SUM) {
 		e_dbg("NVM Checksum Invalid\n");
 		return -E1000_ERR_NVM;
 	}
@@ -606,7 +594,7 @@ s32 e1000e_update_nvm_checksum_generic(struct e1000_hw *hw)
 		}
 		checksum += nvm_data;
 	}
-	checksum = (u16)NVM_SUM - checksum;
+	checksum = NVM_SUM - checksum;
 	ret_val = e1000_write_nvm(hw, NVM_CHECKSUM_REG, 1, &checksum);
 	if (ret_val)
 		e_dbg("NVM Write Error while updating checksum.\n");
diff --git a/drivers/net/ethernet/intel/e1000e/nvm.h b/drivers/net/ethernet/intel/e1000e/nvm.h
index 5d46967e0d1f..6a30dfea4117 100644
--- a/drivers/net/ethernet/intel/e1000e/nvm.h
+++ b/drivers/net/ethernet/intel/e1000e/nvm.h
@@ -1,23 +1,5 @@
-/* Intel PRO/1000 Linux driver
- * Copyright(c) 1999 - 2015 Intel Corporation.
- *
- * This program is free software; you can redistribute it and/or modify it
- * under the terms and conditions of the GNU General Public License,
- * version 2, as published by the Free Software Foundation.
- *
- * This program is distributed in the hope it will be useful, but WITHOUT
- * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
- * FITNESS FOR A PARTICULAR PURPOSE.  See the GNU General Public License for
- * more details.
- *
- * The full GNU General Public License is included in this distribution in
- * the file called "COPYING".
- *
- * Contact Information:
- * Linux NICS <linux.nics@intel.com>
- * e1000-devel Mailing List <e1000-devel@lists.sourceforge.net>
- * Intel Corporation, 5200 N.E. Elam Young Parkway, Hillsboro, OR 97124-6497
- */
+/* SPDX-License-Identifier: GPL-2.0 */
+/* Copyright(c) 1999 - 2018 Intel Corporation. */
 
 #ifndef _E1000E_NVM_H_
 #define _E1000E_NVM_H_
diff --git a/drivers/net/ethernet/intel/e1000e/param.c b/drivers/net/ethernet/intel/e1000e/param.c
index 6d8c39abee16..3132d8f2f207 100644
--- a/drivers/net/ethernet/intel/e1000e/param.c
+++ b/drivers/net/ethernet/intel/e1000e/param.c
@@ -1,23 +1,5 @@
-/* Intel PRO/1000 Linux driver
- * Copyright(c) 1999 - 2015 Intel Corporation.
- *
- * This program is free software; you can redistribute it and/or modify it
- * under the terms and conditions of the GNU General Public License,
- * version 2, as published by the Free Software Foundation.
- *
- * This program is distributed in the hope it will be useful, but WITHOUT
- * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
- * FITNESS FOR A PARTICULAR PURPOSE.  See the GNU General Public License for
- * more details.
- *
- * The full GNU General Public License is included in this distribution in
- * the file called "COPYING".
- *
- * Contact Information:
- * Linux NICS <linux.nics@intel.com>
- * e1000-devel Mailing List <e1000-devel@lists.sourceforge.net>
- * Intel Corporation, 5200 N.E. Elam Young Parkway, Hillsboro, OR 97124-6497
- */
+// SPDX-License-Identifier: GPL-2.0
+/* Copyright(c) 1999 - 2018 Intel Corporation. */
 
 #include <linux/netdevice.h>
 #include <linux/module.h>
@@ -73,17 +55,25 @@ E1000_PARAM(TxAbsIntDelay, "Transmit Absolute Interrupt Delay");
 /* Receive Interrupt Delay in units of 1.024 microseconds
  * hardware will likely hang if you set this to anything but zero.
  *
+ * Burst variant is used as default if device has FLAG2_DMA_BURST.
+ *
  * Valid Range: 0-65535
  */
 E1000_PARAM(RxIntDelay, "Receive Interrupt Delay");
+#define DEFAULT_RDTR	0
+#define BURST_RDTR	0x20
 #define MAX_RXDELAY 0xFFFF
 #define MIN_RXDELAY 0
 
 /* Receive Absolute Interrupt Delay in units of 1.024 microseconds
  *
+ * Burst variant is used as default if device has FLAG2_DMA_BURST.
+ *
  * Valid Range: 0-65535
  */
 E1000_PARAM(RxAbsIntDelay, "Receive Absolute Interrupt Delay");
+#define DEFAULT_RADV	8
+#define BURST_RADV	0x20
 #define MAX_RXABSDELAY 0xFFFF
 #define MIN_RXABSDELAY 0
 
@@ -111,8 +101,6 @@ E1000_PARAM(InterruptThrottleRate, "Interrupt Throttling Rate");
  * demoted to the most advanced interrupt mode available.
  */
 E1000_PARAM(IntMode, "Interrupt Mode");
-#define MAX_INTMODE	2
-#define MIN_INTMODE	0
 
 /* Enable Smart Power Down of the PHY
  *
@@ -297,6 +285,9 @@ void e1000e_check_options(struct e1000_adapter *adapter)
 					 .max = MAX_RXDELAY } }
 		};
 
+		if (adapter->flags2 & FLAG2_DMA_BURST)
+			opt.def = BURST_RDTR;
+
 		if (num_RxIntDelay > bd) {
 			adapter->rx_int_delay = RxIntDelay[bd];
 			e1000_validate_option(&adapter->rx_int_delay, &opt,
@@ -307,7 +298,7 @@ void e1000e_check_options(struct e1000_adapter *adapter)
 	}
 	/* Receive Absolute Interrupt Delay */
 	{
-		static const struct e1000_option opt = {
+		static struct e1000_option opt = {
 			.type = range_option,
 			.name = "Receive Absolute Interrupt Delay",
 			.err  = "using default of "
@@ -317,6 +308,9 @@ void e1000e_check_options(struct e1000_adapter *adapter)
 					 .max = MAX_RXABSDELAY } }
 		};
 
+		if (adapter->flags2 & FLAG2_DMA_BURST)
+			opt.def = BURST_RADV;
+
 		if (num_RxAbsIntDelay > bd) {
 			adapter->rx_abs_int_delay = RxAbsIntDelay[bd];
 			e1000_validate_option(&adapter->rx_abs_int_delay, &opt,
@@ -379,7 +373,7 @@ void e1000e_check_options(struct e1000_adapter *adapter)
 				 "%s Invalid mode - setting default\n",
 				 opt.name);
 			adapter->itr_setting = opt.def;
-			/* fall-through */
+			fallthrough;
 		case 3:
 			dev_info(&adapter->pdev->dev,
 				 "%s set to dynamic conservative mode\n",
diff --git a/drivers/net/ethernet/intel/e1000e/phy.c b/drivers/net/ethernet/intel/e1000e/phy.c
index d78d47b41a71..f7ae0e0aa4a4 100644
--- a/drivers/net/ethernet/intel/e1000e/phy.c
+++ b/drivers/net/ethernet/intel/e1000e/phy.c
@@ -1,25 +1,8 @@
-/* Intel PRO/1000 Linux driver
- * Copyright(c) 1999 - 2015 Intel Corporation.
- *
- * This program is free software; you can redistribute it and/or modify it
- * under the terms and conditions of the GNU General Public License,
- * version 2, as published by the Free Software Foundation.
- *
- * This program is distributed in the hope it will be useful, but WITHOUT
- * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
- * FITNESS FOR A PARTICULAR PURPOSE.  See the GNU General Public License for
- * more details.
- *
- * The full GNU General Public License is included in this distribution in
- * the file called "COPYING".
- *
- * Contact Information:
- * Linux NICS <linux.nics@intel.com>
- * e1000-devel Mailing List <e1000-devel@lists.sourceforge.net>
- * Intel Corporation, 5200 N.E. Elam Young Parkway, Hillsboro, OR 97124-6497
- */
+// SPDX-License-Identifier: GPL-2.0
+/* Copyright(c) 1999 - 2018 Intel Corporation. */
 
 #include "e1000.h"
+#include <linux/ethtool.h>
 
 static s32 e1000_wait_autoneg(struct e1000_hw *hw);
 static s32 e1000_access_phy_wakeup_reg_bm(struct e1000_hw *hw, u32 offset,
@@ -124,6 +107,16 @@ s32 e1000e_phy_reset_dsp(struct e1000_hw *hw)
 	return e1e_wphy(hw, M88E1000_PHY_GEN_CONTROL, 0);
 }
 
+void e1000e_disable_phy_retry(struct e1000_hw *hw)
+{
+	hw->phy.retry_enabled = false;
+}
+
+void e1000e_enable_phy_retry(struct e1000_hw *hw)
+{
+	hw->phy.retry_enabled = true;
+}
+
 /**
  *  e1000e_read_phy_reg_mdic - Read MDI control register
  *  @hw: pointer to the HW structure
@@ -135,57 +128,73 @@ s32 e1000e_phy_reset_dsp(struct e1000_hw *hw)
  **/
 s32 e1000e_read_phy_reg_mdic(struct e1000_hw *hw, u32 offset, u16 *data)
 {
+	u32 i, mdic = 0, retry_counter, retry_max;
 	struct e1000_phy_info *phy = &hw->phy;
-	u32 i, mdic = 0;
+	bool success;
 
 	if (offset > MAX_PHY_REG_ADDRESS) {
 		e_dbg("PHY Address %d is out of range\n", offset);
 		return -E1000_ERR_PARAM;
 	}
 
+	retry_max = phy->retry_enabled ? phy->retry_count : 0;
+
 	/* Set up Op-code, Phy Address, and register offset in the MDI
 	 * Control register.  The MAC will take care of interfacing with the
 	 * PHY to retrieve the desired data.
 	 */
-	mdic = ((offset << E1000_MDIC_REG_SHIFT) |
-		(phy->addr << E1000_MDIC_PHY_SHIFT) |
-		(E1000_MDIC_OP_READ));
+	for (retry_counter = 0; retry_counter <= retry_max; retry_counter++) {
+		success = true;
 
-	ew32(MDIC, mdic);
+		mdic = ((offset << E1000_MDIC_REG_SHIFT) |
+			(phy->addr << E1000_MDIC_PHY_SHIFT) |
+			(E1000_MDIC_OP_READ));
 
-	/* Poll the ready bit to see if the MDI read completed
-	 * Increasing the time out as testing showed failures with
-	 * the lower time out
-	 */
-	for (i = 0; i < (E1000_GEN_POLL_TIMEOUT * 3); i++) {
-		udelay(50);
-		mdic = er32(MDIC);
-		if (mdic & E1000_MDIC_READY)
-			break;
-	}
-	if (!(mdic & E1000_MDIC_READY)) {
-		e_dbg("MDI Read did not complete\n");
-		return -E1000_ERR_PHY;
-	}
-	if (mdic & E1000_MDIC_ERROR) {
-		e_dbg("MDI Error\n");
-		return -E1000_ERR_PHY;
-	}
-	if (((mdic & E1000_MDIC_REG_MASK) >> E1000_MDIC_REG_SHIFT) != offset) {
-		e_dbg("MDI Read offset error - requested %d, returned %d\n",
-		      offset,
-		      (mdic & E1000_MDIC_REG_MASK) >> E1000_MDIC_REG_SHIFT);
-		return -E1000_ERR_PHY;
-	}
-	*data = (u16)mdic;
+		ew32(MDIC, mdic);
 
-	/* Allow some time after each MDIC transaction to avoid
-	 * reading duplicate data in the next MDIC transaction.
-	 */
-	if (hw->mac.type == e1000_pch2lan)
-		udelay(100);
+		/* Poll the ready bit to see if the MDI read completed
+		 * Increasing the time out as testing showed failures with
+		 * the lower time out
+		 */
+		for (i = 0; i < (E1000_GEN_POLL_TIMEOUT * 3); i++) {
+			udelay(50);
+			mdic = er32(MDIC);
+			if (mdic & E1000_MDIC_READY)
+				break;
+		}
+		if (!(mdic & E1000_MDIC_READY)) {
+			e_dbg("MDI Read PHY Reg Address %d did not complete\n",
+			      offset);
+			success = false;
+		}
+		if (mdic & E1000_MDIC_ERROR) {
+			e_dbg("MDI Read PHY Reg Address %d Error\n", offset);
+			success = false;
+		}
+		if (FIELD_GET(E1000_MDIC_REG_MASK, mdic) != offset) {
+			e_dbg("MDI Read offset error - requested %d, returned %d\n",
+			      offset, FIELD_GET(E1000_MDIC_REG_MASK, mdic));
+			success = false;
+		}
 
-	return 0;
+		/* Allow some time after each MDIC transaction to avoid
+		 * reading duplicate data in the next MDIC transaction.
+		 */
+		if (hw->mac.type == e1000_pch2lan)
+			udelay(100);
+
+		if (success) {
+			*data = (u16)mdic;
+			return 0;
+		}
+
+		if (retry_counter != retry_max) {
+			e_dbg("Perform retry on PHY transaction...\n");
+			mdelay(10);
+		}
+	}
+
+	return -E1000_ERR_PHY;
 }
 
 /**
@@ -198,57 +207,72 @@ s32 e1000e_read_phy_reg_mdic(struct e1000_hw *hw, u32 offset, u16 *data)
  **/
 s32 e1000e_write_phy_reg_mdic(struct e1000_hw *hw, u32 offset, u16 data)
 {
+	u32 i, mdic = 0, retry_counter, retry_max;
 	struct e1000_phy_info *phy = &hw->phy;
-	u32 i, mdic = 0;
+	bool success;
 
 	if (offset > MAX_PHY_REG_ADDRESS) {
 		e_dbg("PHY Address %d is out of range\n", offset);
 		return -E1000_ERR_PARAM;
 	}
 
+	retry_max = phy->retry_enabled ? phy->retry_count : 0;
+
 	/* Set up Op-code, Phy Address, and register offset in the MDI
 	 * Control register.  The MAC will take care of interfacing with the
 	 * PHY to retrieve the desired data.
 	 */
-	mdic = (((u32)data) |
-		(offset << E1000_MDIC_REG_SHIFT) |
-		(phy->addr << E1000_MDIC_PHY_SHIFT) |
-		(E1000_MDIC_OP_WRITE));
+	for (retry_counter = 0; retry_counter <= retry_max; retry_counter++) {
+		success = true;
 
-	ew32(MDIC, mdic);
+		mdic = (((u32)data) |
+			(offset << E1000_MDIC_REG_SHIFT) |
+			(phy->addr << E1000_MDIC_PHY_SHIFT) |
+			(E1000_MDIC_OP_WRITE));
 
-	/* Poll the ready bit to see if the MDI read completed
-	 * Increasing the time out as testing showed failures with
-	 * the lower time out
-	 */
-	for (i = 0; i < (E1000_GEN_POLL_TIMEOUT * 3); i++) {
-		udelay(50);
-		mdic = er32(MDIC);
-		if (mdic & E1000_MDIC_READY)
-			break;
-	}
-	if (!(mdic & E1000_MDIC_READY)) {
-		e_dbg("MDI Write did not complete\n");
-		return -E1000_ERR_PHY;
-	}
-	if (mdic & E1000_MDIC_ERROR) {
-		e_dbg("MDI Error\n");
-		return -E1000_ERR_PHY;
-	}
-	if (((mdic & E1000_MDIC_REG_MASK) >> E1000_MDIC_REG_SHIFT) != offset) {
-		e_dbg("MDI Write offset error - requested %d, returned %d\n",
-		      offset,
-		      (mdic & E1000_MDIC_REG_MASK) >> E1000_MDIC_REG_SHIFT);
-		return -E1000_ERR_PHY;
-	}
+		ew32(MDIC, mdic);
 
-	/* Allow some time after each MDIC transaction to avoid
-	 * reading duplicate data in the next MDIC transaction.
-	 */
-	if (hw->mac.type == e1000_pch2lan)
-		udelay(100);
+		/* Poll the ready bit to see if the MDI read completed
+		 * Increasing the time out as testing showed failures with
+		 * the lower time out
+		 */
+		for (i = 0; i < (E1000_GEN_POLL_TIMEOUT * 3); i++) {
+			udelay(50);
+			mdic = er32(MDIC);
+			if (mdic & E1000_MDIC_READY)
+				break;
+		}
+		if (!(mdic & E1000_MDIC_READY)) {
+			e_dbg("MDI Write PHY Reg Address %d did not complete\n",
+			      offset);
+			success = false;
+		}
+		if (mdic & E1000_MDIC_ERROR) {
+			e_dbg("MDI Write PHY Reg Address %d Error\n", offset);
+			success = false;
+		}
+		if (FIELD_GET(E1000_MDIC_REG_MASK, mdic) != offset) {
+			e_dbg("MDI Write offset error - requested %d, returned %d\n",
+			      offset, FIELD_GET(E1000_MDIC_REG_MASK, mdic));
+			success = false;
+		}
 
-	return 0;
+		/* Allow some time after each MDIC transaction to avoid
+		 * reading duplicate data in the next MDIC transaction.
+		 */
+		if (hw->mac.type == e1000_pch2lan)
+			udelay(100);
+
+		if (success)
+			return 0;
+
+		if (retry_counter != retry_max) {
+			e_dbg("Perform retry on PHY transaction...\n");
+			mdelay(10);
+		}
+	}
+
+	return -E1000_ERR_PHY;
 }
 
 /**
@@ -389,7 +413,7 @@ s32 e1000e_read_phy_reg_igp_locked(struct e1000_hw *hw, u32 offset, u16 *data)
 }
 
 /**
- *  e1000e_write_phy_reg_igp - Write igp PHY register
+ *  __e1000e_write_phy_reg_igp - Write igp PHY register
  *  @hw: pointer to the HW structure
  *  @offset: register offset to write to
  *  @data: data to write at register offset
@@ -480,8 +504,8 @@ static s32 __e1000_read_kmrn_reg(struct e1000_hw *hw, u32 offset, u16 *data,
 			return ret_val;
 	}
 
-	kmrnctrlsta = ((offset << E1000_KMRNCTRLSTA_OFFSET_SHIFT) &
-		       E1000_KMRNCTRLSTA_OFFSET) | E1000_KMRNCTRLSTA_REN;
+	kmrnctrlsta = FIELD_PREP(E1000_KMRNCTRLSTA_OFFSET, offset) |
+		      E1000_KMRNCTRLSTA_REN;
 	ew32(KMRNCTRLSTA, kmrnctrlsta);
 	e1e_flush();
 
@@ -553,8 +577,7 @@ static s32 __e1000_write_kmrn_reg(struct e1000_hw *hw, u32 offset, u16 data,
 			return ret_val;
 	}
 
-	kmrnctrlsta = ((offset << E1000_KMRNCTRLSTA_OFFSET_SHIFT) &
-		       E1000_KMRNCTRLSTA_OFFSET) | data;
+	kmrnctrlsta = FIELD_PREP(E1000_KMRNCTRLSTA_OFFSET, offset) | data;
 	ew32(KMRNCTRLSTA, kmrnctrlsta);
 	e1e_flush();
 
@@ -625,7 +648,7 @@ static s32 e1000_set_master_slave_mode(struct e1000_hw *hw)
 		break;
 	case e1000_ms_auto:
 		phy_data &= ~CTL1000_ENABLE_MASTER;
-		/* fall-through */
+		fallthrough;
 	default:
 		break;
 	}
@@ -1029,6 +1052,8 @@ static s32 e1000_phy_setup_autoneg(struct e1000_hw *hw)
 		 */
 		mii_autoneg_adv_reg &=
 		    ~(ADVERTISE_PAUSE_ASYM | ADVERTISE_PAUSE_CAP);
+		phy->autoneg_advertised &=
+		    ~(ADVERTISED_Pause | ADVERTISED_Asym_Pause);
 		break;
 	case e1000_fc_rx_pause:
 		/* Rx Flow control is enabled, and Tx Flow control is
@@ -1042,6 +1067,8 @@ static s32 e1000_phy_setup_autoneg(struct e1000_hw *hw)
 		 */
 		mii_autoneg_adv_reg |=
 		    (ADVERTISE_PAUSE_ASYM | ADVERTISE_PAUSE_CAP);
+		phy->autoneg_advertised |=
+		    (ADVERTISED_Pause | ADVERTISED_Asym_Pause);
 		break;
 	case e1000_fc_tx_pause:
 		/* Tx Flow control is enabled, and Rx Flow control is
@@ -1049,6 +1076,8 @@ static s32 e1000_phy_setup_autoneg(struct e1000_hw *hw)
 		 */
 		mii_autoneg_adv_reg |= ADVERTISE_PAUSE_ASYM;
 		mii_autoneg_adv_reg &= ~ADVERTISE_PAUSE_CAP;
+		phy->autoneg_advertised |= ADVERTISED_Asym_Pause;
+		phy->autoneg_advertised &= ~ADVERTISED_Pause;
 		break;
 	case e1000_fc_full:
 		/* Flow control (both Rx and Tx) is enabled by a software
@@ -1056,6 +1085,8 @@ static s32 e1000_phy_setup_autoneg(struct e1000_hw *hw)
 		 */
 		mii_autoneg_adv_reg |=
 		    (ADVERTISE_PAUSE_ASYM | ADVERTISE_PAUSE_CAP);
+		phy->autoneg_advertised |=
+		    (ADVERTISED_Pause | ADVERTISED_Asym_Pause);
 		break;
 	default:
 		e_dbg("Flow control param set incorrectly\n");
@@ -1744,6 +1775,7 @@ s32 e1000e_phy_has_link_generic(struct e1000_hw *hw, u32 iterations,
 	s32 ret_val = 0;
 	u16 i, phy_status;
 
+	*success = false;
 	for (i = 0; i < iterations; i++) {
 		/* Some PHYs require the MII_BMSR register to be read
 		 * twice due to the link bit being sticky.  No harm doing
@@ -1763,16 +1795,16 @@ s32 e1000e_phy_has_link_generic(struct e1000_hw *hw, u32 iterations,
 		ret_val = e1e_rphy(hw, MII_BMSR, &phy_status);
 		if (ret_val)
 			break;
-		if (phy_status & BMSR_LSTATUS)
+		if (phy_status & BMSR_LSTATUS) {
+			*success = true;
 			break;
+		}
 		if (usec_interval >= 1000)
 			msleep(usec_interval / 1000);
 		else
 			udelay(usec_interval);
 	}
 
-	*success = (i < iterations);
-
 	return ret_val;
 }
 
@@ -1801,8 +1833,7 @@ s32 e1000e_get_cable_length_m88(struct e1000_hw *hw)
 	if (ret_val)
 		return ret_val;
 
-	index = ((phy_data & M88E1000_PSSR_CABLE_LENGTH) >>
-		 M88E1000_PSSR_CABLE_LENGTH_SHIFT);
+	index = FIELD_GET(M88E1000_PSSR_CABLE_LENGTH, phy_data);
 
 	if (index >= M88E1000_CABLE_LENGTH_TABLE_SIZE - 1)
 		return -E1000_ERR_PHY;
@@ -2328,6 +2359,7 @@ s32 e1000e_determine_phy_address(struct e1000_hw *hw)
 /**
  *  e1000_get_phy_addr_for_bm_page - Retrieve PHY page address
  *  @page: page to access
+ *  @reg: register to check
  *
  *  Returns the phy address for the page requested.
  **/
@@ -2713,9 +2745,14 @@ static s32 e1000_access_phy_wakeup_reg_bm(struct e1000_hw *hw, u32 offset,
 void e1000_power_up_phy_copper(struct e1000_hw *hw)
 {
 	u16 mii_reg = 0;
+	int ret;
 
 	/* The PHY will retain its settings across a power down/up cycle */
-	e1e_rphy(hw, MII_BMCR, &mii_reg);
+	ret = e1e_rphy(hw, MII_BMCR, &mii_reg);
+	if (ret) {
+		e_dbg("Error reading PHY register\n");
+		return;
+	}
 	mii_reg &= ~BMCR_PDOWN;
 	e1e_wphy(hw, MII_BMCR, mii_reg);
 }
@@ -2731,9 +2768,14 @@ void e1000_power_up_phy_copper(struct e1000_hw *hw)
 void e1000_power_down_phy_copper(struct e1000_hw *hw)
 {
 	u16 mii_reg = 0;
+	int ret;
 
 	/* The PHY will retain its settings across a power down/up cycle */
-	e1e_rphy(hw, MII_BMCR, &mii_reg);
+	ret = e1e_rphy(hw, MII_BMCR, &mii_reg);
+	if (ret) {
+		e_dbg("Error reading PHY register\n");
+		return;
+	}
 	mii_reg |= BMCR_PDOWN;
 	e1e_wphy(hw, MII_BMCR, mii_reg);
 	usleep_range(1000, 2000);
@@ -2745,6 +2787,7 @@ void e1000_power_down_phy_copper(struct e1000_hw *hw)
  *  @offset: register offset to be read
  *  @data: pointer to the read data
  *  @locked: semaphore has already been acquired or not
+ *  @page_set: BM_WUC_PAGE already set and access enabled
  *
  *  Acquires semaphore, if necessary, then reads the PHY register at offset
  *  and stores the retrieved information in data.  Release any acquired
@@ -2853,6 +2896,7 @@ s32 e1000_read_phy_reg_page_hv(struct e1000_hw *hw, u32 offset, u16 *data)
  *  @offset: register offset to write to
  *  @data: data to write at register offset
  *  @locked: semaphore has already been acquired or not
+ *  @page_set: BM_WUC_PAGE already set and access enabled
  *
  *  Acquires semaphore, if necessary, then writes the data to PHY register
  *  at the offset.  Release any acquired semaphores before exiting.
@@ -2992,7 +3036,7 @@ static u32 e1000_get_phy_addr_for_hv_page(u32 page)
  *  @data: pointer to the data to be read or written
  *  @read: determines if operation is read or write
  *
- *  Reads the PHY register at offset and stores the retreived information
+ *  Reads the PHY register at offset and stores the retrieved information
  *  in data.  Assumes semaphore already acquired.  Note that the procedure
  *  to access these regs uses the address port and data port to read/write.
  *  These accesses done with PHY address 2 and without using pages.
@@ -3051,7 +3095,11 @@ s32 e1000_link_stall_workaround_hv(struct e1000_hw *hw)
 		return 0;
 
 	/* Do not apply workaround if in PHY loopback bit 14 set */
-	e1e_rphy(hw, MII_BMCR, &data);
+	ret_val = e1e_rphy(hw, MII_BMCR, &data);
+	if (ret_val) {
+		e_dbg("Error reading PHY register\n");
+		return ret_val;
+	}
 	if (data & BMCR_LOOPBACK)
 		return 0;
 
@@ -3225,8 +3273,7 @@ s32 e1000_get_cable_length_82577(struct e1000_hw *hw)
 	if (ret_val)
 		return ret_val;
 
-	length = ((phy_data & I82577_DSTATUS_CABLE_LENGTH) >>
-		  I82577_DSTATUS_CABLE_LENGTH_SHIFT);
+	length = FIELD_GET(I82577_DSTATUS_CABLE_LENGTH, phy_data);
 
 	if (length == E1000_CABLE_LENGTH_UNDEFINED)
 		return -E1000_ERR_PHY;
diff --git a/drivers/net/ethernet/intel/e1000e/phy.h b/drivers/net/ethernet/intel/e1000e/phy.h
index 3027f63ee793..049bb325b4b1 100644
--- a/drivers/net/ethernet/intel/e1000e/phy.h
+++ b/drivers/net/ethernet/intel/e1000e/phy.h
@@ -1,23 +1,5 @@
-/* Intel PRO/1000 Linux driver
- * Copyright(c) 1999 - 2015 Intel Corporation.
- *
- * This program is free software; you can redistribute it and/or modify it
- * under the terms and conditions of the GNU General Public License,
- * version 2, as published by the Free Software Foundation.
- *
- * This program is distributed in the hope it will be useful, but WITHOUT
- * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
- * FITNESS FOR A PARTICULAR PURPOSE.  See the GNU General Public License for
- * more details.
- *
- * The full GNU General Public License is included in this distribution in
- * the file called "COPYING".
- *
- * Contact Information:
- * Linux NICS <linux.nics@intel.com>
- * e1000-devel Mailing List <e1000-devel@lists.sourceforge.net>
- * Intel Corporation, 5200 N.E. Elam Young Parkway, Hillsboro, OR 97124-6497
- */
+/* SPDX-License-Identifier: GPL-2.0 */
+/* Copyright(c) 1999 - 2018 Intel Corporation. */
 
 #ifndef _E1000E_PHY_H_
 #define _E1000E_PHY_H_
@@ -69,6 +51,8 @@ s32 e1000e_read_phy_reg_bm2(struct e1000_hw *hw, u32 offset, u16 *data);
 s32 e1000e_write_phy_reg_bm2(struct e1000_hw *hw, u32 offset, u16 data);
 void e1000_power_up_phy_copper(struct e1000_hw *hw);
 void e1000_power_down_phy_copper(struct e1000_hw *hw);
+void e1000e_disable_phy_retry(struct e1000_hw *hw);
+void e1000e_enable_phy_retry(struct e1000_hw *hw);
 s32 e1000e_read_phy_reg_mdic(struct e1000_hw *hw, u32 offset, u16 *data);
 s32 e1000e_write_phy_reg_mdic(struct e1000_hw *hw, u32 offset, u16 data);
 s32 e1000_read_phy_reg_hv(struct e1000_hw *hw, u32 offset, u16 *data);
diff --git a/drivers/net/ethernet/intel/e1000e/ptp.c b/drivers/net/ethernet/intel/e1000e/ptp.c
index b366885487a8..ea3c3eb2ef20 100644
--- a/drivers/net/ethernet/intel/e1000e/ptp.c
+++ b/drivers/net/ethernet/intel/e1000e/ptp.c
@@ -1,23 +1,5 @@
-/* Intel PRO/1000 Linux driver
- * Copyright(c) 1999 - 2015 Intel Corporation.
- *
- * This program is free software; you can redistribute it and/or modify it
- * under the terms and conditions of the GNU General Public License,
- * version 2, as published by the Free Software Foundation.
- *
- * This program is distributed in the hope it will be useful, but WITHOUT
- * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
- * FITNESS FOR A PARTICULAR PURPOSE.  See the GNU General Public License for
- * more details.
- *
- * The full GNU General Public License is included in this distribution in
- * the file called "COPYING".
- *
- * Contact Information:
- * Linux NICS <linux.nics@intel.com>
- * e1000-devel Mailing List <e1000-devel@lists.sourceforge.net>
- * Intel Corporation, 5200 N.E. Elam Young Parkway, Hillsboro, OR 97124-6497
- */
+// SPDX-License-Identifier: GPL-2.0
+/* Copyright(c) 1999 - 2018 Intel Corporation. */
 
 /* PTP 1588 Hardware Clock (PHC)
  * Derived from PTP Hardware Clock driver for Intel 82576 and 82580 (igb)
@@ -33,32 +15,25 @@
 #endif
 
 /**
- * e1000e_phc_adjfreq - adjust the frequency of the hardware clock
+ * e1000e_phc_adjfine - adjust the frequency of the hardware clock
  * @ptp: ptp clock structure
- * @delta: Desired frequency change in parts per billion
+ * @delta: Desired frequency chance in scaled parts per million
  *
  * Adjust the frequency of the PHC cycle counter by the indicated delta from
  * the base frequency.
+ *
+ * Scaled parts per million is ppm but with a 16 bit binary fractional field.
  **/
-static int e1000e_phc_adjfreq(struct ptp_clock_info *ptp, s32 delta)
+static int e1000e_phc_adjfine(struct ptp_clock_info *ptp, long delta)
 {
 	struct e1000_adapter *adapter = container_of(ptp, struct e1000_adapter,
 						     ptp_clock_info);
 	struct e1000_hw *hw = &adapter->hw;
-	bool neg_adj = false;
 	unsigned long flags;
-	u64 adjustment;
-	u32 timinca, incvalue;
+	u64 incvalue;
+	u32 timinca;
 	s32 ret_val;
 
-	if ((delta > ptp->max_adj) || (delta <= -1000000000))
-		return -EINVAL;
-
-	if (delta < 0) {
-		neg_adj = true;
-		delta = -delta;
-	}
-
 	/* Get the System Time Register SYSTIM base frequency */
 	ret_val = e1000e_get_base_timinca(adapter, &timinca);
 	if (ret_val)
@@ -67,12 +42,7 @@ static int e1000e_phc_adjfreq(struct ptp_clock_info *ptp, s32 delta)
 	spin_lock_irqsave(&adapter->systim_lock, flags);
 
 	incvalue = timinca & E1000_TIMINCA_INCVALUE_MASK;
-
-	adjustment = incvalue;
-	adjustment *= delta;
-	adjustment = div_u64(adjustment, 1000000000);
-
-	incvalue = neg_adj ? (incvalue - adjustment) : (incvalue + adjustment);
+	incvalue = adjust_by_scaled_ppm(incvalue, delta);
 
 	timinca &= ~E1000_TIMINCA_INCVALUE_MASK;
 	timinca |= incvalue;
@@ -154,15 +124,16 @@ static int e1000e_phc_get_syncdevicetime(ktime_t *device,
 	sys_cycles = er32(PLTSTMPH);
 	sys_cycles <<= 32;
 	sys_cycles |= er32(PLTSTMPL);
-	*system = convert_art_to_tsc(sys_cycles);
+	system->cycles = sys_cycles;
+	system->cs_id = CSID_X86_ART;
 
 	return 0;
 }
 
 /**
- * e1000e_phc_getsynctime - Reads the current system/device cross timestamp
+ * e1000e_phc_getcrosststamp - Reads the current system/device cross timestamp
  * @ptp: ptp clock structure
- * @cts: structure containing timestamp
+ * @xtstamp: structure containing timestamp
  *
  * Read device and system (ART) clock simultaneously and return the scaled
  * clock values in ns.
@@ -179,22 +150,30 @@ static int e1000e_phc_getcrosststamp(struct ptp_clock_info *ptp,
 #endif/*CONFIG_E1000E_HWTS*/
 
 /**
- * e1000e_phc_gettime - Reads the current time from the hardware clock
+ * e1000e_phc_gettimex - Reads the current time from the hardware clock and
+ *                       system clock
  * @ptp: ptp clock structure
- * @ts: timespec structure to hold the current time value
+ * @ts: timespec structure to hold the current PHC time
+ * @sts: structure to hold the current system time
  *
  * Read the timecounter and return the correct value in ns after converting
  * it into a struct timespec.
  **/
-static int e1000e_phc_gettime(struct ptp_clock_info *ptp, struct timespec64 *ts)
+static int e1000e_phc_gettimex(struct ptp_clock_info *ptp,
+			       struct timespec64 *ts,
+			       struct ptp_system_timestamp *sts)
 {
 	struct e1000_adapter *adapter = container_of(ptp, struct e1000_adapter,
 						     ptp_clock_info);
 	unsigned long flags;
-	u64 ns;
+	u64 cycles, ns;
 
 	spin_lock_irqsave(&adapter->systim_lock, flags);
-	ns = timecounter_read(&adapter->tc);
+
+	/* NOTE: Non-monotonic SYSTIM readings may be returned */
+	cycles = e1000e_read_systim(adapter, sts);
+	ns = timecounter_cyc2time(&adapter->tc, cycles);
+
 	spin_unlock_irqrestore(&adapter->systim_lock, flags);
 
 	*ts = ns_to_timespec64(ns);
@@ -250,9 +229,12 @@ static void e1000e_systim_overflow_work(struct work_struct *work)
 						     systim_overflow_work.work);
 	struct e1000_hw *hw = &adapter->hw;
 	struct timespec64 ts;
+	u64 ns;
 
-	adapter->ptp_clock_info.gettime64(&adapter->ptp_clock_info, &ts);
+	/* Update the timecounter */
+	ns = timecounter_read(&adapter->tc);
 
+	ts = ns_to_timespec64(ns);
 	e_dbg("SYSTIM overflow check at %lld.%09lu\n",
 	      (long long) ts.tv_sec, ts.tv_nsec);
 
@@ -267,9 +249,9 @@ static const struct ptp_clock_info e1000e_ptp_clock_info = {
 	.n_per_out	= 0,
 	.n_pins		= 0,
 	.pps		= 0,
-	.adjfreq	= e1000e_phc_adjfreq,
+	.adjfine	= e1000e_phc_adjfine,
 	.adjtime	= e1000e_phc_adjtime,
-	.gettime64	= e1000e_phc_gettime,
+	.gettimex64	= e1000e_phc_gettimex,
 	.settime64	= e1000e_phc_settime,
 	.enable		= e1000e_phc_enable,
 };
@@ -299,18 +281,34 @@ void e1000e_ptp_init(struct e1000_adapter *adapter)
 
 	switch (hw->mac.type) {
 	case e1000_pch2lan:
+		adapter->ptp_clock_info.max_adj = MAX_PPB_96MHZ;
+		break;
 	case e1000_pch_lpt:
+		if (er32(TSYNCRXCTL) & E1000_TSYNCRXCTL_SYSCFI)
+			adapter->ptp_clock_info.max_adj = MAX_PPB_96MHZ;
+		else
+			adapter->ptp_clock_info.max_adj = MAX_PPB_25MHZ;
+		break;
 	case e1000_pch_spt:
+		adapter->ptp_clock_info.max_adj = MAX_PPB_24MHZ;
+		break;
 	case e1000_pch_cnp:
-		if ((hw->mac.type < e1000_pch_lpt) ||
-		    (er32(TSYNCRXCTL) & E1000_TSYNCRXCTL_SYSCFI)) {
-			adapter->ptp_clock_info.max_adj = 24000000 - 1;
-			break;
-		}
-		/* fall-through */
+	case e1000_pch_tgp:
+	case e1000_pch_adp:
+	case e1000_pch_nvp:
+		if (er32(TSYNCRXCTL) & E1000_TSYNCRXCTL_SYSCFI)
+			adapter->ptp_clock_info.max_adj = MAX_PPB_24MHZ;
+		else
+			adapter->ptp_clock_info.max_adj = MAX_PPB_38400KHZ;
+		break;
+	case e1000_pch_mtp:
+	case e1000_pch_lnp:
+	case e1000_pch_ptp:
+		adapter->ptp_clock_info.max_adj = MAX_PPB_38400KHZ;
+		break;
 	case e1000_82574:
 	case e1000_82583:
-		adapter->ptp_clock_info.max_adj = 600000000 - 1;
+		adapter->ptp_clock_info.max_adj = MAX_PPB_25MHZ;
 		break;
 	default:
 		break;
diff --git a/drivers/net/ethernet/intel/e1000e/regs.h b/drivers/net/ethernet/intel/e1000e/regs.h
index 0cb4d365e5ad..6c0cd8cab3ef 100644
--- a/drivers/net/ethernet/intel/e1000e/regs.h
+++ b/drivers/net/ethernet/intel/e1000e/regs.h
@@ -1,23 +1,5 @@
-/* Intel PRO/1000 Linux driver
- * Copyright(c) 1999 - 2015 Intel Corporation.
- *
- * This program is free software; you can redistribute it and/or modify it
- * under the terms and conditions of the GNU General Public License,
- * version 2, as published by the Free Software Foundation.
- *
- * This program is distributed in the hope it will be useful, but WITHOUT
- * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
- * FITNESS FOR A PARTICULAR PURPOSE.  See the GNU General Public License for
- * more details.
- *
- * The full GNU General Public License is included in this distribution in
- * the file called "COPYING".
- *
- * Contact Information:
- * Linux NICS <linux.nics@intel.com>
- * e1000-devel Mailing List <e1000-devel@lists.sourceforge.net>
- * Intel Corporation, 5200 N.E. Elam Young Parkway, Hillsboro, OR 97124-6497
- */
+/* SPDX-License-Identifier: GPL-2.0 */
+/* Copyright(c) 1999 - 2018 Intel Corporation. */
 
 #ifndef _E1000E_REGS_H_
 #define _E1000E_REGS_H_
@@ -36,11 +18,15 @@
 #define E1000_FEXTNVM	0x00028	/* Future Extended NVM - RW */
 #define E1000_FEXTNVM3	0x0003C	/* Future Extended NVM 3 - RW */
 #define E1000_FEXTNVM4	0x00024	/* Future Extended NVM 4 - RW */
+#define E1000_FEXTNVM5	0x00014	/* Future Extended NVM 5 - RW */
 #define E1000_FEXTNVM6	0x00010	/* Future Extended NVM 6 - RW */
 #define E1000_FEXTNVM7	0x000E4	/* Future Extended NVM 7 - RW */
+#define E1000_FEXTNVM8	0x5BB0	/* Future Extended NVM 8 - RW */
 #define E1000_FEXTNVM9	0x5BB4	/* Future Extended NVM 9 - RW */
 #define E1000_FEXTNVM11	0x5BBC	/* Future Extended NVM 11 - RW */
+#define E1000_FEXTNVM12	0x5BC0	/* Future Extended NVM 12 - RW */
 #define E1000_PCIEANACFG	0x00F18	/* PCIE Analog Config */
+#define E1000_DPGFR	0x00FAC	/* Dynamic Power Gate Force Control Register */
 #define E1000_FCT	0x00030	/* Flow Control Type - RW */
 #define E1000_VET	0x00038	/* VLAN Ether Type - RW */
 #define E1000_ICR	0x000C0	/* Interrupt Cause Read - R/clr */
@@ -227,6 +213,7 @@
 #define E1000_FACTPS	0x05B30	/* Function Active and Power State to MNG */
 #define E1000_SWSM	0x05B50	/* SW Semaphore */
 #define E1000_FWSM	0x05B54	/* FW Semaphore */
+#define E1000_EXFWSM	0x05B58	/* Extended FW Semaphore */
 /* Driver-only SW semaphore (not used by BOOT agents) */
 #define E1000_SWSM2	0x05B58
 #define E1000_FFLT_DBG	0x05F04	/* Debug Register */
@@ -252,4 +239,7 @@
 #define E1000_RXMTRL	0x0B634	/* Time sync Rx EtherType and Msg Type - RW */
 #define E1000_RXUDP	0x0B638	/* Time Sync Rx UDP Port - RW */
 
+/* PHY registers */
+#define I82579_DFT_CTRL	PHY_REG(769, 20)
+
 #endif
diff --git a/drivers/net/ethernet/intel/fm10k/Makefile b/drivers/net/ethernet/intel/fm10k/Makefile
index cac645329cea..26a9746ccb14 100644
--- a/drivers/net/ethernet/intel/fm10k/Makefile
+++ b/drivers/net/ethernet/intel/fm10k/Makefile
@@ -1,25 +1,5 @@
-################################################################################
-#
-# Intel(R) Ethernet Switch Host Interface Driver
-# Copyright(c) 2013 - 2016 Intel Corporation.
-#
-# This program is free software; you can redistribute it and/or modify it
-# under the terms and conditions of the GNU General Public License,
-# version 2, as published by the Free Software Foundation.
-#
-# This program is distributed in the hope it will be useful, but WITHOUT
-# ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
-# FITNESS FOR A PARTICULAR PURPOSE.  See the GNU General Public License for
-# more details.
-#
-# The full GNU General Public License is included in this distribution in
-# the file called "COPYING".
-#
-# Contact Information:
-# e1000-devel Mailing List <e1000-devel@lists.sourceforge.net>
-# Intel Corporation, 5200 N.E. Elam Young Parkway, Hillsboro, OR 97124-6497
-#
-################################################################################
+# SPDX-License-Identifier: GPL-2.0
+# Copyright(c) 2013 - 2018 Intel Corporation.
 
 #
 # Makefile for the Intel(R) Ethernet Switch Host Interface Driver
diff --git a/drivers/net/ethernet/intel/fm10k/fm10k.h b/drivers/net/ethernet/intel/fm10k/fm10k.h
index 689c413b7782..65a2816142d9 100644
--- a/drivers/net/ethernet/intel/fm10k/fm10k.h
+++ b/drivers/net/ethernet/intel/fm10k/fm10k.h
@@ -1,22 +1,5 @@
-/* Intel(R) Ethernet Switch Host Interface Driver
- * Copyright(c) 2013 - 2017 Intel Corporation.
- *
- * This program is free software; you can redistribute it and/or modify it
- * under the terms and conditions of the GNU General Public License,
- * version 2, as published by the Free Software Foundation.
- *
- * This program is distributed in the hope it will be useful, but WITHOUT
- * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
- * FITNESS FOR A PARTICULAR PURPOSE.  See the GNU General Public License for
- * more details.
- *
- * The full GNU General Public License is included in this distribution in
- * the file called "COPYING".
- *
- * Contact Information:
- * e1000-devel Mailing List <e1000-devel@lists.sourceforge.net>
- * Intel Corporation, 5200 N.E. Elam Young Parkway, Hillsboro, OR 97124-6497
- */
+/* SPDX-License-Identifier: GPL-2.0 */
+/* Copyright(c) 2013 - 2019 Intel Corporation. */
 
 #ifndef _FM10K_H_
 #define _FM10K_H_
@@ -58,7 +41,7 @@ struct fm10k_l2_accel {
 	u16 count;
 	u16 dglort;
 	struct rcu_head rcu;
-	struct net_device *macvlan[0];
+	struct net_device *macvlan[];
 };
 
 enum fm10k_ring_state_t {
@@ -194,14 +177,10 @@ static inline struct netdev_queue *txring_txq(const struct fm10k_ring *ring)
 #define MIN_Q_VECTORS	1
 enum fm10k_non_q_vectors {
 	FM10K_MBX_VECTOR,
-#define NON_Q_VECTORS_VF NON_Q_VECTORS_PF
-	NON_Q_VECTORS_PF
+	NON_Q_VECTORS
 };
 
-#define NON_Q_VECTORS(hw)	(((hw)->mac.type == fm10k_mac_pf) ? \
-						NON_Q_VECTORS_PF : \
-						NON_Q_VECTORS_VF)
-#define MIN_MSIX_COUNT(hw)	(MIN_Q_VECTORS + NON_Q_VECTORS(hw))
+#define MIN_MSIX_COUNT(hw)	(MIN_Q_VECTORS + NON_Q_VECTORS)
 
 struct fm10k_q_vector {
 	struct fm10k_intfc *interface;
@@ -210,16 +189,17 @@ struct fm10k_q_vector {
 	struct fm10k_ring_container rx, tx;
 
 	struct napi_struct napi;
+	struct rcu_head rcu;	/* to avoid race with update stats on free */
+
 	cpumask_t affinity_mask;
 	char name[IFNAMSIZ + 9];
 
 #ifdef CONFIG_DEBUG_FS
 	struct dentry *dbg_q_vector;
 #endif /* CONFIG_DEBUG_FS */
-	struct rcu_head rcu;	/* to avoid race with update stats on free */
 
 	/* for dynamic allocation of rings associated with this q_vector */
-	struct fm10k_ring ring[0] ____cacheline_internodealigned_in_smp;
+	struct fm10k_ring ring[] ____cacheline_internodealigned_in_smp;
 };
 
 enum fm10k_ring_f_enum {
@@ -239,13 +219,30 @@ struct fm10k_iov_data {
 	unsigned int		num_vfs;
 	unsigned int		next_vf_mbx;
 	struct rcu_head		rcu;
-	struct fm10k_vf_info	vf_info[0];
+	struct fm10k_vf_info	vf_info[];
 };
 
-struct fm10k_udp_port {
-	struct list_head	list;
-	sa_family_t		sa_family;
-	__be16			port;
+enum fm10k_macvlan_request_type {
+	FM10K_UC_MAC_REQUEST,
+	FM10K_MC_MAC_REQUEST,
+	FM10K_VLAN_REQUEST
+};
+
+struct fm10k_macvlan_request {
+	enum fm10k_macvlan_request_type type;
+	struct list_head list;
+	union {
+		struct fm10k_mac_request {
+			u8 addr[ETH_ALEN];
+			u16 glort;
+			u16 vid;
+		} mac;
+		struct fm10k_vlan_request {
+			u32 vid;
+			u8 vsi;
+		} vlan;
+	};
+	bool set;
 };
 
 /* one work queue for entire driver */
@@ -270,11 +267,15 @@ enum fm10k_flags_t {
 
 enum fm10k_state_t {
 	__FM10K_RESETTING,
+	__FM10K_RESET_DETACHED,
+	__FM10K_RESET_SUSPENDED,
 	__FM10K_DOWN,
 	__FM10K_SERVICE_SCHED,
 	__FM10K_SERVICE_REQUEST,
 	__FM10K_SERVICE_DISABLE,
-	__FM10K_MBX_LOCK,
+	__FM10K_MACVLAN_SCHED,
+	__FM10K_MACVLAN_REQUEST,
+	__FM10K_MACVLAN_DISABLE,
 	__FM10K_LINK_DOWN,
 	__FM10K_UPDATING_STATS,
 	/* This value must be last and determines the BITMAP size */
@@ -344,6 +345,8 @@ struct fm10k_intfc {
 
 	struct fm10k_hw_stats stats;
 	struct fm10k_hw hw;
+	/* Mailbox lock */
+	spinlock_t mbx_lock;
 	u32 __iomem *uc_addr;
 	u32 __iomem *sw_addr;
 	u16 msg_enable;
@@ -362,8 +365,14 @@ struct fm10k_intfc {
 	u32 rssrk[FM10K_RSSRK_SIZE];
 
 	/* UDP encapsulation port tracking information */
-	struct list_head vxlan_port;
-	struct list_head geneve_port;
+	__be16 vxlan_port;
+	__be16 geneve_port;
+
+	/* MAC/VLAN update queue */
+	struct list_head macvlan_requests;
+	struct delayed_work macvlan_task;
+	/* MAC/VLAN update queue lock */
+	spinlock_t macvlan_lock;
 
 #ifdef CONFIG_DEBUG_FS
 	struct dentry *dbg_intfc;
@@ -384,23 +393,17 @@ struct fm10k_intfc {
 
 static inline void fm10k_mbx_lock(struct fm10k_intfc *interface)
 {
-	/* busy loop if we cannot obtain the lock as some calls
-	 * such as ndo_set_rx_mode may be made in atomic context
-	 */
-	while (test_and_set_bit(__FM10K_MBX_LOCK, interface->state))
-		udelay(20);
+	spin_lock(&interface->mbx_lock);
 }
 
 static inline void fm10k_mbx_unlock(struct fm10k_intfc *interface)
 {
-	/* flush memory to make sure state is correct */
-	smp_mb__before_atomic();
-	clear_bit(__FM10K_MBX_LOCK, interface->state);
+	spin_unlock(&interface->mbx_lock);
 }
 
 static inline int fm10k_mbx_trylock(struct fm10k_intfc *interface)
 {
-	return !test_and_set_bit(__FM10K_MBX_LOCK, interface->state);
+	return spin_trylock(&interface->mbx_lock);
 }
 
 /* fm10k_test_staterr - test bits in Rx descriptor status and error fields */
@@ -468,7 +471,6 @@ struct fm10k_cb {
 
 /* main */
 extern char fm10k_driver_name[];
-extern const char fm10k_driver_version[];
 int fm10k_init_queueing_scheme(struct fm10k_intfc *interface);
 void fm10k_clear_queueing_scheme(struct fm10k_intfc *interface);
 __be16 fm10k_tx_encap_offload(struct sk_buff *skb);
@@ -490,10 +492,8 @@ void fm10k_up(struct fm10k_intfc *interface);
 void fm10k_down(struct fm10k_intfc *interface);
 void fm10k_update_stats(struct fm10k_intfc *interface);
 void fm10k_service_event_schedule(struct fm10k_intfc *interface);
+void fm10k_macvlan_schedule(struct fm10k_intfc *interface);
 void fm10k_update_rx_drop_en(struct fm10k_intfc *interface);
-#ifdef CONFIG_NET_POLL_CONTROLLER
-void fm10k_netpoll(struct net_device *netdev);
-#endif
 
 /* Netdev */
 struct net_device *fm10k_alloc_netdev(const struct fm10k_info *info);
@@ -510,6 +510,12 @@ void fm10k_reset_rx_state(struct fm10k_intfc *);
 int fm10k_setup_tc(struct net_device *dev, u8 tc);
 int fm10k_open(struct net_device *netdev);
 int fm10k_close(struct net_device *netdev);
+int fm10k_queue_vlan_request(struct fm10k_intfc *interface, u32 vid,
+			     u8 vsi, bool set);
+int fm10k_queue_mac_request(struct fm10k_intfc *interface, u16 glort,
+			    const unsigned char *addr, u16 vid, bool set);
+void fm10k_clear_macvlan_queue(struct fm10k_intfc *interface,
+			       u16 glort, bool vlans);
 
 /* Ethtool */
 void fm10k_set_ethtool_ops(struct net_device *dev);
@@ -522,14 +528,17 @@ void fm10k_iov_suspend(struct pci_dev *pdev);
 int fm10k_iov_resume(struct pci_dev *pdev);
 void fm10k_iov_disable(struct pci_dev *pdev);
 int fm10k_iov_configure(struct pci_dev *pdev, int num_vfs);
+void fm10k_iov_update_stats(struct fm10k_intfc *interface);
 s32 fm10k_iov_update_pvid(struct fm10k_intfc *interface, u16 glort, u16 pvid);
 int fm10k_ndo_set_vf_mac(struct net_device *netdev, int vf_idx, u8 *mac);
 int fm10k_ndo_set_vf_vlan(struct net_device *netdev,
 			  int vf_idx, u16 vid, u8 qos, __be16 vlan_proto);
-int fm10k_ndo_set_vf_bw(struct net_device *netdev, int vf_idx, int rate,
-			int unused);
+int fm10k_ndo_set_vf_bw(struct net_device *netdev, int vf_idx,
+			int __always_unused min_rate, int max_rate);
 int fm10k_ndo_get_vf_config(struct net_device *netdev,
 			    int vf_idx, struct ifla_vf_info *ivi);
+int fm10k_ndo_get_vf_stats(struct net_device *netdev,
+			   int vf_idx, struct ifla_vf_stats *stats);
 
 /* DebugFS */
 #ifdef CONFIG_DEBUG_FS
diff --git a/drivers/net/ethernet/intel/fm10k/fm10k_common.c b/drivers/net/ethernet/intel/fm10k/fm10k_common.c
index 62a6ad9b3eed..1f919a50c765 100644
--- a/drivers/net/ethernet/intel/fm10k/fm10k_common.c
+++ b/drivers/net/ethernet/intel/fm10k/fm10k_common.c
@@ -1,22 +1,5 @@
-/* Intel(R) Ethernet Switch Host Interface Driver
- * Copyright(c) 2013 - 2016 Intel Corporation.
- *
- * This program is free software; you can redistribute it and/or modify it
- * under the terms and conditions of the GNU General Public License,
- * version 2, as published by the Free Software Foundation.
- *
- * This program is distributed in the hope it will be useful, but WITHOUT
- * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
- * FITNESS FOR A PARTICULAR PURPOSE.  See the GNU General Public License for
- * more details.
- *
- * The full GNU General Public License is included in this distribution in
- * the file called "COPYING".
- *
- * Contact Information:
- * e1000-devel Mailing List <e1000-devel@lists.sourceforge.net>
- * Intel Corporation, 5200 N.E. Elam Young Parkway, Hillsboro, OR 97124-6497
- */
+// SPDX-License-Identifier: GPL-2.0
+/* Copyright(c) 2013 - 2018 Intel Corporation. */
 
 #include "fm10k_common.h"
 
@@ -262,6 +245,7 @@ s32 fm10k_stop_hw_generic(struct fm10k_hw *hw)
  *  fm10k_read_hw_stats_32b - Reads value of 32-bit registers
  *  @hw: pointer to the hardware structure
  *  @addr: address of register containing a 32-bit value
+ *  @stat: pointer to structure holding hw stat information
  *
  *  Function reads the content of the register and returns the delta
  *  between the base and the current value.
@@ -281,6 +265,7 @@ u32 fm10k_read_hw_stats_32b(struct fm10k_hw *hw, u32 addr,
  *  fm10k_read_hw_stats_48b - Reads value of 48-bit registers
  *  @hw: pointer to the hardware structure
  *  @addr: address of register containing the lower 32-bit value
+ *  @stat: pointer to structure holding hw stat information
  *
  *  Function reads the content of 2 registers, combined to represent a 48-bit
  *  statistical value. Extra processing is required to handle overflowing.
@@ -461,19 +446,17 @@ void fm10k_update_hw_stats_q(struct fm10k_hw *hw, struct fm10k_hw_stats_q *q,
 
 /**
  *  fm10k_unbind_hw_stats_q - Unbind the queue counters from their queues
- *  @hw: pointer to the hardware structure
  *  @q: pointer to the ring of hardware statistics queue
- *  @idx: index pointing to the start of the ring iteration
  *  @count: number of queues to iterate over
  *
  *  Function invalidates the index values for the queues so any updates that
  *  may have happened are ignored and the base for the queue stats is reset.
  **/
-void fm10k_unbind_hw_stats_q(struct fm10k_hw_stats_q *q, u32 idx, u32 count)
+void fm10k_unbind_hw_stats_q(struct fm10k_hw_stats_q *q, u32 count)
 {
 	u32 i;
 
-	for (i = 0; i < count; i++, idx++, q++) {
+	for (i = 0; i < count; i++, q++) {
 		q->rx_stats_idx = 0;
 		q->tx_stats_idx = 0;
 	}
@@ -517,8 +500,8 @@ s32 fm10k_get_host_state_generic(struct fm10k_hw *hw, bool *host_ready)
 		goto out;
 	}
 
-	/* verify Mailbox is still valid */
-	if (!mbx->ops.tx_ready(mbx, FM10K_VFMBX_MSG_MTU))
+	/* verify Mailbox is still open */
+	if (mbx->state != FM10K_STATE_OPEN)
 		goto out;
 
 	/* interface cannot receive traffic without logical ports */
diff --git a/drivers/net/ethernet/intel/fm10k/fm10k_common.h b/drivers/net/ethernet/intel/fm10k/fm10k_common.h
index d51f9c7a47ff..13fca6a91a01 100644
--- a/drivers/net/ethernet/intel/fm10k/fm10k_common.h
+++ b/drivers/net/ethernet/intel/fm10k/fm10k_common.h
@@ -1,22 +1,5 @@
-/* Intel(R) Ethernet Switch Host Interface Driver
- * Copyright(c) 2013 - 2016 Intel Corporation.
- *
- * This program is free software; you can redistribute it and/or modify it
- * under the terms and conditions of the GNU General Public License,
- * version 2, as published by the Free Software Foundation.
- *
- * This program is distributed in the hope it will be useful, but WITHOUT
- * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
- * FITNESS FOR A PARTICULAR PURPOSE.  See the GNU General Public License for
- * more details.
- *
- * The full GNU General Public License is included in this distribution in
- * the file called "COPYING".
- *
- * Contact Information:
- * e1000-devel Mailing List <e1000-devel@lists.sourceforge.net>
- * Intel Corporation, 5200 N.E. Elam Young Parkway, Hillsboro, OR 97124-6497
- */
+/* SPDX-License-Identifier: GPL-2.0 */
+/* Copyright(c) 2013 - 2018 Intel Corporation. */
 
 #ifndef _FM10K_COMMON_H_
 #define _FM10K_COMMON_H_
@@ -60,6 +43,6 @@ u32 fm10k_read_hw_stats_32b(struct fm10k_hw *hw, u32 addr,
 void fm10k_update_hw_stats_q(struct fm10k_hw *hw, struct fm10k_hw_stats_q *q,
 			     u32 idx, u32 count);
 #define fm10k_unbind_hw_stats_32b(s) ((s)->base_h = 0)
-void fm10k_unbind_hw_stats_q(struct fm10k_hw_stats_q *q, u32 idx, u32 count);
+void fm10k_unbind_hw_stats_q(struct fm10k_hw_stats_q *q, u32 count);
 s32 fm10k_get_host_state_generic(struct fm10k_hw *hw, bool *host_ready);
 #endif /* _FM10K_COMMON_H_ */
diff --git a/drivers/net/ethernet/intel/fm10k/fm10k_dcbnl.c b/drivers/net/ethernet/intel/fm10k/fm10k_dcbnl.c
index db4bd8bf9722..86397c564dfc 100644
--- a/drivers/net/ethernet/intel/fm10k/fm10k_dcbnl.c
+++ b/drivers/net/ethernet/intel/fm10k/fm10k_dcbnl.c
@@ -1,22 +1,5 @@
-/* Intel(R) Ethernet Switch Host Interface Driver
- * Copyright(c) 2013 - 2016 Intel Corporation.
- *
- * This program is free software; you can redistribute it and/or modify it
- * under the terms and conditions of the GNU General Public License,
- * version 2, as published by the Free Software Foundation.
- *
- * This program is distributed in the hope it will be useful, but WITHOUT
- * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
- * FITNESS FOR A PARTICULAR PURPOSE.  See the GNU General Public License for
- * more details.
- *
- * The full GNU General Public License is included in this distribution in
- * the file called "COPYING".
- *
- * Contact Information:
- * e1000-devel Mailing List <e1000-devel@lists.sourceforge.net>
- * Intel Corporation, 5200 N.E. Elam Young Parkway, Hillsboro, OR 97124-6497
- */
+// SPDX-License-Identifier: GPL-2.0
+/* Copyright(c) 2013 - 2019 Intel Corporation. */
 
 #include "fm10k.h"
 
@@ -53,7 +36,7 @@ static int fm10k_dcbnl_ieee_getets(struct net_device *dev, struct ieee_ets *ets)
 static int fm10k_dcbnl_ieee_setets(struct net_device *dev, struct ieee_ets *ets)
 {
 	u8 num_tc = 0;
-	int i, err;
+	int i;
 
 	/* verify type and determine num_tcs needed */
 	for (i = 0; i < IEEE_8021QAZ_MAX_TCS; i++) {
@@ -74,7 +57,7 @@ static int fm10k_dcbnl_ieee_setets(struct net_device *dev, struct ieee_ets *ets)
 
 	/* update TC hardware mapping if necessary */
 	if (num_tc != netdev_get_num_tc(dev)) {
-		err = fm10k_setup_tc(dev, num_tc);
+		int err = fm10k_setup_tc(dev, num_tc);
 		if (err)
 			return err;
 	}
@@ -122,7 +105,7 @@ static int fm10k_dcbnl_ieee_setpfc(struct net_device *dev, struct ieee_pfc *pfc)
 }
 
 /**
- * fm10k_dcbnl_ieee_getdcbx - get the DCBX configuration for the device
+ * fm10k_dcbnl_getdcbx - get the DCBX configuration for the device
  * @dev: netdev interface for the device
  *
  * Returns that we support only IEEE DCB for this interface
@@ -133,7 +116,7 @@ static u8 fm10k_dcbnl_getdcbx(struct net_device __always_unused *dev)
 }
 
 /**
- * fm10k_dcbnl_ieee_setdcbx - get the DCBX configuration for the device
+ * fm10k_dcbnl_setdcbx - get the DCBX configuration for the device
  * @dev: netdev interface for the device
  * @mode: new mode for this device
  *
diff --git a/drivers/net/ethernet/intel/fm10k/fm10k_debugfs.c b/drivers/net/ethernet/intel/fm10k/fm10k_debugfs.c
index 5116fd043630..5c77054d67c6 100644
--- a/drivers/net/ethernet/intel/fm10k/fm10k_debugfs.c
+++ b/drivers/net/ethernet/intel/fm10k/fm10k_debugfs.c
@@ -1,22 +1,5 @@
-/* Intel(R) Ethernet Switch Host Interface Driver
- * Copyright(c) 2013 - 2016 Intel Corporation.
- *
- * This program is free software; you can redistribute it and/or modify it
- * under the terms and conditions of the GNU General Public License,
- * version 2, as published by the Free Software Foundation.
- *
- * This program is distributed in the hope it will be useful, but WITHOUT
- * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
- * FITNESS FOR A PARTICULAR PURPOSE.  See the GNU General Public License for
- * more details.
- *
- * The full GNU General Public License is included in this distribution in
- * the file called "COPYING".
- *
- * Contact Information:
- * e1000-devel Mailing List <e1000-devel@lists.sourceforge.net>
- * Intel Corporation, 5200 N.E. Elam Young Parkway, Hillsboro, OR 97124-6497
- */
+// SPDX-License-Identifier: GPL-2.0
+/* Copyright(c) 2013 - 2018 Intel Corporation. */
 
 #include "fm10k.h"
 
@@ -52,9 +35,9 @@ static void fm10k_dbg_desc_seq_stop(struct seq_file __always_unused *s,
 static void fm10k_dbg_desc_break(struct seq_file *s, int i)
 {
 	while (i--)
-		seq_puts(s, "-");
+		seq_putc(s, '-');
 
-	seq_puts(s, "\n");
+	seq_putc(s, '\n');
 }
 
 static int fm10k_dbg_tx_desc_seq_show(struct seq_file *s, void *v)
@@ -177,8 +160,6 @@ void fm10k_dbg_q_vector_init(struct fm10k_q_vector *q_vector)
 	snprintf(name, sizeof(name), "q_vector.%03d", q_vector->v_idx);
 
 	q_vector->dbg_q_vector = debugfs_create_dir(name, interface->dbg_intfc);
-	if (!q_vector->dbg_q_vector)
-		return;
 
 	/* Generate a file for each rx ring in the q_vector */
 	for (i = 0; i < q_vector->tx.count; i++) {
@@ -204,7 +185,7 @@ void fm10k_dbg_q_vector_init(struct fm10k_q_vector *q_vector)
 }
 
 /**
- * fm10k_dbg_free_q_vector_dir - setup debugfs for the q_vectors
+ * fm10k_dbg_q_vector_exit - setup debugfs for the q_vectors
  * @q_vector: q_vector to allocate directories for
  **/
 void fm10k_dbg_q_vector_exit(struct fm10k_q_vector *q_vector)
diff --git a/drivers/net/ethernet/intel/fm10k/fm10k_ethtool.c b/drivers/net/ethernet/intel/fm10k/fm10k_ethtool.c
index c7234f35f8ff..bf2029144c1d 100644
--- a/drivers/net/ethernet/intel/fm10k/fm10k_ethtool.c
+++ b/drivers/net/ethernet/intel/fm10k/fm10k_ethtool.c
@@ -1,39 +1,33 @@
-/* Intel(R) Ethernet Switch Host Interface Driver
- * Copyright(c) 2013 - 2017 Intel Corporation.
- *
- * This program is free software; you can redistribute it and/or modify it
- * under the terms and conditions of the GNU General Public License,
- * version 2, as published by the Free Software Foundation.
- *
- * This program is distributed in the hope it will be useful, but WITHOUT
- * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
- * FITNESS FOR A PARTICULAR PURPOSE.  See the GNU General Public License for
- * more details.
- *
- * The full GNU General Public License is included in this distribution in
- * the file called "COPYING".
- *
- * Contact Information:
- * e1000-devel Mailing List <e1000-devel@lists.sourceforge.net>
- * Intel Corporation, 5200 N.E. Elam Young Parkway, Hillsboro, OR 97124-6497
- */
+// SPDX-License-Identifier: GPL-2.0
+/* Copyright(c) 2013 - 2019 Intel Corporation. */
 
+#include <linux/ethtool.h>
 #include <linux/vmalloc.h>
 
 #include "fm10k.h"
 
 struct fm10k_stats {
+	/* The stat_string is expected to be a format string formatted using
+	 * vsnprintf by fm10k_add_stat_strings. Every member of a stats array
+	 * should use the same format specifiers as they will be formatted
+	 * using the same variadic arguments.
+	 */
 	char stat_string[ETH_GSTRING_LEN];
 	int sizeof_stat;
 	int stat_offset;
 };
 
-#define FM10K_NETDEV_STAT(_net_stat) { \
-	.stat_string = #_net_stat, \
-	.sizeof_stat = FIELD_SIZEOF(struct net_device_stats, _net_stat), \
-	.stat_offset = offsetof(struct net_device_stats, _net_stat) \
+#define FM10K_STAT_FIELDS(_type, _name, _stat) { \
+	.stat_string = _name, \
+	.sizeof_stat = sizeof_field(_type, _stat), \
+	.stat_offset = offsetof(_type, _stat) \
 }
 
+/* netdevice statistics */
+#define FM10K_NETDEV_STAT(_net_stat) \
+	FM10K_STAT_FIELDS(struct net_device_stats, __stringify(_net_stat), \
+			  _net_stat)
+
 static const struct fm10k_stats fm10k_gstrings_net_stats[] = {
 	FM10K_NETDEV_STAT(tx_packets),
 	FM10K_NETDEV_STAT(tx_bytes),
@@ -51,11 +45,9 @@ static const struct fm10k_stats fm10k_gstrings_net_stats[] = {
 
 #define FM10K_NETDEV_STATS_LEN	ARRAY_SIZE(fm10k_gstrings_net_stats)
 
-#define FM10K_STAT(_name, _stat) { \
-	.stat_string = _name, \
-	.sizeof_stat = FIELD_SIZEOF(struct fm10k_intfc, _stat), \
-	.stat_offset = offsetof(struct fm10k_intfc, _stat) \
-}
+/* General interface statistics */
+#define FM10K_STAT(_name, _stat) \
+	FM10K_STAT_FIELDS(struct fm10k_intfc, _name, _stat)
 
 static const struct fm10k_stats fm10k_gstrings_global_stats[] = {
 	FM10K_STAT("tx_restart_queue", restart_queue),
@@ -92,11 +84,9 @@ static const struct fm10k_stats fm10k_gstrings_pf_stats[] = {
 	FM10K_STAT("nodesc_drop", stats.nodesc_drop.count),
 };
 
-#define FM10K_MBX_STAT(_name, _stat) { \
-	.stat_string = _name, \
-	.sizeof_stat = FIELD_SIZEOF(struct fm10k_mbx_info, _stat), \
-	.stat_offset = offsetof(struct fm10k_mbx_info, _stat) \
-}
+/* mailbox statistics */
+#define FM10K_MBX_STAT(_name, _stat) \
+	FM10K_STAT_FIELDS(struct fm10k_mbx_info, _name, _stat)
 
 static const struct fm10k_stats fm10k_gstrings_mbx_stats[] = {
 	FM10K_MBX_STAT("mbx_tx_busy", tx_busy),
@@ -110,15 +100,13 @@ static const struct fm10k_stats fm10k_gstrings_mbx_stats[] = {
 	FM10K_MBX_STAT("mbx_rx_mbmem_pushed", rx_mbmem_pushed),
 };
 
-#define FM10K_QUEUE_STAT(_name, _stat) { \
-	.stat_string = _name, \
-	.sizeof_stat = FIELD_SIZEOF(struct fm10k_ring, _stat), \
-	.stat_offset = offsetof(struct fm10k_ring, _stat) \
-}
+/* per-queue ring statistics */
+#define FM10K_QUEUE_STAT(_name, _stat) \
+	FM10K_STAT_FIELDS(struct fm10k_ring, _name, _stat)
 
 static const struct fm10k_stats fm10k_gstrings_queue_stats[] = {
-	FM10K_QUEUE_STAT("packets", stats.packets),
-	FM10K_QUEUE_STAT("bytes", stats.bytes),
+	FM10K_QUEUE_STAT("%s_queue_%u_packets", stats.packets),
+	FM10K_QUEUE_STAT("%s_queue_%u_bytes", stats.bytes),
 };
 
 #define FM10K_GLOBAL_STATS_LEN ARRAY_SIZE(fm10k_gstrings_global_stats)
@@ -148,49 +136,44 @@ enum {
 static const char fm10k_prv_flags[FM10K_PRV_FLAG_LEN][ETH_GSTRING_LEN] = {
 };
 
-static void fm10k_add_stat_strings(u8 **p, const char *prefix,
-				   const struct fm10k_stats stats[],
-				   const unsigned int size)
+static void __fm10k_add_stat_strings(u8 **p, const struct fm10k_stats stats[],
+				     const unsigned int size, ...)
 {
 	unsigned int i;
 
 	for (i = 0; i < size; i++) {
-		snprintf(*p, ETH_GSTRING_LEN, "%s%s",
-			 prefix, stats[i].stat_string);
+		va_list args;
+
+		va_start(args, size);
+		vsnprintf(*p, ETH_GSTRING_LEN, stats[i].stat_string, args);
 		*p += ETH_GSTRING_LEN;
+		va_end(args);
 	}
 }
 
+#define fm10k_add_stat_strings(p, stats, ...) \
+	__fm10k_add_stat_strings(p, stats, ARRAY_SIZE(stats), ## __VA_ARGS__)
+
 static void fm10k_get_stat_strings(struct net_device *dev, u8 *data)
 {
 	struct fm10k_intfc *interface = netdev_priv(dev);
 	unsigned int i;
 
-	fm10k_add_stat_strings(&data, "", fm10k_gstrings_net_stats,
-			       FM10K_NETDEV_STATS_LEN);
+	fm10k_add_stat_strings(&data, fm10k_gstrings_net_stats);
 
-	fm10k_add_stat_strings(&data, "", fm10k_gstrings_global_stats,
-			       FM10K_GLOBAL_STATS_LEN);
+	fm10k_add_stat_strings(&data, fm10k_gstrings_global_stats);
 
-	fm10k_add_stat_strings(&data, "", fm10k_gstrings_mbx_stats,
-			       FM10K_MBX_STATS_LEN);
+	fm10k_add_stat_strings(&data, fm10k_gstrings_mbx_stats);
 
 	if (interface->hw.mac.type != fm10k_mac_vf)
-		fm10k_add_stat_strings(&data, "", fm10k_gstrings_pf_stats,
-				       FM10K_PF_STATS_LEN);
+		fm10k_add_stat_strings(&data, fm10k_gstrings_pf_stats);
 
 	for (i = 0; i < interface->hw.mac.max_queues; i++) {
-		char prefix[ETH_GSTRING_LEN];
+		fm10k_add_stat_strings(&data, fm10k_gstrings_queue_stats,
+				       "tx", i);
 
-		snprintf(prefix, ETH_GSTRING_LEN, "tx_queue_%u_", i);
-		fm10k_add_stat_strings(&data, prefix,
-				       fm10k_gstrings_queue_stats,
-				       FM10K_QUEUE_STATS_LEN);
-
-		snprintf(prefix, ETH_GSTRING_LEN, "rx_queue_%u_", i);
-		fm10k_add_stat_strings(&data, prefix,
-				       fm10k_gstrings_queue_stats,
-				       FM10K_QUEUE_STATS_LEN);
+		fm10k_add_stat_strings(&data, fm10k_gstrings_queue_stats,
+				       "rx", i);
 	}
 }
 
@@ -235,12 +218,11 @@ static int fm10k_get_sset_count(struct net_device *dev, int sset)
 	}
 }
 
-static void fm10k_add_ethtool_stats(u64 **data, void *pointer,
-				    const struct fm10k_stats stats[],
-				    const unsigned int size)
+static void __fm10k_add_ethtool_stats(u64 **data, void *pointer,
+				      const struct fm10k_stats stats[],
+				      const unsigned int size)
 {
 	unsigned int i;
-	char *p;
 
 	if (!pointer) {
 		/* memory is not zero allocated so we have to clear it */
@@ -250,7 +232,7 @@ static void fm10k_add_ethtool_stats(u64 **data, void *pointer,
 	}
 
 	for (i = 0; i < size; i++) {
-		p = (char *)pointer + stats[i].stat_offset;
+		char *p = (char *)pointer + stats[i].stat_offset;
 
 		switch (stats[i].sizeof_stat) {
 		case sizeof(u64):
@@ -266,11 +248,16 @@ static void fm10k_add_ethtool_stats(u64 **data, void *pointer,
 			*((*data)++) = *(u8 *)p;
 			break;
 		default:
+			WARN_ONCE(1, "unexpected stat size for %s",
+				  stats[i].stat_string);
 			*((*data)++) = 0;
 		}
 	}
 }
 
+#define fm10k_add_ethtool_stats(data, pointer, stats) \
+	__fm10k_add_ethtool_stats(data, pointer, stats, ARRAY_SIZE(stats))
+
 static void fm10k_get_ethtool_stats(struct net_device *netdev,
 				    struct ethtool_stats __always_unused *stats,
 				    u64 *data)
@@ -281,20 +268,16 @@ static void fm10k_get_ethtool_stats(struct net_device *netdev,
 
 	fm10k_update_stats(interface);
 
-	fm10k_add_ethtool_stats(&data, net_stats, fm10k_gstrings_net_stats,
-				FM10K_NETDEV_STATS_LEN);
+	fm10k_add_ethtool_stats(&data, net_stats, fm10k_gstrings_net_stats);
 
-	fm10k_add_ethtool_stats(&data, interface, fm10k_gstrings_global_stats,
-				FM10K_GLOBAL_STATS_LEN);
+	fm10k_add_ethtool_stats(&data, interface, fm10k_gstrings_global_stats);
 
 	fm10k_add_ethtool_stats(&data, &interface->hw.mbx,
-				fm10k_gstrings_mbx_stats,
-				FM10K_MBX_STATS_LEN);
+				fm10k_gstrings_mbx_stats);
 
 	if (interface->hw.mac.type != fm10k_mac_vf) {
 		fm10k_add_ethtool_stats(&data, interface,
-					fm10k_gstrings_pf_stats,
-					FM10K_PF_STATS_LEN);
+					fm10k_gstrings_pf_stats);
 	}
 
 	for (i = 0; i < interface->hw.mac.max_queues; i++) {
@@ -302,13 +285,11 @@ static void fm10k_get_ethtool_stats(struct net_device *netdev,
 
 		ring = interface->tx_ring[i];
 		fm10k_add_ethtool_stats(&data, ring,
-					fm10k_gstrings_queue_stats,
-					FM10K_QUEUE_STATS_LEN);
+					fm10k_gstrings_queue_stats);
 
 		ring = interface->rx_ring[i];
 		fm10k_add_ethtool_stats(&data, ring,
-					fm10k_gstrings_queue_stats,
-					FM10K_QUEUE_STATS_LEN);
+					fm10k_gstrings_queue_stats);
 	}
 }
 
@@ -467,12 +448,10 @@ static void fm10k_get_drvinfo(struct net_device *dev,
 {
 	struct fm10k_intfc *interface = netdev_priv(dev);
 
-	strncpy(info->driver, fm10k_driver_name,
-		sizeof(info->driver) - 1);
-	strncpy(info->version, fm10k_driver_version,
-		sizeof(info->version) - 1);
-	strncpy(info->bus_info, pci_name(interface->pdev),
-		sizeof(info->bus_info) - 1);
+	strscpy(info->driver, fm10k_driver_name,
+		sizeof(info->driver));
+	strscpy(info->bus_info, pci_name(interface->pdev),
+		sizeof(info->bus_info));
 }
 
 static void fm10k_get_pauseparam(struct net_device *dev,
@@ -523,7 +502,9 @@ static void fm10k_set_msglevel(struct net_device *netdev, u32 data)
 }
 
 static void fm10k_get_ringparam(struct net_device *netdev,
-				struct ethtool_ringparam *ring)
+				struct ethtool_ringparam *ring,
+				struct kernel_ethtool_ringparam *kernel_ring,
+				struct netlink_ext_ack *extack)
 {
 	struct fm10k_intfc *interface = netdev_priv(netdev);
 
@@ -538,7 +519,9 @@ static void fm10k_get_ringparam(struct net_device *netdev,
 }
 
 static int fm10k_set_ringparam(struct net_device *netdev,
-			       struct ethtool_ringparam *ring)
+			       struct ethtool_ringparam *ring,
+			       struct kernel_ethtool_ringparam *kernel_ring,
+			       struct netlink_ext_ack *extack)
 {
 	struct fm10k_intfc *interface = netdev_priv(netdev);
 	struct fm10k_ring *temp_ring;
@@ -577,7 +560,7 @@ static int fm10k_set_ringparam(struct net_device *netdev,
 
 	/* allocate temporary buffer to store rings in */
 	i = max_t(int, interface->num_tx_queues, interface->num_rx_queues);
-	temp_ring = vmalloc(i * sizeof(struct fm10k_ring));
+	temp_ring = vmalloc_array(i, sizeof(struct fm10k_ring));
 
 	if (!temp_ring) {
 		err = -ENOMEM;
@@ -653,7 +636,9 @@ clear_reset:
 }
 
 static int fm10k_get_coalesce(struct net_device *dev,
-			      struct ethtool_coalesce *ec)
+			      struct ethtool_coalesce *ec,
+			      struct kernel_ethtool_coalesce *kernel_coal,
+			      struct netlink_ext_ack *extack)
 {
 	struct fm10k_intfc *interface = netdev_priv(dev);
 
@@ -667,10 +652,11 @@ static int fm10k_get_coalesce(struct net_device *dev,
 }
 
 static int fm10k_set_coalesce(struct net_device *dev,
-			      struct ethtool_coalesce *ec)
+			      struct ethtool_coalesce *ec,
+			      struct kernel_ethtool_coalesce *kernel_coal,
+			      struct netlink_ext_ack *extack)
 {
 	struct fm10k_intfc *interface = netdev_priv(dev);
-	struct fm10k_q_vector *qv;
 	u16 tx_itr, rx_itr;
 	int i;
 
@@ -696,7 +682,8 @@ static int fm10k_set_coalesce(struct net_device *dev,
 
 	/* update q_vectors */
 	for (i = 0; i < interface->num_q_vectors; i++) {
-		qv = interface->q_vector[i];
+		struct fm10k_q_vector *qv = interface->q_vector[i];
+
 		qv->tx.itr = tx_itr;
 		qv->rx.itr = rx_itr;
 	}
@@ -704,9 +691,11 @@ static int fm10k_set_coalesce(struct net_device *dev,
 	return 0;
 }
 
-static int fm10k_get_rss_hash_opts(struct fm10k_intfc *interface,
-				   struct ethtool_rxnfc *cmd)
+static int fm10k_get_rssh_fields(struct net_device *dev,
+				 struct ethtool_rxfh_fields *cmd)
 {
+	struct fm10k_intfc *interface = netdev_priv(dev);
+
 	cmd->data = 0;
 
 	/* Report default options for RSS on fm10k */
@@ -714,12 +703,12 @@ static int fm10k_get_rss_hash_opts(struct fm10k_intfc *interface,
 	case TCP_V4_FLOW:
 	case TCP_V6_FLOW:
 		cmd->data |= RXH_L4_B_0_1 | RXH_L4_B_2_3;
-		/* fall through */
+		fallthrough;
 	case UDP_V4_FLOW:
 		if (test_bit(FM10K_FLAG_RSS_FIELD_IPV4_UDP,
 			     interface->flags))
 			cmd->data |= RXH_L4_B_0_1 | RXH_L4_B_2_3;
-		/* fall through */
+		fallthrough;
 	case SCTP_V4_FLOW:
 	case SCTP_V6_FLOW:
 	case AH_ESP_V4_FLOW:
@@ -756,9 +745,6 @@ static int fm10k_get_rxnfc(struct net_device *dev, struct ethtool_rxnfc *cmd,
 		cmd->data = interface->num_rx_queues;
 		ret = 0;
 		break;
-	case ETHTOOL_GRXFH:
-		ret = fm10k_get_rss_hash_opts(interface, cmd);
-		break;
 	default:
 		break;
 	}
@@ -766,9 +752,11 @@ static int fm10k_get_rxnfc(struct net_device *dev, struct ethtool_rxnfc *cmd,
 	return ret;
 }
 
-static int fm10k_set_rss_hash_opt(struct fm10k_intfc *interface,
-				  struct ethtool_rxnfc *nfc)
+static int fm10k_set_rssh_fields(struct net_device *dev,
+				 const struct ethtool_rxfh_fields *nfc,
+				 struct netlink_ext_ack *extack)
 {
+	struct fm10k_intfc *interface = netdev_priv(dev);
 	int rss_ipv4_udp = test_bit(FM10K_FLAG_RSS_FIELD_IPV4_UDP,
 				    interface->flags);
 	int rss_ipv6_udp = test_bit(FM10K_FLAG_RSS_FIELD_IPV6_UDP,
@@ -884,22 +872,6 @@ static int fm10k_set_rss_hash_opt(struct fm10k_intfc *interface,
 	return 0;
 }
 
-static int fm10k_set_rxnfc(struct net_device *dev, struct ethtool_rxnfc *cmd)
-{
-	struct fm10k_intfc *interface = netdev_priv(dev);
-	int ret = -EOPNOTSUPP;
-
-	switch (cmd->cmd) {
-	case ETHTOOL_SRXFH:
-		ret = fm10k_set_rss_hash_opt(interface, cmd);
-		break;
-	default:
-		break;
-	}
-
-	return ret;
-}
-
 static int fm10k_mbx_test(struct fm10k_intfc *interface, u64 *data)
 {
 	struct fm10k_hw *hw = &interface->hw;
@@ -1070,16 +1042,16 @@ static u32 fm10k_get_rssrk_size(struct net_device __always_unused *netdev)
 	return FM10K_RSSRK_SIZE * FM10K_RSSRK_ENTRIES_PER_REG;
 }
 
-static int fm10k_get_rssh(struct net_device *netdev, u32 *indir, u8 *key,
-			  u8 *hfunc)
+static int fm10k_get_rssh(struct net_device *netdev,
+			  struct ethtool_rxfh_param *rxfh)
 {
 	struct fm10k_intfc *interface = netdev_priv(netdev);
+	u8 *key = rxfh->key;
 	int i, err;
 
-	if (hfunc)
-		*hfunc = ETH_RSS_HASH_TOP;
+	rxfh->hfunc = ETH_RSS_HASH_TOP;
 
-	err = fm10k_get_reta(netdev, indir);
+	err = fm10k_get_reta(netdev, rxfh->indir);
 	if (err || !key)
 		return err;
 
@@ -1089,23 +1061,25 @@ static int fm10k_get_rssh(struct net_device *netdev, u32 *indir, u8 *key,
 	return 0;
 }
 
-static int fm10k_set_rssh(struct net_device *netdev, const u32 *indir,
-			  const u8 *key, const u8 hfunc)
+static int fm10k_set_rssh(struct net_device *netdev,
+			  struct ethtool_rxfh_param *rxfh,
+			  struct netlink_ext_ack *extack)
 {
 	struct fm10k_intfc *interface = netdev_priv(netdev);
 	struct fm10k_hw *hw = &interface->hw;
 	int i, err;
 
 	/* We do not allow change in unsupported parameters */
-	if (hfunc != ETH_RSS_HASH_NO_CHANGE && hfunc != ETH_RSS_HASH_TOP)
+	if (rxfh->hfunc != ETH_RSS_HASH_NO_CHANGE &&
+	    rxfh->hfunc != ETH_RSS_HASH_TOP)
 		return -EOPNOTSUPP;
 
-	err = fm10k_set_reta(netdev, indir);
-	if (err || !key)
+	err = fm10k_set_reta(netdev, rxfh->indir);
+	if (err || !rxfh->key)
 		return err;
 
-	for (i = 0; i < FM10K_RSSRK_SIZE; i++, key += 4) {
-		u32 rssrk = le32_to_cpu(*(__le32 *)key);
+	for (i = 0; i < FM10K_RSSRK_SIZE; i++, rxfh->key += 4) {
+		u32 rssrk = le32_to_cpu(*(__le32 *)rxfh->key);
 
 		if (interface->rssrk[i] == rssrk)
 			continue;
@@ -1134,13 +1108,12 @@ static void fm10k_get_channels(struct net_device *dev,
 			       struct ethtool_channels *ch)
 {
 	struct fm10k_intfc *interface = netdev_priv(dev);
-	struct fm10k_hw *hw = &interface->hw;
 
 	/* report maximum channels */
 	ch->max_combined = fm10k_max_channels(dev);
 
 	/* report info for other vector */
-	ch->max_other = NON_Q_VECTORS(hw);
+	ch->max_other = NON_Q_VECTORS;
 	ch->other_count = ch->max_other;
 
 	/* record RSS queues */
@@ -1152,14 +1125,13 @@ static int fm10k_set_channels(struct net_device *dev,
 {
 	struct fm10k_intfc *interface = netdev_priv(dev);
 	unsigned int count = ch->combined_count;
-	struct fm10k_hw *hw = &interface->hw;
 
 	/* verify they are not requesting separate vectors */
 	if (!count || ch->rx_count || ch->tx_count)
 		return -EINVAL;
 
 	/* verify other_count has not changed */
-	if (ch->other_count != NON_Q_VECTORS(hw))
+	if (ch->other_count != NON_Q_VECTORS)
 		return -EINVAL;
 
 	/* verify the number of channels does not exceed hardware limits */
@@ -1173,6 +1145,8 @@ static int fm10k_set_channels(struct net_device *dev,
 }
 
 static const struct ethtool_ops fm10k_ethtool_ops = {
+	.supported_coalesce_params = ETHTOOL_COALESCE_USECS |
+				     ETHTOOL_COALESCE_USE_ADAPTIVE,
 	.get_strings		= fm10k_get_strings,
 	.get_sset_count		= fm10k_get_sset_count,
 	.get_ethtool_stats      = fm10k_get_ethtool_stats,
@@ -1187,7 +1161,6 @@ static const struct ethtool_ops fm10k_ethtool_ops = {
 	.get_coalesce		= fm10k_get_coalesce,
 	.set_coalesce		= fm10k_set_coalesce,
 	.get_rxnfc		= fm10k_get_rxnfc,
-	.set_rxnfc		= fm10k_set_rxnfc,
 	.get_regs               = fm10k_get_regs,
 	.get_regs_len           = fm10k_get_regs_len,
 	.self_test		= fm10k_self_test,
@@ -1197,6 +1170,8 @@ static const struct ethtool_ops fm10k_ethtool_ops = {
 	.get_rxfh_key_size	= fm10k_get_rssrk_size,
 	.get_rxfh		= fm10k_get_rssh,
 	.set_rxfh		= fm10k_set_rssh,
+	.get_rxfh_fields	= fm10k_get_rssh_fields,
+	.set_rxfh_fields	= fm10k_set_rssh_fields,
 	.get_channels		= fm10k_get_channels,
 	.set_channels		= fm10k_set_channels,
 	.get_ts_info		= ethtool_op_get_ts_info,
diff --git a/drivers/net/ethernet/intel/fm10k/fm10k_iov.c b/drivers/net/ethernet/intel/fm10k/fm10k_iov.c
index 5f4dac0d36ef..8c50a128df29 100644
--- a/drivers/net/ethernet/intel/fm10k/fm10k_iov.c
+++ b/drivers/net/ethernet/intel/fm10k/fm10k_iov.c
@@ -1,22 +1,5 @@
-/* Intel(R) Ethernet Switch Host Interface Driver
- * Copyright(c) 2013 - 2016 Intel Corporation.
- *
- * This program is free software; you can redistribute it and/or modify it
- * under the terms and conditions of the GNU General Public License,
- * version 2, as published by the Free Software Foundation.
- *
- * This program is distributed in the hope it will be useful, but WITHOUT
- * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
- * FITNESS FOR A PARTICULAR PURPOSE.  See the GNU General Public License for
- * more details.
- *
- * The full GNU General Public License is included in this distribution in
- * the file called "COPYING".
- *
- * Contact Information:
- * e1000-devel Mailing List <e1000-devel@lists.sourceforge.net>
- * Intel Corporation, 5200 N.E. Elam Young Parkway, Hillsboro, OR 97124-6497
- */
+// SPDX-License-Identifier: GPL-2.0
+/* Copyright(c) 2013 - 2019 Intel Corporation. */
 
 #include "fm10k.h"
 #include "fm10k_vf.h"
@@ -35,10 +18,133 @@ static s32 fm10k_iov_msg_error(struct fm10k_hw *hw, u32 **results,
 	return fm10k_tlv_msg_error(hw, results, mbx);
 }
 
+/**
+ *  fm10k_iov_msg_queue_mac_vlan - Message handler for MAC/VLAN request from VF
+ *  @hw: Pointer to hardware structure
+ *  @results: Pointer array to message, results[0] is pointer to message
+ *  @mbx: Pointer to mailbox information structure
+ *
+ *  This function is a custom handler for MAC/VLAN requests from the VF. The
+ *  assumption is that it is acceptable to directly hand off the message from
+ *  the VF to the PF's switch manager. However, we use a MAC/VLAN message
+ *  queue to avoid overloading the mailbox when a large number of requests
+ *  come in.
+ **/
+static s32 fm10k_iov_msg_queue_mac_vlan(struct fm10k_hw *hw, u32 **results,
+					struct fm10k_mbx_info *mbx)
+{
+	struct fm10k_vf_info *vf_info = (struct fm10k_vf_info *)mbx;
+	struct fm10k_intfc *interface = hw->back;
+	u8 mac[ETH_ALEN];
+	u32 *result;
+	int err = 0;
+	bool set;
+	u16 vlan;
+	u32 vid;
+
+	/* we shouldn't be updating rules on a disabled interface */
+	if (!FM10K_VF_FLAG_ENABLED(vf_info))
+		err = FM10K_ERR_PARAM;
+
+	if (!err && !!results[FM10K_MAC_VLAN_MSG_VLAN]) {
+		result = results[FM10K_MAC_VLAN_MSG_VLAN];
+
+		/* record VLAN id requested */
+		err = fm10k_tlv_attr_get_u32(result, &vid);
+		if (err)
+			return err;
+
+		set = !(vid & FM10K_VLAN_CLEAR);
+		vid &= ~FM10K_VLAN_CLEAR;
+
+		/* if the length field has been set, this is a multi-bit
+		 * update request. For multi-bit requests, simply disallow
+		 * them when the pf_vid has been set. In this case, the PF
+		 * should have already cleared the VLAN_TABLE, and if we
+		 * allowed them, it could allow a rogue VF to receive traffic
+		 * on a VLAN it was not assigned. In the single-bit case, we
+		 * need to modify requests for VLAN 0 to use the default PF or
+		 * SW vid when assigned.
+		 */
+
+		if (vid >> 16) {
+			/* prevent multi-bit requests when PF has
+			 * administratively set the VLAN for this VF
+			 */
+			if (vf_info->pf_vid)
+				return FM10K_ERR_PARAM;
+		} else {
+			err = fm10k_iov_select_vid(vf_info, (u16)vid);
+			if (err < 0)
+				return err;
+
+			vid = err;
+		}
+
+		/* update VSI info for VF in regards to VLAN table */
+		err = hw->mac.ops.update_vlan(hw, vid, vf_info->vsi, set);
+	}
+
+	if (!err && !!results[FM10K_MAC_VLAN_MSG_MAC]) {
+		result = results[FM10K_MAC_VLAN_MSG_MAC];
+
+		/* record unicast MAC address requested */
+		err = fm10k_tlv_attr_get_mac_vlan(result, mac, &vlan);
+		if (err)
+			return err;
+
+		/* block attempts to set MAC for a locked device */
+		if (is_valid_ether_addr(vf_info->mac) &&
+		    !ether_addr_equal(mac, vf_info->mac))
+			return FM10K_ERR_PARAM;
+
+		set = !(vlan & FM10K_VLAN_CLEAR);
+		vlan &= ~FM10K_VLAN_CLEAR;
+
+		err = fm10k_iov_select_vid(vf_info, vlan);
+		if (err < 0)
+			return err;
+
+		vlan = (u16)err;
+
+		/* Add this request to the MAC/VLAN queue */
+		err = fm10k_queue_mac_request(interface, vf_info->glort,
+					      mac, vlan, set);
+	}
+
+	if (!err && !!results[FM10K_MAC_VLAN_MSG_MULTICAST]) {
+		result = results[FM10K_MAC_VLAN_MSG_MULTICAST];
+
+		/* record multicast MAC address requested */
+		err = fm10k_tlv_attr_get_mac_vlan(result, mac, &vlan);
+		if (err)
+			return err;
+
+		/* verify that the VF is allowed to request multicast */
+		if (!(vf_info->vf_flags & FM10K_VF_FLAG_MULTI_ENABLED))
+			return FM10K_ERR_PARAM;
+
+		set = !(vlan & FM10K_VLAN_CLEAR);
+		vlan &= ~FM10K_VLAN_CLEAR;
+
+		err = fm10k_iov_select_vid(vf_info, vlan);
+		if (err < 0)
+			return err;
+
+		vlan = (u16)err;
+
+		/* Add this request to the MAC/VLAN queue */
+		err = fm10k_queue_mac_request(interface, vf_info->glort,
+					      mac, vlan, set);
+	}
+
+	return err;
+}
+
 static const struct fm10k_msg_data iov_mbx_data[] = {
 	FM10K_TLV_MSG_TEST_HANDLER(fm10k_tlv_msg_test),
 	FM10K_VF_MSG_MSIX_HANDLER(fm10k_iov_msg_msix_pf),
-	FM10K_VF_MSG_MAC_VLAN_HANDLER(fm10k_iov_msg_mac_vlan_pf),
+	FM10K_VF_MSG_MAC_VLAN_HANDLER(fm10k_iov_msg_queue_mac_vlan),
 	FM10K_VF_MSG_LPORT_STATE_HANDLER(fm10k_iov_msg_lport_state_pf),
 	FM10K_TLV_MSG_ERROR_HANDLER(fm10k_iov_msg_error),
 };
@@ -66,25 +172,21 @@ s32 fm10k_iov_event(struct fm10k_intfc *interface)
 		goto read_unlock;
 
 	/* read VFLRE to determine if any VFs have been reset */
-	do {
-		vflre = fm10k_read_reg(hw, FM10K_PFVFLRE(0));
-		vflre <<= 32;
-		vflre |= fm10k_read_reg(hw, FM10K_PFVFLRE(1));
-		vflre = (vflre << 32) | (vflre >> 32);
-		vflre |= fm10k_read_reg(hw, FM10K_PFVFLRE(0));
+	vflre = fm10k_read_reg(hw, FM10K_PFVFLRE(1));
+	vflre <<= 32;
+	vflre |= fm10k_read_reg(hw, FM10K_PFVFLRE(0));
 
-		i = iov_data->num_vfs;
+	i = iov_data->num_vfs;
 
-		for (vflre <<= 64 - i; vflre && i--; vflre += vflre) {
-			struct fm10k_vf_info *vf_info = &iov_data->vf_info[i];
+	for (vflre <<= 64 - i; vflre && i--; vflre += vflre) {
+		struct fm10k_vf_info *vf_info = &iov_data->vf_info[i];
 
-			if (vflre >= 0)
-				continue;
+		if (vflre >= 0)
+			continue;
 
-			hw->iov.ops.reset_resources(hw, vf_info);
-			vf_info->mbx.ops.connect(hw, &vf_info->mbx);
-		}
-	} while (i != iov_data->num_vfs);
+		hw->iov.ops.reset_resources(hw, vf_info);
+		vf_info->mbx.ops.connect(hw, &vf_info->mbx);
+	}
 
 read_unlock:
 	rcu_read_unlock();
@@ -126,9 +228,14 @@ process_mbx:
 		struct fm10k_mbx_info *mbx = &vf_info->mbx;
 		u16 glort = vf_info->glort;
 
+		/* process the SM mailbox first to drain outgoing messages */
+		hw->mbx.ops.process(hw, &hw->mbx);
+
 		/* verify port mapping is valid, if not reset port */
-		if (vf_info->vf_flags && !fm10k_glort_valid_pf(hw, glort))
+		if (vf_info->vf_flags && !fm10k_glort_valid_pf(hw, glort)) {
 			hw->iov.ops.reset_lport(hw, vf_info);
+			fm10k_clear_macvlan_queue(interface, glort, false);
+		}
 
 		/* reset VFs that have mailbox timed out */
 		if (!mbx->timeout) {
@@ -137,9 +244,14 @@ process_mbx:
 		}
 
 		/* guarantee we have free space in the SM mailbox */
-		if (!hw->mbx.ops.tx_ready(&hw->mbx, FM10K_VFMBX_MSG_MTU)) {
+		if (hw->mbx.state == FM10K_STATE_OPEN &&
+		    !hw->mbx.ops.tx_ready(&hw->mbx, FM10K_VFMBX_MSG_MTU)) {
 			/* keep track of how many times this occurs */
 			interface->hw_sm_mbx_full++;
+
+			/* make sure we try again momentarily */
+			fm10k_service_event_schedule(interface);
+
 			break;
 		}
 
@@ -187,9 +299,30 @@ void fm10k_iov_suspend(struct pci_dev *pdev)
 
 		hw->iov.ops.reset_resources(hw, vf_info);
 		hw->iov.ops.reset_lport(hw, vf_info);
+		fm10k_clear_macvlan_queue(interface, vf_info->glort, false);
 	}
 }
 
+static void fm10k_mask_aer_comp_abort(struct pci_dev *pdev)
+{
+	u32 err_mask;
+	int pos;
+
+	pos = pci_find_ext_capability(pdev, PCI_EXT_CAP_ID_ERR);
+	if (!pos)
+		return;
+
+	/* Mask the completion abort bit in the ERR_UNCOR_MASK register,
+	 * preventing the device from reporting these errors to the upstream
+	 * PCIe root device. This avoids bringing down platforms which upgrade
+	 * non-fatal completer aborts into machine check exceptions. Completer
+	 * aborts can occur whenever a VF reads a queue it doesn't own.
+	 */
+	pci_read_config_dword(pdev, pos + PCI_ERR_UNCOR_MASK, &err_mask);
+	err_mask |= PCI_ERR_UNC_COMP_ABORT;
+	pci_write_config_dword(pdev, pos + PCI_ERR_UNCOR_MASK, err_mask);
+}
+
 int fm10k_iov_resume(struct pci_dev *pdev)
 {
 	struct fm10k_intfc *interface = pci_get_drvdata(pdev);
@@ -205,6 +338,12 @@ int fm10k_iov_resume(struct pci_dev *pdev)
 	if (!iov_data)
 		return -ENOMEM;
 
+	/* Lower severity of completer abort error reporting as
+	 * the VFs can trigger this any time they read a queue
+	 * that they don't own.
+	 */
+	fm10k_mask_aer_comp_abort(pdev);
+
 	/* allocate hardware resources for the VFs */
 	hw->iov.ops.assign_resources(hw, num_vfs, num_vfs);
 
@@ -224,7 +363,7 @@ int fm10k_iov_resume(struct pci_dev *pdev)
 		struct fm10k_vf_info *vf_info = &iov_data->vf_info[i];
 
 		/* allocate all but the last GLORT to the VFs */
-		if (i == ((~hw->mac.dglort_map) >> FM10K_DGLORTMAP_MASK_SHIFT))
+		if (i == (~hw->mac.dglort_map >> FM10K_DGLORTMAP_MASK_SHIFT))
 			break;
 
 		/* assign GLORT to VF, and restrict it to multicast */
@@ -287,7 +426,7 @@ static s32 fm10k_iov_alloc_data(struct pci_dev *pdev, int num_vfs)
 	struct fm10k_iov_data *iov_data = interface->iov_data;
 	struct fm10k_hw *hw = &interface->hw;
 	size_t size;
-	int i, err;
+	int i;
 
 	/* return error if iov_data is already populated */
 	if (iov_data)
@@ -313,6 +452,7 @@ static s32 fm10k_iov_alloc_data(struct pci_dev *pdev, int num_vfs)
 	/* loop through vf_info structures initializing each entry */
 	for (i = 0; i < num_vfs; i++) {
 		struct fm10k_vf_info *vf_info = &iov_data->vf_info[i];
+		int err;
 
 		/* Record VF VSI value */
 		vf_info->vsi = i + 1;
@@ -348,20 +488,6 @@ void fm10k_iov_disable(struct pci_dev *pdev)
 	fm10k_iov_free_data(pdev);
 }
 
-static void fm10k_disable_aer_comp_abort(struct pci_dev *pdev)
-{
-	u32 err_sev;
-	int pos;
-
-	pos = pci_find_ext_capability(pdev, PCI_EXT_CAP_ID_ERR);
-	if (!pos)
-		return;
-
-	pci_read_config_dword(pdev, pos + PCI_ERR_UNCOR_SEVER, &err_sev);
-	err_sev &= ~PCI_ERR_UNC_COMP_ABORT;
-	pci_write_config_dword(pdev, pos + PCI_ERR_UNCOR_SEVER, err_sev);
-}
-
 int fm10k_iov_configure(struct pci_dev *pdev, int num_vfs)
 {
 	int current_vfs = pci_num_vf(pdev);
@@ -382,13 +508,7 @@ int fm10k_iov_configure(struct pci_dev *pdev, int num_vfs)
 		return err;
 
 	/* allocate VFs if not already allocated */
-	if (num_vfs && (num_vfs != current_vfs)) {
-		/* Disable completer abort error reporting as
-		 * the VFs can trigger this any time they read a queue
-		 * that they don't own.
-		 */
-		fm10k_disable_aer_comp_abort(pdev);
-
+	if (num_vfs && num_vfs != current_vfs) {
 		err = pci_enable_sriov(pdev, num_vfs);
 		if (err) {
 			dev_err(&pdev->dev,
@@ -400,6 +520,27 @@ int fm10k_iov_configure(struct pci_dev *pdev, int num_vfs)
 	return num_vfs;
 }
 
+/**
+ * fm10k_iov_update_stats - Update stats for all VFs
+ * @interface: device private structure
+ *
+ * Updates the VF statistics for all enabled VFs. Expects to be called by
+ * fm10k_update_stats and assumes that locking via the __FM10K_UPDATING_STATS
+ * bit is already handled.
+ */
+void fm10k_iov_update_stats(struct fm10k_intfc *interface)
+{
+	struct fm10k_iov_data *iov_data = interface->iov_data;
+	struct fm10k_hw *hw = &interface->hw;
+	int i;
+
+	if (!iov_data)
+		return;
+
+	for (i = 0; i < iov_data->num_vfs; i++)
+		hw->iov.ops.update_stats(hw, iov_data->vf_info[i].stats, i);
+}
+
 static inline void fm10k_reset_vf_info(struct fm10k_intfc *interface,
 				       struct fm10k_vf_info *vf_info)
 {
@@ -411,6 +552,8 @@ static inline void fm10k_reset_vf_info(struct fm10k_intfc *interface,
 	/* disable LPORT for this VF which clears switch rules */
 	hw->iov.ops.reset_lport(hw, vf_info);
 
+	fm10k_clear_macvlan_queue(interface, vf_info->glort, false);
+
 	/* assign new MAC+VLAN for this VF */
 	hw->iov.ops.assign_default_mac_vlan(hw, vf_info);
 
@@ -482,7 +625,7 @@ int fm10k_ndo_set_vf_vlan(struct net_device *netdev, int vf_idx, u16 vid,
 }
 
 int fm10k_ndo_set_vf_bw(struct net_device *netdev, int vf_idx,
-			int __always_unused unused, int rate)
+			int __always_unused min_rate, int max_rate)
 {
 	struct fm10k_intfc *interface = netdev_priv(netdev);
 	struct fm10k_iov_data *iov_data = interface->iov_data;
@@ -493,14 +636,15 @@ int fm10k_ndo_set_vf_bw(struct net_device *netdev, int vf_idx,
 		return -EINVAL;
 
 	/* rate limit cannot be less than 10Mbs or greater than link speed */
-	if (rate && ((rate < FM10K_VF_TC_MIN) || rate > FM10K_VF_TC_MAX))
+	if (max_rate &&
+	    (max_rate < FM10K_VF_TC_MIN || max_rate > FM10K_VF_TC_MAX))
 		return -EINVAL;
 
 	/* store values */
-	iov_data->vf_info[vf_idx].rate = rate;
+	iov_data->vf_info[vf_idx].rate = max_rate;
 
 	/* update hardware configuration */
-	hw->iov.ops.configure_tc(hw, vf_idx, rate);
+	hw->iov.ops.configure_tc(hw, vf_idx, max_rate);
 
 	return 0;
 }
@@ -527,3 +671,30 @@ int fm10k_ndo_get_vf_config(struct net_device *netdev,
 
 	return 0;
 }
+
+int fm10k_ndo_get_vf_stats(struct net_device *netdev,
+			   int vf_idx, struct ifla_vf_stats *stats)
+{
+	struct fm10k_intfc *interface = netdev_priv(netdev);
+	struct fm10k_iov_data *iov_data = interface->iov_data;
+	struct fm10k_hw *hw = &interface->hw;
+	struct fm10k_hw_stats_q *hw_stats;
+	u32 idx, qpp;
+
+	/* verify SR-IOV is active and that vf idx is valid */
+	if (!iov_data || vf_idx >= iov_data->num_vfs)
+		return -EINVAL;
+
+	qpp = fm10k_queues_per_pool(hw);
+	hw_stats = iov_data->vf_info[vf_idx].stats;
+
+	for (idx = 0; idx < qpp; idx++) {
+		stats->rx_packets += hw_stats[idx].rx_packets.count;
+		stats->tx_packets += hw_stats[idx].tx_packets.count;
+		stats->rx_bytes += hw_stats[idx].rx_bytes.count;
+		stats->tx_bytes += hw_stats[idx].tx_bytes.count;
+		stats->rx_dropped += hw_stats[idx].rx_drops.count;
+	}
+
+	return 0;
+}
diff --git a/drivers/net/ethernet/intel/fm10k/fm10k_main.c b/drivers/net/ethernet/intel/fm10k/fm10k_main.c
index 9dffaba85ae6..b8c15b837fda 100644
--- a/drivers/net/ethernet/intel/fm10k/fm10k_main.c
+++ b/drivers/net/ethernet/intel/fm10k/fm10k_main.c
@@ -1,22 +1,5 @@
-/* Intel(R) Ethernet Switch Host Interface Driver
- * Copyright(c) 2013 - 2017 Intel Corporation.
- *
- * This program is free software; you can redistribute it and/or modify it
- * under the terms and conditions of the GNU General Public License,
- * version 2, as published by the Free Software Foundation.
- *
- * This program is distributed in the hope it will be useful, but WITHOUT
- * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
- * FITNESS FOR A PARTICULAR PURPOSE.  See the GNU General Public License for
- * more details.
- *
- * The full GNU General Public License is included in this distribution in
- * the file called "COPYING".
- *
- * Contact Information:
- * e1000-devel Mailing List <e1000-devel@lists.sourceforge.net>
- * Intel Corporation, 5200 N.E. Elam Young Parkway, Hillsboro, OR 97124-6497
- */
+// SPDX-License-Identifier: GPL-2.0
+/* Copyright(c) 2013 - 2019 Intel Corporation. */
 
 #include <linux/types.h>
 #include <linux/module.h>
@@ -28,18 +11,14 @@
 
 #include "fm10k.h"
 
-#define DRV_VERSION	"0.21.7-k"
 #define DRV_SUMMARY	"Intel(R) Ethernet Switch Host Interface Driver"
-const char fm10k_driver_version[] = DRV_VERSION;
 char fm10k_driver_name[] = "fm10k";
 static const char fm10k_driver_string[] = DRV_SUMMARY;
 static const char fm10k_copyright[] =
-	"Copyright(c) 2013 - 2017 Intel Corporation.";
+	"Copyright(c) 2013 - 2019 Intel Corporation.";
 
-MODULE_AUTHOR("Intel Corporation, <linux.nics@intel.com>");
 MODULE_DESCRIPTION(DRV_SUMMARY);
-MODULE_LICENSE("GPL");
-MODULE_VERSION(DRV_VERSION);
+MODULE_LICENSE("GPL v2");
 
 /* single workqueue for entire fm10k driver */
 struct workqueue_struct *fm10k_workqueue;
@@ -52,16 +31,26 @@ struct workqueue_struct *fm10k_workqueue;
  **/
 static int __init fm10k_init_module(void)
 {
-	pr_info("%s - version %s\n", fm10k_driver_string, fm10k_driver_version);
+	int ret;
+
+	pr_info("%s\n", fm10k_driver_string);
 	pr_info("%s\n", fm10k_copyright);
 
 	/* create driver workqueue */
-	fm10k_workqueue = alloc_workqueue("%s", WQ_MEM_RECLAIM, 0,
+	fm10k_workqueue = alloc_workqueue("%s", WQ_MEM_RECLAIM | WQ_PERCPU, 0,
 					  fm10k_driver_name);
+	if (!fm10k_workqueue)
+		return -ENOMEM;
 
 	fm10k_dbg_init();
 
-	return fm10k_register_pci_driver();
+	ret = fm10k_register_pci_driver();
+	if (ret) {
+		fm10k_dbg_exit();
+		destroy_workqueue(fm10k_workqueue);
+	}
+
+	return ret;
 }
 module_init(fm10k_init_module);
 
@@ -212,17 +201,12 @@ static void fm10k_reuse_rx_page(struct fm10k_ring *rx_ring,
 					 DMA_FROM_DEVICE);
 }
 
-static inline bool fm10k_page_is_reserved(struct page *page)
-{
-	return (page_to_nid(page) != numa_mem_id()) || page_is_pfmemalloc(page);
-}
-
 static bool fm10k_can_reuse_rx_page(struct fm10k_rx_buffer *rx_buffer,
 				    struct page *page,
 				    unsigned int __maybe_unused truesize)
 {
-	/* avoid re-using remote pages */
-	if (unlikely(fm10k_page_is_reserved(page)))
+	/* avoid re-using remote and pfmemalloc pages */
+	if (!dev_page_is_reusable(page))
 		return false;
 
 #if (PAGE_SIZE < 8192)
@@ -283,8 +267,8 @@ static bool fm10k_add_rx_frag(struct fm10k_rx_buffer *rx_buffer,
 	if (likely(size <= FM10K_RX_HDR_LEN)) {
 		memcpy(__skb_put(skb, size), va, ALIGN(size, sizeof(long)));
 
-		/* page is not reserved, we can reuse buffer as-is */
-		if (likely(!fm10k_page_is_reserved(page)))
+		/* page is reusable, we can reuse buffer as-is */
+		if (dev_page_is_reusable(page))
 			return true;
 
 		/* this page cannot be reused so discard it */
@@ -295,7 +279,7 @@ static bool fm10k_add_rx_frag(struct fm10k_rx_buffer *rx_buffer,
 	/* we need the header to contain the greater of either ETH_HLEN or
 	 * 60 bytes if the skb->len is less than 60 for skb_pad.
 	 */
-	pull_len = eth_get_headlen(va, FM10K_RX_HDR_LEN);
+	pull_len = eth_get_headlen(skb->dev, va, FM10K_RX_HDR_LEN);
 
 	/* align pull length to size of long to optimize memcpy performance */
 	memcpy(__skb_put(skb, pull_len), va, ALIGN(pull_len, sizeof(long)));
@@ -328,10 +312,7 @@ static struct sk_buff *fm10k_fetch_rx_buffer(struct fm10k_ring *rx_ring,
 				  rx_buffer->page_offset;
 
 		/* prefetch first cache line of first page */
-		prefetch(page_addr);
-#if L1_CACHE_BYTES < 128
-		prefetch(page_addr + L1_CACHE_BYTES);
-#endif
+		net_prefetch(page_addr);
 
 		/* allocate a skb to store the frags */
 		skb = napi_alloc_skb(&rx_ring->q_vector->napi,
@@ -444,15 +425,14 @@ static void fm10k_type_trans(struct fm10k_ring *rx_ring,
 			l2_accel = NULL;
 	}
 
-	skb->protocol = eth_type_trans(skb, dev);
-
+	/* Record Rx queue, or update macvlan statistics */
 	if (!l2_accel)
-		return;
+		skb_record_rx_queue(skb, rx_ring->queue_index);
+	else
+		macvlan_count_rx(netdev_priv(dev), skb->len + ETH_HLEN, true,
+				 false);
 
-	/* update MACVLAN statistics */
-	macvlan_count_rx(netdev_priv(dev), skb->len + ETH_HLEN, 1,
-			 !!(rx_desc->w.hdr_info &
-			    cpu_to_le16(FM10K_RXD_HDR_INFO_XC_MASK)));
+	skb->protocol = eth_type_trans(skb, dev);
 }
 
 /**
@@ -479,8 +459,6 @@ static unsigned int fm10k_process_skb_fields(struct fm10k_ring *rx_ring,
 
 	FM10K_CB(skb)->fi.w.vlan = rx_desc->w.vlan;
 
-	skb_record_rx_queue(skb, rx_ring->queue_index);
-
 	FM10K_CB(skb)->fi.d.glort = rx_desc->d.glort;
 
 	if (rx_desc->w.vlan) {
@@ -656,15 +634,8 @@ static int fm10k_clean_rx_irq(struct fm10k_q_vector *q_vector,
 static struct ethhdr *fm10k_port_is_vxlan(struct sk_buff *skb)
 {
 	struct fm10k_intfc *interface = netdev_priv(skb->dev);
-	struct fm10k_udp_port *vxlan_port;
 
-	/* we can only offload a vxlan if we recognize it as such */
-	vxlan_port = list_first_entry_or_null(&interface->vxlan_port,
-					      struct fm10k_udp_port, list);
-
-	if (!vxlan_port)
-		return NULL;
-	if (vxlan_port->port != udp_hdr(skb)->dest)
+	if (interface->vxlan_port != udp_hdr(skb)->dest)
 		return NULL;
 
 	/* return offset of udp_hdr plus 8 bytes for VXLAN header */
@@ -806,9 +777,10 @@ static int fm10k_tso(struct fm10k_ring *tx_ring,
 	tx_desc->mss = cpu_to_le16(skb_shinfo(skb)->gso_size);
 
 	return 1;
+
 err_vxlan:
 	tx_ring->netdev->features &= ~NETIF_F_GSO_UDP_TUNNEL;
-	if (!net_ratelimit())
+	if (net_ratelimit())
 		netdev_err(tx_ring->netdev,
 			   "TSO requested for unsupported tunnel, disabling offload\n");
 	return -1;
@@ -876,6 +848,7 @@ static void fm10k_tx_csum(struct fm10k_ring *tx_ring,
 	case IPPROTO_GRE:
 		if (skb->encapsulation)
 			break;
+		fallthrough;
 	default:
 		if (unlikely(net_ratelimit())) {
 			dev_warn(tx_ring->dev,
@@ -962,7 +935,7 @@ static void fm10k_tx_map(struct fm10k_ring *tx_ring,
 	struct sk_buff *skb = first->skb;
 	struct fm10k_tx_buffer *tx_buffer;
 	struct fm10k_tx_desc *tx_desc;
-	struct skb_frag_struct *frag;
+	skb_frag_t *frag;
 	unsigned char *data;
 	dma_addr_t dma;
 	unsigned int data_len, size;
@@ -1053,13 +1026,8 @@ static void fm10k_tx_map(struct fm10k_ring *tx_ring,
 	fm10k_maybe_stop_tx(tx_ring, DESC_NEEDED);
 
 	/* notify HW of packet */
-	if (netif_xmit_stopped(txring_txq(tx_ring)) || !skb->xmit_more) {
+	if (netif_xmit_stopped(txring_txq(tx_ring)) || !netdev_xmit_more()) {
 		writel(i, tx_ring->tail);
-
-		/* we need this if more than one processor can write to our tail
-		 * at a time, it synchronizes IO on IA64/Altix systems
-		 */
-		mmiowb();
 	}
 
 	return;
@@ -1094,8 +1062,11 @@ netdev_tx_t fm10k_xmit_frame_ring(struct sk_buff *skb,
 	 *       + 2 desc gap to keep tail from touching head
 	 * otherwise try next time
 	 */
-	for (f = 0; f < skb_shinfo(skb)->nr_frags; f++)
-		count += TXD_USE_COUNT(skb_shinfo(skb)->frags[f].size);
+	for (f = 0; f < skb_shinfo(skb)->nr_frags; f++) {
+		skb_frag_t *frag = &skb_shinfo(skb)->frags[f];
+
+		count += TXD_USE_COUNT(skb_frag_size(frag));
+	}
 
 	if (fm10k_maybe_stop_tx(tx_ring, count + 3)) {
 		tx_ring->tx_stats.tx_busy++;
@@ -1229,7 +1200,7 @@ static bool fm10k_clean_tx_irq(struct fm10k_q_vector *q_vector,
 			break;
 
 		/* prevent any other reads prior to eop_desc */
-		read_barrier_depends();
+		smp_rmb();
 
 		/* if DD is not set pending work has not been completed */
 		if (!(eop_desc->flags & FM10K_TXD_FLAG_DONE))
@@ -1483,11 +1454,11 @@ static int fm10k_poll(struct napi_struct *napi, int budget)
 	if (!clean_complete)
 		return budget;
 
-	/* all work done, exit the polling mode */
-	napi_complete_done(napi, work_done);
-
-	/* re-enable the q_vector */
-	fm10k_qv_enable(q_vector);
+	/* Exit the polling mode, but don't re-enable interrupts if stack might
+	 * poll us due to busy-polling
+	 */
+	if (likely(napi_complete_done(napi, work_done)))
+		fm10k_qv_enable(q_vector);
 
 	return min(work_done, budget - 1);
 }
@@ -1575,7 +1546,7 @@ static bool fm10k_set_rss_queues(struct fm10k_intfc *interface)
  * important, starting with the "most" number of features turned on at once,
  * and ending with the smallest set of features.  This way large combinations
  * can be allocated if they're turned on, and smaller combinations are the
- * fallthrough conditions.
+ * fall through conditions.
  *
  **/
 static void fm10k_set_num_queues(struct fm10k_intfc *interface)
@@ -1621,20 +1592,17 @@ static int fm10k_alloc_q_vector(struct fm10k_intfc *interface,
 {
 	struct fm10k_q_vector *q_vector;
 	struct fm10k_ring *ring;
-	int ring_count, size;
+	int ring_count;
 
 	ring_count = txr_count + rxr_count;
-	size = sizeof(struct fm10k_q_vector) +
-	       (sizeof(struct fm10k_ring) * ring_count);
 
 	/* allocate q_vector and rings */
-	q_vector = kzalloc(size, GFP_KERNEL);
+	q_vector = kzalloc(struct_size(q_vector, ring, ring_count), GFP_KERNEL);
 	if (!q_vector)
 		return -ENOMEM;
 
 	/* initialize NAPI */
-	netif_napi_add(interface->netdev, &q_vector->napi,
-		       fm10k_poll, NAPI_POLL_WEIGHT);
+	netif_napi_add(interface->netdev, &q_vector->napi, fm10k_poll);
 
 	/* tie q_vector and interface together */
 	interface->q_vector[v_idx] = q_vector;
@@ -1812,7 +1780,7 @@ static void fm10k_free_q_vectors(struct fm10k_intfc *interface)
 }
 
 /**
- * f10k_reset_msix_capability - reset MSI-X capability
+ * fm10k_reset_msix_capability - reset MSI-X capability
  * @interface: board private structure to initialize
  *
  * Reset the MSI-X capability back to its starting state
@@ -1825,7 +1793,7 @@ static void fm10k_reset_msix_capability(struct fm10k_intfc *interface)
 }
 
 /**
- * f10k_init_msix_capability - configure MSI-X capability
+ * fm10k_init_msix_capability - configure MSI-X capability
  * @interface: board private structure to initialize
  *
  * Attempt to configure the interrupts using the best available
@@ -1846,7 +1814,7 @@ static int fm10k_init_msix_capability(struct fm10k_intfc *interface)
 	v_budget = min_t(u16, v_budget, num_online_cpus());
 
 	/* account for vectors not related to queues */
-	v_budget += NON_Q_VECTORS(hw);
+	v_budget += NON_Q_VECTORS;
 
 	/* At the same time, hardware can only support a maximum of
 	 * hw.mac->max_msix_vectors vectors.  With features
@@ -1878,7 +1846,7 @@ static int fm10k_init_msix_capability(struct fm10k_intfc *interface)
 	}
 
 	/* record the number of queues available for q_vectors */
-	interface->num_q_vectors = v_budget - NON_Q_VECTORS(hw);
+	interface->num_q_vectors = v_budget - NON_Q_VECTORS;
 
 	return 0;
 }
@@ -1892,7 +1860,7 @@ static int fm10k_init_msix_capability(struct fm10k_intfc *interface)
 static bool fm10k_cache_ring_qos(struct fm10k_intfc *interface)
 {
 	struct net_device *dev = interface->netdev;
-	int pc, offset, rss_i, i, q_idx;
+	int pc, offset, rss_i, i;
 	u16 pc_stride = interface->ring_feature[RING_F_QOS].mask + 1;
 	u8 num_pcs = netdev_get_num_tc(dev);
 
@@ -1902,7 +1870,8 @@ static bool fm10k_cache_ring_qos(struct fm10k_intfc *interface)
 	rss_i = interface->ring_feature[RING_F_RSS].indices;
 
 	for (pc = 0, offset = 0; pc < num_pcs; pc++, offset += rss_i) {
-		q_idx = pc;
+		int q_idx = pc;
+
 		for (i = 0; i < rss_i; i++) {
 			interface->tx_ring[offset + i]->reg_idx = q_idx;
 			interface->tx_ring[offset + i]->qos_pc = pc;
diff --git a/drivers/net/ethernet/intel/fm10k/fm10k_mbx.c b/drivers/net/ethernet/intel/fm10k/fm10k_mbx.c
index 334088a101c3..87fa5874f16e 100644
--- a/drivers/net/ethernet/intel/fm10k/fm10k_mbx.c
+++ b/drivers/net/ethernet/intel/fm10k/fm10k_mbx.c
@@ -1,22 +1,5 @@
-/* Intel(R) Ethernet Switch Host Interface Driver
- * Copyright(c) 2013 - 2016 Intel Corporation.
- *
- * This program is free software; you can redistribute it and/or modify it
- * under the terms and conditions of the GNU General Public License,
- * version 2, as published by the Free Software Foundation.
- *
- * This program is distributed in the hope it will be useful, but WITHOUT
- * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
- * FITNESS FOR A PARTICULAR PURPOSE.  See the GNU General Public License for
- * more details.
- *
- * The full GNU General Public License is included in this distribution in
- * the file called "COPYING".
- *
- * Contact Information:
- * e1000-devel Mailing List <e1000-devel@lists.sourceforge.net>
- * Intel Corporation, 5200 N.E. Elam Young Parkway, Hillsboro, OR 97124-6497
- */
+// SPDX-License-Identifier: GPL-2.0
+/* Copyright(c) 2013 - 2019 Intel Corporation. */
 
 #include "fm10k_common.h"
 
@@ -314,13 +297,14 @@ static u16 fm10k_mbx_validate_msg_size(struct fm10k_mbx_info *mbx, u16 len)
 {
 	struct fm10k_mbx_fifo *fifo = &mbx->rx;
 	u16 total_len = 0, msg_len;
-	u32 *msg;
 
 	/* length should include previous amounts pushed */
 	len += mbx->pushed;
 
 	/* offset in message is based off of current message size */
 	do {
+		u32 *msg;
+
 		msg = fifo->buffer + fm10k_fifo_tail_offset(fifo, total_len);
 		msg_len = FM10K_TLV_DWORD_LEN(*msg);
 		total_len += msg_len;
@@ -708,7 +692,7 @@ static bool fm10k_mbx_tx_complete(struct fm10k_mbx_info *mbx)
 }
 
 /**
- *  fm10k_mbx_deqeueue_rx - Dequeues the message from the head in the Rx FIFO
+ *  fm10k_mbx_dequeue_rx - Dequeues the message from the head in the Rx FIFO
  *  @hw: pointer to hardware structure
  *  @mbx: pointer to mailbox
  *
@@ -825,7 +809,7 @@ static s32 fm10k_mbx_read(struct fm10k_hw *hw, struct fm10k_mbx_info *mbx)
  *  @hw: pointer to hardware structure
  *  @mbx: pointer to mailbox
  *
- *  This function copies the message from the the message array to mbmem
+ *  This function copies the message from the message array to mbmem
  **/
 static void fm10k_mbx_write(struct fm10k_hw *hw, struct fm10k_mbx_info *mbx)
 {
@@ -983,7 +967,7 @@ static s32 fm10k_mbx_validate_msg_hdr(struct fm10k_mbx_info *mbx)
 		if (tail != mbx->head)
 			return FM10K_MBX_ERR_TAIL;
 
-		/* fall through */
+		fallthrough;
 	case FM10K_MSG_DATA:
 		/* validate that head is moving correctly */
 		if (!head || (head == FM10K_MSG_HDR_MASK(HEAD)))
@@ -1003,7 +987,7 @@ static s32 fm10k_mbx_validate_msg_hdr(struct fm10k_mbx_info *mbx)
 		if ((size < FM10K_VFMBX_MSG_MTU) || (size & (size + 1)))
 			return FM10K_MBX_ERR_SIZE;
 
-		/* fall through */
+		fallthrough;
 	case FM10K_MSG_ERROR:
 		if (!head || (head == FM10K_MSG_HDR_MASK(HEAD)))
 			return FM10K_MBX_ERR_HEAD;
@@ -1055,6 +1039,7 @@ static s32 fm10k_mbx_create_reply(struct fm10k_hw *hw,
 	case FM10K_STATE_CLOSED:
 		/* generate new header based on data */
 		fm10k_mbx_create_disconnect_hdr(mbx);
+		break;
 	default:
 		break;
 	}
@@ -1586,7 +1571,7 @@ s32 fm10k_pfvf_mbx_init(struct fm10k_hw *hw, struct fm10k_mbx_info *mbx,
 			mbx->mbmem_reg = FM10K_MBMEM_VF(id, 0);
 			break;
 		}
-		/* fallthough */
+		fallthrough;
 	default:
 		return FM10K_MBX_ERR_NO_MBX;
 	}
@@ -1840,7 +1825,7 @@ static void fm10k_sm_mbx_process_error(struct fm10k_mbx_info *mbx)
 		fm10k_sm_mbx_connect_reset(mbx);
 		break;
 	case FM10K_STATE_CONNECT:
-		/* try connnecting at lower version */
+		/* try connecting at lower version */
 		if (mbx->remote) {
 			while (mbx->local > 1)
 				mbx->local--;
@@ -1937,7 +1922,6 @@ static void fm10k_sm_mbx_transmit(struct fm10k_hw *hw,
 	/* reduce length by 1 to convert to a mask */
 	u16 mbmem_len = mbx->mbmem_len - 1;
 	u16 tail_len, len = 0;
-	u32 *msg;
 
 	/* push head behind tail */
 	if (mbx->tail < head)
@@ -1947,6 +1931,8 @@ static void fm10k_sm_mbx_transmit(struct fm10k_hw *hw,
 
 	/* determine msg aligned offset for end of buffer */
 	do {
+		u32 *msg;
+
 		msg = fifo->buffer + fm10k_fifo_head_offset(fifo, len);
 		tail_len = len;
 		len += FM10K_TLV_DWORD_LEN(*msg);
@@ -2032,6 +2018,7 @@ static s32 fm10k_sm_mbx_process_reset(struct fm10k_hw *hw,
 	case FM10K_STATE_CONNECT:
 		/* Update remote value to match local value */
 		mbx->remote = mbx->local;
+		break;
 	default:
 		break;
 	}
@@ -2149,7 +2136,8 @@ fifo_err:
  *  DWORDs, not bytes.  Any invalid values will cause the mailbox to return
  *  error.
  **/
-s32 fm10k_sm_mbx_init(struct fm10k_hw *hw, struct fm10k_mbx_info *mbx,
+s32 fm10k_sm_mbx_init(struct fm10k_hw __always_unused *hw,
+		      struct fm10k_mbx_info *mbx,
 		      const struct fm10k_msg_data *msg_data)
 {
 	mbx->mbx_reg = FM10K_GMBX;
diff --git a/drivers/net/ethernet/intel/fm10k/fm10k_mbx.h b/drivers/net/ethernet/intel/fm10k/fm10k_mbx.h
index 35c1dbad1330..56d1abff04e2 100644
--- a/drivers/net/ethernet/intel/fm10k/fm10k_mbx.h
+++ b/drivers/net/ethernet/intel/fm10k/fm10k_mbx.h
@@ -1,22 +1,5 @@
-/* Intel(R) Ethernet Switch Host Interface Driver
- * Copyright(c) 2013 - 2016 Intel Corporation.
- *
- * This program is free software; you can redistribute it and/or modify it
- * under the terms and conditions of the GNU General Public License,
- * version 2, as published by the Free Software Foundation.
- *
- * This program is distributed in the hope it will be useful, but WITHOUT
- * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
- * FITNESS FOR A PARTICULAR PURPOSE.  See the GNU General Public License for
- * more details.
- *
- * The full GNU General Public License is included in this distribution in
- * the file called "COPYING".
- *
- * Contact Information:
- * e1000-devel Mailing List <e1000-devel@lists.sourceforge.net>
- * Intel Corporation, 5200 N.E. Elam Young Parkway, Hillsboro, OR 97124-6497
- */
+/* SPDX-License-Identifier: GPL-2.0 */
+/* Copyright(c) 2013 - 2018 Intel Corporation. */
 
 #ifndef _FM10K_MBX_H_
 #define _FM10K_MBX_H_
diff --git a/drivers/net/ethernet/intel/fm10k/fm10k_netdev.c b/drivers/net/ethernet/intel/fm10k/fm10k_netdev.c
index 24f2f6f86f5a..34ab5ff9823b 100644
--- a/drivers/net/ethernet/intel/fm10k/fm10k_netdev.c
+++ b/drivers/net/ethernet/intel/fm10k/fm10k_netdev.c
@@ -1,26 +1,10 @@
-/* Intel(R) Ethernet Switch Host Interface Driver
- * Copyright(c) 2013 - 2017 Intel Corporation.
- *
- * This program is free software; you can redistribute it and/or modify it
- * under the terms and conditions of the GNU General Public License,
- * version 2, as published by the Free Software Foundation.
- *
- * This program is distributed in the hope it will be useful, but WITHOUT
- * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
- * FITNESS FOR A PARTICULAR PURPOSE.  See the GNU General Public License for
- * more details.
- *
- * The full GNU General Public License is included in this distribution in
- * the file called "COPYING".
- *
- * Contact Information:
- * e1000-devel Mailing List <e1000-devel@lists.sourceforge.net>
- * Intel Corporation, 5200 N.E. Elam Young Parkway, Hillsboro, OR 97124-6497
- */
+// SPDX-License-Identifier: GPL-2.0
+/* Copyright(c) 2013 - 2019 Intel Corporation. */
 
 #include "fm10k.h"
 #include <linux/vmalloc.h>
 #include <net/udp_tunnel.h>
+#include <linux/if_macvlan.h>
 
 /**
  * fm10k_setup_tx_resources - allocate Tx resources (Descriptors)
@@ -70,7 +54,7 @@ err:
  **/
 static int fm10k_setup_all_tx_resources(struct fm10k_intfc *interface)
 {
-	int i, err = 0;
+	int i, err;
 
 	for (i = 0; i < interface->num_tx_queues; i++) {
 		err = fm10k_setup_tx_resources(interface->tx_ring[i]);
@@ -137,7 +121,7 @@ err:
  **/
 static int fm10k_setup_all_rx_resources(struct fm10k_intfc *interface)
 {
-	int i, err = 0;
+	int i, err;
 
 	for (i = 0; i < interface->num_rx_queues; i++) {
 		err = fm10k_setup_rx_resources(interface->rx_ring[i]);
@@ -185,7 +169,6 @@ void fm10k_unmap_and_free_tx_resource(struct fm10k_ring *ring,
  **/
 static void fm10k_clean_tx_ring(struct fm10k_ring *tx_ring)
 {
-	struct fm10k_tx_buffer *tx_buffer;
 	unsigned long size;
 	u16 i;
 
@@ -195,7 +178,8 @@ static void fm10k_clean_tx_ring(struct fm10k_ring *tx_ring)
 
 	/* Free all the Tx ring sk_buffs */
 	for (i = 0; i < tx_ring->count; i++) {
-		tx_buffer = &tx_ring->tx_buffer[i];
+		struct fm10k_tx_buffer *tx_buffer = &tx_ring->tx_buffer[i];
+
 		fm10k_unmap_and_free_tx_resource(tx_ring, tx_buffer);
 	}
 
@@ -269,8 +253,7 @@ static void fm10k_clean_rx_ring(struct fm10k_ring *rx_ring)
 	if (!rx_ring->rx_buffer)
 		return;
 
-	if (rx_ring->skb)
-		dev_kfree_skb(rx_ring->skb);
+	dev_kfree_skb(rx_ring->skb);
 	rx_ring->skb = NULL;
 
 	/* Free all the Rx ring sk_buffs */
@@ -384,39 +367,6 @@ static void fm10k_request_glort_range(struct fm10k_intfc *interface)
 }
 
 /**
- * fm10k_free_udp_port_info
- * @interface: board private structure
- *
- * This function frees both geneve_port and vxlan_port structures
- **/
-static void fm10k_free_udp_port_info(struct fm10k_intfc *interface)
-{
-	struct fm10k_udp_port *port;
-
-	/* flush all entries from vxlan list */
-	port = list_first_entry_or_null(&interface->vxlan_port,
-					struct fm10k_udp_port, list);
-	while (port) {
-		list_del(&port->list);
-		kfree(port);
-		port = list_first_entry_or_null(&interface->vxlan_port,
-						struct fm10k_udp_port,
-						list);
-	}
-
-	/* flush all entries from geneve list */
-	port = list_first_entry_or_null(&interface->geneve_port,
-					struct fm10k_udp_port, list);
-	while (port) {
-		list_del(&port->list);
-		kfree(port);
-		port = list_first_entry_or_null(&interface->vxlan_port,
-						struct fm10k_udp_port,
-						list);
-	}
-}
-
-/**
  * fm10k_restore_udp_port_info
  * @interface: board private structure
  *
@@ -425,131 +375,52 @@ static void fm10k_free_udp_port_info(struct fm10k_intfc *interface)
 static void fm10k_restore_udp_port_info(struct fm10k_intfc *interface)
 {
 	struct fm10k_hw *hw = &interface->hw;
-	struct fm10k_udp_port *port;
 
 	/* only the PF supports configuring tunnels */
 	if (hw->mac.type != fm10k_mac_pf)
 		return;
 
-	port = list_first_entry_or_null(&interface->vxlan_port,
-					struct fm10k_udp_port, list);
-
 	/* restore tunnel configuration register */
 	fm10k_write_reg(hw, FM10K_TUNNEL_CFG,
-			(port ? ntohs(port->port) : 0) |
+			ntohs(interface->vxlan_port) |
 			(ETH_P_TEB << FM10K_TUNNEL_CFG_NVGRE_SHIFT));
 
-	port = list_first_entry_or_null(&interface->geneve_port,
-					struct fm10k_udp_port, list);
-
 	/* restore Geneve tunnel configuration register */
 	fm10k_write_reg(hw, FM10K_TUNNEL_CFG_GENEVE,
-			(port ? ntohs(port->port) : 0));
-}
-
-static struct fm10k_udp_port *
-fm10k_remove_tunnel_port(struct list_head *ports,
-			 struct udp_tunnel_info *ti)
-{
-	struct fm10k_udp_port *port;
-
-	list_for_each_entry(port, ports, list) {
-		if ((port->port == ti->port) &&
-		    (port->sa_family == ti->sa_family)) {
-			list_del(&port->list);
-			return port;
-		}
-	}
-
-	return NULL;
-}
-
-static void fm10k_insert_tunnel_port(struct list_head *ports,
-				     struct udp_tunnel_info *ti)
-{
-	struct fm10k_udp_port *port;
-
-	/* remove existing port entry from the list so that the newest items
-	 * are always at the tail of the list.
-	 */
-	port = fm10k_remove_tunnel_port(ports, ti);
-	if (!port) {
-		port = kmalloc(sizeof(*port), GFP_ATOMIC);
-		if  (!port)
-			return;
-		port->port = ti->port;
-		port->sa_family = ti->sa_family;
-	}
-
-	list_add_tail(&port->list, ports);
+			ntohs(interface->geneve_port));
 }
 
 /**
- * fm10k_udp_tunnel_add
- * @netdev: network interface device structure
- * @ti: Tunnel endpoint information
+ * fm10k_udp_tunnel_sync - Called when UDP tunnel ports change
+ * @dev: network interface device structure
+ * @table: Tunnel table (according to tables of @fm10k_udp_tunnels)
  *
- * This function is called when a new UDP tunnel port has been added.
+ * This function is called when a new UDP tunnel port is added or deleted.
  * Due to hardware restrictions, only one port per type can be offloaded at
- * once.
+ * once. Core will send to the driver a port of its choice.
  **/
-static void fm10k_udp_tunnel_add(struct net_device *dev,
-				 struct udp_tunnel_info *ti)
+static int fm10k_udp_tunnel_sync(struct net_device *dev, unsigned int table)
 {
 	struct fm10k_intfc *interface = netdev_priv(dev);
+	struct udp_tunnel_info ti;
 
-	/* only the PF supports configuring tunnels */
-	if (interface->hw.mac.type != fm10k_mac_pf)
-		return;
-
-	switch (ti->type) {
-	case UDP_TUNNEL_TYPE_VXLAN:
-		fm10k_insert_tunnel_port(&interface->vxlan_port, ti);
-		break;
-	case UDP_TUNNEL_TYPE_GENEVE:
-		fm10k_insert_tunnel_port(&interface->geneve_port, ti);
-		break;
-	default:
-		return;
-	}
+	udp_tunnel_nic_get_port(dev, table, 0, &ti);
+	if (!table)
+		interface->vxlan_port = ti.port;
+	else
+		interface->geneve_port = ti.port;
 
 	fm10k_restore_udp_port_info(interface);
+	return 0;
 }
 
-/**
- * fm10k_udp_tunnel_del
- * @netdev: network interface device structure
- * @ti: Tunnel endpoint information
- *
- * This function is called when a new UDP tunnel port is deleted. The freed
- * port will be removed from the list, then we reprogram the offloaded port
- * based on the head of the list.
- **/
-static void fm10k_udp_tunnel_del(struct net_device *dev,
-				 struct udp_tunnel_info *ti)
-{
-	struct fm10k_intfc *interface = netdev_priv(dev);
-	struct fm10k_udp_port *port = NULL;
-
-	if (interface->hw.mac.type != fm10k_mac_pf)
-		return;
-
-	switch (ti->type) {
-	case UDP_TUNNEL_TYPE_VXLAN:
-		port = fm10k_remove_tunnel_port(&interface->vxlan_port, ti);
-		break;
-	case UDP_TUNNEL_TYPE_GENEVE:
-		port = fm10k_remove_tunnel_port(&interface->geneve_port, ti);
-		break;
-	default:
-		return;
-	}
-
-	/* if we did remove a port we need to free its memory */
-	kfree(port);
-
-	fm10k_restore_udp_port_info(interface);
-}
+static const struct udp_tunnel_nic_info fm10k_udp_tunnels = {
+	.sync_table	= fm10k_udp_tunnel_sync,
+	.tables		= {
+		{ .n_entries = 1, .tunnel_types = UDP_TUNNEL_TYPE_VXLAN,  },
+		{ .n_entries = 1, .tunnel_types = UDP_TUNNEL_TYPE_GENEVE, },
+	},
+};
 
 /**
  * fm10k_open - Called when a network interface is made active
@@ -597,8 +468,6 @@ int fm10k_open(struct net_device *netdev)
 	if (err)
 		goto err_set_queues;
 
-	udp_tunnel_get_rx_info(netdev);
-
 	fm10k_up(interface);
 
 	return 0;
@@ -632,8 +501,6 @@ int fm10k_close(struct net_device *netdev)
 
 	fm10k_qv_free_irq(interface);
 
-	fm10k_free_udp_port_info(interface);
-
 	fm10k_free_all_tx_resources(interface);
 	fm10k_free_all_rx_resources(interface);
 
@@ -643,9 +510,13 @@ int fm10k_close(struct net_device *netdev)
 static netdev_tx_t fm10k_xmit_frame(struct sk_buff *skb, struct net_device *dev)
 {
 	struct fm10k_intfc *interface = netdev_priv(dev);
+	int num_tx_queues = READ_ONCE(interface->num_tx_queues);
 	unsigned int r_idx = skb->queue_mapping;
 	int err;
 
+	if (!num_tx_queues)
+		return NETDEV_TX_BUSY;
+
 	if ((skb->protocol == htons(ETH_P_8021Q)) &&
 	    !skb_vlan_tag_present(skb)) {
 		/* FM10K only supports hardware tagging, any tags in frame
@@ -698,8 +569,8 @@ static netdev_tx_t fm10k_xmit_frame(struct sk_buff *skb, struct net_device *dev)
 		__skb_put(skb, pad_len);
 	}
 
-	if (r_idx >= interface->num_tx_queues)
-		r_idx %= interface->num_tx_queues;
+	if (r_idx >= num_tx_queues)
+		r_idx %= num_tx_queues;
 
 	err = fm10k_xmit_frame_ring(skb, interface->tx_ring[r_idx]);
 
@@ -709,21 +580,24 @@ static netdev_tx_t fm10k_xmit_frame(struct sk_buff *skb, struct net_device *dev)
 /**
  * fm10k_tx_timeout - Respond to a Tx Hang
  * @netdev: network interface device structure
+ * @txqueue: the index of the Tx queue that timed out
  **/
-static void fm10k_tx_timeout(struct net_device *netdev)
+static void fm10k_tx_timeout(struct net_device *netdev, unsigned int txqueue)
 {
 	struct fm10k_intfc *interface = netdev_priv(netdev);
+	struct fm10k_ring *tx_ring;
 	bool real_tx_hang = false;
-	int i;
-
-#define TX_TIMEO_LIMIT 16000
-	for (i = 0; i < interface->num_tx_queues; i++) {
-		struct fm10k_ring *tx_ring = interface->tx_ring[i];
 
-		if (check_for_tx_hang(tx_ring) && fm10k_check_tx_hang(tx_ring))
-			real_tx_hang = true;
+	if (txqueue >= interface->num_tx_queues) {
+		WARN(1, "invalid Tx queue index %d", txqueue);
+		return;
 	}
 
+	tx_ring = interface->tx_ring[txqueue];
+	if (check_for_tx_hang(tx_ring) && fm10k_check_tx_hang(tx_ring))
+		real_tx_hang = true;
+
+#define TX_TIMEO_LIMIT 16000
 	if (real_tx_hang) {
 		fm10k_tx_timeout_reset(interface);
 	} else {
@@ -754,23 +628,141 @@ static bool fm10k_host_mbx_ready(struct fm10k_intfc *interface)
 	return (hw->mac.type == fm10k_mac_vf || interface->host_ready);
 }
 
+/**
+ * fm10k_queue_vlan_request - Queue a VLAN update request
+ * @interface: the fm10k interface structure
+ * @vid: the VLAN vid
+ * @vsi: VSI index number
+ * @set: whether to set or clear
+ *
+ * This function queues up a VLAN update. For VFs, this must be sent to the
+ * managing PF over the mailbox. For PFs, we'll use the same handling so that
+ * it's similar to the VF. This avoids storming the PF<->VF mailbox with too
+ * many VLAN updates during reset.
+ */
+int fm10k_queue_vlan_request(struct fm10k_intfc *interface,
+			     u32 vid, u8 vsi, bool set)
+{
+	struct fm10k_macvlan_request *request;
+	unsigned long flags;
+
+	/* This must be atomic since we may be called while the netdev
+	 * addr_list_lock is held
+	 */
+	request = kzalloc(sizeof(*request), GFP_ATOMIC);
+	if (!request)
+		return -ENOMEM;
+
+	request->type = FM10K_VLAN_REQUEST;
+	request->vlan.vid = vid;
+	request->vlan.vsi = vsi;
+	request->set = set;
+
+	spin_lock_irqsave(&interface->macvlan_lock, flags);
+	list_add_tail(&request->list, &interface->macvlan_requests);
+	spin_unlock_irqrestore(&interface->macvlan_lock, flags);
+
+	fm10k_macvlan_schedule(interface);
+
+	return 0;
+}
+
+/**
+ * fm10k_queue_mac_request - Queue a MAC update request
+ * @interface: the fm10k interface structure
+ * @glort: the target glort for this update
+ * @addr: the address to update
+ * @vid: the vid to update
+ * @set: whether to add or remove
+ *
+ * This function queues up a MAC request for sending to the switch manager.
+ * A separate thread monitors the queue and sends updates to the switch
+ * manager. Return 0 on success, and negative error code on failure.
+ **/
+int fm10k_queue_mac_request(struct fm10k_intfc *interface, u16 glort,
+			    const unsigned char *addr, u16 vid, bool set)
+{
+	struct fm10k_macvlan_request *request;
+	unsigned long flags;
+
+	/* This must be atomic since we may be called while the netdev
+	 * addr_list_lock is held
+	 */
+	request = kzalloc(sizeof(*request), GFP_ATOMIC);
+	if (!request)
+		return -ENOMEM;
+
+	if (is_multicast_ether_addr(addr))
+		request->type = FM10K_MC_MAC_REQUEST;
+	else
+		request->type = FM10K_UC_MAC_REQUEST;
+
+	ether_addr_copy(request->mac.addr, addr);
+	request->mac.glort = glort;
+	request->mac.vid = vid;
+	request->set = set;
+
+	spin_lock_irqsave(&interface->macvlan_lock, flags);
+	list_add_tail(&request->list, &interface->macvlan_requests);
+	spin_unlock_irqrestore(&interface->macvlan_lock, flags);
+
+	fm10k_macvlan_schedule(interface);
+
+	return 0;
+}
+
+/**
+ * fm10k_clear_macvlan_queue - Cancel pending updates for a given glort
+ * @interface: the fm10k interface structure
+ * @glort: the target glort to clear
+ * @vlans: true to clear VLAN messages, false to ignore them
+ *
+ * Cancel any outstanding MAC/VLAN requests for a given glort. This is
+ * expected to be called when a logical port goes down.
+ **/
+void fm10k_clear_macvlan_queue(struct fm10k_intfc *interface,
+			       u16 glort, bool vlans)
+
+{
+	struct fm10k_macvlan_request *r, *tmp;
+	unsigned long flags;
+
+	spin_lock_irqsave(&interface->macvlan_lock, flags);
+
+	/* Free any outstanding MAC/VLAN requests for this interface */
+	list_for_each_entry_safe(r, tmp, &interface->macvlan_requests, list) {
+		switch (r->type) {
+		case FM10K_MC_MAC_REQUEST:
+		case FM10K_UC_MAC_REQUEST:
+			/* Don't free requests for other interfaces */
+			if (r->mac.glort != glort)
+				break;
+			fallthrough;
+		case FM10K_VLAN_REQUEST:
+			if (vlans) {
+				list_del(&r->list);
+				kfree(r);
+			}
+			break;
+		}
+	}
+
+	spin_unlock_irqrestore(&interface->macvlan_lock, flags);
+}
+
 static int fm10k_uc_vlan_unsync(struct net_device *netdev,
 				const unsigned char *uc_addr)
 {
 	struct fm10k_intfc *interface = netdev_priv(netdev);
-	struct fm10k_hw *hw = &interface->hw;
 	u16 glort = interface->glort;
 	u16 vid = interface->vid;
 	bool set = !!(vid / VLAN_N_VID);
-	int err = -EHOSTDOWN;
+	int err;
 
 	/* drop any leading bits on the VLAN ID */
 	vid &= VLAN_N_VID - 1;
 
-	if (fm10k_host_mbx_ready(interface))
-		err = hw->mac.ops.update_uc_addr(hw, glort, uc_addr,
-						 vid, set, 0);
-
+	err = fm10k_queue_mac_request(interface, glort, uc_addr, vid, set);
 	if (err)
 		return err;
 
@@ -782,18 +774,15 @@ static int fm10k_mc_vlan_unsync(struct net_device *netdev,
 				const unsigned char *mc_addr)
 {
 	struct fm10k_intfc *interface = netdev_priv(netdev);
-	struct fm10k_hw *hw = &interface->hw;
 	u16 glort = interface->glort;
 	u16 vid = interface->vid;
 	bool set = !!(vid / VLAN_N_VID);
-	int err = -EHOSTDOWN;
+	int err;
 
 	/* drop any leading bits on the VLAN ID */
 	vid &= VLAN_N_VID - 1;
 
-	if (fm10k_host_mbx_ready(interface))
-		err = hw->mac.ops.update_mc_addr(hw, glort, mc_addr, vid, set);
-
+	err = fm10k_queue_mac_request(interface, glort, mc_addr, vid, set);
 	if (err)
 		return err;
 
@@ -804,7 +793,9 @@ static int fm10k_mc_vlan_unsync(struct net_device *netdev,
 static int fm10k_update_vid(struct net_device *netdev, u16 vid, bool set)
 {
 	struct fm10k_intfc *interface = netdev_priv(netdev);
+	struct fm10k_l2_accel *l2_accel = interface->l2_accel;
 	struct fm10k_hw *hw = &interface->hw;
+	u16 glort;
 	s32 err;
 	int i;
 
@@ -815,8 +806,12 @@ static int fm10k_update_vid(struct net_device *netdev, u16 vid, bool set)
 	if (vid >= VLAN_N_VID)
 		return -EINVAL;
 
-	/* Verify we have permission to add VLANs */
-	if (hw->mac.vlan_override)
+	/* Verify that we have permission to add VLANs. If this is a request
+	 * to remove a VLAN, we still want to allow the user to remove the
+	 * VLAN device. In that case, we need to clear the bit in the
+	 * active_vlans bitmask.
+	 */
+	if (set && hw->mac.vlan_override)
 		return -EACCES;
 
 	/* update active_vlans bitmask */
@@ -835,6 +830,12 @@ static int fm10k_update_vid(struct net_device *netdev, u16 vid, bool set)
 			rx_ring->vid &= ~FM10K_VLAN_CLEAR;
 	}
 
+	/* If our VLAN has been overridden, there is no reason to send VLAN
+	 * removal requests as they will be silently ignored.
+	 */
+	if (hw->mac.vlan_override)
+		return 0;
+
 	/* Do not remove default VLAN ID related entries from VLAN and MAC
 	 * tables
 	 */
@@ -851,21 +852,33 @@ static int fm10k_update_vid(struct net_device *netdev, u16 vid, bool set)
 
 	/* only need to update the VLAN if not in promiscuous mode */
 	if (!(netdev->flags & IFF_PROMISC)) {
-		err = hw->mac.ops.update_vlan(hw, vid, 0, set);
+		err = fm10k_queue_vlan_request(interface, vid, 0, set);
 		if (err)
 			goto err_out;
 	}
 
-	/* update our base MAC address if host's mailbox is ready */
-	if (fm10k_host_mbx_ready(interface))
-		err = hw->mac.ops.update_uc_addr(hw, interface->glort,
-						 hw->mac.addr, vid, set, 0);
-	else
-		err = -EHOSTDOWN;
-
+	/* Update our base MAC address */
+	err = fm10k_queue_mac_request(interface, interface->glort,
+				      hw->mac.addr, vid, set);
 	if (err)
 		goto err_out;
 
+	/* Update L2 accelerated macvlan addresses */
+	if (l2_accel) {
+		for (i = 0; i < l2_accel->size; i++) {
+			struct net_device *sdev = l2_accel->macvlan[i];
+
+			if (!sdev)
+				continue;
+
+			glort = l2_accel->dglort + 1 + i;
+
+			fm10k_queue_mac_request(interface, glort,
+						sdev->dev_addr,
+						vid, set);
+		}
+	}
+
 	/* set VLAN ID prior to syncing/unsyncing the VLAN */
 	interface->vid = vid + (set ? VLAN_N_VID : 0);
 
@@ -906,7 +919,6 @@ static u16 fm10k_find_next_vlan(struct fm10k_intfc *interface, u16 vid)
 
 static void fm10k_clear_unused_vlans(struct fm10k_intfc *interface)
 {
-	struct fm10k_hw *hw = &interface->hw;
 	u32 vid, prev_vid;
 
 	/* loop through and find any gaps in the table */
@@ -918,7 +930,7 @@ static void fm10k_clear_unused_vlans(struct fm10k_intfc *interface)
 
 		/* send request to clear multiple bits at a time */
 		prev_vid += (vid - prev_vid - 1) << FM10K_VLAN_LENGTH_SHIFT;
-		hw->mac.ops.update_vlan(hw, prev_vid, 0, false);
+		fm10k_queue_vlan_request(interface, prev_vid, 0, false);
 	}
 }
 
@@ -926,22 +938,17 @@ static int __fm10k_uc_sync(struct net_device *dev,
 			   const unsigned char *addr, bool sync)
 {
 	struct fm10k_intfc *interface = netdev_priv(dev);
-	struct fm10k_hw *hw = &interface->hw;
 	u16 vid, glort = interface->glort;
 	s32 err;
 
 	if (!is_valid_ether_addr(addr))
 		return -EADDRNOTAVAIL;
 
-	/* update table with current entries if host's mailbox is ready */
-	if (!fm10k_host_mbx_ready(interface))
-		return -EHOSTDOWN;
-
-	for (vid = hw->mac.default_vid ? fm10k_find_next_vlan(interface, 0) : 1;
+	for (vid = fm10k_find_next_vlan(interface, 0);
 	     vid < VLAN_N_VID;
 	     vid = fm10k_find_next_vlan(interface, vid)) {
-		err = hw->mac.ops.update_uc_addr(hw, glort, addr,
-						 vid, sync, 0);
+		err = fm10k_queue_mac_request(interface, glort,
+					      addr, vid, sync);
 		if (err)
 			return err;
 	}
@@ -983,7 +990,7 @@ static int fm10k_set_mac(struct net_device *dev, void *p)
 	}
 
 	if (!err) {
-		ether_addr_copy(dev->dev_addr, addr->sa_data);
+		eth_hw_addr_set(dev, addr->sa_data);
 		ether_addr_copy(hw->mac.addr, addr->sa_data);
 		dev->addr_assign_type &= ~NET_ADDR_RANDOM;
 	}
@@ -996,17 +1003,19 @@ static int __fm10k_mc_sync(struct net_device *dev,
 			   const unsigned char *addr, bool sync)
 {
 	struct fm10k_intfc *interface = netdev_priv(dev);
-	struct fm10k_hw *hw = &interface->hw;
 	u16 vid, glort = interface->glort;
+	s32 err;
 
-	/* update table with current entries if host's mailbox is ready */
-	if (!fm10k_host_mbx_ready(interface))
-		return 0;
+	if (!is_multicast_ether_addr(addr))
+		return -EADDRNOTAVAIL;
 
-	for (vid = hw->mac.default_vid ? fm10k_find_next_vlan(interface, 0) : 1;
+	for (vid = fm10k_find_next_vlan(interface, 0);
 	     vid < VLAN_N_VID;
 	     vid = fm10k_find_next_vlan(interface, vid)) {
-		hw->mac.ops.update_mc_addr(hw, glort, addr, vid, sync);
+		err = fm10k_queue_mac_request(interface, glort,
+					      addr, vid, sync);
+		if (err)
+			return err;
 	}
 
 	return 0;
@@ -1044,9 +1053,12 @@ static void fm10k_set_rx_mode(struct net_device *dev)
 
 	/* update xcast mode first, but only if it changed */
 	if (interface->xcast_mode != xcast_mode) {
-		/* update VLAN table */
+		/* update VLAN table when entering promiscuous mode */
 		if (xcast_mode == FM10K_XCAST_MODE_PROMISC)
-			hw->mac.ops.update_vlan(hw, FM10K_VLAN_ALL, 0, true);
+			fm10k_queue_vlan_request(interface, FM10K_VLAN_ALL,
+						 0, true);
+
+		/* clear VLAN table when exiting promiscuous mode */
 		if (interface->xcast_mode == FM10K_XCAST_MODE_PROMISC)
 			fm10k_clear_unused_vlans(interface);
 
@@ -1068,9 +1080,10 @@ static void fm10k_set_rx_mode(struct net_device *dev)
 
 void fm10k_restore_rx_state(struct fm10k_intfc *interface)
 {
+	struct fm10k_l2_accel *l2_accel = interface->l2_accel;
 	struct net_device *netdev = interface->netdev;
 	struct fm10k_hw *hw = &interface->hw;
-	int xcast_mode;
+	int xcast_mode, i;
 	u16 vid, glort;
 
 	/* record glort for this interface */
@@ -1094,22 +1107,33 @@ void fm10k_restore_rx_state(struct fm10k_intfc *interface)
 					       interface->glort_count, true);
 
 	/* update VLAN table */
-	hw->mac.ops.update_vlan(hw, FM10K_VLAN_ALL, 0,
-				xcast_mode == FM10K_XCAST_MODE_PROMISC);
-
-	/* Add filter for VLAN 0 */
-	hw->mac.ops.update_vlan(hw, 0, 0, true);
+	fm10k_queue_vlan_request(interface, FM10K_VLAN_ALL, 0,
+				 xcast_mode == FM10K_XCAST_MODE_PROMISC);
 
 	/* update table with current entries */
-	for (vid = hw->mac.default_vid ? fm10k_find_next_vlan(interface, 0) : 1;
+	for (vid = fm10k_find_next_vlan(interface, 0);
 	     vid < VLAN_N_VID;
 	     vid = fm10k_find_next_vlan(interface, vid)) {
-		hw->mac.ops.update_vlan(hw, vid, 0, true);
+		fm10k_queue_vlan_request(interface, vid, 0, true);
 
-		/* Update unicast entries if host's mailbox is ready */
-		if (fm10k_host_mbx_ready(interface))
-			hw->mac.ops.update_uc_addr(hw, glort, hw->mac.addr,
-						   vid, true, 0);
+		fm10k_queue_mac_request(interface, glort,
+					hw->mac.addr, vid, true);
+
+		/* synchronize macvlan addresses */
+		if (l2_accel) {
+			for (i = 0; i < l2_accel->size; i++) {
+				struct net_device *sdev = l2_accel->macvlan[i];
+
+				if (!sdev)
+					continue;
+
+				glort = l2_accel->dglort + 1 + i;
+
+				fm10k_queue_mac_request(interface, glort,
+							sdev->dev_addr,
+							vid, true);
+			}
+		}
 	}
 
 	/* update xcast mode before synchronizing addresses if host's mailbox
@@ -1122,6 +1146,24 @@ void fm10k_restore_rx_state(struct fm10k_intfc *interface)
 	__dev_uc_sync(netdev, fm10k_uc_sync, fm10k_uc_unsync);
 	__dev_mc_sync(netdev, fm10k_mc_sync, fm10k_mc_unsync);
 
+	/* synchronize macvlan addresses */
+	if (l2_accel) {
+		for (i = 0; i < l2_accel->size; i++) {
+			struct net_device *sdev = l2_accel->macvlan[i];
+
+			if (!sdev)
+				continue;
+
+			glort = l2_accel->dglort + 1 + i;
+
+			hw->mac.ops.update_xcast_mode(hw, glort,
+						      FM10K_XCAST_MODE_NONE);
+			fm10k_queue_mac_request(interface, glort,
+						sdev->dev_addr,
+						hw->mac.default_vid, true);
+		}
+	}
+
 	fm10k_mbx_unlock(interface);
 
 	/* record updated xcast mode state */
@@ -1136,6 +1178,13 @@ void fm10k_reset_rx_state(struct fm10k_intfc *interface)
 	struct net_device *netdev = interface->netdev;
 	struct fm10k_hw *hw = &interface->hw;
 
+	/* Wait for MAC/VLAN work to finish */
+	while (test_bit(__FM10K_MACVLAN_SCHED, interface->state))
+		usleep_range(1000, 2000);
+
+	/* Cancel pending MAC/VLAN requests */
+	fm10k_clear_macvlan_queue(interface, interface->glort, true);
+
 	fm10k_mbx_lock(interface);
 
 	/* clear the logical port state on lower device if host's mailbox is
@@ -1180,10 +1229,10 @@ static void fm10k_get_stats64(struct net_device *netdev,
 			continue;
 
 		do {
-			start = u64_stats_fetch_begin_irq(&ring->syncp);
+			start = u64_stats_fetch_begin(&ring->syncp);
 			packets = ring->stats.packets;
 			bytes   = ring->stats.bytes;
-		} while (u64_stats_fetch_retry_irq(&ring->syncp, start));
+		} while (u64_stats_fetch_retry(&ring->syncp, start));
 
 		stats->rx_packets += packets;
 		stats->rx_bytes   += bytes;
@@ -1196,10 +1245,10 @@ static void fm10k_get_stats64(struct net_device *netdev,
 			continue;
 
 		do {
-			start = u64_stats_fetch_begin_irq(&ring->syncp);
+			start = u64_stats_fetch_begin(&ring->syncp);
 			packets = ring->stats.packets;
 			bytes   = ring->stats.bytes;
-		} while (u64_stats_fetch_retry_irq(&ring->syncp, start));
+		} while (u64_stats_fetch_retry(&ring->syncp, start));
 
 		stats->tx_packets += packets;
 		stats->tx_bytes   += bytes;
@@ -1265,25 +1314,27 @@ err_queueing_scheme:
 	return err;
 }
 
-static int __fm10k_setup_tc(struct net_device *dev, u32 handle, __be16 proto,
-			    struct tc_to_netdev *tc)
+static int __fm10k_setup_tc(struct net_device *dev, enum tc_setup_type type,
+			    void *type_data)
 {
-	if (tc->type != TC_SETUP_MQPRIO)
-		return -EINVAL;
+	struct tc_mqprio_qopt *mqprio = type_data;
+
+	if (type != TC_SETUP_QDISC_MQPRIO)
+		return -EOPNOTSUPP;
 
-	tc->mqprio->hw = TC_MQPRIO_HW_OFFLOAD_TCS;
+	mqprio->hw = TC_MQPRIO_HW_OFFLOAD_TCS;
 
-	return fm10k_setup_tc(dev, tc->mqprio->num_tc);
+	return fm10k_setup_tc(dev, mqprio->num_tc);
 }
 
 static void fm10k_assign_l2_accel(struct fm10k_intfc *interface,
 				  struct fm10k_l2_accel *l2_accel)
 {
-	struct fm10k_ring *ring;
 	int i;
 
 	for (i = 0; i < interface->num_rx_queues; i++) {
-		ring = interface->rx_ring[i];
+		struct fm10k_ring *ring = interface->rx_ring[i];
+
 		rcu_assign_pointer(ring->l2_accel, l2_accel);
 	}
 
@@ -1298,8 +1349,15 @@ static void *fm10k_dfwd_add_station(struct net_device *dev,
 	struct fm10k_l2_accel *old_l2_accel = NULL;
 	struct fm10k_dglort_cfg dglort = { 0 };
 	struct fm10k_hw *hw = &interface->hw;
-	int size = 0, i;
-	u16 glort;
+	int size, i;
+	u16 vid, glort;
+
+	/* The hardware supported by fm10k only filters on the destination MAC
+	 * address. In order to avoid issues we only support offloading modes
+	 * where the hardware can actually provide the functionality.
+	 */
+	if (!macvlan_supports_dest_filter(sdev))
+		return ERR_PTR(-EMEDIUMTYPE);
 
 	/* allocate l2 accel structure if it is not available */
 	if (!l2_accel) {
@@ -1365,12 +1423,18 @@ static void *fm10k_dfwd_add_station(struct net_device *dev,
 
 	glort = l2_accel->dglort + 1 + i;
 
-	if (fm10k_host_mbx_ready(interface)) {
+	if (fm10k_host_mbx_ready(interface))
 		hw->mac.ops.update_xcast_mode(hw, glort,
-					      FM10K_XCAST_MODE_MULTI);
-		hw->mac.ops.update_uc_addr(hw, glort, sdev->dev_addr,
-					   0, true, 0);
-	}
+					      FM10K_XCAST_MODE_NONE);
+
+	fm10k_queue_mac_request(interface, glort, sdev->dev_addr,
+				hw->mac.default_vid, true);
+
+	for (vid = fm10k_find_next_vlan(interface, 0);
+	     vid < VLAN_N_VID;
+	     vid = fm10k_find_next_vlan(interface, vid))
+		fm10k_queue_mac_request(interface, glort, sdev->dev_addr,
+					vid, true);
 
 	fm10k_mbx_unlock(interface);
 
@@ -1384,8 +1448,8 @@ static void fm10k_dfwd_del_station(struct net_device *dev, void *priv)
 	struct fm10k_dglort_cfg dglort = { 0 };
 	struct fm10k_hw *hw = &interface->hw;
 	struct net_device *sdev = priv;
+	u16 vid, glort;
 	int i;
-	u16 glort;
 
 	if (!l2_accel)
 		return;
@@ -1405,12 +1469,18 @@ static void fm10k_dfwd_del_station(struct net_device *dev, void *priv)
 
 	glort = l2_accel->dglort + 1 + i;
 
-	if (fm10k_host_mbx_ready(interface)) {
+	if (fm10k_host_mbx_ready(interface))
 		hw->mac.ops.update_xcast_mode(hw, glort,
 					      FM10K_XCAST_MODE_NONE);
-		hw->mac.ops.update_uc_addr(hw, glort, sdev->dev_addr,
-					   0, false, 0);
-	}
+
+	fm10k_queue_mac_request(interface, glort, sdev->dev_addr,
+				hw->mac.default_vid, false);
+
+	for (vid = fm10k_find_next_vlan(interface, 0);
+	     vid < VLAN_N_VID;
+	     vid = fm10k_find_next_vlan(interface, vid))
+		fm10k_queue_mac_request(interface, glort, sdev->dev_addr,
+					vid, false);
 
 	fm10k_mbx_unlock(interface);
 
@@ -1460,13 +1530,9 @@ static const struct net_device_ops fm10k_netdev_ops = {
 	.ndo_set_vf_vlan	= fm10k_ndo_set_vf_vlan,
 	.ndo_set_vf_rate	= fm10k_ndo_set_vf_bw,
 	.ndo_get_vf_config	= fm10k_ndo_get_vf_config,
-	.ndo_udp_tunnel_add	= fm10k_udp_tunnel_add,
-	.ndo_udp_tunnel_del	= fm10k_udp_tunnel_del,
+	.ndo_get_vf_stats	= fm10k_ndo_get_vf_stats,
 	.ndo_dfwd_add_station	= fm10k_dfwd_add_station,
 	.ndo_dfwd_del_station	= fm10k_dfwd_del_station,
-#ifdef CONFIG_NET_POLL_CONTROLLER
-	.ndo_poll_controller	= fm10k_netpoll,
-#endif
 	.ndo_features_check	= fm10k_features_check,
 };
 
@@ -1511,6 +1577,8 @@ struct net_device *fm10k_alloc_netdev(const struct fm10k_info *info)
 				       NETIF_F_SG;
 
 		dev->features |= NETIF_F_GSO_UDP_TUNNEL;
+
+		dev->udp_tunnel_nic_info = &fm10k_udp_tunnels;
 	}
 
 	/* all features defined to this point should be changeable */
diff --git a/drivers/net/ethernet/intel/fm10k/fm10k_pci.c b/drivers/net/ethernet/intel/fm10k/fm10k_pci.c
index 3e26d27ad213..ae5fe34659cf 100644
--- a/drivers/net/ethernet/intel/fm10k/fm10k_pci.c
+++ b/drivers/net/ethernet/intel/fm10k/fm10k_pci.c
@@ -1,26 +1,8 @@
-/* Intel(R) Ethernet Switch Host Interface Driver
- * Copyright(c) 2013 - 2017 Intel Corporation.
- *
- * This program is free software; you can redistribute it and/or modify it
- * under the terms and conditions of the GNU General Public License,
- * version 2, as published by the Free Software Foundation.
- *
- * This program is distributed in the hope it will be useful, but WITHOUT
- * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
- * FITNESS FOR A PARTICULAR PURPOSE.  See the GNU General Public License for
- * more details.
- *
- * The full GNU General Public License is included in this distribution in
- * the file called "COPYING".
- *
- * Contact Information:
- * e1000-devel Mailing List <e1000-devel@lists.sourceforge.net>
- * Intel Corporation, 5200 N.E. Elam Young Parkway, Hillsboro, OR 97124-6497
- */
+// SPDX-License-Identifier: GPL-2.0
+/* Copyright(c) 2013 - 2019 Intel Corporation. */
 
 #include <linux/module.h>
 #include <linux/interrupt.h>
-#include <linux/aer.h>
 
 #include "fm10k.h"
 
@@ -29,7 +11,7 @@ static const struct fm10k_info *fm10k_info_tbl[] = {
 	[fm10k_device_vf] = &fm10k_vf_info,
 };
 
-/**
+/*
  * fm10k_pci_tbl - PCI Device ID Table
  *
  * Wildcard entries (PCI_ANY_ID) should come last
@@ -40,6 +22,8 @@ static const struct fm10k_info *fm10k_info_tbl[] = {
  */
 static const struct pci_device_id fm10k_pci_tbl[] = {
 	{ PCI_VDEVICE(INTEL, FM10K_DEV_ID_PF), fm10k_device_pf },
+	{ PCI_VDEVICE(INTEL, FM10K_DEV_ID_SDI_FM10420_QDA2), fm10k_device_pf },
+	{ PCI_VDEVICE(INTEL, FM10K_DEV_ID_SDI_FM10420_DA2), fm10k_device_pf },
 	{ PCI_VDEVICE(INTEL, FM10K_DEV_ID_VF), fm10k_device_vf },
 	/* required last entry */
 	{ 0, }
@@ -91,6 +75,76 @@ static int fm10k_hw_ready(struct fm10k_intfc *interface)
 	return FM10K_REMOVED(hw->hw_addr) ? -ENODEV : 0;
 }
 
+/**
+ * fm10k_macvlan_schedule - Schedule MAC/VLAN queue task
+ * @interface: fm10k private interface structure
+ *
+ * Schedule the MAC/VLAN queue monitor task. If the MAC/VLAN task cannot be
+ * started immediately, request that it be restarted when possible.
+ */
+void fm10k_macvlan_schedule(struct fm10k_intfc *interface)
+{
+	/* Avoid processing the MAC/VLAN queue when the service task is
+	 * disabled, or when we're resetting the device.
+	 */
+	if (!test_bit(__FM10K_MACVLAN_DISABLE, interface->state) &&
+	    !test_and_set_bit(__FM10K_MACVLAN_SCHED, interface->state)) {
+		clear_bit(__FM10K_MACVLAN_REQUEST, interface->state);
+		/* We delay the actual start of execution in order to allow
+		 * multiple MAC/VLAN updates to accumulate before handling
+		 * them, and to allow some time to let the mailbox drain
+		 * between runs.
+		 */
+		queue_delayed_work(fm10k_workqueue,
+				   &interface->macvlan_task, 10);
+	} else {
+		set_bit(__FM10K_MACVLAN_REQUEST, interface->state);
+	}
+}
+
+/**
+ * fm10k_stop_macvlan_task - Stop the MAC/VLAN queue monitor
+ * @interface: fm10k private interface structure
+ *
+ * Wait until the MAC/VLAN queue task has stopped, and cancel any future
+ * requests.
+ */
+static void fm10k_stop_macvlan_task(struct fm10k_intfc *interface)
+{
+	/* Disable the MAC/VLAN work item */
+	set_bit(__FM10K_MACVLAN_DISABLE, interface->state);
+
+	/* Make sure we waited until any current invocations have stopped */
+	cancel_delayed_work_sync(&interface->macvlan_task);
+
+	/* We set the __FM10K_MACVLAN_SCHED bit when we schedule the task.
+	 * However, it may not be unset of the MAC/VLAN task never actually
+	 * got a chance to run. Since we've canceled the task here, and it
+	 * cannot be rescheuled right now, we need to ensure the scheduled bit
+	 * gets unset.
+	 */
+	clear_bit(__FM10K_MACVLAN_SCHED, interface->state);
+}
+
+/**
+ * fm10k_resume_macvlan_task - Restart the MAC/VLAN queue monitor
+ * @interface: fm10k private interface structure
+ *
+ * Clear the __FM10K_MACVLAN_DISABLE bit and, if a request occurred, schedule
+ * the MAC/VLAN work monitor.
+ */
+static void fm10k_resume_macvlan_task(struct fm10k_intfc *interface)
+{
+	/* Re-enable the MAC/VLAN work item */
+	clear_bit(__FM10K_MACVLAN_DISABLE, interface->state);
+
+	/* We might have received a MAC/VLAN request while disabled. If so,
+	 * kick off the queue now.
+	 */
+	if (test_bit(__FM10K_MACVLAN_REQUEST, interface->state))
+		fm10k_macvlan_schedule(interface);
+}
+
 void fm10k_service_event_schedule(struct fm10k_intfc *interface)
 {
 	if (!test_bit(__FM10K_SERVICE_DISABLE, interface->state) &&
@@ -118,60 +172,66 @@ static void fm10k_service_event_complete(struct fm10k_intfc *interface)
 		fm10k_service_event_schedule(interface);
 }
 
-/**
- * fm10k_service_timer - Timer Call-back
- * @data: pointer to interface cast into an unsigned long
- **/
-static void fm10k_service_timer(unsigned long data)
+static void fm10k_stop_service_event(struct fm10k_intfc *interface)
 {
-	struct fm10k_intfc *interface = (struct fm10k_intfc *)data;
+	set_bit(__FM10K_SERVICE_DISABLE, interface->state);
+	cancel_work_sync(&interface->service_task);
 
-	/* Reset the timer */
-	mod_timer(&interface->service_timer, (HZ * 2) + jiffies);
+	/* It's possible that cancel_work_sync stopped the service task from
+	 * running before it could actually start. In this case the
+	 * __FM10K_SERVICE_SCHED bit will never be cleared. Since we know that
+	 * the service task cannot be running at this point, we need to clear
+	 * the scheduled bit, as otherwise the service task may never be
+	 * restarted.
+	 */
+	clear_bit(__FM10K_SERVICE_SCHED, interface->state);
+}
 
+static void fm10k_start_service_event(struct fm10k_intfc *interface)
+{
+	clear_bit(__FM10K_SERVICE_DISABLE, interface->state);
 	fm10k_service_event_schedule(interface);
 }
 
-static void fm10k_detach_subtask(struct fm10k_intfc *interface)
+/**
+ * fm10k_service_timer - Timer Call-back
+ * @t: pointer to timer data
+ **/
+static void fm10k_service_timer(struct timer_list *t)
 {
-	struct net_device *netdev = interface->netdev;
-	u32 __iomem *hw_addr;
-	u32 value;
-
-	/* do nothing if device is still present or hw_addr is set */
-	if (netif_device_present(netdev) || interface->hw.hw_addr)
-		return;
-
-	/* check the real address space to see if we've recovered */
-	hw_addr = READ_ONCE(interface->uc_addr);
-	value = readl(hw_addr);
-	if (~value) {
-		interface->hw.hw_addr = interface->uc_addr;
-		netif_device_attach(netdev);
-		set_bit(FM10K_FLAG_RESET_REQUESTED, interface->flags);
-		netdev_warn(netdev, "PCIe link restored, device now attached\n");
-		return;
-	}
+	struct fm10k_intfc *interface = timer_container_of(interface, t,
+							   service_timer);
 
-	rtnl_lock();
-
-	if (netif_running(netdev))
-		dev_close(netdev);
+	/* Reset the timer */
+	mod_timer(&interface->service_timer, (HZ * 2) + jiffies);
 
-	rtnl_unlock();
+	fm10k_service_event_schedule(interface);
 }
 
-static void fm10k_prepare_for_reset(struct fm10k_intfc *interface)
+/**
+ * fm10k_prepare_for_reset - Prepare the driver and device for a pending reset
+ * @interface: fm10k private data structure
+ *
+ * This function prepares for a device reset by shutting as much down as we
+ * can. It does nothing and returns false if __FM10K_RESETTING was already set
+ * prior to calling this function. It returns true if it actually did work.
+ */
+static bool fm10k_prepare_for_reset(struct fm10k_intfc *interface)
 {
 	struct net_device *netdev = interface->netdev;
 
-	WARN_ON(in_interrupt());
-
 	/* put off any impending NetWatchDogTimeout */
 	netif_trans_update(netdev);
 
-	while (test_and_set_bit(__FM10K_RESETTING, interface->state))
-		usleep_range(1000, 2000);
+	/* Nothing to do if a reset is already in progress */
+	if (test_and_set_bit(__FM10K_RESETTING, interface->state))
+		return false;
+
+	/* As the MAC/VLAN task will be accessing registers it must not be
+	 * running while we reset. Although the task will not be scheduled
+	 * once we start resetting it may already be running
+	 */
+	fm10k_stop_macvlan_task(interface);
 
 	rtnl_lock();
 
@@ -189,6 +249,8 @@ static void fm10k_prepare_for_reset(struct fm10k_intfc *interface)
 	interface->last_reset = jiffies + (10 * HZ);
 
 	rtnl_unlock();
+
+	return true;
 }
 
 static int fm10k_handle_reset(struct fm10k_intfc *interface)
@@ -197,6 +259,8 @@ static int fm10k_handle_reset(struct fm10k_intfc *interface)
 	struct fm10k_hw *hw = &interface->hw;
 	int err;
 
+	WARN_ON(!test_bit(__FM10K_RESETTING, interface->state));
+
 	rtnl_lock();
 
 	pci_set_master(interface->pdev);
@@ -235,7 +299,7 @@ static int fm10k_handle_reset(struct fm10k_intfc *interface)
 		if (is_valid_ether_addr(hw->mac.perm_addr)) {
 			ether_addr_copy(hw->mac.addr, hw->mac.perm_addr);
 			ether_addr_copy(netdev->perm_addr, hw->mac.perm_addr);
-			ether_addr_copy(netdev->dev_addr, hw->mac.perm_addr);
+			eth_hw_addr_set(netdev, hw->mac.perm_addr);
 			netdev->addr_assign_type &= ~NET_ADDR_RANDOM;
 		}
 
@@ -253,6 +317,8 @@ static int fm10k_handle_reset(struct fm10k_intfc *interface)
 
 	rtnl_unlock();
 
+	fm10k_resume_macvlan_task(interface);
+
 	clear_bit(__FM10K_RESETTING, interface->state);
 
 	return err;
@@ -270,27 +336,81 @@ reinit_err:
 	return err;
 }
 
-static void fm10k_reinit(struct fm10k_intfc *interface)
+static void fm10k_detach_subtask(struct fm10k_intfc *interface)
 {
-	int err;
+	struct net_device *netdev = interface->netdev;
+	u32 __iomem *hw_addr;
+	u32 value;
 
-	fm10k_prepare_for_reset(interface);
+	/* do nothing if netdev is still present or hw_addr is set */
+	if (netif_device_present(netdev) || interface->hw.hw_addr)
+		return;
 
-	err = fm10k_handle_reset(interface);
-	if (err)
-		dev_err(&interface->pdev->dev,
-			"fm10k_handle_reset failed: %d\n", err);
+	/* We've lost the PCIe register space, and can no longer access the
+	 * device. Shut everything except the detach subtask down and prepare
+	 * to reset the device in case we recover. If we actually prepare for
+	 * reset, indicate that we're detached.
+	 */
+	if (fm10k_prepare_for_reset(interface))
+		set_bit(__FM10K_RESET_DETACHED, interface->state);
+
+	/* check the real address space to see if we've recovered */
+	hw_addr = READ_ONCE(interface->uc_addr);
+	value = readl(hw_addr);
+	if (~value) {
+		int err;
+
+		/* Make sure the reset was initiated because we detached,
+		 * otherwise we might race with a different reset flow.
+		 */
+		if (!test_and_clear_bit(__FM10K_RESET_DETACHED,
+					interface->state))
+			return;
+
+		/* Restore the hardware address */
+		interface->hw.hw_addr = interface->uc_addr;
+
+		/* PCIe link has been restored, and the device is active
+		 * again. Restore everything and reset the device.
+		 */
+		err = fm10k_handle_reset(interface);
+		if (err) {
+			netdev_err(netdev, "Unable to reset device: %d\n", err);
+			interface->hw.hw_addr = NULL;
+			return;
+		}
+
+		/* Re-attach the netdev */
+		netif_device_attach(netdev);
+		netdev_warn(netdev, "PCIe link restored, device now attached\n");
+		return;
+	}
 }
 
 static void fm10k_reset_subtask(struct fm10k_intfc *interface)
 {
+	int err;
+
 	if (!test_and_clear_bit(FM10K_FLAG_RESET_REQUESTED,
 				interface->flags))
 		return;
 
+	/* If another thread has already prepared to reset the device, we
+	 * should not attempt to handle a reset here, since we'd race with
+	 * that thread. This may happen if we suspend the device or if the
+	 * PCIe link is lost. In this case, we'll just ignore the RESET
+	 * request, as it will (eventually) be taken care of when the thread
+	 * which actually started the reset is finished.
+	 */
+	if (!fm10k_prepare_for_reset(interface))
+		return;
+
 	netdev_err(interface->netdev, "Reset interface\n");
 
-	fm10k_reinit(interface);
+	err = fm10k_handle_reset(interface);
+	if (err)
+		dev_err(&interface->pdev->dev,
+			"fm10k_handle_reset failed: %d\n", err);
 }
 
 /**
@@ -360,6 +480,10 @@ static void fm10k_watchdog_update_host_state(struct fm10k_intfc *interface)
  **/
 static void fm10k_mbx_subtask(struct fm10k_intfc *interface)
 {
+	/* If we're resetting, bail out */
+	if (test_bit(__FM10K_RESETTING, interface->state))
+		return;
+
 	/* process upstream mailbox and update device state */
 	fm10k_watchdog_update_host_state(interface);
 
@@ -503,12 +627,15 @@ void fm10k_update_stats(struct fm10k_intfc *interface)
 	net_stats->rx_errors = rx_errors;
 	net_stats->rx_dropped = interface->stats.nodesc_drop.count;
 
+	/* Update VF statistics */
+	fm10k_iov_update_stats(interface);
+
 	clear_bit(__FM10K_UPDATING_STATS, interface->state);
 }
 
 /**
  * fm10k_watchdog_flush_tx - flush queues on host not ready
- * @interface - pointer to the device interface structure
+ * @interface: pointer to the device interface structure
  **/
 static void fm10k_watchdog_flush_tx(struct fm10k_intfc *interface)
 {
@@ -538,7 +665,7 @@ static void fm10k_watchdog_flush_tx(struct fm10k_intfc *interface)
 
 /**
  * fm10k_watchdog_subtask - check and bring link up
- * @interface - pointer to the device interface structure
+ * @interface: pointer to the device interface structure
  **/
 static void fm10k_watchdog_subtask(struct fm10k_intfc *interface)
 {
@@ -562,7 +689,7 @@ static void fm10k_watchdog_subtask(struct fm10k_intfc *interface)
 
 /**
  * fm10k_check_hang_subtask - check for hung queues and dropped interrupts
- * @interface - pointer to the device interface structure
+ * @interface: pointer to the device interface structure
  *
  * This function serves two purposes.  First it strobes the interrupt lines
  * in order to make certain interrupts are occurring.  Secondly it sets the
@@ -571,8 +698,6 @@ static void fm10k_watchdog_subtask(struct fm10k_intfc *interface)
  */
 static void fm10k_check_hang_subtask(struct fm10k_intfc *interface)
 {
-	int i;
-
 	/* If we're down or resetting, just bail */
 	if (test_bit(__FM10K_DOWN, interface->state) ||
 	    test_bit(__FM10K_RESETTING, interface->state))
@@ -584,6 +709,8 @@ static void fm10k_check_hang_subtask(struct fm10k_intfc *interface)
 	interface->next_tx_hang_check = jiffies + (2 * HZ);
 
 	if (netif_carrier_ok(interface->netdev)) {
+		int i;
+
 		/* Force detection of hung controller */
 		for (i = 0; i < interface->num_tx_queues; i++)
 			set_check_for_tx_hang(interface->tx_ring[i]);
@@ -609,9 +736,11 @@ static void fm10k_service_task(struct work_struct *work)
 
 	interface = container_of(work, struct fm10k_intfc, service_task);
 
+	/* Check whether we're detached first */
+	fm10k_detach_subtask(interface);
+
 	/* tasks run even when interface is down */
 	fm10k_mbx_subtask(interface);
-	fm10k_detach_subtask(interface);
 	fm10k_reset_subtask(interface);
 
 	/* tasks only run when interface is up */
@@ -623,6 +752,112 @@ static void fm10k_service_task(struct work_struct *work)
 }
 
 /**
+ * fm10k_macvlan_task - send queued MAC/VLAN requests to switch manager
+ * @work: pointer to work_struct containing our data
+ *
+ * This work item handles sending MAC/VLAN updates to the switch manager. When
+ * the interface is up, it will attempt to queue mailbox messages to the
+ * switch manager requesting updates for MAC/VLAN pairs. If the Tx fifo of the
+ * mailbox is full, it will reschedule itself to try again in a short while.
+ * This ensures that the driver does not overload the switch mailbox with too
+ * many simultaneous requests, causing an unnecessary reset.
+ **/
+static void fm10k_macvlan_task(struct work_struct *work)
+{
+	struct fm10k_macvlan_request *item;
+	struct fm10k_intfc *interface;
+	struct delayed_work *dwork;
+	struct list_head *requests;
+	struct fm10k_hw *hw;
+	unsigned long flags;
+
+	dwork = to_delayed_work(work);
+	interface = container_of(dwork, struct fm10k_intfc, macvlan_task);
+	hw = &interface->hw;
+	requests = &interface->macvlan_requests;
+
+	do {
+		/* Pop the first item off the list */
+		spin_lock_irqsave(&interface->macvlan_lock, flags);
+		item = list_first_entry_or_null(requests,
+						struct fm10k_macvlan_request,
+						list);
+		if (item)
+			list_del_init(&item->list);
+
+		spin_unlock_irqrestore(&interface->macvlan_lock, flags);
+
+		/* We have no more items to process */
+		if (!item)
+			goto done;
+
+		fm10k_mbx_lock(interface);
+
+		/* Check that we have plenty of space to send the message. We
+		 * want to ensure that the mailbox stays low enough to avoid a
+		 * change in the host state, otherwise we may see spurious
+		 * link up / link down notifications.
+		 */
+		if (!hw->mbx.ops.tx_ready(&hw->mbx, FM10K_VFMBX_MSG_MTU + 5)) {
+			hw->mbx.ops.process(hw, &hw->mbx);
+			set_bit(__FM10K_MACVLAN_REQUEST, interface->state);
+			fm10k_mbx_unlock(interface);
+
+			/* Put the request back on the list */
+			spin_lock_irqsave(&interface->macvlan_lock, flags);
+			list_add(&item->list, requests);
+			spin_unlock_irqrestore(&interface->macvlan_lock, flags);
+			break;
+		}
+
+		switch (item->type) {
+		case FM10K_MC_MAC_REQUEST:
+			hw->mac.ops.update_mc_addr(hw,
+						   item->mac.glort,
+						   item->mac.addr,
+						   item->mac.vid,
+						   item->set);
+			break;
+		case FM10K_UC_MAC_REQUEST:
+			hw->mac.ops.update_uc_addr(hw,
+						   item->mac.glort,
+						   item->mac.addr,
+						   item->mac.vid,
+						   item->set,
+						   0);
+			break;
+		case FM10K_VLAN_REQUEST:
+			hw->mac.ops.update_vlan(hw,
+						item->vlan.vid,
+						item->vlan.vsi,
+						item->set);
+			break;
+		default:
+			break;
+		}
+
+		fm10k_mbx_unlock(interface);
+
+		/* Free the item now that we've sent the update */
+		kfree(item);
+	} while (true);
+
+done:
+	WARN_ON(!test_bit(__FM10K_MACVLAN_SCHED, interface->state));
+
+	/* flush memory to make sure state is correct */
+	smp_mb__before_atomic();
+	clear_bit(__FM10K_MACVLAN_SCHED, interface->state);
+
+	/* If a MAC/VLAN request was scheduled since we started, we should
+	 * re-schedule. However, there is no reason to re-schedule if there is
+	 * no work to do.
+	 */
+	if (test_bit(__FM10K_MACVLAN_REQUEST, interface->state))
+		fm10k_macvlan_schedule(interface);
+}
+
+/**
  * fm10k_configure_tx_ring - Configure Tx ring after Reset
  * @interface: board private structure
  * @ring: structure containing ring specific data
@@ -663,7 +898,7 @@ static void fm10k_configure_tx_ring(struct fm10k_intfc *interface,
 
 	/* Map interrupt */
 	if (ring->q_vector) {
-		txint = ring->q_vector->v_idx + NON_Q_VECTORS(hw);
+		txint = ring->q_vector->v_idx + NON_Q_VECTORS;
 		txint |= FM10K_INT_MAP_TIMER0;
 	}
 
@@ -802,7 +1037,7 @@ static void fm10k_configure_rx_ring(struct fm10k_intfc *interface,
 
 	/* Map interrupt */
 	if (ring->q_vector) {
-		rxint = ring->q_vector->v_idx + NON_Q_VECTORS(hw);
+		rxint = ring->q_vector->v_idx + NON_Q_VECTORS;
 		rxint |= FM10K_INT_MAP_TIMER1;
 	}
 
@@ -978,28 +1213,6 @@ static irqreturn_t fm10k_msix_mbx_vf(int __always_unused irq, void *data)
 	return IRQ_HANDLED;
 }
 
-#ifdef CONFIG_NET_POLL_CONTROLLER
-/**
- *  fm10k_netpoll - A Polling 'interrupt' handler
- *  @netdev: network interface device structure
- *
- *  This is used by netconsole to send skbs without having to re-enable
- *  interrupts. It's not called while the normal interrupt routine is executing.
- **/
-void fm10k_netpoll(struct net_device *netdev)
-{
-	struct fm10k_intfc *interface = netdev_priv(netdev);
-	int i;
-
-	/* if interface is down do nothing */
-	if (test_bit(__FM10K_DOWN, interface->state))
-		return;
-
-	for (i = 0; i < interface->num_q_vectors; i++)
-		fm10k_msix_clean_rings(0, interface->q_vector[i]);
-}
-
-#endif
 #define FM10K_ERR_MSG(type) case (type): error = #type; break
 static void fm10k_handle_fault(struct fm10k_intfc *interface, int type,
 			       struct fm10k_fault *fault)
@@ -1156,7 +1369,6 @@ static irqreturn_t fm10k_msix_mbx_pf(int __always_unused irq, void *data)
 	struct fm10k_hw *hw = &interface->hw;
 	struct fm10k_mbx_info *mbx = &hw->mbx;
 	u32 eicr;
-	s32 err = 0;
 
 	/* unmask any set bits related to this interrupt */
 	eicr = fm10k_read_reg(hw, FM10K_EICR);
@@ -1172,15 +1384,16 @@ static irqreturn_t fm10k_msix_mbx_pf(int __always_unused irq, void *data)
 
 	/* service mailboxes */
 	if (fm10k_mbx_trylock(interface)) {
-		err = mbx->ops.process(hw, mbx);
+		s32 err = mbx->ops.process(hw, mbx);
+
+		if (err == FM10K_ERR_RESET_REQUESTED)
+			set_bit(FM10K_FLAG_RESET_REQUESTED, interface->flags);
+
 		/* handle VFLRE events */
 		fm10k_iov_event(interface);
 		fm10k_mbx_unlock(interface);
 	}
 
-	if (err == FM10K_ERR_RESET_REQUESTED)
-		set_bit(FM10K_FLAG_RESET_REQUESTED, interface->flags);
-
 	/* if switch toggled state we should reset GLORTs */
 	if (eicr & FM10K_EICR_SWITCHNOTREADY) {
 		/* force link down for at least 4 seconds */
@@ -1507,10 +1720,9 @@ int fm10k_mbx_request_irq(struct fm10k_intfc *interface)
 void fm10k_qv_free_irq(struct fm10k_intfc *interface)
 {
 	int vector = interface->num_q_vectors;
-	struct fm10k_hw *hw = &interface->hw;
 	struct msix_entry *entry;
 
-	entry = &interface->msix_entries[NON_Q_VECTORS(hw) + vector];
+	entry = &interface->msix_entries[NON_Q_VECTORS + vector];
 
 	while (vector) {
 		struct fm10k_q_vector *q_vector;
@@ -1544,25 +1756,25 @@ int fm10k_qv_request_irq(struct fm10k_intfc *interface)
 	struct net_device *dev = interface->netdev;
 	struct fm10k_hw *hw = &interface->hw;
 	struct msix_entry *entry;
-	int ri = 0, ti = 0;
+	unsigned int ri = 0, ti = 0;
 	int vector, err;
 
-	entry = &interface->msix_entries[NON_Q_VECTORS(hw)];
+	entry = &interface->msix_entries[NON_Q_VECTORS];
 
 	for (vector = 0; vector < interface->num_q_vectors; vector++) {
 		struct fm10k_q_vector *q_vector = interface->q_vector[vector];
 
 		/* name the vector */
 		if (q_vector->tx.count && q_vector->rx.count) {
-			snprintf(q_vector->name, sizeof(q_vector->name) - 1,
-				 "%s-TxRx-%d", dev->name, ri++);
+			snprintf(q_vector->name, sizeof(q_vector->name),
+				 "%s-TxRx-%u", dev->name, ri++);
 			ti++;
 		} else if (q_vector->rx.count) {
-			snprintf(q_vector->name, sizeof(q_vector->name) - 1,
-				 "%s-rx-%d", dev->name, ri++);
+			snprintf(q_vector->name, sizeof(q_vector->name),
+				 "%s-rx-%u", dev->name, ri++);
 		} else if (q_vector->tx.count) {
-			snprintf(q_vector->name, sizeof(q_vector->name) - 1,
-				 "%s-tx-%d", dev->name, ti++);
+			snprintf(q_vector->name, sizeof(q_vector->name),
+				 "%s-tx-%u", dev->name, ti++);
 		} else {
 			/* skip this unused q_vector */
 			continue;
@@ -1746,6 +1958,7 @@ skip_tx_dma_drain:
 /**
  * fm10k_sw_init - Initialize general software structures
  * @interface: host interface private structure to initialize
+ * @ent: PCI device ID entry
  *
  * fm10k_sw_init initializes the interface private data structure.
  * Fields are initialized based on PCI device information and
@@ -1800,9 +2013,6 @@ static int fm10k_sw_init(struct fm10k_intfc *interface,
 		netdev->vlan_features |= NETIF_F_HIGHDMA;
 	}
 
-	/* delay any future reset requests */
-	interface->last_reset = jiffies + (10 * HZ);
-
 	/* reset and initialize the hardware so it is in a known state */
 	err = hw->mac.ops.reset_hw(hw);
 	if (err) {
@@ -1834,7 +2044,7 @@ static int fm10k_sw_init(struct fm10k_intfc *interface,
 		netdev->addr_assign_type |= NET_ADDR_RANDOM;
 	}
 
-	ether_addr_copy(netdev->dev_addr, hw->mac.addr);
+	eth_hw_addr_set(netdev, hw->mac.addr);
 	ether_addr_copy(netdev->perm_addr, hw->mac.addr);
 
 	if (!is_valid_ether_addr(netdev->perm_addr)) {
@@ -1853,13 +2063,16 @@ static int fm10k_sw_init(struct fm10k_intfc *interface,
 	interface->tx_itr = FM10K_TX_ITR_DEFAULT;
 	interface->rx_itr = FM10K_ITR_ADAPTIVE | FM10K_RX_ITR_DEFAULT;
 
-	/* initialize udp port lists */
-	INIT_LIST_HEAD(&interface->vxlan_port);
-	INIT_LIST_HEAD(&interface->geneve_port);
+	/* Initialize the MAC/VLAN queue */
+	INIT_LIST_HEAD(&interface->macvlan_requests);
 
 	netdev_rss_key_fill(rss_key, sizeof(rss_key));
 	memcpy(interface->rssrk, rss_key, sizeof(rss_key));
 
+	/* Initialize the mailbox lock */
+	spin_lock_init(&interface->mbx_lock);
+	spin_lock_init(&interface->macvlan_lock);
+
 	/* Start off interface as being down */
 	set_bit(__FM10K_DOWN, interface->state);
 	set_bit(__FM10K_UPDATING_STATS, interface->state);
@@ -1867,91 +2080,6 @@ static int fm10k_sw_init(struct fm10k_intfc *interface,
 	return 0;
 }
 
-static void fm10k_slot_warn(struct fm10k_intfc *interface)
-{
-	enum pcie_link_width width = PCIE_LNK_WIDTH_UNKNOWN;
-	enum pci_bus_speed speed = PCI_SPEED_UNKNOWN;
-	struct fm10k_hw *hw = &interface->hw;
-	int max_gts = 0, expected_gts = 0;
-
-	if (pcie_get_minimum_link(interface->pdev, &speed, &width) ||
-	    speed == PCI_SPEED_UNKNOWN || width == PCIE_LNK_WIDTH_UNKNOWN) {
-		dev_warn(&interface->pdev->dev,
-			 "Unable to determine PCI Express bandwidth.\n");
-		return;
-	}
-
-	switch (speed) {
-	case PCIE_SPEED_2_5GT:
-		/* 8b/10b encoding reduces max throughput by 20% */
-		max_gts = 2 * width;
-		break;
-	case PCIE_SPEED_5_0GT:
-		/* 8b/10b encoding reduces max throughput by 20% */
-		max_gts = 4 * width;
-		break;
-	case PCIE_SPEED_8_0GT:
-		/* 128b/130b encoding has less than 2% impact on throughput */
-		max_gts = 8 * width;
-		break;
-	default:
-		dev_warn(&interface->pdev->dev,
-			 "Unable to determine PCI Express bandwidth.\n");
-		return;
-	}
-
-	dev_info(&interface->pdev->dev,
-		 "PCI Express bandwidth of %dGT/s available\n",
-		 max_gts);
-	dev_info(&interface->pdev->dev,
-		 "(Speed:%s, Width: x%d, Encoding Loss:%s, Payload:%s)\n",
-		 (speed == PCIE_SPEED_8_0GT ? "8.0GT/s" :
-		  speed == PCIE_SPEED_5_0GT ? "5.0GT/s" :
-		  speed == PCIE_SPEED_2_5GT ? "2.5GT/s" :
-		  "Unknown"),
-		 hw->bus.width,
-		 (speed == PCIE_SPEED_2_5GT ? "20%" :
-		  speed == PCIE_SPEED_5_0GT ? "20%" :
-		  speed == PCIE_SPEED_8_0GT ? "<2%" :
-		  "Unknown"),
-		 (hw->bus.payload == fm10k_bus_payload_128 ? "128B" :
-		  hw->bus.payload == fm10k_bus_payload_256 ? "256B" :
-		  hw->bus.payload == fm10k_bus_payload_512 ? "512B" :
-		  "Unknown"));
-
-	switch (hw->bus_caps.speed) {
-	case fm10k_bus_speed_2500:
-		/* 8b/10b encoding reduces max throughput by 20% */
-		expected_gts = 2 * hw->bus_caps.width;
-		break;
-	case fm10k_bus_speed_5000:
-		/* 8b/10b encoding reduces max throughput by 20% */
-		expected_gts = 4 * hw->bus_caps.width;
-		break;
-	case fm10k_bus_speed_8000:
-		/* 128b/130b encoding has less than 2% impact on throughput */
-		expected_gts = 8 * hw->bus_caps.width;
-		break;
-	default:
-		dev_warn(&interface->pdev->dev,
-			 "Unable to determine expected PCI Express bandwidth.\n");
-		return;
-	}
-
-	if (max_gts >= expected_gts)
-		return;
-
-	dev_warn(&interface->pdev->dev,
-		 "This device requires %dGT/s of bandwidth for optimal performance.\n",
-		 expected_gts);
-	dev_warn(&interface->pdev->dev,
-		 "A %sslot with x%d lanes is suggested.\n",
-		 (hw->bus_caps.speed == fm10k_bus_speed_2500 ? "2.5GT/s " :
-		  hw->bus_caps.speed == fm10k_bus_speed_5000 ? "5.0GT/s " :
-		  hw->bus_caps.speed == fm10k_bus_speed_8000 ? "8.0GT/s " : ""),
-		 hw->bus_caps.width);
-}
-
 /**
  * fm10k_probe - Device Initialization Routine
  * @pdev: PCI device information struct
@@ -1998,8 +2126,6 @@ static int fm10k_probe(struct pci_dev *pdev, const struct pci_device_id *ent)
 		goto err_pci_reg;
 	}
 
-	pci_enable_pcie_error_reporting(pdev);
-
 	pci_set_master(pdev);
 	pci_save_state(pdev);
 
@@ -2063,15 +2189,17 @@ static int fm10k_probe(struct pci_dev *pdev, const struct pci_device_id *ent)
 	/* Initialize service timer and service task late in order to avoid
 	 * cleanup issues.
 	 */
-	setup_timer(&interface->service_timer, &fm10k_service_timer,
-		    (unsigned long)interface);
+	timer_setup(&interface->service_timer, fm10k_service_timer, 0);
 	INIT_WORK(&interface->service_task, fm10k_service_task);
 
+	/* Setup the MAC/VLAN queue */
+	INIT_DELAYED_WORK(&interface->macvlan_task, fm10k_macvlan_task);
+
 	/* kick off service timer now, even when interface is down */
 	mod_timer(&interface->service_timer, (HZ * 2) + jiffies);
 
 	/* print warning for non-optimal configurations */
-	fm10k_slot_warn(interface);
+	pcie_print_link_status(interface->pdev);
 
 	/* report MAC address for logging */
 	dev_info(&pdev->dev, "%pM\n", netdev->dev_addr);
@@ -2079,8 +2207,9 @@ static int fm10k_probe(struct pci_dev *pdev, const struct pci_device_id *ent)
 	/* enable SR-IOV after registering netdev to enforce PF/VF ordering */
 	fm10k_iov_configure(pdev, 0);
 
-	/* clear the service task disable bit to allow service task to start */
+	/* clear the service task disable bit and kick off service task */
 	clear_bit(__FM10K_SERVICE_DISABLE, interface->state);
+	fm10k_service_event_schedule(interface);
 
 	return 0;
 
@@ -2116,10 +2245,13 @@ static void fm10k_remove(struct pci_dev *pdev)
 	struct fm10k_intfc *interface = pci_get_drvdata(pdev);
 	struct net_device *netdev = interface->netdev;
 
-	del_timer_sync(&interface->service_timer);
+	timer_delete_sync(&interface->service_timer);
 
-	set_bit(__FM10K_SERVICE_DISABLE, interface->state);
-	cancel_work_sync(&interface->service_task);
+	fm10k_stop_service_event(interface);
+	fm10k_stop_macvlan_task(interface);
+
+	/* Remove all pending MAC/VLAN requests */
+	fm10k_clear_macvlan_queue(interface, interface->glort, true);
 
 	/* free netdev, this may bounce the interrupts due to setup_tc */
 	if (netdev->reg_state == NETREG_REGISTERED)
@@ -2145,8 +2277,6 @@ static void fm10k_remove(struct pci_dev *pdev)
 
 	pci_release_mem_regions(pdev);
 
-	pci_disable_pcie_error_reporting(pdev);
-
 	pci_disable_device(pdev);
 }
 
@@ -2156,11 +2286,14 @@ static void fm10k_prepare_suspend(struct fm10k_intfc *interface)
 	 * a surprise remove if the PCIe device is disabled while we're
 	 * stopped. We stop the watchdog task until after we resume software
 	 * activity.
+	 *
+	 * Note that the MAC/VLAN task will be stopped as part of preparing
+	 * for reset so we don't need to handle it here.
 	 */
-	set_bit(__FM10K_SERVICE_DISABLE, interface->state);
-	cancel_work_sync(&interface->service_task);
+	fm10k_stop_service_event(interface);
 
-	fm10k_prepare_for_reset(interface);
+	if (fm10k_prepare_for_reset(interface))
+		set_bit(__FM10K_RESET_SUSPENDED, interface->state);
 }
 
 static int fm10k_handle_resume(struct fm10k_intfc *interface)
@@ -2168,6 +2301,13 @@ static int fm10k_handle_resume(struct fm10k_intfc *interface)
 	struct fm10k_hw *hw = &interface->hw;
 	int err;
 
+	/* Even if we didn't properly prepare for reset in
+	 * fm10k_prepare_suspend, we'll attempt to resume anyways.
+	 */
+	if (!test_and_clear_bit(__FM10K_RESET_SUSPENDED, interface->state))
+		dev_warn(&interface->pdev->dev,
+			 "Device was shut down as part of suspend... Attempting to recover\n");
+
 	/* reset statistics starting values */
 	hw->mac.ops.rebind_hw_stats(hw, &interface->stats);
 
@@ -2185,45 +2325,29 @@ static int fm10k_handle_resume(struct fm10k_intfc *interface)
 	interface->link_down_event = jiffies + (HZ);
 	set_bit(__FM10K_LINK_DOWN, interface->state);
 
-	/* clear the service task disable bit to allow service task to start */
-	clear_bit(__FM10K_SERVICE_DISABLE, interface->state);
-	fm10k_service_event_schedule(interface);
+	/* restart the service task */
+	fm10k_start_service_event(interface);
 
-	return err;
+	/* Restart the MAC/VLAN request queue in-case of outstanding events */
+	fm10k_macvlan_schedule(interface);
+
+	return 0;
 }
 
-#ifdef CONFIG_PM
 /**
- * fm10k_resume - Restore device to pre-sleep state
- * @pdev: PCI device information struct
+ * fm10k_resume - Generic PM resume hook
+ * @dev: generic device structure
  *
- * fm10k_resume is called after the system has powered back up from a sleep
- * state and is ready to resume operation.  This function is meant to restore
- * the device back to its pre-sleep state.
+ * Generic PM hook used when waking the device from a low power state after
+ * suspend or hibernation. This function does not need to handle lower PCIe
+ * device state as the stack takes care of that for us.
  **/
-static int fm10k_resume(struct pci_dev *pdev)
+static int fm10k_resume(struct device *dev)
 {
-	struct fm10k_intfc *interface = pci_get_drvdata(pdev);
+	struct fm10k_intfc *interface = dev_get_drvdata(dev);
 	struct net_device *netdev = interface->netdev;
 	struct fm10k_hw *hw = &interface->hw;
-	u32 err;
-
-	pci_set_power_state(pdev, PCI_D0);
-	pci_restore_state(pdev);
-
-	/* pci_restore_state clears dev->state_saved so call
-	 * pci_save_state to restore it.
-	 */
-	pci_save_state(pdev);
-
-	err = pci_enable_device_mem(pdev);
-	if (err) {
-		dev_err(&pdev->dev, "Cannot enable PCI device from suspend\n");
-		return err;
-	}
-	pci_set_master(pdev);
-
-	pci_wake_from_d3(pdev, false);
+	int err;
 
 	/* refresh hw_addr in case it was dropped */
 	hw->hw_addr = interface->uc_addr;
@@ -2238,36 +2362,25 @@ static int fm10k_resume(struct pci_dev *pdev)
 }
 
 /**
- * fm10k_suspend - Prepare the device for a system sleep state
- * @pdev: PCI device information struct
+ * fm10k_suspend - Generic PM suspend hook
+ * @dev: generic device structure
  *
- * fm10k_suspend is meant to shutdown the device prior to the system entering
- * a sleep state.  The fm10k hardware does not support wake on lan so the
- * driver simply needs to shut down the device so it is in a low power state.
+ * Generic PM hook used when setting the device into a low power state for
+ * system suspend or hibernation. This function does not need to handle lower
+ * PCIe device state as the stack takes care of that for us.
  **/
-static int fm10k_suspend(struct pci_dev *pdev,
-			 pm_message_t __always_unused state)
+static int fm10k_suspend(struct device *dev)
 {
-	struct fm10k_intfc *interface = pci_get_drvdata(pdev);
+	struct fm10k_intfc *interface = dev_get_drvdata(dev);
 	struct net_device *netdev = interface->netdev;
-	int err = 0;
 
 	netif_device_detach(netdev);
 
 	fm10k_prepare_suspend(interface);
 
-	err = pci_save_state(pdev);
-	if (err)
-		return err;
-
-	pci_disable_device(pdev);
-	pci_wake_from_d3(pdev, false);
-	pci_set_power_state(pdev, PCI_D3hot);
-
 	return 0;
 }
 
-#endif /* CONFIG_PM */
 /**
  * fm10k_io_error_detected - called when PCI error is detected
  * @pdev: Pointer to PCI device
@@ -2321,8 +2434,6 @@ static pci_ers_result_t fm10k_io_slot_reset(struct pci_dev *pdev)
 		result = PCI_ERS_RESULT_RECOVERED;
 	}
 
-	pci_cleanup_aer_uncorrect_error_status(pdev);
-
 	return result;
 }
 
@@ -2343,39 +2454,42 @@ static void fm10k_io_resume(struct pci_dev *pdev)
 
 	if (err)
 		dev_warn(&pdev->dev,
-			 "fm10k_io_resume failed: %d\n", err);
+			 "%s failed: %d\n", __func__, err);
 	else
 		netif_device_attach(netdev);
 }
 
 /**
- * fm10k_io_reset_notify - called when PCI function is reset
+ * fm10k_io_reset_prepare - called when PCI function is about to be reset
  * @pdev: Pointer to PCI device
  *
- * This callback is called when the PCI function is reset such as from
- * /sys/class/net/<enpX>/device/reset or similar. When prepare is true, it
- * means we should prepare for a function reset. If prepare is false, it means
- * the function reset just occurred.
+ * This callback is called when the PCI function is about to be reset,
+ * allowing the device driver to prepare for it.
  */
-static void fm10k_io_reset_notify(struct pci_dev *pdev, bool prepare)
+static void fm10k_io_reset_prepare(struct pci_dev *pdev)
 {
-	struct fm10k_intfc *interface = pci_get_drvdata(pdev);
-	int err = 0;
-
-	if (prepare) {
-		/* warn incase we have any active VF devices */
-		if (pci_num_vf(pdev))
-			dev_warn(&pdev->dev,
-				 "PCIe FLR may cause issues for any active VF devices\n");
+	/* warn incase we have any active VF devices */
+	if (pci_num_vf(pdev))
+		dev_warn(&pdev->dev,
+			 "PCIe FLR may cause issues for any active VF devices\n");
+	fm10k_prepare_suspend(pci_get_drvdata(pdev));
+}
 
-		fm10k_prepare_suspend(interface);
-	} else {
-		err = fm10k_handle_resume(interface);
-	}
+/**
+ * fm10k_io_reset_done - called when PCI function has finished resetting
+ * @pdev: Pointer to PCI device
+ *
+ * This callback is called just after the PCI function is reset, such as via
+ * /sys/class/net/<enpX>/device/reset or similar.
+ */
+static void fm10k_io_reset_done(struct pci_dev *pdev)
+{
+	struct fm10k_intfc *interface = pci_get_drvdata(pdev);
+	int err = fm10k_handle_resume(interface);
 
 	if (err) {
 		dev_warn(&pdev->dev,
-			 "fm10k_io_reset_notify failed: %d\n", err);
+			 "%s failed: %d\n", __func__, err);
 		netif_device_detach(interface->netdev);
 	}
 }
@@ -2384,18 +2498,18 @@ static const struct pci_error_handlers fm10k_err_handler = {
 	.error_detected = fm10k_io_error_detected,
 	.slot_reset = fm10k_io_slot_reset,
 	.resume = fm10k_io_resume,
-	.reset_notify = fm10k_io_reset_notify,
+	.reset_prepare = fm10k_io_reset_prepare,
+	.reset_done = fm10k_io_reset_done,
 };
 
+static DEFINE_SIMPLE_DEV_PM_OPS(fm10k_pm_ops, fm10k_suspend, fm10k_resume);
+
 static struct pci_driver fm10k_driver = {
 	.name			= fm10k_driver_name,
 	.id_table		= fm10k_pci_tbl,
 	.probe			= fm10k_probe,
 	.remove			= fm10k_remove,
-#ifdef CONFIG_PM
-	.suspend		= fm10k_suspend,
-	.resume			= fm10k_resume,
-#endif
+	.driver.pm		= pm_sleep_ptr(&fm10k_pm_ops),
 	.sriov_configure	= fm10k_iov_configure,
 	.err_handler		= &fm10k_err_handler
 };
diff --git a/drivers/net/ethernet/intel/fm10k/fm10k_pf.c b/drivers/net/ethernet/intel/fm10k/fm10k_pf.c
index 40ee0242a80a..3394645a18fe 100644
--- a/drivers/net/ethernet/intel/fm10k/fm10k_pf.c
+++ b/drivers/net/ethernet/intel/fm10k/fm10k_pf.c
@@ -1,23 +1,7 @@
-/* Intel(R) Ethernet Switch Host Interface Driver
- * Copyright(c) 2013 - 2016 Intel Corporation.
- *
- * This program is free software; you can redistribute it and/or modify it
- * under the terms and conditions of the GNU General Public License,
- * version 2, as published by the Free Software Foundation.
- *
- * This program is distributed in the hope it will be useful, but WITHOUT
- * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
- * FITNESS FOR A PARTICULAR PURPOSE.  See the GNU General Public License for
- * more details.
- *
- * The full GNU General Public License is included in this distribution in
- * the file called "COPYING".
- *
- * Contact Information:
- * e1000-devel Mailing List <e1000-devel@lists.sourceforge.net>
- * Intel Corporation, 5200 N.E. Elam Young Parkway, Hillsboro, OR 97124-6497
- */
+// SPDX-License-Identifier: GPL-2.0
+/* Copyright(c) 2013 - 2019 Intel Corporation. */
 
+#include <linux/bitfield.h>
 #include "fm10k_pf.h"
 #include "fm10k_vf.h"
 
@@ -866,7 +850,7 @@ static s32 fm10k_iov_assign_default_mac_vlan_pf(struct fm10k_hw *hw,
 	/* Determine correct default VLAN ID. The FM10K_VLAN_OVERRIDE bit is
 	 * used here to indicate to the VF that it will not have privilege to
 	 * write VLAN_TABLE. All policy is enforced on the PF but this allows
-	 * the VF to correctly report errors to userspace rqeuests.
+	 * the VF to correctly report errors to userspace requests.
 	 */
 	if (vf_info->pf_vid)
 		vf_vid = vf_info->pf_vid | FM10K_VLAN_OVERRIDE;
@@ -882,8 +866,7 @@ static s32 fm10k_iov_assign_default_mac_vlan_pf(struct fm10k_hw *hw,
 	 * register is RO from the VF, so the PF must do this even in the
 	 * case of notifying the VF of a new VID via the mailbox.
 	 */
-	txqctl = ((u32)vf_vid << FM10K_TXQCTL_VID_SHIFT) &
-		 FM10K_TXQCTL_VID_MASK;
+	txqctl = FIELD_PREP(FM10K_TXQCTL_VID_MASK, vf_vid);
 	txqctl |= (vf_idx << FM10K_TXQCTL_TC_SHIFT) |
 		  FM10K_TXQCTL_VF | vf_idx;
 
@@ -1165,11 +1148,11 @@ static void fm10k_iov_update_stats_pf(struct fm10k_hw *hw,
  *  @results: Pointer array to message, results[0] is pointer to message
  *  @mbx: Pointer to mailbox information structure
  *
- *  This function is a default handler for MSI-X requests from the VF.  The
+ *  This function is a default handler for MSI-X requests from the VF. The
  *  assumption is that in this case it is acceptable to just directly
  *  hand off the message from the VF to the underlying shared code.
  **/
-s32 fm10k_iov_msg_msix_pf(struct fm10k_hw *hw, u32 **results,
+s32 fm10k_iov_msg_msix_pf(struct fm10k_hw *hw, u32 __always_unused **results,
 			  struct fm10k_mbx_info *mbx)
 {
 	struct fm10k_vf_info *vf_info = (struct fm10k_vf_info *)mbx;
@@ -1180,13 +1163,13 @@ s32 fm10k_iov_msg_msix_pf(struct fm10k_hw *hw, u32 **results,
 
 /**
  * fm10k_iov_select_vid - Select correct default VLAN ID
- * @hw: Pointer to hardware structure
+ * @vf_info: pointer to VF information structure
  * @vid: VLAN ID to correct
  *
  * Will report an error if the VLAN ID is out of range. For VID = 0, it will
  * return either the pf_vid or sw_vid depending on which one is set.
  */
-static s32 fm10k_iov_select_vid(struct fm10k_vf_info *vf_info, u16 vid)
+s32 fm10k_iov_select_vid(struct fm10k_vf_info *vf_info, u16 vid)
 {
 	if (!vid)
 		return vf_info->pf_vid ? vf_info->pf_vid : vf_info->sw_vid;
@@ -1197,126 +1180,6 @@ static s32 fm10k_iov_select_vid(struct fm10k_vf_info *vf_info, u16 vid)
 }
 
 /**
- *  fm10k_iov_msg_mac_vlan_pf - Message handler for MAC/VLAN request from VF
- *  @hw: Pointer to hardware structure
- *  @results: Pointer array to message, results[0] is pointer to message
- *  @mbx: Pointer to mailbox information structure
- *
- *  This function is a default handler for MAC/VLAN requests from the VF.
- *  The assumption is that in this case it is acceptable to just directly
- *  hand off the message from the VF to the underlying shared code.
- **/
-s32 fm10k_iov_msg_mac_vlan_pf(struct fm10k_hw *hw, u32 **results,
-			      struct fm10k_mbx_info *mbx)
-{
-	struct fm10k_vf_info *vf_info = (struct fm10k_vf_info *)mbx;
-	u8 mac[ETH_ALEN];
-	u32 *result;
-	int err = 0;
-	bool set;
-	u16 vlan;
-	u32 vid;
-
-	/* we shouldn't be updating rules on a disabled interface */
-	if (!FM10K_VF_FLAG_ENABLED(vf_info))
-		err = FM10K_ERR_PARAM;
-
-	if (!err && !!results[FM10K_MAC_VLAN_MSG_VLAN]) {
-		result = results[FM10K_MAC_VLAN_MSG_VLAN];
-
-		/* record VLAN id requested */
-		err = fm10k_tlv_attr_get_u32(result, &vid);
-		if (err)
-			return err;
-
-		set = !(vid & FM10K_VLAN_CLEAR);
-		vid &= ~FM10K_VLAN_CLEAR;
-
-		/* if the length field has been set, this is a multi-bit
-		 * update request. For multi-bit requests, simply disallow
-		 * them when the pf_vid has been set. In this case, the PF
-		 * should have already cleared the VLAN_TABLE, and if we
-		 * allowed them, it could allow a rogue VF to receive traffic
-		 * on a VLAN it was not assigned. In the single-bit case, we
-		 * need to modify requests for VLAN 0 to use the default PF or
-		 * SW vid when assigned.
-		 */
-
-		if (vid >> 16) {
-			/* prevent multi-bit requests when PF has
-			 * administratively set the VLAN for this VF
-			 */
-			if (vf_info->pf_vid)
-				return FM10K_ERR_PARAM;
-		} else {
-			err = fm10k_iov_select_vid(vf_info, (u16)vid);
-			if (err < 0)
-				return err;
-
-			vid = err;
-		}
-
-		/* update VSI info for VF in regards to VLAN table */
-		err = hw->mac.ops.update_vlan(hw, vid, vf_info->vsi, set);
-	}
-
-	if (!err && !!results[FM10K_MAC_VLAN_MSG_MAC]) {
-		result = results[FM10K_MAC_VLAN_MSG_MAC];
-
-		/* record unicast MAC address requested */
-		err = fm10k_tlv_attr_get_mac_vlan(result, mac, &vlan);
-		if (err)
-			return err;
-
-		/* block attempts to set MAC for a locked device */
-		if (is_valid_ether_addr(vf_info->mac) &&
-		    !ether_addr_equal(mac, vf_info->mac))
-			return FM10K_ERR_PARAM;
-
-		set = !(vlan & FM10K_VLAN_CLEAR);
-		vlan &= ~FM10K_VLAN_CLEAR;
-
-		err = fm10k_iov_select_vid(vf_info, vlan);
-		if (err < 0)
-			return err;
-
-		vlan = (u16)err;
-
-		/* notify switch of request for new unicast address */
-		err = hw->mac.ops.update_uc_addr(hw, vf_info->glort,
-						 mac, vlan, set, 0);
-	}
-
-	if (!err && !!results[FM10K_MAC_VLAN_MSG_MULTICAST]) {
-		result = results[FM10K_MAC_VLAN_MSG_MULTICAST];
-
-		/* record multicast MAC address requested */
-		err = fm10k_tlv_attr_get_mac_vlan(result, mac, &vlan);
-		if (err)
-			return err;
-
-		/* verify that the VF is allowed to request multicast */
-		if (!(vf_info->vf_flags & FM10K_VF_FLAG_MULTI_ENABLED))
-			return FM10K_ERR_PARAM;
-
-		set = !(vlan & FM10K_VLAN_CLEAR);
-		vlan &= ~FM10K_VLAN_CLEAR;
-
-		err = fm10k_iov_select_vid(vf_info, vlan);
-		if (err < 0)
-			return err;
-
-		vlan = (u16)err;
-
-		/* notify switch of request for new multicast address */
-		err = hw->mac.ops.update_mc_addr(hw, vf_info->glort,
-						 mac, vlan, set);
-	}
-
-	return err;
-}
-
-/**
  *  fm10k_iov_supported_xcast_mode_pf - Determine best match for xcast mode
  *  @vf_info: VF info structure containing capability flags
  *  @mode: Requested xcast mode
@@ -1334,19 +1197,19 @@ static u8 fm10k_iov_supported_xcast_mode_pf(struct fm10k_vf_info *vf_info,
 	case FM10K_XCAST_MODE_PROMISC:
 		if (vf_flags & FM10K_VF_FLAG_PROMISC_CAPABLE)
 			return FM10K_XCAST_MODE_PROMISC;
-		/* fallthough */
+		fallthrough;
 	case FM10K_XCAST_MODE_ALLMULTI:
 		if (vf_flags & FM10K_VF_FLAG_ALLMULTI_CAPABLE)
 			return FM10K_XCAST_MODE_ALLMULTI;
-		/* fallthough */
+		fallthrough;
 	case FM10K_XCAST_MODE_MULTI:
 		if (vf_flags & FM10K_VF_FLAG_MULTI_CAPABLE)
 			return FM10K_XCAST_MODE_MULTI;
-		/* fallthough */
+		fallthrough;
 	case FM10K_XCAST_MODE_NONE:
 		if (vf_flags & FM10K_VF_FLAG_NONE_CAPABLE)
 			return FM10K_XCAST_MODE_NONE;
-		/* fallthough */
+		fallthrough;
 	default:
 		break;
 	}
@@ -1369,7 +1232,6 @@ s32 fm10k_iov_msg_lport_state_pf(struct fm10k_hw *hw, u32 **results,
 				 struct fm10k_mbx_info *mbx)
 {
 	struct fm10k_vf_info *vf_info = (struct fm10k_vf_info *)mbx;
-	u32 *result;
 	s32 err = 0;
 	u32 msg[2];
 	u8 mode = 0;
@@ -1379,7 +1241,7 @@ s32 fm10k_iov_msg_lport_state_pf(struct fm10k_hw *hw, u32 **results,
 		return FM10K_ERR_PARAM;
 
 	if (!!results[FM10K_LPORT_STATE_MSG_XCAST_MODE]) {
-		result = results[FM10K_LPORT_STATE_MSG_XCAST_MODE];
+		u32 *result = results[FM10K_LPORT_STATE_MSG_XCAST_MODE];
 
 		/* XCAST mode update requested */
 		err = fm10k_tlv_attr_get_u8(result, &mode);
@@ -1435,7 +1297,7 @@ s32 fm10k_iov_msg_lport_state_pf(struct fm10k_hw *hw, u32 **results,
 }
 
 /**
- *  fm10k_update_stats_hw_pf - Updates hardware related statistics of PF
+ *  fm10k_update_hw_stats_pf - Updates hardware related statistics of PF
  *  @hw: pointer to hardware structure
  *  @stats: pointer to the stats structure to update
  *
@@ -1527,7 +1389,7 @@ static void fm10k_rebind_hw_stats_pf(struct fm10k_hw *hw,
 	fm10k_unbind_hw_stats_32b(&stats->nodesc_drop);
 
 	/* Unbind Queue Statistics */
-	fm10k_unbind_hw_stats_q(stats->q, 0, hw->mac.max_queues);
+	fm10k_unbind_hw_stats_q(stats->q, hw->mac.max_queues);
 
 	/* Reinitialize bases for all stats */
 	fm10k_update_hw_stats_pf(hw, stats);
@@ -1583,7 +1445,7 @@ static s32 fm10k_get_fault_pf(struct fm10k_hw *hw, int type,
 	/* read remaining fields */
 	fault->address = fm10k_read_reg(hw, type + FM10K_FAULT_ADDR_HI);
 	fault->address <<= 32;
-	fault->address = fm10k_read_reg(hw, type + FM10K_FAULT_ADDR_LO);
+	fault->address |= fm10k_read_reg(hw, type + FM10K_FAULT_ADDR_LO);
 	fault->specinfo = fm10k_read_reg(hw, type + FM10K_FAULT_SPECINFO);
 
 	/* clear valid bit to allow for next error */
@@ -1593,8 +1455,7 @@ static s32 fm10k_get_fault_pf(struct fm10k_hw *hw, int type,
 	if (func & FM10K_FAULT_FUNC_PF)
 		fault->func = 0;
 	else
-		fault->func = 1 + ((func & FM10K_FAULT_FUNC_VF_MASK) >>
-				   FM10K_FAULT_FUNC_VF_SHIFT);
+		fault->func = 1 + FIELD_GET(FM10K_FAULT_FUNC_VF_MASK, func);
 
 	/* record fault type */
 	fault->type = func & FM10K_FAULT_FUNC_TYPE_MASK;
@@ -1659,7 +1520,7 @@ const struct fm10k_tlv_attr fm10k_lport_map_msg_attr[] = {
  *  switch API.
  **/
 s32 fm10k_msg_lport_map_pf(struct fm10k_hw *hw, u32 **results,
-			   struct fm10k_mbx_info *mbx)
+			   struct fm10k_mbx_info __always_unused *mbx)
 {
 	u16 glort, mask;
 	u32 dglort_map;
@@ -1702,7 +1563,7 @@ const struct fm10k_tlv_attr fm10k_update_pvid_msg_attr[] = {
  *  This handler configures the default VLAN for the PF
  **/
 static s32 fm10k_msg_update_pvid_pf(struct fm10k_hw *hw, u32 **results,
-				    struct fm10k_mbx_info *mbx)
+				    struct fm10k_mbx_info __always_unused *mbx)
 {
 	u16 glort, pvid;
 	u32 pvid_update;
@@ -1763,7 +1624,7 @@ const struct fm10k_tlv_attr fm10k_err_msg_attr[] = {
  *  messages that the PF has sent.
  **/
 s32 fm10k_msg_err_pf(struct fm10k_hw *hw, u32 **results,
-		     struct fm10k_mbx_info *mbx)
+		     struct fm10k_mbx_info __always_unused *mbx)
 {
 	struct fm10k_swapi_error err_msg;
 	s32 err;
diff --git a/drivers/net/ethernet/intel/fm10k/fm10k_pf.h b/drivers/net/ethernet/intel/fm10k/fm10k_pf.h
index 3336d3c10760..ad3696893cb1 100644
--- a/drivers/net/ethernet/intel/fm10k/fm10k_pf.h
+++ b/drivers/net/ethernet/intel/fm10k/fm10k_pf.h
@@ -1,22 +1,5 @@
-/* Intel(R) Ethernet Switch Host Interface Driver
- * Copyright(c) 2013 - 2016 Intel Corporation.
- *
- * This program is free software; you can redistribute it and/or modify it
- * under the terms and conditions of the GNU General Public License,
- * version 2, as published by the Free Software Foundation.
- *
- * This program is distributed in the hope it will be useful, but WITHOUT
- * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
- * FITNESS FOR A PARTICULAR PURPOSE.  See the GNU General Public License for
- * more details.
- *
- * The full GNU General Public License is included in this distribution in
- * the file called "COPYING".
- *
- * Contact Information:
- * e1000-devel Mailing List <e1000-devel@lists.sourceforge.net>
- * Intel Corporation, 5200 N.E. Elam Young Parkway, Hillsboro, OR 97124-6497
- */
+/* SPDX-License-Identifier: GPL-2.0 */
+/* Copyright(c) 2013 - 2018 Intel Corporation. */
 
 #ifndef _FM10K_PF_H_
 #define _FM10K_PF_H_
@@ -114,9 +97,8 @@ extern const struct fm10k_tlv_attr fm10k_err_msg_attr[];
 #define FM10K_PF_MSG_ERR_HANDLER(msg, func) \
 	FM10K_MSG_HANDLER(FM10K_PF_MSG_ID_##msg, fm10k_err_msg_attr, func)
 
+s32 fm10k_iov_select_vid(struct fm10k_vf_info *vf_info, u16 vid);
 s32 fm10k_iov_msg_msix_pf(struct fm10k_hw *, u32 **, struct fm10k_mbx_info *);
-s32 fm10k_iov_msg_mac_vlan_pf(struct fm10k_hw *, u32 **,
-			      struct fm10k_mbx_info *);
 s32 fm10k_iov_msg_lport_state_pf(struct fm10k_hw *, u32 **,
 				 struct fm10k_mbx_info *);
 
diff --git a/drivers/net/ethernet/intel/fm10k/fm10k_tlv.c b/drivers/net/ethernet/intel/fm10k/fm10k_tlv.c
index f8e87bf086b9..75cbdf2dbbe3 100644
--- a/drivers/net/ethernet/intel/fm10k/fm10k_tlv.c
+++ b/drivers/net/ethernet/intel/fm10k/fm10k_tlv.c
@@ -1,22 +1,5 @@
-/* Intel(R) Ethernet Switch Host Interface Driver
- * Copyright(c) 2013 - 2016 Intel Corporation.
- *
- * This program is free software; you can redistribute it and/or modify it
- * under the terms and conditions of the GNU General Public License,
- * version 2, as published by the Free Software Foundation.
- *
- * This program is distributed in the hope it will be useful, but WITHOUT
- * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
- * FITNESS FOR A PARTICULAR PURPOSE.  See the GNU General Public License for
- * more details.
- *
- * The full GNU General Public License is included in this distribution in
- * the file called "COPYING".
- *
- * Contact Information:
- * e1000-devel Mailing List <e1000-devel@lists.sourceforge.net>
- * Intel Corporation, 5200 N.E. Elam Young Parkway, Hillsboro, OR 97124-6497
- */
+// SPDX-License-Identifier: GPL-2.0
+/* Copyright(c) 2013 - 2019 Intel Corporation. */
 
 #include "fm10k_tlv.h"
 
@@ -95,7 +78,7 @@ static s32 fm10k_tlv_attr_put_null_string(u32 *msg, u16 attr_id,
  *  @string: Pointer to location of destination string
  *
  *  This function pulls the string back out of the attribute and will place
- *  it in the array pointed by by string.  It will return success if provided
+ *  it in the array pointed by string.  It will return success if provided
  *  with a valid pointers.
  **/
 static s32 fm10k_tlv_attr_get_null_string(u32 *attr, unsigned char *string)
@@ -120,6 +103,7 @@ static s32 fm10k_tlv_attr_get_null_string(u32 *attr, unsigned char *string)
  *  @msg: Pointer to message block
  *  @attr_id: Attribute ID
  *  @mac_addr: MAC address to be stored
+ *  @vlan: VLAN to be stored
  *
  *  This function will reorder a MAC address to be CPU endian and store it
  *  in the attribute buffer.  It will return success if provided with a
@@ -155,11 +139,11 @@ s32 fm10k_tlv_attr_put_mac_vlan(u32 *msg, u16 attr_id,
 /**
  *  fm10k_tlv_attr_get_mac_vlan - Get MAC/VLAN stored in attribute
  *  @attr: Pointer to attribute
- *  @attr_id: Attribute ID
  *  @mac_addr: location of buffer to store MAC address
+ *  @vlan: location of buffer to store VLAN
  *
  *  This function pulls the MAC address back out of the attribute and will
- *  place it in the array pointed by by mac_addr.  It will return success
+ *  place it in the array pointed by mac_addr.  It will return success
  *  if provided with a valid pointers.
  **/
 s32 fm10k_tlv_attr_get_mac_vlan(u32 *attr, u8 *mac_addr, u16 *vlan)
@@ -488,7 +472,7 @@ static s32 fm10k_tlv_attr_parse(u32 *attr, u32 **results,
 				const struct fm10k_tlv_attr *tlv_attr)
 {
 	u32 i, attr_id, offset = 0;
-	s32 err = 0;
+	s32 err;
 	u16 len;
 
 	/* verify pointers are not NULL */
@@ -549,7 +533,7 @@ static s32 fm10k_tlv_attr_parse(u32 *attr, u32 **results,
  *  @hw: Pointer to hardware structure
  *  @msg: Pointer to message
  *  @mbx: Pointer to mailbox information structure
- *  @func: Function array containing list of message handling functions
+ *  @data: Pointer to message handler data structure
  *
  *  This function should be the first function called upon receiving a
  *  message.  The handler will identify the message type and call the correct
@@ -600,11 +584,12 @@ s32 fm10k_tlv_msg_parse(struct fm10k_hw *hw, u32 *msg,
  *  @mbx: Unused mailbox pointer
  *
  *  This function is a default handler for unrecognized messages.  At a
- *  a minimum it just indicates that the message requested was
+ *  minimum it just indicates that the message requested was
  *  unimplemented.
  **/
-s32 fm10k_tlv_msg_error(struct fm10k_hw *hw, u32 **results,
-			struct fm10k_mbx_info *mbx)
+s32 fm10k_tlv_msg_error(struct fm10k_hw __always_unused *hw,
+			u32 __always_unused **results,
+			struct fm10k_mbx_info __always_unused *mbx)
 {
 	return FM10K_NOT_IMPLEMENTED;
 }
diff --git a/drivers/net/ethernet/intel/fm10k/fm10k_tlv.h b/drivers/net/ethernet/intel/fm10k/fm10k_tlv.h
index a1f1027fe184..ceb9b791f799 100644
--- a/drivers/net/ethernet/intel/fm10k/fm10k_tlv.h
+++ b/drivers/net/ethernet/intel/fm10k/fm10k_tlv.h
@@ -1,22 +1,5 @@
-/* Intel(R) Ethernet Switch Host Interface Driver
- * Copyright(c) 2013 - 2016 Intel Corporation.
- *
- * This program is free software; you can redistribute it and/or modify it
- * under the terms and conditions of the GNU General Public License,
- * version 2, as published by the Free Software Foundation.
- *
- * This program is distributed in the hope it will be useful, but WITHOUT
- * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
- * FITNESS FOR A PARTICULAR PURPOSE.  See the GNU General Public License for
- * more details.
- *
- * The full GNU General Public License is included in this distribution in
- * the file called "COPYING".
- *
- * Contact Information:
- * e1000-devel Mailing List <e1000-devel@lists.sourceforge.net>
- * Intel Corporation, 5200 N.E. Elam Young Parkway, Hillsboro, OR 97124-6497
- */
+/* SPDX-License-Identifier: GPL-2.0 */
+/* Copyright(c) 2013 - 2019 Intel Corporation. */
 
 #ifndef _FM10K_TLV_H_
 #define _FM10K_TLV_H_
@@ -93,8 +76,8 @@ struct fm10k_tlv_attr {
 #define FM10K_TLV_ATTR_S32(id)		    { id, FM10K_TLV_SIGNED, 4 }
 #define FM10K_TLV_ATTR_S64(id)		    { id, FM10K_TLV_SIGNED, 8 }
 #define FM10K_TLV_ATTR_LE_STRUCT(id, len)   { id, FM10K_TLV_LE_STRUCT, len }
-#define FM10K_TLV_ATTR_NESTED(id)	    { id, FM10K_TLV_NESTED }
-#define FM10K_TLV_ATTR_LAST		    { FM10K_TLV_ERROR }
+#define FM10K_TLV_ATTR_NESTED(id)	    { id, FM10K_TLV_NESTED, 0 }
+#define FM10K_TLV_ATTR_LAST		    { FM10K_TLV_ERROR, 0, 0 }
 
 struct fm10k_msg_data {
 	unsigned int		    id;
diff --git a/drivers/net/ethernet/intel/fm10k/fm10k_type.h b/drivers/net/ethernet/intel/fm10k/fm10k_type.h
index 6bb16c13d9d6..63968c5d7c5d 100644
--- a/drivers/net/ethernet/intel/fm10k/fm10k_type.h
+++ b/drivers/net/ethernet/intel/fm10k/fm10k_type.h
@@ -1,22 +1,5 @@
-/* Intel(R) Ethernet Switch Host Interface Driver
- * Copyright(c) 2013 - 2016 Intel Corporation.
- *
- * This program is free software; you can redistribute it and/or modify it
- * under the terms and conditions of the GNU General Public License,
- * version 2, as published by the Free Software Foundation.
- *
- * This program is distributed in the hope it will be useful, but WITHOUT
- * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
- * FITNESS FOR A PARTICULAR PURPOSE.  See the GNU General Public License for
- * more details.
- *
- * The full GNU General Public License is included in this distribution in
- * the file called "COPYING".
- *
- * Contact Information:
- * e1000-devel Mailing List <e1000-devel@lists.sourceforge.net>
- * Intel Corporation, 5200 N.E. Elam Young Parkway, Hillsboro, OR 97124-6497
- */
+/* SPDX-License-Identifier: GPL-2.0 */
+/* Copyright(c) 2013 - 2019 Intel Corporation. */
 
 #ifndef _FM10K_TYPE_H_
 #define _FM10K_TYPE_H_
@@ -32,6 +15,8 @@ struct fm10k_hw;
 
 #define FM10K_DEV_ID_PF			0x15A4
 #define FM10K_DEV_ID_VF			0x15A5
+#define FM10K_DEV_ID_SDI_FM10420_QDA2	0x15D0
+#define FM10K_DEV_ID_SDI_FM10420_DA2	0x15D5
 
 #define FM10K_MAX_QUEUES		256
 #define FM10K_MAX_QUEUES_PF		128
@@ -596,6 +581,7 @@ struct fm10k_vf_info {
 	 * at the same offset as the mailbox
 	 */
 	struct fm10k_mbx_info	mbx;		/* PF side of VF mailbox */
+	struct fm10k_hw_stats_q	stats[FM10K_MAX_QUEUES_POOL];
 	int			rate;		/* Tx BW cap as defined by OS */
 	u16			glort;		/* resource tag for this VF */
 	u16			sw_vid;		/* Switch API assigned VLAN */
diff --git a/drivers/net/ethernet/intel/fm10k/fm10k_vf.c b/drivers/net/ethernet/intel/fm10k/fm10k_vf.c
index 337ba65a9411..6861a0bdc14e 100644
--- a/drivers/net/ethernet/intel/fm10k/fm10k_vf.c
+++ b/drivers/net/ethernet/intel/fm10k/fm10k_vf.c
@@ -1,23 +1,7 @@
-/* Intel(R) Ethernet Switch Host Interface Driver
- * Copyright(c) 2013 - 2016 Intel Corporation.
- *
- * This program is free software; you can redistribute it and/or modify it
- * under the terms and conditions of the GNU General Public License,
- * version 2, as published by the Free Software Foundation.
- *
- * This program is distributed in the hope it will be useful, but WITHOUT
- * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
- * FITNESS FOR A PARTICULAR PURPOSE.  See the GNU General Public License for
- * more details.
- *
- * The full GNU General Public License is included in this distribution in
- * the file called "COPYING".
- *
- * Contact Information:
- * e1000-devel Mailing List <e1000-devel@lists.sourceforge.net>
- * Intel Corporation, 5200 N.E. Elam Young Parkway, Hillsboro, OR 97124-6497
- */
+// SPDX-License-Identifier: GPL-2.0
+/* Copyright(c) 2013 - 2019 Intel Corporation. */
 
+#include <linux/bitfield.h>
 #include "fm10k_vf.h"
 
 /**
@@ -143,15 +127,14 @@ static s32 fm10k_init_hw_vf(struct fm10k_hw *hw)
 	hw->mac.max_queues = i;
 
 	/* fetch default VLAN and ITR scale */
-	hw->mac.default_vid = (fm10k_read_reg(hw, FM10K_TXQCTL(0)) &
-			       FM10K_TXQCTL_VID_MASK) >> FM10K_TXQCTL_VID_SHIFT;
+	hw->mac.default_vid = FIELD_GET(FM10K_TXQCTL_VID_MASK,
+					fm10k_read_reg(hw, FM10K_TXQCTL(0)));
 	/* Read the ITR scale from TDLEN. See the definition of
 	 * FM10K_TDLEN_ITR_SCALE_SHIFT for more information about how TDLEN is
 	 * used here.
 	 */
-	hw->mac.itr_scale = (fm10k_read_reg(hw, FM10K_TDLEN(0)) &
-			     FM10K_TDLEN_ITR_SCALE_MASK) >>
-			    FM10K_TDLEN_ITR_SCALE_SHIFT;
+	hw->mac.itr_scale = FIELD_GET(FM10K_TDLEN_ITR_SCALE_MASK,
+				      fm10k_read_reg(hw, FM10K_TDLEN(0)));
 
 	return 0;
 
@@ -215,7 +198,7 @@ static s32 fm10k_update_vlan_vf(struct fm10k_hw *hw, u32 vid, u8 vsi, bool set)
  *  This function should determine the MAC address for the VF
  **/
 s32 fm10k_msg_mac_vlan_vf(struct fm10k_hw *hw, u32 **results,
-			  struct fm10k_mbx_info *mbx)
+			  struct fm10k_mbx_info __always_unused *mbx)
 {
 	u8 perm_addr[ETH_ALEN];
 	u16 vid;
@@ -284,8 +267,10 @@ static s32 fm10k_read_mac_addr_vf(struct fm10k_hw *hw)
  *  This function is used to add or remove unicast MAC addresses for
  *  the VF.
  **/
-static s32 fm10k_update_uc_addr_vf(struct fm10k_hw *hw, u16 glort,
-				   const u8 *mac, u16 vid, bool add, u8 flags)
+static s32 fm10k_update_uc_addr_vf(struct fm10k_hw *hw,
+				   u16 __always_unused glort,
+				   const u8 *mac, u16 vid, bool add,
+				   u8 __always_unused flags)
 {
 	struct fm10k_mbx_info *mbx = &hw->mbx;
 	u32 msg[7];
@@ -326,7 +311,8 @@ static s32 fm10k_update_uc_addr_vf(struct fm10k_hw *hw, u16 glort,
  *  This function is used to add or remove multicast MAC addresses for
  *  the VF.
  **/
-static s32 fm10k_update_mc_addr_vf(struct fm10k_hw *hw, u16 glort,
+static s32 fm10k_update_mc_addr_vf(struct fm10k_hw *hw,
+				   u16 __always_unused glort,
 				   const u8 *mac, u16 vid, bool add)
 {
 	struct fm10k_mbx_info *mbx = &hw->mbx;
@@ -390,7 +376,7 @@ const struct fm10k_tlv_attr fm10k_lport_state_msg_attr[] = {
  *  are ready to bring up the interface.
  **/
 s32 fm10k_msg_lport_state_vf(struct fm10k_hw *hw, u32 **results,
-			     struct fm10k_mbx_info *mbx)
+			     struct fm10k_mbx_info __always_unused *mbx)
 {
 	hw->mac.dglort_map = !results[FM10K_LPORT_STATE_MSG_READY] ?
 			     FM10K_DGLORTMAP_NONE : FM10K_DGLORTMAP_ZERO;
@@ -409,8 +395,9 @@ s32 fm10k_msg_lport_state_vf(struct fm10k_hw *hw, u32 **results,
  *  enabled we can add filters, if it is disabled all filters for this
  *  logical port are flushed.
  **/
-static s32 fm10k_update_lport_state_vf(struct fm10k_hw *hw, u16 glort,
-				       u16 count, bool enable)
+static s32 fm10k_update_lport_state_vf(struct fm10k_hw *hw,
+				       u16 __always_unused glort,
+				       u16 __always_unused count, bool enable)
 {
 	struct fm10k_mbx_info *mbx = &hw->mbx;
 	u32 msg[2];
@@ -437,7 +424,8 @@ static s32 fm10k_update_lport_state_vf(struct fm10k_hw *hw, u16 glort,
  *  so that it can enable either multicast, multicast promiscuous, or
  *  promiscuous mode of operation.
  **/
-static s32 fm10k_update_xcast_mode_vf(struct fm10k_hw *hw, u16 glort, u8 mode)
+static s32 fm10k_update_xcast_mode_vf(struct fm10k_hw *hw,
+				      u16 __always_unused glort, u8 mode)
 {
 	struct fm10k_mbx_info *mbx = &hw->mbx;
 	u32 msg[3];
@@ -477,7 +465,7 @@ static void fm10k_rebind_hw_stats_vf(struct fm10k_hw *hw,
 				     struct fm10k_hw_stats *stats)
 {
 	/* Unbind Queue Statistics */
-	fm10k_unbind_hw_stats_q(stats->q, 0, hw->mac.max_queues);
+	fm10k_unbind_hw_stats_q(stats->q, hw->mac.max_queues);
 
 	/* Reinitialize bases for all stats */
 	fm10k_update_hw_stats_vf(hw, stats);
@@ -492,7 +480,7 @@ static void fm10k_rebind_hw_stats_vf(struct fm10k_hw *hw,
  *  that information to then populate a DGLORTMAP/DEC entry and the queues
  *  to which it has been assigned.
  **/
-static s32 fm10k_configure_dglort_map_vf(struct fm10k_hw *hw,
+static s32 fm10k_configure_dglort_map_vf(struct fm10k_hw __always_unused *hw,
 					 struct fm10k_dglort_cfg *dglort)
 {
 	/* verify the dglort pointer */
diff --git a/drivers/net/ethernet/intel/fm10k/fm10k_vf.h b/drivers/net/ethernet/intel/fm10k/fm10k_vf.h
index 2662f33c0c71..787d0d570a28 100644
--- a/drivers/net/ethernet/intel/fm10k/fm10k_vf.h
+++ b/drivers/net/ethernet/intel/fm10k/fm10k_vf.h
@@ -1,22 +1,5 @@
-/* Intel(R) Ethernet Switch Host Interface Driver
- * Copyright(c) 2013 - 2016 Intel Corporation.
- *
- * This program is free software; you can redistribute it and/or modify it
- * under the terms and conditions of the GNU General Public License,
- * version 2, as published by the Free Software Foundation.
- *
- * This program is distributed in the hope it will be useful, but WITHOUT
- * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
- * FITNESS FOR A PARTICULAR PURPOSE.  See the GNU General Public License for
- * more details.
- *
- * The full GNU General Public License is included in this distribution in
- * the file called "COPYING".
- *
- * Contact Information:
- * e1000-devel Mailing List <e1000-devel@lists.sourceforge.net>
- * Intel Corporation, 5200 N.E. Elam Young Parkway, Hillsboro, OR 97124-6497
- */
+/* SPDX-License-Identifier: GPL-2.0 */
+/* Copyright(c) 2013 - 2018 Intel Corporation. */
 
 #ifndef _FM10K_VF_H_
 #define _FM10K_VF_H_
diff --git a/drivers/net/ethernet/intel/i40e/Makefile b/drivers/net/ethernet/intel/i40e/Makefile
index 3da482c3d68d..9faa4339a76c 100644
--- a/drivers/net/ethernet/intel/i40e/Makefile
+++ b/drivers/net/ethernet/intel/i40e/Makefile
@@ -1,28 +1,5 @@
-################################################################################
-#
-# Intel Ethernet Controller XL710 Family Linux Driver
-# Copyright(c) 2013 - 2015 Intel Corporation.
-#
-# This program is free software; you can redistribute it and/or modify it
-# under the terms and conditions of the GNU General Public License,
-# version 2, as published by the Free Software Foundation.
-#
-# This program is distributed in the hope it will be useful, but WITHOUT
-# ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
-# FITNESS FOR A PARTICULAR PURPOSE.  See the GNU General Public License for
-# more details.
-#
-# You should have received a copy of the GNU General Public License along
-# with this program.  If not, see <http://www.gnu.org/licenses/>.
-#
-# The full GNU General Public License is included in this distribution in
-# the file called "COPYING".
-#
-# Contact Information:
-# e1000-devel Mailing List <e1000-devel@lists.sourceforge.net>
-# Intel Corporation, 5200 N.E. Elam Young Parkway, Hillsboro, OR 97124-6497
-#
-################################################################################
+# SPDX-License-Identifier: GPL-2.0
+# Copyright(c) 2013 - 2018 Intel Corporation.
 
 #
 # Makefile for the Intel(R) Ethernet Connection XL710 (i40e.ko) driver
@@ -33,7 +10,7 @@ subdir-ccflags-y += -I$(src)
 
 obj-$(CONFIG_I40E) += i40e.o
 
-i40e-objs := i40e_main.o \
+i40e-y := i40e_main.o \
 	i40e_ethtool.o	\
 	i40e_adminq.o	\
 	i40e_common.o	\
@@ -44,7 +21,10 @@ i40e-objs := i40e_main.o \
 	i40e_diag.o	\
 	i40e_txrx.o	\
 	i40e_ptp.o	\
+	i40e_ddp.o \
 	i40e_client.o   \
-	i40e_virtchnl_pf.o
+	i40e_virtchnl_pf.o \
+	i40e_xsk.o	\
+	i40e_devlink.o
 
 i40e-$(CONFIG_I40E_DCB) += i40e_dcb.o i40e_dcb_nl.o
diff --git a/drivers/net/ethernet/intel/i40e/i40e.h b/drivers/net/ethernet/intel/i40e/i40e.h
index 25bf336c5f38..801a57a925da 100644
--- a/drivers/net/ethernet/intel/i40e/i40e.h
+++ b/drivers/net/ethernet/intel/i40e/i40e.h
@@ -1,71 +1,31 @@
-/*******************************************************************************
- *
- * Intel Ethernet Controller XL710 Family Linux Driver
- * Copyright(c) 2013 - 2016 Intel Corporation.
- *
- * This program is free software; you can redistribute it and/or modify it
- * under the terms and conditions of the GNU General Public License,
- * version 2, as published by the Free Software Foundation.
- *
- * This program is distributed in the hope it will be useful, but WITHOUT
- * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
- * FITNESS FOR A PARTICULAR PURPOSE.  See the GNU General Public License for
- * more details.
- *
- * You should have received a copy of the GNU General Public License along
- * with this program.  If not, see <http://www.gnu.org/licenses/>.
- *
- * The full GNU General Public License is included in this distribution in
- * the file called "COPYING".
- *
- * Contact Information:
- * e1000-devel Mailing List <e1000-devel@lists.sourceforge.net>
- * Intel Corporation, 5200 N.E. Elam Young Parkway, Hillsboro, OR 97124-6497
- *
- ******************************************************************************/
+/* SPDX-License-Identifier: GPL-2.0 */
+/* Copyright(c) 2013 - 2021 Intel Corporation. */
 
 #ifndef _I40E_H_
 #define _I40E_H_
 
-#include <net/tcp.h>
-#include <net/udp.h>
-#include <linux/types.h>
-#include <linux/errno.h>
-#include <linux/module.h>
+#include <linux/linkmode.h>
 #include <linux/pci.h>
-#include <linux/aer.h>
-#include <linux/netdevice.h>
-#include <linux/ioport.h>
-#include <linux/iommu.h>
-#include <linux/slab.h>
-#include <linux/list.h>
-#include <linux/hashtable.h>
-#include <linux/string.h>
-#include <linux/in.h>
-#include <linux/ip.h>
-#include <linux/sctp.h>
-#include <linux/pkt_sched.h>
-#include <linux/ipv6.h>
-#include <net/checksum.h>
-#include <net/ip6_checksum.h>
-#include <linux/ethtool.h>
-#include <linux/if_vlan.h>
-#include <linux/if_bridge.h>
-#include <linux/clocksource.h>
-#include <linux/net_tstamp.h>
 #include <linux/ptp_clock_kernel.h>
-#include "i40e_type.h"
+#include <linux/types.h>
+#include <linux/avf/virtchnl.h>
+#include <linux/net/intel/i40e_client.h>
+#include <net/devlink.h>
+#include <net/pkt_cls.h>
+#include <net/udp_tunnel.h>
+#include "i40e_dcb.h"
+#include "i40e_debug.h"
+#include "i40e_devlink.h"
+#include "i40e_io.h"
 #include "i40e_prototype.h"
-#include "i40e_client.h"
-#include "i40e_virtchnl.h"
-#include "i40e_virtchnl_pf.h"
+#include "i40e_register.h"
 #include "i40e_txrx.h"
-#include "i40e_dcb.h"
 
 /* Useful i40e defaults */
 #define I40E_MAX_VEB			16
 
 #define I40E_MAX_NUM_DESCRIPTORS	4096
+#define I40E_MAX_NUM_DESCRIPTORS_XL710	8160
 #define I40E_MAX_CSR_SPACE		(4 * 1024 * 1024 - 64 * 1024)
 #define I40E_DEFAULT_NUM_DESCRIPTORS	512
 #define I40E_REQ_DESCRIPTOR_MULTIPLE	32
@@ -75,40 +35,38 @@
 #define I40E_MIN_VSI_ALLOC		83 /* LAN, ATR, FCOE, 64 VF */
 /* max 16 qps */
 #define i40e_default_queues_per_vmdq(pf) \
-		(((pf)->flags & I40E_FLAG_RSS_AQ_CAPABLE) ? 4 : 1)
+	(test_bit(I40E_HW_CAP_RSS_AQ, (pf)->hw.caps) ? 4 : 1)
 #define I40E_DEFAULT_QUEUES_PER_VF	4
-#define I40E_DEFAULT_QUEUES_PER_TC	1 /* should be a power of 2 */
+#define I40E_MAX_VF_QUEUES		16
 #define i40e_pf_get_max_q_per_tc(pf) \
-		(((pf)->flags & I40E_FLAG_128_QP_RSS_CAPABLE) ? 128 : 64)
-#define I40E_FDIR_RING			0
+	(test_bit(I40E_HW_CAP_128_QP_RSS, (pf)->hw.caps) ? 128 : 64)
 #define I40E_FDIR_RING_COUNT		32
 #define I40E_MAX_AQ_BUF_SIZE		4096
 #define I40E_AQ_LEN			256
+#define I40E_MIN_ARQ_LEN		1
+#define I40E_MIN_ASQ_LEN		2
 #define I40E_AQ_WORK_LIMIT		66 /* max number of VFs + a little */
 #define I40E_MAX_USER_PRIORITY		8
 #define I40E_DEFAULT_TRAFFIC_CLASS	BIT(0)
-#define I40E_DEFAULT_MSG_ENABLE		4
 #define I40E_QUEUE_WAIT_RETRY_LIMIT	10
 #define I40E_INT_NAME_STR_LEN		(IFNAMSIZ + 16)
 
-#define I40E_NVM_VERSION_LO_SHIFT	0
-#define I40E_NVM_VERSION_LO_MASK	(0xff << I40E_NVM_VERSION_LO_SHIFT)
-#define I40E_NVM_VERSION_HI_SHIFT	12
-#define I40E_NVM_VERSION_HI_MASK	(0xf << I40E_NVM_VERSION_HI_SHIFT)
-#define I40E_OEM_VER_BUILD_MASK		0xffff
-#define I40E_OEM_VER_PATCH_MASK		0xff
-#define I40E_OEM_VER_BUILD_SHIFT	8
-#define I40E_OEM_VER_SHIFT		24
 #define I40E_PHY_DEBUG_ALL \
 	(I40E_AQ_PHY_DEBUG_DISABLE_LINK_FW | \
 	I40E_AQ_PHY_DEBUG_DISABLE_ALL_LINK_FW)
 
-/* The values in here are decimal coded as hex as is the case in the NVM map*/
-#define I40E_CURRENT_NVM_VERSION_HI	0x2
-#define I40E_CURRENT_NVM_VERSION_LO	0x40
+#define I40E_OEM_EETRACK_ID		0xffffffff
+#define I40E_NVM_VERSION_LO_MASK	GENMASK(7, 0)
+#define I40E_NVM_VERSION_HI_MASK	GENMASK(15, 12)
+#define I40E_OEM_VER_BUILD_MASK		GENMASK(23, 8)
+#define I40E_OEM_VER_PATCH_MASK		GENMASK(7, 0)
+#define I40E_OEM_VER_MASK		GENMASK(31, 24)
+#define I40E_OEM_GEN_MASK		GENMASK(31, 24)
+#define I40E_OEM_SNAP_MASK		GENMASK(23, 16)
+#define I40E_OEM_RELEASE_MASK		GENMASK(15, 0)
 
 #define I40E_RX_DESC(R, i)	\
-	(&(((union i40e_32byte_rx_desc *)((R)->desc))[i]))
+	(&(((union i40e_rx_desc *)((R)->desc))[i]))
 #define I40E_TX_DESC(R, i)	\
 	(&(((struct i40e_tx_desc *)((R)->desc))[i]))
 #define I40E_TX_CTXTDESC(R, i)	\
@@ -116,11 +74,13 @@
 #define I40E_TX_FDIRDESC(R, i)	\
 	(&(((struct i40e_filter_program_desc *)((R)->desc))[i]))
 
-/* default to trying for four seconds */
-#define I40E_TRY_LINK_TIMEOUT	(4 * HZ)
+/* BW rate limiting */
+#define I40E_BW_CREDIT_DIVISOR		50 /* 50Mbps per BW credit */
+#define I40E_BW_MBPS_DIVISOR		125000 /* rate / (1000000 / 8) Mbps */
+#define I40E_MAX_BW_INACTIVE_ACCUM	4 /* accumulate 4 credits max */
 
 /* driver state flags */
-enum i40e_state_t {
+enum i40e_state {
 	__I40E_TESTING,
 	__I40E_CONFIG_BUSY,
 	__I40E_CONFIG_DONE,
@@ -128,39 +88,114 @@ enum i40e_state_t {
 	__I40E_SERVICE_SCHED,
 	__I40E_ADMINQ_EVENT_PENDING,
 	__I40E_MDD_EVENT_PENDING,
+	__I40E_MDD_VF_PRINT_PENDING,
 	__I40E_VFLR_EVENT_PENDING,
 	__I40E_RESET_RECOVERY_PENDING,
+	__I40E_TIMEOUT_RECOVERY_PENDING,
+	__I40E_MISC_IRQ_REQUESTED,
 	__I40E_RESET_INTR_RECEIVED,
 	__I40E_REINIT_REQUESTED,
 	__I40E_PF_RESET_REQUESTED,
+	__I40E_PF_RESET_AND_REBUILD_REQUESTED,
 	__I40E_CORE_RESET_REQUESTED,
 	__I40E_GLOBAL_RESET_REQUESTED,
-	__I40E_EMP_RESET_REQUESTED,
 	__I40E_EMP_RESET_INTR_RECEIVED,
 	__I40E_SUSPENDED,
 	__I40E_PTP_TX_IN_PROGRESS,
 	__I40E_BAD_EEPROM,
 	__I40E_DOWN_REQUESTED,
 	__I40E_FD_FLUSH_REQUESTED,
+	__I40E_FD_ATR_AUTO_DISABLED,
+	__I40E_FD_SB_AUTO_DISABLED,
 	__I40E_RESET_FAILED,
 	__I40E_PORT_SUSPENDED,
 	__I40E_VF_DISABLE,
+	__I40E_MACVLAN_SYNC_PENDING,
+	__I40E_TEMP_LINK_POLLING,
+	__I40E_CLIENT_SERVICE_REQUESTED,
+	__I40E_CLIENT_L2_CHANGE,
+	__I40E_CLIENT_RESET,
+	__I40E_VIRTCHNL_OP_PENDING,
+	__I40E_RECOVERY_MODE,
+	__I40E_VF_RESETS_DISABLED,	/* disable resets during i40e_remove */
+	__I40E_IN_REMOVE,
+	__I40E_VFS_RELEASING,
 	/* This must be last as it determines the size of the BITMAP */
 	__I40E_STATE_SIZE__,
 };
 
+#define I40E_PF_RESET_FLAG	BIT_ULL(__I40E_PF_RESET_REQUESTED)
+#define I40E_PF_RESET_AND_REBUILD_FLAG	\
+	BIT_ULL(__I40E_PF_RESET_AND_REBUILD_REQUESTED)
+
 /* VSI state flags */
-enum i40e_vsi_state_t {
+enum i40e_vsi_state {
 	__I40E_VSI_DOWN,
 	__I40E_VSI_NEEDS_RESTART,
 	__I40E_VSI_SYNCING_FILTERS,
 	__I40E_VSI_OVERFLOW_PROMISC,
 	__I40E_VSI_REINIT_REQUESTED,
 	__I40E_VSI_DOWN_REQUESTED,
+	__I40E_VSI_RELEASING,
 	/* This must be last as it determines the size of the BITMAP */
 	__I40E_VSI_STATE_SIZE__,
 };
 
+enum i40e_pf_flags {
+	I40E_FLAG_MSI_ENA,
+	I40E_FLAG_MSIX_ENA,
+	I40E_FLAG_RSS_ENA,
+	I40E_FLAG_VMDQ_ENA,
+	I40E_FLAG_SRIOV_ENA,
+	I40E_FLAG_DCB_CAPABLE,
+	I40E_FLAG_DCB_ENA,
+	I40E_FLAG_FD_SB_ENA,
+	I40E_FLAG_FD_ATR_ENA,
+	I40E_FLAG_MFP_ENA,
+	I40E_FLAG_HW_ATR_EVICT_ENA,
+	I40E_FLAG_VEB_MODE_ENA,
+	I40E_FLAG_VEB_STATS_ENA,
+	I40E_FLAG_LINK_POLLING_ENA,
+	I40E_FLAG_TRUE_PROMISC_ENA,
+	I40E_FLAG_LEGACY_RX_ENA,
+	I40E_FLAG_PTP_ENA,
+	I40E_FLAG_IWARP_ENA,
+	I40E_FLAG_LINK_DOWN_ON_CLOSE_ENA,
+	I40E_FLAG_SOURCE_PRUNING_DIS,
+	I40E_FLAG_TC_MQPRIO_ENA,
+	I40E_FLAG_FD_SB_INACTIVE,
+	I40E_FLAG_FD_SB_TO_CLOUD_FILTER,
+	I40E_FLAG_FW_LLDP_DIS,
+	I40E_FLAG_RS_FEC,
+	I40E_FLAG_BASE_R_FEC,
+	/* TOTAL_PORT_SHUTDOWN_ENA
+	 * Allows to physically disable the link on the NIC's port.
+	 * If enabled, (after link down request from the OS)
+	 * no link, traffic or led activity is possible on that port.
+	 *
+	 * If I40E_FLAG_TOTAL_PORT_SHUTDOWN_ENA is set, the
+	 * I40E_FLAG_LINK_DOWN_ON_CLOSE_ENA must be explicitly forced
+	 * to true and cannot be disabled by system admin at that time.
+	 * The functionalities are exclusive in terms of configuration, but
+	 * they also have similar behavior (allowing to disable physical
+	 * link of the port), with following differences:
+	 * - LINK_DOWN_ON_CLOSE_ENA is configurable at host OS run-time and
+	 *   is supported by whole family of 7xx Intel Ethernet Controllers
+	 * - TOTAL_PORT_SHUTDOWN_ENA may be enabled only before OS loads
+	 *   (in BIOS) only if motherboard's BIOS and NIC's FW has support of it
+	 * - when LINK_DOWN_ON_CLOSE_ENABLED is used, the link is being brought
+	 *   down by sending phy_type=0 to NIC's FW
+	 * - when TOTAL_PORT_SHUTDOWN_ENA is used, phy_type is not altered,
+	 *   instead the link is being brought down by clearing
+	 *   bit (I40E_AQ_PHY_ENABLE_LINK) in abilities field of
+	 *   i40e_aq_set_phy_config structure
+	 */
+	I40E_FLAG_TOTAL_PORT_SHUTDOWN_ENA,
+	I40E_FLAG_VF_VLAN_PRUNING_ENA,
+	I40E_FLAG_MDD_AUTO_RESET_VF,
+	I40E_PF_FLAGS_NBITS,		/* must be last */
+};
+
 enum i40e_interrupt_policy {
 	I40E_INTERRUPT_BEST_CASE,
 	I40E_INTERRUPT_MEDIUM,
@@ -169,8 +204,7 @@ enum i40e_interrupt_policy {
 
 struct i40e_lump_tracking {
 	u16 num_entries;
-	u16 search_hint;
-	u16 list[0];
+	u16 list[];
 #define I40E_PILE_VALID_BIT  0x8000
 #define I40E_IWARP_IRQ_PILE_ID  (I40E_PILE_VALID_BIT - 2)
 };
@@ -212,14 +246,18 @@ struct i40e_fdir_filter {
 	struct hlist_node fdir_node;
 	/* filter ipnut set */
 	u8 flow_type;
-	u8 ip4_proto;
+	u8 ipl4_proto;
 	/* TX packet view of src and dst */
 	__be32 dst_ip;
 	__be32 src_ip;
+	__be32 dst_ip6[4];
+	__be32 src_ip6[4];
 	__be16 src_port;
 	__be16 dst_port;
 	__be32 sctp_v_tag;
 
+	__be16 vlan_etype;
+	__be16 vlan_tag;
 	/* Flexible data to match within the packet payload */
 	__be16 flex_word;
 	u16 flex_offset;
@@ -236,7 +274,57 @@ struct i40e_fdir_filter {
 	u32 fd_id;
 };
 
-#define I40E_ETH_P_LLDP			0x88cc
+#define I40E_CLOUD_FIELD_OMAC		BIT(0)
+#define I40E_CLOUD_FIELD_IMAC		BIT(1)
+#define I40E_CLOUD_FIELD_IVLAN		BIT(2)
+#define I40E_CLOUD_FIELD_TEN_ID		BIT(3)
+#define I40E_CLOUD_FIELD_IIP		BIT(4)
+
+#define I40E_CLOUD_FILTER_FLAGS_OMAC	I40E_CLOUD_FIELD_OMAC
+#define I40E_CLOUD_FILTER_FLAGS_IMAC	I40E_CLOUD_FIELD_IMAC
+#define I40E_CLOUD_FILTER_FLAGS_IMAC_IVLAN	(I40E_CLOUD_FIELD_IMAC | \
+						 I40E_CLOUD_FIELD_IVLAN)
+#define I40E_CLOUD_FILTER_FLAGS_IMAC_TEN_ID	(I40E_CLOUD_FIELD_IMAC | \
+						 I40E_CLOUD_FIELD_TEN_ID)
+#define I40E_CLOUD_FILTER_FLAGS_OMAC_TEN_ID_IMAC (I40E_CLOUD_FIELD_OMAC | \
+						  I40E_CLOUD_FIELD_IMAC | \
+						  I40E_CLOUD_FIELD_TEN_ID)
+#define I40E_CLOUD_FILTER_FLAGS_IMAC_IVLAN_TEN_ID (I40E_CLOUD_FIELD_IMAC | \
+						   I40E_CLOUD_FIELD_IVLAN | \
+						   I40E_CLOUD_FIELD_TEN_ID)
+#define I40E_CLOUD_FILTER_FLAGS_IIP	I40E_CLOUD_FIELD_IIP
+
+struct i40e_cloud_filter {
+	struct hlist_node cloud_node;
+	unsigned long cookie;
+	/* cloud filter input set follows */
+	u8 dst_mac[ETH_ALEN];
+	u8 src_mac[ETH_ALEN];
+	__be16 vlan_id;
+	u16 seid;       /* filter control */
+	__be16 dst_port;
+	__be16 src_port;
+	u32 tenant_id;
+	union {
+		struct {
+			struct in_addr dst_ip;
+			struct in_addr src_ip;
+		} v4;
+		struct {
+			struct in6_addr dst_ip6;
+			struct in6_addr src_ip6;
+		} v6;
+	} ip;
+#define dst_ipv6	ip.v6.dst_ip6.s6_addr32
+#define src_ipv6	ip.v6.src_ip6.s6_addr32
+#define dst_ipv4	ip.v4.dst_ip.s_addr
+#define src_ipv4	ip.v4.src_ip.s_addr
+	u16 n_proto;    /* Ethernet Protocol */
+	u8 ip_proto;    /* IPPROTO value */
+	u8 flags;
+#define I40E_CLOUD_TNL_TYPE_NONE        0xff
+	u8 tunnel_type;
+};
 
 #define I40E_DCB_PRIO_TYPE_STRICT	0
 #define I40E_DCB_PRIO_TYPE_ETS		1
@@ -255,10 +343,12 @@ struct i40e_tc_configuration {
 	struct i40e_tc_info tc_info[I40E_MAX_TRAFFIC_CLASS];
 };
 
+#define I40E_UDP_PORT_INDEX_UNUSED	255
 struct i40e_udp_port_config {
 	/* AdminQ command interface expects port number in Host byte order */
 	u16 port;
 	u8 type;
+	u8 filter_index;
 };
 
 /* macros related to FLX_PIT */
@@ -275,15 +365,6 @@ struct i40e_udp_port_config {
 					     I40E_FLEX_SET_FSIZE(fsize) | \
 					     I40E_FLEX_SET_SRC_WORD(src))
 
-#define I40E_FLEX_PIT_GET_SRC(flex) (((flex) & \
-				     I40E_PRTQF_FLX_PIT_SOURCE_OFF_MASK) >> \
-				     I40E_PRTQF_FLX_PIT_SOURCE_OFF_SHIFT)
-#define I40E_FLEX_PIT_GET_DST(flex) (((flex) & \
-				     I40E_PRTQF_FLX_PIT_DEST_OFF_MASK) >> \
-				     I40E_PRTQF_FLX_PIT_DEST_OFF_SHIFT)
-#define I40E_FLEX_PIT_GET_FSIZE(flex) (((flex) & \
-				       I40E_PRTQF_FLX_PIT_FSIZE_MASK) >> \
-				       I40E_PRTQF_FLX_PIT_FSIZE_SHIFT)
 
 #define I40E_MAX_FLEX_SRC_OFFSET 0x1F
 
@@ -308,7 +389,6 @@ struct i40e_udp_port_config {
 #define I40E_L4_GLQF_ORT_IDX		35
 
 /* Flex PIT register index */
-#define I40E_FLEX_PIT_IDX_START_L2	0
 #define I40E_FLEX_PIT_IDX_START_L3	3
 #define I40E_FLEX_PIT_IDX_START_L4	6
 
@@ -324,25 +404,78 @@ struct i40e_udp_port_config {
 				 I40E_FLEX_54_MASK | I40E_FLEX_55_MASK | \
 				 I40E_FLEX_56_MASK | I40E_FLEX_57_MASK)
 
+#define I40E_QINT_TQCTL_VAL(qp, vector, nextq_type) \
+	(I40E_QINT_TQCTL_CAUSE_ENA_MASK | \
+	(I40E_TX_ITR << I40E_QINT_TQCTL_ITR_INDX_SHIFT) | \
+	((vector) << I40E_QINT_TQCTL_MSIX_INDX_SHIFT) | \
+	((qp) << I40E_QINT_TQCTL_NEXTQ_INDX_SHIFT) | \
+	(I40E_QUEUE_TYPE_##nextq_type << I40E_QINT_TQCTL_NEXTQ_TYPE_SHIFT))
+
+#define I40E_QINT_RQCTL_VAL(qp, vector, nextq_type) \
+	(I40E_QINT_RQCTL_CAUSE_ENA_MASK | \
+	(I40E_RX_ITR << I40E_QINT_RQCTL_ITR_INDX_SHIFT) | \
+	((vector) << I40E_QINT_RQCTL_MSIX_INDX_SHIFT) | \
+	((qp) << I40E_QINT_RQCTL_NEXTQ_INDX_SHIFT) | \
+	(I40E_QUEUE_TYPE_##nextq_type << I40E_QINT_RQCTL_NEXTQ_TYPE_SHIFT))
+
 struct i40e_flex_pit {
 	struct list_head list;
 	u16 src_offset;
 	u8 pit_index;
 };
 
+struct i40e_fwd_adapter {
+	struct net_device *netdev;
+	int bit_no;
+};
+
+struct i40e_channel {
+	struct list_head list;
+	bool initialized;
+	u8 type;
+	u16 vsi_number; /* Assigned VSI number from AQ 'Add VSI' response */
+	u16 stat_counter_idx;
+	u16 base_queue;
+	u16 num_queue_pairs; /* Requested by user */
+	u16 seid;
+
+	u8 enabled_tc;
+	struct i40e_aqc_vsi_properties_data info;
+
+	u64 max_tx_rate;
+	struct i40e_fwd_adapter *fwd;
+
+	/* track this channel belongs to which VSI */
+	struct i40e_vsi *parent_vsi;
+};
+
+struct i40e_ptp_pins_settings;
+
+static inline bool i40e_is_channel_macvlan(struct i40e_channel *ch)
+{
+	return !!ch->fwd;
+}
+
+static inline const u8 *i40e_channel_mac(struct i40e_channel *ch)
+{
+	if (i40e_is_channel_macvlan(ch))
+		return ch->fwd->netdev->dev_addr;
+	else
+		return NULL;
+}
+
 /* struct that defines the Ethernet device */
 struct i40e_pf {
 	struct pci_dev *pdev;
+	struct devlink_port devlink_port;
 	struct i40e_hw hw;
 	DECLARE_BITMAP(state, __I40E_STATE_SIZE__);
 	struct msix_entry *msix_entries;
-	bool fc_autoneg_status;
 
-	u16 eeprom_version;
 	u16 num_vmdq_vsis;         /* num vmdq vsis this PF has set up */
 	u16 num_vmdq_qps;          /* num queue pairs per vmdq pool */
 	u16 num_vmdq_msix;         /* num queue vectors per vmdq pool */
-	u16 num_req_vfs;           /* num VFs requested for this VF */
+	u16 num_req_vfs;           /* num VFs requested for this PF */
 	u16 num_vf_qps;            /* num queue pairs per VF */
 	u16 num_lan_qps;           /* num lan queues this PF has set up */
 	u16 num_lan_msix;          /* num queue vectors for the base PF vsi */
@@ -354,7 +487,6 @@ struct i40e_pf {
 	u16 rss_size_max;          /* HW defined max RSS queues */
 	u16 fdir_pf_filter_count;  /* num of guaranteed filters for this PF */
 	u16 num_alloc_vsi;         /* num VSIs this driver supports */
-	u8 atr_sample_rate;
 	bool wol_en;
 
 	struct hlist_head fdir_filter_list;
@@ -373,6 +505,11 @@ struct i40e_pf {
 	u16 fd_sctp4_filter_cnt;
 	u16 fd_ip4_filter_cnt;
 
+	u16 fd_tcp6_filter_cnt;
+	u16 fd_udp6_filter_cnt;
+	u16 fd_sctp6_filter_cnt;
+	u16 fd_ip6_filter_cnt;
+
 	/* Flexible filter table values that need to be programmed into
 	 * hardware, which expects L3 and L4 to be programmed separately. We
 	 * need to ensure that the values are in ascended order and don't have
@@ -381,67 +518,22 @@ struct i40e_pf {
 	struct list_head l3_flex_pit_list;
 	struct list_head l4_flex_pit_list;
 
-	struct i40e_udp_port_config udp_ports[I40E_MAX_PF_UDP_OFFLOAD_PORTS];
-	u16 pending_udp_bitmap;
+	struct udp_tunnel_nic_shared udp_tunnel_shared;
+	struct udp_tunnel_nic_info udp_tunnel_nic;
+
+	struct hlist_head cloud_filter_list;
+	u16 num_cloud_filters;
 
-	enum i40e_interrupt_policy int_policy;
 	u16 rx_itr_default;
 	u16 tx_itr_default;
 	u32 msg_enable;
 	char int_name[I40E_INT_NAME_STR_LEN];
-	u16 adminq_work_limit; /* num of admin receive queue desc to process */
 	unsigned long service_timer_period;
 	unsigned long service_timer_previous;
 	struct timer_list service_timer;
 	struct work_struct service_task;
 
-	u64 flags;
-#define I40E_FLAG_RX_CSUM_ENABLED		BIT_ULL(1)
-#define I40E_FLAG_MSI_ENABLED			BIT_ULL(2)
-#define I40E_FLAG_MSIX_ENABLED			BIT_ULL(3)
-#define I40E_FLAG_RSS_ENABLED			BIT_ULL(6)
-#define I40E_FLAG_VMDQ_ENABLED			BIT_ULL(7)
-#define I40E_FLAG_IWARP_ENABLED			BIT_ULL(10)
-#define I40E_FLAG_FILTER_SYNC			BIT_ULL(15)
-#define I40E_FLAG_SERVICE_CLIENT_REQUESTED	BIT_ULL(16)
-#define I40E_FLAG_SRIOV_ENABLED			BIT_ULL(19)
-#define I40E_FLAG_DCB_ENABLED			BIT_ULL(20)
-#define I40E_FLAG_FD_SB_ENABLED			BIT_ULL(21)
-#define I40E_FLAG_FD_ATR_ENABLED		BIT_ULL(22)
-#define I40E_FLAG_FD_SB_AUTO_DISABLED		BIT_ULL(23)
-#define I40E_FLAG_FD_ATR_AUTO_DISABLED		BIT_ULL(24)
-#define I40E_FLAG_PTP				BIT_ULL(25)
-#define I40E_FLAG_MFP_ENABLED			BIT_ULL(26)
-#define I40E_FLAG_UDP_FILTER_SYNC		BIT_ULL(27)
-#define I40E_FLAG_PORT_ID_VALID			BIT_ULL(28)
-#define I40E_FLAG_DCB_CAPABLE			BIT_ULL(29)
-#define I40E_FLAG_RSS_AQ_CAPABLE		BIT_ULL(31)
-#define I40E_FLAG_HW_ATR_EVICT_CAPABLE		BIT_ULL(32)
-#define I40E_FLAG_OUTER_UDP_CSUM_CAPABLE	BIT_ULL(33)
-#define I40E_FLAG_128_QP_RSS_CAPABLE		BIT_ULL(34)
-#define I40E_FLAG_WB_ON_ITR_CAPABLE		BIT_ULL(35)
-#define I40E_FLAG_VEB_STATS_ENABLED		BIT_ULL(37)
-#define I40E_FLAG_MULTIPLE_TCP_UDP_RSS_PCTYPE	BIT_ULL(38)
-#define I40E_FLAG_LINK_POLLING_ENABLED		BIT_ULL(39)
-#define I40E_FLAG_VEB_MODE_ENABLED		BIT_ULL(40)
-#define I40E_FLAG_GENEVE_OFFLOAD_CAPABLE	BIT_ULL(41)
-#define I40E_FLAG_NO_PCI_LINK_CHECK		BIT_ULL(42)
-#define I40E_FLAG_100M_SGMII_CAPABLE		BIT_ULL(43)
-#define I40E_FLAG_RESTART_AUTONEG		BIT_ULL(44)
-#define I40E_FLAG_NO_DCB_SUPPORT		BIT_ULL(45)
-#define I40E_FLAG_USE_SET_LLDP_MIB		BIT_ULL(46)
-#define I40E_FLAG_STOP_FW_LLDP			BIT_ULL(47)
-#define I40E_FLAG_PHY_CONTROLS_LEDS		BIT_ULL(48)
-#define I40E_FLAG_PF_MAC			BIT_ULL(50)
-#define I40E_FLAG_TRUE_PROMISC_SUPPORT		BIT_ULL(51)
-#define I40E_FLAG_HAVE_CRT_RETIMER		BIT_ULL(52)
-#define I40E_FLAG_PTP_L4_CAPABLE		BIT_ULL(53)
-#define I40E_FLAG_CLIENT_RESET			BIT_ULL(54)
-#define I40E_FLAG_TEMP_LINK_POLLING		BIT_ULL(55)
-#define I40E_FLAG_CLIENT_L2_CHANGE		BIT_ULL(56)
-#define I40E_FLAG_WOL_MC_MAGIC_PKT_WAKE		BIT_ULL(57)
-#define I40E_FLAG_LEGACY_RX			BIT_ULL(58)
-
+	DECLARE_BITMAP(flags, I40E_PF_FLAGS_NBITS);
 	struct i40e_client_instance *cinst;
 	bool stat_offsets_loaded;
 	struct i40e_hw_port_stats stats;
@@ -449,7 +541,6 @@ struct i40e_pf {
 	u32 tx_timeout_count;
 	u32 tx_timeout_recovery_level;
 	unsigned long tx_timeout_last_recovery;
-	u32 tx_sluggish_count;
 	u32 hw_csum_rx_error;
 	u32 led_status;
 	u16 corer_count; /* Core reset count */
@@ -457,6 +548,7 @@ struct i40e_pf {
 	u16 empr_count; /* EMP reset count */
 	u16 pfr_count; /* PF reset count */
 	u16 sw_int_count; /* SW interrupt count */
+	u32 link_down_events;
 
 	struct mutex switch_mutex;
 	u16 lan_vsi;       /* our default LAN VSI */
@@ -471,23 +563,19 @@ struct i40e_pf {
 	struct i40e_lump_tracking *irq_pile;
 
 	/* switch config info */
-	u16 pf_seid;
 	u16 main_vsi_seid;
 	u16 mac_seid;
-	struct kobject *switch_kobj;
 #ifdef CONFIG_DEBUG_FS
 	struct dentry *i40e_dbg_pf;
 #endif /* CONFIG_DEBUG_FS */
 	bool cur_promisc;
 
-	u16 instance; /* A unique number per i40e_pf instance in the system */
-
 	/* sr-iov config info */
 	struct i40e_vf *vf;
 	int num_alloc_vfs;	/* actual number of VFs allocated */
 	u32 vf_aq_requests;
 	u32 arq_overflows;	/* Not fatal, possibly indicative of problems */
-
+	struct ratelimit_state mdd_message_rate_limit;
 	/* DCBx/DCBNL capability for PF that indicates
 	 * whether DCBx is managed by firmware or host
 	 * based agent (LLDPAD). Also, indicates what
@@ -498,14 +586,87 @@ struct i40e_pf {
 	u16 dcbx_cap;
 
 	struct i40e_filter_control_settings filter_settings;
+	struct i40e_rx_pb_config pb_cfg; /* Current Rx packet buffer config */
+	struct i40e_dcbx_config tmp_cfg;
+
+/* GPIO defines used by PTP */
+#define I40E_SDP3_2			18
+#define I40E_SDP3_3			19
+#define I40E_GPIO_4			20
+#define I40E_LED2_0			26
+#define I40E_LED2_1			27
+#define I40E_LED3_0			28
+#define I40E_LED3_1			29
+#define I40E_GLGEN_GPIO_SET_SDP_DATA_HI \
+	(1 << I40E_GLGEN_GPIO_SET_SDP_DATA_SHIFT)
+#define I40E_GLGEN_GPIO_SET_DRV_SDP_DATA \
+	(1 << I40E_GLGEN_GPIO_SET_DRIVE_SDP_SHIFT)
+#define I40E_GLGEN_GPIO_CTL_PRT_NUM_0 \
+	(0 << I40E_GLGEN_GPIO_CTL_PRT_NUM_SHIFT)
+#define I40E_GLGEN_GPIO_CTL_PRT_NUM_1 \
+	(1 << I40E_GLGEN_GPIO_CTL_PRT_NUM_SHIFT)
+#define I40E_GLGEN_GPIO_CTL_RESERVED	BIT(2)
+#define I40E_GLGEN_GPIO_CTL_PRT_NUM_NA_Z \
+	(1 << I40E_GLGEN_GPIO_CTL_PRT_NUM_NA_SHIFT)
+#define I40E_GLGEN_GPIO_CTL_DIR_OUT \
+	(1 << I40E_GLGEN_GPIO_CTL_PIN_DIR_SHIFT)
+#define I40E_GLGEN_GPIO_CTL_TRI_DRV_HI \
+	(1 << I40E_GLGEN_GPIO_CTL_TRI_CTL_SHIFT)
+#define I40E_GLGEN_GPIO_CTL_OUT_HI_RST \
+	(1 << I40E_GLGEN_GPIO_CTL_OUT_CTL_SHIFT)
+#define I40E_GLGEN_GPIO_CTL_TIMESYNC_0 \
+	(3 << I40E_GLGEN_GPIO_CTL_PIN_FUNC_SHIFT)
+#define I40E_GLGEN_GPIO_CTL_TIMESYNC_1 \
+	(4 << I40E_GLGEN_GPIO_CTL_PIN_FUNC_SHIFT)
+#define I40E_GLGEN_GPIO_CTL_NOT_FOR_PHY_CONN \
+	(0x3F << I40E_GLGEN_GPIO_CTL_PHY_PIN_NAME_SHIFT)
+#define I40E_GLGEN_GPIO_CTL_OUT_DEFAULT \
+	(1 << I40E_GLGEN_GPIO_CTL_OUT_DEFAULT_SHIFT)
+#define I40E_GLGEN_GPIO_CTL_PORT_0_IN_TIMESYNC_0 \
+	(I40E_GLGEN_GPIO_CTL_NOT_FOR_PHY_CONN | \
+	 I40E_GLGEN_GPIO_CTL_TIMESYNC_0 | \
+	 I40E_GLGEN_GPIO_CTL_RESERVED | I40E_GLGEN_GPIO_CTL_PRT_NUM_0)
+#define I40E_GLGEN_GPIO_CTL_PORT_1_IN_TIMESYNC_0 \
+	(I40E_GLGEN_GPIO_CTL_NOT_FOR_PHY_CONN | \
+	 I40E_GLGEN_GPIO_CTL_TIMESYNC_0 | \
+	 I40E_GLGEN_GPIO_CTL_RESERVED | I40E_GLGEN_GPIO_CTL_PRT_NUM_1)
+#define I40E_GLGEN_GPIO_CTL_PORT_0_OUT_TIMESYNC_1 \
+	(I40E_GLGEN_GPIO_CTL_NOT_FOR_PHY_CONN | \
+	 I40E_GLGEN_GPIO_CTL_TIMESYNC_1 | I40E_GLGEN_GPIO_CTL_OUT_HI_RST | \
+	 I40E_GLGEN_GPIO_CTL_TRI_DRV_HI | I40E_GLGEN_GPIO_CTL_DIR_OUT | \
+	 I40E_GLGEN_GPIO_CTL_RESERVED | I40E_GLGEN_GPIO_CTL_PRT_NUM_0)
+#define I40E_GLGEN_GPIO_CTL_PORT_1_OUT_TIMESYNC_1 \
+	(I40E_GLGEN_GPIO_CTL_NOT_FOR_PHY_CONN | \
+	 I40E_GLGEN_GPIO_CTL_TIMESYNC_1 | I40E_GLGEN_GPIO_CTL_OUT_HI_RST | \
+	 I40E_GLGEN_GPIO_CTL_TRI_DRV_HI | I40E_GLGEN_GPIO_CTL_DIR_OUT | \
+	 I40E_GLGEN_GPIO_CTL_RESERVED | I40E_GLGEN_GPIO_CTL_PRT_NUM_1)
+#define I40E_GLGEN_GPIO_CTL_LED_INIT \
+	(I40E_GLGEN_GPIO_CTL_PRT_NUM_NA_Z | \
+	 I40E_GLGEN_GPIO_CTL_DIR_OUT | \
+	 I40E_GLGEN_GPIO_CTL_TRI_DRV_HI | \
+	 I40E_GLGEN_GPIO_CTL_OUT_HI_RST | \
+	 I40E_GLGEN_GPIO_CTL_OUT_DEFAULT | \
+	 I40E_GLGEN_GPIO_CTL_NOT_FOR_PHY_CONN)
+#define I40E_PRTTSYN_AUX_1_INSTNT \
+	(1 << I40E_PRTTSYN_AUX_1_INSTNT_SHIFT)
+#define I40E_PRTTSYN_AUX_0_OUT_ENABLE \
+	(1 << I40E_PRTTSYN_AUX_0_OUT_ENA_SHIFT)
+#define I40E_PRTTSYN_AUX_0_OUT_CLK_MOD	(3 << I40E_PRTTSYN_AUX_0_OUTMOD_SHIFT)
+#define I40E_PRTTSYN_AUX_0_OUT_ENABLE_CLK_MOD \
+	(I40E_PRTTSYN_AUX_0_OUT_ENABLE | I40E_PRTTSYN_AUX_0_OUT_CLK_MOD)
+#define I40E_PTP_HALF_SECOND		500000000LL /* nano seconds */
+#define I40E_PTP_2_SEC_DELAY		2
 
 	struct ptp_clock *ptp_clock;
 	struct ptp_clock_info ptp_caps;
 	struct sk_buff *ptp_tx_skb;
 	unsigned long ptp_tx_start;
-	struct hwtstamp_config tstamp_config;
+	struct kernel_hwtstamp_config tstamp_config;
+	struct timespec64 ptp_prev_hw_time;
+	struct work_struct ptp_extts0_work;
+	ktime_t ptp_reset_start;
 	struct mutex tmreg_lock; /* Used to protect the SYSTIME registers. */
-	u64 ptp_base_adj;
+	u32 ptp_adj_mult;
 	u32 tx_hwtstamp_timeouts;
 	u32 tx_hwtstamp_skipped;
 	u32 rx_hwtstamp_cleared;
@@ -514,18 +675,71 @@ struct i40e_pf {
 	unsigned long latch_events[4];
 	bool ptp_tx;
 	bool ptp_rx;
+	struct i40e_ptp_pins_settings *ptp_pins;
 	u16 rss_table_size; /* HW RSS table size */
-	/* These are only valid in NPAR modes */
-	u32 npar_max_bw;
-	u32 npar_min_bw;
+	u32 max_bw;
+	u32 min_bw;
 
 	u32 ioremap_len;
 	u32 fd_inv;
 	u16 phy_led_val;
+
+	u16 last_sw_conf_flags;
+	u16 last_sw_conf_valid_flags;
+	/* List to keep previous DDP profiles to be rolled back in the future */
+	struct list_head ddp_old_prof;
 };
 
 /**
- * i40e_mac_to_hkey - Convert a 6-byte MAC Address to a u64 hash key
+ * __i40e_pf_next_vsi - get next valid VSI
+ * @pf: pointer to the PF struct
+ * @idx: pointer to start position number
+ *
+ * Find and return next non-NULL VSI pointer in pf->vsi array and
+ * updates idx position. Returns NULL if no VSI is found.
+ **/
+static __always_inline struct i40e_vsi *
+__i40e_pf_next_vsi(struct i40e_pf *pf, int *idx)
+{
+	while (*idx < pf->num_alloc_vsi) {
+		if (pf->vsi[*idx])
+			return pf->vsi[*idx];
+		(*idx)++;
+	}
+	return NULL;
+}
+
+#define i40e_pf_for_each_vsi(_pf, _i, _vsi)			\
+	for (_i = 0, _vsi = __i40e_pf_next_vsi(_pf, &_i);	\
+	     _vsi;						\
+	     _i++, _vsi = __i40e_pf_next_vsi(_pf, &_i))
+
+/**
+ * __i40e_pf_next_veb - get next valid VEB
+ * @pf: pointer to the PF struct
+ * @idx: pointer to start position number
+ *
+ * Find and return next non-NULL VEB pointer in pf->veb array and
+ * updates idx position. Returns NULL if no VEB is found.
+ **/
+static __always_inline struct i40e_veb *
+__i40e_pf_next_veb(struct i40e_pf *pf, int *idx)
+{
+	while (*idx < I40E_MAX_VEB) {
+		if (pf->veb[*idx])
+			return pf->veb[*idx];
+		(*idx)++;
+	}
+	return NULL;
+}
+
+#define i40e_pf_for_each_veb(_pf, _i, _veb)			\
+	for (_i = 0, _veb = __i40e_pf_next_veb(_pf, &_i);	\
+	     _veb;						\
+	     _i++, _veb = __i40e_pf_next_veb(_pf, &_i))
+
+/**
+ * i40e_addr_to_hkey - Convert a 6-byte MAC Address to a u64 hash key
  * @macaddr: the MAC Address as the base key
  *
  * Simply copies the address and returns it as a u64 for hashing
@@ -544,6 +758,7 @@ enum i40e_filter_state {
 	I40E_FILTER_ACTIVE,		/* Added to switch by FW */
 	I40E_FILTER_FAILED,		/* Rejected by FW */
 	I40E_FILTER_REMOVE,		/* To be removed */
+	I40E_FILTER_NEW_SYNC,		/* New, not sent yet, is in i40e_sync_vsi_filters() */
 /* There is no 'removed' state; the filter struct is freed */
 };
 struct i40e_mac_filter {
@@ -573,13 +788,11 @@ struct i40e_new_mac_filter {
 struct i40e_veb {
 	struct i40e_pf *pf;
 	u16 idx;
-	u16 veb_idx;		/* index of VEB parent */
 	u16 seid;
 	u16 uplink_seid;
 	u16 stats_idx;		/* index of VEB parent */
 	u8  enabled_tc;
 	u16 bridge_mode;	/* Bridge Mode (VEB/VEPA) */
-	u16 flags;
 	u16 bw_limit;
 	u8  bw_max_quanta;
 	bool is_abs_credits;
@@ -618,16 +831,22 @@ struct i40e_vsi {
 	struct rtnl_link_stats64 net_stats_offsets;
 	struct i40e_eth_stats eth_stats;
 	struct i40e_eth_stats eth_stats_offsets;
-	u32 tx_restart;
-	u32 tx_busy;
+	u64 tx_restart;
+	u64 tx_busy;
 	u64 tx_linearize;
 	u64 tx_force_wb;
-	u32 rx_buf_failed;
-	u32 rx_page_failed;
+	u64 tx_stopped;
+	u64 rx_buf_failed;
+	u64 rx_page_failed;
+	u64 rx_page_reuse;
+	u64 rx_page_alloc;
+	u64 rx_page_waive;
+	u64 rx_page_busy;
 
 	/* These are containers of ring pointers, allocated at run-time */
 	struct i40e_ring **rx_rings;
 	struct i40e_ring **tx_rings;
+	struct i40e_ring **xdp_rings; /* XDP Tx rings */
 
 	u32  active_filters;
 	u32  promisc_threshold;
@@ -644,6 +863,8 @@ struct i40e_vsi {
 	u16 max_frame;
 	u16 rx_buf_len;
 
+	struct bpf_prog *xdp_prog;
+
 	/* List of q_vectors allocated to this VSI */
 	struct i40e_q_vector **q_vectors;
 	int num_q_vectors;
@@ -658,10 +879,12 @@ struct i40e_vsi {
 	u16 alloc_queue_pairs;	/* Allocated Tx/Rx queues */
 	u16 req_queue_pairs;	/* User requested queue pairs */
 	u16 num_queue_pairs;	/* Used tx and rx pairs */
-	u16 num_desc;
+	u16 num_tx_desc;
+	u16 num_rx_desc;
 	enum i40e_vsi_type type;  /* VSI type, e.g., LAN, FCoE, etc */
 	s16 vf_id;		/* Virtual function ID for SRIOV VSIs */
 
+	struct tc_mqprio_qopt_offload mqprio_qopt; /* queue parameters */
 	struct i40e_tc_configuration tc_config;
 	struct i40e_aqc_vsi_properties_data info;
 
@@ -683,16 +906,38 @@ struct i40e_vsi {
 	bool current_isup;	/* Sync 'link up' logging */
 	enum i40e_aq_link_speed current_speed;	/* Sync link speed logging */
 
+	/* channel specific fields */
+	u16 cnt_q_avail;	/* num of queues available for channel usage */
+	u16 orig_rss_size;
+	u16 current_rss_size;
+	bool reconfig_rss;
+
+	u16 next_base_queue;	/* next queue to be used for channel setup */
+
+	struct list_head ch_list;
+	u16 tc_seid_map[I40E_MAX_TRAFFIC_CLASS];
+
+	/* macvlan fields */
+#define I40E_MAX_MACVLANS		128 /* Max HW vectors - 1 on FVL */
+#define I40E_MIN_MACVLAN_VECTORS	2   /* Min vectors to enable macvlans */
+	DECLARE_BITMAP(fwd_bitmask, I40E_MAX_MACVLANS);
+	struct list_head macvlan_list;
+	int macvlan_cnt;
+
 	void *priv;	/* client driver data reference. */
 
 	/* VSI specific handlers */
 	irqreturn_t (*irq_handler)(int irq, void *data);
+
+	unsigned long *af_xdp_zc_qps; /* tracks AF_XDP ZC enabled qps */
 } ____cacheline_internodealigned_in_smp;
 
 struct i40e_netdev_priv {
 	struct i40e_vsi *vsi;
 };
 
+extern struct ida i40e_client_ida;
+
 /* struct that defines an interrupt vector */
 struct i40e_q_vector {
 	struct i40e_vsi *vsi;
@@ -701,20 +946,21 @@ struct i40e_q_vector {
 	u16 reg_idx;		/* register index of the interrupt */
 
 	struct napi_struct napi;
+	struct rcu_head rcu;	/* to avoid race with update stats on free */
 
 	struct i40e_ring_container rx;
 	struct i40e_ring_container tx;
 
+	u8 itr_countdown;	/* when 0 should adjust adaptive ITR */
 	u8 num_ringpairs;	/* total number of ring pairs in vector */
 
 	cpumask_t affinity_mask;
 	struct irq_affinity_notify affinity_notify;
 
-	struct rcu_head rcu;	/* to avoid race with update stats on free */
 	char name[I40E_INT_NAME_STR_LEN];
 	bool arm_wb_state;
-#define ITR_COUNTDOWN_START 100
-	u8 itr_countdown;	/* when 0 should adjust ITR */
+	bool in_busy_poll;
+	int irq_num;		/* IRQ assigned to this q_vector */
 } ____cacheline_internodealigned_in_smp;
 
 /* lan device */
@@ -724,29 +970,104 @@ struct i40e_device {
 };
 
 /**
+ * i40e_info_nvm_ver - format the NVM version string
+ * @hw: ptr to the hardware info
+ * @buf: string buffer to store
+ * @len: buffer size
+ *
+ * Formats NVM version string as:
+ * <gen>.<snap>.<release> when eetrackid == I40E_OEM_EETRACK_ID
+ * <nvm_major>.<nvm_minor> otherwise
+ **/
+static inline void i40e_info_nvm_ver(struct i40e_hw *hw, char *buf, size_t len)
+{
+	struct i40e_nvm_info *nvm = &hw->nvm;
+
+	if (nvm->eetrack == I40E_OEM_EETRACK_ID) {
+		u32 full_ver = nvm->oem_ver;
+		u8 gen, snap;
+		u16 release;
+
+		gen = FIELD_GET(I40E_OEM_GEN_MASK, full_ver);
+		snap = FIELD_GET(I40E_OEM_SNAP_MASK, full_ver);
+		release = FIELD_GET(I40E_OEM_RELEASE_MASK, full_ver);
+		snprintf(buf, len, "%x.%x.%x", gen, snap, release);
+	} else {
+		u8 major, minor;
+
+		major = FIELD_GET(I40E_NVM_VERSION_HI_MASK, nvm->version);
+		minor = FIELD_GET(I40E_NVM_VERSION_LO_MASK, nvm->version);
+		snprintf(buf, len, "%x.%02x", major, minor);
+	}
+}
+
+/**
+ * i40e_info_eetrack - format the EETrackID string
+ * @hw: ptr to the hardware info
+ * @buf: string buffer to store
+ * @len: buffer size
+ *
+ * Returns hexadecimally formated EETrackID if it is
+ * different from I40E_OEM_EETRACK_ID or empty string.
+ **/
+static inline void i40e_info_eetrack(struct i40e_hw *hw, char *buf, size_t len)
+{
+	struct i40e_nvm_info *nvm = &hw->nvm;
+
+	buf[0] = '\0';
+	if (nvm->eetrack != I40E_OEM_EETRACK_ID)
+		snprintf(buf, len, "0x%08x", nvm->eetrack);
+}
+
+/**
+ * i40e_info_civd_ver - format the NVM version strings
+ * @hw: ptr to the hardware info
+ * @buf: string buffer to store
+ * @len: buffer size
+ *
+ * Returns formated combo image version if adapter's EETrackID is
+ * different from I40E_OEM_EETRACK_ID or empty string.
+ **/
+static inline void i40e_info_civd_ver(struct i40e_hw *hw, char *buf, size_t len)
+{
+	struct i40e_nvm_info *nvm = &hw->nvm;
+
+	buf[0] = '\0';
+	if (nvm->eetrack != I40E_OEM_EETRACK_ID) {
+		u32 full_ver = nvm->oem_ver;
+		u8 major, minor;
+		u16 build;
+
+		major = FIELD_GET(I40E_OEM_VER_MASK, full_ver);
+		build = FIELD_GET(I40E_OEM_VER_BUILD_MASK, full_ver);
+		minor = FIELD_GET(I40E_OEM_VER_PATCH_MASK, full_ver);
+		snprintf(buf, len, "%d.%d.%d", major, build, minor);
+	}
+}
+
+/**
  * i40e_nvm_version_str - format the NVM version strings
  * @hw: ptr to the hardware info
+ * @buf: string buffer to store
+ * @len: buffer size
  **/
-static inline char *i40e_nvm_version_str(struct i40e_hw *hw)
+static inline char *i40e_nvm_version_str(struct i40e_hw *hw, char *buf,
+					 size_t len)
 {
-	static char buf[32];
-	u32 full_ver;
-	u8 ver, patch;
-	u16 build;
-
-	full_ver = hw->nvm.oem_ver;
-	ver = (u8)(full_ver >> I40E_OEM_VER_SHIFT);
-	build = (u16)((full_ver >> I40E_OEM_VER_BUILD_SHIFT) &
-		 I40E_OEM_VER_BUILD_MASK);
-	patch = (u8)(full_ver & I40E_OEM_VER_PATCH_MASK);
-
-	snprintf(buf, sizeof(buf),
-		 "%x.%02x 0x%x %d.%d.%d",
-		 (hw->nvm.version & I40E_NVM_VERSION_HI_MASK) >>
-			I40E_NVM_VERSION_HI_SHIFT,
-		 (hw->nvm.version & I40E_NVM_VERSION_LO_MASK) >>
-			I40E_NVM_VERSION_LO_SHIFT,
-		 hw->nvm.eetrack, ver, build, patch);
+	char ver[16] = " ";
+
+	/* Get NVM version */
+	i40e_info_nvm_ver(hw, buf, len);
+
+	/* Append EETrackID if provided */
+	i40e_info_eetrack(hw, &ver[1], sizeof(ver) - 1);
+	if (strlen(ver) > 1)
+		strlcat(buf, ver, len);
+
+	/* Append combo image version if provided */
+	i40e_info_civd_ver(hw, &ver[1], sizeof(ver) - 1);
+	if (strlen(ver) > 1)
+		strlcat(buf, ver, len);
 
 	return buf;
 }
@@ -817,11 +1138,25 @@ static inline void i40e_write_fd_input_set(struct i40e_pf *pf,
 			  (u32)(val & 0xFFFFFFFFULL));
 }
 
+/**
+ * i40e_get_pf_count - get PCI PF count.
+ * @hw: pointer to a hw.
+ *
+ * Reports the function number of the highest PCI physical
+ * function plus 1 as it is loaded from the NVM.
+ *
+ * Return: PCI PF count.
+ **/
+static inline u32 i40e_get_pf_count(struct i40e_hw *hw)
+{
+	return FIELD_GET(I40E_GLGEN_PCIFCNCNT_PCIPFCNT_MASK,
+			 rd32(hw, I40E_GLGEN_PCIFCNCNT));
+}
+
 /* needed by i40e_ethtool.c */
 int i40e_up(struct i40e_vsi *vsi);
 void i40e_down(struct i40e_vsi *vsi);
 extern const char i40e_driver_name[];
-extern const char i40e_driver_version_str[];
 void i40e_do_reset_safe(struct i40e_pf *pf, u32 reset_flags);
 void i40e_do_reset(struct i40e_pf *pf, u32 reset_flags, bool lock_acquired);
 int i40e_config_rss(struct i40e_vsi *vsi, u8 *seed, u8 *lut, u16 lut_size);
@@ -837,18 +1172,17 @@ struct i40e_vsi *i40e_find_vsi_from_id(struct i40e_pf *pf, u16 id);
 static inline struct i40e_vsi *
 i40e_find_vsi_by_type(struct i40e_pf *pf, u16 type)
 {
+	struct i40e_vsi *vsi;
 	int i;
 
-	for (i = 0; i < pf->num_alloc_vsi; i++) {
-		struct i40e_vsi *vsi = pf->vsi[i];
-
-		if (vsi && vsi->type == type)
+	i40e_pf_for_each_vsi(pf, i, vsi)
+		if (vsi->type == type)
 			return vsi;
-	}
 
 	return NULL;
 }
 void i40e_update_stats(struct i40e_vsi *vsi);
+void i40e_update_veb_stats(struct i40e_veb *veb);
 void i40e_update_eth_stats(struct i40e_vsi *vsi);
 struct rtnl_link_stats64 *i40e_get_vsi_stats_struct(struct i40e_vsi *vsi);
 int i40e_fetch_switch_configuration(struct i40e_pf *pf,
@@ -858,7 +1192,6 @@ int i40e_add_del_fdir(struct i40e_vsi *vsi,
 		      struct i40e_fdir_filter *input, bool add);
 void i40e_fdir_check_and_reenable(struct i40e_pf *pf);
 u32 i40e_get_current_fd_count(struct i40e_pf *pf);
-u32 i40e_get_cur_guaranteed_fd_count(struct i40e_pf *pf);
 u32 i40e_get_current_atr_cnt(struct i40e_pf *pf);
 u32 i40e_get_global_fd_count(struct i40e_pf *pf);
 bool i40e_set_ntuple(struct i40e_pf *pf, netdev_features_t features);
@@ -866,7 +1199,6 @@ void i40e_set_ethtool_ops(struct net_device *netdev);
 struct i40e_mac_filter *i40e_add_filter(struct i40e_vsi *vsi,
 					const u8 *macaddr, s16 vlan);
 void __i40e_del_filter(struct i40e_vsi *vsi, struct i40e_mac_filter *f);
-void i40e_del_filter(struct i40e_vsi *vsi, const u8 *macaddr, s16 vlan);
 int i40e_sync_vsi_filters(struct i40e_vsi *vsi);
 struct i40e_vsi *i40e_vsi_setup(struct i40e_pf *pf, u8 type,
 				u16 uplink, u32 param1);
@@ -875,12 +1207,15 @@ void i40e_service_event_schedule(struct i40e_pf *pf);
 void i40e_notify_client_of_vf_msg(struct i40e_vsi *vsi, u32 vf_id,
 				  u8 *msg, u16 len);
 
+int i40e_control_wait_tx_q(int seid, struct i40e_pf *pf, int pf_q, bool is_xdp,
+			   bool enable);
+int i40e_control_wait_rx_q(struct i40e_pf *pf, int pf_q, bool enable);
 int i40e_vsi_start_rings(struct i40e_vsi *vsi);
 void i40e_vsi_stop_rings(struct i40e_vsi *vsi);
 void i40e_vsi_stop_rings_no_wait(struct  i40e_vsi *vsi);
 int i40e_vsi_wait_queues_disabled(struct i40e_vsi *vsi);
 int i40e_reconfig_rss_queues(struct i40e_pf *pf, int queue_count);
-struct i40e_veb *i40e_veb_setup(struct i40e_pf *pf, u16 flags, u16 uplink_seid,
+struct i40e_veb *i40e_veb_setup(struct i40e_pf *pf, u16 uplink_seid,
 				u16 downlink_seid, u8 enabled_tc);
 void i40e_veb_release(struct i40e_veb *veb);
 
@@ -904,10 +1239,11 @@ static inline void i40e_dbg_exit(void) {}
 int i40e_lan_add_device(struct i40e_pf *pf);
 int i40e_lan_del_device(struct i40e_pf *pf);
 void i40e_client_subtask(struct i40e_pf *pf);
-void i40e_notify_client_of_l2_param_changes(struct i40e_vsi *vsi);
-void i40e_notify_client_of_netdev_close(struct i40e_vsi *vsi, bool reset);
+void i40e_notify_client_of_l2_param_changes(struct i40e_pf *pf);
+void i40e_notify_client_of_netdev_close(struct i40e_pf *pf, bool reset);
 void i40e_notify_client_of_vf_enable(struct i40e_pf *pf, u32 num_vfs);
 void i40e_notify_client_of_vf_reset(struct i40e_pf *pf, u32 vf_id);
+void i40e_client_update_msix_info(struct i40e_pf *pf);
 int i40e_vf_client_capable(struct i40e_pf *pf, u32 vf_id);
 /**
  * i40e_irq_dynamic_enable - Enable default interrupt generation settings
@@ -920,9 +1256,6 @@ static inline void i40e_irq_dynamic_enable(struct i40e_vsi *vsi, int vector)
 	struct i40e_hw *hw = &pf->hw;
 	u32 val;
 
-	/* definitely clear the PBA here, as this function is meant to
-	 * clean out all previous interrupts AND enable the interrupt
-	 */
 	val = I40E_PFINT_DYN_CTLN_INTENA_MASK |
 	      I40E_PFINT_DYN_CTLN_CLEARPBA_MASK |
 	      (I40E_ITR_NONE << I40E_PFINT_DYN_CTLN_ITR_INDX_SHIFT);
@@ -931,7 +1264,7 @@ static inline void i40e_irq_dynamic_enable(struct i40e_vsi *vsi, int vector)
 }
 
 void i40e_irq_dynamic_disable_icr0(struct i40e_pf *pf);
-void i40e_irq_dynamic_enable_icr0(struct i40e_pf *pf, bool clearpba);
+void i40e_irq_dynamic_enable_icr0(struct i40e_pf *pf);
 int i40e_ioctl(struct net_device *netdev, struct ifreq *ifr, int cmd);
 int i40e_open(struct net_device *netdev);
 int i40e_close(struct net_device *netdev);
@@ -945,8 +1278,15 @@ struct i40e_mac_filter *i40e_add_mac_filter(struct i40e_vsi *vsi,
 					    const u8 *macaddr);
 int i40e_del_mac_filter(struct i40e_vsi *vsi, const u8 *macaddr);
 bool i40e_is_vsi_in_vlan(struct i40e_vsi *vsi);
+int i40e_count_all_filters(struct i40e_vsi *vsi);
+int i40e_count_active_filters(struct i40e_vsi *vsi);
 struct i40e_mac_filter *i40e_find_mac(struct i40e_vsi *vsi, const u8 *macaddr);
 void i40e_vlan_stripping_enable(struct i40e_vsi *vsi);
+static inline bool i40e_is_sw_dcb(struct i40e_pf *pf)
+{
+	return test_bit(I40E_FLAG_FW_LLDP_DIS, pf->flags);
+}
+
 #ifdef CONFIG_I40E_DCB
 void i40e_dcbnl_flush_apps(struct i40e_pf *pf,
 			   struct i40e_dcbx_config *old_cfg,
@@ -956,19 +1296,126 @@ void i40e_dcbnl_setup(struct i40e_vsi *vsi);
 bool i40e_dcb_need_reconfig(struct i40e_pf *pf,
 			    struct i40e_dcbx_config *old_cfg,
 			    struct i40e_dcbx_config *new_cfg);
+int i40e_hw_dcb_config(struct i40e_pf *pf, struct i40e_dcbx_config *new_cfg);
+int i40e_dcb_sw_default_config(struct i40e_pf *pf);
 #endif /* CONFIG_I40E_DCB */
 void i40e_ptp_rx_hang(struct i40e_pf *pf);
 void i40e_ptp_tx_hang(struct i40e_pf *pf);
 void i40e_ptp_tx_hwtstamp(struct i40e_pf *pf);
 void i40e_ptp_rx_hwtstamp(struct i40e_pf *pf, struct sk_buff *skb, u8 index);
 void i40e_ptp_set_increment(struct i40e_pf *pf);
-int i40e_ptp_set_ts_config(struct i40e_pf *pf, struct ifreq *ifr);
-int i40e_ptp_get_ts_config(struct i40e_pf *pf, struct ifreq *ifr);
+int i40e_ptp_hwtstamp_get(struct net_device *netdev,
+			  struct kernel_hwtstamp_config *config);
+int i40e_ptp_hwtstamp_set(struct net_device *netdev,
+			  struct kernel_hwtstamp_config *config,
+			  struct netlink_ext_ack *extack);
+void i40e_ptp_save_hw_time(struct i40e_pf *pf);
+void i40e_ptp_restore_hw_time(struct i40e_pf *pf);
 void i40e_ptp_init(struct i40e_pf *pf);
 void i40e_ptp_stop(struct i40e_pf *pf);
+int i40e_ptp_alloc_pins(struct i40e_pf *pf);
+int i40e_update_adq_vsi_queues(struct i40e_vsi *vsi, int vsi_offset);
 int i40e_is_vsi_uplink_mode_veb(struct i40e_vsi *vsi);
-i40e_status i40e_get_npar_bw_setting(struct i40e_pf *pf);
-i40e_status i40e_set_npar_bw_setting(struct i40e_pf *pf);
-i40e_status i40e_commit_npar_bw_setting(struct i40e_pf *pf);
+int i40e_get_partition_bw_setting(struct i40e_pf *pf);
+int i40e_set_partition_bw_setting(struct i40e_pf *pf);
 void i40e_print_link_message(struct i40e_vsi *vsi, bool isup);
+
+void i40e_set_fec_in_flags(u8 fec_cfg, unsigned long *flags);
+
+static inline bool i40e_enabled_xdp_vsi(struct i40e_vsi *vsi)
+{
+	return !!READ_ONCE(vsi->xdp_prog);
+}
+
+int i40e_create_queue_channel(struct i40e_vsi *vsi, struct i40e_channel *ch);
+int i40e_set_bw_limit(struct i40e_vsi *vsi, u16 seid, u64 max_tx_rate);
+int i40e_add_del_cloud_filter(struct i40e_vsi *vsi,
+			      struct i40e_cloud_filter *filter,
+			      bool add);
+int i40e_add_del_cloud_filter_big_buf(struct i40e_vsi *vsi,
+				      struct i40e_cloud_filter *filter,
+				      bool add);
+
+/**
+ * i40e_is_tc_mqprio_enabled - check if TC MQPRIO is enabled on PF
+ * @pf: pointer to a pf.
+ *
+ * Check and return state of flag I40E_FLAG_TC_MQPRIO.
+ *
+ * Return: true/false if I40E_FLAG_TC_MQPRIO is set or not
+ **/
+static inline bool i40e_is_tc_mqprio_enabled(struct i40e_pf *pf)
+{
+	return test_bit(I40E_FLAG_TC_MQPRIO_ENA, pf->flags);
+}
+
+/**
+ * i40e_hw_to_pf - get pf pointer from the hardware structure
+ * @hw: pointer to the device HW structure
+ **/
+static inline struct i40e_pf *i40e_hw_to_pf(struct i40e_hw *hw)
+{
+	return container_of(hw, struct i40e_pf, hw);
+}
+
+struct device *i40e_hw_to_dev(struct i40e_hw *hw);
+
+/**
+ * i40e_pf_get_vsi_by_seid - find VSI by SEID
+ * @pf: pointer to a PF
+ * @seid: SEID of the VSI
+ **/
+static inline struct i40e_vsi *
+i40e_pf_get_vsi_by_seid(struct i40e_pf *pf, u16 seid)
+{
+	struct i40e_vsi *vsi;
+	int i;
+
+	i40e_pf_for_each_vsi(pf, i, vsi)
+		if (vsi->seid == seid)
+			return vsi;
+
+	return NULL;
+}
+
+/**
+ * i40e_pf_get_main_vsi - get pointer to main VSI
+ * @pf: pointer to a PF
+ *
+ * Return: pointer to main VSI or NULL if it does not exist
+ **/
+static inline struct i40e_vsi *i40e_pf_get_main_vsi(struct i40e_pf *pf)
+{
+	return (pf->lan_vsi != I40E_NO_VSI) ? pf->vsi[pf->lan_vsi] : NULL;
+}
+
+/**
+ * i40e_pf_get_veb_by_seid - find VEB by SEID
+ * @pf: pointer to a PF
+ * @seid: SEID of the VSI
+ **/
+static inline struct i40e_veb *
+i40e_pf_get_veb_by_seid(struct i40e_pf *pf, u16 seid)
+{
+	struct i40e_veb *veb;
+	int i;
+
+	i40e_pf_for_each_veb(pf, i, veb)
+		if (veb->seid == seid)
+			return veb;
+
+	return NULL;
+}
+
+/**
+ * i40e_pf_get_main_veb - get pointer to main VEB
+ * @pf: pointer to a PF
+ *
+ * Return: pointer to main VEB or NULL if it does not exist
+ **/
+static inline struct i40e_veb *i40e_pf_get_main_veb(struct i40e_pf *pf)
+{
+	return (pf->lan_veb != I40E_NO_VEB) ? pf->veb[pf->lan_veb] : NULL;
+}
+
 #endif /* _I40E_H_ */
diff --git a/drivers/net/ethernet/intel/i40e/i40e_adminq.c b/drivers/net/ethernet/intel/i40e/i40e_adminq.c
index ba04988e0598..096ec46bb619 100644
--- a/drivers/net/ethernet/intel/i40e/i40e_adminq.c
+++ b/drivers/net/ethernet/intel/i40e/i40e_adminq.c
@@ -1,83 +1,24 @@
-/*******************************************************************************
- *
- * Intel Ethernet Controller XL710 Family Linux Driver
- * Copyright(c) 2013 - 2016 Intel Corporation.
- *
- * This program is free software; you can redistribute it and/or modify it
- * under the terms and conditions of the GNU General Public License,
- * version 2, as published by the Free Software Foundation.
- *
- * This program is distributed in the hope it will be useful, but WITHOUT
- * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
- * FITNESS FOR A PARTICULAR PURPOSE.  See the GNU General Public License for
- * more details.
- *
- * You should have received a copy of the GNU General Public License along
- * with this program.  If not, see <http://www.gnu.org/licenses/>.
- *
- * The full GNU General Public License is included in this distribution in
- * the file called "COPYING".
- *
- * Contact Information:
- * e1000-devel Mailing List <e1000-devel@lists.sourceforge.net>
- * Intel Corporation, 5200 N.E. Elam Young Parkway, Hillsboro, OR 97124-6497
- *
- ******************************************************************************/
+// SPDX-License-Identifier: GPL-2.0
+/* Copyright(c) 2013 - 2018 Intel Corporation. */
 
-#include "i40e_status.h"
-#include "i40e_type.h"
+#include <linux/delay.h>
+#include "i40e_alloc.h"
 #include "i40e_register.h"
-#include "i40e_adminq.h"
 #include "i40e_prototype.h"
 
 static void i40e_resume_aq(struct i40e_hw *hw);
 
 /**
- *  i40e_adminq_init_regs - Initialize AdminQ registers
- *  @hw: pointer to the hardware structure
- *
- *  This assumes the alloc_asq and alloc_arq functions have already been called
- **/
-static void i40e_adminq_init_regs(struct i40e_hw *hw)
-{
-	/* set head and tail registers in our local struct */
-	if (i40e_is_vf(hw)) {
-		hw->aq.asq.tail = I40E_VF_ATQT1;
-		hw->aq.asq.head = I40E_VF_ATQH1;
-		hw->aq.asq.len  = I40E_VF_ATQLEN1;
-		hw->aq.asq.bal  = I40E_VF_ATQBAL1;
-		hw->aq.asq.bah  = I40E_VF_ATQBAH1;
-		hw->aq.arq.tail = I40E_VF_ARQT1;
-		hw->aq.arq.head = I40E_VF_ARQH1;
-		hw->aq.arq.len  = I40E_VF_ARQLEN1;
-		hw->aq.arq.bal  = I40E_VF_ARQBAL1;
-		hw->aq.arq.bah  = I40E_VF_ARQBAH1;
-	} else {
-		hw->aq.asq.tail = I40E_PF_ATQT;
-		hw->aq.asq.head = I40E_PF_ATQH;
-		hw->aq.asq.len  = I40E_PF_ATQLEN;
-		hw->aq.asq.bal  = I40E_PF_ATQBAL;
-		hw->aq.asq.bah  = I40E_PF_ATQBAH;
-		hw->aq.arq.tail = I40E_PF_ARQT;
-		hw->aq.arq.head = I40E_PF_ARQH;
-		hw->aq.arq.len  = I40E_PF_ARQLEN;
-		hw->aq.arq.bal  = I40E_PF_ARQBAL;
-		hw->aq.arq.bah  = I40E_PF_ARQBAH;
-	}
-}
-
-/**
  *  i40e_alloc_adminq_asq_ring - Allocate Admin Queue send rings
  *  @hw: pointer to the hardware structure
  **/
-static i40e_status i40e_alloc_adminq_asq_ring(struct i40e_hw *hw)
+static int i40e_alloc_adminq_asq_ring(struct i40e_hw *hw)
 {
-	i40e_status ret_code;
+	int ret_code;
 
 	ret_code = i40e_allocate_dma_mem(hw, &hw->aq.asq.desc_buf,
-					 i40e_mem_atq_ring,
 					 (hw->aq.num_asq_entries *
-					 sizeof(struct i40e_aq_desc)),
+					 sizeof(struct libie_aq_desc)),
 					 I40E_ADMINQ_DESC_ALIGNMENT);
 	if (ret_code)
 		return ret_code;
@@ -97,14 +38,13 @@ static i40e_status i40e_alloc_adminq_asq_ring(struct i40e_hw *hw)
  *  i40e_alloc_adminq_arq_ring - Allocate Admin Queue receive rings
  *  @hw: pointer to the hardware structure
  **/
-static i40e_status i40e_alloc_adminq_arq_ring(struct i40e_hw *hw)
+static int i40e_alloc_adminq_arq_ring(struct i40e_hw *hw)
 {
-	i40e_status ret_code;
+	int ret_code;
 
 	ret_code = i40e_allocate_dma_mem(hw, &hw->aq.arq.desc_buf,
-					 i40e_mem_arq_ring,
 					 (hw->aq.num_arq_entries *
-					 sizeof(struct i40e_aq_desc)),
+					 sizeof(struct libie_aq_desc)),
 					 I40E_ADMINQ_DESC_ALIGNMENT);
 
 	return ret_code;
@@ -138,11 +78,11 @@ static void i40e_free_adminq_arq(struct i40e_hw *hw)
  *  i40e_alloc_arq_bufs - Allocate pre-posted buffers for the receive queue
  *  @hw: pointer to the hardware structure
  **/
-static i40e_status i40e_alloc_arq_bufs(struct i40e_hw *hw)
+static int i40e_alloc_arq_bufs(struct i40e_hw *hw)
 {
-	i40e_status ret_code;
-	struct i40e_aq_desc *desc;
+	struct libie_aq_desc *desc;
 	struct i40e_dma_mem *bi;
+	int ret_code;
 	int i;
 
 	/* We'll be allocating the buffer info memory first, then we can
@@ -160,7 +100,6 @@ static i40e_status i40e_alloc_arq_bufs(struct i40e_hw *hw)
 	for (i = 0; i < hw->aq.num_arq_entries; i++) {
 		bi = &hw->aq.arq.r.arq_bi[i];
 		ret_code = i40e_allocate_dma_mem(hw, bi,
-						 i40e_mem_arq_buf,
 						 hw->aq.arq_buf_size,
 						 I40E_ADMINQ_DESC_ALIGNMENT);
 		if (ret_code)
@@ -169,9 +108,9 @@ static i40e_status i40e_alloc_arq_bufs(struct i40e_hw *hw)
 		/* now configure the descriptors for use */
 		desc = I40E_ADMINQ_DESC(hw->aq.arq, i);
 
-		desc->flags = cpu_to_le16(I40E_AQ_FLAG_BUF);
+		desc->flags = cpu_to_le16(LIBIE_AQ_FLAG_BUF);
 		if (hw->aq.arq_buf_size > I40E_AQ_LARGE_BUF)
-			desc->flags |= cpu_to_le16(I40E_AQ_FLAG_LB);
+			desc->flags |= cpu_to_le16(LIBIE_AQ_FLAG_LB);
 		desc->opcode = 0;
 		/* This is in accordance with Admin queue design, there is no
 		 * register for buffer size configuration
@@ -180,12 +119,12 @@ static i40e_status i40e_alloc_arq_bufs(struct i40e_hw *hw)
 		desc->retval = 0;
 		desc->cookie_high = 0;
 		desc->cookie_low = 0;
-		desc->params.external.addr_high =
+		desc->params.generic.addr_high =
 			cpu_to_le32(upper_32_bits(bi->pa));
-		desc->params.external.addr_low =
+		desc->params.generic.addr_low =
 			cpu_to_le32(lower_32_bits(bi->pa));
-		desc->params.external.param0 = 0;
-		desc->params.external.param1 = 0;
+		desc->params.generic.param0 = 0;
+		desc->params.generic.param1 = 0;
 	}
 
 alloc_arq_bufs:
@@ -205,10 +144,10 @@ unwind_alloc_arq_bufs:
  *  i40e_alloc_asq_bufs - Allocate empty buffer structs for the send queue
  *  @hw: pointer to the hardware structure
  **/
-static i40e_status i40e_alloc_asq_bufs(struct i40e_hw *hw)
+static int i40e_alloc_asq_bufs(struct i40e_hw *hw)
 {
-	i40e_status ret_code;
 	struct i40e_dma_mem *bi;
+	int ret_code;
 	int i;
 
 	/* No mapped memory needed yet, just the buffer info structures */
@@ -222,7 +161,6 @@ static i40e_status i40e_alloc_asq_bufs(struct i40e_hw *hw)
 	for (i = 0; i < hw->aq.num_asq_entries; i++) {
 		bi = &hw->aq.asq.r.asq_bi[i];
 		ret_code = i40e_allocate_dma_mem(hw, bi,
-						 i40e_mem_asq_buf,
 						 hw->aq.asq_buf_size,
 						 I40E_ADMINQ_DESC_ALIGNMENT);
 		if (ret_code)
@@ -289,25 +227,25 @@ static void i40e_free_asq_bufs(struct i40e_hw *hw)
  *
  *  Configure base address and length registers for the transmit queue
  **/
-static i40e_status i40e_config_asq_regs(struct i40e_hw *hw)
+static int i40e_config_asq_regs(struct i40e_hw *hw)
 {
-	i40e_status ret_code = 0;
+	int ret_code = 0;
 	u32 reg = 0;
 
 	/* Clear Head and Tail */
-	wr32(hw, hw->aq.asq.head, 0);
-	wr32(hw, hw->aq.asq.tail, 0);
+	wr32(hw, I40E_PF_ATQH, 0);
+	wr32(hw, I40E_PF_ATQT, 0);
 
 	/* set starting point */
-	wr32(hw, hw->aq.asq.len, (hw->aq.num_asq_entries |
+	wr32(hw, I40E_PF_ATQLEN, (hw->aq.num_asq_entries |
 				  I40E_PF_ATQLEN_ATQENABLE_MASK));
-	wr32(hw, hw->aq.asq.bal, lower_32_bits(hw->aq.asq.desc_buf.pa));
-	wr32(hw, hw->aq.asq.bah, upper_32_bits(hw->aq.asq.desc_buf.pa));
+	wr32(hw, I40E_PF_ATQBAL, lower_32_bits(hw->aq.asq.desc_buf.pa));
+	wr32(hw, I40E_PF_ATQBAH, upper_32_bits(hw->aq.asq.desc_buf.pa));
 
 	/* Check one register to verify that config was applied */
-	reg = rd32(hw, hw->aq.asq.bal);
+	reg = rd32(hw, I40E_PF_ATQBAL);
 	if (reg != lower_32_bits(hw->aq.asq.desc_buf.pa))
-		ret_code = I40E_ERR_ADMIN_QUEUE_ERROR;
+		ret_code = -EIO;
 
 	return ret_code;
 }
@@ -318,28 +256,28 @@ static i40e_status i40e_config_asq_regs(struct i40e_hw *hw)
  *
  * Configure base address and length registers for the receive (event queue)
  **/
-static i40e_status i40e_config_arq_regs(struct i40e_hw *hw)
+static int i40e_config_arq_regs(struct i40e_hw *hw)
 {
-	i40e_status ret_code = 0;
+	int ret_code = 0;
 	u32 reg = 0;
 
 	/* Clear Head and Tail */
-	wr32(hw, hw->aq.arq.head, 0);
-	wr32(hw, hw->aq.arq.tail, 0);
+	wr32(hw, I40E_PF_ARQH, 0);
+	wr32(hw, I40E_PF_ARQT, 0);
 
 	/* set starting point */
-	wr32(hw, hw->aq.arq.len, (hw->aq.num_arq_entries |
+	wr32(hw, I40E_PF_ARQLEN, (hw->aq.num_arq_entries |
 				  I40E_PF_ARQLEN_ARQENABLE_MASK));
-	wr32(hw, hw->aq.arq.bal, lower_32_bits(hw->aq.arq.desc_buf.pa));
-	wr32(hw, hw->aq.arq.bah, upper_32_bits(hw->aq.arq.desc_buf.pa));
+	wr32(hw, I40E_PF_ARQBAL, lower_32_bits(hw->aq.arq.desc_buf.pa));
+	wr32(hw, I40E_PF_ARQBAH, upper_32_bits(hw->aq.arq.desc_buf.pa));
 
 	/* Update tail in the HW to post pre-allocated buffers */
-	wr32(hw, hw->aq.arq.tail, hw->aq.num_arq_entries - 1);
+	wr32(hw, I40E_PF_ARQT, hw->aq.num_arq_entries - 1);
 
 	/* Check one register to verify that config was applied */
-	reg = rd32(hw, hw->aq.arq.bal);
+	reg = rd32(hw, I40E_PF_ARQBAL);
 	if (reg != lower_32_bits(hw->aq.arq.desc_buf.pa))
-		ret_code = I40E_ERR_ADMIN_QUEUE_ERROR;
+		ret_code = -EIO;
 
 	return ret_code;
 }
@@ -357,20 +295,20 @@ static i40e_status i40e_config_arq_regs(struct i40e_hw *hw)
  *  Do *NOT* hold the lock when calling this as the memory allocation routines
  *  called are not going to be atomic context safe
  **/
-static i40e_status i40e_init_asq(struct i40e_hw *hw)
+static int i40e_init_asq(struct i40e_hw *hw)
 {
-	i40e_status ret_code = 0;
+	int ret_code = 0;
 
 	if (hw->aq.asq.count > 0) {
 		/* queue already initialized */
-		ret_code = I40E_ERR_NOT_READY;
+		ret_code = -EBUSY;
 		goto init_adminq_exit;
 	}
 
 	/* verify input for valid configuration */
 	if ((hw->aq.num_asq_entries == 0) ||
 	    (hw->aq.asq_buf_size == 0)) {
-		ret_code = I40E_ERR_CONFIG;
+		ret_code = -EIO;
 		goto init_adminq_exit;
 	}
 
@@ -416,20 +354,20 @@ init_adminq_exit:
  *  Do *NOT* hold the lock when calling this as the memory allocation routines
  *  called are not going to be atomic context safe
  **/
-static i40e_status i40e_init_arq(struct i40e_hw *hw)
+static int i40e_init_arq(struct i40e_hw *hw)
 {
-	i40e_status ret_code = 0;
+	int ret_code = 0;
 
 	if (hw->aq.arq.count > 0) {
 		/* queue already initialized */
-		ret_code = I40E_ERR_NOT_READY;
+		ret_code = -EBUSY;
 		goto init_adminq_exit;
 	}
 
 	/* verify input for valid configuration */
 	if ((hw->aq.num_arq_entries == 0) ||
 	    (hw->aq.arq_buf_size == 0)) {
-		ret_code = I40E_ERR_CONFIG;
+		ret_code = -EIO;
 		goto init_adminq_exit;
 	}
 
@@ -468,23 +406,23 @@ init_adminq_exit:
  *
  *  The main shutdown routine for the Admin Send Queue
  **/
-static i40e_status i40e_shutdown_asq(struct i40e_hw *hw)
+static int i40e_shutdown_asq(struct i40e_hw *hw)
 {
-	i40e_status ret_code = 0;
+	int ret_code = 0;
 
 	mutex_lock(&hw->aq.asq_mutex);
 
 	if (hw->aq.asq.count == 0) {
-		ret_code = I40E_ERR_NOT_READY;
+		ret_code = -EBUSY;
 		goto shutdown_asq_out;
 	}
 
 	/* Stop firmware AdminQ processing */
-	wr32(hw, hw->aq.asq.head, 0);
-	wr32(hw, hw->aq.asq.tail, 0);
-	wr32(hw, hw->aq.asq.len, 0);
-	wr32(hw, hw->aq.asq.bal, 0);
-	wr32(hw, hw->aq.asq.bah, 0);
+	wr32(hw, I40E_PF_ATQH, 0);
+	wr32(hw, I40E_PF_ATQT, 0);
+	wr32(hw, I40E_PF_ATQLEN, 0);
+	wr32(hw, I40E_PF_ATQBAL, 0);
+	wr32(hw, I40E_PF_ATQBAH, 0);
 
 	hw->aq.asq.count = 0; /* to indicate uninitialized queue */
 
@@ -502,23 +440,23 @@ shutdown_asq_out:
  *
  *  The main shutdown routine for the Admin Receive Queue
  **/
-static i40e_status i40e_shutdown_arq(struct i40e_hw *hw)
+static int i40e_shutdown_arq(struct i40e_hw *hw)
 {
-	i40e_status ret_code = 0;
+	int ret_code = 0;
 
 	mutex_lock(&hw->aq.arq_mutex);
 
 	if (hw->aq.arq.count == 0) {
-		ret_code = I40E_ERR_NOT_READY;
+		ret_code = -EBUSY;
 		goto shutdown_arq_out;
 	}
 
 	/* Stop firmware AdminQ processing */
-	wr32(hw, hw->aq.arq.head, 0);
-	wr32(hw, hw->aq.arq.tail, 0);
-	wr32(hw, hw->aq.arq.len, 0);
-	wr32(hw, hw->aq.arq.bal, 0);
-	wr32(hw, hw->aq.arq.bah, 0);
+	wr32(hw, I40E_PF_ARQH, 0);
+	wr32(hw, I40E_PF_ARQT, 0);
+	wr32(hw, I40E_PF_ARQLEN, 0);
+	wr32(hw, I40E_PF_ARQBAL, 0);
+	wr32(hw, I40E_PF_ARQBAH, 0);
 
 	hw->aq.arq.count = 0; /* to indicate uninitialized queue */
 
@@ -531,6 +469,98 @@ shutdown_arq_out:
 }
 
 /**
+ *  i40e_set_hw_caps - set HW flags
+ *  @hw: pointer to the hardware structure
+ **/
+static void i40e_set_hw_caps(struct i40e_hw *hw)
+{
+	bitmap_zero(hw->caps, I40E_HW_CAPS_NBITS);
+
+	switch (hw->mac.type) {
+	case I40E_MAC_XL710:
+		if (i40e_is_aq_api_ver_ge(hw, 1,
+					  I40E_MINOR_VER_GET_LINK_INFO_XL710)) {
+			set_bit(I40E_HW_CAP_AQ_PHY_ACCESS, hw->caps);
+			set_bit(I40E_HW_CAP_FW_LLDP_STOPPABLE, hw->caps);
+			/* The ability to RX (not drop) 802.1ad frames */
+			set_bit(I40E_HW_CAP_802_1AD, hw->caps);
+		}
+		if (i40e_is_aq_api_ver_ge(hw, 1, 5)) {
+			/* Supported in FW API version higher than 1.4 */
+			set_bit(I40E_HW_CAP_GENEVE_OFFLOAD, hw->caps);
+		}
+		if (i40e_is_fw_ver_lt(hw, 4, 33)) {
+			set_bit(I40E_HW_CAP_RESTART_AUTONEG, hw->caps);
+			/* No DCB support  for FW < v4.33 */
+			set_bit(I40E_HW_CAP_NO_DCB_SUPPORT, hw->caps);
+		}
+		if (i40e_is_fw_ver_lt(hw, 4, 3)) {
+			/* Disable FW LLDP if FW < v4.3 */
+			set_bit(I40E_HW_CAP_STOP_FW_LLDP, hw->caps);
+		}
+		if (i40e_is_fw_ver_ge(hw, 4, 40)) {
+			/* Use the FW Set LLDP MIB API if FW >= v4.40 */
+			set_bit(I40E_HW_CAP_USE_SET_LLDP_MIB, hw->caps);
+		}
+		if (i40e_is_fw_ver_ge(hw, 6, 0)) {
+			/* Enable PTP L4 if FW > v6.0 */
+			set_bit(I40E_HW_CAP_PTP_L4, hw->caps);
+		}
+		break;
+	case I40E_MAC_X722:
+		set_bit(I40E_HW_CAP_AQ_SRCTL_ACCESS_ENABLE, hw->caps);
+		set_bit(I40E_HW_CAP_NVM_READ_REQUIRES_LOCK, hw->caps);
+		set_bit(I40E_HW_CAP_RSS_AQ, hw->caps);
+		set_bit(I40E_HW_CAP_128_QP_RSS, hw->caps);
+		set_bit(I40E_HW_CAP_ATR_EVICT, hw->caps);
+		set_bit(I40E_HW_CAP_WB_ON_ITR, hw->caps);
+		set_bit(I40E_HW_CAP_MULTI_TCP_UDP_RSS_PCTYPE, hw->caps);
+		set_bit(I40E_HW_CAP_NO_PCI_LINK_CHECK, hw->caps);
+		set_bit(I40E_HW_CAP_USE_SET_LLDP_MIB, hw->caps);
+		set_bit(I40E_HW_CAP_GENEVE_OFFLOAD, hw->caps);
+		set_bit(I40E_HW_CAP_PTP_L4, hw->caps);
+		set_bit(I40E_HW_CAP_WOL_MC_MAGIC_PKT_WAKE, hw->caps);
+		set_bit(I40E_HW_CAP_OUTER_UDP_CSUM, hw->caps);
+
+		if (rd32(hw, I40E_GLQF_FDEVICTENA(1)) !=
+		    I40E_FDEVICT_PCTYPE_DEFAULT) {
+			hw_warn(hw, "FD EVICT PCTYPES are not right, disable FD HW EVICT\n");
+			clear_bit(I40E_HW_CAP_ATR_EVICT, hw->caps);
+		}
+
+		if (i40e_is_aq_api_ver_ge(hw, 1,
+					  I40E_MINOR_VER_FW_LLDP_STOPPABLE_X722))
+			set_bit(I40E_HW_CAP_FW_LLDP_STOPPABLE, hw->caps);
+
+		if (i40e_is_aq_api_ver_ge(hw, 1,
+					  I40E_MINOR_VER_GET_LINK_INFO_X722))
+			set_bit(I40E_HW_CAP_AQ_PHY_ACCESS, hw->caps);
+
+		if (i40e_is_aq_api_ver_ge(hw, 1,
+					  I40E_MINOR_VER_FW_REQUEST_FEC_X722))
+			set_bit(I40E_HW_CAP_X722_FEC_REQUEST, hw->caps);
+
+		fallthrough;
+	default:
+		break;
+	}
+
+	/* Newer versions of firmware require lock when reading the NVM */
+	if (i40e_is_aq_api_ver_ge(hw, 1, 5))
+		set_bit(I40E_HW_CAP_NVM_READ_REQUIRES_LOCK, hw->caps);
+
+	/* The ability to RX (not drop) 802.1ad frames was added in API 1.7 */
+	if (i40e_is_aq_api_ver_ge(hw, 1, 7))
+		set_bit(I40E_HW_CAP_802_1AD, hw->caps);
+
+	if (i40e_is_aq_api_ver_ge(hw, 1, 8))
+		set_bit(I40E_HW_CAP_FW_LLDP_PERSISTENT, hw->caps);
+
+	if (i40e_is_aq_api_ver_ge(hw, 1, 9))
+		set_bit(I40E_HW_CAP_AQ_PHY_ACCESS_EXTENDED, hw->caps);
+}
+
+/**
  *  i40e_init_adminq - main initialization routine for Admin Queue
  *  @hw: pointer to the hardware structure
  *
@@ -541,25 +571,22 @@ shutdown_arq_out:
  *     - hw->aq.arq_buf_size
  *     - hw->aq.asq_buf_size
  **/
-i40e_status i40e_init_adminq(struct i40e_hw *hw)
+int i40e_init_adminq(struct i40e_hw *hw)
 {
 	u16 cfg_ptr, oem_hi, oem_lo;
 	u16 eetrack_lo, eetrack_hi;
-	i40e_status ret_code;
 	int retry = 0;
+	int ret_code;
 
 	/* verify input for valid configuration */
 	if ((hw->aq.num_arq_entries == 0) ||
 	    (hw->aq.num_asq_entries == 0) ||
 	    (hw->aq.arq_buf_size == 0) ||
 	    (hw->aq.asq_buf_size == 0)) {
-		ret_code = I40E_ERR_CONFIG;
+		ret_code = -EIO;
 		goto init_adminq_exit;
 	}
 
-	/* Set up register offsets */
-	i40e_adminq_init_regs(hw);
-
 	/* setup ASQ command write back timeout */
 	hw->aq.asq_cmd_timeout = I40E_ASQ_CMD_TIMEOUT;
 
@@ -585,15 +612,20 @@ i40e_status i40e_init_adminq(struct i40e_hw *hw)
 							&hw->aq.api_maj_ver,
 							&hw->aq.api_min_ver,
 							NULL);
-		if (ret_code != I40E_ERR_ADMIN_QUEUE_TIMEOUT)
+		if (ret_code != -EIO)
 			break;
 		retry++;
 		msleep(100);
 		i40e_resume_aq(hw);
 	} while (retry < 10);
-	if (ret_code != I40E_SUCCESS)
+	if (ret_code != 0)
 		goto init_adminq_free_arq;
 
+	/* Some features were introduced in different FW API version
+	 * for different MAC type.
+	 */
+	i40e_set_hw_caps(hw);
+
 	/* get the NVM version info */
 	i40e_read_nvm_word(hw, I40E_SR_NVM_DEV_STARTER_VERSION,
 			   &hw->nvm.version);
@@ -607,8 +639,8 @@ i40e_status i40e_init_adminq(struct i40e_hw *hw)
 			   &oem_lo);
 	hw->nvm.oem_ver = ((u32)oem_hi << 16) | oem_lo;
 
-	if (hw->aq.api_maj_ver > I40E_FW_API_VERSION_MAJOR) {
-		ret_code = I40E_ERR_FIRMWARE_API_VERSION;
+	if (i40e_is_aq_api_ver_ge(hw, I40E_FW_API_VERSION_MAJOR + 1, 0)) {
+		ret_code = -EIO;
 		goto init_adminq_free_arq;
 	}
 
@@ -636,10 +668,8 @@ init_adminq_exit:
  *  i40e_shutdown_adminq - shutdown routine for the Admin Queue
  *  @hw: pointer to the hardware structure
  **/
-i40e_status i40e_shutdown_adminq(struct i40e_hw *hw)
+void i40e_shutdown_adminq(struct i40e_hw *hw)
 {
-	i40e_status ret_code = 0;
-
 	if (i40e_check_asq_alive(hw))
 		i40e_aq_queue_shutdown(hw, true);
 
@@ -648,8 +678,6 @@ i40e_status i40e_shutdown_adminq(struct i40e_hw *hw)
 
 	if (hw->nvm_buff.va)
 		i40e_free_virt_mem(hw, &hw->nvm_buff);
-
-	return ret_code;
 }
 
 /**
@@ -663,14 +691,14 @@ static u16 i40e_clean_asq(struct i40e_hw *hw)
 	struct i40e_adminq_ring *asq = &(hw->aq.asq);
 	struct i40e_asq_cmd_details *details;
 	u16 ntc = asq->next_to_clean;
-	struct i40e_aq_desc desc_cb;
-	struct i40e_aq_desc *desc;
+	struct libie_aq_desc desc_cb;
+	struct libie_aq_desc *desc;
 
 	desc = I40E_ADMINQ_DESC(*asq, ntc);
 	details = I40E_ADMINQ_DETAILS(*asq, ntc);
-	while (rd32(hw, hw->aq.asq.head) != ntc) {
-		i40e_debug(hw, I40E_DEBUG_AQ_MESSAGE,
-			   "ntc %d head %d.\n", ntc, rd32(hw, hw->aq.asq.head));
+	while (rd32(hw, I40E_PF_ATQH) != ntc) {
+		i40e_debug(hw, I40E_DEBUG_AQ_COMMAND,
+			   "ntc %d head %d.\n", ntc, rd32(hw, I40E_PF_ATQH));
 
 		if (details->callback) {
 			I40E_ADMINQ_CALLBACK cb_func =
@@ -704,51 +732,52 @@ static bool i40e_asq_done(struct i40e_hw *hw)
 	/* AQ designers suggest use of head for better
 	 * timing reliability than DD bit
 	 */
-	return rd32(hw, hw->aq.asq.head) == hw->aq.asq.next_to_use;
+	return rd32(hw, I40E_PF_ATQH) == hw->aq.asq.next_to_use;
 
 }
 
 /**
- *  i40e_asq_send_command - send command to Admin Queue
+ *  i40e_asq_send_command_atomic_exec - send command to Admin Queue
  *  @hw: pointer to the hw struct
  *  @desc: prefilled descriptor describing the command (non DMA mem)
  *  @buff: buffer to use for indirect commands
  *  @buff_size: size of buffer for indirect commands
  *  @cmd_details: pointer to command details structure
+ *  @is_atomic_context: is the function called in an atomic context?
  *
  *  This is the main send command driver routine for the Admin Queue send
  *  queue.  It runs the queue, cleans the queue, etc
  **/
-i40e_status i40e_asq_send_command(struct i40e_hw *hw,
-				struct i40e_aq_desc *desc,
-				void *buff, /* can be NULL */
-				u16  buff_size,
-				struct i40e_asq_cmd_details *cmd_details)
+static int
+i40e_asq_send_command_atomic_exec(struct i40e_hw *hw,
+				  struct libie_aq_desc *desc,
+				  void *buff, /* can be NULL */
+				  u16  buff_size,
+				  struct i40e_asq_cmd_details *cmd_details,
+				  bool is_atomic_context)
 {
-	i40e_status status = 0;
 	struct i40e_dma_mem *dma_buff = NULL;
 	struct i40e_asq_cmd_details *details;
-	struct i40e_aq_desc *desc_on_ring;
+	struct libie_aq_desc *desc_on_ring;
 	bool cmd_completed = false;
 	u16  retval = 0;
+	int status = 0;
 	u32  val = 0;
 
-	mutex_lock(&hw->aq.asq_mutex);
-
 	if (hw->aq.asq.count == 0) {
 		i40e_debug(hw, I40E_DEBUG_AQ_MESSAGE,
 			   "AQTX: Admin queue not initialized.\n");
-		status = I40E_ERR_QUEUE_EMPTY;
+		status = -EIO;
 		goto asq_send_command_error;
 	}
 
-	hw->aq.asq_last_status = I40E_AQ_RC_OK;
+	hw->aq.asq_last_status = LIBIE_AQ_RC_OK;
 
-	val = rd32(hw, hw->aq.asq.head);
+	val = rd32(hw, I40E_PF_ATQH);
 	if (val >= hw->aq.num_asq_entries) {
 		i40e_debug(hw, I40E_DEBUG_AQ_MESSAGE,
 			   "AQTX: head overrun at %d\n", val);
-		status = I40E_ERR_QUEUE_EMPTY;
+		status = -ENOSPC;
 		goto asq_send_command_error;
 	}
 
@@ -779,7 +808,7 @@ i40e_status i40e_asq_send_command(struct i40e_hw *hw,
 			   I40E_DEBUG_AQ_MESSAGE,
 			   "AQTX: Invalid buffer size: %d.\n",
 			   buff_size);
-		status = I40E_ERR_INVALID_SIZE;
+		status = -EINVAL;
 		goto asq_send_command_error;
 	}
 
@@ -787,7 +816,7 @@ i40e_status i40e_asq_send_command(struct i40e_hw *hw,
 		i40e_debug(hw,
 			   I40E_DEBUG_AQ_MESSAGE,
 			   "AQTX: Async flag not set along with postpone flag");
-		status = I40E_ERR_PARAM;
+		status = -EINVAL;
 		goto asq_send_command_error;
 	}
 
@@ -802,7 +831,7 @@ i40e_status i40e_asq_send_command(struct i40e_hw *hw,
 		i40e_debug(hw,
 			   I40E_DEBUG_AQ_MESSAGE,
 			   "AQTX: Error queue is full.\n");
-		status = I40E_ERR_ADMIN_QUEUE_FULL;
+		status = -ENOSPC;
 		goto asq_send_command_error;
 	}
 
@@ -822,21 +851,21 @@ i40e_status i40e_asq_send_command(struct i40e_hw *hw,
 		/* Update the address values in the desc with the pa value
 		 * for respective buffer
 		 */
-		desc_on_ring->params.external.addr_high =
+		desc_on_ring->params.generic.addr_high =
 				cpu_to_le32(upper_32_bits(dma_buff->pa));
-		desc_on_ring->params.external.addr_low =
+		desc_on_ring->params.generic.addr_low =
 				cpu_to_le32(lower_32_bits(dma_buff->pa));
 	}
 
 	/* bump the tail */
-	i40e_debug(hw, I40E_DEBUG_AQ_MESSAGE, "AQTX: desc and buffer:\n");
+	i40e_debug(hw, I40E_DEBUG_AQ_COMMAND, "AQTX: desc and buffer:\n");
 	i40e_debug_aq(hw, I40E_DEBUG_AQ_COMMAND, (void *)desc_on_ring,
 		      buff, buff_size);
 	(hw->aq.asq.next_to_use)++;
 	if (hw->aq.asq.next_to_use == hw->aq.asq.count)
 		hw->aq.asq.next_to_use = 0;
 	if (!details->postpone)
-		wr32(hw, hw->aq.asq.tail, hw->aq.asq.next_to_use);
+		wr32(hw, I40E_PF_ATQT, hw->aq.asq.next_to_use);
 
 	/* if cmd_details are not defined or async flag is not set,
 	 * we need to wait for desc write back
@@ -850,7 +879,12 @@ i40e_status i40e_asq_send_command(struct i40e_hw *hw,
 			 */
 			if (i40e_asq_done(hw))
 				break;
-			udelay(50);
+
+			if (is_atomic_context)
+				udelay(50);
+			else
+				usleep_range(40, 60);
+
 			total_delay += 50;
 		} while (total_delay < hw->aq.asq_cmd_timeout);
 	}
@@ -871,14 +905,16 @@ i40e_status i40e_asq_send_command(struct i40e_hw *hw,
 			retval &= 0xff;
 		}
 		cmd_completed = true;
-		if ((enum i40e_admin_queue_err)retval == I40E_AQ_RC_OK)
+		if ((enum libie_aq_err)retval == LIBIE_AQ_RC_OK)
 			status = 0;
+		else if ((enum libie_aq_err)retval == LIBIE_AQ_RC_EBUSY)
+			status = -EBUSY;
 		else
-			status = I40E_ERR_ADMIN_QUEUE_ERROR;
-		hw->aq.asq_last_status = (enum i40e_admin_queue_err)retval;
+			status = -EIO;
+		hw->aq.asq_last_status = (enum libie_aq_err)retval;
 	}
 
-	i40e_debug(hw, I40E_DEBUG_AQ_MESSAGE,
+	i40e_debug(hw, I40E_DEBUG_AQ_COMMAND,
 		   "AQTX: desc and buffer writeback:\n");
 	i40e_debug_aq(hw, I40E_DEBUG_AQ_COMMAND, (void *)desc, buff, buff_size);
 
@@ -889,13 +925,93 @@ i40e_status i40e_asq_send_command(struct i40e_hw *hw,
 	/* update the error if time out occurred */
 	if ((!cmd_completed) &&
 	    (!details->async && !details->postpone)) {
-		i40e_debug(hw,
-			   I40E_DEBUG_AQ_MESSAGE,
-			   "AQTX: Writeback timeout.\n");
-		status = I40E_ERR_ADMIN_QUEUE_TIMEOUT;
+		if (rd32(hw, I40E_PF_ATQLEN) & I40E_GL_ATQLEN_ATQCRIT_MASK) {
+			i40e_debug(hw, I40E_DEBUG_AQ_MESSAGE,
+				   "AQTX: AQ Critical error.\n");
+			status = -EIO;
+		} else {
+			i40e_debug(hw, I40E_DEBUG_AQ_MESSAGE,
+				   "AQTX: Writeback timeout.\n");
+			status = -EIO;
+		}
 	}
 
 asq_send_command_error:
+	return status;
+}
+
+/**
+ *  i40e_asq_send_command_atomic - send command to Admin Queue
+ *  @hw: pointer to the hw struct
+ *  @desc: prefilled descriptor describing the command (non DMA mem)
+ *  @buff: buffer to use for indirect commands
+ *  @buff_size: size of buffer for indirect commands
+ *  @cmd_details: pointer to command details structure
+ *  @is_atomic_context: is the function called in an atomic context?
+ *
+ *  Acquires the lock and calls the main send command execution
+ *  routine.
+ **/
+int
+i40e_asq_send_command_atomic(struct i40e_hw *hw,
+			     struct libie_aq_desc *desc,
+			     void *buff, /* can be NULL */
+			     u16  buff_size,
+			     struct i40e_asq_cmd_details *cmd_details,
+			     bool is_atomic_context)
+{
+	int status;
+
+	mutex_lock(&hw->aq.asq_mutex);
+	status = i40e_asq_send_command_atomic_exec(hw, desc, buff, buff_size,
+						   cmd_details,
+						   is_atomic_context);
+
+	mutex_unlock(&hw->aq.asq_mutex);
+	return status;
+}
+
+int
+i40e_asq_send_command(struct i40e_hw *hw, struct libie_aq_desc *desc,
+		      void *buff, /* can be NULL */ u16  buff_size,
+		      struct i40e_asq_cmd_details *cmd_details)
+{
+	return i40e_asq_send_command_atomic(hw, desc, buff, buff_size,
+					    cmd_details, false);
+}
+
+/**
+ *  i40e_asq_send_command_atomic_v2 - send command to Admin Queue
+ *  @hw: pointer to the hw struct
+ *  @desc: prefilled descriptor describing the command (non DMA mem)
+ *  @buff: buffer to use for indirect commands
+ *  @buff_size: size of buffer for indirect commands
+ *  @cmd_details: pointer to command details structure
+ *  @is_atomic_context: is the function called in an atomic context?
+ *  @aq_status: pointer to Admin Queue status return value
+ *
+ *  Acquires the lock and calls the main send command execution
+ *  routine. Returns the last Admin Queue status in aq_status
+ *  to avoid race conditions in access to hw->aq.asq_last_status.
+ **/
+int
+i40e_asq_send_command_atomic_v2(struct i40e_hw *hw,
+				struct libie_aq_desc *desc,
+				void *buff, /* can be NULL */
+				u16  buff_size,
+				struct i40e_asq_cmd_details *cmd_details,
+				bool is_atomic_context,
+				enum libie_aq_err *aq_status)
+{
+	int status;
+
+	mutex_lock(&hw->aq.asq_mutex);
+	status = i40e_asq_send_command_atomic_exec(hw, desc, buff,
+						   buff_size,
+						   cmd_details,
+						   is_atomic_context);
+	if (aq_status)
+		*aq_status = hw->aq.asq_last_status;
 	mutex_unlock(&hw->aq.asq_mutex);
 	return status;
 }
@@ -907,13 +1023,13 @@ asq_send_command_error:
  *
  *  Fill the desc with default values
  **/
-void i40e_fill_default_direct_cmd_desc(struct i40e_aq_desc *desc,
+void i40e_fill_default_direct_cmd_desc(struct libie_aq_desc *desc,
 				       u16 opcode)
 {
 	/* zero out the desc */
-	memset((void *)desc, 0, sizeof(struct i40e_aq_desc));
+	memset((void *)desc, 0, sizeof(struct libie_aq_desc));
 	desc->opcode = cpu_to_le16(opcode);
-	desc->flags = cpu_to_le16(I40E_AQ_FLAG_SI);
+	desc->flags = cpu_to_le16(LIBIE_AQ_FLAG_SI);
 }
 
 /**
@@ -926,14 +1042,14 @@ void i40e_fill_default_direct_cmd_desc(struct i40e_aq_desc *desc,
  *  the contents through e.  It can also return how many events are
  *  left to process through 'pending'
  **/
-i40e_status i40e_clean_arq_element(struct i40e_hw *hw,
-					     struct i40e_arq_event_info *e,
-					     u16 *pending)
+int i40e_clean_arq_element(struct i40e_hw *hw,
+			   struct i40e_arq_event_info *e,
+			   u16 *pending)
 {
-	i40e_status ret_code = 0;
 	u16 ntc = hw->aq.arq.next_to_clean;
-	struct i40e_aq_desc *desc;
+	struct libie_aq_desc *desc;
 	struct i40e_dma_mem *bi;
+	int ret_code = 0;
 	u16 desc_idx;
 	u16 datalen;
 	u16 flags;
@@ -948,15 +1064,15 @@ i40e_status i40e_clean_arq_element(struct i40e_hw *hw,
 	if (hw->aq.arq.count == 0) {
 		i40e_debug(hw, I40E_DEBUG_AQ_MESSAGE,
 			   "AQRX: Admin queue not initialized.\n");
-		ret_code = I40E_ERR_QUEUE_EMPTY;
+		ret_code = -EIO;
 		goto clean_arq_element_err;
 	}
 
 	/* set next_to_use to head */
-	ntu = (rd32(hw, hw->aq.arq.head) & I40E_PF_ARQH_ARQH_MASK);
+	ntu = rd32(hw, I40E_PF_ARQH) & I40E_PF_ARQH_ARQH_MASK;
 	if (ntu == ntc) {
 		/* nothing to do - shouldn't need to update ring's values */
-		ret_code = I40E_ERR_ADMIN_QUEUE_NO_WORK;
+		ret_code = -EALREADY;
 		goto clean_arq_element_out;
 	}
 
@@ -965,10 +1081,10 @@ i40e_status i40e_clean_arq_element(struct i40e_hw *hw,
 	desc_idx = ntc;
 
 	hw->aq.arq_last_status =
-		(enum i40e_admin_queue_err)le16_to_cpu(desc->retval);
+		(enum libie_aq_err)le16_to_cpu(desc->retval);
 	flags = le16_to_cpu(desc->flags);
-	if (flags & I40E_AQ_FLAG_ERR) {
-		ret_code = I40E_ERR_ADMIN_QUEUE_ERROR;
+	if (flags & LIBIE_AQ_FLAG_ERR) {
+		ret_code = -EIO;
 		i40e_debug(hw,
 			   I40E_DEBUG_AQ_MESSAGE,
 			   "AQRX: Event received with error 0x%X.\n",
@@ -982,7 +1098,7 @@ i40e_status i40e_clean_arq_element(struct i40e_hw *hw,
 		memcpy(e->msg_buf, hw->aq.arq.r.arq_bi[desc_idx].va,
 		       e->msg_len);
 
-	i40e_debug(hw, I40E_DEBUG_AQ_MESSAGE, "AQRX: desc and buffer:\n");
+	i40e_debug(hw, I40E_DEBUG_AQ_COMMAND, "AQRX: desc and buffer:\n");
 	i40e_debug_aq(hw, I40E_DEBUG_AQ_COMMAND, (void *)desc, e->msg_buf,
 		      hw->aq.arq_buf_size);
 
@@ -991,17 +1107,17 @@ i40e_status i40e_clean_arq_element(struct i40e_hw *hw,
 	 * size
 	 */
 	bi = &hw->aq.arq.r.arq_bi[ntc];
-	memset((void *)desc, 0, sizeof(struct i40e_aq_desc));
+	memset((void *)desc, 0, sizeof(struct libie_aq_desc));
 
-	desc->flags = cpu_to_le16(I40E_AQ_FLAG_BUF);
+	desc->flags = cpu_to_le16(LIBIE_AQ_FLAG_BUF);
 	if (hw->aq.arq_buf_size > I40E_AQ_LARGE_BUF)
-		desc->flags |= cpu_to_le16(I40E_AQ_FLAG_LB);
+		desc->flags |= cpu_to_le16(LIBIE_AQ_FLAG_LB);
 	desc->datalen = cpu_to_le16((u16)bi->size);
-	desc->params.external.addr_high = cpu_to_le32(upper_32_bits(bi->pa));
-	desc->params.external.addr_low = cpu_to_le32(lower_32_bits(bi->pa));
+	desc->params.generic.addr_high = cpu_to_le32(upper_32_bits(bi->pa));
+	desc->params.generic.addr_low = cpu_to_le32(lower_32_bits(bi->pa));
 
 	/* set tail = the last cleaned desc index. */
-	wr32(hw, hw->aq.arq.tail, ntc);
+	wr32(hw, I40E_PF_ARQT, ntc);
 	/* ntc is updated to tail + 1 */
 	ntc++;
 	if (ntc == hw->aq.num_arq_entries)
@@ -1009,7 +1125,7 @@ i40e_status i40e_clean_arq_element(struct i40e_hw *hw,
 	hw->aq.arq.next_to_clean = ntc;
 	hw->aq.arq.next_to_use = ntu;
 
-	i40e_nvmupd_check_wait_event(hw, le16_to_cpu(e->desc.opcode));
+	i40e_nvmupd_check_wait_event(hw, le16_to_cpu(e->desc.opcode), &e->desc);
 clean_arq_element_out:
 	/* Set pending if needed, unlock and return */
 	if (pending)
diff --git a/drivers/net/ethernet/intel/i40e/i40e_adminq.h b/drivers/net/ethernet/intel/i40e/i40e_adminq.h
index 2349fbe04bd2..1be97a3a86ce 100644
--- a/drivers/net/ethernet/intel/i40e/i40e_adminq.h
+++ b/drivers/net/ethernet/intel/i40e/i40e_adminq.h
@@ -1,38 +1,15 @@
-/*******************************************************************************
- *
- * Intel Ethernet Controller XL710 Family Linux Driver
- * Copyright(c) 2013 - 2014 Intel Corporation.
- *
- * This program is free software; you can redistribute it and/or modify it
- * under the terms and conditions of the GNU General Public License,
- * version 2, as published by the Free Software Foundation.
- *
- * This program is distributed in the hope it will be useful, but WITHOUT
- * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
- * FITNESS FOR A PARTICULAR PURPOSE.  See the GNU General Public License for
- * more details.
- *
- * You should have received a copy of the GNU General Public License along
- * with this program.  If not, see <http://www.gnu.org/licenses/>.
- *
- * The full GNU General Public License is included in this distribution in
- * the file called "COPYING".
- *
- * Contact Information:
- * e1000-devel Mailing List <e1000-devel@lists.sourceforge.net>
- * Intel Corporation, 5200 N.E. Elam Young Parkway, Hillsboro, OR 97124-6497
- *
- ******************************************************************************/
+/* SPDX-License-Identifier: GPL-2.0 */
+/* Copyright(c) 2013 - 2018 Intel Corporation. */
 
 #ifndef _I40E_ADMINQ_H_
 #define _I40E_ADMINQ_H_
 
-#include "i40e_osdep.h"
-#include "i40e_status.h"
+#include <linux/mutex.h>
+#include "i40e_alloc.h"
 #include "i40e_adminq_cmd.h"
 
 #define I40E_ADMINQ_DESC(R, i)   \
-	(&(((struct i40e_aq_desc *)((R).desc_buf.va))[i]))
+	(&(((struct libie_aq_desc *)((R).desc_buf.va))[i]))
 
 #define I40E_ADMINQ_DESC_ALIGNMENT 4096
 
@@ -52,13 +29,6 @@ struct i40e_adminq_ring {
 	/* used for interrupt processing */
 	u16 next_to_use;
 	u16 next_to_clean;
-
-	/* used for queue tracking */
-	u32 head;
-	u32 tail;
-	u32 len;
-	u32 bah;
-	u32 bal;
 };
 
 /* ASQ transaction details */
@@ -69,7 +39,7 @@ struct i40e_asq_cmd_details {
 	u16 flags_dis;
 	bool async;
 	bool postpone;
-	struct i40e_aq_desc *wb_desc;
+	struct libie_aq_desc *wb_desc;
 };
 
 #define I40E_ADMINQ_DETAILS(R, i)   \
@@ -77,7 +47,7 @@ struct i40e_asq_cmd_details {
 
 /* ARQ event information */
 struct i40e_arq_event_info {
-	struct i40e_aq_desc desc;
+	struct libie_aq_desc desc;
 	u16 msg_len;
 	u16 buf_len;
 	u8 *msg_buf;
@@ -102,14 +72,14 @@ struct i40e_adminq_info {
 	struct mutex arq_mutex; /* Receive queue lock */
 
 	/* last status values on send and receive queues */
-	enum i40e_admin_queue_err asq_last_status;
-	enum i40e_admin_queue_err arq_last_status;
+	enum libie_aq_err asq_last_status;
+	enum libie_aq_err arq_last_status;
 };
 
 /**
  * i40e_aq_rc_to_posix - convert errors to user-land codes
- * aq_ret: AdminQ handler error code can override aq_rc
- * aq_rc: AdminQ firmware error code to convert
+ * @aq_ret: AdminQ handler error code can override aq_rc
+ * @aq_rc: AdminQ firmware error code to convert
  **/
 static inline int i40e_aq_rc_to_posix(int aq_ret, int aq_rc)
 {
@@ -139,10 +109,6 @@ static inline int i40e_aq_rc_to_posix(int aq_ret, int aq_rc)
 		-EFBIG,      /* I40E_AQ_RC_EFBIG */
 	};
 
-	/* aq_rc is invalid if AQ timed out */
-	if (aq_ret == I40E_ERR_ADMIN_QUEUE_TIMEOUT)
-		return -EAGAIN;
-
 	if (!((u32)aq_rc < (sizeof(aq_to_posix) / sizeof((aq_to_posix)[0]))))
 		return -ERANGE;
 
@@ -153,7 +119,7 @@ static inline int i40e_aq_rc_to_posix(int aq_ret, int aq_rc)
 #define I40E_AQ_LARGE_BUF	512
 #define I40E_ASQ_CMD_TIMEOUT	250000  /* usecs */
 
-void i40e_fill_default_direct_cmd_desc(struct i40e_aq_desc *desc,
+void i40e_fill_default_direct_cmd_desc(struct libie_aq_desc *desc,
 				       u16 opcode);
 
 #endif /* _I40E_ADMINQ_H_ */
diff --git a/drivers/net/ethernet/intel/i40e/i40e_adminq_cmd.h b/drivers/net/ethernet/intel/i40e/i40e_adminq_cmd.h
index 5eb04114e13f..cc02a85ad42b 100644
--- a/drivers/net/ethernet/intel/i40e/i40e_adminq_cmd.h
+++ b/drivers/net/ethernet/intel/i40e/i40e_adminq_cmd.h
@@ -1,32 +1,14 @@
-/*******************************************************************************
- *
- * Intel Ethernet Controller XL710 Family Linux Driver
- * Copyright(c) 2013 - 2016 Intel Corporation.
- *
- * This program is free software; you can redistribute it and/or modify it
- * under the terms and conditions of the GNU General Public License,
- * version 2, as published by the Free Software Foundation.
- *
- * This program is distributed in the hope it will be useful, but WITHOUT
- * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
- * FITNESS FOR A PARTICULAR PURPOSE.  See the GNU General Public License for
- * more details.
- *
- * You should have received a copy of the GNU General Public License along
- * with this program.  If not, see <http://www.gnu.org/licenses/>.
- *
- * The full GNU General Public License is included in this distribution in
- * the file called "COPYING".
- *
- * Contact Information:
- * e1000-devel Mailing List <e1000-devel@lists.sourceforge.net>
- * Intel Corporation, 5200 N.E. Elam Young Parkway, Hillsboro, OR 97124-6497
- *
- ******************************************************************************/
+/* SPDX-License-Identifier: GPL-2.0 */
+/* Copyright(c) 2013 - 2021 Intel Corporation. */
 
 #ifndef _I40E_ADMINQ_CMD_H_
 #define _I40E_ADMINQ_CMD_H_
 
+#include <linux/net/intel/libie/adminq.h>
+
+#include <linux/bits.h>
+#include <linux/types.h>
+
 /* This header file defines the i40e Admin Queue commands and is shared between
  * i40e Firmware and Software.
  *
@@ -34,88 +16,21 @@
  */
 
 #define I40E_FW_API_VERSION_MAJOR	0x0001
-#define I40E_FW_API_VERSION_MINOR	0x0005
-
-struct i40e_aq_desc {
-	__le16 flags;
-	__le16 opcode;
-	__le16 datalen;
-	__le16 retval;
-	__le32 cookie_high;
-	__le32 cookie_low;
-	union {
-		struct {
-			__le32 param0;
-			__le32 param1;
-			__le32 param2;
-			__le32 param3;
-		} internal;
-		struct {
-			__le32 param0;
-			__le32 param1;
-			__le32 addr_high;
-			__le32 addr_low;
-		} external;
-		u8 raw[16];
-	} params;
-};
-
-/* Flags sub-structure
- * |0  |1  |2  |3  |4  |5  |6  |7  |8  |9  |10 |11 |12 |13 |14 |15 |
- * |DD |CMP|ERR|VFE| * *  RESERVED * * |LB |RD |VFC|BUF|SI |EI |FE |
- */
-
-/* command flags and offsets*/
-#define I40E_AQ_FLAG_DD_SHIFT	0
-#define I40E_AQ_FLAG_CMP_SHIFT	1
-#define I40E_AQ_FLAG_ERR_SHIFT	2
-#define I40E_AQ_FLAG_VFE_SHIFT	3
-#define I40E_AQ_FLAG_LB_SHIFT	9
-#define I40E_AQ_FLAG_RD_SHIFT	10
-#define I40E_AQ_FLAG_VFC_SHIFT	11
-#define I40E_AQ_FLAG_BUF_SHIFT	12
-#define I40E_AQ_FLAG_SI_SHIFT	13
-#define I40E_AQ_FLAG_EI_SHIFT	14
-#define I40E_AQ_FLAG_FE_SHIFT	15
-
-#define I40E_AQ_FLAG_DD		BIT(I40E_AQ_FLAG_DD_SHIFT)  /* 0x1    */
-#define I40E_AQ_FLAG_CMP	BIT(I40E_AQ_FLAG_CMP_SHIFT) /* 0x2    */
-#define I40E_AQ_FLAG_ERR	BIT(I40E_AQ_FLAG_ERR_SHIFT) /* 0x4    */
-#define I40E_AQ_FLAG_VFE	BIT(I40E_AQ_FLAG_VFE_SHIFT) /* 0x8    */
-#define I40E_AQ_FLAG_LB		BIT(I40E_AQ_FLAG_LB_SHIFT)  /* 0x200  */
-#define I40E_AQ_FLAG_RD		BIT(I40E_AQ_FLAG_RD_SHIFT)  /* 0x400  */
-#define I40E_AQ_FLAG_VFC	BIT(I40E_AQ_FLAG_VFC_SHIFT) /* 0x800  */
-#define I40E_AQ_FLAG_BUF	BIT(I40E_AQ_FLAG_BUF_SHIFT) /* 0x1000 */
-#define I40E_AQ_FLAG_SI		BIT(I40E_AQ_FLAG_SI_SHIFT)  /* 0x2000 */
-#define I40E_AQ_FLAG_EI		BIT(I40E_AQ_FLAG_EI_SHIFT)  /* 0x4000 */
-#define I40E_AQ_FLAG_FE		BIT(I40E_AQ_FLAG_FE_SHIFT)  /* 0x8000 */
-
-/* error codes */
-enum i40e_admin_queue_err {
-	I40E_AQ_RC_OK		= 0,  /* success */
-	I40E_AQ_RC_EPERM	= 1,  /* Operation not permitted */
-	I40E_AQ_RC_ENOENT	= 2,  /* No such element */
-	I40E_AQ_RC_ESRCH	= 3,  /* Bad opcode */
-	I40E_AQ_RC_EINTR	= 4,  /* operation interrupted */
-	I40E_AQ_RC_EIO		= 5,  /* I/O error */
-	I40E_AQ_RC_ENXIO	= 6,  /* No such resource */
-	I40E_AQ_RC_E2BIG	= 7,  /* Arg too long */
-	I40E_AQ_RC_EAGAIN	= 8,  /* Try again */
-	I40E_AQ_RC_ENOMEM	= 9,  /* Out of memory */
-	I40E_AQ_RC_EACCES	= 10, /* Permission denied */
-	I40E_AQ_RC_EFAULT	= 11, /* Bad address */
-	I40E_AQ_RC_EBUSY	= 12, /* Device or resource busy */
-	I40E_AQ_RC_EEXIST	= 13, /* object already exists */
-	I40E_AQ_RC_EINVAL	= 14, /* Invalid argument */
-	I40E_AQ_RC_ENOTTY	= 15, /* Not a typewriter */
-	I40E_AQ_RC_ENOSPC	= 16, /* No space left or alloc failure */
-	I40E_AQ_RC_ENOSYS	= 17, /* Function not implemented */
-	I40E_AQ_RC_ERANGE	= 18, /* Parameter out of range */
-	I40E_AQ_RC_EFLUSHED	= 19, /* Cmd flushed due to prev cmd error */
-	I40E_AQ_RC_BAD_ADDR	= 20, /* Descriptor contains a bad pointer */
-	I40E_AQ_RC_EMODE	= 21, /* Op not allowed in current dev mode */
-	I40E_AQ_RC_EFBIG	= 22, /* File too large */
-};
+#define I40E_FW_API_VERSION_MINOR_X722	0x000C
+#define I40E_FW_API_VERSION_MINOR_X710	0x000F
+
+#define I40E_FW_MINOR_VERSION(_h) ((_h)->mac.type == I40E_MAC_XL710 ? \
+					I40E_FW_API_VERSION_MINOR_X710 : \
+					I40E_FW_API_VERSION_MINOR_X722)
+
+/* API version 1.7 implements additional link and PHY-specific APIs  */
+#define I40E_MINOR_VER_GET_LINK_INFO_XL710 0x0007
+/* API version 1.9 for X722 implements additional link and PHY-specific APIs */
+#define I40E_MINOR_VER_GET_LINK_INFO_X722 0x0009
+/* API version 1.6 for X722 devices adds ability to stop FW LLDP agent */
+#define I40E_MINOR_VER_FW_LLDP_STOPPABLE_X722 0x0006
+/* API version 1.10 for X722 devices adds ability to request FEC encoding */
+#define I40E_MINOR_VER_FW_REQUEST_FEC_X722 0x000A
 
 /* Admin Queue command opcodes */
 enum i40e_admin_queue_opc {
@@ -190,13 +105,14 @@ enum i40e_admin_queue_opc {
 	i40e_aqc_opc_add_mirror_rule	= 0x0260,
 	i40e_aqc_opc_delete_mirror_rule	= 0x0261,
 
-	/* Pipeline Personalization Profile */
+	/* Dynamic Device Personalization */
 	i40e_aqc_opc_write_personalization_profile	= 0x0270,
 	i40e_aqc_opc_get_personalization_profile_list	= 0x0271,
 
 	/* DCB commands */
 	i40e_aqc_opc_dcb_ignore_pfc	= 0x0301,
 	i40e_aqc_opc_dcb_updated	= 0x0302,
+	i40e_aqc_opc_set_dcb_parameters = 0x0303,
 
 	/* TX scheduler */
 	i40e_aqc_opc_configure_vsi_bw_limit		= 0x0400,
@@ -236,6 +152,8 @@ enum i40e_admin_queue_opc {
 	i40e_aqc_opc_set_phy_debug		= 0x0622,
 	i40e_aqc_opc_upload_ext_phy_fm		= 0x0625,
 	i40e_aqc_opc_run_phy_activity		= 0x0626,
+	i40e_aqc_opc_set_phy_register		= 0x0628,
+	i40e_aqc_opc_get_phy_register		= 0x0629,
 
 	/* NVM commands */
 	i40e_aqc_opc_nvm_read			= 0x0701,
@@ -271,6 +189,7 @@ enum i40e_admin_queue_opc {
 	i40e_aqc_opc_get_cee_dcb_cfg	= 0x0A07,
 	i40e_aqc_opc_lldp_set_local_mib	= 0x0A08,
 	i40e_aqc_opc_lldp_stop_start_spec_agent	= 0x0A09,
+	i40e_aqc_opc_lldp_restore		= 0x0A0A,
 
 	/* Tunnel commands */
 	i40e_aqc_opc_add_udp_tunnel	= 0x0B00,
@@ -334,21 +253,6 @@ struct i40e_aqc_get_version {
 	__le16 api_minor;
 };
 
-I40E_CHECK_CMD_LENGTH(i40e_aqc_get_version);
-
-/* Send driver version (indirect 0x0002) */
-struct i40e_aqc_driver_version {
-	u8	driver_major_ver;
-	u8	driver_minor_ver;
-	u8	driver_build_ver;
-	u8	driver_subbuild_ver;
-	u8	reserved[4];
-	__le32	address_high;
-	__le32	address_low;
-};
-
-I40E_CHECK_CMD_LENGTH(i40e_aqc_driver_version);
-
 /* Queue Shutdown (direct 0x0003) */
 struct i40e_aqc_queue_shutdown {
 	__le32	driver_unloading;
@@ -366,93 +270,9 @@ struct i40e_aqc_set_pf_context {
 
 I40E_CHECK_CMD_LENGTH(i40e_aqc_set_pf_context);
 
-/* Request resource ownership (direct 0x0008)
- * Release resource ownership (direct 0x0009)
- */
-#define I40E_AQ_RESOURCE_NVM			1
-#define I40E_AQ_RESOURCE_SDP			2
-#define I40E_AQ_RESOURCE_ACCESS_READ		1
-#define I40E_AQ_RESOURCE_ACCESS_WRITE		2
-#define I40E_AQ_RESOURCE_NVM_READ_TIMEOUT	3000
-#define I40E_AQ_RESOURCE_NVM_WRITE_TIMEOUT	180000
-
-struct i40e_aqc_request_resource {
-	__le16	resource_id;
-	__le16	access_type;
-	__le32	timeout;
-	__le32	resource_number;
-	u8	reserved[4];
-};
-
-I40E_CHECK_CMD_LENGTH(i40e_aqc_request_resource);
-
-/* Get function capabilities (indirect 0x000A)
- * Get device capabilities (indirect 0x000B)
- */
-struct i40e_aqc_list_capabilites {
-	u8 command_flags;
-#define I40E_AQ_LIST_CAP_PF_INDEX_EN	1
-	u8 pf_index;
-	u8 reserved[2];
-	__le32 count;
-	__le32 addr_high;
-	__le32 addr_low;
-};
-
-I40E_CHECK_CMD_LENGTH(i40e_aqc_list_capabilites);
-
-struct i40e_aqc_list_capabilities_element_resp {
-	__le16	id;
-	u8	major_rev;
-	u8	minor_rev;
-	__le32	number;
-	__le32	logical_id;
-	__le32	phys_id;
-	u8	reserved[16];
-};
-
-/* list of caps */
-
-#define I40E_AQ_CAP_ID_SWITCH_MODE	0x0001
-#define I40E_AQ_CAP_ID_MNG_MODE		0x0002
-#define I40E_AQ_CAP_ID_NPAR_ACTIVE	0x0003
-#define I40E_AQ_CAP_ID_OS2BMC_CAP	0x0004
-#define I40E_AQ_CAP_ID_FUNCTIONS_VALID	0x0005
-#define I40E_AQ_CAP_ID_ALTERNATE_RAM	0x0006
-#define I40E_AQ_CAP_ID_WOL_AND_PROXY	0x0008
-#define I40E_AQ_CAP_ID_SRIOV		0x0012
-#define I40E_AQ_CAP_ID_VF		0x0013
-#define I40E_AQ_CAP_ID_VMDQ		0x0014
-#define I40E_AQ_CAP_ID_8021QBG		0x0015
-#define I40E_AQ_CAP_ID_8021QBR		0x0016
-#define I40E_AQ_CAP_ID_VSI		0x0017
-#define I40E_AQ_CAP_ID_DCB		0x0018
-#define I40E_AQ_CAP_ID_FCOE		0x0021
-#define I40E_AQ_CAP_ID_ISCSI		0x0022
-#define I40E_AQ_CAP_ID_RSS		0x0040
-#define I40E_AQ_CAP_ID_RXQ		0x0041
-#define I40E_AQ_CAP_ID_TXQ		0x0042
-#define I40E_AQ_CAP_ID_MSIX		0x0043
-#define I40E_AQ_CAP_ID_VF_MSIX		0x0044
-#define I40E_AQ_CAP_ID_FLOW_DIRECTOR	0x0045
-#define I40E_AQ_CAP_ID_1588		0x0046
-#define I40E_AQ_CAP_ID_IWARP		0x0051
-#define I40E_AQ_CAP_ID_LED		0x0061
-#define I40E_AQ_CAP_ID_SDP		0x0062
-#define I40E_AQ_CAP_ID_MDIO		0x0063
-#define I40E_AQ_CAP_ID_WSR_PROT		0x0064
-#define I40E_AQ_CAP_ID_NVM_MGMT		0x0080
-#define I40E_AQ_CAP_ID_FLEX10		0x00F1
-#define I40E_AQ_CAP_ID_CEM		0x00F2
-
 /* Set CPPM Configuration (direct 0x0103) */
 struct i40e_aqc_cppm_configuration {
 	__le16	command_flags;
-#define I40E_AQ_CPPM_EN_LTRC	0x0800
-#define I40E_AQ_CPPM_EN_DMCTH	0x1000
-#define I40E_AQ_CPPM_EN_DMCTLX	0x2000
-#define I40E_AQ_CPPM_EN_HPTC	0x4000
-#define I40E_AQ_CPPM_EN_DMARC	0x8000
 	__le16	ttlx;
 	__le32	dmacr;
 	__le16	dmcth;
@@ -466,15 +286,8 @@ I40E_CHECK_CMD_LENGTH(i40e_aqc_cppm_configuration);
 /* Set ARP Proxy command / response (indirect 0x0104) */
 struct i40e_aqc_arp_proxy_data {
 	__le16	command_flags;
-#define I40E_AQ_ARP_INIT_IPV4	0x0800
-#define I40E_AQ_ARP_UNSUP_CTL	0x1000
-#define I40E_AQ_ARP_ENA		0x2000
-#define I40E_AQ_ARP_ADD_IPV4	0x4000
-#define I40E_AQ_ARP_DEL_IPV4	0x8000
 	__le16	table_id;
 	__le32	enabled_offloads;
-#define I40E_AQ_ARP_DIRECTED_OFFLOAD_ENABLE	0x00000020
-#define I40E_AQ_ARP_OFFLOAD_ENABLE		0x00000800
 	__le32	ip_addr;
 	u8	mac_addr[6];
 	u8	reserved[2];
@@ -489,19 +302,6 @@ struct i40e_aqc_ns_proxy_data {
 	__le16	table_idx_ipv6_0;
 	__le16	table_idx_ipv6_1;
 	__le16	control;
-#define I40E_AQ_NS_PROXY_ADD_0		0x0001
-#define I40E_AQ_NS_PROXY_DEL_0		0x0002
-#define I40E_AQ_NS_PROXY_ADD_1		0x0004
-#define I40E_AQ_NS_PROXY_DEL_1		0x0008
-#define I40E_AQ_NS_PROXY_ADD_IPV6_0	0x0010
-#define I40E_AQ_NS_PROXY_DEL_IPV6_0	0x0020
-#define I40E_AQ_NS_PROXY_ADD_IPV6_1	0x0040
-#define I40E_AQ_NS_PROXY_DEL_IPV6_1	0x0080
-#define I40E_AQ_NS_PROXY_COMMAND_SEQ	0x0100
-#define I40E_AQ_NS_PROXY_INIT_IPV6_TBL	0x0200
-#define I40E_AQ_NS_PROXY_INIT_MAC_TBL	0x0400
-#define I40E_AQ_NS_PROXY_OFFLOAD_ENABLE	0x0800
-#define I40E_AQ_NS_PROXY_DIRECTED_OFFLOAD_ENABLE	0x1000
 	u8	mac_addr_0[6];
 	u8	mac_addr_1[6];
 	u8	local_mac_addr[6];
@@ -514,7 +314,6 @@ I40E_CHECK_STRUCT_LEN(0x3c, i40e_aqc_ns_proxy_data);
 /* Manage LAA Command (0x0106) - obsolete */
 struct i40e_aqc_mng_laa {
 	__le16	command_flags;
-#define I40E_AQ_LAA_FLAG_WR	0x8000
 	u8	reserved[2];
 	__le32	sal;
 	__le16	sah;
@@ -527,11 +326,7 @@ I40E_CHECK_CMD_LENGTH(i40e_aqc_mng_laa);
 struct i40e_aqc_mac_address_read {
 	__le16	command_flags;
 #define I40E_AQC_LAN_ADDR_VALID		0x10
-#define I40E_AQC_SAN_ADDR_VALID		0x20
 #define I40E_AQC_PORT_ADDR_VALID	0x40
-#define I40E_AQC_WOL_ADDR_VALID		0x80
-#define I40E_AQC_MC_MAG_EN_VALID	0x100
-#define I40E_AQC_ADDR_VALID_MASK	0x1F0
 	u8	reserved[6];
 	__le32	addr_high;
 	__le32	addr_low;
@@ -555,9 +350,7 @@ struct i40e_aqc_mac_address_write {
 #define I40E_AQC_WOL_PRESERVE_ON_PFR	0x0200
 #define I40E_AQC_WRITE_TYPE_LAA_ONLY	0x0000
 #define I40E_AQC_WRITE_TYPE_LAA_WOL	0x4000
-#define I40E_AQC_WRITE_TYPE_PORT	0x8000
 #define I40E_AQC_WRITE_TYPE_UPDATE_MC_MAG	0xC000
-#define I40E_AQC_WRITE_TYPE_MASK	0xC000
 
 	__le16	mac_sah;
 	__le32	mac_sal;
@@ -580,22 +373,9 @@ I40E_CHECK_CMD_LENGTH(i40e_aqc_clear_pxe);
 
 struct i40e_aqc_set_wol_filter {
 	__le16 filter_index;
-#define I40E_AQC_MAX_NUM_WOL_FILTERS	8
-#define I40E_AQC_SET_WOL_FILTER_TYPE_MAGIC_SHIFT	15
-#define I40E_AQC_SET_WOL_FILTER_TYPE_MAGIC_MASK	(0x1 << \
-		I40E_AQC_SET_WOL_FILTER_TYPE_MAGIC_SHIFT)
-
-#define I40E_AQC_SET_WOL_FILTER_INDEX_SHIFT		0
-#define I40E_AQC_SET_WOL_FILTER_INDEX_MASK	(0x7 << \
-		I40E_AQC_SET_WOL_FILTER_INDEX_SHIFT)
+
 	__le16 cmd_flags;
-#define I40E_AQC_SET_WOL_FILTER				0x8000
-#define I40E_AQC_SET_WOL_FILTER_NO_TCO_WOL		0x4000
-#define I40E_AQC_SET_WOL_FILTER_ACTION_CLEAR		0
-#define I40E_AQC_SET_WOL_FILTER_ACTION_SET		1
 	__le16 valid_flags;
-#define I40E_AQC_SET_WOL_FILTER_ACTION_VALID		0x8000
-#define I40E_AQC_SET_WOL_FILTER_NO_TCO_ACTION_VALID	0x4000
 	u8 reserved[2];
 	__le32	address_high;
 	__le32	address_low;
@@ -615,12 +395,6 @@ I40E_CHECK_STRUCT_LEN(0x90, i40e_aqc_set_wol_filter_data);
 struct i40e_aqc_get_wake_reason_completion {
 	u8 reserved_1[2];
 	__le16 wake_reason;
-#define I40E_AQC_GET_WAKE_UP_REASON_WOL_REASON_MATCHED_INDEX_SHIFT	0
-#define I40E_AQC_GET_WAKE_UP_REASON_WOL_REASON_MATCHED_INDEX_MASK (0xFF << \
-		I40E_AQC_GET_WAKE_UP_REASON_WOL_REASON_MATCHED_INDEX_SHIFT)
-#define I40E_AQC_GET_WAKE_UP_REASON_WOL_REASON_RESERVED_SHIFT	8
-#define I40E_AQC_GET_WAKE_UP_REASON_WOL_REASON_RESERVED_MASK	(0xFF << \
-		I40E_AQC_GET_WAKE_UP_REASON_WOL_REASON_RESERVED_SHIFT)
 	u8 reserved_2[12];
 };
 
@@ -653,25 +427,12 @@ I40E_CHECK_CMD_LENGTH(i40e_aqc_get_switch_config_header_resp);
 
 struct i40e_aqc_switch_config_element_resp {
 	u8	element_type;
-#define I40E_AQ_SW_ELEM_TYPE_MAC	1
-#define I40E_AQ_SW_ELEM_TYPE_PF		2
-#define I40E_AQ_SW_ELEM_TYPE_VF		3
-#define I40E_AQ_SW_ELEM_TYPE_EMP	4
-#define I40E_AQ_SW_ELEM_TYPE_BMC	5
-#define I40E_AQ_SW_ELEM_TYPE_PV		16
-#define I40E_AQ_SW_ELEM_TYPE_VEB	17
-#define I40E_AQ_SW_ELEM_TYPE_PA		18
-#define I40E_AQ_SW_ELEM_TYPE_VSI	19
 	u8	revision;
-#define I40E_AQ_SW_ELEM_REV_1		1
 	__le16	seid;
 	__le16	uplink_seid;
 	__le16	downlink_seid;
 	u8	reserved[3];
 	u8	connection_type;
-#define I40E_AQ_CONN_TYPE_REGULAR	0x1
-#define I40E_AQ_CONN_TYPE_DEFAULT	0x2
-#define I40E_AQ_CONN_TYPE_CASCADED	0x3
 	__le16	scheduler_id;
 	__le16	element_info;
 };
@@ -704,12 +465,7 @@ I40E_CHECK_CMD_LENGTH(i40e_aqc_add_remove_statistics);
 /* Set Port Parameters command (direct 0x0203) */
 struct i40e_aqc_set_port_parameters {
 	__le16	command_flags;
-#define I40E_AQ_SET_P_PARAMS_SAVE_BAD_PACKETS	1
-#define I40E_AQ_SET_P_PARAMS_PAD_SHORT_PACKETS	2 /* must set! */
-#define I40E_AQ_SET_P_PARAMS_DOUBLE_VLAN_ENA	4
 	__le16	bad_frame_vsi;
-#define I40E_AQ_SET_P_PARAMS_BFRAME_SEID_SHIFT	0x0
-#define I40E_AQ_SET_P_PARAMS_BFRAME_SEID_MASK	0x3FF
 	__le16	default_seid;        /* reserved for command */
 	u8	reserved[10];
 };
@@ -729,25 +485,6 @@ I40E_CHECK_CMD_LENGTH(i40e_aqc_get_switch_resource_alloc);
 /* expect an array of these structs in the response buffer */
 struct i40e_aqc_switch_resource_alloc_element_resp {
 	u8	resource_type;
-#define I40E_AQ_RESOURCE_TYPE_VEB		0x0
-#define I40E_AQ_RESOURCE_TYPE_VSI		0x1
-#define I40E_AQ_RESOURCE_TYPE_MACADDR		0x2
-#define I40E_AQ_RESOURCE_TYPE_STAG		0x3
-#define I40E_AQ_RESOURCE_TYPE_ETAG		0x4
-#define I40E_AQ_RESOURCE_TYPE_MULTICAST_HASH	0x5
-#define I40E_AQ_RESOURCE_TYPE_UNICAST_HASH	0x6
-#define I40E_AQ_RESOURCE_TYPE_VLAN		0x7
-#define I40E_AQ_RESOURCE_TYPE_VSI_LIST_ENTRY	0x8
-#define I40E_AQ_RESOURCE_TYPE_ETAG_LIST_ENTRY	0x9
-#define I40E_AQ_RESOURCE_TYPE_VLAN_STAT_POOL	0xA
-#define I40E_AQ_RESOURCE_TYPE_MIRROR_RULE	0xB
-#define I40E_AQ_RESOURCE_TYPE_QUEUE_SETS	0xC
-#define I40E_AQ_RESOURCE_TYPE_VLAN_FILTERS	0xD
-#define I40E_AQ_RESOURCE_TYPE_INNER_MAC_FILTERS	0xF
-#define I40E_AQ_RESOURCE_TYPE_IP_FILTERS	0x10
-#define I40E_AQ_RESOURCE_TYPE_GRE_VN_KEYS	0x11
-#define I40E_AQ_RESOURCE_TYPE_VN2_KEYS		0x12
-#define I40E_AQ_RESOURCE_TYPE_TUNNEL_PORTS	0x13
 	u8	reserved1;
 	__le16	guaranteed;
 	__le16	total;
@@ -763,9 +500,44 @@ struct i40e_aqc_set_switch_config {
 	__le16	flags;
 /* flags used for both fields below */
 #define I40E_AQ_SET_SWITCH_CFG_PROMISC		0x0001
-#define I40E_AQ_SET_SWITCH_CFG_L2_FILTER	0x0002
 	__le16	valid_flags;
-	u8	reserved[12];
+	/* The ethertype in switch_tag is dropped on ingress and used
+	 * internally by the switch. Set this to zero for the default
+	 * of 0x88a8 (802.1ad). Should be zero for firmware API
+	 * versions lower than 1.7.
+	 */
+	__le16	switch_tag;
+	/* The ethertypes in first_tag and second_tag are used to
+	 * match the outer and inner VLAN tags (respectively) when HW
+	 * double VLAN tagging is enabled via the set port parameters
+	 * AQ command. Otherwise these are both ignored. Set them to
+	 * zero for their defaults of 0x8100 (802.1Q). Should be zero
+	 * for firmware API versions lower than 1.7.
+	 */
+	__le16	first_tag;
+	__le16	second_tag;
+	/* Next byte is split into following:
+	 * Bit 7    : 0 : No action, 1: Switch to mode defined by bits 6:0
+	 * Bit 6    : 0 : Destination Port, 1: source port
+	 * Bit 5..4 : L4 type
+	 * 0: rsvd
+	 * 1: TCP
+	 * 2: UDP
+	 * 3: Both TCP and UDP
+	 * Bits 3:0 Mode
+	 * 0: default mode
+	 * 1: L4 port only mode
+	 * 2: non-tunneled mode
+	 * 3: tunneled mode
+	 */
+#define I40E_AQ_SET_SWITCH_BIT7_VALID		0x80
+
+
+#define I40E_AQ_SET_SWITCH_L4_TYPE_TCP		0x10
+
+#define I40E_AQ_SET_SWITCH_MODE_NON_TUNNEL	0x02
+	u8	mode;
+	u8	rsvd5[5];
 };
 
 I40E_CHECK_CMD_LENGTH(i40e_aqc_set_switch_config);
@@ -798,19 +570,13 @@ struct i40e_aqc_add_get_update_vsi {
 	__le16	uplink_seid;
 	u8	connection_type;
 #define I40E_AQ_VSI_CONN_TYPE_NORMAL	0x1
-#define I40E_AQ_VSI_CONN_TYPE_DEFAULT	0x2
-#define I40E_AQ_VSI_CONN_TYPE_CASCADED	0x3
 	u8	reserved1;
 	u8	vf_id;
 	u8	reserved2;
 	__le16	vsi_flags;
-#define I40E_AQ_VSI_TYPE_SHIFT		0x0
-#define I40E_AQ_VSI_TYPE_MASK		(0x3 << I40E_AQ_VSI_TYPE_SHIFT)
 #define I40E_AQ_VSI_TYPE_VF		0x0
 #define I40E_AQ_VSI_TYPE_VMDQ2		0x1
 #define I40E_AQ_VSI_TYPE_PF		0x2
-#define I40E_AQ_VSI_TYPE_EMP_MNG	0x3
-#define I40E_AQ_VSI_FLAG_CASCADED_PV	0x4
 	__le32	addr_high;
 	__le32	addr_low;
 };
@@ -834,24 +600,18 @@ struct i40e_aqc_vsi_properties_data {
 #define I40E_AQ_VSI_PROP_SWITCH_VALID		0x0001
 #define I40E_AQ_VSI_PROP_SECURITY_VALID		0x0002
 #define I40E_AQ_VSI_PROP_VLAN_VALID		0x0004
-#define I40E_AQ_VSI_PROP_CAS_PV_VALID		0x0008
-#define I40E_AQ_VSI_PROP_INGRESS_UP_VALID	0x0010
-#define I40E_AQ_VSI_PROP_EGRESS_UP_VALID	0x0020
 #define I40E_AQ_VSI_PROP_QUEUE_MAP_VALID	0x0040
 #define I40E_AQ_VSI_PROP_QUEUE_OPT_VALID	0x0080
-#define I40E_AQ_VSI_PROP_OUTER_UP_VALID		0x0100
 #define I40E_AQ_VSI_PROP_SCHED_VALID		0x0200
 	/* switch section */
 	__le16	switch_id; /* 12bit id combined with flags below */
 #define I40E_AQ_VSI_SW_ID_SHIFT		0x0000
 #define I40E_AQ_VSI_SW_ID_MASK		(0xFFF << I40E_AQ_VSI_SW_ID_SHIFT)
-#define I40E_AQ_VSI_SW_ID_FLAG_NOT_STAG	0x1000
 #define I40E_AQ_VSI_SW_ID_FLAG_ALLOW_LB	0x2000
 #define I40E_AQ_VSI_SW_ID_FLAG_LOCAL_LB	0x4000
 	u8	sw_reserved[2];
 	/* security section */
 	u8	sec_flags;
-#define I40E_AQ_VSI_SEC_FLAG_ALLOW_DEST_OVRD	0x01
 #define I40E_AQ_VSI_SEC_FLAG_ENABLE_VLAN_CHK	0x02
 #define I40E_AQ_VSI_SEC_FLAG_ENABLE_MAC_CHK	0x04
 	u8	sec_reserved;
@@ -863,78 +623,33 @@ struct i40e_aqc_vsi_properties_data {
 #define I40E_AQ_VSI_PVLAN_MODE_MASK	(0x03 << \
 					 I40E_AQ_VSI_PVLAN_MODE_SHIFT)
 #define I40E_AQ_VSI_PVLAN_MODE_TAGGED	0x01
-#define I40E_AQ_VSI_PVLAN_MODE_UNTAGGED	0x02
 #define I40E_AQ_VSI_PVLAN_MODE_ALL	0x03
 #define I40E_AQ_VSI_PVLAN_INSERT_PVID	0x04
 #define I40E_AQ_VSI_PVLAN_EMOD_SHIFT	0x03
 #define I40E_AQ_VSI_PVLAN_EMOD_MASK	(0x3 << \
 					 I40E_AQ_VSI_PVLAN_EMOD_SHIFT)
 #define I40E_AQ_VSI_PVLAN_EMOD_STR_BOTH	0x0
-#define I40E_AQ_VSI_PVLAN_EMOD_STR_UP	0x08
 #define I40E_AQ_VSI_PVLAN_EMOD_STR	0x10
 #define I40E_AQ_VSI_PVLAN_EMOD_NOTHING	0x18
 	u8	pvlan_reserved[3];
 	/* ingress egress up sections */
 	__le32	ingress_table; /* bitmap, 3 bits per up */
-#define I40E_AQ_VSI_UP_TABLE_UP0_SHIFT	0
-#define I40E_AQ_VSI_UP_TABLE_UP0_MASK	(0x7 << \
-					 I40E_AQ_VSI_UP_TABLE_UP0_SHIFT)
-#define I40E_AQ_VSI_UP_TABLE_UP1_SHIFT	3
-#define I40E_AQ_VSI_UP_TABLE_UP1_MASK	(0x7 << \
-					 I40E_AQ_VSI_UP_TABLE_UP1_SHIFT)
-#define I40E_AQ_VSI_UP_TABLE_UP2_SHIFT	6
-#define I40E_AQ_VSI_UP_TABLE_UP2_MASK	(0x7 << \
-					 I40E_AQ_VSI_UP_TABLE_UP2_SHIFT)
-#define I40E_AQ_VSI_UP_TABLE_UP3_SHIFT	9
-#define I40E_AQ_VSI_UP_TABLE_UP3_MASK	(0x7 << \
-					 I40E_AQ_VSI_UP_TABLE_UP3_SHIFT)
-#define I40E_AQ_VSI_UP_TABLE_UP4_SHIFT	12
-#define I40E_AQ_VSI_UP_TABLE_UP4_MASK	(0x7 << \
-					 I40E_AQ_VSI_UP_TABLE_UP4_SHIFT)
-#define I40E_AQ_VSI_UP_TABLE_UP5_SHIFT	15
-#define I40E_AQ_VSI_UP_TABLE_UP5_MASK	(0x7 << \
-					 I40E_AQ_VSI_UP_TABLE_UP5_SHIFT)
-#define I40E_AQ_VSI_UP_TABLE_UP6_SHIFT	18
-#define I40E_AQ_VSI_UP_TABLE_UP6_MASK	(0x7 << \
-					 I40E_AQ_VSI_UP_TABLE_UP6_SHIFT)
-#define I40E_AQ_VSI_UP_TABLE_UP7_SHIFT	21
-#define I40E_AQ_VSI_UP_TABLE_UP7_MASK	(0x7 << \
-					 I40E_AQ_VSI_UP_TABLE_UP7_SHIFT)
 	__le32	egress_table;   /* same defines as for ingress table */
 	/* cascaded PV section */
 	__le16	cas_pv_tag;
 	u8	cas_pv_flags;
-#define I40E_AQ_VSI_CAS_PV_TAGX_SHIFT		0x00
-#define I40E_AQ_VSI_CAS_PV_TAGX_MASK		(0x03 << \
-						 I40E_AQ_VSI_CAS_PV_TAGX_SHIFT)
-#define I40E_AQ_VSI_CAS_PV_TAGX_LEAVE		0x00
-#define I40E_AQ_VSI_CAS_PV_TAGX_REMOVE		0x01
-#define I40E_AQ_VSI_CAS_PV_TAGX_COPY		0x02
-#define I40E_AQ_VSI_CAS_PV_INSERT_TAG		0x10
-#define I40E_AQ_VSI_CAS_PV_ETAG_PRUNE		0x20
-#define I40E_AQ_VSI_CAS_PV_ACCEPT_HOST_TAG	0x40
 	u8	cas_pv_reserved;
 	/* queue mapping section */
 	__le16	mapping_flags;
 #define I40E_AQ_VSI_QUE_MAP_CONTIG	0x0
 #define I40E_AQ_VSI_QUE_MAP_NONCONTIG	0x1
 	__le16	queue_mapping[16];
-#define I40E_AQ_VSI_QUEUE_SHIFT		0x0
-#define I40E_AQ_VSI_QUEUE_MASK		(0x7FF << I40E_AQ_VSI_QUEUE_SHIFT)
 	__le16	tc_mapping[8];
 #define I40E_AQ_VSI_TC_QUE_OFFSET_SHIFT	0
-#define I40E_AQ_VSI_TC_QUE_OFFSET_MASK	(0x1FF << \
-					 I40E_AQ_VSI_TC_QUE_OFFSET_SHIFT)
 #define I40E_AQ_VSI_TC_QUE_NUMBER_SHIFT	9
-#define I40E_AQ_VSI_TC_QUE_NUMBER_MASK	(0x7 << \
-					 I40E_AQ_VSI_TC_QUE_NUMBER_SHIFT)
 	/* queueing option section */
 	u8	queueing_opt_flags;
-#define I40E_AQ_VSI_QUE_OPT_MULTICAST_UDP_ENA	0x04
-#define I40E_AQ_VSI_QUE_OPT_UNICAST_UDP_ENA	0x08
 #define I40E_AQ_VSI_QUE_OPT_TCP_ENA	0x10
-#define I40E_AQ_VSI_QUE_OPT_FCOE_ENA	0x20
-#define I40E_AQ_VSI_QUE_OPT_RSS_LUT_PF	0x00
 #define I40E_AQ_VSI_QUE_OPT_RSS_LUT_VSI	0x40
 	u8	queueing_opt_reserved[3];
 	/* scheduler section */
@@ -959,10 +674,6 @@ I40E_CHECK_STRUCT_LEN(128, i40e_aqc_vsi_properties_data);
  */
 struct i40e_aqc_add_update_pv {
 	__le16	command_flags;
-#define I40E_AQC_PV_FLAG_PV_TYPE		0x1
-#define I40E_AQC_PV_FLAG_FWD_UNKNOWN_STAG_EN	0x2
-#define I40E_AQC_PV_FLAG_FWD_UNKNOWN_ETAG_EN	0x4
-#define I40E_AQC_PV_FLAG_IS_CTRL_PORT		0x8
 	__le16	uplink_seid;
 	__le16	connected_seid;
 	u8	reserved[10];
@@ -973,10 +684,6 @@ I40E_CHECK_CMD_LENGTH(i40e_aqc_add_update_pv);
 struct i40e_aqc_add_update_pv_completion {
 	/* reserved for update; for add also encodes error if rc == ENOSPC */
 	__le16	pv_seid;
-#define I40E_AQC_PV_ERR_FLAG_NO_PV	0x1
-#define I40E_AQC_PV_ERR_FLAG_NO_SCHED	0x2
-#define I40E_AQC_PV_ERR_FLAG_NO_COUNTER	0x4
-#define I40E_AQC_PV_ERR_FLAG_NO_ENTRY	0x8
 	u8	reserved[14];
 };
 
@@ -990,9 +697,6 @@ struct i40e_aqc_get_pv_params_completion {
 	__le16	seid;
 	__le16	default_stag;
 	__le16	pv_flags; /* same flags as add_pv */
-#define I40E_AQC_GET_PV_PV_TYPE			0x1
-#define I40E_AQC_GET_PV_FRWD_UNKNOWN_STAG	0x2
-#define I40E_AQC_GET_PV_FRWD_UNKNOWN_ETAG	0x4
 	u8	reserved[8];
 	__le16	default_port_seid;
 };
@@ -1005,12 +709,8 @@ struct i40e_aqc_add_veb {
 	__le16	downlink_seid;
 	__le16	veb_flags;
 #define I40E_AQC_ADD_VEB_FLOATING		0x1
-#define I40E_AQC_ADD_VEB_PORT_TYPE_SHIFT	1
-#define I40E_AQC_ADD_VEB_PORT_TYPE_MASK		(0x3 << \
-					I40E_AQC_ADD_VEB_PORT_TYPE_SHIFT)
 #define I40E_AQC_ADD_VEB_PORT_TYPE_DEFAULT	0x2
 #define I40E_AQC_ADD_VEB_PORT_TYPE_DATA		0x4
-#define I40E_AQC_ADD_VEB_ENABLE_L2_FILTER	0x8     /* deprecated */
 #define I40E_AQC_ADD_VEB_ENABLE_DISABLE_STATS	0x10
 	u8	enable_tcs;
 	u8	reserved[9];
@@ -1023,10 +723,6 @@ struct i40e_aqc_add_veb_completion {
 	__le16	switch_seid;
 	/* also encodes error if rc == ENOSPC; codes are the same as add_pv */
 	__le16	veb_seid;
-#define I40E_AQC_VEB_ERR_FLAG_NO_VEB		0x1
-#define I40E_AQC_VEB_ERR_FLAG_NO_SCHED		0x2
-#define I40E_AQC_VEB_ERR_FLAG_NO_COUNTER	0x4
-#define I40E_AQC_VEB_ERR_FLAG_NO_ENTRY		0x8
 	__le16	statistic_index;
 	__le16	vebs_used;
 	__le16	vebs_free;
@@ -1059,9 +755,6 @@ I40E_CHECK_CMD_LENGTH(i40e_aqc_get_veb_parameters_completion);
 struct i40e_aqc_macvlan {
 	__le16	num_addresses;
 	__le16	seid[3];
-#define I40E_AQC_MACVLAN_CMD_SEID_NUM_SHIFT	0
-#define I40E_AQC_MACVLAN_CMD_SEID_NUM_MASK	(0x3FF << \
-					I40E_AQC_MACVLAN_CMD_SEID_NUM_SHIFT)
 #define I40E_AQC_MACVLAN_CMD_SEID_VALID		0x8000
 	__le32	addr_high;
 	__le32	addr_low;
@@ -1075,18 +768,11 @@ struct i40e_aqc_add_macvlan_element_data {
 	__le16	vlan_tag;
 	__le16	flags;
 #define I40E_AQC_MACVLAN_ADD_PERFECT_MATCH	0x0001
-#define I40E_AQC_MACVLAN_ADD_HASH_MATCH		0x0002
 #define I40E_AQC_MACVLAN_ADD_IGNORE_VLAN	0x0004
-#define I40E_AQC_MACVLAN_ADD_TO_QUEUE		0x0008
 #define I40E_AQC_MACVLAN_ADD_USE_SHARED_MAC	0x0010
 	__le16	queue_number;
-#define I40E_AQC_MACVLAN_CMD_QUEUE_SHIFT	0
-#define I40E_AQC_MACVLAN_CMD_QUEUE_MASK		(0x7FF << \
-					I40E_AQC_MACVLAN_CMD_SEID_NUM_SHIFT)
 	/* response section */
 	u8	match_method;
-#define I40E_AQC_MM_PERFECT_MATCH	0x01
-#define I40E_AQC_MM_HASH_MATCH		0x02
 #define I40E_AQC_MM_ERR_NO_RES		0xFF
 	u8	reserved1[3];
 };
@@ -1112,14 +798,10 @@ struct i40e_aqc_remove_macvlan_element_data {
 	__le16	vlan_tag;
 	u8	flags;
 #define I40E_AQC_MACVLAN_DEL_PERFECT_MATCH	0x01
-#define I40E_AQC_MACVLAN_DEL_HASH_MATCH		0x02
 #define I40E_AQC_MACVLAN_DEL_IGNORE_VLAN	0x08
-#define I40E_AQC_MACVLAN_DEL_ALL_VSIS		0x10
 	u8	reserved[3];
 	/* reply section */
 	u8	error_code;
-#define I40E_AQC_REMOVE_MACVLAN_SUCCESS		0x0
-#define I40E_AQC_REMOVE_MACVLAN_FAIL		0xFF
 	u8	reply_reserved[3];
 };
 
@@ -1130,30 +812,8 @@ struct i40e_aqc_remove_macvlan_element_data {
 struct i40e_aqc_add_remove_vlan_element_data {
 	__le16	vlan_tag;
 	u8	vlan_flags;
-/* flags for add VLAN */
-#define I40E_AQC_ADD_VLAN_LOCAL			0x1
-#define I40E_AQC_ADD_PVLAN_TYPE_SHIFT		1
-#define I40E_AQC_ADD_PVLAN_TYPE_MASK	(0x3 << I40E_AQC_ADD_PVLAN_TYPE_SHIFT)
-#define I40E_AQC_ADD_PVLAN_TYPE_REGULAR		0x0
-#define I40E_AQC_ADD_PVLAN_TYPE_PRIMARY		0x2
-#define I40E_AQC_ADD_PVLAN_TYPE_SECONDARY	0x4
-#define I40E_AQC_VLAN_PTYPE_SHIFT		3
-#define I40E_AQC_VLAN_PTYPE_MASK	(0x3 << I40E_AQC_VLAN_PTYPE_SHIFT)
-#define I40E_AQC_VLAN_PTYPE_REGULAR_VSI		0x0
-#define I40E_AQC_VLAN_PTYPE_PROMISC_VSI		0x8
-#define I40E_AQC_VLAN_PTYPE_COMMUNITY_VSI	0x10
-#define I40E_AQC_VLAN_PTYPE_ISOLATED_VSI	0x18
-/* flags for remove VLAN */
-#define I40E_AQC_REMOVE_VLAN_ALL	0x1
 	u8	reserved;
 	u8	result;
-/* flags for add VLAN */
-#define I40E_AQC_ADD_VLAN_SUCCESS	0x0
-#define I40E_AQC_ADD_VLAN_FAIL_REQUEST	0xFE
-#define I40E_AQC_ADD_VLAN_FAIL_RESOURCE	0xFF
-/* flags for remove VLAN */
-#define I40E_AQC_REMOVE_VLAN_SUCCESS	0x0
-#define I40E_AQC_REMOVE_VLAN_FAIL	0xFF
 	u8	reserved1[3];
 };
 
@@ -1175,11 +835,9 @@ struct i40e_aqc_set_vsi_promiscuous_modes {
 #define I40E_AQC_SET_VSI_PROMISC_BROADCAST	0x04
 #define I40E_AQC_SET_VSI_DEFAULT		0x08
 #define I40E_AQC_SET_VSI_PROMISC_VLAN		0x10
-#define I40E_AQC_SET_VSI_PROMISC_TX		0x8000
+#define I40E_AQC_SET_VSI_PROMISC_RX_ONLY	0x8000
 	__le16	seid;
-#define I40E_AQC_VSI_PROM_CMD_SEID_MASK		0x3FF
 	__le16	vlan_tag;
-#define I40E_AQC_SET_VSI_VLAN_MASK		0x0FFF
 #define I40E_AQC_SET_VSI_VLAN_VALID		0x8000
 	u8	reserved[8];
 };
@@ -1191,11 +849,7 @@ I40E_CHECK_CMD_LENGTH(i40e_aqc_set_vsi_promiscuous_modes);
  */
 struct i40e_aqc_add_tag {
 	__le16	flags;
-#define I40E_AQC_ADD_TAG_FLAG_TO_QUEUE		0x0001
 	__le16	seid;
-#define I40E_AQC_ADD_TAG_CMD_SEID_NUM_SHIFT	0
-#define I40E_AQC_ADD_TAG_CMD_SEID_NUM_MASK	(0x3FF << \
-					I40E_AQC_ADD_TAG_CMD_SEID_NUM_SHIFT)
 	__le16	tag;
 	__le16	queue_number;
 	u8	reserved[8];
@@ -1216,9 +870,6 @@ I40E_CHECK_CMD_LENGTH(i40e_aqc_add_remove_tag_completion);
  */
 struct i40e_aqc_remove_tag {
 	__le16	seid;
-#define I40E_AQC_REMOVE_TAG_CMD_SEID_NUM_SHIFT	0
-#define I40E_AQC_REMOVE_TAG_CMD_SEID_NUM_MASK	(0x3FF << \
-					I40E_AQC_REMOVE_TAG_CMD_SEID_NUM_SHIFT)
 	__le16	tag;
 	u8	reserved[12];
 };
@@ -1254,9 +905,6 @@ I40E_CHECK_CMD_LENGTH(i40e_aqc_add_remove_mcast_etag_completion);
 /* Update S/E-Tag (direct 0x0259) */
 struct i40e_aqc_update_tag {
 	__le16	seid;
-#define I40E_AQC_UPDATE_TAG_CMD_SEID_NUM_SHIFT	0
-#define I40E_AQC_UPDATE_TAG_CMD_SEID_NUM_MASK	(0x3FF << \
-					I40E_AQC_UPDATE_TAG_CMD_SEID_NUM_SHIFT)
 	__le16	old_tag;
 	__le16	new_tag;
 	u8	reserved[10];
@@ -1283,13 +931,9 @@ struct i40e_aqc_add_remove_control_packet_filter {
 	__le16	flags;
 #define I40E_AQC_ADD_CONTROL_PACKET_FLAGS_IGNORE_MAC	0x0001
 #define I40E_AQC_ADD_CONTROL_PACKET_FLAGS_DROP		0x0002
-#define I40E_AQC_ADD_CONTROL_PACKET_FLAGS_TO_QUEUE	0x0004
 #define I40E_AQC_ADD_CONTROL_PACKET_FLAGS_TX		0x0008
 #define I40E_AQC_ADD_CONTROL_PACKET_FLAGS_RX		0x0000
 	__le16	seid;
-#define I40E_AQC_ADD_CONTROL_PACKET_CMD_SEID_NUM_SHIFT	0
-#define I40E_AQC_ADD_CONTROL_PACKET_CMD_SEID_NUM_MASK	(0x3FF << \
-				I40E_AQC_ADD_CONTROL_PACKET_CMD_SEID_NUM_SHIFT)
 	__le16	queue;
 	u8	reserved[2];
 };
@@ -1315,17 +959,16 @@ struct i40e_aqc_add_remove_cloud_filters {
 	u8	num_filters;
 	u8	reserved;
 	__le16	seid;
-#define I40E_AQC_ADD_CLOUD_CMD_SEID_NUM_SHIFT	0
-#define I40E_AQC_ADD_CLOUD_CMD_SEID_NUM_MASK	(0x3FF << \
-					I40E_AQC_ADD_CLOUD_CMD_SEID_NUM_SHIFT)
-	u8	reserved2[4];
+	u8	big_buffer_flag;
+#define I40E_AQC_ADD_CLOUD_CMD_BB	1
+	u8	reserved2[3];
 	__le32	addr_high;
 	__le32	addr_low;
 };
 
 I40E_CHECK_CMD_LENGTH(i40e_aqc_add_remove_cloud_filters);
 
-struct i40e_aqc_add_remove_cloud_filters_element_data {
+struct i40e_aqc_cloud_filters_element_data {
 	u8	outer_mac[6];
 	u8	inner_mac[6];
 	__le16	inner_vlan;
@@ -1337,13 +980,13 @@ struct i40e_aqc_add_remove_cloud_filters_element_data {
 		struct {
 			u8 data[16];
 		} v6;
+		struct {
+			__le16 data[8];
+		} raw_v6;
 	} ipaddr;
 	__le16	flags;
-#define I40E_AQC_ADD_CLOUD_FILTER_SHIFT			0
-#define I40E_AQC_ADD_CLOUD_FILTER_MASK	(0x3F << \
-					I40E_AQC_ADD_CLOUD_FILTER_SHIFT)
 /* 0x0000 reserved */
-#define I40E_AQC_ADD_CLOUD_FILTER_OIP			0x0001
+/* 0x0001 reserved */
 /* 0x0002 reserved */
 #define I40E_AQC_ADD_CLOUD_FILTER_IMAC_IVLAN		0x0003
 #define I40E_AQC_ADD_CLOUD_FILTER_IMAC_IVLAN_TEN_ID	0x0004
@@ -1355,40 +998,44 @@ struct i40e_aqc_add_remove_cloud_filters_element_data {
 #define I40E_AQC_ADD_CLOUD_FILTER_IMAC			0x000A
 #define I40E_AQC_ADD_CLOUD_FILTER_OMAC_TEN_ID_IMAC	0x000B
 #define I40E_AQC_ADD_CLOUD_FILTER_IIP			0x000C
+/* 0x000D reserved */
+/* 0x000E reserved */
+/* 0x000F reserved */
+/* 0x0010 to 0x0017 is for custom filters */
+#define I40E_AQC_ADD_CLOUD_FILTER_IP_PORT		0x0010 /* Dest IP + L4 Port */
+#define I40E_AQC_ADD_CLOUD_FILTER_MAC_PORT		0x0011 /* Dest MAC + L4 Port */
+#define I40E_AQC_ADD_CLOUD_FILTER_MAC_VLAN_PORT		0x0012 /* Dest MAC + VLAN + L4 Port */
 
-#define I40E_AQC_ADD_CLOUD_FLAGS_TO_QUEUE		0x0080
-#define I40E_AQC_ADD_CLOUD_VNK_SHIFT			6
-#define I40E_AQC_ADD_CLOUD_VNK_MASK			0x00C0
 #define I40E_AQC_ADD_CLOUD_FLAGS_IPV4			0
 #define I40E_AQC_ADD_CLOUD_FLAGS_IPV6			0x0100
 
 #define I40E_AQC_ADD_CLOUD_TNL_TYPE_SHIFT		9
 #define I40E_AQC_ADD_CLOUD_TNL_TYPE_MASK		0x1E00
-#define I40E_AQC_ADD_CLOUD_TNL_TYPE_VXLAN		0
-#define I40E_AQC_ADD_CLOUD_TNL_TYPE_NVGRE_OMAC		1
 #define I40E_AQC_ADD_CLOUD_TNL_TYPE_GENEVE		2
-#define I40E_AQC_ADD_CLOUD_TNL_TYPE_IP			3
-#define I40E_AQC_ADD_CLOUD_TNL_TYPE_RESERVED		4
-#define I40E_AQC_ADD_CLOUD_TNL_TYPE_VXLAN_GPE		5
 
-#define I40E_AQC_ADD_CLOUD_FLAGS_SHARED_OUTER_MAC	0x2000
-#define I40E_AQC_ADD_CLOUD_FLAGS_SHARED_INNER_MAC	0x4000
-#define I40E_AQC_ADD_CLOUD_FLAGS_SHARED_OUTER_IP	0x8000
 
 	__le32	tenant_id;
 	u8	reserved[4];
 	__le16	queue_number;
-#define I40E_AQC_ADD_CLOUD_QUEUE_SHIFT		0
-#define I40E_AQC_ADD_CLOUD_QUEUE_MASK		(0x7FF << \
-						 I40E_AQC_ADD_CLOUD_QUEUE_SHIFT)
 	u8	reserved2[14];
 	/* response section */
 	u8	allocation_result;
-#define I40E_AQC_ADD_CLOUD_FILTER_SUCCESS	0x0
-#define I40E_AQC_ADD_CLOUD_FILTER_FAIL		0xFF
 	u8	response_reserved[7];
 };
 
+I40E_CHECK_STRUCT_LEN(0x40, i40e_aqc_cloud_filters_element_data);
+
+/* i40e_aqc_cloud_filters_element_bb is used when
+ * I40E_AQC_CLOUD_CMD_BB flag is set.
+ */
+struct i40e_aqc_cloud_filters_element_bb {
+	struct i40e_aqc_cloud_filters_element_data element;
+	u16     general_fields[32];
+#define I40E_AQC_ADD_CLOUD_FV_FLU_0X16_WORD0	15
+};
+
+I40E_CHECK_STRUCT_LEN(0x80, i40e_aqc_cloud_filters_element_bb);
+
 struct i40e_aqc_remove_cloud_filters_completion {
 	__le16 perfect_ovlan_used;
 	__le16 perfect_ovlan_free;
@@ -1400,6 +1047,36 @@ struct i40e_aqc_remove_cloud_filters_completion {
 
 I40E_CHECK_CMD_LENGTH(i40e_aqc_remove_cloud_filters_completion);
 
+/* Replace filter Command 0x025F
+ * uses the i40e_aqc_replace_cloud_filters,
+ * and the generic indirect completion structure
+ */
+struct i40e_filter_data {
+	u8 filter_type;
+	u8 input[3];
+};
+
+I40E_CHECK_STRUCT_LEN(4, i40e_filter_data);
+
+struct i40e_aqc_replace_cloud_filters_cmd {
+	u8      valid_flags;
+	u8      old_filter_type;
+	u8      new_filter_type;
+	u8      tr_bit;
+	u8      reserved[4];
+	__le32 addr_high;
+	__le32 addr_low;
+};
+
+I40E_CHECK_CMD_LENGTH(i40e_aqc_replace_cloud_filters_cmd);
+
+struct i40e_aqc_replace_cloud_filters_cmd_buf {
+	u8      data[32];
+	struct i40e_filter_data filters[8];
+};
+
+I40E_CHECK_STRUCT_LEN(0x40, i40e_aqc_replace_cloud_filters_cmd_buf);
+
 /* Add Mirror Rule (indirect or direct 0x0260)
  * Delete Mirror Rule (indirect or direct 0x0261)
  * note: some rule types (4,5) do not use an external buffer.
@@ -1411,8 +1088,6 @@ struct i40e_aqc_add_delete_mirror_rule {
 #define I40E_AQC_MIRROR_RULE_TYPE_SHIFT		0
 #define I40E_AQC_MIRROR_RULE_TYPE_MASK		(0x7 << \
 						I40E_AQC_MIRROR_RULE_TYPE_SHIFT)
-#define I40E_AQC_MIRROR_RULE_TYPE_VPORT_INGRESS	1
-#define I40E_AQC_MIRROR_RULE_TYPE_VPORT_EGRESS	2
 #define I40E_AQC_MIRROR_RULE_TYPE_VLAN		3
 #define I40E_AQC_MIRROR_RULE_TYPE_ALL_INGRESS	4
 #define I40E_AQC_MIRROR_RULE_TYPE_ALL_EGRESS	5
@@ -1435,7 +1110,7 @@ struct i40e_aqc_add_delete_mirror_rule_completion {
 
 I40E_CHECK_CMD_LENGTH(i40e_aqc_add_delete_mirror_rule_completion);
 
-/* Pipeline Personalization Profile */
+/* Dynamic Device Personalization */
 struct i40e_aqc_write_personalization_profile {
 	u8      flags;
 	u8      reserved[3];
@@ -1446,7 +1121,7 @@ struct i40e_aqc_write_personalization_profile {
 
 I40E_CHECK_CMD_LENGTH(i40e_aqc_write_personalization_profile);
 
-struct i40e_aqc_write_ppp_resp {
+struct i40e_aqc_write_ddp_resp {
 	__le32 error_offset;
 	__le32 error_info;
 	__le32 addr_high;
@@ -1455,8 +1130,6 @@ struct i40e_aqc_write_ppp_resp {
 
 struct i40e_aqc_get_applied_profiles {
 	u8      flags;
-#define I40E_AQC_GET_PPP_GET_CONF	0x1
-#define I40E_AQC_GET_PPP_GET_RDPU_CONF	0x2
 	u8      rsv[3];
 	__le32  reserved;
 	__le32  addr_high;
@@ -1473,8 +1146,6 @@ I40E_CHECK_CMD_LENGTH(i40e_aqc_get_applied_profiles);
 struct i40e_aqc_pfc_ignore {
 	u8	tc_bitmap;
 	u8	command_flags; /* unused on response */
-#define I40E_AQC_PFC_IGNORE_SET		0x80
-#define I40E_AQC_PFC_IGNORE_CLEAR	0x0
 	u8	reserved[14];
 };
 
@@ -1591,7 +1262,6 @@ struct i40e_aqc_configure_switching_comp_ets_data {
 	u8	reserved[4];
 	u8	tc_valid_bits;
 	u8	seepage;
-#define I40E_AQ_ETS_SEEPAGE_EN_MASK	0x1
 	u8	tc_strict_priority_flags;
 	u8	reserved1[17];
 	u8	tc_bw_share_credits[8];
@@ -1726,6 +1396,8 @@ enum i40e_aq_phy_type {
 	I40E_PHY_TYPE_10GBASE_CR1_CU		= 0xB,
 	I40E_PHY_TYPE_10GBASE_AOC		= 0xC,
 	I40E_PHY_TYPE_40GBASE_AOC		= 0xD,
+	I40E_PHY_TYPE_UNRECOGNIZED		= 0xE,
+	I40E_PHY_TYPE_UNSUPPORTED		= 0xF,
 	I40E_PHY_TYPE_100BASE_TX		= 0x11,
 	I40E_PHY_TYPE_1000BASE_T		= 0x12,
 	I40E_PHY_TYPE_10GBASE_T			= 0x13,
@@ -1744,20 +1416,72 @@ enum i40e_aq_phy_type {
 	I40E_PHY_TYPE_25GBASE_CR		= 0x20,
 	I40E_PHY_TYPE_25GBASE_SR		= 0x21,
 	I40E_PHY_TYPE_25GBASE_LR		= 0x22,
-	I40E_PHY_TYPE_MAX
-};
-
+	I40E_PHY_TYPE_25GBASE_AOC		= 0x23,
+	I40E_PHY_TYPE_25GBASE_ACC		= 0x24,
+	I40E_PHY_TYPE_2_5GBASE_T		= 0x26,
+	I40E_PHY_TYPE_5GBASE_T			= 0x27,
+	I40E_PHY_TYPE_2_5GBASE_T_LINK_STATUS	= 0x30,
+	I40E_PHY_TYPE_5GBASE_T_LINK_STATUS	= 0x31,
+	I40E_PHY_TYPE_MAX,
+	I40E_PHY_TYPE_NOT_SUPPORTED_HIGH_TEMP	= 0xFD,
+	I40E_PHY_TYPE_EMPTY			= 0xFE,
+	I40E_PHY_TYPE_DEFAULT			= 0xFF,
+};
+
+#define I40E_PHY_TYPES_BITMASK (BIT_ULL(I40E_PHY_TYPE_SGMII) | \
+				BIT_ULL(I40E_PHY_TYPE_1000BASE_KX) | \
+				BIT_ULL(I40E_PHY_TYPE_10GBASE_KX4) | \
+				BIT_ULL(I40E_PHY_TYPE_10GBASE_KR) | \
+				BIT_ULL(I40E_PHY_TYPE_40GBASE_KR4) | \
+				BIT_ULL(I40E_PHY_TYPE_XAUI) | \
+				BIT_ULL(I40E_PHY_TYPE_XFI) | \
+				BIT_ULL(I40E_PHY_TYPE_SFI) | \
+				BIT_ULL(I40E_PHY_TYPE_XLAUI) | \
+				BIT_ULL(I40E_PHY_TYPE_XLPPI) | \
+				BIT_ULL(I40E_PHY_TYPE_40GBASE_CR4_CU) | \
+				BIT_ULL(I40E_PHY_TYPE_10GBASE_CR1_CU) | \
+				BIT_ULL(I40E_PHY_TYPE_10GBASE_AOC) | \
+				BIT_ULL(I40E_PHY_TYPE_40GBASE_AOC) | \
+				BIT_ULL(I40E_PHY_TYPE_UNRECOGNIZED) | \
+				BIT_ULL(I40E_PHY_TYPE_UNSUPPORTED) | \
+				BIT_ULL(I40E_PHY_TYPE_100BASE_TX) | \
+				BIT_ULL(I40E_PHY_TYPE_1000BASE_T) | \
+				BIT_ULL(I40E_PHY_TYPE_10GBASE_T) | \
+				BIT_ULL(I40E_PHY_TYPE_10GBASE_SR) | \
+				BIT_ULL(I40E_PHY_TYPE_10GBASE_LR) | \
+				BIT_ULL(I40E_PHY_TYPE_10GBASE_SFPP_CU) | \
+				BIT_ULL(I40E_PHY_TYPE_10GBASE_CR1) | \
+				BIT_ULL(I40E_PHY_TYPE_40GBASE_CR4) | \
+				BIT_ULL(I40E_PHY_TYPE_40GBASE_SR4) | \
+				BIT_ULL(I40E_PHY_TYPE_40GBASE_LR4) | \
+				BIT_ULL(I40E_PHY_TYPE_1000BASE_SX) | \
+				BIT_ULL(I40E_PHY_TYPE_1000BASE_LX) | \
+				BIT_ULL(I40E_PHY_TYPE_1000BASE_T_OPTICAL) | \
+				BIT_ULL(I40E_PHY_TYPE_20GBASE_KR2) | \
+				BIT_ULL(I40E_PHY_TYPE_25GBASE_KR) | \
+				BIT_ULL(I40E_PHY_TYPE_25GBASE_CR) | \
+				BIT_ULL(I40E_PHY_TYPE_25GBASE_SR) | \
+				BIT_ULL(I40E_PHY_TYPE_25GBASE_LR) | \
+				BIT_ULL(I40E_PHY_TYPE_25GBASE_AOC) | \
+				BIT_ULL(I40E_PHY_TYPE_25GBASE_ACC) | \
+				BIT_ULL(I40E_PHY_TYPE_2_5GBASE_T) | \
+				BIT_ULL(I40E_PHY_TYPE_5GBASE_T))
+
+#define I40E_LINK_SPEED_2_5GB_SHIFT	0x0
 #define I40E_LINK_SPEED_100MB_SHIFT	0x1
 #define I40E_LINK_SPEED_1000MB_SHIFT	0x2
 #define I40E_LINK_SPEED_10GB_SHIFT	0x3
 #define I40E_LINK_SPEED_40GB_SHIFT	0x4
 #define I40E_LINK_SPEED_20GB_SHIFT	0x5
 #define I40E_LINK_SPEED_25GB_SHIFT	0x6
+#define I40E_LINK_SPEED_5GB_SHIFT	0x7
 
 enum i40e_aq_link_speed {
 	I40E_LINK_SPEED_UNKNOWN	= 0,
 	I40E_LINK_SPEED_100MB	= BIT(I40E_LINK_SPEED_100MB_SHIFT),
 	I40E_LINK_SPEED_1GB	= BIT(I40E_LINK_SPEED_1000MB_SHIFT),
+	I40E_LINK_SPEED_2_5GB	= (1 << I40E_LINK_SPEED_2_5GB_SHIFT),
+	I40E_LINK_SPEED_5GB	= (1 << I40E_LINK_SPEED_5GB_SHIFT),
 	I40E_LINK_SPEED_10GB	= BIT(I40E_LINK_SPEED_10GB_SHIFT),
 	I40E_LINK_SPEED_40GB	= BIT(I40E_LINK_SPEED_40GB_SHIFT),
 	I40E_LINK_SPEED_20GB	= BIT(I40E_LINK_SPEED_20GB_SHIFT),
@@ -1780,36 +1504,18 @@ struct i40e_aq_get_phy_abilities_resp {
 	u8	abilities;
 #define I40E_AQ_PHY_FLAG_PAUSE_TX	0x01
 #define I40E_AQ_PHY_FLAG_PAUSE_RX	0x02
-#define I40E_AQ_PHY_FLAG_LOW_POWER	0x04
-#define I40E_AQ_PHY_LINK_ENABLED	0x08
-#define I40E_AQ_PHY_AN_ENABLED		0x10
-#define I40E_AQ_PHY_FLAG_MODULE_QUAL	0x20
-#define I40E_AQ_PHY_FEC_ABILITY_KR	0x40
-#define I40E_AQ_PHY_FEC_ABILITY_RS	0x80
 	__le16	eee_capability;
-#define I40E_AQ_EEE_100BASE_TX		0x0002
-#define I40E_AQ_EEE_1000BASE_T		0x0004
-#define I40E_AQ_EEE_10GBASE_T		0x0008
-#define I40E_AQ_EEE_1000BASE_KX		0x0010
-#define I40E_AQ_EEE_10GBASE_KX4		0x0020
-#define I40E_AQ_EEE_10GBASE_KR		0x0040
 	__le32	eeer_val;
 	u8	d3_lpan;
-#define I40E_AQ_SET_PHY_D3_LPAN_ENA	0x01
 	u8	phy_type_ext;
 #define I40E_AQ_PHY_TYPE_EXT_25G_KR	0X01
 #define I40E_AQ_PHY_TYPE_EXT_25G_CR	0X02
 #define I40E_AQ_PHY_TYPE_EXT_25G_SR	0x04
 #define I40E_AQ_PHY_TYPE_EXT_25G_LR	0x08
 	u8	fec_cfg_curr_mod_ext_info;
-#define I40E_AQ_ENABLE_FEC_KR		0x01
-#define I40E_AQ_ENABLE_FEC_RS		0x02
 #define I40E_AQ_REQUEST_FEC_KR		0x04
 #define I40E_AQ_REQUEST_FEC_RS		0x08
 #define I40E_AQ_ENABLE_FEC_AUTO		0x10
-#define I40E_AQ_FEC
-#define I40E_AQ_MODULE_TYPE_EXT_MASK	0xE0
-#define I40E_AQ_MODULE_TYPE_EXT_SHIFT	5
 
 	u8	ext_comp_code;
 	u8	phy_id[4];
@@ -1855,20 +1561,7 @@ I40E_CHECK_CMD_LENGTH(i40e_aq_set_phy_config);
 struct i40e_aq_set_mac_config {
 	__le16	max_frame_size;
 	u8	params;
-#define I40E_AQ_SET_MAC_CONFIG_CRC_EN		0x04
-#define I40E_AQ_SET_MAC_CONFIG_PACING_MASK	0x78
-#define I40E_AQ_SET_MAC_CONFIG_PACING_SHIFT	3
-#define I40E_AQ_SET_MAC_CONFIG_PACING_NONE	0x0
-#define I40E_AQ_SET_MAC_CONFIG_PACING_1B_13TX	0xF
-#define I40E_AQ_SET_MAC_CONFIG_PACING_1DW_9TX	0x9
-#define I40E_AQ_SET_MAC_CONFIG_PACING_1DW_4TX	0x8
-#define I40E_AQ_SET_MAC_CONFIG_PACING_3DW_7TX	0x7
-#define I40E_AQ_SET_MAC_CONFIG_PACING_2DW_3TX	0x6
-#define I40E_AQ_SET_MAC_CONFIG_PACING_1DW_1TX	0x5
-#define I40E_AQ_SET_MAC_CONFIG_PACING_3DW_2TX	0x4
-#define I40E_AQ_SET_MAC_CONFIG_PACING_7DW_3TX	0x3
-#define I40E_AQ_SET_MAC_CONFIG_PACING_4DW_1TX	0x2
-#define I40E_AQ_SET_MAC_CONFIG_PACING_9DW_1TX	0x1
+#define I40E_AQ_SET_MAC_CONFIG_CRC_EN	BIT(2)
 	u8	tx_timer_priority; /* bitmap */
 	__le16	tx_timer_value;
 	__le16	fc_refresh_threshold;
@@ -1890,8 +1583,6 @@ I40E_CHECK_CMD_LENGTH(i40e_aqc_set_link_restart_an);
 /* Get Link Status cmd & response data structure (direct 0x0607) */
 struct i40e_aqc_get_link_status {
 	__le16	command_flags; /* only field set on command */
-#define I40E_AQ_LSE_MASK		0x3
-#define I40E_AQ_LSE_NOP			0x0
 #define I40E_AQ_LSE_DISABLE		0x2
 #define I40E_AQ_LSE_ENABLE		0x3
 /* only response uses this flag */
@@ -1900,53 +1591,32 @@ struct i40e_aqc_get_link_status {
 	u8	link_speed;  /* i40e_aq_link_speed */
 	u8	link_info;
 #define I40E_AQ_LINK_UP			0x01    /* obsolete */
-#define I40E_AQ_LINK_UP_FUNCTION	0x01
-#define I40E_AQ_LINK_FAULT		0x02
-#define I40E_AQ_LINK_FAULT_TX		0x04
-#define I40E_AQ_LINK_FAULT_RX		0x08
-#define I40E_AQ_LINK_FAULT_REMOTE	0x10
-#define I40E_AQ_LINK_UP_PORT		0x20
 #define I40E_AQ_MEDIA_AVAILABLE		0x40
-#define I40E_AQ_SIGNAL_DETECT		0x80
 	u8	an_info;
 #define I40E_AQ_AN_COMPLETED		0x01
-#define I40E_AQ_LP_AN_ABILITY		0x02
-#define I40E_AQ_PD_FAULT		0x04
-#define I40E_AQ_FEC_EN			0x08
-#define I40E_AQ_PHY_LOW_POWER		0x10
 #define I40E_AQ_LINK_PAUSE_TX		0x20
 #define I40E_AQ_LINK_PAUSE_RX		0x40
 #define I40E_AQ_QUALIFIED_MODULE	0x80
 	u8	ext_info;
-#define I40E_AQ_LINK_PHY_TEMP_ALARM	0x01
-#define I40E_AQ_LINK_XCESSIVE_ERRORS	0x02
-#define I40E_AQ_LINK_TX_SHIFT		0x02
-#define I40E_AQ_LINK_TX_MASK		(0x03 << I40E_AQ_LINK_TX_SHIFT)
-#define I40E_AQ_LINK_TX_ACTIVE		0x00
-#define I40E_AQ_LINK_TX_DRAINED		0x01
-#define I40E_AQ_LINK_TX_FLUSHED		0x03
-#define I40E_AQ_LINK_FORCED_40G		0x10
-/* 25G Error Codes */
-#define I40E_AQ_25G_NO_ERR		0X00
-#define I40E_AQ_25G_NOT_PRESENT		0X01
-#define I40E_AQ_25G_NVM_CRC_ERR		0X02
-#define I40E_AQ_25G_SBUS_UCODE_ERR	0X03
-#define I40E_AQ_25G_SERDES_UCODE_ERR	0X04
-#define I40E_AQ_25G_NIMB_UCODE_ERR	0X05
 	u8	loopback; /* use defines from i40e_aqc_set_lb_mode */
+/* Since firmware API 1.7 loopback field keeps power class info as well */
+#define I40E_AQ_LOOPBACK_MASK		0x07
 	__le16	max_frame_size;
 	u8	config;
 #define I40E_AQ_CONFIG_FEC_KR_ENA	0x01
 #define I40E_AQ_CONFIG_FEC_RS_ENA	0x02
 #define I40E_AQ_CONFIG_CRC_ENA		0x04
 #define I40E_AQ_CONFIG_PACING_MASK	0x78
-	u8	power_desc;
-#define I40E_AQ_LINK_POWER_CLASS_1	0x00
-#define I40E_AQ_LINK_POWER_CLASS_2	0x01
-#define I40E_AQ_LINK_POWER_CLASS_3	0x02
-#define I40E_AQ_LINK_POWER_CLASS_4	0x03
-#define I40E_AQ_PWR_CLASS_MASK		0x03
-	u8	reserved[4];
+	union {
+		struct {
+			u8	power_desc;
+			u8	reserved[4];
+		};
+		struct {
+			u8	link_type[4];
+			u8	link_type_ext;
+		};
+	};
 };
 
 I40E_CHECK_CMD_LENGTH(i40e_aqc_get_link_status);
@@ -1957,13 +1627,7 @@ struct i40e_aqc_set_phy_int_mask {
 	__le16	event_mask;
 #define I40E_AQ_EVENT_LINK_UPDOWN	0x0002
 #define I40E_AQ_EVENT_MEDIA_NA		0x0004
-#define I40E_AQ_EVENT_LINK_FAULT	0x0008
-#define I40E_AQ_EVENT_PHY_TEMP_ALARM	0x0010
-#define I40E_AQ_EVENT_EXCESSIVE_ERRORS	0x0020
-#define I40E_AQ_EVENT_SIGNAL_DETECT	0x0040
-#define I40E_AQ_EVENT_AN_COMPLETED	0x0080
 #define I40E_AQ_EVENT_MODULE_QUAL_FAIL	0x0100
-#define I40E_AQ_EVENT_PORT_TX_SUSPENDED	0x0200
 	u8	reserved1[6];
 };
 
@@ -1984,9 +1648,11 @@ I40E_CHECK_CMD_LENGTH(i40e_aqc_an_advt_reg);
 /* Set Loopback mode (0x0618) */
 struct i40e_aqc_set_lb_mode {
 	__le16	lb_mode;
-#define I40E_AQ_LB_PHY_LOCAL	0x01
-#define I40E_AQ_LB_PHY_REMOTE	0x02
-#define I40E_AQ_LB_MAC_LOCAL	0x04
+#define I40E_LEGACY_LOOPBACK_NVM_VER	0x6000
+#define I40E_AQ_LB_MAC_LOCAL		0x01
+#define I40E_AQ_LB_PHY_LOCAL		0x05
+#define I40E_AQ_LB_PHY_REMOTE		0x06
+#define I40E_AQ_LB_MAC_LOCAL_LEGACY	0x04
 	u8	reserved[14];
 };
 
@@ -1995,13 +1661,6 @@ I40E_CHECK_CMD_LENGTH(i40e_aqc_set_lb_mode);
 /* Set PHY Debug command (0x0622) */
 struct i40e_aqc_set_phy_debug {
 	u8	command_flags;
-#define I40E_AQ_PHY_DEBUG_RESET_INTERNAL	0x02
-#define I40E_AQ_PHY_DEBUG_RESET_EXTERNAL_SHIFT	2
-#define I40E_AQ_PHY_DEBUG_RESET_EXTERNAL_MASK	(0x03 << \
-					I40E_AQ_PHY_DEBUG_RESET_EXTERNAL_SHIFT)
-#define I40E_AQ_PHY_DEBUG_RESET_EXTERNAL_NONE	0x00
-#define I40E_AQ_PHY_DEBUG_RESET_EXTERNAL_HARD	0x01
-#define I40E_AQ_PHY_DEBUG_RESET_EXTERNAL_SOFT	0x02
 /* Disable link manageability on a single port */
 #define I40E_AQ_PHY_DEBUG_DISABLE_LINK_FW	0x10
 /* Disable link manageability on all ports */
@@ -2029,14 +1688,39 @@ struct i40e_aqc_run_phy_activity {
 
 I40E_CHECK_CMD_LENGTH(i40e_aqc_run_phy_activity);
 
+/* Set PHY Register command (0x0628) */
+/* Get PHY Register command (0x0629) */
+struct i40e_aqc_phy_register_access {
+	u8	phy_interface;
+#define I40E_AQ_PHY_REG_ACCESS_EXTERNAL	1
+#define I40E_AQ_PHY_REG_ACCESS_EXTERNAL_MODULE	2
+	u8	dev_address;
+	u8	cmd_flags;
+#define I40E_AQ_PHY_REG_ACCESS_DONT_CHANGE_QSFP_PAGE	0x01
+#define I40E_AQ_PHY_REG_ACCESS_SET_MDIO_IF_NUMBER	0x02
+#define I40E_AQ_PHY_REG_ACCESS_MDIO_IF_NUMBER_SHIFT	2
+#define I40E_AQ_PHY_REG_ACCESS_MDIO_IF_NUMBER_MASK	(0x3 << \
+		I40E_AQ_PHY_REG_ACCESS_MDIO_IF_NUMBER_SHIFT)
+	u8	reserved1;
+	__le32	reg_address;
+	__le32	reg_value;
+	u8	reserved2[4];
+};
+
+I40E_CHECK_CMD_LENGTH(i40e_aqc_phy_register_access);
+
 /* NVM Read command (indirect 0x0701)
  * NVM Erase commands (direct 0x0702)
  * NVM Update commands (indirect 0x0703)
  */
 struct i40e_aqc_nvm_update {
 	u8	command_flags;
-#define I40E_AQ_NVM_LAST_CMD	0x01
-#define I40E_AQ_NVM_FLASH_ONLY	0x80
+#define I40E_AQ_NVM_LAST_CMD			0x01
+#define I40E_AQ_NVM_REARRANGE_TO_FLAT		0x20
+#define I40E_AQ_NVM_REARRANGE_TO_STRUCT		0x40
+#define I40E_AQ_NVM_PRESERVATION_FLAGS_SHIFT	1
+#define I40E_AQ_NVM_PRESERVATION_FLAGS_SELECTED	0x03
+#define I40E_AQ_NVM_PRESERVATION_FLAGS_ALL	0x01
 	u8	module_pointer;
 	__le16	length;
 	__le32	offset;
@@ -2049,9 +1733,6 @@ I40E_CHECK_CMD_LENGTH(i40e_aqc_nvm_update);
 /* NVM Config Read (indirect 0x0704) */
 struct i40e_aqc_nvm_config_read {
 	__le16	cmd_flags;
-#define I40E_AQ_ANVM_SINGLE_OR_MULTIPLE_FEATURES_MASK	1 
-#define I40E_AQ_ANVM_READ_SINGLE_FEATURE		0 
-#define I40E_AQ_ANVM_READ_MULTIPLE_FEATURES		1
 	__le16	element_count;
 	__le16	element_id;	/* Feature/field ID */
 	__le16	element_id_msw;	/* MSWord of field ID */
@@ -2073,16 +1754,8 @@ struct i40e_aqc_nvm_config_write {
 I40E_CHECK_CMD_LENGTH(i40e_aqc_nvm_config_write);
 
 /* Used for 0x0704 as well as for 0x0705 commands */
-#define I40E_AQ_ANVM_FEATURE_OR_IMMEDIATE_SHIFT		1
-#define I40E_AQ_ANVM_FEATURE_OR_IMMEDIATE_MASK \
-				BIT(I40E_AQ_ANVM_FEATURE_OR_IMMEDIATE_SHIFT)
-#define I40E_AQ_ANVM_FEATURE		0
-#define I40E_AQ_ANVM_IMMEDIATE_FIELD	BIT(FEATURE_OR_IMMEDIATE_SHIFT)
 struct i40e_aqc_nvm_config_data_feature {
 	__le16 feature_id;
-#define I40E_AQ_ANVM_FEATURE_OPTION_OEM_ONLY		0x01
-#define I40E_AQ_ANVM_FEATURE_OPTION_DWORD_MAP		0x08
-#define I40E_AQ_ANVM_FEATURE_OPTION_POR_CSR		0x10
 	__le16 feature_options;
 	__le16 feature_selection;
 };
@@ -2102,7 +1775,6 @@ I40E_CHECK_STRUCT_LEN(0xc, i40e_aqc_nvm_config_data_immediate_field);
  * no command data struct used
  */
 struct i40e_aqc_nvm_oem_post_update {
-#define I40E_AQ_NVM_OEM_POST_UPDATE_EXTERNAL_DATA	0x01
 	u8 sel_data;
 	u8 reserved[7];
 };
@@ -2124,9 +1796,6 @@ I40E_CHECK_STRUCT_LEN(0x28, i40e_aqc_nvm_oem_post_update_buffer);
  */
 struct i40e_aqc_thermal_sensor {
 	u8 sensor_action;
-#define I40E_AQ_THERMAL_SENSOR_READ_CONFIG	0
-#define I40E_AQ_THERMAL_SENSOR_SET_CONFIG	1
-#define I40E_AQ_THERMAL_SENSOR_READ_TEMP	2
 	u8 reserved[7];
 	__le32	addr_high;
 	__le32	addr_low;
@@ -2179,10 +1848,6 @@ I40E_CHECK_CMD_LENGTH(i40e_aqc_alternate_ind_write);
  */
 struct i40e_aqc_alternate_write_done {
 	__le16	cmd_flags;
-#define I40E_AQ_ALTERNATE_MODE_BIOS_MASK	1
-#define I40E_AQ_ALTERNATE_MODE_BIOS_LEGACY	0
-#define I40E_AQ_ALTERNATE_MODE_BIOS_UEFI	1
-#define I40E_AQ_ALTERNATE_RESET_NEEDED		2
 	u8	reserved[14];
 };
 
@@ -2191,8 +1856,6 @@ I40E_CHECK_CMD_LENGTH(i40e_aqc_alternate_write_done);
 /* Set OEM mode (direct 0x0905) */
 struct i40e_aqc_alternate_set_mode {
 	__le32	mode;
-#define I40E_AQ_ALTERNATE_MODE_NONE	0
-#define I40E_AQ_ALTERNATE_MODE_OEM	1
 	u8	reserved[12];
 };
 
@@ -2218,13 +1881,9 @@ struct i40e_aqc_lldp_get_mib {
 #define I40E_AQ_LLDP_MIB_TYPE_MASK		0x3
 #define I40E_AQ_LLDP_MIB_LOCAL			0x0
 #define I40E_AQ_LLDP_MIB_REMOTE			0x1
-#define I40E_AQ_LLDP_MIB_LOCAL_AND_REMOTE	0x2
 #define I40E_AQ_LLDP_BRIDGE_TYPE_MASK		0xC
 #define I40E_AQ_LLDP_BRIDGE_TYPE_SHIFT		0x2
 #define I40E_AQ_LLDP_BRIDGE_TYPE_NEAREST_BRIDGE	0x0
-#define I40E_AQ_LLDP_BRIDGE_TYPE_NON_TPMR	0x1
-#define I40E_AQ_LLDP_TX_SHIFT			0x4
-#define I40E_AQ_LLDP_TX_MASK			(0x03 << I40E_AQ_LLDP_TX_SHIFT)
 /* TX pause flags use I40E_AQ_LINK_TX_* above */
 	__le16	local_len;
 	__le16	remote_len;
@@ -2240,7 +1899,6 @@ I40E_CHECK_CMD_LENGTH(i40e_aqc_lldp_get_mib);
  */
 struct i40e_aqc_lldp_update_mib {
 	u8	command;
-#define I40E_AQ_LLDP_MIB_UPDATE_ENABLE	0x0
 #define I40E_AQ_LLDP_MIB_UPDATE_DISABLE	0x1
 	u8	reserved[7];
 	__le32	addr_high;
@@ -2279,23 +1937,34 @@ I40E_CHECK_CMD_LENGTH(i40e_aqc_lldp_update_tlv);
 /* Stop LLDP (direct 0x0A05) */
 struct i40e_aqc_lldp_stop {
 	u8	command;
-#define I40E_AQ_LLDP_AGENT_STOP		0x0
-#define I40E_AQ_LLDP_AGENT_SHUTDOWN	0x1
+#define I40E_AQ_LLDP_AGENT_SHUTDOWN		0x1
+#define I40E_AQ_LLDP_AGENT_STOP_PERSIST		0x2
 	u8	reserved[15];
 };
 
 I40E_CHECK_CMD_LENGTH(i40e_aqc_lldp_stop);
 
 /* Start LLDP (direct 0x0A06) */
-
 struct i40e_aqc_lldp_start {
 	u8	command;
-#define I40E_AQ_LLDP_AGENT_START	0x1
+#define I40E_AQ_LLDP_AGENT_START		0x1
+#define I40E_AQ_LLDP_AGENT_START_PERSIST	0x2
 	u8	reserved[15];
 };
 
 I40E_CHECK_CMD_LENGTH(i40e_aqc_lldp_start);
 
+/* Set DCB (direct 0x0303) */
+struct i40e_aqc_set_dcb_parameters {
+	u8 command;
+#define I40E_AQ_DCB_SET_AGENT	0x1
+#define I40E_DCB_VALID		0x1
+	u8 valid_flags;
+	u8 reserved[14];
+};
+
+I40E_CHECK_CMD_LENGTH(i40e_aqc_set_dcb_parameters);
+
 /* Get CEE DCBX Oper Config (0x0A07)
  * uses the generic descriptor struct
  * returns below as indirect response
@@ -2374,11 +2043,12 @@ I40E_CHECK_STRUCT_LEN(0x20, i40e_aqc_get_cee_dcb_cfg_resp);
  */
 struct i40e_aqc_lldp_set_local_mib {
 #define SET_LOCAL_MIB_AC_TYPE_DCBX_SHIFT	0
-#define SET_LOCAL_MIB_AC_TYPE_DCBX_MASK	BIT(SET_LOCAL_MIB_AC_TYPE_DCBX_SHIFT)
+#define SET_LOCAL_MIB_AC_TYPE_DCBX_MASK	(1 << \
+					SET_LOCAL_MIB_AC_TYPE_DCBX_SHIFT)
 #define SET_LOCAL_MIB_AC_TYPE_LOCAL_MIB	0x0
 #define SET_LOCAL_MIB_AC_TYPE_NON_WILLING_APPS_SHIFT	(1)
-#define SET_LOCAL_MIB_AC_TYPE_NON_WILLING_APPS_MASK \
-			BIT(SET_LOCAL_MIB_AC_TYPE_NON_WILLING_APPS_SHIFT)
+#define SET_LOCAL_MIB_AC_TYPE_NON_WILLING_APPS_MASK	(1 << \
+				SET_LOCAL_MIB_AC_TYPE_NON_WILLING_APPS_SHIFT)
 #define SET_LOCAL_MIB_AC_TYPE_NON_WILLING_APPS		0x1
 	u8	type;
 	u8	reserved0;
@@ -2394,15 +2064,21 @@ I40E_CHECK_CMD_LENGTH(i40e_aqc_lldp_set_local_mib);
  *	Used for stopping/starting specific LLDP agent. e.g. DCBx
  */
 struct i40e_aqc_lldp_stop_start_specific_agent {
-#define I40E_AQC_START_SPECIFIC_AGENT_SHIFT	0
-#define I40E_AQC_START_SPECIFIC_AGENT_MASK \
-				BIT(I40E_AQC_START_SPECIFIC_AGENT_SHIFT)
 	u8	command;
 	u8	reserved[15];
 };
 
 I40E_CHECK_CMD_LENGTH(i40e_aqc_lldp_stop_start_specific_agent);
 
+/* Restore LLDP Agent factory settings (direct 0x0A0A) */
+struct i40e_aqc_lldp_restore {
+	u8	command;
+#define I40E_AQ_LLDP_AGENT_RESTORE		0x1
+	u8	reserved[15];
+};
+
+I40E_CHECK_CMD_LENGTH(i40e_aqc_lldp_restore);
+
 /* Add Udp Tunnel command and completion (direct 0x0B00) */
 struct i40e_aqc_add_udp_tunnel {
 	__le16	udp_port;
@@ -2410,8 +2086,6 @@ struct i40e_aqc_add_udp_tunnel {
 	u8	protocol_type;
 #define I40E_AQC_TUNNEL_TYPE_VXLAN	0x00
 #define I40E_AQC_TUNNEL_TYPE_NGE	0x01
-#define I40E_AQC_TUNNEL_TYPE_TEREDO	0x10
-#define I40E_AQC_TUNNEL_TYPE_VXLAN_GPE	0x11
 	u8	reserved1[10];
 };
 
@@ -2421,8 +2095,6 @@ struct i40e_aqc_add_udp_tunnel_completion {
 	__le16	udp_port;
 	u8	filter_entry_index;
 	u8	multiple_pfs;
-#define I40E_AQC_SINGLE_PF		0x0
-#define I40E_AQC_MULTIPLE_PFS		0x1
 	u8	total_filters;
 	u8	reserved[11];
 };
@@ -2495,16 +2167,7 @@ struct i40e_aqc_tunnel_key_structure {
 	u8	key1_len;  /* 0 to 15 */
 	u8	key2_len;  /* 0 to 15 */
 	u8	flags;
-#define I40E_AQC_TUNNEL_KEY_STRUCT_OVERRIDE	0x01
-/* response flags */
-#define I40E_AQC_TUNNEL_KEY_STRUCT_SUCCESS	0x01
-#define I40E_AQC_TUNNEL_KEY_STRUCT_MODIFIED	0x02
-#define I40E_AQC_TUNNEL_KEY_STRUCT_OVERRIDDEN	0x03
 	u8	network_key_index;
-#define I40E_AQC_NETWORK_KEY_INDEX_VXLAN		0x0
-#define I40E_AQC_NETWORK_KEY_INDEX_NGE			0x1
-#define I40E_AQC_NETWORK_KEY_INDEX_FLEX_MAC_IN_UDP	0x2
-#define I40E_AQC_NETWORK_KEY_INDEX_GRE			0x3
 	u8	reserved[10];
 };
 
@@ -2513,9 +2176,6 @@ I40E_CHECK_CMD_LENGTH(i40e_aqc_tunnel_key_structure);
 /* OEM mode commands (direct 0xFE0x) */
 struct i40e_aqc_oem_param_change {
 	__le32	param_type;
-#define I40E_AQ_OEM_PARAM_TYPE_PF_CTL	0
-#define I40E_AQ_OEM_PARAM_TYPE_BW_CTL	1
-#define I40E_AQ_OEM_PARAM_MAC		2
 	__le32	param_value1;
 	__le16	param_value2;
 	u8	reserved[6];
@@ -2525,8 +2185,6 @@ I40E_CHECK_CMD_LENGTH(i40e_aqc_oem_param_change);
 
 struct i40e_aqc_oem_state_change {
 	__le32	state;
-#define I40E_AQ_OEM_STATE_LINK_DOWN	0x0
-#define I40E_AQ_OEM_STATE_LINK_UP	0x1
 	u8	reserved[12];
 };
 
@@ -2562,14 +2220,8 @@ I40E_CHECK_CMD_LENGTH(i40e_aqc_opc_oem_ocbb_initialize);
 
 struct i40e_acq_set_test_mode {
 	u8	mode;
-#define I40E_AQ_TEST_PARTIAL	0
-#define I40E_AQ_TEST_FULL	1
-#define I40E_AQ_TEST_NVM	2
 	u8	reserved[3];
 	u8	command;
-#define I40E_AQ_TEST_OPEN	0
-#define I40E_AQ_TEST_CLOSE	1
-#define I40E_AQ_TEST_INC	2
 	u8	reserved2[3];
 	__le32	address_high;
 	__le32	address_low;
@@ -2610,20 +2262,6 @@ struct i40e_aqc_debug_modify_reg {
 I40E_CHECK_CMD_LENGTH(i40e_aqc_debug_modify_reg);
 
 /* dump internal data (0xFF08, indirect) */
-
-#define I40E_AQ_CLUSTER_ID_AUX		0
-#define I40E_AQ_CLUSTER_ID_SWITCH_FLU	1
-#define I40E_AQ_CLUSTER_ID_TXSCHED	2
-#define I40E_AQ_CLUSTER_ID_HMC		3
-#define I40E_AQ_CLUSTER_ID_MAC0		4
-#define I40E_AQ_CLUSTER_ID_MAC1		5
-#define I40E_AQ_CLUSTER_ID_MAC2		6
-#define I40E_AQ_CLUSTER_ID_MAC3		7
-#define I40E_AQ_CLUSTER_ID_DCB		8
-#define I40E_AQ_CLUSTER_ID_EMP_MEM	9
-#define I40E_AQ_CLUSTER_ID_PKT_BUF	10
-#define I40E_AQ_CLUSTER_ID_ALTRAM	11
-
 struct i40e_aqc_debug_dump_internals {
 	u8	cluster_id;
 	u8	table_id;
diff --git a/drivers/net/ethernet/intel/i40e/i40e_alloc.h b/drivers/net/ethernet/intel/i40e/i40e_alloc.h
index 926811ad44ac..e0dde326255d 100644
--- a/drivers/net/ethernet/intel/i40e/i40e_alloc.h
+++ b/drivers/net/ethernet/intel/i40e/i40e_alloc.h
@@ -1,58 +1,35 @@
-/*******************************************************************************
- *
- * Intel Ethernet Controller XL710 Family Linux Driver
- * Copyright(c) 2013 - 2014 Intel Corporation.
- *
- * This program is free software; you can redistribute it and/or modify it
- * under the terms and conditions of the GNU General Public License,
- * version 2, as published by the Free Software Foundation.
- *
- * This program is distributed in the hope it will be useful, but WITHOUT
- * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
- * FITNESS FOR A PARTICULAR PURPOSE.  See the GNU General Public License for
- * more details.
- *
- * You should have received a copy of the GNU General Public License along
- * with this program.  If not, see <http://www.gnu.org/licenses/>.
- *
- * The full GNU General Public License is included in this distribution in
- * the file called "COPYING".
- *
- * Contact Information:
- * e1000-devel Mailing List <e1000-devel@lists.sourceforge.net>
- * Intel Corporation, 5200 N.E. Elam Young Parkway, Hillsboro, OR 97124-6497
- *
- ******************************************************************************/
+/* SPDX-License-Identifier: GPL-2.0 */
+/* Copyright(c) 2013 - 2018 Intel Corporation. */
 
 #ifndef _I40E_ALLOC_H_
 #define _I40E_ALLOC_H_
 
+#include <linux/types.h>
+
 struct i40e_hw;
 
-/* Memory allocation types */
-enum i40e_memory_type {
-	i40e_mem_arq_buf = 0,		/* ARQ indirect command buffer */
-	i40e_mem_asq_buf = 1,
-	i40e_mem_atq_buf = 2,		/* ATQ indirect command buffer */
-	i40e_mem_arq_ring = 3,		/* ARQ descriptor ring */
-	i40e_mem_atq_ring = 4,		/* ATQ descriptor ring */
-	i40e_mem_pd = 5,		/* Page Descriptor */
-	i40e_mem_bp = 6,		/* Backing Page - 4KB */
-	i40e_mem_bp_jumbo = 7,		/* Backing Page - > 4KB */
-	i40e_mem_reserved
+/* memory allocation tracking */
+struct i40e_dma_mem {
+	void *va;
+	dma_addr_t pa;
+	u32 size;
+};
+
+struct i40e_virt_mem {
+	void *va;
+	u32 size;
 };
 
 /* prototype for functions used for dynamic memory allocation */
-i40e_status i40e_allocate_dma_mem(struct i40e_hw *hw,
-					    struct i40e_dma_mem *mem,
-					    enum i40e_memory_type type,
-					    u64 size, u32 alignment);
-i40e_status i40e_free_dma_mem(struct i40e_hw *hw,
-					struct i40e_dma_mem *mem);
-i40e_status i40e_allocate_virt_mem(struct i40e_hw *hw,
-					     struct i40e_virt_mem *mem,
-					     u32 size);
-i40e_status i40e_free_virt_mem(struct i40e_hw *hw,
-					 struct i40e_virt_mem *mem);
+int i40e_allocate_dma_mem(struct i40e_hw *hw,
+			  struct i40e_dma_mem *mem,
+			  u64 size, u32 alignment);
+int i40e_free_dma_mem(struct i40e_hw *hw,
+		      struct i40e_dma_mem *mem);
+int i40e_allocate_virt_mem(struct i40e_hw *hw,
+			   struct i40e_virt_mem *mem,
+			   u32 size);
+int i40e_free_virt_mem(struct i40e_hw *hw,
+		       struct i40e_virt_mem *mem);
 
 #endif /* _I40E_ALLOC_H_ */
diff --git a/drivers/net/ethernet/intel/i40e/i40e_client.c b/drivers/net/ethernet/intel/i40e/i40e_client.c
index 088b4a43bd2a..518bc738ea3b 100644
--- a/drivers/net/ethernet/intel/i40e/i40e_client.c
+++ b/drivers/net/ethernet/intel/i40e/i40e_client.c
@@ -1,40 +1,15 @@
-/*******************************************************************************
- *
- * Intel Ethernet Controller XL710 Family Linux Driver
- * Copyright(c) 2013 - 2015 Intel Corporation.
- *
- * This program is free software; you can redistribute it and/or modify it
- * under the terms and conditions of the GNU General Public License,
- * version 2, as published by the Free Software Foundation.
- *
- * This program is distributed in the hope it will be useful, but WITHOUT
- * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
- * FITNESS FOR A PARTICULAR PURPOSE.  See the GNU General Public License for
- * more details.
- *
- * You should have received a copy of the GNU General Public License along
- * with this program.  If not, see <http://www.gnu.org/licenses/>.
- *
- * The full GNU General Public License is included in this distribution in
- * the file called "COPYING".
- *
- * Contact Information:
- * e1000-devel Mailing List <e1000-devel@lists.sourceforge.net>
- * Intel Corporation, 5200 N.E. Elam Young Parkway, Hillsboro, OR 97124-6497
- *
- ******************************************************************************/
+// SPDX-License-Identifier: GPL-2.0
+/* Copyright(c) 2013 - 2018 Intel Corporation. */
 
 #include <linux/list.h>
 #include <linux/errno.h>
+#include <linux/net/intel/i40e_client.h>
 
 #include "i40e.h"
-#include "i40e_prototype.h"
-#include "i40e_client.h"
 
-static const char i40e_client_interface_version_str[] = I40E_CLIENT_VERSION_STR;
-static struct i40e_client *registered_client;
 static LIST_HEAD(i40e_devices);
 static DEFINE_MUTEX(i40e_device_mutex);
+DEFINE_IDA(i40e_client_ida);
 
 static int i40e_client_virtchnl_send(struct i40e_info *ldev,
 				     struct i40e_client *client,
@@ -63,7 +38,7 @@ static struct i40e_ops i40e_lan_ops = {
 /**
  * i40e_client_get_params - Get the params that can change at runtime
  * @vsi: the VSI with the message
- * @param: clinet param struct
+ * @params: client param struct
  *
  **/
 static
@@ -126,25 +101,26 @@ i40e_notify_client_of_vf_msg(struct i40e_vsi *vsi, u32 vf_id, u8 *msg, u16 len)
 
 /**
  * i40e_notify_client_of_l2_param_changes - call the client notify callback
- * @vsi: the VSI with l2 param changes
+ * @pf: PF device pointer
  *
- * If there is a client to this VSI, call the client
+ * If there is a client, call its callback
  **/
-void i40e_notify_client_of_l2_param_changes(struct i40e_vsi *vsi)
+void i40e_notify_client_of_l2_param_changes(struct i40e_pf *pf)
 {
-	struct i40e_pf *pf = vsi->back;
+	struct i40e_vsi *vsi = i40e_pf_get_main_vsi(pf);
 	struct i40e_client_instance *cdev = pf->cinst;
 	struct i40e_params params;
 
 	if (!cdev || !cdev->client)
 		return;
 	if (!cdev->client->ops || !cdev->client->ops->l2_param_change) {
-		dev_dbg(&vsi->back->pdev->dev,
+		dev_dbg(&pf->pdev->dev,
 			"Cannot locate client instance l2_param_change routine\n");
 		return;
 	}
 	if (!test_bit(__I40E_CLIENT_INSTANCE_OPENED, &cdev->state)) {
-		dev_dbg(&vsi->back->pdev->dev, "Client is not open, abort l2 param change\n");
+		dev_dbg(&pf->pdev->dev,
+			"Client is not open, abort l2 param change\n");
 		return;
 	}
 	memset(&params, 0, sizeof(params));
@@ -173,8 +149,6 @@ static void i40e_client_release_qvlist(struct i40e_info *ldev)
 		u32 reg_idx;
 
 		qv_info = &qvlist_info->qv_info[i];
-		if (!qv_info)
-			continue;
 		reg_idx = I40E_PFINT_LNKLSTN(qv_info->v_idx - 1);
 		wr32(&pf->hw, reg_idx, I40E_PFINT_LNKLSTN_FIRSTQ_INDX_MASK);
 	}
@@ -184,23 +158,26 @@ static void i40e_client_release_qvlist(struct i40e_info *ldev)
 
 /**
  * i40e_notify_client_of_netdev_close - call the client close callback
- * @vsi: the VSI with netdev closed
+ * @pf: PF device pointer
  * @reset: true when close called due to a reset pending
  *
  * If there is a client to this netdev, call the client with close
  **/
-void i40e_notify_client_of_netdev_close(struct i40e_vsi *vsi, bool reset)
+void i40e_notify_client_of_netdev_close(struct i40e_pf *pf, bool reset)
 {
-	struct i40e_pf *pf = vsi->back;
 	struct i40e_client_instance *cdev = pf->cinst;
 
 	if (!cdev || !cdev->client)
 		return;
 	if (!cdev->client->ops || !cdev->client->ops->close) {
-		dev_dbg(&vsi->back->pdev->dev,
+		dev_dbg(&pf->pdev->dev,
 			"Cannot locate client instance close routine\n");
 		return;
 	}
+	if (!test_bit(__I40E_CLIENT_INSTANCE_OPENED, &cdev->state)) {
+		dev_dbg(&pf->pdev->dev, "Client is not open, abort close\n");
+		return;
+	}
 	cdev->client->ops->close(&cdev->lan_info, cdev->client, reset);
 	clear_bit(__I40E_CLIENT_INSTANCE_OPENED, &cdev->state);
 	i40e_client_release_qvlist(&cdev->lan_info);
@@ -273,8 +250,8 @@ int i40e_vf_client_capable(struct i40e_pf *pf, u32 vf_id)
 	if (!cdev || !cdev->client)
 		goto out;
 	if (!cdev->client->ops || !cdev->client->ops->vf_capable) {
-		dev_info(&pf->pdev->dev,
-			 "Cannot locate client instance VF capability routine\n");
+		dev_dbg(&pf->pdev->dev,
+			"Cannot locate client instance VF capability routine\n");
 		goto out;
 	}
 	if (!test_bit(__I40E_CLIENT_INSTANCE_OPENED, &cdev->state))
@@ -287,21 +264,78 @@ out:
 	return capable;
 }
 
+void i40e_client_update_msix_info(struct i40e_pf *pf)
+{
+	struct i40e_client_instance *cdev = pf->cinst;
+
+	if (!cdev || !cdev->client)
+		return;
+
+	cdev->lan_info.msix_count = pf->num_iwarp_msix;
+	cdev->lan_info.msix_entries = &pf->msix_entries[pf->iwarp_base_vector];
+}
+
+static void i40e_auxiliary_dev_release(struct device *dev)
+{
+	struct i40e_auxiliary_device *i40e_aux_dev =
+			container_of(dev, struct i40e_auxiliary_device, aux_dev.dev);
+
+	ida_free(&i40e_client_ida, i40e_aux_dev->aux_dev.id);
+	kfree(i40e_aux_dev);
+}
+
+static int i40e_register_auxiliary_dev(struct i40e_info *ldev, const char *name)
+{
+	struct i40e_auxiliary_device *i40e_aux_dev;
+	struct pci_dev *pdev = ldev->pcidev;
+	struct auxiliary_device *aux_dev;
+	int ret;
+
+	i40e_aux_dev = kzalloc(sizeof(*i40e_aux_dev), GFP_KERNEL);
+	if (!i40e_aux_dev)
+		return -ENOMEM;
+
+	i40e_aux_dev->ldev = ldev;
+
+	aux_dev = &i40e_aux_dev->aux_dev;
+	aux_dev->name = name;
+	aux_dev->dev.parent = &pdev->dev;
+	aux_dev->dev.release = i40e_auxiliary_dev_release;
+	ldev->aux_dev = aux_dev;
+
+	ret = ida_alloc(&i40e_client_ida, GFP_KERNEL);
+	if (ret < 0) {
+		kfree(i40e_aux_dev);
+		return ret;
+	}
+	aux_dev->id = ret;
+
+	ret = auxiliary_device_init(aux_dev);
+	if (ret < 0) {
+		ida_free(&i40e_client_ida, aux_dev->id);
+		kfree(i40e_aux_dev);
+		return ret;
+	}
+
+	ret = auxiliary_device_add(aux_dev);
+	if (ret) {
+		auxiliary_device_uninit(aux_dev);
+		return ret;
+	}
+
+	return ret;
+}
+
 /**
  * i40e_client_add_instance - add a client instance struct to the instance list
  * @pf: pointer to the board struct
- * @client: pointer to a client struct in the client list.
- * @existing: if there was already an existing instance
  *
  **/
 static void i40e_client_add_instance(struct i40e_pf *pf)
 {
+	struct i40e_vsi *vsi = i40e_pf_get_main_vsi(pf);
 	struct i40e_client_instance *cdev = NULL;
 	struct netdev_hw_addr *mac = NULL;
-	struct i40e_vsi *vsi = pf->vsi[pf->lan_vsi];
-
-	if (!registered_client || pf->cinst)
-		return;
 
 	cdev = kzalloc(sizeof(*cdev), GFP_KERNEL);
 	if (!cdev)
@@ -322,24 +356,29 @@ static void i40e_client_add_instance(struct i40e_pf *pf)
 	cdev->lan_info.fw_build = pf->hw.aq.fw_build;
 	set_bit(__I40E_CLIENT_INSTANCE_NONE, &cdev->state);
 
-	if (i40e_client_get_params(vsi, &cdev->lan_info.params)) {
-		kfree(cdev);
-		cdev = NULL;
-		return;
-	}
-
-	cdev->lan_info.msix_count = pf->num_iwarp_msix;
-	cdev->lan_info.msix_entries = &pf->msix_entries[pf->iwarp_base_vector];
+	if (i40e_client_get_params(vsi, &cdev->lan_info.params))
+		goto free_cdev;
 
-	mac = list_first_entry(&cdev->lan_info.netdev->dev_addrs.list,
-			       struct netdev_hw_addr, list);
+	mac = list_first_entry_or_null(&cdev->lan_info.netdev->dev_addrs.list,
+				       struct netdev_hw_addr, list);
 	if (mac)
 		ether_addr_copy(cdev->lan_info.lanmac, mac->addr);
 	else
 		dev_err(&pf->pdev->dev, "MAC address list is empty!\n");
 
-	cdev->client = registered_client;
 	pf->cinst = cdev;
+
+	cdev->lan_info.msix_count = pf->num_iwarp_msix;
+	cdev->lan_info.msix_entries = &pf->msix_entries[pf->iwarp_base_vector];
+
+	if (i40e_register_auxiliary_dev(&cdev->lan_info, "iwarp"))
+		goto free_cdev;
+
+	return;
+
+free_cdev:
+	kfree(cdev);
+	pf->cinst = NULL;
 }
 
 /**
@@ -360,14 +399,13 @@ void i40e_client_del_instance(struct i40e_pf *pf)
  **/
 void i40e_client_subtask(struct i40e_pf *pf)
 {
-	struct i40e_client *client = registered_client;
+	struct i40e_vsi *vsi = i40e_pf_get_main_vsi(pf);
 	struct i40e_client_instance *cdev;
-	struct i40e_vsi *vsi = pf->vsi[pf->lan_vsi];
+	struct i40e_client *client;
 	int ret = 0;
 
-	if (!(pf->flags & I40E_FLAG_SERVICE_CLIENT_REQUESTED))
+	if (!test_and_clear_bit(__I40E_CLIENT_SERVICE_REQUESTED, pf->state))
 		return;
-	pf->flags &= ~I40E_FLAG_SERVICE_CLIENT_REQUESTED;
 	cdev = pf->cinst;
 
 	/* If we're down or resetting, just bail */
@@ -375,14 +413,16 @@ void i40e_client_subtask(struct i40e_pf *pf)
 	    test_bit(__I40E_CONFIG_BUSY, pf->state))
 		return;
 
-	if (!client || !cdev)
+	if (!cdev || !cdev->client)
 		return;
 
-	/* Here we handle client opens. If the client is down, but
-	 * the netdev is up, then open the client.
+	client = cdev->client;
+
+	/* Here we handle client opens. If the client is down, and
+	 * the netdev is registered, then open the client.
 	 */
 	if (!test_bit(__I40E_CLIENT_INSTANCE_OPENED, &cdev->state)) {
-		if (!test_bit(__I40E_VSI_DOWN, vsi->state) &&
+		if (vsi->netdev_registered &&
 		    client->ops && client->ops->open) {
 			set_bit(__I40E_CLIENT_INSTANCE_OPENED, &cdev->state);
 			ret = client->ops->open(&cdev->lan_info, client);
@@ -390,20 +430,22 @@ void i40e_client_subtask(struct i40e_pf *pf)
 				/* Remove failed client instance */
 				clear_bit(__I40E_CLIENT_INSTANCE_OPENED,
 					  &cdev->state);
-				i40e_client_del_instance(pf);
+				return;
 			}
 		}
-	} else {
-	/* Likewise for client close. If the client is up, but the netdev
-	 * is down, then close the client.
-	 */
-		if (test_bit(__I40E_VSI_DOWN, vsi->state) &&
-		    client->ops && client->ops->close) {
-			clear_bit(__I40E_CLIENT_INSTANCE_OPENED, &cdev->state);
-			client->ops->close(&cdev->lan_info, client, false);
-			i40e_client_release_qvlist(&cdev->lan_info);
-		}
 	}
+
+	/* enable/disable PE TCP_ENA flag based on netdev down/up
+	 */
+	if (test_bit(__I40E_VSI_DOWN, vsi->state))
+		i40e_client_update_vsi_ctxt(&cdev->lan_info, client,
+					    0, 0, 0,
+					    I40E_CLIENT_VSI_FLAG_TCP_ENABLE);
+	else
+		i40e_client_update_vsi_ctxt(&cdev->lan_info, client,
+					    0, 0,
+					    I40E_CLIENT_VSI_FLAG_TCP_ENABLE,
+					    I40E_CLIENT_VSI_FLAG_TCP_ENABLE);
 }
 
 /**
@@ -436,17 +478,9 @@ int i40e_lan_add_device(struct i40e_pf *pf)
 		 pf->hw.pf_id, pf->hw.bus.bus_id,
 		 pf->hw.bus.device, pf->hw.bus.func);
 
-	/* If a client has already been registered, we need to add an instance
-	 * of it to our new LAN device.
-	 */
-	if (registered_client)
-		i40e_client_add_instance(pf);
+	i40e_client_add_instance(pf);
 
-	/* Since in some cases register may have happened before a device gets
-	 * added, we can schedule a subtask to go initiate the clients if
-	 * they can be launched at probe time.
-	 */
-	pf->flags |= I40E_FLAG_SERVICE_CLIENT_REQUESTED;
+	set_bit(__I40E_CLIENT_SERVICE_REQUESTED, pf->state);
 	i40e_service_event_schedule(pf);
 
 out:
@@ -462,9 +496,13 @@ out:
  **/
 int i40e_lan_del_device(struct i40e_pf *pf)
 {
+	struct auxiliary_device *aux_dev = pf->cinst->lan_info.aux_dev;
 	struct i40e_device *ldev, *tmp;
 	int ret = -ENODEV;
 
+	auxiliary_device_delete(aux_dev);
+	auxiliary_device_uninit(aux_dev);
+
 	/* First, remove any client instance. */
 	i40e_client_del_instance(pf);
 
@@ -485,69 +523,6 @@ int i40e_lan_del_device(struct i40e_pf *pf)
 }
 
 /**
- * i40e_client_release - release client specific resources
- * @client: pointer to the registered client
- *
- **/
-static void i40e_client_release(struct i40e_client *client)
-{
-	struct i40e_client_instance *cdev;
-	struct i40e_device *ldev;
-	struct i40e_pf *pf;
-
-	mutex_lock(&i40e_device_mutex);
-	list_for_each_entry(ldev, &i40e_devices, list) {
-		pf = ldev->pf;
-		cdev = pf->cinst;
-		if (!cdev)
-			continue;
-
-		while (test_and_set_bit(__I40E_SERVICE_SCHED,
-					pf->state))
-			usleep_range(500, 1000);
-
-		if (test_bit(__I40E_CLIENT_INSTANCE_OPENED, &cdev->state)) {
-			if (client->ops && client->ops->close)
-				client->ops->close(&cdev->lan_info, client,
-						   false);
-			i40e_client_release_qvlist(&cdev->lan_info);
-			clear_bit(__I40E_CLIENT_INSTANCE_OPENED, &cdev->state);
-
-			dev_warn(&pf->pdev->dev,
-				 "Client %s instance for PF id %d closed\n",
-				 client->name, pf->hw.pf_id);
-		}
-		/* delete the client instance */
-		i40e_client_del_instance(pf);
-		dev_info(&pf->pdev->dev, "Deleted client instance of Client %s\n",
-			 client->name);
-		clear_bit(__I40E_SERVICE_SCHED, pf->state);
-	}
-	mutex_unlock(&i40e_device_mutex);
-}
-
-/**
- * i40e_client_prepare - prepare client specific resources
- * @client: pointer to the registered client
- *
- **/
-static void i40e_client_prepare(struct i40e_client *client)
-{
-	struct i40e_device *ldev;
-	struct i40e_pf *pf;
-
-	mutex_lock(&i40e_device_mutex);
-	list_for_each_entry(ldev, &i40e_devices, list) {
-		pf = ldev->pf;
-		i40e_client_add_instance(pf);
-		/* Start the client subtask */
-		pf->flags |= I40E_FLAG_SERVICE_CLIENT_REQUESTED;
-		i40e_service_event_schedule(pf);
-	}
-	mutex_unlock(&i40e_device_mutex);
-}
-
-/**
  * i40e_client_virtchnl_send - TBD
  * @ldev: pointer to L2 context
  * @client: Client pointer
@@ -563,9 +538,9 @@ static int i40e_client_virtchnl_send(struct i40e_info *ldev,
 {
 	struct i40e_pf *pf = ldev->pf;
 	struct i40e_hw *hw = &pf->hw;
-	i40e_status err;
+	int err;
 
-	err = i40e_aq_send_msg_to_vf(hw, vf_id, I40E_VIRTCHNL_OP_IWARP,
+	err = i40e_aq_send_msg_to_vf(hw, vf_id, VIRTCHNL_OP_RDMA,
 				     0, msg, len, NULL);
 	if (err)
 		dev_err(&pf->pdev->dev, "Unable to send iWarp message to VF, error %d, aq status %d\n",
@@ -578,7 +553,7 @@ static int i40e_client_virtchnl_send(struct i40e_info *ldev,
  * i40e_client_setup_qvlist
  * @ldev: pointer to L2 context.
  * @client: Client pointer.
- * @qv_info: queue and vector list
+ * @qvlist_info: queue and vector list
  *
  * Return 0 on success or < 0 on error
  **/
@@ -590,19 +565,15 @@ static int i40e_client_setup_qvlist(struct i40e_info *ldev,
 	struct i40e_hw *hw = &pf->hw;
 	struct i40e_qv_info *qv_info;
 	u32 v_idx, i, reg_idx, reg;
-	u32 size;
 
-	size = sizeof(struct i40e_qvlist_info) +
-	       (sizeof(struct i40e_qv_info) * (qvlist_info->num_vectors - 1));
-	ldev->qvlist_info = kzalloc(size, GFP_KERNEL);
+	ldev->qvlist_info = kzalloc(struct_size(ldev->qvlist_info, qv_info,
+				    qvlist_info->num_vectors), GFP_KERNEL);
 	if (!ldev->qvlist_info)
 		return -ENOMEM;
 	ldev->qvlist_info->num_vectors = qvlist_info->num_vectors;
 
 	for (i = 0; i < qvlist_info->num_vectors; i++) {
 		qv_info = &qvlist_info->qv_info[i];
-		if (!qv_info)
-			continue;
 		v_idx = qv_info->v_idx;
 
 		/* Validate vector id belongs to this client */
@@ -653,7 +624,7 @@ err:
  * i40e_client_request_reset
  * @ldev: pointer to L2 context.
  * @client: Client pointer.
- * @level: reset level
+ * @reset_level: reset level
  **/
 static void i40e_client_request_reset(struct i40e_info *ldev,
 				      struct i40e_client *client,
@@ -694,11 +665,11 @@ static int i40e_client_update_vsi_ctxt(struct i40e_info *ldev,
 				       bool is_vf, u32 vf_id,
 				       u32 flag, u32 valid_flag)
 {
+	struct i40e_vsi *vsi = i40e_pf_get_main_vsi(ldev->pf);
 	struct i40e_pf *pf = ldev->pf;
-	struct i40e_vsi *vsi = pf->vsi[pf->lan_vsi];
 	struct i40e_vsi_context ctxt;
 	bool update = true;
-	i40e_status err;
+	int err;
 
 	/* TODO: for now do not allow setting VF's VSI setting */
 	if (is_vf)
@@ -710,20 +681,18 @@ static int i40e_client_update_vsi_ctxt(struct i40e_info *ldev,
 	ctxt.flags = I40E_AQ_VSI_TYPE_PF;
 	if (err) {
 		dev_info(&pf->pdev->dev,
-			 "couldn't get PF vsi config, err %s aq_err %s\n",
-			 i40e_stat_str(&pf->hw, err),
-			 i40e_aq_str(&pf->hw,
-				     pf->hw.aq.asq_last_status));
+			 "couldn't get PF vsi config, err %pe aq_err %s\n",
+			 ERR_PTR(err), libie_aq_str(pf->hw.aq.asq_last_status));
 		return -ENOENT;
 	}
 
-	if ((valid_flag & I40E_CLIENT_VSI_FLAG_TCP_PACKET_ENABLE) &&
-	    (flag & I40E_CLIENT_VSI_FLAG_TCP_PACKET_ENABLE)) {
+	if ((valid_flag & I40E_CLIENT_VSI_FLAG_TCP_ENABLE) &&
+	    (flag & I40E_CLIENT_VSI_FLAG_TCP_ENABLE)) {
 		ctxt.info.valid_sections =
 			cpu_to_le16(I40E_AQ_VSI_PROP_QUEUE_OPT_VALID);
 		ctxt.info.queueing_opt_flags |= I40E_AQ_VSI_QUE_OPT_TCP_ENA;
-	} else if ((valid_flag & I40E_CLIENT_VSI_FLAG_TCP_PACKET_ENABLE) &&
-		  !(flag & I40E_CLIENT_VSI_FLAG_TCP_PACKET_ENABLE)) {
+	} else if ((valid_flag & I40E_CLIENT_VSI_FLAG_TCP_ENABLE) &&
+		  !(flag & I40E_CLIENT_VSI_FLAG_TCP_ENABLE)) {
 		ctxt.info.valid_sections =
 			cpu_to_le16(I40E_AQ_VSI_PROP_QUEUE_OPT_VALID);
 		ctxt.info.queueing_opt_flags &= ~I40E_AQ_VSI_QUE_OPT_TCP_ENA;
@@ -738,90 +707,42 @@ static int i40e_client_update_vsi_ctxt(struct i40e_info *ldev,
 		err = i40e_aq_update_vsi_params(&vsi->back->hw, &ctxt, NULL);
 		if (err) {
 			dev_info(&pf->pdev->dev,
-				 "update VSI ctxt for PE failed, err %s aq_err %s\n",
-				 i40e_stat_str(&pf->hw, err),
-				 i40e_aq_str(&pf->hw,
-					     pf->hw.aq.asq_last_status));
+				 "update VSI ctxt for PE failed, err %pe aq_err %s\n",
+				 ERR_PTR(err),
+				 libie_aq_str(pf->hw.aq.asq_last_status));
 		}
 	}
 	return err;
 }
 
-/**
- * i40e_register_client - Register a i40e client driver with the L2 driver
- * @client: pointer to the i40e_client struct
- *
- * Returns 0 on success or non-0 on error
- **/
-int i40e_register_client(struct i40e_client *client)
+void i40e_client_device_register(struct i40e_info *ldev, struct i40e_client *client)
 {
-	int ret = 0;
-
-	if (!client) {
-		ret = -EIO;
-		goto out;
-	}
-
-	if (strlen(client->name) == 0) {
-		pr_info("i40e: Failed to register client with no name\n");
-		ret = -EIO;
-		goto out;
-	}
-
-	if (registered_client) {
-		pr_info("i40e: Client %s has already been registered!\n",
-			client->name);
-		ret = -EEXIST;
-		goto out;
-	}
-
-	if ((client->version.major != I40E_CLIENT_VERSION_MAJOR) ||
-	    (client->version.minor != I40E_CLIENT_VERSION_MINOR)) {
-		pr_info("i40e: Failed to register client %s due to mismatched client interface version\n",
-			client->name);
-		pr_info("Client is using version: %02d.%02d.%02d while LAN driver supports %s\n",
-			client->version.major, client->version.minor,
-			client->version.build,
-			i40e_client_interface_version_str);
-		ret = -EIO;
-		goto out;
-	}
-
-	registered_client = client;
-
-	i40e_client_prepare(client);
+	struct i40e_pf *pf = ldev->pf;
 
-	pr_info("i40e: Registered client %s\n", client->name);
-out:
-	return ret;
+	pf->cinst->client = client;
+	set_bit(__I40E_CLIENT_SERVICE_REQUESTED, pf->state);
+	i40e_service_event_schedule(pf);
 }
-EXPORT_SYMBOL(i40e_register_client);
+EXPORT_SYMBOL_GPL(i40e_client_device_register);
 
-/**
- * i40e_unregister_client - Unregister a i40e client driver with the L2 driver
- * @client: pointer to the i40e_client struct
- *
- * Returns 0 on success or non-0 on error
- **/
-int i40e_unregister_client(struct i40e_client *client)
+void i40e_client_device_unregister(struct i40e_info *ldev)
 {
-	int ret = 0;
+	struct i40e_pf *pf = ldev->pf;
+	struct i40e_client_instance *cdev = pf->cinst;
 
-	if (registered_client != client) {
-		pr_info("i40e: Client %s has not been registered\n",
-			client->name);
-		ret = -ENODEV;
-		goto out;
+	if (!cdev)
+		return;
+
+	while (test_and_set_bit(__I40E_SERVICE_SCHED, pf->state))
+		usleep_range(500, 1000);
+
+	if (test_bit(__I40E_CLIENT_INSTANCE_OPENED, &cdev->state)) {
+		cdev->client->ops->close(&cdev->lan_info, cdev->client, false);
+		clear_bit(__I40E_CLIENT_INSTANCE_OPENED, &cdev->state);
+		i40e_client_release_qvlist(&cdev->lan_info);
 	}
-	registered_client = NULL;
-	/* When a unregister request comes through we would have to send
-	 * a close for each of the client instances that were opened.
-	 * client_release function is called to handle this.
-	 */
-	i40e_client_release(client);
 
-	pr_info("i40e: Unregistered client %s\n", client->name);
-out:
-	return ret;
+	pf->cinst->client = NULL;
+	clear_bit(__I40E_SERVICE_SCHED, pf->state);
 }
-EXPORT_SYMBOL(i40e_unregister_client);
+EXPORT_SYMBOL_GPL(i40e_client_device_unregister);
diff --git a/drivers/net/ethernet/intel/i40e/i40e_client.h b/drivers/net/ethernet/intel/i40e/i40e_client.h
deleted file mode 100644
index 15b21a5315b5..000000000000
--- a/drivers/net/ethernet/intel/i40e/i40e_client.h
+++ /dev/null
@@ -1,226 +0,0 @@
-/*******************************************************************************
- *
- * Intel Ethernet Controller XL710 Family Linux Driver
- * Copyright(c) 2013 - 2015 Intel Corporation.
- *
- * This program is free software; you can redistribute it and/or modify it
- * under the terms and conditions of the GNU General Public License,
- * version 2, as published by the Free Software Foundation.
- *
- * This program is distributed in the hope it will be useful, but WITHOUT
- * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
- * FITNESS FOR A PARTICULAR PURPOSE.  See the GNU General Public License for
- * more details.
- *
- * You should have received a copy of the GNU General Public License along
- * with this program.  If not, see <http://www.gnu.org/licenses/>.
- *
- * The full GNU General Public License is included in this distribution in
- * the file called "COPYING".
- *
- * Contact Information:
- * e1000-devel Mailing List <e1000-devel@lists.sourceforge.net>
- * Intel Corporation, 5200 N.E. Elam Young Parkway, Hillsboro, OR 97124-6497
- *
- ******************************************************************************/
-
-#ifndef _I40E_CLIENT_H_
-#define _I40E_CLIENT_H_
-
-#define I40E_CLIENT_STR_LENGTH 10
-
-/* Client interface version should be updated anytime there is a change in the
- * existing APIs or data structures.
- */
-#define I40E_CLIENT_VERSION_MAJOR 0
-#define I40E_CLIENT_VERSION_MINOR 01
-#define I40E_CLIENT_VERSION_BUILD 00
-#define I40E_CLIENT_VERSION_STR     \
-	__stringify(I40E_CLIENT_VERSION_MAJOR) "." \
-	__stringify(I40E_CLIENT_VERSION_MINOR) "." \
-	__stringify(I40E_CLIENT_VERSION_BUILD)
-
-struct i40e_client_version {
-	u8 major;
-	u8 minor;
-	u8 build;
-	u8 rsvd;
-};
-
-enum i40e_client_state {
-	__I40E_CLIENT_NULL,
-	__I40E_CLIENT_REGISTERED
-};
-
-enum i40e_client_instance_state {
-	__I40E_CLIENT_INSTANCE_NONE,
-	__I40E_CLIENT_INSTANCE_OPENED,
-};
-
-struct i40e_ops;
-struct i40e_client;
-
-/* HW does not define a type value for AEQ; only for RX/TX and CEQ.
- * In order for us to keep the interface simple, SW will define a
- * unique type value for AEQ.
- */
-#define I40E_QUEUE_TYPE_PE_AEQ  0x80
-#define I40E_QUEUE_INVALID_IDX	0xFFFF
-
-struct i40e_qv_info {
-	u32 v_idx; /* msix_vector */
-	u16 ceq_idx;
-	u16 aeq_idx;
-	u8 itr_idx;
-};
-
-struct i40e_qvlist_info {
-	u32 num_vectors;
-	struct i40e_qv_info qv_info[1];
-};
-
-#define I40E_CLIENT_MSIX_ALL 0xFFFFFFFF
-
-/* set of LAN parameters useful for clients managed by LAN */
-
-/* Struct to hold per priority info */
-struct i40e_prio_qos_params {
-	u16 qs_handle; /* qs handle for prio */
-	u8 tc; /* TC mapped to prio */
-	u8 reserved;
-};
-
-#define I40E_CLIENT_MAX_USER_PRIORITY        8
-/* Struct to hold Client QoS */
-struct i40e_qos_params {
-	struct i40e_prio_qos_params prio_qos[I40E_CLIENT_MAX_USER_PRIORITY];
-};
-
-struct i40e_params {
-	struct i40e_qos_params qos;
-	u16 mtu;
-};
-
-/* Structure to hold Lan device info for a client device */
-struct i40e_info {
-	struct i40e_client_version version;
-	u8 lanmac[6];
-	struct net_device *netdev;
-	struct pci_dev *pcidev;
-	u8 __iomem *hw_addr;
-	u8 fid;	/* function id, PF id or VF id */
-#define I40E_CLIENT_FTYPE_PF 0
-#define I40E_CLIENT_FTYPE_VF 1
-	u8 ftype; /* function type, PF or VF */
-	void *pf;
-
-	/* All L2 params that could change during the life span of the PF
-	 * and needs to be communicated to the client when they change
-	 */
-	struct i40e_qvlist_info *qvlist_info;
-	struct i40e_params params;
-	struct i40e_ops *ops;
-
-	u16 msix_count;	 /* number of msix vectors*/
-	/* Array down below will be dynamically allocated based on msix_count */
-	struct msix_entry *msix_entries;
-	u16 itr_index; /* Which ITR index the PE driver is suppose to use */
-	u16 fw_maj_ver;                 /* firmware major version */
-	u16 fw_min_ver;                 /* firmware minor version */
-	u32 fw_build;                   /* firmware build number */
-};
-
-#define I40E_CLIENT_RESET_LEVEL_PF   1
-#define I40E_CLIENT_RESET_LEVEL_CORE 2
-#define I40E_CLIENT_VSI_FLAG_TCP_PACKET_ENABLE  BIT(1)
-
-struct i40e_ops {
-	/* setup_q_vector_list enables queues with a particular vector */
-	int (*setup_qvlist)(struct i40e_info *ldev, struct i40e_client *client,
-			    struct i40e_qvlist_info *qv_info);
-
-	int (*virtchnl_send)(struct i40e_info *ldev, struct i40e_client *client,
-			     u32 vf_id, u8 *msg, u16 len);
-
-	/* If the PE Engine is unresponsive, RDMA driver can request a reset.
-	 * The level helps determine the level of reset being requested.
-	 */
-	void (*request_reset)(struct i40e_info *ldev,
-			      struct i40e_client *client, u32 level);
-
-	/* API for the RDMA driver to set certain VSI flags that control
-	 * PE Engine.
-	 */
-	int (*update_vsi_ctxt)(struct i40e_info *ldev,
-			       struct i40e_client *client,
-			       bool is_vf, u32 vf_id,
-			       u32 flag, u32 valid_flag);
-};
-
-struct i40e_client_ops {
-	/* Should be called from register_client() or whenever PF is ready
-	 * to create a specific client instance.
-	 */
-	int (*open)(struct i40e_info *ldev, struct i40e_client *client);
-
-	/* Should be called when netdev is unavailable or when unregister
-	 * call comes in. If the close is happenening due to a reset being
-	 * triggered set the reset bit to true.
-	 */
-	void (*close)(struct i40e_info *ldev, struct i40e_client *client,
-		      bool reset);
-
-	/* called when some l2 managed parameters changes - mtu */
-	void (*l2_param_change)(struct i40e_info *ldev,
-				struct i40e_client *client,
-				struct i40e_params *params);
-
-	int (*virtchnl_receive)(struct i40e_info *ldev,
-				struct i40e_client *client, u32 vf_id,
-				u8 *msg, u16 len);
-
-	/* called when a VF is reset by the PF */
-	void (*vf_reset)(struct i40e_info *ldev,
-			 struct i40e_client *client, u32 vf_id);
-
-	/* called when the number of VFs changes */
-	void (*vf_enable)(struct i40e_info *ldev,
-			  struct i40e_client *client, u32 num_vfs);
-
-	/* returns true if VF is capable of specified offload */
-	int (*vf_capable)(struct i40e_info *ldev,
-			  struct i40e_client *client, u32 vf_id);
-};
-
-/* Client device */
-struct i40e_client_instance {
-	struct list_head list;
-	struct i40e_info lan_info;
-	struct i40e_client *client;
-	unsigned long  state;
-};
-
-struct i40e_client {
-	struct list_head list;		/* list of registered clients */
-	char name[I40E_CLIENT_STR_LENGTH];
-	struct i40e_client_version version;
-	unsigned long state;		/* client state */
-	atomic_t ref_cnt;  /* Count of all the client devices of this kind */
-	u32 flags;
-#define I40E_CLIENT_FLAGS_LAUNCH_ON_PROBE	BIT(0)
-#define I40E_TX_FLAGS_NOTIFY_OTHER_EVENTS	BIT(2)
-	u8 type;
-#define I40E_CLIENT_IWARP 0
-	const struct i40e_client_ops *ops; /* client ops provided by the client */
-};
-
-static inline bool i40e_client_is_registered(struct i40e_client *client)
-{
-	return test_bit(__I40E_CLIENT_REGISTERED, &client->state);
-}
-
-/* used by clients */
-int i40e_register_client(struct i40e_client *client);
-int i40e_unregister_client(struct i40e_client *client);
-
-#endif /* _I40E_CLIENT_H_ */
diff --git a/drivers/net/ethernet/intel/i40e/i40e_common.c b/drivers/net/ethernet/intel/i40e/i40e_common.c
index 24f020655291..59f5c1e810eb 100644
--- a/drivers/net/ethernet/intel/i40e/i40e_common.c
+++ b/drivers/net/ethernet/intel/i40e/i40e_common.c
@@ -1,33 +1,15 @@
-/*******************************************************************************
- *
- * Intel Ethernet Controller XL710 Family Linux Driver
- * Copyright(c) 2013 - 2016 Intel Corporation.
- *
- * This program is free software; you can redistribute it and/or modify it
- * under the terms and conditions of the GNU General Public License,
- * version 2, as published by the Free Software Foundation.
- *
- * This program is distributed in the hope it will be useful, but WITHOUT
- * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
- * FITNESS FOR A PARTICULAR PURPOSE.  See the GNU General Public License for
- * more details.
- *
- * You should have received a copy of the GNU General Public License along
- * with this program.  If not, see <http://www.gnu.org/licenses/>.
- *
- * The full GNU General Public License is included in this distribution in
- * the file called "COPYING".
- *
- * Contact Information:
- * e1000-devel Mailing List <e1000-devel@lists.sourceforge.net>
- * Intel Corporation, 5200 N.E. Elam Young Parkway, Hillsboro, OR 97124-6497
- *
- ******************************************************************************/
-
-#include "i40e_type.h"
-#include "i40e_adminq.h"
+// SPDX-License-Identifier: GPL-2.0
+/* Copyright(c) 2013 - 2021 Intel Corporation. */
+
+#include <linux/avf/virtchnl.h>
+#include <linux/bitfield.h>
+#include <linux/delay.h>
+#include <linux/etherdevice.h>
+#include <linux/pci.h>
+#include "i40e_adminq_cmd.h"
+#include "i40e_devids.h"
 #include "i40e_prototype.h"
-#include "i40e_virtchnl.h"
+#include "i40e_register.h"
 
 /**
  * i40e_set_mac_type - Sets MAC type
@@ -36,9 +18,9 @@
  * This function sets the mac type of the adapter based on the
  * vendor ID and device ID stored in the hw structure.
  **/
-static i40e_status i40e_set_mac_type(struct i40e_hw *hw)
+int i40e_set_mac_type(struct i40e_hw *hw)
 {
-	i40e_status status = 0;
+	int status = 0;
 
 	if (hw->vendor_id == PCI_VENDOR_ID_INTEL) {
 		switch (hw->device_id) {
@@ -49,12 +31,19 @@ static i40e_status i40e_set_mac_type(struct i40e_hw *hw)
 		case I40E_DEV_ID_QSFP_A:
 		case I40E_DEV_ID_QSFP_B:
 		case I40E_DEV_ID_QSFP_C:
+		case I40E_DEV_ID_1G_BASE_T_BC:
+		case I40E_DEV_ID_5G_BASE_T_BC:
 		case I40E_DEV_ID_10G_BASE_T:
 		case I40E_DEV_ID_10G_BASE_T4:
+		case I40E_DEV_ID_10G_BASE_T_BC:
+		case I40E_DEV_ID_10G_B:
+		case I40E_DEV_ID_10G_SFP:
 		case I40E_DEV_ID_20G_KR2:
 		case I40E_DEV_ID_20G_KR2_A:
 		case I40E_DEV_ID_25G_B:
 		case I40E_DEV_ID_25G_SFP28:
+		case I40E_DEV_ID_X710_N3000:
+		case I40E_DEV_ID_XXV710_N3000:
 			hw->mac.type = I40E_MAC_XL710;
 			break;
 		case I40E_DEV_ID_KX_X722:
@@ -63,6 +52,7 @@ static i40e_status i40e_set_mac_type(struct i40e_hw *hw)
 		case I40E_DEV_ID_1G_BASE_T_X722:
 		case I40E_DEV_ID_10G_BASE_T_X722:
 		case I40E_DEV_ID_SFP_I_X722:
+		case I40E_DEV_ID_SFP_X722_A:
 			hw->mac.type = I40E_MAC_X722;
 			break;
 		default:
@@ -70,7 +60,7 @@ static i40e_status i40e_set_mac_type(struct i40e_hw *hw)
 			break;
 		}
 	} else {
-		status = I40E_ERR_DEVICE_NOT_SUPPORTED;
+		status = -ENODEV;
 	}
 
 	hw_dbg(hw, "i40e_set_mac_type found mac: %d, returns: %d\n",
@@ -79,212 +69,6 @@ static i40e_status i40e_set_mac_type(struct i40e_hw *hw)
 }
 
 /**
- * i40e_aq_str - convert AQ err code to a string
- * @hw: pointer to the HW structure
- * @aq_err: the AQ error code to convert
- **/
-const char *i40e_aq_str(struct i40e_hw *hw, enum i40e_admin_queue_err aq_err)
-{
-	switch (aq_err) {
-	case I40E_AQ_RC_OK:
-		return "OK";
-	case I40E_AQ_RC_EPERM:
-		return "I40E_AQ_RC_EPERM";
-	case I40E_AQ_RC_ENOENT:
-		return "I40E_AQ_RC_ENOENT";
-	case I40E_AQ_RC_ESRCH:
-		return "I40E_AQ_RC_ESRCH";
-	case I40E_AQ_RC_EINTR:
-		return "I40E_AQ_RC_EINTR";
-	case I40E_AQ_RC_EIO:
-		return "I40E_AQ_RC_EIO";
-	case I40E_AQ_RC_ENXIO:
-		return "I40E_AQ_RC_ENXIO";
-	case I40E_AQ_RC_E2BIG:
-		return "I40E_AQ_RC_E2BIG";
-	case I40E_AQ_RC_EAGAIN:
-		return "I40E_AQ_RC_EAGAIN";
-	case I40E_AQ_RC_ENOMEM:
-		return "I40E_AQ_RC_ENOMEM";
-	case I40E_AQ_RC_EACCES:
-		return "I40E_AQ_RC_EACCES";
-	case I40E_AQ_RC_EFAULT:
-		return "I40E_AQ_RC_EFAULT";
-	case I40E_AQ_RC_EBUSY:
-		return "I40E_AQ_RC_EBUSY";
-	case I40E_AQ_RC_EEXIST:
-		return "I40E_AQ_RC_EEXIST";
-	case I40E_AQ_RC_EINVAL:
-		return "I40E_AQ_RC_EINVAL";
-	case I40E_AQ_RC_ENOTTY:
-		return "I40E_AQ_RC_ENOTTY";
-	case I40E_AQ_RC_ENOSPC:
-		return "I40E_AQ_RC_ENOSPC";
-	case I40E_AQ_RC_ENOSYS:
-		return "I40E_AQ_RC_ENOSYS";
-	case I40E_AQ_RC_ERANGE:
-		return "I40E_AQ_RC_ERANGE";
-	case I40E_AQ_RC_EFLUSHED:
-		return "I40E_AQ_RC_EFLUSHED";
-	case I40E_AQ_RC_BAD_ADDR:
-		return "I40E_AQ_RC_BAD_ADDR";
-	case I40E_AQ_RC_EMODE:
-		return "I40E_AQ_RC_EMODE";
-	case I40E_AQ_RC_EFBIG:
-		return "I40E_AQ_RC_EFBIG";
-	}
-
-	snprintf(hw->err_str, sizeof(hw->err_str), "%d", aq_err);
-	return hw->err_str;
-}
-
-/**
- * i40e_stat_str - convert status err code to a string
- * @hw: pointer to the HW structure
- * @stat_err: the status error code to convert
- **/
-const char *i40e_stat_str(struct i40e_hw *hw, i40e_status stat_err)
-{
-	switch (stat_err) {
-	case 0:
-		return "OK";
-	case I40E_ERR_NVM:
-		return "I40E_ERR_NVM";
-	case I40E_ERR_NVM_CHECKSUM:
-		return "I40E_ERR_NVM_CHECKSUM";
-	case I40E_ERR_PHY:
-		return "I40E_ERR_PHY";
-	case I40E_ERR_CONFIG:
-		return "I40E_ERR_CONFIG";
-	case I40E_ERR_PARAM:
-		return "I40E_ERR_PARAM";
-	case I40E_ERR_MAC_TYPE:
-		return "I40E_ERR_MAC_TYPE";
-	case I40E_ERR_UNKNOWN_PHY:
-		return "I40E_ERR_UNKNOWN_PHY";
-	case I40E_ERR_LINK_SETUP:
-		return "I40E_ERR_LINK_SETUP";
-	case I40E_ERR_ADAPTER_STOPPED:
-		return "I40E_ERR_ADAPTER_STOPPED";
-	case I40E_ERR_INVALID_MAC_ADDR:
-		return "I40E_ERR_INVALID_MAC_ADDR";
-	case I40E_ERR_DEVICE_NOT_SUPPORTED:
-		return "I40E_ERR_DEVICE_NOT_SUPPORTED";
-	case I40E_ERR_MASTER_REQUESTS_PENDING:
-		return "I40E_ERR_MASTER_REQUESTS_PENDING";
-	case I40E_ERR_INVALID_LINK_SETTINGS:
-		return "I40E_ERR_INVALID_LINK_SETTINGS";
-	case I40E_ERR_AUTONEG_NOT_COMPLETE:
-		return "I40E_ERR_AUTONEG_NOT_COMPLETE";
-	case I40E_ERR_RESET_FAILED:
-		return "I40E_ERR_RESET_FAILED";
-	case I40E_ERR_SWFW_SYNC:
-		return "I40E_ERR_SWFW_SYNC";
-	case I40E_ERR_NO_AVAILABLE_VSI:
-		return "I40E_ERR_NO_AVAILABLE_VSI";
-	case I40E_ERR_NO_MEMORY:
-		return "I40E_ERR_NO_MEMORY";
-	case I40E_ERR_BAD_PTR:
-		return "I40E_ERR_BAD_PTR";
-	case I40E_ERR_RING_FULL:
-		return "I40E_ERR_RING_FULL";
-	case I40E_ERR_INVALID_PD_ID:
-		return "I40E_ERR_INVALID_PD_ID";
-	case I40E_ERR_INVALID_QP_ID:
-		return "I40E_ERR_INVALID_QP_ID";
-	case I40E_ERR_INVALID_CQ_ID:
-		return "I40E_ERR_INVALID_CQ_ID";
-	case I40E_ERR_INVALID_CEQ_ID:
-		return "I40E_ERR_INVALID_CEQ_ID";
-	case I40E_ERR_INVALID_AEQ_ID:
-		return "I40E_ERR_INVALID_AEQ_ID";
-	case I40E_ERR_INVALID_SIZE:
-		return "I40E_ERR_INVALID_SIZE";
-	case I40E_ERR_INVALID_ARP_INDEX:
-		return "I40E_ERR_INVALID_ARP_INDEX";
-	case I40E_ERR_INVALID_FPM_FUNC_ID:
-		return "I40E_ERR_INVALID_FPM_FUNC_ID";
-	case I40E_ERR_QP_INVALID_MSG_SIZE:
-		return "I40E_ERR_QP_INVALID_MSG_SIZE";
-	case I40E_ERR_QP_TOOMANY_WRS_POSTED:
-		return "I40E_ERR_QP_TOOMANY_WRS_POSTED";
-	case I40E_ERR_INVALID_FRAG_COUNT:
-		return "I40E_ERR_INVALID_FRAG_COUNT";
-	case I40E_ERR_QUEUE_EMPTY:
-		return "I40E_ERR_QUEUE_EMPTY";
-	case I40E_ERR_INVALID_ALIGNMENT:
-		return "I40E_ERR_INVALID_ALIGNMENT";
-	case I40E_ERR_FLUSHED_QUEUE:
-		return "I40E_ERR_FLUSHED_QUEUE";
-	case I40E_ERR_INVALID_PUSH_PAGE_INDEX:
-		return "I40E_ERR_INVALID_PUSH_PAGE_INDEX";
-	case I40E_ERR_INVALID_IMM_DATA_SIZE:
-		return "I40E_ERR_INVALID_IMM_DATA_SIZE";
-	case I40E_ERR_TIMEOUT:
-		return "I40E_ERR_TIMEOUT";
-	case I40E_ERR_OPCODE_MISMATCH:
-		return "I40E_ERR_OPCODE_MISMATCH";
-	case I40E_ERR_CQP_COMPL_ERROR:
-		return "I40E_ERR_CQP_COMPL_ERROR";
-	case I40E_ERR_INVALID_VF_ID:
-		return "I40E_ERR_INVALID_VF_ID";
-	case I40E_ERR_INVALID_HMCFN_ID:
-		return "I40E_ERR_INVALID_HMCFN_ID";
-	case I40E_ERR_BACKING_PAGE_ERROR:
-		return "I40E_ERR_BACKING_PAGE_ERROR";
-	case I40E_ERR_NO_PBLCHUNKS_AVAILABLE:
-		return "I40E_ERR_NO_PBLCHUNKS_AVAILABLE";
-	case I40E_ERR_INVALID_PBLE_INDEX:
-		return "I40E_ERR_INVALID_PBLE_INDEX";
-	case I40E_ERR_INVALID_SD_INDEX:
-		return "I40E_ERR_INVALID_SD_INDEX";
-	case I40E_ERR_INVALID_PAGE_DESC_INDEX:
-		return "I40E_ERR_INVALID_PAGE_DESC_INDEX";
-	case I40E_ERR_INVALID_SD_TYPE:
-		return "I40E_ERR_INVALID_SD_TYPE";
-	case I40E_ERR_MEMCPY_FAILED:
-		return "I40E_ERR_MEMCPY_FAILED";
-	case I40E_ERR_INVALID_HMC_OBJ_INDEX:
-		return "I40E_ERR_INVALID_HMC_OBJ_INDEX";
-	case I40E_ERR_INVALID_HMC_OBJ_COUNT:
-		return "I40E_ERR_INVALID_HMC_OBJ_COUNT";
-	case I40E_ERR_INVALID_SRQ_ARM_LIMIT:
-		return "I40E_ERR_INVALID_SRQ_ARM_LIMIT";
-	case I40E_ERR_SRQ_ENABLED:
-		return "I40E_ERR_SRQ_ENABLED";
-	case I40E_ERR_ADMIN_QUEUE_ERROR:
-		return "I40E_ERR_ADMIN_QUEUE_ERROR";
-	case I40E_ERR_ADMIN_QUEUE_TIMEOUT:
-		return "I40E_ERR_ADMIN_QUEUE_TIMEOUT";
-	case I40E_ERR_BUF_TOO_SHORT:
-		return "I40E_ERR_BUF_TOO_SHORT";
-	case I40E_ERR_ADMIN_QUEUE_FULL:
-		return "I40E_ERR_ADMIN_QUEUE_FULL";
-	case I40E_ERR_ADMIN_QUEUE_NO_WORK:
-		return "I40E_ERR_ADMIN_QUEUE_NO_WORK";
-	case I40E_ERR_BAD_IWARP_CQE:
-		return "I40E_ERR_BAD_IWARP_CQE";
-	case I40E_ERR_NVM_BLANK_MODE:
-		return "I40E_ERR_NVM_BLANK_MODE";
-	case I40E_ERR_NOT_IMPLEMENTED:
-		return "I40E_ERR_NOT_IMPLEMENTED";
-	case I40E_ERR_PE_DOORBELL_NOT_ENABLED:
-		return "I40E_ERR_PE_DOORBELL_NOT_ENABLED";
-	case I40E_ERR_DIAG_TEST_FAILED:
-		return "I40E_ERR_DIAG_TEST_FAILED";
-	case I40E_ERR_NOT_READY:
-		return "I40E_ERR_NOT_READY";
-	case I40E_NOT_SUPPORTED:
-		return "I40E_NOT_SUPPORTED";
-	case I40E_ERR_FIRMWARE_API_VERSION:
-		return "I40E_ERR_FIRMWARE_API_VERSION";
-	}
-
-	snprintf(hw->err_str, sizeof(hw->err_str), "%d", stat_err);
-	return hw->err_str;
-}
-
-/**
  * i40e_debug_aq
  * @hw: debug mask related to admin queue
  * @mask: debug mask
@@ -297,48 +81,50 @@ const char *i40e_stat_str(struct i40e_hw *hw, i40e_status stat_err)
 void i40e_debug_aq(struct i40e_hw *hw, enum i40e_debug_mask mask, void *desc,
 		   void *buffer, u16 buf_len)
 {
-	struct i40e_aq_desc *aq_desc = (struct i40e_aq_desc *)desc;
+	struct libie_aq_desc *aq_desc = (struct libie_aq_desc *)desc;
+	u32 effective_mask = hw->debug_mask & mask;
+	char prefix[27];
 	u16 len;
 	u8 *buf = (u8 *)buffer;
 
-	if ((!(mask & hw->debug_mask)) || (desc == NULL))
+	if (!effective_mask || !desc)
 		return;
 
 	len = le16_to_cpu(aq_desc->datalen);
 
-	i40e_debug(hw, mask,
+	i40e_debug(hw, mask & I40E_DEBUG_AQ_DESCRIPTOR,
 		   "AQ CMD: opcode 0x%04X, flags 0x%04X, datalen 0x%04X, retval 0x%04X\n",
 		   le16_to_cpu(aq_desc->opcode),
 		   le16_to_cpu(aq_desc->flags),
 		   le16_to_cpu(aq_desc->datalen),
 		   le16_to_cpu(aq_desc->retval));
-	i40e_debug(hw, mask, "\tcookie (h,l) 0x%08X 0x%08X\n",
+	i40e_debug(hw, mask & I40E_DEBUG_AQ_DESCRIPTOR,
+		   "\tcookie (h,l) 0x%08X 0x%08X\n",
 		   le32_to_cpu(aq_desc->cookie_high),
 		   le32_to_cpu(aq_desc->cookie_low));
-	i40e_debug(hw, mask, "\tparam (0,1)  0x%08X 0x%08X\n",
-		   le32_to_cpu(aq_desc->params.internal.param0),
-		   le32_to_cpu(aq_desc->params.internal.param1));
-	i40e_debug(hw, mask, "\taddr (h,l)   0x%08X 0x%08X\n",
-		   le32_to_cpu(aq_desc->params.external.addr_high),
-		   le32_to_cpu(aq_desc->params.external.addr_low));
-
-	if ((buffer != NULL) && (aq_desc->datalen != 0)) {
+	i40e_debug(hw, mask & I40E_DEBUG_AQ_DESCRIPTOR,
+		   "\tparam (0,1)  0x%08X 0x%08X\n",
+		   le32_to_cpu(aq_desc->params.generic.param0),
+		   le32_to_cpu(aq_desc->params.generic.param1));
+	i40e_debug(hw, mask & I40E_DEBUG_AQ_DESCRIPTOR,
+		   "\taddr (h,l)   0x%08X 0x%08X\n",
+		   le32_to_cpu(aq_desc->params.generic.addr_high),
+		   le32_to_cpu(aq_desc->params.generic.addr_low));
+
+	if (buffer && buf_len != 0 && len != 0 &&
+	    (effective_mask & I40E_DEBUG_AQ_DESC_BUFFER)) {
 		i40e_debug(hw, mask, "AQ CMD Buffer:\n");
 		if (buf_len < len)
 			len = buf_len;
-		/* write the full 16-byte chunks */
-		if (hw->debug_mask & mask) {
-			char prefix[20];
-
-			snprintf(prefix, 20,
-				 "i40e %02x:%02x.%x: \t0x",
-				 hw->bus.bus_id,
-				 hw->bus.device,
-				 hw->bus.func);
-
-			print_hex_dump(KERN_INFO, prefix, DUMP_PREFIX_OFFSET,
-				       16, 1, buf, len, false);
-		}
+
+		snprintf(prefix, sizeof(prefix),
+			 "i40e %02x:%02x.%x: \t0x",
+			 hw->bus.bus_id,
+			 hw->bus.device,
+			 hw->bus.func);
+
+		print_hex_dump(KERN_INFO, prefix, DUMP_PREFIX_OFFSET,
+			       16, 1, buf, len, false);
 	}
 }
 
@@ -350,11 +136,11 @@ void i40e_debug_aq(struct i40e_hw *hw, enum i40e_debug_mask mask, void *desc,
  **/
 bool i40e_check_asq_alive(struct i40e_hw *hw)
 {
-	if (hw->aq.asq.len)
-		return !!(rd32(hw, hw->aq.asq.len) &
-			  I40E_PF_ATQLEN_ATQENABLE_MASK);
-	else
+	/* Check if the queue is initialized */
+	if (!hw->aq.asq.count)
 		return false;
+
+	return !!(rd32(hw, I40E_PF_ATQLEN) & I40E_PF_ATQLEN_ATQENABLE_MASK);
 }
 
 /**
@@ -365,17 +151,17 @@ bool i40e_check_asq_alive(struct i40e_hw *hw)
  * Tell the Firmware that we're shutting down the AdminQ and whether
  * or not the driver is unloading as well.
  **/
-i40e_status i40e_aq_queue_shutdown(struct i40e_hw *hw,
-					     bool unloading)
+int i40e_aq_queue_shutdown(struct i40e_hw *hw,
+			   bool unloading)
 {
-	struct i40e_aq_desc desc;
-	struct i40e_aqc_queue_shutdown *cmd =
-		(struct i40e_aqc_queue_shutdown *)&desc.params.raw;
-	i40e_status status;
+	struct i40e_aqc_queue_shutdown *cmd;
+	struct libie_aq_desc desc;
+	int status;
 
 	i40e_fill_default_direct_cmd_desc(&desc,
 					  i40e_aqc_opc_queue_shutdown);
 
+	cmd = libie_aq_raw(&desc);
 	if (unloading)
 		cmd->driver_unloading = cpu_to_le32(I40E_AQ_DRIVER_UNLOADING);
 	status = i40e_asq_send_command(hw, &desc, NULL, 0, NULL);
@@ -394,15 +180,15 @@ i40e_status i40e_aq_queue_shutdown(struct i40e_hw *hw,
  *
  * Internal function to get or set RSS look up table
  **/
-static i40e_status i40e_aq_get_set_rss_lut(struct i40e_hw *hw,
-					   u16 vsi_id, bool pf_lut,
-					   u8 *lut, u16 lut_size,
-					   bool set)
+static int i40e_aq_get_set_rss_lut(struct i40e_hw *hw,
+				   u16 vsi_id, bool pf_lut,
+				   u8 *lut, u16 lut_size,
+				   bool set)
 {
-	i40e_status status;
-	struct i40e_aq_desc desc;
-	struct i40e_aqc_get_set_rss_lut *cmd_resp =
-		   (struct i40e_aqc_get_set_rss_lut *)&desc.params.raw;
+	struct i40e_aqc_get_set_rss_lut *cmd_resp;
+	struct libie_aq_desc desc;
+	int status;
+	u16 flags;
 
 	if (set)
 		i40e_fill_default_direct_cmd_desc(&desc,
@@ -411,27 +197,23 @@ static i40e_status i40e_aq_get_set_rss_lut(struct i40e_hw *hw,
 		i40e_fill_default_direct_cmd_desc(&desc,
 						  i40e_aqc_opc_get_rss_lut);
 
+	cmd_resp = libie_aq_raw(&desc);
 	/* Indirect command */
-	desc.flags |= cpu_to_le16((u16)I40E_AQ_FLAG_BUF);
-	desc.flags |= cpu_to_le16((u16)I40E_AQ_FLAG_RD);
+	desc.flags |= cpu_to_le16((u16)LIBIE_AQ_FLAG_BUF);
+	desc.flags |= cpu_to_le16((u16)LIBIE_AQ_FLAG_RD);
 
-	cmd_resp->vsi_id =
-			cpu_to_le16((u16)((vsi_id <<
-					  I40E_AQC_SET_RSS_LUT_VSI_ID_SHIFT) &
-					  I40E_AQC_SET_RSS_LUT_VSI_ID_MASK));
-	cmd_resp->vsi_id |= cpu_to_le16((u16)I40E_AQC_SET_RSS_LUT_VSI_VALID);
+	vsi_id = FIELD_PREP(I40E_AQC_SET_RSS_LUT_VSI_ID_MASK, vsi_id) |
+		 FIELD_PREP(I40E_AQC_SET_RSS_LUT_VSI_VALID, 1);
+	cmd_resp->vsi_id = cpu_to_le16(vsi_id);
 
 	if (pf_lut)
-		cmd_resp->flags |= cpu_to_le16((u16)
-					((I40E_AQC_SET_RSS_LUT_TABLE_TYPE_PF <<
-					I40E_AQC_SET_RSS_LUT_TABLE_TYPE_SHIFT) &
-					I40E_AQC_SET_RSS_LUT_TABLE_TYPE_MASK));
+		flags = FIELD_PREP(I40E_AQC_SET_RSS_LUT_TABLE_TYPE_MASK,
+				   I40E_AQC_SET_RSS_LUT_TABLE_TYPE_PF);
 	else
-		cmd_resp->flags |= cpu_to_le16((u16)
-					((I40E_AQC_SET_RSS_LUT_TABLE_TYPE_VSI <<
-					I40E_AQC_SET_RSS_LUT_TABLE_TYPE_SHIFT) &
-					I40E_AQC_SET_RSS_LUT_TABLE_TYPE_MASK));
+		flags = FIELD_PREP(I40E_AQC_SET_RSS_LUT_TABLE_TYPE_MASK,
+				   I40E_AQC_SET_RSS_LUT_TABLE_TYPE_VSI);
 
+	cmd_resp->flags = cpu_to_le16(flags);
 	status = i40e_asq_send_command(hw, &desc, lut, lut_size, NULL);
 
 	return status;
@@ -447,8 +229,8 @@ static i40e_status i40e_aq_get_set_rss_lut(struct i40e_hw *hw,
  *
  * get the RSS lookup table, PF or VSI type
  **/
-i40e_status i40e_aq_get_rss_lut(struct i40e_hw *hw, u16 vsi_id,
-				bool pf_lut, u8 *lut, u16 lut_size)
+int i40e_aq_get_rss_lut(struct i40e_hw *hw, u16 vsi_id,
+			bool pf_lut, u8 *lut, u16 lut_size)
 {
 	return i40e_aq_get_set_rss_lut(hw, vsi_id, pf_lut, lut, lut_size,
 				       false);
@@ -464,8 +246,8 @@ i40e_status i40e_aq_get_rss_lut(struct i40e_hw *hw, u16 vsi_id,
  *
  * set the RSS lookup table, PF or VSI type
  **/
-i40e_status i40e_aq_set_rss_lut(struct i40e_hw *hw, u16 vsi_id,
-				bool pf_lut, u8 *lut, u16 lut_size)
+int i40e_aq_set_rss_lut(struct i40e_hw *hw, u16 vsi_id,
+			bool pf_lut, u8 *lut, u16 lut_size)
 {
 	return i40e_aq_get_set_rss_lut(hw, vsi_id, pf_lut, lut, lut_size, true);
 }
@@ -479,16 +261,15 @@ i40e_status i40e_aq_set_rss_lut(struct i40e_hw *hw, u16 vsi_id,
  *
  * get the RSS key per VSI
  **/
-static i40e_status i40e_aq_get_set_rss_key(struct i40e_hw *hw,
-				      u16 vsi_id,
-				      struct i40e_aqc_get_set_rss_key_data *key,
-				      bool set)
-{
-	i40e_status status;
-	struct i40e_aq_desc desc;
-	struct i40e_aqc_get_set_rss_key *cmd_resp =
-			(struct i40e_aqc_get_set_rss_key *)&desc.params.raw;
+static int i40e_aq_get_set_rss_key(struct i40e_hw *hw,
+				   u16 vsi_id,
+				   struct i40e_aqc_get_set_rss_key_data *key,
+				   bool set)
+{
 	u16 key_size = sizeof(struct i40e_aqc_get_set_rss_key_data);
+	struct i40e_aqc_get_set_rss_key *cmd_resp;
+	struct libie_aq_desc desc;
+	int status;
 
 	if (set)
 		i40e_fill_default_direct_cmd_desc(&desc,
@@ -497,15 +278,14 @@ static i40e_status i40e_aq_get_set_rss_key(struct i40e_hw *hw,
 		i40e_fill_default_direct_cmd_desc(&desc,
 						  i40e_aqc_opc_get_rss_key);
 
+	cmd_resp = libie_aq_raw(&desc);
 	/* Indirect command */
-	desc.flags |= cpu_to_le16((u16)I40E_AQ_FLAG_BUF);
-	desc.flags |= cpu_to_le16((u16)I40E_AQ_FLAG_RD);
+	desc.flags |= cpu_to_le16((u16)LIBIE_AQ_FLAG_BUF);
+	desc.flags |= cpu_to_le16((u16)LIBIE_AQ_FLAG_RD);
 
-	cmd_resp->vsi_id =
-			cpu_to_le16((u16)((vsi_id <<
-					  I40E_AQC_SET_RSS_KEY_VSI_ID_SHIFT) &
-					  I40E_AQC_SET_RSS_KEY_VSI_ID_MASK));
-	cmd_resp->vsi_id |= cpu_to_le16((u16)I40E_AQC_SET_RSS_KEY_VSI_VALID);
+	vsi_id = FIELD_PREP(I40E_AQC_SET_RSS_KEY_VSI_ID_MASK, vsi_id) |
+		 FIELD_PREP(I40E_AQC_SET_RSS_KEY_VSI_VALID, 1);
+	cmd_resp->vsi_id = cpu_to_le16(vsi_id);
 
 	status = i40e_asq_send_command(hw, &desc, key, key_size, NULL);
 
@@ -519,9 +299,9 @@ static i40e_status i40e_aq_get_set_rss_key(struct i40e_hw *hw,
  * @key: pointer to key info struct
  *
  **/
-i40e_status i40e_aq_get_rss_key(struct i40e_hw *hw,
-				u16 vsi_id,
-				struct i40e_aqc_get_set_rss_key_data *key)
+int i40e_aq_get_rss_key(struct i40e_hw *hw,
+			u16 vsi_id,
+			struct i40e_aqc_get_set_rss_key_data *key)
 {
 	return i40e_aq_get_set_rss_key(hw, vsi_id, key, false);
 }
@@ -534,378 +314,13 @@ i40e_status i40e_aq_get_rss_key(struct i40e_hw *hw,
  *
  * set the RSS key per VSI
  **/
-i40e_status i40e_aq_set_rss_key(struct i40e_hw *hw,
-				u16 vsi_id,
-				struct i40e_aqc_get_set_rss_key_data *key)
+int i40e_aq_set_rss_key(struct i40e_hw *hw,
+			u16 vsi_id,
+			struct i40e_aqc_get_set_rss_key_data *key)
 {
 	return i40e_aq_get_set_rss_key(hw, vsi_id, key, true);
 }
 
-/* The i40e_ptype_lookup table is used to convert from the 8-bit ptype in the
- * hardware to a bit-field that can be used by SW to more easily determine the
- * packet type.
- *
- * Macros are used to shorten the table lines and make this table human
- * readable.
- *
- * We store the PTYPE in the top byte of the bit field - this is just so that
- * we can check that the table doesn't have a row missing, as the index into
- * the table should be the PTYPE.
- *
- * Typical work flow:
- *
- * IF NOT i40e_ptype_lookup[ptype].known
- * THEN
- *      Packet is unknown
- * ELSE IF i40e_ptype_lookup[ptype].outer_ip == I40E_RX_PTYPE_OUTER_IP
- *      Use the rest of the fields to look at the tunnels, inner protocols, etc
- * ELSE
- *      Use the enum i40e_rx_l2_ptype to decode the packet type
- * ENDIF
- */
-
-/* macro to make the table lines short */
-#define I40E_PTT(PTYPE, OUTER_IP, OUTER_IP_VER, OUTER_FRAG, T, TE, TEF, I, PL)\
-	{	PTYPE, \
-		1, \
-		I40E_RX_PTYPE_OUTER_##OUTER_IP, \
-		I40E_RX_PTYPE_OUTER_##OUTER_IP_VER, \
-		I40E_RX_PTYPE_##OUTER_FRAG, \
-		I40E_RX_PTYPE_TUNNEL_##T, \
-		I40E_RX_PTYPE_TUNNEL_END_##TE, \
-		I40E_RX_PTYPE_##TEF, \
-		I40E_RX_PTYPE_INNER_PROT_##I, \
-		I40E_RX_PTYPE_PAYLOAD_LAYER_##PL }
-
-#define I40E_PTT_UNUSED_ENTRY(PTYPE) \
-		{ PTYPE, 0, 0, 0, 0, 0, 0, 0, 0, 0 }
-
-/* shorter macros makes the table fit but are terse */
-#define I40E_RX_PTYPE_NOF		I40E_RX_PTYPE_NOT_FRAG
-#define I40E_RX_PTYPE_FRG		I40E_RX_PTYPE_FRAG
-#define I40E_RX_PTYPE_INNER_PROT_TS	I40E_RX_PTYPE_INNER_PROT_TIMESYNC
-
-/* Lookup table mapping the HW PTYPE to the bit field for decoding */
-struct i40e_rx_ptype_decoded i40e_ptype_lookup[] = {
-	/* L2 Packet types */
-	I40E_PTT_UNUSED_ENTRY(0),
-	I40E_PTT(1,  L2, NONE, NOF, NONE, NONE, NOF, NONE, PAY2),
-	I40E_PTT(2,  L2, NONE, NOF, NONE, NONE, NOF, TS,   PAY2),
-	I40E_PTT(3,  L2, NONE, NOF, NONE, NONE, NOF, NONE, PAY2),
-	I40E_PTT_UNUSED_ENTRY(4),
-	I40E_PTT_UNUSED_ENTRY(5),
-	I40E_PTT(6,  L2, NONE, NOF, NONE, NONE, NOF, NONE, PAY2),
-	I40E_PTT(7,  L2, NONE, NOF, NONE, NONE, NOF, NONE, PAY2),
-	I40E_PTT_UNUSED_ENTRY(8),
-	I40E_PTT_UNUSED_ENTRY(9),
-	I40E_PTT(10, L2, NONE, NOF, NONE, NONE, NOF, NONE, PAY2),
-	I40E_PTT(11, L2, NONE, NOF, NONE, NONE, NOF, NONE, NONE),
-	I40E_PTT(12, L2, NONE, NOF, NONE, NONE, NOF, NONE, PAY3),
-	I40E_PTT(13, L2, NONE, NOF, NONE, NONE, NOF, NONE, PAY3),
-	I40E_PTT(14, L2, NONE, NOF, NONE, NONE, NOF, NONE, PAY3),
-	I40E_PTT(15, L2, NONE, NOF, NONE, NONE, NOF, NONE, PAY3),
-	I40E_PTT(16, L2, NONE, NOF, NONE, NONE, NOF, NONE, PAY3),
-	I40E_PTT(17, L2, NONE, NOF, NONE, NONE, NOF, NONE, PAY3),
-	I40E_PTT(18, L2, NONE, NOF, NONE, NONE, NOF, NONE, PAY3),
-	I40E_PTT(19, L2, NONE, NOF, NONE, NONE, NOF, NONE, PAY3),
-	I40E_PTT(20, L2, NONE, NOF, NONE, NONE, NOF, NONE, PAY3),
-	I40E_PTT(21, L2, NONE, NOF, NONE, NONE, NOF, NONE, PAY3),
-
-	/* Non Tunneled IPv4 */
-	I40E_PTT(22, IP, IPV4, FRG, NONE, NONE, NOF, NONE, PAY3),
-	I40E_PTT(23, IP, IPV4, NOF, NONE, NONE, NOF, NONE, PAY3),
-	I40E_PTT(24, IP, IPV4, NOF, NONE, NONE, NOF, UDP,  PAY4),
-	I40E_PTT_UNUSED_ENTRY(25),
-	I40E_PTT(26, IP, IPV4, NOF, NONE, NONE, NOF, TCP,  PAY4),
-	I40E_PTT(27, IP, IPV4, NOF, NONE, NONE, NOF, SCTP, PAY4),
-	I40E_PTT(28, IP, IPV4, NOF, NONE, NONE, NOF, ICMP, PAY4),
-
-	/* IPv4 --> IPv4 */
-	I40E_PTT(29, IP, IPV4, NOF, IP_IP, IPV4, FRG, NONE, PAY3),
-	I40E_PTT(30, IP, IPV4, NOF, IP_IP, IPV4, NOF, NONE, PAY3),
-	I40E_PTT(31, IP, IPV4, NOF, IP_IP, IPV4, NOF, UDP,  PAY4),
-	I40E_PTT_UNUSED_ENTRY(32),
-	I40E_PTT(33, IP, IPV4, NOF, IP_IP, IPV4, NOF, TCP,  PAY4),
-	I40E_PTT(34, IP, IPV4, NOF, IP_IP, IPV4, NOF, SCTP, PAY4),
-	I40E_PTT(35, IP, IPV4, NOF, IP_IP, IPV4, NOF, ICMP, PAY4),
-
-	/* IPv4 --> IPv6 */
-	I40E_PTT(36, IP, IPV4, NOF, IP_IP, IPV6, FRG, NONE, PAY3),
-	I40E_PTT(37, IP, IPV4, NOF, IP_IP, IPV6, NOF, NONE, PAY3),
-	I40E_PTT(38, IP, IPV4, NOF, IP_IP, IPV6, NOF, UDP,  PAY4),
-	I40E_PTT_UNUSED_ENTRY(39),
-	I40E_PTT(40, IP, IPV4, NOF, IP_IP, IPV6, NOF, TCP,  PAY4),
-	I40E_PTT(41, IP, IPV4, NOF, IP_IP, IPV6, NOF, SCTP, PAY4),
-	I40E_PTT(42, IP, IPV4, NOF, IP_IP, IPV6, NOF, ICMP, PAY4),
-
-	/* IPv4 --> GRE/NAT */
-	I40E_PTT(43, IP, IPV4, NOF, IP_GRENAT, NONE, NOF, NONE, PAY3),
-
-	/* IPv4 --> GRE/NAT --> IPv4 */
-	I40E_PTT(44, IP, IPV4, NOF, IP_GRENAT, IPV4, FRG, NONE, PAY3),
-	I40E_PTT(45, IP, IPV4, NOF, IP_GRENAT, IPV4, NOF, NONE, PAY3),
-	I40E_PTT(46, IP, IPV4, NOF, IP_GRENAT, IPV4, NOF, UDP,  PAY4),
-	I40E_PTT_UNUSED_ENTRY(47),
-	I40E_PTT(48, IP, IPV4, NOF, IP_GRENAT, IPV4, NOF, TCP,  PAY4),
-	I40E_PTT(49, IP, IPV4, NOF, IP_GRENAT, IPV4, NOF, SCTP, PAY4),
-	I40E_PTT(50, IP, IPV4, NOF, IP_GRENAT, IPV4, NOF, ICMP, PAY4),
-
-	/* IPv4 --> GRE/NAT --> IPv6 */
-	I40E_PTT(51, IP, IPV4, NOF, IP_GRENAT, IPV6, FRG, NONE, PAY3),
-	I40E_PTT(52, IP, IPV4, NOF, IP_GRENAT, IPV6, NOF, NONE, PAY3),
-	I40E_PTT(53, IP, IPV4, NOF, IP_GRENAT, IPV6, NOF, UDP,  PAY4),
-	I40E_PTT_UNUSED_ENTRY(54),
-	I40E_PTT(55, IP, IPV4, NOF, IP_GRENAT, IPV6, NOF, TCP,  PAY4),
-	I40E_PTT(56, IP, IPV4, NOF, IP_GRENAT, IPV6, NOF, SCTP, PAY4),
-	I40E_PTT(57, IP, IPV4, NOF, IP_GRENAT, IPV6, NOF, ICMP, PAY4),
-
-	/* IPv4 --> GRE/NAT --> MAC */
-	I40E_PTT(58, IP, IPV4, NOF, IP_GRENAT_MAC, NONE, NOF, NONE, PAY3),
-
-	/* IPv4 --> GRE/NAT --> MAC --> IPv4 */
-	I40E_PTT(59, IP, IPV4, NOF, IP_GRENAT_MAC, IPV4, FRG, NONE, PAY3),
-	I40E_PTT(60, IP, IPV4, NOF, IP_GRENAT_MAC, IPV4, NOF, NONE, PAY3),
-	I40E_PTT(61, IP, IPV4, NOF, IP_GRENAT_MAC, IPV4, NOF, UDP,  PAY4),
-	I40E_PTT_UNUSED_ENTRY(62),
-	I40E_PTT(63, IP, IPV4, NOF, IP_GRENAT_MAC, IPV4, NOF, TCP,  PAY4),
-	I40E_PTT(64, IP, IPV4, NOF, IP_GRENAT_MAC, IPV4, NOF, SCTP, PAY4),
-	I40E_PTT(65, IP, IPV4, NOF, IP_GRENAT_MAC, IPV4, NOF, ICMP, PAY4),
-
-	/* IPv4 --> GRE/NAT -> MAC --> IPv6 */
-	I40E_PTT(66, IP, IPV4, NOF, IP_GRENAT_MAC, IPV6, FRG, NONE, PAY3),
-	I40E_PTT(67, IP, IPV4, NOF, IP_GRENAT_MAC, IPV6, NOF, NONE, PAY3),
-	I40E_PTT(68, IP, IPV4, NOF, IP_GRENAT_MAC, IPV6, NOF, UDP,  PAY4),
-	I40E_PTT_UNUSED_ENTRY(69),
-	I40E_PTT(70, IP, IPV4, NOF, IP_GRENAT_MAC, IPV6, NOF, TCP,  PAY4),
-	I40E_PTT(71, IP, IPV4, NOF, IP_GRENAT_MAC, IPV6, NOF, SCTP, PAY4),
-	I40E_PTT(72, IP, IPV4, NOF, IP_GRENAT_MAC, IPV6, NOF, ICMP, PAY4),
-
-	/* IPv4 --> GRE/NAT --> MAC/VLAN */
-	I40E_PTT(73, IP, IPV4, NOF, IP_GRENAT_MAC_VLAN, NONE, NOF, NONE, PAY3),
-
-	/* IPv4 ---> GRE/NAT -> MAC/VLAN --> IPv4 */
-	I40E_PTT(74, IP, IPV4, NOF, IP_GRENAT_MAC_VLAN, IPV4, FRG, NONE, PAY3),
-	I40E_PTT(75, IP, IPV4, NOF, IP_GRENAT_MAC_VLAN, IPV4, NOF, NONE, PAY3),
-	I40E_PTT(76, IP, IPV4, NOF, IP_GRENAT_MAC_VLAN, IPV4, NOF, UDP,  PAY4),
-	I40E_PTT_UNUSED_ENTRY(77),
-	I40E_PTT(78, IP, IPV4, NOF, IP_GRENAT_MAC_VLAN, IPV4, NOF, TCP,  PAY4),
-	I40E_PTT(79, IP, IPV4, NOF, IP_GRENAT_MAC_VLAN, IPV4, NOF, SCTP, PAY4),
-	I40E_PTT(80, IP, IPV4, NOF, IP_GRENAT_MAC_VLAN, IPV4, NOF, ICMP, PAY4),
-
-	/* IPv4 -> GRE/NAT -> MAC/VLAN --> IPv6 */
-	I40E_PTT(81, IP, IPV4, NOF, IP_GRENAT_MAC_VLAN, IPV6, FRG, NONE, PAY3),
-	I40E_PTT(82, IP, IPV4, NOF, IP_GRENAT_MAC_VLAN, IPV6, NOF, NONE, PAY3),
-	I40E_PTT(83, IP, IPV4, NOF, IP_GRENAT_MAC_VLAN, IPV6, NOF, UDP,  PAY4),
-	I40E_PTT_UNUSED_ENTRY(84),
-	I40E_PTT(85, IP, IPV4, NOF, IP_GRENAT_MAC_VLAN, IPV6, NOF, TCP,  PAY4),
-	I40E_PTT(86, IP, IPV4, NOF, IP_GRENAT_MAC_VLAN, IPV6, NOF, SCTP, PAY4),
-	I40E_PTT(87, IP, IPV4, NOF, IP_GRENAT_MAC_VLAN, IPV6, NOF, ICMP, PAY4),
-
-	/* Non Tunneled IPv6 */
-	I40E_PTT(88, IP, IPV6, FRG, NONE, NONE, NOF, NONE, PAY3),
-	I40E_PTT(89, IP, IPV6, NOF, NONE, NONE, NOF, NONE, PAY3),
-	I40E_PTT(90, IP, IPV6, NOF, NONE, NONE, NOF, UDP,  PAY4),
-	I40E_PTT_UNUSED_ENTRY(91),
-	I40E_PTT(92, IP, IPV6, NOF, NONE, NONE, NOF, TCP,  PAY4),
-	I40E_PTT(93, IP, IPV6, NOF, NONE, NONE, NOF, SCTP, PAY4),
-	I40E_PTT(94, IP, IPV6, NOF, NONE, NONE, NOF, ICMP, PAY4),
-
-	/* IPv6 --> IPv4 */
-	I40E_PTT(95,  IP, IPV6, NOF, IP_IP, IPV4, FRG, NONE, PAY3),
-	I40E_PTT(96,  IP, IPV6, NOF, IP_IP, IPV4, NOF, NONE, PAY3),
-	I40E_PTT(97,  IP, IPV6, NOF, IP_IP, IPV4, NOF, UDP,  PAY4),
-	I40E_PTT_UNUSED_ENTRY(98),
-	I40E_PTT(99,  IP, IPV6, NOF, IP_IP, IPV4, NOF, TCP,  PAY4),
-	I40E_PTT(100, IP, IPV6, NOF, IP_IP, IPV4, NOF, SCTP, PAY4),
-	I40E_PTT(101, IP, IPV6, NOF, IP_IP, IPV4, NOF, ICMP, PAY4),
-
-	/* IPv6 --> IPv6 */
-	I40E_PTT(102, IP, IPV6, NOF, IP_IP, IPV6, FRG, NONE, PAY3),
-	I40E_PTT(103, IP, IPV6, NOF, IP_IP, IPV6, NOF, NONE, PAY3),
-	I40E_PTT(104, IP, IPV6, NOF, IP_IP, IPV6, NOF, UDP,  PAY4),
-	I40E_PTT_UNUSED_ENTRY(105),
-	I40E_PTT(106, IP, IPV6, NOF, IP_IP, IPV6, NOF, TCP,  PAY4),
-	I40E_PTT(107, IP, IPV6, NOF, IP_IP, IPV6, NOF, SCTP, PAY4),
-	I40E_PTT(108, IP, IPV6, NOF, IP_IP, IPV6, NOF, ICMP, PAY4),
-
-	/* IPv6 --> GRE/NAT */
-	I40E_PTT(109, IP, IPV6, NOF, IP_GRENAT, NONE, NOF, NONE, PAY3),
-
-	/* IPv6 --> GRE/NAT -> IPv4 */
-	I40E_PTT(110, IP, IPV6, NOF, IP_GRENAT, IPV4, FRG, NONE, PAY3),
-	I40E_PTT(111, IP, IPV6, NOF, IP_GRENAT, IPV4, NOF, NONE, PAY3),
-	I40E_PTT(112, IP, IPV6, NOF, IP_GRENAT, IPV4, NOF, UDP,  PAY4),
-	I40E_PTT_UNUSED_ENTRY(113),
-	I40E_PTT(114, IP, IPV6, NOF, IP_GRENAT, IPV4, NOF, TCP,  PAY4),
-	I40E_PTT(115, IP, IPV6, NOF, IP_GRENAT, IPV4, NOF, SCTP, PAY4),
-	I40E_PTT(116, IP, IPV6, NOF, IP_GRENAT, IPV4, NOF, ICMP, PAY4),
-
-	/* IPv6 --> GRE/NAT -> IPv6 */
-	I40E_PTT(117, IP, IPV6, NOF, IP_GRENAT, IPV6, FRG, NONE, PAY3),
-	I40E_PTT(118, IP, IPV6, NOF, IP_GRENAT, IPV6, NOF, NONE, PAY3),
-	I40E_PTT(119, IP, IPV6, NOF, IP_GRENAT, IPV6, NOF, UDP,  PAY4),
-	I40E_PTT_UNUSED_ENTRY(120),
-	I40E_PTT(121, IP, IPV6, NOF, IP_GRENAT, IPV6, NOF, TCP,  PAY4),
-	I40E_PTT(122, IP, IPV6, NOF, IP_GRENAT, IPV6, NOF, SCTP, PAY4),
-	I40E_PTT(123, IP, IPV6, NOF, IP_GRENAT, IPV6, NOF, ICMP, PAY4),
-
-	/* IPv6 --> GRE/NAT -> MAC */
-	I40E_PTT(124, IP, IPV6, NOF, IP_GRENAT_MAC, NONE, NOF, NONE, PAY3),
-
-	/* IPv6 --> GRE/NAT -> MAC -> IPv4 */
-	I40E_PTT(125, IP, IPV6, NOF, IP_GRENAT_MAC, IPV4, FRG, NONE, PAY3),
-	I40E_PTT(126, IP, IPV6, NOF, IP_GRENAT_MAC, IPV4, NOF, NONE, PAY3),
-	I40E_PTT(127, IP, IPV6, NOF, IP_GRENAT_MAC, IPV4, NOF, UDP,  PAY4),
-	I40E_PTT_UNUSED_ENTRY(128),
-	I40E_PTT(129, IP, IPV6, NOF, IP_GRENAT_MAC, IPV4, NOF, TCP,  PAY4),
-	I40E_PTT(130, IP, IPV6, NOF, IP_GRENAT_MAC, IPV4, NOF, SCTP, PAY4),
-	I40E_PTT(131, IP, IPV6, NOF, IP_GRENAT_MAC, IPV4, NOF, ICMP, PAY4),
-
-	/* IPv6 --> GRE/NAT -> MAC -> IPv6 */
-	I40E_PTT(132, IP, IPV6, NOF, IP_GRENAT_MAC, IPV6, FRG, NONE, PAY3),
-	I40E_PTT(133, IP, IPV6, NOF, IP_GRENAT_MAC, IPV6, NOF, NONE, PAY3),
-	I40E_PTT(134, IP, IPV6, NOF, IP_GRENAT_MAC, IPV6, NOF, UDP,  PAY4),
-	I40E_PTT_UNUSED_ENTRY(135),
-	I40E_PTT(136, IP, IPV6, NOF, IP_GRENAT_MAC, IPV6, NOF, TCP,  PAY4),
-	I40E_PTT(137, IP, IPV6, NOF, IP_GRENAT_MAC, IPV6, NOF, SCTP, PAY4),
-	I40E_PTT(138, IP, IPV6, NOF, IP_GRENAT_MAC, IPV6, NOF, ICMP, PAY4),
-
-	/* IPv6 --> GRE/NAT -> MAC/VLAN */
-	I40E_PTT(139, IP, IPV6, NOF, IP_GRENAT_MAC_VLAN, NONE, NOF, NONE, PAY3),
-
-	/* IPv6 --> GRE/NAT -> MAC/VLAN --> IPv4 */
-	I40E_PTT(140, IP, IPV6, NOF, IP_GRENAT_MAC_VLAN, IPV4, FRG, NONE, PAY3),
-	I40E_PTT(141, IP, IPV6, NOF, IP_GRENAT_MAC_VLAN, IPV4, NOF, NONE, PAY3),
-	I40E_PTT(142, IP, IPV6, NOF, IP_GRENAT_MAC_VLAN, IPV4, NOF, UDP,  PAY4),
-	I40E_PTT_UNUSED_ENTRY(143),
-	I40E_PTT(144, IP, IPV6, NOF, IP_GRENAT_MAC_VLAN, IPV4, NOF, TCP,  PAY4),
-	I40E_PTT(145, IP, IPV6, NOF, IP_GRENAT_MAC_VLAN, IPV4, NOF, SCTP, PAY4),
-	I40E_PTT(146, IP, IPV6, NOF, IP_GRENAT_MAC_VLAN, IPV4, NOF, ICMP, PAY4),
-
-	/* IPv6 --> GRE/NAT -> MAC/VLAN --> IPv6 */
-	I40E_PTT(147, IP, IPV6, NOF, IP_GRENAT_MAC_VLAN, IPV6, FRG, NONE, PAY3),
-	I40E_PTT(148, IP, IPV6, NOF, IP_GRENAT_MAC_VLAN, IPV6, NOF, NONE, PAY3),
-	I40E_PTT(149, IP, IPV6, NOF, IP_GRENAT_MAC_VLAN, IPV6, NOF, UDP,  PAY4),
-	I40E_PTT_UNUSED_ENTRY(150),
-	I40E_PTT(151, IP, IPV6, NOF, IP_GRENAT_MAC_VLAN, IPV6, NOF, TCP,  PAY4),
-	I40E_PTT(152, IP, IPV6, NOF, IP_GRENAT_MAC_VLAN, IPV6, NOF, SCTP, PAY4),
-	I40E_PTT(153, IP, IPV6, NOF, IP_GRENAT_MAC_VLAN, IPV6, NOF, ICMP, PAY4),
-
-	/* unused entries */
-	I40E_PTT_UNUSED_ENTRY(154),
-	I40E_PTT_UNUSED_ENTRY(155),
-	I40E_PTT_UNUSED_ENTRY(156),
-	I40E_PTT_UNUSED_ENTRY(157),
-	I40E_PTT_UNUSED_ENTRY(158),
-	I40E_PTT_UNUSED_ENTRY(159),
-
-	I40E_PTT_UNUSED_ENTRY(160),
-	I40E_PTT_UNUSED_ENTRY(161),
-	I40E_PTT_UNUSED_ENTRY(162),
-	I40E_PTT_UNUSED_ENTRY(163),
-	I40E_PTT_UNUSED_ENTRY(164),
-	I40E_PTT_UNUSED_ENTRY(165),
-	I40E_PTT_UNUSED_ENTRY(166),
-	I40E_PTT_UNUSED_ENTRY(167),
-	I40E_PTT_UNUSED_ENTRY(168),
-	I40E_PTT_UNUSED_ENTRY(169),
-
-	I40E_PTT_UNUSED_ENTRY(170),
-	I40E_PTT_UNUSED_ENTRY(171),
-	I40E_PTT_UNUSED_ENTRY(172),
-	I40E_PTT_UNUSED_ENTRY(173),
-	I40E_PTT_UNUSED_ENTRY(174),
-	I40E_PTT_UNUSED_ENTRY(175),
-	I40E_PTT_UNUSED_ENTRY(176),
-	I40E_PTT_UNUSED_ENTRY(177),
-	I40E_PTT_UNUSED_ENTRY(178),
-	I40E_PTT_UNUSED_ENTRY(179),
-
-	I40E_PTT_UNUSED_ENTRY(180),
-	I40E_PTT_UNUSED_ENTRY(181),
-	I40E_PTT_UNUSED_ENTRY(182),
-	I40E_PTT_UNUSED_ENTRY(183),
-	I40E_PTT_UNUSED_ENTRY(184),
-	I40E_PTT_UNUSED_ENTRY(185),
-	I40E_PTT_UNUSED_ENTRY(186),
-	I40E_PTT_UNUSED_ENTRY(187),
-	I40E_PTT_UNUSED_ENTRY(188),
-	I40E_PTT_UNUSED_ENTRY(189),
-
-	I40E_PTT_UNUSED_ENTRY(190),
-	I40E_PTT_UNUSED_ENTRY(191),
-	I40E_PTT_UNUSED_ENTRY(192),
-	I40E_PTT_UNUSED_ENTRY(193),
-	I40E_PTT_UNUSED_ENTRY(194),
-	I40E_PTT_UNUSED_ENTRY(195),
-	I40E_PTT_UNUSED_ENTRY(196),
-	I40E_PTT_UNUSED_ENTRY(197),
-	I40E_PTT_UNUSED_ENTRY(198),
-	I40E_PTT_UNUSED_ENTRY(199),
-
-	I40E_PTT_UNUSED_ENTRY(200),
-	I40E_PTT_UNUSED_ENTRY(201),
-	I40E_PTT_UNUSED_ENTRY(202),
-	I40E_PTT_UNUSED_ENTRY(203),
-	I40E_PTT_UNUSED_ENTRY(204),
-	I40E_PTT_UNUSED_ENTRY(205),
-	I40E_PTT_UNUSED_ENTRY(206),
-	I40E_PTT_UNUSED_ENTRY(207),
-	I40E_PTT_UNUSED_ENTRY(208),
-	I40E_PTT_UNUSED_ENTRY(209),
-
-	I40E_PTT_UNUSED_ENTRY(210),
-	I40E_PTT_UNUSED_ENTRY(211),
-	I40E_PTT_UNUSED_ENTRY(212),
-	I40E_PTT_UNUSED_ENTRY(213),
-	I40E_PTT_UNUSED_ENTRY(214),
-	I40E_PTT_UNUSED_ENTRY(215),
-	I40E_PTT_UNUSED_ENTRY(216),
-	I40E_PTT_UNUSED_ENTRY(217),
-	I40E_PTT_UNUSED_ENTRY(218),
-	I40E_PTT_UNUSED_ENTRY(219),
-
-	I40E_PTT_UNUSED_ENTRY(220),
-	I40E_PTT_UNUSED_ENTRY(221),
-	I40E_PTT_UNUSED_ENTRY(222),
-	I40E_PTT_UNUSED_ENTRY(223),
-	I40E_PTT_UNUSED_ENTRY(224),
-	I40E_PTT_UNUSED_ENTRY(225),
-	I40E_PTT_UNUSED_ENTRY(226),
-	I40E_PTT_UNUSED_ENTRY(227),
-	I40E_PTT_UNUSED_ENTRY(228),
-	I40E_PTT_UNUSED_ENTRY(229),
-
-	I40E_PTT_UNUSED_ENTRY(230),
-	I40E_PTT_UNUSED_ENTRY(231),
-	I40E_PTT_UNUSED_ENTRY(232),
-	I40E_PTT_UNUSED_ENTRY(233),
-	I40E_PTT_UNUSED_ENTRY(234),
-	I40E_PTT_UNUSED_ENTRY(235),
-	I40E_PTT_UNUSED_ENTRY(236),
-	I40E_PTT_UNUSED_ENTRY(237),
-	I40E_PTT_UNUSED_ENTRY(238),
-	I40E_PTT_UNUSED_ENTRY(239),
-
-	I40E_PTT_UNUSED_ENTRY(240),
-	I40E_PTT_UNUSED_ENTRY(241),
-	I40E_PTT_UNUSED_ENTRY(242),
-	I40E_PTT_UNUSED_ENTRY(243),
-	I40E_PTT_UNUSED_ENTRY(244),
-	I40E_PTT_UNUSED_ENTRY(245),
-	I40E_PTT_UNUSED_ENTRY(246),
-	I40E_PTT_UNUSED_ENTRY(247),
-	I40E_PTT_UNUSED_ENTRY(248),
-	I40E_PTT_UNUSED_ENTRY(249),
-
-	I40E_PTT_UNUSED_ENTRY(250),
-	I40E_PTT_UNUSED_ENTRY(251),
-	I40E_PTT_UNUSED_ENTRY(252),
-	I40E_PTT_UNUSED_ENTRY(253),
-	I40E_PTT_UNUSED_ENTRY(254),
-	I40E_PTT_UNUSED_ENTRY(255)
-};
-
 /**
  * i40e_init_shared_code - Initialize the shared code
  * @hw: pointer to hardware structure
@@ -918,10 +333,10 @@ struct i40e_rx_ptype_decoded i40e_ptype_lookup[] = {
  * hw_addr, back, device_id, vendor_id, subsystem_device_id,
  * subsystem_vendor_id, and revision_id
  **/
-i40e_status i40e_init_shared_code(struct i40e_hw *hw)
+int i40e_init_shared_code(struct i40e_hw *hw)
 {
-	i40e_status status = 0;
 	u32 port, ari, func_rid;
+	int status = 0;
 
 	i40e_set_mac_type(hw);
 
@@ -930,26 +345,23 @@ i40e_status i40e_init_shared_code(struct i40e_hw *hw)
 	case I40E_MAC_X722:
 		break;
 	default:
-		return I40E_ERR_DEVICE_NOT_SUPPORTED;
+		return -ENODEV;
 	}
 
 	hw->phy.get_link_info = true;
 
 	/* Determine port number and PF number*/
-	port = (rd32(hw, I40E_PFGEN_PORTNUM) & I40E_PFGEN_PORTNUM_PORT_NUM_MASK)
-					   >> I40E_PFGEN_PORTNUM_PORT_NUM_SHIFT;
+	port = FIELD_GET(I40E_PFGEN_PORTNUM_PORT_NUM_MASK,
+			 rd32(hw, I40E_PFGEN_PORTNUM));
 	hw->port = (u8)port;
-	ari = (rd32(hw, I40E_GLPCI_CAPSUP) & I40E_GLPCI_CAPSUP_ARI_EN_MASK) >>
-						 I40E_GLPCI_CAPSUP_ARI_EN_SHIFT;
+	ari = FIELD_GET(I40E_GLPCI_CAPSUP_ARI_EN_MASK,
+			rd32(hw, I40E_GLPCI_CAPSUP));
 	func_rid = rd32(hw, I40E_PF_FUNC_RID);
 	if (ari)
 		hw->pf_id = (u8)(func_rid & 0xff);
 	else
 		hw->pf_id = (u8)(func_rid & 0x7);
 
-	if (hw->mac.type == I40E_MAC_X722)
-		hw->flags |= I40E_HW_FLAG_AQ_SRCTL_ACCESS_ENABLE;
-
 	status = i40e_init_nvm(hw);
 	return status;
 }
@@ -961,18 +373,19 @@ i40e_status i40e_init_shared_code(struct i40e_hw *hw)
  * @addrs: the requestor's mac addr store
  * @cmd_details: pointer to command details structure or NULL
  **/
-static i40e_status i40e_aq_mac_address_read(struct i40e_hw *hw,
-				   u16 *flags,
-				   struct i40e_aqc_mac_address_read_data *addrs,
-				   struct i40e_asq_cmd_details *cmd_details)
+static int
+i40e_aq_mac_address_read(struct i40e_hw *hw,
+			 u16 *flags,
+			 struct i40e_aqc_mac_address_read_data *addrs,
+			 struct i40e_asq_cmd_details *cmd_details)
 {
-	struct i40e_aq_desc desc;
-	struct i40e_aqc_mac_address_read *cmd_data =
-		(struct i40e_aqc_mac_address_read *)&desc.params.raw;
-	i40e_status status;
+	struct i40e_aqc_mac_address_read *cmd_data;
+	struct libie_aq_desc desc;
+	int status;
 
 	i40e_fill_default_direct_cmd_desc(&desc, i40e_aqc_opc_mac_address_read);
-	desc.flags |= cpu_to_le16(I40E_AQ_FLAG_BUF);
+	cmd_data = libie_aq_raw(&desc);
+	desc.flags |= cpu_to_le16(LIBIE_AQ_FLAG_BUF);
 
 	status = i40e_asq_send_command(hw, &desc, addrs,
 				       sizeof(*addrs), cmd_details);
@@ -988,17 +401,17 @@ static i40e_status i40e_aq_mac_address_read(struct i40e_hw *hw,
  * @mac_addr: address to write
  * @cmd_details: pointer to command details structure or NULL
  **/
-i40e_status i40e_aq_mac_address_write(struct i40e_hw *hw,
-				    u16 flags, u8 *mac_addr,
-				    struct i40e_asq_cmd_details *cmd_details)
+int i40e_aq_mac_address_write(struct i40e_hw *hw,
+			      u16 flags, u8 *mac_addr,
+			      struct i40e_asq_cmd_details *cmd_details)
 {
-	struct i40e_aq_desc desc;
-	struct i40e_aqc_mac_address_write *cmd_data =
-		(struct i40e_aqc_mac_address_write *)&desc.params.raw;
-	i40e_status status;
+	struct i40e_aqc_mac_address_write *cmd_data;
+	struct libie_aq_desc desc;
+	int status;
 
 	i40e_fill_default_direct_cmd_desc(&desc,
 					  i40e_aqc_opc_mac_address_write);
+	cmd_data = libie_aq_raw(&desc);
 	cmd_data->command_flags = cpu_to_le16(flags);
 	cmd_data->mac_sah = cpu_to_le16((u16)mac_addr[0] << 8 | mac_addr[1]);
 	cmd_data->mac_sal = cpu_to_le32(((u32)mac_addr[2] << 24) |
@@ -1018,11 +431,11 @@ i40e_status i40e_aq_mac_address_write(struct i40e_hw *hw,
  *
  * Reads the adapter's MAC address from register
  **/
-i40e_status i40e_get_mac_addr(struct i40e_hw *hw, u8 *mac_addr)
+int i40e_get_mac_addr(struct i40e_hw *hw, u8 *mac_addr)
 {
 	struct i40e_aqc_mac_address_read_data addrs;
-	i40e_status status;
 	u16 flags = 0;
+	int status;
 
 	status = i40e_aq_mac_address_read(hw, &flags, &addrs, NULL);
 
@@ -1039,11 +452,11 @@ i40e_status i40e_get_mac_addr(struct i40e_hw *hw, u8 *mac_addr)
  *
  * Reads the adapter's Port MAC address
  **/
-i40e_status i40e_get_port_mac_addr(struct i40e_hw *hw, u8 *mac_addr)
+int i40e_get_port_mac_addr(struct i40e_hw *hw, u8 *mac_addr)
 {
 	struct i40e_aqc_mac_address_read_data addrs;
-	i40e_status status;
 	u16 flags = 0;
+	int status;
 
 	status = i40e_aq_mac_address_read(hw, &flags, &addrs, NULL);
 	if (status)
@@ -1052,7 +465,7 @@ i40e_status i40e_get_port_mac_addr(struct i40e_hw *hw, u8 *mac_addr)
 	if (flags & I40E_AQC_PORT_ADDR_VALID)
 		ether_addr_copy(mac_addr, addrs.port_mac);
 	else
-		status = I40E_ERR_INVALID_MAC_ADDR;
+		status = -EINVAL;
 
 	return status;
 }
@@ -1090,62 +503,72 @@ void i40e_pre_tx_queue_cfg(struct i40e_hw *hw, u32 queue, bool enable)
 }
 
 /**
- *  i40e_read_pba_string - Reads part number string from EEPROM
+ *  i40e_get_pba_string - Reads part number string from EEPROM
  *  @hw: pointer to hardware structure
- *  @pba_num: stores the part number string from the EEPROM
- *  @pba_num_size: part number string buffer length
  *
- *  Reads the part number string from the EEPROM.
+ *  Reads the part number string from the EEPROM and stores it
+ *  into newly allocated buffer and saves resulting pointer
+ *  to i40e_hw->pba_id field.
  **/
-i40e_status i40e_read_pba_string(struct i40e_hw *hw, u8 *pba_num,
-				 u32 pba_num_size)
+void i40e_get_pba_string(struct i40e_hw *hw)
 {
-	i40e_status status = 0;
+#define I40E_NVM_PBA_FLAGS_BLK_PRESENT	0xFAFA
 	u16 pba_word = 0;
 	u16 pba_size = 0;
 	u16 pba_ptr = 0;
-	u16 i = 0;
+	int status;
+	char *ptr;
+	u16 i;
 
 	status = i40e_read_nvm_word(hw, I40E_SR_PBA_FLAGS, &pba_word);
-	if (status || (pba_word != 0xFAFA)) {
-		hw_dbg(hw, "Failed to read PBA flags or flag is invalid.\n");
-		return status;
+	if (status) {
+		hw_dbg(hw, "Failed to read PBA flags.\n");
+		return;
+	}
+	if (pba_word != I40E_NVM_PBA_FLAGS_BLK_PRESENT) {
+		hw_dbg(hw, "PBA block is not present.\n");
+		return;
 	}
 
 	status = i40e_read_nvm_word(hw, I40E_SR_PBA_BLOCK_PTR, &pba_ptr);
 	if (status) {
 		hw_dbg(hw, "Failed to read PBA Block pointer.\n");
-		return status;
+		return;
 	}
 
 	status = i40e_read_nvm_word(hw, pba_ptr, &pba_size);
 	if (status) {
 		hw_dbg(hw, "Failed to read PBA Block size.\n");
-		return status;
+		return;
 	}
 
 	/* Subtract one to get PBA word count (PBA Size word is included in
-	 * total size)
+	 * total size) and advance pointer to first PBA word.
 	 */
 	pba_size--;
-	if (pba_num_size < (((u32)pba_size * 2) + 1)) {
-		hw_dbg(hw, "Buffer to small for PBA data.\n");
-		return I40E_ERR_PARAM;
+	pba_ptr++;
+	if (!pba_size) {
+		hw_dbg(hw, "PBA ID is empty.\n");
+		return;
 	}
 
+	ptr = devm_kzalloc(i40e_hw_to_dev(hw), pba_size * 2 + 1, GFP_KERNEL);
+	if (!ptr)
+		return;
+	hw->pba_id = ptr;
+
 	for (i = 0; i < pba_size; i++) {
-		status = i40e_read_nvm_word(hw, (pba_ptr + 1) + i, &pba_word);
+		status = i40e_read_nvm_word(hw, pba_ptr + i, &pba_word);
 		if (status) {
 			hw_dbg(hw, "Failed to read PBA Block word %d.\n", i);
-			return status;
+			devm_kfree(i40e_hw_to_dev(hw), hw->pba_id);
+			hw->pba_id = NULL;
+			return;
 		}
 
-		pba_num[(i * 2)] = (pba_word >> 8) & 0xFF;
-		pba_num[(i * 2) + 1] = pba_word & 0xFF;
+		*ptr++ = (pba_word >> 8) & 0xFF;
+		*ptr++ = pba_word & 0xFF;
 	}
-	pba_num[(pba_size * 2)] = '\0';
-
-	return status;
 }
 
 /**
@@ -1169,6 +592,8 @@ static enum i40e_media_type i40e_get_media_type(struct i40e_hw *hw)
 		break;
 	case I40E_PHY_TYPE_100BASE_TX:
 	case I40E_PHY_TYPE_1000BASE_T:
+	case I40E_PHY_TYPE_2_5GBASE_T_LINK_STATUS:
+	case I40E_PHY_TYPE_5GBASE_T_LINK_STATUS:
 	case I40E_PHY_TYPE_10GBASE_T:
 		media = I40E_MEDIA_TYPE_BASET;
 		break;
@@ -1180,6 +605,8 @@ static enum i40e_media_type i40e_get_media_type(struct i40e_hw *hw)
 	case I40E_PHY_TYPE_40GBASE_AOC:
 	case I40E_PHY_TYPE_10GBASE_AOC:
 	case I40E_PHY_TYPE_25GBASE_CR:
+	case I40E_PHY_TYPE_25GBASE_AOC:
+	case I40E_PHY_TYPE_25GBASE_ACC:
 		media = I40E_MEDIA_TYPE_DA;
 		break;
 	case I40E_PHY_TYPE_1000BASE_KX:
@@ -1203,6 +630,29 @@ static enum i40e_media_type i40e_get_media_type(struct i40e_hw *hw)
 	return media;
 }
 
+/**
+ * i40e_poll_globr - Poll for Global Reset completion
+ * @hw: pointer to the hardware structure
+ * @retry_limit: how many times to retry before failure
+ **/
+static int i40e_poll_globr(struct i40e_hw *hw,
+			   u32 retry_limit)
+{
+	u32 cnt, reg = 0;
+
+	for (cnt = 0; cnt < retry_limit; cnt++) {
+		reg = rd32(hw, I40E_GLGEN_RSTAT);
+		if (!(reg & I40E_GLGEN_RSTAT_DEVSTATE_MASK))
+			return 0;
+		msleep(100);
+	}
+
+	hw_dbg(hw, "Global reset failed.\n");
+	hw_dbg(hw, "I40E_GLGEN_RSTAT = 0x%x\n", reg);
+
+	return -EIO;
+}
+
 #define I40E_PF_RESET_WAIT_COUNT_A0	200
 #define I40E_PF_RESET_WAIT_COUNT	200
 /**
@@ -1212,7 +662,7 @@ static enum i40e_media_type i40e_get_media_type(struct i40e_hw *hw)
  * Assuming someone else has triggered a global reset,
  * assure the global reset is complete and then reset the PF
  **/
-i40e_status i40e_pf_reset(struct i40e_hw *hw)
+int i40e_pf_reset(struct i40e_hw *hw)
 {
 	u32 cnt = 0;
 	u32 cnt1 = 0;
@@ -1223,9 +673,8 @@ i40e_status i40e_pf_reset(struct i40e_hw *hw)
 	 * The grst delay value is in 100ms units, and we'll wait a
 	 * couple counts longer to be sure we don't just miss the end.
 	 */
-	grst_del = (rd32(hw, I40E_GLGEN_RSTCTL) &
-		    I40E_GLGEN_RSTCTL_GRSTDEL_MASK) >>
-		    I40E_GLGEN_RSTCTL_GRSTDEL_SHIFT;
+	grst_del = FIELD_GET(I40E_GLGEN_RSTCTL_GRSTDEL_MASK,
+			     rd32(hw, I40E_GLGEN_RSTCTL));
 
 	/* It can take upto 15 secs for GRST steady state.
 	 * Bump it to 16 secs max to be safe.
@@ -1240,7 +689,7 @@ i40e_status i40e_pf_reset(struct i40e_hw *hw)
 	}
 	if (reg & I40E_GLGEN_RSTAT_DEVSTATE_MASK) {
 		hw_dbg(hw, "Global reset polling failed to complete.\n");
-		return I40E_ERR_RESET_FAILED;
+		return -EIO;
 	}
 
 	/* Now Wait for the FW to be ready */
@@ -1259,13 +708,14 @@ i40e_status i40e_pf_reset(struct i40e_hw *hw)
 		     I40E_GLNVM_ULD_CONF_GLOBAL_DONE_MASK))) {
 		hw_dbg(hw, "wait for FW Reset complete timedout\n");
 		hw_dbg(hw, "I40E_GLNVM_ULD = 0x%x\n", reg);
-		return I40E_ERR_RESET_FAILED;
+		return -EIO;
 	}
 
 	/* If there was a Global Reset in progress when we got here,
 	 * we don't need to do the PF Reset
 	 */
 	if (!cnt) {
+		u32 reg2 = 0;
 		if (hw->revision_id == 0)
 			cnt = I40E_PF_RESET_WAIT_COUNT_A0;
 		else
@@ -1277,11 +727,17 @@ i40e_status i40e_pf_reset(struct i40e_hw *hw)
 			reg = rd32(hw, I40E_PFGEN_CTRL);
 			if (!(reg & I40E_PFGEN_CTRL_PFSWR_MASK))
 				break;
+			reg2 = rd32(hw, I40E_GLGEN_RSTAT);
+			if (reg2 & I40E_GLGEN_RSTAT_DEVSTATE_MASK)
+				break;
 			usleep_range(1000, 2000);
 		}
-		if (reg & I40E_PFGEN_CTRL_PFSWR_MASK) {
+		if (reg2 & I40E_GLGEN_RSTAT_DEVSTATE_MASK) {
+			if (i40e_poll_globr(hw, grst_del))
+				return -EIO;
+		} else if (reg & I40E_PFGEN_CTRL_PFSWR_MASK) {
 			hw_dbg(hw, "PF reset polling failed to complete.\n");
-			return I40E_ERR_RESET_FAILED;
+			return -EIO;
 		}
 	}
 
@@ -1301,36 +757,31 @@ i40e_status i40e_pf_reset(struct i40e_hw *hw)
 void i40e_clear_hw(struct i40e_hw *hw)
 {
 	u32 num_queues, base_queue;
-	u32 num_pf_int;
-	u32 num_vf_int;
+	s32 num_pf_int;
+	s32 num_vf_int;
 	u32 num_vfs;
-	u32 i, j;
+	s32 i;
+	u32 j;
 	u32 val;
 	u32 eol = 0x7ff;
 
 	/* get number of interrupts, queues, and VFs */
 	val = rd32(hw, I40E_GLPCI_CNF2);
-	num_pf_int = (val & I40E_GLPCI_CNF2_MSI_X_PF_N_MASK) >>
-		     I40E_GLPCI_CNF2_MSI_X_PF_N_SHIFT;
-	num_vf_int = (val & I40E_GLPCI_CNF2_MSI_X_VF_N_MASK) >>
-		     I40E_GLPCI_CNF2_MSI_X_VF_N_SHIFT;
+	num_pf_int = FIELD_GET(I40E_GLPCI_CNF2_MSI_X_PF_N_MASK, val);
+	num_vf_int = FIELD_GET(I40E_GLPCI_CNF2_MSI_X_VF_N_MASK, val);
 
 	val = rd32(hw, I40E_PFLAN_QALLOC);
-	base_queue = (val & I40E_PFLAN_QALLOC_FIRSTQ_MASK) >>
-		     I40E_PFLAN_QALLOC_FIRSTQ_SHIFT;
-	j = (val & I40E_PFLAN_QALLOC_LASTQ_MASK) >>
-	    I40E_PFLAN_QALLOC_LASTQ_SHIFT;
-	if (val & I40E_PFLAN_QALLOC_VALID_MASK)
+	base_queue = FIELD_GET(I40E_PFLAN_QALLOC_FIRSTQ_MASK, val);
+	j = FIELD_GET(I40E_PFLAN_QALLOC_LASTQ_MASK, val);
+	if (val & I40E_PFLAN_QALLOC_VALID_MASK && j >= base_queue)
 		num_queues = (j - base_queue) + 1;
 	else
 		num_queues = 0;
 
 	val = rd32(hw, I40E_PF_VT_PFALLOC);
-	i = (val & I40E_PF_VT_PFALLOC_FIRSTVF_MASK) >>
-	    I40E_PF_VT_PFALLOC_FIRSTVF_SHIFT;
-	j = (val & I40E_PF_VT_PFALLOC_LASTVF_MASK) >>
-	    I40E_PF_VT_PFALLOC_LASTVF_SHIFT;
-	if (val & I40E_PF_VT_PFALLOC_VALID_MASK)
+	i = FIELD_GET(I40E_PF_VT_PFALLOC_FIRSTVF_MASK, val);
+	j = FIELD_GET(I40E_PF_VT_PFALLOC_LASTVF_MASK, val);
+	if (val & I40E_PF_VT_PFALLOC_VALID_MASK && j >= i)
 		num_vfs = (j - i) + 1;
 	else
 		num_vfs = 0;
@@ -1420,12 +871,11 @@ static u32 i40e_led_is_mine(struct i40e_hw *hw, int idx)
 	u32 gpio_val = 0;
 	u32 port;
 
-	if (!hw->func_caps.led[idx])
+	if (!I40E_IS_X710TL_DEVICE(hw->device_id) &&
+	    !hw->func_caps.led[idx])
 		return 0;
-
 	gpio_val = rd32(hw, I40E_GLGEN_GPIO_CTL(idx));
-	port = (gpio_val & I40E_GLGEN_GPIO_CTL_PRT_NUM_MASK) >>
-		I40E_GLGEN_GPIO_CTL_PRT_NUM_SHIFT;
+	port = FIELD_GET(I40E_GLGEN_GPIO_CTL_PRT_NUM_MASK, gpio_val);
 
 	/* if PRT_NUM_NA is 1 then this LED is not port specific, OR
 	 * if it is not our port then ignore
@@ -1437,12 +887,15 @@ static u32 i40e_led_is_mine(struct i40e_hw *hw, int idx)
 	return gpio_val;
 }
 
-#define I40E_COMBINED_ACTIVITY 0xA
-#define I40E_FILTER_ACTIVITY 0xE
-#define I40E_LINK_ACTIVITY 0xC
-#define I40E_MAC_ACTIVITY 0xD
+#define I40E_FW_LED BIT(4)
+#define I40E_LED_MODE_VALID (I40E_GLGEN_GPIO_CTL_LED_MODE_MASK >> \
+			     I40E_GLGEN_GPIO_CTL_LED_MODE_SHIFT)
+
 #define I40E_LED0 22
 
+#define I40E_PIN_FUNC_SDP 0x0
+#define I40E_PIN_FUNC_LED 0x1
+
 /**
  * i40e_led_get - return current on/off mode
  * @hw: pointer to the hw struct
@@ -1454,7 +907,6 @@ static u32 i40e_led_is_mine(struct i40e_hw *hw, int idx)
  **/
 u32 i40e_led_get(struct i40e_hw *hw)
 {
-	u32 current_mode = 0;
 	u32 mode = 0;
 	int i;
 
@@ -1467,22 +919,7 @@ u32 i40e_led_get(struct i40e_hw *hw)
 		if (!gpio_val)
 			continue;
 
-		/* ignore gpio LED src mode entries related to the activity
-		 * LEDs
-		 */
-		current_mode = ((gpio_val & I40E_GLGEN_GPIO_CTL_LED_MODE_MASK)
-				>> I40E_GLGEN_GPIO_CTL_LED_MODE_SHIFT);
-		switch (current_mode) {
-		case I40E_COMBINED_ACTIVITY:
-		case I40E_FILTER_ACTIVITY:
-		case I40E_MAC_ACTIVITY:
-			continue;
-		default:
-			break;
-		}
-
-		mode = (gpio_val & I40E_GLGEN_GPIO_CTL_LED_MODE_MASK) >>
-			I40E_GLGEN_GPIO_CTL_LED_MODE_SHIFT;
+		mode = FIELD_GET(I40E_GLGEN_GPIO_CTL_LED_MODE_MASK, gpio_val);
 		break;
 	}
 
@@ -1500,11 +937,12 @@ u32 i40e_led_get(struct i40e_hw *hw)
  **/
 void i40e_led_set(struct i40e_hw *hw, u32 mode, bool blink)
 {
-	u32 current_mode = 0;
 	int i;
 
-	if (mode & 0xfffffff0)
+	if (mode & ~I40E_LED_MODE_VALID) {
 		hw_dbg(hw, "invalid mode passed in %X\n", mode);
+		return;
+	}
 
 	/* as per the documentation GPIO 22-29 are the LED
 	 * GPIO pins named LED0..LED7
@@ -1515,27 +953,23 @@ void i40e_led_set(struct i40e_hw *hw, u32 mode, bool blink)
 		if (!gpio_val)
 			continue;
 
-		/* ignore gpio LED src mode entries related to the activity
-		 * LEDs
-		 */
-		current_mode = ((gpio_val & I40E_GLGEN_GPIO_CTL_LED_MODE_MASK)
-				>> I40E_GLGEN_GPIO_CTL_LED_MODE_SHIFT);
-		switch (current_mode) {
-		case I40E_COMBINED_ACTIVITY:
-		case I40E_FILTER_ACTIVITY:
-		case I40E_MAC_ACTIVITY:
-			continue;
-		default:
-			break;
-		}
+		if (I40E_IS_X710TL_DEVICE(hw->device_id)) {
+			u32 pin_func = 0;
+
+			if (mode & I40E_FW_LED)
+				pin_func = I40E_PIN_FUNC_SDP;
+			else
+				pin_func = I40E_PIN_FUNC_LED;
 
+			gpio_val &= ~I40E_GLGEN_GPIO_CTL_PIN_FUNC_MASK;
+			gpio_val |=
+				FIELD_PREP(I40E_GLGEN_GPIO_CTL_PIN_FUNC_MASK,
+					   pin_func);
+		}
 		gpio_val &= ~I40E_GLGEN_GPIO_CTL_LED_MODE_MASK;
 		/* this & is a bit of paranoia, but serves as a range check */
-		gpio_val |= ((mode << I40E_GLGEN_GPIO_CTL_LED_MODE_SHIFT) &
-			     I40E_GLGEN_GPIO_CTL_LED_MODE_MASK);
-
-		if (mode == I40E_LINK_ACTIVITY)
-			blink = false;
+		gpio_val |= FIELD_PREP(I40E_GLGEN_GPIO_CTL_LED_MODE_MASK,
+				       mode);
 
 		if (blink)
 			gpio_val |= BIT(I40E_GLGEN_GPIO_CTL_LED_BLINK_SHIFT);
@@ -1559,42 +993,69 @@ void i40e_led_set(struct i40e_hw *hw, u32 mode, bool blink)
  *
  * Returns the various PHY abilities supported on the Port.
  **/
-i40e_status i40e_aq_get_phy_capabilities(struct i40e_hw *hw,
-			bool qualified_modules, bool report_init,
-			struct i40e_aq_get_phy_abilities_resp *abilities,
-			struct i40e_asq_cmd_details *cmd_details)
+int
+i40e_aq_get_phy_capabilities(struct i40e_hw *hw,
+			     bool qualified_modules, bool report_init,
+			     struct i40e_aq_get_phy_abilities_resp *abilities,
+			     struct i40e_asq_cmd_details *cmd_details)
 {
-	struct i40e_aq_desc desc;
-	i40e_status status;
 	u16 abilities_size = sizeof(struct i40e_aq_get_phy_abilities_resp);
+	u16 max_delay = I40E_MAX_PHY_TIMEOUT, total_delay = 0;
+	struct libie_aq_desc desc;
+	int status;
 
 	if (!abilities)
-		return I40E_ERR_PARAM;
+		return -EINVAL;
 
-	i40e_fill_default_direct_cmd_desc(&desc,
-					  i40e_aqc_opc_get_phy_abilities);
+	do {
+		i40e_fill_default_direct_cmd_desc(&desc,
+					       i40e_aqc_opc_get_phy_abilities);
 
-	desc.flags |= cpu_to_le16((u16)I40E_AQ_FLAG_BUF);
-	if (abilities_size > I40E_AQ_LARGE_BUF)
-		desc.flags |= cpu_to_le16((u16)I40E_AQ_FLAG_LB);
+		desc.flags |= cpu_to_le16((u16)LIBIE_AQ_FLAG_BUF);
+		if (abilities_size > I40E_AQ_LARGE_BUF)
+			desc.flags |= cpu_to_le16((u16)LIBIE_AQ_FLAG_LB);
 
-	if (qualified_modules)
-		desc.params.external.param0 |=
+		if (qualified_modules)
+			desc.params.generic.param0 |=
 			cpu_to_le32(I40E_AQ_PHY_REPORT_QUALIFIED_MODULES);
 
-	if (report_init)
-		desc.params.external.param0 |=
+		if (report_init)
+			desc.params.generic.param0 |=
 			cpu_to_le32(I40E_AQ_PHY_REPORT_INITIAL_VALUES);
 
-	status = i40e_asq_send_command(hw, &desc, abilities, abilities_size,
-				       cmd_details);
+		status = i40e_asq_send_command(hw, &desc, abilities,
+					       abilities_size, cmd_details);
 
-	if (hw->aq.asq_last_status == I40E_AQ_RC_EIO)
-		status = I40E_ERR_UNKNOWN_PHY;
+		switch (hw->aq.asq_last_status) {
+		case LIBIE_AQ_RC_EIO:
+			status = -EIO;
+			break;
+		case LIBIE_AQ_RC_EAGAIN:
+			usleep_range(1000, 2000);
+			total_delay++;
+			status = -EIO;
+			break;
+		/* also covers LIBIE_AQ_RC_OK */
+		default:
+			break;
+		}
+
+	} while ((hw->aq.asq_last_status == LIBIE_AQ_RC_EAGAIN) &&
+		(total_delay < max_delay));
+
+	if (status)
+		return status;
 
 	if (report_init) {
-		hw->phy.phy_types = le32_to_cpu(abilities->phy_type);
-		hw->phy.phy_types |= ((u64)abilities->phy_type_ext << 32);
+		if (hw->mac.type ==  I40E_MAC_XL710 &&
+		    i40e_is_aq_api_ver_ge(hw, I40E_FW_API_VERSION_MAJOR,
+					  I40E_MINOR_VER_GET_LINK_INFO_XL710)) {
+			status = i40e_aq_get_link_info(hw, true, NULL, NULL);
+		} else {
+			hw->phy.phy_types = le32_to_cpu(abilities->phy_type);
+			hw->phy.phy_types |=
+					((u64)abilities->phy_type_ext << 32);
+		}
 	}
 
 	return status;
@@ -1612,21 +1073,21 @@ i40e_status i40e_aq_get_phy_capabilities(struct i40e_hw *hw,
  * of the PHY Config parameters. This status will be indicated by the
  * command response.
  **/
-enum i40e_status_code i40e_aq_set_phy_config(struct i40e_hw *hw,
-				struct i40e_aq_set_phy_config *config,
-				struct i40e_asq_cmd_details *cmd_details)
+int i40e_aq_set_phy_config(struct i40e_hw *hw,
+			   struct i40e_aq_set_phy_config *config,
+			   struct i40e_asq_cmd_details *cmd_details)
 {
-	struct i40e_aq_desc desc;
-	struct i40e_aq_set_phy_config *cmd =
-			(struct i40e_aq_set_phy_config *)&desc.params.raw;
-	enum i40e_status_code status;
+	struct i40e_aq_set_phy_config *cmd;
+	struct libie_aq_desc desc;
+	int status;
 
 	if (!config)
-		return I40E_ERR_PARAM;
+		return -EINVAL;
 
 	i40e_fill_default_direct_cmd_desc(&desc,
 					  i40e_aqc_opc_set_phy_config);
 
+	cmd = libie_aq_raw(&desc);
 	*cmd = *config;
 
 	status = i40e_asq_send_command(hw, &desc, NULL, 0, cmd_details);
@@ -1634,23 +1095,15 @@ enum i40e_status_code i40e_aq_set_phy_config(struct i40e_hw *hw,
 	return status;
 }
 
-/**
- * i40e_set_fc
- * @hw: pointer to the hw struct
- *
- * Set the requested flow control mode using set_phy_config.
- **/
-enum i40e_status_code i40e_set_fc(struct i40e_hw *hw, u8 *aq_failures,
-				  bool atomic_restart)
+static noinline_for_stack int
+i40e_set_fc_status(struct i40e_hw *hw,
+		   struct i40e_aq_get_phy_abilities_resp *abilities,
+		   bool atomic_restart)
 {
-	enum i40e_fc_mode fc_mode = hw->fc.requested_mode;
-	struct i40e_aq_get_phy_abilities_resp abilities;
 	struct i40e_aq_set_phy_config config;
-	enum i40e_status_code status;
+	enum i40e_fc_mode fc_mode = hw->fc.requested_mode;
 	u8 pause_mask = 0x0;
 
-	*aq_failures = 0x0;
-
 	switch (fc_mode) {
 	case I40E_FC_FULL:
 		pause_mask |= I40E_AQ_PHY_FLAG_PAUSE_TX;
@@ -1666,6 +1119,48 @@ enum i40e_status_code i40e_set_fc(struct i40e_hw *hw, u8 *aq_failures,
 		break;
 	}
 
+	memset(&config, 0, sizeof(struct i40e_aq_set_phy_config));
+	/* clear the old pause settings */
+	config.abilities = abilities->abilities & ~(I40E_AQ_PHY_FLAG_PAUSE_TX) &
+			   ~(I40E_AQ_PHY_FLAG_PAUSE_RX);
+	/* set the new abilities */
+	config.abilities |= pause_mask;
+	/* If the abilities have changed, then set the new config */
+	if (config.abilities == abilities->abilities)
+		return 0;
+
+	/* Auto restart link so settings take effect */
+	if (atomic_restart)
+		config.abilities |= I40E_AQ_PHY_ENABLE_ATOMIC_LINK;
+	/* Copy over all the old settings */
+	config.phy_type = abilities->phy_type;
+	config.phy_type_ext = abilities->phy_type_ext;
+	config.link_speed = abilities->link_speed;
+	config.eee_capability = abilities->eee_capability;
+	config.eeer = abilities->eeer_val;
+	config.low_power_ctrl = abilities->d3_lpan;
+	config.fec_config = abilities->fec_cfg_curr_mod_ext_info &
+			    I40E_AQ_PHY_FEC_CONFIG_MASK;
+
+	return i40e_aq_set_phy_config(hw, &config, NULL);
+}
+
+/**
+ * i40e_set_fc
+ * @hw: pointer to the hw struct
+ * @aq_failures: buffer to return AdminQ failure information
+ * @atomic_restart: whether to enable atomic link restart
+ *
+ * Set the requested flow control mode using set_phy_config.
+ **/
+int i40e_set_fc(struct i40e_hw *hw, u8 *aq_failures,
+		bool atomic_restart)
+{
+	struct i40e_aq_get_phy_abilities_resp abilities;
+	int status;
+
+	*aq_failures = 0x0;
+
 	/* Get the current phy config */
 	status = i40e_aq_get_phy_capabilities(hw, false, false, &abilities,
 					      NULL);
@@ -1674,31 +1169,10 @@ enum i40e_status_code i40e_set_fc(struct i40e_hw *hw, u8 *aq_failures,
 		return status;
 	}
 
-	memset(&config, 0, sizeof(struct i40e_aq_set_phy_config));
-	/* clear the old pause settings */
-	config.abilities = abilities.abilities & ~(I40E_AQ_PHY_FLAG_PAUSE_TX) &
-			   ~(I40E_AQ_PHY_FLAG_PAUSE_RX);
-	/* set the new abilities */
-	config.abilities |= pause_mask;
-	/* If the abilities have changed, then set the new config */
-	if (config.abilities != abilities.abilities) {
-		/* Auto restart link so settings take effect */
-		if (atomic_restart)
-			config.abilities |= I40E_AQ_PHY_ENABLE_ATOMIC_LINK;
-		/* Copy over all the old settings */
-		config.phy_type = abilities.phy_type;
-		config.phy_type_ext = abilities.phy_type_ext;
-		config.link_speed = abilities.link_speed;
-		config.eee_capability = abilities.eee_capability;
-		config.eeer = abilities.eeer_val;
-		config.low_power_ctrl = abilities.d3_lpan;
-		config.fec_config = abilities.fec_cfg_curr_mod_ext_info &
-				    I40E_AQ_PHY_FEC_CONFIG_MASK;
-		status = i40e_aq_set_phy_config(hw, &config, NULL);
+	status = i40e_set_fc_status(hw, &abilities, atomic_restart);
+	if (status)
+		*aq_failures |= I40E_SET_FC_AQ_FAIL_SET;
 
-		if (status)
-			*aq_failures |= I40E_SET_FC_AQ_FAIL_SET;
-	}
 	/* Update the link info */
 	status = i40e_update_link_info(hw);
 	if (status) {
@@ -1716,23 +1190,57 @@ enum i40e_status_code i40e_set_fc(struct i40e_hw *hw, u8 *aq_failures,
 }
 
 /**
+ * i40e_aq_set_mac_config - Configure MAC settings
+ * @hw: pointer to the hw struct
+ * @max_frame_size: Maximum Frame Size to be supported by the port
+ * @cmd_details: pointer to command details structure or NULL
+ *
+ * Set MAC configuration (0x0603). Note that max_frame_size must be greater
+ * than zero.
+ *
+ * Return: 0 on success, or a negative error code on failure.
+ */
+int i40e_aq_set_mac_config(struct i40e_hw *hw, u16 max_frame_size,
+			   struct i40e_asq_cmd_details *cmd_details)
+{
+	struct i40e_aq_set_mac_config *cmd;
+	struct libie_aq_desc desc;
+
+	cmd = libie_aq_raw(&desc);
+
+	if (max_frame_size == 0)
+		return -EINVAL;
+
+	i40e_fill_default_direct_cmd_desc(&desc, i40e_aqc_opc_set_mac_config);
+
+	cmd->max_frame_size = cpu_to_le16(max_frame_size);
+	cmd->params = I40E_AQ_SET_MAC_CONFIG_CRC_EN;
+
+#define I40E_AQ_SET_MAC_CONFIG_FC_DEFAULT_THRESHOLD	0x7FFF
+	cmd->fc_refresh_threshold =
+		cpu_to_le16(I40E_AQ_SET_MAC_CONFIG_FC_DEFAULT_THRESHOLD);
+
+	return i40e_asq_send_command(hw, &desc, NULL, 0, cmd_details);
+}
+
+/**
  * i40e_aq_clear_pxe_mode
  * @hw: pointer to the hw struct
  * @cmd_details: pointer to command details structure or NULL
  *
  * Tell the firmware that the driver is taking over from PXE
  **/
-i40e_status i40e_aq_clear_pxe_mode(struct i40e_hw *hw,
-				struct i40e_asq_cmd_details *cmd_details)
+int i40e_aq_clear_pxe_mode(struct i40e_hw *hw,
+			   struct i40e_asq_cmd_details *cmd_details)
 {
-	i40e_status status;
-	struct i40e_aq_desc desc;
-	struct i40e_aqc_clear_pxe *cmd =
-		(struct i40e_aqc_clear_pxe *)&desc.params.raw;
+	struct i40e_aqc_clear_pxe *cmd;
+	struct libie_aq_desc desc;
+	int status;
 
 	i40e_fill_default_direct_cmd_desc(&desc,
 					  i40e_aqc_opc_clear_pxe_mode);
 
+	cmd = libie_aq_raw(&desc);
 	cmd->rx_cnt = 0x2;
 
 	status = i40e_asq_send_command(hw, &desc, NULL, 0, cmd_details);
@@ -1750,18 +1258,18 @@ i40e_status i40e_aq_clear_pxe_mode(struct i40e_hw *hw,
  *
  * Sets up the link and restarts the Auto-Negotiation over the link.
  **/
-i40e_status i40e_aq_set_link_restart_an(struct i40e_hw *hw,
-					bool enable_link,
-					struct i40e_asq_cmd_details *cmd_details)
+int i40e_aq_set_link_restart_an(struct i40e_hw *hw,
+				bool enable_link,
+				struct i40e_asq_cmd_details *cmd_details)
 {
-	struct i40e_aq_desc desc;
-	struct i40e_aqc_set_link_restart_an *cmd =
-		(struct i40e_aqc_set_link_restart_an *)&desc.params.raw;
-	i40e_status status;
+	struct i40e_aqc_set_link_restart_an *cmd;
+	struct libie_aq_desc desc;
+	int status;
 
 	i40e_fill_default_direct_cmd_desc(&desc,
 					  i40e_aqc_opc_set_link_restart_an);
 
+	cmd = libie_aq_raw(&desc);
 	cmd->command = I40E_AQ_PHY_RESTART_AN;
 	if (enable_link)
 		cmd->command |= I40E_AQ_PHY_LINK_ENABLE;
@@ -1782,20 +1290,20 @@ i40e_status i40e_aq_set_link_restart_an(struct i40e_hw *hw,
  *
  * Returns the link status of the adapter.
  **/
-i40e_status i40e_aq_get_link_info(struct i40e_hw *hw,
-				bool enable_lse, struct i40e_link_status *link,
-				struct i40e_asq_cmd_details *cmd_details)
+int i40e_aq_get_link_info(struct i40e_hw *hw,
+			  bool enable_lse, struct i40e_link_status *link,
+			  struct i40e_asq_cmd_details *cmd_details)
 {
-	struct i40e_aq_desc desc;
-	struct i40e_aqc_get_link_status *resp =
-		(struct i40e_aqc_get_link_status *)&desc.params.raw;
 	struct i40e_link_status *hw_link_info = &hw->phy.link_info;
-	i40e_status status;
+	struct i40e_aqc_get_link_status *resp;
+	struct libie_aq_desc desc;
 	bool tx_pause, rx_pause;
 	u16 command_flags;
+	int status;
 
 	i40e_fill_default_direct_cmd_desc(&desc, i40e_aqc_opc_get_link_status);
 
+	resp = libie_aq_raw(&desc);
 	if (enable_lse)
 		command_flags = I40E_AQ_LSE_ENABLE;
 	else
@@ -1819,7 +1327,7 @@ i40e_status i40e_aq_get_link_info(struct i40e_hw *hw,
 	hw_link_info->fec_info = resp->config & (I40E_AQ_CONFIG_FEC_KR_ENA |
 						 I40E_AQ_CONFIG_FEC_RS_ENA);
 	hw_link_info->ext_info = resp->ext_info;
-	hw_link_info->loopback = resp->loopback;
+	hw_link_info->loopback = resp->loopback & I40E_AQ_LOOPBACK_MASK;
 	hw_link_info->max_frame_size = le16_to_cpu(resp->max_frame_size);
 	hw_link_info->pacing = resp->config & I40E_AQ_CONFIG_PACING_MASK;
 
@@ -1845,11 +1353,19 @@ i40e_status i40e_aq_get_link_info(struct i40e_hw *hw,
 	else
 		hw_link_info->lse_enable = false;
 
-	if ((hw->mac.type == I40E_MAC_XL710) &&
-	    (hw->aq.fw_maj_ver < 4 || (hw->aq.fw_maj_ver == 4 &&
-	     hw->aq.fw_min_ver < 40)) && hw_link_info->phy_type == 0xE)
+	if (hw->mac.type == I40E_MAC_XL710 && i40e_is_fw_ver_lt(hw, 4, 40) &&
+	    hw_link_info->phy_type == 0xE)
 		hw_link_info->phy_type = I40E_PHY_TYPE_10GBASE_SFPP_CU;
 
+	if (test_bit(I40E_HW_CAP_AQ_PHY_ACCESS, hw->caps) &&
+	    hw->mac.type != I40E_MAC_X722) {
+		__le32 tmp;
+
+		memcpy(&tmp, resp->link_type, sizeof(tmp));
+		hw->phy.phy_types = le32_to_cpu(tmp);
+		hw->phy.phy_types |= ((u64)resp->link_type_ext << 32);
+	}
+
 	/* save link status information */
 	if (link)
 		*link = *hw_link_info;
@@ -1869,18 +1385,18 @@ aq_get_link_info_exit:
  *
  * Set link interrupt mask.
  **/
-i40e_status i40e_aq_set_phy_int_mask(struct i40e_hw *hw,
-				     u16 mask,
-				     struct i40e_asq_cmd_details *cmd_details)
+int i40e_aq_set_phy_int_mask(struct i40e_hw *hw,
+			     u16 mask,
+			     struct i40e_asq_cmd_details *cmd_details)
 {
-	struct i40e_aq_desc desc;
-	struct i40e_aqc_set_phy_int_mask *cmd =
-		(struct i40e_aqc_set_phy_int_mask *)&desc.params.raw;
-	i40e_status status;
+	struct i40e_aqc_set_phy_int_mask *cmd;
+	struct libie_aq_desc desc;
+	int status;
 
 	i40e_fill_default_direct_cmd_desc(&desc,
 					  i40e_aqc_opc_set_phy_int_mask);
 
+	cmd = libie_aq_raw(&desc);
 	cmd->event_mask = cpu_to_le16(mask);
 
 	status = i40e_asq_send_command(hw, &desc, NULL, 0, cmd_details);
@@ -1889,6 +1405,32 @@ i40e_status i40e_aq_set_phy_int_mask(struct i40e_hw *hw,
 }
 
 /**
+ * i40e_aq_set_mac_loopback
+ * @hw: pointer to the HW struct
+ * @ena_lpbk: Enable or Disable loopback
+ * @cmd_details: pointer to command details structure or NULL
+ *
+ * Enable/disable loopback on a given port
+ */
+int i40e_aq_set_mac_loopback(struct i40e_hw *hw, bool ena_lpbk,
+			     struct i40e_asq_cmd_details *cmd_details)
+{
+	struct i40e_aqc_set_lb_mode *cmd;
+	struct libie_aq_desc desc;
+
+	i40e_fill_default_direct_cmd_desc(&desc, i40e_aqc_opc_set_lb_modes);
+	cmd = libie_aq_raw(&desc);
+	if (ena_lpbk) {
+		if (hw->nvm.version <= I40E_LEGACY_LOOPBACK_NVM_VER)
+			cmd->lb_mode = cpu_to_le16(I40E_AQ_LB_MAC_LOCAL_LEGACY);
+		else
+			cmd->lb_mode = cpu_to_le16(I40E_AQ_LB_MAC_LOCAL);
+	}
+
+	return i40e_asq_send_command(hw, &desc, NULL, 0, cmd_details);
+}
+
+/**
  * i40e_aq_set_phy_debug
  * @hw: pointer to the hw struct
  * @cmd_flags: debug command flags
@@ -1896,17 +1438,17 @@ i40e_status i40e_aq_set_phy_int_mask(struct i40e_hw *hw,
  *
  * Reset the external PHY.
  **/
-i40e_status i40e_aq_set_phy_debug(struct i40e_hw *hw, u8 cmd_flags,
-				  struct i40e_asq_cmd_details *cmd_details)
+int i40e_aq_set_phy_debug(struct i40e_hw *hw, u8 cmd_flags,
+			  struct i40e_asq_cmd_details *cmd_details)
 {
-	struct i40e_aq_desc desc;
-	struct i40e_aqc_set_phy_debug *cmd =
-		(struct i40e_aqc_set_phy_debug *)&desc.params.raw;
-	i40e_status status;
+	struct i40e_aqc_set_phy_debug *cmd;
+	struct libie_aq_desc desc;
+	int status;
 
 	i40e_fill_default_direct_cmd_desc(&desc,
 					  i40e_aqc_opc_set_phy_debug);
 
+	cmd = libie_aq_raw(&desc);
 	cmd->command_flags = cmd_flags;
 
 	status = i40e_asq_send_command(hw, &desc, NULL, 0, cmd_details);
@@ -1922,30 +1464,30 @@ i40e_status i40e_aq_set_phy_debug(struct i40e_hw *hw, u8 cmd_flags,
  *
  * Add a VSI context to the hardware.
 **/
-i40e_status i40e_aq_add_vsi(struct i40e_hw *hw,
-				struct i40e_vsi_context *vsi_ctx,
-				struct i40e_asq_cmd_details *cmd_details)
+int i40e_aq_add_vsi(struct i40e_hw *hw,
+		    struct i40e_vsi_context *vsi_ctx,
+		    struct i40e_asq_cmd_details *cmd_details)
 {
-	struct i40e_aq_desc desc;
-	struct i40e_aqc_add_get_update_vsi *cmd =
-		(struct i40e_aqc_add_get_update_vsi *)&desc.params.raw;
-	struct i40e_aqc_add_get_update_vsi_completion *resp =
-		(struct i40e_aqc_add_get_update_vsi_completion *)
-		&desc.params.raw;
-	i40e_status status;
+	struct i40e_aqc_add_get_update_vsi_completion *resp;
+	struct i40e_aqc_add_get_update_vsi *cmd;
+	struct libie_aq_desc desc;
+	int status;
 
 	i40e_fill_default_direct_cmd_desc(&desc,
 					  i40e_aqc_opc_add_vsi);
 
+	resp = libie_aq_raw(&desc);
+	cmd = libie_aq_raw(&desc);
 	cmd->uplink_seid = cpu_to_le16(vsi_ctx->uplink_seid);
 	cmd->connection_type = vsi_ctx->connection_type;
 	cmd->vf_id = vsi_ctx->vf_num;
 	cmd->vsi_flags = cpu_to_le16(vsi_ctx->flags);
 
-	desc.flags |= cpu_to_le16((u16)(I40E_AQ_FLAG_BUF | I40E_AQ_FLAG_RD));
+	desc.flags |= cpu_to_le16((u16)(LIBIE_AQ_FLAG_BUF | LIBIE_AQ_FLAG_RD));
 
-	status = i40e_asq_send_command(hw, &desc, &vsi_ctx->info,
-				    sizeof(vsi_ctx->info), cmd_details);
+	status = i40e_asq_send_command_atomic(hw, &desc, &vsi_ctx->info,
+					      sizeof(vsi_ctx->info),
+					      cmd_details, true);
 
 	if (status)
 		goto aq_add_vsi_exit;
@@ -1965,19 +1507,18 @@ aq_add_vsi_exit:
  * @seid: vsi number
  * @cmd_details: pointer to command details structure or NULL
  **/
-i40e_status i40e_aq_set_default_vsi(struct i40e_hw *hw,
-				    u16 seid,
-				    struct i40e_asq_cmd_details *cmd_details)
+int i40e_aq_set_default_vsi(struct i40e_hw *hw,
+			    u16 seid,
+			    struct i40e_asq_cmd_details *cmd_details)
 {
-	struct i40e_aq_desc desc;
-	struct i40e_aqc_set_vsi_promiscuous_modes *cmd =
-		(struct i40e_aqc_set_vsi_promiscuous_modes *)
-		&desc.params.raw;
-	i40e_status status;
+	struct i40e_aqc_set_vsi_promiscuous_modes *cmd;
+	struct libie_aq_desc desc;
+	int status;
 
 	i40e_fill_default_direct_cmd_desc(&desc,
 					  i40e_aqc_opc_set_vsi_promiscuous_modes);
 
+	cmd = libie_aq_raw(&desc);
 	cmd->promiscuous_flags = cpu_to_le16(I40E_AQC_SET_VSI_DEFAULT);
 	cmd->valid_flags = cpu_to_le16(I40E_AQC_SET_VSI_DEFAULT);
 	cmd->seid = cpu_to_le16(seid);
@@ -1993,19 +1534,18 @@ i40e_status i40e_aq_set_default_vsi(struct i40e_hw *hw,
  * @seid: vsi number
  * @cmd_details: pointer to command details structure or NULL
  **/
-i40e_status i40e_aq_clear_default_vsi(struct i40e_hw *hw,
-				      u16 seid,
-				      struct i40e_asq_cmd_details *cmd_details)
+int i40e_aq_clear_default_vsi(struct i40e_hw *hw,
+			      u16 seid,
+			      struct i40e_asq_cmd_details *cmd_details)
 {
-	struct i40e_aq_desc desc;
-	struct i40e_aqc_set_vsi_promiscuous_modes *cmd =
-		(struct i40e_aqc_set_vsi_promiscuous_modes *)
-		&desc.params.raw;
-	i40e_status status;
+	struct i40e_aqc_set_vsi_promiscuous_modes *cmd;
+	struct libie_aq_desc desc;
+	int status;
 
 	i40e_fill_default_direct_cmd_desc(&desc,
 					  i40e_aqc_opc_set_vsi_promiscuous_modes);
 
+	cmd = libie_aq_raw(&desc);
 	cmd->promiscuous_flags = cpu_to_le16(0);
 	cmd->valid_flags = cpu_to_le16(I40E_AQC_SET_VSI_DEFAULT);
 	cmd->seid = cpu_to_le16(seid);
@@ -2023,34 +1563,32 @@ i40e_status i40e_aq_clear_default_vsi(struct i40e_hw *hw,
  * @cmd_details: pointer to command details structure or NULL
  * @rx_only_promisc: flag to decide if egress traffic gets mirrored in promisc
  **/
-i40e_status i40e_aq_set_vsi_unicast_promiscuous(struct i40e_hw *hw,
-				u16 seid, bool set,
-				struct i40e_asq_cmd_details *cmd_details,
-				bool rx_only_promisc)
-{
-	struct i40e_aq_desc desc;
-	struct i40e_aqc_set_vsi_promiscuous_modes *cmd =
-		(struct i40e_aqc_set_vsi_promiscuous_modes *)&desc.params.raw;
-	i40e_status status;
+int i40e_aq_set_vsi_unicast_promiscuous(struct i40e_hw *hw,
+					u16 seid, bool set,
+					struct i40e_asq_cmd_details *cmd_details,
+					bool rx_only_promisc)
+{
+	struct i40e_aqc_set_vsi_promiscuous_modes *cmd;
+	struct libie_aq_desc desc;
 	u16 flags = 0;
+	int status;
 
 	i40e_fill_default_direct_cmd_desc(&desc,
 					i40e_aqc_opc_set_vsi_promiscuous_modes);
 
+	cmd = libie_aq_raw(&desc);
 	if (set) {
 		flags |= I40E_AQC_SET_VSI_PROMISC_UNICAST;
-		if (rx_only_promisc &&
-		    (((hw->aq.api_maj_ver == 1) && (hw->aq.api_min_ver >= 5)) ||
-		     (hw->aq.api_maj_ver > 1)))
-			flags |= I40E_AQC_SET_VSI_PROMISC_TX;
+		if (rx_only_promisc && i40e_is_aq_api_ver_ge(hw, 1, 5))
+			flags |= I40E_AQC_SET_VSI_PROMISC_RX_ONLY;
 	}
 
 	cmd->promiscuous_flags = cpu_to_le16(flags);
 
 	cmd->valid_flags = cpu_to_le16(I40E_AQC_SET_VSI_PROMISC_UNICAST);
-	if (((hw->aq.api_maj_ver >= 1) && (hw->aq.api_min_ver >= 5)) ||
-	    (hw->aq.api_maj_ver > 1))
-		cmd->valid_flags |= cpu_to_le16(I40E_AQC_SET_VSI_PROMISC_TX);
+	if (i40e_is_aq_api_ver_ge(hw, 1, 5))
+		cmd->valid_flags |=
+			cpu_to_le16(I40E_AQC_SET_VSI_PROMISC_RX_ONLY);
 
 	cmd->seid = cpu_to_le16(seid);
 	status = i40e_asq_send_command(hw, &desc, NULL, 0, cmd_details);
@@ -2065,18 +1603,19 @@ i40e_status i40e_aq_set_vsi_unicast_promiscuous(struct i40e_hw *hw,
  * @set: set multicast promiscuous enable/disable
  * @cmd_details: pointer to command details structure or NULL
  **/
-i40e_status i40e_aq_set_vsi_multicast_promiscuous(struct i40e_hw *hw,
-				u16 seid, bool set, struct i40e_asq_cmd_details *cmd_details)
+int i40e_aq_set_vsi_multicast_promiscuous(struct i40e_hw *hw,
+					  u16 seid, bool set,
+					  struct i40e_asq_cmd_details *cmd_details)
 {
-	struct i40e_aq_desc desc;
-	struct i40e_aqc_set_vsi_promiscuous_modes *cmd =
-		(struct i40e_aqc_set_vsi_promiscuous_modes *)&desc.params.raw;
-	i40e_status status;
+	struct i40e_aqc_set_vsi_promiscuous_modes *cmd;
+	struct libie_aq_desc desc;
 	u16 flags = 0;
+	int status;
 
 	i40e_fill_default_direct_cmd_desc(&desc,
 					i40e_aqc_opc_set_vsi_promiscuous_modes);
 
+	cmd = libie_aq_raw(&desc);
 	if (set)
 		flags |= I40E_AQC_SET_VSI_PROMISC_MULTICAST;
 
@@ -2098,20 +1637,20 @@ i40e_status i40e_aq_set_vsi_multicast_promiscuous(struct i40e_hw *hw,
  * @vid: The VLAN tag filter - capture any multicast packet with this VLAN tag
  * @cmd_details: pointer to command details structure or NULL
  **/
-enum i40e_status_code i40e_aq_set_vsi_mc_promisc_on_vlan(struct i40e_hw *hw,
-							 u16 seid, bool enable,
-							 u16 vid,
-				struct i40e_asq_cmd_details *cmd_details)
+int i40e_aq_set_vsi_mc_promisc_on_vlan(struct i40e_hw *hw,
+				       u16 seid, bool enable,
+				       u16 vid,
+				       struct i40e_asq_cmd_details *cmd_details)
 {
-	struct i40e_aq_desc desc;
-	struct i40e_aqc_set_vsi_promiscuous_modes *cmd =
-		(struct i40e_aqc_set_vsi_promiscuous_modes *)&desc.params.raw;
-	enum i40e_status_code status;
+	struct i40e_aqc_set_vsi_promiscuous_modes *cmd;
+	struct libie_aq_desc desc;
 	u16 flags = 0;
+	int status;
 
 	i40e_fill_default_direct_cmd_desc(&desc,
 					  i40e_aqc_opc_set_vsi_promiscuous_modes);
 
+	cmd = libie_aq_raw(&desc);
 	if (enable)
 		flags |= I40E_AQC_SET_VSI_PROMISC_MULTICAST;
 
@@ -2120,7 +1659,8 @@ enum i40e_status_code i40e_aq_set_vsi_mc_promisc_on_vlan(struct i40e_hw *hw,
 	cmd->seid = cpu_to_le16(seid);
 	cmd->vlan_tag = cpu_to_le16(vid | I40E_AQC_SET_VSI_VLAN_VALID);
 
-	status = i40e_asq_send_command(hw, &desc, NULL, 0, cmd_details);
+	status = i40e_asq_send_command_atomic(hw, &desc, NULL, 0,
+					      cmd_details, true);
 
 	return status;
 }
@@ -2133,29 +1673,36 @@ enum i40e_status_code i40e_aq_set_vsi_mc_promisc_on_vlan(struct i40e_hw *hw,
  * @vid: The VLAN tag filter - capture any unicast packet with this VLAN tag
  * @cmd_details: pointer to command details structure or NULL
  **/
-enum i40e_status_code i40e_aq_set_vsi_uc_promisc_on_vlan(struct i40e_hw *hw,
-							 u16 seid, bool enable,
-							 u16 vid,
-				struct i40e_asq_cmd_details *cmd_details)
+int i40e_aq_set_vsi_uc_promisc_on_vlan(struct i40e_hw *hw,
+				       u16 seid, bool enable,
+				       u16 vid,
+				       struct i40e_asq_cmd_details *cmd_details)
 {
-	struct i40e_aq_desc desc;
-	struct i40e_aqc_set_vsi_promiscuous_modes *cmd =
-		(struct i40e_aqc_set_vsi_promiscuous_modes *)&desc.params.raw;
-	enum i40e_status_code status;
+	struct i40e_aqc_set_vsi_promiscuous_modes *cmd;
+	struct libie_aq_desc desc;
 	u16 flags = 0;
+	int status;
 
 	i40e_fill_default_direct_cmd_desc(&desc,
 					  i40e_aqc_opc_set_vsi_promiscuous_modes);
 
-	if (enable)
+	cmd = libie_aq_raw(&desc);
+	if (enable) {
 		flags |= I40E_AQC_SET_VSI_PROMISC_UNICAST;
+		if (i40e_is_aq_api_ver_ge(hw, 1, 5))
+			flags |= I40E_AQC_SET_VSI_PROMISC_RX_ONLY;
+	}
 
 	cmd->promiscuous_flags = cpu_to_le16(flags);
 	cmd->valid_flags = cpu_to_le16(I40E_AQC_SET_VSI_PROMISC_UNICAST);
+	if (i40e_is_aq_api_ver_ge(hw, 1, 5))
+		cmd->valid_flags |=
+			cpu_to_le16(I40E_AQC_SET_VSI_PROMISC_RX_ONLY);
 	cmd->seid = cpu_to_le16(seid);
 	cmd->vlan_tag = cpu_to_le16(vid | I40E_AQC_SET_VSI_VLAN_VALID);
 
-	status = i40e_asq_send_command(hw, &desc, NULL, 0, cmd_details);
+	status = i40e_asq_send_command_atomic(hw, &desc, NULL, 0,
+					      cmd_details, true);
 
 	return status;
 }
@@ -2168,15 +1715,14 @@ enum i40e_status_code i40e_aq_set_vsi_uc_promisc_on_vlan(struct i40e_hw *hw,
  * @vid: The VLAN tag filter - capture any broadcast packet with this VLAN tag
  * @cmd_details: pointer to command details structure or NULL
  **/
-i40e_status i40e_aq_set_vsi_bc_promisc_on_vlan(struct i40e_hw *hw,
-				u16 seid, bool enable, u16 vid,
-				struct i40e_asq_cmd_details *cmd_details)
+int i40e_aq_set_vsi_bc_promisc_on_vlan(struct i40e_hw *hw,
+				       u16 seid, bool enable, u16 vid,
+				       struct i40e_asq_cmd_details *cmd_details)
 {
-	struct i40e_aq_desc desc;
-	struct i40e_aqc_set_vsi_promiscuous_modes *cmd =
-		(struct i40e_aqc_set_vsi_promiscuous_modes *)&desc.params.raw;
-	i40e_status status;
+	struct i40e_aqc_set_vsi_promiscuous_modes *cmd;
+	struct libie_aq_desc desc;
 	u16 flags = 0;
+	int status;
 
 	i40e_fill_default_direct_cmd_desc(&desc,
 					i40e_aqc_opc_set_vsi_promiscuous_modes);
@@ -2184,6 +1730,7 @@ i40e_status i40e_aq_set_vsi_bc_promisc_on_vlan(struct i40e_hw *hw,
 	if (enable)
 		flags |= I40E_AQC_SET_VSI_PROMISC_BROADCAST;
 
+	cmd = libie_aq_raw(&desc);
 	cmd->promiscuous_flags = cpu_to_le16(flags);
 	cmd->valid_flags = cpu_to_le16(I40E_AQC_SET_VSI_PROMISC_BROADCAST);
 	cmd->seid = cpu_to_le16(seid);
@@ -2203,18 +1750,18 @@ i40e_status i40e_aq_set_vsi_bc_promisc_on_vlan(struct i40e_hw *hw,
  *
  * Set or clear the broadcast promiscuous flag (filter) for a given VSI.
  **/
-i40e_status i40e_aq_set_vsi_broadcast(struct i40e_hw *hw,
-				u16 seid, bool set_filter,
-				struct i40e_asq_cmd_details *cmd_details)
+int i40e_aq_set_vsi_broadcast(struct i40e_hw *hw,
+			      u16 seid, bool set_filter,
+			      struct i40e_asq_cmd_details *cmd_details)
 {
-	struct i40e_aq_desc desc;
-	struct i40e_aqc_set_vsi_promiscuous_modes *cmd =
-		(struct i40e_aqc_set_vsi_promiscuous_modes *)&desc.params.raw;
-	i40e_status status;
+	struct i40e_aqc_set_vsi_promiscuous_modes *cmd;
+	struct libie_aq_desc desc;
+	int status;
 
 	i40e_fill_default_direct_cmd_desc(&desc,
 					i40e_aqc_opc_set_vsi_promiscuous_modes);
 
+	cmd = libie_aq_raw(&desc);
 	if (set_filter)
 		cmd->promiscuous_flags
 			    |= cpu_to_le16(I40E_AQC_SET_VSI_PROMISC_BROADCAST);
@@ -2230,60 +1777,28 @@ i40e_status i40e_aq_set_vsi_broadcast(struct i40e_hw *hw,
 }
 
 /**
- * i40e_aq_set_vsi_vlan_promisc - control the VLAN promiscuous setting
- * @hw: pointer to the hw struct
- * @seid: vsi number
- * @enable: set MAC L2 layer unicast promiscuous enable/disable for a given VLAN
- * @cmd_details: pointer to command details structure or NULL
- **/
-i40e_status i40e_aq_set_vsi_vlan_promisc(struct i40e_hw *hw,
-				       u16 seid, bool enable,
-				       struct i40e_asq_cmd_details *cmd_details)
-{
-	struct i40e_aq_desc desc;
-	struct i40e_aqc_set_vsi_promiscuous_modes *cmd =
-		(struct i40e_aqc_set_vsi_promiscuous_modes *)&desc.params.raw;
-	i40e_status status;
-	u16 flags = 0;
-
-	i40e_fill_default_direct_cmd_desc(&desc,
-					i40e_aqc_opc_set_vsi_promiscuous_modes);
-	if (enable)
-		flags |= I40E_AQC_SET_VSI_PROMISC_VLAN;
-
-	cmd->promiscuous_flags = cpu_to_le16(flags);
-	cmd->valid_flags = cpu_to_le16(I40E_AQC_SET_VSI_PROMISC_VLAN);
-	cmd->seid = cpu_to_le16(seid);
-
-	status = i40e_asq_send_command(hw, &desc, NULL, 0, cmd_details);
-
-	return status;
-}
-
-/**
- * i40e_get_vsi_params - get VSI configuration info
+ * i40e_aq_get_vsi_params - get VSI configuration info
  * @hw: pointer to the hw struct
  * @vsi_ctx: pointer to a vsi context struct
  * @cmd_details: pointer to command details structure or NULL
  **/
-i40e_status i40e_aq_get_vsi_params(struct i40e_hw *hw,
-				struct i40e_vsi_context *vsi_ctx,
-				struct i40e_asq_cmd_details *cmd_details)
+int i40e_aq_get_vsi_params(struct i40e_hw *hw,
+			   struct i40e_vsi_context *vsi_ctx,
+			   struct i40e_asq_cmd_details *cmd_details)
 {
-	struct i40e_aq_desc desc;
-	struct i40e_aqc_add_get_update_vsi *cmd =
-		(struct i40e_aqc_add_get_update_vsi *)&desc.params.raw;
-	struct i40e_aqc_add_get_update_vsi_completion *resp =
-		(struct i40e_aqc_add_get_update_vsi_completion *)
-		&desc.params.raw;
-	i40e_status status;
+	struct i40e_aqc_add_get_update_vsi_completion *resp;
+	struct i40e_aqc_add_get_update_vsi *cmd;
+	struct libie_aq_desc desc;
+	int status;
 
 	i40e_fill_default_direct_cmd_desc(&desc,
 					  i40e_aqc_opc_get_vsi_parameters);
 
+	resp = libie_aq_raw(&desc);
+	cmd = libie_aq_raw(&desc);
 	cmd->uplink_seid = cpu_to_le16(vsi_ctx->seid);
 
-	desc.flags |= cpu_to_le16((u16)I40E_AQ_FLAG_BUF);
+	desc.flags |= cpu_to_le16((u16)LIBIE_AQ_FLAG_BUF);
 
 	status = i40e_asq_send_command(hw, &desc, &vsi_ctx->info,
 				    sizeof(vsi_ctx->info), NULL);
@@ -2308,26 +1823,26 @@ aq_get_vsi_params_exit:
  *
  * Update a VSI context.
  **/
-i40e_status i40e_aq_update_vsi_params(struct i40e_hw *hw,
-				struct i40e_vsi_context *vsi_ctx,
-				struct i40e_asq_cmd_details *cmd_details)
+int i40e_aq_update_vsi_params(struct i40e_hw *hw,
+			      struct i40e_vsi_context *vsi_ctx,
+			      struct i40e_asq_cmd_details *cmd_details)
 {
-	struct i40e_aq_desc desc;
-	struct i40e_aqc_add_get_update_vsi *cmd =
-		(struct i40e_aqc_add_get_update_vsi *)&desc.params.raw;
-	struct i40e_aqc_add_get_update_vsi_completion *resp =
-		(struct i40e_aqc_add_get_update_vsi_completion *)
-		&desc.params.raw;
-	i40e_status status;
+	struct i40e_aqc_add_get_update_vsi_completion *resp;
+	struct i40e_aqc_add_get_update_vsi *cmd;
+	struct libie_aq_desc desc;
+	int status;
 
 	i40e_fill_default_direct_cmd_desc(&desc,
 					  i40e_aqc_opc_update_vsi_parameters);
+	resp = libie_aq_raw(&desc);
+	cmd = libie_aq_raw(&desc);
 	cmd->uplink_seid = cpu_to_le16(vsi_ctx->seid);
 
-	desc.flags |= cpu_to_le16((u16)(I40E_AQ_FLAG_BUF | I40E_AQ_FLAG_RD));
+	desc.flags |= cpu_to_le16((u16)(LIBIE_AQ_FLAG_BUF | LIBIE_AQ_FLAG_RD));
 
-	status = i40e_asq_send_command(hw, &desc, &vsi_ctx->info,
-				    sizeof(vsi_ctx->info), cmd_details);
+	status = i40e_asq_send_command_atomic(hw, &desc, &vsi_ctx->info,
+					      sizeof(vsi_ctx->info),
+					      cmd_details, true);
 
 	vsi_ctx->vsis_allocated = le16_to_cpu(resp->vsi_used);
 	vsi_ctx->vsis_unallocated = le16_to_cpu(resp->vsi_free);
@@ -2345,21 +1860,21 @@ i40e_status i40e_aq_update_vsi_params(struct i40e_hw *hw,
  *
  * Fill the buf with switch configuration returned from AdminQ command
  **/
-i40e_status i40e_aq_get_switch_config(struct i40e_hw *hw,
-				struct i40e_aqc_get_switch_config_resp *buf,
-				u16 buf_size, u16 *start_seid,
-				struct i40e_asq_cmd_details *cmd_details)
+int i40e_aq_get_switch_config(struct i40e_hw *hw,
+			      struct i40e_aqc_get_switch_config_resp *buf,
+			      u16 buf_size, u16 *start_seid,
+			      struct i40e_asq_cmd_details *cmd_details)
 {
-	struct i40e_aq_desc desc;
-	struct i40e_aqc_switch_seid *scfg =
-		(struct i40e_aqc_switch_seid *)&desc.params.raw;
-	i40e_status status;
+	struct i40e_aqc_switch_seid *scfg;
+	struct libie_aq_desc desc;
+	int status;
 
 	i40e_fill_default_direct_cmd_desc(&desc,
 					  i40e_aqc_opc_get_switch_config);
-	desc.flags |= cpu_to_le16((u16)I40E_AQ_FLAG_BUF);
+	scfg = libie_aq_raw(&desc);
+	desc.flags |= cpu_to_le16((u16)LIBIE_AQ_FLAG_BUF);
 	if (buf_size > I40E_AQ_LARGE_BUF)
-		desc.flags |= cpu_to_le16((u16)I40E_AQ_FLAG_LB);
+		desc.flags |= cpu_to_le16((u16)LIBIE_AQ_FLAG_LB);
 	scfg->seid = cpu_to_le16(*start_seid);
 
 	status = i40e_asq_send_command(hw, &desc, buf, buf_size, cmd_details);
@@ -2372,26 +1887,33 @@ i40e_status i40e_aq_get_switch_config(struct i40e_hw *hw,
  * i40e_aq_set_switch_config
  * @hw: pointer to the hardware structure
  * @flags: bit flag values to set
+ * @mode: cloud filter mode
  * @valid_flags: which bit flags to set
+ * @mode: cloud filter mode
  * @cmd_details: pointer to command details structure or NULL
  *
  * Set switch configuration bits
  **/
-enum i40e_status_code i40e_aq_set_switch_config(struct i40e_hw *hw,
-						u16 flags,
-						u16 valid_flags,
-				struct i40e_asq_cmd_details *cmd_details)
+int i40e_aq_set_switch_config(struct i40e_hw *hw,
+			      u16 flags,
+			      u16 valid_flags, u8 mode,
+			      struct i40e_asq_cmd_details *cmd_details)
 {
-	struct i40e_aq_desc desc;
-	struct i40e_aqc_set_switch_config *scfg =
-		(struct i40e_aqc_set_switch_config *)&desc.params.raw;
-	enum i40e_status_code status;
+	struct i40e_aqc_set_switch_config *scfg;
+	struct libie_aq_desc desc;
+	int status;
 
 	i40e_fill_default_direct_cmd_desc(&desc,
 					  i40e_aqc_opc_set_switch_config);
+	scfg = libie_aq_raw(&desc);
 	scfg->flags = cpu_to_le16(flags);
 	scfg->valid_flags = cpu_to_le16(valid_flags);
-
+	scfg->mode = mode;
+	if (test_bit(I40E_HW_CAP_802_1AD, hw->caps)) {
+		scfg->switch_tag = cpu_to_le16(hw->switch_tag);
+		scfg->first_tag = cpu_to_le16(hw->first_tag);
+		scfg->second_tag = cpu_to_le16(hw->second_tag);
+	}
 	status = i40e_asq_send_command(hw, &desc, NULL, 0, cmd_details);
 
 	return status;
@@ -2409,17 +1931,17 @@ enum i40e_status_code i40e_aq_set_switch_config(struct i40e_hw *hw,
  *
  * Get the firmware version from the admin queue commands
  **/
-i40e_status i40e_aq_get_firmware_version(struct i40e_hw *hw,
-				u16 *fw_major_version, u16 *fw_minor_version,
-				u32 *fw_build,
-				u16 *api_major_version, u16 *api_minor_version,
-				struct i40e_asq_cmd_details *cmd_details)
+int i40e_aq_get_firmware_version(struct i40e_hw *hw,
+				 u16 *fw_major_version, u16 *fw_minor_version,
+				 u32 *fw_build,
+				 u16 *api_major_version, u16 *api_minor_version,
+				 struct i40e_asq_cmd_details *cmd_details)
 {
-	struct i40e_aq_desc desc;
-	struct i40e_aqc_get_version *resp =
-		(struct i40e_aqc_get_version *)&desc.params.raw;
-	i40e_status status;
+	struct i40e_aqc_get_version *resp;
+	struct libie_aq_desc desc;
+	int status;
 
+	resp = libie_aq_raw(&desc);
 	i40e_fill_default_direct_cmd_desc(&desc, i40e_aqc_opc_get_version);
 
 	status = i40e_asq_send_command(hw, &desc, NULL, 0, cmd_details);
@@ -2448,26 +1970,26 @@ i40e_status i40e_aq_get_firmware_version(struct i40e_hw *hw,
  *
  * Send the driver version to the firmware
  **/
-i40e_status i40e_aq_send_driver_version(struct i40e_hw *hw,
+int i40e_aq_send_driver_version(struct i40e_hw *hw,
 				struct i40e_driver_version *dv,
 				struct i40e_asq_cmd_details *cmd_details)
 {
-	struct i40e_aq_desc desc;
-	struct i40e_aqc_driver_version *cmd =
-		(struct i40e_aqc_driver_version *)&desc.params.raw;
-	i40e_status status;
+	struct libie_aqc_driver_ver *cmd;
+	struct libie_aq_desc desc;
+	int status;
 	u16 len;
 
 	if (dv == NULL)
-		return I40E_ERR_PARAM;
+		return -EINVAL;
 
+	cmd = libie_aq_raw(&desc);
 	i40e_fill_default_direct_cmd_desc(&desc, i40e_aqc_opc_driver_version);
 
-	desc.flags |= cpu_to_le16(I40E_AQ_FLAG_BUF | I40E_AQ_FLAG_RD);
-	cmd->driver_major_ver = dv->major_version;
-	cmd->driver_minor_ver = dv->minor_version;
-	cmd->driver_build_ver = dv->build_version;
-	cmd->driver_subbuild_ver = dv->subbuild_version;
+	desc.flags |= cpu_to_le16(LIBIE_AQ_FLAG_BUF | LIBIE_AQ_FLAG_RD);
+	cmd->major_ver = dv->major_version;
+	cmd->minor_ver = dv->minor_version;
+	cmd->build_ver = dv->build_version;
+	cmd->subbuild_ver = dv->subbuild_version;
 
 	len = 0;
 	while (len < sizeof(dv->driver_string) &&
@@ -2490,9 +2012,9 @@ i40e_status i40e_aq_send_driver_version(struct i40e_hw *hw,
  *
  * Side effect: LinkStatusEvent reporting becomes enabled
  **/
-i40e_status i40e_get_link_status(struct i40e_hw *hw, bool *link_up)
+int i40e_get_link_status(struct i40e_hw *hw, bool *link_up)
 {
-	i40e_status status = 0;
+	int status = 0;
 
 	if (hw->phy.get_link_info) {
 		status = i40e_update_link_info(hw);
@@ -2508,13 +2030,13 @@ i40e_status i40e_get_link_status(struct i40e_hw *hw, bool *link_up)
 }
 
 /**
- * i40e_updatelink_status - update status of the HW network link
+ * i40e_update_link_info - update status of the HW network link
  * @hw: pointer to the hw struct
  **/
-i40e_status i40e_update_link_info(struct i40e_hw *hw)
+noinline_for_stack int i40e_update_link_info(struct i40e_hw *hw)
 {
 	struct i40e_aq_get_phy_abilities_resp abilities;
-	i40e_status status = 0;
+	int status = 0;
 
 	status = i40e_aq_get_link_info(hw, true, NULL, NULL);
 	if (status)
@@ -2529,6 +2051,17 @@ i40e_status i40e_update_link_info(struct i40e_hw *hw)
 		if (status)
 			return status;
 
+		if (abilities.fec_cfg_curr_mod_ext_info &
+		    I40E_AQ_ENABLE_FEC_AUTO)
+			hw->phy.link_info.req_fec_info =
+				(I40E_AQ_REQUEST_FEC_KR |
+				 I40E_AQ_REQUEST_FEC_RS);
+		else
+			hw->phy.link_info.req_fec_info =
+				abilities.fec_cfg_curr_mod_ext_info &
+				(I40E_AQ_REQUEST_FEC_KR |
+				 I40E_AQ_REQUEST_FEC_RS);
+
 		memcpy(hw->phy.link_info.module_type, &abilities.module_type,
 		       sizeof(hw->phy.link_info.module_type));
 	}
@@ -2550,24 +2083,24 @@ i40e_status i40e_update_link_info(struct i40e_hw *hw)
  * This asks the FW to add a VEB between the uplink and downlink
  * elements.  If the uplink SEID is 0, this will be a floating VEB.
  **/
-i40e_status i40e_aq_add_veb(struct i40e_hw *hw, u16 uplink_seid,
-				u16 downlink_seid, u8 enabled_tc,
-				bool default_port, u16 *veb_seid,
-				bool enable_stats,
-				struct i40e_asq_cmd_details *cmd_details)
+int i40e_aq_add_veb(struct i40e_hw *hw, u16 uplink_seid,
+		    u16 downlink_seid, u8 enabled_tc,
+		    bool default_port, u16 *veb_seid,
+		    bool enable_stats,
+		    struct i40e_asq_cmd_details *cmd_details)
 {
-	struct i40e_aq_desc desc;
-	struct i40e_aqc_add_veb *cmd =
-		(struct i40e_aqc_add_veb *)&desc.params.raw;
-	struct i40e_aqc_add_veb_completion *resp =
-		(struct i40e_aqc_add_veb_completion *)&desc.params.raw;
-	i40e_status status;
+	struct i40e_aqc_add_veb_completion *resp;
+	struct i40e_aqc_add_veb *cmd;
+	struct libie_aq_desc desc;
 	u16 veb_flags = 0;
+	int status;
 
 	/* SEIDs need to either both be set or both be 0 for floating VEB */
 	if (!!uplink_seid != !!downlink_seid)
-		return I40E_ERR_PARAM;
+		return -EINVAL;
 
+	resp = libie_aq_raw(&desc);
+	cmd = libie_aq_raw(&desc);
 	i40e_fill_default_direct_cmd_desc(&desc, i40e_aqc_opc_add_veb);
 
 	cmd->uplink_seid = cpu_to_le16(uplink_seid);
@@ -2608,21 +2141,20 @@ i40e_status i40e_aq_add_veb(struct i40e_hw *hw, u16 uplink_seid,
  * This retrieves the parameters for a particular VEB, specified by
  * uplink_seid, and returns them to the caller.
  **/
-i40e_status i40e_aq_get_veb_parameters(struct i40e_hw *hw,
-				u16 veb_seid, u16 *switch_id,
-				bool *floating, u16 *statistic_index,
-				u16 *vebs_used, u16 *vebs_free,
-				struct i40e_asq_cmd_details *cmd_details)
+int i40e_aq_get_veb_parameters(struct i40e_hw *hw,
+			       u16 veb_seid, u16 *switch_id,
+			       bool *floating, u16 *statistic_index,
+			       u16 *vebs_used, u16 *vebs_free,
+			       struct i40e_asq_cmd_details *cmd_details)
 {
-	struct i40e_aq_desc desc;
-	struct i40e_aqc_get_veb_parameters_completion *cmd_resp =
-		(struct i40e_aqc_get_veb_parameters_completion *)
-		&desc.params.raw;
-	i40e_status status;
+	struct i40e_aqc_get_veb_parameters_completion *cmd_resp;
+	struct libie_aq_desc desc;
+	int status;
 
 	if (veb_seid == 0)
-		return I40E_ERR_PARAM;
+		return -EINVAL;
 
+	cmd_resp = libie_aq_raw(&desc);
 	i40e_fill_default_direct_cmd_desc(&desc,
 					  i40e_aqc_opc_get_veb_parameters);
 	cmd_resp->seid = cpu_to_le16(veb_seid);
@@ -2653,33 +2185,27 @@ get_veb_exit:
 }
 
 /**
- * i40e_aq_add_macvlan
- * @hw: pointer to the hw struct
- * @seid: VSI for the mac address
+ * i40e_prepare_add_macvlan
  * @mv_list: list of macvlans to be added
+ * @desc: pointer to AQ descriptor structure
  * @count: length of the list
- * @cmd_details: pointer to command details structure or NULL
+ * @seid: VSI for the mac address
  *
- * Add MAC/VLAN addresses to the HW filtering
+ * Internal helper function that prepares the add macvlan request
+ * and returns the buffer size.
  **/
-i40e_status i40e_aq_add_macvlan(struct i40e_hw *hw, u16 seid,
-			struct i40e_aqc_add_macvlan_element_data *mv_list,
-			u16 count, struct i40e_asq_cmd_details *cmd_details)
-{
-	struct i40e_aq_desc desc;
-	struct i40e_aqc_macvlan *cmd =
-		(struct i40e_aqc_macvlan *)&desc.params.raw;
-	i40e_status status;
+static u16
+i40e_prepare_add_macvlan(struct i40e_aqc_add_macvlan_element_data *mv_list,
+			 struct libie_aq_desc *desc, u16 count, u16 seid)
+{
+	struct i40e_aqc_macvlan *cmd = libie_aq_raw(desc);
 	u16 buf_size;
 	int i;
 
-	if (count == 0 || !mv_list || !hw)
-		return I40E_ERR_PARAM;
-
 	buf_size = count * sizeof(*mv_list);
 
 	/* prep the rest of the request */
-	i40e_fill_default_direct_cmd_desc(&desc, i40e_aqc_opc_add_macvlan);
+	i40e_fill_default_direct_cmd_desc(desc, i40e_aqc_opc_add_macvlan);
 	cmd->num_addresses = cpu_to_le16(count);
 	cmd->seid[0] = cpu_to_le16(I40E_AQC_MACVLAN_CMD_SEID_VALID | seid);
 	cmd->seid[1] = 0;
@@ -2690,184 +2216,160 @@ i40e_status i40e_aq_add_macvlan(struct i40e_hw *hw, u16 seid,
 			mv_list[i].flags |=
 			       cpu_to_le16(I40E_AQC_MACVLAN_ADD_USE_SHARED_MAC);
 
-	desc.flags |= cpu_to_le16((u16)(I40E_AQ_FLAG_BUF | I40E_AQ_FLAG_RD));
+	desc->flags |= cpu_to_le16((u16)(LIBIE_AQ_FLAG_BUF | LIBIE_AQ_FLAG_RD));
 	if (buf_size > I40E_AQ_LARGE_BUF)
-		desc.flags |= cpu_to_le16((u16)I40E_AQ_FLAG_LB);
+		desc->flags |= cpu_to_le16((u16)LIBIE_AQ_FLAG_LB);
 
-	status = i40e_asq_send_command(hw, &desc, mv_list, buf_size,
-				       cmd_details);
-
-	return status;
+	return buf_size;
 }
 
 /**
- * i40e_aq_remove_macvlan
+ * i40e_aq_add_macvlan
  * @hw: pointer to the hw struct
  * @seid: VSI for the mac address
- * @mv_list: list of macvlans to be removed
+ * @mv_list: list of macvlans to be added
  * @count: length of the list
  * @cmd_details: pointer to command details structure or NULL
  *
- * Remove MAC/VLAN addresses from the HW filtering
+ * Add MAC/VLAN addresses to the HW filtering
  **/
-i40e_status i40e_aq_remove_macvlan(struct i40e_hw *hw, u16 seid,
-			struct i40e_aqc_remove_macvlan_element_data *mv_list,
-			u16 count, struct i40e_asq_cmd_details *cmd_details)
-{
-	struct i40e_aq_desc desc;
-	struct i40e_aqc_macvlan *cmd =
-		(struct i40e_aqc_macvlan *)&desc.params.raw;
-	i40e_status status;
+int
+i40e_aq_add_macvlan(struct i40e_hw *hw, u16 seid,
+		    struct i40e_aqc_add_macvlan_element_data *mv_list,
+		    u16 count, struct i40e_asq_cmd_details *cmd_details)
+{
+	struct libie_aq_desc desc;
 	u16 buf_size;
 
 	if (count == 0 || !mv_list || !hw)
-		return I40E_ERR_PARAM;
-
-	buf_size = count * sizeof(*mv_list);
-
-	/* prep the rest of the request */
-	i40e_fill_default_direct_cmd_desc(&desc, i40e_aqc_opc_remove_macvlan);
-	cmd->num_addresses = cpu_to_le16(count);
-	cmd->seid[0] = cpu_to_le16(I40E_AQC_MACVLAN_CMD_SEID_VALID | seid);
-	cmd->seid[1] = 0;
-	cmd->seid[2] = 0;
+		return -EINVAL;
 
-	desc.flags |= cpu_to_le16((u16)(I40E_AQ_FLAG_BUF | I40E_AQ_FLAG_RD));
-	if (buf_size > I40E_AQ_LARGE_BUF)
-		desc.flags |= cpu_to_le16((u16)I40E_AQ_FLAG_LB);
-
-	status = i40e_asq_send_command(hw, &desc, mv_list, buf_size,
-				       cmd_details);
+	buf_size = i40e_prepare_add_macvlan(mv_list, &desc, count, seid);
 
-	return status;
+	return i40e_asq_send_command_atomic(hw, &desc, mv_list, buf_size,
+					    cmd_details, true);
 }
 
 /**
- * i40e_mirrorrule_op - Internal helper function to add/delete mirror rule
+ * i40e_aq_add_macvlan_v2
  * @hw: pointer to the hw struct
- * @opcode: AQ opcode for add or delete mirror rule
- * @sw_seid: Switch SEID (to which rule refers)
- * @rule_type: Rule Type (ingress/egress/VLAN)
- * @id: Destination VSI SEID or Rule ID
+ * @seid: VSI for the mac address
+ * @mv_list: list of macvlans to be added
  * @count: length of the list
- * @mr_list: list of mirrored VSI SEIDs or VLAN IDs
  * @cmd_details: pointer to command details structure or NULL
- * @rule_id: Rule ID returned from FW
- * @rule_used: Number of rules used in internal switch
- * @rule_free: Number of rules free in internal switch
+ * @aq_status: pointer to Admin Queue status return value
  *
- * Add/Delete a mirror rule to a specific switch. Mirror rules are supported for
- * VEBs/VEPA elements only
+ * Add MAC/VLAN addresses to the HW filtering.
+ * The _v2 version returns the last Admin Queue status in aq_status
+ * to avoid race conditions in access to hw->aq.asq_last_status.
+ * It also calls _v2 versions of asq_send_command functions to
+ * get the aq_status on the stack.
  **/
-static i40e_status i40e_mirrorrule_op(struct i40e_hw *hw,
-				u16 opcode, u16 sw_seid, u16 rule_type, u16 id,
-				u16 count, __le16 *mr_list,
-				struct i40e_asq_cmd_details *cmd_details,
-				u16 *rule_id, u16 *rules_used, u16 *rules_free)
-{
-	struct i40e_aq_desc desc;
-	struct i40e_aqc_add_delete_mirror_rule *cmd =
-		(struct i40e_aqc_add_delete_mirror_rule *)&desc.params.raw;
-	struct i40e_aqc_add_delete_mirror_rule_completion *resp =
-	(struct i40e_aqc_add_delete_mirror_rule_completion *)&desc.params.raw;
-	i40e_status status;
+int
+i40e_aq_add_macvlan_v2(struct i40e_hw *hw, u16 seid,
+		       struct i40e_aqc_add_macvlan_element_data *mv_list,
+		       u16 count, struct i40e_asq_cmd_details *cmd_details,
+		       enum libie_aq_err *aq_status)
+{
+	struct libie_aq_desc desc;
 	u16 buf_size;
 
-	buf_size = count * sizeof(*mr_list);
+	if (count == 0 || !mv_list || !hw)
+		return -EINVAL;
 
-	/* prep the rest of the request */
-	i40e_fill_default_direct_cmd_desc(&desc, opcode);
-	cmd->seid = cpu_to_le16(sw_seid);
-	cmd->rule_type = cpu_to_le16(rule_type &
-				     I40E_AQC_MIRROR_RULE_TYPE_MASK);
-	cmd->num_entries = cpu_to_le16(count);
-	/* Dest VSI for add, rule_id for delete */
-	cmd->destination = cpu_to_le16(id);
-	if (mr_list) {
-		desc.flags |= cpu_to_le16((u16)(I40E_AQ_FLAG_BUF |
-						I40E_AQ_FLAG_RD));
-		if (buf_size > I40E_AQ_LARGE_BUF)
-			desc.flags |= cpu_to_le16((u16)I40E_AQ_FLAG_LB);
-	}
+	buf_size = i40e_prepare_add_macvlan(mv_list, &desc, count, seid);
 
-	status = i40e_asq_send_command(hw, &desc, mr_list, buf_size,
-				       cmd_details);
-	if (!status ||
-	    hw->aq.asq_last_status == I40E_AQ_RC_ENOSPC) {
-		if (rule_id)
-			*rule_id = le16_to_cpu(resp->rule_id);
-		if (rules_used)
-			*rules_used = le16_to_cpu(resp->mirror_rules_used);
-		if (rules_free)
-			*rules_free = le16_to_cpu(resp->mirror_rules_free);
-	}
-	return status;
+	return i40e_asq_send_command_atomic_v2(hw, &desc, mv_list, buf_size,
+					       cmd_details, true, aq_status);
 }
 
 /**
- * i40e_aq_add_mirrorrule - add a mirror rule
+ * i40e_aq_remove_macvlan
  * @hw: pointer to the hw struct
- * @sw_seid: Switch SEID (to which rule refers)
- * @rule_type: Rule Type (ingress/egress/VLAN)
- * @dest_vsi: SEID of VSI to which packets will be mirrored
+ * @seid: VSI for the mac address
+ * @mv_list: list of macvlans to be removed
  * @count: length of the list
- * @mr_list: list of mirrored VSI SEIDs or VLAN IDs
  * @cmd_details: pointer to command details structure or NULL
- * @rule_id: Rule ID returned from FW
- * @rule_used: Number of rules used in internal switch
- * @rule_free: Number of rules free in internal switch
  *
- * Add mirror rule. Mirror rules are supported for VEBs or VEPA elements only
+ * Remove MAC/VLAN addresses from the HW filtering
  **/
-i40e_status i40e_aq_add_mirrorrule(struct i40e_hw *hw, u16 sw_seid,
-			u16 rule_type, u16 dest_vsi, u16 count, __le16 *mr_list,
-			struct i40e_asq_cmd_details *cmd_details,
-			u16 *rule_id, u16 *rules_used, u16 *rules_free)
-{
-	if (!(rule_type == I40E_AQC_MIRROR_RULE_TYPE_ALL_INGRESS ||
-	    rule_type == I40E_AQC_MIRROR_RULE_TYPE_ALL_EGRESS)) {
-		if (count == 0 || !mr_list)
-			return I40E_ERR_PARAM;
-	}
+int
+i40e_aq_remove_macvlan(struct i40e_hw *hw, u16 seid,
+		       struct i40e_aqc_remove_macvlan_element_data *mv_list,
+		       u16 count, struct i40e_asq_cmd_details *cmd_details)
+{
+	struct i40e_aqc_macvlan *cmd;
+	struct libie_aq_desc desc;
+	u16 buf_size;
+	int status;
+
+	if (count == 0 || !mv_list || !hw)
+		return -EINVAL;
+
+	buf_size = count * sizeof(*mv_list);
+
+	/* prep the rest of the request */
+	i40e_fill_default_direct_cmd_desc(&desc, i40e_aqc_opc_remove_macvlan);
+	cmd = libie_aq_raw(&desc);
+	cmd->num_addresses = cpu_to_le16(count);
+	cmd->seid[0] = cpu_to_le16(I40E_AQC_MACVLAN_CMD_SEID_VALID | seid);
+	cmd->seid[1] = 0;
+	cmd->seid[2] = 0;
+
+	desc.flags |= cpu_to_le16((u16)(LIBIE_AQ_FLAG_BUF | LIBIE_AQ_FLAG_RD));
+	if (buf_size > I40E_AQ_LARGE_BUF)
+		desc.flags |= cpu_to_le16((u16)LIBIE_AQ_FLAG_LB);
 
-	return i40e_mirrorrule_op(hw, i40e_aqc_opc_add_mirror_rule, sw_seid,
-				  rule_type, dest_vsi, count, mr_list,
-				  cmd_details, rule_id, rules_used, rules_free);
+	status = i40e_asq_send_command_atomic(hw, &desc, mv_list, buf_size,
+					      cmd_details, true);
+
+	return status;
 }
 
 /**
- * i40e_aq_delete_mirrorrule - delete a mirror rule
+ * i40e_aq_remove_macvlan_v2
  * @hw: pointer to the hw struct
- * @sw_seid: Switch SEID (to which rule refers)
- * @rule_type: Rule Type (ingress/egress/VLAN)
+ * @seid: VSI for the mac address
+ * @mv_list: list of macvlans to be removed
  * @count: length of the list
- * @rule_id: Rule ID that is returned in the receive desc as part of
- *		add_mirrorrule.
- * @mr_list: list of mirrored VLAN IDs to be removed
  * @cmd_details: pointer to command details structure or NULL
- * @rule_used: Number of rules used in internal switch
- * @rule_free: Number of rules free in internal switch
+ * @aq_status: pointer to Admin Queue status return value
  *
- * Delete a mirror rule. Mirror rules are supported for VEBs/VEPA elements only
+ * Remove MAC/VLAN addresses from the HW filtering.
+ * The _v2 version returns the last Admin Queue status in aq_status
+ * to avoid race conditions in access to hw->aq.asq_last_status.
+ * It also calls _v2 versions of asq_send_command functions to
+ * get the aq_status on the stack.
  **/
-i40e_status i40e_aq_delete_mirrorrule(struct i40e_hw *hw, u16 sw_seid,
-			u16 rule_type, u16 rule_id, u16 count, __le16 *mr_list,
-			struct i40e_asq_cmd_details *cmd_details,
-			u16 *rules_used, u16 *rules_free)
-{
-	/* Rule ID has to be valid except rule_type: INGRESS VLAN mirroring */
-	if (rule_type == I40E_AQC_MIRROR_RULE_TYPE_VLAN) {
-		/* count and mr_list shall be valid for rule_type INGRESS VLAN
-		 * mirroring. For other rule_type, count and rule_type should
-		 * not matter.
-		 */
-		if (count == 0 || !mr_list)
-			return I40E_ERR_PARAM;
-	}
+int
+i40e_aq_remove_macvlan_v2(struct i40e_hw *hw, u16 seid,
+			  struct i40e_aqc_remove_macvlan_element_data *mv_list,
+			  u16 count, struct i40e_asq_cmd_details *cmd_details,
+			  enum libie_aq_err *aq_status)
+{
+	struct i40e_aqc_macvlan *cmd;
+	struct libie_aq_desc desc;
+	u16 buf_size;
 
-	return i40e_mirrorrule_op(hw, i40e_aqc_opc_delete_mirror_rule, sw_seid,
-				  rule_type, rule_id, count, mr_list,
-				  cmd_details, NULL, rules_used, rules_free);
+	if (count == 0 || !mv_list || !hw)
+		return -EINVAL;
+
+	buf_size = count * sizeof(*mv_list);
+
+	/* prep the rest of the request */
+	i40e_fill_default_direct_cmd_desc(&desc, i40e_aqc_opc_remove_macvlan);
+	cmd = libie_aq_raw(&desc);
+	cmd->num_addresses = cpu_to_le16(count);
+	cmd->seid[0] = cpu_to_le16(I40E_AQC_MACVLAN_CMD_SEID_VALID | seid);
+	cmd->seid[1] = 0;
+	cmd->seid[2] = 0;
+
+	desc.flags |= cpu_to_le16((u16)(LIBIE_AQ_FLAG_BUF | LIBIE_AQ_FLAG_RD));
+	if (buf_size > I40E_AQ_LARGE_BUF)
+		desc.flags |= cpu_to_le16((u16)LIBIE_AQ_FLAG_LB);
+
+	return i40e_asq_send_command_atomic_v2(hw, &desc, mv_list, buf_size,
+						 cmd_details, true, aq_status);
 }
 
 /**
@@ -2882,25 +2384,25 @@ i40e_status i40e_aq_delete_mirrorrule(struct i40e_hw *hw, u16 sw_seid,
  *
  * send msg to vf
  **/
-i40e_status i40e_aq_send_msg_to_vf(struct i40e_hw *hw, u16 vfid,
-				u32 v_opcode, u32 v_retval, u8 *msg, u16 msglen,
-				struct i40e_asq_cmd_details *cmd_details)
+int i40e_aq_send_msg_to_vf(struct i40e_hw *hw, u16 vfid,
+			   u32 v_opcode, u32 v_retval, u8 *msg, u16 msglen,
+			   struct i40e_asq_cmd_details *cmd_details)
 {
-	struct i40e_aq_desc desc;
-	struct i40e_aqc_pf_vf_message *cmd =
-		(struct i40e_aqc_pf_vf_message *)&desc.params.raw;
-	i40e_status status;
+	struct i40e_aqc_pf_vf_message *cmd;
+	struct libie_aq_desc desc;
+	int status;
 
 	i40e_fill_default_direct_cmd_desc(&desc, i40e_aqc_opc_send_msg_to_vf);
+	cmd = libie_aq_raw(&desc);
 	cmd->id = cpu_to_le32(vfid);
 	desc.cookie_high = cpu_to_le32(v_opcode);
 	desc.cookie_low = cpu_to_le32(v_retval);
-	desc.flags |= cpu_to_le16((u16)I40E_AQ_FLAG_SI);
+	desc.flags |= cpu_to_le16((u16)LIBIE_AQ_FLAG_SI);
 	if (msglen) {
-		desc.flags |= cpu_to_le16((u16)(I40E_AQ_FLAG_BUF |
-						I40E_AQ_FLAG_RD));
+		desc.flags |= cpu_to_le16((u16)(LIBIE_AQ_FLAG_BUF |
+						LIBIE_AQ_FLAG_RD));
 		if (msglen > I40E_AQ_LARGE_BUF)
-			desc.flags |= cpu_to_le16((u16)I40E_AQ_FLAG_LB);
+			desc.flags |= cpu_to_le16((u16)LIBIE_AQ_FLAG_LB);
 		desc.datalen = cpu_to_le16(msglen);
 	}
 	status = i40e_asq_send_command(hw, &desc, msg, msglen, cmd_details);
@@ -2917,20 +2419,20 @@ i40e_status i40e_aq_send_msg_to_vf(struct i40e_hw *hw, u16 vfid,
  *
  * Read the register using the admin queue commands
  **/
-i40e_status i40e_aq_debug_read_register(struct i40e_hw *hw,
+int i40e_aq_debug_read_register(struct i40e_hw *hw,
 				u32 reg_addr, u64 *reg_val,
 				struct i40e_asq_cmd_details *cmd_details)
 {
-	struct i40e_aq_desc desc;
-	struct i40e_aqc_debug_reg_read_write *cmd_resp =
-		(struct i40e_aqc_debug_reg_read_write *)&desc.params.raw;
-	i40e_status status;
+	struct i40e_aqc_debug_reg_read_write *cmd_resp;
+	struct libie_aq_desc desc;
+	int status;
 
 	if (reg_val == NULL)
-		return I40E_ERR_PARAM;
+		return -EINVAL;
 
 	i40e_fill_default_direct_cmd_desc(&desc, i40e_aqc_opc_debug_read_reg);
 
+	cmd_resp = libie_aq_raw(&desc);
 	cmd_resp->address = cpu_to_le32(reg_addr);
 
 	status = i40e_asq_send_command(hw, &desc, NULL, 0, cmd_details);
@@ -2952,17 +2454,17 @@ i40e_status i40e_aq_debug_read_register(struct i40e_hw *hw,
  *
  * Write to a register using the admin queue commands
  **/
-i40e_status i40e_aq_debug_write_register(struct i40e_hw *hw,
-					u32 reg_addr, u64 reg_val,
-					struct i40e_asq_cmd_details *cmd_details)
+int i40e_aq_debug_write_register(struct i40e_hw *hw,
+				 u32 reg_addr, u64 reg_val,
+				 struct i40e_asq_cmd_details *cmd_details)
 {
-	struct i40e_aq_desc desc;
-	struct i40e_aqc_debug_reg_read_write *cmd =
-		(struct i40e_aqc_debug_reg_read_write *)&desc.params.raw;
-	i40e_status status;
+	struct i40e_aqc_debug_reg_read_write *cmd;
+	struct libie_aq_desc desc;
+	int status;
 
 	i40e_fill_default_direct_cmd_desc(&desc, i40e_aqc_opc_debug_write_reg);
 
+	cmd = libie_aq_raw(&desc);
 	cmd->address = cpu_to_le32(reg_addr);
 	cmd->value_high = cpu_to_le32((u32)(reg_val >> 32));
 	cmd->value_low = cpu_to_le32((u32)(reg_val & 0xFFFFFFFF));
@@ -2983,22 +2485,22 @@ i40e_status i40e_aq_debug_write_register(struct i40e_hw *hw,
  *
  * requests common resource using the admin queue commands
  **/
-i40e_status i40e_aq_request_resource(struct i40e_hw *hw,
-				enum i40e_aq_resources_ids resource,
-				enum i40e_aq_resource_access_type access,
-				u8 sdp_number, u64 *timeout,
-				struct i40e_asq_cmd_details *cmd_details)
+int i40e_aq_request_resource(struct i40e_hw *hw,
+			     enum i40e_aq_resources_ids resource,
+			     enum i40e_aq_resource_access_type access,
+			     u8 sdp_number, u64 *timeout,
+			     struct i40e_asq_cmd_details *cmd_details)
 {
-	struct i40e_aq_desc desc;
-	struct i40e_aqc_request_resource *cmd_resp =
-		(struct i40e_aqc_request_resource *)&desc.params.raw;
-	i40e_status status;
+	struct libie_aqc_req_res *cmd_resp;
+	struct libie_aq_desc desc;
+	int status;
 
 	i40e_fill_default_direct_cmd_desc(&desc, i40e_aqc_opc_request_resource);
 
-	cmd_resp->resource_id = cpu_to_le16(resource);
+	cmd_resp = libie_aq_raw(&desc);
+	cmd_resp->res_id = cpu_to_le16(resource);
 	cmd_resp->access_type = cpu_to_le16(access);
-	cmd_resp->resource_number = cpu_to_le32(sdp_number);
+	cmd_resp->res_number = cpu_to_le32(sdp_number);
 
 	status = i40e_asq_send_command(hw, &desc, NULL, 0, cmd_details);
 	/* The completion specifies the maximum time in ms that the driver
@@ -3007,7 +2509,7 @@ i40e_status i40e_aq_request_resource(struct i40e_hw *hw,
 	 * busy return value and the timeout field indicates the maximum time
 	 * the current owner of the resource has to free it.
 	 */
-	if (!status || hw->aq.asq_last_status == I40E_AQ_RC_EBUSY)
+	if (!status || hw->aq.asq_last_status == LIBIE_AQ_RC_EBUSY)
 		*timeout = le32_to_cpu(cmd_resp->timeout);
 
 	return status;
@@ -3022,20 +2524,20 @@ i40e_status i40e_aq_request_resource(struct i40e_hw *hw,
  *
  * release common resource using the admin queue commands
  **/
-i40e_status i40e_aq_release_resource(struct i40e_hw *hw,
-				enum i40e_aq_resources_ids resource,
-				u8 sdp_number,
-				struct i40e_asq_cmd_details *cmd_details)
+int i40e_aq_release_resource(struct i40e_hw *hw,
+			     enum i40e_aq_resources_ids resource,
+			     u8 sdp_number,
+			     struct i40e_asq_cmd_details *cmd_details)
 {
-	struct i40e_aq_desc desc;
-	struct i40e_aqc_request_resource *cmd =
-		(struct i40e_aqc_request_resource *)&desc.params.raw;
-	i40e_status status;
+	struct libie_aqc_req_res *cmd;
+	struct libie_aq_desc desc;
+	int status;
 
 	i40e_fill_default_direct_cmd_desc(&desc, i40e_aqc_opc_release_resource);
 
-	cmd->resource_id = cpu_to_le16(resource);
-	cmd->resource_number = cpu_to_le32(sdp_number);
+	cmd = libie_aq_raw(&desc);
+	cmd->res_id = cpu_to_le16(resource);
+	cmd->res_number = cpu_to_le32(sdp_number);
 
 	status = i40e_asq_send_command(hw, &desc, NULL, 0, cmd_details);
 
@@ -3054,24 +2556,24 @@ i40e_status i40e_aq_release_resource(struct i40e_hw *hw,
  *
  * Read the NVM using the admin queue commands
  **/
-i40e_status i40e_aq_read_nvm(struct i40e_hw *hw, u8 module_pointer,
-				u32 offset, u16 length, void *data,
-				bool last_command,
-				struct i40e_asq_cmd_details *cmd_details)
+int i40e_aq_read_nvm(struct i40e_hw *hw, u8 module_pointer,
+		     u32 offset, u16 length, void *data,
+		     bool last_command,
+		     struct i40e_asq_cmd_details *cmd_details)
 {
-	struct i40e_aq_desc desc;
-	struct i40e_aqc_nvm_update *cmd =
-		(struct i40e_aqc_nvm_update *)&desc.params.raw;
-	i40e_status status;
+	struct i40e_aqc_nvm_update *cmd;
+	struct libie_aq_desc desc;
+	int status;
 
 	/* In offset the highest byte must be zeroed. */
 	if (offset & 0xFF000000) {
-		status = I40E_ERR_PARAM;
+		status = -EINVAL;
 		goto i40e_aq_read_nvm_exit;
 	}
 
 	i40e_fill_default_direct_cmd_desc(&desc, i40e_aqc_opc_nvm_read);
 
+	cmd = libie_aq_raw(&desc);
 	/* If this is the last command in a series, set the proper flag. */
 	if (last_command)
 		cmd->command_flags |= I40E_AQ_NVM_LAST_CMD;
@@ -3079,9 +2581,9 @@ i40e_status i40e_aq_read_nvm(struct i40e_hw *hw, u8 module_pointer,
 	cmd->offset = cpu_to_le32(offset);
 	cmd->length = cpu_to_le16(length);
 
-	desc.flags |= cpu_to_le16((u16)I40E_AQ_FLAG_BUF);
+	desc.flags |= cpu_to_le16((u16)LIBIE_AQ_FLAG_BUF);
 	if (length > I40E_AQ_LARGE_BUF)
-		desc.flags |= cpu_to_le16((u16)I40E_AQ_FLAG_LB);
+		desc.flags |= cpu_to_le16((u16)LIBIE_AQ_FLAG_LB);
 
 	status = i40e_asq_send_command(hw, &desc, data, length, cmd_details);
 
@@ -3100,23 +2602,23 @@ i40e_aq_read_nvm_exit:
  *
  * Erase the NVM sector using the admin queue commands
  **/
-i40e_status i40e_aq_erase_nvm(struct i40e_hw *hw, u8 module_pointer,
-			      u32 offset, u16 length, bool last_command,
-			      struct i40e_asq_cmd_details *cmd_details)
+int i40e_aq_erase_nvm(struct i40e_hw *hw, u8 module_pointer,
+		      u32 offset, u16 length, bool last_command,
+		      struct i40e_asq_cmd_details *cmd_details)
 {
-	struct i40e_aq_desc desc;
-	struct i40e_aqc_nvm_update *cmd =
-		(struct i40e_aqc_nvm_update *)&desc.params.raw;
-	i40e_status status;
+	struct i40e_aqc_nvm_update *cmd;
+	struct libie_aq_desc desc;
+	int status;
 
 	/* In offset the highest byte must be zeroed. */
 	if (offset & 0xFF000000) {
-		status = I40E_ERR_PARAM;
+		status = -EINVAL;
 		goto i40e_aq_erase_nvm_exit;
 	}
 
 	i40e_fill_default_direct_cmd_desc(&desc, i40e_aqc_opc_nvm_erase);
 
+	cmd = libie_aq_raw(&desc);
 	/* If this is the last command in a series, set the proper flag. */
 	if (last_command)
 		cmd->command_flags |= I40E_AQ_NVM_LAST_CMD;
@@ -3143,15 +2645,16 @@ static void i40e_parse_discover_capabilities(struct i40e_hw *hw, void *buff,
 				     u32 cap_count,
 				     enum i40e_admin_queue_opc list_type_opc)
 {
-	struct i40e_aqc_list_capabilities_element_resp *cap;
+	struct libie_aqc_list_caps_elem *cap;
 	u32 valid_functions, num_functions;
 	u32 number, logical_id, phys_id;
 	struct i40e_hw_capabilities *p;
+	u16 id, ocp_cfg_word0;
 	u8 major_rev;
+	int status;
 	u32 i = 0;
-	u16 id;
 
-	cap = (struct i40e_aqc_list_capabilities_element_resp *) buff;
+	cap = (struct libie_aqc_list_caps_elem *)buff;
 
 	if (list_type_opc == i40e_aqc_opc_list_dev_capabilities)
 		p = &hw->dev_caps;
@@ -3161,17 +2664,17 @@ static void i40e_parse_discover_capabilities(struct i40e_hw *hw, void *buff,
 		return;
 
 	for (i = 0; i < cap_count; i++, cap++) {
-		id = le16_to_cpu(cap->id);
+		id = le16_to_cpu(cap->cap);
 		number = le32_to_cpu(cap->number);
 		logical_id = le32_to_cpu(cap->logical_id);
 		phys_id = le32_to_cpu(cap->phys_id);
-		major_rev = cap->major_rev;
+		major_rev = cap->major_ver;
 
 		switch (id) {
-		case I40E_AQ_CAP_ID_SWITCH_MODE:
+		case LIBIE_AQC_CAPS_SWITCH_MODE:
 			p->switch_mode = number;
 			break;
-		case I40E_AQ_CAP_ID_MNG_MODE:
+		case LIBIE_AQC_CAPS_MNG_MODE:
 			p->management_mode = number;
 			if (major_rev > 1) {
 				p->mng_protocols_over_mctp = logical_id;
@@ -3182,76 +2685,76 @@ static void i40e_parse_discover_capabilities(struct i40e_hw *hw, void *buff,
 				p->mng_protocols_over_mctp = 0;
 			}
 			break;
-		case I40E_AQ_CAP_ID_NPAR_ACTIVE:
+		case LIBIE_AQC_CAPS_NPAR_ACTIVE:
 			p->npar_enable = number;
 			break;
-		case I40E_AQ_CAP_ID_OS2BMC_CAP:
+		case LIBIE_AQC_CAPS_OS2BMC_CAP:
 			p->os2bmc = number;
 			break;
-		case I40E_AQ_CAP_ID_FUNCTIONS_VALID:
+		case LIBIE_AQC_CAPS_VALID_FUNCTIONS:
 			p->valid_functions = number;
 			break;
-		case I40E_AQ_CAP_ID_SRIOV:
+		case LIBIE_AQC_CAPS_SRIOV:
 			if (number == 1)
 				p->sr_iov_1_1 = true;
 			break;
-		case I40E_AQ_CAP_ID_VF:
+		case LIBIE_AQC_CAPS_VF:
 			p->num_vfs = number;
 			p->vf_base_id = logical_id;
 			break;
-		case I40E_AQ_CAP_ID_VMDQ:
+		case LIBIE_AQC_CAPS_VMDQ:
 			if (number == 1)
 				p->vmdq = true;
 			break;
-		case I40E_AQ_CAP_ID_8021QBG:
+		case LIBIE_AQC_CAPS_8021QBG:
 			if (number == 1)
 				p->evb_802_1_qbg = true;
 			break;
-		case I40E_AQ_CAP_ID_8021QBR:
+		case LIBIE_AQC_CAPS_8021QBR:
 			if (number == 1)
 				p->evb_802_1_qbh = true;
 			break;
-		case I40E_AQ_CAP_ID_VSI:
+		case LIBIE_AQC_CAPS_VSI:
 			p->num_vsis = number;
 			break;
-		case I40E_AQ_CAP_ID_DCB:
+		case LIBIE_AQC_CAPS_DCB:
 			if (number == 1) {
 				p->dcb = true;
 				p->enabled_tcmap = logical_id;
 				p->maxtc = phys_id;
 			}
 			break;
-		case I40E_AQ_CAP_ID_FCOE:
+		case LIBIE_AQC_CAPS_FCOE:
 			if (number == 1)
 				p->fcoe = true;
 			break;
-		case I40E_AQ_CAP_ID_ISCSI:
+		case LIBIE_AQC_CAPS_ISCSI:
 			if (number == 1)
 				p->iscsi = true;
 			break;
-		case I40E_AQ_CAP_ID_RSS:
+		case LIBIE_AQC_CAPS_RSS:
 			p->rss = true;
 			p->rss_table_size = number;
 			p->rss_table_entry_width = logical_id;
 			break;
-		case I40E_AQ_CAP_ID_RXQ:
+		case LIBIE_AQC_CAPS_RXQS:
 			p->num_rx_qp = number;
 			p->base_queue = phys_id;
 			break;
-		case I40E_AQ_CAP_ID_TXQ:
+		case LIBIE_AQC_CAPS_TXQS:
 			p->num_tx_qp = number;
 			p->base_queue = phys_id;
 			break;
-		case I40E_AQ_CAP_ID_MSIX:
+		case LIBIE_AQC_CAPS_MSIX:
 			p->num_msix_vectors = number;
 			i40e_debug(hw, I40E_DEBUG_INIT,
 				   "HW Capability: MSIX vector count = %d\n",
 				   p->num_msix_vectors);
 			break;
-		case I40E_AQ_CAP_ID_VF_MSIX:
+		case LIBIE_AQC_CAPS_VF_MSIX:
 			p->num_msix_vectors_vf = number;
 			break;
-		case I40E_AQ_CAP_ID_FLEX10:
+		case LIBIE_AQC_CAPS_FLEX10:
 			if (major_rev == 1) {
 				if (number == 1) {
 					p->flex10_enable = true;
@@ -3267,42 +2770,42 @@ static void i40e_parse_discover_capabilities(struct i40e_hw *hw, void *buff,
 			p->flex10_mode = logical_id;
 			p->flex10_status = phys_id;
 			break;
-		case I40E_AQ_CAP_ID_CEM:
+		case LIBIE_AQC_CAPS_CEM:
 			if (number == 1)
 				p->mgmt_cem = true;
 			break;
-		case I40E_AQ_CAP_ID_IWARP:
+		case LIBIE_AQC_CAPS_RDMA:
 			if (number == 1)
 				p->iwarp = true;
 			break;
-		case I40E_AQ_CAP_ID_LED:
+		case LIBIE_AQC_CAPS_LED:
 			if (phys_id < I40E_HW_CAP_MAX_GPIO)
 				p->led[phys_id] = true;
 			break;
-		case I40E_AQ_CAP_ID_SDP:
+		case LIBIE_AQC_CAPS_SDP:
 			if (phys_id < I40E_HW_CAP_MAX_GPIO)
 				p->sdp[phys_id] = true;
 			break;
-		case I40E_AQ_CAP_ID_MDIO:
+		case LIBIE_AQC_CAPS_MDIO:
 			if (number == 1) {
 				p->mdio_port_num = phys_id;
 				p->mdio_port_mode = logical_id;
 			}
 			break;
-		case I40E_AQ_CAP_ID_1588:
+		case LIBIE_AQC_CAPS_1588:
 			if (number == 1)
 				p->ieee_1588 = true;
 			break;
-		case I40E_AQ_CAP_ID_FLOW_DIRECTOR:
+		case LIBIE_AQC_CAPS_FD:
 			p->fd = true;
 			p->fd_filters_guaranteed = number;
 			p->fd_filters_best_effort = logical_id;
 			break;
-		case I40E_AQ_CAP_ID_WSR_PROT:
+		case LIBIE_AQC_CAPS_WSR_PROT:
 			p->wr_csr_prot = (u64)number;
 			p->wr_csr_prot |= (u64)logical_id << 32;
 			break;
-		case I40E_AQ_CAP_ID_NVM_MGMT:
+		case LIBIE_AQC_CAPS_NVM_MGMT:
 			if (number & I40E_NVM_MGMT_SEC_REV_DISABLED)
 				p->sec_rev_disabled = true;
 			if (number & I40E_NVM_MGMT_UPDATE_DISABLED)
@@ -3336,6 +2839,26 @@ static void i40e_parse_discover_capabilities(struct i40e_hw *hw, void *buff,
 			hw->num_ports++;
 	}
 
+	/* OCP cards case: if a mezz is removed the Ethernet port is at
+	 * disabled state in PRTGEN_CNF register. Additional NVM read is
+	 * needed in order to check if we are dealing with OCP card.
+	 * Those cards have 4 PFs at minimum, so using PRTGEN_CNF for counting
+	 * physical ports results in wrong partition id calculation and thus
+	 * not supporting WoL.
+	 */
+	if (hw->mac.type == I40E_MAC_X722) {
+		if (!i40e_acquire_nvm(hw, I40E_RESOURCE_READ)) {
+			status = i40e_aq_read_nvm(hw, I40E_SR_EMP_MODULE_PTR,
+						  2 * I40E_SR_OCP_CFG_WORD0,
+						  sizeof(ocp_cfg_word0),
+						  &ocp_cfg_word0, true, NULL);
+			if (!status &&
+			    (ocp_cfg_word0 & I40E_SR_OCP_ENABLED))
+				hw->num_ports = 4;
+			i40e_release_nvm(hw);
+		}
+	}
+
 	valid_functions = p->valid_functions;
 	num_functions = 0;
 	while (valid_functions) {
@@ -3369,28 +2892,28 @@ static void i40e_parse_discover_capabilities(struct i40e_hw *hw, void *buff,
  *
  * Get the device capabilities descriptions from the firmware
  **/
-i40e_status i40e_aq_discover_capabilities(struct i40e_hw *hw,
-				void *buff, u16 buff_size, u16 *data_size,
-				enum i40e_admin_queue_opc list_type_opc,
-				struct i40e_asq_cmd_details *cmd_details)
+int i40e_aq_discover_capabilities(struct i40e_hw *hw,
+				  void *buff, u16 buff_size, u16 *data_size,
+				  enum i40e_admin_queue_opc list_type_opc,
+				  struct i40e_asq_cmd_details *cmd_details)
 {
-	struct i40e_aqc_list_capabilites *cmd;
-	struct i40e_aq_desc desc;
-	i40e_status status = 0;
+	struct libie_aqc_list_caps *cmd;
+	struct libie_aq_desc desc;
+	int status = 0;
 
-	cmd = (struct i40e_aqc_list_capabilites *)&desc.params.raw;
+	cmd = libie_aq_raw(&desc);
 
 	if (list_type_opc != i40e_aqc_opc_list_func_capabilities &&
 		list_type_opc != i40e_aqc_opc_list_dev_capabilities) {
-		status = I40E_ERR_PARAM;
+		status = -EINVAL;
 		goto exit;
 	}
 
 	i40e_fill_default_direct_cmd_desc(&desc, list_type_opc);
 
-	desc.flags |= cpu_to_le16((u16)I40E_AQ_FLAG_BUF);
+	desc.flags |= cpu_to_le16((u16)LIBIE_AQ_FLAG_BUF);
 	if (buff_size > I40E_AQ_LARGE_BUF)
-		desc.flags |= cpu_to_le16((u16)I40E_AQ_FLAG_LB);
+		desc.flags |= cpu_to_le16((u16)LIBIE_AQ_FLAG_LB);
 
 	status = i40e_asq_send_command(hw, &desc, buff, buff_size, cmd_details);
 	*data_size = le16_to_cpu(desc.datalen);
@@ -3413,38 +2936,49 @@ exit:
  * @length: length of the section to be written (in bytes from the offset)
  * @data: command buffer (size [bytes] = length)
  * @last_command: tells if this is the last command in a series
+ * @preservation_flags: Preservation mode flags
  * @cmd_details: pointer to command details structure or NULL
  *
  * Update the NVM using the admin queue commands
  **/
-i40e_status i40e_aq_update_nvm(struct i40e_hw *hw, u8 module_pointer,
-			       u32 offset, u16 length, void *data,
-			       bool last_command,
-			       struct i40e_asq_cmd_details *cmd_details)
+int i40e_aq_update_nvm(struct i40e_hw *hw, u8 module_pointer,
+		       u32 offset, u16 length, void *data,
+		       bool last_command, u8 preservation_flags,
+		       struct i40e_asq_cmd_details *cmd_details)
 {
-	struct i40e_aq_desc desc;
-	struct i40e_aqc_nvm_update *cmd =
-		(struct i40e_aqc_nvm_update *)&desc.params.raw;
-	i40e_status status;
+	struct i40e_aqc_nvm_update *cmd;
+	struct libie_aq_desc desc;
+	int status;
 
 	/* In offset the highest byte must be zeroed. */
 	if (offset & 0xFF000000) {
-		status = I40E_ERR_PARAM;
+		status = -EINVAL;
 		goto i40e_aq_update_nvm_exit;
 	}
 
 	i40e_fill_default_direct_cmd_desc(&desc, i40e_aqc_opc_nvm_update);
 
+	cmd = libie_aq_raw(&desc);
 	/* If this is the last command in a series, set the proper flag. */
 	if (last_command)
 		cmd->command_flags |= I40E_AQ_NVM_LAST_CMD;
+	if (hw->mac.type == I40E_MAC_X722) {
+		if (preservation_flags == I40E_NVM_PRESERVATION_FLAGS_SELECTED)
+			cmd->command_flags |=
+				(I40E_AQ_NVM_PRESERVATION_FLAGS_SELECTED <<
+				 I40E_AQ_NVM_PRESERVATION_FLAGS_SHIFT);
+		else if (preservation_flags == I40E_NVM_PRESERVATION_FLAGS_ALL)
+			cmd->command_flags |=
+				(I40E_AQ_NVM_PRESERVATION_FLAGS_ALL <<
+				 I40E_AQ_NVM_PRESERVATION_FLAGS_SHIFT);
+	}
 	cmd->module_pointer = module_pointer;
 	cmd->offset = cpu_to_le32(offset);
 	cmd->length = cpu_to_le16(length);
 
-	desc.flags |= cpu_to_le16((u16)(I40E_AQ_FLAG_BUF | I40E_AQ_FLAG_RD));
+	desc.flags |= cpu_to_le16((u16)(LIBIE_AQ_FLAG_BUF | LIBIE_AQ_FLAG_RD));
 	if (length > I40E_AQ_LARGE_BUF)
-		desc.flags |= cpu_to_le16((u16)I40E_AQ_FLAG_LB);
+		desc.flags |= cpu_to_le16((u16)LIBIE_AQ_FLAG_LB);
 
 	status = i40e_asq_send_command(hw, &desc, data, length, cmd_details);
 
@@ -3465,34 +2999,33 @@ i40e_aq_update_nvm_exit:
  *
  * Requests the complete LLDP MIB (entire packet).
  **/
-i40e_status i40e_aq_get_lldp_mib(struct i40e_hw *hw, u8 bridge_type,
-				u8 mib_type, void *buff, u16 buff_size,
-				u16 *local_len, u16 *remote_len,
-				struct i40e_asq_cmd_details *cmd_details)
+int i40e_aq_get_lldp_mib(struct i40e_hw *hw, u8 bridge_type,
+			 u8 mib_type, void *buff, u16 buff_size,
+			 u16 *local_len, u16 *remote_len,
+			 struct i40e_asq_cmd_details *cmd_details)
 {
-	struct i40e_aq_desc desc;
-	struct i40e_aqc_lldp_get_mib *cmd =
-		(struct i40e_aqc_lldp_get_mib *)&desc.params.raw;
-	struct i40e_aqc_lldp_get_mib *resp =
-		(struct i40e_aqc_lldp_get_mib *)&desc.params.raw;
-	i40e_status status;
+	struct i40e_aqc_lldp_get_mib *resp;
+	struct i40e_aqc_lldp_get_mib *cmd;
+	struct libie_aq_desc desc;
+	int status;
 
 	if (buff_size == 0 || !buff)
-		return I40E_ERR_PARAM;
+		return -EINVAL;
 
 	i40e_fill_default_direct_cmd_desc(&desc, i40e_aqc_opc_lldp_get_mib);
 	/* Indirect Command */
-	desc.flags |= cpu_to_le16((u16)I40E_AQ_FLAG_BUF);
+	desc.flags |= cpu_to_le16((u16)LIBIE_AQ_FLAG_BUF);
 
+	resp = libie_aq_raw(&desc);
+	cmd = libie_aq_raw(&desc);
 	cmd->type = mib_type & I40E_AQ_LLDP_MIB_TYPE_MASK;
-	cmd->type |= ((bridge_type << I40E_AQ_LLDP_BRIDGE_TYPE_SHIFT) &
-		       I40E_AQ_LLDP_BRIDGE_TYPE_MASK);
+	cmd->type |= FIELD_PREP(I40E_AQ_LLDP_BRIDGE_TYPE_MASK, bridge_type);
 
 	desc.datalen = cpu_to_le16(buff_size);
 
-	desc.flags |= cpu_to_le16((u16)I40E_AQ_FLAG_BUF);
+	desc.flags |= cpu_to_le16((u16)LIBIE_AQ_FLAG_BUF);
 	if (buff_size > I40E_AQ_LARGE_BUF)
-		desc.flags |= cpu_to_le16((u16)I40E_AQ_FLAG_LB);
+		desc.flags |= cpu_to_le16((u16)LIBIE_AQ_FLAG_LB);
 
 	status = i40e_asq_send_command(hw, &desc, buff, buff_size, cmd_details);
 	if (!status) {
@@ -3506,6 +3039,46 @@ i40e_status i40e_aq_get_lldp_mib(struct i40e_hw *hw, u8 bridge_type,
 }
 
 /**
+ * i40e_aq_set_lldp_mib - Set the LLDP MIB
+ * @hw: pointer to the hw struct
+ * @mib_type: Local, Remote or both Local and Remote MIBs
+ * @buff: pointer to a user supplied buffer to store the MIB block
+ * @buff_size: size of the buffer (in bytes)
+ * @cmd_details: pointer to command details structure or NULL
+ *
+ * Set the LLDP MIB.
+ **/
+int
+i40e_aq_set_lldp_mib(struct i40e_hw *hw,
+		     u8 mib_type, void *buff, u16 buff_size,
+		     struct i40e_asq_cmd_details *cmd_details)
+{
+	struct i40e_aqc_lldp_set_local_mib *cmd;
+	struct libie_aq_desc desc;
+	int status;
+
+	cmd = libie_aq_raw(&desc);
+	if (buff_size == 0 || !buff)
+		return -EINVAL;
+
+	i40e_fill_default_direct_cmd_desc(&desc,
+					  i40e_aqc_opc_lldp_set_local_mib);
+	/* Indirect Command */
+	desc.flags |= cpu_to_le16((u16)(LIBIE_AQ_FLAG_BUF | LIBIE_AQ_FLAG_RD));
+	if (buff_size > I40E_AQ_LARGE_BUF)
+		desc.flags |= cpu_to_le16((u16)LIBIE_AQ_FLAG_LB);
+	desc.datalen = cpu_to_le16(buff_size);
+
+	cmd->type = mib_type;
+	cmd->length = cpu_to_le16(buff_size);
+	cmd->address_high = cpu_to_le32(upper_32_bits((uintptr_t)buff));
+	cmd->address_low = cpu_to_le32(lower_32_bits((uintptr_t)buff));
+
+	status = i40e_asq_send_command(hw, &desc, buff, buff_size, cmd_details);
+	return status;
+}
+
+/**
  * i40e_aq_cfg_lldp_mib_change_event
  * @hw: pointer to the hw struct
  * @enable_update: Enable or Disable event posting
@@ -3514,17 +3087,17 @@ i40e_status i40e_aq_get_lldp_mib(struct i40e_hw *hw, u8 bridge_type,
  * Enable or Disable posting of an event on ARQ when LLDP MIB
  * associated with the interface changes
  **/
-i40e_status i40e_aq_cfg_lldp_mib_change_event(struct i40e_hw *hw,
-				bool enable_update,
-				struct i40e_asq_cmd_details *cmd_details)
+int i40e_aq_cfg_lldp_mib_change_event(struct i40e_hw *hw,
+				      bool enable_update,
+				      struct i40e_asq_cmd_details *cmd_details)
 {
-	struct i40e_aq_desc desc;
-	struct i40e_aqc_lldp_update_mib *cmd =
-		(struct i40e_aqc_lldp_update_mib *)&desc.params.raw;
-	i40e_status status;
+	struct i40e_aqc_lldp_update_mib *cmd;
+	struct libie_aq_desc desc;
+	int status;
 
 	i40e_fill_default_direct_cmd_desc(&desc, i40e_aqc_opc_lldp_update_mib);
 
+	cmd = libie_aq_raw(&desc);
 	if (!enable_update)
 		cmd->command |= I40E_AQ_LLDP_MIB_UPDATE_DISABLE;
 
@@ -3537,23 +3110,33 @@ i40e_status i40e_aq_cfg_lldp_mib_change_event(struct i40e_hw *hw,
  * i40e_aq_stop_lldp
  * @hw: pointer to the hw struct
  * @shutdown_agent: True if LLDP Agent needs to be Shutdown
+ * @persist: True if stop of LLDP should be persistent across power cycles
  * @cmd_details: pointer to command details structure or NULL
  *
  * Stop or Shutdown the embedded LLDP Agent
  **/
-i40e_status i40e_aq_stop_lldp(struct i40e_hw *hw, bool shutdown_agent,
-				struct i40e_asq_cmd_details *cmd_details)
+int i40e_aq_stop_lldp(struct i40e_hw *hw, bool shutdown_agent,
+		      bool persist,
+		      struct i40e_asq_cmd_details *cmd_details)
 {
-	struct i40e_aq_desc desc;
-	struct i40e_aqc_lldp_stop *cmd =
-		(struct i40e_aqc_lldp_stop *)&desc.params.raw;
-	i40e_status status;
+	struct i40e_aqc_lldp_stop *cmd;
+	struct libie_aq_desc desc;
+	int status;
 
 	i40e_fill_default_direct_cmd_desc(&desc, i40e_aqc_opc_lldp_stop);
 
+	cmd = libie_aq_raw(&desc);
 	if (shutdown_agent)
 		cmd->command |= I40E_AQ_LLDP_AGENT_SHUTDOWN;
 
+	if (persist) {
+		if (test_bit(I40E_HW_CAP_FW_LLDP_PERSISTENT, hw->caps))
+			cmd->command |= I40E_AQ_LLDP_AGENT_STOP_PERSIST;
+		else
+			i40e_debug(hw, I40E_DEBUG_ALL,
+				   "Persistent Stop LLDP not supported by current FW version.\n");
+	}
+
 	status = i40e_asq_send_command(hw, &desc, NULL, 0, cmd_details);
 
 	return status;
@@ -3562,22 +3145,62 @@ i40e_status i40e_aq_stop_lldp(struct i40e_hw *hw, bool shutdown_agent,
 /**
  * i40e_aq_start_lldp
  * @hw: pointer to the hw struct
+ * @persist: True if start of LLDP should be persistent across power cycles
  * @cmd_details: pointer to command details structure or NULL
  *
  * Start the embedded LLDP Agent on all ports.
  **/
-i40e_status i40e_aq_start_lldp(struct i40e_hw *hw,
-				struct i40e_asq_cmd_details *cmd_details)
+int i40e_aq_start_lldp(struct i40e_hw *hw, bool persist,
+		       struct i40e_asq_cmd_details *cmd_details)
 {
-	struct i40e_aq_desc desc;
-	struct i40e_aqc_lldp_start *cmd =
-		(struct i40e_aqc_lldp_start *)&desc.params.raw;
-	i40e_status status;
+	struct i40e_aqc_lldp_start *cmd;
+	struct libie_aq_desc desc;
+	int status;
 
 	i40e_fill_default_direct_cmd_desc(&desc, i40e_aqc_opc_lldp_start);
 
+	cmd = libie_aq_raw(&desc);
 	cmd->command = I40E_AQ_LLDP_AGENT_START;
 
+	if (persist) {
+		if (test_bit(I40E_HW_CAP_FW_LLDP_PERSISTENT, hw->caps))
+			cmd->command |= I40E_AQ_LLDP_AGENT_START_PERSIST;
+		else
+			i40e_debug(hw, I40E_DEBUG_ALL,
+				   "Persistent Start LLDP not supported by current FW version.\n");
+	}
+
+	status = i40e_asq_send_command(hw, &desc, NULL, 0, cmd_details);
+
+	return status;
+}
+
+/**
+ * i40e_aq_set_dcb_parameters
+ * @hw: pointer to the hw struct
+ * @cmd_details: pointer to command details structure or NULL
+ * @dcb_enable: True if DCB configuration needs to be applied
+ *
+ **/
+int
+i40e_aq_set_dcb_parameters(struct i40e_hw *hw, bool dcb_enable,
+			   struct i40e_asq_cmd_details *cmd_details)
+{
+	struct i40e_aqc_set_dcb_parameters *cmd;
+	struct libie_aq_desc desc;
+	int status;
+
+	if (!test_bit(I40E_HW_CAP_FW_LLDP_STOPPABLE, hw->caps))
+		return -ENODEV;
+
+	i40e_fill_default_direct_cmd_desc(&desc,
+					  i40e_aqc_opc_set_dcb_parameters);
+
+	cmd = libie_aq_raw(&desc);
+	if (dcb_enable) {
+		cmd->valid_flags = I40E_DCB_VALID;
+		cmd->command = I40E_AQ_DCB_SET_AGENT;
+	}
 	status = i40e_asq_send_command(hw, &desc, NULL, 0, cmd_details);
 
 	return status;
@@ -3592,19 +3215,19 @@ i40e_status i40e_aq_start_lldp(struct i40e_hw *hw,
  *
  * Get CEE DCBX mode operational configuration from firmware
  **/
-i40e_status i40e_aq_get_cee_dcb_config(struct i40e_hw *hw,
-				       void *buff, u16 buff_size,
-				       struct i40e_asq_cmd_details *cmd_details)
+int i40e_aq_get_cee_dcb_config(struct i40e_hw *hw,
+			       void *buff, u16 buff_size,
+			       struct i40e_asq_cmd_details *cmd_details)
 {
-	struct i40e_aq_desc desc;
-	i40e_status status;
+	struct libie_aq_desc desc;
+	int status;
 
 	if (buff_size == 0 || !buff)
-		return I40E_ERR_PARAM;
+		return -EINVAL;
 
 	i40e_fill_default_direct_cmd_desc(&desc, i40e_aqc_opc_get_cee_dcb_cfg);
 
-	desc.flags |= cpu_to_le16((u16)I40E_AQ_FLAG_BUF);
+	desc.flags |= cpu_to_le16((u16)LIBIE_AQ_FLAG_BUF);
 	status = i40e_asq_send_command(hw, &desc, (void *)buff, buff_size,
 				       cmd_details);
 
@@ -3614,26 +3237,29 @@ i40e_status i40e_aq_get_cee_dcb_config(struct i40e_hw *hw,
 /**
  * i40e_aq_add_udp_tunnel
  * @hw: pointer to the hw struct
- * @udp_port: the UDP port to add
- * @header_len: length of the tunneling header length in DWords
+ * @udp_port: the UDP port to add in Host byte order
  * @protocol_index: protocol index type
  * @filter_index: pointer to filter index
  * @cmd_details: pointer to command details structure or NULL
+ *
+ * Note: Firmware expects the udp_port value to be in Little Endian format,
+ * and this function will call cpu_to_le16 to convert from Host byte order to
+ * Little Endian order.
  **/
-i40e_status i40e_aq_add_udp_tunnel(struct i40e_hw *hw,
-				u16 udp_port, u8 protocol_index,
-				u8 *filter_index,
-				struct i40e_asq_cmd_details *cmd_details)
+int i40e_aq_add_udp_tunnel(struct i40e_hw *hw,
+			   u16 udp_port, u8 protocol_index,
+			   u8 *filter_index,
+			   struct i40e_asq_cmd_details *cmd_details)
 {
-	struct i40e_aq_desc desc;
-	struct i40e_aqc_add_udp_tunnel *cmd =
-		(struct i40e_aqc_add_udp_tunnel *)&desc.params.raw;
-	struct i40e_aqc_del_udp_tunnel_completion *resp =
-		(struct i40e_aqc_del_udp_tunnel_completion *)&desc.params.raw;
-	i40e_status status;
+	struct i40e_aqc_del_udp_tunnel_completion *resp;
+	struct i40e_aqc_add_udp_tunnel *cmd;
+	struct libie_aq_desc desc;
+	int status;
 
 	i40e_fill_default_direct_cmd_desc(&desc, i40e_aqc_opc_add_udp_tunnel);
 
+	resp = libie_aq_raw(&desc);
+	cmd = libie_aq_raw(&desc);
 	cmd->udp_port = cpu_to_le16(udp_port);
 	cmd->protocol_type = protocol_index;
 
@@ -3651,16 +3277,16 @@ i40e_status i40e_aq_add_udp_tunnel(struct i40e_hw *hw,
  * @index: filter index
  * @cmd_details: pointer to command details structure or NULL
  **/
-i40e_status i40e_aq_del_udp_tunnel(struct i40e_hw *hw, u8 index,
-				struct i40e_asq_cmd_details *cmd_details)
+int i40e_aq_del_udp_tunnel(struct i40e_hw *hw, u8 index,
+			   struct i40e_asq_cmd_details *cmd_details)
 {
-	struct i40e_aq_desc desc;
-	struct i40e_aqc_remove_udp_tunnel *cmd =
-		(struct i40e_aqc_remove_udp_tunnel *)&desc.params.raw;
-	i40e_status status;
+	struct i40e_aqc_remove_udp_tunnel *cmd;
+	struct libie_aq_desc desc;
+	int status;
 
 	i40e_fill_default_direct_cmd_desc(&desc, i40e_aqc_opc_del_udp_tunnel);
 
+	cmd = libie_aq_raw(&desc);
 	cmd->index = index;
 
 	status = i40e_asq_send_command(hw, &desc, NULL, 0, cmd_details);
@@ -3676,22 +3302,23 @@ i40e_status i40e_aq_del_udp_tunnel(struct i40e_hw *hw, u8 index,
  *
  * This deletes a switch element from the switch.
  **/
-i40e_status i40e_aq_delete_element(struct i40e_hw *hw, u16 seid,
-				struct i40e_asq_cmd_details *cmd_details)
+int i40e_aq_delete_element(struct i40e_hw *hw, u16 seid,
+			   struct i40e_asq_cmd_details *cmd_details)
 {
-	struct i40e_aq_desc desc;
-	struct i40e_aqc_switch_seid *cmd =
-		(struct i40e_aqc_switch_seid *)&desc.params.raw;
-	i40e_status status;
+	struct i40e_aqc_switch_seid *cmd;
+	struct libie_aq_desc desc;
+	int status;
 
 	if (seid == 0)
-		return I40E_ERR_PARAM;
+		return -EINVAL;
 
 	i40e_fill_default_direct_cmd_desc(&desc, i40e_aqc_opc_delete_element);
 
+	cmd = libie_aq_raw(&desc);
 	cmd->seid = cpu_to_le16(seid);
 
-	status = i40e_asq_send_command(hw, &desc, NULL, 0, cmd_details);
+	status = i40e_asq_send_command_atomic(hw, &desc, NULL, 0,
+					      cmd_details, true);
 
 	return status;
 }
@@ -3705,11 +3332,11 @@ i40e_status i40e_aq_delete_element(struct i40e_hw *hw, u16 seid,
  * recomputed and modified. The retval field in the descriptor
  * will be set to 0 when RPB is modified.
  **/
-i40e_status i40e_aq_dcb_updated(struct i40e_hw *hw,
-				struct i40e_asq_cmd_details *cmd_details)
+int i40e_aq_dcb_updated(struct i40e_hw *hw,
+			struct i40e_asq_cmd_details *cmd_details)
 {
-	struct i40e_aq_desc desc;
-	i40e_status status;
+	struct libie_aq_desc desc;
+	int status;
 
 	i40e_fill_default_direct_cmd_desc(&desc, i40e_aqc_opc_dcb_updated);
 
@@ -3729,15 +3356,14 @@ i40e_status i40e_aq_dcb_updated(struct i40e_hw *hw,
  *
  * Generic command handler for Tx scheduler AQ commands
  **/
-static i40e_status i40e_aq_tx_sched_cmd(struct i40e_hw *hw, u16 seid,
+static int i40e_aq_tx_sched_cmd(struct i40e_hw *hw, u16 seid,
 				void *buff, u16 buff_size,
-				 enum i40e_admin_queue_opc opcode,
+				enum i40e_admin_queue_opc opcode,
 				struct i40e_asq_cmd_details *cmd_details)
 {
-	struct i40e_aq_desc desc;
-	struct i40e_aqc_tx_sched_ind *cmd =
-		(struct i40e_aqc_tx_sched_ind *)&desc.params.raw;
-	i40e_status status;
+	struct i40e_aqc_tx_sched_ind *cmd;
+	struct libie_aq_desc desc;
+	int status;
 	bool cmd_param_flag = false;
 
 	switch (opcode) {
@@ -3758,17 +3384,18 @@ static i40e_status i40e_aq_tx_sched_cmd(struct i40e_hw *hw, u16 seid,
 		cmd_param_flag = false;
 		break;
 	default:
-		return I40E_ERR_PARAM;
+		return -EINVAL;
 	}
 
 	i40e_fill_default_direct_cmd_desc(&desc, opcode);
 
+	cmd = libie_aq_raw(&desc);
 	/* Indirect command */
-	desc.flags |= cpu_to_le16((u16)I40E_AQ_FLAG_BUF);
+	desc.flags |= cpu_to_le16((u16)LIBIE_AQ_FLAG_BUF);
 	if (cmd_param_flag)
-		desc.flags |= cpu_to_le16((u16)I40E_AQ_FLAG_RD);
+		desc.flags |= cpu_to_le16((u16)LIBIE_AQ_FLAG_RD);
 	if (buff_size > I40E_AQ_LARGE_BUF)
-		desc.flags |= cpu_to_le16((u16)I40E_AQ_FLAG_LB);
+		desc.flags |= cpu_to_le16((u16)LIBIE_AQ_FLAG_LB);
 
 	desc.datalen = cpu_to_le16(buff_size);
 
@@ -3787,18 +3414,18 @@ static i40e_status i40e_aq_tx_sched_cmd(struct i40e_hw *hw, u16 seid,
  * @max_credit: Max BW limit credits
  * @cmd_details: pointer to command details structure or NULL
  **/
-i40e_status i40e_aq_config_vsi_bw_limit(struct i40e_hw *hw,
+int i40e_aq_config_vsi_bw_limit(struct i40e_hw *hw,
 				u16 seid, u16 credit, u8 max_credit,
 				struct i40e_asq_cmd_details *cmd_details)
 {
-	struct i40e_aq_desc desc;
-	struct i40e_aqc_configure_vsi_bw_limit *cmd =
-		(struct i40e_aqc_configure_vsi_bw_limit *)&desc.params.raw;
-	i40e_status status;
+	struct i40e_aqc_configure_vsi_bw_limit *cmd;
+	struct libie_aq_desc desc;
+	int status;
 
 	i40e_fill_default_direct_cmd_desc(&desc,
 					  i40e_aqc_opc_configure_vsi_bw_limit);
 
+	cmd = libie_aq_raw(&desc);
 	cmd->vsi_seid = cpu_to_le16(seid);
 	cmd->credit = cpu_to_le16(credit);
 	cmd->max_credit = max_credit;
@@ -3815,10 +3442,10 @@ i40e_status i40e_aq_config_vsi_bw_limit(struct i40e_hw *hw,
  * @bw_data: Buffer holding enabled TCs, relative TC BW limit/credits
  * @cmd_details: pointer to command details structure or NULL
  **/
-i40e_status i40e_aq_config_vsi_tc_bw(struct i40e_hw *hw,
-			u16 seid,
-			struct i40e_aqc_configure_vsi_tc_bw_data *bw_data,
-			struct i40e_asq_cmd_details *cmd_details)
+int i40e_aq_config_vsi_tc_bw(struct i40e_hw *hw,
+			     u16 seid,
+			     struct i40e_aqc_configure_vsi_tc_bw_data *bw_data,
+			     struct i40e_asq_cmd_details *cmd_details)
 {
 	return i40e_aq_tx_sched_cmd(hw, seid, (void *)bw_data, sizeof(*bw_data),
 				    i40e_aqc_opc_configure_vsi_tc_bw,
@@ -3830,13 +3457,15 @@ i40e_status i40e_aq_config_vsi_tc_bw(struct i40e_hw *hw,
  * @hw: pointer to the hw struct
  * @seid: seid of the switching component connected to Physical Port
  * @ets_data: Buffer holding ETS parameters
+ * @opcode: Tx scheduler AQ command opcode
  * @cmd_details: pointer to command details structure or NULL
  **/
-i40e_status i40e_aq_config_switch_comp_ets(struct i40e_hw *hw,
-		u16 seid,
-		struct i40e_aqc_configure_switching_comp_ets_data *ets_data,
-		enum i40e_admin_queue_opc opcode,
-		struct i40e_asq_cmd_details *cmd_details)
+int
+i40e_aq_config_switch_comp_ets(struct i40e_hw *hw,
+			       u16 seid,
+			       struct i40e_aqc_configure_switching_comp_ets_data *ets_data,
+			       enum i40e_admin_queue_opc opcode,
+			       struct i40e_asq_cmd_details *cmd_details)
 {
 	return i40e_aq_tx_sched_cmd(hw, seid, (void *)ets_data,
 				    sizeof(*ets_data), opcode, cmd_details);
@@ -3849,7 +3478,8 @@ i40e_status i40e_aq_config_switch_comp_ets(struct i40e_hw *hw,
  * @bw_data: Buffer holding enabled TCs, relative/absolute TC BW limit/credits
  * @cmd_details: pointer to command details structure or NULL
  **/
-i40e_status i40e_aq_config_switch_comp_bw_config(struct i40e_hw *hw,
+int
+i40e_aq_config_switch_comp_bw_config(struct i40e_hw *hw,
 	u16 seid,
 	struct i40e_aqc_configure_switching_comp_bw_config_data *bw_data,
 	struct i40e_asq_cmd_details *cmd_details)
@@ -3866,10 +3496,11 @@ i40e_status i40e_aq_config_switch_comp_bw_config(struct i40e_hw *hw,
  * @bw_data: Buffer to hold VSI BW configuration
  * @cmd_details: pointer to command details structure or NULL
  **/
-i40e_status i40e_aq_query_vsi_bw_config(struct i40e_hw *hw,
-			u16 seid,
-			struct i40e_aqc_query_vsi_bw_config_resp *bw_data,
-			struct i40e_asq_cmd_details *cmd_details)
+int
+i40e_aq_query_vsi_bw_config(struct i40e_hw *hw,
+			    u16 seid,
+			    struct i40e_aqc_query_vsi_bw_config_resp *bw_data,
+			    struct i40e_asq_cmd_details *cmd_details)
 {
 	return i40e_aq_tx_sched_cmd(hw, seid, (void *)bw_data, sizeof(*bw_data),
 				    i40e_aqc_opc_query_vsi_bw_config,
@@ -3883,10 +3514,11 @@ i40e_status i40e_aq_query_vsi_bw_config(struct i40e_hw *hw,
  * @bw_data: Buffer to hold VSI BW configuration per TC
  * @cmd_details: pointer to command details structure or NULL
  **/
-i40e_status i40e_aq_query_vsi_ets_sla_config(struct i40e_hw *hw,
-			u16 seid,
-			struct i40e_aqc_query_vsi_ets_sla_config_resp *bw_data,
-			struct i40e_asq_cmd_details *cmd_details)
+int
+i40e_aq_query_vsi_ets_sla_config(struct i40e_hw *hw,
+				 u16 seid,
+				 struct i40e_aqc_query_vsi_ets_sla_config_resp *bw_data,
+				 struct i40e_asq_cmd_details *cmd_details)
 {
 	return i40e_aq_tx_sched_cmd(hw, seid, (void *)bw_data, sizeof(*bw_data),
 				    i40e_aqc_opc_query_vsi_ets_sla_config,
@@ -3900,10 +3532,11 @@ i40e_status i40e_aq_query_vsi_ets_sla_config(struct i40e_hw *hw,
  * @bw_data: Buffer to hold switching component's per TC BW config
  * @cmd_details: pointer to command details structure or NULL
  **/
-i40e_status i40e_aq_query_switch_comp_ets_config(struct i40e_hw *hw,
-		u16 seid,
-		struct i40e_aqc_query_switching_comp_ets_config_resp *bw_data,
-		struct i40e_asq_cmd_details *cmd_details)
+int
+i40e_aq_query_switch_comp_ets_config(struct i40e_hw *hw,
+				     u16 seid,
+				     struct i40e_aqc_query_switching_comp_ets_config_resp *bw_data,
+				     struct i40e_asq_cmd_details *cmd_details)
 {
 	return i40e_aq_tx_sched_cmd(hw, seid, (void *)bw_data, sizeof(*bw_data),
 				   i40e_aqc_opc_query_switching_comp_ets_config,
@@ -3917,10 +3550,11 @@ i40e_status i40e_aq_query_switch_comp_ets_config(struct i40e_hw *hw,
  * @bw_data: Buffer to hold current ETS configuration for the Physical Port
  * @cmd_details: pointer to command details structure or NULL
  **/
-i40e_status i40e_aq_query_port_ets_config(struct i40e_hw *hw,
-			u16 seid,
-			struct i40e_aqc_query_port_ets_config_resp *bw_data,
-			struct i40e_asq_cmd_details *cmd_details)
+int
+i40e_aq_query_port_ets_config(struct i40e_hw *hw,
+			      u16 seid,
+			      struct i40e_aqc_query_port_ets_config_resp *bw_data,
+			      struct i40e_asq_cmd_details *cmd_details)
 {
 	return i40e_aq_tx_sched_cmd(hw, seid, (void *)bw_data, sizeof(*bw_data),
 				    i40e_aqc_opc_query_port_ets_config,
@@ -3934,10 +3568,11 @@ i40e_status i40e_aq_query_port_ets_config(struct i40e_hw *hw,
  * @bw_data: Buffer to hold switching component's BW configuration
  * @cmd_details: pointer to command details structure or NULL
  **/
-i40e_status i40e_aq_query_switch_comp_bw_config(struct i40e_hw *hw,
-		u16 seid,
-		struct i40e_aqc_query_switching_comp_bw_config_resp *bw_data,
-		struct i40e_asq_cmd_details *cmd_details)
+int
+i40e_aq_query_switch_comp_bw_config(struct i40e_hw *hw,
+				    u16 seid,
+				    struct i40e_aqc_query_switching_comp_bw_config_resp *bw_data,
+				    struct i40e_asq_cmd_details *cmd_details)
 {
 	return i40e_aq_tx_sched_cmd(hw, seid, (void *)bw_data, sizeof(*bw_data),
 				    i40e_aqc_opc_query_switching_comp_bw_config,
@@ -3956,11 +3591,11 @@ i40e_status i40e_aq_query_switch_comp_bw_config(struct i40e_hw *hw,
  * Returns 0 if the values passed are valid and within
  * range else returns an error.
  **/
-static i40e_status i40e_validate_filter_settings(struct i40e_hw *hw,
-				struct i40e_filter_control_settings *settings)
+static int
+i40e_validate_filter_settings(struct i40e_hw *hw,
+			      struct i40e_filter_control_settings *settings)
 {
 	u32 fcoe_cntx_size, fcoe_filt_size;
-	u32 pe_cntx_size, pe_filt_size;
 	u32 fcoe_fmax;
 	u32 val;
 
@@ -3976,7 +3611,7 @@ static i40e_status i40e_validate_filter_settings(struct i40e_hw *hw,
 		fcoe_filt_size <<= (u32)settings->fcoe_filt_num;
 		break;
 	default:
-		return I40E_ERR_PARAM;
+		return -EINVAL;
 	}
 
 	switch (settings->fcoe_cntx_num) {
@@ -3988,7 +3623,7 @@ static i40e_status i40e_validate_filter_settings(struct i40e_hw *hw,
 		fcoe_cntx_size <<= (u32)settings->fcoe_cntx_num;
 		break;
 	default:
-		return I40E_ERR_PARAM;
+		return -EINVAL;
 	}
 
 	/* Validate PE settings passed */
@@ -4004,11 +3639,9 @@ static i40e_status i40e_validate_filter_settings(struct i40e_hw *hw,
 	case I40E_HASH_FILTER_SIZE_256K:
 	case I40E_HASH_FILTER_SIZE_512K:
 	case I40E_HASH_FILTER_SIZE_1M:
-		pe_filt_size = I40E_HASH_FILTER_BASE_SIZE;
-		pe_filt_size <<= (u32)settings->pe_filt_num;
 		break;
 	default:
-		return I40E_ERR_PARAM;
+		return -EINVAL;
 	}
 
 	switch (settings->pe_cntx_num) {
@@ -4022,19 +3655,16 @@ static i40e_status i40e_validate_filter_settings(struct i40e_hw *hw,
 	case I40E_DMA_CNTX_SIZE_64K:
 	case I40E_DMA_CNTX_SIZE_128K:
 	case I40E_DMA_CNTX_SIZE_256K:
-		pe_cntx_size = I40E_DMA_CNTX_BASE_SIZE;
-		pe_cntx_size <<= (u32)settings->pe_cntx_num;
 		break;
 	default:
-		return I40E_ERR_PARAM;
+		return -EINVAL;
 	}
 
 	/* FCHSIZE + FCDSIZE should not be greater than PMFCOEFMAX */
 	val = rd32(hw, I40E_GLHMC_FCOEFMAX);
-	fcoe_fmax = (val & I40E_GLHMC_FCOEFMAX_PMFCOEFMAX_MASK)
-		     >> I40E_GLHMC_FCOEFMAX_PMFCOEFMAX_SHIFT;
+	fcoe_fmax = FIELD_GET(I40E_GLHMC_FCOEFMAX_PMFCOEFMAX_MASK, val);
 	if (fcoe_filt_size + fcoe_cntx_size >  fcoe_fmax)
-		return I40E_ERR_INVALID_SIZE;
+		return -EINVAL;
 
 	return 0;
 }
@@ -4048,15 +3678,15 @@ static i40e_status i40e_validate_filter_settings(struct i40e_hw *hw,
  * for a single PF. It is expected that these settings are programmed
  * at the driver initialization time.
  **/
-i40e_status i40e_set_filter_control(struct i40e_hw *hw,
-				struct i40e_filter_control_settings *settings)
+int i40e_set_filter_control(struct i40e_hw *hw,
+			    struct i40e_filter_control_settings *settings)
 {
-	i40e_status ret = 0;
 	u32 hash_lut_size = 0;
+	int ret = 0;
 	u32 val;
 
 	if (!settings)
-		return I40E_ERR_PARAM;
+		return -EINVAL;
 
 	/* Validate the input settings */
 	ret = i40e_validate_filter_settings(hw, settings);
@@ -4068,30 +3698,25 @@ i40e_status i40e_set_filter_control(struct i40e_hw *hw,
 
 	/* Program required PE hash buckets for the PF */
 	val &= ~I40E_PFQF_CTL_0_PEHSIZE_MASK;
-	val |= ((u32)settings->pe_filt_num << I40E_PFQF_CTL_0_PEHSIZE_SHIFT) &
-		I40E_PFQF_CTL_0_PEHSIZE_MASK;
+	val |= FIELD_PREP(I40E_PFQF_CTL_0_PEHSIZE_MASK, settings->pe_filt_num);
 	/* Program required PE contexts for the PF */
 	val &= ~I40E_PFQF_CTL_0_PEDSIZE_MASK;
-	val |= ((u32)settings->pe_cntx_num << I40E_PFQF_CTL_0_PEDSIZE_SHIFT) &
-		I40E_PFQF_CTL_0_PEDSIZE_MASK;
+	val |= FIELD_PREP(I40E_PFQF_CTL_0_PEDSIZE_MASK, settings->pe_cntx_num);
 
 	/* Program required FCoE hash buckets for the PF */
 	val &= ~I40E_PFQF_CTL_0_PFFCHSIZE_MASK;
-	val |= ((u32)settings->fcoe_filt_num <<
-			I40E_PFQF_CTL_0_PFFCHSIZE_SHIFT) &
-		I40E_PFQF_CTL_0_PFFCHSIZE_MASK;
+	val |= FIELD_PREP(I40E_PFQF_CTL_0_PFFCHSIZE_MASK,
+			  settings->fcoe_filt_num);
 	/* Program required FCoE DDP contexts for the PF */
 	val &= ~I40E_PFQF_CTL_0_PFFCDSIZE_MASK;
-	val |= ((u32)settings->fcoe_cntx_num <<
-			I40E_PFQF_CTL_0_PFFCDSIZE_SHIFT) &
-		I40E_PFQF_CTL_0_PFFCDSIZE_MASK;
+	val |= FIELD_PREP(I40E_PFQF_CTL_0_PFFCDSIZE_MASK,
+			  settings->fcoe_cntx_num);
 
 	/* Program Hash LUT size for the PF */
 	val &= ~I40E_PFQF_CTL_0_HASHLUTSIZE_MASK;
 	if (settings->hash_lut_size == I40E_HASH_LUT_SIZE_512)
 		hash_lut_size = 1;
-	val |= (hash_lut_size << I40E_PFQF_CTL_0_HASHLUTSIZE_SHIFT) &
-		I40E_PFQF_CTL_0_HASHLUTSIZE_MASK;
+	val |= FIELD_PREP(I40E_PFQF_CTL_0_HASHLUTSIZE_MASK, hash_lut_size);
 
 	/* Enable FDIR, Ethertype and MACVLAN filters for PF and VFs */
 	if (settings->enable_fdir)
@@ -4122,24 +3747,22 @@ i40e_status i40e_set_filter_control(struct i40e_hw *hw,
  * In return it will update the total number of perfect filter count in
  * the stats member.
  **/
-i40e_status i40e_aq_add_rem_control_packet_filter(struct i40e_hw *hw,
-				u8 *mac_addr, u16 ethtype, u16 flags,
-				u16 vsi_seid, u16 queue, bool is_add,
-				struct i40e_control_filter_stats *stats,
-				struct i40e_asq_cmd_details *cmd_details)
+int i40e_aq_add_rem_control_packet_filter(struct i40e_hw *hw,
+					  u8 *mac_addr, u16 ethtype, u16 flags,
+					  u16 vsi_seid, u16 queue, bool is_add,
+					  struct i40e_control_filter_stats *stats,
+					  struct i40e_asq_cmd_details *cmd_details)
 {
-	struct i40e_aq_desc desc;
-	struct i40e_aqc_add_remove_control_packet_filter *cmd =
-		(struct i40e_aqc_add_remove_control_packet_filter *)
-		&desc.params.raw;
-	struct i40e_aqc_add_remove_control_packet_filter_completion *resp =
-		(struct i40e_aqc_add_remove_control_packet_filter_completion *)
-		&desc.params.raw;
-	i40e_status status;
+	struct i40e_aqc_add_remove_control_packet_filter_completion *resp;
+	struct i40e_aqc_add_remove_control_packet_filter *cmd;
+	struct libie_aq_desc desc;
+	int status;
 
 	if (vsi_seid == 0)
-		return I40E_ERR_PARAM;
+		return -EINVAL;
 
+	resp = libie_aq_raw(&desc);
+	cmd = libie_aq_raw(&desc);
 	if (is_add) {
 		i40e_fill_default_direct_cmd_desc(&desc,
 				i40e_aqc_opc_add_control_packet_filter);
@@ -4173,15 +3796,15 @@ i40e_status i40e_aq_add_rem_control_packet_filter(struct i40e_hw *hw,
  * @hw: pointer to the hw struct
  * @seid: VSI seid to add ethertype filter from
  **/
-#define I40E_FLOW_CONTROL_ETHTYPE 0x8808
 void i40e_add_filter_to_drop_tx_flow_control_frames(struct i40e_hw *hw,
 						    u16 seid)
 {
+#define I40E_FLOW_CONTROL_ETHTYPE 0x8808
 	u16 flag = I40E_AQC_ADD_CONTROL_PACKET_FLAGS_IGNORE_MAC |
 		   I40E_AQC_ADD_CONTROL_PACKET_FLAGS_DROP |
 		   I40E_AQC_ADD_CONTROL_PACKET_FLAGS_TX;
 	u16 ethtype = I40E_FLOW_CONTROL_ETHTYPE;
-	i40e_status status;
+	int status;
 
 	status = i40e_aq_add_rem_control_packet_filter(hw, NULL, ethtype, flag,
 						       seid, 0, true, NULL,
@@ -4203,19 +3826,19 @@ void i40e_add_filter_to_drop_tx_flow_control_frames(struct i40e_hw *hw,
  * is not passed then only register at 'reg_addr0' is read.
  *
  **/
-static i40e_status i40e_aq_alternate_read(struct i40e_hw *hw,
-					  u32 reg_addr0, u32 *reg_val0,
-					  u32 reg_addr1, u32 *reg_val1)
+static int i40e_aq_alternate_read(struct i40e_hw *hw,
+				  u32 reg_addr0, u32 *reg_val0,
+				  u32 reg_addr1, u32 *reg_val1)
 {
-	struct i40e_aq_desc desc;
-	struct i40e_aqc_alternate_write *cmd_resp =
-		(struct i40e_aqc_alternate_write *)&desc.params.raw;
-	i40e_status status;
+	struct i40e_aqc_alternate_write *cmd_resp;
+	struct libie_aq_desc desc;
+	int status;
 
 	if (!reg_val0)
-		return I40E_ERR_PARAM;
+		return -EINVAL;
 
 	i40e_fill_default_direct_cmd_desc(&desc, i40e_aqc_opc_alternate_read);
+	cmd_resp = libie_aq_raw(&desc);
 	cmd_resp->address0 = cpu_to_le32(reg_addr0);
 	cmd_resp->address1 = cpu_to_le32(reg_addr1);
 
@@ -4232,17 +3855,40 @@ static i40e_status i40e_aq_alternate_read(struct i40e_hw *hw,
 }
 
 /**
+ * i40e_aq_suspend_port_tx
+ * @hw: pointer to the hardware structure
+ * @seid: port seid
+ * @cmd_details: pointer to command details structure or NULL
+ *
+ * Suspend port's Tx traffic
+ **/
+int i40e_aq_suspend_port_tx(struct i40e_hw *hw, u16 seid,
+			    struct i40e_asq_cmd_details *cmd_details)
+{
+	struct i40e_aqc_tx_sched_ind *cmd;
+	struct libie_aq_desc desc;
+	int status;
+
+	cmd = libie_aq_raw(&desc);
+	i40e_fill_default_direct_cmd_desc(&desc, i40e_aqc_opc_suspend_port_tx);
+	cmd->vsi_seid = cpu_to_le16(seid);
+	status = i40e_asq_send_command(hw, &desc, NULL, 0, cmd_details);
+
+	return status;
+}
+
+/**
  * i40e_aq_resume_port_tx
  * @hw: pointer to the hardware structure
  * @cmd_details: pointer to command details structure or NULL
  *
  * Resume port's Tx traffic
  **/
-i40e_status i40e_aq_resume_port_tx(struct i40e_hw *hw,
-				   struct i40e_asq_cmd_details *cmd_details)
+int i40e_aq_resume_port_tx(struct i40e_hw *hw,
+			   struct i40e_asq_cmd_details *cmd_details)
 {
-	struct i40e_aq_desc desc;
-	i40e_status status;
+	struct libie_aq_desc desc;
+	int status;
 
 	i40e_fill_default_direct_cmd_desc(&desc, i40e_aqc_opc_resume_port_tx);
 
@@ -4307,32 +3953,33 @@ void i40e_set_pci_config_data(struct i40e_hw *hw, u16 link_status)
  * @ret_buff_size: actual buffer size returned
  * @ret_next_table: next block to read
  * @ret_next_index: next index to read
+ * @cmd_details: pointer to command details structure or NULL
  *
  * Dump internal FW/HW data for debug purposes.
  *
  **/
-i40e_status i40e_aq_debug_dump(struct i40e_hw *hw, u8 cluster_id,
-			       u8 table_id, u32 start_index, u16 buff_size,
-			       void *buff, u16 *ret_buff_size,
-			       u8 *ret_next_table, u32 *ret_next_index,
-			       struct i40e_asq_cmd_details *cmd_details)
+int i40e_aq_debug_dump(struct i40e_hw *hw, u8 cluster_id,
+		       u8 table_id, u32 start_index, u16 buff_size,
+		       void *buff, u16 *ret_buff_size,
+		       u8 *ret_next_table, u32 *ret_next_index,
+		       struct i40e_asq_cmd_details *cmd_details)
 {
-	struct i40e_aq_desc desc;
-	struct i40e_aqc_debug_dump_internals *cmd =
-		(struct i40e_aqc_debug_dump_internals *)&desc.params.raw;
-	struct i40e_aqc_debug_dump_internals *resp =
-		(struct i40e_aqc_debug_dump_internals *)&desc.params.raw;
-	i40e_status status;
+	struct i40e_aqc_debug_dump_internals *resp;
+	struct i40e_aqc_debug_dump_internals *cmd;
+	struct libie_aq_desc desc;
+	int status;
 
 	if (buff_size == 0 || !buff)
-		return I40E_ERR_PARAM;
+		return -EINVAL;
 
 	i40e_fill_default_direct_cmd_desc(&desc,
 					  i40e_aqc_opc_debug_dump_internals);
+	resp = libie_aq_raw(&desc);
+	cmd = libie_aq_raw(&desc);
 	/* Indirect Command */
-	desc.flags |= cpu_to_le16((u16)I40E_AQ_FLAG_BUF);
+	desc.flags |= cpu_to_le16((u16)LIBIE_AQ_FLAG_BUF);
 	if (buff_size > I40E_AQ_LARGE_BUF)
-		desc.flags |= cpu_to_le16((u16)I40E_AQ_FLAG_LB);
+		desc.flags |= cpu_to_le16((u16)LIBIE_AQ_FLAG_LB);
 
 	cmd->cluster_id = cluster_id;
 	cmd->table_id = table_id;
@@ -4363,12 +4010,12 @@ i40e_status i40e_aq_debug_dump(struct i40e_hw *hw, u8 cluster_id,
  *
  * Read bw from the alternate ram for the given pf
  **/
-i40e_status i40e_read_bw_from_alt_ram(struct i40e_hw *hw,
-				      u32 *max_bw, u32 *min_bw,
-				      bool *min_valid, bool *max_valid)
+int i40e_read_bw_from_alt_ram(struct i40e_hw *hw,
+			      u32 *max_bw, u32 *min_bw,
+			      bool *min_valid, bool *max_valid)
 {
-	i40e_status status;
 	u32 max_bw_addr, min_bw_addr;
+	int status;
 
 	/* Calculate the address of the min/max bw registers */
 	max_bw_addr = I40E_ALT_STRUCT_FIRST_PF_OFFSET +
@@ -4403,23 +4050,24 @@ i40e_status i40e_read_bw_from_alt_ram(struct i40e_hw *hw,
  *
  * Configure partitions guaranteed/max bw
  **/
-i40e_status i40e_aq_configure_partition_bw(struct i40e_hw *hw,
-			struct i40e_aqc_configure_partition_bw_data *bw_data,
-			struct i40e_asq_cmd_details *cmd_details)
+int
+i40e_aq_configure_partition_bw(struct i40e_hw *hw,
+			       struct i40e_aqc_configure_partition_bw_data *bw_data,
+			       struct i40e_asq_cmd_details *cmd_details)
 {
-	i40e_status status;
-	struct i40e_aq_desc desc;
 	u16 bwd_size = sizeof(*bw_data);
+	struct libie_aq_desc desc;
+	int status;
 
 	i40e_fill_default_direct_cmd_desc(&desc,
 					  i40e_aqc_opc_configure_partition_bw);
 
 	/* Indirect command */
-	desc.flags |= cpu_to_le16((u16)I40E_AQ_FLAG_BUF);
-	desc.flags |= cpu_to_le16((u16)I40E_AQ_FLAG_RD);
+	desc.flags |= cpu_to_le16((u16)LIBIE_AQ_FLAG_BUF);
+	desc.flags |= cpu_to_le16((u16)LIBIE_AQ_FLAG_RD);
 
 	if (bwd_size > I40E_AQ_LARGE_BUF)
-		desc.flags |= cpu_to_le16((u16)I40E_AQ_FLAG_LB);
+		desc.flags |= cpu_to_le16((u16)LIBIE_AQ_FLAG_LB);
 
 	desc.datalen = cpu_to_le16(bwd_size);
 
@@ -4433,16 +4081,16 @@ i40e_status i40e_aq_configure_partition_bw(struct i40e_hw *hw,
  * i40e_read_phy_register_clause22
  * @hw: pointer to the HW structure
  * @reg: register address in the page
- * @phy_adr: PHY address on MDIO interface
+ * @phy_addr: PHY address on MDIO interface
  * @value: PHY register value
  *
  * Reads specified PHY register value
  **/
-i40e_status i40e_read_phy_register_clause22(struct i40e_hw *hw,
-					    u16 reg, u8 phy_addr, u16 *value)
+int i40e_read_phy_register_clause22(struct i40e_hw *hw,
+				    u16 reg, u8 phy_addr, u16 *value)
 {
-	i40e_status status = I40E_ERR_TIMEOUT;
 	u8 port_num = (u8)hw->func_caps.mdio_port_num;
+	int status = -EIO;
 	u32 command = 0;
 	u16 retry = 1000;
 
@@ -4467,8 +4115,7 @@ i40e_status i40e_read_phy_register_clause22(struct i40e_hw *hw,
 			   "PHY: Can't write command to external PHY.\n");
 	} else {
 		command = rd32(hw, I40E_GLGEN_MSRWD(port_num));
-		*value = (command & I40E_GLGEN_MSRWD_MDIRDDATA_MASK) >>
-			 I40E_GLGEN_MSRWD_MDIRDDATA_SHIFT;
+		*value = FIELD_GET(I40E_GLGEN_MSRWD_MDIRDDATA_MASK, command);
 	}
 
 	return status;
@@ -4478,16 +4125,16 @@ i40e_status i40e_read_phy_register_clause22(struct i40e_hw *hw,
  * i40e_write_phy_register_clause22
  * @hw: pointer to the HW structure
  * @reg: register address in the page
- * @phy_adr: PHY address on MDIO interface
+ * @phy_addr: PHY address on MDIO interface
  * @value: PHY register value
  *
  * Writes specified PHY register value
  **/
-i40e_status i40e_write_phy_register_clause22(struct i40e_hw *hw,
-					     u16 reg, u8 phy_addr, u16 value)
+int i40e_write_phy_register_clause22(struct i40e_hw *hw,
+				     u16 reg, u8 phy_addr, u16 value)
 {
-	i40e_status status = I40E_ERR_TIMEOUT;
 	u8 port_num = (u8)hw->func_caps.mdio_port_num;
+	int status = -EIO;
 	u32 command  = 0;
 	u16 retry = 1000;
 
@@ -4519,18 +4166,18 @@ i40e_status i40e_write_phy_register_clause22(struct i40e_hw *hw,
  * @hw: pointer to the HW structure
  * @page: registers page number
  * @reg: register address in the page
- * @phy_adr: PHY address on MDIO interface
+ * @phy_addr: PHY address on MDIO interface
  * @value: PHY register value
  *
  * Reads specified PHY register value
  **/
-i40e_status i40e_read_phy_register_clause45(struct i40e_hw *hw,
-				u8 page, u16 reg, u8 phy_addr, u16 *value)
+int i40e_read_phy_register_clause45(struct i40e_hw *hw,
+				    u8 page, u16 reg, u8 phy_addr, u16 *value)
 {
-	i40e_status status = I40E_ERR_TIMEOUT;
+	u8 port_num = hw->func_caps.mdio_port_num;
+	int status = -EIO;
 	u32 command = 0;
 	u16 retry = 1000;
-	u8 port_num = hw->func_caps.mdio_port_num;
 
 	command = (reg << I40E_GLGEN_MSCA_MDIADD_SHIFT) |
 		  (page << I40E_GLGEN_MSCA_DEVADD_SHIFT) |
@@ -4562,7 +4209,7 @@ i40e_status i40e_read_phy_register_clause45(struct i40e_hw *hw,
 		  (I40E_MDIO_CLAUSE45_STCODE_MASK) |
 		  (I40E_GLGEN_MSCA_MDICMD_MASK) |
 		  (I40E_GLGEN_MSCA_MDIINPROGEN_MASK);
-	status = I40E_ERR_TIMEOUT;
+	status = -EIO;
 	retry = 1000;
 	wr32(hw, I40E_GLGEN_MSCA(port_num), command);
 	do {
@@ -4577,8 +4224,7 @@ i40e_status i40e_read_phy_register_clause45(struct i40e_hw *hw,
 
 	if (!status) {
 		command = rd32(hw, I40E_GLGEN_MSRWD(port_num));
-		*value = (command & I40E_GLGEN_MSRWD_MDIRDDATA_MASK) >>
-			 I40E_GLGEN_MSRWD_MDIRDDATA_SHIFT;
+		*value = FIELD_GET(I40E_GLGEN_MSRWD_MDIRDDATA_MASK, command);
 	} else {
 		i40e_debug(hw, I40E_DEBUG_PHY,
 			   "PHY: Can't read register value from external PHY.\n");
@@ -4593,18 +4239,18 @@ phy_read_end:
  * @hw: pointer to the HW structure
  * @page: registers page number
  * @reg: register address in the page
- * @phy_adr: PHY address on MDIO interface
+ * @phy_addr: PHY address on MDIO interface
  * @value: PHY register value
  *
  * Writes value to specified PHY register
  **/
-i40e_status i40e_write_phy_register_clause45(struct i40e_hw *hw,
-				u8 page, u16 reg, u8 phy_addr, u16 value)
+int i40e_write_phy_register_clause45(struct i40e_hw *hw,
+				     u8 page, u16 reg, u8 phy_addr, u16 value)
 {
-	i40e_status status = I40E_ERR_TIMEOUT;
-	u32 command = 0;
-	u16 retry = 1000;
 	u8 port_num = hw->func_caps.mdio_port_num;
+	int status = -EIO;
+	u16 retry = 1000;
+	u32 command = 0;
 
 	command = (reg << I40E_GLGEN_MSCA_MDIADD_SHIFT) |
 		  (page << I40E_GLGEN_MSCA_DEVADD_SHIFT) |
@@ -4638,7 +4284,7 @@ i40e_status i40e_write_phy_register_clause45(struct i40e_hw *hw,
 		  (I40E_MDIO_CLAUSE45_STCODE_MASK) |
 		  (I40E_GLGEN_MSCA_MDICMD_MASK) |
 		  (I40E_GLGEN_MSCA_MDIINPROGEN_MASK);
-	status = I40E_ERR_TIMEOUT;
+	status = -EIO;
 	retry = 1000;
 	wr32(hw, I40E_GLGEN_MSCA(port_num), command);
 	do {
@@ -4656,82 +4302,9 @@ phy_write_end:
 }
 
 /**
- * i40e_write_phy_register
- * @hw: pointer to the HW structure
- * @page: registers page number
- * @reg: register address in the page
- * @phy_adr: PHY address on MDIO interface
- * @value: PHY register value
- *
- * Writes value to specified PHY register
- **/
-i40e_status i40e_write_phy_register(struct i40e_hw *hw,
-				    u8 page, u16 reg, u8 phy_addr, u16 value)
-{
-	i40e_status status;
-
-	switch (hw->device_id) {
-	case I40E_DEV_ID_1G_BASE_T_X722:
-		status = i40e_write_phy_register_clause22(hw, reg, phy_addr,
-							  value);
-		break;
-	case I40E_DEV_ID_10G_BASE_T:
-	case I40E_DEV_ID_10G_BASE_T4:
-	case I40E_DEV_ID_10G_BASE_T_X722:
-	case I40E_DEV_ID_25G_B:
-	case I40E_DEV_ID_25G_SFP28:
-		status = i40e_write_phy_register_clause45(hw, page, reg,
-							  phy_addr, value);
-		break;
-	default:
-		status = I40E_ERR_UNKNOWN_PHY;
-		break;
-	}
-
-	return status;
-}
-
-/**
- * i40e_read_phy_register
- * @hw: pointer to the HW structure
- * @page: registers page number
- * @reg: register address in the page
- * @phy_adr: PHY address on MDIO interface
- * @value: PHY register value
- *
- * Reads specified PHY register value
- **/
-i40e_status i40e_read_phy_register(struct i40e_hw *hw,
-				   u8 page, u16 reg, u8 phy_addr, u16 *value)
-{
-	i40e_status status;
-
-	switch (hw->device_id) {
-	case I40E_DEV_ID_1G_BASE_T_X722:
-		status = i40e_read_phy_register_clause22(hw, reg, phy_addr,
-							 value);
-		break;
-	case I40E_DEV_ID_10G_BASE_T:
-	case I40E_DEV_ID_10G_BASE_T4:
-	case I40E_DEV_ID_10G_BASE_T_X722:
-	case I40E_DEV_ID_25G_B:
-	case I40E_DEV_ID_25G_SFP28:
-		status = i40e_read_phy_register_clause45(hw, page, reg,
-							 phy_addr, value);
-		break;
-	default:
-		status = I40E_ERR_UNKNOWN_PHY;
-		break;
-	}
-
-	return status;
-}
-
-/**
  * i40e_get_phy_address
  * @hw: pointer to the HW structure
  * @dev_num: PHY port num that address we want
- * @phy_addr: Returned PHY address
  *
  * Gets PHY address for current port
  **/
@@ -4744,76 +4317,70 @@ u8 i40e_get_phy_address(struct i40e_hw *hw, u8 dev_num)
 }
 
 /**
- * i40e_blink_phy_led
+ * i40e_led_get_reg - read LED register
  * @hw: pointer to the HW structure
- * @time: time how long led will blinks in secs
- * @interval: gap between LED on and off in msecs
- *
- * Blinks PHY link LED
+ * @led_addr: LED register address
+ * @reg_val: read register value
  **/
-i40e_status i40e_blink_phy_link_led(struct i40e_hw *hw,
-				    u32 time, u32 interval)
+static int i40e_led_get_reg(struct i40e_hw *hw, u16 led_addr,
+			    u32 *reg_val)
 {
-	i40e_status status = 0;
-	u32 i;
-	u16 led_ctl;
-	u16 gpio_led_port;
-	u16 led_reg;
-	u16 led_addr = I40E_PHY_LED_PROV_REG_1;
 	u8 phy_addr = 0;
 	u8 port_num;
+	int status;
+	u32 i;
 
-	i = rd32(hw, I40E_PFGEN_PORTNUM);
-	port_num = (u8)(i & I40E_PFGEN_PORTNUM_PORT_NUM_MASK);
-	phy_addr = i40e_get_phy_address(hw, port_num);
-
-	for (gpio_led_port = 0; gpio_led_port < 3; gpio_led_port++,
-	     led_addr++) {
+	*reg_val = 0;
+	if (test_bit(I40E_HW_CAP_AQ_PHY_ACCESS, hw->caps)) {
+		status =
+		       i40e_aq_get_phy_register(hw,
+						I40E_AQ_PHY_REG_ACCESS_EXTERNAL,
+						I40E_PHY_COM_REG_PAGE, true,
+						I40E_PHY_LED_PROV_REG_1,
+						reg_val, NULL);
+	} else {
+		i = rd32(hw, I40E_PFGEN_PORTNUM);
+		port_num = (u8)(i & I40E_PFGEN_PORTNUM_PORT_NUM_MASK);
+		phy_addr = i40e_get_phy_address(hw, port_num);
 		status = i40e_read_phy_register_clause45(hw,
 							 I40E_PHY_COM_REG_PAGE,
 							 led_addr, phy_addr,
-							 &led_reg);
-		if (status)
-			goto phy_blinking_end;
-		led_ctl = led_reg;
-		if (led_reg & I40E_PHY_LED_LINK_MODE_MASK) {
-			led_reg = 0;
-			status = i40e_write_phy_register_clause45(hw,
-							 I40E_PHY_COM_REG_PAGE,
-							 led_addr, phy_addr,
-							 led_reg);
-			if (status)
-				goto phy_blinking_end;
-			break;
-		}
+							 (u16 *)reg_val);
 	}
+	return status;
+}
 
-	if (time > 0 && interval > 0) {
-		for (i = 0; i < time * 1000; i += interval) {
-			status = i40e_read_phy_register_clause45(hw,
-						I40E_PHY_COM_REG_PAGE,
-						led_addr, phy_addr, &led_reg);
-			if (status)
-				goto restore_config;
-			if (led_reg & I40E_PHY_LED_MANUAL_ON)
-				led_reg = 0;
-			else
-				led_reg = I40E_PHY_LED_MANUAL_ON;
-			status = i40e_write_phy_register_clause45(hw,
-						I40E_PHY_COM_REG_PAGE,
-						led_addr, phy_addr, led_reg);
-			if (status)
-				goto restore_config;
-			msleep(interval);
-		}
-	}
+/**
+ * i40e_led_set_reg - write LED register
+ * @hw: pointer to the HW structure
+ * @led_addr: LED register address
+ * @reg_val: register value to write
+ **/
+static int i40e_led_set_reg(struct i40e_hw *hw, u16 led_addr,
+			    u32 reg_val)
+{
+	u8 phy_addr = 0;
+	u8 port_num;
+	int status;
+	u32 i;
 
-restore_config:
-	status = i40e_write_phy_register_clause45(hw,
-						  I40E_PHY_COM_REG_PAGE,
-						  led_addr, phy_addr, led_ctl);
+	if (test_bit(I40E_HW_CAP_AQ_PHY_ACCESS, hw->caps)) {
+		status =
+		       i40e_aq_set_phy_register(hw,
+						I40E_AQ_PHY_REG_ACCESS_EXTERNAL,
+						I40E_PHY_COM_REG_PAGE, true,
+						I40E_PHY_LED_PROV_REG_1,
+						reg_val, NULL);
+	} else {
+		i = rd32(hw, I40E_PFGEN_PORTNUM);
+		port_num = (u8)(i & I40E_PFGEN_PORTNUM_PORT_NUM_MASK);
+		phy_addr = i40e_get_phy_address(hw, port_num);
+		status = i40e_write_phy_register_clause45(hw,
+							  I40E_PHY_COM_REG_PAGE,
+							  led_addr, phy_addr,
+							  (u16)reg_val);
+	}
 
-phy_blinking_end:
 	return status;
 }
 
@@ -4824,17 +4391,29 @@ phy_blinking_end:
  * @val: original value of register to use
  *
  **/
-i40e_status i40e_led_get_phy(struct i40e_hw *hw, u16 *led_addr,
-			     u16 *val)
+int i40e_led_get_phy(struct i40e_hw *hw, u16 *led_addr,
+		     u16 *val)
 {
-	i40e_status status = 0;
 	u16 gpio_led_port;
 	u8 phy_addr = 0;
-	u16 reg_val;
+	u32 reg_val_aq;
+	int status = 0;
 	u16 temp_addr;
+	u16 reg_val;
 	u8 port_num;
 	u32 i;
 
+	if (test_bit(I40E_HW_CAP_AQ_PHY_ACCESS, hw->caps)) {
+		status =
+		      i40e_aq_get_phy_register(hw,
+					       I40E_AQ_PHY_REG_ACCESS_EXTERNAL,
+					       I40E_PHY_COM_REG_PAGE, true,
+					       I40E_PHY_LED_PROV_REG_1,
+					       &reg_val_aq, NULL);
+		if (status == 0)
+			*val = (u16)reg_val_aq;
+		return status;
+	}
 	temp_addr = I40E_PHY_LED_PROV_REG_1;
 	i = rd32(hw, I40E_PFGEN_PORTNUM);
 	port_num = (u8)(i & I40E_PFGEN_PORTNUM_PORT_NUM_MASK);
@@ -4861,59 +4440,48 @@ i40e_status i40e_led_get_phy(struct i40e_hw *hw, u16 *led_addr,
  * i40e_led_set_phy
  * @hw: pointer to the HW structure
  * @on: true or false
+ * @led_addr: address of led register to use
  * @mode: original val plus bit for set or ignore
+ *
  * Set led's on or off when controlled by the PHY
  *
  **/
-i40e_status i40e_led_set_phy(struct i40e_hw *hw, bool on,
-			     u16 led_addr, u32 mode)
+int i40e_led_set_phy(struct i40e_hw *hw, bool on,
+		     u16 led_addr, u32 mode)
 {
-	i40e_status status = 0;
-	u16 led_ctl = 0;
-	u16 led_reg = 0;
-	u8 phy_addr = 0;
-	u8 port_num;
-	u32 i;
+	u32 led_ctl = 0;
+	u32 led_reg = 0;
+	int status = 0;
 
-	i = rd32(hw, I40E_PFGEN_PORTNUM);
-	port_num = (u8)(i & I40E_PFGEN_PORTNUM_PORT_NUM_MASK);
-	phy_addr = i40e_get_phy_address(hw, port_num);
-	status = i40e_read_phy_register_clause45(hw, I40E_PHY_COM_REG_PAGE,
-						 led_addr, phy_addr, &led_reg);
+	status = i40e_led_get_reg(hw, led_addr, &led_reg);
 	if (status)
 		return status;
 	led_ctl = led_reg;
 	if (led_reg & I40E_PHY_LED_LINK_MODE_MASK) {
 		led_reg = 0;
-		status = i40e_write_phy_register_clause45(hw,
-							  I40E_PHY_COM_REG_PAGE,
-							  led_addr, phy_addr,
-							  led_reg);
+		status = i40e_led_set_reg(hw, led_addr, led_reg);
 		if (status)
 			return status;
 	}
-	status = i40e_read_phy_register_clause45(hw, I40E_PHY_COM_REG_PAGE,
-						 led_addr, phy_addr, &led_reg);
+	status = i40e_led_get_reg(hw, led_addr, &led_reg);
 	if (status)
 		goto restore_config;
 	if (on)
 		led_reg = I40E_PHY_LED_MANUAL_ON;
 	else
 		led_reg = 0;
-	status = i40e_write_phy_register_clause45(hw, I40E_PHY_COM_REG_PAGE,
-						  led_addr, phy_addr, led_reg);
+
+	status = i40e_led_set_reg(hw, led_addr, led_reg);
 	if (status)
 		goto restore_config;
 	if (mode & I40E_PHY_LED_MODE_ORIG) {
 		led_ctl = (mode & I40E_PHY_LED_MODE_MASK);
-		status = i40e_write_phy_register_clause45(hw,
-						 I40E_PHY_COM_REG_PAGE,
-						 led_addr, phy_addr, led_ctl);
+		status = i40e_led_set_reg(hw, led_addr, led_ctl);
 	}
 	return status;
+
 restore_config:
-	status = i40e_write_phy_register_clause45(hw, I40E_PHY_COM_REG_PAGE,
-						  led_addr, phy_addr, led_ctl);
+	status = i40e_led_set_reg(hw, led_addr, led_ctl);
 	return status;
 }
 
@@ -4927,20 +4495,20 @@ restore_config:
  * Use the firmware to read the Rx control register,
  * especially useful if the Rx unit is under heavy pressure
  **/
-i40e_status i40e_aq_rx_ctl_read_register(struct i40e_hw *hw,
-				u32 reg_addr, u32 *reg_val,
-				struct i40e_asq_cmd_details *cmd_details)
+int i40e_aq_rx_ctl_read_register(struct i40e_hw *hw,
+				 u32 reg_addr, u32 *reg_val,
+				 struct i40e_asq_cmd_details *cmd_details)
 {
-	struct i40e_aq_desc desc;
-	struct i40e_aqc_rx_ctl_reg_read_write *cmd_resp =
-		(struct i40e_aqc_rx_ctl_reg_read_write *)&desc.params.raw;
-	i40e_status status;
+	struct i40e_aqc_rx_ctl_reg_read_write *cmd_resp;
+	struct libie_aq_desc desc;
+	int status;
 
 	if (!reg_val)
-		return I40E_ERR_PARAM;
+		return -EINVAL;
 
 	i40e_fill_default_direct_cmd_desc(&desc, i40e_aqc_opc_rx_ctl_reg_read);
 
+	cmd_resp = libie_aq_raw(&desc);
 	cmd_resp->address = cpu_to_le32(reg_addr);
 
 	status = i40e_asq_send_command(hw, &desc, NULL, 0, cmd_details);
@@ -4958,18 +4526,18 @@ i40e_status i40e_aq_rx_ctl_read_register(struct i40e_hw *hw,
  **/
 u32 i40e_read_rx_ctl(struct i40e_hw *hw, u32 reg_addr)
 {
-	i40e_status status = 0;
-	bool use_register;
+	bool use_register = false;
+	int status = 0;
 	int retry = 5;
 	u32 val = 0;
 
-	use_register = (((hw->aq.api_maj_ver == 1) &&
-			(hw->aq.api_min_ver < 5)) ||
-			(hw->mac.type == I40E_MAC_X722));
+	if (i40e_is_aq_api_ver_lt(hw, 1, 5) || hw->mac.type == I40E_MAC_X722)
+		use_register = true;
+
 	if (!use_register) {
 do_retry:
 		status = i40e_aq_rx_ctl_read_register(hw, reg_addr, &val, NULL);
-		if (hw->aq.asq_last_status == I40E_AQ_RC_EAGAIN && retry) {
+		if (hw->aq.asq_last_status == LIBIE_AQ_RC_EAGAIN && retry) {
 			usleep_range(1000, 2000);
 			retry--;
 			goto do_retry;
@@ -4993,17 +4561,17 @@ do_retry:
  * Use the firmware to write to an Rx control register,
  * especially useful if the Rx unit is under heavy pressure
  **/
-i40e_status i40e_aq_rx_ctl_write_register(struct i40e_hw *hw,
-				u32 reg_addr, u32 reg_val,
-				struct i40e_asq_cmd_details *cmd_details)
+int i40e_aq_rx_ctl_write_register(struct i40e_hw *hw,
+				  u32 reg_addr, u32 reg_val,
+				  struct i40e_asq_cmd_details *cmd_details)
 {
-	struct i40e_aq_desc desc;
-	struct i40e_aqc_rx_ctl_reg_read_write *cmd =
-		(struct i40e_aqc_rx_ctl_reg_read_write *)&desc.params.raw;
-	i40e_status status;
+	struct i40e_aqc_rx_ctl_reg_read_write *cmd;
+	struct libie_aq_desc desc;
+	int status;
 
 	i40e_fill_default_direct_cmd_desc(&desc, i40e_aqc_opc_rx_ctl_reg_write);
 
+	cmd = libie_aq_raw(&desc);
 	cmd->address = cpu_to_le32(reg_addr);
 	cmd->value = cpu_to_le32(reg_val);
 
@@ -5020,18 +4588,18 @@ i40e_status i40e_aq_rx_ctl_write_register(struct i40e_hw *hw,
  **/
 void i40e_write_rx_ctl(struct i40e_hw *hw, u32 reg_addr, u32 reg_val)
 {
-	i40e_status status = 0;
-	bool use_register;
+	bool use_register = false;
+	int status = 0;
 	int retry = 5;
 
-	use_register = (((hw->aq.api_maj_ver == 1) &&
-			(hw->aq.api_min_ver < 5)) ||
-			(hw->mac.type == I40E_MAC_X722));
+	if (i40e_is_aq_api_ver_lt(hw, 1, 5) || hw->mac.type == I40E_MAC_X722)
+		use_register = true;
+
 	if (!use_register) {
 do_retry:
 		status = i40e_aq_rx_ctl_write_register(hw, reg_addr,
 						       reg_val, NULL);
-		if (hw->aq.asq_last_status == I40E_AQ_RC_EAGAIN && retry) {
+		if (hw->aq.asq_last_status == LIBIE_AQ_RC_EAGAIN && retry) {
 			usleep_range(1000, 2000);
 			retry--;
 			goto do_retry;
@@ -5044,7 +4612,123 @@ do_retry:
 }
 
 /**
- * i40e_aq_write_ppp - Write pipeline personalization profile (ppp)
+ * i40e_mdio_if_number_selection - MDIO I/F number selection
+ * @hw: pointer to the hw struct
+ * @set_mdio: use MDIO I/F number specified by mdio_num
+ * @mdio_num: MDIO I/F number
+ * @cmd: pointer to PHY Register command structure
+ **/
+static void i40e_mdio_if_number_selection(struct i40e_hw *hw, bool set_mdio,
+					  u8 mdio_num,
+					  struct i40e_aqc_phy_register_access *cmd)
+{
+	if (!set_mdio ||
+	    cmd->phy_interface != I40E_AQ_PHY_REG_ACCESS_EXTERNAL)
+		return;
+
+	if (test_bit(I40E_HW_CAP_AQ_PHY_ACCESS_EXTENDED, hw->caps)) {
+		cmd->cmd_flags |=
+			I40E_AQ_PHY_REG_ACCESS_SET_MDIO_IF_NUMBER |
+			FIELD_PREP(I40E_AQ_PHY_REG_ACCESS_MDIO_IF_NUMBER_MASK,
+				   mdio_num);
+	} else {
+		i40e_debug(hw, I40E_DEBUG_PHY, "MDIO I/F number selection not supported by current FW version.\n");
+	}
+}
+
+/**
+ * i40e_aq_set_phy_register_ext
+ * @hw: pointer to the hw struct
+ * @phy_select: select which phy should be accessed
+ * @dev_addr: PHY device address
+ * @page_change: flag to indicate if phy page should be updated
+ * @set_mdio: use MDIO I/F number specified by mdio_num
+ * @mdio_num: MDIO I/F number
+ * @reg_addr: PHY register address
+ * @reg_val: new register value
+ * @cmd_details: pointer to command details structure or NULL
+ *
+ * Write the external PHY register.
+ * NOTE: In common cases MDIO I/F number should not be changed, thats why you
+ * may use simple wrapper i40e_aq_set_phy_register.
+ **/
+int i40e_aq_set_phy_register_ext(struct i40e_hw *hw,
+				 u8 phy_select, u8 dev_addr, bool page_change,
+				 bool set_mdio, u8 mdio_num,
+				 u32 reg_addr, u32 reg_val,
+				 struct i40e_asq_cmd_details *cmd_details)
+{
+	struct i40e_aqc_phy_register_access *cmd;
+	struct libie_aq_desc desc;
+	int status;
+
+	i40e_fill_default_direct_cmd_desc(&desc,
+					  i40e_aqc_opc_set_phy_register);
+
+	cmd = libie_aq_raw(&desc);
+	cmd->phy_interface = phy_select;
+	cmd->dev_address = dev_addr;
+	cmd->reg_address = cpu_to_le32(reg_addr);
+	cmd->reg_value = cpu_to_le32(reg_val);
+
+	i40e_mdio_if_number_selection(hw, set_mdio, mdio_num, cmd);
+
+	if (!page_change)
+		cmd->cmd_flags = I40E_AQ_PHY_REG_ACCESS_DONT_CHANGE_QSFP_PAGE;
+
+	status = i40e_asq_send_command(hw, &desc, NULL, 0, cmd_details);
+
+	return status;
+}
+
+/**
+ * i40e_aq_get_phy_register_ext
+ * @hw: pointer to the hw struct
+ * @phy_select: select which phy should be accessed
+ * @dev_addr: PHY device address
+ * @page_change: flag to indicate if phy page should be updated
+ * @set_mdio: use MDIO I/F number specified by mdio_num
+ * @mdio_num: MDIO I/F number
+ * @reg_addr: PHY register address
+ * @reg_val: read register value
+ * @cmd_details: pointer to command details structure or NULL
+ *
+ * Read the external PHY register.
+ * NOTE: In common cases MDIO I/F number should not be changed, thats why you
+ * may use simple wrapper i40e_aq_get_phy_register.
+ **/
+int i40e_aq_get_phy_register_ext(struct i40e_hw *hw,
+				 u8 phy_select, u8 dev_addr, bool page_change,
+				 bool set_mdio, u8 mdio_num,
+				 u32 reg_addr, u32 *reg_val,
+				 struct i40e_asq_cmd_details *cmd_details)
+{
+	struct i40e_aqc_phy_register_access *cmd;
+	struct libie_aq_desc desc;
+	int status;
+
+	i40e_fill_default_direct_cmd_desc(&desc,
+					  i40e_aqc_opc_get_phy_register);
+
+	cmd = libie_aq_raw(&desc);
+	cmd->phy_interface = phy_select;
+	cmd->dev_address = dev_addr;
+	cmd->reg_address = cpu_to_le32(reg_addr);
+
+	i40e_mdio_if_number_selection(hw, set_mdio, mdio_num, cmd);
+
+	if (!page_change)
+		cmd->cmd_flags = I40E_AQ_PHY_REG_ACCESS_DONT_CHANGE_QSFP_PAGE;
+
+	status = i40e_asq_send_command(hw, &desc, NULL, 0, cmd_details);
+	if (!status)
+		*reg_val = le32_to_cpu(cmd->reg_value);
+
+	return status;
+}
+
+/**
+ * i40e_aq_write_ddp - Write dynamic device personalization (ddp)
  * @hw: pointer to the hw struct
  * @buff: command buffer (size in bytes = buff_size)
  * @buff_size: buffer size in bytes
@@ -5053,25 +4737,23 @@ do_retry:
  * @error_info: returns error information
  * @cmd_details: pointer to command details structure or NULL
  **/
-enum
-i40e_status_code i40e_aq_write_ppp(struct i40e_hw *hw, void *buff,
-				   u16 buff_size, u32 track_id,
-				   u32 *error_offset, u32 *error_info,
-				   struct i40e_asq_cmd_details *cmd_details)
-{
-	struct i40e_aq_desc desc;
-	struct i40e_aqc_write_personalization_profile *cmd =
-		(struct i40e_aqc_write_personalization_profile *)
-		&desc.params.raw;
-	struct i40e_aqc_write_ppp_resp *resp;
-	i40e_status status;
+int i40e_aq_write_ddp(struct i40e_hw *hw, void *buff,
+		      u16 buff_size, u32 track_id,
+		      u32 *error_offset, u32 *error_info,
+		      struct i40e_asq_cmd_details *cmd_details)
+{
+	struct i40e_aqc_write_personalization_profile *cmd;
+	struct i40e_aqc_write_ddp_resp *resp;
+	struct libie_aq_desc desc;
+	int status;
 
 	i40e_fill_default_direct_cmd_desc(&desc,
 					  i40e_aqc_opc_write_personalization_profile);
 
-	desc.flags |= cpu_to_le16(I40E_AQ_FLAG_BUF | I40E_AQ_FLAG_RD);
+	cmd = libie_aq_raw(&desc);
+	desc.flags |= cpu_to_le16(LIBIE_AQ_FLAG_BUF | LIBIE_AQ_FLAG_RD);
 	if (buff_size > I40E_AQ_LARGE_BUF)
-		desc.flags |= cpu_to_le16((u16)I40E_AQ_FLAG_LB);
+		desc.flags |= cpu_to_le16((u16)LIBIE_AQ_FLAG_LB);
 
 	desc.datalen = cpu_to_le16(buff_size);
 
@@ -5079,7 +4761,7 @@ i40e_status_code i40e_aq_write_ppp(struct i40e_hw *hw, void *buff,
 
 	status = i40e_asq_send_command(hw, &desc, buff, buff_size, cmd_details);
 	if (!status) {
-		resp = (struct i40e_aqc_write_ppp_resp *)&desc.params.raw;
+		resp = libie_aq_raw(&desc);
 		if (error_offset)
 			*error_offset = le32_to_cpu(resp->error_offset);
 		if (error_info)
@@ -5090,28 +4772,28 @@ i40e_status_code i40e_aq_write_ppp(struct i40e_hw *hw, void *buff,
 }
 
 /**
- * i40e_aq_get_ppp_list - Read pipeline personalization profile (ppp)
+ * i40e_aq_get_ddp_list - Read dynamic device personalization (ddp)
  * @hw: pointer to the hw struct
  * @buff: command buffer (size in bytes = buff_size)
  * @buff_size: buffer size in bytes
+ * @flags: AdminQ command flags
  * @cmd_details: pointer to command details structure or NULL
  **/
-enum
-i40e_status_code i40e_aq_get_ppp_list(struct i40e_hw *hw, void *buff,
-				      u16 buff_size, u8 flags,
-				      struct i40e_asq_cmd_details *cmd_details)
+int i40e_aq_get_ddp_list(struct i40e_hw *hw, void *buff,
+			 u16 buff_size, u8 flags,
+			 struct i40e_asq_cmd_details *cmd_details)
 {
-	struct i40e_aq_desc desc;
-	struct i40e_aqc_get_applied_profiles *cmd =
-		(struct i40e_aqc_get_applied_profiles *)&desc.params.raw;
-	i40e_status status;
+	struct i40e_aqc_get_applied_profiles *cmd;
+	struct libie_aq_desc desc;
+	int status;
 
 	i40e_fill_default_direct_cmd_desc(&desc,
 					  i40e_aqc_opc_get_personalization_profile_list);
 
-	desc.flags |= cpu_to_le16((u16)I40E_AQ_FLAG_BUF);
+	cmd = libie_aq_raw(&desc);
+	desc.flags |= cpu_to_le16((u16)LIBIE_AQ_FLAG_BUF);
 	if (buff_size > I40E_AQ_LARGE_BUF)
-		desc.flags |= cpu_to_le16((u16)I40E_AQ_FLAG_LB);
+		desc.flags |= cpu_to_le16((u16)LIBIE_AQ_FLAG_LB);
 	desc.datalen = cpu_to_le16(buff_size);
 
 	cmd->flags = flags;
@@ -5150,67 +4832,186 @@ i40e_find_segment_in_package(u32 segment_type,
 	return NULL;
 }
 
+/* Get section table in profile */
+#define I40E_SECTION_TABLE(profile, sec_tbl)				\
+	do {								\
+		struct i40e_profile_segment *p = (profile);		\
+		u32 count;						\
+		u32 *nvm;						\
+		count = p->device_table_count;				\
+		nvm = (u32 *)&p->device_table[count];			\
+		sec_tbl = (struct i40e_section_table *)&nvm[nvm[0] + 1]; \
+	} while (0)
+
+/* Get section header in profile */
+#define I40E_SECTION_HEADER(profile, offset)				\
+	(struct i40e_profile_section_header *)((u8 *)(profile) + (offset))
+
 /**
- * i40e_write_profile
+ * i40e_ddp_exec_aq_section - Execute generic AQ for DDP
+ * @hw: pointer to the hw struct
+ * @aq: command buffer containing all data to execute AQ
+ **/
+static int i40e_ddp_exec_aq_section(struct i40e_hw *hw,
+				    struct i40e_profile_aq_section *aq)
+{
+	struct libie_aq_desc desc;
+	u8 *msg = NULL;
+	u16 msglen;
+	int status;
+
+	i40e_fill_default_direct_cmd_desc(&desc, aq->opcode);
+	desc.flags |= cpu_to_le16(aq->flags);
+	memcpy(desc.params.raw, aq->param, sizeof(desc.params.raw));
+
+	msglen = aq->datalen;
+	if (msglen) {
+		desc.flags |= cpu_to_le16((u16)(LIBIE_AQ_FLAG_BUF |
+						LIBIE_AQ_FLAG_RD));
+		if (msglen > I40E_AQ_LARGE_BUF)
+			desc.flags |= cpu_to_le16((u16)LIBIE_AQ_FLAG_LB);
+		desc.datalen = cpu_to_le16(msglen);
+		msg = &aq->data[0];
+	}
+
+	status = i40e_asq_send_command(hw, &desc, msg, msglen, NULL);
+
+	if (status) {
+		i40e_debug(hw, I40E_DEBUG_PACKAGE,
+			   "unable to exec DDP AQ opcode %u, error %d\n",
+			   aq->opcode, status);
+		return status;
+	}
+
+	/* copy returned desc to aq_buf */
+	memcpy(aq->param, desc.params.raw, sizeof(desc.params.raw));
+
+	return 0;
+}
+
+/**
+ * i40e_validate_profile
  * @hw: pointer to the hardware structure
- * @profile: pointer to the profile segment of the package to be downloaded
+ * @profile: pointer to the profile segment of the package to be validated
  * @track_id: package tracking id
+ * @rollback: flag if the profile is for rollback.
  *
- * Handles the download of a complete package.
+ * Validates supported devices and profile's sections.
  */
-enum i40e_status_code
-i40e_write_profile(struct i40e_hw *hw, struct i40e_profile_segment *profile,
-		   u32 track_id)
+static int
+i40e_validate_profile(struct i40e_hw *hw, struct i40e_profile_segment *profile,
+		      u32 track_id, bool rollback)
 {
-	i40e_status status = 0;
-	struct i40e_section_table *sec_tbl;
 	struct i40e_profile_section_header *sec = NULL;
-	u32 dev_cnt;
+	struct i40e_section_table *sec_tbl;
 	u32 vendor_dev_id;
-	u32 *nvm;
-	u32 section_size = 0;
-	u32 offset = 0, info = 0;
+	int status = 0;
+	u32 dev_cnt;
+	u32 sec_off;
 	u32 i;
 
-	if (!track_id) {
-		i40e_debug(hw, I40E_DEBUG_PACKAGE, "Track_id can't be 0.");
-		return I40E_NOT_SUPPORTED;
+	if (track_id == I40E_DDP_TRACKID_INVALID) {
+		i40e_debug(hw, I40E_DEBUG_PACKAGE, "Invalid track_id\n");
+		return -EOPNOTSUPP;
 	}
 
 	dev_cnt = profile->device_table_count;
-
 	for (i = 0; i < dev_cnt; i++) {
 		vendor_dev_id = profile->device_table[i].vendor_dev_id;
-		if ((vendor_dev_id >> 16) == PCI_VENDOR_ID_INTEL)
-			if (hw->device_id == (vendor_dev_id & 0xFFFF))
-				break;
+		if ((vendor_dev_id >> 16) == PCI_VENDOR_ID_INTEL &&
+		    hw->device_id == (vendor_dev_id & 0xFFFF))
+			break;
 	}
-	if (i == dev_cnt) {
-		i40e_debug(hw, I40E_DEBUG_PACKAGE, "Device doesn't support PPP");
-		return I40E_ERR_DEVICE_NOT_SUPPORTED;
+	if (dev_cnt && i == dev_cnt) {
+		i40e_debug(hw, I40E_DEBUG_PACKAGE,
+			   "Device doesn't support DDP\n");
+		return -ENODEV;
 	}
 
-	nvm = (u32 *)&profile->device_table[dev_cnt];
-	sec_tbl = (struct i40e_section_table *)&nvm[nvm[0] + 1];
+	I40E_SECTION_TABLE(profile, sec_tbl);
 
+	/* Validate sections types */
 	for (i = 0; i < sec_tbl->section_count; i++) {
-		sec = (struct i40e_profile_section_header *)((u8 *)profile +
-					     sec_tbl->section_offset[i]);
+		sec_off = sec_tbl->section_offset[i];
+		sec = I40E_SECTION_HEADER(profile, sec_off);
+		if (rollback) {
+			if (sec->section.type == SECTION_TYPE_MMIO ||
+			    sec->section.type == SECTION_TYPE_AQ ||
+			    sec->section.type == SECTION_TYPE_RB_AQ) {
+				i40e_debug(hw, I40E_DEBUG_PACKAGE,
+					   "Not a roll-back package\n");
+				return -EOPNOTSUPP;
+			}
+		} else {
+			if (sec->section.type == SECTION_TYPE_RB_AQ ||
+			    sec->section.type == SECTION_TYPE_RB_MMIO) {
+				i40e_debug(hw, I40E_DEBUG_PACKAGE,
+					   "Not an original package\n");
+				return -EOPNOTSUPP;
+			}
+		}
+	}
+
+	return status;
+}
 
-		/* Skip 'AQ', 'note' and 'name' sections */
+/**
+ * i40e_write_profile
+ * @hw: pointer to the hardware structure
+ * @profile: pointer to the profile segment of the package to be downloaded
+ * @track_id: package tracking id
+ *
+ * Handles the download of a complete package.
+ */
+int
+i40e_write_profile(struct i40e_hw *hw, struct i40e_profile_segment *profile,
+		   u32 track_id)
+{
+	struct i40e_profile_section_header *sec = NULL;
+	struct i40e_profile_aq_section *ddp_aq;
+	struct i40e_section_table *sec_tbl;
+	u32 offset = 0, info = 0;
+	u32 section_size = 0;
+	int status = 0;
+	u32 sec_off;
+	u32 i;
+
+	status = i40e_validate_profile(hw, profile, track_id, false);
+	if (status)
+		return status;
+
+	I40E_SECTION_TABLE(profile, sec_tbl);
+
+	for (i = 0; i < sec_tbl->section_count; i++) {
+		sec_off = sec_tbl->section_offset[i];
+		sec = I40E_SECTION_HEADER(profile, sec_off);
+		/* Process generic admin command */
+		if (sec->section.type == SECTION_TYPE_AQ) {
+			ddp_aq = (struct i40e_profile_aq_section *)&sec[1];
+			status = i40e_ddp_exec_aq_section(hw, ddp_aq);
+			if (status) {
+				i40e_debug(hw, I40E_DEBUG_PACKAGE,
+					   "Failed to execute aq: section %d, opcode %u\n",
+					   i, ddp_aq->opcode);
+				break;
+			}
+			sec->section.type = SECTION_TYPE_RB_AQ;
+		}
+
+		/* Skip any non-mmio sections */
 		if (sec->section.type != SECTION_TYPE_MMIO)
 			continue;
 
 		section_size = sec->section.size +
 			sizeof(struct i40e_profile_section_header);
 
-		/* Write profile */
-		status = i40e_aq_write_ppp(hw, (void *)sec, (u16)section_size,
+		/* Write MMIO section */
+		status = i40e_aq_write_ddp(hw, (void *)sec, (u16)section_size,
 					   track_id, &offset, &info, NULL);
 		if (status) {
 			i40e_debug(hw, I40E_DEBUG_PACKAGE,
-				   "Failed to write profile: offset %d, info %d",
-				   offset, info);
+				   "Failed to write profile: section %d, offset %d, info %d\n",
+				   i, offset, info);
 			break;
 		}
 	}
@@ -5218,39 +5019,238 @@ i40e_write_profile(struct i40e_hw *hw, struct i40e_profile_segment *profile,
 }
 
 /**
- * i40e_add_pinfo_to_list
+ * i40e_rollback_profile
  * @hw: pointer to the hardware structure
- * @profile: pointer to the profile segment of the package
- * @profile_info_sec: buffer for information section
+ * @profile: pointer to the profile segment of the package to be removed
  * @track_id: package tracking id
  *
- * Register a profile to the list of loaded profiles.
+ * Rolls back previously loaded package.
  */
-enum i40e_status_code
-i40e_add_pinfo_to_list(struct i40e_hw *hw,
-		       struct i40e_profile_segment *profile,
-		       u8 *profile_info_sec, u32 track_id)
+int
+i40e_rollback_profile(struct i40e_hw *hw, struct i40e_profile_segment *profile,
+		      u32 track_id)
 {
-	i40e_status status = 0;
 	struct i40e_profile_section_header *sec = NULL;
-	struct i40e_profile_info *pinfo;
+	struct i40e_section_table *sec_tbl;
 	u32 offset = 0, info = 0;
+	u32 section_size = 0;
+	int status = 0;
+	u32 sec_off;
+	int i;
+
+	status = i40e_validate_profile(hw, profile, track_id, true);
+	if (status)
+		return status;
+
+	I40E_SECTION_TABLE(profile, sec_tbl);
+
+	/* For rollback write sections in reverse */
+	for (i = sec_tbl->section_count - 1; i >= 0; i--) {
+		sec_off = sec_tbl->section_offset[i];
+		sec = I40E_SECTION_HEADER(profile, sec_off);
+
+		/* Skip any non-rollback sections */
+		if (sec->section.type != SECTION_TYPE_RB_MMIO)
+			continue;
+
+		section_size = sec->section.size +
+			sizeof(struct i40e_profile_section_header);
+
+		/* Write roll-back MMIO section */
+		status = i40e_aq_write_ddp(hw, (void *)sec, (u16)section_size,
+					   track_id, &offset, &info, NULL);
+		if (status) {
+			i40e_debug(hw, I40E_DEBUG_PACKAGE,
+				   "Failed to write profile: section %d, offset %d, info %d\n",
+				   i, offset, info);
+			break;
+		}
+	}
+	return status;
+}
+
+/**
+ * i40e_aq_add_cloud_filters
+ * @hw: pointer to the hardware structure
+ * @seid: VSI seid to add cloud filters from
+ * @filters: Buffer which contains the filters to be added
+ * @filter_count: number of filters contained in the buffer
+ *
+ * Set the cloud filters for a given VSI.  The contents of the
+ * i40e_aqc_cloud_filters_element_data are filled in by the caller
+ * of the function.
+ *
+ **/
+int
+i40e_aq_add_cloud_filters(struct i40e_hw *hw, u16 seid,
+			  struct i40e_aqc_cloud_filters_element_data *filters,
+			  u8 filter_count)
+{
+	struct i40e_aqc_add_remove_cloud_filters *cmd;
+	struct libie_aq_desc desc;
+	u16 buff_len;
+	int status;
+
+	i40e_fill_default_direct_cmd_desc(&desc,
+					  i40e_aqc_opc_add_cloud_filters);
+
+	cmd = libie_aq_raw(&desc);
+	buff_len = filter_count * sizeof(*filters);
+	desc.datalen = cpu_to_le16(buff_len);
+	desc.flags |= cpu_to_le16((u16)(LIBIE_AQ_FLAG_BUF | LIBIE_AQ_FLAG_RD));
+	cmd->num_filters = filter_count;
+	cmd->seid = cpu_to_le16(seid);
+
+	status = i40e_asq_send_command(hw, &desc, filters, buff_len, NULL);
+
+	return status;
+}
+
+/**
+ * i40e_aq_add_cloud_filters_bb
+ * @hw: pointer to the hardware structure
+ * @seid: VSI seid to add cloud filters from
+ * @filters: Buffer which contains the filters in big buffer to be added
+ * @filter_count: number of filters contained in the buffer
+ *
+ * Set the big buffer cloud filters for a given VSI.  The contents of the
+ * i40e_aqc_cloud_filters_element_bb are filled in by the caller of the
+ * function.
+ *
+ **/
+int
+i40e_aq_add_cloud_filters_bb(struct i40e_hw *hw, u16 seid,
+			     struct i40e_aqc_cloud_filters_element_bb *filters,
+			     u8 filter_count)
+{
+	struct i40e_aqc_add_remove_cloud_filters *cmd;
+	struct libie_aq_desc desc;
+	u16 buff_len;
+	int status;
+	int i;
+
+	i40e_fill_default_direct_cmd_desc(&desc,
+					  i40e_aqc_opc_add_cloud_filters);
+
+	cmd = libie_aq_raw(&desc);
+	buff_len = filter_count * sizeof(*filters);
+	desc.datalen = cpu_to_le16(buff_len);
+	desc.flags |= cpu_to_le16((u16)(LIBIE_AQ_FLAG_BUF | LIBIE_AQ_FLAG_RD));
+	cmd->num_filters = filter_count;
+	cmd->seid = cpu_to_le16(seid);
+	cmd->big_buffer_flag = I40E_AQC_ADD_CLOUD_CMD_BB;
+
+	for (i = 0; i < filter_count; i++) {
+		u16 tnl_type;
+		u32 ti;
+
+		tnl_type = le16_get_bits(filters[i].element.flags,
+					 I40E_AQC_ADD_CLOUD_TNL_TYPE_MASK);
+
+		/* Due to hardware eccentricities, the VNI for Geneve is shifted
+		 * one more byte further than normally used for Tenant ID in
+		 * other tunnel types.
+		 */
+		if (tnl_type == I40E_AQC_ADD_CLOUD_TNL_TYPE_GENEVE) {
+			ti = le32_to_cpu(filters[i].element.tenant_id);
+			filters[i].element.tenant_id = cpu_to_le32(ti << 8);
+		}
+	}
+
+	status = i40e_asq_send_command(hw, &desc, filters, buff_len, NULL);
+
+	return status;
+}
+
+/**
+ * i40e_aq_rem_cloud_filters
+ * @hw: pointer to the hardware structure
+ * @seid: VSI seid to remove cloud filters from
+ * @filters: Buffer which contains the filters to be removed
+ * @filter_count: number of filters contained in the buffer
+ *
+ * Remove the cloud filters for a given VSI.  The contents of the
+ * i40e_aqc_cloud_filters_element_data are filled in by the caller
+ * of the function.
+ *
+ **/
+int
+i40e_aq_rem_cloud_filters(struct i40e_hw *hw, u16 seid,
+			  struct i40e_aqc_cloud_filters_element_data *filters,
+			  u8 filter_count)
+{
+	struct i40e_aqc_add_remove_cloud_filters *cmd;
+	struct libie_aq_desc desc;
+	u16 buff_len;
+	int status;
+
+	i40e_fill_default_direct_cmd_desc(&desc,
+					  i40e_aqc_opc_remove_cloud_filters);
+
+	cmd = libie_aq_raw(&desc);
+	buff_len = filter_count * sizeof(*filters);
+	desc.datalen = cpu_to_le16(buff_len);
+	desc.flags |= cpu_to_le16((u16)(LIBIE_AQ_FLAG_BUF | LIBIE_AQ_FLAG_RD));
+	cmd->num_filters = filter_count;
+	cmd->seid = cpu_to_le16(seid);
+
+	status = i40e_asq_send_command(hw, &desc, filters, buff_len, NULL);
+
+	return status;
+}
+
+/**
+ * i40e_aq_rem_cloud_filters_bb
+ * @hw: pointer to the hardware structure
+ * @seid: VSI seid to remove cloud filters from
+ * @filters: Buffer which contains the filters in big buffer to be removed
+ * @filter_count: number of filters contained in the buffer
+ *
+ * Remove the big buffer cloud filters for a given VSI.  The contents of the
+ * i40e_aqc_cloud_filters_element_bb are filled in by the caller of the
+ * function.
+ *
+ **/
+int
+i40e_aq_rem_cloud_filters_bb(struct i40e_hw *hw, u16 seid,
+			     struct i40e_aqc_cloud_filters_element_bb *filters,
+			     u8 filter_count)
+{
+	struct i40e_aqc_add_remove_cloud_filters *cmd;
+	struct libie_aq_desc desc;
+	u16 buff_len;
+	int status;
+	int i;
+
+	i40e_fill_default_direct_cmd_desc(&desc,
+					  i40e_aqc_opc_remove_cloud_filters);
+
+	cmd = libie_aq_raw(&desc);
+	buff_len = filter_count * sizeof(*filters);
+	desc.datalen = cpu_to_le16(buff_len);
+	desc.flags |= cpu_to_le16((u16)(LIBIE_AQ_FLAG_BUF | LIBIE_AQ_FLAG_RD));
+	cmd->num_filters = filter_count;
+	cmd->seid = cpu_to_le16(seid);
+	cmd->big_buffer_flag = I40E_AQC_ADD_CLOUD_CMD_BB;
+
+	for (i = 0; i < filter_count; i++) {
+		u16 tnl_type;
+		u32 ti;
+
+		tnl_type = le16_get_bits(filters[i].element.flags,
+					 I40E_AQC_ADD_CLOUD_TNL_TYPE_MASK);
+
+		/* Due to hardware eccentricities, the VNI for Geneve is shifted
+		 * one more byte further than normally used for Tenant ID in
+		 * other tunnel types.
+		 */
+		if (tnl_type == I40E_AQC_ADD_CLOUD_TNL_TYPE_GENEVE) {
+			ti = le32_to_cpu(filters[i].element.tenant_id);
+			filters[i].element.tenant_id = cpu_to_le32(ti << 8);
+		}
+	}
+
+	status = i40e_asq_send_command(hw, &desc, filters, buff_len, NULL);
 
-	sec = (struct i40e_profile_section_header *)profile_info_sec;
-	sec->tbl_size = 1;
-	sec->data_end = sizeof(struct i40e_profile_section_header) +
-			sizeof(struct i40e_profile_info);
-	sec->section.type = SECTION_TYPE_INFO;
-	sec->section.offset = sizeof(struct i40e_profile_section_header);
-	sec->section.size = sizeof(struct i40e_profile_info);
-	pinfo = (struct i40e_profile_info *)(profile_info_sec +
-					     sec->section.offset);
-	pinfo->track_id = track_id;
-	pinfo->version = profile->version;
-	pinfo->op = I40E_PPP_ADD_TRACKID;
-	memcpy(pinfo->name, profile->name, I40E_PPP_NAME_SIZE);
-
-	status = i40e_aq_write_ppp(hw, (void *)sec, sec->data_end,
-				   track_id, &offset, &info, NULL);
 	return status;
 }
diff --git a/drivers/net/ethernet/intel/i40e/i40e_dcb.c b/drivers/net/ethernet/intel/i40e/i40e_dcb.c
index 0fab3a9b51d9..9e0c9597aeb9 100644
--- a/drivers/net/ethernet/intel/i40e/i40e_dcb.c
+++ b/drivers/net/ethernet/intel/i40e/i40e_dcb.c
@@ -1,32 +1,11 @@
-/*******************************************************************************
- *
- * Intel Ethernet Controller XL710 Family Linux Driver
- * Copyright(c) 2013 - 2014 Intel Corporation.
- *
- * This program is free software; you can redistribute it and/or modify it
- * under the terms and conditions of the GNU General Public License,
- * version 2, as published by the Free Software Foundation.
- *
- * This program is distributed in the hope it will be useful, but WITHOUT
- * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
- * FITNESS FOR A PARTICULAR PURPOSE.  See the GNU General Public License for
- * more details.
- *
- * You should have received a copy of the GNU General Public License along
- * with this program.  If not, see <http://www.gnu.org/licenses/>.
- *
- * The full GNU General Public License is included in this distribution in
- * the file called "COPYING".
- *
- * Contact Information:
- * e1000-devel Mailing List <e1000-devel@lists.sourceforge.net>
- * Intel Corporation, 5200 N.E. Elam Young Parkway, Hillsboro, OR 97124-6497
- *
- ******************************************************************************/
+// SPDX-License-Identifier: GPL-2.0
+/* Copyright(c) 2013 - 2021 Intel Corporation. */
 
+#include <linux/bitfield.h>
 #include "i40e_adminq.h"
-#include "i40e_prototype.h"
+#include "i40e_alloc.h"
 #include "i40e_dcb.h"
+#include "i40e_prototype.h"
 
 /**
  * i40e_get_dcbx_status
@@ -35,16 +14,15 @@
  *
  * Get the DCBX status from the Firmware
  **/
-i40e_status i40e_get_dcbx_status(struct i40e_hw *hw, u16 *status)
+int i40e_get_dcbx_status(struct i40e_hw *hw, u16 *status)
 {
 	u32 reg;
 
 	if (!status)
-		return I40E_ERR_PARAM;
+		return -EINVAL;
 
 	reg = rd32(hw, I40E_PRTDCB_GENS);
-	*status = (u16)((reg & I40E_PRTDCB_GENS_DCBX_STATUS_MASK) >>
-			I40E_PRTDCB_GENS_DCBX_STATUS_SHIFT);
+	*status = FIELD_GET(I40E_PRTDCB_GENS_DCBX_STATUS_MASK, reg);
 
 	return 0;
 }
@@ -73,12 +51,9 @@ static void i40e_parse_ieee_etscfg_tlv(struct i40e_lldp_org_tlv *tlv,
 	 * |1bit | 1bit|3 bits|3bits|
 	 */
 	etscfg = &dcbcfg->etscfg;
-	etscfg->willing = (u8)((buf[offset] & I40E_IEEE_ETS_WILLING_MASK) >>
-			       I40E_IEEE_ETS_WILLING_SHIFT);
-	etscfg->cbs = (u8)((buf[offset] & I40E_IEEE_ETS_CBS_MASK) >>
-			   I40E_IEEE_ETS_CBS_SHIFT);
-	etscfg->maxtcs = (u8)((buf[offset] & I40E_IEEE_ETS_MAXTC_MASK) >>
-			      I40E_IEEE_ETS_MAXTC_SHIFT);
+	etscfg->willing = FIELD_GET(I40E_IEEE_ETS_WILLING_MASK, buf[offset]);
+	etscfg->cbs = FIELD_GET(I40E_IEEE_ETS_CBS_MASK, buf[offset]);
+	etscfg->maxtcs = FIELD_GET(I40E_IEEE_ETS_MAXTC_MASK, buf[offset]);
 
 	/* Move offset to Priority Assignment Table */
 	offset++;
@@ -92,11 +67,9 @@ static void i40e_parse_ieee_etscfg_tlv(struct i40e_lldp_org_tlv *tlv,
 	 *        -----------------------------------------
 	 */
 	for (i = 0; i < 4; i++) {
-		priority = (u8)((buf[offset] & I40E_IEEE_ETS_PRIO_1_MASK) >>
-				I40E_IEEE_ETS_PRIO_1_SHIFT);
-		etscfg->prioritytable[i * 2] =  priority;
-		priority = (u8)((buf[offset] & I40E_IEEE_ETS_PRIO_0_MASK) >>
-				I40E_IEEE_ETS_PRIO_0_SHIFT);
+		priority = FIELD_GET(I40E_IEEE_ETS_PRIO_1_MASK, buf[offset]);
+		etscfg->prioritytable[i * 2] = priority;
+		priority = FIELD_GET(I40E_IEEE_ETS_PRIO_0_MASK, buf[offset]);
 		etscfg->prioritytable[i * 2 + 1] = priority;
 		offset++;
 	}
@@ -147,12 +120,10 @@ static void i40e_parse_ieee_etsrec_tlv(struct i40e_lldp_org_tlv *tlv,
 	 *        -----------------------------------------
 	 */
 	for (i = 0; i < 4; i++) {
-		priority = (u8)((buf[offset] & I40E_IEEE_ETS_PRIO_1_MASK) >>
-				I40E_IEEE_ETS_PRIO_1_SHIFT);
-		dcbcfg->etsrec.prioritytable[i*2] =  priority;
-		priority = (u8)((buf[offset] & I40E_IEEE_ETS_PRIO_0_MASK) >>
-				I40E_IEEE_ETS_PRIO_0_SHIFT);
-		dcbcfg->etsrec.prioritytable[i*2 + 1] = priority;
+		priority = FIELD_GET(I40E_IEEE_ETS_PRIO_1_MASK, buf[offset]);
+		dcbcfg->etsrec.prioritytable[i * 2] = priority;
+		priority = FIELD_GET(I40E_IEEE_ETS_PRIO_0_MASK, buf[offset]);
+		dcbcfg->etsrec.prioritytable[(i * 2) + 1] = priority;
 		offset++;
 	}
 
@@ -193,12 +164,9 @@ static void i40e_parse_ieee_pfccfg_tlv(struct i40e_lldp_org_tlv *tlv,
 	 * -----------------------------------------
 	 * |1bit | 1bit|2 bits|4bits| 1 octet      |
 	 */
-	dcbcfg->pfc.willing = (u8)((buf[0] & I40E_IEEE_PFC_WILLING_MASK) >>
-				   I40E_IEEE_PFC_WILLING_SHIFT);
-	dcbcfg->pfc.mbc = (u8)((buf[0] & I40E_IEEE_PFC_MBC_MASK) >>
-			       I40E_IEEE_PFC_MBC_SHIFT);
-	dcbcfg->pfc.pfccap = (u8)((buf[0] & I40E_IEEE_PFC_CAP_MASK) >>
-				  I40E_IEEE_PFC_CAP_SHIFT);
+	dcbcfg->pfc.willing = FIELD_GET(I40E_IEEE_PFC_WILLING_MASK, buf[0]);
+	dcbcfg->pfc.mbc = FIELD_GET(I40E_IEEE_PFC_MBC_MASK, buf[0]);
+	dcbcfg->pfc.pfccap = FIELD_GET(I40E_IEEE_PFC_CAP_MASK, buf[0]);
 	dcbcfg->pfc.pfcenable = buf[1];
 }
 
@@ -219,8 +187,7 @@ static void i40e_parse_ieee_app_tlv(struct i40e_lldp_org_tlv *tlv,
 	u8 *buf;
 
 	typelength = ntohs(tlv->typelength);
-	length = (u16)((typelength & I40E_LLDP_TLV_LEN_MASK) >>
-		       I40E_LLDP_TLV_LEN_SHIFT);
+	length = FIELD_GET(I40E_LLDP_TLV_LEN_MASK, typelength);
 	buf = tlv->tlvinfo;
 
 	/* The App priority table starts 5 octets after TLV header */
@@ -238,12 +205,10 @@ static void i40e_parse_ieee_app_tlv(struct i40e_lldp_org_tlv *tlv,
 	 *        -----------------------------------------
 	 */
 	while (offset < length) {
-		dcbcfg->app[i].priority = (u8)((buf[offset] &
-						I40E_IEEE_APP_PRIO_MASK) >>
-					       I40E_IEEE_APP_PRIO_SHIFT);
-		dcbcfg->app[i].selector = (u8)((buf[offset] &
-						I40E_IEEE_APP_SEL_MASK) >>
-					       I40E_IEEE_APP_SEL_SHIFT);
+		dcbcfg->app[i].priority = FIELD_GET(I40E_IEEE_APP_PRIO_MASK,
+						    buf[offset]);
+		dcbcfg->app[i].selector = FIELD_GET(I40E_IEEE_APP_SEL_MASK,
+						    buf[offset]);
 		dcbcfg->app[i].protocolid = (buf[offset + 1] << 0x8) |
 					     buf[offset + 2];
 		/* Move to next app */
@@ -257,7 +222,7 @@ static void i40e_parse_ieee_app_tlv(struct i40e_lldp_org_tlv *tlv,
 }
 
 /**
- * i40e_parse_ieee_etsrec_tlv
+ * i40e_parse_ieee_tlv
  * @tlv: IEEE 802.1Qaz TLV
  * @dcbcfg: Local store to update ETS REC data
  *
@@ -271,8 +236,7 @@ static void i40e_parse_ieee_tlv(struct i40e_lldp_org_tlv *tlv,
 	u8 subtype;
 
 	ouisubtype = ntohl(tlv->ouisubtype);
-	subtype = (u8)((ouisubtype & I40E_LLDP_TLV_SUBTYPE_MASK) >>
-		       I40E_LLDP_TLV_SUBTYPE_SHIFT);
+	subtype = FIELD_GET(I40E_LLDP_TLV_SUBTYPE_MASK, ouisubtype);
 	switch (subtype) {
 	case I40E_IEEE_SUBTYPE_ETS_CFG:
 		i40e_parse_ieee_etscfg_tlv(tlv, dcbcfg);
@@ -322,11 +286,9 @@ static void i40e_parse_cee_pgcfg_tlv(struct i40e_cee_feat_tlv *tlv,
 	 *        -----------------------------------------
 	 */
 	for (i = 0; i < 4; i++) {
-		priority = (u8)((buf[offset] & I40E_CEE_PGID_PRIO_1_MASK) >>
-				 I40E_CEE_PGID_PRIO_1_SHIFT);
-		etscfg->prioritytable[i * 2] =  priority;
-		priority = (u8)((buf[offset] & I40E_CEE_PGID_PRIO_0_MASK) >>
-				 I40E_CEE_PGID_PRIO_0_SHIFT);
+		priority = FIELD_GET(I40E_CEE_PGID_PRIO_1_MASK, buf[offset]);
+		etscfg->prioritytable[i * 2] = priority;
+		priority = FIELD_GET(I40E_CEE_PGID_PRIO_0_MASK, buf[offset]);
 		etscfg->prioritytable[i * 2 + 1] = priority;
 		offset++;
 	}
@@ -383,13 +345,14 @@ static void i40e_parse_cee_app_tlv(struct i40e_cee_feat_tlv *tlv,
 	u8 i;
 
 	typelength = ntohs(tlv->hdr.typelen);
-	length = (u16)((typelength & I40E_LLDP_TLV_LEN_MASK) >>
-		       I40E_LLDP_TLV_LEN_SHIFT);
+	length = FIELD_GET(I40E_LLDP_TLV_LEN_MASK, typelength);
 
 	dcbcfg->numapps = length / sizeof(*app);
 
 	if (!dcbcfg->numapps)
 		return;
+	if (dcbcfg->numapps > I40E_DCBX_MAX_APPS)
+		dcbcfg->numapps = I40E_DCBX_MAX_APPS;
 
 	for (i = 0; i < dcbcfg->numapps; i++) {
 		u8 up, selector;
@@ -438,15 +401,13 @@ static void i40e_parse_cee_tlv(struct i40e_lldp_org_tlv *tlv,
 	u32 ouisubtype;
 
 	ouisubtype = ntohl(tlv->ouisubtype);
-	subtype = (u8)((ouisubtype & I40E_LLDP_TLV_SUBTYPE_MASK) >>
-		       I40E_LLDP_TLV_SUBTYPE_SHIFT);
+	subtype = FIELD_GET(I40E_LLDP_TLV_SUBTYPE_MASK, ouisubtype);
 	/* Return if not CEE DCBX */
 	if (subtype != I40E_CEE_DCBX_TYPE)
 		return;
 
 	typelength = ntohs(tlv->typelength);
-	tlvlen = (u16)((typelength & I40E_LLDP_TLV_LEN_MASK) >>
-			I40E_LLDP_TLV_LEN_SHIFT);
+	tlvlen = FIELD_GET(I40E_LLDP_TLV_LEN_MASK, typelength);
 	len = sizeof(tlv->typelength) + sizeof(ouisubtype) +
 	      sizeof(struct i40e_cee_ctrl_tlv);
 	/* Return if no CEE DCBX Feature TLVs */
@@ -456,11 +417,8 @@ static void i40e_parse_cee_tlv(struct i40e_lldp_org_tlv *tlv,
 	sub_tlv = (struct i40e_cee_feat_tlv *)((char *)tlv + len);
 	while (feat_tlv_count < I40E_CEE_MAX_FEAT_TYPE) {
 		typelength = ntohs(sub_tlv->hdr.typelen);
-		sublen = (u16)((typelength &
-				I40E_LLDP_TLV_LEN_MASK) >>
-				I40E_LLDP_TLV_LEN_SHIFT);
-		subtype = (u8)((typelength & I40E_LLDP_TLV_TYPE_MASK) >>
-				I40E_LLDP_TLV_TYPE_SHIFT);
+		sublen = FIELD_GET(I40E_LLDP_TLV_LEN_MASK, typelength);
+		subtype = FIELD_GET(I40E_LLDP_TLV_TYPE_MASK, typelength);
 		switch (subtype) {
 		case I40E_CEE_SUBTYPE_PG_CFG:
 			i40e_parse_cee_pgcfg_tlv(sub_tlv, dcbcfg);
@@ -497,8 +455,7 @@ static void i40e_parse_org_tlv(struct i40e_lldp_org_tlv *tlv,
 	u32 oui;
 
 	ouisubtype = ntohl(tlv->ouisubtype);
-	oui = (u32)((ouisubtype & I40E_LLDP_TLV_OUI_MASK) >>
-		    I40E_LLDP_TLV_OUI_SHIFT);
+	oui = FIELD_GET(I40E_LLDP_TLV_OUI_MASK, ouisubtype);
 	switch (oui) {
 	case I40E_IEEE_8021QAZ_OUI:
 		i40e_parse_ieee_tlv(tlv, dcbcfg);
@@ -518,28 +475,26 @@ static void i40e_parse_org_tlv(struct i40e_lldp_org_tlv *tlv,
  *
  * Parse DCB configuration from the LLDPDU
  **/
-i40e_status i40e_lldp_to_dcb_config(u8 *lldpmib,
-				    struct i40e_dcbx_config *dcbcfg)
+int i40e_lldp_to_dcb_config(u8 *lldpmib,
+			    struct i40e_dcbx_config *dcbcfg)
 {
-	i40e_status ret = 0;
 	struct i40e_lldp_org_tlv *tlv;
-	u16 type;
-	u16 length;
 	u16 typelength;
 	u16 offset = 0;
+	int ret = 0;
+	u16 length;
+	u16 type;
 
 	if (!lldpmib || !dcbcfg)
-		return I40E_ERR_PARAM;
+		return -EINVAL;
 
 	/* set to the start of LLDPDU */
 	lldpmib += ETH_HLEN;
 	tlv = (struct i40e_lldp_org_tlv *)lldpmib;
 	while (1) {
 		typelength = ntohs(tlv->typelength);
-		type = (u16)((typelength & I40E_LLDP_TLV_TYPE_MASK) >>
-			     I40E_LLDP_TLV_TYPE_SHIFT);
-		length = (u16)((typelength & I40E_LLDP_TLV_LEN_MASK) >>
-			       I40E_LLDP_TLV_LEN_SHIFT);
+		type = FIELD_GET(I40E_LLDP_TLV_TYPE_MASK, typelength);
+		length = FIELD_GET(I40E_LLDP_TLV_LEN_MASK, typelength);
 		offset += sizeof(typelength) + length;
 
 		/* END TLV or beyond LLDPDU size */
@@ -572,12 +527,12 @@ i40e_status i40e_lldp_to_dcb_config(u8 *lldpmib,
  *
  * Query DCB configuration from the Firmware
  **/
-i40e_status i40e_aq_get_dcb_config(struct i40e_hw *hw, u8 mib_type,
-				   u8 bridgetype,
-				   struct i40e_dcbx_config *dcbcfg)
+int i40e_aq_get_dcb_config(struct i40e_hw *hw, u8 mib_type,
+			   u8 bridgetype,
+			   struct i40e_dcbx_config *dcbcfg)
 {
-	i40e_status ret = 0;
 	struct i40e_virt_mem mem;
+	int ret = 0;
 	u8 *lldpmib;
 
 	/* Allocate the LLDPDU */
@@ -613,19 +568,22 @@ static void i40e_cee_to_dcb_v1_config(
 {
 	u16 status, tlv_status = le16_to_cpu(cee_cfg->tlv_status);
 	u16 app_prio = le16_to_cpu(cee_cfg->oper_app_prio);
-	u8 i, tc, err;
+	u8 i, err;
 
 	/* CEE PG data to ETS config */
 	dcbcfg->etscfg.maxtcs = cee_cfg->oper_num_tc;
 
+	/* Note that the FW creates the oper_prio_tc nibbles reversed
+	 * from those in the CEE Priority Group sub-TLV.
+	 */
 	for (i = 0; i < 4; i++) {
-		tc = (u8)((cee_cfg->oper_prio_tc[i] &
-			 I40E_CEE_PGID_PRIO_1_MASK) >>
-			 I40E_CEE_PGID_PRIO_1_SHIFT);
-		dcbcfg->etscfg.prioritytable[i*2] =  tc;
-		tc = (u8)((cee_cfg->oper_prio_tc[i] &
-			 I40E_CEE_PGID_PRIO_0_MASK) >>
-			 I40E_CEE_PGID_PRIO_0_SHIFT);
+		u8 tc;
+
+		tc = FIELD_GET(I40E_CEE_PGID_PRIO_0_MASK,
+			       cee_cfg->oper_prio_tc[i]);
+		dcbcfg->etscfg.prioritytable[i * 2] = tc;
+		tc = FIELD_GET(I40E_CEE_PGID_PRIO_1_MASK,
+			       cee_cfg->oper_prio_tc[i]);
 		dcbcfg->etscfg.prioritytable[i*2 + 1] = tc;
 	}
 
@@ -647,8 +605,7 @@ static void i40e_cee_to_dcb_v1_config(
 	dcbcfg->pfc.pfcenable = cee_cfg->oper_pfc_en;
 	dcbcfg->pfc.pfccap = I40E_MAX_TRAFFIC_CLASS;
 
-	status = (tlv_status & I40E_AQC_CEE_APP_STATUS_MASK) >>
-		  I40E_AQC_CEE_APP_STATUS_SHIFT;
+	status = FIELD_GET(I40E_AQC_CEE_APP_STATUS_MASK, tlv_status);
 	err = (status & I40E_TLV_STATUS_ERR) ? 1 : 0;
 	/* Add APPs if Error is False */
 	if (!err) {
@@ -657,22 +614,19 @@ static void i40e_cee_to_dcb_v1_config(
 
 		/* FCoE APP */
 		dcbcfg->app[0].priority =
-			(app_prio & I40E_AQC_CEE_APP_FCOE_MASK) >>
-			 I40E_AQC_CEE_APP_FCOE_SHIFT;
+			FIELD_GET(I40E_AQC_CEE_APP_FCOE_MASK, app_prio);
 		dcbcfg->app[0].selector = I40E_APP_SEL_ETHTYPE;
 		dcbcfg->app[0].protocolid = I40E_APP_PROTOID_FCOE;
 
 		/* iSCSI APP */
 		dcbcfg->app[1].priority =
-			(app_prio & I40E_AQC_CEE_APP_ISCSI_MASK) >>
-			 I40E_AQC_CEE_APP_ISCSI_SHIFT;
+			FIELD_GET(I40E_AQC_CEE_APP_ISCSI_MASK, app_prio);
 		dcbcfg->app[1].selector = I40E_APP_SEL_TCPIP;
 		dcbcfg->app[1].protocolid = I40E_APP_PROTOID_ISCSI;
 
 		/* FIP APP */
 		dcbcfg->app[2].priority =
-			(app_prio & I40E_AQC_CEE_APP_FIP_MASK) >>
-			 I40E_AQC_CEE_APP_FIP_SHIFT;
+			FIELD_GET(I40E_AQC_CEE_APP_FIP_MASK, app_prio);
 		dcbcfg->app[2].selector = I40E_APP_SEL_ETHTYPE;
 		dcbcfg->app[2].protocolid = I40E_APP_PROTOID_FIP;
 	}
@@ -691,7 +645,7 @@ static void i40e_cee_to_dcb_config(
 {
 	u32 status, tlv_status = le32_to_cpu(cee_cfg->tlv_status);
 	u16 app_prio = le16_to_cpu(cee_cfg->oper_app_prio);
-	u8 i, tc, err, sync, oper;
+	u8 i, err, sync, oper;
 
 	/* CEE PG data to ETS config */
 	dcbcfg->etscfg.maxtcs = cee_cfg->oper_num_tc;
@@ -700,13 +654,13 @@ static void i40e_cee_to_dcb_config(
 	 * from those in the CEE Priority Group sub-TLV.
 	 */
 	for (i = 0; i < 4; i++) {
-		tc = (u8)((cee_cfg->oper_prio_tc[i] &
-			 I40E_CEE_PGID_PRIO_0_MASK) >>
-			 I40E_CEE_PGID_PRIO_0_SHIFT);
-		dcbcfg->etscfg.prioritytable[i * 2] =  tc;
-		tc = (u8)((cee_cfg->oper_prio_tc[i] &
-			 I40E_CEE_PGID_PRIO_1_MASK) >>
-			 I40E_CEE_PGID_PRIO_1_SHIFT);
+		u8 tc;
+
+		tc = FIELD_GET(I40E_CEE_PGID_PRIO_0_MASK,
+			       cee_cfg->oper_prio_tc[i]);
+		dcbcfg->etscfg.prioritytable[i * 2] = tc;
+		tc = FIELD_GET(I40E_CEE_PGID_PRIO_1_MASK,
+			       cee_cfg->oper_prio_tc[i]);
 		dcbcfg->etscfg.prioritytable[i * 2 + 1] = tc;
 	}
 
@@ -729,8 +683,7 @@ static void i40e_cee_to_dcb_config(
 	dcbcfg->pfc.pfccap = I40E_MAX_TRAFFIC_CLASS;
 
 	i = 0;
-	status = (tlv_status & I40E_AQC_CEE_FCOE_STATUS_MASK) >>
-		  I40E_AQC_CEE_FCOE_STATUS_SHIFT;
+	status = FIELD_GET(I40E_AQC_CEE_FCOE_STATUS_MASK, tlv_status);
 	err = (status & I40E_TLV_STATUS_ERR) ? 1 : 0;
 	sync = (status & I40E_TLV_STATUS_SYNC) ? 1 : 0;
 	oper = (status & I40E_TLV_STATUS_OPER) ? 1 : 0;
@@ -738,15 +691,13 @@ static void i40e_cee_to_dcb_config(
 	if (!err && sync && oper) {
 		/* FCoE APP */
 		dcbcfg->app[i].priority =
-			(app_prio & I40E_AQC_CEE_APP_FCOE_MASK) >>
-			 I40E_AQC_CEE_APP_FCOE_SHIFT;
+			FIELD_GET(I40E_AQC_CEE_APP_FCOE_MASK, app_prio);
 		dcbcfg->app[i].selector = I40E_APP_SEL_ETHTYPE;
 		dcbcfg->app[i].protocolid = I40E_APP_PROTOID_FCOE;
 		i++;
 	}
 
-	status = (tlv_status & I40E_AQC_CEE_ISCSI_STATUS_MASK) >>
-		  I40E_AQC_CEE_ISCSI_STATUS_SHIFT;
+	status = FIELD_GET(I40E_AQC_CEE_ISCSI_STATUS_MASK, tlv_status);
 	err = (status & I40E_TLV_STATUS_ERR) ? 1 : 0;
 	sync = (status & I40E_TLV_STATUS_SYNC) ? 1 : 0;
 	oper = (status & I40E_TLV_STATUS_OPER) ? 1 : 0;
@@ -754,15 +705,13 @@ static void i40e_cee_to_dcb_config(
 	if (!err && sync && oper) {
 		/* iSCSI APP */
 		dcbcfg->app[i].priority =
-			(app_prio & I40E_AQC_CEE_APP_ISCSI_MASK) >>
-			 I40E_AQC_CEE_APP_ISCSI_SHIFT;
+			FIELD_GET(I40E_AQC_CEE_APP_ISCSI_MASK, app_prio);
 		dcbcfg->app[i].selector = I40E_APP_SEL_TCPIP;
 		dcbcfg->app[i].protocolid = I40E_APP_PROTOID_ISCSI;
 		i++;
 	}
 
-	status = (tlv_status & I40E_AQC_CEE_FIP_STATUS_MASK) >>
-		  I40E_AQC_CEE_FIP_STATUS_SHIFT;
+	status = FIELD_GET(I40E_AQC_CEE_FIP_STATUS_MASK, tlv_status);
 	err = (status & I40E_TLV_STATUS_ERR) ? 1 : 0;
 	sync = (status & I40E_TLV_STATUS_SYNC) ? 1 : 0;
 	oper = (status & I40E_TLV_STATUS_OPER) ? 1 : 0;
@@ -770,8 +719,7 @@ static void i40e_cee_to_dcb_config(
 	if (!err && sync && oper) {
 		/* FIP APP */
 		dcbcfg->app[i].priority =
-			(app_prio & I40E_AQC_CEE_APP_FIP_MASK) >>
-			 I40E_AQC_CEE_APP_FIP_SHIFT;
+			FIELD_GET(I40E_AQC_CEE_APP_FIP_MASK, app_prio);
 		dcbcfg->app[i].selector = I40E_APP_SEL_ETHTYPE;
 		dcbcfg->app[i].protocolid = I40E_APP_PROTOID_FIP;
 		i++;
@@ -785,9 +733,9 @@ static void i40e_cee_to_dcb_config(
  *
  * Get IEEE mode DCB configuration from the Firmware
  **/
-static i40e_status i40e_get_ieee_dcb_config(struct i40e_hw *hw)
+static int i40e_get_ieee_dcb_config(struct i40e_hw *hw)
 {
-	i40e_status ret = 0;
+	int ret = 0;
 
 	/* IEEE mode */
 	hw->local_dcbx_config.dcbx_mode = I40E_DCBX_MODE_IEEE;
@@ -802,7 +750,7 @@ static i40e_status i40e_get_ieee_dcb_config(struct i40e_hw *hw)
 				     I40E_AQ_LLDP_BRIDGE_TYPE_NEAREST_BRIDGE,
 				     &hw->remote_dcbx_config);
 	/* Don't treat ENOENT as an error for Remote MIBs */
-	if (hw->aq.asq_last_status == I40E_AQ_RC_ENOENT)
+	if (hw->aq.asq_last_status == LIBIE_AQ_RC_ENOENT)
 		ret = 0;
 
 out:
@@ -815,21 +763,18 @@ out:
  *
  * Get DCB configuration from the Firmware
  **/
-i40e_status i40e_get_dcb_config(struct i40e_hw *hw)
+int i40e_get_dcb_config(struct i40e_hw *hw)
 {
-	i40e_status ret = 0;
-	struct i40e_aqc_get_cee_dcb_cfg_resp cee_cfg;
 	struct i40e_aqc_get_cee_dcb_cfg_v1_resp cee_v1_cfg;
+	struct i40e_aqc_get_cee_dcb_cfg_resp cee_cfg;
+	int ret = 0;
 
 	/* If Firmware version < v4.33 on X710/XL710, IEEE only */
-	if ((hw->mac.type == I40E_MAC_XL710) &&
-	    (((hw->aq.fw_maj_ver == 4) && (hw->aq.fw_min_ver < 33)) ||
-	      (hw->aq.fw_maj_ver < 4)))
+	if (hw->mac.type == I40E_MAC_XL710 && i40e_is_fw_ver_lt(hw, 4, 33))
 		return i40e_get_ieee_dcb_config(hw);
 
 	/* If Firmware version == v4.33 on X710/XL710, use old CEE struct */
-	if ((hw->mac.type == I40E_MAC_XL710) &&
-	    ((hw->aq.fw_maj_ver == 4) && (hw->aq.fw_min_ver == 33))) {
+	if (hw->mac.type == I40E_MAC_XL710 && i40e_is_fw_ver_eq(hw, 4, 33)) {
 		ret = i40e_aq_get_cee_dcb_config(hw, &cee_v1_cfg,
 						 sizeof(cee_v1_cfg), NULL);
 		if (!ret) {
@@ -854,7 +799,7 @@ i40e_status i40e_get_dcb_config(struct i40e_hw *hw)
 	}
 
 	/* CEE mode not enabled try querying IEEE data */
-	if (hw->aq.asq_last_status == I40E_AQ_RC_ENOENT)
+	if (hw->aq.asq_last_status == LIBIE_AQ_RC_ENOENT)
 		return i40e_get_ieee_dcb_config(hw);
 
 	if (ret)
@@ -871,7 +816,7 @@ i40e_status i40e_get_dcb_config(struct i40e_hw *hw)
 				     I40E_AQ_LLDP_BRIDGE_TYPE_NEAREST_BRIDGE,
 				     &hw->remote_dcbx_config);
 	/* Don't treat ENOENT as an error for Remote MIBs */
-	if (hw->aq.asq_last_status == I40E_AQ_RC_ENOENT)
+	if (hw->aq.asq_last_status == LIBIE_AQ_RC_ENOENT)
 		ret = 0;
 
 out:
@@ -881,22 +826,41 @@ out:
 /**
  * i40e_init_dcb
  * @hw: pointer to the hw struct
+ * @enable_mib_change: enable mib change event
  *
  * Update DCB configuration from the Firmware
  **/
-i40e_status i40e_init_dcb(struct i40e_hw *hw)
+int i40e_init_dcb(struct i40e_hw *hw, bool enable_mib_change)
 {
-	i40e_status ret = 0;
 	struct i40e_lldp_variables lldp_cfg;
 	u8 adminstatus = 0;
+	int ret = 0;
 
 	if (!hw->func_caps.dcb)
-		return ret;
+		return -EOPNOTSUPP;
 
 	/* Read LLDP NVM area */
-	ret = i40e_read_lldp_cfg(hw, &lldp_cfg);
+	if (test_bit(I40E_HW_CAP_FW_LLDP_PERSISTENT, hw->caps)) {
+		u8 offset = 0;
+
+		if (hw->mac.type == I40E_MAC_XL710)
+			offset = I40E_LLDP_CURRENT_STATUS_XL710_OFFSET;
+		else if (hw->mac.type == I40E_MAC_X722)
+			offset = I40E_LLDP_CURRENT_STATUS_X722_OFFSET;
+		else
+			return -EOPNOTSUPP;
+
+		ret = i40e_read_nvm_module_data(hw,
+						I40E_SR_EMP_SR_SETTINGS_PTR,
+						offset,
+						I40E_LLDP_CURRENT_STATUS_OFFSET,
+						I40E_LLDP_CURRENT_STATUS_SIZE,
+						&lldp_cfg.adminstatus);
+	} else {
+		ret = i40e_read_lldp_cfg(hw, &lldp_cfg);
+	}
 	if (ret)
-		return ret;
+		return -EBUSY;
 
 	/* Get the LLDP AdminStatus for the current port */
 	adminstatus = lldp_cfg.adminstatus >> (hw->port * 4);
@@ -905,7 +869,7 @@ i40e_status i40e_init_dcb(struct i40e_hw *hw)
 	/* LLDP agent disabled */
 	if (!adminstatus) {
 		hw->dcbx_status = I40E_DCBX_STATUS_DISABLED;
-		return ret;
+		return -EBUSY;
 	}
 
 	/* Get DCBX status */
@@ -914,27 +878,996 @@ i40e_status i40e_init_dcb(struct i40e_hw *hw)
 		return ret;
 
 	/* Check the DCBX Status */
-	switch (hw->dcbx_status) {
-	case I40E_DCBX_STATUS_DONE:
-	case I40E_DCBX_STATUS_IN_PROGRESS:
+	if (hw->dcbx_status == I40E_DCBX_STATUS_DONE ||
+	    hw->dcbx_status == I40E_DCBX_STATUS_IN_PROGRESS) {
 		/* Get current DCBX configuration */
 		ret = i40e_get_dcb_config(hw);
 		if (ret)
 			return ret;
+	} else if (hw->dcbx_status == I40E_DCBX_STATUS_DISABLED) {
+		return -EBUSY;
+	}
+
+	/* Configure the LLDP MIB change event */
+	if (enable_mib_change)
+		ret = i40e_aq_cfg_lldp_mib_change_event(hw, true, NULL);
+
+	return ret;
+}
+
+/**
+ * i40e_get_fw_lldp_status
+ * @hw: pointer to the hw struct
+ * @lldp_status: pointer to the status enum
+ *
+ * Get status of FW Link Layer Discovery Protocol (LLDP) Agent.
+ * Status of agent is reported via @lldp_status parameter.
+ **/
+int
+i40e_get_fw_lldp_status(struct i40e_hw *hw,
+			enum i40e_get_fw_lldp_status_resp *lldp_status)
+{
+	struct i40e_virt_mem mem;
+	u8 *lldpmib;
+	int ret;
+
+	if (!lldp_status)
+		return -EINVAL;
+
+	/* Allocate buffer for the LLDPDU */
+	ret = i40e_allocate_virt_mem(hw, &mem, I40E_LLDPDU_SIZE);
+	if (ret)
+		return ret;
+
+	lldpmib = (u8 *)mem.va;
+	ret = i40e_aq_get_lldp_mib(hw, 0, 0, (void *)lldpmib,
+				   I40E_LLDPDU_SIZE, NULL, NULL, NULL);
+
+	if (!ret) {
+		*lldp_status = I40E_GET_FW_LLDP_STATUS_ENABLED;
+	} else if (hw->aq.asq_last_status == LIBIE_AQ_RC_ENOENT) {
+		/* MIB is not available yet but the agent is running */
+		*lldp_status = I40E_GET_FW_LLDP_STATUS_ENABLED;
+		ret = 0;
+	} else if (hw->aq.asq_last_status == LIBIE_AQ_RC_EPERM) {
+		*lldp_status = I40E_GET_FW_LLDP_STATUS_DISABLED;
+		ret = 0;
+	}
+
+	i40e_free_virt_mem(hw, &mem);
+	return ret;
+}
+
+/**
+ * i40e_add_ieee_ets_tlv - Prepare ETS TLV in IEEE format
+ * @tlv: Fill the ETS config data in IEEE format
+ * @dcbcfg: Local store which holds the DCB Config
+ *
+ * Prepare IEEE 802.1Qaz ETS CFG TLV
+ **/
+static void i40e_add_ieee_ets_tlv(struct i40e_lldp_org_tlv *tlv,
+				  struct i40e_dcbx_config *dcbcfg)
+{
+	u8 priority0, priority1, maxtcwilling = 0;
+	struct i40e_dcb_ets_config *etscfg;
+	u16 offset = 0, typelength, i;
+	u8 *buf = tlv->tlvinfo;
+	u32 ouisubtype;
+
+	typelength = (u16)((I40E_TLV_TYPE_ORG << I40E_LLDP_TLV_TYPE_SHIFT) |
+			I40E_IEEE_ETS_TLV_LENGTH);
+	tlv->typelength = htons(typelength);
+
+	ouisubtype = (u32)((I40E_IEEE_8021QAZ_OUI << I40E_LLDP_TLV_OUI_SHIFT) |
+			I40E_IEEE_SUBTYPE_ETS_CFG);
+	tlv->ouisubtype = htonl(ouisubtype);
+
+	/* First Octet post subtype
+	 * --------------------------
+	 * |will-|CBS  | Re-  | Max |
+	 * |ing  |     |served| TCs |
+	 * --------------------------
+	 * |1bit | 1bit|3 bits|3bits|
+	 */
+	etscfg = &dcbcfg->etscfg;
+	if (etscfg->willing)
+		maxtcwilling = BIT(I40E_IEEE_ETS_WILLING_SHIFT);
+	maxtcwilling |= etscfg->maxtcs & I40E_IEEE_ETS_MAXTC_MASK;
+	buf[offset] = maxtcwilling;
+
+	/* Move offset to Priority Assignment Table */
+	offset++;
+
+	/* Priority Assignment Table (4 octets)
+	 * Octets:|    1    |    2    |    3    |    4    |
+	 *        -----------------------------------------
+	 *        |pri0|pri1|pri2|pri3|pri4|pri5|pri6|pri7|
+	 *        -----------------------------------------
+	 *   Bits:|7  4|3  0|7  4|3  0|7  4|3  0|7  4|3  0|
+	 *        -----------------------------------------
+	 */
+	for (i = 0; i < 4; i++) {
+		priority0 = etscfg->prioritytable[i * 2] & 0xF;
+		priority1 = etscfg->prioritytable[i * 2 + 1] & 0xF;
+		buf[offset] = (priority0 << I40E_IEEE_ETS_PRIO_1_SHIFT) |
+				priority1;
+		offset++;
+	}
+
+	/* TC Bandwidth Table (8 octets)
+	 * Octets:| 1 | 2 | 3 | 4 | 5 | 6 | 7 | 8 |
+	 *        ---------------------------------
+	 *        |tc0|tc1|tc2|tc3|tc4|tc5|tc6|tc7|
+	 *        ---------------------------------
+	 */
+	for (i = 0; i < I40E_MAX_TRAFFIC_CLASS; i++)
+		buf[offset++] = etscfg->tcbwtable[i];
+
+	/* TSA Assignment Table (8 octets)
+	 * Octets:| 1 | 2 | 3 | 4 | 5 | 6 | 7 | 8 |
+	 *        ---------------------------------
+	 *        |tc0|tc1|tc2|tc3|tc4|tc5|tc6|tc7|
+	 *        ---------------------------------
+	 */
+	for (i = 0; i < I40E_MAX_TRAFFIC_CLASS; i++)
+		buf[offset++] = etscfg->tsatable[i];
+}
+
+/**
+ * i40e_add_ieee_etsrec_tlv - Prepare ETS Recommended TLV in IEEE format
+ * @tlv: Fill ETS Recommended TLV in IEEE format
+ * @dcbcfg: Local store which holds the DCB Config
+ *
+ * Prepare IEEE 802.1Qaz ETS REC TLV
+ **/
+static void i40e_add_ieee_etsrec_tlv(struct i40e_lldp_org_tlv *tlv,
+				     struct i40e_dcbx_config *dcbcfg)
+{
+	struct i40e_dcb_ets_config *etsrec;
+	u16 offset = 0, typelength, i;
+	u8 priority0, priority1;
+	u8 *buf = tlv->tlvinfo;
+	u32 ouisubtype;
+
+	typelength = (u16)((I40E_TLV_TYPE_ORG << I40E_LLDP_TLV_TYPE_SHIFT) |
+			I40E_IEEE_ETS_TLV_LENGTH);
+	tlv->typelength = htons(typelength);
+
+	ouisubtype = (u32)((I40E_IEEE_8021QAZ_OUI << I40E_LLDP_TLV_OUI_SHIFT) |
+			I40E_IEEE_SUBTYPE_ETS_REC);
+	tlv->ouisubtype = htonl(ouisubtype);
+
+	etsrec = &dcbcfg->etsrec;
+	/* First Octet is reserved */
+	/* Move offset to Priority Assignment Table */
+	offset++;
+
+	/* Priority Assignment Table (4 octets)
+	 * Octets:|    1    |    2    |    3    |    4    |
+	 *        -----------------------------------------
+	 *        |pri0|pri1|pri2|pri3|pri4|pri5|pri6|pri7|
+	 *        -----------------------------------------
+	 *   Bits:|7  4|3  0|7  4|3  0|7  4|3  0|7  4|3  0|
+	 *        -----------------------------------------
+	 */
+	for (i = 0; i < 4; i++) {
+		priority0 = etsrec->prioritytable[i * 2] & 0xF;
+		priority1 = etsrec->prioritytable[i * 2 + 1] & 0xF;
+		buf[offset] = (priority0 << I40E_IEEE_ETS_PRIO_1_SHIFT) |
+				priority1;
+		offset++;
+	}
+
+	/* TC Bandwidth Table (8 octets)
+	 * Octets:| 1 | 2 | 3 | 4 | 5 | 6 | 7 | 8 |
+	 *        ---------------------------------
+	 *        |tc0|tc1|tc2|tc3|tc4|tc5|tc6|tc7|
+	 *        ---------------------------------
+	 */
+	for (i = 0; i < I40E_MAX_TRAFFIC_CLASS; i++)
+		buf[offset++] = etsrec->tcbwtable[i];
+
+	/* TSA Assignment Table (8 octets)
+	 * Octets:| 1 | 2 | 3 | 4 | 5 | 6 | 7 | 8 |
+	 *        ---------------------------------
+	 *        |tc0|tc1|tc2|tc3|tc4|tc5|tc6|tc7|
+	 *        ---------------------------------
+	 */
+	for (i = 0; i < I40E_MAX_TRAFFIC_CLASS; i++)
+		buf[offset++] = etsrec->tsatable[i];
+}
+
+/**
+ * i40e_add_ieee_pfc_tlv - Prepare PFC TLV in IEEE format
+ * @tlv: Fill PFC TLV in IEEE format
+ * @dcbcfg: Local store to get PFC CFG data
+ *
+ * Prepare IEEE 802.1Qaz PFC CFG TLV
+ **/
+static void i40e_add_ieee_pfc_tlv(struct i40e_lldp_org_tlv *tlv,
+				  struct i40e_dcbx_config *dcbcfg)
+{
+	u8 *buf = tlv->tlvinfo;
+	u32 ouisubtype;
+	u16 typelength;
+
+	typelength = (u16)((I40E_TLV_TYPE_ORG << I40E_LLDP_TLV_TYPE_SHIFT) |
+			I40E_IEEE_PFC_TLV_LENGTH);
+	tlv->typelength = htons(typelength);
+
+	ouisubtype = (u32)((I40E_IEEE_8021QAZ_OUI << I40E_LLDP_TLV_OUI_SHIFT) |
+			I40E_IEEE_SUBTYPE_PFC_CFG);
+	tlv->ouisubtype = htonl(ouisubtype);
+
+	/* ----------------------------------------
+	 * |will-|MBC  | Re-  | PFC |  PFC Enable  |
+	 * |ing  |     |served| cap |              |
+	 * -----------------------------------------
+	 * |1bit | 1bit|2 bits|4bits| 1 octet      |
+	 */
+	if (dcbcfg->pfc.willing)
+		buf[0] = BIT(I40E_IEEE_PFC_WILLING_SHIFT);
+
+	if (dcbcfg->pfc.mbc)
+		buf[0] |= BIT(I40E_IEEE_PFC_MBC_SHIFT);
+
+	buf[0] |= dcbcfg->pfc.pfccap & 0xF;
+	buf[1] = dcbcfg->pfc.pfcenable;
+}
+
+/**
+ * i40e_add_ieee_app_pri_tlv -  Prepare APP TLV in IEEE format
+ * @tlv: Fill APP TLV in IEEE format
+ * @dcbcfg: Local store to get APP CFG data
+ *
+ * Prepare IEEE 802.1Qaz APP CFG TLV
+ **/
+static void i40e_add_ieee_app_pri_tlv(struct i40e_lldp_org_tlv *tlv,
+				      struct i40e_dcbx_config *dcbcfg)
+{
+	u16 typelength, length, offset = 0;
+	u8 priority, selector, i = 0;
+	u8 *buf = tlv->tlvinfo;
+	u32 ouisubtype;
+
+	/* No APP TLVs then just return */
+	if (dcbcfg->numapps == 0)
+		return;
+	ouisubtype = (u32)((I40E_IEEE_8021QAZ_OUI << I40E_LLDP_TLV_OUI_SHIFT) |
+			I40E_IEEE_SUBTYPE_APP_PRI);
+	tlv->ouisubtype = htonl(ouisubtype);
+
+	/* Move offset to App Priority Table */
+	offset++;
+	/* Application Priority Table (3 octets)
+	 * Octets:|         1          |    2    |    3    |
+	 *        -----------------------------------------
+	 *        |Priority|Rsrvd| Sel |    Protocol ID    |
+	 *        -----------------------------------------
+	 *   Bits:|23    21|20 19|18 16|15                0|
+	 *        -----------------------------------------
+	 */
+	while (i < dcbcfg->numapps) {
+		priority = dcbcfg->app[i].priority & 0x7;
+		selector = dcbcfg->app[i].selector & 0x7;
+		buf[offset] = (priority << I40E_IEEE_APP_PRIO_SHIFT) | selector;
+		buf[offset + 1] = (dcbcfg->app[i].protocolid >> 0x8) & 0xFF;
+		buf[offset + 2] = dcbcfg->app[i].protocolid & 0xFF;
+		/* Move to next app */
+		offset += 3;
+		i++;
+		if (i >= I40E_DCBX_MAX_APPS)
+			break;
+	}
+	/* length includes size of ouisubtype + 1 reserved + 3*numapps */
+	length = sizeof(tlv->ouisubtype) + 1 + (i * 3);
+	typelength = (u16)((I40E_TLV_TYPE_ORG << I40E_LLDP_TLV_TYPE_SHIFT) |
+		(length & 0x1FF));
+	tlv->typelength = htons(typelength);
+}
+
+/**
+ * i40e_add_dcb_tlv - Add all IEEE TLVs
+ * @tlv: pointer to org tlv
+ * @dcbcfg: pointer to modified dcbx config structure *
+ * @tlvid: tlv id to be added
+ * add tlv information
+ **/
+static void i40e_add_dcb_tlv(struct i40e_lldp_org_tlv *tlv,
+			     struct i40e_dcbx_config *dcbcfg,
+			     u16 tlvid)
+{
+	switch (tlvid) {
+	case I40E_IEEE_TLV_ID_ETS_CFG:
+		i40e_add_ieee_ets_tlv(tlv, dcbcfg);
+		break;
+	case I40E_IEEE_TLV_ID_ETS_REC:
+		i40e_add_ieee_etsrec_tlv(tlv, dcbcfg);
+		break;
+	case I40E_IEEE_TLV_ID_PFC_CFG:
+		i40e_add_ieee_pfc_tlv(tlv, dcbcfg);
+		break;
+	case I40E_IEEE_TLV_ID_APP_PRI:
+		i40e_add_ieee_app_pri_tlv(tlv, dcbcfg);
+		break;
+	default:
 		break;
-	case I40E_DCBX_STATUS_DISABLED:
+	}
+}
+
+/**
+ * i40e_set_dcb_config - Set the local LLDP MIB to FW
+ * @hw: pointer to the hw struct
+ *
+ * Set DCB configuration to the Firmware
+ **/
+int i40e_set_dcb_config(struct i40e_hw *hw)
+{
+	struct i40e_dcbx_config *dcbcfg;
+	struct i40e_virt_mem mem;
+	u8 mib_type, *lldpmib;
+	u16 miblen;
+	int ret;
+
+	/* update the hw local config */
+	dcbcfg = &hw->local_dcbx_config;
+	/* Allocate the LLDPDU */
+	ret = i40e_allocate_virt_mem(hw, &mem, I40E_LLDPDU_SIZE);
+	if (ret)
 		return ret;
-	case I40E_DCBX_STATUS_NOT_STARTED:
-	case I40E_DCBX_STATUS_MULTIPLE_PEERS:
+
+	mib_type = SET_LOCAL_MIB_AC_TYPE_LOCAL_MIB;
+	if (dcbcfg->app_mode == I40E_DCBX_APPS_NON_WILLING) {
+		mib_type |= SET_LOCAL_MIB_AC_TYPE_NON_WILLING_APPS <<
+			    SET_LOCAL_MIB_AC_TYPE_NON_WILLING_APPS_SHIFT;
+	}
+	lldpmib = (u8 *)mem.va;
+	i40e_dcb_config_to_lldp(lldpmib, &miblen, dcbcfg);
+	ret = i40e_aq_set_lldp_mib(hw, mib_type, (void *)lldpmib, miblen, NULL);
+
+	i40e_free_virt_mem(hw, &mem);
+	return ret;
+}
+
+/**
+ * i40e_dcb_config_to_lldp - Convert Dcbconfig to MIB format
+ * @lldpmib: pointer to mib to be output
+ * @miblen: pointer to u16 for length of lldpmib
+ * @dcbcfg: store for LLDPDU data
+ *
+ * send DCB configuration to FW
+ **/
+int i40e_dcb_config_to_lldp(u8 *lldpmib, u16 *miblen,
+			    struct i40e_dcbx_config *dcbcfg)
+{
+	u16 length, offset = 0, tlvid, typelength;
+	struct i40e_lldp_org_tlv *tlv;
+
+	tlv = (struct i40e_lldp_org_tlv *)lldpmib;
+	tlvid = I40E_TLV_ID_START;
+	do {
+		i40e_add_dcb_tlv(tlv, dcbcfg, tlvid++);
+		typelength = ntohs(tlv->typelength);
+		length = FIELD_GET(I40E_LLDP_TLV_LEN_MASK, typelength);
+		if (length)
+			offset += length + I40E_IEEE_TLV_HEADER_LENGTH;
+		/* END TLV or beyond LLDPDU size */
+		if (tlvid >= I40E_TLV_ID_END_OF_LLDPPDU ||
+		    offset >= I40E_LLDPDU_SIZE)
+			break;
+		/* Move to next TLV */
+		if (length)
+			tlv = (struct i40e_lldp_org_tlv *)((char *)tlv +
+			      sizeof(tlv->typelength) + length);
+	} while (tlvid < I40E_TLV_ID_END_OF_LLDPPDU);
+	*miblen = offset;
+	return 0;
+}
+
+/**
+ * i40e_dcb_hw_rx_fifo_config
+ * @hw: pointer to the hw struct
+ * @ets_mode: Strict Priority or Round Robin mode
+ * @non_ets_mode: Strict Priority or Round Robin
+ * @max_exponent: Exponent to calculate max refill credits
+ * @lltc_map: Low latency TC bitmap
+ *
+ * Configure HW Rx FIFO as part of DCB configuration.
+ **/
+void i40e_dcb_hw_rx_fifo_config(struct i40e_hw *hw,
+				enum i40e_dcb_arbiter_mode ets_mode,
+				enum i40e_dcb_arbiter_mode non_ets_mode,
+				u32 max_exponent,
+				u8 lltc_map)
+{
+	u32 reg = rd32(hw, I40E_PRTDCB_RETSC);
+
+	reg &= ~I40E_PRTDCB_RETSC_ETS_MODE_MASK;
+	reg |= FIELD_PREP(I40E_PRTDCB_RETSC_ETS_MODE_MASK, ets_mode);
+
+	reg &= ~I40E_PRTDCB_RETSC_NON_ETS_MODE_MASK;
+	reg |= FIELD_PREP(I40E_PRTDCB_RETSC_NON_ETS_MODE_MASK, non_ets_mode);
+
+	reg &= ~I40E_PRTDCB_RETSC_ETS_MAX_EXP_MASK;
+	reg |= FIELD_PREP(I40E_PRTDCB_RETSC_ETS_MAX_EXP_MASK, max_exponent);
+
+	reg &= ~I40E_PRTDCB_RETSC_LLTC_MASK;
+	reg |= FIELD_PREP(I40E_PRTDCB_RETSC_LLTC_MASK, lltc_map);
+	wr32(hw, I40E_PRTDCB_RETSC, reg);
+}
+
+/**
+ * i40e_dcb_hw_rx_cmd_monitor_config
+ * @hw: pointer to the hw struct
+ * @num_tc: Total number of traffic class
+ * @num_ports: Total number of ports on device
+ *
+ * Configure HW Rx command monitor as part of DCB configuration.
+ **/
+void i40e_dcb_hw_rx_cmd_monitor_config(struct i40e_hw *hw,
+				       u8 num_tc, u8 num_ports)
+{
+	u32 threshold;
+	u32 fifo_size;
+	u32 reg;
+
+	/* Set the threshold and fifo_size based on number of ports */
+	switch (num_ports) {
+	case 1:
+		threshold = I40E_DCB_1_PORT_THRESHOLD;
+		fifo_size = I40E_DCB_1_PORT_FIFO_SIZE;
+		break;
+	case 2:
+		if (num_tc > 4) {
+			threshold = I40E_DCB_2_PORT_THRESHOLD_HIGH_NUM_TC;
+			fifo_size = I40E_DCB_2_PORT_FIFO_SIZE_HIGH_NUM_TC;
+		} else {
+			threshold = I40E_DCB_2_PORT_THRESHOLD_LOW_NUM_TC;
+			fifo_size = I40E_DCB_2_PORT_FIFO_SIZE_LOW_NUM_TC;
+		}
+		break;
+	case 4:
+		if (num_tc > 4) {
+			threshold = I40E_DCB_4_PORT_THRESHOLD_HIGH_NUM_TC;
+			fifo_size = I40E_DCB_4_PORT_FIFO_SIZE_HIGH_NUM_TC;
+		} else {
+			threshold = I40E_DCB_4_PORT_THRESHOLD_LOW_NUM_TC;
+			fifo_size = I40E_DCB_4_PORT_FIFO_SIZE_LOW_NUM_TC;
+		}
+		break;
 	default:
+		i40e_debug(hw, I40E_DEBUG_DCB, "Invalid num_ports %u.\n",
+			   (u32)num_ports);
+		return;
+	}
+
+	/* The hardware manual describes setting up of I40E_PRT_SWR_PM_THR
+	 * based on the number of ports and traffic classes for a given port as
+	 * part of DCB configuration.
+	 */
+	reg = rd32(hw, I40E_PRT_SWR_PM_THR);
+	reg &= ~I40E_PRT_SWR_PM_THR_THRESHOLD_MASK;
+	reg |= FIELD_PREP(I40E_PRT_SWR_PM_THR_THRESHOLD_MASK, threshold);
+	wr32(hw, I40E_PRT_SWR_PM_THR, reg);
+
+	reg = rd32(hw, I40E_PRTDCB_RPPMC);
+	reg &= ~I40E_PRTDCB_RPPMC_RX_FIFO_SIZE_MASK;
+	reg |= FIELD_PREP(I40E_PRTDCB_RPPMC_RX_FIFO_SIZE_MASK, fifo_size);
+	wr32(hw, I40E_PRTDCB_RPPMC, reg);
+}
+
+/**
+ * i40e_dcb_hw_pfc_config
+ * @hw: pointer to the hw struct
+ * @pfc_en: Bitmap of PFC enabled priorities
+ * @prio_tc: priority to tc assignment indexed by priority
+ *
+ * Configure HW Priority Flow Controller as part of DCB configuration.
+ **/
+void i40e_dcb_hw_pfc_config(struct i40e_hw *hw,
+			    u8 pfc_en, u8 *prio_tc)
+{
+	u16 refresh_time = (u16)I40E_DEFAULT_PAUSE_TIME / 2;
+	u32 link_speed = hw->phy.link_info.link_speed;
+	u8 first_pfc_prio = 0;
+	u8 num_pfc_tc = 0;
+	u8 tc2pfc = 0;
+	u32 reg;
+	u8 i;
+
+	/* Get Number of PFC TCs and TC2PFC map */
+	for (i = 0; i < I40E_MAX_USER_PRIORITY; i++) {
+		if (pfc_en & BIT(i)) {
+			if (!first_pfc_prio)
+				first_pfc_prio = i;
+			/* Set bit for the PFC TC */
+			tc2pfc |= BIT(prio_tc[i]);
+			num_pfc_tc++;
+		}
+	}
+
+	switch (link_speed) {
+	case I40E_LINK_SPEED_10GB:
+		reg = rd32(hw, I40E_PRTDCB_MFLCN);
+		reg |= BIT(I40E_PRTDCB_MFLCN_DPF_SHIFT) &
+			I40E_PRTDCB_MFLCN_DPF_MASK;
+		reg &= ~I40E_PRTDCB_MFLCN_RFCE_MASK;
+		reg &= ~I40E_PRTDCB_MFLCN_RPFCE_MASK;
+		if (pfc_en) {
+			reg |= FIELD_PREP(I40E_PRTDCB_MFLCN_RPFCM_MASK, 1);
+			reg |= FIELD_PREP(I40E_PRTDCB_MFLCN_RPFCE_MASK,
+					  pfc_en);
+		}
+		wr32(hw, I40E_PRTDCB_MFLCN, reg);
+
+		reg = rd32(hw, I40E_PRTDCB_FCCFG);
+		reg &= ~I40E_PRTDCB_FCCFG_TFCE_MASK;
+		if (pfc_en)
+			reg |= FIELD_PREP(I40E_PRTDCB_FCCFG_TFCE_MASK,
+					  I40E_DCB_PFC_ENABLED);
+		wr32(hw, I40E_PRTDCB_FCCFG, reg);
+
+		/* FCTTV and FCRTV to be set by default */
+		break;
+	case I40E_LINK_SPEED_40GB:
+		reg = rd32(hw, I40E_PRTMAC_HSEC_CTL_RX_ENABLE_GPP);
+		reg &= ~I40E_PRTMAC_HSEC_CTL_RX_ENABLE_GPP_MASK;
+		wr32(hw, I40E_PRTMAC_HSEC_CTL_RX_ENABLE_GPP, reg);
+
+		reg = rd32(hw, I40E_PRTMAC_HSEC_CTL_RX_ENABLE_PPP);
+		reg &= ~I40E_PRTMAC_HSEC_CTL_RX_ENABLE_GPP_MASK;
+		reg |= BIT(I40E_PRTMAC_HSEC_CTL_RX_ENABLE_PPP_SHIFT) &
+			I40E_PRTMAC_HSEC_CTL_RX_ENABLE_PPP_MASK;
+		wr32(hw, I40E_PRTMAC_HSEC_CTL_RX_ENABLE_PPP, reg);
+
+		reg = rd32(hw, I40E_PRTMAC_HSEC_CTL_RX_PAUSE_ENABLE);
+		reg &= ~I40E_PRTMAC_HSEC_CTL_RX_PAUSE_ENABLE_MASK;
+		reg |= FIELD_PREP(I40E_PRTMAC_HSEC_CTL_RX_PAUSE_ENABLE_MASK,
+				  pfc_en);
+		wr32(hw, I40E_PRTMAC_HSEC_CTL_RX_PAUSE_ENABLE, reg);
+
+		reg = rd32(hw, I40E_PRTMAC_HSEC_CTL_TX_PAUSE_ENABLE);
+		reg &= ~I40E_PRTMAC_HSEC_CTL_TX_PAUSE_ENABLE_MASK;
+		reg |= FIELD_PREP(I40E_PRTMAC_HSEC_CTL_TX_PAUSE_ENABLE_MASK,
+				  pfc_en);
+		wr32(hw, I40E_PRTMAC_HSEC_CTL_TX_PAUSE_ENABLE, reg);
+
+		for (i = 0; i < I40E_PRTMAC_HSEC_CTL_TX_PAUSE_REFRESH_TIMER_MAX_INDEX; i++) {
+			reg = rd32(hw, I40E_PRTMAC_HSEC_CTL_TX_PAUSE_REFRESH_TIMER(i));
+			reg &= ~I40E_PRTMAC_HSEC_CTL_TX_PAUSE_REFRESH_TIMER_MASK;
+			if (pfc_en) {
+				reg |= FIELD_PREP(I40E_PRTMAC_HSEC_CTL_TX_PAUSE_REFRESH_TIMER_MASK,
+						  refresh_time);
+			}
+			wr32(hw, I40E_PRTMAC_HSEC_CTL_TX_PAUSE_REFRESH_TIMER(i), reg);
+		}
+		/* PRTMAC_HSEC_CTL_TX_PAUSE_QUANTA default value is 0xFFFF
+		 * for all user priorities
+		 */
 		break;
 	}
 
-	/* Configure the LLDP MIB change event */
-	ret = i40e_aq_cfg_lldp_mib_change_event(hw, true, NULL);
+	reg = rd32(hw, I40E_PRTDCB_TC2PFC);
+	reg &= ~I40E_PRTDCB_TC2PFC_TC2PFC_MASK;
+	reg |= FIELD_PREP(I40E_PRTDCB_TC2PFC_TC2PFC_MASK, tc2pfc);
+	wr32(hw, I40E_PRTDCB_TC2PFC, reg);
+
+	reg = rd32(hw, I40E_PRTDCB_RUP);
+	reg &= ~I40E_PRTDCB_RUP_NOVLANUP_MASK;
+	reg |= FIELD_PREP(I40E_PRTDCB_RUP_NOVLANUP_MASK, first_pfc_prio);
+	wr32(hw, I40E_PRTDCB_RUP, reg);
+
+	reg = rd32(hw, I40E_PRTDCB_TDPMC);
+	reg &= ~I40E_PRTDCB_TDPMC_TCPM_MODE_MASK;
+	if (num_pfc_tc > I40E_DCB_PFC_FORCED_NUM_TC) {
+		reg |= BIT(I40E_PRTDCB_TDPMC_TCPM_MODE_SHIFT) &
+			I40E_PRTDCB_TDPMC_TCPM_MODE_MASK;
+	}
+	wr32(hw, I40E_PRTDCB_TDPMC, reg);
+
+	reg = rd32(hw, I40E_PRTDCB_TCPMC);
+	reg &= ~I40E_PRTDCB_TCPMC_TCPM_MODE_MASK;
+	if (num_pfc_tc > I40E_DCB_PFC_FORCED_NUM_TC) {
+		reg |= BIT(I40E_PRTDCB_TCPMC_TCPM_MODE_SHIFT) &
+			I40E_PRTDCB_TCPMC_TCPM_MODE_MASK;
+	}
+	wr32(hw, I40E_PRTDCB_TCPMC, reg);
+}
+
+/**
+ * i40e_dcb_hw_set_num_tc
+ * @hw: pointer to the hw struct
+ * @num_tc: number of traffic classes
+ *
+ * Configure number of traffic classes in HW
+ **/
+void i40e_dcb_hw_set_num_tc(struct i40e_hw *hw, u8 num_tc)
+{
+	u32 reg = rd32(hw, I40E_PRTDCB_GENC);
+
+	reg &= ~I40E_PRTDCB_GENC_NUMTC_MASK;
+	reg |= FIELD_PREP(I40E_PRTDCB_GENC_NUMTC_MASK, num_tc);
+	wr32(hw, I40E_PRTDCB_GENC, reg);
+}
+
+/**
+ * i40e_dcb_hw_rx_ets_bw_config
+ * @hw: pointer to the hw struct
+ * @bw_share: Bandwidth share indexed per traffic class
+ * @mode: Strict Priority or Round Robin mode between UP sharing same
+ * traffic class
+ * @prio_type: TC is ETS enabled or strict priority
+ *
+ * Configure HW Rx ETS bandwidth as part of DCB configuration.
+ **/
+void i40e_dcb_hw_rx_ets_bw_config(struct i40e_hw *hw, u8 *bw_share,
+				  u8 *mode, u8 *prio_type)
+{
+	u32 reg;
+	u8 i;
+
+	for (i = 0; i <= I40E_PRTDCB_RETSTCC_MAX_INDEX; i++) {
+		reg = rd32(hw, I40E_PRTDCB_RETSTCC(i));
+		reg &= ~(I40E_PRTDCB_RETSTCC_BWSHARE_MASK     |
+			 I40E_PRTDCB_RETSTCC_UPINTC_MODE_MASK |
+			 I40E_PRTDCB_RETSTCC_ETSTC_MASK);
+		reg |= FIELD_PREP(I40E_PRTDCB_RETSTCC_BWSHARE_MASK,
+				  bw_share[i]);
+		reg |= FIELD_PREP(I40E_PRTDCB_RETSTCC_UPINTC_MODE_MASK,
+				  mode[i]);
+		reg |= FIELD_PREP(I40E_PRTDCB_RETSTCC_ETSTC_MASK,
+				  prio_type[i]);
+		wr32(hw, I40E_PRTDCB_RETSTCC(i), reg);
+	}
+}
+
+/**
+ * i40e_dcb_hw_rx_up2tc_config
+ * @hw: pointer to the hw struct
+ * @prio_tc: priority to tc assignment indexed by priority
+ *
+ * Configure HW Rx UP2TC map as part of DCB configuration.
+ **/
+void i40e_dcb_hw_rx_up2tc_config(struct i40e_hw *hw, u8 *prio_tc)
+{
+	u32 reg = rd32(hw, I40E_PRTDCB_RUP2TC);
+#define I40E_UP2TC_REG(val, i) \
+		(((val) << I40E_PRTDCB_RUP2TC_UP##i##TC_SHIFT) & \
+		  I40E_PRTDCB_RUP2TC_UP##i##TC_MASK)
+
+	reg |= I40E_UP2TC_REG(prio_tc[0], 0);
+	reg |= I40E_UP2TC_REG(prio_tc[1], 1);
+	reg |= I40E_UP2TC_REG(prio_tc[2], 2);
+	reg |= I40E_UP2TC_REG(prio_tc[3], 3);
+	reg |= I40E_UP2TC_REG(prio_tc[4], 4);
+	reg |= I40E_UP2TC_REG(prio_tc[5], 5);
+	reg |= I40E_UP2TC_REG(prio_tc[6], 6);
+	reg |= I40E_UP2TC_REG(prio_tc[7], 7);
+
+	wr32(hw, I40E_PRTDCB_RUP2TC, reg);
+}
+
+/**
+ * i40e_dcb_hw_calculate_pool_sizes - configure dcb pool sizes
+ * @hw: pointer to the hw struct
+ * @num_ports: Number of available ports on the device
+ * @eee_enabled: EEE enabled for the given port
+ * @pfc_en: Bit map of PFC enabled traffic classes
+ * @mfs_tc: Array of max frame size for each traffic class
+ * @pb_cfg: pointer to packet buffer configuration
+ *
+ * Calculate the shared and dedicated per TC pool sizes,
+ * watermarks and threshold values.
+ **/
+void i40e_dcb_hw_calculate_pool_sizes(struct i40e_hw *hw,
+				      u8 num_ports, bool eee_enabled,
+				      u8 pfc_en, u32 *mfs_tc,
+				      struct i40e_rx_pb_config *pb_cfg)
+{
+	u32 pool_size[I40E_MAX_TRAFFIC_CLASS];
+	u32 high_wm[I40E_MAX_TRAFFIC_CLASS];
+	u32 low_wm[I40E_MAX_TRAFFIC_CLASS];
+	u32 total_pool_size = 0;
+	int shared_pool_size; /* Need signed variable */
+	u32 port_pb_size;
+	u32 mfs_max = 0;
+	u32 pcirtt;
+	u8 i;
+
+	/* Get the MFS(max) for the port */
+	for (i = 0; i < I40E_MAX_TRAFFIC_CLASS; i++) {
+		if (mfs_tc[i] > mfs_max)
+			mfs_max = mfs_tc[i];
+	}
+
+	pcirtt = I40E_BT2B(I40E_PCIRTT_LINK_SPEED_10G);
+
+	/* Calculate effective Rx PB size per port */
+	port_pb_size = I40E_DEVICE_RPB_SIZE / num_ports;
+	if (eee_enabled)
+		port_pb_size -= I40E_BT2B(I40E_EEE_TX_LPI_EXIT_TIME);
+	port_pb_size -= mfs_max;
+
+	/* Step 1 Calculating tc pool/shared pool sizes and watermarks */
+	for (i = 0; i < I40E_MAX_TRAFFIC_CLASS; i++) {
+		if (pfc_en & BIT(i)) {
+			low_wm[i] = (I40E_DCB_WATERMARK_START_FACTOR *
+				     mfs_tc[i]) + pcirtt;
+			high_wm[i] = low_wm[i];
+			high_wm[i] += ((mfs_max > I40E_MAX_FRAME_SIZE)
+					? mfs_max : I40E_MAX_FRAME_SIZE);
+			pool_size[i] = high_wm[i];
+			pool_size[i] += I40E_BT2B(I40E_STD_DV_TC(mfs_max,
+								mfs_tc[i]));
+		} else {
+			low_wm[i] = 0;
+			pool_size[i] = (I40E_DCB_WATERMARK_START_FACTOR *
+					mfs_tc[i]) + pcirtt;
+			high_wm[i] = pool_size[i];
+		}
+		total_pool_size += pool_size[i];
+	}
+
+	shared_pool_size = port_pb_size - total_pool_size;
+	if (shared_pool_size > 0) {
+		pb_cfg->shared_pool_size = shared_pool_size;
+		pb_cfg->shared_pool_high_wm = shared_pool_size;
+		pb_cfg->shared_pool_low_wm = 0;
+		for (i = 0; i < I40E_MAX_TRAFFIC_CLASS; i++) {
+			pb_cfg->shared_pool_low_thresh[i] = 0;
+			pb_cfg->shared_pool_high_thresh[i] = shared_pool_size;
+			pb_cfg->tc_pool_size[i] = pool_size[i];
+			pb_cfg->tc_pool_high_wm[i] = high_wm[i];
+			pb_cfg->tc_pool_low_wm[i] = low_wm[i];
+		}
+
+	} else {
+		i40e_debug(hw, I40E_DEBUG_DCB,
+			   "The shared pool size for the port is negative %d.\n",
+			   shared_pool_size);
+	}
+}
+
+/**
+ * i40e_dcb_hw_rx_pb_config
+ * @hw: pointer to the hw struct
+ * @old_pb_cfg: Existing Rx Packet buffer configuration
+ * @new_pb_cfg: New Rx Packet buffer configuration
+ *
+ * Program the Rx Packet Buffer registers.
+ **/
+void i40e_dcb_hw_rx_pb_config(struct i40e_hw *hw,
+			      struct i40e_rx_pb_config *old_pb_cfg,
+			      struct i40e_rx_pb_config *new_pb_cfg)
+{
+	u32 old_val;
+	u32 new_val;
+	u32 reg;
+	u8 i;
+
+	/* The Rx Packet buffer register programming needs to be done in a
+	 * certain order and the following code is based on that
+	 * requirement.
+	 */
+
+	/* Program the shared pool low water mark per port if decreasing */
+	old_val = old_pb_cfg->shared_pool_low_wm;
+	new_val = new_pb_cfg->shared_pool_low_wm;
+	if (new_val < old_val) {
+		reg = rd32(hw, I40E_PRTRPB_SLW);
+		reg &= ~I40E_PRTRPB_SLW_SLW_MASK;
+		reg |= FIELD_PREP(I40E_PRTRPB_SLW_SLW_MASK, new_val);
+		wr32(hw, I40E_PRTRPB_SLW, reg);
+	}
+
+	/* Program the shared pool low threshold and tc pool
+	 * low water mark per TC that are decreasing.
+	 */
+	for (i = 0; i < I40E_MAX_TRAFFIC_CLASS; i++) {
+		old_val = old_pb_cfg->shared_pool_low_thresh[i];
+		new_val = new_pb_cfg->shared_pool_low_thresh[i];
+		if (new_val < old_val) {
+			reg = rd32(hw, I40E_PRTRPB_SLT(i));
+			reg &= ~I40E_PRTRPB_SLT_SLT_TCN_MASK;
+			reg |= FIELD_PREP(I40E_PRTRPB_SLT_SLT_TCN_MASK,
+					  new_val);
+			wr32(hw, I40E_PRTRPB_SLT(i), reg);
+		}
+
+		old_val = old_pb_cfg->tc_pool_low_wm[i];
+		new_val = new_pb_cfg->tc_pool_low_wm[i];
+		if (new_val < old_val) {
+			reg = rd32(hw, I40E_PRTRPB_DLW(i));
+			reg &= ~I40E_PRTRPB_DLW_DLW_TCN_MASK;
+			reg |= FIELD_PREP(I40E_PRTRPB_DLW_DLW_TCN_MASK,
+					  new_val);
+			wr32(hw, I40E_PRTRPB_DLW(i), reg);
+		}
+	}
+
+	/* Program the shared pool high water mark per port if decreasing */
+	old_val = old_pb_cfg->shared_pool_high_wm;
+	new_val = new_pb_cfg->shared_pool_high_wm;
+	if (new_val < old_val) {
+		reg = rd32(hw, I40E_PRTRPB_SHW);
+		reg &= ~I40E_PRTRPB_SHW_SHW_MASK;
+		reg |= FIELD_PREP(I40E_PRTRPB_SHW_SHW_MASK, new_val);
+		wr32(hw, I40E_PRTRPB_SHW, reg);
+	}
+
+	/* Program the shared pool high threshold and tc pool
+	 * high water mark per TC that are decreasing.
+	 */
+	for (i = 0; i < I40E_MAX_TRAFFIC_CLASS; i++) {
+		old_val = old_pb_cfg->shared_pool_high_thresh[i];
+		new_val = new_pb_cfg->shared_pool_high_thresh[i];
+		if (new_val < old_val) {
+			reg = rd32(hw, I40E_PRTRPB_SHT(i));
+			reg &= ~I40E_PRTRPB_SHT_SHT_TCN_MASK;
+			reg |= FIELD_PREP(I40E_PRTRPB_SHT_SHT_TCN_MASK,
+					  new_val);
+			wr32(hw, I40E_PRTRPB_SHT(i), reg);
+		}
+
+		old_val = old_pb_cfg->tc_pool_high_wm[i];
+		new_val = new_pb_cfg->tc_pool_high_wm[i];
+		if (new_val < old_val) {
+			reg = rd32(hw, I40E_PRTRPB_DHW(i));
+			reg &= ~I40E_PRTRPB_DHW_DHW_TCN_MASK;
+			reg |= FIELD_PREP(I40E_PRTRPB_DHW_DHW_TCN_MASK,
+					  new_val);
+			wr32(hw, I40E_PRTRPB_DHW(i), reg);
+		}
+	}
+
+	/* Write Dedicated Pool Sizes per TC */
+	for (i = 0; i < I40E_MAX_TRAFFIC_CLASS; i++) {
+		new_val = new_pb_cfg->tc_pool_size[i];
+		reg = rd32(hw, I40E_PRTRPB_DPS(i));
+		reg &= ~I40E_PRTRPB_DPS_DPS_TCN_MASK;
+		reg |= FIELD_PREP(I40E_PRTRPB_DPS_DPS_TCN_MASK, new_val);
+		wr32(hw, I40E_PRTRPB_DPS(i), reg);
+	}
+
+	/* Write Shared Pool Size per port */
+	new_val = new_pb_cfg->shared_pool_size;
+	reg = rd32(hw, I40E_PRTRPB_SPS);
+	reg &= ~I40E_PRTRPB_SPS_SPS_MASK;
+	reg |= FIELD_PREP(I40E_PRTRPB_SPS_SPS_MASK, new_val);
+	wr32(hw, I40E_PRTRPB_SPS, reg);
+
+	/* Program the shared pool low water mark per port if increasing */
+	old_val = old_pb_cfg->shared_pool_low_wm;
+	new_val = new_pb_cfg->shared_pool_low_wm;
+	if (new_val > old_val) {
+		reg = rd32(hw, I40E_PRTRPB_SLW);
+		reg &= ~I40E_PRTRPB_SLW_SLW_MASK;
+		reg |= FIELD_PREP(I40E_PRTRPB_SLW_SLW_MASK, new_val);
+		wr32(hw, I40E_PRTRPB_SLW, reg);
+	}
+
+	/* Program the shared pool low threshold and tc pool
+	 * low water mark per TC that are increasing.
+	 */
+	for (i = 0; i < I40E_MAX_TRAFFIC_CLASS; i++) {
+		old_val = old_pb_cfg->shared_pool_low_thresh[i];
+		new_val = new_pb_cfg->shared_pool_low_thresh[i];
+		if (new_val > old_val) {
+			reg = rd32(hw, I40E_PRTRPB_SLT(i));
+			reg &= ~I40E_PRTRPB_SLT_SLT_TCN_MASK;
+			reg |= FIELD_PREP(I40E_PRTRPB_SLT_SLT_TCN_MASK,
+					  new_val);
+			wr32(hw, I40E_PRTRPB_SLT(i), reg);
+		}
+
+		old_val = old_pb_cfg->tc_pool_low_wm[i];
+		new_val = new_pb_cfg->tc_pool_low_wm[i];
+		if (new_val > old_val) {
+			reg = rd32(hw, I40E_PRTRPB_DLW(i));
+			reg &= ~I40E_PRTRPB_DLW_DLW_TCN_MASK;
+			reg |= FIELD_PREP(I40E_PRTRPB_DLW_DLW_TCN_MASK,
+					  new_val);
+			wr32(hw, I40E_PRTRPB_DLW(i), reg);
+		}
+	}
+
+	/* Program the shared pool high water mark per port if increasing */
+	old_val = old_pb_cfg->shared_pool_high_wm;
+	new_val = new_pb_cfg->shared_pool_high_wm;
+	if (new_val > old_val) {
+		reg = rd32(hw, I40E_PRTRPB_SHW);
+		reg &= ~I40E_PRTRPB_SHW_SHW_MASK;
+		reg |= FIELD_PREP(I40E_PRTRPB_SHW_SHW_MASK, new_val);
+		wr32(hw, I40E_PRTRPB_SHW, reg);
+	}
+
+	/* Program the shared pool high threshold and tc pool
+	 * high water mark per TC that are increasing.
+	 */
+	for (i = 0; i < I40E_MAX_TRAFFIC_CLASS; i++) {
+		old_val = old_pb_cfg->shared_pool_high_thresh[i];
+		new_val = new_pb_cfg->shared_pool_high_thresh[i];
+		if (new_val > old_val) {
+			reg = rd32(hw, I40E_PRTRPB_SHT(i));
+			reg &= ~I40E_PRTRPB_SHT_SHT_TCN_MASK;
+			reg |= FIELD_PREP(I40E_PRTRPB_SHT_SHT_TCN_MASK,
+					  new_val);
+			wr32(hw, I40E_PRTRPB_SHT(i), reg);
+		}
+
+		old_val = old_pb_cfg->tc_pool_high_wm[i];
+		new_val = new_pb_cfg->tc_pool_high_wm[i];
+		if (new_val > old_val) {
+			reg = rd32(hw, I40E_PRTRPB_DHW(i));
+			reg &= ~I40E_PRTRPB_DHW_DHW_TCN_MASK;
+			reg |= FIELD_PREP(I40E_PRTRPB_DHW_DHW_TCN_MASK,
+					  new_val);
+			wr32(hw, I40E_PRTRPB_DHW(i), reg);
+		}
+	}
+}
+
+/**
+ * _i40e_read_lldp_cfg - generic read of LLDP Configuration data from NVM
+ * @hw: pointer to the HW structure
+ * @lldp_cfg: pointer to hold lldp configuration variables
+ * @module: address of the module pointer
+ * @word_offset: offset of LLDP configuration
+ *
+ * Reads the LLDP configuration data from NVM using passed addresses
+ **/
+static int _i40e_read_lldp_cfg(struct i40e_hw *hw,
+			       struct i40e_lldp_variables *lldp_cfg,
+			       u8 module, u32 word_offset)
+{
+	u32 address, offset = (2 * word_offset);
+	__le16 raw_mem;
+	int ret;
+	u16 mem;
+
+	ret = i40e_acquire_nvm(hw, I40E_RESOURCE_READ);
+	if (ret)
+		return ret;
+
+	ret = i40e_aq_read_nvm(hw, 0x0, module * 2, sizeof(raw_mem), &raw_mem,
+			       true, NULL);
+	i40e_release_nvm(hw);
+	if (ret)
+		return ret;
+
+	mem = le16_to_cpu(raw_mem);
+	/* Check if this pointer needs to be read in word size or 4K sector
+	 * units.
+	 */
+	if (mem & I40E_PTR_TYPE)
+		address = (0x7FFF & mem) * 4096;
+	else
+		address = (0x7FFF & mem) * 2;
+
+	ret = i40e_acquire_nvm(hw, I40E_RESOURCE_READ);
+	if (ret)
+		goto err_lldp_cfg;
+
+	ret = i40e_aq_read_nvm(hw, module, offset, sizeof(raw_mem), &raw_mem,
+			       true, NULL);
+	i40e_release_nvm(hw);
 	if (ret)
 		return ret;
 
+	mem = le16_to_cpu(raw_mem);
+	offset = mem + word_offset;
+	offset *= 2;
+
+	ret = i40e_acquire_nvm(hw, I40E_RESOURCE_READ);
+	if (ret)
+		goto err_lldp_cfg;
+
+	ret = i40e_aq_read_nvm(hw, 0, address + offset,
+			       sizeof(struct i40e_lldp_variables), lldp_cfg,
+			       true, NULL);
+	i40e_release_nvm(hw);
+
+err_lldp_cfg:
 	return ret;
 }
 
@@ -945,25 +1878,38 @@ i40e_status i40e_init_dcb(struct i40e_hw *hw)
  *
  * Reads the LLDP configuration data from NVM
  **/
-i40e_status i40e_read_lldp_cfg(struct i40e_hw *hw,
-			       struct i40e_lldp_variables *lldp_cfg)
+int i40e_read_lldp_cfg(struct i40e_hw *hw,
+		       struct i40e_lldp_variables *lldp_cfg)
 {
-	i40e_status ret = 0;
-	u32 offset = (2 * I40E_NVM_LLDP_CFG_PTR);
+	int ret = 0;
+	u32 mem;
 
 	if (!lldp_cfg)
-		return I40E_ERR_PARAM;
+		return -EINVAL;
 
 	ret = i40e_acquire_nvm(hw, I40E_RESOURCE_READ);
 	if (ret)
-		goto err_lldp_cfg;
+		return ret;
 
-	ret = i40e_aq_read_nvm(hw, I40E_SR_EMP_MODULE_PTR, offset,
-			       sizeof(struct i40e_lldp_variables),
-			       (u8 *)lldp_cfg,
-			       true, NULL);
+	ret = i40e_aq_read_nvm(hw, I40E_SR_NVM_CONTROL_WORD, 0, sizeof(mem),
+			       &mem, true, NULL);
 	i40e_release_nvm(hw);
+	if (ret)
+		return ret;
+
+	/* Read a bit that holds information whether we are running flat or
+	 * structured NVM image. Flat image has LLDP configuration in shadow
+	 * ram, so there is a need to pass different addresses for both cases.
+	 */
+	if (mem & I40E_SR_NVM_MAP_STRUCTURE_TYPE) {
+		/* Flat NVM case */
+		ret = _i40e_read_lldp_cfg(hw, lldp_cfg, I40E_SR_EMP_MODULE_PTR,
+					  I40E_SR_LLDP_CFG_PTR);
+	} else {
+		/* Good old structured NVM image */
+		ret = _i40e_read_lldp_cfg(hw, lldp_cfg, I40E_EMP_MODULE_PTR,
+					  I40E_NVM_LLDP_CFG_PTR);
+	}
 
-err_lldp_cfg:
 	return ret;
 }
diff --git a/drivers/net/ethernet/intel/i40e/i40e_dcb.h b/drivers/net/ethernet/intel/i40e/i40e_dcb.h
index 92d01042c1f6..d5662c639c41 100644
--- a/drivers/net/ethernet/intel/i40e/i40e_dcb.h
+++ b/drivers/net/ethernet/intel/i40e/i40e_dcb.h
@@ -1,28 +1,5 @@
-/*******************************************************************************
- *
- * Intel Ethernet Controller XL710 Family Linux Driver
- * Copyright(c) 2013 - 2014 Intel Corporation.
- *
- * This program is free software; you can redistribute it and/or modify it
- * under the terms and conditions of the GNU General Public License,
- * version 2, as published by the Free Software Foundation.
- *
- * This program is distributed in the hope it will be useful, but WITHOUT
- * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
- * FITNESS FOR A PARTICULAR PURPOSE.  See the GNU General Public License for
- * more details.
- *
- * You should have received a copy of the GNU General Public License along
- * with this program.  If not, see <http://www.gnu.org/licenses/>.
- *
- * The full GNU General Public License is included in this distribution in
- * the file called "COPYING".
- *
- * Contact Information:
- * e1000-devel Mailing List <e1000-devel@lists.sourceforge.net>
- * Intel Corporation, 5200 N.E. Elam Young Parkway, Hillsboro, OR 97124-6497
- *
- ******************************************************************************/
+/* SPDX-License-Identifier: GPL-2.0 */
+/* Copyright(c) 2013 - 2021 Intel Corporation. */
 
 #ifndef _I40E_DCB_H_
 #define _I40E_DCB_H_
@@ -53,6 +30,11 @@
 #define I40E_CEE_SUBTYPE_APP_PRI	4
 
 #define I40E_CEE_MAX_FEAT_TYPE		3
+#define I40E_LLDP_CURRENT_STATUS_XL710_OFFSET	0x2B
+#define I40E_LLDP_CURRENT_STATUS_X722_OFFSET	0x31
+#define I40E_LLDP_CURRENT_STATUS_OFFSET		1
+#define I40E_LLDP_CURRENT_STATUS_SIZE		1
+
 /* Defines for LLDP TLV header */
 #define I40E_LLDP_TLV_LEN_SHIFT		0
 #define I40E_LLDP_TLV_LEN_MASK		(0x01FF << I40E_LLDP_TLV_LEN_SHIFT)
@@ -61,7 +43,7 @@
 #define I40E_LLDP_TLV_SUBTYPE_SHIFT	0
 #define I40E_LLDP_TLV_SUBTYPE_MASK	(0xFF << I40E_LLDP_TLV_SUBTYPE_SHIFT)
 #define I40E_LLDP_TLV_OUI_SHIFT		8
-#define I40E_LLDP_TLV_OUI_MASK		(0xFFFFFF << I40E_LLDP_TLV_OUI_SHIFT)
+#define I40E_LLDP_TLV_OUI_MASK		(0xFFFFFFU << I40E_LLDP_TLV_OUI_SHIFT)
 
 /* Defines for IEEE ETS TLV */
 #define I40E_IEEE_ETS_MAXTC_SHIFT	0
@@ -85,6 +67,8 @@
 #define I40E_IEEE_TSA_ETS		2
 
 /* Defines for IEEE PFC TLV */
+#define I40E_DCB_PFC_ENABLED		2
+#define I40E_DCB_PFC_FORCED_NUM_TC	2
 #define I40E_IEEE_PFC_CAP_SHIFT		0
 #define I40E_IEEE_PFC_CAP_MASK		(0xF << I40E_IEEE_PFC_CAP_SHIFT)
 #define I40E_IEEE_PFC_MBC_SHIFT		6
@@ -98,9 +82,30 @@
 #define I40E_IEEE_APP_PRIO_SHIFT	5
 #define I40E_IEEE_APP_PRIO_MASK		(0x7 << I40E_IEEE_APP_PRIO_SHIFT)
 
+/* TLV definitions for preparing MIB */
+#define I40E_TLV_ID_CHASSIS_ID		0
+#define I40E_TLV_ID_PORT_ID		1
+#define I40E_TLV_ID_TIME_TO_LIVE	2
+#define I40E_IEEE_TLV_ID_ETS_CFG	3
+#define I40E_IEEE_TLV_ID_ETS_REC	4
+#define I40E_IEEE_TLV_ID_PFC_CFG	5
+#define I40E_IEEE_TLV_ID_APP_PRI	6
+#define I40E_TLV_ID_END_OF_LLDPPDU	7
+#define I40E_TLV_ID_START		I40E_IEEE_TLV_ID_ETS_CFG
 
-#pragma pack(1)
+#define I40E_IEEE_TLV_HEADER_LENGTH	2
+#define I40E_IEEE_ETS_TLV_LENGTH	25
+#define I40E_IEEE_PFC_TLV_LENGTH	6
+#define I40E_IEEE_APP_TLV_LENGTH	11
 
+/* Defines for default SW DCB config */
+#define I40E_IEEE_DEFAULT_ETS_TCBW	100
+#define I40E_IEEE_DEFAULT_ETS_WILLING	1
+#define I40E_IEEE_DEFAULT_PFC_WILLING	1
+#define I40E_IEEE_DEFAULT_NUM_APPS	1
+#define I40E_IEEE_DEFAULT_APP_PRIO	3
+
+#pragma pack(1)
 /* IEEE 802.1AB LLDP Organization specific TLV */
 struct i40e_lldp_org_tlv {
 	__be16 typelength;
@@ -139,13 +144,139 @@ struct i40e_cee_app_prio {
 };
 #pragma pack()
 
-i40e_status i40e_get_dcbx_status(struct i40e_hw *hw,
-					   u16 *status);
-i40e_status i40e_lldp_to_dcb_config(u8 *lldpmib,
-					      struct i40e_dcbx_config *dcbcfg);
-i40e_status i40e_aq_get_dcb_config(struct i40e_hw *hw, u8 mib_type,
-					     u8 bridgetype,
-					     struct i40e_dcbx_config *dcbcfg);
-i40e_status i40e_get_dcb_config(struct i40e_hw *hw);
-i40e_status i40e_init_dcb(struct i40e_hw *hw);
+enum i40e_get_fw_lldp_status_resp {
+	I40E_GET_FW_LLDP_STATUS_DISABLED = 0,
+	I40E_GET_FW_LLDP_STATUS_ENABLED = 1
+};
+
+/* Data structures to pass for SW DCBX */
+struct i40e_rx_pb_config {
+	u32	shared_pool_size;
+	u32	shared_pool_high_wm;
+	u32	shared_pool_low_wm;
+	u32	shared_pool_high_thresh[I40E_MAX_TRAFFIC_CLASS];
+	u32	shared_pool_low_thresh[I40E_MAX_TRAFFIC_CLASS];
+	u32	tc_pool_size[I40E_MAX_TRAFFIC_CLASS];
+	u32	tc_pool_high_wm[I40E_MAX_TRAFFIC_CLASS];
+	u32	tc_pool_low_wm[I40E_MAX_TRAFFIC_CLASS];
+};
+
+enum i40e_dcb_arbiter_mode {
+	I40E_DCB_ARB_MODE_STRICT_PRIORITY = 0,
+	I40E_DCB_ARB_MODE_ROUND_ROBIN = 1
+};
+
+#define I40E_DCB_DEFAULT_MAX_EXPONENT		0xB
+#define I40E_DEFAULT_PAUSE_TIME			0xffff
+#define I40E_MAX_FRAME_SIZE			4608 /* 4.5 KB */
+
+#define I40E_DEVICE_RPB_SIZE			968000 /* 968 KB */
+
+/* BitTimes (BT) conversion */
+#define I40E_BT2KB(BT) (((BT) + (8 * 1024 - 1)) / (8 * 1024))
+#define I40E_B2BT(BT) ((BT) * 8)
+#define I40E_BT2B(BT) (((BT) + (8 - 1)) / 8)
+
+/* Max Frame(TC) = MFS(max) + MFS(TC) */
+#define I40E_MAX_FRAME_TC(mfs_max, mfs_tc)	I40E_B2BT((mfs_max) + (mfs_tc))
+
+/* EEE Tx LPI Exit time in Bit Times */
+#define I40E_EEE_TX_LPI_EXIT_TIME		142500
+
+/* PCI Round Trip Time in Bit Times */
+#define I40E_PCIRTT_LINK_SPEED_10G		20000
+#define I40E_PCIRTT_BYTE_LINK_SPEED_20G		40000
+#define I40E_PCIRTT_BYTE_LINK_SPEED_40G		80000
+
+/* PFC Frame Delay Bit Times */
+#define I40E_PFC_FRAME_DELAY			672
+
+/* Worst case Cable (10GBase-T) Delay Bit Times */
+#define I40E_CABLE_DELAY			5556
+
+/* Higher Layer Delay @10G Bit Times */
+#define I40E_HIGHER_LAYER_DELAY_10G		6144
+
+/* Interface Delays in Bit Times */
+/* TODO: Add for other link speeds 20G/40G/etc. */
+#define I40E_INTERFACE_DELAY_10G_MAC_CONTROL	8192
+#define I40E_INTERFACE_DELAY_10G_MAC		8192
+#define I40E_INTERFACE_DELAY_10G_RS		8192
+
+#define I40E_INTERFACE_DELAY_XGXS		2048
+#define I40E_INTERFACE_DELAY_XAUI		2048
+
+#define I40E_INTERFACE_DELAY_10G_BASEX_PCS	2048
+#define I40E_INTERFACE_DELAY_10G_BASER_PCS	3584
+#define I40E_INTERFACE_DELAY_LX4_PMD		512
+#define I40E_INTERFACE_DELAY_CX4_PMD		512
+#define I40E_INTERFACE_DELAY_SERIAL_PMA		512
+#define I40E_INTERFACE_DELAY_PMD		512
+
+#define I40E_INTERFACE_DELAY_10G_BASET		25600
+
+/* Hardware RX DCB config related defines */
+#define I40E_DCB_1_PORT_THRESHOLD		0xF
+#define I40E_DCB_1_PORT_FIFO_SIZE		0x10
+#define I40E_DCB_2_PORT_THRESHOLD_LOW_NUM_TC	0xF
+#define I40E_DCB_2_PORT_FIFO_SIZE_LOW_NUM_TC	0x10
+#define I40E_DCB_2_PORT_THRESHOLD_HIGH_NUM_TC	0xC
+#define I40E_DCB_2_PORT_FIFO_SIZE_HIGH_NUM_TC	0x8
+#define I40E_DCB_4_PORT_THRESHOLD_LOW_NUM_TC	0x9
+#define I40E_DCB_4_PORT_FIFO_SIZE_LOW_NUM_TC	0x8
+#define I40E_DCB_4_PORT_THRESHOLD_HIGH_NUM_TC	0x6
+#define I40E_DCB_4_PORT_FIFO_SIZE_HIGH_NUM_TC	0x4
+#define I40E_DCB_WATERMARK_START_FACTOR		0x2
+
+/* delay values for with 10G BaseT in Bit Times */
+#define I40E_INTERFACE_DELAY_10G_COPPER	\
+	(I40E_INTERFACE_DELAY_10G_MAC + (2 * I40E_INTERFACE_DELAY_XAUI) \
+	 + I40E_INTERFACE_DELAY_10G_BASET)
+#define I40E_DV_TC(mfs_max, mfs_tc) \
+		((2 * I40E_MAX_FRAME_TC(mfs_max, mfs_tc)) \
+		 + I40E_PFC_FRAME_DELAY \
+		 + (2 * I40E_CABLE_DELAY) \
+		 + (2 * I40E_INTERFACE_DELAY_10G_COPPER) \
+		 + I40E_HIGHER_LAYER_DELAY_10G)
+static inline u32 I40E_STD_DV_TC(u32 mfs_max, u32 mfs_tc)
+{
+	return I40E_DV_TC(mfs_max, mfs_tc) + I40E_B2BT(mfs_max);
+}
+
+/* APIs for SW DCBX */
+void i40e_dcb_hw_rx_fifo_config(struct i40e_hw *hw,
+				enum i40e_dcb_arbiter_mode ets_mode,
+				enum i40e_dcb_arbiter_mode non_ets_mode,
+				u32 max_exponent, u8 lltc_map);
+void i40e_dcb_hw_rx_cmd_monitor_config(struct i40e_hw *hw,
+				       u8 num_tc, u8 num_ports);
+void i40e_dcb_hw_pfc_config(struct i40e_hw *hw,
+			    u8 pfc_en, u8 *prio_tc);
+void i40e_dcb_hw_set_num_tc(struct i40e_hw *hw, u8 num_tc);
+void i40e_dcb_hw_rx_ets_bw_config(struct i40e_hw *hw, u8 *bw_share,
+				  u8 *mode, u8 *prio_type);
+void i40e_dcb_hw_rx_up2tc_config(struct i40e_hw *hw, u8 *prio_tc);
+void i40e_dcb_hw_calculate_pool_sizes(struct i40e_hw *hw,
+				      u8 num_ports, bool eee_enabled,
+				      u8 pfc_en, u32 *mfs_tc,
+				      struct i40e_rx_pb_config *pb_cfg);
+void i40e_dcb_hw_rx_pb_config(struct i40e_hw *hw,
+			      struct i40e_rx_pb_config *old_pb_cfg,
+			      struct i40e_rx_pb_config *new_pb_cfg);
+int i40e_get_dcbx_status(struct i40e_hw *hw,
+			 u16 *status);
+int i40e_lldp_to_dcb_config(u8 *lldpmib,
+			    struct i40e_dcbx_config *dcbcfg);
+int i40e_aq_get_dcb_config(struct i40e_hw *hw, u8 mib_type,
+			   u8 bridgetype,
+			   struct i40e_dcbx_config *dcbcfg);
+int i40e_get_dcb_config(struct i40e_hw *hw);
+int i40e_init_dcb(struct i40e_hw *hw,
+		  bool enable_mib_change);
+int
+i40e_get_fw_lldp_status(struct i40e_hw *hw,
+			enum i40e_get_fw_lldp_status_resp *lldp_status);
+int i40e_set_dcb_config(struct i40e_hw *hw);
+int i40e_dcb_config_to_lldp(u8 *lldpmib, u16 *miblen,
+			    struct i40e_dcbx_config *dcbcfg);
 #endif /* _I40E_DCB_H_ */
diff --git a/drivers/net/ethernet/intel/i40e/i40e_dcb_nl.c b/drivers/net/ethernet/intel/i40e/i40e_dcb_nl.c
index 886e667f2f1c..a2ccf4c5e30b 100644
--- a/drivers/net/ethernet/intel/i40e/i40e_dcb_nl.c
+++ b/drivers/net/ethernet/intel/i40e/i40e_dcb_nl.c
@@ -1,33 +1,14 @@
-/*******************************************************************************
- *
- * Intel Ethernet Controller XL710 Family Linux Driver
- * Copyright(c) 2013 - 2014 Intel Corporation.
- *
- * This program is free software; you can redistribute it and/or modify it
- * under the terms and conditions of the GNU General Public License,
- * version 2, as published by the Free Software Foundation.
- *
- * This program is distributed in the hope it will be useful, but WITHOUT
- * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
- * FITNESS FOR A PARTICULAR PURPOSE.  See the GNU General Public License for
- * more details.
- *
- * You should have received a copy of the GNU General Public License along
- * with this program.  If not, see <http://www.gnu.org/licenses/>.
- *
- * The full GNU General Public License is included in this distribution in
- * the file called "COPYING".
- *
- * Contact Information:
- * e1000-devel Mailing List <e1000-devel@lists.sourceforge.net>
- * Intel Corporation, 5200 N.E. Elam Young Parkway, Hillsboro, OR 97124-6497
- *
- ******************************************************************************/
+// SPDX-License-Identifier: GPL-2.0
+/* Copyright(c) 2013 - 2021 Intel Corporation. */
 
 #ifdef CONFIG_I40E_DCB
-#include "i40e.h"
 #include <net/dcbnl.h>
+#include "i40e.h"
 
+#define I40E_DCBNL_STATUS_SUCCESS	0
+#define I40E_DCBNL_STATUS_ERROR		1
+static bool i40e_dcbnl_find_app(struct i40e_dcbx_config *cfg,
+				struct i40e_dcb_app_priority_table *app);
 /**
  * i40e_get_pfc_delay - retrieve PFC Link Delay
  * @hw: pointer to hardware struct
@@ -40,13 +21,12 @@ static void i40e_get_pfc_delay(struct i40e_hw *hw, u16 *delay)
 	u32 val;
 
 	val = rd32(hw, I40E_PRTDCB_GENC);
-	*delay = (u16)((val & I40E_PRTDCB_GENC_PFCLDA_MASK) >>
-		       I40E_PRTDCB_GENC_PFCLDA_SHIFT);
+	*delay = FIELD_GET(I40E_PRTDCB_GENC_PFCLDA_MASK, val);
 }
 
 /**
  * i40e_dcbnl_ieee_getets - retrieve local IEEE ETS configuration
- * @netdev: the corresponding netdev
+ * @dev: the corresponding netdev
  * @ets: structure to hold the ETS information
  *
  * Returns local IEEE ETS configuration
@@ -56,14 +36,13 @@ static int i40e_dcbnl_ieee_getets(struct net_device *dev,
 {
 	struct i40e_pf *pf = i40e_netdev_to_pf(dev);
 	struct i40e_dcbx_config *dcbxcfg;
-	struct i40e_hw *hw = &pf->hw;
 
 	if (!(pf->dcbx_cap & DCB_CAP_DCBX_VER_IEEE))
 		return -EINVAL;
 
-	dcbxcfg = &hw->local_dcbx_config;
+	dcbxcfg = &pf->hw.local_dcbx_config;
 	ets->willing = dcbxcfg->etscfg.willing;
-	ets->ets_cap = dcbxcfg->etscfg.maxtcs;
+	ets->ets_cap = I40E_MAX_TRAFFIC_CLASS;
 	ets->cbs = dcbxcfg->etscfg.cbs;
 	memcpy(ets->tc_tx_bw, dcbxcfg->etscfg.tcbwtable,
 		sizeof(ets->tc_tx_bw));
@@ -85,8 +64,8 @@ static int i40e_dcbnl_ieee_getets(struct net_device *dev,
 
 /**
  * i40e_dcbnl_ieee_getpfc - retrieve local IEEE PFC configuration
- * @netdev: the corresponding netdev
- * @ets: structure to hold the PFC information
+ * @dev: the corresponding netdev
+ * @pfc: structure to hold the PFC information
  *
  * Returns local IEEE PFC configuration
  **/
@@ -107,7 +86,7 @@ static int i40e_dcbnl_ieee_getpfc(struct net_device *dev,
 	pfc->mbc = dcbxcfg->pfc.mbc;
 	i40e_get_pfc_delay(hw, &pfc->delay);
 
-	/* Get Requests/Indicatiosn */
+	/* Get Requests/Indications */
 	for (i = 0; i < I40E_MAX_TRAFFIC_CLASS; i++) {
 		pfc->requests[i] = pf->stats.priority_xoff_tx[i];
 		pfc->indications[i] = pf->stats.priority_xoff_rx[i];
@@ -117,8 +96,715 @@ static int i40e_dcbnl_ieee_getpfc(struct net_device *dev,
 }
 
 /**
- * i40e_dcbnl_getdcbx - retrieve current DCBx capability
+ * i40e_dcbnl_ieee_setets - set IEEE ETS configuration
+ * @netdev: the corresponding netdev
+ * @ets: structure to hold the ETS information
+ *
+ * Set IEEE ETS configuration
+ **/
+static int i40e_dcbnl_ieee_setets(struct net_device *netdev,
+				  struct ieee_ets *ets)
+{
+	struct i40e_pf *pf = i40e_netdev_to_pf(netdev);
+	struct i40e_dcbx_config *old_cfg;
+	int i, ret;
+
+	if (!(pf->dcbx_cap & DCB_CAP_DCBX_VER_IEEE) ||
+	    (pf->dcbx_cap & DCB_CAP_DCBX_LLD_MANAGED))
+		return -EINVAL;
+
+	old_cfg = &pf->hw.local_dcbx_config;
+	/* Copy current config into temp */
+	pf->tmp_cfg = *old_cfg;
+
+	/* Update the ETS configuration for temp */
+	pf->tmp_cfg.etscfg.willing = ets->willing;
+	pf->tmp_cfg.etscfg.maxtcs = I40E_MAX_TRAFFIC_CLASS;
+	pf->tmp_cfg.etscfg.cbs = ets->cbs;
+	for (i = 0; i < I40E_MAX_TRAFFIC_CLASS; i++) {
+		pf->tmp_cfg.etscfg.tcbwtable[i] = ets->tc_tx_bw[i];
+		pf->tmp_cfg.etscfg.tsatable[i] = ets->tc_tsa[i];
+		pf->tmp_cfg.etscfg.prioritytable[i] = ets->prio_tc[i];
+		pf->tmp_cfg.etsrec.tcbwtable[i] = ets->tc_reco_bw[i];
+		pf->tmp_cfg.etsrec.tsatable[i] = ets->tc_reco_tsa[i];
+		pf->tmp_cfg.etsrec.prioritytable[i] = ets->reco_prio_tc[i];
+	}
+
+	/* Commit changes to HW */
+	ret = i40e_hw_dcb_config(pf, &pf->tmp_cfg);
+	if (ret) {
+		dev_info(&pf->pdev->dev,
+			 "Failed setting DCB ETS configuration err %pe aq_err %s\n",
+			 ERR_PTR(ret),
+			 libie_aq_str(pf->hw.aq.asq_last_status));
+		return -EINVAL;
+	}
+
+	return 0;
+}
+
+/**
+ * i40e_dcbnl_ieee_setpfc - set local IEEE PFC configuration
+ * @netdev: the corresponding netdev
+ * @pfc: structure to hold the PFC information
+ *
+ * Sets local IEEE PFC configuration
+ **/
+static int i40e_dcbnl_ieee_setpfc(struct net_device *netdev,
+				  struct ieee_pfc *pfc)
+{
+	struct i40e_pf *pf = i40e_netdev_to_pf(netdev);
+	struct i40e_dcbx_config *old_cfg;
+	int ret;
+
+	if (!(pf->dcbx_cap & DCB_CAP_DCBX_VER_IEEE) ||
+	    (pf->dcbx_cap & DCB_CAP_DCBX_LLD_MANAGED))
+		return -EINVAL;
+
+	old_cfg = &pf->hw.local_dcbx_config;
+	/* Copy current config into temp */
+	pf->tmp_cfg = *old_cfg;
+	if (pfc->pfc_cap)
+		pf->tmp_cfg.pfc.pfccap = pfc->pfc_cap;
+	else
+		pf->tmp_cfg.pfc.pfccap = I40E_MAX_TRAFFIC_CLASS;
+	pf->tmp_cfg.pfc.pfcenable = pfc->pfc_en;
+
+	ret = i40e_hw_dcb_config(pf, &pf->tmp_cfg);
+	if (ret) {
+		dev_info(&pf->pdev->dev,
+			 "Failed setting DCB PFC configuration err %pe aq_err %s\n",
+			 ERR_PTR(ret),
+			 libie_aq_str(pf->hw.aq.asq_last_status));
+		return -EINVAL;
+	}
+
+	return 0;
+}
+
+/**
+ * i40e_dcbnl_ieee_setapp - set local IEEE App configuration
  * @netdev: the corresponding netdev
+ * @app: structure to hold the Application information
+ *
+ * Sets local IEEE App configuration
+ **/
+static int i40e_dcbnl_ieee_setapp(struct net_device *netdev,
+				  struct dcb_app *app)
+{
+	struct i40e_pf *pf = i40e_netdev_to_pf(netdev);
+	struct i40e_dcb_app_priority_table new_app;
+	struct i40e_dcbx_config *old_cfg;
+	int ret;
+
+	if (!(pf->dcbx_cap & DCB_CAP_DCBX_VER_IEEE) ||
+	    (pf->dcbx_cap & DCB_CAP_DCBX_LLD_MANAGED))
+		return -EINVAL;
+
+	old_cfg = &pf->hw.local_dcbx_config;
+	if (old_cfg->numapps == I40E_DCBX_MAX_APPS)
+		return -EINVAL;
+
+	ret = dcb_ieee_setapp(netdev, app);
+	if (ret)
+		return ret;
+
+	new_app.selector = app->selector;
+	new_app.protocolid = app->protocol;
+	new_app.priority = app->priority;
+	/* Already internally available */
+	if (i40e_dcbnl_find_app(old_cfg, &new_app))
+		return 0;
+
+	/* Copy current config into temp */
+	pf->tmp_cfg = *old_cfg;
+	/* Add the app */
+	pf->tmp_cfg.app[pf->tmp_cfg.numapps++] = new_app;
+
+	ret = i40e_hw_dcb_config(pf, &pf->tmp_cfg);
+	if (ret) {
+		dev_info(&pf->pdev->dev,
+			 "Failed setting DCB configuration err %pe aq_err %s\n",
+			 ERR_PTR(ret),
+			 libie_aq_str(pf->hw.aq.asq_last_status));
+		return -EINVAL;
+	}
+
+	return 0;
+}
+
+/**
+ * i40e_dcbnl_ieee_delapp - delete local IEEE App configuration
+ * @netdev: the corresponding netdev
+ * @app: structure to hold the Application information
+ *
+ * Deletes local IEEE App configuration other than the first application
+ * required by firmware
+ **/
+static int i40e_dcbnl_ieee_delapp(struct net_device *netdev,
+				  struct dcb_app *app)
+{
+	struct i40e_pf *pf = i40e_netdev_to_pf(netdev);
+	struct i40e_dcbx_config *old_cfg;
+	int i, j, ret;
+
+	if (!(pf->dcbx_cap & DCB_CAP_DCBX_VER_IEEE) ||
+	    (pf->dcbx_cap & DCB_CAP_DCBX_LLD_MANAGED))
+		return -EINVAL;
+
+	ret = dcb_ieee_delapp(netdev, app);
+	if (ret)
+		return ret;
+
+	old_cfg = &pf->hw.local_dcbx_config;
+	/* Need one app for FW so keep it */
+	if (old_cfg->numapps == 1)
+		return 0;
+
+	/* Copy current config into temp */
+	pf->tmp_cfg = *old_cfg;
+
+	/* Find and reset the app */
+	for (i = 1; i < pf->tmp_cfg.numapps; i++) {
+		if (app->selector == pf->tmp_cfg.app[i].selector &&
+		    app->protocol == pf->tmp_cfg.app[i].protocolid &&
+		    app->priority == pf->tmp_cfg.app[i].priority) {
+			/* Reset the app data */
+			pf->tmp_cfg.app[i].selector = 0;
+			pf->tmp_cfg.app[i].protocolid = 0;
+			pf->tmp_cfg.app[i].priority = 0;
+			break;
+		}
+	}
+
+	/* If the specific DCB app not found */
+	if (i == pf->tmp_cfg.numapps)
+		return -EINVAL;
+
+	pf->tmp_cfg.numapps--;
+	/* Overwrite the tmp_cfg app */
+	for (j = i; j < pf->tmp_cfg.numapps; j++)
+		pf->tmp_cfg.app[j] = old_cfg->app[j + 1];
+
+	ret = i40e_hw_dcb_config(pf, &pf->tmp_cfg);
+	if (ret) {
+		dev_info(&pf->pdev->dev,
+			 "Failed setting DCB configuration err %pe aq_err %s\n",
+			 ERR_PTR(ret),
+			 libie_aq_str(pf->hw.aq.asq_last_status));
+		return -EINVAL;
+	}
+
+	return 0;
+}
+
+/**
+ * i40e_dcbnl_getstate - Get DCB enabled state
+ * @netdev: the corresponding netdev
+ *
+ * Get the current DCB enabled state
+ **/
+static u8 i40e_dcbnl_getstate(struct net_device *netdev)
+{
+	struct i40e_pf *pf = i40e_netdev_to_pf(netdev);
+
+	dev_dbg(&pf->pdev->dev, "DCB state=%d\n",
+		test_bit(I40E_FLAG_DCB_ENA, pf->flags) ? 1 : 0);
+	return test_bit(I40E_FLAG_DCB_ENA, pf->flags) ? 1 : 0;
+}
+
+/**
+ * i40e_dcbnl_setstate - Set DCB state
+ * @netdev: the corresponding netdev
+ * @state: enable or disable
+ *
+ * Set the DCB state
+ **/
+static u8 i40e_dcbnl_setstate(struct net_device *netdev, u8 state)
+{
+	struct i40e_pf *pf = i40e_netdev_to_pf(netdev);
+	int ret = I40E_DCBNL_STATUS_SUCCESS;
+
+	if (!(pf->dcbx_cap & DCB_CAP_DCBX_VER_CEE) ||
+	    (pf->dcbx_cap & DCB_CAP_DCBX_LLD_MANAGED))
+		return ret;
+
+	dev_dbg(&pf->pdev->dev, "new state=%d current state=%d\n",
+		state, test_bit(I40E_FLAG_DCB_ENA, pf->flags) ? 1 : 0);
+	/* Nothing to do */
+	if (!state == !test_bit(I40E_FLAG_DCB_ENA, pf->flags))
+		return ret;
+
+	if (i40e_is_sw_dcb(pf)) {
+		if (state) {
+			set_bit(I40E_FLAG_DCB_ENA, pf->flags);
+			memcpy(&pf->hw.desired_dcbx_config,
+			       &pf->hw.local_dcbx_config,
+			       sizeof(struct i40e_dcbx_config));
+		} else {
+			clear_bit(I40E_FLAG_DCB_ENA, pf->flags);
+		}
+	} else {
+		/* Cannot directly manipulate FW LLDP Agent */
+		ret = I40E_DCBNL_STATUS_ERROR;
+	}
+	return ret;
+}
+
+/**
+ * i40e_dcbnl_set_pg_tc_cfg_tx - Set CEE PG Tx config
+ * @netdev: the corresponding netdev
+ * @tc: the corresponding traffic class
+ * @prio_type: the traffic priority type
+ * @bwg_id: the BW group id the traffic class belongs to
+ * @bw_pct: the BW percentage for the corresponding BWG
+ * @up_map: prio mapped to corresponding tc
+ *
+ * Set Tx PG settings for CEE mode
+ **/
+static void i40e_dcbnl_set_pg_tc_cfg_tx(struct net_device *netdev, int tc,
+					u8 prio_type, u8 bwg_id, u8 bw_pct,
+					u8 up_map)
+{
+	struct i40e_pf *pf = i40e_netdev_to_pf(netdev);
+	int i;
+
+	if (!(pf->dcbx_cap & DCB_CAP_DCBX_VER_CEE) ||
+	    (pf->dcbx_cap & DCB_CAP_DCBX_LLD_MANAGED))
+		return;
+
+	/* LLTC not supported yet */
+	if (tc >= I40E_MAX_TRAFFIC_CLASS)
+		return;
+
+	/* prio_type, bwg_id and bw_pct per UP are not supported */
+
+	/* Use only up_map to map tc */
+	for (i = 0; i < I40E_MAX_TRAFFIC_CLASS; i++) {
+		if (up_map & BIT(i))
+			pf->tmp_cfg.etscfg.prioritytable[i] = tc;
+	}
+	pf->tmp_cfg.etscfg.tsatable[tc] = I40E_IEEE_TSA_ETS;
+	dev_dbg(&pf->pdev->dev,
+		"Set PG config tc=%d bwg_id=%d prio_type=%d bw_pct=%d up_map=%d\n",
+		tc, bwg_id, prio_type, bw_pct, up_map);
+}
+
+/**
+ * i40e_dcbnl_set_pg_bwg_cfg_tx - Set CEE PG Tx BW config
+ * @netdev: the corresponding netdev
+ * @pgid: the corresponding traffic class
+ * @bw_pct: the BW percentage for the specified traffic class
+ *
+ * Set Tx BW settings for CEE mode
+ **/
+static void i40e_dcbnl_set_pg_bwg_cfg_tx(struct net_device *netdev, int pgid,
+					 u8 bw_pct)
+{
+	struct i40e_pf *pf = i40e_netdev_to_pf(netdev);
+
+	if (!(pf->dcbx_cap & DCB_CAP_DCBX_VER_CEE) ||
+	    (pf->dcbx_cap & DCB_CAP_DCBX_LLD_MANAGED))
+		return;
+
+	/* LLTC not supported yet */
+	if (pgid >= I40E_MAX_TRAFFIC_CLASS)
+		return;
+
+	pf->tmp_cfg.etscfg.tcbwtable[pgid] = bw_pct;
+	dev_dbg(&pf->pdev->dev, "Set PG BW config tc=%d bw_pct=%d\n",
+		pgid, bw_pct);
+}
+
+/**
+ * i40e_dcbnl_set_pg_tc_cfg_rx - Set CEE PG Rx config
+ * @netdev: the corresponding netdev
+ * @prio: the corresponding traffic class
+ * @prio_type: the traffic priority type
+ * @pgid: the BW group id the traffic class belongs to
+ * @bw_pct: the BW percentage for the corresponding BWG
+ * @up_map: prio mapped to corresponding tc
+ *
+ * Set Rx BW settings for CEE mode. The hardware does not support this
+ * so we won't allow setting of this parameter.
+ **/
+static void i40e_dcbnl_set_pg_tc_cfg_rx(struct net_device *netdev,
+					int __always_unused prio,
+					u8 __always_unused prio_type,
+					u8 __always_unused pgid,
+					u8 __always_unused bw_pct,
+					u8 __always_unused up_map)
+{
+	struct i40e_pf *pf = i40e_netdev_to_pf(netdev);
+
+	dev_dbg(&pf->pdev->dev, "Rx TC PG Config Not Supported.\n");
+}
+
+/**
+ * i40e_dcbnl_set_pg_bwg_cfg_rx - Set CEE PG Rx config
+ * @netdev: the corresponding netdev
+ * @pgid: the corresponding traffic class
+ * @bw_pct: the BW percentage for the specified traffic class
+ *
+ * Set Rx BW settings for CEE mode. The hardware does not support this
+ * so we won't allow setting of this parameter.
+ **/
+static void i40e_dcbnl_set_pg_bwg_cfg_rx(struct net_device *netdev, int pgid,
+					 u8 bw_pct)
+{
+	struct i40e_pf *pf = i40e_netdev_to_pf(netdev);
+
+	dev_dbg(&pf->pdev->dev, "Rx BWG PG Config Not Supported.\n");
+}
+
+/**
+ * i40e_dcbnl_get_pg_tc_cfg_tx - Get CEE PG Tx config
+ * @netdev: the corresponding netdev
+ * @prio: the corresponding user priority
+ * @prio_type: traffic priority type
+ * @pgid: the BW group ID the traffic class belongs to
+ * @bw_pct: BW percentage for the corresponding BWG
+ * @up_map: prio mapped to corresponding TC
+ *
+ * Get Tx PG settings for CEE mode
+ **/
+static void i40e_dcbnl_get_pg_tc_cfg_tx(struct net_device *netdev, int prio,
+					u8 __always_unused *prio_type,
+					u8 *pgid,
+					u8 __always_unused *bw_pct,
+					u8 __always_unused *up_map)
+{
+	struct i40e_pf *pf = i40e_netdev_to_pf(netdev);
+
+	if (!(pf->dcbx_cap & DCB_CAP_DCBX_VER_CEE) ||
+	    (pf->dcbx_cap & DCB_CAP_DCBX_LLD_MANAGED))
+		return;
+
+	if (prio >= I40E_MAX_USER_PRIORITY)
+		return;
+
+	*pgid = pf->hw.local_dcbx_config.etscfg.prioritytable[prio];
+	dev_dbg(&pf->pdev->dev, "Get PG config prio=%d tc=%d\n",
+		prio, *pgid);
+}
+
+/**
+ * i40e_dcbnl_get_pg_bwg_cfg_tx - Get CEE PG BW config
+ * @netdev: the corresponding netdev
+ * @pgid: the corresponding traffic class
+ * @bw_pct: the BW percentage for the corresponding TC
+ *
+ * Get Tx BW settings for given TC in CEE mode
+ **/
+static void i40e_dcbnl_get_pg_bwg_cfg_tx(struct net_device *netdev, int pgid,
+					 u8 *bw_pct)
+{
+	struct i40e_pf *pf = i40e_netdev_to_pf(netdev);
+
+	if (!(pf->dcbx_cap & DCB_CAP_DCBX_VER_CEE) ||
+	    (pf->dcbx_cap & DCB_CAP_DCBX_LLD_MANAGED))
+		return;
+
+	if (pgid >= I40E_MAX_TRAFFIC_CLASS)
+		return;
+
+	*bw_pct = pf->hw.local_dcbx_config.etscfg.tcbwtable[pgid];
+	dev_dbg(&pf->pdev->dev, "Get PG BW config tc=%d bw_pct=%d\n",
+		pgid, *bw_pct);
+}
+
+/**
+ * i40e_dcbnl_get_pg_tc_cfg_rx - Get CEE PG Rx config
+ * @netdev: the corresponding netdev
+ * @prio: the corresponding user priority
+ * @prio_type: the traffic priority type
+ * @pgid: the PG ID
+ * @bw_pct: the BW percentage for the corresponding BWG
+ * @up_map: prio mapped to corresponding TC
+ *
+ * Get Rx PG settings for CEE mode. The UP2TC map is applied in same
+ * manner for Tx and Rx (symmetrical) so return the TC information for
+ * given priority accordingly.
+ **/
+static void i40e_dcbnl_get_pg_tc_cfg_rx(struct net_device *netdev, int prio,
+					u8 *prio_type, u8 *pgid, u8 *bw_pct,
+					u8 *up_map)
+{
+	struct i40e_pf *pf = i40e_netdev_to_pf(netdev);
+
+	if (!(pf->dcbx_cap & DCB_CAP_DCBX_VER_CEE) ||
+	    (pf->dcbx_cap & DCB_CAP_DCBX_LLD_MANAGED))
+		return;
+
+	if (prio >= I40E_MAX_USER_PRIORITY)
+		return;
+
+	*pgid = pf->hw.local_dcbx_config.etscfg.prioritytable[prio];
+}
+
+/**
+ * i40e_dcbnl_get_pg_bwg_cfg_rx - Get CEE PG BW Rx config
+ * @netdev: the corresponding netdev
+ * @pgid: the corresponding traffic class
+ * @bw_pct: the BW percentage for the corresponding TC
+ *
+ * Get Rx BW settings for given TC in CEE mode
+ * The adapter doesn't support Rx ETS and runs in strict priority
+ * mode in Rx path and hence just return 0.
+ **/
+static void i40e_dcbnl_get_pg_bwg_cfg_rx(struct net_device *netdev, int pgid,
+					 u8 *bw_pct)
+{
+	struct i40e_pf *pf = i40e_netdev_to_pf(netdev);
+
+	if (!(pf->dcbx_cap & DCB_CAP_DCBX_VER_CEE) ||
+	    (pf->dcbx_cap & DCB_CAP_DCBX_LLD_MANAGED))
+		return;
+	*bw_pct = 0;
+}
+
+/**
+ * i40e_dcbnl_set_pfc_cfg - Set CEE PFC configuration
+ * @netdev: the corresponding netdev
+ * @prio: the corresponding user priority
+ * @setting: the PFC setting for given priority
+ *
+ * Set the PFC enabled/disabled setting for given user priority
+ **/
+static void i40e_dcbnl_set_pfc_cfg(struct net_device *netdev, int prio,
+				   u8 setting)
+{
+	struct i40e_pf *pf = i40e_netdev_to_pf(netdev);
+
+	if (!(pf->dcbx_cap & DCB_CAP_DCBX_VER_CEE) ||
+	    (pf->dcbx_cap & DCB_CAP_DCBX_LLD_MANAGED))
+		return;
+
+	if (prio >= I40E_MAX_USER_PRIORITY)
+		return;
+
+	pf->tmp_cfg.pfc.pfccap = I40E_MAX_TRAFFIC_CLASS;
+	if (setting)
+		pf->tmp_cfg.pfc.pfcenable |= BIT(prio);
+	else
+		pf->tmp_cfg.pfc.pfcenable &= ~BIT(prio);
+	dev_dbg(&pf->pdev->dev,
+		"Set PFC Config up=%d setting=%d pfcenable=0x%x\n",
+		prio, setting, pf->tmp_cfg.pfc.pfcenable);
+}
+
+/**
+ * i40e_dcbnl_get_pfc_cfg - Get CEE PFC configuration
+ * @netdev: the corresponding netdev
+ * @prio: the corresponding user priority
+ * @setting: the PFC setting for given priority
+ *
+ * Get the PFC enabled/disabled setting for given user priority
+ **/
+static void i40e_dcbnl_get_pfc_cfg(struct net_device *netdev, int prio,
+				   u8 *setting)
+{
+	struct i40e_pf *pf = i40e_netdev_to_pf(netdev);
+
+	if (!(pf->dcbx_cap & DCB_CAP_DCBX_VER_CEE) ||
+	    (pf->dcbx_cap & DCB_CAP_DCBX_LLD_MANAGED))
+		return;
+
+	if (prio >= I40E_MAX_USER_PRIORITY)
+		return;
+
+	*setting = (pf->hw.local_dcbx_config.pfc.pfcenable >> prio) & 0x1;
+	dev_dbg(&pf->pdev->dev,
+		"Get PFC Config up=%d setting=%d pfcenable=0x%x\n",
+		prio, *setting, pf->hw.local_dcbx_config.pfc.pfcenable);
+}
+
+/**
+ * i40e_dcbnl_cee_set_all - Commit CEE DCB settings to hardware
+ * @netdev: the corresponding netdev
+ *
+ * Commit the current DCB configuration to hardware
+ **/
+static u8 i40e_dcbnl_cee_set_all(struct net_device *netdev)
+{
+	struct i40e_pf *pf = i40e_netdev_to_pf(netdev);
+	int err;
+
+	if (!(pf->dcbx_cap & DCB_CAP_DCBX_VER_CEE) ||
+	    (pf->dcbx_cap & DCB_CAP_DCBX_LLD_MANAGED))
+		return I40E_DCBNL_STATUS_ERROR;
+
+	dev_dbg(&pf->pdev->dev, "Commit DCB Configuration to the hardware\n");
+	err = i40e_hw_dcb_config(pf, &pf->tmp_cfg);
+
+	return err ? I40E_DCBNL_STATUS_ERROR : I40E_DCBNL_STATUS_SUCCESS;
+}
+
+/**
+ * i40e_dcbnl_get_cap - Get DCBX capabilities of adapter
+ * @netdev: the corresponding netdev
+ * @capid: the capability type
+ * @cap: the capability value
+ *
+ * Return the capability value for a given capability type
+ **/
+static u8 i40e_dcbnl_get_cap(struct net_device *netdev, int capid, u8 *cap)
+{
+	struct i40e_pf *pf = i40e_netdev_to_pf(netdev);
+
+	if (!test_bit(I40E_FLAG_DCB_CAPABLE, pf->flags))
+		return I40E_DCBNL_STATUS_ERROR;
+
+	switch (capid) {
+	case DCB_CAP_ATTR_PG:
+	case DCB_CAP_ATTR_PFC:
+		*cap = true;
+		break;
+	case DCB_CAP_ATTR_PG_TCS:
+	case DCB_CAP_ATTR_PFC_TCS:
+		*cap = 0x80;
+		break;
+	case DCB_CAP_ATTR_DCBX:
+		*cap = pf->dcbx_cap;
+		break;
+	case DCB_CAP_ATTR_UP2TC:
+	case DCB_CAP_ATTR_GSP:
+	case DCB_CAP_ATTR_BCN:
+	default:
+		*cap = false;
+		break;
+	}
+
+	dev_dbg(&pf->pdev->dev, "Get Capability cap=%d capval=0x%x\n",
+		capid, *cap);
+	return I40E_DCBNL_STATUS_SUCCESS;
+}
+
+/**
+ * i40e_dcbnl_getnumtcs - Get max number of traffic classes supported
+ * @netdev: the corresponding netdev
+ * @tcid: the TC id
+ * @num: total number of TCs supported by the device
+ *
+ * Return the total number of TCs supported by the adapter
+ **/
+static int i40e_dcbnl_getnumtcs(struct net_device *netdev, int tcid, u8 *num)
+{
+	struct i40e_pf *pf = i40e_netdev_to_pf(netdev);
+
+	if (!test_bit(I40E_FLAG_DCB_CAPABLE, pf->flags))
+		return -EINVAL;
+
+	*num = I40E_MAX_TRAFFIC_CLASS;
+	return 0;
+}
+
+/**
+ * i40e_dcbnl_setnumtcs - Set CEE number of traffic classes
+ * @netdev: the corresponding netdev
+ * @tcid: the TC id
+ * @num: total number of TCs
+ *
+ * Set the total number of TCs (Unsupported)
+ **/
+static int i40e_dcbnl_setnumtcs(struct net_device *netdev, int tcid, u8 num)
+{
+	return -EINVAL;
+}
+
+/**
+ * i40e_dcbnl_getpfcstate - Get CEE PFC mode
+ * @netdev: the corresponding netdev
+ *
+ * Get the current PFC enabled state
+ **/
+static u8 i40e_dcbnl_getpfcstate(struct net_device *netdev)
+{
+	struct i40e_pf *pf = i40e_netdev_to_pf(netdev);
+
+	/* Return enabled if any PFC enabled UP */
+	if (pf->hw.local_dcbx_config.pfc.pfcenable)
+		return 1;
+	else
+		return 0;
+}
+
+/**
+ * i40e_dcbnl_setpfcstate - Set CEE PFC mode
+ * @netdev: the corresponding netdev
+ * @state: required state
+ *
+ * The PFC state to be set; this is enabled/disabled based on the PFC
+ * priority settings and not via this call for i40e driver
+ **/
+static void i40e_dcbnl_setpfcstate(struct net_device *netdev, u8 state)
+{
+	struct i40e_pf *pf = i40e_netdev_to_pf(netdev);
+
+	dev_dbg(&pf->pdev->dev, "PFC State is modified via PFC config.\n");
+}
+
+/**
+ * i40e_dcbnl_getapp - Get CEE APP
+ * @netdev: the corresponding netdev
+ * @idtype: the App selector
+ * @id: the App ethtype or port number
+ *
+ * Return the CEE mode app for the given idtype and id
+ **/
+static int i40e_dcbnl_getapp(struct net_device *netdev, u8 idtype, u16 id)
+{
+	struct i40e_pf *pf = i40e_netdev_to_pf(netdev);
+	struct dcb_app app = {
+				.selector = idtype,
+				.protocol = id,
+			     };
+
+	if (!(pf->dcbx_cap & DCB_CAP_DCBX_VER_CEE) ||
+	    (pf->dcbx_cap & DCB_CAP_DCBX_LLD_MANAGED))
+		return -EINVAL;
+
+	return dcb_getapp(netdev, &app);
+}
+
+/**
+ * i40e_dcbnl_setdcbx - set required DCBx capability
+ * @netdev: the corresponding netdev
+ * @mode: new DCB mode managed or CEE+IEEE
+ *
+ * Set DCBx capability features
+ **/
+static u8 i40e_dcbnl_setdcbx(struct net_device *netdev, u8 mode)
+{
+	struct i40e_pf *pf = i40e_netdev_to_pf(netdev);
+
+	/* Do not allow to set mode if managed by Firmware */
+	if (pf->dcbx_cap & DCB_CAP_DCBX_LLD_MANAGED)
+		return I40E_DCBNL_STATUS_ERROR;
+
+	/* No support for LLD_MANAGED modes or CEE+IEEE */
+	if ((mode & DCB_CAP_DCBX_LLD_MANAGED) ||
+	    ((mode & DCB_CAP_DCBX_VER_IEEE) && (mode & DCB_CAP_DCBX_VER_CEE)) ||
+	    !(mode & DCB_CAP_DCBX_HOST))
+		return I40E_DCBNL_STATUS_ERROR;
+
+	/* Already set to the given mode no change */
+	if (mode == pf->dcbx_cap)
+		return I40E_DCBNL_STATUS_SUCCESS;
+
+	pf->dcbx_cap = mode;
+	if (mode & DCB_CAP_DCBX_VER_CEE)
+		pf->hw.local_dcbx_config.dcbx_mode = I40E_DCBX_MODE_CEE;
+	else
+		pf->hw.local_dcbx_config.dcbx_mode = I40E_DCBX_MODE_IEEE;
+
+	dev_dbg(&pf->pdev->dev, "mode=%d\n", mode);
+	return I40E_DCBNL_STATUS_SUCCESS;
+}
+
+/**
+ * i40e_dcbnl_getdcbx - retrieve current DCBx capability
+ * @dev: the corresponding netdev
  *
  * Returns DCBx capability features
  **/
@@ -131,7 +817,8 @@ static u8 i40e_dcbnl_getdcbx(struct net_device *dev)
 
 /**
  * i40e_dcbnl_get_perm_hw_addr - MAC address used by DCBx
- * @netdev: the corresponding netdev
+ * @dev: the corresponding netdev
+ * @perm_addr: buffer to store the MAC address
  *
  * Returns the SAN MAC address used for LLDP exchange
  **/
@@ -139,22 +826,43 @@ static void i40e_dcbnl_get_perm_hw_addr(struct net_device *dev,
 					u8 *perm_addr)
 {
 	struct i40e_pf *pf = i40e_netdev_to_pf(dev);
-	int i, j;
+	int i;
 
 	memset(perm_addr, 0xff, MAX_ADDR_LEN);
 
 	for (i = 0; i < dev->addr_len; i++)
 		perm_addr[i] = pf->hw.mac.perm_addr[i];
-
-	for (j = 0; j < dev->addr_len; j++, i++)
-		perm_addr[i] = pf->hw.mac.san_addr[j];
 }
 
 static const struct dcbnl_rtnl_ops dcbnl_ops = {
 	.ieee_getets	= i40e_dcbnl_ieee_getets,
 	.ieee_getpfc	= i40e_dcbnl_ieee_getpfc,
 	.getdcbx	= i40e_dcbnl_getdcbx,
-	.getpermhwaddr  = i40e_dcbnl_get_perm_hw_addr,
+	.getpermhwaddr	= i40e_dcbnl_get_perm_hw_addr,
+	.ieee_setets	= i40e_dcbnl_ieee_setets,
+	.ieee_setpfc	= i40e_dcbnl_ieee_setpfc,
+	.ieee_setapp	= i40e_dcbnl_ieee_setapp,
+	.ieee_delapp	= i40e_dcbnl_ieee_delapp,
+	.getstate	= i40e_dcbnl_getstate,
+	.setstate	= i40e_dcbnl_setstate,
+	.setpgtccfgtx	= i40e_dcbnl_set_pg_tc_cfg_tx,
+	.setpgbwgcfgtx	= i40e_dcbnl_set_pg_bwg_cfg_tx,
+	.setpgtccfgrx	= i40e_dcbnl_set_pg_tc_cfg_rx,
+	.setpgbwgcfgrx	= i40e_dcbnl_set_pg_bwg_cfg_rx,
+	.getpgtccfgtx	= i40e_dcbnl_get_pg_tc_cfg_tx,
+	.getpgbwgcfgtx	= i40e_dcbnl_get_pg_bwg_cfg_tx,
+	.getpgtccfgrx	= i40e_dcbnl_get_pg_tc_cfg_rx,
+	.getpgbwgcfgrx	= i40e_dcbnl_get_pg_bwg_cfg_rx,
+	.setpfccfg	= i40e_dcbnl_set_pfc_cfg,
+	.getpfccfg	= i40e_dcbnl_get_pfc_cfg,
+	.setall		= i40e_dcbnl_cee_set_all,
+	.getcap		= i40e_dcbnl_get_cap,
+	.getnumtcs	= i40e_dcbnl_getnumtcs,
+	.setnumtcs	= i40e_dcbnl_setnumtcs,
+	.getpfcstate	= i40e_dcbnl_getpfcstate,
+	.setpfcstate	= i40e_dcbnl_setpfcstate,
+	.getapp		= i40e_dcbnl_getapp,
+	.setdcbx	= i40e_dcbnl_setdcbx,
 };
 
 /**
@@ -174,12 +882,16 @@ void i40e_dcbnl_set_all(struct i40e_vsi *vsi)
 	u8 prio, tc_map;
 	int i;
 
+	/* SW DCB taken care by DCBNL set calls */
+	if (pf->dcbx_cap & DCB_CAP_DCBX_HOST)
+		return;
+
 	/* DCB not enabled */
-	if (!(pf->flags & I40E_FLAG_DCB_ENABLED))
+	if (!test_bit(I40E_FLAG_DCB_ENA, pf->flags))
 		return;
 
 	/* MFP mode but not an iSCSI PF so return */
-	if ((pf->flags & I40E_FLAG_MFP_ENABLED) && !(pf->hw.func_caps.iscsi))
+	if (test_bit(I40E_FLAG_MFP_ENA, pf->flags) && !(hw->func_caps.iscsi))
 		return;
 
 	dcbxcfg = &hw->local_dcbx_config;
@@ -235,16 +947,16 @@ static int i40e_dcbnl_vsi_del_app(struct i40e_vsi *vsi,
 static void i40e_dcbnl_del_app(struct i40e_pf *pf,
 			       struct i40e_dcb_app_priority_table *app)
 {
+	struct i40e_vsi *vsi;
 	int v, err;
 
-	for (v = 0; v < pf->num_alloc_vsi; v++) {
-		if (pf->vsi[v] && pf->vsi[v]->netdev) {
-			err = i40e_dcbnl_vsi_del_app(pf->vsi[v], app);
+	i40e_pf_for_each_vsi(pf, v, vsi)
+		if (vsi->netdev) {
+			err = i40e_dcbnl_vsi_del_app(vsi, app);
 			dev_dbg(&pf->pdev->dev, "Deleting app for VSI seid=%d err=%d sel=%d proto=0x%x prio=%d\n",
-				pf->vsi[v]->seid, err, app->selector,
+				vsi->seid, err, app->selector,
 				app->protocolid, app->priority);
 		}
-	}
 }
 
 /**
@@ -286,7 +998,7 @@ void i40e_dcbnl_flush_apps(struct i40e_pf *pf,
 	int i;
 
 	/* MFP mode but not an iSCSI PF so return */
-	if ((pf->flags & I40E_FLAG_MFP_ENABLED) && !(pf->hw.func_caps.iscsi))
+	if (test_bit(I40E_FLAG_MFP_ENA, pf->flags) && !(pf->hw.func_caps.iscsi))
 		return;
 
 	for (i = 0; i < old_cfg->numapps; i++) {
@@ -309,7 +1021,7 @@ void i40e_dcbnl_setup(struct i40e_vsi *vsi)
 	struct i40e_pf *pf = i40e_netdev_to_pf(dev);
 
 	/* Not DCB capable */
-	if (!(pf->flags & I40E_FLAG_DCB_CAPABLE))
+	if (!test_bit(I40E_FLAG_DCB_CAPABLE, pf->flags))
 		return;
 
 	dev->dcbnl_ops = &dcbnl_ops;
diff --git a/drivers/net/ethernet/intel/i40e/i40e_ddp.c b/drivers/net/ethernet/intel/i40e/i40e_ddp.c
new file mode 100644
index 000000000000..daa9f2c42f70
--- /dev/null
+++ b/drivers/net/ethernet/intel/i40e/i40e_ddp.c
@@ -0,0 +1,499 @@
+// SPDX-License-Identifier: GPL-2.0
+/* Copyright(c) 2013 - 2018 Intel Corporation. */
+
+#include <linux/firmware.h>
+#include "i40e.h"
+
+#define I40_DDP_FLASH_REGION		100
+#define I40E_PROFILE_INFO_SIZE		48
+#define I40E_MAX_PROFILE_NUM		16
+#define I40E_PROFILE_LIST_SIZE		\
+	(I40E_PROFILE_INFO_SIZE * I40E_MAX_PROFILE_NUM + 4)
+#define I40E_DDP_PROFILE_PATH		"intel/i40e/ddp/"
+#define I40E_DDP_PROFILE_NAME_MAX	64
+
+struct i40e_ddp_profile_list {
+	u32 p_count;
+	struct i40e_profile_info p_info[];
+};
+
+struct i40e_ddp_old_profile_list {
+	struct list_head list;
+	size_t old_ddp_size;
+	u8 old_ddp_buf[];
+};
+
+/**
+ * i40e_ddp_profiles_eq - checks if DDP profiles are the equivalent
+ * @a: new profile info
+ * @b: old profile info
+ *
+ * checks if DDP profiles are the equivalent.
+ * Returns true if profiles are the same.
+ **/
+static bool i40e_ddp_profiles_eq(struct i40e_profile_info *a,
+				 struct i40e_profile_info *b)
+{
+	return a->track_id == b->track_id &&
+		!memcmp(&a->version, &b->version, sizeof(a->version)) &&
+		!memcmp(&a->name, &b->name, I40E_DDP_NAME_SIZE);
+}
+
+/**
+ * i40e_ddp_does_profile_exist - checks if DDP profile loaded already
+ * @hw: HW data structure
+ * @pinfo: DDP profile information structure
+ *
+ * checks if DDP profile loaded already.
+ * Returns >0 if the profile exists.
+ * Returns  0 if the profile is absent.
+ * Returns <0 if error.
+ **/
+static int i40e_ddp_does_profile_exist(struct i40e_hw *hw,
+				       struct i40e_profile_info *pinfo)
+{
+	struct i40e_ddp_profile_list *profile_list;
+	u8 buff[I40E_PROFILE_LIST_SIZE];
+	int status;
+	int i;
+
+	status = i40e_aq_get_ddp_list(hw, buff, I40E_PROFILE_LIST_SIZE, 0,
+				      NULL);
+	if (status)
+		return -1;
+
+	profile_list = (struct i40e_ddp_profile_list *)buff;
+	for (i = 0; i < profile_list->p_count; i++) {
+		if (i40e_ddp_profiles_eq(pinfo, &profile_list->p_info[i]))
+			return 1;
+	}
+	return 0;
+}
+
+/**
+ * i40e_ddp_profiles_overlap - checks if DDP profiles overlap.
+ * @new: new profile info
+ * @old: old profile info
+ *
+ * checks if DDP profiles overlap.
+ * Returns true if profiles are overlap.
+ **/
+static bool i40e_ddp_profiles_overlap(struct i40e_profile_info *new,
+				      struct i40e_profile_info *old)
+{
+	unsigned int group_id_old = FIELD_GET(0x00FF0000, old->track_id);
+	unsigned int group_id_new = FIELD_GET(0x00FF0000, new->track_id);
+
+	/* 0x00 group must be only the first */
+	if (group_id_new == 0)
+		return true;
+	/* 0xFF group is compatible with anything else */
+	if (group_id_new == 0xFF || group_id_old == 0xFF)
+		return false;
+	/* otherwise only profiles from the same group are compatible*/
+	return group_id_old != group_id_new;
+}
+
+/**
+ * i40e_ddp_does_profile_overlap - checks if DDP overlaps with existing one.
+ * @hw: HW data structure
+ * @pinfo: DDP profile information structure
+ *
+ * checks if DDP profile overlaps with existing one.
+ * Returns >0 if the profile overlaps.
+ * Returns  0 if the profile is ok.
+ * Returns <0 if error.
+ **/
+static int i40e_ddp_does_profile_overlap(struct i40e_hw *hw,
+					 struct i40e_profile_info *pinfo)
+{
+	struct i40e_ddp_profile_list *profile_list;
+	u8 buff[I40E_PROFILE_LIST_SIZE];
+	int status;
+	int i;
+
+	status = i40e_aq_get_ddp_list(hw, buff, I40E_PROFILE_LIST_SIZE, 0,
+				      NULL);
+	if (status)
+		return -EIO;
+
+	profile_list = (struct i40e_ddp_profile_list *)buff;
+	for (i = 0; i < profile_list->p_count; i++) {
+		if (i40e_ddp_profiles_overlap(pinfo,
+					      &profile_list->p_info[i]))
+			return 1;
+	}
+	return 0;
+}
+
+/**
+ * i40e_add_pinfo
+ * @hw: pointer to the hardware structure
+ * @profile: pointer to the profile segment of the package
+ * @profile_info_sec: buffer for information section
+ * @track_id: package tracking id
+ *
+ * Register a profile to the list of loaded profiles.
+ */
+static int
+i40e_add_pinfo(struct i40e_hw *hw, struct i40e_profile_segment *profile,
+	       u8 *profile_info_sec, u32 track_id)
+{
+	struct i40e_profile_section_header *sec;
+	struct i40e_profile_info *pinfo;
+	u32 offset = 0, info = 0;
+	int status;
+
+	sec = (struct i40e_profile_section_header *)profile_info_sec;
+	sec->tbl_size = 1;
+	sec->data_end = sizeof(struct i40e_profile_section_header) +
+			sizeof(struct i40e_profile_info);
+	sec->section.type = SECTION_TYPE_INFO;
+	sec->section.offset = sizeof(struct i40e_profile_section_header);
+	sec->section.size = sizeof(struct i40e_profile_info);
+	pinfo = (struct i40e_profile_info *)(profile_info_sec +
+					     sec->section.offset);
+	pinfo->track_id = track_id;
+	pinfo->version = profile->version;
+	pinfo->op = I40E_DDP_ADD_TRACKID;
+
+	/* Clear reserved field */
+	memset(pinfo->reserved, 0, sizeof(pinfo->reserved));
+	memcpy(pinfo->name, profile->name, I40E_DDP_NAME_SIZE);
+
+	status = i40e_aq_write_ddp(hw, (void *)sec, sec->data_end,
+				   track_id, &offset, &info, NULL);
+	return status;
+}
+
+/**
+ * i40e_del_pinfo - delete DDP profile info from NIC
+ * @hw: HW data structure
+ * @profile: DDP profile segment to be deleted
+ * @profile_info_sec: DDP profile section header
+ * @track_id: track ID of the profile for deletion
+ *
+ * Removes DDP profile from the NIC.
+ **/
+static int
+i40e_del_pinfo(struct i40e_hw *hw, struct i40e_profile_segment *profile,
+	       u8 *profile_info_sec, u32 track_id)
+{
+	struct i40e_profile_section_header *sec;
+	struct i40e_profile_info *pinfo;
+	u32 offset = 0, info = 0;
+	int status;
+
+	sec = (struct i40e_profile_section_header *)profile_info_sec;
+	sec->tbl_size = 1;
+	sec->data_end = sizeof(struct i40e_profile_section_header) +
+			sizeof(struct i40e_profile_info);
+	sec->section.type = SECTION_TYPE_INFO;
+	sec->section.offset = sizeof(struct i40e_profile_section_header);
+	sec->section.size = sizeof(struct i40e_profile_info);
+	pinfo = (struct i40e_profile_info *)(profile_info_sec +
+					     sec->section.offset);
+	pinfo->track_id = track_id;
+	pinfo->version = profile->version;
+	pinfo->op = I40E_DDP_REMOVE_TRACKID;
+
+	/* Clear reserved field */
+	memset(pinfo->reserved, 0, sizeof(pinfo->reserved));
+	memcpy(pinfo->name, profile->name, I40E_DDP_NAME_SIZE);
+
+	status = i40e_aq_write_ddp(hw, (void *)sec, sec->data_end,
+				   track_id, &offset, &info, NULL);
+	return status;
+}
+
+/**
+ * i40e_ddp_is_pkg_hdr_valid - performs basic pkg header integrity checks
+ * @netdev: net device structure (for logging purposes)
+ * @pkg_hdr: pointer to package header
+ * @size_huge: size of the whole DDP profile package in size_t
+ *
+ * Checks correctness of pkg header: Version, size too big/small, and
+ * all segment offsets alignment and boundaries. This function lets
+ * reject non DDP profile file to be loaded by administrator mistake.
+ **/
+static bool i40e_ddp_is_pkg_hdr_valid(struct net_device *netdev,
+				      struct i40e_package_header *pkg_hdr,
+				      size_t size_huge)
+{
+	u32 size = 0xFFFFFFFFU & size_huge;
+	u32 pkg_hdr_size;
+	u32 segment;
+
+	if (!pkg_hdr)
+		return false;
+
+	if (pkg_hdr->version.major > 0) {
+		struct i40e_ddp_version ver = pkg_hdr->version;
+
+		netdev_err(netdev, "Unsupported DDP profile version %u.%u.%u.%u",
+			   ver.major, ver.minor, ver.update, ver.draft);
+		return false;
+	}
+	if (size_huge > size) {
+		netdev_err(netdev, "Invalid DDP profile - size is bigger than 4G");
+		return false;
+	}
+	if (size < (sizeof(struct i40e_package_header) + sizeof(u32) +
+		sizeof(struct i40e_metadata_segment) + sizeof(u32) * 2)) {
+		netdev_err(netdev, "Invalid DDP profile - size is too small.");
+		return false;
+	}
+
+	pkg_hdr_size = sizeof(u32) * (pkg_hdr->segment_count + 2U);
+	if (size < pkg_hdr_size) {
+		netdev_err(netdev, "Invalid DDP profile - too many segments");
+		return false;
+	}
+	for (segment = 0; segment < pkg_hdr->segment_count; ++segment) {
+		u32 offset = pkg_hdr->segment_offset[segment];
+
+		if (0xFU & offset) {
+			netdev_err(netdev,
+				   "Invalid DDP profile %u segment alignment",
+				   segment);
+			return false;
+		}
+		if (pkg_hdr_size > offset || offset >= size) {
+			netdev_err(netdev,
+				   "Invalid DDP profile %u segment offset",
+				   segment);
+			return false;
+		}
+	}
+
+	return true;
+}
+
+/**
+ * i40e_ddp_load - performs DDP loading
+ * @netdev: net device structure
+ * @data: buffer containing recipe file
+ * @size: size of the buffer
+ * @is_add: true when loading profile, false when rolling back the previous one
+ *
+ * Checks correctness and loads DDP profile to the NIC. The function is
+ * also used for rolling back previously loaded profile.
+ **/
+static int i40e_ddp_load(struct net_device *netdev, const u8 *data, size_t size,
+			 bool is_add)
+{
+	u8 profile_info_sec[sizeof(struct i40e_profile_section_header) +
+			    sizeof(struct i40e_profile_info)];
+	struct i40e_metadata_segment *metadata_hdr;
+	struct i40e_profile_segment *profile_hdr;
+	struct i40e_profile_info pinfo;
+	struct i40e_package_header *pkg_hdr;
+	struct i40e_netdev_priv *np = netdev_priv(netdev);
+	struct i40e_vsi *vsi = np->vsi;
+	struct i40e_pf *pf = vsi->back;
+	u32 track_id;
+	int istatus;
+	int status;
+
+	pkg_hdr = (struct i40e_package_header *)data;
+	if (!i40e_ddp_is_pkg_hdr_valid(netdev, pkg_hdr, size))
+		return -EINVAL;
+
+	if (size < (sizeof(struct i40e_package_header) + sizeof(u32) +
+		    sizeof(struct i40e_metadata_segment) + sizeof(u32) * 2)) {
+		netdev_err(netdev, "Invalid DDP recipe size.");
+		return -EINVAL;
+	}
+
+	/* Find beginning of segment data in buffer */
+	metadata_hdr = (struct i40e_metadata_segment *)
+		i40e_find_segment_in_package(SEGMENT_TYPE_METADATA, pkg_hdr);
+	if (!metadata_hdr) {
+		netdev_err(netdev, "Failed to find metadata segment in DDP recipe.");
+		return -EINVAL;
+	}
+
+	track_id = metadata_hdr->track_id;
+	profile_hdr = (struct i40e_profile_segment *)
+		i40e_find_segment_in_package(SEGMENT_TYPE_I40E, pkg_hdr);
+	if (!profile_hdr) {
+		netdev_err(netdev, "Failed to find profile segment in DDP recipe.");
+		return -EINVAL;
+	}
+
+	pinfo.track_id = track_id;
+	pinfo.version = profile_hdr->version;
+	if (is_add)
+		pinfo.op = I40E_DDP_ADD_TRACKID;
+	else
+		pinfo.op = I40E_DDP_REMOVE_TRACKID;
+
+	memcpy(pinfo.name, profile_hdr->name, I40E_DDP_NAME_SIZE);
+
+	/* Check if profile data already exists*/
+	istatus = i40e_ddp_does_profile_exist(&pf->hw, &pinfo);
+	if (istatus < 0) {
+		netdev_err(netdev, "Failed to fetch loaded profiles.");
+		return istatus;
+	}
+	if (is_add) {
+		if (istatus > 0) {
+			netdev_err(netdev, "DDP profile already loaded.");
+			return -EINVAL;
+		}
+		istatus = i40e_ddp_does_profile_overlap(&pf->hw, &pinfo);
+		if (istatus < 0) {
+			netdev_err(netdev, "Failed to fetch loaded profiles.");
+			return istatus;
+		}
+		if (istatus > 0) {
+			netdev_err(netdev, "DDP profile overlaps with existing one.");
+			return -EINVAL;
+		}
+	} else {
+		if (istatus == 0) {
+			netdev_err(netdev,
+				   "DDP profile for deletion does not exist.");
+			return -EINVAL;
+		}
+	}
+
+	/* Load profile data */
+	if (is_add) {
+		status = i40e_write_profile(&pf->hw, profile_hdr, track_id);
+		if (status) {
+			if (status == -ENODEV) {
+				netdev_err(netdev,
+					   "Profile is not supported by the device.");
+				return -EPERM;
+			}
+			netdev_err(netdev, "Failed to write DDP profile.");
+			return -EIO;
+		}
+	} else {
+		status = i40e_rollback_profile(&pf->hw, profile_hdr, track_id);
+		if (status) {
+			netdev_err(netdev, "Failed to remove DDP profile.");
+			return -EIO;
+		}
+	}
+
+	/* Add/remove profile to/from profile list in FW */
+	if (is_add) {
+		status = i40e_add_pinfo(&pf->hw, profile_hdr, profile_info_sec,
+					track_id);
+		if (status) {
+			netdev_err(netdev, "Failed to add DDP profile info.");
+			return -EIO;
+		}
+	} else {
+		status = i40e_del_pinfo(&pf->hw, profile_hdr, profile_info_sec,
+					track_id);
+		if (status) {
+			netdev_err(netdev, "Failed to restore DDP profile info.");
+			return -EIO;
+		}
+	}
+
+	return 0;
+}
+
+/**
+ * i40e_ddp_restore - restore previously loaded profile and remove from list
+ * @pf: PF data struct
+ *
+ * Restores previously loaded profile stored on the list in driver memory.
+ * After rolling back removes entry from the list.
+ **/
+static int i40e_ddp_restore(struct i40e_pf *pf)
+{
+	struct i40e_vsi *vsi = i40e_pf_get_main_vsi(pf);
+	struct net_device *netdev = vsi->netdev;
+	struct i40e_ddp_old_profile_list *entry;
+	int status = 0;
+
+	if (!list_empty(&pf->ddp_old_prof)) {
+		entry = list_first_entry(&pf->ddp_old_prof,
+					 struct i40e_ddp_old_profile_list,
+					 list);
+		status = i40e_ddp_load(netdev, entry->old_ddp_buf,
+				       entry->old_ddp_size, false);
+		list_del(&entry->list);
+		kfree(entry);
+	}
+	return status;
+}
+
+/**
+ * i40e_ddp_flash - callback function for ethtool flash feature
+ * @netdev: net device structure
+ * @flash: kernel flash structure
+ *
+ * Ethtool callback function used for loading and unloading DDP profiles.
+ **/
+int i40e_ddp_flash(struct net_device *netdev, struct ethtool_flash *flash)
+{
+	const struct firmware *ddp_config;
+	struct i40e_netdev_priv *np = netdev_priv(netdev);
+	struct i40e_vsi *vsi = np->vsi;
+	struct i40e_pf *pf = vsi->back;
+	int status = 0;
+
+	/* Check for valid region first */
+	if (flash->region != I40_DDP_FLASH_REGION) {
+		netdev_err(netdev, "Requested firmware region is not recognized by this driver.");
+		return -EINVAL;
+	}
+	if (pf->hw.bus.func != 0) {
+		netdev_err(netdev, "Any DDP operation is allowed only on Phy0 NIC interface");
+		return -EINVAL;
+	}
+
+	/* If the user supplied "-" instead of file name rollback previously
+	 * stored profile.
+	 */
+	if (strncmp(flash->data, "-", 2) != 0) {
+		struct i40e_ddp_old_profile_list *list_entry;
+		char profile_name[sizeof(I40E_DDP_PROFILE_PATH)
+				  + I40E_DDP_PROFILE_NAME_MAX];
+
+		scnprintf(profile_name, sizeof(profile_name), "%s%s",
+			  I40E_DDP_PROFILE_PATH, flash->data);
+
+		/* Load DDP recipe. */
+		status = request_firmware(&ddp_config, profile_name,
+					  &netdev->dev);
+		if (status) {
+			netdev_err(netdev, "DDP recipe file request failed.");
+			return status;
+		}
+
+		status = i40e_ddp_load(netdev, ddp_config->data,
+				       ddp_config->size, true);
+
+		if (!status) {
+			list_entry =
+			  kzalloc(sizeof(struct i40e_ddp_old_profile_list) +
+				  ddp_config->size, GFP_KERNEL);
+			if (!list_entry) {
+				netdev_info(netdev, "Failed to allocate memory for previous DDP profile data.");
+				netdev_info(netdev, "New profile loaded but roll-back will be impossible.");
+			} else {
+				memcpy(list_entry->old_ddp_buf,
+				       ddp_config->data, ddp_config->size);
+				list_entry->old_ddp_size = ddp_config->size;
+				list_add(&list_entry->list, &pf->ddp_old_prof);
+			}
+		}
+
+		release_firmware(ddp_config);
+	} else {
+		if (!list_empty(&pf->ddp_old_prof)) {
+			status = i40e_ddp_restore(pf);
+		} else {
+			netdev_warn(netdev, "There is no DDP profile to restore.");
+			status = -ENOENT;
+		}
+	}
+	return status;
+}
diff --git a/drivers/net/ethernet/intel/i40e/i40e_debug.h b/drivers/net/ethernet/intel/i40e/i40e_debug.h
new file mode 100644
index 000000000000..e9871dfb32bd
--- /dev/null
+++ b/drivers/net/ethernet/intel/i40e/i40e_debug.h
@@ -0,0 +1,48 @@
+/* SPDX-License-Identifier: GPL-2.0 */
+/* Copyright(c) 2023 Intel Corporation. */
+
+#ifndef _I40E_DEBUG_H_
+#define _I40E_DEBUG_H_
+
+#include <linux/dev_printk.h>
+
+/* debug masks - set these bits in hw->debug_mask to control output */
+enum i40e_debug_mask {
+	I40E_DEBUG_INIT			= 0x00000001,
+	I40E_DEBUG_RELEASE		= 0x00000002,
+
+	I40E_DEBUG_LINK			= 0x00000010,
+	I40E_DEBUG_PHY			= 0x00000020,
+	I40E_DEBUG_HMC			= 0x00000040,
+	I40E_DEBUG_NVM			= 0x00000080,
+	I40E_DEBUG_LAN			= 0x00000100,
+	I40E_DEBUG_FLOW			= 0x00000200,
+	I40E_DEBUG_DCB			= 0x00000400,
+	I40E_DEBUG_DIAG			= 0x00000800,
+	I40E_DEBUG_FD			= 0x00001000,
+	I40E_DEBUG_PACKAGE		= 0x00002000,
+	I40E_DEBUG_IWARP		= 0x00F00000,
+	I40E_DEBUG_AQ_MESSAGE		= 0x01000000,
+	I40E_DEBUG_AQ_DESCRIPTOR	= 0x02000000,
+	I40E_DEBUG_AQ_DESC_BUFFER	= 0x04000000,
+	I40E_DEBUG_AQ_COMMAND		= 0x06000000,
+	I40E_DEBUG_AQ			= 0x0F000000,
+
+	I40E_DEBUG_USER			= 0xF0000000,
+
+	I40E_DEBUG_ALL			= 0xFFFFFFFF
+};
+
+struct i40e_hw;
+struct device *i40e_hw_to_dev(struct i40e_hw *hw);
+
+#define hw_dbg(hw, S, A...) dev_dbg(i40e_hw_to_dev(hw), S, ##A)
+#define hw_warn(hw, S, A...) dev_warn(i40e_hw_to_dev(hw), S, ##A)
+
+#define i40e_debug(h, m, s, ...)				\
+do {								\
+	if (((m) & (h)->debug_mask))				\
+		dev_info(i40e_hw_to_dev(hw), s, ##__VA_ARGS__);	\
+} while (0)
+
+#endif /* _I40E_DEBUG_H_ */
diff --git a/drivers/net/ethernet/intel/i40e/i40e_debugfs.c b/drivers/net/ethernet/intel/i40e/i40e_debugfs.c
index 8f326f87a815..c17b5d290f0a 100644
--- a/drivers/net/ethernet/intel/i40e/i40e_debugfs.c
+++ b/drivers/net/ethernet/intel/i40e/i40e_debugfs.c
@@ -1,70 +1,36 @@
-/*******************************************************************************
- *
- * Intel Ethernet Controller XL710 Family Linux Driver
- * Copyright(c) 2013 - 2016 Intel Corporation.
- *
- * This program is free software; you can redistribute it and/or modify it
- * under the terms and conditions of the GNU General Public License,
- * version 2, as published by the Free Software Foundation.
- *
- * This program is distributed in the hope it will be useful, but WITHOUT
- * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
- * FITNESS FOR A PARTICULAR PURPOSE.  See the GNU General Public License for
- * more details.
- *
- * You should have received a copy of the GNU General Public License along
- * with this program.  If not, see <http://www.gnu.org/licenses/>.
- *
- * The full GNU General Public License is included in this distribution in
- * the file called "COPYING".
- *
- * Contact Information:
- * e1000-devel Mailing List <e1000-devel@lists.sourceforge.net>
- * Intel Corporation, 5200 N.E. Elam Young Parkway, Hillsboro, OR 97124-6497
- *
- ******************************************************************************/
+// SPDX-License-Identifier: GPL-2.0
+/* Copyright(c) 2013 - 2018 Intel Corporation. */
 
 #ifdef CONFIG_DEBUG_FS
 
 #include <linux/fs.h>
 #include <linux/debugfs.h>
-
+#include <linux/if_bridge.h>
 #include "i40e.h"
+#include "i40e_virtchnl_pf.h"
 
 static struct dentry *i40e_dbg_root;
 
+enum ring_type {
+	RING_TYPE_RX,
+	RING_TYPE_TX,
+	RING_TYPE_XDP
+};
+
 /**
  * i40e_dbg_find_vsi - searches for the vsi with the given seid
- * @pf - the PF structure to search for the vsi
- * @seid - seid of the vsi it is searching for
+ * @pf: the PF structure to search for the vsi
+ * @seid: seid of the vsi it is searching for
  **/
 static struct i40e_vsi *i40e_dbg_find_vsi(struct i40e_pf *pf, int seid)
 {
-	int i;
-
-	if (seid < 0)
+	if (seid < 0) {
 		dev_info(&pf->pdev->dev, "%d: bad seid\n", seid);
-	else
-		for (i = 0; i < pf->num_alloc_vsi; i++)
-			if (pf->vsi[i] && (pf->vsi[i]->seid == seid))
-				return pf->vsi[i];
-
-	return NULL;
-}
 
-/**
- * i40e_dbg_find_veb - searches for the veb with the given seid
- * @pf - the PF structure to search for the veb
- * @seid - seid of the veb it is searching for
- **/
-static struct i40e_veb *i40e_dbg_find_veb(struct i40e_pf *pf, int seid)
-{
-	int i;
+		return NULL;
+	}
 
-	for (i = 0; i < I40E_MAX_VEB; i++)
-		if (pf->veb[i] && pf->veb[i]->seid == seid)
-			return pf->veb[i];
-	return NULL;
+	return i40e_pf_get_vsi_by_seid(pf, seid);
 }
 
 /**************************************************************
@@ -74,47 +40,6 @@ static struct i40e_veb *i40e_dbg_find_veb(struct i40e_pf *pf, int seid)
  * setup, adding or removing filters, or other things.  Many of
  * these will be useful for some forms of unit testing.
  **************************************************************/
-static char i40e_dbg_command_buf[256] = "";
-
-/**
- * i40e_dbg_command_read - read for command datum
- * @filp: the opened file
- * @buffer: where to write the data for the user to read
- * @count: the size of the user's buffer
- * @ppos: file position offset
- **/
-static ssize_t i40e_dbg_command_read(struct file *filp, char __user *buffer,
-				     size_t count, loff_t *ppos)
-{
-	struct i40e_pf *pf = filp->private_data;
-	int bytes_not_copied;
-	int buf_size = 256;
-	char *buf;
-	int len;
-
-	/* don't allow partial reads */
-	if (*ppos != 0)
-		return 0;
-	if (count < buf_size)
-		return -ENOSPC;
-
-	buf = kzalloc(buf_size, GFP_KERNEL);
-	if (!buf)
-		return -ENOSPC;
-
-	len = snprintf(buf, buf_size, "%s: %s\n",
-		       pf->vsi[pf->lan_vsi]->netdev->name,
-		       i40e_dbg_command_buf);
-
-	bytes_not_copied = copy_to_user(buffer, buf, len);
-	kfree(buf);
-
-	if (bytes_not_copied)
-		return -EFAULT;
-
-	*ppos = len;
-	return len;
-}
 
 static char *i40e_filter_state_string[] = {
 	"INVALID",
@@ -122,6 +47,7 @@ static char *i40e_filter_state_string[] = {
 	"ACTIVE",
 	"FAILED",
 	"REMOVE",
+	"NEW_SYNC",
 };
 
 /**
@@ -156,18 +82,15 @@ static void i40e_dbg_dump_vsi_seid(struct i40e_pf *pf, int seid)
 			 (unsigned long int)nd->vlan_features);
 	}
 	dev_info(&pf->pdev->dev,
-		 "    vlgrp: & = %p\n", vsi->active_vlans);
-	dev_info(&pf->pdev->dev,
 		 "    flags = 0x%08lx, netdev_registered = %i, current_netdev_flags = 0x%04x\n",
 		 vsi->flags, vsi->netdev_registered, vsi->current_netdev_flags);
 	for (i = 0; i < BITS_TO_LONGS(__I40E_VSI_STATE_SIZE__); i++)
 		dev_info(&pf->pdev->dev,
 			 "    state[%d] = %08lx\n",
 			 i, vsi->state[i]);
-	if (vsi == pf->vsi[pf->lan_vsi])
-		dev_info(&pf->pdev->dev, "    MAC address: %pM SAN MAC: %pM Port MAC: %pM\n",
+	if (vsi->type == I40E_VSI_MAIN)
+		dev_info(&pf->pdev->dev, "    MAC address: %pM Port MAC: %pM\n",
 			 pf->hw.mac.addr,
-			 pf->hw.mac.san_addr,
 			 pf->hw.mac.port_addr);
 	hash_for_each(vsi->mac_filter_hash, bkt, f, hlist) {
 		dev_info(&pf->pdev->dev,
@@ -259,27 +182,19 @@ static void i40e_dbg_dump_vsi_seid(struct i40e_pf *pf, int seid)
 		 (unsigned long int)vsi->net_stats_offsets.rx_compressed,
 		 (unsigned long int)vsi->net_stats_offsets.tx_compressed);
 	dev_info(&pf->pdev->dev,
-		 "    tx_restart = %d, tx_busy = %d, rx_buf_failed = %d, rx_page_failed = %d\n",
+		 "    tx_restart = %llu, tx_busy = %llu, rx_buf_failed = %llu, rx_page_failed = %llu\n",
 		 vsi->tx_restart, vsi->tx_busy,
 		 vsi->rx_buf_failed, vsi->rx_page_failed);
 	rcu_read_lock();
 	for (i = 0; i < vsi->num_queue_pairs; i++) {
-		struct i40e_ring *rx_ring = ACCESS_ONCE(vsi->rx_rings[i]);
+		struct i40e_ring *rx_ring = READ_ONCE(vsi->rx_rings[i]);
 
 		if (!rx_ring)
 			continue;
 
 		dev_info(&pf->pdev->dev,
-			 "    rx_rings[%i]: desc = %p\n",
-			 i, rx_ring->desc);
-		dev_info(&pf->pdev->dev,
-			 "    rx_rings[%i]: dev = %p, netdev = %p, rx_bi = %p\n",
-			 i, rx_ring->dev,
-			 rx_ring->netdev,
-			 rx_ring->rx_bi);
-		dev_info(&pf->pdev->dev,
-			 "    rx_rings[%i]: state = %li, queue_index = %d, reg_idx = %d\n",
-			 i, rx_ring->state,
+			 "    rx_rings[%i]: state = %lu, queue_index = %d, reg_idx = %d\n",
+			 i, *rx_ring->state,
 			 rx_ring->queue_index,
 			 rx_ring->reg_idx);
 		dev_info(&pf->pdev->dev,
@@ -302,40 +217,26 @@ static void i40e_dbg_dump_vsi_seid(struct i40e_pf *pf, int seid)
 			 rx_ring->rx_stats.alloc_page_failed,
 			 rx_ring->rx_stats.alloc_buff_failed);
 		dev_info(&pf->pdev->dev,
-			 "    rx_rings[%i]: rx_stats: realloc_count = %lld, page_reuse_count = %lld\n",
+			 "    rx_rings[%i]: rx_stats: realloc_count = 0, page_reuse_count = %lld\n",
 			 i,
-			 rx_ring->rx_stats.realloc_count,
 			 rx_ring->rx_stats.page_reuse_count);
 		dev_info(&pf->pdev->dev,
-			 "    rx_rings[%i]: size = %i, dma = 0x%08lx\n",
-			 i, rx_ring->size,
-			 (unsigned long int)rx_ring->dma);
+			 "    rx_rings[%i]: size = %i\n",
+			 i, rx_ring->size);
 		dev_info(&pf->pdev->dev,
-			 "    rx_rings[%i]: vsi = %p, q_vector = %p\n",
-			 i, rx_ring->vsi,
-			 rx_ring->q_vector);
-		dev_info(&pf->pdev->dev,
-			 "    rx_rings[%i]: rx_itr_setting = %d (%s)\n",
-			 i, rx_ring->rx_itr_setting,
-			 ITR_IS_DYNAMIC(rx_ring->rx_itr_setting) ? "dynamic" : "fixed");
+			 "    rx_rings[%i]: itr_setting = %d (%s)\n",
+			 i, rx_ring->itr_setting,
+			 ITR_IS_DYNAMIC(rx_ring->itr_setting) ? "dynamic" : "fixed");
 	}
 	for (i = 0; i < vsi->num_queue_pairs; i++) {
-		struct i40e_ring *tx_ring = ACCESS_ONCE(vsi->tx_rings[i]);
+		struct i40e_ring *tx_ring = READ_ONCE(vsi->tx_rings[i]);
 
 		if (!tx_ring)
 			continue;
 
 		dev_info(&pf->pdev->dev,
-			 "    tx_rings[%i]: desc = %p\n",
-			 i, tx_ring->desc);
-		dev_info(&pf->pdev->dev,
-			 "    tx_rings[%i]: dev = %p, netdev = %p, tx_bi = %p\n",
-			 i, tx_ring->dev,
-			 tx_ring->netdev,
-			 tx_ring->tx_bi);
-		dev_info(&pf->pdev->dev,
-			 "    tx_rings[%i]: state = %li, queue_index = %d, reg_idx = %d\n",
-			 i, tx_ring->state,
+			 "    tx_rings[%i]: state = %lu, queue_index = %d, reg_idx = %d\n",
+			 i, *tx_ring->state,
 			 tx_ring->queue_index,
 			 tx_ring->reg_idx);
 		dev_info(&pf->pdev->dev,
@@ -350,25 +251,62 @@ static void i40e_dbg_dump_vsi_seid(struct i40e_pf *pf, int seid)
 			 tx_ring->stats.bytes,
 			 tx_ring->tx_stats.restart_queue);
 		dev_info(&pf->pdev->dev,
-			 "    tx_rings[%i]: tx_stats: tx_busy = %lld, tx_done_old = %lld\n",
+			 "    tx_rings[%i]: tx_stats: tx_busy = %lld, tx_done_old = %lld, tx_stopped = %lld\n",
 			 i,
 			 tx_ring->tx_stats.tx_busy,
-			 tx_ring->tx_stats.tx_done_old);
+			 tx_ring->tx_stats.tx_done_old,
+			 tx_ring->tx_stats.tx_stopped);
 		dev_info(&pf->pdev->dev,
-			 "    tx_rings[%i]: size = %i, dma = 0x%08lx\n",
-			 i, tx_ring->size,
-			 (unsigned long int)tx_ring->dma);
-		dev_info(&pf->pdev->dev,
-			 "    tx_rings[%i]: vsi = %p, q_vector = %p\n",
-			 i, tx_ring->vsi,
-			 tx_ring->q_vector);
+			 "    tx_rings[%i]: size = %i\n",
+			 i, tx_ring->size);
 		dev_info(&pf->pdev->dev,
 			 "    tx_rings[%i]: DCB tc = %d\n",
 			 i, tx_ring->dcb_tc);
 		dev_info(&pf->pdev->dev,
-			 "    tx_rings[%i]: tx_itr_setting = %d (%s)\n",
-			 i, tx_ring->tx_itr_setting,
-			 ITR_IS_DYNAMIC(tx_ring->tx_itr_setting) ? "dynamic" : "fixed");
+			 "    tx_rings[%i]: itr_setting = %d (%s)\n",
+			 i, tx_ring->itr_setting,
+			 ITR_IS_DYNAMIC(tx_ring->itr_setting) ? "dynamic" : "fixed");
+	}
+	if (i40e_enabled_xdp_vsi(vsi)) {
+		for (i = 0; i < vsi->num_queue_pairs; i++) {
+			struct i40e_ring *xdp_ring = READ_ONCE(vsi->xdp_rings[i]);
+
+			if (!xdp_ring)
+				continue;
+
+			dev_info(&pf->pdev->dev,
+				 "    xdp_rings[%i]: state = %lu, queue_index = %d, reg_idx = %d\n",
+				 i, *xdp_ring->state,
+				 xdp_ring->queue_index,
+				 xdp_ring->reg_idx);
+			dev_info(&pf->pdev->dev,
+				 "    xdp_rings[%i]: next_to_use = %d, next_to_clean = %d, ring_active = %i\n",
+				 i,
+				 xdp_ring->next_to_use,
+				 xdp_ring->next_to_clean,
+				 xdp_ring->ring_active);
+			dev_info(&pf->pdev->dev,
+				 "    xdp_rings[%i]: tx_stats: packets = %lld, bytes = %lld, restart_queue = %lld\n",
+				 i, xdp_ring->stats.packets,
+				 xdp_ring->stats.bytes,
+				 xdp_ring->tx_stats.restart_queue);
+			dev_info(&pf->pdev->dev,
+				 "    xdp_rings[%i]: tx_stats: tx_busy = %lld, tx_done_old = %lld\n",
+				 i,
+				 xdp_ring->tx_stats.tx_busy,
+				 xdp_ring->tx_stats.tx_done_old);
+			dev_info(&pf->pdev->dev,
+				 "    xdp_rings[%i]: size = %i\n",
+				 i, xdp_ring->size);
+			dev_info(&pf->pdev->dev,
+				 "    xdp_rings[%i]: DCB tc = %d\n",
+				 i, xdp_ring->dcb_tc);
+			dev_info(&pf->pdev->dev,
+				 "    xdp_rings[%i]: itr_setting = %d (%s)\n",
+				 i, xdp_ring->itr_setting,
+				 ITR_IS_DYNAMIC(xdp_ring->itr_setting) ?
+				 "dynamic" : "fixed");
+		}
 	}
 	rcu_read_unlock();
 	dev_info(&pf->pdev->dev,
@@ -384,8 +322,9 @@ static void i40e_dbg_dump_vsi_seid(struct i40e_pf *pf, int seid)
 		 "    seid = %d, id = %d, uplink_seid = %d\n",
 		 vsi->seid, vsi->id, vsi->uplink_seid);
 	dev_info(&pf->pdev->dev,
-		 "    base_queue = %d, num_queue_pairs = %d, num_desc = %d\n",
-		 vsi->base_queue, vsi->num_queue_pairs, vsi->num_desc);
+		 "    base_queue = %d, num_queue_pairs = %d, num_tx_desc = %d, num_rx_desc = %d\n",
+		 vsi->base_queue, vsi->num_queue_pairs, vsi->num_tx_desc,
+		 vsi->num_rx_desc);
 	dev_info(&pf->pdev->dev, "    type = %i\n", vsi->type);
 	if (vsi->type == I40E_VSI_SRIOV)
 		dev_info(&pf->pdev->dev, "    VF ID = %i\n", vsi->vf_id);
@@ -466,8 +405,6 @@ static void i40e_dbg_dump_vsi_seid(struct i40e_pf *pf, int seid)
 		 vsi->info.resp_reserved[6], vsi->info.resp_reserved[7],
 		 vsi->info.resp_reserved[8], vsi->info.resp_reserved[9],
 		 vsi->info.resp_reserved[10], vsi->info.resp_reserved[11]);
-	if (vsi->back)
-		dev_info(&pf->pdev->dev, "    PF = %p\n", vsi->back);
 	dev_info(&pf->pdev->dev, "    idx = %d\n", vsi->idx);
 	dev_info(&pf->pdev->dev,
 		 "    tc_config: numtc = %d, enabled_tc = 0x%x\n",
@@ -510,7 +447,7 @@ static void i40e_dbg_dump_aq_desc(struct i40e_pf *pf)
 	dev_info(&pf->pdev->dev, "AdminQ Tx Ring\n");
 	ring = &(hw->aq.asq);
 	for (i = 0; i < ring->count; i++) {
-		struct i40e_aq_desc *d = I40E_ADMINQ_DESC(*ring, i);
+		struct libie_aq_desc *d = I40E_ADMINQ_DESC(*ring, i);
 
 		dev_info(&pf->pdev->dev,
 			 "   at[%02d] flags=0x%04x op=0x%04x dlen=0x%04x ret=0x%04x cookie_h=0x%08x cookie_l=0x%08x\n",
@@ -523,7 +460,7 @@ static void i40e_dbg_dump_aq_desc(struct i40e_pf *pf)
 	dev_info(&pf->pdev->dev, "AdminQ Rx Ring\n");
 	ring = &(hw->aq.arq);
 	for (i = 0; i < ring->count; i++) {
-		struct i40e_aq_desc *d = I40E_ADMINQ_DESC(*ring, i);
+		struct libie_aq_desc *d = I40E_ADMINQ_DESC(*ring, i);
 
 		dev_info(&pf->pdev->dev,
 			 "   ar[%02d] flags=0x%04x op=0x%04x dlen=0x%04x ret=0x%04x cookie_h=0x%08x cookie_l=0x%08x\n",
@@ -541,11 +478,12 @@ static void i40e_dbg_dump_aq_desc(struct i40e_pf *pf)
  * @ring_id: ring id entered by user
  * @desc_n: descriptor number entered by user
  * @pf: the i40e_pf created in command write
- * @is_rx_ring: true if rx, false if tx
+ * @type: enum describing whether ring is RX, TX or XDP
  **/
 static void i40e_dbg_dump_desc(int cnt, int vsi_seid, int ring_id, int desc_n,
-			       struct i40e_pf *pf, bool is_rx_ring)
+			       struct i40e_pf *pf, enum ring_type type)
 {
+	bool is_rx_ring = type == RING_TYPE_RX;
 	struct i40e_tx_desc *txd;
 	union i40e_rx_desc *rxd;
 	struct i40e_ring *ring;
@@ -557,6 +495,18 @@ static void i40e_dbg_dump_desc(int cnt, int vsi_seid, int ring_id, int desc_n,
 		dev_info(&pf->pdev->dev, "vsi %d not found\n", vsi_seid);
 		return;
 	}
+	if (vsi->type != I40E_VSI_MAIN &&
+	    vsi->type != I40E_VSI_FDIR &&
+	    vsi->type != I40E_VSI_VMDQ2) {
+		dev_info(&pf->pdev->dev,
+			 "vsi %d type %d descriptor rings not available\n",
+			 vsi_seid, vsi->type);
+		return;
+	}
+	if (type == RING_TYPE_XDP && !i40e_enabled_xdp_vsi(vsi)) {
+		dev_info(&pf->pdev->dev, "XDP not enabled on VSI %d\n", vsi_seid);
+		return;
+	}
 	if (ring_id >= vsi->num_queue_pairs || ring_id < 0) {
 		dev_info(&pf->pdev->dev, "ring %d not found\n", ring_id);
 		return;
@@ -568,15 +518,35 @@ static void i40e_dbg_dump_desc(int cnt, int vsi_seid, int ring_id, int desc_n,
 		return;
 	}
 
-	ring = kmemdup(is_rx_ring
-		       ? vsi->rx_rings[ring_id] : vsi->tx_rings[ring_id],
-		       sizeof(*ring), GFP_KERNEL);
+	switch (type) {
+	case RING_TYPE_RX:
+		ring = kmemdup(vsi->rx_rings[ring_id], sizeof(*ring), GFP_KERNEL);
+		break;
+	case RING_TYPE_TX:
+		ring = kmemdup(vsi->tx_rings[ring_id], sizeof(*ring), GFP_KERNEL);
+		break;
+	case RING_TYPE_XDP:
+		ring = kmemdup(vsi->xdp_rings[ring_id], sizeof(*ring), GFP_KERNEL);
+		break;
+	default:
+		ring = NULL;
+		break;
+	}
 	if (!ring)
 		return;
 
 	if (cnt == 2) {
-		dev_info(&pf->pdev->dev, "vsi = %02i %s ring = %02i\n",
-			 vsi_seid, is_rx_ring ? "rx" : "tx", ring_id);
+		switch (type) {
+		case RING_TYPE_RX:
+			dev_info(&pf->pdev->dev, "VSI = %02i Rx ring = %02i\n", vsi_seid, ring_id);
+			break;
+		case RING_TYPE_TX:
+			dev_info(&pf->pdev->dev, "VSI = %02i Tx ring = %02i\n", vsi_seid, ring_id);
+			break;
+		case RING_TYPE_XDP:
+			dev_info(&pf->pdev->dev, "VSI = %02i XDP ring = %02i\n", vsi_seid, ring_id);
+			break;
+		}
 		for (i = 0; i < ring->count; i++) {
 			if (!is_rx_ring) {
 				txd = I40E_TX_DESC(ring, i);
@@ -587,10 +557,9 @@ static void i40e_dbg_dump_desc(int cnt, int vsi_seid, int ring_id, int desc_n,
 			} else {
 				rxd = I40E_RX_DESC(ring, i);
 				dev_info(&pf->pdev->dev,
-					 "   d[%03x] = 0x%016llx 0x%016llx 0x%016llx 0x%016llx\n",
+					 "   d[%03x] = 0x%016llx 0x%016llx\n",
 					 i, rxd->read.pkt_addr,
-					 rxd->read.hdr_addr,
-					 rxd->read.rsvd1, rxd->read.rsvd2);
+					 rxd->read.hdr_addr);
 			}
 		}
 	} else if (cnt == 3) {
@@ -608,13 +577,12 @@ static void i40e_dbg_dump_desc(int cnt, int vsi_seid, int ring_id, int desc_n,
 		} else {
 			rxd = I40E_RX_DESC(ring, desc_n);
 			dev_info(&pf->pdev->dev,
-				 "vsi = %02i rx ring = %02i d[%03x] = 0x%016llx 0x%016llx 0x%016llx 0x%016llx\n",
+				 "vsi = %02i rx ring = %02i d[%03x] = 0x%016llx 0x%016llx\n",
 				 vsi_seid, ring_id, desc_n,
-				 rxd->read.pkt_addr, rxd->read.hdr_addr,
-				 rxd->read.rsvd1, rxd->read.rsvd2);
+				 rxd->read.pkt_addr, rxd->read.hdr_addr);
 		}
 	} else {
-		dev_info(&pf->pdev->dev, "dump desc rx/tx <vsi_seid> <ring_id> [<desc_n>]\n");
+		dev_info(&pf->pdev->dev, "dump desc rx/tx/xdp <vsi_seid> <ring_id> [<desc_n>]\n");
 	}
 
 out:
@@ -627,16 +595,15 @@ out:
  **/
 static void i40e_dbg_dump_vsi_no_seid(struct i40e_pf *pf)
 {
+	struct i40e_vsi *vsi;
 	int i;
 
-	for (i = 0; i < pf->num_alloc_vsi; i++)
-		if (pf->vsi[i])
-			dev_info(&pf->pdev->dev, "dump vsi[%d]: %d\n",
-				 i, pf->vsi[i]->seid);
+	i40e_pf_for_each_vsi(pf, i, vsi)
+		dev_info(&pf->pdev->dev, "dump vsi[%d]: %d\n", i, vsi->seid);
 }
 
 /**
- * i40e_dbg_dump_stats - handles dump stats write into command datum
+ * i40e_dbg_dump_eth_stats - handles dump stats write into command datum
  * @pf: the i40e_pf created in command write
  * @estats: the eth stats structure to be dumped
  **/
@@ -670,15 +637,14 @@ static void i40e_dbg_dump_veb_seid(struct i40e_pf *pf, int seid)
 {
 	struct i40e_veb *veb;
 
-	veb = i40e_dbg_find_veb(pf, seid);
+	veb = i40e_pf_get_veb_by_seid(pf, seid);
 	if (!veb) {
 		dev_info(&pf->pdev->dev, "can't find veb %d\n", seid);
 		return;
 	}
 	dev_info(&pf->pdev->dev,
-		 "veb idx=%d,%d stats_ic=%d  seid=%d uplink=%d mode=%s\n",
-		 veb->idx, veb->veb_idx, veb->stats_idx, veb->seid,
-		 veb->uplink_seid,
+		 "veb idx=%d stats_ic=%d  seid=%d uplink=%d mode=%s\n",
+		 veb->idx, veb->stats_idx, veb->seid, veb->uplink_seid,
 		 veb->bridge_mode == BRIDGE_MODE_VEPA ? "VEPA" : "VEB");
 	i40e_dbg_dump_eth_stats(pf, &veb->stats);
 }
@@ -692,11 +658,8 @@ static void i40e_dbg_dump_veb_all(struct i40e_pf *pf)
 	struct i40e_veb *veb;
 	int i;
 
-	for (i = 0; i < I40E_MAX_VEB; i++) {
-		veb = pf->veb[i];
-		if (veb)
-			i40e_dbg_dump_veb_seid(pf, veb->seid);
-	}
+	i40e_pf_for_each_veb(pf, i, veb)
+		i40e_dbg_dump_veb_seid(pf, veb->seid);
 }
 
 /**
@@ -716,10 +679,8 @@ static void i40e_dbg_dump_vf(struct i40e_pf *pf, int vf_id)
 		vsi = pf->vsi[vf->lan_vsi_idx];
 		dev_info(&pf->pdev->dev, "vf %2d: VSI id=%d, seid=%d, qps=%d\n",
 			 vf_id, vf->lan_vsi_id, vsi->seid, vf->num_queue_pairs);
-		dev_info(&pf->pdev->dev, "       num MDD=%lld, invalid msg=%lld, valid msg=%lld\n",
-			 vf->num_mdd_events,
-			 vf->num_invalid_msgs,
-			 vf->num_valid_msgs);
+		dev_info(&pf->pdev->dev, "       num MDD=%lld\n",
+			 vf->mdd_tx_events.count + vf->mdd_rx_events.count);
 	} else {
 		dev_info(&pf->pdev->dev, "invalid VF id %d\n", vf_id);
 	}
@@ -740,7 +701,6 @@ static void i40e_dbg_dump_vf_all(struct i40e_pf *pf)
 			i40e_dbg_dump_vf(pf, i);
 }
 
-#define I40E_MAX_DEBUG_OUT_BUFFER (4096*4)
 /**
  * i40e_dbg_command_write - write into command datum
  * @filp: the opened file
@@ -786,7 +746,8 @@ static ssize_t i40e_dbg_command_write(struct file *filp,
 		cnt = sscanf(&cmd_buf[7], "%i", &vsi_seid);
 		if (cnt == 0) {
 			/* default to PF VSI */
-			vsi_seid = pf->vsi[pf->lan_vsi]->seid;
+			vsi = i40e_pf_get_main_vsi(pf);
+			vsi_seid = vsi->seid;
 		} else if (vsi_seid < 0) {
 			dev_info(&pf->pdev->dev, "add VSI %d: bad vsi seid\n",
 				 vsi_seid);
@@ -796,10 +757,9 @@ static ssize_t i40e_dbg_command_write(struct file *filp,
 		/* By default we are in VEPA mode, if this is the first VF/VMDq
 		 * VSI to be added switch to VEB mode.
 		 */
-		if (!(pf->flags & I40E_FLAG_VEB_MODE_ENABLED)) {
-			pf->flags |= I40E_FLAG_VEB_MODE_ENABLED;
-			i40e_do_reset_safe(pf,
-					   BIT_ULL(__I40E_PF_RESET_REQUESTED));
+		if (!test_bit(I40E_FLAG_VEB_MODE_ENA, pf->flags)) {
+			set_bit(I40E_FLAG_VEB_MODE_ENA, pf->flags);
+			i40e_do_reset_safe(pf, I40E_PF_RESET_FLAG);
 		}
 
 		vsi = i40e_vsi_setup(pf, I40E_VSI_VMDQ2, vsi_seid, 0);
@@ -829,10 +789,14 @@ static ssize_t i40e_dbg_command_write(struct file *filp,
 
 	} else if (strncmp(cmd_buf, "add relay", 9) == 0) {
 		struct i40e_veb *veb;
-		int uplink_seid, i;
+		u8 enabled_tc = 0x1;
+		int uplink_seid;
 
 		cnt = sscanf(&cmd_buf[9], "%i %i", &uplink_seid, &vsi_seid);
-		if (cnt != 2) {
+		if (cnt == 0) {
+			uplink_seid = 0;
+			vsi_seid = 0;
+		} else if (cnt != 2) {
 			dev_info(&pf->pdev->dev,
 				 "add relay: bad command string, cnt=%d\n",
 				 cnt);
@@ -844,33 +808,36 @@ static ssize_t i40e_dbg_command_write(struct file *filp,
 			goto command_write_done;
 		}
 
-		vsi = i40e_dbg_find_vsi(pf, vsi_seid);
-		if (!vsi) {
-			dev_info(&pf->pdev->dev,
-				 "add relay: VSI %d not found\n", vsi_seid);
-			goto command_write_done;
-		}
-
-		for (i = 0; i < I40E_MAX_VEB; i++)
-			if (pf->veb[i] && pf->veb[i]->seid == uplink_seid)
-				break;
-		if (i >= I40E_MAX_VEB && uplink_seid != 0 &&
-		    uplink_seid != pf->mac_seid) {
+		if (uplink_seid != 0 && uplink_seid != pf->mac_seid) {
 			dev_info(&pf->pdev->dev,
 				 "add relay: relay uplink %d not found\n",
 				 uplink_seid);
 			goto command_write_done;
+		} else if (uplink_seid) {
+			vsi = i40e_pf_get_vsi_by_seid(pf, vsi_seid);
+			if (!vsi) {
+				dev_info(&pf->pdev->dev,
+					 "add relay: VSI %d not found\n",
+					 vsi_seid);
+				goto command_write_done;
+			}
+			enabled_tc = vsi->tc_config.enabled_tc;
+		} else if (vsi_seid) {
+			dev_info(&pf->pdev->dev,
+				 "add relay: VSI must be 0 for floating relay\n");
+			goto command_write_done;
 		}
 
-		veb = i40e_veb_setup(pf, 0, uplink_seid, vsi_seid,
-				     vsi->tc_config.enabled_tc);
+		veb = i40e_veb_setup(pf, uplink_seid, vsi_seid, enabled_tc);
 		if (veb)
 			dev_info(&pf->pdev->dev, "added relay %d\n", veb->seid);
 		else
 			dev_info(&pf->pdev->dev, "add relay failed\n");
 
 	} else if (strncmp(cmd_buf, "del relay", 9) == 0) {
+		struct i40e_veb *veb;
 		int i;
+
 		cnt = sscanf(&cmd_buf[9], "%i", &veb_seid);
 		if (cnt != 1) {
 			dev_info(&pf->pdev->dev,
@@ -884,9 +851,10 @@ static ssize_t i40e_dbg_command_write(struct file *filp,
 		}
 
 		/* find the veb */
-		for (i = 0; i < I40E_MAX_VEB; i++)
-			if (pf->veb[i] && pf->veb[i]->seid == veb_seid)
+		i40e_pf_for_each_veb(pf, i, veb)
+			if (veb->seid == veb_seid)
 				break;
+
 		if (i >= I40E_MAX_VEB) {
 			dev_info(&pf->pdev->dev,
 				 "del relay: relay %d not found\n", veb_seid);
@@ -894,11 +862,11 @@ static ssize_t i40e_dbg_command_write(struct file *filp,
 		}
 
 		dev_info(&pf->pdev->dev, "deleting relay %d\n", veb_seid);
-		i40e_veb_release(pf->veb[i]);
+		i40e_veb_release(veb);
 	} else if (strncmp(cmd_buf, "add pvid", 8) == 0) {
-		i40e_status ret;
-		u16 vid;
 		unsigned int v;
+		int ret;
+		u16 vid;
 
 		cnt = sscanf(&cmd_buf[8], "%i %u", &vsi_seid, &v);
 		if (cnt != 2) {
@@ -973,13 +941,19 @@ static ssize_t i40e_dbg_command_write(struct file *filp,
 				cnt = sscanf(&cmd_buf[12], "%i %i %i",
 					     &vsi_seid, &ring_id, &desc_n);
 				i40e_dbg_dump_desc(cnt, vsi_seid, ring_id,
-						   desc_n, pf, true);
+						   desc_n, pf, RING_TYPE_RX);
 			} else if (strncmp(&cmd_buf[10], "tx", 2)
 					== 0) {
 				cnt = sscanf(&cmd_buf[12], "%i %i %i",
 					     &vsi_seid, &ring_id, &desc_n);
 				i40e_dbg_dump_desc(cnt, vsi_seid, ring_id,
-						   desc_n, pf, false);
+						   desc_n, pf, RING_TYPE_TX);
+			} else if (strncmp(&cmd_buf[10], "xdp", 3)
+					== 0) {
+				cnt = sscanf(&cmd_buf[13], "%i %i %i",
+					     &vsi_seid, &ring_id, &desc_n);
+				i40e_dbg_dump_desc(cnt, vsi_seid, ring_id,
+						   desc_n, pf, RING_TYPE_XDP);
 			} else if (strncmp(&cmd_buf[10], "aq", 2) == 0) {
 				i40e_dbg_dump_aq_desc(pf);
 			} else {
@@ -987,6 +961,8 @@ static ssize_t i40e_dbg_command_write(struct file *filp,
 					 "dump desc tx <vsi_seid> <ring_id> [<desc_n>]\n");
 				dev_info(&pf->pdev->dev,
 					 "dump desc rx <vsi_seid> <ring_id> [<desc_n>]\n");
+				dev_info(&pf->pdev->dev,
+					 "dump desc xdp <vsi_seid> <ring_id> [<desc_n>]\n");
 				dev_info(&pf->pdev->dev, "dump desc aq\n");
 			}
 		} else if (strncmp(&cmd_buf[5], "reset stats", 11) == 0) {
@@ -998,9 +974,6 @@ static ssize_t i40e_dbg_command_write(struct file *filp,
 				 "emp reset count: %d\n", pf->empr_count);
 			dev_info(&pf->pdev->dev,
 				 "pf reset count: %d\n", pf->pfr_count);
-			dev_info(&pf->pdev->dev,
-				 "pf tx sluggish count: %d\n",
-				 pf->tx_sluggish_count);
 		} else if (strncmp(&cmd_buf[5], "port", 4) == 0) {
 			struct i40e_aqc_query_port_ets_config_resp *bw_data;
 			struct i40e_dcbx_config *cfg =
@@ -1018,7 +991,7 @@ static ssize_t i40e_dbg_command_write(struct file *filp,
 				goto command_write_done;
 			}
 
-			vsi = pf->vsi[pf->lan_vsi];
+			vsi = i40e_pf_get_main_vsi(pf);
 			switch_id =
 				le16_to_cpu(vsi->info.switch_id) &
 					    I40E_AQ_VSI_SW_ID_MASK;
@@ -1157,7 +1130,7 @@ static ssize_t i40e_dbg_command_write(struct file *filp,
 			buff = NULL;
 		} else {
 			dev_info(&pf->pdev->dev,
-				 "dump desc tx <vsi_seid> <ring_id> [<desc_n>], dump desc rx <vsi_seid> <ring_id> [<desc_n>],\n");
+				 "dump desc tx <vsi_seid> <ring_id> [<desc_n>], dump desc rx <vsi_seid> <ring_id> [<desc_n>], dump desc xdp <vsi_seid> <ring_id> [<desc_n>],\n");
 			dev_info(&pf->pdev->dev, "dump switch\n");
 			dev_info(&pf->pdev->dev, "dump vsi [seid]\n");
 			dev_info(&pf->pdev->dev, "dump reset stats\n");
@@ -1178,10 +1151,6 @@ static ssize_t i40e_dbg_command_write(struct file *filp,
 		dev_info(&pf->pdev->dev, "debugfs: forcing GlobR\n");
 		i40e_do_reset_safe(pf, BIT(__I40E_GLOBAL_RESET_REQUESTED));
 
-	} else if (strncmp(cmd_buf, "empr", 4) == 0) {
-		dev_info(&pf->pdev->dev, "debugfs: forcing EMPR\n");
-		i40e_do_reset_safe(pf, BIT(__I40E_EMP_RESET_REQUESTED));
-
 	} else if (strncmp(cmd_buf, "read", 4) == 0) {
 		u32 address;
 		u32 value;
@@ -1228,8 +1197,8 @@ static ssize_t i40e_dbg_command_write(struct file *filp,
 			if (cnt == 0) {
 				int i;
 
-				for (i = 0; i < pf->num_alloc_vsi; i++)
-					i40e_vsi_reset_stats(pf->vsi[i]);
+				i40e_pf_for_each_vsi(pf, i, vsi)
+					i40e_vsi_reset_stats(vsi);
 				dev_info(&pf->pdev->dev, "vsi clear stats called for all vsi's\n");
 			} else if (cnt == 1) {
 				vsi = i40e_dbg_find_vsi(pf, vsi_seid);
@@ -1257,10 +1226,10 @@ static ssize_t i40e_dbg_command_write(struct file *filp,
 			dev_info(&pf->pdev->dev, "clear_stats vsi [seid] or clear_stats port\n");
 		}
 	} else if (strncmp(cmd_buf, "send aq_cmd", 11) == 0) {
-		struct i40e_aq_desc *desc;
-		i40e_status ret;
+		struct libie_aq_desc *desc;
+		int ret;
 
-		desc = kzalloc(sizeof(struct i40e_aq_desc), GFP_KERNEL);
+		desc = kzalloc(sizeof(*desc), GFP_KERNEL);
 		if (!desc)
 			goto command_write_done;
 		cnt = sscanf(&cmd_buf[11],
@@ -1268,10 +1237,10 @@ static ssize_t i40e_dbg_command_write(struct file *filp,
 			     &desc->flags,
 			     &desc->opcode, &desc->datalen, &desc->retval,
 			     &desc->cookie_high, &desc->cookie_low,
-			     &desc->params.internal.param0,
-			     &desc->params.internal.param1,
-			     &desc->params.internal.param2,
-			     &desc->params.internal.param3);
+			     &desc->params.generic.param0,
+			     &desc->params.generic.param1,
+			     &desc->params.generic.addr_high,
+			     &desc->params.generic.addr_low);
 		if (cnt != 10) {
 			dev_info(&pf->pdev->dev,
 				 "send aq_cmd: bad command string, cnt=%d\n",
@@ -1283,7 +1252,7 @@ static ssize_t i40e_dbg_command_write(struct file *filp,
 		ret = i40e_asq_send_command(&pf->hw, desc, NULL, 0, NULL);
 		if (!ret) {
 			dev_info(&pf->pdev->dev, "AQ command sent Status : Success\n");
-		} else if (ret == I40E_ERR_ADMIN_QUEUE_ERROR) {
+		} else if (ret == -EIO) {
 			dev_info(&pf->pdev->dev,
 				 "AQ command send failed Opcode %x AQ Error: %d\n",
 				 desc->opcode, pf->hw.aq.asq_last_status);
@@ -1296,19 +1265,19 @@ static ssize_t i40e_dbg_command_write(struct file *filp,
 			 "AQ desc WB 0x%04x 0x%04x 0x%04x 0x%04x 0x%08x 0x%08x 0x%08x 0x%08x 0x%08x 0x%08x\n",
 			 desc->flags, desc->opcode, desc->datalen, desc->retval,
 			 desc->cookie_high, desc->cookie_low,
-			 desc->params.internal.param0,
-			 desc->params.internal.param1,
-			 desc->params.internal.param2,
-			 desc->params.internal.param3);
+			 desc->params.generic.param0,
+			 desc->params.generic.param1,
+			 desc->params.generic.addr_high,
+			 desc->params.generic.addr_low);
 		kfree(desc);
 		desc = NULL;
 	} else if (strncmp(cmd_buf, "send indirect aq_cmd", 20) == 0) {
-		struct i40e_aq_desc *desc;
-		i40e_status ret;
+		struct libie_aq_desc *desc;
 		u16 buffer_len;
 		u8 *buff;
+		int ret;
 
-		desc = kzalloc(sizeof(struct i40e_aq_desc), GFP_KERNEL);
+		desc = kzalloc(sizeof(*desc), GFP_KERNEL);
 		if (!desc)
 			goto command_write_done;
 		cnt = sscanf(&cmd_buf[20],
@@ -1316,10 +1285,10 @@ static ssize_t i40e_dbg_command_write(struct file *filp,
 			     &desc->flags,
 			     &desc->opcode, &desc->datalen, &desc->retval,
 			     &desc->cookie_high, &desc->cookie_low,
-			     &desc->params.internal.param0,
-			     &desc->params.internal.param1,
-			     &desc->params.internal.param2,
-			     &desc->params.internal.param3,
+			     &desc->params.generic.param0,
+			     &desc->params.generic.param1,
+			     &desc->params.generic.addr_high,
+			     &desc->params.generic.addr_low,
 			     &buffer_len);
 		if (cnt != 11) {
 			dev_info(&pf->pdev->dev,
@@ -1339,12 +1308,12 @@ static ssize_t i40e_dbg_command_write(struct file *filp,
 			desc = NULL;
 			goto command_write_done;
 		}
-		desc->flags |= cpu_to_le16((u16)I40E_AQ_FLAG_BUF);
+		desc->flags |= cpu_to_le16((u16)LIBIE_AQ_FLAG_BUF);
 		ret = i40e_asq_send_command(&pf->hw, desc, buff,
 					    buffer_len, NULL);
 		if (!ret) {
 			dev_info(&pf->pdev->dev, "AQ command sent Status : Success\n");
-		} else if (ret == I40E_ERR_ADMIN_QUEUE_ERROR) {
+		} else if (ret == -EIO) {
 			dev_info(&pf->pdev->dev,
 				 "AQ command send failed Opcode %x AQ Error: %d\n",
 				 desc->opcode, pf->hw.aq.asq_last_status);
@@ -1357,10 +1326,10 @@ static ssize_t i40e_dbg_command_write(struct file *filp,
 			 "AQ desc WB 0x%04x 0x%04x 0x%04x 0x%04x 0x%08x 0x%08x 0x%08x 0x%08x 0x%08x 0x%08x\n",
 			 desc->flags, desc->opcode, desc->datalen, desc->retval,
 			 desc->cookie_high, desc->cookie_low,
-			 desc->params.internal.param0,
-			 desc->params.internal.param1,
-			 desc->params.internal.param2,
-			 desc->params.internal.param3);
+			 desc->params.generic.param0,
+			 desc->params.generic.param1,
+			 desc->params.generic.addr_high,
+			 desc->params.generic.addr_low);
 		print_hex_dump(KERN_INFO, "AQ buffer WB: ",
 			       DUMP_PREFIX_OFFSET, 16, 1,
 			       buff, buffer_len, true);
@@ -1372,10 +1341,13 @@ static ssize_t i40e_dbg_command_write(struct file *filp,
 		dev_info(&pf->pdev->dev, "FD current total filter count for this interface: %d\n",
 			 i40e_get_current_fd_count(pf));
 	} else if (strncmp(cmd_buf, "lldp", 4) == 0) {
+		/* Get main VSI */
+		struct i40e_vsi *main_vsi = i40e_pf_get_main_vsi(pf);
+
 		if (strncmp(&cmd_buf[5], "stop", 4) == 0) {
 			int ret;
 
-			ret = i40e_aq_stop_lldp(&pf->hw, false, NULL);
+			ret = i40e_aq_stop_lldp(&pf->hw, false, false, NULL);
 			if (ret) {
 				dev_info(&pf->pdev->dev,
 					 "Stop LLDP AQ command failed =0x%x\n",
@@ -1383,10 +1355,9 @@ static ssize_t i40e_dbg_command_write(struct file *filp,
 				goto command_write_done;
 			}
 			ret = i40e_aq_add_rem_control_packet_filter(&pf->hw,
-						pf->hw.mac.addr,
-						I40E_ETH_P_LLDP, 0,
-						pf->vsi[pf->lan_vsi]->seid,
-						0, true, NULL, NULL);
+						pf->hw.mac.addr, ETH_P_LLDP, 0,
+						main_vsi->seid, 0, true, NULL,
+						NULL);
 			if (ret) {
 				dev_info(&pf->pdev->dev,
 					"%s: Add Control Packet Filter AQ command failed =0x%x\n",
@@ -1401,10 +1372,9 @@ static ssize_t i40e_dbg_command_write(struct file *filp,
 			int ret;
 
 			ret = i40e_aq_add_rem_control_packet_filter(&pf->hw,
-						pf->hw.mac.addr,
-						I40E_ETH_P_LLDP, 0,
-						pf->vsi[pf->lan_vsi]->seid,
-						0, false, NULL, NULL);
+						pf->hw.mac.addr, ETH_P_LLDP, 0,
+						main_vsi->seid, 0, false, NULL,
+						NULL);
 			if (ret) {
 				dev_info(&pf->pdev->dev,
 					"%s: Remove Control Packet Filter AQ command failed =0x%x\n",
@@ -1412,7 +1382,7 @@ static ssize_t i40e_dbg_command_write(struct file *filp,
 				/* Continue and start FW LLDP anyways */
 			}
 
-			ret = i40e_aq_start_lldp(&pf->hw, NULL);
+			ret = i40e_aq_start_lldp(&pf->hw, false, NULL);
 			if (ret) {
 				dev_info(&pf->pdev->dev,
 					 "Start LLDP AQ command failed =0x%x\n",
@@ -1577,6 +1547,7 @@ static ssize_t i40e_dbg_command_write(struct file *filp,
 		dev_info(&pf->pdev->dev, "  dump vsi [seid]\n");
 		dev_info(&pf->pdev->dev, "  dump desc tx <vsi_seid> <ring_id> [<desc_n>]\n");
 		dev_info(&pf->pdev->dev, "  dump desc rx <vsi_seid> <ring_id> [<desc_n>]\n");
+		dev_info(&pf->pdev->dev, "  dump desc xdp <vsi_seid> <ring_id> [<desc_n>]\n");
 		dev_info(&pf->pdev->dev, "  dump desc aq\n");
 		dev_info(&pf->pdev->dev, "  dump reset stats\n");
 		dev_info(&pf->pdev->dev, "  dump debug fwdata <cluster_id> <table_id> <index>\n");
@@ -1608,7 +1579,6 @@ command_write_done:
 static const struct file_operations i40e_dbg_command_fops = {
 	.owner = THIS_MODULE,
 	.open =  simple_open,
-	.read =  i40e_dbg_command_read,
 	.write = i40e_dbg_command_write,
 };
 
@@ -1617,47 +1587,6 @@ static const struct file_operations i40e_dbg_command_fops = {
  * The netdev_ops entry in debugfs is for giving the driver commands
  * to be executed from the netdev operations.
  **************************************************************/
-static char i40e_dbg_netdev_ops_buf[256] = "";
-
-/**
- * i40e_dbg_netdev_ops - read for netdev_ops datum
- * @filp: the opened file
- * @buffer: where to write the data for the user to read
- * @count: the size of the user's buffer
- * @ppos: file position offset
- **/
-static ssize_t i40e_dbg_netdev_ops_read(struct file *filp, char __user *buffer,
-					size_t count, loff_t *ppos)
-{
-	struct i40e_pf *pf = filp->private_data;
-	int bytes_not_copied;
-	int buf_size = 256;
-	char *buf;
-	int len;
-
-	/* don't allow partal reads */
-	if (*ppos != 0)
-		return 0;
-	if (count < buf_size)
-		return -ENOSPC;
-
-	buf = kzalloc(buf_size, GFP_KERNEL);
-	if (!buf)
-		return -ENOSPC;
-
-	len = snprintf(buf, buf_size, "%s: %s\n",
-		       pf->vsi[pf->lan_vsi]->netdev->name,
-		       i40e_dbg_netdev_ops_buf);
-
-	bytes_not_copied = copy_to_user(buffer, buf, len);
-	kfree(buf);
-
-	if (bytes_not_copied)
-		return -EFAULT;
-
-	*ppos = len;
-	return len;
-}
 
 /**
  * i40e_dbg_netdev_ops_write - write into netdev_ops datum
@@ -1671,58 +1600,36 @@ static ssize_t i40e_dbg_netdev_ops_write(struct file *filp,
 					 size_t count, loff_t *ppos)
 {
 	struct i40e_pf *pf = filp->private_data;
+	char *cmd_buf, *buf_tmp;
 	int bytes_not_copied;
 	struct i40e_vsi *vsi;
-	char *buf_tmp;
 	int vsi_seid;
 	int i, cnt;
 
 	/* don't allow partial writes */
 	if (*ppos != 0)
 		return 0;
-	if (count >= sizeof(i40e_dbg_netdev_ops_buf))
-		return -ENOSPC;
 
-	memset(i40e_dbg_netdev_ops_buf, 0, sizeof(i40e_dbg_netdev_ops_buf));
-	bytes_not_copied = copy_from_user(i40e_dbg_netdev_ops_buf,
-					  buffer, count);
-	if (bytes_not_copied)
+	cmd_buf = kzalloc(count + 1, GFP_KERNEL);
+	if (!cmd_buf)
+		return count;
+	bytes_not_copied = copy_from_user(cmd_buf, buffer, count);
+	if (bytes_not_copied) {
+		kfree(cmd_buf);
 		return -EFAULT;
-	i40e_dbg_netdev_ops_buf[count] = '\0';
+	}
+	cmd_buf[count] = '\0';
 
-	buf_tmp = strchr(i40e_dbg_netdev_ops_buf, '\n');
+	buf_tmp = strchr(cmd_buf, '\n');
 	if (buf_tmp) {
 		*buf_tmp = '\0';
-		count = buf_tmp - i40e_dbg_netdev_ops_buf + 1;
+		count = buf_tmp - cmd_buf + 1;
 	}
 
-	if (strncmp(i40e_dbg_netdev_ops_buf, "tx_timeout", 10) == 0) {
-		cnt = sscanf(&i40e_dbg_netdev_ops_buf[11], "%i", &vsi_seid);
-		if (cnt != 1) {
-			dev_info(&pf->pdev->dev, "tx_timeout <vsi_seid>\n");
-			goto netdev_ops_write_done;
-		}
-		vsi = i40e_dbg_find_vsi(pf, vsi_seid);
-		if (!vsi) {
-			dev_info(&pf->pdev->dev,
-				 "tx_timeout: VSI %d not found\n", vsi_seid);
-		} else if (!vsi->netdev) {
-			dev_info(&pf->pdev->dev, "tx_timeout: no netdev for VSI %d\n",
-				 vsi_seid);
-		} else if (test_bit(__I40E_VSI_DOWN, vsi->state)) {
-			dev_info(&pf->pdev->dev, "tx_timeout: VSI %d not UP\n",
-				 vsi_seid);
-		} else if (rtnl_trylock()) {
-			vsi->netdev->netdev_ops->ndo_tx_timeout(vsi->netdev);
-			rtnl_unlock();
-			dev_info(&pf->pdev->dev, "tx_timeout called\n");
-		} else {
-			dev_info(&pf->pdev->dev, "Could not acquire RTNL - please try again\n");
-		}
-	} else if (strncmp(i40e_dbg_netdev_ops_buf, "change_mtu", 10) == 0) {
+	if (strncmp(cmd_buf, "change_mtu", 10) == 0) {
 		int mtu;
 
-		cnt = sscanf(&i40e_dbg_netdev_ops_buf[11], "%i %i",
+		cnt = sscanf(&cmd_buf[11], "%i %i",
 			     &vsi_seid, &mtu);
 		if (cnt != 2) {
 			dev_info(&pf->pdev->dev, "change_mtu <vsi_seid> <mtu>\n");
@@ -1744,8 +1651,8 @@ static ssize_t i40e_dbg_netdev_ops_write(struct file *filp,
 			dev_info(&pf->pdev->dev, "Could not acquire RTNL - please try again\n");
 		}
 
-	} else if (strncmp(i40e_dbg_netdev_ops_buf, "set_rx_mode", 11) == 0) {
-		cnt = sscanf(&i40e_dbg_netdev_ops_buf[11], "%i", &vsi_seid);
+	} else if (strncmp(cmd_buf, "set_rx_mode", 11) == 0) {
+		cnt = sscanf(&cmd_buf[11], "%i", &vsi_seid);
 		if (cnt != 1) {
 			dev_info(&pf->pdev->dev, "set_rx_mode <vsi_seid>\n");
 			goto netdev_ops_write_done;
@@ -1765,8 +1672,8 @@ static ssize_t i40e_dbg_netdev_ops_write(struct file *filp,
 			dev_info(&pf->pdev->dev, "Could not acquire RTNL - please try again\n");
 		}
 
-	} else if (strncmp(i40e_dbg_netdev_ops_buf, "napi", 4) == 0) {
-		cnt = sscanf(&i40e_dbg_netdev_ops_buf[4], "%i", &vsi_seid);
+	} else if (strncmp(cmd_buf, "napi", 4) == 0) {
+		cnt = sscanf(&cmd_buf[4], "%i", &vsi_seid);
 		if (cnt != 1) {
 			dev_info(&pf->pdev->dev, "napi <vsi_seid>\n");
 			goto netdev_ops_write_done;
@@ -1784,22 +1691,20 @@ static ssize_t i40e_dbg_netdev_ops_write(struct file *filp,
 			dev_info(&pf->pdev->dev, "napi called\n");
 		}
 	} else {
-		dev_info(&pf->pdev->dev, "unknown command '%s'\n",
-			 i40e_dbg_netdev_ops_buf);
+		dev_info(&pf->pdev->dev, "unknown command '%s'\n", cmd_buf);
 		dev_info(&pf->pdev->dev, "available commands\n");
-		dev_info(&pf->pdev->dev, "  tx_timeout <vsi_seid>\n");
 		dev_info(&pf->pdev->dev, "  change_mtu <vsi_seid> <mtu>\n");
 		dev_info(&pf->pdev->dev, "  set_rx_mode <vsi_seid>\n");
 		dev_info(&pf->pdev->dev, "  napi <vsi_seid>\n");
 	}
 netdev_ops_write_done:
+	kfree(cmd_buf);
 	return count;
 }
 
 static const struct file_operations i40e_dbg_netdev_ops_fops = {
 	.owner = THIS_MODULE,
 	.open = simple_open,
-	.read = i40e_dbg_netdev_ops_read,
 	.write = i40e_dbg_netdev_ops_write,
 };
 
@@ -1809,29 +1714,15 @@ static const struct file_operations i40e_dbg_netdev_ops_fops = {
  **/
 void i40e_dbg_pf_init(struct i40e_pf *pf)
 {
-	struct dentry *pfile;
 	const char *name = pci_name(pf->pdev);
-	const struct device *dev = &pf->pdev->dev;
 
 	pf->i40e_dbg_pf = debugfs_create_dir(name, i40e_dbg_root);
-	if (!pf->i40e_dbg_pf)
-		return;
-
-	pfile = debugfs_create_file("command", 0600, pf->i40e_dbg_pf, pf,
-				    &i40e_dbg_command_fops);
-	if (!pfile)
-		goto create_failed;
 
-	pfile = debugfs_create_file("netdev_ops", 0600, pf->i40e_dbg_pf, pf,
-				    &i40e_dbg_netdev_ops_fops);
-	if (!pfile)
-		goto create_failed;
+	debugfs_create_file("command", 0600, pf->i40e_dbg_pf, pf,
+			    &i40e_dbg_command_fops);
 
-	return;
-
-create_failed:
-	dev_info(dev, "debugfs dir/file for %s failed\n", name);
-	debugfs_remove_recursive(pf->i40e_dbg_pf);
+	debugfs_create_file("netdev_ops", 0600, pf->i40e_dbg_pf, pf,
+			    &i40e_dbg_netdev_ops_fops);
 }
 
 /**
@@ -1850,7 +1741,7 @@ void i40e_dbg_pf_exit(struct i40e_pf *pf)
 void i40e_dbg_init(void)
 {
 	i40e_dbg_root = debugfs_create_dir(i40e_driver_name, NULL);
-	if (!i40e_dbg_root)
+	if (IS_ERR(i40e_dbg_root))
 		pr_info("init of debugfs failed\n");
 }
 
diff --git a/drivers/net/ethernet/intel/i40e/i40e_devids.h b/drivers/net/ethernet/intel/i40e/i40e_devids.h
index 8e46098bad57..d9c51a238dcc 100644
--- a/drivers/net/ethernet/intel/i40e/i40e_devids.h
+++ b/drivers/net/ethernet/intel/i40e/i40e_devids.h
@@ -1,33 +1,12 @@
-/*******************************************************************************
- *
- * Intel Ethernet Controller XL710 Family Linux Driver
- * Copyright(c) 2013 - 2015 Intel Corporation.
- *
- * This program is free software; you can redistribute it and/or modify it
- * under the terms and conditions of the GNU General Public License,
- * version 2, as published by the Free Software Foundation.
- *
- * This program is distributed in the hope it will be useful, but WITHOUT
- * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
- * FITNESS FOR A PARTICULAR PURPOSE.  See the GNU General Public License for
- * more details.
- *
- * You should have received a copy of the GNU General Public License along
- * with this program.  If not, see <http://www.gnu.org/licenses/>.
- *
- * The full GNU General Public License is included in this distribution in
- * the file called "COPYING".
- *
- * Contact Information:
- * e1000-devel Mailing List <e1000-devel@lists.sourceforge.net>
- * Intel Corporation, 5200 N.E. Elam Young Parkway, Hillsboro, OR 97124-6497
- *
- ******************************************************************************/
+/* SPDX-License-Identifier: GPL-2.0 */
+/* Copyright(c) 2013 - 2018 Intel Corporation. */
 
 #ifndef _I40E_DEVIDS_H_
 #define _I40E_DEVIDS_H_
 
 /* Device IDs */
+#define I40E_DEV_ID_X710_N3000		0x0CF8
+#define I40E_DEV_ID_XXV710_N3000	0x0D58
 #define I40E_DEV_ID_SFP_XL710		0x1572
 #define I40E_DEV_ID_QEMU		0x1574
 #define I40E_DEV_ID_KX_B		0x1580
@@ -41,15 +20,22 @@
 #define I40E_DEV_ID_10G_BASE_T4		0x1589
 #define I40E_DEV_ID_25G_B		0x158A
 #define I40E_DEV_ID_25G_SFP28		0x158B
+#define I40E_DEV_ID_10G_BASE_T_BC	0x15FF
+#define I40E_DEV_ID_10G_B		0x104F
+#define I40E_DEV_ID_10G_SFP		0x104E
+#define I40E_DEV_ID_5G_BASE_T_BC	0x101F
+#define I40E_DEV_ID_1G_BASE_T_BC	0x0DD2
+#define I40E_IS_X710TL_DEVICE(d) \
+	(((d) == I40E_DEV_ID_1G_BASE_T_BC) || \
+	 ((d) == I40E_DEV_ID_5G_BASE_T_BC) || \
+	 ((d) == I40E_DEV_ID_10G_BASE_T_BC))
 #define I40E_DEV_ID_KX_X722		0x37CE
 #define I40E_DEV_ID_QSFP_X722		0x37CF
 #define I40E_DEV_ID_SFP_X722		0x37D0
 #define I40E_DEV_ID_1G_BASE_T_X722	0x37D1
 #define I40E_DEV_ID_10G_BASE_T_X722	0x37D2
 #define I40E_DEV_ID_SFP_I_X722		0x37D3
+#define I40E_DEV_ID_SFP_X722_A		0x0DDA
 
-#define i40e_is_40G_device(d)		((d) == I40E_DEV_ID_QSFP_A  || \
-					 (d) == I40E_DEV_ID_QSFP_B  || \
-					 (d) == I40E_DEV_ID_QSFP_C)
 
 #endif /* _I40E_DEVIDS_H_ */
diff --git a/drivers/net/ethernet/intel/i40e/i40e_devlink.c b/drivers/net/ethernet/intel/i40e/i40e_devlink.c
new file mode 100644
index 000000000000..cc4e9e2addb7
--- /dev/null
+++ b/drivers/net/ethernet/intel/i40e/i40e_devlink.c
@@ -0,0 +1,235 @@
+// SPDX-License-Identifier: GPL-2.0
+/* Copyright(c) 2023 Intel Corporation. */
+
+#include <net/devlink.h>
+#include "i40e.h"
+#include "i40e_devlink.h"
+
+static void i40e_info_get_dsn(struct i40e_pf *pf, char *buf, size_t len)
+{
+	u8 dsn[8];
+
+	put_unaligned_be64(pci_get_dsn(pf->pdev), dsn);
+
+	snprintf(buf, len, "%8phD", dsn);
+}
+
+static void i40e_info_fw_mgmt(struct i40e_hw *hw, char *buf, size_t len)
+{
+	struct i40e_adminq_info *aq = &hw->aq;
+
+	snprintf(buf, len, "%u.%u", aq->fw_maj_ver, aq->fw_min_ver);
+}
+
+static void i40e_info_fw_mgmt_build(struct i40e_hw *hw, char *buf, size_t len)
+{
+	struct i40e_adminq_info *aq = &hw->aq;
+
+	snprintf(buf, len, "%05d", aq->fw_build);
+}
+
+static void i40e_info_fw_api(struct i40e_hw *hw, char *buf, size_t len)
+{
+	struct i40e_adminq_info *aq = &hw->aq;
+
+	snprintf(buf, len, "%u.%u", aq->api_maj_ver, aq->api_min_ver);
+}
+
+static void i40e_info_pba(struct i40e_hw *hw, char *buf, size_t len)
+{
+	buf[0] = '\0';
+	if (hw->pba_id)
+		strscpy(buf, hw->pba_id, len);
+}
+
+enum i40e_devlink_version_type {
+	I40E_DL_VERSION_FIXED,
+	I40E_DL_VERSION_RUNNING,
+};
+
+static int i40e_devlink_info_put(struct devlink_info_req *req,
+				 enum i40e_devlink_version_type type,
+				 const char *key, const char *value)
+{
+	if (!strlen(value))
+		return 0;
+
+	switch (type) {
+	case I40E_DL_VERSION_FIXED:
+		return devlink_info_version_fixed_put(req, key, value);
+	case I40E_DL_VERSION_RUNNING:
+		return devlink_info_version_running_put(req, key, value);
+	}
+	return 0;
+}
+
+static int i40e_devlink_info_get(struct devlink *dl,
+				 struct devlink_info_req *req,
+				 struct netlink_ext_ack *extack)
+{
+	struct i40e_pf *pf = devlink_priv(dl);
+	struct i40e_hw *hw = &pf->hw;
+	char buf[32];
+	int err;
+
+	i40e_info_get_dsn(pf, buf, sizeof(buf));
+	err = devlink_info_serial_number_put(req, buf);
+	if (err)
+		return err;
+
+	i40e_info_fw_mgmt(hw, buf, sizeof(buf));
+	err = i40e_devlink_info_put(req, I40E_DL_VERSION_RUNNING,
+				    DEVLINK_INFO_VERSION_GENERIC_FW_MGMT, buf);
+	if (err)
+		return err;
+
+	i40e_info_fw_mgmt_build(hw, buf, sizeof(buf));
+	err = i40e_devlink_info_put(req, I40E_DL_VERSION_RUNNING,
+				    "fw.mgmt.build", buf);
+	if (err)
+		return err;
+
+	i40e_info_fw_api(hw, buf, sizeof(buf));
+	err = i40e_devlink_info_put(req, I40E_DL_VERSION_RUNNING,
+				    DEVLINK_INFO_VERSION_GENERIC_FW_MGMT_API,
+				    buf);
+	if (err)
+		return err;
+
+	i40e_info_nvm_ver(hw, buf, sizeof(buf));
+	err = i40e_devlink_info_put(req, I40E_DL_VERSION_RUNNING,
+				    "fw.psid.api", buf);
+	if (err)
+		return err;
+
+	i40e_info_eetrack(hw, buf, sizeof(buf));
+	err = i40e_devlink_info_put(req, I40E_DL_VERSION_RUNNING,
+				    DEVLINK_INFO_VERSION_GENERIC_FW_BUNDLE_ID,
+				    buf);
+	if (err)
+		return err;
+
+	i40e_info_civd_ver(hw, buf, sizeof(buf));
+	err = i40e_devlink_info_put(req, I40E_DL_VERSION_RUNNING,
+				    DEVLINK_INFO_VERSION_GENERIC_FW_UNDI, buf);
+	if (err)
+		return err;
+
+	i40e_info_pba(hw, buf, sizeof(buf));
+	err = i40e_devlink_info_put(req, I40E_DL_VERSION_FIXED,
+				    DEVLINK_INFO_VERSION_GENERIC_BOARD_ID, buf);
+
+	return err;
+}
+
+static const struct devlink_ops i40e_devlink_ops = {
+	.info_get = i40e_devlink_info_get,
+};
+
+/**
+ * i40e_alloc_pf - Allocate devlink and return i40e_pf structure pointer
+ * @dev: the device to allocate for
+ *
+ * Allocate a devlink instance for this device and return the private
+ * area as the i40e_pf structure.
+ **/
+struct i40e_pf *i40e_alloc_pf(struct device *dev)
+{
+	struct devlink *devlink;
+
+	devlink = devlink_alloc(&i40e_devlink_ops, sizeof(struct i40e_pf), dev);
+	if (!devlink)
+		return NULL;
+
+	return devlink_priv(devlink);
+}
+
+/**
+ * i40e_free_pf - Free i40e_pf structure and associated devlink
+ * @pf: the PF structure
+ *
+ * Free i40e_pf structure and devlink allocated by devlink_alloc.
+ **/
+void i40e_free_pf(struct i40e_pf *pf)
+{
+	struct devlink *devlink = priv_to_devlink(pf);
+
+	devlink_free(devlink);
+}
+
+/**
+ * i40e_devlink_register - Register devlink interface for this PF
+ * @pf: the PF to register the devlink for.
+ *
+ * Register the devlink instance associated with this physical function.
+ **/
+void i40e_devlink_register(struct i40e_pf *pf)
+{
+	devlink_register(priv_to_devlink(pf));
+}
+
+/**
+ * i40e_devlink_unregister - Unregister devlink resources for this PF.
+ * @pf: the PF structure to cleanup
+ *
+ * Releases resources used by devlink and cleans up associated memory.
+ **/
+void i40e_devlink_unregister(struct i40e_pf *pf)
+{
+	devlink_unregister(priv_to_devlink(pf));
+}
+
+/**
+ * i40e_devlink_set_switch_id - Set unique switch id based on pci dsn
+ * @pf: the PF to create a devlink port for
+ * @ppid: struct with switch id information
+ */
+static void i40e_devlink_set_switch_id(struct i40e_pf *pf,
+				       struct netdev_phys_item_id *ppid)
+{
+	u64 id = pci_get_dsn(pf->pdev);
+
+	ppid->id_len = sizeof(id);
+	put_unaligned_be64(id, &ppid->id);
+}
+
+/**
+ * i40e_devlink_create_port - Create a devlink port for this PF
+ * @pf: the PF to create a port for
+ *
+ * Create and register a devlink_port for this PF. Note that although each
+ * physical function is connected to a separate devlink instance, the port
+ * will still be numbered according to the physical function id.
+ *
+ * Return: zero on success or an error code on failure.
+ **/
+int i40e_devlink_create_port(struct i40e_pf *pf)
+{
+	struct devlink *devlink = priv_to_devlink(pf);
+	struct devlink_port_attrs attrs = {};
+	struct device *dev = &pf->pdev->dev;
+	int err;
+
+	attrs.flavour = DEVLINK_PORT_FLAVOUR_PHYSICAL;
+	attrs.phys.port_number = pf->hw.pf_id;
+	i40e_devlink_set_switch_id(pf, &attrs.switch_id);
+	devlink_port_attrs_set(&pf->devlink_port, &attrs);
+	err = devlink_port_register(devlink, &pf->devlink_port, pf->hw.pf_id);
+	if (err) {
+		dev_err(dev, "devlink_port_register failed: %d\n", err);
+		return err;
+	}
+
+	return 0;
+}
+
+/**
+ * i40e_devlink_destroy_port - Destroy the devlink_port for this PF
+ * @pf: the PF to cleanup
+ *
+ * Unregisters the devlink_port structure associated with this PF.
+ **/
+void i40e_devlink_destroy_port(struct i40e_pf *pf)
+{
+	devlink_port_unregister(&pf->devlink_port);
+}
diff --git a/drivers/net/ethernet/intel/i40e/i40e_devlink.h b/drivers/net/ethernet/intel/i40e/i40e_devlink.h
new file mode 100644
index 000000000000..469fb3d2ee25
--- /dev/null
+++ b/drivers/net/ethernet/intel/i40e/i40e_devlink.h
@@ -0,0 +1,18 @@
+/* SPDX-License-Identifier: GPL-2.0 */
+/* Copyright (c) 2023, Intel Corporation. */
+
+#ifndef _I40E_DEVLINK_H_
+#define _I40E_DEVLINK_H_
+
+#include <linux/device.h>
+
+struct i40e_pf;
+
+struct i40e_pf *i40e_alloc_pf(struct device *dev);
+void i40e_free_pf(struct i40e_pf *pf);
+void i40e_devlink_register(struct i40e_pf *pf);
+void i40e_devlink_unregister(struct i40e_pf *pf);
+int i40e_devlink_create_port(struct i40e_pf *pf);
+void i40e_devlink_destroy_port(struct i40e_pf *pf);
+
+#endif /* _I40E_DEVLINK_H_ */
diff --git a/drivers/net/ethernet/intel/i40e/i40e_diag.c b/drivers/net/ethernet/intel/i40e/i40e_diag.c
index f141e78d409e..b1ad7c4259b9 100644
--- a/drivers/net/ethernet/intel/i40e/i40e_diag.c
+++ b/drivers/net/ethernet/intel/i40e/i40e_diag.c
@@ -1,28 +1,5 @@
-/*******************************************************************************
- *
- * Intel Ethernet Controller XL710 Family Linux Driver
- * Copyright(c) 2013 - 2014 Intel Corporation.
- *
- * This program is free software; you can redistribute it and/or modify it
- * under the terms and conditions of the GNU General Public License,
- * version 2, as published by the Free Software Foundation.
- *
- * This program is distributed in the hope it will be useful, but WITHOUT
- * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
- * FITNESS FOR A PARTICULAR PURPOSE.  See the GNU General Public License for
- * more details.
- *
- * You should have received a copy of the GNU General Public License along
- * with this program.  If not, see <http://www.gnu.org/licenses/>.
- *
- * The full GNU General Public License is included in this distribution in
- * the file called "COPYING".
- *
- * Contact Information:
- * e1000-devel Mailing List <e1000-devel@lists.sourceforge.net>
- * Intel Corporation, 5200 N.E. Elam Young Parkway, Hillsboro, OR 97124-6497
- *
- ******************************************************************************/
+// SPDX-License-Identifier: GPL-2.0
+/* Copyright(c) 2013 - 2018 Intel Corporation. */
 
 #include "i40e_diag.h"
 #include "i40e_prototype.h"
@@ -33,10 +10,12 @@
  * @reg: reg to be tested
  * @mask: bits to be touched
  **/
-static i40e_status i40e_diag_reg_pattern_test(struct i40e_hw *hw,
-							u32 reg, u32 mask)
+static int i40e_diag_reg_pattern_test(struct i40e_hw *hw,
+				      u32 reg, u32 mask)
 {
-	const u32 patterns[] = {0x5A5A5A5A, 0xA5A5A5A5, 0x00000000, 0xFFFFFFFF};
+	static const u32 patterns[] = {
+		0x5A5A5A5A, 0xA5A5A5A5, 0x00000000, 0xFFFFFFFF
+	};
 	u32 pat, val, orig_val;
 	int i;
 
@@ -49,7 +28,7 @@ static i40e_status i40e_diag_reg_pattern_test(struct i40e_hw *hw,
 			i40e_debug(hw, I40E_DEBUG_DIAG,
 				   "%s: reg pattern test failed - reg 0x%08x pat 0x%08x val 0x%08x\n",
 				   __func__, reg, pat, val);
-			return I40E_ERR_DIAG_TEST_FAILED;
+			return -EIO;
 		}
 	}
 
@@ -59,13 +38,13 @@ static i40e_status i40e_diag_reg_pattern_test(struct i40e_hw *hw,
 		i40e_debug(hw, I40E_DEBUG_DIAG,
 			   "%s: reg restore test failed - reg 0x%08x orig_val 0x%08x val 0x%08x\n",
 			   __func__, reg, orig_val, val);
-		return I40E_ERR_DIAG_TEST_FAILED;
+		return -EIO;
 	}
 
 	return 0;
 }
 
-struct i40e_diag_reg_test_info i40e_reg_list[] = {
+const struct i40e_diag_reg_test_info i40e_reg_list[] = {
 	/* offset               mask         elements   stride */
 	{I40E_QTX_CTL(0),       0x0000FFBF, 1,
 		I40E_QTX_CTL(1) - I40E_QTX_CTL(0)},
@@ -95,31 +74,32 @@ struct i40e_diag_reg_test_info i40e_reg_list[] = {
  *
  * Perform registers diagnostic test
  **/
-i40e_status i40e_diag_reg_test(struct i40e_hw *hw)
+int i40e_diag_reg_test(struct i40e_hw *hw)
 {
-	i40e_status ret_code = 0;
+	int ret_code = 0;
 	u32 reg, mask;
+	u32 elements;
 	u32 i, j;
 
 	for (i = 0; i40e_reg_list[i].offset != 0 &&
 					     !ret_code; i++) {
 
+		elements = i40e_reg_list[i].elements;
 		/* set actual reg range for dynamically allocated resources */
 		if (i40e_reg_list[i].offset == I40E_QTX_CTL(0) &&
 		    hw->func_caps.num_tx_qp != 0)
-			i40e_reg_list[i].elements = hw->func_caps.num_tx_qp;
+			elements = hw->func_caps.num_tx_qp;
 		if ((i40e_reg_list[i].offset == I40E_PFINT_ITRN(0, 0) ||
 		     i40e_reg_list[i].offset == I40E_PFINT_ITRN(1, 0) ||
 		     i40e_reg_list[i].offset == I40E_PFINT_ITRN(2, 0) ||
 		     i40e_reg_list[i].offset == I40E_QINT_TQCTL(0) ||
 		     i40e_reg_list[i].offset == I40E_QINT_RQCTL(0)) &&
 		    hw->func_caps.num_msix_vectors != 0)
-			i40e_reg_list[i].elements =
-				hw->func_caps.num_msix_vectors - 1;
+			elements = hw->func_caps.num_msix_vectors - 1;
 
 		/* test register access */
 		mask = i40e_reg_list[i].mask;
-		for (j = 0; j < i40e_reg_list[i].elements && !ret_code; j++) {
+		for (j = 0; j < elements && !ret_code; j++) {
 			reg = i40e_reg_list[i].offset +
 			      (j * i40e_reg_list[i].stride);
 			ret_code = i40e_diag_reg_pattern_test(hw, reg, mask);
@@ -135,9 +115,9 @@ i40e_status i40e_diag_reg_test(struct i40e_hw *hw)
  *
  * Perform EEPROM diagnostic test
  **/
-i40e_status i40e_diag_eeprom_test(struct i40e_hw *hw)
+int i40e_diag_eeprom_test(struct i40e_hw *hw)
 {
-	i40e_status ret_code;
+	int ret_code;
 	u16 reg_val;
 
 	/* read NVM control word and if NVM valid, validate EEPROM checksum*/
@@ -147,5 +127,5 @@ i40e_status i40e_diag_eeprom_test(struct i40e_hw *hw)
 	     BIT(I40E_SR_CONTROL_WORD_1_SHIFT)))
 		return i40e_validate_nvm_checksum(hw, NULL);
 	else
-		return I40E_ERR_DIAG_TEST_FAILED;
+		return -EIO;
 }
diff --git a/drivers/net/ethernet/intel/i40e/i40e_diag.h b/drivers/net/ethernet/intel/i40e/i40e_diag.h
index 0b5911652084..ab20202a3da3 100644
--- a/drivers/net/ethernet/intel/i40e/i40e_diag.h
+++ b/drivers/net/ethernet/intel/i40e/i40e_diag.h
@@ -1,33 +1,14 @@
-/*******************************************************************************
- *
- * Intel Ethernet Controller XL710 Family Linux Driver
- * Copyright(c) 2013 - 2014 Intel Corporation.
- *
- * This program is free software; you can redistribute it and/or modify it
- * under the terms and conditions of the GNU General Public License,
- * version 2, as published by the Free Software Foundation.
- *
- * This program is distributed in the hope it will be useful, but WITHOUT
- * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
- * FITNESS FOR A PARTICULAR PURPOSE.  See the GNU General Public License for
- * more details.
- *
- * You should have received a copy of the GNU General Public License along
- * with this program.  If not, see <http://www.gnu.org/licenses/>.
- *
- * The full GNU General Public License is included in this distribution in
- * the file called "COPYING".
- *
- * Contact Information:
- * e1000-devel Mailing List <e1000-devel@lists.sourceforge.net>
- * Intel Corporation, 5200 N.E. Elam Young Parkway, Hillsboro, OR 97124-6497
- *
- ******************************************************************************/
+/* SPDX-License-Identifier: GPL-2.0 */
+/* Copyright(c) 2013 - 2018 Intel Corporation. */
 
 #ifndef _I40E_DIAG_H_
 #define _I40E_DIAG_H_
 
-#include "i40e_type.h"
+#include <linux/types.h>
+#include "i40e_adminq_cmd.h"
+
+/* forward-declare the HW struct for the compiler */
+struct i40e_hw;
 
 enum i40e_lb_mode {
 	I40E_LB_MODE_NONE       = 0x0,
@@ -43,9 +24,9 @@ struct i40e_diag_reg_test_info {
 	u32 stride;	/* bytes between each element */
 };
 
-extern struct i40e_diag_reg_test_info i40e_reg_list[];
+extern const struct i40e_diag_reg_test_info i40e_reg_list[];
 
-i40e_status i40e_diag_reg_test(struct i40e_hw *hw);
-i40e_status i40e_diag_eeprom_test(struct i40e_hw *hw);
+int i40e_diag_reg_test(struct i40e_hw *hw);
+int i40e_diag_eeprom_test(struct i40e_hw *hw);
 
 #endif /* _I40E_DIAG_H_ */
diff --git a/drivers/net/ethernet/intel/i40e/i40e_ethtool.c b/drivers/net/ethernet/intel/i40e/i40e_ethtool.c
index 35a246f05520..86c72596617a 100644
--- a/drivers/net/ethernet/intel/i40e/i40e_ethtool.c
+++ b/drivers/net/ethernet/intel/i40e/i40e_ethtool.c
@@ -1,54 +1,243 @@
-/*******************************************************************************
- *
- * Intel Ethernet Controller XL710 Family Linux Driver
- * Copyright(c) 2013 - 2016 Intel Corporation.
- *
- * This program is free software; you can redistribute it and/or modify it
- * under the terms and conditions of the GNU General Public License,
- * version 2, as published by the Free Software Foundation.
- *
- * This program is distributed in the hope it will be useful, but WITHOUT
- * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
- * FITNESS FOR A PARTICULAR PURPOSE.  See the GNU General Public License for
- * more details.
- *
- * You should have received a copy of the GNU General Public License along
- * with this program.  If not, see <http://www.gnu.org/licenses/>.
- *
- * The full GNU General Public License is included in this distribution in
- * the file called "COPYING".
- *
- * Contact Information:
- * e1000-devel Mailing List <e1000-devel@lists.sourceforge.net>
- * Intel Corporation, 5200 N.E. Elam Young Parkway, Hillsboro, OR 97124-6497
- *
- ******************************************************************************/
+// SPDX-License-Identifier: GPL-2.0
+/* Copyright(c) 2013 - 2018 Intel Corporation. */
 
 /* ethtool support for i40e */
 
-#include "i40e.h"
+#include <linux/net/intel/libie/pctype.h>
+#include "i40e_devids.h"
 #include "i40e_diag.h"
+#include "i40e_txrx_common.h"
+#include "i40e_virtchnl_pf.h"
 
+/* ethtool statistics helpers */
+
+/**
+ * struct i40e_stats - definition for an ethtool statistic
+ * @stat_string: statistic name to display in ethtool -S output
+ * @sizeof_stat: the sizeof() the stat, must be no greater than sizeof(u64)
+ * @stat_offset: offsetof() the stat from a base pointer
+ *
+ * This structure defines a statistic to be added to the ethtool stats buffer.
+ * It defines a statistic as offset from a common base pointer. Stats should
+ * be defined in constant arrays using the I40E_STAT macro, with every element
+ * of the array using the same _type for calculating the sizeof_stat and
+ * stat_offset.
+ *
+ * The @sizeof_stat is expected to be sizeof(u8), sizeof(u16), sizeof(u32) or
+ * sizeof(u64). Other sizes are not expected and will produce a WARN_ONCE from
+ * the i40e_add_ethtool_stat() helper function.
+ *
+ * The @stat_string is interpreted as a format string, allowing formatted
+ * values to be inserted while looping over multiple structures for a given
+ * statistics array. Thus, every statistic string in an array should have the
+ * same type and number of format specifiers, to be formatted by variadic
+ * arguments to the i40e_add_stat_string() helper function.
+ **/
 struct i40e_stats {
 	char stat_string[ETH_GSTRING_LEN];
 	int sizeof_stat;
 	int stat_offset;
 };
 
+/* Helper macro to define an i40e_stat structure with proper size and type.
+ * Use this when defining constant statistics arrays. Note that @_type expects
+ * only a type name and is used multiple times.
+ */
 #define I40E_STAT(_type, _name, _stat) { \
 	.stat_string = _name, \
-	.sizeof_stat = FIELD_SIZEOF(_type, _stat), \
+	.sizeof_stat = sizeof_field(_type, _stat), \
 	.stat_offset = offsetof(_type, _stat) \
 }
 
+/* Helper macro for defining some statistics directly copied from the netdev
+ * stats structure.
+ */
 #define I40E_NETDEV_STAT(_net_stat) \
-		I40E_STAT(struct rtnl_link_stats64, #_net_stat, _net_stat)
+	I40E_STAT(struct rtnl_link_stats64, #_net_stat, _net_stat)
+
+/* Helper macro for defining some statistics related to queues */
+#define I40E_QUEUE_STAT(_name, _stat) \
+	I40E_STAT(struct i40e_ring, _name, _stat)
+
+/* Stats associated with a Tx or Rx ring */
+static const struct i40e_stats i40e_gstrings_queue_stats[] = {
+	I40E_QUEUE_STAT("%s-%u.packets", stats.packets),
+	I40E_QUEUE_STAT("%s-%u.bytes", stats.bytes),
+};
+
+/**
+ * i40e_add_one_ethtool_stat - copy the stat into the supplied buffer
+ * @data: location to store the stat value
+ * @pointer: basis for where to copy from
+ * @stat: the stat definition
+ *
+ * Copies the stat data defined by the pointer and stat structure pair into
+ * the memory supplied as data. Used to implement i40e_add_ethtool_stats and
+ * i40e_add_queue_stats. If the pointer is null, data will be zero'd.
+ */
+static void
+i40e_add_one_ethtool_stat(u64 *data, void *pointer,
+			  const struct i40e_stats *stat)
+{
+	char *p;
+
+	if (!pointer) {
+		/* ensure that the ethtool data buffer is zero'd for any stats
+		 * which don't have a valid pointer.
+		 */
+		*data = 0;
+		return;
+	}
+
+	p = (char *)pointer + stat->stat_offset;
+	switch (stat->sizeof_stat) {
+	case sizeof(u64):
+		*data = *((u64 *)p);
+		break;
+	case sizeof(u32):
+		*data = *((u32 *)p);
+		break;
+	case sizeof(u16):
+		*data = *((u16 *)p);
+		break;
+	case sizeof(u8):
+		*data = *((u8 *)p);
+		break;
+	default:
+		WARN_ONCE(1, "unexpected stat size for %s",
+			  stat->stat_string);
+		*data = 0;
+	}
+}
+
+/**
+ * __i40e_add_ethtool_stats - copy stats into the ethtool supplied buffer
+ * @data: ethtool stats buffer
+ * @pointer: location to copy stats from
+ * @stats: array of stats to copy
+ * @size: the size of the stats definition
+ *
+ * Copy the stats defined by the stats array using the pointer as a base into
+ * the data buffer supplied by ethtool. Updates the data pointer to point to
+ * the next empty location for successive calls to __i40e_add_ethtool_stats.
+ * If pointer is null, set the data values to zero and update the pointer to
+ * skip these stats.
+ **/
+static void
+__i40e_add_ethtool_stats(u64 **data, void *pointer,
+			 const struct i40e_stats stats[],
+			 const unsigned int size)
+{
+	unsigned int i;
+
+	for (i = 0; i < size; i++)
+		i40e_add_one_ethtool_stat((*data)++, pointer, &stats[i]);
+}
+
+/**
+ * i40e_add_ethtool_stats - copy stats into ethtool supplied buffer
+ * @data: ethtool stats buffer
+ * @pointer: location where stats are stored
+ * @stats: static const array of stat definitions
+ *
+ * Macro to ease the use of __i40e_add_ethtool_stats by taking a static
+ * constant stats array and passing the ARRAY_SIZE(). This avoids typos by
+ * ensuring that we pass the size associated with the given stats array.
+ *
+ * The parameter @stats is evaluated twice, so parameters with side effects
+ * should be avoided.
+ **/
+#define i40e_add_ethtool_stats(data, pointer, stats) \
+	__i40e_add_ethtool_stats(data, pointer, stats, ARRAY_SIZE(stats))
+
+/**
+ * i40e_add_queue_stats - copy queue statistics into supplied buffer
+ * @data: ethtool stats buffer
+ * @ring: the ring to copy
+ *
+ * Queue statistics must be copied while protected by
+ * u64_stats_fetch_begin, so we can't directly use i40e_add_ethtool_stats.
+ * Assumes that queue stats are defined in i40e_gstrings_queue_stats. If the
+ * ring pointer is null, zero out the queue stat values and update the data
+ * pointer. Otherwise safely copy the stats from the ring into the supplied
+ * buffer and update the data pointer when finished.
+ *
+ * This function expects to be called while under rcu_read_lock().
+ **/
+static void
+i40e_add_queue_stats(u64 **data, struct i40e_ring *ring)
+{
+	const unsigned int size = ARRAY_SIZE(i40e_gstrings_queue_stats);
+	const struct i40e_stats *stats = i40e_gstrings_queue_stats;
+	unsigned int start;
+	unsigned int i;
+
+	/* To avoid invalid statistics values, ensure that we keep retrying
+	 * the copy until we get a consistent value according to
+	 * u64_stats_fetch_retry. But first, make sure our ring is
+	 * non-null before attempting to access its syncp.
+	 */
+	do {
+		start = !ring ? 0 : u64_stats_fetch_begin(&ring->syncp);
+		for (i = 0; i < size; i++) {
+			i40e_add_one_ethtool_stat(&(*data)[i], ring,
+						  &stats[i]);
+		}
+	} while (ring && u64_stats_fetch_retry(&ring->syncp, start));
+
+	/* Once we successfully copy the stats in, update the data pointer */
+	*data += size;
+}
+
+/**
+ * __i40e_add_stat_strings - copy stat strings into ethtool buffer
+ * @p: ethtool supplied buffer
+ * @stats: stat definitions array
+ * @size: size of the stats array
+ *
+ * Format and copy the strings described by stats into the buffer pointed at
+ * by p.
+ **/
+static void __i40e_add_stat_strings(u8 **p, const struct i40e_stats stats[],
+				    const unsigned int size, ...)
+{
+	unsigned int i;
+
+	for (i = 0; i < size; i++) {
+		va_list args;
+
+		va_start(args, size);
+		vsnprintf(*p, ETH_GSTRING_LEN, stats[i].stat_string, args);
+		*p += ETH_GSTRING_LEN;
+		va_end(args);
+	}
+}
+
+/**
+ * i40e_add_stat_strings - copy stat strings into ethtool buffer
+ * @p: ethtool supplied buffer
+ * @stats: stat definitions array
+ *
+ * Format and copy the strings described by the const static stats value into
+ * the buffer pointed at by p.
+ *
+ * The parameter @stats is evaluated twice, so parameters with side effects
+ * should be avoided. Additionally, stats must be an array such that
+ * ARRAY_SIZE can be called on it.
+ **/
+#define i40e_add_stat_strings(p, stats, ...) \
+	__i40e_add_stat_strings(p, stats, ARRAY_SIZE(stats), ## __VA_ARGS__)
+
 #define I40E_PF_STAT(_name, _stat) \
-		I40E_STAT(struct i40e_pf, _name, _stat)
+	I40E_STAT(struct i40e_pf, _name, _stat)
 #define I40E_VSI_STAT(_name, _stat) \
-		I40E_STAT(struct i40e_vsi, _name, _stat)
+	I40E_STAT(struct i40e_vsi, _name, _stat)
 #define I40E_VEB_STAT(_name, _stat) \
-		I40E_STAT(struct i40e_veb, _name, _stat)
+	I40E_STAT(struct i40e_veb, _name, _stat)
+#define I40E_VEB_TC_STAT(_name, _stat) \
+	I40E_STAT(struct i40e_cp_veb_tc_stats, _name, _stat)
+#define I40E_PFC_STAT(_name, _stat) \
+	I40E_STAT(struct i40e_pfc_stats, _name, _stat)
 
 static const struct i40e_stats i40e_gstrings_net_stats[] = {
 	I40E_NETDEV_STAT(rx_packets),
@@ -58,6 +247,7 @@ static const struct i40e_stats i40e_gstrings_net_stats[] = {
 	I40E_NETDEV_STAT(rx_errors),
 	I40E_NETDEV_STAT(tx_errors),
 	I40E_NETDEV_STAT(rx_dropped),
+	I40E_NETDEV_STAT(rx_missed_errors),
 	I40E_NETDEV_STAT(tx_dropped),
 	I40E_NETDEV_STAT(collisions),
 	I40E_NETDEV_STAT(rx_length_errors),
@@ -65,18 +255,32 @@ static const struct i40e_stats i40e_gstrings_net_stats[] = {
 };
 
 static const struct i40e_stats i40e_gstrings_veb_stats[] = {
-	I40E_VEB_STAT("rx_bytes", stats.rx_bytes),
-	I40E_VEB_STAT("tx_bytes", stats.tx_bytes),
-	I40E_VEB_STAT("rx_unicast", stats.rx_unicast),
-	I40E_VEB_STAT("tx_unicast", stats.tx_unicast),
-	I40E_VEB_STAT("rx_multicast", stats.rx_multicast),
-	I40E_VEB_STAT("tx_multicast", stats.tx_multicast),
-	I40E_VEB_STAT("rx_broadcast", stats.rx_broadcast),
-	I40E_VEB_STAT("tx_broadcast", stats.tx_broadcast),
-	I40E_VEB_STAT("rx_discards", stats.rx_discards),
-	I40E_VEB_STAT("tx_discards", stats.tx_discards),
-	I40E_VEB_STAT("tx_errors", stats.tx_errors),
-	I40E_VEB_STAT("rx_unknown_protocol", stats.rx_unknown_protocol),
+	I40E_VEB_STAT("veb.rx_bytes", stats.rx_bytes),
+	I40E_VEB_STAT("veb.tx_bytes", stats.tx_bytes),
+	I40E_VEB_STAT("veb.rx_unicast", stats.rx_unicast),
+	I40E_VEB_STAT("veb.tx_unicast", stats.tx_unicast),
+	I40E_VEB_STAT("veb.rx_multicast", stats.rx_multicast),
+	I40E_VEB_STAT("veb.tx_multicast", stats.tx_multicast),
+	I40E_VEB_STAT("veb.rx_broadcast", stats.rx_broadcast),
+	I40E_VEB_STAT("veb.tx_broadcast", stats.tx_broadcast),
+	I40E_VEB_STAT("veb.rx_discards", stats.rx_discards),
+	I40E_VEB_STAT("veb.tx_discards", stats.tx_discards),
+	I40E_VEB_STAT("veb.tx_errors", stats.tx_errors),
+	I40E_VEB_STAT("veb.rx_unknown_protocol", stats.rx_unknown_protocol),
+};
+
+struct i40e_cp_veb_tc_stats {
+	u64 tc_rx_packets;
+	u64 tc_rx_bytes;
+	u64 tc_tx_packets;
+	u64 tc_tx_bytes;
+};
+
+static const struct i40e_stats i40e_gstrings_veb_tc_stats[] = {
+	I40E_VEB_TC_STAT("veb.tc_%u_tx_packets", tc_tx_packets),
+	I40E_VEB_TC_STAT("veb.tc_%u_tx_bytes", tc_tx_bytes),
+	I40E_VEB_TC_STAT("veb.tc_%u_rx_packets", tc_rx_packets),
+	I40E_VEB_TC_STAT("veb.tc_%u_rx_bytes", tc_rx_bytes),
 };
 
 static const struct i40e_stats i40e_gstrings_misc_stats[] = {
@@ -89,8 +293,15 @@ static const struct i40e_stats i40e_gstrings_misc_stats[] = {
 	I40E_VSI_STAT("rx_unknown_protocol", eth_stats.rx_unknown_protocol),
 	I40E_VSI_STAT("tx_linearize", tx_linearize),
 	I40E_VSI_STAT("tx_force_wb", tx_force_wb),
+	I40E_VSI_STAT("tx_busy", tx_busy),
+	I40E_VSI_STAT("tx_stopped", tx_stopped),
 	I40E_VSI_STAT("rx_alloc_fail", rx_buf_failed),
 	I40E_VSI_STAT("rx_pg_alloc_fail", rx_page_failed),
+	I40E_VSI_STAT("rx_cache_reuse", rx_page_reuse),
+	I40E_VSI_STAT("rx_cache_alloc", rx_page_alloc),
+	I40E_VSI_STAT("rx_cache_waive", rx_page_waive),
+	I40E_VSI_STAT("rx_cache_busy", rx_page_busy),
+	I40E_VSI_STAT("tx_restart", tx_restart),
 };
 
 /* These PF_STATs might look like duplicates of some NETDEV_STATs,
@@ -104,92 +315,103 @@ static const struct i40e_stats i40e_gstrings_misc_stats[] = {
  * is queried on the base PF netdev, not on the VMDq or FCoE netdev.
  */
 static const struct i40e_stats i40e_gstrings_stats[] = {
-	I40E_PF_STAT("rx_bytes", stats.eth.rx_bytes),
-	I40E_PF_STAT("tx_bytes", stats.eth.tx_bytes),
-	I40E_PF_STAT("rx_unicast", stats.eth.rx_unicast),
-	I40E_PF_STAT("tx_unicast", stats.eth.tx_unicast),
-	I40E_PF_STAT("rx_multicast", stats.eth.rx_multicast),
-	I40E_PF_STAT("tx_multicast", stats.eth.tx_multicast),
-	I40E_PF_STAT("rx_broadcast", stats.eth.rx_broadcast),
-	I40E_PF_STAT("tx_broadcast", stats.eth.tx_broadcast),
-	I40E_PF_STAT("tx_errors", stats.eth.tx_errors),
-	I40E_PF_STAT("rx_dropped", stats.eth.rx_discards),
-	I40E_PF_STAT("tx_dropped_link_down", stats.tx_dropped_link_down),
-	I40E_PF_STAT("rx_crc_errors", stats.crc_errors),
-	I40E_PF_STAT("illegal_bytes", stats.illegal_bytes),
-	I40E_PF_STAT("mac_local_faults", stats.mac_local_faults),
-	I40E_PF_STAT("mac_remote_faults", stats.mac_remote_faults),
-	I40E_PF_STAT("tx_timeout", tx_timeout_count),
-	I40E_PF_STAT("rx_csum_bad", hw_csum_rx_error),
-	I40E_PF_STAT("rx_length_errors", stats.rx_length_errors),
-	I40E_PF_STAT("link_xon_rx", stats.link_xon_rx),
-	I40E_PF_STAT("link_xoff_rx", stats.link_xoff_rx),
-	I40E_PF_STAT("link_xon_tx", stats.link_xon_tx),
-	I40E_PF_STAT("link_xoff_tx", stats.link_xoff_tx),
-	I40E_PF_STAT("rx_size_64", stats.rx_size_64),
-	I40E_PF_STAT("rx_size_127", stats.rx_size_127),
-	I40E_PF_STAT("rx_size_255", stats.rx_size_255),
-	I40E_PF_STAT("rx_size_511", stats.rx_size_511),
-	I40E_PF_STAT("rx_size_1023", stats.rx_size_1023),
-	I40E_PF_STAT("rx_size_1522", stats.rx_size_1522),
-	I40E_PF_STAT("rx_size_big", stats.rx_size_big),
-	I40E_PF_STAT("tx_size_64", stats.tx_size_64),
-	I40E_PF_STAT("tx_size_127", stats.tx_size_127),
-	I40E_PF_STAT("tx_size_255", stats.tx_size_255),
-	I40E_PF_STAT("tx_size_511", stats.tx_size_511),
-	I40E_PF_STAT("tx_size_1023", stats.tx_size_1023),
-	I40E_PF_STAT("tx_size_1522", stats.tx_size_1522),
-	I40E_PF_STAT("tx_size_big", stats.tx_size_big),
-	I40E_PF_STAT("rx_undersize", stats.rx_undersize),
-	I40E_PF_STAT("rx_fragments", stats.rx_fragments),
-	I40E_PF_STAT("rx_oversize", stats.rx_oversize),
-	I40E_PF_STAT("rx_jabber", stats.rx_jabber),
-	I40E_PF_STAT("VF_admin_queue_requests", vf_aq_requests),
-	I40E_PF_STAT("arq_overflows", arq_overflows),
-	I40E_PF_STAT("rx_hwtstamp_cleared", rx_hwtstamp_cleared),
-	I40E_PF_STAT("tx_hwtstamp_skipped", tx_hwtstamp_skipped),
-	I40E_PF_STAT("fdir_flush_cnt", fd_flush_cnt),
-	I40E_PF_STAT("fdir_atr_match", stats.fd_atr_match),
-	I40E_PF_STAT("fdir_atr_tunnel_match", stats.fd_atr_tunnel_match),
-	I40E_PF_STAT("fdir_atr_status", stats.fd_atr_status),
-	I40E_PF_STAT("fdir_sb_match", stats.fd_sb_match),
-	I40E_PF_STAT("fdir_sb_status", stats.fd_sb_status),
+	I40E_PF_STAT("port.rx_bytes", stats.eth.rx_bytes),
+	I40E_PF_STAT("port.tx_bytes", stats.eth.tx_bytes),
+	I40E_PF_STAT("port.rx_unicast", stats.eth.rx_unicast),
+	I40E_PF_STAT("port.tx_unicast", stats.eth.tx_unicast),
+	I40E_PF_STAT("port.rx_multicast", stats.eth.rx_multicast),
+	I40E_PF_STAT("port.tx_multicast", stats.eth.tx_multicast),
+	I40E_PF_STAT("port.rx_broadcast", stats.eth.rx_broadcast),
+	I40E_PF_STAT("port.tx_broadcast", stats.eth.tx_broadcast),
+	I40E_PF_STAT("port.tx_errors", stats.eth.tx_errors),
+	I40E_PF_STAT("port.rx_discards", stats.eth.rx_discards),
+	I40E_PF_STAT("port.tx_dropped_link_down", stats.tx_dropped_link_down),
+	I40E_PF_STAT("port.rx_crc_errors", stats.crc_errors),
+	I40E_PF_STAT("port.illegal_bytes", stats.illegal_bytes),
+	I40E_PF_STAT("port.mac_local_faults", stats.mac_local_faults),
+	I40E_PF_STAT("port.mac_remote_faults", stats.mac_remote_faults),
+	I40E_PF_STAT("port.tx_timeout", tx_timeout_count),
+	I40E_PF_STAT("port.rx_csum_bad", hw_csum_rx_error),
+	I40E_PF_STAT("port.rx_length_errors", stats.rx_length_errors),
+	I40E_PF_STAT("port.link_xon_rx", stats.link_xon_rx),
+	I40E_PF_STAT("port.link_xoff_rx", stats.link_xoff_rx),
+	I40E_PF_STAT("port.link_xon_tx", stats.link_xon_tx),
+	I40E_PF_STAT("port.link_xoff_tx", stats.link_xoff_tx),
+	I40E_PF_STAT("port.rx_size_64", stats.rx_size_64),
+	I40E_PF_STAT("port.rx_size_127", stats.rx_size_127),
+	I40E_PF_STAT("port.rx_size_255", stats.rx_size_255),
+	I40E_PF_STAT("port.rx_size_511", stats.rx_size_511),
+	I40E_PF_STAT("port.rx_size_1023", stats.rx_size_1023),
+	I40E_PF_STAT("port.rx_size_1522", stats.rx_size_1522),
+	I40E_PF_STAT("port.rx_size_big", stats.rx_size_big),
+	I40E_PF_STAT("port.tx_size_64", stats.tx_size_64),
+	I40E_PF_STAT("port.tx_size_127", stats.tx_size_127),
+	I40E_PF_STAT("port.tx_size_255", stats.tx_size_255),
+	I40E_PF_STAT("port.tx_size_511", stats.tx_size_511),
+	I40E_PF_STAT("port.tx_size_1023", stats.tx_size_1023),
+	I40E_PF_STAT("port.tx_size_1522", stats.tx_size_1522),
+	I40E_PF_STAT("port.tx_size_big", stats.tx_size_big),
+	I40E_PF_STAT("port.rx_undersize", stats.rx_undersize),
+	I40E_PF_STAT("port.rx_fragments", stats.rx_fragments),
+	I40E_PF_STAT("port.rx_oversize", stats.rx_oversize),
+	I40E_PF_STAT("port.rx_jabber", stats.rx_jabber),
+	I40E_PF_STAT("port.VF_admin_queue_requests", vf_aq_requests),
+	I40E_PF_STAT("port.arq_overflows", arq_overflows),
+	I40E_PF_STAT("port.tx_hwtstamp_timeouts", tx_hwtstamp_timeouts),
+	I40E_PF_STAT("port.rx_hwtstamp_cleared", rx_hwtstamp_cleared),
+	I40E_PF_STAT("port.tx_hwtstamp_skipped", tx_hwtstamp_skipped),
+	I40E_PF_STAT("port.fdir_flush_cnt", fd_flush_cnt),
+	I40E_PF_STAT("port.fdir_atr_match", stats.fd_atr_match),
+	I40E_PF_STAT("port.fdir_atr_tunnel_match", stats.fd_atr_tunnel_match),
+	I40E_PF_STAT("port.fdir_atr_status", stats.fd_atr_status),
+	I40E_PF_STAT("port.fdir_sb_match", stats.fd_sb_match),
+	I40E_PF_STAT("port.fdir_sb_status", stats.fd_sb_status),
 
 	/* LPI stats */
-	I40E_PF_STAT("tx_lpi_status", stats.tx_lpi_status),
-	I40E_PF_STAT("rx_lpi_status", stats.rx_lpi_status),
-	I40E_PF_STAT("tx_lpi_count", stats.tx_lpi_count),
-	I40E_PF_STAT("rx_lpi_count", stats.rx_lpi_count),
+	I40E_PF_STAT("port.tx_lpi_status", stats.tx_lpi_status),
+	I40E_PF_STAT("port.rx_lpi_status", stats.rx_lpi_status),
+	I40E_PF_STAT("port.tx_lpi_count", stats.tx_lpi_count),
+	I40E_PF_STAT("port.rx_lpi_count", stats.rx_lpi_count),
 };
 
-#define I40E_QUEUE_STATS_LEN(n) \
-	(((struct i40e_netdev_priv *)netdev_priv((n)))->vsi->num_queue_pairs \
-	    * 2 /* Tx and Rx together */                                     \
-	    * (sizeof(struct i40e_queue_stats) / sizeof(u64)))
-#define I40E_GLOBAL_STATS_LEN	ARRAY_SIZE(i40e_gstrings_stats)
-#define I40E_NETDEV_STATS_LEN   ARRAY_SIZE(i40e_gstrings_net_stats)
+struct i40e_pfc_stats {
+	u64 priority_xon_rx;
+	u64 priority_xoff_rx;
+	u64 priority_xon_tx;
+	u64 priority_xoff_tx;
+	u64 priority_xon_2_xoff;
+};
+
+static const struct i40e_stats i40e_gstrings_pfc_stats[] = {
+	I40E_PFC_STAT("port.tx_priority_%u_xon_tx", priority_xon_tx),
+	I40E_PFC_STAT("port.tx_priority_%u_xoff_tx", priority_xoff_tx),
+	I40E_PFC_STAT("port.rx_priority_%u_xon_rx", priority_xon_rx),
+	I40E_PFC_STAT("port.rx_priority_%u_xoff_rx", priority_xoff_rx),
+	I40E_PFC_STAT("port.rx_priority_%u_xon_2_xoff", priority_xon_2_xoff),
+};
+
+#define I40E_NETDEV_STATS_LEN	ARRAY_SIZE(i40e_gstrings_net_stats)
+
 #define I40E_MISC_STATS_LEN	ARRAY_SIZE(i40e_gstrings_misc_stats)
-#define I40E_VSI_STATS_LEN(n)   (I40E_NETDEV_STATS_LEN + \
-				 I40E_MISC_STATS_LEN + \
-				 I40E_QUEUE_STATS_LEN((n)))
-#define I40E_PFC_STATS_LEN ( \
-		(FIELD_SIZEOF(struct i40e_pf, stats.priority_xoff_rx) + \
-		 FIELD_SIZEOF(struct i40e_pf, stats.priority_xon_rx) + \
-		 FIELD_SIZEOF(struct i40e_pf, stats.priority_xoff_tx) + \
-		 FIELD_SIZEOF(struct i40e_pf, stats.priority_xon_tx) + \
-		 FIELD_SIZEOF(struct i40e_pf, stats.priority_xon_2_xoff)) \
-		 / sizeof(u64))
-#define I40E_VEB_TC_STATS_LEN ( \
-		(FIELD_SIZEOF(struct i40e_veb, tc_stats.tc_rx_packets) + \
-		 FIELD_SIZEOF(struct i40e_veb, tc_stats.tc_rx_bytes) + \
-		 FIELD_SIZEOF(struct i40e_veb, tc_stats.tc_tx_packets) + \
-		 FIELD_SIZEOF(struct i40e_veb, tc_stats.tc_tx_bytes)) \
-		 / sizeof(u64))
-#define I40E_VEB_STATS_LEN	ARRAY_SIZE(i40e_gstrings_veb_stats)
-#define I40E_VEB_STATS_TOTAL	(I40E_VEB_STATS_LEN + I40E_VEB_TC_STATS_LEN)
-#define I40E_PF_STATS_LEN(n)	(I40E_GLOBAL_STATS_LEN + \
+
+#define I40E_VSI_STATS_LEN	(I40E_NETDEV_STATS_LEN + I40E_MISC_STATS_LEN)
+
+#define I40E_PFC_STATS_LEN	(ARRAY_SIZE(i40e_gstrings_pfc_stats) * \
+				 I40E_MAX_USER_PRIORITY)
+
+#define I40E_VEB_STATS_LEN	(ARRAY_SIZE(i40e_gstrings_veb_stats) + \
+				 (ARRAY_SIZE(i40e_gstrings_veb_tc_stats) * \
+				  I40E_MAX_TRAFFIC_CLASS))
+
+#define I40E_GLOBAL_STATS_LEN	ARRAY_SIZE(i40e_gstrings_stats)
+
+#define I40E_PF_STATS_LEN	(I40E_GLOBAL_STATS_LEN + \
 				 I40E_PFC_STATS_LEN + \
-				 I40E_VSI_STATS_LEN((n)))
+				 I40E_VEB_STATS_LEN + \
+				 I40E_VSI_STATS_LEN)
+
+/* Length of stats for a single queue */
+#define I40E_QUEUE_STATS_LEN	ARRAY_SIZE(i40e_gstrings_queue_stats)
 
 enum i40e_ethtool_test_id {
 	I40E_ETH_TEST_REG = 0,
@@ -209,24 +431,37 @@ static const char i40e_gstrings_test[][ETH_GSTRING_LEN] = {
 
 struct i40e_priv_flags {
 	char flag_string[ETH_GSTRING_LEN];
-	u64 flag;
+	u8 bitno;
 	bool read_only;
 };
 
-#define I40E_PRIV_FLAG(_name, _flag, _read_only) { \
+#define I40E_PRIV_FLAG(_name, _bitno, _read_only) { \
 	.flag_string = _name, \
-	.flag = _flag, \
+	.bitno = _bitno, \
 	.read_only = _read_only, \
 }
 
 static const struct i40e_priv_flags i40e_gstrings_priv_flags[] = {
 	/* NOTE: MFP setting cannot be changed */
-	I40E_PRIV_FLAG("MFP", I40E_FLAG_MFP_ENABLED, 1),
-	I40E_PRIV_FLAG("LinkPolling", I40E_FLAG_LINK_POLLING_ENABLED, 0),
-	I40E_PRIV_FLAG("flow-director-atr", I40E_FLAG_FD_ATR_ENABLED, 0),
-	I40E_PRIV_FLAG("veb-stats", I40E_FLAG_VEB_STATS_ENABLED, 0),
-	I40E_PRIV_FLAG("hw-atr-eviction", I40E_FLAG_HW_ATR_EVICT_CAPABLE, 0),
-	I40E_PRIV_FLAG("legacy-rx", I40E_FLAG_LEGACY_RX, 0),
+	I40E_PRIV_FLAG("MFP", I40E_FLAG_MFP_ENA, 1),
+	I40E_PRIV_FLAG("total-port-shutdown",
+		       I40E_FLAG_TOTAL_PORT_SHUTDOWN_ENA, 1),
+	I40E_PRIV_FLAG("LinkPolling", I40E_FLAG_LINK_POLLING_ENA, 0),
+	I40E_PRIV_FLAG("flow-director-atr", I40E_FLAG_FD_ATR_ENA, 0),
+	I40E_PRIV_FLAG("veb-stats", I40E_FLAG_VEB_STATS_ENA, 0),
+	I40E_PRIV_FLAG("hw-atr-eviction", I40E_FLAG_HW_ATR_EVICT_ENA, 0),
+	I40E_PRIV_FLAG("link-down-on-close",
+		       I40E_FLAG_LINK_DOWN_ON_CLOSE_ENA, 0),
+	I40E_PRIV_FLAG("legacy-rx", I40E_FLAG_LEGACY_RX_ENA, 0),
+	I40E_PRIV_FLAG("disable-source-pruning",
+		       I40E_FLAG_SOURCE_PRUNING_DIS, 0),
+	I40E_PRIV_FLAG("disable-fw-lldp", I40E_FLAG_FW_LLDP_DIS, 0),
+	I40E_PRIV_FLAG("rs-fec", I40E_FLAG_RS_FEC, 0),
+	I40E_PRIV_FLAG("base-r-fec", I40E_FLAG_BASE_R_FEC, 0),
+	I40E_PRIV_FLAG("vf-vlan-pruning",
+		       I40E_FLAG_VF_VLAN_PRUNING_ENA, 0),
+	I40E_PRIV_FLAG("mdd-auto-reset-vf",
+		       I40E_FLAG_MDD_AUTO_RESET_VF, 0),
 };
 
 #define I40E_PRIV_FLAGS_STR_LEN ARRAY_SIZE(i40e_gstrings_priv_flags)
@@ -234,7 +469,7 @@ static const struct i40e_priv_flags i40e_gstrings_priv_flags[] = {
 /* Private flags with a global effect, restricted to PF 0 */
 static const struct i40e_priv_flags i40e_gl_gstrings_priv_flags[] = {
 	I40E_PRIV_FLAG("vf-true-promisc-support",
-		       I40E_FLAG_TRUE_PROMISC_SUPPORT, 0),
+		       I40E_FLAG_TRUE_PROMISC_ENA, 0),
 };
 
 #define I40E_GL_PRIV_FLAGS_STR_LEN ARRAY_SIZE(i40e_gl_gstrings_priv_flags)
@@ -251,459 +486,756 @@ static void i40e_partition_setting_complaint(struct i40e_pf *pf)
 
 /**
  * i40e_phy_type_to_ethtool - convert the phy_types to ethtool link modes
- * @phy_types: PHY types to convert
- * @supported: pointer to the ethtool supported variable to fill in
- * @advertising: pointer to the ethtool advertising variable to fill in
+ * @pf: PF struct with phy_types
+ * @ks: ethtool link ksettings struct to fill out
  *
  **/
-static void i40e_phy_type_to_ethtool(struct i40e_pf *pf, u32 *supported,
-				     u32 *advertising)
+static void i40e_phy_type_to_ethtool(struct i40e_pf *pf,
+				     struct ethtool_link_ksettings *ks)
 {
 	struct i40e_link_status *hw_link_info = &pf->hw.phy.link_info;
 	u64 phy_types = pf->hw.phy.phy_types;
 
-	*supported = 0x0;
-	*advertising = 0x0;
+	ethtool_link_ksettings_zero_link_mode(ks, supported);
+	ethtool_link_ksettings_zero_link_mode(ks, advertising);
 
 	if (phy_types & I40E_CAP_PHY_TYPE_SGMII) {
-		*supported |= SUPPORTED_Autoneg |
-			      SUPPORTED_1000baseT_Full;
-		*advertising |= ADVERTISED_Autoneg;
+		ethtool_link_ksettings_add_link_mode(ks, supported,
+						     1000baseT_Full);
 		if (hw_link_info->requested_speeds & I40E_LINK_SPEED_1GB)
-			*advertising |= ADVERTISED_1000baseT_Full;
-		if (pf->flags & I40E_FLAG_100M_SGMII_CAPABLE) {
-			*supported |= SUPPORTED_100baseT_Full;
-			*advertising |= ADVERTISED_100baseT_Full;
+			ethtool_link_ksettings_add_link_mode(ks, advertising,
+							     1000baseT_Full);
+		if (test_bit(I40E_HW_CAP_100M_SGMII, pf->hw.caps)) {
+			ethtool_link_ksettings_add_link_mode(ks, supported,
+							     100baseT_Full);
+			ethtool_link_ksettings_add_link_mode(ks, advertising,
+							     100baseT_Full);
 		}
 	}
 	if (phy_types & I40E_CAP_PHY_TYPE_XAUI ||
 	    phy_types & I40E_CAP_PHY_TYPE_XFI ||
 	    phy_types & I40E_CAP_PHY_TYPE_SFI ||
 	    phy_types & I40E_CAP_PHY_TYPE_10GBASE_SFPP_CU ||
-	    phy_types & I40E_CAP_PHY_TYPE_10GBASE_AOC)
-		*supported |= SUPPORTED_10000baseT_Full;
-	if (phy_types & I40E_CAP_PHY_TYPE_10GBASE_CR1_CU ||
-	    phy_types & I40E_CAP_PHY_TYPE_10GBASE_CR1 ||
-	    phy_types & I40E_CAP_PHY_TYPE_10GBASE_T ||
-	    phy_types & I40E_CAP_PHY_TYPE_10GBASE_SR ||
-	    phy_types & I40E_CAP_PHY_TYPE_10GBASE_LR) {
-		*supported |= SUPPORTED_Autoneg |
-			      SUPPORTED_10000baseT_Full;
-		*advertising |= ADVERTISED_Autoneg;
+	    phy_types & I40E_CAP_PHY_TYPE_10GBASE_AOC) {
+		ethtool_link_ksettings_add_link_mode(ks, supported,
+						     10000baseT_Full);
+		if (hw_link_info->requested_speeds & I40E_LINK_SPEED_10GB)
+			ethtool_link_ksettings_add_link_mode(ks, advertising,
+							     10000baseT_Full);
+	}
+	if (phy_types & I40E_CAP_PHY_TYPE_10GBASE_T) {
+		ethtool_link_ksettings_add_link_mode(ks, supported,
+						     10000baseT_Full);
 		if (hw_link_info->requested_speeds & I40E_LINK_SPEED_10GB)
-			*advertising |= ADVERTISED_10000baseT_Full;
+			ethtool_link_ksettings_add_link_mode(ks, advertising,
+							     10000baseT_Full);
+	}
+	if (phy_types & I40E_CAP_PHY_TYPE_2_5GBASE_T) {
+		ethtool_link_ksettings_add_link_mode(ks, supported,
+						     2500baseT_Full);
+		if (hw_link_info->requested_speeds & I40E_LINK_SPEED_2_5GB)
+			ethtool_link_ksettings_add_link_mode(ks, advertising,
+							     2500baseT_Full);
+	}
+	if (phy_types & I40E_CAP_PHY_TYPE_5GBASE_T) {
+		ethtool_link_ksettings_add_link_mode(ks, supported,
+						     5000baseT_Full);
+		if (hw_link_info->requested_speeds & I40E_LINK_SPEED_5GB)
+			ethtool_link_ksettings_add_link_mode(ks, advertising,
+							     5000baseT_Full);
 	}
 	if (phy_types & I40E_CAP_PHY_TYPE_XLAUI ||
 	    phy_types & I40E_CAP_PHY_TYPE_XLPPI ||
 	    phy_types & I40E_CAP_PHY_TYPE_40GBASE_AOC)
-		*supported |= SUPPORTED_40000baseCR4_Full;
+		ethtool_link_ksettings_add_link_mode(ks, supported,
+						     40000baseCR4_Full);
 	if (phy_types & I40E_CAP_PHY_TYPE_40GBASE_CR4_CU ||
 	    phy_types & I40E_CAP_PHY_TYPE_40GBASE_CR4) {
-		*supported |= SUPPORTED_Autoneg |
-			      SUPPORTED_40000baseCR4_Full;
-		*advertising |= ADVERTISED_Autoneg;
+		ethtool_link_ksettings_add_link_mode(ks, supported,
+						     40000baseCR4_Full);
 		if (hw_link_info->requested_speeds & I40E_LINK_SPEED_40GB)
-			*advertising |= ADVERTISED_40000baseCR4_Full;
+			ethtool_link_ksettings_add_link_mode(ks, advertising,
+							     40000baseCR4_Full);
 	}
 	if (phy_types & I40E_CAP_PHY_TYPE_100BASE_TX) {
-		*supported |= SUPPORTED_Autoneg |
-			      SUPPORTED_100baseT_Full;
-		*advertising |= ADVERTISED_Autoneg;
+		ethtool_link_ksettings_add_link_mode(ks, supported,
+						     100baseT_Full);
 		if (hw_link_info->requested_speeds & I40E_LINK_SPEED_100MB)
-			*advertising |= ADVERTISED_100baseT_Full;
+			ethtool_link_ksettings_add_link_mode(ks, advertising,
+							     100baseT_Full);
 	}
-	if (phy_types & I40E_CAP_PHY_TYPE_1000BASE_T ||
-	    phy_types & I40E_CAP_PHY_TYPE_1000BASE_SX ||
-	    phy_types & I40E_CAP_PHY_TYPE_1000BASE_LX ||
-	    phy_types & I40E_CAP_PHY_TYPE_1000BASE_T_OPTICAL) {
-		*supported |= SUPPORTED_Autoneg |
-			      SUPPORTED_1000baseT_Full;
-		*advertising |= ADVERTISED_Autoneg;
+	if (phy_types & I40E_CAP_PHY_TYPE_1000BASE_T) {
+		ethtool_link_ksettings_add_link_mode(ks, supported,
+						     1000baseT_Full);
 		if (hw_link_info->requested_speeds & I40E_LINK_SPEED_1GB)
-			*advertising |= ADVERTISED_1000baseT_Full;
+			ethtool_link_ksettings_add_link_mode(ks, advertising,
+							     1000baseT_Full);
+	}
+	if (phy_types & I40E_CAP_PHY_TYPE_40GBASE_SR4) {
+		ethtool_link_ksettings_add_link_mode(ks, supported,
+						     40000baseSR4_Full);
+		ethtool_link_ksettings_add_link_mode(ks, advertising,
+						     40000baseSR4_Full);
+	}
+	if (phy_types & I40E_CAP_PHY_TYPE_40GBASE_LR4) {
+		ethtool_link_ksettings_add_link_mode(ks, supported,
+						     40000baseLR4_Full);
+		ethtool_link_ksettings_add_link_mode(ks, advertising,
+						     40000baseLR4_Full);
 	}
-	if (phy_types & I40E_CAP_PHY_TYPE_40GBASE_SR4)
-		*supported |= SUPPORTED_40000baseSR4_Full;
-	if (phy_types & I40E_CAP_PHY_TYPE_40GBASE_LR4)
-		*supported |= SUPPORTED_40000baseLR4_Full;
 	if (phy_types & I40E_CAP_PHY_TYPE_40GBASE_KR4) {
-		*supported |= SUPPORTED_40000baseKR4_Full |
-			      SUPPORTED_Autoneg;
-		*advertising |= ADVERTISED_40000baseKR4_Full |
-				ADVERTISED_Autoneg;
+		ethtool_link_ksettings_add_link_mode(ks, supported,
+						     40000baseKR4_Full);
+		ethtool_link_ksettings_add_link_mode(ks, advertising,
+						     40000baseKR4_Full);
 	}
 	if (phy_types & I40E_CAP_PHY_TYPE_20GBASE_KR2) {
-		*supported |= SUPPORTED_20000baseKR2_Full |
-			      SUPPORTED_Autoneg;
-		*advertising |= ADVERTISED_Autoneg;
+		ethtool_link_ksettings_add_link_mode(ks, supported,
+						     20000baseKR2_Full);
 		if (hw_link_info->requested_speeds & I40E_LINK_SPEED_20GB)
-			*advertising |= ADVERTISED_20000baseKR2_Full;
+			ethtool_link_ksettings_add_link_mode(ks, advertising,
+							     20000baseKR2_Full);
 	}
-	if (phy_types & I40E_CAP_PHY_TYPE_10GBASE_KR) {
-		if (!(pf->flags & I40E_FLAG_HAVE_CRT_RETIMER))
-			*supported |= SUPPORTED_10000baseKR_Full |
-				      SUPPORTED_Autoneg;
-		*advertising |= ADVERTISED_Autoneg;
+	if (phy_types & I40E_CAP_PHY_TYPE_10GBASE_KX4) {
+		ethtool_link_ksettings_add_link_mode(ks, supported,
+						     10000baseKX4_Full);
 		if (hw_link_info->requested_speeds & I40E_LINK_SPEED_10GB)
-			if (!(pf->flags & I40E_FLAG_HAVE_CRT_RETIMER))
-				*advertising |= ADVERTISED_10000baseKR_Full;
+			ethtool_link_ksettings_add_link_mode(ks, advertising,
+							     10000baseKX4_Full);
 	}
-	if (phy_types & I40E_CAP_PHY_TYPE_10GBASE_KX4) {
-		*supported |= SUPPORTED_10000baseKX4_Full |
-			      SUPPORTED_Autoneg;
-		*advertising |= ADVERTISED_Autoneg;
+	if (phy_types & I40E_CAP_PHY_TYPE_10GBASE_KR &&
+	    !test_bit(I40E_HW_CAP_CRT_RETIMER, pf->hw.caps)) {
+		ethtool_link_ksettings_add_link_mode(ks, supported,
+						     10000baseKR_Full);
 		if (hw_link_info->requested_speeds & I40E_LINK_SPEED_10GB)
-			*advertising |= ADVERTISED_10000baseKX4_Full;
+			ethtool_link_ksettings_add_link_mode(ks, advertising,
+							     10000baseKR_Full);
 	}
-	if (phy_types & I40E_CAP_PHY_TYPE_1000BASE_KX) {
-		if (!(pf->flags & I40E_FLAG_HAVE_CRT_RETIMER))
-			*supported |= SUPPORTED_1000baseKX_Full |
-				      SUPPORTED_Autoneg;
-		*advertising |= ADVERTISED_Autoneg;
+	if (phy_types & I40E_CAP_PHY_TYPE_1000BASE_KX &&
+	    !test_bit(I40E_HW_CAP_CRT_RETIMER, pf->hw.caps)) {
+		ethtool_link_ksettings_add_link_mode(ks, supported,
+						     1000baseKX_Full);
 		if (hw_link_info->requested_speeds & I40E_LINK_SPEED_1GB)
-			if (!(pf->flags & I40E_FLAG_HAVE_CRT_RETIMER))
-				*advertising |= ADVERTISED_1000baseKX_Full;
+			ethtool_link_ksettings_add_link_mode(ks, advertising,
+							     1000baseKX_Full);
+	}
+	/* need to add 25G PHY types */
+	if (phy_types & I40E_CAP_PHY_TYPE_25GBASE_KR) {
+		ethtool_link_ksettings_add_link_mode(ks, supported,
+						     25000baseKR_Full);
+		if (hw_link_info->requested_speeds & I40E_LINK_SPEED_25GB)
+			ethtool_link_ksettings_add_link_mode(ks, advertising,
+							     25000baseKR_Full);
+	}
+	if (phy_types & I40E_CAP_PHY_TYPE_25GBASE_CR) {
+		ethtool_link_ksettings_add_link_mode(ks, supported,
+						     25000baseCR_Full);
+		if (hw_link_info->requested_speeds & I40E_LINK_SPEED_25GB)
+			ethtool_link_ksettings_add_link_mode(ks, advertising,
+							     25000baseCR_Full);
+	}
+	if (phy_types & I40E_CAP_PHY_TYPE_25GBASE_SR ||
+	    phy_types & I40E_CAP_PHY_TYPE_25GBASE_LR) {
+		ethtool_link_ksettings_add_link_mode(ks, supported,
+						     25000baseSR_Full);
+		if (hw_link_info->requested_speeds & I40E_LINK_SPEED_25GB)
+			ethtool_link_ksettings_add_link_mode(ks, advertising,
+							     25000baseSR_Full);
+	}
+	if (phy_types & I40E_CAP_PHY_TYPE_25GBASE_AOC ||
+	    phy_types & I40E_CAP_PHY_TYPE_25GBASE_ACC) {
+		ethtool_link_ksettings_add_link_mode(ks, supported,
+						     25000baseCR_Full);
+		if (hw_link_info->requested_speeds & I40E_LINK_SPEED_25GB)
+			ethtool_link_ksettings_add_link_mode(ks, advertising,
+							     25000baseCR_Full);
 	}
 	if (phy_types & I40E_CAP_PHY_TYPE_25GBASE_KR ||
 	    phy_types & I40E_CAP_PHY_TYPE_25GBASE_CR ||
 	    phy_types & I40E_CAP_PHY_TYPE_25GBASE_SR ||
-	    phy_types & I40E_CAP_PHY_TYPE_25GBASE_LR) {
-		*supported |= SUPPORTED_Autoneg;
-		*advertising |= ADVERTISED_Autoneg;
+	    phy_types & I40E_CAP_PHY_TYPE_25GBASE_LR ||
+	    phy_types & I40E_CAP_PHY_TYPE_25GBASE_AOC ||
+	    phy_types & I40E_CAP_PHY_TYPE_25GBASE_ACC) {
+		ethtool_link_ksettings_add_link_mode(ks, supported, FEC_NONE);
+		ethtool_link_ksettings_add_link_mode(ks, supported, FEC_RS);
+		ethtool_link_ksettings_add_link_mode(ks, supported, FEC_BASER);
+		if (hw_link_info->requested_speeds & I40E_LINK_SPEED_25GB) {
+			ethtool_link_ksettings_add_link_mode(ks, advertising,
+							     FEC_NONE);
+			ethtool_link_ksettings_add_link_mode(ks, advertising,
+							     FEC_RS);
+			ethtool_link_ksettings_add_link_mode(ks, advertising,
+							     FEC_BASER);
+		}
+	}
+	/* need to add new 10G PHY types */
+	if (phy_types & I40E_CAP_PHY_TYPE_10GBASE_CR1 ||
+	    phy_types & I40E_CAP_PHY_TYPE_10GBASE_CR1_CU) {
+		ethtool_link_ksettings_add_link_mode(ks, supported,
+						     10000baseCR_Full);
+		if (hw_link_info->requested_speeds & I40E_LINK_SPEED_10GB)
+			ethtool_link_ksettings_add_link_mode(ks, advertising,
+							     10000baseCR_Full);
+	}
+	if (phy_types & I40E_CAP_PHY_TYPE_10GBASE_SR) {
+		ethtool_link_ksettings_add_link_mode(ks, supported,
+						     10000baseSR_Full);
+		if (hw_link_info->requested_speeds & I40E_LINK_SPEED_10GB)
+			ethtool_link_ksettings_add_link_mode(ks, advertising,
+							     10000baseSR_Full);
+	}
+	if (phy_types & I40E_CAP_PHY_TYPE_10GBASE_LR) {
+		ethtool_link_ksettings_add_link_mode(ks, supported,
+						     10000baseLR_Full);
+		if (hw_link_info->requested_speeds & I40E_LINK_SPEED_10GB)
+			ethtool_link_ksettings_add_link_mode(ks, advertising,
+							     10000baseLR_Full);
+	}
+	if (phy_types & I40E_CAP_PHY_TYPE_1000BASE_SX ||
+	    phy_types & I40E_CAP_PHY_TYPE_1000BASE_LX ||
+	    phy_types & I40E_CAP_PHY_TYPE_1000BASE_T_OPTICAL) {
+		ethtool_link_ksettings_add_link_mode(ks, supported,
+						     1000baseX_Full);
+		if (hw_link_info->requested_speeds & I40E_LINK_SPEED_1GB)
+			ethtool_link_ksettings_add_link_mode(ks, advertising,
+							     1000baseX_Full);
+	}
+	/* Autoneg PHY types */
+	if (phy_types & I40E_CAP_PHY_TYPE_SGMII ||
+	    phy_types & I40E_CAP_PHY_TYPE_40GBASE_KR4 ||
+	    phy_types & I40E_CAP_PHY_TYPE_40GBASE_CR4_CU ||
+	    phy_types & I40E_CAP_PHY_TYPE_40GBASE_CR4 ||
+	    phy_types & I40E_CAP_PHY_TYPE_25GBASE_SR ||
+	    phy_types & I40E_CAP_PHY_TYPE_25GBASE_LR ||
+	    phy_types & I40E_CAP_PHY_TYPE_25GBASE_KR ||
+	    phy_types & I40E_CAP_PHY_TYPE_25GBASE_CR ||
+	    phy_types & I40E_CAP_PHY_TYPE_20GBASE_KR2 ||
+	    phy_types & I40E_CAP_PHY_TYPE_10GBASE_SR ||
+	    phy_types & I40E_CAP_PHY_TYPE_10GBASE_LR ||
+	    phy_types & I40E_CAP_PHY_TYPE_10GBASE_KX4 ||
+	    phy_types & I40E_CAP_PHY_TYPE_10GBASE_KR ||
+	    phy_types & I40E_CAP_PHY_TYPE_10GBASE_CR1_CU ||
+	    phy_types & I40E_CAP_PHY_TYPE_10GBASE_CR1 ||
+	    phy_types & I40E_CAP_PHY_TYPE_10GBASE_T ||
+	    phy_types & I40E_CAP_PHY_TYPE_5GBASE_T ||
+	    phy_types & I40E_CAP_PHY_TYPE_2_5GBASE_T ||
+	    phy_types & I40E_CAP_PHY_TYPE_1000BASE_T_OPTICAL ||
+	    phy_types & I40E_CAP_PHY_TYPE_1000BASE_T ||
+	    phy_types & I40E_CAP_PHY_TYPE_1000BASE_SX ||
+	    phy_types & I40E_CAP_PHY_TYPE_1000BASE_LX ||
+	    phy_types & I40E_CAP_PHY_TYPE_1000BASE_KX ||
+	    phy_types & I40E_CAP_PHY_TYPE_100BASE_TX) {
+		ethtool_link_ksettings_add_link_mode(ks, supported,
+						     Autoneg);
+		ethtool_link_ksettings_add_link_mode(ks, advertising,
+						     Autoneg);
+	}
+}
+
+/**
+ * i40e_get_settings_link_up_fec - Get the FEC mode encoding from mask
+ * @req_fec_info: mask request FEC info
+ * @ks: ethtool ksettings to fill in
+ **/
+static void i40e_get_settings_link_up_fec(u8 req_fec_info,
+					  struct ethtool_link_ksettings *ks)
+{
+	ethtool_link_ksettings_add_link_mode(ks, supported, FEC_NONE);
+	ethtool_link_ksettings_add_link_mode(ks, supported, FEC_RS);
+	ethtool_link_ksettings_add_link_mode(ks, supported, FEC_BASER);
+
+	if ((I40E_AQ_SET_FEC_REQUEST_RS & req_fec_info) &&
+	    (I40E_AQ_SET_FEC_REQUEST_KR & req_fec_info)) {
+		ethtool_link_ksettings_add_link_mode(ks, advertising,
+						     FEC_NONE);
+		ethtool_link_ksettings_add_link_mode(ks, advertising,
+						     FEC_BASER);
+		ethtool_link_ksettings_add_link_mode(ks, advertising, FEC_RS);
+	} else if (I40E_AQ_SET_FEC_REQUEST_RS & req_fec_info) {
+		ethtool_link_ksettings_add_link_mode(ks, advertising, FEC_RS);
+	} else if (I40E_AQ_SET_FEC_REQUEST_KR & req_fec_info) {
+		ethtool_link_ksettings_add_link_mode(ks, advertising,
+						     FEC_BASER);
+	} else {
+		ethtool_link_ksettings_add_link_mode(ks, advertising,
+						     FEC_NONE);
 	}
 }
 
 /**
  * i40e_get_settings_link_up - Get the Link settings for when link is up
  * @hw: hw structure
- * @ecmd: ethtool command to fill in
+ * @ks: ethtool ksettings to fill in
  * @netdev: network interface device structure
- *
+ * @pf: pointer to physical function struct
  **/
 static void i40e_get_settings_link_up(struct i40e_hw *hw,
-				      struct ethtool_link_ksettings *cmd,
+				      struct ethtool_link_ksettings *ks,
 				      struct net_device *netdev,
 				      struct i40e_pf *pf)
 {
 	struct i40e_link_status *hw_link_info = &hw->phy.link_info;
+	struct ethtool_link_ksettings cap_ksettings;
 	u32 link_speed = hw_link_info->link_speed;
-	u32 e_advertising = 0x0;
-	u32 e_supported = 0x0;
-	u32 supported, advertising;
-
-	ethtool_convert_link_mode_to_legacy_u32(&supported,
-						cmd->link_modes.supported);
-	ethtool_convert_link_mode_to_legacy_u32(&advertising,
-						cmd->link_modes.advertising);
 
 	/* Initialize supported and advertised settings based on phy settings */
 	switch (hw_link_info->phy_type) {
 	case I40E_PHY_TYPE_40GBASE_CR4:
 	case I40E_PHY_TYPE_40GBASE_CR4_CU:
-		supported = SUPPORTED_Autoneg |
-			    SUPPORTED_40000baseCR4_Full;
-		advertising = ADVERTISED_Autoneg |
-			      ADVERTISED_40000baseCR4_Full;
+		ethtool_link_ksettings_add_link_mode(ks, supported, Autoneg);
+		ethtool_link_ksettings_add_link_mode(ks, supported,
+						     40000baseCR4_Full);
+		ethtool_link_ksettings_add_link_mode(ks, advertising, Autoneg);
+		ethtool_link_ksettings_add_link_mode(ks, advertising,
+						     40000baseCR4_Full);
 		break;
 	case I40E_PHY_TYPE_XLAUI:
 	case I40E_PHY_TYPE_XLPPI:
 	case I40E_PHY_TYPE_40GBASE_AOC:
-		supported = SUPPORTED_40000baseCR4_Full;
+		ethtool_link_ksettings_add_link_mode(ks, supported,
+						     40000baseCR4_Full);
+		ethtool_link_ksettings_add_link_mode(ks, advertising,
+						     40000baseCR4_Full);
 		break;
 	case I40E_PHY_TYPE_40GBASE_SR4:
-		supported = SUPPORTED_40000baseSR4_Full;
+		ethtool_link_ksettings_add_link_mode(ks, supported,
+						     40000baseSR4_Full);
+		ethtool_link_ksettings_add_link_mode(ks, advertising,
+						     40000baseSR4_Full);
 		break;
 	case I40E_PHY_TYPE_40GBASE_LR4:
-		supported = SUPPORTED_40000baseLR4_Full;
+		ethtool_link_ksettings_add_link_mode(ks, supported,
+						     40000baseLR4_Full);
+		ethtool_link_ksettings_add_link_mode(ks, advertising,
+						     40000baseLR4_Full);
 		break;
+	case I40E_PHY_TYPE_25GBASE_SR:
+	case I40E_PHY_TYPE_25GBASE_LR:
 	case I40E_PHY_TYPE_10GBASE_SR:
 	case I40E_PHY_TYPE_10GBASE_LR:
 	case I40E_PHY_TYPE_1000BASE_SX:
 	case I40E_PHY_TYPE_1000BASE_LX:
-		supported = SUPPORTED_10000baseT_Full;
+		ethtool_link_ksettings_add_link_mode(ks, supported, Autoneg);
+		ethtool_link_ksettings_add_link_mode(ks, advertising, Autoneg);
+		ethtool_link_ksettings_add_link_mode(ks, supported,
+						     25000baseSR_Full);
+		ethtool_link_ksettings_add_link_mode(ks, advertising,
+						     25000baseSR_Full);
+		i40e_get_settings_link_up_fec(hw_link_info->req_fec_info, ks);
+		ethtool_link_ksettings_add_link_mode(ks, supported,
+						     10000baseSR_Full);
+		ethtool_link_ksettings_add_link_mode(ks, advertising,
+						     10000baseSR_Full);
+		ethtool_link_ksettings_add_link_mode(ks, supported,
+						     10000baseLR_Full);
+		ethtool_link_ksettings_add_link_mode(ks, advertising,
+						     10000baseLR_Full);
+		ethtool_link_ksettings_add_link_mode(ks, supported,
+						     1000baseX_Full);
+		ethtool_link_ksettings_add_link_mode(ks, advertising,
+						     1000baseX_Full);
+		ethtool_link_ksettings_add_link_mode(ks, supported,
+						     10000baseT_Full);
 		if (hw_link_info->module_type[2] &
 		    I40E_MODULE_TYPE_1000BASE_SX ||
 		    hw_link_info->module_type[2] &
 		    I40E_MODULE_TYPE_1000BASE_LX) {
-			supported |= SUPPORTED_1000baseT_Full;
+			ethtool_link_ksettings_add_link_mode(ks, supported,
+							     1000baseT_Full);
 			if (hw_link_info->requested_speeds &
 			    I40E_LINK_SPEED_1GB)
-				advertising |= ADVERTISED_1000baseT_Full;
+				ethtool_link_ksettings_add_link_mode(
+				     ks, advertising, 1000baseT_Full);
 		}
 		if (hw_link_info->requested_speeds & I40E_LINK_SPEED_10GB)
-			advertising |= ADVERTISED_10000baseT_Full;
+			ethtool_link_ksettings_add_link_mode(ks, advertising,
+							     10000baseT_Full);
 		break;
 	case I40E_PHY_TYPE_10GBASE_T:
+	case I40E_PHY_TYPE_5GBASE_T_LINK_STATUS:
+	case I40E_PHY_TYPE_2_5GBASE_T_LINK_STATUS:
 	case I40E_PHY_TYPE_1000BASE_T:
 	case I40E_PHY_TYPE_100BASE_TX:
-		supported = SUPPORTED_Autoneg |
-			    SUPPORTED_10000baseT_Full |
-			    SUPPORTED_1000baseT_Full |
-			    SUPPORTED_100baseT_Full;
-		advertising = ADVERTISED_Autoneg;
+		ethtool_link_ksettings_add_link_mode(ks, supported, Autoneg);
+		ethtool_link_ksettings_add_link_mode(ks, supported,
+						     10000baseT_Full);
+		ethtool_link_ksettings_add_link_mode(ks, supported,
+						     5000baseT_Full);
+		ethtool_link_ksettings_add_link_mode(ks, supported,
+						     2500baseT_Full);
+		ethtool_link_ksettings_add_link_mode(ks, supported,
+						     1000baseT_Full);
+		ethtool_link_ksettings_add_link_mode(ks, supported,
+						     100baseT_Full);
+		ethtool_link_ksettings_add_link_mode(ks, advertising, Autoneg);
 		if (hw_link_info->requested_speeds & I40E_LINK_SPEED_10GB)
-			advertising |= ADVERTISED_10000baseT_Full;
+			ethtool_link_ksettings_add_link_mode(ks, advertising,
+							     10000baseT_Full);
+		if (hw_link_info->requested_speeds & I40E_LINK_SPEED_5GB)
+			ethtool_link_ksettings_add_link_mode(ks, advertising,
+							     5000baseT_Full);
+		if (hw_link_info->requested_speeds & I40E_LINK_SPEED_2_5GB)
+			ethtool_link_ksettings_add_link_mode(ks, advertising,
+							     2500baseT_Full);
 		if (hw_link_info->requested_speeds & I40E_LINK_SPEED_1GB)
-			advertising |= ADVERTISED_1000baseT_Full;
+			ethtool_link_ksettings_add_link_mode(ks, advertising,
+							     1000baseT_Full);
 		if (hw_link_info->requested_speeds & I40E_LINK_SPEED_100MB)
-			advertising |= ADVERTISED_100baseT_Full;
+			ethtool_link_ksettings_add_link_mode(ks, advertising,
+							     100baseT_Full);
 		break;
 	case I40E_PHY_TYPE_1000BASE_T_OPTICAL:
-		supported = SUPPORTED_Autoneg |
-			    SUPPORTED_1000baseT_Full;
-		advertising = ADVERTISED_Autoneg |
-			      ADVERTISED_1000baseT_Full;
+		ethtool_link_ksettings_add_link_mode(ks, supported, Autoneg);
+		ethtool_link_ksettings_add_link_mode(ks, supported,
+						     1000baseT_Full);
+		ethtool_link_ksettings_add_link_mode(ks, advertising, Autoneg);
+		ethtool_link_ksettings_add_link_mode(ks, advertising,
+						     1000baseT_Full);
 		break;
 	case I40E_PHY_TYPE_10GBASE_CR1_CU:
 	case I40E_PHY_TYPE_10GBASE_CR1:
-		supported = SUPPORTED_Autoneg |
-			    SUPPORTED_10000baseT_Full;
-		advertising = ADVERTISED_Autoneg |
-			      ADVERTISED_10000baseT_Full;
+		ethtool_link_ksettings_add_link_mode(ks, supported, Autoneg);
+		ethtool_link_ksettings_add_link_mode(ks, supported,
+						     10000baseT_Full);
+		ethtool_link_ksettings_add_link_mode(ks, advertising, Autoneg);
+		ethtool_link_ksettings_add_link_mode(ks, advertising,
+						     10000baseT_Full);
 		break;
 	case I40E_PHY_TYPE_XAUI:
 	case I40E_PHY_TYPE_XFI:
 	case I40E_PHY_TYPE_SFI:
 	case I40E_PHY_TYPE_10GBASE_SFPP_CU:
 	case I40E_PHY_TYPE_10GBASE_AOC:
-		supported = SUPPORTED_10000baseT_Full;
-		advertising = SUPPORTED_10000baseT_Full;
+		ethtool_link_ksettings_add_link_mode(ks, supported,
+						     10000baseT_Full);
+		if (hw_link_info->requested_speeds & I40E_LINK_SPEED_10GB)
+			ethtool_link_ksettings_add_link_mode(ks, advertising,
+							     10000baseT_Full);
+		i40e_get_settings_link_up_fec(hw_link_info->req_fec_info, ks);
 		break;
 	case I40E_PHY_TYPE_SGMII:
-		supported = SUPPORTED_Autoneg |
-			    SUPPORTED_1000baseT_Full;
+		ethtool_link_ksettings_add_link_mode(ks, supported, Autoneg);
+		ethtool_link_ksettings_add_link_mode(ks, supported,
+						     1000baseT_Full);
 		if (hw_link_info->requested_speeds & I40E_LINK_SPEED_1GB)
-			advertising |= ADVERTISED_1000baseT_Full;
-		if (pf->flags & I40E_FLAG_100M_SGMII_CAPABLE) {
-			supported |= SUPPORTED_100baseT_Full;
+			ethtool_link_ksettings_add_link_mode(ks, advertising,
+							     1000baseT_Full);
+		if (test_bit(I40E_HW_CAP_100M_SGMII, pf->hw.caps)) {
+			ethtool_link_ksettings_add_link_mode(ks, supported,
+							     100baseT_Full);
 			if (hw_link_info->requested_speeds &
 			    I40E_LINK_SPEED_100MB)
-				advertising |= ADVERTISED_100baseT_Full;
+				ethtool_link_ksettings_add_link_mode(
+				      ks, advertising, 100baseT_Full);
 		}
 		break;
 	case I40E_PHY_TYPE_40GBASE_KR4:
+	case I40E_PHY_TYPE_25GBASE_KR:
 	case I40E_PHY_TYPE_20GBASE_KR2:
 	case I40E_PHY_TYPE_10GBASE_KR:
 	case I40E_PHY_TYPE_10GBASE_KX4:
 	case I40E_PHY_TYPE_1000BASE_KX:
-		supported |= SUPPORTED_40000baseKR4_Full |
-			     SUPPORTED_20000baseKR2_Full |
-			     SUPPORTED_10000baseKR_Full |
-			     SUPPORTED_10000baseKX4_Full |
-			     SUPPORTED_1000baseKX_Full |
-			     SUPPORTED_Autoneg;
-		advertising |= ADVERTISED_40000baseKR4_Full |
-			       ADVERTISED_20000baseKR2_Full |
-			       ADVERTISED_10000baseKR_Full |
-			       ADVERTISED_10000baseKX4_Full |
-			       ADVERTISED_1000baseKX_Full |
-			       ADVERTISED_Autoneg;
+		ethtool_link_ksettings_add_link_mode(ks, supported,
+						     40000baseKR4_Full);
+		ethtool_link_ksettings_add_link_mode(ks, supported,
+						     25000baseKR_Full);
+		ethtool_link_ksettings_add_link_mode(ks, supported,
+						     20000baseKR2_Full);
+		ethtool_link_ksettings_add_link_mode(ks, supported,
+						     10000baseKR_Full);
+		ethtool_link_ksettings_add_link_mode(ks, supported,
+						     10000baseKX4_Full);
+		ethtool_link_ksettings_add_link_mode(ks, supported,
+						     1000baseKX_Full);
+		ethtool_link_ksettings_add_link_mode(ks, supported, Autoneg);
+		ethtool_link_ksettings_add_link_mode(ks, advertising,
+						     40000baseKR4_Full);
+		ethtool_link_ksettings_add_link_mode(ks, advertising,
+						     25000baseKR_Full);
+		i40e_get_settings_link_up_fec(hw_link_info->req_fec_info, ks);
+		ethtool_link_ksettings_add_link_mode(ks, advertising,
+						     20000baseKR2_Full);
+		ethtool_link_ksettings_add_link_mode(ks, advertising,
+						     10000baseKR_Full);
+		ethtool_link_ksettings_add_link_mode(ks, advertising,
+						     10000baseKX4_Full);
+		ethtool_link_ksettings_add_link_mode(ks, advertising,
+						     1000baseKX_Full);
+		ethtool_link_ksettings_add_link_mode(ks, advertising, Autoneg);
 		break;
-	case I40E_PHY_TYPE_25GBASE_KR:
 	case I40E_PHY_TYPE_25GBASE_CR:
-	case I40E_PHY_TYPE_25GBASE_SR:
-	case I40E_PHY_TYPE_25GBASE_LR:
-		supported = SUPPORTED_Autoneg;
-		advertising = ADVERTISED_Autoneg;
-		/* TODO: add speeds when ethtool is ready to support*/
+		ethtool_link_ksettings_add_link_mode(ks, supported, Autoneg);
+		ethtool_link_ksettings_add_link_mode(ks, advertising, Autoneg);
+		ethtool_link_ksettings_add_link_mode(ks, supported,
+						     25000baseCR_Full);
+		ethtool_link_ksettings_add_link_mode(ks, advertising,
+						     25000baseCR_Full);
+		i40e_get_settings_link_up_fec(hw_link_info->req_fec_info, ks);
+
+		break;
+	case I40E_PHY_TYPE_25GBASE_AOC:
+	case I40E_PHY_TYPE_25GBASE_ACC:
+		ethtool_link_ksettings_add_link_mode(ks, supported, Autoneg);
+		ethtool_link_ksettings_add_link_mode(ks, advertising, Autoneg);
+		ethtool_link_ksettings_add_link_mode(ks, supported,
+						     25000baseCR_Full);
+		ethtool_link_ksettings_add_link_mode(ks, advertising,
+						     25000baseCR_Full);
+		i40e_get_settings_link_up_fec(hw_link_info->req_fec_info, ks);
+
+		ethtool_link_ksettings_add_link_mode(ks, supported,
+						     10000baseCR_Full);
+		ethtool_link_ksettings_add_link_mode(ks, advertising,
+						     10000baseCR_Full);
 		break;
 	default:
 		/* if we got here and link is up something bad is afoot */
-		netdev_info(netdev, "WARNING: Link is up but PHY type 0x%x is not recognized.\n",
+		netdev_info(netdev,
+			    "WARNING: Link is up but PHY type 0x%x is not recognized, or incorrect cable is in use\n",
 			    hw_link_info->phy_type);
 	}
 
 	/* Now that we've worked out everything that could be supported by the
-	 * current PHY type, get what is supported by the NVM and them to
-	 * get what is truly supported
+	 * current PHY type, get what is supported by the NVM and intersect
+	 * them to get what is truly supported
 	 */
-	i40e_phy_type_to_ethtool(pf, &e_supported,
-				 &e_advertising);
-
-	supported = supported & e_supported;
-	advertising = advertising & e_advertising;
+	memset(&cap_ksettings, 0, sizeof(struct ethtool_link_ksettings));
+	i40e_phy_type_to_ethtool(pf, &cap_ksettings);
+	ethtool_intersect_link_masks(ks, &cap_ksettings);
 
 	/* Set speed and duplex */
 	switch (link_speed) {
 	case I40E_LINK_SPEED_40GB:
-		cmd->base.speed = SPEED_40000;
+		ks->base.speed = SPEED_40000;
 		break;
 	case I40E_LINK_SPEED_25GB:
-#ifdef SPEED_25000
-		cmd->base.speed = SPEED_25000;
-#else
-		netdev_info(netdev,
-			    "Speed is 25G, display not supported by this version of ethtool.\n");
-#endif
+		ks->base.speed = SPEED_25000;
 		break;
 	case I40E_LINK_SPEED_20GB:
-		cmd->base.speed = SPEED_20000;
+		ks->base.speed = SPEED_20000;
 		break;
 	case I40E_LINK_SPEED_10GB:
-		cmd->base.speed = SPEED_10000;
+		ks->base.speed = SPEED_10000;
+		break;
+	case I40E_LINK_SPEED_5GB:
+		ks->base.speed = SPEED_5000;
+		break;
+	case I40E_LINK_SPEED_2_5GB:
+		ks->base.speed = SPEED_2500;
 		break;
 	case I40E_LINK_SPEED_1GB:
-		cmd->base.speed = SPEED_1000;
+		ks->base.speed = SPEED_1000;
 		break;
 	case I40E_LINK_SPEED_100MB:
-		cmd->base.speed = SPEED_100;
+		ks->base.speed = SPEED_100;
 		break;
 	default:
+		ks->base.speed = SPEED_UNKNOWN;
 		break;
 	}
-	cmd->base.duplex = DUPLEX_FULL;
-
-	ethtool_convert_legacy_u32_to_link_mode(cmd->link_modes.supported,
-						supported);
-	ethtool_convert_legacy_u32_to_link_mode(cmd->link_modes.advertising,
-						advertising);
+	ks->base.duplex = DUPLEX_FULL;
 }
 
 /**
  * i40e_get_settings_link_down - Get the Link settings for when link is down
  * @hw: hw structure
- * @ecmd: ethtool command to fill in
+ * @ks: ethtool ksettings to fill in
+ * @pf: pointer to physical function struct
  *
  * Reports link settings that can be determined when link is down
  **/
 static void i40e_get_settings_link_down(struct i40e_hw *hw,
-					struct ethtool_link_ksettings *cmd,
+					struct ethtool_link_ksettings *ks,
 					struct i40e_pf *pf)
 {
-	u32 supported, advertising;
-
 	/* link is down and the driver needs to fall back on
 	 * supported phy types to figure out what info to display
 	 */
-	i40e_phy_type_to_ethtool(pf, &supported, &advertising);
-
-	ethtool_convert_legacy_u32_to_link_mode(cmd->link_modes.supported,
-						supported);
-	ethtool_convert_legacy_u32_to_link_mode(cmd->link_modes.advertising,
-						advertising);
+	i40e_phy_type_to_ethtool(pf, ks);
 
 	/* With no link speed and duplex are unknown */
-	cmd->base.speed = SPEED_UNKNOWN;
-	cmd->base.duplex = DUPLEX_UNKNOWN;
+	ks->base.speed = SPEED_UNKNOWN;
+	ks->base.duplex = DUPLEX_UNKNOWN;
 }
 
 /**
- * i40e_get_settings - Get Link Speed and Duplex settings
+ * i40e_get_link_ksettings - Get Link Speed and Duplex settings
  * @netdev: network interface device structure
- * @ecmd: ethtool command
+ * @ks: ethtool ksettings
  *
  * Reports speed/duplex settings based on media_type
  **/
 static int i40e_get_link_ksettings(struct net_device *netdev,
-				   struct ethtool_link_ksettings *cmd)
+				   struct ethtool_link_ksettings *ks)
 {
 	struct i40e_netdev_priv *np = netdev_priv(netdev);
 	struct i40e_pf *pf = np->vsi->back;
 	struct i40e_hw *hw = &pf->hw;
 	struct i40e_link_status *hw_link_info = &hw->phy.link_info;
 	bool link_up = hw_link_info->link_info & I40E_AQ_LINK_UP;
-	u32 advertising;
+
+	ethtool_link_ksettings_zero_link_mode(ks, supported);
+	ethtool_link_ksettings_zero_link_mode(ks, advertising);
 
 	if (link_up)
-		i40e_get_settings_link_up(hw, cmd, netdev, pf);
+		i40e_get_settings_link_up(hw, ks, netdev, pf);
 	else
-		i40e_get_settings_link_down(hw, cmd, pf);
+		i40e_get_settings_link_down(hw, ks, pf);
 
 	/* Now set the settings that don't rely on link being up/down */
 	/* Set autoneg settings */
-	cmd->base.autoneg = ((hw_link_info->an_info & I40E_AQ_AN_COMPLETED) ?
-			  AUTONEG_ENABLE : AUTONEG_DISABLE);
+	ks->base.autoneg = ((hw_link_info->an_info & I40E_AQ_AN_COMPLETED) ?
+			    AUTONEG_ENABLE : AUTONEG_DISABLE);
 
+	/* Set media type settings */
 	switch (hw->phy.media_type) {
 	case I40E_MEDIA_TYPE_BACKPLANE:
-		ethtool_link_ksettings_add_link_mode(cmd, supported,
-						     Autoneg);
-		ethtool_link_ksettings_add_link_mode(cmd, supported,
-						     Backplane);
-		ethtool_link_ksettings_add_link_mode(cmd, advertising,
-						     Autoneg);
-		ethtool_link_ksettings_add_link_mode(cmd, advertising,
+		ethtool_link_ksettings_add_link_mode(ks, supported, Autoneg);
+		ethtool_link_ksettings_add_link_mode(ks, supported, Backplane);
+		ethtool_link_ksettings_add_link_mode(ks, advertising, Autoneg);
+		ethtool_link_ksettings_add_link_mode(ks, advertising,
 						     Backplane);
-		cmd->base.port = PORT_NONE;
+		ks->base.port = PORT_NONE;
 		break;
 	case I40E_MEDIA_TYPE_BASET:
-		ethtool_link_ksettings_add_link_mode(cmd, supported, TP);
-		ethtool_link_ksettings_add_link_mode(cmd, advertising, TP);
-		cmd->base.port = PORT_TP;
+		ethtool_link_ksettings_add_link_mode(ks, supported, TP);
+		ethtool_link_ksettings_add_link_mode(ks, advertising, TP);
+		ks->base.port = PORT_TP;
 		break;
 	case I40E_MEDIA_TYPE_DA:
 	case I40E_MEDIA_TYPE_CX4:
-		ethtool_link_ksettings_add_link_mode(cmd, supported, FIBRE);
-		ethtool_link_ksettings_add_link_mode(cmd, advertising, FIBRE);
-		cmd->base.port = PORT_DA;
+		ethtool_link_ksettings_add_link_mode(ks, supported, FIBRE);
+		ethtool_link_ksettings_add_link_mode(ks, advertising, FIBRE);
+		ks->base.port = PORT_DA;
 		break;
 	case I40E_MEDIA_TYPE_FIBER:
-		ethtool_link_ksettings_add_link_mode(cmd, supported, FIBRE);
-		cmd->base.port = PORT_FIBRE;
+		ethtool_link_ksettings_add_link_mode(ks, supported, FIBRE);
+		ethtool_link_ksettings_add_link_mode(ks, advertising, FIBRE);
+		ks->base.port = PORT_FIBRE;
 		break;
 	case I40E_MEDIA_TYPE_UNKNOWN:
 	default:
-		cmd->base.port = PORT_OTHER;
+		ks->base.port = PORT_OTHER;
 		break;
 	}
 
 	/* Set flow control settings */
-	ethtool_link_ksettings_add_link_mode(cmd, supported, Pause);
+	ethtool_link_ksettings_add_link_mode(ks, supported, Pause);
+	ethtool_link_ksettings_add_link_mode(ks, supported, Asym_Pause);
 
 	switch (hw->fc.requested_mode) {
 	case I40E_FC_FULL:
-		ethtool_link_ksettings_add_link_mode(cmd, advertising,
-						     Pause);
+		ethtool_link_ksettings_add_link_mode(ks, advertising, Pause);
 		break;
 	case I40E_FC_TX_PAUSE:
-		ethtool_link_ksettings_add_link_mode(cmd, advertising,
+		ethtool_link_ksettings_add_link_mode(ks, advertising,
 						     Asym_Pause);
 		break;
 	case I40E_FC_RX_PAUSE:
-		ethtool_link_ksettings_add_link_mode(cmd, advertising,
-						     Pause);
-		ethtool_link_ksettings_add_link_mode(cmd, advertising,
+		ethtool_link_ksettings_add_link_mode(ks, advertising, Pause);
+		ethtool_link_ksettings_add_link_mode(ks, advertising,
 						     Asym_Pause);
 		break;
 	default:
-		ethtool_convert_link_mode_to_legacy_u32(
-			&advertising, cmd->link_modes.advertising);
-
-		advertising &= ~(ADVERTISED_Pause | ADVERTISED_Asym_Pause);
-
-		ethtool_convert_legacy_u32_to_link_mode(
-			cmd->link_modes.advertising, advertising);
+		ethtool_link_ksettings_del_link_mode(ks, advertising, Pause);
+		ethtool_link_ksettings_del_link_mode(ks, advertising,
+						     Asym_Pause);
 		break;
 	}
 
 	return 0;
 }
 
+#define I40E_LBIT_SIZE 8
+/**
+ * i40e_speed_to_link_speed - Translate decimal speed to i40e_aq_link_speed
+ * @speed: speed in decimal
+ * @ks: ethtool ksettings
+ *
+ * Return i40e_aq_link_speed based on speed
+ **/
+static enum i40e_aq_link_speed
+i40e_speed_to_link_speed(__u32 speed, const struct ethtool_link_ksettings *ks)
+{
+	enum i40e_aq_link_speed link_speed = I40E_LINK_SPEED_UNKNOWN;
+	bool speed_changed = false;
+	int i, j;
+
+	static const struct {
+		__u32 speed;
+		enum i40e_aq_link_speed link_speed;
+		__u8 bit[I40E_LBIT_SIZE];
+	} i40e_speed_lut[] = {
+#define I40E_LBIT(mode) ETHTOOL_LINK_MODE_ ## mode ##_Full_BIT
+		{SPEED_100, I40E_LINK_SPEED_100MB, {I40E_LBIT(100baseT)} },
+		{SPEED_1000, I40E_LINK_SPEED_1GB,
+		 {I40E_LBIT(1000baseT), I40E_LBIT(1000baseX),
+		  I40E_LBIT(1000baseKX)} },
+		{SPEED_10000, I40E_LINK_SPEED_10GB,
+		 {I40E_LBIT(10000baseT), I40E_LBIT(10000baseKR),
+		  I40E_LBIT(10000baseLR), I40E_LBIT(10000baseCR),
+		  I40E_LBIT(10000baseSR), I40E_LBIT(10000baseKX4)} },
+
+		{SPEED_25000, I40E_LINK_SPEED_25GB,
+		 {I40E_LBIT(25000baseCR), I40E_LBIT(25000baseKR),
+		  I40E_LBIT(25000baseSR)} },
+		{SPEED_40000, I40E_LINK_SPEED_40GB,
+		 {I40E_LBIT(40000baseKR4), I40E_LBIT(40000baseCR4),
+		  I40E_LBIT(40000baseSR4), I40E_LBIT(40000baseLR4)} },
+		{SPEED_20000, I40E_LINK_SPEED_20GB,
+		 {I40E_LBIT(20000baseKR2)} },
+		{SPEED_2500, I40E_LINK_SPEED_2_5GB, {I40E_LBIT(2500baseT)} },
+		{SPEED_5000, I40E_LINK_SPEED_5GB, {I40E_LBIT(2500baseT)} }
+#undef I40E_LBIT
+};
+
+	for (i = 0; i < ARRAY_SIZE(i40e_speed_lut); i++) {
+		if (i40e_speed_lut[i].speed == speed) {
+			for (j = 0; j < I40E_LBIT_SIZE; j++) {
+				if (test_bit(i40e_speed_lut[i].bit[j],
+					     ks->link_modes.supported)) {
+					speed_changed = true;
+					break;
+				}
+				if (!i40e_speed_lut[i].bit[j])
+					break;
+			}
+			if (speed_changed) {
+				link_speed = i40e_speed_lut[i].link_speed;
+				break;
+			}
+		}
+	}
+	return link_speed;
+}
+
+#undef I40E_LBIT_SIZE
+
 /**
- * i40e_set_settings - Set Speed and Duplex
+ * i40e_set_link_ksettings - Set Speed and Duplex
  * @netdev: network interface device structure
- * @ecmd: ethtool command
+ * @ks: ethtool ksettings
  *
  * Set speed/duplex per media_types advertised/forced
  **/
 static int i40e_set_link_ksettings(struct net_device *netdev,
-				   const struct ethtool_link_ksettings *cmd)
+				   const struct ethtool_link_ksettings *ks)
 {
 	struct i40e_netdev_priv *np = netdev_priv(netdev);
 	struct i40e_aq_get_phy_abilities_resp abilities;
+	struct ethtool_link_ksettings safe_ks;
+	struct ethtool_link_ksettings copy_ks;
 	struct i40e_aq_set_phy_config config;
 	struct i40e_pf *pf = np->vsi->back;
+	enum i40e_aq_link_speed link_speed;
 	struct i40e_vsi *vsi = np->vsi;
 	struct i40e_hw *hw = &pf->hw;
-	struct ethtool_link_ksettings safe_cmd;
-	struct ethtool_link_ksettings copy_cmd;
-	i40e_status status = 0;
-	bool change = false;
+	bool autoneg_changed = false;
 	int timeout = 50;
+	int status = 0;
 	int err = 0;
-	u32 autoneg;
-	u32 advertise;
-	u32 tmp;
+	__u32 speed;
+	u8 autoneg;
 
 	/* Changing port settings is not supported if this isn't the
 	 * port's controlling PF
@@ -712,51 +1244,58 @@ static int i40e_set_link_ksettings(struct net_device *netdev,
 		i40e_partition_setting_complaint(pf);
 		return -EOPNOTSUPP;
 	}
-
-	if (vsi != pf->vsi[pf->lan_vsi])
+	if (vsi->type != I40E_VSI_MAIN)
 		return -EOPNOTSUPP;
-
 	if (hw->phy.media_type != I40E_MEDIA_TYPE_BASET &&
 	    hw->phy.media_type != I40E_MEDIA_TYPE_FIBER &&
 	    hw->phy.media_type != I40E_MEDIA_TYPE_BACKPLANE &&
 	    hw->phy.media_type != I40E_MEDIA_TYPE_DA &&
 	    hw->phy.link_info.link_info & I40E_AQ_LINK_UP)
 		return -EOPNOTSUPP;
-
 	if (hw->device_id == I40E_DEV_ID_KX_B ||
 	    hw->device_id == I40E_DEV_ID_KX_C ||
 	    hw->device_id == I40E_DEV_ID_20G_KR2 ||
-	    hw->device_id == I40E_DEV_ID_20G_KR2_A) {
+	    hw->device_id == I40E_DEV_ID_20G_KR2_A ||
+	    hw->device_id == I40E_DEV_ID_25G_B ||
+	    hw->device_id == I40E_DEV_ID_KX_X722) {
 		netdev_info(netdev, "Changing settings is not supported on backplane.\n");
 		return -EOPNOTSUPP;
 	}
 
-	/* copy the cmd to copy_cmd to avoid modifying the origin */
-	memcpy(&copy_cmd, cmd, sizeof(struct ethtool_link_ksettings));
+	/* copy the ksettings to copy_ks to avoid modifying the origin */
+	memcpy(&copy_ks, ks, sizeof(struct ethtool_link_ksettings));
 
-	/* get our own copy of the bits to check against */
-	memset(&safe_cmd, 0, sizeof(struct ethtool_link_ksettings));
-	i40e_get_link_ksettings(netdev, &safe_cmd);
-
-	/* save autoneg and speed out of cmd */
-	autoneg = cmd->base.autoneg;
-	ethtool_convert_link_mode_to_legacy_u32(&advertise,
-						cmd->link_modes.advertising);
+	/* save autoneg out of ksettings */
+	autoneg = copy_ks.base.autoneg;
+	speed = copy_ks.base.speed;
 
-	/* set autoneg and speed back to what they currently are */
-	copy_cmd.base.autoneg = safe_cmd.base.autoneg;
-	ethtool_convert_link_mode_to_legacy_u32(
-		&tmp, safe_cmd.link_modes.advertising);
-	ethtool_convert_legacy_u32_to_link_mode(
-		copy_cmd.link_modes.advertising, tmp);
+	/* get our own copy of the bits to check against */
+	memset(&safe_ks, 0, sizeof(struct ethtool_link_ksettings));
+	safe_ks.base.cmd = copy_ks.base.cmd;
+	safe_ks.base.link_mode_masks_nwords =
+		copy_ks.base.link_mode_masks_nwords;
+	i40e_get_link_ksettings(netdev, &safe_ks);
+
+	/* Get link modes supported by hardware and check against modes
+	 * requested by the user.  Return an error if unsupported mode was set.
+	 */
+	if (!bitmap_subset(copy_ks.link_modes.advertising,
+			   safe_ks.link_modes.supported,
+			   __ETHTOOL_LINK_MODE_MASK_NBITS))
+		return -EINVAL;
 
-	copy_cmd.base.cmd = safe_cmd.base.cmd;
+	/* set autoneg back to what it currently is */
+	copy_ks.base.autoneg = safe_ks.base.autoneg;
+	copy_ks.base.speed  = safe_ks.base.speed;
 
-	/* If copy_cmd and safe_cmd are not the same now, then they are
-	 * trying to set something that we do not support
+	/* If copy_ks.base and safe_ks.base are not the same now, then they are
+	 * trying to set something that we do not support.
 	 */
-	if (memcmp(&copy_cmd, &safe_cmd, sizeof(struct ethtool_link_ksettings)))
+	if (memcmp(&copy_ks.base, &safe_ks.base,
+		   sizeof(struct ethtool_link_settings))) {
+		netdev_err(netdev, "Only speed and autoneg are supported.\n");
 		return -EOPNOTSUPP;
+	}
 
 	while (test_and_set_bit(__I40E_CONFIG_BUSY, pf->state)) {
 		timeout--;
@@ -784,8 +1323,9 @@ static int i40e_set_link_ksettings(struct net_device *netdev,
 		/* If autoneg was not already enabled */
 		if (!(hw->phy.link_info.an_info & I40E_AQ_AN_COMPLETED)) {
 			/* If autoneg is not supported, return error */
-			if (!ethtool_link_ksettings_test_link_mode(
-				    &safe_cmd, supported, Autoneg)) {
+			if (!ethtool_link_ksettings_test_link_mode(&safe_ks,
+								   supported,
+								   Autoneg)) {
 				netdev_info(netdev, "Autoneg not supported on this phy\n");
 				err = -EINVAL;
 				goto done;
@@ -793,7 +1333,7 @@ static int i40e_set_link_ksettings(struct net_device *netdev,
 			/* Autoneg is allowed to change */
 			config.abilities = abilities.abilities |
 					   I40E_AQ_PHY_ENABLE_AN;
-			change = true;
+			autoneg_changed = true;
 		}
 	} else {
 		/* If autoneg is currently enabled */
@@ -801,10 +1341,10 @@ static int i40e_set_link_ksettings(struct net_device *netdev,
 			/* If autoneg is supported 10GBASE_T is the only PHY
 			 * that can disable it, so otherwise return error
 			 */
-			if (ethtool_link_ksettings_test_link_mode(
-				    &safe_cmd, supported, Autoneg) &&
-			    hw->phy.link_info.phy_type !=
-			    I40E_PHY_TYPE_10GBASE_T) {
+			if (ethtool_link_ksettings_test_link_mode(&safe_ks,
+								  supported,
+								  Autoneg) &&
+			    hw->phy.media_type != I40E_MEDIA_TYPE_BASET) {
 				netdev_info(netdev, "Autoneg cannot be disabled on this phy\n");
 				err = -EINVAL;
 				goto done;
@@ -812,42 +1352,87 @@ static int i40e_set_link_ksettings(struct net_device *netdev,
 			/* Autoneg is allowed to change */
 			config.abilities = abilities.abilities &
 					   ~I40E_AQ_PHY_ENABLE_AN;
-			change = true;
+			autoneg_changed = true;
 		}
 	}
 
-	ethtool_convert_link_mode_to_legacy_u32(&tmp,
-						safe_cmd.link_modes.supported);
-	if (advertise & ~tmp) {
-		err = -EINVAL;
-		goto done;
-	}
-
-	if (advertise & ADVERTISED_100baseT_Full)
+	if (ethtool_link_ksettings_test_link_mode(ks, advertising,
+						  100baseT_Full))
 		config.link_speed |= I40E_LINK_SPEED_100MB;
-	if (advertise & ADVERTISED_1000baseT_Full ||
-	    advertise & ADVERTISED_1000baseKX_Full)
+	if (ethtool_link_ksettings_test_link_mode(ks, advertising,
+						  1000baseT_Full) ||
+	    ethtool_link_ksettings_test_link_mode(ks, advertising,
+						  1000baseX_Full) ||
+	    ethtool_link_ksettings_test_link_mode(ks, advertising,
+						  1000baseKX_Full))
 		config.link_speed |= I40E_LINK_SPEED_1GB;
-	if (advertise & ADVERTISED_10000baseT_Full ||
-	    advertise & ADVERTISED_10000baseKX4_Full ||
-	    advertise & ADVERTISED_10000baseKR_Full)
+	if (ethtool_link_ksettings_test_link_mode(ks, advertising,
+						  10000baseT_Full) ||
+	    ethtool_link_ksettings_test_link_mode(ks, advertising,
+						  10000baseKX4_Full) ||
+	    ethtool_link_ksettings_test_link_mode(ks, advertising,
+						  10000baseKR_Full) ||
+	    ethtool_link_ksettings_test_link_mode(ks, advertising,
+						  10000baseCR_Full) ||
+	    ethtool_link_ksettings_test_link_mode(ks, advertising,
+						  10000baseSR_Full) ||
+	    ethtool_link_ksettings_test_link_mode(ks, advertising,
+						  10000baseLR_Full))
 		config.link_speed |= I40E_LINK_SPEED_10GB;
-	if (advertise & ADVERTISED_20000baseKR2_Full)
+	if (ethtool_link_ksettings_test_link_mode(ks, advertising,
+						  2500baseT_Full))
+		config.link_speed |= I40E_LINK_SPEED_2_5GB;
+	if (ethtool_link_ksettings_test_link_mode(ks, advertising,
+						  5000baseT_Full))
+		config.link_speed |= I40E_LINK_SPEED_5GB;
+	if (ethtool_link_ksettings_test_link_mode(ks, advertising,
+						  20000baseKR2_Full))
 		config.link_speed |= I40E_LINK_SPEED_20GB;
-	if (advertise & ADVERTISED_40000baseKR4_Full ||
-	    advertise & ADVERTISED_40000baseCR4_Full ||
-	    advertise & ADVERTISED_40000baseSR4_Full ||
-	    advertise & ADVERTISED_40000baseLR4_Full)
+	if (ethtool_link_ksettings_test_link_mode(ks, advertising,
+						  25000baseCR_Full) ||
+	    ethtool_link_ksettings_test_link_mode(ks, advertising,
+						  25000baseKR_Full) ||
+	    ethtool_link_ksettings_test_link_mode(ks, advertising,
+						  25000baseSR_Full))
+		config.link_speed |= I40E_LINK_SPEED_25GB;
+	if (ethtool_link_ksettings_test_link_mode(ks, advertising,
+						  40000baseKR4_Full) ||
+	    ethtool_link_ksettings_test_link_mode(ks, advertising,
+						  40000baseCR4_Full) ||
+	    ethtool_link_ksettings_test_link_mode(ks, advertising,
+						  40000baseSR4_Full) ||
+	    ethtool_link_ksettings_test_link_mode(ks, advertising,
+						  40000baseLR4_Full))
 		config.link_speed |= I40E_LINK_SPEED_40GB;
 
+	/* Autonegotiation must be disabled to change speed */
+	if ((speed != SPEED_UNKNOWN && safe_ks.base.speed != speed) &&
+	    (autoneg == AUTONEG_DISABLE ||
+	    (safe_ks.base.autoneg == AUTONEG_DISABLE && !autoneg_changed))) {
+		link_speed = i40e_speed_to_link_speed(speed, ks);
+		if (link_speed == I40E_LINK_SPEED_UNKNOWN) {
+			netdev_info(netdev, "Given speed is not supported\n");
+			err = -EOPNOTSUPP;
+			goto done;
+		} else {
+			config.link_speed = link_speed;
+		}
+	} else {
+		if (safe_ks.base.speed != speed) {
+			netdev_info(netdev,
+				    "Unable to set speed, disable autoneg\n");
+			err = -EOPNOTSUPP;
+			goto done;
+		}
+	}
+
 	/* If speed didn't get set, set it to what it currently is.
 	 * This is needed because if advertise is 0 (as it is when autoneg
 	 * is disabled) then speed won't get set.
 	 */
 	if (!config.link_speed)
 		config.link_speed = abilities.link_speed;
-
-	if (change || (abilities.link_speed != config.link_speed)) {
+	if (autoneg_changed || abilities.link_speed != config.link_speed) {
 		/* copy over the rest of the abilities */
 		config.phy_type = abilities.phy_type;
 		config.phy_type_ext = abilities.phy_type_ext;
@@ -874,18 +1459,20 @@ static int i40e_set_link_ksettings(struct net_device *netdev,
 		/* make the aq call */
 		status = i40e_aq_set_phy_config(hw, &config, NULL);
 		if (status) {
-			netdev_info(netdev, "Set phy config failed, err %s aq_err %s\n",
-				    i40e_stat_str(hw, status),
-				    i40e_aq_str(hw, hw->aq.asq_last_status));
+			netdev_info(netdev,
+				    "Set phy config failed, err %pe aq_err %s\n",
+				    ERR_PTR(status),
+				    libie_aq_str(hw->aq.asq_last_status));
 			err = -EAGAIN;
 			goto done;
 		}
 
 		status = i40e_update_link_info(hw);
 		if (status)
-			netdev_dbg(netdev, "Updating link info failed with err %s aq_err %s\n",
-				   i40e_stat_str(hw, status),
-				   i40e_aq_str(hw, hw->aq.asq_last_status));
+			netdev_dbg(netdev,
+				   "Updating link info failed with err %pe aq_err %s\n",
+				   ERR_PTR(status),
+				   libie_aq_str(hw->aq.asq_last_status));
 
 	} else {
 		netdev_info(netdev, "Nothing changed, exiting without setting anything.\n");
@@ -897,6 +1484,150 @@ done:
 	return err;
 }
 
+static int i40e_set_fec_cfg(struct net_device *netdev, u8 fec_cfg)
+{
+	struct i40e_netdev_priv *np = netdev_priv(netdev);
+	struct i40e_aq_get_phy_abilities_resp abilities;
+	struct i40e_pf *pf = np->vsi->back;
+	struct i40e_hw *hw = &pf->hw;
+	int status = 0;
+	int err = 0;
+
+	/* Get the current phy config */
+	memset(&abilities, 0, sizeof(abilities));
+	status = i40e_aq_get_phy_capabilities(hw, false, false, &abilities,
+					      NULL);
+	if (status) {
+		err = -EAGAIN;
+		goto done;
+	}
+
+	if (abilities.fec_cfg_curr_mod_ext_info != fec_cfg) {
+		struct i40e_aq_set_phy_config config;
+
+		memset(&config, 0, sizeof(config));
+		config.phy_type = abilities.phy_type;
+		config.abilities = abilities.abilities |
+				   I40E_AQ_PHY_ENABLE_ATOMIC_LINK;
+		config.phy_type_ext = abilities.phy_type_ext;
+		config.link_speed = abilities.link_speed;
+		config.eee_capability = abilities.eee_capability;
+		config.eeer = abilities.eeer_val;
+		config.low_power_ctrl = abilities.d3_lpan;
+		config.fec_config = fec_cfg & I40E_AQ_PHY_FEC_CONFIG_MASK;
+		status = i40e_aq_set_phy_config(hw, &config, NULL);
+		if (status) {
+			netdev_info(netdev,
+				    "Set phy config failed, err %pe aq_err %s\n",
+				    ERR_PTR(status),
+				    libie_aq_str(hw->aq.asq_last_status));
+			err = -EAGAIN;
+			goto done;
+		}
+		i40e_set_fec_in_flags(fec_cfg, pf->flags);
+		status = i40e_update_link_info(hw);
+		if (status)
+			/* debug level message only due to relation to the link
+			 * itself rather than to the FEC settings
+			 * (e.g. no physical connection etc.)
+			 */
+			netdev_dbg(netdev,
+				   "Updating link info failed with err %pe aq_err %s\n",
+				   ERR_PTR(status),
+				   libie_aq_str(hw->aq.asq_last_status));
+	}
+
+done:
+	return err;
+}
+
+static int i40e_get_fec_param(struct net_device *netdev,
+			      struct ethtool_fecparam *fecparam)
+{
+	struct i40e_netdev_priv *np = netdev_priv(netdev);
+	struct i40e_aq_get_phy_abilities_resp abilities;
+	struct i40e_pf *pf = np->vsi->back;
+	struct i40e_hw *hw = &pf->hw;
+	int status = 0;
+	int err = 0;
+	u8 fec_cfg;
+
+	/* Get the current phy config */
+	memset(&abilities, 0, sizeof(abilities));
+	status = i40e_aq_get_phy_capabilities(hw, false, false, &abilities,
+					      NULL);
+	if (status) {
+		err = -EAGAIN;
+		goto done;
+	}
+
+	fecparam->fec = 0;
+	fec_cfg = abilities.fec_cfg_curr_mod_ext_info;
+	if (fec_cfg & I40E_AQ_SET_FEC_AUTO)
+		fecparam->fec |= ETHTOOL_FEC_AUTO;
+	else if (fec_cfg & (I40E_AQ_SET_FEC_REQUEST_RS |
+		 I40E_AQ_SET_FEC_ABILITY_RS))
+		fecparam->fec |= ETHTOOL_FEC_RS;
+	else if (fec_cfg & (I40E_AQ_SET_FEC_REQUEST_KR |
+		 I40E_AQ_SET_FEC_ABILITY_KR))
+		fecparam->fec |= ETHTOOL_FEC_BASER;
+	if (fec_cfg == 0)
+		fecparam->fec |= ETHTOOL_FEC_OFF;
+
+	if (hw->phy.link_info.fec_info & I40E_AQ_CONFIG_FEC_KR_ENA)
+		fecparam->active_fec = ETHTOOL_FEC_BASER;
+	else if (hw->phy.link_info.fec_info & I40E_AQ_CONFIG_FEC_RS_ENA)
+		fecparam->active_fec = ETHTOOL_FEC_RS;
+	else
+		fecparam->active_fec = ETHTOOL_FEC_OFF;
+done:
+	return err;
+}
+
+static int i40e_set_fec_param(struct net_device *netdev,
+			      struct ethtool_fecparam *fecparam)
+{
+	struct i40e_netdev_priv *np = netdev_priv(netdev);
+	struct i40e_pf *pf = np->vsi->back;
+	struct i40e_hw *hw = &pf->hw;
+	u8 fec_cfg = 0;
+
+	if (hw->device_id != I40E_DEV_ID_25G_SFP28 &&
+	    hw->device_id != I40E_DEV_ID_25G_B &&
+	    hw->device_id != I40E_DEV_ID_KX_X722)
+		return -EPERM;
+
+	if (hw->mac.type == I40E_MAC_X722 &&
+	    !test_bit(I40E_HW_CAP_X722_FEC_REQUEST, hw->caps)) {
+		netdev_err(netdev, "Setting FEC encoding not supported by firmware. Please update the NVM image.\n");
+		return -EOPNOTSUPP;
+	}
+
+	switch (fecparam->fec) {
+	case ETHTOOL_FEC_AUTO:
+		fec_cfg = I40E_AQ_SET_FEC_AUTO;
+		break;
+	case ETHTOOL_FEC_RS:
+		fec_cfg = (I40E_AQ_SET_FEC_REQUEST_RS |
+			     I40E_AQ_SET_FEC_ABILITY_RS);
+		break;
+	case ETHTOOL_FEC_BASER:
+		fec_cfg = (I40E_AQ_SET_FEC_REQUEST_KR |
+			     I40E_AQ_SET_FEC_ABILITY_KR);
+		break;
+	case ETHTOOL_FEC_OFF:
+	case ETHTOOL_FEC_NONE:
+		fec_cfg = 0;
+		break;
+	default:
+		dev_warn(&pf->pdev->dev, "Unsupported FEC mode: %d",
+			 fecparam->fec);
+		return -EINVAL;
+	}
+
+	return i40e_set_fec_cfg(netdev, fec_cfg);
+}
+
 static int i40e_nway_reset(struct net_device *netdev)
 {
 	/* restart autonegotiation */
@@ -904,13 +1635,13 @@ static int i40e_nway_reset(struct net_device *netdev)
 	struct i40e_pf *pf = np->vsi->back;
 	struct i40e_hw *hw = &pf->hw;
 	bool link_up = hw->phy.link_info.link_info & I40E_AQ_LINK_UP;
-	i40e_status ret = 0;
+	int ret = 0;
 
 	ret = i40e_aq_set_link_restart_an(hw, link_up, NULL);
 	if (ret) {
-		netdev_info(netdev, "link restart failed, err %s aq_err %s\n",
-			    i40e_stat_str(hw, ret),
-			    i40e_aq_str(hw, hw->aq.asq_last_status));
+		netdev_info(netdev, "link restart failed, err %pe aq_err %s\n",
+			    ERR_PTR(ret),
+			    libie_aq_str(hw->aq.asq_last_status));
 		return -EIO;
 	}
 
@@ -919,6 +1650,9 @@ static int i40e_nway_reset(struct net_device *netdev)
 
 /**
  * i40e_get_pauseparam -  Get Flow Control status
+ * @netdev: netdevice structure
+ * @pause: buffer to return pause parameters
+ *
  * Return tx/rx-pause status
  **/
 static void i40e_get_pauseparam(struct net_device *netdev,
@@ -966,9 +1700,10 @@ static int i40e_set_pauseparam(struct net_device *netdev,
 	struct i40e_link_status *hw_link_info = &hw->phy.link_info;
 	struct i40e_dcbx_config *dcbx_cfg = &hw->local_dcbx_config;
 	bool link_up = hw_link_info->link_info & I40E_AQ_LINK_UP;
-	i40e_status status;
 	u8 aq_failures;
 	int err = 0;
+	int status;
+	u32 is_an;
 
 	/* Changing the port's flow control is not supported if this isn't the
 	 * port's controlling PF
@@ -978,18 +1713,17 @@ static int i40e_set_pauseparam(struct net_device *netdev,
 		return -EOPNOTSUPP;
 	}
 
-	if (vsi != pf->vsi[pf->lan_vsi])
+	if (vsi->type != I40E_VSI_MAIN)
 		return -EOPNOTSUPP;
 
-	if (pause->autoneg != ((hw_link_info->an_info & I40E_AQ_AN_COMPLETED) ?
-	    AUTONEG_ENABLE : AUTONEG_DISABLE)) {
+	is_an = hw_link_info->an_info & I40E_AQ_AN_COMPLETED;
+	if (pause->autoneg != is_an) {
 		netdev_info(netdev, "To change autoneg please use: ethtool -s <dev> autoneg <on|off>\n");
 		return -EOPNOTSUPP;
 	}
 
 	/* If we have link and don't have autoneg */
-	if (!test_bit(__I40E_DOWN, pf->state) &&
-	    !(hw_link_info->an_info & I40E_AQ_AN_COMPLETED)) {
+	if (!test_bit(__I40E_DOWN, pf->state) && !is_an) {
 		/* Send message that it might not necessarily work*/
 		netdev_info(netdev, "Autoneg did not complete so changing settings may not result in an actual change.\n");
 	}
@@ -1009,7 +1743,7 @@ static int i40e_set_pauseparam(struct net_device *netdev,
 	else if (!pause->rx_pause && !pause->tx_pause)
 		hw->fc.requested_mode = I40E_FC_NONE;
 	else
-		 return -EINVAL;
+		return -EINVAL;
 
 	/* Tell the OS link is going down, the link will go back up when fw
 	 * says it is ready asynchronously
@@ -1022,25 +1756,25 @@ static int i40e_set_pauseparam(struct net_device *netdev,
 	status = i40e_set_fc(hw, &aq_failures, link_up);
 
 	if (aq_failures & I40E_SET_FC_AQ_FAIL_GET) {
-		netdev_info(netdev, "Set fc failed on the get_phy_capabilities call with err %s aq_err %s\n",
-			    i40e_stat_str(hw, status),
-			    i40e_aq_str(hw, hw->aq.asq_last_status));
+		netdev_info(netdev, "Set fc failed on the get_phy_capabilities call with err %pe aq_err %s\n",
+			    ERR_PTR(status),
+			    libie_aq_str(hw->aq.asq_last_status));
 		err = -EAGAIN;
 	}
 	if (aq_failures & I40E_SET_FC_AQ_FAIL_SET) {
-		netdev_info(netdev, "Set fc failed on the set_phy_config call with err %s aq_err %s\n",
-			    i40e_stat_str(hw, status),
-			    i40e_aq_str(hw, hw->aq.asq_last_status));
+		netdev_info(netdev, "Set fc failed on the set_phy_config call with err %pe aq_err %s\n",
+			    ERR_PTR(status),
+			    libie_aq_str(hw->aq.asq_last_status));
 		err = -EAGAIN;
 	}
 	if (aq_failures & I40E_SET_FC_AQ_FAIL_UPDATE) {
-		netdev_info(netdev, "Set fc failed on the get_link_info call with err %s aq_err %s\n",
-			    i40e_stat_str(hw, status),
-			    i40e_aq_str(hw, hw->aq.asq_last_status));
+		netdev_info(netdev, "Set fc failed on the get_link_info call with err %pe aq_err %s\n",
+			    ERR_PTR(status),
+			    libie_aq_str(hw->aq.asq_last_status));
 		err = -EAGAIN;
 	}
 
-	if (!test_bit(__I40E_DOWN, pf->state)) {
+	if (!test_bit(__I40E_DOWN, pf->state) && is_an) {
 		/* Give it a little more time to try to come back */
 		msleep(75);
 		if (!test_bit(__I40E_DOWN, pf->state))
@@ -1091,7 +1825,7 @@ static void i40e_get_regs(struct net_device *netdev, struct ethtool_regs *regs,
 	struct i40e_pf *pf = np->vsi->back;
 	struct i40e_hw *hw = &pf->hw;
 	u32 *reg_buf = p;
-	int i, j, ri;
+	unsigned int i, j, ri;
 	u32 reg;
 
 	/* Tell ethtool which driver-version-specific regs output we have.
@@ -1180,17 +1914,17 @@ static int i40e_get_eeprom(struct net_device *netdev,
 			len = eeprom->len - (I40E_NVM_SECTOR_SIZE * i);
 			last = true;
 		}
-		offset = eeprom->offset + (I40E_NVM_SECTOR_SIZE * i),
+		offset = eeprom->offset + (I40E_NVM_SECTOR_SIZE * i);
 		ret_val = i40e_aq_read_nvm(hw, 0x0, offset, len,
 				(u8 *)eeprom_buff + (I40E_NVM_SECTOR_SIZE * i),
 				last, NULL);
-		if (ret_val && hw->aq.asq_last_status == I40E_AQ_RC_EPERM) {
+		if (ret_val && hw->aq.asq_last_status == LIBIE_AQ_RC_EPERM) {
 			dev_info(&pf->pdev->dev,
 				 "read NVM failed, invalid offset 0x%x\n",
 				 offset);
 			break;
 		} else if (ret_val &&
-			   hw->aq.asq_last_status == I40E_AQ_RC_EACCES) {
+			   hw->aq.asq_last_status == LIBIE_AQ_RC_EACCES) {
 			dev_info(&pf->pdev->dev,
 				 "read NVM failed, access, offset 0x%x\n",
 				 offset);
@@ -1221,9 +1955,8 @@ static int i40e_get_eeprom_len(struct net_device *netdev)
 		val = X722_EEPROM_SCOPE_LIMIT + 1;
 		return val;
 	}
-	val = (rd32(hw, I40E_GLPCI_LBARCTRL)
-		& I40E_GLPCI_LBARCTRL_FL_SIZE_MASK)
-		>> I40E_GLPCI_LBARCTRL_FL_SIZE_SHIFT;
+	val = FIELD_GET(I40E_GLPCI_LBARCTRL_FL_SIZE_MASK,
+			rd32(hw, I40E_GLPCI_LBARCTRL));
 	/* register returns value in power of 2, 64Kbyte chunks. */
 	val = (64 * 1024) * BIT(val);
 	return val;
@@ -1270,27 +2003,39 @@ static void i40e_get_drvinfo(struct net_device *netdev,
 	struct i40e_vsi *vsi = np->vsi;
 	struct i40e_pf *pf = vsi->back;
 
-	strlcpy(drvinfo->driver, i40e_driver_name, sizeof(drvinfo->driver));
-	strlcpy(drvinfo->version, i40e_driver_version_str,
-		sizeof(drvinfo->version));
-	strlcpy(drvinfo->fw_version, i40e_nvm_version_str(&pf->hw),
-		sizeof(drvinfo->fw_version));
-	strlcpy(drvinfo->bus_info, pci_name(pf->pdev),
+	strscpy(drvinfo->driver, i40e_driver_name, sizeof(drvinfo->driver));
+	i40e_nvm_version_str(&pf->hw, drvinfo->fw_version,
+			     sizeof(drvinfo->fw_version));
+	strscpy(drvinfo->bus_info, pci_name(pf->pdev),
 		sizeof(drvinfo->bus_info));
 	drvinfo->n_priv_flags = I40E_PRIV_FLAGS_STR_LEN;
 	if (pf->hw.pf_id == 0)
 		drvinfo->n_priv_flags += I40E_GL_PRIV_FLAGS_STR_LEN;
 }
 
+static u32 i40e_get_max_num_descriptors(struct i40e_pf *pf)
+{
+	struct i40e_hw *hw = &pf->hw;
+
+	switch (hw->mac.type) {
+	case I40E_MAC_XL710:
+		return I40E_MAX_NUM_DESCRIPTORS_XL710;
+	default:
+		return I40E_MAX_NUM_DESCRIPTORS;
+	}
+}
+
 static void i40e_get_ringparam(struct net_device *netdev,
-			       struct ethtool_ringparam *ring)
+			       struct ethtool_ringparam *ring,
+			       struct kernel_ethtool_ringparam *kernel_ring,
+			       struct netlink_ext_ack *extack)
 {
 	struct i40e_netdev_priv *np = netdev_priv(netdev);
 	struct i40e_pf *pf = np->vsi->back;
-	struct i40e_vsi *vsi = pf->vsi[pf->lan_vsi];
+	struct i40e_vsi *vsi = i40e_pf_get_main_vsi(pf);
 
-	ring->rx_max_pending = I40E_MAX_NUM_DESCRIPTORS;
-	ring->tx_max_pending = I40E_MAX_NUM_DESCRIPTORS;
+	ring->rx_max_pending = i40e_get_max_num_descriptors(pf);
+	ring->tx_max_pending = i40e_get_max_num_descriptors(pf);
 	ring->rx_mini_max_pending = 0;
 	ring->rx_jumbo_max_pending = 0;
 	ring->rx_pending = vsi->rx_rings[0]->count;
@@ -1299,29 +2044,44 @@ static void i40e_get_ringparam(struct net_device *netdev,
 	ring->rx_jumbo_pending = 0;
 }
 
+static bool i40e_active_tx_ring_index(struct i40e_vsi *vsi, u16 index)
+{
+	if (i40e_enabled_xdp_vsi(vsi)) {
+		return index < vsi->num_queue_pairs ||
+			(index >= vsi->alloc_queue_pairs &&
+			 index < vsi->alloc_queue_pairs + vsi->num_queue_pairs);
+	}
+
+	return index < vsi->num_queue_pairs;
+}
+
 static int i40e_set_ringparam(struct net_device *netdev,
-			      struct ethtool_ringparam *ring)
+			      struct ethtool_ringparam *ring,
+			      struct kernel_ethtool_ringparam *kernel_ring,
+			      struct netlink_ext_ack *extack)
 {
+	u32 new_rx_count, new_tx_count, max_num_descriptors;
 	struct i40e_ring *tx_rings = NULL, *rx_rings = NULL;
 	struct i40e_netdev_priv *np = netdev_priv(netdev);
 	struct i40e_hw *hw = &np->vsi->back->hw;
 	struct i40e_vsi *vsi = np->vsi;
 	struct i40e_pf *pf = vsi->back;
-	u32 new_rx_count, new_tx_count;
+	u16 tx_alloc_queue_pairs;
 	int timeout = 50;
 	int i, err = 0;
 
 	if ((ring->rx_mini_pending) || (ring->rx_jumbo_pending))
 		return -EINVAL;
 
-	if (ring->tx_pending > I40E_MAX_NUM_DESCRIPTORS ||
+	max_num_descriptors = i40e_get_max_num_descriptors(pf);
+	if (ring->tx_pending > max_num_descriptors ||
 	    ring->tx_pending < I40E_MIN_NUM_DESCRIPTORS ||
-	    ring->rx_pending > I40E_MAX_NUM_DESCRIPTORS ||
+	    ring->rx_pending > max_num_descriptors ||
 	    ring->rx_pending < I40E_MIN_NUM_DESCRIPTORS) {
 		netdev_info(netdev,
 			    "Descriptors requested (Tx: %d / Rx: %d) out of range [%d-%d]\n",
 			    ring->tx_pending, ring->rx_pending,
-			    I40E_MIN_NUM_DESCRIPTORS, I40E_MAX_NUM_DESCRIPTORS);
+			    I40E_MIN_NUM_DESCRIPTORS, max_num_descriptors);
 		return -EINVAL;
 	}
 
@@ -1333,6 +2093,13 @@ static int i40e_set_ringparam(struct net_device *netdev,
 	    (new_rx_count == vsi->rx_rings[0]->count))
 		return 0;
 
+	/* If there is a AF_XDP page pool attached to any of Rx rings,
+	 * disallow changing the number of descriptors -- regardless
+	 * if the netdev is running or not.
+	 */
+	if (i40e_xsk_any_rx_ring_enabled(vsi))
+		return -EBUSY;
+
 	while (test_and_set_bit(__I40E_CONFIG_BUSY, pf->state)) {
 		timeout--;
 		if (!timeout)
@@ -1345,7 +2112,11 @@ static int i40e_set_ringparam(struct net_device *netdev,
 		for (i = 0; i < vsi->num_queue_pairs; i++) {
 			vsi->tx_rings[i]->count = new_tx_count;
 			vsi->rx_rings[i]->count = new_rx_count;
+			if (i40e_enabled_xdp_vsi(vsi))
+				vsi->xdp_rings[i]->count = new_tx_count;
 		}
+		vsi->num_tx_desc = new_tx_count;
+		vsi->num_rx_desc = new_rx_count;
 		goto done;
 	}
 
@@ -1354,20 +2125,24 @@ static int i40e_set_ringparam(struct net_device *netdev,
 	 * to the Tx and Rx ring structs.
 	 */
 
-	/* alloc updated Tx resources */
+	/* alloc updated Tx and XDP Tx resources */
+	tx_alloc_queue_pairs = vsi->alloc_queue_pairs *
+			       (i40e_enabled_xdp_vsi(vsi) ? 2 : 1);
 	if (new_tx_count != vsi->tx_rings[0]->count) {
 		netdev_info(netdev,
 			    "Changing Tx descriptor count from %d to %d.\n",
 			    vsi->tx_rings[0]->count, new_tx_count);
-		tx_rings = kcalloc(vsi->alloc_queue_pairs,
+		tx_rings = kcalloc(tx_alloc_queue_pairs,
 				   sizeof(struct i40e_ring), GFP_KERNEL);
 		if (!tx_rings) {
 			err = -ENOMEM;
 			goto done;
 		}
 
-		for (i = 0; i < vsi->num_queue_pairs; i++) {
-			/* clone ring and setup updated count */
+		for (i = 0; i < tx_alloc_queue_pairs; i++) {
+			if (!i40e_active_tx_ring_index(vsi, i))
+				continue;
+
 			tx_rings[i] = *vsi->tx_rings[i];
 			tx_rings[i].count = new_tx_count;
 			/* the desc and bi pointers will be reallocated in the
@@ -1379,6 +2154,8 @@ static int i40e_set_ringparam(struct net_device *netdev,
 			if (err) {
 				while (i) {
 					i--;
+					if (!i40e_active_tx_ring_index(vsi, i))
+						continue;
 					i40e_free_tx_resources(&tx_rings[i]);
 				}
 				kfree(tx_rings);
@@ -1402,7 +2179,6 @@ static int i40e_set_ringparam(struct net_device *netdev,
 		}
 
 		for (i = 0; i < vsi->num_queue_pairs; i++) {
-			struct i40e_ring *ring;
 			u16 unused;
 
 			/* clone ring and setup updated count */
@@ -1413,6 +2189,8 @@ static int i40e_set_ringparam(struct net_device *netdev,
 			 */
 			rx_rings[i].desc = NULL;
 			rx_rings[i].rx_bi = NULL;
+			/* Clear cloned XDP RX-queue info before setup call */
+			memset(&rx_rings[i].xdp_rxq, 0, sizeof(rx_rings[i].xdp_rxq));
 			/* this is to allow wr32 to have something to write to
 			 * during early allocation of Rx buffers
 			 */
@@ -1424,9 +2202,8 @@ static int i40e_set_ringparam(struct net_device *netdev,
 			/* now allocate the Rx buffers to make sure the OS
 			 * has enough memory, any failure here means abort
 			 */
-			ring = &rx_rings[i];
-			unused = I40E_DESC_UNUSED(ring);
-			err = i40e_alloc_rx_buffers(ring, unused);
+			unused = I40E_DESC_UNUSED(&rx_rings[i]);
+			err = i40e_alloc_rx_buffers(&rx_rings[i], unused);
 rx_unwind:
 			if (err) {
 				do {
@@ -1446,9 +2223,11 @@ rx_unwind:
 	i40e_down(vsi);
 
 	if (tx_rings) {
-		for (i = 0; i < vsi->num_queue_pairs; i++) {
-			i40e_free_tx_resources(vsi->tx_rings[i]);
-			*vsi->tx_rings[i] = tx_rings[i];
+		for (i = 0; i < tx_alloc_queue_pairs; i++) {
+			if (i40e_active_tx_ring_index(vsi, i)) {
+				i40e_free_tx_resources(vsi->tx_rings[i]);
+				*vsi->tx_rings[i] = tx_rings[i];
+			}
 		}
 		kfree(tx_rings);
 		tx_rings = NULL;
@@ -1474,13 +2253,17 @@ rx_unwind:
 		rx_rings = NULL;
 	}
 
+	vsi->num_tx_desc = new_tx_count;
+	vsi->num_rx_desc = new_rx_count;
 	i40e_up(vsi);
 
 free_tx:
 	/* error cleanup if the Rx allocations failed after getting Tx */
 	if (tx_rings) {
-		for (i = 0; i < vsi->num_queue_pairs; i++)
-			i40e_free_tx_resources(&tx_rings[i]);
+		for (i = 0; i < tx_alloc_queue_pairs; i++) {
+			if (i40e_active_tx_ring_index(vsi, i))
+				i40e_free_tx_resources(vsi->tx_rings[i]);
+		}
 		kfree(tx_rings);
 		tx_rings = NULL;
 	}
@@ -1491,6 +2274,51 @@ done:
 	return err;
 }
 
+/**
+ * i40e_get_stats_count - return the stats count for a device
+ * @netdev: the netdev to return the count for
+ *
+ * Returns the total number of statistics for this netdev. Note that even
+ * though this is a function, it is required that the count for a specific
+ * netdev must never change. Basing the count on static values such as the
+ * maximum number of queues or the device type is ok. However, the API for
+ * obtaining stats is *not* safe against changes based on non-static
+ * values such as the *current* number of queues, or runtime flags.
+ *
+ * If a statistic is not always enabled, return it as part of the count
+ * anyways, always return its string, and report its value as zero.
+ **/
+static int i40e_get_stats_count(struct net_device *netdev)
+{
+	struct i40e_netdev_priv *np = netdev_priv(netdev);
+	struct i40e_vsi *vsi = np->vsi;
+	struct i40e_pf *pf = vsi->back;
+	int stats_len;
+
+	if (vsi->type == I40E_VSI_MAIN && pf->hw.partition_id == 1)
+		stats_len = I40E_PF_STATS_LEN;
+	else
+		stats_len = I40E_VSI_STATS_LEN;
+
+	/* The number of stats reported for a given net_device must remain
+	 * constant throughout the life of that device.
+	 *
+	 * This is because the API for obtaining the size, strings, and stats
+	 * is spread out over three separate ethtool ioctls. There is no safe
+	 * way to lock the number of stats across these calls, so we must
+	 * assume that they will never change.
+	 *
+	 * Due to this, we report the maximum number of queues, even if not
+	 * every queue is currently configured. Since we always allocate
+	 * queues in pairs, we'll just use netdev->num_tx_queues * 2. This
+	 * works because the num_tx_queues is set at device creation and never
+	 * changes.
+	 */
+	stats_len += I40E_QUEUE_STATS_LEN * 2 * netdev->num_tx_queues;
+
+	return stats_len;
+}
+
 static int i40e_get_sset_count(struct net_device *netdev, int sset)
 {
 	struct i40e_netdev_priv *np = netdev_priv(netdev);
@@ -1501,16 +2329,7 @@ static int i40e_get_sset_count(struct net_device *netdev, int sset)
 	case ETH_SS_TEST:
 		return I40E_TEST_LEN;
 	case ETH_SS_STATS:
-		if (vsi == pf->vsi[pf->lan_vsi] && pf->hw.partition_id == 1) {
-			int len = I40E_PF_STATS_LEN(netdev);
-
-			if ((pf->lan_veb != I40E_NO_VEB) &&
-			    (pf->flags & I40E_FLAG_VEB_STATS_ENABLED))
-				len += I40E_VEB_STATS_TOTAL;
-			return len;
-		} else {
-			return I40E_VSI_STATS_LEN(netdev);
-		}
+		return i40e_get_stats_count(netdev);
 	case ETH_SS_PRIV_FLAGS:
 		return I40E_PRIV_FLAGS_STR_LEN +
 			(pf->hw.pf_id == 0 ? I40E_GL_PRIV_FLAGS_STR_LEN : 0);
@@ -1519,194 +2338,210 @@ static int i40e_get_sset_count(struct net_device *netdev, int sset)
 	}
 }
 
+/**
+ * i40e_get_veb_tc_stats - copy VEB TC statistics to formatted structure
+ * @tc: the TC statistics in VEB structure (veb->tc_stats)
+ * @i: the index of traffic class in (veb->tc_stats) structure to copy
+ *
+ * Copy VEB TC statistics from structure of arrays (veb->tc_stats) to
+ * one dimensional structure i40e_cp_veb_tc_stats.
+ * Produce formatted i40e_cp_veb_tc_stats structure of the VEB TC
+ * statistics for the given TC.
+ **/
+static struct i40e_cp_veb_tc_stats
+i40e_get_veb_tc_stats(struct i40e_veb_tc_stats *tc, unsigned int i)
+{
+	struct i40e_cp_veb_tc_stats veb_tc = {
+		.tc_rx_packets = tc->tc_rx_packets[i],
+		.tc_rx_bytes = tc->tc_rx_bytes[i],
+		.tc_tx_packets = tc->tc_tx_packets[i],
+		.tc_tx_bytes = tc->tc_tx_bytes[i],
+	};
+
+	return veb_tc;
+}
+
+/**
+ * i40e_get_pfc_stats - copy HW PFC statistics to formatted structure
+ * @pf: the PF device structure
+ * @i: the priority value to copy
+ *
+ * The PFC stats are found as arrays in pf->stats, which is not easy to pass
+ * into i40e_add_ethtool_stats. Produce a formatted i40e_pfc_stats structure
+ * of the PFC stats for the given priority.
+ **/
+static inline struct i40e_pfc_stats
+i40e_get_pfc_stats(struct i40e_pf *pf, unsigned int i)
+{
+#define I40E_GET_PFC_STAT(stat, priority) \
+	.stat = pf->stats.stat[priority]
+
+	struct i40e_pfc_stats pfc = {
+		I40E_GET_PFC_STAT(priority_xon_rx, i),
+		I40E_GET_PFC_STAT(priority_xoff_rx, i),
+		I40E_GET_PFC_STAT(priority_xon_tx, i),
+		I40E_GET_PFC_STAT(priority_xoff_tx, i),
+		I40E_GET_PFC_STAT(priority_xon_2_xoff, i),
+	};
+	return pfc;
+}
+
+/**
+ * i40e_get_ethtool_stats - copy stat values into supplied buffer
+ * @netdev: the netdev to collect stats for
+ * @stats: ethtool stats command structure
+ * @data: ethtool supplied buffer
+ *
+ * Copy the stats values for this netdev into the buffer. Expects data to be
+ * pre-allocated to the size returned by i40e_get_stats_count.. Note that all
+ * statistics must be copied in a static order, and the count must not change
+ * for a given netdev. See i40e_get_stats_count for more details.
+ *
+ * If a statistic is not currently valid (such as a disabled queue), this
+ * function reports its value as zero.
+ **/
 static void i40e_get_ethtool_stats(struct net_device *netdev,
 				   struct ethtool_stats *stats, u64 *data)
 {
 	struct i40e_netdev_priv *np = netdev_priv(netdev);
-	struct i40e_ring *tx_ring, *rx_ring;
 	struct i40e_vsi *vsi = np->vsi;
 	struct i40e_pf *pf = vsi->back;
-	int i = 0;
-	char *p;
-	int j;
-	struct rtnl_link_stats64 *net_stats = i40e_get_vsi_stats_struct(vsi);
-	unsigned int start;
+	struct i40e_veb *veb = NULL;
+	unsigned int i;
+	bool veb_stats;
+	u64 *p = data;
 
 	i40e_update_stats(vsi);
 
-	for (j = 0; j < I40E_NETDEV_STATS_LEN; j++) {
-		p = (char *)net_stats + i40e_gstrings_net_stats[j].stat_offset;
-		data[i++] = (i40e_gstrings_net_stats[j].sizeof_stat ==
-			sizeof(u64)) ? *(u64 *)p : *(u32 *)p;
-	}
-	for (j = 0; j < I40E_MISC_STATS_LEN; j++) {
-		p = (char *)vsi + i40e_gstrings_misc_stats[j].stat_offset;
-		data[i++] = (i40e_gstrings_misc_stats[j].sizeof_stat ==
-			    sizeof(u64)) ? *(u64 *)p : *(u32 *)p;
-	}
-	rcu_read_lock();
-	for (j = 0; j < vsi->num_queue_pairs; j++) {
-		tx_ring = ACCESS_ONCE(vsi->tx_rings[j]);
+	i40e_add_ethtool_stats(&data, i40e_get_vsi_stats_struct(vsi),
+			       i40e_gstrings_net_stats);
 
-		if (!tx_ring)
-			continue;
+	i40e_add_ethtool_stats(&data, vsi, i40e_gstrings_misc_stats);
 
-		/* process Tx ring statistics */
-		do {
-			start = u64_stats_fetch_begin_irq(&tx_ring->syncp);
-			data[i] = tx_ring->stats.packets;
-			data[i + 1] = tx_ring->stats.bytes;
-		} while (u64_stats_fetch_retry_irq(&tx_ring->syncp, start));
-		i += 2;
-
-		/* Rx ring is the 2nd half of the queue pair */
-		rx_ring = &tx_ring[1];
-		do {
-			start = u64_stats_fetch_begin_irq(&rx_ring->syncp);
-			data[i] = rx_ring->stats.packets;
-			data[i + 1] = rx_ring->stats.bytes;
-		} while (u64_stats_fetch_retry_irq(&rx_ring->syncp, start));
-		i += 2;
+	rcu_read_lock();
+	for (i = 0; i < netdev->num_tx_queues; i++) {
+		i40e_add_queue_stats(&data, READ_ONCE(vsi->tx_rings[i]));
+		i40e_add_queue_stats(&data, READ_ONCE(vsi->rx_rings[i]));
 	}
 	rcu_read_unlock();
-	if (vsi != pf->vsi[pf->lan_vsi] || pf->hw.partition_id != 1)
-		return;
 
-	if ((pf->lan_veb != I40E_NO_VEB) &&
-	    (pf->flags & I40E_FLAG_VEB_STATS_ENABLED)) {
-		struct i40e_veb *veb = pf->veb[pf->lan_veb];
+	if (vsi->type != I40E_VSI_MAIN || pf->hw.partition_id != 1)
+		goto check_data_pointer;
 
-		for (j = 0; j < I40E_VEB_STATS_LEN; j++) {
-			p = (char *)veb;
-			p += i40e_gstrings_veb_stats[j].stat_offset;
-			data[i++] = (i40e_gstrings_veb_stats[j].sizeof_stat ==
-				     sizeof(u64)) ? *(u64 *)p : *(u32 *)p;
-		}
-		for (j = 0; j < I40E_MAX_TRAFFIC_CLASS; j++) {
-			data[i++] = veb->tc_stats.tc_tx_packets[j];
-			data[i++] = veb->tc_stats.tc_tx_bytes[j];
-			data[i++] = veb->tc_stats.tc_rx_packets[j];
-			data[i++] = veb->tc_stats.tc_rx_bytes[j];
+	veb = i40e_pf_get_main_veb(pf);
+	veb_stats = veb && test_bit(I40E_FLAG_VEB_STATS_ENA, pf->flags);
+
+	if (veb_stats)
+		i40e_update_veb_stats(veb);
+
+	/* If veb stats aren't enabled, pass NULL instead of the veb so that
+	 * we initialize stats to zero and update the data pointer
+	 * intelligently
+	 */
+	i40e_add_ethtool_stats(&data, veb_stats ? veb : NULL,
+			       i40e_gstrings_veb_stats);
+
+	for (i = 0; i < I40E_MAX_TRAFFIC_CLASS; i++)
+		if (veb_stats) {
+			struct i40e_cp_veb_tc_stats veb_tc =
+				i40e_get_veb_tc_stats(&veb->tc_stats, i);
+
+			i40e_add_ethtool_stats(&data, &veb_tc,
+					       i40e_gstrings_veb_tc_stats);
+		} else {
+			i40e_add_ethtool_stats(&data, NULL,
+					       i40e_gstrings_veb_tc_stats);
 		}
+
+	i40e_add_ethtool_stats(&data, pf, i40e_gstrings_stats);
+
+	for (i = 0; i < I40E_MAX_USER_PRIORITY; i++) {
+		struct i40e_pfc_stats pfc = i40e_get_pfc_stats(pf, i);
+
+		i40e_add_ethtool_stats(&data, &pfc, i40e_gstrings_pfc_stats);
 	}
-	for (j = 0; j < I40E_GLOBAL_STATS_LEN; j++) {
-		p = (char *)pf + i40e_gstrings_stats[j].stat_offset;
-		data[i++] = (i40e_gstrings_stats[j].sizeof_stat ==
-			     sizeof(u64)) ? *(u64 *)p : *(u32 *)p;
-	}
-	for (j = 0; j < I40E_MAX_USER_PRIORITY; j++) {
-		data[i++] = pf->stats.priority_xon_tx[j];
-		data[i++] = pf->stats.priority_xoff_tx[j];
-	}
-	for (j = 0; j < I40E_MAX_USER_PRIORITY; j++) {
-		data[i++] = pf->stats.priority_xon_rx[j];
-		data[i++] = pf->stats.priority_xoff_rx[j];
+
+check_data_pointer:
+	WARN_ONCE(data - p != i40e_get_stats_count(netdev),
+		  "ethtool stats count mismatch!");
+}
+
+/**
+ * i40e_get_stat_strings - copy stat strings into supplied buffer
+ * @netdev: the netdev to collect strings for
+ * @data: supplied buffer to copy strings into
+ *
+ * Copy the strings related to stats for this netdev. Expects data to be
+ * pre-allocated with the size reported by i40e_get_stats_count. Note that the
+ * strings must be copied in a static order and the total count must not
+ * change for a given netdev. See i40e_get_stats_count for more details.
+ **/
+static void i40e_get_stat_strings(struct net_device *netdev, u8 *data)
+{
+	struct i40e_netdev_priv *np = netdev_priv(netdev);
+	struct i40e_vsi *vsi = np->vsi;
+	struct i40e_pf *pf = vsi->back;
+	unsigned int i;
+	u8 *p = data;
+
+	i40e_add_stat_strings(&data, i40e_gstrings_net_stats);
+
+	i40e_add_stat_strings(&data, i40e_gstrings_misc_stats);
+
+	for (i = 0; i < netdev->num_tx_queues; i++) {
+		i40e_add_stat_strings(&data, i40e_gstrings_queue_stats,
+				      "tx", i);
+		i40e_add_stat_strings(&data, i40e_gstrings_queue_stats,
+				      "rx", i);
 	}
-	for (j = 0; j < I40E_MAX_USER_PRIORITY; j++)
-		data[i++] = pf->stats.priority_xon_2_xoff[j];
+
+	if (vsi->type != I40E_VSI_MAIN || pf->hw.partition_id != 1)
+		goto check_data_pointer;
+
+	i40e_add_stat_strings(&data, i40e_gstrings_veb_stats);
+
+	for (i = 0; i < I40E_MAX_TRAFFIC_CLASS; i++)
+		i40e_add_stat_strings(&data, i40e_gstrings_veb_tc_stats, i);
+
+	i40e_add_stat_strings(&data, i40e_gstrings_stats);
+
+	for (i = 0; i < I40E_MAX_USER_PRIORITY; i++)
+		i40e_add_stat_strings(&data, i40e_gstrings_pfc_stats, i);
+
+check_data_pointer:
+	WARN_ONCE(data - p != i40e_get_stats_count(netdev) * ETH_GSTRING_LEN,
+		  "stat strings count mismatch!");
 }
 
-static void i40e_get_strings(struct net_device *netdev, u32 stringset,
-			     u8 *data)
+static void i40e_get_priv_flag_strings(struct net_device *netdev, u8 *data)
 {
 	struct i40e_netdev_priv *np = netdev_priv(netdev);
 	struct i40e_vsi *vsi = np->vsi;
 	struct i40e_pf *pf = vsi->back;
-	char *p = (char *)data;
-	int i;
+	unsigned int i;
+	u8 *p = data;
 
+	for (i = 0; i < I40E_PRIV_FLAGS_STR_LEN; i++)
+		ethtool_puts(&p, i40e_gstrings_priv_flags[i].flag_string);
+	if (pf->hw.pf_id != 0)
+		return;
+	for (i = 0; i < I40E_GL_PRIV_FLAGS_STR_LEN; i++)
+		ethtool_puts(&p, i40e_gl_gstrings_priv_flags[i].flag_string);
+}
+
+static void i40e_get_strings(struct net_device *netdev, u32 stringset,
+			     u8 *data)
+{
 	switch (stringset) {
 	case ETH_SS_TEST:
 		memcpy(data, i40e_gstrings_test,
 		       I40E_TEST_LEN * ETH_GSTRING_LEN);
 		break;
 	case ETH_SS_STATS:
-		for (i = 0; i < I40E_NETDEV_STATS_LEN; i++) {
-			snprintf(p, ETH_GSTRING_LEN, "%s",
-				 i40e_gstrings_net_stats[i].stat_string);
-			p += ETH_GSTRING_LEN;
-		}
-		for (i = 0; i < I40E_MISC_STATS_LEN; i++) {
-			snprintf(p, ETH_GSTRING_LEN, "%s",
-				 i40e_gstrings_misc_stats[i].stat_string);
-			p += ETH_GSTRING_LEN;
-		}
-		for (i = 0; i < vsi->num_queue_pairs; i++) {
-			snprintf(p, ETH_GSTRING_LEN, "tx-%d.tx_packets", i);
-			p += ETH_GSTRING_LEN;
-			snprintf(p, ETH_GSTRING_LEN, "tx-%d.tx_bytes", i);
-			p += ETH_GSTRING_LEN;
-			snprintf(p, ETH_GSTRING_LEN, "rx-%d.rx_packets", i);
-			p += ETH_GSTRING_LEN;
-			snprintf(p, ETH_GSTRING_LEN, "rx-%d.rx_bytes", i);
-			p += ETH_GSTRING_LEN;
-		}
-		if (vsi != pf->vsi[pf->lan_vsi] || pf->hw.partition_id != 1)
-			return;
-
-		if ((pf->lan_veb != I40E_NO_VEB) &&
-		    (pf->flags & I40E_FLAG_VEB_STATS_ENABLED)) {
-			for (i = 0; i < I40E_VEB_STATS_LEN; i++) {
-				snprintf(p, ETH_GSTRING_LEN, "veb.%s",
-					i40e_gstrings_veb_stats[i].stat_string);
-				p += ETH_GSTRING_LEN;
-			}
-			for (i = 0; i < I40E_MAX_TRAFFIC_CLASS; i++) {
-				snprintf(p, ETH_GSTRING_LEN,
-					 "veb.tc_%d_tx_packets", i);
-				p += ETH_GSTRING_LEN;
-				snprintf(p, ETH_GSTRING_LEN,
-					 "veb.tc_%d_tx_bytes", i);
-				p += ETH_GSTRING_LEN;
-				snprintf(p, ETH_GSTRING_LEN,
-					 "veb.tc_%d_rx_packets", i);
-				p += ETH_GSTRING_LEN;
-				snprintf(p, ETH_GSTRING_LEN,
-					 "veb.tc_%d_rx_bytes", i);
-				p += ETH_GSTRING_LEN;
-			}
-		}
-		for (i = 0; i < I40E_GLOBAL_STATS_LEN; i++) {
-			snprintf(p, ETH_GSTRING_LEN, "port.%s",
-				 i40e_gstrings_stats[i].stat_string);
-			p += ETH_GSTRING_LEN;
-		}
-		for (i = 0; i < I40E_MAX_USER_PRIORITY; i++) {
-			snprintf(p, ETH_GSTRING_LEN,
-				 "port.tx_priority_%d_xon", i);
-			p += ETH_GSTRING_LEN;
-			snprintf(p, ETH_GSTRING_LEN,
-				 "port.tx_priority_%d_xoff", i);
-			p += ETH_GSTRING_LEN;
-		}
-		for (i = 0; i < I40E_MAX_USER_PRIORITY; i++) {
-			snprintf(p, ETH_GSTRING_LEN,
-				 "port.rx_priority_%d_xon", i);
-			p += ETH_GSTRING_LEN;
-			snprintf(p, ETH_GSTRING_LEN,
-				 "port.rx_priority_%d_xoff", i);
-			p += ETH_GSTRING_LEN;
-		}
-		for (i = 0; i < I40E_MAX_USER_PRIORITY; i++) {
-			snprintf(p, ETH_GSTRING_LEN,
-				 "port.rx_priority_%d_xon_2_xoff", i);
-			p += ETH_GSTRING_LEN;
-		}
-		/* BUG_ON(p - data != I40E_STATS_LEN * ETH_GSTRING_LEN); */
+		i40e_get_stat_strings(netdev, data);
 		break;
 	case ETH_SS_PRIV_FLAGS:
-		for (i = 0; i < I40E_PRIV_FLAGS_STR_LEN; i++) {
-			snprintf(p, ETH_GSTRING_LEN, "%s",
-				 i40e_gstrings_priv_flags[i].flag_string);
-			p += ETH_GSTRING_LEN;
-		}
-		if (pf->hw.pf_id != 0)
-			break;
-		for (i = 0; i < I40E_GL_PRIV_FLAGS_STR_LEN; i++) {
-			snprintf(p, ETH_GSTRING_LEN, "%s",
-				 i40e_gl_gstrings_priv_flags[i].flag_string);
-			p += ETH_GSTRING_LEN;
-		}
+		i40e_get_priv_flag_strings(netdev, data);
 		break;
 	default:
 		break;
@@ -1714,25 +2549,21 @@ static void i40e_get_strings(struct net_device *netdev, u32 stringset,
 }
 
 static int i40e_get_ts_info(struct net_device *dev,
-			    struct ethtool_ts_info *info)
+			    struct kernel_ethtool_ts_info *info)
 {
 	struct i40e_pf *pf = i40e_netdev_to_pf(dev);
 
 	/* only report HW timestamping if PTP is enabled */
-	if (!(pf->flags & I40E_FLAG_PTP))
+	if (!test_bit(I40E_FLAG_PTP_ENA, pf->flags))
 		return ethtool_op_get_ts_info(dev, info);
 
 	info->so_timestamping = SOF_TIMESTAMPING_TX_SOFTWARE |
-				SOF_TIMESTAMPING_RX_SOFTWARE |
-				SOF_TIMESTAMPING_SOFTWARE |
 				SOF_TIMESTAMPING_TX_HARDWARE |
 				SOF_TIMESTAMPING_RX_HARDWARE |
 				SOF_TIMESTAMPING_RAW_HARDWARE;
 
 	if (pf->ptp_clock)
 		info->phc_index = ptp_clock_index(pf->ptp_clock);
-	else
-		info->phc_index = -1;
 
 	info->tx_types = BIT(HWTSTAMP_TX_OFF) | BIT(HWTSTAMP_TX_ON);
 
@@ -1741,7 +2572,7 @@ static int i40e_get_ts_info(struct net_device *dev,
 			   BIT(HWTSTAMP_FILTER_PTP_V2_L2_SYNC) |
 			   BIT(HWTSTAMP_FILTER_PTP_V2_L2_DELAY_REQ);
 
-	if (pf->flags & I40E_FLAG_PTP_L4_CAPABLE)
+	if (test_bit(I40E_HW_CAP_PTP_L4, pf->hw.caps))
 		info->rx_filters |= BIT(HWTSTAMP_FILTER_PTP_V1_L4_SYNC) |
 				    BIT(HWTSTAMP_FILTER_PTP_V1_L4_DELAY_REQ) |
 				    BIT(HWTSTAMP_FILTER_PTP_V2_EVENT) |
@@ -1754,12 +2585,12 @@ static int i40e_get_ts_info(struct net_device *dev,
 	return 0;
 }
 
-static int i40e_link_test(struct net_device *netdev, u64 *data)
+static u64 i40e_link_test(struct net_device *netdev, u64 *data)
 {
 	struct i40e_netdev_priv *np = netdev_priv(netdev);
 	struct i40e_pf *pf = np->vsi->back;
-	i40e_status status;
 	bool link_up = false;
+	int status;
 
 	netif_info(pf, hw, netdev, "link test\n");
 	status = i40e_get_link_status(&pf->hw, &link_up);
@@ -1777,7 +2608,7 @@ static int i40e_link_test(struct net_device *netdev, u64 *data)
 	return *data;
 }
 
-static int i40e_reg_test(struct net_device *netdev, u64 *data)
+static u64 i40e_reg_test(struct net_device *netdev, u64 *data)
 {
 	struct i40e_netdev_priv *np = netdev_priv(netdev);
 	struct i40e_pf *pf = np->vsi->back;
@@ -1788,7 +2619,7 @@ static int i40e_reg_test(struct net_device *netdev, u64 *data)
 	return *data;
 }
 
-static int i40e_eeprom_test(struct net_device *netdev, u64 *data)
+static u64 i40e_eeprom_test(struct net_device *netdev, u64 *data)
 {
 	struct i40e_netdev_priv *np = netdev_priv(netdev);
 	struct i40e_pf *pf = np->vsi->back;
@@ -1802,7 +2633,7 @@ static int i40e_eeprom_test(struct net_device *netdev, u64 *data)
 	return *data;
 }
 
-static int i40e_intr_test(struct net_device *netdev, u64 *data)
+static u64 i40e_intr_test(struct net_device *netdev, u64 *data)
 {
 	struct i40e_netdev_priv *np = netdev_priv(netdev);
 	struct i40e_pf *pf = np->vsi->back;
@@ -1850,15 +2681,16 @@ static void i40e_diag_test(struct net_device *netdev,
 
 		set_bit(__I40E_TESTING, pf->state);
 
+		if (test_bit(__I40E_RESET_RECOVERY_PENDING, pf->state) ||
+		    test_bit(__I40E_RESET_INTR_RECEIVED, pf->state)) {
+			dev_warn(&pf->pdev->dev,
+				 "Cannot start offline testing when PF is in reset state.\n");
+			goto skip_ol_tests;
+		}
+
 		if (i40e_active_vfs(pf) || i40e_active_vmdqs(pf)) {
 			dev_warn(&pf->pdev->dev,
 				 "Please take active VFs and Netqueues offline and restart the adapter before running NIC diagnostics\n");
-			data[I40E_ETH_TEST_REG]		= 1;
-			data[I40E_ETH_TEST_EEPROM]	= 1;
-			data[I40E_ETH_TEST_INTR]	= 1;
-			data[I40E_ETH_TEST_LINK]	= 1;
-			eth_test->flags |= ETH_TEST_FL_FAILED;
-			clear_bit(__I40E_TESTING, pf->state);
 			goto skip_ol_tests;
 		}
 
@@ -1905,9 +2737,26 @@ static void i40e_diag_test(struct net_device *netdev,
 		data[I40E_ETH_TEST_INTR] = 0;
 	}
 
+	netif_info(pf, drv, netdev, "testing finished\n");
+	return;
+
 skip_ol_tests:
+	data[I40E_ETH_TEST_REG]		= 1;
+	data[I40E_ETH_TEST_EEPROM]	= 1;
+	data[I40E_ETH_TEST_INTR]	= 1;
+	data[I40E_ETH_TEST_LINK]	= 1;
+	eth_test->flags |= ETH_TEST_FL_FAILED;
+	clear_bit(__I40E_TESTING, pf->state);
+	netif_info(pf, drv, netdev, "testing failed\n");
+}
 
-	netif_info(pf, drv, netdev, "testing finished\n");
+static void i40e_get_link_ext_stats(struct net_device *netdev,
+				    struct ethtool_link_ext_stats *stats)
+{
+	struct i40e_netdev_priv *np = netdev_priv(netdev);
+	struct i40e_pf *pf = np->vsi->back;
+
+	stats->link_down_events = pf->link_down_events;
 }
 
 static void i40e_get_wol(struct net_device *netdev,
@@ -1948,7 +2797,7 @@ static int i40e_set_wol(struct net_device *netdev, struct ethtool_wolinfo *wol)
 		return -EOPNOTSUPP;
 	}
 
-	if (vsi != pf->vsi[pf->lan_vsi])
+	if (vsi->type != I40E_VSI_MAIN)
 		return -EOPNOTSUPP;
 
 	/* NVM bit on means WoL disabled for the port */
@@ -1957,7 +2806,7 @@ static int i40e_set_wol(struct net_device *netdev, struct ethtool_wolinfo *wol)
 		return -EOPNOTSUPP;
 
 	/* only magic packet is supported */
-	if (wol->wolopts && (wol->wolopts != WAKE_MAGIC))
+	if (wol->wolopts & ~WAKE_MAGIC)
 		return -EOPNOTSUPP;
 
 	/* is this a new value? */
@@ -1973,52 +2822,55 @@ static int i40e_set_phys_id(struct net_device *netdev,
 			    enum ethtool_phys_id_state state)
 {
 	struct i40e_netdev_priv *np = netdev_priv(netdev);
-	i40e_status ret = 0;
 	struct i40e_pf *pf = np->vsi->back;
 	struct i40e_hw *hw = &pf->hw;
 	int blink_freq = 2;
 	u16 temp_status;
+	int ret = 0;
 
 	switch (state) {
 	case ETHTOOL_ID_ACTIVE:
-		if (!(pf->flags & I40E_FLAG_PHY_CONTROLS_LEDS)) {
+		if (!test_bit(I40E_HW_CAP_PHY_CONTROLS_LEDS, pf->hw.caps)) {
 			pf->led_status = i40e_led_get(hw);
 		} else {
-			i40e_aq_set_phy_debug(hw, I40E_PHY_DEBUG_ALL, NULL);
+			if (!test_bit(I40E_HW_CAP_AQ_PHY_ACCESS, hw->caps))
+				i40e_aq_set_phy_debug(hw, I40E_PHY_DEBUG_ALL,
+						      NULL);
 			ret = i40e_led_get_phy(hw, &temp_status,
 					       &pf->phy_led_val);
 			pf->led_status = temp_status;
 		}
 		return blink_freq;
 	case ETHTOOL_ID_ON:
-		if (!(pf->flags & I40E_FLAG_PHY_CONTROLS_LEDS))
+		if (!test_bit(I40E_HW_CAP_PHY_CONTROLS_LEDS, pf->hw.caps))
 			i40e_led_set(hw, 0xf, false);
 		else
 			ret = i40e_led_set_phy(hw, true, pf->led_status, 0);
 		break;
 	case ETHTOOL_ID_OFF:
-		if (!(pf->flags & I40E_FLAG_PHY_CONTROLS_LEDS))
+		if (!test_bit(I40E_HW_CAP_PHY_CONTROLS_LEDS, pf->hw.caps))
 			i40e_led_set(hw, 0x0, false);
 		else
 			ret = i40e_led_set_phy(hw, false, pf->led_status, 0);
 		break;
 	case ETHTOOL_ID_INACTIVE:
-		if (!(pf->flags & I40E_FLAG_PHY_CONTROLS_LEDS)) {
+		if (!test_bit(I40E_HW_CAP_PHY_CONTROLS_LEDS, pf->hw.caps)) {
 			i40e_led_set(hw, pf->led_status, false);
 		} else {
 			ret = i40e_led_set_phy(hw, false, pf->led_status,
 					       (pf->phy_led_val |
 					       I40E_PHY_LED_MODE_ORIG));
-			i40e_aq_set_phy_debug(hw, 0, NULL);
+			if (!test_bit(I40E_HW_CAP_AQ_PHY_ACCESS, hw->caps))
+				i40e_aq_set_phy_debug(hw, 0, NULL);
 		}
 		break;
 	default:
 		break;
 	}
-		if (ret)
-			return -ENOENT;
-		else
-			return 0;
+	if (ret)
+		return -ENOENT;
+	else
+		return 0;
 }
 
 /* NOTE: i40e hardware uses a conversion factor of 2 for Interrupt
@@ -2045,29 +2897,26 @@ static int __i40e_get_coalesce(struct net_device *netdev,
 	struct i40e_vsi *vsi = np->vsi;
 
 	ec->tx_max_coalesced_frames_irq = vsi->work_limit;
-	ec->rx_max_coalesced_frames_irq = vsi->work_limit;
 
-	/* rx and tx usecs has per queue value. If user doesn't specify the queue,
-	 * return queue 0's value to represent.
+	/* rx and tx usecs has per queue value. If user doesn't specify the
+	 * queue, return queue 0's value to represent.
 	 */
-	if (queue < 0) {
+	if (queue < 0)
 		queue = 0;
-	} else if (queue >= vsi->num_queue_pairs) {
+	else if (queue >= vsi->num_queue_pairs)
 		return -EINVAL;
-	}
 
 	rx_ring = vsi->rx_rings[queue];
 	tx_ring = vsi->tx_rings[queue];
 
-	if (ITR_IS_DYNAMIC(rx_ring->rx_itr_setting))
+	if (ITR_IS_DYNAMIC(rx_ring->itr_setting))
 		ec->use_adaptive_rx_coalesce = 1;
 
-	if (ITR_IS_DYNAMIC(tx_ring->tx_itr_setting))
+	if (ITR_IS_DYNAMIC(tx_ring->itr_setting))
 		ec->use_adaptive_tx_coalesce = 1;
 
-	ec->rx_coalesce_usecs = rx_ring->rx_itr_setting & ~I40E_ITR_DYNAMIC;
-	ec->tx_coalesce_usecs = tx_ring->tx_itr_setting & ~I40E_ITR_DYNAMIC;
-
+	ec->rx_coalesce_usecs = rx_ring->itr_setting & ~I40E_ITR_DYNAMIC;
+	ec->tx_coalesce_usecs = tx_ring->itr_setting & ~I40E_ITR_DYNAMIC;
 
 	/* we use the _usecs_high to store/set the interrupt rate limit
 	 * that the hardware supports, that almost but not quite
@@ -2085,13 +2934,17 @@ static int __i40e_get_coalesce(struct net_device *netdev,
  * i40e_get_coalesce - get a netdev's coalesce settings
  * @netdev: the netdev to check
  * @ec: ethtool coalesce data structure
+ * @kernel_coal: ethtool CQE mode setting structure
+ * @extack: extack for reporting error messages
  *
  * Gets the coalesce settings for a particular netdev. Note that if user has
  * modified per-queue settings, this only guarantees to represent queue 0. See
  * __i40e_get_coalesce for more details.
  **/
 static int i40e_get_coalesce(struct net_device *netdev,
-			     struct ethtool_coalesce *ec)
+			     struct ethtool_coalesce *ec,
+			     struct kernel_ethtool_coalesce *kernel_coal,
+			     struct netlink_ext_ack *extack)
 {
 	return __i40e_get_coalesce(netdev, ec, -1);
 }
@@ -2118,42 +2971,44 @@ static int i40e_get_per_queue_coalesce(struct net_device *netdev, u32 queue,
  *
  * Change the ITR settings for a specific queue.
  **/
-
 static void i40e_set_itr_per_queue(struct i40e_vsi *vsi,
 				   struct ethtool_coalesce *ec,
 				   int queue)
 {
+	struct i40e_ring *rx_ring = vsi->rx_rings[queue];
+	struct i40e_ring *tx_ring = vsi->tx_rings[queue];
 	struct i40e_pf *pf = vsi->back;
 	struct i40e_hw *hw = &pf->hw;
 	struct i40e_q_vector *q_vector;
-	u16 vector, intrl;
+	u16 intrl;
 
 	intrl = i40e_intrl_usec_to_reg(vsi->int_rate_limit);
 
-	vsi->rx_rings[queue]->rx_itr_setting = ec->rx_coalesce_usecs;
-	vsi->tx_rings[queue]->tx_itr_setting = ec->tx_coalesce_usecs;
+	rx_ring->itr_setting = ITR_REG_ALIGN(ec->rx_coalesce_usecs);
+	tx_ring->itr_setting = ITR_REG_ALIGN(ec->tx_coalesce_usecs);
 
 	if (ec->use_adaptive_rx_coalesce)
-		vsi->rx_rings[queue]->rx_itr_setting |= I40E_ITR_DYNAMIC;
+		rx_ring->itr_setting |= I40E_ITR_DYNAMIC;
 	else
-		vsi->rx_rings[queue]->rx_itr_setting &= ~I40E_ITR_DYNAMIC;
+		rx_ring->itr_setting &= ~I40E_ITR_DYNAMIC;
 
 	if (ec->use_adaptive_tx_coalesce)
-		vsi->tx_rings[queue]->tx_itr_setting |= I40E_ITR_DYNAMIC;
+		tx_ring->itr_setting |= I40E_ITR_DYNAMIC;
 	else
-		vsi->tx_rings[queue]->tx_itr_setting &= ~I40E_ITR_DYNAMIC;
+		tx_ring->itr_setting &= ~I40E_ITR_DYNAMIC;
+
+	q_vector = rx_ring->q_vector;
+	q_vector->rx.target_itr = ITR_TO_REG(rx_ring->itr_setting);
 
-	q_vector = vsi->rx_rings[queue]->q_vector;
-	q_vector->rx.itr = ITR_TO_REG(vsi->rx_rings[queue]->rx_itr_setting);
-	vector = vsi->base_vector + q_vector->v_idx;
-	wr32(hw, I40E_PFINT_ITRN(I40E_RX_ITR, vector - 1), q_vector->rx.itr);
+	q_vector = tx_ring->q_vector;
+	q_vector->tx.target_itr = ITR_TO_REG(tx_ring->itr_setting);
 
-	q_vector = vsi->tx_rings[queue]->q_vector;
-	q_vector->tx.itr = ITR_TO_REG(vsi->tx_rings[queue]->tx_itr_setting);
-	vector = vsi->base_vector + q_vector->v_idx;
-	wr32(hw, I40E_PFINT_ITRN(I40E_TX_ITR, vector - 1), q_vector->tx.itr);
+	/* The interrupt handler itself will take care of programming
+	 * the Tx and Rx ITR values based on the values we have entered
+	 * into the q_vector, no need to write the values now.
+	 */
 
-	wr32(hw, I40E_PFINT_RATEN(vector - 1), intrl);
+	wr32(hw, I40E_PFINT_RATEN(q_vector->reg_idx), intrl);
 	i40e_flush(hw);
 }
 
@@ -2170,14 +3025,29 @@ static int __i40e_set_coalesce(struct net_device *netdev,
 			       int queue)
 {
 	struct i40e_netdev_priv *np = netdev_priv(netdev);
+	u16 intrl_reg, cur_rx_itr, cur_tx_itr;
 	struct i40e_vsi *vsi = np->vsi;
 	struct i40e_pf *pf = vsi->back;
-	u16 intrl_reg;
 	int i;
 
-	if (ec->tx_max_coalesced_frames_irq || ec->rx_max_coalesced_frames_irq)
+	if (ec->tx_max_coalesced_frames_irq)
 		vsi->work_limit = ec->tx_max_coalesced_frames_irq;
 
+	if (queue < 0) {
+		cur_rx_itr = vsi->rx_rings[0]->itr_setting;
+		cur_tx_itr = vsi->tx_rings[0]->itr_setting;
+	} else if (queue < vsi->num_queue_pairs) {
+		cur_rx_itr = vsi->rx_rings[queue]->itr_setting;
+		cur_tx_itr = vsi->tx_rings[queue]->itr_setting;
+	} else {
+		netif_info(pf, drv, netdev, "Invalid queue value, queue range is 0 - %d\n",
+			   vsi->num_queue_pairs - 1);
+		return -EINVAL;
+	}
+
+	cur_tx_itr &= ~I40E_ITR_DYNAMIC;
+	cur_rx_itr &= ~I40E_ITR_DYNAMIC;
+
 	/* tx_coalesce_usecs_high is ignored, use rx-usecs-high instead */
 	if (ec->tx_coalesce_usecs_high != vsi->int_rate_limit) {
 		netif_info(pf, drv, netdev, "tx-usecs-high is not used, please program rx-usecs-high\n");
@@ -2190,15 +3060,34 @@ static int __i40e_set_coalesce(struct net_device *netdev,
 		return -EINVAL;
 	}
 
-	if (ec->rx_coalesce_usecs == 0) {
-		if (ec->use_adaptive_rx_coalesce)
-			netif_info(pf, drv, netdev, "rx-usecs=0, need to disable adaptive-rx for a complete disable\n");
-	} else if ((ec->rx_coalesce_usecs < (I40E_MIN_ITR << 1)) ||
-		   (ec->rx_coalesce_usecs > (I40E_MAX_ITR << 1))) {
-			netif_info(pf, drv, netdev, "Invalid value, rx-usecs range is 0-8160\n");
-			return -EINVAL;
+	if (ec->rx_coalesce_usecs != cur_rx_itr &&
+	    ec->use_adaptive_rx_coalesce) {
+		netif_info(pf, drv, netdev, "RX interrupt moderation cannot be changed if adaptive-rx is enabled.\n");
+		return -EINVAL;
+	}
+
+	if (ec->rx_coalesce_usecs > I40E_MAX_ITR) {
+		netif_info(pf, drv, netdev, "Invalid value, rx-usecs range is 0-8160\n");
+		return -EINVAL;
+	}
+
+	if (ec->tx_coalesce_usecs != cur_tx_itr &&
+	    ec->use_adaptive_tx_coalesce) {
+		netif_info(pf, drv, netdev, "TX interrupt moderation cannot be changed if adaptive-tx is enabled.\n");
+		return -EINVAL;
+	}
+
+	if (ec->tx_coalesce_usecs > I40E_MAX_ITR) {
+		netif_info(pf, drv, netdev, "Invalid value, tx-usecs range is 0-8160\n");
+		return -EINVAL;
 	}
 
+	if (ec->use_adaptive_rx_coalesce && !cur_rx_itr)
+		ec->rx_coalesce_usecs = I40E_MIN_ITR;
+
+	if (ec->use_adaptive_tx_coalesce && !cur_tx_itr)
+		ec->tx_coalesce_usecs = I40E_MIN_ITR;
+
 	intrl_reg = i40e_intrl_usec_to_reg(ec->rx_coalesce_usecs_high);
 	vsi->int_rate_limit = INTRL_REG_TO_USEC(intrl_reg);
 	if (vsi->int_rate_limit != ec->rx_coalesce_usecs_high) {
@@ -2206,27 +3095,14 @@ static int __i40e_set_coalesce(struct net_device *netdev,
 			   vsi->int_rate_limit);
 	}
 
-	if (ec->tx_coalesce_usecs == 0) {
-		if (ec->use_adaptive_tx_coalesce)
-			netif_info(pf, drv, netdev, "tx-usecs=0, need to disable adaptive-tx for a complete disable\n");
-	} else if ((ec->tx_coalesce_usecs < (I40E_MIN_ITR << 1)) ||
-		   (ec->tx_coalesce_usecs > (I40E_MAX_ITR << 1))) {
-			netif_info(pf, drv, netdev, "Invalid value, tx-usecs range is 0-8160\n");
-			return -EINVAL;
-	}
-
-	/* rx and tx usecs has per queue value. If user doesn't specify the queue,
-	 * apply to all queues.
+	/* rx and tx usecs has per queue value. If user doesn't specify the
+	 * queue, apply to all queues.
 	 */
 	if (queue < 0) {
 		for (i = 0; i < vsi->num_queue_pairs; i++)
 			i40e_set_itr_per_queue(vsi, ec, i);
-	} else if (queue < vsi->num_queue_pairs) {
-		i40e_set_itr_per_queue(vsi, ec, queue);
 	} else {
-		netif_info(pf, drv, netdev, "Invalid queue value, queue range is 0 - %d\n",
-			   vsi->num_queue_pairs - 1);
-		return -EINVAL;
+		i40e_set_itr_per_queue(vsi, ec, queue);
 	}
 
 	return 0;
@@ -2236,11 +3112,15 @@ static int __i40e_set_coalesce(struct net_device *netdev,
  * i40e_set_coalesce - set coalesce settings for every queue on the netdev
  * @netdev: the netdev to change
  * @ec: ethtool coalesce settings
+ * @kernel_coal: ethtool CQE mode setting structure
+ * @extack: extack for reporting error messages
  *
  * This will set each queue to the same coalesce settings.
  **/
 static int i40e_set_coalesce(struct net_device *netdev,
-			     struct ethtool_coalesce *ec)
+			     struct ethtool_coalesce *ec,
+			     struct kernel_ethtool_coalesce *kernel_coal,
+			     struct netlink_ext_ack *extack)
 {
 	return __i40e_set_coalesce(netdev, ec, -1);
 }
@@ -2259,15 +3139,12 @@ static int i40e_set_per_queue_coalesce(struct net_device *netdev, u32 queue,
 	return __i40e_set_coalesce(netdev, ec, queue);
 }
 
-/**
- * i40e_get_rss_hash_opts - Get RSS hash Input Set for each flow type
- * @pf: pointer to the physical function struct
- * @cmd: ethtool rxnfc command
- *
- * Returns Success if the flow is supported, else Invalid Input.
- **/
-static int i40e_get_rss_hash_opts(struct i40e_pf *pf, struct ethtool_rxnfc *cmd)
+static int i40e_get_rxfh_fields(struct net_device *netdev,
+				struct ethtool_rxfh_fields *cmd)
 {
+	struct i40e_netdev_priv *np = netdev_priv(netdev);
+	struct i40e_vsi *vsi = np->vsi;
+	struct i40e_pf *pf = vsi->back;
 	struct i40e_hw *hw = &pf->hw;
 	u8 flow_pctype = 0;
 	u64 i_set = 0;
@@ -2276,16 +3153,16 @@ static int i40e_get_rss_hash_opts(struct i40e_pf *pf, struct ethtool_rxnfc *cmd)
 
 	switch (cmd->flow_type) {
 	case TCP_V4_FLOW:
-		flow_pctype = I40E_FILTER_PCTYPE_NONF_IPV4_TCP;
+		flow_pctype = LIBIE_FILTER_PCTYPE_NONF_IPV4_TCP;
 		break;
 	case UDP_V4_FLOW:
-		flow_pctype = I40E_FILTER_PCTYPE_NONF_IPV4_UDP;
+		flow_pctype = LIBIE_FILTER_PCTYPE_NONF_IPV4_UDP;
 		break;
 	case TCP_V6_FLOW:
-		flow_pctype = I40E_FILTER_PCTYPE_NONF_IPV6_TCP;
+		flow_pctype = LIBIE_FILTER_PCTYPE_NONF_IPV6_TCP;
 		break;
 	case UDP_V6_FLOW:
-		flow_pctype = I40E_FILTER_PCTYPE_NONF_IPV6_UDP;
+		flow_pctype = LIBIE_FILTER_PCTYPE_NONF_IPV6_UDP;
 		break;
 	case SCTP_V4_FLOW:
 	case AH_ESP_V4_FLOW:
@@ -2321,10 +3198,17 @@ static int i40e_get_rss_hash_opts(struct i40e_pf *pf, struct ethtool_rxnfc *cmd)
 
 		if (cmd->flow_type == TCP_V4_FLOW ||
 		    cmd->flow_type == UDP_V4_FLOW) {
-			if (i_set & I40E_L3_SRC_MASK)
-				cmd->data |= RXH_IP_SRC;
-			if (i_set & I40E_L3_DST_MASK)
-				cmd->data |= RXH_IP_DST;
+			if (hw->mac.type == I40E_MAC_X722) {
+				if (i_set & I40E_X722_L3_SRC_MASK)
+					cmd->data |= RXH_IP_SRC;
+				if (i_set & I40E_X722_L3_DST_MASK)
+					cmd->data |= RXH_IP_DST;
+			} else {
+				if (i_set & I40E_L3_SRC_MASK)
+					cmd->data |= RXH_IP_SRC;
+				if (i_set & I40E_L3_DST_MASK)
+					cmd->data |= RXH_IP_DST;
+			}
 		} else if (cmd->flow_type == TCP_V6_FLOW ||
 			  cmd->flow_type == UDP_V6_FLOW) {
 			if (i_set & I40E_L3_V6_SRC_MASK)
@@ -2340,7 +3224,7 @@ static int i40e_get_rss_hash_opts(struct i40e_pf *pf, struct ethtool_rxnfc *cmd)
 /**
  * i40e_check_mask - Check whether a mask field is set
  * @mask: the full mask value
- * @field; mask of the field to check
+ * @field: mask of the field to check
  *
  * If the given mask is fully set, return positive value. If the mask for the
  * field is fully unset, return zero. Otherwise return a negative error code.
@@ -2401,7 +3285,7 @@ static int i40e_parse_rx_flow_user_data(struct ethtool_rx_flow_spec *fsp,
 	} else if (valid) {
 		data->flex_word = value & I40E_USERDEF_FLEX_WORD;
 		data->flex_offset =
-			(value & I40E_USERDEF_FLEX_OFFSET) >> 16;
+			FIELD_GET(I40E_USERDEF_FLEX_OFFSET, value);
 		data->flex_filter = true;
 	}
 
@@ -2411,6 +3295,7 @@ static int i40e_parse_rx_flow_user_data(struct ethtool_rx_flow_spec *fsp,
 /**
  * i40e_fill_rx_flow_user_data - Fill in user-defined data field
  * @fsp: pointer to rx_flow specification
+ * @data: pointer to return userdef data
  *
  * Reads the userdef data structure and properly fills in the user defined
  * fields of the rx_flow_spec.
@@ -2487,6 +3372,7 @@ static int i40e_get_ethtool_fdir_entry(struct i40e_pf *pf,
 	struct i40e_rx_flow_userdef userdef = {0};
 	struct i40e_fdir_filter *rule = NULL;
 	struct hlist_node *node2;
+	struct i40e_vsi *vsi;
 	u64 input_set;
 	u16 index;
 
@@ -2506,26 +3392,55 @@ static int i40e_get_ethtool_fdir_entry(struct i40e_pf *pf,
 		fsp->m_u.usr_ip4_spec.proto = 0;
 	}
 
-	/* Reverse the src and dest notion, since the HW views them from
-	 * Tx perspective where as the user expects it from Rx filter view.
-	 */
-	fsp->h_u.tcp_ip4_spec.psrc = rule->dst_port;
-	fsp->h_u.tcp_ip4_spec.pdst = rule->src_port;
-	fsp->h_u.tcp_ip4_spec.ip4src = rule->dst_ip;
-	fsp->h_u.tcp_ip4_spec.ip4dst = rule->src_ip;
+	if (fsp->flow_type == IPV6_USER_FLOW ||
+	    fsp->flow_type == UDP_V6_FLOW ||
+	    fsp->flow_type == TCP_V6_FLOW ||
+	    fsp->flow_type == SCTP_V6_FLOW) {
+		/* Reverse the src and dest notion, since the HW views them
+		 * from Tx perspective where as the user expects it from
+		 * Rx filter view.
+		 */
+		fsp->h_u.tcp_ip6_spec.psrc = rule->dst_port;
+		fsp->h_u.tcp_ip6_spec.pdst = rule->src_port;
+		memcpy(fsp->h_u.tcp_ip6_spec.ip6dst, rule->src_ip6,
+		       sizeof(__be32) * 4);
+		memcpy(fsp->h_u.tcp_ip6_spec.ip6src, rule->dst_ip6,
+		       sizeof(__be32) * 4);
+	} else {
+		/* Reverse the src and dest notion, since the HW views them
+		 * from Tx perspective where as the user expects it from
+		 * Rx filter view.
+		 */
+		fsp->h_u.tcp_ip4_spec.psrc = rule->dst_port;
+		fsp->h_u.tcp_ip4_spec.pdst = rule->src_port;
+		fsp->h_u.tcp_ip4_spec.ip4src = rule->dst_ip;
+		fsp->h_u.tcp_ip4_spec.ip4dst = rule->src_ip;
+	}
 
 	switch (rule->flow_type) {
 	case SCTP_V4_FLOW:
-		index = I40E_FILTER_PCTYPE_NONF_IPV4_SCTP;
+		index = LIBIE_FILTER_PCTYPE_NONF_IPV4_SCTP;
 		break;
 	case TCP_V4_FLOW:
-		index = I40E_FILTER_PCTYPE_NONF_IPV4_TCP;
+		index = LIBIE_FILTER_PCTYPE_NONF_IPV4_TCP;
 		break;
 	case UDP_V4_FLOW:
-		index = I40E_FILTER_PCTYPE_NONF_IPV4_UDP;
+		index = LIBIE_FILTER_PCTYPE_NONF_IPV4_UDP;
+		break;
+	case SCTP_V6_FLOW:
+		index = LIBIE_FILTER_PCTYPE_NONF_IPV6_SCTP;
+		break;
+	case TCP_V6_FLOW:
+		index = LIBIE_FILTER_PCTYPE_NONF_IPV6_TCP;
+		break;
+	case UDP_V6_FLOW:
+		index = LIBIE_FILTER_PCTYPE_NONF_IPV6_UDP;
 		break;
 	case IP_USER_FLOW:
-		index = I40E_FILTER_PCTYPE_NONF_IPV4_OTHER;
+		index = LIBIE_FILTER_PCTYPE_NONF_IPV4_OTHER;
+		break;
+	case IPV6_USER_FLOW:
+		index = LIBIE_FILTER_PCTYPE_NONF_IPV6_OTHER;
 		break;
 	default:
 		/* If we have stored a filter with a flow type not listed here
@@ -2542,26 +3457,47 @@ static int i40e_get_ethtool_fdir_entry(struct i40e_pf *pf,
 	input_set = i40e_read_fd_input_set(pf, index);
 
 no_input_set:
+	if (input_set & I40E_L3_V6_SRC_MASK) {
+		fsp->m_u.tcp_ip6_spec.ip6src[0] = htonl(0xFFFFFFFF);
+		fsp->m_u.tcp_ip6_spec.ip6src[1] = htonl(0xFFFFFFFF);
+		fsp->m_u.tcp_ip6_spec.ip6src[2] = htonl(0xFFFFFFFF);
+		fsp->m_u.tcp_ip6_spec.ip6src[3] = htonl(0xFFFFFFFF);
+	}
+
+	if (input_set & I40E_L3_V6_DST_MASK) {
+		fsp->m_u.tcp_ip6_spec.ip6dst[0] = htonl(0xFFFFFFFF);
+		fsp->m_u.tcp_ip6_spec.ip6dst[1] = htonl(0xFFFFFFFF);
+		fsp->m_u.tcp_ip6_spec.ip6dst[2] = htonl(0xFFFFFFFF);
+		fsp->m_u.tcp_ip6_spec.ip6dst[3] = htonl(0xFFFFFFFF);
+	}
+
 	if (input_set & I40E_L3_SRC_MASK)
-		fsp->m_u.tcp_ip4_spec.ip4src = htonl(0xFFFF);
+		fsp->m_u.tcp_ip4_spec.ip4src = htonl(0xFFFFFFFF);
 
 	if (input_set & I40E_L3_DST_MASK)
-		fsp->m_u.tcp_ip4_spec.ip4dst = htonl(0xFFFF);
+		fsp->m_u.tcp_ip4_spec.ip4dst = htonl(0xFFFFFFFF);
 
 	if (input_set & I40E_L4_SRC_MASK)
-		fsp->m_u.tcp_ip4_spec.psrc = htons(0xFFFFFFFF);
+		fsp->m_u.tcp_ip4_spec.psrc = htons(0xFFFF);
 
 	if (input_set & I40E_L4_DST_MASK)
-		fsp->m_u.tcp_ip4_spec.pdst = htons(0xFFFFFFFF);
+		fsp->m_u.tcp_ip4_spec.pdst = htons(0xFFFF);
 
 	if (rule->dest_ctl == I40E_FILTER_PROGRAM_DESC_DEST_DROP_PACKET)
 		fsp->ring_cookie = RX_CLS_FLOW_DISC;
 	else
 		fsp->ring_cookie = rule->q_index;
 
-	if (rule->dest_vsi != pf->vsi[pf->lan_vsi]->id) {
-		struct i40e_vsi *vsi;
+	if (rule->vlan_tag) {
+		fsp->h_ext.vlan_etype = rule->vlan_etype;
+		fsp->m_ext.vlan_etype = htons(0xFFFF);
+		fsp->h_ext.vlan_tci = rule->vlan_tag;
+		fsp->m_ext.vlan_tci = htons(0xFFFF);
+		fsp->flow_type |= FLOW_EXT;
+	}
 
+	vsi = i40e_pf_get_main_vsi(pf);
+	if (rule->dest_vsi != vsi->id) {
 		vsi = i40e_find_vsi_from_id(pf, rule->dest_vsi);
 		if (vsi && vsi->type == I40E_VSI_SRIOV) {
 			/* VFs are zero-indexed by the driver, but ethtool
@@ -2589,6 +3525,7 @@ no_input_set:
  * i40e_get_rxnfc - command to get RX flow classification rules
  * @netdev: network interface device structure
  * @cmd: ethtool rxnfc command
+ * @rule_locs: pointer to store rule data
  *
  * Returns Success if the command is supported.
  **/
@@ -2602,12 +3539,9 @@ static int i40e_get_rxnfc(struct net_device *netdev, struct ethtool_rxnfc *cmd,
 
 	switch (cmd->cmd) {
 	case ETHTOOL_GRXRINGS:
-		cmd->data = vsi->num_queue_pairs;
+		cmd->data = vsi->rss_size;
 		ret = 0;
 		break;
-	case ETHTOOL_GRXFH:
-		ret = i40e_get_rss_hash_opts(pf, cmd);
-		break;
 	case ETHTOOL_GRXCLSRLCNT:
 		cmd->rule_cnt = pf->fdir_pf_active_filters;
 		/* report total rule count */
@@ -2629,12 +3563,15 @@ static int i40e_get_rxnfc(struct net_device *netdev, struct ethtool_rxnfc *cmd,
 
 /**
  * i40e_get_rss_hash_bits - Read RSS Hash bits from register
+ * @hw: hw structure
  * @nfc: pointer to user request
- * @i_setc bits currently set
+ * @i_setc: bits currently set
  *
  * Returns value of bits to be set per user request
  **/
-static u64 i40e_get_rss_hash_bits(struct ethtool_rxnfc *nfc, u64 i_setc)
+static u64 i40e_get_rss_hash_bits(struct i40e_hw *hw,
+				  const struct ethtool_rxfh_fields *nfc,
+				  u64 i_setc)
 {
 	u64 i_set = i_setc;
 	u64 src_l3 = 0, dst_l3 = 0;
@@ -2653,8 +3590,13 @@ static u64 i40e_get_rss_hash_bits(struct ethtool_rxnfc *nfc, u64 i_setc)
 		dst_l3 = I40E_L3_V6_DST_MASK;
 	} else if (nfc->flow_type == TCP_V4_FLOW ||
 		  nfc->flow_type == UDP_V4_FLOW) {
-		src_l3 = I40E_L3_SRC_MASK;
-		dst_l3 = I40E_L3_DST_MASK;
+		if (hw->mac.type == I40E_MAC_X722) {
+			src_l3 = I40E_X722_L3_SRC_MASK;
+			dst_l3 = I40E_X722_L3_DST_MASK;
+		} else {
+			src_l3 = I40E_L3_SRC_MASK;
+			dst_l3 = I40E_L3_DST_MASK;
+		}
 	} else {
 		/* Any other flow type are not supported here */
 		return i_set;
@@ -2672,21 +3614,28 @@ static u64 i40e_get_rss_hash_bits(struct ethtool_rxnfc *nfc, u64 i_setc)
 	return i_set;
 }
 
-/**
- * i40e_set_rss_hash_opt - Enable/Disable flow types for RSS hash
- * @pf: pointer to the physical function struct
- * @cmd: ethtool rxnfc command
- *
- * Returns Success if the flow input set is supported.
- **/
-static int i40e_set_rss_hash_opt(struct i40e_pf *pf, struct ethtool_rxnfc *nfc)
+#define FLOW_PCTYPES_SIZE 64
+static int i40e_set_rxfh_fields(struct net_device *netdev,
+				const struct ethtool_rxfh_fields *nfc,
+				struct netlink_ext_ack *extack)
 {
+	struct i40e_netdev_priv *np = netdev_priv(netdev);
+	struct i40e_vsi *vsi = np->vsi;
+	struct i40e_pf *pf = vsi->back;
 	struct i40e_hw *hw = &pf->hw;
 	u64 hena = (u64)i40e_read_rx_ctl(hw, I40E_PFQF_HENA(0)) |
 		   ((u64)i40e_read_rx_ctl(hw, I40E_PFQF_HENA(1)) << 32);
-	u8 flow_pctype = 0;
+	DECLARE_BITMAP(flow_pctypes, FLOW_PCTYPES_SIZE);
 	u64 i_set, i_setc;
 
+	bitmap_zero(flow_pctypes, FLOW_PCTYPES_SIZE);
+
+	if (test_bit(I40E_FLAG_MFP_ENA, pf->flags)) {
+		dev_err(&pf->pdev->dev,
+			"Change of RSS hash input set is not supported when MFP mode is enabled\n");
+		return -EOPNOTSUPP;
+	}
+
 	/* RSS does not support anything other than hashing
 	 * to queues on src and dst IPs and ports
 	 */
@@ -2696,37 +3645,40 @@ static int i40e_set_rss_hash_opt(struct i40e_pf *pf, struct ethtool_rxnfc *nfc)
 
 	switch (nfc->flow_type) {
 	case TCP_V4_FLOW:
-		flow_pctype = I40E_FILTER_PCTYPE_NONF_IPV4_TCP;
-		if (pf->flags & I40E_FLAG_MULTIPLE_TCP_UDP_RSS_PCTYPE)
-			hena |=
-			  BIT_ULL(I40E_FILTER_PCTYPE_NONF_IPV4_TCP_SYN_NO_ACK);
+		set_bit(LIBIE_FILTER_PCTYPE_NONF_IPV4_TCP, flow_pctypes);
+		if (test_bit(I40E_HW_CAP_MULTI_TCP_UDP_RSS_PCTYPE,
+			     pf->hw.caps))
+			set_bit(LIBIE_FILTER_PCTYPE_NONF_IPV4_TCP_SYN_NO_ACK,
+				flow_pctypes);
 		break;
 	case TCP_V6_FLOW:
-		flow_pctype = I40E_FILTER_PCTYPE_NONF_IPV6_TCP;
-		if (pf->flags & I40E_FLAG_MULTIPLE_TCP_UDP_RSS_PCTYPE)
-			hena |=
-			  BIT_ULL(I40E_FILTER_PCTYPE_NONF_IPV4_TCP_SYN_NO_ACK);
-		if (pf->flags & I40E_FLAG_MULTIPLE_TCP_UDP_RSS_PCTYPE)
-			hena |=
-			  BIT_ULL(I40E_FILTER_PCTYPE_NONF_IPV6_TCP_SYN_NO_ACK);
+		set_bit(LIBIE_FILTER_PCTYPE_NONF_IPV6_TCP, flow_pctypes);
+		if (test_bit(I40E_HW_CAP_MULTI_TCP_UDP_RSS_PCTYPE,
+			     pf->hw.caps))
+			set_bit(LIBIE_FILTER_PCTYPE_NONF_IPV6_TCP_SYN_NO_ACK,
+				flow_pctypes);
 		break;
 	case UDP_V4_FLOW:
-		flow_pctype = I40E_FILTER_PCTYPE_NONF_IPV4_UDP;
-		if (pf->flags & I40E_FLAG_MULTIPLE_TCP_UDP_RSS_PCTYPE)
-			hena |=
-			  BIT_ULL(I40E_FILTER_PCTYPE_NONF_UNICAST_IPV4_UDP) |
-			  BIT_ULL(I40E_FILTER_PCTYPE_NONF_MULTICAST_IPV4_UDP);
-
-		hena |= BIT_ULL(I40E_FILTER_PCTYPE_FRAG_IPV4);
+		set_bit(LIBIE_FILTER_PCTYPE_NONF_IPV4_UDP, flow_pctypes);
+		if (test_bit(I40E_HW_CAP_MULTI_TCP_UDP_RSS_PCTYPE,
+			     pf->hw.caps)) {
+			set_bit(LIBIE_FILTER_PCTYPE_NONF_UNICAST_IPV4_UDP,
+				flow_pctypes);
+			set_bit(LIBIE_FILTER_PCTYPE_NONF_MULTICAST_IPV4_UDP,
+				flow_pctypes);
+		}
+		hena |= BIT_ULL(LIBIE_FILTER_PCTYPE_FRAG_IPV4);
 		break;
 	case UDP_V6_FLOW:
-		flow_pctype = I40E_FILTER_PCTYPE_NONF_IPV6_UDP;
-		if (pf->flags & I40E_FLAG_MULTIPLE_TCP_UDP_RSS_PCTYPE)
-			hena |=
-			  BIT_ULL(I40E_FILTER_PCTYPE_NONF_UNICAST_IPV6_UDP) |
-			  BIT_ULL(I40E_FILTER_PCTYPE_NONF_MULTICAST_IPV6_UDP);
-
-		hena |= BIT_ULL(I40E_FILTER_PCTYPE_FRAG_IPV6);
+		set_bit(LIBIE_FILTER_PCTYPE_NONF_IPV6_UDP, flow_pctypes);
+		if (test_bit(I40E_HW_CAP_MULTI_TCP_UDP_RSS_PCTYPE,
+			     pf->hw.caps)) {
+			set_bit(LIBIE_FILTER_PCTYPE_NONF_UNICAST_IPV6_UDP,
+				flow_pctypes);
+			set_bit(LIBIE_FILTER_PCTYPE_NONF_MULTICAST_IPV6_UDP,
+				flow_pctypes);
+		}
+		hena |= BIT_ULL(LIBIE_FILTER_PCTYPE_FRAG_IPV6);
 		break;
 	case AH_ESP_V4_FLOW:
 	case AH_V4_FLOW:
@@ -2735,7 +3687,7 @@ static int i40e_set_rss_hash_opt(struct i40e_pf *pf, struct ethtool_rxnfc *nfc)
 		if ((nfc->data & RXH_L4_B_0_1) ||
 		    (nfc->data & RXH_L4_B_2_3))
 			return -EINVAL;
-		hena |= BIT_ULL(I40E_FILTER_PCTYPE_NONF_IPV4_OTHER);
+		hena |= BIT_ULL(LIBIE_FILTER_PCTYPE_NONF_IPV4_OTHER);
 		break;
 	case AH_ESP_V6_FLOW:
 	case AH_V6_FLOW:
@@ -2744,31 +3696,34 @@ static int i40e_set_rss_hash_opt(struct i40e_pf *pf, struct ethtool_rxnfc *nfc)
 		if ((nfc->data & RXH_L4_B_0_1) ||
 		    (nfc->data & RXH_L4_B_2_3))
 			return -EINVAL;
-		hena |= BIT_ULL(I40E_FILTER_PCTYPE_NONF_IPV6_OTHER);
+		hena |= BIT_ULL(LIBIE_FILTER_PCTYPE_NONF_IPV6_OTHER);
 		break;
 	case IPV4_FLOW:
-		hena |= BIT_ULL(I40E_FILTER_PCTYPE_NONF_IPV4_OTHER) |
-			BIT_ULL(I40E_FILTER_PCTYPE_FRAG_IPV4);
+		hena |= BIT_ULL(LIBIE_FILTER_PCTYPE_NONF_IPV4_OTHER) |
+			BIT_ULL(LIBIE_FILTER_PCTYPE_FRAG_IPV4);
 		break;
 	case IPV6_FLOW:
-		hena |= BIT_ULL(I40E_FILTER_PCTYPE_NONF_IPV6_OTHER) |
-			BIT_ULL(I40E_FILTER_PCTYPE_FRAG_IPV6);
+		hena |= BIT_ULL(LIBIE_FILTER_PCTYPE_NONF_IPV6_OTHER) |
+			BIT_ULL(LIBIE_FILTER_PCTYPE_FRAG_IPV6);
 		break;
 	default:
 		return -EINVAL;
 	}
 
-	if (flow_pctype) {
-		i_setc = (u64)i40e_read_rx_ctl(hw, I40E_GLQF_HASH_INSET(0,
-					       flow_pctype)) |
-			((u64)i40e_read_rx_ctl(hw, I40E_GLQF_HASH_INSET(1,
-					       flow_pctype)) << 32);
-		i_set = i40e_get_rss_hash_bits(nfc, i_setc);
-		i40e_write_rx_ctl(hw, I40E_GLQF_HASH_INSET(0, flow_pctype),
-				  (u32)i_set);
-		i40e_write_rx_ctl(hw, I40E_GLQF_HASH_INSET(1, flow_pctype),
-				  (u32)(i_set >> 32));
-		hena |= BIT_ULL(flow_pctype);
+	if (bitmap_weight(flow_pctypes, FLOW_PCTYPES_SIZE)) {
+		u8 flow_id;
+
+		for_each_set_bit(flow_id, flow_pctypes, FLOW_PCTYPES_SIZE) {
+			i_setc = (u64)i40e_read_rx_ctl(hw, I40E_GLQF_HASH_INSET(0, flow_id)) |
+				 ((u64)i40e_read_rx_ctl(hw, I40E_GLQF_HASH_INSET(1, flow_id)) << 32);
+			i_set = i40e_get_rss_hash_bits(&pf->hw, nfc, i_setc);
+
+			i40e_write_rx_ctl(hw, I40E_GLQF_HASH_INSET(0, flow_id),
+					  (u32)i_set);
+			i40e_write_rx_ctl(hw, I40E_GLQF_HASH_INSET(1, flow_id),
+					  (u32)(i_set >> 32));
+			hena |= BIT_ULL(flow_id);
+		}
 	}
 
 	i40e_write_rx_ctl(hw, I40E_PFQF_HENA(0), (u32)hena);
@@ -3068,7 +4023,7 @@ static int i40e_add_flex_offset(struct list_head *flex_pit_list,
  * __i40e_reprogram_flex_pit - Re-program specific FLX_PIT table
  * @pf: Pointer to the PF structure
  * @flex_pit_list: list of flexible src offsets in use
- * #flex_pit_start: index to first entry for this section of the table
+ * @flex_pit_start: index to first entry for this section of the table
  *
  * In order to handle flexible data, the hardware uses a table of values
  * called the FLX_PIT table. This table is used to indicate which sections of
@@ -3182,7 +4137,7 @@ static void i40e_reprogram_flex_pit(struct i40e_pf *pf)
 
 /**
  * i40e_flow_str - Converts a flow_type into a human readable string
- * @flow_type: the flow type from a flow specification
+ * @fsp: the flow specification
  *
  * Currently only flow types we support are included here, and the string
  * value attempts to match what ethtool would use to configure this flow type.
@@ -3198,6 +4153,14 @@ static const char *i40e_flow_str(struct ethtool_rx_flow_spec *fsp)
 		return "sctp4";
 	case IP_USER_FLOW:
 		return "ip4";
+	case TCP_V6_FLOW:
+		return "tcp6";
+	case UDP_V6_FLOW:
+		return "udp6";
+	case SCTP_V6_FLOW:
+		return "sctp6";
+	case IPV6_USER_FLOW:
+		return "ip6";
 	default:
 		return "unknown";
 	}
@@ -3333,9 +4296,14 @@ static int i40e_check_fdir_input_set(struct i40e_vsi *vsi,
 				     struct ethtool_rx_flow_spec *fsp,
 				     struct i40e_rx_flow_userdef *userdef)
 {
-	struct i40e_pf *pf = vsi->back;
+	static const __be32 ipv6_full_mask[4] = {cpu_to_be32(0xffffffff),
+		cpu_to_be32(0xffffffff), cpu_to_be32(0xffffffff),
+		cpu_to_be32(0xffffffff)};
+	struct ethtool_tcpip6_spec *tcp_ip6_spec;
+	struct ethtool_usrip6_spec *usr_ip6_spec;
 	struct ethtool_tcpip4_spec *tcp_ip4_spec;
 	struct ethtool_usrip4_spec *usr_ip4_spec;
+	struct i40e_pf *pf = vsi->back;
 	u64 current_mask, new_mask;
 	bool new_flex_offset = false;
 	bool flex_l3 = false;
@@ -3346,22 +4314,39 @@ static int i40e_check_fdir_input_set(struct i40e_vsi *vsi,
 
 	switch (fsp->flow_type & ~FLOW_EXT) {
 	case SCTP_V4_FLOW:
-		index = I40E_FILTER_PCTYPE_NONF_IPV4_SCTP;
+		index = LIBIE_FILTER_PCTYPE_NONF_IPV4_SCTP;
 		fdir_filter_count = &pf->fd_sctp4_filter_cnt;
 		break;
 	case TCP_V4_FLOW:
-		index = I40E_FILTER_PCTYPE_NONF_IPV4_TCP;
+		index = LIBIE_FILTER_PCTYPE_NONF_IPV4_TCP;
 		fdir_filter_count = &pf->fd_tcp4_filter_cnt;
 		break;
 	case UDP_V4_FLOW:
-		index = I40E_FILTER_PCTYPE_NONF_IPV4_UDP;
+		index = LIBIE_FILTER_PCTYPE_NONF_IPV4_UDP;
 		fdir_filter_count = &pf->fd_udp4_filter_cnt;
 		break;
+	case SCTP_V6_FLOW:
+		index = LIBIE_FILTER_PCTYPE_NONF_IPV6_SCTP;
+		fdir_filter_count = &pf->fd_sctp6_filter_cnt;
+		break;
+	case TCP_V6_FLOW:
+		index = LIBIE_FILTER_PCTYPE_NONF_IPV6_TCP;
+		fdir_filter_count = &pf->fd_tcp6_filter_cnt;
+		break;
+	case UDP_V6_FLOW:
+		index = LIBIE_FILTER_PCTYPE_NONF_IPV6_UDP;
+		fdir_filter_count = &pf->fd_udp6_filter_cnt;
+		break;
 	case IP_USER_FLOW:
-		index = I40E_FILTER_PCTYPE_NONF_IPV4_OTHER;
+		index = LIBIE_FILTER_PCTYPE_NONF_IPV4_OTHER;
 		fdir_filter_count = &pf->fd_ip4_filter_cnt;
 		flex_l3 = true;
 		break;
+	case IPV6_USER_FLOW:
+		index = LIBIE_FILTER_PCTYPE_NONF_IPV6_OTHER;
+		fdir_filter_count = &pf->fd_ip6_filter_cnt;
+		flex_l3 = true;
+		break;
 	default:
 		return -EOPNOTSUPP;
 	}
@@ -3382,7 +4367,7 @@ static int i40e_check_fdir_input_set(struct i40e_vsi *vsi,
 	switch (fsp->flow_type & ~FLOW_EXT) {
 	case SCTP_V4_FLOW:
 		new_mask &= ~I40E_VERIFY_TAG_MASK;
-		/* Fall through */
+		fallthrough;
 	case TCP_V4_FLOW:
 	case UDP_V4_FLOW:
 		tcp_ip4_spec = &fsp->m_u.tcp_ip4_spec;
@@ -3424,6 +4409,53 @@ static int i40e_check_fdir_input_set(struct i40e_vsi *vsi,
 			return -EOPNOTSUPP;
 
 		break;
+	case SCTP_V6_FLOW:
+		new_mask &= ~I40E_VERIFY_TAG_MASK;
+		fallthrough;
+	case TCP_V6_FLOW:
+	case UDP_V6_FLOW:
+		tcp_ip6_spec = &fsp->m_u.tcp_ip6_spec;
+
+		/* Check if user provided IPv6 source address. */
+		if (ipv6_addr_equal((struct in6_addr *)&tcp_ip6_spec->ip6src,
+				    (struct in6_addr *)&ipv6_full_mask))
+			new_mask |= I40E_L3_V6_SRC_MASK;
+		else if (ipv6_addr_any((struct in6_addr *)
+				       &tcp_ip6_spec->ip6src))
+			new_mask &= ~I40E_L3_V6_SRC_MASK;
+		else
+			return -EOPNOTSUPP;
+
+		/* Check if user provided destination address. */
+		if (ipv6_addr_equal((struct in6_addr *)&tcp_ip6_spec->ip6dst,
+				    (struct in6_addr *)&ipv6_full_mask))
+			new_mask |= I40E_L3_V6_DST_MASK;
+		else if (ipv6_addr_any((struct in6_addr *)
+				       &tcp_ip6_spec->ip6dst))
+			new_mask &= ~I40E_L3_V6_DST_MASK;
+		else
+			return -EOPNOTSUPP;
+
+		/* L4 source port */
+		if (tcp_ip6_spec->psrc == htons(0xFFFF))
+			new_mask |= I40E_L4_SRC_MASK;
+		else if (!tcp_ip6_spec->psrc)
+			new_mask &= ~I40E_L4_SRC_MASK;
+		else
+			return -EOPNOTSUPP;
+
+		/* L4 destination port */
+		if (tcp_ip6_spec->pdst == htons(0xFFFF))
+			new_mask |= I40E_L4_DST_MASK;
+		else if (!tcp_ip6_spec->pdst)
+			new_mask &= ~I40E_L4_DST_MASK;
+		else
+			return -EOPNOTSUPP;
+
+		/* Filtering on Traffic Classes is not supported. */
+		if (tcp_ip6_spec->tclass)
+			return -EOPNOTSUPP;
+		break;
 	case IP_USER_FLOW:
 		usr_ip4_spec = &fsp->m_u.usr_ip4_spec;
 
@@ -3444,11 +4476,7 @@ static int i40e_check_fdir_input_set(struct i40e_vsi *vsi,
 			return -EOPNOTSUPP;
 
 		/* First 4 bytes of L4 header */
-		if (usr_ip4_spec->l4_4_bytes == htonl(0xFFFFFFFF))
-			new_mask |= I40E_L4_SRC_MASK | I40E_L4_DST_MASK;
-		else if (!usr_ip4_spec->l4_4_bytes)
-			new_mask &= ~(I40E_L4_SRC_MASK | I40E_L4_DST_MASK);
-		else
+		if (usr_ip4_spec->l4_4_bytes)
 			return -EOPNOTSUPP;
 
 		/* Filtering on Type of Service is not supported. */
@@ -3464,10 +4492,58 @@ static int i40e_check_fdir_input_set(struct i40e_vsi *vsi,
 			return -EINVAL;
 
 		break;
+	case IPV6_USER_FLOW:
+		usr_ip6_spec = &fsp->m_u.usr_ip6_spec;
+
+		/* Check if user provided IPv6 source address. */
+		if (ipv6_addr_equal((struct in6_addr *)&usr_ip6_spec->ip6src,
+				    (struct in6_addr *)&ipv6_full_mask))
+			new_mask |= I40E_L3_V6_SRC_MASK;
+		else if (ipv6_addr_any((struct in6_addr *)
+				       &usr_ip6_spec->ip6src))
+			new_mask &= ~I40E_L3_V6_SRC_MASK;
+		else
+			return -EOPNOTSUPP;
+
+		/* Check if user provided destination address. */
+		if (ipv6_addr_equal((struct in6_addr *)&usr_ip6_spec->ip6dst,
+				    (struct in6_addr *)&ipv6_full_mask))
+			new_mask |= I40E_L3_V6_DST_MASK;
+		else if (ipv6_addr_any((struct in6_addr *)
+				       &usr_ip6_spec->ip6dst))
+			new_mask &= ~I40E_L3_V6_DST_MASK;
+		else
+			return -EOPNOTSUPP;
+
+		if (usr_ip6_spec->l4_4_bytes)
+			return -EOPNOTSUPP;
+
+		/* Filtering on Traffic class is not supported. */
+		if (usr_ip6_spec->tclass)
+			return -EOPNOTSUPP;
+
+		/* Filtering on L4 protocol is not supported */
+		if (usr_ip6_spec->l4_proto)
+			return -EINVAL;
+
+		break;
 	default:
 		return -EOPNOTSUPP;
 	}
 
+	if (fsp->flow_type & FLOW_EXT) {
+		/* Allow only 802.1Q and no etype defined, as
+		 * later it's modified to 0x8100
+		 */
+		if (fsp->h_ext.vlan_etype != htons(ETH_P_8021Q) &&
+		    fsp->h_ext.vlan_etype != 0)
+			return -EOPNOTSUPP;
+		if (fsp->m_ext.vlan_tci == htons(0xFFFF))
+			new_mask |= I40E_VLAN_SRC_MASK;
+		else
+			new_mask &= ~I40E_VLAN_SRC_MASK;
+	}
+
 	/* First, clear all flexible filter entries */
 	new_mask &= ~I40E_FLEX_INPUT_MASK;
 
@@ -3574,7 +4650,7 @@ static int i40e_check_fdir_input_set(struct i40e_vsi *vsi,
 	 * main port cannot change them when in MFP mode as this would impact
 	 * any filters on the other ports.
 	 */
-	if (pf->flags & I40E_FLAG_MFP_ENABLED) {
+	if (test_bit(I40E_FLAG_MFP_ENA, pf->flags)) {
 		netif_err(pf, drv, vsi->netdev, "Cannot change Flow Director input sets while MFP is enabled\n");
 		return -EOPNOTSUPP;
 	}
@@ -3597,6 +4673,16 @@ static int i40e_check_fdir_input_set(struct i40e_vsi *vsi,
 
 	i40e_write_fd_input_set(pf, index, new_mask);
 
+	/* IP_USER_FLOW filters match both IPv4/Other and IPv4/Fragmented
+	 * frames. If we're programming the input set for IPv4/Other, we also
+	 * need to program the IPv4/Fragmented input set. Since we don't have
+	 * separate support, we'll always assume and enforce that the two flow
+	 * types must have matching input sets.
+	 */
+	if (index == LIBIE_FILTER_PCTYPE_NONF_IPV4_OTHER)
+		i40e_write_fd_input_set(pf, LIBIE_FILTER_PCTYPE_FRAG_IPV4,
+					new_mask);
+
 	/* Add the new offset and update table, if necessary */
 	if (new_flex_offset) {
 		err = i40e_add_flex_offset(&pf->l4_flex_pit_list, src_offset,
@@ -3619,6 +4705,89 @@ static int i40e_check_fdir_input_set(struct i40e_vsi *vsi,
 }
 
 /**
+ * i40e_match_fdir_filter - Return true of two filters match
+ * @a: pointer to filter struct
+ * @b: pointer to filter struct
+ *
+ * Returns true if the two filters match exactly the same criteria. I.e. they
+ * match the same flow type and have the same parameters. We don't need to
+ * check any input-set since all filters of the same flow type must use the
+ * same input set.
+ **/
+static bool i40e_match_fdir_filter(struct i40e_fdir_filter *a,
+				   struct i40e_fdir_filter *b)
+{
+	/* The filters do not much if any of these criteria differ. */
+	if (a->dst_ip != b->dst_ip ||
+	    a->src_ip != b->src_ip ||
+	    a->dst_port != b->dst_port ||
+	    a->src_port != b->src_port ||
+	    a->flow_type != b->flow_type ||
+	    a->ipl4_proto != b->ipl4_proto ||
+	    a->vlan_tag != b->vlan_tag ||
+	    a->vlan_etype != b->vlan_etype)
+		return false;
+
+	return true;
+}
+
+/**
+ * i40e_disallow_matching_filters - Check that new filters differ
+ * @vsi: pointer to the targeted VSI
+ * @input: new filter to check
+ *
+ * Due to hardware limitations, it is not possible for two filters that match
+ * similar criteria to be programmed at the same time. This is true for a few
+ * reasons:
+ *
+ * (a) all filters matching a particular flow type must use the same input
+ * set, that is they must match the same criteria.
+ * (b) different flow types will never match the same packet, as the flow type
+ * is decided by hardware before checking which rules apply.
+ * (c) hardware has no way to distinguish which order filters apply in.
+ *
+ * Due to this, we can't really support using the location data to order
+ * filters in the hardware parsing. It is technically possible for the user to
+ * request two filters matching the same criteria but which select different
+ * queues. In this case, rather than keep both filters in the list, we reject
+ * the 2nd filter when the user requests adding it.
+ *
+ * This avoids needing to track location for programming the filter to
+ * hardware, and ensures that we avoid some strange scenarios involving
+ * deleting filters which match the same criteria.
+ **/
+static int i40e_disallow_matching_filters(struct i40e_vsi *vsi,
+					  struct i40e_fdir_filter *input)
+{
+	struct i40e_pf *pf = vsi->back;
+	struct i40e_fdir_filter *rule;
+	struct hlist_node *node2;
+
+	/* Loop through every filter, and check that it doesn't match */
+	hlist_for_each_entry_safe(rule, node2,
+				  &pf->fdir_filter_list, fdir_node) {
+		/* Don't check the filters match if they share the same fd_id,
+		 * since the new filter is actually just updating the target
+		 * of the old filter.
+		 */
+		if (rule->fd_id == input->fd_id)
+			continue;
+
+		/* If any filters match, then print a warning message to the
+		 * kernel message buffer and bail out.
+		 */
+		if (i40e_match_fdir_filter(rule, input)) {
+			dev_warn(&pf->pdev->dev,
+				 "Existing user defined filter %d already matches this flow.\n",
+				 rule->fd_id);
+			return -EINVAL;
+		}
+	}
+
+	return 0;
+}
+
+/**
  * i40e_add_fdir_ethtool - Add/Remove Flow Director filters
  * @vsi: pointer to the targeted VSI
  * @cmd: command to get or set RX flow classification rules
@@ -3641,10 +4810,10 @@ static int i40e_add_fdir_ethtool(struct i40e_vsi *vsi,
 		return -EINVAL;
 	pf = vsi->back;
 
-	if (!(pf->flags & I40E_FLAG_FD_SB_ENABLED))
+	if (!test_bit(I40E_FLAG_FD_SB_ENA, pf->flags))
 		return -EOPNOTSUPP;
 
-	if (pf->flags & I40E_FLAG_FD_SB_AUTO_DISABLED)
+	if (test_bit(__I40E_FD_SB_AUTO_DISABLED, pf->state))
 		return -ENOSPC;
 
 	if (test_bit(__I40E_RESET_RECOVERY_PENDING, pf->state) ||
@@ -3714,15 +4883,38 @@ static int i40e_add_fdir_ethtool(struct i40e_vsi *vsi,
 	input->dst_ip = fsp->h_u.tcp_ip4_spec.ip4src;
 	input->src_ip = fsp->h_u.tcp_ip4_spec.ip4dst;
 	input->flow_type = fsp->flow_type & ~FLOW_EXT;
-	input->ip4_proto = fsp->h_u.usr_ip4_spec.proto;
 
-	/* Reverse the src and dest notion, since the HW expects them to be from
-	 * Tx perspective where as the input from user is from Rx filter view.
-	 */
-	input->dst_port = fsp->h_u.tcp_ip4_spec.psrc;
-	input->src_port = fsp->h_u.tcp_ip4_spec.pdst;
-	input->dst_ip = fsp->h_u.tcp_ip4_spec.ip4src;
-	input->src_ip = fsp->h_u.tcp_ip4_spec.ip4dst;
+	input->vlan_etype = fsp->h_ext.vlan_etype;
+	if (!fsp->m_ext.vlan_etype && fsp->h_ext.vlan_tci)
+		input->vlan_etype = cpu_to_be16(ETH_P_8021Q);
+	if (fsp->m_ext.vlan_tci && input->vlan_etype)
+		input->vlan_tag = fsp->h_ext.vlan_tci;
+	if (input->flow_type == IPV6_USER_FLOW ||
+	    input->flow_type == UDP_V6_FLOW ||
+	    input->flow_type == TCP_V6_FLOW ||
+	    input->flow_type == SCTP_V6_FLOW) {
+		/* Reverse the src and dest notion, since the HW expects them
+		 * to be from Tx perspective where as the input from user is
+		 * from Rx filter view.
+		 */
+		input->ipl4_proto = fsp->h_u.usr_ip6_spec.l4_proto;
+		input->dst_port = fsp->h_u.tcp_ip6_spec.psrc;
+		input->src_port = fsp->h_u.tcp_ip6_spec.pdst;
+		memcpy(input->dst_ip6, fsp->h_u.ah_ip6_spec.ip6src,
+		       sizeof(__be32) * 4);
+		memcpy(input->src_ip6, fsp->h_u.ah_ip6_spec.ip6dst,
+		       sizeof(__be32) * 4);
+	} else {
+		/* Reverse the src and dest notion, since the HW expects them
+		 * to be from Tx perspective where as the input from user is
+		 * from Rx filter view.
+		 */
+		input->ipl4_proto = fsp->h_u.usr_ip4_spec.proto;
+		input->dst_port = fsp->h_u.tcp_ip4_spec.psrc;
+		input->src_port = fsp->h_u.tcp_ip4_spec.pdst;
+		input->dst_ip = fsp->h_u.tcp_ip4_spec.ip4src;
+		input->src_ip = fsp->h_u.tcp_ip4_spec.ip4dst;
+	}
 
 	if (userdef.flex_filter) {
 		input->flex_filter = true;
@@ -3730,19 +4922,25 @@ static int i40e_add_fdir_ethtool(struct i40e_vsi *vsi,
 		input->flex_offset = userdef.flex_offset;
 	}
 
-	ret = i40e_add_del_fdir(vsi, input, true);
+	/* Avoid programming two filters with identical match criteria. */
+	ret = i40e_disallow_matching_filters(vsi, input);
 	if (ret)
-		goto free_input;
+		goto free_filter_memory;
 
 	/* Add the input filter to the fdir_input_list, possibly replacing
 	 * a previous filter. Do not free the input structure after adding it
 	 * to the list as this would cause a use-after-free bug.
 	 */
 	i40e_update_ethtool_fdir_entry(vsi, input, fsp->location, NULL);
-
+	ret = i40e_add_del_fdir(vsi, input, true);
+	if (ret)
+		goto remove_sw_rule;
 	return 0;
 
-free_input:
+remove_sw_rule:
+	hlist_del(&input->fdir_node);
+	pf->fdir_pf_active_filters--;
+free_filter_memory:
 	kfree(input);
 	return ret;
 }
@@ -3758,13 +4956,9 @@ static int i40e_set_rxnfc(struct net_device *netdev, struct ethtool_rxnfc *cmd)
 {
 	struct i40e_netdev_priv *np = netdev_priv(netdev);
 	struct i40e_vsi *vsi = np->vsi;
-	struct i40e_pf *pf = vsi->back;
 	int ret = -EOPNOTSUPP;
 
 	switch (cmd->cmd) {
-	case ETHTOOL_SRXFH:
-		ret = i40e_set_rss_hash_opt(pf, cmd);
-		break;
 	case ETHTOOL_SRXCLSRLINS:
 		ret = i40e_add_fdir_ethtool(vsi, cmd);
 		break;
@@ -3790,7 +4984,7 @@ static unsigned int i40e_max_channels(struct i40e_vsi *vsi)
 
 /**
  * i40e_get_channels - Get the current channels enabled and max supported etc.
- * @netdev: network interface device structure
+ * @dev: network interface device structure
  * @ch: ethtool channels structure
  *
  * We don't support separate tx and rx queues as channels. The other count
@@ -3799,7 +4993,7 @@ static unsigned int i40e_max_channels(struct i40e_vsi *vsi)
  * q_vectors since we support a lot more queue pairs than q_vectors.
  **/
 static void i40e_get_channels(struct net_device *dev,
-			       struct ethtool_channels *ch)
+			      struct ethtool_channels *ch)
 {
 	struct i40e_netdev_priv *np = netdev_priv(dev);
 	struct i40e_vsi *vsi = np->vsi;
@@ -3809,7 +5003,7 @@ static void i40e_get_channels(struct net_device *dev,
 	ch->max_combined = i40e_max_channels(vsi);
 
 	/* report info for other vector */
-	ch->other_count = (pf->flags & I40E_FLAG_FD_SB_ENABLED) ? 1 : 0;
+	ch->other_count = test_bit(I40E_FLAG_FD_SB_ENA, pf->flags) ? 1 : 0;
 	ch->max_other = ch->other_count;
 
 	/* Note: This code assumes DCB is disabled for now. */
@@ -3818,14 +5012,14 @@ static void i40e_get_channels(struct net_device *dev,
 
 /**
  * i40e_set_channels - Set the new channels count.
- * @netdev: network interface device structure
+ * @dev: network interface device structure
  * @ch: ethtool channels structure
  *
  * The new channels count may not be the same as requested by the user
  * since it gets rounded down to a power of 2 value.
  **/
 static int i40e_set_channels(struct net_device *dev,
-			      struct ethtool_channels *ch)
+			     struct ethtool_channels *ch)
 {
 	const u8 drop = I40E_FILTER_PROGRAM_DESC_DEST_DROP_PACKET;
 	struct i40e_netdev_priv *np = netdev_priv(dev);
@@ -3841,12 +5035,18 @@ static int i40e_set_channels(struct net_device *dev,
 	if (vsi->type != I40E_VSI_MAIN)
 		return -EINVAL;
 
+	/* We do not support setting channels via ethtool when TCs are
+	 * configured through mqprio
+	 */
+	if (i40e_is_tc_mqprio_enabled(pf))
+		return -EINVAL;
+
 	/* verify they are not requesting separate vectors */
 	if (!count || ch->rx_count || ch->tx_count)
 		return -EINVAL;
 
 	/* verify other_count has not changed */
-	if (ch->other_count != ((pf->flags & I40E_FLAG_FD_SB_ENABLED) ? 1 : 0))
+	if (ch->other_count != (test_bit(I40E_FLAG_FD_SB_ENA, pf->flags) ? 1 : 0))
 		return -EINVAL;
 
 	/* verify the number of channels does not exceed hardware limits */
@@ -3908,8 +5108,16 @@ static u32 i40e_get_rxfh_indir_size(struct net_device *netdev)
 	return I40E_HLUT_ARRAY_SIZE;
 }
 
-static int i40e_get_rxfh(struct net_device *netdev, u32 *indir, u8 *key,
-			 u8 *hfunc)
+/**
+ * i40e_get_rxfh - get the rx flow hash indirection table
+ * @netdev: network interface device structure
+ * @rxfh: pointer to param struct (indir, key, hfunc)
+ *
+ * Reads the indirection table directly from the hardware. Returns 0 on
+ * success.
+ **/
+static int i40e_get_rxfh(struct net_device *netdev,
+			 struct ethtool_rxfh_param *rxfh)
 {
 	struct i40e_netdev_priv *np = netdev_priv(netdev);
 	struct i40e_vsi *vsi = np->vsi;
@@ -3917,13 +5125,12 @@ static int i40e_get_rxfh(struct net_device *netdev, u32 *indir, u8 *key,
 	int ret;
 	u16 i;
 
-	if (hfunc)
-		*hfunc = ETH_RSS_HASH_TOP;
+	rxfh->hfunc = ETH_RSS_HASH_TOP;
 
-	if (!indir)
+	if (!rxfh->indir)
 		return 0;
 
-	seed = key;
+	seed = rxfh->key;
 	lut = kzalloc(I40E_HLUT_ARRAY_SIZE, GFP_KERNEL);
 	if (!lut)
 		return -ENOMEM;
@@ -3931,7 +5138,7 @@ static int i40e_get_rxfh(struct net_device *netdev, u32 *indir, u8 *key,
 	if (ret)
 		goto out;
 	for (i = 0; i < I40E_HLUT_ARRAY_SIZE; i++)
-		indir[i] = (u32)(lut[i]);
+		rxfh->indir[i] = (u32)(lut[i]);
 
 out:
 	kfree(lut);
@@ -3942,14 +5149,15 @@ out:
 /**
  * i40e_set_rxfh - set the rx flow hash indirection table
  * @netdev: network interface device structure
- * @indir: indirection table
- * @key: hash key
+ * @rxfh: pointer to param struct (indir, key, hfunc)
+ * @extack: extended ACK from the Netlink message
  *
  * Returns -EINVAL if the table specifies an invalid queue id, otherwise
  * returns 0 after programming the table.
  **/
-static int i40e_set_rxfh(struct net_device *netdev, const u32 *indir,
-			 const u8 *key, const u8 hfunc)
+static int i40e_set_rxfh(struct net_device *netdev,
+			 struct ethtool_rxfh_param *rxfh,
+			 struct netlink_ext_ack *extack)
 {
 	struct i40e_netdev_priv *np = netdev_priv(netdev);
 	struct i40e_vsi *vsi = np->vsi;
@@ -3957,17 +5165,18 @@ static int i40e_set_rxfh(struct net_device *netdev, const u32 *indir,
 	u8 *seed = NULL;
 	u16 i;
 
-	if (hfunc != ETH_RSS_HASH_NO_CHANGE && hfunc != ETH_RSS_HASH_TOP)
+	if (rxfh->hfunc != ETH_RSS_HASH_NO_CHANGE &&
+	    rxfh->hfunc != ETH_RSS_HASH_TOP)
 		return -EOPNOTSUPP;
 
-	if (key) {
+	if (rxfh->key) {
 		if (!vsi->rss_hkey_user) {
 			vsi->rss_hkey_user = kzalloc(I40E_HKEY_ARRAY_SIZE,
 						     GFP_KERNEL);
 			if (!vsi->rss_hkey_user)
 				return -ENOMEM;
 		}
-		memcpy(vsi->rss_hkey_user, key, I40E_HKEY_ARRAY_SIZE);
+		memcpy(vsi->rss_hkey_user, rxfh->key, I40E_HKEY_ARRAY_SIZE);
 		seed = vsi->rss_hkey_user;
 	}
 	if (!vsi->rss_lut_user) {
@@ -3977,9 +5186,9 @@ static int i40e_set_rxfh(struct net_device *netdev, const u32 *indir,
 	}
 
 	/* Each 32 bits pointed by 'indir' is stored with a lut entry */
-	if (indir)
+	if (rxfh->indir)
 		for (i = 0; i < I40E_HLUT_ARRAY_SIZE; i++)
-			vsi->rss_lut_user[i] = (u8)(indir[i]);
+			vsi->rss_lut_user[i] = (u8)(rxfh->indir[i]);
 	else
 		i40e_fill_rss_lut(pf, vsi->rss_lut_user, I40E_HLUT_ARRAY_SIZE,
 				  vsi->rss_size);
@@ -4006,11 +5215,11 @@ static u32 i40e_get_priv_flags(struct net_device *dev)
 	u32 i, j, ret_flags = 0;
 
 	for (i = 0; i < I40E_PRIV_FLAGS_STR_LEN; i++) {
-		const struct i40e_priv_flags *priv_flags;
+		const struct i40e_priv_flags *priv_flag;
 
-		priv_flags = &i40e_gstrings_priv_flags[i];
+		priv_flag = &i40e_gstrings_priv_flags[i];
 
-		if (priv_flags->flag & pf->flags)
+		if (test_bit(priv_flag->bitno, pf->flags))
 			ret_flags |= BIT(i);
 	}
 
@@ -4018,11 +5227,11 @@ static u32 i40e_get_priv_flags(struct net_device *dev)
 		return ret_flags;
 
 	for (j = 0; j < I40E_GL_PRIV_FLAGS_STR_LEN; j++) {
-		const struct i40e_priv_flags *priv_flags;
+		const struct i40e_priv_flags *priv_flag;
 
-		priv_flags = &i40e_gl_gstrings_priv_flags[j];
+		priv_flag = &i40e_gl_gstrings_priv_flags[j];
 
-		if (priv_flags->flag & pf->flags)
+		if (test_bit(priv_flag->bitno, pf->flags))
 			ret_flags |= BIT(i + j);
 	}
 
@@ -4036,48 +5245,110 @@ static u32 i40e_get_priv_flags(struct net_device *dev)
  **/
 static int i40e_set_priv_flags(struct net_device *dev, u32 flags)
 {
+	DECLARE_BITMAP(changed_flags, I40E_PF_FLAGS_NBITS);
+	DECLARE_BITMAP(orig_flags, I40E_PF_FLAGS_NBITS);
+	DECLARE_BITMAP(new_flags, I40E_PF_FLAGS_NBITS);
 	struct i40e_netdev_priv *np = netdev_priv(dev);
 	struct i40e_vsi *vsi = np->vsi;
 	struct i40e_pf *pf = vsi->back;
-	u64 changed_flags;
+	enum libie_aq_err adq_err;
+	u32 reset_needed = 0;
+	int status;
 	u32 i, j;
 
-	changed_flags = pf->flags;
+	bitmap_copy(orig_flags, pf->flags, I40E_PF_FLAGS_NBITS);
+	bitmap_copy(new_flags, pf->flags, I40E_PF_FLAGS_NBITS);
 
 	for (i = 0; i < I40E_PRIV_FLAGS_STR_LEN; i++) {
-		const struct i40e_priv_flags *priv_flags;
+		const struct i40e_priv_flags *priv_flag;
+		bool new_val;
 
-		priv_flags = &i40e_gstrings_priv_flags[i];
+		priv_flag = &i40e_gstrings_priv_flags[i];
+		new_val = (flags & BIT(i)) ? true : false;
 
-		if (priv_flags->read_only)
-			continue;
+		/* If this is a read-only flag, it can't be changed */
+		if (priv_flag->read_only &&
+		    test_bit(priv_flag->bitno, orig_flags) != new_val)
+			return -EOPNOTSUPP;
 
-		if (flags & BIT(i))
-			pf->flags |= priv_flags->flag;
+		if (new_val)
+			set_bit(priv_flag->bitno, new_flags);
 		else
-			pf->flags &= ~(priv_flags->flag);
+			clear_bit(priv_flag->bitno, new_flags);
 	}
 
 	if (pf->hw.pf_id != 0)
 		goto flags_complete;
 
 	for (j = 0; j < I40E_GL_PRIV_FLAGS_STR_LEN; j++) {
-		const struct i40e_priv_flags *priv_flags;
+		const struct i40e_priv_flags *priv_flag;
+		bool new_val;
 
-		priv_flags = &i40e_gl_gstrings_priv_flags[j];
+		priv_flag = &i40e_gl_gstrings_priv_flags[j];
+		new_val = (flags & BIT(i + j)) ? true : false;
 
-		if (priv_flags->read_only)
-			continue;
+		/* If this is a read-only flag, it can't be changed */
+		if (priv_flag->read_only &&
+		    test_bit(priv_flag->bitno, orig_flags) != new_val)
+			return -EOPNOTSUPP;
 
-		if (flags & BIT(i + j))
-			pf->flags |= priv_flags->flag;
+		if (new_val)
+			set_bit(priv_flag->bitno, new_flags);
 		else
-			pf->flags &= ~(priv_flags->flag);
+			clear_bit(priv_flag->bitno, new_flags);
 	}
 
 flags_complete:
-	/* check for flags that changed */
-	changed_flags ^= pf->flags;
+	bitmap_xor(changed_flags, new_flags, orig_flags, I40E_PF_FLAGS_NBITS);
+
+	if (test_bit(I40E_FLAG_FW_LLDP_DIS, changed_flags))
+		reset_needed = I40E_PF_RESET_AND_REBUILD_FLAG;
+
+	if (test_bit(I40E_FLAG_VEB_STATS_ENA, changed_flags) ||
+	    test_bit(I40E_FLAG_LEGACY_RX_ENA, changed_flags) ||
+	    test_bit(I40E_FLAG_SOURCE_PRUNING_DIS, changed_flags))
+		reset_needed = BIT(__I40E_PF_RESET_REQUESTED);
+
+	/* Before we finalize any flag changes, we need to perform some
+	 * checks to ensure that the changes are supported and safe.
+	 */
+
+	/* ATR eviction is not supported on all devices */
+	if (test_bit(I40E_FLAG_HW_ATR_EVICT_ENA, new_flags) &&
+	    !test_bit(I40E_HW_CAP_ATR_EVICT, pf->hw.caps))
+		return -EOPNOTSUPP;
+
+	/* If the driver detected FW LLDP was disabled on init, this flag could
+	 * be set, however we do not support _changing_ the flag:
+	 * - on XL710 if NPAR is enabled or FW API version < 1.7
+	 * - on X722 with FW API version < 1.6
+	 * There are situations where older FW versions/NPAR enabled PFs could
+	 * disable LLDP, however we _must_ not allow the user to enable/disable
+	 * LLDP with this flag on unsupported FW versions.
+	 */
+	if (test_bit(I40E_FLAG_FW_LLDP_DIS, changed_flags) &&
+	    !test_bit(I40E_HW_CAP_FW_LLDP_STOPPABLE, pf->hw.caps)) {
+		dev_warn(&pf->pdev->dev,
+			 "Device does not support changing FW LLDP\n");
+		return -EOPNOTSUPP;
+	}
+
+	if (test_bit(I40E_FLAG_RS_FEC, changed_flags) &&
+	    pf->hw.device_id != I40E_DEV_ID_25G_SFP28 &&
+	    pf->hw.device_id != I40E_DEV_ID_25G_B) {
+		dev_warn(&pf->pdev->dev,
+			 "Device does not support changing FEC configuration\n");
+		return -EOPNOTSUPP;
+	}
+
+	if (test_bit(I40E_FLAG_BASE_R_FEC, changed_flags) &&
+	    pf->hw.device_id != I40E_DEV_ID_25G_SFP28 &&
+	    pf->hw.device_id != I40E_DEV_ID_25G_B &&
+	    pf->hw.device_id != I40E_DEV_ID_KX_X722) {
+		dev_warn(&pf->pdev->dev,
+			 "Device does not support changing FEC configuration\n");
+		return -EOPNOTSUPP;
+	}
 
 	/* Process any additional changes needed as a result of flag changes.
 	 * The changed_flags value reflects the list of bits that were
@@ -4085,51 +5356,456 @@ flags_complete:
 	 */
 
 	/* Flush current ATR settings if ATR was disabled */
-	if ((changed_flags & I40E_FLAG_FD_ATR_ENABLED) &&
-	    !(pf->flags & I40E_FLAG_FD_ATR_ENABLED)) {
-		pf->flags |= I40E_FLAG_FD_ATR_AUTO_DISABLED;
+	if (test_bit(I40E_FLAG_FD_ATR_ENA, changed_flags) &&
+	    !test_bit(I40E_FLAG_FD_ATR_ENA, new_flags)) {
+		set_bit(__I40E_FD_ATR_AUTO_DISABLED, pf->state);
 		set_bit(__I40E_FD_FLUSH_REQUESTED, pf->state);
 	}
 
-	/* Only allow ATR evict on hardware that is capable of handling it */
-	if (pf->flags & I40E_FLAG_HW_ATR_EVICT_CAPABLE)
-		pf->flags &= ~I40E_FLAG_HW_ATR_EVICT_CAPABLE;
-
-	if (changed_flags & I40E_FLAG_TRUE_PROMISC_SUPPORT) {
+	if (test_bit(I40E_FLAG_TRUE_PROMISC_ENA, changed_flags)) {
 		u16 sw_flags = 0, valid_flags = 0;
 		int ret;
 
-		if (!(pf->flags & I40E_FLAG_TRUE_PROMISC_SUPPORT))
+		if (!test_bit(I40E_FLAG_TRUE_PROMISC_ENA, new_flags))
 			sw_flags = I40E_AQ_SET_SWITCH_CFG_PROMISC;
 		valid_flags = I40E_AQ_SET_SWITCH_CFG_PROMISC;
 		ret = i40e_aq_set_switch_config(&pf->hw, sw_flags, valid_flags,
-						NULL);
-		if (ret && pf->hw.aq.asq_last_status != I40E_AQ_RC_ESRCH) {
+						0, NULL);
+		if (ret && pf->hw.aq.asq_last_status != LIBIE_AQ_RC_ESRCH) {
 			dev_info(&pf->pdev->dev,
-				 "couldn't set switch config bits, err %s aq_err %s\n",
-				 i40e_stat_str(&pf->hw, ret),
-				 i40e_aq_str(&pf->hw,
-					     pf->hw.aq.asq_last_status));
+				 "couldn't set switch config bits, err %pe aq_err %s\n",
+				 ERR_PTR(ret),
+				 libie_aq_str(pf->hw.aq.asq_last_status));
 			/* not a fatal problem, just keep going */
 		}
 	}
 
+	if (test_bit(I40E_FLAG_RS_FEC, changed_flags) ||
+	    test_bit(I40E_FLAG_BASE_R_FEC, changed_flags)) {
+		u8 fec_cfg = 0;
+
+		if (test_bit(I40E_FLAG_RS_FEC, new_flags) &&
+		    test_bit(I40E_FLAG_BASE_R_FEC, new_flags)) {
+			fec_cfg = I40E_AQ_SET_FEC_AUTO;
+		} else if (test_bit(I40E_FLAG_RS_FEC, new_flags)) {
+			fec_cfg = (I40E_AQ_SET_FEC_REQUEST_RS |
+				   I40E_AQ_SET_FEC_ABILITY_RS);
+		} else if (test_bit(I40E_FLAG_BASE_R_FEC, new_flags)) {
+			fec_cfg = (I40E_AQ_SET_FEC_REQUEST_KR |
+				   I40E_AQ_SET_FEC_ABILITY_KR);
+		}
+		if (i40e_set_fec_cfg(dev, fec_cfg))
+			dev_warn(&pf->pdev->dev, "Cannot change FEC config\n");
+	}
+
+	if (test_bit(I40E_FLAG_LINK_DOWN_ON_CLOSE_ENA, changed_flags) &&
+	    test_bit(I40E_FLAG_TOTAL_PORT_SHUTDOWN_ENA, orig_flags)) {
+		dev_err(&pf->pdev->dev,
+			"Setting link-down-on-close not supported on this port (because total-port-shutdown is enabled)\n");
+		return -EOPNOTSUPP;
+	}
+
+	if (test_bit(I40E_FLAG_VF_VLAN_PRUNING_ENA, changed_flags) &&
+	    pf->num_alloc_vfs) {
+		dev_warn(&pf->pdev->dev,
+			 "Changing vf-vlan-pruning flag while VF(s) are active is not supported\n");
+		return -EOPNOTSUPP;
+	}
+
+	if (test_bit(I40E_FLAG_LEGACY_RX_ENA, changed_flags) &&
+	    I40E_2K_TOO_SMALL_WITH_PADDING) {
+		dev_warn(&pf->pdev->dev,
+			 "2k Rx buffer is too small to fit standard MTU and skb_shared_info\n");
+		return -EOPNOTSUPP;
+	}
+
+	if (test_bit(I40E_FLAG_LINK_DOWN_ON_CLOSE_ENA, changed_flags) &&
+	    test_bit(I40E_FLAG_LINK_DOWN_ON_CLOSE_ENA, new_flags) &&
+	    test_bit(I40E_FLAG_MFP_ENA, new_flags))
+		dev_warn(&pf->pdev->dev,
+			 "Turning on link-down-on-close flag may affect other partitions\n");
+
+	if (test_bit(I40E_FLAG_FW_LLDP_DIS, changed_flags)) {
+		if (test_bit(I40E_FLAG_FW_LLDP_DIS, new_flags)) {
+#ifdef CONFIG_I40E_DCB
+			i40e_dcb_sw_default_config(pf);
+#endif /* CONFIG_I40E_DCB */
+			i40e_aq_cfg_lldp_mib_change_event(&pf->hw, false, NULL);
+			i40e_aq_stop_lldp(&pf->hw, true, false, NULL);
+		} else {
+			status = i40e_aq_start_lldp(&pf->hw, false, NULL);
+			if (status) {
+				adq_err = pf->hw.aq.asq_last_status;
+				switch (adq_err) {
+				case LIBIE_AQ_RC_EEXIST:
+					dev_warn(&pf->pdev->dev,
+						 "FW LLDP agent is already running\n");
+					reset_needed = 0;
+					break;
+				case LIBIE_AQ_RC_EPERM:
+					dev_warn(&pf->pdev->dev,
+						 "Device configuration forbids SW from starting the LLDP agent.\n");
+					return -EINVAL;
+				case LIBIE_AQ_RC_EAGAIN:
+					dev_warn(&pf->pdev->dev,
+						 "Stop FW LLDP agent command is still being processed, please try again in a second.\n");
+					return -EBUSY;
+				default:
+					dev_warn(&pf->pdev->dev,
+						 "Starting FW LLDP agent failed: error: %pe, %s\n",
+						 ERR_PTR(status),
+						 libie_aq_str(adq_err));
+					return -EINVAL;
+				}
+			}
+		}
+	}
+
+	/* Now that we've checked to ensure that the new flags are valid, load
+	 * them into place. Since we only modify flags either (a) during
+	 * initialization or (b) while holding the RTNL lock, we don't need
+	 * anything fancy here.
+	 */
+	bitmap_copy(pf->flags, new_flags, I40E_PF_FLAGS_NBITS);
+
 	/* Issue reset to cause things to take effect, as additional bits
 	 * are added we will need to create a mask of bits requiring reset
 	 */
-	if ((changed_flags & I40E_FLAG_VEB_STATS_ENABLED) ||
-	    ((changed_flags & I40E_FLAG_LEGACY_RX) && netif_running(dev)))
-		i40e_do_reset(pf, BIT(__I40E_PF_RESET_REQUESTED), true);
+	if (reset_needed)
+		i40e_do_reset(pf, reset_needed, true);
+
+	return 0;
+}
+
+/**
+ * i40e_get_module_info - get (Q)SFP+ module type info
+ * @netdev: network interface device structure
+ * @modinfo: module EEPROM size and layout information structure
+ **/
+static int i40e_get_module_info(struct net_device *netdev,
+				struct ethtool_modinfo *modinfo)
+{
+	struct i40e_netdev_priv *np = netdev_priv(netdev);
+	struct i40e_vsi *vsi = np->vsi;
+	struct i40e_pf *pf = vsi->back;
+	struct i40e_hw *hw = &pf->hw;
+	u32 sff8472_comp = 0;
+	u32 sff8472_swap = 0;
+	u32 sff8636_rev = 0;
+	u32 type = 0;
+	int status;
+
+	/* Check if firmware supports reading module EEPROM. */
+	if (!test_bit(I40E_HW_CAP_AQ_PHY_ACCESS, hw->caps)) {
+		netdev_err(vsi->netdev, "Module EEPROM memory read not supported. Please update the NVM image.\n");
+		return -EINVAL;
+	}
+
+	status = i40e_update_link_info(hw);
+	if (status)
+		return -EIO;
+
+	if (hw->phy.link_info.phy_type == I40E_PHY_TYPE_EMPTY) {
+		netdev_err(vsi->netdev, "Cannot read module EEPROM memory. No module connected.\n");
+		return -EINVAL;
+	}
+
+	type = hw->phy.link_info.module_type[0];
+
+	switch (type) {
+	case I40E_MODULE_TYPE_SFP:
+		status = i40e_aq_get_phy_register(hw,
+				I40E_AQ_PHY_REG_ACCESS_EXTERNAL_MODULE,
+				I40E_I2C_EEPROM_DEV_ADDR, true,
+				I40E_MODULE_SFF_8472_COMP,
+				&sff8472_comp, NULL);
+		if (status)
+			return -EIO;
+
+		status = i40e_aq_get_phy_register(hw,
+				I40E_AQ_PHY_REG_ACCESS_EXTERNAL_MODULE,
+				I40E_I2C_EEPROM_DEV_ADDR, true,
+				I40E_MODULE_SFF_8472_SWAP,
+				&sff8472_swap, NULL);
+		if (status)
+			return -EIO;
+
+		/* Check if the module requires address swap to access
+		 * the other EEPROM memory page.
+		 */
+		if (sff8472_swap & I40E_MODULE_SFF_ADDR_MODE) {
+			netdev_warn(vsi->netdev, "Module address swap to access page 0xA2 is not supported.\n");
+			modinfo->type = ETH_MODULE_SFF_8079;
+			modinfo->eeprom_len = ETH_MODULE_SFF_8079_LEN;
+		} else if (sff8472_comp == 0x00) {
+			/* Module is not SFF-8472 compliant */
+			modinfo->type = ETH_MODULE_SFF_8079;
+			modinfo->eeprom_len = ETH_MODULE_SFF_8079_LEN;
+		} else if (!(sff8472_swap & I40E_MODULE_SFF_DDM_IMPLEMENTED)) {
+			/* Module is SFF-8472 compliant but doesn't implement
+			 * Digital Diagnostic Monitoring (DDM).
+			 */
+			modinfo->type = ETH_MODULE_SFF_8079;
+			modinfo->eeprom_len = ETH_MODULE_SFF_8079_LEN;
+		} else {
+			modinfo->type = ETH_MODULE_SFF_8472;
+			modinfo->eeprom_len = ETH_MODULE_SFF_8472_LEN;
+		}
+		break;
+	case I40E_MODULE_TYPE_QSFP_PLUS:
+		/* Read from memory page 0. */
+		status = i40e_aq_get_phy_register(hw,
+				I40E_AQ_PHY_REG_ACCESS_EXTERNAL_MODULE,
+				0, true,
+				I40E_MODULE_REVISION_ADDR,
+				&sff8636_rev, NULL);
+		if (status)
+			return -EIO;
+		/* Determine revision compliance byte */
+		if (sff8636_rev > 0x02) {
+			/* Module is SFF-8636 compliant */
+			modinfo->type = ETH_MODULE_SFF_8636;
+			modinfo->eeprom_len = I40E_MODULE_QSFP_MAX_LEN;
+		} else {
+			modinfo->type = ETH_MODULE_SFF_8436;
+			modinfo->eeprom_len = I40E_MODULE_QSFP_MAX_LEN;
+		}
+		break;
+	case I40E_MODULE_TYPE_QSFP28:
+		modinfo->type = ETH_MODULE_SFF_8636;
+		modinfo->eeprom_len = I40E_MODULE_QSFP_MAX_LEN;
+		break;
+	default:
+		netdev_dbg(vsi->netdev, "SFP module type unrecognized or no SFP connector used.\n");
+		return -EOPNOTSUPP;
+	}
+	return 0;
+}
+
+/**
+ * i40e_get_module_eeprom - fills buffer with (Q)SFP+ module memory contents
+ * @netdev: network interface device structure
+ * @ee: EEPROM dump request structure
+ * @data: buffer to be filled with EEPROM contents
+ **/
+static int i40e_get_module_eeprom(struct net_device *netdev,
+				  struct ethtool_eeprom *ee,
+				  u8 *data)
+{
+	struct i40e_netdev_priv *np = netdev_priv(netdev);
+	struct i40e_vsi *vsi = np->vsi;
+	struct i40e_pf *pf = vsi->back;
+	struct i40e_hw *hw = &pf->hw;
+	bool is_sfp = false;
+	u32 value = 0;
+	int status;
+	int i;
+
+	if (!ee || !ee->len || !data)
+		return -EINVAL;
+
+	if (hw->phy.link_info.module_type[0] == I40E_MODULE_TYPE_SFP)
+		is_sfp = true;
 
+	for (i = 0; i < ee->len; i++) {
+		u32 offset = i + ee->offset;
+		u32 addr = is_sfp ? I40E_I2C_EEPROM_DEV_ADDR : 0;
+
+		/* Check if we need to access the other memory page */
+		if (is_sfp) {
+			if (offset >= ETH_MODULE_SFF_8079_LEN) {
+				offset -= ETH_MODULE_SFF_8079_LEN;
+				addr = I40E_I2C_EEPROM_DEV_ADDR2;
+			}
+		} else {
+			while (offset >= ETH_MODULE_SFF_8436_LEN) {
+				/* Compute memory page number and offset. */
+				offset -= ETH_MODULE_SFF_8436_LEN / 2;
+				addr++;
+			}
+		}
+
+		status = i40e_aq_get_phy_register(hw,
+				I40E_AQ_PHY_REG_ACCESS_EXTERNAL_MODULE,
+				addr, true, offset, &value, NULL);
+		if (status)
+			return -EIO;
+		data[i] = value;
+	}
 	return 0;
 }
 
+static void i40e_eee_capability_to_kedata_supported(__le16 eee_capability_,
+						    unsigned long *supported)
+{
+	const int eee_capability = le16_to_cpu(eee_capability_);
+	static const int lut[] = {
+		ETHTOOL_LINK_MODE_100baseT_Full_BIT,
+		ETHTOOL_LINK_MODE_1000baseT_Full_BIT,
+		ETHTOOL_LINK_MODE_10000baseT_Full_BIT,
+		ETHTOOL_LINK_MODE_1000baseKX_Full_BIT,
+		ETHTOOL_LINK_MODE_10000baseKX4_Full_BIT,
+		ETHTOOL_LINK_MODE_10000baseKR_Full_BIT,
+		ETHTOOL_LINK_MODE_40000baseKR4_Full_BIT,
+	};
+
+	linkmode_zero(supported);
+	for (unsigned int i = ARRAY_SIZE(lut); i--; )
+		if (eee_capability & BIT(i + 1))
+			linkmode_set_bit(lut[i], supported);
+}
+
+static int i40e_get_eee(struct net_device *netdev, struct ethtool_keee *edata)
+{
+	struct i40e_netdev_priv *np = netdev_priv(netdev);
+	struct i40e_aq_get_phy_abilities_resp phy_cfg;
+	struct i40e_vsi *vsi = np->vsi;
+	struct i40e_pf *pf = vsi->back;
+	struct i40e_hw *hw = &pf->hw;
+	int status;
+
+	/* Get initial PHY capabilities */
+	status = i40e_aq_get_phy_capabilities(hw, false, true, &phy_cfg, NULL);
+	if (status)
+		return -EAGAIN;
+
+	/* Check whether NIC configuration is compatible with Energy Efficient
+	 * Ethernet (EEE) mode.
+	 */
+	if (phy_cfg.eee_capability == 0)
+		return -EOPNOTSUPP;
+
+	i40e_eee_capability_to_kedata_supported(phy_cfg.eee_capability,
+						edata->supported);
+	linkmode_copy(edata->lp_advertised, edata->supported);
+
+	/* Get current configuration */
+	status = i40e_aq_get_phy_capabilities(hw, false, false, &phy_cfg, NULL);
+	if (status)
+		return -EAGAIN;
+
+	linkmode_zero(edata->advertised);
+	if (phy_cfg.eee_capability)
+		linkmode_copy(edata->advertised, edata->supported);
+	edata->eee_enabled = !!phy_cfg.eee_capability;
+	edata->tx_lpi_enabled = pf->stats.tx_lpi_status;
+
+	edata->eee_active = pf->stats.tx_lpi_status && pf->stats.rx_lpi_status;
+
+	return 0;
+}
+
+static int i40e_is_eee_param_supported(struct net_device *netdev,
+				       struct ethtool_keee *edata)
+{
+	struct i40e_netdev_priv *np = netdev_priv(netdev);
+	struct i40e_vsi *vsi = np->vsi;
+	struct i40e_pf *pf = vsi->back;
+	struct i40e_ethtool_not_used {
+		bool value;
+		const char *name;
+	} param[] = {
+		{!!(edata->advertised[0] & ~edata->supported[0]), "advertise"},
+		{!!edata->tx_lpi_timer, "tx-timer"},
+		{edata->tx_lpi_enabled != pf->stats.tx_lpi_status, "tx-lpi"}
+	};
+	int i;
+
+	for (i = 0; i < ARRAY_SIZE(param); i++) {
+		if (param[i].value) {
+			netdev_info(netdev,
+				    "EEE setting %s not supported\n",
+				    param[i].name);
+			return -EOPNOTSUPP;
+		}
+	}
+
+	return 0;
+}
+
+static int i40e_set_eee(struct net_device *netdev, struct ethtool_keee *edata)
+{
+	struct i40e_netdev_priv *np = netdev_priv(netdev);
+	struct i40e_aq_get_phy_abilities_resp abilities;
+	struct i40e_aq_set_phy_config config;
+	struct i40e_vsi *vsi = np->vsi;
+	struct i40e_pf *pf = vsi->back;
+	struct i40e_hw *hw = &pf->hw;
+	__le16 eee_capability;
+	int status;
+
+	/* Deny parameters we don't support */
+	if (i40e_is_eee_param_supported(netdev, edata))
+		return -EOPNOTSUPP;
+
+	/* Get initial PHY capabilities */
+	status = i40e_aq_get_phy_capabilities(hw, false, true, &abilities,
+					      NULL);
+	if (status)
+		return -EAGAIN;
+
+	/* Check whether NIC configuration is compatible with Energy Efficient
+	 * Ethernet (EEE) mode.
+	 */
+	if (abilities.eee_capability == 0)
+		return -EOPNOTSUPP;
+
+	/* Cache initial EEE capability */
+	eee_capability = abilities.eee_capability;
+
+	/* Get current PHY configuration */
+	status = i40e_aq_get_phy_capabilities(hw, false, false, &abilities,
+					      NULL);
+	if (status)
+		return -EAGAIN;
+
+	/* Cache current PHY configuration */
+	config.phy_type = abilities.phy_type;
+	config.phy_type_ext = abilities.phy_type_ext;
+	config.link_speed = abilities.link_speed;
+	config.abilities = abilities.abilities |
+			   I40E_AQ_PHY_ENABLE_ATOMIC_LINK;
+	config.eeer = abilities.eeer_val;
+	config.low_power_ctrl = abilities.d3_lpan;
+	config.fec_config = abilities.fec_cfg_curr_mod_ext_info &
+			    I40E_AQ_PHY_FEC_CONFIG_MASK;
+
+	/* Set desired EEE state */
+	if (edata->eee_enabled) {
+		config.eee_capability = eee_capability;
+		config.eeer |= cpu_to_le32(I40E_PRTPM_EEER_TX_LPI_EN_MASK);
+	} else {
+		config.eee_capability = 0;
+		config.eeer &= cpu_to_le32(~I40E_PRTPM_EEER_TX_LPI_EN_MASK);
+	}
+
+	/* Apply modified PHY configuration */
+	status = i40e_aq_set_phy_config(hw, &config, NULL);
+	if (status)
+		return -EAGAIN;
+
+	return 0;
+}
+
+static const struct ethtool_ops i40e_ethtool_recovery_mode_ops = {
+	.get_drvinfo		= i40e_get_drvinfo,
+	.set_eeprom		= i40e_set_eeprom,
+	.get_eeprom_len		= i40e_get_eeprom_len,
+	.get_eeprom		= i40e_get_eeprom,
+};
+
 static const struct ethtool_ops i40e_ethtool_ops = {
+	.supported_coalesce_params = ETHTOOL_COALESCE_USECS |
+				     ETHTOOL_COALESCE_TX_MAX_FRAMES_IRQ |
+				     ETHTOOL_COALESCE_USE_ADAPTIVE |
+				     ETHTOOL_COALESCE_RX_USECS_HIGH |
+				     ETHTOOL_COALESCE_TX_USECS_HIGH,
 	.get_drvinfo		= i40e_get_drvinfo,
 	.get_regs_len		= i40e_get_regs_len,
 	.get_regs		= i40e_get_regs,
 	.nway_reset		= i40e_nway_reset,
 	.get_link		= ethtool_op_get_link,
+	.get_link_ext_stats	= i40e_get_link_ext_stats,
 	.get_wol		= i40e_get_wol,
 	.set_wol		= i40e_set_wol,
 	.set_eeprom		= i40e_set_eeprom,
@@ -4145,6 +5821,8 @@ static const struct ethtool_ops i40e_ethtool_ops = {
 	.set_rxnfc		= i40e_set_rxnfc,
 	.self_test		= i40e_diag_test,
 	.get_strings		= i40e_get_strings,
+	.get_eee		= i40e_get_eee,
+	.set_eee		= i40e_set_eee,
 	.set_phys_id		= i40e_set_phys_id,
 	.get_sset_count		= i40e_get_sset_count,
 	.get_ethtool_stats	= i40e_get_ethtool_stats,
@@ -4154,8 +5832,12 @@ static const struct ethtool_ops i40e_ethtool_ops = {
 	.get_rxfh_indir_size	= i40e_get_rxfh_indir_size,
 	.get_rxfh		= i40e_get_rxfh,
 	.set_rxfh		= i40e_set_rxfh,
+	.get_rxfh_fields	= i40e_get_rxfh_fields,
+	.set_rxfh_fields	= i40e_set_rxfh_fields,
 	.get_channels		= i40e_get_channels,
 	.set_channels		= i40e_set_channels,
+	.get_module_info	= i40e_get_module_info,
+	.get_module_eeprom	= i40e_get_module_eeprom,
 	.get_ts_info		= i40e_get_ts_info,
 	.get_priv_flags		= i40e_get_priv_flags,
 	.set_priv_flags		= i40e_set_priv_flags,
@@ -4163,9 +5845,18 @@ static const struct ethtool_ops i40e_ethtool_ops = {
 	.set_per_queue_coalesce	= i40e_set_per_queue_coalesce,
 	.get_link_ksettings	= i40e_get_link_ksettings,
 	.set_link_ksettings	= i40e_set_link_ksettings,
+	.get_fecparam = i40e_get_fec_param,
+	.set_fecparam = i40e_set_fec_param,
+	.flash_device = i40e_ddp_flash,
 };
 
 void i40e_set_ethtool_ops(struct net_device *netdev)
 {
-	netdev->ethtool_ops = &i40e_ethtool_ops;
+	struct i40e_netdev_priv *np = netdev_priv(netdev);
+	struct i40e_pf		*pf = np->vsi->back;
+
+	if (!test_bit(__I40E_RECOVERY_MODE, pf->state))
+		netdev->ethtool_ops = &i40e_ethtool_ops;
+	else
+		netdev->ethtool_ops = &i40e_ethtool_recovery_mode_ops;
 }
diff --git a/drivers/net/ethernet/intel/i40e/i40e_fcoe.c b/drivers/net/ethernet/intel/i40e/i40e_fcoe.c
deleted file mode 100644
index b077ef8b00fa..000000000000
--- a/drivers/net/ethernet/intel/i40e/i40e_fcoe.c
+++ /dev/null
@@ -1,1571 +0,0 @@
-/*******************************************************************************
- *
- * Intel Ethernet Controller XL710 Family Linux Driver
- * Copyright(c) 2013 - 2016 Intel Corporation.
- *
- * This program is free software; you can redistribute it and/or modify it
- * under the terms and conditions of the GNU General Public License,
- * version 2, as published by the Free Software Foundation.
- *
- * This program is distributed in the hope it will be useful, but WITHOUT
- * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
- * FITNESS FOR A PARTICULAR PURPOSE.  See the GNU General Public License for
- * more details.
- *
- * You should have received a copy of the GNU General Public License along
- * with this program.  If not, see <http://www.gnu.org/licenses/>.
- *
- * The full GNU General Public License is included in this distribution in
- * the file called "COPYING".
- *
- * Contact Information:
- * e1000-devel Mailing List <e1000-devel@lists.sourceforge.net>
- * Intel Corporation, 5200 N.E. Elam Young Parkway, Hillsboro, OR 97124-6497
- *
- ******************************************************************************/
-
-#include <linux/if_ether.h>
-#include <scsi/scsi_cmnd.h>
-#include <scsi/scsi_device.h>
-#include <scsi/fc/fc_fs.h>
-#include <scsi/fc/fc_fip.h>
-#include <scsi/fc/fc_fcoe.h>
-#include <scsi/libfc.h>
-#include <scsi/libfcoe.h>
-#include <uapi/linux/dcbnl.h>
-
-#include "i40e.h"
-#include "i40e_fcoe.h"
-
-/**
- * i40e_fcoe_sof_is_class2 - returns true if this is a FC Class 2 SOF
- * @sof: the FCoE start of frame delimiter
- **/
-static inline bool i40e_fcoe_sof_is_class2(u8 sof)
-{
-	return (sof == FC_SOF_I2) || (sof == FC_SOF_N2);
-}
-
-/**
- * i40e_fcoe_sof_is_class3 - returns true if this is a FC Class 3 SOF
- * @sof: the FCoE start of frame delimiter
- **/
-static inline bool i40e_fcoe_sof_is_class3(u8 sof)
-{
-	return (sof == FC_SOF_I3) || (sof == FC_SOF_N3);
-}
-
-/**
- * i40e_fcoe_sof_is_supported - returns true if the FC SOF is supported by HW
- * @sof: the input SOF value from the frame
- **/
-static inline bool i40e_fcoe_sof_is_supported(u8 sof)
-{
-	return i40e_fcoe_sof_is_class2(sof) ||
-	       i40e_fcoe_sof_is_class3(sof);
-}
-
-/**
- * i40e_fcoe_fc_sof - pull the SOF from FCoE header in the frame
- * @skb: the frame whose EOF is to be pulled from
- **/
-static inline int i40e_fcoe_fc_sof(struct sk_buff *skb, u8 *sof)
-{
-	*sof = ((struct fcoe_hdr *)skb_network_header(skb))->fcoe_sof;
-
-	if (!i40e_fcoe_sof_is_supported(*sof))
-		return -EINVAL;
-	return 0;
-}
-
-/**
- * i40e_fcoe_eof_is_supported - returns true if the EOF is supported by HW
- * @eof:     the input EOF value from the frame
- **/
-static inline bool i40e_fcoe_eof_is_supported(u8 eof)
-{
-	return (eof == FC_EOF_N) || (eof == FC_EOF_T) ||
-	       (eof == FC_EOF_NI) || (eof == FC_EOF_A);
-}
-
-/**
- * i40e_fcoe_fc_eof - pull EOF from FCoE trailer in the frame
- * @skb: the frame whose EOF is to be pulled from
- **/
-static inline int i40e_fcoe_fc_eof(struct sk_buff *skb, u8 *eof)
-{
-	/* the first byte of the last dword is EOF */
-	skb_copy_bits(skb, skb->len - 4, eof, 1);
-
-	if (!i40e_fcoe_eof_is_supported(*eof))
-		return -EINVAL;
-	return 0;
-}
-
-/**
- * i40e_fcoe_ctxt_eof - convert input FC EOF for descriptor programming
- * @eof: the input eof value from the frame
- *
- * The FC EOF is converted to the value understood by HW for descriptor
- * programming. Never call this w/o calling i40e_fcoe_eof_is_supported()
- * first and that already checks for all supported valid eof values.
- **/
-static inline u32 i40e_fcoe_ctxt_eof(u8 eof)
-{
-	switch (eof) {
-	case FC_EOF_N:
-		return I40E_TX_DESC_CMD_L4T_EOFT_EOF_N;
-	case FC_EOF_T:
-		return I40E_TX_DESC_CMD_L4T_EOFT_EOF_T;
-	case FC_EOF_NI:
-		return I40E_TX_DESC_CMD_L4T_EOFT_EOF_NI;
-	case FC_EOF_A:
-		return I40E_TX_DESC_CMD_L4T_EOFT_EOF_A;
-	default:
-		/* Supported valid eof shall be already checked by
-		 * calling i40e_fcoe_eof_is_supported() first,
-		 * therefore this default case shall never hit.
-		 */
-		WARN_ON(1);
-		return -EINVAL;
-	}
-}
-
-/**
- * i40e_fcoe_xid_is_valid - returns true if the exchange id is valid
- * @xid: the exchange id
- **/
-static inline bool i40e_fcoe_xid_is_valid(u16 xid)
-{
-	return (xid != FC_XID_UNKNOWN) && (xid < I40E_FCOE_DDP_MAX);
-}
-
-/**
- * i40e_fcoe_ddp_unmap - unmap the mapped sglist associated
- * @pf: pointer to PF
- * @ddp: sw DDP context
- *
- * Unmap the scatter-gather list associated with the given SW DDP context
- *
- * Returns: data length already ddp-ed in bytes
- *
- **/
-static inline void i40e_fcoe_ddp_unmap(struct i40e_pf *pf,
-				       struct i40e_fcoe_ddp *ddp)
-{
-	if (test_and_set_bit(__I40E_FCOE_DDP_UNMAPPED, &ddp->flags))
-		return;
-
-	if (ddp->sgl) {
-		dma_unmap_sg(&pf->pdev->dev, ddp->sgl, ddp->sgc,
-			     DMA_FROM_DEVICE);
-		ddp->sgl = NULL;
-		ddp->sgc = 0;
-	}
-
-	if (ddp->pool) {
-		dma_pool_free(ddp->pool, ddp->udl, ddp->udp);
-		ddp->pool = NULL;
-	}
-}
-
-/**
- * i40e_fcoe_ddp_clear - clear the given SW DDP context
- * @ddp - SW DDP context
- **/
-static inline void i40e_fcoe_ddp_clear(struct i40e_fcoe_ddp *ddp)
-{
-	memset(ddp, 0, sizeof(struct i40e_fcoe_ddp));
-	ddp->xid = FC_XID_UNKNOWN;
-	ddp->flags = __I40E_FCOE_DDP_NONE;
-}
-
-/**
- * i40e_fcoe_progid_is_fcoe - check if the prog_id is for FCoE
- * @id: the prog id for the programming status Rx descriptor write-back
- **/
-static inline bool i40e_fcoe_progid_is_fcoe(u8 id)
-{
-	return (id == I40E_RX_PROG_STATUS_DESC_FCOE_CTXT_PROG_STATUS) ||
-	       (id == I40E_RX_PROG_STATUS_DESC_FCOE_CTXT_INVL_STATUS);
-}
-
-/**
- * i40e_fcoe_fc_get_xid - get xid from the frame header
- * @fh: the fc frame header
- *
- * In case the incoming frame's exchange is originated from
- * the initiator, then received frame's exchange id is ANDed
- * with fc_cpu_mask bits to get the same cpu on which exchange
- * was originated, otherwise just use the current cpu.
- *
- * Returns ox_id if exchange originator, rx_id if responder
- **/
-static inline u16 i40e_fcoe_fc_get_xid(struct fc_frame_header *fh)
-{
-	u32 f_ctl = ntoh24(fh->fh_f_ctl);
-
-	return (f_ctl & FC_FC_EX_CTX) ?
-		be16_to_cpu(fh->fh_ox_id) :
-		be16_to_cpu(fh->fh_rx_id);
-}
-
-/**
- * i40e_fcoe_fc_frame_header - get fc frame header from skb
- * @skb: packet
- *
- * This checks if there is a VLAN header and returns the data
- * pointer to the start of the fc_frame_header.
- *
- * Returns pointer to the fc_frame_header
- **/
-static inline struct fc_frame_header *i40e_fcoe_fc_frame_header(
-	struct sk_buff *skb)
-{
-	void *fh = skb->data + sizeof(struct fcoe_hdr);
-
-	if (eth_hdr(skb)->h_proto == htons(ETH_P_8021Q))
-		fh += sizeof(struct vlan_hdr);
-
-	return (struct fc_frame_header *)fh;
-}
-
-/**
- * i40e_fcoe_ddp_put - release the DDP context for a given exchange id
- * @netdev: the corresponding net_device
- * @xid: the exchange id that corresponding DDP context will be released
- *
- * This is the implementation of net_device_ops.ndo_fcoe_ddp_done
- * and it is expected to be called by ULD, i.e., FCP layer of libfc
- * to release the corresponding ddp context when the I/O is done.
- *
- * Returns : data length already ddp-ed in bytes
- **/
-static int i40e_fcoe_ddp_put(struct net_device *netdev, u16 xid)
-{
-	struct i40e_netdev_priv *np = netdev_priv(netdev);
-	struct i40e_pf *pf = np->vsi->back;
-	struct i40e_fcoe *fcoe = &pf->fcoe;
-	int len = 0;
-	struct i40e_fcoe_ddp *ddp = &fcoe->ddp[xid];
-
-	if (!fcoe || !ddp)
-		goto out;
-
-	if (test_bit(__I40E_FCOE_DDP_DONE, &ddp->flags))
-		len = ddp->len;
-	i40e_fcoe_ddp_unmap(pf, ddp);
-out:
-	return len;
-}
-
-/**
- * i40e_fcoe_sw_init - sets up the HW for FCoE
- * @pf: pointer to PF
- **/
-void i40e_init_pf_fcoe(struct i40e_pf *pf)
-{
-	struct i40e_hw *hw = &pf->hw;
-	u32 val;
-
-	pf->flags &= ~I40E_FLAG_FCOE_ENABLED;
-	pf->num_fcoe_qps = 0;
-	pf->fcoe_hmc_cntx_num = 0;
-	pf->fcoe_hmc_filt_num = 0;
-
-	if (!pf->hw.func_caps.fcoe) {
-		dev_dbg(&pf->pdev->dev, "FCoE capability is disabled\n");
-		return;
-	}
-
-	if (!pf->hw.func_caps.dcb) {
-		dev_warn(&pf->pdev->dev,
-			 "Hardware is not DCB capable not enabling FCoE.\n");
-		return;
-	}
-
-	/* enable FCoE hash filter */
-	val = i40e_read_rx_ctl(hw, I40E_PFQF_HENA(1));
-	val |= BIT(I40E_FILTER_PCTYPE_FCOE_OX - 32);
-	val |= BIT(I40E_FILTER_PCTYPE_FCOE_RX - 32);
-	val &= I40E_PFQF_HENA_PTYPE_ENA_MASK;
-	i40e_write_rx_ctl(hw, I40E_PFQF_HENA(1), val);
-
-	/* enable flag */
-	pf->flags |= I40E_FLAG_FCOE_ENABLED;
-	pf->num_fcoe_qps = I40E_DEFAULT_FCOE;
-
-	/* Reserve 4K DDP contexts and 20K filter size for FCoE */
-	pf->fcoe_hmc_cntx_num = BIT(I40E_DMA_CNTX_SIZE_4K) *
-				I40E_DMA_CNTX_BASE_SIZE;
-	pf->fcoe_hmc_filt_num = pf->fcoe_hmc_cntx_num +
-				BIT(I40E_HASH_FILTER_SIZE_16K) *
-				I40E_HASH_FILTER_BASE_SIZE;
-
-	/* FCoE object: max 16K filter buckets and 4K DMA contexts */
-	pf->filter_settings.fcoe_filt_num = I40E_HASH_FILTER_SIZE_16K;
-	pf->filter_settings.fcoe_cntx_num = I40E_DMA_CNTX_SIZE_4K;
-
-	/* Setup max frame with FCoE_MTU plus L2 overheads */
-	val = i40e_read_rx_ctl(hw, I40E_GLFCOE_RCTL);
-	val &= ~I40E_GLFCOE_RCTL_MAX_SIZE_MASK;
-	val |= ((FCOE_MTU + ETH_HLEN + VLAN_HLEN + ETH_FCS_LEN)
-		 << I40E_GLFCOE_RCTL_MAX_SIZE_SHIFT);
-	i40e_write_rx_ctl(hw, I40E_GLFCOE_RCTL, val);
-
-	dev_info(&pf->pdev->dev, "FCoE is supported.\n");
-}
-
-/**
- * i40e_get_fcoe_tc_map - Return TC map for FCoE APP
- * @pf: pointer to PF
- *
- **/
-u8 i40e_get_fcoe_tc_map(struct i40e_pf *pf)
-{
-	struct i40e_dcb_app_priority_table app;
-	struct i40e_hw *hw = &pf->hw;
-	u8 enabled_tc = 0;
-	u8 tc, i;
-	/* Get the FCoE APP TLV */
-	struct i40e_dcbx_config *dcbcfg = &hw->local_dcbx_config;
-
-	for (i = 0; i < dcbcfg->numapps; i++) {
-		app = dcbcfg->app[i];
-		if (app.selector == IEEE_8021QAZ_APP_SEL_ETHERTYPE &&
-		    app.protocolid == ETH_P_FCOE) {
-			tc = dcbcfg->etscfg.prioritytable[app.priority];
-			enabled_tc |= BIT(tc);
-			break;
-		}
-	}
-
-	/* TC0 if there is no TC defined for FCoE APP TLV */
-	enabled_tc = enabled_tc ? enabled_tc : 0x1;
-
-	return enabled_tc;
-}
-
-/**
- * i40e_fcoe_vsi_init - prepares the VSI context for creating a FCoE VSI
- * @vsi: pointer to the associated VSI struct
- * @ctxt: pointer to the associated VSI context to be passed to HW
- *
- * Returns 0 on success or < 0 on error
- **/
-int i40e_fcoe_vsi_init(struct i40e_vsi *vsi, struct i40e_vsi_context *ctxt)
-{
-	struct i40e_aqc_vsi_properties_data *info = &ctxt->info;
-	struct i40e_pf *pf = vsi->back;
-	struct i40e_hw *hw = &pf->hw;
-	u8 enabled_tc = 0;
-
-	if (!(pf->flags & I40E_FLAG_FCOE_ENABLED)) {
-		dev_err(&pf->pdev->dev,
-			"FCoE is not enabled for this device\n");
-		return -EPERM;
-	}
-
-	/* initialize the hardware for FCoE */
-	ctxt->pf_num = hw->pf_id;
-	ctxt->vf_num = 0;
-	ctxt->uplink_seid = vsi->uplink_seid;
-	ctxt->connection_type = I40E_AQ_VSI_CONN_TYPE_NORMAL;
-	ctxt->flags = I40E_AQ_VSI_TYPE_PF;
-
-	/* FCoE VSI would need the following sections */
-	info->valid_sections |= cpu_to_le16(I40E_AQ_VSI_PROP_QUEUE_OPT_VALID);
-
-	/* FCoE VSI does not need these sections */
-	info->valid_sections &= cpu_to_le16(~(I40E_AQ_VSI_PROP_SECURITY_VALID |
-					    I40E_AQ_VSI_PROP_VLAN_VALID |
-					    I40E_AQ_VSI_PROP_CAS_PV_VALID |
-					    I40E_AQ_VSI_PROP_INGRESS_UP_VALID |
-					    I40E_AQ_VSI_PROP_EGRESS_UP_VALID));
-
-	if (i40e_is_vsi_uplink_mode_veb(vsi)) {
-		info->valid_sections |=
-				cpu_to_le16(I40E_AQ_VSI_PROP_SWITCH_VALID);
-		info->switch_id =
-				cpu_to_le16(I40E_AQ_VSI_SW_ID_FLAG_ALLOW_LB);
-	}
-	enabled_tc = i40e_get_fcoe_tc_map(pf);
-	i40e_vsi_setup_queue_map(vsi, ctxt, enabled_tc, true);
-
-	/* set up queue option section: only enable FCoE */
-	info->queueing_opt_flags = I40E_AQ_VSI_QUE_OPT_FCOE_ENA;
-
-	return 0;
-}
-
-/**
- * i40e_fcoe_enable - this is the implementation of ndo_fcoe_enable,
- * indicating the upper FCoE protocol stack is ready to use FCoE
- * offload features.
- *
- * @netdev: pointer to the netdev that FCoE is created on
- *
- * Returns 0 on success
- *
- * in RTNL
- *
- **/
-int i40e_fcoe_enable(struct net_device *netdev)
-{
-	struct i40e_netdev_priv *np = netdev_priv(netdev);
-	struct i40e_vsi *vsi = np->vsi;
-	struct i40e_pf *pf = vsi->back;
-	struct i40e_fcoe *fcoe = &pf->fcoe;
-
-	if (!(pf->flags & I40E_FLAG_FCOE_ENABLED)) {
-		netdev_err(netdev, "HW does not support FCoE.\n");
-		return -ENODEV;
-	}
-
-	if (vsi->type != I40E_VSI_FCOE) {
-		netdev_err(netdev, "interface does not support FCoE.\n");
-		return -EBUSY;
-	}
-
-	atomic_inc(&fcoe->refcnt);
-
-	return 0;
-}
-
-/**
- * i40e_fcoe_disable- disables FCoE for upper FCoE protocol stack.
- * @dev: pointer to the netdev that FCoE is created on
- *
- * Returns 0 on success
- *
- **/
-int i40e_fcoe_disable(struct net_device *netdev)
-{
-	struct i40e_netdev_priv *np = netdev_priv(netdev);
-	struct i40e_vsi *vsi = np->vsi;
-	struct i40e_pf *pf = vsi->back;
-	struct i40e_fcoe *fcoe = &pf->fcoe;
-
-	if (!(pf->flags & I40E_FLAG_FCOE_ENABLED)) {
-		netdev_err(netdev, "device does not support FCoE\n");
-		return -ENODEV;
-	}
-	if (vsi->type != I40E_VSI_FCOE)
-		return -EBUSY;
-
-	if (!atomic_dec_and_test(&fcoe->refcnt))
-		return -EINVAL;
-
-	netdev_info(netdev, "FCoE disabled\n");
-
-	return 0;
-}
-
-/**
- * i40e_fcoe_dma_pool_free - free the per cpu pool for FCoE DDP
- * @fcoe: the FCoE sw object
- * @dev: the device that the pool is associated with
- * @cpu: the cpu for this pool
- *
- **/
-static void i40e_fcoe_dma_pool_free(struct i40e_fcoe *fcoe,
-				    struct device *dev,
-				    unsigned int cpu)
-{
-	struct i40e_fcoe_ddp_pool *ddp_pool;
-
-	ddp_pool = per_cpu_ptr(fcoe->ddp_pool, cpu);
-	if (!ddp_pool->pool) {
-		dev_warn(dev, "DDP pool already freed for cpu %d\n", cpu);
-		return;
-	}
-	dma_pool_destroy(ddp_pool->pool);
-	ddp_pool->pool = NULL;
-}
-
-/**
- * i40e_fcoe_dma_pool_create - per cpu pool for FCoE DDP
- * @fcoe: the FCoE sw object
- * @dev: the device that the pool is associated with
- * @cpu: the cpu for this pool
- *
- * Returns 0 on successful or non zero on failure
- *
- **/
-static int i40e_fcoe_dma_pool_create(struct i40e_fcoe *fcoe,
-				     struct device *dev,
-				     unsigned int cpu)
-{
-	struct i40e_fcoe_ddp_pool *ddp_pool;
-	struct dma_pool *pool;
-	char pool_name[32];
-
-	ddp_pool = per_cpu_ptr(fcoe->ddp_pool, cpu);
-	if (ddp_pool && ddp_pool->pool) {
-		dev_warn(dev, "DDP pool already allocated for cpu %d\n", cpu);
-		return 0;
-	}
-	snprintf(pool_name, sizeof(pool_name), "i40e_fcoe_ddp_%d", cpu);
-	pool = dma_pool_create(pool_name, dev, I40E_FCOE_DDP_PTR_MAX,
-			       I40E_FCOE_DDP_PTR_ALIGN, PAGE_SIZE);
-	if (!pool) {
-		dev_err(dev, "dma_pool_create %s failed\n", pool_name);
-		return -ENOMEM;
-	}
-	ddp_pool->pool = pool;
-	return 0;
-}
-
-/**
- * i40e_fcoe_free_ddp_resources - release FCoE DDP resources
- * @vsi: the vsi FCoE is associated with
- *
- **/
-void i40e_fcoe_free_ddp_resources(struct i40e_vsi *vsi)
-{
-	struct i40e_pf *pf = vsi->back;
-	struct i40e_fcoe *fcoe = &pf->fcoe;
-	int cpu, i;
-
-	/* do nothing if not FCoE VSI */
-	if (vsi->type != I40E_VSI_FCOE)
-		return;
-
-	/* do nothing if no DDP pools were allocated */
-	if (!fcoe->ddp_pool)
-		return;
-
-	for (i = 0; i < I40E_FCOE_DDP_MAX; i++)
-		i40e_fcoe_ddp_put(vsi->netdev, i);
-
-	for_each_possible_cpu(cpu)
-		i40e_fcoe_dma_pool_free(fcoe, &pf->pdev->dev, cpu);
-
-	free_percpu(fcoe->ddp_pool);
-	fcoe->ddp_pool = NULL;
-
-	netdev_info(vsi->netdev, "VSI %d,%d FCoE DDP resources released\n",
-		    vsi->id, vsi->seid);
-}
-
-/**
- * i40e_fcoe_setup_ddp_resources - allocate per cpu DDP resources
- * @vsi: the VSI FCoE is associated with
- *
- * Returns 0 on successful or non zero on failure
- *
- **/
-int i40e_fcoe_setup_ddp_resources(struct i40e_vsi *vsi)
-{
-	struct i40e_pf *pf = vsi->back;
-	struct device *dev = &pf->pdev->dev;
-	struct i40e_fcoe *fcoe = &pf->fcoe;
-	unsigned int cpu;
-	int i;
-
-	if (vsi->type != I40E_VSI_FCOE)
-		return -ENODEV;
-
-	/* do nothing if no DDP pools were allocated */
-	if (fcoe->ddp_pool)
-		return -EEXIST;
-
-	/* allocate per CPU memory to track DDP pools */
-	fcoe->ddp_pool = alloc_percpu(struct i40e_fcoe_ddp_pool);
-	if (!fcoe->ddp_pool) {
-		dev_err(&pf->pdev->dev, "failed to allocate percpu DDP\n");
-		return -ENOMEM;
-	}
-
-	/* allocate pci pool for each cpu */
-	for_each_possible_cpu(cpu) {
-		if (!i40e_fcoe_dma_pool_create(fcoe, dev, cpu))
-			continue;
-
-		dev_err(dev, "failed to alloc DDP pool on cpu:%d\n", cpu);
-		i40e_fcoe_free_ddp_resources(vsi);
-		return -ENOMEM;
-	}
-
-	/* initialize the sw context */
-	for (i = 0; i < I40E_FCOE_DDP_MAX; i++)
-		i40e_fcoe_ddp_clear(&fcoe->ddp[i]);
-
-	netdev_info(vsi->netdev, "VSI %d,%d FCoE DDP resources allocated\n",
-		    vsi->id, vsi->seid);
-
-	return 0;
-}
-
-/**
- * i40e_fcoe_handle_status - check the Programming Status for FCoE
- * @rx_ring: the Rx ring for this descriptor
- * @rx_desc: the Rx descriptor for Programming Status, not a packet descriptor.
- *
- * Check if this is the Rx Programming Status descriptor write-back for FCoE.
- * This is used to verify if the context/filter programming or invalidation
- * requested by SW to the HW is successful or not and take actions accordingly.
- **/
-void i40e_fcoe_handle_status(struct i40e_ring *rx_ring,
-			     union i40e_rx_desc *rx_desc, u8 prog_id)
-{
-	struct i40e_pf *pf = rx_ring->vsi->back;
-	struct i40e_fcoe *fcoe = &pf->fcoe;
-	struct i40e_fcoe_ddp *ddp;
-	u32 error;
-	u16 xid;
-	u64 qw;
-
-	/* we only care for FCoE here */
-	if (!i40e_fcoe_progid_is_fcoe(prog_id))
-		return;
-
-	xid = le32_to_cpu(rx_desc->wb.qword0.hi_dword.fcoe_param) &
-	      (I40E_FCOE_DDP_MAX - 1);
-
-	if (!i40e_fcoe_xid_is_valid(xid))
-		return;
-
-	ddp = &fcoe->ddp[xid];
-	WARN_ON(xid != ddp->xid);
-
-	qw = le64_to_cpu(rx_desc->wb.qword1.status_error_len);
-	error = (qw & I40E_RX_PROG_STATUS_DESC_QW1_ERROR_MASK) >>
-		I40E_RX_PROG_STATUS_DESC_QW1_ERROR_SHIFT;
-
-	/* DDP context programming status: failure or success */
-	if (prog_id == I40E_RX_PROG_STATUS_DESC_FCOE_CTXT_PROG_STATUS) {
-		if (I40E_RX_PROG_FCOE_ERROR_TBL_FULL(error)) {
-			dev_err(&pf->pdev->dev, "xid %x ddp->xid %x TABLE FULL\n",
-				xid, ddp->xid);
-			ddp->prerr |= I40E_RX_PROG_FCOE_ERROR_TBL_FULL_BIT;
-		}
-		if (I40E_RX_PROG_FCOE_ERROR_CONFLICT(error)) {
-			dev_err(&pf->pdev->dev, "xid %x ddp->xid %x CONFLICT\n",
-				xid, ddp->xid);
-			ddp->prerr |= I40E_RX_PROG_FCOE_ERROR_CONFLICT_BIT;
-		}
-	}
-
-	/* DDP context invalidation status: failure or success */
-	if (prog_id == I40E_RX_PROG_STATUS_DESC_FCOE_CTXT_INVL_STATUS) {
-		if (I40E_RX_PROG_FCOE_ERROR_INVLFAIL(error)) {
-			dev_err(&pf->pdev->dev, "xid %x ddp->xid %x INVALIDATION FAILURE\n",
-				xid, ddp->xid);
-			ddp->prerr |= I40E_RX_PROG_FCOE_ERROR_INVLFAIL_BIT;
-		}
-		/* clear the flag so we can retry invalidation */
-		clear_bit(__I40E_FCOE_DDP_ABORTED, &ddp->flags);
-	}
-
-	/* unmap DMA */
-	i40e_fcoe_ddp_unmap(pf, ddp);
-	i40e_fcoe_ddp_clear(ddp);
-}
-
-/**
- * i40e_fcoe_handle_offload - check ddp status and mark it done
- * @adapter: i40e adapter
- * @rx_desc: advanced rx descriptor
- * @skb: the skb holding the received data
- *
- * This checks ddp status.
- *
- * Returns : < 0 indicates an error or not a FCOE ddp, 0 indicates
- * not passing the skb to ULD, > 0 indicates is the length of data
- * being ddped.
- *
- **/
-int i40e_fcoe_handle_offload(struct i40e_ring *rx_ring,
-			     union i40e_rx_desc *rx_desc,
-			     struct sk_buff *skb)
-{
-	struct i40e_pf *pf = rx_ring->vsi->back;
-	struct i40e_fcoe *fcoe = &pf->fcoe;
-	struct fc_frame_header *fh = NULL;
-	struct i40e_fcoe_ddp *ddp = NULL;
-	u32 status, fltstat;
-	u32 error, fcerr;
-	int rc = -EINVAL;
-	u16 ptype;
-	u16 xid;
-	u64 qw;
-
-	/* check this rxd is for programming status */
-	qw = le64_to_cpu(rx_desc->wb.qword1.status_error_len);
-	/* packet descriptor, check packet type */
-	ptype = (qw & I40E_RXD_QW1_PTYPE_MASK) >> I40E_RXD_QW1_PTYPE_SHIFT;
-	if (!i40e_rx_is_fcoe(ptype))
-		goto out_no_ddp;
-
-	error = (qw & I40E_RXD_QW1_ERROR_MASK) >> I40E_RXD_QW1_ERROR_SHIFT;
-	fcerr = (error >> I40E_RX_DESC_ERROR_L3L4E_SHIFT) &
-		 I40E_RX_DESC_FCOE_ERROR_MASK;
-
-	/* check stateless offload error */
-	if (unlikely(fcerr == I40E_RX_DESC_ERROR_L3L4E_PROT)) {
-		dev_err(&pf->pdev->dev, "Protocol Error\n");
-		skb->ip_summed = CHECKSUM_NONE;
-	} else {
-		skb->ip_summed = CHECKSUM_UNNECESSARY;
-	}
-
-	/* check hw status on ddp */
-	status = (qw & I40E_RXD_QW1_STATUS_MASK) >> I40E_RXD_QW1_STATUS_SHIFT;
-	fltstat = (status >> I40E_RX_DESC_STATUS_FLTSTAT_SHIFT) &
-		   I40E_RX_DESC_FLTSTAT_FCMASK;
-
-	/* now we are ready to check DDP */
-	fh = i40e_fcoe_fc_frame_header(skb);
-	xid = i40e_fcoe_fc_get_xid(fh);
-	if (!i40e_fcoe_xid_is_valid(xid))
-		goto out_no_ddp;
-
-	/* non DDP normal receive, return to the protocol stack */
-	if (fltstat == I40E_RX_DESC_FLTSTAT_NOMTCH)
-		goto out_no_ddp;
-
-	/* do we have a sw ddp context setup ? */
-	ddp = &fcoe->ddp[xid];
-	if (!ddp->sgl)
-		goto out_no_ddp;
-
-	/* fetch xid from hw rxd wb, which should match up the sw ctxt */
-	xid = le16_to_cpu(rx_desc->wb.qword0.lo_dword.mirr_fcoe.fcoe_ctx_id);
-	if (ddp->xid != xid) {
-		dev_err(&pf->pdev->dev, "xid 0x%x does not match ctx_xid 0x%x\n",
-			ddp->xid, xid);
-		goto out_put_ddp;
-	}
-
-	/* the same exchange has already errored out */
-	if (ddp->fcerr) {
-		dev_err(&pf->pdev->dev, "xid 0x%x fcerr 0x%x reported fcer 0x%x\n",
-			xid, ddp->fcerr, fcerr);
-		goto out_put_ddp;
-	}
-
-	/* fcoe param is valid by now with correct DDPed length */
-	ddp->len = le32_to_cpu(rx_desc->wb.qword0.hi_dword.fcoe_param);
-	ddp->fcerr = fcerr;
-	/* header posting only, useful only for target mode and debugging */
-	if (fltstat == I40E_RX_DESC_FLTSTAT_DDP) {
-		/* For target mode, we get header of the last packet but it
-		 * does not have the FCoE trailer field, i.e., CRC and EOF
-		 * Ordered Set since they are offloaded by the HW, so fill
-		 * it up correspondingly to allow the packet to pass through
-		 * to the upper protocol stack.
-		 */
-		u32 f_ctl = ntoh24(fh->fh_f_ctl);
-
-		if ((f_ctl & FC_FC_END_SEQ) &&
-		    (fh->fh_r_ctl == FC_RCTL_DD_SOL_DATA)) {
-			struct fcoe_crc_eof *crc = NULL;
-
-			crc = (struct fcoe_crc_eof *)skb_put(skb, sizeof(*crc));
-			crc->fcoe_eof = FC_EOF_T;
-		} else {
-			/* otherwise, drop the header only frame */
-			rc = 0;
-			goto out_no_ddp;
-		}
-	}
-
-out_put_ddp:
-	/* either we got RSP or we have an error, unmap DMA in both cases */
-	i40e_fcoe_ddp_unmap(pf, ddp);
-	if (ddp->len && !ddp->fcerr) {
-		int pkts;
-
-		rc = ddp->len;
-		i40e_fcoe_ddp_clear(ddp);
-		ddp->len = rc;
-		pkts = DIV_ROUND_UP(rc, 2048);
-		rx_ring->stats.bytes += rc;
-		rx_ring->stats.packets += pkts;
-		rx_ring->q_vector->rx.total_bytes += rc;
-		rx_ring->q_vector->rx.total_packets += pkts;
-		set_bit(__I40E_FCOE_DDP_DONE, &ddp->flags);
-	}
-
-out_no_ddp:
-	return rc;
-}
-
-/**
- * i40e_fcoe_ddp_setup - called to set up ddp context
- * @netdev: the corresponding net_device
- * @xid: the exchange id requesting ddp
- * @sgl: the scatter-gather list for this request
- * @sgc: the number of scatter-gather items
- * @target_mode: indicates this is a DDP request for target
- *
- * Returns : 1 for success and 0 for no DDP on this I/O
- **/
-static int i40e_fcoe_ddp_setup(struct net_device *netdev, u16 xid,
-			       struct scatterlist *sgl, unsigned int sgc,
-			       int target_mode)
-{
-	static const unsigned int bufflen = I40E_FCOE_DDP_BUF_MIN;
-	struct i40e_netdev_priv *np = netdev_priv(netdev);
-	struct i40e_fcoe_ddp_pool *ddp_pool;
-	struct i40e_pf *pf = np->vsi->back;
-	struct i40e_fcoe *fcoe = &pf->fcoe;
-	unsigned int i, j, dmacount;
-	struct i40e_fcoe_ddp *ddp;
-	unsigned int firstoff = 0;
-	unsigned int thisoff = 0;
-	unsigned int thislen = 0;
-	struct scatterlist *sg;
-	dma_addr_t addr = 0;
-	unsigned int len;
-
-	if (xid >= I40E_FCOE_DDP_MAX) {
-		dev_warn(&pf->pdev->dev, "xid=0x%x out-of-range\n", xid);
-		return 0;
-	}
-
-	/* no DDP if we are already down or resetting */
-	if (test_bit(__I40E_DOWN, &pf->state) ||
-	    test_bit(__I40E_NEEDS_RESTART, &pf->state)) {
-		dev_info(&pf->pdev->dev, "xid=0x%x device in reset/down\n",
-			 xid);
-		return 0;
-	}
-
-	ddp = &fcoe->ddp[xid];
-	if (ddp->sgl) {
-		dev_info(&pf->pdev->dev, "xid 0x%x w/ non-null sgl=%p nents=%d\n",
-			 xid, ddp->sgl, ddp->sgc);
-		return 0;
-	}
-	i40e_fcoe_ddp_clear(ddp);
-
-	if (!fcoe->ddp_pool) {
-		dev_info(&pf->pdev->dev, "No DDP pool, xid 0x%x\n", xid);
-		return 0;
-	}
-
-	ddp_pool = per_cpu_ptr(fcoe->ddp_pool, get_cpu());
-	if (!ddp_pool->pool) {
-		dev_info(&pf->pdev->dev, "No percpu ddp pool, xid 0x%x\n", xid);
-		goto out_noddp;
-	}
-
-	/* setup dma from scsi command sgl */
-	dmacount = dma_map_sg(&pf->pdev->dev, sgl, sgc, DMA_FROM_DEVICE);
-	if (dmacount == 0) {
-		dev_info(&pf->pdev->dev, "dma_map_sg for sgl %p, sgc %d failed\n",
-			 sgl, sgc);
-		goto out_noddp_unmap;
-	}
-
-	/* alloc the udl from our ddp pool */
-	ddp->udl = dma_pool_alloc(ddp_pool->pool, GFP_ATOMIC, &ddp->udp);
-	if (!ddp->udl) {
-		dev_info(&pf->pdev->dev,
-			 "Failed allocated ddp context, xid 0x%x\n", xid);
-		goto out_noddp_unmap;
-	}
-
-	j = 0;
-	ddp->len = 0;
-	for_each_sg(sgl, sg, dmacount, i) {
-		addr = sg_dma_address(sg);
-		len = sg_dma_len(sg);
-		ddp->len += len;
-		while (len) {
-			/* max number of buffers allowed in one DDP context */
-			if (j >= I40E_FCOE_DDP_BUFFCNT_MAX) {
-				dev_info(&pf->pdev->dev,
-					 "xid=%x:%d,%d,%d:addr=%llx not enough descriptors\n",
-					 xid, i, j, dmacount, (u64)addr);
-				goto out_noddp_free;
-			}
-
-			/* get the offset of length of current buffer */
-			thisoff = addr & ((dma_addr_t)bufflen - 1);
-			thislen = min_t(unsigned int, (bufflen - thisoff), len);
-			/* all but the 1st buffer (j == 0)
-			 * must be aligned on bufflen
-			 */
-			if ((j != 0) && (thisoff))
-				goto out_noddp_free;
-
-			/* all but the last buffer
-			 * ((i == (dmacount - 1)) && (thislen == len))
-			 * must end at bufflen
-			 */
-			if (((i != (dmacount - 1)) || (thislen != len)) &&
-			    ((thislen + thisoff) != bufflen))
-				goto out_noddp_free;
-
-			ddp->udl[j] = (u64)(addr - thisoff);
-			/* only the first buffer may have none-zero offset */
-			if (j == 0)
-				firstoff = thisoff;
-			len -= thislen;
-			addr += thislen;
-			j++;
-		}
-	}
-	/* only the last buffer may have non-full bufflen */
-	ddp->lastsize = thisoff + thislen;
-	ddp->firstoff = firstoff;
-	ddp->list_len = j;
-	ddp->pool = ddp_pool->pool;
-	ddp->sgl = sgl;
-	ddp->sgc = sgc;
-	ddp->xid = xid;
-	if (target_mode)
-		set_bit(__I40E_FCOE_DDP_TARGET, &ddp->flags);
-	set_bit(__I40E_FCOE_DDP_INITALIZED, &ddp->flags);
-
-	put_cpu();
-	return 1; /* Success */
-
-out_noddp_free:
-	dma_pool_free(ddp->pool, ddp->udl, ddp->udp);
-	i40e_fcoe_ddp_clear(ddp);
-
-out_noddp_unmap:
-	dma_unmap_sg(&pf->pdev->dev, sgl, sgc, DMA_FROM_DEVICE);
-out_noddp:
-	put_cpu();
-	return 0;
-}
-
-/**
- * i40e_fcoe_ddp_get - called to set up ddp context in initiator mode
- * @netdev: the corresponding net_device
- * @xid: the exchange id requesting ddp
- * @sgl: the scatter-gather list for this request
- * @sgc: the number of scatter-gather items
- *
- * This is the implementation of net_device_ops.ndo_fcoe_ddp_setup
- * and is expected to be called from ULD, e.g., FCP layer of libfc
- * to set up ddp for the corresponding xid of the given sglist for
- * the corresponding I/O.
- *
- * Returns : 1 for success and 0 for no ddp
- **/
-static int i40e_fcoe_ddp_get(struct net_device *netdev, u16 xid,
-			     struct scatterlist *sgl, unsigned int sgc)
-{
-	return i40e_fcoe_ddp_setup(netdev, xid, sgl, sgc, 0);
-}
-
-/**
- * i40e_fcoe_ddp_target - called to set up ddp context in target mode
- * @netdev: the corresponding net_device
- * @xid: the exchange id requesting ddp
- * @sgl: the scatter-gather list for this request
- * @sgc: the number of scatter-gather items
- *
- * This is the implementation of net_device_ops.ndo_fcoe_ddp_target
- * and is expected to be called from ULD, e.g., FCP layer of libfc
- * to set up ddp for the corresponding xid of the given sglist for
- * the corresponding I/O. The DDP in target mode is a write I/O request
- * from the initiator.
- *
- * Returns : 1 for success and 0 for no ddp
- **/
-static int i40e_fcoe_ddp_target(struct net_device *netdev, u16 xid,
-				struct scatterlist *sgl, unsigned int sgc)
-{
-	return i40e_fcoe_ddp_setup(netdev, xid, sgl, sgc, 1);
-}
-
-/**
- * i40e_fcoe_program_ddp - programs the HW DDP related descriptors
- * @tx_ring: transmit ring for this packet
- * @skb:     the packet to be sent out
- * @sof: the SOF to indicate class of service
- *
- * Determine if it is READ/WRITE command, and finds out if there is
- * a matching SW DDP context for this command. DDP is applicable
- * only in case of READ if initiator or WRITE in case of
- * responder (via checking XFER_RDY).
- *
- * Note: caller checks sof and ddp sw context
- *
- * Returns : none
- *
- **/
-static void i40e_fcoe_program_ddp(struct i40e_ring *tx_ring,
-				  struct sk_buff *skb,
-				  struct i40e_fcoe_ddp *ddp, u8 sof)
-{
-	struct i40e_fcoe_filter_context_desc *filter_desc = NULL;
-	struct i40e_fcoe_queue_context_desc *queue_desc = NULL;
-	struct i40e_fcoe_ddp_context_desc *ddp_desc = NULL;
-	struct i40e_pf *pf = tx_ring->vsi->back;
-	u16 i = tx_ring->next_to_use;
-	struct fc_frame_header *fh;
-	u64 flags_rsvd_lanq = 0;
-	bool target_mode;
-
-	/* check if abort is still pending */
-	if (test_bit(__I40E_FCOE_DDP_ABORTED, &ddp->flags)) {
-		dev_warn(&pf->pdev->dev,
-			 "DDP abort is still pending xid:%hx and ddp->flags:%lx:\n",
-			 ddp->xid, ddp->flags);
-		return;
-	}
-
-	/* set the flag to indicate this is programmed */
-	if (test_and_set_bit(__I40E_FCOE_DDP_PROGRAMMED, &ddp->flags)) {
-		dev_warn(&pf->pdev->dev,
-			 "DDP is already programmed for xid:%hx and ddp->flags:%lx:\n",
-			 ddp->xid, ddp->flags);
-		return;
-	}
-
-	/* Prepare the DDP context descriptor */
-	ddp_desc = I40E_DDP_CONTEXT_DESC(tx_ring, i);
-	i++;
-	if (i == tx_ring->count)
-		i = 0;
-
-	ddp_desc->type_cmd_foff_lsize =
-				cpu_to_le64(I40E_TX_DESC_DTYPE_DDP_CTX	|
-				((u64)I40E_FCOE_DDP_CTX_DESC_BSIZE_4K  <<
-				I40E_FCOE_DDP_CTX_QW1_CMD_SHIFT)	|
-				((u64)ddp->firstoff		       <<
-				I40E_FCOE_DDP_CTX_QW1_FOFF_SHIFT)	|
-				((u64)ddp->lastsize		       <<
-				I40E_FCOE_DDP_CTX_QW1_LSIZE_SHIFT));
-	ddp_desc->rsvd = cpu_to_le64(0);
-
-	/* target mode needs last packet in the sequence  */
-	target_mode = test_bit(__I40E_FCOE_DDP_TARGET, &ddp->flags);
-	if (target_mode)
-		ddp_desc->type_cmd_foff_lsize |=
-			cpu_to_le64(I40E_FCOE_DDP_CTX_DESC_LASTSEQH);
-
-	/* Prepare queue_context descriptor */
-	queue_desc = I40E_QUEUE_CONTEXT_DESC(tx_ring, i++);
-	if (i == tx_ring->count)
-		i = 0;
-	queue_desc->dmaindx_fbase = cpu_to_le64(ddp->xid | ((u64)ddp->udp));
-	queue_desc->flen_tph = cpu_to_le64(ddp->list_len |
-				((u64)(I40E_FCOE_QUEUE_CTX_DESC_TPHRDESC |
-				I40E_FCOE_QUEUE_CTX_DESC_TPHDATA) <<
-				I40E_FCOE_QUEUE_CTX_QW1_TPH_SHIFT));
-
-	/* Prepare filter_context_desc */
-	filter_desc = I40E_FILTER_CONTEXT_DESC(tx_ring, i);
-	i++;
-	if (i == tx_ring->count)
-		i = 0;
-
-	fh = (struct fc_frame_header *)skb_transport_header(skb);
-	filter_desc->param = cpu_to_le32(ntohl(fh->fh_parm_offset));
-	filter_desc->seqn = cpu_to_le16(ntohs(fh->fh_seq_cnt));
-	filter_desc->rsvd_dmaindx = cpu_to_le16(ddp->xid <<
-				I40E_FCOE_FILTER_CTX_QW0_DMAINDX_SHIFT);
-
-	flags_rsvd_lanq = I40E_FCOE_FILTER_CTX_DESC_CTYP_DDP;
-	flags_rsvd_lanq |= (u64)(target_mode ?
-			I40E_FCOE_FILTER_CTX_DESC_ENODE_RSP :
-			I40E_FCOE_FILTER_CTX_DESC_ENODE_INIT);
-
-	flags_rsvd_lanq |= (u64)((sof == FC_SOF_I2 || sof == FC_SOF_N2) ?
-			I40E_FCOE_FILTER_CTX_DESC_FC_CLASS2 :
-			I40E_FCOE_FILTER_CTX_DESC_FC_CLASS3);
-
-	flags_rsvd_lanq |= ((u64)skb->queue_mapping <<
-				I40E_FCOE_FILTER_CTX_QW1_LANQINDX_SHIFT);
-	filter_desc->flags_rsvd_lanq = cpu_to_le64(flags_rsvd_lanq);
-
-	/* By this time, all offload related descriptors has been programmed */
-	tx_ring->next_to_use = i;
-}
-
-/**
- * i40e_fcoe_invalidate_ddp - invalidates DDP in case of abort
- * @tx_ring: transmit ring for this packet
- * @skb: the packet associated w/ this DDP invalidation, i.e., ABTS
- * @ddp: the SW DDP context for this DDP
- *
- * Programs the Tx context descriptor to do DDP invalidation.
- **/
-static void i40e_fcoe_invalidate_ddp(struct i40e_ring *tx_ring,
-				     struct sk_buff *skb,
-				     struct i40e_fcoe_ddp *ddp)
-{
-	struct i40e_tx_context_desc *context_desc;
-	int i;
-
-	if (test_and_set_bit(__I40E_FCOE_DDP_ABORTED, &ddp->flags))
-		return;
-
-	i = tx_ring->next_to_use;
-	context_desc = I40E_TX_CTXTDESC(tx_ring, i);
-	i++;
-	if (i == tx_ring->count)
-		i = 0;
-
-	context_desc->tunneling_params = cpu_to_le32(0);
-	context_desc->l2tag2 = cpu_to_le16(0);
-	context_desc->rsvd = cpu_to_le16(0);
-	context_desc->type_cmd_tso_mss = cpu_to_le64(
-		I40E_TX_DESC_DTYPE_FCOE_CTX |
-		(I40E_FCOE_TX_CTX_DESC_OPCODE_DDP_CTX_INVL <<
-		I40E_TXD_CTX_QW1_CMD_SHIFT) |
-		(I40E_FCOE_TX_CTX_DESC_OPCODE_SINGLE_SEND <<
-		I40E_TXD_CTX_QW1_CMD_SHIFT));
-	tx_ring->next_to_use = i;
-}
-
-/**
- * i40e_fcoe_handle_ddp - check we should setup or invalidate DDP
- * @tx_ring: transmit ring for this packet
- * @skb: the packet to be sent out
- * @sof: the SOF to indicate class of service
- *
- * Determine if it is ABTS/READ/XFER_RDY, and finds out if there is
- * a matching SW DDP context for this command. DDP is applicable
- * only in case of READ if initiator or WRITE in case of
- * responder (via checking XFER_RDY). In case this is an ABTS, send
- * just invalidate the context.
- **/
-static void i40e_fcoe_handle_ddp(struct i40e_ring *tx_ring,
-				 struct sk_buff *skb, u8 sof)
-{
-	struct i40e_pf *pf = tx_ring->vsi->back;
-	struct i40e_fcoe *fcoe = &pf->fcoe;
-	struct fc_frame_header *fh;
-	struct i40e_fcoe_ddp *ddp;
-	u32 f_ctl;
-	u8 r_ctl;
-	u16 xid;
-
-	fh = (struct fc_frame_header *)skb_transport_header(skb);
-	f_ctl = ntoh24(fh->fh_f_ctl);
-	r_ctl = fh->fh_r_ctl;
-	ddp = NULL;
-
-	if ((r_ctl == FC_RCTL_DD_DATA_DESC) && (f_ctl & FC_FC_EX_CTX)) {
-		/* exchange responder? if so, XFER_RDY for write */
-		xid = ntohs(fh->fh_rx_id);
-		if (i40e_fcoe_xid_is_valid(xid)) {
-			ddp = &fcoe->ddp[xid];
-			if ((ddp->xid == xid) &&
-			    (test_bit(__I40E_FCOE_DDP_TARGET, &ddp->flags)))
-				i40e_fcoe_program_ddp(tx_ring, skb, ddp, sof);
-		}
-	} else if (r_ctl == FC_RCTL_DD_UNSOL_CMD) {
-		/* exchange originator, check READ cmd */
-		xid = ntohs(fh->fh_ox_id);
-		if (i40e_fcoe_xid_is_valid(xid)) {
-			ddp = &fcoe->ddp[xid];
-			if ((ddp->xid == xid) &&
-			    (!test_bit(__I40E_FCOE_DDP_TARGET, &ddp->flags)))
-				i40e_fcoe_program_ddp(tx_ring, skb, ddp, sof);
-		}
-	} else if (r_ctl == FC_RCTL_BA_ABTS) {
-		/* exchange originator, check ABTS */
-		xid = ntohs(fh->fh_ox_id);
-		if (i40e_fcoe_xid_is_valid(xid)) {
-			ddp = &fcoe->ddp[xid];
-			if ((ddp->xid == xid) &&
-			    (!test_bit(__I40E_FCOE_DDP_TARGET, &ddp->flags)))
-				i40e_fcoe_invalidate_ddp(tx_ring, skb, ddp);
-		}
-	}
-}
-
-/**
- * i40e_fcoe_tso - set up FCoE TSO
- * @tx_ring:  ring to send buffer on
- * @skb:      send buffer
- * @tx_flags: collected send information
- * @hdr_len:  the tso header length
- * @sof: the SOF to indicate class of service
- *
- * Note must already have sof checked to be either class 2 or class 3 before
- * calling this function.
- *
- * Returns 1 to indicate sequence segmentation offload is properly setup
- * or returns 0 to indicate no tso is needed, otherwise returns error
- * code to drop the frame.
- **/
-static int i40e_fcoe_tso(struct i40e_ring *tx_ring,
-			 struct sk_buff *skb,
-			 u32 tx_flags, u8 *hdr_len, u8 sof)
-{
-	struct i40e_tx_context_desc *context_desc;
-	u32 cd_type, cd_cmd, cd_tso_len, cd_mss;
-	struct fc_frame_header *fh;
-	u64 cd_type_cmd_tso_mss;
-
-	/* must match gso type as FCoE */
-	if (!skb_is_gso(skb))
-		return 0;
-
-	/* is it the expected gso type for FCoE ?*/
-	if (skb_shinfo(skb)->gso_type != SKB_GSO_FCOE) {
-		netdev_err(skb->dev,
-			   "wrong gso type %d:expecting SKB_GSO_FCOE\n",
-			   skb_shinfo(skb)->gso_type);
-		return -EINVAL;
-	}
-
-	/* header and trailer are inserted by hw */
-	*hdr_len = skb_transport_offset(skb) + sizeof(struct fc_frame_header) +
-		   sizeof(struct fcoe_crc_eof);
-
-	/* check sof to decide a class 2 or 3 TSO */
-	if (likely(i40e_fcoe_sof_is_class3(sof)))
-		cd_cmd = I40E_FCOE_TX_CTX_DESC_OPCODE_TSO_FC_CLASS3;
-	else
-		cd_cmd = I40E_FCOE_TX_CTX_DESC_OPCODE_TSO_FC_CLASS2;
-
-	/* param field valid? */
-	fh = (struct fc_frame_header *)skb_transport_header(skb);
-	if (fh->fh_f_ctl[2] & FC_FC_REL_OFF)
-		cd_cmd |= I40E_FCOE_TX_CTX_DESC_RELOFF;
-
-	/* fill the field values */
-	cd_type = I40E_TX_DESC_DTYPE_FCOE_CTX;
-	cd_tso_len = skb->len - *hdr_len;
-	cd_mss = skb_shinfo(skb)->gso_size;
-	cd_type_cmd_tso_mss =
-		((u64)cd_type  << I40E_TXD_CTX_QW1_DTYPE_SHIFT)     |
-		((u64)cd_cmd     << I40E_TXD_CTX_QW1_CMD_SHIFT)	    |
-		((u64)cd_tso_len << I40E_TXD_CTX_QW1_TSO_LEN_SHIFT) |
-		((u64)cd_mss     << I40E_TXD_CTX_QW1_MSS_SHIFT);
-
-	/* grab the next descriptor */
-	context_desc = I40E_TX_CTXTDESC(tx_ring, tx_ring->next_to_use);
-	tx_ring->next_to_use++;
-	if (tx_ring->next_to_use == tx_ring->count)
-		tx_ring->next_to_use = 0;
-
-	context_desc->tunneling_params = 0;
-	context_desc->l2tag2 = cpu_to_le16((tx_flags & I40E_TX_FLAGS_VLAN_MASK)
-					    >> I40E_TX_FLAGS_VLAN_SHIFT);
-	context_desc->type_cmd_tso_mss = cpu_to_le64(cd_type_cmd_tso_mss);
-
-	return 1;
-}
-
-/**
- * i40e_fcoe_tx_map - build the tx descriptor
- * @tx_ring:  ring to send buffer on
- * @skb:      send buffer
- * @first:    first buffer info buffer to use
- * @tx_flags: collected send information
- * @hdr_len:  ptr to the size of the packet header
- * @eof:      the frame eof value
- *
- * Note, for FCoE, sof and eof are already checked
- **/
-static void i40e_fcoe_tx_map(struct i40e_ring *tx_ring,
-			     struct sk_buff *skb,
-			     struct i40e_tx_buffer *first,
-			     u32 tx_flags, u8 hdr_len, u8 eof)
-{
-	u32 td_offset = 0;
-	u32 td_cmd = 0;
-	u32 maclen;
-
-	/* insert CRC */
-	td_cmd = I40E_TX_DESC_CMD_ICRC;
-
-	/* setup MACLEN */
-	maclen = skb_network_offset(skb);
-	if (tx_flags & I40E_TX_FLAGS_SW_VLAN)
-		maclen += sizeof(struct vlan_hdr);
-
-	if (skb->protocol == htons(ETH_P_FCOE)) {
-		/* for FCoE, maclen should exclude ether type */
-		maclen -= 2;
-		/* setup type as FCoE and EOF insertion */
-		td_cmd |= (I40E_TX_DESC_CMD_FCOET | i40e_fcoe_ctxt_eof(eof));
-		/* setup FCoELEN and FCLEN */
-		td_offset |= ((((sizeof(struct fcoe_hdr) + 2) >> 2) <<
-				I40E_TX_DESC_LENGTH_IPLEN_SHIFT) |
-			      ((sizeof(struct fc_frame_header) >> 2) <<
-				I40E_TX_DESC_LENGTH_L4_FC_LEN_SHIFT));
-		/* trim to exclude trailer */
-		pskb_trim(skb, skb->len - sizeof(struct fcoe_crc_eof));
-	}
-
-	/* MACLEN is ether header length in words not bytes */
-	td_offset |= (maclen >> 1) << I40E_TX_DESC_LENGTH_MACLEN_SHIFT;
-
-	i40e_tx_map(tx_ring, skb, first, tx_flags, hdr_len, td_cmd, td_offset);
-}
-
-/**
- * i40e_fcoe_set_skb_header - adjust skb header point for FIP/FCoE/FC
- * @skb: the skb to be adjusted
- *
- * Returns true if this skb is a FCoE/FIP or VLAN carried FCoE/FIP and then
- * adjusts the skb header pointers correspondingly. Otherwise, returns false.
- **/
-static inline int i40e_fcoe_set_skb_header(struct sk_buff *skb)
-{
-	__be16 protocol = skb->protocol;
-
-	skb_reset_mac_header(skb);
-	skb->mac_len = sizeof(struct ethhdr);
-	if (protocol == htons(ETH_P_8021Q)) {
-		struct vlan_ethhdr *veth = (struct vlan_ethhdr *)eth_hdr(skb);
-
-		protocol = veth->h_vlan_encapsulated_proto;
-		skb->mac_len += sizeof(struct vlan_hdr);
-	}
-
-	/* FCoE or FIP only */
-	if ((protocol != htons(ETH_P_FIP)) &&
-	    (protocol != htons(ETH_P_FCOE)))
-		return -EINVAL;
-
-	/* set header to L2 of FCoE/FIP */
-	skb_set_network_header(skb, skb->mac_len);
-	if (protocol == htons(ETH_P_FIP))
-		return 0;
-
-	/* set header to L3 of FC */
-	skb_set_transport_header(skb, skb->mac_len + sizeof(struct fcoe_hdr));
-	return 0;
-}
-
-/**
- * i40e_fcoe_xmit_frame - transmit buffer
- * @skb:     send buffer
- * @netdev:  the fcoe netdev
- *
- * Returns 0 if sent, else an error code
- **/
-static netdev_tx_t i40e_fcoe_xmit_frame(struct sk_buff *skb,
-					struct net_device *netdev)
-{
-	struct i40e_netdev_priv *np = netdev_priv(skb->dev);
-	struct i40e_vsi *vsi = np->vsi;
-	struct i40e_ring *tx_ring = vsi->tx_rings[skb->queue_mapping];
-	struct i40e_tx_buffer *first;
-	u32 tx_flags = 0;
-	int fso, count;
-	u8 hdr_len = 0;
-	u8 sof = 0;
-	u8 eof = 0;
-
-	if (i40e_fcoe_set_skb_header(skb))
-		goto out_drop;
-
-	count = i40e_xmit_descriptor_count(skb);
-	if (i40e_chk_linearize(skb, count)) {
-		if (__skb_linearize(skb))
-			goto out_drop;
-		count = i40e_txd_use_count(skb->len);
-		tx_ring->tx_stats.tx_linearize++;
-	}
-
-	/* need: 1 descriptor per page * PAGE_SIZE/I40E_MAX_DATA_PER_TXD,
-	 *       + 1 desc for skb_head_len/I40E_MAX_DATA_PER_TXD,
-	 *       + 4 desc gap to avoid the cache line where head is,
-	 *       + 1 desc for context descriptor,
-	 * otherwise try next time
-	 */
-	if (i40e_maybe_stop_tx(tx_ring, count + 4 + 1)) {
-		tx_ring->tx_stats.tx_busy++;
-		return NETDEV_TX_BUSY;
-	}
-
-	/* prepare the xmit flags */
-	if (i40e_tx_prepare_vlan_flags(skb, tx_ring, &tx_flags))
-		goto out_drop;
-
-	/* record the location of the first descriptor for this packet */
-	first = &tx_ring->tx_bi[tx_ring->next_to_use];
-
-	/* FIP is a regular L2 traffic w/o offload */
-	if (skb->protocol == htons(ETH_P_FIP))
-		goto out_send;
-
-	/* check sof and eof, only supports FC Class 2 or 3 */
-	if (i40e_fcoe_fc_sof(skb, &sof) || i40e_fcoe_fc_eof(skb, &eof)) {
-		netdev_err(netdev, "SOF/EOF error:%02x - %02x\n", sof, eof);
-		goto out_drop;
-	}
-
-	/* always do FCCRC for FCoE */
-	tx_flags |= I40E_TX_FLAGS_FCCRC;
-
-	/* check we should do sequence offload */
-	fso = i40e_fcoe_tso(tx_ring, skb, tx_flags, &hdr_len, sof);
-	if (fso < 0)
-		goto out_drop;
-	else if (fso)
-		tx_flags |= I40E_TX_FLAGS_FSO;
-	else
-		i40e_fcoe_handle_ddp(tx_ring, skb, sof);
-
-out_send:
-	/* send out the packet */
-	i40e_fcoe_tx_map(tx_ring, skb, first, tx_flags, hdr_len, eof);
-
-	i40e_maybe_stop_tx(tx_ring, DESC_NEEDED);
-	return NETDEV_TX_OK;
-
-out_drop:
-	dev_kfree_skb_any(skb);
-	return NETDEV_TX_OK;
-}
-
-/**
- * i40e_fcoe_change_mtu - NDO callback to change the Maximum Transfer Unit
- * @netdev: network interface device structure
- * @new_mtu: new value for maximum frame size
- *
- * Returns error as operation not permitted
- *
- **/
-static int i40e_fcoe_change_mtu(struct net_device *netdev, int new_mtu)
-{
-	netdev_warn(netdev, "MTU change is not supported on FCoE interfaces\n");
-	return -EPERM;
-}
-
-/**
- * i40e_fcoe_set_features - set the netdev feature flags
- * @netdev: ptr to the netdev being adjusted
- * @features: the feature set that the stack is suggesting
- *
- **/
-static int i40e_fcoe_set_features(struct net_device *netdev,
-				  netdev_features_t features)
-{
-	struct i40e_netdev_priv *np = netdev_priv(netdev);
-	struct i40e_vsi *vsi = np->vsi;
-
-	if (features & NETIF_F_HW_VLAN_CTAG_RX)
-		i40e_vlan_stripping_enable(vsi);
-	else
-		i40e_vlan_stripping_disable(vsi);
-
-	return 0;
-}
-
-static const struct net_device_ops i40e_fcoe_netdev_ops = {
-	.ndo_open		= i40e_open,
-	.ndo_stop		= i40e_close,
-	.ndo_get_stats64	= i40e_get_netdev_stats_struct,
-	.ndo_set_rx_mode	= i40e_set_rx_mode,
-	.ndo_validate_addr	= eth_validate_addr,
-	.ndo_set_mac_address	= i40e_set_mac,
-	.ndo_change_mtu		= i40e_fcoe_change_mtu,
-	.ndo_do_ioctl		= i40e_ioctl,
-	.ndo_tx_timeout		= i40e_tx_timeout,
-	.ndo_vlan_rx_add_vid	= i40e_vlan_rx_add_vid,
-	.ndo_vlan_rx_kill_vid	= i40e_vlan_rx_kill_vid,
-	.ndo_setup_tc		= __i40e_setup_tc,
-
-#ifdef CONFIG_NET_POLL_CONTROLLER
-	.ndo_poll_controller	= i40e_netpoll,
-#endif
-	.ndo_start_xmit		= i40e_fcoe_xmit_frame,
-	.ndo_fcoe_enable	= i40e_fcoe_enable,
-	.ndo_fcoe_disable	= i40e_fcoe_disable,
-	.ndo_fcoe_ddp_setup	= i40e_fcoe_ddp_get,
-	.ndo_fcoe_ddp_done	= i40e_fcoe_ddp_put,
-	.ndo_fcoe_ddp_target	= i40e_fcoe_ddp_target,
-	.ndo_set_features	= i40e_fcoe_set_features,
-};
-
-/* fcoe network device type */
-static struct device_type fcoe_netdev_type = {
-	.name = "fcoe",
-};
-
-/**
- * i40e_fcoe_config_netdev - prepares the VSI context for creating a FCoE VSI
- * @vsi: pointer to the associated VSI struct
- * @ctxt: pointer to the associated VSI context to be passed to HW
- *
- * Returns 0 on success or < 0 on error
- **/
-void i40e_fcoe_config_netdev(struct net_device *netdev, struct i40e_vsi *vsi)
-{
-	struct i40e_hw *hw = &vsi->back->hw;
-	struct i40e_pf *pf = vsi->back;
-
-	if (vsi->type != I40E_VSI_FCOE)
-		return;
-
-	netdev->features = (NETIF_F_HW_VLAN_CTAG_TX |
-			    NETIF_F_HW_VLAN_CTAG_RX |
-			    NETIF_F_HW_VLAN_CTAG_FILTER);
-
-	netdev->vlan_features = netdev->features;
-	netdev->vlan_features &= ~(NETIF_F_HW_VLAN_CTAG_TX |
-				   NETIF_F_HW_VLAN_CTAG_RX |
-				   NETIF_F_HW_VLAN_CTAG_FILTER);
-	netdev->fcoe_ddp_xid = I40E_FCOE_DDP_MAX - 1;
-	netdev->features |= NETIF_F_ALL_FCOE;
-	netdev->vlan_features |= NETIF_F_ALL_FCOE;
-	netdev->hw_features |= netdev->features;
-	netdev->priv_flags |= IFF_UNICAST_FLT;
-	netdev->priv_flags |= IFF_SUPP_NOFCS;
-
-	strlcpy(netdev->name, "fcoe%d", IFNAMSIZ-1);
-	netdev->mtu = FCOE_MTU;
-	SET_NETDEV_DEV(netdev, &pf->pdev->dev);
-	SET_NETDEV_DEVTYPE(netdev, &fcoe_netdev_type);
-	/* set different dev_port value 1 for FCoE netdev than the default
-	 * zero dev_port value for PF netdev, this helps biosdevname user
-	 * tool to differentiate them correctly while both attached to the
-	 * same PCI function.
-	 */
-	netdev->dev_port = 1;
-	spin_lock_bh(&vsi->mac_filter_hash_lock);
-	i40e_add_filter(vsi, hw->mac.san_addr, 0);
-	i40e_add_filter(vsi, (u8[6]) FC_FCOE_FLOGI_MAC, 0);
-	i40e_add_filter(vsi, FIP_ALL_FCOE_MACS, 0);
-	i40e_add_filter(vsi, FIP_ALL_ENODE_MACS, 0);
-	spin_unlock_bh(&vsi->mac_filter_hash_lock);
-
-	/* use san mac */
-	ether_addr_copy(netdev->dev_addr, hw->mac.san_addr);
-	ether_addr_copy(netdev->perm_addr, hw->mac.san_addr);
-	/* fcoe netdev ops */
-	netdev->netdev_ops = &i40e_fcoe_netdev_ops;
-}
-
-/**
- * i40e_fcoe_vsi_setup - allocate and set up FCoE VSI
- * @pf: the PF that VSI is associated with
- *
- **/
-void i40e_fcoe_vsi_setup(struct i40e_pf *pf)
-{
-	struct i40e_vsi *vsi;
-	u16 seid;
-	int i;
-
-	if (!(pf->flags & I40E_FLAG_FCOE_ENABLED))
-		return;
-
-	for (i = 0; i < pf->num_alloc_vsi; i++) {
-		vsi = pf->vsi[i];
-		if (vsi && vsi->type == I40E_VSI_FCOE) {
-			dev_warn(&pf->pdev->dev,
-				 "FCoE VSI already created\n");
-			return;
-		}
-	}
-
-	seid = pf->vsi[pf->lan_vsi]->seid;
-	vsi = i40e_vsi_setup(pf, I40E_VSI_FCOE, seid, 0);
-	if (vsi) {
-		dev_dbg(&pf->pdev->dev,
-			"Successfully created FCoE VSI seid %d id %d uplink_seid %d PF seid %d\n",
-			vsi->seid, vsi->id, vsi->uplink_seid, seid);
-	} else {
-		dev_info(&pf->pdev->dev, "Failed to create FCoE VSI\n");
-	}
-}
diff --git a/drivers/net/ethernet/intel/i40e/i40e_fcoe.h b/drivers/net/ethernet/intel/i40e/i40e_fcoe.h
deleted file mode 100644
index a93174ddeaba..000000000000
--- a/drivers/net/ethernet/intel/i40e/i40e_fcoe.h
+++ /dev/null
@@ -1,127 +0,0 @@
-/*******************************************************************************
- *
- * Intel Ethernet Controller XL710 Family Linux Driver
- * Copyright(c) 2013 - 2014 Intel Corporation.
- *
- * This program is free software; you can redistribute it and/or modify it
- * under the terms and conditions of the GNU General Public License,
- * version 2, as published by the Free Software Foundation.
- *
- * This program is distributed in the hope it will be useful, but WITHOUT
- * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
- * FITNESS FOR A PARTICULAR PURPOSE.  See the GNU General Public License for
- * more details.
- *
- * You should have received a copy of the GNU General Public License along
- * with this program.  If not, see <http://www.gnu.org/licenses/>.
- *
- * The full GNU General Public License is included in this distribution in
- * the file called "COPYING".
- *
- * Contact Information:
- * e1000-devel Mailing List <e1000-devel@lists.sourceforge.net>
- * Intel Corporation, 5200 N.E. Elam Young Parkway, Hillsboro, OR 97124-6497
- *
- ******************************************************************************/
-
-#ifndef _I40E_FCOE_H_
-#define _I40E_FCOE_H_
-
-/* FCoE HW context helper macros */
-#define I40E_DDP_CONTEXT_DESC(R, i)     \
-	(&(((struct i40e_fcoe_ddp_context_desc *)((R)->desc))[i]))
-
-#define I40E_QUEUE_CONTEXT_DESC(R, i)   \
-	(&(((struct i40e_fcoe_queue_context_desc *)((R)->desc))[i]))
-
-#define I40E_FILTER_CONTEXT_DESC(R, i)  \
-	(&(((struct i40e_fcoe_filter_context_desc *)((R)->desc))[i]))
-
-/* receive queue descriptor filter status for FCoE */
-#define I40E_RX_DESC_FLTSTAT_FCMASK	0x3
-#define I40E_RX_DESC_FLTSTAT_NOMTCH	0x0	/* no ddp context match */
-#define I40E_RX_DESC_FLTSTAT_NODDP	0x1	/* no ddp due to error */
-#define I40E_RX_DESC_FLTSTAT_DDP	0x2	/* DDPed payload, post header */
-#define I40E_RX_DESC_FLTSTAT_FCPRSP	0x3	/* FCP_RSP */
-
-/* receive queue descriptor error codes for FCoE */
-#define I40E_RX_DESC_FCOE_ERROR_MASK		\
-	(I40E_RX_DESC_ERROR_L3L4E_PROT |	\
-	 I40E_RX_DESC_ERROR_L3L4E_FC |		\
-	 I40E_RX_DESC_ERROR_L3L4E_DMAC_ERR |	\
-	 I40E_RX_DESC_ERROR_L3L4E_DMAC_WARN)
-
-/* receive queue descriptor programming error */
-#define I40E_RX_PROG_FCOE_ERROR_TBL_FULL(e)	\
-	(((e) >> I40E_RX_PROG_STATUS_DESC_FCOE_TBL_FULL_SHIFT) & 0x1)
-
-#define I40E_RX_PROG_FCOE_ERROR_CONFLICT(e)	\
-	(((e) >> I40E_RX_PROG_STATUS_DESC_FCOE_CONFLICT_SHIFT) & 0x1)
-
-#define I40E_RX_PROG_FCOE_ERROR_TBL_FULL_BIT	\
-	BIT(I40E_RX_PROG_STATUS_DESC_FCOE_TBL_FULL_SHIFT)
-#define I40E_RX_PROG_FCOE_ERROR_CONFLICT_BIT	\
-	BIT(I40E_RX_PROG_STATUS_DESC_FCOE_CONFLICT_SHIFT)
-
-#define I40E_RX_PROG_FCOE_ERROR_INVLFAIL(e)	\
-	I40E_RX_PROG_FCOE_ERROR_CONFLICT(e)
-#define I40E_RX_PROG_FCOE_ERROR_INVLFAIL_BIT	\
-	I40E_RX_PROG_FCOE_ERROR_CONFLICT_BIT
-
-/* FCoE DDP related definitions */
-#define I40E_FCOE_MIN_XID	0x0000  /* the min xid supported by fcoe_sw */
-#define I40E_FCOE_MAX_XID	0x0FFF  /* the max xid supported by fcoe_sw */
-#define I40E_FCOE_DDP_BUFFCNT_MAX	512	/* 9 bits bufcnt */
-#define I40E_FCOE_DDP_PTR_ALIGN		16
-#define I40E_FCOE_DDP_PTR_MAX	(I40E_FCOE_DDP_BUFFCNT_MAX * sizeof(dma_addr_t))
-#define I40E_FCOE_DDP_BUF_MIN	4096
-#define I40E_FCOE_DDP_MAX	2048
-#define I40E_FCOE_FILTER_CTX_QW1_PCTYPE_SHIFT	8
-
-/* supported netdev features for FCoE */
-#define I40E_FCOE_NETIF_FEATURES (NETIF_F_ALL_FCOE | \
-	NETIF_F_HW_VLAN_CTAG_TX | \
-	NETIF_F_HW_VLAN_CTAG_RX | \
-	NETIF_F_HW_VLAN_CTAG_FILTER)
-
-/* DDP context flags */
-enum i40e_fcoe_ddp_flags {
-	__I40E_FCOE_DDP_NONE = 1,
-	__I40E_FCOE_DDP_TARGET,
-	__I40E_FCOE_DDP_INITALIZED,
-	__I40E_FCOE_DDP_PROGRAMMED,
-	__I40E_FCOE_DDP_DONE,
-	__I40E_FCOE_DDP_ABORTED,
-	__I40E_FCOE_DDP_UNMAPPED,
-};
-
-/* DDP SW context struct */
-struct i40e_fcoe_ddp {
-	int len;
-	u16 xid;
-	u16 firstoff;
-	u16 lastsize;
-	u16 list_len;
-	u8 fcerr;
-	u8 prerr;
-	unsigned long flags;
-	unsigned int sgc;
-	struct scatterlist *sgl;
-	dma_addr_t udp;
-	u64 *udl;
-	struct dma_pool *pool;
-
-};
-
-struct i40e_fcoe_ddp_pool {
-	struct dma_pool *pool;
-};
-
-struct i40e_fcoe {
-	unsigned long mode;
-	atomic_t refcnt;
-	struct i40e_fcoe_ddp_pool __percpu *ddp_pool;
-	struct i40e_fcoe_ddp ddp[I40E_FCOE_DDP_MAX];
-};
-
-#endif /* _I40E_FCOE_H_ */
diff --git a/drivers/net/ethernet/intel/i40e/i40e_hmc.c b/drivers/net/ethernet/intel/i40e/i40e_hmc.c
index a7c7b1d9b7c8..1742624ca62e 100644
--- a/drivers/net/ethernet/intel/i40e/i40e_hmc.c
+++ b/drivers/net/ethernet/intel/i40e/i40e_hmc.c
@@ -1,33 +1,8 @@
-/*******************************************************************************
- *
- * Intel Ethernet Controller XL710 Family Linux Driver
- * Copyright(c) 2013 - 2014 Intel Corporation.
- *
- * This program is free software; you can redistribute it and/or modify it
- * under the terms and conditions of the GNU General Public License,
- * version 2, as published by the Free Software Foundation.
- *
- * This program is distributed in the hope it will be useful, but WITHOUT
- * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
- * FITNESS FOR A PARTICULAR PURPOSE.  See the GNU General Public License for
- * more details.
- *
- * You should have received a copy of the GNU General Public License along
- * with this program.  If not, see <http://www.gnu.org/licenses/>.
- *
- * The full GNU General Public License is included in this distribution in
- * the file called "COPYING".
- *
- * Contact Information:
- * e1000-devel Mailing List <e1000-devel@lists.sourceforge.net>
- * Intel Corporation, 5200 N.E. Elam Young Parkway, Hillsboro, OR 97124-6497
- *
- ******************************************************************************/
+// SPDX-License-Identifier: GPL-2.0
+/* Copyright(c) 2013 - 2018 Intel Corporation. */
 
-#include "i40e_osdep.h"
-#include "i40e_register.h"
-#include "i40e_status.h"
 #include "i40e_alloc.h"
+#include "i40e_debug.h"
 #include "i40e_hmc.h"
 #include "i40e_type.h"
 
@@ -39,43 +14,39 @@
  * @type: what type of segment descriptor we're manipulating
  * @direct_mode_sz: size to alloc in direct mode
  **/
-i40e_status i40e_add_sd_table_entry(struct i40e_hw *hw,
-					      struct i40e_hmc_info *hmc_info,
-					      u32 sd_index,
-					      enum i40e_sd_entry_type type,
-					      u64 direct_mode_sz)
+int i40e_add_sd_table_entry(struct i40e_hw *hw,
+			    struct i40e_hmc_info *hmc_info,
+			    u32 sd_index,
+			    enum i40e_sd_entry_type type,
+			    u64 direct_mode_sz)
 {
-	enum i40e_memory_type mem_type __attribute__((unused));
 	struct i40e_hmc_sd_entry *sd_entry;
 	bool dma_mem_alloc_done = false;
 	struct i40e_dma_mem mem;
-	i40e_status ret_code = I40E_SUCCESS;
+	int ret_code = 0;
 	u64 alloc_len;
 
 	if (NULL == hmc_info->sd_table.sd_entry) {
-		ret_code = I40E_ERR_BAD_PTR;
+		ret_code = -EINVAL;
 		hw_dbg(hw, "i40e_add_sd_table_entry: bad sd_entry\n");
 		goto exit;
 	}
 
 	if (sd_index >= hmc_info->sd_table.sd_cnt) {
-		ret_code = I40E_ERR_INVALID_SD_INDEX;
+		ret_code = -EINVAL;
 		hw_dbg(hw, "i40e_add_sd_table_entry: bad sd_index\n");
 		goto exit;
 	}
 
 	sd_entry = &hmc_info->sd_table.sd_entry[sd_index];
 	if (!sd_entry->valid) {
-		if (I40E_SD_TYPE_PAGED == type) {
-			mem_type = i40e_mem_pd;
+		if (type == I40E_SD_TYPE_PAGED)
 			alloc_len = I40E_HMC_PAGED_BP_SIZE;
-		} else {
-			mem_type = i40e_mem_bp_jumbo;
+		else
 			alloc_len = direct_mode_sz;
-		}
 
 		/* allocate a 4K pd page or 2M backing page */
-		ret_code = i40e_allocate_dma_mem(hw, &mem, mem_type, alloc_len,
+		ret_code = i40e_allocate_dma_mem(hw, &mem, alloc_len,
 						 I40E_HMC_PD_BP_BUF_ALIGNMENT);
 		if (ret_code)
 			goto exit;
@@ -128,22 +99,22 @@ exit:
  *	   aligned on 4K boundary and zeroed memory.
  *	2. It should be 4K in size.
  **/
-i40e_status i40e_add_pd_table_entry(struct i40e_hw *hw,
-					      struct i40e_hmc_info *hmc_info,
-					      u32 pd_index,
-					      struct i40e_dma_mem *rsrc_pg)
+int i40e_add_pd_table_entry(struct i40e_hw *hw,
+			    struct i40e_hmc_info *hmc_info,
+			    u32 pd_index,
+			    struct i40e_dma_mem *rsrc_pg)
 {
-	i40e_status ret_code = 0;
 	struct i40e_hmc_pd_table *pd_table;
 	struct i40e_hmc_pd_entry *pd_entry;
 	struct i40e_dma_mem mem;
 	struct i40e_dma_mem *page = &mem;
 	u32 sd_idx, rel_pd_idx;
-	u64 *pd_addr;
+	int ret_code = 0;
 	u64 page_desc;
+	u64 *pd_addr;
 
 	if (pd_index / I40E_HMC_PD_CNT_IN_SD >= hmc_info->sd_table.sd_cnt) {
-		ret_code = I40E_ERR_INVALID_PAGE_DESC_INDEX;
+		ret_code = -EINVAL;
 		hw_dbg(hw, "i40e_add_pd_table_entry: bad pd_index\n");
 		goto exit;
 	}
@@ -163,7 +134,7 @@ i40e_status i40e_add_pd_table_entry(struct i40e_hw *hw,
 			page = rsrc_pg;
 		} else {
 			/* allocate a 4K backing page */
-			ret_code = i40e_allocate_dma_mem(hw, page, i40e_mem_bp,
+			ret_code = i40e_allocate_dma_mem(hw, page,
 						I40E_HMC_PAGED_BP_SIZE,
 						I40E_HMC_PD_BP_BUF_ALIGNMENT);
 			if (ret_code)
@@ -197,7 +168,6 @@ exit:
  * @hw: pointer to our HW structure
  * @hmc_info: pointer to the HMC configuration information structure
  * @idx: the page index
- * @is_pf: distinguishes a VF from a PF
  *
  * This function:
  *	1. Marks the entry in pd tabe (for paged address mode) or in sd table
@@ -208,28 +178,28 @@ exit:
  *	1. Caller can deallocate the memory used by backing storage after this
  *	   function returns.
  **/
-i40e_status i40e_remove_pd_bp(struct i40e_hw *hw,
-					struct i40e_hmc_info *hmc_info,
-					u32 idx)
+int i40e_remove_pd_bp(struct i40e_hw *hw,
+		      struct i40e_hmc_info *hmc_info,
+		      u32 idx)
 {
-	i40e_status ret_code = 0;
 	struct i40e_hmc_pd_entry *pd_entry;
 	struct i40e_hmc_pd_table *pd_table;
 	struct i40e_hmc_sd_entry *sd_entry;
 	u32 sd_idx, rel_pd_idx;
+	int ret_code = 0;
 	u64 *pd_addr;
 
 	/* calculate index */
 	sd_idx = idx / I40E_HMC_PD_CNT_IN_SD;
 	rel_pd_idx = idx % I40E_HMC_PD_CNT_IN_SD;
 	if (sd_idx >= hmc_info->sd_table.sd_cnt) {
-		ret_code = I40E_ERR_INVALID_PAGE_DESC_INDEX;
+		ret_code = -EINVAL;
 		hw_dbg(hw, "i40e_remove_pd_bp: bad idx\n");
 		goto exit;
 	}
 	sd_entry = &hmc_info->sd_table.sd_entry[sd_idx];
 	if (I40E_SD_TYPE_PAGED != sd_entry->entry_type) {
-		ret_code = I40E_ERR_INVALID_SD_TYPE;
+		ret_code = -EINVAL;
 		hw_dbg(hw, "i40e_remove_pd_bp: wrong sd_entry type\n");
 		goto exit;
 	}
@@ -264,17 +234,17 @@ exit:
  * @hmc_info: pointer to the HMC configuration information structure
  * @idx: the page index
  **/
-i40e_status i40e_prep_remove_sd_bp(struct i40e_hmc_info *hmc_info,
-					     u32 idx)
+int i40e_prep_remove_sd_bp(struct i40e_hmc_info *hmc_info,
+			   u32 idx)
 {
-	i40e_status ret_code = 0;
 	struct i40e_hmc_sd_entry *sd_entry;
+	int ret_code = 0;
 
 	/* get the entry and decrease its ref counter */
 	sd_entry = &hmc_info->sd_table.sd_entry[idx];
 	I40E_DEC_BP_REFCNT(&sd_entry->u.bp);
 	if (sd_entry->u.bp.ref_cnt) {
-		ret_code = I40E_ERR_NOT_READY;
+		ret_code = -EBUSY;
 		goto exit;
 	}
 	I40E_DEC_SD_REFCNT(&hmc_info->sd_table);
@@ -292,14 +262,14 @@ exit:
  * @idx: the page index
  * @is_pf: used to distinguish between VF and PF
  **/
-i40e_status i40e_remove_sd_bp_new(struct i40e_hw *hw,
-					    struct i40e_hmc_info *hmc_info,
-					    u32 idx, bool is_pf)
+int i40e_remove_sd_bp_new(struct i40e_hw *hw,
+			  struct i40e_hmc_info *hmc_info,
+			  u32 idx, bool is_pf)
 {
 	struct i40e_hmc_sd_entry *sd_entry;
 
 	if (!is_pf)
-		return I40E_NOT_SUPPORTED;
+		return -EOPNOTSUPP;
 
 	/* get the entry and decrease its ref counter */
 	sd_entry = &hmc_info->sd_table.sd_entry[idx];
@@ -313,16 +283,16 @@ i40e_status i40e_remove_sd_bp_new(struct i40e_hw *hw,
  * @hmc_info: pointer to the HMC configuration information structure
  * @idx: segment descriptor index to find the relevant page descriptor
  **/
-i40e_status i40e_prep_remove_pd_page(struct i40e_hmc_info *hmc_info,
-					       u32 idx)
+int i40e_prep_remove_pd_page(struct i40e_hmc_info *hmc_info,
+			     u32 idx)
 {
-	i40e_status ret_code = 0;
 	struct i40e_hmc_sd_entry *sd_entry;
+	int ret_code = 0;
 
 	sd_entry = &hmc_info->sd_table.sd_entry[idx];
 
 	if (sd_entry->u.pd_table.ref_cnt) {
-		ret_code = I40E_ERR_NOT_READY;
+		ret_code = -EBUSY;
 		goto exit;
 	}
 
@@ -341,14 +311,14 @@ exit:
  * @idx: segment descriptor index to find the relevant page descriptor
  * @is_pf: used to distinguish between VF and PF
  **/
-i40e_status i40e_remove_pd_page_new(struct i40e_hw *hw,
-					      struct i40e_hmc_info *hmc_info,
-					      u32 idx, bool is_pf)
+int i40e_remove_pd_page_new(struct i40e_hw *hw,
+			    struct i40e_hmc_info *hmc_info,
+			    u32 idx, bool is_pf)
 {
 	struct i40e_hmc_sd_entry *sd_entry;
 
 	if (!is_pf)
-		return I40E_NOT_SUPPORTED;
+		return -EOPNOTSUPP;
 
 	sd_entry = &hmc_info->sd_table.sd_entry[idx];
 	I40E_CLEAR_PF_SD_ENTRY(hw, idx, I40E_SD_TYPE_PAGED);
diff --git a/drivers/net/ethernet/intel/i40e/i40e_hmc.h b/drivers/net/ethernet/intel/i40e/i40e_hmc.h
index d90669211392..480e3a883cc7 100644
--- a/drivers/net/ethernet/intel/i40e/i40e_hmc.h
+++ b/drivers/net/ethernet/intel/i40e/i40e_hmc.h
@@ -1,32 +1,13 @@
-/*******************************************************************************
- *
- * Intel Ethernet Controller XL710 Family Linux Driver
- * Copyright(c) 2013 - 2014 Intel Corporation.
- *
- * This program is free software; you can redistribute it and/or modify it
- * under the terms and conditions of the GNU General Public License,
- * version 2, as published by the Free Software Foundation.
- *
- * This program is distributed in the hope it will be useful, but WITHOUT
- * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
- * FITNESS FOR A PARTICULAR PURPOSE.  See the GNU General Public License for
- * more details.
- *
- * You should have received a copy of the GNU General Public License along
- * with this program.  If not, see <http://www.gnu.org/licenses/>.
- *
- * The full GNU General Public License is included in this distribution in
- * the file called "COPYING".
- *
- * Contact Information:
- * e1000-devel Mailing List <e1000-devel@lists.sourceforge.net>
- * Intel Corporation, 5200 N.E. Elam Young Parkway, Hillsboro, OR 97124-6497
- *
- ******************************************************************************/
+/* SPDX-License-Identifier: GPL-2.0 */
+/* Copyright(c) 2013 - 2018 Intel Corporation. */
 
 #ifndef _I40E_HMC_H_
 #define _I40E_HMC_H_
 
+#include "i40e_alloc.h"
+#include "i40e_io.h"
+#include "i40e_register.h"
+
 #define I40E_HMC_MAX_BP_COUNT 512
 
 /* forward-declare the HW struct for the compiler */
@@ -37,7 +18,6 @@ struct i40e_hw;
 #define I40E_HMC_DIRECT_BP_SIZE		0x200000 /* 2M */
 #define I40E_HMC_PAGED_BP_SIZE		4096
 #define I40E_HMC_PD_BP_BUF_ALIGNMENT	4096
-#define I40E_FIRST_VF_FPM_ID		16
 
 struct i40e_hmc_obj_info {
 	u64 base;	/* base addr in FPM */
@@ -211,28 +191,28 @@ struct i40e_hmc_info {
 	/* add one more to the limit to correct our range */		\
 	*(pd_limit) += 1;						\
 }
-i40e_status i40e_add_sd_table_entry(struct i40e_hw *hw,
-					      struct i40e_hmc_info *hmc_info,
-					      u32 sd_index,
-					      enum i40e_sd_entry_type type,
-					      u64 direct_mode_sz);
-
-i40e_status i40e_add_pd_table_entry(struct i40e_hw *hw,
-					      struct i40e_hmc_info *hmc_info,
-					      u32 pd_index,
-					      struct i40e_dma_mem *rsrc_pg);
-i40e_status i40e_remove_pd_bp(struct i40e_hw *hw,
-					struct i40e_hmc_info *hmc_info,
-					u32 idx);
-i40e_status i40e_prep_remove_sd_bp(struct i40e_hmc_info *hmc_info,
-					     u32 idx);
-i40e_status i40e_remove_sd_bp_new(struct i40e_hw *hw,
-					    struct i40e_hmc_info *hmc_info,
-					    u32 idx, bool is_pf);
-i40e_status i40e_prep_remove_pd_page(struct i40e_hmc_info *hmc_info,
-					       u32 idx);
-i40e_status i40e_remove_pd_page_new(struct i40e_hw *hw,
-					      struct i40e_hmc_info *hmc_info,
-					      u32 idx, bool is_pf);
+
+int i40e_add_sd_table_entry(struct i40e_hw *hw,
+			    struct i40e_hmc_info *hmc_info,
+			    u32 sd_index,
+			    enum i40e_sd_entry_type type,
+			    u64 direct_mode_sz);
+int i40e_add_pd_table_entry(struct i40e_hw *hw,
+			    struct i40e_hmc_info *hmc_info,
+			    u32 pd_index,
+			    struct i40e_dma_mem *rsrc_pg);
+int i40e_remove_pd_bp(struct i40e_hw *hw,
+		      struct i40e_hmc_info *hmc_info,
+		      u32 idx);
+int i40e_prep_remove_sd_bp(struct i40e_hmc_info *hmc_info,
+			   u32 idx);
+int i40e_remove_sd_bp_new(struct i40e_hw *hw,
+			  struct i40e_hmc_info *hmc_info,
+			  u32 idx, bool is_pf);
+int i40e_prep_remove_pd_page(struct i40e_hmc_info *hmc_info,
+			     u32 idx);
+int i40e_remove_pd_page_new(struct i40e_hw *hw,
+			    struct i40e_hmc_info *hmc_info,
+			    u32 idx, bool is_pf);
 
 #endif /* _I40E_HMC_H_ */
diff --git a/drivers/net/ethernet/intel/i40e/i40e_io.h b/drivers/net/ethernet/intel/i40e/i40e_io.h
new file mode 100644
index 000000000000..2a2ed9a1d476
--- /dev/null
+++ b/drivers/net/ethernet/intel/i40e/i40e_io.h
@@ -0,0 +1,16 @@
+/* SPDX-License-Identifier: GPL-2.0 */
+/* Copyright(c) 2023 Intel Corporation. */
+
+#ifndef _I40E_IO_H_
+#define _I40E_IO_H_
+
+/* get readq/writeq support for 32 bit kernels, use the low-first version */
+#include <linux/io-64-nonatomic-lo-hi.h>
+
+#define wr32(a, reg, value)	writel((value), ((a)->hw_addr + (reg)))
+#define rd32(a, reg)		readl((a)->hw_addr + (reg))
+
+#define rd64(a, reg)		readq((a)->hw_addr + (reg))
+#define i40e_flush(a)		readl((a)->hw_addr + I40E_GLGEN_STAT)
+
+#endif /* _I40E_IO_H_ */
diff --git a/drivers/net/ethernet/intel/i40e/i40e_lan_hmc.c b/drivers/net/ethernet/intel/i40e/i40e_lan_hmc.c
index daa9204426d4..beaaf5c309d5 100644
--- a/drivers/net/ethernet/intel/i40e/i40e_lan_hmc.c
+++ b/drivers/net/ethernet/intel/i40e/i40e_lan_hmc.c
@@ -1,35 +1,10 @@
-/*******************************************************************************
- *
- * Intel Ethernet Controller XL710 Family Linux Driver
- * Copyright(c) 2013 - 2014 Intel Corporation.
- *
- * This program is free software; you can redistribute it and/or modify it
- * under the terms and conditions of the GNU General Public License,
- * version 2, as published by the Free Software Foundation.
- *
- * This program is distributed in the hope it will be useful, but WITHOUT
- * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
- * FITNESS FOR A PARTICULAR PURPOSE.  See the GNU General Public License for
- * more details.
- *
- * You should have received a copy of the GNU General Public License along
- * with this program.  If not, see <http://www.gnu.org/licenses/>.
- *
- * The full GNU General Public License is included in this distribution in
- * the file called "COPYING".
- *
- * Contact Information:
- * e1000-devel Mailing List <e1000-devel@lists.sourceforge.net>
- * Intel Corporation, 5200 N.E. Elam Young Parkway, Hillsboro, OR 97124-6497
- *
- ******************************************************************************/
+// SPDX-License-Identifier: GPL-2.0
+/* Copyright(c) 2013 - 2018 Intel Corporation. */
 
-#include "i40e_osdep.h"
-#include "i40e_register.h"
-#include "i40e_type.h"
-#include "i40e_hmc.h"
+#include "i40e_alloc.h"
+#include "i40e_debug.h"
 #include "i40e_lan_hmc.h"
-#include "i40e_prototype.h"
+#include "i40e_type.h"
 
 /* lan specific interface functions */
 
@@ -96,12 +71,12 @@ static u64 i40e_calculate_l2fpm_size(u32 txq_num, u32 rxq_num,
  * Assumptions:
  *   - HMC Resource Profile has been selected before calling this function.
  **/
-i40e_status i40e_init_lan_hmc(struct i40e_hw *hw, u32 txq_num,
-					u32 rxq_num, u32 fcoe_cntx_num,
-					u32 fcoe_filt_num)
+int i40e_init_lan_hmc(struct i40e_hw *hw, u32 txq_num,
+		      u32 rxq_num, u32 fcoe_cntx_num,
+		      u32 fcoe_filt_num)
 {
 	struct i40e_hmc_obj_info *obj, *full_obj;
-	i40e_status ret_code = 0;
+	int ret_code = 0;
 	u64 l2fpm_size;
 	u32 size_exp;
 
@@ -133,7 +108,7 @@ i40e_status i40e_init_lan_hmc(struct i40e_hw *hw, u32 txq_num,
 
 	/* validate values requested by driver don't exceed HMC capacity */
 	if (txq_num > obj->max_cnt) {
-		ret_code = I40E_ERR_INVALID_HMC_OBJ_COUNT;
+		ret_code = -EINVAL;
 		hw_dbg(hw, "i40e_init_lan_hmc: Tx context: asks for 0x%x but max allowed is 0x%x, returns error %d\n",
 			  txq_num, obj->max_cnt, ret_code);
 		goto init_lan_hmc_out;
@@ -156,7 +131,7 @@ i40e_status i40e_init_lan_hmc(struct i40e_hw *hw, u32 txq_num,
 
 	/* validate values requested by driver don't exceed HMC capacity */
 	if (rxq_num > obj->max_cnt) {
-		ret_code = I40E_ERR_INVALID_HMC_OBJ_COUNT;
+		ret_code = -EINVAL;
 		hw_dbg(hw, "i40e_init_lan_hmc: Rx context: asks for 0x%x but max allowed is 0x%x, returns error %d\n",
 			  rxq_num, obj->max_cnt, ret_code);
 		goto init_lan_hmc_out;
@@ -179,7 +154,7 @@ i40e_status i40e_init_lan_hmc(struct i40e_hw *hw, u32 txq_num,
 
 	/* validate values requested by driver don't exceed HMC capacity */
 	if (fcoe_cntx_num > obj->max_cnt) {
-		ret_code = I40E_ERR_INVALID_HMC_OBJ_COUNT;
+		ret_code = -EINVAL;
 		hw_dbg(hw, "i40e_init_lan_hmc: FCoE context: asks for 0x%x but max allowed is 0x%x, returns error %d\n",
 			  fcoe_cntx_num, obj->max_cnt, ret_code);
 		goto init_lan_hmc_out;
@@ -202,7 +177,7 @@ i40e_status i40e_init_lan_hmc(struct i40e_hw *hw, u32 txq_num,
 
 	/* validate values requested by driver don't exceed HMC capacity */
 	if (fcoe_filt_num > obj->max_cnt) {
-		ret_code = I40E_ERR_INVALID_HMC_OBJ_COUNT;
+		ret_code = -EINVAL;
 		hw_dbg(hw, "i40e_init_lan_hmc: FCoE filter: asks for 0x%x but max allowed is 0x%x, returns error %d\n",
 			  fcoe_filt_num, obj->max_cnt, ret_code);
 		goto init_lan_hmc_out;
@@ -251,11 +226,11 @@ init_lan_hmc_out:
  *	1. caller can deallocate the memory used by pd after this function
  *	   returns.
  **/
-static i40e_status i40e_remove_pd_page(struct i40e_hw *hw,
-						 struct i40e_hmc_info *hmc_info,
-						 u32 idx)
+static int i40e_remove_pd_page(struct i40e_hw *hw,
+			       struct i40e_hmc_info *hmc_info,
+			       u32 idx)
 {
-	i40e_status ret_code = 0;
+	int ret_code = 0;
 
 	if (!i40e_prep_remove_pd_page(hmc_info, idx))
 		ret_code = i40e_remove_pd_page_new(hw, hmc_info, idx, true);
@@ -278,11 +253,11 @@ static i40e_status i40e_remove_pd_page(struct i40e_hw *hw,
  *	1. caller can deallocate the memory used by backing storage after this
  *	   function returns.
  **/
-static i40e_status i40e_remove_sd_bp(struct i40e_hw *hw,
-					       struct i40e_hmc_info *hmc_info,
-					       u32 idx)
+static int i40e_remove_sd_bp(struct i40e_hw *hw,
+			     struct i40e_hmc_info *hmc_info,
+			     u32 idx)
 {
-	i40e_status ret_code = 0;
+	int ret_code = 0;
 
 	if (!i40e_prep_remove_sd_bp(hmc_info, idx))
 		ret_code = i40e_remove_sd_bp_new(hw, hmc_info, idx, true);
@@ -298,43 +273,43 @@ static i40e_status i40e_remove_sd_bp(struct i40e_hw *hw,
  * This will allocate memory for PDs and backing pages and populate
  * the sd and pd entries.
  **/
-static i40e_status i40e_create_lan_hmc_object(struct i40e_hw *hw,
-				struct i40e_hmc_lan_create_obj_info *info)
+static int i40e_create_lan_hmc_object(struct i40e_hw *hw,
+				      struct i40e_hmc_lan_create_obj_info *info)
 {
-	i40e_status ret_code = 0;
 	struct i40e_hmc_sd_entry *sd_entry;
 	u32 pd_idx1 = 0, pd_lmt1 = 0;
 	u32 pd_idx = 0, pd_lmt = 0;
 	bool pd_error = false;
 	u32 sd_idx, sd_lmt;
+	int ret_code = 0;
 	u64 sd_size;
 	u32 i, j;
 
 	if (NULL == info) {
-		ret_code = I40E_ERR_BAD_PTR;
+		ret_code = -EINVAL;
 		hw_dbg(hw, "i40e_create_lan_hmc_object: bad info ptr\n");
 		goto exit;
 	}
 	if (NULL == info->hmc_info) {
-		ret_code = I40E_ERR_BAD_PTR;
+		ret_code = -EINVAL;
 		hw_dbg(hw, "i40e_create_lan_hmc_object: bad hmc_info ptr\n");
 		goto exit;
 	}
 	if (I40E_HMC_INFO_SIGNATURE != info->hmc_info->signature) {
-		ret_code = I40E_ERR_BAD_PTR;
+		ret_code = -EINVAL;
 		hw_dbg(hw, "i40e_create_lan_hmc_object: bad signature\n");
 		goto exit;
 	}
 
 	if (info->start_idx >= info->hmc_info->hmc_obj[info->rsrc_type].cnt) {
-		ret_code = I40E_ERR_INVALID_HMC_OBJ_INDEX;
+		ret_code = -EINVAL;
 		hw_dbg(hw, "i40e_create_lan_hmc_object: returns error %d\n",
 			  ret_code);
 		goto exit;
 	}
 	if ((info->start_idx + info->count) >
 	    info->hmc_info->hmc_obj[info->rsrc_type].cnt) {
-		ret_code = I40E_ERR_INVALID_HMC_OBJ_COUNT;
+		ret_code = -EINVAL;
 		hw_dbg(hw, "i40e_create_lan_hmc_object: returns error %d\n",
 			  ret_code);
 		goto exit;
@@ -346,8 +321,8 @@ static i40e_status i40e_create_lan_hmc_object(struct i40e_hw *hw,
 				 &sd_idx, &sd_lmt);
 	if (sd_idx >= info->hmc_info->sd_table.sd_cnt ||
 	    sd_lmt > info->hmc_info->sd_table.sd_cnt) {
-			ret_code = I40E_ERR_INVALID_SD_INDEX;
-			goto exit;
+		ret_code = -EINVAL;
+		goto exit;
 	}
 	/* find pd index */
 	I40E_FIND_PD_INDEX_LIMIT(info->hmc_info, info->rsrc_type,
@@ -415,7 +390,7 @@ static i40e_status i40e_create_lan_hmc_object(struct i40e_hw *hw,
 						     j, sd_entry->entry_type);
 				break;
 			default:
-				ret_code = I40E_ERR_INVALID_SD_TYPE;
+				ret_code = -EINVAL;
 				goto exit;
 			}
 		}
@@ -439,7 +414,7 @@ exit_sd_error:
 			i40e_remove_sd_bp(hw, info->hmc_info, (j - 1));
 			break;
 		default:
-			ret_code = I40E_ERR_INVALID_SD_TYPE;
+			ret_code = -EINVAL;
 			break;
 		}
 		j--;
@@ -457,13 +432,13 @@ exit:
  * - This function will be called after i40e_init_lan_hmc() and before
  *   any LAN/FCoE HMC objects can be created.
  **/
-i40e_status i40e_configure_lan_hmc(struct i40e_hw *hw,
-					     enum i40e_hmc_model model)
+int i40e_configure_lan_hmc(struct i40e_hw *hw,
+			   enum i40e_hmc_model model)
 {
 	struct i40e_hmc_lan_create_obj_info info;
-	i40e_status ret_code = 0;
 	u8 hmc_fn_id = hw->hmc.hmc_fn_id;
 	struct i40e_hmc_obj_info *obj;
+	int ret_code = 0;
 
 	/* Initialize part of the create object info struct */
 	info.hmc_info = &hw->hmc;
@@ -496,7 +471,7 @@ try_type_paged:
 		break;
 	default:
 		/* unsupported type */
-		ret_code = I40E_ERR_INVALID_SD_TYPE;
+		ret_code = -EINVAL;
 		hw_dbg(hw, "i40e_configure_lan_hmc: Unknown SD type: %d\n",
 			  ret_code);
 		goto configure_lan_hmc_out;
@@ -533,7 +508,7 @@ configure_lan_hmc_out:
 }
 
 /**
- * i40e_delete_hmc_object - remove hmc objects
+ * i40e_delete_lan_hmc_object - remove hmc objects
  * @hw: pointer to the HW structure
  * @info: pointer to i40e_hmc_delete_obj_info struct
  *
@@ -542,44 +517,44 @@ configure_lan_hmc_out:
  * caller should deallocate memory allocated previously for
  * book-keeping information about PDs and backing storage.
  **/
-static i40e_status i40e_delete_lan_hmc_object(struct i40e_hw *hw,
-				struct i40e_hmc_lan_delete_obj_info *info)
+static int i40e_delete_lan_hmc_object(struct i40e_hw *hw,
+				      struct i40e_hmc_lan_delete_obj_info *info)
 {
-	i40e_status ret_code = 0;
 	struct i40e_hmc_pd_table *pd_table;
 	u32 pd_idx, pd_lmt, rel_pd_idx;
 	u32 sd_idx, sd_lmt;
+	int ret_code = 0;
 	u32 i, j;
 
 	if (NULL == info) {
-		ret_code = I40E_ERR_BAD_PTR;
+		ret_code = -EINVAL;
 		hw_dbg(hw, "i40e_delete_hmc_object: bad info ptr\n");
 		goto exit;
 	}
 	if (NULL == info->hmc_info) {
-		ret_code = I40E_ERR_BAD_PTR;
+		ret_code = -EINVAL;
 		hw_dbg(hw, "i40e_delete_hmc_object: bad info->hmc_info ptr\n");
 		goto exit;
 	}
 	if (I40E_HMC_INFO_SIGNATURE != info->hmc_info->signature) {
-		ret_code = I40E_ERR_BAD_PTR;
+		ret_code = -EINVAL;
 		hw_dbg(hw, "i40e_delete_hmc_object: bad hmc_info->signature\n");
 		goto exit;
 	}
 
 	if (NULL == info->hmc_info->sd_table.sd_entry) {
-		ret_code = I40E_ERR_BAD_PTR;
+		ret_code = -EINVAL;
 		hw_dbg(hw, "i40e_delete_hmc_object: bad sd_entry\n");
 		goto exit;
 	}
 
 	if (NULL == info->hmc_info->hmc_obj) {
-		ret_code = I40E_ERR_BAD_PTR;
+		ret_code = -EINVAL;
 		hw_dbg(hw, "i40e_delete_hmc_object: bad hmc_info->hmc_obj\n");
 		goto exit;
 	}
 	if (info->start_idx >= info->hmc_info->hmc_obj[info->rsrc_type].cnt) {
-		ret_code = I40E_ERR_INVALID_HMC_OBJ_INDEX;
+		ret_code = -EINVAL;
 		hw_dbg(hw, "i40e_delete_hmc_object: returns error %d\n",
 			  ret_code);
 		goto exit;
@@ -587,7 +562,7 @@ static i40e_status i40e_delete_lan_hmc_object(struct i40e_hw *hw,
 
 	if ((info->start_idx + info->count) >
 	    info->hmc_info->hmc_obj[info->rsrc_type].cnt) {
-		ret_code = I40E_ERR_INVALID_HMC_OBJ_COUNT;
+		ret_code = -EINVAL;
 		hw_dbg(hw, "i40e_delete_hmc_object: returns error %d\n",
 			  ret_code);
 		goto exit;
@@ -621,7 +596,7 @@ static i40e_status i40e_delete_lan_hmc_object(struct i40e_hw *hw,
 				 &sd_idx, &sd_lmt);
 	if (sd_idx >= info->hmc_info->sd_table.sd_cnt ||
 	    sd_lmt > info->hmc_info->sd_table.sd_cnt) {
-		ret_code = I40E_ERR_INVALID_SD_INDEX;
+		ret_code = -EINVAL;
 		goto exit;
 	}
 
@@ -654,10 +629,10 @@ exit:
  * This must be called by drivers as they are shutting down and being
  * removed from the OS.
  **/
-i40e_status i40e_shutdown_lan_hmc(struct i40e_hw *hw)
+int i40e_shutdown_lan_hmc(struct i40e_hw *hw)
 {
 	struct i40e_hmc_lan_delete_obj_info info;
-	i40e_status ret_code;
+	int ret_code;
 
 	info.hmc_info = &hw->hmc;
 	info.rsrc_type = I40E_HMC_LAN_FULL;
@@ -681,7 +656,7 @@ i40e_status i40e_shutdown_lan_hmc(struct i40e_hw *hw)
 
 #define I40E_HMC_STORE(_struct, _ele)		\
 	offsetof(struct _struct, _ele),		\
-	FIELD_SIZEOF(struct _struct, _ele)
+	sizeof_field(struct _struct, _ele)
 
 struct i40e_context_ele {
 	u16 offset;
@@ -937,9 +912,9 @@ static void i40e_write_qword(u8 *hmc_bits,
  * @context_bytes: pointer to the context bit array (DMA memory)
  * @hmc_type: the type of HMC resource
  **/
-static i40e_status i40e_clear_hmc_context(struct i40e_hw *hw,
-					u8 *context_bytes,
-					enum i40e_hmc_lan_rsrc_type hmc_type)
+static int i40e_clear_hmc_context(struct i40e_hw *hw,
+				  u8 *context_bytes,
+				  enum i40e_hmc_lan_rsrc_type hmc_type)
 {
 	/* clean the bit array */
 	memset(context_bytes, 0, (u32)hw->hmc.hmc_obj[hmc_type].size);
@@ -953,9 +928,9 @@ static i40e_status i40e_clear_hmc_context(struct i40e_hw *hw,
  * @ce_info:  a description of the struct to be filled
  * @dest:     the struct to be filled
  **/
-static i40e_status i40e_set_hmc_context(u8 *context_bytes,
-					struct i40e_context_ele *ce_info,
-					u8 *dest)
+static int i40e_set_hmc_context(u8 *context_bytes,
+				struct i40e_context_ele *ce_info,
+				u8 *dest)
 {
 	int f;
 
@@ -986,7 +961,7 @@ static i40e_status i40e_set_hmc_context(u8 *context_bytes,
 
 /**
  * i40e_hmc_get_object_va - retrieves an object's virtual address
- * @hmc_info: pointer to i40e_hmc_info struct
+ * @hw: the hardware struct, from which we obtain the i40e_hmc_info pointer
  * @object_base: pointer to u64 to get the va
  * @rsrc_type: the hmc resource type
  * @obj_idx: hmc object index
@@ -995,43 +970,43 @@ static i40e_status i40e_set_hmc_context(u8 *context_bytes,
  * base pointer.  This function is used for LAN Queue contexts.
  **/
 static
-i40e_status i40e_hmc_get_object_va(struct i40e_hmc_info *hmc_info,
-					u8 **object_base,
-					enum i40e_hmc_lan_rsrc_type rsrc_type,
-					u32 obj_idx)
+int i40e_hmc_get_object_va(struct i40e_hw *hw, u8 **object_base,
+			   enum i40e_hmc_lan_rsrc_type rsrc_type,
+			   u32 obj_idx)
 {
+	struct i40e_hmc_info *hmc_info = &hw->hmc;
 	u32 obj_offset_in_sd, obj_offset_in_pd;
-	i40e_status ret_code = 0;
 	struct i40e_hmc_sd_entry *sd_entry;
 	struct i40e_hmc_pd_entry *pd_entry;
 	u32 pd_idx, pd_lmt, rel_pd_idx;
 	u64 obj_offset_in_fpm;
 	u32 sd_idx, sd_lmt;
+	int ret_code = 0;
 
 	if (NULL == hmc_info) {
-		ret_code = I40E_ERR_BAD_PTR;
+		ret_code = -EINVAL;
 		hw_dbg(hw, "i40e_hmc_get_object_va: bad hmc_info ptr\n");
 		goto exit;
 	}
 	if (NULL == hmc_info->hmc_obj) {
-		ret_code = I40E_ERR_BAD_PTR;
+		ret_code = -EINVAL;
 		hw_dbg(hw, "i40e_hmc_get_object_va: bad hmc_info->hmc_obj ptr\n");
 		goto exit;
 	}
 	if (NULL == object_base) {
-		ret_code = I40E_ERR_BAD_PTR;
+		ret_code = -EINVAL;
 		hw_dbg(hw, "i40e_hmc_get_object_va: bad object_base ptr\n");
 		goto exit;
 	}
 	if (I40E_HMC_INFO_SIGNATURE != hmc_info->signature) {
-		ret_code = I40E_ERR_BAD_PTR;
+		ret_code = -EINVAL;
 		hw_dbg(hw, "i40e_hmc_get_object_va: bad hmc_info->signature\n");
 		goto exit;
 	}
 	if (obj_idx >= hmc_info->hmc_obj[rsrc_type].cnt) {
 		hw_dbg(hw, "i40e_hmc_get_object_va: returns error %d\n",
 			  ret_code);
-		ret_code = I40E_ERR_INVALID_HMC_OBJ_INDEX;
+		ret_code = -EINVAL;
 		goto exit;
 	}
 	/* find sd index and limit */
@@ -1064,13 +1039,13 @@ exit:
  * @hw:    the hardware struct
  * @queue: the queue we care about
  **/
-i40e_status i40e_clear_lan_tx_queue_context(struct i40e_hw *hw,
-						      u16 queue)
+int i40e_clear_lan_tx_queue_context(struct i40e_hw *hw,
+				    u16 queue)
 {
-	i40e_status err;
 	u8 *context_bytes;
+	int err;
 
-	err = i40e_hmc_get_object_va(&hw->hmc, &context_bytes,
+	err = i40e_hmc_get_object_va(hw, &context_bytes,
 				     I40E_HMC_LAN_TX, queue);
 	if (err < 0)
 		return err;
@@ -1084,14 +1059,14 @@ i40e_status i40e_clear_lan_tx_queue_context(struct i40e_hw *hw,
  * @queue: the queue we care about
  * @s:     the struct to be filled
  **/
-i40e_status i40e_set_lan_tx_queue_context(struct i40e_hw *hw,
-						    u16 queue,
-						    struct i40e_hmc_obj_txq *s)
+int i40e_set_lan_tx_queue_context(struct i40e_hw *hw,
+				  u16 queue,
+				  struct i40e_hmc_obj_txq *s)
 {
-	i40e_status err;
 	u8 *context_bytes;
+	int err;
 
-	err = i40e_hmc_get_object_va(&hw->hmc, &context_bytes,
+	err = i40e_hmc_get_object_va(hw, &context_bytes,
 				     I40E_HMC_LAN_TX, queue);
 	if (err < 0)
 		return err;
@@ -1105,13 +1080,13 @@ i40e_status i40e_set_lan_tx_queue_context(struct i40e_hw *hw,
  * @hw:    the hardware struct
  * @queue: the queue we care about
  **/
-i40e_status i40e_clear_lan_rx_queue_context(struct i40e_hw *hw,
-						      u16 queue)
+int i40e_clear_lan_rx_queue_context(struct i40e_hw *hw,
+				    u16 queue)
 {
-	i40e_status err;
 	u8 *context_bytes;
+	int err;
 
-	err = i40e_hmc_get_object_va(&hw->hmc, &context_bytes,
+	err = i40e_hmc_get_object_va(hw, &context_bytes,
 				     I40E_HMC_LAN_RX, queue);
 	if (err < 0)
 		return err;
@@ -1125,14 +1100,14 @@ i40e_status i40e_clear_lan_rx_queue_context(struct i40e_hw *hw,
  * @queue: the queue we care about
  * @s:     the struct to be filled
  **/
-i40e_status i40e_set_lan_rx_queue_context(struct i40e_hw *hw,
-						    u16 queue,
-						    struct i40e_hmc_obj_rxq *s)
+int i40e_set_lan_rx_queue_context(struct i40e_hw *hw,
+				  u16 queue,
+				  struct i40e_hmc_obj_rxq *s)
 {
-	i40e_status err;
 	u8 *context_bytes;
+	int err;
 
-	err = i40e_hmc_get_object_va(&hw->hmc, &context_bytes,
+	err = i40e_hmc_get_object_va(hw, &context_bytes,
 				     I40E_HMC_LAN_RX, queue);
 	if (err < 0)
 		return err;
diff --git a/drivers/net/ethernet/intel/i40e/i40e_lan_hmc.h b/drivers/net/ethernet/intel/i40e/i40e_lan_hmc.h
index e74128db5be5..305a276953b0 100644
--- a/drivers/net/ethernet/intel/i40e/i40e_lan_hmc.h
+++ b/drivers/net/ethernet/intel/i40e/i40e_lan_hmc.h
@@ -1,32 +1,11 @@
-/*******************************************************************************
- *
- * Intel Ethernet Controller XL710 Family Linux Driver
- * Copyright(c) 2013 - 2014 Intel Corporation.
- *
- * This program is free software; you can redistribute it and/or modify it
- * under the terms and conditions of the GNU General Public License,
- * version 2, as published by the Free Software Foundation.
- *
- * This program is distributed in the hope it will be useful, but WITHOUT
- * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
- * FITNESS FOR A PARTICULAR PURPOSE.  See the GNU General Public License for
- * more details.
- *
- * You should have received a copy of the GNU General Public License along
- * with this program.  If not, see <http://www.gnu.org/licenses/>.
- *
- * The full GNU General Public License is included in this distribution in
- * the file called "COPYING".
- *
- * Contact Information:
- * e1000-devel Mailing List <e1000-devel@lists.sourceforge.net>
- * Intel Corporation, 5200 N.E. Elam Young Parkway, Hillsboro, OR 97124-6497
- *
- ******************************************************************************/
+/* SPDX-License-Identifier: GPL-2.0 */
+/* Copyright(c) 2013 - 2018 Intel Corporation. */
 
 #ifndef _I40E_LAN_HMC_H_
 #define _I40E_LAN_HMC_H_
 
+#include "i40e_hmc.h"
+
 /* forward-declare the HW struct for the compiler */
 struct i40e_hw;
 
@@ -160,22 +139,22 @@ struct i40e_hmc_lan_delete_obj_info {
 	u32 count;
 };
 
-i40e_status i40e_init_lan_hmc(struct i40e_hw *hw, u32 txq_num,
-					u32 rxq_num, u32 fcoe_cntx_num,
-					u32 fcoe_filt_num);
-i40e_status i40e_configure_lan_hmc(struct i40e_hw *hw,
-					     enum i40e_hmc_model model);
-i40e_status i40e_shutdown_lan_hmc(struct i40e_hw *hw);
-
-i40e_status i40e_clear_lan_tx_queue_context(struct i40e_hw *hw,
-						      u16 queue);
-i40e_status i40e_set_lan_tx_queue_context(struct i40e_hw *hw,
-						    u16 queue,
-						    struct i40e_hmc_obj_txq *s);
-i40e_status i40e_clear_lan_rx_queue_context(struct i40e_hw *hw,
-						      u16 queue);
-i40e_status i40e_set_lan_rx_queue_context(struct i40e_hw *hw,
-						    u16 queue,
-						    struct i40e_hmc_obj_rxq *s);
+int i40e_init_lan_hmc(struct i40e_hw *hw, u32 txq_num,
+		      u32 rxq_num, u32 fcoe_cntx_num,
+		      u32 fcoe_filt_num);
+int i40e_configure_lan_hmc(struct i40e_hw *hw,
+			   enum i40e_hmc_model model);
+int i40e_shutdown_lan_hmc(struct i40e_hw *hw);
+
+int i40e_clear_lan_tx_queue_context(struct i40e_hw *hw,
+				    u16 queue);
+int i40e_set_lan_tx_queue_context(struct i40e_hw *hw,
+				  u16 queue,
+				  struct i40e_hmc_obj_txq *s);
+int i40e_clear_lan_rx_queue_context(struct i40e_hw *hw,
+				    u16 queue);
+int i40e_set_lan_rx_queue_context(struct i40e_hw *hw,
+				  u16 queue,
+				  struct i40e_hmc_obj_rxq *s);
 
 #endif /* _I40E_LAN_HMC_H_ */
diff --git a/drivers/net/ethernet/intel/i40e/i40e_main.c b/drivers/net/ethernet/intel/i40e/i40e_main.c
index e4eb97832413..50be0a60ae13 100644
--- a/drivers/net/ethernet/intel/i40e/i40e_main.c
+++ b/drivers/net/ethernet/intel/i40e/i40e_main.c